<%BANNER%>

Self-Assembly of Purines by Quadruple Hydrogen Bonding

Permanent Link: http://ufdc.ufl.edu/UFE0021745/00001

Material Information

Title: Self-Assembly of Purines by Quadruple Hydrogen Bonding Design, Synthesis, and Structure-Property Relationships
Physical Description: 1 online resource (165 p.)
Language: english
Creator: Martin, Alisha Michelle
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: diaminopurine, quadruple, ureidopurine
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The development of synthetically accessible, quadruple hydrogen bonded (QHB) systems that are built on nucleobase platforms has revolutionized the field of supramolecular chemistry in the last decade. Herein are introduced the first QHB systems built on a ureidodiaminopurine (UDAP) platform. This unit offers conformational stability with modest dimerization constants, Kdim ~ 530-1100 M^-1, in chloroform. The structure-property relationships of five UDAPs were examined through synthetic modifications to the urea and N9 substituents. The urea substituents were varied from phenyl to hexyl to probe electronic effects. Likewise, the N9 substituent was varied from aryl to alkyl to investigate steric consequences. The synthesis of aryl UDAP is regioselective for N_C2 substitution, interestingly, without isolation of an N_C6 substituted product. However, substitution occurs exclusively at N_C6 with hexyl isocyanate. From this reactivity difference, a study into the nucleophilicity of the amino groups of DAP with common acylating agents was undertaken. Characterization by ^1H, ^13C, ^1H-^13C gHMBC, ^1H-^15N gHMBC unambiguously assigned the site of acylation as N_C2 or N_C6. The nucleophilic reactivity of DAP was explored, thermodynamic equilibrium was eliminated, and the reactivity determined to favor N_C2 kinetically with acetic anhydride and methoxy acetyl chloride.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Alisha Michelle Martin.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Castellano, Ronald K.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021745:00001

Permanent Link: http://ufdc.ufl.edu/UFE0021745/00001

Material Information

Title: Self-Assembly of Purines by Quadruple Hydrogen Bonding Design, Synthesis, and Structure-Property Relationships
Physical Description: 1 online resource (165 p.)
Language: english
Creator: Martin, Alisha Michelle
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2007

Subjects

Subjects / Keywords: diaminopurine, quadruple, ureidopurine
Chemistry -- Dissertations, Academic -- UF
Genre: Chemistry thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: The development of synthetically accessible, quadruple hydrogen bonded (QHB) systems that are built on nucleobase platforms has revolutionized the field of supramolecular chemistry in the last decade. Herein are introduced the first QHB systems built on a ureidodiaminopurine (UDAP) platform. This unit offers conformational stability with modest dimerization constants, Kdim ~ 530-1100 M^-1, in chloroform. The structure-property relationships of five UDAPs were examined through synthetic modifications to the urea and N9 substituents. The urea substituents were varied from phenyl to hexyl to probe electronic effects. Likewise, the N9 substituent was varied from aryl to alkyl to investigate steric consequences. The synthesis of aryl UDAP is regioselective for N_C2 substitution, interestingly, without isolation of an N_C6 substituted product. However, substitution occurs exclusively at N_C6 with hexyl isocyanate. From this reactivity difference, a study into the nucleophilicity of the amino groups of DAP with common acylating agents was undertaken. Characterization by ^1H, ^13C, ^1H-^13C gHMBC, ^1H-^15N gHMBC unambiguously assigned the site of acylation as N_C2 or N_C6. The nucleophilic reactivity of DAP was explored, thermodynamic equilibrium was eliminated, and the reactivity determined to favor N_C2 kinetically with acetic anhydride and methoxy acetyl chloride.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Alisha Michelle Martin.
Thesis: Thesis (Ph.D.)--University of Florida, 2007.
Local: Adviser: Castellano, Ronald K.
Electronic Access: RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2008-06-30

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2007
System ID: UFE0021745:00001


This item has the following downloads:


Full Text

PAGE 1

1 SELF-ASSEMBLY OF PURINES BY QU ADRUPLE HYDROGEN BONDING: DESIGN, SYNTHESIS, AND STRUCTURE-PROPERTY RELATIONSHIPS BY ALISHA M. MARTIN 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 2007

PAGE 2

2 2007 Alisha M. Martin

PAGE 3

3 To my brother Tony Martin Jr., riding wave s of peace (10-21-06); and to my nephew Curren Nesta Martin

PAGE 4

4 ACKNOWLEDGMENTS I would like to thank my fa mil y, especially my parents. I th ank my mother and father, for their love, guidance, encouragement, and suppor t. I thank Tony for our long talks. His support meant so much then, but means even more now. T ony, you always said life is too short. I thank my brother Grant for making me laugh in the tough times. I thank Dr. Ram Bhat (Mr. M.I.A.) for his encouragement and guidance. I thank my oldest and dearest friends, Crissie Bonser and Stacy Solano, for getting me through the tough personal ti mes as well as the challenging times while in school. Andrea, Amy, Stephanie, and Susans sup port is also appreciated. To my friend Tim Steckler, thanks for helping me through in the general sense, encouraging me, and talking me out of the rough spots, but also the COOKIES (I than k Mrs. Steckler for that). Of course I cannot forget the furry ones who make my house a home, Fifi, Sailor, Ella, and Peaches. I would also like to thank R on Castellano, my advisor, for teaching me with patience and understanding. I am thankful for his creative science, his scientific expertise, and his diligence. I thank the Castellano group past a nd present. I thank Yan for being a good lab mate. I thank Pam, whom I wish had come along sooner to share her chemistry knowledge with me. Ling, it was a pleasure. I thank Ion Ghiviriga immensely; his assi stance to me was a great factor to my success. I thank goodness he has patience and understand ing. I spent many hours in the NMR labs, and want to thank the staff, Robert Harker and Da vid Richardson, for putting up with my ridiculous questions and guiding me to the pr oper solutions. Also, I thank my committee for their guidance.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS...............................................................................................................4 LIST OF TABLES................................................................................................................. ..........7 LIST OF FIGURES.........................................................................................................................8 LIST OF ABBREVIATIONS........................................................................................................ 11 ABSTRACT...................................................................................................................................12 CHAPTER 1 INTRODUCTION..................................................................................................................13 Inspiration...............................................................................................................................13 Supramolecular Chemistry ..................................................................................................... 14 Hydrogen Bonding in Supramolecular Chemistry. ........................................................ 17 Supramolecular M aterials and Self-Assembly................................................................ 19 Strength and Specificity of Linear Systems by Multip le Hydrogen Bonding........................ 20 Systems of T wo and Three Linear Hydrogen Bonds....................................................... 20 Linear Arrays of Four Hydrogen Bonds .........................................................................23 Scope and Organization of Thesis................................................................................... 27 2 DESIGN AND SYNTHESIS OF COMP LEMEN TARY UREIDODIAMINOPURINES BY QUADRUPLE HYDROGEN BONDING....................................................................... 29 Design: Structural, Conformational, and Synthetic Considerations....................................... 29 Structural and Synthetic Design...................................................................................... 29 Synthesis..........................................................................................................................33 Solution Phase Characterization of 1a Supports the Mode of Dim erization..........................34 Intermolecular Hydrogen Bonding Strength................................................................... 38 Dimerization in the Solid State........................................................................................42 Summary..........................................................................................................................44 Experimental M ethods .................................................................................................... 45 General protocols.....................................................................................................45 Synthetic details.......................................................................................................46 Computational details............................................................................................... 52 NMR Experimental parameters............................................................................... 53 3 STRUCTURE-PROPERTY RELATIONS HIPS IN UDAP DERIVATIVES....................... 54 Introduction................................................................................................................... ..........54 Structure Property Relationships............................................................................................ 57

PAGE 6

6 Synthesis of 7, 8, and 9.................................................................................................... 57 Properties of UDAP in Solution ......................................................................................61 Interactions between N9 and the Urea ............................................................................ 64 Electronics Effects........................................................................................................... 67 Solubility..................................................................................................................... ....69 Summary..........................................................................................................................69 Experimental M ethods .................................................................................................... 70 Computational Details...................................................................................................... 81 4 REACTIVITY OF 2,6-DIAMINOPURINE........................................................................... 82 Introduction................................................................................................................... ..........82 Reactivity of 2,6-Diaminopurine............................................................................................84 Acylation with M ethoxy Acetyl Chloride........................................................................ 84 Determination of Substitution Site.................................................................................. 86 Substitution at NC6 and Reversibility............................................................................... 88 Reactivity of DAP with Acetic Anhydride...................................................................... 91 Summary and Conclusions..............................................................................................93 Experimental M ethods .................................................................................................... 94 NMR Experimental Parameters..................................................................................... 101 5 CONCLUSIONS AND FUTURE DIRECTIONS............................................................... 103 Summary and Conclusions...................................................................................................103 Future Directions..................................................................................................................105 Reversible M aterials......................................................................................................105 Alternative QHB Platforms........................................................................................... 109 Preliminary Experimental Results ................................................................................. 114 APPENDIX A NMR DATA.........................................................................................................................119 B SOLID STATE DATA.........................................................................................................156 LIST OF REFERERENCES........................................................................................................157 BIOGRAPHICAL SKETCH.......................................................................................................165

PAGE 7

7 LIST OF TABLES Table page 2-1 Relative energies (kcal mol1) of four possible UDAP conformers by computation........ 32 2-2 Hydrogen bond distances () for QHB 1a in the solid state............................................. 43 2-3 Hydrogen bond distance of 1a participating in Hoogsteen dim erization in the solid state....................................................................................................................................44 3-1 Dimerization constants and fr ee energy of form ation data for 1a, 1b and 79 ................64 4-1 Reaction conditions and outcomes for nucleophilic substitu tion of DAP with me thoxy acetyl chloride..................................................................................................... 85 4-2 Chemical shifts of the imidazole portion of the bicyclic purine compounds 4c (2am ino-6-chloro-9-heptylpurine), 5c (2,6-diamino-9-heptylpurine), and 18c (monosubstituted product 18cx or 18cy )...........................................................................87 4-3 Chemical shift data for the pyrimid ine portion of the purine of compounds 4c 5c and 18cx .............................................................................................................................88 4-4 Reaction outcomes for high con centration (40 mM) reactions of 5c and 5d with electrophiles A B and C ...................................................................................................90

PAGE 8

8 LIST OF FIGURES Figure page 1-1 Association of potassium and sodium by crown ethers..................................................... 15 1-2 Calixarene structures can vary from simp le to complex and are proposed as mi mics to enzyme binding pockets................................................................................................. 16 1-3 Molecular sensors couple bi nding to an observable response........................................... 16 1-4 Rotaxane and catenane are m echanically interlocked rings.............................................. 17 1-5 Secondary interactions that result from adjacent sites in a hydrogen-bonded com plex.... 19 1-6 Rigidly linked pyridone un its are predisposed to assemble by the arrangem ent of hydrogen bond donors and acceptors................................................................................. 21 1-7 For the diacylpyridin e/Npropylthymine complex B the Ka of 800 M in CDCl3 is an order of magnitude greater than the a ssociation of diami nopyridine with Npropylthymine (Ka = 84 M, in CDCl3) complex A Acylation of diaminotriazine ( C ) resulted in a lowered Ka ~ 6 M due to electrostatic repulsion......................................... 22 1-8 Electrostatic repulsion of amide carbonyl groups yields a lin ear array of five hydrogen bond donors and accepto rs................................................................................. 22 1-9 Strength of triazine dimerization was in creased by structural modifications and preorganization................................................................................................................ ..23 1-10 Tautomeric structures of UPy are preorganized for different modes of dimerization and association................................................................................................................ ...24 1-11 Hydroxy telechelic polymers func tionalized with UPy molecules. ................................... 25 1-12 Ureidoguanosine dimerizes with DAN..............................................................................26 1-13 Purines offer many opportunites for prepari ng controlled self-assembling structures. .....27 2-1 Dimerization equilibrium for UDAP, 1N3..........................................................................30 2-2 Four possible low energy conformers of 9-m ethyl-2-N-phenylureidodiaminopurine (UDAP) considered computationally................................................................................. 31 2-3 Single CN bond rotation interconverts 1N1 to 1N3............................................................32 2-4 X-ray crystal stru cture of model compound 2 ...................................................................33 2-5 Regioselective synthesis of UDAP.................................................................................... 34

PAGE 9

9 2-6 A NOESY spectrum...........................................................................................................36 2-7 VT NMR of compound 1a .................................................................................................37 2-8 The possible dimers that can be form ed through two intermolecular hydrogen bonds via 1N1................................................................................................................................38 2-9 Representative dilution of 1a.............................................................................................41 2-10 Dilution data for 1a and calculation of Kdim......................................................................41 2-11 1N3 Conformer QHB motif in the solid state..................................................................... 43 2-12 Hydrogen bonding to the Hoogsteen face of 1a ................................................................44 3-1 Ureidotriazine and ureidopyrimidine form highly stable DADA QHB dimers................. 54 3-2 UDAP derivatives designed to probe structure-property relationships............................. 55 3-3 Ester hydrolysis of 9..........................................................................................................57 3-4 The reaction of phenyl isocyanate with DAP.................................................................... 58 3-5 Synthesis of 4c and 4d using TBAF.................................................................................. 59 3-6 Deprotonation of DAP with n-BuLi or NaH and reaction of the anion with hexyl isocyanate..................................................................................................................... ......60 3-7 The synthesis of UDAP 7 and 8 by deprotonation, substitution with hexyl isocyanate, and displaceme nt of the 6-chloro group............................................................................. 60 3-8 The synthesis of 9 was achieved through deprotonation of 4d with LTMP followed by substitution with hexyl isocyanate to form 6d ..............................................................61 3-9 1H VT-NMR of compound 8.............................................................................................63 3-10 Computational studies to probe possi ble edge -to-face aromatic interactions.................... 66 3-11 Proton NMR of 7 and 8 show similar chem ical shifts for Hc............................................66 3-12 Proton NMR illustrates decreased acidity of Hc when the urea substituent is changed from aryl to alkyl...............................................................................................................67 3-13 Compound 9 arranged for bifurcated hydrogen bonding of Hc to N3............................... 68 3-14 Proton NMR of 9 ...............................................................................................................68 4-1 DAP derivatives investigated in pharmaceutical research as sp ecific kinase inhibitors...82

PAGE 10

10 4-2 Examples of DAP acylation found in the literature111,112..................................................83 4-3 Reaction of DAP with 2.5 e q. me thoxy acetyl chloride....................................................85 4-4 NMR coupling assignments of the imidazole portion of 2-am ino-6-chloropurine............86 4-5 Correlations proved the point of substitution to be NC2.....................................................88 4-6 Spectrum showing correlations for 18cx proves the site of substitu tion........................... 89 4-7 Electrophiles A C were reacted with DAP....................................................................... 89 4-8 Reaction of DAP with acetic anhydride yielded a mi xture of NC2 and NC6 products........91 4-9 Reaction of 5d with acetic anhydride................................................................................ 92 5-1 Synthesis of UDAP functi onalized lo w molecular weight poly(THF) for functional materials...................................................................................................................... .....107 5-2 Original synthesis of compound 23 (UA)........................................................................ 108 5-3 Proposed synthesis of UA................................................................................................109 5-4 Pyrimidine rings fused at the [4,5] carbons show potential for a new class of ditopic self-assembling m olecules............................................................................................... 110 5-5 Synthesis of 4,8-diamino[4,5-d]pyrimi do-2,6-bisphenylureidopyrimidine by a sim ple two-step process...............................................................................................................111 5-6 Products obtained in preliminary synt hesis aimed at form ing 4,8-diamino-6chloro[4,5-d]pyrimido-2-N-phenylureidopyrimidine...................................................... 111 5-7 Synthesis of 4,8-dibenzyloxy[4,5-d] pyrimido-2,6-N-hexylureidopyrim idine ( 34) and unsuccessful reduction by hydrogenation........................................................................112 5-8 The suggested synthesis for 28 by catalytic hydrogen transfer from a mmonium formate in refluxing methanol......................................................................................... 113 5-9 The mono-faced counterpart s to the ditopic systems 36 and 37 derived from 2,4chloroquinazolines...........................................................................................................113 5-10 The proposed synthesis of 4-amino-2-Nphenylureidoquinazoline is similar to the ditopic counterpart. .......................................................................................................... 114 5-11 Synthesis of 2,6-dichloroquinazoline 29 from 2-am inobenzonitrile 27 ..........................114

PAGE 11

11 LIST OF ABBREVIATIONS DAP Diaminopurine DMF Di methylformamide DMSO Dimethyl sulfoxide DP Degree of polymerization ESI-FT-ICR Electrospray ionization-Fourie r transform-ion cyclotron resonance HRMS High resolution mass spectrometry NMR Nuclear magnetic resonance NOESY Nuclear Overhauser Spectroscopy PNA Peptide nucleic acid QHB Quadruple hydrogen bond SMP Supramolecular polymer THF Tetrahydrofuran TLC Thin layer chromatography UDAP Ureidodiaminopurine VT Variable temperature

PAGE 12

12 Abstract of Dissertation Pres ented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy SELF-ASSEMBLY OF PURINES BY QU ADRUPLE HYDROGEN BONDING: DESIGN, SYNTHESIS, AND STRUCTURE-PROPERTY RELATIONSHIPS By Alisha M. Martin December 2007 Chair: Ronald K. Castellano Major: Chemistry The development of synthetically accessibl e, quadruple hydrogen bonded (QHB) systems that are built on nucleobase platforms has revolutionized the field of supramolecular chemistry in the last decade. Herein are introduced the firs t QHB systems built on a ureidodiaminopurine (UDAP) platform. This unit offers conformational stability with modest dimerization constants, Kdim ~ 530 M, in chloroform. The structure-propert y relationships of five UDAPs were examined through synthetic modifications to the urea and N9 substituents. The urea substituents were varied from phenyl to hexyl to probe electronic effects. Li kewise, the N9 substituent was varied from aryl to alkyl to investigate steric consequences. The synthesis of aryl UDAP is regioselective for NC2 substitution, interestingly, w ithout isolation of an NC6 substituted product. However, substitution occurs exclusively at NC6 with hexyl isocyanate. From this reactivity difference, a study into the nucleophilicity of the amino groups of DAP with common acylating agents was undertaken. Characterization by 1H, 13C, 1H13C gHMBC, 1H15N gHMBC unambiguously assigned the si te of acylation as NC2 or NC6. The nucleophilic reactivity of DAP was explored, thermodynamic equilibrium was elimin ated, and the reactivity determined to favor NC2 kinetically with acetic anhydrid e and methoxy acetyl chloride.

PAGE 13

13 CHAPTER 1 INTRODUCTION Inspiration Synthetic chemists often draw inspiration fr om the enormous complexity and specificity that has evolved in nature.1,2 From only a few atoms, nature forms a variety of functional groups, which together make up a small number of biol ogical molecules (compared to the synthetic opportunities). These biological mo lecules (monomers) are combin ed to form macromolecules with exact molecular weight, with specific bran ching points, and with predisposed but dynamic conformations. Natural macromolecules are formed using a variety of functional groups that control polarity, electrostatic in teractions, and hydrophobicity. Fu rther, biology uses covalent and noncovalent interactions for a ssembly with precise control. These biological ensembles then elicit some unique function. Common biomacromolecules or biopolymers th at display function, which arises from complex structure, conformation, and asse mbly include DNA/RNA and polypeptides.3-6 The nucleobases of a single DNA strand are held together through c ovalent phosphate linkages, likewise polypeptides are held together by cova lent amidic bonds. The primary sequence of the biopolymer represents the chemical information fr om which secondary structure emerges, that in turn leads to defined macromolecular architecture, functional group placement, and function.1 Under physiological conditions the amino acid side chains can link cova lently (by disulfide bonds), or noncovalently, primarily by hydrogen bonds, to form local regions of ordered structure. In DNA the nucleobase sequence stores information, and specificity. Formation of tertiary structure is achieved through complementary pairing by hydrogen bonding between purine and pyrimidine bases. In a double helix the two strands are additionally held together by hydrophobic and electrostatic interactions. The noncovalent forces by which recognition or

PAGE 14

14 association occurs in natural systems consists primarily of stacking (aromatic interactions), van der Waals interactions, ionic interactions and hydrogen bonding, which work in concert to stabilize the overall assembly. Supramolecular Chemistry It is through mi micking the in teractions of biological system s that organic chemists can create synthetic materials with enhanced func tions through the combination of covalent and noncovalent chemistry. There are many concepts in supramolecular chemistry which have been founded on the basis of modeling th e interactions of nature, including a) molecular recognition, b) molecular self-assembly, and c) dynamic covalent chemistry. Molecular recognition, involves tw o or more entities recogniz ing one another, and is, by definition an inte rmolecular process.7,8 Molecular recognition is ubiqu itous in biological systems and is at the basis of the various physiological responses elic ited by highly specific host-guest interactions at the cel lular level (enzyme-substrate complexes, antibody/antigen, and membrane receptors).9 Study of molecular recogniti on extends to increasing th e diversity of synthetic systems that are intended to mimic the host-guest interactions of biological systems. Molecular receptors are developed to be capable of select ively binding specific substrates by noncovalent interactions which rely on information storag e, and processing info rmation stored in the interacting species. Examples of synthetic molecular recognition units include crown et hers, calixarenes, and molecular sensors. Association of metals with crown ethers is based on size, shape, and charge.10,11Cavity size is proportional to ring size, thus selectivity of complexation is based on the size of the metal ion. For example, potassium ion, 2.66 is nicely accommodated by 18crown-6 with a cavity size of 2.6.2 whereas, sodium, 1.94 is accommodated by 15crown-5 with a cavity size of 1.7.2 (figure 1-1). Sodium/potassium balance is critical to

PAGE 15

15 cellular function, and disruption equa tes to cell death. This is importa nt in terms of antimicrobial compounds. Hydrophilic ions cannot pass through hydrophobic cell membranes. Transport of ions across cell membranes, by crown et hers, results in bacterial death. Figure 1-1. Association of potassi um and sodium by crown ethers Calixarenes are basket or cup-like structures that serve as hosts to guest from metals to neutral organic molecules by size complementarity and intermolecular interactions.12,13 As illustrated in figure 1-2, structures can vary from simple to complex by modification of the top or bottom rims, and are proposed as enzyme mimics and molecular capsules. These versatile binding pockets have helped to unravel many of the my steries of molecular recognition and are useful for rational drug design. Molecular sensors speak more to the analytical side of chemistry rather than the biological.14 Sensors are specifically designed molecule s which change properties in response to the presence of a molecular complement. The spec ific reversible interaction of a guest with a host produces a change which can be readily mon itored such as color, photoluminescence, or redox potential. The chosen method is photolum inescence; for example, crown ethers are tethered to luminescent moieties, and the specificity of crown ethers for metals is used to elicit a response upon binding (figure 1-3, left).15-17 Anions are typically bound using sets of hydrogen bond donors and acceptors as illustrated by the central structure of figure 1-3,18-20 where small molecules such as CO or O2, and organic compounds with donor atoms, can be coordinated to metal atoms.21

PAGE 16

16 Figure 1-2. Calixarene structures can vary from simple to complex and are proposed as mimics to enzyme binding pockets. Figure 1-3. Molecular sensors c ouple binding to an observable response (photoluminescence in these cases). Molecular self-assembly falls under the umbre lla of molecular recognition and is defined by components designed (programmed) or predispos ed to reversibly organize themselves into desired patterns and functions (exemplif ied by protein folding or DNA/RNA duplex formation).7,8 Self-assembly can be interor intramolecu lar. The scope of self-assembly is broad and applies to biological chemistry, polymer chemistry, and materials science and engineering. Molecular self-assembly is within the general scope of this thesis and will be exemplified herein. Dynamic covalent chemistry describes syst ems in which covalent bonds are formed reversibly, under thermodynamic control. The product composition is thus governed by the relative stability of the species.22-29 Figure 1-4 shows examples of such dynamic systems including mechanically interlocked molecular assemblies such as rotaxanes (dumbbell-shaped

PAGE 17

17 component with a trapped ring, left), and catenanes (two interlocked rings, right). These systems have been developed for molecular machines and electronics, and display readily controlled internal movements of one compone nt with respect to the other. Figure 1-4. Rotaxane (left) and cat enane (right) are mechanically in terlocked rings that can act as molecular machines or electronics.22 Molecular electronics and molecular motors derived from catenanes and rotaxanes are formed by template directed synthesis, and si nce the noncovalent template component remains within the system they can be activated so that they respond to a stimulus. The recognition element can be chemical, electrical, or optical ly controlled to switch off and on. Molecular motors are conceptually similar; however, th e response to stimuli re sults in the physical, reversible movement of the system components. Hydrogen Bonding in Supramolecular Chemistry Of the dynami c interactions that define supramolecular chemistry, hydrogen bonding interactions have a prominent role due to their intermediate strength, between van der Waals interactions (> 5 kJ mol) and covalent bonds (< 250 kJ mol).30 Hydrogen bonding also offers directionality and specific ity. Directionality is a conseque nce of an energetic preference for a hydrogen atom of a donor to arrange linearly, or close to linearly, with the lone pair of an acceptor atom (however, hydrogen bonding is dominate d by electrostatics which allows for some

PAGE 18

18 flexibility of the bond angle),30 and specificity is the result of the presence of a hydrogen bond donor and hydrogen bond acceptor (atoms which are more electronegative that H, generally N, O, or F, but in some cases C, P, S, Cl, Se, Br, and I).31 Hydrogen bond preferences can generally be discerned intuitively based on proton acidities and the basicity of the corresponding acceptor.32 Likewise, the strengths of individual hydrogen bonds can generally be related to donor strength (i.e., can be de fined by the distance of X H; X = donor) that follows a general ranking of OH > NH > SH > CH.33 As a general rule the donor strength is increased by neighboring electron-withdraw ing groups and reduced by electron-donating groups. In consequence, the ranking of OH donor strengths follows p Ka, where H3O+ > OCOH > PhOH > C(sp3)OH > H2O > OH. When multiple hydrogen bonding entities are pr esent the strength of association is considered additive, however, the additivity of the individual hydrogen bonds is complicated when multiple sites are linearly arranged. In the formation of complexes (even the dimer formed between thymine and adenine) there is a si zeable repulsive contribution by secondary interactions between adjacent sites of the complex. First stated by Jorgensen for simple homodimers, the interaction between identical sites at the H-bonding interface is repulsive (shown with black arrows, figure 1-5), while the interaction between diffe rent sites at the Hbonding interface is attractive (gray arrows, figure 1-5).34,35 Schneider later drew linear freeenergy relationships that help to predict the strength of linear multiple hydrogen-bonded complexes in CDCl3 ( 8 kJ/mol for each primary interaction and 2.9 kJ/mol for each attractive and repulsive secondary interaction).36 These values are truly estimates when conformational effects are considered;37,38 preorganization increases stab ility of the hydrogen-bonding motif by

