Citation
Important Interactions in Peptide-Based Transfection Agents, Sugars as Chiral Scaffolds, and Molecular Imprinting of Nerve Gases

Material Information

Title:
Important Interactions in Peptide-Based Transfection Agents, Sugars as Chiral Scaffolds, and Molecular Imprinting of Nerve Gases
Creator:
ALLAIS, FLORENT ( Author, Primary )
Copyright Date:
2008

Subjects

Subjects / Keywords:
Bromides ( jstor )
Chlorides ( jstor )
Esters ( jstor )
Lewis acids ( jstor )
Molecules ( jstor )
Nerves ( jstor )
Polymerization ( jstor )
Polymers ( jstor )
Sensors ( jstor )
Solvents ( jstor )

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright Florent Allais. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Embargo Date:
4/30/2005
Resource Identifier:
436097543 ( OCLC )

Downloads

This item is only available as the following downloads:


Full Text

PAGE 1

IMPORTANT INTERACTIONS IN PEPTIDE-BASED TRANSFECTION AGENTS, SUGARS AS CHIRAL SC AFFOLDS, AND MOLECULAR IMPRINTING OF NERVE GASES By FLORENT ALLAIS A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLOR IDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2004

PAGE 2

Copyright 2004 by Florent Allais

PAGE 3

This is dedicated to my parents Nadine and J acques. This is for Stanislas, especially, who inspires me daily.

PAGE 4

ACKNOWLEDGMENTS First and foremost I would like to thank my parents and my grandmother Maria. Their love, support, and understanding are a continuous source of strength. They have always continued to support me in whatever endeavor I choose and I will forever be grateful. I would also like to thank my great-grandmother Alberte, who taught me a lot while I was a child; and my grandfather Georges, whom I didn’t have the joy to really know. I am sure they would be really proud to see the individual I became and where I am now. There are other important persons in my life: Stanislas, the greatest man I could have ever found; my family (although some of them did or still behave in a way that disgusts and upsets me); Nathan, (though I haven’t known him for a long time, I know that we will be friends forever); and finally Xavier, who is someone I won’t forget, in spite of huge misunderstandings about life and relationships. Next I would like to thank Pr. Eric Enholm for being a great advisor, but more importantly an excellent teacher. I valued the environment he created in his lab and the many discussions on topics at hand. He made me feel at home when I came to his lab the first time, 5 years ago. Having such a warm and friendly welcome helped me make up my mind when it came to choosing a place for my Ph.D. He cares about his students, and I thank him for his support and guidance over the years. I would also like to thank Dr. Randy Duran, the one who made my dream come true. For years I wanted to come to the USA for my education. Thanks to the REU program, I had the opportunity to spend Summer at the University of Florida and meet amazing and interesting people, and a iv

PAGE 5

great laboratory. I will always be grateful to the REU program. I would like to thank Dr. Merle Battiste, Dr. Dolbier, Dr. Randy Duran and Dr. Ken Sloan for serving on my committee for their help and sharing of their wisdom over the years; and also for dealing with my terrible French accent (oops, sorry, Freedom accent). Dr. Kenneth Sloan is a great committee member and we should all be so lucky. I would like to thank the Enholm Group members past and present. Jennifer Cottone, Maria Gallagher, Ashwin Bharadwaj, Sophie Klein, Aarti Joshi, Jed Hastings, Emilie Claudel, Rayanne Mohammed, Christophe Blaszykowski, and Sbastian Bareyt made the time spent in the lab a joy, with loud music, constant joking (about Indians, Americans and French), and dance moves. Sometimes we argued, but I don’t know any workplace where everyone gets along 24/7! Special thanks go to Jennifer for training me and taking me under her wing upon arrival. Ryan Martin is a great undergraduate to work with, and I wish him nothing but the best. Despite profound disagreements on specific topics, I enjoyed working with him. He’s a great individual and deserves to succeed. Thanks also go to the many other students in the department who enriched my experience. Thanks go to the “French Mafia”: Chaya Pooput, Christophe Grenier, Thomas Joncheray, Emilie Galand, Daniel Serra, Sophie Bernard, Thomas Cardolaccia, and Florence Courchay. I will miss our parties and delicious lunches/dinners. They were great occasions to cheer us up when home sickness struck. I also want to thank Maria Estrella, Lynn Usher, Ravi and many others for their friendship and valuable discussion. There is chemistry and then there is the home side. After leaving work, it can be difficult to keep work at work. Without these people, I would have had no life outside the work. Thus I would like to thank Chaya Pooput and Sophie Klein for being great v

PAGE 6

roommates and persons who know what is truly important. All the people named above are not just friends but part of my family and I will miss them. We had many good times and I will miss them. I also want to give special thanks for having been awarded the MacLaughlin Dissertation Fellowship to finance my last semester here at the University of Florida. I really appreciated their recognition of my work and career. I would like to thank the “French faculty”: Pr. Yannick Landais, who taught me rigour and precision; Catherine Grosdemange-Billard, who is an excellent professor and also a great friend who helped (and still helps) me construct my future and career; and the REU coordinators Dr. Anne-Lise Dhimane. Dr. Yves Gnanou. Finally I thank Pr. Janine Cossy for welcoming me into her research group as a postdoctoral fellow. I sincerely hope I’ll fulfill her expectations. vi

PAGE 7

TABLE OF CONTENTS page ACKNOWLEDGMENTS .................................................................................................iv LIST OF TABLES .............................................................................................................ix LIST OF FIGURES .............................................................................................................x ABSTRACT ......................................................................................................................xii CHAPTER 1 HISTORICAL BACKGROUND.................................................................................1 History of Free-Radicals...............................................................................................1 Generation of Radicals.................................................................................................2 Free-Radical Cyclizations.............................................................................................4 5-Hexenyl Radical Cyclizations...................................................................................6 Chiral Templates for Free-Radical Chemistry..............................................................8 DNA Transfection......................................................................................................12 RGD Analogues..........................................................................................................13 History of Molecular Imprinting................................................................................19 Methodology and Principles of Molecular Imprinting...............................................20 Fluorescence and Fluorescent Molecules in MIP.......................................................22 Nerve Gases................................................................................................................24 2 ALKENE-ALDEHYDE CYCLIZATION VIA O-KETYL RADICAL USING LEWIS ACIDS AND SUGARS AS AUXILIARIES................................................28 Introduction.................................................................................................................28 Preliminary Study.......................................................................................................30 Synthesis of the Chiral Precursors..............................................................................31 Results and Discussion...............................................................................................34 Conclusion..................................................................................................................38 3 MOLECULAR IMPRINTING WITH ROMP METHODOLOGY: EFFECTS ON POLYMERIZATION AND IMPRINTED PROPERTIES........................................39 Introduction.................................................................................................................39 Comparison of ROMP versus Radical Induced Polymerization................................39 vii

PAGE 8

Synthesis of the MIPs.................................................................................................40 Templates....................................................................................................................40 Radical Polymerization...............................................................................................41 ROMP Polymerization................................................................................................42 Synthesis of the Monomers........................................................................................45 Analytical Method, Results and Discussion...............................................................47 Control Experiments...................................................................................................54 Using MIP Technology in the Synthesis of Selective Nerve Gas Detectors..............55 Synthesis of the Sensors.............................................................................................58 Analytical Studies.......................................................................................................63 Conclusion..................................................................................................................63 4 NEW SYNTHETIC VECTORS FOR DNA TRANSFECTION BASED ON A PROLINE MOIETY...................................................................................................65 Introduction.................................................................................................................65 Synthesis of the Proline Moiety..................................................................................68 Synthesis of the Tartrate Derivative...........................................................................69 Convergent Synthesis: Formation of the Acetal.........................................................71 Solid-Phase Synthesis of the Cyclic Tetrapeptide......................................................72 Solution-Phase Synthesis of the Tetrapeptide............................................................74 Synthesis of the DNA Transfection Agent Precursor.................................................77 Conclusion..................................................................................................................78 5 EXPERIMENTAL METHODS.................................................................................79 APPENDIX A SPECTRAL DATA..................................................................................................114 B HPLC TRACES........................................................................................................128 LIST OF REFERENCES.................................................................................................132 BIOGRAPHICAL SKETCH...........................................................................................138 viii

PAGE 9

LIST OF TABLES Table page 2-1 Diastereoselectivities achieved with the precursor..................................................31 2-2 Diastereoselectivities achieved with isosorbide.......................................................35 2-3 Diastereoselectivities achieved with isomannide.....................................................36 3-1 Control experiments.................................................................................................55 ix

PAGE 10

LIST OF FIGURES Figure page 1-1. Cyclization of an RGD analogue.............................................................................14 1-2. RGD recognition sequence.......................................................................................15 1-3. RGD binding to an integrin......................................................................................16 1-4. Kessler’s sugar-based cyclic tripeptide....................................................................17 1-5. RGD analogues........................................................................................................18 1-6. L-Proline based cyclic RGD....................................................................................18 1-7. Nerve gases..............................................................................................................26 3-1. Templates.................................................................................................................41 3-2. Grubbs ruthenium catalysts......................................................................................42 3-3. Radical polymerization with edrophonium chloride test 1....................................49 3-4. ROMP polymerization with edrophonium chloride test 1.....................................49 3-5. Radical polymerization with edrophonium chloride test 2....................................50 3-6. ROMP polymerization with edrophonium chloride test 2.....................................51 3-7. Radical polymerization with pyridostigmine bromide test 1.................................52 3-8. ROMP polymerization with pyridostigmine bromide test 1..................................52 3-9. Radical polymerization with pyridostigmine bromide test 2.................................53 3-10. ROMP polymerization with pyridostigmine bromide test 2..................................54 3-11. Nerve gases..............................................................................................................56 3-12. Detection device.......................................................................................................57 3-13. Fluorescent sensors..................................................................................................58 x

PAGE 11

4-1. Commonly used lipoplexes......................................................................................67 4-2. Lipoplex precursor...................................................................................................67 4-3. Synthetic pathway to the RGD-based lipoplex........................................................68 A-1. 1 H NMR of isosorbide precursor agent..................................................................115 A-2. 1 H NMR of isomannide precursor agent................................................................116 A-3. 1 H NMR of isosorbide cyclized adduct..................................................................117 A-4. 1 H NMR of the diester cross-linking agent............................................................118 A-5. 1 H NMR of the fluorescent detector.......................................................................119 A-6. 1 H NMR of 2-Bromo-4-methyl pyridine................................................................120 A-7. 1 H NMR of methyl-2-trimethylstannyl pyridine....................................................121 A-8. 1 H NMR of (8-Bromo-naphthalen-1-ylmethoxy)-trimethyl-silane........................122 A-9. 1 H NMR of Fmoc-keto-Proline..............................................................................123 A-10. 1 H NMR of Fmoc-keto-Proline methyl ester.........................................................124 A-11. 1 H NMR of oleyl bromide......................................................................................125 A-12. 1 H NMR of the dioleyl ether isopropylidene.........................................................126 A-13. 1 H NMR of the DNA transfection agent precursor................................................127 B-1. Cyclization of isosorbide at -78 o C in presence of ZnCl 2 .......................................129 B-2. Cyclization of isomannide at 80 o C without any Lewis acid..................................130 B-3. Cyclization of isomannide at -78 o C with MgBr 2 ...................................................131 xi

PAGE 12

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy IMPORTANT INTERACTIONS IN PEPTIDE-BASED TRANSFECTION AGENTS, SUGARS AS CHIRAL SCAFFOLDS, AND MOLECULAR IMPRINTING OF NERVE GASES By FLORENT ALLAIS May 2004 Chair: Eric J. Enholm Department of Chemistry Progress toward a library of new DNA transfection agents that feature the L-proline motif was examined. Lipid and aliphatic chains were grafted on a pseudo-L-proline core. Other investigations discussed include asymmetric free radical cyclizations using sugars as chiral auxiliaries; and the use of Ring Opening Metathesis Polymerization (ROMP) in a new method of polymerization of molecular imprinted polymers as well as the synthesis of new selective nerve gas detectors. DNA transfection agents are cationic molecules that possess lipidic chains. These chains allow the formation of supramolecular edifices and also hydrophobicity properties suitable for in vivo use. The cationic charges trap the DNA helixes (which are negatively charged) and protect the DNA from DNAses. Synthesis of a new proline based agent was undertaken. With this analogue in hand, biological testing is the next step. Later on, our agents were submitted to a solid-phase peptide synthesis to graft an RGD sequence on the proline core. RGD is a tripeptide recognition sequence that is xii

PAGE 13

composed of three amino acids (arginine, glycine, and aspartic acid). This sequence is recognized by receptors on the extracellular matrix that, in turn, are responsible for the adhesive properties of cells. Asymmetric free-radical cyclizations under the influence of a carbohydrate scaffold were also investigated. O-ketyl cyclization processes were studied under a number of conditions to find optimal conditions for stereochemical closure. With variation in temperature, solvent, and Lewis acid, a potentially useful synthetic process was shown. Finally, Molecular Imprinted Polymers (MIPs) were made by using ROMP methodology. This new methodology allowed the synthesis of highly selective MIPs while using milder conditions and simplifying work-up steps compared to the commonly used radical polymerization. Once the superiority of ROMP methodology was proven, the synthesis of new selective nerve-gas detectors was undertaken. Fluorescent sensors were synthesized, and their incorporation into MIPs is still in progress in our group. xiii

PAGE 14

CHAPTER 1 HISTORICAL BACKGROUND History of Free-Radicals Gomberg 1 identified the first radical in 1900 (in the form of a trivalent carbon compound, triphenylmethyl). However the first application of radical pathways in synthetic transformations did not appear until 1937, with the work of Hey and Waters 2 in the area of hemolytic phenylation of aromatic substrates; and the work of Karasch 3 in the regioselectivity of HBr addition to alkenes. An interesting note in cyclization reactions is that many aspects of radical chemistry were outlined in detail by Surzur 4 and Julia 5 in the 1960s, but were not applied to synthetic endeavors until the 1980s (by Stork, Curran, Enholm, and others). 6 Radical species are generated when a covalent bond is cleaved homolytically. Those intermediates are neutral in nature. Therefore solvation effects, racemization at adjacent centers, and unwanted synthetic transformations of labile functional groups are less pertinent issues with this type of chemistry. Although radicals are considered as neutral species, their reactivity is vast, making carbon-carbon free-radical bond formation a powerful tool in synthetic chemistry. Scheme 1-1 AD+AD Radical synthetic pathways can be used in various areas, such as cyclizations, alkylations, and atom transfer in an intramolecular as well as intermolecular fashion. A radical can be defined as a compound or more frequently an atom or functional group with an unpaired electron. 7 All of these species contain an odd number of electrons, and 1

PAGE 15

2 the simplest examples are atoms such as Br and H. These species are generally very reactive and quite unstable. The unpaired electrons combine to form a covalent bond (such as bromine and hydrogen radical combining to form hydrogen bromide). Radicals are generally formed by a homolysis (Scheme 1-1) of bonds in nonpolar solvents. Polar solvents tend to favor heterolytic bond cleavage, leaving charged species. Generation of Radicals One rule in generating radicals is that bonds must be sufficiently weak for homolysis to occur. Many sources of energy can lead to radicals. The most common sources are thermolysis, photolysis, ultraviolet, X-ray, and oxidation and reductive processes. Most of these methods require high temperatures (such as thermal initiators) or harsh conditions for metal-promoted oxidation. A typical initiating system is tributyl tin hydride (HSnBu 3 ) and AIBN (2,2-azobisbutyronitrile). This particular system is very common and is used in a wide array of organic transformations (such as reduction of halogens and cyclizations). 8-10 Tributyltin hydride (TBTH) is extremely popular, because the rates of initiation and propagation are well known. TBTH has a bond-dissociation energy of 74 kcal per mol. With this low bond dissociation energy, the tributyl tin radical is usually readily formed and stable. B r AIBN, SnBu3H Benzene1-11-2 Scheme 1-2 Another popular radical source similar to TBTH is tristrimethylsilyl silane, whose bond-dissociation energy is 79 kcal per mol. This is a viable alternative to TBTH. Termination, caused by the propagating radicals combining, can be minimized with lower

PAGE 16

3 concentrations of the propagating species. A general example of the reductive capability of tributyltin hydride is shown where a cyclic halide (1-1) can be reduced to the cyclopentane 1-2 (Scheme 1-2). Solvent choice in this transformation is generally high-boiling nonpolar solvents such as benzene and toluene. OOEtRR1BEtEtEtXREtR1RXRR1RR1RXRBEtEtOO+++ EtX++1-31-3 Scheme 1-3 Interest has grown, in recent years, in performing radical reactions in a stereospecific manner. Since most of the initiating systems require high temperatures for homolysis, stereochemical induction with radical chemistry was based on steric and electronic effects. Lack of an appropriate initiating system at lower temperatures hindered progress of stereoselective radical reactions; however, in the last 10 years new initiating systems have been developed to address this issue. 9, 11-12 Lowering the temperature tends to lock out reactive conformations. The most used of these systems is the triethyl borane oxygen system (Scheme 1-3). The mechanism starts by initiation with triplet oxygen ejecting an ethyl radical (1-3) from borane. The ethyl radical then abstracts

PAGE 17

4 a halogen. The propagating radical continues to do the chemistry desired. The mechanism shown, involves an atom-transfer process. With this system, a wider variety of solvents (such as methylene chloride, ether, and tetrahydrofuran) can be used. Free-Radical Cyclizations Radical cyclizations usually consist of intramolecular additions to double or triple bonds. Use of intramolecular radical closures in organic synthesis 6 began in the 1980s. Many mechanistic studies have shown that a wide range of monoand polycyclic products can be obtained with regioand stereoselectively (natural products, alkaloids, etc). 6 Because the conditions are neutral and therefore compatible with most functional and protecting groups, this free-radical transformation can be applied to many precursors. To illustrate that, very often a hydroxyl group doesn’t need to be protected before a radical transformation. The mechanism of radical cyclization involves a controlled chain reaction: initiation, propagation and termination, as previously described (Scheme 1-4). A site-specific radical is generated from an organic substrate by atom or group abstraction. The radical then reacts with tin hydride to generate the reduced product (or with itself to obtain a cyclized adduct). The transferability of various substrate X to the tin radical is generally in the order of I>Br>SePh = OC(S)SMe>Cl>SPh. 8, 9 Although steric interactions influence the cyclization energy barrier, highly complex molecules nonetheless can be constructed through radical additions.When radical closures occur on a double bond, the radical center is either external to the newly formed bond (exo) or resides internally in the generated cyclic structure (endo).

PAGE 18

5 Initiation: NCNNCNNCNCNCN2+.2.nBu3SnHnBu3Sn.+ Propagation: nBu3Sn.XnBu3SnX...++ Termination: .nBu3SnH Scheme 1-4 Of the two possible cyclizations, exo and endo, generally the former is preferred kinetically (scheme 1-5). 13 CCX.CC.C.Cexoendokexokendo...exo (98%)endo (2%)+XX Scheme 1-5

PAGE 19

6 Cyclizations are sensitive to the same thermochemical, steric and polar effects as intermolecular additions, but they are subject to an additional constraint of ring formation. 5-Hexenyl Radical Cyclizations The hexenyl radical is unique in that radical cyclization predominates and favors exo regioselectivity. According to Baldwin, exo-cyclizations (which predominate for a 5-membered ring closure) are favored over endo-cyclizations (which result in the formation of cyclohexyl systems). 13 Less-favorable entropies and steric effects in the 6-endo cyclization transition state (such as 1,3-diaxal interactions) also account for the favored formation of the 5-exo cyclization adduct. 4, 14 Because of these energy factors and steric considerations, cyclizations usually are faster for the formation of 5-membered rings than any other ring size. Since the 5-hexenyl radical cyclizations are twenty times faster than 6-heptenyl closures, 5-membered ring formation is less prone to competitive side reactions than the slower reacting 6-membered radical closures. 15 An additional advantage to formation of a cyclopentyl system is its predictable and outstanding regioselectivity. Moreover, the stereoselectivity of the reaction can be foreseen successfully by the Beckwith transition state model. 16 According to Beckwith’s theory, the early transition state of a 5-exo radical cyclization resembles a cyclohexane ring, preferring a chair instead of a boat conformation and pseudo-equatorial substituents to pseudo-axial substituents (Scheme 1-6). Furthermore, the Beckwith model suggests that C1 or C3 substitution of the 5-hexenyl radical gives primarily cis-disubstituted cyclopentanes; whereas C2 or C4 substitution results in a trans molecule. 17 This stereoselectivity is explained by a

PAGE 20

7 transition state in the chair configuration, the substituents preferably being in the equatorial position. . Scheme 1-6. Beckwith’s chair model for the transition state of a 5-exo-hexenyl radical closure Through the formation of these cyclopentyl skeletons, highly complex or hindered structures can result. For instance, in Nagarajan’s synthesis of an angular triquinane silphilene, a neopentyl/quartinary center is produced (Scheme 1-7). 6, 7 In this case, a derivative of p-tolythionocarbonate 1-4 was treated with tributyltin hydride. The generated radical added stereoselectively to the double bond of the enone, yielding angular triquinane 1-5, exclusively. This radical, intramolecular cyclization was among the first approaches to the tricyclo[6.3.0.0 1, 5 ]undecane skeleton, and is a clear example of the complexity that can be achieved with the versatile radical-cyclization methodology. OSOOOOn-Bu3SnH, AIBN PhH, 80oC1-41-5 Scheme 1-7

PAGE 21

8 Chiral Templates for Free-Radical Chemistry The triethylborane and oxygen initiation system aids in design of asymmetric radical reactions. An initiating system is not the only ticket for a desired stereochemical result. Other factors can and must be exploited. One of these factors is the use of a chiral template. Use of chiral auxiliaries in free-radical chemistry is not a new idea, and has been investigated thoroughly for the last 20 years. Many different appendages have been used to assess free-radical reactions on chiral support. Templates have been used to study asymmetric radical transformations (such as conjugate type additions, cyclizations, and atom transfer processes). A wide variety of chiral moieties have been explored in recent literature (Scheme 1-8). 18 Curran’s derivative of Kemp’s triacid 19 received great acclaim for its selectivity, although construction of the appending molecule is quite tedious. NOOROSOROROPhOOOOBnOR:Sibi's oxazolidinone20Malacria's chiral sulfoxide ligand21Nishida's chiral ester22Enholm's carbohydrate template25 Scheme 1-8

PAGE 22

9 Many investigators have used Evan’s oxazolidinone; perhaps most notable was Sibi’s attempts at intermolecular stereoselective free-radical additions with such a template. 20 Other scaffolds such as sulfoxides and chiral esters also were used in radical transformations with moderate success. 21, 22 Carbohydrate derivatives, despite their complex chiral nature and ready availability, curiously are underexploited in this realm of chemistry. Ours is one of the few studies in this area. Sugar derivatives have been regarded as too complex to be useful in asymmetric synthesis, containing too many chiral centers, and with too many functional groups. It was thought that the chiral information they contain could not be exploited in a stereodifferentiating selection process, in an organized and surveyable fashion. Therefore, carbohydrates have been transformed and converted into diverse, interesting chiral natural products (as opposed to taking advantage of their inherent asymmetry for auxiliary purposes). 23, 24 The steric, stereoelectronic, and coordinating properties of carbohydrate templates also can be used selectively to attain high levels of induction in processes such as Diels-Alder reactions, [2+2] cycloadditions, cyclopropanations, and Michael additions; however, their application to free radical chemistry remained mostly unexplored until Enholm and coworkers investigated the use of (+)-isosorbide as a chiral appendage. 25 Although the asymmetric influence of the appended chiral auxiliary is apparent in the free-radical process, asymmetric induction has not been all that successful in this area of chemistry. In fact, it was not until Renaud and Porter demonstrated the powerful tandem of chiral auxiliaries and Lewis acids in the free radical reactions, illustrating that true selectivity is observed. 9, 26 Before the utility of combining Lewis acids and chiral auxiliaries was realized, the first application of Lewis acids in radical reactions involved polymerization reactions. 27

