SELECTIVE INCORPORATION OF THE C-F BOND AS A CONFORMATIONAL TOOL IN QUADRUPLEX DNA LIGAND DESIGN Daniel L. Smith A Thesis Submitted for the Degree of PhD at the University of St Andrews 2012 Full metadata for this item is available in St Andrews Research Repository at: http://research-repository.st-andrews.ac.uk/ Please use this identifier to cite or link to this item: http://hdl.handle.net/10023/3169 This item is protected by original copyright Selective incorporation of the C–F bond as a conformational tool in quadruplex DNA ligand design School of Chemistry Daniel L. Smith June 2012 Thesis submitted to the University of St Andrews for the degree of Doctor of Philosophy Supervisor: Prof. David O’Hagan ii 1. Candidate’s declarations: I, Daniel Smith, hereby certify that this thesis, which is approximately 50,000 words in length, has been written by me, that it is the record of work carried out by me and that it has not been submitted in any previous application for a higher degree. I was admitted as a research student in September 2008 and as a candidate for the degree of PhD in September 2009; the higher study for which this is a record was carried out in the University of St Andrews between 2008 and 2012 Date: Signature of Candidate: 2. Supervisor’s declaration: I hereby certify that the candidate has fulfilled the conditions of the Resolution and Regulations appropriate for the degree of Doctor of Philosophy in the University of St Andrews and that the candidate is qualified to submit this thesis in application for that degree. Date: Signature of Supervisor: iii 3. Permission for electronic publication: In submitting this thesis to the University of St Andrews I understand that I am giving permission for it to be made available for use in accordance with the regulations of the University Library for the time being in force, subject to any copyright vested in the work not being affected thereby. I also understand that the title and the abstract will be published, and that a copy of the work may be made and supplied to any bona fide library or research worker, that my thesis will be electronically accessible for personal or research use unless exempt by award of an embargo as requested below, and that the library has the right to migrate my thesis into new electronic forms as required to ensure continued access to the thesis. I have obtained any third-party copyright permissions that may be required in order to allow such access and migration, or have requested the appropriate embargo below. The following is an agreed request by candidate and supervisor regarding the electronic publication of this thesis: Embargo on both all of printed copy and electronic copy for the same fixed period of two years on the following ground: publication would preclude future publication. Date: Signature of Candidate: Signature of Supervisor: Dedicated to my Papa v Acknowledgements I would like to thank my supervisor Prof. David O’Hagan for the opportunity to conduct research under his supervision. Our many conversations around the project and about science in general have been instrumental throughout my PhD and in my professional development. I would also like to extend my gratitude towards Dr Nancy Campbell for her tireless work and effort during our collaboration. Her patience and expertise has been of great value. The work presented in this thesis would have not been possible if it was not for the help, support and skill of all of the technical staff at the School of Chemistry and the BSRC. I am particularly indebted to Melanja Smith for NMR assistance and Caroline Horsburgh for mass spectroscopic analysis. I am also thankful for the assistance of Dr Tomas Lebl with NMR analysis and to Prof. Alexandra Slawin for X-ray crystallography and advice. Many people have made my time in St Andrews a very memorable and enjoyable experience. All members of the DOH group and the BSRC past and present have contributed in their own special way, however, I would like to particularly acknowledge the following people for their respective contributions: Andrew Nortcliffe for his friendship, invaluable support and advice; Dr Daniel Farran, Dr Jason Schmidberger, Romain Cadou and Thomas Moraux; the forbidden BSRC quartet of Alan, Ray, Stan and Stevie; Nawaf and Fraser for the Dundee commute; Dr Neil Keddie and Dr Michael Corr for their careful proof reading; and finally to my thesis office partner Joanna Wlochal for her sympathetic and supportive attitude. I am also immensely grateful for the love and support of my family, Marian and Neville Mansell, Steven’s Gogo, Robert Kennedy and my friends. Finally and most importantly, I want to acknowledge the love, understanding and motivation from my partner Steven, whom has been instrumental throughout my PhD. Cancer Research UK generously sponsored this work. vi Table of Contents Abbreviations .............................................................................................................. x!Abstract ...................................................................................................................... xv Chapter 1: Synthesis & properties of fluorinated compounds 1.1 - A brief history ........................................................................................................ 1!1.2 - Fluorination techniques ........................................................................................... 1!1.3 - Electrophilic fluorinating reagents .......................................................................... 2!1.4 - Nucleophilic fluorinating reagents .......................................................................... 6!1.5 - Trifluoromethylation of aromatic and heteroaromatic rings ................................... 9!1.6 - Properties of fluorine in organic molecules .......................................................... 12!1.7 - Acidity and basicity .............................................................................................. 13!1.8 - Fluorine in drug metabolism ................................................................................. 13!1.9 - Fluorine based suicide inhibitors .......................................................................... 14!1.10 - 19F NMR probes in chemical biology ................................................................. 15!1.10.1 - General applications ................................................................................... 15!1.10.2 - rRNA conformation probes .......................................................................... 15!1.10.3 - Membrane transport kinetics ....................................................................... 16 1.11 - Conformational effects of fluorine ..................................................................... 17!1.11.1 - The gauche effect ........................................................................................ 17!1.11.2 - The α-fluoroamide effect .............................................................................. 20 1.11.3 - The charge-dipole effect .............................................................................. 22 1.12 - Applications of the charge-dipole effect ............................................................. 24!1.12.1 - Organocatalysts .......................................................................................... 24!1.12.2 - Biological exploitation of the C–F bond ....................................................... 25 1.13 - Synthesis of fluorinated β-amino acids ............................................................... 28!1.13.1 - General methods ......................................................................................... 28!1.13.2 - Evans oxazolidine approach ........................................................................ 29!1.13.3 - Davies’ lithium amide approach .................................................................. 30 1.14 - Conclusion .......................................................................................................... 31! vii Chapter 2: Telomeres, telomerase and quadruplex DNA 2.1 - The 2009 Nobel prize to telomeres ....................................................................... 32!2.2 - Telomeres and telomerase .................................................................................... 32!2.3 - Telomerase and cancer .......................................................................................... 34!2.4 - Shelterin complex at the telomere ........................................................................ 34!2.5 - Self-assembly of guanosine .................................................................................. 35!2.6 - Quadruplex DNA folding and topology ............................................................... 36!2.7 - Telomerase inhibition ........................................................................................... 41!2.8 - Assessing telomerase inhibition and quadruplex stability .................................... 41!2.9 - Quadruplex DNA stabilising ligands .................................................................... 43!2.9.1 - Natural products and analogues .................................................................... 44!2.9.2 - Porphyrin based inhibitors ............................................................................ 45!2.9.3 - Quinacridine ligands ..................................................................................... 46!2.9.4 - Anthraquinone and fluorenone ligands ........................................................... 47!2.9.5 - Acridone and di- and tri-substituted acridine ligands ...................................... 48 2.10 - BRACO-19 142a in vitro & in vivo studies ........................................................ 49!2.11 - X-ray crystallographic studies with acridine based ligands ................................ 51 2.11.1 - BSU6039 141a with O. nova DNA ................................................................ 51 2.11.2 - BRACO-19 142a with quadruplex DNA ........................................................ 53 2.12 - Conclusions ......................................................................................................... 55! Chapter 3: Synthesis and evaluation of fluorinated BSU6039 analogues 3.1 - Introduction ........................................................................................................... 56!3.2 - Synthesis of BSU6039 141a analogues ................................................................ 58!3.3 - Characterisation of (S,S)- and (R,R)-144 .............................................................. 60 3.4 - Fluoropyrrolidine ring conformation in 144.HCl ................................................. 62 3.5 - Co-crystallisation with quadruplex DNA ............................................................. 65!3.5.1 - Background and crystallisation ...................................................................... 66!3.5.2 - General observations in the co-crystals with (S,S)- and (R,R)-144 .................... 68!3.5.3 - Detailed assessment of the DNA co-crystal with (S,S)-144 ............................... 71!3.5.4 - Detailed assessment of the DNA co-crystal with (R,R)-144 .............................. 73 viii 3.6 - FRET studies with quadruplex DNA .................................................................... 75!3.7 - Conclusion ............................................................................................................ 76 Chapter 4: Synthesis of C–F bond incorporated BRACO-19 analogues 4.1 - Introduction ........................................................................................................... 77!4.2 - Aims ...................................................................................................................... 79!4.3 - Synthesis of an α-fluoro-β-amino acid ................................................................. 80!4.3.1 - Pyrrolidine functionalisation of 166 ............................................................... 83 4.4 - Acridone 155 synthesis ......................................................................................... 84!4.5 - Coupling reactions with diaminoacridine 155 ...................................................... 85!4.5.1 - Debenzylation of acridone 169 ....................................................................... 89 4.6 - Alternative protecting groups ............................................................................... 90!4.6.2 - Ring closing metathesis approach with 188 .................................................... 92!4.7 - Acridone coupling with ester 188 ......................................................................... 93!4.7.1 - Allyl deprotection of acridone 195 ................................................................. 94!4.7.2 - Functionalisation of acridone 196 .................................................................. 97!4.8 - Trisubstituted acridine 206 synthesis .................................................................... 98!4.8.1 - Allyl deprotection of acridine 206 ................................................................ 100 4.9 - Alternative side chain functionalisation ............................................................. 103!4.9.1 - Acridone coupling with ester 209 ................................................................. 103!4.9.2 - 1H-19F HOESY analysis of (S,S)-212 ............................................................. 107 4.10 - Non-fluorinated BRACO-19 142a analogues ................................................... 108!4.11 - Crystallographic assessment ............................................................................. 111!4.12 - Conclusions ....................................................................................................... 112! Chapter 5: Studies on the selective fluorination of dipeptides 5.1 - Introduction ......................................................................................................... 113!5.2 - Carboxylic acid synthesis for peptide couplings ................................................ 114!5.3 - Peptide couplings ................................................................................................ 115!5.3.1 - Fluorination reactions of dipeptides 224a-c with DAST 32 ............................ 118 5.3.2 - Dipeptide conformation in 227a ................................................................... 123 ix 5.4 - Preparation of α-amino acid N–H and N–CH3 amide derivatives ...................... 124!5.4.1 - Fluorination reactions of amides 241a/b ...................................................... 126 5.5 - Extending the applicability to useful N-substituted amides ................................ 129!5.6 - Tertiary allylamide from secondary dipeptides .................................................. 132!5.6.1 - N-Allyl amide 244 dipeptide fluorination with DAST 32 ................................ 138 5.7 - N-Allyl amide 245 deprotection ......................................................................... 139!5.8 - Conclusions ......................................................................................................... 143! Chapter 6: Future Work 6.1 - Future work for Chapter 3 .................................................................................. 144 6.2 - Future work for Chapter 4 .................................................................................. 144 6.3 - Future work for Chapter 5 .................................................................................. 146 Chapter 7: Experimental 7.1 - General experimental procedures ....................................................................... 148 7.2 - Experimental for Chapter 3 ................................................................................. 152 7.3 - Experimental for Chapter 4 ................................................................................. 158 7.4 - Experimental for Chapter 5 ................................................................................. 208 7.4.1 - General proceedures ................................................................................... 208 References ........................................................................................................... 245 Appendix Appendix 1.1 - Crystallographic information for (R,R)-144 ....................................... 258 Appendix 1.2 - Crystallographic information for (S,S)-144 ....................................... 259 Appendix 1.3 - Crystallographic information for 234 ................................................ 260 Appendix 1.4 - Selected NMR .................................................................................... 262 x Abbreviations {1H} - proton decoupled 5-FU - 5-fluorouracil A - adenine Å - Angstrom Ala - alanine ap - antiperiplanar aq - aqueous Ar - aryl ASAP MS - atmospheric solids probe analysis mass spectrometry atm - atmospheric pressure ax - axial Bn - benzyl Boc - tert-butoxycarbonyl br. - broad c - concentration calc. - calculated CD - circular dichroism CDI - 1,1′-carbonyldiimidazole cf. - compare concd - concentrated COSY - correlation spectroscopy CSD - Cambridge Structural Database d - doublet d6-DMSO - deuterated dimethyl sulfoxide DAST - diethylaminosulfur trifluoride dba - dibenzylideneacetone DBU - 1,8-diazabicycloundec-7-ene DCC - dicyclohexylcarbodiimide dd - doublet of doublets xi de - diastereomeric excess dec. - decomposition δ - Nuclear magnetic resonance chemical shift parts per million downfield from a standard ΔTm - difference in melting temperature Deoxo-Fluor® - dimethoxyethylaminosulfur trifluoride DEPT - distortionless enhancement by polarization transfer DFI - 2,2-difluoro-1,3-dimethylimidazolidine DFT - density functional theory DIC - N,N′-diisopropylcarbodiimide DIPEA - N,N-diisopropylethylamine DMAc - N,N-dimethylacetamide DMAP - N,N-dimethylaminopyridine DMF - N,N-dimethylformamide DMSO - dimethyl sulfoxide DNA - deoxyribonucleic acid dppb - 1,4-bis(diphenylphosphino)butane dr - diastereomeric ratio ε - molar extinction coefficient ED50 - dose that is effect in 50% of test subjects EDCI - N-(3-dimethylaminopropyl)-N′-ethylcarbodiimide hydrochloride ee - enantiomeric excess EI - electron ionisation eq - equivalent ES - electrospray ionisation EXSY - exchange spectroscopy FAM - carboxyfluorescein FRET - Förster resonance energy transfer FT-IR - Fourier transform infrared spectroscopy G - guanine g - grams g- - gauche torsion angle xii g+ - gauche torsion angle G0 - resting phase of the cell cycle GABA - γ-aminobutyric acid GABAA/B/C - γ-aminobutyric acid receptor subclass A/B/C GMP - guanosine monophosphate Go - any number of guanine nucleotides GP - general procedure h - hour HATU - O-(7-azabenzotriazol-1-yl)-N,N,N′,N′- tetramethyluronium hexafluorophosphate HBTU - O-(benzotriazol-1-yl)-N,N,N′,N′- tetramethyluronium hexafluorophosphate Heq - equatorial hydrogen HMBC - heteronuclear multiple-bond correlation spectroscopy HMDS - 1,1,1,3,3,3-hexamethyldisilazane HOBt - hydroxybenzotriazole HOESY - heteronuclear Overhauser effect spectroscopy HPLC - high-performance liquid chromatography HRMS - high resolution mass spectroscopy HSQC - heteronuclear single-quantum correlation spectroscopy hTERT - human telomerase reverse transcriptase hTR - human telomerase ribonucleic acid Hz - Hertz ID50 - dose that inhibits at 50% of maximum response IR - infrared J - coupling constant L - litre ℓ - path length Lit. - literature reference LHS - left hand side M - molar m - multiplet xiii m/z - mass to charge ratio mg - milligrams MHz - megahertz min - minutes mL - milliliters MO - molecular orbital MOST - morpholinosulfur trifluoride mp - melting point n - number N/T - not tested NFSI - N-fluorobenzenesulfonamide NHEJ - non-homologous end joining NMM - N-methylmorpholine NMR - nuclear magnetic resonance nOe - nuclear Overhauser effect NOESY - nuclear Overhauser effect spectroscopy O. nova - Oxytricha nova t-Bu P4 - tert-butyl P4 phosphazene P450 - specific cytochrome enzyme subtype PDB - protein database PET - positron emission tomography Ph - phenyl Phe - phenylalanine Phen - phenanthroline π* - antibonding π orbital POT1 - protection of telomeres 1 protein ppm - parts per million PyBrop - bromotripyrrolidinophosphonium hexafluorophosphate q - quartet Rf - retention factor RHS - right hand side rRNA - ribosomal ribonucleic acid rt - room temperature xiv s - singlet Ser - serine SET - single electron transfer σ* - antibonding σ orbital soln. - solution ssDNA - single stranded deoxyribonucleic acid T - thymine t - triplet T3P® - propylphosphonic anhydride TAMRA - carboxytetramethylrhodamine TBAF - tetrabutylammonium fluoride TBAI - tetrabutylammonium iodide TBDMS - tert-butyldimethylsilyl TBSOTf - tert-butyldimethylsilyl trifluoromethanesulfonate telEC50 - dose that is 50% effective against the action of telomerase telIC50 - dose that results in 50% inhibition of telomerase temp - temperature TFAc - trifluoroacetate THF - tetrahydrofuran TLC - thin layer chromatography TRAP - telomeric repeat amplification protocol tRNA - transfer ribonucleic acid UV-Vis - ultraviolet-visible v/v - volume per volume Val - valine w/v - weight per volume w/w - weight per weight Xn - any number non-guanine nucleotides Xp - any number of non-guanine nucleotides involved in loop formation XtalFluorE® - morpholinodifluorosulfonium tetrafluoroborate XtalFluorM® - (diethylamino)difluorosulfonium tetrafluoroborate xv Abstract Chapter 1 provides a general introduction to organofluorine chemistry and focuses on recent developments in fluorination techniques. It also details how the C–F bond influences conformational and physiochemical properties of organic molecules. Chapter 2 highlights the biological role of the telomere, telomerase and quadruplex DNA in cells. It discusses the inhibition of telomerase with small molecules that stabilise quadruplex DNA as a treatment for cancer. An overview of the development of structurally related telomerase inhibitors and recent X-ray crystallographic structural data with BSU6039 and BRACO-19 telomeric DNA is presented. Chapter 3 discusses the synthesis of fluorinated BSU6039 analogues for the investigation of the conformational effects of fluorine in 5-membered rings and its influence on binding with quadruplex DNA. These compounds have been successfully co-crystallised with telomeric DNA and their relative stabilisation of telomeric DNA has been assessed. The latter half of this chapter focuses on the co-crystal structures between (S,S)- and (R,R)-144 with Oxytricha nova telomeric DNA, discussing the key differences between the two stereoisomers. Chapter 4 details the synthesis of fluorinated BRACO-19 analogues. The syntheses of such fluorinated analogues were achieved through a base mediated coupling between 3,6-diaminoacridone and an α-fluorinated-β-amino ester. The α-fluorinated-β-amino ester was synthesised through a deoxyfluorination-mediated approach, using the stereochemistry of natural amino acids. Chapter 5 describes the stereo- and regio- selectivity of deoxyfluorination reactions with dipeptides bearing the β-amino alcohol functionality. Understanding this selectivity enabled the development of a method towards α-fluorination of tertiary amides. The application of this fluorination method with an orthogonally protected tertiary amide is described. Chapter 1 Synthesis & properties of fluorinated compounds 1.1 - A brief history Alchemists of the 17th century first harnessed the power of fluoric acids by treating fluorspar (CaF2) with strong acid to liberate hydrofluoric acid vapour. This vapour was used to etch glass for decorative purposes. However, the isolation of elemental fluorine remained elusive until the end of the 19th century, with many of great experimentalists, including H. Davy and A.-M. Ampère, dedicating their efforts. The Frenchman Henri Moissan finally isolated elemental fluorine (F2) in 1886, by the electrolysis of KHF2/HF, an accomplishment that contributed to Moissan being awarded the Nobel Prize for Chemistry in 1906.1 During his attempts to isolate F2, Moissan would often experience the apparatus exploding into flames, as the liberated gas reacted with silicon grease. Moissan took this to conclude that he had in fact produced F2 and proceeded to inform the National Academy with the following statement: “One can indeed make various hypotheses on the nature of the liberated gas: the simplest would be we are in the presence of fluorine”.2 1.2 - Fluorination techniques Elemental fluorine will react with practically any organic material. In today’s chemical research environment, fluorine gas is still actively used by academic groups around the world, despite the requirement for rigorous safety considerations. Chambers and Sandford have developed a method employing microflow reactors to tame F2. With this approach, 1,3-diketones such as 1, can be mildly fluorinated with 10% F2 in N2, and Chapter 1 2 then subsequent cyclisation with hydrazine through to mono-fluorinated pyrazoles such as 3. These represent fluorinated structural motifs for medicinal chemistry applications (Scheme 1.01).3 This approach results in higher yields over direct fluorination of pyrazoles with electrophilic fluorinating reagents or with elemental fluorine. OO F2 in N2 (10%)MeCN FO O NH2NH2MeCN NHNF Scheme 1.01. Synthesis of pyrazoles employing a fluorination flow method. Despite controlling the reactivity of F2 with flow reactors, elemental fluorine remains difficult to handle and therefore in the last half-century, there have been many novel fluorination methods reported in the literature.4,5 1.3 - Electrophilic fluorinating reagents In the 1960’s Derek Barton explored the development of the electrophilic reagent CF3OF. Fluoroxy-trifluoromethane can fluorinate activated enolates, but the reagent is toxic and difficult to use.6 Various second generation electrophilic fluorinating reagents initially inspired by the power of CF3OF have now become commonplace in organic chemistry (Figure 1.01). N+F BF4- N SFSO OF3C O OCF3 N+N+ ClF 2 BF4- Figure 1.01. Common electrophilic fluorinating reagents. The first of these reagents were the N-fluoropyridinium salts 4 of Umemoto.7 The reactivity of these salts can be tuned through modification of the pyridinium ring with electron-withdrawing and -donating functional groups. The mode of fluorination is thought to proceed by a single electron transfer (SET) mechanism. 1 2 3 - 77% 4 5 6 Chapter 1 3 The sulfone amide, N-fluorobis[(trifluoromethyl)sulfonyl]imide 5 (NFSI), developed by Desmarteau, is among the most powerful electrophilic fluorinating reagent developed.8 Differding and co-workers demonstrated the first enantioselective electrophilic fluorination with the chiral N-fluoro sultam 7 NFSI.9 Fluorination of cyclic enolates 8 with sulfone 7, enabled a modest enantioselectivities of up to 70% ee (Scheme 1.02). NSO O FO COOEt Base O COOEtF Scheme 1.02. The first example of an enantioselective fluorination reaction. More recently, however, there have been remarkable developments in enantioselective fluorination as demonstrated by MacMillan,10 Jørgenson11 and Barbas.12 In their separate approaches they have demonstrated how organocatalysts in the presence of NFSI can achieve α-fluorination of various aldehydes, with enantioselectivities of up to 99% ee in the case of MacMillan’s system (Scheme 1.03). HO NFSI 20 mol% 12THF/iPrOH, -10 ºC, 12 hFollowed by NaBH4 OHFHO NFSI 1 mol% 15MeOtBu, rt, 2 h HFHO NFSI 1 eq 18DMF, 4 ºC, 2 h F O HO MacMillanJørgensonBarbas N NHO PhHN OTMSCF3F3C CF3CF3NHN tBuO Ph 121518 H Scheme 1.03. Organocatalytic approaches to the α-fluorination of aldehydes with NFSI (5). 8 7 9 13 16 11 90%, 99% ee 14 >90 %, 97% ee 17 74%, 96% ee 10 Chapter 1 4 In these examples, the formation of a chiral enamine intermediate results in a diastereoselective interaction with the fluorinating reagent (Scheme 1.04). RNNH (S)O Ph HOR NN+ OPhR NN+ OPhNN+ OPhFR R (S) HOF R NN+ OPhN SS F OPhOO PhO HSi-faceRe-face NS SFOPhO OPh OVia H H OOClCl Scheme 1.04. Proposed catalytic cycle for fluorination where R = aryl or alkyl. The counterion for intermediates 20-22 is dichloroacetic acid. Transfer of fluorine from the bottom face (the Re-face) of the chiral intermediate is blocked by the bulky phenyl moiety (Scheme 1.04, red clash). Thus, NFSI approaches the Si-face of the enaminium intermediate 21 to furnish the α-fluoro iminium 22, which generates α-fluoro aldehyde 23 following hydrolysis. Jørgenson used the bulky proline derivative 15 at very low catalyst loading (1 mol%), much lower than that of MacMillan and Barbas. The MacMillan and Barbas catalysts both suffer from higher catalyst loadings, however in the system developed by MacMillan, the yields and level of enantiocontrol were significantly improved (Scheme 1.03). 21 20 19 23 12 22 N + N R H Me Me O Bn H Re-face Si-face N S S Ph O O Ph O O F N SS Ph O O Ph O O F Chapter 1 5 The most widely used electrophilic fluorinating reagent is the 1,4-diazabicyclo[2.2.2]octane based reagent, selectfluor (6) (Figure 1.01).13 Selectfluor, developed by Banks, is a highly stabile, versatile and reliable fluorinating reagent that has found wide application in synthesis.13,14 A recent publication in Science by Toste and co-workers demonstrated the applicability of selectfluor (6) in the enantioselective fluorocyclisation of dihydropyran based substrates such as 24 to fluorinated spiro-oxazoline, such as 26. (Scheme 1.05).15 O NHO R O NO RF5 mol% Cat*1.25 eq Selectfluor1.1 eq proton spongeC6H5F, -20 ºC, 24 hOO P OOHR'R' iPr iPriPriPr iPriPrCat* - Scheme 1.05. Phase transfer catalyst with selectfluor to induce fluorocyclisation. R = H, alkyl, aryl, halide In this particular transformation, Toste et al. formed a chiral cationic fluorinating reagent in situ, with phosphoric acid catalyst 25. Selectfluor is insoluble in non-polar solvents and in this protocol the formation of the chiral selectfluor salt with 25 enables the reaction to proceed in a non-polar solvent and thus increases the substrate scope. Conducting the reaction in a polar medium was found to lead to multiple unidentifiable products. Ritter and co-workers have pioneered an approach to the fluorination of aromatic heterocycles through the use of the selectfluor PF6- salt 28.16 They demonstrated that treatment of aryl tributylstannanes (27) with silver triflate and selectfluor resulted in the isolation of aryl fluorides such as 29 within 20 min at room temperature (Scheme 1.06). 24 26 25 6 Chapter 1 6 tBu3Sn R N+N+ ClF 2 PF6- 1.2 eq2.0 eq AgOTfacetone, 23 ºC, 20 min F R Scheme 1.06. Fluorodestannylation through a silver-mediated approach. This proceeds through a mechanistically complex process that preliminary experimental data demonstrates the involvement of two silver cations in the catalytically active species. Trans-metalation of the aryl stannane followed by oxidative insertion of the fluorine to form a speculative [(Aryl-Ag-F)Ag]n+ species is though to be key to the process. Reductive elimination to form the Aryl-fluorine bond completes the process. Gouverneur et al. have recently demonstrated this fluorodestannylation in the preparation of fluorine-18 labelled heterocycles with modest radiochemical yields of up to 18% for positron emission tomography (PET).17 1.4 - Nucleophilic fluorinating reagents Fluoride ion is a hard nucleophile with high solvation energy. The use of polar coordinating solvents greatly diminishes its nucleophilicity.18,19 Typical nucleophilic sources of fluoride are: NaF, KF and CsF.20–22 These alkali metal fluorides can be used to displace activated alcohols, such as 30, to furnish the corresponding fluorinated derivatives 31 (Scheme 1.07).4 R OMs KF/18F R F/18F Scheme 1.07. General nucleophilic displacement of an activated alcohol with KF. Radiolabelled K18F can be used for the generation of PET radiotracers. R = H, alkyl, aryl. Many nucleophilic fluorinating reagents are commercially available and offer an array of options for chemists to introduce one more fluorine atoms into organic molecules. 30 31 R=alkyl or aryl 27 29 where R=aryl, alkyl, halide or H 28 Chapter 1 7 One of the most powerful fluorinating reagents is sulfur tetrafluoride (SF4), which can convert alcohols, ketones and carboxylic acids through to their respective mono-, di- and tri-fluoro analogues.4 The development of safer and easier to use sulfur-based fluorination regents was led by Middleton at DuPont in the 1970’s.23 Middleton developed diethylaminosulfur trifluoride 32 (DAST 32, Figure 1.02), which represented a convenient nucleophilic fluorinating reagent (Scheme 1.08) for the deoxyfluorination of alcohols, with wide applications in synthesis. NSF3 NSF3MeO OMe NSF3ON FF F FF F NF FF Cl NN F F+N SF2 +N SF2OBF4- BF4- Figure 1.02. Nucleophilic fluorinating reagents. DAST 32 exists in equilibrium between the neutral form 32 and the charged activated species 32' with a fluoride counter ion. Nucleophilic attack of the lone pair of an alcohol (40), to the sulfur of the activated DAST 32' results in intermediate 41 (Scheme 1.08). This activates the alcohol ready for nucleophilic displacement by the liberated fluoride to yield the fluorinated product 42 (Scheme 1.08). DAST 32 does, however, suffer from thermal sensitivity and can decompose exothermically at temperatures above 90 ºC. 32 DAST 33 Deoxo-Fluor® 34 MOST 35 XtalFluor-E® 36 XtalFluor-M® 37 Ishikawa’s Reagent 38 Yarovenko’s Reagent 39 DFI Chapter 1 8 Et2N SF3Et2N SF2F-+ RHO R' OR' RSF NEt2F H+F- R'RF S OEt2N F HF+ +H Scheme 1.08. General mechanism for fluorination of alcohols (where R = aryl and alkyl) with DAST 32 or Deoxo-Fluor® 33. Deoxo-Fluor® 33 and MOST 34 were developed as thermally stable alternatives to DAST 32 (Figure 1.02).23 Sulfur trifluoride based fluorinating reagents can fluorinate a variety of substrates, however, they are mostly employed for the fluorination of alcohols. DAST 32 can also convert ketones into their respective gem-difluoro analogues.24 Ishikawa’s reagent 37, Yarovenko’s reagent 38 and the 2,2-difluoro-1,3-dimethylimidazolidine (DFI) 39 are nucleophilic fluorinating reagents that are less commonly employed.4 This lack of application can be explained through their propensity to form side products. For example, DFI 39 can be used to fluorinate alcohols and ketones such as 43 and 46, however, this often results in elimination to give vinyl and fluoro vinyl side products such as 44 and 48 (Scheme 1.09).25 O F FF+OHOH F F5 5+NN F F Scheme 1.09. Fluorination of alcohols and ketones with DFI 39. 42 40 43 44 45 46 47 48 49 50 32' 41 32 39 Chapter 1 9 The tetrafluoroborate salts, XtalFluorE ® 35 and M® 36 are recent additions to the list of nucleophilic fluorinating regents (Figure 1.02).26,27 These air and thermally stable solid salts have been shown to have similar selectivity to DAST 32 or Deoxo-Fluor® 33 with few elimination products observed. However, these reagents require promoter additives such as DBU or HF.Et3N to enable effective conversion to gem-difluoro 51, mono-fluoro 53 and acyl fluoride 55 derivatives (Scheme 1.10).27 These reagents, unlike DAST 32 and Deoxo-Fluor® 33, do not have free fluoride, therefore the promoters are important in generating fluoride for nucleophilic attack at the activated carbon centers. O O OH +N SF2 BF4-DBU or Et3N.3HF FFFPhOHPhPh O FPh Scheme 1.10. Fluorination with XtalFluorE ® (35). 1.5 - Trifluoromethylation of aromatic and heteroaromatic rings Recent high profile publications have appeared in the literature addressing the development of effective aromatic trifluoromethylation. The most notable of these are from the laboratories of Buchwald,28 Baran29 and MacMillan,30 appearing in Science, PNAS and Nature respectively. Buchwald et al., developed a low catalyst loading Pd phosphine based trifluoromethylation procedure that displayed wide substrate tolerance (Scheme 1.11). This method proceeds via a classic Pd catalysed mechanism with reductive elimination furnishing the Ar-CF3 product, such as 57, in yields >70%. 55 54 53 51 50 52 35 Chapter 1 10 Cl 3 mol% [(allyl)PdCl]2 9 mol% PhosphineTESCF3 (2eq), KF (2 eq)dioxane, 130 ºCup to 20 h CF3 OMeMeO PCy2OiPriPrO OiPrnBu nBuN NaSO2CF3 (6.0 eq)tBuOOH (10 eq)DCM:H2O (2.5:1)rt 3 N2Me O Me O CF3N N CF3SO2Cl (4 eq)2 mol% photocatalystK2HPO4, MeCN, 23 ºC26-W light N N N NNNCl2Ru NNMe Me Me Me CF3 BuchwaldBaranMacMillan Scheme 1.11. Recent advances in the trifluoromethylation of aromatic rings. The Buchwald palladium catalysed approach provides an efficient procedure for the trifluoromethylation of various building blocks for later assembly into structurally important compounds. The contributions by Baran29 and MacMillan30 employ radical trifluoromethylation approaches, which proceed at ambient temperature (Scheme 1.11). Baran et al., have shown that the Langlois reagent (NaSO2CF3),31 along with tert-butyl hydroperoxide can achieve radical trifluoromethylation of unactivated heteroaromatics such as 59 (Scheme 1.11). This occurs in a non-selective fashion under biphasic conditions with trifluoromethylation of 59 occurring at the 2/3- positions (Scheme 1.11). Extending this approach to the trifluoromethylation of medicinally active agents enabled the synthesis of trifluoromethylated analogues 64-66 with some selectivity for the more nucleophilic carbon center (Figure 1.03). N N NNO ONHNO HO N N25 NH CF3CF3F3C Figure 1.03. Selected products using the Baran’s trifluoromethylation method. 57 58 56 60 59 62 61 63 65 (78%) CF3-Caffeine 66 (50%), 4:1 (C5/C2) CF3 - Varenicline 64 (49%) CF3-dihydroqunine Chapter 1 11 In a subsequent development, MacMillan et al. have demonstrated the applicability of Ru(phen)3 photocatalyst 63 for the initiation of radical trifluoromethylation with various unactivated heteroaromatics such as 61 (Scheme 1.11).30 This approach also works at ambient temperature and is highly suitable to late stage modifications of pharmacological relevant compounds such as 67-72 (Figure 1.04). The authors report both selective and promiscuous trifluoromethylation with this approach (Figure 1.04). While both outcomes are beneficial, the promiscuous trifluoromethylation is significantly more powerful, enabling the synthesis of various trifluoromethylated regioisomers for the systematic studies of their effects in vitro/vivo. N NHO O 4 3 HN O NO3 2MeO OH OMeOMeO 2 5H O OMe OMeCF3 CF3 F3C CF3 CF3 O CF3SelectivePromiscious Figure 1.04. Typical products of the MacMillan’s trifluoromethylation procedure. Both of the procedures from the MacMillan and Baran laboratories represent significant advances in the trifluoromethylation of aromatic compounds with wide ranging applications. The application of these techniques to rapid profiling of trifluoromethylated drugs in a medicinal chemistry context will be a key aspect of their future application.32,33 67 (92%) CF3-methyluracil 68 (94%) CF3-Aricept 69 (85% ) CF3-flavone 72 (78%), 2:1 (C3/C4) CF3-lidocaine 71 (78%), 1.4:1 (C3/C2) CF3-ibuprofen 70 (82%), 5:1 (C2/C5) CF3-methylvanillin Chapter 1 12 1.6 - Properties of fluorine in organic molecules Organofluorine compounds have found a multitude of applications from functional materials34 to pharmaceuticals,35–37 and agrochemicals.38 They form a significant proportion of pharmaceutical compounds and agrochemicals, with organofluorine compounds representing approximately 20% of products on the market.39 Fluorine has a small van der Waals radius (1.47 Å)40 and is often regarded as an isostere for hydrogen or oxygen, as its steric influence is intermediate between these atoms.41 Due to its high electronegativity, fluorine holds onto valence electrons tightly, and as a result, it has a high ionisation energy potential of 1681 kJ mol−1 (cf. chlorine 1251 kJ mol-1). Fluorine has never been observed as a ‘fluoronium’ ion (F+), unlike other halogens.19 The C–F bond is the strongest single bond in organic chemistry with a dissociation energy of 115.0 kcal mol-1, significantly higher than that of other carbon halogen bonds (cf. C–Br - 69 kcal mol-1, C–Cl - 79 kcal mol-1).42 This can be attributed to the electronegativity and relative size of the fluorine atom, with the carbon donating electrons to the fluorine such that the carbon becomes δ+ and the fluorine δ-. Therefore, it can be assumed that the single bond has some electrostatic character in addition to its covalent nature.42 Chapter 1 13 1.7 - Acidity and basicity The strong C–F dipole alters the pKa or pKb of neighbouring functional groups (Table 1.1).43,44 For example, sequential introductions of fluorine on the α-carbon of acetic acid results in an increase of acidity of the carboxylic acid, with the pKa going from 4.76 in acetic acid to 0.52 for trifluoroacetic acid (Table 1.01). This can be explained by a greater electropositive nature of the α-carbon supporting the negative charge of the carboxylate through the inductive effect. Carboxylic acids pKa Alcohols pKa Bases pKb CH3CO2H 4.76 CH3CH2OH 15.9 CH3CH2NH2 10.7 CH2FCO2H 2.59 CF3CH2OH 12.4 CH2FCH2NH2 8.97 CHF2CO2H 1.34 (CH3)3OH 19.2 CHF2CH2NH2 7.52 CF3CO2H 0.52 (CF3)3OH 5.1 CF3CH2NH2 5.70 Table 1.01. pKa and pKb of non-fluorinated and fluorinated acids, alcohols and bases. This effect on acidity or basicity can affect a drugs ability to be transported through biological membranes such as the blood brain barrier. Thus, fluorine incorporation can be used to tailor properties to improve pharmacokinetic profile of a molecule.36 1.8 - Fluorine in drug metabolism Within medicinal chemistry, the strong C–F bond has been used to limit the susceptibility of pharmaceuticals to P450 oxidation in the liver.35 This has been widely applied to the modification of aryl rings for example, in Ezetimibe 74 (Scheme 1.12). Lead optimisation of 73 resulted in an overall increase in potency and stability in vivo.45 Chapter 1 14 NOMeO OMeNOHO FF Lead optimisationOxidation Blocked Oxidation Blocked Scheme 1.12. Metabolic stability introduced through lead optimisation resulting in greater potency. 1.9 - Fluorine based suicide inhibitors One of the earliest and most successful applications of fluorine within medicinal chemistry was the antineoplastic drug 5-fluorouracil (5-FU) 75 (Scheme 1.13).46 5-FU acts by inhibiting the enzyme thymidylate synthase disrupting the biosynthesis of nucleotides for DNA synthesis resulting in cell death (Figure 1.13). HN NHO O F HN NHO O FSEnzFAH Nucleotide biosynthesis interrupted- Cell DeathCovalent inhibition of Thymidylate synthaseThymidylate Synthase Scheme 1.13. 5-FU inhibition of thymidylate synthase through the irreversible covalent blocking of the active site. Inhibition of the enzyme occurs during the methylation stage. To release the methylated nucleotide, the fluorine must leave as ‘F+’, which cannot happen, and so the 5-FU becomes covalently bound in the active site, thus leaving the enzyme inactive (Scheme 1.13).47 76 75 74 73 F S FAH Enz Chapter 1 15 1.10 - 19F NMR probes in chemical biology 1.10.1 - General applications The use of fluorine NMR has been particularly important in chemical biology for example, during the elucidation of the metabolism of fluoro-containing pharmaceuticals,48,49 or following the biosynthesis of fluorinated natural products in vitro.50,51 The power of 19F NMR signals in these systems is that the chemical shift of fluorinated compounds varies over a large range (~ 300 ppm) and their spectral complexities are lower compared with 1H NMR spectra. 19F NMR has been extended to the study of small molecule-protein or DNA binding studies.52,53 The synthesis of proteins containing fluorine modified amino acids, offers a route to monitor the binding of small molecules to an active site by observing the chemical shift changes in the 19F NMR spectrum.54 1.10.2 - rRNA conformation probes The application of 19F NMR in the study of small molecule probes for determining RNA tertiary structure was reported in 2010 by Micouin and co-workers.55 Based on previous work, they demonstrated that small diaminocyclopentanes 77 (Figure 1.05) were able to bind to various transfer RNAs without altering the global structure of specific tRNAs.56,57 H2N NH2F H2N NH2F Figure 1.05. Small fluorinated NMR probes for RNA structure. Monitoring the chemical shift differences of the two diastereoisomers of 77 upon binding to tRNA sequences demonstrated the formation of a diastereomeric pair with tRNA (Figure 1.06). The relative chemical shift was found to be tRNA dependent. In each case this shift was dependent on the tertiary structure of the tRNA, as confirmed by variable temperature-NMR (VT-NMR).55 (R)-77 (S)-77 Chapter 1 16 Δδ 1 Δδ 2 19 F NMR Figure 1.06. Changes in the 19F NMR chemical shift of rac-77 in response to RNA addition. This technique allows a potential method to investigate the topological changes in tRNA structure, however the future assessment, in a complex biological environment, may be complicated by non-specific binding of 77. 