University of Southern California Dissertations and Theses

Synthesis of fluorescent conjugates of risedronate and related analogues for bone imaging

(USC Thesis Other)

Synthesis of fluorescent conjugates of risedronate and related analogues for bone imaging

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content I. SYNTHESIS OF FLUORESCENT CONJUGATES OF RISEDRONATE AND
RELATED ANALOGUES FOR BONE IMAGING.
II. SYNTHESIS OF A NOVEL BISPHOSPHONIC ACID ALKENE MONOMER.

by

Joy Lynn Fabile Bala

A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(CHEMISTRY)

August 2009

Copyright 2009 Joy Lynn Fabile Bala
ii
Dedication

This work is dedicated to my loving father, Prof. Juan L. Bala, Jr., who waited so
patiently for this day.
iii
Acknowledgements

To my parents, I cannot express how thankful I am to both of you,
unconditionally loving and believing in me in whatever strange adventure or crazy antic I
decide to do next. Without your support, I would not have made it this far, and I am
forever grateful for all the sacrifices you have both made to give me a better life. Both of
you are my inspiration, and I thank you. To my brother, thank you for always looking out
for your little sister and being the best brother anyone could ask for. To Christopher,
thank you for being my rock these last 5 years.
I would like to thank Prof. Charles McKenna for giving me an opportunity to
grow into an independent scientist, for opening the intertwined world of science and law
to me, and for supporting me during my graduate studies and beyond. I would also like to
thank Dr. Boris Kashemirov for believing in me, for his patience, and for passing on his
wealth of knowledge to make me a better chemist. I also thank Dr. F. Hal Ebetino of
Procter & Gamble Pharmaceuticals for all his guidance and assistance. Additionally, I
would like to thank all of my countless collaborators for all of their help and
contributions to the research projects that I feel lucky to have been involved with.
I owe a great deal of gratitude to the USC Department of Chemistry. Thank you to
my committee for all your support: Prof. Surya Prakash, Prof. Peter Qin, Prof. Robert
Bau, Prof. Clay Wang, and Prof. Ian Haworth. To Prof. Curt Wittig and Prof. Hanna
Reisler, thank you for all you have done; your kindness and guidance has sweetened my
USC experience. A huge thanks goes to the USC staff, working so hard just to make my
life that much easier: Michele Dea, Heather Connor, Danielle Hayes, Katie McKissick,
iv
Marie de la Torre, Jaime Avila, Bruno Herreros, and the VWR staff (Marie, Darryl, and
Leo).
From Creighton University, I would like to thank Prof. Mark Kearley for making
chemistry fun for me and for continually supporting me. Thank you to Prof. Robert
Dobberpuhl; without your advice I may never have reached graduate school, and I
definitely would not have be the same person I am today.
To past and present members of the McKenna group, thank you for the advice,
support, and, most of all, the good times! To Ulrika Tehler, Mong Marma, and Greg
Sanchez, thank you for all your patience and help when I was new to the lab. To James
Hogan and Thomas Upton, thank you for all the uncountable lending hands you were
always willing to give. To Larryn Peterson, thank you is not enough for all your help and
support every step of the way; the ride would not have been the same without you. Thank
you to Ivan Krylov and Brian Chamberlain for keeping me laughing.
And finally, a great big thank you to the rest of my family and friends, near and
far. Thank you for helping me achieve one milestone after another.
v
Table of Contents

Dedication ii

Acknowledgements iii

List of Figures vii

List of Schemes xiii

Abstract xiv

Chapter 1: Bone-Targeting Bisphosphonic Acids 1
Anti-resorptive Drugs 1
Bone-Seeking Agents 5
Conclusion 10
Chapter 1 References 12

Chapter 2: Synthesis of Fluorescent and Near-IR Labeled Risedronate and
Related Analogues 14
Introduction 14
Results and Discussion 19
Conclusion 36
Experimental Methods 37
Chapter 2 References 51

Chapter 3: Evaluation of Fluorescent Risedronate and Related
Analogues as Imaging Probes 56
Introduction 56
Results and Discussion 59
Conclusion 68
Experimental Methods 69
Chapter 3 References 70

Chapter 4: Synthesis of a Novel Bisphosphonic Acid Alkene Monomer 72
Introduction 72
Results and Discussion 75
Table 4.1 86
Conclusion 91
Experimental Methods 91
Chapter 4 References 100

vi
Chapter 5: Chiral Phosphonocarboxylate Analogues of
Risedronate and Minodronate: 3-PEHPC and 3-IPEHPC 102
Introduction 102
Results and Discussion 104
Conclusion 112
Experimental Methods 113
Chapter 5 References 116

Bibliography 118

Appendices
Appendix A 126
Appendix B 172
Appendix C 190

vii
List of Figures

Figure 1.1. Structure of pyrophosphate (left) versus general structure of a
bisphosphonate (right). 1

Figure 1.2. Representative structures of 1st generation (clodronate and
etidronate), 2nd generation (pamidronate, alendronate, and
ibandronate), and 3rd generation (risedronate, zoledronate, and
minodronate) bisphosphonates. 3

Figure 1.3. Examples of bisphosphonate prodrugs. 7

Figure 1.4. Structures of exemplary polymerizable BPs. 9

Figure 2.1. Structures of risedronate (1) and its related analogues, 3-PEHPC
(2) and desoxyRIS (3). 15

Figure 2.2. Structures of previously reported fluorescent alkylamino BPs. 17

Figure 2.3. UV absorption spectra of 13 µM solution of 5(6)-FAM after
exposure to 3 different light levels in 30 minutes. 20

Figure 2.4. Analytical HPLC chromatograms FL-ALN using 2 different
gradients (scalable for purification conditions). 21

Figure 2.5. General structures of dihalide or N-protected aminoalkyl halide
alkylating agents studied for alkylation of RIS. 23

Figure 2.6. Two methods of following the progression of the reaction between
epoxide 4 and RIS. 27

Figure 2.7. Structures of reported “Alexa” dye versus proposed AF647. 34

Figure 3.1. Structures of labeled RIS compounds discussed in this chapter. 59

Figure 3.2. Comparison of retention times of RIS vs. 5- and 6-FAMRIS on
hydroxyapatite column. 60

Figure 3.3. Binding and “recycling” of FAMRIS to surface of dentine. 61

Figure 3.4. Cellular uptake of FAMRIS and FAMRISPC proceeds by fluid-
phase endocytosis. 62

viii
Figure 3.5. Western blot assays for unprenylated Rap1A (uRap1A) or
unprenylated Rab11 (uRab11). 64

Figure 3.6. FAMRIS inhibits prenylation and bone resorption in vivo. 65

Figure 3.7. Presence of FAMRIS in osteocytic lacunae. 66

Figure 4.1. Structures of polymerizable BPs. 74

Figure 4.2.
31
P NMR of reaction mixture of the alkylation of TIPMFBP (4a). 78
Figure 4.3.
31
P NMR spectrum of reaction mixture of alkylation of TIPMBP
with 8-bromooctyl phthalimide. 80

Figure 4.4.
31
P NMR of Bis(propan-2-yl) {1-[bis(propan-2-yloxy)phosphoryl]-
7-bromoheptyl}phosphonate (4c). 83

Figure 4.5.
31
P NMR of partial ester of Bis(propan-2-yl) {1-[bis(propan-2
yloxy)phosphoryl]-7-(1,3-dioxo-2,3-dihydro-1H-isoindol-2-yl)-1
fluoroheptyl} phosphonate (6). 84

Figure 4.6. Reaction progression of synthesis of [1-fluoro-1-phosphono-7-
(prop-2-enamido)heptyl]phosphonic acid (1) by
1
H NMR. 88

Figure 5.1. Structures of the nitrogen-containing BPs, risedronate (1) and
minodronate (3), and their corresponding phosphonocarboxylate
analogues, 3-PEHPC (2) and 3-IPEHPC (4), respectively. 103

Figure 5.2. A general schematic of the mevalonate pathway, depicting known
inhibitors of specific enzymes (shown in boxes). 103

Figure 5.3. Preparative HPLC chromatogram of purification of racemic 4
(R
t
= 12.5 min). 106

Figure 5.4. Chiral stationary phase of Prontosil AX QN-1 (8S, 9R) and QD-1
(8R,9S) column. 107

Figure 5.5. Chromatogram of preparative chiral reverse-phase HPLC for the
separation of enantiomers of 4. 108

Figure 5.6. Chromatogram of analysis of separated enantiomers of 4 on column
with opposite chirality as AX QN column. 109

Figure 5.7. Preliminary biological data of 3-IPEHPC as a racemate. 110

ix

Figure 5.8. Chromatogram of chiral resolution of 3-PEHPC by ligand exchange
reverse-phase HPLC. 112

Figure A.1. Analytical HPLC chromatogram of FAM, TLC purified FL-ALN,
versus TLC purified FL-ALN spiked with FAM. 126

Figure A.2. Chromatograms of semi-preparative HPLC purification of
5(6)-FAMRIS. 127

Figure A.3. Chromatograms for semi-preprative HPLC purifications of
FAM-labeled compounds. 128

Figure A.4. Chromatogram of semi-preparative HPLC purification of
RhR-X-RisPC. 129

Figure A.5. Chromatogram of semi-preparative HPLC purification of AF647RIS. 130

Figure A.6. Chromatogram of semi-preparative HPLC purification of
AF647RISPC. 131

Figure A.7.
1
H NMR of 5. 132

Figure A.8.
31
P{
1
H} NMR of 5. 133

Figure A.9.
1
H NMR of 6. 134

Figure A.10.
31
P{
1
H} NMR of 6. 135

Figure A.11.
1
H NMR of 7. 136

Figure A.12.
31
P{
1
H} NMR of 7. 137

Figure A.13.
1
H NMR of 8. 138

Figure A.14.
31
P{
1
H} NMR of 8. 139

Figure A.15.
1
H NMR of 9. 140

Figure A.16.
31
P{
1
H} NMR of 9. 141

Figure A.17.
1
H NMR of 10. 142

Figure A.18.
31
P{
1
H} NMR of 10. 143

Figure A.19.
1
H NMR of 11. 144
x
Figure A.20.
31
P{
1
H} NMR of 11. 145

Figure A.21.
1
H NMR of 12. 146

Figure A.22.
31
P{
1
H} NMR of 12. 147

Figure A.23.
1
H NMR of 13. 148

Figure A.24.
31
P{
1
H} NMR of 13. 149

Figure A.25.
1
H NMR of 14. 150

Figure A.26.
31
P{
1
H} NMR of 14. 151

Figure A.27.
1
H NMR of 15. 152

Figure A.28.
1
H NMR of 16. 153

Figure A.29.
1
H NMR of 17. 154

Figure A.30.
31
P{
1
H} NMR of 17. 155

Figure A.31.
1
H NMR of 18. 156

Figure A.32.
31
P{
1
H} NMR of 18. 157

Figure A.33. Normalized UV absorption spectra of 5(6)-FAM vs. FAM-labeled
RIS compounds. 158

Figure A.34. Fluorescence emission spectra of 5(6)-FAM vs. FAM-labeled RIS
compounds. 159

Figure A.35. UV absorption spectra of RhR-X-RisPC. 160

Figure A.36. Fluorescence emission spectra of RhR-X-RisPC. 161

Figure A.37. UV absorption spectra of AF647RIS. 162

Figure A.38. Fluorescence emission spectra of AF647RIS. 163

Figure A.39. HRMS of 12. 164

Figure A.40. HRMS of 13. 165

Figure A.41. HRMS of 15. 166
xi

Figure A.42. HRMS of 16. 167

Figure A.43. HRMS of 17. 168

Figure A.44. HRMS of 18. 169

Figure A.45. MS of 19. 170

Figure A.46. MS of 20. 171

Figure B.1.
1
H NMR of 2. 172

Figure B.2.
31
P{
1
H} NMR of 2. 173

Figure B.3.
1
H NMR of 4c. 174

Figure B.4.
31
P{
1
H} NMR of 4c. 175

Figure B.5.
1
H NMR of 5. 176

Figure B.6.
31
P{
1
H} NMR of 5. 177

Figure B.7.
19
F NMR of 5. 178

Figure B.8.
1
H NMR of 6. 179

Figure B.9.
31
P{
1
H} NMR of 6. 180

Figure B.10.
19
F NMR of 6. 181

Figure B.11.
1
H NMR of 7. 182

Figure B.12.
31
P{
1
H} NMR of 7. 183

Figure B.13.
19
F NMR of 7. 184

Figure B.14. ESI-MS of 7. 185

Figure B.15.
1
H NMR of 1. 186

Figure B.16.
31
P{
1
H} NMR of 1. 187

Figure B.17.
19
F NMR of 1. 188

xii
Figure B.18. ESI-MS of 1. 189

Figure C.1.
1
H NMR of racemic 3-IPEHPC. 190

Figure C.2.
31
P{
1
H} NMR of 3-IPEHPC. 191

Figure C.3.
1
H NMR of 3-IPEHPC, “A” enantiomer. 192

Figure C.4.
31
P{
1
H} NMR of 3-IPEHPC, “A” enantiomer. 193

Figure C.5.
1
H NMR of 3-IPEHPC, “B” enantiomer. 194

Figure C.6.
31
P{
1
H} NMR of 3-IPEHPC, “B” enantiomer. 195

Figure C.7. HRMS of racemic 4. 196

xiii
List of Schemes

Scheme 2.1. Two different approaches towards a fluorescently labeled
analogue of 1 based on ester linkages. 24

Scheme 2.2. Alkylation of 1 by 1, 5-hexadiene diepoxide. 25

Scheme 2.3. Alkylation of 1 by oxiran-2-ylmethyl-2-(tert-butylamino)
propanoate. 26

Scheme 2.4. Synthesis of linker-containing RIS analogues using tert-butyl
(oxiran-2-ylmethyl)carbamate (4). 26

Scheme 2.5. Synthesis of fluorescent and near-IR labeled risedronate
analogues, 11-20. 29

Scheme 4.1. Synthesis of Pam monomer (2). 76

Scheme 4.2. Synthesis of Bis(propan-2-yl){1-[bis(propan-2-yloxy)phosphoryl]
-7-bromo-1-fluoroheptyl} phosphonate (5) by alkylation of
TIPMFBP (4a). 78

Scheme 4.3. Alkylation of TIPMBP (3) by 8-bromooctyl phthalimide. 79

Scheme 4.4. Total synthesis of target compound, [1-fluoro-1-phosphono
-7-(prop-2-enamido)heptyl]phosphonic acid (1). 81

Scheme 4.5. General synthesis of target compound, 1, from key
intermediate 7 and various acryl starting materials. 85

Scheme 5.1. Total synthesis of 3-IPEHPC. 105

xiv
Abstract

Bisphosphonates are structural analogues of pyrophosphate with increased
hydrolytic stability. The P-C-P backbone of BPs gives rise to two additional side chains,
which can be “tuned” to alter the BP’s characteristics. The ionizable phosphonate groups
chelate metal ions, such as Ca
2+
, giving BPs exceptional bone affinity. Thus, BPs have
been used for several applications due to their bone-targeting efficiency.
Nitrogen-containing bisphosphonates (N-BPs) are potent anti-resorptive agents
clinically used for the treatment of various metabolic bone diseases associated with
excessive bone resorption. N-BPs, such as risedronate (RIS), inhibit farnesyl diphosphate
synthase (FPPS), impairing the prenylation of small GTPase proteins and inducing
apoptosis of osteoclasts. N-BPs and their phosphonocarboxylate (PC) analogues have
also shown promising anti-tumor effects. However, the exact mechanism by which these
promising anti-cancer drugs exert their effects remain unknown. Their cellular and
skeletal distributions are also relatively unclear. Therefore, specific imaging probes to
elucidate these properties of BP drugs are highly desirable.
A coupling reaction was identified that introduces a universal linker group to RIS
and its related analogues, 3-PEHPC (where a phosphonate is replaced by a carboxylate)
and desoxyRIS (where α–OH is replaced by α-H), in a mild (aqueous, pH near neutral,
40-50 ˚C) synthetic step, giving high yields and regioselectivity. The key linker includes
a primary amine necessary for direct acylation by an activated ester of an fluorophore,
leading to the first syntheses of fluorescent and near-IR analogues of RIS. The BP
xv
imaging probes have been visualized in vitro and in vivo and continue to be critical tools
for investigating these drugs.
Self-etching enamel-dentin adhesives, based on polymerizable, strongly acidic
monomers, are crucial to modern dentistry but are limited by their technique sensitivity
and inadequate storage stability at ambient temperature. The introduction of
hydrolytically stable bonds between the polymerizable group and a strongly acidic group
are believed to improve stability. Thus, polymerizable BP derivatives appear to be
promising targets. The synthesis of a novel α-fluorinated, polymerizable bisphosphonic
acid, [1-fluoro-1-phosphono-7-(prop-2-enamido)heptyl]phosphonic acid, was
accomplished to achieve a compound that includes the following ideal characteristics:
enhanced acidity, increased hydrophobicity, and improved hydrolytic stability.

1
Chapter 1
Bone-Targeting Bisphosphonic Acids

Bisphosphonates (BPs) are structural analogues of inorganic pyrophosphate (PP
i
),
where the P-O-P backbone is replaced with P-C-P, with increased resistance to enzymatic
and chemical hydrolysis (Figure 1.1)
1-3
. The ionizable phosphonate groups are capable of
chelating metal ions, such as Ca
2+
, in a divalent manner, and are responsible for the BPs’
strong affinity to hydroxyapatite (HAP), an inorganic mineral found in bone
1-3
. The
introduction of the central carbon gives rise to two additional side chains, which can be
“tuned” to alter the BP’s characteristics
1, 3
. Thus, these properties of BPs have been
exploited to design and synthesize a vast array of bone-targeting compounds. This
introductory chapter will first briefly describe the clinically important nitrogen-
containing BPs as anti-resorptive agents and their mechanism of action and then discuss
several examples of bisphosphonates as bone-targeting agents for various applications.
Figure 1.1. Structure of pyrophosphate (left) versus general structure of a bisphosphonate (right).

Anti-Resorptive Drugs
BPs are the major class of drugs for the treatment of diseases associated with
excessive bone resorption; clinical uses include osteoporosis, Paget’s disease, bone
problems associated with malignancy (such as hypercalcemia and/or increased bone
R
1
P
P R
2
OH
O
OH
OH
O
OH
O
P
P OH
O
OH
OH
O
OH
2
destruction), and osteogenesis imperfecta
1
. Rapid clearance from circulation and high
concentration of BPs in bone brings them in close vicinity to osteoclasts
4
. During bone
resorption, the acidic environment of the resorption lacuna increases the BPs’
dissociation from HAP by protonating the phosphonate groups, which in turn allows BPs
to become internalized by osteoclasts via fluid phase endocytosis
5-7
. Subsequent
acidification of the intracellular endocytic vesicles releases the BPs into the cytosol,
where they exert their biochemical effects
5-7
.
Altering R
1
and R
2
substituents of BPs significantly affect HAP affinity and
enzymatic activity. Structure-activity relationships have revealed that amino or hydroxyl
groups at R
1
allow the BP to bind in a tridentate manner to bone mineral and enhance its
bone affinity
1, 2
. Moreover, for BPs that include a nitrogen within their R
2
side chain, the
charge on the nitrogen, particularly at physiological pH, may also affect HAP binding
1, 2,
8
.
First generation BPs, such as etidronate and clodronate, include simple alkyl or
halogen groups at the R
2
position (Figure 1.2). Due to their structural mimicry of PP
i
,
these BPs are metabolized into β, γ-methylene analogues of adenosine triphosphate
(ATP), forming non-hydrolyzable AppCp-type nucleotides that accumulate in the
osteoclast and induce apoptosis
1, 4
. However, nitrogen-containing BPs (N-BPs), which
include alkylamino (second generation; e.g. alendronate, pamidronate, ibandronate) and
heterocyclic (third generation; e.g. risedronate, zoledronate, minodronate) R
2
groups, are
known to inhibit farnesyl diphosphate synthase (FPPS), an enzyme involved in the
mevalonate pathway (Figure 1.2)
1, 4
. Isoprenoid diphosphates, such as farnesyl
diphosphate (FPP) and geranylgeranyl diphosphate (GGPP), are utilized by enzymes of
3
this pathway for the biosynthesis of sterols and are critically important for post-
translational modification (prenylation) of small GTPase proteins, a process necessary for
proper cell function. Thus, inhibition of FPPS leads to the reduction of concentrations of
both FPP and GGPP, which in turn prevents prenylation and results in the loss of
osteoclast function and eventually apoptosis
1, 4
. Moreover, FPPS inhibition also causes
the accumulation of isopentyl pyrophosphate (IPP), resulting in an acute phase reaction
(fever and flu-like symptoms)
1, 4
. Furthermore, IPP accumulation has also recently been
shown to lead to the condensation of IPP to adenosine monophosphate (giving ApppI),
leading to the inhibition of adenine nucleotide translocase and osteoclast apoptosis
1, 4
.

Figure 1.2. Representative structures of 1st generation (clodronate and etidronate), 2nd generation
(pamidronate, alendronate, and ibandronate), and 3rd generation (risedronate, zoledronate, and
minodronate) bisphosphonates.

Crystallographic studies have shown that the orientation of the nitrogen of N-BPs
with respect to the phosphonate moieties affects the drug’s ability to inhibit FPPS. A
Cl P
P Cl OH
O
OH
OH
O
OH
H
3
C P
P HO OH
O
OH
OH
O
OH
P
P OH
O
OH
OH
O
OH
H
2
N
HO
P
P OH
O
OH
OH
O
OH
HO
H
2
N
P
P OH
O
OH
OH
O
OH
N
HO
P
P OH
O
OH
OH
O
OH
HO
N
P
P OH
O
OH
OH
O
OH
N
HO
N
P
P OH
O
OH
OH
O
OH
HO
N
N
clodronate etidronate
ibandronate alendronate pamidronate
minodronate zoledronate risedronate
4
distance of 3 Å between the nitrogen of the BP and the hydroxyl group of threonine-201
and/or the backbone carbonyl of lysine-200 reportedly gives optimal binding
1, 9
. This
effect can be observed with risedronate and zoledronate, which are able to bind more
tightly and exert greater inhibitory effects as compared with alendronate and pamidronate
(which are unable to have similar interactions with these key residues of the target
enzyme)
1, 9
. Additionally, upon binding to FPPS, BPs induce a conformational change of
the enzyme, which is then followed by the binding of IPP. This forms a tightly-bound
inhibition complex that increases the stabilization of the FPPS and N-BP complex, an
effect that is less pronounced in the weaker bound BPs, such as alendronate and
pamidronate
9
.
Although BPs have been clinically used for several decades, their mechanism of
action has only recently been elucidated, and much remains unknown. Ongoing studies
are focused on further understanding their exact anti-tumor properties as well as
identifying if non-osteoclastic cells (such as osteocytes, osteoblasts, and cancer cells) are
also affected by BPs
4, 6, 10
. It has also recently been proposed that lower affinity BPs (and
their phosphonocarboxylate (PC) analogues) may diffuse differently to skeletal
compartments and may be potentially more attractive as anti-cancer therapeutics than
their higher affinity counterparts; however, our understanding of skeletal distribution
must be improved
11
. Moreover, osteonecrosis of the jaw has only recently been
associated with BPs as a serious side effect although the exact pathology of this disease
remains unclear
12
. Clearly, much remains to be understood about these highly potent and
clinically important therapeutic agents.

5
Bone-Seeking Agents
The chelating properties and exclusive localization of BPs to bone mineral has
been extensively exploited to create an interesting variety of bone-targeting agents,
including radiopharmaceuticals, diagnostic tools, drug and protein delivery systems, and
polymers. Due to the extensive work done in this area, the following presents only a few
notable examples that demonstrate the utility of BPs as bone-seeking agents.
Due to their ability to chelate radioactive isotopes while maintaining bone
affinity, BPs have been utilized for radiodiagnostic and radiotherapeutic applications. For
example, the β-emitter
153
Samarium-tetraphosphonate ethylenediamine-
tetramethylenephosphonic acid (
153
Sm-EDTMP), where EDTMP chelates
153
Sm in a 1:1
ratio, accumulated in metastatic areas four times greater than in normal bone and was
recently approved for clinical use by the US FDA for pain palliation in cancer
metastasized to bone
3
. BPs have also been useful in skeletal radiodiagnostics, such as the
ϒ-emitter
99m
Technetium-methylene diphosphonate (
99m
Tc-MDP), which was visualized
at osseous sites of tumor burden
2
. Moreover, the ϒ-emitter
186
Rhenium-1-hydroxy-
ethylidene-1,1-diphosphonate (
186
Re-HEDP) has been used as a diagnostic tool for
metastatic cancers within bone, but due to its ability to emit β-particles, was also found to
be palliative for treatment of painful bone metastases even at small doses
2
. However,
both of these compounds are dependent on radioactive emitters for detection and thus
also present significant health hazards related to exposure
2
.
Near-infrared (near-IR) labels are thus an attractive alternative for radiolabels.
Previously, the near-IR analogue of pamidronate was visualized in bone and also
provided real-time intraoperative near-IR imaging of HAP in bone and soft tissue in
6
animal models
13, 14
. In addition to avoiding exposure to radioactive isotopes, this imaging
technique produces a much higher resolution compared to
99m
Tc-MDP imaging
2
.
In delivery systems, BPs are typically covalently attached to either a drug or
protein for site-specific transport to bone. Anticancer, antibacterial, or anti-osteoporosis
agents have previously been coupled to BPs to form novel delivery systems
2
. For
example, although hormone replacement therapy is considered an effective treatment for
postmenopausal osteoporosis, systemic adverse side effects, including risk of cancer, are
possible due to the presence of estrogen receptors in several tissues
3
. Thus, BP prodrugs,
in the hope that the BP will deliver the drug to bone and enzymatic and/or chemical
mechanisms will subsequently release the active drug, are believed to alleviate this issue.
Thus, estradiol was conjugated to a BP via a cleavable ester bond and was shown to be
much more effective than the parent drug alone, thus demonstrating its selective and
sustained therapeutic effects on bone (Figure 1.3A)
3
. More recently, bisphosphonate
prodrugs of fluoroquinolone conjugates have also been reported for the treatment of
osteomyelitis, an inflammatory process accompanied by bone necrosis and resulting from
an underlying microbial infection
15
. Antibacterial agents to treat this disease face many
challenges, including a physiological environment that is difficult for the immune system
to access and that highly favors bacterial cells to increase their resistance. Thus, prodrugs
that deliver the antibacterial directly to the bone where the drug would be released would
address these therapeutic challenges. 94.7% of the exemplary BP-derived compound
shown in Figure 1.3B was bound to bone within 1 hour alone, indicating that the prodrug
maintained a high affinity for osseous tissues
15
. Additionally, the compound not only
demonstrated reasonable rates for releasing the parent drug, but also efficaciously
7
prevented infection in a rat model of osteomyelitis, an effect that is not observed by the
parent drug gatifloxacin alone; further in vitro and in vivo analyses are currently
ongoing
15
. Thus, BP prodrugs can clearly improve therapeutic efficiency, especially in
cases were site-specific delivery is of critical importance.
Figure 1.3. Examples of bisphosphonate prodrugs. A) Structure of estradiol conjugated to a
bisphosphonate via a cleavable ester bond. B) Structure of bisphosphonate prodrug of gatifloxacin.

Although a large number of proteins may be considered as potential therapeutic
agents, their clinical use is limited by their instability under in vivo conditions and short
half-lives, resulting in the need for repeated administration, which may alternately cause
adverse side effects on other tissues
2
. To conjugate a BP to proteins, only hydrophilic
BPs with readily-reactive functional groups that can be easily conjugated to proteins
(such as amines, thiols, and carboxylates) have been studied in order to avoid abolishing
protein activity
2
. The first reported example of bone-specific protein delivery involved
the conjugation of a BP to bovine serum albumin, which demonstrate an affinity for HAP
and, upon systemic injection, were found to exhibit up to 7-fold increased bone
O
NH
O
HO
H
2
N
O OH
O
O
O
NH
P P
HO OH
OH HO
O O
O OH
OH
N
O
HO
O
F
N
N
O
OMe
O
P
P
O
OH
O
OH
OH
OH
A
B
-
O CF
3
O
8
deposition versus the protein alone
2, 3
. Proteins are considerably much larger than small
molecule drugs and thus present a special challenge in terms of delivery. Proteins are,
therefore, often substituted with multiple copies of BPs to improve their targeting
efficiency, but minimal protein modification is also critical. Thus, to achieve a higher
substitution efficiency with minimal effect on the protein, dendritic BP compounds
(containing two or four bisphosphonic moieties) have been reported and attached to the
protein albumin via a carboxylate group on the BP, yielding conjugates with sufficient
bone targeting properties
2
.
Bone-targeting BPs have also been incorporated into polymers for site-specific
delivery and/or therapeutic capabilities. Polymers are an attractive platform for drug
delivery due to controlled release of drugs, the possible attachment of several molecules
of a drug and/or combination of drugs, and improved efficacy with decreased side effects.
For example, a pamidronate-based monomer (where the polymerizable acryl group was
conjugated to pamidronate’s primary amine via an amide linkage) was polymerized to
form a hydrogel, giving a novel therapeutic biomaterial that maintained affinity for HAP
(Figure 1.4A)
16
. N-(2-hydroxypropyl)methacrylamide (HPMA)-based copolymers,
containing alendronate as a bone-targeting and anti-resorptive agent, have demonstrated
their affinity for HAP in vitro and accumulated in bone in vivo (Figure 1.4B)
17
. More
recently, alendronate (again employed as a bone-seeking moiety) and the well-known
chemotherapeutic agent, paclitaxel, were recently attached to a HPMA-based copolymer,
resulting in a conjugate that showed promise as an effective antiangiogenic agent in vitro
(Figure 1.4C)
18
.

9
*
NH O
P P
OH
O
HO
HO
O
OH
OH
n
*
NH O
P P
OH
O
HO
HO
O
OH
OH
x
*
O NH
y
Pamidronate-containing
homopolymer
Pamidronate-containing
copolymer
C C
CH
2
CH
2
CH
2
NH
O
NH
CO
2
H
O HO O
NH
O
CH
2
CH HO
CH
3
H
2
C
H
2
C
H
2
C C *
O
NH
CH
2
O
NH
CH
2
O
NH
CH
2
CH
2
CH
2
P
OH
P
O O
OH
OH HO
HO
m n l
Alendronate-containing HPMA copolymer
A B
H
N
O O
O
O
O
O
O
HO
O
O
O
O
H
O
OH
O
O
NH
O
NH
HN
O
O
NH
O
O
CH
2
NH
O
NH
O
NH
O
CH
2
NH
O
H
2
C *
CH
3
C
z
CH
3 H
2
C
CH
3 H
2
C
O
NH
CH
2
HC OH
CH
3
C
O
NH
CH
2
O
NH
O
NH
O
NH
CH
2
NH
P
P
OH
HO
O
HO
HO OH
O
H
2
C
CH
3
*
O
NH
CH
2
HC OH
CH
3
y x x
O
paclitaxel-alendronate conjugate
x=90%
y=2.5%
z=4%
C

Figure 1.4. Structures of exemplary polymerizable BPs. A) pamidronate-based homopolymer and
copolymer; B) alendronate-containing HPMA polymer; C) HPMA-based delivery system of alendronate
and paclitaxel.

The above examples illustrate the wide variety of molecules with exceedingly
different physicochemical properties that have been successfully utilized for bone-
targeted delivery, indicating the utility of BPs as universal carriers. One limitation is that
each BP derivative must be designed and optimized on a compound-to-compound basis
10
to maximize efficiency
3
. Moreover, after rapid injection at high doses, BP conjugates
may form precipitates of calcium salt/complex in plasma, which may result in adverse
effects in other tissues
3
. However, bone-seeking agents remain promising, and
exploration in this area is still ongoing, including optimization of cleavable linkages,
length of chains between the BP and drug/protein, and, in cases of protein delivery,
targeting efficiency (the ratio of BP-protein conjugate delivery to bone versus native
protein delivery to bone)
2
.

