PART 1. SYNTHESIS OF N-15 LABELED (R)-DEUTERIOGLYCINE
PART 2. SYNTHESES OF CARBON-LINKED ANALOGS OF RETINOID GLYCOSIDE CONJUGATES
DISSERTATION
Presented in Partial Fulfillment of the Requirements
for the Degree Doctor of Philosophy
in the Graduate School of The Ohio State University
By
Joel R. Walker
* * * * *
The Ohio State Univeristy
2003
Dissertation Committee: Approved by Robert W. Curley, Jr., Ph.D., Adviser
Werner Tjarks, Ph.D. ______Adviser Karl A. Werbovetz, Ph.D. College of Pharmacy
ABSTRACT
(R)-Glycine-d-15N has been used to permit assignments of the prochiral α-protons
of glycine residues in the FK-506 binding protein. A key and low yielding step in the
synthetic route to (R)-glycine-d-15N occurred in the ruthenium tetraoxide-mediated
degradation of N-t-BOC-p-methoxybenzyl amine to the N-t-BOC-glycine after both 2H
and 15N are incorporated. In order to improve this step, investigation of the oxidation
reaction conditions along with various aromatic ring carboxylate precursors were
undertaken. It was found that using ruthenium chloride, periodic acid as the
stoichiometric re-oxidant, and N-(p-methoxyphenylmethylamine)-2,2,2-trichloroethyl carbamate were the optimal conditions and substrate. This improvement was paramount for the applicability of this route for large scale production of labeled glycine that could be used in other biological applications.
The retinoic acid analog N-(4-hydroxyphenyl)retinamide (4-HPR) is an effective chemopreventative and chemotherapeutic for numerous types of cancer. In vivo, 4-HPR is metabolized to 4-HPR-O-glucuronide (4-HPROG), which has been shown to be more effective than the parent molecule in rat mammary tumor models. To investigate whether
4-HPROG was an active agent, the carbon linked analog (4-HPRCG) was synthesized
ii and subsequently found to be a more effective chemopreventative than 4-HPROG or 4-
HPR. In the original synthesis of 4-HPRCG, the route to a key C-glycoside was lengthy and inefficient. In order to investigate 4-HPRCG as a chemotherapeutic, the synthesis was redesigned and significantly improved by access to a key C-benzyl-glucuronide intermediate through employment of a Suzuki coupling reaction between an exoanomeric methylene sugar and an aryl bromide. Subsequently, 4-HPRCG was tested in an animal model and shown to possess effective chemotherapeutic actions.
In vivo, potential cleavage of the amide bond of 4-HPR would liberate retinoic acid, which may explain some of its side effects. This same cleavage may occur with 4-
HPRCG, thus the fully carbon linked analog of 4-HPROG (4-HBRCG) was proposed and synthesized. The synthetic route for 4-HBRCG focused on the production of a key C- benzylbromide-glucuronide obtained through the employment of Suzuki coupling chemistry. Once the benzylbromide was obtained, a key alkylation of a retinal Umpolung derivative yielded the carbon skeleton for 4-HBRCG. Subsequent biological testing will reveal the actions and potency of 4-HBRCG.
iii
Dedicated to Mom, Dad, and Aphayphone
iv
ACKNOWLEDGMENTS
For all the guidance, training, assistance, and answered questions, I would like to
thank my advisor, Dr. Robert W. Curley, Jr. From your instruction, I will take with me patience, critical thinking, ethics, and a little bit of musical talent. Also, I would like to thank the rest of the Medicinal Chemistry faculty for their teaching and experience, including John Fowble and Charles Cottrell. I would also like to show appreciation for
Joan Dandrea and Kathy Brooks for their help with administrative matters.
I am grateful to my classmates, Dave, James, and Tanit, for all your brainstorming
during our coursework. Often times, it seemed I would not have made it through without
your assistance. I would like to thank Young for his lightning speed with fresh chemistry
ideas and on the basketball court. Also, I would like to recognize former and present
students in the department for enriching both my academic and social lives.
I would like to acknowledge my fellow dungeon dwellers, Derek, Kevin, and
Serena, for their help in numerous matters. From physical labor to mental exercises to bouncing off the walls, we managed to sneak in lots of fun. Kevin, thanks for all the sound advice and the standards, “Is it in the literature?” and “What about a model reaction?” I would especially like to recognize Derek for being detail oriented, logical, and always available for assistance in and out of the lab. Also thanks for being the comic
v relief of the group and an associate member of Benzene. Without your solid rhythm section, Benzene would have been especially horrible. I am grateful that we became good friends over the infinite amount of time we were locked up together. Let’s just say,
I’ll never forget your birthday.
Finally, I am eternally grateful for the support given to me from my wife,
Aphayphone. Between the long hours studying, the long hours in the lab, and the occasional basketball game, you probably were tired of wondering, “When is he coming home?” Without your sacrifices, this goal would not have been reached. Also, I would like to thank my Dad for his unwavering encouragement to persevere until the problems were solved. Furthermore, I would like to thank the rest of my family, my wife’s family, and all my friends for asking every month or so, “When are you graduating?”
vi
VITA
May 11, 1972……………………….Born – Muncie, Indiana
1995....……………………………...B.S. Chemistry, University of Nevada, Las Vegas
1994-1997…………………………..Associate Scientist, Lockheed Environmental Systems and Technologies Company Las Vegas, Nevada
1997-2003…………………………..Teaching and Research Assistant, The Ohio State University
PUBLICATIONS
Steinberg, S.; Walker, J. R. Colorimetric Analysis of Benzene for Use in Environmental Screening. Chemosphere 1995, 31, 3771-3781.
Walker, J. R.; Curley, R. W., Jr. Improved Synthesis of (R)-Glycine-d-15N. Tetrahedron 2001, 57, 6695-6701.
Walker, J. R.; Alshafie, G.; Abou-Issa, H.; Curley, R. W., Jr. An Improved Synthesis of the C-Linked Glucuronide of N-(4-Hydroxyphenyl)retinamide. Bioorg. Med. Chem. Lett. 2002, 12, 2447-2450.
vii FIELDS OF STUDY
Major Field: Pharmacy
Specialization: Medicinal Chemistry
viii
TABLE OF CONTENTS
Page Dedication………………………………………………………………………………...iv
Acknowledgments…………………………………………………………………………v
Vita……………………………………………………………………………………....vii
List of Figures…………………………………………………………………………...xiii
List of Schemes……………………………………………………………………….....xvi
List of Spectra………………………………………………………………………….xviii
List of Tables…………………………………………………………………………....xix
List of Abbreviations…………………………………………………………………….xx
Part 1 – Synthesis of N-15 Labeled (R)-Deuterioglycine
Chapter 1 – Stereospecific Synthesis of Doubly Labeled (R)-2H,15N-Glycine
1.1 Introduction……………………………………………………………………...... 2 1.1.1 NMR Theory………………………………………………………………2
ix 1.1.2 Uses of Labeled Amino Acids in Enzymology…………………………....4 1.1.3 Uses of Labeled Amino Acids in Protein NMR…………………………..5 1.1.4 Uses of Labeled Chiral Glycine…………………………………………...8 1.2 Synthesis of Chiral Glycines……………………………………………………..11 1.3 Synthesis of Doubly Labeled Glycine…………………………………………...12 1.4 Conclusions………………………………………………………………………27
Part 2 – Syntheses of Carbon-linked Analogs of Retinoid Glycoside Conjugates
Chapter 2 – Introduction for Retinoids
2.1 Discovery……………………………………………………………………...... 30 2.2 Structures……………………………………………………………………...... 31 2.3 Natural and Synthetic Sources of Vitamin A………………………………….....33 2.4 Absorption, Storage, and Metabolism………………………………………...... 35 2.5 Biological Effects of Retinoids………………………………………………...... 39 2.5.1 Vision………………………………………………………………….....39 2.5.2 Growth and Differentiation……………………………………………....42 2.5.3 Embryonic Development……………………………………………...... 43 2.5.4 Immune System……………………………………………………….....44 2.6 Cellular Mechanisms of Retinoids…………………………………………….....45 2.7 Therapeutic Vitamin A and Retinoids………………………………………...... 48 2.8 Toxicity of Retinoids………………………………………………………….....52 2.9 Synthetic Retinoids……………………………………………………………....53 2.9.1 Receptor Dependent Retinoids………………………………………...... 53 2.9.2 Receptor Independent Retinoids……………………………………...... 56 2.10 N-(4-Hydroxyphenyl)retinamide………………………………………………...58 2.11 N-(4-Hydroxyphenyl)retinamide-O-glucuronide………………………………...60
x Chapter 3 – Synthesis of the C-Linked Analog of 4-HPR-O-glucuronide
3.1 Rationale…………………………………………………………………………64 3.2 C-Glycosides……………………………………………………………………..65 3.3 Previous Studies with 4-HPR-C-glucuronide……………………………………69 3.4 Redesign of the Synthetic Route to 4-HPRCG…………………………………..74 3.4.1 Carbon Monoxide Insertion Route…………………………………...... 74 3.4.2 Suzuki Coupling Route…………………………………………………..79 3.5 Synthesis of 4-HPRCG………………………………………………………...... 79 3.6 Biological Evaluation of 4-HPRCG…………………………………………...... 92 3.7 Conclusions……………………………………………………………………....96
Chapter 4 – Synthesis of the Fully C-Linked Analog of 4-HPR-O-glucuronide
4.1 Rationale……………………………………………………………………...... 97 4.2 Retinoid C-Glycosides……………………………………………………….....100 4.3 Synthesis of 4-HBRCG………………………………………………………....100 4.4 Synthesis of 4-HBRC-glucoside……………………………………………...... 114 4.5 Biological Evaluation of 4-HBRCG…………………………………………....114 4.6 Efforts toward the Synthesis of Retinoyl-β-C-glucuronide…………………….117 4.7 Conclusions……………………………………………………………………..120
Chapter 5 – Summary and Conclusions
5.1 Part 1……………………………………………………………………………122 5.2 Part 2……………………………………………………………………………125
Chapter 6 – Experimental Section
6.1 General Methods………………………………………………………………..131 6.2 Glycine Synthesis…………………………………………………………….....132
xi 6.3 4-HPRCG Synthesis……………………………………………………………145 6.4 4-HBRCG Synthesis…………………………………………………………....154 6.5 Miscellaneous Compounds……………………………………………………..167
Bibliography……………………………………………………………………………175
xii
LIST OF FIGURES
Figure Page
1.1 Prochiral methylene protons of glycine………………………………………...... 9
1.2 15N slices of 15N edited TOSCY experiment. Left panels show 1H-1H crosspeaks from [U-15N]FKBP and right panels show 1H-1H crosspeaks from (R)-[(2-2H), 15N]glycine-FKBP…………………………………………….11
1.3 Hypothetical mechanism for ruthenium tetraoxide oxidative degradation………15
1.4 N-2,2,2-trichloroethyl aryl methylamine carbamates……………………………19
1.5 Theoretical mechanism of the chiral reduction by R-Alpine-Borane®…………..22
1.6 Circular dichroism spectrum of (R)-glycine-d-15N (1.24) (diamonds) and unlabeled glycine (squares) in D2O……………………………………………...26
2.1 Important naturally occurring retinoids………………………………………….32
2.2 Commercial routes to retinyl acetate, (A) La Roche and (B) BASF procedures...34
2.3 Major metabolic routes of retinol………………………………………………..38
2.4 Summary of the visual cycle……………………………………………………..40
xiii 2.5 Overview of the cellular mechanism of retinoic acid. Solid lines indicate known processes and dashed lines indicate postulated transformations…………46
2.6 Retinoids used in the treatment of skin disease and cancer……………………...50
2.7 Receptor dependent retinoic acid analogs………………………………………..55
2.8 Structure of important atypical retinoids………………………………………...57
2.9 Structure of major metabolites of 4-HPR………………………………………..59
2.10 Synthesis of N-(4-hydroxyphenyl)retinamide-O-glucuronide…………………...62
3.1 Structures of 4-HPROG and 4-HPRCG………………………………………….65
3.2 First syntheses of C-glycosides…………………………………………………..66
3.3 General routes to C-glycosides………………………………………...... 67
3.4 C-glycoside analogs of 4-HPROG……………………………………………….71
3.5 Proposed retrosynthetic routes to 4-HPRCG…………………………………….75
3.6 General scheme for organocuprate formation and alkylation…………………..78
3.7 Structure and theoretical mechanism of ester alkenylation reagents…………….81
3.8 General catalytic cycle for the Suzuki-Miyaura cross-coupling…………………83
3.9 Theoretical mechanism of hydroboration with 9-BBN………………………….85
3.10 The catalytic cycle for TEMPO mediated oxidations…………………………....87
xiv
3.11 Effect of retinoid treatment on plasma total triglyceride levels……………….....94
4.1 Structures of 4-HPR and analogs………………………………………………...98
4.2 Retrosyntheses of retinoid C-glycosides………………………………………..101
4.3 Structures of ester byproducts…………………………………………………..112
4.4 Diagram of key long range coupling for 4.17…………………………………..115
xv
LIST OF SCHEMES
Scheme Page
1.1 Previous synthesis of (R)-[(2-2H), 15N]glycine hydrochloride…………………..13
1.2 Synthesis of N-protected aryl amines……………………………………………17
1.3 Synthesis of (S)-p-methoxyphenyl benzyl alcohol-d (1.4)……………………...21
1.4 Final steps in the synthesis of (R)-glycine-d-15N (1.24)………………………....25
3.1 The first synthesis of 4-HPRCG………………………………………………....70
3.2 Synthesis of siloxymethyl-C-glucuronide 3.14…………………………………..76
3.3 Synthesis of MOM-protected enol ether sugar 3.11…………………………...... 80
3.4 Study of the Suzuki coupling reaction conditions……………………………….84
3.5 Synthesis of key intermediate 3.19………………………………………………86
3.6 Synthesis of glucuronide-retinoid conjugate 3.21……………………………….90
3.7 Final steps in the synthesis of 4-HPRCG………………………………………...91
xvi
4.1 Synthesis of C-glucoside methyl ether 4.5……………………………………..102
4.2 Synthesis of the key intermediate 4.9 and possible oxidation by-products…….104
4.3 Synthesis of benzyl bromide 4.12………………………………………………106
4.4 Synthesis of retinal silylcyanohydrins………………………………………….107
4.5 Model alkyation reactions………………………………………………………108
4.6 Key alkylation and final production of 4-HBRCG……………………………..109
4.7 Synthesis of 4-HBRC-glucoside 4.23…………………………………………..116
4.8 Proposed alternative synthetic route towards bromomethyl-glucuronide 3.8….118
4.9 Progress toward the synthesis of retinoyl-β-C-glucuronide……………………119
xvii
LIST OF SPECTRA
Spectra Page
1 1.1 Partial 800 MHz H NMR spectrum of MTPA ester of 1.4 in CDCl3…………...23
13 15 1.2 Partial 100 MHz C NMR spectrum of (R)-glycine-d- N (1.24) in D2O………25
1.3 Partial 800 MHz 1H NMR spectrum of the (R)-glycine-d-15N camphanate derivative 1.25 in CDCl3………………………………………………………....27
1 3.1 400 MHz H NMR spectra of lactone 3.15 and enol 3.11 in DMK-d6…………..82
3.2 Difference spectra from steady state NOE experiments on C-glucuronide 3.19 in DMK-d6 at 400 MHz…...... 89
1 3.3 400 MHz H NMR spectrum of 4-HPRCG in DMK-d6…………………………92
1 4.1 400 MHz H NMR spectrum of glucuronide bromide 4.12 in DMK-d6……….106
1 4.2 400 MHz H NMR spectrum of final product 4-HBRCG in MeOH-d4………..111
4.3 400 MHz 1H NMR specta of protected 4-HBRCG 4.17 and ester 4.18 in DMK-d6……………………………………………………………………...113
4.4 Partial HMBC plot of protected 4-HBRCG 4.17 in DMK-d6 at 400 MHz……..115
xviii
LIST OF TABLES
Table Page
1.1 Ruthenium oxidation studies of aromatic substrates…………………………….18
1.2 Acidic ruthenium oxidation studies of aromatic substrates……………………...20
3.1 Effect of retinoid treatment on DMBA-induced rat mammary tumor volume…..95
3.2 Effect of retinoid treatment on liver weights…………………………………….95
xix
LIST OF ABBREVIATIONS
Abbreviation Term °C degrees in Celsius 2D 2-dimensional 4-HBR (4-hydroxybenzyl)retinone 4-HBRCG (4-hydroxybenzyl)retinone-C-glucuronide 4-HPR N-(4-hydroxyphenyl)retinamide (4-hydroxyphenyl)retinamide-C- 4-HPRCG glucuronide 4-MePR N-(4-methoxyphenyl)retinamide 9-BBN 9-borabicyclo[3.1.1]nonane A adenine Ac acetyl BOC butoxycarbamate Bu butyl c concentration (g/100 mL) C cytosine cGMP 3’,5’-cyclic guanosine monophosphate chicken ovalbumin upstream promoter COUP-TF transcription factor CRABP cellular retinoic acid binding protein CRALBP cellular retinal binding protein CRBP cellular retinol binding protein cyp cytochrome P-450 de diasteriomeric excess
xx DEAD diethyl azodicarboxylate DMAP 4-(dimethylamino)pyridine DMBA 7,12-dimethylbenz[α]anthracene DMF N,N-dimethylformamide
DMK-d6 deuterated acetone DNA deoxyribonucleic acid dppf 1-1’-bis(diphenylphosphino)-ferrocene ee enantiomeric excess eq equivalents ES electrospray Et ethyl FKBP FK-506 binding protein g gram G guanine GMP guanosine 5’-monophosphate GTP guanosine 5’-triphosphate h hour HMBC heteronuclear multiple-bond correlation HMDS bis(trimethylsilyl)amide HMPA hexamethylphosphoramide HPLC high performance liquid chromatography HRE hormone response element HRMS high resolution mass spectroscopy HSQC heteronuclear single quantum coherence i iso IR infra-red IRBP interphotoreceptor retinoid binding protein IU international units kDa kilodalton L liter LDA lithium diisopropylamide
xxi LRMS low resolution mass spectroscopy m meter M molar Me methyl
MeOH-d4 deuterated methanol MHz megahertz min minute mol mole MOM methoxymethyl mp melting point MTPA methoxy-α-trichloromethylphenylacetyl N normality NADH nicotinamide adenine dinucleotide nicotinamide adenine dinucleotide NADPH phosphate NMR nuclear magnetic resonance NOESY nuclear Overhauser effect spectroscopy o ortho p para Ph phenyl PPAR peroxisomal proliferator activating receptor ppm parts per million Pr propyl psi pounds per square inch pyr pyridine RAR retinoic acid receptor RBP retinol binding protein RDA recommended daily allowance RPE retinal pigment epithelium rt room temperature RXR retinoid X receptor
xxii SAR structure activity relationship T thymine t tert TBAF tetrabutylammonium fluoride TBDMSCN tert-butyldimethylsilyl cyanide 2,2,6,6-tetramethyl-1-piperidinyloxy free TEMPO radical TFAA trifluoroacetic acid THF tetrahydrofuran TLC thin layer chromatography TMSCN trimethylsilyl cyanide TMSI trimethylsilyl iodide TOC tracheal organ culture TOCSY total correlation spectroscopy
tR retention time TR thyroid hormone receptor UDP uridine diphosphate UV ultraviolet VAD vitamin A deficient VDR vitamin D receptor
xxiii
PART 1 – SYNTHESIS OF N-15 LABELED (R)-DEUTERIOGLYCINE
1
CHAPTER 1
STEREOSPECIFIC SYNTHESIS OF DOUBLY LABELED (R)-2H, 15N-GLYCINE
1.1 INTRODUCTION
1.1.1 NMR THEORY
Isotopes are defined as two or more forms of an element that have the same number of protons but a different number of neutrons, which gives rise to a different atomic weight. Some of these isotopes are unstable and subsequently decay by liberating radioactive energy. Three main types of decay are alpha, beta, and gamma, which are a helium nucleus, an electron or antielectron, and a photon, respectively. Conversely, there are many stable isotopes, which have many uses in a variety of sciences. One of these areas is their application towards nuclear magnetic resonance (NMR).
In the basic structure of an atom, electrons orbit the nucleus, which contains neutrons and protons. The protons in the nucleus spin on their axes and this spinning creates a magnetic dipole that is perpendicular to the plane of the spin. The magnitude of
the dipole is defined as the nuclear magnetic moment. The angular momentum of these
spinning protons are classified into spin quantum numbers (I), e.g. 0, 1/2, 1, 3/2, etc. The
relationship between atomic number and atomic mass determines the quantum spin
number. If the atomic mass is an odd number and the atomic number is odd or even, then 2 I is a half integer. If the atomic mass is even and the atomic number is odd or even, then
I is an integer or zero, respectively. NMR active nuclei (e.g. 1H, 13C, 15N) typically have
spin numbers of 1/2, whereas nuclei with spin numbers of zero or whole numbers usually
are insensitive or silent in NMR spectroscopy.
In the absence of an external magnetic field, the spinning dipoles are randomly
oriented. However when they are placed into a magnetic field, the direction of the
dipoles are organized into two possible spin states, for example, +1/2 (α state) and -1/2 (β
state). The energy difference between these two states is proportional to the strength of
the applied magnetic field. The observation of an absorbance in NMR is greatly
dependent on this energy difference and therefore, the stronger magnet will result in a
greater sensitivity. According to Boltzmann’s distribution, there will be a slight excess of protons in the lower energy state (α state), which is aligned with the magnetic field.
Once the two populations are established, transitions between energy spin states are possible by the application of an external energy in the form of radiofrequency radiation.
The introduced radiofrequency pulse needed for the spin flip is defined in megahertz
(MHz). The excitation, relaxation, and detection of the proton spin in a magnetic field is the basis of NMR spectroscopy.1
At equilibrium, the total population of proton dipoles can be represented as a
single vector that is aligned in the Z-axis with the direction of the applied magnetic field.
To detect a signal, an applied electromagnetic pulse changes the direction of the vector
by 90 ° into the XY plane after which the dipole magnetization relaxes to the direction of
the magnetic field. During relaxation, the magnitude of the representative vector in the
XY plane is measured over time as it decays back to equilibrium. Time and amplitude 3 measurements are plotted to show the free induction decay. To generate the traditional
NMR spectrum, the time dimension is Fourier transformed to show the frequency versus amplitude plot, which can then be interpreted for structural information.
After pulse excitation, the rate of relaxation of the nuclei is dependent on its environment and these different environments can either prolong or expedite the relaxation process. Differences in the relaxation of the nuclei give rise to different frequencies. In any particular molecule, neighboring nuclei can either be close in space or close through a covalent bond, which gives rise to dipole-dipole cross relaxation and scalar relaxation, respectively. Scalar relaxation results in scalar coupling constants and this information can be used to determine bond connectivities. Dipole-dipole cross relaxation is the major mode of relaxation and this information can be used to interpret the three dimensional conformation of the molecule. Therefore, the measurement of excitation and relaxation of spin 1/2 nuclei is the basis of structure determination by
NMR spectroscopy.
1.1.2 USES OF LABELED AMINO ACIDS IN ENZYMOLOGY
The fundamental building blocks of all proteins in living organisms are amino acids. Therefore, the primary and secondary metabolism of these essential molecules has always been of interest. The main methodology for studying their metabolism is through the tracking of isotopes. Due to the scarce natural abundances of important NMR active nuclei (13C and 15N), the synthesis of labeled amino acids or precursors becomes essential in order to enhance the signal obtained in NMR. Conversely, the elimination of a signal from the NMR spectra becomes a valuable technique, for example, with the substitution 4 of deuterium for hydrogen. Since this substitution can be done stereospecifically,
interpretations of enzyme mechanisms can be assessed pertaining to processes involving
carbon-hydrogen bonds. Experimentally, synthetic isotopically labeled amino acids or
other labeled precursors are given to the organism of interest after which the structure of
the natural product can be analyzed by NMR. Subsequently, retro-biosynthetic analysis
is performed by tracking the supplemented isotopes.
The metabolism of amino acids has been extensively studied with the use of 13C,
15N, and 2H isotopes.2,3 Labeled precursors have been used to elucidate the biosynthesis of all amino acids, which can be divided into two main pathways. The aromatic amino acids tend to derive from the shikimate pathway while the aliphatic amino acids derive from carbohydrate metabolism and the citric acid cycle. Secondary metabolisms of amino acids have been elucidated with specifically labeled amino acids. For example, many alkaloids are derived from amino acids like lysine (atropine, sparteine, and nicotine), phenylalanine and tyrosine (ephedrine, mescaline, morphine, and berberine), and tryptophan (serotonin, strychnine, lysergic acid, vinca alkaloids, and ergot alkaloids).
Also, valine and serine are used in the biosynthesis of various β-lactam antibiotics.
Through the biosynthesis of the amino acid or the use of the amino acid in a natural product synthesis, isotope tracking can be used to deduce enzyme mechanisms or determine biosynthetic pathways.
1.1.3 USES OF LABELED AMINO ACIDS IN PROTEIN NMR
Determination of protein structure is an essential part of biology and medicine.
The traditional method of solving protein structures is through X-ray crystallography, in 5 which the protein is crystallized and then analyzed for its diffraction of X-ray radiation.
An alternative way to solve protein structures is through NMR. Since the first complete solution structure of a protein,4 substantial advances in technology have brought NMR to the forefront of structural biology. These advances have combined to enable protein
NMR structures to be solved at atomic resolution, which previously was reserved for crystallography. The main advantage of using protein NMR is that the protein is in solution instead of a solid crystal. This difference is important because, in solution, it is possible for proteins to fold and flex to their natural conformation. Furthermore, dynamic processes such as chelation, binding, folding, and catalysis may be monitored by
NMR. These reasons make protein NMR a powerful tool for structural biology and drug discovery.5,6
The use of isotopically labeled amino acids is essential to obtaining a solution
NMR structure. Not only do spin 1/2 nuclei enhance the signal generated in the NMR but they also enable multidimensional, isotope-edited experiments. Experimentally, labeled proteins used in NMR are obtained in ample quantities by recombinant DNA technologies. The gene for the protein of interest is spliced into the plasmid of a microorganism, after which the protein is isolated. For the organism to biosynthesize
13 15 2 labeled amino acids, C-glucose, NH4Cl, and H2O are used in the growth medium with various combinations to obtain the desired labeling pattern. With differently labeled
13C-glucose isotopomers, the labeling pattern of the carbons on each amino acid residue can be varied. Also groups of amino acids like leucine, isoleucine, valine, alanine, and phenylalanine can be specifically labeled with 13C-labeled precursors like ketobutyrate, ketoisovalerate, glycerol, and pyruvate.7 Uniform deuterium labeling of non- 6 2 exchangeable protons is done with H2O in the growth medium and site specific
deuteration is done with labeled pyruvates and other precursors to label methyl groups on
valine, leucine, isoleucine, and alanine.8-10
Once the proteins are labeled in the desired pattern, a gambit of multidimensional
NMR experiments are performed to make resonance assignments for the protons,
carbons, and nitrogens. The use of six triple resonance experiments usually leads to
complete and unambiguous resonance assignments of most proteins. These experiments
are HNCA,11,12 HN(CO)CA,11,13 HN(CA)CB,13,14 HN(CACO)CB,13,15 HNCO,13,15 and
HN(CA)CO16,17 which show the correlations of amide proton to the α carbon, to the
neighboring α carbon, to the α and β carbons, to the α and β carbon of the neighboring
residue, to the neighboring carbonyl carbon, and to the carbonyl carbon of each residue,
respectively. The data generated enables the protein backbone chemical shifts to be
assigned and this eventually leads to the determination of the secondary structure through
a series of predictive charts based on the shifts and coupling constants of the backbone
atoms. The dihedral angles can be used to predict secondary structure motifs such as α-
helices and β-sheets. If the Cβ and Cα resonances are assigned from the above
experiments, then side chain assignments can be done using 13C-edited total correlation spectroscopy (TOCSY) experiments. Finally, once all the assignments are made, tertiary structure refinement begins. Experiments like 13C- and 15N-NOESY-HSQC, which are
isotope edited nuclear Overhauser effect spectroscopy coupled with heteronuclear single
quantum coherence, provides distance constraints, which are input into iterative modeling
programs, to finally yield a total protein structure.5
7 In some cases, the above routine is not enough to unambiguously assign each resonance of all amino acid residues. To address these problems, specially prepared single amino acids can be incorporated into a target protein. Specifically labeled amino acids can be synthesized and used in the production of a recombinant protein. However to avoid scrambling of the label in the amino acid, particular auxotrophs are needed to produce the protein of interest. Since amino acids are part of the regular metabolism in an organism, auxotrophs are specific strains which lack the biochemical pathway to process the amino acid of interest. This aspect is particularly important for labeled amino acids with stereospecifically incorporated deuterium. With the particular auxotroph, the specifically labeled amino acid is supplemented into the growth medium, thereby producing a protein product with a specifically labeled residue. Subsequent NMR experiments will easily allow for resonance assignments and other structural information.
1.1.4 USES OF LABELED CHIRAL GLYCINE
Glycine is an amino acid that plays an important role in defining protein structure and is involved in many metabolic processes. It is a unique amino acid because it lacks chirality and a side chain. In protein NMR, resonance assignments of glycine prochiral methylene protons are particularly useful because it is important in tertiary defining structure. From the above methods, backbone resonance assignments are used to predict structural motifs, however, they do not predict the tertiary structure of turn regions that are located between structural motifs. Glycines usually occur in turn regions because of their lack of hydrophobicity and steric bulk. The stereospecific assignments of the
3 glycine pro-R and pro-S protons can be used to determine the coupling constant ( JNH,Hα) 8 with the amide bond proton (Figure 1.1). This value is then used to estimate the dihedral
angle between the amide proton and the methylene protons using the Karplus
equation.18,19 Determination of this angle will help predict the conformation of the backbone of the protein, especially in turn regions where there is typically ambiguity.
O H HS HR HS HR HO N NH2 N O O H R
glycine peptide
Figure 1.1 Prochiral methylene protons of glycine.
(R)-[2-2H]glycine has been used in NMR studies of the cyclic peptide enkephalin.20 The pentapeptide, L-tyrosine-D-penicillamine-glycine-L-phenylalanine-D- penicillamine, where penicillamine is β,β-dimethylcysteine, is a potent selective ligand
for the δ opioid receptor. In order to determine the solution conformation of this flexible
cyclic peptide, NMR studies were done with a labeled enkephalin analog containing the
labeled glycine and also stereoselectively β-deuterated phenylalanine and tyrosine. After
stereospecific resonance assignments were made, NOESY experiments were done to
obtain distance constraints between the nuclei. These constraints were used to perform
energy minimizations and finally propose a solution structure.
Even more useful than chirally deuterated glycine is doubly labeled chiral glycine.
Doubly labeled (R)-[(2-2H), 15N]glycine was made in our lab and used in protein NMR
studies of the FK-506 binding protein (FKBP).21 Once incorporated into a protein, the
15N nucleus enables for 15N edited NMR experiments, which greatly simplify data 9 interpretation. FKBP is a protein involved in the immune response and FK-506 is an
immunosuppressant natural product. The labeled FKBP was produced with recombinant
technology and two molecules, uniformly 15N labeled FKBP and doubly labeled R-2H,
15N-glycine FKBP, were prepared for the NMR studies. With the 15N nucleus
incorporated into the proteins, 15N resolved TOCSY experiments were done to correlate
proton spin systems. With glycine, the spin system only consists of the amide proton and
the prochiral methylene protons. From the TOCSY experiments, 15N slices of the
uniformly 15N labeled FKBP and the (R)-[(2-2H), 15N]glycine FKBP were compared to
stereospecifically assign the prochiral methylene protons. From Figure 1.2, the doubly
protonated glycine slices show two crosspeaks while the deuterated slices have one
crosspeak. Therefore from only one experiment, it is possible to make stereospecific
resonance assignments of the glycine methylene protons. This information was used to
3 determine the JNH,Hα values, dihedral angles, and more distance constraints to further
refine the generated solution structure of FKBP.
Although chiral glycine can be useful in protein and peptide NMR, it has mostly
been used to elucidate enzyme mechanisms and biochemical pathways. Serine
hydroxymethyltransferase (EC 2.1.2.1) reversibly catalyzes the transfer of the β-
methylene group of serine to tetrahydrofolate, which yields glycine and 5,10-
methylenetetrahydrofolate. This pyridoxal phosphate mediated enzyme process has been
extensively studied with variously labeled glycines, serines, and tetrahydrofolates to
determine that the stereochemical course of the methylene transfers is partially
stereoselective.2,22-26 Glycine can be further metabolized in the glycine cleavage pathway, which yields carbon dioxide, ammonia, and 5,10-methylenetetrahydrofolate. 10
Figure 1.2 15N slices of 15N edited TOSCY experiment. Left panels show 1H-1H crosspeaks from [U-15N]FKBP and right panels show 1H-1H crosspeaks from (R)-[(2-2H), 15N]glycine-FKBP (taken from ref. 12).
This system has been studied with chirally deuterated glycine to show the transfer to the
folate is only partially stereoselective.27 Also, methylene bridges in heme are derived
from glycine units and it was shown that preservation of one prochiral proton is
stereospecific.28 Due to the important applications of labeled glycine, the development of
efficient syntheses of these isotopomers also becomes important.
1.2 SYNTHESIS OF CHIRAL GLYCINES
There are several methods for the synthesis of singly labeled, stereospecifically
deuterated glycine. Different amino acids such as serine,29 phenylalanine,30 and
11 glutamate31 have been used to serve as chiral templates for the synthesis of labeled glycine. Other molecules in the chiral pool such as D-ribose32, D-glucose33, and (-)-8-
phenylmenthol34 have also been used to induce stereochemistry in the production of
chiral glycine. It has also been shown that cobalt complexes of amino acids can be used
to synthesize chiral glycine and other labeled amino acids.35,36 The majority of reported
methods for chiral glycine include a combination of chemical and enzymatic
transformations to incorporate deuterium and chirality, respectively. Several methods
start with a deuterated benzaldehyde, which is reduced with various dehydrogenases, to
give a chiral benzyl alcohol. The alcohol is functionalized to the amine and the aromatic
ring is oxidized by ozonolysis to the carboxylic acid, eventually yielding the chiral glycine.37-39 A similar approach used furfuryl aldehyde as a starting material.40 Also, an
2 enzymatic decarboxylation of glutamate in H2O with glutamate decarboxylase afforded chirally deuterated aminobutanoic acids, which were eventually converted to labeled glycine.41
1.3 SYNTHESIS OF DOUBLY LABELED GLYCINE
Although there are many routes to deuterated glycine, our lab has published syntheses of the more useful doubly labeled glycine.21,42 The synthetic route (Scheme
1.1) for the glycine used in the stereospecific assignments of FK-506 binding protein was
adapted from procedures reported by Woodard and others.43 Like other classical
methods, the starting material is the benzaldehyde 1.1. Conversion to the Umpolung
derivative enables proton-deuterium exchange, yielding 1.2. Instead of an enzymatic
reduction, chiral reducing agent R-Alpine-Borane® was used to reduce the aldehyde 1.3, 12 O
O O 1) morpholine, N HClO , NaCN 2N HCl 4 D H CN D 2) NaH, D2O H3CO H3CO H3CO 1.1 70 %, >98 % D 1.2 88 % 1.3
O D H H D 15 R-Alpine-Borane [ N]- phthalimide OH N 15 Ph3P, DEAD, THF H CO H3CO 3 O 95 % 1.4 49 % 1.5
D H O 1) NaBH4, CH3COOH RuCl3 ⋅ H2O, NaIO4 N O 15 CH CN, CCl , H O 2) ((CH3)3COCO)2O H 3 4 2 H3CO 86 % 1.6 45 %
D O H 1) TFAA, CH2Cl2 DH HO 15 N O 15 2) DOWEX 50W-X8 HOOC NH3Cl O H
1.7 92 %, 80 %ee 1.8
Scheme 1.1 Previous synthesis of (R)-[(2-2H), 15N]glycine hydrochloride.
13 producing the chiral alcohol 1.4. This was advantageous because chemically induced
chirality can be done on a much larger scale than enzymatic. The 15N nucleus was
introduced through a Mitsunobu reaction with labeled phthalimide to afford the conjugate
1.5. The amine protecting group was exchanged to the tert-butoxy-carbamate (t-BOC)
yielding 1.6. The next step was to oxidatively degrade the aromatic ring to a carboxylic
acid with catalytic ruthenium tetraoxide to give N-t-BOC glycine 1.7. Although the
reported yield was 45 %, this key, penultimate step proceeded sporadically, was
problematic, and on average, very low yielding. Finally the acid labile amine protecting
group was removed and (R)-[(2-2H), 15N]glycine hydrochloride 1.8 was isolated with ion exchange chromatography.
Due to an interest in applying doubly labeled glycine to other protein structural problems, additional glycine was required. In order to improve the ruthenium tetraoxide reaction, the above synthesis was repeated to obtain the precursor to the oxidation reaction, the protected benzylamine 1.6. The reaction was then studied mimicking conditions used in the past.21,43,44 In summary, the ruthenium source was ruthenium
trichloride hydrate (2.2 mol %), the stoichiometric re-oxidant was sodium periodate (18
equivalents), and the solvent mixture used was CH3CN/CCl4/H2O (2:2:3). These conditions were reported to give good yields but in our hands, the doubly labeled glycine carbamate 1.7 was never obtained in greater that 10 % yield due to extensive benzylic oxidation and production of complex mixtures of other oxidized products. Therefore, investigations toward improving the oxidation reaction were initiated.
The oxidation of organic compounds with ruthenium tetraoxide (RuO4) to give a
carboxylic acid functionality is commonly employed.45 The oxidation reaction is usually 14 done in biphasic conditions (Figure 1.3). The catalytic ruthenium salt is oxidized by the
stoichiometric re-oxidant in the aqueous layer to give the active oxidant RuO4, which can enter the organic phase. The double bonds on the aromatic ring are dihydroxylated and then undergo further oxidations until bond cleavage occurs, releasing carbon dioxide.
