Wayne State University
Wayne State University Dissertations
1-1-2015
Synthesis, Characterization, And Rna Reactivity Of
Amino-Acid-Linked Cisplatin Analogues
Xun Bao
Wayne State University,
Follow this and additional works at: http://digitalcommons.wayne.edu/oa_dissertations
Part of the Chemistry Commons
Recommended Citation
Bao, Xun, "Synthesis, Characterization, And Rna Reactivity Of Amino-Acid-Linked Cisplatin Analogues" (2015). Wayne State
University Dissertations. Paper 1115.
This Open Access Dissertation is brought to you for free and open access by DigitalCommons@WayneState. It has been accepted for inclusion in
Wayne State University Dissertations by an authorized administrator of DigitalCommons@WayneState.
SYNTHESIS, CHARACTERIZATION, AND RNA REACTIVITY OF
AMINO-ACID-LINKED CISPLATIN ANALOGUES
by
XUN BAO
DISSERTATION
Submitted to the Graduate School
of Wayne State University,
Detroit, Michigan
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
2015
MAJOR: CHEMISTRY (BIOCHEMISTRY)
Approved by:
_______________________________
Advisor
Date
_______________________________
________________________________
________________________________
________________________________
DEDICATION
In memory of my grandfather
ii
ACKNOWLEDGMENTS
Pursuing a Ph.D. degree overseas was never an easy decision for me, and the
journey to this dissertation was not that easy either. None of this would happen
without all the support and help from others. I am grateful to all the great people in my
life for their help and company. You have given me endless power and let me know
that I am never alone in this world.
As in chronological order, I want to first thank my parents for their support. No
one in my family has ever lived outside of China. As a single child, I understand you
must have a lot of worries. However, you let me go for the life and career I want and
bury all the worries in your heart. Second, I am heartily thankful to Prof. Christine
Chow for accepting me in her lab. Ever since then, her guidance, support, and
determination have continuously inspired me and led me to this finish line. She is an
amazing mentor who has given me valuable advice in both my career and life. I am
also most grateful to my second advisor, Prof. Mary T. Rodgers, for her generous
support in our collaboration project and her confidence in me. Her guidance and
encouragement has opened a new direction in my career path and provided further
benefits for my future. I also want to express my thanks to all my committee members,
Dr. Tamara Hendrickson, Dr. Charles Winter, and Dr. Ahmad Heydari for all their
constructive criticism and suggestions in every stage of my Ph.D. study.
I am extremely fortunate to have the opportunity to work with a lot of awesome
people in the last five and a half years. I want to first thank my previous lab members,
Keshab, Yogo, Geethi, Papa Nii, Rajesh, Dananjaya, and Heba for their warm
iii
welcome and help when I first joined the group. I also want to thank Danielle, Gayani,
Daya, Hyosuk, Nisansala, Supuni, Prabuddha, and Bett for sharing their laughter and
tears with me. Thanks to Jun for all his assistance with NMR experiments and in
research discussions. I would also like to thank our collaborators, Chenchen, Dr.
Zhihua Yang and Dr. Al-Katib, for their great help and encouragement in our
collaborative projects. I appreciate all the help that I got from the Verani, Rueda, Ahn
and Feig labs, and for permission to use their instruments.
I also want to use this opportunity to thank all the great staff members. First,
thanks to Dr. Brian Shay for help with the LC-MS experiments and to Nestor for
solving all my computer problems. Second, I want to express my thanks to the office
staff, especially Melissa for her patience, help and advice with all the paper work.
I am sincerely grateful to the best friends that I met here, Beixi and Chenchen,
for always being there for me, listening to me, and giving me advice in my life and
research.
Last but not least, I want to express my gratitude to my dearest family. Thanks
again to my parents for always trusting me and respecting my decision. My special
appreciation goes to my husband, Wei, for his unconditional love and support to me
and our family.
Thank you all for being such an important part in my life. Because of you, I am
here and I have become what I am now.
iv
TABLE OF CONTENTS
DEDICATION.…………………………………………………………..…………...ii
ACKNOWLEDGEMENTS……………………………………………..………….iii
LIST OF TABLES…………………………………………………………………..xi
LIST OF FIGURES………………………………………………………………xii
CHAPTER 1 Introduction................................................................................1
1.1 Introduction to RNA........................................................................1
1.2 The ribosome as a potential drug target......................................3
1.2.1 Introduction to ribosomes.........................................................3
1.2.2 The ribosome as a drug target in eukaryotic cells....................5
1.2.3 The ribosome as a drug target in bacteria................................7
1.2.4 The ribosome and antibiotic resistance....................................8
1.3 Introduction to cisplatin................................................................14
1.3.1 The discovery of cisplatin.......................................................14
1.3.2 The structural components of cisplatin...................................15
1.3.3 Pharmacological mechanism of cisplatin...............................16
1.3.4 Cisplatin and RNA..................................................................23
1.3.5 Cisplatin as a chemical probe................................................28
1.3.6 Cisplatin derivatives...............................................................31
1.4 Objectives of this project..............................................................34
v
CHAPTER 2 Synthesis and characterization of amino-acid-linked
cisplatin analogues.......................................................................................37
2.1 Abstract...........................................................................................37
2.2 Introduction....................................................................................37
2.3 Materials and methods..................................................................42
2.3.1 General..................................................................................42
2.3.2 Synthesis
of
glycine-
and
ornithine-linked
cisplatin
analogues..................................................................................43
2.3.3 Nuclear magnetic resonance (NMR) spectroscopy, infrared (IR)
spectroscopy, and mass spectrometry (MS)..............................45
2.4 Results and discussion.................................................................46
2.5 Conclusions....................................................................................53
CHAPTER 3 Binding studies of Oplatin and Gplatin with single RNA
nucleosides...................................................................................................55
3.1 Abstract...........................................................................................55
3.2 Introduction....................................................................................55
3.3 Materials and methods..................................................................58
3.3.1 General..................................................................................58
3.3.2 Ultraviolet (UV) spectroscopy................................................58
3.3.3 Single nucleosides.................................................................59
3.3.4 Monoaquation and bisaquation of Pt complex.......................60
3.3.5 Platination of RNA nucleosides..............................................61
vi
3.3.6 High performance liquid chromatography (HPLC).................62
3.3.7 HPLC standard calibration.....................................................63
3.3.8 HPLC
analysis
of
digested
products
of
23S
rRNA
platination..................................................................................63
3.3.9 Liquid chromatography-mass spectrometry (LC-MS)............64
3.4 Results and discussion.................................................................65
3.4.1 HPLC standard calibration of single nucleosides...................65
3.4.2 Quantitative LC analysis of digested products of 23S rRNA
platination..................................................................................66
3.4.3 Binding studies of Oplatin with single nucleoside...................69
3.4.4 Preliminary product analysis of Oplatin with adenosine.........76
3.4.5 Binding studies of Gplatin with single nucleoside...................80
3.5 Future directions............................................................................83
CHAPTER 4 Product characterization and kinetic studies of Oplatin with
RNA nucleosides...........................................................................................85
4.1 Abstract...........................................................................................85
4.2 Introduction....................................................................................85
4.3 Materials and methods..................................................................88
4.3.1 General..................................................................................88
4.3.2 Product isolation....................................................................89
4.3.3 Mass spectrometry.................................................................89
4.3.4 NMR spectroscopy.................................................................90
vii
4.3.5 Ultraviolet (UV) spectroscopy................................................92
4.4 Results............................................................................................92
4.4.1 Product characterization of Oplatin (2) on the nucleoside
level...........................................................................................92
4.4.1.1 Mass spectrometry analysis............................................92
4.4.1.2 NMR analysis..................................................................99
4.4.1.3 UV spectroscopy analysis..............................................102
4.4.2 Kinetic characterization of Oplatin-nucleoside reactions......104
4.5 Discussion....................................................................................107
CHAPTER 5 Tandem mass spectrometry analysis of Oplatin and
Oplatin-nucleoside adducts.......................................................................110
5.1 Abstract.........................................................................................110
5.2 Introduction..................................................................................111
5.3 Materials and methods................................................................113
5.3.1 General................................................................................113
5.3.2 Sample preparation..............................................................113
5.3.3 FT-ICR mass spectrometry..................................................113
5.3.4 Glycosidic bond survival yield analysis................................114
5.3.5 IRMPD analysis...................................................................116
5.4 Results and discussion...............................................................117
5.4.1 CID studies of Oplatin (2).....................................................117
5.4.2 CID
study
and
glycosidic
bond
stability
analysis
of
2a-nucleoside adducts (5a, 5b and 7)......................................120
viii
5.4.3 IRMPD analysis of 5a, 5b and 7...........................................124
5.5 Conclusions..................................................................................126
CHAPTER 6 Gaussian calculations of Oplatin-nucleoside products.....127
6.1 Abstract.........................................................................................127
6.2 Introduction..................................................................................127
6.3 Methods........................................................................................131
6.4 Results and discussion...............................................................131
6.4.1 Basis set evaluation.............................................................131
6.4.2 Further examination on the coordination mode of glycine to
Pt(II).........................................................................................132
6.4.3 Comparison of monohydrated Oplatin (trans- and cis-2a)
structures.................................................................................135
6.4.4 Structure optimization on possible 1a-nucleoside adducts..136
6.4.5 Structure optimization and IR spectrum computation on
[Pt(Orn)(NH3)(N1-Ado)]+..........................................................142
6.4.6 Structure optimization and IR spectrum computation on
[PtCl(Orn)(N7-Ado)]+ (5b)........................................................148
6.4.7 Structure optimization and IR spectrum computation on
[PtCl(Orn)(N7-Guo)]+ (7)..........................................................156
5.6 Conclusions and future directions.............................................161
CHAPTER 7 General conclusions and future directions.........................165
7.1 General conclusions....................................................................165
ix
7.2 Future directions..........................................................................165
APPENDIX....................................................................................................169
REFERENCES.............................................................................................178
ABSTRACT..................................................................................................212
AUTOBIOGRAPHICAL STATEMENT.........................................................214
x
LIST OF TABLES
Table 1.1. Binding sites, mechanisms of action, and target-related resistance
mechanisms of selected 30S ribosome-targeting antibiotics are
summarized....................................................................................................10
Table 1.2. Binding sites, mechanisms of action, and target-related resistance
mechanisms of selected 50S ribosome-targeting antibiotics are
summarized....................................................................................................11
Table 3.1. Extinction coefficients for RNA nucleosides...................................59
Table 3.2: Retention times of guanosine, adenosine, and major platination
products with 2a and 2b are summarized.......................................................72
Table 4.1. Detailed assignments of the mass spectra of 2a-nucleosides
adducts...........................................................................................................95
Table 5.1. Detailed assignments of the CID results of 5a and 5b..................121
Table 5.2. Detailed assignments of the CID results of 7................................123
xi
LIST OF FIGURES
Figure 1.1. Representative scheme of the central dogma and three types of
RNA ..................................................................................................................2
Figure 1.2. Ribosome composition....................................................................4
Figure 1.3. Chemical structures of ribosome-targeting antibiotics.....................6
Figure 1.4. Overview of selected 30S ribosome-targeting antibiotics on the 30S
subunit ribosome.............................................................................................12
Figure 1.5. Overview of selected 50S ribosome-targeting antibiotics on the 50S
subunit ribosome.............................................................................................13
Figure 1.6. Chemical structures of cisplatin and transplatin............................16
Figure 1.7. Schematic representation of in vivo reactivity and aquation
chemistry of cisplatin.......................................................................................18
Figure 1.8. Cisplatin-DNA cross-links and ratios............................................ 20
Figure 1.9. Crystal structures of three platinum-DNA adducts........................22
Figure 1.10. Cisplatin modes of action............................................................23
Figure 1.11. Sequence and secondary structures of RNAs with cisplatin
coordination sites............................................................................................25
Figure 1.12. Schematic representation of RNA secondary structures.............26
Figure 1.13. Target sites and representative RNA chemical probes...............29
Figure 1.14. Ligand exchange equilibria for cisplatin......................................32
Figure 1.15. Chemical structures of FDA-approved cisplatin-derivative
anticancer drugs..............................................................................................34
Figure 1.16. Project overview..........................................................................36
Figure 2.1. Chemical structures of six Pt-amino acid compounds...................39
Figure 2.2. Possible chemical structures of cis-[PtCl2(Orn)] compounds....... 42
xii
Figure 2.3. Chemical structures of four Pt-amino acid compounds
synthesized.................................................................................................... 43
Figure 2.4. IR spectra of L-ornithine hydrochloride and 2................................47
Figure 2.5. NMR spectra of L-ornithine hydrochloride and 2...........................49
Figure 2.6. Mass analyses of 2........................................................................50
Figure 2.7. IR spectra of glycine and 1............................................................51
Figure 2.8. Mass analyses of 1........................................................................53
Figure 3.1. Probing results with Pt(II)-based complexes on helix 24 (h24) of
16S rRNA........................................................................................................57
Figure 3.2. Chemical structures of adenosine, guanosine, cytidine, and
uridine.............................................................................................................59
Figure 3.3. HPLC standard curves of U, C, A and G.......................................65
Figure 3.4. HPLC analysis of rRNA nucleosides.............................................67
Figure 3.5. HPLC results of time-dependent monoaquated Oplatin (2a)
reaction with adenosine in a 1:1 molar ratio....................................................70
Figure 3.6. HPLC results of time-dependent bisaquated Oplatin (2b) reaction
with adenosine in a 1:1 molar ratio..................................................................71
Figure 3.7 HPLC results of time-dependent monoaquated Oplatin (2a) reaction
with guanosine in a 1:1 molar ratio..................................................................71
Figure 3.8. HPLC results of time-dependent bisaquated Oplatin (2b) reaction
with guanosine in a 1:1 molar ratio..................................................................72
Figure 3.9. HPLC results of a competition reaction of monoaquated Oplatin (2a)
reaction with adenosine and guanosine in a 1:1:1 molar ratio.........................73
Figure 3.10 HPLC results of a competition reaction of bisaquated Oplatin (2b)
reaction with adenosine and guanosine in a 1:1:1 molar ratio.........................74
Figure 3.11. Excess 2a reactions with adenosine or guanosine......................75
xiii
Figure 3.12. Competition reactions of excess monoaquated Oplatin (2a) with
adenosine and guanosine...............................................................................76
Figure 3.13. LC-MS analysis of monoaquated Oplatin (2a) reacted with
adenosine....................................................................................................... 77
Figure 3.14. Structure and calculated mass of 2a, adenosine, and proposed
reaction products.............................................................................................79
Figure 3.15. HPLC results of time-dependent monoaquated Gplatin (1a)
reaction with adenosine in a 1:1 molar ratio....................................................81
Figure 3.16. HPLC results of time-dependent monoaquated Gplatin (1a)
reaction with guanosine in a 1:1 molar ratio....................................................81
Figure 3.17. Competition reaction of monoaquated Gplatin (1a) reaction with
adenosine and guanosine in a 1:1:1 molar ratio is shown. The left figure (A) is
the control without 1a......................................................................................82
Figure 4.1. Optimized structures of cisplatin reaction products with adenine
and guanine....................................................................................................86
Figure 4.2. Schematic for radioactive labeling and primer extension on
platinated rRNA...............................................................................................87
Figure 4.3. HPLC analysis of reactions of monoaquated Oplatin (2a) with
adenosine or guanosine in 1:1 molar ratios.....................................................93
Figure 4.4. MS analysis of 2a-nucleoside reaction products...........................94
Figure 4.5. Structures of hypothesized monofunctional 2a adducts with
guanosine or adenosine..................................................................................96
Figure 4.6 Experimental and simulated spectra of 2a-adenosine minor
products with m/Z 990.1 and 860.2.................................................................98
Figure 4.7. HPLC chromatographs of isolated A1 and A2 fractions................99
Figure 4.8. HMQC analysis of HPLC fractions A1, A2 and G1......................100
Figure 4.9. Ultraviolet spectra of 2a-nucleoside product fractions compared to
methylated nucleosides.................................................................................103
Figure 4.10. Kinetics of 2a-nuceloside reactions...........................................106
xiv
Figure 4.11. Assigned 2a product structures with nucleosides......................108
Figure 5.1. Schematic diagram of glycosidc bond and depurination
reaction.........................................................................................................111
Figure 5.2. CID result of Oplatin (396.0 m/Z).................................................117
Figure 5.3. Comparison of experimental and simulated spectra of Oplatin (2)
fragments......................................................................................................118
Figure 5.4. Possible mechanism of 2 fragmentation......................................119
Figure 5.5. CID results of 5a and 5b (629.1 m/Z)...........................................121
Figure 5.6. The glycosidic bond survival yield of 5a and 5b..........................122
Figure 5.7. CID result of 7 (645.1 m/Z)..........................................................123
Figure 5.8. The glycosidic bond survival yield of 7........................................124
Figure 5.9. IRMPD spectra of cis-[Pt(N1-Ado)(NH3)(Orn)]+, 5b and 7...........125
Figure 6.1. Chemical structure of glycine......................................................133
Figure 6.2. Geometries and relative free energies of [PtCl(Gly)(N7-Gua)]....134
Figure 6.3. Reaction scheme of Oplatin monoaquation and two possible
monoaquated Oplatin structures...................................................................136
Figure 6.4. Geometries and relative energies of trans- and cis-2a................136
Figure 6.5. Geometries and relative energies of [PtCl(Gly)(N1-Ade)]+,
[PtCl(Gly)(N3-Ade)]+, and [PtCl(Gly)(N7-Ade)]+............................................138
Figure 6.6. Geometries and relative energies of [PtCl(Gly)(N3-Gua)]+,
[PtCl(Gly)(O6-Gua)]+ and [PtCl(Gly)(N7-Gua)]+............................................140
Figure 6.7. Geometries and relative energies of [PtCl(Gly)(N1-Ado)]+..........144
Figure 6.8. Geometries and relative energies of [Pt(N1-Ado)(NH3)(Orn)]+....146
Figure 6.9. Comparison of experimental IRMPD data with the predicted IR
spectra on optimized [Pt(N1-Ado)(NH3)(Orn)]+ structures.............................147
xv
Figure 6.10. Geometries and relative energies of [PtCl(Gly)(N7-Ado)]+........150
Figure 6.11. Geometries and relative energies of [PtCl(N7-Ado)(Orn)]+ (5b)
......................................................................................................................151
Figure 6.12. Comparison of experimental IRMPD data with the computed IR
spectra on [PtCl(N7-Ado)(Orn)]+ (5b) structures...........................................153
Figure 6.13. Geometries and relative energies of [PtCl(Gly)(N7-Guo)]+........157
Figure 6.14. Geometries and relative energies of [PtCl(N7-Guo)(Orn)]+ (7)..158
Figure 6.15. Comparison of experimental IRMPD data with the computed IR
spectra on [PtCl(N7-Guo)(Orn)]+ (7) structures.............................................160
xvi
1
CHAPTER 1
Introduction
1.1 Introduction to RNA
Since discovery of the DNA double-helical structure, intensive research
has focused on solving the mysteries of how the genetic information flows from
DNA to cellular components (1). With establishment of the central dogma, RNA
was identified as a key intermediate (Figure 1.1) (2, 3). The role of RNA in the
cell life cycle was first recognized in protein synthesis (1-3). The information is
delivered from DNA to protein by transcription and translation (1-3). Three
types of RNA were first identified. Messenger RNA (mRNA) is a short-lived
RNA species that carries the genetic information transcribed from DNA (1-3).
Transfer RNAs (tRNAs) possess the second genetic code corresponding to
various amino acids (1-3). The tRNA connects the nucleic acid and amino acid
sequences (1-3). The ribosome is a large riboprotein assembly composed of
multiple proteins and RNAs (4-6). The ribosome is the protein synthesis
machinery responsible for translating information encoded by the mRNA to the
polypeptide chain (4-6). The ribosome can be viewed as the working station,
while mRNA is the template, and tRNA carries the building blocks. Recently
obtained crystal structures of ribosomal particles demonstrated unambiguously
that ribosomal RNA (rRNA) is the major actor in protein biosynthesis, since the
active sites exclusively comprise RNA, and the fundamental steps involve
RNA-RNA interactions (5-7).
2
Figure 1.1. A representative scheme of the central dogma (upper) and three types of
RNA (lower) in protein synthesis, namely tRNA, rRNA (PDB: 2I2T and 2I2P) and mRNA,
are shown.
Messanger RNA can also undergo a maturation process, which is
catalyzed by the RNA itself (8, 9). These catalytic RNAs are called ribozymes
(8, 9). Moreover, novel regulatory roles of RNA in both transcription and
translation have been revealed. Small interfering RNAs (siRNAs) and
microRNAs (miRNAs) are 20-30 nucleotide-long, non-translated RNAs with
regulatory functions (10, 11). The common action of siRNA and miRNA in gene
regulation is that they recognize target mRNAs with complementary
sequences, and further direct their destruction or inhibit their utilization in
translation (10-12). Long non-coding RNA (lncRNA) further expanded the
world of regulatory RNAs (13, 14). The lncRNAs are generally more than 200
nucleotides long and regulate genes in a wide variety of important biological
processes (1, 13-15). The biochemical nature of lncRNA regulation is still
poorly understood (1, 13-15). New roles for RNA continue to emerge, and the
3
known biological functions of RNA continue to expand in scope (1, 13-15).
RNA is recognized as an important player in all of molecular biology and is
essential for cell survival.
The linear sequence of RNA is built up from different ribonucleosides
(adenosine, guanosine, cytidine, uridine) and their modified derivatives (16).
RNA folds into a wide range of three-dimensional structures by following a
hierarchical pathway analogous to that observed for proteins (17, 18). The
primary structure dictates the type of secondary structure via complementary
base pairing, which then leads to tertiary structure through multiple interactions
between secondary structures (17, 18). The biochemical function of RNA is
tightly connected to its conformation, and the folded structures create binding
pockets similar to those of proteins for substrate or ligand recognition (19). The
number of reports on detailed structures and functions of RNA, particularly
rRNA, has grown rapidly over the past decade. With its crucial cellular roles
and ability to form substrate-binding pockets, RNA has presented itself as an
ideal target for drug discovery.
1.2 The ribosome as a potential drug target
1.2.1 Introduction to ribosomes
As established in the central dogma, the ribosome was identified as the
protein synthesis machinery (2, 3). Ribosomes are located in the cytoplasm (2,
3). They are composed of multiple proteins and RNAs, with rRNA dominating
4
Figure 1.2. The ribosome is composed of two unequally sized assemblies, one large
and one small subunit. Each subunit is also composed of ribosomal RNAs (bottom, left)
and proteins (bottom, right) (PDB: 2I2T and 2I2P).
the main functional steps (two-thirds of the ribosome is composed of RNA,
Figure 1.2) (20). All ribosomes consist of two unequally sized assemblies, one
large and one small subunit (Figure 1.2) (4). The small subunit is involved in
the initiation step of translation and also controls fidelity in decoding the
genetic information from mRNA (4). The large subunit catalyzes peptide-bond
5
formation, and also guides the newly synthesized protein chain through the
dynamic exit tunnel (4). The rapid and precise assembly of the two subunits is
important for accurate function (21, 22). The ribosome assembles in vivo
through an intersubunit interface that relies heavily on RNA-RNA interactions
(21, 22). The intersubunit bridges contribute to the maintenance and function
of the assembled complex (4). Accurate conformational dynamics of the
ribosome are indispensable for protein biosynthesis, which is necessary for
any organism's survival. It is therefore not surprising that the ribosome has
drawn great attention among various RNAs as an ideal drug target.
1.2.2 The ribosome as a drug target in eukaryotic cells
It is well known that most cancer cells possess hyper-proliferative behavior.
In order to maintain their fast and unbridled growth, cancer cells require a
global increase in protein synthesis. Excessive and hypoactive ribosomes are
generally observed in cancer cells due to deregulated signaling (23, 24). It is
now apparent that targeting the ribosome in cancer cells could be a powerful
strategy in antitumor therapy.
Cycloheximide (CHX) is one of the known glutarimide antibiotics that
targets the eukaryotic ribosome (Figure 1.3A) (25). It inhibits protein
biosynthesis by blocking the elongation phase and eEF2-mediated
translocation (25, 26). As reported in 2010, the Shen and Liu groups used a
6
Figure 1.3. Chemical structures of ribosome-targeting antibiotics are shown. A) CHX
and LTM target the E site of eukaryotic ribosome. B) Streptomycin was the
first-discovered aminoglycoside-type antibiotic. Streptomycin and neomycin are
decoding-site targeting antibiotics. C) Three macrolide-type antibiotics are shown.
The large subunit of bacterial ribosome is their molecular target (27-30). The charges at
pH 7.0 are indicated.
cell-based screening assay to discover a CHX structural analogue,
lactimidomycin (LTM) (Figure 1.3A), with potent antiproliferative activity (26).
Similar to CHX, LTM maintains the glutarimide group, but has an added
7
12-membered macrocycle that was later shown to be essential for its activity
(26). The in vitro study suggested that LTM is a better inhibitor than CHX, even
though they have a common binding site at the E site of 60S rRNA (26).
Further mouse experiments established the potential of LTM as an anticancer
drug targeting tumor cell ribosomes (Figure 1.3A) (26). With increasing
evidence pointing to the connection between protein synthesis and cancer cell
growth, attention on cellular translation is rapidly increasing in cancer therapy
(26). As a major cellular player, the placement of ribosomes on the list of
potential antitumor therapeutic targets is reinforced.
1.2.3 The ribosome as a drug target in bacteria
In the history of antibacterial drug discovery, the ribosome holds its
position as a highly validated and versatile target (31, 32). Approximately half
of the antibiotic classes used in the clinic target ribosomes (31, 32). One of the
most important antibiotic families is the aminoglycosides (Figure 1.3B) (28).
Aminoglycosides were established as antibiotics when streptomycin was
discovered in the 1940s (Figure 1.3B) (33, 34). Streptomycin has been widely
used
in
treating
bacterial
infections,
including
Gram-positive
and
Gram-negative pathogens (33, 34). Its antibacterial mechanism has been well
characterized, and 16S rRNA in the small subunit of bacterial ribosomes was
identified as the molecular target (33).
In addition to aminoglycosides, the macrolide family is also a well-known
8
group of antibiotics that use the bacterial ribosome as the pharmacological
target (Figure 1.3C) (29, 30). Macrolides bind to the large subunit of the
bacterial ribosome and inhibit peptide transfer and translocation (30). Due to
high affinity to bacterial ribosomes combined with high structural conservation
of ribosomes across virtually all of the bacterial families, macrolides
demonstrate broad-spectrum antibacterial activity (30). Ribosome-targeting
antimicrobials have already demonstrated their therapeutic effects in treating
bacterial infections (32). In addition to their toxicity against bacteria, these
compounds also present a surprising selectivity for bacterial targets because
of different transport efficiencies with higher organisms (32).
1.2.4 The ribosome and antibiotic resistance
During the 1960s, antibiotic research reached its golden age. It was once
believed that the bacterial infection was completely defeated by these miracle
small molecules. Unfortunately, due to their excessive and heavy clinical
usage, many bacterial strains have developed resistance to the antibiotics on
the market (35, 36). There has been a steady increase in the emergence of
multi-antibiotic-resistant pathogenic bacteria in the last few decades (35, 36).
In a report released in 2014, the World Health Organization (WHO) warned
that antibacterial resistance (ABR) within a wide range of infectious agents has
become a growing public health threat (37). Furthernore, the WHO announced
that we have entered a “post-antibiotic era” in which infections caused by
9
bacteria, parasites, viruses, and fungi can no longer be treated (37).
Aimed at defeating the drug-resistance threat, numerous efforts have been
taken to find novel antibiotics, and most of the work can be divided into two
directions. The first direction is developing derivatives of existing potent
antibiotics in a given class (36, 38). This strategy offers a safety margin since
the structural core of the developed small molecule is known to be active;
however, it also carries the risk that a resistance mechanism already exists (36,
38). The other direction is finding novel targets, possibly with no cross
resistance, and further using these targets in future rational drug design. Even
though this latter strategy has elements of uncertainty, it still presents great
potential for successful outcomes.
As mentioned in the previous section, the ribosome is not only a validated
target for small-molecule ligands in the anticancer field, but also one of the
main antibiotic targets in bacteria. The bacterial ribosome is composed of three
rRNAs (16S, 23S, and 5S) and about 54 ribosomal proteins (5). Despite its
large size, only a few sites on the ribosome are targeted by currently marketed
antibiotics,as revealed by biochemical studies and crystal structures (Tables
1.1 and 1.2; Figures 1.4 and 1.5) (39). The 30S-targeting antibiotics bind near
the P site and interfere with the initiation step of translation by preventing
interactions between the initiator tRNA and the start codon (Table 1.1 and
Figure 1.4) (39). The other 30S-targeting antibiotics bind mainly to locations
around the A site to inhibit either tRNA delivery or mRNA-tRNA translocation
10
(Table 1.1 and Figure 1.4) (39). The peptidyl-transferase center (PTC) is
where the binding sites for most of the 50S-targeting antibiotics are located
(Table 1.2 and Figure 1.5) (39-41). There are a few exceptions, such as the
orthosomycins, which interact with 23S rRNA H89 and H91, and the
thiopeptides, which bind to H43 and H44 also in the 23S rRNA (39-41).
Table 1.1. Binding sites, mechanisms of action, and target-related resistance
mechanisms of selected 30S ribosome-targeting antibiotics.
Antibiotic
Binding
Inhibition
site
mechanism
edeine
P site
initiation
neomycin
helix 44
translocation
paromomycin
A site
translocation
spectinomycin
helix 34
translocation
streptomycin
A site
translocation
thermorubin
helix 44
initiation
Resistance mechanism
References
target modification
(42)
target modification, target
mutation
target modification, target
mutation
target modification, target
mutation
target modification, target
mutation
not determined
(43, 44)
(45-47)
(48-50)
(48, 51)
(52, 53)
11
Table 1.2. Binding sites, mechanisms of action, and target-related resistance
mechanisms of selected 50S ribosome-targeting antibiotics.
Antibiotic
Binding
Inhibition
Resistance
site
mechanism
mechanism
peptide bond
target modification,
formation
target mutation
References
chloramphenicol
A site
clindamycin,
peptide exit
peptide bond
target modification,
(54, 57, 60, 62,
lincomycin
tunnel
formation
target mutation
63)
peptide exit
nascent chain
target modification,
(55, 57, 59, 62,
tunnel
elongation
target mutation
64, 65)
peptide bond
target modification,
formation
target mutation
erythromycin
linezolid
A site
neomycin
helix 69
translocation
(54-61)
(66-71)
target modification,
(43, 44)
target mutation
puromycin
A site
sparsomycin
A site
thermorubin
helix 69
peptide bond
no target-related
formation
mechanism detected
peptide bond
formation
initiation
(72)
target modification
(55, 72, 73)
not determined
(52, 53)
12
Figure 1.4. Overview is given of the binding sites of selected 30S ribosome-targeting
antibiotics
on
the
30S
subunit
ribosome,
including
spectinomycin
(green),
paromomycin (red), streptomycin (blue) (PDB: 1FJG), neomycin (magenta, PDB:
4GAQ/R/S/U) and edeine (yellow, PDB: 1I95). The h44 region is highlighted in pink for
reference.