Conclusion
The P-C-P backbone of BPs provides chemical and enzymatic resistance against
hydrolysis and imparts the ability to strongly chelate ions, such as Ca
2+
found in HAP,
resulting in the BPs’ high affinity for bone mineral. Thus, BPs are not only clinically
used as therapeutic agents, but have also become widely exploited for various
applications due to their bone-targeting properties. Although BPs are the major class of
anti-resorptive agents for the treatment of metabolic bone diseases, our understanding of
their cellular and skeletal distribution, anti-cancer properties, and relation to ONJ must
urgently be improved. The synthesis and biological evaluation of bone-targeting BPs as
imaging probes will be discussed in chapters 2 and 3, respectively; these fluorescent and
near-IR labeled BPs are proving to be useful tools for increasing our current knowledge
of these important drugs. The synthesis of an α-fluorinated bisphosphonic acid monomer
is presented in chapter 4; in this case, we take advantage of the chelating properties and
acidity of this BP for its application as a component of dental materials. Finally,
decreased affinity to bone can also be useful in certain applications, and chapter 5 will
11
discuss chiral PC analogues of N-BPs, which exhibit decreased bone affinity but also
represent a promising new direction for anticancer therapeutics.
12
Chapter 1 References

1. Russell, R. G. G.; Watts, N. B.; Ebetino, F. H.; Rogers, M. J., Mechanisms of
Action of Bisphosphonates: Similarities and Differences and Their Potential Influence on
Clinical Efficacy. Osteoporosis International 2008, 19, (6), 733-759.
2. Zhang, S.; Gangal, G.; Uludag, H., Magic Bullets for Bone Diseases: Progress in
Rational Design of Bone-seeking Medicinal Agents. Chemical Society Reviews 2007, 36,
(3), 507-531.
3. Hirabayashi, H.; Fujisaki, J., Bone-specific Drug Delivery Systems: Approaches
Via Chemical Modification of Bone-seeking Agents. Clinical Pharmacokinetics 2003,
42, (15), 1319-1330.
4. Roelofs, A. J.; Thompson, K.; Gordon, S.; Rogers, M. J., Molecular Mechanisms
of Action of Bisphosphonates: Current Status. Clinical Cancer Research 2006, 12, (20,
Pt. 2), 6222s-6230s.
5. Thompson, K.; Rogers, M. J.; Coxon, F. P.; Crockett, J. C., Cytosolic Entry of
Bisphosphonate Drugs Requires Acidification of Vesicles After Fluid-phase Endocytosis.
Molecular Pharmacology 2006, 69, (5), 1624-1632.
6. Coxon, F. P.; Thompson, K.; Roelofs Anke, J.; Ebetino, F. H.; Rogers Michael, J.,
Visualizing Mineral Binding and Uptake of Bisphosphonate by Osteoclasts and Non-
resorbing Cells. Bone 2008, 42, (5), 848-60.
7. Sato, M.; Grasser, W.; Endo, N.; Akins, R.; Simmons, H.; Thompson, D. D.;
Golub, E.; Rodan, G. A., Bisphosphonate Action. Alendronate Localization in Rat Bone
and Effects on Osteoclast Ultrastructure. Journal of Clinical Investigation 1991, 88, (6),
2095-105.
8. Nancollas, G. H.; Tang, R.; Phipps, R. J.; Henneman, Z.; Gulde, S.; Wu, W.;
Mangood, A.; Russell, R. G. G.; Ebetino, F. H., Novel Insights into Actions of
Bisphosphonates on Bone: Differences in Interactions with Hydroxyapatite. Bone 2006,
38, (5), 617-627.
9. Kavanagh, K. L.; Guo, K.; Dunford, J. E.; Wu, X.; Knapp, S.; Ebetino, F. H.;
Rogers, M. J.; Russell, R. G. G.; Oppermann, U., The Molecular Mechanism of Nitrogen-
containing Bisphosphonates as Antiosteoporosis Drugs. Proceedings of the National
Academy of Sciences of the United States of America 2006, 103, (20), 7829-7834.
10. Lueftner, D.; Henschke, P.; Possinger, K., Clinical Value of Bisphosphonates in
Cancer Therapy. Anticancer Research 2007, 27, (4A), 1759-1768.

13
11. Fournier, P. G. J.; Daubine, F.; Lundy, M. W.; Rogers, M. J.; Ebetino, F. H.;
Clezardin, P., Lowering Bone Mineral Affinity of Bisphosphonates as a Therapeutic
Strategy to Optimize Skeletal Tumor Growth Inhibition In Vivo. Cancer Research 2008,
68, (21), 8945-8953.
12. Krueger, C. D.; West, P. M.; Sargent, M.; Lodolce, A. E.; Pickard, A. S.,
Bisphosphonate-induced Osteonecrosis of the Jaw. Annals of Pharmacotherapy 2007, 41,
(2), 276-284.
13. Bhushan, K. R.; Tanaka, E.; Frangioni, J. V., Synthesis of Conjugatable
Bisphosphonates for Molecular Imaging of Large Animals. Angewandte Chemie,
International Edition 2007, 46, (42), 7969-7971.
14. Zaheer, A.; Lenkinski, R. E.; Mahmood, A.; Jones, A. G.; Cantley, L. C.;
Frangioni, J. V., In Vivo Near-Infrared Fluorescence Imaging of Osteoblastic Activity.
Nature Biotechnology 2001, 19, (12), 1148-1154.
15. Houghton, T. J.; Tanaka, K. S. E.; Kang, t.; Dietrich, E.; Lafontaine, Y.; Delorme,
D.; Ferreira, S. S.; Viens, F.; Arhin, F. F.; Sarmiento, I.; Lehoux, D.; Fadhil, I.; Laquerre,
K.; Liu, J.; Ostiguy, V.; Poirer, H.; Moeck, G.; Parr, J., Thomas R.; Far, A. R., Linking
Bisphosphonates to the Free Amino Groups in Fluoroquinolones: Preparation of
Osteotropic Prodrugs for the Prevention of Osteomyelitis. Journal of Medicinal
Chemistry 2008, 51, (21), 6955-6969.
16. Wang, L.; Zhang, M.; Yang, Z.; Xu, B., The First Pamidronate Containing
Polymer and Copolymer. Chemical Communications 2006, 2006, (26), 2795-2797.
17. Wang, D.; Miller, S. C.; Sima, M.; Kopeckova, P.; Kopecek, J., Synthesis and
Evaluation of Water-Soluble Polymeric Bone-Targeted Drug Delivery Systems.
Bioconjugate Chemistry 2003, 14, (5), 853-859.
18. Miller, K.; Erez, R.; Segal, E.; Shabat, D.; Satchi-Fainaro, R., Targeting Bone
Metastases with a Bispecific Anticancer and Antiangiogenic Polymer-alendronate-taxane
Conjugate. Angewandte Chemie, International Edition 2009, 48, (16), 2949-2954.

14
Chapter 2
Synthesis of Fluorescent and Near-IR Labeled Risedronate and Related
Analogues

Introduction
Nitrogen-containing bisphosphonates (N-BPs), including risedronate (RIS, 1) are
potent anti-resorptive agents clinically used for the treatment of various metabolic bone
diseases associated with excessive bone resorption (Figure 2.1)
1, 2
. By inhibiting farnesyl
diphosphate synthase (FPPS), prenylation of small GTPase proteins is prevented, which
impairs their correct function necessary for osteoclast activity and induces apoptosis
1, 2
.
The P-C-P backbone gives N-BPs its exceptionally high affinity for
hydroxyapatite (HAP), an inorganic material found in bone; the α-OH group is believed
to enhance this characteristic by allowing for a tridentate interaction with Ca
2+
in bone
mineral
1
. Thus, modification of a phosphonate and/or the α-OH moiety results in
decreased bone mineral affinity
1
. Phosphonocarboxylate (PC) analogues of BPs, where
one phosphonate group is replaced with a carboxylate (such as 3-PEHPC, 2), also exhibit
decreased bone affinity but still retain anti-resorptive properties (Figure 2.1)
3
. Unlike
their parent compounds, PCs have a different enzymatic target, inhibiting Rab
geranylgeranyl transferase (RGGT), thereby selectively preventing the prenylation of Rab
proteins
3, 4
. In contrast, desoxyRIS (3), an analogue of RIS where the α-OH is replaced
with α-H, exhibits lowered bone affinity and decreased potency versus its parent drug
(Figure 2.1)
5
.
15
Figure 2.1. Structures of RIS (1) and its related analogues, 3-PEHPC (2) and desoxyRIS (3).

N-BPs and their PC analogues have recently shown promising anti-tumor effects,
demonstrating the ability to inhibit tumor cell adhesion, invasion, angiogenesis, and
proliferation
2, 6, 7
. Moreover, it has been suggested that employing compounds with
decreased bone affinity (such as 2) may be an effective strategy to inhibit tumor growth
in vivo since it appears that bone affinity and anti-tumor potential may be inversely
related
2, 7
. However, questions regarding the exact mechanisms underlying these effects
remain unclear: do these observed effects occur directly in tumor cells or as a result of
bone resorption inhibition? Moreover, the skeletal distribution of BPs and PCs in bone
and other tissue and their ability to affect other non-osteoclast cells are not well
understood
1, 8
. Furthermore, osteonecrosis of the jaw (ONJ) has only recently been
reported as a serious side effect in patients taking N-BPs
9, 10
. Clearly, our understanding
of these key mechanisms and properties of N-BPs must be urgently improved.
The radiolabeled 4-amino-1-hydroxybutylidene-1,1-bisphosphonic acid, [
3
H]-
alendronate, was previously used to study cellular uptake and localization
11
. The BP was
taken up into osteoclasts and no other cell types; however, experimental limitations
involved with radiolabels may have hampered detection sensitivity
11-13
.
Alternatively, fluorescent or near-IR labels have become invaluable tools in
proteomics, genomics, and pharmacological drug development due to their high
N N
P
P HO OH
O
OH
OH
O
OH
P
HO OH
O
OH
O
OH
N
P
P H OH
O
OH
OH
O
OH
1 2 3
16
sensitivity and ease of detection
14, 15
. Compared to radiolabeled tags, fluorescent and
near-IR labels give the following advantages: higher resolution, reduced radiation
exposure, rapid acquisition times, visualization of skeletal events over extended time
periods, and simultaneous multichannel imaging with other fluorescent markers
12, 13, 16
.
For drug discovery, in particular, labels with carboxylic acid moieties can be conjugated
to a primary amine of a parent drug with traditional coupling reagents, such as N,N’-
dicyclohexylcarbodiimide (DCC)
17
. Imaging agents are also often commercially available
as activated esters, allowing for facile acylation to the drug’s primary amine. A widely
known disadvantage of this method involves the rapid decomposition of the label’s
activated ester in the aqueous environment of the conjugation reaction mixture, often
resulting in contamination by free label. Both labeling approaches result in the formation
of a stable amide bond, a linkage that typically resists hydrolysis in vitro and in vivo,
between the label and drug.
Imaging probes as analogues of N-BPs are ideal tools for elucidating key
properties to further our understanding of these drugs and their actions. Fluorescently
labeled isoprenoid and aminomethylene bisphosphonates have been previously used to
study skeletal distribution and cellular uptake
18-20
. Near-IR analogues of pamidronate,
including Pam78, Pam800, and OsteoSense™680 and 750 have also been visualized in
vitro and in vivo although pharmacological activity was not reported (Figure 2.2)
13, 21, 22
.
Fluorescent analogues of alendronate, AF-ALN and FL-ALN (labeled with Alexa Fluor
488™ and carboxyfluorescein, respectively), have previously been used to study cellular
uptake; however, the labeled drugs were not completely purified and were unable to
inhibit protein prenylation (Figure 2.2)
8, 23
.
17
In the synthesis of fluorescent alkylamino BPs, the conversion of the terminal
amino group of the drug to an amide linkage decreases the basicity of the nitrogen and
alters the inhibitory effects of the resulting BP
13
. This type of direct acylation to the label
by the BP is also not applicable to the more potent and pharmalogically important
heterocyclic N-BPs, such as RIS, which lack a primary amine necessary for facile
acylation to activated esters of fluorescent labels.
P
P HO OH
O
OH
OH
O
OH
N
H
R
1
P
P HO OH
O
OH
OH
O
OH
H
N
R
2
O H
2
N NH
2
SO
3
H SO
3
H
CO
2
H
O
O HO O
CO
2
H
O
N
SO
3
H
HO
3
S O
N
SO
3
H
SO
3
H
O
N
SO
3
H
HO
3
S O
SO
3
H
N
SO
3
H
O
FL-ALN,
R
1
=
Pam78,
R
2
=
Pam800,
R
2
=
AF-ALN,
R
1
=
Alendronate,
R
1
= H
Pamidronate,
R
2
= H

Figure 2.2. Structures of previously reported fluorescent alkylamino BPs. Osteosense™ 680 and 750
structures have not been reported.

In order to label heterocyclic N-BPs, a universal linker must be incorporated to
the drug that introduces a functional group, such as a primary amine, to allow for
18
subsequent conjugation to any activated ester of a fluorophore. The ideal linker should
not adversely influence bone mineral binding affinity while allowing the drug analogue to
retain some ability to inhibit protein prenylation. Therefore, the phosphonate/carboxylate
and α-OH moieties were not considered as suitable sites for attachment of a linker
moiety. Thus, the pyridine nitrogen appeared to be an attractive alternative, and its
conversion to a positively charged quaternary nitrogen may even simulate a carbocation-
like transition state that could aid the resultant conjugate to retain activity
24, 25
.
Importantly, the introduction of the linker group must also take place in aqueous media,
considering the solubility of the parent BP and PC drugs.
In this chapter, studies on the photostability of carboxyfluorescein (FAM) and
synthesis and purification methods of FL-ALN are first discussed. Then, the first
syntheses of fluorescent and near-IR labeled analogues of heterocyclic N-BPs
(specifically, RIS and its related compounds) are presented. Different possible synthetic
pathways are described, including several N-alkylation reactions that were studied in an
attempt to form a suitable drug-linker complex that could be subsequently conjugated to
an imaging agent. The synthesis of the fluorescent BPs centers on an important coupling
reaction that introduces a universal linker group to compounds 1, 2, and 3 in a
exceptionally mild (aqueous, 40 ˚C) and highly regioselective synthetic step, providing a
primary amine for direct acylation by an activated ester of a label
25, 26
. This technology
has been used towards the synthesis of imaging probes with different labels, such as
FAM, Rhodamine Red-X™ (RhR-X) and Alexa Fluor® 647 (AF647). This differential
labeling approach produces a “matrix” of compounds, opening the doors to multicolor
biological experiments where different BPs and PC compounds may be visualized within
19
the same assay and allowing us to further understand the structural variants of each
compound that may affect its localization and cellular uptake.

Results and Discussion
Although 5(6)-carboxyfluorescein (5(6)-FAM) is widely used and relatively
inexpensive, the label suffers from extreme sensitivity to light and changes in pH
14, 27
. To
understand what precautions are necessary to minimize the compound’s decomposition
during a possible “strenuous” work-up of our labeling reactions, its photostability after
exposure to various levels of light was studied. A 13 µM solution of 5(6)-FAM was
divided into 3 samples, and each sample was subjected to a different level of light
exposure and analyzed by UV absorption spectroscopy (Figure 2.3). After only 30 min
under direct sunlight, the absorption of the label appeared to rapidly drop, while the label
kept in darkness showed no decrease in absorption. Under laboratory light, the label
showed only a slight decrease in absorption. Thus, it appeared that the label would be
fairly photostable under typical laboratory light conditions and was deemed appropriate
for our syntheses. As an extra precautionary measure, the following synthetic procedures
regarding fluorescent labels were performed under minimal exposure to light.

20
Figure 2.3. UV absorption spectra of 13 µM solution of 5(6)-FAM (in buffer containing 5 mM sodium
acetate and 5 mM KH
2
PO
4
) after exposure to 3 different light levels in 30 minutes.

The synthesis of the FAM-labeled analogue of alendronate, FL-ALN, was then
studied. According to the published synthesis of this compound, CaCl
2
is used to
precipitate the desired product from the reaction mixture, which also contains unlabeled
alendronate, FAM, and NaHCO
3
; then, ethylene-bis(oxyethylenenitrilo)tetraacetic acid
(EGTA, equimolar to CaCl
2
) is used to chelate Ca
2+
and thus resolubilize alendronate
8
.
However, CaCl
2
is likely to precipitate all compounds present in the reaction mixture
while EGTA will re-introduce these compounds into the final sample.
Therefore, a two-step purification method of FL-ALN may be employed to yield
a much cleaner compound. By purifying the reaction mixture by TLC (eluted with
MeOH), the free label moved with the eluant while all phosphonate-containing
compounds remained at the origin. The bottom band was then extracted and purified by
21
semi-preparative HPLC to remove unlabeled ALN from the desired product, which
presumably (since it cannot be detected by UV absorption) eluted at a much shorter
retention time. The eluant gradient used for HPLC may also be additionally adjusted to
separate the 5- and 6- isomers of FL-ALN (Figure 2.4).

Figure 2.4. Analytical HPLC chromatograms FL-ALN using 2 different gradients (scalable for purification
conditions). Conditions: Varian C18 (250 x 4.6 mm, 5 µm, 100 Ǻ pore size) column, flow rate 1.5 mL/min,
UV detection at 490 nm., eluted with 50% CH
3
CN in 0.1 N triethylammonium acetate (TEAAc) buffer (pH
6). Gradient as follows: HPLC trace 1 (in blue): 5% to 20% eluant in 15 seconds, linearly increasing to
80% of eluant in 6:40 min, linearly increasing to 100% of eluant in 15 min; HPLC trace 2 (in red): 5% to
1% of eluant 7:20 min, linearly increasing to 70% of eluant in 11:70 min, linearly increasing to 100% of
eluant in 13:20 min.

However, although TLC purification removes the majority of FAM, traces of the
free label may remain with the desired product after HPLC. To help alleviate this issue,
we attempted to extract the free label with an organic solvent, such as MeOH and ethyl
acetate, but this did not appear successful. Thus, HPLC conditions may need to be
further optimized to ensure that traces of FAM may be separable from the product.
Although a useful analytical HPLC method to determine a sample’s purity was identified
(refer to Appendix A for representative chromatogram), the citrate-phosphate buffer used
22
in this case would not be suitable for HPLC purifications. However, purification of FL-
ALN by TLC and HPLC still provides a much cleaner desired product, with only traces
of FAM remaining according to TLC, than the original synthetic procedure, which yielded
a product that was estimated to contain only 30% FL-ALN
8
.
We then turned our attention towards the synthesis of fluorescent heterocyclic N-
BPs, where two different approaches are possible. First, the imaging agent can be coupled
to a linker moiety, which can be subsequently attached to the BP or PC drug. On the
other hand, the linker can be first attached to the drug, forming a drug-linker complex
that can be directly acylated by an imaging agent. Since RIS analogues containing a
quaternary nitrogen have been previously reported, we therefore began our studies by
applying a similar approach towards our synthetic goal
28-30
.
We initially studied the alkylation of the pyridine nitrogen of 1 with various alkyl
dihalides, where an amino group would have to be subsequently introduced to the drug-
linker complex, and N-protected aminoalkyl halides under several reaction conditions
(Figure 2.5). However, these reactions typically resulted in low yields, possibly due to
solubility issues (the alkylating agent being less hydrophilic than the BP). Moreover, the
release of the halide (iodide, bromide, or chloride) in the aqueous environment (needed to
dissolve hydrophilic 1) often lowered the pH of the reaction mixture, resulting in a
protonated, non-nucleophilic pyridinium nitrogen and/or the cleavage of the protecting
group of the amine. If these protecting groups are lost, cyclization of the alkylating agent
may also be possible.

23

Figure 2.5. General structures of dihalide or N-protected aminoalkyl halide alkylating agents studied for
alkylation of RIS.

Fortunately, we found that glycidol alkylates the nitrogen of 1 under fairly mild
conditions, giving a drug-linker complex containing a primary alcohol. The newly
introduced hydroxyl group could then be utilized to form an ester bond to a fluorescent or
near-IR label. However, use of an activating agent, such as DCC, to couple this
intermediate to FAM was not successful (Scheme 2.1A). An epoxide conjugated to a
FAM-based analogue was previously shown to selectively alkylate the nitrogen of a
histidine side chain in human carbonic anhydrase II
31
. Thus, we attempted to label 1 with
an epoxide linked to FAM via an ester bond (Scheme 2.1B). However, yields were fairly
low, perhaps due to the decomposition of the hydrolytically labile ester bond during the
aqueous reaction or TLC purification.
X = Br or I
X (CH
2
)
2
X
X = Br or I
n = 2 or 3
R = tBOC, Me, Ph, or phthalimide
X (CH
2
)n NHR
24
OH
O
N
P
P HO OH
O
OH
OH
O
OH
OH
OH
O HO O
CO
2
H
O
O
OH
N
P
HO P
OH
OH
O OH
OH
O
+ 1
O HO O
CO
2
H
O
O
+
N
P
P HO OH
O
OH
OH
O
OH
O
HO
2
C
O OH
1
i
ii A
B
O
OH
O
O

Scheme 2.1. Two different approaches towards a fluorescently labeled analogue of 1 based on ester
linkages. A) Alkylation of RIS with glycidol, followed by conjugation to FAM. Conditions: i) H
2
O,
tetrabutylammonium hydroxide, pH 7, 50 ˚C; ii) DCC, DMF, catalytic amount of DMAP, rt, overnight. B)
Alkylation of 1 by an epoxide-containing derivative of FAM. Conditions: H
2
O, pH 9-10, 50 ˚C.

Other epoxide-based strategies were also studied, including employing 1, 5-
hexadiene diepoxide as an alkylating agent in the hopes of attaching the drug through one
epoxide ring-opening while a label could be attached similarly at the second epoxide.
However, instead of simply alkylating the pyridinium nitrogen of 1, dimerization may be
occurring, yielding an undesired product (Scheme 2.2).

25

Scheme 2.2. Alkylation of 1 by 1, 5-hexadiene diepoxide. Conditions: H
2
O, pH 5-6.5, 50 ˚C.

Additionally, the alkylation of RIS with oxiran-2-ylmethyl-2-(tert-
butylamino)propanoate, which was synthesized from the conjugation of glycidol and the
commercially available t-BOC protected alanine, was studied (Scheme 2.3). After
deprotection, the primary amine of the linker could be conjugated to an activated ester of
an imaging agent. However, the ester linkage appeared to be susceptible to hydrolysis.
With these results, we decided to pursue linkers containing primary amines in order to
form a more hydrolytically stable amide linkage, as compared to its ester counterpart,
between the drug-linker complex and a fluorophore.
+
O
O
N
P
P HO OH
O
OH
OH
O
OH
OH
HO
N
P
P HO
OH
OH
O
OH
O
OH
1
N
P
P HO OH
O
OH
OH
O
OH
OH
O
Attach appropriate label through
2nd ring opening
X
26
Scheme 2.3. Alkylation of 1 by oxiran-2-ylmethyl-2-(tert-butylamino)propanoate. Conditions: H
2
O/DMF,
pH 6-7, 50 ˚C.

We then identified that the t-BOC protected epoxide 4 alkylated the pyridine
nitrogen of 1 under exceedingly mild (pH near neutral, 40 °C) aqueous conditions, with
high yields and regioselectivity (Scheme 2.4)
25, 26
. The epoxide is dissolved in minimal
MeOH before its addition to the drug in an aqueous environment; care is needed in this
step, since excess MeOH may cause the BP or PC starting material to precipitate.
Scheme 2.4. Synthesis of linker-containing RIS analogues using tert-butyl (oxiran-2-ylmethyl)carbamate
(4). i) H
2
O/MeOH, 40-50 °C, pH 5-6. ii) 50:50 TFA/H
2
O.

The reaction may be followed by reverse-phase HPLC or
31
P NMR (Figure 2.6A-
B). Chuiko et al. previously synthesized aminohydroxypropane esters of 1-
hydroxyethylidene-1,1-bisphosphonic acids in aqueous conditions near neutral at 60-70
°C
32
. However, by altering the temperature in our reaction, a change in regioselectivity
O
O
O
NHBOC
N
P
P HO OH
O
OH
OH
O
OH
OH
O
O
NHBOC
+ 1
NHBOC
O
N
P
O
+
N
P
O
N
P
O
i ii
1 R
1
= OH, R
2
= P(O)(OH)
2
2 R
1
= OH, R
2
= CO
2
H
3 R
1
= H, R
2
= P(O)(OH)
2
5 R
1
= OH, R
2
= P(O)(OH)
2
6 R
1
= OH, R
2
= CO
2
H
7 R
1
= H, R
2
= P(O)(OH)
2
8 R
1
= OH, R
2
= P(O)(OH)
2
9 R
1
= OH, R
2
= CO
2
H
10 R
1
= H, R
2
= P(O)(OH)
2
4
OH
R
1
R
2
R
1
R
2
R
1
R
2
NHBOC
OH
NH
2
OH OH
OH
OH
OH
OH
27
can be observed: at higher temperatures, the generation of esters can be observed while,
at lower temperatures, the regioselectivity becomes absolute, forming only the N-
alkylated drug analogues (Figure 2.6B-C). Since we wished to avoid perturbing the bone
affinity of the resulting conjugates, the tunability of the regioselectivity was an
unexpected but gratifying advantage. After treatment of the BOC-protected linker-drug
intermediates 5-7 with TFA, the corresponding free amino forms 8-10 are achieved
(Scheme 2.4). Each resulting drug-linker complex advantageously contains a primary
amine (after suitable deprotection) for facile conjugation to any activated ester of any
fluorophore, a positively charged pyridinium nitrogen, and an additional hydroxyl group
that may counteract the addition of the hydrophobic alkyl chain.
Figure 2.6. Two methods of following the progression of the reaction between epoxide 4 and RIS. A)
Chromatogram of analytical reverse-phase HPLC of reaction mixture. Conditions: Varian C18 (250 x 4.6
mm, 5 µm, 100 Ǻ pore size) column, flow rate 1.5 mL/min of 10% CH
3
CN in 0.1 N phosphate buffer with
3mM tetrabutylammonium dihydrogen phosphate (pH 7), UV detection at 260 nm. B)
31
P NMR of reaction
mixture of 4 and RIS after 18 h at 40 ˚C (pH 5-6). C)
31
P NMR of reaction mixture of 4 and RIS at 50 ˚C
overnight (pH 6).
28
Although our synthetic approach first attached the linker to the drug, this
universal linker theoretically may also be first conjugated to an activated ester of the
label, and the resultant epoxidized label-linker complex may then be conjugated to the
drug. The amide linkage between linker and label will be less susceptible to hydrolysis
than its ester counterpart (as shown in the epoxidized FAM derivative described above).
However, additional synthetic steps involving the fluorescent or near-IR label may
increase the label’s exposure to light and lower its fluorescence capabilities. Additionally,
solubility issues may arise when attempting to attach the label-linker complex to the
highly hydrophilic drug. Thus, our approach appears to be much more advantageous.
To achieve the first synthesis of fluorescent RIS analogues, 5(6)-
carboxyfluorescein, succinimidyl ester (5(6)-FAM, SE, 21) was readily conjugated to
compounds 8-10 (Scheme 2.5). Aqueous conditions are important for this reaction to
dissolve the BP or PC starting materials; a slightly basic pH is necessary to ensure that
the primary amine of our drug-linker complex is deprotonated, and therefore
nucleophilic. However, these conditions also give rise to an unwanted side reaction: the
hydrolysis of the succinimidyl ester of the imaging agent. In an attempt to achieve
maximum yield, the activated ester is dissolved in organic solvent prior to its addition to
the aqueous solution containing an excess of phosphonate starting material; however,
decomposition of the fluorescent starting material remains difficult to avoid, giving free
label and unlabeled drug as contaminants.
29
Scheme 2.5. Synthesis of fluorescent and near-IR risedronate analogues, 11-20. For AF647 structure, see
discussion below. i) FAM, SE (for compounds 11-17), RhR-X, SE (for 18), or AF647, SE (19-20),
NaHCO
3
, DMF, pH 8-9, 3 h to overnight, in darkness.

Several purification steps for FAM-labeled compounds were attempted, including
size exclusion chromatography, HPLC, and TLC. In the hope that one purification step
could simultaneously remove both contaminants, Sephadex G10 columns eluted with
triethylammonium acetate (TEAAc) buffer were initially used to purify labeling reaction
mixtures. The FAM-labeled compound was clearly visible by eye, and the first orange
eluant was collected as product. However, free label and unlabeled drug may form a salt
with the positively charged quaternary nitrogen of the desired product, and both
N
P
O
OH
R
1
R
2
NH
OH
OH
R
3
11 5(6)-FAMRIS. R
1
= OH, R
2
= P(O)(OH)
2
, R
3
= FAM
12 5-FAMRIS. R
1
= OH, R
2
= P(O)(OH)
2
, R
3
= FAM
13 6-FAMRIS. R
1
= OH, R
2
= P(O)(OH)
2
, R
3
= FAM
14 5(6)-FAMRISPC. R
1
= OH, R
2
= CO
2
H, R
3
= FAM
15 5-FAMRISPC. R
1
= OH, R
2
= CO
2
H, R
3
= FAM
16 6-FAMRISPC. R
1
= OH, R
2
= CO
2
H, R
3
= FAM
17 5(6)-FAMdRIS. R
1
= H, R
2
= P(O)(OH)
2
, R
3
= FAM
18 RhR-X-RISPC. R
1
= OH, R
2
= CO
2
H, R
3
= RhR-X
19 AF647RIS. R
1
= OH, R
2
= P(O)(OH)
2
, R
3
= AF647
20 AF647RISPC. R
1
= OH, R
2
= CO
2
H, R
3
= AF647
O HO O
CO
2
H
O
FAM:
O Et
2
N NEt
2
SO
3
H
SO
2
HN (CH
2
)
5
O
RhR-X:
i
8, 9, or 10
30
contaminants appeared to elute with the product. Thus, reverse-phase HPLC was needed
as an additional purification step.
Similar to our FL-ALN purification method, we then employed preparative TLC
eluted with 100% MeOH to remove any free label; thus, any free label moves with the
solvent (R
f
= 1) while all BP/PC compounds remain at the baseline. Extraction from
silica is performed with water as the solvent, which may not entirely extract the desired
compound from silica and may result in lowered yields. Chelex (sodium form) may be
used to aid the extraction process.
Reverse-phase HPLC purification without prior removal of the majority of FAM
may result in overlapping FAM peaks with the desired product. However, HPLC
conditions suitable for simultaneously removing unlabeled BPs, which have much shorter
retention times, and traces of FAM in the desired products have been identified. Although
TEAAc is an appropriate buffer for such HPLC separations, the higher volatility of
triethylamine versus acetic acid causes a drop in pH, resulting in pH 4-5, when removing
the buffer under vacuum. We have found that, especially for larger scale purifications,
the FAM-labeled products tend to precipitate under these conditions due to their
decreased hydrophilicity in acidic pH
14, 27
. Although the acetate buffer appears to work
fairly well with smaller amounts of compound, triethylammonium carbonate (TEAC) is
capable of maintaining a more basic pH (and thus precipitation of our FAM-labeled
products may be avoided) and is therefore a more suitable buffer for larger scale
purifications.
Individual isomers (5- and 6-FAM) of the labeled products may also be isolated
by semi-preparative reverse-phase HPLC, a more cost effective method than direct
31
synthesis from their respective isomerically pure FAM, SE starting materials. The
1
H and
31
P NMR spectra and HPLC retention time of 5-FAMRIS (12) directly synthesized from
5-FAM, SE correspond to the peak eluting at 44 min, thus confirming our isomer
assignments given to compounds 12 and 6-FAMRIS (13) that were separated by HPLC.
In the
1
H NMR spectra, 12 synthesized from 5(6)-FAM, SE contains a minor singlet at δ
7.56, indicating a trace of the 6-isomer, which is absent from the product directly
synthesized from the pure 5-isomer starting material.
The stability of 11 was studied by storing the compound in various buffers
typically used for biological assays, including phosphate,
tris(hydroxymethyl)aminomethane (TRIS), and 4-(2-Hydroxyethyl)piperazine-1-
ethanesulfonic acid (HEPES)
33
; the solutions were analyzed by TLC (eluted with
MeOH). After compound 11 in 0.1 N phosphate buffer (pH near neutral) was kept frozen
for 1 week, followed by storage at room temperature for 6 days (continuously kept in
darkness), no hydrolyzed free label was detected. Additionally, 11 showed no
decomposition in 50 mM TRIS and 50 mM HEPES (pH near neutral) buffers after 1 day
at room temperature in darkness. Thus, our FAM-labeled compounds appear to be stable
under conditions often employed during biological studies appropriate for these types of
compounds.
Due to their shift towards the red in their absorption and emission spectra,
rhodamine labels encounter little interference from the fluorescence of fluorescein, and
thus the two labels are often used in combination for dual or multiparameter fluorescence
analysis
34
. Therefore, in order to perform multicolor labeling experiments with BP and PC
compounds, compound 9 was also conjugated to Rhodamine Red-X™, succinimidyl ester
32
(RhR-X, SE, 22), commercially available as a single isomer. The presence of the sulfonate
groups of RhR-X, compared to the weakly acidic carboxylate groups of FAM, appeared
to increase the difficulty of extracting the desired products from silica after TLC
purification. We also attempted to employ similar TLC purification conditions (Whatman
LK6F plates eluted with 40% CH
3
CN in H
2
O) previously utilized for Pam78, which was
able to remove unlabeled BP and free label in one step
22
. However, this purification
method appeared to be unsuccessful in our case, which may be due to the permanently
charged pyridinium nitrogen of unlabeled PC compounds forming a salt with our desired
product. To help circumvent these issues, TLC plates were saturated with the crude
reaction mixture and eluted 3-4x with 100% MeOH or until pink bands no longer
appeared to be moving from the baseline.
In addition, HPLC conditions used for purifications of FAM-labeled conjugates
resulted in extremely low yields for RhR-X conjugates. Thus, we again employed the
“saturation” idea towards HPLC and utilized a smaller semi-preparative C18 column than
previously used for FAM products, resulting in much improved yields (up to 10-fold
increase) for RhR-X conjugates. Thus, the increased hydrophobicity (due to the larger
structure of the label) and/or the highly acidic sulfonate groups appear to increase the
difficulty of removing the desired RhR-X conjugate from media such as silica and
stationary phases of reverse-phase HPLC columns, but this problem may be somewhat
alleviated by saturating the media.
33
Near-IR labeled BPs were also of high interest due to their increased sensitivity in
detection for in vivo biological studies
14
. Therefore, RIS analogues 8 and 9 were
conjugated to Alexa Fluor 647®, succinimidyl ester (AF647, SE, 23). TLC purification
was bypassed in hopes of maximizing yield (considering the much higher cost of this
starting material), and the reaction mixture was directly purified by HPLC. We initially
attempted to apply the HPLC conditions utilized for FAM-labeled products towards the
purification of 19 and 20. However, we found that a similar HPLC method to that
employed for purifying the RhR-X conjugate 18 was much more successful for the
purification of AF647-labeled products, and both unlabeled phosphonate compounds and
free label were simultaneously separated from the desired products.
Although the structure of AF647 is proprietary, White et al. cited a 2004 patent
application and published the structure shown in Figure 2.7A
35, 36
. However, further
investigation of the patent application lists the structure as simply “ALEXA dye series
NHS ester”
35
. Characterization of AF647-labeled compounds, such as Alexa-ATP, have
typically not been performed, and thus the reported structure of AF647 has not been
confirmed
37
. By comparing the experimentally determined masses for HPLC purified 19
and 20 (obtained by positive ion ESI-MS), we were able to identify the molecular weight
of AF647 (as a carboxylic, tetrasulfonic acid) as 860.02. Although the observed mass for
AF647 does not correspond to the reported structure, the structure of AF647 may be more
accurately attributed to that shown in Figure 2.7B. The synthesis of the proposed
structure can also be found in a 2002 patent
38
. As an additional confirmation, the
experimentally determined mass of an AF647-labeled peptide ([M + H]
+
= 1954.11 m/z,
34
previously not compared to a theoretical mass) can be compared to the observed mass of
the corresponding unlabeled peptide ([M + H]
+
= 1113 m/z), giving AF647 a molecular
weight that independently corresponds to that of our findings
39, 40
.
N
HO
3
S
C
H
C
H
C
H
C
H
C
H
N
SO
3
H
HO
3
S SO
3
H
O
HO
A B
N
C
H
C
H
C
H
C
H
C
H
N
+
K
-
O
3
S SO
3
-
O
O N
O
O
SO
3
-
K
+
SO
3
-
K
+

Figure 2.7. Structures of reported “Alexa” dye versus proposed AF647. A) Reported Alexa dye structure,
calcd MW (as tripotassium salt): 1097.0985; B) Proposed AF647 structure, calcd MW (acidic form):
860.0228.