1.10.3 - Membrane transport kinetics 19F NMR has also been used to explore the transport of the fluorinated glucose analogue α/β-78 across erythrocyte (red blood cells) membranes (Figure 1.07).58 OO FF F OHOHFFF Outside Cell Inside Cell Glut1OO FF F OHOHFFF Glut1 β−efflux α−effluxErythrocyte cell membraneSlow Fast Figure 1.07. Trifluoro-glucose analogues α/β-78 as a 19F NMR probes to study efflux enzymes through 2D EXSY experiments. rac-77 + RNA Type 1 rac-77 + RNA Type 2 rac-77 19F NMR α-78 β-78 α-78' β-78' Chapter 1 17 In the study by O’Hagan et al., membrane transport kinetics were assessed by 2D 19F Exchange Correlation Spectroscopy NMR (EXSY NMR).59 The 19F NMR signals for both of the intra- and extra-cellular populations of the trifluoroglucose anomers’ α/β-78 were distinguishable. In the EXSY experiment, the intensity of the cross peaks formed from the retained polarisation from internal to external populations could be correlated to the transmembrane rate constant. It was found that this method generated rate constants that were similar to glucose itself.60 1.11 - Conformational effects of fluorine 1.11.1 - The gauche effect The incorporation of a carbon-fluorine bond into organic molecules can have an influence on molecular conformation. A well-documented case is the gauche effect in 1,2-difluoroethane 79 (Scheme 1.14). It has been shown that the gauche conformers are favoured over the anti-conformer by approximately 0.8 kcal mol-1.19 HF HF HH HH FF HHFH HF HHFavoured by ~ 0.8 kcal mol-1Disfavoured high energy state HH ClCl HH ClH HCl HH~ 50:50HH BrBr HH BrH HBr HH Scheme 1.14. The gauche effect in 1,2-dihaloethanes. This is in contrast with 1,2-dichloroethane 80, which does not show conformational preference and with 1,2-dibromoethane 81, which has an anti-conformer preference. In the case of 1,2-difluoroethane 79, the gauche-conformer is stabilised through hyperconjugative interactions (Figure 1.08). g- 80 81 79 g+ ap Chapter 1 18 FHσCH σ*CF Figure 1.08. Hyperconjugation in 1,2-difluoroethane 79. This donation of electron density from the σCH orbital antiperiplanar to the σ*CF orbital stabilises the gauche conformer.61,62 The σ*CF orbital is low in energy and a good acceptor of electron density compared to the other halogens.63 In the case of 1,2-difluoroethane 79, there are two hyperconjugative interactions with both C–F bonds orientating antiperiplanar to C–H bonds. The gauche effect is also observed in 1,2-fluorohydrins and in other systems whereby the fluorine has a vicinal arrangement with various electron withdrawing groups.64 Stereoelectronic effects are important in the molecular preorganization of linear fluoroalkanes containing multiple contiguous vicinal fluorine atoms.65 The synthesis of the all-syn tetra-, penta- and hexa-fluoroalkanes, 82-8467–69 respectively, results in a defined conformation arising from gauche effect contributions and most significantly 1,3-fluorine-fluorine repulsion of ~3.0 kcal mol-1 (Figure 1.09).66 R F F F F R R F F F F F R R F F F F F F R1,3-repulsion Figure 1.09. The all syn-vicinal fluorinated alkane motifs. This is particularly striking in the hexa-fluoroalkane 84, where a helical structure is observed in both the solid and solution states (Figure 1.10).69 79 82 83 84 Chapter 1 19 Figure 1.10. X-ray crystal structure of the all-syn hexa-fluoro alkane demonstrating the helicity induced by 1,3-fluorine-fluorine repulsions. In a systematic study with various fluorohydrin stereoisomers 87-90 of the HIV-1 protease inhibitor Indinavir 85, the gauche effect has been used to stabilise the preferred extended binding conformation (Figure 1.11).70 N N OH PhO HN OHNHtBuON N N OH PhO HN OHNHtBuONN N OH PhO HN OHNHtBuON N N OH PhO HN OHNHtBuONN N OH PhO HN OHNHtBuON N N OH PhO HN OHNHtBuONFF FF Indinavir Ki (nM) - 1.9 epi-Indinavir Ki (nM) - 160syn,syn Ki (nM) - 2.0anti,anti Ki (nM) - 27 anti,syn Ki (nM) - 5900 syn,anti Ki (nM) - 20 Figure 1.11. Fluorinated Indinavir 85 and epi-Indinavir 86 demonstrating a conformational preference for target specificity. The optimal conformation is achieved in the syn,syn isomer 87, which has the same efficacy as Indinavir 85. This conformation is not accommodated in the anti,anti isomer 89, which has a lower activity (10 fold decrease). There is a more significant effect in the fluorinated stereoisomers of the less active epi-Indinavir 86. Interestingly the syn,anti isomer 88 reverses the loss in activity in 86 over 85 by 8 fold, whereas, the anti,syn 90 shows a dramatic decrease in activity into the millimolar range (Figure 1.11).70 85 85 86 87 88 89 90 Chapter 1 20 1.11.2 - The α-fluoroamide effect α-Fluoro amides have a clear conformational preference as a result of the C–F bond dipole (1.85 Debye in fluoromethane).71 In α-fluoro-amides there is a strong preference for the C–F bond to lie antiperiplanar to the dipole of the amide carbonyl (Scheme 1.15).72–74 This is particularly striking for 91 and 92 where there is a stabilisation of 7.5 and 8.0 kcal mol-1 respectively for the anti conformation.72 NO RHMeH F NO RHF HMeanti cis Scheme 1.15. The α-fluoroamide effect with anti-preference relative to the amide carbonyl. Molecular orbital (MO) calculations by O’Hagan et al., demonstrated this preference in N-methyl-2-fluoropropionamide.72 The rotational energy profile for this system is shown in Figure 1.12. There is a clear energy well of ~8.0 kcal mol-1 when the C–F bond and amide carbonyl are anti to each other.72 Figure 1.12. Rotational energy profile for α-fluoroacetamide 92 showing a conformational well for the trans- (anti) conformation. 7.5 kcal mol-1 - 91, R = H 8.0 kcal mol-1 - 92, R = Me NO MeHMeHF NO MeHMeH FNO MeHHF Me Chapter 1 21 This strong preference arises from at least three stabilizing factors: primarily the relaxation of the dipoles from the C–F and the C=O bonds, such that their combined vectors cancel (Figure 1.13, A); a favourable C–F⋯H-N electrostatic interaction (Figure 1.13, B); and finally, a stabilising orbital interaction between the amide π*C(O)N orbital and the F np orbital (Figure 1.13, C).75 N HFHOπ*C(O)N npO NH MeFHH MeHO NH MeFHH Figure 1.13. Stabilising factors for the anti-conformation in α-fluoroamides. A study of the Cambridge Structural Database (CSD) by Seebach et al., reveals that this conformational preference is reflected across a range of open chain compounds in the solid state with typical F–C–C–O dihedral angles of 147º<φ<190º (Figure 1.14).76 Figure 1.14. Dihedral angle prevalence in α-fluoro amides. The notable exception to this is the synclinal example with a dihedral angle ~50-60º. Seebach et al., also observed this deviation from the anticipated antiperiplanar orientation in a study of fluorinated β-amino acid conformational effects on peptide conformation.76 They demonstrated through NMR analysis of the 4JHF coupling constants that the C–F bond orientates perpendicular to the amide plane in the tridecapeptide 93 if it can not adopt an anti orientation, thus destroying the helicity of the peptide (Figure 1.15, A).77 In this example, it was reasoned that the global energy 92 92 A B C O N F H H H Me p π ∗ C(O)N 92 Chapter 1 22 minimum of the peptide was enough to ‘override’ the local stabilisation of the C–F bond, which was forced into its next favoured conformation (Figure 1.15, B). This energy preference is apparent in the rotational energy diagram (Figure 1.12) with a plateau around 60º.72 NH NHOFO PhNHH2N O ONH FRH Figure 1.15. The fluorinated tridecapeptide has a favoured solution structure with the C–F bond orientating 90º relative to the carbonyl of the amide. 1.11.3 - The Charge-dipole effect If the C–F bond is located proximal to a positive charged species, a conformational preference arises resulting from a strong electrostatic interaction.78 This favours a conformation that might otherwise be unfavourable. For example, in 2-fluoroethylammonium 94, the C–F and C–NH3+ orientate preferentially in a gauche rather than an anti alignment (Scheme 1.16). F+H3N HF HNH3+HH HH FNH3+HHFH HNH3+HHCharge - Dipole interaction Favoured by ~ 5.8 kcal mol-1Disfavoured high energy state Scheme 1.16. Charge-dipole effect in 2-fluoroethylammonium 94. 93 94 A B A B g – g + Chapter 1 23 DFT calculations on 2-fluoroethylammonium 94 have demonstrated that the gauche conformation is preferred by about 5.8 kcal mol-1 (Figure 1.16), thus, exerting significant stabilisation for that particular conformation.78,79 This conformational preference is also observed for protonated alcohol 95 and the N-fluoroethylpyridinium ion 96 (Figure 1.16). N+ F O+ F N FH HH H H Figure 1.16. Conformational preference as a result of the charge-dipole interaction. In studies by Lankin and Synder80 the charge-dipole effect was shown to result in a strong axial orientation of the C–F bond in 3-fluoropiperidinium rings 97-99, where the stabilisation is ~5 kcal mol-1. The axial preference was confirmed by NMR and DFT calculations, in addition to X-ray crystallographic analysis of selected compounds. The 3,5-difluoropiperidinium 100 also demonstrated an axial preference, clearly overcoming the repulsive 1,3-diaxial fluorine-fluorine interaction (Figure 1.17). N+ FFMeMeN+MeMeN+MeMe PhPhFFN+HH F Figure 1.17. Charge-dipole effect in fluorinated piperidinium systems with their respective DFT (Becke3 LYP/6–311G(d,p)) calculated energies. Following from these studies, Gooseman et al. demonstrated the charge-dipole effect in 4-and 5-membered rings, 101 and 102 respectively (Figure 1.18), which have no particular conformational preference in their non-fluorinated forms unlike six membered rings.81,82 97 5.4 kcal mol-1 98 4.0 kcal mol-1 99 3.7 kcal mol-1 100 8.9 kcal mol-1 94 5.8 kcal mol-1 95 7.2 kcal mol-1 96 3.7 kcal mol-1 Chapter 1 24 FN+N+HH HHN+ FH H N+ FHH Figure 1.18. Charge-dipole effect in small and large fluorinated nitrogen containing heterocycles. This was confirmed through DFT and X-ray crystallographic analysis, both of which are in strong agreement with that observed in six-membered rings. The effect in larger rings, such as 103 (Figure 1.18), was also found to favour the axial orientation of the C–F bond.81 It follows that the incorporation of the C–F bond into protonated nitrogen heterocycles offers a way to influence the conformation in a non-covalent manner, which could have an application in drug discovery. 1.12 - Applications of the charge-dipole effect 1.12.1 - Organocatalysts The charge-dipole effect has been explored in both medicinal and organocatalytic arenas.83 In a study by Gilmour et al.84 an exocyclic C–F bond situated pendant to a pyrrolidine ring (104) was shown to induce stereocontrol in the epoxidation of α,β-unsaturated aldehydes such as 105a/b (Scheme 1.17) with excellent enantiomeric control, up to 96% ee in 107b with good to excellent diastereomeric excess. NH F PhPh R CHO N+ F PhPhR O CHORN+ F PhPhR N+ PhR F H2O2 Scheme 1.17. Organocatalytic epoxidation of α,β-unsaturated aldehydes with the C–F charge-dipole effect driving enantiocontrol. 107a R=Ph 92% 82:18 dr, 96% ee 107b R=iPr 90% >95:<5 dr, 90% ee 106a/b 104 -10 mol% 105a/b 106a/b 101 102 103 9.2 kcal mol-1 Chapter 1 25 In this reaction, the intermediate C–F-iminium dihedral angle in 106 was shown to be 58º, thus directing the phenyl moiety to shield one face of the π-system (3.8/4.3 kcal mol-1 gauche stabilisation for the E-/Z- geometry in 106a/b). This therefore delivers the nucleophile to the opposite Si-face. The non-fluorinated catalyst 108 proceeds with 23% ee, albeit under different conditions (Scheme 1.18).85 N PhHPh O CHOPhN F PhPhHOPh (i) or (ii) Scheme 1.18. Demonstration of the fluorine charge dipole effect and its application in organocatalysis. 1.12.2 - Biological exploitation of the C–F bond The charge-dipole effect has also been used to probe the molecular binding conformation of GABA 109 to GABAA/C receptors and GABA metabolising enzymes (Figure 1.19). F CO2-+H3NF CO2-+H3NCO2-+H3N Figure 1.19. Fluoro-GABA analogues. Deniau et al. demonstrated that GABA transaminase could discriminate between (R)-and (S)-3-fluoroGABA 110.86 The preferred transaminase binding conformation is readily adopted by (S)-110, however, this conformation is unfavoured in (R)-110 (Figure 1.20, A).86 Both (R)- and (S)-110 exhibited a similar efficacy for the GABAA receptor87 (Figure 1.20, B), however (R)-110 was more active at the GABAC receptor (Figure 1.20, C).88 (R)-110 (S)-110 109 (i) - 22%, 23 ee [%] (ii) - 92%, 96 ee [%] Condition set (i) Condition set (ii) 108 - 30 mol% Hexane, TBHP 140 h, rt 106 -10 mol% CHCl3, H2O2 3 h, rt 105a 107a NH F PhPhNH HPhPh Chapter 1 26 CH2CO2-F HNH3+HH FH CH2CO2-NH3+HHH-O2CH2C FNH3+HH F-O2CH2C HNH3+HH CH2CO2-H FNH3+HH HF CH2CO2-NH3+HHB CATransaminasebinding conformation GABAA receptorbindingconformation F CO2-+H3N F CO2-+H3NGABAc receptorbinding conformation Figure 1.20. Preferred binding conformations of the fluorinated GABA analogues 110 to the GABAA receptor and GABA transaminase. The charge-dipole effect governs these conformations leading to the observed activities. These studies demonstrate that the fluorine charge-dipole effect is significant in a biological context and can offer information on the favoured binding mode of small molecules to large proteins (Figure 1.21). -OOC H HHH HH NH3+ -OOC HH HH HH NH3+GABAC Binding GABAABinding Figure 1.21. Binding conformation to GABAA/C receptors of GABA based on fluorinated probes. It also demonstrated the applicability of this approach in the development of probes to study binding conformations of bioactives in complex molecular environments; thus providing information for the development of new inhibitors to target these receptors.89 (R)-110 (S)-110 109 109 A B C Chapter 1 27 In a comprehensive study by Hunter et al., the α-fluoro amide, charge-dipole and fluorine-fluorine gauche effects were collectively considered for a conformational study on α,β-difluoro-γ-amino amides 111 and 112 (Figure 1.22).90 N OPhO O F F N OOH N OPhO O F F N OOH 'Bent'Linear Figure 1.22. Conformational differences between syn- and anti-isomers of the α,β-difluoro-γ-amino amides 111 and 112. In this study, it was possible to demonstrate through NMR solution studies and by X-ray crystallography that both 111 and 112 exhibited an antiperiplanar orientation of the C–F bond relative to the amide carbonyl and that the vicinal C–F bonds are gauche to each other. The β-C–F and γ-C–N bonds are also gauche in the solid and solution state. The accumulative effect of these interactions results in two distinct and predictable conformations for 111 and 112.90 From these compounds, the vicinally 2,3-difluorinated GABA analogues 113-116 were prepared and assessed for their respective activity on GABAC receptors.91 It was found that the syn-isomers exhibited potent activity over the anti-isomers (Figure 1.23). O-OFF+H3NO-OFF+H3NO-OFF+H3N O-OFF+H3NAgonist - EC50 155 µM Antagonist - IC50 128 µM 15.4 % activtyat 100 µM 14.8% activtyat 100 µM Figure 1.23. The four stereoisomers of 2,3-difluorinated GABA analogues and their activity for the GABAC receptors. In the syn-isomer series, 113 acts as an agonist, whereas 114 was an antagonist. Molecular docking studies of the energy-minimised structures of 113 demonstrate it could adopt the correct binding conformation to accommodate the key contacts important for GABAC binding, thus supporting its agonist response. Docking of 114 also demonstrated the ability of the carboxylate and amine groups to orientate in the correct manner, however, this also highlighted additional steric interactions with the 113 114 115 116 111 112 Chapter 1 28 receptor that may explain its antagonist activity. The predicted conformation of 113 corresponds with the binding conformation proposed earlier for GABA 109 (Figure 1.21) to the GABAC receptor. Interestingly 113 and 114 did not exhibit any GABAA activity whereas 115 and 116 did.91 From these studies, it is clear that the charge-dipole and other conformational effects can be used to influence the conformation of otherwise flexible organic molecules. This predictable preorganization has enabled detailed studies on the binding of GABA to its receptors. It is envisaged that this information will enable a better understanding of how to design tailored inhibitors for these receptors. 1.13 - Synthesis of fluorinated β-amino acids 1.13.1 - General methods The most common way of incorporating a fluorine substituent into an amino acid involves deoxyfluorination reactions with DAST 32. Takei et al., have employed DAST 32 in the synthesis of α-fluoro-β2,3-homophenylalanine 118 in studies exploring the inhibition of chromotrypsin (Scheme 1.19, A).92 F OOPh NHBn2N OOFBocOH OOPh NHBoc DASTHO OONBn2 DAST Scheme 1.19. The application of DAST 32 to the synthesis of α-fluorinated-β-amino acids. DAST 32 was first used in this context in a synthesis of benzyl protected α-fluoro-β-alanine 120 by Shomek (Scheme 1.19, B).93 This method has been widely used by Seebach et al. for studies on the influence of α-fluorinated-β-amino acids on β-amino peptide structures and exploring their metabolic stability.94 In example B (Figure 1.19), the fluorination proceeds through an aziridinium intermediate, a mechanism, which will be discussed in more detail in Chapter 4.95 117 119 118 120 A B 32 32 Chapter 1 29 These methods rely on stereospecific reactions manipulating the existing stereochemistry of the starting material. Other methods have used stereoselective fluorination reactions to generate new stereogenic centers with fluorine. This approach is discussed in sections 1.13.2 and 1.13.3. 1.13.2 - Evans oxazolidine approach In 2008, Abell et al. demonstrated an Evans oxazolidinone based strategy for the construction of mono-fluorinated β2,2-amino acids bearing a quaternary stereogenic center (Scheme 1.20).96 In this approach they were able to fluorinate the oxazolidine derivative of cyclohexyl and phenyl propanoic acids, 120a/b respectively, by deprotonation and treatment with N-fluorobenzenesulfonamide. This gave 121a/b in high diastereomeric excess and in good yield (>90% de, 79%). R NO OOPh R NO OOPhF R NO OOPhFOBn1) LDA, NFBS 1) TiCl4, DiPEABnOCH2Cl Scheme 1.20. Evans auxiliary approach to the synthesis of mono-fluorinated quaternary centers The alkylation of 121a/b was achieved with benzyl chloromethyl ether in the presence of base and TiCl4 (Scheme 1.20). Further transformations with 122 enabled the synthesis of 124a/b in >95% diastereomeric excess (Scheme 1.21). This was the first synthesis of a mono-fluorinated β2,2 substituted amino acid and offered a versatile approach to the synthesis of diverse fluorinated amino acids (Scheme 1.21). R NO OOPhFOBn R OOFN3 R OOFNHFmoc3 Steps 3 Steps Scheme 1.21. Functionalisation of the fluorinated Evans auxiliaries 122a/b mono-fluorinated β2,2 amino acids. 122a - 70% 122b - 69% 121a - 79%, 12:1 dr 121b - 75%, 12:1 dr 120a - R = Ph 120b - R = C6H12 122a/b 124a - R = Ph 124b - R = C6H12 123a/b Chapter 1 30 1.13.3 - Davies’ lithium amide approach In a modification of the Davies’97 diastereoselective addition of lithium amides 126 to α,β-unsaturated esters 125, Duggan and co-workers have demonstrated that quenching the intermediate enolates 127 with NFSI 5 results in the isolation of α-fluorinated β2,3-amino acids 128 with high diastereoselectivity (Scheme 1.22).98 OtBuOR Ph NLi PhTHF, -78 ºC2) NFSI1) OtBuOR FN PhPhOtBuOLiR NPh PhVia Scheme 1.22. Chiral lithium amide addition to α,β-unsaturated esters in the tandem approach to α-fluorinated β2,3 amino acids. R = alkyl, aryl. Various α,β-unsaturated esters 125 could be employed with the chiral lithium amide 126 to provide synthetically useful β-amino acids 128 through this tandem approach. In a proof of concept study, a stepwise addition-fluorination protocol via 129a/b demonstrated that the isolated β-amino acids 128a/b were of low diastereomeric excess (Figure 1.23). Thus, demonstrating the power of their tandem approach (Figure 1.22). OtBuOR Ph NLi PhTHF, -78 ºC2) NH4Cl1) OtBuOR N PhPh OtBuOR N PhPh F1) LDA, -78 ºC2) NFSI Scheme 1.23. Stepwise approach with poor diastereomeric control of the fluorination step. 125a (R = Ph) 125b (R = TBDMSO(CH2)2) 129a 129b 128a' - quant, anti:syn = 61:39 128b' - 93% anti:syn = 72:28 128 50% to quant. anti:syn 80:20 to 99:1 125 a-f 127 126 Chapter 1 31 Using the tandem approach, the orthogonally protected α-fluoro-β2,3-lysine derivative 130 starting from 128c was synthesised (Scheme 1.24). OtBuOFN PhPhTBDMSO OtBuOFFMocHNBocHNFGI Steps Scheme 1.24. Further functionalisation through to an orthogonally protected α-fluoro-β2,3-lysine for synthesis applications. This method offers a route for generating an array of structurally diverse fluorinated β-amino acids for the incorporation into medicinally relevant compounds or for the use as chemical probes. The synthesis of fluorinated amino acids has received a significant level of attention as indicated by the number of reviews and publications in the field.99 Structural and biological applications of the conformational influence of the C–F bond continue to appear, and the role of the C–F bond in this context may become more widely recognised. 1.14 - Conclusion It is clear, from the discussion in this chapter, that the strategic incorporation of a C–F bond into an organic molecule can result in a conformational bias. This has enabled the elucidation of binding modes of small molecules to enzymes and large receptors. The following chapter will address the development of quadruplex DNA stabilizing ligands and will offer a biological context for investigating the conformational influence of the C–F bond in a drug-DNA complex. 128c 130 Chapter 2 Telomeres, telomerase and quadruplex DNA 2.1 - The 2009 Nobel prize to telomeres In 2009, the Nobel Prize for Medicine was awarded to Elizabeth H. Blackburn, Carol W. Greider and Jack W. Szostak for their contributions towards our understanding of the telomere and telomerase.100 Their contributions have budded what is now a very active research area. Blackburn and Szostak were first to demonstrate that the telomeric sequence was conserved across a range of distantly related organisms and that it was fundamental in cell biology.101 Following this, Blackburn and Greider provided evidence for the enzyme telomerase (Christmas Day 1984), which is responsible for elongation of the telomere.102 They later demonstrated that telomerase required an RNA component to be catalytically active.103 These discoveries underpin many of the biochemical and genetic studies focused around the telomeres and its role in cancer, ageing, and inheritable diseases, amongst many others. The following will provide a general review of the telomere and inhibitors of telomerase. 2.2 - Telomeres and telomerase One of the major limitations of replication in eukaryotic cells is for the replication machinery to completely copy to the 3' end of DNA. This is known as the end replication problem.104 Thus, with each round of replication, a short section of DNA is cleaved from the chromosome. DNA sequences coded at the end of chromosomes are not faithfully copied and this jeopardises genomic stability following each round of cell division. To guard against this, chromosomes have a protective ending known as the telomere (Scheme 2.01). Chapter 2 33 Cell Division Replicative Senescence p53 induced apoptosis Chromosome Telomere Eroded Telomere T e l o m e r a s e ≡ Telomere extension Damage Response Telomeres Visualised Maintained Telomere Scheme 2.01. Generalised scheme for telomere erosion and elongation at the chromosome end. Knitted chromosomes copyright Science Museum/Science and Society Picture Library 2012. The telomere is a non-coding sequence of DNA consisting of tandem hexanucleotide repeats dTTAGGG, with an approximate length of 5-8 kilobase pairs.105 Therefore, each round of cell division results in the loss of approximately 50-200 bases from this non-coding DNA.106 The cell will replicate until the length of the telomere reaches a critical length, its Hayflick limit,107 before entering a state of replicative senescence (G0 state), which then initiates p53-mediated cell death.108 Therefore, the average telomeric length in a colony of cells will decrease over time until it reaches its Hayflick length, unless the cells are replenished by their respective stem cell, which generally have longer, maintained telomeric DNA lengths. The telomeres in stem and germ-line cells are maintained by an RNA-dependent DNA polymerase.102,109 This reverse transcriptase adds the TTAGGG nucleotides to the 3' end of the telomere following each round of cell division. The RNA component (hTR) is critical for activity and anneals the 3' end of the telomere,110,111 templating it into the active site of the catalytic subunit, hTERT. The hTR component is found expressed in somatic cells, however hTERT expression is silenced through various transcriptional regulators,106 thus somatic cells contain no constitutively active telomerase.112–114 Both hTR and hTERT components are expressed in stem and germ-line cells, thus active telomerase can maintain the telomeric length in these cells.113 Chapter 2 34 2.3 - Telomerase and cancer The nature of the telomere in somatic cells, with its critical length, raises a question around its role in cancer. Cancer cells typically replicate without control and so it would be natural to assume that the average telomeric DNA length in cancer is critically low. However, this is not observed and the telomere lengths in cancers are maintained. This telomeric maintenance can be attributed to the loss of the transcriptional control of hTERT, with 85% of cancers expressing this catalytic domain.115,116 With both hTR and hTERT found in cancer cells, an active telomerase maintains the telomere length and these cells subsequently avoid activation of cell death pathways. Targeting the action of telomerase offers a unique method of cancer therapy with the bulk of non-cancerous cells not affected by this approach.108 Even though the telomere is a section of non-coding DNA, its maintenance and function is controlled by a multitude of protein interactions.117 The majority of the telomere adheres to normal topological parameters of duplex DNA, with the exception of approximately 200 nucleobases at the 3' end.118–120 These nucleobases form a single stranded overhang and it is this section of the telomere where the most interesting biological interactions occur and structures form. 2.4 - Shelterin complex at the telomere Single-stranded DNA (ssDNA) in other regions of the genome are quickly recognised as damaged by the cell.121–123 Detection of ssDNA results in the activation of repair mechanisms or causes the cell to activate apoptotic pathways. To circumvent this for the telomere, there are various proteins that interact with the 3' single stranded overhang and protect it.117,124 These core protective proteins form the complex known as shelterin and specifically bind the TTAGGG repeat of the telomere. Shelterin is comprised of eight core protein units, TRF1/2, TIN2, TPP1 and POT1 (Figure 2.01), with five domains available to recognize the telomeric DNA sequence, making it highly specific for telomeric DNA. Knockout studies of the shelterin component POT1, resulted in cell-cycle arrest and chromosomal end-to-end fusion as a direct result of shelterin complex disruption.125 For telomerase to elongate the telomere following cell division, Chapter 2 35 the telomere must be linear and free of any tertiary structure to enable the association of the catalytic and recognition domains of telomerase.126 TIN2 TRF1 TRF2 T P P 1 POT1 5' 5' 3' 3' Thursday, 1 March 2012 Figure 2.01. A simplified representation of shelterin proteins associating with telomeric DNA. 2.5 - Self-assembly of guanosine In the 1960’s, it was established that solutions of guanosine-5'-monophosphate (5'-GMP) base 131 formed gel-like solutions upon standing (Scheme 2.02).127 These gel solutions resulted from the formation of square planar quartets of hydrogen-bonded ionophores, which rapidly formed in the presence of monovalent cations. NH1 2N345 6N9N7 O NH2O OHHOOP-O-O O 8 Scheme 2.02. The self-assembly of guanosine nucleotides 131 into quartet structures in the presence of mono-valent counter ions. Quartet representation extracted from PDB 1I1H (Ref.-128) with the image created using PyMol.129 Distances are in Å. These monovalent cations, primarily Na+ and K+, increase the stability of G-quartets by favourable electrostatic interactions with the O6 atoms of each base. 5'-GMP 131 can form four hydrogen bonds per base by utilizing its Hoogsteen edge along with the 131 Hoogsteen face Watson-Crick face M+ Chapter 2 36 typical Watson-Crick edge as shown in 131 (Scheme 2.02). The hydrogen bond acceptors N7 and O6 on the Hoogsteen edge form hydrogen bonds with the hydrogen bond donors, N1 and N2, on the Watson-Crick edge. Therefore, four individual 5'-GMP 131 nucleotides can form a G-quartet with eight hydrogen bonds in total. 130 Multiple G-quartets can self-assemble into a G-quadruplex structure through favourable π-π stacking interactions and stabilisation brought about by a bipyramidal antiprismatic bound cation, primarily K+.131 A further thermodynamic driving force for this assembly is found from the displacement of water that forms unfavourable interactions with each G-quartet. G-quadruplex self-assembly motifs can be between 8 to 30 nm long with each G-quartet rotated round the central axis (Figure 2.02).132–134 G G G G K 0.34 nm 8-30 nm Thursday, 1 March 2012 Figure 2.02. Representation of guanine tetrads stacking to form a self-assembled G-quadruplex. 2.6 - Quadruplex DNA folding and topology It is likely that guanine-rich telomeric sequences in the human genome will form quadruplexes in vivo.135-137 There are two main types of quadruplex structures that can form with guanine-rich sequences, intramolecular (unimolecular) and intermolecular (bimolecular) forms. Sequences of the type Go·Xp·Go form intermolecular structures and Xn·Go·Xp·Go·Xp·Go·Xp·Go·Xn form intramolecular folds (where Xn is any number non-guanine nucleotide, Go is any number of guanine nucleotides and Xp is any number non-guanine nucleotides involved in loop formation).135 The topology that one sequence may form over another is firstly governed by the linking nucleotide length and by the presence of different monovalent cations. 138 Go·Xp·Go sequences, such as the Oxytricha nova telomeric sequence, G4T4G4, will form bimolecular quadruplexes Chapter 2 37 where two sequences associate to form the quadruplex structure (Figure 2.03). In these structures, Xp will orientate relative to the guanine nucleotides to accommodate their assembly into a quadruplex (Figure 2.03). a) b) c) d) External Loop Diagonal Loop (D) e) Lateral Loop (L) External Loop (E) LateralLoop Figure 10 Schematic diagrams showing potential linking loop arrangements and resulting topologies for bimolecular quadruplexes. (a) Lateral loops connecting anti- parallel strands arranged on the same face; (b) lateral loops connecting anti- parallel strands arranged on either end. (c) Diagonal loops connecting opposite anti-parallel strands. (d) External chain reversal loops connecting parallel strands together; (e) Diagonal loops connecting opposite anti-parallel strands 18 Chapter 1 a) b) c) d) External Loop Diagonal Loop (D) e) Lateral Loop (L) External Loop (E) LateralLoop Figure 10 Schematic diagrams showing potential linking loop arrangements and resulting topologies for bimolecular quadruplexes. (a) Lateral loops connecting anti- parallel strands arranged on the same face; (b) lateral loops connecting anti- parallel strands arranged on either end. (c) Diagonal loops connecting opposite anti-parallel strands. (d) External chain reversal loops connecting parallel strands together; (e) Diagonal loops connecting opposite anti-parallel strands 18 Chapter 1 In the forming of dimeric quadruplexes we first need two strands to be linked together. If we link two of the strands by a lateral loop the strands will be constrained to be anti-parallel and require an inversion of one of the paired bases and the glycosidic torsion angle, in order to retain an appropriate hydrogen bonding arrangement. This is usually accommodated in one of the two ways, either each strand has a run of alternating syn, anti glycosidic angles for the guanine bases, or one strand has exclusively syn and the other strand Syn Anti O3' O5' Counter Ion Backbone a) b) c) d) e) 1 4 32 Figure 9 Schematic diagrams showing potential strand polarity and related glycosidic torsional angles for intermoleculer quadruplexes. (a) Parallel with all anti glycosidic torsional angles; (b) parallel all syn glycosidic torsional angles; (c– e) alternating strands arrangements with the corresponding mixed syn anti relationships 16 Chapter 1 Lateral op External Loop Diagonal Loop Key: Anti yn Backbone 3′ 5′ a) b) c) d) e) Figure 2.03. Arrangement of gua in s d linking nucl otides in bimolecular quadruplexes.139 The Xp nucleotides can either form external, diagonal or lateral links relative to the quadruplex core, which can result in many complex and diverse quadruplex structures.130,140 In addition to this, the glycosidic bond in the DNA monomers will also reverse from the favoured anti to syn bo d angl , in an attempt to accommodate the formation of hydrogen bonded quartets.138 In the O. nova sequence, G4T4G4, the thymine linking nucleotides associate in a diagonal manner (Figure 2.03, e) relative to the guanine nucleotides (Figure 2.04).141–143 Chapter 2 38 Figure 2.04. Crystal structure of the O. nova G4T4G4 bimolecular quadruplex with two aspect views. Image created with PyMOL129 using PDB file 1JPQ.141 However, the mutant O. nova sequence, G3T4G4 forms a mismatched quadruplex, which results in a distinct fold compared to G4T4G4, with the thymine nucleotides associating in a lateral and diagonal manner.144 A small sequence change can result in a significant structural modification, highlighting the complexity of quadruplex folds in the solid state, and clearly the dynamics in solution. A further example of this structural complexity is exemplified by reducing the number of nucleotides available for cross-linking. For example, the G4T3G4 sequence forms a bimolecular structure with lateral thymine linkages.145 In this case, two quadruplex structures in the unit cell are observed with both the head-to-head (Figure 2.03, a) and head-to-tail (Figure 2.03, b) quadruplexes formed (Figure 2.05). Chapter 2 39 Figure 2.05. The two quadruplex DNA structures formed from the G4T3G4 sequence. Images created with PDB codes 2AVH & 2AVJ respectively using the PyMOL package.129 Unimolecular quadruplexes also form different topological structures (Figure 2.06).138 Such unimolecular quadruplexes are biologically relevant and are likely to form in vivo at the ssDNA ends of the telomere, in addition to regions throughout the genome that have a high guanine content.146 1.3.1 Loop Length, Sequence, and StabilityAn additional constraint for the folding topologies for bimolecular and inter-molecular quadruplexes relates to the number (length) of the linking nucleo-tides. Short linker lengths, two or less, will prevent diagonal loops fromforming due to the distance to be spanned across G tetrad. However, the short b) DiagonalLoop O3'O5' c) O5' O3' ExternalLoop O3' O5' d) e) O3' O5' f) O3' O5' a) LLL b) LDL c) EEE d) LLE e) EDL f ) EDEg) ELLh) LDE i) No D** or **D a) LateralLoop O5' O3' 1 4 32 Figure 11 Schematic diagrams showing potential linking loop arrangements and resultingtopologies for intramolecular quadruplexes. (a) Lateral loops connecting anti-parallel strands arranged on the same face; (b) mixed lateral and diagonalloops connecting anti-parallel strands. (c) External chain reversal loops con-necting parallel strands together. (d) Mixed external and lateral loops con-necting strands; (e) Mixed external, diagonal, and lateral loops connectingstrands together; (f) Mixed external and diagonal loops connecting strandstogether 19Fundamentals of Quadruplex Structures 1.3.1 Loop Length, Sequence, and Stability An additional constraint for the folding topologies for bimolecular and inter- molecular quadruplexes relates to the number (length) of the linking nucleo- tides. Short linker lengths, two or less, will prevent diagonal loops from forming due to the distance to be spanned across G tetrad. However, the short b) Diagonal Loop O3' O5' c) O5' O3' External Loop O3' O5' d) e) O3' O5' f) O3' O5' a) LLL b) LDL c) EEE d) LLE e) EDL f ) EDE g) ELL h) LDE i) No D** or **D a) Lateral Loop O5' O3' 1 4 32 Figure 11 Schematic diagrams showing potential linking loop arrangements and resulting topologies for intramolecular quadruplexes. (a) Lateral loops connecting anti- parallel strands arranged on the same face; (b) mixed lateral and diagonal loops connecting anti-parallel strands. (c) External chain reversal loops con- necting parallel strands together. (d) Mixed external and lateral loops con- necting strands; (e) Mixed external, diagonal, and lateral loops connecting strands together; (f) Mixed external and diagonal loops connecting strands together 19Fundamentals of Quadruplex Structures a) b) c) d) e) f) O3′ O3′ O3′ O3′ O3′ 3′ O5′ O5′ O5′ O5′ O5′ O5′ teral Loop i gonal Loop External Loop Figure 2.06. Arrangement of guanines and linking nucleotides in unimolecular quadruplexes.139 Chapter 2 40 In addition to the various ways that the linking polynucleotides can arrange relative to the guanine core, there is also a structural effect from the presence of different monovalent cations.147 For example, Na+ ion coordinates within the plane of the G-quartet while the larger K+ ion coordinates offset between two G-quartets. As a result, as shown by solution NMR (Figure 2.07, A)148 and X-ray149 crystallographic studies (Figure 2.07, B), structures for the human telomeric sequence d(TG3(T2AG3)3) adopt very distinct folds. In the Na+ NMR structure the T2A linking nucleotides associate in a lateral and diagonal fashion relative to the guanine tetrads (Figure 2.07, A). In contrast, the linking nucleotides assemble in a diagonal manner (Figure 2.06, C) in the crystal structure with K+, resulting in a distinctive propeller like structure (Figure 2.07, B). Figure 2.07. Unimolecular quadruplex DNA with different counterions A – d[AG3(T2AG3)3] sequence with Na+ counter ion & B – with K+ counter ion. Images created from PDB codes 143D (Ref.-148) and 1KF1 (Ref.-149) with PyMol.129 Structural studies of the human telomeric DNA sequences are starting to provide a strong foundation in the understanding of how quadruplex DNA topologies arise. This information is important for the rational design of drugs that may interact with these sequences.140 A B Chapter 2 41 2.7 - Telomerase inhibition Zahler and co-workers observed that increasing the K+ concentration induced an inhibitory effect on telomerase action in vivo.150 They concluded that the increased [K+] stabilises quadruplex folds in vivo and in vitro, and that this may act as a negative feedback mechanism for the maintenance of telomere length. This inhibition of telomerase can be attributed to the disruption of the association between the telomere and telomerase, with the interaction critically dependent on the 3' ssDNA overhang being free and linear. This free topology is required such that the hTERT RNA subunit can anneal with the telomere, thus stable quadruplex DNA structures will inhibit the action of telomerase as they will not uncoil.151 2.8 - Assessing telomerase inhibition and quadruplex stability The development of quadruplex DNA stabilising ligands has called for the development of standardised techniques to assess their action in vitro and in vivo. The following section is a brief overview of the current techniques used.152 Circularly polarised light spectroscopy is used to assess the topological arrangement of guanine-rich sequences. Circular dichroism (CD) is a powerful tool to probe and monitor structural changes of quadruplex DNA upon altering the nature and concentration of counter ions.153,154 CD can also be used to monitor structural changes upon binding of drugs that interact with quadruplex DNA.152 Other spectroscopic techniques such as NMR155,156 and X-ray crystallography157 play significant roles in the structural evaluation of how quadruplex stabilising ligands bind to quadruplex DNA. These techniques taken together provide information on the interactions between drugs and quadruplex DNA and can be used for the systematic development and rational design of new agents. Chapter 2 42 Fluorescence resonance energy transfer (FRET) assays are used to quantify the level of stabilisation that specific ligands provide to the quadruplex folds.158 For this technique, modified quadruplex DNA sequences are attached with fluorescent donor and acceptor chromophores at the 3' and 5' ends (Scheme 2.03). Commonly used chromophores are 6-carboxyfluorescein 132 (FAM) and 6-carboxytetramethyl rhodamine 133 (TAMRA) (Scheme 2.03). When the donor and acceptor are in close contact, such as when the quadruplex is folded, the excitation of the donor ligand results in FRET transfer to the acceptor which results in emission of a different wavelength (Scheme 2.03).159 An increase in temperature will unfold the quadruplex structure to its linear form, thus increasing the distance between the two chromophores, resulting in poor energy transfer with subsequent fluorescence decrease. The addition of a stabilising ligand will cause an increase in the melting temperature of the quadruplex DNA, and the increased stabilisation of the fold results in a higher melting temperature. ON N+OOHO OHOHO OO OHOHO Scheme 2.03. General overview of FRET-based analysis of quadruplex stability. Telomerase inhibition assays provide specific values for the inhibition of telomerase activity in vitro. The protocol for this is known as TRAP - Telomeric repeat amplification protocol.160 The use of a fluorescent primer for the telomerase enzyme enables the evaluation of telomerase inhibition by quadruplex-stabilising drugs. This is Donor Acceptor 132 133 488 nm 520 nm 488 nm 488 nm Chapter 2 43 a widely employed protocol in the literature providing telIC50/telEC50 values for a range of ligands. Good experimental correlation between the FRET and the TRAP assays are observed and generally a higher melting temperature results in better inhibition of the telomerase enzyme. However, a direct correlation and prediction of one value based on the other cannot easily be made. 2.9 - Quadruplex DNA stabilising ligands Various quadruplex DNA stabilising ligands have been identified.106,161 These include natural products along with various other motifs arising from structure based design strategies.140,161 A concurrent theme arises with these stabilising ligands, as many are based on a large polyaromatic core with various peripheral cationic side chain substituents. These side chain substituents often interact with the negatively charged phosphate grooves and polynucleotide linkages, while the flat polyaromatic cores capitalise on favourable π-π stacking with a free G-tetrad at the end of the quadruplex DNA fold. It has been observed that larger aromatic cores give rise to greater selectivity of quadruplex DNA over that of duplex DNA, due to the greater π-bonding surface available from the quadruplex.135 A critical evaluation of quadruplex DNA ligands is in their classical cytotoxicity. For an effective quadruplex DNA stabilising ligand, the level of cell cytotoxicity should be at least 10 times higher than the telEC50. Classical anti-cancer/proliferative drugs work on being cytotoxic to the cell, this is not the case for quadruplex DNA ligands. Typically the use of a drug that inhibits telomerase through the stabilisation of quadruplex structures will not demonstrate any signs of activity until multiple rounds of cell division. This would lead to the gradual erosion of the telomere, initiating expression of proteins associated with short or damaged telomeres, resulting in characteristics protein foci at the chromosomal ends. The use of other anti-cancer agents would accelerate this process by targeting the cancerous cells from two approaches, with each agent working synergistically with one another. Chapter 2 44 2.9.1 - Natural products and analogues The natural product telomestatin 134 is currently the most efficient telomerase inhibitor known, with an telIC50 of 5 nM and is often used as the benchmark for assessment of other telomerase inhibitors (Figure 2.08).162,163 Telomestatin 134 was isolated from Streptomyces anulatus and is comprised of seven oxazole rings with one dehydrothiazole ring.162 A subsequent total synthesis identified the natural configuration of telomestatin 134 as the (R)-enantiomer, in-line with the natural configuration of the amino acid cysteine.164 N OON N OONNO ONNO S N Figure 2.08. The natural product telomestatin, a potent inhibitor of telomerase. Telomestatin has a 70-fold binding selectivity for quadruplex DNA over that of duplex DNA.163 Quadruplex selectivity is critical for non-specific binding of the ligand to other regions of genomic DNA, with unspecific binding resulting in unwanted cytotoxic effects. Synthetic analogues of telomestatin 134, such as 135a/b (Figure 2.09) have been shown to be entirely quadruplex selective over duplex DNA, with the acetate analogue 135b demonstrating a 2 µM inhibition of telomerase (Figure 2.09).165,166 N OON N OOHNNO ONNO NHO N OON N OOHNNO ONNO NHO OAcAcO Figure 2.09. Analogues of the natural product telomestatin. 135a 134 telIC50 = 5 nM 135b telIC50 = 2 μM Chapter 2 45 2.9.2 - Porphyrin based inhibitors Porphyrin based inhibitors such as TMPyP4 136 have poor selectivity for quadruplex DNA, however CD and NMR based studies have shown that porphyrin 136 significantly stabilises quadruplex DNA (Figure 2.10).167,168 TMPyP4 136 was shown to inhibit telomerase with an telIC50 of 6.5 µM as determined by the TRAP assay and a ΔTm of 17 ºC from FRET analysis.169 Subsequent X-ray crystallographic studies of TMPyP4 136 with bimolecular quadruplex d(TAG3T2AG3) detailed an unusual major groove complexation between the ligand and the DNA rather than complexing to the G-tetrad face.170 This may offer an explanation for the poor duplex/quadruplex selectivity.167,171 NNH N HNN+ N+N+N+ Figure 2.10. Structure of the porphyrin based quadruplex DNA stabilizing ligand, TMPyP4 136. 136 ΔTm = 17.0 ºC telIC50 = 6.5 μM Chapter 2 46 2.9.3 - Quinacridine ligands Dibenzophenanthroline ligands such as 137a-c have been shown to have good stabilisation potential for quadruplex DNA. Assessing their stabilisation through FRET analysis, found that 137a and 137b stabilized the human telomeric sequence by +19.7 ºC and +12.8 ºC respectively (Figure 2.11).158 In TRAP assays, 137a and 137b have also showed an inhibitory effect on telomerase with telIC50 values of 0.028 µM and 0.5 µM respectively. These values correlate with the FRET-based observations with the higher stabilising ligand resulting in a greater inhibition of telomerase. The cyclic derivative 137c was shown to bind by both π-stacking and groove intercalation with quadruplex DNA with a ΔTm of 28 ºC and telIC50 of 0.13 µM (Figure 2.11).172 NNHN HNNN N N NHHNN NN N NHNHNN HNHNHN NH Figure 2.11. Structures of the quinacridine based quadruplex DNA stabilizing ligands. 137b - ΔTm = 12.8 ºC telIC50 = 0.5 μM 137a - ΔTm = 19.7 ºC telIC50 = 0.028 μM 137c - ΔTm = 28.0 ºC telIC50 = 0.13 μM Chapter 2 47 2.9.4 - Anthraquinone and fluorenone ligands Quadruplex DNA-stabilising ligands developed by Neidle, Hurley and co-workers have led to a plethora of papers detailing the improvements and achievements of designing new ligands based on their early work.173 Initially, bisamidoanthraquinone ligands 138a-h had attractive telEC50 values for telomerase inhibition (Figure 2.12). 32 14 109 8 765OONH NHO O NR2R2N ONH NHO O NR2R2N N+ N N+NN O N+ ONR2 = N N Figure 2.12. Structures and telEC50 values of the anthraquinone and fluorenone based ligands. Systematic studies on the substitution patterns of anthraquinones 138a-h through the 1,4- 1,8-, 2,6- and 2,7- regioisomers identified the 2,7-regioisomer 138a as the most potent, with an inhibitory value of 2.0 µM. The side groups are protonated at physiological pH with the exception of morpholine 138c. The detrimental effect of neutral ligands can be observed in the telEC50 value for 138c (>>50 µM), thus this ligand does not appear to form strong interactions with quadruplex DNA folds. Interestingly, the generation of the N+−Me salts 138b/d/f/h resulted in an increase in telEC50 values. This may be due to the removal of hydrogen bonds between the ligands and the quadruplex DNA. However, the metabolic cytotoxic effects of 138a-h were problematic. This lead to the development of fluorenone analogues 139a-h, which do not suffer the same metabolic fate as 138a-h. Fluorenones 139a-h demonstrated telIC50 values between 8-12 µM and a decrease in metabolic related cytotoxicity compared to the anthraquinones 138a-h (Figure 2.12). 138 139 138a - 2.0 μM 139a - N/T 138b - 16.0 μM 139b - N/T 138c - >50.0 μM 139c - >50.0 μM 138d - 16.5 μM 139d - 27.3 μM 138e - 3.1 μM 139e - 9.0 μM 138f - 7.8 μM 139f - 21.0 μM 138g - 4.7 μM 139g - 16.2 μM 138h - 4.3 μM 139h - 15.5 μM Chapter 2 48 2.9.5 - Acridone and di- and tri-substituted acridine ligands Subsequent development of acridone-based ligands 140a-h,174 demonstrated that the incorporation of a nitrogen in the aromatic core results in an increased interaction with quadruplex DNA (Figure 2.13). Following these observations, the acridine-based compounds 141a-h175,176 were developed in an attempt to arrange the central protonated acridine nitrogen over the negatively polarised central quadruplex core (Figure 2.13). However, the acridine series 141a-h had low selectivity (1.3:1) for quadruplex DNA over duplex DNA.177 Of the series, BSU6039 141a, which exhibited a telEC50 value of 5.2 µM, was chosen for subsequent studies. 