13
Figure 1.5. Overview is given of the binding sites of selected 50S ribosome-targeting
ribosome
antibiotics on the 50S subunit ribosome with H69 highlighted in pink for reference. Top:
sparsomycin (red, PDB: 1M90), macrolides (green, PDB: 1K9M), puromycin (blue, PDB:
1Q82),
and
d
neomycin
(magenta,
PDB:
4GAQ/R/S/U)
are
shown.
Bottom:
chloramphenical (red, PDB: 3OFA-D),
3OFA D), clindamycin (green, PDB: 3OFX-0),
3OFX
and
erythromycin (blue, PDB: 3OFO-R)
3OFO
are shown.
14
Although resistance mechanisms have already been observed in some of
the ribosome-targeting antibiotics, most of them have arisen from target
alteration via modification or target mutation (47-78). There is still great
potential in finding novel targeting sites in the structurally diverse yet
sequence-conserved portions of the ribosome. With respect to finding possible
novel drug-binding sites, chemical probes are needed to examine the
structural details of rRNA both in vivo and in vitro. Similarly, investigating how
small charged molecules interact with the ribosome will also be important for
providing insight into future drug design.
1.3 Introduction to cisplatin
1.3.1 The discovery of cisplatin
As an inorganic metal complex, cisplatin was first synthesized in 1844 by
Michel Peyrone (74). More than one hundred years later (1966), an X-ray
crystal structure of cisplatin was published, confirming its cis configuration (75).
Not long after that, Barnett Rosenberg observed the curious phenomenon that
cell division ceased and cell morphology dramatically changed during an
investigation of electric field effects on bacterial cell growth (74, 76). He later
discovered that this effect was caused by a Pt-based compound generated
from the platinum electrode and buffer (74, 76). This work led to the discovery
of the antitumor activity of cisplatin (74, 76). In 1969, Rosenberg and his
coworkers published an article entitiled "Platinum compounds: a new class of
15
potent antitumor agents" in the journal Nature (77). After further evaluation,
cisplatin was successfully pushed through clinical trials and was approved by
the Food and Drug Administration (FDA) in 1978 as the first platinum-based
anticancer drug (74). This officially started the era of platinum anticancer drug
discovery with cisplatin taking the lead.
1.3.2 The structural components of cisplatin
Cisplatin is well known as a square-planar compound with a platinum
metal center (Figure 1.6). As a third row transition metal, Pt has two dominant
valence states, +2 and +4, and, in cisplatin, its charge is +2. Platinum(II) has a
dsp2 hybridization state on its first four empty d orbitals and a square-planar
configuration. Two groups of ligands are attached to Pt: two chlorido ligands,
each with a -1 valence state, that coordinate to Pt in a cis fashion, and two
neutral ammine ligands. Based on the hard-soft acid-base (HSAB) principle,
Pt(II) is a relatively soft acid that can coordinate with soft bases (78). The Clligand is a harder base compared to NH3, which makes its coordination to Pt(II)
relatively weaker. Prior to the action of cisplatin against its cellular targets (this
will be discussed in detail in the following sections), the Cl- ligand must
dissociate from Pt and is therefore referred to as the "leaving group" or "labile
ligand" (74, 76). On the other hand, ammine-ligand coordination with Pt(II) is
more thermodynamically stable, so NH3 is referred to as a "non-leaving group"
or "carrier ligand" (74, 76). Each ligand has its role in determining cisplatin's
16
reactivity and mechanism of action. The cis configuration is also critical to its
anticancer activity (79, 80). During the discovery of cisplatin activity, its trans
isomer, transplatin, was also synthesized and tested for antitumor activity;
however, no comparably useful pharmacological effects were exhibited
(Figure 1.6) (79, 80).
Figure 1.6. The chemical structures of cisplatin and transplatin are shown (74, 76).
1.3.3 Pharmacological mechanism of cisplatin
Cisplatin demonstrates great efficacy in a variety of solid tumors, with over
90% cure rates in testicular and ovarian cancer when tumors are discovered
early (74, 81). The biological mechanism behind its antitumor activity has been
thoroughly investigated since its serendipitous discovery. It is widely accepted
that DNA is the main target (74). This section will focus on the cellular
processes that lead to the formation of cisplatin-DNA adducts and the
subsequent cellular responses, which eventually induce cancer cell death.
The first question that needed to be answered in this field was, how does
cisplatin enter the cancer cell? Although the biological mechanism of cell entry
is not fully understood, it is believed that passive diffusion is a main contributor
(82). After intravenous administration to a cancer patient, cisplatin rapidly
17
diffuses into tissues. The chloride concentration in blood serum is around 100
mM; therefore, the aquation of cisplatin is inhibited and the compound remains
in its neutral form when it reaches the outer membrane of the cancer cell
(83-86). After cisplatin diffuses through the lipid bilayer, it faces the cellular
environment, which has a much lower chloride concentration (around 4-12 mM)
(81, 87). This environment facilitates aquation of cisplatin to its cationic forms,
[PtCl(H2O)(NH3)2]+ and [Pt(H2O)2(NH3)2]2+. Cationic molecules with no
hydrocarbon components are rarely able to diffuse through the biological
membranes of higher organisms; therefore, mono- and bis-aquated cisplatin
cannot diffuse back out of the cell, which results in its intracellular
accumulation (Figure 1.7) (83-86).
Other than passive diffusion, some facilitated or active transport
mechanisms may also contribute to cisplatin cellular uptake (Figure 1.7) (88).
The connection between a copper transporter and cisplatin uptake has been
observed in some transposon mutagenesis experiments (89, 90). Cells lacking
copper uptake protein Ctr1 displayed increased resistance to cisplatin and
analogues; whereas, overexpression of Ctr1 led to greater cisplatin
intracellular accumulation (89-91). Some organic cation transporters (OCT)
were also associated with cisplatin uptake (92).
Once inside the cell, cisplatin is activated by aquation (Figure 1.7). The
mono- and bis-aqua cations are reactive towards various biomolecules with
nucleophilic centers, including DNA, since H2O is a better leaving group than
18
Figure 1.7. A schematic representation of the in vivo reactivity and aquation chemistry
of cisplatin and the obstacles that the drug finds before reaching nuclear DNA is
shown.
Cl- (93). Before aquated cisplatin can reach the DNA in the nucleus, it can
coordinate to many cellular constituents such as thiol-containing proteins,
S-donor ligands, and RNA (93). It is estimated that less than 1% of cellular
accumulated cisplatin actually ends up binding to nuclear DNA (Figure 1.7)
(94). This non-specific targeting of cisplatin to cytoplasmic components may
contribute to its mechanism of cytotoxicity.
As mentioned previously, after 40 years of research, evidence has
19
accumulated that indicates the nuclear DNA as the major biological target of
cisplatin (88). Once it enters the nucleus, because of the high concentration
and the nucleophilicity of the nucleotides, cisplatin will react primarily with DNA
(88).
The
aquated
cisplatin
derivatives,
[PtCl(H2O)(NH3)2]+
or
[Pt(H2O)2(NH3)2]2+, readily modify DNA at the N7 positions of deoxyguanosine
and deoxyadenosine to form monofunctional or bifunctional adducts (88, 95).
The metal binding reaction occurs mainly in the DNA major groove due to the
higher accessibility compared to the minor groove (88, 95).
Research done using DNA repair-deficient cells has provided convincing
evidence that DNA platination is responsible for cisplatin's anticancer activity
(96, 97). As shown in Figure 1.8, the reaction of cisplatin with DNA leads to the
formation of various structural adducts. First, monofunctional adducts are
formed, then most of them further react with a second nucleotide to produce
inter- or intrastrand cross-links (98, 99). Among all of the known DNA-cisplatin
adducts, at least 95% are intrastrand cross-links with 65% 1,2-d(GpG), 25%
1,2-d(ApG), and 5-10% 1,3-d(GpNpG) adducts (100). Minor adducts are also
present in a small percentage, including monofunctional adducts, DNA-DNA
interstrand cross-links and DNA-protein cross-links (81, 87, 88). Cisplatin
almost exclusively coordinates to the N7 position of purine bases, with
preference for deoxyguanosine (101).
Among the various platinum-DNA adducts, intrastrand cross-links have
been the focal point of interest. Formation of cisplatin adducts, especially
20
Figure 1.8. The types of cisplatin-DNA cross-links and their corresponding ratios are
shown. The platinum atom of cisplatin coordinates to the N7 position of purines, with
preference to deoxyguanosine over deoxyadenosine, to form 1,2- or 1,3-intrastrand
cross-links, interstrand cross-links, and monofunctional adducts.
intrastrand cross-links, significantly alters the DNA structure and destabilizes
the DNA by unwinding and bending the duplex (102, 103). The trans isomer of
cisplatin, trans-DDP or transplatin (Figure 1.6), is a clinically ineffective
compound. Due to steric reasons, it is unable to form intrastrandnadducts,
21
especially 1,2-intrastrand cross-links (87). The major cisplatin adduct, the
1,2-intrastrand cross-link, bend the duplex toward the major groove and leave
a widened and shallow minor groove to which several classes of proteins bind
(Figure 1.9A). In contrast, interstand cross-links bend the duplex toward the
minor groove (Figure 1.9C) (104, 105). The structural details of 1,2- and 1,3intrastrand and interstrand cross-link platinum-DNA adducts are illustrated in
Figure 1.9 (104-106).
The presence of a kinetically inert bond between platinum and DNA can
lead to several consequences, such as inhibiting the ability of DNA to function
in replication and transcription (81). These effects then trigger subsequent
cellular responses such as cell-cycle arrest and DNA repair (107).
Cisplatin can induce not just apoptosis, but also the necrosis (Figure 1.10)
(108, 109). Apoptosis and necrosis are different in both morphology and
mechanism (110, 111). Apoptosis, or programmed cell death, generally results
from the activation of the caspase family of aspartate-specific proteases and
occurs as a defense mechanism (110, 111). It is also the main response of
cells to chemotherapeutic agents such as cisplatin (107). Necrosis is
traditionally defined as an accidental cell death caused by traumatic damage
(110, 111). It is often characterized by cytosolic swelling and early loss of
plasma-membrane integrity (110, 111). The mode of cell death induced by
cisplatin is concentration dependent (108). Excessive cisplatin-induced DNA
damage has been shown to hyper-activate the poly(ADP-ribose) polymerase
22
(PARP),
Figure 1.9. The crystal structures of three platinum-DNA adducts are shown. Cisplatin
A) 1,2-d(GpG) intrastrand (PDB: 1A84), B) 1,3-d(GpTpG) intrastrand (1DA4), and C)
interstrand cross-link (1A2E) structures were generated with Pymol. In the left panel,
deoxyguanosine residues are indicated in green, with deoxyadenosines in blue,
deoxycytidine in yellow and deoxythymine in magenta; in the right panel, two strands of
DNA are marked separately in deep teal and pale green. Cisplatin is indicated in red.
23
which cleaves the glycolytic coenzyme NAD+ and provokes the formation of
poly(ADP-ribose) moieties, thereby causing NAD+/ATP depletion, resulting in
necrosis (Figure 1.10) (112).
Figure 1.10. Cisplatin-induced DNA damage triggers two different modes of cell death,
apoptosis and necrosis (TCR is transcription-coupled repair.) (104-109).
1.3.4 Cisplatin and RNA
As an important part of the translation process and a modulator of gene
expression, RNA has critical and essential roles in the life cycle of cells. As a
result, targeting RNA has peaked the interest of scientists as a promising
direction in drug discovery and development. Cisplatin is known to form
24
exchange-inert complexes with DNA, and its antitumor activity is believed to
result from its ability to cause DNA damage (81, 88). As mentioned previously,
it is well established that before cisplatin enters the nucleus, most of the
activated complexes (mono- or bis-aquated species) bind to cytoplasmic
components, such as RNA, and possibly contribute to cytotoxicity (91, 92).
Similar to DNA, RNA is also abundant in nucleophilic nucleotides. As roughly
estimated in reported molecular biology experiments, there are approximately
6 picograms of DNA and 32 picograms of RNA in a mammalian cell (113, 114).
However, compared to DNA, there have been limited studies on the
interactions between RNA and Pt(II)-based compounds.
In the last ten years, there were a growing number of reports on
cisplatin-RNA interactions, which helped to provide evidence that Pt(II)
coordination to cellular RNA may potentiate the established DNA-based
mechanism of cisplatin (115-122). A number of recent studies have
characterized cisplatin reactivity with isolated RNAs by using radioactive
labeling and enzymatic footprinting (115-122). The types of RNAs in these
reports varied, and the sizes ranged from small interfering RNA (~20
nucleotides) to the whole bacterial ribosomal RNA (~4500 nucleotides) (Figure
1.11) (115-122). A biochemical study conducted by Hägerlöf and coworkers
focused on comparing the reactivity of two RNA and DNA hairpins, namely
r(CGCGUUGUUCGCG) and d(CGCGTTGTTCGCG), towards cisplatin and its
derivatives (115). The results indicated that similar to DNA, cisplatin tends to
25
Figure 1.11. The sequence and secondary structures of RNAs in which the coordination
sites of cisplatin have been mapped and identified are summarized. The RNA structures
Ala
that have been studied include A) tRNA
Ala
(left) and truncated models of its acceptor
stem (middle, sMh ) and anticondon loop (right, acMh
Ala
) (117), B) an RNA hairpin
(RNAI) (115), C) a purine-rich internal loop derived from the spliceosome (BBD) (118),
and D) helix 24 of the small subunit ribosomal RNA in E. coli. (119). The binding sites
identified in these RNAs are indicated by stars or labeled with Pt in red.
coordinate with G residues in RNA, and binding to RNA was kinetically
preferred over DNA (Figure 1.11B) (115). Furthermore, the Elmroth group
26
employed full-length tRNAAla as well as hairpin models of its acceptor stem and
anticodon loop in cisplatin reactions, to closely analyze cisplatin's binding
preferences to RNA secondary structures (for examples of RNA secondary
structure types, see Figure 1.12) (116, 117). Combined, these data showed
that coordination of cisplatin to RNA is highly sequence and structure
dependent with a common preference for G-C-rich single-stranded regions
(Figure 1.11A) (116, 117).
Figure 1.12. Schematic representation of RNA secondary structures is given.
Early studies on cisplatin-treated cells or cell extracts showed that cisplatin
27
is able to inhibit important RNA-related processes such as translation and
splicing (123-125). These processes depend heavily on RNA structure and are
mediated by RNA-RNA and RNA-protein interactions (123-125). This finding
was later supported by the discovery of cisplatin coordination to tRNA and
cross-linking to the internal loop of spliceosomal RNAs (Figure 1.11A and C)
(117, 118). DNA-cisplatin interactions and adduct distribution are now well
understood. In contrast, little is known about the number or distribution of
RNA-Pt adducts. In a recent report conducted by DeRose and coworkers,
RNA-Pt adducts from cisplatin-treated Saccharomyces cerevisiae were
quantified (121). The in-cell Pt concentration and Pt accumulation levels were
measured by using inductively coupled plasma mass spectrometry (ICP-MS)
(121). The results revealed that, compared to DNA, cisplatin can accumulate
faster and be retained on RNA, especially rRNA (121). This work suggested
that rRNA can be a target candidate for Pt(II)-based drugs. It was also pointed
out that a single [Pt(NH3)2] adduct on an RNA strand is enough to interfere with
several enzymatic activities, including exonuclease, endonuclease, and
reverse transcriptase (120).
Platinum modification is capable of disrupting the native RNA structure,
causing possible alterations in RNA tertiary structure and enhancing
cytotoxicity in vivo. Comprehensive analysis by our laboratory has
documented cisplatin coordination within the translation machinery (119).
Stable platinum adducts have been mapped to various locations within E. coli
28
rRNA through primer extension and enzymatic digestion of platinated rRNA
(Figure 1.11D). Based on previous findings, we were prompted to carry out
further studies, especially with rRNA, to explore novel targets for Pt(II)-based
drugs.
1.3.5 Cisplatin as a chemical probe
As mentioned in Section 1.2, ribosomal RNA is a validated target for both
anticancer and antibacterial drugs, and it still has great potential for further
exploration. In the interest of finding possible target sites, selective chemical
reagents are needed to probe the structural details of rRNA in vivo and in vitro.
Several chemical probes (Figure 1.13) have been commonly used in RNA
structure studies, such as dimethyl sulfate (DMS) and 2'-hydroxyl acylation
analyzed by primer extension (SHAPE) (126-130). Chemical probing has been
revealed as a great approach to determine the reactivity of individual
nucleotides under near physiological conditions or at different pH or salt
concentrations. Compared to crystallography and NMR methods, chemical
probing is convenient and has a low requirement for sample quantity and
homogeneity. This further allows its application in monitoring the effects of
various solution conditions on RNA conformations. Despite these advantages,
chemical probes still have limitations. Most have poor cell membrane
permeability and are not suitable for in vivo experiments. A few exceptions do
exit, such as DMS and Pb2+ (128, 129). The mechanisms of the chemical
29
Figure 1.13. The target sites (A) and representative chemical probes (B) are shown.
Underlining shows chemical probing detected after RNA strand scission by aniline
treatment, whereas others are detected by direct reverse transcription analysis
(128-131).
probes rely on their reactivity with RNA and their product stability, such that the
newly produced chemical adducts can be detected later with analytical
methods (128). For example, DMS has been used for probing a variety of
RNAs in vivo (128, 130, 131). It has the major experimental convenience that it
can rapidly penetrate the cell and react with RNA without the need for harsh
solution conditions (128, 130, 131). It reacts mainly with adenosine N1 and
cytidine N3 (128, 130, 131). DMS can also react with guanosine N7; however,
since the guanosine N7 is not involved in Watson-Crick base pairing,
methylation at this site does not inhibit reverse transcription, making it difficult
30
to detect through standard biochemical procedures. Aniline treatment is
needed to induce RNA strand scission at these sites before detection (Figure
1.13) (128, 130, 131).
Enzymatic probing is another method commonly used for RNA structure
studies. Each enzyme has unique specificity to cleave RNA in a
structure-dependent or sequence-dependent manner (132). For example,
RNase A cleaves RNA single-stranded regions with UpN and CpN sequences,
and RNase T1 reacts on the 3' side of guanosine residues (131, 133). RNase
V1 is a double-strand-specific enzyme with no sequence preference (131, 133).
Enzymatic probing can detect binding sites of small molecules that could
cause severe steric hinderance or conformational changes that inhibit RNase
activity. Due to the size of these enzymes and their inability to enter the cell
easily, enzymatic probing is limited to in vitro studies.
The use of cisplatin as a probe of RNA was reported by a previous lab
member, Dr. Keshab Rijal. His study focused on Pt(II) complexes, especially
cisplatin, as chemical probes for identifying solvent-accessible nucleotides in
intact bacterial ribosomes in vivo and in vitro (119, 122). Cisplatin forms stable
and inert bonds with RNA, which allows the platinated rRNA to be extracted
and analyzed (119, 122). The rRNA platination sites are identified as "stops" by
radioactive labeling with a primer extension reaction (119, 122). His results not
only provided valuable information on rRNA structural accessibility to charged
small molecules, but also showed RNA-cisplatin interactions in vivo.
31
Platinum-based compounds as chemical probes have specific advantages
such as 1) efficient cell membrane permeability, and 2) easy quenching of the
cisplatin reactivity in high chloride concentrations. Another key advantage of
platinum-based probes is the ease with which their properties such as size and
charge can be altered. Due to cisplatin's preferred reactivity, the structural
information obtained has been limited to G-rich regions. In order to vary the
potential coordination sites and gain further information about accessibility of
rRNA towards various ligands, platinum(II) complexes were modified to
achieve altered reactivity.
1.3.6 Cisplatin derivatives
Cisplatin, as the first FDA-approved Pt(II)-based anticancer drug, has
been presented as belonging to a unique class of antitumor agents. During
early clinical trials, it was soon realized that cisplatin's toxic side effects hinder
its wide application (134-136). In addition, the emergence of drug-resistant
tumor cells has further limited the overall efficiency of cisplatin (137, 138).
Cisplatin's success and limitations prompted a parallel synthetic effort to
design more effective and less toxic as well as orally available platinum
analogues (139).
As discussed in Section 1.3.2, cisplatin has several major structural
components that are critical for its reactivity, such as leaving groups (chlorido),
non-leaving groups (ammine) and cis configuration. These structural features
32
can be strategically modified in designing new platinum anticancer agents.
Since cisplatin is administered as a relatively unreactive complex, its
anticancer mechanism requires aquation of the chlorido ligands to yield
monoaquated and bisaquated complexes (140, 141). The rate-determining
step for cisplatin's initial binding to DNA is hydrolysis of the first chlorido ligand
via association of solvent water (Figure 1.14) (140, 141). Complexes that react
quickly are generally more toxic because of indiscriminate binding to off-target
biological molecules. It has been hypothesized that modification of cisplatin
with less labile leaving groups will reduce or alter its toxicity.
Figure 1.14. Ligand exchange equilibria for cisplatin with values taken from the
literature (140, 141) are shown.
33
In the search for a less toxic agent, carboplatin was designed and
synthesized at the Institute for Cancer Research in the U.K. (Figure 1.15)
(142). Compared to cisplatin, the leaving groups in carboplatin are replaced by
a cyclobutanedicarboxylate (CBDCA) ligand, which is a relatively strong
chelator to Pt(II). Using a murine screen for nephrotoxicity, carboplatin does
indeed demonstrate diminished renal effects and retains antitumor activity
(143, 144). As a result of its lower toxicity profile, carboplatin is able to be
administered at higher doses (145-148). However, since the activated form
(bisaquated) of carboplatin and cisplatin are still the same, they exhibit similar
spectra of activity and cross-resistance (145-149). Carboplatin demonstrates
that modifications on the leaving ligands can result in different toxicity profiles,
indicating that the non-leaving group determines the adduct structures. It has
also been hypothesized that modifications on the non-leaving group could
affect the profile of the resulting Pt(II) adducts and further influence the cellular
repair pathway responses to those adducts (150-152). With the above
understanding and goals, oxaliplatin was developed by substituting the
ammine ligands of cisplatin with 1,2-diaminocyclohexane and the chlorido
ligands with oxalate group (153-156). With different carrier (non-leaving)
ligands, oxaliplatin possesses a different spectrum of activity and displays no
cross-resistance with cisplatin or carboplatin (Figure 1.15) (150, 151, 157).
Over one thousand complexes have been synthesized and tested for better
anticancer profiles than cisplatin. Only carboplatin and oxaplatin have been
34
approved by the FDA for clinical use worldwide (158-160).
Figure 1.15. The chemical structures of the FDA-approved cisplatin-derivative
anticancer drugs are shown. Red-dashed boxes indicate the non-leaving group/carrier
ligands, and the blue-dashed boxes indicate the leaving group/labile ligands.
1.4 Objectives of this project
The rRNA binding sites for currently available ribosome-targeting
antibiotics and potential anticancer drugs are limited to a relatively small region
of the entire ribonuclear complex. The ribosome still has great potential to
provide novel sites for new drug discovery in the cancer field. At the same time,
increasing antibiotic resistance also necessitates the search for new bacterial
target sites and novel drug design. Cisplatin as an active metal compound can
form stable products with RNA, which are detectable by analytical methods.
Probing structured RNA with cisplatin can provide information about potential
“druggable” sites on RNA, but the reactive sites are mostly limited to
35
accessible guanosine sites. Modified cisplatin analogues with altered
functionality are needed to obtain a more comprehensive understanding of the
rRNA structural details as well as RNA's overall interactions with different sized
and charged compounds. In addition, it was reported recently that besides
targeting DNA, cisplatin can accumulate faster and be retained on RNA, likely
causing damage to the RNA structure and potentiating its established
DNA-based mechanism (118, 120, 121). These results promote the
development of RNA-targeting Pt compounds as a promising direction to
overcome DNA repair-related resistance and to reduce dose-related side
effects. For the dual purpose of developing both RNA-targeting chemical
probes and anticancer drugs, understanding the structure-activity relationships
between RNA and Pt compounds is especially important.
The objectives of my Ph.D. work are to: 1) develop cisplatin analogues to
achieve altered reactivity from the parent compound, 2) use model systems to
characterize the product profiles and kinetics of cisplatin derivatives, 3) obtain
information on how ligands alter the Pt(II)-RNA interaction by comparing
analogue structure and reactivity with cisplatin, and 4) develop new
methodologies to further investigate the platination effects on chemical stability
of RNA as well as explore the potential of Pt(II) complexes to be RNA-targeting
anticancer drugs.
The information obtained by using the approaches described above can
help enrich our understanding of the structure-activity relationships (SAR)
36
between Pt(II) complexes and RNA, and further benefit the future development
of both RNA-targeting chemical probes and anticancer drugs (Figure 1.16).
Figure 1.16. The project overview is shown (PDB: 2I2T, 2I2P).
37
CHAPTER 2
Synthesis and characterization of amino-acid-linked cisplatin
analogues
2.1 Abstract
In the last few decades, thousands of platinum-based complexes have
been synthesized and screened for potential biological activity, but only a few
were successful. Based on previous analogue studies, we hypothesize that
modification of cisplatin can lead to compounds that have the potential to
produce different platination profiles than the parent compound. In this chapter,
amino acids were employed as ligands to obtain platinum(II) compounds with
different sizes and charge distributions. The products were then characterized
by NMR spectroscopy, infrared spectroscopy, and mass spectrometry to verify
their chemical compositions and structures before testing them in subsequent
reactions with RNA.
2.2 Introduction
Cisplatin was the first FDA-approved platinum anticancer drug. Its
pharmacological mechanism has been studied in detail, as summarized in the
previous chapter. Treatment with cisplatin is limited by severe side effects,
including nephrotoxicity, neurotoxicity, and emetogenesis, as well as
increasing resistance that is related to systematic DNA repair (161). In order to
overcome these limitations, thousands of cisplatin analogues have been
38
synthesized (74, 162). Among all of the studied analogues, amino-acid-linked
platinum compounds have drawn the most attention for our work.
An amino acid (aa) is the monomer unit of a protein; it has the general
structure of one carboxylate group, one amino group, and a side chain
connected to a chiral carbon (except glycine). With different side chains
properties, amino acids can be divided into four groups: 1) acidic amino acids,
such as aspartic acid (Asp, D) and glutamic acid (Glu, E); 2) basic amino acids,
like lysine (Lys, K), arginine (Arg, R), and histidine (His, H); 3) hydrophobic
ones, with alanine (Ala, A), leucine (Leu, L) and phenylalanine (Phe, F) as
examples; and 4) polar amino acids, such as methionine (Met, M) and cysteine
(Cys, C) (163). There are 22 common amino acids that are naturally abundant,
and even more non-standard amino acids occur in biology due to
post-translational modifications (163). They also serve as metabolic
intermediates. With their great variety and abundance, amino acids comprise a
convenient ligand pool for platinum-amino acid chemistry studies.
Platinum(II) is a relatively soft metal center. Amino acids could coordinate
to platinum through nitrogen, oxygen, or sulfur donors. There is also the
possibility for a single amino acid to coordinate to Pt(II) in a monodentate,
bidentate, or tridentate fashion. In addition, the binding of more than one
amino acid to a single platinum center is possible. All of the potential binding
modes and the possible formation of geometric isomers contribute to the
39
molecular diversity of platinum-amino acid compound libraries, which makes
them ideal candidates for studying ligand effects on RNA-Pt(II) reactivity.
There have only been a few reports on platinum-amino acid compounds
and their properties. In 1994, Lorraine Webster's group reported on an
O-methyl-methionine-linked Pt(II) complex with methionine coordinating to Pt
through sulfur and nitrogen (Figure 2.1A) (164).
Figure 2.1. The chemical structures of six Pt-amino acid compounds are shown. A) Two
methionine-linked Pt compounds are shown, one with a cis configuation (left) and the
other with a trans configuration (right) (164, 165); B) two basic-amino-acid-linked
cisplatin analogues are shown (166-168); C) two histidine-linked cisplatin analogues are
shown (169, 170). The charges at pH 7.0 are indicated.
40
Subsequent cell studies showed no significant activity against leukemia
(164). Almost 20 years later, a methionine-linked Pt(II) trans complex was
reported with improved platination rates with DNA compared to the trans
isomer of cisplatin (Figure 2.1A) (165).
Janina Altman, Meir Wilcheck, Avi Bino, and coworkers published two
articles in the 1980s on amino-acid-linked platinum compounds (166, 171).
Lysine and ornithine were chosen as the ligands and the crystal structures
were determined (Figure 2.1B) (166, 171). Both lysine and ornithine formed a
1:1 Pt:aa bidentate complex through the α-amine and carboxylate group
(referred to as Kplatin and Oplatin, respectively); however, their anticancer
activity was not tested nor mentioned in these papers (166, 171). Later, two
different groups (Lippard's group, in 1997 and Noto's group in 2006) both
reported on the cytotoxicity of lysine- and ornithine-linked cisplatin analogues
(167, 168).
In an attempt to find cisplatin analogues with improved anticancer activities,
Lippard's group screened amino-acid-linked Pt(II) complexes for cytotoxicity in
a solid-phase assay by testing the binding of the corresponding DNA adducts
to the high mobility group (HMG) domain protein (167, 172, 173). The HMG
protein binds specifically to 1,2-intrastrand Pt(II) cross-linked DNA and further
distorts the double-helical structure (174-177). HMG proteins can protect the
cisplatin-induced DNA damage from repair, or be further hijacked by the
platinum adducts from other essential cellular functions (174-177). The
41
screening results suggested the potential of the basic-amino-acid-linked Pt
complexes (Figure 2.1) to be antitumor drug candidates. This activity was
confirmed in toxicity assays (172, 178). Both Kplatin and Oplatin demonstrated
comparable toxicity as cisplatin towards HeLa cells (172, 178).
In 2006, Dalla Via et al. proposed a synthetic pathway for the
ornithine-linked platinum complex, and solved its crystal structure (168). In the
structure, it was revealed that the amino acid coordinates to the platinum
center in a bidentate fashion, and Oplatin maintains a square-planar geometry
similar to that of cisplatin (168). As reported in the above-mentioned literature,
both Kplatin and Oplatin showed comparable antiproliferative activity to
cisplatin (167, 168). These results further drew our attention and interest in
evaluating their reactivity toward RNA and possibly developing them into
RNA-targeting chemical probes or anticancer drugs.
Amino acids have a common structure with one amine group and one
carboxylic acid group, both of which can be electron donors and coordinate to
Pt. As summarized above, many Pt(II) complexes with amino acids have a
common structure, PtCl2(N,O-A) (where A is the amino acid) (164-169). A cis
coordination has been observed with the amino and carboxylate groups
forming a five-membered chelating ring with Pt (166, 171, 172, 179). However,
the chelates of basic amino acids sometimes demonstrate other coordinating
modes due to the nitrogen-containing side chains. Reports have suggested
that histidine can bind to Pt through both the α-amine and the nitrogens of the
42
imidazole ring (Figure 2.1C) (160, 161). Similar to histidine, molecules such as
ornithine may theoretically react through three possible pathways, leading to
the formation of N,O-eight-membered (Figure 2.2, left), N,N-seven-membered
(Figure 2.2, right), or N,O-five-membered (Figure 2.1B, left) using
compounds.
Figure 2.2. The possible chemical structures of the eight-membered (left) and
seven-membered (right) cis-[PtCl2(Orn)] compounds are shown. The charges at pH 7.0
are indicated.