Based on the attached fluorescent or near-IR label, each class of RIS imaging
probes synthesized to date has required the identification of its own specific purification
method. FAM, SE (as a mixture or isomerically pure) is the least expensive of the studied
labels, but the FAM-labeled products appear to be more pH sensitive (and may
precipitate at pH < 7) and may require extra care to protect from light
14, 27
. Additionally,
complete removal of the products from HPLC columns and TLC plates is difficult to
achieve, leading to product loss. RhR-X™, SE, isomerically pure, contains a larger
heterocyclic ring system, long alkyl chain, and highly acidic sulfonate group; due to its
structural features, the RhR-X conjugates may remain strongly attracted to silica of TLC
plates and the stationary phase of a reverse-phase HPLC column, perhaps resulting in
lower yields. AF647®, SE, containing 4 highly acidic sulfonate groups, may also interact
with media similar to RhR-X conjugates. Fortunately, unlike the RhR-X and FAM
35
conjugates, HPLC conditions for AF647-labeled analogues were identified that separate
both free label and phosphonate-containing compounds in 1 step. Importantly, this allows
us to bypass TLC purification, a step that appears necessary in other cases where
separation from free label may be difficult to achieve with HPLC alone, and thus
eliminate issues regarding extraction from silica. Considering the relatively high cost of
this starting material, avoiding product loss is of utmost importance; fortunately, the
purification of this class of labels appears to require the least manipulation. Although our
purification methods may require tuning for each appropriate class of imaging probes,
moderate yields and exceptional purity of all labeled products are achieved.
All labeled FAM and RhR-X labeled products were characterized by high
resolution mass spectrometry and
1
H and
31
P NMR, UV absorption, and fluorescence
emission spectra. AF647-labeled imaging probes were not characterized by NMR
spectroscopy due to the limited amount of product, but ESI-MS, UV absorption, and
fluorescence emission spectra correctly correspond to the expected results of an AF647
conjugate. UV spectra of all labeled RIS analogues exhibit similar spectra to their
respective parent dyes
34, 41-43
. Additionally, as compared to 5(6)-FAM, a slightly
increased extinction coefficient at 260 nm can been seen in UV absorption spectra of the
labeled compounds 11-17 due to the presence of the risedronate pyridinium
chromophore.
The emission spectra of FAM-labeled compounds indicate a loss of 10-20% of
fluorescence intensity as compared to 5(6)-FAM, which may be due to slight exposure to
light during work-up or may be an inherent characteristic of the labeled compounds.
Emission spectra of RhR-X and AF647 compounds were not compared to their parent
36
dyes, although labels belonging to the rhodamine and Alexa Fluor families are known for
their increased photostability compared to FAM
14, 27
.
In previously reported fluorescent and near-IR labeled alkylamino BPs, the
pharmacologically important primary amine is converted to an amide linkage, thereby
altering the inhibitory properties of the resultant imaging probe
13
. However, our synthetic
approach for fluorescent RIS analogues produces a permanently positively charged
quaternary pyridinium ring, which may simulate a carbocation-like transition state and
thus contribute to its pharmacological activity
24, 25
. Thus, the design and syntheses of
fluorescent and near-IR labeled RIS analogues allow for the conjugates to maintain their
affinity to bone and inhibitory effects, and are thus useful tools for elucidating key
properties of BPs currently not well understood.

Conclusion
In summary, the photostability of FAM after exposure to various levels of light
and the synthesis and purification of FL-ALN were investigated. Several different
pathways towards the synthesis of fluorescent risedronate analogues were also discussed.
To achieve the first syntheses of fluorescent and near-IR labeled RIS, 3-PEHPC, and
desoxyRIS analogues, a key coupling reaction that introduces a universal linker and
proceeds in exceedingly mild conditions, giving quantitative yields and high
regioselectivity, was identified. By attaching the linker to the nitrogen of the pyridine
ring, the labeled conjugates maintain or acquire additional structural features that may aid
in retaining bone affinity and inhibitory effects. Furthermore, our synthetic strategy not
only allows for the creation of a “matrix” of labeled analogues of RIS for multicolor
37
labeling experiments, but may also be applicable to any nitrogen-containing heterocyclic
compound.

Experimental Methods
Reagents and Spectral Measurements. Funding was provided by Procter & Gamble
Pharmaceuticals. 5(6)-, 5-, and 6-carboxyfluorescein, succinimidyl ester (FAM, SE) were
purchased from Sigma Aldrich. Rhodamine Red-X™, SE and Alexa Fluor® 647, SE
were purchased from Molecular Probes-Invitrogen. Compounds 1, 2, and 3 were kind
gifts from Procter & Gamble Pharmaceuticals. All other compounds were purchased from
Aldrich or Alfa Aesar. Triethylamine (TEA) was distilled from KOH; CH
2
Cl
2
was
distilled from P
2
O
5
; and allylamine was distilled under N
2
. All other compounds were
used as supplied by the manufacturer. Thin layer chromatography was performed on
Merck Silica Gel 60 F
254
plates, and the developed plates were visualized under a UV
lamp at 354 nm. HPLC separations were performed on a Rainan Dynamax Model SD-
200 system with a Rainan Dynamax absorbance detector Model UV-DII. NMR spectra
were recorded on either 400 MHz Varian or 500 MHz Bruker spectrometers. Chemical
shifts are reported relative to internal D
2
O standard (for
1
H NMR spectra) or 85% H
3
PO
4

(for
31
P NMR spectra). UV spectra were recorded on a DU 800 spectrometer, and
fluorescence emission spectra were taken on a Jobin Yvon Horiba FluoroMax-3
fluorimeter equipped with a DataMax Software version 2.20 (Jobin Yvon Inc). High
resolution mass spectra were performed by Dr. Ron New at UC Riverside High
Resolution Mass Spectrometry Facility on a PE Biosystems DE-STR MALDI TOF
38
spectrometer with a WinNT (2000) Data System. All relevant spectral data and HPLC
chromatograms may be found in Appendix A.

Tert-butyl (oxiran-2-ylmethyl)carbamate (4): The epoxide 4 was synthesized as
previously described
44
. Briefly, freshly distilled allylamine (2.3 mL, 30 mmol, 1.0 eq) in
10 mL dry CH
2
Cl
2
was cooled in an ice bath (0 °C). To this cold solution, 6.54 g di-tert-
butyl dicarbonate (30.0 mmol, 1.00 eq) in 20 mL dry CH
2
Cl
2
was added. The solution
was brought to rt and stirred for 4 h. The reaction mixture was then diluted with
additional CH
2
Cl
2
and washed with 5% citric acid solution, followed by brine. The
organic layer was dried over Na
2
SO
4
and concentrated in vacuo, yielding 3.3 g (68%) of
the tert-butyl allylcarbamate. The
1
H NMR spectral data matched previously reported
values for this compound
44
.
1
H NMR (CDCl
3
): δ 1.38 (s, 9H), 3.68 (br, 2H), 4.53 (brs,
1H), 5.02-5.16 (m, 2H), 5.72-5.84 (m, 1H).
Tert-butyl allylcarbamate (1.0 g, 6.4 mmol, 1.0 eq) was dissolved in 50 mL dry CH
2
Cl
2
.
The solution was cooled to 0 °C and kept cold during addition of 2.8 g of 3-
chlorobenzenecarboperoxoic acid (MCPBA, commercially available as 77% pure), 12
mmol, 1.9 eq). The solution was then brought to rt and stirred overnight. The reaction
mixture was then diluted with additional CH
2
Cl
2
. The solution was washed with 10%
Na
2
SO
3
, followed by washing with saturated NaHCO
3
3x, and finally by washing with
water. The organic layer was dried over Na
2
SO
4
and concentrated in vacuo, yielding
crude epoxide 4. According to the
1
H NMR spectrum, the yield was 85%.
1
H NMR
(CDCl
3
): δ 1.44 (s, 9H), 2.59-2.78 (m, 2H), 3.04-3.54 (m, 3H), 4.75 (brs, 1H).

39
General Method for Preparation of N-alkylated, t-BOC protected Drug
Intermediates 5, 6, and 7. The parent drug 1, 2, or 3 (1 eq) was dissolved in water and
the pH adjusted to 5.7-6.2 with 1 N NaOH. Epoxide 4 (1-1.2 eq) was dissolved in
minimal methanol and added to the water solution, causing a slight precipitation to occur.
The precipitate disappeared on heating (40-50 °C) and as the reaction progresses. The
reaction was monitored by
31
P NMR or by analytical reverse-phase HPLC. After 90-95%
of the desired product was obtained (
31
P NMR), the solvent was removed in vacuo, and
the resulting white powder was washed with diethyl ether, filtered, and dried in a
dessicator. The product was then used without further purification.
1-(3-{[(tert-butoxy)carbonyl]amino}-2-hydroxypropyl)-3-(2-hydroxy-2,2diphosphon
oethyl)pyridinium (5): Synthesis performed according to the general method described
above with the monosodium salt of (1-hydroxy-2-pyridin-3-ylethane-1,1-
diyl)bis(phosphonic acid), 1 (287 mg, 0.94 mmol, 1.00 eq), in 4 mL water and 163 mg of
4 (0.94 mmol, 1.00 eq) in minimal MeOH. The reaction mixture was stirred at 40 °C for
18.5 h, yielding 90% of 5 by
31
P NMR.
1
H NMR (D
2
O): δ 1.27 (s, 9H), 3.07-3.30 (m,
4H), 3.95-4.03 (m, 1H), 4.18-4.27 (dd, J = 13.7 Hz, 3.7 Hz, 1H), 4.58-4.65 (part.
obscured by HDO, 1H), 7.75 (t, J = 6.8 Hz, 1H), 8.39 (d, J = 7.8 Hz, 1H), 8.43 (d, J = 6.5
Hz, 1H), 8.65 (s, 1H).
31
P{
1
H} NMR (D
2
O): δ 16.33 (d, J = 21.1 Hz, 1P), 16.55 (d, J =
21.1 Hz, 1P).
1-(3-{[(tert-butoxy)carbonyl]amino}-2-hydroxypropyl)-3-(2-carboxy-2-hydroxy-2-
phosphonoethyl)pyridinium (6): Synthesis performed according to the general method
described above with 2-hydroxy-2-phosphono-3-pyridin-3-ylpropanoic acid, 2, (0.3 g, 1.2
mmol, 1.0 eq), in 4 mL water and 0.23 g of 4 (1.33 mmol, 1.08 eq) in minimal MeOH.
40
The reaction mixture was stirred at 40 °C overnight. Then, an additional 46 mg of 4 (0.3
mmol, 0.3 eq) was added, and the reaction mixture was again stirred at 40 °C overnight,
yielding 90% of 6 (
31
P NMR). The product is isolated as a diastereomeric mixture.
1
H
NMR (D
2
O): δ 1.30 (s, 9H), 3.02-3.26 (m, 3H), 3.48 (dd, J = 14 Hz, 3.8 Hz, 1H), 3.94-
4.04 (m, 1H), 4.22-4.31 (m, 1H), 4.60-4.67 (br, 1H), 7.81 (dd, J = 8.3 Hz, 6.3 Hz, 1H),
8.27-8.32 (m, 1H), 8.48-8.56 (m, 2H).
31
P{
1
H} NMR (D
2
O): δ 14.99 (brs, 2P).
1-(3-{[(tert-butoxy)carbonyl]amino}-2-hydroxypropyl)-3-(2,2-diphosphonoethyl)
pyridinium (7): Synthesis performed according to the general method described above
with (2-pyridin-3-ylethane-1,1-diyl)bis(phosphonic acid), 3, (38.0 mg, 0.14 mmol, 1.00
eq), in 1 mL water and 25.5 mg of 4 (0.15 mmol, 1.07 eq) in minimal MeOH. The
reaction mixture was stirred at 40 °C overnight. After 19 h, 20% (
31
P NMR) of starting
material 3 remained. Thus, an additional 5.30 mg (0.03 mmol, 0.21 eq) of 4 in MeOH
was added to the reaction mixture. After 42 h, 95% of the desired product was obtained
(
31
P NMR).
1
H NMR (D
2
O): δ 1.28 (s, 9H), 2.09 (tt, J = 20.9 Hz, 7.5 Hz, 1H), 3.07-3.26
(m, 4H), 3.95-4.03 (brm, 1H), 4.23 (dd, J = 13.9 Hz, 9.9 Hz, 1H), 4.59-4.66 (m. 1H), 7.80
(dd, J = 8.1 Hz, 5.8 Hz, 1H), 8.38 (d, J = 7.0 Hz, 1H), 8.45 (d, J = 5.8 Hz, 1H), 8.66 (s,
1H).
31
P{
1
H} NMR (D
2
O): δ 17.23 (s, 2P).

General Method for Deprotection of Intermediates 5-7. Standard deprotection was
performed with 1:1 trifluoroacetic acid (TFA): H
2
O. After the reaction mixture was
stirred for 3-4 h at rt, the solvent was removed in vacuo. The resulting crystals were
washed with diethyl ether, filtered, and dried to yield the drug-linker conjugates 8-10 in
quantitative yields (by
1
H NMR), which were used without further purification.
41
1-(3-amino-2-hydroxypropyl)-3-(2-hydroxy-2,2-diphosphonoethyl)pyridinium (8):
1
H NMR (D
2
O): 3.01 (t, J = 11 Hz, 1H), 3.31 (d, J = 12.6 Hz, 1H), 3.39 (t, J = 11.6 Hz,
2H), 4.25-4.33 (m, 1H), 4.36-4.44 (br, 1H), 7.88 (t, J = 6.6 Hz, 1H), 8.48 (d, J = 8.0 Hz,
1H), 8.58 (d, J = 5.5 Hz, 1H), 8.77 (s, 1H).
31
P{
1
H} NMR (D
2
O): δ 16.04 (d, J = 27.5 Hz,
1P), 16.40 (d, J = 27.5 Hz, 1P).
1-(3-amino-2-hydroxypropyl)-3-(2-carboxy-2-hydroxy-2-phosphonoethyl)
pyridinium (9): (isolated as diastereomeric mixture)
1
H NMR (D
2
O): δ 2.92 (m, 1H),
3.20-3.30 (m, 2H), 3.48-3.56 (m, 1H), 4.17-4.25 (m, 1H), 4.34-4.42 (m, 1H), 4.71-4.76
(m, 1H), 7.89 (dd, J = 8.3 Hz, 6.0 Hz, 1H), 8.36-8.41 (brd, 1H), 8.58-8.62 (brd, 1H), 8.65
(s, 1H).
31
P{
1
H} NMR (D
2
O): δ 12.69 (s, 1P), 12.56 (s, 1P).
1-(3-amino-2-hydroxypropyl)-3-(2,2-diphosphonoethyl)pyridinium (10):
1
H NMR
(D
2
O): δ 2.18-2.36 (brt, 1H), 2.92 (brt, 1H), 3.14-3.28 (m, 3H), 4.16-4.24 (m, 1H), 4.60
(dd, J = 13.5 Hz, 9.5 Hz, 1H), 7.79 (dd, J = 8.7 Hz, 6.1 Hz, 1H), 8.37 (d, J = 8.2 Hz, 1H),
8.44 (d, J = 6.1 Hz, 1H), 8.64 (s, 1H).
31
P{
1
H} NMR (D
2
O): δ 17.54 (brm, 2P).

General Method of Preparation of Labeled Compounds 11-22. The following
synthesis and purification steps were performed under minimal lighting. 8, 9, or 10 (3-5
eq) is dissolved in 0.1 N NaHCO
3
. The pH was adjusted to 8.3 with solid Na
2
CO
3
. 5(6)-
carboxyfluorescein, N-hydroxysuccinimide ester (5(6)-FAM, SE, 21, 1 eq), Rhodamine
Red-X™, N-hydroxysuccinimide ester (RhR-X, SE, 22), or Alexa Fluor 647, N-
hydroxysuccinimide ester (AF7647, SE, 23) was dissolved in anhydrous DMF and
combined with the water solution. The pH may be re-adjusted to 8.9 with Na
2
CO
3
,
dissolving any precipitate, and the reaction mixture was stirred for 3 h to rt overnight.
42
Crude products 11-17 were purified by TLC on plates 20 x 20 cm (crude 11-17) or 7 x 20
cm (crude 18) eluted with 100% MeOH (crude 19 and 20 were not purified by TLC). The
phosphonate-containing compounds remaining at the origin were extracted with water;
the combined aqueous extracts may be treated with Chelex (sodium form) to aid the
extraction process. The solution was centrifuged, then concentrated in vacuo. The
resulting solids were then dissolved in either water, 20% MeOH in 0.1 N
triethylammonium acetate (TEAAc) or triethylammonium carbonate (TEAC) buffer (pH
7) and filtered through Nanosep 30K Omega filters. The solution was then purified by
semi-preparative reverse-phase HPLC according to the appropriate method. The final
amount of labeled product was calculated from the UV absorption spectrum, and the
isolated eluate was lyophilized. HPLC conditions:
Method A: Dynamax C18 (21.4 mm x 25 cm, 5 µm, 100 Å pore size) column, flow rate
8.0 mL/min, UV detection at 260 nm, gradient as follows: linearly increasing from 10%
MeOH in 0.1 N TEAAc or TEAC (pH 7) to 40% of 75% MeOH in 0.1 N TEAAc or
TEAC (pH 7) in 12 min, increasing to 70% of Buffer B in 100 min;
Method B: Similar to Method A except isocratic elution with 20% MeOH in 0.1 N TEAC
(pH 7) for 12 min, linearly increasing to 100% of 70% MeOH in 0.1 N TEAC (pH 7) in
22 min;
Method C: Beckman Ultrasphere ODS C18 (250 x 10 mm, 5 µm, 80 Å pore size), flow
rate 3.0 mL/min, UV detection at 260 nm, isocratic elution of 20% MeOH in 0.1 N
TEAC (pH 7.5) for 4 min, linearly increasing to 100% of 70% MeOH in 0.1 N TEAC
(pH 7.5) in 8 min;
43
Method D: Similar to Method C except flow rate 4.0 mL/min, UV detection at 260 nm
and 568 nm, isocratic elution of 20 % MeOH in 0.1 N TEAAc (pH 5) for 5 min, linearly
increasing to 40% of 70% MeOH in 0.1 N TEAAc (pH 5) in 25 min;
Method E: Similar to Method A except isocratic elution with 20% MeOH 0.1 N TEAAc
(pH 7) for 15 min, linearly increasing to 100% of 45% MeOH 0.1 N TEAAc (pH 7) in 25
min.
5(6)-FAMRIS (5-FAMRIS: 1-(3-{[3-carboxy-4-(6-hydroxy-3-oxo-3H-xanthen-9-
yl)benzoyl]amino}-2-hydroxypropyl)-3-(2-hydroxy-2,2-diphosphonoethyl)
pyridinium; 6-FAMRIS: 1-(3-{[4-carboxy-3-(6-hydroxy-3-oxo-3H-xanthen-9-
yl)benzoyl]amino}-2-hydroxypropyl)-3-(2-hydroxy-2,2-diphosphonoethyl)
pyridinium, 11): Synthesized according to the method above with 177 mg of 8 (0.36
mmol, 4.50 eq) in 2 mL of 0.1 N NaHCO
3
and 41.0 mg of 21 (0.08 mmol, 1.00 eq) in 200
µL anhydrous DMF; the reaction mixture was then stirred at rt for 3 h. After TLC
purification, the product was purified by HPLC according to Method A with TEAAc
buffers. Peaks eluting from 25-75 min were collected. During evaporation of the buffer
solution, product precipitated from the solution. Consequently, a second HPLC
purification was performed according to Method A but eluting with TEAC buffers (pH
7.5). Obtained 26 mg, 42% yield (triethylammonium bicarbonate salt);
1
H NMR (D
2
O): δ
3.33-3.39 (m, 2H), 3.43-3.66 (m, 2H), 4.08-4.42 (m, 3H), 4.72-4.80 (brd, 1H), 6.36-6.48
(m, 4H), 6.93 (d, 2H), 7.09 (s, 1H), 7.43 (s, 0.4 H), 7.68-7.87 (m, 2H), 8.06 (s, 0.6 H),
8.37-8.56 (m, 2H), 8.69-8.78 (2 s, 1H).
31
P{
1
H} NMR (D
2
O): δ 16.01 (brm, 2P).
HPLC Separation of 5- and 6-FAMRIS (12 and 13): Synthesized according to method
described for 11. Under HPLC conditions described as Method A, 6-FAMRIS and 5-
44
FAMRIS elute at very different retention times, 27 and 44 min, respectively. Each isomer
was collected separately and then concentrated in vacuo to remove buffer. Compound 12
was also directly synthesized from 5-FAM, SE according to the method described above.
Compound 12 samples obtained from 5(6)-FAM, SE and from 5-FAM, SE were
combined. Detailed NMR descriptions given below correspond to the HPLC-separated
products. 12 (triethylammonium acetate salt): obtained 11.9 mg, 8% yield;
1
H NMR
(D
2
O): δ 3.37 (brt, 2H), 3.57 (dd, J = 14.1 Hz, 6.7 Hz, 1H), 3.61-3.69 (m, 1H), 4.22-4.31
(m, 1H), 4.36-4.45 (m, 1H), 4.79 (d, 1H), 6.61-6.71 (m, 4H), 7.11 (d, J = 9.2 Hz, 2H),
7.33 (d, J = 8.0 Hz, 1H), 7.56 (s, 0.14H), 7.83 (t, J = 7.2 Hz,1H), 7.91 (d, J = 8.1 Hz, 1H),
8.15 (s, 1H), 8.44 (d, J = 8.3 Hz, 1H), 8.56 (d, J = 5.9 Hz, 1H), 8.74 (s, 1H).
31
P{
1
H}
NMR (D
2
O): δ 16.47 (brs). HRMS (positive ion MALDI): calcd 715.1089 m/z; found
[M]
+
= 715.1055 m/z; MS (negative ion MALDI): found [M-H]- = 713 m/z.
13 (triethylammonium acetate salt): 5.5 mg, 4% yield;
1
H NMR (D
2
O): δ 3.27-3.37 (m,
2H), 3.44 (dd, J = 14 Hz, 6.9 Hz,1H), 3.58 (dd, J = 14 Hz, 4.9 Hz, 1H), 4.19 (brs, 1H),
4.35 (dd, J = 14 Hz, 9.3 Hz, 1H), 6.60-6.73 (m, 4H), 7.04 (d, J = 9.4 Hz, 2H), 7.53 (s,
1H), 7.81 (dd, J = 8.2 Hz, 6.4 Hz, 1H), 7.88 (d, J = 8.1 Hz, 1H), 7.95 (dd, J = 8.0 Hz, 1.6
Hz, 1H), 8.43 (d, J = 8.1 Hz, 1H), 8.53 (d, J = 6.5 Hz, 1H), 8.70 (s, 1H).
31
P{
1
H} NMR
(D
2
O): δ 16.51 (brs). HRMS (positive ion MALDI): calcd 715.1089 m/z; found [M]
+
=
715.1082 m/z.
5(6)-FAMRISPC (5-FAMRISPC: 1-(3-{[3-carboxy-4-(6-hydroxy-3- oxo-3H-
xanthen-9-yl)benzoyl]amino}-2-hydroxypropyl)-3-(2-carboxy-2-hydroxy-2-
phosphonoethyl)pyridinium; 6-FAMRISPC: 1-(3-{[4-carboxy-3-(6-hydroxy-3-oxo-
3H-xanthen-9-yl)benzoyl]amino}-2-hydroxypropyl)-3-(2-carboxy-2-hydroxy-2-
45
phosphonoethyl)pyridinium, 14): Synthesized according to method described above
with 158 mg of intermediate 9 (0.36 mmol, 3.60 eq) in 1 mL of water and 53 mg of 21
(0.1 mmol, 1.0 eq) in 200 µL anhydrous DMF. After TLC purification, the mixture was
purified by HPLC according to Method B. Peaks eluting at 25-40 min were collected
together as 14. Obtained 9.6 mg, 13% yield (triethylammonium bicarbonate salt).
1
H
NMR (D
2
O): δ 3.27-3.62 (m, 3H), 4.03-4.40 (m, 2H), 6.42 (m, 4H), 6.92 (dd, J = 9.5 Hz,
3.5 Hz, 2H), 7.10 (d, J = 8.1 Hz, 1H), 7.42 (s, 0.4H), 7.70-7.86 (m, 2H), 8.06 (s, 0.6H),
8.27 (brs, 1H), 8.46-8.63 (m, 2H).
31
P{
1
H} NMR (D
2
O): δ 15.28 (brs, 1P). HRMS
(positive ion MALDI): calcd 679.1324 m/z; found [M]
+
= 679.1321 m/z.
5-FAMRISPC (15) and 6-FAMRISPC (16): Single isomers of FAMRISPC were
synthesized directly from isomerically pure starting materials (5-FAM, SE or 6-FAM,
SE) according to the method above. Synthesized in collaboration with Dr. X. Chen. 15
(retention time 28 min, triethylammonium bicarbonate salt, 2 mg, 29% yield):
1
H NMR
(D
2
O): δ 3.39-3.52 (m, 2H), 3.55-3.63 (m, 1H), 4.14-4.23 (m, 1H), 4.30-4.40 (m, 1H),
6.43-6.60 (m, 4H), 7.04 (d, J = 9.0 Hz, 2H), 7.25 (d, J = 8.2 Hz, 1H), 7.77-7.86 (m, 2H),
8.08 (s, 1H), 8.23-8.33 (brs, 1H), 8.45-8.62 (m, 2H). HRMS (positive ion MALDI): calcd
679.1324 m/z; found [M]
+
= 679.1356 m/z.
16 (retention time 25 min, triethylammonium bicarbonate salt, 2.5 mg, 36% yield):
1
H
NMR (D
2
O): δ 3.31-3.45 (m, 2H), 3.47-3.55 (m, 1H), 4.05-4.15 (m, 1H), 4.24-4.34 (m,
1H), 6.45-6.59 (m, 4H), 7.00 (d, J = 9.0 Hz, 2H), 7.48 (s, 1H), 7.71-7.80 (m, 2H), 7.85
(dd, J = 8.1 Hz, 1.7 Hz, 1H), 8.21-8.27 (m, 1H), 8.47- 8.55 (m, 2H). HRMS (positive ion
MALDI): calcd 679.1324 m/z; found [M]
+
= 679.1321 m/z.
46
5(6)-FAMdRIS (5-FAMdRIS: 1-(3-{[3-carboxy-4-(6-hydroxy-3-oxo- 3H-xanthen-9-
yl)benzoyl]amino}-2-hydroxypropyl)-3-(2,2-diphosphonoethyl)pyridinium; 6-
FAMdRIS: 1-(3-{[4-carboxy-3-(6-hydroxy-3-oxo-3H-xanthen-9-yl)benzoyl]amino}-
2-hydroxypropyl)-3-(2,2-diphosphonoethyl)pyridinium, 17): Synthesized according to
method described above with 53 mg of intermediate 10 (0.1 mmol, 2.5 eq) in 0.5 mL
HPLC water and 0.5 mL of 0.1 N NaHCO3 and 18.0 mg of 21 (0.04 mmol, 1.00 eq) in
100 µL anhydrous DMF. After TLC purification, the mixture was purified according to
Method B. Peaks eluting from 27-45 min were collected as 17. Obtained 9.4 mg, 36%
yield (triethylammonium acetate salt).
1
H NMR (D
2
O): δ 2.07- 2.27 (m, 1H), 3.10-3.27
(m, 2H), 3.35 (dd, J = 14.2 Hz, 6.8 Hz, 0.4H), 3.44-3.53 (m, 1H), 3.60 (dd, J = 14.2 Hz,
4.7 Hz, 0.6H), 4.08-4.16 (m, 0.4H), 4.18-4.25 (m, 0.6H), 4.25-4.40 (m, 1H), 4.71-4.77
(m, 1H), 6.40-6.52 (m, 4H), 6.96 (dd, J = 8.1 Hz, 4.1 Hz, 2H), 7.17 (d, J = 8.1 Hz, 0.6H),
7.44 (s, 0.4H), 7.72 (d, J = 8.1 Hz, 0.4H), 7.76-7.86 (m, 2H), 8.07 (d, J = 1.7 Hz, 0.6 H),
8.38 (t, J = 8.7 Hz, 1H), 8.47 (d, J = 5.9 Hz, 1H), 8.53 (d, J = 6.1 Hz, 1H), 8.67-8.71 (2 s,
1H).
31
P{
1
H} NMR (D
2
O): δ 16.51 (brs). HRMS (positive ion MALDI): calcd 699.1139
m/z; found [M]
+
= 699.1137 m/z.
RhR-X-RisPC (3-(2-carboxy-2-hydroxy-2-phosphonoethyl)-1-{3-[6-({4[6(diethyl
amino)-3-(diethylimino)-3H-xanthen-9-yl]-3-sulfobenzene}sulfonamido)hexanamido
]-2-hydroxypropyl}pyridinium, 18): Synthesized according to method described above
with 10.9 mg of compound 9 (0.04 mmol, 3 eq) in 0.5 mL of 0.1 N NaHCO
3
and 5 mg of
22 in 500 µL DMF. After TLC purification, the solution was then purified by HPLC
according to Method C. Peak eluting at 13 min is collected as 18. Obtained 2.1 mg, 33%
yield (triethylammonium bicarbonate salt).
1
H NMR (D
2
O): δ 1.11-1.25 (m, 20H
47
including TEA salt), 1.29-1.39 (m, 2H), 1.41-1.50 (m, 2H), 2.16 (t, J = 7.6 Hz, 2H), 2.97-
3.05 (m, 3H), 3.09-3.25 (m, 5H including TEA), 3.27-3.36 (m, 1H), 3.43-3.57 (m, 8H),
3.98-4.06 (m, 1H), 4.16-4.27 (m, 1H), 4.55-4.62 (m, 1H), 6.69 (s, 2H), 6.74-6.88 (m,
4H), 7.47 (d, J = 6.8 Hz, 1H), 7.81 (t, J = 6.8 Hz, 1H), 8.15 (d, J = 7.9 Hz, 1H), 8.32-
8.37 (m, 1H), 8.45-8.53 (m, 2H), 8.55-8.60 (m, 1H).
31
P{
1
H} NMR (D
2
O): 15.33 (s).
HRMS (positive ion MALDI): calcd 975.3148 m/z, found [M-H]
+
= 974.3118 m/z.
AF647RIS (2-(5-{3-[5-({2-hydroxy-3-[3-(2-hydroxy-2,2-diphosphonoethyl)pyridin-1-
ium-1-yl]propyl}carbamoyl)pentyl]-3-methyl-5-sulfo-1-(3-sulfopropyl)-2,3-dihydro-
1H-indol-2-ylidene}penta-1,3-dien-1-yl)-3,3-dimethyl-5-sulfo-1-(3-sulfopropyl)-3H-
indol-1-ium, 19): 25 mg of 8 was dissolved in 1 mL of H
2
O, and pH was adjusted to 8.3
with Na
2
CO
3
. 200 µL of this aqueous solution (containing 5 mg (10 µmol, 10 eq) of 8)
was added to 1 mg (assuming molecular weight of 957.09 g/mol, 1.0 µmol, 1 eq) of 23 in
50 µL anhydrous DMF, and the solution was stirred overnight. The solvent was
concentrated under vacuum, and the resulting blue residue was dissolved in 20% MeOH
in 0.1 N TEAAc buffer (pH 5) and purified by HPLC (Method D). Peaks eluting at 17
min were collected as 19 and then lyophilized. Obtained 0.5 mg, 40% yield
(triethylammonium acetate salt). MS (positive ion ESI-MS): calcd 1198.2410 m/z, found
[M-H]
+
= 1197.10 m/z.
AF647RISPC (2-(5-{3-[5-({3-[3-(2-carboxy-2-hydroxy-2-phosphonoethyl)pyridin-1-
ium-1-yl]-2-hydroxypropyl}carbamoyl)pentyl]-3-methyl-5-sulfo-1-(3-sulfopropyl)-
2,3-dihydro-1H-indol-2-ylidene}penta-1,3-dien-1-yl)-3,3-dimethyl-5-sulfo-1-(3-
sulfopropyl)-3H-indol-1-ium, 20): Synthesized according to method described above
with 2.3 mg of 9 (5 µmol, 5 eq) in 500 µL of H
2
O and 1 mg (assuming molecular weight
48
of 957.09 g/mol, 1.0 µmol, 1 eq) of 23 in 100 µL anhydrous DMF; reaction mixture was
stirred overnight. The solution was then purified by HPLC (Method E). Peaks eluting at
~33 min were collected as 20. Final yield was not determined. MS (positive ion ESI-MS):
calcd 1162.2645 m/z, found [M-H]
+
= 1162.1 m/z.