The ruthenium dioxide enters back into the aqueous phase where ruthenium tetraoxide is regenerated.
Some common aromatic systems that serve as carboxylic acid precursors are furan and unsubstituted phenyl rings. In our case, the furan ring is not useful due to its instability in the strongly acidic conditions used in the formation of the morpholinonitrile compounds. A phenyl ring possessing a benzylic heteroatom is also not desirable
RuX + IO4-
RuO2 RuO4 Aqueous
Organic
RuO2 RuO4
R R HOOC R OH OH
CO2
Further oxidations
Figure 1.3 Hypothetical mechanism for ruthenium tetraoxide oxidative degradation.
15 because of its susceptibility to benzylic oxidation.44,46 Since the assessment of the para- methoxyphenyl ring showed it was unfavorable, other types of aromatic systems, namely
1- and 2-naphthyl rings, were investigated. These aromatic rings were claimed to be superior to other systems in the ruthenium oxidation reaction.47
In order to investigate the behavior of 1- and 2-substituted naphthyl rings in the
ruthenium oxidation, unlabeled N-protected naphthyl amines were synthesized (Scheme
1.2). From the respective aldehydes 1.9 and 1.12, reduction was followed by a
Mitsunobu reaction with phthalimide to give the conjugates 1.10 and 1.13. Amine
protecting group manipulations gave the precursors, 1.11 and 1.14, for the oxidation
reactions. Also, unlabeled analog 1.15 was prepared to use as a comparison.
Various oxidation conditions were used to determine the optimal ring system and
reaction environment (Table 1.1). Three catalytic ruthenium reagents were investigated
at different temperatures using sodium periodate as the ruthenium re-oxidant; cis-
[Ru(bpy)2Cl]·2H2O (bpy = 2,2’-bipyridine), RuO2·2H2O, and RuCl3·2H2O. The bipyridine ruthenium complex is a thermally stable ruthenium, which enabled use of refluxing reaction conditions.48 This is pertinent because ruthenium tetraoxide sublimes at warmer temperatures. Ruthenium dioxide hydrate was claimed to be superior to the chloride form for the oxidation of aromatic rings to carboxylic acids.49 These reagents
were compared against the original form used, ruthenium chloride hydrate. The aryl
amines were submitted to the various oxidation conditions and the reactions were semi-
quantitatively monitored. The crude products were weighed and examined by 1H NMR
spectroscopy for formation of N-t-butoxy-glycine (1.16) and for the complexity of
16 O H N O OH N 1) NaBH4 1) NaBH4, AcOH O O 2) Ph3P, DEAD, 2) ((CH3)3COCO)2O phthalimide
1.9 44 % 1.10 83 % 1.11
O O 1) NaBH4 H N 2) Ph3P, DEAD, phthalimide O
1.12 55 % 1.13
O O 1) NaBH4, AcOH N O N O 2) ((CH3)3COCO)2O H H H3CO
80 % 1.14 1.15
Scheme 1.2 Synthesis of N-protected aryl amines.
product mixture. The plus and minus signs indicate increasing and decreasing amount of
1.16, respectively, and also decreasing and increasing byproduct formation, respectively.
Entries 1-4 show experiments with the bipyridine ruthenium complex at different temperatures and it was found to be generally unfavorable for any of the tested arylamines. Entries 5-8 were experiments using ruthenium chloride at different temperatures. It gave variable results but the p-methoxyphenyl ring was slightly more favorable than the 2-naphthyl ring. Entries 9-12 were experiments with ruthenium dioxide and it showed similar results as ruthenium chloride except the 2-naphthyl ring
17 O O RuO4 HO Ar N O N O H O H
arylamine 1.16
Entry Compound Oxidation conditions Results
cis-[Ru(bpy) Cl]·2H O, NaIO , 1 1.11 2 2 4 No reaction CH3CN, H2O, 72 h, 35-40°C
cis-[Ru(bpy) Cl]·2H O, NaIO , 2 1.11 2 2 4 - - - CH3CN, H2O, 46 h, 80°C
cis-[Ru(bpy) Cl]·2H O, NaIO , 3 1.14 2 2 4 - CH3CN, H2O, 46 h, 80°C
cis-[Ru(bpy) Cl]·2H O, NaIO , 4 1.15 2 2 4 CH3CN, H2O, 46 h, 80°C +
RuCl ·H O, NaIO , CH CN, 5 1.11 3 2 4 3 - - CCl4, H2O, 72 h, rt
RuCl ·H O, NaIO , CH CN, 6 1.14 3 2 4 3 + CCl4, H2O, 72 h, rt
RuCl ·H O, NaIO , CH CN, 7 1.14 3 2 4 3 + CCl4, H2O, 68 h, 40-45 °C
RuCl ·H O, NaIO , CH CN, 8 1.15 3 2 4 3 + + CCl4, H2O, 68 h, 40-45 °C
RuO ·H O, NaIO , NaHCO , 9 1.11 2 2 4 3 - CH3CN, CCl4, H2O, 72 h, rt
RuO ·H O, NaIO , NaHCO , 10 1.14 2 2 4 3 + CH3CN, CCl4, H2O, 72 h, rt
RuO2·H2O, NaIO4, NaHCO3, 11 1.14 + + CH3CN, CCl4, H2O, 68 h, 40-45 °C
RuO ·H O, NaIO , NaHCO , 12 1.15 2 2 4 3 + CH3CN, CCl4, H2O, 68 h, 40-45 °C
Table 1.1 Ruthenium oxidation studies of aromatic substrates.
18 was slightly more favorable. In general, the 1-naphthyl group was found to behave
poorly in the oxidation reactions. Although the 2-naphthyl and p-methoxyphenyl rings showed some promise, none of the explored conditions gave the desired improvement to make this a generally applicable method for the production of doubly labeled glycine.
Another published ruthenium oxidation uses ruthenium chloride and periodic acid instead of sodium periodate as the stoichiometric re-oxidant.50 This method claimed to
give the greatest yields and shortest reaction times for the oxidative degradation of
aromatic rings. These acidic conditions were not initially explored due to the acid labile
tert-butoxy carbamate amine protecting group. Therefore, in order to use these
conditions, the acid stable 2,2,2-trichloroethyl carbamate protecting group was
employed.51 Since the 2-naphthyl and p-methoxyphenyl rings showed the most promise
from the oxidation study, these rings were further explored. With similar chemistry,
unlabeled N-(2-naphthylmethylamine)- and N-(p-methoxyphenylmethylamine)-2,2,2-
trichloroethyl carbamates 1.17 and 1.18, respectively, were synthesized and subjected to
the acidic oxidation conditions to determine the optimal aromatic ring (Figure 1.4).
O O Cl Cl N O N O Cl Cl H Cl H Cl H3CO 1.17 1.18
Figure 1.4 N-2,2,2-trichloroethyl aryl methylamine carbamates.
19 Table 1.2 shows the semi-quantitative results of the ruthenium oxidation using
H5IO6 as the re-oxidant, where the reactions were monitored for the formation of N-2,2,2- trichloroethylglycine carbamate (1.19). Entry 1 shows the control experiment with tert- butoxy analog 1.15, which resulted in carbamate deprotection and the oxidation of the benzylic position, giving the arylamide. Entry 2 shows the 2-naphthyl analog 1.17, which yielded a typical complex mixture as seen with the sodium periodate conditions.
Entry 3 shows the results with the p-methoxyphenyl analog 1.18 and it yielded product
1.19 and very few by-products. Using the p-methoxyphenyl ring, the acidic condition was a substantial improvement over the neutral, sodium periodate conditions. Therefore, the H5IO6 oxidation conditions were used in the subsequent labeled synthesis.
O O RuO4 Cl HO Cl Ar N O N O Cl Cl H Cl O H Cl
arylamine 1.19
Entry Compound Oxidation conditions Results
RuCl ·H O, H IO , 1 1.15 3 2 5 4 - - CH3CN, CCl4, H2O, 2 h, rt
RuCl ·H O, H IO , 2 1.17 3 2 5 4 - CH3CN, CCl4, H2O, 2 h, rt
RuCl ·H O, H IO , 3 1.18 3 2 5 4 CH3CN, CCl4, H2O, 2 h, rt + + + +
O
Ar NH2
arylamide
Table 1.2 Acidic ruthenium oxidation studies of aromatic substrates. 20 Due to the apparent success of the unlabeled p-methoxyphenyl ring in the periodic
acid oxidation conditions, the synthesis of labeled glycine was repeated. Starting with p-
anisaldehyde (1.1), derivatization to the Umpolung reagent 1.20 enabled for proton-
deuteron exchange, giving the deuterated morpholinonitrile 1.2 in good yield and with
>97 % 2H incorporation as determined by 1H NMR integration. These first three reactions producing the labeled morpholinonitrile were very amenable to large scale and were performed starting with 100 g of 1.1. Following hydrolysis, stereochemistry is introduced by the chiral reduction of 1.3 using R-Alpine-Borane®, which produced the
(S)-alcohol 1.4 (Scheme 1.3). S-Alpine-Borane® will produce the enantiomeric R-alcohol of 1.4.
R-Alpine-Borane® is a trade name for B-isopinocamphenyl-9-borabicyclo[3.3.1]-
nonane (Figure 1.5), which is derived from the hydroboration of (+)-α-pinene with 9-
O
O N 1) morpholine,HClO4 NaH, D2O H CN 2) KCN H3CO H3CO 1.1 96 % 1.20 93 %, >97 % D
O
N O H D R-Alpine-Borane CN 2N HCl D OH D
H3CO H3CO H3CO 1.2 95 % 1.3 90 % 1.4
Scheme 1.3 Synthesis of (S)-p-methoxyphenyl benzyl alcohol-d (1.4).
21 BBN. Midland and others demonstrated that this chiral reducing agent was highly
stereoselective and gave good yields, accepting a wide range of aldehyde substrates.52
Concerning the mechanism, it is hypothesized that initial coordination of the boron and oxygen atoms positions the aldehyde with its alkyl group away from the pinane methyl group.53 After hydride addition, α-pinene and the boronated chiral alcohol are released,
and the organoborane is converted to the free alcohol upon workup.
O H B RD B H O H H CH3 CH3 D R R-Alpine-Borane®
H D H D + ROB ROH
α-pinene chiral alcohol
Figure 1.5 Theoretical mechanism of the chiral reduction by R-Alpine-Borane®.
Once the crude alcohol was obtained, purification of the labeled p-methoxybenzyl
alcohol 1.4 was tedious due to the large amount of pinene and pinene by-products produced in the reduction. The amount of contamination was greatly reduced by partitioning the crude product between aqueous methanol and octane. By sequentially washing the methanol layer with octane, the benzyl alcohol 1.4 could be substantially 22 decontaminated. Furthermore, the product could be purified by vacuum distillation to
give the clean chiral alcohol 1.4. Determination of the optical purity was first
investigated by derivatization with camphanic acid and also with the use of chiral shift
reagents such as Eu(hfc)3, neither method was successful. However, analysis of
enantiomeric purity of the alcohol-d 1.4 was made possible by derivatization to Mosher’s
ester,54,55 which gave the methoxy-α-trifluoromethylphenylacetyl (MTPA) ester 1.21 and
showed 96 % de by 1H NMR analysis of the benzylic methylene proton region (Spectrum
1.1).
Major(S) -derivative
Doubly protonated
Minor(R ) -derivative
5.35 5.30 5.25 5.20 5.15 ppm
H D O O CF H3CO 3 H3CO 1.21
1 Spectrum 1.1 Partial 800 MHz H NMR spectrum of MTPA ester of 1.4 in CDCl3. 23
Incorporation of the 15N nucleus was accomplised with 15N phthalimide, as done
previously, to give conjugate 1.5 in modest yield (Scheme 1.4). The phthalimide
protecting group was removed and replaced with the 2,2,2-trichloroethyl carbamate,
yielding the oxidation precursor 1.22. Next, the acidic ruthenium oxidative degradation
was performed to give the N-protected glycine 1.23 in good yield after chromatography.
Investigation of the optical purity of the penultimate material 1.23 was done by derivatization to the (-)-8-phenylmenthyl ester. Numerous 1H NMR spectra were taken
in various solvents and at different field strengths to fully resolve the glycinate methylene
proton resonances without success. However, 2H NMR at 123 MHz showed fairly
resolved deuterium signals, giving an estimated 90 % de of the derivatized material.
The cleavage of the trichloroethyl carbamate was performed with zinc dust in
acetic acid and water.56 Purification of the final glycine 1.24 was very challenging.
Various ion exchange resins with various elution systems, reverse phase chromatography,
crystallization, and derivatization were exhaustingly investigated without success to
efficiently purify the final glycinate product. Eventually, the crude product was placed
on anion exhange resin (hydroxide form) and then eluted with aqueous HCl. Next, the
concentrated eluent was placed on cation exchange resin (proton form) and then eluted
with dilute NH4OH. This tandem anion-cation exchange method gave clean (R)-glycine-
d-15N free base (1.24) in good yield.
Spectroscopic and physical characteristics confirmed the production of the target compound. The 13C NMR spectrum shows that the methylene carbon is a triplet of
doublets due to the one bond 15N-13C and 2H-13C coupling (Spectrum 1.2). The doublet 24 O H D D H [15N]- phthalimide OH N 15 Ph3P, DEAD, THF H CO H3CO 3 O
1.4 60 % 1.5
D H O 1) NaBH , AcOH RuCl ⋅H O, H IO 4 Cl 3 2 5 6 N O 15 Cl CH CN, CCl , H O 2) 2,2,2-trichloroethyl H Cl 3 4 2 chloroformate H3CO
88 % 1.22 62 %
D O H 1) Zn, AcOH, H2O DH HO Cl N O HOOC NH2 15 Cl 2) AG 1, HCl 15 O H Cl 3) AG 50W, NH OH 4 1.23 73 %, 82 %ee 1.24
Scheme 1.4 Final steps in the synthesis of (R)-glycine-d-15N (1.24).
40.5 41.0 41.5 42.0 42.5 ppm
13 15 Spectrum 1.2 Partial 100 MHz C NMR spectrum of (R)-glycine-d- N (1.24) in D2O. 25 arises from the coupling with 15N (I = 1/2) and the triplet from 2H (I = 1). Chirality was
demonstrated by the analysis of 1.24 with circular dichroism spectropolarimetry, which
gave a maximum absorbance of 208 nm (Figure 1.6). Analysis of optical purity of the
chiral glycine was conducted by derivatization of the amine with (1S)-(-)-camphanic
chloride (Spectrum 1.3). The known camphanate 1.25 derivative was analyzed by 1H
NMR and it showed a de of 82 % by integration of the glycine methylene proton region.57
This number was in agreement with 2H NMR analysis of the same derivative. As with
the MTPA ester, the doubly protonated contaminant and the deuterium epimer of 1.25 are
seen in the 1H NMR spectrum. Most likely, deterioration of stereochemistry from the
3
2.5
2 y it
ic 1.5 pt li l e
1
0.5
0 200 210 220 230 240 250 260 270 280 wavelength (nm) Figure 1.6 Circular dichroism spectrum of (R)-glycine-d-15N (1.24) (diamonds) and unlabeled glycine (squares) in D2O.
26 Major (R) -derivative
O O O 15 HN COOH DH
1.25 Doubly protonated Minor (S ) -derivative
4.45 4.40 4.35 4.30 4.25 4.20 4.15 4.10 4.05 ppm
Spectrum 1.3 Partial 800 MHz 1H NMR spectrum of the (R)-glycine-d-15N camphanate derivative 1.25 in CDCl3.
chiral alcohol 1.4 to the final glycine 1.24 occurred in either of two reactions.
Epimerization may have transpired in the ruthenium tetraoxide oxidation and in the camphanate derivatization of the chiral glycine.
1.4 CONCLUSIONS
Poor yields in the ruthenium oxidation led to the investigation of the optimal set of conditions and optimal aromatic ring synthon for the production of a carboxylic acid.
The overall conclusions from the semi-quantitative oxidation study was that the use of periodate salts as the stoichiometric re-oxidant is not favored for the oxidative
27 degradation of aromatic rings. Subsequently, the optimal oxidation condition was found to be with periodic acid as the stoichiometric re-oxidant and the optimal substrate was N-
(p-methoxyphenylmethylamine)-2,2,2-trichloroethyl carbamate (1.22). Since the oxidation reaction is a key step, optimization was paramount for the applicability of this route. This presented synthesis of the doubly labeled glycine has been substantially improved over the previous work with consistently better overall yields, 18 % versus the inconsistent 10 %, giving nearly double the amount of the title glycine. Furthermore, additional characterizations such as determination of enantiomeric purity of several intermediates, multinuclear NMR experiments, purification improvements, and numerous analyses on the final product make this a bona fide route to the target compound.
28
PART 2 – SYNTHESES OF CARBON-LINKED ANALOGS OF RETINOID GLYCOSIDE CONJUGATES
29
CHAPTER 2
INTRODUCTION FOR RETINOIDS
2.1 DISCOVERY
From 1906-1912, Frederick Gowland Hopkins first described landmark feeding experiments with specialized diets and observed the physiological effects on animals. It was shown that even though animals were given the appropriate calories for survival, they would still lose weight and even die on these purified diets. During the same period,
E. V. McCollum and Marguerite Davis at the University of Wisconsin showed that butter and egg yolk, but not white fat (e.g. lard), contained a lipid substance that was necessary for the growth of rats. They also isolated components that partitioned into water and in
1913 coined the terms “fat-soluble A” and “water soluble B”. Consequently for the first time, a single factor was found to be needed for normal growth in a mammal.58
These important discoveries were the beginning of nutrition research through the investigation of vitamins. The term vitamin, coined by Casimir Funk in 1912, can be broken down into vita, meaning life, and amine, because he thought they were amines.
Vitamins are organic substances required by mammals in trace amounts from exogenous sources to maintain normal metabolism. These molecules, which cannot be biosynthesized by the mammal, are essential for ordinary growth, reproduction, and 30 normal function. In general, vitamins are found in foodstuffs or sometimes produced
synthetically and are not carbohydrates, lipids, proteins, or inorganic salts. Mammals are
thought to have vitamin requirements to conserve biosynthesizing energy by relying on
other organisms to supply these nutrients.
In the 1920’s, Wolbach and Howe in Boston first characterized the deficiency
symptoms of the “fat soluble A”, which included weight loss and keratinization. Another important discovery was accomplished by Moore in England and he showed that a colored extract from plants was converted into a colorless form of the vitamin in the liver after ingestion. In 1930, Karrer in Switzerland proved the structures of both vitamin A and β-carotene, subsequently synthesizing some derivatives which included retinal
(Figure 2.1). In 1935, George Wald of Harvard found retinal in the visual pigments of the eye and was awarded a Nobel Prize for this work.58
2.2 STRUCTURES
Retinoid is a generic term used to describe molecules that resemble the structure
or the actions of vitamin A. Seen in Figure 2.1 are the structures for all-trans retinol
(vitamin A) and other active retinoids. The structures consist of highly conjugated
terpene hydrocarbons with a polar functionality on one end, which gives them
amphiphilic properties. Also due to the conjugation, retinoids are susceptible to UV-Vis
detection and have high extinction coefficients. Some active double bond isomers of
retinol include cis bonds which occur at the 9- and 11-postions. The all-trans isomers
have higher absorption maxima and extinction coefficients than the respective cis-
isomers. The terms “cis” and “trans” is the traditional nomenclature but, for example, 31 20 16 17 19 7 11 15 1 9 13 OH 3 5 18 CHO 11-cis retinal all-trans retinol (Vitamin A)
O
all-trans retinal 9-cis retinoic acid O OH
O O
OH OC15H31
all-trans retinoic acid retinyl palmitate
β-carotene
Figure 2.1 Important naturally occurring retinoids.
the accepted chemical notation for vitamin A is 7E,9E,11E,13E-retinol.59 Furthermore, due to the polyene system, they are highly sensitive to white light and oxidation and thus should be handled using specialized techniques. If not handled in yellow lights, the polyene system will begin to isomerize to cis isomers due to high energy photons found in white light. The molecules are generally low melting solids and insoluble in water but highly soluble in organic solvents.
32 2.3 NATURAL AND SYNTHETIC SOURCES OF VITAMIN A
Most of the vitamin A found in animals are fatty acid esters of retinol, the most common ester being retinyl palmitate. The very richest source of retinyl esters is found in liver tissue of marine fish such as shark and halibut but also marine mammals such as polar bears. However, common human dietary sources are animal products such as milk and eggs. The major provitamin carotenoid, β-carotene, is found in carrots and green leafy vegetables, such as spinach and also some fruits such as papaya and oranges. The richest source of β-carotene is red palm oil which contains 0.5 µg/ml. There are over 600 known carotenoids at present in the plant world, however only approximately 50 contain provitamin A activity for mammals. The natural sources of retinoids are usually isolated through organic extraction of the particular matrix, followed by chromatography or distillation. To hydrolyze retinyl palmitate from animal sources, saponification is traditionally used, followed by low temperature crystallization to obtain pure vitamin A acceptable for human consumption.58
Even though retinoids are readily obtained from natural sources, synthetic vitamin
A is much more practical to obtain. In the late 1940’s, Otto Isler and colleagues at
Hoffman-La Roche designed a synthesis starting with the inexpensive precursor β-ionone and using a Grignard reaction as a key step.60 More than a decade later, BASF published a competing synthesis, also starting from β-ionone, but used a Wittig reaction as a key step.61 Both routes are summarized in Figure 2.2. These important synthetic discoveries decreased the price of the vitamin 10 fold at that time, thus assuring the possibility of
33 O
1) ClCH2CO2Et, NaOEt HNa 2) OH- A B
OH O
H
H2, Lindler's cat BrMg OMgBr
OH
OH OH
1) H2, Pd Ph3P, HCl 2) CH3COCl
PPh3Cl OH OAc
1) NaOMe
HBr (aq), CH2Cl2 2) OHC OAc
OAc
Figure 2.2 Commercial routes to retinyl acetate, (A) La Roche and (B) BASF procedures.
34 using vitamin A as a practical dietary supplement. Furthermore, these basic routes are
still used today to commercially prepare retinyl acetate.
2.4 ABSORPTION, STORAGE, AND METABOLISM
Upon ingestion, retinyl esters from animal tissues and carotenoids from plants
gather with other fat soluble nutrients into fatty globules in the stomach. Entering the
duodenum, the globules are absorbed into enterocytes in the upper half of the small
intestine.62 Before hand, the retinyl esters are readily hydrolyzed into retinol in the
intestinal lumen where a specific transporter exists for the absorption by the
enterocytes.63 The carotenoids enter the cells by passive diffusion of the micelles. Some of the carotenoids are metabolized into active retinoids (predominantly into retinal) in the enterocytes by undergoing oxidative cleavage of the central double bond. The enzyme β-
carotene 15-15’-oxygenase is a monooxygenase and it catalyzes the epoxidation of the
double bond, which is followed by unselective ring opening with water to yield two
aldehydes.64,65 β-carotene is the only carotenoid to be converted to two molecules of
active vitamin A constituents.
After the uptake of retinol into the enterocytes, retinyl palmitate, and to some
extent stearate, oleate, and linoleate, is formed, packaged into chylomicrons, and released
into the lymph duct. In the duct, the chylomicrons are degraded into chylomicron
remnants, which contain retinyl esters, cholesteryl esters, and other lipids. The remnants
are taken up by cell surface receptors on the liver and other tissues, though most of the
dietary retinol is processed by the liver. In the hepatocytes, retinyl palmitate is
hydrolyzed to retinol via retinyl ester hydrolase and subsequently, complexed with 35 cellular retinol-binding protein I (CRBP-I).66 Retinoid-binding proteins play key roles in
the metabolism and function of their ligands. CRBP-I and II serve as co-ligands for
enzymatic transformations such as hydrolysis and formation of fatty acid esters. The
protein complex may continue to circulate in the cell or be transported to the endoplasmic
reticulum to bind with retinol-binding protein (RPB). Retinol binds to these carrier
proteins (1:1 mol/mol) in a hydrophobic pocket formed by flanking β-barrels,67 therefore,
protecting it from oxidation and degradation during transport in the plasma or cytoplasm.
It is important to note that the regulation of retinoids in the cell is tightly controlled
because of several factors including solubility of retinoids in the aqueous intracellular
environment, limiting the level of free retinol sequestering in membranes, and untimely
processing and damaging of the retinoid.
The hepatocytes can then release the retinol-RBP or retinol-RBP complexed
together with transthyretin, which is a tetramer that binds thyroid hormones. After
release, approximately 80% of the retinol-protein complexes are transferred to another
cell type in the liver which is the stellate cell.68 Within the stellate cells, the retinol is
rapidly esterfied again with fatty acids and stored in fat globules, which contain up to 60
% retinyl palmitate, approximately 30 % triacylglycerols and small amounts of
cholesterol and α-tocopherol. In mammals, up to 80 % of the total retinoids are stored in
the liver stellate cells and up to 95% of that amount is stored as retinyl palmitate.69 Since the majority of retinol is stored, there has been a lot of attention given to elucidating the regulation of this storage mechanism. It has been shown that retinyl ester hydrolase, which is closely associated with the storage globules, exists in several forms and is thought to have a key role in the regulation of storage.70 36 After retinol-RBP is released from the liver stellate cells, it is transported through
the plasma complexed with transthyretin in a 1:1 mol/mol ratio. The purpose of the large
transthyretin complex is thought to minimize the loss of RBP through filtration by the
kidney because only the retinol-RBP complex is needed to bind to cell surface receptors
of target tissues.71 Furthermore, since the complex is not lost through the kidneys, the
retinol can be recycled and sent back to the liver for storage. In the general circulation,
the carrier proteins take retinol to its numerous target tissues where cell surface receptors
for RBP have been found, for example in retinal pigment epithelium (50,000 per cell),
liver, skin, placenta, testes, and the blood brain barrier.69
Inside the target cell, many transformations of retinol can occur to affect its
subsequent actions. The major routes of metabolism are summarized in Figure 2.3.62 In many cell types, retinol can be reversibly oxidized to retinal by various NAD-dependent alcohol dehydrogenases. However, specific retinal dehydrogenases and reductases do exist. Next, retinal is irreversibility oxidized in many tissues, except for the eye, to retinoic acid by aldehyde dehydrogenases and oxidases.70 Generally, a very small portion
of retinoid (0.2-5 %) is present as retinoic acid in the plasma and tissues. These three
molecules account for the vast majority of the potent biological activity of retinoids.
Other metabolites are largely devoid of activity such as 4-hydroxyretinoic acid, 5,6-
epoxyretinoic acid, and oxidized chain shortened metabolites. These intermediates are
subjected to further oxidation, polyene chain cleavage, and conjugation reactions to
prepare for excretion into the bile. Glucuronyltransferases accept retinol and retinoic
acid as substrates and form glucuronide conjugates using UDP-glucuronic acid. Unlike
the previous inactive metabolites, the glucuronides retain significant biological activity.70 37 O 5,6-Epoxyretinyl OH ester OH Retinyl ester 5,6-Epoxyretinol 14-OH-4,14-retro retinol
OH O O OH HO O HO Retinol HO Retinyl glucuronide
O Schiff base with proteins Cis Isomers H Retinal
OH O O O HO O HO Other OH HO conjugations Retinoic acid O Retinoyl glucuronide
CO2 4-OH-retinoic acid 5,6-Epoxyretinoic acid Chain shortened metabolites
O 4-Oxoretinoic acid
Figure 2.3 Major metabolic routes of retinol.
38 Retinoyl glucuronide is not hydrolyzed in some cell types, resulting in a slow breakdown
in vivo. Both glucuronides are synthesized in the intestine, liver, and in other tissues and
also both are found as endogenous components of blood. Interestingly, it has also been
shown that the rate of glucuronidation is faster for 9-cis and 13-cis retinoic acid in rat
liver microsomes.72
Quantitatively, vitamin A homeostasis in the human body is greatly affected by the condition of the individual. But generally, when ingested, 10-20 % of the vitamin is not absorbed and is excreted within 1-2 days in the feces. The remaining 80-90 % is absorbed and 20-60 % of that can be conjugated or oxidized into metabolites, which is excreted within a week.62 The remainder of absorbed vitamin is then stored primarily in
the liver, which is metabolized much more slowly. The half life for the overall depletion
rate of vitamin A in humans is 128-156 days.73
2.5 BIOLOGICAL EFFECTS OF RETINOIDS
2.5.1 VISION
11-cis retinal serves as a chromophore in the photosensitivity in insects, crustacea,
arthropods, cephalopods, and vertebrates. In all the organisms, this protein- bound
molecule is responsible for absorbing an incoming photon, which causes an isomerization
of the cis double bond to trans. This fundamental structural change initiates a sequence
of events that is eventually interpreted as sight.66 A summary of the visual cycle is
shown in Figure 2.4.74
39 all-trans-Retinol BLOOD
RETINAL PIGMENT EPITHELIUM all-trans-Retinyl ester all-trans-Retinol
11-cis-Retinyl ester 11-cis-Retinol
NAD+
NADH
11-cis-Retinal
INTERPHOTORECEPTER MATRIX
NEURAL RETINA
NADPH NADP+ 11-cis-Retinal all-trans-Retinal all-trans-Retinol
Opsin
t Membrane Sodium arrest en polarization Rhodopsin segm r
Disk Oute membranes Nerve Photon impulse t en
-Nucleus
r segm -Mitochondrion -Synaptic nne I terminal ROD CELL
Figure 2.4 Summary of the visual cycle.
40 From the general circulation, the donor retinoid to the retinal pigment epithelial
(RPE) is retinol bound to the RBP-transthyretin complex. This complex interacts with
the cell surface receptor and is then internalized into the RPE. Once in the cell, CRBP
binds retinol which is acylated to a retinyl ester. In a concerted reaction, the ester is hydrolyzed and isomerized by the enzyme isomerohydrolase to give 11-cis retinal. Since the all-trans isomer is more thermodynamically stable, it is postulated that the isomerization is coupled to the ester hydrolysis to provide the ample amount of energy needed for the transformation.75 Cellular retinaldehyde-binding protein (CRALBP) binds
the 11-cis retinal and transports it to the interphotoreceptor matrix where the carrier
protein, interphotoreceptor retinoid-binding protein (IRBP), carries the retinal to the rod
cell outer segment of the neural retina. The retinal combines covalently to opsin in the
disk membranes of the rod cell by forming a Schiff base with lysine residue 296,
constructing the complex called rhodopsin.
Located in the dark-adapted retina of the eye, rhodopsin is seven-membered helix
membrane bound G-protein coupled receptor that has a maximum absorbance of 498-500
nm. Retinal resides in a hydrophobic pocket formed by several transmembrane helices
and lies near the cytosolic side of the disk membrane. Three variants of rhodopsin called
iodopsins exists in the cone cells of the eye with maximum absorbances of 420 nm (blue
cones), 534-540 nm (green cones), and 563-570 nm (red cones). When a photon of light
strikes rhodopsin, 11-cis retinal is converted to the trans isomer which puts a high degree
of strain on the protein. The tertiary structure of the protein changes to a conformation
called metarhodopsin II. This new state interacts with transducin, which is a G-protein in
the disk membranes. The α-subunit of transducin binds GTP, which activates cGMP 41 phosphodiesterase, converting cGMP to GMP. In the unactivated state, sodium pores in the membrane of rod cells bind cGMP and permit sodium ions to enter. When light strikes, cGMP is degraded, causing the sodium pores to close, and inducing the membrane to become hyperpolarized. The polarized rod cell triggers a nerve impulse to other cells in the retina through the synaptic terminal, which is eventually interpreted by the brain as visual stimulus.58,74
The all-trans retinal resulting from photon isomerization is released from metarhodopsin II, binds to CRALBP, and reduces back to retinol in the Müller cells of the neural retina. IRBP then transports the retinol through the interphotoreceptor matrix and into the RPE where it enters back into the cycle. It is also interesting to note that retinal is not active in any other cell types in which the other retinoids show potent activity.
2.5.2 GROWTH AND DIFFERENTIATION
From the beginning of nutrition research, it was known that vitamin A causes a growth response in mammals. With the onset of vitamin A deficiency, a growth plateau is seen in experimental animals, which is followed by severe weight loss and death. One cause of this wasting is that initial effects of deficiency is the loss of appetite, occurring in 1-2 days. For normal growth, a rat requires an average dose of 14 nmol (4 µg) per kg body weight per day of vitamin A.76 In humans, the minimum amount is not known, however, there are some guidelines for dietary intake. The RDA for males, females, and pregnant females is 1000, 800, and zero retinol equivalents, respectively. A retinol equivalent is defined as 1 µg of retinol or 6 µg of β-carotene. Other effects of deficiency 42 are that mucus-secreting cells transform into keratin-producing cells in many tissues.
However, upon treatment with vitamin A, the cells return to mucus-secreting cells. Since
they are potent inducers of differentiation, administration of retinoids cause a host of new
proteins to appear in newly differentiating cells. Furthermore, studies have indicated that
retinoic acid is the active metabolite in these effects by the observation that administered
retinoic acid stimulated growth in deficient rats but did not maintain normal vision.77
2.5.3 EMBRYONIC DEVELOPMENT
Vitamin A deficiency as well as excess produces severe malformations during
embryogenesis. These abnormalities involve almost the whole body and include the
central nervous system, eye, face, ear, limb, urogenital tract, skin, lungs, heart, and
hematopoiesis. Known genes that play a key role in the development in several cell types
and several species are the four Hox gene clusters.78 Some of these genes are known to be directly influenced by retinoic acid. Hox a-1 is activated which subsequently activates the other gene clusters containing Hox a-2 through a-13. Other influenced clusters are the Hox b-1 and the Hox d family, which is involved in hindbrain development and limb development, respectively. Numerous other genes are also activated or suppressed by
treatment with retinoic acid. The exact mechanism through which retinoids evoke their
effects is still being elucidated, however, all-trans retinoic acid is presumed to be one of
the host of morphogens that control embryonic development.79
43 2.5.4 IMMUNE SYSTEM
Another effect that retinoids have on human development is on hemotopoietic
progenitor cells, which include the vast array of cells in the immune system. Early in its
discovery, vitamin A was termed the “anti-infective” vitamin because of the increased
number of infections observed in deficient animals and humans. Due to impaired cell-
mediated and mucosal immunity, natural killer cell activity, and phagocytosis, a
decreased response to bacterial, parasitic, and viral infections are seen in deficient
animals. Furthermore, when vitamin A supplementation is resumed, the primary immune
response generally returns to normal capacities. Even though the vitamin is necessary for
the primary response, deficiency does not seem to affect immunological memory, which
is essential for a secondary response.62,80
It is evident that retinoids are essential in the promotion of the complex interplay
of cells and peptide factors which make up the immune system. However the exact role
that retinoids play is still not clarified. At least one aspect of the action of retinoic acid is
now becoming clear. In a normal immune response, an antigen is phagocytized by an
antigen-presenting cell in which the antigen is broken up by special proteases. A
fragment of the antigen is then presented on the cell surface after which a T-helper
lymphocyte becomes activated upon contact with the fragment. Interleukin-2, which is
made by the T-cell, stimulates the proliferation of more T-cells and B-cells. There are
two types of T-helper cells, TH1 and TH2, and each type secretes cytokines which stimulate cell-mediated immunity (TH1) and enhancement of antibody production (TH2).
The ratio of TH1 to TH2 cells will determine the type of immune response when presented with an antigen. In vitamin A deficient rats, the ratio of TH1 to TH2 is increased, which 44 leads to an increase in cell-mediated immunity response but a decrease in antibody
production. These valuable studies elucidated at least one of the sites of action for
retinoids as the T-helper cells.58,81
2.6 CELLULAR MECHANISMS OF RETINOIDS
At the heart of all the biological effects of retinoids (except vision) is the
metabolite retinoic acid. Within cells, all-trans retinol bound to CRBP is oxidized to all- trans retinoic acid and, presumably, the all-trans materials can be transformed at some point to the 9-cis isomers. Either isomer of retinoic acid can bind and be transported with cellular retinoic acid-binding protein (CRABP-I, II) to the cell nucleus. Once inside the nucleus, the retinoic acids tightly bind to either the retinoic acid receptors (RAR) or the retinoid X receptors (RXR). Each nuclear receptor has three subtypes α, β, and γ. RARs can bind either 9-cis or all-trans retinoic acid whereas RXRs bind 9-cis retinoic acid.
Upon ligand binding, RARs form heterodimers with RXRs and RXR form homodimers.
Figure 2.5 shows schematically the cellular actions of retinoic acid.62
Upon the discovery of the RARs82,83 and RXRs,84 it was determined that they
were similar to other proteins in the super family of hormone nuclear receptors. The
receptor contains six domains, each possessing a specific function. Starting from the N-
terminus, domains A and B serve as physiological activators of the receptor. Domain C
is highly conserved across all six subtypes and contains residues called zinc fingers,
which have zinc-sulfhydryl moieties. These domains are positively charged and interact
favorably with the negatively charged phosphates on the DNA backbone. Domain D is a
hinge region on the protein that provides for the appropriate conformational change to 45 PLASMA CYTOPLASM NUCLEUS
all-trans Retinoyl all-trans Retinoyl all-trans Retinoyl β-glucuronide β-glucuronide β-glucuronide
all-trans all-trans Retinol all-trans Retinol-RBP Retinol-CRBP all-trans Retinoic acid- RXR-RAR RARα,β,γ DNA-RE
9-cis Retinol all-trans Retinal-CRBP RXR-RXR DNA-RE
9-cis Retinal RXR-TR all-trans Retinoic acid DNA-RE 9-cis Retinoic acid- RXRα,β,γ RXR-VDR 9-cis Retinoic acid DNA-RE all-trans Retinoic acid-CRABP RXR-?? 9-cis Retinoic acid- DNA-RE CRABP
STIMULATE / INHIBIT PROTEINS RIBOSOME mRNA TRANSCRIPTION
Figure 2.5 Overview of the cellular mechanism of retinoic acid. Solid lines indicate known processes and dashed lines indicate postulated transformations.
occur upon ligand binding. The retinoic acid binds in the domain E region which is also
highly conserved within each receptor type. Finally, domain F, at the C-terminus, is where the two receptors interact to form dimers. As far as size, the receptors range from
410-467 amino acid residues and are 45-51 kDa in size. As far as homology is concerned, each RAR is very similar, differing by only 3 amino acids, but for example,
RARα and RXRα show little homology, 61 % in the DNA binding region and 27 % in
the ligand binding region.85 46 Each of the six retinoic acid receptors show different chromosomal locations, embryonic developmental expression, and organ localization. RARα and RXRβ are ubiquitous in their occurrences in tissues. RARβ and RXRγ are found in high amounts in adult muscle and heart but are different in appearance and amounts in embryonic tissues.