The overall goal of the work reported in this chapter was to optimize the
synthetic scheme for the basic-amino-acid-linked Pt(II) complexes and to verify
their chemical composition and structures (Figure 2.3). The work focused
mainly on the ornithine and glycine complexes although lysine, histidine, and
tyrosine were also examined as potential ligands.
2.3 Materials and methods
2.3.1 General
Potassium tetrachloroplatinate (K2PtCl4) was purchased from Strem
Chemicals, Inc. (Newburyport, MA). L-Ornithine hydrochloride, glycine,
43
aspartic acid and tyrosine were purchased from Alfa-Aesar (Ward Hill, MA). All
other chemicals and reagents were purchased from Sigma (St. Louis, MO) or
Fisher (New York, NY). RNase-free, distilled, deionized H2O (ddH2O) (Millipore,
Billerica, MA) was used for all experiments.
Figure 2.3. The chemical structures of four Pt-amino acid compounds synthesized are
shown. The charges at pH 7.0 are indicated.
2.3.2 Synthesis of glycine- and ornithine-linked cisplatin analogues
Synthesis of cis-[PtCl2(N,O-Gly)](1)
For the glycine-linked platinum complex (1), the general synthesis
procedure was similar to a published synthetic procedure for Oplatin (2), but
was altered and optimized to obtain the best yield of the 1:1 Pt:glycine product
(168). Potassium tetrachloroplatinate (K2PtCl4) (0.070 g, 0.17 mmol) was
44
dissolved in 1 mL ddH2O and stirred at RT for at least 3 h until a red solution
was obtained. Glycine (0.013 g, 0.17 mmole) was added to 1 mL ddH2O and
vortexed
until
dissolved
completely.
The
glycine
and
potassium
tetrachloroplatinate solutions were then mixed in a 1:2 volume ratio (resulting
in a 1:1 molar ratio) and incubated at 70 oC in the dark for 6 h before being
dried under vacuum. The crude product was then washed three times with 90%
ethanol and the remaining precipitate was dried again under vacuum (0.014 g,
0.041 mmol, 24%). νmax/cm-1 (FT-IR, KBr pellet): 3235 (νaNH2), 3139 (νsNH2),
1654 (νaCO2), 1580 (δNH2), 1419 (δCH2), 1362 (νsCO2) and 1315 (δCH2); m/z
(ES-MS-) 338.9.0 [M]-; chemical formular (M): PtC2H4Cl2NO2.
Synthesis of cis-[PtCl2(N,O-Orn)] (2)
The synthesis pathway of the L-ornithine-linked platinum complex (2) is
based on previously published literature (168). Potassium tetrachloroplatinate
(K2PtCl4) (0.070 g, 0.17 mmol) was dissolved in 1 mL ddH2O and stirred at RT
for at least 3 h until a red solution was obtained. L-Ornithine hydrochloride
(0.058 g, 0.34 mmol) was dissolved in 1 mL ddH2O and stirred at RT until a
clear solution was obtained. L-Ornithine and the potassium tetrachloroplatinate
solutions were mixed in a 1:1 volume ratio and the mixture was heated at 50
o
C for 12 h in the dark with gentle shaking until the color of the solution
became dark. Next, the reaction mixture was evaporated to dryness under
vacuum. A yellow powder was obtained, which was washed thoroughly with
45
cold water and dried. Hot H2O was used to dissolve the powder and
recrystallize the product. The crystal was vacuum dried in the dark for 3 days
at room temperature (0.027 g, 0.068 mmol, 40%). νmax/cm-1 (FT-IR, KBr pellet):
3196 (νsNH2), 1643 (νaCO2), 1381 (νsCO2), 1301 (νsCO), 1243 (δNH2), 1180,
1087, 1035 (νCN), 959 (νC-C), 749 (δNH2) and 656 (δCO2); δ/ppm (1H-NMR,
D2O, 400 MHz): 3.48 (1 H, dd, JAB=6 Hz, JAC=12 Hz, -CHCH2-), 2.89 (2 H, t,
-CH2NH3), 1.84 and 1.69 (4 H, m, -CHCH2CH2-); m/z (ES-MS-) 395.98 [M-H]-.
chemical formular (M): PtC5H12Cl2N2O2。
Synthesis of cis-[PtCl2(N,O-Arg)] (3) and cis-[PtCl2(N,O-Tyr)] (4)
Other amino-acid-linked platinum complexes were also synthesized. The
synthesis schemes were altered and tuned differently to achieve the best
purity and yield for each case. For example, in order to improve the solubility,
KOH was added in the synthesis of a tyrosine-linked Pt(II) complex (Yplatin, 4).
For the synthesis of the arginine-linked Pt(II) complex (Rplatin, 3), KCl was
added to prevent the guanidinium group from coordination to a second Pt atom.
However, the synthetic schemes of Yplatin and Rplatin still need to be further
optimized, and this part of the project is being carried out by Bett Kimutai.
2.3.3 Nuclear magnetic resonance (NMR) spectroscopy, infrared (IR)
spectroscopy, and mass spectrometry (MS)
NMR spectroscopy was carried out on a Mercury-400S spectrometer in
46
the Lumigen Instrument Center in the Chemistry Department. All samples were
dissolved in 99.9% D2O, and the signal was locked using D2O. All J values are
given in Hz. All chemical shift values are reported in ppm and referenced to the
solvent peak. Infrared spectra were taken as KBr pellets on a Tensor 27 FT IR
spectrophotometer with the range of 4000-600 cm-1 (Dr. Claudio N. Verani's
Lab, Room 130, Chemistry). Mass spectra were obtained on a 7T Bruker
SolariX FT-ICR mass spectrometer (Dr. Mary T. Rodgers' Lab, Room 10,
Chemistry), and methanol was used as the carrier solvent.
2.4 Results and discussion
The L-ornithine-linked platinum complex (Oplatin, 2) was synthesized and
purified based on Dalla Via's procedure (168). The reaction is shown in
Scheme 1. (40% yield). IR, NMR, and MS were performed to confirm its
structure.
Synthesis
from
starting
material
K2PtCl4
produced
the
cis-PtCl2(N,O-Orn) product 2 in modest yields (40%). It is a one-step, one-pot
reaction done in water. An excess of ornithine was used to obtain the best yield.
The purification step was optimized based on solubility differences between
free ornithine and compound 2 since the latter has poor water solubility
compared to ornithine. The IR and NMR results (discussed in the following
sections) are consistent with the literature, suggesting the successful
generation and isolation of compound 2 (166, 168). MS data further confirmed
the chemical composition of 2.
47
Scheme 1: Synthesis of the ornithine-linked platinum complex (Oplatin, 2)
Figure 2.4. The IR spectra of L-ornithine hydrochloride (upper) and 2 (lower) are shown
(ν = stretching, δ = bending, ζ = torsion, a = asymmetric vibration mode, s = symmetric
vibration mode).
As mentioned in Section 2.2, ornithine theoretically can form three
possible products, containing N,O-eight-membered (Figure 2.2, left),
N,N-seven-membered (Figure 2.2, right), or N,O-five-membered (Figure 2.1B,
left) rings (168); however, formation of the large ring compounds (Figure 2.2)
is expected to be unfavorable. In the IR spectrum of free ligand, L-ornithine
48
(Figure 2.4), a band typical of zwitterionic amino acids was observed at 2078
cm-1 (168). This peak was attributed to a combination of α–NH3+ antisymmetric
bending and torsional oscillation signals. Also, a clear band for COOantisymmetric bending was seen at 1632 cm-1. After the reaction to form 2, the
zwitterionic amino acid band was no longer present in the spectrum,
suggesting that the α-amino group is deprotonated and coordinated to Pt. The
COO- signal of the product shifted from 1632 to 1643 cm-1, a change also
attributed to coordination to Pt through the carboxylate group. The major
changes in the zwitterionic amino acid band and COO- signal suggest that
ornithine coordinates to Pt(II) through the α–NH2 and COO- groups, and this
further suggests that the reaction product is the proposed compound 2. The
presumed five-membered ring structure was expected to localize the electrons
in the acidic oxygen of COO-, causing an increase in the bonding strength of
the carboxylate group. Detailed assignments of the IR signals are summarized
in Figure 2.4.
The 1H-NMR spectra (Figure 2.5) of ornithine and 2 are consistent with the
literature (168). After coordination, a clear peak shift was observed, mainly on
the C1 proton (Δδ = δligand - δcomplex = 3.60 ppm - 3.47 ppm), which suggests
that the C1 proton is located closest to the Pt center. These results further
support the assignment of compound 2 as the five-membered ring structure.
Electron localization effects caused by five-membered ring formation would be
expected to cause the C1 proton to be more shielded than in the free amino
49
acid, causing an upfield shift. Minor chemical shift and peak shape changes
were also observed for the C2 and C3 proton signals. The C4 proton was
affected the least, consistent with a longer distance to the Pt center.
Figure 2.5. The NMR spectra of L-ornithine hydrochloride (upper) and 2 (lower) in D2O
are shown.
In addition to IR and NMR spectroscopy, mass analysis was carried out.
The ESI-MS spectrum of 2 (Figure 2.6) showed a strong, dominant peak with
an m/Z (Z=-1) value of 396.0 and signature Pt isotope distribution, which is in
50
agreement with the calculated exact mass based on the proposed structure of
2 ([m-H]-: calculated 396.0; observed 396.0). This result, combined with the IR
and NMR data, supports the structure of 2, and is consistent with previously
reported compounds and X-ray structures (166, 168). The product was
therefore ready for further reactivity analysis.
Figure 2.6. Mass analyses of 2 are provided. A) The ESI-MS of 2 in negative detection
mode is shown; B) a close-up view of the 395.98 m/Z signal revealed the signature Pt
isotope distribution (upper) and matched the simulated spectrum (lower).
51
Scheme 2: Synthesis of the glycine-linked platinum complex (Gplatin, 1)
Other amino-acid-linked platinum complexes, such as Gplatin (1), were
also characterized by FT-IR and MS to confirm their structures. The reaction
for compound 1 is shown in Scheme 2.
Figure 2.7. The IR spectra of glycine (upper) and 1 (lower) are shown (ν = stretching, δ =
bending, ζ = torsion, a = asymmetric vibration mode, s = symmetric vibration mode).
52
In order to compare the reactivity of different amino-acid-linked platinum
complexes with varying side chains, Gplatin was used as a control compound
in this study because it has no functional group on the Cα position. As
mentioned in Section 2.2.2, different reaction conditions were tested in the
Gplatin (1) synthesis in order to obtain 1:1 Pt:glycine coordination. Compared
to 2, a higher temperature and shorter reaction time was employed. Due to the
competing reaction to form the 1:2 Pt:glycine complex, the overall yield was
very modest (24%). The IR spectrum of the product, 1 (Figure 2.7), showed
the same disappearance of the typical zwitterionic amino acid peak (2118 cm-1
in glycine) as 2, which indicated coordination of the α-amine to the metal
center. In the ligand spectrum (Figure 2.7), the band corresponding to
antisymmetric stretching of the COO- group at 1612 cm-1 was observed, and
shifted to 1654 cm-1 in the product spectrum. This result was suggestive of
five-membered-ring formation through the α-amine and COO- group, which
was similar to Oplatin (2). The MS data provided further evidence for the
structure shown in Figure 2.3. The isotope distribution in electrospray MS
(Figure 2.8) agreed well with the simulated spectrum. The characterization of
1 was therefore in accordance with the proposed structure (Figure 2.3).
Besides Gplatin (1) and Oplatin (2), arginine- (3) and tyrosine-linked (4)
platinum
complexes
were
also
synthesized;
however,
the
synthetic
methodology for the tyrosine-linked platinum complex still needs further
optimization. The need for optimized reaction conditions for each amino acid is
53
likely due to differences in solubility as well as reactivity.
Figure 2.8. Mass analyses of 1 is given. A) The ESI-MS of 1 in negative mode is shown;
B) a close-up view of the 338.93 m/z signal exhibited the signature Pt isotope
distribution (upper) and matched with the simulated spectrum (lower).
2.5 Conclusions
The first goal of this project was to synthesize various amino-acid-linked
cisplatin analogues to achieve altered reactivity. At this point, compounds
Gplatin (1) and Oplatin (2) were synthesized successfully with modest yields
and purities, and were ready for subsequent RNA-reactivity studies (Chapter
3). The MS data for Oplatin (2) provided complementary information on the H
54
atom composition to the reported X-ray structure (168), and further confirmed
the chemical composition. The 1:2 Pt:ornithine complex was not observed
during synthesis and characterization. The preference for the 1:1 Pt:ornithine
complex is believed to be attributed to bulkiness and positive charge of the
basic side chain. In contrast to compound 2, the synthesis of compound 1 was
first reported here and the generation of exclusively the 1:1 complex was much
more challenging. Therefore the product yield of 1 was lower compared to 2.
Despite these challenges, Gplatin (1) serves as a useful control compound in
this project to compare the reactivity of different amino-acid-linked platinum
complexes with varying side chains.
As a hydrophilic and basic amino acid, ornithine can only provide partial
information regarding the basic side-chain effects on platinum reactivity
profiles. In order to fully understand the side-chain effect and expand the
library of Pt(II)-based chemical probes, other closely related amino-acid-linked
cisplatin
analogues
are
needed
for
comparison.
For
example,
an
Nε-dimethylated ornithine platinum complex will give some idea about how
bulky groups (or altering H-bond capabilities) influence compound reactivity
compared to Oplatin (2) (this project is now being continued by Supuni Thalalla
Gamage). Therefore, synthesis, purification, and characterization of different
amino-acid-linked platinum complexes using this methodology will be further
optimized to provide a more complete library in the future.
55
CHAPTER 3
Binding studies of Oplatin and Gplatin with single RNA
nucleosides
3.1 Abstract
The cisplatin anticancer mechanism is based on the ability of platinum(II)
to coordinate to the N7 of purines in DNA. Recent studies showed that RNA
can also be considered as a target; platinum accumulates 10- to 20-fold faster
in RNA compared to DNA, and can form stable products with RNA. In this
chapter, both qualitative and quantitative analysis methods were developed to
study Pt complex reactions with different RNA nucleosides by using high
performance liquid chromatography (HPLC). The reactivities of Gplatin and
Oplatin with adenosine and guanosine were evaluated and compared with
cisplatin using this method. The preliminary product profile of Oplatin with RNA
nucleosides
was
also
determined
by
liquid
chromatography-mass
spectrometry (LC-MS).
3.2 Introduction
Since the initial discovery of cisplatin's anticancer activity, its mechanism
of toxicity has been well established and commonly believed to involve DNA
coordination and formation of irreversible products (81, 180). Recently
published work from our laboratory and others indicates that cisplatin has
56
comparable or even better reactivity towards RNA, raising the possibility of
RNA as a secondary drug target (119, 121). Furthermore, utilizing cisplatin as
a chemical probe provided valuable structural information for rRNA (122);
however, this information was limited to G-rich regions due to cisplatin's
preferred reactivity. Several amino-acid-linked cisplatin analogues were
synthesized (Chapter 2) and their reactivity towards the helix 24 (h24) region
of the small subunit rRNA (16S rRNA in bacteria) was monitored (122). The
h24 region was shown previously to be a strong target site of cisplatin (119,
181). Syntheses of Dplatin, Kplatin and Rplatin were done by K. Rijal (122).
The results with amino-acid derivatives of cisplatin demonstrated that with a
different distribution of positive charge compared to the parental compound
cisplatin, as well as different sizes and hydrogen-bonding capabilities, altered
reactivity occurred (122,181). Cisplatin preferred to react with consecutive Gs,
as well as Gs in mismatched or loop regions of h24 (122, 181). In contrast,
Dplatin (aspartic acid-linked cisplatin analogue) had no significant reactivity,
and both Kplatin and Oplatin exhibited a sequence preference for A residues
and a structural preference for single-stranded regions (122, 181). Only one
coordination site was detected with Rplatin, which was a three-nucleotide
bulge region (A780, A781 and A782, Figure 3.1) (122, 181).
Among the above amino-acid-linked cisplatin analogues, Oplatin (2) has
drawn the most attention and has potential as a dual structure- and
sequence-specific RNA-targeting compound because of its distinct reactivity
57
Figure 3.1. Probing results with Pt(II)-based complexes on helix 24 (h24) of 16S rRNA
are shown (K. Rijal) (119, 120, 122, 181). A) Autoradiogram of probing results with
cisplatin is shown. B) Autoradiogram of probing result with four amino-acid-linked Pt(II)
complexes is shown. C) The chemical structures of the amino-acid-linked complexes in
monoaquated forms are given. D) The secondary structure map of 16S rRNA at h24 is
shown. The reverse transcription stops used to identify platinum sites occur one
nucleotide prior to the coordination site. Strong reactive sites with cisplatin (
and Kplatin (
), and Rplatin (
) are indicated.
), Oplatin
58
from cisplatin. Previous studies focused on identification of the platination
sites by using methods such as radioactive labeling and enzymatic footprinting
(117, 120, 121, 182); however, these techniques provide little information
about how the ligands affect Pt(II) complex reactivity on RNA at the nucloeside
level. With the unique reactivity of monoaquated Oplatin on full-length RNAs,
evaluating its nucleoside reactivity will provide an improved understanding of
its capability as a higher-order structure probe. In order to examine more
closely the nucleoside preferences of the amino-acid-linked analogues,
reactions between the monoaquated Pt(II) complexes and nucleosides were
monitored by HPLC.
3.3 Materials and methods
3.3.1 General
HPLC grade acetonitrile, ammonium acetate (NH4OAc), sodium chloride
(NaCl), sodium dihydrogen phosphate (NaH2PO3), disodium hydrogen
phosphate (Na2HPO3), silver nitrate (AgNO3), N,N-dimethyl formamide (DMF),
adenosine (A), guanosine (G) and all the other chemicals and reagents were
purchased from Sigma-Aldrich or Fisher, unless otherwise stated. RNase-free,
distilled, deionized ddH2O (Millipore water) was used throughout the
experiments.
3.3.2 Ultraviolet (UV) spectroscopy
59
An Aviv 14DS UV-visible spectrometer in Dr. David Rueda's laboratory was
used for all UV measurements.
3.3.3 Single nucleosides
Single RNA nucleosides, guanosine and adenosine (structures and
numbering are shown in Figure 3.2) were dissolved in water at room
temperature. The concentration of each solution was determined by UV-visible
spectroscopy. The extinction coefficients used for the A, G, U, C nucleosides
are shown in Table 3.1 (16).
Figure 3.2. The chemical structures of adenosine, guanosine, cytidine, and uridine are
shown.
Table 3.1. Extinction coefficients for RNA nucleosides
-1
-1
Nucleoside
pH
λmax (nm)
εmax (1/mol cm )
A
6
260
14,900
G
6
253
13,600
U
7
262
10,100
C
7
271
9,100
ref (16)
60
After the nucleoside stock solutions were made, pH 6 or 7 sodium
phosphate buffer (100 mM) was used in the dilution step (Table 3.1). The
difference in UV absorbance between phosphate buffer and water was less
than 0.15%. Dilutions were carried out so that the UV readings were all in the
range of 0.1 to 1 OD to ensure high sensitivity of the UV spectrometer (183).
The dilution factors varied from 100 to 250. All of the samples were prepared
freshly. Beer's Law (Equation 3.1) was used to determine the stock
concentration.
C=
A×D
ε •l
Equation 3.1
in which A is UV absorbance at λmax, D is the dilution factor, l is pathlength (1
cm), and ε is the extinction coefficient at λmax.
3.3.4 Monoaquation and bisaquation of Pt complex
The Pt complex was converted to the monohydrated or bisaquated forms
before reaction with the RNA nucleoside. The aquated forms were prepared by
addition of one or two equivalents of AgNO3 to a DMF solution of
amino-acid-linked Pt complex, respectively (the hydrolysis reaction equations
are shown in Schemes 3.1 and 3.2). The reaction mixture was incubated at
room temperature in the dark for 12 h with constant shaking, and then the
white precipitate, AgCl, was removed by centrifugation at 12,000 RPM for 15
min. The supernatant of the platinum-DMF mixture was transferred into a new
microcentrifuge tube and stored at -20 °C up to one week. The concentration
61
of each stock solution was calculated based on the mass of Pt complex used.
The approximate concentration of stock solution was 5-6 mM.
Scheme 3.1: Preparation of monoaquated amino-acid-linked platinum analogues
Scheme 3.2: Preparation of bisaquated amino-acid-linked platinum analogues
3.3.5 Platination of RNA nucleosides
All platination reactions were carried out in a 25 mM sodium phosphate
buffer, pH 7.0, 5 mM sodium chloride at 37 °C in the dark. The nucleoside
stock solutions were freshly prepared every time as described in Section 3.3.3.
The monoaquated or bisaquated Pt complex stock (1a, 1b, 2a and 2b) was
prepared as described in Section 3.3.4 and diluted as required just prior to use.
The RNA nucleosides were incubated at 37 °C with aquated complex in 1:1
(with 1 mM nucleoside concentration), 1:10 (with 0.1 mM nucleoside
62
concentration), and 1:50 (with 0.1 mM nucleoside concentration) ratios of Pt to
nucleosides. For reactions with 1:10 and 1:50 nucleoside: Pt molar ratios, two
repetitions were done (error was calculated and indicated in the graphs). All
reaction mixtures were incubated at 37 °C in the dark up to 24 h. At each time
point, 25 μL of sample was taken from the reaction mixture and 3 μL of 2 M
NaCl was added into the sample to quench the activity of the Pt complex.
Following each incubation, the samples were immediately frozen on dry ice
and stored in -80 °C until further analysis. All samples were centrifuged at
12,000 RPM for 10 minutes to remove any precipitates before injection onto
the HPLC.
3.3.6 High performance liquid chromatography (HPLC)
Samples were analyzed with a Supelco Discovery C18 column (5 μm
particle diameter; 4.6 mm i.d. × 250 mm, Sigma-Aldrich, St. Louis, MO, USA)
on a Waters 600 LC with a 717 autosampler (Waters, Milford, MA, USA) and
UV-detector set at 254 nm. Forty mM NH4OAc, pH 6.5, was used as the
running buffer, and acetonitrile was used as the elution buffer. A 95% NH4OAc
buffer in H2O, with 2% acetonitrile and 3% H2O, was used for column
equilibration at the beginning of the applied program, and then an acetonitrile
gradient was applied (2 to 40% over 40 min). The average column pressure
within this program was around 2200 psi. The retention times of the
nucleosides were confirmed by injection of known standards. The peaks were
63
quantified using Empower software (Empower Software Solutions, Inc.) with
the baseline set to the baseline integration method (the integration begins
when the UV reading first begins to rise above the baseline and ends when it
returns to the baseline.).
3.3.7 HPLC standard calibration
Stock solutions of single nucleosides were prepared as described in
Section 3.3.3. For each nucleoside, at least three parallel sets of injections
were performed to obtain the standard curves. In each set, 11 samples with
different concentrations (from 25 pmole at 1 μM to 5 nmole at 200 μM) were
prepared, and 25 uL of the sample was injected onto the HPLC and analyzed
as described in Section 3.3.6. Between every two injections, one blank
injection (water only) was made. The mean peak area was used as parameter
Y versus the actual injected amount X (in qmoles). The standard error was
also calculated for each point. The calibration curves of adenosine (A) and
guanosine (G) were repeated two times over a six-month time period (each
time with three parrallel sample sets, and the repeats were made at least one
month apart).
3.3.8 HPLC analysis of digested products of 23S rRNA platination
HPLC analysis of digested products of 23S rRNA platination was done by
Dr. Hanna Hedman and Dr. Keshab Rijal. The experimental procedure was
64
summarized in detail in the thesis work of Dr. Keshab Rijal (181). In short, to
evaluate
the
binding
preferences
of
the
two
platinum
complexes
(monoaquated cisplatin and monoaquated Oplatin (2a)), 23S rRNA was
incubated with approximately equimolar concentrations of metal complex and
total guanosine content. Unplatinated and platinated RNAs were each
digested with P1 nuclease for 14 h at 37 °C in 30 mM NaOAc, pH 5.2, and 250
µM ZnCl2. The digested nucleotides were dephosphorylated with calf intestine
alkaline phosphatase (CIAP) after addition of 10× CIAP reaction buffer
(Promega) to a final concentration of 50 mM Tris-HCl (pH 9.3 at 25 °C), 1 mM
MgCl2, 0.1 mM ZnCl2 and 1 mM spermidine. The resulting nucleosides were
separated and analyzed by the HPLC program described in Section 3.3.6
(above experiments were done by Dr. Hanna Hedman and Dr. Keshab Rijal).
3.3.9 Liquid chromatography-mass spectrometry (LC-MS)
LC-MS was carried out on an ACQUITY Ultra Performance Liquid
Chromatography (UPLC) system (Waters Corporation, MA USA) with an
HSST3 C18 column (1.8 μm particle diameter, 2.1 mm i.d. x 100 mm) coupled
to a LCT Premier XE mass spectrometer(Waters, Milford, MA, USA).
Acetonitrile and 100 mM NH4OAc, pH 6.0, were used as LC buffer. The column
was maintained at 50 °C and elution was carried out with a linear gradient of
acetonitrile and ammonium acetate buffer. The gradient was run for 3 min with
a flow rate 700 µL/min, starting at 100% ammonium acetate and decreasing to
65
60% over the course of the gradient. The coupled mass spectrometer was first
calibrated with the internal standard, dioctyl phthalate (DOP), before use. All
LC-MS data were collected in the positive-ion mode.
3.4 Results and discussion
3.4.1 HPLC standard calibration of single nucleosides
Figure 3.3. HPLC standard curves of U, C, A and G are shown. The detected peak area
(mAU*sec) was plotted versus injected amount (pmole) and fitted to a linear equation.
Each point is an average of three repeat with error bar indicated.
In order to perform quantitative analysis from HPLC-UV, a series of
standard calibration curves were obtained for the nucleosides. From the data
shown in Figure 3.3, clear linear relationships were obtained for all four
nucleosides with R factors greater than 0.99. In Figure 3.3, each data point
66
was an average of three injections, and the error bar is also indicated (in all
cases, the error bars are smaller than the symbol representing each point). As
shown in Figure 3.3, high reproducibility was observed with this LC method,
even when the injections were made months apart. The four linear equations
can therefore be used in future Pt reactivity studies for quantification of product
yields.
3.4.2 Quantitative LC analysis of digested products of 23S rRNA
platination
In order to evaluate the binding preferences of monoaquated cisplatin and
Oplatin, the LC data of digested products of 23S rRNA were further analyzed
using the standard curves (Figure 3.3), and the amount of each nucleoside in
the sample was determined. In addition to the four standard nucleoside peaks,
five product peaks in the chromatogram of nucleosides derived from
cisplatin-modified rRNA were detected and compared to that of the unmodified
RNA (Figure 3.4). Similarly, at least five product peaks were identified in the
Oplatin sample, whose pattern differred from that of the cisplatin sample. The
appearance of these peaks corresponded to a decrease in intensity of the
standard nucleoside peaks.
Uridine was used as an internal standard since it is the least reactive
substrate among nucleosides for Pt(II)-based complexes (74). The relative
nucleoside/uridine value was calculated by dividing the amount (pmole) of
each nucleoside (A, U, G, and C) with the calculated pmole amount of uridine
67
Figure 3.4. HPLC analysis of rRNA nucleosides are shown. A) Overlaid HPLC
chromatogram
of
rRNA
nucleosides
from
digestion
of
unmodified
(upper),
cisplatin-modified (middle) and Oplatin-modified (lower) samples (181). The product
peaks are indicated with arrows. B) The detected peak areas of unreacted rRNA
nucleosides are summarized. C) By using the standard curves of RNA nucleosides, for
each sample, the pmole amount of each nucleoside was calculated and normalized to
the uridine amount.
in the same sample. The HPLC results of the nuclease-digested unmodified
and platinum-modified 23S rRNA reveal different binding preferences for the
two platinum complexes (Figure 3.4). After modification of 23S rRNA with
68
cisplatin, the amount of guanosine was reduced by approximately 50% and the
adenosine amount is reduced by 25% (Figure 3.4). These results suggest that
cisplatin reacted primarily with purine residues, and G residues are the major
target sites. After normalization, the cytidine amount was higher than the
control. This result could occur if uridine is able to react with cisplatin, even
through it has been reported as the least reactive substrate among
nucleosides for Pt(II)-based complexes (74). In contrast to the result with
cisplatin, the detected adenosine amount in 2a-modified 23S rRNA sample
was reduced by more than 50% while the guanosine amount in this case was
only reduced by 25% compared to the unmodified rRNA.
Based on LC-MS analysis, the product peaks in cisplatin-modified RNA
sample
were
assigned
as
cisplatin-RNA
cis-[Pt{1,2-(GpG)}(NH3)2]+,
cis-[Pt{1,2-(GpG)}(NH3)2]+,
intra-cross-linked
adducts
cis-[Pt{1,2-(ApG)}(NH3)2]+,
and
other
bisfunctional
adducts
such
as
cis-[Pt(Guo)2(NH3)2]2+ and cis-[Pt(Ado)(Guo)(NH3)2]2+ (Guo = guanosine, Ado
= adenosine) (181). Five potential product peaks in Oplatin-modified 23S rRNA
sample were assigned as 2a-RNA monofunctional adducts such as
[PtCl(Ado)(Orn)]+,
adduct
[PtCl(Guo)(Orn)]+,
[Pt(Ado)(Guo)(Orn)]+
[Pt(Ado)(H2O)(Orn)]+,
and
bisfunctional
intra-cross-linked
adduct
[Pt{1,2-(ApG)(Orn)}]+ (181).
Overall, monoaquated cisplatin and Oplatin (2a) exhibit different sequence
preferences for platination of large, structured RNA. More specifically, similar
69
to DNA, cisplatin exhibits a preference for guanosine over adenosine in RNA,
whereas the opposite preference was found for Oplatin (2a). This result is
consistent with the primer extension analysis on 23S rRNA (122) and further
confirms Oplatin's altered sequence preference for A.
3.4.3 Binding studies of Oplatin with single nucleosides
Reactions of Oplatin (2a, 2b) with adenosine or guanosine in 1:1 molar
ratios
With the establishment of both qualitative and quantitative HPLC methods
for the evaluation of amino-acid-linked cisplatin analogue reactivity towards
RNA, reactions were then carried out between chosen cisplatin analogues and
RNA
purine
nucleosides.
Ratios
of
1:1
mono-
or
bisaquated
Pt
complex:nucleoside were employed. The results with 2a and adenosine (Ado,
A) are shown in Figure 3.5. Product peaks (19 and 21 min retention times)
were observed within the first 10 min of incubation. After approximately 7 h
incubation, two more product peaks were detected at retention times of 5 and
6 min. Another small peak was observed at approximately 17 min. The results
with bisaquated Oplatin, 2b, and adenosine (Ado, A) (Figure 3.6) were slightly
different from those with monohydrated Oplatin (2a). Similar to the results
shown in Figure 3.5, two major product peaks at 19 and 20 min retention times
could be detected right after mixing; however, no product peaks at retention
times of 5 and 6 min were detected. The same reactivity test was also carried
70
out with 2a and guanosine (Guo, G) (Figures 3.7 and 3.8). As shown in Figure
3.7, a single dominant product peak (9 min retention time) was detected after 1
h. As with the adenosine-2a reaction, additional product peaks (5 min retention
time) were observed at longer reaction times (> 3 h). Compared with 2a, 2b
showed similar reactivity towards guanosine (Figure 3.8). A single product
peak (9 min retention time) was detected after the 1 h reaction time.