PAGE 23

10 These studies observed that the Lewis acid could control the reactivity of radicals. This effect was not applied to synthetic radical chemistry until fairly recently. An early example concerns the cyclization of an aminyl radical precursor, where the PTOC is pyridine-2-thion-N-oxycarbonyl (Scheme 1-9). 28 NPTOCNSPyTi(OPr)Cl3CH2Cl2, h-78oC, 98% Scheme 1-9 Under strictly neutral conditions (no Lewis acid), the cyclic product is not observed. However, by using the titanium metal complex, formation of the cyclization product is quantitative. The amount of product formed exceeds the amount of Lewis acid used, implicating a potential, catalytic cycle. 29 The role of Lewis acids in radical reactions is thought to be multifunctional. Commonly, Lewis acids are responsible for substrate activation (as in the above case, for radical addition), resulting in higher yields of products. 9 Chiral, designer Lewis acids (alone, or in combination with chiral auxiliaries) are used to achieve good selectivity. For instance, in the case of the popular oxazolidinone scaffold, a bidentate Lewis acid is suggested to control the rotamer population so that the N-enolyloxazolidinone can attain a conformation that favors the formation of 1-6 (Scheme 1-10). 20 NOPhOOINOOOPhH1-6 Bu3SnH AIBN, 80oC82%, >98% d.e. Scheme 1-10

PAGE 24

11 This powerful combination instigated the use of amide, ester, and sulfoxide moieties as potential chiral auxiliaries. Chelation by a Lewis acid provides a steric bias toward the oncoming radicophile, and also locks the chiral scaffold into a preferred conformation or rotameric form. 20 This combination coaxed many synthetic chemists back to the benefits of free radical chemistry, gaining both neutrality in reaction conditions and stereoselectivity in a free-radical transformation. Moreover, this combination elucidated the powerful potential of carbohydrate derivatives as chiral scaffolds. The complexation of Lewis acids (which results in an organized arrangement of the functional groups of a carbohydrate) seemed paramount to effective use of sugar templates as stereodifferentiating tools in asymmetric synthesis. Enholm 25 has shown that carbohydrate template 1-7 and zinc chloride as a chelating agent gave a diastereomeric ratio of 100:1 for the cyclization of 1-8. Saponification recovers the carbohydrate template and (S)-(+)-indan acid 1-9. In a different example, similar success using a different auxiliary was examined (Scheme 1-11). Scheme 1-11 BrOOOOHHOBnHSnBu3, BEt3, O2OOOOHHOBnLiOHZnCl2OOH100:11-71-81-9 OOOOOBrBnOOOOOOBnOHnBu3SnHLewis Acid, Et3B, O21-101-11

PAGE 25

12 In this example, a (+)-xylose-derived template 1-10 was used and similar diastereoselectivity was achieved. These studies show that a number of factors are required to achieve a good asymmetric radical reaction. A template with an appropriate choice of solvent, temperature and Lewis acid will affect the outcome of most asymmetric radical processes. Recently, Enholm and coworkers also successfully showed that using D-xylose as a chiral appendage provided a Karasch atom transfer with enantioselectivities up to 71% using Eu(OTf) 3 as a Lewis acid (Scheme 1-12). 30 OOOBnOOOBrOOOOBnOOBEt3, O2, 1-hexene Lewis AcidBrd.e. up to 12.5:1 Scheme 1-12 DNA Transfection The use of DNA as a “drug” is a fairly new concept, and appears to be a promising way for curing DNA-linked illnesses such as mucoviscidosis. When DNA is used in therapeutic applications, the process is called gene-therapy. 31 One of the most important and difficult early stages of the gene-therapy protocol is to transfer the DNA into the target cell (cancer cells for instance). To do so, recombinant viruses or gene-transfection synthetic vectors are developed. 32 Viral vectors are powerful tools in terms of transduction efficiency. However, immunogical problems limit their continued use in clinical studies. Twelve years ago, synthetic vectors were an alternative because, lacking any DNA information, they were not a threat and should not induce immunologic response. The most efficient synthetic vectors currently used lead to the complexation

PAGE 26

13 and the compaction of DNA. In order to complex and compact, lipids, polymers, or cationic peptides able to electrostatically interact with DNA are synthesized. 33 Among cationic polymers, polyamines of primary, secondary and tertiary amines are the most efficient, because they provide a wide spectrum of protonation. 32 Particles of DNA/polyamines are still capable of being protonated when, during their endocytosis, they are trapped in the endosomes. This “proton-sponge” effect would explain the endosomolytic activity associated with the polyamines (which favor a transmembrane crossing of the DNA into the cytosol). So far, we are unable to explain why these synthetic vectors provide a much higher transfection rate than the naked DNA itself. The mechanism remains unclear. This dissertation presents the synthesis of a new DNA transfection-agent precursor based on a pseudo-proline moiety 1-12 (Scheme 1-13). NCO2HFmocOHNR'OOOOOROR1-12R= cationic lipids, amines or steroids Scheme 1-13 RGD Analogues Proteins are a continuous source of possible drug candidates. Much of the focus with respect to peptide-based drugs is on cyclic peptides as drug candidates. 34 Studies show that cyclic peptides exhibit greater stability in vivo. Lack of motion possibly allows masking of an active bioconformation that could aid the study of protein folding. If a conformation of a designed cyclic peptide is not biologically active, then manipulation of functionality can give one a diverse library of cyclic peptide analogues.

PAGE 27

14 A recurring problem with proteins as potential therapeutics is size. In vivo, active peptides show a very short half-life. 35 Most active peptides are degraded by a variety of different proteases, making passing through the diffusion barrier of the cell wall extremely difficult. Many strategies can be used to help the uptake of peptides through cell barriers. One approach is incorporation of proteins or peptides into nanoparticles. The other approach is making the protein a prodrug which is focused on the physiochemical properties; however, the most popular and most effective way to alter peptides is through chemical modification. Incorporation of D-amino acids and other spacers can potentially change the active bioconformation. 35 Natural cyclic peptide analogues are the most frequently used, because of the availability of individual amino acids. An L-amino acid in combination with oligonucleotides is a common variation from an all peptide backbone (Figure 1-1). These analogues are based on nucleic acids along with peptides, to try and mimic nucleotide cleavage. AspGlyArgOIONHOOOHNOHNHNNHRHNOOPd(OAc)2, PPh3, Bu3NCl1-131-14 Figure 1-1. Cyclization of an RGD analogue An example of a cyclic peptide application would be one that incorporates the RGD sequence. RGD (Asp-Gly-Arg) is a tripeptide recognition motif that is important in

PAGE 28

15 cellular adhesive properties. The tripeptide recognition sequence of arginine, glycine, and aspartic acid (R-G-D) is found in the binding domain of many extracellular matrix proteins. Interest in this sequence has increased due to its binding to specific receptors. The processes the binding controls include are hemostasis, cell proliferation, cellular transduction, and tumorgenesis. 36, 37 All seem to be controlled by binding of the sequence (Figure 1-2) to extracellular matrix receptors. The receptors are called integrins. Binding of RGD to an integrin mediates a cellular signal that is sent from the matrix to the interior of the cell. Once inside the cytoplasm, the specific function that the signal caused commences. Integration of messages from the extracellular matrix to the cytoplasm seems to be caused (and to some degree controlled) by initial binding. 36, 39 NHHNNHOHNNHH2NOOOOHAAAAAAAAAAAAGlyAspArgR-G-DAA= amino acid Figure 1-2. RGD recognition sequence Integrins are receptors found on the surface of the extracellular matrix that binds the sequence. A variety of integrin receptors exist, each controlling a number of cellular processes. An important integrin-RGD binding event is seen in the protein fibronectin. Much is known about this protein, since it was the first well-characterized adhesive protein. Though its structure is extremely complex, the protein is folded into specific domains; and several of the domains are exposed to the integrin receptor, allowing RGD sequences to bind. Binding of RGD in fibronectin to integrin receptors helps prevent tumor development. 40 Many other matrix proteins bind to integrins and have RGD as its

PAGE 29

16 initial trigger. The synthetic community’s interest has turned to finding RGD mimics that bind to integrins. 41 A summary of RGD binding with respect to an integrin receptor (Figure 1-3) shows binding to fibrinogen. Figure 1-3. RGD binding to an integrin Binding of RGD sequences on matrix proteins is based on many factors. Conformation of the sequence is extremely important. 42, 43 Changing flanking amino acids can have effects on binding. A simple example is changing a flanking amino acid from an aspartic acid to a glutamic acid. 38 A change in one methylene unit can alter conformation drastically especially in cyclic-polypeptide moieties. Along with a change in flanking amino acids, restricted conformation has been important in design of potential RGD-binding agents. 38 The approach in design of RGD mimics uses Kessler’s spatial screening approach. 44, 45 Spatial screening relies heavily on known domains of integrins. Design of possible mimics is based on what is known of space that is free for a sequence to approach. 46, 47

PAGE 30

17 Figure 1-4. Kessler’s sugar-based cyclic tripeptide. However, in preparing a mimic, uncertainty will still arise as to whether the design fits into the pocket of a receptor. Since conformation is all-important in designing analogues, progress has been made in cyclic peptides and peptidomimetics of the sequence. Carbohydrate-based analogues were the first to anchor the tripeptide in a cyclic array. Kessler 37 used sugar-based amino acids (1-15, 1-16) to anchor the tripeptide (Figure 1-4). The only difference between the two analogues is the stereochemistry at the C 1 position. The subtle change had an effect on binding to different types of IIIb 3 integrins. The IC 50 value for 1-15 was 720 nM as opposed to 13.4 nM for analogue 1-16. A huge difference in potency based of IC 50 values is observed from just one different stereocenter. A second set of analogues shows a similar effect of structure and activity relationships. This set is based on 6,5-fused bicyclic lactams. Key features of the scaffold include a natural proline residue built into the 6,5-system. 48 Inclusion of proline is important in the design because it can mimic a -turn. Evidence suggests that RGD recognition sequences may be in a -turn structure. 35 In this set of analogues (1-17, 1-18, 1-19), the difference is at the stereochemistry of the ring junction. Analogue 1-18 showed an extremely high binding affinity for the integrins associated with the matrix protein O B n O B n O H N O B n O B n O O B n A s p G l y A r g O H N O O B n A s p G l y A r g 1-15 1 16

PAGE 31

18 echisatin (Figure 1-5). Since these analogues have an unnatural amino acid, they are referred to as cyclic pseudopentapeptides. The bicyclic lactam systems are based on natural proline; and interest in our group turned to solely proline-based RGD systems. NOOAspGlyArgHNNOOAspGlyArgHNNOOAspGlyArgHN1-171-181-19 Figure 1-5. RGD analogues No analogue had been made of an actual proline-anchored RGD system (1-20). Enholm and coworkers were the first to propose synthesis of a proline-anchored RGD system. The advantage in design of an analogue with a natural anchor is that there is no required synthesis of an “unnatural” piece. An analogue contains all naturally occurring L-amino acids. Even if the binding of analogue 1-20 is not as that of 1-17, 1-18, and 1-19, a comparable natural analogue may be more attractive because of the availability of the anchor (L-proline). Their study consisted of synthesis of the cyclic RGD (1-20) with L-proline as its anchor via a solid-phase peptide synthesis (Figure 1-6). NOONHOHNNH2NHNHOHOOHN1-20 Figure 1-6. L-Proline based cyclic RGD

PAGE 32

19 Similar to the work done by Enholm et al., the proline-based transfection-agent precursor could be submitted to a similar synthetic pathway and provide RGD analogues bearing lipids and/or cationic chains, leading to a new class of synthetic vectors. Combining cationic/lipophilic properties of the grafted chains and 3-D conformation of the RGD moiety could provide better adhesion to the membrane of the target cell, specific recognition by the cell receptors, and more efficient transfection of the DNA into the cytoplasm. History of Molecular Imprinting Molecularly imprinted polymers (MIPs), discovered in 1972, represent a new class of materials that have artificially created receptor structures. 49 In the last decade, the development of assays, sensors and membrane-capillary electrophoresis was a new area in which MIPs provided opportunities for advancement. Molecular imprinting technology has recently developed into a viable approach for mimicking natural recognition entities, such as antibodies and biological receptors. 50 Materials with very high selectivity can be obtained using this method. This is especially the case in separation science, for which applications ranging from chiral separation to solid-phase extraction and combinatorial library screening have been proposed. 51 Much attention is also being given to applying the technique to diagnostic drug assays (in which the characteristics of the imprinted materials may potentially complement current antibody-based methodologies); and to the replacement of biological receptors involved in biosensor technology. One of the most intriguing challenges for the use of molecularly imprinted polymers (MIPs) however, is their application as enzyme mimics in synthesis and catalysis. 52 In parallel to the work done with catalytic antibodies and supramolecular (model) systems, attempts have been made to build catalytic activity into molecularly

PAGE 33

20 imprinted materials. Given the combination of high selectivity and robustness that can be accomplished with imprinted materials, their use in demanding processes may find many niches, such as operations at elevated temperature and pressure, reactions under acidic or basic conditions, and reactions in organic solvents. Methodology and Principles of Molecular Imprinting Molecular imprinting is a technology in which recognition sites are created in a macromolecular matrix (or plastic) using a molecular template in a casting procedure. The selected ligand or print molecule is first allowed to establish bonds with functional monomers and the resulting complexes are subsequently captured into a rigid twoor three-dimensional network following polymerization. Subsequent extraction of the print molecule then develops the recognition sites that are left in the matrix, where the spatial arrangement of the complementary functional entities of the network, together with the shape image, correspond to the imprinted molecule. The process of molecular imprinting involves complex formation between a macromolecule and low molecular weight template in solution, followed by drying and washing the solid complex with a selective solvent that removes the template but does not dissolve the polymer. Since the molecular mobility of the polymer chain is very low in the solid state, the macromolecule retains the conformation induced by the template even after the template has been removed from the complex (Scheme 1-14). Three particular features have made MIPs the target of intense investigation: their high affinity and selectivity (similar to those of natural receptors), their unique stability which is superior to that demonstrated by natural biomolecules, and finally the simplicity of their preparation and the ease of adaptation to different practical applications.

PAGE 34

21 Scheme 1-14. Principles of molecular imprinting (left) Self-assembly approach. (right) Preorganized approach. Radical polymerization is the most commonly used in MIP technology, this polymerization proved to be adequate and entirely compatible with a wide range of application and templates. However, radical polymerization can’t be used in specific applications: for instance with light or heat sensitive templates or for substrates that react with free radicals like halides. Radical initiation requires either heat or irradiation, the template could be destroyed or altered. It then appears necessary to develop new ways of making MIPs. To date, no studies used Ring Opening Metathesis Polymerization (ROMP), so this dissertation presents the first study of ROMP application in MIP technology. As shown in chapter 3, compared to radical polymerization, ROMP method is more efficient in term of selectivity, speed of reaction, and compatibility. We hope this might lead to a totally new way to perform these polymerizations.

PAGE 35

22 Fluorescence and Fluorescent Molecules in MIP Fluorescence is a radiative process, this is a phenomenon in which absorption of light of a given wavelength by a fluorescent molecule is followed by the emission of light at longer wavelengths. The distribution of wavelength-dependent intensity that causes fluorescence is known as the fluorescence excitation spectrum, and the distribution of wavelength-dependent intensity of emitted energy is known as the fluorescence emission spectrum (Scheme 1-15). Scheme 1-15. Illustration of the fluorescence process: An incident ultraviolet photon is absorbed and transfers its energy to an electron (process 1), the electron loses some of its energy in non-radiative processes, 2, and then the electron makes a transition to the ground state, process 3, emitting a photon with energy in the visible region. Fluorescence has proven to be a versatile tool for a myriad of applications. It's a powerful technique for studying molecular interactions in analytical chemistry, biochemistry, cell biology, physiology, nephrology, cardiology, photochemistry, and environmental science. It boasts phenomenal sensitivity for the analytical chemist or the life scientist working at nanomolar concentrations. But fluorescence offers much more than mere signal-gathering capability. New developments in instrumentation, software, probes, and applications have resulted in a burst of popularity for a technique that was first observed over 150 years ago.

PAGE 36

23 As the theoretical underpinnings of fluorescence became more understood, a more powerful set of applications emerged that gave detailed information about complex molecules and their reaction pathways. The binding of biochemical species can be easily studied in situ . 53 Distances within macromolecules may be measured. The dynamics of the folding of proteins can be studied. Concentrations of ions can be measured inside living cells. Membrane structure and function may be studied with fluorescence probes. Drug interactions with cell receptors can be investigated. Minute traces of fluorescent materials can be detected and identified in mixtures. Oil samples can be finger-printed and identified by their fluorescence. 54 Electronic structure and dynamics of an excited state of a molecule may be elucidated. These are only a few examples of the applications of modern fluorescence techniques. For instance Miller and Copeland 55 have developed molecules that are able to fluoresce in presence of acidic compounds. Meanwhile, others, like Swager 56 , focused on molecules able to fluoresce in presence of phosphorous compounds. Combining fluorescent molecules and MIP technology could provide highly selective detectors. In fact, MIPs provide specific cavities which will only trap the molecules that have been used as template. Thus, if we can graft, on the cavities, molecules that fluoresce in presence of the template, then we would have a selective detector. To illustrate this concept let’s assume that one has a mixture of different sugars and that he wants to detect only glucose. A MIP using glucose as template and a molecule that fluoresces with any sugar could be made; without the MIP the molecule would give a positive response with any sugar, but since the MIP selectively traps the glucose, this detector would only give positive result with the template it has been imprinted with, in that case glucose.

PAGE 37

24 This dissertation presents the first attempted synthesis of the MIP of selective fluorescent detectors of nerve gases such as sarin, soman, tabun or VX, using our new ROMP methodology. Nerve Gases A factory for production of this first nerve agent was built and a total of 12,000 tons of tabun were produced during the years three years (1942-1945). At the end of the second world war the Allies seized large quantities of this nerve agent and other nerve agents that Schrader 57 and his co-workers had synthesized: they had synthesized about 2,000 new organo-phosphorus compounds, including sarin (1938). 58 The third of the "classic" nerve agents, soman, was first produced in 1944. 59 These three nerve agents are known as G agents in the American nomenclature. The manufacture of sarin never started properly and up to 1945 only about 0.5 ton of this nerve agent was produced in a pilot plant. This was in hindsight a very lucky outcome or the history books would read very differently. Enough sarin gas could easily have won Germany the war. There is evidence that suggests that Hitler was advised against using the agents and even stopped their production. Hitler's Minister of Production, Albert Speer, said after the war, "All sensible army people turned gas warfare down as being utterly insane, since, in view of America's superiority in the air, it would not be long before it would bring the most terrible catastrophe upon German cities." Immediately after the war, research was mainly concentrated on studies of the mechanisms of the nerve agents in order to discover more effective forms of protection against these new CW agents. The results of these efforts led not only to better forms of protection but however to new types of agents closely related to the earlier ones. By the mid-1950's a group of more stable nerve agents had been developed, known as the V-agents in the American nomenclature. 60 They are

PAGE 38

25 approximately ten-fold more poisonous than sarin and are thus among the most toxic substances ever synthesized. Nevertheless due to Sarin’s other properties, it is still extremely effective. The first publication of these substances appeared in 1955. Ghosh 61 and J.F. Newman, described one of the new strains of nerve agents substances, known as Amiton, as being particularly effective against mites. At this time, intensive research was being devoted to the organo-phosphorus insecticides both in Europe and in the United States. At least three chemical firms appear to have independently discovered the remarkable toxicity of these phosphorus compounds during the years 1952-53. Surprisingly enough, some of these substances were available on the market as pesticides. A remarkably stupid decision, they were soon withdrawn owing to their considerable toxicity to mammals and therefore humans. This served as a warning that the proper research should be carried out before bringing out new products to spray over our food. Another, more persistent agent, named VX was discovered by British chemist R. Ghosh. 61 It was touted as being even more toxic than the previously synthesized nerve agents. Since the discovery of VX in 1949 there have been only minor advancements in the development of new nerve agents (Figure 1-7). In the United States, the choice fell in 1958 on a substance known by its code name VX as suitable as a chemical warfare agent of persistent type. It had the consistency of thick motor oil but slowly broke-down in the rain. Full-scale production of VX started in April 1961 but its structure was not published until 1972. Stockpiling the existing agents became a way for global heavyweights to flex their power in the later half of this century, but now efforts are being made to destroy the enormous stockpiles.

PAGE 39

26 OPOCNNOPOFOPOFC2H2OPOSNVXTabunSarinSoman Figure 1-7. Nerve gases A famous contemporary use of nerve agents was in the Iran-Iraq war (1984-1988). In this conflict the UN confirmed that Iraq used the nerve agent Tabun and other organophosphorous nerve agents against Iran. This incident is a prime example of how chemical warfare technology was shared during the Cold War. The Soviets would arm their allies while the US did the same for its allies. Iraq was obviously a benefactor and implemented its chemical stockpiles during the war. Another contemporary incident of nerve agent use occurred in Japan. The Aum Shinrikyo Cult was reported to have used the nerve agent Sarin in a Tokyo subway. This incident of use gives some clue as to the new roles that nerve agents play as tools of terrorists instead of powerful nations. While chemists put a lot of time on making nerve agents, they did not think of developing ways of efficiently detect those substances. A few detection systems exist, however they are either not selective enough (they do not detect only nerve gases but also any phosphorous compounds like pesticides) or do not give a result fast enough. 56 Some of them require a minimum of 30 minutes to provide a response, meanwhile people would already die. Swager 56 proposed a promising fluorescent detector which gives a fast and reliable response in a few minutes. Nevertheless his molecule is not selective enough, it can react with pesticides. To be efficient and usable, Swager’s probe needs to be more

PAGE 40

27 selective. This dissertation presents the synthesis of a new detector based on the combination of the ROMP-MIP technology and the use of fluorescent probe.

