Modified Nucleosides Part A: a Platform for the Chemical Tagging of Ribonucleic Acids for Analysis by Mass Spectrometry Part B: Base-Modified Thymidines Exhibiting Cytotoxicity

Home , Deoxyribonucleoside

Modified Nucleosides Part A: A Platform for the Chemical Tagging of Ribonucleic

Acids for Analysis by Mass Spectrometry Part B: Base-Modified Thymidines

Exhibiting Cytotoxicity towards Cancer Cells

A Dissertation

Submitted to the Graduate School

of the University of Cincinnati

in Partial Fulfillment of the

Requirements for the Degree

Doctor of Philosophy (PhD)

In the Department of Chemistry of McMicken College of Arts and Sciences

By:

Kayla Borland

Aug 2018

University of Cincinnati

Cincinnati, Ohio

Abstract

This dissertation is focused on modified nucleosides. Part A focuses on method development for multiplex analysis of modified oligonucleosides while part B is a medicinal chemistry perspective of modified nucleosides as potential anti-cancer therapeutics. Part A uses mass spectrometry (MS) as an enabling technology for the characterization of post- transcriptionally modified nucleosides within ribonucleic acids (RNAs). These modified RNAs tend to be more challenging to completely characterize using conventional genomic-based sequencing technologies. As with many biological molecules, information relating to the presence or absence of a particular compound (i.e., qualitative measurement) is only one step in sample characterization. Additional useful information is found by performing quantitative measurements on the levels of the compound of interest in the sample. To unlock this information within RNA samples, previously reported duplex-based strategies to characterize modified RNAs in two different samples have been examined. Here is reported the use of poly adenosine polymerase (PAP), which – under optimized conditions – can add one 2’ azido modified nucleotide to the 3’-terminus of modified RNA. The addition of this azido-modified nucleotide can allow for the use of click chemistry to uniquely tag each sample. One sample is labeled with an unisotopically labeled alkyne while the other samples are reacted with an isotopically labeled alkyne. The two samples can easily be compared to one another based on doublet separate by difference in isotopic label mass. In part B modified nucleoside are synthesized because, current FDA-approved anti-cancer modified nucleosides elicit severe side effects warranting their improvement. Therefore, a series of compounds with a mechanism of action focused on inhibiting DNA replication was designed. Compound were inspired by the previous discovery that 5-(α-substituted-2-nitrobenzyloxy)methyluridine-5’-triphosphates terminate DNA synthesis. Thus, a library of thymidine analogs were synthesized and evaluated using a cell viability assay in MCF7 breast cancer cells, which were chosen because they had the greatest susceptibility to these nucleosides. The structure-activity relationship study lead to a compounds having α-tert-butyl-2-nitro-4-(phenyl)alkynylbenzyloxy; it caused 50% of MCF7 cell death at 9 ± 1 µM concentration.

iii Acknowledgments

This dissertation is dedicated to my family and my friends that have become like family for their continued support, prayers, and encouragement.

I would like to acknowledge Dr. Limbach and Dr Merino for all their guidance in my journey to become a better scientist.

iv Table of Contents

Abstract ………………………………………………………………………………………....ii

Acknowledgments …………………………………………………………………….….……iv

List of Figures …………………………………………………………………………..……...ix

List of Tables …………………………………………………………………………………xiv

Part A

1 Chapter 1 Introduction ...... 17

1.1 Research goal ...... 17

1.2 Introduction to Mass Spectrometry of Modified Ribonucleic Acids ...... 17

1.3 Stable Isotope Labeling Methods for RNA Mass Spectrometry ...... 20

1.4 18O labeling ...... 23

1.4.1 18O Labeling of RNA – Early Applications for Mass Spectrometry ...... 25

1.4.2 18O Labeling of RNA for Modification Mapping by Mass Spectrometry ...... 30

1.5 Future Outlook ...... 36

2 Chapter 2 Multiplexing ...... 38

2.1 Background ...... 38

v 2.2 Experimental ...... 39

2.2.1 Materials: ...... 39

2.2.2 Identification of Optimal azido-modified dNTP...... 39

2.2.3 Optimization of PAP reactions ...... 39

2.2.4 Click reaction optimization ...... 40

2.2.5 Calibration curve ...... 40

2.2.6 Model duplexes ...... 41

2.2.7 LC-MS ...... 41

2.3 Results and Discussion ...... 42

2.3.1 Identification of optimal azido-modified dNTP ...... 42

2.3.2 Optimization of PAP extension ...... 43

2.3.3 Click reaction optimization ...... 45

2.3.4 Model Duplexing Studies ...... 47

2.3.5 Relative Quantification of Oligonucleotide Modification Levels ...... 49

2.4 Conclusion ...... 50

3 Chapter 3 Conclusions and Future Work ...... 52

3.1 Conclusions ...... 52

3.2 Future work ...... 52

3.3 Additional alkynes tags ...... 53

3.4 T1 removal or inactivation ...... 58

3.5 PAP extension of oligonucleotide ...... 64

vi 3.6 T4 ligation ...... 66

4 Chapter 4 Introduction ...... 70

4.1 Research goal ...... 70

4.2 Introduction ...... 70

4.3 Current nucleobase-modified antimetabolite chemotherapeutic drugs ...... 72

4.3.1 Modified nucleobase drugs ...... 72

4.3.2 Base-modified natural sugar nucleoside drugs ...... 73

4.3.3 Base-modified unnatural sugar nucleoside drugs ...... 75

4.3.4 Development of novel nucleobase-modified antimetabolite drug candidates ...... 78

4.4 Conclusions and outlook ...... 87

5 Chapter 5 Base-Modified Thymidines Capable of Terminating DNA Synthesis as Novel

Drug Candidates Showing Activity in Cancer Cells...... 88

5.1 Background ...... 88

5.2 Chemical synthesis ...... 89

5.2.1 Materials ...... 89

5.2.2 Chemical Synthesis ...... 90

5.2.3 Synthesized Compounds ...... 90

5.3 Biochemistry experiments ...... 110

5.3.1 Cell cytotoxicity assay (MTT) ...... 110

5.3.2 DNA synthesis termination studies ...... 110

5.4 Results/discussion ...... 113

vii 5.4.1 Synthesis ...... 113

5.4.2 Cytotoxicity in MCF7 breast cancer cells: elucidation of structure-activity

relationship ...... 113

5.4.3 Evaluation of selectivity for novel bioactive compound 3a...... 114

5.4.4 Evaluation of activity for novel bioactive compound 3a in other cancer cells ...115

5.4.5 DNA synthesis termination by 5’-triphosphate of 3a ...... 117

5.5 Conclusion ...... 118

6 References ...... 119

7 Appendix A Supporting and Supplemental Nucleoside Information ...... 139

viii List of Figures

Figure 1.1 Endonuclease method for RNase T1 and other RNases. The RNA oligonucleotide

is cleaved and a 3’ cyclic phosphate intermediate is formed. An 18O atom from the

reaction solvent can break the cyclic phosphate and allow for the addition of the stable

isotope label to the final 3’ linear phosphate product. Protease method for Trypsin.

The protein is cleaved at the C-terminus after arginine or lysine residues. The 18O

form the labeled reaction solvent is incorporated into the newly formed carboxylic

acid after the cleavage site ...... 24

Figure 1.2 Expanded view of MALDI mass spectral data obtained from the RNase T1

digestion products of E. coli 5S rRNA. (a) Digestion was done in unlabeled water.

Three major ions are detected (m/z 980, 987 and 998). (b) After digestion in a 50:50

(v/v) mixture of unlabeled and 18O-labeled water, only those oligonucleotide

digestion products that contain a 3’-phosphate will exhibit the characteristic A + 2

doublet. From this, base compositions for the three ions can be made or confirmed.

Reproduced with permission from Berhane, B.T. and P.A. Limbach, Stable isotope

labeling for matrix-assisted laser desorption/ionization mass spectrometry and post-

source decay analysis of ribonucleic acids. J. Mass Spectrom., 2003. 38: p. 872-878,

Figure 1.3 Representative MALDI mass spectrum of RNase T1 digestion products obtained

from a tRNA-Val mixture prepared at a heavy-to-light RNA ratio of 2:1. The asterisk

(*) denotes expected RNase digestion product pairs. Inset: expanded view of the

RNase T1 digestion product 5’-CUCAGp-3’ used for quantifying RNA levels for

tRNA-Val with overlaid calculated isotopic distribution assuming a 2:1 ratio.

ix Reproduced with permission from Meng, Z. and P.A. Limbach, Quantitation of

Ribonucleic Acids Using 18-O Labeling and Mass Spectrometry. Anal. Chem., 2005.

Figure 2.1 To determine which 2’-N₃dNTPs is “ideal” for PAP extension, template S1

(UAACUAUAACGG) was reacted with two different 2’-N₃dNTPs at a 1:10 ratio and

PAP for 1 h at 30 and 37 °C. The reaction was heat inactivated, and reaction products

were separated on a 20% polyacrylamide gel. Oligonucleotide bands were visualized

using Sybr Gold. This gel shows a mixture of extended and unextended template for

the 2’-N₃dGTPs sample, but the 2’-N₃dUTPs shows much more extended than

unextended template. When comparing the 30 and 37 °C for 2’-N₃dUTP, more

extended product is observed in the 30 °C incubated sample ...... 44

Figure 2.2 CAU[Gm]UGG is shown as a representative example from an oligonucleotide

template that was reacted with 2’N3dUTP and PAP. Reaction aliquots were then

clicked with either phenylacetylene in an unlabeled (D0) or deuterium (D5) labeled

form. Samples were run individually or combined in a 1:1 ratio. Both labeling

reactions were successful. Samples labeled with D0 and D5 in the 2- charge state show

a mass difference of 2.5 Daltons. In the 3- charge state, they show a difference of 1.66

Daltons. Even though the samples were mixed 1:1 the mass spectra do not reveal equal

ion abundances due to the decreased ion efficiency of the D5 labeled product...... 48

Figure 2.3 CAU[Gm]UGG is shown as a representative example from a mixture of modified

oligonucleotide and unmodified oligonucleotide pair templates that were reacted with

2’-N3dUTP and PAP. Reactions were then clicked with either phenylacetylene in an

unlabeled (D0) or deuterium (D5) labeled form. Samples were combined in a 1:1 ratio.

x A) 100% modified oligonucleotide sample labeled with D0 mixed with the D5 labeled

0% modified oligonucleotide sample. As would be expected, the peak corresponding

to D0 is much higher than the peak for D5. B) Two 50% modified samples each

bearing a different label. The peak for the D0 sample was higher, but not as drastically

as panel A. This is likely related to the lower ionization efficiency of the D5 label. C)

A 30% modified oligonucleotide sample labeled with D0 and a 70% modified

oligonucleotide sample labeled with D5. As expected the D5 peak was larger than the

D0 peak. The ionization prevents the D5 peak from being higher, but it is still above

the D0 as expected. All of the model duplexing samples visually represent the

increasing and decreasing modification status within the two samples...... 51

Figure 3.1 Extracted ion chromatogram of expected product and product +14 Da. For the

heptyne alkyne reaction for 4 h at 37 C with 50 equivalence of alkyne, the

predominant product based on peak area is the +14 Da product. The top panel shows

the expected product while the bottom panel shows the + 14 Da product. In this

reaction based on peak are the +14 Da in the major product...... 55

Figure 3.2 MS/MS of PropA click product with unexpected CID fragments ...... 55

Figure 3.3 Possibility for new isotopically labeled tags ...... 58

Figure 3.4 Post digestion PAP reaction product ...... 59

Figure 3.5 Indicator fragment GUUCAU incubation post phenol chloroform extraction

followed by ether wash...... 63

Figure 3.6 Extracted ion chromatogram for extension and click product post phenol

chloroform with ether extraction. Top spectrum shows the extension product for the

xi oligonucleotide AAUUCG while the bottom shows the product of the click reaction

with phenylacetylne D0 for the same oligonucleotide ...... 63

Figure 3.7 Ligation reaction plan ...... 67

Figure 4.1 Unnatural nucleobase drugs: 5-fluorouracil (5FU), 6-mercaptopurine (6MP), and

6-thioguanine (6TG) ...... 73

Figure 4.2 Natural sugar base-modified nucleoside drugs: 2-chloro-2'-deoxyadenosine

(2CDA), 5-azacytidine (5-AZC) and 5-aza-2'-deoxycytidine (5-AZCdR) ...... 75

Figure 4.3 Sugar and base modified nucleside drugs: Emtricitabine (FTC), Abacavir (ABC),

Sorivudine (BV-ara-U), Clofarabine (CAFdA), Fludarabine (2FaraA), and Nelarabine

(araG) ...... 76

Figure 4.4 Susceptibility to failure in polymerase promoted DNA incorporation of nucleoside

5'-triphosphates depending upon substitution: blue - large groups are tolerated; purple

- small groups are tolerated; red - no substitution is tolerated ...... 80

Figure 4.5 Chemotherapeutic base-modified purine nucleoside analogs ...... 82

Figure 4.6 Primer extension assay of base modified triphosphates ...... 82

Figure 4.7 Chemotherapeutic base-modified pyrimidine nucleosides ...... 83

Figure 4.8 Chemotherapeutic nucleosides with modified nucleobases having altered hydrogen

bonding pattern ...... 85

Figure 4.9 1'-Non-nucleobase-2'-deoxyribonucleoside analogs: 5-nitroindolyl- (5NI), 3-

ethynyl-5-nitroindolyl- (3Eth5NI), indolyl- (Ind), naphthyl- (Naphth), and 3,4-

difluorophenyl- (34diFPh) 2'-deoxynucleosides ...... 87

xii Figure 5.1 Synthesis of T-nucleoside analogs: Appropriate alcohol, 110-124 C, 1-3 h; (ii) n-

Bu4NF, THF, 0 C to r.t., 2-6 h; (iii) Pd(PPh3)4, appropriate terminal alkyne, CuI,

Et3N, DMF, r.t., 6-18 h; (iv) Benzyl azide, CuI, Et3N, MeCN, r.t., 4 h...... 112

Figure 5.2 Selectivity of 3a. Viability of MCF7 cancer cells (black) and fibroblasts (grey) in

the presence of 3a. Selectivity is greater than 5-fold ...... 115

Figure 5.3 Primer extension assay studies of 3a triphosphate. Left to right: no polymerase,

3aTP at 0, 0.125, 0.25, 0.5, 1, 2 mM terminating nucleotide, acyclo A, C, G (2 mM

each), acylco T at 2, 1, 0.5, 0.125 mM ...... 117

xiii

List of Tables

Table 1.1 Analysis of RNase T1 quantitative signature digestion products from E. coli. Table

reproduced with permission from Castleberry, C.M. and P.A. Limbach, Relative

Quantitation of Transfer RNAs Using Liquid Chromatography-Mass Spectrometry

(LC-MS) and Signature Digestion Products. Nucleic Acids Res (2010) 38 e162...... 31

Table 2.1 Optimization of PAP extension conditions for a set of unmodified oligonucleotide

templates. Extension levels for each template were quantified based on peak area

ratios for extended versus un-extended template ...... 44

Table 2.2 Effect of poly(A) polymerase to template ratio on oligonucleotide extension.

Extension reactions were performed for 240 min at 28 °C. Extension levels for each

template were quantified based on peak area ratios for extended versus un-extended

template ...... 45

Table 2.3 Effect of reaction period on phenylacetylene click efficiency. Extension levels for

each template were quantified based on peak area ratios for extended versus un-

extended template...... 46

Table 2.4 Effect of phenylacetylene equivalence on click reaction efficiency. Extension levels

for each template were quantified based on peak area ratios for oligonucleotides that

were labeled with the click reagent versus extended but not labeled...... 47

xiv Table 2.5 A calibration curve with the template [m3U][Am]ACAAGG was created. Samples

with appropriate alkynes were mixed prior to analysis. Measurements based on peak

area were completed in technical triplicate ...... 49

Table 3.1 Alkyne reaction variations ...... 56

Table 3.2 Summary of PAP reaction products ...... 59

Table 3.3 Summary of purification results ...... 61

Table 3.4 3’ Extension with 2’-N3dUTP ...... 65

Table 3.5 T4 PNK reaction results ...... 68

Table 3.6 Ligation reaction results. Percent ligation was calculated based on peak area with

ligated/ (ligated+unligated) *100% ...... 68

Table 5.1 IC50 values determined by MTT assays using base-modified T-nucleoside analogs

and MCF7 breast cancer cells ...... 111

Table 5.2 Percent growth inhibition of 3a at 10 μM in varied cell lines ...... 116

xv List of Appendices

Appendix A Supporting and Supplemental Nucleoside Information

……………………….Error! Bookmark not defined.

xvi 1 CHAPTER 1 INTRODUCTION

1.1 Research goal

The goal of part one of this dissertation is to advance the field of nucleic acids research, specifically within the field of nucleic acids analyzed by Mass Spectrometry (MS). In this chapter, previous MS techniques for nucleic acids are introduced. The biological relevance of studying ribonucleic acids (RNA) is explained, and a special focus is given to isotopic labeling strategies and duplexing strategies useful for MS analysis of nucleic acids. In Chapter 2, a new method to multiplex oligonucleotide samples for mass spectrometry analysis is described. This work is done to further mass spectrometry method development. It is completed in part to improve the ability to obtain relative quantification of numerous samples in the same analysis.

This work bridges the fields of synthetic chemistry, biochemistry, and analytical chemistry to develop a new labeling and MS multiplexing strategy.

1.2 Introduction to Mass Spectrometry of Modified Ribonucleic Acids

The remainder of this chapter is based around a previously published review article.

(Borland and Limbach 2017) Mass Spectrometry is a powerful and popular analytical platform for the characterization of biomolecules. However, the application of mass spectrometry for characterizing nucleic acids has lagged behind other classes, due to the simplicity, speed and sensitivity of amplification-driven technologies such as Sanger and Next-Gen sequencing. Where 17 mass spectrometry has proven most useful in nucleic acids is in the direct detection of modified nucleosides. In both deoxyribonucleic acids (DNA) and RNA, many nucleosides can be enzymatically or chemically modified from their four canonical bases of adenosine, guanine, cytosine, and thymine/uracil. The central dogma explains how DNA is made into protein. DNA is transcribed into RNA then translated into protein. During the transcription process, DNA is copied to messenger RNA (mRNA). The mRNA is then taken to the ribosome where its codon makes interactions with the transfer RNA’s (tRNA) anti-codon loop leading to the amino acids on the tRNA forming peptide bonds, also known as translation.

Modifications of the nucleosides comprising the tRNA can play a variety of roles in the cell, however modifications are not limited to tRNA. RNA modifications are also present in mRNA, ribosomal RNA (rRNA) and non-coding RNA. (Marbaniang and Vogel 2016) The role of modifications in both rRNA and tRNA have been linked to improved stability. Along with improved stability, modifications in any type of RNA can lead to changes in protein synthesis.

Modification changes can lead to codon biased translation, (Chan, Pang et al. 2012) tRNA- ribosome binding efficiency (Gefter and Russell 1969) or translational infidelity. (Patil, Chan et al. 2012) Based on all these effects to translation, the regulatory role of tRNA modifications is crucial during times of cell stress. (Gu, Begley et al. 2014) This explains why researchers have observed fluctuation of RNA modification levels during stress.(Chan, Dyavaiah et al. 2010) In times of stress it seems logical for the cell to undergo temporary changes, therefore it makes sense that researcher observed RNA modifications altering the cell’s metabolism. (Helm and

Alfonzo 2014) With the variety of roles RNA modifications play one would expect a link between modifications and disease. Researchers have found modifications in non-coding RNA that have now been linked to diseases, specifically those related to hematopoietic stem cell

18 differential. (Bellodi, McMahon et al. 2013) With the confirmed link to diseases, the urgency to study RNA modifications is more pressing than ever before.

These modifications have important biological functions or outcomes, thus techniques and technologies that enable the rapid determination of modified nucleosides remain an on-going interest. Some modifications can affect RNA editing events, RNA stability, and protein expression. (Zhang, Cozen et al. , Nachtergaele and He 2016, Song and Yi 2017) The effectiveness of mass spectrometry as an enabling technology is that it can reveal the mass and the structure of the modified nucleoside, which many amplification-based approaches are unable to perform directly.

A primary focus of our laboratory has been developing mass spectrometry approaches that enable the rapid and accurate identification of modified nucleosides from RNA. More specifically, one goal has been to create a platform that enables RNA modification mapping – placing identified modified nucleosides into the correct RNA sequence context. The basis for

RNA modification mapping by mass spectrometry is a hyphenated liquid chromatography tandem mass spectrometry (LC-MS/MS) (Kowalak, Pomerantz et al. 1993) approach although an alternative method using matrix-assisted laser desorption/ionization mass spectrometry (MALDI-

MS) has been used by us and others. (Bentzley, Johnston et al. 1996)

RNA modification mapping by mass spectrometry was initially developed by McCloskey and co-workers (Kowalak, Pomerantz et al. 1993). The general approach involves two separate experiments. The first experiment allows one to obtain a census of all the modified nucleosides in the RNA sample of interest by completely digesting the intact RNA into individual nucleosides, which are separated and identified by LC-MS/MS.(Kowalak, Bruenger et al. 1995)

The second experiment requires that the intact RNA first be digested using a specific nuclease,

19 which will generate a mixture of oligonucleotides of varying length. This mixture of digestion products is then analyzed by LC-MS/MS as well.(Ni, Pomerantz et al. 1996) Here, the MS/MS step is used to fragment an oligonucleotide by collision-induced dissociation (CID) such that the original sequence can be reconstructed.(McLuckey, Van Berkel et al. 1992) As noted above, a similar approach can be used with MALDI-MS, as demonstrated by Kirpekar and coworkers during the mapping of post-transcriptional modifications to ribosomal RNAs (rRNAs).(Kirpekar,

Douthwaite et al. 2000)

RNA modification mapping by mass spectrometry is facilitated these days by the availability of known RNA sequences, which arise due to advances in genomic sequencing technologies. These sequences reflect the status of the RNA lacking modification, thus one can readily calculate the molecular weights of unmodified RNAs and any subsequent RNase digestion products using a variety of on-line tools. Because nearly all RNA modifications result in an increase in the mass of the canonical nucleoside, digestion products matching the calculated value will not be modified. As such, experimental strategies now limit data analysis primarily to those RNase digestion products whose masses do not match, suggesting the digestion product contains a modification. The interested reader is directed to a number of recent reviews that describe RNA modification mapping by mass spectrometry in more detail. (Meng and Limbach 2006, Li, Xiong et al. 2016, Ross, Cao et al. 2016, Limbach and Paulines 2017)

1.3 Stable Isotope Labeling Methods for RNA Mass Spectrometry

While methods that allow one to identify modified nucleosides and map those nucleosides onto specific sequence locations of an RNA sample are quite powerful, methods that allow for

20 quantitative measurement of modification levels are needed to better inform and understand the biological significance of these molecules. As is commonly conducted in other areas of mass spectrometry, the field of nucleic acid modifications has turned to stable-isotope approaches to improve both the qualitative analysis of modified RNAs and the quantitative measurement of modified nucleosides/nucleic acids.

Two styles of labeling can be employed: in vivo or in vitro. An example of in vivo labeling is when a medium containing stable isotope labeled nutrients (e.g., essential amino acids) is used in the culturing of the organism of interest. The normal biochemistry pathways of the organism will result in the incorporation of the stable isotope, which provides a specific traceable marker for identification in MS. The most common form of in vivo labeling in mass spectrometry is stable isotope labeling by amino acids in cell culture (SILAC), (Ong, Blagoev et al. 2002) which has found widespread application in protoemics. SILAC relies on the addition of Leu-D3 or 13C labeled arginine or lysine to the medium for incorporation in protein synthesis (Ong, Blagoev et al. 2002, Ong and Mann 2006).

The alternative approach is to use some in vitro method for labeling. Most often, these in vitro methods rely on chemical or enzymatic strategies to incorporate the stable isotope into the biomolecule(s) of interest. Due to ionization rates and the potential coupling to liquid chromatography, the use of labels as close to the original product is desired. The goal of isotopic labeling is to change the mass of the ion but minimize affects on ionization rates and chromatographic retention.

Bruckl and co-workers demonstrated parallel isotope-based quantification of modified transfer RNA (tRNA) nucleosides.(Brückl, Globisch et al. 2009) A subset of modified tRNA nucleosides were generated using deuterium labeling of a methyl group, which were used as

21 internal standards to quantify modified nucleoside levels in tumor cells versus healthy tissue. The area under the peak of the known concentration spike was compared back to the calibration curves created for each isotopically labeled modified nucleoside. Kellner and co-workers have developed a method for the absolute quantification of modified ribonucleosides using biosynthetic isotopomers.(Kellner, Ochel et al. 2014) By feeding 13C glucose to bacteria, stable isotope labeled modified RNAs are synthesized by the organism. The stable isotope labeled RNA can be hydrolyzed to nucleosides and those naturally occurring modified nucleosides, generated by the bacterium, are then used as internal standards. This concept is taken a step further to observe when methylation events occur throughout the growth phase by spiking deuterated methionine into the growth media. (Heiss, Reichle et al. 2017)

A different stable-isotope labeling strategy was developed by Dickman and co- workers.(Waghmare and Dickman 2011). By using 15N-labeled medium, they could successfully map post-transcriptional modifications in bacterial 16S rRNA. The approach requires two samples – one cultured in 14N-labeled medium and the other in 15N-labeled medium. This labeling approach allows for the unambiguous identification of base composition in each digestion product, thereby improving the accuracy of RNA modification mapping experiments. While the quantitative applications of this approach were not explored in depth by

Dickman and co-workers, such a strategy clearly pointed towards the additional utility of isotope labeling for relative quantification during RNA modification mapping experiments.

Williamson and coworkers adapted the Dickman methodology to quantify rRNA modification levels.(Popova and Williamson 2014) Cells were cultured in minimal media and minimal media supplemented with 15N ammonium sulfate as the heavy nitrogen source. The heavy labeled culture was used as an internal standard. Known ratios of 15N and 14N cultured cells were combined for relative quantification. Methylated nucleosides in rRNA were quantified

22 using CD3-methionine supplemented medium while psuedouridine modifications were quantified by cultures supplemented with 5,6-D-uracil.

More recently, an alternative strategy has been developed by Taoka and co-workers for the absolute quantification of posttranscriptional modifications in rRNA.(Taoka, Nobe et al.

2015) This approach, deemed stable isotope-labeled ribonucleic acid internal standard

(SILNAS), relies on an internal standard that is generated by in vitro transcription of RNA using

13C labeled nucleoside triphosphates, which generates an unmodified copy of the rRNA uniformly labeled. After RNase digestion of both the sample of interest and the in vitro transcript internal standard, any LC peaks that lack a co-eluding heavy transcript (internal standard) were indicative of a modification in that digestion product. The modified oligonucleotide could be quantified through the ion abundance ratio of spike to sample.

1.4 18O labeling

The origins of 18O labeling in mass spectrometry first arose in the field of proteomics. In

1951, Sprinson and Rittenberg took advantage of 18O to better understand enzyme activity of proteases.(Sprinson and Rittenberg 1951) This idea was adapted and used by Desiderio and Kai in protein sample preparation for mass spectrometry.(Desiderio and Kai 1983) By 1983 they were taking advantage of stable isotope-incorporated peptide internal standards for field desorption mass spectrometry quantification of peptides in biological tissues.(Desiderio and Kai

1983) They were preparing internal standards by digesting proteins with trypsin in 18O-labeled water. Due to the mechanism of trypsin, digestion under these conditions can result in the C- terminus of the peptide being labeled with one or two 18O molecules, which leads to a 2 or 4 Da

23 mass increase in the tryptic peptide (Figure 1). As such, unique heavy internal standards could be generated and applied to the mass spectrometry-based analysis of peptides (and proteins) for identification and quantification.

Figure 1.1 Endonuclease method for RNase T1 and other RNases. The RNA oligonucleotide is cleaved and a 3’ cyclic phosphate intermediate is formed. An 18O atom from the reaction solvent can break the cyclic phosphate and allow for the addition of the stable isotope label to the final 3’ linear phosphate product. Protease method for Trypsin. The protein is cleaved at the C-terminus after arginine or lysine residues. The 18O form the labeled reaction solvent is incorporated into the newly formed carboxylic acid after the cleavage site

In 2000 Mirgorodskaye paired 18O stable isotope labeling of proteins with MALDI-MS for quantification of protein samples.(Mirgorodskaya, Kozmin et al. 2000) The following year

Yao introduced a shotgun comparative proteomics tool based on 16O versus 18O labeling of trypsin digested proteins.(Yao, Freas et al. 2001) This comparative proteomics labeling approach has even been applied to clinical samples when a pooled 18O labeled reference sample was

24 spiked into patient samples.(Qian, Liu et al. 2009) Unlike in vivo approaches, 18O labeling is cost effective due to the minimization of “wasting” the stable isotope as labeling is only performed on extracted protein. On the other hand 18O labeling was found to have drawbacks as compared to in vivo methods including a poorer dynamic range in protein identification and a limit of detection – at best – in the high femtomole range.(Lange, Sylvester et al. 2010)

The use of 18O labeling in mass spectrometry has not been limited to only proteins. In 2013,

Hamasaki and coworkers used solid phase synthesis to incorporate 18O into the oligonucleotide to enable the mass spectrometry-based study of oligonucleotide therapeutics.(Hamasaki,

Matsumoto et al. 2013) Because these labeled standards are generated via solid-phase synthesis, they can be used as quantitative standards for a variety of pharmacokinetic and pharmacodynamics studies, in particular for oligonucleotide therapeutics like small interfering

RNAs (siRNAs). The mechanism of action and drug clearance can be studied directly due to the mass label, which provides advantages over fluorescently tagged siRNAs that may not behave ideally due to the structural differences in the drug caused by the fluorescent tag.

1.4.1 18O Labeling of RNA – Early Applications for Mass Spectrometry

Learning from the field of proteomics, Beniam Berhane in our lab began investigating the applicability of enzyme-mediated labeling of nucleic acids using 18O-labeled water. The initial studies focused on whether the similarity of enzyme mechanisms between proteases, such as trypsin, and nucleases, such as Ribonuclease T1 (RNase T1), would enable a similar labeling method to be used for RNA (Figure 1.1).(Berhane and Limbach 2003) Once it was found that

RNase T1 could be used to incorporate 18O onto the 3’-terminal phosphate of the oligonucleotide digestion product, this approach was exploited to simplify data interpretation in

MALDI post-source decay (PSD) analysis of oligonucleotides.(Berhane and Limbach 2003) 25 Samples were digested in 50:50 light (16O-labeled) and heavy (18O-labeled) water to give the characteristic doublets for samples successfully digested bearing the 3’ phosphate group

(Figure 1.2). The doublet leads to simplified identification of products for further analysis. The only potential drawback was the need to use twice as much sample, because one was digested in light water at the same time as the other half of the sample was digested in the 50/50 mixture.

This approach allowed for the spectra to be directly compared. Without the “normal” spectrum, it would be difficult to identify the +2 doublet of the 18O labeled digest.

Once we determined that certain RNases could be used to enzymatically label terminal phosphates with a single 18O, Zhaojing Meng in the lab next turned to developing an approach for quantifying RNase digestion products.(Meng and Limbach 2005) This time samples were separately digested in 18O-labeled and 16O-labeled water. Method development was performed using commercially available Escherichia coli tRNA-Val to determine the effectiveness of this strategy for sample quantification (Figure 1.3). The heavy and light digestion products were combined in ratios from 1:10 through 10:1 and analyzed using MALDI-MS. The averaged ion abundance ratio (heavy:light) was plotted against the prepared sample ratio to generate a calibration curve

From this study it was determined that more accurate results were obtained when the 18O- labeled sample was more abundant than the 16O-labeled sample. When the 16O-labeled sample is more abundant, other natural isotopes in the digestion product (e.g., 13C, 15N) can interfere with accurate determination of the 18O-labeled peak abundance. By ensuring the more abundant sample is labeled with 18O, those interfering isotope peaks from the 16O-labeled sample are proportionally much less than the 18O peak abundance, which minimizes errors in relative quantification. This information can guide the application of this approach in

Figure 1.2 Expanded view of MALDI mass spectral data obtained from the RNase T1 digestion products of E. coli 5S rRNA. (a) Digestion was done in unlabeled water. Three major ions are detected (m/z 980, 987 and 998). (b) After digestion in a 50:50 (v/v) mixture of unlabeled and 18O-labeled water, only those oligonucleotide digestion products that contain a 3’-phosphate will exhibit the characteristic A + 2 doublet. From this, base compositions for the three ions can be made or confirmed. Reproduced with permission from Berhane, B.T. and P.A. Limbach, Stable isotope labeling for matrix-assisted laser desorption/ionization mass spectrometry and post-source decay analysis of ribonucleic acids. J. Mass Spectrom., 2003. 38: p. 872-878, Copyright 2003.

27 .

Figure 1.3 Representative MALDI mass spectrum of RNase T1 digestion products obtained from a tRNA-Val mixture prepared at a heavy-to-light RNA ratio of 2:1. The asterisk (*) denotes expected RNase digestion product pairs. Inset: expanded view of the RNase T1 digestion product 5’-CUCAGp-3’ used for quantifying RNA levels for tRNA-Val with overlaid calculated isotopic distribution assuming a 2:1 ratio. Reproduced with permission from Meng, Z. and P.A. Limbach, Quantitation of Ribonucleic Acids Using 18-O Labeling and Mass Spectrometry. Anal. Chem., 2005. 77: p. 1891-1895 Copyright 2005.

quantitative analysis. Accurate relative quantification required the generation of a calibration curve for each RNase digestion product of interest. To demonstrate the robust nature of this approach, a blinded analysis of heavy and light E. coli tRNA-Val mixtures was performed using the previously established calibration curve.

One of the more significant limitations of using enzyme-mediated labeling of RNA is that this approach requires complete enzymatic digestion of the RNA. As the mechanism involves a cyclic phosphate intermediate (Figure 1.1), incomplete digestion products will not be labeled, significantly impacting the utility of this approach.(Berhane and Limbach 2003) To circumvent this issue, higher amounts of RNases and a longer incubation time at an optimal temperature

28 have been linked to a decrease in cyclic phosphate digestion products.(Berhane and Limbach

2003, Hartmer, Storm et al. 2003) Another experimental challenge identified is the presence of sodium or potassium salt adducts to the RNase digestion products. These adducts can interfere with accurate detection and quantification, thus sample preparation and desalting are important to ensure accurate relative quantification when using MALDI-MS.

Having demonstrated the applicability of 18O-based quantification of individual RNA samples, our interest next turned to using this approach to examine more complex mixtures of

RNAs. Our specific interest was to characterize the total pool of tRNAs in a cell, which would obviate the need to individually purify tRNAs one by one from a sample. The analysis of total tRNA pools would not only decrease sample preparation time, it would also allow more information to be gained from a single mass spectrometry experiment. When such a strategy is applied to the total cellular pool of tRNAs, information regarding codon usage and potential codon bias can be obtained in a more straightforward fashion.(Castleberry and Limbach 2010)

The initial method developed to identify individual tRNAs within an unseparated mixture of total tRNAs was presented by Mahmud Hossain, who described the signature endonuclease digestion product (SDP) concept for tRNA identification using MALDI-MS.(Hossain and

Limbach 2007) In short, when one performs an in silico digest of known tRNA sequences (e.g., all E. coli tRNAs), the RNase digestion product masses that result will reveal that each individual tRNA will have at least one digestion product that is unique in both mass and sequence. Thus, these unique or signature digestion products can be used as a proxy to confirm the presence of any specific tRNA within the sample mixture.