The Oplatin reactivity profiles with RNA purine nucleosides as determined
by HPLC analysis are summarized in Table 3.2. The major product peaks
consistently eluted 1-2 min earlier than the unreacted nucleosides in the
spectra due to the increase in polarity after platination. Based on the overall
results, both mono- and bisaquated Oplatin (2a and 2b) demonstrate better
reactivity and form multiple adducts with A compared to G. Competition
reactions were carried out next to confirm this reactivity difference.
Figure 3.5. HPLC results of time-dependent monoaquated Oplatin (2a) reaction with
adenosine in a 1:1 molar ratio are shown. The left figure is the control without 2a.
71
Figure 3.6. HPLC results of time-dependent bisaquated Oplatin (2b) reaction with
adenosine in a 1:1 molar ratio are shown. The left figure is the control without 2b.
Figure 3.7 HPLC results of time-dependent monoaquated Oplatin (2a) reaction with
guanosine in a 1:1 molar ratio are shown. The left figure is the control without 2a.
72
Figure 3.8. HPLC results of time-dependent bisaquated Oplatin (2b) reaction with
guanosine in a 1:1 molar ratio are shown. The left figure is the control without 2b.
Table 3.2: Retention times of guanosine, adenosine, and major platination products with
2a and 2b are sumarized.
Retention Time
1
2
G
G-Pt
A
A-Pt
A-Pt
12-13 min
9-10 min
22 min
18-19 min
20-21 min
Competition reactions of Oplatin (2a, 2b) with adenosine and guanosine
in a 1:1:1 molar ratio
In order to confirm Oplatin's reactivity and preference for adenosine over
guanosine in RNA studies, a competition reaction of 2a with both adenosine
and guanosine (1:1:1 molar ratio) was performed and analyzed by HPLC. As
shown in Figure 3.9, the major products observed are derived from adenosine
platination (retention times of 19 and 21 min). Adenosine product peaks were
73
detected immediately after mixing of the reaction components; whereas, the
guanosine platination product were only be seen after 2-3 h of incubation.
Figure 3.9. HPLC results of a competition reaction of monoaquated Oplatin (2a) reaction
with adenosine and guanosine in a 1:1:1 molar ratio are shown. The left figure is the
control without 2a.
To summarize the results of the binding studies with HPLC analysis, monoand bisaquated Oplatin (2a, 2b) can form multiple stable products with
adenosine within a few minutes reaction time. The preference for A over G was
also observed in the single nucleoside reactions, indicating that Oplatin (2)
reactivity at A residues in h24 within the full-length rRNA may result from a
nucleotide preference in addition to structural recognition of the RNA helix
(122).
74
Figure 3.10 HPLC results of a competition reaction of bisaquated Oplatin (2b) reaction
with adenosine and guanosine in a 1:1:1 molar ratio are shown. The left figure is the
control without 2b.
Reactions of excess Oplatin (2a) with adenosine or/and guanosine
In order to have a more quantitative view with respect to reaction kinetics
of 2a with individual nucleosides, the peak areas of adenosine or guanosine
were analyzed following reactions with the platinum complex. As with reaction
profiles shown in Figures 3.9 and 3.10, excess 2a was reacted with
adenosine or guanosine separately in different molar ratios. A control sample
without platinum complex was prepared for each nucleoside reaction under the
same reaction conditions. At each 2 h time point of the reaction, both control
and platinum-containing reactions were analyzed by HPLC to obtain the peak
areas of unreacted nucleoside. As indicated in Equation 3.2, the peak area of
75
unreacted nucleoside in each platinum sample was divided by the nucleoside
peak area in the control as the fraction of unreacted nucleoside remaining in
the reaction mixture (Figure 3.11).
Equation 3.2
in which Y is unreacted nucleoside fraction.
Figure 3.11. Excess 2a reactions with adenosine (left) or guanosine (right) were
quantified and plotted as a function of reaction time.
When incubated with at least 10-fold excess 2a, the decrease of
unreacted adenosine in the reaction mixture was significantly faster than for
guanosine. After 6 h of incubation, less than 50% of unreacted adenosine
remained. For guanosine, more than 75% of the unreacted nucleoside was left
at the same time point and ratio. After 10 h, the reactions carried out with a
1:50 ratio of nucleoside:Pt contained less than 10% unreacted A, whereas, 40%
unreacted G was detected. Competition reactions with excess Pt and both G
and A present were also carried out, and the data are shown in Figure 3.12
76
with direct comparison of G and A reactivity.
Figure 3.12. Competition reactions of excess monoaquated Oplatin (2a) with adenosine
and guanosine were quantified and plotted as a function of reaction time (unreacted
adenosine (■); unreacted guanosine (□)).
The competition reactions with both purine nucleosides present and
10-fold or 50-fold excess monoaquated Oplatin (2a) showed a very clear
preference for A (Figure 3.12). From the 0 to 10 h time points, the decrease of
adenosine was always faster than G in both reactions. This result further
supports the preference of 2a for A over G, which is opposite of the reactivity
preference of cisplatin.
3.4.4 Preliminary product analysis of Oplatin with adenosine
As mentioned previously, Oplatin (2) can form multiple adducts with
adenosine under short incubation times, and has a clear preference for A over
G. In order to further analyze the structures of platinated products, LC-MS
analysis was carried out on an overnight reaction between monoaquated
Oplatin (2a) and adenosine.
77
Figure 3.13. LC-MS analysis of monoaquated Oplatin (2a) with adenosine is shown. The
HPLC trace is shown in panel A. Mass spectra were collected in the positive ion mode
(panels B-G) for each peak at the various retention times indicated. The isotope
distribution is shown in the expanded peak insets.
78
As shown in Figure 3.13A, six peaks were detected by LC and the
corresponding mass data are given. The last peak in the LC spectrum at 2.24
min is assigned as unreacted adenosine based on the mass data and
comparison to control adenosine (Figure 3.13B). The two product peaks at
1.89 and 2.03 min retention times have the same mass of 629.1 m/z, both with
typical platinum isotope distributions (Figure 3.13C and D). The three product
peaks at 0.61, 0.80, and 0.89 min (Figure 3.13E, F and G) all have a mass of
691.1 m/z.
Based on the mass data of adenosine and Oplatin (2), and consistent with
previous literature (122,181), the structures shown in Figure 3.14 are
proposed. There are three nitrogen atoms in adenine available for Pt
coordination, which explains the multiple product peaks in the HPLC trace from
the reaction between 2a and adenosine. All platination reactions were carried
out in 25 mM sodium phosphate buffer and 5 mM NaCl. The phosphate
concentration is four-fold higher than Cl-. As a result, phosphate coordination
to Oplatin is also possible.
A potential hydrogen bond exists between the carboxylate oxygen of
Oplatin and N6-H of adenosine, which may explain why only two major
products are observed at the shorter reaction times. This hydrogen bond would
stabilize coordination at the N1 or N7 position of adenine. We hypothesize that
steric factors and the absence of stabilizing hydrogen bonding would disfavor
A-N3 coordination to 2a. Previous research to characterize cisplatin-DNA
79
Figure 3.14. Structures and calculated masses of 2a, adenosine, and proposed products
for species with 0.61, 0.80, 0.89, 1.89 and 2.03 min HPLC retention times are shown.
80
products showed hydrogen-bonding interactions between the amine group on
cisplatin and O6 of guanine, leading to cisplatin's preference for guanosine
over adenosine (74, 88). Based on this information, we hypothesize that the
Oplatin preference for A over G is due to a similar mechanism, but with
differing hydrogen-bonding interactions. Therefore, we propose that the two
major products observed when Oplatin reacts with adenosine correspond to
structures 5a and 5b in Figure 3.14. Further studies to confirm this hypothesis
are discussed later in this thesis.
3.4.5 Binding studies of Gplatin (1) with single nucleoside
In addition to Oplatin, Gplatin (1) reactivity with RNA nucleosides was also
characterized. The binding studies were performed as described in Section
3.3.3. Compared to Oplatin, Gplatin has no side chain and also has one less
positive charge. In the reaction of monoaquated Gplatin (1a) with adenosine,
two product peaks were observed with similar retention times as those of
Oplatin (Figure 3.15). The rates of increase of product peak area with
corresponding decrease of unreacted adenosine over time were slower
compared to 2a, implying a weaker reactivity of 1a towards RNA nucleosides.
In the reactions of monoaquated Gplatin (1a) with guanosine (Figure
3.16), two major product peaks were observed on reverse-phase HPLC, with
shorter and longer retention times than guanosine. In general, Gplatin (1)
demonstrated similar reactivity towards RNA purine nucleosides as Oplatin (2)
81
in terms of the number of products, but some differences were observed with
respect to reaction rates and product structures or behaviors on the HPLC
column.
Figure 3.15. HPLC result of time-dependent monoaquated Gplatin (1a) reaction with
adenosine in a 1:1 molar ratio is shown. The left figure is the control without 1a.
Figure 3.16. HPLC result of time-dependent monoaquated Gplatin (1a) reaction with
guanosine in a 1:1 molar ratio is shown. The left figure is the control without 1a.
82
Figure 3.17. Results from a competition reaction of monoaquated Gplatin (1a) with
adenosine and guanosine in a 1:1:1 molar ratio were shown. The left figure (A) is the
control without 1a.
Since the Oplatin (2) preference for adenosine was hypothesized to occur
because of a potential hydrogen bond between its carboxylate group and
adenosine N6-H, Gplatin was expected to have a similar reaction preference
towards A with the same five-membered ring coordination pattern. In a
competition reaction of monoaquated Gplatin (1a) with both purine nucleosides
83
present (Figure 3.17A), five product peaks were detected after 30 min, then
two more product peaks appeared after extended reaction times (9 h). The
fractions of nucleoside remaining in the reaction mixture were also determined
by comparing the peak area of the nucleoside in the sample with the peak area
of nucleoside in the control (no Gplatin) (Equation 3.2). As seen in Figure
3.17B, Gplatin also demonstrates an A preference. This competition reaction
was only carried out once. The reaction appeared to reach its first equilibrium
(for monofunctional adduct formation) after 1 h (Figure 3.17B); no decrease in
unreacted nucleoside amount was observed during the 1 to 5 h time points.
Although this result needs to be repeated, we believe that it supports our
hypothesis that the preference for A is due to a hydrogen-bonding interaction
between the amino acid carboxylate and A N6-H.
3.5 Future directions
Coordination of aquated forms of Oplatin (2) and Gplatin (1) with RNA
nucleosides reveals unexpected reactivity and binding preferences for
adenosine in both full-length rRNA and in the context of free nucleosides. This
preference differs from that of cisplatin, which coordinates mainly to guanosine
residues. These results indicate that the RNA reactivity of Pt(II)-based
complexes can be tuned by altering the carrier ligands. By characterizing the
Oplatin-RNA products at the nucleoside level, we can understand the role of
the ligands in platinum reactivity, and separate this activity from the influence
84
of RNA secondary or tertiary structure. The information gained from this study
will contribute to the current profile of chemical modifications of cisplatin
analogues to improve their selectivity or to develop more effective metal-based
chemical probes or drugs.
85
CHAPTER 4
Product characterization and kinetic studies of Oplatin with
RNA nucleosides
4.1 Abstract
The role of RNA in the life cycle of the cell is highlighted by protein
synthesis, in which ribosomal RNA is essential. Therefore, the ribosome is
considered to be an ideal drug target. Platinum-based complexes have been
reported as chemical probes to determine solvent accessibility of rRNAs in
vivo and in vitro, and the reactivity can be tuned by altering the carrier ligand(s)
to amino acids, such as ornithine (Oplatin, 2). In order to further understand
the mechanism behind Oplatin's altered reactivity, the work in this chapter
focused on the reactivity and product profile of 2 with nucleosides by using
high performance liquid chromatography (HPLC), nuclear magnetic resonance
(NMR) spectroscopy, and mass spectrometry (MS). Based on these results
and previous literature, we have gained a better understanding on how the
ligands alter the reactivity of Pt(II) compounds.
4.2 Introduction
Since the discovery of cisplatin's antitumor property, its reactivity to
nucleosides and nucleic acids has been studied intensively (88, 139, 184). The
N7 position of a deoxyguanosine residue was shown to be the primary target
site of cisplatin, with deoxyadenosine as a minor coordination site (74, 88).
86
Based on a theoretical study, the preference of cisplatin for G over A was
attributed predominantly to a strong hydrogen bond between the ammine
ligand of the metal complex and the 6-oxo group of guanine (185). The
Pt-adenine adduct, in which either the Pt-ammine or chlorido ligand serves as
the hydrogen-bond donor, is energetically less favorable compared to the
Pt-guanine adduct (185) (Figure 4.1).
Figure 4.1. Optimized structures of cisplatin platination products with guanine (top) and
adenine (bottom) are shown (185). Geometries were optimized using Jaguar and the
B3LYP functional (186, 187), with the 6-31G** basis set. Platinum was represented by
the Los Alamos LACVP** basis (188-190), which includes relativistic effective core
potentials. The lowest energy structure was determined by additional single point
calculations using Dunning’s correlation consistent triple-ζ basis set cc-pVTZ(-f) (191)
with the standard double set of polarization functions; the potential H bonds (dashed
lines) and bond lengths indicated.
87
Figure 4.2. An example of radioactive labeling and primer extension for platination
residue identification in RNA is given. (A) A schematic illustration of ddNTP sequencing
of RNA is shown. (B) A schematic illustration of reverse transcription (RT)-based primer
extension of Pt(II)-modified RNA is given. (C) Gel electrophoresis analysis of the
sequencing and probing results is illustrated. The observed modified site (stop) on the
gel is one nucleotide prior to the actual modified site.
88
In the previous chapter, Oplatin (2) demonstrated a clear preference for
adenosine over guanosine, within the loop or bulge regions of RNA (122). This
result suggests that Oplatin (2) is a promising candidate not just as a chemical
probe that is complementary to cisplatin, but also as a potential RNA-targeting
anticancer drug. Product characterization of the Oplatin-nucleoside adducts
will offer insight into compound selectivity, and will provide an improved point
of view regarding the ligand effects on Pt(II) reactivity without interference from
the RNA higher-order structure (185). Previous studies on RNA-Pt(II) adducts
focused on identification of platination sites using methods such as radioactive
labeling, primer extension, and enzymatic footprinting (summarized in Figure
4.2); however, there is little information about the chemical structures of the
RNA-Pt(II) adducts from these studies (117, 118, 192-195).
In the work reported in this chapter, we further evaluated Oplatin reactivity
at the nucleoside level using HPLC, NMR spectroscopy, and MS. These
Oplatin-nucleoside
product
profiles
are
expected
to
give
a
better
understanding of how the ligands alter Pt(II) reactivity, and will assist in the
design of RNA-targeting Pt(II) complex design.
4.3 Materials and methods
4.3.1 General
All chemicals and reagents were purchased from Sigma-Aldrich or Fisher,
unless otherwise stated (Section 3.3.1). Deuterium oxide (99.9 atom% D) was
purchased from Cambridge Isotope Laboratories, Inc. The C8-13C labeled
89
adenosine was a generous gift from Dr. Jun Jiang and Dr. John SantaLucia.
RNase-free, distilled, deionized ddH2O (Millipore water) was used throughout
the experiments.
4.3.2 Product isolation
The stock solution of adenosine or guanosine was prepared fresh for each
platination reaction, and the concentration was determined using UV-visible
spectroscopy as described in Section 3.3.3. Oplatin (2) was monoaquated to
2a right before each reaction. The platination reaction was prepared with 1:1
Pt:nucleoside ratio and the reaction conditions are summarized in Section
3.3.5. For product isolation, each platination reaction was carried out for 12 h
before quenching (Section 3.3.5) and the sample was stored at -80 °C until
further analysis. The product fractions were isolated using the HPLC program
described in Section 3.3.6. The isolated fractions were dried and stored at
-20 °C.
4.3.3 Mass spectrometry
MS analysis was performed on a 7T Bruker SolariX Fourier Transform Ion
Cyclotron Resonance Mass Spectrometer (FT-ICR) in the Rodgers' laboratory.
Data collection was done with the help of Dr. Zhihua Yang. All mass spectra
were
obtained
in
the
positive-ionization
mode.
The
protonated
Oplatin-nucleoside adducts were generated by an electrospray ionization (ESI)
90
source from the solutions of 0.01 mM sample dissolved in 50%:50%
methanol:water. Direct infusion ESI experiments were performed at an infusion
rate of 2 µL/min. The nebulizing (NZ) drying gas temperature was set at 180 °C
with a flow rate of 4 L/min, and held at 1.2 bar. The ESI capillary voltage was
set between −3.5 and −4.5 kV. Tandem mass spectra (MS/MS) of the product
ions of interest were generated by collision-induced dissociation (CID) of the
trapped precursor ions (argon collision gas). The fragmentation amplitudes
varied from 0.4 to 1.0 V, and the isolation width was 4.0 m/z. All data and
simulated spectra were analyzed using the DataAnalysis 4.1 software (Bruker
Daltonics).
4.3.4 NMR spectroscopy
The 1H and
13
C NMR (700 MHz) spectra of Pt(II)-nucleoside adducts and
reaction kinetics were recorded on a Bruker Avance 700 MHz spectrometer
equipped with a TXI cryoprobe (Lumigen Instrument Center). Data collection
was done in collaboration with Dr. Jun Jiang. In order to differentiate the H8
and H2 signals in adenosine, C8-13C labeled adenosine was used (Section
4.3.2). The C2 and C8 signals of adenosine were identified by their chemical
shifts. Ten μM trimethylsilyl propionate (TSP) was added to each sample as
the internal standard. The 1H chemical shifts were referenced to the residual
HDO (semi-heavy water) signal, which was calibrated against TSP. For the 1H
and 13C NMR (700 MHz) spectra of Pt(II)-nucleoside adducts, D2O was chosen
91
as the solvent without buffer.
Solutions from the kinetic studies were initially prepared in H2O and then
transferred into D2O by solvent replacement (i.e., dry the sample under
vacuum and then re-dissolve in D2O). Oplatin (2) was converted to the
monohydrated form (2a) by reacting with AgNO3 in a 1:0.7 molar ratio in D2O
for 5 h in the dark. Then, the white precipitate, AgCl, was removed by
centrifugation twice at 12,000 RPM for 10 min. Monoaquated Oplatin (2a) was
freshly prepared before each experiment. The platination reaction was carried
out in 25 mM sodium phosphate buffer, pH 7.0, at 37 °C. The concentration of
nucleoside was 100 μM, and that of 2a was 5 mM. The TSP standard was
added to each sample to a final concentration of 10 μM for calibration of proton
chemical shifts and peak integrations. A 1D version of nuclear Overhauser
effect spectroscopy (NOESY) with presaturation during relaxation delay and
mixing time (noesypr1d) was used to acquire the 1D 1H spectra at 37 °C. Data
collection was run for 6 min and 50 sec (128 scans). Each reaction was carried
out in duplicate. Data were processed with TOPSPIN 2.1 (Bruker Biospin). The
percentage of the unreacted nucleobase proton signal was calculated as a
function of time by using Equation 4.1.
Equation 4.1
92
As shown in Scheme 4.1, the data were fitted to a pseudo-first-order
exponential decay: Y = A exp(−kobs X), in which, Y is the unreacted fraction,
and X is the reaction time.
Scheme 4.1. Reaction scheme for Pt(II)-nucleoside kinetic studies.
4.3.5 Ultraviolet (UV) spectroscopy
An Aviv 14DS UV-visible spectrometer in Dr. David Rueda's laboratory was
used for all UV measurements. The isolated platinated nucleoside was
dissolved in 25 mM NaH2PO4/Na2HPO4, pH 6.0. The sample temperature was
kept at 37 °C for all measurements. Extinction coefficients were given in
Chapter 3, Section 3.3.3.
4.4 Results
4.4.1 Product characterization of Oplatin (2) reactions at the
nucleoside level
4.4.1.1 Mass spectrometry analysis
93
Figure 4.3. HPLC (C18) analysis of reactions of monoaquated Oplatin (2a) with (A)
adenosine, or (B) guanosine, in a 1:1 molar ratio is given.
The
2a-nucleoside
reaction
products
were
separated
by
using
reverse-phase (C18) HPLC-UV as discussed in Section 3.3.6. The individual
fractions were collected and subjected to mass spectrometry (in the
positive-ion mode) for characterization. Two major product peaks (Figure 4.3,
referred to as A1 and A2), with retention times of 19.8 and 21.6 min,
respectively, were identified in the chromatograph of the adenosine-2a
reaction. The guanosine-2a mixture gave only one major product peak (Figure
4.3, referred to as G1) with a retention time of 10.8 min. The same product
profile was revealed in 2a reactions with a mixture of adenosine and
guanosine (1:1:1 molar ratio, Figure 3.9). Relative abundances of the A
adducts were also deduced from the reverse-phase HPLC profile (Figure 4.3).
94
The A product peaks have similar abundances (A1, 23%; A2, 22%) and
appeared simultaneously in the reaction time profile. With higher hydrophilicity,
all of the platination products eluted before the unreacted nucleosides. The LC
results indicate that 2a forms multiple stable adducts with adenosine over a
short reaction time. As described in Chapter 3, Oplatin has a strong
preference for reactions with adenosine over guanosine (122).
Figure 4.4. The 2a-nucleoside reaction products were analyzed by MS in the positive-ion
mode.
Mass spectrometry analysis (Figure 4.4) confirmed the formation of
2a-nucleoside adducts, based on the m/Z values and signature Pt isotope
distribution (details of peak assignments are given in Table 4.1 and Figure
4.5). The data indicate that monofunctional platination adducts are the major
95
Table 4.1. The detailed assignments of the mass spectra of 2a-nucleosides adducts (the
chemical formulas are given for the major product peaks; others are indicated as
fragments of the major products.)
(Orn = ornithine; Ado = adenosine; Guo = guanosine)
species in each of the isolated fractions. A pronounced peak with m/Z 629.1
was observed in both mass spectra of the adenosine reaction fractions A1 and
A2 (Figure 4.4). These products were assigned as monofunctional adducts of
2a with A (m/z: calcd. 629.1; obsd. 629.1), likely resulting from coordination at
different nucleobase nitrogens, N1 and N7. Similarly, a dominant peak was
detected in the G1 fraction and assigned as a monofunctional adduct of 2a
with G(m/Z calcd. 645.1; obsd. 645.1, Figure 4.4).
Combining mass and HPLC data for reactions of 2a with A, some
preliminary assignments could be made. At the nucleoside level, Pt(II)
complexes normally show high coordination ability with both the N1 and N7
positions of adenosine (195-197). In double-stranded DNA and RNA, however,
96
NH3
H3N
O
O
H2N
NH2
N
N
O
Pt
Cl
NH2
Pt O
NH2
N
N
Cl
N
HO
N
HO
N
O
OH
N
O
OH
OH
5a
[PtCl(N1-Ado)(Orn)]+
calculated mass: 629.12
OH
5b
[PtCl(N7-Ado)(Orn)]+
calculated mass: 629.12
NH3
O
H2N
Cl
Pt O
O
N
N
HO
NH
N
NH2
O
OH
OH
7
[PtCl(N7-Guo)(Orn)]+
calculated mass: 645.11
Figure 4.5. The structures of hypothesized monofunctional 2a adducts with guanosine
or adenosine are shown (Orn = ornithine; Ado = adenosine; Guo = guanosine, μ:
bridging ligand).
the A-N1 position is involved in Watson-Crick base pairing, therefore less
accessible than A-N7. In previous probing studies, 2a was highly reactive
towards adenosine residues in the single-stranded regions of RNA, which
implied that 2a could be more reactive towards the adenosine N1 in RNA (122).
Since the A-N7 is not involved with Watson-Crick base pairing, its accessibility
would be expected to be similar in single-stranded and double-stranded
97
regions of the RNA, although the possibility of its involvement in RNA tertiary
structure can not be excluded. Furthermore, the N7 of purines in RNA is
generally less accessible compared to DNA due to the deeper and more
narrow "major" groove of RNA.
Based on the mass data (Figures 4.4, 4.5, and Table 4.1), as well as the
elution time and fraction abundance (Figure 4.3), the major species in the
A1and A2 fractions were assigned as A-N1 or A-N7 Pt-coordinated
monofunctional adducts, [PtCl(N1-Ado)(Orn)]+ or [PtCl(N7-Ado)(Orn)]+. It
should be noted that electrospray MS (Figure 4.4 and Table 4.1) also
identified minor products within the A1 and A2 fractions, including a
bifunctional adduct, [Pt(Ado)2(Orn)]+ (m/z: calcd. 860.2; obsd. 860.2; Ado =
adenosine) and a doubly platinated adenosine adduct, [(PtCl(Orn))2(N1,
N7-Ado)]+, (m/z: calcd. 990.1; obsd. 990.1) (Table 4.1 and Figure 4.6 A).
Tandem mass spectrometry and simulated spectra were used to assign
the composition of the minor products of the 2a reaction with adenosine
(Figure 4.6). In each of the 2a-adenosine product fractions, more than one
product type was identified with a monofunctional adduct as the major species.
Further analysis, in which isolated fractions A1 and A2 were reinjected onto
reverse phase HPLC, suggested that the minor products in each of the HPLC
fractions were produced during the reaction and co-eluted
with the major
adducts (an asymmetric peak shape was observed with both the A1 and A2
HPLC fractions, Figure 4.7).
98
Figure 4.6. Based on the proposed chemical formula of 2a-adenosine minor adducts
from fractions A1 and A2, the experimental and simulated spectra of m/Z 990.1 and
860.2 are compared (A). (B) CID analysis of m/Z 860.2 in the positive-ion mode. The
fragmentation patterns match the proposed chemical composition of [Pt(Ado)2(Orn)]
(Orn = ornithine; Ado = adenosine, μ: bridging ligand).
+
99
Figure 4.7. HPLC chromatographs show isolated A1 and A2 fractions reinjected on a
reverse phase C18 column after NMR and MS analysis.
4.4.1.2 NMR analysis
Proton and
13
C NMR spectroscopy was used to further identify the 2a
coordination site(s). Previous analysis of monofunctional deoxyadenosine
adducts of cisplatin by Eastman showed that the more hydrophilic product
eluted earlier in reverse-phase HPLC, which was assigned as the N1 adduct
due to a major chemical shift change (~ 0.5 ppm) of the H2 proton upon
platination (195). The other product was assigned as the N7 adduct due to the
observed 0.4 ppm downfield shift of the H8 peak (195). C8-13C-labeled
adenosine was used in our initial NMR studies in order to confirm the
assignment of the H8 signal (Figure 4.8A). The peaks for the H8 and H2
protons of fractions A1 and A2 were assigned by employing 2D heteronuclear
multiple quantum correlation (HMQC) spectroscopy in collaboration with Dr.
Jun Jiang (Figure 4.8A). Even though the chemical shifts of the adenosine H8
and H2 protons are located in the same region, they can
100
Figure 4.8. The 2D heteronuclear multiple quantum correlation (HMQC) experiments
were run on the adenosine control (green), A1 (blue), and A2 fraction (red), at 37 °C. The
C8-H8 and C2-H2 regions of the overlaid spectra (A) were shown in upper and lower
panels, respectively. The chemical shifts of C8 and C2 were used to assign the chemical
shifts of H8 and H2. The aromatic region of 1D proton NMR spectra (B and C) of the
isolated 2a-nucleoside products is shown for: (B) adenosine (upper), A1 (middle), and
A2 (lower), and (C) guanosine (upper) and G1 (lower). Proton signals for A-H8, A-H2 and
G-H8 are noted with triangles, circles, and stars, respectively.
101
be differentiated by coupling to the C8 and C2 carbon atoms, respectively. The
peaks corresponding to C8 are between 140 and 145 ppm for adenosine and
product A1 and A2, and those of C2 are downfield shifted to 152-157 ppm. In
the HMQC spectra, crosspeaks between C8 and H2 or C2 and H2 are
observed. From the chemical shift of the carbon atom to which the proton is
attached, an H8 can be differentiated from an H2 proton.
In the 1D 1H spectrum for minor products of fraction A2 (Figure 4.8B,
lower), both A-H8 (triangles, open and closed for major products) peaks were
downfield shifted by approximately 0.5 ppm, while the corresponding changes
of A-H2 chemical shifts were only 0.1 ppm. The stronger deshielding effect
observed for the two A-H8 signals suggests that platination occurs on the
neighboring A-N7 for both products. The dominant signal set (closed symbols)
was assigned as the monofunctional A-N7 adduct, and the other set (open
symbols) was assigned to the bifunctional adduct [Pt(Ado)2(Orn)]+, with
coordination to both N7 positions (Figure 4.6). In the 1D 1H spectrum of
fraction A1 (Figure 4.8B, middle), two sets of A-H8 and A-H2 peaks were
clearly shown in a pattern that differs from that of the A2 fraction. The major
peak corresponding to H8 only showed a small chemical shift change (< 0.1
ppm, A-H8, marked with a closed triangle), whereas the other showed a 0.5
ppm downfield shift (A-H2, marked with a closed circle), which suggested that
the metal center is most likely coordinated to A-N1 (185, 195). For the set of
minor peaks those corresponding to A-H2 (open circle) and A-H8 (open
102
triangle) signals were shifted downfield by about 0.5 ppm (Figure 4.8B,
middle). These products were assigned to the A-N1 monofunctional adduct
and doubly platinated adenosine with coordination of both the N1 and N7
positions (Figures 4.5 and 4.6A, [(PtCl(Orn))2(N1, N7-Ado)]+, m/Z: calcd.
990.1; obsd. 990.1, μ: bridging ligand), respectively.
In contrast to the results with adenosine, the G1 fraction showed only one
peak corresponding to the guanosine H8 (asterisk, Figure 4.8C, lower). The
0.5 ppm downfield shift of this resonance was suggestive of G-N7 platination.
Overall, the NMR results combined with MS data allowed for further
identification of the 2a-nucleoside products. The HPLC peaks referred to A1
and A2 are predominantly A-N1 or A-N7 platinum adducts, respectively; G1 is
a G-N7 monofunctional adduct. The NMR spectrum of the G1 fraction (Figure
4.8C, lower) does not show any minor products. This result is also consistent
with the mass data.
4.4.1.3 UV spectroscopy analysis
The coordination sites of 2a-nucleoside products were also corroborated
by ultraviolet (UV) spectroscopy. The λmax of each fraction, A1 and A2, was
compared with the corresponding methylated nucleoside (Figure 4.9). As
previously reported, cisplatin coordination to the nucleoside can cause a
similar shift in the λmax as methylation (195), which would also allow for UV
spectroscopy as a detection method for monitoring the platination reaction
103
Figure 4.9. Ultraviolet spectral patterns of 2a-nucleoside product fractions are shown
(A), and (B) their λmax are compared to the λmax values of corresponding methylated
nucleosides from the literature (16).
104
(192, 194). Comparing to data published on cisplatin effects on DNA
nucleosides (195), a similar correlation was observed with the Oplatin reaction
with RNA nucleosides. Compared to the λmax (260 nm) of adenosine, the
A-N7-platinum adduct (A2) exhibited red-shifted absorption maximum at 263
nm (Figure 4.9A, upper), similar to N7-methyladenosine (Figure 4.9B, a red
shift to λmax=267 nm) (16). The A1 fraction (A-N1 adducts) revealed an
absorption maximum at 258 nm, which is consistent with methylation on the
same site (a characteristic λmax at 258 nm) (16). The absorption maximum for
G-Pt (fraction G1) occurred at 257 nm, similar to N7-methylguanosine (16).