2 3 41 N+10 9 5 678NHO NH O NR2R2N N N NNNN NNNHNHO NH O NR2R2N O NR2 = H Figure 2.13. Structure and telEC50 values for the acridone and acridine 3,6-disubstituted ligands. Molecular modelling with BSU6039 141a suggested that substitution at the 9-position would result in further interactions with a third phosphate groove in the quadruplex fold (Figure 2.14).177,178 The tri-substituted series 142a-f was synthesised and it was found that 142a (BRACO-19) exhibited the most favourable inhibitor characteristics. BRACO-19, with the 4-(dimethylamino)aniline substituent in the 9-position, was 31-fold more selective for quadruplex over duplex DNA and also showed a 44-fold increase in inhibition of telomerase (telEC50 0.115 µM) when compared to BSU6039.174 140b - 4.3 μM 141b - 8.2 μM 140a - 8.1 μM 141a - 5.2 μM 140d - 5.7 μM 141d - 5.8 μM 140c - 5.9 μM 141c - 2.8 μM 140e - N/T 141e - 3.1 μM 140f - N/T 141f - >50μM 140g - 1.7 μM 141g - 1.3 μM 140h - 2.3 μM 141h - 2.6 μM 140 141 Chapter 2 49 2 3 41 N+109 5 678NHO NH O NN RHHN NH2 HN HNH2NNH2HN NMe2 HN HN FNMe2 Figure 2.14. 3,6,9-Trisubstituted acridine ligands. TelEC50 values are underneath for each 9-position substituent. 2.10 - BRACO-19 142a in vitro & in vivo studies BRACO-19 142a (Figure 2.15) has good in vitro efficacy against the prostate cancer cell line DU145.179 It was shown that after incubation over 7 days with sub-cytotoxic doses of BRACO-19 142a that half of the cells entered the G0 phase. After 21 days there was an increase in the expression of the apoptosis associated proteins p21 and p16. It was also noted that non-homologous end-joining (NHEJ) events were occurring during the metaphase of the cell cycle, a feature of dysfunctional telomeres.121 The data presented here suggests that BRACO-19 142a acts as a telomerase inhibitor through quadruplex DNA stabilization and also competes with telomeric binding proteins such as POT1. It has been demonstrated in vivo that BRACO-19 142a has efficacy towards xenographed uterine cancers in a murine model.180 142 142d 0.074 μM 142e 0.06 μM 143f 0.02 μM 142a 0.115 μM 142b 0.10 μM 142c 0.07 μM Chapter 2 50 NNHO NH O NN HN N OOOOHO OHOOHO HN OOO O OH Figure 2.15. Structure of BRACO-19 142a and Paclitaxel 143 (Taxol). Further studies with BRACO-19 142a and the established clinical anti-cancer agent paclitaxel 143 (Figure 2.15) have shown promising synergistic activities.151 The treatment of A431 human epithelial carcinoma with BRACO-19 142a results in an insignificant decrease in tumour size upon intraperitoneal dosage. However, dosing of BRACO-19 142a post paclitaxel 143 treatment in these carcinomas resulted in greater tumour shrinkage, with a shortening of the average telomere length, than with paclitaxel alone.181 This was the first proof of principal study of quadruplex DNA stabilisation as a method for anti-cancer therapy. 142a 143 O O OH O HO O H O O O OH NH O O O O Chapter 2 51 2.11 - X-ray crystallographic studies with acridine based ligands 2.11.1 - BSU6039 141a and O. nova DNA The crystal structure between BSU6039 141a and the bimolecular quadruplex DNA sequence G4T4G4 was solved to 1.75 Å and provided an insight into how the acridine based ligands 141a-h interact with quadruplex DNA (Figure 2.16).128 N+NHO NH O N+N+ HH H Figure 2.16. Crystal structure of BSU6039 141a bound to the bimolecular quadruplex fold from the Oxytricha nova sequence. A – BSU6039 structure, B – standard representation with BSU6039 binding in the top section and C – Surface representation to show binding cleft. Images created from PDB 1L1H using PyMol.129 In this structure, BSU6039 141a binds with one quadruplex fold with the thymine linking residues orientating in a diagonal manner across the top face of the quadruplex (Figure 2.16, B). This topology is the same for the native crystal structure (Figure 2.4) with potassium ions and generates two wide phosphate grooves with complementary narrower phosphate grooves along the sides of the quadruplex.141 The glycosidic angles of the guanine nucleotides in the sequence alternate syn-anti, such that the G-tetrad has a syn-syn-anti-anti glycosidic arrangement, which enables the hydrogen bonds from the B C A 141a Chapter 2 52 Hoogsteen and Watson-Crick faces to be accommodated (Figure 2.3). Between the G-tetrads, potassium ions are coordinated to the O6 carbonyls of the G-tetrads, thus further stabilizing the quadruplex structure (Figure 2.16, purple crosses). The diagonal orientation of the thymine loops generates a binding cleft into which BSU6039 141a can orientate and π-π stack with the top G-tetrad (Figure 2.16, C). In contrast to the native crystal structure (Figure 2.4), one of the thymine nucleotides twists out of the loop plane and interacts with the central nitrogen and one amide carbonyl of 141a, forming two hydrogen bonds (Figure 2.17). Figure 2.17. Top orientated view of the binding between BSU6039 141a and the quadruplex fold. Distances are in Å. Image generated from PDB file 1L1H using PyMOL.129 Another thymine nucleotide also forms a π-π interaction with the acridine, further stabilising the fold. In the phosphate grooves, there is a highly ordered water lattice. However, the charged pyrrolidino rings of 141a do not interact through salt bridges with the phosphate backbone and only on one side does the substituent form a hydrogen bond with a water molecule (Figure 2.17). This specific water molecule forms a hydrogen bond with the guanine tetrad. There are no other direct hydrogen bonds in the crystal structure between the quadruplex and the ligand. It has been rationalised that the substituents interact in an electrostatic manner as demonstrated in a range of crystal structures of 3,6-substituted acridines 141a-h with the O. nova sequence.182 Chapter 2 53 2.11.2 - BRACO-19 142a and human DNA A subsequent study on the mode of binding of the 3,6,9-substituted acridine ligand 142a has been published.183 The crystal structure of BRACO-19 bound to the bimolecular human telomeric G-quadruplex sequence, d(TAG3T2AG3T), was resolved to 2.5 Å by X-ray crystallography (Figure 2.18).183 Figure 2.18. BRACO-19 bimolecular quadruplex DNA co-complex X-ray crystal structure. A – BRACO-19 in a space filling representation sandwiched between two quadruplex folds, B – Structure of BRACO-19 for comparison and C – Surface representation of the bottom quadruplex clearly demonstrating the phosphate grooves. Images created from the PDB file 3CE5 using PyMol.129 This crystal structure and the binding between the quadruplex DNA and the ligand 142a are very different to that of the O. nova structure presented earlier. Each bimolecular quadruplex has propeller linkages, similar to those observed in the native G3(T2AG3)3 sequence (Figure 2.07, B), with an assembly of three planar stacked guanine tetrads. In contrast to the BSU6039 141a crystal structure, BRACO-19 142a is complexed between two quadruplex folds, forming π-π stacking interactions with the 3' end guanine tetrad of one quadruplex and TA nucleotides of the 5' end from another quadruplex. One linking thymine base is rotated such that it interacts through hydrogen bonding and water-salt bridges with the acridine core of BRACO-19 142a (Figure 2.19). A B N+NHO NH O N+N+ HH HHN N C 142a Chapter 2 54 Figure 2.19. Top orientated view of the binding between BRACO-19 142a and quadruplex DNA. Distances are in Å. This rotated thymine plays a critical role in the interaction and stabilisation of BRACO-19 142a with the quadruplex. The charged 3- and 6- substituents do not form hydrogen bonds with the negatively charged phosphates and interact in an electrostatic manner. The propeller TTA linkages between the guanine tracks generate size specific grooves that accommodate smaller substituents. This is in contrast to the O. nova sequence where the diagonal loops generate a cleft able to accommodate a variety of substituents.182 The overall mode of binding of BRACO-19 142a is typically mediated through hydrogen bonding with water to the quadruplex rather than through direct interactions. The synthesis of BRACO-19 142a analogues with longer alkyl substituents results in decreased stability.184 It is likely that the conformational flexibility in these analogues results in a less ordered binding ligand that will not orientate correctly to form hydrogen bonds, even with the water lattice in the phosphate grooves. The interaction between BRACO-19 142a and two quadruplex folds is a significant observation and may represent a realistic in vivo model. It is reasonable that quadruplex folds occur in a contiguous fashion with one or more quadruplex folds occurring at the ssDNA telomeric end. Therefore, the binding of BRACO-19 142a to one quadruplex in vivo may induce another quadruplex to ‘fold back’ and interact similar to that observed in the crystal structure. To date, however, there are no crystal structures of extended telomeric sequences, such as with d(AG3(T2AG3)n) where n≥7, that demonstrate two or Chapter 2 55 more quadruplexes per sequence.151 The generation of such structural data would be a significant advance in the current understanding of how quadruplex ligands are likely to interact with quadruplex DNA in vivo. 2.12 - Conclusions This chapter has provided a brief overview of the biological and structural aspects of telomeres in the cell. It has also discussed the implication of inhibiting enzymes involved in the maintenance of telomere length. The structures of several classes of small molecule inhibitors of telomerase that act through the stabilisation of quadruplex DNA in vivo have been described. Detailed assessment of their interaction in the solid state by X-ray crystallography has enabled the development of a generalised view of their mode of action. Chapter 3 Synthesis and evaluation of fluorinated BSU6039 analogues 3.1 - Introduction It is common practice in drug discovery to vary the substituents on the pharmacophore to explore structure activity relationships for a particular inhibitor. In the case of BRACO-19 142a and BSU6039 141a, it is clear that small ring substituents on the alkyl amino side chains led to a more efficacious inhibitor (Chapter 2.14).161 These rings are accommodated within the hydrophobic pocket formed by the DNA sequence. The size of the hydrophobic pocket is highly dependent on the individual sequence of the quadruplex DNA used for crystallisation, with the various linker nucleotides assuming different conformations in space (Chapter 2.6).157 In the case of the O. nova sequence, crystallography indicates that the thymine residues allow for a variety of alkyl substituents to be accommodated.182 As highlighted in Chapter 1.11, modification of small nitrogen containing heterocycles with β-fluorine substituents results in a ring conformation whereby the fluorine-carbon bond dipole moment interacts favourably with the N+–H bond dipole on the protonated nitrogen (Figure 3.01).19 N+ FHH N+HH F Figure 3.01. The charge-dipole effect in 5- and 6-membered rings. 101 97 Chapter 3 57 This is a dipole-dipole through space interaction. The possibility of incorporating fluorine in this manner into BSU6039 would enable an investigation into the mode of binding and the importance of ring conformation to the stability of quadruplex DNA. In addition, the C–F bond will be expected to lower the pKaH of the protonated amine and increase the acidity of the hydrogen for hydrogen bonding. In general, however, a smaller ΔpKa between the donor and acceptor results in a stronger hydrogen bond.185 Therefore, the aim of this research was to investigate the effect of substituting the pyrrolidino rings of BSU6039 141a with a C–F bond at the 3-position and to assess the structural influence by X-ray crystallography of co-crystals with O. nova bimolecular quadruplex DNA (Chapter 2.6/2.11). In order to investigate these effects, it was necessary to synthesise the enantiomers of the fluoro- and hydroxyl- analogues 144 and 145 of BSU6039 (Figure 3.02). These analogues would then be co-crystallised with the T4G4T4 O. nova quadruplex DNA sequence. NNH NH O NONF FNNH NH O NONF FNNH NH O NONHO OHNNH NH O NONHO OH Figure 3.02. 3-Fluoro- and 3-hydroxyl pyrrolidine analogues of BSU6039 141a. (R,R) & (S,S)-144 (R,R) & (S,S)-145 Chapter 3 58 3.2 - Synthesis of BSU6039 141a analogues To access the synthetic targets it was required to prepare the bis-chloro intermediate 146 from which 144 and 145 could be synthesised (Scheme 3.01). This intermediate can be accessed by treating proflavin 147 with 3-chloropropionyl chloride 148. NH2N NH2OClNNH NH O ClOCl Cl + Scheme 3.01. Retrosynthetic approach to 146. Treatment of diamine 147 with neat 3-chloropropionyl chloride (148) under forcing conditions generated the known bis-chloro substituted acridine 146 in an excellent yield of 90% (Scheme 3.02).175 N NH2H2N N NHNHO O ClCla) Scheme 3.02. Reagents and conditions: a) 3-Chloropropionyl chloride (neat), 140 ºC, 4 h, 90%. The solubility of bis-chloro 146 in common solvents was very low, thus it was particularly difficult to purify. In the event, bis-chloro 146 was often contaminated with acid chloride 148. This pungent smelling contaminant remained even after recrystallisation and washing with ethanol and then drying under high vacuum. However, the contamination (NMR analysis) was low and so the product was taken on without further purification. In order to access (R,R)-144, it was required to treat bis-chloro 146 with (3R)-fluoropyrrolidine 149 (Scheme 3.03). 147 146 146 148 147 Chapter 3 59 NNH NH O ClOCl NNH NH O NONF FNHF+ a) Scheme 3.03. Reagents and conditions: a) NaI, EtOH, 80 ºC, 3-5 h, 63%. Treating pyrrolidine (R)-149 with bis-chloro 146 in ethanol and with sodium iodide generated (R,R)-144. Purification of the product mixture by column chromatography required up to 5% triethylamine to obtain a pure sample of (R,R)-144 (63% yield). Repeating the procedure with (S)-149 enabled the isolation of (S,S)-144 also in a satisfactory yield of 65% (Scheme 3.04). NHF NNH NH O NONF Fa) Scheme 3.04. Reagents and conditions: a) 146, NaI, EtOH, 80 ºC, 3-5 h, 65%. In order to obtain (S,S)-145 and (R,R)-145, the reactions were repeated with (S)- or (R)-150. Products (S,S)-145 and (R,R)-145 could be isolated in satisfactory yields for both ligands (Scheme 3.05). NHHO NHHO NNH NH O NONHO OHNNH NH O NONHO OHa)a) Scheme 3.05. Reagents and conditions: a) 146, NaI, EtOH, 80 ºC, 3-5 h, 59%. 146 (R,R)-144 (R)-149 (S, S)-144 (S)-149 (S,S)-145 (S)-150 (R,R)-145 (R)-150 Chapter 3 60 3.3 - Characterisation of (S,S)- and (R,R)-144 The proton resonances of the 3-fluoropyrrolidine ring in (S,S)- and (R,R)-144 in the 1H NMR are well defined and can be assigned based on the correlations in the 1H–1H COSY spectrum (Figure 3.03). Figure 3.03. 1H-1H COSY (500 MHz, d6-DMSO/CD3OD) analysis of (S,S)-144. It is likely that the C–F bond does not exhibit a strong preference for a pseudo- axial or equatorial C–F bond conformation. Therefore, the unambiguous assignment of pseudo-axial and equatorial proton resonances cannot be made and have been designated as Ha and Hb for each resonance. Thus, starting from the distinctive 2JHF coupling constants, it was possible to assign the resonance at 5.28 as H-7' in 144 (Figure 3.03, blue line). The cross peak pattern from H-7' enabled the putative assignment of the resonances belonging to H-6'a/b and H-8'a/b. The relative chemical shift of the resonances at 3.15 and 2.84 ppm and their cross peak isolation (Figure 3.03, purple line) support the assignment of these as the H-6'a/b resonances. Therefore, the NNH1' NH1'' 2'' 3''O 4'' N5''2'3' O4'N5'6'7' 8' 9' 9'' 8'' 7''6''F F Chapter 3 61 resonances at 2.29 and 2.10 ppm can be assigned as H-8'a/b. The remaining resonances corresponding to H-9'a/b are confirmed through the clear cross peaks with H-8'a/b (Figure 3.03, orange lines). It is clear from the analysis of 144 that distinct resonances for each proton on the pyrrolidino ring can be observed. The 1H–1H COSY assignments were supported by a 1H–13C HSQC analysis (Figure 3.04). NNH1' NH1'' 2'' 3''O 4'' N5''2'3' O4'N5'6'7' 8' 9' 9'' 8'' 7''6''F F Figure 3.04. 1H–13C HSQC (500/125 MHz, d6-DMSO/CD3OD) analysis of (S,S)-144. Again, the distinctive 1JCF and 2JCF couplings in the 13C NMR spectrum, along with their relative chemical shifts enabled the assignment of C-7' (1JCF = 175 Hz). The resonances at 52.5 and 36.5 ppm in the 13C NMR spectrum were assigned to C-3' and C-4' respectively through cross peaks with the triplets at 3.02 and 2.75 ppm of the 1H NMR spectrum. Of the remaining resonances in the 13C NMR spectrum, two exhibited distinctive couplings of 22.8 and 22.3 Hz at 61.4 and 33.5 ppm respectively, corresponding to 2JCF through bond coupling. Distinguishing between these resonances Chapter 3 62 was possible by referring to the 1H–1H COSY spectrum (Figure 3.03), with the signal at 61.4 ppm corresponding to C-6' (Figure 3.04, purple line) and that at 33.5 ppm to C-8'. The remaining resonance, without any JCF coupling could be assigned to C-9'. The close proximity of the C-9' signal to the strongly correlating peak of C-3' (Figure 3.04, red line) obscured the H-9'a – C-9'correlation (Figure 3.04, green line), however, this could still be observed as a shoulder peak. These assignments were further supported by HMBC and 1H–19F HMBC analyses. The aromatic quaternary carbons of the acridine ring were assigned based on DEPT-Q, HMBC and HSQC analyses. It was possible to assign the spectrum of (R,R)-144 using the same combination of techniques detailed above. The spectra for diols (R,R)-145 and (S,S)-145, were less complex and more readily assigned. 3.4 - Fluoropyrrolidine ring conformation in 144.HCl To assess the conformation of the 3-fluoropyrrolidine ring in 144, the HCl salts were prepared. Acridine (S,S)-144 was dissolved in methanol and treated with HCl (1 M diethyl ether soln.), such that the nitrogen of the pyrrolidino rings were protonated (Scheme 3.06). This resulted in the precipitation of (S,S)-144 as the hydrochloride salt, and thus the salt had to be dissolved in d6-DMSO for NMR analysis. NNH NH O NONF FN+NH NH O N+ON+F FH HH a) Cl-Cl- Cl- Scheme 3.06. Reagents and conditions: a) HCl (1 M in Et2O), MeOH, rt. The 1H NMR spectrum of (S,S)-144.HCl showed significantly broadened signals as a result of further coupling to the N–H of the pyrrolidino ring and no coupling constants could be extrapolated. Interestingly, the 19F NMR of (S,S)-144.HCl, split into two well-defined resonances at -171.9 ppm and -173.4 ppm, upfield from that of the free amine (Figure 3.05). (S,S)-144 (S,S)-144 Chapter 3 63 -200 -200-190-190-180-180-170-170-160-160-150-150-140-140-130-130-120-120-110-110-100-100-90-90ppmppm 0.976 0.9761.011.01 -173.6 -173.6-173.4-173.4-173.2-173.2-173.0-173.0-172.8-172.8-172.6-172.6-172.4-172.4-172.2-172.2-172.0-172.0-171.8-171.8-171.6-171.6ppm -190-180-170-160-150-140-130-120-110-100-90 ppm -173.6 -173.2-172.8-172.4-172.0-171.6 0.98 1.00 Tuesday, 31 January 2012 Figure 3.05. A -19F NMR (470 MHz, d6-DMSO) of (S,S)-144. B - 19F NMR of (S,S)-144.HCl & C - Expanded section of B. The two resonances (Figure 3.05, C) must reasonably correspond to the formation of a diastereomeric pair, as protonation of the nitrogen can occur from either side of the rings, syn or anti to the axially orientated C–F bond (Figure 3.06). The equal distribution of protonation states was supported by DFT calculation, with the relative energy difference insignificant enough to influence the distribution of the two isomers (Figure 3.06). N + H O NH 2 F N + H F O NH 2 Figure 3.06. Protonation creates diastereomers of the pyrrolidine ring, which can be clearly observed in the 19F NMR. The relative energy difference was calculated by DFT (B3LYP/6-316(D)) by Dr Tomas Lebl, St Andrews. R = H. The coupling constants from the two resonances (Figure 3.05, C) can be used to demonstrate an axial solution conformation of the C–F bond, consistent with the literature.79-81 The 2JFH and 3JFH coupling constants could be extracted from the resonance at -171.9 ppm , although the 4JFH couplings are small and could not be B C A anti-151 Erel = 0.218 kJ/mol syn-151 Erel = 0 kJ/mol Chapter 3 64 quantified (Figure 3.07, A). In (S,S)-144.HCl four 3JFH coupling constants at -171.9 ppm could be determined with values of 39.4, 33.0, 25.8 and 19.5 Hz. The experimental coupling constants were entered into a NMR simulation package (iNMR) and the simulated spectrum (Figure 3.07, B) provided a good fit with the experimental data (Figure 3.07, A). The same treatment could not be repeated with the resonance at -173.4 ppm, however this signal was of similar width. -171.6 -171.7 -171.8 -171.9 -172.0 -172.1 -172.2 ppm 2 J FH = 52.8 3 J FH = 39.4 3 J FH = 33.0 3 J FH = 25.8 3 J FH = 19.5 - 1 7 1 . 9 p p m -171.6 -171.7 -171.8 -171.9 -172.0 -172.1 -172.2 ppm Figure 3.07. A - Expanded region of 19F NMR spectrum (470 MHz) for (S,S)-144.HCl & B - Simulated spectrum at 470 MHz. A comparison of the coupling constants observed for (S,S)-144.HCl with the literature was made.186 Thibaudeau et al., have reported trans 3JHF relationships (H–C–C–F torsion angles 160-180º) of between 30-45 Hz in ribose rings. In these systems, the size of the coupling constant is dependent on the substituents. The 3J values of 39.4 and 33.0 Hz suggest an H–C–C–F torsional angle approaching 180º, indicative of a pseudo-axial/axial coupling. The remaining two 3JFH values correspond well with torsion angles in the range 0-60º respectively, suggesting a pseudo-axial/equatorial conformation. These large coupling constants support the view that the fluorine-ammonium interaction in (S,S)-144.HCl leads to a highly ordered conformation with the C–F bond orientating in axial conformation (Figure 3.08, A). Experimental Simulated A B Chapter 3 65 H N+H RH HH H F HN+H RH HHH Fp-ax-eq p-ax-eqp-ax-axp-ax-ax p-eq-eqp-eq-eq Figure 3.08. Representations for the pseudo-axial (A) and equatorial (B) orientation of the C–F bond. If the C–F bond was orientated in a pseudo-equatorial conformation (Figure 3.08, B), for this particular signal, then the 3JFH coupling constants would be between 10-20 Hz. This is consistent with torsion angles approaching 60º, corresponding to a pseudo-equatorial-equatorial coupling. These coupling constants would result in a resonance with a smaller spectral width and less definition, similar to that of the non-protonated system (Figure 3.05, A). 3.5 - Co-crystallisation with quadruplex DNA 3.5.1 - Background and crystallisation Co-crystallisation trials with (S,S)- and (R,R)-144 and 145 with quadruplex DNA were conducted in collaboration with Prof. Steven Neidle’s laboratory at the UCL, School of Pharmacy. The approach used the hanging drop vapour-diffusion method (Figure 3.09). Mother liquor Cover Slip Grease A B C D Figure 3.09. A - Hanging drop vapour diffusion method for growing crystals, B - Top view of crystals growing on the top cover slip, C - Crystal tray with various conditions attempted such as altering the mother liquor concentrations & D - Attempted trays stored at 16 ºC. By this method the ligand and quadruplex DNA are added in various concentrations to a buffer solution, which is placed on a cover slip (Figure 3.09, A/B). This cover slip is then suspended over a concentrated salt solution, forming a closed system with grease A B Chapter 3 66 generating a tight seal. Evaporation of the buffer solution containing the ligand and quadruplex DNA results in super-saturation. Optimisation of this evaporation process by changing variables can result in the crystallisation of DNA-ligand crystals that are suitable for X-ray diffraction. The variables are: crystallisation temperature; buffer constituents; stock buffer concentrations; DNA to ligand ratio; and/or concentration relative to the buffer. Negative results such as DNA precipitation can be used to tune the crystal growing conditions.157 After an extensive exploration of variables by Dr. Nancy Campbell (School of Pharmacy) the optimal conditions for crystal growth were found, resulting in rhombohedral co-crystals of (R,R)-144 and (S,S)-144 with quadruplex DNA. The crystals (Figure 3.10) were subject to X-ray analysis on a synchrotron at the Diamond Light source (Oxfordshire, UK). Figure 3.10. Photograph of diffracted ligand-DNA crystals. The diffraction data was fitted and refined, providing crystal structures at a highly refined 1.18 Å and 1.10 Å resolution for (R,R)-144 and (S,S)-144 respectively. The resulting unit cell for crystals of (R,R)-144 is shown in Figure 3.11. 0.2 mm Chapter 3 67 Figure 3.11. Unit cell of the co-crystal of (R,R)-144 with O. nova bimolecular DNA. This representation and that shown in subsequent figures were created from the files deposited in the Protein Data Bank (PDB) with file names 3NYP and 3NZ7 for (R,R)-and (S,S)-144 respectively. Images were generated using the COOT187 and PyMol129 crystal visualisation packages. The high level of resolution of these co-crystals enabled a detailed assessment of interactions of these ligands with O. nova DNA and they could be compared to the co-crystal structure previously solved with BSU6039 141a to 1.75 Å (Figure 2.16). Thus, the co-crystals with the fluorinated analogues (R,R)- and (S,S)-144 were of particularly high quality. While the fluorinated analogues (R,R)- and (S,S)-144 provided suitable crystals for X-ray diffraction, the hydroxy compounds (R,R)- and (S,S)-145 failed to crystallise under the same conditions Altering the conditions for crystallisation failed to produce any suitable crystals for diffraction (Nancy Campbell, UCL). These compounds would have enabled a useful comparison as an intermediate between hydrogen and fluorine in the pyrrolidine ring, however suitable co-crystals were not forthcoming. Chapter 3 68 3.5.2 - General observations in the co-crystals with (S,S)- and (R,R)-144 The topology observed within the unit cell for both enantiomers of 144 is the same. This corresponds to the quadruplex DNA assembling with the thymine bases forming a loop at either end of the quadruplex in an identical manner to the co-crystals BSU6039 141a discussed in Chapter 2.11 (Figure 3.12). The DNA and ligand are bound in a one to one ratio with the acridine ligand occupying the cavity between an upper guanine tetrad and the thymine base loop (Figure 3.13). Although the two structures are formally diastereomers this does not significantly change the G-quadruplex conformation. [S, S] [R, R] Figure 3.12. Co-crystal structures for fluorinated pyrrolidine analogues. A - (S,S)-144 and B - (R,R)-144 both with quadruplex DNA (G4T4G4). The blue arrows highlight the ligand binding to the top face of the quadruplex. Representation created with PyMol.129 A B Chapter 3 69 Figure 3.13. Surface representation in cyan with 50% transparency. The binding cleft of the fluorinated BSU6039 ligands can be clearly seen at the top of both structures. A - (S,S)-144 & B - (R,R)-144. Generated with PyMOL. 129 The unity of the electron density maps in structures (R,R)- and (S,S)-144 correlates well with the resolution of the data, and enables a high degree of certainty in assigning atom coordinates (Figure 3.14 & 3.15). Where the electron density is poorly resolved, the crystallographer must use chemical intuition in fitting the structure. In Figures 3.14 & 3.15, the structures are represented according to the preset b-factor colours (COOT package), with cool (blue/green) and hot (yellow/red) representing low and high disorder of the atoms relative to one another. The b-factor is an important parameter of the solved data providing an appropriate measurement of the static/dynamic nature of the atoms within the crystal structure. It is intrinsically linked to the occupancy factor for each atom. The occupancy factor (0.10–1.00) is used in conjunction with the b-factor data to provide information about the precise geometry of substituents within the structure. A B Chapter 3 70 Figure 3.14. Electron density maps for (S,S)-144 Oxytricha nova quadruplex DNA. Represented at the σ = 1 level using the COOT package.187 Distances are in Å. Figure 3.15. Electron density maps for (R,R)-144 with Oxytricha nova quadruplex DNA. Represented at the σ = 1 level generated with the COOT package.187 Distances are in Å. The occupancy and b-factors (Table 3.01 and Figure 3.14/3.15) for each atom correlate well with the hydrogen bonding and electrostatic interactions, suggesting their contribution to structural stability. b-factor low high Monday, 5 March 2012 b-factor low high Monday, 5 March 2012 Chapter 3 71 Acridine nitrogen Pyrrolidine-N (LHS/RHS) C–F (LHS/RHS) C–F (LHS/RHS) BSU6039 (2.4 Å) 9.04 33.96 / 25.65 N/A N/A (S,S)-144 (1.10 Å) 6.73 11.03 / 18.65 17.21 / 37.35 11.03 / 18.65 (R,R)-144 (1.18 Å) 8.05 17.54 / 25.77 23.56 / 56.09 17.54 / 25.77 Table 3.01. b-Factors for the crystal structures with BSU6039 141a, (S,S)-144 and (S,S)-144 3.5.3 - Detailed assessment of the DNA co-crystal with (S,S)-144 The structure of BSU6039 141a, discussed in Chapter 2.16 is reproduced for comparison (Figure 3.16). The (S,S)-144 bound ligand is considered in Figure 3.17. Figure 3.16. Hydrogen bonding and electrostatic interaction distances found in the BSU6039 141a crystal structure. Distances are in Å. Chapter 3 72 Figure 3.17. A - Expanded section of the mode of binding for ligand (S,S)-144 with quadruplex DNA & B - graphical summary of angles and bond lengths. Distances are in Å. The most notable difference in the binding of fluorinated (S,S)-144 (Figure 3.17) over non-fluorinated BSU6039 141a (Figure 3.16) is that the fluoropyrrolidine rings have rotated by 180º and are forming new contacts within the crystal structure. Interestingly, in the left hand side (LHS) of the binding pocket (Figure 3.17), the N–H of the pyrrolidine is no longer forming a hydrogen bond with a crystallographically resolved water molecule. This water molecule is observed in all structures of this series and presumably plays an important role in the stabilisation of the guanine tetrad through a hydrogen-bonding network between the pyrrolidine and the guanine base (Figure 3.17). In fact, the pyrrolidine N–H of the structure with (S,S)-144 now forms a hydrogen bond with a phosphate on a neighbouring quadruplex within the unit cell. It may be the change in the pKa of the pyrrolidine substituent, due to fluorine incorporation, is responsible for the change in hydrogen bonding partner. The trajectory of the N+–H⋯-O–P hydrogen bond is close to 180º and the H⋯O distance is short at <1.7 Å (Figure 3.17, B), correlating well with the more acidic hydrogen bonding to the basic A B Chapter 3 73 phosphate to result in a ‘strong’ hydrogen bond. The observed occupancy factors (F = 1.00 and N = 1.00) and a good fit with the electron density map (Figure 3.14), suggest high order and a stable hydrogen bonding interaction. The pyrrolidine ring is clearly puckered as a result of the C–F⋯N–H+ dipole-charge interaction, with a F–C–C–N angle of approximately 90º. A similar situation occurs at the RHS of the crystal structure (Figure 3.17) where again the pyrrolidine N–H directionality has rotated through 180º relative to BSU6039 141a (Figure 3.15) and a hydrogen bond forms now with the phosphate backbone, albeit with a longer N–H⋯O–P contact (Figure 3.17) of 1.98 Å. The b-factors for the atoms in this ring are lower (Table 3.01) perhaps suggesting a weaker hydrogen bond between this pyrrolidine ring and the phosphate. As with the LHS, the RHS pyrrolidine ring is also puckered, with the C–F bond occupying a dramatic axial orientation (fluorine occupancy = 1.00). The distance between the N–H and C–F bond (~3.0 Å) and the narrow angle (< 100º) in both of the rings in (S,S)-144 preclude a reasonable hydrogen bond but are consistent with a charge-dipole interaction as discussed in Chapter 1. 3.5.4 - Detailed assessment of the DNA co-crystal with (R,R)-144 Examination of the binding mode of (R,R)-144 (Figure 3.18) with the quadruplex, reveals a similar change in pyrrolidine ring orientation relative to BSU6039 141a (Figure 3.16). On the LHS of the structure (Figure 3.18, A) again the N–H of the pyrrolidine forms a hydrogen bond (Figure 3.18, B) with a phosphate group of a neighbouring quadruplex within the unit cell. The C–F bond is also axial (fluorine occupancy = 0.90), exhibiting a similar conformational bias to that observed for (S,S)-144 (Figure 3.17). Chapter 3 74 N+F H PO OODNA ODNA N+ FHPO O ODNA ODNA165.1 º90.6 º2.92 Å 1.69 Å1.40 Å acridine 62.8 º132.0 º 3.39 Å2.70 Å 1.38 Å68.9 º65.6 º Figure 3.18. Expanded section of the mode of binding for ligand (R,R)-144 with quadruplex DNA & B - graphical summary of angles and bond lengths. Distances are in Å. The b-factors for the LHS of the X-ray structure of the co-crystalline (R,R)-144 with DNA (Table 3.01) although slightly higher than for (S,S)-144, remain low compared to 141a and indicate high resolution data. In the RHS binding pocket, it is clear that the co-crystalline DNA-(R,R)-144 structure the interaction between the pyrrolidine ring and the phosphate group is more electrostatic in nature. This is deduced from the longer N+–H⋯-O–P contact distance and non-linear contact angle (Figure 3.18, B), which are out-with the topological parameters for a strong hydrogen bond. Based on this observation, it is anticipated that the tolerance of the RHS binding interaction is less favourable to accommodating the fluorine in the pyrrolidine ring. The orientation of the C–F bond, in the case of (R,R)-144, possibly positions the C–F bond closer to the phosphate resulting in a charge-dipole repulsion. The stereospecific incorporation of a C–F bond in this manner can lead to a diastereotopic difference in the topology of binding between the two enantiomers (S,S)- and (R,R)-144, altering their relative modes of binding. A B Chapter 3 75 3.6 - FRET studies with quadruplex DNA The melting temperatures of co-complexes between the 144 and 145 series with quadruplex DNA were measured. This enabled the assessment of the relative stability of quadruplex DNA following addition of (S,S)- or (R,R)-144 and 145. For this application, the monomeric human quadruplex DNA sequence G3(T2AG3)3 with the donor and acceptor appendages TAMRA and FAM, as described in Chapter 2, were employed. These experiments were carried out in triplicate with 1 µM of (S,S)- or (R,R)-144 and 145 (Table 3.02. By Tony Respka, UCL). Sample Concentration Run1 (ºC) Run 2 (ºC) Run 3 (ºC) Average Tm (ºC) dTm (ºC) DNA control N/A 58.6 58.7 58.7 58.7 N/A (R,R)-144 1 μM 62.8 63.4 63.3 63.2 4.5 ± 0.4 (S,S)-144 1 μM 64.9 64.7 64.7 64.8 6.1 ± 0.2 (R,R)-145 1 μM 71.0 71.1 71.2 71.1 12.4 ± 0.2 (S,S)-145 1 μM 71.1 71.3 71.3 71.2 12.6 ± 0.2 BSU6039 1 μM - - - - 13.3 Table 3.02. Results from the FRET based assessment of (R,R)- and (S,S)-144 and 145. Errors by standard deviation of the mean and are reported to 1 DP. Both types of ligands stabilised the quadruplex, however the fluorinated ligands (S,S)-and (R,R)-144 resulted in a lower overall stabilisation of the fold relative to BSU6039 141a. This difference is difficult to quantify empirically, however it is likely that change in orientation of hydrogen bonding observed for the C–F ligands relative to BSU6039 141a weakens the structures. By contrast the hydroxyl compounds (S,S)- and (R,R)-145 demonstrated a much better stabilisation compared to BSU6039 142a. These results show that the C–F bond in the peripheral pyrrolidines perturbs the mode of binding more than that of the C–OH bond. Chapter 3 76 3.7 - Conclusion This chapter reports the synthesis of the (S,S)-/(R,R)-fluoro 144 and (S,S)-/(R,R)-hydroxy 145 analogues of BSU6039 141a. The fluoro analogues were successfully co-crystallised with the bimolecular O. nova quadruplex DNA and the X-ray structures were solved to high resolution. This data enabled a detailed assessment of the mode of binding between the two enantiomers of 144 and relative to BSU6039 141a. This analysis demonstrated that the C–F bond orientated in an axial position, consistent with the anticipated literature and that the hydrogen bonding network between the quadruplex DNA and ligand changed. In these structures with (S,S)- and (R,R)-144 the pyrrolidine N–H had rotated by 180º and paired with the phosphate backbone to form new hydrogen bonding interactions. Further to this, FRET analysis indicates that the fluorinated derivatives increased the stability of the quadruplex fold, however to a lesser extent when compared to BSU6039 141a. This may be explained by the hydrogen-bond differences observed in the co-crystal structures. Overall, the incorporation of a C–F bond in the peripheral pyrrolidines leads to a ring pucker and perturbed the mode of binding to quadruplex DNA. Chapter 4 Synthesis of C–F bond incorporated BRACO-19 analogues 4.1 - Introduction This chapter describes the synthesis of an enantiomeric pair of 3- and 6- substituted di-fluorinated analogues (152) of BRACO-19 (142a) (Figure 4.01), to investigate if the C–F bonds can induce an improved binding conformation over BRACO-19 142a to quadruplex DNA. 3 N+ 6 NHNH αO N+αON+ HN NN+ NHNH O F N+OFN+ HN NN+ NHNH O F N+OFN+ HN N H HHH HHH HH Figure 4.01. BRACO-19 142a & 3,6-C–F bond substituted BRACO-19 152 analogues. (R,R)-152 (S,S)-152 142a Chapter 4 78 In BRACO-19 142a, the amino groups of the pyrrolidine substituents are protonated at physiological pH, thus stereospecific C–F bond incorporation in the α-amide positions will exert a stereoelectronic influence to provide enantiomeric conformations of (R,R)- and (S,S)-152. This conformational bias is anticipated to arise for two reasons, charge-dipole compensation between C–F and N+–H bonds and the α-fluoroamide effect (Figure 4.01). NO RHR'H F NO RHF HHanti cisHigh energydisfavouredLow energy ~7-8 kcal mol-1HF HNH3+HH HH FNH3+HHFH HNH3+HHFavoured by ~ 5.8 kcal mol-1Disfavoured high energy state Scheme 4.01. The charge-dipole effect (A) and α-fluoroamide effect (B). R/R' = H, CH3. Therefore, for each stereoisomer, the C–F bond will orientate anti to the carbonyl, extending the planarity of the amide bond (Figure 4.01). Also, the protonated ring amine will align g+ or g- relative to the C–F bond through the charge-dipole effect (Chapter 1). Thus, the (R,R)-stereoisomer of 152 should have an enantiomeric twist relative to the (S,S)-stereoisomer (Figure 4.02).19 The conformational bias induced in 152 will force the charged pyrrolidine substituents to orientate towards the phosphate grooves of a quadruplex DNA fold. Thus, this conformation will be in contrast to the planar binding mode of BRACO-19 142a with quadruplex DNA (Figure 2.18). This conformational bias coupled with the change in pKaH of the pyrrolidine nitrogen, as a result of C–F bond incorporation, will generate the potential for new hydrogen bonds between 152 and the phosphate backbone of quadruplex DNA structure. This altered binding conformation is hypothesised to result in a stronger complementary binding between 152 and quadruplex DNA and thus stabilise the DNA fold more than that of BRACO-19 142a. A B g- g+ Chapter 4 79 H HFH O NHHHH F O NHN+HN+H H HHFONH N+HHHF HONH N+H NN HN NHN N Figure 4.02. Conformational representations of the influence of the C–F bond in fluorinated BRACO-19 analogues 152. 4.2 - Aims It therefore became a research objective to prepare the (R,R)- and (S,S)- enantiomers of 152. A retrosynthesis is shown in Scheme 4.02. N NHNH O F NO HN NFN NH2N NH2O OFNH2N N O+N NHNH O F NO HN NFN OFN O Scheme 4.02. Retrosynthetic approach to the synthesis of α-fluorinated BRACO-19 152 analogues. A key step in the synthesis involved the preparation of each enantiomer of the fluorinated amino ester 153. (R,R)-152 (R,R)-152 (R)-153 155 154 (S,S)-152 (S)-153 (S, S)-152 Chapter 4 80 4.3 - Synthesis of an α-fluoro-β-amino acid As highlighted in Chapter 1, Deniau et al., employed a DAST 32 mediated approach for the synthesis GABA analogues 110.87 Treating β-amino alcohol (R)-157 with DAST 32 resulted in fluorination to furnish (S)-158 in a stereospecific manner. The benzyl protected amine 158 could be further manipulated to 3-fluoro GABA (S)-110 after oxidation of the aromatic ring and subsequent deprotection of the amine (Scheme 4.03). FBn2N F+H3N O O-NBn2HO i) RuCl3ii) H2, Pd/CNBn2HO O LiAlH4 DAST Scheme 4.03. Synthesis of 3-fluoro GABA (S)-110. The rearrangement and fluorination arise from the aziridinium intermediate 160 formed from SNi attack of the nitrogen lone pair at the β-carbon following activation of the alcohol by DAST 159 (Scheme 4.04).188 The transient intermediate 160 can either be fluorinated in the α- or β-positions respectively by SN2 attack of fluoride (Scheme 4.04). FBn2NNBn2O a)NBn2OSEt2N NBn2F +F F+N SF F -HF β-fluorinationα-fluorinationa bSNi 1 : 4H N+BnBn Scheme 4.04. Mechanism of DAST 32 mediated fluorination of β-amino alcohols (R)-156 (R)-157 (S)-158 (S)-110 (R)-157 (R)-161 (S)-158 (R)-159 (R)-160 N2 Chapter 4 81 Nucleophilic attack of fluoride to the α-position leads to the substitution product (R)-161, whereas attack at the β-position results in the rearranged product (S)-158 and with an inversion of stereochemistry (Scheme 4.04). In this particular example the fluorination proceeds with a 4:1 selectivity in favour of the β-fluorinated product. The fluorination of esters of the natural amino acid L-serine, with DAST 32, also undergoes this rearrangement to furnish esters, (R)-120 and (S)-163 respectively (Scheme 4.05). This transformation typically proceeds with high α-fluorination selectivity and with excellent enantiocontrol.95 NBn2HO OO αβ ON+BnBn OViaDAST F- FBn2N OO NBn2F OO+ Scheme 4.05. Putative aziridinium intermediate formed during fluorination of benzyl protected (S)-serine methyl ester (S)-119. The approach seemed appropriate for the synthesis of (R)- and (S)-153. The only concern was the tolerance of the functional groups on the nitrogen. Only benzyl protecting groups are reported in the literature for this transformation.95 The dialkyl amino acid 119 was first prepared from methyl ester 165. Starting with racemic serine 164, the methyl ester 165 could be synthesised on a gram scale by treatment with thionyl chloride in methanol (Scheme 4.06). Treatment of 165 with benzyl bromide and potassium carbonate enabled a straightforward preparation of 119 in good yield (94%) (Scheme 4.06), ready for fluorination with DAST 32 ONH3ClOHO ONBn2OHOb)ONH2 OHHO a) Scheme 4.06. Reagents and conditions: a) SOCl2, MeOH, rt, 24 h and b) Benzyl bromide (2.2 eq), K2CO3 (4 eq), MeCN, rt, 24 h, 94%. (S)-119 (S)-162 (R)-120 (S)-163 164 119 165 Chapter 4 82 The reaction between alcohol 119 and DAST 32 in THF (Scheme 4.07) was monitored by 19F NMR spectroscopy, which revealed the presence of the α- and β-products, with signals at -190 ppm and -220 ppm. OFBn2N Oa)ONBn2OHO ONBn2OF+ Scheme 4.07. Reagents and conditions: DAST 32 (1.1 eq), THF, 0 ºC, 1 h, 90%. The major product was the α-isomer 120, which was formed with 95:5 selectivity over the β-isomer 163. This reaction scaled well and the α-isomer 120 was isolated in excellent yield (90%). This procedure was repeated with enantiopure D- and L-serine 164 and furnished the individual enantiomers (S)- and (R)-120 in gram quantities with excellent enantiocontrol (>90% ee) as analysed by chiral HPLC (ChiralCel OD-H). Debenzylation of methyl ester 120 was not straightforward. Classical hydrogenation conditions using 10% Pd/C in methanol or ethyl acetate under a hydrogen atmosphere was unsuccessful (Scheme 4.08). O OFBn2N Variousconditions O OFH2N Scheme 4.08. Hydrogenation of benzyl protected ester 120 gave multiple products. In these reactions, analysis of the reaction product by 19F NMR showed various fluorinated products, which could not be separated by standard reverse phase chromatography. These products could not be identified other than containing desired amine 166. The addition of acid during the reaction resulted in elimination forming acrylate type products as confirmed by 19F NMR. Successful debenzylation of 120 was achieved with Pd(OH)2/C in methanol, with ester 166 isolated in quantitative yield (Scheme 4.09). However, this reaction was unreliable and readily gave multiple products despite successful literature reports.189,94,190 120 166 119 120 See text 160 Chapter 4 83 OFBn2N O OFClH3N Oa) Scheme 4.09. Reagents and conditions: a) Pd(OH)2/C, H2, MeOH, 24 h followed by HCl (1 M), 99%. When the debenzylation was followed by 19F{1H} NMR, only one product at -200 ppm was observed, presumably the debenzylated material 166. However, purification through Celite led to degradation. Thus, in an attempt to isolate amine 166 it was protected as its carbamate ester in situ following removal of the palladium catalyst by filtration, without addition of HCl to the mixture (Scheme 4.10). O OFH2N O OFBocHN O OF(Boc)2Na) + Scheme 4.10. Reagents and conditions: a) Boc2O, Et3N, aqueous dioxane (25% v/v), rt, 24 h. The 19F{1H} NMR spectrum of this mixture again indicated a complex mixture from which both 167 and 168 could not be isolated cleanly as they co-eluted with multiple unidentifiable products, although the 1H NMR spectrum had no evidence for aromatic residues. Due to the nature of this complex product mixture this approach proved unsuccessful and was discontinued. 4.3.1 - Pyrrolidine functionalisation of 166 The double alkylation of 166.HCl to form a pyrrolidine ring was next investigated. This involved treating the amine with 1,4-dibromobutane and a catalytic amount of TBAI (Scheme 4.11). OFClH3N O a) OFN O Scheme 4.11. Reagents and conditions: 1,4-Dibromobutane (1.1 eq), TBAI (0.2 eq), Na2CO3 (4.0 eq), MeCN, reflux, 4 h, 77%. 166.HCl 120 166 167 168 166 153 Chapter 4 84 Pyrrolidine 153 was isolated in good yield (77%) following purification by chromatography. However, this route was not practical for the synthesis of 153, with the inefficient debenzylation step limiting the quantity of amine 166 available. Therefore an alternative approach was sought, which involved debenzylation after amide coupling to the acridone core (Scheme 4.12). NH NH2H2NO OFBn2NNH NHNH OO OFBn2N F NBn2OH + Scheme 4.12. Proposed amide coupling route to functionalise acridone 169. In order to investigate amide coupling to acridone 155, the methyl ester 120 required hydrolysis to carboxylic acid 170. This was achieved with KOH in methanol and generated (R)-170 in 97% yield (Scheme 4.13). OFBn2N O OFBn2N OHa) Scheme 4.13. Reagents and conditions: KOH, MeOH, rt, 36 h, 97%. The reaction was repeated also for the (S)-170 enantiomer. Thus with the carboxylic acids (R)- and (S)-170 available, it was next required to synthesise acridone 155. 4.4 - Acridone 155 synthesis The synthesis of acridone 155, the structural core of BRACO-19 142a, was carried out according to a literature method.191,192 The tetranitro diphenylmethane 172 was prepared from diphenyl methane 171 in refluxing sulfuric acid with potassium nitrate (Scheme 4.14). The reaction product was recrystallised from acetic acid and furnished 172 in good yield (74%). Oxidation to benzophenone 173 was high yielding (95%) and 169 170 155 (R)-120 (R)-170 Chapter 4 85 was achieved by refluxing 172 with chromium trioxide in acetic acid (Scheme 4.14). NO2O2N NO2 NO2 NO2O2N NO2 NO2Oa) b) Scheme 4.14. Reagents and conditions: a) KNO3, H2SO4 (conc), 70 °C, 2 h, 74%; b) CrO3, acetic acid, 118 °C, 16 h, 95%. The nitro-groups of the benzophenone derivative 173 were reduced by refluxing with a large excess of tin(II) chloride in hydrochloric acid. This also facilitated ring closure to form acridone 155 with the 3,6-diamino regiochemistry (Scheme 4.15). NHOH2N NH2O NO2O2N NO2 NO2 a) Scheme 4.15. Reagents and conditions: a) SnCl2, HCl (conc.), EtOH, reflux, 3 h, 63%. Acridone 155 is a clay-like solid that was difficult to manipulate and had poor solubility in DMSO, DMF or DMAc. Despite this, it was possible to obtain an 1H NMR spectrum in d6-DMSO where both NH2 and NH resonances could be assigned in-line with the literature.191 4.5 - Coupling reactions with diaminoacridine 155 With suitable quantities of acridone 155 in hand, it was now possible to investigate coupling conditions with the fluorinated amino acid 170. There is limited literature on the chemistry of acridone 155. In the synthesis of BRACO-19 142a, 3-chloropropionyl chloride 148 is used as the reaction solvent under forcing conditions to prepare amide 174 (Scheme 4.16),191 with a microwave irradiation approach also requiring a significant excess of the acid chloride 148.184 Another approach to form amides on acridone 155, employed 1:1 mixtures of acid anhydrides under forcing conditions, to prepare acetamides such as 175 (Scheme 4.16).193 171 172 173 173 155 Chapter 4 86 NHO NH2H2N OCl Cl NHO NHNH O ClOCl145 ºC80 ºCNHO NHNHO OAc2O/AcOH N NHNH O NON HN N3 Steps Scheme 4.16. Synthesis of bis-chloro and acetamide acridones. From these examples it is clear that forcing conditions are required but in our case a large excess of fluorinated amino acid is not practical. Alternative efforts with practical equivalents of acid chloride 176 were investigated. Initially, 2.5 equivalents of the acid chloride 176, which was formed in situ from with thionyl chloride and directly added to acridone 155 in DMAc, was explored (Scheme 4.17). NHO NH2H2NOFBn2N Cl + Various Conditions Scheme 4.17. Attempted coupling of acid chloride 176 to acridone 155. Conditions are summarised in Table 4.01. The use of 19F{1H} NMR with unquenched reaction aliquots was useful in following this reaction as both acid 170 and acid chloride 176 have characteristic chemical shifts. However, these reactions, which were heated over extended time periods, failed to give product as judged by TLC and 19F{1H} NMR. Alternative conditions with different equivalents of acid or with different reagents were attempted (Table 4.01), but in each case no product could be observed. 