RARγ is rich in the adult skin and lung and also has different embryonic distribution.
Lastly, RXRα occurrs in the liver, skin, and kidney of both embryos and adults. In further detail, RXRs outnumber RARs 5 to 1 in the skin. Of the RARs in the skin, γ is the major isoform at 87 % and of the RXRs, α is the only one detected.86 Therefore, each receptor and their subtypes seem to have their own distribution and function, and operate independently. The RXR family of receptors shows the broadest range of activities.
When RXR does not bind 9-cis retinoic acid, it can form heterodimers with other members of the super family of hormone nuclear receptors such as the vitamin D receptor
(VDR), the thyroid hormone receptor (TR), and, obviously, RAR. In addition, RXR forms heterodimers with two orphan receptors, the chicken ovalbumin upstream promoter transcription factor (COUP-TF) and the peroxisomal proliferator activating receptor
(PPAR). Thus, whether RXR forms a hetero or homodimer is greatly dependent on the ratio of concentration of all-trans to 9-cis retinoic acid in the cellular nucleus.58,85
In the absence of ligand, the RAR-RXR dimer recruits histone deacetylase, which is tethered through several corepressors. This results in the silencing of gene transcription by histone processing and chromatin compacting. Upon ligand binding, the complex destabilizes the corepressor binding due to the large allosteric changes in the ligand binding domain. The complex then sequesters different cofactors (e.g. histone
47 acetyltransferase) allowing access to the DNA and binds to a sequence of base pairs on
DNA called hormone response elements (HRE). The HRE for both RXR and RAR is the
consensus sequence AGGTCA. In order for gene activation to occur, a direct repeat of
the sequence is needed and a non-palindromic arrangement is most common (e.g.
AGGTCA – Xn – AGGTCA). The type and the number of bases between each HRE can
define the type of dimer (RXR-RXR or RAR-RXR) to bind. The HRE sequences are
located upstream from a promoter region, which is next to the target gene. Depending on
the particular HRE, promoter, and other soluble co-factors that are recruited to the bound
dimer this will dictate whether there is gene transcription or inhibition. This transcription
occurs in the nucleus after which the mRNA is transported into the cytoplasm, translated
with ribosomes, and finally gives a protein product.
2.7 THERAPEUTIC VITAMIN A AND RETINOIDS
The main therapeutic application of vitamin A is to treat morbidity and mortality
of children in the developing world. Deficiency increases the severity, complications and
the risk of death from measles and vitamin A is used as a prophylactic or as a treatment.
Oral doses of vitamin A in oil are given to children of 2-6 years of age three times a year
as a preventative health initiative in many countries.87 Other approaches such as food
fortification have been used to improve public health. Recently, the β-carotene
biosynthesis gene has been fused into the rice plant genome to produce yellow rice.88
Other campaigns such as nutritional education and government intervention are greatly improving world health.
48 In many types of skin diseases such as acne, psoriasis and ichthyosis, hyperkeratosis is a general manifestation in these malformed cells. A common effect of retinoids is their potent activity in differentiation of epithelial cells and therefore they have widespread use in treating skin disorders. Very effective retinoids (Figure 2.6) that are used for treatment are etretinate, acitretin, all-trans, and 13-cis retinoic acid.
However, at high doses, they are teratogenic when given orally and an irritant when given topically. Other synthetic analogs have been approved for the treatment of acne and psoriasis, such as adapalene and tazarotene, respectively. Conjugated forms such as retinoyl β-glucuronide and N-4-(hydroxyphenyl)retinamide (4-HPR) retain therapeutic actions while decreasing or eliminating their toxicity.89 Furthermore, the cosmetic industry has a large interest in the research and promotion of retinoids due to their reducing activity in wrinkles caused by photo-induced aging.
The mechanisms by which retinoids are effective with skin diseases are multifaceted but likely involve the promotion of epithelial differentiation. Since the receptor subtype distribution in the skin is known, the predominant dimer present is the
RARγ/RXRα heterodimer. Therefore, retinoic acid binds to this dimer and is likely to control the differentiation events in the skin.
Ample evidence suggests that there is a relationship between cancer and retinoids.
Carcinogenesis is a disorder in cell differentiation and under normal conditions, retinoids are regulators of differentiation and proliferation. Thus, it is likely retinol homeostasis inside cells influences the chance for malignant formation. It has been shown that in animals, dietary vitamin A and carotenoids can interfere with the progression of chemically induced neoplastic transformation.90 Furthermore, RARβ has been implicated 49 O
OH HO O all-trans Retinoic acid 13-cis Retinoic acid
O O
O OH
H3CO Etretinate H3CO Acitretin
OH OH O O O HO O N HO HO H O 4-Hydroxyphenylretinamide Retinoyl β-glucuronide
CO2H O
O
H3CO N CO2H
Adapalene S Tazarotene Bexarotene
Figure 2.6 Retinoids used in the treatment of skin disease and cancer.
as a potential earmark in many types of solid tumors. Loss of RARβ expression is associated with human tumor progression and the growth inhibitory effect of retinoic acid on tumors has been shown to induce expression of RARβ.91
Consequently, retinoids are used both as therapeutic and as chemopreventative agents in both experimental animals92 and in humans.93 In animals, most of the work has been in the chemoprevention arena using retinyl esters or other synthetic retinoids, usually with positive results. Chemically induced cancers of the mammary gland, skin, lung, bladder, pancreas, liver, digestive tract, and prostrate gland have been investigated. 50 There has also been success using combination therapy with retinoids and tamoxifen as chemoprevention agents against mammary cancers. In general, retinoids are more effective against tumors that follow discrete promotional stages of carcinogenesis, like mammary and skin tumors. In humans, chemoprevention trials have been performed with retinoids or carotenoids to prevent premalignanat lesions such as oral premalignancies, bronchial metaplasia, laryngeal papillomatosis, and actinic keratosis.
Retinoids have also been investigated for their preventative effects against secondary incidence of primary tumors of the head and neck, skin, breast, lung, and bladder.
In the clinic, all-trans retinoic acid is used to treat acute promyelocytic leukemia
(APL). This is the only known disease which is caused by a genetic defect in vitamin A function. In APL, the patients have a reciprocal translocation between chromosome 15 and 17, where the RARα gene is located. It is postulated that treatment with high levels of retinoic acid is required to activate the defective RARα transcription factor and induce cellular differentiation. However, after several months of treatment, resistance develops due to the increase in absorption and metabolism of retinoic acid. Another retinoid approved for the treatment of cancer is bexarotene (Figure 2.6), which is used for cutaneous T-cell lymphoma, however, it has been shown to be teratogenic. But in general, chemotherapy of carcinoma in animals or humans with natural retinoids is severely limited due to their toxicity at pharmacological doses. Chemotherapy with synthetic retinoids has shown some success and will be discussed in a later section.
51 2.8 TOXICITY OF RETINOIDS
Three types of retinoid toxicities exist, acute, chronic and teratogenic. Acute
toxicity arises from a single dose or large doses over a short period of time. Some
symptoms such as nausea, vomiting, headache, increased spinal fluid pressure, and
impaired vision are observed with a dose of approximately 200 mg (0.7 mmol, 660,000
IU). A lethal dose for a young monkey was 168 mg (0.6 mmol) which extrapolates to
11.8 g (41 mmol) for a 70 kg adult human. Chronic toxicity can arise from a recurrent
dose of 10 mg (35 µmol, 33,000 IU), which is 10 times the RDA. More than 50
symptoms of chronic toxicity have been reported which include alopecia, bone and
muscle pain, hepatotoxicity, hyperostosis, and skin disorders. It has been shown that
daily doses of 26-52 µmol causes liver toxicity and in these individuals, retinol-RBP in the plasma is not elevated but the storage of retinyl esters is significantly increased, consequently giving rise to the hepatotoxicity.94
Natural and some synthetic retinoids are powerful teratogens in female animals
and humans. Large single doses (0.1 mmol) or prolonged elevated doses (26 µmol) of
vitamin A during early pregnancy cause spontaneous abortions or major fetal deformities.
Common birth defects include craniofacial abnormalities, heart disease, kidney defects,
thymic abnormalities, and central nervous system defects. The mechanism of toxicity
most likely involves the increased activation of the nuclear retinoid receptors. The Hox genes are known to be activated by retinoids and are involved in embryogenesis. These effects will influence the timing of cell proliferation, differentiation, migration, and cell death.95
52 2.9 SYNTHETIC RETINOIDS
Because of their potent activity in cell regulation, retinoids are mainly used to
treat various proliferative diseases, however, they are only therapeutically effective at
high doses, at which they have high toxicity. Therefore, the search for a synthetic analog
possessing the optimal therapeutic index carries on. Before the nuclear retinoid receptors
were discovered, many synthetic analogs of vitamin A were prepared to elucidate their
structure activity relationship (SAR). This earlier work also was driven by the fact that
the natural retinoids have the ability to inhibit neoplastic transformations, revealing its
chemopreventative qualities. In the interest of developing synthetic analogs that could be
used to prevent cancer, researchers at the National Caner Institute developed an assay
called the tracheal organ culture assay (TOC).96 This important assay allowed for the
expedient testing of compounds, which spurred the chemistry of producing synthetic
analogs. In general, the SAR of an active retinoid includes a polar terminus with an
acidic pKa, a lipophilic polyene side chain with π electron delocalization, fairly lipophilic
ring opposite the polar terminus, and conformational restriction.97 Upon the discovery of the retinoic acid receptors, researchers continued to design retinoids that were active as antiproliferative agents but also turned their attention on the development of analogs that could selectively inhibit or active particular subtypes of either the RAR or RXR.
2.9.1 RECEPTOR DEPENDENT RETINOIDS
Many compounds were initially designed as therapeutic candidates because of the increased stability over retinoic acid. One of the first synthetic analogs made was
TTNPB, which possesses only one internal double bond, giving it far more stability than 53 the natural retinoids.98 Even today, TTNPB is one of the most potent and teratogenic
retinoids known. Synthesized before the nuclear receptors were discovered, it was later
found that it binds and activates RARs selectively over RXRs. Other compounds that
derived from TTNPB were TTNN99 and TTAB,100 which are fully aromatic retinoid
benzoic acids. TTNN not only activated RAR selectively but RARβ and RARγ specifically. This was the first example of an analog that possessed some subtype selectivity and provided the initial leads toward the pursuit of subtype selective retinoids.
See Figure 2.7 for the structures of retinoic acid analogs.101,102
The first RARα selective retinoids discovered were Am80 and Am580,103 which are analogs of TTNPB with the insertion of an amide bond. The next receptor selective retinoid that was reported was for RARγ. Based upon the structure of TTNN, insertion of a Z-configured oxime group gave rise to SR11254.104 Identification of a RARβ selective
ligand has been somewhat elusive. There have been some analogs reported like CD2019
that have modest selectivity for the RARβ.105 In general, analogs possessing RARα selectivity may be easier to obtain because RARβ and RARγ have greater homology in their ligand binding domains compared to the LBD of RARα. After it was found that 9- cis retinoic acid was the ligand for RXRs, several of the previously made analogs were shown to possess little activity for the activation of RXR. A simple but effective modification to a known analog that showed RXR selectivity was the 3-methyl derivative of TTNPB.106 Shortly after, another RXR-selective retinoid prepared was SR11217,107 showing modest activity. Subsequently, more potent analogs were prepared by replacing
the tetra-substituted double bond of SR11217 with a terminal olefin and adding a 3-
54 CO2H CO2H CO2H
TTNPB TTNN TTAB
CO2H CO2H H O CO2H N N O H N HO Am80 Am580 SR11254
CO2H CO2H CO2H
MeO
CD2019 Me-TTNPB SR11217
CO2H CO2H
CO2H
SO OO
LDG1069 AGN193109 Ro41-5253
F
CO2H O CO2H
N F O H SS O Br CO2H LGD100754 AGN194301 SR11253
Figure 2.7 Receptor dependent retinoic acid analogs.
55 methyl group, giving LDG1069.108 Retinoids with high RXR subtype selectively are still
being pursued.
Instead of binding and activating the retinoid nuclear receptors, synthetic analogs
may also bind and inhibit the actions of the receptor. Retinoid antagonists must bind and
arrest the conformational change needed in the ligand binding domain for the induction
of gene transcription. One of the first inhibitors of RAR transcription was AGN193109,
showing inhibition of all three subtypes.109 Shortly after, subtype selective antagonists
were being discovered. One of the first compounds reported is the modestly RARα
selective Ro41-5253,110 which is a heterocylic variant of TTNPB. An even more
selective α antagonist is AGN194301111, which is a halogenated variant of AGN193109.
The first RARγ selective antagonist reported was SR11253104 and it is a dithiolane derivative of TTNN. Finally, the first RXR antagonist was reported to be LGD100754 and it potently inhibits all three subtypes.112 RXR subtype selective antagonists are still
being pursued.
2.9.2 RECEPTOR INDEPENDENT RETINOIDS
In addition to ligands that bind RARs or RXRs, there is a growing field of
retinoids that do not share the typical profile. These retinoids are being called “atypical
retinoids” and they are intensely being investigated for their chemopreventative and
therapeutic potential. Whether or not atypical retinoids bind and activate receptors is still
unclear, but at most, some may weakly bind and activate RARγ. Their main actions are
that they are growth inhibitory and cause apoptosis in cancer cells selectively over
56 normal cells. These effects are not blocked in the presence of retinoid receptor
antagonists and they are still active in retinoid resistant carcinoma cell lines.113 This evidence supports the theory that atypical retinoids do not affect their actions through the retinoid nuclear receptors. Two main compounds in this class are emerging as significantly effective agents, 4-HPR and CD437 (Figure 2.8). 4-HPR is an aryl amide derivative of retinoic acid and CD437 is adamantyl derivative of the synthetic analog
TTNN.
As far as the mechanism of action of atypical retinoids, three novel apoptotic signaling pathways have been described for CD437.91 One involves the induction of
three types of death receptors, FAS, DR4, and DR5. The activation of these receptors
induces apoptosis through the release of FAS ligand and TNF-related apoptosis inducing
ligand. A second pathway involves induction of the TR3/NGFIB/Nur77 orphan nuclear
receptor. Also, a third signaling option is the eventual activation of p38 mitogen-
activated protein kinase. The exact mechanism of action of atypical retinoids is still
being investigated.
OH COOH O
N H HO 4-Hydroxyphenylretinamide (4-HPR) CD437
Figure 2.8 Structure of important atypical retinoids.
57 2.10 N-(4-HYDROXYPHENYL) RETINAMIDE
The most researched synthetic retinoid analog is N-4-(hydroxyphenyl)retinamide.
First synthesized in 1978, 4-HPR was initially tested in the TOC assay and shown to be highly active in reversing keratinization.114 1978 1978,115 In the same studies it also was demonstrated to be highly efficacious at preventing chemically induced mammary tumors in rats. Retinyl acetate was equally effective but had markedly higher toxicity.
Paramount to the further investigation and development of 4-HPR was the fact that it showed little toxicity when used as a preventative or therapeutic agent. In animal models, 4-HPR has been tested as a chemopreventative against cancers of the mammary gland (spontaneous and chemically induced), bladder, skin (spontaneous and chemically induced), colon, lung, prostate, and T-lymphomas (virally induced) with successful results. It has also been tested as a chemotherapeutic agent against mammary tumors, prostate cancer, and ovarian cancer. Furthermore, in human carcinoma cell lines such as breast, prostate, leukemia, ovarian, cervical, head and neck, and small-cell lung, 4-HPR has shown the ability to cause apoptosis effectively.116
Like CD437, the exact mechanism of action for 4-HPR is not known, however, it has been associated with several mechanisms of apoptosis. Some of which include an influence of bcl-2 production, increasing amount of transforming growth factor β1 expression, and eliciting oxidative stress, which generates reactive oxygen species. 4-
HPR has also been associated with the nuclear receptors, although the details remain unclear. It has been shown to bind poorly to RARα,β,and γ in vitro and only minimally activate the response elements.117,118 RARβ has been implicated as a site of action due to
58 a study where normal mammary epithelial cells increased RARβ production upon
treatment with 4-HPR.119 Alternatively, others have shown that 4-HPR has no measurable binding to any subtype of RAR.120 Furthermore, the apoptotic activity of 4-
HPR is not inhibited by the RAR pan-antagonist AGN193109.
The metabolism of 4-HPR has been studied in animals and in humans. It is
mainly metabolized by methylation of the phenolic hydroxyl group to give 4-
(methoxyphenyl)retinamide (4-Me-HPR). A minor metabolite is the conjugated product
4-HPR-O-glucuronide (4-HPROG).121 The methyl derivative is considered to be an
inactive metabolite, however, in some systems it has been found to be equally potent to
4-HPR. The actions of the glucuronide conjugate were not initially examined. However,
the structure of the glucuronide was confirmed by comparison with synthesized 4-
HPROG.122 In humans, plasma concentrations for 4-HPR are maximally reached at 5-7
hours and 8-12 hours for the metabolite 4-MPR. Subsequently, the half life of 4-HPR is
14-20 hours and 22-28 hours for 4-MPR.116 The metabolite structures are seen in Figure
2.9.
OH O OMe O O HO O HO HO O N N H H
4-MPR 4-HPROG
Figure 2.9 Structures of major metabolites of 4-HPR.
59 4-HPR has been in several clinical trials and has proven to be effective, however
with some variability. It was first investigated for the treatment of chronic plaque
psoriasis and 3 out of 8 patients were graded as being better after treatment (600 mg daily
dose). Subsequently, a phase II trial occurred with advanced breast cancer patients, who
had been treated with other drugs, and 4-HPR gave no benefit at doses of 300-400
mg/day. Other studies include randomized chemoprevention trials in breast cancer, basal
cell carcinoma, and oral leukoplakia in which 4-HPR has shown favorable but not
outstanding results.116 The breast cancer trial was for the chemoprevention of second
malignancies in patients with breast cancer.123 The study revealed that 4-HPR had no
statistical influence on the occurrence of secondary tumors in postmenopausal women,
but did show a benefit for premenopausal women. In general, 4-HPR has been well
tolerated. Classical retinoid toxicities like decreases in bone density, modifications of
lipoprotein levels, dermatological effects, and hepatotoxicity was absent or very low.
The major side effect seen with 20 % of the patients after five years of treatment was
dark adaptation problems.116 This side effect is postulated to be caused by the lowered plasma retinol levels observed during 4-HPR treatment.124
2.11 N-(4-HYDROXYPHENYL) RETINAMIDE-O-GLUCURONIDE
Glucuronidation is a common route of metabolism for the excretion of
endogenous and exogenous compounds. Glucuronidation is regulated by the enzymes
glucuronyltransferase, which attaches the donor sugar, and glucuronidase, which removes
the sugar. A common strategy for the chemoprevention of carcinogenesis is to promote
carcinogen metabolism and excretion. One way to expediate the excretion of the 60 carcinogen is by indirectly promoting glucuronidation by the inhibition of β-
glucuronidase. A potent β-glucuronidase inhibitor, and a chemopreventative, is D- glucaro-1,4-lactone, which can be formed in vivo from calcium glucarate. The main application for 4-HPR is as a chemopreventative and an interesting study showed that a combination of the two agents, 4-HPR and calcium glucarate, could act synergistically.125
Doses of glucarate (32 mmol/kg) and 4-HPR (0.75 mmol/kg), which were almost ineffective alone, inhibited chemically induced mammary tumor incidence and multiplicity by 50 %. Furthermore, it was found that the amount of 4-HPR and 4-
HPROG excreted in the bile of the rats treated with both agents was much lower than the
4-HPR treated rats. With the prevalent glucuronidation of 4-HPR and the slow breakdown of 4-HPROG, a high concentration of 4-HPROG in extrahepatic tissue, e.g. the mammary tissue, might be expected. It is known that the glucuronide conjugates of retinol and retinoic acid retain sufficient activity in maintaining growth and differentiation in normal and cancer cells. In general, retinoid glucuronides are much less
toxic and teratogenic than the parent retinoid. Therefore, a theory developed that the
glucuronide metabolite 4-HPROG may have significant activity in these tumor models.
To explore this hypothesis, 4-HPROG was synthesized (Figure 2.10) to supply
sufficient amounts for the following studies. A major advantage of 4-HPROG was found
during in vitro experiments with MCF-7 human breast carcinoma cell line.126 The results
were that the glucuronide had slightly better growth inhibitory action than 4-HPR,
although more importantly, it showed significantly less toxicity. To extend the in vitro
work, 4-HPROG was compared to 4-HPR in vivo for the ability to prevent chemically
induced carcinogenesis in rat mammary tissue.127 The glucuronide demonstrated superior 61 HO H OMe OMe O 1) NaOMe, MeOH O HBr, AcOH O OOH O O AcO O 2) pyr, Ac2O OAc AcO AcO AcO H OH 3) crystallize AcO AcO Br
OMe OMe AgO O O O H2, Pd / C O AcO O AcO O HO AcO AcO AcO AcO NO NO2 2 NH2
OH O 1) retinoyl chloride, pyr O HO O HO 2) saponification HO O N H
Figure 2.10 Synthesis of N-(4-hydroxyphenyl)retinamide-O-glucuronide
activity in all criteria that were measured which included inhibition of tumor incidence,
tumor multiplicity, and tumor growth. In addition, initial analyses of the blood from the
rats showed no measurable hydrolysis of 4-HPROG during the long term feeding. This
indicated that the intact molecule is active and has better activity and lower toxicity than
the parent molecule.
Another aspect of the antiproliferative activity of 4-HPROG was investigated by
measuring the chemotherapeutic potential for the treatment of chemically induced
mammary tumors.128 This study revealed that the glucuronide analog possessed greater
antitumor action than the parent 4-HPR, where tumor reduction was observed in 75 % of
62 the rats treated with 4-HPROG. Also, it was shown the maximum tolerated dose of 4-
HPROG was significantly higher than that of 4-HPR. In conclusion, these studies indicate the glucuronide is advantageous as a chemopreventative and chemotherapeutic in a rat mammary tumor model due to its greater potency and lower toxicity.
63
CHAPTER 3
SYNTHESIS OF THE C-LINKED ANALOG OF 4-HPR-O-GLUCURONIDE
3.1 RATIONALE
From previous work, it was demonstrated that the glucuronide metabolite of N-4-
(hydroxyphenyl)retinamide (4-HPROG) possessed superior activity over 4-HPR as a chemopreventative and therapeutic.126-128 These studies suggested that 4-HPROG was the active molecule and was not serving as a more bioavailable prodrug of 4-HPR.
Conversely, during in vitro experiments, 4-HPROG was shown to be a substrate for β- glucuronidase, which is an enzyme responsible for conjugate hydrolysis.129 To further investigate the activity of this molecule, an enzymatically stable analog was proposed.
Replacement of the phenolic oxygen link between the benzene ring and the carbohydrate with a methylene unit would render this molecule resistant toward β-glucuronidases. The activity of the C-linked analog might reveal whether or not 4-HPROG was active as an intact molecule. See Figure 3.1 for the structures of 4-HPROG and the C-linked analog
(4-HPRCG).
64 OH O O HO O HO HO O N H N-4-(Hydroxyphenyl)retinamide-O-glucuronide
OH O O HO
HO HO O N H N-4-(Hydroxyphenyl)retinamide-C-glucuronide
Figure 3.1 Structures of 4-HPROG and 4-HPRCG.
3.2 C-GLYCOSIDES
Carbohydrates have long been of interest to scientists because of their diversity
and their importance in cell biology. The synthesis, study, and use of carbohydrates have
often been complicated by the labile nature of the anomeric glycosidic bond. When the
anomeric oxygen is substituted with a carbon atom in a carbohydrate, it results in an
isostere commonly called a C-glycoside. The earliest syntheses for C-glycosides were
serendipitous and only of academic interest. In 1945, Hurd found that a 1-chloroglucose
analog would react with various Grignard reagents to give 1-aryl-C-glucosides.130 Also, in 1961, Helferich obtained an unexpected result when a 1-bromoglucose analog underwent nucleophilic attack from a cyanide ion to give a 1-cyano-C-glucoside131
(Figure 3.2). Interests in C-glycosides again peaked in the 1970s after the discovery of naturally occurring C-nucleosides with potent biological actions including antiviral, 65 OAc OAc O 1) PhMgBr AcO O AcO AcO AcO AcO 2) Ac2O, pyr AcO Cl
OAc OAc Hg(CN)2 O O AcO AcO AcO MeNO CN AcO 2 AcO AcO Br
Figure 3.2 First syntheses of C-glycosides.
antibacterial, and antitumor activity. With the discovery and syntheses of C-glycosides,
the limitations of carbohydrates could be surmounted to provide a new generation of
glycoside based materials. These conformationally similar analogs132 are enzymatically
and chemically stable and have many important uses in enzymology, medicine, and
industry. Recently, in the 1990s, interest arose again upon the robust growth of research
into glycobiology, providing the potential uses of C-glycosides as tools to study these
processes, such as cell surface glycosidation, enzymatic polysaccharide synthesis, and
amioglycoside antibiotics.133 In the pharmaceutical field, therapeutically active
carbohydrates can be studied in vivo using the C-glycoside isostere. This carbon
modification gives rise to antimetabolites, which can inhibit carbohydrate processing
enzymes. Some examples of their uses are as β-glucosidase inhibitors, β-galactosidase
substrates, glycogen phosphorylase inhibitors, sialyl LewisX precursor analogs, lipid A
analogs, hemagglutination inhibitors, and analogs for E. coli receptors.134,135
General synthetic routes for C-glycosides are depicted in Figure 3.3.135 The unique nature of the anomeric position on carbohydrates is advantageous for chemical 66 OH OH OH O RM RCH=PPh3 RR R CHO OH
OR OOH O + allyl silane O
Lewis acid
O OX OR - LiR RM
RCHO
OH O O R O R R
Figure 3.3 General routes to C-glycosides.
manipulations. Due to the delocalization of electrons on the anomeric carbon, it can be used to react in a variety of methodologies including electrophilic and nucleophilic substitutions, transition metal mediated glycosidations, anomeric radical reactions, rearrangements and cycloadditions, and ring closure reactions. Electrophilic substitution reactions encompass the largest scope of carbon-carbon bond formations and usually involve the displacement of a leaving group at the anomeric position. Cyanation reactions are common and occur with a cyanide salt displacing an acetate leaving group.
When a halogen is placed at the 1-position, alkylations are done with carbanions, 67 oganocuprates, Grignard reagents, and other organometallics. Lewis acids are commonly
employed to create a reactive anomeric carbon. They can be used with a leaving group or
the free hydroxyl group at the 1-position and then coupled with olefins or allyl silanes.
Aryl groups can be added directly to the anomeric carbon through methods mainly using
Lewis acid catalyzed displacement reactions with metallated aryl compounds. Lewis
acids have also been used to catalyze allylic ether additions to glycals (a 1,2-unsaturated
glycoside derivative) to give a C-glycoside aldehyde. Metallocarbonyl reagents can be
used to perform carbon monoxide insertions at anomeric positions to readily yield a
carbon linked carbohydrate. Sugar lactones can readily undergo nucleophilic additions
with organometallics. The resultant alcohol at the anomeric position can subsequently be
removed to give the C-glycoside. Sugar lactones can also be derivatized to an enol ether
by olefination reactions and these can be subjected to electrophilic additions.134
Another type of reaction that serves as a synthetic route to C-glycosides is nucleophilic substitution reactions, which usually involve a halogen or hydrogen-metal exchange at the anomeric carbon. This class of reaction is frequently used with glycals, but exchanges can be done using glycosides with a 1-position leaving group. Another method is transition metal mediated coupling reactions of a metallated glycal and an aryl halide. Anomeric radical formation is a popular reaction type and radicals can be generated from a glycoside halide. The radical is then coupled with an acceptor, such as various olefins, to form the coupled C-glycosides or can be used to form intramolecular cyclizations.
Other methodologies involve an open chain form of the carbohydrate in which the free aldehyde is functionalized. The carbonyl carbon of the aldehyde can be used in 68 organometallic nucleophilic attack or reaction with various Wittig reagents.
Subsequently, there are numerous types of methods for the ring closure of the newly
made C-glycoside.134 An important aspect for all of these methods for C-glycoside synthesis is the stereochemical outcome of each reaction type. Since most biological reactions are stereoselective, it is important for these methodologies to possess an isomerically specific process. Most of the reaction types mentioned above have either some stereoselectivity or are stereospecific.
3.3 PREVIOUS STUDIES WITH 4-HPR-C-GLUCURONIDE
The first synthesis (Scheme 3.1) of 4-HPRCG was reported in 1994 along with the synthesis of other C-linked analogs of 4-HPROG including the glucoside, other glucuronides, and xylose derivatives.129 To synthesize 4-HPRCG, tetra-O-benzylglucose
(3.1) was benzoylated with p-nitrobenzoyl chloride to give the conjugate 3.2.
Displacement of the 1-position leaving group with hydrogen bromide gives the α-1- bromoglucuose analog, which then undergoes electrophilic substitution with a benzyl
Grignard reagent to form the β-C-glycoside 3.3. Addition of the phenyl Grignard
reagents gave rise to the retinamidobenzene glycoside analogs (Figure 3.4). Next,
removal of the benzyl protecting groups was accomplished by catalytic hydrogenation,
which is followed by acetylation giving 3.4. The next key step is the selective oxidation
of the 6-position alcohol of deprotected 3.4 to the carboxylic acid with Adams’
catalyst.136 After the reprotected C-glucuronide 3.5 was obtained, the next key step was the generation of a para-nitro group on the phenyl ring. Nitration was accomplished fairly readily using copper nitrate, giving a ratio of 2 to 3 ortho / para substitution 69 O OBn Cl OBn NO 1) HBr, CH2Cl2 O 2 O O2N BnO BnO BnO O 2) PhCH2MgCl BnO BnO BnO OH pyridine O 61 % 3.1 82 % 3.2
OBn OAc 1) K2CO3, MeOH 2) PtO2, O2 O 1) H2, Pd / C O BnO AcO BnO AcO 2) Ac2O, pyr + BnO AcO 3) CH3OH / H 4) Ac O, pyr 3.3 3.4 2 26 % 10 %
O OCH 3 O OCH3 O Cu(NO ) O AcO 3 2 AcO 1) Pd / C, H2 AcO AcO AcO AcO NO2 2) separation 3.5 3.6 66 % 30 % 2:3 o / p
O OCH3 1) Retinoyl chloride, pyr O 2) NaOMe, MeOH AcO AcO 4-HPRCG AcO 3) NaOH, MeOH, H2O NH 3.7 2 28 %
Scheme 3.1 The first synthesis of 4-HPRCG.
70 H R N O HO O HO HO
R = H Retinamidophenyl xylose
R = CH2OH Retinamidophenyl glucose R = COOH Retinamidophenyl glucuronide
R O HO HO O HO
N H
R = H Retinamidobenzyl xylose
R = CH2OH Retinamidobenzyl glucose R = COOH Retinamidobenzyl glucuronide (4-HPRCG)
Figure 3.4 C-glycoside analogs of 4-HPROG.
pattern. Reduction followed by separation gives the p-amino isomer 3.7. Once the
aniline 3.7 is obtained, it is easily coupled to retinoic acid by the same procedure used to
make 4-HPROG to make the target 4-HPRCG.122,127 Finally, saponification of the esters
yields the final product, 4-HPRCG.
All the various C-glycoside 4-HPROG analogs (Figure 3.4) synthesized were prescreened for their ability to inhibit the growth of MCF-7 cells.137 In general, the C-
linked benzyl analogs were more active than the phenyl analogs in the in vitro assay.
Retinamidobenzyl glucoside and glucuronide were the most active analogs and were
further analyzed for chemopreventative action in 7,12-dimethylbenz[a]anthracene
(DMBA)-induced rat mammary tumors.137 Generally in a chemopreventative study, the
animals are administered the treatment retinoid (1-2 mmol/kg diet) for 10 days prior to a 71 single dose of DMBA (15 mg) after which the retinoid treatment continues until the
experiment endpoint. Due to the small quantities of compound synthesized, the animals
were fed with the retinoid for 10 days after DMBA administration, which is only long
enough for the initiation phase of carcinogenesis. The benzyl glucoside and glucuronide
compounds showed a tumor latency (time to appearance of first tumor) of 71 days (42
days for control) and a tumor incidence (number of rats with tumors) of 42 % (92 % for
control). Specifically, 4-HPRCG was most effective in preventing tumor multiplicity
(number of tumors per rat) at 0.67 (1.5 for control and 0.83 for 4-HPROG). Although
this study was preliminary in nature, certain conclusions could be drawn. One, it was not
surprising that 4-HPRCG showed the most potency because it is spatially similar to 4-
HPROG. Furthermore, the potency of these C-linked molecules is not likely due to the
liberation of 4-HPR from in vivo hydrolysis. The observed activities of these analogs
indicate that the glucuronide O- and C-conjugates may indeed be active agents.
Although the synthesis was successful for the production of 4-HPRCG, there are disadvantages to this route. In order to perform a more complete chemopreventative assessment of 4-HPRCG, improvements of some of the key steps were needed to generate sufficient quantities. The main disadvantages of the synthetic route are low yielding steps such as the benzyl Grignard reaction, removing benzyl protecting groups, the oxidation of the C-glucose to the glucuronide, and the aryl nitration reaction. This synthetic route was improved upon by the modification of several of the key steps.138
The new developments were made to improve access to the key intermediate 3.5. The first step to be improved was the debenzylation reaction. Trimethylsilyliodide (TMSI)139 was used to remove the benzyl ether protecting groups in 1 hour instead of using the low 72 yielding catalytic hydrogenation. Another problematic step from the previous route is the
oxidation of the 6-position hydroxyl group of deprotected 3.4 using Adams’ catalyst,
which was very tedious and took up to 21 days to maximize the yield (~ 20 %). Progress
was made when the selective oxidation was accomplished with 2,2,6,6-tetramethyl-1-
piperidinyloxy free radical (TEMPO). This improvement was not only in yield but also it
is inexpensive compared to the platinum catalyst. Another disadvantage of the previous
synthesis is the generation of inseparable nitro isomers 3.6 arising from the aromatic
nitration reaction. However, upon reduction to the amino functionality, the isomers were
readily separable to give 3.7 and the ortho isomer. A further modification was
demonstrated when the unwanted ortho-amino group of ortho-3.7 could be easily removed with dediazotizing conditions (BF3·Et2O, t-butylnitrite, H3PO2) which regenerated more quantities of 3.5. These modified steps improved the yield of the C- benzyl glucuronide 3.5 by 10 times.
These improvements allowed for the quantities needed to facilitate a long term chemopreventative study with 4-HPRCG.120 After 80 days of retinoid feeding post
DMBA administration, tumor latency was 55 days for 4-HPRCG (48 days for 4-HPROG
and 40 days for control), tumor incidence per rat was 27 % for 4-HPRCG (57 % for 4-
HPROG and 79 % for control), and tumor multiplicity was 0.36 per rat for 4-HPRCG
(0.71 for 4-HPROG and 1.43 for control). This study supported the preliminary
chemopreventative study in that 4-HPRCG is a more effective agent for the prevention of
mammary tumors than 4-HPR and its O-glucuronide. Also, body weights were
monitored as an indication for toxicity (chronic retinoid toxicity produces weight loss)
and it showed similar weights for all groups. Also reported, the binding of 4-HPROG 73 and 4-HPRCG to retinoic acid receptors were measured and they confirmed no binding
for either analog or 4-HPR. All these data suggests that the stable C-linked analog of 4-
HPROG have a chemopreventative advantage over the O-glucuronides or free retinoids due to their increased potency.
3.4 REDESIGN OF SYNTHETIC ROUTE TO 4-HPRCG
With the demonstration of the excellent chemopreventative actions of 4-HPRCG, the possibility of it being an effective chemotherapeutic agent needed to be investigated.
However to regenerate sufficient quantities, the modified synthetic route to the target C- glucuronide still warranted further improvements. One reason is that the tetra-O-benzyl glucose starting material 3.1 for the previous route is fairly expensive, approximately
$20/gram (Scheme 3.1). Also, the electrophilic substitution reaction of benzyl Grignard reagent to produce 3.3 is inefficient, prone to rearrangement reactions,140 and gives poor
yields. In general, benzyl Grignard reagents react poorly due to the prominent formation
of the Würtz coupling side product, which would be diphenylethane in this case.
Furthermore, the production of regioisomers in the phenyl ring nitration gives only 20 %
yield over two steps to the para-amino analog 3.7. With a goal of improving the
efficiency of the synthesis, a redesign of the route was undertaken.
3.4.1 CARBON MONOXIDE INSERTION ROUTE
Two possible routes were proposed for the redesign of the 4-HPRCG synthesis
involving different pathways to obtain the p-amino-C-glycoside intermediate 3.7 (Figure
3.5). The first route explored was the carbon monoxide insertion pathway due to 74 O OH O HO HO O OH N H 4-HPRCG
O Me O O AcO AcO + AcO Cl R
NH2 CO Insertion / 3.7 Organocupprate Suzuki Coupling Pathway Pathway
OMOM O Me O O AcO MOMO AcO MOMO AcO MOMO
3.5 3.10 NO2
M OMOM Br O Me O O AcO + MOMO + AcO Br MOMO AcO MOMO NO 3.8 3.11 2
OH O Me O O CO HO AcO OAc + AcO HO AcO HO O 3.9 3.12
Figure 3.5 Proposed retrosynthetic routes to 4-HPRCG.