The UV absorption spectra showed that 2a and methylation have similar
effects on the UV spectroscopic characteristics of ribonucleosides, and also
demonstrate that UV spectroscopy could be used as a characterization
method for future nucleoside-level platination studies. Alkylation sites in RNA
have been shown to cause cytotoxicity and contribute to the anticancer activity
of alkylating drugs (198). Dimethylsulfate (DMS) is a known methylation
reagent that is commonly used as a chemical probe for RNA structures and
dynamics. Similar effects of Oplatin and methylation on the UV spectroscopic
characteristics of nucleoside show the potential of 2a as a chemical probe and
drug candidate in future RNA-related studies.
4.4.2 Kinetic characterization of Oplatin-nucleoside reactions
The binding of platinum compounds to nucleic acids has been shown to
105
be kinetically controlled (140, 184). The Bierbach group previously reported an
A-coordinating Pt(II) compound with a template-dependent preference to A
(199-201). To establish the kinetic preference of monoaquated Oplatin (2a) for
various nucleosides, the rates of 2a and monoaquated cisplatin reacting with
adenosine and guanosine were determined using 1D 1H NMR spectroscopy in
parallel (in collaboration with Dr. Jun Jiang). A 50-fold excess of the Pt
compounds were used to ensure pseudo-first-order reaction conditions. The
percentage of the unreacted nucleobase proton signal was calculated as a
function of time (Section 4.3.4).
The results (Figure 4.10A) showed that 2a reacted with adenosine 12
times faster (k=(69 ± 8) x 10-6 S-1 vs. (5.6 ± 2.1) x 10-6 S-1) than monoaquated
cisplatin under the given conditions. In contrast, the reaction of 2a with
guanosine was 1.5-fold slower (k=(17 ± 2) x 10-6 S-1 vs. (25 ± 2) x 10-6 S-1) than
cisplatin. In the NMR spectra of adenosine reacting with 2a at different time
points (Figure 4.10B), the proton signals corresponding to the assigned
monofunctional adducts were observed first, and then the doubly platinated
and bifunctional products appeared after 2 h. This further supports the peak
assignments in Figures 4.8 and 4.9, since the doubly platinated or bifunctional
products result from a second reaction with the substitution on monofunctional
adducts. Since both of the above-mentioned experiments were carried out at
the nucleoside level without any influence from the structure of RNA, Oplatin (2)
is the first reported Pt(II) compound with a template-independent A preference.
106
Figure 4.10. (A) Reactions of single nucleosides (▲: adenosine, ◆ : guanosine) with a
50-fold excess (pseudo-first-order) of monoaquated cisplatin or 2a were monitored by
1
H NMR spectroscopy. Data were fitted to a first-order exponential decay and k values
1
were determined. (B) The 1D H NMR spectra overlay of the aromatic proton regions of
adenosine with 50-fold excess 2a reaction are shown. All peaks were normalized with
the proton peak of TSP.
107
4.5 Discussion
In this chapter, we first presented the adduct characterization of Oplatin (2)
with adenosine and guanosine. The assigned product structures are
summarized in Figure 4.11. A kinetic study was also carried out to characterize
the A selectivity of 2. Monoaquated Oplatin (2a) was shown to form two major
products with adenosine (the A1 fraction is assigned to [PtCl(N1-Ado)(Orn)]+
(5a) and the A2 fraction is assigned to [PtCl(N7-Ado)(Orn)]+ (5b)) and one
major adduct with G (the G1 fraction is assigned to [PtCl(N7-Guo)(Orn)]+ (7)).
Doubly
platinated
([(PtCl(Orn))2(N1,N7-Ado)]+)
and
bifunctional
([Pt(N7-Ado)2(Orn)]+) adducts were also observed as minor products in the
2a-adenosine reaction. Oplatin is reported here to be the first Pt(II) compound
with template-independent kinetic preference for adenosine when reacting with
RNA. Cisplatin's selectivity for G in DNA has been attributed predominantly to
a strong hydrogen bond between the ammine ligand and 6-oxo group of
guanine (185); Oplatin's preference can be explained by an alternative
hydrogen bond between the carboxylate group of 2 and 6-amine group of
adenine.
Based on the kinetic trans effect, we propose that Oplatin coordinates to
the nucleoside in a trans fashion with the α-amine group and nucleoside N1 or
N7 (202-204). When binding at the A-N1/N7 position, the 6-amino group in
adenosine could form hydrogen-bonding interactions with the carboxylate
ligand on Pt, whereas this structure would not be available in the guanosine
108
Figure 4.11. Assigned 2a product structures with nucleosides are shown (Orn =
ornithine; Ado = adenosine; Guo = guanosine, μ: bridging ligand).
109
adduct (Figure 4.11). Steric factors and the absence of a stabilizing hydrogen
bond could also explain the lack of A-N3 coordination for both Oplatin and
cisplatin. The differences between Oplatin and cisplatin in both their chemical
structures and sequence selectivity support the Lippard hypothesis and
preference of cisplatin to G over A (185). This theory is valuable for future
Pt-based chemical probing work and drug design, while other nucleoside
selectivity (pyrimidines) needs to be achieved.
For analogue coordination at N1 or N7 of adenosine, identifying new
compounds with preferred reactivity at the N1 position could be a promising
direction for selective RNA targeting by Pt compounds, since DNA is generally
lacking single-stranded regions, and its A-N1 would therefore be expected to
be unreactive. Combining the previous probing studies (122) and work in this
chapter, Oplatin has been proven to be a useful RNA structure probe with
sequence complementarity to cisplatin. We have also demonstrated that this
compound has the potential to be a selective RNA-targeting drug if the
preference for N1 over N7 of A can be achieved through ligand modifications.
110
CHAPTER 5
Tandem mass spectrometry analysis of Oplatin and
Oplatin-nucleoside adducts*
* This work was carried out in collaboration with Chenchen He, Dr. Zhihua
Yang, Dr. M.T. Rodgers, and the FELIX facility at the University of Nimengen,
the Netherlands.
5.1 Abstract
Tandem mass spectrometry (MS/MS) is an effective and powerful
technique that is widely used in current proteomic methodologies; however,
MS/MS techniques in RNA-related research are not as well established. In the
work presented in this chapter, different MS/MS techniques (collision induced
dissociation, survival yield analysis, and infrared multiphoton dissociation
spectroscopy) were applied to the study of amino-acid-linked cisplatin
analogues (Gplatin, 1 and Oplatin, 2) to obtain further structural information.
They were also utilized in the product characterization of monoaquated Oplatin
(2a) reaction with adenosine and guanosine. MS/MS analysis has provided
new information that is complementary to the data obtained from nuclear
magnetic resonance (NMR) spectroscopy and high performance liquid
chromatography (HPLC).
111
5.2 Introduction
Mass spectrometry (MS) has played a key role in drug discovery (205).
Tandem mass spectrometry (MS/MS) has also emerged as a powerful tool in
protein structural biology over the last two decades (APPENDIX) (206).
Tandem MS is a sensitive, high-resolution method with wide applications in
proteomics; however, to date it has played only limited roles in RNA-related
research.
In this work, we first applied the collision induced dissociation (CID)
technique to obtain further structural information on Pt(II) complexes, 1 and 2,
and the 2a-nucleoside adducts discussed in previous chapters (Chapter 2, 3
and 4). CID is an effective MS/MS method used to induce direct collisions
between sample molecules and rare gas atoms, and applied frequently in the
analysis of peptides and proteins (CID mechanism, APPENDIX) (207, 208). In
this study, our goal was to use CID to gain information about the glycosidic
bond strengths of nucleosides and various platinum adducts.
Figure 5.1. A schematic diagram of the glycosidic bond and depurination reaction is
given.
112
The effect of platination on stability of the nucleoside glycosidic bond has
been a focus of cisplatin-DNA studies (203). Depurination (Figure 5.1) is a
common damage event in both DNA and RNA, and certain alkylating agents
accelerate the depurination event, thus contributing to their anticancer activity
(209). The ribosome-inactivating protein (RIP) is well known to inactivate
ribosomes by cleaving the glycosidic bond of a single adenosine in the large
subunit rRNA (210, 211). This A residue (A4324 in rat) lies within the α-sarcin
loop, which is conserved in the large subunit rRNAs from bacteria to humans
(211). In a theoretical study published by the Lippard group, the effect of G-N7
platination by cisplatin on the depurination reaction was evaluated, and results
indicated that G-N7 platination has only a minor effect on the stability of
glycosidic bond (212). Due to cisplatin's preferred reactivity for G, the A
products have not evaluated. The effects of different coordination products
have not been examined either. In the work reported in this chapter, we used
glycosidic survival yield analysis to evaluate the platination effect on the
stability of purine nucleosides. Survival yield analysis is a routine method in
mass spectroscopy to assess precursor ion stability (213, 214). Since the
internal energy of the ion is dependent on its chemical composition, the
survival yield is also related to the chemical structure (214).
In addition, aiming at atomic level characterization, infrared multiphoton
dissociation spectroscopy (IRMPD) was applied to each of the monofunctional
2a-nucleoside adducts assigned in Chapters 3 and 4 (5a, 5b and 7). IRMPD is
113
similar to traditional infrared spectroscopy, but with significantly improved
sensitivity (APPENDIX). The obtained IRMPD spectra are compared to results
from density function theory (DFT) calculations in order to make structural
correlations with the IR bands (discussed further in Chapter 6)
5.3 Materials and methods
5.3.1 General
All chemicals and reagents were purchased from Sigma-Aldrich or Fisher,
unless otherwise stated. RNase-free, distilled, deionized ddH2O (Millipore
water) was used for all experiments.
5.3.2 Sample preparation
The Oplatin (2) samples were prepared as described in Section 2.3.2. The
monofunctional 2a-nucleoside adducts (5a, 5b, and 7) were isolated as
described in Section 4.3.2.
5.3.3 FT-ICR mass spectrometry
Data collection was done by Dr. Zhihua Yang in Dr. M. T. Rodgers'
laboratory. All of the samples were dissolved in a 50:50 methanol:water
mixture to a final concentration between 10 and 20 μM. The MS and MS/MS
analyses were performed on a 7T Bruker SolariX Fourier Transform Ion
Cyclotron Resonance Mass Spectrometer (FT-ICR, Bruker Daltonics, Bremen,
114
Germany) (APPENDIX). The mass spectra of compound 2 were obtained in
the negative-ionization mode; the others (5a, 5b and 7) were obtained in the
positive-ionization mode. Direct infusion ESI experiments were performed at
an infusion rate of 2 µL/min. The nebulizing (NZ) drying gas temperature was
set at 180 °C with a flow rate of 4 L/min and held at 1.2 bar. The ESI capillary
voltage was set between 3.5 and 4.5 kV for 2, and between -3.5 and -4.5 kV
for 5a, 5b and 7. The capillary exit was typically held at 120 V with the deflector
plate set at 140 V. Ion funnel 1 was operated at 90 V, with skimmer 1 at 20 V.
The number of ions to fill the instrument for each acquisition was set at 108 to
avoid space charge effects. The trapping potentials of the FT-ICR analyzer
were set at 0.8 V, and the analyzer entrance was maintained at 7 V. The side
kick was set at −4.0 V for 2 with a side kick offset of −4.5 V, or it was set at 4.0
V with a side kick offset of 4.5 V for 5a, 5b and 7 to further optimize peak
shape and signal intensity.
5.3.4 Glycosidic bond survival yield analysis
Data collection was done by Dr. Zhihua Yang and Chenchen He in Dr. M. T.
Rodgers' laboratory. Glycosidic bond survival yield analysis was performed
using a Bruker amaZon ETD quadrupole ion trap mass spectrometer
(APPENDIX)
(Bruker
Daltonics,
Bremen,
Germany).
The
protonated
2a-nucleoside adducts (5a, 5b, and 7) were generated by an Apollo
electrospray ionization (ESI) source from solutions of 0.01 mM sample
115
dissolved in a 50:50 methanol:water mixture. All solutions were infused at a
flow rate of 3 μL/min using a Hamiltonian syringe connected to the Apollo ESI
source. These liquid samples were introduced through the nebulizer assembly
into the spray chamber at a temperature of 200 °C and a pressure of 0.69 bar.
N2 was used as the heated drying gas flowing countercurrent to the stream of
droplets at a flow rate of 3.0 L/min to aid volatilization, ionization, and carry
away any uncharged material. Protonated monofunctional adducts (5a, 5b and
7) were formed and focused directly into the entrance of the glass capillary.
The capillary voltage was held at 4.5 kV to transfer ions into a dual ion funnel.
Helium gas was introduced into the ion trap as the collision gas. The pressure
of the collision gas was about 1 mTorr. The primary rf amplitude was applied to
the ring electrode for trapping ions. An auxiliary dipolar amplitude was applied
to the second end-cap electrode for exciting ions. The qz value was set to 0.25
for ion trapping. The rf excitation amplitude was varied from 0.00 at 0.01 V step
sizes until complete fragmentation of the precursor ion was observed. The
isolation window for the precursor ions was of 0.8 m/z width. Spectra were
recorded in the positive-ion mode for protonated Oplatin-nucleoside adducts.
Each data point was an average of three repeats. Data analysis was
performed using Compass Data Analysis 4.0.
Glycosidic bond survival yield curves were deduced based on the signal
intensity and Equation 5.1, in which P is the precursor ion, and F is the
fragment ion.
116
Equation 5.1
5.3.5 IRMPD analysis
IRMPD action spectra of Oplatin-nucleoside adducts were obtained using
a laboratory-constructed 4.7 T Fourier transform ion cyclotron resonance mass
spectrometer (Cryomagnetics, Inc., Oak Ridge, TN) with a 128 mm i.d.
horizontal bore (215). Photodissociation is induced by the widely tunable free
electron laser for infrared experiments (FELIX) (APPENDIX) (215-217). A final
sample concentration of 0.5 mM in a 50:50 MeOH:H2O solution was used. The
solutions were delivered to a Micromass "Z-spray" electrospray ionization (ESI)
source at a flow rate between 2.5 and 8.5 μL/min. The ions emanating from the
spray were accumulated in an rf hexapole ion trap for several seconds prior to
pulsed extraction through a quadrupole deflector. The precursor ions were
injected into the FT-ICR MS via a 1 meter-long rf octopole ion guide, and
stored in the ICR cell to cool to room temperature by radiative emission. FELIX
typically produces high-energy (50-100 mJ) macropulses of 1-ns-spaced
micropulses for a duration of about 5-10 μs. The monofunctional adducts (5a,
5b, and 7) were isolated using stored waveform inverse Fourier transform
(SWIFT) techniques and irradiated for 3 s corresponding to interaction with 15
macropulses from the free electron laser to induce IR photodissociation over
the wavelength range between ~16.0 μm (~625 cm-1) and ~5.2 μm (~1920
117
cm-1). The IRMPD yield was determined based on Equation 5.2, and then
normalized linearly with laser power to correct for changes in the laser power
as a function of photon energy.
Equation 5.2
5.4 Results and discussion
5.4.1 CID studies of Oplatin (2)
In order to evaluate platination-effects on the glycosidic bond strength of
nucleosides, and further explore the potential of MS/MS in Pt(II)-RNA related
studies, the fragmentation patterns of Oplatin (2) and its reaction products with
adenosine were analyzed. Using the procedure described in Section 5.3.2, 2
was analyzed first using MS in the negative-ion mode. The anticipated 396.0
m/z ion was isolated and the collision voltage was applied further to cause
fragmentation of 2 (Figure 5.2).
Figure 5.2. CID results of Oplatin (396.0 m/z in negative-ion mode) under low collision
voltage are given. The precursor ion is marked with asterisk.
118
Figure 5.3. A comparison of experimental and simulated spectra of Oplatin (2)
fragments is given.
In CID analysis of 2, three major fragment ions were observed. The
proposed chemical compositions of those fragments were compared with the
119
simulated spectra for further confirmation (Figure 5.3). Those product ions
were formed by losing NH3, HCl and (NH3 + HCl) from 2 (Figures 5.2 and 5.3,
obsd. m/Z: 379.0, 360.0 and 343.0, respectively).
Figure 5.4. Possible mechanisms of 2 fragmentation are shown.
Based on the structure of 2 and previous literature (218, 219), we
hypothesized that the increased overall internal energy of 2 due to collisions in
the cell promotes increased flexibility of the ornithine side chain, and the lone
electron pair on the side-chain amine group acts as a nucleophile and attacks
the α-carbon, which further leads to rearrangement and fragmentation as
shown in Figure 5.4. This result was used as a control in the MS/MS analysis
on the fragmentation patterns of 2a-nucleoside adducts
120
5.4.2 CID studies and glycosidic bond stabilities of 2a-nucleoside
adducts (5a, 5b and 7)
As described in Section 5.2, the effect of platination on the glycosidic bond
stability has been one focus of cisplatin-nucleoside research (203). Guanosine
is the primary target site of cisplatin. As a result, the glycosidic bond strength of
A-cisplatin adducts have not been evaluated, and the relative stabilities of
different coordination products have not been compared. To evaluate the
differences in glycosidic bond stabilities between the various monofunctional
2a adducts, 5a, 5b, and 7, CID-MS/MS was employed. Glycosidic bond
survival yields were determined for each adduct (Figures 5.5 and 5.7).
In this part of the project, we compared adducts 5a and 5b, since they
have the same chemical composition. Two groups of fragments were observed
in each CID mass spectrum (Figure 5.5), and all of the positive ions showed a
Pt isotope distribution. The first group contained the fragments formed by
losing neutral molecules of NH3, HCl, and (NH3 + HCl) from 2a (Figure 5.5 and
Table 5.1: for 5a, obsd. m/z: 612.1, 593.1, and 576.1, respectively; for 5b,
obsd. m/z: 612.1, 593.1, and 576.1, respectively); the second group contained
the fragments corresponding to all of the first group of positive ions
sequentially losing a ribose sugar (Figure 5.5 and Table 5.1: for 5a, obsd. m/z:
497.1, 480.0, 461.1 and 444.1; for 5b, obsd. m/z: 497.1, 480.0, 461.1 and
444.1).
121
Figure 5.5. CID results of 5a and 5b (629.1 m/z in positive-ion mode) under low collision
voltage are shown. The precursor ion is marked with an asterisk.
Table 5.1. Detailed assignments of the CID results of 5a and 5b (both are
+
[PtCl(Ado)(Orn)] )
(Orn = ornithine; Ado = adenosine)
As described in Section 5.2, depurination is a common form of damage to
RNA and can contribute to the anticancer activity of certain alkylating drugs
(209). The glycosidic bond survival yield curves were then determined by
fragmentation of 5a and 5b from 0.0 V (rf excitation voltage) until the parent ion
was no longer detectable. The glycosidic bond survival yield curves were then
deduced based on the signal intensity and Equation 5.1. The survival yield
122
curves
were
quite
different
for
5a
and
5b
(Figure
5.6A);
5a
([PtCl(Orn)(N1-Ado)]+) demonstrated a higher glycosidic bond stability
compared to 5b ([PtCl(Orn)(N7-Ado)]+). When higher excitation voltage was
applied, 5a underwent fragmentations that did not involve breaking the
glycosidic bond (Figure 5.6B).
Figure 5.6. The glycosidic bond survival yield (A) and individual fragment abundance
percentage (B) were deduced from the MS/MS data of 5a and 5b by increasing the rf
excitation valtage from 0.0 V until the parent ion was no longer detectable. (A) The
glycosidic bond survival yield was determined by using Equation 5.1, and plotted
versus the excitation voltage applied. (B) The signal intensities of individual fragment
ions were plotted versus the excitation voltage.
In order to further compare N7-platination effects on the glycosidic bond
stability of A and G, a CID mass spectrum of 7 (G-N7 platination) was obtained.
123
Two groups of fragments were observed (Figure 5.7). Similar to the results of
5a and 5b, the first group contained the fragments formed by losing neutral
molecules of NH3, HCl, and (NH3 + HCl) from 7 (Figure 5.7 and Table 5.2: m/z:
628.1, 609.1, and 592.1, respectively); the second group contained the
fragments corresponding to each of the first group of positive ions sequentially
losing a ribose sugar (Figure 5.9 and Table 5.2: obsd. m/z: 513.0, 496.0,
477.1 and 460.0, respectively).
Figure 5.7. CID result of 7 (645.1 m/z in positive-ion mode) under low collision voltage is
shown. The precursor ion is marked with an asterisk.
Table 5.2. Detailed assignments of the CID results of 7
(Orn = ornithine; Guo = guanosine)
124
Figure 5.8. The glycosidic bond survival yield (A) and individual fragment abundance
percentage (B) are deduced from the MS/MS data of 7 by increasing the rf excitation
valtage from 0.0 V until the parent ion was no longer detectable. (A) The glycosidic
bond survival yield was determined by using Equation 5.1, and plotted versus the
excitation voltage applied. (B) The signal intensities of individual fragment ions were
plotted versus the excitation voltage.
The glycosidic bond survival yield of 7 ([PtCl(N7-Guo)(Orn)]+) was also
deduced as mentioned above, and the obtained curve (Figure 5.8) was similar
to that of 5b (Figure 5.6A). Product 7 underwent fragmentation that did involve
breaking the glycosidic bond when higher excitation voltage was applied
(Figure 5.8B)
5.4.3 IRMPD analysis of 5a, 5b and 7
Aiming at atomic-level characterization of the 2a-nucleoside products and
a further understanding for the previously mentioned glycosidic bond stability
differences, infrared multiphoton dissociation (IRMPD) spectroscopy was
applied to each monofunctional product, 5a, 5b, and 7. Similar to the traditional
infrared spectroscopy (IR), precursor ions in the ICR cell can absorb photons
of a certain wavelength (220). This absorption results in a build-up of internal
energy and cause precursor ion dissociation (220). By plotting the extent of ion
125
dissociation
versus
wavelength,
the
IRMPD
spectra
of
the
three
monofunctional adducts, 5a, 5b, and 7, were obtained and are shown in
Figure 5.9 (220). When the IRMPD data was collected for 5a, the
[Pt(N1-Ado)(NH3)(Orn)]+ signal was isolated as a precursor ion instead of 5a
([PtCl(N1-Ado)(Orn)]+) due to its dominant intensity. Since the products are
isolated by using HPLC with 40 mM ammonium acetate buffer, 5b could further
react with ammoniun buffer during product isolation, leading to replacement of
the chlorido ligand by an ammonia. The IRMPD data were later used as an
experimental standard and directly compared to the results from density
function theory (DFT) calculations, which will be discussed in Chapter 6.
Figure 5.9. IRMPD spectra of [Pt(N1-Ado)(NH3)(Orn)]+, 5b, and 7, are given.
126
5.5 Conclusions
In the work discribed in this chapter, we applied tandem mass
spectrometry and survival yield analysis to obtain stability information on
monofunctional 2a-nucleoside products. By direct comparison of 5a and 5b,
we discovered that 5a ([PtCl(N1-Ado)(Orn)]+) has higher glycosidic bond
strength than 5b ([PtCl(N7-Ado)(Orn)]+). These results reveal for the first time
that different platinum coordination sites on the nucleoside have different
effects on glycosidic bond stability. In the future, this information can be utilized
as a reference for identification of platinum coordination sites by tandem mass
spectrometry. This work illustrates the potential of MS/MS in RNA-Pt or other
modification studies to provide stability information on RNA adducts. Other
tandem mass spectrometry techniques such as infrared multiphoton
dissociation (IRMPD) were also applied, and the IRMPD spectra of
2a-nucleoside adducts were obtained. The combination of these experiments
with computational studies (Chapter 6) will provide a more precise view of
2a-nucleoside adduct structures, as well as an explanation for glycosidic bond
stability differences.
127
CHAPTER 6
Gaussian calculations of Oplatin-nucleoside products*
* This work was carried out in collaboration with Chenchen He and Dr. M.T.
Rodgers.
6.1 Abstract
Tandem mass spectrometry was used to obtain an atomic-level
understanding of the product profiles of monoaquated Oplatin (2a) with
adenosine (Ado) and guanosine (Guo). For further structural characterization
of these products, infrared multiple photon dissociation (IRMPD) spectra of the
electrospray ionization (ESI)-generated monofunctional 2a-nucleoside adducts
([Pt(N1-Ado)(NH3)(Orn)]+, 5b and 7) were obtained. These results were
compared to density functional theory (DFT) calculations in order to make
peak assignments. From the calculation results, we observed that
2a-nucleoside adducts had the potential to adopt various orientations with
unique hydrogen bonds between the amino acid ligand and either the
nucleobase functional groups or the sugar moiety.
6.2 Introduction
During the last few decades, extensive research has been carried out to
establish the structure-activity relationships of cisplatin. Computational
methods have made valuable contributions to our understanding of how
128
cisplatin reacts with DNA (185, 212, 221-224). With atomic-level theoretical
studies, Lippard's group rationalized cisplatin's G-N7 preference and the
impact on glycosidic bond strength (185, 212). These findings have
fundamental importance for the future design of new drugs, and also play an
important role in this Ph.D. project. With a deeper understanding of the
cisplatin mechanism of action, issues related to drug resistance and
dose-related side effects can be addressed. Thousands of cisplatin analogues
have been synthesized and screened for improved biological activity, but only
a few have been successful (184, 225-228). Rational drug design is difficult if
structure-activity relationships are not clear. Therefore, geometries, electronic
structures, and energy profiles of the Pt(II) complexes and their corresponding
platination products are important molecular features that need to be
evaluated theoretically in order to establish such relationships.
As a summary of the previous chapters, several amino-acid-linked
cisplatin derivatives were synthesized. Among them, Oplatin (2) demonstrated
an unambiguous coordination preference for adenosine and single-stranded
regions of RNA, which makes it the first reported Pt(II) complex with a
template-independent adenosine preference when reacting with RNA. Further
product characterization shed light on the mechanisms behind the different
selectivity of 2, compared to the parental compound cisplatin, such as its ability
to form a hydrogen bond with the 6-amine group of adenosine. Tandem mass
spectrometry analysis revealed that coordination of Oplatin (2) to different sites
129
on adenosine (e.g., N1 and N7) affects glycosidic bond stability differently. An
explanation for this difference is still needed. Based on previous reports on
cisplatin (176, 203, 214-217), computational methods are considered a reliable
approach to understanding the mechanism behind the glycosidic bond
cleavage of products 5a and 5b.
The core of cisplatin and its derivatives, the Pt(II) metal center, creates a
challenge
in
theoretical
analyses.
This
problem
arises
from
the
incompleteness of the basis set for the third-row transition metals (222-225).
To overcome this limitation, previous published calculations on cisplatin were
done with a suitable valence basis set in combination with an effective core
potential (ECP) for the Pt(II) center (222-225). There are reports on basis set
evaluation of the prediction accuracy on Pt(II)-containing complex structures,
electronic properties, and IR spectra (229-232). In 1998, Pavankumar and
Hausheer et al. compared two different ECPs, referred to as SBK and
LANL2DZ, and over 30 different basis sets, including 3-21G(d), 6-311*G(3d,
3pd), and cc-pVDZ, for predicting the chemical structure and IR spectrum of
cisplatin (229). Based on their performance, two hybrid basis sets (HF
(Hartree-Fock)/6-311G* and MP2 (Moller-Presset 2nd order perturbation
theory)/6-311G*) were chosen to describe the molecular properties of cisplatin
(229). Later in 2007, similar work was reported by Amado, Carvalho and
coworkers (230). Several methods (e.g., HF, MP and DFT) were evaluated and
reviewed for their roles in prediction of conformational and vibrational modes of
130
cisplatin (230). Based on the overall results of both prediction accuracy and
computational demand, and by comparing the results with existing
experimental data such as X-ray crystal structures, the best result was
achieved by using the DFT mPW1PW approach coupled to the 6-31G(d) basis
set for the ligands and the relativistic ECP LANL2DZ for the Pt center (230).
This basis set was shown to provide the most accurate predicted structure and
IR spectrum of cisplatin regardless of the theoretical methods (230). More
recently, Jorge, Berrêdo, and coworkers constructed three all-electron basis
sets (ABS) on Pt(II)-containing compounds: double zeta valence plus
polarization function (DZP), augmented double zeta (ADZP, augmented
nonrelativistic), and DZP-DKH (Douglas-Kroll-Hess) (231, 232). After testing
the three ABS with the one-parameter hybrid functional mPW1PW on cisplatin
and carboplatin, they concluded that the ABS are only slightly superior to the
standard ECP in geometry calculation (231, 232).
In addition to the lack of conclusive results mentioned above, another
challenge in our project is that the amino-acid-linked cisplatin analogues have
an asymmetric configuration that is different from cisplatin and most of its
derivatives. The summarized publications discussed above are based on the
structural properties of cisplatin (or carboplatin), which has a symmetric and
rigid square-planar configuration, and therefore it is possible that the above Pt
complex calculations are not sensitive to the basis set changes (222-225).
Furthermore, the evaluated basis sets are not yet validated with other
131
derivatives, especially ones with distinct reactivity from cisplatin. As a result, in
order to obtain reliable predictions on the structure and energy profiles of
2a-nucleoside adducts (5a, 5b, and 7), evaluations of the different basis sets,
as well as stepwise structure optimizations will be important.
6.3 Methods
All calculations presented here were carried out using the density
functional theory, as implemented in the Gaussian-09 set of programs (233,
234). Geometries were optimized by using the B3LYP functional with the
def2-TZVPPD basis set, without any symmetry constraints (186). Platinum
was represented by the defz-TZVPPD basis set, which includes relativistic
effective core potentials (186). The energies were re-evaluated by additional
single point calculations at each optimized geometry using the B3LYP
functional with the def2-TZVPPD basis set (186). In these single-point
calculations, Pt was also described by the defz-TZVPPD basis set (186).
6.4 Results and discussion
6.4.1 Basis set evaluation
Since this is a collaborative project, two directions were carried out at the
same time, one focused on choosing the best basis set in describing the
asymmetric Pt(II) complex system (done by Chenchen He), and the other
involving geometry optimization of the Pt(II) complexes (1 and 2) and its
132
corresponding nucleoside adducts (5a, 5b, and 7). Chenchen He in Dr.
Rodger's group focused on basis set evaluation by using Gplatin (1) as a
model study. For comparison, the IRMPD action spectrum of 1 was obtained
and used to verify the reliability of the calculations. First, all of the possible
structures of Gplatin were projected into the calculations. Then, by comparing
the calculated spectra with the IRMPD spectra, and their energy profiles, some
structures were ruled out. Lastly, three different levels of theory and basis sets
were tested, including different ECPs and four ABS (this part of the work was
done by Chenchen He). Gplatin (1) has an asymmetric structure (Chapter 2),
and has preference for A over G nucleosides (Chapter 3). Basis set evaluation
by using 1 as a standard helped us choose the most suitable basis set for
describing amino-acid-linked Pt(II) complexes such as Oplatin (2)
6.4.2 Further examination of the coordination mode of glycine to Pt(II)
The second part of this work was done together with Chenchen He
(Rodgers' Laboratory), and started with structure optimization of 1. The hybrid
basis set mentioned in Section 6.3 was used in the calculations. Glycine has
three atoms available for metal coordination, two oxygen atoms from the
carboxylate group and one nitrogen atom in the α-amine group (Figure 6.1).