Colette Castleberry built upon this SDP strategy by first demonstrating LC-MS/MS was just as effective at SDP identification as MALDI-MS.(Castleberry and Limbach 2010) In the

29 same work, she then focused on combining RNase-mediated 18O-labeling and the signature digestion product concept to create quantitative signature digestion products (qSDPs) – those

SDPs that could simultaneously be used for tRNA identification and quantification. The criteria for defining a digestion product as a qSDP include incorporation of the 18O label, a difference in mass by more than 2 Da from other known digestion products, and the labeled SDP must be able to provide a linear response spanning a 5-fold change in SDP amount. By creating a set of qSDPs, studies on how specific tRNA levels change as a result of culturing conditions were examined (Table 1.1).

1.4.2 18O Labeling of RNA for Modification Mapping by Mass Spectrometry

Our lab next turned the focus to how 3’-terminal phosphate labeling can enable alternative strategies to characterize RNA samples. Of particular interest to the lab is the discovery and characterization of post-transcriptionally modified nucleosides in RNA samples.

As noted earlier, RNA modification mapping by mass spectrometry is an analytical approach that is used to identify the specific sequence location for modified nucleosides. Although several different strategies have been developed for modification mapping, it was thought that by using

18O labeling, one could multiplex the analysis. This would reduce analysis time and cost as well as improving run-to-run reproducibility

While 18O labeling only enables duplex analysis (i.e., two different samples in a single analysis), it provides the template for even higher levels of multiplexing in the future. Our first

18O labeled multiplexing investigation was described in 2012 by Siwei Li.(Li and Limbach

2012) Comparative analysis of RNA digests (CARD) pairs a sample with known

30 Table 1.1 Analysis of RNase T1 quantitative signature digestion products from E. coli. Table reproduced with permission from Castleberry, C.M. and P.A. Limbach, Relative Quantitation of Transfer RNAs Using Liquid Chromatography-Mass Spectrometry (LC-MS) and Signature Digestion Products. Nucleic Acids Res (2010) 38 e162. tRNA qSDP Sequence Experimental I18/I16 %CV Decrease in Relative Abundance Cys CA[ms2i6A]ACCGp 0.75 19% Cys U[s4U]AACAAAGp 0.67 25% Tyr 1, 2 ACUQUA[ms2i6A]ACUGp 0.60 18% Increase in Relative Abundance Gly 1 AUUCCCUUCGp 1.44 26% Gly 2 CCU[Um]CCAAGp 1.28 24% Gly 3 AAUAGp 1.88 13% Ser 1, 4, 5 AAAGp 2.43 18% Ser 1, 2 A[ms2i6A]AACCGp 1.42 10% No Change in Relative Abundance Ala 1, 2 [m7G]UCUGp 1.17 15% Arg 1, 2 [m2A]ACCGp 1.04 20% Asn UCCUCUGp 1.21 15% Glu 1, 2, 3 AAUCCCCUAGp 1.06 15% Glu 1, 2, 3 UCCCCUUCGp 1.20 15% Leu 1 UCCCCCCCCUCGp 1.09 15% Phe AA[ms2i6A]ACCCCGp 1.31 13% Phe A[s4U]AGp-3’ 1.22 25% Phe U[m7G][acp3U]CCUUGp 1.21 25% Ser 3 CUCCC[s2C]UGp 1.00 17% Trp UCUCUCCGp 1.21 27% Trp U[Cm]UCCA[ms2i6A]AACCGp 1.29 22% Val 1 AU[s4U]AGp 0.86 16% Indeterminate His UU[m7G]UCGp 1.80 24% His AAUCCCAUUAGp 1.36 21% His [m2A]CCAGp 1.00 27% Ini 1, 2 TCAAAUCCGp 1.51 23% Ini 1, 2 [Cm]UCAUAACCCGp 0.97 41%

posttranscriptional modification with a sample of unknown posttranscriptional modifications.

The idea is that any peaks appearing as doublets separated by 2 Da indicate that the digestion product from the “unknown” is identical to the digestion product of the “known” or reference sample. Thus, by examining all doublets one can quickly identify the similarities of RNA samples. By the same reasoning, digestion products appearing as singlets (either from the 16O- labeled sample alone or the 18O-labeled sample alone), inform one of differences between the two RNA samples. These differences could arise because the unknown sample is modified differently than the known sample or singlets could arise due to sequence differences in the two samples (Figure 1.4)

While CARD was found to significantly improve RNA modification mapping of total tRNA pools from organisms whose tRNA modification patterns were previously unknown, the minimal mass difference of the 18O-label (versus 16O-label) combined with interferences from naturally occurring stable isotopes (e.g., 13C) limited our ability to generate automated methods for identifying singlets and doublets within the sample. To overcome this limitation, Collin

Wetzel worked with Siwei to investigate culturing conditions that would minimize stable isotope interferences. We used 12C-enriched medium during cell culturing to essentially eliminate 13C

(and 15N) isotope interferences during CARD (Figure 1.6).(Wetzel, Li et al. 2014). This culturing strategy leads to identification of singlets and doublets that can be automated due to improved differentiation of doublets. Moreover, this strategy can again be combined with the

SDP approach to provide more targeted tRNA analysis.

Another limitation of the CARD strategy was identified by limiting singlet and doublet measurements to only mass measurements. RNase digestion products having the same mass but different sequences in the two samples could co-elute and appear as doublets, leading to a false

positive in the analysis. To address this challenge, Siwei showed that 16O/18O-labeled digestion products can be differentiated based on MS/MS data,(Li and Limbach 2014) echoing our very first studies of 18O-labeling during MALDI PSD experiments.(Berhane and Limbach 2003)

Reference RNA Candidate RNA RNase T1 in RNase T1 in RNase T1 18 18 RNase T1 in 16 H O H O 16 2 2 H O in H2 O 2

Combine and analyze by LC-MS Combine and analyze by LC-MS

Reference Reference Reference Reference Reference Candidate Reference Candidate Candidate Candidate Candidate Candidate

m/z m/z

Reference Candidate

18 18 16 18 16 16 18 O O 16 O O O O O O m/z

Figure 1.4 Schematic outline of comparative sequencing by isotope labeling and LC-MS where Escherichia coli serves as the reference organism and Citrobacter koseri serves as the candidate (unknown) to be sequenced. tRNA endonuclease digestion products that are equivalent between organisms will appear as doublets (separated by 2 Da) in the mass spectral data; digestion products that are different between the two organisms will appear as a singlet. Reproduced with permission from Li, S. and P.A. Limbach, Method for comparative analysis of ribonucleic acids using isotope labeling and mass spectrometry. Anal Chem, 2012. 84(20): p. 8607-13 Copyright 2012.

Figure 1.5 Mass spectra corresponding to a detected singlet when (a) C. koseri is labeled with 18O and (b) E. coli is labeled with 18O. The singlet U[s4U]AACAAAGp (m/z 1469.6, 2-charge) arises from the E. coli tRNA- Cys(GCA) as confirmed by the +1 increase in the m/z isotopic envelope after 18O-labeling of E. coli. Reproduced with permission from Li, S. and P.A. Limbach, Mass spectrometry sequencing of transfer ribonucleic acids by the comparative analysis of RNA digests (CARD) approach. The Analyst, 2013. 138(5): p. 1386-94 Copyright 2013

When oligonucleotides are fragmented during CID MS/MS, the most abundant fragment ions are the c-type and y-type ions, representing the oligonucleotide sequence from the 5’- and

3’-termini, respectively. The y-type ions contain the 16O or 18O label on the 3’-phosphate.

Knowing this, Siwei demonstrated that the MS/MS data can be used to confirm that doublets detected in the mass spectrum are truly the same sequence rather than sequence isomers.

Doublets detected in the y-type ions during MS/MS can only arise if the two sequences are identical. Sequence isomers are identified by singlets in the y-type ions, which occur wherever sequence differences are present in the original digestion products.

Singlets Doublets a 1201.3 b 1201.6

100 1200.8 100

LB medium 80 80 1200.6 1202.1 1201.8 60 60 1202.6 40 40 1202.3 20 1203.1 Relative Abundance 20

0 0 1200.8 1200.9 c 100 d 100 1201.8 12 80 C -enriched medium 80

60 60

40 40

20 1201.3 1202.4 Relative Abundance 20

0 0 1198 1200 1202 1204 1206 1208 1198 1200 1202 1204 1206 1208 m/z m/z

Figure 1.6 Improvements in singlet and doublet identification using 12C-enriched medium as illustrated with the doubly-charged E. coli total tRNA RNase T1 digestion product A[ms2i6A]AACCGp (MW 2403.4 Da). (a) Mass spectrum from sample grown in rich medium and labeled with 16O during RNase T1 digestion. (b) Same sample as in Figure 6a except labeled with both 16O and 18O during RNase T1 digestion. (c) Mass spectrum obtained when sample grown in 12C-enriched medium and labeled with 16O during RNase T1 digestion. (d) Same sample as in Figure 6c except labeled with both 16O and 18O during RNase T1 digestion. Singlet and doublet identifications are simplified in Figure 6c and d, respectively, by use of 12C-enriched medium. Reproduced with permission from Wetzel, C., S. Li, and P.A. Limbach, Metabolic De-Isotoping for Improved LC-MS Characterization of Modified RNAs. Journal of the American Society for Mass Spectrometry, 2014. 25(7): p. 1114-23. Copyright 2014.

1.5 Future Outlook

As discussed above, phosphate labeling by using enzyme-mediated incorporation of 18O into

RNase digestion products has been used in numerous ways to improve the mass spectrometry- based characterization of modified RNAs. However, a fundamental limitation remains that the minimal mass difference between 16O and 18O limits the overall utility of this approach to those examples discussed previously. It will be of interest to examine alternative strategies for phosphate labeling, which are known in the field of RNA biology, but which have not yet entered the world of mass spectrometry.

T1 ligase has been used in radiolabeling for visualizing RNA since the late 1970’s. (Bruce and Uhlenbeck 1978, England, Bruce et al. 1980) This ligase has not been explored in mass spectrometry applications because it is known to generate a variety of different side products.

However, more recently a T4 RNA ligase was created to reduce these unwanted ligation products.(Viollet, Fuchs et al. 2011) With this advancement, T4 RNA ligase may become a more promising tool for stable isotope labeling of oligonucleotides, including RNase digestion products. Another potential enzyme for RNA labeling is Thg1.(Abad, Rao et al. 2010) The role of Thg1 in the cell is to add a single guanosine nucleotide to the 5’ terminus of tRNA-His.

However, it has been shown that this enzyme has 3’ to 5’ polymerase activity. (Jackman, Gott et al. 2012) With additional study, it may be possible to use this unique function and activity to incorporate stable isotopes into specific RNA samples, which could be part of a broader mass spectrometry strategy for sample characterization.

Another area where RNA mass spectrometry in general, and RNA modification mapping in particular, can look to for inspiration and ideas for phosphate labeling and multiplexing strategies is the field of proteomics. A significant diversity of multiplexing strategies have been developed

in proteomics.(Timms and Cutillas 2010) Given the demonstrated advantages of relatively rapid characterization of multiple samples by these proteomics approaches, one can envision the development of tools and technologies for RNA mass spectrometry that provide similar advantages, even if the particular chemistry and biochemistry may differ due to the unique characteristics of RNA. Regardless, mass spectrometry as a platform for RNA analysis in general, and RNA modification mapping in particular, are now well appreciated. It remains an ongoing challenge for the community to identify and develop the needed sample labeling tools to take full advantage of this platform.

Based on the previous work completed in mass mapping and duplexing strategies, I explore the use of enzymatic and chemical labeling strategies to develop a platform for RNA multiplexing. This platform should allow for rapid characterization of multiple samples. It will take advantage of isotopic labeling strategies for multiplexing samples, and also uses an enzymatic strategy for selective labeling.

2 CHAPTER 2 MULTIPLEXING

2.1 Background

The previous work on duplexing, discussed in Chapter 1, inspired the idea for this enzymatic and chemical labeling platform. Based on a gel based labeling assay of enzymatic labeling with poly adenosine polymerase (PAP), oligonucleotides can be enzymatically labeled with azido modified nucleotides. (Winz, Samanta et al. 2012) This strategy followed by click chemistry can produce uniquely labeled oligonucleotide samples post-RNase digestion. Using alkynes for CUCCA click reactions on RNA is common in a variety of biochemistry experiments including monitoring DNA and RNA synthesis, selectively. (Jao and Salic 2008, Best 2009, Qu,

Zhou et al. 2013, Schulz and Rentmeister 2014) These two strategies, in combination, are used to create an LC-MS multiplexing strategy for oligonucleotides.

By pairing enzymatic labeling with several isotopically labeled alkynes for click chemistry sample can be combined beyond duplexing. The use of isotopically labeled alkynes can have higher molecular weight difference than previously achieved with 18O labeling. This platform is capable of multiplexing although only duplexing is shown here for proof of concept. The reason more samples were not uniquely labeled was due to the limited commercially available of isotopically labeled reagent.

2.2 Experimental

2.2.1 Materials:

Alkynes, CuSO4, tris(3-hydroxypropyltriazolylmethyl)amine (THPTA), 1,3,3,3-

hexafluoro-2-propanol (HFIP), and triethylamine (TEA) were purchased from Sigma-Aldrich

(St. Louis, MO). LC-MS grade water and methanol were purchased from Honeywell B&J

(Morristown, NJ). Synthetic oligonucleotides were purchased from Eurofins (Huntsville, AL) or

Dharmacon (Lafayette, CO). Poly(A) polymerase (PAP) was purchased from Affemetrix

(Cleveland, OH). Azido modified deoxynucleoside triphosphates (dNTPs) were purchased from

TriLink (San Diego, CA).

2.2.2 Identification of Optimal azido-modified dNTP

Template S1 (UAACUAUAACGG) was reacted with two different 2’-N₃dNTPs at a 1:10 ratio

(RNA:2’-N3dNTP) with 215 U PAP/ µg RNA for 1 h at 30 and 37 °C. The reaction was heat

inactivated for 10 min at 62 °C. The reaction products were separated on a 20% polyacrylamide

gel with 1xTBE running buffer and visualized using Sybr Gold.

2.2.3 Optimization of PAP reactions

Oligonucleotide templates (CUG, AAG, CAAG, ACCCUG) were reacted at a 1:2 molar

ratio with 2’N3dUTP with various amounts of PAP. The incubation time and temperature were

varied to identify optimal extension conditions. All reactions were heat inactivated for 10 min at

62 °C.

Another set of oligonucleotide templates (GUG, ACAG, CAU[Gm]UGG,

[m3U][Am]ACAAGG, UAACAUAACG) were reacted at a 1:2 molar ratio with 2’N3dUTP and varying amounts of PAP while holding incubation period and temperature constant (4 h and

28 °C). All reactions were heat inactivated for 10 min at 62 °C.

2.2.4 Click reaction optimization

The set of oligonucleotide templates (GUG, ACAG, CAU[Gm]UGG,

[m3U][Am]ACAAGG, UAACAUAACG) were first reacted at a 1:4 molar ratio with

2’N3dUTP, 139 U PAP/µg RNA for an incubation period of 4 h at 28 °C with heat inactivation for 10 min at 62 °C. Reaction aliquots were then reacted with 20 molar equivalence of CuSO₄,

50 equivalence of THPTA, 100 equivalence of sodium ascorbate, and varying equivalence of alkyne relative to RNA. Click reaction were incubated for various time periods at 37 °C.

2.2.5 Calibration curve

A calibration curve was created using [m3U][Am]ACAAGG, which was reacted at a 1:2 molar ratio with 2’N3dUTP, 139 U PAP/µg RNA for an incubation period of 4 h at 28 °C with heat inactivation for 10 min at 62 °C. Click reactions were then performed using 20 equivalence of CuSO4, 50 equivalence of THPTA, 100 equivalence of sodium ascorbate, and 50 equivalence of alkyne (all relative to initial oligonucleotide equivalence). Samples with appropriate alkynes were mixed prior to analysis.

2.2.6 Model duplexes

Oligonucleotide pairs (GUUCAG/GUUC[I]G, CAUGUGG/CAU[Gm]UGG,

UAACAAGG/[m3U][Am]ACAAGG, and AUCUCGACCG/AUCUCG[m2A]CCG) were mixed in varying ratios of unmodified to modified and then reacted at a 1:2 molar ratio with

2’N3dUTP, 204 U PAP/µg RNA for an incubation period of 4 h at 28 °C with heat inactivation for 10 min at 62 °C. Equal volume aliquots from each pair were then reacted with alkynes of different isotopically labeled states for 90 min at 37 °C. The click reactions were performed using 20 equivalence of CuSO4, 50 equivalence of THPTA, 100 equivalence of sodium ascorbate, and 50 equivalence of alkyne (all relative to initial oligonucleotide equivalence).

Samples with complimentary alkynes were mixed prior to analysis.

2.2.7 LC-MS

Samples were separated using a Poroshell C18 column (50 mm × 1 mm; 2.7 µm –

Agilent, Santa Clara, CA, USA) thermostatted at 45 °C with mobile phase A (MPA) made with

200 mM HFIP, 8 mM TEA and mobile phase B (MPB) composed of 200 mM HFIP, 8 mM TEA in 50:50 (v/v) H₂O/methanol. The mobile phase gradient started at 5 %B, held for 5 min, increased to 73 %B over 35 min, and finally ramped to 100 %B for 5 min before re-equilibrating prior to the next analysis.

LC-MS/MS data were acquired on a Waters Synapt G2-S HDMS mass spectrometer with a source temperature of 120 °C, desolvation temperature of 400 °C, capillary voltage of 2.5 kV, sampling cone of 55V, source offset of 80V, cone gas of 50 L/hr and desolvation gas of 700

L/hr. For all measurements, the mass spectrometer was operated in V-mode (sensitivity) with a typical resolving power of 15,000 FWHM (full width at half maximum). Samples were analyzed

in negative-mode ESI over an m/z range of 405 to 2000 for MS and 300 to 2000 for MS/MS.

Data dependent acquisition was used to collect MS/MS data (1 s scan) for a maximum of three ions per MS scan (0.2 s scan) using a collision energy ramp from 10V to 20V for low masses and

30 V to 40 V for high masses before being added to a dynamic exclusion list for 15 s. Precursor ion selection was performed with an isolation width of 2. Lockspray calibration was performed using a solution of leucine enkephalin (200 pg/µL) infused at 5 µL/min. Lockspray scans were collected for 1 s every 60 s with setmass at m/z 554.2615. A flow rate of 80 µL/min was used throughout. Data acquisition was through the MassLynx software.

2.3 Results and Discussion

The overall scheme for post-RNase digestion labeling of oligonucleotides in preparation for mass spectrometry analysis involves the use of PAP to add a single azido-modified dNTP onto the 3’- terminus of the digestion product. This azido-modified dNTP then provides the platform for copper-catalyzed azide-alkyne cycloaddition to place mass unique tags onto each extended oligonucleotide. To establish this platform by which RNase digestion products from different samples could be uniquely labeled and then combined for subsequent multiplexed LC-MS analysis, each step in the protocol was individually optimized to identify the necessary conditions for this multiplexing strategy.

2.3.1 Identification of optimal azido-modified dNTP

Initial studies were formulated based on the previous report of using azido-modified nucleotides as a chemical handle for labeling oligonucleotides using chemical dyes (Winz,

Samanta et al. 2012). The first task was to identify an appropriate, consistent azido-modified

dNTP which could be extended onto an oligonucleotide substrate at near unit efficiency.

Previous results have shown that yeast PAP is preferred over bacterial PAP in these types of template labeling experiments (Winz, Samanta et al. 2012). Moreover, Winz and co-workers already noted that 2’N3 dGTP and 2’N3 dUTP were more appropriate reagents as they demonstrated good extension efficiency while not leading to significant levels of multiply- extended products (Winz, Samanta et al. 2012).

As seen in Figure 2.1, 2’-N₃dUTP produced more single addition product than 2’-N₃dGTP using yeast PAP and the oligonucleotide UAACUAUAACGG. Multiple analyses revealed that

30 °C provided more favorable extension than 37 °C (data not shown). Based on these initial investigations, all further extension reactions were performed using 2’-N3dUTP.

2.3.2 Optimization of PAP extension

The next goal was to identify appropriate PAP extension conditions, including ratio of reagents, incubation period and temperature (Table 2.1). To understand the influence of the oligonucleotide template length on these extension reactions, a set of oligonucleotides (CUG,

AAG, CAAG, ACCCUG) of varying length and base composition were used. In sum, while base composition appeared to have more of an impact than length, conditions could be identified that would allow for >85% extension of 2’-N3dUTP onto the oligonucleotide template. Based on these experiments, optimized reaction conditions included a 1:2 RNA to 2’N₃dUTP molar ratio,

90 U PAP/µg RNA, incubation period of 4 h at a temperature between 25 °C and 30 °C.

Figure 2.1 To determine which 2’-N₃dNTPs is “ideal” for PAP extension, template S1 (UAACUAUAACGG) was reacted with two different 2’-N₃dNTPs at a 1:10 ratio and PAP for 1 h at 30 and 37 °C. The reaction was heat inactivated, and reaction products were separated on a 20% polyacrylamide gel. Oligonucleotide bands were visualized using Sybr Gold. This gel shows a mixture of extended and unextended template for the 2’- N₃dGTPs sample, but the 2’-N₃dUTPs shows much more extended than unextended template. When comparing the 30 and 37 °C for 2’-N₃dUTP, more extended product is observed in the 30 °C incubated sample

Table 2.1 Optimization of PAP extension conditions for a set of unmodified oligonucleotide templates. Extension levels for each template were quantified based on peak area ratios for extended versus un-extended template U PAP/ ug RNA ratio 45 45 45 45 90 Temperature (°C) 22 25 30 25 25 Incubation period (min) 120 120 120 240 120 CUG 6% 4% 8% 12% 14% AGG 63% 85% 91% 94% 98% CAAG 66% 63% 65% 67% 73% ACCCUG 66% 82% 96% 98% 98%

Next I sought to determine if these conditions would be appropriate for somewhat larger oligonucleotides and whether oligonucleotides containing post-transcriptional modifications would adversely impact the extension reaction. This new set of oligonucleotides included GUG,

ACAG, CAU[Gm]UGG, [m3U][Am]ACAAGG, and UAACAUAACG. Similar to the results with the smaller length unmodified oligonucleotides, increasing the relative amount of

PAP added to the reaction resulted in greater extension yield (Table 2.2). Extension levels >75% were found for the 7-mer and smaller templates, with very low extension efficiencies seen using

[m3U][Am]ACAAGG and UAACAUAACG. Thus, these two sets of extension data suggest that PAP addition of 2’N₃dUTP prefers smaller length oligonucleotide templates.

Table 2.2 Effect of poly(A) polymerase to template ratio on oligonucleotide extension. Extension reactions were performed for 240 min at 28 °C. Extension levels for each template were quantified based on peak area ratios for extended versus un-extended template U PAP/ug RNA ratio 45 70 139 GUG 87% 83% 88% ACAG 43% 34% 86% CAU[Gm]UGG 75% 75% 75% [m3U][Am]ACAAGG 21% 23% 39% UAACAUAACG 1% 6% 15%

2.3.3 Click reaction optimization

To identify optimal click reaction conditions, experiments were performed using

2’N₃dUTP extended oligonucleotides as substrates. The previous set of templates containing modified nucleosides (GUG, ACAG, CAU[Gm]UGG, [m3U][Am]ACAAGG, and

UAACAUAACG) was used in these studies. To mimic the worst case scenario, the ratio of

RNA to 2’N₃dUTP was increased from 1:2 to 1:4 leaving more azido-modified dNTP around to potentially compete against the extended oligonucleotide.

The first set of reaction conditions investigated focused on the reaction period (Table 2.3).

For every template, it was noted that the click reaction with phenylacetylene was more effective at 90 minutes versus 180 minutes. The major product of these reactions was a click-reaction product as investigation of the LC-MS data from the 90 min and 3 h click reactions did not reveal any significant oxidation or degradation products (data not shown). Next, the amount of added alkyne was varied (Table 2.4). Again, the trend was that a lower level of alkyne equivalence was more efficient for the click reaction under these experimental conditions. These results could arise due to the limited solubility of the alkyne in the reaction mixture. Future studies are planned to investigate additional alkynes with varying solubility. However, for these initial developmental studies, it was determined that optimal click reaction conditions were to use 50 equivalence of alkyne with a reaction period of 90 min at 37 °C.

Table 2.3 Effect of reaction period on phenylacetylene click efficiency. Extension levels for each template were quantified based on peak area ratios for extended versus un-extended template. Time (min) 90 180 GUG 87% 67% ACAG 96% 88% CAU[Gm]UGG 80% 49% [m3U][Am]ACAAGG 96% 90% UAACAUAACG 90% 76%

Table 2.4 Effect of phenylacetylene equivalence on click reaction efficiency. Extension levels for each template were quantified based on peak area ratios for oligonucleotides that were labeled with the click reagent versus extended but not labeled. Phenylacetylene equivalence 50 100 GUG 87% 77% ACAG 96% 95% CAU[Gm]UGG 80% 76% [m3U][Am]ACAAGG 96% 91% UAACAUAACG 90% 57%

2.3.4 Model Duplexing Studies

As envisioned, this approach would enable multiplexed analysis of oligonucleotides by using alkynes of different masses. To ensure differentially labeled oligonucleotides would co- elute, the ideal reagent would by isotopically labeled alkynes. To first confirm that this supposition was correct, phenylacetylene in an unlabeled (D0) and deuterium (D5) labeled form was used as a model to test oligonucleotide labeling and mass spectral analysis of duplexed mixtures.

The oligonucleotide templates GUG, ACAG, CAU[Gm]UGG, [m3U][Am]ACAAGG, and UAACAUAACG, were first modified by end-labeling with 2’-N3dUTP through the PAP extension and then split into separate aliquots for subsequent click reaction with either the unlabeled or deuterium labeled phenylacetylene. Representative mass spectral data for individual

(D0 or D5) and equal molar mixtures (D0 and D5) are shown in Figure 2.2 with the extended template CAU[Gm]UGG. As anticipated, the alkyne group does not impact the chromatographic separation of these oligonucleotide.

Figure 2.2 CAU[Gm]UGG is shown as a representative example from an oligonucleotide template that was reacted with 2’N3dUTP and PAP. Reaction aliquots were then clicked with either phenylacetylene in an unlabeled (D0) or deuterium (D5) labeled form. Samples were run individually or combined in a 1:1 ratio. Both labeling reactions were successful. Samples labeled with D0 and D5 in the 2- charge state show a mass difference of 2.5 Daltons. In the 3- charge state, they show a difference of 1.66 Daltons. Even though the samples were mixed 1:1 the mass spectra do not reveal equal ion abundances due to the decreased ion efficiency of the D5 labeled product.

Surprisingly, the ionization efficiency post-extension and click-based labeling was found to be quite sensitive to the deuterium status of the alkyne. This difference in ion abundances was found to be independent of the template used, and was not a result of differences in ion abundance for extended but unclicked templates. When data from each labeled sample was normalized to the unextended template, the relative amounts of extended templates were similar, but the mass spectral response for D5 was always lower than D0. No other side products were observed to account for the lower D5 labeled product. Therefore it is most likely that the deuterium status of the label is contributing to lower ionization efficiency.

I next sought to determine whether the responses seen were linear over a range of sample concentrations. The D0- and D5-labeled oligonucleotides were combined at various ratios, mixed and then analyzed by LC-MS. Table 2.5 provides the relative responses and the results of a linear least-squares fit of these data. Throughout the concentration ratios studied, the samples were found to provide a reproducible and linear response to the relative ratios of the individual components. Thus, while additional work will be required to identify more mass spectrometry friendly labels, the platform can be used in multiplexing experiments where the mass spectral data will report relative amounts of individual components in the mixture.

Table 2.5 A calibration curve with the template [m3U][Am]ACAAGG was created. Samples with appropriate alkynes were mixed prior to analysis. Measurements based on peak area were completed in technical triplicate Prepared (D5/D0) Measured (D5/D0) 0.1 0.139 ± 0.005 0.125 0.127 ± 0.005 0.167 0.173 ± 0.012 0.25 0.220 ± 0.006 0.5 0.285 ± 0.045 1 0.448 ± 0.066 2 0.628 ± 0.074 4 1.000 ± 0.080 6 1.237 ± 0.209 8 1.760 ± 0.708 10 1.885 ± 0.510

2.3.5 Relative Quantification of Oligonucleotide Modification Levels

Another potential use of this labeling strategy would be to characterize relative levels of post-transcriptional modifications within particular RNA sequence contexts. To examine the

utility of this approach, a series of oligonucleotide pairs, which differ in ratio of unmodified to modified oligonucleotide, were generated. Within these pairs, one mixture was labeled with the

D0-phenylacetylene tag and the other was labeled using the D5-phenylacetylene tag. Labeled oligonucleotides were combined and then analyzed by LC-MS/MS. A comparison of peak areas for the eluting oligonucleotides (Figure 2.3) reveals that site-specific modification levels can be determined by this approach.

2.4 Conclusion

A platform that enables the post-RNase digestion labeling of modified oligonucleotides for subsequent multiplexed mass spectrometry analysis has been established. This platform is based upon the controlled extension of the RNase digestion product using an azido-modified dNTP, which then provides the subsequent handle for click-based labeling of different RNA samples.

Further, these labeled oligonucleotides were found to be compatible with subsequent LC-MS analysis, through which the relative amounts could be measured. The future work and direction of this project are discussed further in Chapter 3.

Figure 2.3 CAU[Gm]UGG is shown as a representative example from a mixture of modified oligonucleotide and unmodified oligonucleotide pair templates that were reacted with 2’-N3dUTP and PAP. Reactions were then clicked with either phenylacetylene in an unlabeled (D0) or deuterium (D5) labeled form. Samples were combined in a 1:1 ratio. A) 100% modified oligonucleotide sample labeled with D0 mixed with the D5 labeled 0% modified oligonucleotide sample. As would be expected, the peak corresponding to D0 is much higher than the peak for D5. B) Two 50% modified samples each bearing a different label. The peak for the D0 sample was higher, but not as drastically as panel A. This is likely related to the lower ionization efficiency of the D5 label. C) A 30% modified oligonucleotide sample labeled with D0 and a 70% modified oligonucleotide sample labeled with D5. As expected the D5 peak was larger than the D0 peak. The ionization prevents the D5 peak from being higher, but it is still above the D0 as expected. All of the model duplexing samples visually represent the increasing and decreasing modification status within the two samples.

3 CHAPTER 3 CONCLUSIONS AND FUTURE WORK

3.1 Conclusions

Oligonucleotides can successfully be enzymatically labeled with 2’N₃dUTP followed by click chemistry with isotopically labeled alkynes. These samples can be duplexed and information about each sample can be obtained with a single analysis. Chapter 2 describes a platform for the chemical tagging of ribonucleic acids for analysis by MS. It demonstrates proof of concept as these results were obtained using synthetic oligonucleotide mixtures with and without known modifications. In Chapter 2 conditions were identified that resulted in >75% extension and labeling of oligonucleotides, which can contain modified nucleosides. In addition, the platform was shown to enable the relative quantification of hypomodified oligonucleotides in batch format.

3.2 Future work

To improve the platform future work remains. One of the main challenges is to move from duplexing to multiplexing. For this to occur, more isotopically labeled alkynes should be investigated. A variety of commercially available alkynes have been evaluated to determine their potential usefulness as labels, but none were identified that met the goals of this research. I propose that isotopic labels be synthesized as, based on the availability of starting materials, a wide range of appropriate alkynes could be synthesized.

For the platform to be used on biological samples, a way to remove or inactivate digestive enzyme must be found. Thus far RNase T1, exclusively, is used therefore the work described

focuses on removal or inactivation of RNase T1. Proposed future work in this area should be based on techniques that have already worked or showed promising results. As the platform expands to allow for more samples to be analyzed at the same time, it would be beneficial to investigate post-digestion samples from a variety of RNases. Currently only oligonucleotides ending in G have been investigated, but other RNases leave other 3’ nucleosides. By investigated the 3’ end identity effect on the PAP reaction, the feasibility of using other RNases can be determined.

3.3 Additional alkynes tags

To move from duplexing to multiplexing, more isotopically labeled alkynes should be investigated. In previous work phenylacetylene was used in an unlabeled (D0) and deuterium

(D5) labeled form. These were chosen from a small pool of commercially available isotopically labeled alkynes. Phenylacetylene D5 was chosen due to its large mass increase from the unlabeled form. Other available alkynes have minimal mass differences. The larger the mass difference, the less likely the isotopic distribution will overlap, especially for higher charge states.

I have analyzed a variety of readily available alkynes. Originally alkynes of varying lengths for straight chain hydrocarbons were clicked to 2’N₃dUTP PAP extended oligonucleotides.

Hydrocarbon chains with carboxylic acid functionalization to improve solubility were also tested. Furthermore hydrocarbon cycles and heterocycles alkynes were clicked and analyzed.

One of the two conditions below were used to investigate different alkynes attached to

2’N₃dUTP extended RNA. The GUUCAG oligonucleotide template was reacted at a 1:10 molar

ratio with 2’N3dUTP and 120 U PAP/µg RNA for an incubation period of 2 h at room temperature followed by heat inactivation for 10 min at 62 °C or a 1:2 molar ratio with

2’N3dUTP and 80 U PAP/µg RNA for an incubation period of 3 h at 28 °C with heat inactivation for 10 min at 62 °C. Reaction aliquots were then reacted with 20 molar equivalence of CuSO₄, 50 equivalence of THPTA, 100 equivalence of sodium ascorbate, and 50 or 100 equivalence of alkyne relative to RNA. Click reactions were incubated for either 2.5 or 4 hours at

37 °C.

All click reactions were successful, as the expected product mass was detected by MS

(Table 3.1). The click reaction for some alkynes resulted in an unanticipated +14 Da product.

(Figure 3.1 and Table 3.1) In some cases, the +14 Da product is more prominent than the expected product. The MS/MS of the +14 product revealed that the mass increase was in the 3’- terminus because the c ions stayed the same but 14 Da was added to all y ions. This +14 product was present when the heptyne, pentyne, hexynoic aicd, and cylcohexyl alkynes were used in the click reaction. The +14 Da product could be an oxidation or nitration product occurring at the triazole ring. It is also possible that the addition is occurring on the unbranched carbon chain. To determine the structure of the +14 Da product, 2’N₃dUTP could be reacted with heptyne and a direction infusion MS experiment could be performed. Assuming the product is observed it could be targeted for MS/MS or MS/MS/MS analysis. The fragmentation pattern could provide

Figure 3.1 Extracted ion chromatogram of expected product and product +14 Da. For the heptyne alkyne reaction for 4 h at 37 C with 50 equivalence of alkyne, the predominant product based on peak area is the +14 Da product. The top panel shows the expected product while the bottom panel shows the + 14 Da product. In this reaction based on peak are the +14 Da in the major product.

Figure 3.2 MS/MS of PropA click product with unexpected CID fragments

Table 3.1 Alkyne reaction variations

Alkyne m/z +14 product CID not matching traditional c and

observed y ion m/z

1149.635 X

1135.635 X

1153.635

1155.635 X X

1136.635 X

1157.635 X

information regarding the structure. If needed NMR could be used on product once the reaction was performed on a larger scale.

An unexpected fragmentation pattern was discovered for the propiolic acid click product.