PAGE 41

CHAPTER 2 ALKENE-ALDEHYDE CYCLIZATION VIA O-KETYL RADICAL USING LEWIS ACIDS AND SUGARS AS AUXILIARIES Introduction Radical cyclizations with activated olefins offer synthetic chemists a well-documented and reliable means of carbon-carbon bond formation. 62 Synthetic benefits include neutral reaction conditions and tolerance of various functional groups and protecting schemes. All of these advantages render this methodology attractive for a wide variety of synthetic sequences. 63 Many radical cyclizations utilize precursors to carbon-centered radicals (such as halides, alkenes, alkynes, selenides, and sulfides), which are usually tethered to a suitable free-radical acceptor to produce cyclic derivatives when treated with tributyltin hydride. An aldehyde is not generally considered to be a good carbon-centered radical in these cyclizations. However, a free-radical cyclization promoted by a O-stannyl ketyl has shown that aldehydes and ,-unsaturated ketones and esters can be utilized. 64 In the most cases the diastereoselectivity of free radical reactions has been greatly enhanced due to the use of a combination of Lewis acids and appended chiral auxiliaries, scaffolds, and templates. 9, 65-66 Although radical additions to activated olefins have been studied extensively, 19, 63, 67 alkene-aldehyde cyclizations with chiral auxiliaries such as oxazolidinones, sulfoxides, and chiral esters have been far less investigated. 22, 68 Interestingly, carbohydrates as removable chiral auxiliaries for O-stannyl ketyl free-radical cyclizations have not been studied. In a previous study by Enholm 25 and Cottone, those sugars gave excellent diastereoselectivity for a 5-hexenyl 28

PAGE 42

29 radical cyclization. Our investigation examined the utility of two inexpensive carbohydrate derivatives as sources of asymmetry for an aldehyde-alkene radical cyclization via an O-stannyl ketyl radical at low temperatures (-78 o C) with tributyltin hydride, BEt 3 , and O 2 (Scheme 2-1). Very high diastereomeric ratios were achieved while the effects of temperature dependence, Lewis acids, and solvents were all studied. By correlation with known compounds, the newly generated chiral centers in 11a,b in both carbohydrate-mediated cyclizations are (3R,4S). OOOOOBnOHHHOOOOHHHHOOOOHHHOHOOOOBnOHHRSnBu3SnH, Lewis Acid Et3B, O2, CH2Cl2 -78oC(+)-Isosorbide => 2-11a(+)-Isomannide => 2-11b2-11a,b** Scheme 2-1 Earlier investigations of auxiliaries used in various radical reactions have relied on a chiral oxazolidinone ring. 2, 9, 19, 66-69 However, these investigations examined the commercially available and less expensive sugar auxiliaries (+)-isosorbide and (+)-isomannide. Both carbohydrates are highly oxygenated and provide multiple sites for metal chelation unlike oxazolidinones. In addition, those sugars are isomers, and could provide different chelation properties. Thus, additional affinity that strongly attracts the Lewis acid was a key and desirable trait in selecting these auxiliaries.

PAGE 43

30 Preliminary Study A preliminary study of the cyclization without any chiral appendage has been carried out. Precursor 2-8 was to be constructed via a Wittig reaction; the main framework was to be assembled via an unsaturated ester and an aliphatic aldehyde. The ylide 2-6 was synthesized from bromomethyl acetate and triphenylphosphine with a 98% yield, then reacted with the lactol 2-5 obtained in 98% yield by reduction with DIBAL of commercially available valerolactone. The rate of the Wittig coupling is faster than the rate of the equilibrium between the lactol and its open form (hydroxyl-aldehyde). The cis/trans mixture of the free primary alcohol 2-7 was then oxidized using PDC (pyridinium dichromate). The E and Z isomers 2-7 were not separable by flash chromatography, however after oxidation the trans precursor 2-8 was easily separated from the cis in 85% yield. OOOOHPh3PCO2MeCH2Cl2HOCO2MeOCO2MeH2-52-72-62-8DIBAL-HCH2Cl2PDC, CH2Cl2OOMeOBOOAHOOMeO Bu3SnH, Lewis Acid+2-8 Scheme 2-2 The cyclization of 2-8 gave two products, the trans isomer A and the cis isomer, which through a lactonization provided B (Scheme 2-2). Effect of the temperature and the presence of Lewis acid have been explored. As expected, the lower the temperature the higher the alcohol:lactone (A:B) ratio (thermodynamic control). The different ratios were obtained from the 1 H NMR spectra and also from HPLC. Since A and B are two different compounds their ratios can be easily measured from the spectra but, for more precision

PAGE 44

31 HPLC was used. Many Lewis acids were tested, and after analysis of the crude reaction mixtures, it appeared that the use of Cu (I) OTf and MgBr 2 gave ratios up to >100:1, as shown in Table 2-1. That showed that the metal provided a suitable chelation. Of course, as we didn’t have any source of chirality, no enantioselectivity for A as well as B was observed: A and B were obtained as racemic mixtures. Table 2-1. Diastereoselectivities achieved with the precursor Entry Lewis Ac. Solvent Temperature % Yield A:B 1 a none Benzene 80 o C 97 1.1:1 2 none CH 2 Cl 2 20 o C 70 1.8:1 3 none CH 2 Cl 2 0 o C 58 3:1 4 none CH 2 Cl 2 -78 o C 30 7:1 5 MgBr 2 .OEt 2 CH 2 Cl 2 , ether -78 o C 83 >100:1 6 Cu I (OTf) CH 2 Cl 2 -78 o C 98 >100:1 7 ZnCl 2 CH 2 Cl 2 , THF -78 o C 90 >100:1 a AIBN was used as initiator in entry 1, all other entries used BEt 3 and O 2 . After this preliminary study of the stereocontrol of the radical cyclization in presence of Lewis acid we wanted to see if the use of sugars as chiral auxiliaries could provide a diastereocontrol of this radical reaction. Synthesis of the Chiral Precursors Chiral precursors 2-10a,b were constructed via a Wittig reaction; the main framework was assembled via an unsaturated chiral ester and an aliphatic aldehyde. Carbohydrate-derived Wittig reagents 2-4a and 2-4b were synthesized from monobenzyl

PAGE 45

32 ethers 2-2a and 2-2b, respectively, in two steps in overall 42% and 23% yields, as illustrated in Scheme 2-3. OOOHOHHHOOOHOBnHHClOClOOOOOOOHOHHHCH2Cl2OHHOBnOOOHOOOOOBnOClHHOOOOBnOPPh3HH2-2a,b2-3a,b2-4a,b2-4a,b2-10a,bpyridine, CHCl31) Ph3P, THF2)NaOH, H2O Scheme 2-3 Monobenzylation of (+)-isosorbide 2-1a and (+)-isomannide 2-1b was achieved by Williamson etherification with benzyl bromide and benzyl chloride respectively, affording 2-2a and 2-2b, respectively in 82% and 45% yields. Compound 2-2a was obtained by Williamson etherification using NaH and benzyl bromide in THF with a catalytic amount of TBAI. Another procedure using DMF instead of THF was also performed, however DMF is not a great solvent to deal with: tedious work-up is necessary and traces of DMF always remain. Switching to THF required the use of a phase transfer reagent. After having tried TBABr, it appeared that TBAI gave better yields. The monobenzylation of 2-2b proved to be more difficult, this was probably due to the fact that the two alcohols are cis. Thus they are more sterically hindered and less accessible. The previous procedure (THF, NaH and TBAI) was first tested: 2-2b was obtained with a 30% yield. We finally applied a procedure reported by Loupy. 70 The commercially available isomannide was dissolved in water with potassium hydroxide and

PAGE 46

33 refluxed. Subsequent addition of benzyl chloride and acidic work-up provided 2-2b with 50% yield after recrystallization in ether. The remaining hydroxyl functionality of 2-2a and 2-2b was converted to the chloromethylene esters 2-3a and 2-3b using chloroacetic anhydride and pyridine. 13 Corresponding ylides 2-4a and 2-4b were obtained in 80% yields using the common procedure (triphenyl phosphine followed by a basic treatment). Similarly to the synthesis of the achiral precursor, 2-4a and 2-4b were coupled with the lactol 2-5 in methylene chloride. However, the reaction didn’t give yields higher than 10%. In that case, the rate of the Wittig coupling is not high enough to compete with the lactol-hydroxyaldehyde equilibrium. A way to trap the lactol 2-5 in its open form was needed. The easiest way was to protect the alcohol. OOOTBDPSOHHOBnOOOOSiPhPhOHHOBnOOOOTBDPS2-4a,b2-10'a,b2-15a,b2-52-5'CH2Cl2, RTTBDPSCl, Et3N CH2Cl2TBAFor HFTBDPS=HO Scheme 2-4 Treatment of 2-5 with Et 3 N and TBDPSCl provided the silylether 2-5’ in 83% yield. 2-5’ was then coupled with the ylides and provided the esters 2-10’a and 2-10’b in 80% yield. However, various attempts at deprotection of the alcohol failed. Indeed, as soon as the alcoholate was formed, it added to the -unsaturated esters 2-10’a,b via a 1,4-Michael addition and led to compound 2-15a,b (Scheme 2-4). Another route was devised. Surprisingly, the Wittig reagents simply coupled with freshly distilled

PAGE 47

34 glutaraldehyde 2-6 provided the chiral precursors 2-10a and 2-10b in 75% yield. Once the -unsaturated ester of each carbohydrate derivative was synthesized, various reaction conditions for a radical cyclization were explored (Scheme 2-5). OOOOOBnOHHOOOOBnOHHOHOOOOBnOHHOHOOOO+++ci s -lac t onestrans"alcohols"2-10a HSnBu3,Lewis Acid, BEt3/O22-11a2-11a'2-122-12' Scheme 2-5 Results and Discussion At 80 o C, AIBN was used as a radical initiator; however, triethylborane and oxygen were used for the lower temperature studies. As expected, the radical cyclization proved to be more selective at lower temperatures. We also discovered that the formation of the lactones 2-12 and 2-12’ were unfavored with isosorbide, due probably to the hindrance of the sugar, forcing the alcohol and the ester to be trans (Table 2-2). Without a Lewis acid, the diastereomeric ratios for 2-11a,a’ were very low, ranging from 1:1 to 1.1:1. When a Lewis acid was used, diastereomeric ratios increased substantially, suggesting the formation of an activated metal chelate with the isosorbide-appended auxiliary. Not all Lewis acids gave good diastereomeric ratios,

PAGE 48

35 among the best of the Lewis acids studied, zinc chloride proved to be the most effective for the isosorbide chiral scaffold with a diastereomeric value of 9:1. Table 2-2. Diastereoselectivities achieved with isosorbide Entry Lewis Ac. Solvent Temperature % Yield 2-11a,a’:2-12,12’ d.e.* 1 a none Benzene 80 o C 84 2:1 1:1 2 none CH 2 Cl 2 0 o C 40 3.3:1 1:1 3 Eu(OTf) CH 2 Cl 2 0 o C 65 6:1 1.7:1 4 MgBr 2 CH 2 Cl 2 -78 o C 71 98:2 1.1:1 5 Cu I OTf CH 2 Cl 2 -78 o C 70 98:2 2.3:1 6 ZnCl 2 CH 2 Cl 2 , THF -78 o C 73 98:2 9:1 7 Yb(OTf) 3 CH 2 Cl 2 -78 o C 68 98:2 1.4:1 a AIBN was used as initiator in entry 1, all other entries used BEt 3 and O 2 . * d.e. for the compounds 2-11a and 2-11a’. The radical cyclization of isomannide-adduct 2-10b proved to be even more successful, as shown in Table 2-3 (Scheme 2-6). OOOOOHHBnOOOOOHHOHBnOOOOOHHOHBnOOOOO+++ ci s -lac t ones trans"alcohols"2-10b HSnBu3,Lewis Acid, BEt3/O22-11b2-11b'2-122-12' Scheme 2-6 Notice that formation of the lactones 2-12 and 2-12’ was not observed even at high temperatures and without Lewis acid, showing that isomannide proved to have different

PAGE 49

36 properties compared to isosorbide. We believe those differences are due to the stereochemistry of the hydroxyl groups. Indeed, in the case of isomannide the two hydroxyl groups are in the cup formed by the fused cyclopentoses, providing better steric hindrance of one alcohol. Similar to Table 2-1, temperature dependence and an improvement in diastereomeric ratios with Lewis acids were observed for these cyclizations. Isomannide gave a higher diastereomeric ratio 100:1 for 2-11b,b’ with ZnCl 2 (compared to isosorbide). We assume that the position of the hydroxyl groups provide better chelation with the Lewis acids, hence giving a better diastereoselectivity. Table 2-3. Diastereoselectivities achieved with isomannide Entry Lewis Ac. Solvent Temp. % Yield 2-11b,b’:2-12,12’ d.e. 1 a none benzene 80 o C 96 >100:1 1:1 2 MgBr 2 CH 2 Cl 2 -78 o C 91 >100:1 >100:1 3 Cu (I) OTf CH 2 Cl 2 -78 o C 89 >100:1 >100:1 4 ZnCl 2 CH 2 Cl 2 , THF -78 o C 93 >100:1 >100:1 5 Yb(OTf) 3 CH 2 Cl 2 -78 o C 68 >100:1 2:1 a AIBN was used as initiator in entry 1, all other entries used BEt 3 and O 2 * d.e. for the compounds 2-11b and 2-11b’ For both sugars, lanthanides such as Eu(III) and Yb(III) were uniformly ineffective in enhancing the diastereoselectivity of the cyclization. In some cases, the lanthanide Lewis acids gave ratios that were similar to those reactions lacking any added Lewis acids. This wasn’t unexpected, because lanthanide Lewis acids demonstrated poor results in the previous radical study involving those sugars in our research. 25

PAGE 50

37 OOOOOBnOHHOHOOOOBnOHH nBu3SnH Lewis AcidEt3B, O2, CH2Cl2-78oC2-142-10a,b Scheme 2-7 In addition it is important to report that, aside from 2-11 and 2-12, the reduction of the aldehyde into 2-14 was observed, but remained low (<5% yield) (Scheme 2-7). Once the diastereomeric ratios of each reaction parameter set were determined, elucidation of the absolute stereochemistry of the major diastereomer formed was investigated (Scheme 2-8). OOOOHHOHBnOOOOOHHOHBnORRMeOOOHMeOOOHSS 1) KOH, MeOH2) H2SO4, MeOH2-11b'2-13[]25D=+43.1o 1) KOH, MeOH2) H2SO4, MeOH2-11a'2-13 major isomer[]25D=+34.5o9:1 mixture Scheme 2-8 The major isomer from the cyclization with isomannide in presence of ZnCl 2 (Entry 4 , Table 2-3) was saponified with potassium hydroxide to remove the chiral sugar. Esterification of the corresponding carboxylic acid in methanol was done, affording the known trans-(3R, 4S)-4-hydroxy-cyclopentane acetic acid methyl ester 2-13 with a specific rotation of +43.1 o , indicating that the absolute configuration of the isomer 2-11b is trans-(3R, 4S). 71 The same process was done with the 9:1 mixture of alcohols

PAGE 51

38 from the cyclization of the isosorbide adduct (Entry 6, table 2-2): after transesterification, a specific rotation of +34.5 o was measured, suggesting that the major isomer of 2-11a’ is also trans-(3R, 4S). Conclusion In conclusion, the highly oxygenated nature of carbohydrate derivatives is extremely advantageous, offering ease in functionalization, multiple sites for Lewis acid chelation, inexpensive and ready availability. As chiral templates for free-radical, both (+)-isosorbide and (+)-isomannide gave high diastereomeric ratios. This high selectivity can be attributed to the steric hindrance and the multiple sites of chelation, firmly anchoring the Lewis acid to the polyoxygenated sugar. Nevertheless, (+)-isomannide provided better diastereoselectivity, probably due to its higher steric hindrance.

PAGE 52

CHAPTER 3 MOLECULAR IMPRINTING WITH ROMP METHODOLOGY: EFFECTS ON POLYMERIZATION AND IMPRINTED PROPERTIES Introduction The general approach of the Molecular Imprinting technique is to combine the desired "imprint" molecule with a solution of small monomers which has several desired features: The monomers have a variety of functional groups (e.g., hydrogen bond donors, acceptors, possibly acidic or basic groups, or aliphatic or aromatic groups). One or more of these groups can specifically interact with the imprint molecule (e.g., they can form specific hydrogen bonds) The monomers can be made to polymerize to form a rigid, or semi-rigid polymer The imprint molecule is soluble in the monomer solution The monomers, either in solution or during the polymerization process, do not form covalent bonds with the imprint molecule (the interaction is strictly non-covalent) After polymerization of the monomers (in the presence of the imprint molecule) the general features of the polymer are as follows The polymer is porous enough to allow the imprint molecule to diffuse in and out The polymer is rigid enough that the organization of monomers in response to the imprint molecule is retained Therefore, the stereochemistry (i.e., size, shape and orientation of appropriate functional groups) of the pocket where the imprint molecules were located is retained These sites in the polymer constitute an "induced molecular memory" capable of selectively recognizing the imprint molecule Comparison of ROMP versus Radical Induced Polymerization Our research is focused primarily on the synthesis of sensors. In fact, typical problems that remain unsolved in connection with biological receptors for use in 39

PAGE 53

40 biosensors are: low stability, high cost and absence of enzymes or receptors that are able to recognize certain target analytes. MIPs could be considered as appropriate alternatives to the biological receptors for use in sensors principally because of their high stability. The most promising application area for MIP sensors is medicine and pharmaceutical industry in cases where the stability of MIPs would confer significant advantage. MIP sensors can be made for toxins and immunodepressants, which are difficult targets for the development of immunosensors. Because of easy regeneration, MIP sensors are considered to be particularly suited to the design of high throughput screening (HTS) and toxicity evaluation systems. 72 Synthesis of the MIPs Our contribution to this area consists in the fact that ROMP catalyzed by Grubb’s catalysts was used for the synthesis of our MIPs while all other researchers use radical induced polymerization. To our knowledge, we are the first to report the use of ROMP methodology in molecular imprinting technology. The main advantages of the ROMP versus radical polymerization are: shorter time of reaction, mild synthetic conditions, high stability of the monomers, improved cross-linking, and tolerance of ROMP to a large type of functional groups. Templates We used acetylcholinesterase inhibitors (such as pyridostigmine bromide 3-1) as templates (Figure 3-1). It is important to note that we used these molecules just for a model study. Indeed our first project was to study a new methodology involving ROMP, and after succeeding in the formation of highly selective polymers we decided to apply it in the synthesis of nerve gas detectors. Therefore there are no links whatsoever between the use of acetylcholinesterase inhibitors as templates and the synthesis of nerve gas

PAGE 54

41 detectors. Those molecules have been extensively studied. Recently, pyridostigmine bromide was approved by the FDA as pretreatment against nerve gas. This is the first drug approved to increase survival after exposure to Soman “nerve gas” poisoning. Thus it appeared interesting to study the MIPs on those particular molecules. Those templates are polar and are able to create hydrogen bonds with the acid and esters groups present in the MIPs, providing the formation of a tight complex around the ligand. N+ONOBr-HON+Cl-Pyridostigmine bromideEdrophonium chlorideN-N+OONHEserine3-13-23-3 Figure 3-1. Templates Radical Polymerization The template was allowed to form solution complexes with functional building blocks (methacrylic acid). These complexes were subsequently fixed using a cross-linking monomer (ethylene glycol dimethacrylate) into a rigid network, locking the complexes in position in the resulting material (Scheme 3-1). As usually reported, the [template/monomer/cross-linking agent] ratio was [1/4/16]. AIBN was used as initiator, however irradiation could also been used, but our templates were light sensitive so it could have been an issue. The solid polymer was then ground and removal of the print molecule by refluxing the polymer in ethanol/acetic acid then exposed the recognition

PAGE 55

42 sites. More than 99% of the initial mass of template was recovered. The powder was then filtered, dried and ready for testing. OOHN+ONOBr-OOHOHOOOHOOHN+ONOBr-OOHOHOOOHAIBN, AcetonitrileOOOOOOHOOHOHOOOHRemove Rebind Scheme 3-1 ROMP Polymerization Ring opening metathesis polymerization (ROMP) and acyclic diene metathesis (ADMET) have shown utility in soluble supported organic chemistry. ROMP is based on release of ring strain usually from a norbornene system. The reactive double bond readily reacts with a catalyst to provide supports for organic transformations. NNRuClClPCy3RuClClPCy3PCy3Second Generation Grubbs Catalyst 3-4Grubbs First Generation Catalyst3-5 Figure 3-2. Grubbs ruthenium catalysts

PAGE 56

43 Both of these processes require specific catalysts to trigger the process. The two popular catalysts are the Grubbs’ 73 first generation catalyst 3-5 and Shrock’s 74 catalyst. A second generation Grubbs’ catalyst 3-4 is now more frequently used due to its tolerance of a multitude of functional groups (Figure 3-2). 75 LnM R MLn R LnM R LnM R R MLn n [2+2] retro[2+2] [2+2] retro[2+2] repeat ROMP Polymer 3-6 3-7 3-8 3-9 Scheme 3-2 Ring opening metathesis polymerization works via coordination of the ruthenium catalyst 3-7 to the double bond of the norbornene 3-6 and a [2+2] cycloaddition to form a metallocyclobutane 3-8. A subsequent retro [2+2] cycloaddition places the ruthenium on a double bond terminus 3-9. This process continues to the next norbornene double bond until the reaction is capped (Scheme 3-2). This is accomplished by addition of a capping agent such as ethyl vinyl ether that kills the catalyst. Two beneficial aspects of this process are that the ring stain of the norbornene system provides a driving force for the initial [2+2] addition of the ruthenium and that typically less than 10 mol% of catalyst is required to proceed. With the development of second generation Grubbs’ catalyst 3-4, polar groups that could have coordinated ruthenium can now undergo a polymerization. Norbornene systems are not necessary to perform polymerizations with the two generations of Grubbs’ catalyst and other metathesis catalysts. Molecules that are not

PAGE 57

44 confined to a strained system can undergo polymerization by acyclic diene metathesis (ADMET). R RRepeatR[2+2]RMLnRRMLnR[2+2)RLnMRLoss ofRetro [2+2]LnMRnR Scheme 3-3. ADMET polymerization Although the driving force for release of ring strain does not exist in this polymerization, a more diverse array of polymers can be made since the minimum requirement are two sets of double bonds. XADMET(Acyclic Diene Methathesis)RCM(RingClosing Metathesis)X()n ROMPRing Opening MetathesisPolymerization-C2H4nXn Scheme 3-4 The steps of catalyst loading and chain elongation are the same. The catalyst attaches to a terminal double bond (Scheme 3-3). Subsequent release of ethylene and then

PAGE 58

45 attachment to double bond terminus are the steps in this process. A summary of the processes (RCM and ADMET) is shown in Scheme 3-4. Comparing the two processes, ADMET gives a more diverse array of materials due to a lack of a strained system requirement, while RCM sees more total synthetic applications. Development of these catalysts helped in design of polymers for biological purposes. Polymers that are designed for these purposes can sometimes be called biomaterials. These molecules have a tremendous effect on human health as seen in the wide variety of applications. Possible uses of biomaterials include medical devices, sensors, drug delivery systems, and tissue engineering. 76, 77 Synthesis of the Monomers Our monomers 3-10 and 3-12 are norbornene derivatives. Monomer 3-12 was easily synthesized with excellent overall yields (Scheme 3-5). HOOHClOClO4OOOO4OHPyridine, CHCl3 87%NaBH4, MeoH 98%Cross-linking monomer3-113-113-12OHO3-10 Scheme 3-5

PAGE 59

46 Commercially available 5-norbornene-2-carboxaldehyde was reduced with NaBH 4 , providing quantitatively the 5-norbornene-2-methanol 3-11. Treatment of 3-11 with adipoyl chloride in presence of pyridine gave the diester 3-12 (92% yield) that would be used as cross-linking agent. Grubb’s catalysts 3-5 (5% cat.) provided a cross-linked polymer insoluble in methylene chloride in less than 1 minute, while it took more than 10 hours for radical polymerization, whatever the initiation method (irradiation or thermal activation). OOHN+ONOBr-OOHOHOOOHOOHN+ONOBr-OOHOHOOOHGrubb's and Super Grubb's catalyst cross-linking monomerOOHOOHOHOOOHRemove RebindRR Scheme 3-6 Different [template/monomer/cross-linking agent] ratios have been tested, and it appeared that the best one in term of rate of polymerization and mechanical properties of the polymer was [1/4/16]. Second generation Grubbs’ catalyst (3-4) proved to be more

PAGE 60

47 efficient than the regular one. It allowed not only a faster polymerization (hours versus couple of minutes) but also a higher cross-linking. This higher cross-linking rate would not only give a more robust polymer but also a denser 3-D network which could provide a better encapsulating of the template. After removal of the solvent, the “rubbery” solid was ground into a fine powder which was subsequently refluxed in ethanol to remove the template (Scheme 3-6). After filtration of the ethanol suspension, the powdered MIP was ready for selectivity tests. After concentration of the filtrate in vacuo, more than 99% of the initial mass of template was recovered. Analytical Method, Results and Discussion General Method Gas chromatography and mass spectrometry methods were used to determine the affinity of each polymer for various compounds. Each polymer was suspended in a solution of equimolar amounts of two materials: the template and one molecule that was structurally close to the template (test molecule). Aliquots were taken over time for which the polymer was filtered and the remaining solution was analyzed by GC/MS to determine its composition. Since it was impossible to quantitatively measure the quantity of template and test molecule trapped in the polymer, a back analysis was performed. From the starting amount of material and the amount present in each aliquot, we could quantitatively see the affinity and rate of binding for each polymer to a material. The percentage (%) of what remained in solution for both the template and the test molecule were calculated and, plotting the percentages against time allowed us to see the affinity and rate of binding. Comparing the selectivity properties of the different MIPs was then possible.