Modified purification of FL-ALN (as a mixture of 5- and 6- isomers, 5-FL-ALN: 2-
(6-hydroxy-3-oxo-3H-xanthen-9-yl)-5-[(4-hydroxy-4,4-diphosphonobutyl)carbamoyl
]benzoic acid; 6-FL-ALN: 2-(6-hydroxy-3-oxo-3H-xanthen-9-yl)-4-[(4-hydroxy-4,4-
diphosphonobutyl)carbamoyl]benzoic acid). The following was performed under
minimal lighting. A solution of 6 mg of 21 (0.0123 mmol, 1 eq) in 120 µL DMF was
added to 31.9 mg of alendronate (0.13 mmol, 10 eq) in 1 mL of 0.2 M NaHCO
3
(pH 8.3),
and pH re-adjusted to 8.5 with NaHCO
3
. The reaction mixture was stirred in darkness for
2 h and then purified by TLC eluted with 100% MeOH. The band remaining at the origin
was extracted with H
2
O, and the solution concentrated under vacuum. The sample was
then purified by semi-preparative HPLC: Dynamax C18 (21.4 mm x 25 cm, 5 µm, 100 Å
pore size) column, flow rate 4.0 mL/min, UV detection at 490 nm, gradient as follows:
linearly increasing from 5% CH
3
CN in 0.1 N TEAAc (pH 6) to 20% of 50% CH
3
CN in
0.1 N TEAAc (pH 6) in 20 sec, linearly increasing to 80% of 50% CH
3
CN in 0.1 N
TEAAc (pH 6) in 6:40 min, then linearly increasing to 100% of 50% CH
3
CN in 0.1 N
TEAAc (pH 6) by 15 min. Peaks eluting at 4.4 and 4.8 min (corresponding to two
49
isomers) were collected as product, and concentrated under vacuum. Final yield was not
determined.

Photostability studies of 5(6)-FAM. A 13 µM solution of 5(6)-FAM (in a buffer
containing 5 mM sodium acetate and 5 mM KH
2
PO
4
) was divided into 3 test tubes. Each
sample was placed in one of the following light levels: direct sunlight, indoor lighting of
a typical laboratory, and a completely darkened laboratory. After 30 minutes, UV
absorption spectrum was taken of each sample.

Stability studies of compound 11. A 72 µM solution of 11 in 0.1 N phosphate buffer
(pH 7.2) was frozen for 1 week, then kept at room temperature for 6 days (continously in
darkness). As a standard for comparison, a 1.3 mM solution of 5(6)-FAM in 0.1 N
phosphate buffer (pH 7.2) was used. The solutions were analyzed by analytical TLC
(eluted with 100% MeOH), and the developed plates were visualized under a UV lamp at
354 nm. While 5(6)-FAM always exhibited an R
f
= 1, the solution containing 11
remained at the baseline, indicating no release of hydrolyzed label and thus no
decomposition. Two other solutions of 11 were also studied as described above: 200 µL
of the above solution of 11 was diluted to 1.0 mL with 50 mM TRIS buffer (pH 7.7), and
an additional 200 µL was diluted to 1.0 mL with 50 mM HEPES buffer (pH 7). Similarly,
200 µL of the above 5(6)-FAM solution was diluted to 1.0 mL with 50 mM TRIS buffer
(pH 7.7), and an additional 200 µL was diluted to 1.0 mL with 50 mM HEPES buffer (pH
7). After 1 day in both the TRIS and HEPES buffers at room temperature and in darkness,
no decomposition of 11 was observed by TLC.
50
UV absorption and fluorescence emission spectra of labeled compounds 11-22. All
labeled samples were dissolved in water and diluted with 0.1 N phosphate buffer.
Assuming that the labeled bisphosphonates have the same ε as the free label, the final
concentrations for all labeled products are calculated from UV absorption spectra at λ =
493 nm (ε = 73000 M
-1
cm
-1
at pH 7.2) for FAM conjugates 11-17, λ = 567 nm (ε =
114850 M
-1
cm
-1
at pH 7.2) for RhR-X conjugate 18, or λ = 648 nm (ε = 240000 M
-1
cm
-1

at pH 7.2) for AF647 conjugates 19 and 20
34, 41-43
.
Emission spectra for FAM and RhR-X conjugates were recorded using an
excitation wavelength of 490 nm or 520 nm, respectively, with the excitation and
emission slits set at a 1 nm spectral bandwidth; integration time and increment were set to
0.5 s and 1 nm, respectively
34, 41, 42
. Emission spectra for AF647 conjugates were
recorded using an excitation wavelength of 600 nm; excitation and emission slits were set
at 1 nm and 3 nm, respectively, and integration time and increment were set to 1 s and 1
nm
37
.
51
Chapter 2 References
1. Russell, R. G. G.; Watts, N. B.; Ebetino, F. H.; Rogers, M. J., Mechanisms of
Action of Bisphosphonates: Similarities and Differences and Their Potential Influence on
Clinical Efficacy. Osteoporosis International 2008, 19, (6), 733-759.
2. Roelofs, A. J.; Thompson, K.; Gordon, S.; Rogers, M. J., Molecular Mechanisms
of Action of Bisphosphonates: Current Status. Clinical Cancer Research 2006, 12, (20,
Pt. 2), 6222s-6230s.
3. Coxon, F. P.; Helfrich, M. H.; Larijani, B.; Muzylak, M.; Dunford, J. E.;
Marshall, D.; McKinnon, A. D.; Nesbitt, S. A.; Horton, M. A.; Seabra, M. C.; Ebetino, F.
H.; Rogers, M. J., Identification of a Novel Phosphonocarboxylate Inhibitor of Rab
Geranylgeranyl Transferase that Specifically Prevents Rab Prenylation in Osteoclasts and
Macrophages. Journal of Biological Chemistry 2001, 276, (51), 48213-48222.
4. Coxon, F. P.; Ebetino, F. H.; Mules, E. H.; Seabra, M. C.; McKenna, C. E.;
Rogers, M. J., Phosphonocarboxylate Inhibitors of Rab Geranylgeranyl Transferase
Disrupt the Prenylation and Membrane Localization of Rab Proteins in Osteoclasts In
Vitro and In Vivo. Bone 2005, 37, (3), 349-358.
5. Marma, M. S.; Xia, Z.; Stewart, C.; Coxon, F.; Dunford, J. E.; Baron, R.;
Kashemirov, B. A.; Ebetino, F. H.; Triffitt, J. T.; Russell, R. G. G.; McKenna, C. E.,
Synthesis and Biological Evaluation of α-Halogenated Bisphosphonate and
Phosphonocarboxylate Analogues of Risedronate. Journal of Medicinal Chemistry 2007,
50, (24), 5967-5975.
6. Lueftner, D.; Henschke, P.; Possinger, K., Clinical Value of Bisphosphonates in
Cancer Therapy. Anticancer Research 2007, 27, (4A), 1759-1768.
7. Fournier, P. G. J.; Daubine, F.; Lundy, M. W.; Rogers, M. J.; Ebetino, F. H.;
Clezardin, P., Lowering Bone Mineral Affinity of Bisphosphonates as a Therapeutic
Strategy to Optimize Skeletal Tumor Growth Inhibition In vivo. Cancer Research 2008,
68, (21), 8945-8953.
8. Coxon, F. P.; Thompson, K.; Roelofs Anke, J.; Ebetino, F. H.; Rogers Michael, J.,
Visualizing Mineral Binding and Uptake of Bisphosphonate by Osteoclasts and Non-
resorbing Cells. Bone 2008, 42, (5), 848-860.
9. Woo, S.-B.; Hellstein, J. W.; Kalmar, J. R., Narrative Review: Bisphosphonates
and Osteonecrosis of the Jaws. Annals of Internal Medicine 2006, 144, (10), 753-761.
10. Krueger, C. D.; West, P. M.; Sargent, M.; Lodolce, A. E.; Pickard, A. S.,
Bisphosphonate-induced Osteonecrosis of the Jaw. Annals of Pharmacotherapy 2007, 41,
(2), 276-284.
52
11. Sato, M.; Grasser, W.; Endo, N.; Akins, R.; Simmons, H.; Thompson, D. D.;
Golub, E.; Rodan, G. A., Bisphosphonate action. Alendronate Localization in Rat Bone
and Effects on Osteoclast Ultrastructure. Journal of Clinical Investigation 1991, 88, (6),
2095-2105.
12. Zhang, S.; Gangal, G.; Uludag, H., Magic Bullets for Bone Diseases: Progress in
Rational Design of Bone-seeking Medicinal Agents. Chemical Society Reviews 2007, 36,
(3), 507-531.
13. Kozloff, K. M.; Weissleder, R.; Mahmood, U., Noninvasive Optical Detection of
Bone Mineral. Journal of Bone and Mineral Research 2007, 22, (8), 1208-1216.
14. Waggoner, A., Fluorescent Labels for Proteomics and Genomics. Current
Opinion in Chemical Biology 2006, 10, (1), 62-66.
15. Gumbleton, M.; Stephens, D. J., Coming Out of the Dark: The Evolving Role of
Fluorescence Imaging in Drug Delivery Research. Advanced Drug Delivery Reviews
2004, 57, (1), 5-15.
16. Mayer-Kuckuk, P.; Boskey, A. L., Molecular Imaging Promotes Progress in
Orthopedic Research. Bone 2006, 39, (5), 965-977.
17. Walker, B.; Gray, J.; Burns, D. M.; Wang, Q.; Adrian, T. E.; Nichols, D. H.;
Murphy, R. F.; Nelson, J., Carboxyfluorescein and Biotin Neuromedin C Analogs:
Synthesis and Applications. Peptides 1995, 16, (2), 255-61.
18. Maalouf, M. A.; Wiemer, A. J.; Kuder, C. H.; Hohl, R. J.; Wiemer, D. F.,
Synthesis of Fluorescently Tagged Isoprenoid Bisphosphonates that Inhibit Protein
Geranylgeranylation. Bioorganic & Medicinal Chemistry 2007, 15, (5), 1959-1966.
19. Hirabayashi, H.; Sawamoto, T.; Fujisaki, J.; Tokunaga, Y.; Kimura, S.; Hata, T.,
Relationship Between Physicochemical and Osteotropic Properties of Bisphosphonic
Derivatives: Rational Design for Osteotropic Drug Delivery System (ODDS).
Pharmaceutical Research 2001, 18, (5), 646-651.
20. Fujisaki, J.; Tokunaga, Y.; Takahashi, T.; Hirose, T.; Shimojo, F.; Kagayama, A.;
Hata, T., Osteotropic Drug Delivery System (ODDS) Based on Bisphosphonic Prodrug.
I: Synthesis and In Vivo Characterization of Osteotropic Carboxyfluorescein. Journal of
Drug Targeting 1995, 3, (4), 273-82, 1 plate.
21. Bhushan, K. R.; Tanaka, E.; Frangioni, J. V., Synthesis of Conjugatable
Bisphosphonates for Molecular Imaging of Large Animals. Angewandte Chemie,
International Edition 2007, 46, (42), 7969-7971.
22. Zaheer, A.; Lenkinski, R. E.; Mahmood, A.; Jones, A. G.; Cantley, L. C.;
Frangioni, J. V., In Vivo Near-infrared Fluorescence Imaging of Osteoblastic Activity.
Nature Biotechnology 2001, 19, (12), 1148-1154.
53
23. Thompson, K.; Rogers, M. J.; Coxon, F. P.; Crockett, J. C., Cytosolic Entry of
Bisphosphonate Drugs Requires Acidification of Vesicles After Fluid-phase Endocytosis.
Molecular Pharmacology 2006, 69, (5), 1624-1632.
24. Martin, M. B.; Arnold, W.; Heath, H. T., III; Urbina, J. A.; Oldfield, E., Nitrogen-
containing Bisphosphonates as Carbocation Transition State Analogs for Isoprenoid
Biosynthesis. Biochemical and Biophysical Research Communications 1999, 263, (3),
754-758.
25. Kashemirov, B. A.; Bala, J. L. F.; Chen, X.; Ebetino, F. H.; Xia, Z.; Russell, R. G.
G.; Coxon, F. P.; Roelofs, A. J.; Rogers, M. J.; McKenna, C. E., Fluorescently Labeled
Risedronate and Related Analogues: "Magic Linker" Synthesis. Bioconjugate Chemistry
2008, 19, (12), 2308-2310.
26. McKenna, C. E.; Kashemirov, B. A.; Bala, J. L. F Synthesis of Drug Conjugates
via Reaction with Epoxide-containing Linkers. US Patent Application 20080312440,
December 18, 2008.

27. Slavik, J., Fluorescent Probes in Cellular and Molecular Biology. CRC Press,
Inc.: Boca Raton, FL, 1994.

28. Ebetino, F. H.; Kaas, S. M.; Francis, m. D.; Nelson, D. G. A.; Janusz, J. M.
Quaternary Nitrgen-containing Phosphonate Compounds, Pharmaceutical Compositions,
and Methods of Treating Abnormal Calcium and Phosphate Metabolism and Methods of
Treating and Preventing Dental Calculus and Plaque. US Patent 5391743, April 30, 1993.
29. Boisser, S.; Ferreras, M.; Peyruchand, O.; Magnetto, S.; Ebetino, F. H.;
Colombel, M.; Delmas, P.; Delaisse, J.-M.; Clezardin, P., Bisphosphonates Inhibit Breast
and Prostate Carcinoma Cell Invasion, an Early Event in the Formation of Bone
Metastases. Cancer Research 2000, 60, (11), 2949-2954.
30. Boisser, S.; Magnetto, S.; Frappart, L.; Cuzin, B.; Ebetino, F. H.; Delmas, P. D.;
Clezardin, P., Bisphosphonates Inhibit Prostate and Breast Carcinoma Cell Adhesion to
Unmineralized and Mineralized Bone Extracellular Matrices. Cancer Research 1997, 57,
(18), 3890-3894.
31. Chen, G.; Heim, A.; Riether, D.; Yee, D.; Milgrom, Y.; Gawinowicz Mary, A.;
Sames, D., Reactivity of Functional Groups on the Protein Surface: Development of
Epoxide probes for Protein Labeling. Journal of the American Chemical Society 2003,
125, (27), 8130-8133.
32. Chuiko, A. L.; Filonenko, L. P.; Borisevich, A. N.; Lozinskii, M. O., Synthesis
and Properties of Hydroxyaminoalkyl Esters of (Hydroxyethylidene)bis[phosphonic
acid]. Zhurnal Obshchei Khimii 1993, 63, (5), 1070-1074.

54
33. Dunford, J. E.; Thompson, K.; Coxon, F. P.; Luckman, S. P.; Hahn, F. M.;
Poulter, C. D.; Ebetino, F. H.; Rogers, M. J., Structure-activity Relationships for
Inhibition of Farnesyl Diphosphate Synthase In Vitro and Inhibition of Bone Resorption
In Vivo by Nitrogen-containing Bisphosphonates. Journal of Pharmacology and
Experimental Therapeutics 2001, 296, (2), 235-242.
34. Lefevre, C.; Kang, H. C.; Haugland, R. P.; Malekzadeh, N.; Arttamangkul, S.;
Haugland, R. P., Texas Red-X and Rhodamine Red-X, New Derivatives of
Sulforhodamine 101 and Lissamine Rhodamine B with Improved Labeling and
Fluorescence Properties. Bioconjugate Chemistry 1996, 7, (4), 482-489.
35. Terpetschnig, E. A.; Patsenker, L. D.; Tatarets, A. Luminescent compounds. . US
Patent Application 2003724580, November 28, 2003.

36. White, S. S.; Li, H.; Marsh, R. J.; Piper, J. D.; Leonczek, N. D.; Nicolaou, N.;
Bain, A. J.; Ying, L.; Klenerman, D., Characterization of a Single Molecule DNA Switch
in Free Solution. Journal of the American Chemical Society 2006, 128, (35), 11423-
11432.
37. Balaz, M.; Sundberg, M.; Persson, M.; Kvassman, J.; Mnsson, A., Effects of
Surface Adsorption on Catalytic Activity of Heavy Meromyosin Studied Using a
Fluorescent ATP Analogue. Biochemistry 2007, 46, (24), 7233-7251.
38. Leung, W.-Y.; Cheung, C.-Y.; Yue, S. Modified Carbocyanine Dyes and Their
Conjugates and Their Uses. WO2002026891, April 4, 2002.

39. Albizu, L.; Teppaz, G.; Seyer, R.; Bazin, H.; Ansanay, H.; Manning, M.;
Mouillac, B.; Durroux, T., Toward Efficient Drug Screening by Homogeneous Assays
Based on the Development of New Fluorescent Vasopressin and Oxytocin Receptor
Ligands. Journal of Medicinal Chemistry 2007, 50, (20), 4976-4985.
40. Durroux, T.; Peter, M.; Turcatti, G.; Chollet, A.; Balestre, M.-N.; Barberis, C.;
Seyer, R., Fluorescent Pseudo-Peptide Linear Vasopressin Antagonists: Design,
Synthesis, and Applications. Journal of Medicinal Chemistry 1999, 42, (7), 1312-1319.
41. Kutyavin, I. V.; Lokhov, S. G.; Afonina, I. A.; Dempcy, R.; Gall, A. A.; Gorn, V.
V.; Lukhtanov, E.; Metcalf, M.; Mills, A.; Reed, M. W.; Sanders, S.; Shishkina, I.;
Vermeulen, N. M. J., Reduced Aggregation and Improved Specificity of G-rich
Oligodeoxyribonucleotides Containing Pyrazolo[3,4-d]pyrimidine Guanine Bases.
Nucleic Acids Research 2002, 30, (22), 4952-4959.
42. Mende, I.; Hoffmann, P.; Wolf, A.; Lutterbuese, R.; Kopp, E.; Baeuerle, P. A.; de
Baey, A.; Kufer, P., Highly Efficient Antigen Targeting to M-DC8+ Dendritic Cells Via
FcgRIII/CD16-specific Antibody Conjugates. International Immunology 2005, 17, (5),
539-547.
55
43. Maurel, D.; Kniazeff, J.; Mathis, G.; Trinquet, E.; Pin, J.-P.; Ansanay, H., Cell
Surface Detection of Membrane Protein Interaction with Homogeneous Time-resolved
Fluorescence Resonance Energy Transfer Technology. Analytical Biochemistry 2004,
329, (2), 253-262.
44. Rocheblave, L.; Bihel, F.; De Michelis, C.; Priem, G.; Courcambeck, J.; Bonnet,
B.; Chermann, J.-C.; Kraus, J.-L., Synthesis and Antiviral Activity of New Anti-HIV
Amprenavir Bioisosteres. Journal of Medicinal Chemistry 2002, 45, (15), 3321-3324.

56
Chapter 3
Evaluation of Fluorescent Risedronate and Related Analogues as Imaging
Probes

Introduction
Nitrogen-containing bisphosphonates (N-BPs), such as risedronate (RIS), inhibit
farnesyl pyrophosphate synthase (FPPS), resulting in the accumulation of unprenylated
proteins and the disruption of the proper function of osteoclasts
1
. The presence of two
phosphonate groups enables N-BPs to have an exceptionally high affinity for Ca
2+
ions
found in hydroxyapatite (HAP), inorganic mineral found in bone. Thus, N-BPs rapidly
localize to areas of bone resorption within bone mineral following administration in vivo,
allowing the drugs to come into close contact with osteoclasts
1, 2
. The acidic pH in the
resorption lacuna protonates the phosphonate groups of the N-BPs, thus increasing their
dissociation from HAP; the drugs (perhaps as a complex with calcium and bone organic
matrix proteins) are then internalized by osteoclasts via fluid-phase endocytosis
2, 3
.
After their release from HAP, it is believed that N-BPs may not only be taken up
into osteoclasts but may also be “recycled” by reattaching to HAP on the bone surface
2-4
.
Even after patients stop treatment, small amounts of N-BPs can be detected in their
urinary excretion over several weeks to months, indicating that the drugs remain present
in the circulation and available for re-uptake into bone for extended periods
1
.
“Recycling” may be a reasonable explanation for the ability of the drugs to exert ongoing
57
pharmacological actions and the differences between compounds in their local actions
and duration of effects
1
.
Quantitative distribution of BPs to other sites within bone remains unclear
1
. If
mineral surfaces, such as the osteocyte canalicular network, are accessible to N-BPs
diffusing from the extracellular fluid, they may also be available for BP uptake.
Furthermore, the differences of microanatomical distribution among the various BPs is
also relatively unknown
1
. Higher affinity N-BPs appear to localize on the bone surface
while lower affinity drugs, such as their phosphonocarboxylate (PC) analogues, will
likely concentrate greater in the surrounding solution. Thus, higher affinity N-BPs,
compared to their lower affinity analogues, are more likely to return to bone surfaces
during re-attachment and have a lowered ability to diffuse through bone, including the
osteocyte network
1
.
N-BPs have been shown to affect other cell types in vivo, such as osteoblasts and
cells in the marrow space (including cancer cells that have metastasized in bone)
1, 5
.
Moreover, the ability of N-BPs to exert anti-tumor activity observed in animal models of
cancer may be attributed to their effects on tumor or endothelial cells
5
. Additionally, the
acute-phase response typically observed in patients receiving intravenous N-BP
administration may be a result of the BPs’ direct effects on circulating monocytes and
subsequent activation of peripheral blood Vγ9Vδ2 T cells
5
. However, the exact cell types
that internalize BP, and thus those that are directly affected, remain unclear
1, 5
.
Although [
3
H]alendronate ([
3
H]ALN) was previously detected in osteoclasts in an
animal model, the BP was not detected in other cells
6
. However, experimental challenges
involved in working with radioactive compounds may have limited detection sensitivity
7,
58
8
. In comparison, fluorescent and near-IR labels not only exhibit increased sensitivity but
also reduce radiation exposure and allow for serial visualization of skeletal events over an
extended time period and simultaneous multichannel imaging
8
. Thus, fluorescently
labeled analogues of aminoalkyl N-BPs have also been utilized for cellular uptake and
localization studies. The alendronate (ALN) analogues, Alexa Fluor 488-ALN (AF-ALN)
and fluorescein-ALN (FL-ALN), were previously shown to be taken up into osteoclasts
by fluid-phase endocytosis, demonstrated that the fluorescent N-BP binds to bone mineral
in vitro, and concluded that only osteoclasts efficiently internalize BP from the bone
surface during resorption
2, 3
. Both alendronate analogues, however, were unable to inhibit
protein prenylation
9
. Near-IR labeled analogues of pamidronate, (Pam78, Pam800, and
the commercially available Osteosense
TM
680 and Osteosense
TM
750) have been utilized as
optical markers of bone metabolism
8, 10, 11
. However, cellular uptake or inhibitory
properties of these Pam analogues in vivo has not been reported.
In this chapter, fluorescent and near-IR analogues of RIS are used to study
localization and cellular uptake, demonstrating their usefulness as imaging probes (Figure
3.1). The inhibitory activities and HAP affinity of the labeled compounds are also
discussed. The following exhibits previously reported work performed in the laboratories
of Drs. Rogers (University of Aberdeen, UK) and Russell (University of Oxford, UK)
and Procter & Gamble Pharmaceuticals (Cincinnati, OH).

59
Figure 3.1. Structures of labeled RIS compounds discussed in this chapter: (from left to right) 6-FAMRIS,
5-FAMRIS, 5(6)-FAMRISPC, and RhR-X-RISPC. For structure of AF647RIS and synthetic details of all
compounds, refer to chapter 2.

Results and Discussion
HAP affinity is a critical property to maintain in order for fluorescent or near-IR
BPs to be useful bone-targeting imaging probes. To determine if introducing the linker
moiety and/or fluorescent label to RIS adversely affected the fluorescent BP’s bone
affinity, a column packed with hydroxyapatite ceramic spheres (20 µm) was employed as
a model to compare the HAP affinity between the parent drug and its labeled analogues
12
.
5- and 6-FAMRIS exhibited a slight decrease in retention time (7 min) as compared to its
parent compound RIS (10 min) (Figure 3.2). Thus, these results indicate that the labeled
compounds maintain binding affinity to bone, although slightly less than RIS
9
. Clearly,
HAP affinity is mostly determined by the chelating capabilities of the phosphonate
groups, but perhaps the increased hydrophobicity of the labeled BPs due to the addition
N
P
O
OH
NH
O
O
CO
2
H
O HO
HO P OH
OH
O
OH
OH
N
P
O
OH
NH
O
CO
2
H
O HO
HO P OH
OH
O
OH
OH
O
N
P
O
OH
NH
HO OH
O
OH
OH
O
(CH
2
)
5
NH
SO
2
SO
3
H
O NEt
2
Et
2
N
N
P
O
OH
NH
O
O
CO
2
H
O HO
HO OH
O
OH
OH
6-FAMRIS 5-FAMRIS 5(6)-FAMRISPC
RhR-X-RISPC
60
of FAM and/or the linker group may also affect affinity. Regardless, these imaging
agents retain significant bone affinity and remain suitable for other biological
experiments, including cellular and skeletal distribution studies.
Figure 3.2. Comparison of retention times of RIS vs. 5- and 6-FAMRIS on hydroxyapatite column (HAP
ceramic spheres, 20 µm). BPs eluted by Fast Performance Liquid Chromatography in phosphate buffer (pH
7) with gradient between 1 mM and 1 M over 50 min (flow rate 2.0 mL/min), and detected by UV
absorbance
9
. Adapted from Ref. 9.

Mineral binding capabilities of FAM-labeled compounds can also be studied by
utilizing dentine discs (mineralized substrates). FAMRIS and FAMRISPC were shown to
intensely label the surface of dentine, indicating that both FAM-labeled compounds retain
affinity for physiological mineralized surfaces in vitro (Figure 3.3)
9, 13
. As depicted in
Figure 3.3A-B, after rabbit osteoclasts were seeded onto dentine discs precoated with
FAMRIS, the fluorescent BP labeled newly exposed surfaces of resorption pits beneath
actively resorbing osteoclasts
9
. A similar finding was also observed with FAMRISPC as
shown in Figure 3.3C-D, where green fluorescence can be observed within resorption
pits
13
. Importantly, these results demonstrate that both FAMRIS and FAMRISPC label
newly exposed mineral surfaces after being released from dentine during resorption, or in
other words, “recycle”. By contrast, fluorescent bone-labeling agents with low bone
61
affinity, but are able to bind to dentine, do not exhibit this “recycling” capability (data not
shown); thus, bone affinity appears to be crucial for this effect. Interestingly, considering
higher affinity analogues are believed to be more likely to re-attach to bone, FAMRISPC,
whose parent drug 3-PEHPC exhibits decreased bone affinity, is still able to “recycle”
1
.
Perhaps additional studies can further clarify the structural variations that lead to
differences in skeletal distribution and the relationship between affinity and “recycling”.
Regardless, these in vitro findings are similar to previous results observed with FL-ALN,
proving that “recycling” is a legitimate explanation for the long duration of action that
some potent N-BPs may exert, especially considering that resorbed areas do not become
depleted of BP even in the absence of repeated administration of the drug
1, 3
.
Figure 3.3. Binding and “recycling” of FAMRIS to surface of dentine; bars = 10 µm. Images analyzed by
laser scanning confocal microscopy
9, 13
. (A-B) Rabbit osteoclasts seeded onto dentine discs precoated with
FAMRIS (green), then immunostained for VNR (blue); actin rings of osteoclasts were visualized using
TRIC-phalloidin (red; a,b only). A) 1 m xy image at the surface of dentine; asterisks indicate resorption
pits; B) zx image of the same field-of-view. (C-D) After matrix proteins on dentine surface were labeled
with TAMRA, succinimidyl ester, dentine discs were labeled for 1 h with 1mM FAMRISPC. Discs were
then seeded with rabbit osteoclasts and cultured for 24 h. C) extended focus xy images of FAMRISPC
(green, left panel), TAMRA-labeled dentine (middle), and left and middle panels merged (right). D)
Corresponding zx sections of panels depicted in Fig. 3.3C. Adapted from Ref. 9 and 13.
62
In vitro studies using rabbit osteoclasts and J774 macrophages and analyzed by
laser scanning confocal microscopy were performed to study cellular uptake (Figure
3.4)
9, 13
. As shown in Figure 3.4A, FAMRIS can clearly be observed within intracellular
vesicles in resorbing rabbit osteoclasts
9
. In J774 macrophages, both FAMRIS and
FAMRISPC co-localized with dextran, a marker of fluid-phase endocytosis (Figure
3.4B)
13
. These results clearly demonstrate that FAMRIS and FAMRISPC are internalized
by fluid-phase endocytosis in resorbing osteoclasts, coinciding with results previously
observed with FL-ALN
3
.
Figure 3.4. Cellular uptake of FAMRIS and FAMRISPC proceeds by fluid-phase endocytosis; bars = 10
µM. Analyzed by laser scanning confocal microscopy
9, 13
. A) Uptake of FAMRIS by rabbit osteoclast in
vitro. Rabbit osteoclasts were seeded onto dentine discs that had been precoated with FAMRIS, then
immunostained for VNR (blue). Top panel depicts 1 µm xy image 8 µm above dentine surface; bottom
panel shows zx image of the same osteoclast with (z position of top panel shown by arrow, outline of
resorption pit); detector gain optimized for detection of FAMRIS intracellullarly resulting in saturated
signal from FAMRIS at dentine surface. B) J774 macrophages incubated with FAMRIS or FAMRISPC for
4 h, in the presence or absence of TRITC-dextran (40 kDa). Cells were then fixed; F-actin and nuclei
stained (1µM sections shown). Adapted from Ref. 9 and 13.

63
The ability to inhibit protein prenylation was either abolished or not reported for
previous fluorescent N-BPs. However, we found that FAMRIS exhibited similar activity
as RIS for inhibition of Rap1A prenylation in J774 cells (Figure 3.5A)
9
. This effect may
be reversed in the presence of GGPP, indicating that FAMRIS apparently shares the same
enzymatic target, FPPS, as RIS (Figure 3.5B)
9
. FAMRISPC, on the other hand, is unable
to inhibit the prenylation of Rab11, and thus does not exert inhibitory properties towards
its expected enzymatic target Rab GGTase (the target of its parent compound, 3-PEHPC)
(Figure 3.5C)
13
. Although 3-PEHPC is considered a weak inhibitor of Rab GGTase, its
fluorescent analogue would ideally maintain some sort of inhibition capability. Thus, it
appears that perhaps the fluorescent label and/or the added linker group may adversely
affect the inhibitory properties in some cases. Further studies on compounds such as PCs
may need to be performed with different imaging agents and/or linker moieties to
determine an optimal imaging probe that will maintain enzymatic activity.
64

Figure 3.5. Western blot assays for unprenylated Rap1A (uRap1A) or unprenylated Rab11 (uRab11)
9, 13
.
J774.2 cells were treated for 24 h with: A) 0 - 100 µM 6-FAMRIS or RIS; B) 20 µM 6-FAMRIS or 20 µM
RIS in the presence or absence of 100 µM GGPP; C) 67 - 600 µM 3-PEHPC or FAMRISPC. Adapted
from Ref. 9 and 13.

FAMRIS retains the ability to inhibit prenylation in rabbit osteoclasts in vivo and
bone resorption in the Schenk model (LED = 0.1 mg P/kg) (Figure 3.6)
13
. In isolated
enzyme assays, however, FAMRIS (IC
50
= 2690 ± 1400 nM) appeared to be a much less
potent inhibitor of FPPS versus RIS (IC
50
= 5.7 nM), indicating that these effects may not
be due to FPPS inhibition alone. Interestingly, we then identified that FAMRIS was also
able to inhibit geranylgeranyl diphosphate synthase (IC
50
= 2552.7 ± 764 nM), an effect
that may be contributing to the compound’s inhibitory properties of Rap1A prenylation
13,
14
. Therefore, these results suggest that FAMRIS inhibits both FPPS and GPPS with
similar potencies in isolated enzyme assays
13
. More importantly, FAMRIS is the first
example of a fluorescent BP that is able to maintain the ability to inhibit protein
prenylation
9, 13
. Unlike fluorescent alendronate and pamidronate analogues that altered
the pharmacologically important amino moiety by its conversion to a neutral amide
65
linkage, the attachment of the linker group to the pyridine nitrogen generates a
permanently charged quaternary nitrogen in FAMRIS, which may aid the fluorescent BP
to maintain its inhibitory activities. Furthermore, perhaps the addition of the linker and
fluorophore allows FAMRIS to structurally mimic other known GPPS inhibitors, which
often contain side chains with several ring systems
15
.

Figure 3.6. FAMRIS inhibits prenylation and bone resorption in vivo
13
. A) Newborn rabbits were injected
with 0.5 mg/kg 6-FAMRIS or vehicle, then 24 h later osteoclasts (VNR+) isolated from marrow cells
(VNR-) in the long bones by immunomagnetic bead separation, and unprenylated/prenylated Rap1A
detected by western blotting. B) Assessment of anti-resorptive potency of FAMRIS by Schenk growing rat
model. Adapted from Ref. 13.