PAGE 19

19 reducing entropic penalties upon assembly and is an additional factor that can appear to increase the association strength between uni ts of multiple hydrogen bonding arrays.39 Figure 1-5. Secondary interactions that result from adjacent sites in a hydrogen-bonded complex, and can be attractive (gray arrows ) or repulsive (black arrows) Supramolecular Materials and Self-Assembly Supramolecular m aterials thus far have been described in terms of molecular recognition and dynamic covalent systems. Se lf-assembly of biological systems by H-bonding has inspired the development of synthetic syst ems which mimic these interactions to gain perspective on the interactions which control assembly, and al so the pursuit of functional materials. Functional materials that are self-assembled to obtain some degree of polymerization (DP), as one-dimensional units, in thermodynami c equilibrium are termed supramolecular polymers (SMPs).40-42 Supramolecular polymers/materials o ffer several advantages over their covalent counterparts that often include a simplified synthesis (reduced cost), self-assembly into the most thermodynamically stable form (error checking is inherent), and responsiveness (to external stimuli). These factors make supram olecular strategies attractive for materials applications in areas as diverse as electro-opti cs, thermoplastic materials, information storage, and biomedicine.43 Materials with tightly bi nding supramolecular units give elastic properties similar to covalent polymers, and with the temperature responsiveness of SMPs makes processability a great advantage.44 Self-assembling monomers are also of interest to biomedical areas where there is a need for biocompatible and biodegradable materials. Drug delivery,

PAGE 20

20 polymer diagnostics, and implantable biomaterials for tissue or bone replacement engineering are appealing. The strength of dimerization largely controls material properties. The degree of polymerization is directly related to the ther modynamic stability of the monomeric system or endgroup,45 and the relationship is such that a Ka ~ 1034 in a 1 M solution gives a DP 100, while a Ka ~ 107 offers a DP 100000.40 Association strength, Ka, is also related to the association lifetime, and is thus relevant to viscoelastic properties in the bulk material.44 Strongly associating monomeric units are not the only way to obtain functi onal materials; the effects of weak hydrogen bonding interactions can be amplified by phase segr egation (typically by aromatic interactions) to obtain higher degrees of polymerization by excluded volume effects.4649 In general, the properties of SMPs are simila r to those of traditi onal polymers with the exception of their reversibility which make s the materials thermally and chemically responsive.46-49 Systems of two hydrogen bonds or more fall into two categories, complementary and selfcomplementary. Complementary systems require a bonding partner, and stoichiometry is a concern for maximizing interactions. Thus, self -complementary systems (necessarily involving two or four hydrogen bonds) offer an advantage in some materials applicati ons in that a partner is not necessary for a recognition event to occu r. In general, a higher number of hydrogen bonds increases the strength of associa tion, but also importan tly increases the speci ficity of recognition through preferential pairing.46-49 Strength and Specificity of Linear Systems by Mult iple Hydrogen Bonding Systems of Two and Three Linear Hydrogen Bonds Hydrogen bonding systems have evolved from systems with two linear hydrogen bonding sites to those with four. Early ex amples of the latter include t hose by Boucher and coworkers that

PAGE 21

21 feature two pyridones connect ed through rigid acetylen ic spacers (figure 1-6).50-52 Pyridones linked at the 3 and 6 positions are self-complemen tary and strongly dimerize ( Kdim ~ 104 M) into cyclic dimers as shown in figure 1-6, A Conversely, linear polymeric aggregates are formed by pyridones which are not completely self-com plementary, and are linked at the 6 and 6 positions ( B figure 1-6). Expanded cyclic structures have also b een prepared through further strategic reorganization of the pyridones about the spacer (not shown). In terms of dimerization motifs the dimers held together by four hydroge n bonds are ~ two orders of magnitude more stable ( B Kdim ~ 102 M based on the dimerization of single pyridone units). Figure 1-6. Rigidly linked pyridone units are predisposed to assemble by the arrangement of hydrogen bond donors and acceptors. An acetylene spacer links two pyridones from the 3 and 6 positions to arrange four hydrogen bonds for strong dimerization ( A Kdim ~ 104 M); incorrect programming give s mostly reversible polymers ( B ).50-52 Functional systems with three hydrogen bondi ng sites included, ear ly on, work with melamine and imide pairs, which were easily accessible, and dimerized on the order of ~ 102 M 1.38,53,54 Related complexation of acylated diaminopyridines with thym ine derivatives was explored by Lehn and coworkers and laid th e groundwork for the quadruple hydrogen-bonded systems that are widely used today.46-49 They importantly discovered that the diacylpyridine/ Npropylthymine complex B ( Ka = 800 M), figure 1-7, is an order of magnitude more stable than the complex formed from diami nopyridine and N-propylthymine, A ( Ka = 84 M). Beijer and coworkers have similarly investigated diaminotriazines.55,56 Surprisingly with acyl substitution of

PAGE 22

22 the diaminotriazine, the strength of association with propylthymine decreased from 890 to 6 M, the apparent result of electrostatic repulsion between the heterocyclic nitrogens and amide oxygens ( C figure 2-1). Figure 1-7. For the diacylpyrid ine/N-propylthymine complex B the Ka of 800 M in CDCl3 is an order of magnitude greater than the a ssociation of diami nopyridine with Npropylthymine (Ka = 84 M, in CDCl3) complex A Acylation of diaminotriazine ( C ) resulted in a lowered Ka ~ 6 M due to electros tatic repulsion. Repulsive effects in C (figure 1-2) led to unanticipat ed self-association via the mode illustrated in figure 1-8. When the amides are in the cis conformation, an ADADA array of five hydrogen bonding sites is formed. Four of the five sites can dimerize by a DADA QHB array ( Kdim = 37 M). This low value speaks to the unfavorable interactions even in this conformation, between the acetamide methyl groups and triazine nitrogen atoms. Nonetheless, the design was presented to explore systems using four hydrogen bonds. Figure 1-8. Electrostatic repulsion of amide car bonyl groups yields a lin ear array of five hydrogen bond donors and acceptors and dimerization via four hydrogen bonds ( Kdim = 37 M, in CDCl3).

PAGE 23

23 Linear Arrays of Four H ydrogen Bonds Subsequent work on QHB systems by Meijer and coworkers,53 particularly on selfcomplementary systems, has been profound. The di merization strength of DADA QHB arrays is predicted to be ~ 3.1 102 M by Jorgensens rules for s econdary interactions; the Kdim of 37 M 1 for C in figure 1-7 was much lower. Dimer stability was increased by acylating only one amino group ( F figure 1-9) to reduce electros tatic repulsions to give a Kdim = 530 M. The strength of dimerization was further increased to 2 104 M by preorganizing the QHB face via an intramolecular hydrogen bond of a urea hydrogen to a triazine nitrogen as shown in figure 1-9 F Several amido and ureidopyrimidine units have been synthesized since then that have dimerization strengths between 170 and 2 105 M. The strength of dimerization predicted by Jorgensens rules have generally been far exceeded by these DADA self-complementary QHB systems; the examination of the remaining self-complementary QHB array, DDAA, naturally followed. Figure 1-9. Strength of triazine dimerization was increased by structural modifications and preorganization. Dimerization st rength was increased to 530 M and 2 104 M in CDCl3, for F and G respectively. The pursuit for a strong QHB hydrogen bonded DDA A array lead quickly to the discovery of a isocytosine-derived ureidopyrimidinone unit (UPy).53 This unit exists in three tautomeric forms, shown in figure 1-10 as H I and J; where the equilibrium lies depends on solvent,

PAGE 24

24 concentration, and substitution at the C6 position. Two ke to forms are accessible; H is preferred when R is electron donating, and is arra nged for DDAA QHB dimerization ( Kdim ~ 107 M). Figure 1-10. Tautomeric structures of UPy are pr eorganized for different modes of dimerization and association. Tautomer H forms a strong self-complementary DDAA motif (Kdim ~ 107 M). Tautomer J forms a strong DADA hydrogen-bonding motif ( Kdim ~ 105 M). Tautomer I is organized for triple hydrogen bonding with an appropriate partner. The keto tautomer I is preorganized for association by a triple hydrogen bonding array. The third tautomer, J, is an enol tautomer which is preferred when R is electron withdrawing, and is preorganized for DADA QHB association ( Kdim ~ 105 M). The large dimerization stability of the UPy unit made it appealing for functionalization and for investigation as a building block in reversible polymeric materials. Ureidopyrimidone (UPy) has been extensivel y explored in the area of functional materials, and is being commercia lly marketed by SupraPolix as SupraB.54,57,58 A variety of low molecular weight (2000 MW) hydroxyl term inated telechelic polymers including polyethers, polyesters, and poly carbonates have been functionaliz ed with UPy end groups (figure 1-11). The resulting materials have the mechanical properties of covalent polymers, but the melt viscosity of small molecules. Fo r example, polyethylene butylene ( H1 ) is a viscous liquid at room temperature and when functionalized with UPy is a rubber-like solid.

PAGE 25

25 Figure 1-11. Hydroxy telechelic polymers func tionalized with UPy fo rm materials with properties similar to covalent polymers but melt viscosity of small organic molecules. The UPy unit has also been investigated in materials useful for tissue engineering.59 Bioactive materials were formed by incorpor ating cell adhesion promoting polypeptides functionalized with UPy that were then combined with UPy functionalized low molecular weight oligocaprolactone. Scaffolds were fabricated and implanted subcutaneously. Although the UPy UPy dimerization is weak in an aqueous e nvironment, hydrophobic shielding of the H-bonding face within the polymer film renders the dynamic binding strong. Cell adhesion was shown in vitro, and signaling of cells and a ngiogenesis were seen in vivo. Synthetically modified nucleobases are interes ting platforms for supramolecular materials. Natural nucleobases hardly self-associate, a nd weakly heteroassociat e through two or three hydrogen bonds (adenine-thymine; Ka = 50 M, and guanine-cytidine; Ka = 103 M).36,60 Through simple modifications, multiple hydrog en bonding units can be formed from these heterocycles. This has been e xplified by Meijer and coworkers with the functional unit derived from isocytosine, UPy. Also Zimmerman and coworkers have introduced a complementary ureidoguanosine (UG) unit; the first purine derivative shown in figure 1-12.61-64 With an appropriate partner, UG associates strongly; a stability of 5 107 M is obtained with DAN (2,7-diamido-1,8-naphthyridine). Association of UG is decreased by competition from

PAGE 26

26 Hoogsteen dimerization (involving N7), and a stronger dimer ( Ka 10 times larger) is obtained by conversion of N7 to CH (DeUG).61 Other nucleobases have been functionalized for use in multiple hydrogen bonding systems. Ditopic recognition un its, termed Janus bases, have been introduced to recognize mismatches in DNA.65 Further, Sessler and coworkers have intr oduced nucleobase-functionalized porphyrins as recognition units.66-68 Figure 1-12. Ureidoguanosine ( UG top) dimerizes with DAN Ka ~ 5 107 M (CHCl3), dimerization at the Hoogsteen face (bottom left) is eliminated with the synthesis of a deazaguanine (DeUG, Ka ~ 108, in CDCl3, bottom right).

PAGE 27

27 The allure of nucleobases is multifaceted. DNA or DNA-like materials offer opportunities for preparing controlled se lf-assembled structures.69 The heterocycles are inexpensive and readily available. Beyond this, purine bases in troduce attractive adva ntages over pyrimidine bases, although these have been largely untapped. Purines possess polarizable aromatic surfaces which are available for and hydrophobic interactions (figur e 1-13). These interactions can lend additional stability for functional material s. Further, purines have multiple sites for functionalization. The C2, C6, and C8 positions as well as N7 and N9 are all viable sites for modification. Further, purine nucleobases offer two H-bonding faces (Hoogsteen and WatsonCrick), which can provide homodimers. These ad ditional H-bonding sites can play important roles in self-assembly. Figure 1-13. Purines offer many opportunites for pr eparing controlled self -assembling structures. Scope and Organization of Thesis It was the interes t of the author for this thesis to contribute to the development of the biologically-inspired QHB systems for the better understand ing of these types of systems. Herein is introduced the first adenine derived self-complementary QHB system through a DADA array. The rational design of this system, which is the topic of Chapter 2, includes structural, conformational, and synthetic cons iderations. These factors are di scussed in detail, and followed with the synthesis of N9 aryl protected phe nyl ureidodiaminopurine. The methods for solution phase characterization are presented, and followed by characterization of subsequent ureidodiaminopurines.

PAGE 28

28 Reactivity issues complicated the synthesis of alkyl UDAP These complications were overcome by alternative syntheses in Chapter 3, and the systems were characterized. The structure-property relationships were examined with alterations made to the urea and N9 substituents, to consider possible electronic and steric effects. The reactivity preferences of DAP that were encountered in Chapter 3 are the ba sis of Chapter 4. Reaction of DAP with common acylating agents was investigated to shed lig ht on the differential reactivity of the NC2 and NC6 amino groups. Future directions are discussed in Chapter 5 in terms of applications of UDAP, and other interesting QHB platforms.

PAGE 29

29 CHAPTER 2 DESIGN AND SYNTHESIS OF COMPLEMEN TARY UREIDODIAMINOPURINES BY QUADRUPLE HYDROGEN BONDING Nucleobases are readily available heterocycles as platforms for the construction of selfcomplementary quadruple hydrogen bonding (QHB ) complexes. Of the nucleobases, the pyrimidines have been extensively explored as QHB systems.57,70-72 For example, in the last decade the UPy unit of Meijer et al. (derived from isocytosine) has found applications that span materials science due to its exceedingly high dimerization constant in organic solution ( Kdim ~ 107 M1 in CDCl3) via an accessible DDAA hydrogen bond ing arrangement. Despite the extended surfaces, which can be used to control ring electronics, and additional sites for functionalization, purines have onl y recently been explored in this arrangement by Zimmerman et al. with a QHB ureideoguanine.61-64 In this discourse is presen ted the first discrete, selfcomplementary, QHB unit based on a DAP platform. Design: Structural, Conformational, and Synthetic Considerations Structural and Synthetic Design The general design of a self-comple mentary QHB dimer derived from DAP is shown in figure 2-1; it entails formation of a DADA a rray along the Watson-Crick edge of the purine through regioselective urea formation at NC2. Intramolecular hydrogen bond formation between the Hc proton of the urea to N3 of the purine core serves to preorganize the monomer for dimer formation, similar to the strategy introduced for related systems in Chapter 1. Also apparent from the figure is that selection of the purine N9 (R) and urea (R1) substituents is important given that they can modulate solubility, conforma tional equilibria, and dimer stability. Two requirements surface for the purine N9 subs tituent designated R; it must a) be sterically compatible with the urea substituent (R1) so as not to discour age formation of the desired folded conformation (conversely, an inte raction which is complementary between R and

PAGE 30

30 the urea substituent, such as an aromatic interaction, could confer additional stability to the desired folded conformation) a nd b) impart suitable solubility to the monomer and dimer to allow convenient spectroscopic stud y in solution. Regarding the la tter, the purines are relatively polar heterocyclic structures and thus show pr oblematic solubility in most organic solvents. While the nucleoside ribose hydroxyl groups of na tural nucleobase precursors can be outfitted with organic protecting groups (e .g., TMS or TBDMS) for these stud ies, sugar substituents at N9 present challenges with respect to stability and size. These modifications are a topic of further discussion in Chapter 5. Benzyl substituents (CH2Ar), on the other hand, are a) known to increase the organic solubility of purines, b) chemically stable, and c) conveniently installed by SN2 chemistry at N9.73 Thus, for the initial design, benzylic N9 substituents were selected. Figure 2-1. Dimerization equilibrium for UDAP, 1N3 The urea substituent, R1, influences both intraand in termolecular hydrogen bonding in the design. For example, an electron withdrawing group might promote preorganization of the monomer by rendering Hc more acidic (e.g., the p Ka of an aryl urea (biphe nyl urea) is ~20 versus ~27 for urea).74 The same substituent could, however, diminish the basicity of the urea carbonyl group that is important for intermolecular Hbonding. Despite the risk that aromatic urea substituents might impair solubility via aromatic aggregation, phenyl was the initial choice for R1 primarily for reasons of synthetic convenience. Ongoing studies in the lab revealed that aryl

PAGE 31

31 isocyanates enjoy a considerable reactivity advantage over alkyl isocyanates in reaction with DAP (vide infra).75 Further guiding the init ial choice of R and R1 groups were results from computation and Xray crystallography. Four conformations can th eoretically be accessed by the UDAP urea in the monomeric state, two designated syn and two designated anti (figure 2-2). Of these possible conformers two are expected to be preferred due to the formation of an enthalpically favored intramolecular hydrogen bond involving Hc and either N1 ( 1N1; anti 1 ) or N3 ( 1N3; syn 1 ). These two conformers are interconverted through a single CN bond rotation (figure 2-3) and their arrangement is prerequisite for the formation of a QHB dimer. Figure 2-2. Four possible low energy conforme rs of 9-methyl-2-N-phenylureidodiaminopurine (UDAP) considered computationally. A computational approach was taken to investigate the equlibrium among the four conformers (table 2-1). When a substituent in the N9 position is too small to interact appreciably with a urea phenyl group (R1 = Ph), such as methyl (R = CH3), computation shows that the anti 2 and syn 2 conformers are ~ 5 kcal mol1 higher in energy than the 1N1 ( anti 1) and 1N3 ( syn 1) conformers (Monte Carlo conforma tional searching using MacroModel v 9.0 and the MCMM method; solvent (GB/SA) = CHCl3, force field = Amber*). Higher-level calculations were performed on the two lo west energy conformers, 1N1 ( anti 1 ) and 1N3 ( syn 1 ), and show that they are essentially isoenergetic (~ 0.55 kcal mol1 in favor of the desired 1N3 conformer) in the gas

PAGE 32

32 phase (Gaussian 03 (revision D.01), MP2/6G*//H F/6G*). The results suggest that the design shown in figure 2-1 is thermodynamically f easible, where the slight energetic preference for 1N3 will leave the monomer conformational equilibrium, and also likely Kdim, easily perturbed. Figure 2-3. Single CN bond rotation interconverts 1N1 to 1N3. Table 2-1. Relative energies (kcal mol1) of four possible UDAP conformers by computation Method 1N3 syn 1 1N1 anti 1 anti 2 syn 2 Amber* 0 0.19 5.0 4.5 MM3* 0 0.06 5.7 5.5 HF/3-21G* 0 0.75 HF/6-31G* 0 0.50 MP2/6-31G*// HF/6-31G* 0 0.55 Further evidence for the accessibility of the 1N3 conformer, when the N9 substituent is a bulky aryl group, comes in the form of a crystal structure of model compound 2 (figure 2-4). The dimethylamino substituent at C6 obviously prev ents QHB formation, but the desired mode of intramolecular hydrogen bonding nonetheless is obta ined through a planar arrangement (where N12 N3 = 2.74 ). Interestingly, in the solid st ate there does not appe ar to be a steric consequence for the bulky N9 and urea substituents but what appears as a potentially favorable interaction between the urea phenyl and the protec ting group in the form of a near edge-to-face

PAGE 33

33 Figure 2-4. X-ray crystal st ructure of model compound 2 (ellipsoids drawn at 50% probability level). The crystal structure shows an intramolecular hydrogen bond between N3 and N12 and a near edge-to-face interaction between the N9 and urea phenyl groups. A space-filling model illustrates the close cont act between the aromatic N9 and urea substituents on the right. aromatic interaction (angle between the least-sq uares planes of the aromatic rings = 86.3). Further, the space filling model to the right of the ORTEP plot shows the extent of contact between the two aromatic groups. The rings are sl ightly offset (center-to-center distance = 5.42 ) with respect to one anothe r, however, still in close co ntact (the closest carboncarbon distance equals 3.67 ). Although asse ssing a priori whether this interaction would be important to the solution-phase conformati on of the molecule is difficult, the solid-state structure does suggest that the desired conformation is sterical ly accessible. Indeed, consistent with the solidstate data, the chemical shif t of the N12H proton in CDCl3 (~ 5 mM) is significantly deshielded to 11.4 ppm (relative to TMS); N10H appears at ~ 7.2 ppm. Synthesis The three step synthesis of UDAP (fi gure 2-5) began from routinely prepared76 or commercially available 2-amino-6-chloropurine 3 Standard alkylation of N9 yielded compound 4a and 4b in the presence of K2CO3, overnight. The yields of the desired N9 product were

PAGE 34

34 diminished by competitive N7 alkylation with the ratio of the two products sensitive to the size of the benzylic bromide; 4a (N9) was obtained in 40 % yiel d with its N7 regioisomer ( 4a ) in Figure 2-5. Regioselective synthesis of UDAP 12 %, while 4b (N9) was obtained in 85 %. For this step, benzyl bromides derived from mesitylene ( a)77 and 3,5-heptyloxybenzene ( b )78,79 were selected. The methyl substituents of the mesitylene-derived protecting group were intended to sterically disrupt aromatic interactions between dimers thereby increasing solubility, while the heptyloxy derivative featured long solubilizing arms. Subsequent addition of am monia and elimination of the 6-chloro group afforded DAP, 5. Regioselective formation of 1 ensued by reaction of 5 with phenyl isocyanate in the presence of pyridine. This interesting regi ochemical preference is the topic of Chapter 4. Solution Phase Characterization of 1a Supports the Mode of Dimeriza tion Solution phase data was acquired by gHMBC, NOESY, and variable temperature (VT) NMR techniques. This data el ucidates that the desired 1N3 conformer is populated in solution and participates in QHB.

PAGE 35

35 The proton chemical shifts of 1a were unequivocally assigned by 1H13C gHMBC (figure 2-6, only pertinent information shown). The chemical shifts that are relevant to the discussion which follows are those of the urea Hb ( 8.92) and Hc ( 11.91), the methylene protons ( 5.29; N9 CH2, 2.32; CH3) of the protecting group, and the or tho proton of the urea phenyl ( 7.62). From these chemical shifts the most direct conformational eviden ce was observed by NOESY NMR (~ 2 mM, 25 C, in CDCl3). Key NOEs were found between the methylene protons of the N9 benzyl to Hc and to the ortho phenyl proton. Signifi cant NOEs were also seen from the methyl substituents of the protecting group to the ortho proton of the urea phenyl as well as to the urea proton Hc, which indicates the close spatial a rrangement of the aromatic protecting group and urea phenyl in solution. Further, the chemical shif t for the assignments of Hb and Hc ( 11.91 ppm) is in agreement with the ch emical shift of the model compound 2 (Hc, 11.4, ~ 5 mM, CDCl3). In considering the presence of the 1N1 conformer in solution, one would expect to see exchange of the Hc proton, as is seen for Hb. Exchange can occur with any other protic species in solution including solvent or H2O, but such exchange is not r eadily apparent. Although, this is not conclusive evidence for the absence of the 1N1 conformer it does speak to the stability of the 1N3 comformer in solution at the con centration and temperature studied. Conformational aspects were further clarified, and in addition, the mode of dimerization was supported by 1H VT-NMR (figure 2-7). Upon heating of a solution 1a in CDCl3 from 55 C to 55 C, Hc makes a moderate upfield shift from ~ 12.4 to ~ 11.8. On the contrary, Hb makes a substantial upfield shift from ~ 10.2 to ~ 8.1, consistent with intermolecular hydrogen bonding. Also consistent with intermol ecular hydrogen bonding is the broadening of the Hb proton signal as the temperature increases. As the temperature is raised from 55 C the

PAGE 36

36 Figure 2-6. A NOESY spectrum (~ 2 mM, 25 C, 500 MHz, in CDCl3 relative to TMS) shows the desired 1N3 conformer (proton chemi cal shifts assigned by gHMBC, not shown). signal broadens as the dimerization equilibrium becomes faster and then reaches the NMR time scale (between 5 and 15 C). The rate of exchange gradually increases and the signal is sharpened at 55 C. The intermolecularl y hydrogen bonded amino proton signal for Ha, visible at ~ 9.3 ( 55 C), shifts to ~ 8.4 (5 C) where the signal disappears by peak broadening through an intermediate exchange regime with Ha ; the latter remains at ~ 5.9 until 5 C. Fast exchange

PAGE 37

37 of the amino protons relative to the NMR time scale is reflected by an averaged peak at ~ 6.2 (55 C). Interestingly, VT NMR of 1a did not indicate dimerizat ion was occurring via QHB. This is likely a steric consequence of the mes itylene protecting group at N7 interacting with the amino protons. A subtle but important point is that based on th e trends of the chemical shifts for the amino protons, intermolecular hydrogen bonding via the H oogsteen face is not a contributing mode of dimerization at the concentration and te mperature range studied (figure 2-8, A ). Zimmerman and coworkers encountered Hoogsteen binding as a competing motif in the association of ureidoguanine as described in the introductory chapter. Like wise, two-point hydrogen bonding between the urea groups in the 1N1 conformation ( B ) is not consistent with the NOESY data and is expected to be much weaker than QHB in solution. Figure 2-7. VT NMR of compound 1a (~ 2 mM, 500 MHz, in CDCl3 relative to TMS)

PAGE 38

38 Figure 2-8. The possible dimers that can be formed through two intermolecular hydrogen bonds via 1N1. Dimer A is formed through the urea carbonyl and Hb of one purine ( b ) to the Hoogsteen face of the partner ( a). Complementary dimer B is formed through hydrogen bonds between the urea carbonyl and Hb proton. Intermolecular Hydrogen Bonding Strength The theory80-82 derived for a system consisting of a monomer and dimer in equilibrium begins with equation (1). 2 M D (1) and the dimerization constant described by (2). Kdim = 2M D (2) The concentration of monomer and dimer at a par ticular time sum to the initial concentration of monomer ([M]0, Eq. 3). 0M= M+ 2 D (3) The chemical shift that is observed for Hb is the chemical shift of th e monomer less the change in chemical shift from dimer to monomer multiplie d by the mole fraction of the initial monomer concentration that is dimer (Eq. 4a ). The mole fraction of dimer can also be expressed in terms of the monomer concentration [M] (Eq. 4b). Solving equations 4a and 4b for [D] and [M] and inserting them into the equilibrium expression and rearranging for Kdim gives expression (5).