PAGE 61

48 Tests: Two different tests were performed, the first one consisted in evaluating the binding properties of the MIP while the other one would give us more insight about the increase of selectivity provided by the ROMP. For the first test, after having made the polymer (radical induced or ROMP) using x moles of template, a solution of x/2 moles of template and x/2 moles of test molecule was prepared. The ground polymer was then added to the solution and aliquots were taken over time. This test didn’t provide any absolute proof of selectivity, but showed the binding properties of each polymer. The second test differed from the first one only by the fact that instead of using x/2 moles we used x moles of both the template and the test molecule. This method undoubtedly shows the selectivity of the polymer. Both tests were performed five times for accuracy. The first template studied was Edrophonium Chloride (3-2). Test 1: The polymer formed via radical propagation trapped 91% of the initial edrophonium chloride (3-2) and 68% of the initial eserine (3-3) (Figure 3-3). When the ROMP polymer was used, 93% of the original edrophonium chloride was trapped (increase of 5% compared to the radical) while the eserine concentration in solution increased (46% versus 32%) (Figure 3-4). An increase in the affinity for the template was shown in the ROMP method. No change was shown for the affinity of the test molecule. It also appeared that the ROMP MIP trapped more selectively the template than the test molecule, the second test would confirm it.

PAGE 62

49 0102030405060708090100024681012Time (minutes)% remaining Edrophonium chloride Eserine Figure 3-3. Radical polymerization with edrophonium chloride test 1 0102030405060708090100024681012141618Time (minutes)% remaining in solution 20 Edrophonium chloride Eserine Figure 3-4. ROMP polymerization with edrophonium chloride test 1

PAGE 63

50 Test 2: The radical induced polymer showed a really poor, not to say zero selectivity (Figure 3-5). When the eserine and edrophonium chloride molecules were in competition (test 2), the polymer seemed unable of differentiating them. There seemed to be an exchange process: the concentration of the test molecule and the templates kept changing through time, the cavities provided by the radical induced MIP were not selective enough to selectively trap the templates molecules. The templates and test molecules were not differentiated by the MIP. 0102030405060708090100024681012141618Time (minutes)% remaining in solution 20 Edrophonium chloride Eserine Figure 3-5. Radical polymerization with edrophonium chloride test 2 However, when the ROMP polymer was used, the selectivity was greatly enhanced and only the templates molecule (i.e., edrophonium chloride) got trapped in the matrix (Figure 3-6).

PAGE 64

51 In that case the cavities formed in the MIP were shaped on the template and did not allow the test molecules to enter. The trapping process for the template was faster than the one for the test molecule. This result proved, in that particular case, that the ROMP methodology permitted the formation of more selective and better shaped cavities than the ones present in the radical induced MIP. 0102030405060708090100024681012141618Time (minutes)% remaining in solution 20 Edrophonium chloride Eserine Figure 3-6. ROMP polymerization with edrophonium chloride test 2 The second template was Pyridostigmine Bromide (3-1). Test 1: The polymer formed via radical propagation trapped 70% of the initial pyridostigmine bromide and 50% of the initial eserine (Figure 3-7). When the ROMP polymer was used, 90% of the original pyridostigmine bromide was trapped (increase of 29% compared to the radical) while the eserine concentration remained the same (52% versus 50%) (Figure 3-8). An increase in the affinity for the template was once again shown in the ROMP method. No change was shown for the affinity of the test molecule.

PAGE 65

52 010203040506070809010002468101214161820Time (minutes)% remaining in solution Pyridostigmine bromide Eserine Figure 3-7. Radical polymerization with pyridostigmine bromide test 1 010203040506070809010002468101214161820Time (minutes)% remaining in solution Pyridostigmine bromide Eserine Figure 3-8. ROMP polymerization with pyridostigmine bromide test 1

PAGE 66

53 Test 2: As reported with edrophonium chloride, once again the radical induced polymer showed a really poor, not to say zero selectivity (Figure 3-9). When the eserine and pyridostigmine bromide molecules were in competition (test 2), the polymer seemed unable of differentiating them. However, with the ROMP polymer the selectivity was greatly enhanced and only the templates molecules (i.e., edrophonium chloride) got trapped in the matrix (Figure 3-10). With this second template it also appeared that the ROMP MIP more selectively trapped the template molecule than the test molecule. In both cases, we were pleased to see that the trapping process of the template took less than 20 minutes. Two different tests were necessary to prove both the binding properties and the high selectivity provided. 01020304050607080901000246810121416182Time (minutes)% remaining in solution 0 Pyridostigmine bromide Eserine Figure 3-9. Radical polymerization with pyridostigmine bromide test 2

PAGE 67

54 010203040506070809010002468101214161820Time (minutes)% remaining in solution Pyridostigmine bromide Eserine Figure 3-10. ROMP polymerization with pyridostigmine bromide test 2 Control Experiments One can wonder whether the molecules were in the cavities within the polymer or simply attached on the surface of the polymer. In the second case, the selectivity provided by the MIP wouldn’t come from the cavities but only from the nature of the interactions between the molecules and the surface of the polymer. Control experiments were needed. To make sure that the molecules actually got trapped in the cavities and did not stay on the surface of the polymer (adsorption), a third type of analysis was performed with blank polymers for both radical and ROMP MIPs. Each blank polymer was mixed with one solution of a single molecule (pyridostigmine bromide, edrophonium chloride or eserine) of known weight (the conditions and quantities used in these control tests were identical to those used in the synthesis of the printed polymers). The mixture was stirred under argon and after 10 minutes, the suspension was filtered. The filtrate was concentrated in

PAGE 68

55 vacuo, the mass measured and a 1 H NMR as well as a GC trace were taken to check the purity of the residue. For each analysis the percentage of mass recovered was calculated. The results are reported in Table 3-1. The errors are due to the precision of the balance ( 0.1 mg). Table 3-1. Control experiments Entry MIP Molecule % recovery % trapped 1 Radical Eserine 96 3 4 3 2 Radical Pyridostigmine bromide 98 4 2 4 3 Radical Edrophonium chloride 95 1 5 1 4 ROMP Eserine 96 3 4 3 5 ROMP Pyridostigmine bromide 95 1 5 1 6 ROMP Edrophonium chloride 98 4 2 4 The control experiments showed that a small quantity (average of 3.5%) of molecules was trapped by the blank polymers, meaning that the surface wasn’t responsible for the high selectivities previously observed (Figures 3-3 to 3-10). When a template was used during the formation of the MIPs, the percentage of molecules trapped greatly increased, proving that the cavities present in the MIPs were responsible for the trapping of the molecules. These control experiments proved that the majority of the molecules retained by the MIP was actually trapped in the cavities and not adsorbed on the surface. Using MIP Technology in the Synthesis of Selective Nerve Gas Detectors Many times in history chemists created molecules that could be used in warfare, and unfortunately they did not think about ways of detecting them. The proliferation of

PAGE 69

56 toxic gases like Soman, Sarin and Tabun forced governments to encourage the development of detectors in order to protect soldiers from exposure. As terrorist acts and threats are growing, many scientists recently put interest in the synthesis and the study of nerve gas sensors. As the common frame of these gases are P=O bonds, they develop chemicals (sensors) able to fluoresce in presence of the gas molecule itself or even its hydrolyzed form (Figure 3-11). 56, 78 These molecules proved to be effective in solution and also on polymeric materials, and gave positive results even for a small quantity of nerve gas. However, these sensors are not particularly selective, they react not only with nerve gas, but also with any substance containing phosphorous like pesticides. Besides, these fluorescent molecules indicate the presence of toxic gas but do not give its actual nature (identity): it is then impossible to accurately respond to the threat and apply a destructive (deactivation) process. OPOCNNOPOFOPOFC2H2OPOSNVXTabunSarinSoman3-133-143-153-16 Figure 3-11. Nerve gases Our approach involves the use of our new ROMP-based MIP in combination with fluorescent sensors to create a selective nerve gas detector able to, not only detect nerve gas in the atmosphere, but also tell us its nature. One could then launch an effective response to inactivate the gas and protect soldiers (and/or targeted civilians). We proved that ROMP based MIPs are more selective than regularly used radical induced MIPs.

PAGE 70

57 Using nerve gas molecules (or their hydrolyzed form) as templates will provide specific cavities. This MIP’s property coupled with the presence of fluorescent sensors directly in the MIP cavities will give a specific response: the MIPs are selective, thus only the molecules used as templates would get into the cavities and react with the chemosensors. The idea is to create one MIP for every single nerve gas: one for Sarin, one for Soman and so forth. It is then possible to make a device containing each MIP and use it as a detector (Figure 3-12). This device will be fast and reliable. To illustrate let us assume the following: “A cult decides to spray Soman in a mall. Our detector will “react” and only the MIP imprinted on Soman will fluoresce. Authorities will directly know that an attack with Soman is in progress and will respond accordingly by using known deactivation of Soman processes.” VXVXVXSomanSarinTabunTabunSarinSomanSarinSomanTabunTabunSoman= non-fluorescent= fluorescentEach MIP as been imprinted with a different template, and will fluoresce ONLY in presence of the molecules of template.MIP polymers Figure 3-12. Detection device

PAGE 71

58 Synthesis of the Sensors Our current work is the synthesis of fluorescent sensors for nerve gases based on ROMP-MIP chemistry. We have tried to obtain two different sensors in order to compare their properties in the imprinted polymers. N O OMe OY N 3-22 3-18 Figure 3-13. Fluorescent sensors The molecules of interest are a cyclic amine 3-17 and an aryl-pyridine 3-18 which will fluoresce in the presence of nerve gas molecules (Figure 3-13). A molecule synthesized by Copelandand Miller was chosen. This sensor is a pH sensitive fluorophore which undergoes PET (Photo-induced Electron Transfer) processes when in the free amine form 3-17, but fluoresces when protonated. As the phosphore oxygen bounds are very polar, these molecules are acidic enough to protonate 3-17 (Scheme 3-7). 78 N O OH N+ O OH H BB= Base No Fluorescence Intense Fluorescence 3-17 3-17+BScheme 3-7 The first step was the esterification of isonipecotic 3-19 acid with HCl gas in methanol to give 3-20. In order to get 3-22, the methyl ester underwent a reductive

PAGE 72

59 amination with 9-anthraldehyde 3-21 (Scheme 3-8). After different procedures, we noticed that the amination was possible only with powdered molecular sieves. NH O OMe NH O OH N O OMe CHO 82% 72% 3-19 3-20 3-21 3-22 1) HClg, MeOH 2) NH3g NaCNBH3, THF, 4A powdered molecular sieves Scheme 3-8 The contact surface needs to be increased because of the high hydrophilicity of the methyl ester and also because of the water formed during the reductive amination. N O OMe N O OH N O O 75% 3-22 NaOH aq. THF/MeOH 3-17 .HCl 83% 3-23 3-11, DIC, DMAP, CH2Cl2 Scheme 3-9 Then, to be able to graft a sensor in the imprint polymer, methyl ester 3-21 was saponified with NaOH to give 3-17. By this way, the monomer 3-23 was synthesized by DIC (N, N’-diisopropylcarbodiimide) coupling of 3-17 and 3-11 (Scheme 3-9). To increase the chance of getting a successful sensor, we tried to synthesize another kind of fluorescent molecule. The second molecule we focused on was Swager’s sensor. 56 The biaryl compound 3-18 detects nerve agents by reacting with nerve gas molecules. This biaryl is a naphthalene derivative bearing a pyridyl and a hydroxyl (or siloxyl)

PAGE 73

60 substituent. Nerve agents (such as Sarin 3-14) are volatile organophosphorous compounds that react with a hydroxyl group in biological tissue to form a phosphate ester. When this reaction occurs at the catalytic site of acetylcholinesterase an enzyme critical to nerve function the enzyme’s activity is inhibited, leading to convulsions and death. In the detection system, organophosphates (which are used as model compounds for the nerve agents) react with the hydroxyl or siloxyl group of the indicator to form a phosphate ester. As this ester is a good leaving group, the indicator molecule spontaneously undergoes an intramolecular cyclization reaction, yielding a tetracyclic product 3-18 + B that fluoresce at a different wavelength and intensity than the starting compound (Scheme 3-10). OY N NO P O OR OR N+ B OP(OR)2 X 3-18 3-18+BB= PO2(OR)2Y = H, SiMe3 Blue Green -YX Scheme 3-10 Swager 56 and Zhang have shown that the sensor, in the form of a thin film, responds in seconds to 10 ppm of diisopropylfluorophosphate vapor, which is a less reactive than sarin gas. Visually, the sensor signals the presence of a toxic compound by shifting its fluorescence emission from blue to green. A more favorable color change would be green to red, noting the dye could be modified to optimize the response even if this sensor has the potential to detect nerve agents in the sub-parts-per-billion range as Swager has said. However, this system in solution is not designed to identify the agents it detects. Then, this “turn on” fluorescent detector is going to be used in an imprint

PAGE 74

61 polymer, hence the idea of the monomer 3-24a/b, which was tend to be synthesized by esterification of the biaryl 3-18a/b with 3-11 (Scheme 3-11). 79 OTMS N OTMS N O O OH HO2C 3-18a/b DIC, DMAP + 3-11 3-24a/b CH2Cl2 Scheme 3-11 To obtain the monomer 3-24a/b, the biaryl compound must first be synthesized. For that, a convergent synthesis has been devised (Scheme 3-12). First, naphthalic anhydride 3-25 reacted with mercury acetate (Hg(OAc) 2 ) and acetic acid to give anhydro-8-(hydroxymercuri)-1-naphthoic acid 3-26. The 8-Bromo-1-naphthoic acid 3-27 was obtained by treating 3-26 with NaBr/Br 2 . Successive treatment of 3-27 with SOCl 2 and LiAlH 4 gave the alcohol 3-28. Protection of the alcohol with TMSCl provided the aryl bromide 3-29. O Br OH O O O O O Hg Br TMSO Br OH O Br OH 98% 62% 64% 90% 3-26 3-27 3-28 3-29 3-27 1) AcOH, H2O, 0oC 1) NaOH, H2O Et3N, TMSBr 1) SOCl2 3-25 2) AcOH, Hg(OAc)2 2) NaBr, Br2 2) LiAlH4, Et2O CH2Cl2 Scheme 3-12

PAGE 75

62 In a second part, aromatic substitution of 2-amino-4-methylpyridine with NaNO 3 and NaBr/Br 2 gave the 2-bromo-4-methylpyridine 3-31. N NH2 N Br N Sn N Br N Sn 3-31 3-32a BuLi, Me3SnCl H2O NaNO2, 2) HBr, Br2 1) NaOH aq 3) 95% 72% 3-30 BuLi, Me3SnCl 3-32b 96% 3-33 Et2O, -78oC Et2O, -78oC Scheme 3-13 After substitution of the bromine 3-31 by Me 3 SnCl, 3-32a was obtained. Commercially available 2-Bromo-5-methylpyridine was substituted using the same reagent and gave 3-32b (Scheme 3-13). 81, 82 The choice of substituted pyridine was not random. In order to graft our fluorescent molecule on a monomer it was necessary to have an alkyl group that will be functionalized later on in the synthetic pathway. To obtain a sensor, 3-33a/b, capable of being coupled with the norbornene alcohol, 3-11, an attempt was made to couple 3-32a/b and 3-29 by Stille’s coupling and then oxidize the coupled product with KMnO 4 to get an acid for subsequent esterification (Scheme 3-14). 56 In fact, the presence of a carboxylic acid on one of the aryl shuts down the coupling. Br OTMS N Sn OTMS N CH3 HO2C + 3-32a/b 3-31 3-33a/b 1) Pd(PPh3)4, Toluene 2) KMnO4, H2O Scheme 3-14

PAGE 76

63 Different conditions have been tested in order to get 3-33a/b. The desired cross-coupled product has yet to be successfully obtained. Thus, the future work is the synthesis and optimization of 3-33a/b and the development of a procedure to graft the fluorescent sensor onto the monomer. Analytical Studies With the different monomers (3-24a,b, 3-12) in hand, the MIPs will be made. The monomer carrying the sensor (3-24,b or 3-23) will be mixed with the template in a stoechiometric amount for a few minutes in order to allow bonding. Through this process we will only have the chemosensors in the specific cavities and not all around the polymer. In fact, if sensors are present everywhere in the polymer it will loose the selectivity offered by the MIP. Then norbornene-amine and monomer 3-12 will be added to the mixture. In the model study we used norbornene-carboxilic acid 3-10 as monomer; but in the synthesis of the nerve gas detectors, since the sensors are acid-sensitive, we must use neutral or slightly basic monomers like norbornene-amine. After adding Grubbs’ catalyst the MIP will be ground and the template will be removed. MIPS will be made for every single nerve gas. For every template we will make 2 different polymers: one with the amine as sensor, and the other with the biaryl. Analytical studies will tell us which one provides the best results. Conclusion We report one the first uses of ROMP in molecular imprinting technology, this method of polymerization not only improved the binding properties of the template but also the selectivity of the MIPs. Compared to the usually used radical induced polymerization, our method is faster, uses milder conditions and requires easier work-up. It is now possible to synthesize highly selective MIPs in a few minutes; time-consuming

PAGE 77

64 work-up is no longer needed. Beside, due to the tolerance of Grubbs’ catalyst to a large number of functional groups, our technique provides a larger field of applications. It is now possible to print heat and light sensitive templates. The synthesis of new highly selective nerve gas detectors are in progress in our group. The new MIP technology based on ROMP methodology provides material in a timely manner offering very high selectivity and binding properties.

PAGE 78

CHAPTER 4 NEW SYNTHETIC VECTORS FOR DNA TRANSFECTION BASED ON A PROLINE MOIETY Introduction Gene delivery into cultured cells or in vivo is a promising approach to the treatment of diseases. Several gene delivery systems have been developed to promote gene expression either in vitro or in vivo. Concerns about viral-induced immune response, the risk associated with replication-competent viruses, and production issues have stimulated efforts toward the development of alternative gene delivery systems such as cationic lipids and polymers. These positively charged molecules interact through electrostatic forces with DNA. This results in the formation of highly organized supramolecular structures where DNA molecules are condensed and protected against DNAse degradation. Association of DNA with cationic lipids under a micellar or liposomal form leads to lamellar organization with DNA molecules sandwiched between lipids layers. DNA delivery systems are used in the field of gene therapy to introduce foreign DNA encoding therapeutic protein sequences into cells. Ideally, these DNA delivery systems should: Be safe, non toxic, non-immunogenic, and well-tolerated Contain DNA in small particles; below 500 nm in diameter in order to be compatible with cell penetration through endocytosis Provide a stealthy behavior in order to improve in vivo bioavailability (i.e., prevent the removal of particles from the blood by the monophagocytic system) Target particular cells 65

PAGE 79

66 Penetrate the target cell Escape from the endosomes Protect DNA molecules from extracellular and intracellular DNAse degradation Target the nucleus for transgenic expression The future of gene therapy will depend on the ability to engineer all the properties listed above into DNA delivery systems. Cationic lipid/DNA and cationic polymer/DNA complexes are called lipoplexes and polyplexes, respectively. Despite the extensive use and diverse applications of lipolexes and polyplexes, it must be stressed that their supramolecular characteristics and their mechanisms of transfection are in many ways surprisingly ill-defined. The efficient transfection of eukaryotic cells using cationic liposomes was first described in 1987 by Felgner et al. 83 These cationic liposomes, composed of a cationic lipid {N-[1-(2,3-dioleyloxy)propyl]-N,N,N-trimethylammonium chloride, DOTMA} and a natural neutral phospholipids (dioleoyl phosphatidylethanolamine, DOPE) in a ratio 1:1, were shown to bind DNA efficiently, leading to cellular uptake of plasmid DNA and to high levels of transgene expression. Many synthetic amphiphiles have been synthesized since then that present the common features of vesicles formation in aqueous solutions, DNA binding, and more or less efficient gene transfer. The lipoplexes are usually described by the theoretical charge ratio (express as moles of positive charges per mole of negative charge). Cationic lipids are composed of a hydrophobic moiety (cholesterol, acyl chains, etc.) linked to a cationic headgroup including quartenary ammonium such as dioctadecyl-ammonium bromide (DODAB), or dioleoyl trimethylammonium propane (DOTAP), or a polycation such as a lipopolyamine (RPR120535) or a bisguanidinium trencholesterol (BGTC) (Figure 4-1).

PAGE 80

67 NHNONHNHOHNNH2N+OON+HOOONHONNHH2NNHNHNHH2NDODABDOTAPCl_177171717777 R P R 120535BGTCBrFigure 4-1. Commonly used lipoplexes Our interest was to synthesize a library of a new kind of lipoplexes 4-3 based on a pseudo-proline moiety 4-1. Proline provides a -turn which is thought to be an important feature in the activity (potency) of RGD transfection agent. 35 OO7788OONOOOO4-2 Figure 4-2. Lipoplex precursor

PAGE 81

68 After studying the feasibility of our methodology on a precursor 4-2 (Figure 4-2), 4-1 was submitted to a solid-phase peptidic synthesis to provide synthetic vectors (4-3) bearing the cyclic RGD sequence (Figure 4-3). NCO2HOOOHNOONHOHNNH2NHNHOHOOHNOOOOlipidlipidNOONHOHNNH2NHNHOHOOHNO4-14-3 Figure 4-3. Synthetic pathway to the RGD-based lipoplex We hope that the presence of RGD sequence on our lipoplexes will provide improved selectivity properties. RGD is known to specifically bind to integrin receptors present on fibronectin, so one can assume that the lipoplexes will preferably target those cells. Synthesis of the Proline Moiety Construction of our lipoplexes was based on a chiral proline structure. This structure can be arrived by starting with the commercially available pure enantiomer of trans-4-hydroxy-L-proline. A convergent synthesis was devised to get to the precursor 4-2. A total of 8 steps were required.