75 concurrent work in that area in our lab. From the work of Chatani and others,141 it was
shown that a 1-position acetoxy group on a carbohydrate could be displaced with a
siloxymethyl group using carbon monoxide, catalytic octacarbonyl dicobalt (Co2(CO)8) and an alkylsilane. This work was done with glucosyl analogs but not with glucuronide type analogs. To obtain the necessary glucuronide, the very inexpensive D-glucurono-
6,3-lactone 3.13 was used as a starting material (Scheme 3.2). The cyclized sugar can be opened by saponification and protected to give the β-anomer glucuronide 3.9 after crystallization.142 Although Chatani demonstrated that the mixture of 1-position epimers
can be used in the CO insertion to yield the β-anomer product preferentially, only the β-
glucuronide was used in this case. Using procedures adapted from Spak and others,143 our lab demonstrated that glucuronides react fairly well in the carbon monoxide insertion to give the siloxymethyl β-C-glucuronide 3.14.
HO H OMe O 1) NaOMe, MeOH O OOH O 2) HClO , Ac O AcO OAc O 4 2 AcO H OH 3) crystallize AcO 3.13 20 % 3.9
OMe HSiEt2Me, CH2Cl2 O O AcO Co2(CO)8, CO AcO AcO OSiEt2Me
3.14
Scheme 3.2 Synthesis of siloxymethyl-C-glucuronide 3.14.
76 However, the CO insertion reaction turned out to be very problematic. Initial
successful reactions were done by our collaborators with the Co2(CO)8 catalyst supplied
by Strem® chemicals (Newburyport, MA). Subsequent attempts were done using catalyst supplied by Aldrich® (Milwaukee, WI). The source of materials was found to be paramount for the success of the siloxymethyl insertion. Before the importance of the catalyst source was realized, several unsuccessful reactions were done using the Aldrich® supplied catalyst. Numerous modifications of the reaction conditions were made including reaction time, carbon monoxide pressure, order of addition, catalyst stoichiometry, catalyst air-free crystallization, and fresh preparation of other non-gaseous reagents. None of these changes that were made to the CO insertion using the Aldrich® supplied catalyst increased the yield of 3.14 by more than 7 %.
Once the siloxymethyl C-glycoside would be attained, easy conversion to the bromomethyl material 3.8 could be accomplished with dibromotriphenyl phosphorane.
At the same time of the CO insertion studies, work also began on modeling the addition of a phenyl group to an alkyl bromide. Seen in Figure 3.5, the proposed CO insertion route to 4-HPRCG included addition of a phenyl anion equivalent to the bromomethyl-C- glucuronide 3.8 to yield C-glycoside 3.5. Once the C-arylmethyl compound 3.5 would be obtained, similar chemistry would eventually be used to couple the aniline 3.7 and
retinoic acid to make the target 4-HPRCG.
The explored approach to generate a phenyl anion equivalent was with the use of
organocuprates. The uses of organocopper reagents in substitution reactions have a wide
scope and typically are generated with alkyl halides and metal exchange reactions.144 To
determine if this approach was viable, several model reactions were done (Figure 3.6). In 77 R R-Li CuX Br R CuX Li 2 Br Li
Figure 3.6 General scheme for organocuprate formation and alkylation.
general, bromobenzene was reacted with a lithium base followed by the addition of a copper salt, which forms the organocuprate, and this is alkylated with an alkyl bromide.
The bromobenzene was a model for a potential p-amino- or p-nitrobromobenzene compound and the alkyl bromide compound was a model for the bromomethyl sugar 3.8 to be used in the real synthesis. The first step in the model reactions is to generate phenyl lithium, which was attempted with sources including n-butyl lithium, t-butyl lithium, and lithium metal. Later, phenyl lithium was purchased for use in the organocuprate generation. Once the phenyl lithium was obtained, copper salts, such as CuBr, CuI, and
CuCN, were added to form either homocuprates or heterocuprates, which have different reactivities in various situations. The copper halide and the copper cyanide salts would produce homocuprates and heterocuprates, respectively. Lastly, alkyl bromides, such as ethyl-3-bromopropionate, ethyl-2-bromoacetate, and benzyl bromide, were added for the alkylation with the organocuprate. The results of the model reactions indicated the generation and reactivity of the organocopper reagents was problematic. Furthermore, it was never confirmed whether the model of bromobenzene could make the appropriate organolithium reagent, which would be important for the substituted phenyl rings in the desired synthesis. Although numerous variations were attempted, none of the model reactions showed enough promise to pursue further. From the above studies and 78 difficulties with the siloxymethylation reaction, it was determined that the CO insertion
pathway was not feasible for the production of 4-HPRCG.
3.4.2 SUZUKI COUPLING ROUTE
Attention for the redesign of the 4-HPRCG synthesis was turned to the Suzuki
coupling route due to some interesting papers by Johnson and others wherein C- arylglycosides were made from various Suzuki coupling reactions.145,146 In their study,
numerous types of aryl halides were coupled with a boronated sugar. One particular
entry piqued our interest in which they used p-aminobromobenze. This was applicable to
our case because it would give direct access to the carbon skeleton of key intermediate
3.7. Therefore, making 4-HPRCG via the Suzuki coupling pathway was undertaken147
(Figure 3.5).
3.5 SYNTHESIS OF 4-HPRCG
To obtain materials needed for the Suzuki coupling, known procedures were followed.146 The starting material for this route was δ-gluconolactone 3.12 (Scheme 3.3),
which is very inexpensive at $0.70/gram. This hydrolytically sensitive lactone was
protected with methoxymethyl (MOM) groups using mild conditions. MOM protecting
groups are advantageous due to their ease of assembly and removal. The hydroxyls were
derivatized using MOMCl and Hünig’s base ((i-Pr)2NEt), which is used to insure the absence of free hydrolytic protons to yield the lactone 3.15.
The protected lactone 3.15 was then subjected to an olefination reaction.
Generally, alkenylations are done with Wittig or Horner-Emmons type ylides when the 79 OH OMOM 4 6 O MOMCl, (i-Pr)2NEt O HO MOMO HO Bu NI, CH Cl MOMO HO O 4 2 2 3 MOMO O 83% 3.12 3.15
OMOM 46 Cp2Ti(CH3)2 O MOMO MOMO PhCH3 MOMO 3 1' 87% 3.11
Scheme 3.3 Synthesis of MOM-protected enol ether sugar 3.11.
reactant is a ketone or aldehyde. For carboxylic esters and lactones, Tebbe’s or Petasis’
reagents convert esters to the corresponding enol ethers in good yields (Figure 3.7).
Early studies were done by RajanBabu and others which demonstrated the use of Tebbe’s
reagent to perform olefinations of sugar lactones.148 Although Tebbe’s reagent is generally efficient in these reactions, it was shown that Petasis’ reagent was preferred in the case of sugar lactones.149,150 Another advantage of Petasis' reagent is that it is much more stable to air and moisture than Tebbe’s reagent. The theoretical mechanism of the olefination reaction is energetically driven by the formation of the strong titanium- oxygen bond.
Since Petasis’ reagent (dimethyl titanocene) is stable to air and moisture, preparation of the reagent can be done without the use of specialized laboratory techniques. Titanocene dichloride (CpTiCl2) is methylated with an excess of methyl
lithium, followed by an aqueous workup. After concentration, the orange solid can be
stored as a toluene solution at 0 ºC in the dark for at least two weeks without significant 80 Al Ti Ti Cl
Tebbe's reagent Petasis reagent
R' R' O O R' O O O O R R R' + Ti O RO Cp Cp R Cp Ti CH3 Cp Ti CH2 Cp CH3 Cp
Figure 3.7 Stucture and theoretical mechanism of ester alkenylation reagents.
decomposition. The 0.5 M solution of dimethyl titanocene was used without
characterization to olefinate the protected gluconolactone 3.15 to give the enol ether 3.11
in good yield (Scheme 3.3). The enol ether 3.11 is sensitive to hydrolysis and can
degrade over time in air. However, it is stable enough for expedient silica gel
chromatography. Upon isolation, characterization data of the exocyclic olefin matched
the published report.146 Identification of the newly generated methylene protons was
done by comparison of the 1H NMR spectra of 3.15 and 3.11 (Spectrum 3.1).
The next step was the key Suzuki-Miyaura coupling reaction,151 which is a palladium catalyzed cross coupling of an organoborane and an aryl halide to form a bond between sp2 and sp3 carbon atoms (Figure 3.8). Liganded palladium (0) is the source
reagent and in a rate limiting step, it is oxidized to palladium (II) by the addition of an
aryl halide, which displaces two original ligands. Next, the complex is transmetallated
81 MOM-groups
MOM-groups
2-H 3,4,5,6-H 3.15
3,4,5,6-H 1'-H 2-H 3.11
4.9 4.8 4.7 4.6 4.5 4.4 4.3 4.2 4.1 4.0 3.9 3.8 3.7 3.6 3.5 3.4 3.3 3.2 ppm
1 Spectrum 3.1 400 MHz H NMR spectra of lactone 3.15 and enol 3.11 in DMK-d6.
with an oganoboron reagent. The source of the organoborane is usually from the hydroboration of an alkene or alkyne. Various bases are used with the organoborane and the base is thought to increase the nucleophilicity of the alkyl group to facilitate transfer to the palladium. Reductive elimination of the organic partners produces the carbon- carbon bond and also regenerates the palladium (0) complex, which can enter back into the reaction cycle.152,153
In our case, the first coupling reactions attempted were with p-bromoaniline. This model was chosen in order to mimic conditions reported with p-bromoaniline and a benzyl protected analog of 3.11.145 Our attempts were successful in making the amino analog of 3.10 (Figure 3.5) with yields similar to reported amounts. However, the aniline
82 L ArBr Alkyl-Ar 0 L Pd L
L
Reductive elimination Oxidative addition
L L
+2 +2 Alkyl Pd Ar Br Pd Ar
L L Transmetallation
Br BR2 Alkyl BR2 + Base
Figure 3.8 General catalytic cycle for the Suzuki-Miyaura cross-coupling.
was not desired at this stage in the synthesis due to its reactivity in subsequent oxidative
chemistry. Therefore, with these results, work began on optimization of the coupling
with 1-bromo-4-nitrobenzene. In a one pot sequence, the organoborane was prepared by
hydroboration of the exocyclic double bond of 3.11 with 9-borabicyclo[3.1.1]nonane (9-
BBN). Next, addition of base was followed by addition of palladium catalyst (PdCl2
(dppf)) and aryl bromide, which yielded the C-glucoside 3.10 (Scheme 3.4). It was postulated that with an electron withdrawing nitro group on the aryl bromide, the yields would exceed the reported yield for p-bromoaniline. However, the yields for the coupled product 3.10 were lower than with the aniline trials from above. Therefore, efforts were 83 OMOM OMOM 1) 9-BBN-H, THF O O MOMO MOMO MOMO MOMO MOMO 2) PdCl2(dppf), K3PO4, DMF MOMO 1-bromo-4-nitrobenzene NO2 3.11 54% 3.10
Trial Hydroboration Suzuki coupling Yield
time temp 9-BBN time temp Pd base Br
1 5 h reflux 2.5 eq 18 h 25 °C 4 mol% 3 M K3PO4 1.1 eq 50 % 2.1 eq
2 6 h reflux 2.5 eq 18 h 25 °C 8 mol% 3 M K3PO4 1.1 eq 54 % 2.4 eq
3 4.5 h reflux 2.5 eq 18 h 25 °C 8 mol% 3 M Cs2O3 1.1 eq 54 % 2.4 eq
4 18 h reflux 2.5 eq 18 h 25 °C 10 mol% 3 M K3PO4 1.1 eq 38 % 2.4 eq
5 6 h reflux 2.5 eq 18 h 50 °C 4 mol% 3 M K3PO4 1.1 eq 29 % 2.1 eq
6 5 h reflux 2.5 eq 18 h 25 °C 8 mol% 3 M K3PO4 2 eq 29 % 2.4 eq
Scheme 3.4 Study of the Suzuki coupling reaction conditions.
made to improve the efficiency of the coupling by varying conditions, which included reaction time, temperature, stoichiometry, and bases. Also seen in Scheme 3.4 are the results from this study, where equivalency is based upon 1 equivalent of the olefin 3.11.
It was concluded from this study that the conditions (Trial 2) originally reported by
Johnson and others were indeed optimal.145
Another advantage of the Suzuki coupling route is that the stereochemical course of the hydroboration reaction is selective. It has previously been demonstrated that the
84 hydroboration of a perbenzylated derivative of 3.11 with 9-BBN, followed by oxidative
workup, resulted in the detection of the β isomer only.148 The stereochemical outcome of a Suzuki coupling was also shown to be selective using olefin 3.11 and 3-iodotoluene.146
The stereoselectivity is dictated by the sterically hindered 9-BBN. The hydroboration follows a cis anti-Markovnikov addition of hydrogen and boron. Approach of the boron can occur either from the top or bottom face of the double bond (Figure 3.9). The top face of the molecule is more sterically hindered than the bottom face due to the axial 2- position proton. While the bottom face has two axial protons, they are both attached to carbons two atoms away and present less of a steric barrier. With the coordination of the
H B 9-BBN
O O O O
H O O O O O O O O O H O O O H O 3.11 O H B δ+
H B O O O O O O O B O O H
Figure 3.9 Theoretical mechanism of hydroboration with 9-BBN. 85 9-BBN to the double bond, a partial positive charge builds on the boron inducing the
addition of hydrogen and boron, giving the organoboron reagent used in the Suzuki
coupling reaction.
Once the β-C-glucoside 3.10 was obtained, subsequent chemistry focused on
conversion to the glucuronide analog (Scheme 3.5). Deprotection of the MOM groups
was easily accomplished in dilute acid, which exposes all the hydroxyl groups. Next, the
crude product 3.16 was oxidized at the 6-position alcohol to the carboxylic acid, which is
accomplished selectively using 2,2,6,6-tetramethyl-1-piperidinyloxy free radical154,155
(TEMPO) as previously described.138 The catalytic TEMPO selectively reacts with the
OMOM OH O O MOMO HCl HO TEMPO, NaOCl MOMO HO MOMO OH MeOH KBr, NaHCO3 NO2 NO2 3.10 3.16
OH OMe O O O HCl (g) O HO HO HO HO OH MeOH OH
NO2 NO2 3.17 3.18
OMe Ac O, DMAP O 2 O AcO pyridine AcO AcO
84% NO2 3.19
Scheme 3.5 Synthesis of key intermediate 3.19. 86 primary alcohol due to steric bulk around the piperidine nitrogen. In the reaction mechanism, the stable radical is oxidized to the nitrosonium ion, which is the active oxidant, indirectly by the stoichiometric oxidant NaOCl (Figure 3.10).156 The hydroxyl oxygen from the primary alcohol adds to the charged nitrogen, which is followed by α- proton extraction and this leads to the oxidized product. Addition of secondary alcohols to the nitrogen is much slower than for primary alcohols, which gives rise to its selectivity. When these oxidations are done in aqueous systems, the aldehyde is not detected and the carboxylic acid is isolated.156 The TEMPO N-hydroxide is released and the radical is regenerated, which enters back into the catalytic cycle. The resultant oxidized sugar was generated and the crude product was acidified to give the protonated carboxylic acid 3.17.
R HO
- + + NaOCl OBr N N - O O O - H R Cl KBr H - N OH O TEMPO
O H2O O + N HO R HR OH
Figure 3.10 The catalytic cycle for TEMPO mediated oxidations.
87 With the production of the glucuronide, the crude oxidized product was then
exposed to reprotection chemistry. The first step was to methylate the carboxylic acid
which is facilitated with methanol and gaseous hydrochloric acid to give 3.18. Using
standard conditions, acetylation of the remaining alcohol groups gave the key
intermediate β-C-glucuronide 3.19 (Scheme 3.5). The overall yield over four chemical
transformations without isolation was efficient and consistent. Due to purification issues, the acetate protected 3.19, which is a crystalline material unlike MOM protected 3.10, was used to investigate the stereochemical outcome of the hydroboration reaction. One way to analyze relative stereochemistry is with nuclear Overhauser effect (NOE) spectroscopy. Three steady-state NOE experiments were performed on 3.19 to obtain difference spectra (Spectrum 3.2). Irradiation at H1 (proton assignments accomplished by COSY) resulted in enhancement of H5 (12 %), H3 (9%), H2’ (5 %), and H1’ (4 %).
Irradiation at H3 resulted in enhancement of H5 (7 %) and H1 (6 %). Irradiation of H5 resulted in enhancement of H1 (10 %) and H3 (8 %). These results are consistent for the
β-isomer structure and are also consistent with other reports of the stereoselectivity of the
hydroboration-Sukuki reaction with the sugar enol ether 3.11.146
After facile reduction of the nitro group of 3.19, the final glycoside 3.7 was
obtained (Scheme 3.6). Coupling of the aniline and the retinoid was accomplished by the
known procedure previously used.122,129 Dried retinoic acid in pyridine was exposed to
freshly distilled thionyl chloride, which generated the acyl chloride 3.20 in situ. Careful
attention to stoichiometry and use of excess base insured the absence of hydrochloric acid
by forming pyridine hydrochloride. The C-glucuronide-retinoid conjugate 3.21 was
obtained and isolated by crystallization. This was important when large quantities of the 88 OAc OMe water 4-H 3'-H 2'-H 3-H 2-H 5-H 1-H 1'-H 3.19
Irradiation at H-3
Irradiation at H-5
Irradiation at H-1
8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm
OMe O 5 O AcO 1' 2' AcO 3' AcO 3 1 2' NO2 3.19 3'
Irradiation at H3 Irradiation at H5 Irradiation at H1
OMe OMe OMe O O O H O H H H O H H H O H H AcO AcO AcO AcO AcO AcO AcO AcO AcO H H H H H H H H H H NO H NO2 H NO2 2
Spectrum 3.2 Difference spectra from steady state NOE experiments on C-glucuronide 3.19 in DMK-d6 at 400 MHz.
89 OMe OMe O Pd / C, EtOAc O O O AcO AcO AcO H2 (40 psi) AcO AcO AcO
3.19 NO2 98% 3.7 NH2
+ O O thionyl chloride Cl OH pyr, THF Retinoic acid 3.20
86% O OMe O AcO AcO O AcO N H 3.21
Scheme 3.6 Synthesis of glucuronide-retinoid conjugate 3.21.
retinoid conjugate were generated because chromatography could be avoided. Even though the solid was extensively washed with hexanes, a small amount of retinoic acid was present in the mixture. Since retinoic acid is extremely potent in receptor binding experiments, trace amounts would give false positive results in cellular assays. Since the product 3.21 and retinoic acid poorly separate on silica gel chromatography, the mixture was exposed to diazomethane to generate methyl retinoate, which is easily separated from 3.21. The amount of the methyl retinoate generated was 0.8 % of the total material,
as estimated by HPLC. However, the vast majority of material used in the subsequent
animal studies was washed with an aqueous sodium bicarbonate solution, which reduced
the amount of retinoic acid contamination to 0.2 % of the total material. The remaining 90 amount of retinoic acid contamination was acceptable for the forthcoming
chemotherapeutic studies because it only shrinks tumors at high doses and the remaining
retinoic acid contamination would not affect the size of the tumors.
Finally, the C-glucuronide-retinoid conjugate 3.21 was deprotected using a mild two-step hydrolysis (Scheme 3.7). In a one pot procedure, catalytic potassium carbonate in methanol generates methoxide ion, which transesterifies the acetate protecting groups to form methyl acetate and the triol 3.22. Next, aqueous potassium hydroxide is added to saponify the methyl ester of 3.22 to yield the final product 4-HPRCG, which was purified by reverse phase chromatography (Spectrum 3.3). In the end, approximately 6.5 grams
O OMe O AcO AcO O AcO K2CO3, MeOH N H 3.21
O OMe O HO HO O HO KOH, MeOH N
H 3.22 86 %
O OH O HO HO O HO N H 4-HPRCG
Scheme 3.7 Final steps in the synthesis of 4-HPRCG. 91 16,17-H
O OH 4' O H bz H 4,19-H HO ar 20 19 HO O 16 17 3' HO 1' 11 7 N 1 14 9 H 18 3
4-HPRCG 20-H 18-H
7,8,10,12,14-H Har 2,3-H Hbz,1',2',3',4',5'-H Hbz N-H 11-H
9.5 9.0 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 ppm
1 Spectum 3.3 400 MHz H NMR spectrum of 4-HPRCG in DMK-d6.
of 4-HPRCG was made with an overall yield of 24 % over 13 chemical transformations.
This was more than a 10 fold increase from the previously reported overall yield of 2.2
%.
3.6 BIOLOGICAL EVALUATION OF 4-HPRCG
To investigate 4-HPRCG as a chemotherapeutic agent, female Sprague-Dawley rats were administered the highly effective mammary carcinogen DMBA as previously described.124,128 The tumors were allowed to fully develop for 50 days before drug
treatment was given. The tumor bearing rats were randomly divided into 4 groups, 12 92 rats per group, which consisted of control, retinoic acid, 4-HPR, and 4-HPRCG treatment groups. The treatment retinoids were given as additives in the rat chow at a concentration of 2 mmol/kg diet.
It is known that 4-HPR causes apoptosis in tumor cell lines, but to investigate the apoptotic activity of the treatment retinoids in vivo, three rats per group were sacrificed and their tumors harvested after 10 days of treatment. These tumors will subsequently be analyzed to determine apoptotic status. Also in the small group of rats, other aspects were explored including drug concentration in tissues and plasma and effects on endogenous plasma retinol concentration and triglyceride levels. At this first 10 day time point, tumor volumes showed a decrease of 40, 25, and 21 % for retinoic acid, 4-HPR, and 4-HPRCG, respectively and control tumors increased by 224 %.147 It is important to note that retinoic acid does shrink tumors at high doses, however, the toxicity at these effective doses was very significant. While only a few rats were evaluated and variability was high in the retinoic acid treatment group, total triglycerides were observed to be elevated by treatment with retinoic acid. 4-HPRCG treated rats had triglyceride levels identical to control rats, indicating low toxicity (Figure 3.11). The rest of the cell based assays and pharmacokinetic analyses are yet to be performed.
The remaining animals, nine rats per group, received treatment for a total of 24 days, after which they were sacrificed. Plasma was collected, tumor volumes and liver weights were measured, and femurs were taken to check for effects on bone mineral density, which decreases with retinoic acid chronic toxicity. At this 24 day end point, tumor volume decreased 51, 49, and 48 % for retinoic acid, 4-HPR, and 4-HPRCG, respectively (Table 3.1). An increase in liver volume and weight is a sign of chronic 93
450
350
L / d 250
mg
150
50
Control Retinoic acid 4-HPR 4-HPRCG
Figure 3.11 Effect of retinoid treatment on plasma total triglyceride levels.
toxicity by retinoic acid and 4-HPRCG treated rats had liver weights similar to control
(Table 3.2). Other important cellular assays of 4-HPRCG action are underway which include binding to nuclear retinoid receptors and determination of apoptosis and growth inhibition effects in MCF-7 cells in culture.
The data for the chemotherapeutic experiment and cellular assays with 4-HPRCG were generated by our collaborators, Drs. Abou-Issa, Alshafie (Ohio State University) and Clagget-Dame (University of Wisconsin-Madison).
94 Average volume (cm3) Number of Group % Change tumors Initial volume Final volume
Control 34 0.214 ± 0.05 0.385 ± 0.07 + 180
Retinoic acid 33 0.410 ± 0.10 0.202 ± 0.04 - 51
4-HPR 34 0.631 ± 0.18 0.321 ± 0.11 - 49
4-HPRCG 22 0.656 ± 0.23 0.341 ± 0.14 - 48
Table 3.1 Effect of retinoid treatment on DMBA-induced rat mammary tumor volume.
Group Endpoint (day) Mean weight ± SD (g)
10 8.08 ± 0.99 Control 24 9.19 ± 1.02
10 8.02 ± 0.88 Retinoic acid 24 11.60 ± 3.36
10 9.54 ± 1.65 4-HPR 24 15.56 ± 3.64
10 6.01 ± 0.41 4-HPRCG 24 9.85 ± 1.23
Table 3.2 Effect of retinoid treatment on liver weights.
95 3.7 CONCLUSIONS
The redesign of the synthesis for the important retinoid 4-HPRCG was
accomplished yielding a much more efficient route. One reason is that the starting
material for the current synthesis, 3.12, is considerably less expensive than the previous
sugar 3.1, $0.70 versus $20 per gram. Also, immediate access to the p-nitro isomer by
the Suzuki coupling is advantageous over the tedious reduction and separation of nitro
and amino regioisomers, respectively. Furthermore, with new steps and modifications to
previously used steps, the overall yield of 4-HPRCG was increased from 2.2 % to 24%.
With the material generated from the new route, the chemotherapeutic activity of
4-HPRCG was able to be investigated. It was shown to be just as effective as 4-HPR at
shrinking mammary tumors in rats. However, it did have significantly less toxicity
indicated by the serum triglyceride levels and liver weights. Further details on the
biological actions of this important retinoid will be forthcoming. But since this important retinoid is now easier to obtain, it will facilitate these further biological studies. The combination of a more effective chemopreventative, equieffective chemotherapeutic, and decreased toxicity, make 4-HPRCG a better anticancer agent than 4-HPR.
96
CHAPTER 4
SYNTHESIS OF THE FULLY C-LINKED ANALOG OF 4-HPR-O-GLUCURONIDE
4.1 RATIONALE
Even though the retinoid 4-HPR has been shown to be an effective chemopreventative and therapeutic with very low toxicity, some of its side effects may be attributed to in vivo hydrolysis of the amide bond, liberating retinoic acid. To investigate this possibility, an unhydrolyzable analog of 4-HPR, 4-hydroxybenzyl retinone (4-HBR) was proposed and synthesized in our lab.157 Figure 4.1 shows the structures of pertinent
4-HPR analogs.
Through a series of published and unpublished in vivo and in vitro experiments by
our collaborators, it was demonstrated that 4-HBR and 4-HPR have similar biological
actions as antiproliferative agents.124,158 It was shown that 4-HPR and 4-HBR possess
equipotent growth inhibition and apoptotic activity in MCF-7 cells. In a DMBA-induced
rat mammary tumor model, both agents were shown to be equiactive chemotherapeutics.
Another set of important experiments was with vitamin A deficient (VAD) rats.158
Typically, normal rats are depleted of vitamin A over several weeks by feeding a vitamin
97 OH O X
X = NH 4-HPR
X = CH2 4-HBR
O OH O HO HO O HO X
X = NH 4-HPRCG
X = CH2 4-HBRCG
Figure 4.1 Structures of 4-HPR and analogs.
A-free diet, after which the investigated agent is administered.159,160 Compounds exhibiting classic retinoid characteristics will support growth of the deficient animals, thus behaving as retinoic acid. Initial experiments were done with retinoic acid and 4-
HPR and when both agents were given at equal doses, 4-HPR did not support growth.
However, when 4-HPR was given at chemotherapeutic doses, it did support growth in the
VAD rats, while 4-HBR at equal therapeutic doses was not active in maintaining growth.
Subsequently, lung extracts from retinoid treated VAD rats were measured for induced transcription of the messenger RNA for cyp26, an enzyme that metabolizes retinoic acid to 4-oxo-retinoic acid. Induction of cyp26 occurs through retinoic acid receptor activation.161 Extracts from 4-HPR treated rats induced more cyp26 mRNA than 4-HBR extracts, which indicates the production of retinoic acid from the metabolism of 4-HPR.
Also, plasma taken from the treatment groups of these VAD animals was further assayed for the existence of retinoic acid. Plasma preparations were applied to cells containing a 98 retinoic acid receptor-luciferase construct. Plasma containing trace amounts of retinoic acid would show light emission due to construct activation. These experiments revealed that plasma from 4-HPR and 4-HBR treated rats indeed contained and confirmed the absence of retinoic acid, respectively. The above data strongly indicates that 4-HPR does liberate small quantities of retinoic acid in vivo and that 4-HBR does not liberate retinoic acid. Furthermore either retinoid does not interact with the nuclear receptors to induce apoptosis.
Because retinoic acid is potent teratogen, a great concern for any retinoid is the possibility of the analog being teratogenic. 4-HPR has been shown to be 100 times less teratogenic than retinoic acid and the slight toxicity may be caused by the liberation of retinoic acid. For the effective antiproliferative agent 4-HPRCG, amide bond hydrolysis, at least in theory, could still occur in vivo liberating retinoic acid by similar mechanisms as for 4-HPR. To investigate this possibility, the C-linked analog (4-HBRCG) was proposed and replacement of the amide bond of 4-HPRCG with a methylene group would give the fully C-linked derivative of 4-HPR-O-glucuronide (Figure 4.1). Like 4-HBR, this ketone-type analog should prove to be as effective as the amide and possibly non- teratogenic, which is a very important goal for any retinoid. Furthermore, since the glucuronide conjugates, 4-HPROG and 4-HPRCG, are more potent than the parent molecule, 4-HPR, 4-HBRCG may prove to have the highest therapeutic index of any retinoid in this series of analogs.
99 4.2 RETINOID C-GLYCOSIDES
Another major advantage of using the Suzuki coupling pathway for the synthesis of 4-HPRCG is the versatility of the route to make other C-glycosides. Seen in Figure
4.2 are the retrosynthetic analyses for other retinoid C-glucuronides of interest. Using the enol ether 3.11 as a key intermediate, Suzuki coupling with p-nitrobromobenzene gives the carbon skeleton of the C-glycoside 4.1, which was used to make 4-HPRCG as demonstrated in the previous chapter. The C-linked analog of the natural glucuronide metabolite of retinoic acid also could be made using these methodologies. C-linked retinoyl β-glucuronide has previously been made in our lab.162 However, the synthetic
route was very inefficient and yielded only milligram quantities containing double bond
isomers. Starting from enol ether 3.11, hydroboration with an oxidative work up,
followed by halogenation could give the bromomethyl sugar 4.2. This bromide could be
used to alkylate a retinal equivalent, which would give ready access to retinoyl-β-C-
glucuronide. Furthermore, this chemistry can also be used to make the fully C-linked
analog 4-HBRCG. After a Suzuki coupling reaction with 3.11 and an aryl halide,
generation of benzyl bromide 4.3 would be followed by alkylation of a retinal equivalent to yield the target 4-HBRCG. In all cases, the glucoside analogs of the retinoids can readily be obtained using the same chemistry.
4.3 SYNTHESIS OF 4-HBRCG
While redesigning the synthesis of 4-HPRCG, another entry in the report by
Johnson and others145 was noticed in which 3-iodobenzyl alcohol was used to make a C- aryl-glycoside benzyl alcohol. This piqued our interest because this chemistry could 100 OMOM R R O O O AcO MOMO AcO AcO MOMO AcO AcO MOMO AcO Br 4.1 NO2 3.11 4.3
R O AcO AcO AcO Br 4.2
R O HO HO HO O Retinoyl-β-C-glucuronide R O HO HO O HO N H 4-HPRCG
R O HO HO O HO
4-HBRCG
R = CH2OH, COOH
Figure 4.2 Retrosyntheses of retinoid C-glycosides.
101 potentially be applied toward the synthesis of 4-HBRCG. Therefore, starting from δ- gluconolactone, the enol ether 3.11 was obtained and used to perform Suzuki reactions with 4-halobenzyl alcohols (Scheme 4.1). Trials with both bromo- and iodobenzyl alcohols were done, which showed no significant difference in the yield of the coupled product 4.4. Due to expense, 4-bromobenzyl alcohol thus was regularly used as the aryl halide. The coupling was done without the need to protect the alcohol, demonstrating the versatility of the Suzuki reaction. However, because of the planned subsequent chemistry, the C-glucoside benzyl alcohol 4.4 was easily protected as the methyl ether to give 4.5.
OMOM OMOM 1) 9-BBN-H, THF O O MOMO MOMO 2) PdCl (dppf), 3 M K PO , MOMO MOMO MOMO 2 3 4 MOMO DMF, p-bromobenzyl alcohol OH 3.11 67% 4.4
OMOM NaH, THF O MOMO MOMO CH3I MOMO OMe 90% 4.5
Scheme 4.1 Synthesis of C-glucoside methyl ether 4.5.
As in the 4-HPRCG synthesis, the C-glucoside needed to be converted to the C- glucuronide. Deprotection of the MOM groups was accomplished in dilute acid and the crude product 4.7 was oxidized at the 6-position with TEMPO (Scheme 4.2).
102 Unfortunately, the conditions used for the oxidation of the p-nitro analog 3.16 resulted in
a complex mixture of numerous products. As a result, many trial oxidations were done to
find the optimal conditions. Variations in the time, temperature, base, amount of
TEMPO, amount of sodium hypochlorite, and order of addition were tried without any
success in cleanly generating the desired compound. Due to the oxidative sensitivity of
benzylic centers, it was assumed that one or more benzylic carbons were being oxidized.
Potential compounds such as 4.10 and 4.11 were suspected to be present in product
mixtures (Scheme 4.2). From past experience in our lab, other oxidatively sensitive
sugar-type molecules can undergo selective oxidations when excess instead of catalytic
TEMPO is used.162 When excess amounts of TEMPO, KBr, and NaOCl are premixed in a NaHCO3 solution, the deprotected sugar 4.6 can be added and selectively oxidized,
efficiently yielding the sodium salt 4.7. Upon acidification, the carboxylic acid was
exposed to gaseous hydrochloric acid in methanol in attempt to form the methyl ester, as
previously done. However, these conditions also formed unidentified isomers.
Consequently, the methyl ester 4.8 was formed upon treatment of the sodium salt 4.7 with methyl iodide. Lastly, reprotection of the remaining alcohols was done using standard conditions to afford the protected C-glucuronide methyl ether 4.9. Once the problems with the oxidation and the methylation were solved, the execution of the four chemical transformations without isolation gave consistently high yields (Scheme 4.2).
In the design of this synthesis, the methyl ether protecting group for the benzyl alcohol was chosen for several reasons. One, methyl ethers are robust under a number of conditions, including the acidic conditions needed for deprotection of MOM groups.
Two, methyl ethers are stable to the basic oxidative conditions used in the TEMPO 103 OMOM OH O 6 N HCl, MeOH O MOMO HO MOMO HO MOMO HO OMe OMe 4.5 4.6
ONa TEMPO (4 eq), O NaOCl, KBr, NaHCO O CH3I, DMF 3 HO HO HO OMe 4.7
OMe OMe O O O Ac O, pyridine, DMAP O HO 2 AcO HO AcO HO AcO OMe OMe 82% 4.8 4.9
OMe OMe O O O O AcO O AcO AcO AcO AcO AcO OMe OMe 4.10 4.11 O
Scheme 4.2 Synthesis of the key intermediate 4.9 and possible oxidation by-products.
104 reactions. Lastly, it has been reported that benzylic methyl ethers can be displaced by bromide from hydrobromic acid.163,164 In a dry atmosphere, key intermediate 4.9 was
exposed to a large excess of a saturated solution of HBr in acetic acid, giving the benzyl
bromide C-glucuronide 4.12 (Scheme 4.3, Spectrum 4.1). This surprisingly facile
reaction yielded a very stable benzyl bromide, which was isolated by crystallization.
The next step in this route was the key alkylation of electrophile 4.12 by a retinal
anion equivalent. Aldehyde anion synthons are commonly referred to as Umpolung
derivatives. These synthons are usually produced by derivatizing the carbonyl group,
making the aldehyde proton acidic, and thereby facilitating anion production upon
exposure to basic conditions. Common aldehyde derivatives that serve as acyl anion
sythons are 1,3-dithianes, cyanohydrins, α-dialkylaminonitriles, and α-alkyl-N-
acylaminoacetonitriles.165,166 These are used to perform alkylations with various
electrophiles, and then subsequently deprotected to eventually yield an alkylated ketone.
The most suitable Umpolung strategy for the chemically sensitive retinal is to make the
protected cyanohydrin derivative,167 more particularly, the silylcyanohydrin of retinal.168
The employment of silylcyanohydrins of retinal for the synthesis of 4-HBR has been demonstrated in our lab157 and a great deal of the reaction troubleshooting has been
done.169
These silylcyanohydrins are formed by exposing the aldehyde to a silylcyanide in
the presence of a Lewis acid. 1,2-Addition to the carbonyl with the cyanide group and
the silyl group gives the Umpolung derivative (Scheme 4.4). In the 4-HBr synthesis,
trimethylsilylcyanide (TMSCN) was used to make the cyanohydrin 4.13, which was then
alkylated with a p-siloxybenzyl bromide to eventually yield 4-HBR. One of the problems 105 OMe OMe O O O HBr, AcOH O AcO AcO AcO AcO AcO AcO OMe 86% Br 4.9 4.12
Scheme 4.3 Synthesis of benzyl bromide 4.12.
OAc
OMe O OMe 4 O 1' AcO Har AcO AcO 3 1 Br
4.12 Hbz water
Hbz
Har
2,3,4-H 5-H 1'-H 1-H
7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 ppm
1 Spectrum 4.1 400 MHz H NMR spectrum of glucuronide bromide 4.12 in DMK-d6.
106 O CN trialkylsilylcyanide SiR3 H O NEt3, CH2Cl2 Retinal silylcyanohydrin 84 %
CN CN Si Si O O
4.13 4.14
Scheme 4.4 Synthesis of retinal silylcyanohydrins.
with this route was that 4.13 could not be isolated due to hydrolysis on silica gel. Due to
byproduct formation in the derivatization of retinal, subsequent chemistry was
complicated by purification problems. It was then proposed that a bulkier silicon reagent,
like tert-butyldimethylsilylcyanide (TBDMSCN), could be used to make the retinal
cyanohydrin 4.14, which could be more stable. The cyanohydrin 4.14 proved to be stable
on silica gel, which facilitated isolation, and was formed in consistently good yields.