155 175 148 174 142a 176 155 Table 4.01 various conditions Chapter 4 87 Entry Acid equiv. Reagent Base Solvent Temp (ºC) Time (h) 1 2.5 SOCl2 Na2CO3 DMAc 100 24-48 2 2.5 SOCl2 Na2CO3 DMAc 160 24 3 20 SOCl2 - DMAc 160 24 4 20 SOCl2 Na2CO3 DMAc 100 48 5 20 SOCl2 pyridine DMAc 100 24 6 20 SOCl2 DMAP DMAc 100 24 7 2.5 cyanuric fluoride pyridine DMAc 100 16 8 4 Ethyl chloroformate NMM DMAc rt >24 9 4 Benzyl chloroformate NMM DMAc rt >24 10 2.5 EDC + HOBt NMM DMF rt >24 11 2.5 EDC + HOBt DiPEA DMF rt 24 12 2.5 EDC + HOBt DiPEA + DMAP (cat.) DMF rt 24 13 2.5 EDC + HOBt DiPEA DMF 60 24 14 2.5 HBTU DiPEA DMF rt 24 15 2.5 HATU DiPEA DMF rt 24 Table 4.01. Attempted conditions for the coupling of acid 170 with acridone 155. In all cases there was no evidence for the formation of product 166. It was clear that the nucleophilicity of the amine in acridone 155 is insufficient for amide bond formation by classical means. An alternative literature approach employing metal amides, such as 178, for the synthesis of aromatic amides such as 180 was subsequently explored (Scheme 4.18).194,195 Chapter 4 88 N O O CH3BnOCH3Li2N THF-78 ºC N OBn OCH3NH+ OCH3H2NBuLi THF Scheme 4.18. Generation of the bis-lithium amide for the aminolysis reaction. This was an attractive alternative for the construction of the desired acridone 169 (Scheme 4.19). NHO NHNH O F NBn2OFBn2NOFBn2N Oa)NHO NH2H2N Scheme 4.19. Reagents and conditions: BuLi (4 eq), THF, -78 ºC, 1 h, followed by 120, -78 ºC, 3 h, <10%. The solubility of acridone 155 in THF was problematic, however the addition of BuLi brought the acridone into solution, resulting in a homogenous yellow solution. The addition of fluorinated ester 120 maintained this colour. Monitoring of the reaction by TLC and 19F{1H} NMR indicated multiple products. Separation of these products by column chromatography enabled the isolation of the coupled acridone 169, although in low yield (<10%). The remaining products from the reaction, which were both non-fluorinated and fluorinated, could not be identified. Subsequent optimisation of the reaction conditions showed that treatment of acridone 155 with KHMDS followed by the addition of ester 120 in THF at -78 ºC via cannula, resulted in a modest improvement in the reaction yield (19%). 177 178 155 120 169 179 180 Chapter 4 89 Treating acridone 169 with neat phosphorus oxychloride gave the 9-chloro intermediate 181, which was reacted directly with N,N-dimethylaminoaniline 154 in chloroform (Scheme 4.20). NHO NHNH O F NBn2OFBn2N N NHNH O F NBn2OFBn2N HN Na), b)N Clvia Scheme 4.20. Reagents and conditions: POCl3 (neat), 105 ºC, 3 h and b) N,N-dimethylaminoaniline 154 (10 eq), CHCl3, reflux, 2 h, 23% over two steps. Decomposition of aniline 154 complicated both TLC analysis of the reaction and also the subsequent purification of racemic acridine 182. Multiple columns were required to obtain a pure sample of the trisubstituted product 182. The HCl salt of 182 was particularly insoluble and was not suitable for DNA quadruplex binding studies. 4.5.1 - Debenzylation of acridone 169 It was evident that the benzyl groups of 182 gave an unsuitable non-drug like compound, thus removal of the benzyl groups of acridone 169 was explored. Various conditions (Table 4.02) were attempted (Scheme 4.21), however, multiple products were observed (19F{1H} NMR analysis), similar to that obtained with ester 120. 169 182 181 Chapter 4 90 NHO NHNH O F NH2OFH2NNH O NHNH O F NBn2OFBn2N Various conditions Scheme 4.21. Debenzylation of acridone 169. Entry Catalyst Solvent Temp. Pressure 1 10% Pd/C CH3OH rt atm 2 20% Pd(OH)2/C CH3OH rt atm 3 20% Pd(OH)2/C CH3OH + Acetic acid rt atm 4 20% Pd(OH)2/C Ethyl Acetate rt atm 5 Pd black CH3OH rt atm 6 (H-Cube®) Pd/C MeOH rt 1 bar 7 (H-Cube®) 20% Pd(OH)2/C MeOH rt 1 bar 8 (H-Cube®) 20% Pd(OH)2/C MeOH 40 ºC 1 bar 9 (H-Cube®) 20% Pd(OH)2/C MeOH rt 50 bar Table 4.02. Attempted hydrogenation conditions on tetra-benzylated 169. All flow reactions were conducted on a 1 mmol scale with a flow rate of 1 mL/min. In all cases multiple products were observed. 4.6 - Alternative protecting groups Alternative dialkyl protecting groups for serine were next explored. Cossy reported in 2010 that treating allyl protected 184 with DAST 32 resulted in fluorination and rearrangement, consistent with other amino alcohols, to furnish 185 with high selectivity and excellent enantiocontrol (99% ee). Deprotection of amine 185 was achieved to furnish 186 with a palladium based catalyst (Scheme 4.22).196 169 183 Table 4.02 Chapter 4 91 N(allyl)2OHR F N(allyl)2RDAST Pd(dppb)Thiosalicylic Acid F NH2R Scheme 4.22. Fluorination of quaternary-β-amino alcohols. Thus, the route was investigated, with the potential for this motif to undergo RCM to form the 5-membered pyrrolidine ring. Starting from the respective serine methyl ester, both (R)- and (S)-187 were prepared in good yields (62% and 57%) (Scheme 4.23). ON OHOONH3ClOHO a) Scheme 4.23. Reagents and conditions: a) Allyl bromide (4 eq), K2CO3, MeCN, reflux, 16 h, 57%. Fluorination of alcohols (R)- and (S)-187 was achieved by treatment with DAST 32 and this gave their respective α-fluorinated isomers (S)- and (R)-188 in good yield (69% and 61% respectively, >95% ee) (Scheme 4.24). Analysis of the reaction mixture with 19F{1H} NMR showed that fluorination to the α-product 188 proceeded with 95:5 regioselectivity over the β-product 189, consistent with that found with the benzyl moiety. OFN Oa)ON OHO ON OF+ Scheme 4.24. Reagents and conditions: DAST 32 (1.1 eq), THF, 0 ºC, 1 h. The removal of the allyl groups was next explored. Treatment of allyl protected 188 using various literature procedures (Table 4.03), failed to furnish free amine 166.197 184 185 186 (S)-187 (R)-188 164 187 (S)-189 Chapter 4 92 OFN O OFClH3N OVarious conditions Scheme 4.25. Attempted de-allylation reaction with conditions summarised in Table 4.03 Entry Catalyst Solvent Temperature 1 RhCl(PPh3)3 THF reflux 2 PdCl2 THF reflux 3 Pd(dppb) THF reflux 4 PdCl2 THF rt Table 4.03. Summary of the conditions attempted for the cleavage of the allyl groups in 188. In all cases multiple products were observed by TLC and NMR. It was found that diallyl amine 188 exhibited a similar side-reactivity to hydrogenation of benzyl protected 120, with multiple products observed in the 19F{1H} NMR spectrum. Thus an alternative approach was sought. 4.6.1 - Ring closing metathesis approach with 188 The allyl groups in 188 offered the opportunity for ring-closing metathesis through to dehydropyrrolidine 190 (Scheme 4.26). Initial attempts of the RCM reaction of ester 188, with 10 mol% of Grubbs 1st generation catalyst 191 failed to provide the desired cyclised product 190, as judged by NMR and MS analyses. OFN O OFN OVarious conditions Scheme 4.26. Ring closing metathesis strategy to dehydropyrrolidine 190. The addition of Ti(OiPr)4 or acetic acid with the Grubbs I catalyst 191, in CH2Cl2 or toluene at room temperature or reflux also failed to provide dehydropyrrolidine 190.198,199 Alternative catalysts such as Hoveyda-Grubbs 193 or the temperature stable (R)-188 (R)-166 (R)-188 (R)-190 Chapter 4 93 indenylidene based 194 at 1.0 mol% to 10 mol% catalytic loadings were also unsuccessful (Figure 4.03).200 In each case multiple products were observed by 19F{1H} NMR analysis. RuClCl P(Cy)3NN PhP(Cy)3RuCl ClP(Cy)3 Ph P(Cy)3RuClCl ORuCl ClP(Cy)3 PhNN Figure 4.03. Catalysts employed in the investigations of the RCM reaction of (R)-188. 4.7 - Acridone coupling with ester 188 As an alternative strategy, the diallyl ester 188 was coupled to acridone 155 and then chemistry on the side chains was subsequently explored. Thus (S)-188 was treated with the optimised base mediated coupling procedure (Scheme 4.27) and diamide acridone (S,S)-195 was isolated in a modest yield (23%). NHO NHNH O F NOFNOFN Oa)NHO NH2H2N Scheme 4.27. Reagents and conditions: KHMDS, THF, -78 ºC 1 h, followed by (S)-188, 78 ºC to rt, 12 h, 23%. The 1H and 13C NMR spectra of (S,S)-195 were readily assigned with the overlapping terminal allyl resonance and CHF resonance confirmed by 1H-19F HMQC analysis (Figure 4.04) 191 192 193 194 (S)-188 155 (S,S)-195 Chapter 4 94 2.0 3.05.06.0 4.0 -220 -200 -180 -170 -190 -210 ppm Figure 4.04. 1H-19F HMQC (300/282 MHz, CD3OD) spectrum of acridone (S,S)-195. Repeating the procedure starting from (R)-188 enabled isolation of the enantiomeric (R,R)-195, which had identical spectroscopic characteristics. 4.7.1 - Allyl deprotection of acridone 195 With suitable quantities of both stereoisomers of (R,R)- and (S,S)-195 in hand, RCM and de-allylation of the allyl groups was explored. Ring closing metathesis failed again with the conditions previously attempted for ester 188. This was presumably due to the lone pair of the secondary amine, but also to the insolubility of 195 in toluene or CH2Cl2. However, the de-allylation of acridone 195 to diamine 196 was achieved following a modification of a literature procedure. This used thiosalicylic acid 197 and a palladium phosphine based catalyst 198 (Scheme 4.28). Chapter 4 95 NHO NHNH O F NOFN NHO NHNH O F NH3ClOFClH3Na) Scheme 4.28. Reagents and conditions: Pd2(dba)3 (10 mol%/allyl group), DPPB, thiosalicylic acid 197, THF, reflux followed by HCl, 89%. This reaction proceeds via the Pd-allyl cation 199 facilitating nucleophilic attack of thiosalicylic acid 197, which acts as an allyl scavenger (Scheme 4.29). Pd+Pd0 NHRHN RSHCO2-SCOOH SHCO2H SHCO2-NR NH2RRepeat cycle Scheme 4.29. Mechanism for Pd catalysed de-allylation with stoichiometric thiosalicylic acid. R/R' = alkyl, aryl. This reaction proceeded smoothly, with an acidic work up enabling the isolation of amine (S,S)-196.HCl by aqueous extraction. Purification by reverse phase chromatography followed by freeze-drying provided amine (S,S)-196.HCl in an almost quantitative yield (95%). By contrast, when ester 188 had been treated under these conditions multiple products were observed. However, this post coupling strategy proved much more successful with acridone 195. 1H NMR analysis confirmed that the (S, S)-195 (S, S)-196.HCl 197 200 198 199 195 201 196 Chapter 4 96 acridone core had remained intact and that the allyl groups were cleanly removed (Figure 4.05). 5.0 3.0 4.06.07.0 8.0 Figure 4.05. 1H NMR (300 MHz, D2O) spectrum of acridone (S,S)-196 after allyl deprotection. Repeating the procedure with (R,R)-195 enabled the isolation of the complementary diamine (R,R)-196 (Figure 4.06). With both diamines available in suitable quantities, an investigation into the functionalisation of the terminal amines was now explored. NHO NHNH O F NH3ClOFClH3N NHO NHNH O F NH3ClOFClH3N Figure 4.06. Both enantiomers of the deallylated acridone. (S,S)-196.HCl (R,R)-196. HCl 2 3 41 NH109 5 678O NHNH O F NH3ClOFClH3N Chapter 4 97 4.7.2 - Functionalisation of acridone 196 Pyrrolidine formation of the terminal amines of 196 with 1,4-dibromobutane was explored (Scheme 4.30), however these reactions were unsuccessful despite trying a range of conditions (Table 4.04). NHO NHNH O F NOFNVarious conditionsNHO NHNH O F NH2OFH2N Scheme 4.30. Attempted functionalisation of the amino group in acridone 196. Conditions tested are summarised in Table 4.04. 1,4-dihalobutane Base Solvent Conditions Time (h) Outcome Bromo- DiPEA MeCN reflux 48 No Rx Bromo- + TBAI (cat) DiPEA MeCN reflux 48 Inconclusive Bromo- K2CO3 THF reflux 48 Inconclusive Iodo- K2CO3 THF reflux 48 No Rx Iodo- K2CO3 DMF 100 ºC 24 Cleavage of amide Iodo- K2CO3 THF/MeCN (1:1) Microwave, 120W 0.5 No Rx Iodo- K2CO3 DMF Microwave, 120W 0.5 Cleavage of amide Iodo- K2CO3 Et3N (2 eq) THF/DMF (9:1) rt 48 No Rx Iodo- Et3N (6 eq) DMF rt 48 No Rx cis-1,4-dichloro-2-butene Et3N (6 eq) DMF rt 48 No Rx Table 4.04. Attempted conditions for the reaction detailed in Scheme 4.30. 202 196 Table 4.04 Chapter 4 98 An alternative approach involved the double reductive amination of acridone 196 with 1,4-butanedial 204 (Scheme 4.32). The required succinate dialdehyde 204 was accessed by hydrolysis of 2,5-dimethoxytetrahydrofuran 203 (Scheme 4.31). O OO H O O Ha) Scheme 4.31. Reagents and conditions: HCl (1 M), rt, 20 min, basified and distilled.201 NH NHNH OO OFH2N F NH2OOH H NH NHNH OO OFN F NVariousconditions+ Scheme 4.32. Attempted route towards pyrrolidine functionalised acridone 202. Again, the solubility of acridone 196 was a limiting factor in this reaction (Scheme 4.32). Various borohydride reagents such as NMe4BH(OAc)3 and NaBH(OMe)3 in THF were used with and without acetic acid.202 However, the desired product was not identified. 4.8 - Trisubstituted acridine 206 synthesis An alternative strategy where the acridone moiety in 195 was converted to the appropriately trisubstituted acridine was taken. This would enable the synthesis of analogues for DNA binding and telomerase assays. To achieve this, acridone 195 was treated with neat phosphorus oxychloride to access the 9-chloro intermediate 205 (Scheme 4.33). The use of elevated temperatures resulted in the elimination of diallyl amine as indicated by a resonance at -120 ppm in the 19F NMR spectrum. 203 204 204 See text 196 202 Chapter 4 99 NHO NHNH O F NOFN N NHNH O F NOFN HN Na), b)N Clvia Scheme 4.33. Reagents and conditions: a) POCl3 (neat), rt, 24 h & b) N,N-dimethylaminoaniline 154 (20 eq), CHCl3, reflux, 5 h, 59%. The intermediate was used straight away in a SNAr reaction with aniline 154, using the stable monohydrochloride 207, which was neutralised and used immediately to avoid decomposition (Scheme 4.34). ClH3N N a) H2N N Scheme 4.34. Reagents and conditions: Na2CO3 (sat. aq. soln.), Et2O, rt, quantitative. Exclusion of light and air minimised the decomposition of the extracted aniline prior to the reaction. Refluxing aniline 154 with (S,S)-195 in chloroform for 5 h enabled the formation of the desired acridine product in good yield over two steps (59%). The 1H NMR spectrum of (S,S)-195 was readily assigned with the aniline substituent clearly defined (Figure 4.07). Starting from (R,R)-195, it was also possible to access the complementary (R,R)-195 enantiomer, in a yield of 61%. (S, S)-195 207 154 (S, S)-206 205 Chapter 4 100 5.0 3.0 3.54.04.5 5.5 6.06.57.07.5 8.08.5 9.8 10.04.02.0 2.0 1.9 1.82.0 Figure 4.07. 1H NMR (400 MHz, CD3OD) spectrum of trisubstituted acridine (S,S)-206. The UV-vis absorption spectrum of acridine 206 (maxima at 268, 294, 365 and 425 nm) has a broad absorption extending up to 700 nm. This made it difficult to record an optical rotation value. 4.8.1 - Allyl deprotection of acridine 206 With practical quantities of acridine (R,R)- and (S,S)-206 in hand, allyl group deprotection was next investigated. This was successfully achieved following the protocol developed for the deallylation of acridone 195 (Scheme 4.35). 2 3 41 N109 5 678 NHNH O F NOFN HN1213 14 15 N Chapter 4 101 N NHNH O F NOFN HN N a) N NHNH O F NH3ClOFClH3N HN N Scheme 4.35. Reagents and conditions: Pd2(dba)3 (10 mol%/allyl group), dppb, thiosalicylic acid, THF, reflux followed by HCl, 61%. Analysis of the crude reaction product by NMR indicated that the acridine heterocyclic core was unaffected by the conditions and that the allyl groups were cleaved to furnish (S,S)-208 as its HCl salt. This required careful purification by C-18 reverse phase chromatography. The 1H NMR confirmed cleavage of the allyl groups and the CH2CHF, CHF and aromatic resonances were clearly resolved (Figure 4.08). The procedure was repeated for the other (R,R)-206 enantiomer, to yield amine (R,R)-208 in 51% yield. (S, S)-206 (S, S)-208.HCl Chapter 4 102 3.0 4.06.07.0 8.0 5.0 9.0 3.80 3.75 3.70 3.65 3.60 3.55 3 . 7 2 3 . 7 1 3 . 6 9 3 . 6 8 3 . 6 7 3 . 6 4 3 . 6 3 3 . 6 3 3 . 6 2 3 . 6 0 3 . 5 9 3 . 7 8 3 . 7 7 3 . 7 5 3 . 7 4 3 . 6 6 Figure 4.08. 1H NMR (500 MHz, D2O) spectrum of acridine (S,S)-208.HCl after reverse phase chromatography. The two stereoisomers (R,R)- and (S,S)-208.HCl are undergoing binding assays with human quadruplex DNA at the UCL School of Pharmacy (Figure 4.09). Co-crystallisation trials with the G3(T2AG3)3 telomeric sequence are also underway to enable a structural assessment of the influence of the C–F bond on ligand binding to quadruplex DNA. NNH NHOFClH3N O F NH3ClHN NNNH NHOFClH3N O F NH3ClHN N Figure 4.09. Selectively fluorinated acridines for studies with quadruplex DNA. (R,R)-208.HCl (S,S)-208.HCl 2 3 41 N109 5 678 NHNH O F NH3ClOFClH3N HN1213 14 15 N Chapter 4 103 4.9 - Alternative side chain functionalisation A hydrogenation reaction of the allyl groups of ester (S)-188 was explored (Scheme 4.36). This proceeded smoothly with 10% Pd/C as a catalyst, and gave the N,N-dipropyl product (S)-209 in good yield (61%) OFN O OFN Oa) Scheme 4.36. Reagents and conditions: a) 10% Pd/C (20 mol%), H2, EtOAc, 24 h, rt, 61%. Repeating this reaction for (R)-209 enabled the isolation of the other enantiomer in a yield of 51%. 4.9.1 - Acridone coupling with ester 209 The N,N-dipropyl ester (S)-209 was thus subject to the general coupling procedure described above, to generate acridone (S,S)-210 (Scheme 4.37). This reaction went smoothly and purification of acridone (S,S)-210 was relatively straightforward. NHO NHNH O F NOFNOFN Oa)NHO NH2H2N Scheme 4.37. Reagents and conditions: KHMDS, THF, -78 ºC, 1 h, followed by (S)-209, 78 ºC to rt, 12 h, 26%. The propyl groups in (S,S)-210 simplified the analysis of the 1H (Figure 4.10) and 13C NMR assignment, relative to the diallyl product 195. (S)-188 (S)-209 154 (S)-209 (S,S)-210 Chapter 4 104 4.0 1.0 2.03.05.06.07.0 8.0 Figure 4.10. 1H NMR (300 MHz, CD3OD) spectrum of acridone (S,S)-210. This acridone coupling procedure was then carried out with the opposite enantiomer of ester (R)-188 to furnish (R,R)-210 acridone in 30% yield. The N,N-dipropyl derivatised acridones (S,S)- and (R,R)-210 were again treated with phosphorus oxychloride followed by the addition of aniline 154. These reactions went smoothly to give the (S,S)- and (R,R)- enantiomers of acridine 212 (Scheme 4.38). 2 3 41 NH109 5 678O NHNH O F NOFN Chapter 4 105 NHO NHNH O F NOFN N NHNH O F NOFN HN Na), b)N Clvia Scheme 4.38. Reagents and conditions a) POCl3 (neat), rt, 24 h & b) N,N-dimethylaminoaniline 154 (20 eq), CHCl3, reflux, 5 h, 21%. In each case purification was achieved by silica gel chromatography, furnishing the (S,S)- and (R,R)-212 enantiomers in reasonable yields (21% and 25% respectively) over two steps. The 1H NMR spectrum of (R,R)-212 is shown in Figure 4.11 by way of example, with each resonance clearly resolved. (S,S)-210 211 (S,S)-212 Chapter 4 106 4.0 1.0 2.03.05.06.07.0 8.09.0 12.4 1.04 0.99 2.06 2.09 2.02 2.071.95 8.00 7.9910.4 Figure 4.11. 1H NMR (400 MHz, CD3OD) spectrum of trisubstituted acridine (R,R)-212. The two enantiomers of the propyl functionalised acridines 212 (Figure 4.12) have also been submitted for binding assays to human telomeric DNA. NNH NHOFN O F NHN NNNH NHOFN O F NHN N Figure 4.12. Selectively fluorinated enantiomers of BRACO-19 analogues. (R,R)-212 (S,S)-212 2 3 41 N109 5 678 NHNH O F NOFN HN1213 14 15 N Chapter 4 107 4.9.2 - 1H-19F HOESY analysis of (S,S)-212 To investigate the solution conformation of the α-fluoroamide moiety in 212 a 1D 1H-19F HOESY NMR of (S,S)-212 was recorded. In this experiment, irradiation of the fluorine resonance strongly enhances the NH and the CHF resonances in the 1H NMR spectrum (Figure 4.13). This is consistent with the C–F and N–H bonds orientated close in space, as expected for the anticipated α-fluoroamide conformation (Figure 4.14). 4.0 5.08.09.0 10.0 7.011.0 6.0 ! Figure 4.13. HOESY analysis of (S,S)-212. Lower NMR– 1H NMR (500 MHz, CDCl3) of acridine 212 with broad peaks as a result of CDCl3 and Top NMR – 1H NMR (500 MHz, CDCl3) recorded during selective irradiation of the 19F signal at -191.1 ppm. 23 4 1 NHN RNHOFR2N HHH Figure 4.14. Simplified representation of acridine 212 with arrows detailing the main NOE enhancements in the HOESY spectrum in Figure 4.14. R = propyl, R' = N,N-dimethylaminoaniline. (S,S)-212 Chapter 4 108 4.10 - Non-fluorinated BRACO-19 142a analogues The non-fluorinated compounds 213.HCl and 214 have not been previously synthesised and were required as reference compounds to compare the effects of selective fluorination on the stabilisation of quadruplex DNA (Figure 4.16). NHNNH NHOClH3N O NH3ClNNHNNH NHON O NN Figure 4.15. Non-fluorinated BRACO-19 analogues for comparative studies. Initially, bis-chloro amide 174 was prepared by refluxing acridone 154 in neat 3-chloropropionyl chloride 148 (Scheme 4.39). NHO NH2H2N NHO NHNH O ClOCla) Scheme 4.39. Reagents and conditions: 3-Chloropropionyl chloride (neat), 145 ºC, 3 h, 27% Acridone 174 proved difficult to isolate and purify, however gram scale reactions enabled the isolation of sufficient quantities for subsequent reactions. The resultant acridone 174 was then treated with either diallylamine (215) or dipropyl amine (216), and sodium iodide to furnish 140i and 140j respectively (Scheme 4.40). 154 174 213.HCl 214 Chapter 4 109 NHONH NHOCl O Cl NHONH NHOR2N O NR2a) Scheme 4.40. Reagents and conditions: for 140i a) Diallylamine, EtOH, 80 ºC, 3 h, 46% and for 140j a) dipropylamine, EtOH, 80 ºC, 3 h, 26%. The literature purification of similar acridone compounds calls for the recrystallisation from DMF and ethanol, however acridones 140i/j were much more reasonably purified by column chromatography, albeit with the need to pre-basify the column with triethylamine.191 In the event both 140i/j were isolated in acceptable yields (46% & 26% respectively). Treatment of acridones 140i/j with neat phosphorus oxychloride at reflux afforded the corresponding 9-chloro intermediates 216 and 217 respectively. Again the chlorides were not purified but were reacted immediately, following a brief work-up, with aniline 154 (Scheme 4.41). Both of the trisubstituted acridines 219 and 214 were isolated in acceptable yields (37% & 29% respectively) following purification by column chromatography. 140i R - allyl, 46% 140j R - propyl, 26% 174 Chapter 4 110 NHONH NHOR2N O NR2NCl NNH NHOR2N O NR2HN Na), b) via Scheme 4.41. Reagents and conditions: For both a) POCl3 (neat), 140 ºC, 3 h, 37% & b) N,N-dimethylaminoaniline, CHCl3, 80 ºC, 4 h, 29%. The purification of 219 and 214 by column chromatography proved to be less straightforward than with the fluorinated analogues 206 and 212. Acridines 219 and 214 were loaded on the column as their HCl salts and the column was flushed with chloroform and methanol (95:5) to remove any impurities. Addition of triethylamine to the eluant neutralised the salts, enabling the isolation of analytically pure tetra-allyl protected 219. Repeated chromatography was required to provide a pure sample of the propyl functionalised acridine 214. 219 R - allyl, 37% 214 R - propyl, 29% 140i R - allyl 140j R - propyl 217/218 Chapter 4 111 Finally, removal of the allyl groups in 219 to furnish diamine 213 was achieved following the standard procedure using Pd2(dba)3 and thiosalicylic acid (Scheme 4.42). NNH NHON O NHN N NNH NHOH2N O NH2HN Na) Scheme 4.42. Reagents and conditions: Pd2(dba)3 (10 mol%/allyl group), dppb, thiosalicylic acid, THF, reflux followed by HCl, 17%. Purification of acridine 213 by reverse phase chromatography was particularly problematic and amine 213 could only be isolated in milligram quantities following the addition of formic acid to the eluant. Thus 213 was isolated as its formic acid salt and in a poor yield (17%). However, sufficient material was prepared for comparative biological assessment with the fluorinated analogues (S,S)- and (R,R)-208. 4.11 - Crystallographic assessment To date, no suitable co-crystals of the fluorinated trisubstituted acridines have been achieved for X-ray diffraction despite repeated attempts to identify good crystallisation conditions. Crystals of (R,R)- and (S,S)-212 with the human telomeric sequence have formed, however with poor morphology. Diffraction of these crystals has so far only provided low-resolution data (>6 Å). However this preliminary data demonstrated that the propyl functionalised acridine (S,S)-212 does form a quadruplex fold with the DNA as highlighted by the characteristic π-π stacking observed in the diffraction pattern. Studies have now focused on investigating conditions with the O. nova quadruplex 219 213 Chapter 4 112 DNA sequence, which generally accommodates a wider variety of ligand substituents.182 Whole cell based assays and in vitro analysis by FRET are currently underway with our collaborators at the UCL School of Pharmacy in London. 4.12 - Conclusions The chapter has demonstrated the successful synthesis of (S,S)- and (R,R)- stereoisomers of fluorinated 208 and 212. In addition, the complementary non-fluorinated ligands 213 and 214 were also prepared for assessment by X-ray crystallography and in vitro based assays. Chapter 5 Studies on the selective fluorination of dipeptides 5.1 - Introduction In this chapter an alternative approach for the synthesis of α-fluoroamides is explored. The fluorination of β-alcohol amino esters, as discussed in Chapters 1 and 4, with sulfur trifluoride reagents provides an efficient method for the generation of an array of fluorinated building blocks. The reaction of DAST 32 with peptides bearing the β-hydroxy functionality, such as in 220, is commonly employed for the synthesis of oxazoles (222) and oxazolines (223) (Scheme 5.01) without incorporation of fluorine. The approach allows the synthesis of these heterocycles with various degrees of functionalisation and structural diversity.203 Such activation with DAST 32 has even been employed in the synthesis of the telomerase inhibitor telomestatin 134 (Chapter 2.9.1), which contains repeating oxazole units.162 R'' NHO CO2R'OHR DAST R'' NH+O- CO2R'OR SF2N(Et)2O NR CO2R'R'' BrCCl3, Base O NR CO2R'R'' Scheme 5.01. Synthesis of oxazoles and oxazolines with sulfur trifluoride reagents. R,R', R'' = H, alkyl, aryl. 220 222 223 221 Chapter 5 114 However, the use of dehydroxyfluorination reagents on α-amino-β-hydroxyamides such as 224 has not been reported. At the outset the aim was to explore the scope and potential of the fluorination of dipeptides 224 with a terminal serine residue (Scheme 5.02). NH R' O OONR2HO NH R' O OONR R NH R' O OOR2N NH R' O OONR2F NH R' O OOR2N FN SF3 α−fluorinationβ−fluorination Scheme 5.02. Fluorination pathway for β-hydroxy amines bearing the amide functionality. The double bond character could support the intermediate or facilitate in the formation of an α-carbocation leading to racemisation. R = allyl/benzyl & R' = H, alkyl, aryl. The influence of the amide bond on the opening of the aziridinium ring in intermediate 225 and its effect on the distribution of α- and β-fluorinated regioisomers 227 and 228 was a key consideration (Scheme 5.02). The amide resonance in 226 may also stabilise a carbocation at the α-position and promote racemisation in the α-fluorinated product 227. 5.2 - Carboxylic acid synthesis for peptide couplings The serine derivative (S)-229 was first prepared (Scheme 5.03). N,N-Diallylation of L-serine methyl ester 165 was achieved by reaction with allyl bromide as summarised in Scheme 5.03. This proceeded straightforwardly generating gram quantities of alcohol (S)-187. The terminal hydroxyl group was then protected as its TBDMS ether (S)-229 in preparation for peptide coupling (Scheme 5.03). 224 227 225 226 228 Chapter 5 115 a) ONO OSiONHO OONH3ClHO O b) Scheme 5.03. Reagents and conditions: a) Allyl bromide (2.5 eq), K2CO3 (4.0 eq), CH3CN, 60 ºC, 16 h, 57%; b) TBDMSOTf (1.1 eq), Et3N (5.0 eq), CH2Cl2, rt, 16 h, 83%. Methyl ester (S)-229 was hydrolysed using lithium hydroxide in methanol to give carboxylic acid (S)-230 (Scheme 5.04). The purity of the extracted product was sufficient to be used directly in peptide coupling reactions. OONTBDMSO OHONTBDMSOa) Scheme 5.04. Reagents and conditions: a) LiOH (4.0 eq), CH3OH:THF:H2O (3:1:1), 24 h, 90%. 5.3 - Peptide couplings Peptides 323a-c were prepared by coupling to a variety of commercially available amino acid methyl esters 231a-c (Scheme 5.05). OHONTBDMSO NHO OORNTBDMSONH3ClOOR+ a) Scheme 5.05. Reagents and conditions: a) T3P® (1.5 eq), amine 231a/b/c (2.0 eq), diisopropylethylamine (4.0 eq), CH2Cl2, 0 ºC to rt, 1-12 h, 79-85%. Both HBTU and T3P® were effective for the coupling reactions, although from a practical point of view, T3P® offered an advantage over HBTU. This was because the hydrolysis product is water-soluble and could be removed along with excess unreacted (S)-165 (S)-187 (S)-229 231a-c (S)-229 (S)-230 (S)-230 232a - 81%, R = CH2Ph 232b - 85%, R = CH3 232c - 79%, R = CH(CH3)2 Chapter 5 116 amine by an acidic wash upon work-up (Scheme 5.05). More generally, T3P® mediated peptide couplings proceed in high yields with low levels of epimerisation.204 The synthesis of the L-phenylalanine dipeptide 232a proved amenable to scale-up and was isolated in 81% yield. Dipeptides 232b/c derived from L-alanine and L-valine, were synthesised in 85% and 79% yields respectively. All three dipeptides 232a-c, were isolated with good diastereoselectivity, with only a very low level of epimerization at the α-carbon (95:5 dr). Fluorination of 232a-c required that the silyl protecting group be removed to release alcohols 224a-c. Initial silyl ether deprotection of 232a with TBAF in THF cleaved the silyl ether to yield 224a but also resulted in the hydrolysis of the methyl ester to give 233. This gave rise to an unexpected cyclisation to the cyclic dimer 234 (Scheme 5.06). NHO OONTBDMSO Ph NHO OHONHO Ph NHO OONHO Ph HNO ONH O NN O O OPh Ph++a) Scheme 5.06. Reagents and conditions: a) TBAF (2 eq), THF, 0 ºC to rt, 2 h. Both carboxylic acid 233 and cyclic dimer 234 co-eluted during purification with the structure of the cyclic depsipeptide confirmed by single crystal X-ray crystallographic analysis (Figure 5.01). The formation of 234 is clearly a dimeric condensation, although the detailed sequence of events is not clear. 232a 233 234 224a Chapter 5 117 Figure 5.01. Crystal structure of the 14-membered cyclic (R,R) 234. Structure A shows two cyclic dimers in the unit cell. The nature of the allyl groups results in some disorder. Structure B represents one ring with the peripheral benzyl and allyl groups removed so that both amide and ester bonds are clearly observed. In order to circumvent the problematic methyl ester hydrolysis and cyclisation, the silyl ether was removed with an acetic acid buffered TBAF solution in dry THF, at rt (Scheme 5.07). NHO OORNTBDMSO a) NHO OORNHO Scheme 5.07. Reagents and conditions: a) TBAF (4.0 eq, 1 M THF soln.), AcOH (5.0 eq), THF, rt, 12-24 h, 73-88%. This gave an excellent conversion to the free alcohols 224a-c (73-88%), which could then be purified by chromatography in a straightforward manner. Importantly, no further epimerisation was observed in the products. A B 232a-c 224a - 73%, R = CH2Ph 224b - 88%, R = CH3 224c - 75%, R = CH(CH3)2 Chapter 5 118 5.3.1 - Fluorination reactions of dipeptides with DAST 32 The fluorination of L-phenylalanine derived dipeptide 224a with DAST 32 was initially explored (Scheme 5.08). NHO OONHO Ph NHO OONF PhNHO OOFN Pha) + Scheme 5.08. Reagents and conditions: a) DAST 32, THF, 0 ºC, 1 h, 81% (total fluorinated yield). TLC analysis indicated that the starting material was consumed after 1 h stirring at 0 ºC showing a similar reactivity to the ester substrates 119/187 in Chapter 4. Analysis of the product mixture by 19F and 1H NMR indicated a ratio of 60:40 for the α- and β-fluorinated products (227a:228a). The α- and β- regioisomers could be distinguished by their distinctive coupling patterns in both the 19F and 1H NMR spectra and chemical shift in the 19F NMR spectrum. This is a significant change in selectivity relative to the fluorination of esters 119/187 (Chapter 4). In those cases the α-products were favoured in a 95:5 (α:β) ratio (Table 5.1). However, despite the poorer selectivity in 224a, the reaction proceeds cleanly, with the 1H NMR and 19F NMR spectra correlating accordingly. OONR2HO ONR2FOOFR2N +DAST O Entry R α-NMR ratio β-NMR ratio 1 Benzyl (119) 95 (120) 5 (160) 2 Allyl (187) 95 (188) 5 (189) Table 5.1. Fluorination selectivity ratio for esters, 119 and 187 (Chapter 4.3/4.7). The two fluorinated regioisomers 227a and 228a were readily separated by column chromatography. This is in contrast to the fluorinated esters 119/187, which could not be easily separated. Both 227a and 228a were isolated in 43% and 38% representing an 228a - 38% 224a 227a - 43% Chapter 5 119 efficient overall transformation. The diastereomeric ratio of the α-fluorinated product 227a was determined to be 95:5 from the 1H and 19F NMR spectra. This suggests that the reaction has good stereocontrol and does not proceed via substantial carbocation character at the α-position. The significant shift from high α-selectivity in esters 119/187 to a poorer α-selectivity for amide 224a was unexpected. Accordingly, dipeptides 224b/c were explored to assess the influence of less bulky amino acid side chain substituents on the outcome of the fluorination. NHO R'NR2HO NHO R'NR2FNHO R'FR2N +a) O OO OO O + Scheme 5.09. Reagents and conditions: a) DAST 32, THF, 0 ºC, 1 h, 47-54% (total fluorination). R = allyl The reactions of 224b/c with methyl and isopropyl side chains respectively, with DAST 32 demonstrated a shift towards the β-fluorinated product (Table 5.2, Entries 2/3) with an α:β ratio of 35:65 and 25:75 for 227b/228b and 227c/228c respectively. The fluorination ratios for 224a-c are summarised in Table 5.2 along with their respective diastereomeric ratios. Entry Substitution α:β ratio dr 1 Phe, 224a 60:40 95:5 2 Ala, 224b 35:65 85:15 3 Val, 224c 25:75 92:8 Table 5.2. Fluorination ratios and diastereomeric ratios of products 224a-c. An overlay and progressive offset display of the 19F{1H} NMR spectra of the crude reaction mixtures illustrates the α:β fluorination ratio of the three dipeptides 227/228a-c products (Figure 5.02). 224b 224c 227b - 18%, R' = CH3 227c - 12%, R' = CH(CH3)2 228b - 36%, R ' = CH3 228c - 35%, R' = CH(CH3)2 Chapter 5 120 -190 -200 -210 -220 -230 ppm - 1 9 1 . 3 - 2 2 7 . 6 - 1 9 0 . 7 - 2 2 7 . 6 - 1 9 0 . 4 Phe Ala Val 1 0.8 2.5 3 1 1 - 2 2 7 . 1 ] ] ] ] ] ] -195 -205 -215 -225-185 α-Fluoro β-FluoroNHO OOFN NHO OOFN NHO OOFN Ph NHO OONF NHO OONF NHO OONF Ph19F{1H} NMR Figure 5.02. Overlay of the 19F{1H} NMR (376 MHz, CDCl3) spectra of the crude products from the fluorination of amides 224a-c. These fluorination reactions proceed via an aziridinium intermediate with subsequent fluorination at either the α- or β-positions of 236a-c (Scheme 5.10).95 It is assumed that the β-product results from nucleophilic fluoride attack at the β-position of aziridinium 236a-c, rather than by direct SN2 attack of fluoride at the activated alcohol 235a-c. It is not obvious that the β carbon will be significantly more electropositive in the aziridinium intermediate for the amides 224a-c over that of the esters 119/187. However, the product ratios observed for the fluorinations from 224a-c suggest an increased β-reactivity (Scheme 5.10). Chapter 5 121 NO R'NO HR R NO R'NO HRRSNF Fk3NO R'NHO HR RSNF F NO R'NF HR R N O R'NR R NO R'R2N Hβ attack α attackF Hk2 k1 k-3 k5 k4 k5'ActivationSN2 SNi SN2SN2 Scheme 5.10. Two proposed reaction pathways that could be occurring to explain the product distribution. R = allyl, R'= Phe (a), Ala (b), Val (c). For the aziridinium intermediate 236a-c to form, terminal alcohol 224a-c requires to be activated (235a-c) such that the lone pair of the α-amine nitrogen can undergo SNi attack at the β-carbon. The rate at which this intramolecular SNi reaction (k4) occurs is governed by the availability of the nitrogen lone pairs of 235a-c (Scheme 5.10). This will be compromised if the lone pairs form a hydrogen bond to the N–H of the amide. Such 5-membered hydrogen bonded rings are observed in the solid state within this structural class.205 Therefore, the rate of aziridinium formation (k3) over direct SN2 substitution (k2) at the β-carbon will be affected by this hydrogen bond (Scheme 5.10). On the other hand, if the nitrogen lone pair is free to form an aziridinium intermediate 236a-c - as in the case of esters 119/187 - then the rate of this intramolecular reaction (k4) will be significantly faster than the intermolecular fluoride ion attack (k2) (Scheme 5.10). If k2>>k3 then β-fluorination dominates, if k2 ≥ k3 then this results in a mixture of α- and β-products and if k3>>k2 then α-fluorination dominates. Fluoride attack of aziridinium intermediate 236 (Scheme 5.10) can either be at the α- or β-position (k5 or k5′). Based on the observation with esters 119/187, k5′ is significantly faster than k5 and so α-fluorination will dominate as a result of the more electropositive α-carbon. 224a-c 227a-c 228a-c 235a-c 235'a-c -c 236a-c Chapter 5 122 Therefore, in the case of the L-phenylalanine dipeptide 224a this analysis suggests that the benzyl group dictates a less favourable hydrogen bonding interaction than the methyl and isopropyl groups in 224b/c. By comparing the chemical shift difference between the amide NH 1H NMR resonance of 224a-c, a correlation between the fluorination ratios and the relative chemical shifts is apparent (Figure 5.03). A higher NH chemical shift and therefore a relatively stronger hydrogen bond, appears to correlate with a higher proportion of the β-product in the L-alanine and L-valine dipeptides. NON HNON F H R' O OR' O O NONHO H R' O OF δH: 7.75 (Phe, R' - CH2Ph) 7.83 (Ala, R' - CH3) 7.87 (Val, R' - CH(CH3)2)δH: 7.75 (Phe, R' - CH2Ph) 7.88 (Ala, R' - CH3) 7.91 (Val, R' - CH(CH3)2)δH: 6.94 (Phe, R' - CH2Ph) 7.03 (Ala, R' - CH3) 6.97 (Val, R' - CH(CH3)2) Amide-NH 1H NMR ΔδH = ~ 0.03ΔδH = ~ 0.875-memberedHydrogen bond5-memberedHydrogen bondno 5-memberedHydrogen bond Figure 5.03. Analysis of the NH 1H NMR resonance between non-fluoro, α and β fluorinated products detailing the chemical shift disparities. Chemical shifts are reported in ppm and are quoted relative to CDCl3. The analysis of the amide NH chemical shifts in the 1H NMR spectrum in both the dipeptide starting materials 224a-c and α/β fluorinated products 227/228a-c provides further evidence of hydrogen bonding. In the α-fluorinated products 227a-c, the NH resonance is approximately 0.9 ppm upfield relative to the respective starting materials 224a-c and β-fluorinated products 228a-c. Without the hydrogen bonding in these compounds, the direction of the chemical shift change of the α-fluorinated products may appear counter intuitive based on the electronegativity of the α-C–F bond. Thus these observations further support the evidence for a hydrogen bond in the alcohols 224a-c resulting in a 5-membered ring. 224a-c 228a-c 227a-c Chapter 5 123 To test this hypothesis further, the effect of temperature on the regioselectivity of the reaction was explored. It was anticipated at lower temperatures, that the selectivity for the β-fluorination in 224a would increase relative to the α-fluorinated. Carrying out the reaction at -78 ºC gave rise to low levels of fluorination, however, the 19F NMR spectrum of the reaction mixture after 5 h showed only the β-fluorinated product. The reaction was then conducted at -20 ºC. After 2 h the 19F NMR spectrum showed a greater level of the α-fluorinated product, with an α:β ratio of 75:25. At ambient temperature, a black solution is immediately formed upon the addition of DAST 32. After 20 minutes the 19F NMR spectrum of the crude mixture revealed an unexpected shift towards β-fluorinated product in a ratio of 50:50. The temperature profile was more complex than expected, but indicated a tendency to β-product at low temperature (-78 ºC). 5.3.2 - Dipeptide conformation in 227a 19F NMR analysis of α-fluorinated products 227a-c, shows a through space 4JFH coupling constant, between the fluorine and the amide proton (Figure 5.04). These coupling constants are: 4.0, 3.7 and 4.3 Hz, corresponding to compounds 227a, 227b and 227c respectively. This is indicative of an NMR solution structure with the expected ~180º relationship between the C–F bond and the amide (Chapter 1.11.2).77 Chapter 5 124 -190.60 -190.65 -190.70 -190.75 -190.80 -190.85 - 1 9 0 . 5 9 6 - 1 9 0 . 6 0 4 - 1 9 0 . 6 4 6 - 1 9 0 . 6 5 9 - 1 9 0 . 6 5 5 - 1 9 0 . 6 6 8 - 1 9 0 . 7 0 1 - 1 9 0 . 7 1 0 - 1 9 0 . 7 1 9 - 1 9 0 . 7 5 2 - 1 9 0 . 7 6 1 - 1 9 0 . 7 6 6 - 1 9 7 . 7 7 4 - 1 9 0 . 8 1 7 - 1 9 7 . 8 2 5 - 1 9 0 . 6 0 - 1 9 0 . 6 0 - 1 9 0 . 6 5 - 1 9 0 . 6 5 - 1 9 0 . 6 6 - 1 9 0 . 6 7 - 1 9 0 . 7 0 - 1 9 0 . 7 1 - 1 9 0 . 7 2 - 1 9 0 . 7 5 - 1 9 0 . 7 6 - 1 9 0 . 7 7 - 1 9 0 . 7 7 - 1 9 0 . 8 2 - 1 9 0 . 8 2 -220 -250 -240-230 -210 -200-190 NO HFN PhOOH4JFH = 4.0 Hz Figure 5.04. 19F NMR (470 MHz, CDCl3) of 227a detailing the coupling pattern of the CHF resonance. 5.4 - Preparation of α-amino acid N–H and N–CH3 amide derivatives In order to study the effect of hydrogen bonding on the α/β fluorination ratio, incorporation of an N-methyl group on the amide nitrogen was explored. This removes any possibility of a hydrogen bond and may show an increase in α-fluorination selectivity. An initial effort to synthesise N-methylated 238 proved unsuccessful, even though a number of reported literature conditions were attempted (Scheme 5.11).206,207 In our hands, the treatment of Boc protected L-phenylalanine with various bases and methyl iodide failed to furnish the desired N-methylated product 238 in suitable quantities and only complex mixtures resulted. NBocOOPhVarious conditionsNHBocOOPh Scheme 5.11. Attempted synthesis of N-methyl phenylalanine. 237 238 NO HFN PhOOH4JFH = 4.0 Hz Chapter 5 125 An alternative strategy was taken for the syntheses of appropriate N-methyl analogues and was achieved with secondary benzylamines 239a/b. Starting from amine 239a and following the same coupling procedure with T3P®, amide 240a was synthesised in good yield (Scheme 5.12). a)ONTBDMSO OH + ONTBDMSO NR PhHNR Ph Scheme 5.12. Reagents and conditions: a) T3P® (1.5 eq), amine 239a/b (2.0 eq), diisopropylethylamine (4.0 eq), CH2Cl2, 0 ºC to rt, 1-12 h. Amide 240a showed a significant cis-trans isomer ratio in the 1H NMR spectrum with a 75:25 preference for the trans isomer (Scheme 5.13). Isomerisation in this class of compound arises from the higher steric demand of the amide N′–CH3 over that of an N–H in 232a-c. The identification of the resonances attributed to both the cis and trans conformers in 240a was possible by analysis of the 1H-1H COSY spectrum. The resonance of the N′-CH3 corresponding to the trans isomer was 0.2 ppm downfield of the cis resonance, with a ratio of 25:75 (cis:trans). This ratio was also observed for the α-CH3 resonance with the trans isomer 0.07 ppm upfield from that of the corresponding cis isomer. In 1D and 2D nOe experiments, irradiation of the α-CH3 (1.47 ppm) of the trans isomer resulted in a transfer of magnetization to both cis and trans conformers. This highlights that the two conformations are interconverting on the NMR timescale, complicating the unambiguous assignment of the cis and trans isomers. NO Me PhNR2TBDMSO NONR2TBDMSO PhMe Scheme 5.13. Cis-trans isomerisation of the tertiary amide 240a. The trans conformation is preferred by 3:1 as indicated from integration of the 1H NMR signals. R = allyl. 230 240a, 82%, R = CH3 240b, 78%, R = H 239a, R = CH3 239b, R = H cis - 240a trans - 240a Chapter 5 126 For a direct comparison of the fluorination ratio between N′-Me and N-H amides, amide 240b was synthesized from α-methylbenzylamine 239b (Scheme 5.12). The amide 240b, like the dipeptides 232a-c, did not show any obvious cis-trans isomerisation by 1H NMR. The silyl protecting groups of amides 240a/b were removed using acetic acid buffered TBAF (Scheme 5.14), enabling the isolation of alcohols 241a/b in good yields. The cis-trans isomer ratio in amide 241a was also 25:75. ONTBDMSO NR Ph a) ONHO NR Ph Scheme 5.14. Reagents and conditions: b) TBAF (4.0 eq, 1 M THF soln.), AcOH (5.0 eq), THF, rt, 12-24 h. 5.4.1 - Fluorination Reactions of amides 241a/b With the N′-alkylated substrates in hand, alcohols 241a/b were treated with DAST 32 under the same conditions to that used previously, such that the α:β ratios could be directly compared (Scheme 5.15 & 5.16). NO PhFNNO PhNHO NO PhNF+a)Me Me Me Scheme 5.15. Reagents and conditions: a) DAST 32 (1.1 eq), THF, 0 ºC, 1 h. NO PhNHO H NO PhNF HNO PhN F Ha) + Scheme 5.16. Reagents and conditions: a) DAST 32 (1.1 eq), THF, 0 ºC, 1 h. 242b - 70% (ratio) 240a/b 241a, 77%, R = Me 241b, 78%, R = H 243b - 30% (ratio) 241b 241a 242a - 81% 243a - 0% Chapter 5 127 Direct analysis of the reaction mixture of N–H amide 241b, which was still capable of forming a 5-membered intramolecular hydrogen bond, showed a 70:30 (α:β) bias for fluoride attack at the α–carbon (Figure 5.05, lower NMR). This reaction proceeded with a diastereomeric ratio of 92:8. Although the bias has shifted, it represents a significantly lower selectivity compared to esters 119 and 187. This shift back to an α-fluorination preference suggests that the mechanism for selectivity may be more complex than anticipated. By contrast, analysis of the α:β ratio from N′-methylated amide 241a was 99:1 in favour of α-fluorination (Figure 5.05, top NMR). However, analysis of the crude 1H NMR indicated a diastereomeric ratio of 89:11. This represents a drop in diastereoselectivity for the fluorination of the N-substituted substrate 241b compared to 224a-c, suggesting that the reaction may have some SN1 character. The 19F NMR of α-fluorinated 242a, shows cis and trans products with resonances of -184.6 and -186.8 ppm respectively. -190 -200 -210 -220 -230 ppm -195 -205 -215 -225-185 -335 - 1 9 0 . 6 - 1 8 6 . 8 - 1 8 4 . 6 - 2 2 8 . 3 cis - 25% trans - 75% α-Fluoro -190 -200 -210 -220 -230 ppm -195 -205 -215 -225-185 -335 NO CH3FN PhNO CH3FN Ph NO CH3NF Phβ-FluoroNot observed β-FluoroNHONF PhNHOFN Phα-Fluoro Figure 5.05. Overlay of the crude 19F{1H} NMR (376 MHz, CDCl3) from the fluorination of 241a/b with DAST 32. 