(Figure 3.2) All expected c ions were present implying that the product was there, but none of the expected y ions were present. The y ions all showed a loss of -45 which is consistent with the loss of the terminal carboxylic acid. Another alkyne that showed unexpected fragmentation was cyclohexyl. However the addition of two carbons between the triazole ring and the carboxylic

acid dramatically decreased this fragmentation effect. This was observed when analyzing the

MS/MS of hexyoinc acid verses proponic acid. The hexyoinic acid click reaction showed the +14 peak, but the fragmentation from the expected mass was as predicted. Additionally -45 y ions were at or below the limit of detection unlike the proponic acid. Its MS/MS showed the - 45 from all the detected y ions. The information about fragmentation could be beneficial moving forward if a future direction was to create a tag capable of cleaving during fragmentation to release a reporter ion

The knowledge of the +14 product and the unexpected fragmentation pattern is important as I am proposing isotopically labeled alkynes to be custom synthesized. The study revealed that long straight carbon chains are likely to show an extra +14 product in addition to the expected product. This is undesirable as it adds more products to an already complex spectra, which could overshadow other less abundant labeled oligonucleotides that elute near the same time. This data also reveals that carboxylic acid groups directly adjacent to triazole ring are susceptible to cleavage during MS/MS fragmentation. This results in fragment ions not easily predicted. The unexpected fragmentation pattern could dramatically increase the time and ease of data analysis.

Based on data from these alkynes, an alternate tag strategy could utilize isotopically labeled amino acid based label that can have an alkyne added to either the C or N terminus. This straightforward synthesis of tags can enable movement from duplexing to multiplexing. Several forms of isotopically labeled amino acids are readily available from Cambridge isotopes. One amino acid that has multiple labeled states that vary in mass enough to avoid overlap is phenylalanine. Through amide bond formation a terminal alkyne can be added to either the N or

C terminus of the isotopically labeled phenylalanine. (Figure 3.3b) Based on previously published synthesis, I propose that propargly amine undergo N,N’-dicycloxylcarbodiimide

(DCC) coupling to phenyalanine (of different isotopically labeled states) to create synthetic multiplexing tags. (Pagoti, Dutta et al. 2013) This would be the next natural step in improving the enzymatic and chemical labeling procedure already described. Additionally an alternative to the peptide coupling to propargly amine would be the coupling of phenylalanine to propiolic acid. (Figure 3.3a)

Figure 3.3 Possibility for new isotopically labeled tags

3.4 T1 removal or inactivation

To expand multiplexing to biological samples, the RNA must first undergo digestion by

RNases with BAP to leave shorter oligonucleotides with a 3’ hydroxyl group. RNase T1 cleaves after a specific nucleobase (G) while BAP removes the remaining 3’ phosphate. BAP is easily inactivated with incubation at room temperature for several hours; however the digestive enzyme

RNase T1 is more robust. It recovers from heat inactivation and remains active through dramatic pH shifts. RNases need to be removed or inactivated prior to the PAP reaction to successfully label oligonucleotides, otherwise once PAP adds on 2’N₃dUTP any remaining active RNase would clip off the 2’N₃dUTP leaving a 3’ phosphate. This competitive reaction was observed for an RNase T1 digestion of the 22mer oligonucleotide G CAAG CUG ACCCUG AAG 58

UUCAU. In the initial sample, the expected digestion products were present. In the PAP reacted sample, new peaks were observed but those peaks were not 2’N₃dUTP extended. Those peaks DDA, TRAP 20-40, Source 2.7, 30, 10 werenormal_PAPrxn_digest_82516 digestion products with 3’ phosphate implying that active RNase removed the 2’N1: TOF₃dUTP MS ES- 14.0614.60 100 15.22 1081.14+720.424 20.00PPM 11.97 24.62 123 (Table 3.2). 13.55 23.25 15.30 4.81 5.36 16.57 % 22.87 0.68 6.57 11.25 18.16 19.89 24.32 8.51 25.19 26.55 2.59 4.64 7.14 10.13 20.94 26.92 0 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 normal_PAPrxn_digest_82516RNAse T1 1: TOF MS ES- 6.24 100 1832.283+915.637+610.089 20.00PPM digestion 7.60e4

ACCCUG %

0 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 normal_PAPrxn_digest_82516 1: TOF MS ES- 17.17 100PAP reaction 1912.249+955.62+636.744 20.00PPM 17.25 7.55e4

ACCCUGp %

0 2.00 4.00 6.00 8.00 10.00 12.00 14.00 16.00 18.00 20.00 22.00 24.00 26.00 FigurePAP_22_digest82516 3.4 Post digestion PAP reaction product 1: TOF MS ES- 6.08 100 TIC 3.34e6 Table 3.2 Summary of PAP reaction products 1.44 % 2.24 4.22 4.44 Digest PAP Reaction 19.12 20.48 0 2.00 4.003’ OH6.00 8.003’ p 10.00 12.003’ OH14.00 16.00 3’18.00 p 20.00 2’N3dUTP22.00 24.00 26.00 AGGnormal_PAPrxn_digest_82516 X ND X X ND 1: TOF MS ES- 0.57 100 TIC CAAG1.64 X ND X X ND 2.44e6 ACCCUG X 6.24 ND X X ND % 2.17 17.17 4.30 4.68 7.31 8.24 12.16 0 Time Purification2.00 4.00 methods6.00 8.00 are needed10.00 12.00to remove14.00 or16.00 inactivate18.00 the20.00 RNases22.00 without24.00 interfering26.00 with the subsequent PAP reaction. Purification methods were tested using tRNA phenyalanine from yeast as a model because of its simplicity and well characterized post translational modifications. Digestion product size and modification status is important to ensure the

purification or inactivation methods do not result in complete loss of a given fragment or remove modifications. tRNA Phe was digested with RNase T1 at a concentration of 28U of RNase T1/ug of RNA and BAP (0.01 U/ug of RNA). Digestions were incubated at 37 °C for 2 h. Samples were purified using a number of different methods. Purification and inactivation techniques were tested by room temperature incubation for 1-2 h with a short oligonucleotide (GUUCAU or

GUUCAA) indicator. If RNase T1 was still active post-purification, the oligonucleotide would be cleaved leaving the fragment UUCAU or UUCAA, respectively. Samples were analyzed using MS.

A variety of purification methods were attempted to remove or inactivate RNases. One removal strategy with a lot of effort devoted to it was molecular weight cut off filters. They were used to isolate RNase T1 inside the filter and RNA digestion products were to flow through the filter to the collection tube. Attempts were made to degrade the filter to allow more RNA to exit the filter. Filters were treated with high heat for several days or organic solvents.

Other strategies attempted included liquid-liquid extractions. Digestion reactions were purified through Tri Reagent extractions. Phenol chloroform reactions were also analyzed.

Purifications were attempted with C₁₈ spin column chromatography. RNase T1 would wash through the column while RNA would weakly bind to the column until eluted with some organic solvent. Inactivation of RNase T1 was attempted through proteinase K enzyme treatment to cleave RNase T1. Results of purification and inactivation methods were categorized into the following groups: significant loss of RNA fragments, failed RNase T1 removal, failure to allow

PAP reaction to proceed, and successful. (Table 3.2) Using 3kD MWCO filters, TriReagent

Table 3.3 Summary of purification results

Significant loss of Failed RNase T1 Failure to allow PAP Success

RNA fragments, removal reaction to proceed

-3kD MWCO (even if -5kD MWCO -Phenol/Chloroform -Phenol/Chloroform chemical or heat -10kD MWCO Extraction with ether wash degraded) -Acetone -Proteinase K

-Tri Reagent precipitation digestion

-Size exclusion -C18 spin columns column chromatography

purification and size exclusion column chromatography much of the sample was lost. Many

RNA digest products fell below the limits of detection. This means over a >70% loss of RNA.

Loss is expected in any purification technique, but this much is impractical for biological samples.

Other techniques allow RNase T1 to remain active and present in the sample as evident by the clipped oligonucleotide indicator fragment. C₁₈ spin columns allowed some RNase T1 to stayed weakly bound to the column, which then eluted with the RNA. The 5 or 10 kD MWCO filters with their larger size allowed RNA digestion products to pass through, but they also allowed RNase T1 to escape the filter. Acetone precipitation also allowed the RNA and RNAse

T1 to co-precipitate and thus remain in the sample.

Another category of problems encountered when trying to purify RNA digestion products from active RNase T1 was the inactivation of PAP enzyme in the next step. The indicator RNA remained unclipped, but in the next step no clipped or extended products were detected.

Proteinase K was used to successfully cleave RNase T1, but even after a heat inactivation

Proteinase K was not inactivated and thus cleaved PAP resulting in digestion fragments remaining unchanged. Phenol/Chloroform extractions removed the RNase T1 from the RNA as proven by the full indicator fragment, but in the next step no clipped or extended products were detected. Most likely, the left over phenol lead to the inactivation of the PAP enzyme.

Recently phenol/chloroform extractions with an ether wash were successful. This method allowed for PAP extension. RNase T1 was removed as confirmed by full GUUCAU and PAP remained active. (Figure 3.5) A phenol/chloroform with ether wash purified sample was able to be treated with PAP and clicked with phenylacetyelene. (Figure 3.6) All products were confirmed with CID fragmentation. The only drawback is the variability introduced by the extraction. Purification should be reproducible but due to the labor intensity of this method variability could be introduced. Two samples would go through the same extraction process but a drop more of one sample left behind in comparison to the other would alter relative quantification. I am proposing this variability be corrected for by the addition of a known concentration of a single synthetic oligonucleotide. A poly dT of a known length could be used as it would not be present as part of the sample. Although phenol/chloroform/ether was is not the ideal purification technique it is successful. With the proposed dT internal standard, its major drawback can be overcome and corrected for.

Figure 3.5 Indicator fragment GUUCAU incubation post phenol chloroform extraction followed by ether wash.

Figure 3.6 Extracted ion chromatogram for extension and click product post phenol chloroform with ether extraction. Top spectrum shows the extension product for the oligonucleotide AAUUCG while the bottom shows the product of the click reaction with phenylacetylne D0 for the same oligonucleotide

A less labor intensive purification strategy includes Proteinase K. I propose purification attempts with Proteinase K paired with another purification technique like 10kD MWCO filters.

An alternative strategy could be using Proteinase K immobilized on agarose beads. They would inactivate RNase T1 and the beads could easily be filtered out and washed to free the RNA digestion products. This would avoid the need for Proteinase K inactivation.

3.5 PAP extension of oligonucleotide

To expand this labeling strategy, enzymes other than RNase T1 could be used to digest the

RNA sample. Using other enzymes could help improve the sequence coverage and the analysis of longer digestion products. When RNase T1 cleaves G rich sequences, it can result in short

RNA fragments. Digestion fragments of less than 3 nucleotides in length are difficult to detect.

They are also difficult to unambiguously map back onto a tRNA sequence as the sequence may appear multiple times or on multiple tRNAs. If non-G specific enzymes were used to cleave the same sequence that same segment of RNA could result in larger digestion products more specific to a particular tRNA. If RNA was cleaved by cusativin, U2, or MC1 it would result in digestion products with a different 3’ nucleobase identity. (Addepalli, Lesner et al. 2015, Houser, Butterer et al. 2015, Addepalli, Venus et al. 2017) If cleaved by cusativin then the resulting nucleotide would end in C, for U2 C or U, for MC1 any base as it cleave 5’ of U. Before a PAP extension is attempted with any non RNase T1 digestion products, the effect of the identity of the 3’ end on

PAP extension would need to be investigated.

The template GUUCA[N], where N = C, U, A, G was reacted in a 1:10 ratio of RNA to

2’N3dUTP with 425 U PAP/μg RNA and incubated 1.5 h at 30 °C and heat inactivated was

completed for 10 min at 62 °C. Post reaction, samples were purified through an Illustra Spin

Column (GE Healthcare), dried, and resuspended in mobile phase A (mobile phase A matching that described in Chapter 2) to be ran on an LC-LTQ with an Xbridge-MS C18 1.0x150 mm,

3.5um particle size and 150A pore size column. Percent extension was based on peak area of product over product plus unreacted oligonucleotide.

Table 3.4 3’ Extension with 2’-N3dUTP Template Extension (%) GUUCAC 99 GUUCAA 99 GUUCAG 99 GUUCAU 40

With the exception of U, all templates showed the same level of extension when high

PAP concentrations are used. (Table 3.4) This is not a big problem as currently there is no base specific RNase enzyme that leaves a 3’ U. RNase A cleaves 3’ to C or U, but it is not commonly used due to the resulting short fragments.

The future work I am proposing here would be to react digestion products from Cusativin and MC1 with PAP and 2’-N₃dUTP to verify compatibility. I also think it would be worthwhile to examine the effect of a 3’ modified G on the ability of PAP to add on 2’-N3dUTP. If two oligonucleotides both ended in the same G modification, it would not affect the method as the effect would be the same. It would however affect the method if the 3’ was changing in percent methylation. This could be tested with synthetic oligonucleotides with the same sequence with the exception of the 3’ end. I propose an experiment where one oligonucleotide would end in G the other would end in m2G, 8oxG or another methylated G other than Gm. Gm would not be present at the 3’ end of an oligonucleotide as RNase T1 does not cleave there. Another variable

that needs to be tested to use this platform to be useful on biological sample would be to optimize the concentration of PAP per μg of tRNA. This could be conducted with commercially available total tRNA isolated from E. coli.

3.6 T4 ligation

Ligation is a powerful tool to create designed oligonucleotides. Ligations are common place in molecular biology for insertion into a plasmid, for DNA sequence amplification, or recombinant protein expression. Ligation can be used to create chimeric type oligonucleotide therapeutic. Ligation labeling can useful by moving labeling to the 5’ end of RNA to avoid issues with removing or selectively inactivating RNase T1 related to 3’ end addition cleavage.

Another graduate student’s previous work in the lab on ligation was focused on adding NAD+ to the 3’OH of digestion products. This ligation idea can easily be adapted to labeling the 5’ end of the RNA. Instead of the 3’ NAD+, a short synthetic oligonucleotide could be used to label. The digestion product in its 3’ phosphate state is be treated with T4 PNK to add a 5’ phosphate to prepare the fragment for ligation. The synthetic oligonucleotide with its 3’OH can be ligated onto the 5’ end of the digestion product. This strategy will prevent self-ligation of the digestion product. T4 Polynucleotide Kinase (3’ phosphate minus) must be used to ensure that the additional phosphate is from ATP and not from the 3’ end of oligonucleotide. Once the reaction is complete, the product will have 5’ OH and 3’ p therefore avoiding the potential issue of double ligation. Then T4 ligase 1 (ssRNA ligase) can be used to add on a labeling RNA to the 5’ end.

Figure 3.7 Ligation reaction plan

A synthetic oligonucleotide with the sequence CAUGUGG was digested with RNase T1 to produce CAUGp. Varying amounts of T4 Polynucleotide Kinase (PNK) (3’ phosphatase minus) (New England BioLabs) were incubated at 37 °C for various times followed by a heat inactivation for 20 min at 65 °C to produce pCAUGp. A T4 ligation was setup with varying ratios of ligate (CCUCCUUUCU), varying units of T4 RNA ligase 1 (Thermo Scientific), and for varying times at 37 °C followed by 15 min heat inactivation at 62 °C. Samples were analyzed using a Waters G2 Synapt mass spectrometer with conditions matching those from

Chapter 2.

The percent ligated is based on peak area. It is calculated by the expected product over the expected product plus the unreacted oligonucleotide. Because analysis was based on percentages relative to one another it reveals which conditions works best; however it is not truly the percent reaction completion as the ionization efficiency for all products is not the same.

Table 3.5 T4 PNK reaction results Time 40min 20min 20min 1h 20min 1h 20min Enzyme 2.5 2.5 7.5 1 2 % P added 100 99.2 99.6 98.1 99.7

Table 3.6 Ligation reaction results. Percent ligation was calculated based on peak area with ligated/ (ligated+unligated) *100% Time 3h 3h 3h 3h 3h 3h 2h 2h Ratio to 1 to 1 1 to 1 1 to 1 1 to 2 1 to 2 1 to 5 1 to 1 1 to 2 RNA to ligate T4 10U 5U 1U 5U 2U 5U 5U 5U % ligated 1 0.4 0.3 1.7 0.7 12.8 1 1.5

For PNK reactions all conditions worked well with minimal unreacted oligonucleotide detected. With enzyme concentrations above 2.5 U/ 100 pmol RNA time does not seem to be a factor as long as it is greater than 20 minutes. To conserve enzyme, reactions were incubated with 2 U/ 100pmol RNA for 90 minutes. (Table 3.5) This produced the same result as the 2.5U/

100pmol RNA for greater than 20 minutes.

Although attempts were made, a single nucleotide could not be successfully ligated onto the

5’ of the digestion products. At this point a non-G containing oligonucleotide was ligated onto the 5’ end of a digestion product. Ligation reactions show low percent of ligation product relative to starting oligonucleotide. (Table 3.6) The best results were achieved when the ratio of digestion

product to ligate was increased. For enzyme concentration, with a 1 to 1 ratio, the enzyme concentration increase did not result in a dramatic increase in ligation product. When the 1 to 2 ratio was constant but the enzyme concentration was increased, the ligation product increased.

Higher enzyme concentration seems to improve ligation product yields, but a minimum ratio of 1 to 3 (RNA to labeling oligonucleotide) is required for significant ligation product to be observed.

Although this requires excessive amounts of labeling oligonucleotide, it is critical for the ligation to occur.

In regards to future work, the initial focus should be RNase T1 removal and inactivation.

Although the phenol/chloroform/ether extraction shows promise, the method of purification used must be reproducible. The next challenge to overcome should be developing better isotopic alkynes for labeling. This would enable labeling to move past duplexing and onto multiplexing.

PAP extension of non-G ending oligonucleotides and ligase labeling should be saved as strategies to be continued only if other areas fail.

4 CHAPTER 4 INTRODUCTION

4.1 Research goal

The goal of part two of this dissertation is to advance the field of medicinal chemistry by creating base modified nucleosides compounds that have anti-cancer properties. In this chapter, previous nucleoside mimic therapeutics are highlighted. Their structure and mechanism of action are considered when looking for a starting point to create a more selective potential cancer therapeutic. Most of this chapter can be found in a previously published review article. (Burke,

Borland et al. 2016) In Chapter 5, compounds are synthesized based off of previous compounds exhibiting polymerase chain terminating properties for sequencing. Structure activity relationship studies were conducted until a lead compound immerged. Mechanisms of action studies are conducted and a new potential anti-cancer agent resulted. This work is done to further the field of medicinal chemistry. It is completed in part to improve the selective toxicity of therapeutic towards cancerous cells instead of all rapidly dividing cells. This work bridges the fields of synthetic chemistry and biochemistry to produce a lead compound.

4.2 Introduction

Over half a century, nucleoside analogs, otherwise termed antimetabolites, have substantially impacted treatment of cancer, bacterial, and viral infectious diseases. (Ritter,

Jedlltschky et al. 2005, Tiwari 2012, Capasso and Supuran 2014) Their structural similarity to physiological nucleosides allows their passage into cells by nucleoside transporters, where they are metabolized e.g. into 5'-triphosphates (other active metabolites are described), the active species capable of interfering with a large variety of intracellular targets. (Eriksson 2002, Zhang,

Visser et al. 2007) As antineoplastic agents and antibiotics, antimetabolites interfere with key intracellular processes by inhibiting enzymes involved in the synthesis of nucleic acids and nucleotides, signal DNA damage upon their incorporation, obstruct DNA repair, and trigger apoptosis by directly affecting mitochondria (Hill and Bennett 1969, Fairchild, Maybaum et al.

1986, Parker, Shaddix et al. 1991, Kuchta, Ilsley et al. 1992, Peters, van der Wilt et al. 1994,

Swann, Waters et al. 1996, Genini, Adachi et al. 2000, Ji, Rha et al. 2005, Chen, Plunkett et al.

2008). The antiviral action of antimetabolites is connected to inhibition of viral DNA replication in doses that are not toxic to the host cells. (Hostetler, Stuhmiller et al. 1990, Clouser, Holtz et al.

2012)

All clinically approved antimetabolite drugs elicit adverse effects to various degrees as they also affect rapidly proliferating normal human cells, lymphocytes and sometimes even non- dividing cells, such as neurons which substantially narrows their therapeutic windows. (Wang,

Tzeng et al. 1997, Besirli, Deckwerth et al. 2003, Koros and Kitraki 2009, Krishna, Vanaja et al.

2009, Cordier, Nau et al. 2011, Lolkema, Arkenau et al. 2011, Cannas, Pautas et al. 2012)

Additionally, their efficiency is somewhat limited to a relatively short list of malignancies that are predominantly hematological, although 5-fluorouracil and gemcitabine have proven to be effective against several solid tumors. (Hertel, Boder et al. 1990, Lennard 1992, Peters, van der

Wilt et al. 1994, Gandhi, Estey et al. 1996, Ando, Watanabe et al. 1998, Bocci, Fioravanti et al.

2005, Tallman 2006) Consequently, there is a critical need to discover novel anti-cancer chemotherapeutics with higher selectivity towards cancer cells. This chapter will be focused on the development of base-modified nucleoside antimetabolites.

4.3 Current nucleobase-modified antimetabolite chemotherapeutic drugs

4.3.1 Modified nucleobase drugs

Minor modifications of both pyrimidine and purine nucleobases (Figure 4.1) are well tolerated in terms of enzymatic recognition. Thus, modified nucleobases, such as 5-fluorouracil

(5FU) 6-mercaptopurine (6-MP) and 6-thioguanine (6TG) undergo intracellular metabolism into the corresponding nucleotides. (Chaudhuri, Montag et al. 1958, Moore and LePage 1958,

Sartorelli and LePage 1958, Sartorelli, LePage et al. 1958, Mukherjee and Heidelberger 1962,

Reyes and Hall 1969, Higuchi, Nakamura et al. 1976, Liliemark, Pettersson et al. 1990)

The primary mode of action of 5FU, a chemotherapeutic agent used against breast and colorectal cancer, is inhibition of thymidilate synthase by one of its metabolites, 5-fluoro-2’- deoxyuridine-5’-monophosphate (FdUMP), which results in depletion of thymidine-5’- triphosphate (TTP), ultimately leading to a condition termed thymineless cell death. (Cohen

1971, Santi and McHenry 1972, Francini, Petrioli et al. 1993, Sobrero, Aschele et al. 1997) The other nucleotide metabolite of 5FU, 5-fluoro-2’-deoxyuridine-5’-triphosphate (FdUTP), is capable of misincorporating into DNA in place of TTP, which results in DNA damage, and its ribose analog, FUTP, incorporates into RNA, which is suggested to cause gastrointestinal toxicity. (Houghton, Houghton et al. 1979, Canman, Tang et al. 1992, Parsels, Parsels et al.

1998) The origin of neurotoxicity of 5FU has not been fully explored, but based on structure many researchers strongly suspect interaction with intracellular targets other than replicating

DNA and enzymes involved in this process. (Cordier, Nau et al. 2011)

Figure 4.1 Unnatural nucleobase drugs: 5-fluorouracil (5FU), 6-mercaptopurine (6MP), and 6-thioguanine (6TG)

The main mode of action of 6TG and 6MP, common chemotherapeutic agents used against leukemia, is inhibition of purine synthesis by their ribonucleotide 5’-monophosphates, although their 5'-triphosphates, 6TGTP and 6MPTP, have recently been demonstrated to block activation of the small GTPase Rac1 by competition with GTP, which leads to apoptosis, and also to block DNA synthesis. (Sartorelli and LePage 1958, McCollister, Gilbert et al. 1964,

Bourgine, Garat et al. 2011, Marinković, Kroon et al. 2014, Munshi, Lubin et al. 2014) Their most dangerous toxicity is connected to the development of myelosuppression as well as liver damage. (Yates, Krynetski et al. 1997, Dubinsky, Vasiliauskas et al. 2003)

4.3.2 Base-modified natural sugar nucleoside drugs

Certain modified nucleobases do not exhibit activity on their own, but their ribo- or 2’- dideoxyribonucleosides are used as chemotherapeutic agents (Figure 4.2). 2-

Chlorodeoxyadenosine (2CDA) otherwise known as Cladribine, the drug that exhibits specific toxicity toward proliferating human lymphocytes is currently used against hairy cell leukemia and was in clinical trials for treatment of multiple sclerosis, although has not been approved by the FDA in the USA for treatment of that disease due to insufficient data on the benefit:risk ratio.

(Carson, Wasson et al. 1980, Carson, Wasson et al. 1983, von Rohr, Schmitz et al. 2002,

Giovannoni, Comi et al. 2010) Even though the exact mechanism of action of 2CDA remains unclear, the most likely effect is presumed to be the incorporation of the 5’-triphosphate of

2CDA into DNA, which leads to DNA strand breaks and eventual apoptosis. (Robak, Korycka et al. 2005, Robak, Lech-Maranda et al. 2006, Johnston 2011) The most serious side effects include infection, which is responsible for 3% treatment related mortality. (Juliusson, Lenkei et al. 1995) The 2-fluoro analog of 2CDA also exhibited activity, but was not as selective. (Carson,

Wasson et al. 1980, Parsons, Bowman et al. 1986) 5-Azacytidine (5AZC) and 5-aza-2’- deoxycytidine (5AZCdR) , otherwise known as Vidaza and Decitabine, respectively, are potent drugs against myelodysplastic syndromes and acute myeloid leukemia. (Sorm and Vesely 1964,

Doskočil and ŠOrm 1970, Kantarjian, Issa et al. 2006, Momparler, Cote et al. 2014) In addition,

5-AZC is active against human immunodeficiency HIV-1 and adult T-cell lymphoma HTLV-1 viruses. (Dapp, Clouser et al. 2009, Diamantopoulos, Michael et al. 2012) These cytidine nucleoside analogs are metabolized into the 5’-triphosphates followed by their incorporation into nucleic acids in the place of their natural counterparts, resulting in covalent binding and inactivation of methyltransferases, which abrogates DNA synthesis and leads to cell death

(Doskočil and ŠOrm 1969, Bouchard and Momparler 1983, Lu and Randerath 1984, Pinto, Maio et al. 1984, Huschtscha, Bartier et al. 1995, Stresemann and Lyko 2008). Incorporation of 5-

AZC into RNA causes defective methylation of tRNA thus impairing ribosomal protein synthesis. (Harris and Randerath 1978, Skripal', Babichev et al. 1993, Kuo, Krasich et al. 2010)

Serious side effects include hematopoietic toxicity and myelosuppression (Momparler, Bouffard et al. 1997, Arce, Segura-Pacheco et al. 2006).

Figure 4.2 Natural sugar base-modified nucleoside drugs: 2-chloro-2'-deoxyadenosine (2CDA), 5-azacytidine (5-AZC) and 5-aza-2'-deoxycytidine (5-AZCdR)

With regard to the nucleobases corresponding to 2CDA and 5AZC, 5-azauracil does not undergo intracellular metabolism to form either 5AZC or 5AZCdR, as do 5FU, 6TG, and 6MP.

Although the conversion of 2-chloroadenine into 2CDA and subsequently to its 5’-triphosphate has been reported, the activity of the nucleobase is significantly lower than that of the nucleoside. (Bontemps, Delacauw et al. 2000, Vande Voorde, Liekens et al. 2013). In fact, the nucleoside 2CDA could lose efficacy due to depurination caused by infection of cancer cells with mycoplasma species that express certain nucleoside phosphorylases. (Vande Voorde,

Liekens et al. 2013) Catabolism of 5AZC was also reported in connection with its activity decrease. (Veselý, Seifert et al. 1966)

4.3.3 Base-modified unnatural sugar nucleoside drugs

Some modified nucleobases have to be attached to a modified ribose fragment to exhibit efficiency. Examples include Emtricitabine (FTC) , Abacavir (ABC) , Sorivudine (BVaraU) ,

Clofarabine (CAFdA) , Fludarabine (2FaraA) , and Nelarabine (araG). (Dow, Bell et al. 1980,

Machida, Sakata et al. 1981, Averett, Koszalka et al. 1991, Parker, Shaddix et al. 1991, Schinazi,

Boudinot et al. 1992, Crimmins and King 1996) (Figure 4.3)

Figure 4.3 Sugar and base modified nucleside drugs: Emtricitabine (FTC), Abacavir (ABC), Sorivudine (BV-ara-U), Clofarabine (CAFdA), Fludarabine (2FaraA), and Nelarabine (araG)

The first three drugs are used against viral infections. Emtricitabine (FTC) is used in the treatment of HIV and is also active against hepatitis B (HBV), although not approved by FDA for its treatment. (Doong, Tsai et al. 1991, Herzmann, Arastèh et al. 2005) FTC is metabolized into its 5’-triphosphate that inhibits viral reverse transcriptase (Wilson, Martin et al. 1993).

Severe side effects are rare; sometimes, hepatoxicity and lactic acidosis may take place.

(Dragovic and Jevtovic 2012, Macías, Neukam et al. 2012) Abacavir (ABC) is also used against

HIV and works in the same way as Emtricitabine. (Fleury, De Boer et al. 1998, Hughes,

McDowell et al. 1999, Kumar, Sweet et al. 1999, Miller, Ait-Khaled et al. 2000) The side effects of ABC are usually mild, the most serious incidents include hyperlactatemia and lactic acidosis.

(Dragovic and Jevtovic 2012) Sorivudine (BVaraU) is used against the herpes virus family, particularly herpes simplex (HSV-1), varicella zoster (VZV), and Epstein-Barr (EBV) viruses.

(Alrabiah and Sacks 1996, Snoeck, Andrei et al. 1999, Saijo, Suzutani et al. 2008) The 5’- triphosphate of BVaraU is a competitive inhibitor of the viral DNA polymerase. (Yokota, Konno et al. 1989) The drug does not elicit serious side effects. (Hiraoka, Masaoka et al. 1991)

The other three nucleotide analogs are used against hematological malignancies.

Clofarabine (CAFdA) is a close analog of previously described 2-chloro-2’-deoxyadenosine, with the 2'-arabino-2’-fluoro sugar instead of natural 2’-deoxyribose. This change makes CAFdA more active than 2CDA, presumably due to improved acid and metabolic stability. (Carson,

Wasson et al. 1992, Lotfi, Månsson et al. 1999) Clofarabine is approved for the use against acute lymphoblastic leukemia. (Cooper, Kantarjian et al. 2004, Jeha, Gandhi et al. 2004) Its main mechanisms of action are the inhibition of ribonucleotide reductase and DNA polymerases by its

5'-triphosphate, the latter also incorporates into DNA incurring its damage, eventually resulting in apoptosis. (Parker, Shaddix et al. 1991, Xie and Plunkett 1996, Genini, Adachi et al. 2000,

Takahashi, Shimizu et al. 2002) Serious side effects include tumor lysis syndrome, bacterial infections, and renal insufficiency. (Nabhan, Davis et al. 2011, Simko, Tran et al. 2014)

Fludarabine (2FaraA) differs from Clofarabine by substituents in the 2-position of the nucleobase

(fluorine instead of chlorine) and in the 2’-position of the sugar (hydroxyl group instead of

fluorine), but its principle mode of action is the same as that of CAFdA. (Brockman, Cheng et al. 1980, Sato, Montgomery et al. 1984, Spriggs, Robbins et al. 1986, Huang and Plunkett 1991,

Huang and Plunkett 1992, Huang and Plunkett 1995) It is used against chronic lymphocytic leukemia. (Johnson, Richardson et al. 1993) The most serious side effects are severe autoimmune hemolytic anemia and pulmonary infections. (Eftekhari, Lassoued et al. 1998,

Gamberale, Fernandez-Calotti et al. 2006) Nelarabine (araG) is the 6-O-methylguanine analog of 2FaraA, initially developed against varicella-zoster virus (VZV), with anti-malignant activity later revealed. (Averett, Koszalka et al. 1991, Lambe, Averett et al. 1995) The mode of action of ara-G is the same that of the other two pyrimidine analogs, and likewise, it is used against T-cell acute lymphoblastic leukemia. (Rodriguez, Legha et al. 1997, Hochster, Oken et al. 2000, Cohen,

Johnson et al. 2006) This drug, however, exhibits serious side effects, such as neurologic toxicity and increased susceptibility to infections (Testi, Moleti et al. 1997, DeAngelo 2009).

4.3.4 Development of novel nucleobase-modified antimetabolite drug candidates

As can be seen, the active species of the existing nucleoside antimetabolites are their nucleotides, most commonly 5’-triphosphates whose direct use is hampered by the low cellular membrane permeability due to the negative charge and their enzymatic dephosphorylation in vivo. (Wagner, Iyer et al. 2000, Jordheim, Cros et al. 2006) Despite the recent advances in the development of nanogel delivery systems suitable for intracellular delivery of cytotoxic nucleoside 5’-triphosphate analogs into cancer cells, and silica nanoparticles that effected intracellular delivery of fluorescent labeled nucleotides, this approach to utilize the active nucleotide drugs is still in early stages of development. (Vinogradov, Zeman et al. 2005,

Vinogradov, Kohli et al. 2006, Kohli, Han et al. 2007, Vasilyeva, Silnikov et al. 2013) Other

approaches to facilitate the formation of active species include the use of phosphoramidates, e.g.

‘ProTide’ technology or monoester prodrugs, the pronucleotides that undergo intracellular enzymatic activation to form 5’-monophosphates, which circumvents the first 5’-phosphorylation of the parent nucleoside, the rate determining step in the activation process. (Chang, Griesgraber et al. 2001, Murakami, Tolstykh et al. 2010, Slusarczyk, Lopez et al. 2014). Similarly, chemically cleavable S-Acyl-2-thioethyl (SATE) or salicyl alcohol (cycloSal) can be used as protecting groups for 5’-monophosphates. (Meier 2002, Peyrottes, Egron et al. 2004) These strategies, too, require further development and optimization. (Peterson and McKenna 2009)

Therefore, the development of novel base-modified nucleoside chemotherapeutic prodrugs remains one of the key directions for medicinal organic chemists. The success is contingent upon passing the following three processes: (a) cellular uptake by nucleoside/nucleobase transporters; (b) intracellular 5’-phosphorylation into the active nucleotide species; and (c) recognition of the produced drug molecules by enzymes involved in DNA replication. (Cabrita, Baldwin et al. 2002, Galmarini, Popowycz et al. 2008, Kose and Schiedel

2009, Lauridsen, Rothnagel et al. 2012, Robak and Robak 2013) It is important that the modifying moieties interfere with these processes as little as possible.

One of the important aspects in the antimetabolite drug design and development is the recognition of the base-modified nucleoside by a polymerase that carries out a template driven

DNA or RNA synthesis. This template-dependent polymerization reaction incurs building of a

DNA strand that is complementary to the template strand, and is effected according to proper

Watson−Crick base pairing rules for complementary nucleobases. Therefore, the overall approach to the attachment of modifying moieties to nucleobases is usually based on the retention of the hydrogen bonding donor-acceptor pattern (Figure 4.4). In general, substitutions

at C5 of pyrimidines and C7 of 7-deazapurines are well tolerated, whereas enzymatic incorporation of C6-modified pyrimidines or C8-modified purines remains challenging, with only a few exceptions. (Akerblom, Pontis et al. 1982, Perrin, Garestier et al. 2001, Jäger,

Rasched et al. 2005, Srivatsan and Tor 2009, Kajiyama, Kuwahara et al. 2011, Litosh, Wu et al.