Based on the reported X-ray crystal structures of Oplatin (2) and Kplatin,
the Pt(II) center is coordinated to one of the carboxylate oxygen atoms and the
α-amine nitrogen atom (158-160). The structure centered around the Pt(II)
133
metal center was square planar (158-160). Since the IRMPD data were
obtained in the gas phase, we took extra precautions to confirm the structure
that was based on X-ray data. We used the published structure of cisplatin
coordinated to G-N7 as a starting point, and the coordination modes and sites
were varied while maintaining a square-planar configuration around the metal
center. Different initial structures were proposed and calculated at the
def2-TZVPPD level, and the results are summarized in Figure 6.2. The
relative energy profiles are provided as a reference and expressed in kJ/mol.
Figure 6.1. The chemical structure of glycine is shown. The oxygen and nitrogen
atoms that are available for metal coordination are highlighted in red.
NH2-Pt-G_N7_Gplatin_1 was identified as the optimized structure with the
lowest energy profile. In this structure, Gplatin (1) still maintained a
square-planar configuration and G-N7 was trans with respect to the α-amine
group. Glycine coordinates to Pt(II) through the α-amine nitrogen and one of
the carboxylate oxygen atoms. This arrangement is consistent with the
literature (158-160) and the proposed structure (Chapter 2). Due to the
asymmetry of Gplatin (1), another structure, OCO-Pt-G_N7_Gplatin_1 (G-N7
is trans with respect to the carboxylate group) was also obtained with a slightly
higher energy value (8.3 kJ/mol) than NH 2 -Pt-G_N7_Gplatin_1. In
OCO-Pt-G_N7_Gplatin_1, the structure of 1 is the same as that in
134
Figure 6.2. Geometries and relative free energies (kJ/mol, in parenthesis, in italics)
of [PtCl(Gly)(N7-Gua)] (Gua = guanine) were calculated at the def2-TZVPPD level
with various initial Gplatin (1) structures. All of the unlabeled atoms are hydrogen.
135
NH2-Pt-G_N7_Gplatin_1, and a potential hydrogen bond was observed
between the α-amine and G-O6. Calculations resulted in different Gplatin
structures
(e.g.,
in
both
OCO-Pt-G_N7_Gplatin_3)
structures
glycine
OCO-Pt-G_N7_Gplatin_2
coordinates
to
Pt(II)
through
and
two
carboxylate oxygen atoms) were also obtained; however, the energy values of
those structures are significantly higher than that of NH2-Pt-G_N7_Gplatin_1
(68.3 kJ/mol and 67.5 kJ/mol, respectively). Based on the calculation results
discussed above, the structure of 1 in [PtCl(Gly)(N7-Gua)] is optimal, which is
consistent with the data presented in Section 2.3. The two structures,
NH2-Pt-G_N7_Gplatin_1 and OCO-Pt-G_N7_Gplatin_1, were used later as
the starting point in the next step of structural optimization.
6.4.3 Comparison of monohydrated Oplatin (cis- and trans-2a)
structures
Due to the asymmetry of 2, monoaquation of Oplatin can theoretically yield
two activated forms, cis- and trans-2a (Figure 6.3). Based on the kinetic trans
effect (193-195) discussed in Section 4.3, we concluded that the
monoaquation of Oplatin (2) would yield one activated form, cis-2a. Here, in
this section, further calculations were done on both cis- and trans-2a, and their
energy profiles were compared.
The results of these computations (Figure 6.4) further confirmed our
conclusion (Section 4.3) that monoaquation of 2 yields the activated form
136
cis-2a. When cis-2a further reacts with RNA or nucleosides, a hydrogen bond
could potentially form with adenosine 6-amine group when binding at the A-N1
or N7 position. This stabilizing factor was proposed predominantly to its
template-independent adenosine selectivity.
Figure 6.3. The reaction scheme of Oplatin monoaquation and the two possible
monoaquated Oplatin structures are shown.
Figure 6.4. Geometries and relative energies (kJ/mol, in parentheses, in italics) of
cis- and trans-2a were calculated at the def2-TZVPPD level. All of the unlabeled
atoms are hydrogen.
6.4.4 Structure optimization on possible 1a-nucleoside adducts
After optimizing the structures of 1 and cis-2a, the structures of possible
1a-nucleobase adducts were also evaluated, and their energy profiles were
compared for further calculation. Besides the reported common platination
sites, A-N1, A-N7 and G-N7, the A-N3, G-N3, and G-O6 positions were also
137
taken into consideration. For each binding site, both cis- and trans-2a (Figure
6.4) were used in the initial structure constructions. The overall optimized
structures and corresponding energy profiles are summarized in Figure 6.5
and 6.6. The optimized structures were further utilized in the next step of
calculation on the Gplatin-nucleoside adducts.
As summarized in Figure 6.5, the two NH2-Pt-A-N3 structures
(NH2-Pt-A-N3_1 and NH2-Pt-A-N3_2) are the most stable adducts with the
lowest energy values (0.0 kJ/mol and 9.1 kJ/mol, respectively); however, this
situation was expected to change in the next step of calculation with the
nucleoside. With the addition of a ribose sugar at the N9 position of adenine,
steric factors will occur and further destabilize N3-platination. In addition,
elimination of the hydrogen bond between the chlorido ligand or carboxylate
group in Gplatin (1) with A-H9 will disfavor N3-platinated adducts. Besides the
N3-platination adducts, it is also worth noting that when adenine is trans with
respect to the α-amine, the overall energy profile is significantly lower than
those adducts in the cis configuration (comparing the energy values of
NH2-Pt-A-N1_1, NH2-Pt-A-N1_2 with OCO-Pt-A-N1, or NH2-Pt-A-N7_1,
NH2-Pt-A-N7_2 with OCO-Pt-A-N7_1 and OCO-Pt-A-N7_2). When comparing
N1 and N7 platination, the energy difference of the optimized structures is
negligible, which is also consistent with the HPLC data in Section 3.4.3, in
which no preference was observed in peak intensities and increasing
formation rates of the two fractions (A1 and A2).
138
139
Figure 6.5. Geometries and relative energies (kJ/mol, in parentheses, in italics) of
[PtCl(Gly)(N1-Ade)]+, [PtCl(Gly)(N3-Ade)]+, and [PtCl(Gly)(N7-Ade)]+ (Ade = adenine)
were calculated at the def2-TZVPPD level. All of the unlabeled atoms are hydrogen.
140
141
Figure 6.6. Geometries and relative energies (kJ/mol, in parentheses, in italics) of
[PtCl(Gly)(N3-Gua)]+, [PtCl(Gly)(O6-Gua)]+ and [PtCl(Gly)(N7-Gua)]+ (Gua = guanine)
were calculated at the def2-TZVPPD level. All of the unlabeled atoms are hydrogen.
142
Similar to the 1-adenine results (Figure 6.5), platination at the G-N3
position also yielded the structures with the lowest energy profiles (Figure 6.6)
among all of the possible monofunctional 1-guanine adducts. This scenario
also changed after addition of the ribose sugar at the G-N9 position due to the
introduction of steric factors and elimination of the hydrogen bond. Besides the
G-N3 adducts, the most stable structure was NH2-Pt-G-N7. In this structure,
platination occurred at the guanine N7 position, which was consistent with the
observed experimental data presented in Section 4.4.1.2. Guanine and the
α-amine group are in a trans configuration. In contrast to the optimized 2-A-N7
structures (Figure 6.5), no hydrogen bond was observed between the Pt(II)
complex and guanine. Instead, a 3.3 Å distance occurred in order to minimize
the possible repulsive interaction between the negatively charged oxygen
atom in 1 and G-O6. This result further supports the proposed product profiles
and mechanism of A (Ado) preference as mentioned in Section 4.5. The
optimized structures (Figure 6.5 and 6.6), with platination occurring at A-N1,
A-N7 or G-N7 were utilized in further calculations on the 2a-nucleoside
adducts.
6.4.5 Structure optimization and IR spectrum computation on
[Pt(N1-Ado)(NH3)(Orn)]+ (Ado = adenosine)
Aiming at an atomic-level understanding of both the structure and energy
aspects of 2a-nucleoside adducts, IRMPD action spectroscopy was applied in
143
Chapter 5, and the "IR" spectra were obtained as guides to finalize the DFT
calculations. To compute the energy profiles of 5a, 5b, and 7, it was first
necessary to assemble the structures. The optimized structures in the last two
sections (Sections 6.4.3 and 6.4.4) were used as a base to further build the
final adducts. As summarized in Section 4.3, the three major monofunctional
adducts between 2a and nucleosides were A-N1, A-N7, and G-N7
monofunctional adducts. In this and the next few sections, the ribose sugar
and ornithine side chain were then added stepwise onto the optimized
Gplatin-AN1, Gplatin-AN7 and Gplatin-GN7 structures. The IR spectra of all
the structures were also computated and compared with the experimental
data.
After adding the ribose sugar, the structures of [PtCl(Gly)(N1-Ado)]+ (Ado =
adenosine) were calculated and the results are summarized in Figure 6.7. The
ribose sugar adopted a C3'-endo conformation with the 5'-hydroxyl group in
relatively close distance to the A-H8 (~3.2 Å). Two different types of hydrogen
bonds were present in structures of 1 coordinated to adenosine. One type was
between the chlorido ligand and the adenosine 6-amine (Gplatin-A-N1_1,
Gplatin-A-N1_2, Gplatin-A-N1_5 and Gplatin-A-N1_6), and the other type was
between the carboxylate group and adenosine 6-amine (Gplatin-A-N1_3 and
Gplatin-A-N1_4). The latter showed a better stabilizing effect with a relatively
shorter hydrogen bond (1.8 and 1.9 Å instead of 2.3 Å). As a result, the
structures Gplatin-A-N1_3 and Gplatin-A-N1_4 were most thermodynamically
144
stable at this step.
Figure 6.7. Geometries and relative energies (kJ/mol, in parentheses) of
[PtCl(Gly)(N1-Ado)]+ (Ado = adenosine) were calculated at the def2-TZVPPD level.
All of the unlabeled atoms are hydrogen.
145
As a final step, the side chain of ornithine was added to the structures.
Another thing worth mentioning is that when the IRMPD data (for the A-N1
monofunctional adduct) were collected, the [Pt(N1-Ado)(NH3)(Orn)]+ signal
was isolated as a precursor ion instead of [PtCl(N1-Ado)(Orn)]+ due to its
dominant intensity. Therefore, the Cl ligand was replaced by NH3 in order to
make the structure composition consistent with the experimental data. The
theoretical IR spectra of the computed structures (Figure 6.8) were compared
with the experimental IRMPD data (Figure 6.9).
After the final addition of the ornithine side chain and adjustment, the
optimized geometries of [Pt(NH3)(N1-Ado)(Orn)]+ were obtained and are
summarized in Figure 6.8. The ribose sugar still adopted a C3'-endo
conformation; however, the side-chain orientation varied. For example, in the
structures, Oplatin-A-N1_1, Oplatin-A-N1_4 and Oplatin-A-N1_5, the side
chain amine was unbound and did not interact with either Pt(II)-five-member
ring or adenosine. On the other hand, the side chain of ornithine folded back
and close to the α-amine in the Oplatin-A-N1_2 and Oplatin-A-N1_6 structures.
Even though the possible hydrogen-bonding interaction between two amine
groups could provide certain stabilizing effects, tension in the folded side chain
also contributed additional destabilization and resulted in a slightly less stable
energy value. Among the finalized structures, the Oplatin-A-N1_5 was the
most stable; however, in this structure, A-N1 was cis with respect to the
α-amine ligand. As suggested by our previous data (Section 6.4.3), the
146
Figure 6.8. Geometries and relative energies (kJ/mol, in parentheses, in italics) of
+
[Pt(N1-Ado)(NH3)(Orn)] (Ado = adenosine) were calculated at the def2-TZVPPD level.
All of the unlabeled atoms are hydrogen.
147
Figure 6.9. A comparison of experimental IRMPD data (black, top) with the predicted
IR spectra (with 0.98 scaling factor) of the optimized [Pt(N1-Ado)(NH3)(Orn)]+ (Ado =
adenosine) structures is given. The experimental data are also overlayed (grey) onto
each of the computed spectra.
148
monoaquation of Oplatin yielded the activated form cis-2a, and this activation
step was complete before reacting with the nucleoside. Therefore, the best
representatives for the adduct [Pt(N1-Ado)(NH3)(Orn)]+ were expected to be
Oplatin-A-N1_1 and Oplatin-A-N1_2, and their computed spectroscopic
characteristics further confirmed that hypothesis (Figure 6.9).
The predicted IR spectra (Figure 6.9) showed some agreement with the
IRMPD data; however, further optimization on the details such as the scaling
factor is still needed in order to obtain a better match. At this point, the parallel
project carried out by Chenchen He in Dr. Rodgers' group on the evaluation of
the basis sets in amino-acid-linked Pt(II) complex calculations provided a
preliminary conclusion. It is currently believed that the def2-TZVPPD basis set
is not a very efficient or accurate method for predicting the energy and
spectroscopic characteristics of these Pt(II) complexes. This deficiency could
also contribute to the lack of overlap between the predicted and experimental
IR spectra (Figure 6.9). More information is still needed to choose the best
basis set for this project. Oplatin-A-N1_1 and Oplatin-A-N1_2 computed IR
spectra were shown most closely match with experimental IR spectra.
6.4.6 Structure optimization and IR spectrum computation on
[PtCl(N7-Ado)(Orn)]+ (5b) (Ado = adenosine)
Similar to Section 6.4.5, the final steps in structure optimization of
[PtCl(N7-Ado)(Orn)]+ (5b) were continued using the structures shown in
149
Figure 6.5. First, the ribose sugar was added; the results of both structures
and energy profiles are summarized in Figure 6.10. Similar to the results
presented in Figure 6.7 and also consistent with our conclusions in Section
4.3, the most thermodynamically stable configuration of [PtCl(Gly)(N7-Ado)]+
was Gplatin-A-N7_1 (Figure 6.10). In this structure, a hydrogen bond formed
between the oxygen in the carboxylate group and the adenosine 6-amine. In
addition, another hydrogen bond was formed between the chlorido ligand and
5'-OH group with the ribose sugar in a C3'-endo conformation. A-N7 is trans
with respect to the α-amine ligand. All of the above factors contributed to
stabilization of the overall structure. Another structure, Gplatin-A-N7_2, also
presented a relatively stable energy profile (1.5 kJ/mol). The only difference
from the Gplatin-A-N7_1 structure was the hydrogen bonds: two hydrogen
bonds were involved in this structure, one between the chlorido ligand and the
6-amine; the other one between the carboxylate group and the 5'-OH.
Lastly, similar to Section 6.4.5, the side chain of ornithine was added to
the above structures and calculated. The theoretical IR spectra (Figure 6.12)
of the final structures (Figure 6.11) were also computed and compared with
experimental IRMPD data. Based on the structures summarized in Figure 6.11,
5b appeared to be able to adopt more various orientations than
[Pt(N1-Ado)(NH3)(Orn)]+. Both monoaquated Oplatin forms (cis- and trans-2a)
were used in this step of the calculations. The types of electrostatic
interactions between the Pt(II) center and nucleobase region were divided
150
Figure 6.10. Geometries and relative energies (kJ/mol, in parentheses, in italics) of
+
[PtCl(Gly)(N7-Ado)] (5b) (Ado = adenosine) were calculated at the def2-TZVPPD
level. All of the unlabeled atoms are hydrogen.
151
152
Figure 6.11. Geometries and relative energies (kJ/mol, in parentheses, in italics) of
+
[PtCl(N7-Ado)(Orn)] (5b) were calculated at the def2-TZVPPD level. All of the
unlabeled atoms are hydrogen.
153
154
Figure 6.12. A comparison of experimental IRMPD data (black, top) with the
computed IR spetra (with 0.98 scaling factor) of [PtCl(N7-Ado)(Orn)]+ (5b) (Ado =
adenosine) structures is given. The experimental data are also overlayed (grey) onto
each of the computed spectra.
155
into three types: the first type forms a hydrogen bond between the carboxylate
group and adenosine 6-amine (Oplatin-A-N7_3, and Oplatin-A-N7_4); the
second type of hydrogen bond was between the chlorido ligand and
adenosine6-amine (Oplatin-A-N7_1, Oplatin-A-N7_2, Oplatin-A-N7_5, and
Oplatin-A-N7_7), and in the last type, the adenosine 6-amine interacted with
the
α-amine
group
in
Oplatin
(Oplatin-A-N7_6,
Oplatin-A-N7_8,
Oplatin-A-N7_9, and Oplatin-A-N7_10). The orientations of the side chain
were categorized into three types as well. In the majority of the above
structures, the side-chain amine participated in a close electrostatic interaction
with the five-membered ring moeity of 2 (Oplatin-A-N7_1, Oplatin-A-N7_3,
Oplatin-A-N7_5, Oplatin-A-N7_7, Oplatin-A-N7_8 and Oplatin-A-N7_10); in
another group, the ornithine side chain was in an unbound position relatively
far
away
from
any
other
functional
groups
(Oplatin-A-N7_6
and
Oplatin-A-N7_9); in the last category, which was the most interesting, the side
chain was folded in a certain conformation which allowed the side chain amine
to form multiple hydrogen bonds with both the Pt(II)-five-membered-ring center
and the 5'-OH group in the ribose (Oplatin-A-N7_2 and Oplatin-A-N7_4).
Further information is still needed to conclude whether this configuration can
affect glycosidic bond stability. In this round of calculations, the lowest energy
structures were Oplatin-A-N7_1 and Oplatin-A-N7_3. In both of these
structures, the side chain amine participated in an electrostatic interaction only
with the Pt-five-member ring.
156
The computed IR spectra of all the [PtCl(N7-Ado)(Orn)]+ structures are
shown in Figure 6.12. Due to the current choice of scale factor and basis set,
consistency between computed IR spectra and experimental IRMPD data is
not great. After finishing the evaluation of the basis sets, a second computation
of IR spectra is needed, which would offer improved consistency and further
information on the structure and energy characteristics of adduct 5b.
6.4.7 Structure optimization and IR spectrum computation on
[PtCl(N7-Guo)(Orn)]+ (7) (Guo = guanosine)
Structure optimization was also continued on adduct 7. As in Sections
6.4.5 and 6.4.6, the ribose sugar was added in the first step. The results on
both structures and energy profiles are summarized in Figure 6.13. The most
thermodynamically stable structure among all of the calculated structures
above (Figure 6.13) was Gplatin-G-N7_1. In this structure, Gplatin
coordinated to G-N7 trans with respect to the α-amine. The distance between
the oxygen in the carboxylate group and G-O6 was over 3 Å. The ribose sugar
was in a C3'-endo conformation, which was the same as in the Gplatin-A-N7
structures. The side chain of ornithine was then added (Figure 6.14) and the
theoretical IR spectra (Figure 6.15) were also computed and compared with
experimental IRMPD data. The summarized [PtCl(N7-Guo)(Orn)]+ (7)
structures are shown in Figure 6.14. Several differences were noticed when
compared to the 5b results. First, only one orientation was observed between
157
Figure 6.13. Geometries and relative energies (kJ/mol, in parentheses, in italics) for
+
[PtCl(Gly)(N7-Guo)] (Guo = guanosine) were calculated at the def2-TZVPPD level.
All of the unlabeled atoms are hydrogen.
158
159
Figure 6.14. Geometries and relative energies (kJ/mol, in parentheses, in italics) for
+
[PtCl(N7-Guo)(Orn)] (7) (Guo = guanosine) were calculated at the def2-TZVPPD level.
All of the unlabeled atoms are hydrogen.
160
161
Figure 6.15. A comparison of experimental IRMPD data (black, top) with the
computed IR spetra (with 0.98 scaling factor) of [PtCl(N7-Guo)(Orn)]+ (7) (Guo =
guanosine) structures is given. The experimental data are also overlayed (grey)
onto each of the computed spectra.
162
the Pt(II)-five-membered ring and nucleobase moiety when cis-2a was used as
the monoaquated Oplatin form. The distance between the negative-charged
carboxylate group and the G-O6 was approximately 3.2 Å. The side chain
amine of Oplatin forms electrostatic interactions with the Pt(II)-five-membered
ring (Oplatin-G-N7_1, Oplatin-G-N7_3, Oplatin-G-N7_6 and Oplatin-G-N7_8)
or the 5'-OH in the ribose sugar (Oplatin-G-N7_4 and Oplatin-G-N7_5). In
addition, one more type of orientation was also observed in the structure
Oplatin-G-N7_2, in which the side chain amine not only formed a hydrogen
bond with the carboxylate group near the Pt(II) center, but also interacted with
G-O6. The most stable structures of 7 were Oplatin-G-N7_1 and
Oplatin-G-N7_2, in which the side-chain amine of Oplatin formed a hydrogen
bond with the carboxylate group.
The computed IR spectra of the above-mentioned structures are shown in
Figure 6.15. Similar to adduct 5b, the consistency between the computed IR
spectra and the experimental IRMPD data is lacking. Further evaluation of the
basis sets and scale factor is definitely important and essential for the
completion of this project.
6.5 Conclusions and future directions
In this chapter, a computational study was applied to further obtain an
atomic-level understanding of the structural, energy, and spectroscopic
characteristics of 2a-nucleoside adducts, especially 5a, 5b, and 7. In order to
163
achieve this goal, two directions were carried out in parallel. First, due to the
incompleteness of the basis set design on the third-row transition metals,
evaluation on the basis set effect on computation accuracy was required to
choose a suitable method. This part of the project is currently being continued
by Chenchen He in Dr. Rodgers' group. In the second direction, as a
collaboration with Chenchen He, we began geometry optimization on the
Gplatin component of [PtCl(Gly)(N7-Gua)]+, and the results were consistent
with our data (Chapter 2). Then, the two monoaquated Oplatin forms (cis- and
trans-2a) were computed and their energy profiles were compared. Product
cis-2a was determined to be the preferred monoaquated form. This result
provides further support of our proposed theory on product structures and the
mechanism for adenosine-preference of 2 (Chapter 4). Next, various possible
platination sites on both nucleosides, adenosine and guanosine, were
compared. The A-N1, A-N7, and G-N7 positions were more favored than other
sites such as G-O6. Lastly, further structure optimization was performed on
[Pt(N1-Ado)(NH3)(Orn)]+, 5b and 7 by adding the ribose sugar and ornithine
side chain in a stepwise fashion. The resulting calculated structures were
summarized, and their IR spectra were computed as well. From the resulting
structures, we observed that different orientations occur on the ornithine side
chain. More specifically, the side chain of ornithine has the possibility to
interact with the nucleoside sugar moeity in adducts 5b and 7; however, due to
the fact that the current basis set is not the best in describing the Pt(II)
164
complexes, the computed IR spectra did not match very well with the
experimental IRMPD data. Further work on evaluation of the basis set is
needed to obtain improved IR computations. In addition, the possibility that an
interaction between the side chain amine and the ribose sugar causes the
difference in glycosidic bond stability is still under investigation.
165
CHAPTER 7
General conclusions and future directions
7.1 General conclusions
With this multidisciplinary research, we have demonstrated that: 1) altered
reactivity of cisplatin can be achieved by replacing the carrier ligands with
different amino acids; 2) Oplatin (2) displays a template-independent kinetic
preference for adenosine (A) over other nucleosides/nucleotides when
reacting with RNA; 3) Oplatin forms monofunctional adducts with adenosine at
the N1 or N7 positions (5a and 5b, respectively), and with guanosine at the N7
position (7). In addition, the use of tandem mass spectrometry reveals for the
first time that the different coordination sites of 2a on adenosine can lead to
different stabilities of the glycosidic bond. The A-N1 adduct (5a) has higher
glycosidic bond stability than the A-N7 adduct (5b). Lastly, through
computational calculations and IRMPD studies, we have obtained atomic-level
details of the adducts, [Pt(N1-Ado)(NH3)(Orn)]+, 5b, and 7. These results not
only enrich our understanding of structure-activity relationships between
amino-acid-linked cisplatin analogues and RNA, but also illustrate the potential
of utilizing various tandem mass spectrometry techniques for future RNA-Pt(II)
studies.
7.2 Future directions
166
The synthesis of amino-acid-linked platinum complexes is relatively
forwardstraight. There are multiple choices of amino acids and related ligands
to use for future studies, including both natural and unnatural amino acids. As
an example, Gplatin was synthesized to understand the side-chain effect on
the RNA reactivity of amino-acid-linked Pt(II) complexes. In order to further
confirm the proposed mechanism of Oplatin (2)'s altered preference, the
detailed adduct profile and kinetic characterization of 2 with nucleosides is
certainly helpful without the possible influence from the basic side chain. The
synthesis of Gplatin is being continued by Bett Kimutai.
From the coordination studies done in this Ph.D. work, it is clear that
altered reactivity is observed with carrier ligand modification, so it is possible
that by expanding the category of synthesized amino-acid-linked cisplatin
analogues, there is still additional potential to obtain analogues with distinct
reactivity and altered specificity.
Ribosomal RNA is a structurally diverse molecule. Its structural dynamics
are known to be important for both biological function and potential drug
interactions (235). Previously, Dr. Keshab Rijal successfully developed a
methodology of using cisplatin to probe rRNA structure in vitro and in vivo;
however, probes with varying specificity are still needed (119, 236). Compared
to cisplatin, Oplatin is relatively larger and has a different charge distribution
(158-160). Its structural characteristics result in an alternative coordination
behavior with a preference for A over G; it also react exclusively to
167
single-stranded RNA regions. This result not only provides additional
information on solvent accessibility of RNA, but also gives us the first piece of
the puzzle regarding how modification affects Pt complex reactivity toward
RNA. With more amino-acid-linked platinum complexes available, further
binding and probing studies can be achieved. RNA structure information and
the structure-activity relationship of Pt(II) complexes with RNA can be further
obtained and applied to the rational design of RNA-targeting Pt(II) complexes
or drugs.
Two recent reports suggest that cisplatin can coordinate quickly and
accumulate more in RNA than DNA (119, 121). This discovery provides a new
view with respect to using RNA as an anticancer drug target. As discussed in
Chapter 1, RNA is located in the cytoplasm, which makes it more accessible
than DNA. As a result, it has also been hypothesized that if the anticancer drug
can be modified to be more reactive towards RNA, the dose-related toxicity of
Pt(II)-based drug could be greatly reduced (119, 121). Also, since RNA repair
mechanisms are not as systematic as those for DNA, resistance caused by the
DNA repair system may be avoided. From the data presented in this thesis,
Oplatin (2) has shown high reactivity in an RNA loop region, which is a unique
secondary structure in RNA (122). This makes Oplatin (2) a potential Pt-based
drug lead for targeting RNA.
Tandem mass spectrometry is a relatively new and powerful technique.
Based on its application in proteomics and metabolomics studies, it also
168
presents great potential in RNA-related studies. The combination of
computational calculations with IRMPD studies can provide detailed
information on the structure, energy, and spectroscopic characteristics of
RNA-Pt adducts, and this will also be helpful in the interpretation of MS/MS
data. However, due the immaturity and incompleteness of methodology in this
area, further work is still needed that focuses on optimizing the experimental
procedures and computation methods.
169
APPENDIX
Information in this appendix is the instrumentation introduction for Chapter 5.
Figure 1. Diagrams of three beam mass analysers are shown. (A) Time-of-flight (TOF)
spectrometry analyzes mass-to-charge ratios based on the principle that different m/z
ions have different velocities and therefore reach the detector at different times. (B) A
double-focusing mass analyzer analyzes ions based on their different kinetic
energy-to-charge ratios. Ions with the same kinetic-energy-to-charge value are
dispersed on the same path. (C) A quadrupole mass analyzer separates ions by
maintaining stable trajectories of the ions with the same m/z, while the others are
ejected (205).
170
Based on the different types of mass analyzer, the types of mass
spectrometers can be divided into two general groups: beam analyzer
instruments (Figure 1) and trapping analyzer instruments (205). In the beam
analyzer, ions travel in a beam to the detector (205). In the trapping analyzer,
ions are trapped in a defined space for separation and detection (205). In this
section, two major types of trapping instruments are introduced and their mass
separation and detection mechanism are given.
Fourier-transform ion cyclotron resonance (FTICR) mass spectrometer
A Fourier-transform ion cyclotron resonance (FT-ICR) mass spectrometer
is a type of instrumentation that is considered one of the most complicated
mass analysis and detection methods. Its m/z determination depends on the
cyclotron frequency of an ion when a fixed magnetic field is applied (Figure 2)
(207).
When an ion is generated (in our experimental setting, the ions are
produced with electrospray ionization), a series of pumping stages and an ion
guide system lead the ions into the ICR cell where low pressure with a range of
10-10 to 10-11 mBar is present (Figure 2A) (207). The ICR cell is surrounded by
a spatial uniform static superconducting high field magnet (7 Tesla in our
experimental setting), which is consistently cooled by liquid helium and liquid
nitrogen in order to keep the temperature in the cell close to absolute zero
(Figure 2A) (207). Due to the charge of the ion and the magnetic field, the ion
moves in a cyclic motion in a plane perpendicular to the magnetic field due to
171
the Lorentz Force, and it is trapped in a Penning trap (Figure 2A) (207). With a
certain magnetic field, the angular frequency of the ion is directly related to its
m/Z value (Equation 1) (207).
Equation 1
in which B is the magnetic field strength (constant), ω is the induced cyclotron
frequency, m is the mass of the ion, and Z is the charge on the ion.
At this step, due to the small radius of the ion cycle motion, no signal can
be detected (Figure 2B, left) (207). As demonstrated in Figure 2A, besides
the trapping plates, two excitation plates are also installed at the sides of the
ICR cell (207). By generating an oscillating electric field orthogonal to the
magnetic field, the ions can be excited at their resonant frequencies to a higher
radius (Figure 2B, middle) without affecting the cyclotron frequency (207).
Excitation of each individual m/z can be achieved by a swept radio frequency
pulse across the excitation plates of the cell (207). When the excited ion
rotates at a higher orbit, an image current is induced and detected between the
detection plates (Figure 2B, right and 2C) (207). The frequency of this current
is the same as the cyclotron of the ion and the intensity is proportional to the
number of ions (207). The resulting signal is called a free induction decay (FID)
(207). The detected signal is then extracted by performing a Fourier
transform to give a mass spectrum (Figure 2A) (207). Fourier transform ion
cyclotron resonance (FTICR) mass spectrometry is a high resolution mass
spectrometry technique with high accuracy. Due to its ion-trap nature, it does
172
not need to destroy the ions in order for the measurement to be performed.
This enables further analysis on the desired ions.
Figure 2. A schematic of the basic set up of an ICR trap configuration (A) and its
detection mechanisms (B and C) are shown. (A) A schematic diagram of a cubic ICR cell
is shown; (B) After the ions enter the ICR cell, they undergo excitation with a sweep RF
and (C) induce a current in the detector plates (237, 238).
173
Quadruple ion trap (QIT) mass spectrometer
Besides the FTICR mass spectrometer, another predominant instrument in
the category of a trapping instrument is the quadrupole ion trap (QIT) mass
spectrometer (205, 239). Compared to FTICR, a QIT mass spectrometer is a
relatively inexpensive and simple bench-top instrument commonly used for
routine applications. Its mass separation and detection mechanism is similar to
FTICR, which results from the dynamic electric field (RF frequency) and is
dependent on the m/Z value (205, 239).