In the alkylation reaction, the cyanohydrin is exposed to an organic base at low
temperatures (Scheme 4.5). Upon anion formation, which has a very dark red color, the
electrophile is slowly added and left to stir until the anion is quenched. The intermediate
alkylated cyanohydrin is then exposed to fluoride ion to remove the silyl group, resulting
in the production of the alkylated ketone product. Due to the increased bulk of the
eventual nucleophile 4.14, initial model alkylation reactions were done to compare the
reactivity of the trimethyl- and the tert-butyldimethylsilylcyanohydrins 4.13 and 4.14, respectively. Using benzyl bromide as a model electrophile, numerous variables were changed including base (LDA, NaH, LiHMDS, and NaHMDS), additives (HMPA), 107 CN Base CN benzyl bromide SiR3 O SiR3 H -78 °C - O silylcyanohydrin
CN O SiR TBAF O 3
4.15 64 % benzyl retinone
Scheme 4.5 Model alkyation reactions.
stoichiometry, temperature, and time. After addition of tert-butylammonium fluoride
(TBAF) to the crude alkylated product 4.15, benzyl retinone was isolated and on average, it was determined that using the TMS-cyanohydrins gave slightly better yields. However, due to isolation issues, the TBDMS-cyanohydrin was considered to be advantageous and was chosen to be the Umpolung derivative. Therefore, from these model alkylation studies, it was determined the optimal reaction conditions were the use of 4.14, slight excess of LiHMDS as a base, and 2 equivalents of bromide.
Initial alkylations with the bromoglucuronide 4.12 and the cyanohydrin 4.14 were not successful and revealed another variable not examined in the model studies. When following the procedure from the synthesis of 4-HBR,169 which used the
trimethylsilylcyanohydrin 4.13, anion formation occurs within 5 minutes. However, with
the bulkier 4.14, the anion needs at least 30 minutes to efficiently form. Once the anion
was fully formed, the bromide 4.4 was slowly added and allowed to stir until the anion
was quenched (Scheme 4.6). The alkylated cyanohydrin 4.16 was analyzed for stability
108 CN 1) LiHMDS, THF, -78 °C Si O O OMe O 2) AcO 4.12 4.14 AcO AcO Br
47 %
O OMe O AcO Si AcO O TBAF, THF/H2O AcO
CN 75 % 4.16
O OMe O AcO AcO O AcO 1) K2CO3, MeOH 2) KOH, MeOH
4.17 82 %
O OH O HO HO O HO
4-HBRCG
Scheme 4.6 Key alkylation and final production of 4-HBRCG.
109 on 2D TLC and found to be stable. At this point, the alkylated product 4.16 and unreacted bromide 4.4 were chromatographically separable, however, the deprotected ketone 4.17 is poorly separated from the unreacted bromide. Therefore, the alkylated product 4.16 was isolated and the unreacted bromide was recovered, which is important because it could be recycled in subsequent alkylations. Steric bulk of the tert-butyl group was a concern for possibly causing poor alkylation yields. Therefore, alkylations with the trimethylsilylcyanohydrin 4.13 and the sugar electrophiles were explored and it was found that the yields were better. However, since the trimethylsilyl analog of 4.16 was not stable on silica gel, separation of resulting product 4.17 and unreacted bromide 4.12 was tedious with chromatography. Thereafter, the tert-butyl 4.14 was used to generate the nucleophiles in subsequent reactions and, on average, the yields of the alkylations were only 30 %. Next, the ketone was easily unmasked with TBAF to afford the penultimate material 4.17. Lastly, standard ester deprotection chemistry followed by reverse phase chromatography, was used to isolate the final product 4-HBRCG (Scheme
4.6, Spectrum 4.2).
It is important to note that the preparation of the silylcyanohydrin and bromide are paramount to a successful alkylation. The chromatographed 4.14 is required to be actively dried over sodium or magnesium sulfate followed by at least 18 hours of evacuation on the vacuum pump. This insures that residual solvents or moisture are not trapped in the gel-like matrix of the cyanohydrin. In addition, the isolated crude bromide
4.4 is a foam, which can also harbor moisture and residual solvents. Before use in the alkylation, the bromide should be crystallized with ether and subsequently subjected to vacuum at elevated temperatures for at least 18 hours. With both of the reagents 110 16,17-H
water H'bz,1',2,3',4',5'-H
4,19-H
20-H 18-H
Har,11-H Hbz
7,8,10,12,14-H 2,3-H
H'bz
7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm
O OH 4' O H'bz H HO ar 20 19 HO O 16 17 3' HO 1' 11 7
9 1 Hbz 14 3 18
4-HBRCG
1 Spectrum 4.2 400 MHz H NMR spectrum of final product 4-HBRCG in MeOH-d4.
111 specially prepared, alkylations were consistent and were able to produce ample quantities
of the alkylated product.
Before these special preparations were recognized as being important, alkylations
were carried out to yield a small quantity of a deceptively similar ester linked product
4.18 (Figure 4.3). The 1H NMR spectra for the ketone 4.17 and the ester 4.18 were almost identical except for the chemical shifts of the benzylic protons, aryl protons, and the 14-position proton (Spectrum 4.3). At the time, the ester was thought to be the target compound 4.17 and it was subsequently subjected to chemistry used to remove the carbohydrate protection groups. To our surprise, only methyl retinoate was isolated from the organic extract of the reaction. From this result, we assumed the product “4.17” to be very unstable and therefore, a large amount of work went into developing conditions that would mildly hydrolyze the sugar ester groups. When a small amount of deprotected
O OMe O AcO AcO AcO O O O OH O 4.18 HO HO HO O O 4.19
Figure 4.3 Structures of ester byproducts
112 OMe 16,17-H
18-H 20-H
Har,11-H Hbz
7,8,10,12,14-H 2',3',4'-H 5'-H 2,3-H H'bz 1'-H ketone 4.17
Hbz
14-H 11-H ester 4.18
7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm
Spectrum 4.3 400 MHz 1H NMR specta of protected 4-HBRCG 4.17 and ester 4.18 in DMK-d6.
4.19 was obtained and submitted for mass spectral analysis, the molecular ion peak
measured showed a mass of 16 greater than expected. Shortly after, it was realized that
the structure of the isolated material was the ester 4.19.
Upon synthesis of the real target ketone 4.17, the structures were confirmed with
a series of NMR experiments. The first detail noticed was the difference in 13C NMR spectra of 4.17 and 4.18. The ester 4.18 did not have a typical ketone peak (~200 ppm),
whereas the product 4.17 did have a carbonyl ketone peak. Secondly, heteronuclear
multiple-bond correlation (HMBC) experiments were used to show long range coupling
between 1H and 13C nuclei and were performed on ketone 4.17 (Figure 4.4, Spectrum 113 4.4). The benzylic protons (3.7 ppm) showed cross peaks with the ketone carbon (199
ppm), the phenyl quaternary carbon, and the phenyl tertiary carbons (128-137 ppm).
Furthermore, the ketone carbon showed cross peaks with the neighboring 14- and 20-
position protons (6.3 and 2.3 ppm, respectively). Additionally, the more conjugated
ketone 4.17 showed a UV λmax absorbance at 379 nm, while the ester 4.18 had a UV λmax of 358 nm. These combined data, along with mass spectra, confirmed the structure of the target 4.17.
4.4 SYNTHESIS OF 4-HBRC-GLUCOSIDE
Along with the synthesis of the glucuronide 4-HBRCG, the glucoside was also produced (Scheme 4.7). Starting from the methyl ether 4.6, MOM groups were removed and replaced with acetate groups to give the tetra-O-acetyl-C-glucoside 4.20. As previously done, bromide displacement of the methyl ether gave the benzylbromide 4.21.
This was alkylated with tert-butyldimethylsilylcyanohydrin 4.14 and after TBAF treatement, yielded the alkylated ketone 4.22. Possium carbonate in methanol was used to remove the acetate groups giving the tetra-hydroxyl-glucoside analog 4.23. Key novel steps like the bromination and the alkylation reactions were initially done with the glucose analog as a model due to the increased difficulties in obtaining the glucuronide.
4.5 BIOLOGICAL EVALUATION OF 4-HBRCG
In the end, the final amount of material synthesized was approximately 2 g of 4-
HBRCG and 50 mg of the glucose analog 4.23. Both analogs will be tested for their ability to bind the nuclear retinoid receptors, inhibit growth, and cause apoptosis of MCF- 114 O OMe O H AcO H H AcO O AcO
H H H 4.17
Figure 4.4 Diagram of key long range coupling for 4.17.
ppm
110
120
130
140
150
160
170
180
190
200
210 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm
Spectrum 4.4 Partial HMBC plot of protected 4-HBRCG 4.17 in DMK-d6 at 400 MHz.
115 OMOM OH O 6 N HCl, MeOH O MOMO HO MOMO HO MOMO HO OMe OMe 4.6 4.7
OAc OAc
Ac2O, pyridine O HBr, AcOH O AcO AcO AcO AcO DMAP AcO AcO OMe 97 % Br 90 % 4.20 4.21
OAc 1) LiHMDS, 4.14, O THF, -78 °C AcO AcO AcO O 2) 4.21
3) TBAF, THF/H2O 4.22 36 %
OH O K2CO3, MeOH HO HO HO O 49 %
4.23
Scheme 4.7 Synthesis of 4-HBRC-glucoside 4.23.
116 7 cells in culture by our collaborators. The glucuronide 4-HBRCG will be tested for its
chemotherapeutic activity in DMBA-induced rat mammary tumors, similarly as for 4-
HPRCG. Due to the small quantities produced, this study will be preliminary in nature to
assess its potency relative to 4-HPR. This pilot study should indicate whether 4-HBRCG
should be pursued for further chemotherapeutic or preventative studies as well as
mechanism of action studies.
4.6 EFFORTS TOWARD THE SYNTHESIS OF RETINOYL-β-C- GLUCURONIDE
Because the glucuronide metabolite of retinoic acid possesses interesting activities, improved access to the C-linked analog (retinoyl-β-C-glucuronide) is desired for further studies. To further apply the versatility of the exocyclic sugar olefin 3.11 for
C-glycosides, work started toward the synthesis of retinoyl-β-C-glucuronide (Figure 4.2).
Enol ether 3.11 was hydroborated as previously done and then followed by a basic oxidative workup, using standard conditions.170 However, several trials were performed
without any success in isolating the hydroxymethyl sugar 4.24. Investigation of this
reaction was not fully explored but the production of 4.24 through this route should be
possible (Scheme 4.8). The next proposed steps would be to methylate the alcohol,
oxidize the 6-position alcohol group, reprotect, and brominate to give the bromomethyl-
C-glucuronide 3.8. With the bromide in hand, similar type alkylation chemistry used to
make 4-HBR and 4-HBRCG would be used to attempt making the target retinoyl-β-C-
glucuronide.
117 OMOM OMOM O 1) 9-BBN-H, THF O NaH, THF MOMO MOMO MOMO MOMO CH I MOMO 2) NaOH, H2O2 MOMO OH 3 3.11 4.24
OMOM OMe 1) HCl, MeOH O HBr, AcOH O 2) TEMPO O MOMO AcO MOMO OMe AcO OMe MOMO 3) CH3I, DMF AcO 4) Ac O, pyr 4.25 2 4.26
OMe O O AcO Retinoyl-β-C-glucuronide AcO AcO Br 3.8
Scheme 4.8 Proposed alternative synthetic route towards bromomethyl-glucuronide 3.8.
Access to the bromomethyl 3.8 can also be obtained from the siloxymethylation
reaction as discussed earlier in Chapter 3. Through the work of our collaborators, more quantities of the siloxymethyl-glucuronide analog 3.14 were obtained through CO insertion chemistry (Scheme 3.2). Seen in Scheme 4.9, the siloxymethyl group was smoothly displaced by bromide with dibromotriphenylphosphorane to obtain the bromomethyl-glucuronide 3.8. This bromide used in alkylation trials with silylcyanohydrin 4.14. Unfortunately, alkylation attempts did not result in the production of retinoyl-β-C-glucuronide 4.27. From these results, it was hypothesized that the bromomethyl-glucuronide lacked the needed reactivity towards the TBDMS-cyanohydrin
4.14. Therefore, the iodomethyl-glucuronide 4.28 was prepared by halogen exchange via the Finkelstein reaction. Since the iodo group is sufficiently bulky, the less sterically 118 OMe O O OMe O Ph PBr , CH Cl O AcO 3 2 2 2 AcO AcO AcO AcO OSiEt2Me AcO Br 3.14 88 % 3.8
O OMe 1) LiHMDS, 4.14, O THF, -78 °C AcO AcO AcO 2) 3.8 O 3) TBAF, THF/H2O 4.27
O OMe O OMe 1) LiHMDS, 4.13, O NaI O THF, -78 °C AcO AcO AcO AcO AcO Br DMK, reflux AcO I 2) 4.28 3) TBAF, THF/H2O 3.8 95 % 4.28
Scheme 4.9 Progress toward the synthesis of retinoyl-β-C-glucuronide.
hindered TMS-cyanohydrin 4.13 was used as the the retinal Umpolung derivative in the subsequent alkylation. Unfortunately, these modifications to the electrophile and the cyanohydrin still did not result in a successful alkylation to yield the target retinoid 4.27.
Since there was a significant amout of exploration into this synthetic approach, perhaps the halomethyl-glucuronides are not useful electrophiles for the silylcyanohydrins of retinal.
119 4.7 CONCLUSIONS
The fully C-linked analog of the glucuronide metabolite of 4-HPR was prepared
as a novel retinoid. The synthetic route to 4-HBRCG borrowed from strategies used in
the syntheses of 4-HPRCG and 4-HBR, which were access to the C-glycoside through
Suzuki coupling chemistry and the alkylation of the retinal Umpolung derivative, respectively. Even though preliminary efforts towards retinoyl-β-C-glucuronide were unsuccessful, the exocyclic sugar olefin 3.11 may still serve as a common precursor for numerous C-glycosides, especially C-linked retinoids that are important in our lab.
Starting from δ-gluconolactone, the overall yield of 4-HBRCG over 15 chemical transformations was 8.9 %, and approximately 2 grams were produced. Even though the synthetic route was successful, further optimizations are needed in the key, yet low yielding, alkylation reaction. As an alternative, preliminary literature investigations were done to determine the feasibility of alkylating with smaller isopentenyl units to build up the retinoid component. From known methodologies, it appeared that there was not an advantage to that approach due to the addition of needed steps.
The biological experiments with 4-HBRCG should reveal its activity as an antiproliferative. Since 4-HBRCG is more difficult to synthesize, determination of its potency will dictate whether it should be pursued or if the more readily accessible 4-
HPRCG should be studied further as a chemotherapeutic agent. The fully C-linked analog was originally proposed in order to investigate the possibility of teratogenicity due to retinoic acid liberation in vivo deriving from amide bond cleavage of 4-HPRCG. The root question may be clarified with cell based assays done with 4-HBRCG. Eventually, a vitamin A deficient animal study will need to be done with 4-HPRCG and 4-HBRCG to 120 answer the question definitively. However, if the chemotherapeutic actions are not robust, the hydrolytically stable 4-HBRCG can serve as a biological tool to study the exact target and mechanism of action for these interesting, biologically active, and structurally unique retinoids.
121
CHAPTER 5
SUMMARY AND CONCLUSIONS
5.1 PART 1
Stable isotope labeled amino acids are useful tools for protein structure
elucidation and mechanistic enzymology. The primary and secondary metabolism of
these essential molecules has been extensively studied by tracking isotopes. Retro-
biosynthetic analysis of amino acids and other natural products has been accomplished
with the incorporation of synthetic labeled precursors and amino acids, respectively.
Determination of protein structure is an essential part of biology and medicine. With the advent of numerous methodologies to produce and incorporate labeled amino acids into proteins, dynamic protein structures can now be obtained with the use of NMR. These featured isotopes, usually 13C and 15N, not only serve to greatly enhance the signal generated in NMR, but also can be used for multidimensional, isotope-edited experiments. Conversely, deuterium can be used to eliminate a signal, thereby, simplifying data interpretation. Also, deuterium can be substituted stereospecifically for hydrogen and, therefore, be used for spatial interpretations. Through a various combination of labeling patterns, isotopes, and multidimensional NMR experiments,
122 resonance assignments of the nuclei in the protein can be accomplished, after which
interpretations of secondary and tertiary structure can be facilitated.
Although other amino acids are important in enzymology and protein NMR,
glycine is a particularly essential amino acid. Due to its size and polarity, glycine
residues are usually instrumental in determining tertiary structure because they
commonly occur in turn regions, which are situated between common structural motifs.
In protein NMR, resonance assignments of the pro-R and pro-S glycine prochiral
methylene protons are particularly useful in the highly variable turn regions. These
assignments can be used to determine the coupling constant between the amide proton
and one of the methylene protons. This value is then used to estimate the dihedral angle,
which can be used to help determine the backbone conformation of the protein.
Stereospecific resonance assignments of glycine and other amino acids can
usually be obtained through the typical gambit of multidimensional NMR experiments.
However, there is sometimes a need for the synthesis and incorporation of specifically
labeled amino acids for further resonance assignments. Previously, our lab prepared and used (R)-[2-2H, 15N]glycine in NMR studies of the FK-506 binding protein. The labeled
glycine was chemically synthesized and incorporated into FKBP and used to facilitate stereospecific resonance assignments of prochiral methylene protons in glycine residues.
From 15N-edited TOCSY experiments, 1H-1H cross slices from the (R)-glycine-d-15N-
FKBP showed only one crosspeak, revealing the resonance position of the pro-S glycine
methylene proton, while the cross slices from 15N-FKBP gave the pro-R and pro-S
resonance positions.
123 The synthetic route to the doubly labeled glycine used in the above study was
plagued by a key, highly variable, and low yielding, oxidative degradation reaction.
After both isotopes are incorporated, catalytic ruthenium tetraoxide oxidized N-t-BOC-p-
methoxyphenyl methylamine 1.6 (Scheme 1.1), leaving the labeled N-t-BOC-glycine 1.7.
Due to interest in applying the doubly labeled glycine to other protein structural
problems, work was undertaken in order to optimize the ruthenium tetraoxide reaction.
In this reaction, aromatic rings serve as carboxylic acid synthons. Therefore, other
aromatic rings, such as 1- and 2-naphthyl rings, that were compatible with preceding
reaction conditions, were studied. From these studies, it became clear that the original
ring, p-methoxyphenyl, was optimal and the reaction conditions were in need of
modification. The original conditions used catalytic ruthenium chloride, stoichiometric
re-oxidant NaIO4, and a biphasic solvent system. Numerous oxidation conditions were
investigated, which included various ruthenium sources, temperatures, times, and
stoichiometry (Table 1.1). None of the modifications were satisfactory.
Another set of conditions that were not initially explored was using catalytic
ruthenium chloride and the stoichiometric re-oxidant periodic acid. In order to
investigate these conditions, a different amine protecting group was needed and the acid
labile tert-butoxy carbamate was replaced by the acid stable 2,2,2-trichloroethyl
carbamate. With these new conditions, the ruthenium tetraoxide mediated oxidative
degradation of the p-methoxyphenyl analog 1.22 was efficient and consistently produced
labeled N-2,2,2-trichloroethyl-glycine carbamate 1.23 (Scheme 1.4). Subsequently, the removal of the protecting group was investigated and optimized to produce the doubly labeled amino acid, (R)-glycine-d-15N, essentially doubling the reported overall yield 124 from the previous synthesis. These modifications to the synthetic route would enable large scale production for this important labeled amino acid, making (R)-glycine-d-15N
(1.24) available for researchers with specific enzymatic or NMR protein problems.
5.2 PART 2
Vitamin A and its metabolites are involved in many biological processes including vision, cell differentiation, and growth. Although retinol is the major form of retinoid in the body, most known actions, except for vision, are caused by the metabolite retinoic acid. The absorption, transport, and metabolism of retinol in the body are highly regulated due to its wide range of actions. Retinoic acid potently binds with nuclear receptors RAR and RXR, which are members of the super family of hormone-mediated nuclear receptors. Through these interactions, transcription can be induced or inhibited to affect cell function.
Besides being essential for normal cell processes, retinoic acid has antiproliferative actions in skin diseases and certain types of cancers. Because carcinomas are malfunctions in cell proliferation, retinoic acid is effective because it has potent cell differentiation actions. However, retinoic acid is an effective cancer therapeutic only at doses in which it causes severe toxicity as a hepatotoxic and a teratogen. Therefore, retinoic acid analogs possessing a higher therapeutic index as treatment for cancers are of interest.
One of the most investigated synthetic retinoid is the retinoic acid conjugate N-(4- hydroxyphenyl)retinamide. This derivative has been shown to be very effective in numerous tumor models in vitro and in vivo and has been involved in phase III clinical 125 trials for breast cancer. By derivatization to the p-hydroxyphenyl amide, the toxicity of retinoic acid is not eliminated but it is decreased by approximately two orders of magnitude. 4-HPR does not cause cell differentiation like retinoic acid but is an effective chemopreventative and therapeutic due to its ability to cause apoptosis in cancers cells selectively over normal cells. The exact target and mechanism of action for 4-HPR remains unknown, however, it is known that 4-HPR does not affect its actions though activation of the retinoic acid receptors.
Glucuronidation of drugs and natural products is a common metabolic pathway that usually facilitates excretion. In vivo, the hydrophobic 4-HPR is metabolized mainly to the methylated derivative N-(4-methoxyphenyl)retinamide and the glucuronide conjugate 4-HPR-O-glucuronide, in which the phenolic hydroxyl group is linked to the sugar. The methyl derivative is thought to be devoid of significant biological actions, however, the actions of 4-HPROG was first indicated through a study with 4-HPR and the β-glucuronidase inhibitor calcium glucarate. Subsequently, 4-HPROG has been shown to have superior chemopreventative and therapeutic actions over 4-HPR in mammary tumor models. Since 4-HPROG is a substrate for β-glucuronidase, it was not known whether 4-HPROG was advantageous due to improved bioavailability, liberating
4-HPR, or had activity as an intact molecule.
To study this issue, the enzymatically stable carbon linked analog of 4-HPROG
(4-HPRCG) was proposed, with a methylene group replacing the phenolic linkage between 4-HPR and the sugar. 4-HPRCG was synthesized and was shown to have chemopreventative qualities superior to the O-linked analog and the parent 4-HPR in a
126 mammary tumor model. These results strongly indicated that 4-HPROG and 4-HPRCG
are effective agents as intact molecules and are not prodrugs of 4-HPR.
The previous synthetic route to the important retinoid 4-HPRCG concentrated on
the production of a key C-glucuronide aniline 3.7, after which the aniline would be coupled to retinoyl chloride in a penultimate step (Scheme 3.1). Access of the key C- glucuronide through the previous route was lengthy, inefficient, and warranted improvement. The formation of the C-glycoside 3.3 relied on a modest yielding electrophilic addition of benzyl Grignard reagent to an expensive and unstable perbenzylated bromoglucose derivative. Subsequent difficult transformations included removal of the benzyl protecting groups, oxidation of the C-glucoside to the glucuronide, and an aryl nitration reaction which produced nitro regioisomers (Scheme 3.1).
Due to an interest in investigating the chemotherapeutic actions of this interesting retinoid, the synthetic route for 4-HPRCG was redesigned (Figure 3.5). Similarly to the previous synthesis, the current route concentrated on efficiently gaining access to the key
C-glucuronide aniline 3.7. Starting from the very inexpensive δ-gluconolactone (3.12), efficient production of a key exocyclic sugar olefin 3.11, using the organometallic Petasis reagent, enabled access to various C-glycosides. The enol ether was used in the versatile
Suzuki-Miyaura coupling reaction with 1-bromo-4-nitrobenzene in a key step to directly produce the C-glucoside 3.10. Subsequently, numerous functional group interconversions and protecting group manipulations efficiently yielded the key C- glucuronide aniline 3.7 (Scheme 3.5). Lastly, the aniline was coupled to retinoyl chloride and then deprotected to give the target retinoid 4-HPRCG (Scheme 3.6). Furthermore,
127 the new route introduced chemistry techniques previously unexplored in our lab and also improved upon the known steps used in the previous route.
The new route to 4-HPRCG enabled a more efficient production of 4-HPRCG, which yielded amounts sufficient for a preliminary and a complete chemotherapeutic animal study. The rat mammary tumor model revealed that 4-HPRCG was an effective chemotherapeutic, however, it was only equieffective as 4-HPR at shrinking tumors.
Other data from the animal study indicated that 4-HPRCG may have lower toxicity than
4-HPR. Therefore, the actions of 4-HPRCG as a superior chemopreventative, an equieffective chemotherapeutic, and combined with its apparent lower toxicity, show that
4-HPRCG may prove to be a better antiproliferative agent than the parent 4-HPR.
Besides the pursuit of increasing the therapeutic index of 4-HPR by the utilization of glucuronide conjugates, investigation of the mechanism of action and toxicity of 4-
HPR is also of interest. A possible explanation for the minor toxicities of 4-HPR is in vivo amide cleavage, which would liberate retinoic acid. To study this issue, an unhydrolyzable analog (4-HBR) was proposed and synthesized, where the amide is replaced by a methylene functionality. In vitro and in vivo studies showed that 4-HPR and 4-HBR are equally effective chemotherapeutics and both affect their actions by apoptosis. Their differences were revealed in experiments with vitamin A deficient rats.
By treatment with both retinoids, it was shown that high doses of 4-HPR could maintain growth in the rats and 4-HBR did not support growth. Furthermore, in vitro assays of the blood and tissues from these deficient animals did reveal the presence of retinoic acid in
4-HPR treated rats. Conversely, samples from the 4-HBR treated rats did not contain
128 retinoic acid. Therefore, the toxicities seen with 4-HPR may be explained by the in vivo production of retinoic acid.
Since it was shown that the amide bond of 4-HPR is unstable in vivo, the possibility of amide bond cleavage may also occur with the effective agent 4-HPRCG.
To investigate this issue, the carbon-linked analog of 4-HPRCG (4-HBRCG) was proposed by replacing the amide functionality with a methylene group, which would give the fully C-linked analog of the natural metabolite 4-HPR-O-glucuronide. Like 4-HBR, the methyl retinone derivative 4-HBRCG may have the possibility of being devoid of teratogenic toxicities due to its inability to liberate retinoic acid. Furthermore, because 4-
HBRCG is a glucuronide conjugate, its potency as an antiproliferative should be increased over the parent 4-HBR.
One advantage of using the Suzuki coupling reaction as the key step in producing
C-glycosides is its versatility. The synthetic route to 4-HBRCG adapts methods developed from the synthesis of 4-HPRCG and 4-HBR. Using the key exocyclic sugar olefin 3.11, p-bromobenzyl alcohol was coupled in the Suzuki reaction to give ready access to the C-glycoside 4.4 (Scheme 4.1). Similar to the 4-HPRCG synthesis, numerous functional group interconversions and protecting group manipulations readily yielded the appropriate C-glucuronide 4.9 (Scheme 4.2). Once the necessary benzyl bromide electrophile 4.12 was obtained, the key alkylation with a retinal Umpolung derivative 4.14, similarly done with the 4-HBR synthesis, gave the carbon skeleton for 4-
HBRCG (Scheme 4.6). Even though this route enabled the production of the novel glucuronide-retinoid conjugate, the key alkylation of the benzyl bromide glucuronide was fairly inefficient and warrants improvement. 129 Upcoming biological testing such as receptor binding, cell based assays, and a
preliminary chemotherapy experiment should reveal the potency and actions of 4-
HBRCG and dictate whether it should be further pursued as a potential drug candidate.
To investigate the possibility of 4-HPRCG amide bond cleavage, a comparative study
with 4-HBRCG on vitamin A deficient rats will be necessary. Because of its probable in
vivo stability, 4-HBRCG should prove to be a valuable biological tool for the elucidation of the mechanism of action for these glucuronide-retinoid conjugates.
130
CHAPTER 6
EXPERIMENTAL SECTION
6.1 GENERAL METHODS
Anhydrous THF and CH2Cl2 were obtained using distillation from sodium
benzophenone ketyl and calcium hydride, respectively. Sigma-Aldrich (Milwaukee, WI)
supplied starting materials and reagents. Cambridge Isotopes Laboratories (Cambridge,
MA) supplied isotope labeled reagents. All reactions and handling of retinoid containing
compounds were done under gold fluorescent lights. TLC was performed on Merck
(Gibbstown, NJ) silica gel 60 F254 aluminum plates. Column chromatography was
performed with Merck silica gel 60 and reverse phase flash chromatography with Merck
Lichroprep® RP-18. Ion exchange chromatography was performed using Bio-Rad
(Hercules, CA) AG 1 X8 and AG 50W X8, 200-400 mesh. Double deionized, demineralized water was used in the resin preparation and elution solutions. Analytical
HPLC was done on Beckman Instruments (San Ramon, CA), with pump component 127 and detector module 166, equipped with a Metachem Polaris (Varian), 5 µm C-18, 250 X
4.6 mm column. All retinoids were detected at a wavelength of 350 nm. Catalytic
hydrogenation was conducted on a Parr (Moline, IL) Hydrogenation Apparatus Model
131 3911. Melting points were determined using a Thomas-Hoover (Philadelphia, PA)
capillary aparatus and are uncorrected. Circular dichroism was performed uncalibrated
on a JASCO (Victoria, British Columbia) J-500A spectropolarimeter and reported in
molar ellipticity (degrees·deciliters·mol-1·decimeter-1). Optical rotations were conducted
on a Perkin-Elmer (Wellesley, MA) 241 polarimeter and reported in mol·dm-1·gram-1.
Ultraviolet spectra were recorded on a Beckman Instruments DU-40 spectrophotometer.
Infrared spectra were recorded as films on silver chloride plates using a Nicolet
(Madison, WI) Protégé 460 spectrophotometer. NMR spectra were recorded on a Bruker
15 (Billerica, MA) DRX 400 spectrometer. Standard N-glycine in D2O was used as an
external reference for 15N NMR (31.5 ppm). Analysis of the MTPA ester was recorded
on a Bruker DRX 800 spectrometer. Mass spectra were recorded on a Micromass
(Milford, MA) QTOF Electrospray mass spectrometer.
6.2 GLYCINE SYNTHESIS
O
N CN
H3CO
α-(p-Methoxyphenyl)-4-morpholine acetonitrile (1.20).
To a 250 mL flask chilled to -15°C, was added 60 mL of morpholine and 28.4 mL
(0.33 mol) of perchloric acid (70 %) dropwise and stirred for 15 min. The flask was removed from the cold bath and 40.8 g (0.3 mol) of p-anisaldehyde (1.1) was added. The
132 mixture was heated to 60 °C for 2 h. A saturated aqueous solution of 21.45 g (0.33 mol)
of potassium cyanide was then added in portions and the reaction heated to 90 °C for 2 h.
The mixture was cooled to ~40 °C and then poured onto ice while stirring. Yellowish
solid formed which was washed with water via vacuum filtration. Recrystallization
(ethanol / water) yielded 67.2 g of white needles (96 %), mp 79-80 °C (lit. mp 81-82
171 -1 1 °C). IR (cm ) 2965 (m), 1506 (s), 1255 (s), 1117 (s); H NMR (CDCl3) δ 2.53-2.57
(m, 4H), 3.69-3.71 (m, 4H), 3.80 (s, 3H), 4.73 (s, 1H), 6.90 (d, 2H, J = 8.8 Hz), 7.41 (d,
13 2H, J = 8.8 Hz); C NMR (CDCl3) δ 49.86, 55.34, 61.79, 66.61, 114.12, 115.35, 124.25,
129.31, 160.10; HRMS (ES) calcd for C13H16N2O2 (M + Na) 255.1109, found 255.1120.
O
N
CN D
H3CO
α-(p-Methoxyphenyl)-4-morpholine acetonitrile-α-d (1.2).
Under an argon atmosphere, a dry 3-neck 250 mL flask equipped with a reflux condenser was charged with 22.4 g (0.097 mol) of the aryl-morpholinonitrile 1.20 into 75 mL of dry THF. Under an argon envolope, 4.9 g (0.58 mol) of sodium hydride (95 %) was weighed and added to the reaction mixture in ~1.2 g portions. The gray solution was heated to 50 °C for 6 h. After cooling the solution to 0 °C, 17.3 mL (10 equiv.) of D2O
was carefully added dropwise and allowed to stir for 30 min. Thionyl chloride was then
added dropwise until the mixture was slightly acidic (pH <5). The entire solution was
poured onto ice while stirring until white solid precipitated. The solid was dried via
133 vacuum filtration to give 20.9 g (93 %; 97 % 2H), mp 79-80 °C (lit. mp 81-82 °C).171 IR
-1 1 (cm ) 2959 (m), 1511 (s), 1250 (s), 1111 (s), 839 (m); H NMR (CDCl3) δ 2.54 (s, 4H),
3.67-3.71 (m, 4H), 3.80 (s, 3H), 6.90 (d, 2H, J = 8.8 Hz), 7.41 (d, 2H, J = 8.8 Hz); 13C
NMR (CDCl3) δ 49.80, 55.34, 61.50 (t, J = 22.6 Hz), 66.63, 114.11, 115.38, 124.29,
129.29, 160.09; HRMS (ES) calcd for C13H15DN2O2 (M + H) 234.1353, found 234.1354.
O
D
H3CO
4-Methoxy-benzaldehyde-formyl-d (1.3).
To a 1 L flask equipped with a reflux condenser was added 62.5 g (0.27 mol) of the deuterated aryl-mopholinonitrile 1.2 along with 600 mL of 2 M HCl. The suspension
was refluxed for 14 h. The mixture was allowed to cool and extracted with CH2Cl2. The
organic layers were combined and washed with saturated NaHCO3 solution, water, and
- dried (MgSO4). Evaporation of the solvent yielded 34.7 g of orange oil (95 %). IR (cm
1 1 ) 3016 (w), 2843 (m), 1685 (s), 1603 (s), 1271 (s), 1168 (s); H NMR (CDCl3) δ 3.86 (s,
13 3H), 6.98 (d, 2H, J = 8.3 Hz), 7.82 (d, 2H, J = 8.7 Hz); C NMR (CDCl3) δ 55.52,
114.27, 129.81, 131.92, 164.58, 190.49 (t, J = 26.7 Hz); HRMS (ES) calcd for C8H7DO2
(M + H) 138.0665, found 138.0667.
134
H D
OH
H3CO
(αS)-4-Methoxy-benzenemethan-d-ol (1.4).
Under an argon atmosphere, a dry 1 L flask equipped with a reflux condenser was charged with 500 mL (0.25 mol) of 0.5 M THF solution of R-Alpine-Borane® along with
20.5 g (0.15 mol) of deuterated aldehyde 1.3. The solution was stirred for 20 h followed by reflux for 1.5 h. After cooling to rt, 23 mL of acetaldehyde was added and the mixture stirred for 1 h. Rotary evaporation removed the solvent and pinene was partially removed by vacuum pump at 50 °C for 3 h. The resultant orange oil was dissolved in
250 mL of ether and cooled to 0 °C. 2-Aminoethanol (15.25 g, 0.25 mol) was added and left to stir at rt for 1 h. The white precipitate was removed by vacuum filtration and washed with ether. The organic fractions were combined, washed with water, and dried
(MgSO4). Evaporation of the solvent yielded an orange oil which was further
decontaminated by partitioning between 10 % aqueous methanol and octane. The
methanol layer was concentrated and the alcohol was isolated by vacuum distillation or
silica gel chromatography (95:5 followed by 90:10 hexanes / ethyl acetate) to yield 18.8 g
of yellow oil (90 %). IR (cm-1) 3348 (br), 2936 (w), 1608 (m), 1511 (s), 1247 (s), 1033
1 (s), 804 (m); H NMR (CDCl3) δ 3.79 (s, 3H), 4.57-4.60 (m, 1H), 6.87 (d, 2H, J = 8.7
13 Hz), 7.27 (d, 2H, J = 8.4 Hz); C NMR (CDCl3) δ 55.26, 64.50 (t, J = 22.0), 113.89,
128.63, 133.09, 159.12; HRMS (ES) calcd for C8H9DO2 (M + Na) 162.0641, found
162.0645. 135
O D H
N 15 H3CO O
(R)-2-[(Methoxy phenyl) methyl-d]-1H-isoindole-1,3(2H)-dione-2-15N (1.5).
Under an argon atmosphere, a dry 3-neck 500 mL flask equipped with a reflux
condenser was charged with 5.78 g (41.6 mmol) of the chiral deuterated alcohol 1.4,
11.96 g (45.7 mmol) of triphenylphosphine, and 6.76 g (45.7 mmol) of 15N-phthalimide
stirred in 300 mL of dry THF. The solution was chilled to 0 °C and 7.95 g (45.7 mmol)
of diethyl azodicarboxylate was added dropwise. The solution stirred at 0 °C for 4 h and
then at rt for 18 h. The solvent was evaporated to give a yellow solid. The resultant
mixture was dissolved and passed through silica gel using 3:2 hexanes / ethyl acetate to
remove triphenylphosphine oxide. The eluent was subjected to silica gel chromatography
using 4:1 hexanes / acetone and the product recrystallized using acetone / H2O to give
6.71 g (60 %) of white solid, mp 129-131 °C (lit. mp 133-134 °C).43 IR (cm-1) 3045 (w),
1 2967 (w), 1705 (s), 1515 (m), 1243 (m), 707 (m); H NMR (CDCl3) δ 3.75 (s, 3H), 4.74-
4.76 (m, 1H), 6.82 (d, 2H, J = 8.7 Hz), 7.37 (d, 2H, J = 8.7 Hz), 7.66-7.68 (m, 2H), 7.80-
13 7.82 (m, 2H); C NMR (CDCl3) δ 40.75 (dt, J = 8.8, 22.7 Hz), 55.21, 113.95, 123.23,
128.59, 130.12, 132.09, 132.16, 133.88, 159.18, 168.01 (d, J = 12.7 Hz); HRMS (ES)
15 calcd for C16H12D NO3 (M + Na) 292.0826, found 292.0830.
136
D H O Cl N O 15 Cl H Cl H3CO
(R)-2,2,2-Trichloroethyl [(4-methoxyphenyl) methyl-d]-carbamate-15N (1.22).