PAGE 39

39 Taking the log of this expression and rearranging gives equation (6). A plot of the right side of this equation vs. log [M]0 will give a straight line with a y-intercept equal to log Kdim + log 2 log MD obs = M + MD 2 0M D, obs = M + MD (1 0M M) (4a and 4b) Kdim = 0 2 M2 MobsD M obs D (5) MD2 log+ log Kdim + log 0M = log M obs 2 log obsD (6) In order to determine the numerical value for Kdim, it is apparent that one must know the endpoints, D and M with reasonable certainty. It is po ssible to extrapolate a value for D from a plot of chemical shift vs. concentration. It is also possible to obtain D at low temperature or high concentration. On the contrary, M can be determined by low concentration and/or high temperature. However, extrapolation to M is typically limited by NMR instrument sensitivity. It has been illustrated that the experimental chemical shift data obtained in CDCl3 and (CDCl2)2 are essentially equivalent.6 Tetrachloroethane is an appealing al ternative for determining the limiting chemical shift values, as the higher boiling poin t (~ 146 C) in comparison to chloroform (~ 62 C) is helpful in achieving complete dissoci ation of the dimer into monomer units for measurement of M The concentration range selected for Kdim determination should fall within the range of 20 80 % of saturation (ie. fraction of solution which is dimer), and is directly related to the accuracy of the experimental values obtained for M and D .83 Outside of this saturation range the error associated increases exponentially.83 The relationship between Kdim and fraction of monomer can

PAGE 40

40 be calculated by solving equation (3) for D and inserting that value into the equilibrium expression to yield equation (7). It then follows that the fraction of monomer is 0M M. The fraction of dimer can therefore be calculated as 1 0M M. M = (1 + 8Kdim 0M) dim4 1K (7) An observed chemical shift value which is central to M and D if accurate, will correspond to a solution in which the fraction of dimer is 0.5. Dilution studies were performed at 25 C, and the chemical shift of Hb monitored by 1H NMR to obtain the intermolecular dimerization constant, Kdim for 1a and 1b. The Kdim values were determined as the averag e of four runs which were cal culated based on the theory of Nogales82 and then compared to values cal culated using standard software.84,85 Dilution studies were performed at starting conc entrations between 3.5.7 mM for each of the four sets, and the chemical shift of Hb monitored. A stock solution was diluted sequentially to a total volume of 1 mL in increments of 0. 05 mL with 8 iterati ons (figure 2-9). The dimerization constants were calculated for each dilution set using the experimentally determined D and M and obtaining graphical repr esentations of the data by equation 6 (figure 2-10), and the values averaged. The dimerization constants were calculated to be 820 160 M1 (20 % dimer) and 980 290 M1 (20 % dimer) for compounds 1a and 1b, respectively. Dimerization constants were al so calculated by non-linear leas t squares analysis using a program developed by Sanderson et al.85 This program uses the same theoretical principles as described previously, with the exception that theM and D values are calcul ated by an iterative approach. The Kdim values obtained from this method were in close agre ement with the values

PAGE 41

41 Figure 2-9. Representative dilution of 1a (25 C, 500 MHz, in CDCl3). Figure 2-10. Dilution data in figure 2-7 for 1a and calculation of Kdim calculated using experimentally determined e ndpoints. The average calculated values were 1100 360 M1 (20 % dimer) for 1a and 1600 380 M1 (20 % dimer) for 1b. The deviation in Kdim is due to the calculated endpoints. However, the calcula ted monomer chemical shift ( 7.2 b = log Kdim + log 2 log MD ; 2.649 = log Kdim + log 2 log (10.2 7.2) log Kdim = 2.825 K dim ~ 670 M1

PAGE 42

42 0.1) was nearly equivalent to the experimental value ( 7.2 0.01; 120 C, (CDCl2)2, ~ 1.5 mM)). The chemical shift of the fully di merized species was calculated to be 10.1 0.1 and experimentally determined to be 10.2 0.01 ( 55 C, (CDCl2)2, ~ 3 mM). Even a discrepancy of 0.1 ppm in D is enough to introduce significant uncertainty in the value of Kdim. Clearly the level of error by approximation is less favorable when using the previ ously developed program, however, the values for the equi librium constant are in clos e agreement when the endpoint values are taken into consideration. The values of Kdim for 1a and 1b are in close agreement given the experimental error. Chapter 3 discusses the structure-property relatio nships for this system in terms of R and R1. It will become apparent that much larger groups can be accommodated than originally thought. Dimerization in the Solid State X-ray crystallography is a pow erful tool for confirming hydrogen bonding modes. There are clear preferences for sp ecific hydrogen bonding patterns in crystal structures. These preferences are combined hierarchally to form e mpirical rules which can be used for predicting hydrogen bond patterns from a limited number of functional groups.32 Further, in the absence of other forces, these rules can be used as indi cators for hydrogen bond preferences in the solid state or in solution.32 In the solid state, compound 1a features both the 1N1 and 1N3 conformers and the expected hydrogen bonding and aromatic in teractions. As per the empirical rules mapped out by Margaret Etter,32 all intramolecular hydrogen bonding possi bilities are utili zed, and all the hydrogen bond donor and acceptor sites available ar e satisfied in the crystal lattice. The DADA QHB mode of dimerizat ion is confirmed in the so lid state (figure 2-11). The 1N3 conformer is organized through an intramolecular hydrogen bond (2.678 table 2-2), and intermolecular hydrogen bonds are pr esent as expected. Intermol ecular hydrogen bonds are seen

PAGE 43

43 from NH19O1 and NH10N1 of each UDAP with averaged distances of 2.786 and 3.291, respectively. Other interesting conformational information is available. The plane of the urea phenyl is twisted with respect to the plane of th e purine core by 20.6. More interestingly, the N9 2,4,6-trimethylbenzyl substituent is in an anti conformation 86.5 out of plane with the purine core, in contrast to the N9 be nzyl substituent of model compound 2. Figure 2-11. 1N3 Conformer QHB motif in the solid stat e (hydrogens removed for clarity). Table 2-2. Hydrogen bond distances () for QHB 1a in the solid state. DHA Distance () N12HN3 2.678 N19HO1 2.786 N10HN1 3.291 The extended crystal structure of 1a also shows the presence of a preorganized (N12 HN3) 1N1 conformer. Hydrogen bonding from the urea of a preorganized 1N1 conformer (a; N10HN7) to the Hoogsteen edge of a 1N3 conformer (b; N19HO1) is seen (figure 2-12, for bond distances see table 2-3). Similar to the 1N3 conformer, the N9 substituent is nearly perpendicular to the purine core (83. 81). It also follows that for the 1N1 conformer the urea

PAGE 44

44 phenyl is twisted out of the plane of the purine core by 25.86 degrees, similar to the intramolecular arrangement for 1N3. Figure 2-12. Hydrogen bonding to the Hoogsteen face of 1a in the 1N3 conformation to the urea of a UDAP in the 1N1 conformation Table 2-3. Hydrogen bond distance of 1a participating in Hoogsteen dimerization in the solid state. DHA Distance () N12HN1 2.664 N10HN7 3.061 N19HO1 2.748 Summary The general design and synthesis of a self-c omplementary QHB dimer derived from DAP has been outlined. Regioselective urea formation occurs at NC2, and a DADA QHB motif is presented via preorganiza tion of the urea into a 1N3 syn 1 conformation. This conformation was shown to exist in solution by NOESY NMR in CDCl3 (~ 2 mM). Further, by VT NMR, the conformation is again verified, and QHB is shown to be the predominant mode of assembly at

PAGE 45

45 the temperature and concentration ranges studi ed. Dimerization constant s were derived from sequential dilution and monitoring of 1H chemical shifts by NMR. Calculated Kdim values were 820 160 M1 (20 % dimer) and 980 290 M1 (20 % dimer) for compounds 1a and 1b, respectively. X-ray crystallo graphy confirmed QHB via the desired DADA motif, but also confirmed the possible Hoogsteen dimer. Experimental Methods General protocols Reagents were purchased from Acros or Aldrich, and were used without further purification unless stated otherwise. Dry solvents were de gassed and purified under an atmosphere of argon using the GlassContour solven t system (GlassContour, Inc.). Pyridine was distilled onto 3 activated molecular sieves. Column chromatography was carried out using Whatman 230 mesh silica gel. Thin layer chromatography (TLC) was performed on Duracil TLC aluminum sheets with visualization by UV li ght. Melting points (Mp) were determined on a MEL-TEMP melting apparatus and are uncorrected. 1H (300, 500 MHz) and 13C (75, 125 MHz) nuclear magnetic resonance (NMR) spectra were recorded on Varian Gemini 300, Mercury 300BB, and Inova 500 spectrometer at room temp erature unless otherwis e specified. Chemical shifts ( ) are given in parts per million (ppm) relative to TMS and referenced to residual protonated solvent (CHCl3: 1H 7.24 ppm, 13C 77.0 ppm; DMSO: 1H 2.49 ppm, 13C 39.5 ppm). Abbreviations used are singlet (s), doublet (d), triplet (t), multiplet (m), and broad (b). High resolution mass spectrometry (HRMS) spectra were recorded on a Finnigan LCQ-Ion Trap Spectrometer.

PAGE 46

46 Synthetic details N N N N Cl NH2 2-Amino-6-chloro-N-9-(2,4,6-tr imethylbenzyl)purine (4a). 6-Chloro-2-aminopurine76 3 (2.08 g, 11.1 mmol), 2-bromomethyl-1,3,5-trimethyl-benzene77 (3.52 g, 16.5 mmol), and K2CO3 (2.30 g, 16.6 mmol) were placed in an oven-dried round-bottomed flask and dried under argon. Dry DMF (180 mL) was added and the mixture was stirred overnight at ro om temperature. The solvent was removed under vacuum and the re sidue was purified by column chromatography on silica gel (5% CH3OH/CH2Cl2) to yield a yellow/white powder (1.42 g, 40%). Mp 205 C; 1H NMR (CDCl3) 2.25 (s, 6H), 2.31 (s, 4H), 5.17 (s, 4H), 6.94 (s, 2H), 7.25 (s, 1H); 1H NMR (DMSO-d6) 2.23 (s, 3H), 2.25 (s, 6H), 5.16 (s, 2H), 6.92 (s, 2H), 6.95 (s, 2H), 7.55 (s, 1H); 13C NMR (DMSO-d6) 19.4, 20.5, 123.1, 128.2, 128.3, 129.1, 129.2, 137.5, 141.7, 149.3, 153.9, 159.7; HRMS (ESI-FT-ICR) calculated for C15H16N5Cl (M + H)+ 302.1167, found 302.1166. N N N N Cl NH2 2-Amino-6-chloro-N-7-(2,4,6-tr imethylbenzyl)purine (4a ). Using conditions identical to those designed for 4a, the N7 regioisomer was isolated as a ye llow/white solid from 5.30 g (28.1 mmol) of starting material 3 to give 1.04 g (12%) of 4a. Mp 223 C; 1H NMR (CDCl3) 2.25 (s, 6H), 2.30 (s, 3H), 5.16 (s, 2H), 5. 26 (bs, 2H), 6.94 (s, 2H), 7.25 (s, 1H); 1H NMR

PAGE 47

47 (DMSO-d6) 2.22 (s, 3H), 2.24 (s, 6H), 5.14 (s, 2H), 6.90 (s, 2H), 6.95 (bs, 2H), 7.54 (s, 1H); 13C NMR (DMSO-d6; 100 C) 19.4, 20.5, 41.1, 123.2, 128.3, 129.2, 137.5, 137.6, 141.8, 149.4, 154.0, 159.8. N N N N NH2 NH2 2,6-Diamino-N-9-(2,4,6-trimethylbenzyl)purine (5a). Vacuum dried starting material 4a (1.05 g, 3.30 mmol) was placed in a 330 mL pre ssure tube. Methanolic ammonia (95 mL, 7 N) was added to the solid followed by heating under pressure to 90 C for 16 h. The solvent was removed under vacuum and the residue was purified by chromatogr aphy on silica gel (5% CH3OH/CH2Cl2) to yield 5a (0.640 g, 50%) as a white powder. Mp 275 C; 1H NMR (CDCl3) 2.24 (s, 6H), 2.29 (s, 3H), 4.72 (bs, 2H), 5. 11 (s, 2H), 5.30 (bs, 2H), 6.92 (s, 2H), 6.97 (s, 1H); 1H NMR (DMSO-d6) 2.24 (s, 9H), 5.04 (s, 2H), 5.80 (bs, 2H), 6.65 (bs, 2H), 6.91 (s, 2H), 7.04 (s, 1H); 13C NMR (DMSO-d6; 100 C) 18.9, 20.1, 40.2, 112.9, 128.7, 128.7, 135.6, 136.9, 137.1, 151.7, 155.8, 159.9; HRMS (ESI-FT-ICR) calculated for C15H18N6 (M + H)+ 283.1666, found 283.1668. N N N N NH2 NH2 2,6-Diamino-N-7-(2,4,6-trimethylbenzyl)purine (5a ). This compound was prepared by the same procedure as 5a from 1.00 g (3.11 mmol) of 4a to yield 5a as a white solid (0.530 g,

PAGE 48

48 42%). Mp 282 C; 1H NMR (CDCl3) 2.24 (s, 6H), 2.29 (s, 3H), 4.72 (bs, 2H), 5.11 (s, 2H), 5.29 (bs, 2H), 6.92 (s, 2H), 6.97 (s, 1H); 1H NMR (DMSO-d6) 2.24 (s, 6H), 2.26 (s, 6H) 5.08 (s, 2H), 5.46 (bs, 2H), 6.27 (bs, 2H), 6.91 (s, 2H), 7.05 (s, 1H); 13C NMR (DMSO-d6; 100 C) 18.7, 19.9, 113.0, 128.5, 128.6, 135.5, 136.8, 136.9, 151.6, 155.7, 159.8. N N N N NH2 N H N H O 6-Amino-N-9-(2,4,6-trimethylbenzyl)-2 -N-(4-phenylamino)ureidopurine (1a). Compound 5a (0.050 g, 0.21 mmol) was placed in an oven-dried two-necked round-bottomed flask and dried under vacuum. Under argon atmosphere CH2Cl2 (29 mL) was added to the solid and the mixture was heated to 50 C to dissolve 5a. When the starting material was completely dissolved, the temperature was lowe red to 40 C. Pyridine (0.034 mL, 0.43 mmol) and phenyl isocyanate (0.051 mL, 0.47 mmol) were added. The mixture was stirred at 40 C for 20.5 h followed by evaporation. Purification by column chromatography on silica gel (1% MeOH/CH2Cl2) afforded 1a (0.058 g, 70%). Mp 263 C; 1H NMR (CDCl3) 2.29 (s, 6H) 2.32 (s, 3H), 5.26 (s, 2H), 6.96 (s, 2H), 7.09 (t, J = 7.1 Hz, 3H), 7.16 (bs, 1H), 7.34 (t, J = 7.3 Hz, 2H), 7.59 (d, J = 7.6 Hz, 2H), 9.34 (bs, 1H), 12.04 (s, 1H); 1H NMR (DMSO-d6) 2.24 (s, 3H), 2.26 (s, 6H), 5.20 (s, 2H), 6.92 (s, 2H), 7.02 (t, J = 7.0 Hz, 1H), 7.30 (t, J = 7.3 Hz, 2H), 7.37 (s, 1H), 7.61 (s, 2H), 7.73 (d, J = 7.7 Hz, 2H), 9.31 (s, 1H), 11.82 (s, 1H); 13C NMR (DMSO-d6; 100 C) 18.8, 19.9, 40.7, 114.6, 119.3, 122.2, 128.1, 128.7, 137.0, 137.1, 137.9, 138.6, 149.8, 151.4, 153.1, 155.2; HRMS (ESI-FT-ICR) calculated for C19H17N7O (M + H)+ 402.2037, found 402.2027; calculated for 2(C19H17N7O) (2M + H)+ 803.4001, found 803.4088.

PAGE 49

49 N N N N NH2 N H N H O 6-Amino-N-7-(2,4,6-trimethylbenzyl)-2 -N-(4-phenylamino)ureidopurine (1a ). Starting material 5a (0.052 g, 0.22 mmol) was dried under v acuum in a 50 mL two-necked roundbottomed flask fitted with a reflux condenser. CH2Cl2 (45 mL) was added and the mixture was heated to reflux until 5a was dissolved. The temperature was reduced to 40 C. Without equilibration of the temperature, pyridine (0.035 mL, 0.44 mmol) was added followed by phenyl isocyanate (0.024 mL, 0.22 mmol). The mixtur e was stirred 24 h. The solvent was removed under vacuum and the residue was purified by column chromatography on silica gel (2% CH3OH/CH2Cl2) to yield 1a (0.064 g, 72%). Mp 243 C; 1H NMR (CDCl3) 2.27 (s, 3H), 2.32 (s, 3H), 4.94 (s, 2H), 5.17 (s, 2H), 6.95 (s, 2H), 7.13 (t, J = 7.1 Hz, 1H), 7.35 (m, 2H), 7.36 (s, 1H), 7.59 (d, J = 7.5 Hz, 2H), 7.81 (s, 1H), 11.52 (s, 1H); 1H NMR (DMSO-d6) 2.23 (s, 3H), 2.26 (s, 6H), 5.13 (s, 2H), 6.83 (s, 2H), 6.92 (s, 1H), 7.05 (t, J = 7.0 Hz, 1H), 7.32 (t, J = 7.3 Hz, 2H), 7.38 (s, 1H), 7.75 (d, J = 7.7 Hz, 2H), 9.35 (s, 1H), 11.81 (s, 1H); 13C NMR (DMSO-d6) 19.4, 20.6, 113.1, 119.9, 123.0, 128.6, 128.7, 129.1, 137.4, 137.5, 138.5, 138.9, 150.0, 151.2, 153.1, 158.6. N N N N Cl NH2 C7H15O OC7H15

PAGE 50

50 2-Amino-6-chloro-N-9-(3,5-bis-heptyloxybenzyl)purine (4b). 6-Chloro-2-aminopurine76 3 (0.175 g, 0.931 mmol), 2-brom omethyl-3,5-bis-heptyloxybenzene78,79 (0.464 g, 1.16 mmol), and K2CO3 (0.322 g, 2.33 mmol) were placed in an oven-dried round-bottomed flask and dried under argon. Dry DMF (50 mL) was added to the solid mixture and the mixture was stirred overnight at room temperature. The solvent was removed under vacuum and the residue was purified by column chromatography on silica gel (5% MeOH/CH2Cl2) to yield a yellow/white powder (0.386 g, 85%). Mp 110 C; 1H NMR (CDCl3) 0.86 (t, J = 6.6 Hz, 6H), 1.35 (m, 16H), 1.72 (m, 4H), 3.87 (t, J = 3.9 Hz, 4H), 5.14 (s, 2H), 5.20 (s, 2H), 6.34 (m, 3H), 7.74 (s, 1H); 1H NMR (DMSO-d6) 0.85 (t, J = 6.6 Hz, 6H), 1.32 (m, 16H), 1.64 (m, 4H), 3.88 (t, J = 3.9 Hz, 4H), 5.17 (s, 2H), 6.38 (s, 3H), 6.96 (s, 2H), 8.21 (s, 1H); 13C NMR (DMSO-d6) 13.9, 22.0, 25.4, 28.4, 28.6, 31.2, 38.7, 46.1, 67.4, 99.8, 105.7, 123.2, 138.7, 143.15, 149.43, 154.0, 159.9, 160.0; HRMS (ESI-FT-ICR) calculated for C26H40N6O2 (M + H)+ 488.2787, found 488.2790. N N N N NH2 NH2 C7H15O OC7H15 2,6-Diamino-N-9-(3,5-bis-heptyloxybenzyl)purine (5b).Vacuum dried starting material 4b (0.323 g, 0.717 mmol) was placed in a 100 mL pressure tube. Methanolic ammonia (60 mL, 7 N) was added to the solid followed by heating unde r pressure to 90 C for 19 h. The solvent was removed under vacuum and the residue was pur ified by column chromatography on silica gel (5% CH3OH/CH2Cl2) to yield 5b (0.231 g, 69%) as a white powder. Mp 151 C; 1H NMR (CDCl3) 0.85 (t, J = 6.6 Hz, 6H), 1.34 (m, 16H), 1.70 (m, 4H), 3.84 (t, J = 3.9 Hz, 4H), 4.88

PAGE 51

51 (bs, 2H), 5.067 (s, 2H), 5.68 (bs, 2H), 6.33 (m, 3H), 7.44 (s, 1H); 1H NMR (DMSO-d6) 0.85 (t, J = 6.6 Hz, 6H), 1.30 (m, 16H), 1.64 (m, 4H), 3.87 (t, J = 3.9 Hz, 4H), 5.06 (s 2H), 6.34 (s, 2H), 6.66 (s, 2H), 7.75 (s, 1H); 13C NMR (CDCl3) 14.3, 22.8, 26.2, 29.2, 29.4, 32.0, 46.9, 68.3, 100.8, 106.3, 114.4, 138.2, 152.4, 156.1, 160.2, 160.9; HRMS (ESI-FT-ICR) calculated for C26H40N6O2 (M + H)+ 469.3286, found 469.3285. N N N N NH2 N H N H C7H15O OC7H15 O 6-Amino-N-9-(3,5-bis-heptyloxybenzyl)-2 -N-(4-phenylamino)ureidopurine (1b). Compound 5b (0.055 g, 0.12 mmol) was placed in an ove n-dried round-bottomed flask and dried under vacuum. Methylene chloride (6 mL), pyridine (0.012 mL, 0.015 mmol), and phenyl isocyanate (0.017 mL, 0.015 mmol) were added sequentially. The mixture was stirred at room temperature for 19 h. Solvent was removed unde r vacuum. Purification of the residue was performed by column chromatography on silica gel (1% MeOH/CH2Cl2) to afford 1b (0.050 g, 73%). Mp 217 C; 1H NMR (CDCl3) 0.86 (t, J = 6.6 Hz, 6H), 1.26 (m, 16H), 1.69 (m, 4H), 3.80 (t, J = 3.9 Hz, 4H), 5.23 (s, 2H), 6.33 (m, 1H), 6.37 (m, 2H), 7.05 (t, J = 7.0 Hz, 1H), 7.27 (t, J = 7.3 Hz, 3H), 7.37 (d, J = 7.4 Hz 3H), 7.65 (s, 1H), 9.41 (s, 1H), 11.84 (s, 1H); 1H NMR (DMSO-d6) 0.83 (t, J = 6.6 Hz, 6H), 1.22 (m, 16H), 1.59 (m, 4H), 3.82 (t, J = 3.8 Hz, 4H), 5.25 (s, 2H), 6.35 (s, 1H), 6.41 (s, 2H), 7.00 (t, J = 7.0 Hz, 1H), 7.26 (t, J = 7.2 Hz, 2H), 7.58 (d, J = 7.6 Hz, 2H), 7.61 (s, 1H), 8.11 (s, 1H), 9.30 (s, 1H), 11.75 (s, 1H); 13C NMR (DMSO-d6; 100 C) 13.3, 21.5, 25.0, 27.9, 28.2, 28.5, 30.7, 46.0, 67.4, 100.4, 105.7, 114.8, 119.2, 122.2, 128.2, 138.7, 139.7, 149.5, 151.4, 153.4, 155.5, 159.9; HRMS (ESI-FT-ICR) calculated for C33H45N7O3

PAGE 52

52 (M + H)+ 588.3657, found 588.3678; calculated for 2(C33H45N7O3) (2M + H)+ 1175.7241, found 1175.7484. N N N N N N H N H O N-9-Benzyl-6-dimethylamino-2-N-(4 -phenylamino)ureidopurine (2). Phenyl isocyanate (0.65 mL, 6.0 mmol) was added dropwise to 2-am ino-6-dimethylamino-9-be nzyl purine (0.10 g, 0.37 mmol) dissolved in dry pyridine (7.5 mL). After stirring at room temperature for 1 h the crude reaction mixture was concentrated under reduced pressure and the crude solid was recrystallized with ethano l to give a white solid (0.137 g, 94%). Mp 234 C; 1H NMR (CDCl3) 3.56 (bs, 6H), 5.31 (s, 2H), 7.16 (s, 1H), 7.33 (m, 10H), 7.61 (s, 1H) 11.39 (s, 1H); 1H NMR (DMSO-d6) 3.62 (bs, 6H), 5.40 (s, 2H), 7.28 (m, 10H), 8.13 (s, 1H), 9.41 (1.37, 1H), 11.45 (s, 1H); 13C NMR (DMSO-d6) 46.1, 60.0, 115.46, 119.17, 122.7, 126.8, 126.9, 127.6, 128.7, 128.8, 136.8, 138.7, 138.9, 151.8, 153.0, 154.1. HRMS (ESI-FT-ICR) calculated for C21H22N7O (M + H)+ 388.1886, found 388.1921. Computational details Monte Carlo conformational searching was done on a Dell PC (2.4 GHz) running the Fedora Core using MacroModel v. 9.0 (Schrodinger, LLC)86 and the MCMM method (relevant parameters include: steps = 100, itera tions = 2000, solv ent (GB/SA) = CHCl3, force field = Amber*). Only 1N3 and 1N1 were further refined using ab in itio methods (using Gaussian 03 (revision D.01)87) as implemented through the National Ce nter for Supercomputing Applications, SGI Altix cluster Cobalt (http://www.ncsa.uiuc.e du/ UserInfo/Resources/Hardware/SGIAltix/).

PAGE 53

53 NMR Experimental parameters General. NMR spectra were recorded at 25 C on a Varian Inova spectrometer equipped with a 5 mm indirect detecti on probe, operating at 500 MHz for 1H and at 125 MHz for 13C. Chemical shifts are reported in ppm relative to TMS. gHMBC. The gHMBC spectrum was recorded with the standard pulse sequence in vnmr, in 512 increments, each acquired in 16 transients. The number of points in the FIDs was 4k, and the same number was used for the spectrum in f2 The number of points for the spectrum in f1 was 2k. The preacquisition delay was 0.5 s. The 8 Hz spectrum was taken in CDCl3 at an approximate concentration of 2 mM. All shifts are reported in ppm dow nfield of TMS, and 13C shifts are listed in bold. NOESY. The NOESY spectrum was recorded at 25 C with the standard pulse sequence in vnmr, in 2k increments, each acquired in 32 transients. The number of points in the FIDs was 4k, and the same number was used for the spectrum, in both f1 and f2. The preacquisition delay was 1 s and the mixing time 0.5 s. Spectrum was obtained in CDCl3 at an approximate concentration of 2 mM. VT-NMR. The variable temperature spectrum was recorded on a sample in CDCl3 (~ 4 mM), on automation, arra ying the temperature from 55 C to 55 C in steps of 10 C. For each change in temperature, a delay of 300 s allowed for the temperat ure equilibration, followed by shimming z1z2 on the lock level, then acquisition in 128 transients with an acquisition time of 5 s.

PAGE 54

54 CHAPTER 3 STRUCTURE-PROPERTY RELATIONSHIPS IN UDAP DERIVATIVES Introduction Meijer and coworkers have developed self-complementary DADA quadruple hydrogen bonding (QHB) motifs from both triazines (A, figure 3-1) and pyrimidines (B, figure 3-1). The ureidotriazine unit (A) enjoys a Kdim of 2 104 M1 in CDCl3; the pyrimidine unit B dimerizes even more efficiently, with a Ka on the order of 105 M1 in CDCl3.53 The increase in dimerization strength from A to B is apparently realized through a) the incr eased basicity of the pyrimidine nitrogen (diaminotriazine pKa ~ 2, diaminopyrimidine pKa ~ 7) and b) acylation of the 6-amino group that increases the acidit y of the pyrimidine H-bond donor.88,89 Given the structural similarity of A and B to 1a and 1b, it is initially surprising that the UDAP systems dimerize at least an order of magnitude more weakly. To wit, the basicity of the nitrogen of interest for UDAP dimerization, N1, has a pKa (~ 5; diaminopurine) that falls between the values for triazine and pyrimidine.90-92 The goal of this chapter is to draw structure-property relationships within the UDAP systems that, once id entified, will rationalize Kdim and potentially suggest ways that it can be tuned. Figure 3-1. Ureidotriazine (A; Kdim = 2 4 M 1) and ureidopyrimidine (B; Kdim = 2 5 M1) of Meijer and coworkers form hi ghly stable DADA QHB dimers.