PAGE 82

69 Commercially available trans-4-hydroxy-L-proline 4-4 was protected using Fmoc. Adding Fmoc-chloroformate to a mixture of 4-4/DIPEA (diisopropyl ethylamine a.k.a. Hunig’s base) in H 2 O/dioxane gave the Fmoc-trans-4-hydroxy-L-proline derivative 4-5 in 100% yield as a white solid. 84 Treatment of 4-5 with Jones reagent (CrO 3 , H 2 SO 4 ) provided the ketone 4-6, once again with a quantitative yield. Finally, esterification of 4-6 in presence of thionyl chloride in MeOH gave the Fmoc-keto-L-proline methyl ester 4-7 in 60% yield (59% overall yield from trans-4-hydroxy-L-proline) (Scheme 4-1). 85 NCO2HHOHNCO2HOOONCO2HOOOHNCO2MeOOONCO2HOOO4-54-64-4DIPEA, Fmoc-Cl H2O/dioxaneJones, acetoneSOCl2, MeOH4-64-7 Scheme 4-1 Synthesis of the Tartrate Derivative Tartrate derivatives were chosen to graft the lipids (oleyl alcohol) to the proline moiety. The interesting feature of this tartrate derivative is that we can easily create a library of lipoplexes by grafting different kinds of lipids and cationic chains. Commercially available diethyl-L-tartrate 4-8 was protected with an isopropylidene by treatment with dimethoxypropane and TsOH in acetone, giving 4-9 in 82%. Reduction of

PAGE 83

70 the esters of 4-9 in presence of NaBH 4 in MeOH provided the diol 4-10 with 96% yield (Scheme 4-2). OOOOHOOHMeOOMeOOOOOOHOOHOO4-94-104-8TsOH, acetoneNaBH4, MeOH Scheme 4-2 The fatty chains were to be grafted to the diol 4-10 using Williamson’s ether synthesis. Oleyl alcohol 4-11 was treated with DDQ, triphenylphosphine and Bu 4 NBr in CH 2 Cl 2 to give the oleyl bromide 4-12 in 85% yield. 86 Etherification of 4-10 with bromide 4-12 required Bu 4 NBr in order to get a decent yield. In fact the solubility of the oleyl bromide 4-12 was an issue and a phase transfer agent (i.e., Bu 4 NBr) was the easiest solution. The etherification of 4-10 provided the diether 4-13 in 40% yield. Subsequent acidic removal of the acetal, in THF, provided the diol 4-14 in quantitative yield from 4-13 (Scheme 4-3). OOOO4-13 DDQ, PPh3Bu4NBr, CH2Cl2NaH, Bu4NBr, 4-10, THFOHBr787877884-114-12OOOO4-137788OOHOOH4-147788HCl, THF Scheme 4-3

PAGE 84

71 Convergent Synthesis: Formation of the Acetal The final step of the convergent synthesis was to couple the proline derivative 4-7 with the diol 4-14. OOHOOH4-147788NOOOOOOO7788OONOOOOTsOH, toluene +4-74-2 Scheme 4-4 To do so, typical conditions were used: refluxing 4-7 and 4-14 in toluene in presence of a catalytic amount of TsOH and using a Dean-Starke trap. The acetal 4-2 was successfully synthesized in 81% yield (22% overall yield in 8 steps from commercially available starting materials) (Scheme 4-4). Once the methodology proved to be successful, the RGD solid-phase synthesis previously developed in our group was carried out. With this new route, the peptide can be synthesized on a resin and then cleaved. Cyclization and deprotection would be attempted off of the solid-phase support. Use of a solid-phase support allows one to use higher boiling solvents such as DMF since it can be washed off of the bead. 87 By choosing this resin, it was observed that when the peptide was cleaved, the side chain protecting groups remained unaffected. If the protecting groups were removed many problems in isolation could occur, specifically formation of oligomers. Solid-phase

PAGE 85

72 synthesis of the peptide was linear. However, the site of cyclization remained between the C-terminus of glycine and the N-terminus of aspartic acid. Solid-Phase Synthesis of the Cyclic Tetrapeptide The resin chosen was the SASRIN (Super Acid Sensitive Resin) 4-15a. 88, 89 It is a derivative of the WANG resin 4-15b. A major difference, other than subtle structural differences, is the condition of cleavage (Scheme 4-5). 4-15bOOHOOHO4-15a Scheme 4-5 Wang resin 90 requires harsh conditions typically 50% TFA. This is not compatible with our strategy due to potential cleavage of the protecting groups on arginine and aspartic acid. SASRIN resin cleavage requires only 1%TFA in methylene chloride. Side chain protecting groups will survive these conditions. Starting with the SASRIN-Gly-NHFmoc 4-16, the solid phase synthesis was developed. The sequence to the tetrapeptide on the resin involved a number of similar steps. The steps were deprotection of the Fmoc moiety to free the amine under standard conditions with piperidine followed by coupling with diisopropylcarbodiimide (DIC), 1-hydroxybenzotriazole (HOBt), 2,4,6-collidine, and the appropriate amino-acid. The amino groups were protected with Fmoc, which is base labile. The -position of the aspartic acid and the terminal nitrogen of the guanidine side chain of arginine were protected as a t-Butyl ester and acid-sensitive Pbf (where Pbf stands for

PAGE 86

73 2,2,4,6,7-pentamethyldihydrobenzofuran-5-sulfonyl) respectively. Both are acid labile. TFA removes both in one step, leaving the N-protected Fmoc intact. All the reagents in the coupling reactions were used in excess, typically threefold, leading to the tetrapeptide 4-19 on the solid support (Scheme 4-6). OONHF m ocOONHONHF m ocHNNHNHPbfOONHOHNNHNHPbfHNONFmocOOONHOHNNHNHPbfHNONOONHFmocOtBuOONHFmocNHNHPbfHNHONFmocOOHOONHFmocOtBuOHO1) 20% piperidine2) DIC, HOBt1) 20% piperidine2) DIC, HOBt, 2,4,6 collidine1) 20% piperidine2) DIC, HOBt, 2,4,6 collidine4-164-174-184-19 Scheme 4-6 Yields of the steps were all over 95% based solely on the recovery of the bead. Tetrapeptide 4-19 was subjected to mild cleavage conditions: OONHOHNNHPbfHNNHONONHFmocOtBuOONOONH2ONHHNNHP b fNHONHOHOOO 1) 20% piperidine2) 1% TFA, CH2Cl24-194-20 Scheme 4-7

PAGE 87

74 It was treated with 1% TFA in methylene chloride gave 4-20 which was purified via size exclusion chromatography. Linear proline-based RGD 4-20 will be treated with HATU (N,N,N' N'-Tetramethyl-O-(7-azabenzotriazol-1-yl)uranium hexafluorophosphate), HOBt and 2,4,6-collidine to yield 4-21 (Scheme 4-7). Cyclic RGD 4-21 will be treated with a mixture of trifluoroacetic acid, thioanisole, anisole, and water (90:5:3:2). Thioanisole and anisole are used as ion (i.e., pyridinium salts) scavengers. Solvent like 1,2-ethanedithiol can also be added as another trap for the cation but it is unnecessary in this sequence (Scheme 4-8). After purification, the isolated cyclic peptide 4-21 was obtained in 23% yield. NOONH2ONHHNNHP b fNHONHOHOOONOOONHHNNHPbfNHONHOOOHNHATU, HOBt, collidine4-204-21 Scheme 4-8 Solution-Phase Synthesis of the Tetrapeptide Since solid-phase synthesis do not allow for large scale production because of the price of the resin, a solution-phase synthesis was also carried out. A convergent approach was chosen with the synthesis of two separate dipeptide fragments 4-29 and 4-31 (Scheme 4-9).

PAGE 88

75 NHFmocNOOOOOHOOPhONHONH2HNNHNHPbf4-314-29 Scheme 4-9 Cyclization was envisioned between aspartic acid and glycine to avoid steric congestion. From retrosynthetic analysis, C-protected keto-proline 4-22, Fmoc-Arg(Pbf)-OH 4-23, C-protected glycine 4-24 and Fmoc-Asp(OtBu)-OH 4-25 were needed. Fmoc-protected 4-keto-L-proline 4-6 was reacted with thionyl chloride in benzyl alcohol to give Fmoc-4-keto-L-proline benzyl ester 4-26 in 79% yield. Amino-acid 4-26 was then deprotected using 20% piperidine in DMF to yield 4-27 (86%). NCO2HOFmocFmocHNHOOOONCO2BnOFmocNHFmocNOOOOOHONHFmocNOOOOOOPhHNCO2BnO4-64-264-27BnOH, SOCl24-27EDCI, HOBt, Et3N4-254-284-29H2, Pd(OH)220% piperidine Scheme 4-10 Coupling of 4-27 with commercially available Fmoc-Asp(OtBu)-OH 4-25 using the water-soluble carbodiimide EDCI (to ease the purification), HOBt and triethylamine will

PAGE 89

76 yield the protected dipeptide 4-28. Dipeptide 4-28 will be treated with palladium hydroxide under a hydrogen atmosphere (1 atm) to give 4-29 (Scheme 4-10). The second dipeptide was made from commercially available Fmoc-Arg(Pbf)-OH 4-23 and glycine benzyl ester 4-24. The two protected amino-acids were coupled using EDCI, HOBt and triethylamine and yield the dipeptide 4-30 in 71% yield. The Fmoc group was removed using standard conditions with 20% piperidine to give 4-31 in 57% yield (Scheme 4-11). ONHFmocNHNHPbfHNHOOPhONHONHFmocHNNHNHPbfONH2OOPhONHONH2HNNHNHPbfOPhONHONHFmocHNNHNHPbf4-30EDCI, HOBt, Et3N+4-234-244-304-3120% piperidine, CHCl3 Scheme 4-11 Protected amino-acids 4-31 and 4-29, under similar coupling conditions, will give the protected tetrapeptide 4-32. It was decided to deprotect the amine first under standard conditions. Linear peptide 4-20 will be treated with a 20% piperidine solution in DMF followed by hydrogenation with Pearlman’s catalyst to yield the cyclization precursor 4-20. Compound 4-20 will not be purified but taken on to the cyclization step (same procedure that the one used in the solid-phase peptide synthesis): it will be treated with HATU, HOBt in DMF to give 4-21 (Scheme 4-12).

PAGE 90

77 NHFmocNOOOOOHONOONHFmocONHHNNHPbfNHONHOBnOOOOPhONHONH2HNNHNHPbfNOOONHHNNHPbfNHONHOOOHNNOONH2ONHHNNHPbfNHONHOHOOO4-314-324-214-294-20+EDCI, HOBt,1) piperidine, CHCl32) H2, Pd(OH)2HATU, HOBt Scheme 4-12 Synthesis of the DNA Transfection Agent Precursor With the cyclic tetrapeptide in hand, the acetonide can be made by reacting the diol 4-14 bearing the two oleyl moieties with 4-21 in benzene in presence of TsOH in catalytic amount, and a Dean-Starke trap (Scheme 4-13). The acetonide 4-22 will then be obtained and purified. The side chains protecting groups (t-Butyl and Pbf) should survive the acidic treatment. If not, then it wouldn’t be an issue since our goal is to get a non-protected RGD transfection agent (to bind to the integrins). When the agents are made they will be tested in vivo and in vitro to check their biological activity.

PAGE 91

78 NOOONHHNNHPbfNHONHOOOHN4-21OOHOOH4-147788NOONHHNNHPbfNHONHOOOHNOOOO8877TsOH, benzene 4-22 Scheme 4-13 Conclusion The methodology we developed on the proline core was successful, the SPPS of the RGD fragment was achieved with good yield. Meanwhile the solution-phase synthesis needs to be done. The molecule 4-22 would be the first RGD-based DNA transfection agent precursor ever made. It opens a new area of interest in gene therapy and provides an easily way to create libraries thanks to the versatility offered by the acetonide 4-14. Our methodology allows the grafting of different type of groups on the acetonide, leading to a wide library of compounds. Another advantage of the tartrate is its acid sensitivity, allowing the agent to be easily degraded in the body.

PAGE 92

CHAPTER 5 EXPERIMENTAL METHODS General Methods All moisture and air-sensitive reactions were performed in oven dried glassware under a positive pressure of Argon. Sensitive liquids and solutions (e.g., n-butyllithium) were transferred by syringe and introduced into the reaction vessels through rubber septa. All commercial solvents and reagents were used without further purification unless otherwise indicated. Solvents were freshly distilled before use. All yields reported refer to isolated material determined to be pure by NMR spectroscopy and thin layer chromatography. Infrared (IR) spectra were recorded on a Perkin-Elmer 1600 FT-IR spectrometer and are reported in wave numbers (cm -1 ). 1 H Nuclear magnetic resonance spectra were recorded on a Varian Gemini-300 (300 MHz). 13 C NMR spectra were recorded at 75 MHz on the same spectrometers. Chemical shifts are reported in ppm downfield relative to tetramethylsilane (TMS) as an internal standard. All mass spectroscopy was performed by the Mass Spectroscopy Service at the University of Florida-Department of Chemistry. Elemental analysis was performed by Atlantic Microlab Inc. in Norcross, GA. The GC/MS analyses were performed on a Shimadzu GC-17-A/GCMS-QP5000. HPLC analysis were performed on a Shimadzu LC-6 AD with a RID 10 A detector and a CHIRALCEL OD-H from Darcel Inc.. 79

PAGE 93

80 Monobenzylated isosorbide 2-2a. OOOHOBnHH Sodium hydride (60% mass) (15.0 mmol, 0.6 g) was added to a flask flushed with argon. The gray powder was washed three times with pentane to remove the protective oil. A dilute solution of isosorbide 2-1a (13.7 mmol, 2.0 g) in DMF (30 mL) was added to the sodium hydride. The mixture was stirred for 20 minutes. Benzylbromide (14.0 mmol, 1.7 mL) was added to the reaction flask and allowed to stir for 15 hours at ambient temperature. The reaction was monitored by TLC. Workup entailed an extraction with brine and diethyl ether. The organic layer was dried with anhydrous magnesium sulfate and the solution concentrated. The product was isolated by flash chromatography with silica gel (hexane/ether 35:65), leaving a colorless, residual oil (2.7 g, 82%). 1 H NMR (CDCl 3 ) 7.35 (m, 5H), 4.6 (m, 3H), 4.25 (m, 1H), 4.1 (m, 2H), 3.85 (m, 2H), 3.5 (dd, 9.3 Hz, 1H), 2.8 (s, 1H). All spectral data is in agreement with known literature values. 91 Chloro-acetate of isosorbide 2-3a. OOOOBnOClHH The monobenzylated isosorbide 2-2a (17.2 mmol, 4.07 g) was dissolved in chloroform (80 mL). The solution was cooled to 0 o C in an ice-bath and allowed to stir

PAGE 94

81 for 10 minutes. After addition of pyridine (60.3 mmol, 4.9 mL), the reaction was stirred at 0 o C for 15 minutes. In a seperate vessel, chloroacetic anhydride (34.5 mmol, 5.9 g) was dissolved in chloroform (70 mL). This solution was added dropwise to the above reaction mixture. The reaction was stirred at room temperature for 12 hours. After consumption of starting material, dilute HCl was added to neutralize the pH. Workup entailed an extraction with brine and chloroform. The organic layer was dried with anhydrous magnesium sulfate and the solution was concentrated. The product was isolated by flash chromatography on silica gel (hexane/ether 45 : 55), leaving a residual oil (4.5 g, 84%). 1 H NMR (CDCl 3 ) 7.35 (m, 5H), 5.25 (d, 3.5 Hz, 1H), 4.75 (d, 11.9 Hz, 1H), 4.67 (t, 4.6 Hz, 1H), 4.55 (d, 11.9 Hz, 1H), 4.5 (m, 1H), 4.1 (m, 5H), 3.86 (dd, 8.8 Hz, 1H), 3.65 (dd, 8.7 Hz, 1H); 13C NMR (CDCl 3 ) 166.3, 137.5, 128.4, 127.8, 127.7, 85.4, 80.5, 80.0, 78.8, 73.2, 72.3, 70.4, 40.5 ; IR (neat): 2877.7, 1757.8, 1174.1, 699.8 cm -1 ; HRMS calcd for [C 15 H 17 O 5 Cl] + 312.0764, found 312.0764; Anal. Calcd for C 15 H 17 O 5 Cl: C, 57.61; H, 5.48. Found C, 57.73; H, 5.49. Wittig reagent 2-4a OOOOBnOPPh3HH Ester 2-3a (39.4 mmol, 12.3 g) was mixed with triphenylphosphine (118.2 mmol, 31.0 g) and benzene (50 mL) and allowed to reflux for 15 hours. The resulting solution was cooled to room temperature. Hot water was added to dissolve the salt formed in the flask. The benzene layer was discarded. Then 0.1 M NaOH was added dropwise until no

PAGE 95

82 further white precipitate was formed. More precipitate was observed upon cooling of the flask in an ice-bath for 1 hour. Workup entailed an extraction with brine and methylene chloride. The organic layer was dried with anhydrous magnesium sulfate and the solution was concentrated. The product was thoroughly dried by a vacuo pump (17.2 g, 81%). 1 H NMR (CDCl 3 ) 7.63 (m, 5H), 7.42 (m, 10H), 7.29 (m, 5H), 5.1 (d, 12 Hz, 1H), 4.77-4.4 (m, 4H), 4.07-3.4 (m, 6H), 1.98 (d, 13 Hz, 1H); 13 C NMR (CDCl 3 ) 166.3, 137.5, 128.4, 127.8, 127.7, 85.4, 80.5, 80.0, 78.8, 73.2, 72.3, 70.4, 40.5 ; IR (neat): 3048.5, 2954.5, 2355.1, 1619.3, 1434.3, 1190.4, 1108.1, 884.8, 743.8, 696.8 cm-1; HRMS calcd for [C 33 H 32 PO 5 ] + 538.1909, found 538.1987. Isosorbide Wittig adduct 2-10a. OOOOOBnOHH Freshly distilled glutaric dialdehyde (14.64 mmol, 465 mg) and Wittig reagent from isosorbide 2-4a (5.57 mmol, 3 g) were combined in an oven dried flask and dissolved in freshly distilled methylene chloride (1 M). The reaction was stirred for 15 hours at room temperature. After completion of the reaction, the solvent was evaporated and the remaining residue was purified with column chromatography (hexane/ether 50:50) (1.75 g, 75%). 1 H NMR 1.75 (quint, J=6.4 Hz, 2H), 2.2 (q, J=6.4 Hz, 2H), 2.45 (dt, J=6.4, 8.9 Hz, 2H), 3.4-4.08 (m, 7H), 4.1 (q, J=6.5 Hz, 1H), 4.8 (t, J=6.3 Hz, 1H), 5.15 (q, J=6.4 Hz, 1H), 5.82 (d, J=12.7 Hz, 1H), 6.9 (dt, J=8.9,12.7 Hz, 1H), 7.12 (m, 5H), 9.7 (s, 1H). HRMS calcd for [C 20 H 24 O 6 ] + 360.1573, found 360.1575.

PAGE 96

83 Monobenzylated isomannide 2-2b. OOOHOBnHH Isomannide 2-1b (6.85 mmol, 1.0 g), potassium hydroxide (6.85 mmol, 0.44 g) were dissolved in H 2 O (3.5 mL) and the resulting solution was heated to reflux for 20 minutes. The mixture was cooled to room temperature, benzyl chloride was added dropwise (6.85 mmol, 0.78 mL). The solution was refluxed for additional 3 hours before an acidic quench (HCl 2N, 3.5 mL) and extraction with ethyl acetate (3 X 2mL). The combined organic layers were dried (MgSO 4 ) and concentrated under vacuo. The crude product was then precipitated in cold Et 2 O (5 mL) to give the monobenzylated isomannide 2-2b (0.73 g, 45%) as a white solid. Mp=93 o C. 1 H NMR 2.85 (d, J=8.5 Hz, 1H), 3.74.80 (m, 2H), 4.02.05 (m, 2H), 4.08 (dd, J=4.8, 8.1 Hz, 1H), 4.29 (dq, J=5.5, 8.5 Hz, 1H), 4.50 (dd, J=4.6, 5.5 Hz, 1H), 4.57 (dd, J=4.6, 4.8 Hz, 1H), 4.58 (d, J=11.8 Hz, 1H), 4.78 (d, J=11.8 Hz, 1H), 7.30.40 (m, 5H); 13 C NMR 71.4, 72.3, 72.6, 74.9, 78.9, 80.6, 81.7, 128.0, 131.0; IR (max, cm-1, film) 1220, 2850, 3400; [] D 20 =+138.0 (c=1, CHCl 3 ); Anal. Calcd for C 13 H 16 O 4 : C, 66.09; H, 6.83. Found: C, 65.69; H, 6.76.

PAGE 97

84 Chloro-esterified isomannide 2-3b. OOOOBnOClHH The monobenzylated isomannide 2-2b (17.2 mmol, 4.07 g) was dissolved in chloroform (80 mL). The solution was cooled to 0 o C in an ice-bath and allowed to stir for 10 minutes. After addition of pyridine (60.3 mmol, 4.9 mL), the reaction was stirred at 0 o C for 15 minutes. In a separate vessel, chloroacetic anhydride (34.5 mmol, 5.9 g) was dissolved in chloroform (70 mL). This solution was added dropwise to the above reaction mixture. The reaction was stirred at room temperature for 12 hours. After consumption of starting material shown by TLC, dilute HCl was added to neutralize the pH. Workup entailed an extraction with brine and chloroform. The organic layer was dried with anhydrous magnesium sulfate and the solution was concentrated. The product was isolated by flash chromatography on silica gel (hexane/ether 45 : 55), leaving a residual oil (4.5 g, 84%). [] 25 D = +77.6 o (C=0.105, chloroform), 1 H NMR (CDCl 3 ) 7.35 (m, 5H), 5.25 (d, J=3.5 Hz, 1H), 4.75 (d, J=11.9 Hz, 1H), 4.67 (t, J= 4.6 Hz, 1H), 4.55 (d, J=11.9 Hz, 1H), 4.5 (m, 1H), 4.1 (m, 5H), 3.86 (dd, J=8.8 Hz, 1H), 3.65 (dd, J=8.7 Hz, 1H); 13 C NMR (CDCl 3 ) 166.3, 137.5, 128.4, 127.8, 127.7, 85.4, 80.5, 80.0, 78.8, 73.2, 72.3, 70.4, 40.5 ; IR (neat): 2877.7, 1757.8, 1174.1, 699.8 cm-1; MS (EI) m/z 312 (M + ), 234, 183, 136, 112, 107, 91; HRMS calcd for [C 15 H 17 O 5 Cl] + 312.0765, found 312.0769; Anal. Calcd for C 15 H 17 O 5 Cl: C, 57.61; H, 5.48. Found: C, 57.73; H, 5.49.