242b 243b 243a 242a Chapter 5 128 The high fluorination selectivity for the N′-Me substrate for the α-product 242a is greater than that observed for the fluorination of the amino esters 119 and 187 (Table 5.1). This is the first example of selective α-fluorination of an N'-substituted amide by deoxyfluorination. The difference in fluorination ratio between the N′-substituted amide 241a and amide 241b is consistent with the involvement of hydrogen bonding influencing the product profile. It was possible to observe spin polarization in the 1D HOESY (19F-1H nOe) following selective irradiation of the fluorine signal in the two isomers of 242a.59 Selective irradiation of the cis isomer at -184.6 ppm in the 19F NMR (Figure 5.06, A) showed an enhancement of the trans Cα+1–H proton. 7 ppm 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 2.57.5 non selective selective major selective minor Friday, 27 January 2012 Figure 5.06. 1D HOESY (500 MHz, d6-DMSO) of 242a. A - Selective cis C–F irradiation at -184.6 ppm in 242a; B - Selective trans C–F irradiation at -186.8 ppm in 242a; C - Non-selective dual irradiation of cis and trans 242a; D - standard 1H NMR. C B A D NO PhMeN F Me NO MeN FH MeH PhHOSEYCorrelation HOSEYCorrelationMajor Irridation Minor Irridationtrans C–F irradiation HOSEY correlation cis C–F irradiation HOSEY correlation Cα+1–H N'–CH3 Chapter 5 129 This appears counterintuitive, however, this enhancement is observed as the amide is equilibrating on the NMR timescale, where during !m the spin polarization and enhancement is ‘carried’ over to the trans isomer. These is a strong enhancement of the allyl resonances, which overwhelms the cis Cα+1–H resonance. Such an enhancement of the trans Cα+1–H is not observed when selective irradiation of the signal at -186.8 ppm is carried out (Figure 5.06, B). An averaging of the enhancement is observed when a non-selective irradiation of the fluorine is conducted. The dynamic nature of the cis-trans isomerisation in 242a clearly complicates the assignment. 5.5 - Extending the applicability to useful N-substituted amides In order to extend the potential of this methodology, the N-allylamide 244 was prepared carrying non-orthogonal protecting groups. This protection strategy was designed to enable a one-pot deprotection to furnish a synthetically useful dipeptide for further peptide coupling through the free β-amine in 246 (Scheme 5.17). NON F OO NHO+H3N F OO NONHO PhOOFluorination OnepotDeprotection Further manipulationPh Ph Scheme 5.17. Synthesis and deprotection strategy to peptide 246. The synthesis of N-allyl amine 247 was achieved by treating methyl ester 231a with allyl bromide in DMF and diisopropylethylamine (Scheme 5.18). The addition of the reagents at 0 ºC and slow warming to room temperature enabled isolation of mono-allylated 247, with no dialkylation observed. This method was efficient and gave 244 245 246 Chapter 5 130 rise to sufficient quantities of allyl amine 247 for peptide coupling experiments to be conducted. H2N OOPh NH OOPha).HCl Scheme 5.18. Reagents and conditions: a) Allyl bromide (5.0 eq), diisopropylethylamine (4.0 eq), DMF, 0 ºC to rt, 24 h, 38%. Initial efforts to prepare the allyl protected tertiary amide 248 utilising the conditions employed so far with T3P®, only provided a trace of the desired tertiary amide 248 (Table 5.3, Entry 1 & 2). Other peptide coupling reagents such as CDI, EDCI and HATU in either DMF or CH2Cl2 at room or elevated temperatures were also unsuccessful (Table 5.3, Entries 3-7 It is clear from these unsuccessful conditions that the N-allylated nitrogen is poorly nucleophilic as a result of the high steric hindrance from both the allyl and benzyl moieties in 230. OHONR2TBDMSO NO OONR2TBDMSONH OO+ a)Ph Ph Entry Reagent Solvent Temp (ºC) Product 1 T3P® CH2Cl2 20 <5% (48 h) 2 T3P® DMF 20 Trace 3 HATU CH2Cl2 20 No 4 HATU DMF 20 No 5 HATU DMF 60 No 6 CDI CH2Cl2 20 No 7 EDCI CH2Cl2 20 No 8 PyBrop CH2Cl2 20 No 9 PyBrop DMF 20 No 10 PyBrop DMF 60 No 11 SOCl2 CH2Cl2 20 Multiple Table 5.3. Attempted peptide coupling conditions between 230 and 247. 247 230 231a 247 248, R = Allyl Conditions: see below Chapter 5 131 Entries 1 - 7 (Table 5.3) were also explored adding a catalytic amount of DMAP, however, this failed to generate the desired amide 248. Alternative and more tailored peptide coupling reagents were explored. Attempts with the phosphonium salt, PyBrop208 249 (Figure 5.07) were also unsuccessful in the preparation of the coupled amide (Table 5.3, Entries 8-10). In each case starting materials were recovered from the reaction. P+BrN NN PF6-PyBroP Figure 5.07. Tailored phosphonium salt for peptide coupling with sterically hindered substrates. A final approach to prepare 248 was attempted using the acid chloride of 250 (Table 5.3, Entry 11) (Scheme 5.19). OHONTBDMSO FONTBDMSO Cl ONTBDMSO OMeONTBDMSO N ONTBDMSO PhOOa)Cyanuric Fluoride b) Scheme 5.19. Reagents and conditions: a) SOCl2, CH2Cl2, DMF (cat.), rt, 1 h followed by the addition of 247; b) CH3OH quench. Acid chloride 250 was prepared by treatment of the carboxylic acid 230 with thionyl chloride in CH2Cl2, and a catalytic amount of DMF. Its formation was confirmed by a methanol quench to furnish the corresponding methyl ester 229 (Scheme 5.19). Once the acid chloride was formed, secondary amine 247 was added. TLC analysis indicated 249 250 248 229 230 Chapter 5 132 that multiple products had formed after 1 h, and a complex mixture was observed by 1H NMR analysis. In light of these results, this route was not investigated any further. 5.6 - Tertiary allylamide from secondary dipeptides With the coupling between N-allyl amine 247, and carboxylic acid 230, failing to furnish the desired amide, attention turned to direct amide allylation (251 to 252). The tri-allylated dipeptide 232a, could clearly be accessed using an appropriate base with an allyl halide or by the use of a transition metal catalysed allylation reaction (Scheme 5.20). R O N R'R O NH R' BaseAllyl bromideorPd0Allyl alcohol Scheme 5.20. Base and metal mediated approaches for the allylation of amides. R/R' = allyl/alkyl Treatment of amide 232a with NaH in THF . (Table 5.4, Entry 1 and 2) in the presence of an excess of allylbromide resulted in a complex product mixture, of which, the β-alanyl 253 species was one of the major side-products as judged by crude 1H NMR (Scheme 5.21). 251 252 Chapter 5 133 NO OONTBDMSO PhNHO OONTBDMSO Ph Br Entry Base Equivalents Solvent Temp ºC Products 1 NaH 1 THF 20 Multiple 2 NaH 1 DMF 20 Multiple 3 KHMDS 1 THF -78 Yes 4 KHMDS 0.9 THF -78 Yes 5 KHMDS 1 THF -100 Yes 6 KHMDS 0.9 THF -100 Yes 7 LiHMDS 1 THF -78 Yes 8 KOtBu 1 THF 20 No 9 Cs2CO3 4 THF 20 Multiple 10 Cs2CO3 4 THF 60 Multiple 11 K2CO3 4 THF 20 Multiple Table 5.4. Reagents and conditions used for the attempted allylation of amide 248. NHO OONTBDMSO PhNH OOPh NH OOPh+b) a) NHO OON PhONTBDMSO ONTBDMSO Scheme 5.21. Reagents and conditions: a) NaH (1.0 eq), allyl bromide (4.0 eq), DMF, 20 ºC; b) KHMDS (1.0 eq), allyl bromide (4.0 eq), THF, -78 ºC, 1-16 h, 53%. 232a 248 232a 253 (R,R)-254 (R, S)-254 Conditions see below Chapter 5 134 When KHMDS and LiHMDS were used (Table 5.4, entries 3-7) at -78 ºC, the Cα+1 allylated product 254 was formed (Scheme 5.21). This product could also be observed even when a substoichiometric equivalent of KHMDS was employed at -78 ºC and -100 ºC (Table 5.5, entry 4 & 6). The Cα+1 product, an α-allyl phenylalanine dipeptide 254 was isolated as a diastereomeric pair, from the reaction with KHMDS and allyl bromide. The diastereomers could not be separated by column chromatography but were characterised as a mixture by 1D and 2D NMR. In the 1H NMR of 254, the loss of the Cα+1 proton resonance, and retention of the amide NH resonance were indicative of the formation of 254. This was confirmed by DEPT and HMBC analyses of 254, which identified the expected quaternary carbon. The fact that this is a 1:1 diastereomeric pair is consistent with α-proton abstraction and subsequent enolization. This diastereomer mixture was also observed when KtOBu was used as a base (Table 5.4, entry 8). From these observations it is clear that the pKa of the Cα+1 proton was similar or more acidic than that of the amide proton and that the bases were able to deprotonate this hindered proton, furnishing the Cα+1 allylated product. In an effort to circumvent this, the strong, hindered base tert-butyl P4 phosphazene 255 was explored (Figure 5.08). Developed in the late 90’s by Reinhard Schwesinger, tBu P4 255 has a MeCNpKBH+ of 42.7 resulting form the potential large charge delocalisation available upon protonation (Figure 5.08). This class of base is well documented in the synthesis of N′-benzylated peptides.209,210 PNN NNPNNN P NN NtBu P NN N Figure 5.08. The strong hindered base tBu P4 255. 255 Chapter 5 135 Initial treatment of 232 with of tBu P4 255 at -78 ºC in THF with allyl bromide again gave rise to multiple products, of which both the Cα+1 allylated product 254 and the desired N′-allyl amide 248 could be isolated (Scheme 5.22). NHO OONTBDMSO Ph OOPh+ N OOPh+a) ONTBDMSO ONTBDMSO NHOOPhONTBDMSO NH Scheme 5.22. Reagents and conditions: a) tBu P4 255 (0.9 eq), allyl bromide (4.0 eq), THF, -78 ºC, 16 h, 30-50%. The N′-allyl amide 248 was only isolated in a 10 % yield, however, interestingly, the Cα+1 allylated product 254 was isolated again with a low but significantly improved selectivity for one diastereomer (2:1 from 1H NMR). The observation that 254 was still formed, even when using a particularly hindered base, highlights the similar pKa of both protons. Using phosphazene base 255 with benzyl bromide in place of allyl bromide, three benzylated products could be isolated after chromatography. Analysis identified these as the Cα+1 benzyl 256, N′-benzyl 257, and benzyl ester 258 (Scheme 5.23) which were isolated in yields of 20%, 18% and 15% respectively. 232a (S, R)-254 (S, S)-254 248 Chapter 5 136 NHO OONTBDMSO Ph OOPh+ N OOPhPh+a) ONTBDMSOONTBDMSO NHOOPhPhONTBDMSO NH Ph Scheme 5.23. Reagents and conditions: a) tBu P4 (0.9 eq), benzyl bromide (4.0 eq), THF, -78 ºC, 16 h, 15-20%. N′-Benzylated amide 257 was formed in an improved yield (16%) compared to the N′-allyl amide 248 (10%). This protecting group is however not so attractive as deprotection of N′-benzyl amides involves the use of sodium metal in napthalene, a reagent combination incompatible with the functional groups in amide 257.197 The apparent hydrolysis of the methyl ester and subsequent carboxylate alkylation to generate an ester in 258, was not observed with allyl bromide. Optimisation of the reaction by increasing the scale and initial treatment of amide with tBu P4 at -100 ºC followed by gradual warming through to -78 ºC before the addition of allyl bromide, furnished amide 248 in a significantly improved yield of 53% (Scheme 5.24). This enabled sufficient quantities of the N′-allyl amide 248 to be isolated for subsequent transformations. NO OONTBDMSO PhNHO OONTBDMSO Ph a) Scheme 5.24. Reagents and conditions: a) tBu P4 (0.93 eq), allyl bromide (5.0 eq), THF, -100 ºC to -78 ºC to rt, 20 h, 53%. The 1H NMR spectrum of 248 clearly indicates two rotamers, designated here as the cis and trans allylic resonances for the N′-allyl group (Figure 5.09). The other resonances 232a 256 257 258 232a 248 Chapter 5 137 were less easy to assign with confidence, however, from Figure 5.09 it would appear that the trans rotamer predominates in a ratio of 95:5. 4.9 4.95.05.05.15.15.25.25.35.35.45.45.55.55.65.65.75.75.85.85.95.96.06.0ppm 6.0 ppm 5.9 5.8 5.7 5.6 5.5 5.4 5.3 5.2 5.1 5.0 4.9 1 H NMR Cis Trans Cis Trans Figure 5.09. 1H NMR (400 MHz, CDCl3) showing the allylic hydrogen spin system of 248 detailing the cis/trans resonances of the N′-allyl group. The cis-trans relationship in 248 is significantly less than that observed in 240a. It is likely that the steric demand of the N′-CH3 in 240a and the benzyl moiety are similar. Whereas in 248 the CH2 of the N′-allyl group is significantly less demanding compared to phenylalanine. Thus, the trans isomer in 248 is likely to be the more stable conformation. Silyl ether cleavage of amide 248 was achieved with acetic acid buffered TBAF to furnish alcohol 244 in good yield (Scheme 5.25). The nature of the cis-trans isomerism in the 1H NMR was similar to that of allyl amide 248. NO OONTBDMSO Ph a) NO OONHO Ph Scheme 5.25. Reagents and conditions: a) TBAF (4.0 eq), AcOH (5.0 eq), THF, rt, 12 h, 84%. trans - 248 cis - 248 NONTBSO PhOO NONTBSO O OPh 248 244 Chapter 5 138 5.6.1 - N-allyl amide dipeptide 244 fluorination with DAST 32 Treatment of 244 with DAST 32 gave the expected α-fluorinated rearranged product 245 and proved to be high yielding (73%) and also showed a high level of selectivity (Scheme 5.26), similar to that observed for fluorination of N′-methyl amide 241a. Direct 19F NMR analysis of the reaction showed <1% β-fluorinated product 259 (Figure 5.10). However, amide 245 was isolated in a diastereomeric ratio of 90:10 suggesting a little SN1 character, as observed with the N′-methyl amide 241a. Presumably this arises from the steric bulk of the amide and perhaps some stabilisation of a carbocation intermediate by the amide bond. NO OONHO Ph NO OOFN Pha) Scheme 5.26. Reagents and conditions: a) DAST 32, THF, rt, 1 h, 73%. Purification of 245 from residual starting material was easily achieved by column chromatography. A combination of 1D and 2D NMR techniques were used to assign the resonances corresponding to the major trans isomer, which was favoured by 90:10 to the cis isomer. 244 245 Chapter 5 139 -190 -200 -210 -220 -230 ppm 1.00 0.11 ] ] 0.88 ] 0.008 cis - 12% trans - 88% α-Fluoro β-Fluoro 99% : 1% - 2 2 7 . 1 - 1 8 7 . 2 - 1 8 5 . 4 NONF O OPhNOFN PhO ONOFN Ph O O Figure 5.10. 19F{1H} NMR (470 MHz, CDCl3) of tertiary amide 244 directly after reaction with DAST 32. High selectivity for the α- over the β-fluorinated product. Both cis and trans amide C–F resonances can be clearly observed, which were confirmed by VT. 5.7 - N-Allyl amide 245 deprotection The next stage in the synthesis required complete N-deallylation of 245 by metal catalysis (Scheme 5.27). NO PhOOFN Catalyst NHO PhOOF+H3N Scheme 5.27. Proposed one pot de-allylation with a metal catalyst for further functionalisation. Initially, deprotection of all three allyl protecting groups in one step was attempted following the procedure set out in Chapter 4. This involved Pd2(dba)3 as a catalyst and a stoichiometric amount of thiosalicylic acid in THF (Scheme 5.28). 245 246 259 trans-245 cis-245 Chapter 5 140 NO PhOOFN a) NO PhOOF+H3N Scheme 5.28. Reagents and conditions: a) Pd2(dba)3 (25 mol%), dppb (30 mol%), thiosalicylic acid (3.3 eq), THF, reflux, 5 h, 93%. This approach was only partially successful and led to the deprotection of the primary allyl groups, furnishing N-allyl amide 260. Increasing the reaction time resulted in decomposition of the starting material and there was no evidence that the N′-allylamide could be cleaved by this approach. The lack of reactivity of the N′-allyl amide moiety can be rationalised as illustrated in Scheme 5.29. Protonation of the allyl amine nitrogen by the thiosalicylic acid enables a π-allyl Pd complex to be formed. Without protonation, cleavage of the allyl groups is not possible. The partial double bond character of the amide bond and subsequent partial positive charge on the amide nitrogen is not sufficiently activating for this purpose and does not enable the π-allyl Pd complex to form. Thus the N′-allyl amide is less reactive than the N,N-diallylamine towards this particular catalytic system. R O- N+R' R O N R'Pd+Pd0 NSHCO2-SCOOH R ROH H2O R O NH R' Scheme 5.29. Proposed mechanism for the unsuccesful cleavage of the allylamide protecting group. The literature contains a wealth of methods for the removal of allyl ethers and amines, however there are few examples for the removal of N′-allyl amides.211–213 The successful reports all use the same principal involving a metal mediated isomerisation of an allyl amide (263) to the enamide 264 (Scheme 5.30), followed by an oxidative cleavage, rather than direct cleavage of the N′-allyl amide.214–217 200 261 262 197 199 198 245 260 Chapter 5 141 R O N R' R O N R' R O N R'+ Scheme 5.30. Metal mediated isomerisation of an allylamide to cis and trans enamide. R/R′ = various alkyl and aryl substituents. M = metal catalyst For this purpose the free amine of N-allyl amide 245 was protected as the tert-butyl carbamate ester 265 in aqueous dioxane with Boc2O and triethylamine (Scheme 5.32). NO PhOOFN NHO PhOOF+H3NNO PhOOFH2N a) NO PhOOFBocHNb) Scheme 5.32. Reagents and conditions: a) Pd2(dba)3 (25 mol%), dppb (30 mol%), thiosalicylic acid (2.9 eq), THF, 60 ºC, 3 h, 93%; b) Boc2O (1.3 eq), diisopropylethylamine (3.0 eq), aqueous dioxane (25% v/v), rt, 24 h, 71%. Purification of 265 was easily achieved by chromatography and the carbamate protected allyl amide 265 was isolated 71% yield. The 1H NMR was complex due to the NH carbamate proton coupling to the diastereotopic CH2 protons and also the CHF group, resulting in signal broadening. With the amine protected, it was now possible to screen an array of catalysts for allylamide deprotection. Initial efforts with a variety of catalysts proved unsuccessful however the reaction with 10 mol% of RuHCO(PPh3)2 in toluene did generate the isomerised enamide 266 (Scheme 5.33) as indicated by direct 1H NMR analysis of the reaction product. The crude enamide and catalyst were then committed directly for oxidative cleavage by RuCl3 and NaIO4 (Scheme 5.33). TLC indicated that enamide 266 was consumed and then hydrolysis of the putative N-formyl intermediate was attempted by treatment with aqueous NaHCO3 during work-up. However, this work 263 trans-264 245 246 260 265 cis-264 Chapter 5 142 up was insufficient to cleave the formyl group with the N-formylamide 267 isolated after chromatography. NO PhOOFBocHN NO PhOOFBocHNNO PhOOFBocHN OHa), b) Scheme 5.33. Reagents and conditions: a) RuHCO(PPh3)2 (10 mol%), toluene, reflux, 3 h; b) RuCl3, NaIO4, 1,2-dichloroethane, water, rt, 12 h, 59%. The distinctive chemical shift of the aldehyde proton (~9 ppm) and the carbonyl resonances observed in the 13C NMR supported the structure of 267. Subsequent hydrolysis of the formyl group in 267, to yield amide 268 was achieved by stirring in basic aqueous acetone (Scheme 5.34). NO OOPhFBocHN O NHO OOPhFBocHNNO OOPhFBocHNNO OOPhBocHN F a, b c Scheme 5.34. Reagents and conditions: a) NaHCO3 (1.0 eq), Na2CO3 (0.1 eq), acetone, water, 10 h, 46%. Purification was achieved by column chromatography to furnish 268 in a yield of 46%. Amide 268, did not show any cis isomer by NMR. The successful deprotection of the allyl amide in 265 demonstrates that it is possible to use this methodology to access synthetically useful peptides and demonstrates the applicability of selective fluorination of amides by DAST 32. 265 267 266 267 268 a) Chapter 5 143 5.8 - Conclusions This chapter reports the scope and limitations of the fluorination of dipeptides and amide analogues bearing the hydroxy amine motif. Initial studies with amides 224a-c demonstrated poor selectivity for α-fluorination, however for N′-alkylated amides 241a and 244 the selectivity for α-fluorination returns. As discussed throughout the Chapter, this can be attributed to a hydrogen bond between the dialkylated amine and the amide N–H. With this methodology, it was possible to synthesize an α-fluorinated N′-allylamide with high fluorination selectivity and satisfactory diastereomeric excess. It was also possible to demonstrate that this product could be successfully deprotected yielding a synthetically useful fluorinated dipeptide 268. Chapter 6 Future work 6.1 - Future Work for Chapter 3 Solving of the crystal structure with the hydroxy compounds (S,S)- and (R,R)-145 would supplement the data presented in Chapter 3.5. This would enable a direct comparison to be made between the hydroxy-145 and fluoro-144 ligands. To achieve this it will be required to reconsider the conditions already attempted and to optimise these for suitable crystal growth. This process is timely, however careful optimisation should yield crystals that diffract at a suitable resolution. 6.2 - Future Work for Chapter 4 The key limitation of the chemistry in Chapter 4 was the amide bond synthesis between 155 and 120/188/209, which typically returned poor yields of <30% (Chapter 4.5, 4.7 and 4.9.1). Alternative routes to access the coupled material may be possible through a copper or palladium mediated coupling between 3,6-dichloroacridone218 269 and fluoropropanamide 270 (Scheme 6.01).219,220 Cl Cl N H O NH 2 O F N + Cu(I) Ref 218 Ref 219-220 Scheme 6.01. Proposed coupling route between 3,6-dichloroacridone 269 and primary amide 270. X = halide 269 270 NHO XX Chapter 6 145 In addition to this, future work should be focused on the synthesis of a genuine BRACO-19 142a analogue. To achieve this, it may be required to re-assess the synthesis of the fluorinated amino acid. One promising route would be to treat methyl 3-chloropropanoate 271 with potassium phthalate 272 (Scheme 6.02) to form the protected β-amino acid 273 following simple functional group interconversions. This would be followed by a metal mediated asymmetric electrophilic fluorination of the acid chloride 273 to yield the α-fluoro product 274 as developed by Lectka (Scheme 6.02).221 This reaction can be quenched by various nucleophiles to provide esters and amides in good yields with high enantiomeric excess. The quenching of the intermediate by the dianion of 3,6-diaminoacridone 155 may prove fruitful to explore. O Nu F N O O O ClN O O 1) BQd, (PPh3)2PdCl2 DiPEA, NFSI 2) Nucleophile O O Cl K N O O + O Nu F N Scheme 6.02. Proposed alternative route to the synthesis of true BRACO-19 142a analogues. Nu = -OMe, -NHaryl, -NH2 In addition to this, the incorporation of two C–F bonds into 208 and 212 may enable a useful NMR probe to study the interactions between 208 and 212 with quadruplex DNA. The investigation between quadruplex DNA stabilising ligands and DNA is a very complicated and multifaceted arena, however, 1D or 2D NMR experiments involving 19F NMR may offer further information in addition to standard techniques employed. 271 272 273 274 275 Chapter 6 146 6.3 - Future work for Chapter 5 In order to fully investigate the finer details of the reaction mechanism, the synthesis of an aziridine dipeptide such as 276 from 275 would be advantageous (Scheme 6.03). The synthesis of similar aziridine containing dipeptides has been previously reported in the literature.222 N H O N H O O N H O NHTrt O O HO HF.Pyridine N H O O O NH 2 F N H O O O F H 2 N + ref. 222 Scheme 6.03. Proposed synthesis of an aziridine containing dipeptide 276 to probe α/β-fluorination distribution. Treatment of 276 with a nucleophilic source of fluorine, such as HF.pyridine, would enable the α/β-fluorination selectivity to be probed (Scheme 6.03). Further to this, the preparation of 276 as its PF6- salt would allow for X-ray crystallographic evaluation of a pseudo-intermediate similar to that of 236 in the DAST pathway mechanism (Scheme 5.10). This would provide conformational information and thus allow for further assessment of the origin of α/β-selectivity in these systems. In addition to the aforementioned investigations, the synthesis of the structures in Figure 6.01 followed by evaluation of the fluorination ratios by treating with DAST 32. This would add to the overall quality of this work by providing a comprehensive evaluation, scope and limitation survey. 276 277 278 279 Chapter 6 147 N N O HO N NH 2 O HO N N O HO N N O HO O O N N O HO O O N N O HO N Figure 6.01. Target compounds to further understand the scope of the fluorination reaction. The evaluation of other fluorinating reagents such as Deoxo-Fluor® or MOST would be of general interest, however, solvent investigations and work focused on improving the overall diastereomeric excess would vastly improve the synthetic use of this methodology. 280 281 282 283 284 285 Chapter 7 Experimental 7.1 - General experimental procedures All glassware was flame dried under high vacuum other than in situations where aqueous solutions were employed. Reactions were carried out under an atmosphere of argon, unless otherwise noted. Compressed argon was passed through a drying column packed with 4 Å molecular sieves, potassium hydroxide and self-indicating desiccant, before reaching a double manifold. All reactions involving the use of organometallic reagents were conducted by standard air-free techniques in Schlenk tubes or flasks. Hydrogenations were conducted in multi-neck flasks and the atmosphere exchanged with hydrogen by a pump-purge method. Dry CH2Cl2, Et2O, and THF were obtained from an mBRAUN SPS-800 solvent purification machine by passage through a drying column packed with 4Å molecular sieves and dispensed under an inert atmosphere when required. NOTE: THF from this purification system was unstabilised. Dry CH3OH was achieved by reflux over calcium hydride and collected in a still head when required. Where appropriate, solvents were degassed by the standard freeze-pump-thaw technique at least three times with freshly dispensed dry solvent.223 1H NMR spectra were recorded on 300, 400 or 500 MHz Bruker Avance/Avance II spectrometers. All spectra were acquired in deuterated solvents, and calibrated to the chemical shift of that residual solvent. Proton assignments are made according to chemical shift, multiplicity and 2D NMR experiments. Coupling constants (J) are reported to 0.1 Hz and are averaged for coupling nuclei. Complex spectra are numbered for ease of interpretation. All other resonances are described based on their chemical environment. NMR spectra were interpreted using iNMR or TopSpin. Experimental 149 13C NMR spectra were recorded at 75, 101, 126 MHz on Bruker Avance/Avance II spectrometers. Resonances were assigned by reference to DEPTQ, HMBC and HSQC spectra with coupling constants reported to 0.1 Hz, where appropriate. 19F NMR spectra were recorded at 282, 376, 470 MHz on Bruker Avance/Avance II spectrometers. Resonances were assigned according to chemical shift, multiplicity, and reference to the literature. Coupling constants are reported to 0.1 Hz and are averaged for coupling nuclei. Dr Tomas Lebl recorded all HOESY spectra. NMR Multiplicities are reported as follows: s - singlet; br s - broad singlet; m - multiplet; d - doublet; dd - doublet of doublets; ddd - doublet of doublet of doublets; dddd - doublet of doublet of doublet of doublets; dq - doublet of quartets; t - triplet; q - quartet; tq - triplet of quartets; qqd - quartet of quartet of doublets In vacuo refers to the use of a diaphragm vacuum pump to remove solvent under reduced pressure on a Büchi Rotavapor at 40 ºC. The bath temperature was reduced to 0 ºC with ice when removing solvents from volatile compounds. Drying under vacuum refers to the use of an Edwards RV-5 rotary-vane oil pump at a pressure of <0.1 mbar. Lypholisation refers to the removal of water by sublimation on a Christ Alpha 1-2 LD Plus freeze dryer equipped with an Edwards RV3 rotary-vane oil pump. Optical rotations were recorded on a Perkin Elmer optical rotation model 341 machine with a cell path length of 1 dm. The vast majority of samples were recorded at 589 nm (sodium D-line) at ambient temperature (20 ºC) and are denoted as [α]!!". For acridone and acridine compounds the light source was maintained at either 365 nm, 436 nm, 546 nm or 578 nm in an attempt to achieve satisfactory beam transmission. Concentrations (c) are reported in g/dm and specific optical rotations are denoted as [α]!!" in the implied units of 10-1 deg cm3 g-1. HPLC analysis was conducted on a Varian Prostar HPLC machine equipped with a Prostar Auto Sampler model 400 and a Prostar 240 solvent delivery system. Compound elution was monitored with a Prostar UV-Vis 325 module at a wavelength of 230 nm or 250 nm. Column for chiral analysis – Chiralcel OD (25 cm × 4.6mm, 10 µM), Chrialcel Experimental 150 OD-H (25 cm × 4.6mm, 5 µM) or Chiralcel AD-H (25 cm × 4.6mm, 5 µM). Reverse phase analysis – Nucleosil 100-5 C-18 RP column (25 mm × 3.2 mm, 5 µM). HPLC grade solvents were degassed prior to use and the column was preconditioned with the solvent system for at least 20 min before injection. Melting points were measured using a Gallenkamp Griffin MPA350 or Electrothermal 9100 digital melting point apparatus and are uncorrected. Mass Spectroscopic analyses at the Biomedical Sciences Research Complex (BSRC) were conducted by Mrs. Caroline Horsburgh on a Micromass LCT electrospray time of flight mass spectrometer by electrospray ionisation. Samples sent to the EPSRC mass spectrometry service in Swansea were analysed on a Thermofisher LTQ Orbitrap XL mass spectrometer using either electrospray ionisation (ES) or atmospheric solids analysis probe techniques (ASAP). X-ray analysis of single crystals was conducted by Prof. Alexandra Slawin at the University of St Andrews on a Rigaku Cu MM007 high brilliance generator with Saturn 92 CCD and XStream LT accessories. IR spectra were recorded a Perkin Elmer Spectrum GX FT-IR machine as either a KBr disc, neat on NaCl plates or on PTFE cards. Peptide samples were recorded neat on a Shimadzu Raffinity-1 FT-IR machine. UV-Vis spectra were recorded on Perkin Elmer Lambda 35 UV/VIS spectrometer with a quartz cell with a 1 cm path length using spectrophotometric grade methanol. Samples were prepared at a concentration of 1 mg in 1 mL and diluted until suitable spectra could be obtained. Extinction coefficients (ε) were calculated using the Beer-Lambert law: A = εcℓ and are quoted as log10 ε values in M-1 cm-1. TLC analysis was conducted on aluminium backed TLC silica gel 60 F254 plates, and followed by visualisation with UV light (254 or 365 nm) and/or staining with the appropriate staining solution. Typical solutions used included: aqueous alkaline potassium permanganate; ninhydrin spray; vanillin solution; or ethanolic ceric ammonium molybdate. Flash column chromatography was achieved using Merck Experimental 151 Geduran Si-60 silica-gel (40-63 µM particle size).224 Where required, reverse phase C-18 silica gel (2-10 µM particle size) aluminium backed TLC plates with 254nm indicator were used. Reverse phase flask chromatography was achieved with C-18 fully end-capped reverse phase silica gel (15-25 µM particle size). Miscellaneous - Brine referrers to a saturated solution of sodium chloride in deionised water. Reactions at -78 ºC were readily achieved with isopropanol and dry ice in a Dewar vacuum flask, -100 ºC was achieved with CH3OH and liquid nitrogen. Reactions that required a sustained low temperature were cooled with a LabPlant RP-60 Refrigerated Immersion Probe. Celite was washed with aqueous HCl (0.1 M), followed by water and CH3OH prior to use. Chemicals - All chemicals were purchased from: Sigma-Aldrich, Alfa-Aesar, Fluorochem, TCI Europe or Fisher Scientific and were used as supplied unless otherwise noted. Triethylamine (distilled from KOH, stored over KOH), diisopropylethylamine (distilled from KOH, stored under N2), diallylamine (distilled from NaOH, stored under inert atmosphere) and dipropylamine (distilled from KOH, stored under inert atmosphere) were distilled before use. Phosphorus oxychloride (distilled and stored in the dark under inert atmosphere), 3-chloropropionyl chloride (distilled under reduced pressure, stored in dark under inert atmosphere) and DMF (distilled from CaH2 and stored over 3A molecular sieves under inert atmosphere) were distilled before use.223 Experimental 152 7.2 - Experimental for Chapter 3 7.2.1 - 3,6-Bis(3-chloropropionamido)acridine[175] 146 3,6-Diaminoacridine (1.00 g, 4.78 mmol, 1.0 eq) was heated under reflux in neat 3-chloropropionyl chloride (5 mL) for 3 h. The solution was cooled to rt and ice-cold Et2O (20 mL) was added, resulting in formation of a precipitate. The precipitate was isolated by filtration, washed with Et2O (2 × 10 mL) and dried under vacuum. This solid was recrystallized from ethanol and DMF (1:5), to yield 3,6-bis(3-chloropropionamido)acridine 146 (1.80 g, 4.30 mmol, 90%) as an orange amorphous solid: mp >300 °C (ethanol:DMF); 1H NMR (300 MHz, d6-DMSO) δH 11.55 (2H, br s, NHCO), 9.61 (1H, s, Ar H-9), 8.91 (2H, d, J 1.6 Hz, Ar H-4,5), 8.37 (2H, d, J 9.2 Hz, Ar H-1,8), 7.92 (2H, dd, J 9.2, 1.6 Hz, Ar H-2,7), 4.00 (4H, t, J 6.2 Hz, 2 × COCH2), 3.11 (4H, t, J 6.2 Hz, 2 × CH2Cl). 7654b8a 8N104a9a 9432 1 NHNH OOCl Cl Experimental 153 7.2.2 - 3,6-Bis(3-(3'-(R)-fluoropyrrolidin)propionamido)acridine (R,R)-144 (3R)-Fluoropyrrolidine (340 mg, 3.8 mmol, 10.0 eq) in ethanol (1 mL) was added to a solution of 3,6-bis(3-chloropropionamido)acridine (150 mg, 0.38 mmol, 1.0 eq) and NaI (58 mg, 0.38 mmol 2.0 eq) in ethanol (5 mL) and the mixture was heated under reflux for 5 h. The reaction was cooled to 0 °C, resulting in formation of a precipitate, which was isolated by filtration and washed with Et2O (10 mL). The product was purified by silica gel column chromatography, eluting with CH2Cl2, CH3OH and Et3N (85:10:5), to yield 3,6-bis(3-(3'-(R)-fluoropyrrolindino)propionamido)acridine (R,R)-144 (121 mg, 0.24 mmol, 63%) as an orange amorphous solid: Rf 0.1 (CH2Cl2:CH3OH:Et3N, 94:5:1); IR (film/cm-1) 3689, 2958, 2918, 2780, 2310, 2315, 1644, 1445, 1304, 1215, 1154, 1084, 1026; mp >300 °C (dec.); [!]!"#!" -8.0 (c 0.44, CH3OH); 1H NMR (500 MHz, CD3OD/d6-DMSO) δH 8.82 (1H, s, Ar H-9), 8.54 (2H, d, J 1.9 Hz, Ar H-4,5), 8.01 (2H, d, J 9.1 Hz, Ar H-1,8), 7.64 (2H, dd, J 9.1, 1.9 Hz, Ar H-2,7), 5.36-5.23 (2H, m, 2 × CHF-7',7''), 3.20-3.10 (4H, m, 2 × CHa -6',6'' and 2 × CHa-9',9''), 3.04 (4H, t, J 7.0 Hz, 2 × CH2-3',3''), 2.89-2.79 (2H, m, 2 × CHb-6',6''), 2.76 (4H, t, J 7.0 Hz, 2 × CH2-4',4''), 2.68-2.63 (2H, m, 2 × CHb-9',9''), 2.36-2.23 (2H, m, 2 × CHa-8',8''), 2.16-2.05 (2H, m, 2 × CHb-8',8''); 13C NMR (126 MHz, CD3OD/d6-DMSO) δC 172.6 (2 × CONH), 150.7(2 × Ar C-8a,9a), 142.5 (2 × Ar C-4a,4b), 137.9 (Ar CH-9), 130.5 (2 × Ar CH-1,8), 124.5 (2 × Ar C-3,6), 7654b8a 8N104a9a 9432 1 NH1''NH1'2' 2''O 3'' 4'' N5''3' O4'N5' 9'' 8'' 7''6''6'7' 8' 9' FF Experimental 154 121.6 (2 × Ar CH-2,7), 114.8 (2 × Ar CH-4,5), 94.3 (d, J 174.9 Hz, 2 × CHF-7',7''), 61.3 (d, J 22.8 Hz, 2 × CH2-6',6''), 53.1 (2 × CH2-9',9''), 52.5 (2 × CH2-3',3''), 36.3 (2 × CH2-4',4''), 33.4 (d, J 22.3 Hz, 2 × CH2-8',8''); 19F NMR (470 MHz, CD3OD/d6-DMSO) δF -169.0 (2F, m, 2 × CHF); HRMS m/z (ES+) calcd. for C27H32F2N5O2 [M+H]+ requires 496.2524, found 496.2540; m/z (ES+) 496 ([M+H]+, 100%). 7.2.3 - 3,6-Bis(3-(3'-(S)-fluoropyrrolindino)propionamido)acridine (S,S)-144 (3S)-Fluoropyrrolidine (340 mg, 3.8 mmol, 10.0 eq) in ethanol (1 mL) was added to a solution of 3,6-bis(3-chloropropionamido)acridine (150 mg, 0.38 mmol, 1.0 eq) and NaI (58 mg, 0.38 mmol 2.0 eq) in ethanol (5 mL) and the mixture was heated under reflux for 5 h. The reaction was cooled to 0 °C, resulting in formation of a precipitate, which was isolated by filtration and washed with Et2O (10 mL). The product was purified by silica gel column chromatography, eluting with CH2Cl2, CH3OH and Et3N (85:10:5), to yield 3,6-bis(3-(3'-(S)-fluoropyrrolindino)propionamido)acridine (S,S)-144 (123 mg, 0.26 mmol, 65%) as an orange amorphous solid: Rf 0.1 (CH2Cl2:CH3OH:Et3N, 94:5:1); IR (film/cm-1) 3689, 2958, 2918, 2780, 2310, 2315, 1644, 1445, 1304, 1304, 1215, 1154, 1084, 1038, 1026; mp >300 °C (dec.); [!]!"#!" +8.1 (c 0.44, CH3OH); 1H NMR (500 MHz, CD3OD/d6-DMSO) δH 8.85 (1H, s, Ar H-9), 8.56 (2H, d, J 1.9 Hz, Ar H-4,5), 8.04 (2H, d, J 9.1 Hz, Ar H-1,8), 7.66 (2H, 7654b8a 8N104a9a 9432 1 NH1''NH1'2' 2''O 3'' 4'' N5''3' O4'N5' 9'' 8'' 7''6''6'7' 8' 9' FF Experimental 155 dd, J 9.1, 1.9 Hz, Ar H-2,7), 5.34-5.21 (2H, m, 2 × CHF-7',7''), 3.18-3.08 (4H, m, 2 × CHa-6',6'' and 2 × CHa-9',9''), 3.02 (4H, t, J 7.0 Hz, 2 × CH2-3',3''), 2.89-2.79 (2H, m, 2 × CHb-6',6''), 2.75 (4H, t, J 7.0 Hz, 2 × CH2-4',4''), 2.64-2.59 (2H, m, 2 × CHb-9',9''), 2.35-2.22 (2H, m, 2 × CHa-8',8''), 2.15-2.04 (2H, m, 2 × CHb-8',8''); 13C NMR (126 MHz, CD3OD/d6-DMSO) δC 172.6 (2 × CONH), 150.8 (2 × Ar C-8a,9a), 142.5 (2 × Ar C-4a,4b), 137.8 (Ar CH-9), 130.5 (2 × Ar CH-1,8), 124.5 (2 × Ar C-3,6), 121.6 (2 × Ar CH-2,7), 114.9 (2 × Ar CH-4,5), 94.5 (d, J 174.9 Hz, 2 × CHF-7',7''), 61.4 (d, J 22.8 Hz, 2 × CH2-6', 6''), 53.1 (2 × CH2-9',9''), 52.5 (2 × CH2-3',3''), 36.5 (2 × CH2-4',4''), 33.5 (d, J 22.3 Hz, 2 × CH2-8', 8''); 19F NMR (470 MHz, CD3OD/d6-DMSO) δF -168.5 (2F, m, 2 × CHF); HRMS m/z (ES+) calcd. for C27H32F2N5O2 [M+H]+ requires 496.2524, found 496.2520; m/z (ES+) 496 ([M+H]+, 100%). 7.2.4 - 3,6-Bis(3-(3'-(R)-hydroxypyrrolindino)propionamido)acridine (R,R)-145 (3R)-Hydroxypyrrolidine hydrochloride (335 mg, 3.8 mmol, 10.0 eq) in ethanol (1 mL) was added to a solution of 3,6-bis(3-chloropropionamido)acridine (150 mg, 0.38 mmol, 1.0 eq) and NaI (58 mg, 0.38 mmol 2.0 eq) in ethanol (5 mL) and the mixture was heated under reflux for 5 h. The reaction was cooled to 0 °C, resulting in formation of a precipitate, which was isolated by filtration and washed with Et2O (10 mL). The product 7654b8a 8N104a9a 9432 1 NH1''NH1'2' 2''O 3'' 4'' N5''3' O4'N5' 9'' 8'' 7''6''6'7' 8' 9' OHHO Experimental 156 was purified by silica gel column chromatography, eluting with CH2Cl2, CH3OH and Et3N (85:10:5), to yield 3,6-bis(3-(3'-(R)-hydroxypyrrolindino)propionamido)acridine (R,R)-145 (118 mg, 0.23 mmol, 59%) as an orange amorphous solid: mp >300 °C (dec.); IR (film/cm-1) 3680, 2963, 2918, 2789, 2352, 1672, 1549, 1448, 1205, 1150, 1068, 1020; [!]!"#!" +5.6 (c 0.53, CH3OH); 1H NMR (500 MHz, CD3OD/d6-DMSO) δH 8.82 (1H, s, Ar H-9), 8.53 (2H, d, J 1.7 Hz, Ar H-4,5), 8.01 (2H, d, J 9.1 Hz, Ar H-1,8), 7.67 (2H, dd, J 9.0, 1.7 Hz, Ar H-2,7), 4.44-4.41 (2H, m, 2 × CH-7',7''), 3.03 (4H, t, J 6.9 Hz, 2 × CH2-3',3''), 3.04-2.93 (4H, m, 2 × CH2-6',6''), 2.73 (4H, t, J 6.9 Hz, 2 × CH2-4',4''), 2.77-2.71 (4H, m, 2 × CH2-9',9''), 2.24-2.17 (2H, m, 2 × CHa-8',8''), 1.84-1.78 (2H, m, 2 × CHb-8',8''); 13C NMR (126 MHz, CD3OD/d6-DMSO) δC 172.9 (2 × CONH), 150.9 (2 × Ar C-8a,9a), 142.4 (2 × Ar C-4a,4b), 137.8 (Ar CH-9), 130.4 (2 × Ar CH-1,8), 124.7 (2 × Ar C-3,6), 121.7 (2 × Ar CH-2,7), 115.2 (2 × Ar CH-4,5), 71.4 (2 × CH-7',7''), 63.3 (2 × CH2-6',6''), 53.5 (2 × CH2-9',9''), 52.8 (2 × CH2-3',3''), 36.2 (2 × CH2-4',4''), 35.2 (2 × CH2-5',5''); HRMS m/z (ES+) calcd. for C27H34N5O4 [M+H]+ requires 492.2611, found 492.2601; m/z (ES+) 492 ([M+H]+, 100%). 7.2.5 - 3,6-Bis(3-(3'-(S)-hydroxypyrrolindino)propionamido)acridine (S,S)-145 (3S)-Hydroxypyrrolidine hydrochloride (335 mg, 3.8 mmol, 10.0 eq) in ethanol (1 mL) was added to a solution of 3,6-bis(3-chloropropionamido)acridine (150 mg, 0.38 mmol, 7654b8a 8N104a9a 9432 1 NH1''NH1'2' 2''O 3'' 4'' N5''3' O4'N5' 9'' 8'' 7''6''6'7' 8' 9' OHHO Experimental 157 1.0 eq) and NaI (58 mg, 0.38 mmol 2.0 eq) in ethanol (5 mL) and the mixture was heated under reflux for 5 h. The reaction was cooled to 0 °C, resulting in formation of a precipitate, which was isolated by filtration and washed with Et2O (10 mL). The product was purified by silica gel column chromatography, eluting with CH2Cl2, CH3OH and Et3N (85:10:5), to yield 3,6-bis(3-(3'-(S)-hydroxypyrrolindino)propionamido)acridine (S,S)-144 (110.9 mg, 0.225 mmol, 59%) as an orange amorphous solid: [!]!"#!" -7.0 (c 0.44, CH3OH); 1H NMR (500 MHz, CD3OD/d6-DMSO) δH 8.80 (1H, s, Ar H-9), 8.51 (2H, d, J 1.7 Hz, Ar H-4,5), 7.99 (2H, d, J 9.1 Hz, Ar H-1,8), 7.67 (2H, dd, J 9.0, 1.7 Hz, Ar H-2,7), 4.44-4.41 (2H, m, 2 × CH-7',7''), 2.99 (4H, t, J 6.9 Hz, 2 × CH2-3',3''), 3.01-2.90 (4H, m, 2 × CH2-6',6''), 2.71 (4H, t, J 6.9 Hz, 2 × CH2-4',4''), 2.74-2.67 (4H, m, 2 × CH2-9',9''), 2.24-2.17 (2H, m, 2 × CHa-8',8''), 1.84-1.78 (2H, m, 2 × CHb-8',8''); 13C NMR (126 MHz, CD3OD/d6-DMSO) δC 173.0 (2 × CONH), 150.9 (2 × Ar C-8a,9a), 142.4 (2 × Ar C-4a,4b), 137.7 (Ar CH-9), 130.4 (2 × Ar CH-1,8), 124.7 (2 × Ar C-3,6), 121.7 (2 × Ar CH-2,7), 115.2 (2 × Ar CH-4,5), 71.4 (2 × CH-7',7''), 63.3 (2 × CH2-6',6''), 53.5 (2 × CH2-9',9''), 52.8 (2 × CH2-3',3''), 36.4 (2 × CH2-4',4''), 35.1 (2 × CH2-5',5''); HRMS m/z (ES+) calcd. for C27H34N5O4 [M+H]+ requires 492.2611 found 492.2606; m/z (ES+) 492 ([M+H]+, 100%). Experimental 158 7.3 - Experimental for Chapter 4 7.3.1 - D-Serine methyl ester hydrochloride[225] (R)-164 Thionyl chloride (13.7 mL, 192 mmol, 1.1 eq) was added dropwise to CH3OH (180 mL) over 30 min at rt followed by D-serine (18.0 g, 170 mmol, 1.0 eq) portion wise. Following consumption of the starting material as indicated by TLC, the solvent was removed in vacuo and the resulting solids were triturated with petroleum ether. Trituration and subsequent evaporation was repeated to remove excess thionyl chloride. The product was recrystallised from CH3OH to yield D-serine methyl ester hydrochloride (R)-164 (21.0 g, 140 mmol, 80%) as a white crystalline solid: mp 163-165 °C (CH3OH) [Lit.[225] 163-166 ºC]; [!]!!" -4.2 (c 4.0, CH3OH) [Lit.[225] [!]!!" -3.7 (c 4.0, CH3OH)]; 1H NMR (400 MHz, CD3OD) δH 4.91 (1H, br s, OH), 4.19 (1H, dd, J 4.4, 3.5 Hz, CHN), 4.04 (1H, dd, J 11.9, 4.4 Hz, CHaHbOH), 3.98 (1H, dd, J 11.9, 3.5 Hz, CHaHbOH), 3.88 (3H, s, CH3). ONH2HO O Experimental 159 7.3.2 - L-Serine methyl ester hydrochloride[225] (S)-164 Thionyl chloride (13.7 mL, 190 mmol, 1.1 eq) was added dropwise to CH3OH (180 mL) over 30 min at rt followed by L-serine (18.0 g, 170 mmol, 1.0 eq) in a portion wise manner. Following consumption of the starting material as indicated by TLC, the solvent was removed in vacuo and the solids were triturated with petroleum ether. Trituration and subsequent evaporation was repeated to remove excess thionyl chloride. The product was recrystallised from CH3OH to yield L-serine methyl ester hydrochloride (S)-164 (21.3 g, 140 mmol, 80%) as a white crystalline solid: mp 162-165 ºC [Lit.[225] 163-166 ºC]; [!]!!" +4.3 (c 4.0, CH3OH), [Lit.[225] [!]!!" +3.7 (c 4.0, CH3OH)]; 1H NMR (400 MHz, CD3OD) δH 4.91 (1H, br s, OH), 4.19 (1H, dd, J 4.4, 3.5 Hz, CHN), 4.04 (1H, dd, J 11.9, 4.4 Hz, CHaHbOH), 3.98 (1H, dd, J 11.9, 3.5 Hz, CHaHbOH), 3.88 (3H, s, CH3). ONH2HO O Experimental 160 7.3.3 - Methyl (±)-2-(dibenzylamino)-3-hydroxypropanoate[226] 119 Benzyl bromide (19.3 mL, 162 mmol, 2.5 eq) was added to a solution of DL-serine methyl ester hydrochloride (10.0 g, 64.0 mmol, 1.0 eq) and K2CO3 (44.7 g, 323 mmol, 5.0 eq) in acetonitrile. This mixture was stirred for 24 h at rt and quenched by the addition of water (300 mL). The aqueous phase was extracted with ethyl acetate (3 × 300 mL) and the combined organic fractions were dried over Na2SO4, filtered and concentrated in vacuo. The product was purified by silica gel column chromatography, eluting with hexane and ethyl acetate (80:20), to yield methyl 2-(dibenzylamino)-3-hydroxypropanoate 119 (18.1 g, 60.0 mmol, 94%) as a colorless oil: 1H NMR (400 MHz, CD3OD) δH 7.40-7.22 (10H, m, 10 × Ar H), 3.94 (1H, dd, J 10.9, 7.8 Hz, CHaHbOH), 3.88 (2H, d, J 13.8 Hz, 2 × NCHaHb-benzyl), 3.80 (3H, s, OCH3), 3.75 (1H, dd, J 10.9, 5.9 Hz, CHaHbOH), 3.61 (2H, d, J 13.8 Hz, 2 × NCHaHb-benzyl), 3.47 (1H, dd, J 7.8, 5.9 Hz, CHN); m/z (ES+) 300 ([M+H]+, 100%). OON OHPhPh Experimental 161 7.3.4 - Methyl (+)-(2R)-(dibenzylamino)-3-hydroxypropanoate[226] (R)-119 Following the procedure set out for methyl 2-(dibenzylamino)-3-hydroxypropanoate 119, starting from D-serine methyl ester hydrochloride (R)-164 (10.0 g, 64.0 mmol), methyl (2R)-(dibenzylamino)-3-hydroxypropanoate (R)-119 (18.3 g, 61.0 mmol, 95%) was obtained as a colourless oil: [!]!!" +144 (c 1.0, CHCl3), [Lit.[226] ![!]!!" +147 (c 0.96, CHCl3)]; 1H NMR (400 MHz, CD3OD) δH 7.40-7.22 (10H, m, 10 × Ar H), 3.94 (1H, dd, J 10.9, 7.8 Hz, CHaHbOH), 3.88 (2H, d, J 13.8 Hz, 2 × NCHaHb-benzyl), 3.80 (3H, s, OCH3), 3.75 (1H, dd, J 10.9, 5.9 Hz, CHaHbOH), 3.61 (2H, d, J 13.8 Hz, 2 × NCHaHb-benzyl), 3.47 (1H, dd, J 7.8, 5.9 Hz, CHN); m/z (ES+) 300 ([M+H]+, 100%). 7.3.5 - Methyl (–)-(2S)-(dibenzylamino)-3-hydroxypropanoate[227] (S)-119 Following the procedure set out for methyl 2-(dibenzylamino)-3-hydroxypropanoate 119, starting from L-serine methyl ester hydrochloride (S)-164 (10.0 g, 64.0 mmol), OON OHPhPh OON OHPhPh Experimental 162 methyl (2S)-(dibenzylamino)-3-hydroxypropanoate (S)-119 (18.0 g, 60.0 mmol, 94%) was obtained as a colourless oil: [!]!!" -140 (c 1.0, CHCl3) [Lit.[221] [!]!!" -105 (c 1.21, CH3OH)]; 1H NMR (300 MHz, CD3OD) δH 7.37-7.19 (10H, m, 10 × Ar H), 3.91 (1H, dd, J 10.9, 7.8 Hz, CHaHbCHN), 3.85 (2H, d, J 13.8 Hz, 2 × CHaHb-benzyl), 3.77 (3H, s, OCH3), 3.72 (1H, dd, J 10.9, 5.9 Hz, CHaHbCHN), 3.58 (2H, d, J 13.8 Hz, 2 × CHaHb-benzyl), 3.47 (1H, dd, J 7.8, 5.9 Hz, CHN); m/z (ES+) 300 ([M+H]+, 100%). 7.3.6 - Methyl (±)-3-(dibenzylamino)-2-fluoropropanoate 120 Diethylaminosulfur trifluoride 32 (4.20 mL, 32.0 mmol, 1.2 eq) was added to a solution of methyl 2-(dibenzylamino)-3-hydroxypropanoate 119 (8.00 g, 26.7 mmol, 1.0 eq) in THF (45 mL) and the reaction was cooled to 0 °C. This solution was stirred for 1 h at 0 °C before being quenched by the addition of cold water (90 mL), followed by an excess of solid K2CO3 and Et2O (90 mL). The organic phase was separated and the aqueous mixture re-extracted with Et2O (2 × 90 mL). The organic fractions were combined, dried over Na2SO4 and concentrated in vacuo. The product was purified by silica gel column chromatography, eluting with hexane and ethyl acetate (80:20), to yield methyl 3-(dibenzylamino)-2-fluoropropanoate 120 (7.25 g, 24.1 mmol, 90%) as a colourless oil: IR νmax (film, cm-1) 3027, 2802, 1763, 1494, 1438, 1352, 1368, 1291, 1207, 1150, 1066; 1H NMR (400 MHz, CDCl3) δH 7.25-7.15 (10H, m, 10 × Ar-H), 4.98 (1H, ddd, J 49.3, 5.8, 3.3 Hz, CHF), 3.76 (2H, d, J 13.6 Hz, 2 × NCHaHb-benzyl), OOFNPhPh Experimental 163 3.61 (3H, s, OCH3), 3.45 (2H, d, J 13.6 Hz, 2 × NCHaHb-benzyl), 3.03-2.85 (2H, m, CH2CHF); 13C NMR (101 MHz, CDCl3) δC 169.3 (d, J 24.2 Hz, CO2CH3), 138.9 (2 × Ar-C), 129.1 (4 × Ar-CH), 128.4 (4 × Ar-CH), 127.2 (2 × Ar-CH), 89.5 (d, J 186.1 Hz, CHF), 58.9 (2 × CH2Ar), 54.3 (d, J 20.1 Hz, NCH2), 52.3 (OCH3); 19F NMR (283 MHz, CDCl3) δF -192.2 (ddd, J 49.3, 29.2, 22.1 Hz, CHF); HRMS m/z (ES+) calcd. for C18H20NO2FNa [M+Na]+ requires 324.1376, found 324.1369; m/z (ES+) 324 ([M+Na]+, 100%). 7.3.7 - Methyl (–)-(2S)-3-(dibenzylamino)-2-fluoropropanoate (S)-120 Following the procedure set out for methyl 3-(dibenzylamino)-2-fluoropropanoate 120, starting from methyl (2R)-(dibenzylamino)-3-hydroxypropanoate (R)-119 (8.00 g, 26.7 mmol) with diethylaminosulfur trifluoride 32 (4.20 mL, 32.0 mmol, 1.2 eq) in THF (45 mL), furnished methyl (2S)-3-(dibenzylamino)-2-fluoropropanoate (S)-120 (7.19 g, 23.9 mmol, 89%) as a colourless oil: [!]!!" -19.0 (c 2.0, CH3OH); 1H NMR (300 MHz, CDCl3) δH 7.26-7.14 (10H, m, 10 × Ar H), 4.97 (1H, ddd, J 49.5, 5.7, 3.3 Hz, CHF), 3.75 (2H, d, J 13.6 Hz, 2 × CHaHbPh), 3.61 (3H, s, OCH3), 3.44 (2H, d, J 13.6 Hz, 2 × CHaHbPh), 2.97 (1H, ddd, J 26.9, 14.7, 5.7 Hz, 2 × CHaHbCHF), 2.90 (1H, ddd, J 24.3, 14.7, 3.3 Hz, 2 × CHaHbCHF); 13C NMR (75 MHz, CDCl3) δC 169.3 (d, J 24.2 Hz, CO2CH3), 138.9 (2 × Ar C), 129.1 (4 × Ar CH), 128.4 (4 × Ar CH), 127.2 (2 × Ar CH), 89.5 (d, J 186.1 Hz, CHF), 58.9 (2 × CH2Ph), OOFNPhPh Experimental 164 54.3 (d, J 20.1 Hz, CH2CHF), 52.3 (OCH3); 19F NMR (282 MHz, CDCl3) δF -191.0 (ddd, J 49.5, 26.9, 24.3 Hz, CHF); HRMS m/z (ES+) calcd. for C18H20NO2FNa [M+Na]+ requires 324.1376, found 324.1375; m/z (ES+) 324 ([M+Na]+, 100%). Enantiomeric excess determined by chiral HPLC (Chiralcel OD 5% iPrOH in hexane, 0.25 mL/min, tr maj = 14.51 min >95%, tr min = 15.70 min <5%). 7.3.8 - Methyl (+)-(2R)-3-(dibenzylamino)-2-fluoropropanoate (R)-120 Following the procedure set out for methyl 3-(dibenzylamino)-2-fluoropropanoate 120, starting from methyl (2S)-(dibenzylamino)-3-hydroxypropanoate (S)-119 (8.00 g, 26.7 mmol) with diethylaminosulfur trifluoride 32 (4.20 mL, 32.0 mmol, 1.2 eq) in THF (45 mL), furnished methyl (2R)-3-(dibenzylamino)-2-fluoropropanoate (R)-120 (7.23 g, 24.0 mmol, 90%) as a colourless oil: [!]