2011, Holzberger, Strohmeier et al. 2012, Stupi, Li et al. 2012, Goubet, Chardon et al. 2013,

Hollenstein 2013, Mori, Ozasa et al. 2013). While 2-halo-2’-deoxyadenosine-5’-triphosphates have been shown to be good substrates for DNA polymerases, including human, as well as other

DNA processing enzymes, incorporation of dATP analogs with larger substituents at 2-position is difficult. (Parker, Bapat et al. 1988, Hentosh, Koob et al. 1990, Moore, Jalluri et al. 1996,

Moore, Li et al. 1996, Foley, Hentosh et al. 2004) N6-methyl- , N6-benzyl- , and N4-alkylated

2’-deoxycytidine-5’-triphosphates are successfully incorporated into DNA by natural polymerases, but not N2-substituted 2’-deoxyguanosines. (Wu, Stupi et al. 2007, Petruseva,

Tikhanovich et al. 2008, Zhang, Motea et al. 2010, Stupi, Li et al. 2012)

Figure 4.4 Susceptibility to failure in polymerase promoted DNA incorporation of nucleoside 5'-triphosphates depending upon substitution: blue - large groups are tolerated; purple - small groups are tolerated; red - no substitution is tolerated

N6-monoalkylated adenosines and 2’-deoxyadenosines (Figure 4.6) have a strong potential as chemotherapeutic agents. Examples include N6-isopentenyladenosine acting as a modulator of farnesyl pyrophosphate synthase (FPPS) activity, N6-ureidoadenosine inhibiting protein kinases, and N6-hydrazyladenosine presumably inhibiting ribonucleotide reductase.

(Cappellacci, Petrelli et al. 2011, Shelton, Cutler et al. 2012, Scrima, Lauro et al. 2014) 2-

Chloro-3-deaza-N6-cyclooctyladenosine and 2’-deoxyadenosines with diverse substituents attached at N6 show antibacterial and anticancer activities, respectively, although the mechanisms are not clear. (Ottria, Casati et al. 2010, Vitali, Petrelli et al. 2012) 2-(N-aryl)amino-

2’-deoxyadenosines were found to inhibit DNA polymerases in cancer cells, but expectedly, no

DNA incorporation was observed. (Höltje, Richartz et al. 2010, Schwanke, Murruzzu et al. 2010)

Mono-substituted 7-deazapurines, such as 7-cyano-, 7-iodo-, 7-amido adenosines and 7- thioamido-2’-deoxyadenosine exhibit various antiviral and anticancer activity. (Hecht, Frye et al.

1976, Gupta, Daunert et al. 1989, Krawczyk, Renau et al. 1995, Nord, Stolfi et al. 1997, Zhang,

Wisniewski et al. 2015) 7-Aryl substituted 7-deazaadenosine analogs show modest to high activity in promyelocytic leukemia, cervix carcinoma, lymphoblastoid, and hepatocellular carcinoma cell lines, but when the substituent is attached at the 8-position, anticancer activity is lost, which is consistent with the inability of polymerases to incorporate their triphosphates into nucleic acids. (Cottam, Kazimierczuk et al. 1985, Nauš, Pohl et al. 2010) Kinase inhibitors, such as 7-amido-8-bromo- and 7-amido-8-hydrazino-7-deazaadenosines, cause apoptosis in cancer cells, and the latter agent also inhibits HIV-1 and HCV replication. (Nekhai, Bhat et al. 2006,

Radhakrishnan and Gartel 2006, Cho, Lee et al. 2010, Dolloff, Allen et al. 2012) Another important base-modified nucleoside that exhibits antitumor and antiviral activity are 4-

aminopyrrolo[2,3-d]pyridazine adenosine analogs that thwart proliferation of murine leukemic and human cervical cancer cells and to inhibit human cytomegalovirus and herpes simplex viruses. (Meade, Wotring et al. 1992)

Figure 4.5 Chemotherapeutic base-modified purine nucleoside analogs

Figure 4.6 Primer extension assay of base modified triphosphates

In the course of developing terminators for a cyclic reversible termination protocol commonly used in DNA sequencing, it was recently discovered that N6-(2-nitro)benzyl-2’- deoxyadenosine-,(Wu, Stupi et al. 2007) 5-(2-nitro)benzyloxymethylpyrimidine-,(Litosh, Wu et

al. 2011) and 7-deaza-7-benzyloxymethylpurine-2’-deoxy-5’-triphosphates are incorporated into partial double helix DNA primer by natural polymerases more efficiently than the corresponding natural nucleotides, and then terminating further DNA synthesis by obstructing the subsequent nucleotide incorporation. A DNA template containing A in the sequence was reacted with

Vent(exo-) polymerase and synthesized base modified nucleotides. In this primer extension assay, upon incorporation of the base modified compounds extension of the template strand was terminated. It was also evident that termination of DNA synthesis occurs only in the presence of a bulky group, such as a branched alkyl (e.g. isopropyl or tert-butyl) linked to the α-benzylic carbon. The bulkier the modification the shorter the resulting DNA fragment as extension was terminated upon the first incorporation event. Considering the high recognition of the novel thymidine 5’-triphosphate DNA terminators by polymerases, we presumed that they would compete with natural nucleotides for incorporation into DNA within cells.

Figure 4.7 Chemotherapeutic base-modified pyrimidine nucleosides

The initial discovery of 5-(α-substituted)benzyloxymethyl-2’-deoxyuridines whose 5’- triphosphates undergo successful incorporation into DNA and terminate further elongation after

a single incorporation event has allowed for the hypothesis that treatment of cancer cells with the corresponding nucleosides or nucleobases would result in halting of their DNA replication due to incorporation of a DNA synthesis terminator into the DNA replication fork. (Figure 4.6) (Litosh,

Wu et al. 2011) Thymidine nucleoside analogs are successfully converted into their 5’- triphosphates by nucleotide kinases; hence, we hypothesized that base-modified thymidine derivatives would be metabolized within cells obstruct further addition of natural nucleotides.

(Magnani, Casabianca et al. 1996, Olivero, Vazquez et al. 2010) Therefore, these thymidine analogs were synthesized and their activity was examined in breast cancer cells (Figure 4.8).

The initial studies revealed the cytotoxicity of the nucleosides to be in correlated with the ability of their 5’-triphosphate to terminate DNA synthesis. (Burke, Borland et al. 2013, Borland, Burke et al. 2014) The corresponding T nucleobases, however, showed overall weaker activity, and did not follow the same trend as the nucleosides. (Borland, Lawson et al. 2014) The mechanism of action of either species has not been proven yet. Other examples of bioactive T-nucleoside analogs include 5-aryloxy-2’-deoxyuridines that exhibit moderate cytotoxicity in cervical cancer cells, and 5-alkoxymethyluridines and their 5-fluoro-2’-deoxy analogs that exhibit activity in T- lymphoblastic and acute myeloid leukemia, lung adenocarcinoma, and colorectal cancer cell lines, but likewise their mechanism of action was not reported. (Kim, Lee et al. 2005, Brulíková,

Džubák et al. 2011) Consistent with previous observations, the activity of the uracil nucleobases is generally lower than that of their uridine analogs. (Brulíková, Džubák et al. 2011)

N4-alkylated cytidines work as inhibitors of methyltransferase 1 in cervical and kidney cancer cells, and their 5-fluoro-2’-deoxy analogs exhibiting activity in a variety of cancer cells,

but the detailed mechanism of action was not reported.(Ludwig, Schwendener et al. 2005,

Plitta, Adamska et al. 2012)

Figure 4.8 Chemotherapeutic nucleosides with modified nucleobases having altered hydrogen bonding pattern

With regard to the nucleoside transporters-mediated cellular uptake, the recognition can be increased, if necessary, using pro-drug approaches, such as, for instance, functionalization of the 3’ and 5’-hydroxyl with amino acids. (Song, Lorenzi et al. 2005, Molina-Arcas, Moreno-

Bueno et al. 2006) Intracellular phosphorylation of base-modified nucleosides, even those containing very large groups, such as carboranes, attached at the C5 of uridines and 2’- deoxyuridines or at C6 of 7-deaza-2’-deoxyadenosine, is not impaired, let alone minor modifications (e.g. 5-fluoro-U , 5-alkynyl-C , 2-chloro-A , and 6-thio-G. (Azuma, Huang et al.

2001, Ceruti, Mazzola et al. 2003, Pratt, Shepard et al. 2005, Nauš, Pohl et al. 2010, Guan, van der Heijden et al. 2011, Prüfer, Schuchardt et al. 2014)

Sometimes, active species can have modified nucleobases with altered hydrogen bonding pattern (Figure 4.9). Besides araG, the anti-leukemic drug, triciribine, a 7-deaza-6-alkylated

adenosine derivative is capable of inhibiting DNA synthesis, showing activity against breast cancer likely by inhibiting Akt kinase, although its antiviral activity is low. (Schram and

Townsend 1971, Wotring, Townsend et al. 1990, Porcari, Ptak et al. 2000, Wang, Ding et al.

2014) In contrast, bicyclic thymidine derivatives missing 3-H exhibit excellent potency and selectivity in Varicella-Zoster Virus (VZV) inhibition. (McGuigan, Yarnold et al. 1999) Some

N3-alkylated thymidine derivatives also show anti-HIV activity. (Velázquez, Tuñón et al. 1999)

Similar bicyclic C nucleosides are active against other viruses. (Kifli, De Clercq et al. 2004) 1,3-

Imidazoles attached to ribose have even lower resemblance to nucleobases, yet exhibit anticancer activity by inhibiting inosinate dehydrogenase. (Balzarini, Karlsson et al. 1993)

There are several exceptions from the necessity for the nucleobases to be capable of forming Watson−Crick pairing at all, particularly in the case of leukemic cells that express deoxynucleotidyl transferase that performs DNA synthesis without the use of a templating strand. (Bollum 1964) Thus, 5-nitroindolyl- (5NI) and 3-ethynyl-5-nitroindolyl- (3Eth5NI) 2′- deoxynucleosides (Figure 4.9) whose nucleobases are incapable of forming Watson−Crick pairing, show activity against acute lymphoblastic leukemia where deoxynucleotidyl transferase is implicated. (Reineks and Berdis 2003, Motea, Lee et al. 2012) Incorporation of the 5’- triphosphates of these nucleosides, particularly of 3Eth5NI, results in halting of the DNA synthesis, which seems to be the primary mechanism of action. (Motea, Lee et al. 2012) These compounds do not interfere with DNA synthesis in normal cells, which is template driven, so they are very selective, but the scope is limited to only one type of leukemia. Other similar 2’- deoxyriboses with arbitrary aryl moiety attached at the 1’-position exhibit moderate antibacterial activity, but neither the mechanism of action nor the toxicity toward normal human cells has been reported. (Hatano, Nishimura et al. 2009)

Figure 4.9 1'-Non-nucleobase-2'-deoxyribonucleoside analogs: 5-nitroindolyl- (5NI), 3-ethynyl-5- nitroindolyl- (3Eth5NI), indolyl- (Ind), naphthyl- (Naphth), and 3,4-difluorophenyl- (34diFPh) 2'- deoxynucleosides

4.4 Conclusions and outlook

Overall, modification of nucleobases has provided several potent anti-tumor and anti- microbal chemotherapeutic drugs and a greater number of drug candidates. However, the high activity exhibited due to close similarity to natural nucleosides, comes with off-target-mediated adverse effects, especially severe in case of cancer chemotherapeutic agents. Therefore, there is a critical need to explore the therapeutic potential of alternative anti-cancer agents, with higher efficiency and lower incidence of the side effects.

5 CHAPTER 5 BASE-MODIFIED THYMIDINES CAPABLE OF TERMINATING DNA SYNTHESIS AS NOVEL DRUG CANDIDATES SHOWING ACTIVITY IN CANCER CELLS

5.1 Background

Cancer is one of the most sinister diseases known to mankind claiming over half a million lives in the US alone in 2014 with more than one and a half million new diagnoses. (Siegel, Ma et al. 2014) In contrast to normal cells, cancer cells undergo rapid, abnormal, and uncontrolled division, resulting in a constant requirement for DNA production. Therefore, tampering with this process preferentially affects them and represents a plausible approach to cancer chemotherapy, a major component of cancer treatment, particularly if the tumor is inoperable or has metastasized. (Calvo 2008) Current FDA-approved anti-cancer nucleosides elicit severe side effects that warrant their improvement; therefore, we designed compounds with a mechanism of action focused on inhibiting DNA replication rather than targeting multiple pathways. We previously discovered that 5-(α-substituted-2-nitrobenzyloxy)methyluridine-5’-triphosphates were exquisite DNA synthesis terminators; therefore, we synthesized a library of 35 thymidine analogs and evaluated their activity using an MTT cell viability assay of MCF7 breast cancer cells chosen for their vulnerability to these nucleoside derivatives.

In this chapter, I report the synthesis of thymidine analogs bearing a 2-nitrobenzyl modifying moiety attached at the C-5 of the uracil nucleobase and evaluate their cytotoxic and cytostatic activity. The lead compound identified from these structure-activity relationship studies was further tested for toxicity to normal cells, and PCR termination by its corresponding

5’-triphosphate. Our contribution to the field of anti-cancer drug discovery is significant as it facilitates the exploration of the therapeutic potential of novel base-modified nucleoside species.

These compounds are unlikely to affect other targets than replicating DNA, and show promise to have a wider therapeutic window than present antimetabolites. This chapter is based on previously published work.(Borland, AbdulSalam et al. 2015)

5.2 Chemical synthesis

5.2.1 Materials

All chemicals, reagents, and solvents were purchased from Sigma-Aldrich Inc., TCI, and

Fisher Scientific, Inc., and used as received unless stated otherwise. All reactions were carried out under an atmosphere of dry argon in oven-dried glassware. Indicated reaction temperatures refer to those of the reaction bath, while room temperature is noted as 25°C. Pure reaction products were typically dried under high vacuum in the presence of phosphorus pentoxide.

Analytical thin layer chromatography (TLC) was performed with glass backed silica plates (5 x

20 cm, 60 Å, 250 μm). Visualization was accomplished using a 254 nm UV lamp. 1H and 13C

NMR spectra were recorded on either a Bruker Avance 400 MHz spectrometer or Bruker DPX

500 MHz spectrophotometer using solutions of samples in either of the deturated solvents: chloroform, methanol, acetonitrile, or water. Chemical shifts are reported in ppm with tetramethylsilane as standard. Data are reported as follows: chemical shift, number of protons, multiplicity (s = singlet, d = doublet, dd = doublet of doublet, t = triplet, q = quartet, b = broad, m = multiplet, abq = ab quartet), and coupling constants. High resolution mass spectral data were collected on a Shimadzu Q-TOF 6500. All novel compounds were characterized by 1H, 13C,

DEPT 13C, 31P (3aTP) NMR spectroscopy and high resolution mass spectrometry. The identity of previously made nucleoside derivatives was confirmed by comparison of their 1H NMR to the

published data (reference provided). HPLC analysis of final products was performed on an

Agilent 1200 HPLC with UV detection. Compounds biologically tested were greater than 95% pure as judged by 1H NMR and HPLC.

5.2.2 Chemical Synthesis

General procedure for preparation of base-modified nucleosides 2a-w. N3-tert-

Butyloxycarbonyl-5-bromomethyl-3’,5’-bis-O-tert-butyldimethylsilyl-2’-deoxyuridine (Litosh,

Wu et al. 2011) and appropriate alcohol (4-20 eq) were heated neat at 110-120 C for 0.5-3 hours under argon atmosphere. The mixture was cooled down to room temperature, dissolved in ethyl acetate (ca 5 ml), and silica (0.5-1.0 g) was added. The mixture was evaporated, and the solid was applied onto a silica gel chromatography column (hexane/ethyl acetate = 15:1 to 2:1, then dichloromethane/methanol = 0:1 to 10:1). Fractions that were not the starting alcohol were collected, evaporated under reduced pressure, dissolved in tetrahydrofuran (ca 5 mL), and to this solution chilled at 0 C tetra-n-butylammonium fluoride trihydrate (TBAF) was added (2.5 eq.).

The reaction mixture was stirred for 2-3 hours while gradually warming up to room temperature.

The solvent was removed under reduced pressure and the residue was purified by silica gel (ethyl acetate/methanol = 1:0 to 20:1) to afford product as waxy solid.

5.2.3 Synthesized Compounds

5-(benyl)oxymethyl-2'-deoxyuridine (2a). NOTE: no TBAF treatment was necessary. Heating 1

(86 mg, 0.132 mmol) with benzyl alcohol (286 mg, 2.346 mmol) for 1.5 hours at 118 C after column chromatography afforded 25 mg (54%) of product as 1:1 mixture of diastereomers. 1H

NMR (400 MHz, CD3OD): δ 8.05 7.33 (m, 5 H), 6.27 (t, 1 H, J = 6.7 Hz), 4.57 (s, 2 H), 4.39 (m,

1H), 4.30 (AB d, 1 H, J = 12.6 Hz), 4.24 (AB d, 1 H, J = 12.6 Hz), 3.93 (q, 1 H, J = 3.5 Hz), 3.78

(AB dd, 1 H, J = 12.0, 3.4 Hz), 3.78 (AB dd, 1 H, J = 12.0, 3.6 Hz), 2.27 (m, 2 H), 2.19 (m, 2 H),

1.42 (d, 3 H, J = 6.5 Hz).

5-[1-(phenyl)ethoxymethyl]-2'-deoxyuridine (2b). Heating 1 (121 mg, 0.186 mmol) with α- methylbenzyl alcohol (1-phenyl-1-ethanol) (0.228 g, 1.862 mmol) for 1 hour at 114 C followed by treatment with TBAF (0.303 g, 0.930 mmol) afforded after purification (method A) 20 mg

(30%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 7.96 and 7.95 (s, 1 H), 7.33 (br. m, 5 H), 6.27 (m, 1 H), 4.54 (m, 1H), 4.38 (m, 1

H), 4.10 (m, 2 H), 3.92 (m, 1 H), 3.75 (m, 2 H) 2.26 (m, 1 H), 2.19 (m, 2 H), 1.42 (m, 3 H). 13C

NMR (400 MHz, CD3OD) for diastereomers: δ 165.06 and 165.03 (C), 152.09 (C), 144.91 and

144.82 (C), 140.55 and 140.52 (CH), 129.53 (CH), 128.58 and 128.56 (CH), 127.32 and 127.28

(CH), 112.75 and 112.71 (C), 88.93 (CH), 86.50 (CH), 79.42 and 79.34 (CH), 72.22 and 72.18

(CH), 64.41 and 64.22 (CH2), 62.84 and 62.81 (CH2), 41.37 and 41.32 (CH2), 24.53 and 24.42

(CH3). HRMS (ESI) for [MH]+ C18H23N2O6 calculated: 363.15506, observed: 363.15516; for

[MNa]+ C18H22N2O6Na calculated: 385.13701, observed: 385.13712.

5-[1-(phenyl)-2-(methyl)-1-propoxymethyl]-2'-deoxyuridine (2c). Heating 1 (250 mg, 0.385 mmol) with α-isopropylbenzyl alcohol (2-methyl-1-phenyl-1-propanol) (1.16 g, 7.70 mmol) for 2 hours at 124 C followed by treatment with TBAF (303 mg, 0.963 mmol) afforded after purification (method A) 92 mg (61%) of product as 1:1 mixture of diastereomers. 1H NMR (500

MHz, CD3OD) for diastereomers: δ 7.91 (s, 1 H), 7.29 (m, 5 H), 6.27 (t, 1 H, J = 6.7 Hz), 4.39

(m, 1 H), 4.05 (m, 3 H), 3.93 (m, 1 H), 3.75 (m, 2 H), 2.27 (m, 1 H), 2.18 (m, 1 H), 1.91 (m, 1

H), 0.99 (m, 3 H), 0.72 (m, 3 H). 13C NMR (125 MHz, CD3OD) for diastereomers: δ 163.63

(C), 150.72 (C), 141.10 and 141.02 (C), 138.94 (CH), 127.80 (CH), 127.23 (CH), 127.14 (CH),

111.56 (C), 87.69 (CH), 87.54 (CH), 85.14 and 85.06 (CH), 70.94 (CH) and 70.87 (CH), 63.39 and 63.20 (CH2), 61.51 (CH2), 39.93 (CH2), 34.67 and 34.61 (CH), 18.09 (CH3), 18.00 (CH3).

HRMS (ESI) for [MH]+ C20H27N2O6 calculated: 391.18636, observed: 391.18644; for

[MNa]+ C20H26N2O6Na calculated: 413.16831, observed: 413.16836.

5-[1-(phenyl)-1-(cyclohexyl)methoxymethyl]-2'-deoxyuridine (2d). Heating 1 (97 mg, 0.149 mmol) with α-cyclohexylbenzyl alcohol (550 mg, 2.890 mmol) for 2.5 hours at 132 C followed by purification of bis- and mono-TBS products with subsequent treatment with TBAF (103 mg,

0.326 mmol) afforded after purification (method B) 22 mg (34%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 7.90 (s, 1 H), 7.28 (m, 5 H),

6.26 (m, 1 H), 4.39 (m, 1 H), 4.06 (m, 3 H), 3.93 (m, 1 H), 3.75 (m, 2 H), 2.28 (m, 1 H), 2.17 (m,

1 H), 2.04 (m, 1 H), 1.73 (m, 1 H), 1.60 (m, 3 H), 1.05 (m, 6 H). 13C NMR (100 MHz, CD3OD) for diastereomers δ 165.00 and 164.98 (C), 154.08 (C), 142.40 (C), 140.29 and 140.26 (CH),

129.15 (CH), 128.64 and 128.57 (CH), 128.48 and 128.46 (CH), 112.92 (C), 88.94 and 88.90

(CH), 88.14 (CH), 86.54 and 86.47 (CH), 72.36 and 72.27 (CH), 64.66 and 64.45 (CH2), 62.94 and 62.90 (CH2), 45.71 and 45.70 (CH), 41.32 (CH2), 30.62 and 30.41 (CH2), 27.65 (CH2),

27.17 and 27.12 (CH2). HRMS (ESI) for [MH]+ C23H31N2O6 calculated: 431.21766, observed: 431.21781; [MNa]+ C23H30N2O6Na calculated: 453.19961, observed: 453.19977.

5-[(diphenyl)methoxymethyl]-2'-deoxyuridine (2e). Heating 1 (250 mg, 0.385 mmol) with diphenylmethanol (1.42 g, 7.70 mmol) for 2.5 hours at 120 C followed by treatment with TBAF

(607 mg, 1.925 mmol) afforded after purification (method A) 6 mg (3%) of product. 1H NMR

(400 MHz, CD3OD) δ 7.97 (s, 1 H), 7.36 (d, 4 H, J = 7.9 Hz), 7.29 (m, 4 H), 7.21 (m, 2 H), 6.26

(t, 1 H, J = 6.7 Hz), 5.51 (s, 1 H), 4.37 (q, 1 H, J = 3.5 Hz), 4.29 (AB d, 1 H, J = 12.1 Hz), 4.24

(AB d, 1 H, J = 12.1 Hz), 3.93 (m, 1 H), 3.75 (AB dd, 1 H, J = 12.0, 3.5), 3.70 (AB dd, 1 H, J =

12.0, 3.9), 2.27 (m, 1 H), 2.17 (m, 1 H). 13C NMR (100 MHz, CD3OD) δ 165.04 (C), 152.06

(C), 143.52 (C), 140.53 (CH), 129.34 (CH), 128.47 (CH), 128.09 (CH), 112.64 (C), 88.91 (CH),

86.56 (CH), 84.76 (CH), 72.22 (CH), 64.84 (CH2), 62.91 (CH2), 41.28 (CH2). HRMS (ESI) for

[MH]+ C23H25N2O6 calculated: 425.17071, observed: 425.17082; [MNa]+ C23H24N2O6Na calculated: 447.15266, observed: 447.15276.

5-[1-(phenyl)-3,3-(dimethyl)-1-butoxymethyl]-2'-deoxyuridine (2f). Heating 1 (44 mg, 0.068 mmol) with α-neo-pentylbenzyl alcohol (3,3-dimethyl-1-phenyl-1-butanol) (41 mg, 0.239 mmol) for 2.5 hours at 110 C followed by treatment with TBAF (53 mg, 0.170 mmol) afforded after purification (method A) 12 mg (43%) of product as 1:1 mixture of diastereomers. 1H NMR (400

MHz, CD3OD) for diastereomers: δ 7.97 and 7.95 (s, 1 H), 7.34 (m, 5 H), 6.29 (m, 1 H), 4.51 (m,

1 H), 4.41 (m, 1 H), 4.05 (m, 2 H), 3.94 (m, 1 H), 3.78 (m, 2 H), 2.28 (m, 1 H), 2.21 (m, 1 H),

1.82 (m, 1 H), 1.44 (m, 1 H), 0.99 and 0.98 (2 s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers δ 163.60 (C), 150.63 (C), 143.72 (C), 139.23 and 139.06 (CH), 128.13 (CH),

127.00 (CH), 126.28 and 126.21 (CH), 111.47 (C), 86.51 (CH), 85.09 and 84.93 (CH), 80.08 and

79.97 (CH), 70.85 (CH), 62.71 and 62.52 (CH2), 61.52 and 61.49 (CH2), 51.94 and 51.82

(CH2), 40.01 (CH2), 30.01 (C), 29.32 (CH3). HRMS (ESI) for [MH]+ C22H31N2O6 calculated: 419.21766, observed: 419.21780; [MNa]+ C22H30N2O6Na calculated: 441.19961, observed: 441.19974.

5-[1-(phenyl)-2,2-(dimethyl)-1-propoxymethyl]-2'-deoxyuridine (2g). Heating 1 (250 mg, 0.385 mmol) with α-tert-butylbenzyl alcohol (2,2-dimethyl-1-phenyl-1-propanol) (1.26 g, 7.70 mmol) for 2 hours at 120 C followed by treatment with TBAF (607 mg, 1.925 mmol) afforded after purification (method A) 22 mg (21%) of product as 1:1 mixture of diastereomers. 1H NMR (400

MHz, CD3OD) for diastereomers: δ 7.93 and 7.92 (2 s, 1 H), 7.32 (m, 5 H), 6.28 (m, 1 H), 4.42

(m, 1 H), 4.06 (m, 3 H), 3.95 (m, 1 H), 3.77 (m, 2 H), 2.30 (m, 1 H), 2.19 (m, 1 H), 0.91 (s, 9 H).

13C NMR (100 MHz, CD3OD) for diastereomers: δ 163.61 (C), 150.68 (C), 139.42 (C), 138.73 and 138.62 (CH), 128.24 and 128.20 (CH), 127.18 (CH), 126.95 (CH), 111.61 (C), 87.66 and

87.56 (CH), 87.58 and 87.51 (CH), 85.14 and 84.99 (CH), 70.99 and 70.89 (CH), 63.79 and

63.53 (CH2), 61.59 (CH2), 39.94 and 39.87 (CH2), 35.11 (C), 25.38 (CH3). HRMS (ESI) for

[MH]+ C21H29N2O6 calculated: 405.20201, observed: 405.20210; for [MNa]+

C21H28N2O6Na calculated: 427.18396, observed: 427.18409.

5-[1-(2-methyl)phenyl-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (2h). Heating 1 (125 mg,

0.195 mmol) with α-tert-butyl-2-methybenzyl alcohol (3,3-dimethyl-1-(2-methyl)phenyl-1- propanol) (174 mg, 0.776 mmol) for 1 hour at 112 C using purification (method A) afforded 8 mg (10%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 7.89 and 7.88 (2 s, 1 H), 7.41 (m, 1 H), 7.16 (m, 3 H), 6.28 (m, 1 H), 4.46 (s, 1

H), 4.41 (m, 1 H), 4.03 and 4.01 (2 s, 2 H), 3.95 (m, 1 H), 3.76 (m, 2 H), 2.38 and 2.37 (2 s, 3 H),

2.30 (m, 1 H), 2.19 (m, 1 H), 0.95 (s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers δ

163.57 (C), 150.67 (C), 138.45 and 138.39 (CH), 136.84 (C), 136.74 (C), 129.76 (CH), 127.76 and 127.72 (CH), 126.71 and 126.66 (CH), 124.97 and 124.93 (CH), 111.82 (C), 87.54 and

87.52 (CH), 85.10 and 85.03 (CH), 83.83 and 83.79 (CH), 71.02 (CH), 63.41 and 63.31 (CH2),

61.59 (CH2), 39.88 and 39.85 (CH2), 36.41 and 36.37 (C), 25.43 (CH3), 19.29 and 19.23 (CH3).

HRMS (ES+ TOF) for [MNa]+ C22H30N2O6Na calculated: 441.20020 observed: 441.19960.

5-[1-(2-cyanophenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (2i). Heating 1 (114 mg,

0.176 mmol) with α-tert-butyl-2-cyanobenzyl alcohol (2,2-dimethyl-1-(2-cyano)phenyl-1- propanol) (1.61 g, 7.70 mmol) for 2 hours at 120 C followed by treatment with TBAF (607 mg,

1.925 mmol) afforded after purification (method A) 14 mg (18%) of product as 1:1 mixture of

diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 7.99 and 7.98 (2 s, 1 H), 7.89

(d, J = 7.7 Hz, 1 H), 7.54 (m, 3 H), 6.32 (m, 1 H), 5.33 (s, 1 H), 4.38 (m, 3 H), 3.91 (m, 1 H),

3.68 (m, 2 H), 2.24 (m, 2 H), 1.01 (s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers: δ

164.07 (C), 150.86 (C), 145.06 (C), 137.80 (CH), 131.33 (CH), 130.46 (C), 128.57 (CH), 123.06 and 122.93 (CH), 115.25 (CH), 112.98 (C), 91.70 (CH), 87.56 and 87.52 (CH), 85.11 and 85.02

(CH), 71.00 and 70.92 (CH), 61.55 (CH2), 58.09 (CH2), 39.83 (CH2), 35.58 (C), 24.29 (CH3).

HRMS (ES+ TOF) for [MH]+ C22H28N3O6 calculated: 430.19870, observed: 430.19700; for

[MNa]+ C22H27N3O6Na calculated: 452.17970, observed: 452.18040.

5-[1-(2-chlorophenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (2j). Heating 1 (250 mg,

0.385 mmol) with α-tert-butyl-2-chlorobenzyl alcohol (3,3-dimethyl-1-(2-chloro)phenyl-1- propanol) (540 mg, 2.718 mmol) for 3 hours at 118 C followed by treatment with TBAF (43 mg, 0.136 mmol) afforded after purification (method A) 19 mg (11%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 7.91 and 7.89 (2 s, 1 H),

7.53 (d, 1 H, J = 7.6 Hz), 7.31 (m, 3 H), 6.27 (m, 1 H), 4.67 (s, 1 H), 4.41 (m, 1 H), 4.10 (m, 1

H), 4.00 (m, 1 H), 3.95 (m, 1 H), 3.76 (m, 2 H), 2.27 (m, 2 H), 0.97 and 0.96 (2 s, 9 H). 13C

NMR (100 MHz, CD3OD) for diastereomers δ 163.50 and 163.47 (C), 150.68 (C), 139.10 and

138.68 (CH), 137.28 (C), 134.38 (C), 129.76 and 129.70 (CH), 128.89 and 128.80 (CH), 128.40 and 128.36 (CH), 111.36 and 111.10 (C), 109.90 (CH), 87.57 and 87.52 (CH), 85.24 and 85.07

(CH), 83.89 and 83.36 (CH), 70.97 (CH), 64.20 and 63.72 (CH2), 61.54 (CH2), 39.89 and 39.83

(CH2), 36.37 and 36.33 (C), 25.15 (CH3). HRMS (ESI+) for [MNa]+ C21H2735ClN2O6Na calculated: 461.14499 observed: 461.14504; for [MNa]+ C21H2737ClN2O6Na calculated:

463.14213 observed: 463.14210. HRMS (ESI-) for [M-H]- C21H2635ClN2O6 calculated:

437.14849 observed: 437.14851; C21H2637ClN2O6 calculated: 439.14561 observed:

439.14656.

5-[1-(2-bromomethyl)phenyl-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (2k). Heating 1

(208 mg, 0.320 mmol) with α-tert-butyl-2-bromobenzyl alcohol (3,3-dimethyl-1-(2- bromo)phenyl-1-propanol) (389 mg, 1.607 mmol) for 1 hour at 112 C. with purification

(method A) afforded 3 mg (2%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz,

CD3OD) for diastereomers: δ 7.89 and 7.86 (2 s, 1 H), 7.54 (m, 2 H), 7.36 (m, 1 H), 7.18 (m, 1

H), 6.25 (m, 1 H), 4.64 and 4.63 (2 s, 1 H), 4.40 (m, 1 H), 4.04 (m, 2 H), 3.92 (m, 1 H), 3.75 (m,

2 H), 2.24 (m, 2 H), 0.97 and 0.96 (s, 9 H); 13C NMR (100 MHz, CD3OD) for diastereomers δ

165.03 and 164.88 (C), 152.12 (C), 140.55 and 140.08 (CH), 140.36 (C), 140.26 (C), 133.72 and

133.66 (CH), 131.35 and 131.32 (CH), 130.18 and 130.15 (CH), 128.20 and 128.18 (CH),

112.77 and 112.52 (C), 88.97 and 88.94 (CH), 87.64 and 86.54 (CH), 87.10 and 86.71 (CH),

72.41 and 72.39 (CH), 65.50 and 65.10 (CH2), 62.99 and 62.97 (CH2), 41.30 and 41.27 (CH2),

37.92 and 37.88 (C), 26.68 (CH3). HRMS (ESI) for [MH]+ C21H2879BrN2O6 calculated:

483.11307 observed: 483.11264, C21H2881BrN2O6 calculated: 485.11103 observed:

483.11055; for [MNa]+ C21H2779BrN2O6Na calculated: 505.09502 observed: 505.09452,

C21H2781BrN2O6Na calculated: 507.09297 observed: 507.09212.

5-[1-(2-nitrophenyl)ethoxymethyl]-2'-deoxyuridine (2o).18 Heating 1 (175 mg, 0.270 mmol) with α-isopropyl-2-nitrobenzyl alcohol (1-(2-nitro)phenyl-2-methyl-1-propanol) (400 mg, 2.050 mmol) for 1 hour at 105-114 C afforded after purification (method A) 16 mg (14%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.01 and

7.98 (2 s, 1 H), 7.90 (d, 1 H, J = 8.5 Hz), 7.77 (m, 1 H), 7.51 (m, 1 H), 6.27 (m, 1 H), 4.78 (m, 1

H), 4.41 (m, 1 H), 4.13 (m, 2 H), 3.94 (m, 1 H), 3.77 (m, 2 H), 2.25 (m, 2 H), 1.96 (m, 1 H), 0.97

and 0.96 (2 d, 3 H, J = 6.7 Hz), 0.88 and 0.86 (2 d, 3 H, J = 7.0 Hz). 13C NMR (100 MHz,

CD3OD) for diastereomers δ 163.63 and 163.56 (C), 150.66 and 150.64 (C), 149.52 (C), 139.66 and 139.50 (CH), 136.52 and 136.51 (C), 132.56 and 132.54 (CH), 129.01 and 128.95 (CH),

128.05 (CH), 123.71 and 123.65 (CH), 111.05 and 110.86 (C), 87.57 (CH), 85.08 and 85.07

(CH), 81.08 and 80.82 (CH), 70.90 (CH), 64.23 and 63.96 (CH2), 61.48 and 61.45 (CH2), 39.95 and 39.90 (CH2), 34.67 (CH), 18.31 and 18.26 (CH3), 16.64 and 16.57 (CH3). HRMS (ESI+) for [MH]+ C20H26N3O8 calculated: 436.17144, observed: 436.17149; for [MNa]+

C20H25N3O8Na calculated: 458.15339, observed: 458.15342. HRMS (ESI-) for [M-H]-

C20H24N3O8 calculated: 434.15689, observed: 434.15669.