After ions are produced from the source (ESI source is used in our
experimental setting), the ions enter the cell through the inlet (205, 239). The
basic set up of a QIT cell is composed of three electrodes: one donut-shaped
ring electrode is connected to radio frequency (RF) voltages, and two endcap
electrodes are powered with direct current (DC) voltages (Figure 3A) (205,
239). The combination of these electrodes results in a three-dimensional
electric field and traps ions in the field (240). Similar to the FTICR, each m/z is
directly related to a certain rf voltage (Figure 3B) (205, 239). By increasing the
rf voltage, different m/z ions are excited separately (205, 239, 240). Under a
certain rf voltage, the trajectories of the ions with corresponding m/z become
unstable, which cause the ions to be ejected from the cell to strike a standard
detector; meanwhile, those ions with stable trajectories are still trapped inside
the electrodes (205, 239).
Due to the high sensitivity, throughput, ease of use and low cost, the
174
quadrupole ion trap has become the workhorse for routine lab applications.
Figure 3. A schematic of the basic set up of a QIT mass analyzer is shown. (A) The
major components of a QIT analyzer are shown; (B) A schematic diagram of positively
charged ions responding to the applied RF and DC field is given (205, 239, 240).
Collision-induced dissociation (CID)
CID is an effective fragmentation method used to induce the direct
collision between a sample molecule with a rare gas atom in the cell (207, 208).
Sample molecules dissociate as a result of the collision between a fast-speed
sample and a neutral species (207, 208).
175
Figure 4. An analogy is given, illustrating a typical CID experiment (207, 208).
As demonstrated in Figure 4, the target molecule will first be accelerated,
also known as being "activated", and then it will enter the collision cell filled
with the inert gas argon (207, 208). The target molecule will have multiple
collisions with gas atoms, and the kinetic energy of the precursor will transfer
to internal energy (207, 208). When the internal energy is accumulated enough
to overcome the energy barrier, dissociation occurs (207, 208). The precursor
breaks down from isomerization or direct breakage, and the resulting
fragments are further captured by the detector (207, 208). In peptide-CID
studies, the production of peptide fragments that only differ in a few amino
acids occurs, providing sequencing information (241). This results in a gradual
build-up of internal energy, and eventually causes the dissociation of parent
ions. CID is the most used method in the tandem mass spectrometry analysis
of peptides and proteins.
176
Infrared multiphoton dissociation (IRMPD)
Infrared multiphoton dissociation (IRMPD) is also a powerful technique
used in tandem mass spectrometry for the structural characterization of
gaseous ions. Its mechanism is similar to the traditional infrared spectroscopy;
however, it has higher sensitivity. This technique is normally implemented with
trapping instrumentation, in which the isolated precursor ions are trapped and
maintained in the laser beam for extended periods of time (208, 213). Similar
to IR spectroscopy, the ions can absorb light of a certain wavelength (Figure
5). After absorbing a photon, the ions are promoted to the absorbing mode
from the ground state, and then the absorbed energy is redistributed to the
vibrational degrees of freedom within the molecule via intramolecular
vibrational relaxation (IVR) (Figure 5) (208, 213). IVR continuously
depopulates the absorbing states and allows photon absorption to occur over
and over again (Figure 5) (208, 213). This repeating process results in a
gradual build up of internal energy, and eventually induces dissociation of the
parent ions (Figure 5) (208, 213). The charged fragments are later detected
with the second stage MS. Each MS/MS result from IRMPD technique is
directly related to a certain wavelength IR laser. After plotting the extent of ion
dissociation versus wavelength, the IRMPD spectrum is obtained (Figure 5).
Normally, an intense laser source is used in this technique, such as a free
electron laser, which is generated by projecting a high speed electron into a
periodic magnetic structure (208, 213).
177
Figure 5. An schematic illustration is given, illustrating a typical IRMPD (208,
213).
178
REFERENCES
1.
Sharp, P. A. (2009) The centrality of RNA, Cell 136, 577-580.
2.
Crick, F. (1970) Central dogma of molecular biology, Nature 227,
561-563.
3.
Cooper, S. (1981) The central dogma of cell biology, Cell Biol. Int. Rep.
5, 539-549.
4.
Yonath, A. (2003) Structural insight into functional aspects of ribosomal
RNA targeting, Chembiochem 4, 1008-1017.
5.
Ben-Shem, A., Jenner, L., Yusupova, G., and Yusupov, M. (2010)
Crystal structure of the eukaryotic ribosome, Science 330, 1203-1209.
6.
Feng, S., Chen, Y., and Gao, Y. G. (2013) Crystal structure of 70S
ribosome with both cognate tRNAs in the E and P sites representing an
authentic elongation complex, PLoS One 8, e58829.
7.
Jenner, L., Melnikov, S., de Loubresse, N. G., Ben-Shem, A., Iskakova,
M., Urzhumtsev, A., Meskauskas, A., Dinman, J., Yusupova, G., and
Yusupov, M. (2012) Crystal structure of the 80S yeast ribosome, Curr.
Opin. Struct. Biol. 22, 759-767.
8.
Guerrier-Takada, C., and Altman, S. (1984) Catalytic activity of an RNA
molecule prepared by transcription in vitro, Science 223, 285-286.
9.
Guerriertakada, C., Gardiner, K., Marsh, T., Pace, N., and Altman, S.
(1983) The RNA moiety of ribonuclease-P Is the catalytic subunit of the
enzyme, Cell 35, 849-857.
179
10.
Agrawal, N., Dasaradhi, P. V., Mohmmed, A., Malhotra, P., Bhatnagar, R.
K., and Mukherjee, S. K. (2003) RNA interference: biology, mechanism,
and applications, Microbiol. Mol. Biol. Rev. 67, 657-685.
11.
Hannon, G. J. (2002) RNA interference, Nature 418, 244-251.
12.
McManus, M. T., Petersen, C. P., Haines, B. B., Chen, J., and Sharp, P.
A. (2002) Gene silencing using micro-RNA designed hairpins, RNA 8,
842-850.
13.
Ernst, C., and Morton, C. C. (2013) Identification and function of long
non-coding RNA, Front Cell Neurosci. 7, 168.
14.
Qi, W., Song, X., and Li, L. (2013) Long non-coding RNA-guided
regulation in organisms, Sci. China Life Sci. 56, 891-896.
15.
Ma, L., Bajic, V. B., and Zhang, Z. (2013) On the classification of long
non-coding RNAs, RNA Biol. 10, 925-933.
16.
Lundblad, R. L., and Macdonald, F. (2010) Handbook of biochemistry
and molecular biology, CRC Press, Boca Raton.
17.
Thirumalai, D. (1998) Native secondary structure formation in RNA may
be a slave to tertiary folding, Proc. Natl. Acad. Sci. U. S. A. 95,
11506-11508.
18.
Schuster, P., Stadler, P. F., and Renner, A. (1997) RNA structures and
folding: from conventional to new issues in structure predictions, Curr.
Opin. Struct. Biol. 7, 229-235.
19.
Woese, C. R., Winker, S., and Gutell, R. R. (1990) Architecture of
180
ribosomal RNA: constraints on the sequence of "tetra-loops", Proc. Natl.
Acad. Sci. U. S. A. 87, 8467-8471.
20.
Korostelev, A., and Noller, H. F. (2007) The ribosome in focus: new
structures bring new insights, Trends Biochem. Sci. 32, 434-441.
21.
Warner, J. R., Vilardell, J., and Sohn, J. H. (2001) Economics of
ribosome biosynthesis, Cold Spring Harb. Symp. Quant. Biol. 66,
567-574.
22.
Warner, J. R. (1999) The economics of ribosome biosynthesis in yeast,
Trends Biochem. Sci. 24, 437-440.
23.
Drygin, D., Siddiqui-Jain, A., O'Brien, S., Schwaebe, M., Lin, A.,
Bliesath, J., Ho, C. B., Proffitt, C., Trent, K., Whitten, J. P., Lim, J. K.,
Von Hoff, D., Anderes, K., and Rice, W. G. (2009) Anticancer activity of
CX-3543: a direct inhibitor of rRNA biogenesis, Cancer Res. 69,
7653-7661.
24.
Marcel, V., Catez, F., and Diaz, J. J. (2013) Ribosomes: the future of
targeted therapies?, Oncotarget 4, 1554-1555.
25.
Obrig, T. G., Culp, W. J., McKeehan, W. L., and Hardesty, B. (1971) The
mechanism by which cycloheximide and related glutarimide antibiotics
inhibit peptide synthesis on reticulocyte ribosomes, J. Biol. Chem. 246,
174-181.
26.
Schneider-Poetsch, T., Ju, J., Eyler, D. E., Dang, Y., Bhat, S., Merrick,
W. C., Green, R., Shen, B., and Liu, J. O. (2010) Inhibition of eukaryotic
181
translation elongation by cycloheximide and lactimidomycin, Nat. Chem.
Biol. 6, 209-217.
27.
Moazed, D., and Noller, H. F. (1987) Interaction of antibiotics with
functional sites in 16S ribosomal RNA, Nature 327, 389-394.
28.
Schroeder, R., Waldsich, C., and Wank, H. (2000) Modulation of RNA
function by aminoglycoside antibiotics, EMBO J. 19, 1-9.
29.
Katz, L., and Ashley, G. W. (2005) Translation and protein synthesis:
macrolides, Chem. Rev. 105, 499-528.
30.
Xiong, L., Korkhin, Y., and Mankin, A. S. (2005) Binding site of the
bridged macrolides in the Escherichia coli ribosome, Antimicrob. Agents
Chemother. 49, 281-288.
31.
Rodriguez-Fonseca, C., and Garrett, R. A. (1998) Chemical and
enzymatic probing of antibiotic-ribosome complexes, Methods Mol. Biol.
77, 311-323.
32.
Franceschi, F. (2007) Back to the future: the ribosome as an antibiotic
target, Future Microbiol. 2, 571-574.
33.
Houghton, J. L., Green, K. D., Chen, W., and Garneau-Tsodikova, S.
(2010) The future of aminoglycosides: the end or renaissance?,
Chembiochem 11, 880-902.
34.
Davies, J., and Wright, G. D. (1997) Bacterial resistance to
aminoglycoside antibiotics, Trends Microbiol. 5, 234-240.
35.
Reardon, S. (2014) Antibiotic resistance sweeping developing world,
182
Nature 509, 141-142.
36.
McKenna, M. (2013) Antibiotic resistance: the last resort, Nature 499,
394-396.
37.
Organization, W. H. (2014) Antimicrobial resistance: global report on
surveillance 2014, 257.
38.
Gould, I. M., and Bal, A. M. (2013) New antibiotic agents in the pipeline
and how they can help overcome microbial resistance, Virulence 4,
185-191.
39.
Wilson, D. N. (2014) Ribosome-targeting antibiotics and mechanisms of
bacterial resistance, Nat. Rev. Microbiol. 12, 35-48.
40.
Belova, L., Tenson, T., Xiong, L., McNicholas, P. M., and Mankin, A. S.
(2001) A novel site of antibiotic action in the ribosome: interaction of
evernimicin with the large ribosomal subunit, Proc. Natl. Acad. Sci. U. S.
A. 98, 3726-3731.
41.
Mikolajka, A., Liu, H., Chen, Y., Starosta, A. L., Marquez, V., Ivanova, M.,
Cooperman, B. S., and Wilson, D. N. (2011) Differential effects of
thiopeptide and orthosomycin antibiotics on translational GTPases,
Chem. Biol. 18, 589-600.
42.
Pioletti, M., Schlunzen, F., Harms, J., Zarivach, R., Glühmann, M., Avila,
H., Bashan, A., Bartels, H., Auerbach, T., Jacobi, C., Hartsch, T., Yonath,
A., and Franceschi, F. (2001) Crystal structures of complexes of the
small ribosomal subunit with tetracycline, edeine and IF3, EMBO J. 20,
183
1829-1839.
43.
Feldman, M. B., Terry, D. S., Altman, R. B., and Blanchard, S. C. (2010)
Aminoglycoside activity observed on single pre-translocation ribosome
complexes, Nat. Chem. Biol. 6, 54-62.
44.
Wang, L., Pulk, A., Wasserman, M. R., Feldman, M. B., Altman, R. B.,
Cate, J. H., and Blanchard, S. C. (2012) Allosteric control of the
ribosome by small-molecule antibiotics, Nat. Struct. Mol. Biol. 19,
957-963.
45.
Ogle, J. M., Brodersen, D. E., Clemons, W. M., Jr., Tarry, M. J., Carter, A.
P., and Ramakrishnan, V. (2001) Recognition of cognate transfer RNA
by the 30S ribosomal subunit, Science 292, 897-902.
46.
Ogle, J. M., and Ramakrishnan, V. (2005) Structural insights into
translational fidelity, Annu. Rev. Biochem. 74, 129-177.
47.
Demeshkina, N., Jenner, L., Westhof, E., Yusupov, M., and Yusupova,
G. (2012) A new understanding of the decoding principle on the
ribosome, Nature 484, 256-259.
48.
Carter, A. P., Clemons, W. M., Brodersen, D. E., Morgan-Warren, R. J.,
Wimberly, B. T., and Ramakrishnan, V. (2000) Functional insights from
the structure of the 30S ribosomal subunit and its interactions with
antibiotics, Nature 407, 340-348.
49.
Thom, G., and Prescott, C. D. (1997) The selection in vivo and
characterization of an RNA recognition motif for spectinomycin, Bioorg.
184
Med. Chem. 5, 1081-1086.
50.
Borovinskaya, M. A., Shoji, S., Holton, J. M., Fredrick, K., and Cate, J. H.
(2007) A steric block in translation caused by the antibiotic
spectinomycin, ACS Chem. Biol. 2, 545-552.
51.
Okamoto, S., Tamaru, A., Nakajima, C., Nishimura, K., Tanaka, Y.,
Tokuyama, S., Suzuki, Y., and Ochi, K. (2007) Loss of a conserved
7-methylguanosine modification in 16S rRNA confers low-level
streptomycin resistance in bacteria, Mol. Microbiol. 63, 1096-1106.
52.
Bulkley, D., Johnson, F., and Steitz, T. A. (2012) The antibiotic
thermorubin inhibits protein synthesis by binding to inter-subunit bridge
B2a of the ribosome, J. Mol. Biol. 416, 571-578.
53.
Cavalleri, B., Turconi, M., and Pallanza, R. (1985) Synthesis and
antibacterial activity of some derivatives of the antibiotic thermorubin, J.
Antibiot. 38, 1752-1760.
54.
Schlünzen, F., Zarivach, R., Harms, J., Bashan, A., Tocilj, A., Albrecht,
R., Yonath, A., and Franceschi, F. (2001) Structural basis for the
interaction of antibiotics with the peptidyl transferase centre in
eubacteria, Nature 413, 814-821.
55.
Hansen, J. L., Moore, P. B., and Steitz, T. A. (2003) Structures of five
antibiotics bound at the peptidyl transferase center of the large
ribosomal subunit, J. Mol. Biol. 330, 1061-1075.
56.
Bulkley, D., Innis, C. A., Blaha, G., and Steitz, T. A. (2010) Revisiting the
185
structures of several antibiotics bound to the bacterial ribosome, P. Natl.
Acad. Sci. U. S. A. 107, 17158-17163.
57.
Dunkle, J. A., Xiong, L., Mankin, A. S., and Cate, J. H. (2010) Structures
of the Escherichia coli ribosome with antibiotics bound near the peptidyl
transferase center explain spectra of drug action, Proc. Natl. Acad. Sci.
U. S. A. 107, 17152-17157.
58.
Mankin, A. S., and Garrett, R. A. (1991) Chloramphenicol resistance
mutations in the single 23S rRNA gene of the archaeon Halobacterium
halobium, J. Bacteriol. 173, 3559-3563.
59.
Champney, W. S. (2006) The other target for ribosomal antibiotics:
inhibition of bacterial ribosomal subunit formation, Infect. Disord. Drug
Targets 6, 377-390.
60.
Ermolenko, D. N., Cornish, P. V., Ha, T., and Noller, H. F. (2013)
Antibiotics that bind to the A site of the large ribosomal subunit can
induce mRNA translocation, RNA 19, 158-166.
61.
Long, K. S., Poehlsgaard, J., Kehrenberg, C., Schwarz, S., and Vester,
B. (2006) The Cfr rRNA methyltransferase confers resistance to
phenicols,
lincosamides,
oxazolidinones,
pleuromutilins,
and
streptogramin A antibiotics, Antimicrob. Agents Chemother. 50,
2500-2505.
62.
Wilson, D. N. (2009) The A-Z of bacterial translation inhibitors, Crit. Rev.
Biochem. Mol. 44, 393-433.
186
63.
Tu, D., Blaha, G., Moore, P. B., and Steitz, T. A. (2005) Structures of
MLSBK antibiotics bound to mutated large ribosomal subunits provide a
structural explanation for resistance, Cell 121, 257-270.
64.
Siibak, T., Peil, L., Xiong, L., Mankin, A., Remme, J., and Tenson, T.
(2009)
Erythromycin-
and
chloramphenicol-induced
ribosomal
assembly defects are secondary effects of protein synthesis inhibition,
Antimicrob. Agents Chemother. 53, 563-571.
65.
Tenson, T., Xiong, L., Kloss, P., and Mankin, A. S. (1997) Erythromycin
resistance peptides selected from random peptide libraries, J. Biol.
Chem. 272, 17425-17430.
66.
Leach, K. L., Swaney, S. M., Colca, J. R., McDonald, W. G., Blinn, J. R.,
Thomasco, L. M., Gadwood, R. C., Shinabarger, D., Xiong, L., and
Mankin, A. S. (2007) The site of action of oxazolidinone antibiotics in
living bacteria and in human mitochondria, Mol. Cell. 26, 393-402.
67.
Wilson, D. N., Schluenzen, F., Harms, J. M., Starosta, A. L., Connell, S.
R., and Fucini, P. (2008) The oxazolidinone antibiotics perturb the
ribosomal peptidyl-transferase center and effect tRNA positioning, Proc.
Natl. Acad. Sci. U. S. A. 105, 13339-13344.
68.
Ippolito, J. A., Kanyo, Z. F., Wang, D., Franceschi, F. J., Moore, P. B.,
Steitz, T. A., and Duffy, E. M. (2008) Crystal structure of the
oxazolidinone antibiotic linezolid bound to the 50S ribosomal subunit, J.
Med. Chem. 51, 3353-3356.
187
69.
LaMarre, J., Mendes, R. E., Szal, T., Schwarz, S., Jones, R. N., and
Mankin, A. S. (2013) The genetic environment of the cfr gene and the
presence of other mechanisms account for the very high linezolid
resistance
of
Staphylococcus
epidermidis
isolate
426-3147L,
Antimicrob. Agents Chemother. 57, 1173-1179.
70.
Billal, D. S., Feng, J., Leprohon, P., Legare, D., and Ouellette, M. (2011)
Whole genome analysis of linezolid resistance in Streptococcus
pneumoniae reveals resistance and compensatory mutations, BMC
Genomics 12.
71.
LaMarre, J. M., Howden, B. P., and Mankin, A. S. (2011) Inactivation of
the indigenous methyltransferase RlmN in Staphylococcus aureus
increases linezolid resistance, Antimicrob. Agents Chemother. 55,
2989-2991.
72.
Hansen, J. L., Schmeing, T. M., Moore, P. B., and Steitz, T. A. (2002)
Structural insights into peptide bond formation, Proc. Natl. Acad. Sci. U.
S. A. 99, 11670-11675.
73.
Fredrick, K., and Noller, H. F. (2003) Catalysis of ribosomal
translocation by sparsomycin, Science 300, 1159-1162.
74.
Lippert, B. (1999) Cisplatin : chemistry and biochemistry of a leading
anticancer drug, Verlag Helvetica Chimica Acta; Wiley-VCH, Zürich
Weinheim ; New York.
75.
Milburn, G. H. W. T., M. R. . (1966) The crystal structures of cis- and
188
trans-dichlorodiammineplatinum(II), J. Chem. Soc. A. 1609-1615
76.
Rosenberg, B. (1978) Platinum complexes for treatment of cancer,
Interdiscipl. Sci. Rev. 3, 134-147.
77.
Rosenberg, B., VanCamp, L., Trosko, J. E., and Mansour, V. H. (1969)
Platinum compounds: a new class of potent antitumour agents, Nature
222, 385-386.
78.
Ho, T. L. (1975) Hard soft acids bases (HSAB) principle and
organic-chemistry, Chem. Rev. 75, 1-20.
79.
Leng, M., and Brabec., V. (1994) DNA adducts of cisplatin, transplatin
and platinum intercalating drugs, in DNA adducts: identification and
biological significance, International Agency for Research on Cancer
125, 339-348.
80.
Farrell, N., Kelland, L. R., Roberts, J. D. and van Beusichem, M. (1992)
Activation of the trans geometry in platinum antitumor complexes: a
survey of the cytotoxicity of trans complexes containing planar ligands
in murine L1210 and human tumor panels and studies on their
mechanism of action., Cancer Res. 52, 5065-5072.
81.
Jung, Y., and Lippard, S. J. (2007) Direct cellular responses to
platinum-induced DNA damage, Chem. Rev. 107, 1387-1407.
82.
Judson, I., and Kelland, L. R. (2000) New developments and
approaches in the platinum arena, Drugs 59 Suppl. 4, 29-36; discussion
37-28.
189
83.
Gately, D. R., and Howell, S. B. (1993) Cellular accumulation of the
anticancer agent cisplatin: a review., Br. J. Cancer. 67, 1171-1176.
84.
Binks, S. P., and Dobrota, M. (1990) Kinetics and mechanism of uptake
of platinum-based pharmaceuticals by the rat small intestine., Biochem.
Pharmacol. 40, 1329-1336.
85.
Hromas, R. A., North, J. A., and Burns, C. P. (1987) Decreased cisplatin
uptake by resistant L1210 leukemia cells., Cancer Lett. 36, 197-201.
86.
Gale, G. R., Morris, C. R., Atkins, L. M., and Smith, A. B. (1973) Binding
of an antitumor platinum compound to cells as influenced by physical
factors and pharmacologically active agents., Cancer Res. 33, 813-818.
87.
Fuertes, M. A., Castilla, J., Alonso, C., and Pérez, J. M. (2003) Cisplatin
biochemical mechanism of action: from cytotoxicity to induction of cell
death through interconnections between apoptotic and necrotic
pathways, Curr. Med. Chem. 10, 257-266.
88.
Jamieson, E. R., and Lippard, S. J. (1999) Structure, recognition, and
processing of cisplatin-DNA Adducts, Chem. Rev. 99, 2467-2498.
89.
Ishida, S., Lee, J., Thiele, D. J., and Herskowitz, I. (2002) Uptake of the
anticancer drug cisplatin mediated by the copper transporter Ctr1 in
yeast and mammals, Proc. Natl. Acad. Sci. U. S. A. 99, 14298-14302.
90.
Lin, X., Okuda, T., Holzer, A., and Howell, S. B. (2002) The copper
transporter Ctr1 regulates cisplatin uptake in Saccharomyces cerevisiae,
Mol. Pharmacol. 62, 1154-1159.
190
91.
Holzer, A. K., Samimi, G., Katano, K., Naerdemann, W., Lin, X., Safaei,
R., and Howell, S. B. (2004) The copper influx transporter human
copper transport protein 1 regulates the uptake of cisplatin in human
ovarian carcinoma cells, Mol. Pharmacol. 66, 817-823.
92.
Ciarimboli, G., Ludwig, T., Lang, D., Pavenstadt, H., Koepsell, H.,
Piechota, H. J., Haier, J., Jaehde, U., Zisowsky, J., and Schlatter, E.
(2005) Cisplatin nephrotoxicity is critically mediated via the human
organic cation transporter 2, Am. J. Pathol. 167, 1477-1484.
93.
Jordan, P., and Carmo-Fonseca, M. (2000) Molecular mechanisms
involved in cisplatin cytotoxicity, Cell Mol. Life Sci. 57, 1229-1235.
94.
Perez, R. P. (1998) Cellular and molecular determinants of cisplatin
resistance, Eur. J. Cancer 34, 1535-1542.
95.
Yang, X. L., and Wang, A. H. J. (1999) Structural studies of
atom-specific anticancer drugs acting on DNA, Pharmacol. Therapeut.
83, 181-215.
96.
Brouwer, J., Vandeputte, P., Fichtingerschepman, A. M. J., and Reedijk,
J. (1981) Base-pair substitution hotspots in GAG and GCG
nucleotide-sequences
in
Escherichia
coli
K-12
induced
by
cis-diamminedichloroplatinum-(II), Proc. Natl. Acad. Sci.-Biol. 78,
7010-7014.
97.
Popoff, S. C., Beck, D. J., and Rupp, W. D. (1987) Repair of plasmid
DNAdamaged
in
vitro
with
cis-diamminedichloroplatinum(II)
or
191
trans-diamminedichloroplatinum(II) in Escherichia coli, Mutat. Res. 183,
129-137.
98.
Payet, D., Gaucheron, F., Sip, M., and Leng, M. (1993) Instability of the
monofunctional
adducts
in
cis-[Pt(NH3)2(N7-N-methyl-2-diazapyrenium)Cl]2+-modified DNA: rates
of cross-linking reactions in cis-platinum-modified DNA, Nucleic Acids
Res. 21, 5846-5851.
99.
Barnham, K. J., Bernersprice, S. J., Frenkiel, T. A., Frey, U., and Sadler,
P. J. (1995) Platination pathways for reactions of cisplatin with GG
single-stranded and double-stranded decanucleotides, Angew. Chem.
Int. Edit. 34, 1874-1877.
100.
Kartalou, M., and Essigmann, J. M. (2001) Recognition of cisplatin
adducts by cellular proteins, Mutat. Res. 478, 1-21.
101.
Cohen, G. L., Ledner, J. A., Bauer, W. R., Ushay, H. M., Caravana, C.,
and
Lippard,
S.
J.
(1980)
Sequence
dependent
binding
of
cis-dichlorodiammineplatinum(II) to DNA, J. Am. Chem. Soc. 102,
2487-2488.
102.
Poklar, N., Pilch, D. S., Lippard, S. J., Redding, E. A., Dunham, S. U.,
and Breslauer, K. J. (1996) Influence of cisplatin intrastrand crosslinking
on the conformation, thermal stability, and energetics of a 20-mer DNA
duplex, Proc. Natl. Acad. Sci. U. S. A. 93, 7606-7611.
103.
Cohen, G. L., Bauer, W. R., Barton, J. K., and Lippard, S. J. (1979)
192
Binding of cis- and trans-dichlorodiammineplatinum(II) to DNA:
evidence for unwinding and shortening of the double helix, Science 203,
1014-1016.
104.
Gelasco, A., and Lippard, S. J. (1998) NMR solution structure of a DNA
dodecamer duplex containing a cis-diammineplatinum(II) d(GpG)
intrastrand cross-link, the major adduct of the anticancer drug cisplatin,
Biochemistry 37, 9230-9239.
105.
Coste, F., Malinge, J. M., Serre, L., Shepard, W., Roth, M., Leng, M.,
and Zelwer, C. (1999) Crystal structure of a double-stranded DNA
containing a cisplatin interstrand cross-link at 1.63 angstrom resolution:
hydration at the platinated site, Nucleic Acids Res. 27, 1837-1846.
106.
Vangarderen, C. J., and Vanhoute, L. P. A. (1994) The solution structure
of a DNA duplex containing the cis-Pt(NH3)2[d(-GTG-)-N7(G), N7(G)]
adduct,
as
determined
with
high-field
NMR
and
molecular
mechanics/dynamics, Eur. J. Biochem. 225, 1169-1179.
107.
Wang, D., and Lippard, S. J. (2005) Cellular processing of platinum
anticancer drugs, Nat. Rev. Drug Discov. 4, 307-320.
108.
Gonzalez, V. M., Fuertes, M. A., Alonso, C., and Perez, J. M. (2001) Is
cisplatin-induced cell death always produced by apoptosis?, Mol.
Pharmacol. 59, 657-663.
109.
Lieberthal, W., Triaca, V., and Levine, J. (1996) Mechanisms of death
induced by cisplatin in proximal tubular epithelial cells: apoptosis vs.
193
necrosis, Am. J. Physiol-Renal 270, F700-F708.
110.
Kanduc, D., Mittelman, A., Serpico, R., Sinigaglia, E., Sinha, A. A.,
Natale, C., Santacroce, R., Di Corcia, M. G., Lucchese, A., Dini, L., Pani,
P., Santacroce, S., Simone, S., Bucci, R., and Farber, E. (2002) Cell
death: apoptosis versus necrosis, Int. J. Oncol. 21, 165-170.
111.
Farber, E. (1994) Programmed cell death: necrosis versus apoptosis,
Mod. Pathol. 7, 605-609.
112.
Herceg, Z., and Wang, Z. Q. (2001) Functions of poly(ADP-ribose)
polymerase (PARP) in DNA repair, genomic integrity and cell death,
Mutat. Res. 477, 97-110.
113.
Georgiou, C. D., Papapostolou, I., and Grintzalis, K. (2009) Protocol for
the quantitative assessment of DNA concentration and damage
(fragmentation and nicks), Nat. Protoc. 4, 125-131.
114.
Sellin Jeffries, M. K., Kiss, A. J., Smith, A. W., and Oris, J. T. (2014) A
comparison
of
commercially-available
automated
and
manual
extraction kits for the isolation of total RNA from small tissue samples,
BMC Biotechnol. 14, 94.
115.
Hägerlöf, M., Papsai, P., Chow, C. S., and Elmroth, S. K. (2006) More
pronounced salt dependence and higher reactivity for platination of the
hairpin r(CGCGUUGUUCGCG) compared with d(CGCGTTGTTCGCG),
J. Biol. Inorg. Chem. 11, 974-990.
116.
Papsai, P., Aldag, J., Persson, T., and Elmroth, S. K. (2006) Kinetic
194
preference for interaction of cisplatin with the G-C-rich wobble basepair
region in both tRNAAla and MhAla, Dalton Trans., 3515-3517.
117.
Papsai, P., Snygg, A. S., Aldag, J., and Elmroth, S. K. (2008) Platination
of full length tRNAAla and truncated versions of the acceptor stem and
anticodon loop, Dalton Trans., 5225-5234.
118.
Hostetter, A. A., Chapman, E. G., and DeRose, V. J. (2009) Rapid
cross-linking of an RNA internal loop by the anticancer drug cisplatin, J.
Am. Chem. Soc. 131, 9250-9257.
119.
Rijal, K., and Chow, C. S. (2009) A new role for cisplatin: probing
ribosomal RNA structure, Chem. Commun., 107-109.
120.
Chapman, E. G., and DeRose, V. J. (2010) Enzymatic processing of
platinated RNAs, J. Am. Chem. Soc. 132, 1946-1952.
121.
Hostetter, A. A., Osborn, M. F., and DeRose, V. J. (2012) RNA-Pt
adducts following cisplatin treatment of Saccharomyces cerevisiae,
ACS Chem. Biol. 7, 218-225.
122.
Rijal, K., Bao, X., and Chow, C. S. (2014) Amino acid-linked platinum(ii)
analogues have altered specificity for RNA compared to cisplatin, Chem.
Commun. 50, 3918-3920.