!!" +19.1 (c 2.0, CH3OH); 1H NMR (300 MHz, CDCl3) δH 7.29-7.12 (10H, m, 10 × Ar H), 4.97 (1H, ddd, J 49.5, 5.7, 3.3 Hz, CHF), 3.75 (2H, d, J 13.6 Hz, 2 × CHaHbPh), 3.61 (3H, s, OCH3), 3.44 (2H, d, J 13.6 Hz, 2 × CHaHbPh), 2.97 (1H, ddd, J 26.9, 14.7, 5.7 Hz, 2 × CHaHbCHF), 2.90 (1H, ddd, J 24.3, 14.7, 3.3 Hz, 2 × CHaHbCHF); 13C NMR (75 MHz, CDCl3) δC 169.3 (d, J 24.2 Hz, CO2CH3), 138.9 (2 × Ar C), 129.1 (4 × Ar CH), 128.4 (4 × Ar CH), 127.2 (2 × Ar CH), 89.5 (d, J 186.1 Hz, CHF), 58.9 (2 × CH2Ph), 54.3 (d, J 20.1 Hz, CH2CHF), 52.3 (OCH3); 19F NMR (282 MHz, CDCl3) δF -191.0 (ddd, J 49.5, 26.9, 24.3 Hz, CHF); HRMS m/z (ES+) calcd. for OOFNPhPh Experimental 165 C18H21NO2F [M+H]+ requires 302.1556, found 302.1548; m/z (ES+) 324 ([M+Na]+, 100%), 302 ([M+H], 80%). Enantiomeric excess determined by chiral HPLC (Chiralcel OD-H 5% iPrOH in hexane, 0.25 mL/min, tr maj = 15.70 min >95%, tr min = 14.51 min <5%). 7.3.9 - Methyl (±)-3-amino-2-fluoropropanoate hydrochloride 166.HCl A solution of methyl (±)-3-(dibenzylamino)-2-fluoropropanoate 120 (200 mg, 0.66 mmol, 1.0 eq) and 20% Pd(OH)2/C (40.0 mg, 10 mol%) in CH3OH (10 mL) was stirred under an H2 atmosphere. This suspension was stirred vigorously until TLC/19F NMR analysis had indicated complete debenzylation and HCl (0.5 M, 1.5 mL) was added. The mixture was filtered through a pad of Celite and the residue was washed with CH3OH (30 mL). The filtrate was concentrated in vacuo to yield methyl (±)-3-amino-2-fluoropropanoate hydrochloride 166.HCl (80 mg, 99%) as a colourless solid, which was used without purification: 1H NMR (400 MHz, CD3OD) δH 5.37 (1H, ddd, J 47.6, 7.5, 3.4 Hz, CHF), 3.86 (3H, s, OCH3), 3.61-3.42 (2H, m, CH2); 13C NMR (101 MHz, CD3OD) δC 171.2 (d, J 22.9 Hz, CO2CH3), 86.7 (d, J 185.9 Hz, CHF), 53.6 (OCH3), 41.1 (d, J 21.4 Hz, CH2); 19F NMR (376 MHz, CD3OD) δF -200.3 (ddd, J 48.0, 25.0, 23.0 Hz, CHF); HRMS m/z (ES+) calcd. for C4H9NO2F [M+H]+ requires 122.0614, found 122.0621; m/z (ES+) 122 ([M+H]+, 100%). OOClH3N F Experimental 166 7.3.10 - Methyl (±)-3-(pyrrolidin-1-yl)-2-fluoropropanoate 153 1,4-Dibromobutane (85 µL, 712 µmol, 1.1 eq) was added to a solution of tetrabutylammonium iodide (45 mg, 146 µmol, 0.2 eq), sodium carbonate (270 mg, 2.55 mmol, 4.0 eq) and methyl (±)-3-amino-2-fluoropropanoate 166 (100 mg, 634 µmol, 1.0 eq) in THF. The resulting solution was heated under reflux for 4 hr, cooled to rt and quenched with water (2 mL) and ethyl acetate (4 mL). The organics were separated and the aqueous layer further extracted with ethyl acetate (4 mL). The organic phases were combined, washed with brine (5 mL), dried over sodium sulfate, filtered and the solvent removed in vacuo. The product was purified by silica gel column chromatography eluting with CH2Cl2 and CH3OH (100:0, 99:1), to furnish methyl (±)-3-(pyrrolidin-1-yl)-2-fluoropropanoate 153 (85 mg, 485 µmol, 77%) as a colourless oil: Rf 0.13 (CH2Cl2:CH3OH, 99:1); 1H NMR (400 MHz, CDCl3) δH 5.09 (1H, ddd, J 49.5, 6.8, 2.8 Hz, CHF), 3.81 (3H, s, OCH3), 3.02 (1H, ddd, J 26.2, 14.0, 6.8 Hz, CHaHbCHF), 2.95 (1H, ddd, J 28.6, 14.0, 2.8 Hz, CHaHbCHF), 2.67-2.56 (4H, m, 2 × CH2), 1.79-1.76 (4H, m, 2 × CH2); 13C NMR (101 MHz, CDCl3) δC 169.5 (d, J 24.0 Hz, CO2CH3), 89.5 (d, J 187.0 Hz, CHF), 57.2 (d, J 20.3 Hz, CH2CHF), 54.9 (2 × CH2), 52.5 (OCH3), 23.8 (2 × CH2); 19F{1H} NMR (376 MHz, CDCl3) δF -192.2 (s, CHF); HRMS m/z (ES+) calcd. for C8H15NO2F [M+H]+ requires 176.1087, found 176.1087; m/z (ES+) 194 ([M+Na]+, 10%), 176 ([M+H]+, 100%). OON F Experimental 167 7.3.11 - (+)-(2S)-3-(Dibenzylamino)-2-fluoropropanoic acid[228] (S)-170 Methyl (2S)-3-(dibenzylamino)-2-fluoropropanoate (S)-120 (1.00 g, 3.32 mmol, 1.0 eq) was added to a solution of KOH (3.20 g, 57.1 mmol, 10.0 eq) in CH3OH (10 mL). The resulting solution was stirred for 36 h at rt. The reaction was diluted with HCl (1 M, 10 mL) and the aqueous phase extracted with ethyl acetate (3 × 5 mL). The combined organic fractions were dried over Na2SO4, filtered and the solvent removed in vacuo to furnished (2S)-3-(dibenzylamino)-2-fluoropropanoic acid (S)-170 (0.91 g, 98%) as a colourless oil, which was used without any further purification: [!]!!" +0.8 (c 2.5, CHCl3); 1H NMR (400 MHz, CD3OD) δH 7.31 (10H, m, 10 × Ar-H), 4.98 (1H, ddd, J 49.5, 7.3, 3.3 Hz, CHF), 3.79 (2H, d, J 13.8 Hz, 2 × NCHaHb), 3.66 (2H, d, J 13.8 Hz, 2 × NCHaHb), 3.01 (2H, m, CH2CHF); 19F NMR (282 MHz, CD3OD) δF -189.0 (ddd, J 49.5, 22.5, 22.5 Hz, CHF); m/z (ES+) 310 ([M+Na]+, 80%), 288 ([M+H]+, 100%). OHOFNPhPh Experimental 168 7.3.12 - (–)-(2R)-3-(Dibenzylamino)-2-fluoropropanoic acid[228,229] (R)-170 Methyl (2R)-3-(dibenzylamino)-2-fluoropropanoate (R)-120 (1.00 g, 3.32 mmol, 1.0 eq) was added to a solution of KOH (3.20 g, 57.1 mmol, 10.0 eq) in CH3OH (10 mL). The resulting solution was stirred for 36 h at rt. The reaction was diluted with HCl (1 M, 10 mL) and the aqueous phase extracted with ethyl acetate (3 × 5 mL). The combined organic fractions were dried over Na2SO4, filtered and the solvent removed in vacuo to furnish (2R)-3-(dibenzylamino)-2-fluoropropanoic acid (R)-170 (0.90 g, 97%) as a colourless oil which was used without any further purification: [!]!!" -0.8 (c 2.5, CHCl3); 1H NMR (300 MHz, CD3OD) δH 7.44-7.33 (10H, m, 10 × Ar-H), 5.08 (1H, ddd, J 49.5, 7.3, 3.3 Hz, CHF), 4.09 (2H, d, J 13.4 Hz, 2 × CHaHbPh), 3.95 (2H, d, J 13.4 Hz, 2 × CHaHbPh), 3.31-3.19 (2H, m, CH2CHF); 19F NMR (282 MHz, CD3OD) δF -189.0 (ddd, J 49.5, 22.5, 22.5 Hz, CHF); m/z (ES+) 288 ([M+H]+, 100%). OHOFNPhPh Experimental 169 7.3.13 - 2,2',4,4'-Tetranitrodiphenylmethane[191, 230] 172 Finely powdered potassium nitrate (27.7 g, 270 mmol, 4.5 eq) was added to a solution of aqueous sulfuric acid (200 mL, 15 M) over 0.5 h at 30 °C. Diphenylmethane (10.0 mL, 60 mmol, 1.0 eq) was added dropwise over 1 h, with the temperature maintained below 30 °C. After stirring for a further 0.5 h at rt, the solution was heated to 70 °C for 1 h, cooled to rt before iced water (1.5 L) was added, which resulted in the immediate precipitation of yellow solid that was isolated by filtration. This solid was suspended in ethanol (75 mL) and was heated under reflux for 5 min, after which the solid was re-collected by hot filtration and recrystallised from acetic acid (~75 mL), to yield 2,2',4,4'-tetranitrodiphenylmethane 172 (15.3 g, 44 mmol, 74%) as large yellow crystals: mp 172-173 °C (acetic acid) [Lit. [191, 230] 173 ºC]; 1H NMR (300 MHz, d6-DMSO) δH 8.85 (2H, d, J 2.5 Hz, Ar H-3,3′), 8.52 (2H, dd, J 8.6, 2.5 Hz, Ar H-5,5′), 7.62 (2H, d, J 8.6 Hz, Ar H-6,6′), 3.36 (2H, s, CH2); m/z (ASAP) 349 ([M+H]+, 100%), 348 ([M]+, 55%). 3 4 5 6 1 2 1' 6' 5' 4' 3' 2' NO 2 O 2 N NO 2 NO 2 Experimental 170 7.3.14 - 2,2',4,4'-Tetranitrobenzophenone[230] 173 Chromium trioxide (6.90 g, 68.9 mmol, 2.0 eq) was slowly added to a solution of 2,2',4,4'-tetranitrodiphenylmethane 172 (12.0 g, 34.5 mmol, 1.0 eq) in acetic acid (100 mL) under reflux. The resulting dark green solution was stirred under reflux for 16 h, cooled to rt with the precipitate isolated by filtration and washed with acetic acid (20 mL). The precipitate was further washed with ethanol (200 mL), water (200 mL). and was re-crystallised from acetic acid to yield 2,2',4,4'-tetranitrobenzophenone 173 (11.8 g, 32.7 mmol, 95%) as small light yellow crystals: mp 235-238 °C (acetic acid) [lit.[230] 232 °C]; 1H NMR (300 MHz, d6-DMSO) δH 8.98 (2H, d, J 2.1 Hz, Ar H-3,3′), 8.67 (2H, dd, J 8.5, 2.1 Hz, Ar H-5,5′), 8.05 (2H, d, J 8.5 Hz, Ar H-6,6′); m/z (ASAP) 363 ([M+H]+, 100%). 3 4 5 6 1 2 1' O 6' 5' 4' 3' 2' NO 2 O 2 N NO 2 NO 2 Experimental 171 7.3.15 - 3,6-Diamino-9-(10H)-acridone[230, 231] 155 A solution of stannous chloride (69.1 g, 360 mmol, 12 eq) in concentrated HCl (200 mL, 1.18 specific gravity) was heated under reflux for 0.5 h while a steady flow of argon removed HCl gas evolved from the reaction [NOTE: Evolved HCl was neutralised by passing through double Drechsel flask set up containing a solution of NaOH (30% w/v)]. 2,2',4,4'-tetranitrobenzophenone 174 (11.0 g, 30.4 mmol) and ethanol (30 mL) were added, followed by a further portion of concentrated HCl (30 mL, 1.18 specific gravity). This mixture was heated under reflux for 3 h, cooled to rt and concentrated HCl (50 mL, 1.18 specific gravity) was added. The mixture was stirred for 16 h at rt resulting in precipitation of the hydrochloride salt of 155, which was isolated by filtration. This salt was dissolved in hot aqueous HCl (0.1 M, 200 mL) and heated under reflux for 1 h before activated carbon was added. This suspension was heated under reflux for 1 h, filtered and the filtrate was basified (pH 13) with NaOH (30% w/v). The resulting precipitate was isolated by hot filtration, washed with hot aqueous NaOH (50 mL, 2 M) and hot water (200 mL), until the filtrate was neutral, to furnish 3,6-diamino-9-(10H)-acridone 155 (4.30 g, 19.0 mmol, 63%) as a light brown solid: mp >300 °C [lit. [230, 231] >300 ºC]; 1H NMR (300 MHz, d6-DMSO) δH 10.85 (1H, s, NH), 7.83 (2H, d, J 8.7 Hz, Ar H-1,8), 6.45 (2H, dd, J 8.7, 2.0 Hz, Ar H-2,7), 23 4 4a9a1 NH10 4b8a9 5 678H2N NH2O Experimental 172 6.40 (2H, d, J 2.0 Hz, Ar H-4,5), 4.00 (4H, br s, 2 × NH2); m/z (ES+) 289 ([M+Na+MeCN]+, 20%), 226 ([M+H]+, 80%). 7.3.16 - (±)-3,6-Bis(3-N,N-dibenzylamino-2-fluoropropionamido)-9-(10H)-acridone rac-169 Potassium hexamethyldisilazide (1.8 mL, 1 M in THF, 4.0 eq) was added dropwise to a suspension of 3,6-diaminoacridone 155 (102 mg, 0.45 mmol, 1.0 eq) in THF (5.0 mL) over 0.5 h at -78 °C and the mixture stirred for a further 1 h at -78 ºC. Methyl (±)-3-dibenzylamino-2-fluoropropanoate rac-120 (300 mg, 1.0 mmol, 2.2 eq) in THF (5.0 mL) was added and the mixture was stirred for 16 h whist warming to rt. The reaction was quenched by the addition of saturated aqueous NH4OH (20 mL) and ethyl acetate (20 mL) resulting in significant precipitation. The organic phase was separated and the solids were isolated from the aqueous phase by filtration. The aqueous filtrate was extracted with ethyl acetate (3 × 10 mL) and the combined organic phases were washed with brine (30 mL), dried over Na2SO4, filtered and solvent removed in vacuo to yield a dark orange solid. This solid was absorbed onto Na2SO4 for purification by silica gel column chromatography, eluting with ethyl acetate and hexane (60:40, 90:10, 100:0) to furnish (±)-3,6-bis(3-N,N-dibenzylamino-2-fluoropropionamido)-9-(10H)-acridone rac-169 (64.0 mg, 83.8 µmol, 19%) as a pale yellow solid: 1H NMR (400 MHz, d6-DMSO) δH 11.84 (1H, s, NH-10), 10.45 (2H, s, 2 × CONH), 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O N PhPhONPhPh F F Experimental 173 8.23 (2H, d, J 1.7 Hz, Ar H-4/5), 8.15 (2H, d, J 8.8 Hz, Ar H-1/8), 7.39-7.16 (22H, m, 20 × ArH, Ar H-2/7), 5.38 (2H, ddd, J 49.2, 5.9, 3.6 Hz, 2 × CHF), 3.78 (4H, d, J 13.9 Hz, 4 × CHaHbPh), 3.58 (4H, d, J 13.9 Hz, 4 × CHaHbPh), 3.10-2.94 (4H, m, 2 × CH2CHF); 19F{1H} NMR (376 MHz, d6-DMSO) δF -187.9 (2F, dt, J 49.2, 24.9 Hz, 2 × CHF); HRMS m/z (ES-) calcd. for C47H42F2N5O3 [M-H]- 762.3256, found 762.3246; m/z (ES-) 762 ([M-H]-, 100%). 7.3.17 - (±)-3,6-Bis(3-N,N-dibenzylamino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino)acridine rac-182 Phosphorous oxychloride (5.0 mL) was added to (±)-3,6-bis(3-N,N-dibenzylamino-2-fluoropropionamido)-9-(10H)-acridone rac-169 (20.0 mg, 26.2 µmol, 1.0 eq), with the resulting suspension stirred at reflux for 3 hr. The solution was cooled to 0 °C and cold Et2O (10 mL) was added, resulting in formation of a precipitate. The precipitate was isolated by filtration and washed with further Et2O (2 × 5 mL) and dissolved in CHCl3 (5 mL). The organic phase was washed with aqueous NH4OH (1 M, 5 mL), and brine (5 mL), dried over Na2SO4, filtered and the solvent removed in vacuo to yield (±)-3,6-bis(3-N,N-dibenzylamino-2-fluoropropionamido)-9-chloroacridine (18.2 mg) as a red brown solid, which was used in the next step without further purification. 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O N PhPhONPhPh 13 141514'13' NF F Experimental 174 N,N-dimethylaminoaniline (63.0 mg, 466 mmol, 20 eq) in CHCl3 (2 mL) was added dropwise to a refluxing solution of (±)-3,6-bis((3-N,N-dibenzylamino-2-fluoropropionamido)-9-chloroacridine in CHCl3 (2 mL). The mixture was heated under reflux until TLC analysis had indicated the consumption of the chloride, at which point the solvent was removed in vacuo to yield a black oil. Cold Et2O (excess) was added, resulting in the precipitation of a red-brown solid. The solids were isolated by filtration and washed with further Et2O (5 mL), dissolved in CHCl3 (5 mL) and washed with aqueous NH4OH (1 M, 5 mL) followed by brine (5 mL). The organic phase was dried over Na2SO4, filtered and the solvent removed in vacuo to yield a red solid. The solid was purified by silica gel column chromatography, eluting with CHCl3 and CH3OH (95:5), to yield (±)-3,6-bis(3-N,N-dibenzylamino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino)acridine rac-182 (4.9 mg, 5.5 µmol, 23%) as a dark red solid: 1H NMR (500 MHz, CD3OD) δH 8.28 (2H, s, Ar H-4/5), 7.93 (2H, d, J 9.0 Hz, Ar H-1/8), 7.29-7.00 (24H, m, 20 × Ar-H, Ar H-2/7, Ar H-14/14′), 6.77 (2H, d, J 8.9 Hz, Ar H-13/13′), 5.13 (2H, ddd, J 49.1, 5.3, 3.7 Hz, 2 × CHF), 3.73 (4H, d, J 13.7 Hz, 4 × CHaHbPh), 3.50 (4H, d, J 13.7 Hz, 4 × CHaHbPh), 3.07-2.95 (4H, m, 2 × CH2CHF), 2.89 (6H, s, 2 × NCH3); 19F{1H} NMR (470 MHz, CD3OD) -187.5 (2F, s, 2 × CHF); HRMS m/z (ES+) calcd. for C55H54F2N7O2 [M+H]+ 882.4307, found 882.4299; m/z (ES+) 882 ([M+H]+, 100%). Experimental 175 7.3.18 - Methyl (+)-(2R)-(diallylamino)-3-hydroxypropanoate (R)-187 Allyl bromide (12.2 mL, 141 mmol, 2.2 eq) was added to a suspension of D-serine methyl ester hydrochloride (R)-164 (10.0 g, 64.8 mmol, 1.0 eq) and K2CO3 (35.6 g, 258 mmol, 4.0 eq) in acetonitrile (300 mL) and the resulting suspension was heated under reflux for 24 h. The reaction was cooled to rt, diluted with water (300 mL) and extracted with ethyl acetate (3 × 100 mL). The organic fractions were combined, washed with brine (100 mL), dried over Na2SO4, filtered and the solvent removed in vacuo. The resulting oil was purified by silica gel column chromatography, eluting with hexane and ethyl acetate (95:5 to 90:10), to yield methyl (2R)-3-hydroxy-2-N,N- bisallylaminopropanoate (R)-187 (7.66 g, 39.8 mmol, 62%) as a colourless oil: Rf 0.1 (hexane:ethyl acetate, 90:10); ![!]!!" +80.3 (c 3.0, CHCl3); IR νmax (neat, cm-1) 3446 (OH), 2953 (C=CH), 1730 (C=O), 1645, 993, 920; 1H NMR (400 MHz, CDCl3) δH 5.75 (2H, dddd, J 17.2, 10.1, 7.9, 4.8 Hz, 2 × =CH), 5.20 (2H, dddd, J 17.2, 1.8, 1.1, 1.1 Hz, 2 × CHaHb=), 5.14 (2H, dddd, J 10.1, 1.8, 0.9, 0.9 Hz, 2 × CHaHb=), 3.75 (1H, dd, J 9.2, 4.6 Hz, CHN), 3.70 (3H, s, OCH3), 3.67-3.64 (2H, m, OCH2), 3.36 (2H, dddd, J 14.3, 4.8, 1.1, 0.9 Hz, 2 × NCHaHb), 3.20-3.14 (2H, dddd, J 14.3, 7.9, 1.1, 0.9 Hz, 2 × NCHaHb), 2.63 (1H, br s, OH); 13C NMR (101 MHz, CDCl3) δc 171.8 (CO2CH3), 135.9 (2 × =CH), 118.1 (2 × CH2=), 62.5 (CHN), 59.1 (CH2OH), 53.7 (2 × NCH2-allyl), 51.5 (OCH3); HRMS m/z (ES+) calcd. for C10H17NO3Na [M+Na]+ 222.1106, found 222.1100; m/z (ES+) 222 ([M+Na]+, 100%). ONHO O Experimental 176 7.3.19 - Methyl (–)-(2S)-(diallylamino)-3-hydroxypropanoate (S)-187 Allyl bromide (12.2 mL, 141 mmol, 2.2 eq) was added to a suspension of L-serine methyl ester hydrochloride (S)-164 (10.0 g, 64.8 mmol, 1.0 eq) and K2CO3 (35.6 g, 258 mmol, 4.0 eq) in acetonitrile (300 mL) and the resulting suspension was heated under reflux for 24 h. The reaction was cooled to rt, diluted with water (300 mL) and extracted with ethyl acetate (3 × 100 mL). The organic fractions were combined, washed with brine (100 mL), dried over Na2SO4, filtered and the solvent removed in vacuo. The resulting oil was purified by silica gel column chromatography, eluting with hexane and ethyl acetate (95:5 to 90:10), to yield methyl (2S)-3-hydroxy-2-N,N- bisallylaminopropanoate (S)-187 (7.02 g, 36.5 mmol, 57%) as a colourless oil: Rf 0.1 (hexane:ethyl acetate, 90:10); [!]!!" -81.3 (c 2.9, CHCl3); ! IR νmax (neat, cm-1) 3446 (OH), 2926 (C=CH), 1730 (C=O), 1645, 993, 920; 1H NMR (400 MHz, CDCl3) δH 5.75 (2H, dddd, J 17.2, 10.1, 7.9, 4.8 Hz, 2 × =CH), 5.20 (2H, dddd, J 17.2, 1.8, 1.1, 1.1 Hz, 2 × CHaHb=), 5.14 (2H, dddd, J 10.1, 1.8, 0.9, 0.9 Hz, 2 × CHaHb=), 3.70 (3H, s, OCH3), 3.75 (1H, dd, J 9.2, 4.6 Hz, CHN), 3.67 (1H, dd, J 14.3, 4.6 Hz, OCHaHb), 3.64 (1H, d, J 14.3, 9.2 Hz, OCHaHb), 3.36 (2H, dddd, J 14.3, 4.8, 1.1, 0.9 Hz, 2 × NCHaHb-allyl), 3.20-3.14 (2H, m, 2 × NCHaHb-allyl), 2.63 (1H, br s, OH); 13C NMR (101 MHz, CDCl3) δC 171.8 (CO2CH3), 135.9 (2 × =CH), 118.1 (2 × CH2=), ONHO O Experimental 177 62.5 (CHN), 59.1 (CH2OH), 53.7 (2 × CH2N), 51.5 (OCH3); HRMS m/z (ES+) calcd. for C10H18NO3 [M+H]+ 200.1276, found 200.1277; m/z (ES+) 200 ([M+H]+, 100%). 7.3.20 - Methyl (–)-(2S)-3-diallylamino-2-fluoropropanoate (S)-188 Diethylaminosulfur trifluoride 32 (2.2 mL, 18.1 mmol, 1.2 eq) was added to a solution of methyl (2R)-(diallylamino)-3-hydroxypropanoate (R)-187 (3.00 g, 15.1 mmol, 1.0 eq) in THF (80 mL) over a period of 5 min at 0 ºC. The resulting solution was stirred at 0 °C for 1 h and the reaction was quenched by the addition of solid K2CO3 (excess) and water (1 mL). As the effervescence subsided, the solution was diluted further with water (20 mL) and the organic fractions extracted with diethyl ether (3 × 20 mL). The organic fractions were combined and washed with brine (20 mL), dried over Na2SO4, filtered and the solvent removed in vacuo. The resulting oil was purified by silica gel column chromatography, eluting with hexane:ethyl acetate (95:5), to yield methyl (2S)-3-diallylamino-2-fluoropropanoate (S)-188 (2.06 g, 9.2 mmol, 61%) as a colourless oil: Rf 0.15 (hexane:ethyl acetate, 95:5);![!]!!" -9.7 (c 0.97, CHCl3); IR νmax (neat, cm-1) 2956, 2814, 1767 (C=O), 1643, 1440, 1213, 922; 1H NMR (500 MHz, CDCl3) δH 5.80 (2H, dddd, J 16.9, 10.4, 6.9, 5.9 Hz, 2 × =CH), 5.20-5.13 (4H, m, 2 × CH2=), 5.05 (1H, ddd, J 49.7, 6.4, 3.2 Hz, CHF), 3.79 (3H, s, OCH3), 3.28-3.24 (2H, m, 2 × NCHaHb-allyl), 3.14-3.10 (2H, m, 2 × NCHaHb-allyl), ON F O Experimental 178 2.99 (1H, ddd, J 25.8, 14.7, 6.4 Hz, CHaHbCHF), 2.96 (1H, ddd, J 26.6, 14.7, 3.2 Hz, CHaHbCHF); 13C NMR (101 MHz, CDCl3) δC 169.5 (d, J 23.5 Hz, CO2CH3), 135.3 (2 × =CH), 118.0 (2 × CH2=), 89.5 (d, J 187.1 Hz, CHF), 57.6 (2 × NCH2-allyl), 54.2 (d, J 20.2 Hz, CH2), 52.4 (OCH3); 19F NMR (470 MHz, CDCl3) δC -191.4 (ddd, J 49.7, 26.6, 25.8 Hz, CHF); HRMS m/z (ES+) calcd. for C10H16NO2FNa [M+Na]+ 224.1058, found 224.1064; m/z (ES+) 224 ([M+Na]+, 100%), 202 ([M+H]+, 20%). Enantiomeric excess determined by chiral HPLC (Chiralcel OD-H 5% iPrOH in hexane, 0.5 mL/min, tr maj = 9.57 min >95%, tr min = 9.33 min <5%). 7.3.21 - Methyl (+)-(2R)-3-diallylamino-2-fluoropropanoate (R)-188 Following the procedure set out for methyl (2S)-3-diallylamino-2-fluoropropanoate (S)-188, starting from methyl 2-(S)-3-hydroxypropanoate (S)-187 (3.10 g, 15.6 mmol, 1.0 eq) with diethylaminosulfur trifluoride 32 (2.3 mL, 18.7 mmol, 1.2 eq) in THF (80 mL), the reaction yielded methyl (2R)-3-diallylamino-2-fluoropropanoate (R)-188 (2.19 g, 10.9 mmol, 69%) as a colourless oil: Rf 0.15 (hexane:ethyl acetate, 95:5); IR νmax (neat, cm-1) 2956, 2815, 1767 (C=O), 1643, 1440, 1214, 1069, 923;![!]!!" +9.8 (c 0.97, CHCl3); 1H NMR (400 MHz, CDCl3) δH 5.80 (2H, dddd, J 17.2, 10.2, 7.0, 6.0 Hz, 2 × =CH), 5.20-5.13 (4H, m, 2 × CH2=), 5.05 (1H, ddd, J 49.7, 6.3, 3.2 Hz, CHF), 3.73 (3H, s, OCH3), 3.29-3.23 (2H, m, 2 × NCHaHb-allyl), ON F O Experimental 179 3.15-3.09 (2H, m, 2 × NCHaHb-allyl), 2.99 (1H, ddd, J 25.8, 14.7, 6.3 Hz, CHaHbCHF), 2.97 (1H, ddd, J 26.6, 14.7, 3.2 Hz, CHaHbCHF); 13C NMR (101 MHz, CDCl3) δC 169.5 (d, J 23.5 Hz, CONH), 135.3 (2 × =CH), 118.0 (2 × CH2=), 89.5 (d, J 186.1 Hz, CHF), 57.6 (2 × NCH2-allyl), 54.2 (d, J 20.2 Hz, CH2CHF), 52.4 (OCH3); 19F NMR (376 MHz, CDCl3) δF -191.9 (ddd, J 49.7, 26.6, 25.8 Hz, CHF); HRMS m/z (ES+) calcd. for C10H17NO2F [M+H]+, 202.1233, found 202.1235; m/z (ES+) 202 ([M+H]+, 100%). Enantiomeric excess determined by chiral HPLC (Chiralcel OD-H 5% iPrOH in hexane, 0.5 mL/min, tr maj = 9.33 min >95%, tr min = 9.57 min <5%). 7.3.22 - 3,6-Bis((2R)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-195 Potassium hexamethyldisilazide (4.4 mL, 1M in THF, 5.0 eq) was added dropwise to a suspension of 3,6-diaminoacridone 155 (200 mg, 0.88 mmol, 1.0 eq) in THF (10 mL) over 0.5 h at -78 °C and the mixture stirred for a further 1 h at -78 ºC. Methyl (2R)-3-diallylamino-2-fluoropropanoate (R)-188 (450 mg, 2.2 mmol, 2.5 eq) in THF (10 mL) was gradually added via cannula to the homogeneous orange solution and the mixture was stirred for 16 h whist warming to rt. The reaction was quenched by the addition of saturated aqueous Na2CO3 (20 mL) resulting in significant precipitation. The 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON F F Experimental 180 solids were removed by filtration and the filtrate was extracted with ethyl acetate (3 × 20 mL). The combined organic phases were washed with brine (30 mL), dried over Na2SO4, filtered and solvent removed in vacuo to yield a dark orange solid. This solid was absorbed onto Na2SO4 for purification by silica gel column chromatography, eluting with ethyl acetate and hexane (60:40, 90:10, 100:0) to furnish 3,6-bis((2R)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-195 (132 mg, 0.19 mmol, 22%) as a pale yellow solid: Rf 0.31 (ethyl acetate:hexane, 80:20); mp 240 ºC (dec.) ; [!]!"#!" 3.7 (c 0.9, CH3OH); IR νmax (KBr, cm-1) 3271, 3200, 2851, 1688, 1629, 1599, 1462, 1300, 1269, 1190, 1118, 997, 921; 1H NMR (400 MHz, CD3OD) δH 8.14 (2H, d, J 8.9 Hz, Ar H-1,8), 8.08 (2H, d, J 1.8 Hz, Ar H-4,5), 7.17 (2H, dd, J 8.9, 1.8 Hz, Ar H-2,7), 5.77 (4H, dddd, J 17.0, 10.4, 6.6, 6.6 Hz, 4 × =CH), 5.15-5.03 (10H, m, 4 × CH2=, 2 × CHF), 3.21-3.08 (8H, m, 4 × NCH2-allyl), 2.99 (4H, dd, J 25.8, 5.0 Hz, 2 × CH2CHF); 13C NMR (101 MHz, CD3OD) δC 178.5 (Ar C-9), 170.1 (d, J 20.6 Hz, 2 × CONH), 143.7 (2 × C-8a,9a & 2 × Ar C-4a,4b), 136.5 (4 × =CH), 128.2 (2 × Ar CH-1,8), 118.8 (4 × CH2=), 118.6 (2 × Ar C-3,6), 115.9 (2 × Ar CH-2,7), 107.7 (2 × Ar CH-4,5), 92.1 (d, J 187.9 Hz, 2 × CHF), 58.5 (4 × NCH2-allyl), 55.5 (d, J 20.4 Hz, 2 × CH2CHF); 19F NMR (376 MHz, CD3OD) δF -191.2 (2F, dt, J 49.5, 24.7 Hz, 2 × CHF); HRMS m/z (ES-) calcd. for C31H34F2N5O3 [M-H]- requires 562.2630, found 562.2637; m/z (ES-) 562 ([M-H]-, 100%). Experimental 181 7.3.23 - 3,6-Bis((2S)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-195 Following the procedure set out for 3,6-bis((2R)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-195, starting from methyl (2S)-3-diallylamino-2-fluoropropanoate (S)-188 (450 mg, 2.2 mmol, 2.5 eq) the reaction furnished 3,6-bis((2S)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-195 (138 mg, 0.20 mmol, 23%) as a pale yellow solid: Rf 0.31 (ethyl acetate:hexane, 80:20); mp 240 ºC (dec.); [!]!"#!" -3.6 (c 1.1, CH3OH); IR νmax (KBr, cm-1) 3269, 3200, 2850, 1691, 1628, 1598, 1462, 1299, 1268, 1190, 1118, 997, 922; 1H NMR (400 MHz, CD3OD) δH 8.24 (2H, d, J 8.9 Hz, Ar H-1,8), 8.17 (2H, d, J 2.0 Hz, Ar H-4,5), 7.27 (2H, dd, J 8.9, 2.0 Hz, Ar H-2,7), 5.92-5.82 (4H, m, 4 × =CH), 5.26-5.11 (10H, m, 4 × CH2= and 2 × CHF), 3.32-3.17 (8H, m, 4 × NCH2-allyl), 3.09 (4H, dd, J 25.9, 5.0 Hz, 2 × CH2CHF); 13C NMR (101 MHz, CD3OD) δC 178.5 (Ar C-9), 170.0 (d, J 20.6 Hz, 2 × CONH), 143.7 (2 × Ar C-8a,9a & 2 × Ar C-4a,4b), 136.1 (4 × =CH), 128.2 (2 × Ar CH-1,8), 118.7 (4 × CH2=), 118.6 (2 × Ar C-3,6), 115.9 (2 × Ar CH-2,7), 107.7 (2 × Ar CH-4,5), 92.1 (d, J 187.9 Hz, 2 × CHF), 58.4 (4 × NCH2-allyl), 55.6 (d, J 20.4 Hz, 2 × CH2CHF); 19F NMR (376 MHz, CD3OD) δF -191.2 (2F, dt, J 49.5, 24.7 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C31H36F2N5O3 [M+H]+ requires 564.2786, found 564.2789; m/z (ES+) 564 ([M+H]+, 100%), 586 ([M+H]+, 50%). 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON F F Experimental 182 7.3.24 - 3,6-Bis((2R)-3-amino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-196.HCl 1,4-Di(phenylphosphino)butane (30 mg, 71 µmol, 10 mol%/allyl group) was added to a solution of palladium acetylacetonate (33 mg, 36 µmol, 5 mol%/allyl group) in degassed THF (5.0 mL) and the resulting solution was stirred at rt for 15 min. 3,6-Bis((2R)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-195 (100 mg, 0.18 mmol, 1.0 eq) and 2-mercaptosalicylic acid (137 mg, 890 mmol, 5.0 eq) in THF (5.0 mL) were added via cannula to the catalyst solution and the reaction was heated under reflux for 3 h. The reaction was cooled to rt and water (10 mL) and HCl (1 M, 60 µL) were added, resulting in precipitation of a yellow solid. The precipitate was isolated by filtration and the residue was washed repeatedly with water (3 × 10 mL). The filtrate was reduced in vacuo to furnish a yellow solid. This solid was reconstituted in water, filtered and the filtrate was lyophilised to yield 3,6-bis(3-amino-(2R)-fluoropropionamido)-9-(10H)-acridone dihydrochloride (R,R)-196.HCl (81 mg, 0.17 mmol, 95% based on dichloride salt) as a pale yellow amorphous solid: mp 160 ºC (dec.); [!]!"#!" 27.8 (c 0.9, CH3OH); IR (KBr, cm-1) 2960, 2921, 1677, 1563, 1416, 1207, 1149, 1063, 1042, 741; 1H NMR (300 MHz, D2O) δH 7.73 (2H, d, J 9.0 Hz, Ar H-1,8), 7.27 (2H, d, J 1.9 Hz, Ar H-4,5), 6.95 (2H, dd, J 9.0, 1.9 Hz, Ar H-2,7), 5.45 (2H, ddd, J 48.5, 8.1, 3.2 Hz, 2 × CHF), 3.77-3.51 (4H, m, 2 × CH2CHF); 13C NMR (126 MHz, D2O/d6-DMSO) δC 178.3 (Ar C-9), 167.4 (d, 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NH3ClOClH3N F F Experimental 183 J 19.6 Hz, 2 × CONH), 142.6 (2 × Ar C-8a,9a), 142.5 (2 × Ar C-4a,4b), 128.3 (2 × Ar CH-1,8), 118.2 (2 × Ar C-3,6), 116.3 (2 × Ar CH-2,7), 107.9 (2 × Ar CH-4,5), 88.7 (d, J 188.2 Hz, 2 × CHF), 41.9 (d, J 20.6, 2 × CH2CHF); 19F NMR (286 MHz, D2O/d6-DMSO) δF -196.9 (2F, ddd, J 48.5, 29.1, 19.3 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C19H20F2N5O3 [M+H]+ requires 404.1529, found 404.1530; m/z (ES+) 404 ([M+H]+, 100%). 7.3.25 - 3,6-Bis((2S)-3-amino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-196.HCl Following the procedure set out for 3,6-bis((2S)-3-amino-2-fluoropropionamido)-9-(10H)-acridone dihydrochloride (R,R)-196, starting from 3,6-bis((2S)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-195 (51 mg, 0.09 mmol) the reaction yielded 3,6-bis((2S)-3-amino-2-fluoropropionamido)-9-(10H)-acridone dihydrochloride (S,S)-196.HCl (38 mg, 0.08 mmol, 89%) as a pale yellow amorphous solid: mp 160 ºC (dec.); [!]!"#!" -22.5 (c 1.1, CH3OH); 1H NMR (300 MHz, D2O) δH 7.73 (2H, d, J 9.0 Hz, Ar H-1,8), 7.27 (2H, d, J 1.6 Hz, Ar H-4,5), 6.95 (2H, dd, J 9.0, 1.6 Hz, Ar H-2,7), 5.45 (2H, ddd, J 48.4, 8.1, 3.1 Hz, 2 × CHF), 3.68 (2H, ddd, J 29.0, 14.4, 3.1 Hz, 2 × CHaHbCHF), 3.58 (2H, ddd, J 19.4, 14.4, 8.1 Hz, 2 × CHaHbCHF); 13C NMR (126 MHz, D2O) δC 178.3 (Ar C-9), 167.4 (d, J 19.7 Hz, 2 × CONH), 142.6 (2 × Ar C-8a,9a), 142.5 (2 × Ar C-4a,4b), 128.3 (2 × Ar CH-1,8), 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NH3ClOClH3N F F Experimental 184 118.2 (2 × Ar C-3,6), 116.3 (2 × Ar CH-2,7), 107.9 (2 × Ar CH-4,5), 88.7 (d, J 188.2 Hz, 2 × CHF), 41.9 (d, J 20.7 Hz, 2 × CH2CHF); 19F NMR (282 MHz, D2O) δF -195.5 (2F, ddd, J 48.4, 29.0, 19.4 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C19H20F2N5O3 [M+H]+ requires 404.1529, found 404.1528; m/z (ES+) 404 ([M+H]+, 100%). 7.3.26 - 3,6-Bis((2R)-3-N,N-bisallylamino-2-fluoropropionamido)-9-(4-dimethylamino phenylamino)acridine (R,R)-206 Phosphorous oxychloride (10 mL) was added to 3,6-bis((2R)-3-amino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-195 (100 mg, 0.18 mmol, 1.0 eq). The resulting bright orange suspension was stirred at rt until TLC analysis had indicated consumption of the starting material. The solution was cooled to 0 °C and cold Et2O (20 mL) was added, resulting in formation of a precipitate. The precipitate was isolated by filtration and washed with further Et2O (2 × 10 mL) and dissolved in CHCl3 (10 mL). The organic phase was washed with aqueous NH4OH (1 M, 10 mL), and brine (10 mL), dried over Na2SO4, filtered and the solvent removed in vacuo to yield 3,6-bis((2R)-3-N,N-diallylamino-2-fluoropropionamido)-9-chloro 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' NF F Experimental 185 acridine (74 mg, 72%) as a red brown solid, which was used in the next step without further purification. N,N-dimethylaminoaniline monohydrochloride (445 mg, 2.60 mmol, 20 eq) was dissolved in saturated aqueous Na2CO3 and extracted with Et2O (3 × 20 mL). The organic fractions were combined, washed with brine (20 mL), dried over Na2SO4, filtered and the solvent was removed in vacuo to yield N,N-dimethylaminoaniline as a light brown oil. This oil was dissolved in CHCl3 (10 mL) and the solution was gradually added via cannula to a refluxing solution of 3,6-bis((2R)-3-N,N-diallylamino-2-fluoropropionamido)-9-chloroacridine (74 mg, 0.13 mmol, 1.0 eq) in CHCl3 (10 mL) over 30 min. The mixture was heated under reflux until TLC analysis had indicated the consumption of the chloride, at which point the solvent was removed in vacuo to yield a purple oil. Cold Et2O (excess) was added, resulting the precipitation of a red-brown solid. The solids were isolated by filtration and washed with further Et2O (20 mL), dissolved in CHCl3 (20 mL) and washed with aqueous NH4OH (1 M, 10 mL) followed by brine (10 mL). The organic phase was dried over Na2SO4, filtered and the solvent removed in vacuo to yield a red solid. The solid was purified by silica gel column chromatography, eluting with CHCl3 and CH3OH (95:5), to yield 3,6-bis((2R)-3-N,N-bisallylamino-2-fluoropropionamido)-9-(4-dimethylamino phenylamino)acridine (R,R)-206 (53 mg, 0.08 mmol, 61%) as a dark red solid: Rf 0.09 (CHCl3:CH3OH, 95:5); mp 195 ºC (dec.); IR νmax (KBr disc, cm-1) 3428, 3077, 1700 (C=O), 1633 (C=O), 1521, 1469, 1446, 1359, 1257, 1065, 922; 1H NMR (400 MHz, CD3OD) δH 8.37 (2H, d, J 1.9 Hz, Ar H-4,5), 7.96 (2H, d, J 9.4 Hz, Ar H-1,8), 7.27 (2H, dd, J 9.4, 1.9 Hz, Ar H-2,7), 7.10 (2H, d, J 8.9 Hz, Ar H-14,14'), 6.78 (2H, d, J 8.9 Hz, Ar H-13,13'), 5.85 (4H, dddd, J 17.0, 10.4, 6.5, 6.5 Hz, 4 × =CH), 5.24-5.14 (8H, m, 4 × CH2=), 5.20 (2H, dt, J 49.3, 5.0 Hz, 2 × CHF), 3.31-3.25 (4H, m, 4 × NCHaHb-allyl), 3.24-3.19 (4H, m, 4 × NCHaHb-allyl), 3.09 (4H, Experimental 186 dd, J 26.0, 5.0 Hz, 2 × CH2CHF), 3.00 (6H, s, 2 × NCH3); 13C NMR (101 MHz, CD3OD) δC 170.3 (d, J 20.9 Hz, 2 × CONH), 153.9 (Ar C-9), 151.0 (Ar C-15), 144.3 (2 × Ar C-4a,4b), 143.8 (2 × Ar C-8a,9a), 136.2 (4 × =CH), 132.0 (Ar C-12), 127.7 (2 × Ar CH-1,8), 126.2 (2 × Ar CH-14,14'), 118.7 (4 × CH2=), 117.9 (2 × Ar CH-2,7), 114.4 (2 × Ar CH-13,13'), 112.2 (2 × Ar C-3,6), 109.0 (2 × Ar CH-4,5), 92.2 (d, J 187.9 Hz, 2 × CHF), 58.4 (4 × NCH2-allyl), 55.5 (d, J 20.0 Hz, 2 × CH2CHF), 40.9 (2 × NCH3); 19F NMR (376 MHz, CD3OD) δF 191.2 (2F, dt, J 49.3, 26.0 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C39H46F2N7O2 [M+H]+ requires 682.3676, found 682.3670; m/z (ES+) 682 ([M+H]+, 100%). 7.3.27 - 3,6-Bis((2S)-3-N,N-bisallylamino-2-fluoropropionamido)-9-(4-dimethylamino phenylamino) acridine (S,S)-206 Following the procedure set out for 3,6-bis((2R)-3-bisallylamino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino) acridine (R,R)-206, starting from 3,6-bis((2S)-3-N,N-diallylamino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-195 (100 mg, 0.18 mmol), the reaction yielded 3,6-bis((2S)-3-N,N-bisallylamino- 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' N FF Experimental 187 2-fluoropropionamido)-9-(4-dimethylaminophenylamino) acridine (S,S)-206 (41.7 mg, 0.063 mmol, 48%) as a red solid: Rf 0.09 (CHCl3:CH3OH, 95:5); mp 195 ºC (dec.); IR νmax (KBr disc, cm-1) 3453, 3077, 1697 (C=O), 1633 (C=O), 1521, 1469, 1446, 1359, 1257, 1064, 922; UV-vis (CH3OH) λmax 268 nm (log10 ε 11.34), 293 nm (log10 ε 8.04), 364 nm (log10 ε 4.00), 425 nm (log10 ε 1.63); 1H NMR (400 MHz, CD3OD) δH 8.39 (2H, d, J 1.9 Hz, Ar H-4,5), 7.97 (2H, d, J 9.4 Hz, Ar H-1,8), 7.31 (2H, dd, J 9.4, 1.9 Hz, Ar H-2,7), 7.15 (2H, d, J 9.0 Hz, Ar H-14,14'), 6.81 (2H, d, J 9.0 Hz, Ar H-13,13'), 5.85 (4H, dddd, J 17.0, 10.4, 6.5, 6.5 Hz, 4 × =CH), 5.28-5.13 (8H, m, 4 × CH2=), 5.20 (2H, dt, J 49.4, 4.9 Hz, 2 × CHF), 3.32-3.16 (8H, m, 4 × NCH2-allyl), 3.09 (4H, dd, J 26.0, 4.9 Hz, CH2CHF), 3.00 (6H, s, 2 × NCH3); 13C NMR (101 MHz, CD3OD) δC 170.3 (d, J 20.9 Hz, 2 × CONH), 154.4 (Ar C-9), 151.3 (Ar C-15), 144.2 (2 × Ar C-4a,4b), 143.8 (2 × Ar C-8a,9a), 136.2 (4 × =CH), 131.2 (Ar C-12), 127.8 (2 × Ar CH-1,8) 126.6 (2 × Ar CH-14,14'), 118.7 (4 × CH2=), 118.0 (2 × Ar CH-2,7), 114.3 (2 × Ar CH-13,13´), 111.8 (2 × Ar C-3,6), 108.4 (2 × Ar CH-4,5), 92.2 (d, J 187.9 Hz, 2 × CHF), 58.4 (4 × NCH2-allyl), 55.5 (d, J 20.0 Hz, 2 × CH2CHF), 40.8 (2 × NCH3); 19F NMR (376 MHz, CD3OD) δF 191.4 (2F, dt, J 49.4, 26.0 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C39H46F2N7O2 [M+H]+ requires 682.3676, found 682.3666; m/z (ES+) 682 ([M+H]+, 100%). Experimental 188 7.3.28 - 3,6-Bis((2R)-3-amino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino) acridine dihydrochloride (R,R)-208.HCl 1,4-Di(phenylphosphino)butane (25.2 mg, 59 µmol, 20 mol%/allyl group) was added to a solution of tris(dibenzylideneacetone)dipalladium (26.5 mg, 29 µmol, 10 mol%/allyl group) in THF (8 mL) and stirred for 15 min until the solution turned yellow. The solution was added to a solution of 3,6-bis((2R)-3-N,N-bisallylamino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino) acridine (R,R)-206 (48.8 mg, 72 µmol, 1.0 eq) and 2-mercaptosalicylic acid (55.0 mg, 0.356 mmol, 5.0 eq) in THF (5.0 mL) via cannula. The resulting solution was heated under reflux for 3 h before being cooled to rt and diluted with distilled water (5 mL) and HCl (1 M, 60 µL), which caused precipitation of a solid. The precipitate was isolated by filtration and the residue was washed repeatedly with distilled water (3 × 10 mL). The water/THF filtrate was concentrated in vacuo to yield a yellow solid. The solid was redissolved in water and filtered. The filtrate was lyophilised to yield 3,6-bis((2R)-3-amino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino)acridine dihydrochloride (R,R)-208.HCl (25.2 mg, 42 µmol, 59% based on dichloride salt) as a red amorphous solid: IR νmax (KBr disc, cm-1) 3424, 1700 (C=O), 1633 (C=O), 1610, 1594, 1546, 1516, 1467, 1447, 1386, 1257; 1H NMR (500 MHz, D2O) δH 8.01 (2H, s, Ar H-4,5), 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NH3ClOClH3N 13 141514'13' N FF Experimental 189 7.70 (2H, d, J 9.4 Hz, Ar H-1,8), 7.45 (2H, d, J 8.7 Hz, Ar H-14,14´), 7.25 (2H, d, J 8.7 Hz, Ar H-13,13´), 7.19 (2H, d, J 9.4 Hz, Ar H-2,7), 5.49 (2H, ddd, J 48.3, 8.3, 3.0 Hz, 2 × CHF), 3.63 (2H, ddd, J 29.3, 14.4, 3.0 Hz, 2 × CHaHbCHF), 3.53 (2H, ddd, J 19.1, 14.4, 8.3 Hz, 2 × CHaHbCHF), 3.15 (6H, s, 2 × N CH3); 13C NMR (126 MHz, D2O) δC 166.8 (d, J 20.0 Hz, 2 × CONH), 153.2 (Ar C-9), 152.0 (Ar C-15), 142.1 (2 × Ar C-4a,4b), 140.7 (2 × Ar C-8a,9a), 133.1 (Ar C-12), 126.5 (2 × Ar CH-1,8), 125.5 (2 × Ar CH-14,14'), 117.5 (2 × Ar CH-2,7), 110.3 (2 × Ar C-3,6), 106.9 (2 × Ar CH-4,5), 87.4 (d, J 189.2 Hz, 2 × CHF), 65.9 (2 × NCH3), 40.5 (d, J 20.6 Hz, 2 × CH2CHF); 19F NMR (470 MHz, D2O) δF -195.7 (2F, ddd, J 48.3, 29.3, 19.1 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C27H30F2N7O2 [M+H]+ requires 522.2429, found 522.2421; m/z (ES+) 522 ([M+H]+, 100%). 7.3.29 - 3,6-Bis((2S)-3-amino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino) acridine dihydrochloride (S,S)-208.HCl Following the procedure set out for 3,6-bis((2R)-3-amino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino) acridine dihydrochloride (R,R)-208.HCl, starting from 3,6-bis((2S)-3-N,N-bisallylamino-2-fluoropropionamido)-9-(4-dimethylaminophenyl 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NH3ClOClH3N 13 141514'13' N FF Experimental 190 amino)acridine (S,S)-206 (41.7 mg, 61 µmol), the reaction yielded 3,6-bis((2S)-3-amino-2-fluoropropionamido)-9-(4-dimethylaminophenylamino)acridine dihydrochloride (S,S)-208.HCl (19.3 mg, 33 µmol, 53%) as a red solid: IR νmax (KBr disc, cm-1) 3424, 2927, 1700 (C=O), 1633 (C=O), 1610, 1595, 1517, 1446, 1257; 1H NMR (500 MHz, D2O) δH 8.01 (2H, d, J 1.9 Hz, Ar H-4,5), 7.70 (2H, d, J 9.4 Hz, Ar H-1,8), 7.45 (2H, d, J 8.7 Hz, Ar H-14,14'), 7.25 (2H, d, J 8.7 Hz, Ar H-13,13'), 7.19 (2H, dd, J 9.4, 1.9 Hz, Ar H-2,7), 5.49 (2H, ddd, J 48.3, 8.2, 3.0 Hz, 2 × CHF), 3.63 (2H, ddd, J 29.3, 14.4, 3.0 Hz, 2 × CHaHbCHF), 3.53 (2H, ddd, J 19.2, 14.4, 8.2 Hz, 2 × CHaHbCHF), 3.15 (6H, s, 2 × NCH3); 13C NMR (126 MHz, D2O) δC 166.8 (d, J 20.0 Hz, 2 × CONH), 153.2 (Ar C-9), 152.0 (Ar C-15), 142.1 (2 × Ar C-4a,4b), 140.7 (2 × Ar C-8a,9a), 133.1 (Ar C-12), 126.5 (2 × Ar CH-14,14´), 125.5 (2 × Ar CH-1,8), 117.5 (2 × Ar CH-2,7), 110.3 (2 × Ar C-3,6), 106.9 (2 × Ar CH-4,5), 87.4 (d, J 189.2 Hz, 2 × CHF), 65.9 (2 × NCH3), 40.5 (d, J 20.6 Hz, 2 × CH2CHF); 19F NMR (470 MHz, D2O) δF -195.7 (2F, ddd, J 48.3, 29.3, 19.2 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C27H30F2N7O2 [M+H]+ requires 522.2429, found 522.2440; m/z (ES+) 522 ([M+H]+, 100%). Experimental 191 7.3.30 - Methyl (–)-(2R)-3-dipropylamino-2-fluoropropanoate (R)-209 Palladium on carbon (20%, 205 mg, 10 mol%) was added to a solution of methyl (2R)-3-diallylamino-2-fluoropropanoate (R)-188 (755 mg, 3.88 mmol, 1.0 eq) in ethyl acetate (10 mL). The resulting suspension was stirred vigorously under a hydrogen atmosphere at rt for 24 h. The solids were removed by filtration and the residue was washed repeatedly with ethyl acetate (3 × 20 mL). The filtrate was made deliberately wet with water (1 mL) followed by the addition of MgSO4 to remove any remaining fine particulate. The solvent was removed in vacuo to yield a oil that was purified by silica gel column chromatography, eluting with ethyl acetate and hexane (20:80), to yield methyl (2R)-3-dipropylamino-2-fluoropropanoate (R)-209 (393 mg, 1.91 mmol, 51%) as a colourless oil: Rf 0.16 (hexane:ethyl acetate, 95:5); [!]!!" -23.0 (c 1.0, CHCl3); IR νmax (neat, cm-1) 2960, 1769 (C=O), 1462, 1212, 1074; 1H NMR (400 MHz, CDCl3) δH 4.95 (1H, ddd, J 49.8, 5.5, 5.5 Hz, CHF), 3.72 (3H, s, OCH3), 2.90-2.88 (2H, m, CH2CHF), 2.47-2.33 (4H, m, 2 × NCH2CH2), 1.41-1.32 (4H, m, 2 × NCH2CH2), 0.79 (6H, t, J 7.4 Hz, 2 × CH2CH3); 13C NMR (101 MHz, CDCl3) δC 169.8 (d, J 23.9 Hz, CO2CH3), 89.6 (d, J 186.5 Hz, CHF), 57.0 (2 × NCH2CH2), 55.8 (d, J 20.2 Hz, CH2CHF), 52.3 (OCH3), 20.5 (2 × NCH2CH2), 11.8 (2 × CH2CH3); 19F NMR (376 MHz, CDCl3) δF -192.0 (ddd, J 49.8, 25.9, 25.9 Hz, CHF); HRMS m/z (ES+) calcd. for C10H21FNO2 [M+H]+, requires 206.1551, found 206.1550; m/z (ES+) 206 ([M+H]+, 100%). ON F O Experimental 192 7.3.31 - Methyl (+)-(2S)-3-dipropylamino-2-fluoropropanoate (S)-209 Following the procedure set out for methyl 3-dipropylamino-(2R)-fluoropropanoate (R)-209, starting from methyl (2S)-3-diallylamino-2-fluoropropanoate (S)-188 (830 mg, 414 mmol) the reaction yielded methyl (2S)-3-dipropylamino-2-fluoropropanoate (S)-209 (520 mg, 2.25 mmol, 61%) as a colourless oil: Rf 0.16 (hexane:ethyl acetate, 95:5); [!]!!"! +22.5 (c 1.0, CHCl3); IR νmax (neat, cm-1) 2959, 1769 (C=O), 1461, 1212, 1073; 1H NMR (300 MHz, CDCl3) δH 5.01 (1H, ddd, J 49.7, 5.4, 5.4 Hz, CHF), 3.79 (3H, s, OCH3), 3.01-2.92 (2H, m, CH2CHF), 2.55-2.38 (4H, m, 2 × NCH2CH2), 1.43 (4H, tq, J 7.4, 7.4 Hz, 2 × NCH2CH2), 0.86 (6H, t, J 7.4 Hz, 2 × CH2CH3); 13C NMR (75 MHz, CDCl3) δC 169.8 (d, J 23.8, CO2CH3), 89.7 (d, J 187.1, CHF), 57.0 (2 × NCH2CH2), 55.8 (d, J 20.2, CH2CHF), 52.3 (OCH3), 20.5 (2 × CH2CH2CH3), 11.8 (2 × CH2CH3); 19F NMR (282 MHz, CDCl3) δF -191.9 (ddd, J 49.7, 26.0, 26.0 Hz, CHF); HRMS m/z (ES+) calcd. for C10H21FNO2 [M+H]+, requires 206.1551, found 206.1550; m/z (ES+) 206 ([M+H]+, 100%). ON F O Experimental 193 7.3.32 - 3,6-Bis((2R)-3-N,N-dipropylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-210 Potassium hexamethyldisilazide (4.1 mL, 1 M in THF, 5.0 eq) was added dropwise to a suspension of 3,6-diaminoacridone (188 mg, 0.83 mmol, 1.0 eq) in THF (10 mL) at -78 °C over 30 min and stirred for a further 1 h at -78 ºC. Methyl (2R)-3-dipropylamino-2-fluoropropanoate (R)-209 (375 mg, 1.83 mmol, 2.2 eq) in THF (10 mL) was gradually added via cannula to the homogeneous orange solution and the mixture was stirred for 16 h while gently warming through to rt. The reaction was quenched by the addition of saturated aqueous Na2CO3 (20 mL), resulting in significant precipitation, which was removed by filtration. The filtrate was extracted with ethyl acetate (3 × 20 mL). The organic phases were combined and washed with brine (30 mL), dried over Na2SO4, filtered and solvent removed in vacuo to yield a dark orange solid. The solids were absorbed onto Na2SO4 for purification by silica gel column chromatography, eluting with ethyl acetate and hexane (30:70, 90:10, 100:0) to furnish 3,6-bis((2R)-3-N,N-dipropylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-210 (142 mg, 0.14 mmol, 30%) as a pale yellow solid: Rf 0.17 (ethyl acetate:hexane, 80:20); mp 215 ºC (dec.); [!]!"#!" +3.7 (c 0.88, CH3OH), [!]!"#!" +4.0 (c 0.88, CH3OH); IR νmax (KBr, cm-1) 3286, 2961, 1685 (C=O), 1633 (C=O), 1600, 1536, 1464; 1H NMR (300 MHz, CD3OD) δH 8.25 (2H, d, J 8.9 Hz, 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON F F Experimental 194 Ar H-1,8), 8.21 (2H, d, J 1.9 Hz, Ar H-4,5), 7.29 (2H, dd, J 8.9, 1.9 Hz, Ar H-2,7), 5.16 (2H, dt, J 49.4, 4.9 Hz, 2 × CHF), 3.07 (4H, dd, J 27.9, 4.9 Hz, 2 × CH2CHF), 2.62-2.47 (8H, m, 4 × CH2N), 1.54-1.48 (8H, m, 4 × CH3CH2CH2), 0.89 (12H, t, J 7.4 Hz, 4 × CH3CH2); 13C NMR (75.5 MHz, CD3OD) δC 178.6 (Ar C-9), 170.3 (d, 20.6 Hz, 2 × CONH), 143.7 (2 × Ar C-8a,9a & 2 × Ar C-4a,4b), 128.2 (2 × Ar CH-1,8), 118.8 (2 × Ar C-3,6), 115.6 (2 × Ar CH-2,7), 107. (2 × Ar CH-4,5), 92.1 (d, J 187.8 Hz, 2 × CHF), 58.0 (4 × CH2N), 56.9 (d, J 20.6 Hz, 2 × CH2CHF), 21.2 (4 × CH3CH2CH2), 12.1 (4 × CH2CH3); 19F NMR (282 MHz, CD3OD) δF 191.2 (2F, dt, J 49.4, 27.9 Hz, 2 × CHF); HRMS m/z (ES-) calcd. for C31H42F2N5O3 [M-H]- requires 570.3251, found 570.3242; m/z (ES-) 570 ([M-H]-, 100%). 7.3.33 - 3,6-Bis((2S)-3-N,N-dipropylamino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-210 Following the procedure set out for 3,6-bis((2R)-3-N,N-dipropylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-210 starting from methyl 2(S)-fluoropropanoate (S)-209 (370 mg, 1.8 mmol), the reaction yielded 3,6-bis((2S)-3-N,N-dipropylamino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-210 (124 mg, 0.22 mmol, 26%) as a pale yellow solid: Rf 0.17 (ethyl acetate:hexane, 80:20); 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON F F Experimental 195 mp 215 ºC (dec.); [!]!"#!" -11.7 (c 1.02, CH3OH), [!]!"#!" -5.4 (c 1.02, CH3OH), [!]!"#!" -4.1 (c 1.02, CH3OH); IR νmax (KBr, cm-1) 3283, 2959, 1687 (C=O), 1637 (C=O), 1600, 1537, 1464; 1H NMR (300 MHz, CD3OD) δH 8.25 (2H, d, J 8.9 Hz, Ar H-1,8), 8.21 (2H, d, J 1.9 Hz, Ar H-4,5), 7.29 (2H, dd, J 8.9, 1.9 Hz, Ar H-2,7), 5.16 (2H, dt, J 49.4, 4.9 Hz, 2 × CHF), 3.07 (4H, dd, J 27.9, 4.9 Hz, 2 × CH2CHF), 2.62-2.47 (8H, m, 4 × CH2N), 1.54-1.48 (8H, m, 4 × CH3CH2CH2), 0.89 (12H, t, J 7.4 Hz, 4 × CH3CH2); 13C NMR (75.5 MHz, CD3OD) δC 178.5 (Ar C-9), 170.3 (d, J 20.6 Hz, 2 × CONH), 143.7 (2 × Ar C-8a,9a & 2 × Ar C-4a,4b), 128.2 (2 × Ar CH-1,8), 118.6 (2 × Ar C-3,6), 115.9 (2 × Ar CH-2,7), 107.7 (2 × Ar CH-4,5), 92.1 (d, J 187.8-Hz, 2 × CHF), 58.0 (4 × CH2N), 57.0 (d, J 20.6 Hz, 2 × CH2CHF), 21.2 (4 × CH3CH2CH2), 12.1 (4 × CH3CH2); 19F NMR (282 MHz, CD3OD) δF -191.2 (2F, dt, J 49.4, 27.9 Hz, 2 × CHF); HRMS m/z (ES-) calcd. for C31H42F2N5O3 [M-H]- requires 570.3251, found 570.3263; m/z (ES-) 570 ([M-H]-, 100%). Experimental 196 7.3.34 - 3,6-Bis((2R)-3-N,N-bispropylamino-2-fluoropropionamido)-9-(4-dimethylamino phenylamino)acridine (R,R)-212 Phosphorous oxychloride (5 mL) was added to 3,6-bis((2R)-3-dipropylamino-2-fluoropropionamido)-9-(10H)-acridone (R,R)-210 (115 mg, 0.20 mmol, 1.0 eq). The resulting bright orange suspension was stirred at rt until TLC analysis indicated consumption of the starting material. The solution was cooled to 0 °C and cold Et2O (10 mL) was added, resulting in formation of a precipitate. The precipitate was isolated by filtration and washed with further Et2O (2 × 10 mL) and dissolved in CHCl3 (10 mL). The organic phase was washed with aqueous NH4OH (1 M, 10 mL) and brine (10 mL), dried over Na2SO4, filtered and the solvent removed in vacuo to yield 3,6-bis((2R)-3-N,N-dipropylamino-2-fluoropropionamido)-9-chloroacridine (R,R)-211 as a red brown solid, which was used directly in the next step without further purification. N,N-dimethylaminoaniline monohydrochloride (700 mg, 4.05 mmol, 20 eq) was dissolved in saturated aqueous Na2CO3 and extracted with Et2O (3 × 20 mL). The organic fractions were combined, washed with brine, dried over Na2SO4, filtered and the solvent removed in vacuo to yield N,N-dimethylaminoaniline as a light brown oil. This oil was dissolved in CHCl3 (20 mL) and added via cannula to 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' N FF Experimental 197 a refluxing solution of 3,6-bis((2R)-3-N,N-dipropylamino-2-fluoropropionamido)-9-chloroacridine in CHCl3 (10 mL) over 30 min. This mixture was heated under reflux until TLC analysis indicated the consumption of the chloride, at which point the solvent was removed in vacuo to yield a purple oil. Cold Et2O (excess) was added, resulting in the precipitation of a red-brown solid. The solid was isolated by filtration and washed with further Et2O (excess). The residue was dissolved in CHCl3 (20 mL) and washed with aqueous NH4OH (1 M, 10 mL) followed by brine (10 mL). The organic phase was dried over Na2SO4, filtered and the solvent removed in vacuo to yield a red solid, which was purified by silica gel column chromatography, eluting with CHCl3 and CH3OH (95:5), to yield 3,6-bis((2R)-3-bispropylamino-2-fluoropropanamide)-9-(4-dimethylamino phenylamino) acridine (R,R)-212 (34.6 mg, 50.6 µmol, 25% over two steps) as a dark red solid: Rf 0.06 (CHCl3:CH3OH, 95:5); mp 190 ºC (dec.); IR νmax (KBr disc, cm-1) 3421, 2960, 1701 (C=O), 1611 (C=O), 1518, 1277; 1H NMR (500 MHz, CD3OD) δH 8.46 (2H, d, J 1.5 Hz, Ar CH-4,5), 8.02 (2H, d, J 9.4 Hz, Ar CH-1,8), 7.33 (2H, dd, J 9.4, 1.5 Hz, Ar CH-2,7), 7.17 (2H, d, J 8.8 Hz, Ar CH-14,14′), 6.85 (2H, d, J 8.8 Hz, Ar CH-13,13′), 5.18 (2H, dt, J 49.5, 4.8 Hz, 2 × CHF), 3.08 (4H, dd, J 26.0, 4.8 Hz, 2 × CH2CHF), 2.97 (6H, s, 2 × NCH3), 2.60-2.49 (8H, m, 4 × CH2CH2N), 1.54-1.46 (8H, m, 4 × CH3CH2CH2), 0.88 (12H, t, J 7.3 Hz, 4 × CH3CH2); 13C NMR (126 MHz, CD3OD) δC 170.7 (d, J 20.8 Hz, CONH), 154.5 (Ar C-9), 151.3 (Ar C-15), 144.2 (2 × Ar C-4a,4b), 144.0 (2 × Ar C-8a,9a), 131.2 (Ar C-12), 127.8 (2 × Ar CH-1,8), 126.5 (2 × Ar CH-14,14′), 118.0 (2 × Ar CH-2,7), 114.5 (2 × Ar CH-13,13′), 112.0 (2 × Ar C-3,6), 108.7 (2 × Ar CH-4,5), 92.3 (d, J 188.4 Hz, 2 × CHF), 58.0 (4 × CH2CH2N), 56.9 (d, J 19.9 Hz, 2 × CH2CHF), 40.8 (2 × NCH3), 21.3 (4 × CH3CH2CH2), 12.1 (4 × CH3CH2); 19F NMR (470 MHz, CD3OD) δF -191.2 (2F, dt, J 49.5, 26.0 Hz, Experimental 198 2 × CHF); HRMS m/z (ES+) calcd. for C39H54F2N7O2 [M+H]+ requires 690.4302, found 690.4309; m/z (ES+) 690 ([M+H]+, 100%). 