5-[1-(2-nitro)phenyl-1-(cyclohexyl)methoxymethyl]-2'-deoxyuridine (2p). Heating 1 (150 mg,

0.231 mmol) with α-cyclohexyl-2-nitrobenzyl alcohol (440 mg, 1.880 mmol) for 2.5 hours at 116

C followed by purification of bis- and mono-TBS products with subsequent treatment with

TBAF (73 mg, 0.231 mmol) afforded after purification (method B) 28 mg (25%) of product as

1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 7.99 and 7.96

(2 s, 1 H), 7.89 (d, J = 8.1 Hz, 1 H), 7.71 (m, 2 H), 7.50 (d, J = 7.6 Hz, 1 H), 6.27 (t, J = 6.6 Hz, 1

H), 4.78 (m, 1 H), 4.42 (m, 1 H), 4.12 (m, 3 H), 3.94 (m, 1 H), 3.77 (m, 2 H), 2.28 (m, 1 H), 2.22

(m, 1 H), 1.87 (m, 1 H), 1.70 (m, 2 H), 1.30 (m, 4 H), 1.17 (m, 4 H). 13C NMR (100 MHz,

CD3OD) for diastereomers δ 163.57 (C), 150.65 (C), 149.62 and 149.53 (C), 139.61 and 139.51

(CH), 136.21 (C), 132.48 (CH), 129.11 and 129.05 (CH), 128.02 (CH), 123.65 and 123.59 (CH),

111.02 and 110.86 (C), 87.58 (CH), 85.07 (CH), 80.64 and 80.23 (CH), 70.94 and 70.91 (CH),

64.15 and 63.96 (CH2), 61.50 (CH2), 44.50 and 44.47 (CH), 39.92 and 39.89 (CH2), 29.22 and

29.15 (CH2), 28.05 and 27.98 (CH2), 26.09 (CH2) , 25.96 and 25.94 (CH2), 25.78 (CH2).

HRMS (ESI) for [MH]+ C23H30N3O8 calculated: 476.20274, observed: 476.20292; for

[MNa]+ C23H29N3O8Na calculated: 498.18469, observed: 498.18486.

5-[(2-nitrophenyl)phenyl}methoxymethyl]-2'-deoxyuridine (o-2q). Heating 1 (210 mg, 0.323 mmol) with α-phenyl-2-nitrobenzyl alcohol (phenyl(2-nitrophenyl)methanol) (361 mg, 1.576 mmol) for 2.5 hours at 110-117 C followed by purification of bis- and mono-TBS products with subsequent treatment with TBAF (73 mg, 0.231 mmol) afforded after purification (method A) 12 mg (7%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.04 and 8.00 (2 s, 1 H), 7.88 (m, 1 H), 7.68 (m, 2 H), 7.52 (m, 1 H), 7.34 (m, 5

H), 6.28 (m, 1 H), 6.18 and 6.17 (2 s, 1 H), 4.42 (m, 1 H), 4.30 (m, 3 H), 3.94 (m, 1 H), 3.78 (m,

2 H), 2.28 (m, 2 H). 13C NMR (100 MHz, CD3OD) for diastereomers δ 163.71 (C), 159.43 (C),

158.87 (C), 150.64 (C), 140.03 and 139.74 (CH), 136.21 and 136.18 (C), 132.60 (CH), 128.13

(CH), 128.11 (CH), 128.09 (CH), 127.58 (CH), 123.92 (CH), 123.87 (CH), 110.71 (C), 87.59

(CH), 85.13 (CH), 78.29 and 78.16 (CH), 70.82 (CH), 64.01 and 63.91 (CH2), 61.48 (CH2),

39.95 (CH2). HRMS (ES+ TOF) for [MNa]+ C23H29N3O8Na calculated: 492.13820, observed: 492.13830.

5-[(3-nitrophenyl)phenyl}methoxymethyl]-2'-deoxyuridine (m-2q). Heating 1 (200 mg, 0.308 mmol) with α-phenyl-3-nitrobenzyl alcohol (phenyl(3-nitrophenyl)methanol) (350 mg, 1.673 mmol) for 1 hour at 120 C followed by purification of bis- and mono-TBS products with subsequent treatment with TBAF (49 mg, 0.155 mmol) afforded after purification (method B) 16 mg (12%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.25 and 8.25 (2 s, 1 H), 8.11 (m, 1 H), 8.07 (m, 1 H), 7.79 (m, 1 H), 7.55 (dt, 1

H, J = 8.0, 1.3 Hz), 7.42 (m, 2 H), 7.35 (m, 2 H), 7.28 (m, 1 H), 6.26 (m, 1 H), 5.68 (s, 1 H), 4.39

(m, 1 H), 4.32 (m, 2 H), 3.94 (q, 1 H, J = 3.4 Hz), 3.74 (m, 2 H), 2.29 (m, 1 H), 2.20 (m, 1 H).

13C NMR (100 MHz, CD3OD) for diastereomers δ 163.69 (C), 150.66 (C), 148.27 (C), 144.96

(C), 141.11 (C), 139.65 (CH), 132.81 and 132.80 (CH), 129.19 (CH), 128.36 (CH), 127.68 (CH),

126.83 and 126.80 (CH), 121.85 (CH), 121.12 and 121.09 (CH), 110.91 (C), 87.65 (CH), 85.21 and 85.19 (CH), 82.08 and 82.03 (CH), 70.87 (CH), 63.79 and 63.68 (CH2), 61.45 (CH2), 40.04

(CH2). HRMS (ESI+) for [MH]+ C23H24N3O8 calculated: 470.15634, observed: 470.15581; for [MNa]+ C23H23N3O8Na calculated: 492.13828, observed: 492.13777. HRMS (ESI-) for

[M-H]- C23H22N3O8 calculated: 468.14124, observed: 468.14101.

5-[(4-nitrophenyl)phenyl}methoxymethyl]-2'-deoxyuridine (p-2q). Heating 1 (150 mg, 0.231 mmol) with α-phenyl-3-nitrobenzyl alcohol (phenyl(4-nitrophenyl)methanol) (211 mg, 0.923 mmol) for 20 minutes at 115 C followed by treatment with TBAF (113 mg, 0.358 mmol) afforded after purification (method B) 7 mg (6%) of product as 1:1 mixture of diastereomers. 1H

NMR (400 MHz, CD3OD) for diastereomers: δ 8.14 (d, 2 H, J = 8.7 Hz), 8.03 (s, 1 H), 7.61 (d, 2

H, J = 8.7 Hz), 7.37 (m, 2 H), 7.30 (m, 2 H), 7.23 (m, 1 H), 6.23 (m, 1 H), 5.62 (s, 1 H), 4.36 (m,

1 H), 4.26 (m, 2 H), 3.90 (m, 1 H), 3.71 (m, 2 H), 2.25 (m, 1 H), 2.15 (m, 1 H). 13C NMR (100

MHz, CD3OD) for diastereomers δ 166.94 and 165.11 (C), 152.09 and 151.34 (C), 148.53 and

148.00 (C), 142.41 (C), 140.96 (CH), 141.11 (C), 129.77 (CH), 129.64 and 129.54 (C), 129.13

(CH), 128.81 (CH), 128.32 and 128.29 (CH), 124.44 (CH), 112.35 (C), 89.06 (CH), 86.61 (CH),

83.61 (CH), 72.23 (CH), 65.11 and 65.05 (CH2), 62.84 (CH2), 41.47 (CH2). HRMS (ESI+) for

[MH]+ C23H24N3O8 calculated: 470.15634, observed: 470.15582; for [MNa]+

C23H23N3O8Na calculated: 492.13828, observed: 492.13780. HRMS (ESI-) for [M-H]-

C23H22N3O8 calculated: 468.14124, observed: 468.14112.

5-[1-(2-nitrophenyl)-3,3-(dimethyl)butoxymethyl]-2'-deoxyuridine (2r). Heating 1 (210 mg,

0.323 mmol) with α-neo-pentyl-2-nitrobenzyl alcohol (3,3-dimethyl-1-(2-nitro)phenyl-1-butanol)

(480 mg, 2.152 mmol) for 2 hours at 120 C followed by treatment with TBAF (607 mg, 1.925 mmol) afforded after purification (method A) 18 mg (12%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.02 and 8.00 (2 s, 1 H), 7.90

(d, 1 H, J = 8.2 Hz), 7.82 (m, 1 H) 7.70 (t, 1 H, J = 7.6 Hz), 7.48 (m, 1 H), 6.27 (t, 1 H, J = 6.9

Hz), 5.14 (m, 1 H), 4.42 (m, 1 H), 4.05 (m, 2 H), 3.94 (m, 1 H), 3.78 (m, 2 H), 2.28 (m, 1 H),

2.19 (m, 1 H), 1.71 (m, 1 H), 1.53 (m, 1 H), 1.05 and 1.04 (2 s, 9 H). 13C NMR (100 MHz,

CD3OD) for diastereomers: δ 163.55 and 163.51 (C), 150.64 and 150.60 (C), 148.26 and 148.22

(C), 139.90 and 139.66 (CH), 139.16 (C), 137.73 (CH), 133.16 and 133.11 (CH), 128.29 and

127.82 (CH), 123.73 and 123.67 (CH), 111.10 and 110.88 (C), 87.67 and 87.58 (CH), 85.07 and

84.97 (CH), 75.02 and 74.74 (CH), 70.89 (CH), 63.25 and 63.13 (CH2), 61.46 (CH2), 51.05 and

51.01 (CH2), 40.05 and 39.93 (CH2), 30.42 and 30.39 (C), 29.40 and 29.38 (CH3). HRMS

(ESI) for [MNa]+ C22H29N3O8Na calculated: 486.18430, observed: 486.18520.

5-[1-(2-nitrophenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (o-2s). Heating 1 (250 mg,

0.385 mmol) with α-tert-butyl-2-nitrobenzyl alcohol (2,2-dimethyl-1-(2-nitro)phenyl-1-propanol)

(1.61 g, 7.70 mmol) for 2 hours at 120 C followed by treatment with TBAF (607 mg, 1.925 mmol) afforded after purification (method A) 33 mg (19%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.01 and 7.99 (2 s, 1 H), 7.81

(m, 2 H), 7.68 (m, 1 H), 7.51 (m, 1 H), 6.28 (t, 1 H, J = 6.9 Hz), 4.98 (s, 1 H), 4.42 (m, 1 H), 4.20

(m, 2 H), 3.94 (m, 1 H), 3.76 (m, 2 H), 2.26 (m, 2 H), 0.85 and 0.84 (2 s, 9 H). 13C NMR (100

MHz, CD3OD) for diastereomers: δ 163.60 and 163.55 (C), 150.89 and 150.75 (C), 150.70 (C),

139.80 and 139.41 (CH), 133.81 (C), 131.76 and 131.74 (CH), 129.91 and 129.82 (CH), 128.14

(CH), 123.56 and 123.43 (CH), 111.01 and 110.74 (C), 87.55 (CH), 85.13 and 85.04 (CH), 81.76 and 81.04 (CH), 70.98 and 70.95 (CH), 64.49 and 64.18 (CH2), 61.52 and 61.46 (CH2), 39.86

100

and 39.78 (CH2), 36.12 and 36.02 (C), 24.84 and 24.82 (CH3). HRMS (ESI+) for [MH]+

C21H28N3O8 calculated: 450.18709, observed: 450.18708; for [MH]+ C21H28N3O8 calculated: 472.16904, observed: 472.16918. HRMS (ESI-) for [M-H]- C21H26N3O8 calculated: 448.17254, observed: 448.17258.

5-[1-(3-nitrophenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (m-2s). Heating 1 (223 mg,

0.343 mmol) with α-tert-butyl-2-nitrobenzyl alcohol (2,2-dimethyl-1-(3-nitro)phenyl-1-propanol)

(0.575 g, 2.748 mmol) for 45 minutes at 108-112 C followed by treatment with TBAF (87 mg,

1.925 mmol) afforded after purification (method B) 19 mg (12%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.15 (m, 2 H), 8.00 (s, 1 H),

7.70 (m, 1 H), 7.57 (m, 1 H), 6.24 (m, 1 H), 4.41 (m, 1 H), 4.21 and 4.20 (2 s, 1 H), 4.12 (m, 2

H), 3.93 (m, 1 H), 3.77 (m, 2 H), 2.23 (m, 2 H), 0.90 and 0.90 (2 s, 9 H). 13C NMR (100 MHz,

CD3OD) for diastereomers: δ 165.00 (C), 152.02 (C), 149.20 (C), 143.71 (C), 140.95 and 140.92

(CH), 135.94 and 135.89 (CH), 129.81 and 129.79 (CH), 124.13 (CH), 123.28 and 123.25 (CH)

112.54 (C), 89.74 and 89.61 (CH), 89.06 and 89.02 (CH), 86.54 and 86.50 (CH), 72.36 and

72.32 (CH), 65.58 (CH2), 62.86 (CH2), 41.49 and 41.41 (CH2), 36.56 (C), 26.46 (CH3).

HRMS (ESI) for [MH]+ C21H28N3O8 calculated: 450.18709, observed: 450.18711; for

[MNa]+ C21H27N3O8Na calculated: 472.16958, observed: 472.16903; for [M-H]-

C21H26N3O8 calculated: 448.17254, observed: 448.17236.

5-[1-(2,6-dinitrophenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (2t). Heating 1 (259 mg,

0.399 mmol) with α-tert-butyl-2,6-dinitrobenzyl alcohol (2,2-dimethyl-1-(2,6-dinitro)phenyl-1- propanol) (342 g, 1.345 mmol) for 10 minutes at 105 C followed by treatment with TBAF (314 mg, 0.997 mmol) afforded after purification (method A) 23 mg (12%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.05 (m, 1 H), 8.01 and

101

7.86 (2 s, 1 H), 7.74 (m, 2 H), 6.39 and 6.34 (2 t, J = 6.7 Hz, 1 H), 5.20 and 5.19 (2 s, 1 H), 4.44

(m, 1 H), 4.25 (m, 2 H), 3.94 (m, 1 H), 3.74 (m, 2 H), 2.33 (m, 2 H), 0.87 (s, 9 H). 13C NMR

(100 MHz, CD3OD) for diastereomers (NOTE: due to the presence of two ortho-substituents, there is, apparently, restricted rotation of the 2,6-dinitrophenyl group around its 1-C - 4-C axis, which thereby makes 2-CNO2 non-equivalent to 6-CNO2, and accordingly, 3-CH non- equivalent to 5-CH): δ 163.62 and 163.55 (C), 150.89 and 150.74 (C), 151.25 (C), 151.13 (C),

140.62 and 139.45 (CH), 130.13 and 130.04 (CH), 128.06 and 128.01 (CH), 126.26 and 126.16

(CH), 125.47 and 125.32 (C), 109.97 and 109.84 (C), 87.55 and 87.35 (CH), 85.05 and 84.56

(CH), 82.85 and 82.05 (CH), 71.07 and 70.96 (CH), 66.23 and 65.85 (CH2), 61.69 (CH2), 39.82 and 39.50 (CH2), 37.88 and 37.83 (C), 25.72 (CH3). HRMS (ESI) for [MH]+ C21H27N4O10 calculated: 495.17217, observed: 495.17218; for [M-H]- C21H26N4O10 calculated: 493.15762, observed: 493.15754.

5-[1-(2-methoxyphenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (o-2u). Heating 1 (250 mg, 0.385 mmol) with α-tert-butyl-2-methoxybenzyl alcohol (3,3-dimethyl-1-(2- methoxy)phenyl-1-propanol) (625 mg, 3.460 mmol) for 2.5 hours at 114-128 C followed by treatment with TBAF (303 mg, 0.963 mmol) afforded after purification (method A) 94 mg (56%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ

7.82 and 7.81 (2 s, 1 H), 7.33 (d, 1 H, J = 7.8 Hz), 7.21 (m, 1 H), 6.90 (m, 2 H), 6.25 (m, 1 H),

4.62 and 4.61 (2 s, 1 H), 4.40 (m, 1 H), 4.01 (m, 3 H), 3.79 (s, 3 H), 3.73 (m, 2 H), 2.28 (m, 1 H),

2.17 (m, 1 H), 0.89 and 0.88 (2 s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers δ

163.53 (C), 157.92 (C), 150.71 (C), 138.62 and 138.27 (CH), 128.33 (CH), 127.93 (CH), 127.78

(C), 119.65 (CH), 111.82 and 111.56 (C), 109.90 (CH), 87.51 (CH), 85.10 (CH), 81.13 and

80.05 (CH), 71.02 (CH), 63.69 and 63.46 (CH2), 61.62 (CH2), 54.33 (CH3), 39.86 and 39.78

102

(CH2), 35.78 and 35.74 (C), 25.21 (CH3). HRMS (ESI) for [MH]+ C22H31N2O7 calculated:

435.21258 observed: 435.21261; for [MNa]+ C22H30N2O7Na calculated: 457.19452 observed:

457.19451.

5-[1-(3-methoxyphenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (m-2u). Heating 1 (150 mg, 0.231 mmol) with α-tert-butyl-3-methoxybenzyl alcohol (3,3-dimethyl-1-(3- methoxy)phenyl-1-propanol) (200 mg, 1.030 mmol) for 2 hours at 120 C followed by treatment with TBAF (182 mg, 0.578 mmol) afforded after purification (method A) 5 mg (5%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 7.91 and

7.91 (2 s, 1 H), 7.22 (m, 1 H), 6.85 (m, 3 H), 6.27 (m, 1 H), 4.42 (2 s, 1 H), 4.09 (m, 3 H), 3.95

(m, 1 H), 3.81 and 3.81 (2 s, 3 H), 3.77 (m, 2 H), 2.27 (m, 2 H), 0.92 (s, 9 H). 13C NMR (100

MHz, CD3OD) for diastereomers δ 159.22 and 159.19 (C), 150.73 and 150.68 (C), 141.36 and

141.17 (C), 138.86 and 138.72 (CH), 128.12 and 128.10 (CH), 120.77 and 120.67 (CH), 115.38 and 115.31 (C), 113.72 and 113.64 (CH), 112.31 (CH), 111.82 and 111.56 (C), 89.60 and 89.45

(CH), 87.59 and 87.52 (CH), 85.15 and 84.98 (CH), 71.00 and 70.90 (CH), 64.07 and 63.66

(CH2), 61.58 and 61.51 (CH2), 54.20 (CH3), 39.94 and 39.89 (CH2), 35.10 and 35.08 (C), 25.45 and 25.43 (CH3). HRMS (ESI+) for [MH]+ C22H31N2O7 calculated: 435.21258 observed:

435.21259; for [MNa]+ C22H30N2O7Na calculated: 457.19452 observed: 457.19450. HRMS

(ESI-) for [M-H]- C22H29N2O7 calculated: 433.19802 observed: 433.19809.

5-[1-(4-methoxyphenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (p-2u). Heating 1 (346 mg, 0.539 mmol) with α-tert-butyl-4-methoxybenzyl alcohol (3,3-dimethyl-1-(4- methoxy)phenyl-1-propanol) (620 mg, 2.150 mmol) for 2.5 hours at 120 C followed by treatment with TBAF (870 mg, 2.762 mmol) afforded after purification (method A) 11 mg (5%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ

103

7.89 and 7.87 (s, 1 H), 7.19 (d, J = 8.6 Hz, 2 H), 6.87 (m, 2 H), 6.27 (m, 1 H), 4.40 (m, 1 H), 4.05

(m, 3 H), 3.94 (m, 1 H), 3.78 and 3.78 (2 s, 3 H), 3.76 (m, 2 H), 2.28 (m, 1 H), 2.18 (m, 1 H),

0.88 (s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers δ 163.75 (C), 159.05 (C), 150.80

(C), 138.62 and 138.51 (CH), 131.39 and 131.32 (C), 129.24 and 129.22 (CH), 112.62 (CH),

111.78 and 111.72 (C), 89.28 and 89.16 (CH), 87.57 and 87.49 (CH), 85.15 and 85.00 (CH),

71.01 and 70.91 (CH), 63.62 and 63.38 (CH2), 61.61 and 61.57 (CH2), 54.28 (CH3), 39.93 and

39.83 (CH2), 35.20 (C), 25.38 (CH3). HRMS (ES+ TOF) [MNa]+ C22H30N2O7Na calculated:

457.19510 observed: 457.19490.

5-[1-(4-iodo-2-nitrophenyl)-2,2-(dimethyl)propoxymethyl]-2'-deoxyuridine (2v). Heating 1 (400 mg, 0.616 mmol) with α-tert-butyl-4-iodo-2-nitrobenzyl alcohol (2,2-dimethyl-1-(4-iodo-2- nitro)phenyl-1-propanol) (717 mg, 2.140 mmol) for 2 hours at 120 C followed by treatment with TBAF (607 mg, 1.925 mmol) afforded after purification (method A) 163 mg (28%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.14

(m, 1 H), 8.00 (m, 2 H), 7.51 (d, J = 8.4 Hz, 1 H), 6.26 (m, 1 H), 4.42 (m, 1 H), 4.19 (m, 2 H),

3.94 (m, 1 H), 3.76 (m, 2 H), 2.25 (m, 2 H), 0.82 and 0.80 (2 s, 9 H). 13C NMR (100 MHz,

CD3OD) for diastereomers δ 164.98 and 164.92 (C), 152.34 and 152.24 (C), 152.05 (C), 142.25 and 142.21 (CH), 141.38 and 141.11 (CH), 135.01 and 134.97 (C), 133.45 and 133.35 (CH),

133.05 and 132.96 (CH), 112.17 and 111.94 (C), 92.91 (C), 88.95 and 88.94 (CH), 86.62 and

86.48 (CH), 82.93 and 82.36 (CH), 72.34 and 72.27 (CH), 65.91 and 65.66 (CH2), 62.85 and

62.80 (CH2), 41.32 and 41.24 (CH2), 37.45 and 37.36 (C), 26.15 (CH3). HRMS (ESI) for

[MH]+ C21H27IN3O8 calculated: 576.08428, observed: 576.08383; for [MNa]+

C21H26IN3O8Na calculated: 598.06623, observed: 598.06581.

104

N3-tert-Butyloxycarbonyl-5-(di-tert-butylcarbinol)oxymethyl-3’,5’-bis-O-tert- butyldimethylsilyl-2’-deoxyuridine. Compound 1 (255 mg, 0.392 mmol) and di-tert- butylcarbinol (453 mg, 1.140 mmol) were placed in an iron screw-top vial equipped with a ball followed by vigorous shaking at room temperature for 20 hours under argon atmosphere. The contents of the vial were dissolved in ethyl acetate (1 mL) and mixed with silica (ca 500 mg).

The solvent was evaporated, and the powder was applied onto a chromatography column (SiO2, hexane/ethyl acetate = 15:1 to 6:1) to afford 50 mg (18%) of crude product. 1H NMR (500

MHz, CDCl3) δ 7.64 (s, 1 H), 6.27 (t, 1 H, J = 6.7 Hz), 4.49 (m, 1 H), 4.44 (m, 2 H), 3.95 (q, 2

H, J = 3.4 Hz), 3.92 (AB d, 1 H, J = 11.0 Hz), 3.75 (AB d, 1 H, J = 11.0 Hz), 2.82 (s, 1 H), 2.33

(m, 2 H), 1.62 (s, 9 H), 1.05 (s, 18 H), 0.91 (s, 18 H), 0.10 (2 s, 6 H), 0.09 (s, 6 H). The product was not further characterized but introduced into the subsequent transformation as is.

5-(di-tert-butylcarbinol)oxymethyl-3’,5’-bis-O-tert-butyldimethylsilyl-2’-deoxyuridine.

Intermediate from previous reaction (50 mg, 0.070 mmol) was placed into a round bottom flask and purged with argon for 10 minutes. Anhydrous acetonitrile (10 mL) and magnesium perchlorate (2 mg, 0.009 mmol) were added, and the reaction mixture was stirred at reflux for 2.5 hours under argon atmosphere. The solvent was removed under reduced pressure; the crude product was dissolved in ethyl acetate (1 mL) and mixed with silica (ca 500 mg). The solvent was evaporated, and the powder was applied onto a chromatography column (SiO2, hexane/ethyl acetate = 8:1 to 4:1) to afford 22 mg (51%) of 6. 1H NMR (500 MHz, CDCl3) δ 8.36 (s, 1 H),

7.50 (s, 1 H), 6.31 (dd, 1 H, J = 7.7, 5.9 Hz), 4.41 (s, 2 H), 4.38 (m, 1 H), 3.95 (m, 1 H), 3.77

(AB dd, 1 H, J = 10.6, 4.7 Hz), 3.56 (AB dd, 1 H, J = 10.6, 7.0 Hz), 2.80 (s, 1 H), 2.32 (m, 1 H),

1.90 (m, 1 H), 1.03 (s, 18 H), 0.90 and 0.89 (2 s, 18 H), 0.09 and 0.07 (2 s, 12 H). 13C NMR

(125 MHz, CDCl3) δ 163.43 (C), 149.85 (C), 135.41 (CH), 107.73 (C), 96.13 (CH), 87.64 (CH),

105

85.31 (CH), 72.69 (CH), 68.70 (CH2), 63.58 (CH2), 40.04 (CH2), 38.75 (C), 29.45 (CH3), 29.20

(CH3), 25.92 (CH3), 17.98 (C), -4.70 (CH3), -5.39 (CH3).

5-[1-(4-[2-phenylacetylenyl]-2-nitrophenyl)-2,2-(dimethyl)-1-propoxymethyl]-2'-deoxyuridine

(3a). Treatment of 38 mg (0.067 mmol) of 2v with phenylacetylene (21 mg, 0.201 mmol) followed by purification using ethyl acetate/methanol = 20:1 system yielded 12 mg (33%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ

8.02 and 8.00 (2 s, 1 H), 7.93 (m, 1 H), 7.77 (m, 2 H), 7.55 (m, 2 H), 7.39 (m, 2 H), 6.27 (m, 1

H), 4.95 and 4.94 (2 s, 1 H), 4.42 (m, 1 H), 4.19 (m, 2 H), 3.93 (m, 1 H), 3.76 (m, 2 H), 2.25 (m,

2 H), 0.84 and 0.83 (2 s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers: δ 164.97 and

164.91 (C), 152.07 (C), 151.95 (C), 141.44 and 141.02 (CH), 135.58 and 135.55 (CH), 135.34 and 135.31 (C), 132.75 (CH), 131.71 and 131.62 (CH), 130.14 (CH), 129.64 (CH), 127.52 and

127.42 (CH), 125.00 (C), 123.58 (C), 112.27 and 112.02 (C), 92.61 (C), 88.96 (CH), 87.63 (C),

86.61 and 86.50 (CH), 83.06 and 82.47 (CH), 72.35 and 72.29 (CH), 65.97 and 65.70 (CH2),

62.87 and 62.83 (CH2), 41.31 and 41.23 (CH2), 37.66 and 37.57 (C), 26.24 (CH3). HRMS

(ESI+) for [MH]+ C29H32N3O8 calculated: 550.21856, observed: 550.21839; for [MNa]+

C29H31N3O8Na calculated: 572.20034, observed: 572.20050. HRMS (ESI+) for [MH]+

C29H32N3O8 calculated: 550.21856, observed: 550.21839; for [MNa]+ C29H31N3O8Na calculated: 572.20034, observed: 572.20050. HRMS (ESI-) for [M-H]- C29H30N3O8 calculated: 548.20384, observed: 548.20335; for [MCl]- C29H31N3O835Cl calculated:

584.18052, observed: 584.18052; for C29H31N3O837Cl calculated: 586.17914, observed:

586.17754.

5-[1-(4-(trimethylsilyl)acetylenyl-2-nitrophenyl)-2,2-(dimethyl)-1-propoxymethyl]-2'- deoxyuridine (3d). Treatment of 87 mg (0.152 mmol) of 2v with (trimethylsilyl)acetylene (60

106

mg, 0.607 mmol) followed by purification using ethyl acetate/methanol = 1:0 to 100:1 system yielded 32 mg (38%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.01 and 8.00 (2 s, 1 H), 7.84 (m, 1 H), 7.74 (AB d, J= 8.1 Hz, 1 H), 7.68

(m, 1 H), 6.26 (m, 1 H), 4.93 and 4.92 (2 s, 1 H), 4.40 (m, 1 H), 4.18 (m, 2 H), 3.92 (m, 1 H),

3.74 (m, 2 H), 2.24 (m, 2 H), 0.83 and 0.82 (2 s, 9 H), 0.26 (s, 9 H). 13C NMR (100 MHz,

CD3OD) for diastereomers: δ 165.00 and 164.95 (C), 152.10 and 152.00 (C), 151.86 (C), 141.14 and 141.14 (CH), 135.86 and 135.75 (CH), 131.73 and 131.63 (CH), 129.91 (C), 127.84 and

127.74 (CH), 124.83 (C), 112.22 and 111.98 (C), 103.33 (C), 97.79 (C), 89.01 (CH), 86.61 and

86.51 (CH), 83.02 and 82.42 (CH), 72.38 and 72.31 (CH), 66.00 and 65.70 (CH2), 62.88 and

62.84 (CH2), 41.33 and 41.25 (CH2), 37.66 and 37.56 (C), 26.20 (CH3), -0.25 (CH3). HRMS

(ESI) for [MH]+ C26H36N3O8Si calculated: 546.22717, observed: 546.22670; [MNa]+

C26H35N3O8SiNa calculated: 568.20911, observed: 568.20911.

5-[1-(4-acetylenyl-2-nitrophenyl)-2,2-(dimethyl)-1-propoxymethyl]-2'-deoxyuridine (3e).

Compound 3d (30 mg, 0.055 mmol) was dissolved in tetrahydrofuran (2.5 mL) followed by addition of tetra-n-butylammonium fluoride trihydrate (26 mg, 0.083 mmol). The reaction mixture was stirred for 6 hours, then concentrated under reduced pressure, and the residue was purified by column chromatography on silica gel using ethyl acetate/methanol = 1:0 to 40:1 to afford 10 mg (38%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.02 and 8.00 (2 s, 1 H), 7.90 (m, 1 H), 7.76 (AB d, J= 8.2 Hz, 1 H), 7.65

(m, 1 H), 6.26 (m, 1 H), 4.94 and 4.93 (2 s, 1 H), 4.40 (m, 1 H), 4.18 (m, 2 H), 3.92 (m, 1 H),

3.74 (m, 3 H), 2.24 (m, 2 H), 0.84 and 0.82 (2 s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers: δ 163.60 and 163.55 (C), 150.69 and 150.61 (C), 150.48 (C), 140.01 and 139.91

(CH), 134.73 and 134.52 (CH), 130.94 and 130.26 (CH), 128.46 (C), 126.71 and 126.61 (CH),

107

122.79 (C), 110.82 and 110.57 (C), 87.61 (CH), 85.20 and 85.10 (CH), 81.61 and 81.01 (CH),

80.48 (C), 80.17 (CH), 70.97 and 70.90 (CH), 64.57 and 64.30 (CH2), 61.47 and 61.43 (CH2),

39.93 and 39.86 (CH2), 36.93 and 36.14 (C), 24.78 (CH3). HRMS (ESI) for [MH]+

C23H28N3O8 calculated: 474.18764, observed: 474.18715; [MNa]+ C23H27N3O8Na calculated: 496.16958, observed: 496.16915.

5-[1-(4-(1-benzyl-1,2,3-triazo-4-yl)-2-nitrophenyl)-2,2-(dimethyl)propoxymethyl]-2'- deoxyuridine (4). Compound 3e (3.8 mg, 0.008 mmol) was dissolved in acetonitrile (2 mL) followed by addition of benzyl azide (2 mg, 0.014 mmol), diisopropylethylamine (10 mg, 0.08 mmol) and copper(I) iodide (0.1 mg, 0.0008 mmol). The reaction mixture was stirred for 2 hours under argon atmosphere at room temperature. The reaction mixture was then concentrated under reduced pressure and the residue was purified by column chromatography on silica gel using dichloromethane/methanol = 1:0 to 30:1 to afford 3.7 mg (76%) of product as 1:1 mixture of diastereomers. 1H NMR (400 MHz, CD3OD) for diastereomers: δ 8.49 (s, 1 H), 8.27 (m, 1H),

8.07 (m, 1 H), 8.01 and 7.99 (2 s, 1 H), 7.82 (d, 1 H, J= 8.3 Hz), 7.37 (m, 5 H), 6.24 (m, 1 H),

5.66 (s, 2 H), 4.96 and 4.95 (2 s, 1 H), 4.39 (m, 1 H), 4.20 (m, 2 H), 3.91 (m, 1 H), 3.73 (m, 2 H),

2.24 (m, 2 H), 0.85 and 0.84 (2 s, 9 H). 13C NMR (100 MHz, CD3OD) for diastereomers: δ

165.03 and 164.96 (C), 152.63 and 152.48 (C), 152.09 (C), 146.82 (C), 141.34 and 141.05 (CH),

136.62 (C), 134.85 (C), 134.42 and 134.40 (C), 132.16 and 132.06 (CH), 133.10 (CH), 129.75

(CH), 129.69 (CH), 129.19 (CH), 123.32 (CH), 121.67 and 121.55 (CH), 112.32 and 112.06 (C),

88.97 (CH), 86.48 and 86.36 (CH), 83.11 and 82.47 (CH), 72.36 and 72.32 (CH), 65.98 and

65.67 (CH2), 62.90 and 62.84 (CH2), 55.18 (CH2), 41.29 and 41.20 (CH2), 37.63 and 37.53

(CH), 26.26 (CH3). HRMS (ESI) for [MH]+ C30H35N6O8 calculated: 607.25164, observed:

607.25119; for [MNa]+ C30H34N6O8Na calculated: 629.23358, observed: 629.23311.

108

5-[(R/S)-1-(4-(2-phenylacetylenyl)-2-nitrophenyl)-2,2-(dimethyl)-1-propoxymethyl]-2'- deoxyuridine-5’-triphosphate (3aTP) Standard procedure was used as previously described.

POCl3 (15 μL, 0.163 mmol) was added to a solution of 3a (19 mg, 0.035 mmol) and proton sponge (30 mg, 0.140 mmol) in trimethylphosphate (0.7 mL) at 0 C and stirred for two hours under argon atmosphere. Reaction progress was monitored by HPLC and reverse-phase TLC

(C18). Additional POCl3 (10 μL, 0.111 mmol) was added, and the mixture was stirred for another one hour. A solution of tri‐n‐butylammonium pyrophosphate (200 mg, 0.366 mmol) and tri‐n‐butylamine (95 μL) in anhydrous DMF (1.0 mL) was added. After five minutes of stirring, triethylammonium bicarbonate buffer (1 M, pH 7.5; 8 mL) was added and then stirred at room temperature for one hour. The reaction was lyophilized to dryness, and the residue was dissolved in water (5 mL), filtered, and purified by reverse phase chromatography using triethylammonium bicarbonate buffer as eluent. Fractions containing the triphosphate (identified by HRMS) were combined and lyophilized to give product in ca 4 mg (10%) as a solid. 1H NMR

(400 MHz, D2O) for diastereomers: δ 7.90 and 7.89 (2 s, 1 H), 7.82 (m, 1 H), 7.73 (m, 2 H), 7.52

(m, 2 H), 7.40 (m, 2 H), 6.21 (m, 1 H), 4.88 and 4.88 (2 s, 1 H), 4.38 (m, 1 H), 4.29 (m, 2 H),

3.92 (m, 1 H), 3.72 (m, 2 H), 2.53 (m, 24 H), 1.44 (m, 24 H), 1.27 (m, 24 H), 0.89 (t, 36 H, J =

7.3 Hz), 0.79 and 0.77 (2 s, 9 H). 31P NMR (162 MHz, D2O): δ ‐3.53 (d, J = 17.8 Hz), ‐14.63

(d, J = 17.8 Hz), ‐25.32 (m). HRMS (ESI-TOF): For [M‐H]‐ C29H33N3O17P3 calculated:

788.10283, observed: 788.10272.