In a 500 mL flask equipped with a reflux condenser was added 3.45 g (12.8
mmol) of the doubly labeled phthalimide conjugate 1.5 stirred in n-propanol (140 mL)
and water (22 mL). Sodium borohydride (2.43 g, 64.3 mmol) was added in portions and
left to stir at rt for 24 h. Carefully, 14 mL of glacial acetic acid was added dropwise and
the mixture heated to 80-90 °C for 5 h. After cooling to rt, the solvent was removed by
rotary evaporation and dissolved in 50 mL of dioxane / water (2:1). The solution was
chilled to 0 °C and 5.66 g (53.4 mmol) of sodium carbonate was added to give a basic solution (pH ~8). 2,2,2-Trichloroethyl chloroformate (4.08 g, 19.3 mmol), dissolved in
50 mL of dioxane / water (2:1), was added in portions to the reaction which was left to stir on ice and warm to rt overnight. The solvent was removed and the remaining aqueous suspension acidified with 1 M HCl to pH 2. The mixture was extracted with ether and dried (MgSO4). The organic fraction was concentrated and acetic acid removed by rotary evaporation using toluene azeotrope. Silica gel chromatography using 3:2 hexanes / ethyl acetate gave an oil which was crystallized with ether / hexanes yielding
3.55 g (88 %), mp 63-64°C (unlabeled lit. mp 61-62°C).172 IR (cm-1) 3328 (br), 3002
1 (w), 2951 (w), 1713 (s), 1511 (s), 1239 (s), 816 (m), 715 (m); H NMR (CDCl3) δ 3.78 (s,
3H), 4.31-4.34 (m, 1H), 4.73 (s, 2H), 5.18 (dd, 1H, J = 5.8, 91.8 Hz), 6.86 (d, 2H, J = 8.6
13 Hz), 7.21 (d, 2H, J = 8.6 Hz); C NMR (CDCl3) δ 44.22-44.72 (m), 55.28, 74.54, 95.58, 137 114.10, 128.96, 129.79, 154.54 (d, J = 27.7 Hz), 159.16; HRMS (ES) calcd for
15 C11H11D NO3Cl3 (M + Na) 335.9814, found 335.9821.
D H O
15N O H H CO 3
(R)-t-Butyl [(4-methoxyphenyl) methyl-d]-carbamate-15N (1.6).
This compound was obtained by the same procedure as that for the synthesis of
the 2,2,2-trichloroethyl analog 1.22 using di-tert-butyl dicarbonate instead of 2,2,2- trichloroethyl chloroformate to give the t-butyl analog 1.6, a colorless oil, in 88% yield.
1 H NMR (CDCl3) δ 1.44 (s, 9H), 3.77 (s, 3H), 4.20-4.21 (m, 1H), 4.76 (dd, 1H, J = 5.4,
90.4 Hz), 6.84 (d, 2H, J = 8.6 Hz), 7.18 (d, 2H, J = 8.6 Hz); HRMS (EI) calcd for
15 C13H18D NO3 239.1398, found 239.1402.
D H O HO Cl NO 15 Cl O H Cl
(R)-N-[(2,2,2-Trichloroethyoxy) carbonyl]-glycine-d-15N (1.23).
In a water bath, a 50 mL flask was charged with 2.45 g (7.78 mmol) of the
trichloroethyl p-methoxybenzylamine carbamate 1.22 along with carbon tetrachloride
(15.7 mL), acetonitrile (15.7 mL), and water (23.5 mL). Periodic acid (24.86 g, 0.109
mol) was added and stirred for 10 min. Ruthenium chloride hydrate (35.5 mg, 0.171
138 mmol) was added and the mixture stirred at rt for 2 h. The solution immediately turned
black then changed to orange. The solution was placed in an ice bath, diluted with ether,
and stirred for 10 min. The reaction mixture was extracted with ether and the organic
fractions were combined, washed with brine, dried (MgSO4), and concentrated to an
orange oil. The product was isolated with reverse-phase flash chromatography using 4:1
then 1:1 water / methanol to give 1.22 g (62 %) of white solid, mp 123-124 °C (unlabeled
lit. mp 123-125 °C).173 IR (cm-1) 3320 (br), 2955 (m), 1720 (s), 1511 (m), 1227 (m), 815
1 (m); H NMR (CDCl3) δ 4.06-4.09 (m, 1H), 4.74 (s, 2H), 5.43 (dd, 1H, J = 5.6, 93.5),
13 5.82 (dd, 1H, J = 5.6, 93.5) (rotamers); C NMR (DMK-d6) δ 42.34-42.91 (m), 74.92,
15 96.72, 155.76 (d, J = 28.6 Hz), 171.15; N NMR (CDCl3) 75.1, 78.1 (rotamers); HRMS
15 (ES) calcd for C5H5D NO4Cl3 (M + Na) 273.9293, found 273.9297.
D H O HO 15NO O H
(R)-N-[(t-Butoxy) carbonyl]-glycine-d-15N (1.7).
To a 50 mL flask was added 1.0 g (4.18 mmol) of the t-butyl p- methoxybenzylamine carbamate 1.6 dissolved in carbon tetrachloride (4 mL), acetonitrile
(4 mL), and water (6 mL). Sodium periodate (17.9 g, 84 mmol) was added and stirred for
10 min. Ruthenium chloride hydrate (19 mg, 0.09 mmol) was added and the mixture
stirred at rt for 72 h. The mixture was filtered and the solid washed with ethyl acetate.
The organic filtrate was washed with water, dried (MgSO4) and concentrated. The
139 resultant oil was re-dissolved in chloroform, filtered, and concentrated to give an orange
1 oil, crude yield, 10%. H NMR (CDCl3) δ 1.44 (s, 9H), 3.90-3.91 (m, 1H), 5.01 (d, 1H, J
= 97.7).
DH HO 15 NH2 O
(R)-Glycine-d-15N (1.24).
To a 50 mL flask was added 0.56 g (2.2 mmol) of the trichloroethyl glycine carbamate 1.23 dissolved in 7.4 mL of glacial acetic acid. The solution was diluted with
8.6 mL of water and then 1.21 g (18.6 mmol) of zinc dust was added which caused violent bubbling. The solution was stirred for ~2 min after which 6.1 mL of water was added and stirring continued for 10 min. The suspension was vacuum filtered and the aqueous phase washed with CH2Cl2. Water was removed by lyophilization and acetic acid by rotary evaporation using toluene azeotrope. The resultant solid was applied to anion exchange resin AG 1 in the OH- form. The resin was washed with water until
neutral and then glycine was eluted with 1 M HCl. Concentration of the eluent gave a colored gel that was taken up in water and applied to cation exchange resin AG 50W in the H+ form. The resin was washed until neutral and the glycine eluted with 1 M
NH4OH. The eluent was degassed, concentrated, and lyophilized to yield 125 mg (73 %)
29 25 of white solid, mp 229-231 °C dec (deuterated lit. dec. 234 °C). CD, θ 208(max) = +9.65
40 1 13 (c 2.0 H2O); H NMR (D2O, 4.65 ppm) δ 3.36 (s, 1H); C NMR (D2O, CH3NO2 ref.) δ
140 15 41.38 (dt, J = 5.4, 22.3 Hz), 173.10; N NMR (D2O) 30.3; HRMS (ES) calcd for
15 C2H4D NO2 (M + Na) 100.0251, found 100.0254.
O
N
O
2-[(1-Naphthyl) methyl]-1H-isoindole-1,3(2H)-dione (1.10).
To a 500 mL flask was added 3.34 g (21.4 mmol) of 1-naphthaldehyde (1.9)
dissolved in 55 mL of ethanol and 17 mL of water. Sodium borohydride (0.89 g, 23.5
mmol) was added in portions and the mixture stirred at rt overnight. The reaction was
quenched with 1 M HCl until bubble formation ceased. The solution was extracted with
ether and the organic fractions were washed with brine, dried (MgSO4), and concentrated to give 3.35 g (99 %) of a yellow oil. The alcohol was taken to the next step without further purification. The title compound was obtained by the same procedure as that for the synthesis of the phthalimide protected analog 1.5, to yield 2.7 g of a white solid (44
1 %). H NMR (CDCl3) δ 5.32 (s, 2H), 7.34-7.88 (m, 10H), 8.36 (d, 1H, J = 12.5 Hz).
141
O
N
O
2-[(2-Naphthyl) methyl]-1H-isoindole-1,3(2H)-dione (1.13).
This compound was obtained by the same procedure as that for the synthesis of
the 1-naphthyl analog 1.10 using 2-naphthaldehyde (1.12) (5.28 g) instead of 1- naphthaldehyde as the starting material to give 5.38 g of a white solid (55 %). 1H NMR
(CDCl3) δ 4.98 (s, 2H), 7.42-7.88 (m, 11H).
H N O
O
t-Butyl [(1-naphthyl) methyl]-carbamate (1.11).
Starting with 2.0 g (7.0 mmol) of the phthalimide protected 1-naphthyl
methylamine 1.10, the title compound was obtained by the same procedure as that for the
synthesis of the p-methoxybenzyl analog 1.6 to give 1.47 g of a clear solid (83 %). 1H
NMR (CDCl3) δ 1.52 (s, 9H), 4.75 (s, 3H), 7.34-8.1 (m, 7H).
142
O
N O H
t-Butyl [(2-naphthyl) methyl]-carbamate (1.14).
Starting with 2.0 g (7.0 mmol) of the phthalimide protected 2-naphthyl methylamine 1.13, the title compound was obtained by the same procedure as that for the synthesis of the p-methoxy-benzyl analog 1.6 to give 1.43 g a clear solid (80 %). 1H
NMR (CDCl3) δ 1.46 (s, 9H), 4.46 (d, 2H, J = 5.8 Hz), 4.91 (s, 1H), 7.37-7.81 (m, 7H).
H D O
O CF H3CO 3 H3CO
Preparation of the MTPA ester of (αS)-4-methoxybenzenemethan-d-ol (1.21).
To a dry 50 mL flask under an argon atmosphere was added 47 mg (0.34 mmol) of the chiral deuterated benzyl alcohol 1.4 and 58 mg (0.47 mmol) of 4- dimethylaminopyridine dissolved in 6 mL of dry methylene chloride. (R)-(-)-α-methoxy-
α-trifluoromethylphenylacetic acid chloride (100 mg, 0.4 mmol) was added and the mixture stirred for 3 h at rt. The solvent was removed by rotary evaporation and the product isolated by preparative TLC using a 3:1 hexanes / ethyl acetate mobile phase to give a yellowish oil, 98 mg (90 %). The %ee of the (S)-4-MeO-benzyl alcohol was
1 determined by H NMR (800 MHz, CDCl3) integration analysis of the benzylic proton
143 resonances of the MTPA ester: (R)-MTPA ester of (S)-OH, δ 5.28 (98 %); (R)-MTPA
ester of (R)-OH, δ 5.21 (2 %).
O O O 15 HN COOH DH
Preparation of the camphanate amide of (R)-Glycine-d-15N (1.25).
To a 25 mL flask chilled on ice was added 7 mg (0.09 mmol) of the doubly
labeled chiral glycine 1.24 dissolved in 2.2 mL of 0.1 M NaOH. (1S)-(-)-camphanic
chloride (19 mg, 0.1 mmol) dissolved in 2 mL of toluene was added and the mixture
stirred for 3 h. The basic solution was washed with methylene chloride and then acidified
with 1 M HCl. The aqueous layer was extracted with methylene chloride once and then
with diethyl ether three times. The ether layers were combined, dried (MgSO4), and concentrated to give a white solid, 13 mg (56 %). The %ee of the doubly labeled (R)-
1 glycine was determined by H NMR (800 MHz, CDCl3) integration analysis of the α-
proton resonances of the camphanamide derivative: (S)-camphanamide of (R)-glycine, δ
4.34 (d, 91 %, J = 4.9); (S)-camphanamide of (S)-glycine, δ 4.15 (d, 9 %, J = 4.9).
144 6.3 4-HPRCG SYNTHESIS
O O O
O OO O O O O
2,3,4,6-tetra-O-(methoxymethyl)-D-gluconic acid-δ-lactone (3.15).
To a flame dried flask under argon atmosphere was added δ-gluconolactone
(3.12) (7.38 g, 41.4 mmol) and CH2Cl2 (400 mL). Upon cooling the suspension with an ice bath, diisopropylethylamine (57.6 mL, 331 mmol) was added dropwise, followed by careful addition of chloromethyl methyl ether (50 g, 621 mmol) via an addition funnel. A significant amount of white smoke formed in the reaction vessel. Solid tetrabutylammonium iodide (50 g, 134 mmol) was added and the solution was allowed to warm to rt. The reaction stirred in the dark for 48 h upon which the solution gradually turned red. After cooling the vessel to 0 °C, saturated aqueous NH4Cl (75 mL) was
added. The contents were then diluted with water and the layers separated. The organic
layer was washed with brine and the combined aqueous layers were extracted with
CH2Cl2 (3x). The combined organic layers were dried (MgSO4), filtered, and
concentrated. The solids were then triturated with ether (4x) and the ether was
concentrated. The resultant oil was chromatographed on silica gel (1:1 hexanes/ethyl
-1 acetate) to afford 12.04 g (83 %) of clear oil. [α]D 118.4 (c 2.15, CH2Cl2); IR (cm )
2948 (s), 2885 (s), 1757 (s), 1464 (m), 1443 (m), 1213 (s), 1150 (s), 1035 (s), 912 (m); 1H
NMR (CDCl3) δ 3.36-3.42 (m, 12H), 3.77 (dd, 1H, J = 3.8, 11.3 Hz), 3.82 (dd, 1H, J =
145 2.8, 11.3 Hz), 3.99-4.05 (m, 2H), 4.29 (d, 1H, J = 6.6 Hz), 4.55-4.56 (m, 1H), 4.65 (s,
13 2H), 4.69-4.92 (m, 7H); C NMR (CDCl3) δ 55.42, 56.05, 56.11, 56.22, 66.12, 73.69,
74.77, 78.43, 96.56, 96.66, 96.78, 96.91, 97.13, 168.70; HRMS (ES) calcd for C14H26O10
(M+Na) 377.1424, found 377.1408.
Dimethyl titanocene, Cp2Ti(CH3)2 (Petasis’ reagent).
To a flame dried flask under argon atmosphere was added titanocene dichloride
(14.63 g, 58.8 mmol) and absolute ether (300 mL), which was cooled to 10 °C. Methyl lithium (100 mL, 140 mmol, 1.4 M) was carefully added dropwise via an addition funnel in the dark. The cold bath was removed and the red solution was allowed to stir for 10 min. The solution was then cooled to 0 °C and ice water (25 mL) was carefully added to quench the unreacted methyl lithium. The layers were separated and the aqueous layer extracted with ether (2x). The combined organic layers were dried (Na2SO4) under argon for 1 h and concentrated in the dark at 20 °C to give 12.4 g of orange solid. Dry toluene
(100 mL) was added and the reagent was stored at 4 °C and used without characterization.
146
O O
O OO O O O O
2,6-Anhydro-1-deoxy-3,4,5,7-tetra-O-(methoxymethyl)-D-gluco-hept-1-enitol (3.11).
To a flame dried flask under argon atmosphere was added the sugar lactone 3.15
(10.05 g, 28.4 mmol) dissolved in dry toluene (140 mL) via an addition funnel. The toluene solution of dimethyl titanocene (12.4 g, 59 mmol) was then added dropwise via an addition funnel to give a red solution. The flask was then equipped with a reflux condenser and heated to 70 °C and let stir in the dark for 18 h. The resultant black solution was cooled and poured into hexanes (~500 mL). A precipitate formed and was filtered through celite. The supernatant was concentrated to yield a red oil which was chromatographed on silica gel (4:1 then 2:1 hexanes/ethyl acetate) to afford 8.66 g (87
-1 %) of yellowish oil. [α]D 46.8 (c 2.33, CH2Cl2); IR (cm ) 2940 (m), 2895 (m), 1750 (w),
1 1440 (w), 1154 (s), 1032 (s), 918 (m); H NMR (DMK-d6) δ 3.31-3.37 (m, 12H), 3.64-
3.71 (m, 2H), 3.78-3.83 (m, 2H), 3.88-3.89 (m, 1H), 4.12 (d, 1H, J = 5.4 Hz), 4.35 (s,
13 1H), 4.51 (s, 1H), 4.62 (s, 2H), 4.66-4.84 (m, 6H); C NMR (DMK-d6) δ 55.15, 55.87,
56.04, 56.19, 67.50, 75.42, 76.68, 77.36, 81.08, 93.43, 95.35, 97.23, 97.64, 97.81,
156.39; HRMS (ES) calcd for C15H28O9 (M+Na) 375.1631, found 375.1628.
147
O O
ONOO O2 O O O O
2,6-Anhydro-3,4,5,7-tetra-O-(methoxymethyl)-1-deoxy-1-(4-nitrophenyl)-D-glycero-
D-gulo-heptitol (3.10).
To a flame dried flask under argon atmosphere was added the exocyclic olefin
3.11 (7.06 g, 20.0 mmol) dissolved in dry THF (100 mL). 9-BBN-H (100 mL, 50 mmol,
0.5 M) was added via addition funnel. The flask was then equipped with a reflux
condenser, heated to 75-80 °C, and refluxed for 5 h. The mixture was cooled to rt after which K3PO4 (17 mL, 3 M) was added and allowed to stir for 10 min. p-
Nitrobromobenzene (8.07 g, 40.0 mmol) and PdCl2(dppf) (1.3 g, 1.6 mmol) dissolved in
DMF (180 mL) was added via addition funnel and the mixture stirred for 18 h. The
reaction was diluted with water and extracted with ether (3x). The organic layers were
combined, dried (MgSO4), concentrated, and chromatographed (2:1 then 1:1
hexanes/ethyl acetate) to afford 5.09 g (54 %) of orange oil. IR (cm-1) 3079 (w), 2936
(m), 1607 (w), 1517 (m), 1346 (s), 1150 (s), 1101 (s), 1028 (s), 922 (w); 1H NMR (DMK-
d6) δ 2.78-2.84 (m, 1H), 3.23 (s, 3H), 3.30-3.46 (m, 13H), 3.52-3.57 (m, 2H), 3.64 (t, 1H,
J = 8.9 Hz), 3.73 (dd, 1H, J = 1.9, 11.5 Hz), 4.52 (d, 2H, J = 3.6 Hz), 4.70 (d, 1H, J = 6.5
Hz), 4.78-4.86 (m, 4H), 4.94 (d, 1H, J = 6.5 Hz), 7.60 (d, 2H, J = 8.8 Hz), 8.14 (d, 2H, J
13 = 8.8 Hz); C NMR (DMK-d6) δ 38.49, 55.03, 56.46, 56.48, 56.56, 67.47, 77.93, 79.09,
79.34, 81.57, 84.64, 97.17, 99.03, 99.21, 99.35, 123.78, 131.49, 147.49, 148.41; HRMS
(ES) calcd for C21H33NO11 (M+Na) 498.1951, found 498.1993. 148 O O O
OONO O 2 O O O
2,6-Anhydro-7-deoxy-7-(4-nitrophenyl)-3,4,5-tri-O-acetyl-L-glycero-L-gulo- heptinoic acid methyl ester (3.19).
The MOM-protected glucoside 3.10 (4.14 g, 8.71 mmol) dissolved in methanol
(120 mL) was placed in a flask at rt. Aqueous HCl (6 N, 80 mL) was added and the solution stirred for 18 h. The mixture was then concentrated to a paste. Saturated
NaHCO3 solution (150 mL) was added along with KBr (200 mg) and TEMPO (200 mg).
Clorox (150 mL) was then added dropwise while monitoring the pH of the reaction.
NaOH (4 N) was used to maintain the pH between 9-10. After the addition of the
Clorox was complete, the reaction was stirred for 2 h. The mixture was washed (3x) with CH2Cl2 and the aqueous layer was acidified to pH 2 and concentrated to dryness.
The remaining solid was triturated with methanol, which was then transferred to a two-
neck flask where HCl gas was bubbled through the methanol at 40 °C for 5 h. The
reaction mixture was concentrated and placed in an rt water bath. Acetic anhydride (100
mL), pyridine (100 mL), and DMAP (75 mg) were added and the mixture stirred for 18 h.
The contents were diluted with water and extracted (3x) with ethyl acetate. The organic
layers were washed with water, brine, and dried (MgSO4). This suspension was filtered
and concentrated to a brown gum, dissolved in hot ethanol, and upon cooling gave 3.33 g
-1 (84 %) of white crystals, mp 133-134 °C. [α]D -17.8 (c 0.45, CH2Cl2); IR (cm ) 3078
(w), 2952 (w), 2850 (w), 1758 (s), 1603 (m), 1521 (s), 1346 (s), 1219 (s), 1036 (m); 1H 149 NMR (DMK-d6) δ 1.94 (s, 3H), 1.96 (s, 3H), 2.00 (s, 3H), 2.94 (dd, 1H, J = 8.6, 14.7
Hz), 3.12 (dd, 1H, J = 3.1, 14.7 Hz), 3.65 (s, 3H), 4.07 (dt, 1H, J = 3.1, 8.6 Hz), 4.21 (d,
1H, J = 9.8 Hz), 4.92 (t, 1H, J = 9.8 Hz), 5.06 (t, 1H, J = 9.8 Hz), 5.31 (t, 1H, J = 9.8 Hz),
13 7.56 (d, 2H, J = 8.7 Hz), 8.15 (d, 2H, J = 8.7 Hz); C NMR (DMK-d6) δ 20.38, 20.47,
20.62, 37.93, 52.72, 70.47, 72.33, 73.92, 76.30, 77.91, 123.92, 131.55, 146.37, 147.74,
168.26, 169.89, 170.29; HRMS (ES) calcd for C20H23NO11 (M+Na) 476.1169, found
476.1175.
O O O
OONH O 2 O O O
2,6-Anhydro-7-deoxy-7-(4-aminophenyl)-3,4,5-tri-O-acetyl-L-glycero-L-gulo-
heptinoic acid methyl ester (3.7).
To a pressure flask, flushed with argon, was added palladium on carbon (1.0 g)
and the nitro glucuronide 3.19 (5.79 g, 12.78 mmol) dissolved in ethyl acetate (200 mL).
The vessel was placed on a hydrogenator and, after sufficient flushing, hydrogen was
added and maintained at 40 p.s.i. After shaking for 6 h, the reaction mixture was filtered
through a bed of celite and the supernatant concentrated to give 5.3 g (98 %) of a white
-1 solid, mp 160-161 °C. [α]D -17.8 (c 0.675, CH2Cl2); IR (cm ) 3470 (m), 3368 (m), 3009
(w), 2948 (m), 1750 (s), 1623 (m), 1517 (s), 1431 (m), 1366 (s), 1228 (s), 1028 (s); 1H
NMR (CDCl3) δ 1.93-2.02 (m, 9H), 2.72-2.75 (m, 2H), 3.59-3.64 (m, 1H), 3.69 (s, 3H),
3.88 (d, 1H, J = 9.6 Hz), 4.92 (t, 1H, J = 9.6 Hz), 5.10-5.21 (m, 2H), 6.60 (d, 2H, J=8.2
150 13 Hz), 6.96 (d, 2H, J=8.2 Hz); C NMR (CDCl3) δ 19.84, 19.98, 36.53, 52.01, 69.14,
71.17, 73.13, 75.62, 78.20, 114.67, 126.31, 129.71, 143.96, 167.04, 168.78, 168.91,
169.63; HRMS (ES) calcd for C20H25NO9 (M+Na) 446.1427, found 446.1433.
O O O O
OON O H O O O
2,6-Anhydro-7-deoxy-7-[4-(retinamido)phenyl]-3,4,5-tri-O-acetyl-L-glycero-L-gulo- heptinoic acid methyl ester (3.21).
To a flame dried flask under argon atmosphere was added dried retinoic acid (2.0 g, 6.66 mmol) dissolved in dry THF (50 mL). The flask was charged with pyridine (0.8 mL, 10 mmol) and then chilled with an ice bath. Freshly distilled thionyl chloride (0.48 mL, 6.7 mmol) was carefully added and the red solution stirred for 30 min at 0 °C and 15 min at rt. Dried aniline sugar 3.7 (2.82 g, 6.66 mmol) and pyridine (1.1 mL, 13.3 mmol) dissolved in THF (100 mL) was added to the reaction via addition funnel and allowed to stir for 2 h. Methanol (20 mL) was added and stirred for 10 min. The reaction was diluted with water and extracted with ethyl acetate (4x). The organic layers were combined, washed with water and brine, dried (Na2SO4) under argon, and concentrated.
The red gel was dissolved in methanol and crystallized at -20°C. The solid was filtered
and washed with hexanes to give 3.85 g of yellow amorphous solid. HPLC of a
diazomethane treated sample of the solid revealed that there was 0.84 % of methyl
151 retinoate present as determined by area integration. Chromatograhy on silica gel (1:1 hexanes/ethyl acetate) yielded 0.15 g of more product, giving a total of 4.0 g (85 %). UV
λmax = 352 nm (ε = 20988); HPLC tR = 10.1 min (1 mL/min, 90:10 MeOH:H2O both with
-1 10 mM NH4OAc); IR (cm ) 3026 (w), 2916 (m), 1750 (s), 1668 (m), 1594 (m), 1513
1 (m), 1431 (m), 1358 (m), 1244 (s), 1211 (s), 1028 (m); H NMR (DMK-d6) δ 1.03 (s,
6H), 1.46-1.49 (m, 2H), 1.60-1.63 (m, 2H), 1.70 (s, 3H), 1.93 (s, 3H), 1.95 (s, 3H), 1.96
(s, 3H), 2.02 (s, 3H), 2.41 (s, 3H), 2.70-2.78 (m, 1H), 2.87 (dd, 1H, J = 3.3, 14.7 Hz),
3.65 (s, 3H), 3.91-3.96 (m, 1H), 4.18 (d, 1H, J = 9.8 Hz), 4.89 (t, 1H, J = 9.8 Hz), 5.05 (t,
1H, J = 9.8 Hz), 5.29 (t, 1H, J = 9.8 Hz), 6.01 (s, 1H), 6.15-6.36 (m, 4H), 7.07 (dd, 1H, J
= 11.4, 15.0 Hz), 7.17 (d, 2H, J = 8.4 Hz), 7.62 (d, 2H, J = 8.4 Hz), 9.13 (s, 1H); 13C
NMR (DMK-d6) δ 11.36, 12.09, 18.40, 18.90, 19.03, 19.13, 20.40, 32.23, 33.39, 36.40,
38.89, 51.16, 68.34, 71.08, 72.73, 75.02, 77.30, 118.26, 118.35, 121.88, 121.93, 127.19,
128.65, 129.03, 129.10, 129.44, 131.50, 135.48, 136.98, 137.12, 137.65, 137.75, 148.42,
164.05, 164.13, 166.94, 168.37, 168.57, 168.80; HRMS (ES) calcd for C40H51NO10
(M+Na) 728.3411, found 728.3405.
152
O O HO O HO OH N H OH
2,6-Anhydro-7-deoxy-7-[4-(retinamido)phenyl]-L-glycero-L-gulo-heptinoic acid (4-
HPRCG).
To a flask was added the protected glucuronide-retinoid conjugate 3.21 (1.76 g,
2.5 mmol) suspended in methanol (740 mL) to give a ~3.4 mM solution. Potassium carbonate (345 mg, 2.5 mmol) was added and allowed to stir for 20 h. The reaction mixture was concentrated at 25-30 °C to ~400 mL. Adjustment to the original volume with methanol was followed by addition of 5 N KOH (5 ml, 25 mmol). After stirring for
20 h, the reaction was cooled to 0°C and carefully adjusted to pH 7 with 4 N HCl. The reaction mixture was concentrated at 25-30°C to ~100 mL, cooled, and the pH carefully adjusted to 5 with 1 N HCl. The suspension was extracted with ethyl acetate until the aqueous layer was essentially clear. The organic layers were combined, dried (Na2SO4) under argon for 2 h, and carefully concentrated. The residue was chromatographed on reverse phase silica gel (80:20 methanol/water) to yield 1.22 g (86 %). UV λmax = 364
nm (ε = 45821); HPLC tR = 7.3 min (1 mL/min, 85:15 MeOH:H2O both with 10 mM
-1 NH4OAc); IR (cm ) 3319 (br), 3041 (w), 2932 (m), 1733 (m), 1648 (s), 1599 (m), 1517
1 (s), 1407 (m), 1358 (m), 1309 (m), 1256 (m), 1089 (m); H NMR (DMK-d6) δ 1.03 (s,
6H), 1.46-1.49 (m, 2H), 1.59-1.63 (m, 2H), 1.70 (s, 3H), 1.94-2.05 (m, 2H), 2.02 (s, 3H),
2.41 (s, 3H), 2.65-2.70 (m, 1H), 3.14-3.19 (m, 2H), 3.40-3.46 (m, 2H), 3.53 (t, 1H, J =
153 9.5 Hz), 3.72 (d, 1H, 9.5 Hz), 6.01 (s, 1H), 6.15-6.36 (m, 4H), 7.07 (dd, 1H, J = 11.4,
15.0 Hz), 7.21 (d, 2H, J = 8.3 Hz), 7.60 (d, 2H, J = 8.3 Hz), 9.11 (s, 1H); 13C NMR
(DMK-d6) δ 12.85, 13.54, 19.87, 21.91, 33.57, 34.85, 37.84, 40.32, 73.03, 74.02, 79.10,
79.62, 81.66, 119.65, 123.53, 128.61, 130.11, 130.52, 130.66, 130.97, 134.60, 137.01,
138.48, 138.57, 138.75, 139.18, 149.77, 165.57, 170.83; HRMS (ES) calcd for
C33H43NO7 (M+Na) 588.2937, found 588.2939.
6.4 4-HBRCG SYNTHESIS
O O O OH O OO O O O
2,6-Anhydro-1-deoxy-1-[4-(hydroxymethyl)phenyl]-3,4,5,7-tetra-O-
(methoxymethyl)-D-glycero-D-gulo-heptitol (4.4).
To a flame dried flask under argon atmosphere was added the exocyclic olefin
3.11 (3.75 g, 10.6 mmol) dissolved in dry THF (100 mL). 9-BBN-H (53.2 mL, 26.6
mmol, 0.5 M) was added via addition funnel. The flask was then equipped with a reflux
condenser, heated to 75-80 °C, and refluxed for 4.5 h. The mixture was cooled to rt, then
K3PO4 (10 mL, 3 M) was added and allowed to stir for 10 min. p-Bromobenzyl alcohol
(3.98 g, 21.3 mmol) and PdCl2(dppf) (0.686 g, 0.85 mmol) dissolved in DMF (100 mL)
was added via addition funnel and stirred for 18 h. The reaction was diluted with water
and ether, and then the layers were separated. The organic layer was washed with water
154 and brine. The combined aqueous layers were extracted with ether (3x). The organic layers were combined, dried (MgSO4), concentrated, and chromatographed (1:1 then 1:2 hexanes/ethyl acetate) to afford 3.29 g (67 %) of orange oil. [α]D -26.2 (c 1.15, DMK);
IR (cm-1) 3470 (w), 2932 (m), 2887 (m), 1692 (m), 1444 (w), 1150 (s), 1101 (s), 1024 (s),
1 918 (m); H NMR (DMK-d6) δ 2.60 (dd, 1H, J = 9.4, 14.4 Hz), 3.18-3.42 (m, 5H), 3.25
(s, 3H), 3.35 (s, 3H), 3.40 (s, 3H), 3.44 (s, 3H), 3.54-3.61 (m, 2H), 3.73 (dd, 1H, J = 1.8,
11.3 Hz), 4.51-4.58 (m, 4H), 4.70 (d, 1H, J = 6.5 Hz), 4.77-4.85 (m, 4H), 4.93 (d, 1H, J =
13 6.5 Hz), 7.25 (s, 4H); C NMR (DMK-d6) δ 38.35, 55.04, 56.45, 56.55, 64.44, 64.57,
67.42, 77.97, 79.07, 80.32, 81.63, 84.83, 97.20, 99.01, 99.19, 99.32, 127.15, 130.11,
138.75, 141.03; HRMS (ES) calcd for C22H36O10 (M+Na) 483.2206, found 483.2188.
O O O O O OO O O
O
2,6-Anhydro-1-deoxy-1-[4-(methoxymethyl)phenyl]-3,4,5,7-tetra-O-
(methoxymethyl)-D-glycero-D-gulo-heptitol (4.5).
To a flame dried flask under argon atmosphere was added the C-glycoside benzyl
alcohol 4.4 (2.44 g, 5.3 mmol) dissolved in dry THF (100 mL). Sodium hydride (0.63 g,
26.5 mmol) was added to the flask and the suspension stirred for 1.5 h. Iodomethane (4.5
g, 31.7 mmol) dissolved in THF (10 mL) was cannulated into the reaction mixture and
allowed to stir for 18 h. After cooling in an ice bath, water was added carefully to quench
excess NaH. The mixture was extracted with ether (3x), the organic layers combined,
155 washed with brine, dried (MgSO4), concentrated, and then chromatographed (1:1 then 1:2
hexanes/ethyl acetate) to give 2.37 g (94 %) of clear oil. [α]D -27.0 (c 4.70, DMK); IR
(cm-1) 2981 (s), 2883 (s), 1701 (w), 1513 (m), 1444 (m), 1378 (m), 1301 (m), 1158 (s),
1 1105 (s), 1028 (s), 918 (s); H NMR (DMK-d6) δ 2.61 (dd, 1H, J = 9.4, 14.4 Hz), 3.19-
3.42 (m, 5H), 3.24 (s, 3H), 3.30 (s, 3H), 3.35 (s, 3H), 3.40 (s, 3H), 3.44 (s, 3H), 3.54-3.64
(m, 2H), 3.73 (dd, 1H, J = 2.6, 13.5 Hz), 4.38 (s, 2H), 4.50 (d, 1H, J = 6.4 Hz), 4.54 (d,
1H, J = 6.4 Hz), 4.70 (d, 1H, J = 6.5 Hz), 4.77-4.85 (m, 4H), 4.93 (d, 1H, J = 6.5 Hz),
13 7.21 (d, 2H, J = 8.0 Hz), 7.28 (d, 2H, J = 8.0 Hz); C NMR (DMK-d6) δ 33.39, 55.05,
56.47, 56.49, 56.57, 57.97, 67.46, 74.84, 78.00, 79.10, 80.23, 81.66, 84.86, 97.20, 99.01,
99.21, 99.32, 128.15, 130.19, 137.23, 139.45; HRMS (ES) calcd for C23H38O10 (M+Na)
497.2363, found 497.2384.
O O O O OO O O O O
2,6-Anhydro-7-deoxy-7-[4-(methoxymethyl)phenyl]-3,4,5-tri-O-acetyl-L-glycero-L- gulo-heptinoic acid methyl ester (4.9).
The MOM-protected glucoside 4.5 (2.43 g, 5.12 mmol) dissolved in methanol
(500 mL) was placed in a flask at rt. Aqueous HCl (6 N, 26 mL) was added and the solution stirred for 18 h after which the mixture was then concentrated to dryness and set aside. In a separate flask, KBr (2.42 g, 20.38 mmol) and TEMPO (3.19 g, 20.41 mmol) were added to a saturated NaHCO3 solution (400 mL) and stirred for 20 min at 0 °C.
156 Aqueous NaOCl (11.2 mL, 1.6-2.0 M) was then added and stirred for 10 min. The
deprotected sugar was dissolved in saturated NaHCO3 solution (100 mL) and added to
the flask with the TEMPO mixture. The total mixture was stirred for 45 min at 0 °C.
Then the reaction was quenched with EtOH (50 mL) and the contents were washed with
ether in a separatory funnel. The aqueous layer was concentrated to dryness and the
remaining solid was exhaustively triturated with methanol. The methanol was then
concentrated and dried. The dried residue was suspended in DMF (180 mL) and then
iodomethane (6.4 g) dissolved in DMF (10 mL) was added and allowed to stir for 20 h
under argon at rt. The reaction mixture was then supplemented with acetic anhydride (40
mL), pyridine (20 mL), and DMAP (15 mg) and allowed to stir for 18 h. The reaction
mixture was diluted with water and extracted (3x) with ethyl acetate. The organic layers were washed with water, brine, dried (MgSO4), concentrated, and chromatographed (2:1 then 1:1 hexanes/ethyl acetate) to give 1.90 g (82 %) of clear oil that solidified upon
-1 standing, mp 84-86 °C. [α]D -13.04 (c 1.15, DMK); IR (cm ) 2956 (w), 2818 (w), 1750
1 (s), 1440 (m), 1370 (m), 1211 (s), 1105 (m), 1028 (m); H NMR (DMK-d6) δ 1.94 (s,
3H), 1.94 (s, 3H), 1.95 (s, 3H), 2.74-2.81 (m, 1H), 2.90 (dd, 1H, J = 3.4, 7.3 Hz), 3.30 (s,
3H), 3.65 (s, 3H), 3.94-3.99 (m, 1H), 4.18 (d, 1H, J = 9.8 Hz), 4.38 (S, 2H), 4.90 (t, 1H, J
= 9.8 Hz), 5.05 (t, 1H, J = 9.8 Hz), 5.29 (t, 1H, J = 9.8 Hz), 7.22 (s, 4H); 13C NMR
(DMK-d6) δ 20.39, 20.52, 20.60,38.12, 52.67, 58.03, 70.62, 72.53, 74.09, 74.73, 76.41,
78.62, 128.25, 130.16, 137.43, 137.76, 168.40, 169.89, 170.07, 170.30; HRMS (ES)
calcd for C22H28O10 (M+Na) 475.1580, found 475.1577.
157 O O O Br OO O O O O
2,6-Anhydro-7-deoxy-7-[4-(bromomethyl)phenyl]-3,4,5-tri-O-acetyl-L-glycero-L-
gulo-heptinoic acid methyl ester (4.12).