AF647RIS has been utilized for in vivo cellular and skeletal distribution studies.
Cellular uptake was detected in rabbit osteoclasts using confocal microscopy, and similar
to FAMRIS, AF647RIS was easily detected in osteoclasts in vivo
16-18
. Although in vivo
flow cytometry experiments revealed limited uptake of FAMRIS by bone marrow cells
(BMCs), preliminary data indicated rabbit BMCs internalized AF647RIS in vivo,
including a subset that were CD14
+
(data not shown)
16-18
. However, this difference in
uptake between the labeled RIS analogues may be related to decreased detection
sensitivity with FAM versus the near-IR AF647 label
8
. Furthermore, ex vivo treatment of
66
BMCs indicated intracellular uptake of FAMRIS observable by confocal microscopy
17, 18
.
These results suggest that, although osteoclasts are the main cell type affected by BPs,
bone marrow monocytes may also be directly affected by the drugs
17, 18
.
Histological analysis of the ulnae and vertebrae indicated that FAMRIS
concentrated to bone surfaces in vivo, especially around the growth plate and
significantly around vascular channels within the bone matrix
16-18
. Osteocytic lacunae in
close proximity to these channels as well as the lacunae of newly embedded osteocytes
also exhibited FAMRIS binding (Figure 3.7)
16-18
. These findings indicate that, although
osteoclasts are the only cell type that internalize substantial quantities of BP in vivo, BPs
remain capable of exerting direct effects on osteocytes in vivo through extracellular
mechanisms even if the drugs may only interact with a small subset of osteocytes
16-18
.
Continuing this study with FAMRISPC to clarify the differences in distribution between
RIS and 3-PEHPC will be an interesting investigation since higher affinity N-BPs, unlike
their PC analogues, are believed to have a lowered ability to diffuse through bone,
including the osteocyte network
1
.

Figure 3.7. Presence of FAMRIS in osteocytic lacunae. 3-month old mouse injected with 1 mg/kg
FAMRIS (green), and confocal microscopy image taken of 2 µm longitudinal sections through a tibia after
24 h. Sections counterstained with TO-PRO-3 (nuclear dye); bar = 10 µM. Adapted from A. Roelofs, et. al.,
J. Bone Miner. Res., submitted.
67
Multicolor labeling experiments have been accessible due in part to our flexible
synthetic approach towards these fluorescent and near-IR RIS analogues, which allows us
to easily introduce a different fluorophore into a BP or PC drug. Thus, by attaching
various imaging agents to the BP or PC and creating a “matrix” of compounds, we may
investigate the effect of various bone affinity levels on skeletal distribution by
simultaneously visualizing two different imaging agents. Thus, rats were injected
subcutaneously with both FAMRIS and RhR-X-RISPC and sacrificed 1 or 7 days later
16
.
Although some differences in localization were detected after 1 day, FAMRIS had
additionally labeled bone surface that had formed after 7 days while RhR-X-RISPC did
not exhibit the same effect (data not shown)
16
. Thus, these results represent the first in
vivo demonstration of the concept of BP “recycling” to other bone surfaces, similar to
results obtained from in vitro studies
9, 16
. To ensure that these effects are not solely
influenced by the label rather than bone affinity, an interesting next step may be to
perform the same analyses with RhR-X-labeled RIS and FAM-labeled 3-PEHPC. If
influence from the label can be discounted and bone affinity is the only characteristic
responsible for these effects, these findings may strengthen the hypothesis that higher
affinity BPs are more likely return to bone surfaces than their lower affinity PC
counterparts.
Fluorescent and near-IR labeled RIS analogues are demonstrating their usefulness
for elucidating important properties of BPs. Near-IR imaging probes are particularly
advantageous due to their increased sensitivity in detection. However, the labeled RIS
analogues synthesized to date utilize FAM, RhR-X, and AF647, imaging agents that vary
greatly in their structure. It is currently not well understood how each fluorescent/near-IR
68
label may affect uptake or distribution. Therefore, ideal BP/PC imaging agents would be
more structurally similar with only slight modifications that would impart the different
spectral properties. This would eliminate or minimize observable differences that may
solely be attributed to the imaging agent rather than properties of the BP or PC itself, and
at the very least, remove the need for repeating multicolor labeling experiments with
alternate labels on the drugs.

Conclusion
Fluorescent and near-IR analogues of RIS are proving to be exceptional tools for
improving our understanding of cellular and skeletal distribution. Although the
introduction of FAM and/or the linker group has resulted in a slightly decreased retention
time on a HAP column and thus slightly lowered affinity to bone, FAMRIS represents the
first example of a fluorescent BP that exhibits inhibitory activities and has been shown to
inhibit both FPPS and GPPS
9, 13
. Similar to FL-ALN, FAMRIS and FAMRISPC bind to
dentine, a mineralized substrate, and are able to “recycle” onto newly exposed mineral
surfaces, a characteristic that appears to be directly related to bone affinity
9, 13
. In vivo
results also indicate that, although osteoclasts appear to be the only cells that internalize
large quantities of BP, other cells such as osteocytes may also be affected by the drug
16-
18
. Multicolor labeling experiments are also increasing our understanding on the
relationship between bone affinity and skeletal distribution
16
. Ongoing biological studies
with our labeled BP and PC compounds are continuing to garner important information to
improve our current understanding of BPs and PCs.

69
Experimental Methods
Funding provided by Procter & Gamble Pharmaceuticals. Data presented above
collected by Rogers (University of Aberdeen, UK) and Russell (University of Oxford,
UK) laboratories and Procter & Gamble Pharmaceuticals (Cincinnati, OH). For detailed
experimental methods, please refer to references 10, 11, and 14-16.

70
Chapter 3 References
1. Russell, R. G. G.; Watts, N. B.; Ebetino, F. H.; Rogers, M. J., Mechanisms of
Action of Bisphosphonates: Similarities and Differences and Their Potential Influence on
Clinical Efficacy. Osteoporosis International 2008, 19, (6), 733-759.
2. Thompson, K.; Rogers, M. J.; Coxon, F. P.; Crockett, J. C., Cytosolic Entry of
Bisphosphonate Drugs Requires Acidification of Vesicles After Fluid-phase Endocytosis.
Molecular Pharmacology 2006, 69, (5), 1624-1632.
3. Coxon, F. P.; Thompson, K.; Roelofs Anke, J.; Ebetino, F. H.; Rogers Michael, J.,
Visualizing Mineral Binding and Uptake of Bisphosphonate by Osteoclasts and Non-
resorbing Cells. Bone 2008, 42, (5), 848-860.
4. Nancollas, G. H.; Tang, R.; Phipps, R. J.; Henneman, Z.; Gulde, S.; Wu, W.;
Mangood, A.; Russell, R. G. G.; Ebetino, F. H., Novel Insights into Actions of
Bisphosphonates on Bone: Differences in Interactions with Hydroxyapatite. Bone 2006,
38, (5), 617-627.
5. Roelofs, A. J.; Thompson, K.; Gordon, S.; Rogers, M. J., Molecular Mechanisms
of Action of Bisphosphonates: Current Status. Clinical Cancer Research 2006, 12, (20,
Pt. 2), 6222s-6230s.
6. Sato, M.; Grasser, W.; Endo, N.; Akins, R.; Simmons, H.; Thompson, D. D.;
Golub, E.; Rodan, G. A., Bisphosphonate action. Alendronate Localization in Rat Bone
and Effects on Osteoclast Ultrastructure. Journal of Clinical Investigation 1991, 88, (6),
2095-2105.
7. Zhang, S.; Gangal, G.; Uludag, H., Magic Bullets for Bone Diseases: Progress in
Rational Design of Bone-seeking Medicinal Agents. Chemical Society Reviews 2007, 36,
(3), 507-531.
8. Kozloff, K. M.; Weissleder, R.; Mahmood, U., Noninvasive Optical Detection of
Bone Mineral. Journal of Bone and Mineral Research 2007, 22, (8), 1208-1216.
9. Kashemirov, B. A.; Bala, J. L. F.; Chen, X.; Ebetino, F. H.; Xia, Z.; Russell, R. G.
G.; Coxon, F. P.; Roelofs, A. J.; Rogers, M. J.; McKenna, C. E., Fluorescently Labeled
Risedronate and Related Analogues: "Magic Linker" Synthesis. Bioconjugate Chemistry
2008, 19, (12), 2308-2310.
10. Zaheer, A.; Lenkinski, R. E.; Mahmood, A.; Jones, A. G.; Cantley, L. C.;
Frangioni, J. V., In Vivo Near-infrared Fluorescence Imaging of Osteoblastic Activity.
Nature Biotechnology 2001, 19, (12), 1148-1154.
11. Bhushan, K. R.; Tanaka, E.; Frangioni, J. V., Synthesis of Conjugatable
Bisphosphonates for Molecular Imaging of Large Animals. Angewandte Chemie,
International Edition 2007, 46, (42), 7969-7971.
71
12. Marma, M. S.; Xia, Z.; Stewart, C.; Coxon, F.; Dunford, J. E.; Baron, R.;
Kashemirov, B. A.; Ebetino, F. H.; Triffitt, J. T.; Russell, R. G. G.; McKenna, C. E.,
Synthesis and Biological Evaluation of α-Halogenated Bisphosphonate and
Phosphonocarboxylate Analogues of Risedronate. Journal of Medicinal Chemistry 2007,
50, (24), 5967-5975.
13. Coxon, F. P.; Bala, J. L.; Kashemirov, B. A.; Lundy, M. W.; Chen, X.; Xia, Z.;
Dunford, J. E.; Russell, R. G. G.; Roelofs, A. J.; Rogers, M. J.; McKenna, C. E.; Ebetino,
F. H., Fluorescently Labeled Risedronate and Related Analogs: Design and Evaluation as
Imaging Probes. Bone 2008, 42, (S1), S36.
14. Kavanagh, K. L.; Guo, K.; Dunford, J. E.; Wu, X.; Knapp, S.; Ebetino, F. H.;
Rogers, M. J.; Russell, R. G. G.; Oppermann, U., The Molecular Mechanism of Nitrogen-
containing Bisphosphonates as Antiosteoporosis Drugs. Proceedings of the National
Academy of Sciences of the United States of America 2006, 103, (20), 7829-7834.
15. Guo, R.-T.; Cao, R.; Liang, P.-H.; Ko, T.-P.; Chang, T.-H.; Hudock, M. P.; Jeng,
W.-Y.; Chen, C. K.-M.; Zhang, Y.; Song, Y.; Kuo, C.-J.; Yin, F.; Oldfield, E.; Wang, A.
H.-J., Bisphosphonates Target Multiple Sites in Both Cis- and Trans- Prenyltransferases.
Proceedings of the National Academy of Sciences of the United States of America 2007,
104, (24), 10022-10027.
16. Roelofs, A. J.; Coxon, F. P.; Lundy, M. W.; Ebetino, F. H.; Bala, J. F.;
Kashemirov, B. A.; McKenna, C. E.; Rogers, M. J., Studying Cellular Uptake and
Distribution of Bisphosphonate In Vivo Using Fluorescently-labelled Analogues of
Risedronate. Journal of Bone and Mineral Research 2008, 23, S20.
17. Roelofs, A. J.; Coxon, F. P.; Ebetino, F. H.; Bala, J. F.; Kashemirov, B. A.;
McKenna, C. E.; Rogers, M. J., Visualisation of Cellular Uptake and Localisation of
Bisphosphonate In Vivo Using a Fluorescent Analogue of Risedronate. Calcified Tissue
International 2008, 82, (S1), S59.
18. Roelofs, A. J.; Coxon, F. P.; Ebetino, F. H.; Bala, J. L. F.; Kashemirov, B. A.;
McKenna, C. E.; Rogers, M. J., Use of a Fluorescent Analogue of Risedronate to Study
Localisation and Cellular Uptake of Bisphosphonates In Vivo. Bone 2008, 42, (S1), S85.

72
Chapter 4
Synthesis of a Novel Bisphosphonic Acid Alkene Monomer

Introduction
Direct resin-based filling composites and enamel-dentin adhesives are often used
in combination to form durable, aesthetic tooth-colored restorative materials, which are
crucial to modern dentistry
1
. Although filling materials have been widely used in the past
decade, their current formulations, fillers, and enamel and dentin bonding agents urgently
need to be improved to increase their clinical performance
1
. Specifically, self-etching
enamel-dentin adhesives, which are based on polymerizable, strongly acidic monomers,
are used to modify dental hard tissues (enamel and dentin) and achieve strong bonding
between a restorative composite and these tissues
1
. However, major limitations of these
adhesives include technique sensitivity and limited storage stability at ambient
temperature
1
. Ideal monomers must meet the following requirements
1
:
1. High rate of homo- or copolymerization with other monomers in the adhesive;
2. Solubility in adhesive composition (aqueous solutions of ethanol or acetone);
3. Stability of monomer and subsequently formed polymer, including hydrolytic
stability during storage in water;
4. Low water uptake and swelling degree of polymer;
5. Low oral toxicity and cytotoxicity.
Previously reported self-etching enamel-dentin adhesives are composed of
methacrylate (or similar functionalities, such as acrylate or vinyl groups) containing
phosphoric acids
2
. However, in a self-etching adhesive, water is typically employed as a
73
solvent. Under these conditions, methacrylate containing phosphoric acids, are subject to
hydrolysis, resulting in the alteration of the chemical composition of the adhesive and the
deterioration of its performance
3
.
The introduction of hydrolytically stable bonds between the polymerizable group
and the strongly acidic group are believed to improve stability
4, 5
. Thus, the substitution
of the phosphoryl groups with phosphonic acid moieties, generating polymerizable
phosphonic and/or bisphosphonic (BP) acid derivatives, appears to be a promising new
direction. For example, phosphonic acids connected to a polymerizable methacrylate
group via a hydrolytically stable ether bond exhibited an etch pattern similar to that
generated by commercial etching gels based on phosphoric acid
2, 4
. Additionally,
phosphonic acids with long alkyl chains connected to acrylamido groups were shown to
be resistant to hydrolysis and gave moderate to high yields upon polymerization
6
.
In this study, our goal is to design and create an improved, novel polymerizable
BP, ideally affording one or more of the following: increased acidity and hydrophobicity,
satisfactory polymerization, and acceptable or improved hydrolytic stability. We initially
began our studies on a known polymerizable BP, which had demonstrated good
polymerization properties but was not studied as an enamel-dentin adhesive
7
. However,
this compound was not sufficient for dental applications due to its insolubility in typical
solvents used for polymerization, such as 2-hydroxyethylmethacrylate (HEMA).
Therefore, we turned our focus towards the design of a novel polymerizable BP with the
above listed characteristics but enhanced hydrophobicity in order to increase its solubility
in HEMA.
74
A previous patent included certain polymerizable bisphosphonic acids as shown
in Figure 4.1A
5
. Although several different moieties are claimed at the R
2
position
(directly attached to the bridging carbon atom of a methylene BP), halogens were not
specified. Moreover, the method used to synthesize a number of examples detailed in the
patent will typically yield α–hydroxyl BPs. Therefore, R
2
, in these cases, does not
necessarily have any basis in design, but is present due to the synthetic chemistry used to
created the BP.
P
P
HO
OH HO
OH
O
O
R
2
A N
R
4
O
C
H
2
C
R
3
x
x = 1-3
R
2
= H, OH, alkyl, aryl, alkyoxy, aryloxy,
-A-(N(R
4
)-C(O)-C(R
3
)=CH
2
)
x
R
3
= H or CH
3
R
4
= H, alkyl, or joined to A forming cyclic
organic group
A = straight or branched alkyl group
P
P
HO
OH HO
OH
O
O
F (CH
2
)
x
H
N
O
C
H
H
2
C
P
P
HO
OH HO
OH
O
O
F (CH
2
)
6
H
N
O
C
H
H
2
C
A B
C
x = 6 or 8

Figure 4.1. Structures of polymerizable BPs. A) General structure of patented BP; B) General structure of
novel target BP; C) Novel target BP 1.

Fluorine-containing compounds are well-known for their increased lipophilicty
8
.
Additionally, the introduction of fluorine at the α-carbon of phosphonates as an
isoelectronic replacement for an α-hydroxyl group is widely known to affect the
metabolic degradation, hydrogen bonding, and reactivity of organic molecules
9
.
Moreover, due to the electron-withdrawing property of fluorine, the CHF group of an α-
monofluoro BP sterically and electronically mimics oxygen, increasing the acidity of the
BP
9
. Therefore, in order to achieve a novel BP that met these ideal requirements, an α-F
BP was designed with a long alkyl chain connected to a polymerizable moiety (Figure
75
4.1B). The halogen moiety serves two purposes: to increase the acidity of the compound
(an important property for dental etching) and the hydrophobicity of the compound.
Furthermore, the length of the alkyl chain and the linkage between the
polymerizable group and the alkyl chain can be additionally modified to achieve the ideal
bisphosphonic acid. For our initial studies, hexyl and octyl chains were chosen due to the
relatively inexpensive commercially available starting materials necessary for these
synthetic targets. Moreover, amide linkages were utilized due to their increased
hydrolytic stability as compared to their ester counterparts.
The following discussion in this chapter first presents an improved synthesis of a
pamidronic acid-based monomer and studies on its solubility in solvent systems typically
used for polymerizations for dental materials. Also, different strategies to achieve the
synthesis of an α–fluorinated, alklylamino bisphosphonic acid, a key synthon for
subsequent conjugation to an acryl group via an amide linkage, are discussed. Finally,
after several reaction conditions were attempted, the synthesis of [1-fluoro-1-phosphono-
7-(prop-2-enamido)heptyl]phosphonic acid, 1, was achieved (Figure 4.1C).

Results and Discussion
We initially studied a previously reported polymerizable analogue of pamidronate
(Pam), an alkylamino bisphosphonic acid, to identify if this compound was suitable as a
component for enamel-dentin adhesives. Pam monomers were first synthesized according
to published reaction conditions (Scheme 4.1)
7
. N-acryloxysuccinimide and Pam were
combined in equimolar amounts in H
2
O, and pH was then adjusted to 8 with NaOH.
After the reaction mixture was stirred at room temperature overnight, the product 2 was
76
precipitated with ethanol. However, the following problems with this reaction were
identified:
1. N-acryloxysuccinimide is not readily soluble in H
2
O;
2. The pH drops during the reaction, protonating the primary amine of Pam and
thus slowing the progression of our desired reaction;
3. N-hydroxysuccinimde (NHS), a side product of this reaction, precipitates with
the product in ethanol if the pH is above neutral, thereby contaminating the
product;
4. The desired product 2 is slightly soluble in ethanol; thus, the published
precipitation method may entail product loss.
Scheme 4.1. Synthesis of Pam monomer (2). Previously reported conditions: H
2
O, pH 8, rt. Improved
conditions: NaHCO
3
/DMF, pH 8, rt.

This synthesis was improved by the following adjustments. First, N-
acryloxysuccinimide was dissolved in N,N-dimethylformamide (DMF) prior to addition
to Pam to ensure the compound will be in solution during the reaction, while NaHCO
3

buffer was also used to maintain a stable pH. The reaction can also followed by
31
P NMR
so that additional acryloxysuccinimide may be added during the reaction to ensure its
completion; thus, purification from Pam will not be a factor. For a more efficient
purification method, all products (Pam monomer 2, NHS, and acrylic acid) were first
H
2
N P
HO P OH
O OH
OH
OH O
O
O
O
O
a
H
N P
HO P OH
O OH
OH
OH O
O
+
2
77
converted to their acidic forms by eluting the reaction mixture through DOWEX (H
+

form), and 2 was then precipitated from MeOH and acetone. This method yielded the
purified acidic form of 2, which was isolated as a white crystalline solid.
Then, we studied the solubility of 2, as white crystals, under a variety of
conditions that are typically employed for the polymerization process used for these types
of monomers. First, compound 2 was added to HEMA, forming a 20% mixture of 2 in
solution and stirred at rt for one day. However, 2 was not soluble under these conditions.
Thus, to study the solubility of 2 in a mixture of 50:50 HEMA:H
2
O, 23% and 3% of
compound 2 in 50:50 HEMA:H
2
O solutions were stirred at room temperature for one
day. Unfortunately, 2 was also not soluble under any of these conditions.
With these disappointing results, we then attempted to first dissolve BP 2 in H
2
O
prior to its addition to HEMA to increase its solubility. Solutions of 16%, 18%, and 22%
of 2 in H
2
O were kept at room temperature overnight, and no precipitation was observed.
These saturated aqueous solutions were then used to create the following solutions: 7%,
9%, and 10% of compound 2, respectively, in a HEMA/H
2
O mixture. The solutions were
stirred at room temperature overnight, but 2 was not soluble under any of these
conditions.
Thus, we decided to dilute the amount of compound 2 in solution to increase its
solubility in HEMA, forming solutions containing 9%, 4%, and 2% of compound 2 in
H
2
O. These solutions were then used to form solutions of 4%, 2%, and 1% of compound
2 in mixtures of HEMA/H
2
O, respectively; these solutions were then stirred at room
temperature overnight. Of these solutions, compound 2 initially appeared soluble in the
1% mixture, but precipitation was observed after the solution was stirred overnight. Thus,
78
2 (as a crystalline solid or as an aqueous solution) was not soluble in HEMA under the
several conditions studied. Therefore, we aimed to synthesize α–F polymerizable BPs,
compounds with increased hydrophobicity, increased acidity, and hydrolytic stability.
Several synthetic approaches have been attempted. First, the alkylation of
tetraisopropyl monofluorobisphosphonate 4a (TIPMFBP), which is first synthesized from
the fluorination of tetraisopropyl methylenebisphosphonate (TIPMBP, 3), by
dibromohexane was studied (Scheme 4.2). However, this reaction produced a mixture of
various compounds, including starting material 4a according to
31
P NMR (Figure 4.2).
Scheme 4.2. Synthesis of Bis(propan-2-yl){1-[bis(propan-2-yloxy)phosphoryl]-7-bromo-1-fluoroheptyl}
phosphonate (5) by alkylation of TIPMFBP (4a). Conditions: a) NaH, Selectfluor™, anhydrous THF/DMF;
b) NaH, dibromohexane, anhydrous THF.

Figure 4.2.
31
P NMR of reaction mixture of the alkylation of TIPMFBP (4a).
P P
O
O
O
O
O O
P P
O
O
O
O
O O
H
P P
O
O
O
O
O O
a b
3 4a
F F
Br
5
79
Thus, we decided that reversing the order of these steps may prove to be a more
productive synthetic strategy: alkylation of TIPMBP must first be performed, followed by
fluorination of the resulting BP. We attempted to alkylate TIPMBP with the
commercially available 8-bromooctyl phthalimide, which would introduce a
chromophore that could be easily detected by UV during purification (Scheme 4.3). The
corresponding product, 4b, could then be fluorinated. However, according to
31
P NMR,
less than 15% of product 4b was achieved (Figure 4.3); perhaps these results indicate that
the presence of the phthalimide group adversely affects the progression of the desired
reaction
10
.

P P
O
O
O
O
O O
a
3
P P
O
O
O
O
O O
N
O O
4b

Scheme 4.3. Alkylation of TIPMBP (3) by 8-bromooctyl phthalimide. Conditions: a) NaH, 8-bromooctyl
phthalimide, anhydrous THF.

80

Figure 4.3.
31
P NMR spectrum of reaction mixture of alkylation of TIPMBP with 8-bromooctyl
phthalimide.

Next, a synthetic pathway that initially attaches an alkylhalide to methylene
bisphosphonate, followed by fluorination, was pursued. Then, the introduction of the
amino group will be performed in a subsequent step (Scheme 4.4).
81

Scheme 4.4. Total synthesis of target compound, [1-fluoro-1-phosphono-7-(prop-2-
enamido)heptyl]phosphonic acid (1). a) NaH, dibromohexane, anhydrous THF; b) NaH, Selectfluor™,
anhydrous 50:50 THF/DMF; c) potassium phthalimide, anhydrous DMF; d) 6 N HCl; e) hydrazine hydrate;
f) i) acryloyl chloride, NaHCO
3
, H
2
O (pH 8-9), ii) HCl.

First, TIPMBP 3 was alkylated with dibromohexane, yielding 4c in relatively high
yields (60-70% by
31
P NMR). After starting material 3 was converted to the
corresponding carbanion (which can be observed by
31
P NMR), dibromohexane in
anhydrous THF was added at 0 °C; the reaction mixture was then brought to room
temperature and stirred for 2 days. To avoid substituting at both bromine positions of the
alkyl chain and/or forming a cyclized BP
11
, a large excess of dibromohexane (4-5 eq) was
used, which was then separated from the desired product 4c under vacuum and repeated
P
O
O
O
P
O
O
O
P
O
O
O
P
O
O
O
Br
P
O
O
O
P
O
O
O
Br
F
P
O
O
O
P
O
O
O
N
F
P
O
OH
OH
P
O
HO
HO
NH
2
F
a
b
c
d, e
f
P
O
OH
OH
P
O
HO
HO
HN
F
O
O O
3
4c 5
6 7
1
82
co-evaporation with DMF. Both
1
H and
31
P NMR spectra correspond to the structure of
the desired product 4c, which exhibits a triplet of triplets at 2.1 ppm in the
1
H NMR
(corresponding to the α–H of 4c) and a downfield chemical shift from starting material 2
in the
31
P NMR spectrum (from 15 ppm to 20 ppm).
The crude product contained approximately 20-30% of starting material 3.
Although this crude mixture can be used in the next synthetic step, the presence of
compound 3 can lower yields for the next target product, 5, by yielding compounds 4a
and tetraisopropyl difluoromethylene bisphosphonate. Thus, compound 3 and product 4c
were separated based on their differing solubility properties. H
2
O was directly added to
crude 4c, and the solution was emulsified. Two layers separated upon standing, and the
upper aqueous layer, containing 3, was decanted. This washing procedure was repeated
several times until the organic layer no longer contained 3 according to NMR.
Compound 4c was then fluorinated with Selectfluor™ to yield intermediate 5.
After 4c was converted to its corresponding carbanion (which can be observed by
31
P
NMR), Selectfluor™ and anhydrous DMF were added at 0 °C
12
. The reaction mixture
was then brought to room temperature and stirred for 1-3 hrs (the reaction can be
followed by
31
P NMR). When the reaction was complete, the reaction mixture was
quenched with saturated NH
4
Cl, diluted with CH
2
CH
2
, and washed with 0.1 N NaHCO
3
;
the product 5 was used without further purification. According to NMR, ~5-20% of 4c
can be present with the product (Figure 4.4). Compound 5 corresponds to the doublet
upfield (12 ppm) from compound 4c (singlet at 22 ppm) in the
31
P NMR spectrum, an
expected characteristic of an α–F BP compound. According to
31
P NMR, ~70-85% of
product 5 can be achieved.
83
Figure 4.4.
31
P NMR of Bis(propan-2-yl) {1-[bis(propan-2-yloxy)phosphoryl]-7-
bromoheptyl}phosphonate (4c).

The bromo moiety of 5 was then exchanged with phthalimide, yielding compound
6. Potassium phthalimide was combined with 5 in anhydrous DMF, and the reaction
mixture was heated at 100 °C
13
. If this reaction is allowed to proceed for a long period of
time (e.g. overnight), we have found that, under these basic reaction conditions, the α–F
moiety aids in the selective hydrolysis of one isopropyl group, yielding a partial ester of 6
(~20% yield versus 45% yield of tetraisopropyl ester 6 by
31
P NMR). Similar
observations have been observed with various α-halogenated BP esters
14
. The
31
P NMR
of the partial ester of 5 shows two unequivalent phosphorus signals, one as a doublet of
doublets – split by the 2
nd
phosphorus and α–F at 15.5 ppm - and the second as a broad
unresolved peak at 4.5 ppm (Figure 4.5). The introduction of the UV-Visible
chromophore eases purification, and both the tetraisopropyl ester and triisopropyl ester of
6 can be isolated by preparative TLC or column chromatography.
84

Figure 4.5.
31
P NMR of partial ester of Bis(propan-2-yl) {1-[bis(propan-2-yloxy)phosphoryl]-7-(1,3-dioxo-
2,3-dihydro-1H-isoindol-2-yl)-1-fluoroheptyl}phosphonate (6).

However, the complete conversion of 6 from 5 can be completed within 3-4 hours
without any hydrolysis of the isopropyl ester groups.
1
H NMR spectra was used to
confirm completion of the reaction; if necessary, additional potassium phthalimide may
be added to force the reaction to completion. A shortened reaction time resulted in less
complicated crude product mixtures, thus simplifying the purification of 6 (purified via
preparative TLC or silica gel column chromatography).
Cleavage of the isopropyl and phthalimide groups of 6 then yielded key
intermediate 7, which can be readily conjugated to an acryl moiety. According to ESI-
MS, it appeared that refluxing compound 6 in concentrated HCl for 2 days resulted in a
mixture of the free acid of 6 and the free acid of 7, which may result in lower yields of
product 7. Thus, the synthesis of 7 was performed in two stages: cleavage of the
isopropyl groups of 5 with HCl, followed by removal of the phthalimide by hydrazine
hydrate. After complete conversion to 7 was achieved (can be confirmed by
1
H NMR),
85
the solvent was removed under vacuum, and the resulting solids were dissolved in
minimal H
2
O and precipitated with MeOH, removing side products and yielding purified
7 as a hydrazinium salt.
A number of different synthetic approaches to achieve compound 1 from
intermediate 7 were studied (Scheme 4.5) and are summarized in Table 4.1.
P
F P OH
O OH
OH
OH O
X
O
P
F P OH
O OH
OH
OH O
+
1
H
N
O
H
2
N
7

Scheme 4.5. General synthesis of target compound, 1, from key intermediate 7 and various acryl starting
materials. For summary of all attempted reaction conditions, refer to Table 4.1.

Similar to the synthesis of Pam monomers, acryloxysuccinimide has been used as
a starting material for this reaction (Entries 1-2, Table 4.1). However, the side product,
NHS, had similar solubility properties as the target product, 1; thus, HPLC appeared to be
the most useful method for removing this impurity from 1. Since this purification method
would be difficult on a large scale, we decided to pursue other acrylate starting materials,
such as acrylic anhydride or acryloyl chloride.
Acrylic anhydride was an attractive alternative since the only side product from
this reaction would be acrylic acid, a compound easily soluble in a wide range of organic
solvents. However, as shown in Entry 3 of Table 4.1, solubility with the chosen reaction
conditions was low, and thus this direction was not pursued further.

86
Entry
No.
X Base Solvent Solubility Acidification Purification
1 NHS Cs
2
CO
3
H
2
O/MeOH Poor Dowex (H
+

form)*
ppt/extract with
MeOH, EtOH,
iPrOH, acetone,
CH
2
Cl
2
; HPLC
2 NHS TEA H
2
O Very
Good
TFA ppt with acetone
3 anhyd-
ride
TBA H
2
O/MeOH Poor HCl ppt with CH
2
Cl
2
4 Cl Cs
2
CO
3
H
2
O/acetone Low Dowex (H
+

form)*
-
5 Cl TEA H
2
O/dioxane Low Dowex (H
+

form)*
crystallization with
ether
6 Cl TEA H
2
O Very
Good
HCl ppt with EtOAc,
EtOAc/acetone,
EtOH/acetone,
EtOH
7 Cl TBA H
2
O/MeOH Poor - -
8 Cl KOH H
2
O/dioxane Low - -
9 Cl K
2
CO
3
H
2
O/dioxane Good Dowex (H
+

form)*
crystallization with
ether & iPrOH
10 Cl Na
2
CO
3
H
2
O Very
Good
Dowex (H
+

form)*
ppt with
MeOH/EtOH
11 Cl Na
2
CO
3
H
2
O Very
Good
HCl/NaCl ppt/extract with
EtOH
Table 4.1. Summary of reaction conditions and purification methods attempted for 1. TEA = triethylamine,
TBA = tributylamine, TFA = trifluoroacetic acid, ppt = precipitate. *Dowex (H
+
form) removes excess salt
while acidifying compound.

We then focused our efforts on using acryloyl chloride for this reaction (Entries 4-
11, Table 4.1)
5
. Through these studies, the following major conclusions were reached:
87
1. Any organic solvent used for this reaction lowers the solubility of starting
material 7, thus the reaction must take place only in H
2
O;
2. Any solvent that may contain traces of peroxide (e.g. ether, dioxane)
should be avoided during this reaction and subsequent purification;
3. Although Dowex (H
+
form) is an attractive tool to simultaneously acidify
the BP and remove excess salts, yields appeared to drop dramatically when
used.
Thus, the reaction conditions and purification method shown in Entry 11 of Table
4.1 appeared to be the most successful. In order to avoid conjugating the nucleophilic
hydrazine to acryloyl chloride, 7 must first be converted from a hydrazinium salt into a
sodium salt. Compound 7 was dissolved in minimal amount of H
2
O and pH was adjusted
to ~13 with concentrated NaOH. MeOH was added to the solution to precipitate the
sodium salt of 7 and dissolve excess NaOH, and the mixture was centrifuged. The filtrate
was decanted, and the resulting white precipitate was washed with additional MeOH and
the centrifugation process was repeated. The precipitate may finally be washed with
acetone, followed by an additional centrifugation. The white precipitate was then dried in
a desiccator, giving compound 7 as a tetrasodium salt.
Compound 7 was then dissolved in H
2
O, and the solution was brought to 0 °C. To
the cooled solution was added 1 eq of acryloyl chloride dropwise. During this step, the
pH was monitored and Na
2
CO
3
was added to maintain a pH of 8-9. The reaction mixture
was brought to room temperature and stirred an additional 15-20 min. This addition
procedure was repeated 2-4 times during which the reaction time was slowly increased to
45-60 min. The progression of the reaction may be followed by
1
H NMR. If a triplet at
88
2.9 ppm (corresponding to the CH
2
protons adjacent to the amine group of 7) is observed,
an additional portion of acryloyl chloride can be added to push the reaction to completion
(Figure 4.6A). Once the triplet at 2.8 ppm was completely converted to give a triplet at
3.2 ppm (corresponding to the CH
2
protons adjacent to the amide group of 1), the
reaction was deemed complete (Figure 4.6B).