PAGE 98

85 Wittig reagent 2-4b. OOOOBnOPPh3HH Ester 2-3b (39.4 mmol, 12.3 g) was mixed with triphenylphosphine (118.2 mmol, 31.0 g) and THF (50 mL) and allowed to reflux for 15 hours. The resulting solution was cooled to room temperature. Hot water was added to dissolve the salt formed in the flask. The THF layer was disregarded. Then 0.1 M NaOH was added dropwise until no further white precipitate was formed. More precipitate was observed upon cooling of the flask in an ice-bath for 1 hour. Workup entailed an extraction with brine and methylene chloride. The organic layer was dried with anhydrous magnesium sulfate and the solution was concentrated. The product was thoroughly dried by a vacuo pump (17.2 g, 81%). 1 H NMR (CDCl 3 ) 7.63 (m, 5H), 7.42 (m, 10H), 7.29 (m, 5H), 5.1 (d, J=12 Hz, 1H), 4.77-4.4 (m, 4H), 4.07-3.4 (m, 6H), 1.98 (d, J=13 Hz, 1H); 13 C NMR (CDCl 3 ) 166.3, 137.5, 128.4, 127.8, 127.7, 85.4, 80.5, 80.0, 78.8, 73.2, 72.3, 70.4, 40.5 ; IR (neat): 3048.5, 2954.5, 2355.1, 1619.3, 1434.3, 1190.4, 1108.1, 884.8, 743.8, 696.8 cm-1; HRMS calcd for C 33 H 32 PO 5 538.1909, found 538.1987. Isomannide Wittig adduct 2-10b. OOOOOHHBnO

PAGE 99

86 Freshly distilled glutaric dialdehyde (14.64 mmol, 465 mg) and Wittig reagent from isomannide 2-4b (5.57 mmol, 3 g) were combined in an oven dried flask and dissolved in freshly distilled methylene chloride (1 M). The reaction was stirred for 15 hours at room temperature. After completion of the reaction shown by TLC, the solvent was evaporated and the remaining residue was purified with column chromatography (hexane/ether 50:50) (1.75 g, 75%). 1 H NMR 1.75 (quint, J=6.4 Hz, 2H), 2.2 (q, J=6.4 Hz, 2H), 2.45 (dt, J=6.4, 8.9 Hz, 2H), 3.4-4.08 (m, 7H), 4.1 (q, J=6.5 Hz, 1H), 4.8 (t, J=6.4 Hz, 1H), 5.15 (q, J=6.4 Hz, 1H), 5.82 (d, J=12.7 Hz, 1H), 6.9 (dt, J=8.9,12.7 Hz, 1H), 7.12 (m, 5H), 9.7 (s, 1H). HRMS calcd for C 20 H 24 O 6 360.1573, found 360.1575. Tetrahydropyran-2-ol 25. OO H Method 1: 3-4 Dihydropyran (10.93 mmol, 1 mL) was stirred in THF (22 mL) at room temperature under Argon pressure. To the flask was added water (55 mmol, 1 mL) then some drops of concentrated HCl. The reaction was stirred at room temperature for 2 hours and monitored by TLC. Workup entailed an extraction with diethyl ether. The organic layer was dried with anhydrous magnesium sulfate and the solution concentrated under vacuo leaving a residual colorless oil (949 mg, 85%). Method 2: Valerolactone (23.7 mmol, 2.2 mL) was stirred in dry methylene chloride (50 mL) at o C under Argon pressure. To the flask was added Dibal-H (diisobutylaluminum hydride, 1 M in hexane) (26 mmol, 26 mL) dropwise via a syringe over 10 minutes. The reaction was stirred 10 more minutes and monitored by TLC. Then, the reaction was quenched slowly at o C with methanol (5 mL) and the solution was

PAGE 100

87 poured into aqueous saturated Rochelle’s salt solution (300 mL). The mixture was stirred for 1hour then extracted with diethyl ether. The organic layer was dried with anhydrous magnesium sulfate and the solution concentrated under vacuo leaving residual colorless oil (2.37 g, 98%). All spectral data was in agreement with known literature values. 92 (Methoxycarbonylmethylene)triphenylphosphorane 2-6. Ph3PCO2Me Methyl bromoacetate (31.69 mmol, 3 mL) was added to triphenylphosphine (95.1 mmol, 25 g) already dissolved in ethyl acetate (500 mL) at room temperature. The mixture was stirred overnight. The white precipitate was filtered, washed with diethyl ether and dried under vacuo at 40 o C for half an hour. The solid was redissolved in hot water. NaOH was added and ylide 2-6 precipitated (10.4 g, 98%). All spectral data was in agreement with known literature values. 93 7-Hydroxyhept-2-enoic acid methyl ester 2-7. HOCO2Me Tetrahydropyran-2-ol 2-6 (17.1 mmol, 1.75 g) was stirred in chloroform (17 mL) at room temperature. To the flask was added the ylide 6 (25.65 mmol, 8.6 g) and a tip of benzoic acid (10 mg). The reaction ran overnight. Workup entailed an extraction with diethyl ether. The organic layer was dried with anhydrous magnesium sulfate and the solution concentrated under vacuo and triphenylphosphine oxide was removed by flash chromatography (hexane/ether 50:50) leaving a residual light yellow oil. (2.1 g, 76%). All spectral data is in agreement with known literature values. 94

PAGE 101

88 Trans-7-oxohept-2-enoic acid methyl ester 2-8. OCO2MeH 7-Hydroxyhept-2-enoic acid methyl ester 2-7 (12.7 mmol, 2 g) was stirred in freshly distilled methylene chloride (25.4 mL) at room temperature. To the flask was added Celite (12 g) then PDC (19 mmol, 7.1 g) to prevent caking. The mixture was stirred overnight. Workup entailed a dilution with ether, filtration and concentration under vacuo. The trans compound was isolated by flash chromatography on silica gel (hexane/ether 50:50), leaving a white powder (85%). All spectral data was in agreement with known literature values. 95 General Procedure for Radical Cyclizations. Aldehyde 2-8 (0.13 mmol, 0.06 g) and methylene chloride (0.5 M) were added to an argon flushed, oven-dried, round bottom flask. The Lewis acid (1 equiv.) was also added to the reaction vessel and allowed to stir under argon at room temperature for 15 minutes. The reaction mixture was then cooled to o C with a dry-ice/acetone bath. Tributyltin hydride (0.63 mmol, 0.17 mL) and triethylborane (1M in hexane) (0.32 mmol, 0.32 mL) were then added to the solution in that order. After all the reagents were allowed to stir at o C for 10 minutes, oxygen and triethylborane (1M in hexane) (0.32 mmol, 0.32 mL) was injected into the mixture for five minutes intervals every 30 minutes. The reaction was monitored by TLC until completion. Workup entailed an extraction of the crude mixture with ethyl acetate and sodium carbonate saturated solution. The combined organic layers were dried and concentrated under vacuo. The

PAGE 102

89 mixture was applied to a pad of potassium fluoride on top of a column of silica gel. After leaving the residue on the pad for thirty minutes, the cyclized product was eluted (30:60 ether:hexane) yielding a clear oil. Product 2-11a,a’ (3R, 4S). OOOOHHOHBnO [] D 25 =+125.4 (c=0.1, MeOH); 1 H NMR 1.21 (m, 1H), 1.58 (m, 2H), 1.71 (m, 1H), 1.94 (m, 2H), 2.11 (m, 1H), 2.38 (dd, J=16.5, 7.8 Hz, 1H), 2.41 (ddd, J=16.5, 6.3, 1.0 Hz, 1H), 2.60 (br s, 1H), 3.82 (m, 8H), 4.23 (q, J=6.3 Hz, 1H), 4.81 (t, J=6.3 Hz, 1H), 5.12 (q, J=6.6 Hz, 1H), 7.23 (m, 5H); 13 C NMR (CDCl 3 ) 172.8, 137.2, 128.4, 127.6, 83.3, 79.0, 76.8, 75.5, 74.1, 72.2, 64.9, 64.2, 36.5, 32.0, 30.7, 23.7, 17.2. HRMS calcd for [C 20 H 26 O 6 ] + 362.1729, found 362.1736. Anal. Calcd for C 20 H 26 O 2 : C, 66.28; H, 7.23. Found: C, 66.55; H, 7.46. Product 2-11b,b’ (3R, 4S). OOOOHHOHBnO [] D 25 =+137.6 o (c=0.1, MeOH); 1 H NMR 1.16-1.36 (m, 1H), 1.54-164 (m, 2H), 1.67-1.77 (m, 1H), 1.91-1.99 (m, 2H), 2.03-2.13 (m, 1H), 2.39 (dd, J=16.3, 8.0 Hz, 1H), 2.45 (ddd, J=16.3, 6.1, 1.0 Hz, 1H), 2.51 (br s, 1H), 3.40-4.08 (m, 7H), 3.85 (m, 1H),

PAGE 103

90 4.20 (q, J=6.5 Hz, 1H), 4.85 (t, J=6.4 Hz, 1H), 5.15 (q, J=6.4 Hz, 1H), 7.14 (m, 5H); 13 C NMR (CDCl 3 ) 172.0, 137.2, 128.4, 127.6, 82.9, 79.2, 76.4, 75.6, 73.8, 72.2, 65.4, 64.6, 36.1, 31.8, 30.7, 23.5, 16.9 HRMS calcd for [C 20 H 26 O 6 ] + 362.1729, found 362.1732. Anal. Calcd for C 20 H 26 O 2 : C, 66.28; H, 7.23. Found: C, 66.44; H, 7.44. 2-12 and 2-12’. OO 1 H NMR 1.55 (dddt, J=1.0, 4.1, 13.0, 4.1 Hz, 1H), 1.68-1.77 (m, 3H), 1.86 (m, 1H), 2.05 (td, J=4.4, 14.8 Hz, 1H), 2.29 (dd, J=2.5, 18.1 Hz, 1H), 2.82 (dd, J=10.2, 18.1 Hz, 1H), 2.88-2.96 (m, 1H), 5.00 (td, J=1.5, 5.9 Hz, 1H); 13 C NMR (CDCl 3 ) 23.35, 33.43, 33.56, 35.93, 37.89, 86.32, 177.66; MS (EI) m/z 126 (M + ), 98, 97, 83, 80, 67, 54 HRMS (EI) calcd for [C 7 H 10 O 2 ] + 126.0681, found 126.0678. 2-13 (3R, 4S). RMeOOOHS [] D 25 =+43.1 (c=1, MeOH); 1 H NMR 1.16-1.3626 (m, 1H), 1.54-164 (m, 2H), 1.67-1.77 (m, 1H), 1.91-1.99 (m, 2H), 2.03-2.13 (m, 1H), 2.39 (dd, J=16.3, 8.0 Hz, 1H, CH 2 CO), 2.45 (ddd, J=16.3, 6.1, 1.0 Hz, 1H, CH 2 CO), 2.51 (br s, 1H, OH), 3.66 (s, 3H, OMe), 3.85 (m, 1H, CHOH); 13 C NMR (CDCl 3 ) 21.83, 30.70, 34.26, 38.17, 44.35, 51.77, 78.85, 174.64; MS (EI) 159(MH + ), 141, 127, 109, 81

PAGE 104

91 General procedure for radical polymerization Pyridostigmine bromide (0.75 mmol, 195.8 mg) was added to a round bottom flask, methacrylic acid (30.0 mmol, 0.28 mL) was poured and the mixture was allowed to stir for 10 minutes under argon. Ethylene glycol (12.0 mmol, 2.25 mL), AIBN (0.17 mmol, 28 mg) and distilled acetonitrile (3.0 mL) were added. The mixture is refluxed for 8 hours. The white solid obtained was then ground, suspended in ethanol and refluxed overnight to remove the template (pyridostigmine). The powder was dried under vacuo to give the polymer. General procedure for ROMP polymerization: Pinacolyl methylphosphonate (0.76 mmole, 138.0 mg), bicyclo[2.2.1]hept-5-en-2-carboxylic acid (3.06 mmoles, 422.0 mg) and hexanedioic acid dibicyclo[2.2.1]hept-5-en-2-ylmethyl ester 3-12 (12.3 mmoles, 4.4 g) were dissolved in CH 2 Cl 2 (31 mL). After waiting 30 minutes for electrostatic bonding to occur between monomers and templates, Super Grubbs’s catalyst (0.09 mmole, 81.3 mg) was added and the solution was allowed to stir until a solid formed. In few minutes, the solution became a soft gelatin structure with a pink color. To remove Grubbs’s catalyst, the solid was washed up with CH 2 Cl 2 (100 mL) and ethyl vinyl ether after grinding. The solution was stirred 15 minutes and filtered. To remove the excess of ethyl vinyl ether and monomers, the solution was washed with EtOAc and filtered. Then to remove the template, the mixture was refluxed overnight with ethanol, filtered and the solvent was removed under reduce pressure. Bicyclo[2.2.1]hept-5-en-2-yl-methanol or 5-Norbornene-2-methanol 3-11. OH

PAGE 105

92 Aldehyde (1.23 mol, 15 g, 14.9 mL) was dissolved in methanol (150 mL) and the mixture was cooled at 0 o C. Sodium borohydride (0.49 mol, 18.6 g) was added and the solution was stirred overnight. A solution of saturated solution of ammonium chloride was added and the solution was filtered: the white solid was washed with water. The aqueous layer was then extracted with chloroform. The solvent was removed in vacuo and the crude product 3-11 (14.9 g, 98%) was used without further purification. 1 H (CDCl 3 ) 6.18 (m, 1H), 5.96 (m, 1H), 3.38-3.25 (m, 2H), 2.95 (m, 1H), 2.82 (m, 1H), 1.82 (m, 1H), 1.47 (m, 2H), 1.38 (m, 2H); 13 C (CDCl 3 ) 136.93, 136.40, 136.23, 132.01, 66.72, 66.74, 49.22, 44.65, 43.32, 43.03, 41.95, 41.36, 41.26, 41.20, 29.34, 28.62. Hexanedioic acid dibicyclo[2.2.1]hept-5-en-2-ylmethyl ester 3-12. OOOO Norbornene alcohol 3-11 (0.81 mmol, 100 mg) was dissolved in freshly distilled methylene chloride (2 mL), pyridine (2.42 mmol, 191.4 mg, 0.2 mL) was added at room temperature and the solution was allowed to stir for 30 minutes. Then, the reaction was cooled at 0 o C and adipoyl chloride (0.885 mmol, 0.13 mL) was added slowly. The solution was stirred overnight. Brine was added to the mixture and the aqueous layer was extracted with methylene chloride, the organic layers were then dried over magnesium sulfate. The solvent was removed in vacuo and the crude material was purified by flash chromatography with ether/hexane (50/50) to give a clear oil (267 mg, 92%). 1 H (CDCl 3 ) 6.14 (dd, J= 2.6 Hz, J= 3.0 Hz, 2H), 5.93 (m, 2H), 3.82 (dd, J= 6.6 Hz, J= 10.8 Hz, 2H),

PAGE 106

93 2.84 (d, J= 15 Hz, 4H), 2.34 (m, 7H), 1.82 (m, 2H), 1.64 (m, 4H), 1.42 (dq, J= 7.5 Hz, J= 2.5 Hz, 2H), 1.26 (m, 5H); HRMS for [C 22 H 31 O 4 +2H] + calc. 360.2144, found 360.2251. 1-Anthracen-9-ylmethyl-piperidine-4-carboxylic acid methyl ester 3-22 N O OMe Isonipecotic acid (38.7 mmol, 5.0 g) was dissolved in MeOH (125 mL) and HCl gas was bubbled through the solution for 4 hours. The solvent was removed, the resulting white solid was suspended in CH 2 Cl 2 and ammonia was bubbled through the mixture for 20 minutes. The mixture was filtered and the residue washed with CH 2 Cl 2 , and the mother liquor was then concentrated to yield the methyl ester of isonipecotic acid as a clear oil 3-20 (4.7 g, 82%). The methyl ester 3-20 (33.9 mmol, 4.9 g) and 9-anthraldehyde 3-21 (33.9 mmol, 7.0 g) were dissolved in THF (135 mL) and 4 powdered molecular sieves (8 g) were added to the solution. After stirring for 30 minutes, NaCNBH 3 (48.5 mmol, 3.0 g) was added and the mixture was allowed to stir overnight. The solvent was evaporated and the residue was dissolved in CH 2 Cl 2 (170 mL), washed with 1N HCl (3 x 90 mL), saturated NaHCO 3 (3 x 90 mL) and dried over Na 2 SO 4 . Evaporation of solvent yielded sensor 3-22 as a pale yellow crystalline solid (3.6 g, 72%). 1 H (CDCl 3 ) 8.50 (d, J= 9.2 Hz, 2H), 8.42 (s, 1H), 8.01 (d, J= 8.4 Hz, 2H), 7.54-7.45 (m, 4H), 4.41 (s, 2H), 3.66 (s, 3H), 2.94 (m, 2H), 2.38-2.27 (m, 3H), 1.85-1.81 (m, 2H), 1.75-1.66 (m, 2H); 13 C (CDCl 3 ) 175.7, 131.3, 129.9, 128.9, 127.3, 125.5, 124.8, 54.5, 53.0, 51.5, 41.2, 28.4; IR (CH 2 Cl 2 ) 3425,

PAGE 107

94 3355, 1740, 1670, 1672, 1636; HRMS (ES) for [C 22 H 23 NO 2 +H] + calc. 334.1807, found 334.1802. 1-Anthracen-9-ylmethyl-piperidine-4-carboxylic acid 3-17 N O OH .HCl To a solution of sensor 3-22 (23.5 mmol, 8.7 g) in THF/MeOH (260 mL / 104 mL) was added aqueous NaOH solution (516 mmol, 20.6 g, dissolved in 46 mL of H 2 O). The solution was stirred for 4 hours and the solvent was removed in vacuo. The residue was dissolved in H 2 O (195 mL), and the pH was adjusted to <2 using 1N HCl which induced precipitation. The solids were collected, washed with cold (0 o C) H 2 O and dried over MgSO 4 to afford sensor 3-17 (6.5 g, 75%) as the HCl salt. 1 H (DMSO-d 6 ) 12.09 (br s, 1H), 8.55 (s, 1H), 8.47 (d, J= 8.8 Hz, 2H), 8.06 (d, J= 8.4 Hz, 2H), 7.56-7.48 (m, 4H), 4.37 (s, 2H), 2.79 (m, 2H), 2.24 (m, 3H), 1.72 (m, 2H), 1.42 (m, 2H); 13 C (DMSO-d 6 ) 176.1, 130.9, 130.8, 129.8, 128.8, 127.1, 125.7, 125.1, 125.0, 53.6, 52.5, 40.3, 28.2; IR (film) 3389, 3049, 2930, 2848, 1703, 1571, 1406; HRMS (ES) for [C 21 H 21 NO 2 Na] + calc. 342.1470, found 342.1474. 1-Anthracen-9-ylmethyl-piperidine-4-carboxylic acid bicyclo[2.2.1]hept-5-en-2-ylmethyl ester 3-23 N O O

PAGE 108

95 1-Anthracen-9-ylmethyl-piperidine-4-carboxylic acid 3-17 (0.56 mmol, 200.0 mg) and DIC (N, N’, diisopropylcarbodiimide) (1.75 mmol, 277.1 mg) in CH 2 Cl 2 (1 mL) were cooled to 0 o C. DMAP (4-dimethylaminopyridine) (0.23 mmol, 28.6 mg) was added and the solution was stirred for 30 minutes at this temperature. 5-Norbornene-2-methanol 3-11 (0.58 mmol, 72.5 mg) was added dropwise to the solution and warmed to room temperature. The reaction was monitored by TLC (hexane/EtOAc, 65:35). After completion, the volatiles were removed under reduced pressure. The resulting residue was dissolved in CH 2 Cl 2 and hexane (1:1, 100 mL). The solid was removed via filtration and the filtrate was evaporated leaving 3-23 (197.7 g, 83%). Anhydro-8-(hydromercuri)-1-naphthoic acid or 2-Oxa-1-mercura-phenalen-3-one 3-26 O O Hg 1,8-Naphthalic anhydride 3-25 (0.13 mole, 25.0 g) was suspended in aqueous NaOH (0.46 mol, 18.5 g, in 792 mL of H 2 O) and refluxed until the solid material dissolved. The excess base was neutralized with glacial CH 3 COOH (50 mL), and a solution of mercuric acetate (0.132 mol, 42.0 g) was added in one portion. After the mixture was refluxed 30 min, additional glacial CH 3 COOH (90 mL) was added to the white slurry, resulting in slow evolution of carbon dioxide. The slurry was refluxed for 48 h, cooled and filtered. The highly insoluble solid was washed with H 2 O and then dried under vacuo at 105 o C overnight to give 3-26 as a tan storable powder (48.2 g, 98%). All spectral data was in agreement with known literature values. 96

PAGE 109

96 8-Bromo-1-naphthoic acid or 8-Bromonaphthalene-1-carboxylic acid 3-27 O OH Br Anhydro-8-(hydroxymercuri)-l-naphthoic acid 3-26 (0.13 mol, 48.2 g) suspended in glacial acetic acid (190 mL) and H 2 O (30 mL) was stirred vigorously at 0 o C. NaBr (0.42 mol, 42.8 g) in H 2 O (155 mL) and Br 2 (0.134 mol, 21.4 g, 6.9 mL) were added slowly while the reaction temperature was maintained at 0-5 o C. The resulting slurry was then slowly heated to 100 o C and poured on ice (380 g). The precipitate was washed with H 2 O, dissolved in hot aqueous NaOH (0.75 mol, 30.2 g, in 63 mL of H 2 O), and filtered through Celite. When the filtrate was acidified with concentrated HCl and cooled, acid 3-27 was obtained as white crystals (20.1 g, 62%). Mp 174-175 o C (lit. 174-175 o C). All spectra data are in accord with literature. 97 (8-Bromo-1-naphthyl) methanol or (8-Bromonaphthalen-1-y1)methanol 3-28 Br OH A solution of naphthoic acid 3-27 ( 0.08 mol, 20.0 g) and SOCl 2 (0.84 mol, 100.0 g, 62.0 mL) was refluxed 6 hours and concentrated under reduced pressure. Crystallization of the residue from hexane yielded 8-bromo-1-naphthoyl chloride (18.5 g, 86%). O (58 mL) was added to LiAlH 8-Bromo-1-naphthoyl chloride (0.07 mol, 18.5 g) in Et 2 4 (0.054 mol, 2.0 g) suspended in Et 2 O (58 mL). After the addition was complete, the mixture was refluxed 5 hours, cooled, and hydrolyzed with saturated aqueous NaSO . 4

PAGE 110

97 When the ether layer was dried over MgSO 4 and concentrated at reduced pressure, the residue obtained yielded 3-28 as a pale yellow solid (10.2 g, 64%). HRMS for [C 11 H 9 BrO] calc. 235.9837 found 235.9834. All spectra data were in accord with literature. 80 (8-Bromo-naphthalen-1-ylmethoxy)-trimethyl-silane 3-29 Br TMSO To (8-bromo-naphthy1) methanol 3-28 (3.16 mmol, 0.7 g) dissolved in CH 2 Cl 2 (3.5 mL) was added Et 3 N (4.00 mmol, 0.4 g, 0.56 mL) and let stir 30 minutes. Then TMSCl (1.2 eq., 3.67 mmol, 0.4g, 0.47 mL) was added dropwise at 0 o C and let stir 4 hours at room temperature. The resulting solution was neutralized with NH 4 Cl, extracted with Et 2 O (3 x 20 mL), washed with H 2 O (3 x 20 mL) and dried over MgSO 4 . After filtration and evaporation of the solvent, a brownish liquid 3-29 was obtained (0.9 g, 90%). 1 H (CDCl 3 ) 8.76-7.26 (m, 6H), 5.67 (s, 2H), 0.22 (s, 9H); HRMS for [C 14 H 17 OSiBr] + calc. 308.0232, found 308.0234. 2-Bromo-4-methylpyridine 3-31 N Br 2-Amino-4-methylpyridine (9.25 mmol, 1.0 g) was added over 10 minutes to cold (-5 o C) aqueous HBr (51.8 mmol, 4.2 g, 6.1 mL). With stirring and external cooling, Br 2 (27.8 mmol, 4.4 g, 3.0 mL) was added slowly enough to allow the temperature to remain below 0 o C. When the addition was complete, the dark red precipitate was treated with a