PAGE 55

55 Three new targets (79) were conceived to draw struct ure-property relationships in the UDAP platform (figure 3-2); each is specifically designed to probe how th e structure of the urea and/or N9 substituent might alter conformational preferences, solubility, and ultimately Kdim. Exploring substituent effects more broadly should reveal a) the tolerance of the system to functionality in these positions th at could potentially be used to covalently connect the UDAP group to a polymeric backbone or another UDAP unit and b) strategies to tune Kdim by changes that are remote from the hydrogen bonding in terface. The monocyclic ureidotriazines and pyrimidines, shown in figure 3-1, are not easily amenable to the latter studies. Figure 3-2. UDAP derivatives designed to probe structur e-property relationships. An obvious difference between UDAPs 1a/1b and compounds A and B is the class of urea substituent. Alkyl ureas are typically used by Meijer, Zimmerman, and others for solubility reasons and synthetic feasibility The poor reactivity of alkyl isocyanates with DAP precluded their use in the first UDAP molecules (Chapter 2). If suitable synthe tic methods could be developed to prepare a compound like 7, that incorporates a hexyl urea substituent, two

PAGE 56

56 important features could be e xplored. The contribution of a more basic urea carbonyl oxygen (and commensurately less acidic Hc proton) will highlight the role of electronic effects in the dimerization. A priori, it is difficult to determin e which consequence, a weakened intramolecular hydrogen bond between Hc and N3 or a potentially streng thened intermolecular hydrogen bonding interface, is more important to the overall thermodynamics of the system. Secondly, an alkyl urea might expose the influence, if any, of the aromatic inter action (between the N9 substituent and the aryl urea) s uggested by the NOESY NMR data for 1a (and crystal structure of 1a) by deleting this contact. The dependence of the dimerization propert ies on an interaction between the N9 substituent and the urea will be further test ed by replacement of the N9 aryl group of 1a and 1b with a heptyl chain as in 8. Studies with this derivative could potentially also rationalize, in part, the lack of QHB observed for 1a (N7 protected UDAP, Chapter 2). A final target, 9, that bears a tert-butyl acetate group at N9, will test the tole rance of UDAP dimerization to a functional group handle that has been remarkably successful in th e peptide nucleic acid (PNA) field. Subsequent hydrolysis of the ester moiety and coupling to amino acids, for exam ple, could provide a building block appropriate for chain extension (figure 3-3).93,94 The tert-butyl acetate group will also explore possible stabilizat ion of the desired intramolecular hydrogen bonding conformation through introduction of a favorable el ectrostatic inter action between Hc and the ester carbonyl. Discussed herein is how the three ureidodiaminopurine targets 79 have been synthesized by new methods, and studied in solution by the techniques outlined in Chapter 2. The results of the structure-property studies are then discussed with respect to Kdim, conformation, enthalpy and entropy effects, and secondary elect rostatic interactions (e.g. Chapter 1) to leverage future optimization and application of these systems.

PAGE 57

57 Figure 3-3. Ester hydrolysis of 9 and subsequent coupling with an amino acid residue for future application in a PNA. Structure Property Relationships Synthesis of 7, 8, and 9 The synthesis of UDAPs 1a and 1b was described in Chapter 2. The appropriate diaminopurine (DAP) was first prepared from N9 protected 2-amino-6-chloropurine, and the phenyl urea was subsequently fo rmed regioselectively at NC2 by reaction with phenyl isocyanate to give 1a and 1b in fair yields. Preferential re activity of acylating agents at NC2 of DAP has been demonstrated in the literature and is further discussed in depth in Chapter 4.95 It was observed, however, that alkyl isoc yanates did not react readily at NC2 under the same, or even harsher conditions. The overall re activity differences between aryl and alkyl isocyanates toward DAP can likely be rationalized on the basis of trans ition state stabilization a nd electronic effects. Figure 3-4 shows the relative stab ilities of the transition states. The reactivity difference originates from electronic differences between alkyl and aryl isocyana tes and the respective intermediates. The phenyl isocyanate transition state is lower in energy due to resonance stabilization by the phenyl group in the reaction intermediate. Theref ore the reaction is faster due to its lowered activation barrier. This interesti ng observation spurred an investigation into the nucleophilic reactivity difference of the amino group of 2,6-diaminopurine, and is the subject of

PAGE 58

58 Chapter 4. It follows that formation of the relevant NC2 hexylureidopurines 7, 8, and 9 was achieved by exploring other strategies to obtain hexyl ureidodiaminopurines. Figure 3-4. The reaction of phenyl isocyanate with DAP is fa ster than reaction of hexyl isocyanate due to the lowered energy of the transition state. To avoid the N7/N9 regioselectivity comp lications observed in the alkylation of 3 with benzylic bromides to form 4a and 4b (Chapter 2), a modified procedure was used for the synthesis of the heptyl (4c) and ester (4d) protected purines (figure 3-5). Conditions were used reported by Brik and coworkers for the synthesis of N9 alkylated purines under conditions mild enough for combinatorial reactions,96 where deprotonation and alkylation of N9 was afforded in minutes by TBAF. Similarly, 4c and 4d were both afforded in 75 % yield in about ten minutes (figure 3-5). Upon alkylation of 4c it was converted to DAP 5c by amination. Synthesis of 5d

PAGE 59

59 was not amenable to the harsh conditions requir ed for amination and was thus synthesized by direct alkylation of DAP. Figure 3-5. Synthesis of 4c and 4d using TBAF as base yields predominantly the N9 isomer in ten minutes. Amination of 4c yields 5c, and 5d is prepared by alkylation of N9 directly. The synthesis of UG by Zimmerman and co workers was achieved by deprotonation of guanine with sodium hydride.62 Along this vein, deprotonation of DAP by a strong base was investigated. In accordance with the a pparent increased nuc leophilicity of NC2, initial deprotonation occurred at NC6 giving an N6-hexylureidopurine with n-BuLi or NaH (figure 3-6, experimental data presented in Chapter 4). An alternative approach to access NC2 alkyl urea derivatives involves reaction at NC2 before introduction of the competitive amino gr oup at C6, in other words, deprotonation of suitably N9 functionalized 2-amino-6-chloropurines 4ad. Formation of the 6-chloro-2-N-hexyl ureidopurines 7 and 8 was realized through de protonation of 2-amino-6chloropurine derivatives 6a and 6c by n-BuLi followed by reaction of the anion with hexyl isocyanate (figure 3-7).

PAGE 60

60 Sodium hydride and n-BuLi reactions were performed concurrently, and n-BuLi was the preferred method for convenience. Subse quent formation of UDAP was achieved by displacement of the chloro group with ammonia (in methanol). Figure 3-6. Deprotona tion of DAP with n-BuLi or NaH and reaction of the anion with hexyl isocyanate gives 6-N-hexylureidopurine (for general procedures see Chapter 3 experimental). Figure 3-7. The synthesis of UDAP 7 and 8 by deprotonation, substitution with hexyl isocyanate, and displacement of the 6-chloro group Formation of 7 and 8 were low-yielding reactions, 46 % and 34 %, respectively. Interestingly, the reason for the low yields was the apparent displa cement of the chloro substituent by methoxide formed in situ, th e product of which was isolated in 34 % for 13 and 56 % for 14. The formation of the methoxy substitute d products was initially overlooked, and was realized in attempts to synthesize heptyl protec ted N9 phenylureidopurines.

PAGE 61

61 The presence of the N9 ester substituent prompt ed use of a different route for formation of 9. The pKa of the -proton of an ester is ~ 25 while the pKa of the 2-amino group of 2-amino-6chloropurine 4d is likely ~ 30 (value for PhNH2).97 Not surprisingly, co mpetitive deprotonation of the ester was observed when n-BuLi was used. Selective depr otonation of the more accessible amino group was afforded through the use of a bulkier base, LTMP, at low temperature. Treatment of this intermediate with hexyl isocyanate provided th e isocyanate product 6d in respectable yield (69 %, figure 3-8). Installation of the amino functionality at C6 was then achieved via a two-step procedure; displacement of the chloro group by sodium azide (to avoid disruption of the t-butyl ester) followed by catalytic transfer hydrogenation using ammonium formate as the hydrogen source. Figure 3-8. The synthesis of 9 was achieved through deprotonation of 4d with LTMP followed by substitution with hexyl isocyanate to form 6d. The 6-chloro group was substituted by sodium azide and the azide subseque ntly reduced by ammonium formate. Properties of UDAP in Solution The solution-phase properties of 7, 8, and 9 were all investigated analogously to 1a and 1b (Chapter 2). Conformational considerations we re addressed by NOESY NMR, while modes of dimerization were investigated by VT NMR. Like wise, solution-phase di merization strength was quantified by monitoring the chemical shift of Hb upon sequential d ilution in CDCl3. Ureidodiaminopurines 79 all populate the desired N3 hydrogen-bonded conformation in solution at mM concentrations in CDCl3 at 25 C based on NOESY NMR. The key NOEs

PAGE 62

62 observed in each case are similar to 1a and 1b, between the urea Hc proton and the methylene protons at the N9 position. For 7, this conformation was even found to be stable up to 75 C in DMSO (a strong hydrogen-bond competitor). NOE contacts that would reflect the N1 conformer (e.g., between Hc and the 6-amino protons) have not been observed in any case. Even so, this conformation is presumably accessible in solution. The trends identified for 7 by variable temperature NMR spectroscopy are similar to those found for 1a and 1b. Differences do arise in chemical sh ift as a consequence of conversion of the aryl urea to an alkyl urea. Hc, since less acidic, is co mparatively upfield in 79 (e.g. at 25 C and 2 mM: 1a, 12.4; 7, 11.8; 8, 9.6; 9, 9.4); this proton is also split into a triplet (3J = 2.5 Hz) from the neighboring methylene gr oup making its assignment (versus Hb) trivial. A representative VT experiment is shown for 8 in Figure 3-9. At 55 C, Ha appears at ~ 9.5, and Ha ~ 5.9. The Ha resonance moves upfield with increasi ng temperature (as it responds to the dimer concentration) while Ha remains stationary (again, good evidence for negligible Hoogsteen participation under these conditions). The amino protons Ha/a then broaden into the baseline at ~ 0 C and remain in an interm ediate exchange regime through 55 C; at this temperature a broad sing let is observed at ~ 7.1 (Figure 3-8, topmost spectrum). For 8, Ha/a appear as a broad peak at ~ 6.7 at 35 C (similar to 1a) and sharpens to a si nglet at 55 C (~ 6.3; not shown). Finally, in all cases, Hb moves upfield upon increa sing the temperature (for 8 at 33 mM, from ~ 10 to 9.4). Hc H b Ha / a

PAGE 63

63 Figure 3-9. 1H VT-NMR of compound 8 is similar to that of the other structures with the exception for coalescence of Ha and Ha (bottom). Coalescence begins to appear at 55 C (top). The top spectra shows a spectrum taken at 55 C, it is unprocessesed for visualization of coalescence of Ha/a (CDCl3, ~ 33 mM, 300 MHz). Solution-phase dimerization of the newly synthesized UDAP derivatives 79 were investigated by sequential dilution (a technique di scussed in Chapter 2). Dimerization constants, Kdim, were calculated based on experimentally determined D and M for Hb (table 3-1); the former was taken under conditions of complete dimerization (at high concentration (~ 3 mM) and low temperature ( 55 C) in (CD2Cl2)2) and the latter under condi tions of dissociation (at lower concentration (~ 1.5 mM) a nd high temperature (120 C) in (CD2Cl2)2). We and others82 have found (CD2Cl2)2 to be a useful solvent analogue of CDCl3 in these limiting chemical shift determination experiments. The values determined for Kdim (and G) are similar within the estimated error in spite of the significant monomer structural modifica tions, with the exception of 9 that is noticeably weakened. At first glance, it appears that any introduced stabilizing and destabilizing interactions in the alkyl urea series (versus the aryl series) have been offset; signs of compensatory effects often observed in coope rative and reversible as sembly. Rationalizing the Hc H b Ha Ha

PAGE 64

64 similar results and relative stabilities of 1ab, 7, 8, and 9 is the emphasis of the remainder of this Chapter. Table 3-1. Dimerization constants and free energy of formation data for 1a, 1b, and 79 UDAP 1a 1b 7 8 9 Kdim (M1)a 820 160 980 290 1100 250 820 20 530 40 G (kcal/mol) 4.0 0.8 4.1 1.2 4.1 0.9 4.0 0.1 3.7 0.1 dilution range (% dimer) 20 23 22 63 27 D (Hb)b 10.23 10.23 9.89 9.97 9.91 M (Hb)c 7.22 7.22 7.04 7.23 6.98 (Hc)d 12.04 11.82 9.58 9.69 9.38 a error calculated as a average deviat ion for four independently calculated Kdim values N xx xk k ; b Fully dimerized chemical shifts of Hb protons, determined by cooling samples (~ 3 mM) to 55 C in (CD2Cl2)2/300 MHz. c Chemical shifts of fully dissociated Hb determined by heating diluted samples (~ 1.5 mM) to 120 C in (CD2Cl2)2/300 MHz. d Chemical shift of Hc determined at 25 C in CDCl3 at 500 MHz (~ 4 mM for 1a, 1b, 7, 9, and ~ 40 mM for 8) Interactions between N9 and the Urea Any enthalpically favorable interaction between the N9 substituent and the urea substituent might serve to stabilize the conformation require d for dimerization, redu ce the entropic penalty associated with assembly (preorganization), and result in a higher value of Kdim. During the design of the UDAP systems, interactions betw een the urea (e.g., phenyl) substituent and N9 (e.g., benzyl) groups were considered given that a potentially favorable aromatic interaction was identified for model compound 2 in the solid state. The solution-phase contribution of such an interaction in chloroform is of course much sm aller. Hunter and coworkers have used chemical double mutant cycles and model systems to quan tify these contacts between substituted phenyl groups.98-100 Interaction energies ( G) in chloroform can be as high as 1.4 0.5 kJ mol1 at 25 C ( 0.33 kcal mol1) for a benzoyl-diisopropyl aniline interacting with a tert-butyl aniline, but

PAGE 65

65 typically range from destabilizing (0.29 kcal mol1) to stabilizing ( 1.1 kcal mol1) depending on appropriate choice of aromatic ring substituents. Comparison of 1a and 1b provides the starting poin t for the discussion where Kdim = 150 M-1 (in favor of 1b), a value that admittedly falls within the error bars. The ortho methyl (1a) substituents of the N9 protecting groups donate electron density inductively to the aromatic protecting group; similarly the alkoxy (1b) substituents donate by resonance.98 Both substituents are appropriate for edge-to-face ar omatic interactions with the el ectron-deficient edge of the urea phenyl group. The conformations of 1a and 1b were explored by computational means using Monte Carlo methods (MacroModel, Amber* force field, CHCl3 solvation model) to see if edgeto-face interactions are geometrically favored, or even accessible. For these calculations the urea conformation was locked in the appropriate conformation for dimer formation. The lowest energy structures are represented in fi gure 3-10. The N9 substituent of compound 1a appears to be restricted to an anti conf ormation. Edge-to-face interactions are likely prevented by the inability of the substituent to rotate without disrupting intramolecular H-bonding by ortho methyl substituents. The lack of ortho substituen ts and presence of meta substituents for 1b (versus 1a) does offer opportunity for some additional stabilization (long alkyl sitting under urea phenyl), but again this is difficult to quantify. However, comparison by 1H NMR indicates that 1a populates the anti conformati on more frequently than 1b. This is seen in the shielding of the C8 proton. The C8-H proton for 1a appears at 7.58 ppm and appears at 7.65 ppm for 1b in CDCl3 ~ 4 mM at 25 C. This is not a significan t difference, but does indicate the increased flexibility of a substituent with me ta rather than ortho substitution. The incorporation of a hexyl urea substituent in 79 deletes the aforementioned aryl interactions. That the Kdim of 7 is essentially the same as 1a and 1b begins to suggest that the

PAGE 66

66 N9/urea interaction plays a secondary role (and that the large neighboring groups in 1a and 1b are not compromising the dimerization). This is further confirmed by the fact that the Kdim for 9 (530 40 M1), that bears two alkyl substituents, is equivalent to that of 1a (Kdim = 820 160 M1). Figure 3-10. Computational studies to probe possibl e edge-to-face aromatic interactions indicate that the N9 substituent for 1a occupies an anti conformation. The energy minimized structure of compound 1b shows that ortho substi tution does offer some stabilization. A comparison of 7 and 8 reveals shielding of the C8-H proton of 7, consistent with the preferred anti conformation of the mesitylene su bstituent (figure 3-11). Also identified is the similar chemical shift for Hb at similar concentrations of 7 and 8; this of course suggests that the stability of the dimers of the two species is near ly similar, a fact borne out in their calculated Kdim (Table 3-1). Finally, the values of Hc, which are also similar, show that the N9 substituent does not significantly perturb the in tramolecular hydrogen bond. Figure 3-11. Proton NMR of 7 (top) and 8 (bottom) show similar chemical shifts for Hc ( 9.59, 9.60) and Hb ( 9.07, 9.29) at concentrations of 4.53 and 4.55 mM, respectively. The imidazole proton, C8, is deshielded in 8 ( 7.54) in comparison to 7 ( 7.11), presumably due to the aromatic N9 substituent in the latter. (8) (7)

PAGE 67

67 Electronics Effects An alkyl urea substituent was introduced to expl ore electronic in addition to steric effects; consequences on the acidity of Hc (and intramolecular hydrogen bonding) and the basicity of the urea carbonyl oxygen (intermolecular hydr ogen bonding). Indeed the acidity of Hc was decreased, as evidenced by an upfield chemical shift by 1H NMR (figure 3-12; 1a; 12.01, 7; 9.60, table 3-1). From the values give n in table 3-1, the desired effect on Kdim was not realized. This is consistent with the chemical shift of Hb remaining at similar chemical shifts at similar concentrations (~ 4 mM) for 1a and 7 (1a; 9.32, 7; 9.17; figure 3-13). In addition, the 13C chemical shifts of the urea car bonyl groups are also similar for 1a and 7; 155.2, 155.3, respectively (experimental section). The consistency in the Kdim values is probably partially explained in terms of cooperative and offsetti ng electronic effects with in the urea group and pyrimidine ring. Also, a potentially increased carbonyl basicity for 7 may be met with increased repulsive secondary interactions in the model of Jorgensen (Chapter 1). Figure 3-12. Proton NMR illustrates decreased acidity of Hc when the urea substituent is changed from aryl to alkyl, but the chemical shift of Hb remains approximately the same (CHCl3, 25 C, 7 = 4.0 mM; 1a = 4.4 mM). Compound 9, which was synthesized with a tert-butyl acetate group at N9, saw the most noticeable change in dimerization c onstant (nearly a 50 % reduction in Kdim). In considering the ester substituent for N9, the possibilit y of a bifurcated hydrogen bond to Hc of the urea from the ester carbonyl and N3 appeared enthalpically a ppealing (illustrated in figure 3-13). Similar (7)

PAGE 68

68 motifs from the literature101 are shown that are, in general, stabilizing (indicated be the downfield shift of the NH proton, up to ~ 0.5 pp m). If such a hydrogen bond was present in 9, by 1H NMR one may expect Hc of 9 to be more deshielded than Hc of 7 at similar concentration of dimer. This, however, is not the case as shown in figure 3-14 (Hc = 9.37; 8, Hc = 9.39; 7). Figure 3-13. Compound 9 arranged for bifurcated hydrogen bonding of Hc to N3 and the ester carbonyl and changes in chemical shift associ ated with the addition of a bifurcated hydrogen bond (CDCl3).101 Figure 3-14. Proton NMR of 9 (bottom, 3.06 mM, 57 % dimer) and 7 (top, 1.68 mM, 54 % dimer) shows no additional deshielding of Hc (CDCl3, 25 C, 500 MHz). Molecular modeling again suggests reasons why this is the case. Simply put, unfavorable steric interactions between the bulky t-butyl group and urea substituent preclude formation of the potentially stabilizing hydr ogen bond. The reduced relative Kdim then additionally suggests that steric interactions involving th e urea destabilize the conforma tion required for dimerization. Possible remedies include synthesi s of a smaller methyl or ethyl ester, or perhaps even better, conversion of the group to an amide that would introduce a more basic ca rbonyl to the region. (7) (9)

PAGE 69

69 Solubility For the ureidopurines to be easily processed, th ey should be readily so luble. Ureidopurines 1a, 1b, and 7 were near saturation at 4.7.9 mM in deut erated chloroform at room temperature. That value equates to 0.5 g of 1a requiring almost 250 mL of chloroform for solvation. Compounds 8 and 9 were synthesized with a remedy in mind. It was envisioned that the ester at N9 on compound 9 would serve as a point for future functionalization and increased solubility. Solubility was expected to increase with a nonaromatic protecting group due to a decreased possibility of aggregation through surfaces. Interestingly, the so lubility was not significantly increased, with a saturated solution being obtained at ~ 4.5 mM in CDCl3. It was apparently the alkyl chains for 8 which offered superior solubility re lative to the other UDAP derivatives, affording a 38 mM solution in CDCl3 (saturated solution at room te mperature). Therefore, future synthetic designs might include an ester group with a longer methylene linker extending from N9. Summary Structure-property relationships were exam ined for five UDAP whose structural modifications were motivated by enhancing organic solubility, Kdim, and an understanding of the properties that control these phenomena. The urea and N9 substituents were varied to probe interactions between them to affect a change in Kdim and G. Dimer stabilities were similar (within the error limits) with the exception of 7c, and appeared to be governed by cooperative interactions. Weakening of the dimer stability of 7c is the apparent result of steric interactions of the urea and bulky ester t-butyl group. The UDAP design is capable of accommodating bulky aromatic substituents on the urea and N9 positions without compromising the Kdim, an important observation when considering connection of the subunits to covalent scaffolds.

PAGE 70

70 When comparing the UDAP systems to those of Meijer and coworkers, it appears that the differences in Kdim largely derive from the differing elec tronic properties of the heterocyclic scaffolds. These subtleties will be explored in future UDAP designs. Experimental Methods N N N N Cl N H N H O 6-Chloro-N-9-(2,4,6-trimethylbenzyl)-2 -N-(4-hexylamino)ureidopurine (6a). Compound 4a 6-chloro-9-(2,4,6-trimethylbenzyl)-2-a minopurine (0.225 g, 0.746 mmol) was placed in an oven dried two-necked r ound bottom flask and dried under vacuum. Tetrahydrofuran (20 mL) was added and the mixture cooled to 78 C under argon. n-BuLi (0.358 mL, 2.5 M in hexanes, 0.895 mmol) was added dropwise and the mixture stirred for 10 min. Hexyl isocyanate (0.095 mL, 0.895 mmol) was added dropwis e and the mixture stirred a 78 C for 1.75 h. The reaction mixture was stirre d an additional 10 min without cooling followed by quenching with saturated aqueous ammonium chloride. The product mixture was extracted with ethyl acetate and the solvent evaporated Purification by column chromatography (5% CH3OH/CH2Cl2) afforded 6a (0.230 g, 72 %). 1H NMR (CDCl3) 0.86 (m, 3H) 1.30 (m, 4H), 1.63 (m, 2H), 2.24 (s, 6H), 2.30 (s, 3H), 3.40 (q, J = 3.2 Hz, 2H), 5.21 (s, 2H), 6.95 (s, 2H), 7.42 (s, 1H), 7.46 (s, 1H), 8.74 (t, J = 2.9 Hz, 1H). 1H NMR (DMSO-d6) (50 C ) 0.083 (m, 3H), 1.28 (m, 8H), 2.24 (s, 9H), 2.49 (m, 2H), 3.24 (q, J = 3.1 Hz, 2H), 5.33 (s, 2H), 6.93 (s, 2H), 7.90 (s, 1H), 8.41 (t, J = 2.8 Hz, 1H), 9.68 (s,1H). 13C NMR (DMSO-d6) (50 C ) 13.5, 19.2, 20.3, 21.8, 25.9, 29.3, 30.8, 41.8, 125.8, 127.6, 129.1, 137.3, 137.5, 144.4, 149.3, 152.0, 153.1. HRMS (ESI-FT-ICR) calculated for C22H29ClN6NaO (M + Na)+ 451.1984 found 451.2014;

PAGE 71

71 calculated for C44H60Cl2N12O2 (2M + H)+ 857.4256 found 857.4220; calculated for C44H58Cl2N12NaO2 (2M + Na)+ 879.4075 found 879.4230. N N N N NH2 N H N H O 6-Amino-N-9-(2,4,6-trimethylbenzyl)-2 -N-(4-hexylamino)ureidopurine (7). Compound 6a (0.129 g, 0.301 mmol) was placed in a 75 mL pressure tube. Methanolic ammonia (40 mL) was added and the mixture heated to 75 C. The mixture was stirred for 2 h. Solvent was evaporated and the product purifie d by column chromatography (3% CH3OH/CH2Cl2) to yield 7a (0.054 g, 46 %). Mp 239 C; 1H NMR (CDCl3) 0.83 (m, 5H), 1.26 (m, 4H), 1.59 (m, 2H), 2.54 (s, 6H), 2.60 (s, 3H), 3.39 (m, 2H), 4.07 (s, 2H) 5.17 (s, 2H), 6.94 (s, 2H), 7.00 (s, 1H), 9.16 (s, 1H), 9.60 (t, J = 2.4 Hz, 1H). 1H NMR (DMSO-d6) 0.85 (m, 3H), 1.26 (m, 6H), 1.55 (q, J = 3.4 Hz, 2H) 2.25 (s, 9H), 3.23 (q, J = 3.5 Hz, 2H), 5.19 (s, 2H), 6.92 (s, 2H), 6.96 (s, 2H), 7.33 (s, 1H), 7.95 (s, 1H), 9.07 (t, J = 2.2 Hz, 1H). 13C NMR (DMSO-d6) (100 C) 13.0, 18.7, 19.1, 19.9, 25.6, 28.3, 30.4, 31.3, 40.5, 115.5, 128.0, 128.6, 136.8, 136.9, 151.0, 153.5, 155.3, 168.2. HRMS (ESI-FT-ICR) calculated for C22H32N7O (M + H)+ 410.2663 found 410.2703. N N N N O CH3 NH H N O C6H13 N-9-(2,4,6-Trimethylbenzyl)6-methoxy -2 -N-(4-hexylamino) ureidopurine (13). Product was isolated from the sa me reaction mixture as compound 7. The product was isolated as a light yellow solid (0.044 g, 34%). Mp 239 C 1H NMR (CDCl3) 0.83 (m, 3H), 1.28 (m,

PAGE 72

72 6H), 1.61 (m, 2H), 3.39 (q, J = 3.4 Hz, 2H), 4.09 (s, 3H) 5.18 (s 2H), 6.93 (s, 2H), 7.34 (s, 1H), 7.42 (s, 1H), 8.92 (t, J = 2.1 Hz, 1H). 13C NMR (CDCl3) 13.9, 19.6, 20.9, 22.5, 26.7, 29.6, 29.8, 31.4, 42.1, 54.2, 126.9,129.5, 129.6, 129.7, 137.5, 139.8, 152.2, 153.3, 154.3, 161.1. HRMS (ESI-FT-ICR) calculated for C23H33N6O2 (M + H)+ 425.2660 found 425.2647; calculated for C46H64N12NaO4 (2M + Na)+ 871.5066 found 871.5111. N N N N Cl NH2 C 6 H1 3 6-Chloro-N-9-heptyl-2-aminopurine (4c). Compound 3, 6-chloro-2-aminopurine (0.501 g, 2.67 mmol) was placed in a 250 mL oven dried round bottomed flask and suspended in dimethylformamide (100 mL). Tetrabutyl ammonium fluoride (5.60 mL, 5.60 mmol) and iodoheptane (0.813 mL, 3.59 mmol) were added dropwise. The mixture was stirred at room temperature for 20 minutes. The product was extr acted with ethyl acetate and the solvent was evaporated. The residue was recrystallized from a 8:2 ratio of ethyl acetat e:hexanes to yield the product (2.30 g, 75%). 1H NMR (CDCl3) 0.35 (m, 3H), 0.74 (m, 8H), 1.32 (m, 2H), 3.55 (t, J = 3.5 Hz, 2H), 5.24 (s, 2H), 7.68 (s, 1H). 1H NMR (DMSO-d6) 0.84 (s, 3H), 1.24 (m, 8H), 1.78 (m, 2H), 4.05 (t, J = 3.4 Hz, 2H), 6.89 (s, 2H), 8.14 (s, 1H). 13C NMR (DMSO-d6) 13.9, 22.0, 26.0, 28.2, 28.9, 31.1, 43.0, 123.4, 143.2, 149.3, 154.1, 159.8. HRMS (ESI-FT-ICR) calculated for C12H18ClN5 (M + H)+ 268.1323 found 268.1348. N N N N NH2 NH2 C 6 H1 3 2,6-Diamino-N-9-heptylpurine (5c). Vacuum dried 4c (0.207 g, 0.773 mmol) was placed in a pressure tube. Methanolic ammonia was added and the mixture heated to 65C for 20 h.