To a dry flask equipped with a CaSO4 drying tube was added the C-glucuronide methyl ether 4.9 (462 mg, 1.02 mmol) along with 30 % HBr in acetic acid (5 mL, 25 mmol) at 0 °C. The mixture stirred for 30 min and then at rt for 18 h. The mixture was diluted with methylene chloride and then carefully washed with water and saturated
NaHCO3 solution. The organic layer was dried (MgSO4), concentrated, and
chromatographed (2:1 then 1:1 hexanes/ethyl acetate) to give 440 mg (86 %) of white
foam, which was crystallized with ether, mp 116-117 °C. [α]D -12.03 (c 5.57, DMK); IR
(cm-1) 3026 (w), 2952 (w), 1754 (s), 1440 (m), 1370 (m), 1215 (s), 1101 (m), 1036 (m);
1 H NMR (DMK-d6) δ 1.93 (s, 3H), 1.94 (s, 3H), 1.95 (s, 3H), 2.76-2.83 (m, 1H), 2.92
(dd, 1H, J = 3.5, 7.3 Hz), 3.64 (s, 3H), 3.96-3.99 (m, 1H), 4.20 (d, 1H, J = 9.7 Hz), 4.62
(s, 2H), 4.90 (t, 1H, J = 9.7 Hz), 5.05 (t, 1H, J = 9.7 Hz), 5.29 (t, 1H, J = 9.7 Hz), 7.25 (d,
13 2H, J=8.2 Hz), 7.36 (d, 2H, J=8.2 Hz); C NMR (DMK-d6) δ 20.40, 20.52, 20.63, 34.37,
38.12, 52.69, 70.58, 72.52, 74.04, 76.35, 78.43, 129.88, 130.68, 137.26, 138.67, 168.39,
169.91, 170.09, 170.29; HRMS (ES) calcd for C21H25BrO9 (M+Na) 523.0580, found
523.0602.
158 O O O O OO O O O O
2,6-Anhydro-1-deoxy-1-[4-(methoxymethyl)phenyl]-3,4,5,7-tetra-O-acetyl-D-glycero-
D-gulo-heptitol (4.20).
The MOM-protected glucoside 4.6 (0.643 g, 1.35 mmol) dissolved in methanol
(34 mL) was placed in a flask at rt. Aqueous HCl (6 N, 6.7 mL) was added and the
solution stirred for 18 h. The mixture was then concentrated to dryness. Acetic
anhydride (4 mL) and pyridine (3 mL) were added to the paste along with a catalytic amount of DMAP and the mixture was allowed to stir for 18 h at rt. The reaction was diluted with water and extracted with ethyl acetate (3x). The organic layers were combined, washed with water and brine, dried (MgSO4), filtered, concentrated, and
chromatographed (1:1 hexanes/ethyl acetate) to give 570 mg (90 %) of white solid, mp
-1 120-122 °C. [α]D -4.0 (c 0.78, DMK); IR (cm ) 2940 (w), 2862 (w), 1750 (s), 1436 (w),
1 1370 (m), 1224 (s), 1105 (m), 1032 (m); H NMR (CDCl3) δ 1.96-2.02 (m, 12H), 2.78 (s,
2H, J = 5.8 Hz), 3.36 (s, 3H), 3.52-3.57 (m, 2H), 4.02 (dd, 1H, J = 2.3, 12.2 Hz), 4.20
(dd, 1H, J = 5.3, 12.2 Hz), 4.40 (s, 2H), 4.92 (t, 1H, J = 9.6 Hz), 5.03 (t, 1H, J = 9.6 Hz),
5.15 (t, 1H, J = 9.6 Hz), 7.16 (d, 2H, J=8.0 Hz), 7.23 (d, 2H, J=8.0 Hz); 13C NMR
(DMK-d6) δ 20.55, 20.57, 20.65, 38.08, 58.02, 63.05, 69.70, 72.81, 74.76, 74.86, 76.05,
78.60, 128.20, 130.21, 137.65, 137.75, 170.04, 170.16, 170.37, 170.59; HRMS (ES)
calcd for C23H30O10 (M+Na) 489.1737, found 489.1727.
159 O O O Br OO O O O O
2,6-Anhydro-1-deoxy-1-[4-(bromomethyl)phenyl]-3,4,5,7-tetra-O-acetyl-D-glycero-
D-gulo-heptitol (4.21).
To a dry flask equipped with a drying tube was added the C-glycoside methyl ether 4.20 (0.54 g, 1.16 mmol) along with 30 % HBr in acetic acid (5 mL, 25 mmol) at 0
°C. The mixture was stirred for 30 min and then at rt for 18 h. The mixture was diluted with methylene chloride and then carefully washed with water and saturated NaHCO3 solution. The organic layer was dried (MgSO4), concentrated, and chromatographed (2:1
then 1:1 hexanes/ethyl acetate) to give 593 mg (97 %) of white solid, mp 141-142 °C.
-1 [α]D -4.67 (c 2.57, DMK); IR (cm ) 2993 (w), 2952 (w), 1754 (s), 1440 (w), 1374 (m),
1 1244 (s), 1105 (w), 1052 (m); H NMR (DMK-d6) δ 1.92-1.98 (m, 12H), 2.72 (dd, 1H, J
= 7.3, 8.6 Hz), 2.88 (dd, 1H, J = 3.2, 7.3 Hz), 3.77-3.85 (m, 2H), 4.00 (dd, 1H, J = 2.4,
6.1 Hz), 4.21 (dd, 1H, J = 5.9, 6.1 Hz), 4.63 (s, 2H), 4.86 (t, 1H, J = 9.6 Hz), 4.97 (t, 1H,
J = 9.6 Hz), 5.22 (t, 1H, J = 9.6 Hz), 7.25 (d, 2H, J=8.2 Hz), 7.37 (d, 2H, J=8.2 Hz); 13C
NMR (DMK-d6) δ 20.49, 20.60, 34.32, 38.07, 63.03, 69.73, 72.82, 74.85, 76.07, 78.40,
129.72, 130.68, 137.12, 138.92, 169.97, 170.10, 170.30, 170.52; HRMS (ES) calcd for
C22H27BrO9 (M+Na) 537.0736, found 537.0724.
160
Si O
N
tert-Butyl-dimethylsilylcyanohydrin of retinal (4.14).
To a flame dried flask under argon atmosphere was added retinal (1.03 g, 3.62 mmol) dissolved in dry CH2Cl2 (50 mL). A catalytic amount of Et3N (0.1 mL) was added
then tert-butyldimethylsilyl cyanide (1.0 g, 7.08 mmol) dissolved in CH2Cl2 (10 mL) was
added by cannulation. The reaction stirred for 20 h after which the solution was
concentrated, chromatographed (95:5 hexanes/ethyl acetate), dried (Na2SO4) under argon,
and subjected to vacuum overnight to give 1.20 g (78 %) of orange oil. UV λmax = 329 nm (ε = 49462); IR (cm-1) 3042 (w), 2960 (s), 2928 (s), 2850 (s), 2239 (w), 1586 (w),
1 1472 (m), 1358 (m), 1256 (m), 1105 (s), 963 (s), 832 (s), 775 (m); H NMR (DMK-d6) δ
0.16 (s, 3H), 0.20 (s, 3H), 0.90 (s, 9H), 1.02 (s, 6H), 1.45-1.48 (m, 2H), 1.58-1.63 (m,
2H), 1.70 (s, 3H), 1.99 (s, 6H), 5.57-5.61 (m, 2H), 6.13-6.23 (m, 3H), 6.38 (d, 1H, J =
15.2 Hz), 6.86 (dd, 1H, J = 11.3, 15.2 Hz); HRMS (ES) calcd for C27H43NOSi (M+Na)
448.3012, found 448.2982.
161 O O O O
OO O O O O
2,6-Anhydro-7-deoxy-7-[4-(retinoylmethyl)-phenyl]-3,4,5-tri-O-acetyl-L-glycero-L- gulo-heptinoic acid methyl ester (4.17).
To a flame dried flask under argon atmosphere was added THF (40 mL) along with LiHMDS (1.0 M in hexanes, 3.8 mL, 3.8 mmol). The mixture was cooled to –78 °C upon which the silyl cyanohydrin of retinal 4.14 (1.08 g, 2.54 mmol) in THF (15 mL) was added by cannulation into the flask. The dark red solution was allowed to stir for 30 min at –78 °C. The crystalline bromoglucuronide 4.12 (2.78 g, 5.56 mmol) in THF (15 mL) was cannulated into the flask and the mixture stirred for 3 h at –78 °C after which the solution changed to light red. The reaction was taken out of the cold bath and quenched with a solution of 1 M NH4Cl (10 mL). The mixture was extracted with ethyl
acetate (3x) and the organic layers were combined, washed with brine, dried (Na2SO4), filtered, concentrated, and chromatographed (2:1 hexanes/ethyl acetate) to give 1.0 g (47
%) of yellow foam 4.16 and 1.7 g of recovered bromide 4.12. The alkylated product was taken up in 1 % aqueous THF (200 mL) and chilled to 0 °C. TBAF (309 mg, 1.18 mmol) was added and the darkened solution stirred 1 h. The reaction was diluted with water and extracted with ethyl acetate (3x). The organic layers were combined, washed with brine, dried (NaSO4), filtered, concentrated, and chromatographed (2:1 hexanes/ethyl acetate) to
give 628 mg (35 % over two steps) of yellow foam. UV λmax = 379 nm (ε = 36182);
- HPLC tR = 24.0 min, 1 mL/min (85:15 MeOH:H2O both with 10 mM NH4OAc); IR (cm 162 1) 2956 (w), 2924 (w), 2863 (w), 1754 (s), 1672 (w), 1554 (m), 1436 (w), 1362 (w), 1215
1 (s), 1081 (w), 1028 (w), 971 (w); H NMR (DMK-d6) δ 1.02 (s, 6H), 1.45-1.48 (m, 2H),
1.58-1.62 (m, 2H), 1.69 (s, 3H), 1.90 (s, 3H), 1.93 (s, 3H), 1.95 (s, 3H), 2.01 (s, 3H),
2.03-2.05 (m, 2H), 2.28, (s, 3H), 2.75-2.89 (m, 2H), 3.64 (s, 3H), 3.71 (s, 2H), 3.95-3.98
(m, 1H), 4.19 (d, 1H, J = 9.8 Hz), 4.90 (t, 1H, J = 9.8 Hz), 5.05 (t, 1H, J = 9.8 Hz), 5.29
13 (t, 1H, J = 9.8 Hz), 6.15-6.35 (m, 5H), 7.13-7.20 (m, 5H); C NMR (DMK-d6) δ 13.45,
14.68, 20.41, 20.96, 21.08, 21.15, 22.47, 34.15, 35.41, 38.76, 40.86, 52.25, 53.23, 71.18,
73.09, 74.63, 76.93, 79.13, 126.89, 129.86, 130.73, 130.93, 131.01, 131.47, 133.69,
134.89, 135.11, 137.14, 137.26, 138.95, 139.09, 140.96, 152.68, 168.97, 170.45, 170.64,
170.84, 198.78; HRMS (ES) calcd for C41H52O10 (M+Na) 727.3458, found 727.3456.
O O O O
OO O O O O
2,6-Anhydro-7-deoxy-7-[4-(retinoylmethyl)-phenyl]-3,4,5,7-tetra-O-acetyl-D-glycero-
D-gulo-heptinol (4.22).
To a flame dried flask under argon atmosphere was added THF (10 mL) along
with LiHMDS (1.0 M in hexanes, 0.78 mL, 0.78 mmol). The mixture was cooled to –78
°C upon which the silyl cyanohydrin of retinal 4.14 (218 mg, 51 mmol) in THF (5 mL)
was added by cannulation into the flask. The dark red solution was allowed to stir for 30 min at –78 °C. The crystalline glucoside bromide 4.21 (277 mg, 0.53 mmol) in THF (5
mL) was cannulated into the flask and the mixture stirred for 2 h at –78 °C after which
163 the solution changed to light red. The reaction was taken out of the cold bath and
quenched with a solution of 1 M NH4Cl (1 mL). The mixture was extracted with ethyl
acetate (3x) and the organic layers were combined, washed with brine, dried (Na2SO4), filtered, and concentrated. The crude alkylated product was taken up in 1 % aqueous
THF (20 mL) and chilled to 0 °C. TBAF (134 mg, 0.51 mmol) was added and the darkened solution stirred overnight while warming to rt. The reaction was diluted with water and extracted with ethyl acetate (3x). The organic layers were combined, washed with brine, dried (NaSO4), filtered, concentrated, and chromatographed (2:1
hexanes/ethyl acetate) to give 132 mg (36 % over two steps) of yellow foam. 1H NMR
(DMK-d6) δ 1.00 (s, 6H), 1.43-1.46 (m, 2H), 1.58-1.59 (m, 2H), 1.68 (s, 3H), 1.90 (s,
3H), 1.92 (s, 3H), 1.94 (s, 3H), 1.95 (s, 3H), 2.26, (s, 3H), 2.70-2.87 (m, 2H), 3.70 (s,
2H), 3.74-3.83 (m, 2H), 3.97 (d, 1H, J = 12.0 Hz), 4.19 (dd, 1H, J = 5.9, 12.0 Hz) 4.85
(t, 1H, J = 9.6 Hz), 4.95 (t, 1H, J = 9.6 Hz), 5.20 (t, 1H, J = 9.6 Hz), 6.15-6.37 (m, 5H),
7.10-7.17 (m, 5H); HRMS (ES) calcd for C42H54O10 (M+Na) 741.3615, found 741.3617.
O O HO O HO OH OH
2,6-Anhydro-7-deoxy-7-[4-(retinoylmethyl)-phenyl]-L-glycero-L-gulo-heptinoic acid
(4-HBRCG).
To a flask was added the protected glucuronide-retinoid conjugate 4.17 (1.15 g,
1.64 mmol) dissolved in methanol (500 mL) and chilled to 4 ºC. Potassium carbonate
(136 mg, 0.98 mmol) was added and allowed to stir for 20 h. The reaction mixture was 164 concentrated at 25-30 °C to ~200 mL. Adjustment to the original volume with methanol
was followed by addition of 1 N KOH (14 ml, 14 mmol). After stirring for 20 h at 4 ºC,
the reaction was warmed and allowed to stir for 5 h at rt. The reaction was then cooled to
0 °C and carefully adjusted to pH 7 with 4 N HCl. The reaction mixture was
concentrated at 25-30°C to ~100 mL, cooled back to 0 °C, and the pH carefully adjusted
to 3 with 1 N HCl. The suspension was extracted with ethyl acetate and the organic
layers were combined, dried (Na2SO4) under argon for 2 h, and carefully concentrated.
The residue was chromatographed on reverse phase silica gel (gradient 70:30 to 85:15
methanol/water) to yield 759 mg (82 %) of yellow foam. UV λmax = 382 nm (ε = 30019);
-1 HPLC tR = 9.2 min (1 mL/min, 85:15 MeOH:H2O both with 10 mM NH4OAc); IR (cm )
3384 (br), 2920 (s), 1721 (m), 1664 (s), 1550 (s), 1427 (m), 1362 (m), 1232 (w), 1089
1 (m), 1052 (m), 1102 (s), 967 (w); H NMR (MeOH-d4) δ 0.94 (s, 6H), 1.38-1.41 (m, 2H),
1.54-1.58 (m, 2H), 1.61 (s, 3H), 1.91 (s, 3H), 1.93-1.96 (m, 2H), 2.20 (s, 3H), 2.60 (dd,
1H, J = 8.7, 14.4 Hz), 3.03-3.08 (m, 2H), 3.21-3.29 (m, 2H), 3.37 (t, 1H, J = 9.5 Hz),
3.53 (d, 1H, J = 9.5 Hz), 3.62 (s, 2H), 6.03-6.25 (m, 5H), 7.02-7.16 (m, 5H); 13C NMR
(MeOH-d4) δ 12.88, 14.45, 20.29, 21.04, 21.93, 29.40, 34.00, 35.25, 38.34, 40.76, 52.11,
73.44, 74.63, 79.28, 80.36, 82.36, 124.37, 126.26, 129.92, 130.19, 130.86, 131.01,
131.12, 134.00, 134.23, 136.82, 138.70, 138.91, 139.04, 141.12, 154.22, 173.26, 201.21;
HRMS (ES) calcd for C34H44O7 (M+Na) 587.2985, found 587.2989.
165
O HO O
HO OH OH
2,6-Anhydro-7-deoxy-7-[4-(retinoylmethyl)-phenyl]-D-glycero-D-gulo-heptinol
(4.23).
To a flask was added the protected glucoside-retinoid conjugate 4.22 (130 mg,
0.18 mmol) dissolved in methanol (75 mL) and chilled to 4 ºC. Potassium carbonate (25 mg, 0.18 mmol) was added and allowed to stir for 20 h. The reaction was then cooled to
0 °C and carefully adjusted to pH 5 with 1 N HCl. The solution was extracted with ethyl acetate and the organic layers were combined, dried (Na2SO4), and carefully
concentrated. The residue was chromatographed on reverse phase silica gel (gradient
70:30 to 85:15 methanol/water) to yield 49 mg (49 %) of yellow foam. UV λmax = 380
nm (ε = 34567); HPLC tR = 13.8 min (1 mL/min, 85:15 MeOH:H2O both with 10 mM
-1 NH4OAc); IR (cm ) 3388 (br), 2948 (m), 2846 (m), 1652 (m), 1554 (w), 1456 (w), 1415
1 (w), 1113 (w), 1056 (m), 1016 (s), 694 (br); H NMR (DMK-d6) δ 1.02 (s, 6H), 1.45-1.48
(m, 2H), 1.58-1.64 (m, 2H), 1.69 (s, 3H), 2.01 (s, 3H), 2.03-2.05 (m, 2H), 2.29, (s, 3H),
2.65 (dd, 1H, J = 8.3, 14.3 Hz), 3.09-3.38 (m, 6H), 3.57-3.60 (m, 1H), 3.70 (s, 3H), 6.15-
13 6.35 (m, 5H), 7.12-7.26 (m, 5H); C NMR (DMK-d6) δ 13.43, 14.66, 20.41, 22.46,
26.31, 30.95, 34.14, 38.69, 40.85, 52.34,63.81, 72.70, 75.05, 80.37, 81.31, 81.50, 129.82,
130.43, 130.90, 131.20, 131.48, 133.64, 134.46, 137.31, 138.96, 139.08, 140.89, 152.57,
198.97; HRMS (ES) calcd for C34H46O6 (M+Na) 573.3192, found 573.3204.
166 6.5 MISCELLANEOUS COMPOUNDS
O O O O OO O O O O O
2,6-Anhydro-7-deoxy-7-[4-(hydroxymethyl)-phenyl]-3,4,5-tri-O-acetyl-L-glycero-L- gulo-heptinoyl retinoate (4.18).
This product was synthesized unintentionally as a byproduct from the alkylation of glucuronide bromide and silylcyanohydrin due to insufficiently dry starting materials.
UV λmax = 358 nm (ε = 46779); HPLC tR = 17.2 min, 1 mL/min, 90:10 MeOH:H2O (10
-1 mM NH4OAc); IR (cm ) 2960 (w), 2923 (w), 2862 (w), 1754 (s), 1709 (m), 1607 (w),
1578 (w), 1436 (m), 1366 (m), 1240 (s), 1219 (s), 1142 (s), 1028 (w); 1H NMR (DMK- d6) δ 1.02 (s, 6H), 1.45-1.48 (m, 2H), 1.60-1.63 (m, 2H), 1.70 (s, 3H), 1.93 (s, 3H), 1.94
(s, 3H), 1.95 (s, 3H), 2.36, (s, 3H), 2.78-2.90 (m, 2H), 3.64 (s, 3H), 3.97-3.98 (m, 1H),
4.19 (d, 1H, J = 9.8 Hz), 4.90 (t, 1H, J = 9.8 Hz), 5.05 (t, 1H, J = 9.8 Hz), 5.11 (s, 2H),
5.29 (t, 1H, J = 9.8 Hz), 5.86 (s, 1H), 6.15-6.35 (m, 4H), 7.15 (dd, 1H, J = 11.4, 15.0 Hz),
13 7.26 (d, 2H, J = 8.2 Hz), 7.31 (d, 2H, J = 8.2 Hz); C NMR (DMK-d6) δ 12.82, 13.84,
19.81, 20.33, 20.46, 20.52, 21.82, 33.54, 34.81, 38.11, 40.28, 52.60, 65.64, 70.58, 72.51,
74.10, 76.38, 78.53, 118.88, 128.84, 129.16, 130.28, 130.26, 130.63, 132.16, 135.89,
136.00, 138.06, 138.31, 138.48, 140.32, 154.05, 166.97, 168.33, 169.80, 170.02, 170.22;
LRMS (ES) calcd for C41H52O11 (M+Na) 743.3, found 743.7.
167 O O HO
O HO OH OH O
2,6-Anhydro-7-deoxy-7-[4-(hydroxymethyl)-phenyl]-L-glycero-L-gulo-heptinoyl
retinoate (4.19).
The protected ester 4.18 (60 mg, 0.08 mmol) was dissolved in methanol (30 mL)
and placed in a cold box (4 °C). Potassium carbonate (5 mg) was added and the reaction
stirred for 24 h. Next, potassium hydroxide (0.1 M, 5 mL) was added and the mixture
stirred for 24 h. The reaction was then placed on ice and acidified with HCl until pH 7.
The reaction was then concentrated to approximately 5 mL. The mixture was further
acidified until pH 3. The contents were extracted with ethyl acetate (3x) and the organic
fractions were combined, dried (NaSO4), concentrated, and chromatographed on reverse phase silica gel (gradient 50:50 to 90:10 methanol/water) to yield 5 mg (10 %) of orange oil. UV λmax = 358 nm (ε = 37149); HPLC tR = 12.9 min (1 mL/min, 85:15 MeOH:H2O
1 both with 10 mM NH4OAc); H NMR (MeOH-d4) δ 1.02 (s, 6H), 1.47-1.50 (m, 2H),
1.61-1.67 (m, 2H), 1.70 (s, 3H), 2.00 (s, 3H), 2.71-2.77 (m, 1H), 3.11-3.19 (m, 2H), 3.34-
3.54 (m, 4H), 5.09 (s, 2H), 5.83 (s, 1H), 6.15-6.35 (m, 4H), 7.09 (dd, 1H, J = 11.4, 16.0
13 Hz), 7.25 (d, 2H, J = 8.0 Hz), 7.32 (d, 2H, J = 8.0 Hz); C NMR (DMK-d6) δ 12.87,
14.02, 20.31, 21.92, 29.41, 33.99, 35.25, 38.37, 40.76, 66.58, 73.78, 74.64, 79.55, 82.10,
119.15, 128.98, 129.59, 130.73, 130.96, 132.56, 135.59, 136.39, 138.97, 140.34, 140.68,
154.90, 168.45; LRMS (ES) calcd for C34H44O8 (M+Na) 603.3, found 603.6.
168
O HO O HO OH OH O
2,6-Anhydro-7-deoxy-7-[4-(hydroxymethyl)-phenyl]-D-glycero-D-gulo-heptinol
retinoate.
This product was obtained from the deprotection of the tetra-O-acetyl-glucoside
analog of 4.18, which was synthesized unintentionally as a byproduct from the alkylation
of glucoside bromide and silylcyanohydrin due to insufficiently dry starting materials.
UV λmax = 361 nm (ε = 29160); HPLC tR = 20.7 min (1 mL/min, 85:15 MeOH:H2O both
-1 with 10 mM NH4OAc); IR (cm ) 3380 (br), 2956 (m), 2920 (m), 2858 (m), 1701 (s),
1607 (w), 1578 (m), 1436 (m), 1350 (m), 1236 (m), 1142 (s), 1085 (m), 1032 (w); 1H
NMR (DMK-d6) δ 1.02 (s, 6H), 1.45-1.48 (m, 2H), 1.60-1.63 (m, 2H), 1.70 (s, 3H), 2.37
(s, 3H), 2.69 (dd, 1H, J = 8.3, 14.4 Hz), 3.15-3.18 (m, 3H), 3.26-3.28 (m, 1H), 3.56-3.38
(m, 2H), 3.59-3.60 (m, 1H), 3.71-3.76 (m, 1H), 5.11 (s, 2H), 5.86 (s, 1H), 6.17-6.43 (m,
13 4H), 7.15 (dd, 1H, J = 11.4, 15.0 Hz), 7.27-7.32 (m, 4H); C NMR (DMK-d6) δ
12.84,13.91, 19.86, 21.81, 26.08, 33.59, 34.86, 38.25, 40.37, 63.34, 65.80, 72.26, 74.48,
79.84, 80.78, 80.88, 119.00, 128.64, 129.18, 130.59, 130.65, 132.12, 136.06, 138.34,
138.55, 140.10, 140.29, 153.94, 167.05.
169 O OOO O
OO O O O O
Tetra-O-acetyl-β-D-glucopyanuronic acid methyl ester (3.9).
D-Glucurono-3,6-lactone (3.13) (8.8 g, 50 mmol) was mixed with MeOH (50 mL) and stirred until dissolved. Sodium methoxide (150 mg, 2.8 mmol) was added and stirred at rt for 1 h. The reaction was concentrated to a tan foam upon which acetic anhydride
(34 mL) and perchloric acid (70 %, 0.3 mL in 5 mL Ac2O) was added. The reaction was
allowed to stir and left to stand at 0 ºC overnight. The resultant solid was collected and
crystallized with EtOH to give 3.7 g (20 %) of white solid, mp 178-179 °C (lit. mp 177-
142 1 178 ºC). H NMR (CDCl3) δ 2.00 (s, 9H), 2.08 (s, 3H), 3.70 (s, 3H), 4.14 (d, 1H, J =
9.0 Hz), 5.10 (t, 1H, J = 9.0 Hz), 5.18-5.30 (m, 2H), 5.73 (d, 1H, J = 9.0 Hz).
O O Si O O
OO O O O O
2,6-Anhydro-3,4,5-tri-O-acetyl-7-diethylmethylsiloxy-L-glycero-L-gulo-heptinoic
acid methyl ester (3.14).
To a CO flushed pressure flask was added distilled ethyldimethyl silane (0.92 mL,
6.4 mmol) and Co2(CO)8 (15 mg, 0.04 mmol, Sigma-Aldrich). CO gas was applied to a
pressure of 15 psi and the mixture was allowed to stir for 5 min. The pressure was
released and methyl-tetra-O-acetyl-β-glucuronide (3.9) (400 mg, 1.06 mmol) dissolved in 170 dry CH2Cl2 (5 mL) was added. The tan reaction mixture stirred at 15 psi of CO for 65 h
in the dark. The flask was depressurized and pyridine (1 mL) was added to quench the
catalyst. The solution was applied to a silica gel column filter and eluted with CH2Cl2.
The concentrated eluent was chromatographed (2:1 hexanes/ethyl acetate) to give 30 mg
-1 (7.4 %) of clear oil. [α]D -13.4 (c 3.81, DMK); IR (cm ) 2956 (s), 2879 (s), 1754 (s),
1 1436 (m), 1374 (s), 1224 (s), 1113 (s), 1036 (s), 795 (m), 755 (m); H NMR (CDCl3) δ
0.2 (s, 3H), 0.51-0.58 (m, 4H), 0.87-0.95 (m, 6H), 1.98-2.00 (m, 9H), 3.51-3.55 (m, 1H),
3.67-3.68 (m, 2H), 3.70 (s, 3H), 3.95 (d, 1H, J = 9.6 Hz), 5.00 (t, 1H, J = 9.6 Hz), 5.11-
13 5.27 (m, 2H); C NMR (CDCl3) δ -5.07, 6.08, 6.11, 6.58, 20.50, 21.03, 52.62, 62.33,
68.73, 69.55, 73.66, 76.23, 78.92, 167.49, 169.42, 170.33; HRMS (ES) calcd for
C19H32O10Si (M+Na) 471.1662, found 471.1662.
O O O Br
O O O O O O
2,6-Anhydro-3,4,5-tri-O-acetyl-7-bromo-L-glycero-L-gulo-heptinoic acid methyl
ester (3.8).
To a flamed dried flask under argon atmosphere was added the siloxymethyl-
glucuronide analog 3.14 (326 mg, 0.73 mmol) dissolved in dry CH2Cl2 (50 mL) at 0ºC.
Dibromotriphenylphosphorane (400 mg, 1.42 mmol) dissolved in dry CH2Cl2 (10 mL)
was added via cannula and the reaction stirred at rt overnight. The reaction was diluted
with CH2Cl2 and washed with water. The organic layer was dried (MgSO4),
171 concentrated, and chromatographed (2:1 hexanes/ethyl acetate) to give 265 mg (88 %) of
-1 white solid, mp 165-166 °C. [α]D -13.5 (c 2.67, DMK); IR (cm ) 2973 (w), 2948 (w),
1750 (s), 1456 (m), 1431 (m), 1378 (m), 1228, (s), 1109 (m), 1036 (m), 983 (w), 901 (w);
1 H NMR (CDCl3) δ 1.99-2.04 (m, 9H), 3.38 (dd, 1H, J = 6.6, 11.4 Hz), 3.45 (dd, 1H, J =
2.5, 11.4 Hz), 3.69-3.74 (m, 4H), 4.02 (d, 1H, J = 9.6 Hz), 5.01 (t, 1H, J = 9.6 Hz), 5.15-
13 5.27 (m, 2H); C NMR (CDCl3) δ 20.47, 20.60, 20.64, 52.94, 69.37, 70.56, 73.06, 76.24,
167.10, 169.35, 170.18; HRMS (ES) calcd for C14H19BrO9 (M+Na) 433.0110, found
433.0134.
O O O I
O O O O O O
2,6-Anhydro-3,4,5-tri-O-acetyl-7-iodo-L-glycero-L-gulo-heptinoic acid methyl ester
(4.28)
To a dry flask equipped with a reflux condenser was added the bromomethyl- glucuronide 3.8 (216 mg, 0.53 mmol) dissolved in acetone (30 mL). Sodium iodide (1.57 g, 10.5 mmol) was added and the reaction was allowed to stir at reflux under argon for 48 h. Precipatate formed after several hours of refluxing. The reaction was cooled and diluted with ethyl acetate. The organic layer was washed with water (2x) after which the aqueous layer was extracted with ethyl acetate (3x). The organic layers were combined, dried (MgSO4), concentrated, and chromatographed (1:1 hexanes/ethyl acetate) to give
-1 230 mg (95 %) of white solid, mp 177-178 °C. [α]D -20.0 (c 2.40, DMK); IR (cm ) 2952
172 (w), 1758 (s), 1448 (w), 1374 (m), 1207, (s), 1113 (m), 1032 (m), 975 (w), 897 (w); 1H
NMR (CDCl3) δ 2.00-2.04 (m, 9H), 3.16 (dd, 1H, J = 7.4, 11.2 Hz), 3.28 (dd, 1H, J = 2.8,
11.2 Hz), 3.47-3.52 (m, 1H), 3.74 (s, 3H), 4.02 (d, 1H, J = 9.5 Hz), 4.92 (t, 1H, J = 9.5
13 Hz), 5.15-5.26 (m, 2H); C NMR (CDCl3) δ 2.18, 20.47, 20.59, 20.68, 52.92, 69.47,
71.83, 72.84, 76.18, 77.40, 167.11, 169.35, 170.15; HRMS (ES) calcd for C14H19IO9
(M+Na) 480.9972, found 480.9980.
O O O Br
OO O O O O 2,6-Anhydro-1-bromo-1-deoxy-3,4,5,7-tetra-O-acetyl-D-glycero-D-gulo-heptinol.
To a flamed dried flask under argon atmosphere was added the tetra-O-acetyl-
siloxymethyl-glucose analog (352 mg, 0.76 mmol) dissolved in dry CH2Cl2 (25 mL) at
0ºC. Dibromotriphenylphosphorane (400 mg, 0.95 mmol) dissolved in dry CH2Cl2 (10
mL) was added via cannula and the reaction stirred at rt overnight. The reaction was
diluted with CH2Cl2 and washed with water. The organic layer was dried (MgSO4),
concentrated, and chromatographed (1:1 hexanes/ethyl acetate) to give 249 mg (77 %) of
174 - white solid, mp 115-117 °C (lit. mp 119-120 ºC). [α]D -6.84 (c 1.02, CHCl3); IR (cm
1) 2994 (w), 2875 (w), 1758 (s), 1423 (w), 1366 (m), 1232 (s), 1097 (w), 1032 (m) 901
1 (w); H NMR (CDCl3) δ 2.00-2.10 (m, 12H), 3.32-3.49 (m, 2H), 3.64-3.74 (m, 2H), 4.13
(dd, 1H, J = 3.8, 19.8 Hz), 4.27 (dd, 1H, J = 8.0, 19.8 Hz), 4.97-5.25 (m, 3H).
173 O O O OH OO O O O O
2,6-Anhydro-7-deoxy-7-[4-(hydroxymethyl)phenyl]-3,4,5-tri-O-acetyl-L-glycero-L- gulo-heptinoic acid methyl ester.
The bromomethyl glucuronide analog 4.12 (30 mg, 0.06 mmol) was dissolved in
5 mL of acetone / water (90:10). Silver carbonate (48 mg, 0.17 mmol) was added and the suspension stirred overnight. The reaction was filtered through celite and the supernatant concentrated and chromatographed (1:1 hexanes/ethyl acetate) to give 20 mg (76 %) of white solid. IR (cm-1) 3515 (br), 3017 (w), 2952 (w), 2867 (w), 1750 (s), 1440 (m), 1374
1 (m), 1219 (s), 1105 (w), 1028 (m); H NMR (DMK-d6) δ 1.93-1.98 (m, 12H), 2.76 (dd,
1H, J = 8.2, 14.6 Hz), 2.90 (dd, 1H, J = 3.3, 14.6 Hz), 3.65 (s, 3H), 3.93-3.98 (m, 1H),
4.18 (d, 1H, J = 9.7 Hz), 4.58 (d, 1H, J = 5.9 Hz), 4.89 (t, 1H, J = 9.7 Hz), 5.04 (t, 1H, J
= 9.7 Hz), 5.29 (t, 1H, J = 9.7 Hz), 7.20 (d, 2H, J=8.1 Hz), 7.25 (d, 2H, J=8.1 Hz); 13C
NMR (DMK-d6) δ 20.31, 20.43, 20.52, 38.07, 52.57, 64.35, 64.47, 70.62, 72.52, 74.15,
76.45, 78.69, 127.25, 130.05, 136.66, 141.45, 168.34, 169.78, 169.99, 170.22; HRMS
(ES) calcd for C21H26O10 (M+Na) 461.1424, found 461.1420.
174
BIBLIOGRAPHY
1. Silverstein, R. M.; Webster, F. X. In Spectrometric identification of organic compounds; John Wiley and Sons, Inc.: New York, 1998.
2. Young, D. W. Stereochemistry of metabolic reactions of amino acids. Top. Stereochem. 1994, 21, 381-465.
3. Mann, J. In Secondary metabolism; Clarendon Press: Oxford, 1987.
4. Williamson, M. P.; Havel, T. F.; Wuthrich, K. Solution conformation of proteinase inhibitor IIA from bull seminal plasma by proton NMR and distance geometry. J. Molec. Biol. 1985, 182, 295-315.
5. Ferentz, A. E.; Wagner, G. NMR spectroscopy: A multifaceted approach to macromolecular structure. Q. Rev. Biophys. 2000, 33, 29-65.
6. Fesik, S. W. NMR studies of molecular complexes as a tool in drug design. J. Med. Chem. 1991, 34, 2937-2945.
7. Goto, N. K.; Kay, L. E. New developments in isotope labeling strategies for protein solution NMR spectroscopy. Current Opinion in Structural Biology 2000, 10, 585-592.
8. Gardner, K. H.; Kay, L. E. The use of 2H, 13C, 15N multidimensional NMR to study the structure and dynamics of proteins. Annu. Rev. Biophys. Biomol. Struct. 1998, 27, 357-406.
9. LeMaster, D. M. Chiral beta and random fractional deuteration for the determination of protein sidechain conformation by NMR. FEBS Lett. 1987, 223, 191-196.
10. LeMaster, D. M. Deuterium labelling in NMR structural analysis of larger proteins. Q. Rev. Biophys. 1990, 23, 133-174.
175 11. Ikura, M.; Kay, L. E.; Bax, A. A novel approach for sequential assignment of 1H, 13C, and 15N spectra of larger proteins: heteronuclear triple-resonance three-dimensional NMR spectroscopy. Application to calmodulin. Biochemistry 1990, 29, 4659-4667.
12. Yamazaki, T.; Lee, W.; Revingtom, M.; Mattiello, D. L.; Dahlquist, F. W.; Arrowsmith, C. H.; Kay, L. E. An HNCA pulse scheme for the backbone assignment of 15N,13C,2H-labeled proteins: application to a 37-kDa Trp repressor-DNA complex. J. Am. Chem. Soc. 1994, 116, 6464-6465.
13. Yamazaki, T.; Lee, W.; Arrowsmith, C. H.; McHandiram, D. R.; Kay, L. E. A suite of triple resonance NMR experiments for the backbone assignment of 15N, 13C, 2H labeled proteins with high sensitivity. J. Am. Chem. Soc. 1994, 116, 11655-11666.
14. Wittekind, M.; Mueller, L. HNCACB, a high-sensitivity 3D NMR experiment to correlate amide-proton and nitrogen resonances with the alpha- and beta-carbon resonances in proteins. J. Magn. Reson. 1993, B101, 201-205.
15. Grzesiek, S.; Bax, A. Improved 3D triple-resonance NMR techniques applied to a 31 kDa protein. J. Magn. Reson. 1992, 96, 432-440.
16. Clubb, R. T.; Thanabal, V.; Wagner, G. A constant-time three-dimensional triple- resonance pulse scheme to correlate intraresidue proton (1HN), nitrogen-15 and carbon-13 (13C') chemical shifts in nitrogen-15-carbon-13-labeled proteins. J. Magn. Reson. 1992, 97, 213-217.
17. Matsuo, H.; Li, H.; Wagner, G. A sensitive HN(CA)CO experiment for deuterated proteins. J. Magn. Reson. 1996, B110, 112-115.
18. Karplus, M. J. Contact electron-spin coupling of nuclear magnetic moments. J. Chem. Phys. 1959, 30, 11-15.
19. Karplus, M. J. Vicinal proton coupling in nuclear magnetic resonance. J. Am. Chem. Soc. 1963, 85, 2870-2871.
20. Mosberg, H. I.; Sobczyk-Kojiro, K.; Subramanian, P.; Crippen, G. M.; Ramalingam, K.; Woodard, R. W. Combined use of stereospecific deuteration, NMR, distance geometry, and energy minimization for the conformational analysis of the highly δ opioid receptor selective peptide [D-Pen2,D-Pen5]enkephalin. J. Am. Chem. Soc. 1990, 112, 822-829.