PAGE 111

98 solution of NaNO 2 (25.0 mmol, 1.7 g) in H 2 O (3 mL) that was added at such as a rate as to keep the temperature of the reaction mixture below 5 o C. After stirring for one hour while the temperature gradually rose to about 20 o C, a solution of NaOH (94.3 mmol, 3.8 g) in H 2 O (5 mL) was added dropwise with external cooling at a rate such that the temperature of the reaction mixture did not rise above 25 o C. When the addition was complete, the reaction mixture was extracted with Et 2 O (3 x 25 mL), The combined ether extracts were washed with brine (3 x 20 mL), dried over anhydrous MgSO 4 , filtered and concentrated at reduced pressure. Two distillations yielded the bromopicoline 3-31 (1.5g, 95%). 1 H (CDCl 3 ) 8.38 (d, J= 5.8 Hz, 1H), 7.48 (d, J= 0.6 Hz, 1H), 7.25 (dd, J= 5.8 Hz, J= 0.6 Hz, 1H), 2.52 (s, 3H); 13 C (CDCl 3 ) 150.2, 149.6, 142.0, 128.4, 123.6, 20.6; HRMS for [C 6 H 6 NBr+2H] + calc. 172.9840 found 172.9665. Methyl-2-trimethylstannylpyridine 3-32a,b] N Sn 3-32a N Sn 3-32b 2-Bromo-4-methylpyridine 3-31 (5.8 mmol, 1.0g) was charged in a flask, evacuated and put under argon. Freshly distilled THF (11 mL) was added from a syringe under stirring, and the solution was cooled to -78 o C in a dry ice/acetone bath. n-BuLi (2.5 M in cyclohexane, 5.8 mmol, 2.35 mL) was added dropwise from a syringe, and the reaction mixture was stirred for 1/2 h under continued cooling. Then, trimethylstannyl chloride (1.0 M in THF, 6.4 mL, 6.4 mmol) was added dropwise from a syringe. After 1

PAGE 112

99 h, the reaction mixture was allowed to warm slowly to room temperature overnight. Saturated NH 4 Cl was added and the aqueous layer was extracted with ethyl ether, the combined organic layers were dried and concentrated in vacuo. The crude yellow oil was purified on alumina to yield 3-32a (1.3 g, 72%) as a clear oil. Yield for the product 3-32b: 90%. 3-32a: 1 H (CDCl 3 ) 7.37 (m, 3H), 2.31 (s, 3H), 0.58 (s, 9H). 3-32b: 1 H (CDCl 3 ) 7.37 (m, 3H), 2.31 (s, 3H), 0.58 (s, 9H); HRMS for [C 9 H 15 NSn] + calc. 257.0226, found 257.0228. 2-Bromoisonicotinic acid and 2-Bromonicotinic acid 3-33 N Br COOH N Br HOOC 2-Bromomethylpyridine 3-32a or 3-32b (9.3 mmol, 1.6 g) was suspended in H 2 O (32 mL) and KMnO 4 (21.3 mmol, 3.4 g) was added in one portion. The purple-brown solution was refluxed for 6 hours and then an additional portion of KMnO 4 (1.9 mmol, 0.3 g) was added and refluxed for one more hour. The hot solution was filtered with vacuum and the residue washed with boiling H 2 O (10 mL). The cooled aqueous solution was extracted with Et 2 O (2 x 5 mL) and concentrated in vacuo to a volume of about 15 mL and the cool solution was acidified with HCl (6 N) to precipitate the product. Recrystallization of the precipitate from hot H 2 O yielded 2-bromoisonicotinic acid 3-33a (0.6 g, 29%) or 2-bromonicotinic acid 3-33b (0.99 g, 84%). 3-33a 1 H (DMSO-d 6 ) 8.59 (d, J= 4.8 Hz, 1H), 7.99 (d, J= 0.3 Hz, 1H), 7.85 (dd, J= 4.8 Hz, J= 0.3 Hz, 1H); 3-33b 1 H (DMSO-d 6 ) 8.85 (d, J= 2.4 Hz, 1H), 8.16 (dd, J= 8.1, J= 2.4 Hz, 1H), 7.81 (d, J= 8.1 Hz, J= 0.3 Hz, 1H).

PAGE 113

100 N-Fmoc-trans-4-hydroxy-L-proline 4-5 NCO2HOOOH Hydroxy-proline 4-4 (20.0 g, 152 mmol) and DIPEA (66.3 mL, 49.1 g, 380 mmol) were suspended in 150 mL H 2 O + 300 mL dioxane. Fmoc chloroformate (43.25 g, 1.1 eq) in 100 mL dioxane was added dropwise. The temperature of the reaction was not allowed to rise above 10 o C during the addition. The mixture was stirred vigorously overnight, and most of the solvent was removed in vacuo. Water and saturated NaHCO 3 were added (250 mL each), and the solution extracted with 250 mL diethyl ether, which was discarded. The aqueous layer was acidified to pH 1 with concentrated HCl, extracted twice with ethyl acetate (2 X 300 mL), and the organic extracts were washed with brine. The solution was dried over MgSO 4 , filtered and the solvent removed in vacuo. The pure product 4-5 crystallized from concentrated solution (53.75 g, 100% yield). HRMS for [C 20 H 19 NO 5 +H] + calc. 354.1341, found 354.1344. Anal. Calcd for C 20 H 19 NO 5 : C, 67.98; H, 5.42; N, 3.96 Found: C, 67.94; H, 5.54; N, 3.91.

PAGE 114

101 N-Fmoc-keto-L-proline 4-6 NCO2HOOO Fmoc-trans-4-hydroxy-L-proline 4-5 (4 g, 11.32 mmol) was dissolved in 50 mL reagent acetone. A solution of 8 N Jones reagent (CrO 3 , H 2 SO 4 ) was added dropwise until the color of the solution remained dark brown-orange. The mixture was allowed to stir an extra half hour and the excess Jones reagent was quenched with iPrOH. The solution was diluted with EtOAc and washed with brine. The solution was dried over MgSO 4 , filtered and the solvent removed in vacuo. The crude N-Fmoc-keto-L-proline 4-6 was obtained as yellow solid (1.84 g, 90%) and was used in the next steps without further purification. Mp 74-75 o C. 1 H (CDCl 3 ) 9.68 (bs, 1H), 7.75 (m, 2H), 7.56 (m, 2H), 7.30 (m, 4H), 4.78 (m, 1H), 4.46 (m, 2H), 4.21 (m, 1H), 3.9 (m, 2H), 2.81 (m, 2H); 13 C (DMSO-d 6 ) 209.60, 173.85, 155.13, 144.47, 141.61, 128.64, 128.10, 126.11, 121.04, 80.08, 73.37, 68.05, 67.25, 56.93, 53.07, 47.39; HRMS for [C 21 H 19 NO 5 ] + calc. 351.1107, found 351.1097

PAGE 115

102 N-Fmoc-keto-L-proline methyl ester 4-7 NCO2MeOOO N-Fmoc-keto-L-proline 4-6 (500 mg, 1.42 mmol) in MeOH (400 mL) at 0 o C was treated with SOCl 2 (0.21 mL, 338.7 mg, 2.85 mmol), and the resulting solution was heated under reflux for 2 hours. The reaction mixture was concentrated in vacuo and chromatographed on silica gel (EtOAc/hexane 20:80) to give 4-7 as a colorless oil (406 mg, 78%). HRMS for [C 21 H 19 NO 5 ] + calc. 365.1263, found 365.1257 (-)-Diethyl-2,3-O-isoproylidene-L-tartrate 4-9 OOOOOO To a solution of diethyl-L-tartrate 4-8 (23.2 g, 112.3 mmol) and dimethoxypropane (6.4 g, 33.7 mmol) in acetone (280 mL) was added p-toluenesulfonic acid (0.48 g, 2.55 mmol). The vessel was purged of all oxygen by degassing with argon for 20 min. The solution was stirred at room temperature overnight. The resulting solution was neutralized with sodium bicarbonate and extracted with diethyl ether. The concentrate resulted in a yellow oil 4-9 (20.9 g, 85%). 1 H (CDCl 3 ) 4.73 (s, 2H), 4.24 (q, 4H), 1.47 (s, 6H), 1.25 (t, 6H). All data are similar to those published in previous literature. 98

PAGE 116

103 (+)-2,3-isopropylidene-L-threitol 4-10 HOOHOO (-)-Dimethyl-2,3-O-isoproylidene-L-tartrate 4-9 (10 g, 45.83 mmol) was dissolved in MeOH (80 mL) and the solution was cooled to 0 o C. NaBH 4 (3.82 g, 100.84 mmol) was added by portions and the solution was allowed to stir at room temperature for one hour. The reaction mixture was diluted with H 2 O (400 mL) and extracted with ethyl acetate. The organic extracts were dried over MgSO 4 , filtered and concentrated in vacuo. The crude was chromatographed (EtOAc/hexane 1:1) to yield (+)-2,3-isopropylidene-L-threitol 4-10 as a yellow oil (7.14 g, 96%). 1 H (CDCl 3 ) 3.9 (s, 2H), 3.7 (s, 4H), 3.5 (s, 1H), 1.38 (s, 6H). HRMS for [C 7 H 14 O 4 +H] + calc. 163.0970, found 163.0975. Oleyl bromide 4-12 Br78 DDQ (34.4 g, 151.8 mmol), PPh 3 (48.9 g, 151.8 mmol) and Bu 4 NBr (39.8 g, 151.8 mmol) were dissolved in freshly distilled CH 2 Cl 2 (1 L) and cool with a cold water bath. Oleyl alcohol (33.96 g, 126.5 mmol) was added dropwise to the flask, immediately changing the solution from red to dark blue. To ensure completion, the reaction was stirred for 10 minutes. The solvent was removed in vacuo and the residue purified on silica with straight pentane, yielding a clear oil (37.74 g, 90%). 1 H (CDCl 3 ) 5.38 (m, 2H), 3.41 (t, J=6.8 Hz, 2H), 2.06 (m, 4H), 1.85 (q, J=6.8 Hz, 2H), 1.35 (m, 20H), 0.91 (t, J=6.8 Hz, 3H); 13 C (CDCl 3 ) 130.11, 129.97, 33.95, 33.03, 32.83, 32.13, 29.98, 29.90, 29.75, 29.67, 29.55, 29.37, 28.97, 28.38, 27.41, 27.35, 22.90, 14.31; HRMS for

PAGE 117

104 [C 18 H 35 Br] + calc. 330.1922, found 330.1921. Anal. Calcd for C 18 H 35 Br: C, 65.24; H, 10.65 Found: C, 65.64; H, 11.04 (+)-2,3-L-threitol 4-13 OOOO7788 Sodium hydride (60% mass) (514.0 mg, 12.33 mmol) was added to a flask flushed with argon. The gray powder was washed three times with pentane to remove the protective oil. A dilute solution of (+)-2,3-isopropylidene-L-threitol 4-10 (200 mg, 4.93 mmol) in THF (10 mL) was added to the sodium hydride. The mixture was stirred for 20 minutes, then Bu 4 NBr (647 mg, 2.46 mmol) was added. Oleyl bromide 4-12 (1.63 g, 4.93 mmol) was added to the reaction flask and allowed to stir for 15 hours at ambient temperature. The reaction was monitored by TLC. Workup entailed an extraction with brine and diethyl ether. The organic layer was dried with anhydrous magnesium sulfate and the solution concentrated. The product 4-13 was isolated by flash chromatography with silica gel (hexane/EtOAc 3:1), leaving a colorless, residual oil (327 mg, 40%). 1 H (CDCl 3 ) 5.33 (m, 4H), 3.93 (m, 4H), 3.70 (m, 5H), 3.49 (m, 5H), 2.01 (m, 8H), 1.59 (m, 4H), 1.42 (s, 6H), 1.1.26 (m, 40H), 0.87 (m, 6H). HRMS for [C 43 H 82 O 4 ] + calc. 662.6213, found 662.6202.

PAGE 118

105 1,4-Bis-octadec-9-enyloxybutane-2,3-diol 4-14 OOHOOH7788 4-13 (300 mg, 0.45 mmol) was dissolved in freshly distilled THF (1 mL). Concentrated HCl (1.1 mL) was added and the reaction was stirred overnight at room temperature. The solution was concentrated in vacuo, yielding crude 4-14 (280.4 mg, 99%) which was used without further purification. 1 H (CDCl 3 ) 5.33 (m, 4H), 3.80 (m, 2H), 3.67 (m, 4H), 3.56 (m, 4H), 3.44 (m, 4H), 1.99 (m, 4H), 1.84 (m, 4H), 1.68 (m, 4H), 1.54 (m, 4H), 1.32 (m, 40H), 0.86 (m, 6H). 2,3-Bis-octadec-9-enyloxymethyl-1,4-dioxa-7-azaspiro[4.4]nonane-7,8-dicarboxylic acid 7-(9H-fluor-9-ylmethyl) ester 8-methyl ester 4-2 OO7788OONOOOO 1,4-Bis-octadec-9-enyloxy-butane-2,3-diol 4-14 (50 mg, 0.08 mmol) was dissolved in freshly distilled benzene (1 mL). To the solution was added N-Fmoc-keto-L-proline methyl ester 4-7 (59 mg, 0.16 mmol) and TsOH (2 mg, 0.01 mmol). The solution was refluxed overnight and a Dean-Starke trap was used. The reaction mixture was

PAGE 119

106 concentrated in vacuo and chromatographed (EtOAc/hexane 1:3) to yield 4-2 (630 mg, 81%). SASRIN-O-Gly-Arg(Pbf)-NH Fmoc (4-17): OONHONHFmocNHNHP b fHN In a solid phase reaction vessel Fmoc-Gly SASRIN resin (0.300 g, 0.150 mmol) was treated with 20% piperidine in DMF (10 mL, 1 x 3 min, 2 x 25 min). The reaction vessel was shaken for the specified period of time and the liquid drained from the vessel. The beads were then washed with DMF (4 x 15 mL), MeOH (3 x 10 mL), and DCM (dichloromethane, 4 x 10 mL). The deprotection was assessed with a TNBS test. In a separate flask N-Fmoc-Arg-(Pbf)-OH (0.45 mmol, 292 mg) and HOBt (0.45 mmol, 61 mg) were dissolved in DCM/DMF (2:1, 7 mL). At 0 o C DIC (0.45 mmol, 57 mg) was added dropwise and allowed to stir for 10 minutes. The solution was then transferred to the reaction vessel and agitated for 12 hours. The solution was drained and the beads washed with DMF (3 x 10 mL) and DCM (3 x 10 mL). The coupling was assessed by a TNBS test. To account for the presence of free amino groups acetylimidazole (1.50 mmol, 165 mg) was added in DCM (10 mL) and agitated for 4 hours. The solution was drained and the resin washed with DCM (4 x 15 mL).

PAGE 120

107 SASRIN-O-Gly-Arg(Pbf)-keto-Pro-NHFmoc (4-18): OONHOHNNHNHP b fHNONFmocO In a solid phase reaction vessel 4-17 (0.300 g, 0.150 mmol) was treated with 20% piperidine in DMF (10mL, 1 x 3 min, 2 x 25 min). The reaction vessel was shaken for the specified period of time and the liquid drained from the vessel. The beads were then washed with DMF (4 x 15 mL), MeOH (3 x 10 mL), and DCM (4 x 10 mL). The deprotection was assessed with a TNBS test. In a separate flask N-Fmoc-keto-L-Pro-OH (0.300 mmol, 105.4 mg), HOBt (0.300 mmol, 41 mg) and 2, 4, 6-collidine (0.300 mmol, 36 mg) were dissolved in DCM/DMF (3:1, 5 mL). At 0 o C DIC (0.300 mmol, 37 mg) was added dropwise to the flask and allowed to stir for 10 minutes. The solution was then transferred and suspended in the beads. The suspension was agitated for 18 h. The solution was drained and the beads washed with DMF (3 x 10 mL) and DCM (3 x 10 mL). The coupling was assessed by a TNBS test. To account for the presence of free amino groups acetylimidazole (1.50 mmol, 165 mg) was added in DCM (10 mL) and agitated for 4 hours. The solution was drained and the resin was washed with DCM (4 x 15 mL).

PAGE 121

108 SASRIN-O-Gly-Arg(Pbf)-keto-Pro-Asp(Otbu)-NHFmoc (4-19): OONHOHNNHNHP b f HNONOONHFmocOtBuO In a solid phase reaction vessel N-Fmoc-keto-Pro-Arg (Pbf)-Gly-OSASRIN resin (0.300 g) was treated with 20% piperidine in DMF (10 mL, 1 x 3 min, 2 x 25 min). The reaction vessel was shaken for the specified period of time and the liquid drained from the vessel. The beads were then washed with DMF (4 x 15 mL), MeOH (3 x 10 mL), and DCM (4 x 10 mL). The deprotection was assessed with a TNBS test. In a separate flask N-Fmoc-Asp (Otbu) (0.60 mmol, 246 mg), HOBt (0.60 mmol, 81 mg) and 2,4,6-collidine (0.60 mmol, 73 mg) were dissolved in DCM/DMF (2:1, 10 mL). DIC (0.60 mmol, 76 mg) was added dropwise and allowed to stir for 10 minutes. The solution was then transferred and suspended in the beads. The suspension was agitated for 24 h. The solution was drained and the beads washed with DMF (3 x 10 mL) and DCM (3 x 10 mL). The coupling was assessed by a TNBS test. To account for the presence of free amino groups acetylimidazole (1.50 mmol, 165 mg) was added in DCM (10mL) and agitated for 4 hours. The solution was drained and washed with DCM(4 x 15 mL).

PAGE 122

109 HO-Gly-Arg(Pbf)-keto-Pro-Asp(Otbu)-NH 2 (4-20): NOONH2ONHHNNHP b f NHONHOHOOO In a solid phase reaction vessel N-Fmoc-Asp(Otbu)-keto-Pro-Arg (Pbf)-Gly-OSASRIN resin (0.300 g) was treated with 20% piperidine in DMF (10 mL, 1 x 3 min, 2 x 25 min). The reaction vessel was shaken for the specified period of time and the liquid drained from the vessel. The beads were then washed with DMF (4 x 15 mL), MeOH (3 x 10 mL), and DCM (4 x 10 mL). The deprotection was assessed with a TNBS test. To cleave from the bead, 1% TFA in DCM (3 x 7.4 mL) was added and agitated each time for 5 minutes. The bead turns purple. Each filtrate fraction was immediately neutralized with 18% pyridine in MeOH solution (0.89 mL). The neutralized fractions were evaporated leaving a white solid. To remove pyridinium salts, the brown solid was purified by size-exclusion chromatography with Amberlite (XAD-2, H 2 O) resin. The resin was first washed with water and then with methanol (80 mL). The methanol fractions that contained spots as determined by TLC (Chloroform:MeOH 9:1) were combined and evaporated to leave the product as a white solid that required no further purification. HRMS (ESI, M + H) calcd for [C 34 H 51 N 7 O 11 S] + 765.3367 Found: 765.3359

PAGE 123

110 Cyclized Protected Tetrapeptide (4-21): NOOONHHNNHPbfNHONHOOOHN 4-20 (40 mg, 0.053 mmol) in DMF (18 mL) under an argon atmosphere was stirred for 5 minutes. Then HATU (61 mg, 0.16 mmol), HOBt (22 mg, 0.16 mmol) and 2,4,6-collidine (19 mg, 0.16 mmol) in DMF (11 mL) was added dropwise and allowed to stir for 24 hours. DMF was removed by vacuo distillation leaving a residue in the flask that was further purified by flash chromatography (DCM: MeOH 95:5 to 90:10) leaving a yellow foam (9 mg, 33%) HRMS(ESI, M + H) calculated for C 34 H 49 N 7 O 10 S: 747.3262. Actual: 747.3265 Fmoc-keto-Pro benzyl ester 4-26 NCO2BnOFmoc 4-6 (1 g, 2.85 mmol) was added to a mixture of benzyl alcohol (2 mL) and thionyl chloride (0.41 mL, 67.73 mg, 5.7 mmol) at 0 o C. After addition, the reaction was stirred for 15 hours. The solution was then concentrated in vacuo and the crude oil 4-26 (930.0 mg, 76%) was used in the next step without further purification. 1 H (CDCl 3 ) 7.75 (m, 2H), 7.56-7.30 (m, 11H), 4.76 (m, 1H), 4.46 (m, 2H), 4.19 (m, 1H), 3.9 (m, 2H), 3.75 (s, 2H), 2.81 (m, 2H)

PAGE 124

111 Keto-Pro benzyl ester 4-27 HNCO2BnO 4-26 (883 mg, 2 mmol) was added to a 20% solution of piperidine in either DMF or CH 3 Cl 3 (5 mL). The mixture was heated for 30 minutes or until disappearance of the starting material by TLC (CHCl 3 :MeOH, 9:1). To remove the dibenzofulvene the solution was cooled to room temperature then poured into cold water (30 mL). The white solid was removed by vacuum filtration, The filtrate was concentrated under reduced pressure and a stream of nitrogen leaving 4-27 (377 mg, 86%) as an oil which required no further purification. HRMS for C 12 H 13 NO 3 calcd 219.0895, found 219.0889 Fmoc-Arg(Pbf)-Gly benzyl ester 4-30 OPhONHONHF m ocHNNHNHPbf Dissolve Fmoc-Arg(Pbf)-OH (0.160 g, 0.248 mmol) was dissolved in a mixture of DMF/CHCl 3 (3:1, 3.5 mL). To that solution was added EDCI (0.066 g, 0.347 mmol) and HOBt (0.067 g, 0.496 mmol) and allowed to stir a t room temperature for 15 minutes. The solution was cooled to 0 o C and glycine benzyl ester hydrochloride (0.050 g, 0.248 mmol) in DMF (0.20 mL) and triethylamine (0.049 g, 0.491 mmol) was added. After the addition the solution was warmed to room temperature. The solution was stirred for 8 hours and the solid removed by vacuum filtration. Water (15 mL) was added to the filtrate and the aqueous layer was extracted with ethyl acetate (5 x 40 mL). The organic layer was washed with 10% citric acid and brine, and dried with sodium sulfate. After

PAGE 125

112 removal of the volatiles, the resulting oil was purified via flash chromatography (CHCl 3 :MeOH, 99:1 to 96:4) leaving a clear oil (0.113 g, 57%). 1 H NMR (CDCl 3 ) 7.71 (d, 2H, J= 7.7 Hz), 7.51 (t, 2H, J= 7.0 Hz), 7.41-7.38 (m, 3H), 7.35-7.01 (m, 6H), 6.51 (s, 2H), 6.30 (d, 1H, J= 53.9 Hz), 4.81 (m, 2H), 4.41 (m, 1H), 4.29-4.20 (m, 21H), 4.10 (t, 1H), 3.88 (m, 2H), 3.65 (m, 1H), 3.25 (s, 1H), 2.49 (s, 3H), 2.40 (s, 3H). 13 C NMR (CDCl 3 ) 170.05, 156.01, 153.86, 149.49, 143.96, 141.48, 136.94, 135.98, 135.24, 129.40, 129.31, 129.10, 127.97, 127.44, 127.33, 127.20, 126.99, 125.21, 125.40, 120.21, 106.92, 98.41, 81.73, 67.40, 63.71, 57.70, 52.35, 50.51, 49.69, 47.27, 46.67, 33.96, 30.87, 28.24, 25.74, 25.09 HRMS for C 43 H 49 N 6 O 8 S calcd 796.3302, found 796.3301. Glycine-benzyl ester-Arg(Pbf)-NH 2 4-31 OPhONHONH2HNNHNHPbf 4-30 (0.421 g, 1.26 mmol) was added to a 20% solution of piperidine in either DMF or CH 3 Cl 3 (5 mL). The mixture was heated for 30 minutes or until disappearance of starting material by TLC (CHCl 3 :MeOH, 9:1). To remove dibenzofulvene the solution was cooled to room temperature then poured into cold water (30 mL). The white solid was removed by vacuum filtration and the filtrate was concentrated under reduced pressure and a stream of nitrogen leaving 4-31 (0.215 g, 71%) as a white solid which required no further purification. M.p. decomposes rapidly at 245-247 o C. 1 H NMR (CDCl 3 ) 7.42-7.33 (m, 5H), 6.63 (s, 2H), 6.43 (d, 1H, J= 6.0 Hz), 4.81 (m, 2H), 4.41 (m, 1H), 3.88 (m, 2H), 3.65 (m, 1H), 3.25 (s, 1H), 2.49 (s, 3H), 2.40 (s, 3H), 2.00 (s, 3H), 1.93-1.86 (m, 3H), 1.43 (s, 3H), 1.28 (d, 3H, J= 8.3 Hz). IR: 3303.9, 2950.06, 2761.53, 1748.2,

PAGE 126

113 1669.53, 1614.69, 1591.96, 948.63, 745.48 HRMS (FAB, M+H) for C 28 H 39 N 6 O 6 S calcd 574.2621, found 574.2650. [] 25 D =-38.2 o (c=1, MeOH).