PAGE 73

73 Solvent was removed under vacuum and the product purified by column chromatography (5% CH3OH/CH2Cl2) to yield 5c as a white solid (0.102 g, 50 %). Mp 227 C; 1H NMR (DMSO-d6) 0.83 (m, J = 3.5 Hz, 3H), 1.22 (m, 8H), 1.71 (q, J = 3.4 Hz, 2H), 3.91 (t, J = 3.4 Hz, 2H), 5.74 (bs, 2H), 6.60 (bs, 2H), 7.68 (s, 1H). 13C NMR (DMSO-d6) 13.9, 22.0, 26.0, 28.2, 29.3, 31.1, 42.3, 113.2, 137.5, 151.7, 156.0, 160.1. HRMS (ESI-FT-ICR) calculated for C12H20N6 (M + H)+ 249.1822 found 249.1828. N N N N Cl N H C6H13 N H O 6-Chloro-N-9-heptyl-2-N-(4-hexylamino)ureidopurine (6c). 6-Chloro-9-heptyl2aminopurine (0.519 g, 1.81 mmol) was placed in an oven dried two-necked round bottom flask and dried under vacuum. Tetrahydrofuran (15 mL) was added and the mixture cooled to 78 C under argon. n-BuLi (1.25 mL, 1.6 M in hexanes, 1.99 mmol) was added dropwise and the mixture stirred for 30 minutes. Hexyl isocyana te (0.211 mL, 1.99 mmol) was added dropwise and the mixture stirred at 78 C for 20 minutes. The reaction mixture was stirred an additional 10 minutes without cooling follo wed by quenching with saturated aqueous ammonium chloride. The product mixture was extracted with ethyl acet ate and the solvent evap orated. Purification by column chromatography (5% CH3OH/CH2Cl2) afforded 6c (0.610 g, 85 %). Mp 146 C; 1H NMR (CDCl3) 0.86 (m, 5H) 1.30 (m, 4H), 1.63 (m, 2H ), 2.24 (s, 6H), 2.30 (s, 3H), 3.40 (q, J = 3.38 Hz 2H), 5.21 (s, 2H), 6.95 (s, 2H), 7.42 (s, 1H), 7.46 (s, 1H), 8.74 (t, J = 2.5 Hz, 1H). 13C NMR (DMSO-d6) 13.9, 14.0, 21.9, 22.1, 25.7, 26.0, 26.1, 28.2, 29.1, 29.5, 31.0, 31.2, 41.7, 114.6, 140.1, 149.8, 153.1, 155.1, 158.5. HRMS (ESI-FT-ICR) calculated for C19H32ClN6O (M + H)+ 395.2321 found 395.2307.

PAGE 74

74 N N N N NH2 N H C6H13 N H O 6-Amino-N-9-heptyl-2-N-(4-h exylamino)ureidopurine (8). Compound 6c (0.604 g, 1.53 mmol) was placed in a 75 mL pressure tube. Methanolic ammonia (40 mL) was added and the mixture heated to 70 C. The mixture was stirred for 3 h. Solvent was evaporated and the product purified by column chromatography (3% CH3OH/CH2Cl2) to yield 7c (0.194 g, 34 %). Mp 229 231 C; 1H NMR (CDCl3) 0.87 (m, 6H), 1.27 (m, 14H), 1.42 (q, J = 3.6 Hz, 2H), 1.85 (q, J = 3.4 Hz, 2H), 3.38 (q, J = 3.6 Hz 2H), 4.03 (t, J = 3.6 Hz, 2H), 7.57 (s, 1H), 9.68 (s, 1H), 9.69 (t, J = 2.4,1H). 1H NMR (DMSO-d6) 0.83 (m, 6H), 1.27 (m, 14H), 1.52 (q, J = 3.5 Hz, 2H), 1.75 (m, 2H), 3.71 (q, J = 3.6 Hz, 2H) 4.12 (t, J = 3.5 Hz, 2H), 6.87 (s, 2H), 7.94 (s, 1H), 8.64 (s, 1H), 9.36 (t, J = 2.5 Hz, 1H). 13C NMR (DMSO-d6) 13.9, 21.9, 22.1, 25.9, 26.0, 26.1, 28.1, 29.1, 29.5, 31.0, 31.1, 42.5, 113.1, 139.9, 150.2, 153.7, 158.0, 158.7. HRMS (ESI-FT-ICR) calculated for C19H33N7O (M + H)+ 376.2819 found 376.2829 C19H33N7ONa (M + Na)+ 398.2639 found 398.2648. N N N N Cl N H C6H13 N H O 6-Chloro-N-9-heptyl-2-N-(4-phenylamino)ureidopurine (15). Tetrahydrofuran (10 mL) was added to an oven dried 50 mL two-necked round bottomed flask followed by addition of 2,2,6,6-tetramethyl piperidine (0.095 mL, 0.559 mmol). The mixture was cooled to 0 C, and nBuLi (0.467 mL, 0.560 mmol) was added dropwi se. The cooled mixture was stirred for 30 minutes followed by cooling to C. Vacuum dried 2-amino-6-chloro-7-heptylpurine (0.125 g, 0.467 mmol) was dissolved in 20 mL THF a nd added dropwise. The reaction mixture was

PAGE 75

75 stirred for 1 h. Phenylisocyanate (0.102 mL, 0. 938 mmol) was added and stirred for 2 hours at 78 C followed by stirring without cooling for 10 min. The reaction was quenched with aqueous saturated ammonium chloride. The mixture was ex tracted with chloroform and the organic layer washed with brine. Solvent was removed under reduced pressure. The product was purified by column chromatography (2% CH3OH/ CH2Cl2) to yield the final produc t as a white solid (0.152 g, 84 %). Mp 174 C; 1H NMR (CDCl3) 0.84 (m, 3H), 1.28 (m, 8H), 1.89 (m, 2H), 4.18 (t, J = 3.5 Hz, 2H), 7.09 (t, J = 3.8 Hz, 1H), 7.33 (t, J = 4.9 Hz, 2H) 7.58 (d, J = 3.7 Hz, 2H), 7.78 (s, 1H), 7.94 (s, 1H), 11.05 (s, 1H). 13C NMR (CDCl3) 13.9, 22.4, 26.6, 28.6, 29.6, 31.5, 44.5, 119.7, 123.8, 127.4, 129.0, 138.0, 144.4, 150.9, 151.0, 152.5, 152.6. HRMS (ESI-FT-ICR) calculated for C19H24ClN6O (M + H)+ 387.1700 found 387.1691. N N N N O CH3 N H C6H13 N H O N-9-Heptyl-6-methoxy-2-N-(4-phenylamino)ureidopurine (16). Compound (15) (0.152 g, 0.414 mmol) was placed in an oven dried pres sure tube to which me thanolic ammonia (50 mL) was added. The mixture heated to 95 C. The reaction was stirred for 3 h. Solvent was removed under reduced pressure. The product wa s purified by column chromatography (2 % CH3OH/ CH2Cl2) to yield the final product as a white solid (0.071 g, 49 %). 1H NMR (CDCl3) 0.86 (m, 3H), 1.31 (m, 8H), 1.87 (q, J = 3.3 Hz 2H), 4.19 (t, J = 3.0 Hz, 5H), 7.12 (t, J = 3.7 Hz, 1H), 7.36 (t, J = 3.9 Hz, 2H) 7.42 (s, 1H), 7.58 (d, J = 3.8 Hz, 1H), 7.78 (s, 1H), 11.25 (s, 1H). 13C NMR (CDCl3) 13.9, 21.9, 22.1, 25.9, 26.0, 26.1, 28.1, 29.1, 29.9, 30.0, 31.0, 31.1, 42.5, 113.1, 139.9, 150.2, 152.7, 153.7, 158.0, 158.7. HRMS (ESI-FT-ICR) calculated for C20H27N6O2 (M + H)+ 383.2195 found 383.2229.

PAGE 76

76 N N N N Cl N H N H O C6H13 6-Chloro-N-7-heptyl-2-N-(4-phenylamino)ureidopurine (15c ). Tetrahydrofuran (10 mL) was added to an oven dried 50 mL twonecked, round bottomed flask followed by addition of 2,2,6,6-tetramethyl piperidine (0.057 mL, 0.33 6 mmol). The mixture was cooled to 0 C, and n-BuLi (0.210 mL, 0.336 mmol) wa s added dropwise. The cooled mixture was stirred for 30 minutes followed by cooling to 78 C. Vacuum dried 2-amino-6-chloro-7-heptylpurine (0.075 g, 0.280 mmol) was dissolved in 12 mL THF a nd added dropwise. The reaction mixture was stirred for 1 h. Phenylisocyanate (0.061 mL, 0. 560 mmol) was added and stirred for 2 h at 78 C followed by stirring without cooling for 10 min. The reaction was quenched with aqueous saturated ammonium chloride. The mixture was ex tracted with chloroform and the organic layer washed with brine. Solvent was removed under reduced pressure. The product was purified by column chromatography (1% CH3OH/ CH2Cl2) to yield the final produc t as a white solid (0.068 g, 63 %). Mp 151 C; 1H NMR (CDCl3) 0.85 (m, 3H), 1.20 (m, 8H), 1.76 (m, 2H), 4.18 (t, J = 3.4 Hz, 2H), 6.99 (t, J = 3.4 Hz, 1H), 7.31 (t, J = 3.7 Hz, 2H), 7.48 (s, 1H), 7.58 (d, J = 3.4 Hz, 2H), 7.80 (s, 1H) 11.59 (s, 1H).13C NMR (CDCl3) 13.9, 22.5, 26.3, 28.6, 31.0, 31.5, 47.8, 54.4, 119.8, 120.3, 123.5, 128.8, 129.0, 138.3, 152.0, 153.0, 157.8. HRMS (ESI-FT-ICR) calculated for C20H27N6O2 (M + H)+ 383.2195 found 383.2229; C20H26N6O2Na (M + Na)+ 405.2015 found 405.2030. N N N N OCH3 N H N H O C6H13

PAGE 77

77 7-Heptyl-6-methoxy-2-N-(4-phe nylamino)ureidopurine (16c ). Starting material, compound 15c (0.068 g, 0.176 mmol), was placed in an oven-dried pressure tube. Methanolic ammonia (90 mL) was added and the vessel sealed The mixture was heated to 90 C for 18 h. Solvent was removed under vacuum and the product purified by column chromatography (1% CH3OH/ CH2Cl2). The product was isolated as a wh ite solid (0.027 g, 42%). Mp 179 C; 1H NMR (CDCl3) 0.86 (m, 3H), 1.26 (m, 8H), 1.86 (m, 2H), 4.11 (s, 3H), 4.23 (t, J = 3.4 Hz, 2H), 7.06 (t, J = 3.4 Hz, 1H), 7.312 (t, J = 3.4 Hz, 2H), 7.46 (s, 1H), 7.66 (d, J = 3.7 Hz, 2H), 7.86 (s, 1H) 11.52 (s, 1H).13C NMR (CDCl3) 14.0, 22.5, 26.7, 28.7, 29.7, 29.9, 31.6, 44.4, 54.6, 119.7, 123.7, 129.1, 137.1, 138.2, 151.5, 152.8, 156.4. HRMS (ESI-FT-ICR) calculated for C20H27N6O2 (M + H)+ 383.2195 found 383.2229. N N N N Cl NH2 O O 2-Amino-6-chloro-9-( tertbutyl acetate)purine (4d). Dry 6-chloro-2-aminopurine (0.501 g, 2.67 mmol) was placed in a 250 mL oven dried round bottomed flask and suspended in dimethylformamide (100 mL). Tetrabutyl ammonium fluoride (5.60 mL, 5.60 mmol) and iodoheptane (0.813 mL, 3.59 mmol) were added dropwise. The mixture was stirred at room temperature for 20 minutes. The product was extr acted with ethyl acetate and the solvent was evaporated. The residue was recrystallized from an 8:2 ratio of ethyl acetate :hexanes to yield the product (2.30 g, 75%). 1H NMR (CDCl3) 1.46 (s, 9H), 4.67 (s, 2H), 5.33 (s, 2H), 7.56 (s, 1H). 13C NMR (DMSO-d6) 27.6, 44.6, 82.3, 122.9, 143.5, 149.4, 154.3, 160.0, 166.7. HRMS (ESIFT-ICR) calculated for C12H18ClN5 (M + H)+ 268.1323 found 268.1348.

PAGE 78

78 N N N N NH2 NH2 O O 2,6-Diamino-9-( tertbutyl acetate)purine (5d). 2,6-Diaminopurine ( 0.503 g, 3.35 mmol) was placed in a 100 mL oven dried round bottomed fl ask. The starting material was suspended in degassed dimethylformamide (20 mL). Sodium hydride (0.142 g, 60 % dispersion in mineral oil, 3.69 mmol) was added and the mixt ure stirred for 2.5 h. Bromo tert-butyl acetate (0.089 mL, 0.46 mmol) was added dropwise followed with stirring for 2.5 h. The product mixture was filtered through celite and the celite rinsed with hot ethyl acetate to give a pale orange solution. Solvent was removed under vacuum methylene ch loride was added to the solid and the solid filtered to yield 5d as an off-white solid (0.447 g, 50 %). 1H NMR (DMSO-d6) 1.42 (s, 9H), 4.74 (s, 2H), 5.82 (s, 2H), 6.71 (s, 2H), 7.67 (s, 1H). 13C NMR (DMSO-d6) 27.7, 44.0, 81.9, 112.7, 138.0, 152.0, 156.2, 160.4, 167.3. HRMS (ESI-FT-ICR) calculated for C12H20N6 (M + H)+ 249.1822 found 249.1828. N N N N Cl N H O O N H O 9-( tertButyl acetate)-6-chloro-2-N-(4-hexylamino)ureidopurine (6d). Vacuum dried 6chloro-9-heptyl-2aminopurine (4d) (0.224 g, 0.789 mmol) was placed in an oven dried round bottom flask and dried under vacuum. Tetrahyd rofuran (10 mL) was added to a two necked round bottomed flask. n-BuLi (0.741 mL, 1.19 mmol, 1.6 M in hexanes) was added and the

PAGE 79

79 mixture cooled to 78 C under argon. Tetramethylpipe ridine (0.201 mL, 1.19 mmol) was added and the mixture stirred for 30 minutes. Hexyl isocyanate (0.100 mL, 0.947 mmol) was added dropwise and the mixture stirred at 78 C for 20 minutes. The reaction mixture was stirred an additional 10 min without cooling followed by quenching with saturated aqueous ammonium chloride. The product mixture was ex tracted with ethyl a cetate and the solvent evaporated. Purification by column chromatography (1% CH3OH/CH2Cl2) afforded 6d (0.610 g, 85 %). Mp 122 C; 1H NMR (CDCl3) 0.87 (m, 3H) 1.31 (m, 8H), 1.45 (s, 9H) 1.61 (m, 2H), 3.37 (q, J = 2.4 Hz, 2H), 4.77 (s, 2H), 7.55 (s, 1H), 7.97 (s, 1H), 8.68 (t, J = 3.4 Hz, 1H). 13C NMR (CDCl3) 14.0, 22.6, 26.7, 27.9, 29.6, 31.5, 40.1, 45.2, 77.0, 84.2, 126.6, 144.5, 150.8, 152.9, 153.3, 153.5, 165.3. HRMS (ESI-FT-ICR) calculated for C18H27ClN6O3 (M + H)+ 411.1906 found 411.1896. N N N N N3 N H O O N H O 6-Azido-9-( tertbutyl acetate)-2-N-(4-hexylamino)ureidopurine (17). Compound 6d (0.107 g, 0.260 mmol) was placed in an oven dr ied round bottomed flask fitted with a reflux condenser. Dimethylformamide (9 mL) was adde d followed by sodium azide (0.034 g, 0.520 mmol). The mixture was heated 105 C for 4 h. Product was purified by column chromatography (1% CH3OH/CH2Cl2) isolated at a pink oil (0.052 g, 58%). 1H NMR (CDCl3) 0.87 (m, 3H) 1.31 (m, 6H), 1.45 (s, 9H), 1.59 (m, 2H), 3.36 (q, J = 2.4 Hz, 2H), 4.74 (s, 2H), 7.50 (s, 1H), 7.85(s, 1H), 8.64 (t, J =3.4 Hz, 1H). 13C NMR (CDCl3) 14.0, 22.5, 26.7, 27.9, 29.7, 31.5, 40.2,

PAGE 80

80 45.2, 84.0, 119.4, 143.1, 153.1, 153.3, 153.5, 153.8, 165.4. HRMS (ESI-FT-ICR) calculated for C11H16N6O2 (M + H)+ 418.2310 found 418.2355. N N N N NH2 N H O O N H O 6-Amino-9-( tert-butyl acetate)-2-N-(4-he xylamino)ureidopurine (9). Starting material (9) (0.041 g, 0.097 mmol) was dried overnight in a 25 mL oven-dried two-necked round bottomed flask fitted with a reflux condenser. A 0.4 M solution of ammonium formate in methanol (2 mL) was added followed by 10 % pa lladium on carbon (9.0 mg). The mixture was heated to reflux for 1.5 h. The palladium was filtered and rinsed with excess methanol until black. Solvent was removed and the residue dissolved in CH2Cl2 followed by washing with water 3 times. Purification by column chromatography (3% CH3OH/CH2Cl2) yielded compound 7d (0.028 g, 74 %). Mp 239 C. 1H NMR (CDCl3) 0.86 (m, 3H) 1.24 (m, 4H), 1.33 (m, 4H), 1.44 (s, 9H), 3.36 (q, J = 2.4 Hz, 2H), 4.67 (s, 2H), 7.65 (s, 1H), 9.33 (s, 1H), 9.46 (t, J = 3.4 Hz, 1H). 1H NMR (DMSO-d6) (100 C) 0.86 (m, 3H), 1.29 (m, 6H), 1.41, (s, 9H), 1.49 (m, 2H), 3.20 (q, J = 2.5 Hz, 2H), 4.83 (s, 2H), 7.10 (s, 2H), 7.88 (s, 1H), 8.18 (s, 1H), 9.14 (t, J = 3.3 Hz, 1H). 13C NMR (DMSO-d6) (100 C) 13.3, 21.6, 25.9, 27.5, 29.3, 30.7, 44.5, 81.9, 114.3, 139.8, 150.1, 153.9, 155.6, 166.3. HRMS (ESI-FT-ICR) calculated for C18H29N7O3 (M + H)+ 392.2405 found 392.2407. N N N N HN NH2 R N H O

PAGE 81

81 General procedure for deprotonation of DAP by NaH. Starting material was dried overnight under vacuum in an oven dried round bottomed flask and dissolved in THF. The solution was cooled to 0 C and sodium hydride (1.2 eq.) added in portions. The mixture was then heated to 65 C and stirred under Ar fo r 2.5 h followed with cooling again to 0 C. Hexyl isocyanate (1.2 eq.) was added and the mi xture stirred 2 h. Acid (0.2 M HCl) was added dropwise until the evolution of gas ceased. Solven t was evaporated and the residue purified by column chromatography. Characterization for NC6 ureidopurines is given in Chapter 4, experimental methods. General procedure for deprotonation of DAP by n -BuLi. Vacuum dried DAP was placed in an oven dried round bottomed flask. Tetrahydrofuran was added and the solution cooled to 78 C. The mixture was stirred under inert atmosphere as n-BuLi (1.1 eq) was added dropwise. Hexyl isocyanate (1.2 eq.) was added after 30 minu tes and stirred for 30 minutes longer. The ice bath was removed and the mixtur e stirred without cooling for 5 minutes followed with quenching with saturated aqueous ammonium chloride. The product mixture was extracted with ethyl acetate and the solv ent removed. The product residue was then purified by column chromatography. Characterization for NC6 ureidopurines is given in Chapter 4, experimental methods. Computational Details Monte Carlo conformational searching was done on a Dell PC (2.4 GHz) running the Fedora Core using MacroModel v. 9.0 (Schrodinger, LLC)86 and the MCMM method (relevant parameters include: steps = 100, itera tions = 2000, solv ent (GB/SA) = CHCl3, force field = Amber*). Only 1N3 and 1N1 were further refined using ab in itio methods (using Gaussian 03 (revision D.01)87) as implemented through the National Ce nter for Supercomputing Applications, SGI Altix cluster Cobalt (http://www.ncsa.uiuc.e du/ UserInfo/Resources/Hardware/SGIAltix/).

PAGE 82

82 CHAPTER 4 REACTIVITY OF 2,6-DIAMINOPURINE Introduction Diaminopurine is a useful synthon, and serves as a platform for appl ications which span materials to pharmaceuticals (figure 4-1).102-108 These applications use diaminopurine with various modified amino groups at N9, C2, and C6 of the purine core. Interestingly, few are obtained by nucleophilic substituti on with DAP. Rather these molecules are obtained through electrophilic substitution of purines halogenated at C2 and C6,102,103,109 which incorporates additional steps into a synthetic ro ute. Generating a greater understa nding of the reac tivity of the amino groups will assist in advancing purine chemistr y and serves as the thrust for this chapter. Figure 4-1. DAP derivatives investigated in pharm aceutical research as specific kinase inhibitors A second thrust involves the systematic char acterization of diaminopurine derivatives by NMR, particularly 2-D and 15N (e.g., 1H13C gHMBC, and 1H15N gHMBC), for the purposes of elucidating substitution patterns on the core. This level of advanced NMR characterization for diaminopurines is sparse in the literature,106 and to date 15N chemical shift data has only been reported by isotopic labeli ng of purine fragments.110 Described here is the full characterization data for differently functiona lized diaminopurines, including 15N chemical shifts (collected at natural abundance), that may serve as a us eful reference for future derivatives.

PAGE 83

83 Syntheses of aryl and alkyl UDAPs have been described in Chapters two and three, respectively. Phenyl UDAP is formed through re gioselective reaction of phenyl isocyanate at NC2, without the isolation of an NC6 product. This result is consistent with literature precedent (one case) for preferential acylation at NC2.111 Porcher and coworker proposed that 2aminoadenosine was acylated by methoxyacetyl chloride at NC2 in 86 %; no mention was made of the NC6 substituted product (figure 4-2, top). The NC2 acylated product was then subjected to a second acylation step at NC6 with isobutyryl chloride. This work is one of just a couple examples of the apparent differentia l reactivity of the amino groups of 2,6-diaminopurine (NC2 versus NC6) reported in the literature. Figure 4-2. Examples of DAP acylation found in the literature111,112 Synthesis of the alkyl UDAPs required a different approach due to an apparent reactivity difference between aryl and alkyl isocyanates. This ultimately involved deprotonation of N9protected 2-amino-6-chloro purine (4), followed by reaction with hexyl isocyanate to yield compounds 6a, c, and d, and then subsequent amination. Re call that direct deprotonation of DAP yielded the NC6-substituted purine upon trea tment with an alkyl isoc yanate. Interestingly, treatment of DAP with hexyl isocyanate at elevated temperature (in a pressure tube), in pyridine,

PAGE 84

84 was also found to afford the NC6 regioisomer in 97% yield (based on 1H and NOESY NMR analysis). This result, in connection with work from Nielsen and coworkers, which reported the exclusive NC6 substitution of DAP to form a carbamate via reaction with N-benzyloxycarbonyl-N-methylimidazolium triflate (rt, dioxane, 20 h, 88 %),112 Rappaports reagent, speaks to a more complicated story involving NC2 versus NC6 substitution. In other word s, the two best literature examples of DAP electrophilic substitution chemis try provide disparate results, as do examples from our own experiments. Issues related to NC2 versus NC6 substitution quickly expose the additional problem of assigning the (monosubstitution) regiochemistry w ith certainty by NMR. NOE (from 1-D or 2-D techniques) and routine chemical shift data (1H and 13C) are largely inconclusive, the latter since attempts to establish chemical shift trends among the isomers fail due to unpredictable shift changes upon introduction of differe nt electrophiles. In fact, the work of Nielsen and the work of Porcher offers unsatisfying evidence with respect to regiochemical assignments. In addition to changes in solvent, electrophile and temperature, DAP structure is also a mitigating factor. In particular, functionalization at the N9 position that could, in theory, affect reaction at NC2. It has also been demonstrated that the so lubility of the diaminopurines is affected by the N9 substituent (Chapter 3, 8 is ~10 times more soluble than 7), and solubility could also affect apparent rates and selectivities. Finally, la rgely unexplored is the ex tent to which acylation reactions involving DAP might be reversible. Drawing general conclusions about DAP regioselectivity requires a more systematic study. Reactivity of 2,6-Diaminopurine Acylation with Methoxy Acetyl Chloride Acylation of diaminopurine was studied usi ng common acylating agen ts with three N9 protected DAPs (5a, 5c, and 5d), each previously introduced in the synthesis of UDAP. The

PAGE 85

85 three DAPs were intended to probe solvent, st eric, or electronic effects related to the N9 substituents. Experiments were performed in two different solvent systems (3:1 solvent mixture of CH2Cl2 or CH3CN to pyridine) with 2.5 eq. methoxy acetyl chloride (A), a common acylating agent (figure 4-3, table 4-1). The reactions were stirred until the starti ng material disappeared, and the monosubstituted (x or y) and disubstituted products (z) were isolated. By the ratio of the products obtained the reactivity difference betw een the two amino groups under these conditions might be revealed (assuming irreversible bond formation). Figure 4-3. Reaction of DAP w ith 2.5 eq. methoxy acetyl chloride in a 3:1 ratio of CH2Cl2 or CH3CN to pyridine. Products 18 x, y, and z were isolated and characterized. The time for the reactions to reach completion varied from 0.5 h to 1.5 h for CH2Cl2 or CH3CN, respectively (table 4-1). From each of the reaction mixtures two components were isolated in similar ratios. The major product was the monosubstituted product (x or y in Figure 41) and the minor product was the disubstituted product, z (as determined by 1H NMR). Table 4-1. Reaction conditions and outcomes fo r nucleophilic substitution of DAP with methoxy acetyl chloride. Purine R1 time (h) Solvent t (C) Yield x:y:z 5a CH2OCH3 0.5 CH2Cl2/C5H5N Rt 95 9:1:0 5a CH2OCH3 1.5 CH3CN/C5H5NRt 95 9:1:0 5c CH2OCH3 0.5 CH2Cl2/C5H5N Rt 97 9:1:0 5c CH2OCH3 1.5 CH3CN/C5H5NRt 98 4:1:0 5d CH2OCH3 0.5 CH2Cl2/C5H5N Rt 98 9:1:0 5d CH2OCH3 1.5 CH3CN/C5H5NRt 97 4:1:0 Reactions were performed at a co ncentration of 4.0 mM with 2.5 eq. of electrophile. Yields are based on total product masses isolated.