109

5.3 Biochemistry experiments

5.3.1 Cell cytotoxicity assay (MTT)

MCF7 cells were grown in RPMI 1640 media supplemented with 10% fetal bovine serum, 1% penicillin, 10 nM estrogen and 1 mM insulin. Cells were tyripsinized and resuspended at a density of 2.2 x 104 cells per mL. 500 L of this suspension was added to each well in a 24 well plate. The plates were incubated at 37 C and 5% CO2 atmosphere overnight.

The media was changed, and plates were dosed in triplicate with compound dissolved in DMSO

(DMSO concentration not exceeding 0.5%). Cells were dosed to a final concentration of 200,

100, 50, 25, and 6.25 M of compound. 5-Flurouracil was used as a positive control and dosed in the same manner. (Hernández-Vargas, Ballestar et al. 2006) Plates were incubated for 65 hours prior to the addition of MTT solution. 500 L of a 193g/mL MTT and media solution was added to each well. Plates were incubated for 3 hours. The MTT solution was removed and

500 L of DMSO was added to each well. Cells were imaged using GS 800 Bio Rad scanner with Quality One Software or BioTek plate reader with Gene5 software. IC50 curves were determined by plotting viability verses compound concentration. Kaleidagraph software was used to calculate the R value for each logarithmic curve fitting. The results are outlined in Table

5.1.

5.3.2 DNA synthesis termination studies

Primer extension experiments were performed using a 51-mer nucleotide dsDNA, synthesized from pUC19 plasmid vector (New England Biolabs). A 19-nucleotide forward

110

Table 5.1 IC50 values determined by MTT assays using base-modified T-nucleoside analogs and MCF7 breast cancer cells

primer, 5’CACGACGTTGTAAAACGAC3’ was used as a template

5’CACGACGTTGTAAAACGACGGCCAGTGAATTCGAGCTCGGTACCCGGGGAT3’

(representing 370-420 of pUC 19). The primer was fluorescently-labeled with IRDye700. The oligonucleotide was purified using a Cycle Pure Kit (Omega BioTek). Master mix was made at

2X with a concentration of final in PCR tube reaction of 0.5 μM template, 1X ThermoPol

Reaction Buffer, 0.05 U/μL Vent (exo-), and 0.75 μM primer. 100 μM dNTP with varying concentration of Acylco TTP and 3a triphosphate were added with the master mix. Sequencing reactions were conducted using 10 µL of a solutions containing acyclo-dNTP and dNTP having a

111

final PCR reaction concentration of 2 mM and 100 μM, respectively, followed by the addition of

10µl of master mix. For sequencing the labels on the gel correspond to the sequence produced by the 51-mer template. 10 μL PCR product was mixed with 17 μL PAGE denaturing load dye.

After heating at 95 C for 5 min, 5 μL of the resulting solution was loaded into each well containing a 12% denaturing PAGE gel, which was subsequently run at a constant 18 Watts for

35 minutes. The gel was visualized using the Odyssey Infrared Imaging System (LiCor) with

169 μm resolution and the 700-channel laser source which has a solid-state laser diode at 680 nm and ImageQuant 5.0 software was used to determine density measurements. Experiments were performed in triplicate, and standard errors were calculated.

Figure 5.1 Synthesis of T-nucleoside analogs: Appropriate alcohol, 110-124 C, 1-3 h; (ii) n-Bu4NF, THF, 0 C to r.t., 2-6 h; (iii) Pd(PPh3)4, appropriate terminal alkyne, CuI, Et3N, DMF, r.t., 6-18 h; (iv) Benzyl azide, CuI, Et3N, MeCN, r.t., 4 h.

112

5.4 Results/discussion

5.4.1 Synthesis

The bioactive compounds were obtained by heating 5-bromomethyl-3’,5’-bis-(tert- butyl)dimethylsilyl-O-2’-deoxyuridine (Litosh, Wu et al. 2011) with an appropriate neat alcohol in neat, anhydrous conditions. Formation of HBr as a byproduct resulting in the in situ elimination of the N3-Boc group, but also in side reactions that made the purification challenging, which at times resulted in low yields. Removal of the residual TBS groups using tetra-n-butyl-ammonium fluoride yielded 2a-w (Figure 5.1). From our previous studies, (Litosh,

Wu et al. 2011) the attachment of a large group at the para-position of the benzyl ring was known to further improve the DNA synthesis termination properties of the base-modified nucleotides. Consequently, the 4-iodo-2-nitrobenzyl derivative 2v was used for Sonogashira coupling to various terminal alkynes to form derivatives 3a-d. The coupling product with

(trimethylsilyl)acetylene (3d) was exposed to tetra-n-butylammonium fluoride to generate the acetylenyl derivative 3e, which was then “clicked” to benzyl azide to yield compound 4.

5.4.2 Cytotoxicity in MCF7 breast cancer cells: elucidation of structure-activity relationship

The MTT bioassay (Bell-Horwath, Vadukoot et al. 2013) results for the derivatives are summarized in Table 5.1, and they reveal four important trends in the SAR. First, the presence of at least one phenyl group as R1 or R2 required, as neither the bis-isopropyl (2w) nor the bis- tert-butyl (2x) showed significant activity. Second, the presence of one nitro group on the phenyl

113

ring is critical for the anti-cancer activity, similarly to the bulkiness of the substituent attached to the benzylic α-carbon. Consequently, the IC50 values of the non-substituted phenyl derivatives

2d-g are generally much higher compared to the 2-nitro substituted analogs 2p-s. At the same time, the presence of a 2-nitro group in 2m-o does not increase the activities of the smaller α- substituted analogs 2a-c. An activity decrease was observed when the α-tert-butyl group in the first generation lead compound o-2s is replaced with a less bulky (2p, o-2q) or moved away from the benzylic α-carbon by just one CH₂ unit (2r), which is consistent with DNA synthesis terminating properties of their 5’-triphosphates. (Litosh, Wu et al. 2011) Furthermore, neither the replacement of the 2-nitro group with different substituents such as methyl (2h), cyano (2i), halo

(2j-l), methoxy (o-2u), nor the introduction of another ortho-nitro group (2t) improved activity.

Third, the electronic character of the aromatic substituents and their position on the aromatic ring appeared to be related. Therefore, for electron-donating groups in the benzene ring, for example, the methoxy, in the ortho- (o-2s, o-2q) or para- (p-2q) nitro-substituted derivatives are substantially more active than their meta- counterparts (m-2q, m-2s). Fourth, derivatization of the initial lead compound o-2s by the attachment of a large group at the benzyl ring para-position via an acetylene linker substantially improves the IC50 values, with the (phenyl)alkynyl derivative 3a was identified as the second generation lead compound.

5.4.3 Evaluation of selectivity for novel bioactive compound 3a

To evaluate selectivity of the lead compound 3a, we assessed its toxicity using normal fibroblast cells and compared it to that in MCF7 breast cancer cells (Figure 5.2) The IC50 value for fibroblast cells was 55 ± 8 μM, showing a selectivity ratio of 6.3 ± 1.6, whereas the

114

v i

100t

t i

l 60

l l

e 20 C 1 10 100 [3a], uM

Figure 5.2 Selectivity of 3a. Viability of MCF7 cancer cells (black) and fibroblasts (grey) in the presence of 3a. Selectivity is greater than 5-fold

selectivity of 5-fluorouracil, the FDA-approved drug used for treating breast cancer, was 1.8 ±

0.5. Therefore, our second-generation lead compound 3a could potentially have a significantly wider therapeutic window than the current chemotherapeutic drug.

5.4.4 Evaluation of activity for novel bioactive compound 3a in other cancer cells

The NCI-60 human tumor cell line screen based on the SRB assay gave a growth inhibition of 58% for compound 3a at 10 μM in MCF7 cells, which is consistent with the IC50 value of 9 ± 1 μM obtained from the MTT assay. Additionally, the lead compound also showed significant activity in leukemia, prostate, renal, melanoma, central nervous system, and non- small lung cancer cell lines (Table 5.2)

115

Table 5.2 Percent growth inhibition of 3a at 10 μM in varied cell lines

116

Figure 5.3 Primer extension assay studies of 3a triphosphate. Left to right: no polymerase, 3aTP at 0, 0.125, 0.25, 0.5, 1, 2 mM terminating nucleotide, acyclo A, C, G (2 mM each), acylco T at 2, 1, 0.5, 0.125 mM

5.4.5 DNA synthesis termination by 5’-triphosphate of 3a

To gain insight into the mechanism of action of 3a, we synthesized its 5’-triphosphate

(3aTP) as described in our previous work and examined its incorporation into DNA under conditions approximating the intracellular environment. (Litosh, Wu et al. 2011, Vadukoot,

AbdulSalam et al. 2014) I performed a primer extension assay using exo-Vent polymerase, a random 51-mer template, a fluorescently labeled primer, a 100μM cocktail of all four natural dNTPs, and various concentrations of acyclo-NTPs as positive control of 3aTP. Following extension (Figure 5.3), it is obvious that a 3aTP inhibits DNA polymerase rather than terminates

DNA synthesis as evidenced by the fact that acylo-TTP showed termination at nucleotide positions 26, 30, 31, 37, and 41, while termination was not observed for 3aTP. Furthermore, primer extension was halted entirely when the concentration of 3aTP was increased from 1 to

2mM, but this was not observed for any of the acyclo-nucleotide triphosphates at these concentrations.

117

5.5 Conclusion

I have synthesized a library of thymidine derivatives bearing a modifying moiety attached at 5-methyl group. Studies of the structure-activity relationship regarding the cytotoxicity of these base-modified T-nucleosides in MCF7 cancer cells have revealed a lead compound. The

DNA damage signaling elicited by the active nucleoside 3a is consistent with its cellular uptake and 5’-triphosphorylation into the active species 3aTP that interferes with DNA synthesis. PCR studies using 3aTP support a mechanism of action that inhibits DNA polymerase rather than being incorporated into the DNA replication fork and blocking nucleotide addition as initially hypothesized, which warrants future studies of polymerase activity affected by these species.

Importantly, this novel anti-cancer bioactive compound is less toxic to normal cell lines compared to the FDA-approved T-nucleobase analog 5-fluorouracil, which is currently used against breast cancer. Furthermore, we have produced an analog (3e) that can undergo click reaction with various azides, facilitating the synthesis of large libraries of diverse triazole compounds that are analogous to 4. This opportunity offers: (a) the potential for further improvement of cytotoxic activity, particularly in the light of the positive influence of a large substituent when attached at the para-position of the benzene ring, and (b) the attachment of imaging modalities for further investigation of intracellular metabolism.

This work was not continued when Dr. Litosh left the University of Cincinnati and I joined

Dr. Limbach’s research group.

118

6 REFERENCES

Abad, M. G., B. S. Rao and J. E. Jackman (2010). "Template-dependent 3'-5' nucleotide addition is a shared feature of tRNAHis guanylyltransferase enzymes from multiple domains of life." Proc Natl Acad Sci U S A 107(2): 674-679. Addepalli, B., N. P. Lesner and P. A. Limbach (2015). "Detection of RNA nucleoside modifications with the uridine-specific ribonuclease MC1 from Momordica charantia." Rna 21(10): 1746-1756. Addepalli, B., S. Venus, P. Thakur and P. A. Limbach (2017). "Novel ribonuclease activity of cusativin from Cucumis sativus for mapping nucleoside modifications in RNA." Anal Bioanal Chem 409(24): 5645-5654. Akerblom, L., E. Pontis and P. Reichard (1982). "Effects of azidocytidine on DNA synthesis and deoxynucleotide pools of mouse fibroblast cell lines." Journal of Biological Chemistry 257(12): 6776-6782. Alrabiah, F. and S. Sacks (1996). "New Antiherpesvirus Agents." Drugs 52(1): 17-32. Ando, M., T. Watanabe, Y. Sasaki, D. F. Ying, Y. Omuro, N. Katsumata, M. Narabayashi, Y. Tokue, H. Fujii, T. Igarashi, H. Wakita, T. Ohtsu, K. Itoh, I. Adachi and T. Taguchi (1998). "A phase I trial of docetaxel and 5-day continuous infusion of 5-fluorouracil in patients with advanced or recurrent breast cancer." Br J Cancer 77(11): 1937-1943. Arce, C., B. Segura-Pacheco, E. Perez-Cardenas, L. Taja-Chayeb, M. Candelaria and A. Duennas-Gonzalez (2006). "Hydralazine target: From blood vessels to the epigenome." Journal of Translational Medicine 4(1): 10. Averett, D. R., G. W. Koszalka, J. A. Fyfe, G. B. Roberts, D. J. Purifoy and T. A. Krenitsky (1991). "6-Methoxypurine arabinoside as a selective and potent inhibitor of varicella-zoster virus." Antimicrobial Agents and Chemotherapy 35(5): 851-857. Azuma, A., P. Huang, A. Matsuda and W. Plunkett (2001). "Cellular pharmacokinetics and pharmacodynamics of the deoxycytidine analog 2′-C-cyano-2′-deoxy-1-β-d-arabino- pentofuranosylcytosine (CNDAC)." Biochemical Pharmacology 61(12): 1497-1507. Balzarini, J., A. Karlsson, L. Wang, C. Bohman, K. Horská, I. Votruba, A. Fridland, A. Van Aerschot, P. Herdewijn and E. De Clercq (1993). "Eicar (5-ethynyl-1-beta-D- ribofuranosylimidazole-4-carboxamide). A novel potent inhibitor of inosinate dehydrogenase activity and guanylate biosynthesis." Journal of Biological Chemistry 268(33): 24591-24598. Barker, S. and R. Norris (1983). "Nucleophilic substitution reactions in m-nitrobenzylic substrates." Australian Journal of Chemistry 36(1): 81-95. Bell-Horwath, T. R., A. K. Vadukoot, F. S. Thowfeik, G. Li, M. Wunderlich, J. C. Mulloy and E. J. Merino (2013). "Novel ROS-activated agents utilize a tethered amine to selectively target acute myeloid leukemia." Bioorganic & Medicinal Chemistry Letters 23(10): 2951-2954. Bellodi, C., M. McMahon, A. Contreras, D. Juliano, N. Kopmar, T. Nakamura, D. Maltby, A. Burlingame, Sharon A. Savage, A. Shimamura and D. Ruggero (2013). "H/ACA Small RNA

119

Dysfunctions in Disease Reveal Key Roles for Noncoding RNA Modifications in Hematopoietic Stem Cell Differentiation." Cell Reports 3(5): 1493-1502. Bentzley, C. M., M. V. Johnston, B. S. Larsen and S. Gutteridge (1996). "Oligonucleotide Sequence and Composition Determined by Matrix-Assisted Laser Desorption/Ionization." Analytical Chemistry 68(13): 2141-2146. Berhane, B. T. and P. A. Limbach (2003). "Stable isotope labeling for matrix-assisted laser desorption/ionization mass spectrometry and post-source decay analysis of ribonucleic acids." J. Mass Spectrom. 38: 872-878. Besirli, C. G., T. L. Deckwerth, R. J. Crowder, R. S. Freeman and E. M. Johnson, Jr. (2003). "Cytosine arabinoside rapidly activates Bax-dependent apoptosis and a delayed Bax-independent death pathway in sympathetic neurons." Cell Death Differ 10(9): 1045-1058. Best, M. D. (2009). "Click Chemistry and Bioorthogonal Reactions: Unprecedented Selectivity in the Labeling of Biological Molecules." Biochemistry 48(28): 6571-6584. Bickelhaupt, F., C. Jongsma, P. de Koe, R. Lourens, N. R. Mast, G. L. van Mourik, H. Vermeer and R. J. M. Weustink (1976). "Synthesis and application of precursors of hetero-anthracenes: 2- halo- and 2,2′-dihalodiphenylmethanes and methine substituted derivatives." Tetrahedron 32(15): 1921-1930. Bocci, G., A. Fioravanti, P. Orlandi, N. Bernardini, P. Collecchi, M. Del Tacca and R. Danesi (2005). "Fluvastatin synergistically enhances the antiproliferative effect of gemcitabine in human pancreatic cancer MIAPaCa-2 cells." British Journal Of Cancer 93(3): 319-330. Bollum, F. J. (1964). "Chemically Defined Templates and Initiators for Deoxypolynucleotide Synthesis." Science 144(3618): 560. Bontemps, F., A. Delacauw, S. Cardoen, E. Van Den Neste and G. Van Den Berghe (2000). "Metabolism and cytotoxic effects of 2-chloroadenine, the major catabolite of 2-chloro-2′- deoxyadenosine." Biochemical Pharmacology 59(10): 1237-1243. Borland, K. and P. A. Limbach (2017). "Applications and Advantages of Stable Isotope Phosphate Labeling of RNA in Mass Spectrometry." Top Curr Chem (Cham) 375(2): 33. Borland, K. M., S. F. AbdulSalam, M. J. Solivio, M. P. Burke, P. R. Wolfkiel, S. M. Lawson, C. A. Stockman, J. M. Andersen, S. Smith, J. N. Tolstolutskaya, P. N. Gurjar, A. P. Bercz, E. J. Merino and V. A. Litosh (2015). "Base-modified thymidines capable of terminating DNA synthesis are novel bioactive compounds with activity in cancer cells." Bioorganic & Medicinal Chemistry 23(8): 1869-1881. Borland, K. M., M. P. Burke, S. M. Lawson, S. Ventura, B. Zborovska, E. J. Merino and V. A. Litosh (2014). Base-Modified Nucleotide Terminators of DNA Synthesis as Novel Anti-Cancer Agents. Abstracts of Papers, 247th ACS National Meeting & Exposition, Dallas, TX, United States, March 16-20, 2014, Dallas, TX, United States Borland, K. M., S. M. Lawson, S. Ventura, E. J. Merino and V. A. Litosh (2014). Novel modified nucleobases that show cytotoxicity towards breast cancer cells. Abstracts of Papers, 248th ACS National Meeting & Exposition, San Francisco, CA, United States, August 10-14, 2014 San Francisco, CA. Bouchard, J. and R. L. Momparler (1983). "Incorporation of 5-Aza-2'-deoxycytidine-5'- triphosphate into DNA. Interactions with mammalian DNA polymerase alpha and DNA methylase." Molecular Pharmacology 24(1): 109-114. Bourgine, J., A. Garat, D. Allorge, A. Crunelle-Thibaut, J.-M. Lo-Guidice, J.-F. Colombel, F. Broly and I. Billaut-Laden (2011). "Evidence for a functional genetic polymorphism of the Rho-

120

GTPase Rac1. Implication in azathioprine response?" Pharmacogenetics and Genomics 26(6): 313–324. Brockman, R. W., Y.-C. Cheng, F. M. Schabel and J. A. Montgomery (1980). "Metabolism and Chemotherapeutic Activity of 9-β-d-Arabinofuranosyl-2-fluoroadenine against Murine Leukemia L1210 and Evidence for Its Phosphorylation by Deoxycytidine Kinase." Cancer Research 40(10): 3610-3615. Bruce, A. G. and O. C. Uhlenbeck (1978). "Reactions at the termini of tRNA with T4 RNA ligase." Nucleic Acids Research 5(10): 3665-3677. Brückl, T., D. Globisch, M. Wagner, M. Müller and T. Carell (2009). "Parallel Isotope-Based Quantification of Modified tRNA Nucleosides." Angew Chem Int Ed Engl 48(42): 7932-7934. Brulíková, L., P. Džubák, M. Hajdúch and J. Hlaváč (2011). "Synthesis of various 5- alkoxymethyluracil analogues and structure–cytotoxic activity relationship study." Carbohydrate Research 346(14): 2136-2144. Burke, M. P., K. M. Borland, P. N. De Salvo, S. M. Lawson, J. N. Tolstolutskaya, E. J. Merino, R. J. Kirby, R. Papoian and V. A. Litosh (2013). Novel nucleoside and nucleobase anticancer prodrug candidates: Synthesis and investigation of their biological activity in human cancer cell lines. Abstracts of Papers, 246th ACS National Meeting & Exposition, Indianapolis, IN, United States, September 8-12, 2013, Indianapolis, IN, United States. Burke, M. P., K. M. Borland and V. A. Litosh (2016). "Base-Modified Nucleosides as Chemotherapeutic Agents: Past and Future." Curr Top Med Chem 16(11): 1231-1241. Cabrita, M. A., S. A. Baldwin, J. D. Young and C. E. Cass (2002). "Molecular biology and regulation of nucleoside and nucleobase transporter proteins in eukaryotes and prokaryotes." Biochemistry and Cell Biology 80(5): 623-638. Calvo, C. H. T. E. (2008). Principles of oncologic pharmacotherapy, Wiley and Sons. Canman, C. E., H. Y. Tang, D. P. Normolle, T. S. Lawrence and J. Maybaum (1992). "Variations in patterns of DNA damage induced in human colorectal tumor cells by 5-fluorodeoxyuridine: implications for mechanisms of resistance and cytotoxicity." Proceedings of the National Academy of Sciences 89(21): 10474-10478. Cannas, G., C. Pautas, E. Raffoux, B. Quesnel, S. de Botton, T. de Revel, O. Reman, C. Gardin, M. Elhamri, N. Boissel, P. Fenaux, M. Michallet, S. Castaigne, H. Dombret and X. Thomas (2012). "Infectious complications in adult acute myeloid leukemia: analysis of the Acute Leukemia French Association-9802 prospective multicenter clinical trial." Leuk Lymphoma 53(6): 1068-1076. Capasso, C. and C. T. Supuran (2014). "Sulfa and trimethoprim-like drugs - antimetabolites acting as carbonic anhydrase, dihydropteroate synthase and dihydrofolate reductase inhibitors." Journal of Enzyme Inhibition & Medicinal Chemistry 29(3): 379-387. Cappellacci, L., R. Petrelli, P. Franchetti, P. Vita, P. Kusumanchi, M. Kumar, H. N. Jayaram, B. Zhou, Y. Yen and M. Grifantini (2011). "Synthesis and biological activity of novel N6- substituted and 2,N6-disubstituted adenine ribo- and 3′-C-methyl-ribonucleosides as antitumor agents." European Journal of Medicinal Chemistry 46(5): 1499-1504. Carson, D., D. Wasson, R. Taetle and A. Yu (1983). "Specific toxicity of 2- chlorodeoxyadenosine toward resting and proliferating human lymphocytes." Blood 62(4): 737- 743. Carson, D. A., D. B. Wasson, L. M. Esparza, C. J. Carrera, T. J. Kipps and H. B. Cottam (1992). "Oral antilymphocyte activity and induction of apoptosis by 2-chloro-2'-arabino-fluoro-2'- deoxyadenosine." Proceedings of the National Academy of Sciences 89(7): 2970-2974.

121

Carson, D. A., D. B. Wasson, J. Kaye, B. Ullman, D. W. Martin, R. K. Robins and J. A. Montgomery (1980). "Deoxycytidine kinase-mediated toxicity of deoxyadenosine analogs toward malignant human lymphoblasts in vitro and toward murine L1210 leukemia in vivo." Proceedings of the National Academy of Sciences 77(11): 6865-6869. Castleberry, C. M. and P. A. Limbach (2010). "Relative Quantitation of Transfer RNAs Using Liquid Chromatography-Mass Spectrometry (LC-MS) and Signature Digestion Products." Nucleic Acids Res 38(16): e162. Ceruti, S., A. Mazzola, E. Beltrami, D. Passera, E. Piantoni, F. Cattabeni and M. P. Abbracchio (2003). "Intracellular phosphorylation of chloro-adenosine analogs is a prerequisite for activation of caspase-3 and induction of apoptosis in human astrocytoma cells." Drug Development Research 58(4): 396-404. Chan, C. T. Y., M. Dyavaiah, M. S. DeMott, K. Taghizadeh, P. C. Dedon and T. J. Begley (2010). "A Quantitative Systems Approach Reveals Dynamic Control of tRNA Modifications during Cellular Stress." PLOS Genetics 6(12): e1001247. Chan, C. T. Y., Y. L. J. Pang, W. Deng, I. R. Babu, M. Dyavaiah, T. J. Begley and P. C. Dedon (2012). "Reprogramming of tRNA modifications controls the oxidative stress response by codon-biased translation of proteins." Nature Communications 3: 937. Chang, S.-l., G. W. Griesgraber, P. J. Southern and C. R. Wagner (2001). "Amino Acid Phosphoramidate Monoesters of 3‘-Azido-3‘-deoxythymidine: Relationship between Antiviral Potency and Intracellular Metabolism." Journal of Medicinal Chemistry 44(2): 223-231. Chaudhuri, N. K., B. J. Montag and C. Heidelberger (1958). "Studies on Fluorinated Pyrimidines: III. The Metabolism of 5-Fluorouracil-2-C14 and 5-Fluoroorotic-2-C14 Acid in Vivo." Cancer Research 18(3): 318-328. Chen, L. S., W. Plunkett and V. Gandhi (2008). "Polyadenylation inhibition by the triphosphates of deoxyadenosine analogues." Leukemia research 32(10): 1573-1581. Cho, S.-J., S.-S. Lee, Y.-J. Kim, B.-D. Park, J.-S. Choi, L. Liu, Y.-M. Ham, B. Moon Kim and S.-K. Lee (2010). "Xylocydine, a novel Cdk inhibitor, is an effective inducer of apoptosis in hepatocellular carcinoma cells in vitro and in vivo." Cancer Letters 287(2): 196-206. Clouser, C. L., C. M. Holtz, M. Mullett, D. L. Crankshaw, J. E. Briggs, M. G. O'Sullivan, S. E. Patterson and L. M. Mansky (2012). "Activity of a Novel Combined Antiretroviral Therapy of Gemcitabine and Decitabine in a Mouse Model for HIV-1." Antimicrobial Agents and Chemotherapy 56(4): 1942-1948. Cohen, M. H., J. R. Johnson, T. Massie, R. Sridhara, W. D. McGuinn, S. Abraham, B. P. Booth, M. A. Goheer, D. Morse, X. H. Chen, N. Chidambaram, L. Kenna, J. V. Gobburu, R. Justice and R. Pazdur (2006). "Approval Summary: Nelarabine for the Treatment of T-Cell Lymphoblastic Leukemia/Lymphoma." Clinical Cancer Research 12(18): 5329-5335. Cohen, S. S. (1971). "On the Nature of Thymineless Death." Annals of the New York Academy of Sciences 186(1): 292-301. Cooper, T., H. Kantarjian, W. Plunkett and V. Gandhi (2004). "Clofarabine in adult acute leukemias: clinical success and pharmacokinetics." Nucleosides, Nucleotides & Nucleic Acids 23(8-9): 1417-1423. Cordier, P. Y., A. Nau, J. Ciccolini, M. Oliver, C. Mercier, B. Lacarelle and E. Peytel (2011). "5- FU-induced neurotoxicity in cancer patients with profound DPD deficiency syndrome: a report of two cases." Cancer Chemother Pharmacol 68(3): 823-826. Cottam, H. B., Z. Kazimierczuk, S. Geary, P. A. McKernan, G. R. Revankar and R. K. Robins (1985). "Synthesis and biological activity of certain 6-substituted and 2,6-disubstituted 2'-

122

deoxytubercidins prepared via the stereospecific sodium salt glycosylation procedure." Journal of Medicinal Chemistry 28(10): 1461-1467. Crimmins, M. T. and B. W. King (1996). "An Efficient Asymmetric Approach to Carbocyclic Nucleosides: Asymmetric Synthesis of 1592U89, a Potent Inhibitor of HIV Reverse Transcriptase." The Journal of Organic Chemistry 61(13): 4192-4193. Dapp, M. J., C. L. Clouser, S. Patterson and L. M. Mansky (2009). "5-Azacytidine Can Induce Lethal Mutagenesis in Human Immunodeficiency Virus Type 1." Journal of Virology 83(22): 11950-11958. DeAngelo, D. J. (2009). "Nelarabine for the Treatment of Patients with Relapsed or Refractory T-cell Acute Lymphoblastic Leukemia or Lymphoblastic Lymphoma." Hematol. Oncology Clinics North America 23(5): 1121-1135. Desiderio, D. M. and M. Kai (1983). "Preparation of stable isotope-incorporated peptide internal standards for field desorption mass spectrometry quantification of peptides in biologic tissue." Biomed Mass Spectrom 10(8): 471-479. Diamantopoulos, P., M. Michael, O. Benopoulou, E. Bazanis, G. Tzeletas, J. Meletis, G. Vayopoulos and N.-A. Viniou (2012). "Antiretroviral activity of 5-azacytidine during treatment of a HTLV-1 positive myelodysplastic syndrome with autoimmune manifestations." Virology Journal 9(1): 1. Dolloff, N. G., J. E. Allen, D. T. Dicker, N. Aqui, D. Vogl, J. Malysz, G. Talamo and W. S. El- Deiry (2012). "Sangivamycin-like Molecule 6 Exhibits Potent Anti-Multiple Myeloma Activity through Inhibition of Cyclin-Dependent Kinase-9." Molecular Cancer Therapeutics 11(11): 2321-2330. Doong, S. L., C. H. Tsai, R. F. Schinazi, D. C. Liotta and Y. C. Cheng (1991). "Inhibition of the replication of hepatitis B virus in vitro by 2',3'-dideoxy-3'-thiacytidine and related analogues." Proceedings of the National Academy of Sciences 88(19): 8495-8499. Doskočil, J. and F. ŠOrm (1969). "Incorporation and Phosphorylation of 5-Azacytidine by Normal and T4-Phage-Infected Cells of E. coli." European Journal of Biochemistry 8(1): 75-80. Doskočil, J. and F. ŠOrm (1970). "The Mode of Action of 5-Aza-2′-Deoxycytidine in Escherichia coli." European Journal of Biochemistry 13(1): 180-187. Dow, L. W., D. E. Bell, L. Poulakos and A. Fridland (1980). "Differences in Metabolism and Cytotoxicity between 9-β-d-Arabinofuranosyladenine and 9-β-d-Arabinofuranosyl-2- fluoroadenine in Human Leukemic Lymphoblasts." Cancer Research 40(5): 1405-1410. Dragovic, G. and D. Jevtovic (2012). "The role of nucleoside reverse transcriptase inhibitors usage in the incidence of hyperlactatemia and lactic acidosis in HIV/AIDS patients." Biomedicine & Pharmacotherapy 66(4): 308-311. Dubinsky, M. C., E. A. Vasiliauskas, H. Singh, M. T. Abreu, K. A. Papadakis, T. Tran, P. Martin, J. M. Vierling, S. A. Geller, S. R. Targan and F. F. Poordad (2003). "6-thioguanine can cause serious liver injury in inflammatory bowel disease patients." Gastroenterology 125(2): 298-303. Eftekhari, P., K. Lassoued, E. Oksenhendler, C. Scieux and J. P. Clauvel (1998). "Severe respiratory syncytial virus pulmonary infection in a patient treated with fludarabine for chronic lymphocytic leukemia." Annals of Hematology 76(5): 225-226. England, T. E., A. G. Bruce and O. C. Uhlenbeck (1980). "Specific labeling of 3' termini of RNA with T4 RNA ligase." Meth Enzymol 65(1): 65-74. Eriksson, S. (2002). "Structure and function of cellular deoxyribonucleoside kinases." Cellular and Molecular Life Sciences 59(8): 1327-1346.

123

Fairchild, C. R., J. Maybaum and K. A. Kennedy (1986). "Concurrent unilateral chromatid damage and DNA strand breakage in response to 6-thioguanine treatment." Biochemical Pharmacology 35(20): 3533-3541. Fleury, S., R. J. De Boer, G. P. Rizzardi, K. C. Wolthers, S. A. Otto, C. C. Welbon, C. Graziosi, C. Knabenhans, H. Soudeyns, P.-A. Bart, S. Gallant, J.-M. Corpataux, M. Gillet, P. Meylan, P. Schnyder, J.-Y. Meuwly, W. Spreen, M. P. Glauser, F. Miedema and G. Pantaleo (1998). "Limited CD4+ T-cell renewal in early HIV-1 infection: Effect of highly active antiretroviral therapy." Nat Med 4(7): 794-801. Foley, T., P. Hentosh and D. E. Walters (2004). "2-Chloro-2′-deoxyadenosine: alteration of DNA:TATA element binding protein (TBP) interactions." Journal of Molecular Modeling 10(1): 32-37. Francini, G., R. Petrioli, A. Aquino and S. Gonnelli (1993). "Advanced breast cancer treatment with folinic acid, 5-fluorouracil, and mitomycin C." Cancer Chemotherapy and Pharmacology 32(5): 359-364. Galmarini, C. M., F. Popowycz and B. Joseph (2008). "Cytotoxic nucleoside analogues: different strategies to improve their clinical efficacy." Current Medicinal Chemistry 15(11): 1072-1082. Gamberale, R., P. Fernandez-Calotti, J. Sanchez-Avalos, M. Alberto, J. Geffner and M. Giordano (2006). "The effect of fludarabine on interferon-gamma production by lymphoid cells from healthy donors and patients with B-cell chronic lymphocytic leukemia." Haematologica 91(4): 574-576. Gandhi, V., E. Estey, M. Keating, A. Chucrallah and W. Plunkett (1996). "Chlorodeoxyadenosine and arabinosylcytosine in patients with acute myelogenous leukemia: pharmacokinetic, pharmacodynamic, and molecular interactions." Blood 87(1): 256-264. Gefter, M. L. and R. L. Russell (1969). "Role of modifications in tyrosine transfer RNA: A modified base affecting ribosome binding." Journal of Molecular Biology 39(1): 145-157. Genini, D., S. Adachi, Q. Chao, D. W. Rose, C. J. Carrera, H. B. Cottam, D. A. Carson and L. M. Leoni (2000). "Deoxyadenosine analogs induce programmed cell death in chronic lymphocytic leukemia cells by damaging the DNA and by directly affecting the mitochondria." Blood 96(10): 3537-3543. Giovannoni, G., G. Comi, S. Cook, K. Rammohan, P. Rieckmann, P. S. Sørensen, P. Vermersch, P. Chang, A. Hamlett, B. Musch and S. J. Greenberg (2010). "A Placebo-Controlled Trial of Oral Cladribine for Relapsing Multiple Sclerosis." New England Journal of Medicine 362(5): 416- 426. Goubet, A., A. Chardon, P. Kumar, P. K. Sharma and R. N. Veedu (2013). "Synthesis of DNA oligonucleotides containing C5-ethynylbenzenesulfonamide-modified nucleotides (EBNA) by polymerases towards the construction of base functionalized nucleic acids." Bioorganic & Medicinal Chemistry Letters 23(3): 761-763. Gu, C., T. J. Begley and P. C. Dedon (2014). "tRNA modifications regulate translation during cellular stress." FEBS Letters 588(23): 4287-4296. Guan, L., G. W. van der Heijden, A. Bortvin and M. M. Greenberg (2011). "Intracellular Detection of Cytosine Incorporation in Genomic DNA by Using 5-Ethynyl-2′-Deoxycytidine." ChemBioChem 12(14): 2184-2190. Gupta, P. K., S. Daunert, M. R. Nassiri, L. L. Wotring, J. C. Drach and L. B. Townsend (1989). "Synthesis, cytotoxicity, and antiviral activity of some acyclic analogs of the pyrrolo[2,3- d]pyrimidine nucleoside antibiotics tubercidin, toyocamycin, and sangivamycin." Journal of Medicinal Chemistry 32(2): 402-408.