7.3.35 - 3,6-Bis((2S)-3-bispropylamino-2-fluoropropanamide)-9-(4-dimethylamino phenylamino)acridine (S,S)-212 Following the procedure set out for 3,6-bis((2R)-3-bispropylamino-2-fluoropropanamide)-9-(4-dimethylaminophenylamino)acridine (R,R)-212 , starting from 3,6-bis((2S)-3-N,N-dipropylamino-2-fluoropropionamido)-9-(10H)-acridone (S,S)-210 (118 mg, 0.209 mmol), the reaction yielded 3,6-bis((2S)-3-bispropylamino-2-fluoropropanamide)-9-(4-dimethylaminophenylamino) acridine (S,S)-212 (29.7 mg, 44 µmol, 21% over two steps) as a dark red solid: Rf 0.06 (CHCl3:CH3OH, 95:5); mp 190 ºC (dec.); IR νmax (KBr disc, cm-1) 3420, 2959, 1701 (C=O), 1610 (C=O), 1520, 1276; 1H NMR (400 MHz, CD3OD) δH 8.33 (2H, d, J 2.1 Hz, Ar H-4,5), 7.92 (2H, d, J 9.4 Hz, Ar H-1,8), 7.27 (2H, dd, J 9.4, 2.1 Hz, Ar H-2,7), 7.11 (2H, d, J 9.0 Hz, Ar H-14,14′), 6.77 (2H, d, J 9.0 Hz, Ar H-13,13′), 5.16 (2H, dt, J 49.4, 4.9 Hz, 2 × CHF), 3.09-3.02 (4H, m, 2 × CH2CHF), 2.97 (6H, s, 2 × NCH3), 2.58-2.46 (8H, m, 4 × CH2CH2N), 1.54-1.44 (8H, m, 4 × CH3CH2CH2), 0.87 (12H, t, J 7.4 Hz, 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' NF F Experimental 199 4 × CH3CH2); 13C NMR (101 MHz, CD3OD) δC 170.5 (d, J 20.1 Hz, 2 × CONH), 154.1 (Ar C-9), 151.1 (Ar C-15), 144.0 (2 × Ar C-4a,4b), 143.9 (2 × Ar C-8a,9a), 131.1 (Ar C-12), 127.7 (2 × Ar CH-1,8), 126.4 (2 × Ar CH-14,14′), 117.9 (2 × Ar CH-2,7), 114.3 (2 × Ar CH-13,13′), 111.8 (2 × Ar C-3,6), 108.5 (2 × Ar CH-4,5), 92.1 (d, J 188.4 Hz, 2 × CHF), 57.9 (4 × CH2CH2N), 56.9 (d, J 19.9 Hz, 2 × CH2CHF), 40.8 (2 × NCH3), 21.2 (4 × CH3CH2CH2), 12.1 (4 × CH3CH2); 19F NMR (376 MHz, CD3OD) δF -191.1 (2F, dt, J 49.4, 26.2 Hz, 2 × CHF); HRMS m/z (ES+) calcd. for C39H54F2N7O2 [M+H]+ requires 690.4302, found 690.4294; m/z (ES+) 690 ([M+H]+, 100%). 7.3.36 - 3,6-Bis(3-chloropropionamido)-9-(10H)-acridone[191] 174 3,6-Diamino-9-(10H)-acridone 155 (500 mg, 2.2 mmol, 1.0 eq) was heated under reflux in neat 3-chloropropionyl chloride (5.0 mL) for 5 h. The solution was cooled to rt and ice-cold Et2O (30 mL) was added, resulting in formation of a precipitate. The precipitate was isolated by filtration, washed with Et2O (2 × 30 mL) and dried under vacuum to yield 3,6-bis(3-chloropropionamido)-9(10H)-acridone 174 (241 mg, 0.59 mmol, 27%) as an orange amorphous solid which was used without further purification: mp 295-297 ºC [Lit.[191] 300 ºC]; 1H NMR (300 MHz; d6-DMSO) δH 11.48 (2H, s, 2 × CONH), 9.58 (1H, s, NH-10), 8.88 (2H, d, J 1.7 Hz, Ar H-4/5), 8.35 (2H, d, 2 3 4 4a9a1 NH10 4b8a9 5 678NH NH O ClOCl O Experimental 200 J 9.1 Hz, Ar H-1/8), 7.87 (2H, dd, J 9.1, 1.7 Hz, Ar H-2/7), 3.96 (4H, t, J 6.2 Hz, 2 × CH2CH2Cl), 3.06 (4H, t, J 6.2 Hz, 2 × CH2CH2Cl); m/z (ES-) 404 ([M[35Cl]-H], 100%). 7.3.37 - 3,6-Bis(3-N,N-bisallylaminopropionamido)-9-(10H)-acridone 140i Diallylamine (320 µL, 2.6 mmol, 10 eq) was added dropwise to a refluxing solution of 3,6-bis(3-chloropropionamido)-9-(10H)-acridone 174 (100 mg, 0.25 mmol, 1.0 eq) and NaI (96 mg, 0.64 mmol, 2.5 eq) in EtOH (10 mL) and the resulting mixture was stirred for 3 h at reflux. The solution was cooled to rt and the solvent removed in vacuo and Et2O (20 mL) was added, resulting in the formation of a precipitate. The precipitate was isolated by filtration, and washed with further Et2O (2 × 10 mL) and dissolved in ethyl acetate (20 mL). The organic phase was washed with NH4OH (1 M, 2 × 10 mL) and brine (10 mL), dried over Na2SO4, filtered and the solvent removed in vacuo. The material was purified by silica gel chromatography, eluting with CHCl3, CH3OH and triethylamine (90:9:1), to yield 3,6-bis(3-N,N-bisallylaminopropionamido)-9-(10H)-acridone 140i (62 mg, 0.12 mmol, 46%) as a yellow solid: mp 300 ºC dec.; IR νmax (NaCl plate, cm-1) 3417, 2928, 1600 (C=O), 1554, 1493, 1458; 1H NMR (500 MHz, CDCl3) δH 11.17 (1H, br s, ArH-10), 11.05 (2H, br s, 2 × CONH), 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON Experimental 201 8.20 (2H, d, J 8.8 Hz, ArH-1/8), 8.06 (2H, d, J 1.6 Hz, ArH-4/5), 6.82 (2H, dd, J 8.8, 1.6 Hz, ArH-2/7), 5.81-5.76 (4H, m, 4 × =CH), 5.19-5.15 (8H, m, 4 × CH2=), 3.14 (8H, s, 4 × NCH2-allyl), 2.77 (4H, t, J 6.3 Hz, 2 × CH2CH2), 2.50 (4H, t, J 6.3 Hz, 2 × CH2N); 13C NMR (126 MHz, CD3OD) δC 178.5 (ArC-9), 173.5 (2 × CONH), 144.6 (ArC-8a/9a), 143.9 (ArC-4a/4b), 135.6 (4 × =CH), 128.2 (ArCH-1/8), 119.2 (4 × CH2=), 118.1 (ArC-3/6), 115.3 (ArCH-2/7), 106.6 (ArCH-4/5), 57.5 (4 × NCH2-allyl), 50.0 (2 × CH2CH2), 35.1 (2 × CH2CH2N); HRMS m/z (ES+ calcd. for C31H38N5O3 [M+H]+ requires 528.2975, found 528.2963; m/z (ES+) 528 ([M+H]+, 100%), 550 ([M+Na]+, 20%). 7.3.38 - 3,6-Bis(3-N,N-bispropylaminopropionamido)-9-(10H)-acridone 140j Following the proceedure set out for 3,6-bis(3-N,N-bisallylamino-propionamido)-9-(10H)-acridone 140i, starting from diproplyamine (350 µL, 2.5 mmol, 10 eq) with 3,6-bis(3-chloropropionamido)-9-(10H)-acridone 174 (103 mg, 0.25 mmol, 1.0 eq) and NaI (95 mg, 0.63 mmol, 2.5 eq), the reacton yielded 3,6-bis(3-N,N-bispropylaminopropionamido)-9-(10H)-acridone 140j (35.8 mg, 0.067 mmol, 26%) as a yellow solid: mp 320 ºC (dec.); IR νmax (NaCl plate, cm-1) 3420, 2930, 1602 (C=O), 1545, 1459; 1H NMR (500 MHz, CDCl3) δH 11.72 (1H, s, ArH-10), 10.83 (2H, s, 2 × CONH), 8.49 (2H, d, J 1.6 Hz, ArH-4/5), 8.35 (2H, d, J 8.7 Hz, ArH-1/8), 6.73 (2H, 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON Experimental 202 dd, J 8.7, 1.6 Hz, ArH-2/7), 2.81 (4H, t, J 5.6 Hz, 2 × CH2N), 2.68 (4H, t, J 5.6 Hz, 2 × CH2CH2), 2.56-2.53 (8H, m, 4 × NCH2CH2), 1.62-1.55 (8H, m, 4 × CH2CH3), 0.92 (12H, t, J 7.3 Hz, 4 × CH2CH3); 13C NMR (126 MHz, CDCl3) δC 176.9 (ArC-9), 172.3 (2 × CONH), 142.7 (ArC-8a/9a), 142.5 (ArC-4a/4b), 127.9 (ArCH-1/8), 117.7 (ArC-3/6), 113.6 (ArCH-2/7), 106.2 (ArCH-4/5), 55.3 (4 × NCH2), 50.3 (2 × CH2CH2), 33.5 (2 × CH2CH2N) , 19.9 (4 × CH2CH3), 12.1 (4 × CH2CH3); HRMS m/z (ES+) calcd. for C31H45N5O2 [M+H]+ requires 536.3601, found 536.3594; m/z (ES+) 536 ([M+H]+, 100%). 7.3.39 - 3,6-Bis(3-N,N-diallylamino-propanamide)-9-(4-dimethylaminophenylamino) acridine 219 Phosphorous oxychloride (5.0 mL) was added to 3,6-bis(3-N,N-diallylamino-propionamido)-9-(10H)-acridone 140i (30.0 mg, 0.20 mmol, 1.0 eq). The resulting suspension was stirred at rt until TLC analysis indicated consumption of the starting material. The solution was cooled to 0 °C and cold Et2O (10 mL) was added, resulting in formation of a precipitate. The precipitate was isolated by filtration and washed with further Et2O (2 × 10 mL) and dissolved in CHCl3 (10 mL). The organic phase was 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' N Experimental 203 washed with aqueous NH4OH (1 M, 10 mL) and brine (10 mL), dried over Na2SO4, filtered and the solvent removed in vacuo to yield 3,6-bis(3-N,N-diallylamino-propionamido)-9-chloroacridine as a red brown solid, which was used directly in the next step without further purification. N,N-dimethylaminoaniline monohydrochloride (200 mg, 1.16 mmol, 20 eq) was dissolved in saturated aqueous Na2CO3 (10 mL) and extracted with Et2O (3 × 5 mL). The organic fractions were combined, washed with brine, dried over Na2SO4, filtered and the solvent removed in vacuo to yield N,N-dimethylaminoaniline as a light brown oil. This oil was dissolved in CHCl3 (5 mL) and added via cannula to a refluxing solution of 3,6-bis(3-N,N-diallylamino-propionamido)-9-chloroacridine in CHCl3 (5 mL) over 30 min. This mixture was heated under reflux until TLC analysis indicated the consumption of the chloride, at which point the solvent was removed in vacuo to yield a purple oil. Cold Et2O (20 mL) was added, resulting in the precipitation of a red-brown solid. The solid was isolated by filtration and washed with further Et2O (excess). The residue was dissolved in CHCl3 (10 mL) and washed with aqueous NH4OH (1 M, 10 mL) followed by brine (10 mL). The organic phase was dried over Na2SO4, filtered and the solvent reduced in vacuo to yield a dark red solution. The material was prepared as its HCl salt by the addition of HCl (1 M, diethyl ether), which resulted in precipitation. This salt solution was loaded onto silica gel and purified by column chromatography, eluting with CHCl3, CH3OH and triethylamine (100:0:0, 95:5:0, 90:5:5). The desired fractions were collected and washed with NH4OH (1.0 M, 2 × 10 mL), brine (10 mL), dried over Na2SO4, filtered and the solvent was removed in vacuo to yield 3,6-bis(3-N,N-diallylamino-propanamide)-9-(4-dimethylamino phenylamino) acridine 219 (13.5 mg, 20.9 µmol, 37%) as a dark red solid: Rf 0.09 (CHCl3:CH3OH:Et3N, 90:8:2); mp >320 ºC; IR νmax (NaCl plate, cm-1) 3419, Experimental 204 2930, 1603, 1548, 1488, 1398; 1H NMR (500 MHz, CD3OD) δH 8.37 (2H, s, ArH-4/5), 8.00 (2H, d, J 8.5, ArH-1/8), 7.28 (2H, d, J 8.5, ArH-2/7), 7.22 (2H, d, J 8.7, ArH-14/14′), 6.85 (2H, d, J 8.7, ArH-13/13′), 6.03-6.11 (4H, m, 4 × =CH), 5.59-5.69 (8H, m, 4 × CH2=), 3.85 (8H, s, 4 × NCH2-allyl), 3.50 (4H, t, J 6.7 Hz, 2 × CH2CH2), 3.07 (4H, t, J 6.7 Hz, 2 × CH2CH2N), 3.03 (6H, s, 2 × NCH3); 13C NMR (75 MHz, CDCl3) δC 171.9 (2 × CONH), 153.9 (Ar C-9), 150.3 (Ar C-15), 145.7 (2 × Ar C-4a/4b), 143.4 (2 × Ar C-8a/9a), 134.0 (4 × =CH), 133.2 (Ar C-12), 127.3 (2 × Ar CH-1/8), 123.7 (2 × Ar CH-14/14′), 119.4 (4 × CH2=), 116.4 (2 × Ar CH-2/7), 115.9 (2 × Ar CH-13/13′), 113.6 (2 × Ar C-3/6), 107.5 (2 × Ar CH-4/5), 48.8 (2 × CH2CH2), 46.2 (4 × NCH2-allyl), 40.9 (2 × NCH3), 33.7 (2 × CH2CH2N); HRMS m/z (ES+) calcd. for C39H47N7O2 [M+H]+ requires 646.3869, found 646.3870; m/z (ES+) 646 ([M+H]+, 100%). 7.3.40 - 3,6-Bis(3-N,N-dipropylamino-propionamido)-9-(4-dimethylaminophenylamino) acridine 214 Following the proceedure set out for 3,6-bis(3-N,N-diallylamino-propanamide)-9-(4-dimethylaminophenylamino) acridine 219, starting from 3,6-bis(3-N,N-bispropylamino- 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' N Experimental 205 propionamido)-9-(10H)-acridone 140j (25.0 mg, 46.9 µmol, 1.0 eq) and N,N-dimethylaminoaniline monohydrochloride (160 mg, 0.93 mmol, 20 eq), the reacton yielded 3,6-bis(3-N,N-dipropylamino-propanamide)-9-(4-dimethylaminophenylamino) acridine 214 (8.9 mg, 13.6 µmol, 29%) as a red solid: IR νmax (NaCl plate, cm-1) 3401, 2928, 1601, 1491, 1454, 1405; 1H NMR (500 MHz, CD3OD) δH 8.47 (2H, d, J 1.8 Hz, ArH-4/5), 8.05 (2H, d, J 9.3 Hz, ArH-1/8), 7.23 (2H, dd, J 9.3, 1.8 Hz, ArH-2/7), 7.22 (2H, d, J 9.0 Hz, ArH-14/14′), 6.87 (2H, d, J 9.0 Hz, ArH-13/13′), 3.07 (4H, t, J 6.8 Hz, CH2CH2N), 3.03 (6H, s, 2 × NCH3), 2.73 (4H, t, J 6.8 Hz, 2 × CH2CH2N), 2.68-2.65 (8H, m, 4 × NCH2CH2), 1.66-1.58 (8H, m, 4 × CH2CH3), 0.95 (12H, t, J 7.4 Hz, 4 × CH2CH3); 13C NMR (126 MHz, CDCl3) δC 173.6 (2 × CONH), 154.9 (Ar C-9), 151.5 (Ar C-15), 145.4 (2 × Ar C-4a/4b), 143.8 (2 × Ar C-8a/9a), 130.5 (Ar C-12), 127.9 (2 × Ar CH-1/8), 126.9 (2 × Ar CH-14/14′), 117.6 (2 × Ar CH-2/7), 114.3 (2 × Ar CH-13/13′), 111.0 (2 × Ar C-3/6), 106.8 (2 × Ar CH-4/5), 56.7 (4 × NCH2CH2), 50.6 (2 × CH2CH2), 40.7 (2 × NCH3), 34.5 (2 × CH2CH2N), 20.5 (4 × CH2CH3), 12.1 (4 × CH2CH3); HRMS m/z (ES+) calcd. for C39H56N7O2 [M+H]+ requires 654.4495, found 654.4513; m/z (ES+) 654 ([M+H]+, 100%). Experimental 206 7.3.41 - 3,6-Bis(3-amino-propionamido)-9-(4-dimethylaminophenylamino) acridine diformate 213 1,4-Di(phenylphosphino)butane (4.2 mg, 9.8 µmol, 20 mol%/allyl group) was added to a solution of tris(dibenzylideneacetone)dipalladium (4.5 mg, 4.9 µmol, 10 mol%/allyl group) in THF (1.0 mL) and stirred for 15 min until the solution turned yellow. The solution was added to a solution of 3,6-bis(3-N,N-diallylamino-propionamido)-9-(4-dimethylaminophenylamino) acridine 219 (8.1 mg, 13 µmol, 1.0 eq) and 2-mercaptosalicylic acid (9.8 mg, 64 µmol, 5.0 eq) in THF (1.0 mL). The resulting solution was heated under reflux for 3 h before being cooled to rt and diluted with distilled water (2.0 mL) and HCl (0.1 M, 250 µL), which caused precipitation of a solid. The precipitate was isolated by filtration and the residue was washed repeatedly with distilled water (3 × 1.0 mL). The water/THF filtrate was concentrated in vacuo to yield a yellow solid, which was purified by reverse phase silica gel chromatography eluting with water and CH3OH (90:10 with 1% formic acid) to yield 3,6-bis(3-amino-propionamido)-9-(4-dimethylaminophenylamino) acridine 213 as its diformic acid salt (1.2 mg, 2.1 µmol, 17%) as a red powder: Rf 0.12 (H2O:CH3OH:HCO2H, 90:9:1); 1H NMR (500 MHz, D2O) δH 8.14 (2H, s, Ar H-4,5), 7.91 (2H, d, J 9.2 Hz, Ar H-1,8), 7.23 (2H, d, J 8.9 Hz, Ar H-14/14'), 2 3 4 4a 9a 1 9 8a 4b 5 6 7 8 N H N H N 10 HN 12 13' 14' 15 14 13 N O NH 3 + O + H 3 N HCOO - HCOO - Experimental 207 7.15 (2H, d, J 9.2 Hz, Ar H-2,7), 7.02 (2H, d, J 8.9 Hz, Ar H-13,13'), 3.36 (4H, t, J 6.5 Hz, 2 × CH2), 2.94 (4H, t, J 6.5 Hz, 2 × CH2), 2.92 (6H, s, 2 × NCH3); HRMS m/z (ES+) calcd. for C27H32N7O2 [M+H]+ requires 486.2617, found 486.2621; m/z (ES+) 486 ([M+H]+, 100%). Experimental 208 7.4 - Experimental for Chapter 5 7.4.1 - General proceedures General procedure (GP) 1 - The appropriate quantities of (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoic acid (S)-230 (1.0 eq), diisopropylethylamine (4.0 eq) and amino ester (2.0 eq) in CH2Cl2 (1.0 mL mmol-1) were cooled to 0 ºC and propylphosphonic anhydride (T3P) (50% w/w in ethyl acetate, 2.0 eq) was added dropwise. The solution was maintained at 0 ºC for a further 30 min before being warmed to rt and stirred until TLC analysis indicated consumption of the starting material. The reaction mixture was quenched by the addition of HCl (1 M, 10 mL) and the aqueous phase was extracted with ethyl acetate (2 × 10 mL). The combined organic phases were washed sequentially with HCl (1 M, 3 × 10 mL), saturated aqueous Na2CO3 (3 × 10 mL), brine (20 mL) and dried over Na2SO4. The solvent was removed in vacuo and the product purified by silica gel chromatography, eluting with mixtures of ethyl acetate and hexane. General Procedure (GP) 2 - Tetrabutylammonium fluoride (1 M in THF, 4.0 eq) was added dropwise to the appropriate silyl protected dipeptide 232a-c (1.0 eq) and acetic acid (5.0 eq) in THF (8.0 mL mmol-1) and the resulting reaction mixture stirred at rt. The reaction was quenched by the addition of water (5 mL) followed by ethyl acetate (10 mL). The organic phases were washed successively with water (2 × 5 mL) and brine (10 mL), dried over Na2SO4 and the solvent removed in vacuo. The product was purified by silica gel column chromatography, eluting with mixtures of ethyl acetate and hexane. Experimental 209 General procedure (GP) 3 - Diethylaminosulfur trifluoride 32 (1.5 eq) was added dropwise to a solution of the appropriate amino alcohol dipeptide 224a-c (1.0 eq) in THF (5.0 mL mmol-1) at 0 ºC. The resulting solution was stirred at 0 ºC for 1 h before being quenched by the addition of NaHCO3 (solid) and water until the solution was basic (pH >9) and effervescence subsided. The aqueous phase was extracted with diethyl ether (3 × 10 mL) and the combined organic phases were washed with brine, dried over Na2SO4, filtered and the solvent removed in vacuo. The product mixtures were purified by silica gel column chromatography, eluting with mixtures of ethyl acetate and hexane, separating the α- and β-fluorinated regioisomers where applicable. Experimental 210 7.4.2 - Methyl (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoate (S)-229 Triethylamine (16.0 mL, 115 mmol, 4.5 eq) was added dropwise over 30 min to a solution of methyl (2S)-2-diallylamino-3-hydroxy-propanoate (S)-187 (5.00 g, 25.1 mmol, 1.0 eq) and tert-butyldimethylsilyl trifluoromethanesulfonate (9.00 mL, 39.1 mmol, 1.8 eq) in CH2Cl2 (230 mL) at 0 ºC. The mixture was brought to rt and stirred for 16 h, quenched by the addition of CH3OH (40 mL) followed by saturated aqueous Na2CO3 (100 mL). The organic phase was separated and the aqueous phase was extracted with CH2Cl2 (3 × 100 mL). The organic phases were combined, dried over Na2SO4 and the solvent removed in vacuo. The oil was purified by silica gel chromatography, eluting with hexane and ethyl acetate (95:5), to yield methyl (–)-(2S)- 3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoate (S)-229 (6.52 g, 20.8 mmol, 83%) as a colourless oil: Rf 0.5 (hexane:ethyl acetate, 90:10); [!]!!" -18.1 (c 0.6, CHCl3); IR νmax (neat, cm-1) 2951, 2929, 1735 (C=O), 1251 (Si–C), 1103, 918 (Si–C); 1H NMR (400 MHz, CDCl3) δH 5.78 (2H, dddd, J 17.2, 10.1, 7.0, 5.4 Hz, 2 × =CH), 5.21-5.09 (4H, m, 2 × CH2=), 3.93 (1H, dd, J 9.9, 7.0 Hz, OCHaHb), 3.82 (1H, dd, J 9.9, 5.6 Hz, OCHaHb), 3.61 (1H, dd, J 7.0, 5.6 Hz, CHN), 3.38-3.32 (4H, m, 2 × NCH2), 3.15 (3H, OCH3), 0.86 (9H, s, (CH3)3), 0.03 (3H, s, SiCH3), 0.03 (3H, s, SiCH3); 13C NMR (101 MHz, CDCl3) δC 172.4 (CO2CH3), 136.7 (2 × =CH), 117.2 (2 × CH2=), 64.2 (CHN), 62.9 (CH2), 54.6 (2 × NCH2-allyl), 51.2 (OCH3), 25.9 (C(CH3)3), 18.3 (SiC), -5.4 (SiCH3), -5.4 (SiCH3); HRMS m/z (ES+) calcd. OONOSi Experimental 211 for C16H32NO3Si [M+H]+ requires 314.2151, found 314.2156; m/z (ES+) 314 ([M+H]+, 100%). Enantiomeric excess determined by chiral HPLC (Chiralcel OD-H, 5% iPrOH in hexane, 0.25 mL/min, tr maj = 7.08 min). 7.4.3 - (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoic acid (S)-230 Lithium hydroxide monohydrate (3.37 g, 80.4 mmol, 4.0 eq) was added portionwise to a solution of methyl (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoate (S)-229 (6.30 g, 20.1 mmol, 1.0 eq) in THF:H2O:CH3OH (20:20:60, 150 mL) and was stirred for 24 h at rt. The reaction was quenched by neutralisation with HCl (1 M, 60 mL) and the aqueous phase extracted with CH2Cl2 (3 × 50 mL). The combined organics were dried with Na2SO4, filtered and the solvent removed in vacuo to yield (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoic acid (S)-230 (5.41 g, 18.1 mmol, 90%) as a colourless gum, which was used without any further purification: [!]!!" -3.1 (c 1.7, CH3OH); IR νmax (neat, cm-1) 2927, 2856, 1635 (C=O), 1417, 1257 (Si–C), 1087, 918 (Si–C); 1H NMR (300 MHz, CD3OD) δH 5.90 (2H, dddd, J 17.0, 10.3, 6.6, 6.6 Hz, 2 × =CH), 5.27-5.12 (4H, m, 2 × CH2=), 4.01 (1H, dd, J 10.7, 5.2 Hz, OCHaHb), 3.92 (1H, dd, J 10.7, 6.8 Hz, OCHaHb), 3.47 (1H, dd, J 6.8, 5.2 Hz, CHN), 3.40 (4H, m, NCH2-allyl), 0.91 (9H, s, C(CH3)3), 0.08 (3H, s, SiCH3), 0.08 (3H, s SiCH3); 13C NMR (75 MHz, CD3OD) δC 178.3 (CO2H), 135.0 (2 × =CH), 118.5 (2 × CH2=), 68.5 (CHN), 64.4 (OCH2), 55.5 (2 × NCH2-allyl), 26.5 (C(CH3)3), OHONOSi Experimental 212 19.2 (SiC), -5.1 (2 × SiCH3); HRMS m/z (ES+) calcd. for C15H30NO3Si [M+H]+ requires 300.1995, found 300.2004; m/z (ES+) 322 ([M+Na]+, 100%), 300 ([M+H]+, 5%). 7.4.4 - Methyl (+)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 232a Following GP1: Starting with (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoic acid (S)-230 (1.00 g, 3.34 mmol), L-phenylalanine methyl ester hydrochloride 231a (1.44 g, 6.68 mmol), diisopropylethylamine (2.30 mL, 13.2 mmol) and T3P (50% w/w in ethyl acetate, 2.33 mL), the reaction yielded methyl (+)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate  232a (1.25 g, 2.71 mmol, 81%) as a colourless oil: Rf 0.35 (hexane:ethyl acetate, 90:10); [!]!!" +15.0 (c 0.6, CHCl3); IR νmax (neat, cm-1) 3361 (NH), 2953, 1747 (C=O), 1670 (C=O), 1496 (NH), 1093, 920 (Si–C); 1H NMR (400 MHz, CDCl3) δH 7.90 (1H, d, J 7.9 Hz, CONH), 7.28-7.08 (5H, m, 5 × Ar-H), 5.61 (2H, dddd, J 17.0, 10.4, 6.4, 6.4 Hz, 2 × =CH), 5.14-5.04 (4H, m, 2 × CH2=), 4.80 (1H, dt, J 7.9, 6.2 Hz, Cα+1H), 4.14 (1H, dd, J 11.1, 4.0 Hz, OCHaHb), 3.90 (1H, dd, J 11.1, 8.3 Hz, OCHaHb), 3.71 (3H, s, OCH3), 3.53 (1H, dd, J 8.3, 4.0 Hz, CαH), 3.21 (4H, m, 2 × NCH2-allyl), 3.17-3.02 (2H, m, CH2Ph), 0.86 (9H, s, C(CH3)3), 0.03 (3H, s, SiCH3), 0.03 (3H, s, SiCH3); 13C NMR (101 MHz, CDCl3) δC 172.1 (CONH), 171.9 (CO2CH3), α NHO α+1PhOONOSi Experimental 213 136.4 (Ar-C), 136.2 (2 × =CH), 129.3 (2 × Ar-CH), 128.6 (2 × Ar-CH), 127.1 (Ar-CH), 117.3 (2 × CH2=), 63.9 (CαH), 61.3 (OCH2), 53.9 (2 × NCH2-allyl), 53.0 (Cα+1H), 52.3 (OCH3), 38.1 (CH2Ph), 26.0 (C(CH3)3), 18.2 (SiC), -5.5 (SiCH3), -5.5 (SiCH3); HRMS m/z (ES+) calcd. for C25H40N2O4SiNa [M+Na]+ requires 483.2655, found 483.2643; m/z (ES+) 483 ([M+Na]+, 30%), 461 ([M+H]+, 100%). 7.4.5 - Methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-alaninate 232b Following GP1: Starting with (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoic acid (S)-230 (198 mg, 0.661 mmol), L-alanine methyl ester hydrochloride 231b (185 mg, 1.32 mmol), diisopropylethylamine (460 µL, 2.64 mmol) and T3P (460 µL, 50% w/w in ethyl acetate), methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-alaninate 232b (215 mg, 0.559 mmol, 85%) was isolated as a colourless oil: Rf 0.2 (hexane:ethyl acetate 90:10); [!]!!" -7.5 (c 0.4, CHCl3); 1H NMR (400 MHz, CDCl3) δH 7.94 (1H, d, J 7.6 Hz, CONH), 5.83 (2H, dddd, J 17.1, 10.3, 6.8, 5.6 Hz, 2 × =CH), 5.24-5.13 (4H, m, 2 × CH2=), 4.54 (1H, dq, J 7.6, 7.2 Hz, Cα+1H), 4.17 (1H, dd, J 11.1, 4.1 Hz, OCHaHb), 3.97 (1H, dd, J 11.1, 7.6 Hz, OCHaHb), 3.72 (3H, s, OCH3), 3.53 (1H, dd, J 7.6, 4.1 Hz, CαH), 3.39-3.29 (4H, m, 2 × NCH2), 1.37 (3H, d, J 7.2 Hz, CHCH3), 0.89 (9H, s, C(CH3)3), 0.06 (6H, s, 2 × SiCH3); 13C NMR (101 MHz, CDCl3) δC 173.5 (CONH), 171.8 (CO2CH3), 136.2 (2 × =CH), α NHO α+1OONOSi Experimental 214 117.5 (2 × CH2=), 64.1 (CαH), 61.3 (OCH2), 54.0 (2 × NCH2-allyl), 52.5 (OCH3), 47.7 (Cα+1H), 25.9 (C(CH3)3), 18.7 (CHCH3), 18.2 (SiC), -5.4 (SiCH3), -5.5 (SiCH3); HRMS m/z (ES+) calcd. for C19H36N2O4NaSi [M+Na]+ requires 407.2342, found 407.2339; m/z (ES+) 407 ([M+Na]+, 50%), 385 ([M+H]+, 100%). 7.4.6 - Methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-valinate 232c Following GP1: starting with (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoic acid (S)-230 (205 mg, 0.685 mmol), L-valine methyl ester hydrochloride 231c (330 mg, 1.37 mmol), diisopropylethylamine (480 µL, 2.76 mmol) and T3P (480 µL, 50% w/w in ethyl acetate), the reaction yielded methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-valinate 232c (224 mg, 0.590 mmol, 79%) as a colourless oil: Rf 0.3 (hexane:ethyl acetate, 90:10); [!]!!"!-37.0 (c 0.8, CHCl3); IR νmax (neat, cm-1) 3365 (NH), 2954, 1745 (C=O), 1680 (C=O), 1496 (NH), 920 (Si–C); 1H NMR (400 MHz, CDCl3) δH 8.01 (1H, d, J 9.3 Hz, CONH), 5.83 (2H, dddd, J 17.2, 10.2, 7.0, 4.8 Hz, 2 × =CH), 5.26-5.14 (4H, m, 2 × CH2=), 4.49 (1H, dd, J 9.3, 4.7 Hz, Cα+1H), 4.20 (1H, dd, J 11.1, 4.0 Hz, OCHaHb), 4.00 (1H, dd, J 11.1, 8.0 Hz, OCHaHb), 3.75 (3H, s, OCH3), 3.59 (1H, dd, J 8.0, 4.0 Hz, CαH), 3.43-3.30 (4H, m, NCH2-allyl), 2.21-2.13 (1H, m, CH(CH3)2), 0.92-0.86 (15H, m, C(CH3)2 and C(CH3)3), 0.06 (6H, s, 2 × SiCH3); 13C NMR (101 MHz, CDCl3) δC 172.4 (CONH), 172.0 (CO2CH3), 136.2 (2 × =CH), 117.4 (2 × CH2=), 64.0 (CαH), 61.3 (OCH2), α NHO α+1OONOSi Experimental 215 56.9 (Cα+1H), 54.0 (2 × NCH2-allyl), 52.1 (OCH3), 31.3 (CH), 26.0 (C(CH3)3), 19.3 (CH3), 18.2 (SiC), 17.8 (CH3), -5.5 (SiCH3), -5.5 (SiCH3); HRMS m/z (ES+) calcd. for C21H40N2O4SiNa [M+Na]+ requires 435.2655, found 435.2654; m/z (ES+) 435 ([M+Na]+, 5%), 413 ([M+H]+, 100%). 7.4.7 - Cyclo-(N,N-bisallyl-(S)-seryl-(S)-phenylalanine) 234 Tetrabutylammonium fluoride (870 µL, 1 M in THF, 4.0 eq) was added dropwise to methyl (+)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 232a (100 mg, 0.217 mmol, 1.0 eq) in dry THF (1.5 mL) and was stirred at rt for 2 h. The reaction was quenched by the addition of water (5 mL) followed by ethyl acetate (10 mL). The organic phases were washed successively with water (2 × 5 mL) and brine (10 mL), dried over Na2SO4 and the solvent removed in vacuo. The product was purified by silica gel column chromatography, eluting with ethyl acetate and hexane (20:80), to yield cyclo-(N,N-bisallyl-(S)-seryl-(S)-phenylalanine) 234 (18.4 mg, 0.029 mmol, 27%) as a colourless solid: mp 170-172 ºC (ethyl acetate); [!]!!" -83.9 (c 0.3, CH3OH); IR νmax (NaCl plate, cm-1) 3359 (NH), 3303, 2928, 1716 (C=O), 1663 (C=O), 1551 (NH), 1261 (C–O–C); 1H NMR (400 MHz, O Oα+1HNα NHO O NOON PhPh Experimental 216 CDCl3) δH 7.31-7.14 (10H, m, 10 × Ar-H), 6.65 (2H, d, J 7.5 Hz, 2 × NH), 5.57 (4H, dddd, J 17.0, 10.4, 6.6, 5.7 Hz, 4 × =CH), 5.12-5.06 (8H, m, 4 × CH2=), 4.52 (2H, dd, J 11.1, 3.3 Hz, 2 × CHaHbCHN), 4.54-4.38 (2H, m, 2 × Cα+1H), 4.40 (2H, dd, J 11.1, 6.2 Hz, 2 × CHaHbCHN), 3.39 (2H, dd, J 6.2, 3.3 Hz, 2 × CαH), 3.30 (2H, dd, J 14.3, 4.8 Hz, 2 × CHaHbPh), 3.16 (2H, dd, J 14.3, 10.0 Hz, 2 × CHaHbPh), 3.10-3.05 (4H, m, 4 × NCHaHb-allyl), 2.99-2.93 (4H, m, 4 × NCHaHb-allyl); 13C NMR (101 MHz, CDCl3) δC 170.6 (2 × CO2CH2), 170.1 (2 × CONH), 137.5 (2 × Ar C), 135.8 (4 × =CH), 129.3 (4 × Ar CH), 128.8 (4 × Ar CH), 127.0 (2 × Ar CH), 117.9 (4 × CH2=), 61.4 (2 × CαH), 60.1 (2 × CH2CHN), 53.9 (2 × Cα+1H), 53.7 (4 × NCH2-allyl), 35.7 (2 × CH2Ph); HRMS m/z (ES+) calcd. for C36H44N4O6Na [M+Na]+ requires 651.3153, found 651.3154; m/z (ES+) 667 ([M+K]+, 40%), 651 ([M+Na]+, 100%). 7.4.8 - Methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-phenylalaninate 224a Following GP2: Starting with methyl (+)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 232a (220 mg, 0.478 mmol), acetic acid (140 µL, 2.45 mmol) and TBAF (1.88 mL, 1 M in THF), the reaction yielded methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-phenylalaninate 224a (121 mg, 0.349 mmol, 73%) as a colourless oil: Rf 0.10 (hexane:ethyl acetate, 70:30); [!]!!" +28.1 (c 1.15, CHCl3); IR νmax (neat, cm-1) 3349 (OH), 2955, 1744 (C=O), 1659 (C=O), 1513, 1254, 1032; α NHO α+1PhOONHO Experimental 217 1H NMR (500 MHz, CDCl3) δH 7.75 (1H, d, J 7.7 Hz, CONH), 7.31-7.09 (5H, m, 5 × Ar-H), 5.57 (2H, dddd, J 17.3, 10.0, 7.5, 4.5 Hz, 2 × =CH), 5.16-5.10 (4H, m, 2 × CH2=), 4.85 (1H, ddd, J 7.7, 7.3, 5.7 Hz, Cα+1H), 3.85 (1H, dd, J 11.2, 7.6 Hz, OCHaHb), 3.76 (3H, s, OCH3), 3.75 (1H, dd, J 11.2, 4.1 Hz, OCHaHb), 3.41 (1H, dd, J 7.6, 4.1 Hz, CαH), 3.25 (1H, dd, J 14.0, 5.7 Hz, CHaHbPh), 3.15-3.11 (3H, m, 2 × NCHaHb-allyl and CH2OH), 3.06 (1H, dd, J 14.0, 7.3 Hz, CHaHbPh), 2.96-2.92 (2H, m, 2 × NCHaHb-allyl); 13C NMR (126 MHz, CDCl3) δC 174.3 (CONH), 172.0 (CO2CH3), 135.9 (Ar-C), 135.5 (2 × =CH), 129.2 (2 × Ar-CH), 128.8 (2 × Ar-CH), 127.3 (Ar-CH), 118.1 (2 × CH2=), 62.9 (CαH), 58.5 (OCH2), 53.5 (2 × NCH2-allyl), 52.9 (Cα+1H), 52.6 (OCH3), 38.0 (CH2Ph); HRMS m/z (ES+) calcd. for C19H26N2O4Na [M+Na]+ requires 369.1790, found 369.1782; m/z (ES+) 369 ([M+Na]+, 100%). 7.4.9 - Methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-alaninate 224b Following GP2: Starting with methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-alaninate 232b (195 mg, 0.507 mmol), acetic acid (150 µL, 2.62 mmol) and TBAF (2.10 mL, 1 M in THF), the reaction yielded methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-alaninate 224b (121 mg, 0.448 mmol, 88%) as a colourless oil: Rf 0.1 (hexane:ethyl acetate, 80:20); [!]!!" +12.0 (c 0.7, CHCl3); α NHO α+1OONHO Experimental 218 IR νmax (neat, cm-1) 3356 (OH/NH), 3076, 2981, 1743 (C=O), 1653 (C=O), 1521 (NH), 1219, 1155; 1H NMR (400 MHz, CDCl3) δH 7.83 (1H, d, J 7.0 Hz, CONH), 5.79 (2H, dddd, J 17.3, 10.1, 7.3, 4.7 Hz, 2 × =CH), 5.27-5.17 (4H, m, 2 × CH2=), 4.57 (1H, dq, J 7.2, 7.0 Hz, Cα+1H), 3.95 (1H, dd, J 11.2, 7.7 Hz, OCHaHb), 3.84-3.81 (1H, dd, J 11.2, 4.1 Hz, OCHaHb), 3.76 (3H, s, OCH3), 3.48 (1H, dd, J 7.7, 4.1 Hz, CαH), 3.41 (1H, br s, OH), 3.33-3.28 (2H, m, 2 × NCHaHb-allyl), 3.11-3.06 (2H, m, 2 × NCHaHb-allyl), 1.42 (3H, d, J 7.2 Hz, CH3); 13C NMR (101 MHz, CDCl3) δC 174.1 (CONH), 173.2 (CO2CH3), 135.4 (2 × =CH), 118.2 (2 × CH2=), 62.8 (CαH), 58.4 (OCH2), 53.7 (2 × NCH2-allyl), 52.7 (OCH3), 47.8 (Cα+1H), 18.6 (CHCH3); HRMS m/z (ES+) calcd. for C13H22N2O4Na [M+Na]+ requires 293.1477, found 293.1471; m/z (ES+) 293 ([M+Na]+, 100%). 7.4.10 - Methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-valinate 224c Following GP2: Starting with methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-valinate 232c (201 mg, 0.487 mmol), acetic acid (140 µL, 2.45 mmol) and TBAF (1.90 mL, 1 M in THF), the reaction yielded methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-valinate 224c (108 mg, 0.363 mmol, 75%) as a colourless oil: Rf 0.13 (hexane:ethyl acetate, 80:20); [!]!!" +6.9 (c 0.5, CHCl3); IR νmax (neat, cm-1) 3361 (OH), 2960, 2821, 1741 (C=O), 1660 (C=O), 1500 (NH), 1149; 1H NMR (300 MHz, CDCl3) δH 7.87 (1H, d, J 9.0 Hz, CONH), 5.80 (2H, dddd, α NHO α+1OONHO Experimental 219 J 17.3, 10.1, 7.3, 4.4 Hz, 2 × =CH), 5.30-5.18 (4H, m, 2 × CH2=), 4.53 (1H, dd, J 9.0, 4.7 Hz, Cα+1H), 3.99-3.83 (2H, m, CH2OH), 3.75 (3H, s, OCH3), 3.51 (1H, dd, J 7.5, 4.1 Hz, CαH), 3.43 (1H, br s, OH), 3.39-3.32 (2H, m, 2 × NCHaHb-allyl), 3.13-3.06 (2H, m, 2 × NCHaHb-allyl), 2.21 (1H, qqd, J 6.9, 6.9, 4.7 Hz, CH(CH3)2), 0.94 (3H, d, J 6.9 Hz, CH3), 0.90 (3H, d, J 6.9 Hz, CH3); 13C NMR (75.0 MHz, CDCl3) δC 174.4 (CONH), 172.2 (CO2CH3), 135.4 (2 × =CH), 118.1 (2 × CH2=), 63.1 (CαH), 58.5 (Cα+1H), 56.9 (OCH3), 53.6 (2 × NCH2-allyl), 52.3 (OCH2), 31.3 (CH(CH3)2), 19.3 (CHCH3), 17.9 (CHCH3); HRMS m/z (ES+) C15H26N2O4Na [M+Na]+ requires 321.1790, found 321.1787; m/z (ES+) 321 ([M+Na]+, 100%). 7.4.11 - Methyl (+)-N-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-(S)-phenylalanate 227a & Methyl (+)-N-[(2R)-2-(diallylamino)-3-fluoropropanoyl]-(S)-phenylalanate 228a 227a Following GP3: Starting with methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-phenylalaninate 224a (181 mg, 0.522 mmol) and diethylaminosulfur trifluoride 32 (90.0 µL, 0.682 mmol), to yield methyl (+)-N-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-(S)-phenylalanate 227a (78.1 mg, 0.224 mmol, 43%) as a colourless oil: Rf 0.24 (hexane:ethyl acetate, 7:3); [!]!!" +69.8 (c 2.3, CHCl3); IR νmax (neat, cm-1) 3429 (NH), 3070, 2924, 1747 (C=O), 1676 (C=O), 1525 (NH), 1278, 1217; α NHO α+1PhOOFN Experimental 220 1H NMR (500 MHz, CDCl3) δH 7.22-7.03 (5H, m, 5 × Ar-H), 6.94 (1H, br d, J 4.7 Hz, CONH), 5.71 (2H, dddd, J 17.0, 10.3, 6.5, 6.5 Hz, 2 × =CH), 5.11-5.05 (4H, m, 2 × CH2=), 4.90 (1H, ddd, J 49.9, 7.1, 2.8 Hz, CHF), 4.83-4.79 (1H, m, Cα+1H), 3.67 (3H, s, OCH3), 3.14-3.03 (6H, m, 2 × NCH2-allyl and CH2Ph), 2.94 (1H, ddd, J 30.1, 14.9, 2.8 Hz, CHaHbCHF), 2.84 (1H, ddd, J 23.9, 14.9, 7.1 Hz, CHaHbCHF); 13C NMR (125 MHz, CDCl3) δC 171.5 (CO2CH3), 168.6 (d, J 20.0 Hz, CONH), 135.6 (Ar-C), 135.1 (2 × =CH), 129.3 (2 × Ar-CH), 128.7 (2 × Ar-CH), 127.3 (Ar-CH), 118.1 (2 × CH2=), 91.2 (d, J 187.8 Hz, CHF), 57.4 (Cα+1H), 54.5 (d, J 19.3 Hz, CH2CHF), 52.9 (2 × NCH2-allyl), 52.5 (OCH3), 38.0 (CH2Ph); 19F NMR (470 MHz, CDCl3) δF -190.7 (dddd, J 49.9, 30.1, 23.9, 4.0 Hz, CHF); HRMS m/z (ES+) calcd. for C19H25FN2O3Na [M+Na]+ requires 371.1747, found 371.1741; m/z (ES+) 371 ([M+Na]+, 100%). 228a Further elution of the reaction mixture from the above preparation furnished methyl (+)-N-[(2R)-2-(diallylamino)-3-fluoropropanoyl]-(S)-phenylalanate 228a (68.8 mg, 0.197 mmol, 38%) as a colourless oil: Rf 0.15 (hexane:ethyl acetate, 70:30); [!]!!" +21.9 (c 2.8, CHCl3); IR νmax (neat, cm-1) 3360 (NH) , 2951, 1743 (C=O), 1674 (C=O), 1496 (NH), 1201, 1006; 1H NMR (500 MHz, CDCl3) δH 7.75 (1H, d, J 7.7 Hz, CONH), 7.21-7.02 (5H, m, 5 × Ar-H), 5.53 (2H, dddd, J 17.4, 10.0, 7.5, 4.6 Hz, 2 × =CH), 5.10-5.03 (4H, m, 2 × CH2=), 4.89-4.70 (3H, m, CH2F and Cα+1H), 3.68 (3H, s, OCH3), 3.62 (1H, ddd, J 23.9, 6.7, 3.5 Hz, CH2FCαH), 3.17-2.99 (6H, m, α NHO α+1PhOONF Experimental 221 2 × NCH2-allyl and CH2Ph); 13C NMR (125 MHz, CDCl3) δC 172.0 (CO2CH3), 170.1 (d, J 10.3 Hz, CONH), 135.9 (Ar-C), 135.4 (2 × =CH), 129.2 (2 × Ar-CH), 128.7 (2 × Ar-CH), 127.2 (Ar-CH), 118.1 (2 × CH2=), 81.1 (d, J 171.1 Hz, CH2F), 62.6 (d, J 18.9 Hz, CαH), 53.9 (2 × NCH2-allyl), 53.0 (Cα+1H), 52.5 (OCH3), 37.9 (CH2Ph); 19F NMR (470 MHz, CDCl3) δF -227.1 (dt, J 47.2, 23.9 Hz, CH2F); HRMS m/z (ES+) calcd. for C19H25FN2O3Na [M+Na]+ requires 371.1747, found 371.1746; m/z (ES+) 371 ([M+Na]+, 100%). 7.4.12 - Methyl (+)-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-(S)-alaninate 227b & Methyl (–)-N-[(2R)-2-(diallylamino)-3-fluoropropanoyl]-(S)-alaninate 228b 227b Following GP3: Starting with methyl (+)-(N,N-diallyl-(S)-seryl)-(S)-alaninate 224b (98.5 mg, 0.366 mmol) and diethylaminosulfur trifluoride 32 (65.9 mg, 54 µL, 0.409 mmol), the reaction yielded methyl (+)-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-(S)-alaninate 227b (18.9 mg, 69.4 µmol, 18%) as a colourless oil: Rf 0.34 (hexane:ethyl acetate, 80:20); ![!]!!" +15.2 (c 1.8, CHCl3); 1H NMR (400 MHz, CDCl3) δH 7.03-7.02 (1H, br m, CONH), 5.83 (2H, dddd, J 17.0, 10.3, 6.6, 6.6 Hz, 2 × =CH), 5.21-5.14 (4H, m, 2 × CH2=), 5.00 (1H, ddd, J 50.1, 7.2, 2.8 Hz, CHF), 4.63-4.59 (1H, dq, J 7.2, 6.7 Hz, Cα+1H), 3.76 (3H, s, OCH3), 3.26-3.14 (4H, m, 2 × NCH2-allyl), 3.05 (1H, ddd, J 30.7, 14.9, 2.8 Hz, CHaHbCHF), 2.94 (1H, ddd, J 24.0, 14.9, 7.2 Hz, CHaHbCHF), 1.45 (3H, d, J 7.2 Hz, CH3); 13C NMR (101 MHz, α NHO α+1OOFN Experimental 222 CDCl3) δC 173.0 (CO2CH3), 168.6 (d, J 19.1 Hz, CONH), 135.2 (2 × =CH), 118.2 (2 × CH2=), 91.4 (d, J 187.9 Hz, CHF), 57.6 (2 × NCH2-allyl), 54.7 (d, J 19.1 Hz, CH2), 52.7 (Cα+1H), 47.8 (OCH3), 18.5 (CH3); 19F NMR (376 MHz, CDCl3) δ -191.0 (dddd, J 50.1, 30.7, 24.0, 3.7 Hz, CHF); HRMS m/z (ES+) calcd. for C13H22FN2O3 [M+H]+ requires 273.1614, found 273.1621; m/z (ES+) 273 ([M+H]+, 100%). 228b Further elution of the reaction mixture from the above preparation furnished methyl (–)-N-[(2R)-2-(diallylamino)-3-fluoropropanoyl]-(S)-alaninate 228b (36.3 mg, 0.133 mmol, 36%) as a colourless oil: Rf 0.20 (hexane:ethyl acetate, 80:20); [!]!!" -1.7 (c 3.6, CHCl3); IR νmax (neat, cm-1) 3365 (NH), 2983, 1745 (C=O), 1674 (C=O), 1500 (NH), 1450, 1157; 1H NMR (400 MHz, CDCl3) δH 7.88 (1H, d, J 7.2 Hz, CONH), 5.82 (2H, dddd, J 17.3, 10.1, 7.3, 4.9 Hz, 2 × =CH), 5.28-5.18 (4H, m, 2 × CH2=), 4.96 (1H, ddd, J 46.7, 10.3, 3.5 Hz, CHaHbF), 4.90 (1H, ddd, J 47.8, 10.3, 6.6 Hz, CHaHbF), 4.56 (1H, dq, J 7.2, 7.2 Hz, Cα+1H), 3.75 (3H, s, OCH3), 3.74 (1H, ddd, J 23.8, 6.6, 3.5 Hz, CH2FCαH), 3.41-3.19 (4H, m, 2 × NCH2-allyl) 1.39 (3H, d, J 7.2 Hz, CH3); 13C NMR (101 MHz, CDCl3) δC 173.3 (CO2CH3), 169.9 (d, J 16.1 Hz, CONH), 135.3 (2 × =CH), 118.2 (2 × CH2=), 81.1 (d, J 170.8 Hz, CH2F), 62.5 (d, J 19.1 Hz, CαH), 54.0 (2 × NCH2-allyl), 52.6 (Cα+1H), 47.8 (OCH3), 18.6 (CH3); 19F NMR (376 MHz, CDCl3) δF -228.9 (ddd, J 47.8, 46.7, 23.8 Hz, CH2F); HRMS m/z (ES+) calcd. for C13H22FN2O3 [M+H]+ requires 273.1614, found 273.1612; m/z (ES+) 273 ([M+H]+, 100%). α NHO α+1OONF Experimental 223 7.4.13 - Methyl (+)-N-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-(S)-valinate 227c & Methyl (+)-N-[(2R)-2-(diallylamino)-3-fluoropropanoyl]-(S)-valinate 228c 227c Following GP3: Starting with methyl (+)-(N,N-bisallyl-(S)-seryl)-(S)-valinate 224c (136 mg, 0.455 mmol) and diethylaminosulfur trifluoride (65.0 µL, 0.492 mmol), to yield methyl (+)-N-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-(S)-valinate 227c (12.7 mg, 42.3 µmol, 12%) as a colourless oil: Rf 0.50 (hexane:ethyl acetate, 80:20); [!]!!" +24.1 (c 1.3, CHCl3); 1H NMR (400 MHz, CDCl3) δH 6.97 (1H, br d, J 5.3 Hz, CONH), 5.83 (2H, dddd, J 17.0, 10.3, 6.6, 6.6 Hz, 2 × =CH), 5.21-5.14 (4H, m, 2 × CH2=), 5.04 (1H, ddd, J 50.1, 7.0, 2.8 Hz, CHF), 4.56 (1H, dd, J 8.9, 5.3 Hz, Cα+1H), 3.75 (3H, s, OCH3), 3.26-3.14 (4H, m, 2 × NCH2-allyl), 3.05 (1H, ddd, J 29.4, 14.9, 2.8 Hz, CHaHbCHF), 2.95 (1H, ddd, J 24.9, 14.9, 7.0 Hz, CHaHbCHF), 2.24-2.16 (1H, m, CH(CH3)2), 0.95 (3H, d, J 6.9 Hz, CH3), 0.93 (3H, d, J 6.9 Hz, CH3); 13C NMR (101 MHz, CDCl3) δC 172.0 (CO2CH3), 169.0 (d, J 19.1 Hz, CONH), 135.2 (2 × =CH), 118.2 (2 × CH2=), 91.5 (d, J 187.9 Hz, CHF), 57.6 (2 × NCH2-allyl), 56.9 (Cα+1H), 54.6 (d, J 19.1 Hz, CH2CHF), 52.4 (OCH3), 31.5 (CH(CH3)2), 19.1 (CH3), 17.9 (CH3); 19F NMR (376 MHz, CDCl3) δF -190.8 (dddd, J 50.1, 29.4, 24.9, 4.3 Hz, CHF); HRMS m/z (ES+) calcd. for C15H26FN2O3 [M+H]+ requires 301.1927, found 301.1925; m/z (ES+) 301 ([M+H]+, 100%). α NHO α+1OOFN Experimental 224 228c Further elution provided methyl (+)-N-[(2R)-2-(diallylamino)-3-fluoropropanoyl]-(S)-valinate 228c (50.8 mg, 0.169 mmol, 35%) as a colourless oil: Rf 0.4 (hexane:ethyl acetate, 80:20);![!]!!" +2.1 (c 5.0, CHCl3); IR νmax (neat, cm-1) 3367 (NH), 2962, 1741 (C=O), 1680 (C=O), 1500 (NH), 1149,; 1H NMR (400 MHz, CDCl3) δH 7.91 (1H, br d, J 9.2 Hz, CONH), 5.83 (2H, dddd, J 17.3, 10.2, 7.3, 4.6 Hz, 2 × =CH), 5.30-5.19 (4H, m, 2 × CH2=), 4.96 (1H, ddd, J 46.8, 10.3, 3.5 Hz, CHaHbF), 4.92 (1H, ddd, J 47.8, 10.3, 6.3 Hz, CHaHbF), 4.53 (1H, dd, J 9.2, 4.6 Hz, Cα+1H), 3.74 (1H, ddd, J 24.8, 6.3, 3.5 Hz, CH2FCαH), 3.74 (3H, s, OCH3), 3.45-3.39 (2H, m, 2 × NCHaHb-allyl), 3.25-3.20 (2H, m, 2 × NCHaHb-allyl), 2.25-2.15 (1H, m, CH(CH3)2), 0.92 (3H, d, J 6.9 Hz, CHCH3), 0.87 (3H, d, J 6.9 Hz, CHCH3); 13C NMR (101 MHz, CDCl3) δC 172.3 (CO2CH3), 170.2 (d, J 10.1 Hz, CONH), 135.3 (2 × =CH), 118.2 (2 × CH2=), 81.1 (d, J 171.5 Hz, CH2F), 62.8 (d, J 11.1 Hz, CH2FCαH), 56.9 (Cα+1H), 53.9 (2 × NCH2-allyl), 52.3 (OCH3), 31.3 (CH(CH3)2), 19.2 (CH3), 17.7 (CH3); 19F NMR (376 MHz, CDCl3) δF -227.6 (ddd, J 47.8, 46.8, 24.8 Hz, CH2F); HRMS m/z (ES+) calcd. for C15H26FN2O3 [M+H]+ requires 301.1927, found 301.1921; m/z (ES+) 301 ([M+H]+, 100%). α NHO α+1OONF Experimental 225 7.4.14 - (–)-(2S)-3-[(tert-Butyldimethylsilyl)oxy]-2-(diallylamino)-N-methyl-N-[(S)-1-phenylethyl]propanamide 240a Following GP1: Starting with (2S)-3-[(tert-butyldimethylsilyl)oxy]-2-diallylamino-propanoic acid (S)-230 (205 mg, 0.685 mmol), (S)-(–)-N,α-dimethylbenzylamine (188 mg, 200 µL, 1.39 mmol), diisopropylethylamine (450 µL, 2.58 mmol) and T3P (470 µL, 50% w/w ethyl acetate), the reaction yielded (–)-(2S)-3-[(tert- butyldimethylsilyl)oxy]-2-(diallylamino)-N-methyl-N-[(S)-1-phenylethyl]propanamide 240a (231 mg, 0.561 mmol, 82%) as a colourless oil: Rf 0.25 (hexane:ethyl acetate, 90:10); [!]!!" -87.0 (c 1.3, CHCl3); IR νmax (neat, cm-1) 2927, 2854, 1635 (C=O), 1404, 1095, 920 (Si–C); 1H NMR (400 MHz, CDCl3) δH (major rotamer) 7.34-7.21 (5H, m, 5 × Ar-H), 6.07 (1H, q, J 7.1 Hz, Cα+1H), 5.84-5.73 (2H, m, 2 × =CH), 5.18-5.03 (4H, m, 2 × CH2=), 4.07 (1H, dd, J 9.5, 7.6 Hz, OCHaHb), 3.91 (1H, dd, J 9.5, 5.4 Hz, OCHaHb), 3.84 (1H, dd, J 7.6, 5.4 Hz, CαH), 3.32-3.28 (4H, m, 2 × NCH2-allyl), 2.70 (3H, s, NCH3), 1.44 (3H, d, J 7.1 Hz, CHCH3), 0.85 (9H, s, C(CH3)3), 0.05 (3H, s, SiCH3), 0.03 (3H, s, SiCH3); 13C NMR (101 MHz, CDCl3) δC 172.0 (CONH), 140.8 (Ar-C), 137.2 (2 × =CH), 128.4 (2 × Ar-CH), 127.5 (2 × Ar-CH), 117.5 (Ar-CH), 116.9 (2 × CH2=), 62.1 (CH2), 60.9 (CαH), 54.0 (2 × NCH2-allyl), 50.3 (Cα+1H), 29.6 (NCH3), 26.0 (C(CH3)3), 18.4 (SiC), 15.7 (CHCH3), -5.3 (SiCH3), -5.4 (SiCH3); α NO α+1PhNOSi Experimental 226 HRMS m/z (ES+) calcd. for C24H41N2O2Si [M+H]+ requires 417.2937, found 417.2941; m/z (ES+) 417 ([M+H]+, 100%). 7.4.15 - (–)-(2S)-3-[(tert-Butyldimethylsilyl)oxy]-2-(diallylamino)-N-[(S)-1-phenylethyl]propanamide 240b Following GP1: Starting with (2S)-3-[(tert-butyldimethylsilyl)oxy]-2-bisallylamino-propanoic acid (S)-230 (104 mg, 0.347 mmol), (S)-(-)-α-methylbenzylamine (85.0 µL, 0.668 mmol), diisopropylethylamine (230 µL, 1.34 mmol) and T3P (390 µL, 50% w/w in ethyl acetate), the reaction yielded (–)-N′-α-(S)-methylbenzyl-((2S)-N,N- bisallylamino-3-(tert-butyldimethylsilyloxy)) propanamide 240b (109 mg, 0.270 mmol, 78%) as a colourless oil: Rf 0.5 (hexane:ethyl acetate, 80:20); [!]!!" -50.3 (c 1.0, CHCl3); IR νmax (neat, cm-1) 3365 (NH), 2953, 1747 (C=O), 1674 (C=O), 1498 (NH), 1259, 920 (Si–C); 1H NMR (400 MHz, CDCl3) δH 7.73 (1H, d, J 8.0 Hz, CONH), 7.33-7.20 (5H, m, 5 × Ar-H), 5.81-5.71 (2H, m, 2 × =CH), 5.18-5.08 (4H, m, 2 × CH2=), 5.02 (1H, dq, J 8.0, 6.9 Hz, Cα+1H), 4.20 (1H, dd, J 11.1, 4.2 Hz, OCHaHb), 3.98 (1H, dd, J 11.1, 7.8 Hz, OCHaHb), 3.51 (1H, dd, J 7.8, 4.2 Hz, CαH), 3.36-3.27 (4H, m, 2 × NCH2-allyl), 1.42 (3H, d, J 6.9 Hz, CHCH3), 0.88 (9H, s, C(CH3)3), 0.05 (3H, s, SiCH3), 0.04 (3H, s, SiCH3); 13C NMR (101 MHz, CDCl3) δC 171.2 (CONH), 143.6 (Ar-C), 136.2 (2 × =CH), 128.7 (2 × Ar-CH), α NHO α+1PhNOSi Experimental 227 127.3 (Ar-CH), 126.1 (2 × Ar-CH), 117.5 (2 × CH2=), 64.1 (CαH), 61.5 (OCH2), 54.0 (2 × NCH2-allyl), 48.4 (Cα+1H), 26.0 (C(CH3)3), 22.5 (CH3), 18.2 (SiC), -5.4 (SiCH3), -5.5 (SiCH3); HRMS m/z (ES+) C23H39N2O2Si [M+H]+ 403.2781, found 403.2787; m/z (ES+) 425 ([M+Na]+, 30%), 403 ([M+H]+, 100%). 7.4.16 - (–)-(2S)-2-(Diallylamino)-3-hydroxy-N-methyl-N-[(S)-1-phenylethyl]propanamide 241a Following GP2: Starting with (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-(diallylamino)-N-methyl-N-[(S)-1-phenylethyl]propanamide 240a (102 mg, 0.245 mmol), acetic acid (60.0 µL, 1.00 mmol) and TBAF (1.00 mL, 1 M in THF), the reaction yielded (–)-(2S)-2-(diallylamino)-3-hydroxy-N-methyl-N-[(S)-1- phenylethyl]propanamide 241a (58 mg, 0.192 mmol, 78%) as a colourless oil: Rf 0.15 (hexane:ethyl acetate, 80:20); [!]!!" -54.2 (c 0.8, CHCl3); IR νmax (neat, cm-1) 3419 (OH), 2926, 1630 (C=O), 1404, 1282, 1122, 995; 1H NMR (500 MHz, CDCl3) δH (major rotamer) 7.31-7.17 (5H, m, 5 × Ar-H), 5.98 (1H, q, J 7.1 Hz, Cα+1H), 5.74-5.61 (2H, m, 2 × =CH), 5.15-5.11 (4H, m, 2 × CH2=), 3.94 (1H, dd, J 10.9, 6.9 Hz, CαH), 3.76-3.70 (2H, m, CH2OH), 3.41 (1H, br s, CH2OH), 3.32 (2H, m, 2 × NCHaHb-allyl), 3.17 (2H, m, 2 × NCHaHb-allyl), 2.70 (3H, s, OCH3), 1.41 (3H, d, J 7.1 Hz, CH3); 13C NMR (126 MHz, CDCl3) δC 172.5 (CONH), 140.2 (Ar-C), α NO α+1PhNHO Experimental 228 136.1 (2 × =CH), 128.8 (2 × Ar-CH), 127.5 (2 × Ar-CH), 118.4 (Ar-CH), 117.8 (2 × CH2=), 60.8 (CαH), 58.0 (OCH2), 53.8 (2 × NCH2-allyl), 50.7 (Cα+1H), 29.7 (NCH3), 15.7 (CH3); HRMS m/z (ES+) calcd. for C18H26N2O2Na [M+Na]+ requires 325.1892, found 325.1885; m/z (ES+) 325 ([M+Na]+, 100%). 7.4.17 - (–)-(2S)-2-(Diallylamino)-3-hydroxy-N-[(S)-1-phenylethyl]propanamide 241b Following GP2: Starting with (–)-(2S)-3-[(tert-butyldimethylsilyl)oxy]-2-(diallylamino)-N-[(S)-1-phenylethyl]propanamide 240b (93 mg, 0.231 mmol), acetic acid (66 µL, 1.16 mmol) and TBAF (900 µL, 1 M in THF), the reaction yielded (–)-(2S)-2-(diallylamino)-3-hydroxy-N-[(S)-1-phenylethyl]propanamide 241b (51.0 mg, 0.180 mmol, 77%) as a colourless oil: Rf 0.11 (hexane:ethyl acetate, 80:20); [!]!!" -25.2 (c 0.7, CHCl3); IR νmax (neat, cm-1) 3325 (OH/NH), 3064, 2926, 1647 (C=O), 1521 (NH), 1494, 1280, 1128, 993; 1H NMR (300 MHz, CDCl3) δH (major rotamer) 7.63 (1H, d, J 7.8 Hz, CONH), 7.35-7.22 (5H, m, 5 × Ar-H), 5.74 (2H, dddd, J 17.3, 10.1, 7.3, 4.6 Hz, 2 × =CH), 5.23-5.12 (4H, m, 2 × CH2=), 5.06 (1H, dq, J 7.8, 6.9 Hz, Cα+1H), 3.94 (1H, dd, J 11.1, 7.9 Hz, OCHaHb), 3.79 (1H, dd, J 11.1, 3.9 Hz, OCHaHb), 3.58 (1H, br s, OH), 3.42 (1H, dd, J 7.9, 3.9 Hz, CαH), 3.34-3.26 (2H, m, 2 × NCHaHb-allyl), 3.08-3.01 (2H, m, 2 × NCHaHb-allyl), 1.45 (3H, d, J 6.9 Hz, CH3); 13C NMR (75 MHz, CDCl3) δC 173.6 (CONH), 143.1 (Ar-C), α NHO α+1PhNHO Experimental 229 135.4 (2 × =CH), 128.9 (2 × Ar-CH), 127.5 (Ar-CH), 126.0 (2 × Ar-CH), 118.1 (2 × CH2=), 62.7 (CαH), 58.2 (OCH3), 53.7 (2 × CH2-allyl), 48.5 (Cα+1H), 22.4 (CH3); HRMS m/z (ES+) calcd. for C17H24N2O2Na [M+Na]+ requires 311.1735, found 311.1739; m/z (ES+) 311 ([M+Na]+, 100%), 289 ([M+H]+, 20%). 7.4.18 - (–)-(2R)-3-(Diallylamino)-2-fluoro-N-methyl-N-[(1S)-1-phenylethyl]propanamide 242a Following GP3: Starting with (–)-(2S)-2-(diallylamino)-3-hydroxy-N-methyl-N-[(S)-1-phenylethyl]propanamide 241a (50.0 mg, 0.165 mmol) and diethylaminosulfur trifluoride 32 (25 µL, 0.189 mmol), to yield (–)-(2R)-3-(diallylamino)-2-fluoro-N-methyl-N-[(1S)-1-phenylethyl]propanamide 242a (41.0 mg, 0.135 mmol, 81%) as a colourless oil: Rf 0.09 (hexane:ethyl acetate, 90:10); [!]!!" -125 (c 1.8, CHCl3); 1H NMR (400 MHz, CDCl3) δH (major rotamer) 7.32-7.18 (5H, m, 5 × Ar-H), 5.96 (1H, qq, J 7.1, 1.6 Hz, Cα+1H), 5.83-5.70 (2H, m, 2 × =CH), 5.48-5.23 (1H, m, CHF), 5.16-5.03 (4H, m, 2 × CH2=), 3.20-3.10 (4H, m, 2 × NCH2-allyl), 3.04-2.90 (2H, m, CH2CHF), 2.63 (3H, d, J 1.6 Hz, NCH3), 1.43 (3H, d, J 7.1 Hz, CH3); 13C NMR (101 MHz, CDCl3) δC 168.2 (d, J 20.0 Hz, CONH), 139.9 (Ar-C), 135.3 (2 × =CH), 128.6 (2 × Ar-CH), 127.5 (2 × Ar-CH), 126.8 (Ar-CH), 118.2 (2 × CH2=), 89.0 (d, J 181.3 Hz, CαHF), 57.9 (2 × NCH2-allyl), 54.4 (d, α NO α+1PhFN Experimental 230 J 21.9 Hz, CH2CHF), 51.0 (Cα+1H), 28.5 (NCH3), 15.5 (CH3); 19F NMR (376 MHz, CDCl3) δF -184.5 (ddd, J 50.3, 28.9, 21.0 Hz, CHF-minor), -186.8 (ddd, J 49.3, 28.3, 20.6 Hz, CHF-major); HRMS m/z (ES+) calcd. for C18H25FN2O3Na [M+Na]+ requires 327.1849, found 327.1844; m/z (ES+) 327 ([M+Na]+, 100%), 305 ([M+Na]+, 30%). 7.4.19 - Methyl (+)-(2S)-N-allyl-phenylalanine[232] 247 Allyl bromide (1.0 mL, 12 mmol, 2.5 eq) was added to a solution of L-phenylalanine methyl ester hydrochloride 231a (1.00 mg, 4.6 mmol, 1.0 eq) and diisopropylethylamine (2.80 mL, 16 mmol, 3.5 eq) in DMF (15 mL) at 0 ºC. The solution was slowly warmed to rt and stirred for 48 h and quenched by the addition of water (10 mL) and the organics extracted with ether (3 × 20 mL). The organics were combined, washed with brine (20 mL), dried over Na2SO4, filtered and the solvent remove in vacuo. The product was purified by silica gel chromatography eluting with ethyl acetate, hexane and triethylamine (8:2 with 5% triethylamine), to yield methyl (2S)-N-allyl-phenylalaninate 247 (385 mg, 1.76 mmol, 38%) as a colourless oil: [!]!!" +23.2 (c 1.0, CH3OH) [Lit.[233] [!]!!" +23 (c 1.0, CH3OH)]; 1H NMR (300 MHz, CDCl3) δH 7.32-7.09 (5H, m, 5 × Ar-H), 5.80 (1H, dddd, J 17.2, 10.2, 6.0, 6.0 Hz, =CH), 5.15-5.04 (2H, m, CH2=), 3.64 (3H, s, OCH3), 3.56 (1H, t, J 6.9 Hz, CH2CHN), 3.26 (1H, dddd, J 13.9, 6.0, 1.5, 1.5 Hz, NCHaHb-allyl), 3.11 (1H, dddd, J 13.9, 6.0, 1.4, NH PhOO Experimental 231 1.4 Hz, NCHaHb-allyl), 2.96 (2H, d, J 6.9 Hz, CH2Ph), 1.74 (1H, br s, NH); m/z (ES+) 242 ([M+Na]+, 60%), 220 ([M+H]+, 100%). 7.4.20 - Methyl (–)-N-[O-(tert-butyldimethylsilyl)-N',N'-diallyl-(S)-seryl]-N-allyl-(S)-phenylalaninate 248 tBu P4 phosphazene 255 (1 M in hexanes, 1.00 mL, 1.00 mmol, 0.93 eq) was gradually added dropwise to a solution of methyl (+)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 232a (500 mg, 1.09 mmol, 1.0 eq) and allyl bromide (500 µL, 5.79 mmol, 5.3 eq) in THF (15 mL) at -100 ºC. The resulting mixture was gradually warmed to -78 ºC and stirred at this temperature for 20 h before being diluted with ethyl acetate (10 mL) and washed with HCl (1 M, 2 × 10 mL). The organic fractions were dried over Na2SO4, filtered and the solvent removed in vacuo. The reaction mixture was purified by silica gel column chromatography, eluting with hexane and ethyl acetate (90:10), to yield methyl (–)-N-[O-(tert-butyldimethylsilyl)-N',N'-diallyl-(S)-seryl]-N-allyl-(S)-phenylalaninate 248 (286 mg, 0.57 mmol, 53%) as a colourless oil: Rf 0.10 (hexane:ethyl acetate, 90:10); [!]!!" -102.3 (c 2.3, CHCl3); IR νmax (neat, cm-1) 2951, 2856, 1747 (C=O), 1670 (C=O), 1259, 1093, 920 (Si–C); 1H NMR (300 MHz, CDCl3) δH (major rotamer) 7.30-7.17 (5H, m, 5 × Ar-H), 5.79-5.67 (2H, m, 2 × =CH), 5.66-5.58 (1H, m, =CH), 5.17-5.06 (4H, m, 2 × CH2=), α NO α+1PhOONOSi Experimental 232 5.09-4.96 (2H, m, CH2=), 4.19 (1H, dd, J 9.6, 5.4 Hz, Cα+1H), 4.06-3.88 (4H, m, OCH2 and N′CH2-allyl), 3.71-3.67 (1H, m, CαH), 3.64 (3H, s, OCH3), 3.40-3.08 (6H, m, 2 × NCH2-allyl and CH2Ph), 0.91 (9H, s, C(CH3)3), 0.10 (3H, s, SiCH3), 0.08 (3H, s, SiCH3); 13C NMR (101 MHz, CDCl3) δC 171.4 (CONH), 171.3 (CO2CH3), 138.5 (Ar-C), 136.7 (2 × =CH), 134.5 (=CH), 129.6 (2 × Ar-CH), 128.6 (2 × Ar-CH), 126.6 (Ar-CH), 118.0 (CH2=), 117.5 (2 × CH2=), 61.3 (CαH), 60.4 (Cα+1H), 59.9 (OCH2), 53.6 (2 × NCH2-allyl), 52.0 (OCH3), 51.3 (N′CH2-allyl), 34.9 (CH2Ph), 26.1 (C(CH3)3), 18.5 (SiC), -5.2 (SiCH3), -5.3 (SiCH3); HRMS m/z (ES+) calcd. for C28H44N2O4SiNa [M+Na]+ requires 523.2968, found 523.2969; m/z (ES+) 523 ([M+Na]+, 100%), 501 ([M+H]+, 80%). 7.4.21 - Methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(R,S)-a-allyl-phenylalaninate 254 Potassium hexamethyldisilazide (1 M soln. in THF, 0.1 mL, 0.10 mmol, 0.90 eq) was added dropwise to a solution of methyl (+)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 232a (50 mg, 0.11 mmol, 1.0 eq) and allyl bromide (50 µL, 0.58 mmol, 5.3 eq) in dry THF (2.5 mL) at -78 ºC. The resulting mixture was stirred at -78 ºC for 20 h before being diluted with ethyl acetate (5 mL) and washed with HCl (1 M, 2 × 5 mL). The organic fractions were dried over Na2SO4, filtered and the solvent removed in vacuo. The reaction mixture was purified by silica α NHO α+1Ph OONOSi α NHO α+1 OONOSi Ph Experimental 233 gel column chromatography, eluting with hexane and ethyl acetate (90:10), to yield methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(R,S)-a-allyl-phenylalaninate 254 (27.1 mg, 0.31 mmol, 53%) as a colourless oil: [!]