PAGE 86

86 Determination of Substitution Site The protocol for determining the s ite of nucleophilic substitution as NC2 or NC6 of DAP (in all of the studies) involved the comprehensive characterization of the N9 protected 2-amino-6chloropurines (4a, 4c, and 4d), their corresponding diaminopurines (5a, 5c, and 5d), and the monosubstituted reaction products by 1H, 13C, 1H13C gHMBC, and 1H15N gHMBC NMR (in addition to mass spec). The NMR data was obt ained at 25 C, in a common solvent (DMSO-d6), and at natural abundance of 15N (NMR spectra were recorded on a Varian Inova spectrometer equipped with a 5 mm indirect detection pr obe, operating at 500 MHz at 50 MHz for 15N). Concentrated samples (0.1 M) were prepared in DMSO-d6, however, peak broadening was an issue. Although the issue of peak broadening wa s not completely remedied (occurrences were seen in relation to NH2 and NH resonance), chemical shift correlations were still accessible. Chemical shift data for 4c, 5c, and 12c are representative and di scussed here (figure 4-4 and table 4-2); information for other compounds is provided in Appendix A. The chemical shift of the C8-H and -CH2 protons (at N9) were easily assigne d on the basis of chemical shift and integration. These protons were further characterized by their oneand multiple-bond heteronuclear couplings. For H8 H8-C8 coupling was observed by 1H13C gHMQC, H8-C5 and H8-C4 coupling by 1H13C gHMBC, and H8-N7 and H8-N9 coupling by 1H15N gHMBC. These correlations are shown schematically in figure 4-2. Figure 4-4. NMR coupling assignments of the imidazole portion of 2-amino-6-chloropurine by 1H13C and 1H15N gHMBC

PAGE 87

87 Table 4-2. Chemical shifts of the imidazole portion of the bicy clic purine compounds 4c (2amino-6-chloro-9-heptylpurine), 5c (2,6-diamino-9-heptylpurine), and 18c (monosubstituted product 18cx or 18cy) Purine C4a C5a C8a H8a H9a N7b N9b 4c 154.1 123.4 143.2 8.14 4.05 237.8 159.0 5c 151.7 113.2 137.7 7.70 3.94 237.4 157.1 18c 150.3 115.0 140.3 8.03 4.07 237.2 159.8 Chemical shifts at natura l abundance in 0.1 M DMSO-d6 solutions at 25 C. a 1H and 13C values are relative to TMS; b 15N values given on the ammonia scal e relative to nitromethane (379.5 ppm). The chemical shift assignments of the pyrim idine ring for the compounds of this study were made similarly to the imidazole assignments. The C2 amino groups (NC2) of compounds 4c and 5c were assigned by long-range coupling of th e amino protons to two nitrogens in the 1H15N gHMBC spectra, assigned as N1 and N3 (assigned by chemical shifts, 110 experimental shifts given in table 4-3). Likewise, the NC6 amino group of 5c was assigned based on the long-range coupling to only one nitrogen, N1. To fu rther solidify the assignment of NC6 and NC2, coupling was seen for the amino protons NC6H and NC2H to C5 and C2 (respectively) of the purine core in the 1H13C gHMBC spectrum. Assignment for the point of nucleophilic subs titution was made by the simple line of reasoning demonstrated thus far. In all six of th e reactions the point of substitution was the same, and the assignment was made for substitution at NC2. The amidic proton of NC2 was coupled to two nitrogen in the 1H15N gHMBC spectra (scheme shown, fi gure 4-5, and spectrum shown, figure 4-6), N1 and N3. Similarly the NC6 amino protons showed long range coupling to both N1 and to C5 (1H13C gHMBC, not shown, see Appendix A). With the site of monosubstitution confirmed (tab le 4-1), the results of this study indicate that solvent, but more importan tly N9 protecting groups, are not f actors in the reac tivity of DAP under fairly routine acylati on conditions. The results further suggest that NC2 reacts completely and much more quickly than NC6 (the NC6 monosubstituted product was not identified) in

PAGE 88

88 agreement with the results of Porcher and coworkers;111 the NC6 reacts next to provide the disubstituted product under these conditions. Figure 4-5.Correlations proved the point of substitution to be NC2. Long-range coupling is seen from the amidic proton to N1 and N3, and from the amino protons of NC6 to N1 in the 1H15N gHMBC spectrum. Long-range c oupling is also seen from NC6H to C5 in the 1H13C gHMBC spectrum (chemical shifts given tables 4-1 and 4-2) Table 4-3. Chemical shift data for the pyrimidine portion of the purine of compounds 4c, 5c, and 18cx ppurine C2 C4 C5 C6 NC2HNC6HN1 NC2 N3 NC6 4c 149.3 154.1 123.4 159.8 6.89 -232.981.0 196.9-5c 156.2a 151.7 113.2 160.4a6.69 5.81 197.7b75.6 178.4b75.8 18cx 156.1 150.3 116.0 152.3 7.22 9.60 209.8139.5 197.279.8 a Chemical shift value was not measured for compound 5c due to peak broadening. Value given is the chemical shift for the corresponding atom of 5d.b Value given was not measured for 5c, but was obtained for 5a and 5d, which were averaged to give the corresponding value. Substitution at NC6 and Reversibility Introduced earlier, when DAP (~ 40 mM) was r eacted with hexyl isocyanate at 100 C in a sealed reaction vessel in pyridin e (figure 4-7, table 4-4) NC6 monosubstitution was predominantly observed. The increased concen tration of substrate for this reaction was coincidentally similar to the concentration of DAP that provided the NC6 carbamate upon reaction with N-benzyloxycarbonyl-N-methylimidazolium triflate95 in dioxane at room temperature (20 h), reported by Nielsen.112 The literature reaction was reproduced in our

PAGE 89

89 laboratory, and the products from both r eactions were studied thoroughly by the NMR techniques discussed above. A lthough the conditions (e.g. electr ophiles, solvents, etc.) were quite different, characterization proved the products to be NC6 substituted (11cz, 11dz and 12dz). Furthermore, the products were obtained in high yi eld with complete regios electivity (table 4-4). Figure 4-6. Spectrum showing 1H15N gHMBC correlations for 18cx proves the site of substitution to be NC2. Chemical shifts are at natu ral abundance in a 0.1 M DMSO-d6 solution at 25 C. 15N values are given on the ammonia scale relative to nitromethane (379.5 ppm). Figure 4-7. Electrophiles AC were reacted with DAP and the reaction products characterized. Reaction conditions and outcome s are given in table 4-4.

PAGE 90

90 Table 4-4. Reaction outcomes for high concentration (40 mM) reactions of 5c and 5d with electrophiles A, B, and C. purine E+ R1 time (h) Solvent temp (C) yield x:y:z 5c B NHC6H13 3.5 C5H5N 100 97 0:0:10 5d B NHC6H13 3.5 C5H5N 100 97 0:0:10 5d C OBn 20 C4H8O2 97 97a 0:0:10 5d A CH2OCH3 0.3 CH2Cl2/C5H5N 99 99 9:1:0 a Reaction yield based on r ecovered starting material (product yield was 81 %). Before speculating on the or igin of the regioselectivity differences, we took the opportunity to evaluate whether any of these acylations were reversible under the reaction conditions tested. Returning to the methoxy acetyl chloride reactions, a 4 mM solution of DAP 5d in a 3:1 mixture of CH2Cl 2:pyridine was cooled to C a nd treated with the electrophile (2.5 equiv). The starting material completely disappeared within 3 h (determined to be NC2 substituted product, 18dx). Within 4 h TLC showed a second product spot, the disubstituted product (18dy). The reaction was left to stir overnight and showed no change in product ratio or conversion by TLC. More importantly, the reacti on at 30( 35) C yielded the same product ratio (9:1:0, x:y:z) that was isolated from the reaction at room temperature. This result is most consistent with the NC2 regioisomer being the kinetic product. In a second variable temperature experiment, product 18dx was dissolved in a 3:1 mixture of methylene chloride:pyridine and heated to 75 C for 17 h in a sealed reaction ve ssel. Detectable displacement of the acyl group by pyridine (or a purine) would establish a case for an equilibrium. No evidence arose to support an acyl migration from NC2 to NC6 under these conditions, or reaction of the starting material with pyridine; the NC2 product was recovered in 94% yield. A similar reaction was performed with compound 8 (6-amino-2N-hexyl-9-heptylureidopurine). There was also no visible evidence for exchange seen here (8 was recovered in 97% yield).

PAGE 91

91 Reactivity of DAP with Acetic Anhydride Reaction of DAP with acetic anhydride was next pursued to a) explore regioselectivity trends using a generic electrophile and b) shed further light on potential reversibility issues (thermodynamic effects) in a reaction rate regi me that was particularly amenable to NMR analysis. Reaction of a 4 mM solution of 5d in a 3:1 ratio of CH2Cl2:pyridine with acetic anhydride (2.5 eq.) offered a much slower reaction rate (figure 4-8, a discussion follows which addresses the factors involved in the comparatively decreased reaction rate, including significantly different reactive intermediates). After 48 h both monosubstituted products were isolated in a 4.5:1 ratio (NC2(19x):NC6(19z)) in an overall yield of 91 % based on recovered starting material. With the formation of NC2 and NC6 products from the same reaction mixture the study of kinetic and thermodyna mic effects was simplified. Figure 4-8. Reaction of DAP with acetic anhydride yielded a mixture of NC2 and NC6 products. To determine if an equilibrium was present und er the reaction conditions two experiments were performed. Compound 19x was heated to reflux with deuterat ed acetic anhydride (2.5 eq.) in a methylene chloride/pyridine mixture. No deuterium was incorporated into the NC2 position nor was any NC6 monosubstituted product isolated. The experiment showed that under these conditions, the acylated NC2 product was kinetically stab le. Further support for NC2 as the kinetic product was obtained by modifying th e concentration of starting mate rial in the reaction mixture. A 40 mM solution of compound 5d was prepared in a 3:1 mixture of CH2Cl2:pyridine and 2.5

PAGE 92

92 equiv of methoxy acetyl chloride was added. Consis tent with a kinetic effect, the reaction rate was increased and the same 9:1 product mixture of NC2:disubstituted product was isolated (table 4-4). The increased concentration also ruled out possible associative effects in the formation of NC6 substituted products, 11z and 12z. A thermodynamic equilibrium was eliminated conclusively by monitoring the reaction progress by NMR. Diaminopurine, 5d, was treated with acetic a nhydride in a 3:1 mixture of CDCl3:pyridine (6.6 mM). After ~ 48 h the reacti on had only progressed to ~ 30 % consumption of starting material, however, the NC2:NC6 product ratios remained 4:1 throughout the course of the reaction (figure 4-9). Base d on this result, the reaction of DAP with a common acylating agent, acetic anhydride, is a kinetically governed reaction with NC2 substitution being favored. Figure 4-9. Reaction of 5d with acetic anhydride in 3:1 CDCl3:pyridine to yield 15dx and 15dz. A relative product ratio of 4:1 ratio is ma intained throughout the course of reaction indicating a kinetic dist ribution of products. The results thus far point to a ki netically controlled acylation of NC2 under standard conditions that involve pyridine and either an ac id chloride or simple anhydride. Complicating Starting material, 5d NC2, 15dx NC6, 15dz

PAGE 93

93 the overall analysis is what could be different acylation mechanisms related to the identity of the active acylating agent under the c onditions employed. Acylations catalyzed by pyridine are wellestablished that proceed through an acyl pyridinium intermediate.113 It is also possible, for methoxy acetyl chloride, that pyridi ne will facilitate elimin ation of HCl to form a ketene; this has been illustrated in the literatu re (specifically for methoxy acetyl chloride) in the presence of tertiary amines.114,115 Thus, the reaction by which 18x is formed could involve three different electrophilic species: a ketene, a py ridinium cation, or an acid ch loride. Likewise, formation of 19x and 19z can occur through two likely pathways: di splacement of a pyridinium cation, or displacement of acetate. In terestingly, reaction of 5d with acetic anhydride in the absence of pyridine gave a different product composition. The NC6 product was isolated in the absence of the NC2 product after 56 h at room temperature in 23 % yield (53% based on recovered starting material; ~ 5 % of the composition wa s found to be disubstituted product by 1H NMR). The known catalysis by pyridine in acylation reactions is thus perhaps an important factor in the regioselectivity of DAP. Finally, also uncertain is the extent to which deprotonated DAP (generated at high temperature in the presence of pyridine) might l ead to preferential reaction at NC6 in the simple acylation reactions. Summary and Conclusions Nucleophilic substitution r eactions involving DAP have b een studied using various electrophiles, conditions, and DAP protecting groups (at N9). For monosubstitution, the site of substitution has for the first time been una mbiguously assigned by NMR methods including 1H, 13C, 1H13C gHMQC, and 1H15N gHMBC at natural abundance. The characterization has provided a protocol and small database of 15N chemical shift data, through the study of fourteen purine derivatives, that could be used in the structural elucidation of future DAP substitution products.

PAGE 94

94 Reaction of DAP with standa rd acylating reagents, acetic anhydride or methoxy acetyl chloride, in the presence of pyridine, gave exclusively NC2 monosubstitution. The selectivity was insensitive to temperature, concentration, so lvent, or DAP N9 substituent. Thermodynamic effects were ruled out, and the reaction was shown to be kinetical ly governed. Unactivated acetic anhydride, alkyl isocyanates, and Rappoports reagent give primarily NC6 monosubstitution. Substitution at this position is also observed upon formal deprotonation of DAP with n-BuLi. Further studies are required to determine whether the selectivity changes mirror mechanism changes that could involve deprot onation of the purine ( under certain conditions ) or the nature of the electrophile. Also, subtle differences between the reactive centers may need to be analyzed more closely. For example, although nitrogen data does no t suggest different chemical environments for NC2 and NC6, the 13C chemical shift of C6 is slightly more deshielded than C2 ( 160.4 and 156.2 in DMSO-d6, 25 C for 5d). Experimental Methods N N N N HN NH2 C6H13 N H O 6-Amino-N-9-heptyl-6-N-(4-h exylamino)ureidopurine (11). 2,6-Diamino-9-heptylpurine (0.0191 g, 0.0769 mmol) was placed in an oven-dried pressure tube and dissolved in pyridine (2 mL). Hexylisocyanate (0.123 mL, 1.15 mmol) was added and the mixture was heated to 100 C. The reaction was stirred for 3.5 h. The solven t was removed under reduced pressure, and the product was purified by column chromatography (1% CH3OH/ CH2Cl2) to yield 11as a beige solid (0.0281, 97 %). Mp 148 C; 1H NMR (CDCl3) 0.85 (m, 6H), 1.37 (m, 14H), 1.42 (q, J = 3.7 Hz, 2H), 1.85 (q, J = 3.6 Hz, 2H) 3.38 (q, J = 3.0 Hz, 2H), 4.02 (t, J = 3.8 Hz, 2H), 4.84

PAGE 95

95 (s, 2H), 7.68 (s, 1H), 7.88 (s, 1H), 9.16 (t, J = 2.7 Hz, 1H). 1H NMR (DMSO-d6) 0.83 (m, 6H), 1.27 (m, 14H), 1.54 (q, J = 3.4 Hz, 2H), 1.75 (q, J = 3.4 Hz, 2H), 3.23 (t, J = 3.4 Hz, 2H), 3.99 (t, J = 3.6 Hz, 2H), 4.43 (s, 2H), 6.60 (s, 2H), 7.95 (s, 1H), 8.76 (s, 1H), 9.58 (t, J = 2.8 Hz, 1H). 13C NMR (DMSO-d6) 13.9, 21.9, 22.1, 25.9, 26.0, 26.1, 28.1, 29.1, 29.9, 30.0, 31.0, 31.1, 42.5, 113.1, 139.9, 150.2, 152.7, 153.7, 158.0, 158.7. HRMS (ESI-FT-ICR) calculated for C19H4N7O (M + H)+ 376.2819, found 376.2829. N N N N HN NH2 O O N H O 6-Amino-N-9-( tertbutyl acetate)-6-N-(4-hexylamino)ureidopurine (12). 2,6-Diamino-9heptylpurine (0.086 g, 0.325 mmol) was placed in an oven dried pressure tube and dissolved in pyridine (9 mL). Hexylisocyanate (0.517 mL, 4. 88mmol) was added and th e mixture heated to 100 C. The reaction was stirred for 3.5 h. So lvent was removed under reduced pressure. The product was purified by column chromatography (1% CH3OH/ CH2Cl2) to yield the final product as a beige solid (0. 124 g, 97 %). Mp 242 C; 1H NMR (CDCl3) 0.88 (m, 3H), 1.30 (m, 8H), 1.45 (s, 9H), 1.60 (q, J = 3.5 Hz, 2H), 3.37 (t, J = 2.9 Hz, 2H), 4.68 (s, 2H), 4.84 (s, 2H), 7.69 (s, 1H), 7.88 (s, 1H), 9.16 (t, J = 2.7 Hz, 1H). 1H NMR (DMSO-d6) 0.86 (m, 3H), 1.30 (m, H), 1.42 (s, 9H), 1.55 (m, 2H), 3.24 (t, J = 3.0 Hz, 2H), 4.82 (s, 2H), 7.93 (s, 1H), 8.88 (s, 1H), 9.58 (t, J = 2.8 Hz, 1H). 13C NMR (DMSO-d6) 13.9, 21.9, 22.1, 25.9, 26.0, 26.1, 28.1, 29.1, 29.9, 30.0, 31.0, 31.1, 42.5, 113.1, 139.9, 150.2, 152.7, 153.7, 158.0, 158.7. HRMS (ESI-FT-

PAGE 96

96 ICR) calculated for C18H30N7O3 (M + H)+ 392.2404, found 392.2409 calculated for C18H29N7O3 (M + Na)+ 414.2224, found 414.2245. N N N N NH2 N H O O 6-Amino-N-9-(2,4,6-trimethylbenzyl)-2-N-(methoxy-acetamide)purine (18ax). Compound 5a (0.0510 g, 0.181 mmol) was placed in an oven-dried three necked-roundbottomed flask and dried overnight. The starting material was dissolved in a 3:1 mixture of acetonitrile (46 mL) and pyridine (15 mL). Meth oxy acetyl chloride (0.0410 mL, 0.450 mmol) was added. The reaction mixture was stirred until the starting material was consumed (1.5 h). The solvent was evaporated under a flow of n itrogen and the product was purified by column chromatography (2%% CH3OH/CH2Cl2). Purine 18ax was isolated as a white solid (0.056 g, 88%). 1H NMR (CDCl3) 2.25 (s, 6H), 2.30 (s, 3H), 3.51 (s, 3H), 4.16 (s, 2H), 5.24 (s, 2H), 5.27 (s, 2H), 6.93 (s, 2H), 7.17 (s, 1H), 8.86 (s, 1H) 1H NMR (DMSO-d6) (100 C) 2.24 (s, 3H), 2.28 (s, 6H), 3.40 (s, 3H), 4.20 (s, 2H), 5.22 (s, 2H), 6.81 (s, 2H), 6.91 (s, 2H), 7.42 (s, 1H), 9.05 (s, 1H). 13C NMR (DMSO-d6) (100 C) 18.7, 19.9, 40.6, 58.0, 71.7, 115.6, 128.2, 128.6, 136.9, 137.0, 138.5, 150.1, 151.7, 155.7, 167.7. HRMS (ESI-FT-ICR) calculated for C18H22N6O2 (M + H)+ 355.1877, found (M + Na)+ 377.1696. N N N N HN N H O O O O

PAGE 97

97 N-9-(2,4,6-Trimethylbenzyl)-2,6-N-(methoxy-acetamide)purine (18ay). The product was isolated as an off-white solid (0.009 g, 14 %) from the same reaction mixture as compound 18ax. Mp 265 C (decomp); 1H NMR (CDCl3) 2.24 (s, 6H), 2.30 (s, 3H), 3.51 (s, 3H), 3.52 (s, 3H), 4.17 (s, 2H), 4.25 (s, 2H), 5.31 (s, 2H ), 6.93 (s, 2H), 7. 35 (s, 1H), 8.95 (s, 1H), 9.18 (s, 1H). 1H NMR (DMSO-d6) (100 C) 2.24 (s, 3H), 2.28 (s, 6H), 3.40 (s, 3H), 3.41 (s, 3H), 4.28 (s, 2H), 4.35 (s, 2H), 5.31 (s, 2H), 6.92 (s, 2H), 7.76 (s, 1H), 9.76 (s, 1H), 9.94 (s, 1H). 13C NMR (DMSO-d6) (100 C) 18.8, 19.9, 42.9, 58.0, 58.1, 71.8, 72.4, 108.6, 127.7, 128.7, 137.1, 137.1, 139.6, 144.3, 148.4, 149.1, 168.1, 168.1. HRMS (ESI-FT-ICR) calculated for C21H26N6O4 (M + H)+ 427.2088 found (M + Na)+ 449.1908. N N N N NH2 N H C6H13 O O 6-Amino N-9-heptyl-2-N-(methoxy-acetamide)purine (18cx). Compound 5c (0.030 g, 0.121 mmol) was placed in an oven dried th ree necked-round-bottomed flask and dried overnight. The starting material was dissolved in a 3:1 mixture of acetonitrile (31 mL) and pyridine (10 mL). Methoxy acetyl chloride (0 .028 mL, 0.76 mmol) was added. The reaction mixture was stirred until the starting material was consumed (1.5 h). The solvent was evaporated under a flow of nitrogen and the product was purified by column chromatography (2% CH3OH/CH2Cl2). The product was isolated as a wh ite solid (0.032 g, 83%). Mp 201 C (decomp); 1H NMR (CDCl3) 0.85 (m, 3H), 1.26 (m, 8H), 1.86 (m, 2H), 3.50 (s, 3H), 4.09 (t, J = 3.4 Hz, 2H), 4.29 (s, 2H), 7.68 (s, 1H). 1H NMR (DMSO-d6) (100 C) 0.85 (t, 3H), 1.26 (m, 8H), 1.83 (m, 2H), 3.39 (s, 3H), 4.07 (t, J = 3.3 Hz, 2H), 4.23 (s, 2H), 6.78 (s, 2H), 7.94 (s, 1H), 9.05 (s, 1H). 13C NMR (DMSO-d6) (100 C) 13.0, 21.2, 25.4, 27.4, 28.6, 30.4, 42.4, 58.0, 71.9,

PAGE 98

98 115.8, 139.8, 150.1, 151.7, 155.6, 167.8. HRMS (ESI-FT-ICR) calculated for C15H24N6O2 (M + H)+ 321.2034 found (M + H)+ 321.2037. N N N N HN N H C6H13 O O O O N-9-Heptyl-2,6-N-(methoxy-acetamide)purine (18cy). The disubstituted product was isolated as an off-white solid (0.006 g, 16 %) from the same reaction mixture as compound 18cx. Mp 204 C; 1H NMR (CDCl3) 0.85 (m, 3H), 1.27 (m, 8H), 1.89 (m, 2H), 3.50 (s, 3H), 3.53 (s, 3H), 4.15 (m, 2H), 4.18 (s, 2H), 4.26 (s, 2H), 7.89 (s, 1H), 8.92 (s, 1H), 9.25 (s, 1H). 1H NMR (DMSO-d6) 0.83 (m, 3H), 1.22 (m, 8H), 1.78 (m, 2H), 3.34 (s, 3H), 3.37 (s, 3H), 4.12 (t, J = 3.5 Hz, 2H), 4.32 (s, 2H), 4.39 (s, 2H), 8.34 (s, 1H), 10.21 (s, 1H), 10.40 (s, 1H). 13C NMR (DMSO-d6) (100 C) 13.0, 21.2, 25.4, 27.4, 28.4, 30.4, 35.9, 42.8, 58.1, 58.2, 71.8, 71.9, 75.0, 118.9, 143.1, 148.2, 151.0, 160.9, 168.1. HRMS (ESI-FT-ICR) calculated for C18H28N6O4 (M + H)+ 393.2245 found (M + Na)+ 415.2064. 6-Amino-N-9-( tertbutyl acetate)-6-N-(methoxy-acetamide)purine (18cd). Compound 5d (0.0415 g, 0.157 mmol) was placed in an ovendried three-neckedro und bottomed flask and dried overnight. The starting material was dissolv ed in a 3:1 mixture of acetonitrile (40 mL) and

PAGE 99

99 pyridine (13 mL). Methoxy acetyl chloride (0.0358 mL, 0.393 mmol) was added. The reaction mixture was stirred until the st arting material was consumed (1.5 h). Solvent was evaporated under a flow of nitrogen and product pur ified by column chromatography (2% CH3OH/ CH2Cl2). The product was isolated as a white solid (0.044 g, 83%). Mp 196 C; 1H NMR (CDCl3) 1.47 (s, 9H), 3.49 (s, 3H), 4.12 (s, 2H), 4.79 (s, 2H), 5.94 (s, 2H), 7.78 (s, 1H), 8.81 (s, 1H). 1H NMR (DMSO-d6) (100 C) 1.43 (s, 9H), 3.38 (2, 3H), 4.22 (s, 2H), 6.85 (s, 2H), 7.93 (s, 1H), 9.08 (s, 1H). 13C NMR (DMSO-d6) (100 C) 27.2, 29.8, 44.2, 58.0, 115.2, 140.3, 150.3, 151.9, 155.6, 166.1, 167.9. HRMS (ESI-FT-ICR) calculated for C14H20N6O4 (M + H)+ 337.1619, found (M + Na)+ 359.1438. N N N N HN N H O O O O O O 9-( tertbutyl acetate)-2,6-N-(methoxy-acetamide)purine (18dy). The disubstituted product was isolated as an off-white solid (0. 007 g, 13 %) from the same reaction mixture as compound 18dx. Mp 196 C; 1H NMR (CDCl3) 1.47 (s, 9H), 3.50 (s, 3H), 3.36 (s, 3H), 4.14 (s, 2H), 4.24 (s, 2H), 4.87 (s, 2H), 7.97 (s, 1H), 8.94 (s, 1H), 9.26 (s, 1H). 1H NMR (DMSO-d6) 1.43 (2, 9H), 3.39 (s, 3H), 3.43 (s, 3H), 4.28 (s, 2H), 4.36 (s, 2H), 4.95 (s, 2H), 8.23 (s, 1H), 9.67 (s, 1H), 9.85 (s, 1H). 13C NMR (DMSO-d6) (100 C) 27.2, 44.4, 58.0, 58.1, 71.7, 71.9, 82.0, 108.6, 118.4, 143.5, 148.3, 151.3, 165.8, 168.2. HRMS (ESI-FT-ICR) calculated for C17H24N6O6 (M + H)+ 409.1830, found 409.1819.