Figure 4.6. Reaction progression of synthesis of [1-fluoro-1-phosphono-7-(prop-2-
enamido)heptyl]phosphonic acid (1) by
1
H NMR. A) 1 hr at rt after 4th addition (1 eq per addition) of
acryloyl chloride; B) 1 hr at rt after 5
th
addition of acryloyl chloride.
89
To remove sodium acrylate, the crude reaction mixture was dissolved in minimal
H
2
O, and the pH was brought to 11-12 with NaOH. EtOH was then added (~5x volume
of H
2
O present) to precipitate product 1 as a tetrasodium salt and dissolve sodium
acrylate, and the mixture was centrifuged. The filtrate was decanted and concentrated
under vacuum, and the precipitate was dried under vacuum. This precipitation method
was repeated until sodium acrylate was no longer detected by
1
H NMR. Filtrate portions
can be combined and concentrated under vacuum. Although this sample mostly consists
of sodium acrylate and sodium hydroxide, some product 1 may also be present. To
recover this product loss, the sample can be treated as described above.
To isolate the free acid of 1, the purified tetrasodium salt of 1 was dissolved in
H
2
O, and the pH was brought to 0.5 with concentrated HCl. The solvent was removed,
yielding a mixture of solids, consisting of the free acid of 1 and NaCl.
Abuelyaman et al. previously utilized MeOH to isolate their polymerizable BP
products
5
. However, this method most likely will not remove a significant amount of
NaCl. Thus, to avoid excess NaCl in our samples, EtOH was used to extract 1 from
NaCl
15
. To the mixture of free acid 1 and NaCl was added ~50 mL of EtOH; excess
EtOH will dissolve a significant amount of NaCl while an insufficient amount of EtOH
may lead to product loss. After the solution was stirred at room temperature for 30 min
and then filtered, the white precipitate was dried in a desiccator. The filtrate was
concentrated under vacuum, yielding a mixture of white crystals (most likely NaCl) and
viscous yellow oil (our product). To this mixture was added a smaller volume (~10 mL)
of EtOH to dissolve our product while precipitating the remaining amount of NaCl, and
the solution was again stirred at room temperature for 30 min and centrifuged. The
90
precipitate, which may contain traces of 1 as a monosodium salt and can be additionally
extracted with EtOH to increase product yield, was dried in a desiccator. The filtrate was
decanted and concentrated under vacuum, giving a clear yellow viscous oil. The oil was
taken up into H
2
O and lyophilized to constant weight, giving free acid 1 as sticky, foam-
like solids.
Although this final synthetic step proceeds to completion, free acid 1 was isolated
in low yields. During the reaction, a large amount of Na
2
CO
3
was added to maintain a
basic pH, necessary for the deprotonation of the amine group of 7 and thus the
progression of the reaction. In addition, the hydrolysis of acryloyl chloride generated the
side product sodium acrylate. Therefore, the crude reaction mixture at the completion of
the reaction contained mostly these compounds, increasing the difficulty of isolating the
target product.
However, this purification method is advantageous for large scale syntheses since
chromatography is not necessary. The purification utilizes our estimation of the pKa
values of 1 to exploit its varying solubility properties. More specifically, although acrylic
acid has similar solubility properties as the mono- or disodium salt of 1, the tetrasodium
salt of 1 has different solubility properties versus sodium acrylate and can be precipitated
away from this contamination in EtOH. Furthermore, unlike its corresponding sodium
salt forms, the free acid of 1 exhibits dissimilar solubility properties than NaCl and can be
dissolved in minimal EtOH. The use of EtOH instead of MeOH also reduces the amount
of NaCl in the final sample. However, it appears difficult to completely push the
equilibrium between the different species of 1 (e.g. trisalt versus tetrasalt, monosalt
versus free acid), which may result in product loss during purification.
91
Conclusion
In this study, the synthesis and purification of a pamidronate-based monomer was
optimized. Due to its low solubility in solvent systems typically used for polymerizations
of dental materials, we were interested in designing and synthesizing a polymerizable BP
with improved characteristics: increased acidity, improved hydrophobicity, and
hydrolytic stability. Introduction of fluorine at the α-carbon of the BP helps to impart key
characteristics into the target compound. Various synthetic pathways to achieve a novel
α–fluorinated, amino bisphosphonic acid, which can undergo subsequent conjugation to
an acryl monomer, were also studied, and the synthesis of the novel α-fluorinated,
polymerizable BP, 1, was achieved.

Experimental Methods
Reagents and Spectral Measurements. Funding for this research was provided by Kerr-
Sybron Corporation. Pamidronic acid and tetraisopropyl methylene bisphosphonate were
gifts from Novartis and Rhodia, Inc, respectively. THF was obtained by distillation from
sodium metal; anhydrous DMF was purchased from EMD Chemicals. All other
compounds were purchased from Sigma Aldrich or Alfa Aesar and used as supplied by
the manufacturer. Thin layer chromatography was performed on Merck Silica Gel 60 F
254

plates, and the developed plates were visualized under a UV lamp at 254 nm. NMR
spectra were recorded on either 400 MHz Varian or 500 MHz Bruker spectrometers.
Chemical shifts are reported relative to internal D
2
O or CDCl
3
standard (for
1
H NMR
spectra) or 85% H
3
PO
4
(for
31
P NMR spectra). All relevant spectral data may be found in
Appendix B.
92
[1-hydroxy-1-phosphono-3-(prop-2-enamido)propyl]phosphonic acid (2). Pamidronic
acid (0.63, 2.66 mmol, 1 eq) was dissolved in 13 mL of 1 M NaHCO
3
, and the pH was
adjusted to 8 with Na
2
CO
3
. To the solution was added N-acryloxysuccinimide (0.45 g,
2.66 mmol, 1 eq) in 500 µL anhydrous DMF, which was then combined with the
pamidronate solution; pH of reaction mixture was re-adjusted with Na
2
CO
3
. Slight
precipitation occurs but will dissipate as the reaction progresses. The reaction mixture
was stirred at rt.
31
P NMR was used to follow the reaction; additional portions of N-
acryloxysuccinimide may be added to push the reaction to completion. After 3 h at rt,
~11% of Pam remained according to
31
P NMR; therefore, an additional 100 mg of N-
acryloxysuccinimide (0.2 eq) was added and the reaction mixture was stirred at rt
overnight. Then, the reaction mixture was eluted on a DOWEX (H
+
form) column 2x,
converting the mixture of reaction products to their respective acidic forms. The eluant
was isolated and concentrated under vacuum. The resulting compound was dissolved in
MeOH and precipitated with acetone. The solids 2 were filtered and dried in a desiccator,
giving white crystals (28% yield).
1
H NMR (D
2
O): δ 1.99-2.12 (brm, 2H), 3.44 (t, J = 7.8 Hz, 2H), 5.59 (d, J = 10 Hz, 1H),
5.99-6.13 (brm, 2H).
31
P{
1
H} NMR (D
2
O): δ 18.61 (s). ESI-MS (negative ion): calcd [M
- H]
-
= 288.1168 m/z; found [M - H]
-
= 287.9 m/z.
Bis(propan-2-yl) {1-[bis(propan-2-yloxy)phosphoryl]-7-bromoheptyl}phosphonate
(4c). To 515 mg of NaH (95%, 21.4 mmol, 1 eq) was added 40 mL of distilled THF, and
the suspension was stirred at 0 °C for ~20 min. Then, 7.0 g of tetraisopropyl methylene
bisphosphonate (TIPMDP, 3, 20 mmol, 1 eq) in 2 mL distilled THF was added dropwise
to the chilled solution, generating H
2
and forming a clear solution. After the solution was
93
stirred an additional 20-30 min at 0 °C, 22.3 g of dibromohexane (91 mmol, 4.5 eq) was
added. The solution was then brought to rt and stirred at rt for 3 days. The reaction
mixture was diluted with CH
2
Cl
2
and washed with 1.6 M citric acid, followed by 0.1 N
NaHCO
3
. The organic layer was then dried over Na
2
SO
4
, filtered, and concentrated under
vacuum.
H
2
O was added to the resulting solution, and the mixture was emulsified. The
solution was allowed to separate, and the top layer was decanted. This procedure was
repeated (~3-4x) until TIPMDP was not detected by
31
P NMR. Then, the organic layer
was taken up into EtOAc, dried over Na
2
SO
4
, filtered, and concentrated under vacuum.
Excess dibromohexane was removed under vacuum and repeated co-evaporation with
DMF; this process was also repeated until dibromohexane was no longer observed in
1
H
NMR. The crude oil (~90% yield according to
31
P NMR; sample contains slight amount
of other alkylated phosphonate compounds, which were observed in both
1
H and
31
P
NMR spectra) was used without further purification.
1
H NMR (CDCl
3
): δ 1.29-1.41 (brm, 28H), 1.43-1.53 (brm, 2H), 1.57-1.67 (brm, 2H),
1.83-1.94 (brm, 4H), 2.16 (tt, J = 25.1, 5.9 Hz, 1H), 3.44 (t, J = 6.4 Hz, 2H), 4.74-4.89
(brm, 4H).
31
P{
1
H} NMR (CDCl
3
): δ 22.06 (s).
Bis(propan-2-yl){1-[bis(propan-2-yloxy)phosphoryl]-7-bromo-1-fluoroheptyl}
phosphonate (5). To 330 mg of NaH (95%, 13.8 mmol, 1 eq) was added 25 mL of
distilled THF, and the suspension was stirred at 0 °C for ~15 min. Then, 4.7 g of 4c in 5
mL distilled THF was added, and the solution stirred at 0 °C for ~1 h (or until carbanion
was observed according to
31
P NMR). Then, 3.3 g of Selectfluor™ was added to the
reaction mixture (under heavy flushing of N
2
gas), followed by the addition of 30 mL
94
anhydrous DMF. The reaction mixture was stirred an additional 1 h at rt. The reaction
mixture was then quenched with saturated NH
4
Cl and diluted with CH
2
CH
2
. The organic
layer was washed with 0.1 N NaHCO
3
, dried over Na
2
SO
4
, and concentrated under
vacuum, yielding a viscous orange-red oil (~80-85% yield by
31
P NMR). The crude
product was used without further purification.
1
H NMR (CDCl
3
): δ 1.33-1.43 (brm, 27H), 1.45-1.54 (brm , 2H), 1.65-1.74 (brm, 2H),
1.85-1.94 (brm, 2H), 2.04-2.24 (brm, 2H), 3.45 (t, J = 7 Hz, 2H), 4.77-5.00 (brm, 4H).
31
P{
1
H} NMR (CDCl
3
): δ 13.04 (d, J = 77.5 Hz).
19
F NMR (CDCl
3
): δ -192.352 (tt, J=
73.6, 18.9 Hz).
Bis(propan-2-yl) {1-[bis(propan-2-yloxy)phosphoryl]-7-(1,3-dioxo-2,3-dihydro-1H-
isoindol-2-yl)-1-fluoroheptyl}phosphonate (6). To 7.5 g of 5 (14 mmol, 1 eq) in 75 mL
anhydrous DMF was added 2.6 g of potassium phthalimide (14 mmol, 1 eq). The reaction
mixture was stirred at 100 °C for 90 min, and a small portion of solution was
concentrated under vacuum. The resultant residue was analyzed by
1
H NMR. When the
reaction reached completion without cleavage of the tetraisopropyl ester groups, the
reaction mixture was brought to rt, and the solvent was removed under vacuum. The
resulting residue was taken up into CH
2
CH
2
, filtered, and concentrated under vacuum.
The solution was dissolved in small amount of CH
2
CH
2
and purified by silica gel column
chromatography, eluted with slight gradient of 0-20% acetone in CH
2
CH
2
. The eluant
eluting with 10% acetone in CH
2
CH
2
was analyzed by TLC (eluted with 10% acetone in
CH
2
CH
2
); samples with an R
f
value of ~0.4 were collected as product and concentrated
under vacuum, giving 2.8 g of 6 as a clear oil (isolated yield 33%).
95
1
H (CDCl
3
): δ 1.26-1.33 (m, 24H), 1.28-1.65 (m, 4H), 1.93-2.12 (m, 2H), 3.60 (t, J = 7.4
Hz), 4.71-4.86 (m, 4H), 7.61-7.67 (m, 2H), 7.74-7.79 (m, 2H).
31
P{
1
H} NMR (CDCl
3
): δ
12.97 (d, J = 76.4 Hz).
19
F NMR (CDCl
3
): δ -192.67 (tt, J = 73.2, 27.3 Hz).
(7-amino-1-fluoro-1-phosphonoheptyl)phosphonic acid (7). 11.3 g of compound 6 (1
mmol, 1 eq) was suspended in 60-70 mL concentrated HCl and refluxed overnight. The
solvent was then removed under vacuum, and the resultant residue was dissolved in ~70
mL H
2
O. The yellow solution was cooled to 0 °C (all precipitate eventually dissolved),
and 9.6 g of hydrazine hydrate (11 mmol, 10 eq) was added dropwise. The reaction
mixture was brought to rt and stirred for 90 min. A small portion of the reaction mixture
may be concentrated under vacuum and analyzed by
1
H NMR to follow progression of
reaction. When the reaction reached completion (100% yield by
1
H NMR), the solvent
was removed under vacuum, and the resulting residue was dissolved in minimal H
2
O
(requiring slight heating); MeOH was then added to precipitate the product. The mixture
was filtered, and the precipitate was dried in a desiccator overnight. To exchange
hydrazinium salts to sodium salts, the solids were dissolved in H
2
O, and pH was adjusted
to ~13 with NaOH. The solvent was removed under vacuum, and the resulting solids
were then washed with MeOH. The solids were centrifuged, and the filtrate was
decanted. The resulting precipitate was dried in a desiccator, yielding 7 as a tetrasodium
salt.
1
H NMR (D
2
O): δ 1.17-1.31 (m, 4H), 1.42-1.58 (m, 4H), 1.81-2.01 (m, 2H), 2.841 (t, J =
7.3 Hz).
31
P{
1
H} NMR (D
2
O): δ 13.87 (d, J = 69.1 Hz).
19
F NMR (D
2
O): δ -185.98 (tt, J
= 74.0, 27.6 Hz). ESI-MS (negative ion): calcd [M - H]
-
= 292.1668 m/z; found [M - H]
-

= 292.06 m/z.
96
[1-fluoro-1-phosphono-7-(prop-2-enamido)heptyl]phosphonic acid (1). A solution of
8.4 g of 6 (tetrasodium salt, 18.6 mmol, 1 eq, assume actual BP only ~65% by weight) in
30 mL H
2
O was cooled to 0 °C. To the chilled solution was added 1.7 g of acryloyl
chloride (18.7 mmol, 1 eq) dropwise with vigorous stirring and continuous monitoring of
pH. When the pH dropped below 8 during this addition, Na
2
CO
3
was added until pH 8-9.
The reaction mixture was then brought to rt and stirred an additional 20 min at rt. The
addition process was then repeated 2x, adding 1 eq of acryloyl chloride during each
round. After the 3rd equivalent was added, the reaction mixture was stirred at rt for 1 h.
Then, a small portion of reaction mixture was concentrated under vacuum and then
analyzed by
1
H NMR. The addition process was repeated with an additional 1 eq of
acryloyl chloride. After 1 h at rt, the reaction mixture was again checked by
1
H NMR.
Since starting material remained, an additional 1 g of acryloyl chloride (11 mmol, 0.6 eq)
was added. After 1 h at rt, the reaction mixture was checked by
1
H NMR. The reaction
appeared to be complete, and the solvent was then concentrated under vacuum. The
resulting solids were split into 2 portions. The first portion was dissolved in ~15 mL H
2
O,
and pH was adjusted to 12.5 with concentrated NaOH. To the solution was added ~300
mL EtOH, precipitating out product, and the mixture left to stand for ~15 min. The
mixture was then centrifuged, and the filtrate was decanted. The precipitate was re-
dissolved in H
2
O and concentrated under vacuum to co-evaporate EtOH. The second
portion of crude reaction mixture was dissolved in 7 mL H
2
O, and pH was adjusted to
12.5 with concentrated NaOH. To the solution was added 240 mL EtOH, and the mixture
was again centrifuged. The filtrate was decanted and combined with the first EtOH
solution; precipitate samples from both portions were combined and dried in a desiccator.
97
The dried solids were then dissolved in H
2
O, and pH was brought to 0.5 with
concentrated HCl. The solvent was removed under vacuum, giving white solids. The
resulting solids were extracted with 50 mL EtOH for 30 min at rt. The solution was
filtered and concentrated under vacuum. The resulting residue, containing white
precipitate suspended in a yellow oil, was extracted with 10 mL EtOH for 30 min at rt.
The mixture was centrifuged, and the filtrate was decanted. The solution was
concentrated under vacuum, giving 0.9 g of 1 as a yellow oil (free acid form) prior to
lyophilization. The sample was combined with an additional 8 g of 1 (free acid form) and
lyophilized, giving a total of 6.1 g of 1 as a crystalline-type solid (isolated yield 20%).
NMR spectra below taken of lyophilized product but pH of NMR sample adjusted to ~7
with NaHCO
3
.
1
H NMR (D
2
O): δ 1.15-1.28 (m, 4H), 1.37-1.51 (m, 4H), 1.82-2.02 (m, 2H), 3.12 (t, J =
6.9 Hz, 2H), 5.58 (dd, J = 10.2, 1.7 Hz, 1H), 5.98-6.15 (m, 2H).
31
P{
1
H} NMR (D
2
O): δ
13.93 (d, J = 69.1 Hz).
19
F NMR (D
2
O): δ -185.91 (tt, J = 72.6, 27.4 Hz). ESI-MS
(negative ion): calcd [M - H]
-
= 346.2142 m/z; found [M - H]
-
= 346.21 m/z.

Solubility Studies of 2. 50 mg of compound 2 was added to 200 mg of 2-
hydroxyethylmethacrylate (HEMA), forming a 20% mixture of 2 in solution; the mixture
was stirred at rt for one day. Compound 2 was not soluble under these conditions.
50 mg of compound 2 was added to 166 mg of 50:50 HEMA:H
2
O, forming a
mixture with 23% of 2 in the 50:50 HEMA:H
2
O solution. Also, 8 mg of compound 2 was
added to 280 mg of 50:50 HEMA:H
2
O, forming a mixture of 3% bisphosphonate 2 in the
98
HEMA/H
2
O solution. These mixtures were also stirred at rt for one day. Compound 2
was also not soluble under any of these conditions.
209 mg of crystalline 2 was dissolved in 2 mL H
2
O with slight heating, and the
solution was then concentrated under vacuum until 1.155 mL H
2
O remains, yielding a
solution of 16% of 2 in H
2
O (Solution A). 300 µL of Solution A was reserved. The
remaining solution was further concentrated under vacuum until 810 µL of solvent
remains, yielding a solution of 18% of 2 in H
2
O (Solution B). 300 µL of Solution B was
also reserved. The remaining solution was again concentrated under vacuum until 425 µL
remained, yielding a solution of 22% of 2 in H
2
O (Solution C). Solutions A, B, and C
were kept at rt overnight, and no precipitation was observed. Then, 114 mg of Solution A
was added to 114 mg of HEMA, yielding 7% of 2 in a HEMA/H
2
O mixture (Solution D).
Also, 98 mg of Solution B was added to 102 mg of HEMA, yielding 9% of 2 in a
HEMA/H
2
O mixture (Solution E). Additionally, 107 mg of Solution C was added to 107
mg of HEMA, yielding 10% of 2 in a HEMA/H
2
O mixture (Solution F). Solutions D, E,
and F were stirred at rt overnight, but 2 was not soluble under any of these conditions.
Then, 50 µL of Solution B was diluted with 50 µL H
2
O, yielding 9% of
compound 2 in H
2
O (Solution G). 119 mg of Solution G was added to 109 mg HEMA,
yielding 4% of compound 2 in a HEMA/H
2
O mixture (Solution H). 50 µL of Solution A
was diluted with 150 µL H
2
O, yielding 4% of compound 2 in H
2
O (Solution I). 110 mg
of Solution I was added to 105 mg HEMA, yielding 2% of compound 2 in a HEMA/H
2
O
mixture (Solution J). Also, 50 µL of Solution I was additionally diluted with 50 µL of
H
2
O, yielding 2% of 2 in H
2
O (Solution K). 100 mg of Solution K was added to 110 mg
of HEMA, yielding 1% of 2 in a HEMA/H
2
O mixture (Solution L). Solutions H, J, and
99
L were stirred at rt overnight. Of these solutions, 2 was only soluble under the conditions
described for Solution L, but only immediately after the addition of Solution K to
HEMA. After Solution L was stirred at rt overnight, precipitation was observed.

100
Chapter 4 References
1. Moszner, N.; Salz, U., Recent Developments of New Components for Dental
Adhesives and Composites. Macromolecular Mateials and Engineering 2007, 292, (3),
245-271.

2. Qian, X., Two-Part Self-Adhering Dental Compositions. US Patent 7214726, May
8, 2007.

3. Salz, U.; Zimmermann, J.; Zeuner, F.; Moszner, N., Hydrolytic Stability of Self-
Etching Adhesive Systems. Journal of Adhesive Dentistry 2005, 7, (2), 107-116.

4. Moszner, N.; Salz, U.; Zeuner, F.; Zimmermann, J.; Rheinberger, V., Dental
Material on the Basis of Acryloester Phosphonic Acid. EP1374829, January 1, 2002.

5. Abuelyaman, A. S.; Boardman, G. S.; Shukla, B. A.; Aasen, S. M.; Mitra, S. B.;
Mikulla, M.; Cinader, D. K., Dental Compositions Including Polymerizable
Bisphosphonic Acids. US 2004206932, October 21, 2004.

6. Catel, Y.; Degrange, M.; Le Pluart, L.; Madec, P.-J.; Pham, T.-N.; Picton, L.,
Synthesis, Photopolymerization and Adhesive Properties of New Hydrolytically Stable
Phosphonic Acids for Dental Applications. Journal of Polymer Science, Part A: Polymer
Chemistry 2008, 46, (21), 7074-7090.

7. Wang, L.; Zhang, M.; Yang, Z.; Xu, B., The First Pamidronate Containing
Polymer and Copolymer. Chemical Communications 2006, 2006, (26), 2795-2797.

8. Mikami, K.; Itoh, Y.; Yamanaka, M., Fluorinated Carbonyl and Olefinic
Compounds: Basic Character and Asymmetric Catalytic Reactions. Chemical Reviews
2003, 104, (1), 1-16.

9. Romanenko, V. D.; Kukhar, V. P., Fluorinated Phosphonates: Synthesis and
Biomedical Application. Chemical Reviews 2006, 106, (9), 3868-3935.

10. Gourves, J.-P.; Couthon, H.; Sturtz, G., Synthesis of 3,4-Dihydro-2h-Pyrido[1,2-
B]Isoindol-1-One and 3,4-Dihydro-2h-Pyrido[1,2-B]Pyrrolidin-1-One Functionalized At
the C(6) Position By an Intramolecular Horner-Wadsworth-Emmons Reaction. European
Journal of Organic Chemistry 1999, 1999, (12), 3489-3493.

11. Ebetino, F. H.; Degenhardt, C. R.; Jamieson, L. A.; Burdsall, D. C., Recent Work
on the Synthesis of Phosphonate-Containing, Bone-Active Heterocycles. Heterocycles
1990, 30, (2), 855-862.

101
12. Marma, M. S.; Khawli, L. A.; Harutunian, V.; Kashemirov, B. A.; McKenna, C.
E., Synthesis of α-Fluorinated Phosphonoacetate Derivatives Using Electrophilic
Fluorine Reagents: Perchloryl Fluoride Versus 1-Chloromethyl-4-Fluoro-1,4-
Diazoniabicyclo[2.2.2]Octane Bis(Tetrafluoroborate) (Selectfluor®). Journal of Fluorine
Chemistry 2005, 126, (11-12), 1467-1475.

13. Bako, P.; Novak, T.; Ludanyi, K.; Pete, B.; Toke, L.; Keglevich, G., D-Glucose-
Based Aza-Crown Ethers With a Phosphonoalkyl Side Chain: Application as
Enantioselective Phase Transfer Catalysts. Tetrahedron: Asymmetry 1999, 10, (12),
2373-2380.

14. Reddy, G. V.; Jacobs, H.; Gopalan, A. S.; Barrans, R. E., Jr.; Dietz, M. L.;
Stepinski, D.; Herlinger, A., Synthesis of Symmetrical Methylenebis(Alkyl Hydrogen
Phosphonates) By Selective Cleavage of Methylenebis(Dialkyl Phosphonates) With
Morpholine. Synthetic Communications 2004, 34, (2), 331-344.

15. Pinho, S. P.; Macedo, E. A., Solubility of NaCl, NaBr, and KCl in Water,
Methanol, Ethanol, and Their Mixed Solvents. Journal of Chemical and Engineering
Data 2005, 50, (1), 29-32.

102
Chapter 5
Chiral Phosphonocarboxylate Analogues of Risedronate and Minodronate:
3-PEHPC and 3-IPEHPC

Introduction
Some phosphonocarboxylate (PC) analogues of biologically active nitrogen-
containing bisphosphonates (N-BPs) have recently emerged as a promising new direction
for anti-cancer therapeutics. PC compounds typically exhibit decreased bone affinity
compared to their BP parent drugs, a property that may be attractive particularly in cases
where long-term bone retention may be undesirable, such as in the treatment of pediatric
bone diseases
1
. Currently, lowered bone affinity is believed to actually increase the anti-
tumor potential of BPs
1, 2
. BPs with decreased affinity to bone may accumulate in higher
concentrations in the bone marrow microenvironment, suggesting that compounds such
as PCs may not only exhibit a more direct anti-tumor potential but also act more
effectively on tumor cells within the bone marrow
1, 2
.
Our laboratory has been particularly interested in two PC inhibitors, analogues of
the potent N-BPs risedronate (RIS, 1) and minodronate (3): 2-hydroxy-2-phosphono-3-
pyridin-3-yl-propionic acid (3-PEHPC, previously known as NE10790, 2) and 2-
hydroxy-3-imidazo[1,2-a]pyridin-3-yl-2-phosphonopropionic acid (3-IPEHPC, 4),
respectively (Figure 5.1). While their parent compounds inhibit farnesyl diphosphate
synthase (FPPS), both PC compounds have been shown to specifically inhibit Rab
geranylgeranyltransferase (GGTase), an enzyme further downstream than FPPS in the
103
mevalonate pathway (Figure 5.2). Rab GGTase catalyzes the post-translational
geranylgeranylation of typically two C-terminal cysteines in Rab GTPases
3
. Inhibition of
Rab GGTase therefore prevents prenylation and membrane localization of Rab GTPases
2,
4-6
. Recently, compounds 2 and 4 were more specifically shown to inhibit the second
geranylgeranylation event of RGGTase
3
. Interestingly, although 2 is only a weak
inhibitor of bone resorption, it has also been shown to inhibit invasion of breast and
prostate cancer cells, prevent osteolytic bone disease, and reduce tumor burden in vivo
1
.
Figure 5.1. Structures of the nitrogen-containing BPs, risedronate (1) and minodronate (3), and their
corresponding phosphonocarboxylate analogues, 3-PEHPC (2) and 3-IPEHPC (4), respectively.

Figure 5.2. A general schematic of the mevalonate pathway, depicting known inhibitors of specific
enzymes (shown in boxes).
P
O
OH
O
OH
OH
N
N
HO
4
P
P
O
OH
O
OH
OH
N
N
HO
3
OH
N
P
HO P
OH
OH
O
OH
O
OH
N
HO P
OH
OH
O
OH
O
2 1
104

By exchanging the phosphonate moiety to a carboxylate, PC compounds contain a
potential chiral center at the α-carbon, which importantly gives rise to the possibility of
differing biological activity between the two enantiomers. In fact, unlike 2, the
enantiomers of 4 were recently shown to exhibit different inhibitory potencies; the more
active enantiomer is reportedly at least 25-fold more potent than 2 as an inhibitor of Rab
GGTase
3
. Thus, separation of the enantiomers for biological studies is clearly an
important task.
Resolution of the component enantiomers of 3-PEHPC was previously
accomplished in our laboratory by chiral reverse-phase HPLC
7
. In this chapter, the
synthesis and resolution of 3-IPEHPC into its component enantiomers is discussed
8
.
Furthermore, although enantiomeric purity may be closely estimated by chiral reverse-
phase HPLC, we were interested in developing an additional analytical method to
determine this property of a sample and identified a chiral ligand exchange reverse-phase
HPLC method to achieve this goal.

Results and Discussion
Although a previous patent included the conceptual synthesis of 3-IPEHPC, the
described methodology did not appear to lead to the desired product
9
. Thus, the first
synthesis of 4 was accomplished in six steps by Dr. Isabelle Mallard-Favier and Dr. Boris
Kashemirov of our laboratory (Scheme 5.1)
8
. Briefly, a Vilsmeier reaction is employed to
convert the heterocycle 5 to the aldehyde 6. Ethylazidoacetate, synthesized from
ethylbromoacetate, is then added to the aldehyde 6, giving synthon 7, which is then used
105
as a crude product. Catalytic hydrogenation with hydrogen and 10% palladium on
charcoal successfully yielded enanime 8, and the crude product is then converted to the
corresponding enol 9 with acetic acid and water at low temperature. Although 9 exists
mostly as its enol form, diethyl phosphite apparently still reacts with 9, yielding the
desired product 10. The ester groups of crude 10 are then hydrolyzed by concentrated
HCl, producing crude 4 as a racemate.

Scheme 5.1. Total synthesis of 3-IPEHPC. i) POCl
3
/DMF, 30 min at 120 ˚C, then 2 h at 85 ˚C; ii)
N
3
CH
2
CO
2
Et, NaOEt, -30 ˚C, then 3 h at rt; iii) H
2
, 10% Pd/C, CH
3
OH; iv) CH
3
CO
2
H/H
2
O (1:7), -20 ˚C;
v) HP(O)(OEt)
2
, 18 h, 70 ˚C; vi) concentrated HCl, reflux overnight.

Racemic 4 is purified by semi-preparative reverse-phase HPLC. A peak eluting at
12.5 min is collected as product (Figure 5.3). The final amount of the triethylammonium
salt of 4 is calculated by UV absorption spectra. Since the UV absorption spectra of
minodronate has not yet been reported, an extinction coefficient of the structurally similar
imidazo[1,2-a]pyridine (ε ≈ 7000 M
-1
cm
-1
) was used to determine the final amount,
assuming the two compounds exhibit similar absorption spectra
10
. The final product is
confirmed by
1
H and
31
P NMR spectra and high resolution mass spectroscopy.

N
N
i
5
N
N
O
H
6
N
N
H
N
3
CO
2
Et
N
N
H
NH
2
CO
2
Et
N
N
H
OH
CO
2
Et
7
8
9
N
N
OH
10
CO
2
Et
P
OEt
EtO
O
4
ii iii
v
vi
8
iv
N
N
OH
CO
2
H
P
OH
HO
O
106

Figure 5.3. Preparative HPLC chromatogram of purification of racemic 4 (R
t
= 12.5 min). Conditions:
Dynamax C18 (21.4 mm x 25 cm, 5 µm, 100 Å pore size) column, flow rate 8.0 mL/min, isocratic elution
of 5% CH
3
CN in triethylammonium carbonate buffer (pH 7.5), UV detection at 260 nm.

We were then interested in resolving the component enantiomers of 4. Prontosil
Chiral AX QN-1 and AX QD-1 columns are enantioselective weak anion-exchange
columns that contain silica-based stationary phases covalently bonded to chiral selectors:
a carbamoylated quinine or quinidine, respectively (Figure 5.4)
11
. These columns are well
known for their ability to separate acidic chiral compounds, such as amino acids. An
ionic interaction between the positively charged nitrogen in the quinuclidine moiety of
the chiral selector and a negatively charged chiral acid (thereby forming a transient
diastereomer) is the primary mechanism for separation
11
. The enantiomer that forms a
more stable diastereomer will have a longer retention time than the enantiomer that forms
a less stable diastereomer with the chiral selector of the stationary phase. Beyond the ion-
exchange mechanism, the column may also separate compounds based on typical reverse-
phase HPLC conditions
11
. Furthermore, the chiral acid may also interact with the chiral
selector via hydrogen bonding and π-π interactions. Thus, for optimum separations,
107
several factors may need to be altered, including pH, type and content of organic solvent
and buffer, and flow rate
11, 12
.

Figure 5.4. Chiral stationary phase of Prontosil AX QN-1 (8S, 9R) and QD-1 (8R, 9S) column.

The chiral HPLC method previously utilized to separate 3-PEHPC into its
component enantiomers was then adapted towards the enantiomeric separation of 3-
IPEHPC
7
. The Prontosil AX QN-1 column (8.0 x 150 mm) is eluted with 75% MeOH in
triethylammonium acetate buffer (pH = 5.8) at 3.0 mL/min. At this pH, the phosphonate
and carboxylate groups of our compound will contain negative charges, which will
interact with the positively charged chiral selector of the HPLC column. A sample
chromatogram of these separations is depicted in Figure 5.5, and the last two major
peaks, corresponding to enantiomer “A” and enantiomer “B” (R
t
= 7.2 min and 8.5 min,
respectively), are collected separately. For optimum separation, we found that each
injection sample should consist of 20 µL of a 0.2 M solution of crude 4. Interestingly,
according to the reverse-phase HPLC chromatogram of the crude racemic 4, the
compound appears to be fairly clean, resulting in one major peak corresponding to our
product. In comparison, the chiral HPLC chromatogram depicts several peaks that elute
at shorter retention times than the major 2 peaks corresponding to our product. This
N
OMe
O
H
H
N
O
N
S
8
9
H
108
perhaps indicates that 4 may be experiencing several other types of interactions with the
stationary phase of the chiral column as described above.
Figure 5.5. Chromatogram of preparative chiral reverse-phase HPLC of 4. Conditions: Prontosil AX QN-1
column (8.0 x 150 mm, 5 μm), Flow rate 3.0 mL/min of 75% MeOH in triethylammonium acetate buffer
(pH = 5.85), UV detection at 260 nm.