!!" -21.4 (1:1 mix of diastereoisomers, c 2.7, CHCl3); 1H NMR (400 MHz, CDCl3) δH (diastereoisomer A) 8.45 (1H, s, CONH), 7.25-6.98 (5H, m, 5 × Ar-H), 5.63-5.51 (3H, m, 3 × =CH), 5.10-4.99 (6H, m, 3 × CH2=), 4.28 (1H, dd, J 11.0, 4.1 Hz, OCHaHb), 3.99 (1H, dd, J 11.0, 9.1 Hz, OCHaHb), 3.76 (3H, s, OCH3), 3.73 (1H, d, J 13.4 Hz, CHaHbPh), 3.57 (1H, dd, J 9.1, 4.1 Hz, CαH), 3.44-3.06 (6H, m, CHaHbPh, 2 × NCH2-allyl and NCHaHb-allyl), 2.66-2.61 (1H, m, NCHaHb-allyl), 0.92 (9H, s, C(CH3)3), 0.10 (6H, s, 2 × SiCH3); δH (diastereomer B) 8.39 (1H, s, NH), 7.25-6.98 (5H, m, 5 × Ar-H), 5.63-5.51 (3H, m, 3 × =CH), 5.10-4.99 (6H, m, 3 × CH2=), 4.28 (1H, dd, J 11.0, 4.1 Hz, OCHaHb), 3.93 (1H, dd, J 11.0, 9.1 Hz, OCHaHb), 3.79 (3H, s, OCH3), 3.67 (1H, d, J 13.4 Hz, CHaHbPh), 3.56 (1H, dd, J 9.1, 4.1 Hz, CαH), 3.44-3.06 (6H, m, CHaHbPh, 2 × NCH2-allyl and NCHaHb-allyl), 2.61-2.56 (1H, m, NCHaHb-allyl), 0.92 (9H, s, C(CH3)3), 0.10 (6H, s, 2 × SiCH3); 13C NMR (75 MHz, CDCl3) δC (diastereoisomer A) 173.3 (CONH), 171.5 (CO2CH3), 136.8 (2 × =CH), 136.6 (Ar-C), 132.6 (=CH), 129.8 (2 × Ar-CH), 128.3 (2 × Ar-CH), 127.0 (Ar-CH), 119.1 (CH2=), 117.5 (2 × CH2=), 65.9 (Cα+1), 63.6 (CαH), 61.4 (OCH2), 54.1 (2 × NCH2-allyl), 52.6 (OCH3), 40.6 (CH2Ph), 39.6 (NCH2-allyl), 26.1 (C(CH3)3), 18.3 (SiC), -5.4(8) (SiCH3), -5.4(3) (SiCH3); δC (diastereoisomer B) 173.1 (CONH), 171.4 (CO2CH3), 136.7 (2 × =CH), 136.5 (Ar-C), 132.4 (=CH), 129.7 (2 × Ar-CH), 128.3 (2 × Ar-CH), 126.9 (Ar-CH), 118.9 (CH2=), 117.3 (2 × CH2=), 65.5 (Cα+1), 63.6 (CαH), 61.1 (OCH2), 53.9 (2 × NCH2-allyl), 52.5 (OCH3), 40.5 (CH2Ph), 39.5 (NCH2-allyl), 26.1 (C(CH3)3), 18.3 (SiC), -5.4(8) (SiCH3), -5.4(3) (SiCH3); Experimental 234 HRMS m/z (ES+) calcd. for C28H45N2O4Si [M+H]+ requires 501.3149, found 501.3153; m/z (ES+) 523 ([M+Na]+, 100%), 501 ([M+H]+, 70%). 7.4.22 - Methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-dibenzylglycinate 256, Methyl (–)-N-[O-(tert-butyldimethylsilyl)-N',N'-diallyl-(S)-seryl]-N-benzyl-(S)-phenylalaninate 257 & Benzyl (–)-N-[O-(tert-butyldimethylsilyl)-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 258 256 P4 phosphazene 255 (1 M soln. in hexanes, 0.10 mL, 1.00 mmol, 0.90 eq) was added dropwise to a solution of methyl (+)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 232a (50.0 mg, 0.11 mmol, 1.0 eq) and benzyl bromide (50 µL, 0.58 mmol, 5.3 eq) in THF (2 mL) at -78 ºC. The resulting mixture was stirred at -78 ºC for 20 h before being diluted with ethyl acetate (5 mL) and washed with HCl (1 M, 2 × 5 mL). The organic fractions were dried over Na2SO4, filtered and the solvent removed in vacuo. The reaction mixture was purified by silica gel column chromatography, eluting with hexane and ethyl acetate (90:10), to yield methyl (–)-N-[O-(tert-butyldimethyl)silyl-N',N'-diallyl-(S)-seryl]-dibenzylglycinate 256 (12.0 mg, 0.22 µmol, 20%) as a colourless oil: [!]!!" -19.8 (c 0.8, CHCl3); 1H NMR (500 MHz, CDCl3) δH 8.43 (1H, s, NH), 7.39-7.00 (10H, m, 10 × Ar-H), 5.31-5.23 (2H, m, 2 × =CH), 4.93-4.89 (4H, m, 2 × CH2=), 4.35 (1H, dd, J 11.0, 4.0 Hz, OCHaHb), α NHO α+1Ph OONOSi Ph Experimental 235 3.96-3.93 (1H, dd, J 11.0, 9.5 Hz, OCHaHb), 3.94 (2H, AB, J 13.5 Hz, CH2Ph), 3.79 (3H, s, OCH3), 3.55 (1 H, dd, J 9.5, 4.0 Hz, CαH), 3.30 (1H, AB, J 13.5 Hz, CHaHbPh), 3.19 (1H, AB, J 13.3, CHaHbPh), 3.05-2.92 (4H, m, 2 × NCH2-allyl), 0.9 (9H, s, C(CH3)3), 0.13 (3H, s, SiCH3), 0.12 (3H, s, SiCH3); 13C NMR (126 MHz, CDCl3) δC 172.9 (CONH), 171.7 (CO2CH3), 136.7 (Ar-C), 136.7 (Ar-C), 136.5 (2 × =CH), 129.7 (2 × Ar-CH), 129.7 (2 × Ar-CH), 128.3 (2 × Ar-CH), 128.3 (2 × Ar-CH), 127.1 (Ar-CH), 126.9 (Ar-CH), 117.3 (2 × CH2=), 67.4 (Cα+1), 63.4 (CαH), 61.2 (CH2), 53.8 (2 × NCH2-allyl), 52.5 (OCH3), 41.2 (CH2Ph), 40.9 (CH2Ph), 26.1 (C(CH3)3), 18.3 (SiC), -5.3 (SiCH3), -5.4 (SiCH3); HRMS m/z (ES+) calcd. for C32H46N2O4SiNa [M+Na]+ requires 573.3125, found 573.3117; m/z (ES+) 573 ([M+Na]+, 100%). 257 Further separation of the material enabled the isolation of methyl (–)-N-[O-(tert-butyldimethylsilyl)-N',N'-diallyl-(S)-seryl]-N-benzyl-(S)-phenylalaninate 257 (11.0 mg, 20 µmol, 18%) as a colourless oil: [!]!!" -80.1 (c 0.7, CHCl3); 1H NMR (500 MHz, CDCl3) δH 7.30-7.00 (10H, m, 10 × Ar-H), 5.71 (2H, dddd, J 17.2, 10.0, 7.4, 5.5 Hz, 2 × =CH), 5.20-5.03 (4H, m, 2 × CH2=), 4.60 (1H, AB, J 15.6 Hz, N′CHaHbPh), 4.13 (1H, dd, J 9.5, 7.5 Hz, OCHaHb), 4.06 (1H, dd, J 7.7, 6.5 Hz, Cα+1H), 4.01 (1H, AB, J 15.6 Hz, N′CHaHbPh), 3.93 (1H, dd, J 9.5, 5.3 Hz, OCHaHb), 3.81 (1H, dd, J 7.5, 5.3 Hz, CαH), 3.50 (3H, s, OCH3), 3.41-3.07 (6H, m, CH2Ph and 2 × NCH2-allyl), α NO α+1PhOONOSi Ph Experimental 236 0.93 (9H, s, C(CH3)3), 0.12 (3H, s, SiCH3), 0.10 (3H, s, SiCH3); 13C NMR (126 MHz, CDCl3) δC 171.4 (CONH), 170.9 (CO2CH3), 138.5 (Ar-C), 136.4 (2 × =CH), 136.2 (Ar-C), 129.5 (2 × Ar-CH), 128.5 (2 × Ar-CH), 128.4 (2 × Ar-CH), 128.4 (2 × Ar-CH), 127.6 (Ar-CH), 126.5 (Ar-CH), 117.5 (2 × CH2=), 61.3 (CαH), 60.2 (Cα+1H), 59.4 (OCH2), 53.6 (2 × NCH2-allyl), 51.8 (OCH3), 51.6 (N′CH2Ph), 35.2 (CH2Ph), 26.0 (C(CH3)3), 18.4 (SiC), -5.2 (2 × SiCH3); HRMS m/z (ES+) calcd. for C32H46N2O4Si [M+H]+ requires 551.3305, found 551.3320; m/z (ES+) 573 ([M+Na]+, 100%), 551 ([M+H]+, 5%). 258 Yet further separation furnished benzyl (–)-N-[O-(tert-butyldimethylsilyl)-N',N'-diallyl-(S)-seryl]-(S)-phenylalaninate 258 (9.0 mg, 16 µmol, 15%) as a colourless oil: [!]!!"!-9.7 (c 0.6, CHCl3); 1H NMR (500 MHz, CDCl3) δH 7.94 (1H, d, J 7.9 Hz, CONH), 7.37-7.27 (5H, m, 5 × Ar-H), 7.21-7.03 (5H, m, 5 × Ar-H), 5.61-5.53 (2H, m, 2 × =CH), 5.14 (2H, ABq, J 12.2 Hz, OCH2Ph), 5.12-5.02 (4H, m, 2 × CH2=), 4.86 (1H, ddd, J 7.9, 6.4, 5.9 Hz, Cα+1H), 4.16 (1H, dd, J 11.1, 4.1 Hz, OCHaHb), 3.90 (1H, dd, J 11.1, 8.4 Hz, OCHaHb), 3.53 (1H, dd, J 8.4, 4.1 Hz, CαH), 3.21-3.18 (4H, m, 2 × NCH2-allyl), 3.15 (1H, dd, J 14.0, 5.9 Hz, CHaHbPh), 3.08 (1H, dd, J 14.0, 6.4 Hz, CHaHbPh), 0.89 (9H, s, C(CH3)3), 0.06 (6H, s, 2 × SiCH3); 13C NMR (126 MHz, CDCl3) δC 171.9 (CONH), 171.4 (CO2Bn), 136.4 (2 × =CH), 136.1 (Ar-C), 135.3 (Ar-C), 129.4 (2 × Ar-CH), 128.7 (2 × Ar-CH), 128.7 (2 × Ar-CH), 128.6 (2 × Ar-CH), 128.6 (Ar-CH), 127.1 (Ar-CH), 117.4 (2 × CH2=), 67.3 (OCH2Ph), α NHO α+1PhOONOSi Ph Experimental 237 63.8 (CαH), 61.3 (OCH2), 54.0 (2 × NCH2-allyl), 53.1 (Cα+1H), 38.1 (CH2Ph), 26.0 (C(CH3)3), 18.2 (SiC), -5.4 (2 × SiCH3); HRMS m/z (ES+) calcd. for C31H45N2O4Si [M+H]+ 537.3149, found 537.3161; m/z (ES+) 559 ([M+Na]+, 100%), 537 ([M+H]+, 10%). 7.4.23 - Methyl (–)-N-[N',N'-diallyl-(S)-seryl]-N-allyl-(S)-phenylalaninate 244 Following GP2: Starting with methyl (–)-N-[O-(tert-butyldimethylsilyl)-N',N'-diallyl-(S)-seryl]-N-allyl-(S)-phenylalaninate 248 (206 mg, 0.411 mmol), acetic acid (120.0 µL, 2.06 mmol) and TBAF (1.6 mL, 1 M in THF), the reaction yielded methyl (–)-N-[N',N'-diallyl-(S)-seryl]-N-allyl-(S)-phenylalaninate 244 (133 mg, 0.345 mmol, 84%) as a colourless oil: Rf 0.12 (hexane:ethyl acetate, 70:30); ![!]!!" -124 (c 1.2, CHCl3); IR νmax (neat, cm-1) 3446 (OH), 3078, 2949, 1743 (C=O), 1635 (C=O), 1436, 1274, 1195, 993; 1H NMR (400 MHz, CDCl3) δH (major rotamer) 7.25-7.06 (5H, m, 5 × Ar-H), 5.63 (2H, dddd, J 17.3, 10.0, 7.4, 5.4 Hz, 2 × =CH), 5.59-5.50 (1H, m, =CH), 5.11-5.03 (4H, m, 2 × CH2=), 5.03-4.92 (2H, m, CH2=), 4.12 (1H, dd, J 10.2, 5.2 Hz, Cα+1H), 4.00-3.94 (1H, m, N′CHaHb-allyl), 3.88 (1H, dd, J 11.3, 7.3 Hz, OCHaHb), 3.69 (1H, dd, J 11.3, 5.0 Hz, OCHaHb), 3.62 (3H, s, OCH3), 3.60 (1H, dd, J 7.3, 5.0 Hz, CαH), 3.32 (1H, dd, J 14.0, 5.2, CHaHbPh), 3.22-3.04 (6H, m, CHaHbPh, 2 × NCH2-allyl and N′CHaHb-allyl), 2.27 (1H, br s, OH); 13C NMR (101 MHz, α NO α+1PhOONHO Experimental 238 CDCl3) δC 172.5 (CON), 170.9 (CO2CH3), 138.0 (Ar-C), 136.1 (2 × =CH), 133.8 (=CH), 129.4 (2 × Ar-CH), 128.8 (2 × Ar-CH), 126.9 (Ar-CH), 118.8 (CH2=), 118.1 (2 × CH2=), 60.8 (CαH), 60.4 (Cα+1H), 57.8 (OCH2), 53.5 (2 × NCH2-allyl), 52.2 (OCH3), 51.6 (N′CH2-allyl), 34.7 (CH2Ph); HRMS m/z (ES+) calcd. for C22H30N2O4Na [M+Na]+ requires 409.2103, found 409.2090; m/z (ES+) 409 ([M+Na]+, 100%). 7.4.24 - Methyl (–)-N-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-N-allyl-(S)-phenylalanate 245 Following GP3: Starting with methyl (–)-N-[N',N'-diallyl-(S)-seryl]-N-allyl-(S)-phenylalaninate 244 (76.3 mg, 0.197 mmol) and diethylaminosulfur trifluoride 32 (30.0 µL, 0.265 mmol), to yield methyl (–)-N-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-N-allyl-(S)-phenylalanate 245 (56 mg, 0.144 mmol, 73%) as a colourless oil: Rf 0.1 (hexane:ethyl acetate, 90:10); [!]!!" -39.2 (c 1.3, CHCl3); IR νmax (neat, cm-1) 3076, 2949, 1743 (C=O), 1653 (C=O), 1436, 1222, 1166, 993; 1H NMR (500 MHz, CDCl3) δH (major rotamer) 7.25-7.09 (5H, m, 5 × Ar-H), 5.74 (2H, dddd, J 17.0, 10.3, 6.6, 6.6 Hz, 2 × =CH), 5.52-5.44 (1H, m, =CH), 5.12-5.04 (6H, m, 3 × CH2=), 5.06 (1H, ddd, J 49.7, 8.0, 3.2 Hz, CHF), 4.31 (1H, dd, J 10.3, 5.3 Hz, Cα+1H), 3.86-3.82 (2H, m, N′CH2-allyl), 3.64 (3H, s, OCH3), 3.30 (1H, α NO α+1PhOOFN Experimental 239 dd, J 14.1, 5.3 Hz, CHaHbPh), 3.17 (1H, dd, J 14.1, 10.3 Hz, CHaHbPh), 3.13-3.03 (4H, m, 2 × NCH2-allyl), 2.82 (1H, ddd, J 18.1, 15.0, 8.0 Hz, CHaHbCHF), 2.68 (1H, ddd, J 31.8, 15.0, 3.2 Hz, CHaHbCHF); 13C NMR (125 MHz, CDCl3) δC 170.5 (CO2CH3), 168.5 (d, J 20.4 Hz, CON), 137.6 (Ar-C), 135.3 (2 × =CH), 133.1 (=CH), 129.4 (2 × Ar-CH), 128.7 (2 × Ar-CH), 126.9 (Ar-CH), 118.4 (CH2=), 118.1 (2 × CH2=), 88.4 (d, J 180.7 Hz, CHF), 60.8 (Cα+1H), 57.7 (2 × NCH2-allyl), 54.3 (d, J 21.6 Hz, CH2CHF), 52.4 (OCH3), 50.8 (d, J 4.1 Hz, N′CH2-allyl), 34.8 (CH2Ph); 19F NMR (470 MHz, CDCl3) δF -185.4 (ddd, J 49.6, 33.5, 19.0 Hz, CHF-minor rotamer), -187.2 (ddd, J 49.7, 31.8, 18.1 Hz, CHF-major rotamer); HRMS m/z (ES+) calcd. for C22H29FN2O3Na [M+Na]+ requires 411.2060, found 411.2058; m/z (ES+) 411 ([M+Na]+, 100%). 7.4.25 - Methyl (–)-N-[(2R)-3-amino-2-fluoropropanoyl]-N-allyl-(S)-phenylalanate hydrochloride 260 1,4-Di(phenylphosphino)butane (16.5 mg, 37.3 µmol, 30.0 mol%) was added to a solution of tris(dibenzylideneacetone)dipalladium (22.2 mg, 24.2 µmol, 25.0 mol%) in THF (5 mL) and stirred for 15 min until the solution turned yellow. This solution was added via canula to a solution of methyl (–)-N-[(2R)-3-(diallylamino)-2-fluoropropanoyl]-N-allyl-(S)-phenylalanate 254 (51.6 mg, 0.133 mmol, 1.0 eq) and α NO α+1PhOOFClH3N Experimental 240 2-mercaptosalicylic acid (60.0 mg, 0.389 mmol, 2.9 eq) in THF (7.0 mL) and the solution was brought to reflux for 3 h. The reaction was cooled to rt and water (10 mL) and HCl (1 M, 0.2 mL) were added. The precipitate was isolated by filtration and washed repeatedly with water with the filtrate collected and the solvent removed in vacuo to furnish a yellow solid. This solid was reconstituted in water and re-filtered and the sample lyophilised to yield methyl (–)-N-[(2R)-3-amino-2-fluoropropanoyl]-N-allyl-(S)-phenylalanate hydrochloride 260 (41 mg, 0.20 mmol, 93%) as a colourless solid which was used without further purification: [!]!!" -41.7 (c 1.0, D2O); 1H NMR (500 MHz, D2O) δH (major rotamer) 7.36-7.22 (5H, m, 5 × Ar-H), 5.62 (1H, ddd, J 48.0, 7.4, 3.4 Hz, CαHF), 5.58-5.51 (1H, m, =CH), 5.18-5.14 (2H, m, CH2=), 4.65 (1H, dd, J 10.7, 5.1 Hz, Cα+1H), 3.94 (1H, m, N′CHaHb-allyl), 3.71 (3H, s, OCH3), 3.41-3.32 (2H, m, CH2CHF), 3.29-3.16 (3H, m, CH2Ph and N′CHaHb-allyl); 13C NMR (101 MHz, D2O) δC 172.2 (CO2CH3), 167.5 (d, J 19.8 Hz, CONH), 136.9 (Ar-C), 131.6 (=CH), 129.4 (2 × Ar-CH), 128.8 (2 × Ar-CH), 127.1 (Ar-CH), 119.3 (CH2=), 84.3 (d, J 179 Hz, CHF), 61.3 (Cα+1H), 53.0 (OCH3), 51.5 (N′CH2-allyl), 40.0 (d, J 21.3 Hz, CH2CHF), 33.5 (CH2Ph); 19F NMR (376 MHz, D2O) δF -193.9 (ddd, J 48.0, 27.9, 20.2 Hz, CHF-minor rotamer), -194.7 (ddd, J 48.0, 26.8, 21.5 Hz, CHF-major rotamer); HRMS m/z (ES+) calcd. for C16H22FN2O3 [M+H]+ requires 309.1614, found 309.1621; m/z (ES+) 331 ([M+Na]+, 50%), 309 ([M+H]+, 100%). Experimental 241 7.4.26 - Methyl (–)-N-{(2R)-3-[(tert-butoxycarbonyl)amino]-2-fluoropropanoyl}-N-allyl-(S)-phenylalanate 265 Diisopropylethylamine (40.0 µL, 0.230 mmol, 3.0 eq) and di-tert-butyl dicarbonate (22.0 mg, 0.101 mmol, 1.3 eq) were added to a solution of methyl (–)-N-((2R)-3-amino-2-fluoropropanoyl)-N-allyl-(S)-phenylalanate hydrochloride 260 (24.0 mg, 69.6 µmol, 1.0 eq) in aqueous dioxane (2 mL, 25% v/v) and the mixture was stirred at rt for 24 h. The reaction was quenched by the addition of saturated aqueous Na2CO3 (2 mL) and the aqueous phase extracted with ethyl acetate (2 × 2 mL). The organic extracts were combined, dried over Na2SO4, filtered and the solvent removed in vacuo to yield an oil. The product was purified by silica gel column chromatography, eluting with hexane and ethyl acetate (90:10), to yield methyl (–)-N-{(2R)-3-[(tert-butoxycarbonyl)amino]- 2-fluoropropanoyl}-N-allyl-(S)-phenylalanate 265 (20.1 mg, 49.2 µmol, 71%) as a colourless oil: [!]!!" -52.1 (c 2.0, CHCl3); 1H NMR (400 MHz, CDCl3) δH (major rotamer) 7.31-7.16 (5H, m, 5 × Ar-H), 5.55-5.46 (1H, m, =CH), 5.17-5.10 (2H, m, CH2=), 5.15-5.02 (1H, m, CαHF), 4.93 (1H, t, J 6.3 Hz, NHBoc), 4.40 (1H, dd, J 10.4, 5.1 Hz, Cα+1H), 3.96-3.87 (1H, m, N′CHaHb-allyl), 3.73 (3H, s, OCH3), 3.58-3.29 (3H, m, CH2CHF and N′CHaHb-allyl), 3.38 (1H, dd, J 14.2, 5.1 Hz, CHaHbPh), 3.24 (1H, dd, J 14.2, 10.4 Hz, CHaHbPh), 1.44 (9H, s, C(CH3)3); 13C NMR (75 MHz, CDCl3) δC 170.5 (CO2CH3), 167.8 (d, J 20.3 Hz, CONH), α NO α+1PhOOFNHO O Experimental 242 156.0 (OCOtBu), 137.5 (Ar-C), 132.8 (=CH), 129.5 (2 × Ar-CH), 128.7 (2 × Ar-CH), 127.0 (Ar-CH), 119.0 (CH2=), 86.4 (d, J 181 Hz, CHF), 60.9 (Cα+1H), 52.5 (OCH3), 51.1 (N′CH2-allyl), 41.7 (d, J 23.8 Hz, CH2CHF), 34.8 (CH2Ph), 30.0 (C(CH3)3), 28.5 (C(CH3)3); 19F NMR (376 MHz, CDCl3) δF -191.5 (ddd, J 47.3, 25.4, 19.6 Hz, CHF-minor rotamer), -192.5 (ddd, J 48.0, 22.9, 17.4 Hz, CHF-major rotamer); HRMS m/z (ES+) calcd. for C21H29FN2O5Na [M+Na]+ requires 431.1958, found 431.1948; m/z (ES+) 431 ([M+Na]+, 100%). 7.4.27 - Methyl (–)-N-{(2R)-3-[(tert-butoxycarbonyl)amino]-2-fluoropropanoyl}-N-formyl-(S)-phenylalanate 267 Ru(CO)HCl(PPh3)3 (4.7 mg, 4.9 µmol, 10 mol%) was added to a solution of methyl (–)-N-{(2R)-3-[(tert-butoxycarbonyl)amino]-2-fluoropropanoyl}-N-allyl-(S)-phenylalanate 265 (20 mg, 48.9 µmol, 1.0 eq) in toluene (2 mL) and the mixture was brough to reflux for 3 h. The solution was cooled to rt and the solvent removed in vacuo. RuCl3 (1.0 mg, 1.7 µmol, 3.5 mol%) and NaIO4 (20.8 mg, 97.8 mmol, 2 eq) in aqueous 1,2-dichloroethane (50% v/v, 1 mL) were added to the isomerised product and the mixture was stirred at rt for 24 hr. The reaction was quenched by the addition of saturated aqueous Na2CO3 (1 mL) and the organics extracted with ethyl acetate (2 × 2 mL). The organic phases were combined washed with brine (1 mL) and dried over Na2SO4, filtered and the solvent removed in vacuo to yield an oil. The α NO α+1PhOOFNH OO O H Experimental 243 product was purified by silica gel chromatography to yield methyl (–)-N-{(2R)-3-[(tert- butoxycarbonyl)amino]-2-fluoropropanoyl}-N-formyl-(S)-phenylalanate 267 (11.5 mg, 29 µmol, 59%) as a colourless oil: [!]!!" -30.2 (c 1.1, CHCl3); 1H NMR (500 MHz, CDCl3) δH 8.97 (1H, s, CHO), 7.29-7.10 (5H, m, 5 × Ar-H), 5.49 (1H, dd, J 10.2, 5.2 Hz, NHBoc), 5.35-5.26 (1H, br m, CHF), 4.74 (1H, m, Cα+1H), 3.76 (3H, s, OCH3), 3.52 (1H, dd, J 14.2, 5.4 Hz, CHaHbPh), 3.54-3.34 (1H, m, CHaHbCHF), 3.28 (1H, dd, J 14.2, 11.1 Hz, CHaHbPh), 3.32-3.23 (1H, m, CHaHbCHF), 1.44 (9H, s, C(CH3)3); 13C NMR (126 MHz, CDCl3) δC 169.1 (CO2CH3), 168.8 (CHO), 161.3 (d, J 9.6 Hz, CONH), 155.7 (OCOtBu), 136.3 (Ar-C), 129.2 (2 × Ar-CH), 128.6 (2 × Ar-CH), 127.1 (Ar-CH), 87.0 (d, J 186 Hz, CHF), 60.4 (1H, s, Cα+1H), 52.8 (OCH3), 41.8 (d, J 23.1 Hz, CH2CHF), 34.3 (CH2Ph), 29.7 (C(CH3)3), 28.3 (C(CH3)3); 19F NMR (470 MHz, CDCl3) δF -190.5 (br m, CHF-minor rotamer), -191.5 (br m, CHF-major rotamer); HRMS m/z (ES+) calcd. for C19H25FN2O6Na [M+Na]+ 419.1594, found 419.1588; m/z (ES+) 419 ([M+Na]+, 100%). 7.4.28 - Methyl (–)-N-{(2R)-3-[(tert-butoxycarbonyl)amino]-2-fluoropropanoyl}-(S)-phenylalanate 268 Saturated aqueous sodium carbonate (0.5 mL) was added to a solution of NaHCO3 (1.0 mg, 9.4 µmol, 0.33 eq) and methyl (–)-N-{(2R)-3-[(tert-butoxycarbonyl)amino]-2-fluoropropanoyl}-N-formyl-(S)-phenylalanate 267 (11.0 mg, α NHO α+1PhOOFNHO O Experimental 244 28 µmol, 1.0 eq) in aqueous acetone (25% v/v, 1 mL) and the mixture was stirred vigorously for 12 h at rt. The mixture was diluted with water (1 mL) and ethyl acetate (2 mL), the organic phase was separated and the aqueous layer further extracted with ethyl acetate (2 mL). The organic phases were combined, washed with brine (1 mL), dried with Na2SO4, filtered and the solvent removed in vacuo. The oil was purified by silica gel column chromatography, to yield methyl (–)-N-{(2R)-3-[(tert- butoxycarbonyl)amino]-2-fluoropropanoyl}-(S)-phenylalanate 268 (4.7 mg, 12 µmol, 46%) as a colourless solid: [!]!!" -26.1 (c 0.1, CHCl3); 1H NMR (500 MHz, CDCl3) δH 7.34-7.11 (5H, m, 5 × Ar-H), 6.70 (1H, br d, J 4.9 Hz, CONH), 4.95-4.84 (3H, m, CHF, Cα+1H and NHBoc), 3.82-3.72 (1H, m, CHaHbCHF), 3.75 (3H, s, OCH3), 3.54-3.44 (1H, m, CHaHbCHF), 3.19 (1H, dd, J 14.0, 5.7 Hz, CHaHbPh), 3.12 (1H, dd, J 14.0, 6.5 Hz, CHaHbPh), 1.43 (9H, s, C(CH3)3); 13C NMR (126 MHz, CDCl3) δC 171.2 (CO2CH3), 167.9 (d, J 20.7 Hz, CONH), 155.7 (OCOtBu), 135.3 (Ar-C), 129.1 (2 × Ar-CH), 128.8 (2 × Ar-CH), 127.4 (Ar-C), 90.0 (d, J 194.5 Hz, CHF), 52.8 (Cα+1H), 52.6 (OCH3), 42.1 (d, J 21.1 Hz, CH2CHF), 37.7 (CH2Ph), 29.7 (C(CH3)3), 28.3 (C(CH3)3); 19F NMR (470 MHz, CDCl3) δF -195.3 (ddd, J 48.1, 23.7, 23.7 Hz, CHF); HRMS m/z (ES+) calcd. for C18H25FN2O5Na [M+Na]+ requires 391.1645, found 391.1645; m/z (ES+) 391 ([M+Na]+, 100%). References 245 References 1. A. Tressaud, Angew. Chem., Int. Ed., 2006, 45, 6792-6796. 2. N. N. Greenwood and A. Earnshaw, Chemistry of the Elements, Elsevier, Oxford, 2nd edn., 2005, vol. Elsevier. 3. J. R. Breen, G. Sandford, D. S. Yufit, J. A. K. Howard, J. Fray, and B. Patel, Beilstein J. Org. Chem., 2011, 7, 1048-1054. 4. T. Furuya, C. A. Kuttruff, and T. Ritter, Curr. Opin. Drug Discovery Dev., 2008, 11, 803-819. 5. C. Hollingworth and V. Gouverneur, Chem. Commun., 2012, 48, 2929-2942. 6. D. H. R. Barton, L. S. Godinho, R. H. Hesse, and M. M. Pechet, Chem. Commun., 1968, 804-806. 7. T. Umemoto, S. Fukami, G. Tomizawa, K. Harasawa, K. Kawada, and K. Tomita, J. Am. Chem. Soc., 1990, 112, 8563-8575. 8. S. Sing, D. D. DesMarteau, S. S. Zuberi, M. Witz, and H.-N. Haung, J. Am. Chem. Soc., 1987, 109, 7194-7196. 9. E. Differding and R. W. Lang, Tetrahedron Lett., 1988, 29, 6087-6090. 10. T. D. Beeson and D. W. C. MacMillan, J. Am. Chem. Soc., 2005, 127, 8826-8828. 11. M. Marigo, D. Fielenbach, A. Braunton, A. Kjærsgaard, and K. A. Jørgensen, Angew. Chem., Int. Ed., 2005, 44, 3703-3706. 12. D. D. Steiner, N. Mase, and C. F. Barbas, Angew. Chem., Int. Ed., 2005, 44, 3706-3710. 13. R. E. Banks, J. Fluorine Chem., 1998, 87, 1-17. 14. P. T. Nyffeler, S. G. Durón, M. D. Burkart, S. P. Vincent, and C.-H. Wong, Angew. Chem., Int. Ed., 2005, 44, 192-212. 15. V. Rauniyar, A. D. Lackner, G. L. Hamilton, and F. D. Toste, Science, 2011, 334, 1681-1684. 16. T. Furuya, A. E. Strom, and T. Ritter, J. Am. Chem. Soc., 2009, 131, 1662-1663. 17. H. Teare, E. G. Robins, A. Kirjavainen, S. Forsback, G. Sandford, O. Solin, S. K. Luthra, and V. Gouverneur, Angew. Chem., Int. Ed., 2010, 49, 6821-6824. References 246 18. J. H. Clark, Chem. Rev., 1980, 80, 429-452. 19. D. O’Hagan, Chem. Soc. Rev., 2008, 37, 308-319. 20. C. W. Tullock and D. D. Coffman, J. Org. Chem., 1960, 25, 2016-2019. 21. I. Shahak and E. D. Bergmann, J. Chem. Soc. C, 1967, 319-320. 22. M. H. Litt, A.-M. Weidler-Kubanek, and F. P. Avonda, J. Org. Chem., 1968, 33, 1837-1839. 23. W. J. Middleton, J. Org. Chem., 1975, 40, 574-578. 24. M. J. Tozer and T. F. Herpin, Tetrahedron, 1996, 52, 8619-8683. 25. H. Hayashi, H. Sonoda, and T. Nagata, Chem. Commun., 2002, 1618-1619. 26. F. Beaulieu, L.-P. Beauregard, G. Courchesne, M. Couturier, F. LaFlamme, and A. L’Heureux, Org. Lett., 2009, 11, 5050-5053. 27. A. L’Heureux, F. Beaulieu, C. Bennett, D. R. Bill, S. Clayton, F. LaFlamme, M. Mirmehrabi, S. Tadayon, D. Tovell, and M. Couturier, J. Org. Chem., 2010, 75, 3401-3411. 28. E. J. Cho, T. D. Senecal, T. Kinzel, Y. Zhang, D. A. Watson, and S. L. Buchwald, Science, 2010, 328, 1679-1681. 29. Y. Ji, T. Brueckl, R. D. Baxter, Y. Fujiwara, I. B. Seiple, S. Su, D. G. Blackmond, and P. S. Baran, Proc. Natl. Acad. Sci. USA, 2011, 108, 14411-14415. 30. D. A. Nagib and D. W. C. MacMillan, Nature, 2011, 480, 224-248. 31. B. R. Langlois, E. Laurent, and N. Roidot, Tetrahedron Lett., 1991, 32, 7525-7528. 32. A. T. Parsons and S. L. Buchwald, Nature, 2011, 480, 184-185. 33. T. Besset, C. Schneider, and D. Cahard, Angew. Chem., Int. Ed., 2012, 51, 5048-5050. 34. M. Pagliaro and R. Ciriminna, J. Mater. Chem., 2005, 15, 4981-4991. 35. H.-J. Böhm, D. Banner, S. Bendels, M. Kansy, B. Kuhn, K. Müller, U. Obst-Sander, and M. Stahl, ChemBioChem, 2004, 5, 637-643. 36. S. Purser, P. R. Moore, S. Swallow, and V. Gouverneur, Chem. Soc. Rev., 2008, 37, 320-330. 37. K. Müller, C. Faeh, and F. Diederich, Science, 2007, 317, 1881-1886. References 247 38. P. Jeschke, ChemBioChem, 2004, 5, 571-589. 39. W. K. Hagmann, J. Med. Chem., 2008, 51, 4359-4369. 40. A. Bondi, J. Phys. Chem., 1964, 68, 441-451. 41. B. E. Smart, J. Fluorine Chem., 2001, 109, 3-11. 42. D. M. Lemal, J. Org. Chem., 2004, 69, 1-11. 43. M. H. Abraham, P. L. Grellier, D. V. Prior, P. P. Duce, J. J. Morris, and P. J. Taylor, J. Chem. Soc., Perkin Trans. 2, 1989, 699-711. 44. M. H. Abraham, P. L. Grellier, D. V. Prior, J. J. Morris, and P. J. Taylor, J. Chem. Soc., Perkin Trans. 2, 1990, 521-529. 45. S. B. Rosenblum, T. Huynh, A. Afonso, H. R. Davis, N. Yumibe, J. W. Clader, and D. A. Burnett, J. Med. Chem., 1998, 41, 973-980. 46. C. Heidelberger, N. K. Chaudhuri, P. Danneberg, D. Mooren, L. Griesbach, R. Duschinsky, R. J. Schnitzer, E. Pleven, and J. Scheiner, Nature, 1957, 179, 663-666. 47. J. E. Barrett, D. A. Maltby, D. V. Santi, and P. G. Schultz, J. Am. Chem. Soc., 1998, 120, 449-450. 48. J. G. Bundy, E. M. Lenz, D. Osborn, J. M. Weeks, J. C. Lindon, and J. K. Nicholson, Xenobiotica, 2002, 32, 479-490. 49. C. J. Duckett, I. D. Wilson, D. S. Douce, H. J. Walker, F. R. Abou-Shakra, J. C. Lindon, and J. K. Nicholson, Xenobiotica, 2007, 37, 1378-1393. 50. C. Schaffrath, S. L. Cobb, and D. O’Hagan, Angew. Chem., Int. Ed., 2002, 41, 3913-3915. 51. M. Onega, R. P. McGlinchey, H. Deng, J. T. G. Hamilton, and D. O’Hagan, Chem. Commun., 2007, 35, 375-385. 52. S. Klimasauskas, T. Szyperski, S. Serva, and K. Wüthrich, EMBO J., 1998, 17, 317-324. 53. M. A. Danielson and J. J. Falke, Annu. Rev. Biophys. Biomol. Struct., 1996, 25, 163-195. 54. S. L. Cobb and C. Murphy, J. Fluorine Chem., 2009, 130, 132-143. 55. R. Moumne, M. Pasco, E. Prost, T. Lecourt, L. Micouin, and C. Tisne, J. Am. Chem. Soc., 2010, 132, 13111-13113. References 248 56. F. Chung, C. Tisné, T. Lecourt, B. Seijo, F. Dardel, and L. Micouin, Chem. Eur. J., 2009, 15, 7109-7116. 57. F. Chung, C. Tisné, T. Lecourt, F. Dardel, and L. Micouin, Angew. Chem., Int. Ed., 2007, 46, 4489-4491. 58. S. Bresciani, T. Lebl, A. M. Z. Slawin, and D. O’Hagan, Chem. Commun., 2010, 46, 5434-5436. 59. J. Battiste and R. Newmark, Prog. Nucl. Magn. Reson. Spectrosc., 2006, 48, 1-23. 60. T. M. O’Connell, S. A. Gabel, and R. E. London, Biochem., 1994, 33, 10985-10992. 61. F. R. Souza, M. P. Freitas, and R. Rittner, J. Mol. Struct.: THEOCHEM, 2008, 863, 137-140. 62. L. Goodman, H. Gu, and V. Pophristic, J. Phys. Chem. A, 2005, 109, 1223-1229. 63. I. V. Alabugin, K. M. Gilmore, and P. W. Peterson, Wiley Interdiscip. Rev.: Comput. Mol. Sci., 2011, 1, 109-141. 64. D. Y. Buissonneaud, T. van Mourik, and D. O’Hagan, Tetrahedron, 2010, 66, 2196-2202. 65. D. O’Hagan, J. Org. Chem., 2012, 77, 3689-3699. 66. D. Wu, A. Tian, and H. Sun, J. Phys. Chem. A, 1998, 102, 9901-9905. 67. D. Farran, A. M. Z. Slawin, P. Kirsch, and D. O’Hagan, J. Org. Chem., 2009, 74, 7168-7171. 68. L. Hunter, A. M. Z. Slawin, P. Kirsch, and D. O’Hagan, Angew. Chem., Int. Ed., 2007, 46, 7887-7890. 69. L. Hunter, P. Kirsch, A. M. Z. Slawin, and D. O’Hagan, Angew. Chem., Int. Ed., 2009, 48, 5457-5460. 70. A. G. Myers, J. K. Barbay, and B. Zhong, J. Am. Chem. Soc., 2001, 123, 7207-7219. 71. K. B. Wiberg and P. R. Rablen, J. Am. Chem. Soc., 1993, 115, 614-625. 72. J. W. Banks, A. S. Batsanov, J. A. K. Howard, D. O’Hagan, H. S. Rzepa, and S. Martin-Santamaria, J. Chem. Soc., Perkin Trans. 2, 1999, 2409-2411. 73. C. R. S. Briggs, D. O’Hagan, J. A. K. Howard, and D. S. Yu, J. Fluorine Chem., 2003, 119, 9-13. References 249 74. C. F. Tormena, N. S. Amadeu, R. Rittner, and R. J. Abraham, J. Chem. Soc., Perkin Trans. 2, 2002, 773-778. 75. P. R. Olivato, S. A. Guerrero, M. H. Yreijo, R. Rittner, and C. F. Tormena, J. Mol. Struct., 2002, 607, 87-99. 76. R. I. Mathad, B. Jaun, O. Flögel, J. Gardiner, M. Löweneck, J. D. C. Codée, P. H. Seeberger, and D. Seebach, Helv. Chim. Acta, 2007, 90, 2251-2273. 77. B. Jaun, D. Seebach, and R. I. Mathad, Helv. Chim. Acta, 2011, 94, 355-361. 78. C. R. S. Briggs, M. J. Allen, D. O’Hagan, D. J. Tozer, A. M. Z. Slawin, A. E. Goeta, and J. A. K. Howard, Org. Biomol. Chem., 2004, 2, 732-740. 79. N. E. J. Gooseman, D. O’Hagan, M. J. G. Peach, A. M. Z. Slawin, D. J. Tozer, and R. J. Young, Angew. Chem., Int. Ed., 2007, 46, 5904-5908. 80. A. Sun, D. C. Lankin, K. Hardcastle, and J. P. Snyder, Chem.-Eur. J., 2005, 11, 1579-1591. 81. N. E. J. Gooseman, D. O’Hagan, A. M. Z. Slawin, A. M. Teale, D. J. Tozer, and R. J. Young, Chem. Commun., 2006, 3190-3192. 82. N. E. J. Gooseman, Ph.D. Thesis, University of St Andrews, Scotland, 2008. 83. L. E. Zimmer, C. Sparr, and R. Gilmour, Angew. Chem., Int. Ed., 2011, 50, 11860-11871. 84. C. Sparr, W. B. Schweizer, H. M. Senn, and R. Gilmour, Angew. Chem., Int. Ed., 2009, 48, 3065-3068. 85. A. Lattanzi, Org. Lett., 2005, 7, 2579-2582. 86. M. D. Clift, H. Ji, G. P. Deniau, D. O’Hagan, and R. B. Silverman, Biochemistry, 2007, 46, 13819-13828. 87. G. Deniau, A. M. Z. Slawin, T. Lebl, F. Chorki, J. P. Issberner, T. van Mourik, J. M. Heygate, J. J. Lambert, L.-A. Etherington, K. T. Sillar, and D. O’Hagan, ChemBioChem, 2007, 8, 2265-2274. 88. I. Yamamoto, G. P. Deniau, N. Gavande, M. Chebib, G. A. R. Johnston, and D. O’Hagan, Chem. Commun., 2011, 47, 7956-7958. 89. L. Hunter, Beilstein J. Org. Chem., 2010, 6, 14pp. 90. L. Hunter, K. A. Jolliffe, M. J. T. Jordan, P. Jensen, and R. B. Macquart, Chem. Eur. J., 2011, 2340-2343. 91. I. Yamamoto, M. J. T. Jordan, N. Gavande, M. R. Doddareddy, M. Chebib, and L. Hunter, Chem. Commun., 2012, 48, 829-831. References 250 92. T. Ohba, E. Ikeda, and H. Takei, Bioorg. Med. Chem. Lett., 1996, 6, 1875-1880. 93. L. Somekh and A. Shanzer, J. Am. Chem. Soc., 1982, 104, 5836-5837. 94. D. F. Hook, F. Gessier, C. Noti, P. Kast, and D. Seebach, ChemBioChem, 2004, 5, 691-706. 95. T.-X. Métro, B. Duthion, D. G. Pardo, and J. Cossy, Chem. Soc. Rev., 2010, 39, 89-102. 96. M. K. Edmonds, F. H. M. Graichen, J. Gardiner, and A. D. Abell, Org. Lett., 2008, 10, 885-887. 97. S. G. Davies, N. M. Garrido, D. Kruchinin, O. Ichihara, L. J. Kotchie, P. D. Price, A. J. P. Mortimer, A. J. Russell, and A. D. Smith, Tetrahedron: Asymmetry, 2006, 17, 1793-1811. 98. P. J. Duggan, M. Johnston, and T. L. March, J. Org. Chem., 2010, 75, 7365-7372. 99. X.-L. Qiu and F.-L. Qing, Eur. J. Org. Chem., 2011, 3261-3278. 100. Nobel Prize, The 2009 Nobel Prize in Physiology or Medicine Nobelprize.org. Feb 2012 www.nobelprize.org/nobel_prizes/medicine/laureates/2009/press.html, 2009. 101. J. W. Szoztak and E. H. Blackburn, Cell, 1982, 29, 245-255. 102. C. W. Greider and E. H. Blackburn, Cell, 1985, 43, 405-413. 103. C. W. Greider and E. H. Blackburn, Nature, 1989, 337, 331-337. 104. J. A. Londono-Vallejo, Biochimie, 2008, 90, 73-82. 105. R. K. Moyzis, J. M. Buckingham, L. S. Cram, M. Dani, L. L. Deaven, M. D. Jones, J. Meyne, R. L. Ratliff, and J. R. Wu, Proc. Natl. Acad. Sci. USA, 1988, 85, 6622-6626. 106. S. Neidle and G. Parkinson, Nat. Rev. Drug Discovery, 2002, 1, 383-393. 107. J. W. Shay and W. E. Wright, Nat. Rev. Mol. Cell. Biol., 2000, 1, 72-76. 108. J. W. Shay and W. E. Wright, FEBS Letters, 2010, 584, 3819-3825. 109. T. M. Bryan and T. R. Cech, Curr. Opin. Cell Biol., 1999, 11, 318-324. 110. J. Feng, W. D. Funk, S. S. Wang, S. L. Weinrich, A. A. Avilion, C. P. Chiu, R. R. Adams, E. Chang, R. C. Allsopp, and J. Yu, Science, 1995, 269, 1236-1241. References 251 111. J.-L. Chen, M. A. Blasco, and C. W. Greider, Cell, 2000, 100, 503-514. 112. K. Yashima, A. Maitra, B. B. Rogers, C. F. Timmons, A. Rathi, H. Pinar, W. E. Wright, J. W. Shay, and A. F. Gazdar, Cell Growth Differ., 1998, 9, 805-813. 113. W. E. Wright, M. A. Piatyszek, W. E. Rainey, W. Byrd, and J. W. Shay, Dev. Genet., 1996, 18, 173-179. 114. N. R. Forsyth, W. E. Wright, and J. W. Shay, Differentiation, 2002, 69, 188-197. 115. N. W. Kim, M. A. Piatyszek, K. R. Prowse, C. B. Harley, M. D. West, P. L. Ho, G. M. Coviello, W. E. Wright, S. L. Weinrich, and J. W. Shay, Science, 1994, 266, 2011-2015. 116. J. W. Shay and W. E. Wright, Nat. Rev. Drug Discovery, 2006, 5, 577-584. 117. T. de Lange, Genes Dev., 2005, 19, 2100-2110. 118. W. E. Wright, V. M. Tesmer, K. E. Huffman, S. D. Levene, and J. W. Shay, Genes Dev., 1997, 11, 2801-2809. 119. R. McElligott and R. J. Wellinger, EMBO J., 1997, 16, 3705-3714. 120. K. E. Huffman, S. D. Levene, V. M. Tesmer, J. W. Shay, and W. E. Wright, J. Biol. Chem., 2000, 275, 19719-19722. 121. H. Takai, A. Smogorzewska, and T. D. Lange, Curr. Biol., 2003, 13, 1549-1556. 122. F. d’Adda di Fagagna, P. M. Reaper, L. Clay-Farrace, H. Fiegler, P. Carr, T. von Zglinicki, G. Saretzki, N. P. Carter, and S. P. Jackson, Nature, 2003, 426, 194-198. 123. Z. Lou and J. Chen, Exp. Cell Res., 2006, 312, 2641-2646. 124. E. H. Blackburn, Nature, 2000, 408, 53-56. 125. D. Hockemeyer, J.-P. Daniels, H. Takai, and T. de Lange, Cell, 2006, 126, 63-77. 126. S. J. Froelich-Ammon, B. A. Dickinson, J. M. Bevilacqua, S. C. Schultz, and T. R. Cech, Genes & Development, 1998, 12, 1504-1514. 127. M. Gellert, M. N. Lipsett, and D. R. Davies, Proc. Natl. Acad. Sci. USA, 1962, 48, 2013-2018. 128. S. M. Haider, G. N. Parkinson, and S. Neidle, J. Mol. Biol., 2003, 326, 117-125. 129. DeLanoScientific, The PyMOL Molecular Graphics System, 2002, Version 0.99. 130. J. T. Davis, Angew. Chem., Int. Ed., 2004, 43, 668 - 698. References 252 131. T. J. Pinnavaia, C. L. Marshall, C. M. Mettler, C. L. Fisk, T. Miles, and E. D. Becker, J. Am. Chem. Soc., 1978, 100, 3625-3627. 132. P. Tougard, J.-F. Chantot, and W. Guschlbauer, Biochim. Biophys. Acta, Nucleic Acids Protein Synth., 1973, 308, 9-16. 133. S. B. Zimmerman, G. H. Cohen, and D. R. Davies, J. Mol. Biol., 1975, 92, 181-192. 134. Z. B. Steven, J. Mol. Biol., 1976, 106, 663-672. 135. T.-M. Ou, Y.-J. Lu, J.-H. Tan, Z.-S. Huang, and K.-Y. Wong, ChemMedChem, 2008, 3, 690-713. 136. E. Henderson, C. C. Hardin, S. K. Walk, I. Tinoco Jr., and E. H. Blackburn, Cell, 1987, 51, 899-908. 137. J. R. Williamson, M. K. Raghuraman, and T. R. Cech, Cell, 1989, 59, 871-880. 138. S. Burge, G. N. Parkinson, P. Hazel, A. K. Todd, and S. Neidle, Nucleic Acids Res., 2006, 34, 5402-5415. 139. G. N. Parkinson, in Quadruplex Nucleic Acids, eds. S. Neidle and S. Balasubramanian, RSC Publishing, 1st edn., 2006, pp. 17-19. 140. S. Neidle and G. N. Parkinson, Biochemie, 2008, 90, 1184-1196. 141. S. Haider, G. N. Parkinson, and S. Neidle, J. Mol. Biol., 2002, 320, 189-200. 142. F. W. Smith and J. Feigon, Nature, 1992, 356, 164-168. 143. P. Schultze, N. V. Hud, F. W. Smith, and J. Feigon, Nucleic Acids Res., 1999, 27, 3018-3028. 144. M. Crnugelj, P. Sket, and J. Plavec, J. Am. Chem. Soc., 2003, 125, 7866-7871. 145. P. Hazel, G. N. Parkinson, and S. Neidle, J. Am. Chem. Soc., 2006, 128, 5480-5487. 146. R. Rodriguez, K. M. Miller, J. V. Forment, C. R. Bradshaw, M. Nikan, S. Britton, T. Oelschlaegel, B. Xhemalce, S. Balasubramanian, and S. P. Jackson, Nat. Chem. Biol., 2012, 8, 301-310. 147. D. Sen and W. Gilbert, Nature, 1990, 334, 410-414. 148. Y. Wang and D. J. Patel, Structure, 1993, 1, 263-282. 149. G. N. Parkinson, M. P. H. Lee, and S. Neidle, Nature, 2002, 417, 876-880. References 253 150. A. M. Zahler, J. R. Williamson, T. R. Cech, and D. M. Prescott, Nature, 1991, 350, 718-720. 151. S. Neidle, FEBS Journal, 2010, 277, 1118-1125. 152. P. Murat, Y. Singh, and E. Defrancq, Chem. Soc. Rev., 2011, 40, 5293-5307. 153. A. Ambrus, D. Chen, J. Dai, T. Bialis, R. A. Jones, and D. Yang, Nucleic Acids Res., 2006, 34, 2723-2735. 154. C. C. Hardin, T. Watson, M. Corregan, and C. Bailey, Biochemistry, 1992, 31, 833-841. 155. P. K. Patel and R. V. Hosur, Nucleic Acids Res., 1999, 27, 2457-2464. 156. P. K. Patel, A. S. R. Koti, and R. V. Hosur, Nucleic Acids Res., 1999, 27, 3836-3843. 157. N. H. Campbell and G. N. Parkinson, Methods, 2007, 43, 252-263. 158. J.-L. Mergny, L. Lacroix, C. Hounsou, L. Guittat, M. Hoarau, P. B. Arimondo, J.-P. Vigneron, J.-M. Lehn, J.-F. Roui, T. Garestier, and C. Helene, Proc. Natl. Acad. Sci. USA, 2001, 98, 3062-3067. 159. A. D. Cian, L. Guittat, M. Kaiser, B. Saccà, S. Amrane, A. Bourdoncle, P. Alberti, M.-P. Teulade-Fichou, L. Lacroix, and J.-L. Mergny, Methods, 2007, 42, 183-195. 160. B.-S. Herbert, A. E. Hochreiter, W. E. Wright, and J. W. Shay, Nat. Protoc., 2006, 1, 1583-1590. 161. D. Monchaud and M.-P. Teulade-Fichou, Org. Biomol. Chem., 2008, 6, 627-636. 162. K. Shin-Ya, K. Wierzba, K.-I. Matsuo, T. Ohtani, Y. Yamada, K. Furihata, Y. Hayakawa, and H. Seto, J. Am. Chem. Soc., 2001, 123, 1262-1263. 163. M.-Y. Kim, H. Vankayalapati, K. Shin-Ya, K. Wierzba, and L. H. Hurley, J. Am. Chem. Soc., 2002, 124, 2098-2099. 164. T. Doi, M. Yoshida, K. Shin-ya, and T. Takahashi, Org. Lett., 2006, 8, 4165-4167. 165. C. M. Barbieri, A. R. Srinivasan, S. G. Rzuczek, J. E. Rice, E. J. Lavoie, and D. S. Pilch, Nucleic Acids Res., 2007, 35, 3272-3286. 166. M. Tera, Y. Sohtome, H. Ishizuka, T. Doi, M. Takagi, K. Shin-ya, and K. Nagasawa, Heterocycles, 2006, 69, 505-514. 167. R. T. Wheelhouse, D. Sun, H. Han, F. X. Han, and L. H. Hurley, J. Am. Chem. Soc., 1998, 120, 3261-3262. References 254 168. N. V. Anantha, M. Azam, and R. D. Sheardy, Biochemistry, 1998, 37, 2709-2714. 169. F. X. Han, R. T. Wheelhouse, and L. H. Hurley, J. Am. Chem. Soc., 1999, 121, 3561-3570. 170. G. N. Parkinson, R. Ghosh, and S. Neidle, Biochemistry, 2007, 46, 2390-2397. 171. J. Ren and J. B. Chaires, Biochemistry, 1999, 38, 16067-16075. 172. M.-P. Teulade-Fichou, C. Carrasco, L. Guittat, C. Bailly, P. Alberti, J.-L. Mergny, A. David, J.-M. Lehn, and W. D. Wilson, J. Am. Chem. Soc., 2003, 125, 4732-4740. 173. S. Neidle and M. A. Read, Biopolymers, 2001, 56, 195-208. 174. R. J. Harrison, A. P. Reszka, S. M. Haider, B. Romagnoli, J. Morrell, M. A. Read, S. M. Gowan, C. M. Incles, L. R. Kelland, and S. Neidle, Bioorg. Med. Chem. Lett., 2004, 14, 5845-5849. 175. M. A. Read, A. A. Wood, J. R. Harrison, S. M. Gowan, L. R. Kelland, H. S. Dosanjh, and S. Neidle, J. Med. Chem., 1999, 42, 4538-4546. 176. R. J. Harrison, S. M. Gowan, L. R. Kelland, and S. Neidle, Bioorg. Med Chem. Lett., 1999, 9, 2463-2468. 177. M. Read, R. J. Harrison, B. Romagnoli, F. A. Tanious, S. M. Gowan, A. P. Reszka, W. D. Wilson, L. R. Kelland, and S. Neidle, Proc. Natl. Acad. Sci. USA, 2001, 98, 4844-4849. 178. R. J. Harrison, J. Cuesta, G. Chessari, M. A. Read, S. K. Basra, A. P. Reszka, J. Morrell, S. M. Gowan, C. M. Incles, F. A. Tanious, W. D. Wilson, L. R. Kelland, and S. Neidle, J. Med. Chem., 2003, 46, 4463-4476. 179. C. M. Incles, C. M. Schultes, H. Kempski, H. Koehler, L. R. Kelland, and S. Neidle, Mol. Cancer Ther., 2004, 3, 1201-1206. 180. A. M. Burger, F. Dai, C. M. Schultes, A. P. Reszka, M. J. Moore, J. A. Double, and S. Neidle, Cancer Res., 2005, 65, 1489-1496. 181. S. M. Gowan, J. R. Harrison, L. Patterson, M. Valenti, M. A. Read, S. Neidle, and L. R. Kelland, Mol. Pharmacol., 2002, 61, 1154-1162. 182. N. H. Campbell, M. Patel, A. B. Tofa, R. Ghosh, G. N. Parkinson, and S. Neidle, Biochemistry, 2009, 48, 1675-1680. 183. N. H. Campbell, G. N. Parkinson, A. P. Reszka, and S. Neidle, J. Am. Chem. Soc., 2008, 130, 6722-6724. References 255 184. M. J. B. Moore, C. M. Schultes, J. Cuesta, F. Cuenca, M. Gunaratnam, F. A. Tanious, W. D. Wilson, and S. Neidle, J. Med. Chem., 2006, 49, 582-599. 185. C. L. Perrin and J. B. Nielson, Annu. Rev. Phys. Chem., 1997, 48, 511-544. 186. C. Thibaudeau, J. Plavec, and J. Chattopadhyaya, J. Org. Chem., 1998, 63, 4967-4984. 187. P. Emsley and K. Cowtan, Acta Crystallogr., Sect. D: Biol. Crystallogr., 2004, 60, 2126-2132. 188. C. Ye and J. M. Shreeve, J. Fluorine Chem., 2004, 125, 1869-1872. 189. A. Cheguillaume, S. Lacroix, and J. Marchand-Brynaert, Tetrahedron Lett., 2003, 44, 2375-2377. 190. P. E. Floreancig, S. E. Swalley, J. W. Trauger, and P. B. Dervan, J. Am. Chem. Soc., 2000, 122, 6342-6350. 191. S. Neidle, R. J. Harrison, L. Kelland, S. M. Gowan, M. A. Read, and A. Reszka, 2003, US 2003/0207909 A1. 192. K. Matsumura, J. Am. Chem. Soc., 1929, 51, 816-820. 193. S. A. Gamage, N. Tepsiri, P. Wilairat, S. J. Wojcik, D. P. Figgitt, R. K. Ralph, and W. A. Denny, J. Med. Chem., 1994, 37, 1486-1494. 194. C. A. Faler and M. M. Joullie, Tetrahedron, 2006, 47, 7229-7231. 195. K.-W. Yang, J. G. Cannon, and J. G. Rose, Tetrahedron Lett., 1970, 21, 1791-1794. 196. B. Duthion, D. G. Pardo, and J. Cossy, Org. Lett., 2010, 12, 4620-4623. 197. P. G. M. Wuts and T. W. Greene, in Greene’s Protective Groups in Organic Synthesis, John Wiley & Sons, Inc., 2006, p. 4th Edition. 198. Q. Yang, X.-Y. Li, H. Wu, and W.-J. Xiao, Tetrahedron, 2006, 47, 3893-3896. 199. Q. Yang, W.-J. Xiao, and Z. Yu, Org. Lett., 2005, 7, 871-874. 200. O. Songis, A. M. Z. Slawin, and C. S. J. Cazin, Chem. Commun., 2012, 48, 1266-1268. 201. C. Enkisch and C. Schneider, Eur. J. Org. Chem., 2009, 32, 5549-5564. 202. A. F. Abdel-Magid, K. G. Carson, B. D. Harris, C. A. Maryanoff, and R. D. Shah, J. Org. Chem., 1996, 61, 3849-3862. References 256 203. A. J. Phillips, Y. Uto, P. Wipf, M. J. Reno, and D. R. Williams, Org. Lett., 2000, 2, 1165-1168. 204. J. R. Dunetz, Y. Xiang, A. Baldwin, and J. Ringling, Org. Lett., 2011, 13, 5048-5051. 205. B. Kuhn, P. Mohr, and M. Stahl, J. Med. Chem., 2010, 53, 2601-2611. 206. S. T. Cheung and N. L. Benoiton, Can. J. Chem., 1977, 55, 906 -910. 207. L. Aurelio, R. T. C. Brownlee, and A. B. Hughes, Chem. Rev., 2004, 104, 5823-5846. 208. J. Coste, E. Frérot, P. Jouin, and B. Castro, Tetrahedron Lett., 1991, 32, 1967-1970. 209. V. T. Pietzonka and D. Seebach, Angew. Chem., Int. Ed., 1992, 31, 1481-1482. 210. R. Schwesinger, H. Schlemper, C. Hasenfratz, J. Willaredt, T. Dambacher, T. Breuer, C. Ottaway, M. Fletschinger, J. Boele, H. Fritz, D. Putzas, H. W. Rotter, F. G. Bordwell, A. V. Satish, G.-Z. Ji, E.-M. Peters, K. Peters, H. G. V. Schnering, and L. Walz, Liebigs. Ann., 1996, 1055-1081. 211. S. Escoubet, S. Gastaldi, and M. Bertrand, Eur. J. Org. Chem., 27, 2005, 3855-3873. 212. M. Schelhaas and H. Waldmann, Angew. Chem., Int. Ed., 1996, 35, 2056-2083. 213. F. Guibé, Tetrahedron, 1998, 54, 2967-3042. 214. N. Ohmura, A. Nakamura, A. Hamasaki, and M. Tokunaga, Eur. J. Org. Chem., 2008, 30, 5042-5045. 215. B. Alcaide, P. Almendros, and J. M. Alonso, Chem. Eur. J., 2006, 12, 2874-2879. 216. K. Kajihara, M. Arisawa, and S. Shuto, J. Org. Chem., 2008, 73, 9494-9496. 217. M. J. Zacuto and F. Xu, J. Org. Chem., 2007, 72, 6298-6300. 218. D. P. Spalding, G. W. Moersch, H. S. Mosher, and F. C. Whitmore, J. Am. Chem. Soc., 1964, 68, 1596-1598. 219. A. Klapars, X. Huang, and S. L. Buchwald, J. Am. Chem. Soc., 2002, 124, 7421-7428. 220. D. P. Phillips, X.-F. Zhu, T. L. Lau, X. He, K. Yang, and H. Liu, Tetrahedron Lett., 2009, 50, 7293-7296. References 257 221. D. H. Paull, M. T. Scerba, E. Alden-Danforth, L. R. Widger, and T. Lectka, J Am. Chem. Soc., 2008, 130, 17260-17261. 222. D. P. Galonic, N. D. Ide, W. A. V. D. Donk, and D. Y. Gin, J. Am. Chem. Soc., 2005, 127, 7359-7369. 223. W. L. F. Armarego and C. L. L. Chai, Purification of Laboratory Chemicals, Butterworth-Heinemann, Oxford, 6th edn., 2009. 224. W. C. Still, M. Kahn, and A. Mitra, J. Org. Chem., 1978, 43, 2923-2925. 225. B. M. Trost and M. T. Rudd, Org. Lett., 2003, 5, 4599-4602. 226. A. N. Hulme, C. H. Montgomery, and D. K. Henderson, J. Chem. Soc., Perkin Trans. 1, 2000, 1837-1841. 227. D. Gani, P. B. Hitchcock, and D. W. Young, J. Chem. Soc., Perkin Trans. 1, 1985, 1363-1372. 228. G. P. Deniau, Ph.D., University of St Andrews, Scotland, 2007. 229. N. S. Chandrakumar, P. K. Yonan, A. Stapelfeld, M. Savage, E. Rorbacher, P. C. Contreras, and D. Hammond, J. Med. Chem., 1992, 35, 223-233. 230. K. Matsumura, J. Am. Chem. Soc., 1929, 51, 816-820. 231. A. W. McConnaughie and T. C. Jenkins, J. Med. Chem., 1995, 38, 3488-3501. 232. Y. Tong, Y. M. Fobian, M. Wu, N. D. Boyd, and K. D. Moeller, J. Org. Chem., 2000, 65, 2484-2493. 233. J. H. Cho and B. M. Kim, Tetrahedron Lett., 2002, 43, 1273-1276. Appendix 258 Appendix 1.1- Crystallographic information for (R,R)-144 with O.nova telomeric DNA PDB ID - 3NYP* Data collection Total number of reflections collected 85447 Number of unique reflections 23982 Space Group P21212 Cell dimensions: a,b and c (Å) 57.80, 44.46, 28.14 Angle (º) α, β, γ 90.00, 90.00, 90.00 Maximum resolution (Å) 1.18 Rmerge 0.051 I/σI 15.5 I/σ (highest resolution shell) 5.7 Completeness (%) 97.7 Redundancy 3.6 Refinement Resolution range used in refinement (Å) 21.99-1.18 Number of unique reflections used in refinement 23275 Completeness (%) 94.6 Rfactor(%) 16.6 Rfree(%) 18.8 Number of G-quadruplexes/asymmetric unit 1 Number of ligands/asymmetric unit 1 Number of asymmetric units per unit cell 4 Number of atoms DNA 506 Ligand 36 Potassium ions 4 Water 177 Appendix 259 Appendix 1.2 - Crystallographic information for (S,S)-144 with O. nova telomeric DNA PDB ID - 3NZ7 ** Data collection Total number of reflections collected 248167 Number of unique reflections 25994 Space Group P21212 Cell dimensions: a, b and c (Å) 55.57, 42.70, 27.00 Angle (º) α, β, γ 90.00, 90.00, 90.00 Maximum resolution (Å) 1.1 Rmerge 0.073 I/σI 25.4 I/σ (highest resolution shell) 6.8 Completeness (%) 97.1 Redundancy 5 Refinement Resolution range used in refinement (Å) 7.80 - 1.10 Number of unique reflections used in refinement 25901 Completeness (%) 97 Rfactor(%) 13.8 Rfree(%) 15.9 Number of G-quadruplexes/asymmetric unit 1 Number of ligands/asymmetric unit 1 Number of asymmetric units/unit cell 4 Number of atoms DNA 506 Ligand 36 Potassium ions 4 Water 188 Appendix 260 Appendix 1.3 - Crystallographic information for 234 A. Crystal Data dsdh4 Empirical Formula C36H44N4O6 Formula Weight 628.77 Crystal Color, Habit colorless, prism Crystal Dimensions 0.200 × 0.020 × 0.020 mm Crystal System orthorhombic Lattice Type Primitive No. of Reflections Used for Unit Cell Determination (2θ range) 6233 (81.8 - 139.0 °) Lattice Parameters a = 16.681(5) Å b = 17.399(6) Å c = 24.279(8) Å V = 7047(4) Å3 Space Group P212121(#19) Z value 8 Dcalc 1.185 g/cm3 F000 2688.00 μ(CuKα) 6.576 cm-1 B. Intensity Measurements Diffractometer Radiation CuKα (λ = 1.54187 Å) multi-layer mirror monochromated Take-off Angle 2.8 ° Detector Aperture 2.0 - 2.5 mm horizontal, 2.0 mm vertical Crystal to Detector Distance 21 mm Voltage, Current 40kV, 20mA Temperature -100.0 °C Scan Type ω-2θ 2θmax 137.0 ° No. of Reflections Measured Total: 73332, Unique: 12756 (Rint = 0.1119) Friedel pairs: 5732 Corrections Lorentz-polarization Absorption (trans. factors: 0.472 - 0.987) Appendix 261 C. Structure Solution and Refinement Structure Solution Direct Methods Refinement Full-matrix least-squares on F2 Function Minimized Σ w (Fo2 - Fc2)2 Least Squares Weights w = 1/ [ σ2(Fo2) + (0.1213 . P)2 + 0.1830 . P ] where P = (Max(Fo2,0) + 2Fc2)/3 2θmax cutoff 137.0° Anomalous Dispersion All non-hydrogen atoms No. Observations (All reflections) 12756 No. Variables 845 Reflection/Parameter Ratio 15.10 Residuals: R1 (I>2.00σ(I)) 0.0707 Residuals: R (All reflections) 0.0761 Residuals: wR2 (All reflections) 0.1914 Goodness of Fit Indicator 1.059 Flack Parameter (Friedel pairs = 5732) 0.14(17) Max Shift/Error in Final Cycle 0.001 Maximum peak in Final Diff. Map 0.69 e‑/Å3 Minimum peak in Final Diff. Map -0.38 e‑/Å3 ! Append ix! ! ! !262 Appen dix 1.4 - Sele cted N MR 1.4.1 - 1 H NM R of (S ,S)-196 .HCl 3.03.54.04.55.05.56.06.57.07.5 ppm 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NH3ClOClH3N F F ! Append ix! ! ! ! 263 1.4.2 - 19 F NM R of (S ,S)-196 .HCl −220−200−180−160−140−120−100−80−60−40−200 ppm −195.2 −195.4 −195.6 ppm 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NH3ClOClH3N F F ! Append ix! ! ! !264 1.4.3 - 1 H NM R of (S ,S)-206 2.53.03.54.04.55.05.56.06.57.07.58.08.5 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' N FF ! Append ix! ! ! ! 265 1.4.4 - 19 F NM R of (S ,S)-206 −130 −135 −140 −145 −150 −155 −160 −165 −170 −175 −180 −185 −190 −195 −200 ppm −191.2 −191.4 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' N FF ! Append ix! ! ! !266 1.4.5 - 13 C NM R of (S ,S)-206 220 200 180 160 140 120 100 80 60 40 20 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' N FF ! Append ix! ! ! ! 267 1.4.6 - 1 H NMR (S,S)-2 08.HCl 2.53.03.54.04.55.05.56.06.57.07.58.08.5 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NH3ClOClH3N 13 141514'13' N FF ! Append ix! ! ! !268 1.4.7 - 1 H NMR (S,S)-2 08.HCl −185 −190 −195 −200 −205 −210 −215 −220 −225 −230 −235 −240 −245 −250 −255 ppm −195.6 −195.7 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NH3ClOClH3N 13 141514'13' N FF ! Append ix! ! ! ! 269 1.4.8 - 1 H NM R of (S ,S)-210 10 9 8 7 6 5 4 3 2 1 ppm 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON F F ! Append ix! ! ! !270 1.4.9 - 19 F NM R of (S ,S)-210 −180−160−140−120−100−80−60−40−20 ppm −190.0 −190.5 ppm 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON F F ! Append ix! ! ! ! 271 1.4.10 - 13 C N MR of (S,S)-2 10 220 200 180 160 140 120 100 80 60 40 20 ppm 23 4 4a9a1 NH10 4b8a9 5 678NH NHO O NON F F ! Append ix! ! ! !272 1.4.11 - 1 H NM R of (S ,S)-212 9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' NF F ! Append ix! ! ! ! 273 1.4.12 - 19 F N MR of (S,S)-2 12 −185 −190 −195 −200 −205 −210 −215 −220 −225 −230 −235 −240 −245 −250 −255 ppm −191.0 −191.2 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' NF F ! Append ix! ! ! !274 1.4.13 - 13 C NM R of (S ,S)-212 190 180 170 160 150 140 130 120 110 100 90 80 70 60 50 40 30 20 10 ppm 23 4 4a9a1 N10 4b8a9 5 678NH NHHN12 O NON 13 141514'13' NF F ! Append ix! ! ! ! 275 1.4.14 - 1 H NM R of (S )-229 2.53.03.54.04.55.05.56.06.57.07.58.08.5 ppm OONOSi ! Append ix! ! ! !276 1.4.15 - 13 C NM R of (S )-229 220 200 180 160 140 120 100 80 60 40 20 ppm OONOSi ! Append ix! ! ! ! 277 1.4.16 - 1 H NM R 224c 9 8 7 6 5 4 3 2 1 0 ppm α NHO α+1OONHO ! Append ix! ! ! !278 1.4.17 - 13 C NM R of 22 4c 220 200 180 160 140 120 100 80 60 40 20 ppm α NHO α+1OONHO ! Append ix! ! ! ! 279 1.4.18 - 1 H NM R of 22 7c 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm α NHO α+1OOFN ! Append ix! ! ! !280 1.4.19 - 19 F NM R of 22 7c −240−220−200−180−160−140−120−100−80−60−40−200 ppm −190.4 −190.6 −190.8 ppm α NHO α+1OOFN ! Append ix! ! ! ! 281 1.4.20 - 13 C N MR of 227c 220 200 180 160 140 120 100 80 60 40 20 ppm α NHO α+1OOFN ! Append ix! ! ! !282 1.4.21 - 1 H NM R of 22 8c 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm α NHO α+1OONF ! Append ix! ! ! ! 283 1.4.22 - 19 F N MR of 228c −240−220−200−180−160−140−120−100−80−60−40−200 ppm −227.4 −227.5 −227.6 −227.7 ppm α NHO α+1OONF ! Append ix! ! ! !284 1.4.23 - 13 C NM R of 22 8c 220 200 180 160 140 120 100 80 60 40 20 ppm α NHO α+1OONF ! Append ix! ! ! ! 285 1.4.24 - 1 H NM R of 24 8 2.53.03.54.04.55.05.56.06.57.07.58.08.5 ppm α NO α+1PhOOFN ! Append ix! ! ! !286 1.4.25 - 19 F NM R of 24 8 −175 −180 −185 −190 −195 −200 −205 −210 −215 −220 −225 −230 −235 −240 ppm −185 −186 −187 ppm α NO α+1PhOOFN ! Append ix! ! ! ! 287 1.4.26 - 13 C N MR of 248 220 200 180 160 140 120 100 80 60 40 20 ppm α NO α+1PhOOFN