PAGE 100

100 N N N N HN NH2 O O O O 2-Amino-6-N-(benzylcarboxy amino)-N-9-( tert-butyl acetate)purine (12z). 2,6Diamino-9-(tert-butyl acetate)purine (5d) (0.065 g, 0.25 mmol) was pl aced in an oven-dried twonecked round-bottomed flask. Dry degassed 1,4-dioxane (2 mL) was added followed by Nbenzyloxycarbonyl imidazolium triflate95 (0.124 g, 0.356 mmol). The mixture was stirred for 20 h at room temperature. The solvent was rem oved under vacuum and the residue was purified by column chromatography (3% CH3OH/CH2Cl2). The product was isolated as a white powder (0.079 g, 81%). Mp 269 C; 1H NMR (CDCl3) 1.45 (s, 9H), 4.63 (s, 2H), 5.19 (s, 2H), 5.24 (s, 2H), 7.36 (m, 5H), 7.65 (s, 1H), 8.80 (s, 1H). 13C NMR (CDCl3) 28.1, 44.8, 67.7, 83.6, 128.6, 128.7, 128.8, 135.9, 140.8, 150.2, 151.4, 153.4, 160.1, 163.3, 166.4. HRMS (ESI-FTICR) calculated for C17H24N6O6 (M + H)+ 399.1775, found 399.1757. N N N N NH2 N H O O O 6-Amino-2-N-(acetyl)-N-9-( tertbutyl acetate)purine (19x). Compound 5d (0.052 g, 0.196 mmol) was placed in an ove n-dried two-necked round-bottome d flask. Methylene chloride (50 mL) was added followed by pyridine (17 mL). Acetic anhydride (0.046 mL) was added and the mixture was stirred for 43 h at room temp erature. The solvent was removed under vacuum

PAGE 101

101 and the residue purified by column chromatography (3% CH3OH/CH2Cl2) to yield the product as a white solid (0.040 g, 67% (77% ba sed on recovered st arting material)). 1H NMR (DMSO-d6) 1.41 (s, 9H), 2.18 (s, 3H), 4.85 (s, 2H), 7.19 (s, 2H), 7.97 (s, 1H), 9.74 (s, 1H). 13C NMR (DMSO-d6) 27.7, 43.6, 44.4, 82.0, 115.3, 150.5, 153.0, 156.0, 167.0. HRMS (ESI-FT-ICR) calculated for C13H19N6O3 (M + H)+ 307.1513 found 307.1518 calculated for C13H18N6 NaO3 (M + Na)+ 329. 1333, found 329.1346. N N N N HN NH2 O O O 2-Amino-6-N-(acetyl)-9-(tertbutyl acetate)purine (19z). The product was isolated from the above reaction mixture as a ye llow solid (0.009 g, 15%). Mp 259 C; 1H NMR (CDCl3) 1.46 (s, 9H), 2.54 (s, 3H), 4.70 (s, 2H), 4.97 (s, 2H), 7.67 (s, 1H), 8.53 (s, 1H). 1H NMR (DMSO-d6) 1.39 (s, 9H), 2.20 (s, 3H), 4.780 (s, 2H), 6.37 (s, 2H), 7.88 (s, 1H), 10.05 (s, 1H). 13C NMR (DMSO-d6) 24.4, 27.6, 44.1, 82.0, 116.3, 149.7, 159.9, 167.0, 169.1. HRMS (ESIFT-ICR) calculated for C13H19N6O3 (M + H)+ 307.1522 found (M + H)+ 307.1522 calculated for C26H36N12NaO6 (2M + Na)+ 635.2773 found (2M + Na)+ 635.2713. NMR Experimental Parameters NMR spectra were recorded on a Varian Inova spectrometer equipped with a 5 mm indirect detection probe operating at 500 MHz for 1H, 125 MHz for 13C, and at 50 MHz for 15N. The samples were prepared as 0.1 M solutions in dimethyl sulfoxide-d6, containing an equivalent amount of tetramethylsilane and nitromethane. The 1H and 13C chemical are referenced to

PAGE 102

102 tetramethylsilane, 0 ppm. The 15N chemical shifts are given on the liquid ammonia scale, and are referenced to internal nitromethane, 379.5 ppm. The temperature was 25 C. Four experiments were run for each sample: 1H spectrum was run in 1 transient, with an acq uisition time of 5 s, and transformed with no apodization. 13C spectrum was run in 1024 transients, with an acquisition time of 1.3 s, and transformed with a line broadening of 0.5 Hz. The 1H 13C gHMBC had in f2 a spectral window from 10 ppm to -0.5 ppm and 4096 points, yielding a digital re solution of 1.3 Hz. The relaxation delay was 1 s. In f1, the spectral window was from 170 ppm to 5 ppm, and 512 incremen ts were acquired in 1 transient. The total acquisition time was 14 minutes. The experiment was optimized for a 1H 13C long range coupling of 8 Hz. The number of po ints in the f1 dimension was se t to 2048, to provide a digital resolution of 0.085 ppm/point. In both f2 and f1 shifted Gaussian functions were applied, gf = 0.261, gfs = 0.095, gf1 = 0.011, gfs1 = 0.010. The 1H 15N gHMBC had in f2 a spectral window from 10 ppm to 0.5 ppm and 4096 points, yielding a digital re solution of 1.3 Hz. The relaxation delay was 1 s. In f1, the spectral window was from 400 ppm to 50 ppm, and 1024 increm ents were acquired in 1 transient. The total acquisition time was 27 minutes. The experiment was optimized for a 1H 15N long range coupling of 4 Hz. The number of po ints in the f1 dimension was se t to 4096, to provide a digital resolution of 0.085 ppm/point. In both f2 and f1 shifted Gaussian functions were applied, gf = 0.131, gfs = 0.069, gf1 = 0.034, gfs1 = 0.019.

PAGE 103

103 CHAPTER 5 CONCLUSIONS AND FUTURE DIRECTIONS Summary and Conclusions The general design, synthesis, and structure-pr operty relationships of self-complementary QHB ureidodiaminopurines have been described. Urea formation at NC2 affords a DADA motif, which is preorganized into a N3 conformer. The existence of this conformer has been verified by NOESY and VT NMR. Variable te mperature studies also revealed QHB via the N3 conformer to be the predominant mode of assembly at the c oncentration studied. The N3 conformer and its associated dimer were further identified in th e solid state. The stab ility of association (Kdim) was measured by sequential dilution and monitoring the intermolecular hydrogen-bonded urea proton by 1H NMR. The dimerization constants were calcul ated based on the chemical shift of the urea proton using well-known theory, and found to be 820 160 M and 980 290 M for 1a and 1b, respectively. The dimerization constants of the UDAPs were one to two orders of magnitude lower than the comparable ureidodiaminotriazine (104 M) and ureidodiaminopyrimidine (105 M) systems developed by Meijer and coworkers. Three new UDAPs were synthesized to derive structureproperty relationships and understand how modifications at and remote from the hydrogenbonded interface could influence self-association. The urea and N9 substituents were varied to see potential differences between alkyl and aryl ureas, and probe interactions between the N9 and urea substituents with respect to Kdim. Dimers of various structures proved to have similar solution stabilities (Kdim = 820 M), within the margin of error, with the exception of 9. The consistency speaks to the compensating effects that typify most self-assembli ng systems. Apparent steric and electrostatic interactions between the bulky ester N9 substituent and the urea substituent of 9 decreased Kdim

PAGE 104

104 by almost half (530 40 M). Despite the lowered dimerization constant the ester substituent is an attractive point for future functionaliz ation. The heptyl N9 substituent of 8 afforded a nice increase in solubility, and it is foreseen that functionalization of UDAP with a tert-butyl-7heptanoate (or conversion of the tert-butyl to the methyl ester) will relieve unfavorable steric interactions, and afford nice solubility with a po int of functionalization. Additionally, the results indicate that N9 can accommodate substituents of larger size than initially anticipated, even with an aryl urea present. The synthesis of the alkyl UDAPs required new methodology as the reactivity of DAP with alkyl isocyantes wa s found to be sluggish (versus aryl isocyanates), a result rationalized in terms of transition state stabilization. This work ultimately produced the desired alkyl ureas, but found that the site of substitution was NC6 under various conditions. This unexpected regiochemistry spurred a study into the appare nt differential nucleophilic ity of the DAP amino groups. For this purpose an NMR protocol was developed to rigorously confirm the substitution position (as NC6 or NC2). 1H, 13C, 1H13C, and 1H15N 1-D and 2-D NMR techniques nicely elucidated that alkyl isocyanates react regioselectively at NC6 under more pressing conditions (100 C in a pressure tube). Diaminopurine is a useful synthon for pharmaceu tical and materials applications; thus, understanding the reactivity of DAP has practical implications for the development of purine chemistry. The reactivity of DAP was studied under common acylation conditions. The reaction with methoxy acetyl chloride occurred most rapidly at NC2 in agreement with the results of Porcher and coworkers;111 the ratio of the monosubstitution to disubstitution products (9:1) was invariant with solvent (from CH3CN to CH2Cl2) and N9 substituent. These results, along with the seemingly disparate results of Neilsen and coworkers (NC6 substitution of DAP by a

PAGE 105

105 benzyloxycarbonyl triflate),112 prompted a thermodynamic/kinetic study. Acylation of DAP with acetic anhydride and monitoring the reaction pr ogress by NMR generally showed that acylation at the NC2 position was kinetically controlled under st andard pyridine-catalyzed conditions. The conditions that were found to afford NC6 monosubstitution will require further study to elucidate the potential role of deprotonated pu rine intermediates, N7 transition state stabilization, and other more subtle effects. Future Directions Reversible Materials Purines are functional aromatic heterocycles wi th the capacity for core functionalization at multiple sites. Functionalization can be ach ieved through palladium chemistry at C8, by alkylation at N9, and through nucle ophilic/electrophilic s ubstitution at C2 and C6. Not discussed much in this dissertation, but the subject of considerable laborato ry investigation by the author, is covalent functionalization at C8. This chemistry is much less general than the previously mentioned N9 functionalization of DAP through routine alkylation. This position is the most promising, particularly given the apparent insensitivity of the dime rization to its substituents, for connection of UDAP to other scaffolds. Extension at N9 via the ester moiety for UDAP application in PNA chemistry (figure 3-3) could be an early first goal. PNAs are nucleobase analogs that are composed of neutral peptide or peudopeptide backbones with regular or modified nucleobases.116,117 The pseudopeptide backbone typically consists of N-(2-aminoethyl) glycine and replaces the sugar-phosphate backbone of DNA/RNA, which is negatively charged under physiological conditions. The neutral backbone promotes incr eased binding affinity for the minor groove of DNA or RNA by reducing electrostatic repulsion. The PNA backbone features six atoms that separate the

PAGE 106

106 nucleobases to mimic DNA.93,118 PNAs are useful drug targets for gene therapy when the molecular basis of a disease is well known.119 Our interest would be to see whether connect ion of the UDAPs to the PNA scaffold would provide them a framework within which to form self-complementary oligomers, much in the fashion of DNA and RNA. The binding properties of these oligomers with natural nucleobase oligomers could also be probed. Synthesis of the PNA oligomers would be done in two parts. The synthetic protocol for UDAP has been descri bed herein. Saponification of the ester moiety will deliver a unit which is suitable for peptide coupling to ethyl-N-(2-tbutyloxycarbonylaminoethyl)glycina te (DCC and DhbtOH) to afford the PNA monomer after basic hydrolysis of the ester (figure 3-3). 112,118,120 PNA oligomers would be formed through the Merrifield method on solid support.112,120,121 Functionalization at N9 is also appealing for forming linear reversible polymeric systems. Although the degree of polymeri zation (DP) for UDAP is exp ected to be low based on Kdim, excluded volume interactions (phase segregation) can render these intera ctions stronger thereby increasing the DP.40,45 Recently Sivakova and coworkers introduced weakly associating nucleobase derivatives (1.5 1 M for 6-N-(4-methoxy-benzoyl)-9-(dodecyl)aden ine, and 5 2 M for 1-(methoxycarbonylmethyl)-4-N-(4-tert-butylbenzoyl)cytosine, in CDCl3) as selfassembling units to study phase segrega tion aided supramolecular polymerization.122 Low molecular weight amine terminated THF (a soft waxy material, MP = 20 C) was functionalized with the nucleobase derivatives to yield ther moresponsive films (increased MP to 126 C). Attaching the more strongly dimerizing UDAP units to a low molecular weight THF would be a nice extension of this work.

PAGE 107

107 To do this, the N9 ester substituent would serv e as a point of functi onalization. Attachment of the purine moiety would invol ve saponification of the ester a nd then coupling to the amino functionalized poly(THF). Coupling may be realized by traditiona l peptide coupling as in the PNA, or through the method presented by Sivakova and coworkers (coupling of the purine to amino terminated poly(THF) via anhydride form ation of the carboxylic acid and trimethylacetyl chloride and N-methyl morpholine; figure 5-1).122 Figure 5-1. Synthesis of UDA P functionalized low molecu lar weight (< 2000 g mol) poly(THF) for functional materials In addition to the ester functionality, a ribose group at the N9 position can serve as a point for functionalization. Zimmerman and cowork ers have explored this option with a ureidoguanosine unit, and shown that the ribos e nicely accommodates func tionalization from the 5 hydroxyl group without disrupting pr eorganization of the QHB motif.62 A ureido-2aminoadenosine (UA, 23) counterpart is an interesting concept. The results of 1a suggest that a phenyl urea is sterically compatible with bulky N9 substituents. The synthesis of a phenyl ureido-2-aminoadenosine has b een initially explored. The synthesis started from commercia lly available 2,6-diaminoadenosine, 20. The ribose hydroxyl groups were protected following a pro cedure published by Beigelman and coworkers (figure 5-2).123 The 3 and 5 hydroxyl groups were protected by 1,3-

PAGE 108

108 dichlorotetraisopropyldisiloxane (TIPDSiCl2) to yield 21, and the 2 hydroxyl group was methylated (22) under standard conditions. Urea formation was afforded through regioselective reaction at NC2 with phenyl isocyanate (23). Initial results were promising for this unit and further exploration is warranted. Figure 5-2. Original synthesis of compound 23 (UA) The limiting step in the synthesis of 23 was the methylation of the 2 hydroxyl group that employed NaH as the base. As discovered in Chapter 3, the amino groups may also be deprotonated by sodium hydride. Hence the yield suffered as a consequence (31%, 22). The proposed remedy is to deprotonate the hydroxyl groups with pyridine, and protect the 2, 3, and 5 hydroxyl groups in one step with trimethylsilyl chloride to yield 24 (figure 5-3). Ureidodiaminopurines are new scaffolds for s upramolecular applica tions. The quadruple hydrogen bonding face resembles those of system s created by Meijer and coworkers. The UDAPs, however, offer unique structural features that cannot be mimicked by the monocyclic ureidotriazines and ureidopyrimidines. Although the dimerization is weaker by an order of

PAGE 109

109 magnitude, the extended aromatic surfaces offer a dditional sites for functionalization toward the preparation of functional materials. Figure 5-3. Proposed synthesis of UA Alternative QHB Platforms A new class of nucleobase-like compleme ntary and self-complementary quadruple hydrogen bonding systems composed of [4,5]-fused pyrimidine rings we re initially explored as a new generation of QHB units (fig ure 5-4). Urea functionalization of these systems derived from 2,4,6,8-tetrachloro[4,5-d]pyrimido pyrimidine (26) show potential as inte resting supramolecular structures that mimic the urei dopyrimidine units of guanine (28) and diaminopurine (27) (figure 5-4). The urea variants would be the most rapidly accessed monomers for supramolecular polymeric systems yet reported, side-stepp ing the typical requirement for linking QHB recognition units together by a covalent tether (as shown in figure 5-1). It has been confirmed thr ough synthesis beginning from 26 that the 4 and 8 chloro groups are readily displaced by nucleophiles in a matter of minutes at low temperatures.124 Conversely, the 2 and 6 positions require elevated temperatures and extended reaction times for substitution. Based on the reactivity differences between the 2,6, and 4,8 positions of the core, it is expected that there will be a cooperative effect between the two faces of the monomer (essentially a chain effect, or communication from one pyrimidine unit to the next) upon formation of polymeric structures. The reactivity diffe rences around the pyrimidopyrimidine would also be the basis for

PAGE 110

110 possibly differentiating the two hydrogen bonding f aces of the molecule; a hybrid structure of 27 and 28 is conceivable. Figure 5-4. Pyrimidine rings fuse d at the [4,5] carbons show potenti al for a new class of ditopic self-assembling molecules. The synthesis of the self-complementa ry ditopic ureidodiaminopurine mimic, 27, was proposed to be a simple two step procedure (figure 5-5). The starting material, 2,4,6,8tetrachloro[4,5-d]pyrimido pyrimidine, was to be heated in methanolic ammonia in a sealed reaction vessel with a co pper catalyst to yield 29.125 The tetraamino product would then be reacted with phenyl isocya nate to yield the final ditopic DADA QHB motif (27). Preliminary results obtained suggest that harshe r conditions are required for the substitution of the C6 amino group. At 185 C in a sealed re action vessel three of the chloro groups were substituted by methanolic ammonia even upon long reaction times (figure 5-6). A suggested remedy is to use higher temperatures with a stai nless steel reaction vessel rather than a thickwalled glass vessel in the interest of safety. Despite triamino substitution, 30 was reacted with phenyl isocyanate to yield 31 (although the position of phenyl urea has not been rigorously confirmed, only predicted based on the reactivity of 2,6-diaminopurines with phenyl isoc yanate). Limited product solubility necessitated solution studies in DMSO-d6 by 1H NMR for compound 31, where information on hydrogen

PAGE 111

111 bonding could not be obtained. Solid phase analysis (IR or solid-state NMR) might be useful to characterize this polar system. Figure 5-5. Synthesis of 4,8diamino[4,5-d]pyrimido-2,6-bis phenylureidopyrimidine by a simple two-step process. Figure 5-6. Products obtained in preliminary synthesis aimed at forming 4,8-diamino-6chloro[4,5-d]pyrimido-2-N-phenylureidopyrimidine. The proposed synthesis of the complementar y QHB fused guanine mimic also began with 2,4,6,8-tetrachloro[4,5-d]pyrimido pyrimidine 26 (figure 5-7). In this case the differential reactivity of the 2,6 and 4,8 positions was used advantageously to plac e the benzyloxy groups at the 4 and 8 positions to yield 32 quantitatively. The amino groups were placed in the 2 and 6 positions by copper sulfate catalyzed substitu tion with ammonia in methanol to yield 33. Compound 34 was formed (74 %) by heating compound 24 with hexyl isocyanate and pyridine to 100 C in a sealed reaction vessel. Reduction of the benzyloxy groups at C4 and C8 to give 28 was attempted by hydrogenation with Pd/C and H2 as used in the debenzylation of protected guanine.126 The desired product was not visibly present in the inse parable product mixture, and decomposition was evident.

PAGE 112

112 Figure 5-7. Synthesis of 4, 8-dibenzyloxy[4,5-d]pyrimido2,6-N-hexylureidopyrimidine (34) and unsuccessful reduction by hydrogenation. A possible remedy is to use a different hydrogen donor in the re duction (figure 5-8). Ammonium formate reduction has been shown to be compatible with the urea in the azide reduction to form compound 9. It should therefore be feasib le for the formation of compound 28. A mono-faced counterpart to the ditopic systems derived from 2,4-dichloroquinazoline (35) was also explored (figure 5-9). Compound 36 was visualized as a chain stopper for the selfcomplementary ditopic diaminopurine mimic (fig ure 5-9), but a mono-faced guanine mimic (37) will provide an equally interesting platform. Al so, a substituent in the position indicated R could be used to explore interactions with the neighboring urea (e.g. an OCH3 group). Functionalization elsewhere on the quinazoline ring of either 36 or 37 will transition these monomers into functional materials. It was envi sioned that the synthesi s of these units from

PAGE 113

113 dichloroquinazoline 35 would afford similar reactivity patterns at the C2 and C4 positions that were observed in the synthesis of DAP derivatives. Figure 5-8. The sugge sted synthesis for 28 by catalytic hydrogen tr ansfer from ammonium formate in refluxing methanol Figure 5-9. The mono-faced counter parts to the ditopic systems 36 and 37 derived from 2,4chloroquinazolines. The 2,4-dichloroquinazoli ne starting material (35) was obtained from commercially available 2-aminobenzonitrile (39, figure 5-11). Compound 39 in the presence of diphosgene forms a phenyl isocyanate which undergoes ring formation by acetonitrile incorporation and subsequent chlorination to form compound 35.127 As expected, the reaction of compound 35 with ammonia in methanol did appear to show the reactivity preference encoun tered in the synthesis of compounds 29/30. Sites C2 and C4 react at different rates, and compound 40 was obtained. It is projected that harsher conditions will a fford the desired 2,4-diaminoquinazoline (38), which can be urea-functionalized to yield a DADA QHB motif 36. Additionally, the synthesis of the

PAGE 114

114 quinazoline-derived guanine analog is proposed to follow the same reaction sequence as the ditopic system (28). Figure 5-10. The proposed synthesis of 4-amino-2-Nphenylureidoquinazoline is similar to the ditopic counterpart. Figure 5-11. Synthesis of 2,6-dichloroquinazoline 29 from 2-aminobenzonitrile 27 Preliminary Rxperimental Results 2,6-Diamino-9-[3 5 O -tetraisopropyldisiloxane-1, 3diyl)--D-ribofuranosyl]purine (21). Vacuum-dried 2,6-diamino-9(-D-ribofura nosyl)purine (0.499 g, 0.177 mmol) was placed in an oven-dried round bottomed flask. Under argon, DMF (5 mL) and pyridine (8 mL) were added, and the suspension was cooled to 0 C. Th e temperature was maintain ed at 0 C with the dropwise addition (over 30 min) of TIPDSiCl2 (0.700 g, 2.12 mmol). The mixture was allowed to warm to room temperature and the reaction was monitored to completion by TLC. The reaction was quenched with ethanol (1 mL). The solvent was evaporated and the residue was

PAGE 115

115 dissolved in EtOAc and rinsed with saturated aq NaHCO3. Upon evaporation of the solvent, the product was purified by column chromatography (EtOAC/hexanes/EtOH 2:1:1) to afford 21 as a white solid (0.929 g, quantitative). 1H NMR (DMSO-d6) 1.02 (d, J = 3.7 Hz,12H), 1.06 (m, 16H), 3.33 (d, J = 0.5 Hz, 2H), 4.00 (m, 3H), 4.29 (t, J = 4.8 Hz, 2H), 4.44 (q, J = Hz, 1H), 5.58 (d, J = 4.8 Hz, 2H), 5.71 (d, J = 1.4 Hz, 1H), 5.77 (s, 1H), 6.77 (s, 2H), 7.77 (s, 1H). LRMS (ESI-FT-ICR) calculated for C22H41N6O5Si2 (M + H)+ 525, found 525. 2,6-Diamino-9-[3 5 -O-tetraisopropyldisiloxane-1, 3-diyl)-2-O-methyl--Dribofuranosyl]purine (22). Under a blanket of argon to a solution of compound 21 (0.375 g, 0.710 mmol) in DMF (7.3 mL) was added CH3I (0.133 g, 2.10 mmol). The reaction was cooled to 0 C and NaH (0.043 g, 1.8 mmol) was added. The reaction was stirred for 40 min at 0 C, and then quenched with ethanol (1mL). Cold CH2Cl2 was added to the reaction mixture and the solution washed with saturated aq NH4Cl and brine. The solvent was removed and the yellow oil recrystallized with ethanol/w ater (1:1) to yield a white solid (0.121 g, 31%). 1H NMR (CDCl3) 1.07 (m, 29H), 1.25 (m, 2H),1.79 (s, 4H), 3.67 (s, 2H), 3.72 (d, J = 3.4 Hz, 1H), 3.98 (d, J = 3.2 Hz, 1H), 4.02 (d, J = 4.8 Hz, 1H), 4.10 (d, J = 4.3 Hz, 1H), 4.20 (d, J = 6.0 Hz, 1H) 4.60 (q, J = 4.5 Hz, 1H), 4.73 (s, 2H, 5.40 (s, 2H), 5.86 (s, 1H), 7.81 (s, 1H). LRMS (ESI-FT-ICR) calculated for C23H43N6O5Si2 (M + H)+ 535 found 535.