Purified enantiomers of 4 may also be analyzed for their enantiomeric purity by
utilizing the chiral column Prontosil AX QD-1. This resulted in the reversal of elution
times between the two enantiomers: R
t
for “A” was 6.3 min while “B” now elutes first
with an R
t
of 5.1 min (Figure 5.6, runs 1 and 2, respectively). Since the two enantiomers
slightly overlap each other upon elution from the AX QN-1 column, a small amount of
“A” can be seen in the chromatogram of “B” (Figure 5.6, run 2). Then, to ensure that the
major peaks of each chromatogram did in fact correspond to two separate enantiomers,
the enantiomers were re-mixed and also injected into the HPLC (Figure 5.5, run 3), thus
giving two peaks with retention times that correspond to the separated “A” and “B”
samples.
109
Figure 5.6. Chromatogram of analysis of separated enantiomers of 4 on column with opposite chirality as
AX QN column. Conditions: Prontosil AX QD-1 (4 x 150 mm, 5 μm) column, Flow rate 1.0 mL/min of
triethylammonium acetate buffer (pH=5.85), UV detection at 260 nm. 1) “A” enantiomer (R
t
= 6.3 min) ; 2)
“B” enantiomer (R
t
= 5.1 min); 3) Separated enantiomers re-combined.

Biological data of 4 was obtained by our collaborators in the Rogers laboratory
(University of Aberdeen, UK)
8
. Preliminary data is shown in Figure 5.7. As compared to
the potent N-BP risedronate (RIS, 1), 4 was unable to inhibit prenylation of Ras proteins,
thus indicating that 4 exerts its affects on a different enzymatic target than RIS (Figure
5.7A). Furthermore, 4 exhibited the ability to inhibit prenylation of both Rap1A and Rab6
proteins, which demonstrates the compound’s ability to inhibit Rab GGTase I and/or II
(Figure 5.7A). However, in Figure 5.7B, it can be seen that 4 is able to inhibit the
prenylation of Rab6 at much lower concentrations than Rap1A, thus indicating the
enzymatic target of the drug is most likely Rab GGTase II and perhaps the drug is also a
weaker inhibitor of either GGPPS or GGTase I. This conclusion was recently further
confirmed
3
.
110

Figure 5.7. Preliminary biological data of 3-IPEHPC as a racemate. U = unprenylated; P = prenylated. A)
Western blot comparing RIS (1) inhibition of GTPase proteins at 100 µM versus 3-IPEHPC (4) inhibition
at 400 µM. B) Western blot comparing 3-IPEHPC inhibition of Rab6 and Rap1A proteins at various
concentrations. Sample used prepared as described above by Dr. B. A. Kashemirov.

Additionally, we strived to develop an alternative method towards determining
enantiomeric purity. We initially attempted to resolve the component enantiomers of 3-
PEHPC (2) by using chiral amine-containing reagents, hoping to form diastereomeric
salts observable by NMR. However, solubility issues appeared to hamper this
experiment.
Previous studies were able to achieve enantiomeric separation of chiral
phosphonates and carboxylates by employing a chiral ligand exchange reverse-phase
HPLC system
13, 14
. In ligand exchange chromatography, compounds, which are capable
of donating electrons and coordinating to immobilized metal ions, are separated based on
111
their ability to rapidly and reversibly form metal ion complexes. Buffer components
(typically a chiral amino acid) occupy coordination sites on the metal ion (copper is the
most commonly used due to its ability to form complexes with a wide range of
compounds) until they are displaced by ligands from the sample. Enantiomers of the
sample interact with the metal/chiral amino acid complex, forming diastereomeric ion
pairs that can be separated on typical reverse-phase stationary phases and thus achieve
the resolution of the component enantiomers
13, 14
.
Ligand exchange chromatography works particularly well for samples that are
able to form complexes with metal ions. Compound 2 is thus an ideal candidate for this
methodology due to the presence of the highly ionizable phosphonate and carboxylate
groups that are well-known for their chelating power. After several factors were studied
(including column size, buffer, content of organic solvent, and pH), the following
conditions appeared to give optimum separation of the enantiomers of 2 (Figure 5.8): C18
(5 cm x 4.5 mm, 3 µm, 100 Å) column, flow rate 1.0 mL/min of 2% MeOH in solution of
7 mM L-phenylalanine and 4 mM cupric sulfate (pH 3.3), UV detection at 270 nm. Each
injection contained 20 µL of a solution containing 1 mg of 2 in 1 mL of buffer (4 mM
cupric sulfate and 10 mM L-phenylalanine, pH 3.3). Although this method appeared to
separate the enantiomers (eluting at 5.9 min and 7.6 min), experimental issues with
maintaining a steady baseline may occur. Further optimization of these HPLC conditions
may be necessary to increase detection sensitivity for determining exact enantiomeric
purity of samples.

112
Figure 5.8. Chromatogram of chiral resolution of 3-PEHPC by ligand exchange reverse-phase HPLC.
Conditions: C18 column (5 cm x 4.5 mm, 3 µm, 100 Ǻ) column, flow rate 1.0 mL/min of 2% MeOH in
solution of 7 mM L-phenylalanine and 4 mM cupric sulfate (pH 3.3), UV detection at 270 nm. Injection
sample for 1-3: 20 µL of a solution containing 1 mg 3-PEHPC in 1 mL of buffer (4 mM cupric sulfate and
10 mM L-phenylalanine, pH 3.3). R
t
= 5.9 and 7.6 min.

Conclusion
In summary, phosphonocarboxylate analogues of N-BPs appear to be a promising
new class of therapeutic agents. Our laboratory had previously accomplished the first
synthesis of 3-IPEHPC, and purification of the final product is achieved by reverse-phase
HPLC
8
. Additionally, 3-IPEHPC can be resolved into its component enantiomers by
chiral reverse-phase HPLC, an important separation since it has been shown that the
enantiomers have differing enzymatic activities. In addition to chiral reverse-phase
HPLC, an alternative method for analyzing enantiomeric purity of 3-PEHPC by chiral
ligand exchange reverse-phase HPLC was also studied.

113
Experimental Methods
Reagents and Spectral Measurements. Funding and 3-PEHPC were provided by
Procter & Gamble Pharmceuticals. The synthesis of 3-IPEHPC was performed by Dr.
Isabelle Mallard-Favier and Dr. Boris Kashemirov. Biological data was obtained in the
Rogers laboratory at the University of Aberdeen, UK. Triethylamine (TEA) was distilled
from KOH; all other reagents were purchased from Sigma Aldrich or Alfa Aesar and
used as supplied by the manufacturer. HPLC separations were performed on a Rainan
Dynamax Model SD-200 system with a Rainan Dynamax absorbance detector Model
UV-DII. NMR spectra were recorded on either 400 MHz Varian or 500 MHz Bruker
spectrometers. Chemical shifts are reported relative to internal D
2
O standard (for
1
H
NMR spectra) or 85% H
3
PO
4
(for
31
P NMR spectra). UV spectra were recorded on a DU
800 spectrometer. High resolution mass spectra were performed by Dr. Ron New at UC
Riverside High Resolution Mass Spectrometry Facility on a VG ZAB2SE high resolution
mass spectrometer, with Opus V3.1 and DEC 3000 Alpha Station. For biological
experimental methods, please refer to reference 8. All relevent spectra can be found in
Appendix C.
Purification of 2-hydroxy-3-imidazo[1,2-a]pyridin-3-yl-2-phosphonopropionic acid
(3-IPEHPC, 4) by preparative reverse-phase HPLC. A 0.35-0.4 M solution of crude 4
was purified by semi-preparative HPLC according to the following conditions: sample
injection for each run = 250 µL, Dynamax C18 (21.4 mm x 25 cm, 5 µm, 100 Å pore
size) column, flow rate 8.0 mL/min, isocratic elution of 5% CH
3
CN in triethylammonium
carbonate buffer (pH 7.5), UV detection at 260 nm. The peak eluting at 12.5 min was
114
collected as purified 4 and concentrated under vacuum. The final amount was calculated
from UV absorption spectra (ε = 7000 M
-1
L
-1
)
10
, and the sample was lyophilized until
constant weight is reached.
1
H NMR (D
2
O): δ 3.27-3.36 (brm, 1H), 3.64 (brd, 1H), 7.08-
7.14 (brm, 1H), 7.42 (s, 1H), 7.49-7.57 (m, 2H), 8.51 (d, J = 7.1 Hz, 1H).
31
P{
1
H} NMR
of 3-IPEHPC (D
2
O): 15.24 (s). HRMS (negative mode, FAB): calcd [M - H]
-
= 285.1779
m/z; found [M - H]
-
= 285.0283 m/z.
Resolution of component enantiomers of 4 by chiral HPLC. For optimium separation,
each injection sample consisted of 1.3 mg of crude 4 in 20 µL in 75% MeOH in 0.1 N
triethylammonium acetate buffer. The sample was purified according to the following
conditions: Prontosil AX QN-1 column (8.0 x 150 mm, 5 μm), flow rate 3.0 mL/min,
isocratic elution of 75% MeOH in triethylammonium acetate buffer (pH = 5.86), UV
detection at 260 nm. The peak eluting at 7.2 min was collected as enantiomer “A;”
similarly, the peak eluting at 8.5 min was collected as enantiomer “B.” The two peaks
were kept separate and concentrated under vacuum. Final amount of isolated product was
determined by UV absorption spectra (ε = 7000 M
-1
L
-1
)
10
, and the sample was lyophilized
until constant weight was reached. Enantiomer “A”:
1
H NMR (D
2
O): δ 3.32-3.42 (brm,
1H), 3.66 (brd, 1H), 7.25 (t, J = 6.9 Hz, 1H), 7.51 (s, 1H), 7.62-7.73 (m, 2H), 8.61 (d, J =
7.0 Hz, 1H);
31
P{
1
H} NMR (D
2
O): δ 15.24 (s). Enantiomer “B”:
1
H NMR (D
2
O): δ 3.32-
3.41 (brm, 1H), 3.66 (brd, 1H), 7.25 (t, J = 6.9 Hz, 1H), 7.51 (s, 1H), 7.62-7.73 (m, 2H),
8.61 (d, J = 6.9 Hz, 1H);
31
P{
1
H} NMR (D
2
O): δ 15.24 (s).
Separation of component enantiomers of 3-PEHPC (2) by reverse-phase HPLC. A
20 µL injection sample of a 3.5 mM solution of 3-PEHPC in a buffer of 2% MeOH in 4
115
mM cupric sulfate and 7 mM L-phenylalanine (pH 3.3) was used for each run. HPLC
conditions as follows: C18 (5 cm x 4.5 mm, 3 µm, 100 Å) column, flow rate 1.0 mL/min
of 2% MeOH in solution of 7 mM L-phenylalanine and 4 mM cupric sulfate (pH 3.3),
UV detection at 270 nm.

116
Chapter 5 References
1. Roelofs, A. J.; Thompson, K.; Gordon, S.; Rogers, M. J., Molecular Mechanisms
of Action of Bisphosphonates: Current Status. Clinical Cancer Research 2006, 12, (20,
Pt. 2), 6222s-6230s.
2. Fournier, P. G. J.; Daubine, F.; Lundy, M. W.; Rogers, M. J.; Ebetino, F. H.;
Clezardin, P., Lowering Bone Mineral Affinity of Bisphosphonates as a Therapeutic
Strategy to Optimize Skeletal Tumor Growth Inhibition In vivo. Cancer Research 2008,
68, (21), 8945-8953.
3. Baron, R. A.; Tavare, R.; Figueiredo, A. C.; Blazewska, K. M.; Kashemirov, B.
A.; McKenna, C. E.; Ebetino, F. H.; Taylor, A.; Rogers, M. J.; Coxon, F. P.; Seabra, M.
C., Phosphonocarboxylates Inhibit the Second Geranylgeranyl Addition by Rab
Geranylgeranyl Transferase. Journal of Biological Chemistry 2009, 284, (11), 6861-
6868.
4. Lawson, M. A.; Coulton, L.; Ebetino, F. H.; Vanderkerken, K.; Croucher, P. I.,
Geranylgeranyl Transferase Type II Inhibition Prevents Myeloma Bone Disease.
Biochemical and Biophysical Research Communications 2008, 377, (2), 453-457.
5. Coxon, F. P.; Ebetino, F. H.; Mules, E. H.; Seabra, M. C.; McKenna, C. E.;
Rogers, M. J., Phosphonocarboxylate Inhibitors of Rab Geranylgeranyl Transferase
Disrupt the Prenylation and Membrane Localization of Rab Proteins in Osteoclasts In
Vitro and In Vivo. Bone 2005, 37, (3), 349-358.
6. Coxon, F. P.; Helfrich, M. H.; Larijani, B.; Muzylak, M.; Dunford, J. E.;
Marshall, D.; McKinnon, A. D.; Nesbitt, S. A.; Horton, M. A.; Seabra, M. C.; Ebetino, F.
H.; Rogers, M. J., Identification of a Novel Phosphonocarboxylate Inhibitor of Rab
Geranylgeranyl Transferase that Specifically Prevents Rab Prenylation in Osteoclasts and
Macrophages. Journal of Biological Chemistry 2001, 276, (51), 48213-48222.
7. Kashemirov, B. A.; Marma, M. S.; Mallard, I.; Ebetino, F. H.; Coxon, F. P.;
Rojas, J.; Rogers, M. J.; McKenna, C. E., Chiral HPLC Resolution of Bioactive
Phosphonocarboxylate and Relative Potency of the Enantiomers. Book of Abstracts, 16th
International Conference on Phosphorus Chemistry, Birmingham UK, July 4-6 2004,
PS2-041.
8. Bala, J.; Mallard, I.; Kashemirov, B. A.; Coxon, F. P.; Rogers, M. J.; Ebetino, F.
H.; McKenna, C. E., Chemical Synthesis and Chiral Resolution of the
Phosphonocarboxylate Analogue of Minodronate, a Selective GGTase Inhibitor. Book of
Abstracts, 40th Western Regional Meeting of the American Chemical Society, Anaheim,
CA, United States, January 22-25 2006, WRM-129.
117
9. Ebetino, F. H.; Bayless, A. V.; Dansereau, S. M. Phosphonocarboxylate
Compounds Pharmaceutical Compositions for Treating Abnormal Calcium and
Phosphate Metabolism. US Patent 5760021, June 2, 1998.

10. Tomoda, H.; Hirano, T.; Saito, S.; Mutai, T.; Araki, K., Substituent Effects on
Fluorescent Properties of Imidazo[1,2-a]pyridine-Based Compounds. Bulletin of the
Chemical Society of Japan 1999, 72, (6), 1327-1334.
11. Mandl, A.; Nicoletti, L.; Lammerhofer, M.; Lindner, W., Quinine Versus
Carbamoylated Quinine-based Chiral Anion Exchangers. A Comparison Regarding
Enantioselectivity for N-protected Amino Acids and Other Chiral Acids. Journal of
Chromatography, A 1999, 858, (1), 1-11.
12. Lammerhofer, M.; Gyllenhaal, O.; Lindner, W., HPLC Enantiomer Separation of
a Chiral 1,4-dihydropyridine Monocarboxylic Acid. Journal of Pharmaceutical and
Biomedical Analysis 2004, 35, (2), 259-266.
13. Pack, E. J., Jr.; Bleiberg, B.; Rosenberg, I. E., An HPLC Procedure for the
Enantiomeric Purity of an Optically Active Cytosine Analog Using Ligand Exchange.
Chirality 1990, 2, (4), 275-279.
14. Yan, H.; Row, K. H., Direct Determination of Ofloxacin Enantiomers in Human
Urine by Ligand Exchange Chromatography. Journal of Liquid Chromatography &
Related Technologies 2007, 30, (9&10), 1497-1511.

118
Bibliography

Abuelyaman, A. S.; Boardman, G. S.; Shukla, B. A.; Aasen, S. M.; Mitra, S. B.; Mikulla,
M.; Cinader, D. K., Dental Compositions Including Polymerizable Bisphosphonic Acids.
US 2004206932, October 21, 2004.

Albizu, L.; Teppaz, G.; Seyer, R.; Bazin, H.; Ansanay, H.; Manning, M.; Mouillac, B.;
Durroux, T., Toward Efficient Drug Screening by Homogeneous Assays Based on the
Development of New Fluorescent Vasopressin and Oxytocin Receptor Ligands. Journal
of Medicinal Chemistry 2007, 50, (20), 4976-4985.

Bako, P.; Novak, T.; Ludanyi, K.; Pete, B.; Toke, L.; Keglevich, G., D-Glucose-Based
Aza-Crown Ethers With a Phosphonoalkyl Side Chain: Application as Enantioselective
Phase Transfer Catalysts. Tetrahedron: Asymmetry 1999, 10, (12), 2373-2380.

Bala, J.; Mallard, I.; Kashemirov, B. A.; Coxon, F. P.; Rogers, M. J.; Ebetino, F. H.;
McKenna, C. E., Chemical Synthesis and Chiral Resolution of the Phosphonocarboxylate
Analogue of Minodronate, a Selective GGTase Inhibitor. Book of Abstracts, 40th
Western Regional Meeting of the American Chemical Society, Anaheim, CA, United
States, January 22-25 2006, WRM-129.

Balaz, M.; Sundberg, M.; Persson, M.; Kvassman, J.; Mnsson, A., Effects of Surface
Adsorption on Catalytic Activity of Heavy Meromyosin Studied Using a Fluorescent
ATP Analogue. Biochemistry 2007, 46, (24), 7233-7251.

Baron, R. A.; Tavare, R.; Figueiredo, A. C.; Blazewska, K. M.; Kashemirov, B. A.;
McKenna, C. E.; Ebetino, F. H.; Taylor, A.; Rogers, M. J.; Coxon, F. P.; Seabra, M. C.,
Phosphonocarboxylates Inhibit the Second Geranylgeranyl Addition by Rab
Geranylgeranyl Transferase. Journal of Biological Chemistry 2009, 284, (11), 6861-
6868.

Bhushan, K. R.; Tanaka, E.; Frangioni, J. V., Synthesis of Conjugatable Bisphosphonates
for Molecular Imaging of Large Animals. Angewandte Chemie, International Edition
2007, 46, (42), 7969-7971.

Boisser, S.; Ferreras, M.; Peyruchand, O.; Magnetto, S.; Ebetino, F. H.; Colombel, M.;
Delmas, P.; Delaisse, J.-M.; Clezardin, P., Bisphosphonates Inhibit Breast and Prostate
Carcinoma Cell Invasion, an Early Event in the Formation of Bone Metastases. Cancer
Research 2000, 60, (11), 2949-2954.

Boisser, S.; Magnetto, S.; Frappart, L.; Cuzin, B.; Ebetino, F. H.; Delmas, P. D.;
Clezardin, P., Bisphosphonates Inhibit Prostate and Breast Carcinoma Cell Adhesion to
Unmineralized and Mineralized Bone Extracellular Matrices. Cancer Research 1997, 57,
(18), 3890-3894.
119
Catel, Y.; Degrange, M.; Le Pluart, L.; Madec, P.-J.; Pham, T.-N.; Picton, L., Synthesis,
Photopolymerization and Adhesive Properties of New Hydrolytically Stable Phosphonic
Acids for Dental Applications. Journal of Polymer Science, Part A: Polymer Chemistry
2008, 46, (21), 7074-7090.

Chen, G.; Heim, A.; Riether, D.; Yee, D.; Milgrom, Y.; Gawinowicz Mary, A.; Sames,
D., Reactivity of Functional Groups on the Protein Surface: Development of Epoxide
Probes for Protein Labeling. Journal of the American Chemical Society 2003, 125, (27),
8130-8133.

Chuiko, A. L.; Filonenko, L. P.; Borisevich, A. N.; Lozinskii, M. O., Synthesis and
Properties of Hydroxyaminoalkyl Esters of (Hydroxyethylidene)bis[phosphonic acid].
Zhurnal Obshchei Khimii 1993, 63, (5), 1070-1074.
Coxon, F. P.; Bala, J. L.; Kashemirov, B. A.; Lundy, M. W.; Chen, X.; Xia, Z.; Dunford,
J. E.; Russell, R. G. G.; Roelofs, A. J.; Rogers, M. J.; McKenna, C. E.; Ebetino, F. H.,
Fluorescently Labeled Risedronate and Related Analogs: Design and Evaluation as
Imaging Probes. Bone 2008, 42, (S1), S36.

Coxon, F. P.; Ebetino, F. H.; Mules, E. H.; Seabra, M. C.; McKenna, C. E.; Rogers, M. J.,
Phosphonocarboxylate Inhibitors of Rab Geranylgeranyl Transferase Disrupt the
Prenylation and Membrane Localization of Rab Proteins in Osteoclasts In Vitro and In
Vivo. Bone 2005, 37, (3), 349-358.

Coxon, F. P.; Helfrich, M. H.; Larijani, B.; Muzylak, M.; Dunford, J. E.; Marshall, D.;
McKinnon, A. D.; Nesbitt, S. A.; Horton, M. A.; Seabra, M. C.; Ebetino, F. H.; Rogers,
M. J., Identification of a Novel Phosphonocarboxylate Inhibitor of Rab Geranylgeranyl
Transferase that Specifically Prevents Rab Prenylation in Osteoclasts and Macrophages.
Journal of Biological Chemistry 2001, 276, (51), 48213-48222.

Coxon, F. P.; Thompson, K.; Roelofs Anke, J.; Ebetino, F. H.; Rogers Michael, J.,
Visualizing Mineral Binding and Uptake of Bisphosphonate by Osteoclasts and Non-
resorbing Cells. Bone 2008, 42, (5), 848-860.

Dunford, J. E.; Thompson, K.; Coxon, F. P.; Luckman, S. P.; Hahn, F. M.; Poulter, C. D.;
Ebetino, F. H.; Rogers, M. J., Structure-activity Relationships for Inhibition of Farnesyl
Diphosphate Synthase In Vitro and Inhibition of Bone Resorption In Vivo by Nitrogen-
containing Bisphosphonates. Journal of Pharmacology and Experimental Therapeutics
2001, 296, (2), 235-242.

Durroux, T.; Peter, M.; Turcatti, G.; Chollet, A.; Balestre, M.-N.; Barberis, C.; Seyer, R.,
Fluorescent Pseudo-Peptide Linear Vasopressin Antagonists: Design, Synthesis, and
Applications. Journal of Medicinal Chemistry 1999, 42, (7), 1312-1319.

120
Ebetino, F. H.; Bayless, A. V.; Dansereau, S. M. Phosphonocarboxylate Compounds
Pharmaceutical Compositions for Treating Abnormal Calcium and Phosphate
Metabolism. US Patent 5760021, June 2, 1998.

Ebetino, F. H.; Degenhardt, C. R.; Jamieson, L. A.; Burdsall, D. C., Recent Work on the
Synthesis of Phosphonate-Containing, Bone-Active Heterocycles. Heterocycles 1990, 30,
(2), 855-862.

Ebetino, F. H.; Kaas, S. M.; Francis, M. D.; Nelson, D. G. A.; Janusz, J. M. Quaternary
Nitrgen-containing Phosphonate Compounds, Pharmaceutical Compositions, and
Methods of Treating Abnormal Calcium and Phosphate Metabolism and Methods of
Treating and Preventing Dental Calculus and Plaque. US Patent 5391743, April 30, 1993.

Fournier, P. G. J.; Daubine, F.; Lundy, M. W.; Rogers, M. J.; Ebetino, F. H.; Clezardin,
P., Lowering Bone Mineral Affinity of Bisphosphonates as a Therapeutic Strategy to
Optimize Skeletal Tumor Growth Inhibition In Vivo. Cancer Research 2008, 68, (21),
8945-8953.

Fujisaki, J.; Tokunaga, Y.; Takahashi, T.; Hirose, T.; Shimojo, F.; Kagayama, A.; Hata,
T., Osteotropic Drug Delivery System (ODDS) Based on Bisphosphonic Prodrug. I:
Synthesis and In Vivo Characterization of Osteotropic Carboxyfluorescein. Journal of
Drug Targeting 1995, 3, (4), 273-282.

Gourves, J.-P.; Couthon, H.; Sturtz, G., Synthesis of 3,4-Dihydro-2h-Pyrido[1,2-
B]Isoindol-1-One and 3,4-Dihydro-2h-Pyrido[1,2-B]Pyrrolidin-1-One Functionalized At
the C(6) Position By an Intramolecular Horner-Wadsworth-Emmons Reaction. European
Journal of Organic Chemistry 1999, 1999, (12), 3489-3493.

Gumbleton, M.; Stephens, D. J., Coming Out of the Dark: The Evolving Role of
Fluorescence Imaging in Drug Delivery Research. Advanced Drug Delivery Reviews
2004, 57, (1), 5-15.

Guo, R.-T.; Cao, R.; Liang, P.-H.; Ko, T.-P.; Chang, T.-H.; Hudock, M. P.; Jeng, W.-Y.;
Chen, C. K.-M.; Zhang, Y.; Song, Y.; Kuo, C.-J.; Yin, F.; Oldfield, E.; Wang, A. H.-J.,
Bisphosphonates Target Multiple Sites in Both Cis- and Trans- Prenyltransferases.
Proceedings of the National Academy of Sciences of the United States of America 2007,
104, (24), 10022-10027.

Hirabayashi, H.; Fujisaki, J., Bone-specific Drug Delivery Systems: Approaches Via
Chemical Modification of Bone-seeking Agents. Clinical Pharmacokinetics 2003, 42,
(15), 1319-1330.

121
Hirabayashi, H.; Sawamoto, T.; Fujisaki, J.; Tokunaga, Y.; Kimura, S.; Hata, T.,
Relationship Between Physicochemical and Osteotropic Properties of Bisphosphonic
Derivatives: Rational Design for Osteotropic Drug Delivery System (ODDS).
Pharmaceutical Research 2001, 18, (5), 646-651.

Houghton, T. J.; Tanaka, K. S. E.; Kang, t.; Dietrich, E.; Lafontaine, Y.; Delorme, D.;
Ferreira, S. S.; Viens, F.; Arhin, F. F.; Sarmiento, I.; Lehoux, D.; Fadhil, I.; Laquerre, K.;
Liu, J.; Ostiguy, V.; Poirer, H.; Moeck, G.; Parr, J., Thomas R.; Far, A. R., Linking
Bisphosphonates to the Free Amino Groups in Fluoroquinolones: Preparation of
Osteotropic Prodrugs for the Prevention of Osteomyelitis. Journal of Medicinal
Chemistry 2008, 51, (21), 6955-6969.

Kashemirov, B. A.; Bala, J. L. F.; Chen, X.; Ebetino, F. H.; Xia, Z.; Russell, R. G. G.;
Coxon, F. P.; Roelofs, A. J.; Rogers, M. J.; McKenna, C. E., Fluorescently Labeled
Risedronate and Related Analogues: "Magic Linker" Synthesis. Bioconjugate Chemistry
2008, 19, (12), 2308-2310.

Kashemirov, B. A.; Marma, M. S.; Mallard, I.; Ebetino, F. H.; Coxon, F. P.; Rojas, J.;
Rogers, M. J.; McKenna, C. E., Chiral HPLC Resolution of Bioactive
Phosphonocarboxylate and Relative Potency of the Enantiomers. Book of Abstracts, 16th
International Conference on Phosphorus Chemistry, Birmingham UK, July 4-6 2004,
PS2-041.

Kavanagh, K. L.; Guo, K.; Dunford, J. E.; Wu, X.; Knapp, S.; Ebetino, F. H.; Rogers, M.
J.; Russell, R. G. G.; Oppermann, U., The Molecular Mechanism of Nitrogen-containing
Bisphosphonates as Antiosteoporosis Drugs. Proceedings of the National Academy of
Sciences of the United States of America 2006, 103, (20), 7829-7834.

Kozloff, K. M.; Weissleder, R.; Mahmood, U., Noninvasive Optical Detection of Bone
Mineral. Journal of Bone and Mineral Research 2007, 22, (8), 1208-1216.

Krueger, C. D.; West, P. M.; Sargent, M.; Lodolce, A. E.; Pickard, A. S.,
Bisphosphonate-induced Osteonecrosis of the Jaw. Annals of Pharmacotherapy 2007, 41,
(2), 276-284.

Kutyavin, I. V.; Lokhov, S. G.; Afonina, I. A.; Dempcy, R.; Gall, A. A.; Gorn, V. V.;
Lukhtanov, E.; Metcalf, M.; Mills, A.; Reed, M. W.; Sanders, S.; Shishkina, I.;
Vermeulen, N. M. J., Reduced Aggregation and Improved Specificity of G-rich
Oligodeoxyribonucleotides Containing Pyrazolo[3,4-d]pyrimidine Guanine Bases.
Nucleic Acids Research 2002, 30, (22), 4952-4959.

Lammerhofer, M.; Gyllenhaal, O.; Lindner, W., HPLC Enantiomer Separation of a Chiral
1,4-dihydropyridine Monocarboxylic Acid. Journal of Pharmaceutical and Biomedical
Analysis 2004, 35, (2), 259-266.

122
Lawson, M. A.; Coulton, L.; Ebetino, F. H.; Vanderkerken, K.; Croucher, P. I.,
Geranylgeranyl Transferase Type II Inhibition Prevents Myeloma Bone Disease.
Biochemical and Biophysical Research Communications 2008, 377, (2), 453-457.

Lefevre, C.; Kang, H. C.; Haugland, R. P.; Malekzadeh, N.; Arttamangkul, S.; Haugland,
R. P., Texas Red-X and Rhodamine Red-X, New Derivatives of Sulforhodamine 101 and
Lissamine Rhodamine B with Improved Labeling and Fluorescence Properties.
Bioconjugate Chemistry 1996, 7, (4), 482-489.

Leung, W.-Y.; Cheung, C.-Y.; Yue, S. Modified Carbocyanine Dyes and Their
Conjugates and Their Uses. WO2002026891, April 4, 2002.

Lueftner, D.; Henschke, P.; Possinger, K., Clinical Value of Bisphosphonates in Cancer
Therapy. Anticancer Research 2007, 27, (4A), 1759-1768.

Maalouf, M. A.; Wiemer, A. J.; Kuder, C. H.; Hohl, R. J.; Wiemer, D. F., Synthesis of
Fluorescently Tagged Isoprenoid Bisphosphonates that Inhibit Protein
Geranylgeranylation. Bioorganic & Medicinal Chemistry 2007, 15, (5), 1959-1966.

Mandl, A.; Nicoletti, L.; Lammerhofer, M.; Lindner, W., Quinine Versus Carbamoylated
Quinine-based Chiral Anion Exchangers. A Comparison Regarding Enantioselectivity for
N-protected Amino Acids and Other Chiral Acids. Journal of Chromatography, A 1999,
858, (1), 1-11.

Marma, M. S.; Khawli, L. A.; Harutunian, V.; Kashemirov, B. A.; McKenna, C. E.,
Synthesis of α-Fluorinated Phosphonoacetate Derivatives Using Electrophilic Fluorine
Reagents: Perchloryl Fluoride Versus 1-Chloromethyl-4-Fluoro-1,4-
Diazoniabicyclo[2.2.2]Octane Bis(Tetrafluoroborate) (Selectfluor®). Journal of Fluorine
Chemistry 2005, 126, (11-12), 1467-1475.

Marma, M. S.; Xia, Z.; Stewart, C.; Coxon, F.; Dunford, J. E.; Baron, R.; Kashemirov, B.
A.; Ebetino, F. H.; Triffitt, J. T.; Russell, R. G. G.; McKenna, C. E., Synthesis and
Biological Evaluation of α-Halogenated Bisphosphonate and Phosphonocarboxylate
Analogues of Risedronate. Journal of Medicinal Chemistry 2007, 50, (24), 5967-5975.

Martin, M. B.; Arnold, W.; Heath, H. T., III; Urbina, J. A.; Oldfield, E., Nitrogen-
containing Bisphosphonates as Carbocation Transition State Analogs for Isoprenoid
Biosynthesis. Biochemical and Biophysical Research Communications 1999, 263, (3),
754-758.

Maurel, D.; Kniazeff, J.; Mathis, G.; Trinquet, E.; Pin, J.-P.; Ansanay, H., Cell Surface
Detection of Membrane Protein Interaction with Homogeneous Time-resolved
Fluorescence Resonance Energy Transfer Technology. Analytical Biochemistry 2004,
329, (2), 253-262.
123
Mayer-Kuckuk, P.; Boskey, A. L., Molecular Imaging Promotes Progress in Orthopedic
Research. Bone 2006, 39, (5), 965-977.

McKenna, C. E.; Kashemirov, B. A.; Bala, J. L. F., Synthesis of Drug Conjugates via
Reaction with Epoxide-containing Linkers. US Patent Application 20080312440,
December 18, 2008.

Mende, I.; Hoffmann, P.; Wolf, A.; Lutterbuese, R.; Kopp, E.; Baeuerle, P. A.; de Baey,
A.; Kufer, P., Highly Efficient Antigen Targeting to M-DC8+ Dendritic Cells via
FcgRIII/CD16-specific Antibody Conjugates. International Immunology 2005, 17, (5),
539-547.

Mikami, K.; Itoh, Y.; Yamanaka, M., Fluorinated Carbonyl and Olefinic Compounds:
Basic Character and Asymmetric Catalytic Reactions. Chemical Reviews 2003, 104, (1),
1-16.

Miller, K.; Erez, R.; Segal, E.; Shabat, D.; Satchi-Fainaro, R., Targeting Bone Metastases
with a Bispecific Anticancer and Antiangiogenic Polymer-alendronate-taxane Conjugate.
Angewandte Chemie, International Edition 2009, 48, (16), 2949-2954.