21. Curley, R. W., Jr.; Panigot, M. J.; Hansen, A. P.; Fesik, S. W. Stereospecific assignments of glycine in proteins by stereospecific deuteration and 15N labeling. J. Biomol. NMR 1994, 4, 335-340.
176 22. Aberhart, D. J.; Russell, D. J. Steric course of ketopantoate hydroxymethyltransferase in E. coli. J. Am. Chem. Soc. 1984, 106, 4902-4906.
23. Tatum, C. M.; Benkovic, P. A.; Benkovic, S. J.; Potts, R.; Schleicher, E.; Floss, H. G. Stereochemistry of methylene transfer involving 5,10-methylenetetrahydrofolate. Biochemistry 1977, 16, 1093-1102.
24. Slieker, L. J.; Benkovic, S. J. Synthesis of (6R,11S)- and (6R,11R)-5,10- methylene[11-1H,2H]tetrahydrofolate. Stereochemical paths of serine hydroxymethyltransferase, 5,10-methylenetetrahydrofolate dehydrogenase, and thymidylate synthetase catalysis. J. Am. Chem. Soc. 1984, 106, 1833-1839.
25. Jordan, P. M.; Akhtar, M. The mechanism of action of serine transhydroxymethylase. Biochem. J. 1970, 116, 277-286.
26. Malthouse, P. G.; Milne, J. J.; Gariani, L. S. A comparative study of the kinetics and stereochemistry of the serine hydroxymethyltransferase- and tryptophan synthase- catalyzed exchange of the pro-2R and pro-2S protons of glycine. Biochem. J. 1991, 274, 807-812.
27. Aberhart, D. J.; Russell, D. J. Steric course of N5,N10-methylenetetrahydrofolate formation from glycine by the glycine cleavage system in E. coli. J. Am. Chem. Soc. 1984, 106, 4907-4910.
28. Jones, C.; Jordan, P. M.; Akhtar, M. Mechanism and stereochemistry of the porphobilinogen deaminase and protoporphyrinogen IX oxidase reactions: stereospecific manipulation of hydrogen atoms at the four methylene bridges during the biosynthesis of heme. J. Chem. Soc., Perkin Trans. 1 1984, 2625-2633.
29. Armarego, W. L. F.; Milloy, B. A.; Pendergast, W. A highly stereospecific synthesis 2 of (R)-and (S)-[2- H1]glycine. J. Chem. Soc., Perkin Trans. 1 1976, 2229-2237.
30. Hegedus, L. S.; Lastra, E.; Narukawa, Y.; Snustad, D. C. Asymmetric synthesis of 2 (S)- and (R)-[2- H1]glycine via photolysis of optically active chromium carbene complexes. A comparison of stereoselectivity between chromium ketene complexes, ketenes, and ester enolates. J. Am. Chem. Soc. 1992, 114, 2991-2994.
31. Sinclair, P. J.; Zhai, D.; Reibenspies, J.; Williams, R. M. Electrophilic glycinates: new and versatile templates for asymmetric amino acid synthesis. J. Am. Chem. Soc. 1986, 108, 1103-1104.
32. Ohrui, H.; Misawa, T.; Meguro, H. Facile synthesis of chiral glycine from D-ribose. J. Org. Chem. 1985, 50, 3007-3009.
177 33. Eguchi, T.; Koudate, T.; Kakinuma, K. Diacetone glucose architecture as a chirality template. III. The Overman rearrangement on a diacetone-D-glucose template: kinetic and theoretical studies on the chirality transcription. Tetrahedron 1993, 49, 4527-4540.
34. Hamon, D. P. G.; Massy-Westropp, R. A.; Razzino, P. Asymmetric induction in acyclic radical reactions: enantioselective syntheses of (S)-2-deuterioglycine and (R)-2- deuterioglycine. Tetrahedron 1993, 49, 6419-6428.
35. Golding, B. T.; Gainsford, G. J.; Herlt, A. J.; Sargeson, A. M. Stereoselective hydrogen exchange on the methylene group of glycine in λ-R-(N- benzylglycinato)bis(ethylenediamine)cobalt(III) ion. Angew. Chem. 1975, 87, 523-524.
36. Belokon, Y. N.; Melikyan, A. S.; Salel'eva, T. F.; Bakhmutov, V. I.; Vitt, S. V.; Belikov, V. M. A new synthesis of (R)- and (S)-2-deuterio-amino acids, including (R)- and (S)-2-deuterio-glycine via stereochemically inert cobalt(III) complexes. Tetrahedron 1980, 36, 2327-2335.
37. Fuganti, C.; Mazza, M. Biosynthesis of amaryllidaceae alkaloids. Stereochemistry of hydrogen removal α to the tertiary nitrogen atom in the biological conversion of norpluviine into lycorenine. J. Chem. Soc., Perkin Trans. 1 1973, 954-956.
38. Battersby, A. R.; Staunton, J.; Summers, M. C. Studies of enzyme-mediated reactions. Part VI. Stereochemical course of the dehydrogenation of stereospecifically labelled benzylamines by the amine oxidase from pea seedlings (E.C. 1.4.3.6.). J. Chem. Soc., Perkins Trans. 1 1976, 10, 1052-1056.
39. Kajiwara, M.; Lee, S. F.; Scott, A. I.; Akhtar, M.; Jones, C. R.; Jordan, P. M. Stereospecific synthesis of chiral acetic acid from glycine. J. Chem. Soc., Chem. Commun. 1978, 967-968.
40. Yamada, H.; Kurumaya, K.; Eguchi, T.; Kajiwara, M. Stereospecific synthesis of 2 (R)-[2- H1]-glycine by means of microbial reduction and a determination of it's absolute configuration by CD spectra. J. Labelled Compd. Radiopharm. 1987, 24, 561-575.
41. Santaniello, E.; Casati, R.; Manzocchi, A. Convenient synthesis of stereospecifically deuterated glycines from glutamic acid using a combination of enzymic and chemical methods. J. Chem. Soc., Perkin Trans. 1 1985, 2389-2392.
42. Walker, J. R.; Curley, R. W., Jr. Improved synthesis of (R)-glycine-d-15N. Tetrahedron 2001, 57, 6695-6701.
43. Ramalingam, K.; Nanjappan, P.; Kalvin, D. M.; Woodard, R. W. A practical large scale chemical synthesis of chiral glycines. Tetrahedron 1988, 44, 5597-5604.
178 44. Carlsen, P. H. J.; Katsuki, T.; Martin, V. S.; Sharpless, K. B. A greatly improved procedure for ruthenium tetroxide catalyzed oxidations of organic compounds. J. Org. Chem. 1981, 46, 3936-3938.
45. Mijs, W. J.; De Jonge, C. R. H. I.; Editors In Organic Syntheses by Oxidation with Metal Compounds, 1986.
46. Tanaka, K.; Yoshifuji, S.; Nitta, Y. A new method for the synthesis of amides from amines: ruthenium tetroxide oxidation of N-protected alkylamines. Chem. Pharm. Bull. 1988, 36, 3125-3129.
47. Scientists at Isotec, Inc., Miamisburg, OH.
48. Chakraborti, A. K.; Ghatak, U. R. A highly effective ligand-bound ruthenium catalyst for chemoselective degradation of aromatic rings to carboxylic acids. J. Chem. Soc., Perkin Trans. 1 1985, 2605-2609.
49. Giddings, S.; Mills, A. Optimization of a simple system for the oxidation of octan-2- ol with sodium bromate, mediated by ruthenium tetraoxide generated in situ. J. Org. Chem. 1988, 53, 1103-1107.
50. Nunez, M. T.; Martin, V. S. Efficient oxidation of phenyl groups to carboxylic acids with ruthenium tetraoxide. A simple synthesis of (R)-γ-caprolactone, the pheromone of Trogoderma granarium. J. Org. Chem. 1990, 55, 1928-1932.
51. Windholz, T. B.; Johnston, D. B. R. Trichloroethoxycarbonyl. A generally applicable protecting group. Tetrahedron Lett. 1967, 2555-2557.
52. Midland, M. M.; Greer, S.; Tramontano, A.; Zderic, S. A. Chiral trialkylborane reducing agents. Preparation of 1-deuterio primary alcohols of high enantiomeric purity. J. Am. Chem. Soc. 1979, 101, 2352-2355.
53. Rogic, M. M. Conformational analysis and the transition state in asymmetric reductions with boranes based on (+)-α-pinene. 1. Benzaldehyde reduction with Alpine Borane and other B-alkyl-9-borabicyclo[3.3.1]nonanes. A semiempirical study. J. Org. Chem. 1996, 61, 1341-1346.
54. Dale, J. A.; Mosher, H. S. Nuclear magnetic resonance enantiomer regents. Configurational correlations via nuclear magnetic resonance chemical shifts of diastereomeric mandelate, O-methylmandelate, and α-methoxy-α- trifluoromethylphenylacetate (MTPA) esters. J. Am. Chem. Soc. 1973, 95, 512-519.
55. Yamada, I.; Noyori, R. Asymmetric transfer hydrogenation of benzaldehydes. Org. Lett. 2000, 2, 3425-3427.
179 56. Chollet, J. F.; Miginiac, L.; Rudelle, J.; Bonnemain, J. L. A convenient method of protection and mild deprotection of α-amino acid group for the synthesis of functional α- amino acids. Synth. Commun. 1993, 23, 2101-2111.
57. Williams, R. M.; Zhai, D.; Sinclair, P. J. Asymmetric synthesis of (R)- and (S)-[2- 2 H1]glycine. J. Org. Chem. 1986, 51, 5021-5022.
58. Olson, J. A. Vitamin A. In Handbook of Vitamins, 3rd Edition; Rucker, R. B., Suttie, J. W., McCormick, D. B., Machlin, L. J., Eds.; Marcel Dekker, Inc.: New York, 2001; p 1-50.
59. Anonymous. Nomenclature policy: Generic descriptors and trivial names for vitamins and related compounds. J. Nutr. 1990, 120, 12-19.
60. Isler, O.; Huber, W.; Ronco, A.; Kofler, M. Synthesis of vitamin A. Helv. Chim. Acta 1947, 30, 1911-1927.
61. Pommer, H. Synthesis of vitamin A. Angew. Chem. 1960, 72, 811-819.
62. Olson, J. A. Vitamin A, retinoids and carotenoids. In Modern Nutrition in Health and Disease, 8th Edition; Goodhart, R. S., Shils, M. E., Eds.; Lea & Febiger: Philadelphia, 1994; p 287-307.
63. Ong, D. E. Absorption of Vitamin A. In Vitamin A in Health and Disease; Blomhoff, R., Ed.; Marcel Dekker, Inc.: New York, 1994; p 37-72.
64. Leneberger, M. G.; Engeloch-Jarret, C.; Woggon, W.-D. The reaction mechanism of the enzyme-catalyzed central cleavage of β-carotene in retinal. Angew. Chem., Int. Ed. Engl. 2001, 40, 2614-2617.
65. Wyss, A.; Wirtz, G.; Woggon, W. D.; Brugger, R.; Wyss, M.; Friedlein, A.; Bachmann, H.; Hunziker, W. Cloning and expression of β,β-carotene 15,15'- dioxygenase. Biochem. Biophys. Res. Commun. 2000, 271, 334-336.
66. Blomhoff, R. Introduction. In Vitamin A in Health and Disease; Blomhoff, R., Ed.; Marcel Dekker, Inc.: New York, 1994; p 1-35.
67. Newcomer, M. E. Retinoid-binding proteins: structural determinants important for function. FASEB J. 1995, 9, 229-239.
68. Wake, K. Absorption of Vitamin A. In Vitamin A in Health and Disease; Blomhoff, R., Ed.; Marcel Dekker, Inc.: New York, 1994; p 73-86.
69. Blomhoff, R.; Green, M. H.; Norum, K. R. Vitamin A: Physiological and biochemical processing. Annu. Rev. Nutr. 1992, 12, 37-57. 180
70. Blaner, W. S.; Olson, J. A. Plasma retinol-binding protein. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 229-255.
71. Soprano, D. R.; Blaner, W. S. Plasma retinol-binding protein. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 257-281.
72. Genchi, G.; Wang, W.; Barua, A.; Bidlack, W. R.; Olson, J. A. Formation of beta- glucuronides and of beta-galacturonides of various retinoids catalyzed by induced and noninduced microsomal UDP-glucuronosyltransferases of rat liver. Biochim. Biophys. Acta 1996, 1289, 284-290.
73. Olson, J. A. Recommended dietary intakes (RDI) of vitamin A in humans. Am. J. Clin. Nutr. 1987, 45, 704-716.
74. Saari, J. C. Retinoids in Photosensitive Systems. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 351-385.
75. Rando, R. R. Isomerization reactions of retinoids in the visual system. Pure Appl. Chem. 1994, 66, 989-994.
76. Lewis, K. C.; Green, M. H.; Green, J. B.; Zeck, L. A. Retinol metabolism in rats with low vitamin A status - A compartmental model. J. Lipid Res. 1990, 31, 1535-1548.
77. Dowling, J. E.; Wald, G. The role of vitamin A acid. Vitam. Horm. 1960, 18, 515- 541.
78. Gudas, L. J. Retinoids and vertebrate development. J. Biol. Chem. 1994, 269, 15399- 15402.
79. Hofmann, C.; Eichele, G. Retinoids in development. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 387-441.
80. Ross, A. C.; Hammerling, U. G. Retinoids and the immune system. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 257-281.
81. Cantorna, M. T.; Nashold, F. E.; Hayes, C. E. In vitamin A deficiency multiple mechanisms establish a regulatory T helper cell imbalance with excess TH1 and insufficient TH2 function. J. Immunol. 1994, 152, 1515-1522.
181 82. Giguere, V.; Ong, E. S.; Segui, P.; Evans, R. M. Identification of a receptor for the morphogen retinoic acid. Nature 1987, 330, 624-629.
83. Petkovich, M.; Brand, N. J.; Krust, A.; Chambon, P. A human retinoic acid receptor which belongs to the family of nuclear receptors. Nature 1987, 330, 444-450.
84. Manglesdorf, D. J.; Ong, E. S.; Dyck, J. A.; Evans, R. M. Nuclear receptor that identifies a novel retinoic acid response pathway. Nature 1990, 345, 224-229.
85. Manglesdorf, D. J.; Umesono, K.; Evans, R. M. Plasma retinol-binding protein. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 319-349.
86. Fisher, G. J.; Talwar, H. S.; Xiao, J. H.; Datta, S. C.; Reddy, A. P.; Gaubs, M. P.; Rochette-Egly, C.; Chambon, P.; Voorhees, J. J. Immunological identification and functional quantitation of retinoic acid and retinoid X receptor proteins in human skin. J. Biol. Chem. 1994, 269, 20629-20635.
87. Underwood, B. A. Vitamin A in human nutrition: Public health considerations. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 211-227.
88. Ye, X. D.; Al-Babili, S.; Kloti, A.; Zhang, J.; Lucca, P.; Beyer, P.; Potrykus, I. Engineering the provitamin A (beta-carotene) biosynthetic pathway into (carotenoid-free) rice endosperm. Science 2000, 287, 303-305.
89. Gunning, D. B.; Barua, A. B.; Lloyd, R. A.; Olson, J. A. Retinoyl β-glucuronide: A non-toxic retinoid for the topical treatment of acne. J. Dermatol. Treat. 1994, 5, 181-185.
90. Moon, R. C.; Metha, R. G. Cancer chemoprevention by retinoids: animal models. Methods Enzymol. 1990, 190, 395-406.
91. Altucci, L.; Gronemeyer, H. The promise of retinoids to fight against cancer. Nature Rev., Cancer 2001, 1, 181-193.
92. Moon, R. C.; Metha, R. G.; Rao, K. V. N. Retinoids and cancer in experimental animals. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 573-595.
93. Hong, W. K.; Itri, L. M. Retinoids and human cancer. In The Retinoids: Biology, Chemistry, and Medicine, 2nd Edition; Sporn, M. B., Roberts, A. B., Goodman, D. S., Eds.; Raven Press: New York, 1994; p 597-630.
94. Hathcock, J. N.; Hattan, D. G.; Jenkins, M. Y.; McDonald, J. T.; Sundaresan, P. R.; Wilkening, V. L. Evaluation of vitamin A toxicity. Am. J. Clin. Nutr. 1990, 52, 183-202. 182
95. Nau, H.; Chahoud, I.; Dencher, L.; Lammer, E. J.; Scott, W. J. Teratogenicity of vitamin A and retinoids. In Vitamin A in Health and Disease; Blomhoff, R., Ed.; Marcel Dekker, Inc.: New York, 1994; p 615-663.
96. Newton, D. L.; Henderson, W. R.; Sporn, M. B. Structure activity-relationships of retinoids in hamster tracheal organ-culture. Cancer Res. 1980, 40, 3413-3425.
97. Willhite, C. C. Molecular correlates in retinoid pharmacology and toxicology. In Chemistry and Biology of Synthetic Retinoids; Dawson, M. L., Okamura, W. H., Eds.; CRC Press: Boca Raton, FL., 1990; p 539-573.
98. Loeliger, P.; Bollag, W.; Mayer, H. Arotinoids, a new class of highly active compounds. Eur. J. Med. Chem. 1980, 15, 9-15.
99. Dawson, M. L.; Chan, R. S.; Derdzinski, K.; Hobbs, P. D.; Chao, W.; Schiff, L. J. Synthesis and pharmacological activity of 6-[(E)-2-(2,6,6-trimethyl-1-cyclohexen-1- yl)ethen-1-yl]- and 6-(1,2,3,4-tetrahydro-1,1,4,4-tetramethyl-6-naphthyl)-2- naphthalenecarboxylic acids. J. Med. Chem. 1983, 26, 1653-1656.
100. Dawson, M. L.; Hobbs, P. D.; Derdzinski, K.; Chao, W.; Frenking, G.; Loew, G. H.; Jetten, A. M.; Napoli, J. L.; Williams, J. B.; Sani, B. P. Effect of structural modifications in the C7-C11 region of the retinoid skeleton on biological activity in a series of aromatic retinoids. J. Med. Chem. 1989, 32, 1504-1517.
101. Dawson, M. L.; Zhang, X. Discovery and design of retinoic acid receptor and retinoic X receptor class- and subtype-selective synthetic analogs of all-trans-retinoic acid and 9-cis-retinoic acid. Curr. Med. Chem. 2002, 9, 623-637.
102. Dawson, M. L.; Zhang, X.; Hobbs, P. D.; Jong, L. Synthetic retinoids and their usefulness in biology and medicine. In Vitamin A and retinoids: An update of biological aspects and clinical applications; Livrea, M. A., Ed.; Birkhauser Verlag: Basel, 2000; p 161-196.
103. Kagechika, H.; Kawachi, E.; Hashimoto, Y.; Himi, T.; Shudo, K. Retinobenzoic acids 1. Structure-activity relationships of aromatic amides with retinoidal activity. J. Med. Chem. 1988, 31, 2182-2192.
104. Chao, W.; Hobbs, P. D.; Jong, L.; Zhang, X.; Zheng, Y.; Wu, Q.; Shroot, B.; Dawson, M. L. Effects of receptor class- and subtype-selective retinoids and an apoptosis-inducing retinoid on the adherent growth of the NIH:OVCAR-3 ovarian cancer cell line in culture. Cancer Lett. 1997, 113, 1-7.
105. Bernard, B. A.; Bernardon, J. M.; Delescluse, C.; Martin, B.; Lenoir, M. C.; Maignan, J.; Charpentier, B.; Pilgrim, W. R.; Reichert, U.; Shroot, B. Identification of 183 synthetic retinoids with selectivity for human nuclear retinoic acid receptor γ. Biochem. Biophys. Res. Commun. 1992, 186, 977-983.
106. Strickland, S.; Breitman, T. R.; Frickel, F.; Nurrenback, A.; Hadicke, E.; Sporn, M. B. Structure-activity-relationships of a new series of retinoidal benzoic-acid derivatives as measured by induction of differentiation of murine F9 teratocarcinoma cells and human HL-60 promyelocytic leukemia cells. Cancer Res. 1983, 43, 5268-5272.
107. Lehmann, J. M.; Jong, L.; Fanjul, A.; Cameron, J. F.; Lu, X. P.; Haefner, P.; Dawson, M. L. Retinoids selective for retinoid X receptor response pathways. Science 1992, 258, 1944-1946.
108. Boehm, M. F.; Zhang, L.; Badea, B. A.; White, S. K.; Mais, D. E.; Berger, E.; Suto, C. M.; Goldman, M. E.; Heyman, R. A. Synthesis and structure-activity relationships of novel retinoid X receptor-selective retinoids. J. Med. Chem. 1994, 37, 2930-2941.
109. Johnson, A. T.; Klein, E. S.; Gillett, S. J.; Wang, L.; Song, T. K.; Pino, M. E.; Chandraratna, R. A. S. Synthesis and characterization of a highly potent and effective antagonist of retinoic acid receptors. J. Med. Chem. 1995, 38, 4764-4767.
110. Apfel, C.; Bauer, F.; Crettaz, M.; Forni, L.; Kamber, M.; Kaufmann, F.; LeMotte, P.; Pirson, W.; Klaus, M. A retinoic acid receptor α antagonist selectively counteracts retinoic acid effects. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 7129-7133.
111. Teng, M.; Duong, T. T.; Johnson, A. T.; Klein, E. S.; Wang, L.; Khalifa, B.; Chandraratna, R. A. S. Identification of highly potent retinoic acid receptor α-selective antagonists. J. Med. Chem. 1997, 40, 2445-2451.
112. Botling, J.; Castro, D. S.; Oberg, F.; Nilsson, K.; Perlmann, T. Retinoic acid receptor/retinoid X receptor heterodimers can be activated through both subunits providing a basis for synergistic transactivation and cellular differentiation. J. Biol. Chem. 1997, 272, 9443-9449.
113. Delia, D.; Aiello, A.; Lombardi, L.; Pellicci, P. G.; Grignani, F.; Formelli, F.; Menard, S.; Costa, A.; Veronesi, U.; Pierotti, M. A. N-(4-hydroxyphenyl)retinamide induces apoptosis of malignant hematopoietic-cell lines including those unresponsive to retinoic acid. Cancer Res. 1993, 53, 6036-6041.
114. Gander, R. J.; Gurney, J. A. U.S. Patent 4,108,880, August 22, 1978.
115. Moon, R. C.; Thompson, H. J.; Becci, P. J.; Grubbs, R. J.; Newton, D. L.; Smith, J. M.; Phillips, S. L.; Henderson, W. R.; Mullen, L. T.; Brown, C. C.; Sporn, M. B. N-(4- hydroxyphenyl)retinamide, a new retinoid for the prevention of breast cancer in the rat. Cancer Res. 1979, 39, 1339-1346.
184 116. Formelli, F. Bioactivities of N-(4-hydroxyphenyl) retinamide. In Vitamin A and retinoids: An update of biological aspects and clinical applications; Livrea, M. A., Ed.; Birkhauser Verlag: Basel, 2000; p 241-250.
117. Sheikh, M. S.; Shao, Z. M.; Li, Z. S.; Ordenez, J. V.; Conley, B. A.; Wu, S.; Dawson, M. L.; Han, Q. X.; Chao, W.; Quick, T. N-(4-hydroxyphenyl)retinamide (4- HPR)-mediated biological actions involve retinoid receptor-independent pathways in human breast cancinoma. Carcinogenesis 1995, 16, 2477-2486.
118. Fanjul, A.; Delia, D.; Pierotti, M. A.; Rideout, D.; Qiu, J.; Pfahl, M. 4- (Hydroxyphenyl)retinamide is a highly selective activator of retinoid receptors. J. Biol. Chem. 1996, 271, 22441-22446.
119. Swisshelm, K.; Ryan, K.; Lee, X.; Tsou, H. C.; Peacocke, M.; Sager, R. Down- regulation of retinoic acid receptor β in mammary carcinoma cell lines and its up- regulation in senescing normal mammary epithelial cells. Cell Growth Diff. 1994, 5, 133- 141.
120. Abou-Issa, H. M.; Alshafie, G. A.; Curley, R. W., Jr.; Wong, M. F.; Clagett-Dame, M.; Repa, J. J.; Sikri, V. Chemopreventive activity of a C-glucuronide analog of N-(4- hydroxyphenyl) retinamide-O-glucuronide against mammary tumor development and growth. Anticancer Res. 1999, 19, 999-1004.
121. Swanson, B. N.; Newton, D. L.; Roller, P. P.; Sporn, M. B. Biotransformation and biological activity of N-(4-hydroxyphenyl) retinamide derivatives in rodents. J. Pharmacol. Exp. Ther. 1981, 219, 632-637.
122. Dawson, M. J.; Hobbs, P. D. Synthesis of the D-glucuronic acid conjugates of N-(4- hydroxyphenyl) and N-(2-hydroxyethyl) retinamides. Carbohydr. Res. 1980, 85, 121-129.
123. Veronesi, U.; De Palo, G.; Marubini, E.; Costa, A.; Formelli, F.; Mariani, L.; Decensi, A.; Camerini, T.; Rosselli Del Turco, M.; Di Mauro, M. G.; Muraca, M. G.; Del Vecchio, M.; Pinto, C.; D'Aiuto, G.; Boni, C.; Campa, T.; Magni, A.; Miceli, R.; Perloff, M.; Malone, W. F.; Sporn, M. B. Randomized trial of fenretinide to prevent second breast malignancy in women with early breast cancer. J. Natl. Cancer Inst. 1999, 91, 1847- 1856.
124. Abou-Issa, H.; Curley, R. W., Jr.; Alshafie, G. A.; Weiss, K. L.; Clagett-Dame, M.; Chapman, J. S.; Mershon, S. M. Chemotherapeutic evaluation of 4- hydroxybenzylretinone (4-HBR), a nonhydrolyzable C-linked analog of N-(4- hydroxyphenyl) retinamide (4-HPR) against mammary carcinogenesis. Anticancer Res. 2001, 21, 3839-3844.
125. Abou - Issa, H.; Duruibe, V. A.; Minton, J. P.; Larroya, S.; Dwivedi, C.; Webb, T. E. Putative metabolites derived from dietary combinations of calcium glucarate and N-(4- 185 hydroxyphenyl)retinamide act synergistically to inhibit the induction of rat mammary tumors by 7,12-dimethybenz[a]anthracene. Proc. Natl. Acad. Sci. U.S.A. 1988, 85, 4181- 4184.
126. Bhatnagar, R.; Abou-Issa, H.; Curley, R. W., Jr.; Koolemans-Beynen, A.; Moeschberger, M. L.; Webb, T. E. Growth suppression of human breast carcinoma cells in culture by N-(4-hydroxyphenyl)retinamide and its glucuronide and through synergism with glucarate. Biochem. Pharmacol. 1991, 41, 1471-1477.
127. Abou - Issa, H.; Curley, R. W., Jr.; Panigot, M. J.; Wilcox, K. A.; Webb, T. E. In vivo use of N-(4-hydroxyphenyl retinamide)-O-glucuronide as a breast cancer chemopreventive agent. Anticancer Res. 1993, 13, 1431-1436.
128. Abou-Issa, H.; Curley, R. W., Jr.; Panigot, M. J.; Tanagho, S. N.; Sidhu, B. S.; Alshafie, G. A. Chemotherapeutic evaluation of N-(4-hydroxyphenyl) retinamide-O- glucuronide in the rat mammary tumor model. Anticancer Res. 1997, 17, 3335-3339.
129. Panigot, M. J.; Humphries, K. A.; Curley, R. W., Jr. Preparation of 4- retinamidophenyl- and 4-retinamidobenzyl-C-glycosyl and C-glucuronosyl analogs of the glucuronide of 4-hydroxyphenylretinamide as potential stable cancer chemopreventive agents. J. Carbohydr. Chem. 1994, 13, 303-321.
130. Hurd, C. D.; Bonner, W. A. Glycosylation of hydrocarbons by means of the Grignard reagent. J. Am. Chem. Soc. 1945, 67, 1972-1977.
131. Helferich, B.; Bettin, K. L. Note on the synthesis of nonreducing disaccharides. Chem. Ber. 1961, 94, 1159-1160.
132. Goekjian, P. G.; Wu, T. C.; Kishi, Y. Preferred conformation of C-glycosides. 6. Conformational similarity of glycosides and corresponding C-glycosides. J. Org. Chem. 1991, 56, 6412-6422.
133. Wang, P. G.; Bertozzi, C. R. In Glycochemistry: Principles, synthesis, and applications; Marcel Dekker, Inc.: New York, 2001.
134. Levy, D. E.; Tang, C. In The Chemistry of C-glycosides; Pergamon: Oxford, 1995.
135. Nicotra, F. Synthesis of C-glycosides of biological interest. Top. Curr. Chem. 1997, 187, 55-83.
136. Heyns, K.; Paulsen, H. Selective catalytic oxidation of carbohydrates, employing platinum catalysts. Advan. Carbohydrate Chem. 1962, 17, 169-211.
137. Curley, R. W., Jr.; Abou-Issa, H.; Panigot, M. J.; Repa, J. J.; Clagett-Dame, M.; Alshafie, G. Chemopreventive activities of C-glucuronide/glycoside analogs of retinoid- 186 O-glucuronides against breast cancer development and growth. Anticancer Res. 1996, 16, 757-763.
138. Wong, M. F.; Weiss, K. L.; Curley, R. W., Jr. Recent improvements towards the synthesis of the C-glucuronosyl cancer chemopreventive (β-D-glucopyranosyluronate)-4- retinamidophenylmethane. J. Carbohydr. Chem. 1996, 15, 763-768.
139. Jung, M. E.; Lyster, M. A. Quantitative dealkylation of alkyl ethers via treatment with trimethylsilyl iodide. A new method for ether hydrolysis. J. Org. Chem. 1977, 42, 3761-3764.
140. Panigot, M. J.; Curley, R. W., Jr. Reaction of glycosyl halides with benzyl Grignard reagents: Unexpected O-tolyl alkylation of tetra-O-acetylglucopyranosyl bromide and direct synthesis of (β-glycosyl)phenylmethanes. J. Carbohydr. Chem. 1994, 13, 293-302.
141. Chatani, N.; Ikeda, T.; Sano, T.; Sonoda, N.; Kurosawa, H.; Kawasaki, Y.; Murai, S. Catalytic siloxymethylation of glycosides by the HSiR3/CO/Co2(CO)8 reaction. A new entry to C-glycosyl compounds. J. Org. Chem. 1988, 53, 3387-3389.
142. Bollenback, G. N.; Long, J. W.; Benjamin, D. G.; Lindquist, J. A. The synthesis of aryl-D-glucopyranosiduronic acids. J. Am. Chem. Soc. 1955, 77, 3310-3315.
143. Spak, S. J.; Martin, O. R. The synthesis of an O,C-trisaccharide: The O,C-analog of methyl 4'-O-β-D-glucopyanosylgentiobioside. Tetrahedron 2000, 56, 217-224.
144. Taylor, R. J. K. Organocopper chemistry: An overview. In Organocopper reagents: A practical approach; Taylor, R. J. K., Ed.; Oxford University Press: Oxford, 1994; p 1-26.
145. Johnson, C. R.; Johns, B. A. Suzuki cross-coupling of carbohydrates. Synthesis of arylmethyl β-C-glycosides and aryl-scaffolded trisaccharide mimics. Synlett 1997, 1406- 1408.
146. Johns, B. A.; Pan, Y. T.; Elbein, A. D.; Johnson, C. R. Synthesis and biological evaluation of aza-C-disaccharides: (1-6), (1-4), and (1-1) linked sugar mimics. J. Am. Chem. Soc. 1997, 119, 4856-4865.
147. Walker, J. R.; Alshafie, G.; Abou-Issa, H.; Curley, R. W., Jr. An improved synthesis of the C-linked glucuronide of N-(4-Hydroxyphenyl)retinamide. Bioorg. Med. Chem. Lett. 2002, 12, 2447-2450.
148. RajanBabu, T. V.; Reddy, G. S. 1-Methylene sugars as C-glycoside precursors. J. Org. Chem. 1986, 51, 5458-5461.
187 149. Petasis, N. A.; Bzowej, E. I. Titanium-mediated carbonyl olefinations. 1. Methylenations of carbonyl compounds with dimethyltitanocene. J. Am. Chem. Soc. 1990, 112, 6392-6394.
150. Csuk, R.; Glaenzer, B. I. Methylenation of aldonolactones. Tetrahedron 1991, 47, 1655-1664.
151. Miyaura, N.; Ishiyama, T.; Sasaki, H.; Ishikawa, M.; Satoh, M.; Suzuki, A. Palladium-catalyzed inter- and intramolecular cross-coupling reactions with B-alkyl-9- borabicyclo[3.3.1]nonane derivatives with 1-halo-1-alkenes or haloarenes. Synthesis of functionalized alkenes, arenes, and cycloalkenes via a hydroboration-coupling sequence. J. Am. Chem. Soc. 1989, 111, 314-321.
152. Miyaura, N.; Suzuki, A. Palladium-catalyzed cross-coupling reactions of organoboron compounds. Chem. Rev. 1995, 95, 2457-2483.
153. Suzuki, A. Recent advances in the cross-coupling reactions of organoboron derivatives with organic electrophiles, 1995-1998. J. Organomet. Chem. 1999, 576, 147- 168.
154. Davis, N. J.; Flitsch, S. L. Selective oxidation of monosaccharide derivatives to uronic acids. Tetrahedron Lett. 1993, 34, 1181-1184.
155. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H. Highly selective TEMPO mediated oxidation of primary alcohol groups in polysaccharides. Recl. Trav. Chim. Pays-Bas 1994, 113, 165-166.
156. de Nooy, A. E. J.; Besemer, A. C. Selective oxidation of primary alcohols mediated by nitroxyl radical in aqueous solution. Kinetics and mechanism. Tetrahedron 1995, 51, 8023-8032.
157. Weiss, K. L.; Alshafie, G.; Chapman, J. S.; Mershon, S. M.; Abou-Issa, H.; Clagett- Dame, M.; Curley, R. W., Jr. An unhydrolyzable analogue of N-(4- hydroxyphenyl)retinamide synthesis and preliminary biological studies. Bioorg. Med. Chem. Lett. 2001, 11, 1583-1586.
158. Chapman, J. S.; Weiss, K. L.; Curley, R. W., Jr.; Highland, M. A.; Clagett-Dame, M. Hydrolysis of 4-HPR to atRA occurs in vivo but is not required for retinamide- induced apoptosis. 2003, manuscript submitted.
159. Zile, M. H.; Bunge, E. C.; DeLuca, H. F. On the physiological basis of vitamin A- stimulated growth. J. Nutr. 1979, 109, 1787-1796.
160. Zile, M. H.; Inhorn, R. C.; DeLuca, H. F. The biological activity of 5,6- epoxyretinoic acid. J. Nutr. 1980, 110, 2225-2230. 188
161. Abu-Abed, S. S.; Beckett, B. R.; Chiba, H.; Chithalen, J. V.; Jones, G.; Metzger, D.; Chambon, P.; Petkovich, M. Mouse P450RAI (CYP26) expression and retinoic acid- inducible retinoic acid metabolism in F9 cells are regulated by retinoic acid receptor γ and retinoic X acid α. J. Biol. Chem. 1998, 273, 2409-2415.
162. Robarge, M. J. Stable analogues of retinoid-O-glucuronides: Synthesis and biological activity. Dissertation, The Ohio State University, 1996.
163. Kelley, J. L.; Baker, B. R. Irreversible enzyme inhibitors 202. Candidate active- site-directed irreversible inhibitors of 5-fluoro-2'-deoxyuridine phosphorylase from Walker 256 rat tumor derived from 1-benzyl-5-(3-ethoxybenzyl)uracil. J. Med. Chem. 1982, 25, 600-603.
164. Katz, H. E. Chelate and macrocycle effects in the 2,2'-bipyridine N,N'-dioxide complexation of alkyltin trichlorides. J. Org. Chem. 1985, 50, 2086-2091.
165. Hase, T. A. In Umpoled synthons: A survey of sources and uses in synthesis; John Wiley and Sons, Inc.: New York, 1987.
166. Albright, J. D. Reactions of acyl anion equivalents derived from cyanohydrins, protected cyanohydrins and α-dialkylaminonitriles. Tetrahedron 1983, 39, 3207-3233.
167. Stork, G.; Maldonado, L. Anions of protected cyanohydrins as acyl carbanion equivalents and their use in a new synthesis of ketones. J. Am. Chem. Soc. 1971, 93, 5286-5287.
168. Kobayashi, S.; Tsuchiya, Y.; Mukaiyama, T. A facile synthesis of cyanohydrin trimethylsilyl ethers by the addition reaction of trimethylsilyl cyanide with aldehydes under basic conditions. Chem. Lett. 1991, 4, 537-540.
169. Weiss, K. L. Structural probes of retinoid action. Dissertation, The Ohio State University, 2001.
170. Brown, H. C.; Snyder, C.; Subba Rao, B. C.; Zweifel, G. Organoboranes for synthesis 2. Oxidation of organoboranes with alkaline hydrogen peroxide as a convenient route for the cis-hydration of alkenes via hydroboration. Tetrahedron 1986, 42, 5505-5510.
171. Kirby, G. W.; Michael, J. Stereoselective-labelling of aromatic amino-acids with deuterium and tritium. J. Chem. Soc., Perkin Trans. 1 1973, 115-120.
172. Chong, J. M.; Park, S. B. Preparation of alkyl tert-butyl iminodicarbonates. J. Org. Chem. 1993, 58, 7300-7303.
189 173. Carson, J. F. N-(2,2,2-Trichloroethoxycarbonyl)-L-amino acids. Synthesis 1981, 268-270.
174. Hanessian, S.; Pernet, A. G. Carbanions in carbohydrate chemistry. Synthesis of C- glycosyl malonates. Can. J. Chem. 1974, 52, 1266-1279.
190