PAGE 116

116 N N N N NH2 O OCH3 O O NH (iPr)2Si (iPr)2Si O H N O Ph 6-Amino-2-N-(phenylamino)carbonyl-9-[3 5 -O-tetraisopropyldisiloxane-1, 3-diyl)2-O-methyl--D-ribofu ranosyl]purine (23). Previously vacuum-dried 22 (0.05 g, 0.09 mmol) was placed in a 25 mL oven-dried two-necked round bottomed flask and stirred under vacuum for approximately one hour followed by 15 min of stirring under Ar. By syringe, CH2Cl2 (4.0 mL) was added followed by pyridine (0.01 mL) a nd phenylisocyanate (0. 01 mL, 0.12 mmol). The mixture was allowed to stir at room te mperature for 1.75 h until there was no observable change by TLC. The solvent was removed under vacuum and the residue was purified by column chromatography (5% MeOH, CH2Cl2) to yield 23 as a white solid (0.070 g, 89 %). 1H NMR (CDCl3) 1.02 (m, 34H), 1.24 (s, 2H), 3.56 (s, 3H), 3.85 (d, J = 4.68 Hz, 1H), 4.06 (m, 1H), 4.28 (m, 2H), 4.55 (m, 1H), 6.05 (s 1H), 7.08 (t, J = 8.06 Hz, 2H), 7.35 (t, J = 7.68 Hz, 3H), 7.68 (t, J = 7.78 Hz, 2H), 8.04 (s, 1H), 11.8 (s, 1H). 13C NMR (CDCl3) 13.0, 13.4, 13.9, 30.2, 59.9, 60.4, 70.1, 81.6, 84.0, 88.2, 115.5, 137.5, 151.6, 155.3, 159.1. HRMS (ESI-FT-ICR) calculated for C30H47O6N7Si2Na (M + Na)+ 680.3019, found 680.3017. N N N N O Cl Cl O

PAGE 117

117 4,8-Dibenzyloxy-2,6-dichloro-pyrim idino[5,4-d]pyrimidine (32).Tetrahydrofuran (25 mL) was transferred by syringe to an oven dr ied two necked-round-botto med flask under argon. Benzyl alcohol (0.314 mL, 3.02 mmol) and sodi um hydride (0.123 g, 3.02 mmol, 60% dispersion in mineral oil) were added and the mixture was stirred for 30 min at room temperature. 2,4,6,8Tetrachloropyrimido[5,4-d]pyrimidine (0.160 g, 0.610 mmol) was dissolved in THF (10 mL) and added dropwise to the reaction mixture. The mixture was stirred for 50 min after which the reaction was quenched by the addition of water. The product was extracted with chloroform and washed with brine. The organic layers were combined and the solvent was evaporated. The product was used without further purification (0.269 g, quant.). 1H NMR (CDCl3) 5.68 (s, 4H), 7.35 (m, 6H), 7.53 (m, 4H). 13C NMR (CDCl3) 71.1, 128.7, 128.9, 129.3, 134.2, 136.5, 156.6, 165.8. HRMS (ESI-FT-ICR) calculated for C20H15Cl2N4O2 (M + H)+ 413.2568, found 413.2664; calculated for C20H14Cl2N4O2Na (M + Na)+ 435.0391, found 435.0385. 2,6-Diamino-4,8-dibenzyloxypyrim idino[5,4-d]pyrimidine (33). Compound 32 (0.054 g, 0.13 mmol) was placed in a pressure tube. Meth anolic ammonia (15 mL) was added along with copper(II)sulfate (0.002 g, 0.01 mm ol). The mixture was heated to 140 C and stirred 19 h. The solvent was evaporated and the product was purified by column chromatography (5% CH3OH/CH2Cl2) to yield a yellow/white powder (0.058 g, quant.). 1H NMR (DMSO-d6) 5.38 (s, 4H), 7.36 (m, 10H), 7.44 (m, 4H). 13C NMR (DMSO-d6) 67.5, 127.6, 127.9, 128.3, 137.5,

PAGE 118

118 159.6, 162.4. HRMS (ESI-FT-ICR) calculated for C20H18N6O2Na (M + Na)+ 397.1383, found 397.1397; calculated for C20H18N6O2Na2 (2M + Na)+ 771.2880, found 771.2894. N N N N O N H H N O H N N H O O 2,6-N-dihexylamino-4,8-dibenzyloxyureidopyrimidino[5,4-d]pyrimidine (34). Compound 33 (0.020 g, 0.053 mmol) was placed in a 75 mL pressure tube. Pyridine (2 mL) and hexylisocyanate (0.057 mL, 0.037 mmol) were added. The reaction mixture was heated to 100 C and stirred 20 h. Solvent was removed under reduced pressure a nd the product purified by column chromatography (1% CH3OH/CH2Cl2) to yield the product as a white solid (0.0303 g, 89%). 1H NMR (CDCl3) 0.86 (m, 6H), 1.29 (m, 12H ), 1.59 (m, 4H), 3.38 (q, J = 3.2 Hz 4H), 5.42 (s, 4H), 7.42 (m, 10H), 8.47 (s, 2H), 8.99 (t, J = 2.7 Hz, 2H). 13C NMR (CDCl3) 14.2, 22.8, 26.9, 29.7, 31.7, 40.0, 128.2, 128.6, 128.9, 130.2, 135.9, 152.8, 157.2, 159.0. HRMS (ESIFT-ICR) calculated for C34H44N8O4Na (M + Na)+ 651.3377, found 651.3416; calculated for C34H44N8O4Na2 (2M + Na)+ 1279.6863, found 1279.6957.

PAGE 119

119 APPENDIX A NMR DATA

PAGE 120

120

PAGE 121

121

PAGE 122

122

PAGE 123

123

PAGE 124

124

PAGE 125

125

PAGE 126

126

PAGE 127

127

PAGE 128

128

PAGE 129

129

PAGE 130

130

PAGE 131

131

PAGE 132

132

PAGE 133

133

PAGE 134

134

PAGE 135

135

PAGE 136

136

PAGE 137

137

PAGE 138

138

PAGE 139

139

PAGE 140

140

PAGE 141

141

PAGE 142

142

PAGE 143

143

PAGE 144

144

PAGE 145

145

PAGE 146

146

PAGE 147

147

PAGE 148

148

PAGE 149

149

PAGE 150

150

PAGE 151

151

PAGE 152

152

PAGE 153

153

PAGE 154

154

PAGE 155

155

PAGE 156

156 APPENDIX B SOLID STATE DATA Table 1. Crystal data and structure refinement for 1a Identification code mart2 Empirical formula C100 H122 N32 O9 Formula weight 1916.30 Temperature 173(2) K Wavelength 0.71073 Crystal system Triclinic Space group P-1 Unit cell dimensions a = 14.8752(11) = 87.519(2). b = 19.2433(14) = 78.459(1). c = 19.4178(15) = 68.260(2). Volume 5055.7(7) 3 Z 2 Density (calculated) 1.259 Mg/m3 Absorption coefficient 0.085 mm-1 F(000) 2036 Crystal size 0.25 x 0.09 x 0.09 mm3 Theta range for data coll ection 1.07 to 27.50. Index ranges -19 h 9, -25 k 20, -25 l 24 Reflections collected 34510 Independent reflections 22702 [R(int) = 0.0761] Completeness to theta = 27.50 97.7 % Absorption correction Integration Max. and min. transmission 0.9945 and 0.9862 Refinement method Full-matrix least-squares on F2 Data / restraints / parameters 22702 / 0 / 1340 Goodness-of-fit on F2 0.905 Final R indices [I>2sigma(I)] R1 = 0.0592, wR2 = 0.1127 [7753] R indices (all data) R1 = 0.2065, wR2 = 0.1637 Largest diff. peak and hole 0.275 and -0.294 e.-3

PAGE 157

157 LIST OF REFERERENCES (1) Cunliffe, D.; Pennadam S.; Alexander, C. Eur. Polym. J. 2004, 40, 5-25. (2) Fricke, M.; Schadler, V. Macromol. Biosci. 2007, 7, 103-104. (3) Ghosh, A.; Bansal, M. Acta Crystallogr. Sect. D-Biol. Crystallogr. 2003, 59, 620-626. (4) Ponnuswamy, P. K.; Gromiha, M. M. J. Theor. Biol. 1994, 169, 419-432. (5) Ponnuswamy, P. K.; Gromiha, M. M. J. Theor. Biol. 1994, 166, 63-74. (6) Principles of Biochemistry; 4 ed.; Nelson, D. L.; Cox, M., M., Eds.; W. H. Freeman, 2004. (7) Lehn, J. M. Chem. Soc. Rev. 2007, 36, 151-160. (8) Whitesides, G. M.; Grzybowski, B. Science 2002, 295, 2418-2421. (9) Lemieux, R. P. Chem. Soc. Rev. 2007, 36, 2033-2045. (10) Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 7017-&. (11) Pedersen, C. J. J. Am. Chem. Soc. 1967, 89, 2495-&. (12) Ball, P. Designing the Molecular World: Chemistry at the Frontier; Princeton University Press: Princeton, N.J., 1994. (13) Mandolini, L.; Ungaro, R. Calixarenes in Action; Imperial College Press: London, U.K., 2000. (14) Rogers, C. W.; Wolf, M. O. Coord. Chem. Rev. 2002, 233, 341-350. (15) de Silva, A. P.; Fox, D. B.; Huxley, A. J. M.; Moody, T. S. Coord. Chem. Rev. 2000, 205, 41-57. (16) deSilva, A. P.; Gunaratne, H. Q. N.; Gunnl augsson, T.; Huxley, A. J. M.; McCoy, C. P.; Rademacher, J. T.; Rice, T. E. Chem. Rev. 1997, 97, 1515-1566. (17) Valeur, B.; Leray, I. Coord. Chem. Rev. 2000, 205, 3-40.

PAGE 158

158 (18) Beer, P. D.; Cadman, J. Coord. Chem. Rev. 2000, 205, 131-155. (19) Fabbrizzi, L.; Licchelli, M.; Rabaioli, G.; Taglietti, A. Coord. Chem. Rev. 2000, 205, 85108. (20) Watanabe, S.; Onogawa, O.; Komatsu, Y.; Yoshida, K. J. Am. Chem. Soc. 1998, 120, 229-230. (21) Chan, W. H.; Mak, T. C. W.; Che, C. M. J. Chem. Soc., Dalton Trans. 1998, 2275-2276. (22) Rowan, S. J.; Cantrill, S. J.; Cousins, G. R. L.; Sanders, J. K. M.; Stoddart, J. F. Angew. Chem. Int. Ed. 2002, 41, 898-952. (23) Flood, A. H.; Wong, E. W.; Stoddart, J. F. Chem. Phys. 2006, 324, 280-290. (24) Haussmann, P. C.; Khan, S. I.; Stoddart, J. F. J. Org. Chem. 2007, 72, 6708-6713. (25) Miljanic, O. S.; Dichtel, W. R.; Khan, S. I.; Mortezaei, S.; Heath, J. R.; Stoddart, J. F. J. Am. Chem. Soc. 2007, 129, 8236-8246. (26) Miljanic, O. S.; Stoddart, J. F. Proc. Natl. Acad. Sci. U. S. A. 2007, 104, 12966-12970. (27) Nygaard, S.; Leung, K. C. F.; Aprahamian, I.; Ikeda, T.; Saha, S.; Laursen, B. W.; Kim, S. Y.; Hansen, S. W.; Stein, P. C.; Flood, A. H.; Stoddart, J. F.; Jeppesen, J. O. J. Am. Chem. Soc. 2007, 129, 960-970. (28) Saha, S.; Flood, A. H.; Stoddart, J. F.; Impellizzeri, S.; Silvi, S.; Venturi, M.; Credi, A. J. Am. Chem. Soc. 2007, 129, 12159-12171. (29) Saha, S.; Stoddart, J. F. Chem. Soc. Rev. 2007, 36, 77-92. (30) Hoeben, F. J. M.; Jonkheijm, P.; Meijer, E. W.; Schenning, A. Chem. Rev. 2005, 105, 1491-1546. (31) Taylor, R.; Kennard, O. Acc. Chem. Res. 1984, 17, 320-326. (32) Etter, M. C. Acc. Chem. Res. 1990, 23, 120-126. (33) Rowland, R. S.; Taylor, R. J. Phys. Chem. 1996, 100, 7384-7391. (34) Jorgensen, W. L.; Pranata, J. J. Am. Chem. Soc. 1990, 112, 2008-2010.

PAGE 159

159 (35) Pranata, J.; Wierschke, S. G.; Jorgensen, W. L. J. Am. Chem. Soc. 1991, 113, 2810-2819. (36) Sartorius, J.; Schneider, H. J. Chem. Eur. J. 1996, 2, 1446-1452. (37) Corbin, P. S.; Zimmerman, S. C. J. Am. Chem. Soc. 1998, 120, 9710-9711. (38) Lange, R. F. M.; Meijer, E. W. Macromolecules 1995, 28, 782-783. (39) Hirschberg, J.; Brunsveld, L.; Ramzi, A.; Vekemans, J.; Sijbesma, R. P.; Meijer, E. W. Nature 2000, 407, 167-170. (40) Brunsveld, L.; Folmer, B. J. B.; Meijer, E. W.; Sijbesma, R. P. Chem. Rev. 2001, 101, 4071-4097. (41) Ciferri, A. Macromol. Rapid Comm. 2002, 23, 511-529. (42) ten Cate, A. T.; Sijbesma, R. P. Macromol. Rapid Comm. 2002, 23, 1094-1112. (43) Shimizu, L. S. Polym. Int. 2007, 56, 444-452. (44) Bedrov, D.; Smith, G. D.; Douglas, J. F. Europhys. Lett. 2002, 59, 384-390. (45) Martin, R. B. Chem. Rev. 1996, 96, 3043-3064. (46) Gulikkrzywicki, T.; Fouquey, C.; Lehn, J. M. Proc. Natl. Acad. Sci. U. S. A. 1993, 90, 163-167. (47) Kotera, M.; Lehn, J. M.; Vigneron, J. P. J. Chem. Soc., Chem. Commun. 1994, 197-199. (48) Kotera, M.; Lehn, J. M.; Vigneron, J. P. Tetrahedron 1995, 51, 1953-1972. (49) Lehn, J. M. Makromol. Chem.-Macromol. Symp. 1993, 69, 1-17. (50) Ducharme, Y.; Wuest, J. D. J. Org. Chem. 1988, 53, 5787-5789. (51) Gallant, M.; Viet, M. T. P.; Wuest, J. D. J. Org. Chem. 1991, 56, 2284-2286. (52) Boucher, E.; Simard, M.; Wuest, J. D. J. Org. Chem. 1995, 60, 1408-1412. (53) Sijbesma, R. P.; Meijer, E. W. Chem. Commun. 2003, 5-16.

PAGE 160

160 (54) Folmer, B. J. B.; Sijbesma, R. P.; Versteegen, R. M.; van der Rijt, J. A. J.; Meijer, E. W. Adv. Mater. 2000, 12, 874-878. (55) Beijer, F. H., Technisc he Universiteit Eindhoven, 1998. (56) Beijer, F. H.; Sijbesma, R. P.; Vekemans, J.; Meijer, E. W.; Kooijman, H.; Spek, A. L. J. Org. Chem. 1996, 61, 6371-6380. (57) Beijer, F. H.; Sijbesma, R. P.; Kooijman, H.; Spek, A. L.; Meijer, E. W. J. Am. Chem. Soc. 1998, 120, 6761-6769. (58) Lange, R. F. M.; Van Gurp, M.; Meijer, E. W. J. Polym. Sci., Part A: Polym. Chem. 1999, 37, 3657-3670. (59) Dankers, P. Y. W.; Harmsen, M. C.; Brouwer L. A.; Van Luyn, M. J. A.; Meijer, E. W. Nature Mat. 2005, 4, 568-574. (60) Sivakova, S.; Rowan, S. J. Chem. Soc. Rev. 2005, 34, 9-21. (61) Ong, H. C.; Zimmerman, S. C. Org. Lett. 2006, 8, 1589-1592. (62) Park, T.; Todd, E. M.; Nakashima, S.; Zimmerman, S. C. J. Am. Chem. Soc. 2005, 127, 18133-18142. (63) Park, T.; Zimmerman, S. C. J. Am. Chem. Soc. 2006, 128, 14236-14237. (64) Park, T.; Zimmerman, S. C.; Nakashima, S. J. Am. Chem. Soc. 2005, 127, 6520-6521. (65) Asadi, A.; Patrick, B. O.; Perrin, D. M. J. Org. Chem. 2007, 72, 466-475. (66) Sessler, J. L.; Jayawickramarajah, J. Chem. Commun. 2005, 1939-1949. (67) Sessler, J. L.; Jayawickramarajah, J.; Gouloumis, A.; Torres, T.; Guldi, D. M.; Maldonado, S.; Stevenson, K. J. Chem. Commun. 2005, 1892-1894. (68) Sessler, J. L.; Lawrence, C. M.; Jayawickramarajah, J. Chem. Soc. Rev. 2007, 36, 314325. (69) Sivakova, S.; Rowan, S. J. Chem. Soc. Rev. 2005, 34, 9-21. (70) Beijer, F. H.; Kooijman, H.; Spek, A. L.; Sijbesma, R. P.; Meijer, E. W. Angew. Chem. Int. Ed. 1998, 37, 75-78.

PAGE 161

161 (71) Sijbesma, R. P.; Beijer, F. H.; Brunsveld, L.; Folmer, B. J. B.; Hirschberg, J.; Lange, R. F. M.; Lowe, J. K. L.; Meijer, E. W. Science 1997, 278, 1601-1604. (72) Sntjens, S. H. M.; Sijbesma, R. P. ; van Genderen, M. H. P.; Meijer, E. W. J. Am. Chem. Soc. 2000, 122, 7487-7493. (73) Camaioni, E.; Costanzi, S.; Vittori, S.; Volpini, R.; Klotz, K. N.; Cristalli, G. Bioorg. Med. Chem. 1998, 6, 523-533. (74) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463. (75) Butler, R. S., University of Florida, 2007. (76) Eur. Pat., 1993; Vol. EP 543095 (77) van der Made, A. W.; van der Made, R. H. J. Org. Chem. 1993, 58, 1262-1263. (78) Aathimanikandan, S. V.; Sandanaraj, B. S.; Arges, C. G.; Bardeen, C. J.; Thayumanavan, S. Org. Lett. 2005, 7, 2809-2812. (79) Brun, K. A.; Linden, A.; Heimgartner, H. Helv. Chim. Acta 2002, 85, 3422-3443. (80) Connors, K. A. Binding Constants, Wiley: New York, 1987. (81) Fielding, L. Tetrahedron 2000, 56, 6151-6170. (82) Nogales, D. F.; Ma, J. S.; Lightner, D. A. Tetrahedron 1993, 49, 2361-2372. (83) Deranlea, D. A. J. Am. Chem. Soc. 1969, 91, 4050-4054. (84) Bisson, A. P.; Hunter, C. A.; Morales, J. C.; Young, K. Chem. Eur. J. 1998, 4, 845-851. (85) Sanderson, J. M., University of Durham http://www.dur.ac.uk/j.m.sanderson/science/sheets/NMR_Fit_HH.pdf (86) Moham adi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W. C. J. Comput. Chem. 1990, 11, 440-467.

PAGE 162

162 (87) Frisch, M. J. T., G. W. Schlegel, H. B. Scuseria, G. E. Robb, M. A. Cheeseman, J. R. Montgomery, Jr., J. A. Vreven, T. Kudin, K. N. Burant, J. C. Millam, J. M. Iyengar, S. S. Tomasi, J. Barone, V. Mennucci, B. Cossi, M. Scalmani, G. Rega, N. Petersson, G. A. Nakatsuji, H. Hada, M. Ehara, M. Toyota, K. Fukuda, R. Hasegawa, J. Ishida, M. Nakajima, T. Honda, Y. Kitao, O. Nakai, H. Kl ene, M. Li, X. Knox, J. E. Hratchian, H. P. Cross, J. B. Bakken, V. Adamo, C. Jaramillo, J. Gomperts, R. Stratmann, R. E. Yazyev, O. Austin, A. J. Cammi, R. Pomelli, C. Ochterski, J. W. Ayala, P. Y. Morokuma, K. Voth, G. A. Salv ador, P. Dannenberg, J. J. Za krzewski, V. G. Dapprich, S. Daniels, A. D. Strain, M. C. Farkas, O. Malick, D. K. Rabuck, A. D. Raghavachari, K. Foresman, J. B. Ortiz, J. V.Cui, Q. Baboul, A. G. Clifford, S. Cioslowski, J. Stefanov, B. B. Liu, G. Liashenko, A. Piskor z, P. Komaromi, I. Martin, R. L. Fox, D. J. Keith, T. AlLaham, M. A. Peng, C. Y. Nanayakkara, A. Ch allacombe, M. Gill, P. M. W. Johnson, B. Chen, W. Wong, M. W. Gonzalez, C. Pople, J. A. Gaussian, Inc., Wallingford CT, 2004. (88) Agarwal, A.; Srivastava, K.; Puri, S. K.; Chauhan, P. M. S. Bioorg. Med. Chem. Lett. 2005, 15, 531-533. (89) Brown, D. J.; Teitei, T. J. Chem. Soc. 1965, 755-&. (90) Blindauer, C. A.; Sjastad, T. I.; Holy, A.; Sletten, E.; Sigel, H. J. Chem. Soc., Dalton Trans. 1999, 3661-3671. (91) Sigel, H. Pure Appl. Chem. 2004, 76, 1869-1886. (92) Ward, D. C.; Reich, E.; Stryer, L. J. Biol. Chem. 1969, 244, 1228-&. (93) Cerea, P.; Giannini, C.; Dall'Angelo, S. ; Licandro, E.; Maiorana, S.; Marchelli, R. Tetrahedron 2007, 63, 4108-4119. (94) Nelson, K. E.; Levy, M.; Miller, S. L. Proc. Natl. Acad. Sci. U. S. A. 2000, 97, 38683871. (95) Watkins, B. E.; Kiely, J. S.; Rapoport, H. J. Am. Chem. Soc. 1982, 104, 5702-5708. (96) Brik, A.; Wu, C. Y.; Best, M. D.; Wong, C. H. Bioorg. Med. Chem. 2005, 13, 4622-4626. (97) Organic Chemistry; 2nd ed.; Jones, M., Ed.; W. W. Norton and Company, Inc.: New York, N.Y., 2000. (98) Carver, F. J.; Hunter, C. A.; Jones, P. S.; Livingstone, D. J.; McCabe, J. F.; Seward, E. M.; Tiger, P.; Spey, S. E. Chem. Eur. J. 2001, 7, 4854-4862. (99) Martinez, A. G.; Barcina, J. O.; Cerezo, A. D. Chem. Eur. J. 2001, 7, 1171-1175.

PAGE 163

163 (100) Meyer, E. A.; Castellano, R. K.; Diederich, F. Angew. Chem. Int. Ed. 2003, 42, 12101250. (101) Salaun, A.; Favre, A.; Le Gr el, B.; Potel, M.; Le Grel, P. J. Org. Chem. 2006, 71, 150158. (102) Chang, Y. T.; Gray, N. S.; Rosania, G. R.; Sutherlin, D. P.; Kwon, S.; Norman, T. C.; Sarohia, R.; Leost, M.; Meijer, L.; Schultz, P. G. Chem. Biol. 1999, 6, 361-375. (103) Cosyn, L.; Palaniappan, K. K.; Kim, S. K. ; Duong, H. T.; Gao, Z. G.; Jacobson, K. A.; Van Calenbergh, S. J. Med. Chem. 2006, 49, 7373-7383. (104) Gryaznov, S.; Schultz, R. G. Tetrahedron Lett. 1994, 35, 2489-2492. (105) Havlicek, L.; Hanus, J.; Vesely, J.; Lecl erc, S.; Meijer, L.; Shaw, G.; Strnad, M. J. Med. Chem. 1997, 40, 408-412. (106) Malon, M.; Travnicek, Z.; Marek, R.; Strnad, M. J. Inorg. Biochem. 2005, 99, 21272138. (107) Norman, T. C.; Gray, N. S.; Koh, J. T.; Schultz, P. G. J. Am. Chem. Soc. 1996, 118, 7430-7431. (108) Marco, E.; Negri, A.; Luque, F. J.; Gago, F. Nucleic Acids Res. 2005, 33, 6214-6224. (109) Busca, P.; McCort, I.; Prange, T.; Le Merrer, Y. Eur. J. Org. Chem. 2006, 2403-2409. (110) Shallop, A. J.; Gaffney, B. L.; Jones, R. A. J. Org. Chem. 2003, 68, 8657-8661. (111) Porcher, S.; Pitsch, S. Helv. Chim. Acta 2005, 88, 2683-2704. (112) Haaima, G.; Hansen, H. F.; Christ ensen, L.; Dahl, O.; Nielsen, P. E. Nucleic Acids Res. 1997, 25, 4639-4643. (113) Meyers, C. Y.; Malte, A. M. J. Am. Chem. Soc. 1969, 91, 2125-2126. (114) Brady, W. T.; Lloyd, R. M. J. Org. Chem. 1980, 45, 2025-2028. (115) Rappoport, Z.; Lei, Y. X.; Yamataka, H. Helv. Chim. Acta 2001, 84, 1405-1431. (116) Egholm, M.; Buchardt, O.; Christensen, L. ; Behrens, C.; Freier, S. M.; Driver, D. A.; Berg, R. H.; Kim, S. K.; Norden, B.; Nielsen, P. E. Nature 1993, 365, 566-568.

PAGE 164

164 (117) Nielsen, P. E.; Egholm, M.; Berg, R. H.; Buchardt, O. Science 1991, 254, 1497-1500. (118) Ikeda, H.; Kitagawa, F.; Nakamura, Y. Tetrahedron 2007, 63, 5677-5689 (119) Spitzer, G. M.; Wellenzohn, B.; Laggner, C.; Langer, T.; Liedl, K. R. J. Chem. Inf. Model. 2007, 47, 1580-1589. (120) Christensen, L., Fitzpatrick,R., Gild ea,B., Petersen,K.H., Hansen,H.F.,; Koch, T., Egholm,M., Buchardt,O., Nielsen,P.E., Coull,J. and Berg,R.H. J. Pept. Sci. 1995, 3, 175183. (121) Merrifield, R. B. J. Am. Chem. Soc. 1963, 85, 2149-&. (122) Sivakova, S.; Bohnsack, D. A.; Mackay, M. E.; Suwanmala, P.; Rowan, S. J. J. Am. Chem. Soc. 2005, 127, 18202-18211. (123) Beigelman, L.; Haeberli, P.; Sweedler, D.; Karpeisky, A. Tetrahedron 2000, 56, 10471056. (124) Curtin, N. J.; Barlow, H. C.; Bowman, K. J. ; Calvert, A. H.; Davison, R.; Golding, B. T.; Huang, B.; Loughlin, P. J.; Newell, D. R.; Smith, P. G.; Griffin, R. J. J. Med. Chem. 2004, 47, 4905-4922. (125) Pamukcu, R.; Piazza, G. A.; Pa tent, U. S., Ed. 2001; Vol. 6,232,312. (126) Girgis, N. S.; Cottam, H. B.; Larson, S. B.; Robins, R. K. J. Heterocyclic Chem. 1987, 24, 821-827. (127) Lee, J. H.; Lee, B. S.; Shin, H.; Nam, D. H.; Chi, D. Y. Synlett 2006, 65-68.

PAGE 165

165 BIOGRAPHICAL SKETCH Alisha Michelle Martin was born on February 12, 1975, in Ga inesville, Florida. She was the mi ddle child of two brothers, and grew up in Jacksonville, Florida. She was always an independent soul who wanted to do things her way without qu estion. She always held an appreciation for nature, and the way things work Whether it be the physical or the mechanical, Alisha was always curious. She would sit for hours and ponder such realities. All she needed was encouragement to pursue the formal education ne cessary to make the most of her favorite past times; thinking, questioning, and dreaming. Alisha had her own way of thinking. To view the world through her own eyes, but obtain the knowledge of the wiser world that made her pursuit challenging and therefore fufilling. After high school Alisha attended Florida Community College at Jacks onville for a gain in momentum. From there she went on to earn her bachelors degree in Chemistry from the University of North Florida. Upon graduation from UNF, in spring of 2001, she obtained a posi tion as an analytical chemist at an environmental testing compa ny in Jacksonville, Florida, Environmental Conservation Laboratories (ENCO). It was quickly apparent that th is would not satisfy her need to search for the answers to the bigger questions, and decided that continuing education was the best option. She began studies at the University of Florida in fall 2002, as a post-baccalaureate with Dr. Castellano. Alisha remained under the advisement of Ron Castellano to obtain a Ph. D. in organic chemistry in the Fall of 2007. It is through the diversity of her experiences and education that she has become a well rounded individual.