124

Hamasaki, T., T. Matsumoto, N. Sakamoto, A. Shimahara, S. Kato, A. Yoshitake, A. Utsunomiya, H. Yurimoto, E. C. Gabazza and T. Ohgi (2013). "Synthesis of (1)(8)O-labeled RNA for application to kinetic studies and imaging." Nucleic Acids Res 41(12): e126. Harris, J. S. and K. Randerath (1978). "Aminoacylation of undermethylated mammalian transfer RNA." Biochimica et Biophysica Acta (BBA) - Nucleic Acids and Protein Synthesis 521(2): 566-575. Hartmann, R. W. (1986). "Influence of alkyl chain ramification on estradiol receptor binding affinity and intrinsic activity of 1,2-dialkylated 1,2-bis(4- or 3-hydroxyphenyl)ethane estrogens and antiestrogens." Journal of Medicinal Chemistry 29(9): 1668-1674. Hartmer, R., N. Storm, S. Boecker, C. P. Rodi, F. Hillenkamp, C. Jurinke and D. van den Boom (2003). "RNase T1 mediated base-specific cleavage and MALDI-TOF MS for high-throughput comparative sequence analysis." Nucleic Acids Res 31(9): e47. Hatano, A., M. Nishimura and I. Souta (2009). "Impact of Unnatural Nucleosides on the Control of Microbial Growth." Biocontrol Science 14(2): 55-60. Hecht, S. M., R. B. Frye, D. Werner, T. Fukui and S. D. Hawrelak (1976). "Synthesis and biological activity of pyrazolo[3,4-d]pyrimidine nucleosides and nucleotides related to tubercidin, toyocamycin, and sangivamycin." Biochemistry 15(5): 1005-1015. Heiss, M., V. F. Reichle and S. Kellner (2017). "Observing the fate of tRNA and its modifications by nucleic acid isotope labeling mass spectrometry: NAIL-MS." RNA Biology: 1- 9. Helm, M. and Juan D. Alfonzo (2014). "Posttranscriptional RNA Modifications: Playing Metabolic Games in a Cell’s Chemical Legoland." Chemistry & Biology 21(2): 174-185. Hentosh, P., R. Koob and R. L. Blakley (1990). "Incorporation of 2-halogeno-2'-deoxyadenosine 5-triphosphates into DNA during replication by human polymerases alpha and beta." Journal of Biological Chemistry 265(7): 4033-4040. Hernández-Vargas, H., E. Ballestar, P. Carmona-Saez, C. von Kobbe, I. Bañón-Rodríguez, M. Esteller, G. Moreno-Bueno and J. Palacios (2006). "Transcriptional profiling of MCF7 breast cancer cells in response to 5-Fluorouracil: Relationship with cell cycle changes and apoptosis, and identification of novel targets of p53." International Journal of Cancer 119(5): 1164-1175. Hertel, L. W., G. B. Boder, J. S. Kroin, S. M. Rinzel, G. A. Poore, G. C. Todd and G. B. Grindey (1990). "Evaluation of the antitumor activity of gemcitabine (2',2'-difluoro-2'-deoxycytidine)." Cancer Res 50(14): 4417-4422. Herzmann, C., K. Arastèh, R. L. Murphy, H. Schulbin, P. Kreckel, D. Drauz, R. F. Schinazi, A. Beard, L. Cartee and M. J. Otto (2005). "Safety, Pharmacokinetics, and Efficacy of (+/−)-β-2′,3′- Dideoxy-5-Fluoro-3′-Thiacytidine with Efavirenz and Stavudine in Antiretroviral-Naïve Human Immunodeficiency Virus-Infected Patients." Antimicrobial Agents and Chemotherapy 49(7): 2828-2833. Higuchi, T., T. Nakamura and G. Wakisaka (1976). "Metabolism of 6-Mercaptopurine in Human Leukemic Cells." Cancer Research 36(10): 3779-3783. Hill, D. L. and L. L. Bennett (1969). "Purification and properties of 5- phosphoribosylpyrophosphate amidotransferase from adenocarcinoma 755 cells." Biochemistry 8(1): 122-130. Hiraoka, A., T. Masaoka, K. Nagai, A. Horiuchi, A. Kanamaru, M. Niimura, T. Hamada and M. Takahashi (1991). "Clinical effect of BV-araU on varicella-zoster virus infection in immunocompromised patients with haematological malignancies." Journal of Antimicrobial Chemotherapy 27(3): 361-367.

125

Hochster, H. S., M. M. Oken, J. N. Winter, L. I. Gordon, B. G. Raphael, J. M. Bennett and P. A. Cassileth (2000). "Phase I Study of Fludarabine Plus Cyclophosphamide in Patients With Previously Untreated Low-Grade Lymphoma: Results and and Long-Term Follow-Up—A Report From the Eastern Cooperative Oncology Group." Journal of Clinical Oncology 18(5): 987. Hollenstein, M. (2013). "Deoxynucleoside triphosphates bearing histamine, carboxylic acid, and hydroxyl residues - synthesis and biochemical characterization." Organic & Biomolecular Chemistry 11(31): 5162-5172. Höltje, M., A. Richartz, B. Zdrazil, A. Schwanke, B. Dugovic, C. Murruzzu, H.-U. Reißig, H. C. Korting, B. Kleuser, H.-D. Höltje and M. Schäfer-Korting (2010). "Human polymerase α inhibitors for skin tumors. Part 2. Modeling, synthesis and influence on normal and transformed keratinocytes of new thymidine and purine derivatives." Journal of Enzyme Inhibition & Medicinal Chemistry 25(2): 250-265. Holzberger, B., J. Strohmeier, V. Siegmund, U. Diederichsen and A. Marx (2012). "Enzymatic synthesis of 8-vinyl- and 8-styryl-2′-deoxyguanosine modified DNA—novel fluorescent molecular probes." Bioorganic & Medicinal Chemistry Letters 22(9): 3136-3139. Hossain, M. and P. A. Limbach (2007). "Mass spectrometry-based detection of transfer RNAs by their signature endonuclease digestion products." RNA 13(2): 295-303. Hostetler, K. Y., L. M. Stuhmiller, H. B. Lenting, H. van den Bosch and D. D. Richman (1990). "Synthesis and antiretroviral activity of phospholipid analogs of azidothymidine and other antiviral nucleosides." Journal of Biological Chemistry 265(11): 6112-6117. Houghton, J. A., P. J. Houghton and R. S. Wooten (1979). "Mechanism of Induction of Gastrointestinal Toxicity in the Mouse by 5-Fluorouracil, 5-Fluorouridine, and 5-Fluoro-2′- deoxyuridine." Cancer Research 39(7 Part 1): 2406-2413. Houser, W. M., A. Butterer, B. Addepalli and P. A. Limbach (2015). "Combining recombinant ribonuclease U2 and protein phosphatase for RNA modification mapping by liquid chromatography-mass spectrometry." Anal Biochem 478: 52-58. Huang, P. and W. Plunkett (1991). "Action of 9-beta-D-arabinofuranosyl-2-fluoroadenine on RNA metabolism." Molecular Pharmacology 39(4): 449-455. Huang, P. and W. Plunkett (1992). "A quantitative assay for fragmented DNA in apoptotic cells." Analytical Biochemistry 207(1): 163-167. Huang, P. and W. Plunkett (1995). "Fludarabine- and gemcitabine-induced apoptosis: incorporation of analogs into DNA is a critical event." Cancer Chemotherapy and Pharmacology 36(3): 181-188. Hughes, W., J. A. McDowell, J. Shenep, P. Flynn, M. W. Kline, R. Yogev, W. Symonds, Y. Lou and S. Hetherington (1999). "Safety and Single-Dose Pharmacokinetics of Abacavir (1592U89) in Human Immunodeficiency Virus Type 1-Infected Children." Antimicrobial Agents and Chemotherapy 43(3): 609-615. Huschtscha, L. I., W. A. Bartier, A. Malmstrom and M. H. N. Tattersall (1995). "Cell death by apoptosis following anticancer drug treatment in vitro." Int. J. Oncol. 6(3): 585-593. Jackman, J. E., J. M. Gott and M. W. Gray (2012). "Doing it in reverse: 3'-to-5' polymerization by the Thg1 superfamily." Rna 18(5): 886-899. Jäger, S., G. Rasched, H. Kornreich-Leshem, M. Engeser, O. Thum and M. Famulok (2005). "A Versatile Toolbox for Variable DNA Functionalization at High Density." Journal of the American Chemical Society 127(43): 15071-15082.

126

Jao, C. Y. and A. Salic (2008). "Exploring RNA transcription and turnover in vivo by using click chemistry." Proceedings of the National Academy of Sciences of the United States of America 105(41): 15779-15784. Jeha, S., V. Gandhi, K. W. Chan, L. McDonald, I. Ramirez, R. Madden, M. Rytting, M. Brandt, M. Keating, W. Plunkett and H. Kantarjian (2004). Clofarabine, a novel nucleoside analog, is active in pediatric patients with advanced leukemia. Ji, H. J., S. Y. Rha, H. C. Jeung, S. H. Yang, S. W. An and H. C. Chung (2005). "Cyclic induction of senescence with intermittent AZT treatment accelerates both apoptosis and telomere loss." Breast cancer research and treatment 93(3): 227-236. Johnson, S., D. Richardson, J. Hopkins, D. Howe and M. Phillips (1993). "Complete remission after fludarabine for chronic lymphocytic leukemia [letter; comment]." Blood 81(2): 560-560. Johnston, J. B. (2011). "Mechanism of Action of Pentostatin and Cladribine in Hairy Cell Leukemia." Leukemia & Lymphoma 52: 43-45. Jordheim, L. P., E. Cros, C. M. Galmarini, C. Dumontet, A.-S. Bretonnet, I. Krimm, J.-M. Lancelin and M.-C. Gagnieu (2006). "F-ara-AMP is a substrate of cytoplasmic 5'-nucleotidase II (CN-II): HPLC and NMR studies of enzymatic dephosphorylation." Nucleosides, Nucleotides & Nucleic Acids 25(3): 289-297. Juliusson, G., R. Lenkei, G. Tjønnfjord, D. Heldal and J. Liliemark (1995). "Neutropenic fever following cladribine therapy for symptomatic hairy-cell leukemia: Predictive factors and effects of granulocyte-macrophage colony-stimulating factor." Annals of Oncology 6(4): 371-375. Kajiyama, T., M. Kuwahara, M. Goto and H. Kambara (2011). "Optimization of pyrosequencing reads by superior successive incorporation efficiency of improved 2′-deoxyadenosine-5′- triphosphate analogs." Analytical Biochemistry 416(1): 8-17. Kantarjian, H., J.-P. J. Issa, C. S. Rosenfeld, J. M. Bennett, M. Albitar, J. DiPersio, V. Klimek, J. Slack, C. de Castro, F. Ravandi, R. Helmer, L. Shen, S. D. Nimer, R. Leavitt, A. Raza and H. Saba (2006). "Decitabine improves patient outcomes in myelodysplastic syndromes." Cancer 106(8): 1794-1803. Kellner, S., A. Ochel, K. Thuring, F. Spenkuch, J. Neumann, S. Sharma, K. D. Entian, D. Schneider and M. Helm (2014). "Absolute and relative quantification of RNA modifications via biosynthetic isotopomers." Nucleic Acids Res 42(18): e142. Kifli, N., E. De Clercq, J. Balzarini and C. Simons (2004). "Novel imidazo[1,2-c]pyrimidine base-modified nucleosides: synthesis and antiviral evaluation." Bioorganic & Medicinal Chemistry 12(15): 4245-4252. Kim, K. N., J. Lee, D. H. Kim, J.-S. Yoo and H. J. Kwon (2005). "A new synthetic analogue of thymidine, 7-(3-bromo-phenoxy)-thymidine, inhibits the proliferation of tumor cells." Bioorganic & Medicinal Chemistry Letters 15(1): 77-79. Kirpekar, F., S. Douthwaite and P. Roepstorff (2000). "Mapping posttranscriptional modifications in 5S ribosomal RNA by MALDI mass spectrometry." RNA 6: 296-306. Kohli, E., H.-Y. Han, A. D. Zeman and S. V. Vinogradov (2007). "Formulations of biodegradable Nanogel carriers with 5′-triphosphates of nucleoside analogs that display a reduced cytotoxicity and enhanced drug activity." Journal of Controlled Release 121(1–2): 19- 27. Koros, C. and E. Kitraki (2009). "Neurofilament isoform alterations in the rat cerebellum following cytosine arabinoside administration." Toxicology letters 189(3): 215-218. Kose, M. and A. C. Schiedel (2009). "Nucleoside/nucleobase transporters: drug targets of the future?" Future Medicinal Chemistry 1(2): 303-326.

127

Kowalak, J. A., E. Bruenger and J. A. McCloskey (1995). "Posttranscriptional modification of the central loop of domain V in Escherichia coli 23 S ribosomal RNA." J Biol Chem 270(30): 17758-17764. Kowalak, J. A., S. C. Pomerantz, P. F. Crain and J. A. McCloskey (1993). "A novel method for the determination of post-transcriptional modification in RNA by mass spectrometry." Nucleic Acids Res 21(19): 4577-4585. Krawczyk, S. H., T. E. Renau, M. R. Nassiri, A. C. Westerman, L. L. Wotring, J. C. Drach and L. B. Townsend (1995). "Synthesis and Evaluation of Certain Thiosangivamycin Analogs as Potential Inhibitors of Cell Proliferation and Human Cytomegalovirus." Journal of Medicinal Chemistry 38(20): 4115-4119. Krishna, I. V., G. R. Vanaja, N. S. Kumar and G. Suman (2009). "Cytotoxic studies of antineoplastic drugs on human lymphocytes--in vitro studies." Cancer Biomark 5(6): 261-272. Kuchta, R. D., D. Ilsley, K. D. Kravig, S. Schubert and B. Harris (1992). "Inhibition of DNA primase and polymerase .alpha. by arabinofuranosylnucleoside triphosphates and related compounds." Biochemistry 31(19): 4720-4728. Kumar, P. N., D. E. Sweet, J. A. McDowell, W. Symonds, Y. Lou, S. Hetherington and S. LaFon (1999). "Safety and Pharmacokinetics of Abacavir (1592U89) following Oral Administration of Escalating Single Doses in Human Immunodeficiency Virus Type 1-Infected Adults." Antimicrobial Agents and Chemotherapy 43(3): 603-608. Kuo, H. K., R. Krasich, A. S. Bhagwat and K. N. Kreuzer (2010). "Importance of the tmRNA system for cell survival when transcription is blocked by DNA–protein cross-links." Molecular Microbiology 78(3): 686-700. Lambe, C. U., D. R. Averett, M. T. Paff, J. E. Reardon, J. G. Wilson and T. A. Krenitsky (1995). "2-Amino-6-methoxypurine Arabinoside: An Agent for T-Cell Malignancies." Cancer Research 55(15): 3352-3356. Lange, S., M. Sylvester, M. Schumann, C. Freund and E. Krause (2010). "Identification of phosphorylation-dependent interaction partners of the adapter protein ADAP using quantitative mass spectrometry: SILAC vs (18)O-labeling." J Proteome Res 9(8): 4113-4122. Lauridsen, L. H., J. A. Rothnagel and R. N. Veedu (2012). "Enzymatic Recognition of 2′-Modified Ribonucleoside 5′-Triphosphates: Towards the Evolution of Versatile Aptamers." ChemBioChem 13(1): 19-25. Lennard, L. (1992). "The clinical pharmacology of 6-mercaptopurine." Eur J Clin Pharmacol 43(4): 329-339. Li, S. and P. A. Limbach (2012). "Method for comparative analysis of ribonucleic acids using isotope labeling and mass spectrometry." Anal Chem 84(20): 8607-8613. Li, S. and P. A. Limbach (2014). "Identification of RNA Sequence Isomers by Isotope Labeling and LC-MS/MS." J Mass Spectrom 49(11): 1191-1198. Li, X., X. Xiong and C. Yi (2016). "Epitranscriptome sequencing technologies: decoding RNA modifications." Nat Methods 14(1): 23-31. Liao, Y.-X. and Q.-S. Hu (2011). "CuCl/Bipyridine-Catalyzed Addition Reactions of Arylboroxines with Aldehydes, α,β-Unsaturated Ketones, and N-Tosyl Aldimines." The Journal of Organic Chemistry 76(18): 7602-7607. Liliemark, J., B. Pettersson, B. Engberg, P. Lafolie, M. Masquelier and C. Peterson (1990). "On the Paradoxically Concentration-dependent Metabolism of 6-Mercaptopurine in WEHI-3b Murine Leukemia Cells." Cancer Research 50(1): 108-112.

128

Limbach, P. A. and M. J. Paulines (2017). "Going global: the new era of mapping modifications in RNA." Wiley Interdisciplinary Reviews: RNA 8(1): e1367-n/a. Litosh, V. A., W. Wu, B. P. Stupi, J. Wang, S. E. Morris, M. N. Hersh and M. L. Metzker (2011). "Improved nucleotide selectivity and termination of 3'-OH unblocked reversible terminators by molecular tuning of 2-nitrobenzyl alkylated HOMedU triphosphates." Nucleic Acids Res 39(6): e39. Litosh, V. A., W. Wu, B. P. Stupi, J. Wang, S. E. Morris, M. N. Hersh and M. L. Metzker (2011). "Improved nucleotide selectivity and termination of 3′-OH unblocked reversible terminators by molecular tuning of 2-nitrobenzyl alkylated HOMedU triphosphates." Nucleic Acids Research 39(6): e39. Lolkema, M. P., H.-T. Arkenau, K. Harrington, P. Roxburgh, R. Morrison, V. Roulstone, K. Twigger, M. Coffey, K. Mettinger, G. Gill, T. R. J. Evans and J. S. de Bono (2011). "A Phase I Study of the Combination of Intravenous Reovirus Type 3 Dearing and Gemcitabine in Patients with Advanced Cancer." Clinical Cancer Research 17(3): 581-588. Lotfi, K., E. Månsson, T. Spasokoukotskaja, B. Pettersson, J. Liliemark, C. Peterson, S. Eriksson and F. Albertioni (1999). "Biochemical Pharmacology and Resistance to 2-Chloro-2′-arabino- fluoro-2′-deoxyadenosine, a Novel Analogue of Cladribine in Human Leukemic Cells." Clinical Cancer Research 5(9): 2438-2444. Lu, L. J. and K. Randerath (1984). "Long term instability and molecular mechanism of 5- azacytidine-induced DNA hypomethylation in normal and neoplastic tissues in vivo." Molecular Pharmacology 26(3): 594-603. Ludwig, P. S., R. A. Schwendener and H. Schott (2005). "Synthesis and anticancer activities of amphiphilic 5-fluoro-2′-deoxyuridylic acid prodrugs." European Journal of Medicinal Chemistry 40(5): 494-504. Machida, H., S. Sakata, A. Kuninaka and H. Yoshino (1981). "Antiherpesviral and anticellular effects of 1-beta-D-arabinofuranosyl-E-5-(2-halogenovinyl) uracils." Antimicrobial Agents and Chemotherapy 20(1): 47-52. Macías, J., K. Neukam, J. Mallolas, L. López-Cortés, J. Cartón, P. Domingo, S. Moreno, J. Iribarren, B. Clotet, M. Crespo, I. de Los Santos, E. Ortega, H. Knobel, M. Jiménez-Expósito and J. Pineda (2012). "Liver Toxicity of Initial Antiretroviral Drug Regimens Including Two Nucleoside Analogs Plus One Non-Nucleoside Analog or One Ritonavir-Boosted Protease Inhibitor in HIV/HCV-Coinfected Patients." HIV Clinical Trials 13(2): 61-69. Magnani, M., A. Casabianca, A. Fraternale, G. Brandi, S. Gessani, R. Williams, M. Giovine, G. Damonte, A. De Flora and U. Benatti (1996). "Synthesis and targeted delivery of an azidothymidine homodinucleotide conferring protection to macrophages against retroviral infection." Proceedings of the National Academy of Sciences 93(9): 4403-4408. Marbaniang, C. N. and J. Vogel (2016). "Emerging roles of RNA modifications in bacteria." Current Opinion in Microbiology 30(Supplement C): 50-57. Marinković, G., J. Kroon, M. Hoogenboezem, K. A. Hoeben, M. S. Ruiter, K. Kurakula, I. Otermin Rubio, M. Vos, C. J. M. de Vries, J. D. van Buul and V. de Waard (2014). "Inhibition of GTPase Rac1 in Endothelium by 6-Mercaptopurine Results in Immunosuppression in Nonimmune Cells: New Target for an Old Drug." The Journal of Immunology 192(9): 4370- 4378. Maruoka, K., S. Saito, A. B. Concepcion and H. Yamamoto (1993). "Chemoselective functionalization of more hindered aldehyde carbonyls with the methylaluminum bis(2,6-

129

diphenylphenoxide)/alkyllithium system." Journal of the American Chemical Society 115(3): 1183-1184. McCollister, R. J., W. R. Gilbert, D. M. Ashton and J. B. Wyngaarden (1964). "Pseudofeedback Inhibition of Purine Synthesis by 6-Mercaptopurine Ribonucleotide and Other Purine Analogues." Journal of Biological Chemistry 239(5): 1560-1563. McGuigan, C., C. J. Yarnold, G. Jones, S. Velázquez, H. Barucki, A. Brancale, G. Andrei, R. Snoeck, E. De Clercq and J. Balzarini (1999). "Potent and Selective Inhibition of Varicella- Zoster Virus (VZV) by Nucleoside Analogues with an Unusual Bicyclic Base." Journal of Medicinal Chemistry 42(22): 4479-4484. McLuckey, S. A., G. J. Van Berkel and G. L. Glish (1992). "Tandem Mass Spectrometry of Small, Multiply Charged Oligonucleotides." J. Am. Soc. Mass Spectrom. 3: 60–70. Meade, E. A., L. L. Wotring, J. C. Drach and L. B. Townsend (1992). "Synthesis, antiproliferative, and antiviral activity of certain 4-aminopyrrolo[2,3-d]pyridazine nucleosides: an entry into a novel series of adenosine analogs." Journal of Medicinal Chemistry 35(3): 526- 533. Meier, C. (2002). "cycloSal-pronucleotides design of chemical trojan horses." Mini Reviews In Medicinal Chemistry 2(3): 219-234. Meng, Z. and P. A. Limbach (2005). "Quantitation of Ribonucleic Acids Using 18-O Labeling and Mass Spectrometry." Anal. Chem. 77: 1891-1895. Meng, Z. and P. A. Limbach (2006). "Mass spectrometry of RNA: linking the genome to the proteome†." Briefings in Functional Genomics 5(1): 87-95. Miller, V., M. Ait-Khaled, C. Stone, P. Griffin, D. Mesogiti, A. Cutrell, R. Harrigan, S. Staszewski, C. Katlama, G. Pearce and M. Tisdale (2000). "HIV-1 reverse transcriptase (RT) genotype and susceptibility to RT inhibitors during abacavir monotherapy and combination therapy." AIDS 14(2): 163-171. Mirgorodskaya, O. A., Y. P. Kozmin, M. I. Titov, R. Korner, C. P. Sonksen and P. Roepstorff (2000). "Quantitation of peptides and proteins by matrix-assisted laser desorption/ionization mass spectrometry using (18)O-labeled internal standards." Rapid Commun Mass Spectrom 14(14): 1226-1232. Molina-Arcas, M., G. Moreno-Bueno, P. Cano-Soldado, H. Hernández-Vargas, F. J. Casado, J. Palacios and M. Pastor-Anglada (2006). "Human equilibrative nucleoside transporter-1 (hENT1) is required for the transcriptomic response of the nucleoside-derived drug 5′-DFUR in breast cancer MCF7 cells." Biochemical Pharmacology 72(12): 1646-1656. Momparler, R., S. Cote, L. Momparler and Y. Idaghdour (2014). "Epigenetic therapy of acute myeloid leukemia using 5-aza-2'-deoxycytidine (decitabine) in combination with inhibitors of histone methylation and deacetylation." Clinical Epigenetics 6(1): 19. Momparler, R. L., D. Y. Bouffard, L. F. Momparler, J. Dionne, K. Belanger and J. Ayoub (1997). "Pilot phase I-II study on 5-aza-2'-deoxycytidine (Decitabine) in patients with metastatic lung cancer." Anti-Cancer Drugs 8(4): 358-368. Moore, B. M., R. K. Jalluri and M. B. Doughty (1996). "DNA Polymerase Photoprobe 2-[(4- Azidophenacyl)thio]-2‘-deoxyadenosine 5‘-Triphosphate Labels an Escherichia coli DNA Polymerase I Klenow Fragment Substrate Binding Site." Biochemistry 35(36): 11642-11651. Moore, B. M., K. Li and M. B. Doughty (1996). "Deoxyadenosine-Based DNA Polymerase Photoprobes: Design, Synthesis, and Characterization as Inhibitors of the Escherichia coli DNA Polymerase I Klenow Fragment." Biochemistry 35(36): 11634-11641.

130

Moore, E. C. and G. A. LePage (1958). "The Metabolism of 6-Thioguanine in Normal and Neoplastic Tissues." Cancer Research 18(9): 1075-1083. Mori, Y., S. Ozasa, M. Kitaoka, S. Noda, T. Tanaka, H. Ichinose and N. Kamiya (2013). "Aligning an endoglucanase Cel5A from Thermobifida fusca on a DNA scaffold: potent design of an artificial cellulosome." Chemical Communications 49(62): 6971-6973. Motea, E. A., I. Lee and A. J. Berdis (2012). "A Non-natural Nucleoside with Combined Therapeutic and Diagnostic Activities against Leukemia." ACS Chemical Biology 7(6): 988-998. Mukherjee, K. L. and C. Heidelberger (1962). "Studies on Fluorinated Pyrimidines: XV. Inhibition of the Incorporation of Formate-C14 into DNA Thymine of Ehrlich Ascites Carcinoma Cells by 5-Fluoro-2′-deoxyuridine-5′-monophosphate and Related Compounds." Cancer Research 22(7 Part 1): 815-822. Munshi, P. N., M. Lubin and J. R. Bertino (2014). "6-Thioguanine: A Drug With Unrealized Potential for Cancer Therapy." The Oncologist 19(7): 760-765. Murakami, E., T. Tolstykh, H. Bao, C. Niu, H. M. M. Steuer, D. Bao, W. Chang, C. Espiritu, S. Bansal, A. M. Lam, M. J. Otto, M. J. Sofia and P. A. Furman (2010). "Mechanism of Activation of PSI-7851 and Its Diastereoisomer PSI-7977." Journal of Biological Chemistry 285(45): 34337-34347. Nabhan, C., N. Davis, J. D. Bitran, A. Galvez, W. Fried, K. Tolzien, S. Foss, W. M. Dewey and P. Venugopal (2011). "Efficacy and safety of clofarabine in relapsed and/or refractory non- hodgkin lymphoma, including rituximab-refractory patients." Cancer 117(7): 1490-1497. Nachtergaele, S. and C. He (2016). "The emerging biology of RNA post-transcriptional modifications." RNA Biol 12(2): 156-162. Nagaradja, E., F. Chevallier, T. Roisnel, V. Jouikov and F. Mongin (2012). "Deprotonative metalation of aromatic compounds using mixed lithium–iron combinations." Tetrahedron 68(14): 3063-3073. Nauš, P., R. Pohl, I. Votruba, P. Džubák, M. Hajdúch, R. Ameral, G. Birkuš, T. Wang, A. S. Ray, R. Mackman, T. Cihlar and M. Hocek (2010). "6-(Het)aryl-7-Deazapurine Ribonucleosides as Novel Potent Cytostatic Agents." Journal of Medicinal Chemistry 53(1): 460-470. Nekhai, S., U. G. Bhat, T. Ammosova, S. K. Radhakrishnan, M. Jerebtsova, X. Niu, A. Foster, T. J. Layden and A. L. Gartel (2006). "A novel anticancer agent ARC antagonizes HIV-1 and HCV." Oncogene 26(26): 3899-3903. Newman, A. H., S. Izenwasser, M. J. Robarge and R. H. Kline (1999). "CoMFA study of novel phenyl ring-substituted 3alpha-(diphenylmethoxy)tropane analogues at the dopamine transporter." J Med Chem 42(18): 3502-3509. Ni, J., S. C. Pomerantz, J. Rozenski, Y. Zhang and J. A. McCloskey (1996). "Interpretation of Oligonucleotide Mass Spectra for Determination of Sequence Using Electrospray Ionization and Tandem Mass Spectrometry." Analytical Chemistry 68(13): 1989-1999. Nord, L. D., R. L. Stolfi, A. A. Alfieri, G. Netto, V. Reuter, S. S. Sternberg, J. R. Colofiore, J. A. Koutcher and D. S. Martin (1997). "Apoptosis induced in advanced CD8F1-murine mammary tumors by the combination of PALA, MMPR and 6AN precedes tumor regression and is preceded by ATP depletion." Cancer Chemotherapy and Pharmacology 40(5): 376-384. Olivero, O. A., I. L. Vazquez, C. C. Cooch, J. Ming, E. Keller, M. Yu, J. P. Borojerdi, H. M. Braun, E. McKee and M. C. Poirier (2010). "Long-term AZT Exposure Alters the Metabolic Capacity of Cultured Human Lymphoblastoid Cells." Toxicological Sciences 115(1): 109-117.

131

Ong, S. E., B. Blagoev, I. Kratchmarova, D. B. Kristensen, H. Steen, A. Pandey and M. Mann (2002). "Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics." Mol Cell Proteomics 1(5): 376-386. Ong, S. E. and M. Mann (2006). "A practical recipe for stable isotope labeling by amino acids in cell culture (SILAC)." Nat Protoc 1(6): 2650-2660. Ottria, R., S. Casati, E. Baldoli, J. A. M. Maier and P. Ciuffreda (2010). "N6-Alkyladenosines: Synthesis and evaluation of in vitro anticancer activity." Bioorganic & Medicinal Chemistry 18(23): 8396-8402. Pagoti, S., D. Dutta and J. Dash (2013). "A Magnetoclick Imidazolidinone Nanocatalyst for Asymmetric 1,3-Dipolar Cycloadditions." Advanced Synthesis & Catalysis 355(18): 3532-3538. Pajouhesh, H., Z. P. Feng, Y. Ding, L. Zhang, H. Pajouhesh, J. L. Morrison, F. Belardetti, E. Tringham, E. Simonson, T. W. Vanderah, F. Porreca, G. W. Zamponi, L. A. Mitscher and T. P. Snutch (2010). "Structure-activity relationships of diphenylpiperazine N-type calcium channel inhibitors." Bioorg Med Chem Lett 20(4): 1378-1383. Parker, W. B., A. R. Bapat, J. X. Shen, A. J. Townsend and Y. C. Cheng (1988). "Interaction of 2-halogenated dATP analogs (F, Cl, and Br) with human DNA polymerases, DNA primase, and ribonucleotide reductase." Molecular Pharmacology 34(4): 485-491. Parker, W. B., S. C. Shaddix, C.-H. Chang, E. L. White, L. M. Rose, R. W. Brockman, A. T. Shortnacy, J. A. Montgomery, J. A. Secrist and L. L. Bennett (1991). "Effects of 2-Chloro-9-(2- deoxy-2-fluoro-β-d-arabinofuranosyl)adenine on K562 Cellular Metabolism and the Inhibition of Human Ribonucleotide Reductase and DNA Polymerases by Its 5′-Triphosphate." Cancer Research 51(9): 2386-2394. Parker, W. B., S. C. Shaddix, C. H. Chang, E. L. White, L. M. Rose, R. W. Brockman, A. T. Shortnacy, J. A. Montgomery, J. A. Secrist, 3rd and L. L. Bennett, Jr. (1991). "Effects of 2- chloro-9-(2-deoxy-2-fluoro-beta-D-arabinofuranosyl)adenine on K562 cellular metabolism and the inhibition of human ribonucleotide reductase and DNA polymerases by its 5'-triphosphate." Cancer Res 51(9): 2386-2394. Parsels, L. A., J. D. Parsels, L. M. Wagner, T. L. Loney, E. H. Radany and J. Maybaum (1998). "Mechanism and pharmacological specificity of dUTPase-mediated protection from DNA damage and cytotoxicity in human tumor cells." Cancer Chemotherapy and Pharmacology 42(5): 357-362. Parsons, P. G., E. P. W. Bowman and R. L. Blakley (1986). "Selective toxicity of deoxyadenosine analogues in human melanoma cell lines." Biochemical Pharmacology 35(22): 4025-4029. Patil, A., C. Chan, M. Dyavaiah, J. P. Rooney and P. Dedon (2012). "Translational infidelity- induced protein stress results from a deficiency in Trm9-catalyzed tRNA modifications." RNA Biology 9(7): 990-1001. Perrin, D. M., T. Garestier and C. Hélène (2001). "Bridging the Gap between Proteins and Nucleic Acids: A Metal-Independent RNAseA Mimic with Two Protein-Like Functionalities." Journal of the American Chemical Society 123(8): 1556-1563. Peters, G. J., C. L. van der Wilt, C. J. van Groeningen, K. Smid, S. Meijer and H. M. Pinedo (1994). "Thymidylate synthase inhibition after administration of fluorouracil with or without leucovorin in colon cancer patients: implications for treatment with fluorouracil." J Clin Oncol 12(10): 2035-2042. Peterson, L. W. and C. E. McKenna (2009). "Prodrug approaches to improving the oral absorption of antiviral nucleotide analogues." Expert Opinion on Drug Delivery 6(4): 405-420.

132

Petruseva, I. O., I. S. Tikhanovich, B. P. Chelobanov and O. I. Lavrik (2008). "RPA repair recognition of DNA containing pyrimidines bearing bulky adducts." Journal of Molecular Recognition 21(3): 154-162. Peyrottes, S., D. Egron, I. Lefebvre, G. Gosselin, J. L. Imbach and C. Périgaud (2004). "SATE pronucleotide approaches: an overview." Mini Reviews In Medicinal Chemistry 4(4): 395-408. Pinto, A., M. Maio, V. Attadia, S. Zappacosta and R. Cimino (1984). "MODULATION OF HLA-DR ANTIGENS EXPRESSION IN HUMAN MYELOID LEUKAEMIA CELLS BY CYTARABINE AND 5-AZA-2'-DEOXYCYTIDINE." The Lancet 324(8407): 867-868. Plitta, B., E. Adamska, M. Giel-Pietraszuk, A. Fedoruk-Wyszomirska, M. Naskręt-Barciszewska, W. T. Markiewicz and J. Barciszewski (2012). "New cytosine derivatives as inhibitors of DNA methylation." European Journal of Medicinal Chemistry 55(0): 243-254. Popova, A. M. and J. R. Williamson (2014). "Quantitative analysis of rRNA modifications using stable isotope labeling and mass spectrometry." Journal of the American Chemical Society 136(5): 2058-2069. Porcari, A. R., R. G. Ptak, K. Z. Borysko, J. M. Breitenbach, S. Vittori, L. L. Wotring, J. C. Drach and L. B. Townsend (2000). "Deoxy Sugar Analogues of Triciribine: Correlation of Antiviral and Antiproliferative Activity with Intracellular Phosphorylation." Journal of Medicinal Chemistry 43(12): 2438-2448. Pratt, S., R. L. Shepard, R. A. Kandasamy, P. A. Johnston, W. Perry and A. H. Dantzig (2005). "The multidrug resistance protein 5 (ABCC5) confers resistance to 5-fluorouracil and transports its monophosphorylated metabolites." Molecular Cancer Therapeutics 4(5): 855-863. Prüfer, J., M. Schuchardt, M. Tölle, N. Prüfer, M. Höhne, W. Zidek and M. van der Giet (2014). "Harmful Effects of the Azathioprine Metabolite 6-Mercaptopurine in Vascular Cells: Induction of Mineralization." PLoS ONE 9(7): 1-8. Qian, W. J., T. Liu, V. A. Petyuk, M. A. Gritsenko, B. O. Petritis, A. D. Polpitiya, A. Kaushal, W. Xiao, C. C. Finnerty, M. G. Jeschke, N. Jaitly, M. E. Monroe, R. J. Moore, L. L. Moldawer, R. W. Davis, R. G. Tompkins, D. N. Herndon, D. G. Camp and R. D. Smith (2009). "Large-scale multiplexed quantitative discovery proteomics enabled by the use of an (18)O-labeled "universal" reference sample." J Proteome Res 8(1): 290-299. Qu, D., L. Zhou, W. Wang, Z. Wang, G. Wang, W. Chi and B. Zhang (2013). "5-Ethynylcytidine as a new agent for detecting RNA synthesis in live cells by “click” chemistry." Analytical Biochemistry 434(1): 128-135. Radhakrishnan, S. K. and A. L. Gartel (2006). "A Novel Transcriptional Inhibitor Induces Apoptosis in Tumor Cells and Exhibits Antiangiogenic Activity." Cancer Research 66(6): 3264- 3270. Reineks, E. Z. and A. J. Berdis (2003). "Evaluating the Contribution of Base Stacking during Translesion DNA Replication†." Biochemistry 43(2): 393-404. Reyes, P. and T. C. Hall (1969). "Synthesis of 5-fluorouridine 5′-phosphate by a pyrimidine phosphoribosyltransferase of mammalian origin—II. Correlation between the tumor levels of the enzyme and the 5-fluorouracil-promoted increase in survival of tumor-bearing mice." Biochemical Pharmacology 18(10): 2587-2590. Ritter, C. A., G. Jedlltschky, H. Meyer zu Schwabedissen, M. Grube, K. Köck and H. K. Kroerner (2005). "CELLULAR EXPORT OF DRUGS AND SIGNALING MOLECULES BY THE ATP-BINDING CASSETTE TRANSPORTERS MRP4 (ABCC4) AND MRP5 (ABCC5)." Drug Metabolism Reviews 37(1): 253-278.