123.
Pascoe, J. M., and Roberts, J. J. (1974) Interactions between
mammalian cell DNA and inorganic platinum compounds. I. DNA
interstrand cross-linking and cytotoxic properties of platinum(II)
compounds, Biochem. Pharmacol. 23, 1359-1365.
195
124.
Pascoe, J. M., and Roberts, J. J. (1974) Interactions between
mammalian cell DNA and inorganic platinum compounds. II. Interstrand
cross-linking of isolated and cellular DNA by platinum(IV) compounds,
Biochem. Pharmacol. 23, 1345-1357.
125.
Heminger, K. A., Hartson, S. D., Rogers, J., and Matts, R. L. (1997)
Cisplatin inhibits protein synthesis in rabbit reticulocyte lysate by
causing an arrest in elongation, Arch. Biochem. Biophys. 344, 200-207.
126. Lucks, J. B., Mortimer, S. A., Trapnell, C., Luo, S., Aviran, S., Schroth, G.
P., Pachter, L., Doudna, J. A., and Arkin, A. P. (2011) Multiplexed RNA
structure characterization with selective 2'-hydroxyl acylation analyzed
by primer extension sequencing (SHAPE-Seq), Proc. Natl. Acad. Sci. U.
S. A. 108, 11063-11068.
127.
Wilkinson, K. A., Merino, E. J., and Weeks, K. M. (2006) Selective
2'-hydroxyl
acylation
analyzed
by
primer
extension
(SHAPE):
quantitative RNA structure analysis at single nucleotide resolution, Nat.
Protoc. 1, 1610-1616.
128.
Brunel, C., and Romby, P. (2000) Probing RNA structure and
RNA-ligand complexes with chemical probes, Methods Enzymol. 318,
3-21.
129.
Ehresmann, C., Baudin, F., Mougel, M., Romby, P., Ebel, J. P., and
Ehresmann, B. (1987) Probing the structure of RNAs in solution,
Nucleic Acids Res. 15, 9109-9128.
196
130.
Sakakibara, Y., and Chow, C. S. (2012) Role of pseudouridine in
structural rearrangements of helix 69 during bacterial ribosome
assembly, ACS Chem. Biol. 7, 871-878.
131.
Ziehler, W. A., and Engelke, D. R. (2001) Probing RNA structure with
chemical reagents and enzymes, Curr. Protoc. Nucleic Acid Chem.
Chapter 6, Unit 6 1.
132.
Knapp, G. (1989) Enzymatic approaches to probing of RNA secondary
and tertiary structure, Methods Enzymol. 180, 192-212.
133.
Donis-Keller, H., Maxam, A. M., and Gilbert, W. (1977) Mapping
adenines, guanines, and pyrimidines in RNA, Nucleic Acids Res. 4,
2527-2538.
134.
Daugaard, G., and Abildgaard, U. (1989) Cisplatin nephrotoxicity. A
review, Cancer Chemother. Pharmacol. 25, 1-9.
135.
Cvitkovic, E. (1998) Cumulative toxicities from cisplatin therapy and
current cytoprotective measures, Cancer Treat Rev. 24, 265-281.
136.
Screnci, D., and McKeage, M. J. (1999) Platinum neurotoxicity: clinical
profiles, experimental models and neuroprotective approaches, J. Inorg.
Biochem. 77, 105-110.
137.
Galluzzi, L., Senovilla, L., Vitale, I., Michels, J., Martins, I., Kepp, O.,
Castedo, M., and Kroemer, G. (2012) Molecular mechanisms of
cisplatin resistance, Oncogene 31, 1869-1883.
138.
Shen, D. W., Pouliot, L. M., Hall, M. D., and Gottesman, M. M. (2012)
197
Cisplatin resistance: a cellular self-defense mechanism resulting from
multiple epigenetic and genetic changes, Pharmacol. Rev. 64, 706-721.
139.
Wong, E., and Giandomenico, C. M. (1999) Current status of
platinum-based antitumor drugs, Chem. Rev. 99, 2451-2466.
140.
Bancroft, D. P., Lepre, C. A., and Lippard, S. J. (1990) Pt-195 NMR
kinetic and mechanistic studies of cis-diamminedichloroplatinum and
trans-diamminedichloroplatinum(II) binding to DNA, J. Am. Chem. Soc.
112, 6860-6871.
141.
Hindmarsh, K., House, D. A., and Turnbull, M. M. (1997) The hydrolysis
products of cis-diamminedichloroplatinum(II) .9. Chloride and bromide
anation kinetics for some [Pt-II(N)2(OH2)2]2+ complexes and the
structures of [Pt-IVBr4(N)2] ((N)2=en, tn), Inorg. Chim. Acta 257, 11-18.
142.
Harrap, K. R. (1995) Initiatives with platinum- and quinazoline-based
antitumor molecules--Fourteenth Bruce F. Cain Memorial Award Lecture,
Cancer Res. 55, 2761-2768.
143.
Calvert, A. H., Harland, S. J., Newell, D. R., Siddik, Z. H., and Harrap, K.
R. (1985) Phase I studies with carboplatin at the Royal Marsden
Hospital, Cancer Treat Rev. 12 Suppl. A, 51-57.
144.
Harrap, K. R. (1985) Preclinical studies identifying carboplatin as a
viable cisplatin alternative, Cancer Treat Rev. 12 Suppl A, 21-33.
145.
Duffull, S. B., and Robinson, B. A. (1997) Clinical pharmacokinetics and
dose optimisation of carboplatin, Clin. Pharmacokinet. 33, 161-183.
198
146.
Riccardi, R., Riccardi, A., Lasorella, A., Di Rocco, C., Carelli, G.,
Tornesello, A., Servidei, T., Iavarone, A., and Mastrangelo, R. (1994)
Clinical pharmacokinetics of carboplatin in children, Cancer Chemother.
Pharmacol. 33, 477-483.
147. van der Vijgh, W. J. (1991) Clinical pharmacokinetics of carboplatin, Clin.
Pharmacokinet. 21, 242-261.
148.
Oguri, S., Sakakibara, T., Mase, H., Shimizu, T., Ishikawa, K., Kimura,
K., and Smyth, R. D. (1988) Clinical pharmacokinetics of carboplatin, J.
Clin. Pharmacol. 28, 208-215.
149.
Chu, G., Mantin, R., Shen, Y. M., Baskett, G., and Sussman, H. (1993)
Massive cisplatin overdose by accidental substitution for carboplatin.
Toxicity and management, Cancer 72, 3707-3714.
150.
Spingler, B., Whittington, D. A., and Lippard, S. J. (2001) 2.4 Å crystal
structure of an oxaliplatin 1,2-d(GpG) intrastrand cross-link in a DNA
dodecamer duplex, Inorg. Chem. 40, 5596-5602.
151.
Silverman, A. P., Bu, W., Cohen, S. M., and Lippard, S. J. (2002) 2.4- Å
crystal
structure
of
the
asymmetric
platinum
complex
[Pt(ammine)(cyclohexylamine)]2+ bound to a dodecamer DNA duplex, J.
Biol. Chem. 277, 49743-49749.
152.
Lovejoy, K. S., Todd, R. C., Zhang, S., McCormick, M. S., D'Aquino, J.
A., Reardon, J. T., Sancar, A., Giacomini, K. M., and Lippard, S. J.
(2008)
cis-Diammine(pyridine)chloroplatinum(II),
a
monofunctional
199
platinum(II) antitumor agent: uptake, structure, function, and prospects,
Proc. Natl. Acad. Sci. U. S. A. 105, 8902-8907.
153.
Kidani, Y., Suzuki, K., Noji, M., and Tashiro, T. (1989) Antitumor activity
of
steroid-containing
platinum
1R,2R-cyclohexanediamine
and
(II)
complexes
of
2-(aminomethyl)-cyclohexylamine
isomers against leukemia L1210, Biomed. Pharmacother. 43, 261-264.
154.
Tashiro, T., Kawada, Y., Sakurai, Y., and Kidani, Y. (1989) Antitumor
activity
of
a
new
platinum
complex,
oxalato
(trans-l-1,2-diaminocyclohexane)platinum (II): new experimental data,
Biomed. Pharmacother. 43, 251-260.
155.
Extra, J. M., Espie, M., Calvo, F., Ferme, C., Mignot, L., and Marty, M.
(1990) Phase I study of oxaliplatin in patients with advanced cancer,
Cancer Chemother. Pharmacol. 25, 299-303.
156.
Mathé, G., Kidani, Y., Segiguchi, M., Eriguchi, M., Fredj, G., Peytavin,
G., Misset, J. L., Brienza, S., de Vassals, F., Chenu, E., and et al. (1989)
Oxalato-platinum or 1-OHP, a third-generation platinum complex: an
experimental and clinical appraisal and preliminary comparison with
cis-platinum and carboplatinum, Biomed. Pharmacother. 43, 237-250.
157.
Rixe, O., Ortuzar, W., Alvarez, M., Parker, R., Reed, E., Paull, K., and
Fojo, T. (1996) Oxaliplatin, tetraplatin, cisplatin, and carboplatin:
spectrum of activity in drug-resistant cell lines and in the cell lines of the
National Cancer Institute's Anticancer Drug Screen panel, Biochem.
200
Pharmacol. 52, 1855-1865.
158.
Olszewski, U., and Hamilton, G. (2010) A better platinum-based
anticancer drug yet to come?, Anticancer Agents Med. Chem. 10,
293-301.
159.
Galanski, M. (2006) Recent developments in the field of anticancer
platinum complexes, Recent Pat. Anticancer Drug Discov. 1, 285-295.
160.
Kostova, I. (2006) Platinum complexes as anticancer agents, Recent
Pat. Anticancer Drug Discov. 1, 1-22.
161.
Woźniak, K., and Błasiak, J. (2002) Recognition and repair of
DNA-cisplatin adducts., Acta Biochim. Pol. 49, 583-596.
162.
Kelland, L. R., Sharp, S. Y., O'Neill, C. F., Raynaud, F. I., Beale, P. J.,
and Judson, I. R. (1999) Mini-review: discovery and development of
platinum complexes designed to circumvent cisplatin resistance, J.
Inorg. Biochem. 77, 111-115.
163.
Surridge, C. (1998) Oxford dictionary of biochemistry and molecular
biology, Nature 392, 351-352.
164.
Hambley, T. W., and Webster, L. K. (1994) The preparation and in vivo
activity
of
L-O-methyl-methionine-dichloroplatinum(II)
and
D-O-methyl-methionine-dichloroplatinum(II) - the crystal-structure of the
L-enantiomer, J Inorg. Biochem. 55, 175-181.
165.
Li, C., Li, Z. Y., Sletten, E., Arnesano, F., Losacco, M., Natile, G., and
Liu,
Y.
Z.
(2009) Methionine
can favor DNA platination by
201
trans-coordinated platinum antitumor drugs, Angew. Chem. Int. Edit. 48,
8497-8500.
166.
Altman, J., Wilchek, M., and Warshawsky, A. (1985) Platinum(II)
complexes
with
2,4-diaminobutyric
acid,
ornithine,
lysine
and
4,5-diaminovaleric acid, Inorg. Chim. A.-Bioinor. 107, 165-168.
167.
Sandman, K. E., Fuhrmann, P., and Lippard, S. J. (1998) A
mechanism-based, solution-phase method for screening combinatorial
mixtures of potential platinum anticancer drugs, J. Biol. Inorg. Chem. 3,
74-80.
168.
Dalla Via, L., Gia, O., Magno, S. M., Dolmella, A., Marton, D., and Di
Noto, V. (2006) Synthesis, characterization and biological activity of
platinum(II) complexes with L- and D-ornithine ligands, Inorg. Chim.
Acta 359, 4197-4206.
169.
Ye, Q. S., Xie, M. J., Liu, W. P., Chen, X. Z., Yu, Y., Chang, Q. W., and
Hou, S. Q. (2009) Synthesis, characterization and cytotoxicity of
dihalogeno-platinum(II) complexes with L-histidine ligand, Chem.
Pharm. Bull. 57, 424-427.
170.
Baidina, I. A., Slyudkin, O. P., and Borisov, S. V. (1985) Crystal and
molecular-structures of platinum(Ii) diiodo-L-histidinate, J. Struct. Chem.
26, 955-958.
171.
Bino, A., Cohen, S., Altman, J., and Wilchek, M. (1988) The crystal and
molecular-structures
of
dichloro[N,O-(D,L)-diaminopropionic
202
acid]platinum(II) and dichloro[N,O-(L)-lysine]platinum(II) monohydrate,
Inorg. Chim. Acta 147, 99-102.
172.
Sandman, K. E., Fuhrmann, P., and Lippard, S. J. (1998) A
mechanism-based, solution-phase method for screening combinatorial
mixtures of potential platinum anticancer drugs, J. Biol. Inorg. Chem. 3,
74-80.
173.
Ziegler, C. J., Silverman, A. P., and Lippard, S. J. (2000)
High-throughput synthesis and screening of platinum drug candidates, J.
Biol. Inorg. Chem. 5, 774-783.
174.
McA'Nulty, M. M., Whitehead, J. P., and Lippard, S. J. (1996) Binding of
Ixr1, a yeast HMG-domain protein, to cisplatin-DNA adducts in vitro and
in vivo, Biochemistry 35, 6089-6099.
175.
McA'Nulty, M. M., and Lippard, S. J. (1996) The HMG-domain protein
Ixr1 blocks excision repair of cisplatin-DNA adducts in yeast, Mutat. Res.
362, 75-86.
176.
Chow, C. S., Whitehead, J. P., and Lippard, S. J. (1994) HMG domain
proteins induce sharp bends in cisplatin-modified DNA, Biochemistry 33,
15124-15130.
177.
Huang, J. C., Zamble, D. B., Reardon, J. T., Lippard, S. J., and Sancar,
A. (1994) HMG-domain proteins specifically inhibit the repair of the
major DNA adduct of the anticancer drug cisplatin by human excision
nuclease, Proc. Natl. Acad. Sci. U. S. A. 91, 10394-10398.
203
178.
Ziegler, C. J., Sandman, K. E., Liang, C. H., and Lippard, S. J. (1999)
Toxicity of platinum(II) amino acid (N,O) complexes parallels their
binding to DNA as measured in a new solid phase assay involving a
fluorescent HMG1 protein construct readout, J. Biol. Inorg. Chem. 4,
402-411.
179.
Banner, R. J., Charlson, A. J., Gale, R. P., Mcardle, N. T., Trainor, K. E.,
and Watton, E. C. (1977) Antitumor activity of some amino-acid metal
systems, Cancer Treat Rep. 61, 469-470.
180.
Cohen, S. M., and Lippard, S. J. (2001) Cisplatin: from DNA damage to
cancer chemotherapy, Prog. Nucleic Acid Res. Mol. Biol. 67, 93-130.
181.
Rijal, K. (2011) Exploring potential drug target sites in the ribosome
using cisplatin and its analogue, p 259 p.
182.
Hedman, H. K., Kirpekar, F., and Elmroth, S. K. (2011) Platinum
interference with siRNA non-seed regions fine-tunes silencing capacity,
J. Am. Chem. Soc. 133, 11977-11984.
183.
Cavaluzzi, M. J., and Borer, P. N. (2004) Revised UV extinction
coefficients for nucleoside-5'-monophosphates and unpaired DNA and
RNA, Nucleic Acids Res. 32, e13.
184.
Reedijk, J. (2003) New clues for platinum antitumor chemistry:
kinetically controlled metal binding to DNA, Proc. Natl. Acad. Sci. U. S.
A. 100, 3611-3616.
185.
Baik, M. H., Friesner, R. A., and Lippard, S. J. (2003) Theoretical study
204
of cisplatin binding to purine bases: why does cisplatin prefer guanine
over adenine?, J. Am. Chem. Soc. 125, 14082-14092.
186.
Becke, A. D. (1993) Density-functional thermochemistry .3. The role of
exact exchange, J. Chem. Phys. 98, 5648-5652.
187.
Lee, C. T., Yang, W. T., and Parr, R. G. (1988) Development of the
colle-salvetti correlation-energy formula into a functional of the
electron-density, Phys. Rev. B 37, 785-789.
188.
Hay, P. J., and Wadt, W. R. (1985) Ab initio effective core potentials for
molecular calculations - potentials for the transition-metal atoms Sc to
Hg, J. Chem. Phys. 82, 270-283.
189.
Hay, P. J., and Wadt, W. R. (1985) Ab initio effective core potentials for
molecular calculations - potentials for K to Au including the outermost
core orbitals, J. Chem. Phys. 82, 299-310.
190.
Wadt, W. R., and Hay, P. J. (1985) Ab initio effective core potentials for
molecular calculations - potentials for main group elements Na to Bi, J.
Chem. Phys. 82, 284-298.
191.
Dunning, T. H. (1989) Gaussian-basis sets for use in correlated
molecular calculations .1. The atoms boron through neon and hydrogen,
J. Chem. Phys. 90, 1007-1023.
192.
Polonyi, C., and Elmroth, S. K. (2013) Time dependence of
cisplatin-induced duplex dissociation of 15-mer RNAs and mature
miR-146a, Dalton Trans. 42, 14959-14962.
205
193. White, J. D., Osborn, M. F., Moghaddam, A. D., Guzman, L. E., Haley, M.
M., and DeRose, V. J. (2013) Picazoplatin, an azide-containing
platinum(II) derivative for target analysis by click chemistry, J. Am.
Chem. Soc. 135, 11680-11683.
194.
Scovell, W. M., and O'Connor, T. (1977) Interaction of aquated
cis-[(NH3)2Pt(II)] with nucleic acid constituents. 1. Ribonucleosides, J.
Am. Chem. Soc. 99, 120-126.
195.
Eastman, A. (1982) Separation and characterization of products
resulting from the reaction of cis-diamminedichloroplatinum (II) with
deoxyribonucleosides, Biochemistry 21, 6732-6736.
196.
Martin, R. B. (1996) Dichotomy of metal ion binding to N1 and N7 of
purines, Met. Ions Biol. Syst. 32, 61-89.
197.
Kampf, G., Lüth, M. S., Kapinos, L. E., Müller, J., Holý, A., Lippert, B.,
and Sigel, H. (2001) Formation of ternary complexes by coordination of
(diethylenetriamine)-platinum(II) to N1 or N7 of the adenine moiety of
the antiviral nucleotide analogue 9, Chemistry 7, 1899-1908.
198.
Drabløs, F., Feyzi, E., Aas, P. A., Vaagbø, C. B., Kavli, B., Bratlie, M. S.,
Peña-Diaz, J., Otterlei, M., Slupphaug, G., and Krokan, H. E. (2004)
Alkylation damage in DNA and RNA--repair mechanisms and medical
significance, DNA Repair (Amst) 3, 1389-1407.
199.
Barry, C. G., Day, C. S., and Bierbach, U. (2005) Duplex-promoted
platination of adenine-N3 in the minor groove of DNA: challenging a
206
longstanding bioinorganic paradigm, J. Am. Chem. Soc. 127,
1160-1169.
200.
Guddneppanavar, R., and Bierbach, U. (2007) Adenine-N3 in the DNA
minor groove - an emerging target for platinum containing anticancer
pharmacophores, Anticancer Agents Med. Chem. 7, 125-138.
201.
Rao, L., and Bierbach, U. (2007) Kinetically favored platination of
adenine in the G-rich human telomeric repeat, J. Am. Chem. Soc. 129,
15764-15765.
202.
Coe, B. J., and Glenwright, S. J. (2000) Trans-effects in octahedral
transition metal complexes, Coordin. Chem. Rev. 203, 5-80.
203.
Chval, Z., Sip, M., and Burda, J. V. (2008) The trans effect in
square-planar platinum(II) complexes--a density functional study, J.
Comput. Chem. 29, 2370-2381.
204.
Pinter, B., Van Speybroeck, V., Waroquier, M., Geerlings, P., and De
Proft, F. (2013) Trans effect and trans influence: importance of metal
mediated ligand-ligand repulsion, Phys. Chem. Chem. Phys. 15,
17354-17365.
205.
Glish, G. L., and Vachet, R. W. (2003) The basics of mass spectrometry
in the twenty-first century, Nat. Rev. Drug Discov. 2, 140-150.
206.
Wells, J. M., and McLuckey, S. A. (2005) Collision-induced dissociation
(CID) of peptides and proteins, Methods Enzymol. 402, 148-185.
207.
Marshall, A. G., Hendrickson, C. L., and Jackson, G. S. (1998) Fourier
207
transform ion cyclotron resonance mass spectrometry: a primer, Mass
Spectrom. Rev. 17, 1-35.
208.
Coon, J. J. (2009) Collisions or electrons? Protein sequence analysis in
the 21st century, Anal. Chem. 81, 3208-3215.
209.
Tamm, C., Shapiro, H. S., Lipshitz, R., and Chargaff, E. (1953)
Distribution density of nucleotides within a desoxyribonucleic acid chain,
J. Biol. Chem. 203, 673-688.
210.
Taylor, B. E., and Irvin, J. D. (1990) Depurination of plant ribosomes by
pokeweed antiviral protein, FEBS Lett. 273, 144-146.
211.
de Virgilio, M., Lombardi, A., Caliandro, R., and Fabbrini, M. S. (2010)
Ribosome-inactivating proteins: from plant defense to tumor attack,
Toxins (Basel) 2, 2699-2737.
212.
Baik, M. H., Friesner, R. A., and Lippard, S. J. (2002) Theoretical study
on the stability of N-glycosyl bonds: why does N7-platination not
promote depurination?, J. Am. Chem. Soc. 124, 4495-4503.
213.
Kenttämaa, H. I., and Cooks, R. G. (1985) Internal energy-distributions
acquired through collisional activation at low and high-energies, Int. J.
Mass Spectrom. 64, 79-83.
214.
Kertesz, T. M., Hall, L. H., Hill, D. W., and Grant, D. F. (2009) CE50:
Quantifying collision induced dissociation energy for small molecule
characterization and identification, J. Am. Soc. Mass Spectr. 20,
1759-1767.
208
215. Valle, J. J., Eyler, J. R., Oomens, J., Moore, D. T., van der Meer, A. F. G.,
von Helden, G., Meijer, G., Hendrickson, C. L., Marshall, A. G., and
Blakney, G. T. (2005) Free electron laser-Fourier transform ion cyclotron
resonance mass spectrometry facility for obtaining infrared multiphoton
dissociation spectra of gaseous ions, Rev. Sci. Instrum. 76.
216.
Polfer, N. C., and Oomens, J. (2007) Reaction products in mass
spectrometry elucidated with infrared spectroscopy, Phys. Chem. Chem.
Phys. 9, 3804-3817.
217.
Eyler, J. R. (2009) Infrared multiple photon dissociation spectroscopy of
ions in penning traps, Mass Spectrom. Rev. 28, 448-467.
218.
Shin, J. W., Lee, Y. H., Hwang, S., and Lee, S. W. (2007) Observation of
an unusually facile fragmentation pathway of gas-phase peptide ions: a
study on the gas-phase fragmentation mechanism and energetics of
tryptic peptides modified with 4-sulfophenyl isothiocyanate (SPITC) and
4-chlorosulfophenyl isocyanate (SPC) and their 18-crown-6 complexes,
J. Mass. Spectrom. 42, 380-388.
219.
Paizs, B., and Suhai, S. (2005) Fragmentation pathways of protonated
peptides, Mass Spectrom. Rev. 24, 508-548.
220.
Polfer, N. C., and Oomens, J. (2007) Reaction products in mass
spectrometry elucidated with infrared spectroscopy, Phys. Chem. Chem.
Phys. 9, 3804-3817.
221.
Deeth, R. J., and Elding, L. I. (1996) Theoretical Modeling of Water
209
Exchange on [Pd(H2O)4]2+, [Pt(H2O)4]2+, and trans-[PtCl2(H2O)2], Inorg.
Chem. 35, 5019-5026.
222.
Chval, Z., and Sip, M. (2003) Transition states of cisplatin binding to
guanine and adenine: ab initio reactivity study, Collect. Czech. Chem. C.
68, 1105-1118.
223.
Banerjee, S., and Mukherjee, A. K. (2013) Interactions of the aquated
forms of the anticancer drug AMD443 with DNA purine bases: A detailed
computational approach, Inorg. Chim. Acta 400, 130-141.
224.
Chiavarino, B., Crestoni, M. E., Fornarini, S., Scuderi, D., and Salpin, J.
Y. (2013) Interaction of cisplatin with adenine and guanine: a combined
IRMPD, MS/MS, and theoretical study, J. Am. Chem. Soc. 135,
1445-1455.
225.
Barnes, K. R., and Lippard, S. J. (2004) Cisplatin and related anticancer
drugs: recent advances and insights, Met. Ions Biol. Syst. 42, 143-177.
226.
Alonso, A., Almendral, M. J., Curto, Y., Criado, J. J., Rodríguez, E., and
Manzano, J. L. (2007) New fluorescent antitumour cisplatin analogue
complexes. Study of the characteristics of their binding to DNA by flow
injection analysis, J. Fluoresc. 17, 390-400.
227.
Molodkin, A., Esina, N., and Andreeva, O. (2008) Mixed-ligand
platinum(IV) complexes with amino acids and adenine, Russ. J. Inorg.
Chem. 53, 1741-1746.
228.
Wilson, J. J., and Lippard, S. J. (2013) Synthetic methods for the
210
preparation of platinum anticancer complexes, Chem. Rev.
229.
Pavankumar, P. N. V., Seetharamulu, P., Yao, S., Saxe, J. D., Reddy, D.
G., and Hausheer, F. H. (1999) Comprehensive ab initio quantum
mechanical and molecular orbital (MO) analysis of cisplatin: Structure,
bonding, charge density, and vibrational frequencies, J. Comput. Chem.
20, 365-382.
230. Amado, A. M., Fiuza, S. M., Marques, M. P. M., and de Carvalho, L. A. E.
B. (2007) Conformational and vibrational study of platinum(II)
anticancer drugs: cis-diamminedichloroplatinum(II) as a case study, J.
Chem. Phys. 127.
231.
de Berrêdo, R. C., and Jorge, F. E. (2010) All-electron double zeta basis
sets for platinum: Estimating scalar relativistic effects on platinum(II)
anticancer drugs, J. Mol. Struc.-Theochem. 961, 107-112.
232.
Paschoal, D., Marcial, B. L., Lopes, J. F., De Almeida, W. B., and Dos
Santos, H. F. (2012) The role of the basis set and the level of quantum
mechanical theory in the prediction of the structure and reactivity of
cisplatin, J. Comput. Chem. 33, 2292-2302.
233.
Parr, R. G., and Yang, W. T. (1989) Density-functional theory of atoms
and molecules, Oxford University Press; Clarendon Press, New York
Oxford England.
234.
Ziegler, T. (1991) Approximate density functional theory as a practical
tool in molecular energetics and dynamics, Chem. Rev. 91, 651-667.
211
235.
Trappl, K., and Polacek, N. (2011) The ribosome: a molecular machine
powered by RNA, Met. Ions Life Sci. 9, 253-275.
236.
Rijal, K. (2011) Exploring potential drug target sites in the ribosome
using cisplatin and its analogues., In Department of Chemistry, Wayne
State University.
237.
Schrader, W., and Klein, H. W. (2004) Liquid chromatography/Fourier
transform ion cyclotron resonance mass spectrometry (LC-FTICR MS):
an early overview, Anal. Bioanal. Chem. 379, 1013-1024.
238.
Hendrickson, C. L., and Emmett, M. R. (1999) Electrospray ionization
Fourier transform ion cyclotron resonance mass spectrometry, Annu.
Rev. Phys. Chem. 50, 517-536.
239.
Payne, A. H., and Glish, G. L. (2005) Tandem mass spectrometry in
quadrupole ion trap and ion cyclotron resonance mass spectrometers,
Methods Enzymol. 402, 109-148.
240.
McLuckey, S. A., Van Berkel, G. J., Goeringer, D. E., and Glish, G. L.
(1994) Ion trap mass spectrometry. Using high-pressure ionization, Anal.
Chem. 66, 737A-743A.
241.
Bergquist, J., Palmblad, M., Wetterhall, M., Håkansson, P., and
Markides, K. E. (2002) Peptide mapping of proteins in human body
fluids using electrospray ionization Fourier transform ion cyclotron
resonance mass spectrometry, Mass Spectrom. Rev. 21, 2-15.
212
ABSTRACT
SYNTHESIS, CHARACTERIZATION, AND RNA REACTIVITY OF
AMINO-ACID-LINKED CISPLATIN ANALOGUES
by
XUN BAO
May 2015
Advisor : Prof. Christine S. Chow
Major : Chemistry
Degree : Doctor of Philosophy
The essential role of ribosomal RNA (rRNA) in the cell life cycle is
highlighted by protein synthesis; therefore, the ribosome is considered to be an
ideal drug target. Ribosomal RNAs exhibit a high level of structural diversity.
The well-known anticancer drug cisplatin was previously applied as a chemical
probe of rRNA structure to determine solvent accessible purines (guanosine)
in vivo and in vitro. Cisplatin accumulates faster on RNA than DNA, with less
chance of repair. As such, designing new RNA-targeting Pt compounds is not
only a promising direction for chemical-probing applications, but also for the
design of anticancer drugs that could overcome DNA repair-related resistance.
In the present study, amino-acid-linked cisplatin analogues were synthesized.
Their reactivity and product profiles with RNA were evaluated on the
nucleoside level by using high performance liquid chromatography (HPLC),
213
nuclear magnetic resonance (NMR) spectroscopy, and mass spectrometry
(MS). Tandem mass spectrometry (MS/MS) combined with computational
chemistry was also utilized to assess the effects of Pt(II) on the
glycosidic-bond strength of the purine nucleoside, and to obtain an
atomic-level understanding of the nucleoside-Pt(II) products. Results reveal a
cisplatin analogue with the first-reported template-independent adenosine
preference, and demonstrate that Pt(II) reactivity can be tuned by altering the
carrier ligand(s). This project also illustrates the potential of MS/MS to be used
in RNA-Pt studies and provides details of adduct structures, such as the sites
of adduct formation.
214
AUTOBIOGRAPHICAL STATEMENT
Xun Bao
ADVISOR: Prof. Christine S. Chow
THESIS TITLE: Synthesis, Characterization, and RNA Reactivity of
Amino-Acid-Linked Cisplatin Analogues
EDUCATION
• Ph.D; Biological Chemistry, 2015, Wayne State University, MI, USA
•
B. S; Chemistry, 2008, University of Science and Technology of China,
Anhui, China
PUBLICATIONS:
K. Rijal, X. Bao, C. S. Chow Amino-acid-linked platinum(II) analogues have
altered specificity for RNA compared to cisplatin, Chem. Commun., 2014, 50,
3918-3920, cover article
X. Bao, J. Jiang, Z. Yang, M. T. Rodgers, C. S. Chow Spectroscopic studies of
reactivity between nucleosides and cisplatin analogues, manuscript in
preparation
H. Hedman, X. Bao, K. Rijal, B. J. Shay, S. K. Elmroth, C. S. Chow Adduct
formation in ribosomal RNA by an ornithine-platinum(II) complex and parent
cisplatin, manuscript in preparation.
X. Bao, C. He, M. T. Rodgers, C. S. Chow Interaction of ornithine-linked Pt
compound with adenosine and guanosine: a combined IRMPD,CID MS/MS
and theoretical study, manuscript in preparation