PAGE 127

APPENDIX A SPECTRAL DATA Selected spectra of compounds from Chapters 2, 3 and 4. 114

PAGE 128

115 Figure A-1. 1 H NMR of isosorbide precursor agent

PAGE 129

116 Figure A-2. 1 H NMR of isomannide precursor agent

PAGE 130

117 Figure A-3. 1 H NMR of isosorbide cyclized adduct

PAGE 131

118 Figure A-4. 1 H NMR of the diester cross-linking agent

PAGE 132

119 Figure A-5. 1 H NMR of the fluorescent detector

PAGE 133

120 Figure A-6. 1 H NMR of 2-Bromo-4-methyl pyridine

PAGE 134

121 Figure A-7. 1 H NMR of methyl-2-trimethylstannyl pyridine

PAGE 135

122 Figure A-8. 1 H NMR of (8-Bromo-naphthalen-1-ylmethoxy)-trimethyl-silane

PAGE 136

123 Figure A-9. 1 H NMR of Fmoc-keto-Proline

PAGE 137

124 Figure A-10. 1 H NMR of Fmoc-keto-Proline methyl ester

PAGE 138

125 Figure A-11. 1 H NMR of oleyl bromide

PAGE 139

126 Figure A-12. 1 H NMR of the dioleyl ether isopropylidene

PAGE 140

127 Figure A-13. 1 H NMR of the DNA transfection agent precursor

PAGE 141

APPENDIX B HPLC TRACES HPLC traces for the free-radical cyclization of isosorbide and isomannide adducts from Chapter 2. Conditions: Isopropanol/hexanes 20:80 Flow: 0.75 mL/min Temperature: 25 o C 128

PAGE 142

129 Figure B-1. Cyclization of isosorbide at -78 o C in presence of ZnCl 2

PAGE 143

130 Figure B-2. Cyclization of isomannide at 80 o C without any Lewis acid

PAGE 144

131 Figure B-3. Cyclization of isomannide at -78 o C with MgBr 2

PAGE 145

LIST OF REFERENCES 1. (a) Gomberg, M. J. Am. Chem. Soc. 1900, 22, 757; (b) Gomberg, M. Chem. Ber. 1900, 33, 3150. 2. Hey, D. H.; Waters, W. A. Chem. Rev. 1937, 21, 169. 3. Kharasch, M. S.; Margolis, E. T.; Mayo, F. R. J. Org. Chem. 1937, 2, 393. 4. Surzur, J. M. in Reactive Inermediates; Abramovitch R. A., Ed.; Plenum Press: New York, 1982; Vol. 2, Chapter 3. 5. Julia, M. Acc. Chem. Res. 1971, 4, 386. 6. Jasperse, C. P.; Curran, D. P.; Fevig, T. L. Chem. Rev. 1991, 91, 1237. 7. Parsons, A. Free Radical Chemistry; Blackwell: London, 2000. 8. Giese, B. Radical in Organic Synthesis: Formation of Carbon-Carbon Bonds; Ergammon: Oxford, 1986. 9. Renaud, P.; Gerster, M. Angew. Chem. Int. Ed. Engl. 1998, 37, 2562 10. Curran, D.; Jasperse, C.; Feury, T. Chem. Rev. 1991, 91, 1237. 11. Sibi, M.; Ji, J. Angew. Chem. Int. Ed. Engl. 1997, 36, 274. 12. Urabe, H.; Yamashita, K.; Suzuki, K.; Kobayashi, K.; Sato, F. J. Org. Chem. 1995, 60, 3576. 13. Ingold, K. U.; Lusztyk, J.; Scaiano, J. C. J. Am. Chem. Soc. 1984, 106, 343. 14. Fossey, J.; Lefort, D.; Sorba, J. Free Radicals in Organic Chemistry; John Wiley & Sons: Masson, Paris, 1995. 15. (a) Enholm, E. J.; Whitley, P. E.; Xie, Y. P. J. Org. Chem. 1996, 61, 5384; (b) Enholm, E. J.; Whitley, P. E. Tetrahedron Lett. 1996, 37, 559. 16. Ruchardt, C.; Freudenberg, B. C. H. Tetrahedron Lett. 1964, 3623. 17. (a) Beckwith, A. L.; Schiesser, C. H. Tetrahedron 1985, 41, 3925; (b) Beckwith, A. L.; Phillipou, G.; Serelis, A. K. Tetrahedron Lett. 1981, 22, 2811. 132

PAGE 146

133 18. (a) Charette, A. B.; Mellon, C.; Rouillard, L.; Malenfant, E. Pure and Appl. Chem. 1992, 64, 1925; (b) Hultin, P. G.; Earle, M. A.; Sudharshan, M. Tetrahedron 1997, 53, 14823. 19. Stack, J. G.; Curran, D. P.; Geib, S. V.; Rebek, J.; Ballester, P. J. Am. Chem. Soc. 1999, 114, 7007. 20. Sibi, M. P.; Ji, J.; Sausker, J. B.; Jasperse, C. P. J. Am. Chem. Soc. 1999, 121, 7517. 21. Delouvrie, B.; Fensterbank, L.; Lacoste, E.; Malacria, M. J. Am. Chem. Soc. 1999, 121, 11395. 22. Nishida, M.; Ueyama, E.; Hayashi, H.; Ohtake, Y.; Yamaura, Y.; Yanagimuma, E.; Yonemitsu, O.; Nishida, A.; Kawahara, N. J. Am. Chem. Soc. 1994, 116, 6455. 23. Casiraghi, G.; Zanardi, F. Chem. Rev. 1995, 95, 1677. 24. Hanessian, S. Total Synthesis of Natural Products: The Chiron Approach; Pergamon Press: Oxford, 1983. 25. Enholm, E. J.; Cottone, J. S.; Allais, F. Org. Lett. 2001, 3, 2, 145-147. 26. Porter, N. A.; Mero, G. L. J. Am. Chem. Soc. 1999, 121, 5155. 27. Pnelle, J.; Padias, A. B.; Hall, J. H. K.; Tanaka, H. Adv. Polym. Sci. 1992, 102, 73. 28. Newcomb, M.; Ha, C. Tetrahedron Lett. 1991, 32, 6493. 29. Hau, C.; Musa, O. M.; Martinez, F. N.; Newcomb, M. J. Org. Chem. 1997, 62, 2704. 30. Enholm, E. J.; Bhardawaj, A. Tetrahedron Lett. 2003, 44, 3763. 31. (a) Oupicky, D.; Read, M. L.; Bettinger, T. Methods Mol. Biol. 2001, 65, 61-78. (b) Zhdanov, R. I.; Podobed, O. V.; Vlassov, V. V. Bioelectrochemistry 2002, 58, 1, 53-64. (c) Kircheis, R.; Wagner, E. Gene Ther. Reg. 2000, 1, 1, 95-114. (d) Ogris, M.; Wagner, E. Somatic Cell and Molecular Genetics 2002, 27, 1-6, 85-95. 32. (a) Pitard, B. Somatic Cell Mol. Genet. 2002, 27, 1-6, 5-15. (b)Carlisle, R. C. Curr. Opin. Mol. Ther. 2002, 4, 4, 306-312. (c) Rusconi, S.; Ceppi, M. Modern Therapeutics in Rheumatic Diseases 2002, 479-503. (d) Niidome, T.; Huang, L Gene Ther. 2002, 9, 24, 1647-1652. (e) Kataoka, K. Biotherapy 2001, 15, 4, 425-431.

PAGE 147

134 33. (a) Blagbrough, I. S.; Geall, A. J.; Neal, A. P Biochem. Soc. Trans. 2003, 31, 2, 397-406. (b) Safinya, Cyrus R.; Lin, Alison J.; Slack, Nelle L.; Koltover, Ilya. Departments of Materials and Physics and Biomolecular Science and Engineering Program, University of California, Santa Barbara, CA, USA. Editor(s): Mahato, Ram I.; Kim, Sung Wan. Pharmaceutical Perspectives of Nucleic Acid-Based Therapeutics 2002, 190-209. (c) Ilies, M. A.; Seitz, W. A.; Balaban, A.T. Curr. Pharm. Des. 2002, 8, 27, 2441-2473. (d) Audouy, S. A. L.; de Leij, L. F. M. H.; Hoekstra, D.; Molema, G.; Groningen, N. Pharmaceutical Research 2002, 19, 11, 1599-1605. (e) Rocha, A.; Ruiz, S.; Coll, J. M. J. of Physiol.Biochem. 2002, 58, 1, 45-56. (f) Romero, E. L.; Morilla, M. J.; Bakas, L. Acta Bioquimica Clinica Latinoamericana 2001, 35, 3, 373-402. (g) Bos, G. W.; Crommelin, D. J. A.; Hennink, W. E. Methods Mol. Med. 2001, 65, 35-60. 34. Lambert, J; Mitchell, J; Roberts, K. J. Perkin. Tran. 1 2001, 471. 35. Muller, A.; F. Schuman; Koksch, M.; Sewald, N. Lett. Peptide Sci. 1997, 4, 275-281. 36. Ruoslahti, E.; Pierschbacher, M. Science 1987, 238, 491. 37. Haubner, R.; Finsinger, D.; Kessler, H. Angew. Chem. Int. Ed. Engl. 1997, 36, 1374. 38. Sugawara, T.; Matsuda, T. Biomed. Mat. Res. 1995, 29, 1047. 39. (a) Ruoslahti, E. Annu. Rev. Cell Dev. Biol. 1996, 12, 697. (b) Ojima, I.; Chakravarty, S.; Dong, Q. Bioorg. Med. Chem. 1995, 3, 337. 40. (a) Saiki, I. Jpn. Journal of Pharm. 1997, 75, 215. (b) Davies, J.; Enjalbal, C.; Nguyen, C.; Al-Jamri; Naumer, C. J. Chem. Soc. Perkin. Trans I 2000, 2907. (c) Davies, J.; Howe, J.; Jayatilake, J.; Riley, T. Lett. Peptide Sci. 1997, 4, 441. (d) Sow, M.; Molla, A.; Lamaty, F.; Lazaro, R. Lett. Peptide Sci. 1997, 4, 455. 41. Scarborough, R.; Naughton, M.; Teng, W.; Rose, J.; Phillips, D.; Nannizi, L.; Arfsten, A.; Campbell, A.; Charco, I. J. Biol. Chem. 1993, 268, 1066. 42. Well, R.; Overkleet, H.; Gijsbert, A.; Bruss, D.; Thibault, G.; Groot, P. d.; Boom, J.; Overhand, M. Bioorg. Med. Chem Lett. 2003, 13, 331. 43. McDowell, R.; Gadek, T. J. Am. Chem. Soc. 1992, 114, 9245. 44. Muller, G.; Kessler, H.; Gurrath, H. J. Comput-Aided Design 1994, 8, 709. 45. Haubner, R.; Diefenbach, J.; Kessler, H.; Goodman, S.; Jonczyk, A.; Gratias, R. J. Am. Chem. Soc. 1996, 118, 7461. 46. Roedern, E. V.; Lohof, E.; Hessler, G.; Hoffman, M.; Kessler, H. J. Am. Chem. Soc. 1996, 118, 10156.

PAGE 148

135 47. Graf, E.; Kessler, H. Angew. Chem. Int. Ed. Engl. 1994, 106, 684. 48. Belvisi, L.; Bernardi, A.; Checchia, A.; Manzoni, L.; Potenza, D.; Scolastico, C.; Castorina, M.; Cupelli, A.; Gianni, G.; Carminati, P.; Pisano, C. Org. Lett. 2001, 3, 1001. 49. (a) Hishiya, T.; Asanuma, H.; Komiyama, M. Polymer Preprints 2003, 44, 1, 653. (b) Haupt, K. Chem. Commun. 2003, 2, 171-178. (c) Shea, K. J. Polymer Preprints 2003, 44, 1, 653. (d) Ye, Lei; Mosbach, K. J. Inc. Phen. Macro. Chem. 2001, 41, 1-4, 107-113. 50. (a) Haupt, K. Chem. Commun. 2003, 2, 171-178. b) Wulff, G. Studies in Surface Science and Catalysis 2002, 141(Nanoporous Materials III), 35-44. 51. (a) Lei, J.; Tan, T. Prog. Nat. Science 2002, 12, 12, 900-903. (b) Takeuchi, T.; Fukuma, D.; Matsui, J. Chem. Lett. 2001, 6, 530-531. 52. Toorisaka, E.; Uezu, K.; Goto, M.; Furusaki, S. Biochem. Eng. J. 2003, 14, 2, 85-91. 53. (a) Selvarajan, S.; Bay, B.-H.; Choo, A.; Chuah, K.-L.; Sivaswaren, C. R.; Tien, S.-L.; Wong, C.-Y.; Tan, P.-H. J. Histochem. Cytochem. 2002, 50, 12, 1693-1696. (b) Anderson, J.; Webb, S.; Fischer, R.; Smith, C.; Dennis, J.; Di Benedetto, J. J. of Fluorescence 2002, 12, 1, 51-55. 54. (a) Pharr, Daniel Y.; McKenzie, J. Keith; Hickman, Aaron B Ground Water 1992, 30, 4, 484-9. (b) Luedeker, W.; Guenther, Dr. K. P.; Dahn, H. G. Proc. of SPIE-The International Society for Optical Engineering 1995, 2504(Environmental Monitoring and Hazardous Waste Site Remediation), 426-35. 55. a) Miller, S. J.; Copeland, G. T. J. Am. Chem. Soc. 1999, 121, 4306-4307. b) Millers, S. J.; Copeland, G. T. J. Am. Chem. Soc. 2001, 123, 6496-6502. 56. Zhang, S-W and Swager, T. J. Am. Chem. Soc. 2003, 125, 3420-3421. 57. Schrader, G. Angew. Chem. 1950, 62, 471-3. 58. Schrader, Gerhard; Bayer, Otto. Dialkylaminophosphorus fluorides. (I. G. Farbenind. A.-G.). 1938, DE 664438 19380826 Patent 59. Bajgar, J.; Tulach, J.; Jakl, A.; Patocka, J. Purkyne Med. Res. Inst., Hradec Kralove, Czech. Acta Biologica et Medica Germanica 1971, 27(1), 171-8. 60. Emerson, G. A.; Van Liere, E. J. Journal of Laboratory and Clinical Medicine 1943, 28, 700-6. 61. Ghosh, R.; Newman, J. F. Chem. Ind. 1955, 118.

PAGE 149

136 62. Leonard, J.; Diez-Barra, E.; Merino, S. Eur. J. Org. Chem. 1998, 2051. 63. Sibi, M.; Ji, J.; Sausker, J.; Jasperse, C. J. Am. Chem. Soc. 1999, 121, 7517. 64. Enholm, E. J. and Kinter K. S. J. Am. Chem. Soc. 1991, 113, 7784-7785. 65. Murakata, M.; Jono, T.; Mizuno, Y.; Hoshino, O. J. Am. Chem. Soc. 1997, 119, 11713. 66. Porter, N. A.; Giese, B.; Curran, D. P. Acc. Chem. Res. 1991, 24, 296-304. 67. (a) Nozaki, K.; Oshima, K.; Utimoto, K. Bull. Chem. Soc. Jpn. 1991, 64, 403. (b) Ruck, K.; Kunz, H. Synthesis 1993, 1018. (d) Sibi, M. P.; Jasperse, C. P.; Ji, J. J. Am. Chem. Soc. 1995, 117, 10779. 68. (a) Nishida, M.; Hayashi, H.; Yamaura, Y.; Yanaginuma, E.; Yonemitsu, O. Tetrahedron Lett. 1995, 36, 269. (b) Badone, D.; Bernassau, J.-M.; Cardamone, R.; Guzzi, U. Angew. Chem., Int. Ed. Engl. 1996, 35, 535. 69. Sibi, M.; Ji, J. J. Org. Chem. 1996, 61, 6090. 70. Loupy, A.; Monteux, D. Tetrahedron Lett. 1996, 37, 7023. 71. (a) Bonadies, F.; Di Fabio, R. J. Org. Chem. 1984, 49, 1647. (b) Hashimoto, S.; Miyazaki, Y.; Ikegami, S. Synth. Commun. 1992, 2717-2722. 72. Jenkins, A. L.; Uy, O. M.; Murray, G. Anal. Chem. 1999, 71, 2, 373-8. 73. Grubbs, R.; Gilliom, L. R. J. Am. Chem. Soc. 1986, 108, 733. 74. Shrock, R. Acc. Chem. Res. 1990, 23, 158. 75. Grubbs, R.; Trinka, T. Acc. Chem. Res. 2001, 34, 18-29. 76. Langer, R. Nature 1998, 392, 5. 77. Langer, R. Acc. Chem. Res. 2000, 33, 94. 78. (a) Miller, S. J. and Copeland, G. T. J. Am. Chem. Soc. 1999, 121, 4306-4307. (b) Miller, S. J. and Copeland, G. T. J. Am. Chem. Soc. 2001, 123, 6496-6502. 79. Dagani, R.; Chemical & Engineering News 2003, 81, 12. 80. Bailey, R. J.; Card, P. J.; Shechter, H. J. Am. Chem. Soc. 1983, 105, 6096-6103. 81. Michl, J.; Schwab, P. F. H.; Fleischer, F. J. Org. Chem. 2002, 67, 443-449. 82. Reyes-Rivera, H. M.; Hutchins, R. O.; Dalton, D. R. J. Heterocyclic Chem. 1995, 32, 665-669.

PAGE 150

137 83. Zelphati, O.; Nguyen, C.; Ferrari, M.; Felgner, J.; Tsai, Y.; Felgner, P. Gene Therapy 1998, 5, 9, 1272-82 84. Smith, Elizabeth M.; Swiss, Gerald F.; Neustadt, Bernard R.; Gold, Elijah H.; Sommer, Jane A.; Brown, Arthur D.; Chiu, Peter J.S.; Moran, Rosa; Sybertz, Edmund J. and Baum, Thomas J. Med. Chem. 1988, 31, 875-885. 85. Smith, Elizabeth M.; Swiss, Gerald F.; Neustadt, Bernard R.; McNamara, Paul; Gold, Elijah H.; Sybertz, Edmund J.; Baum, Thomas J. Med. Chem. 1989, 32, 7, 1600-1606. 86. Iranpoor, N.; Firouzabadi, H.; Aghapour, Gh. And Vaez zadeh A.R. Tetrahedron 2002, 58, 8689-8693. 87. Bradley, M.; Guillier, F.; Orain, D. Chem. Rev. 2000, 100, 2091. 88. Mergler, M.; Nyfler, N.; Tanner, R.; Gostelli, J.; Groog, P. Tetrahedron Lett. 1988, 29, 4008. 89. Mergler, M.; Tanner, R.; Gostelli, J.; Groog, P. Tetrahedron Lett. 1988, 29, 4005. 90. Wang, S. J. Am. Chem. Soc. 1973, 95, 1328. 91. Loupy, Andre; Monteux, Daphne A. Tetrahedron 2002, 58, 8, 1541-1549. 92. Banwell, Martin G.; Hockless, David C. R.; McLeod, Malcolm D. New J. Chem. 2003, 27, 1, 50-59. 93. Gallagher, Michael J. Australian J. Chem. 1968, 21, 5, 1197-204. 94. Pak, Chwang Siek; Lee, Eun; Lee, Ge Hyeong. J. Org. Chem. 1993, 58, 6, 1523-30. 95. Little, R. Daniel; Fox, Dennis P.; Van Hijfte, Luc; Dannecker, Robert; Sowell, Gregory; Wolin, Ronald L.; Moens, Luc; Baizer, Manuel M. J. Org. Chem. 1988, 53, 10, 2287-94. 96. Leuck, G. J.; Perkins, R. P.; Whitmore, F. C. J. Am. Chem. Soc. 1929, 51,1831. 97. Rule, H. G.; Pursell, W.; Brown, R. R. H. J. Chem. Soc. 1934, 168. 98. Waanders, P. P.; Thijs, L.; Zwanenburg, B. Tetrahedron Lett. 1987, 28, 21, 2409-12.

PAGE 151

BIOGRAPHICAL SKETCH Florent Allais was born in Agen, France on April, 12 th 1977, and grew up an only child, in Buzet sur Base (47), a little town lost in the South West countryside. Growing up in the south of France, he enjoyed the relaxed lifestyle and all the aspects of southern culture (excellent food, family time, country, beach, walks in the woods, biking, etc.). Science was important early in his education: as recalled by his grandmother, at 10 years old, he was already doing electrolysis of salted water in a glass with a 4.5 V battery at his great-grandmother’s! He always enjoyed labs in physics and chemistry (as well as biology) in middle and high school. After studying Sciences in an Honor Class, he graduated with Honors from Lyce Stendhal (Aiguillon, France) in 1995. In Fall 1995, Florent started college at the University of Bordeaux I in Agen (47, France). He began his undergraduate career with hopes of becoming a scientist. That never changed through the years, despite some difficulties during the first semester of the freshman year: God, he hated mathematics! Chemistry was now his new career path. In the fall of 1997 he moved from Agen to Talence (33, France), where he started his Bachelor of Science degree at the Chemistry Department of the University of Bordeaux I. Going from a 17-student classroom to a 200-student amphitheatre wasn’t an easy task. Moreover struggling with hardcore classes and trying to make connections with an unknown environment didn’t ease the process. However Florent succeeded in making friends and also mastering chemistry, and got his bachelor degree in 1998, with honors. Then in 1999, he left France and spent 4 months here, at the University of Florida, as part of the REU program 138

PAGE 152

139 directed by Dr. Randy Duran. Florent discovered for the first time what research really was about. He worked on radical chemistry under Dr. Enholm’s advising, and went back to France to complete his Master of Science in Pr. Yannick Landais’ research group at the University of Bordeaux I, where he worked on asymmetric cyclopropanations. After he got his Master with Honors in Spring 2000, and as his experience in Gainesville was enjoyable, Florent decided to join Dr. Enholm’s group to pursue a Ph.D. He flew back to Gainesville in Fall 2000 and has been working on three projects since. Right now Florent is finishing up his doctorate degree and will go back to France on May 2004. He’s expected in Pr Janine Cossy’s research group at the ENSPCI (Ecole Nationale Suprieure de Physique et Chimie Industrielles de Paris, Paris, France) where he will complete a postdoctoral fellowship. In Paris, Florent will work on the synthesis of an anti-cancer drug with the financial support of the ARC (French Association for the Research against Cancer).