Moszner, N.; Salz, U., Recent Developments of New Components for Dental Adhesives
and Composites. Macromolecular Mateials and Engineering 2007, 292, (3), 245-271.

Moszner, N.; Salz, U.; Zeuner, F.; Zimmermann, J.; Rheinberger, V., Dental Material on
the Basis of Acryloester Phosphonic Acid. EP1374829, January 1, 2002.

Nancollas, G. H.; Tang, R.; Phipps, R. J.; Henneman, Z.; Gulde, S.; Wu, W.; Mangood,
A.; Russell, R. G. G.; Ebetino, F. H., Novel Insights into Actions of Bisphosphonates on
Bone: Differences in Interactions with Hydroxyapatite. Bone 2006, 38, (5), 617-627.

Pack, E. J., Jr.; Bleiberg, B.; Rosenberg, I. E., An HPLC Procedure for the Enantiomeric
Purity of an Optically Active Cytosine Analog Using Ligand Exchange. Chirality 1990,
2, (4), 275-279.

Pinho, S. P.; Macedo, E. A., Solubility of NaCl, NaBr, and KCl in Water, Methanol,
Ethanol, and Their Mixed Solvents. Journal of Chemical and Engineering Data 2005, 50,
(1), 29-32.

Qian, X., Two-Part Self-Adhering Dental Compositions. US Patent 7214726, May 8,
2007.

Reddy, G. V.; Jacobs, H.; Gopalan, A. S.; Barrans, R. E., Jr.; Dietz, M. L.; Stepinski, D.;
Herlinger, A., Synthesis of Symmetrical Methylenebis(Alkyl Hydrogen Phosphonates)
By Selective Cleavage of Methylenebis(Dialkyl Phosphonates) With Morpholine.
Synthetic Communications 2004, 34, (2), 331-344.
124

Rocheblave, L.; Bihel, F.; De Michelis, C.; Priem, G.; Courcambeck, J.; Bonnet, B.;
Chermann, J.-C.; Kraus, J.-L., Synthesis and Antiviral Activity of New Anti-HIV
Amprenavir Bioisosteres. Journal of Medicinal Chemistry 2002, 45, (15), 3321-3324.

Roelofs, A. J.; Coxon, F. P.; Ebetino, F. H.; Bala, J. F.; Kashemirov, B. A.; McKenna, C.
E.; Rogers, M. J., Visualisation of Cellular Uptake and Localisation of Bisphosphonate In
Vivo Using a Fluorescent Analogue of Risedronate. Calcified Tissue International 2008,
82, (S1), S59.

Roelofs, A. J.; Coxon, F. P.; Ebetino, F. H.; Bala, J. L. F.; Kashemirov, B. A.; McKenna,
C. E.; Rogers, M. J., Use of a Fluorescent Analogue of Risedronate to Study Localisation
and Cellular Uptake of Bisphosphonates In Vivo. Bone 2008, 42, (S1), S85.

Roelofs, A. J.; Coxon, F. P.; Lundy, M. W.; Ebetino, F. H.; Bala, J. F.; Kashemirov, B.
A.; McKenna, C. E.; Rogers, M. J., Studying Cellular Uptake and Distribution of
Bisphosphonate In Vivo Using Fluorescently-labelled Analogues of Risedronate. Journal
of Bone and Mineral Research 2008, 23, S20.

Roelofs, A. J.; Thompson, K.; Gordon, S.; Rogers, M. J., Molecular Mechanisms of
Action of Bisphosphonates: Current Status. Clinical Cancer Research 2006, 12, (20, Pt.
2), 6222s-6230s.

Romanenko, V. D.; Kukhar, V. P., Fluorinated Phosphonates: Synthesis and Biomedical
Application. Chemical Reviews 2006, 106, (9), 3868-3935.

Russell, R. G. G.; Watts, N. B.; Ebetino, F. H.; Rogers, M. J., Mechanisms of Action of
Bisphosphonates: Similarities and Differences and Their Potential Influence on Clinical
Efficacy. Osteoporosis International 2008, 19, (6), 733-759.

Salz, U.; Zimmermann, J.; Zeuner, F.; Moszner, N., Hydrolytic Stability of Self-Etching
Adhesive Systems. Journal of Adhesive Dentistry 2005, 7, (2), 107-116.

Sato, M.; Grasser, W.; Endo, N.; Akins, R.; Simmons, H.; Thompson, D. D.; Golub, E.;
Rodan, G. A., Bisphosphonate Action. Alendronate Localization in Rat Bone and Effects
on Osteoclast Ultrastructure. Journal of Clinical Investigation 1991, 88, (6), 2095-2105.

Slavik, J., Fluorescent Probes in Cellular and Molecular Biology. CRC Press, Inc.: Boca
Raton, FL, 1994.

Terpetschnig, E. A.; Patsenker, L. D.; Tatarets, A. Luminescent Compounds. US Patent
Application 2003724580, November 28, 2003.

125
Thompson, K.; Rogers, M. J.; Coxon, F. P.; Crockett, J. C., Cytosolic Entry of
Bisphosphonate Drugs Requires Acidification of Vesicles After Fluid-phase Endocytosis.
Molecular Pharmacology 2006, 69, (5), 1624-1632.

Tomoda, H.; Hirano, T.; Saito, S.; Mutai, T.; Araki, K., Substituent Effects on
Fluorescent Properties of Imidazo[1,2-a]pyridine-Based Compounds. Bulletin of the
Chemical Society of Japan 1999, 72, (6), 1327-1334.

Waggoner, A., Fluorescent Labels for Proteomics and Genomics. Current Opinion in
Chemical Biology 2006, 10, (1), 62-66.

Walker, B.; Gray, J.; Burns, D. M.; Wang, Q.; Adrian, T. E.; Nichols, D. H.; Murphy, R.
F.; Nelson, J., Carboxyfluorescein and Biotin Neuromedin C Analogs: Synthesis and
Applications. Peptides 1995, 16, (2), 255-61.

Wang, D.; Miller, S. C.; Sima, M.; Kopeckova, P.; Kopecek, J., Synthesis and Evaluation
of Water-Soluble Polymeric Bone-Targeted Drug Delivery Systems. Bioconjugate
Chemistry 2003, 14, (5), 853-859.

Wang, L.; Zhang, M.; Yang, Z.; Xu, B., The First Pamidronate Containing Polymer and
Copolymer. Chemical Communications 2006, 2006, (26), 2795-2797.

White, S. S.; Li, H.; Marsh, R. J.; Piper, J. D.; Leonczek, N. D.; Nicolaou, N.; Bain, A. J.;
Ying, L.; Klenerman, D., Characterization of a Single Molecule DNA Switch in Free
Solution. Journal of the American Chemical Society 2006, 128, (35), 11423-11432.

Woo, S.-B.; Hellstein, J. W.; Kalmar, J. R., Narrative Review: Bisphosphonates and
Osteonecrosis of the Jaws. Annals of Internal Medicine 2006, 144, (10), 753-761.

Yan, H.; Row, K. H., Direct Determination of Ofloxacin Enantiomers in Human Urine by
Ligand Exchange Chromatography. Journal of Liquid Chromatography & Related
Technologies 2007, 30, (9&10), 1497-1511.

Zaheer, A.; Lenkinski, R. E.; Mahmood, A.; Jones, A. G.; Cantley, L. C.; Frangioni, J.
V., In Vivo Near-Infrared Fluorescence Imaging of Osteoblastic Activity. Nature
Biotechnology 2001, 19, (12), 1148-1154.

Zhang, S.; Gangal, G.; Uludag, H., Magic Bullets for Bone Diseases: Progress in Rational
Design of Bone-seeking Medicinal Agents. Chemical Society Reviews 2007, 36, (3), 507-
531.

126
Appendix A

Figure A.1. Analytical HPLC chromatogram of FAM, TLC purified FL-ALN, versus TLC purified FL-
ALN spiked with FAM. Conditions: C18 column (5 cm x 4.5 mm, 3 µm, 100 Ǻ), flow rate 1.0 mL/min of
50% MeOH in citrate-phosphate buffer with 10% tetrabutylammonium dihydrogen phosphate (pH 7.4), UV
detection at 490 nm. Injection samples as follows:
1: 20 µL of 0.5 mg FAM in 200 µL NaHCO3 buffer (R
t
(min) = 0.63, 0.85, 1.28, 1.46).
2: 20 µL of TLC purified reaction mixture (R
t
(min) =1.48 and 1.84).
3: 20 µL of TLC purified reaction mixture spiked with FAM (R
t
(min) =1.19, 1.41 and 1.81).

127

Figure A.2. Chromatograms of semi-preparative HPLC purification of 5(6)-FAMRIS, 11. Conditions:
Dynamax C18 (21.4 mm x 25 cm, 5 μm, 100 Ǻ pore size) column, flow rate 8.0 mL/min, UV detection at
260 nm, gradient as follows: 10% MeOH 0.1 N TEAAc or TEAC (pH 7) to 40% of 75% MeOH 0.1 N
TEAAc or TEAC (pH 7) in 12 min, increasing to 70% of Buffer B in 100 min.
1: Reaction mixture after size exclusion chromatography.
2: Reaction mixture after TLC purification (eluant, TEAAc buffer).
3: Reaction mixture after HPLC purification (eluant, TEAC buffer).

128

Figure A.3. Chromatograms for semi-preprative HPLC purifications of FAM-labeled compounds.
Conditions: Dynamax C18 (21.4 mm x 25 cm, 5 μm, 100 Å pore size) column, flow rate 8.0 mL/min, UV
detection at 260 nm, gradient as follows: (Method A, used for 11-13) 10% MeOH 0.1 N TEAAc or TEAC
(pH 7) to 40% of 75% MeOH 0.1 N TEAAc or TEAC (pH 7) in 12 min, increasing to 70% of Buffer B in
100 min or (Method B, used for 14-17) isocratic elution with 20% MeOH 0.1 N TEAC (pH 7) for 12 min,
linearly increasing to 100% of 70% MeOH 0.1 N TEAC (pH 7) in 22 min.
1: 11, 5(6)-FAMRIS.
2: 12, 5-FAMRIS.
3: 14, 5(6)-FAMRISPC.
4: 15, 5-FAMRISPC.
5: 16, 6-FAMRISPC.
6: 17, 5(6)-FAMdRIS.
Unlabeled phosphonate-containing compounds elute at much shorter retention times.

129
Figure A.4. Chromatogram of semi-preparative HPLC purification of RhR-X-RisPC, 18. Conditions:
Beckman Ultrasphere ODS C18 (250 x 10 mm, 5 μm, 80 Å pore size), flow rate 4.0 mL/min, UV detection
at 260 nm, isocratic elution of 20% MeOH in 0.1 N TEAC (pH 7.5) for 4 min, linearly increasing to 100%
of 70% MeOH in 0.1 N TEAC (pH 7.5) in 19 min.

130
Figure A.5. Chromatogram of semi-preparative HPLC purification of AF647RIS, 19. Conditions:
Beckman Ultrasphere ODS C18 (250 x 10 mm, 5 μm, 80 Å pore size), flow rate 4.0 mL/min, UV detection
at 260 nm and 568 nm, isocratic elution of 20 % MeOH in 0.1 N TEAAc for 4 min, linearly increasing to
100% of 70% MeOH in 0.1 N TEAAc in 19 min.

131
Figure A.6. Chromatogram of semi-preparative HPLC purification of AF647RISPC, 20. Conditions:
Dynamax C18 (21.4 mm x 25 cm, 5 µm, 100 Ǻ pore size) column, flow rate 8.0 mL/min, UV detection at
260 nm, isocratic elution with 20% MeOH 0.1 N TEAAc (pH 7) for 15 min, linearly increasing to 100% of
45% MeOH 0.1 N TEAAc (pH 7) in 25 min.

132
Figure A.7.
1
H NMR of 5 (D
2
O, 400 MHz): δ 1.27 (s, 9H), 3.07-3.30 (m, 4H), 3.95-4.03 (m, 1H), 4.18-
4.27 (dd, J = 13.7 Hz, 3.7 Hz, 1H), 4.58-4.65 (part. obscured by HDO, about 1H), 7.75 (t, J = 6.8 Hz, 1H),
8.39 (d, J = 7.8 Hz, 1H), 8.43 (d, J = 6.5 Hz, 1H), 8.65 (s, 1H).

133
Figure A.8.
31
P{
1
H} NMR of 5 (D
2
O, 400 MHz): δ 16.33 (d, J = 21.1 Hz, 1P), 16.55 (d, J = 21.1 Hz, 1P).

134

Figure A.9.
1
H NMR of 6 (D
2
O, 400 MHz): δ 1.30 (s, 9H), 3.02-3.26 (m, 3H), 3.48 (dd, J = 14 Hz, 3.8 Hz,
1H), 3.94-4.04 (m, 1H), 4.22-4.31 (m, 1H), 4.60-4.67 (br, 1H), 7.81 (dd, J = 8.3 Hz, 6.3 Hz, 1H), 8.27-8.32
(m, 1H), 8.48-8.56 (m, 2H).

135

Figure A.10.
31
P{
1
H} NMR of 6 (D
2
O, 400 MHz): δ 14.99 (brs, 2P).

136
Figure A.11.
1
H NMR of 7 (D
2
O, 400 MHz): δ 1.28 (s, 9H), 2.09 (tt, J = 20.9 Hz, 7.5 Hz, 1H), 3.07-3.26
(m, 4H), 3.95-4.03 (brm, 1H), 4.23 (dd, J = 13.9 Hz, 9.9 Hz, 1H), 4.59-4.66 (m. 1H), 7.80 (dd, J = 8.1 Hz,
5.8 Hz, 1H), 8.38 (d, J = 7.0 Hz, 1H), 8.45 (d, J = 5.8 Hz, 1H), 8.66 (s, 1H).

137

Figure A.12.
31
P{
1
H} NMR of 7 (D
2
O, 400 MHz): δ 17.23 (s, 2P).

138
Figure A.13.
1
H NMR of 8 (D
2
O, 400 MHz): 3.01 (t, J = 11 Hz, 1H), 3.31 (d, J = 12.6 Hz, 1H), 3.39 (t, J =
11.6 Hz, 2H), 4.25-4.33 (m, 1H), 4.36-4.44 (br, 1H), 7.88 (t, J = 6.6 Hz, 1H), 8.48 (d, J = 8.0 Hz, 1H), 8.58
(d, J = 5.5 Hz, 1H), 8.77 (s, 1H).

139

Figure A.14.
31
P{
1
H} NMR of 8 (D
2
O, 400 MHz): δ 16.04 (d, J = 27.5 Hz, 1P), 16.40 (d, J = 27.5 Hz, 1P).

140
Figure A.15.
1
H NMR of 9 (D
2
O, 400 MHz): δ 2.92 (m, 1H), 3.20-3.30 (m, 2H), 3.48-3.56 (m, 1H), 4.17-
4.25 (m, 1H), 4.34-4.42 (m, 1H), 4.71-4.76 (m, 1H), 7.89 (dd, J = 8.3 Hz, 6.0 Hz, 1H), 8.36-8.41 (brd, 1H),
8.58-8.62 (brd, 1H), 8.65 (s, 1H).

141

Figure A.16.
31
P{
1
H} NMR of 9 (D
2
O, 400 MHz): δ 12.69 (s, 1P), 12.56 (s, 1P).

142
Figure A.17.
1
H NMR of 10 (D
2
O, 400 MHz): δ 2.18-2.36 (brt, 1H), 2.92 (brt, 1H), 3.14-3.28 (m, 3H),
4.16-4.24 (m, 1H), 4.60 (dd, J = 13.5 Hz, 9.5 Hz, 1H), 7.79 (dd, J = 8.7 Hz, 6.1 Hz, 1H), 8.37 (d, J = 8.2
Hz, 1H), 8.44 (d, J = 6.1 Hz, 1H), 8.64 (s, 1H).

143

Figure A.18.
31
P{
1
H} NMR of 10 (D
2
O, 400 MHz): δ 17.54 (brm, 2P).

144

Figure A.19.
1
H NMR of 11 (D
2
O, 400 MHz): δ 3.33-3.39 (m, 2H), 3.43-3.66 (m, 2H), 4.08-4.42 (m, 3H),
4.72-4.80 (brd, 1H), 6.36-6.48 (m, 4H), 6.93 (d, 2H), 7.09 (s, 1H), 7.43 (s, 0.4 H), 7.68-7.87 (m, 2H), 8.06
(s, 0.6 H), 8.37-8.56 (m, 2H), 8.69-8.78 (2 s, 1H).

145

Figure A.20.
31
P{
1
H} NMR of 11 (D
2
O, 400 MHz): δ 16.01 (brm, 2P).

146
Figure A.21.
1
H NMR of 12 (D
2
O, 400 MHz): δ 3.37 (brt, 2H), 3.57 (dd, J = 14.1 Hz, 6.7 Hz, 1H), 3.61-
3.69 (m, 1H), 4.22-4.31 (m, 1H), 4.36-4.45 (m, 1H), 4.79 (d, 1H), 6.61-6.71 (m, 4H), 7.11 (d, J = 9.2 Hz,
2H), 7.33 (d, J = 8.0 Hz, 1H), 7.56 (s, 0.14H), 7.83 (t, J = 7.2 Hz,1H), 7.91 (d, J = 8.1 Hz, 1H), 8.15 (s,
1H), 8.44 (d, J = 8.3 Hz, 1H), 8.56 (d, J = 5.9 Hz, 1H), 8.74 (s, 1H).

147
Figure A.22.
31
P{
1
H} NMR of 12 (D
2
O, 400 MHz): δ 16.47 (brs).

148
Figure A.23.
1
H NMR of 13 (D
2
O, 400 MHz): δ 3.27-3.37 (m, 2H), 3.44 (dd, J = 14 Hz, 6.9 Hz,1H), 3.58
(dd, J = 14 Hz, 4.9 Hz, 1H), 4.19 (brs, 1H), 4.35 (dd, J = 14 Hz, 9.3 Hz, 1H), 6.60-6.73 (m, 4H), 7.04 (d, J
= 9.4 Hz, 2H), 7.53 (s, 1H), 7.81 (dd, J = 8.2 Hz, 6.4 Hz, 1H), 7.88 (d, J = 8.1 Hz, 1H), 7.95 (dd, J = 8.0
Hz, 1.6 Hz, 1H), 8.43 (d, J = 8.1 Hz, 1H), 8.53 (d, J = 6.5 Hz, 1H), 8.70 (s, 1H).

149
Figure A. 24.
31
P{
1
H} NMR of 13 (D
2
O, 400 MHz): δ 16.51 (brs).

150
Figure A.25.
1
H NMR of 14 (D
2
O, 400 MHz): δ 3.27-3.62 (m, 3H), 4.03-4.40 (m, 2H), 6.42 (m, 4H), 6.92
(dd, J = 9.5 Hz, 3.5 Hz, 2H), 7.10 (d, J = 8.1 Hz, 1H), 7.42 (s, 0.4H), 7.70-7.86 (m, 2H), 8.06 (s, 0.6H),
8.27 (brs, 1H), 8.46-8.63 (m, 2H).

151
Figure A.26.
31
P{
1
H} NMR of 14 (D
2
O, 400 MHz): δ 15.28 (brs, 1P).

152
Figure A.27.
1
H NMR of 15 (D
2
O, 400 MHz): δ 3.39-3.52 (m, 2H), 3.55-3.63 (m, 1H), 4.14-4.23 (m, 1H),
4.30-4.40 (m, 1H), 6.43-6.60 (m, 4H), 7.04 (d, J = 9.0 Hz, 2H), 7.25 (d, J = 8.2 Hz, 1H), 7.77-7.86 (m, 2H),
8.08 (s, 1H), 8.23-8.33 (brs, 1H), 8.45-8.62 (m, 2H).

153
Figure A.28.
1
H NMR of 16 (D
2
O, 400 MHz): δ 3.31-3.45 (m, 2H), 3.47-3.55 (m, 1H), 4.05-4.15 (m, 1H),
4.24-4.34 (m, 1H), 6.45-6.59 (m, 4H), 7.00 (d, J = 9.0 Hz, 2H), 7.48 (s, 1H), 7.71-7.80 (m, 2H), 7.85 (dd, J
= 8.1 Hz, J = 1.7 Hz, 1H), 8.21-8.27 (m, 1H), 8.47- 8.55 (m, 2H).

154
Figure A.29.
1
H NMR of 17 (D
2
O, 400 MHz): δ 2.07- 2.27 (m, 1H), 3.10-3.27 (m, 2H), 3.35 (dd, J = 14.2
Hz, 6.8 Hz, 0.4H), 3.44-3.53 (m, 1H), 3.60 (dd, J = 14.2 Hz, 4.7 Hz, 0.6H), 4.08-4.16 (m, 0.4H), 4.18-4.25
(m, 0.6H), 4.25-4.40 (m, 1H), 4.71-4.77 (m, 1H), 6.40-6.52 (m, 4H), 6.96 (dd, J = 8.1 Hz, 4.1 Hz, 2H), 7.17
(d, J = 8.1 Hz, 0.6H), 7.44 (s, 0.4H), 7.72 (d, J = 8.1 Hz, 0.4H), 7.76-7.86 (m, 2H), 8.07 (d, J = 1.7 Hz, 0.6
H), 8.38 (t, J = 8.7 Hz, 1H), 8.47 (d, J = 5.9 Hz, 1H), 8.53 (d, J = 6.1 Hz, 1H), 8.67-8.71 (2 s, 1H).

155
Figure A.30.
31
P{
1
H} NMR of 17 (D
2
O, 400 MHz): δ 16.51 (brs).

156
Figure A.31.
1
H NMR of 18 (D
2
O, 400 MHz): δ 1.11-1.25 (m, 20H including TEA salt), 1.29-1.39 (m,
2H), 1.41-1.50 (m, 2H), 2.16 (t, J = 7.6 Hz, 2H), 2.97-3.05 (m, 3H), 3.09-3.25 (m, 5H including TEA),
3.27-3.36 (m, 1H), 3.43-3.57 (m, 8H), 3.98-4.06 (m, 1H), 4.16-4.27 (m, 1H), 4.55-4.62 (m, 1H), 6.69 (s,
2H), 6.74-6.88 (m, 4H), 7.47 (d, J = 6.8 Hz, 1H), 7.81 (t, J = 6.8 Hz, 1H), 8.15 (d, J = 7.9 Hz, 1H), 8.32-
8.37 (m, 1H), 8.45-8.53 (m, 2H), 8.55-8.60 (m, 1H).

157

Figure A.32.
31
P{
1
H} NMR of 18 (D
2
O, 40 MHz): 15.33 (s).

158
Figure A.33. Normalized UV absorption spectra of 5(6)-FAM vs. FAM-labeled RIS compounds 11-17. All
samples in 0.1 N phosphate buffer (pH 7.2). Normalized at λ= 492 nm. Spectra taken on DU 800
spectrometer.

159
Figure A.34. Fluorescence emission spectra of 5(6)-FAM vs. FAM-labeled RIS compounds 11-17. All
samples in 0.1 N phosphate buffer (pH 7.2). Sample absorbance normalized to λ = 516 nm. Spectra taken
on Jobin Yvon Horiba FluoroMax-3 Fluorometer.

160

Figure A.35. UV absorption spectra of RhR-X-RisPC, 18. Sample concentration at 3 µM in 0.1 N
phosphate buffer (pH 7.5). Spectra taken on DU 800 spectrometer.

161

Figure A.36. Fluorescence emission spectra of RhR-X-RisPC, 18. Sample concentration at 0.5 µM in 0.1 N
phosphate buffer (pH 7.5). Spectra taken on Jobin Yvon Horiba FluoroMax-3 Fluorometer.

162
Figure A.37. UV absorption spectra of AF647RIS, 19. Sample concentration at 1 µM in 0.1 N phosphate
buffer (pH 7.0). Spectra taken on DU 800 spectrometer.

163
Figure A.38. Fluorescence emission spectra of AF647RIS, 19. Sample concentration at 0.3 µM in 0.1 N
phosphate buffer (pH 7.5). Spectra taken on Jobin Yvon Horiba FluoroMax-3 Fluorometer.

164

Figure A.39. HRMS of 12 (positive ion MALDI): calcd 715.1089 m/z; found [M]
+
= 715.1055 m/z.

165

Figure A.40. HRMS of 13 (positive ion MALDI): calcd 715.1089 m/z; found [M]
+
= 715.1082 m/z.

166

Figure A.41. HRMS of 15 (positive ion MALDI): calcd 679.1324 m/z; found [M]
+
= 679.1356 m/z.

167

Figure A.42. HRMS of 16 (positive ion MALDI): calcd 679.1324 m/z; found [M]
+
= 679.1321 m/z.

168

Figure A.43. HRMS of 17 (positive ion MALDI): calcd 699.1139 m/z; found [M]
+
= 699.1137 m/z.

169

Figure A.44. HRMS of 18 (positive ion MALDI): calcd 975.3148 m/z, found [M-H]
+
= 974.3118 m/z.

170
Figure A.45. MS of 19 (positive ion ESI-MS): calcd 1198.2410 m/z, found [M-H]
+
= 1197.10 m/z.
Corresponding Na
+
salts of 19 can also be observed.

171

Figure A.46. MS of 20 (positive ion ESI-MS): calcd 1162.2645 m/z, found [M-H]
+
= 1162.1 m/z.
Corresponding Na
+
salts of 20 can also be observed.

172
APPENDIX B
Figure B.1.
1
H NMR of 2 (D
2
O, 500 MHz): δ 1.99-2.12 (brm, 2H), 3.44 (J = 7.8 Hz, 2H), 5.59 (d, J = 10
Hz, 1H), 5.99-6.13 (brm, 2H).
173

Figure B.2.
31
P{
1
H} NMR of 2 (D
2
O, 500 MHz): δ 18.61 (s).

174

Figure B.3.
1
H NMR of 4c (CDCl
3
, 400 MHz): δ 1.29-1.41 (brm, 28H), 1.43-1.53 (brm, 2H), 1.57-1.67
(brm, 2H),1.83-1.94 (brm, 4H), 2.16 (tt, J = 25.1, 5.9 Hz, 1H), 3.44 (t, J = 6.4 Hz, 2H), 4.74-4.89 (brm,
4H).

175

Figure B.4.
31
P{
1
H} NMR of 4c (CDCl
3
, 400 MHz): δ 22.06 (s).
176

Figure B.5.
1
H NMR of 5 (CDCl
3
, 400 MHz): δ 1.33-1.43 (brm, 27H), 1.45-1.54 (brm , 2H), 1.65-1.74
(brm, 2H), 1.85-1.94 (brm, 2H), 2.04-2.24 (brm, 2H), 3.45 (t, J = 7 Hz, 2H), 4.77-5.00 (brm, 4H).

177

Figure B.6.
31
P{
1
H} NMR of 5 (CDCl
3
, 400 MHz): δ 13.04 (d, J = 77.5 Hz).
178

Figure B.7.
19
F NMR of 5 (CDCl
3
, 400 MHz): δ -192.352 (tt, J = 73.6, 18.9 Hz).
179

Figure B.8.
1
H NMR of 6 (CDCl
3
, 400 MHz): δ 1.26-1.33 (m, 24H), 1.28-1.65 (m, 4H), 1.93-2.12 (m, 2H),
3.60 (t, J = 7.4 Hz), 4.71-4.86 (m, 4H), 7.61-7.67 (m, 2H), 7.74-7.79 (m, 2H).
180

Figure B.9.
31
P{
1
H} NMR of 6 (CDCl
3
, 400 MHz): δ 12.97 (d, J = 76.4 Hz).
181

Figure B.10.
19
F NMR of 6 (CDCl
3
, 400 MHz): δ -192.67 (tt, J = 73.2, 27.3 Hz).
182

Figure B.11.
1
H NMR of 7 (D
2
O, 400 MHz): δ 1.17-1.31 (m, 4H), 1.42-1.58 (m, 4H), 1.81-2.01 (m, 2H),
2.841 (t, J = 7.3 Hz).
183

Figure B.12.
31
P{
1
H} NMR of 7 (D
2
O, 400 MHz): δ 13.87 (d, J = 69.1 Hz).
184

Figure B.13.
19
F NMR of 7 (D
2
O, 400 MHz): δ -185.98 (tt, J = 74.0, 27.6 Hz).
185

Figure B.14. ESI-MS of 7 (negative ion): calcd [M - H]
-
= 292.1668 m/z; found [M - H]
-
= 292.06 m/z.
186

Figure B.15.
1
H NMR of 1 (D
2
O, 400 MHz): δ 1.15-1.28 (m, 4H), 1.37-1.51 (m, 4H), 1.82-2.02 (m, 2H),
3.12 (t, J = 6.9 Hz, 2H), 5.58 (dd, J = 10.2, 1.7 Hz, 1H), 5.98-6.15 (m, 2H).
187

Figure B.16.
31
P{
1
H} NMR of 1 (D
2
O, 400 MHz): δ 13.93 (d, J = 69.1 Hz).
188

Figure B.17.
19
F NMR of 1 (D
2
O, 400 MHz): δ -185.91 (tt, J = 72.6, 27.4 Hz).
189

Figure B.18. ESI-MS of 1 (negative ion): calcd [M - H]
-
= 346.2142 m/z; found [M - H]
-
= 346.21 m/z.

190
APPENDIX C
Figure C.1.
1
H NMR of racemic 3-IPEHPC (D
2
O, 400 MHz): δ 3.27-3.36 (brm, 1H), 3.64 (brd, 1H),
7.08-7.14 (brm, 1H), 7.42 (s, 1H), 7.49-7.57 (m, 2H), 8.51 (d, J = 7.1 Hz, 1H).

191

Figure C.2.
31
P{
1
H} NMR of 3-IPEHPC (D
2
O, 400 MHz): 15.24 (s).

192
Figure C.3.
1
H NMR of 3-IPEHPC, “A” enantiomer (D
2
O, 400 MHz): δ 3.32-3.42 (brm, 1H), 3.66
(brd, 1H), 7.25 (t, J = 6.8 Hz, 1H), 7.51 (s, 1H), 7.62-7.73 (m, 2H), 8.61 (d, J = 7.1 Hz, 1H).

193

Figure C.4.
31
P{
1
H} NMR of 3-IPEHPC, “A” enantiomer (D
2
O, 400 MHz): δ 15.24 (s).

194
Figure C.5.
1
H NMR of 3-IPEHPC, “B” enantiomer (D
2
O, 400 MHz): δ 3.32-3.41 (brm, 1H), 3.66
(brd, 1H), 7.25 (t, J = 6.9 Hz, 1H), 7.51 (s, 1H), 7.62-7.73 (m, 2H), 8.61 (d, J = 6.9 Hz, 1H).

195

Figure C.6.
31
P{
1
H} NMR of 3-IPEHPC, “B” enantiomer (D
2
O, 400 MHz): δ 15.24 (s).

196
Figure C.7. HRMS of racemic 4 (negative mode, FAB): calcd [M - H]
-
= 285.1779 m/z; found [M -
H]
-
= 285.0283 m/z.

Abstract (if available)

Abstract Bisphosphonates are structural analogues of pyrophosphate with increased hydrolytic stability. The P-C-P backbone of BPs gives rise to two additional side chains, which can be “tuned” to alter the BP’s characteristics. The ionizable phosphonate groups chelate metal ions, such as Ca2+, giving BPs exceptional bone affinity. Thus, BPs have been used for several applications due to their bone-targeting efficiency.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Synthesis of foscarnet-peptide conjugates and detection of pyrophosphate analogues, bisphosphonates and phosphonocarboxylates

PDF

Fluorescent imaging probes of nitrogen-containing bone active drugs: design, synthesis and applications

PDF

Studies of bisphosphate-conjugated fluorescent imaging compounds and 8-oxo-dGTP derivatives

PDF

Design and synthesis of bisphosphonate analogs as inhibitors of farnesyl pyrophosphate synthase

PDF

Design and synthesis of a series of methylenebisphosphonates: a nucleotide analogue toolkit to probe nucleic acid polymerase structure and function

PDF

Syntheses of P-O-C linked foscarnet-peptide conjugates

PDF

Utilization of bisphosphonate drugs in fluorescent imaging and targeted drug delivery

PDF

Novel stereochemical probes for DNA polymerases: nucleoside triphosphate beta,gamma-CXY analogues

PDF

Deoxynucleotide analog probes and model compounds for studying DNA polymerase structure and mechanism: synthesis and evaluation of alkyl-, azido-, and halomethylene bisphosphonate-substituted tri...

PDF

Nucleophilic fluorination of bisphosphonates and its application in PET imaging

PDF

I. Microwave-assisted synthesis of phosphonic acids; II. Design and synthesis of polymerase β lyase domain inhibitors

PDF

Motions and conformations of nucleic acids studied using site-directed spin labeling

Asset Metadata

Creator Bala, Joy Lynn Fabile (author)

Core Title Synthesis of fluorescent conjugates of risedronate and related analogues for bone imaging

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Chemistry

Publication Date 08/03/2009

Defense Date 06/30/2009

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag bisphosphonate,OAI-PMH Harvest

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor McKenna, Charles E. (committee chair), Haworth, Ian S. (committee member), Qin, Peter Z. (committee member)

Creator Email joylynnb@usc.edu,joylynnbala@gmail.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m2452

Unique identifier UC149143

Identifier etd-Bala-3104 (filename),usctheses-m40 (legacy collection record id),usctheses-c127-183651 (legacy record id),usctheses-m2452 (legacy record id)

Legacy Identifier etd-Bala-3104.pdf

Dmrecord 183651

Document Type Dissertation

Rights Bala, Joy Lynn Fabile

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Repository Name Libraries, University of Southern California

Repository Location Los Angeles, California

Repository Email cisadmin@lib.usc.edu