133

Robak, P. and T. Robak (2013). "Older and new purine nucleoside analogs for patients with acute leukemias." Cancer Treatment Reviews 39(8): 851-861. Robak, T., A. Korycka, M. Kasznicki, A. Wrzesien-Kus and P. Smolewski (2005). "Purine nucleoside analogues for the treatment of hematological malignancies: pharmacology and clinical applications." Current Cancer Drug Targets 5(6): 421-444. Robak, T., E. Lech-Maranda, A. Korycka and E. Robak (2006). "Purine nucleoside analogs as immunosuppressive and antineoplastic agents: mechanism of action and clinical activity." Current Medicinal Chemistry 13(26): 3165-3189. Roberts, D. D. and E. W. Hall (1988). "Solvent effects on the rates of solvolysis of pinacolyl derivatives." The Journal of Organic Chemistry 53(11): 2573-2579. Rodriguez, C. O., J. K. Legha, E. Estey, M. J. Keating and V. Gandhi (1997). "Pharmacological and biochemical strategies to increase the accumulation of arabinofuranosylguanine triphosphatein primary human leukemia cells." Clinical Cancer Research 3(11): 2107-2113. Ross, R., X. Cao, N. Yu and P. A. Limbach (2016). "Sequence mapping of transfer RNA chemical modifications by liquid chromatography tandem mass spectrometry." Methods 107: 73- 78. Saijo, M., T. Suzutani, K. Mizuta, I. Kurane and S. Morikawa (2008). "Characterization and susceptibility to antiviral agents of herpes simplex virus type 1 containing a unique thymidine kinase gene with an amber codon between the first and the second initiation codons." Archives of Virology 153(2): 303-314. Santi, D. V. and C. S. McHenry (1972). "5-Fluoro-2′-Deoxyuridylate: Covalent Complex with Thymidylate Synthetase." Proceedings of the National Academy of Sciences of the United States of America 69(7): 1855-1857. Sartorelli, A. C. and G. A. LePage (1958). "Metabolic Effects of 6-Thioguanine: II. Biosynthesis of Nucleic Acid Purines in Vivo and in Vitro." Cancer Research 18(11): 1329-1335. Sartorelli, A. C., G. A. LePage and E. C. Moore (1958). "Metabolic Effects of 6-Thioguanine: I. Studies on Thioguanine-resistant and -sensitive Ehrlich Ascites Cells." Cancer Research 18(10): 1232-1239. Sato, A., J. A. Montgomery and J. G. Cory (1984). "Synergistic Inhibition of Leukemia L1210 Cell Growth in Vitro by Combinations of 2-Fluoroadenine Nucleosides and Hydroxyurea or 2,3- Dihydro-1H-pyrazole[2,3-a]imidazole." Cancer Research 44(8): 3286-3290. Schinazi, R. F., F. D. Boudinot, S. S. Ibrahim, C. Manning, H. M. McClure and D. C. Liotta (1992). "Pharmacokinetics and metabolism of racemic 2',3'-dideoxy-5-fluoro-3'-thiacytidine in rhesus monkeys." Antimicrobial Agents and Chemotherapy 36(11): 2432-2438. Schram, K. H. and L. B. Townsend (1971). "The synthesis of 6-amino-4-methyl-8-(β-D- ribofuranosyl) (4-H,8-H)pyrrolo-[4,3,2-de]pyrimido[4,5-c]pyridazine, a new tricyclic nucleoside." Tetrahedron Letters 12(49): 4757-4760. Schulz, D. and A. Rentmeister (2014). "Current Approaches for RNA Labeling in Vitro and in Cells Based on Click Reactions." ChemBioChem 15(16): 2342-2347. Schwanke, A., C. Murruzzu, B. Zdrazil, R. Zuhse, M. Natek, M. Höltje, H. C. Korting, H.-U. Reissig, H.-D. Höltje and M. Schäfer-Korting (2010). "Antitumor effects of guanosine-analog phosphonates identified by molecular modelling." International Journal of Pharmaceutics 397(1– 2): 9-18. Scrima, M., G. Lauro, M. Grimaldi, S. Di Marino, A. Tosco, P. Picardi, P. Gazzerro, R. Riccio, E. Novellino, M. Bifulco, G. Bifulco and A. M. D’Ursi (2014). "Structural Evidence of N6-

134

Isopentenyladenosine As a New Ligand of Farnesyl Pyrophosphate Synthase." Journal of Medicinal Chemistry 57(18): 7798-7803. Seyferth, D., R. C. Hui and W. L. Wang (1993). "Direct, low temperature, in situ nucleophilic aroylation with aroyllithium reagents." The Journal of Organic Chemistry 58(21): 5843-5845. Shelton, J. R., C. E. Cutler, M. Oliveira, J. Balzarini and M. A. Peterson (2012). "Synthesis, SAR, and preliminary mechanistic evaluation of novel antiproliferative N6,5′-bis-ureido- and 5′- carbamoyl-N6-ureidoadenosine derivatives." Bioorganic & Medicinal Chemistry 20(2): 1008- 1019. Siegel, R., J. Ma, Z. Zou and A. Jemal (2014). "Cancer statistics, 2014." CA Cancer J Clin 64(1): 9-29. Simko, S. J., H. D. Tran, J. Jones, M. Bilgi, L. K. Beaupin, D. Coulter, T. Garrington, T. L. McCavit, C. Moore, F. Rivera-Ortegón, L. Shaffer, L. Stork, L. Turcotte, E. C. Welsh, M. J. Hicks, K. L. McClain and C. E. Allen (2014). "Clofarabine salvage therapy in refractory multifocal histiocytic disorders, including Langerhans cell histiocytosis, juvenile xanthogranuloma and Rosai–Dorfman disease." Pediatric Blood & Cancer 61(3): 479-487. Skripal', I. G., V. V. Babichev, S. V. Bezuglyi, O. V. Egorov, E. S. Korobkova, L. P. Malinovskaia, L. P. Panchenko, I. P. Tokovenko, A. S. Shalamai and I. V. Alekseeva (1993). "The inhibitory action of 6-azacytidine on Mollicutes and its proposed mechanism." Mikrobiologicheskii Zhurnal 55(2): 99-104. Slusarczyk, M., M. H. Lopez, J. Balzarini, M. Mason, W. G. Jiang, S. Blagden, E. Thompson, E. Ghazaly and C. McGuigan (2014). "Application of ProTide Technology to Gemcitabine: A Successful Approach to Overcome the Key Cancer Resistance Mechanisms Leads to a New Agent (NUC-1031) in Clinical Development." Journal of Medicinal Chemistry 57(4): 1531- 1542. Smith, W. B. and O. C. Ho (1990). "Application of the isoamyl nitrite-diiodomethane route to aryl iodides." The Journal of Organic Chemistry 55(8): 2543-2545. Snoeck, R., G. Andrei and E. De Clercq (1999). "Current Pharmacological Approaches to the Therapy of Varicella Zoster Virus Infections." Drugs 57(2): 187-206. Sobrero, A. F., C. Aschele and J. R. Bertino (1997). "Fluorouracil in colorectal cancer--a tale of two drugs: implications for biochemical modulation." Journal of Clinical Oncology 15(1): 368- 381. Song, J. and C. Yi (2017). "Chemical Modifications to RNA: A New Layer of Gene Expression Regulation." ACS Chem Biol. Song, X., P. L. Lorenzi, C. P. Landowski, B. S. Vig, J. M. Hilfinger and G. L. Amidon (2005). "Amino Acid Ester Prodrugs of the Anticancer Agent Gemcitabine: Synthesis, Bioconversion, Metabolic Bioevasion, and hPEPT1-Mediated Transport." Molecular Pharmaceutics 2(2): 157- 167. Sorm, F. and J. Vesely (1964). "The activity of a new antimetabolite, 5-azacytidine, against lymphoid leukaemia in AK mice." Neoplasma 11(2): 123-130. Spriggs, D., G. Robbins, T. Mitchell and D. Kufe (1986). "Incorporation of 9-β-d- arabinofuranosyl-2-fluoroadenine into HL-60 cellular RNA and DNA." Biochemical Pharmacology 35(2): 247-252. Sprinson, D. B. and D. Rittenberg (1951). "Nature of the Activation Process in Enzymatic Reactions." Nature 167(4247): 484-484. Srivatsan, S. G. and Y. Tor (2009). "Enzymatic Incorporation of Emissive Pyrimidine Ribonucleotides." Chemistry – An Asian Journal 4(3): 419-427.

135

Storz, M. P., C. K. Maurer, C. Zimmer, N. Wagner, C. Brengel, J. C. de Jong, S. Lucas, M. Müsken, S. Häussler, A. Steinbach and R. W. Hartmann (2012). "Validation of PqsD as an Anti- biofilm Target in Pseudomonas aeruginosa by Development of Small-Molecule Inhibitors." Journal of the American Chemical Society 134(39): 16143-16146. Stresemann, C. and F. Lyko (2008). "Modes of action of the DNA methyltransferase inhibitors azacytidine and decitabine." International Journal of Cancer 123(1): 8-13. Stupi, B. P., H. Li, J. Wang, W. Wu, S. E. Morris, V. A. Litosh, J. Muniz, M. N. Hersh and M. L. Metzker (2012). "Stereochemistry of Benzylic Carbon Substitution Coupled with Ring Modification of 2-Nitrobenzyl Groups as Key Determinants for Fast-Cleaving Reversible Terminators." Angewandte Chemie International Edition 51(7): 1724-1727. Swann, P. F., T. R. Waters, D. C. Moulton, Y. Z. Xu, Q. Zheng, M. Edwards and R. Mace (1996). "Role of postreplicative DNA mismatch repair in the cytotoxic action of thioguanine." Science 273(5278): 1109-1111. Takahashi, T., M. Shimizu and S. Akinaga (2002). "Mechanisms of the apoptotic activity of Cl- F-araA in a human T-ALL cell line, CCRF-CEM." Cancer Chemotherapy and Pharmacology 50(3): 193-201. Tallman, M. S. (2006). "New agents for the treatment of acute myeloid leukemia." Best Practice & Research Clinical Haematology 19(2): 311-320. Taoka, M., Y. Nobe, M. Hori, A. Takeuchi, S. Masaki, Y. Yamauchi, H. Nakayama, N. Takahashi and T. Isobe (2015). "A mass spectrometry-based method for comprehensive quantitative determination of post-transcriptional RNA modifications: the complete chemical structure of Schizosaccharomyces pombe ribosomal RNAs." Nucleic Acids Res 43(18): e115. Testi, A. M., M. L. Moleti, F. Giona, L. Annino, S. Chiaretti, I. Del Giudice, E. Todisco, G. D'Elia, A. Ferrari, W. Arcese and F. Mandelli (1997). "A single high dose of idarubicin combined with high-dose ARA-C (MSKCC ALL-3 protocol) in adult and pediatric patients with acute lymphoblastic leukemia. Experience at the University "La Sapienza" of Rome." Haematologica 82(5 suppl): 19-22. Timms, J. F. and P. R. Cutillas (2010). "Overview of quantitative LC-MS techniques for proteomics and activitomics." Methods Mol Biol 658: 19-45. Tiwari, M. (2012). "Antimetabolites: Established cancer therapy." Journal of Cancer Research and Therapeutics 8(4): 510. Vadukoot, A. K., S. F. AbdulSalam, M. Wunderlich, E. D. Pullen, J. Landero-Figueroa, J. C. Mulloy and E. J. Merino (2014). "Design of a hydrogen peroxide-activatable agent that specifically targets cancer cells." Bioorganic & Medicinal Chemistry 22(24): 6885-6892. Vande Voorde, J., S. Liekens and J. Balzarini (2013). "Mycoplasma hyorhinis–Encoded Purine Nucleoside Phosphorylase: Kinetic Properties and Its Effect on the Cytostatic Potential of Purine-Based Anticancer Drugs." Molecular Pharmacology 84(6): 865-875. Vasilyeva, S. V., V. N. Silnikov, N. V. Shatskaya, A. S. Levina, M. N. Repkova and V. F. Zarytova (2013). "SiO2 nanoparticles as platform for delivery of nucleoside triphosphate analogues into cells." Bioorganic & Medicinal Chemistry 21(3): 703-711. Velázquez, S., V. Tuñón, M. L. Jimeno, C. Chamorro, E. De Clercq, J. Balzarini and M. J. Camarasa (1999). "Potential Multifunctional Inhibitors of HIV-1 Reverse Transcriptase. Novel [AZT]-[TSAO-T] and [d4T]-[TSAO-T] Heterodimers Modified in the Linker and in the Dideoxynucleoside Region." Journal of Medicinal Chemistry 42(25): 5188-5196.

136

Veselý, J., J. Seifert, A. Čihák and F. Šorm (1966). "Biochemical changes associated with the development of resistance to 5-azacytidine in AKR leukemic mice." International Journal of Cancer 1(1): 31-39. Vinogradov, S., E. Kohli and A. Zeman (2006). "Comparison of Nanogel Drug Carriers and their Formulations with Nucleoside 5′-Triphosphates." Pharmaceutical Research 23(5): 920-930. Vinogradov, S. V., A. D. Zeman, E. V. Batrakova and A. V. Kabanov (2005). "Polyplex Nanogel formulations for drug delivery of cytotoxic nucleoside analogs." Journal of Controlled Release 107(1): 143-157. Viollet, S., R. T. Fuchs, D. B. Munafo, F. Zhuang and G. B. Robb (2011). "T4 RNA ligase 2 truncated active site mutants: improved tools for RNA analysis." BMC Biotechnol 11: 72. Vitali, L. A., D. Petrelli, C. Lambertucci, M. Prenna, R. Volpini and G. Cristalli (2012). "In vitro antibacterial activity of different adenosine analogues." Journal of Medical Microbiology 61(Pt 4): 525-528. von Rohr, A., S.-F. H. Schmitz, A. Tichelli, U. Hess, D. Piguet, M. Wernli, N. Frickhofen, G. Konwalinka, G. Zulian, M. Ghielmini, B. Rufener, C. Racine, M. F. Fey, T. Cerny, D. Betticher and A. Tobler (2002). "Treatment of hairy cell leukemia with cladribine (2- chlorodeoxyadenosine) by subcutaneous bolus injection: a phase II study." Annals of Oncology 13(10): 1641-1649. Waghmare, S. P. and M. J. Dickman (2011). "Characterization and quantification of RNA post- transcriptional modifications using stable isotope labeling of RNA in conjunction with mass spectrometry analysis." Analytical chemistry 83(12): 4894-4901. Wagner, C. R., V. V. Iyer and E. J. McIntee (2000). "Pronucleotides: toward the in vivo delivery of antiviral and anticancer nucleotides." Med Res Rev 20(6): 417-451. Wang, Q., H. Ding, B. Liu, S. Li, P. Li, H. Ge and K. Zhang (2014). "Addition of the Akt inhibitor triciribine overcomes antibody resistance in cells from ErbB2/Neu-positive/PTEN- deficient mammary tumors." International Journal of Oncology 44(4): 1277-1283. Wang, W.-S., C.-H. Tzeng, T.-J. Chiou, J.-H. Liu, R.-K. Hsieh, C.-C. Yen and P.-M. Chen (1997). "High-Dose Cytarabine and Mitoxantrone as Salvage Therapy for Refractory Non- Hodgkin's Lymphoma." Japanese Journal of Clinical Oncology 27(3): 154-157. Wetzel, C., S. Li and P. A. Limbach (2014). "Metabolic De-Isotoping for Improved LC-MS Characterization of Modified RNAs." Journal of the American Society for Mass Spectrometry 25(7): 1114-1123. Wilson, J. E., J. L. Martin, K. Borroto-Esoda, S. Hopkins, G. Painter, D. C. Liotta and P. A. Furman (1993). "The 5'-triphosphates of the (-) and (+) enantiomers of cis-5-fluoro-1-[2- (hydroxymethyl)-1,3-oxathiolane-5-yl]cytosine equally inhibit human immunodeficiency virus type 1 reverse transcriptase." Antimicrobial Agents and Chemotherapy 37(8): 1720-1722. Winz, M.-L., A. Samanta, D. Benzinger and A. Jäschke (2012). "Site-specific terminal and internal labeling of RNA by poly(A) polymerase tailing and copper-catalyzed or copper-free strain-promoted click chemistry." Nucleic Acids Research 40(10): e78-e78. Wotring, L. L., L. B. Townsend, L. M. Jones, K. Z. Borysko, D. L. Gildersleeve and W. B. Parker (1990). "Dual Mechanisms of Inhibition of DNA Synthesis by Triciribine." Cancer Research 50(16): 4891-4899. Wu, W., B. P. Stupi, V. A. Litosh, D. Mansouri, D. Farley, S. Morris, S. Metzker and M. L. Metzker (2007). "Termination of DNA synthesis by N6-alkylated, not 3'-O-alkylated, photocleavable 2'-deoxyadenosine triphosphates." Nucleic Acids Res 35(19): 6339-6349.

137

Xie, K. C. and W. Plunkett (1996). "Deoxynucleotide Pool Depletion and Sustained Inhibition of Ribonucleotide Reductase and DNA Synthesis after Treatment of Human Lymphoblastoid Cells with 2-Chloro-9-(2-deoxy-2-fluoro-β-d-arabinofuranosyl)adenine." Cancer Research 56(13): 3030-3037. Yao, X., A. Freas, J. Ramirez, P. A. Demirev and C. Fenselau (2001). "Proteolytic 18O labeling for comparative proteomics: model studies with two serotypes of adenovirus." Anal Chem 73(13): 2836-2842. Yates, C. R., E. Y. Krynetski, T. Loennechen, M. Y. Fessing, H.-L. Tai, C.-H. Pui, M. V. Relling and W. E. Evans (1997). "Molecular Diagnosis of Thiopurine S-Methyltransferase Deficiency: Genetic Basis for Azathioprine and Mercaptopurine Intolerance." Annals of Internal Medicine 126(8): 608-614. Yokota, T., K. Konno, S. Mori, S. Shigeta, M. Kumagai, Y. Watanabe and H. Machida (1989). "Mechanism of selective inhibition of varicella zoster virus replication by 1-beta-D- arabinofuranosyl-E-5-(2-bromovinyl)uracil." Molecular Pharmacology 36(2): 312-316. Zhang, J., F. Visser, K. M. King, S. A. Baldwin, J. D. Young and C. E. Cass (2007). "The role of nucleoside transporters in cancer chemotherapy with nucleoside drugs." Cancer Metastasis Rev 26(1): 85-110. Zhang, M., J. A. Wisniewski and H. Ji (2015). "AlphaScreen selectivity assay for β-catenin/B- cell lymphoma 9 inhibitors." Analytical Biochemistry 469(0): 43-53. Zhang, X., A. E. Cozen, Y. Liu, Q. Chen and T. M. Lowe "Small RNA Modifications: Integral to Function and Disease." Trends in Molecular Medicine 22(12): 1025-1034. Zhang, X., E. Motea, I. Lee and A. J. Berdis (2010). "Replication of a Universal Nucleobase Provides Unique Insight into the Role of Entropy during DNA Polymerization and Pyrophosphorolysis." Biochemistry 49(14): 3009-3023.

138

7 APPENDIX A SUPPORTING AND SUPPLEMENTAL NUCLEOSIDE INFORMATION

Synthetic procedures for alcohols Benzyl alcohol, 1-phenylethanol, 1-phenyl-2-methyl-1- propanol, 1-phenyl-2,2- dimethyl-1-propanol, 2,4-dimethyl-3-pentanol, 2-nitrobenzyl alcohol, and 1-(2-nitrophenyl)ethanol were available commercially. 1-(2-Nitrophenyl)-2-methyl-1- propanol, (Litosh, Wu et al. 2011) 1-(2-nitro)phenyl- (Litosh, Wu et al. 2011) and 1-(3- nitro)phenyl- (Barker and Norris 1983) 2,2-dimethyl-1-propanols, diphenylmethanol,

(Pajouhesh, Feng et al. 2010) 2,2,4,4-tertramethyl-3-pentanol (di-tert-butylcarbinol), (Roberts and Hall 1988) 1-(3- methoxy)phenyl-2,2-dimethyl-1-propanol, (Hartmann 1986) 1-(4- methoxy)phenyl-2,2-dimethyl-1-propanol, (Hartmann 1986) 1- (2-bromophenyl)-2,2-dimethyl-1- propanol, (Bickelhaupt, Jongsma et al. 1976) 1-(4-iodo-2-nitrophenyl)-2,2-dimethyl-1-propanol,

2-nitro-(α-phenyl)-, (Storz, Maurer et al. 2012) 3-nitro-(α-phenyl)- (Newman, Izenwasser et al.

1999) and 4-nitro-(α-phenyl)- (Newman, Izenwasser et al. 1999) benzyl alcohols were prepared in accordance with literature protocols. 1-Phenyl-3,3-dimethylbutanol (Maruoka, Saito et al.

1993) and (α -cyclohexyl)benzyl alcohol (Liao and Hu 2011) were prepared by Grignard addition of phenylmagnesium bromide to 3,3-dimethylbutanal and cyclohexanecarboxaldehyde, respectively, whereas 1-(2- methoxyphenyl)-2,2-dimethyl-1-propanol (Nagaradja, Chevallier et al. 2012) and 1-(2-methylphenyl)-2,2-dimethyl-1-propanol (Seyferth, Hui et al. 1993) were prepared by addition of tert-butylmagnesium chloride to o-anyzaldehyde and o- methylbenzaldehyde, respectively, and were identified according to their 1H NMR data given in literature.

Preparation of novel α-substituted benzyl alcohols

A. Transmetallation of 1-iodo-2-nitrobenzene followed by reaction with aldehydes

139

General procedure for transmetallation: To a solution of 1-iodo-2-nitrobenzene in anhydrous tetrahydrofuran (0.4 M) cooled at minus 48°C (dry ice-acetone bath) under argon atmosphere, phenylmagnesium chloride (2 M in THF, 1.0-1.1 eq.) was added at a rate to keep the temperature at or below minus 35 °C. Upon completion of the addition the mixture was stirred for five minutes, then the appropriate aldehyde (1.1 -1.5 eq.) was added. The mixture was allowed to gradually warm up to room temperature, then quenched with saturated aqueous NH4Cl (ca 5 mL) and poured into ethyl acetate (25 mL). The organic layer was separated; aqueous layer was extracted three times with ethyl acetate (25 mL each). Combined organic extract was washed with water (20 mL), dried over anhydrous Na2SO4, concentrated under reduced pressure, and purified by silica gel column chromatography.

(±)-1-(2-nitrophenyl)-3,3-dimethyl-1-butanol: 3.512 g (14.11 mmol) of 1-iodo-2-nitrobenzene

(Smith and Ho 1990) and 1.7 mL (13.54 mmol) of cyclohexaldehyde were used. Column chromatography (hexane/ethyl acetate 20:1 to 10:1) afforded (RS)-1-(2-nitrophenyl)-3,3- dimethyl-1-butanol, 1.35g, 45%, as light yellow oil. 1H NMR (400 MHz, CDCl3): δ 7.85 (dd, 1

H, J = 8.2, 1.4 Hz), 7.82 (dd, 1 H, J = 8.2, 1.4 Hz), 7.63 (dt, 1 H, J = 7.6, 1.4 Hz), 7.40 (m, 1 H),

140

5.41 (m, 1 H), 2.40 (d,1 H, J = 4.2 Hz), 1.70 (AB dd, 1 H, J = 14.5, 8.9 Hz), 1.63 (AB dd, 1 H, J

= 14.5, 2.6 Hz), 1.05 (s, 9 H). 13C NMR (100 MHz, CD3OD): δ 147.34 (C), 141.48 (C), 133.45

(CH), 128.37 (CH), 127.86 (CH), 124.16 (CH), 67.52 (CH), 51.55 (CH2), 30.87 (C), 30.21

(CH3). HRMS (TOF ES+) for [MNa]+ C12H17NO3Na calculated: 246.1106, observed:

246.1100.

(±)-(2-nitrophenyl)-cyclohexylmethanol: 1.16 g (4.678 mmol) of 1-iodo-2-nitrobenzene and

0.623 mL (5.143 mmol) of cyclohexaldehyde were used. Column chromatography (hexane/ethyl acetate 20:1 to 8:2) afforded (RS)-(2-nitropheny)-cyclohexylmethanol, 0.838 g, 76%, as light yellow oil. 1H NMR (400 MHz, CDCl3): δ 7.87(dd, 1 H, J = 8.2, 1.3 Hz), 7.74 (dd, 1 H, J = 7.9,

1.5 Hz), 7.63 (dt, 1 H, J = 7.6, 1.3 Hz), 7.43 (dt, 1 H, J = 7.6, 1.5 Hz), 5.06 (d, 1 H, J = 6.3 Hz),

2.50 (s, 1 H), 1.86 (m, 1 H), 1.75 (m, 4 H), 1.43 (d, 1 H, J = 12.7 Hz), 1.18 (m 4 H) , 0.90 (m,

1H). 13C NMR (100 MHz, CD3OD): δ 148.62 (C), 138.60 (C), 132.92 (CH), 129.05 (CH),

127.98 (CH), 124.24 (CH), 73.45 (CH), 44.06 (CH), 29.79 (CH2), 27.75(CH2), 26.30 (CH2),

26.20 (CH2), 25.97 (CH2). HRMS (ESI+) for [MNa]+ C13H17NO3Na+ calculated: 258.1101, observed: 258.1102.

B. Addition of tert-butylmagnesium chlorides to substituted benzaldehydes

141

General procedure for Grignard addition: To a solution of the appropriate aldehyde in anhydrous tetrahydrofuran (0.4 M) cooled at 0 C (ice-water bath) tert-butylmagnesium chloride (2.0 M in

THF, 1.0-1.2 eq) was added to the mixture at a rate to keep the temperature below 10 C. The mixture was allowed to gradually warm up to room temperature, then quenched with saturated aqueous NH4Cl (ca 5 mL) and poured into ethyl acetate (25 mL). The organic layer was separated; aqueous layer was extracted three times with ethyl acetate (25 mL each). Combined organic extract was washed with water (20 mL), dried over anhydrous Na2SO4, concentrated under reduced pressure, and purified by silica gel column chromatography.

142

(±)-1-(2-cyanophenyl)-2,2-dimethyl-1-propanol: 0.5 g (3.8 mmol) of 2-cyanobenzaldehyde and

1.9 mL (3.8 mmol) of tert-butylmagnesium chloride solution were used. Column chromatography (hexane/ethyl acetate 20:1 to 8:1) yielded (RS)-1-(2-cyanophenyl)-3,3-dimethyl

1-butanol, 0.097 g, 13%, as light yellow. 1H NMR (400 MHz, CDCl3): δ 7.77 (d, 1 H, J= 7.5

Hz), 7.37 (m, 3 H), 5.03 (s, 1 H), 0.90 (s, 9 H). 13C NMR (100 MHz, CD3OD): δ 145.18 (C),

131.28 (CH), 130.19 (C), 128.43 (CH), 123.61 (CH), 122.89 (CH), 90.31(CH), 35.83 (C) , 25.19

(CH3). HRMS (ESI+) for [MH]+ C12H16NO+ calculated: 190.12264, observed: 190.12265.

(±)-1-(2-chlorophenyl)-2,2-dimethyl-1-propanol: 1.41 g (10 mmol) of o-chlorobenzaldehyde and

5 mL (12 mmol) of tert-butylmagnesium chloride solution were used. Column chromatography

(hexane/ethyl acetate 20:1 to 10:2) yielded (RS)-1-(2-chlorophenyl)-3,3-dimethyl-1-butanol, 1.3 g, 92%, as light yellow. 1H NMR (400 MHz, CDCl3): δ 7.56 (AB dd, 1 H, J = 7.8, 1.8 Hz), 7.34

(AB dd, 1 H, J = 8.0, 1.4 Hz), 7.28 (dt, 1 H, J = 7.5, 1.3 Hz), 7.21(AB dt, 1 H, J = 7.7, 1.8 Hz),

5.04 (s, 1 H), 2.15 (br. s, 1 H), 1.0 (s, 9 H). 13C NMR (100 MHz,CD3OD): δ 139.98 (C), 133.44

(C), 129.46 (CH), 129.10 (CH), 128.32 (CH), 126.28 (CH), 76.61 (CH), 36.91 (C), 25.82 (CH3).

HRMS (ESI+) for [MNa]+ C11H15O35ClNa+ calculated: 221.03091, observed: 221.03406;

C11H15O37ClNa+ calculated: 223.03196, observed: 223.03108.

(±)-1-(2,6-dichlorophenyl)-2,2-dimethyl-1-propanol: 2 g (11.4 mmol) of 2,6- dichlorobenzaldehyde and 6.86 mL (13.7 mmol) of tert-butylmagnesium chloride solution were used. Column chromatography (hexane/ethyl acetate 20:1 to 10:2) yielded (RS)-1-(2,6- chlorophenyl)-3,3-dimethyl-1-butanol as light yellow oil. 1H NMR (400 MHz, CDCl3, note: due

143

to restricted rotation, H-3 and H-5 are inequivalent): δ 7.36 (dd, 1 H, J = 8.0, 1.4), 7.31 (dd, 1 H,J

=8.1, 1.4 Hz), 7.15 (t, 1 H, J =8.0 Hz), 5.33 (d, 1 H, J =11.2 Hz), 3.25 (d,1 H, J =11.2 Hz), 1.09

(s, 9H). 13C NMR (100 MHz, CDCl3, note: due to restricted rotation, all aromatic carbons are inequivalent to one another): δ 136.48 (C), 136.01 (C), 133.79 (C), 130.63 (CH), 128.98 (CH),

128.54 (CH), 79.69 (CH), 39.17(C), 27.35 (CH3). HRMS (ESI+) for [MNa]+

C11H14O35Cl2Na+ calculated: 255.01194, observed: 255.01285.

(±)-1-(2,6-dinitrophenyl)-2,2-dimethyl-1-propanol: To a solution of 2,6-dinitroiodobenzene14

(0.597 g, 2.03 mmol) in anhydrous tetrahydrofuran (7 mL) chilled at -78 C 2.0 M phenylmagnesium chloride solution (1.22 mL, 2.44 mmol) were added at a rate that the temperature would not increase -55 C. Upon completion of the addition, the mixture was stirred for another 10 minutes (allowing the temperature to return to -78 C), and trimethylacetaldehyde

(0.463 mL, 4.263 mmol) was added. The reaction mixture was stirred for 2 hours while gradually warming up to room temperature, then quenched by a saturated solution of ammonium chloride

(4 mL). The mixture was diluted with water (10 mL) and extracted with ethyl acetate

(3x25 mL). Combined extracts were washed with water (10 mL), brine (10 mL), dried over anhydrous Na2SO4 evaporated, and the residue was purified by column chromatography

(hexane/ethyl acetate 20:1 to 8:2) to yield (RS)-1-(2,6-dinitrophenyl)-3,3-dimethyl-1-butanol as

144

dark oil. 1H NMR (400 MHz, CDCl3): δ 7.73 (br, 2 H), 7.57 (t, 1 H, J =8.0 Hz), 5.44 (d, 1 H, J

=6.2 Hz), 2.69 (d, 1 H, J = 6.2 Hz), 0.92 (s, 9H). 13C NMR (100 MHz, CDCl3): δ 151.28 (br C),

129.18 (CH), 128.34 (CH), 127.15 (br C), 128.54 (CH), 75.99 (CH), 38.76 (C), 26.32 (CH3).

HRMS (ESI+) for [MNa]+ C11H14N2O5Na+ calculated: 277.07949, observed: 277.07948.

NCI-60 Human Tumor Cell Line Screen

The second generation lead compound 3a was submitted to the National Cancer Institute for the in vitro cell line screening project (IVCLSP). The screening consisted in the evaluation of this compound against the 60 cell lines at a single dose of 10 μM. The human tumor cell lines of the cancer screening panel are grown in RPMI 1640 medium containing 5% fetal bovine serum and

2 mM L-glutamine. For a typical screening experiment, cells are inoculated into 96 well microtiter plates in 100 μL at plating densities ranging from 5,000 to 40,000 cells/well depending on the doubling time of individual cell lines. After cell inoculation, the microtiter plates are incubated at 37° C, 5% CO2, 95 % air and 100 % relative humidity for 24 h prior to addition of experimental drugs. After 24 h, two plates of each cell line are fixed in situ with TCA, to represent a measurement of the cell population for each cell line at the time of drug addition

(Tz). Experimental drugs are solubilized in dimethyl sulfoxide at 400-fold the desired final maximum test concentration and stored frozen prior to use. At the time of drug addition, an aliquot of frozen concentrate is thawed and diluted to twice the desired final maximum test concentration with complete medium containing 50 μg/ml gentamicin. Additional four, 10-fold or ½ log serial dilutions are made to provide a total of five drug concentrations plus control.

Aliquots of 100 μl of these different drug dilutions are added to the appropriate microtiter wells already containing 100 μl of medium, resulting in the required final drug concentrations.

145

Following drug addition, the plates are incubated for an additional 48 h at 37°C, 5 % CO2, 95 % air, and 100 % relative humidity. For adherent cells, the assay is terminated by the addition of cold TCA. Cells are fixed in situ by the gentle addition of 50 μl of cold 50 % (w/v) TCA (final concentration, 10 % TCA) and incubated for 60 minutes at 4°C. The supernatant is discarded, and the plates are washed five times with tap water and air dried. Sulforhodamine B (SRB) solution (100 μl) at 0.4 % (w/v) in 1 % acetic acid is added to each well, and plates are incubated for 10 minutes at room temperature. After staining, unbound dye is removed by washing five times with 1 % acetic acid and the plates are air dried. Bound stain is subsequently solubilized with 10 mM trizma base, and the absorbance is read on an automated plate reader at a wavelength of 515 nm. For suspension cells, the methodology is the same except that the assay is terminated by fixing settled cells at the bottom of the wells by gently adding 50 μl of 80 % TCA

(final concentration, 16 % TCA). Using the seven absorbance measurements [time zero, (Tz), control growth, (C), and test growth in the presence of drug at the five concentration levels (Ti)], the percentage growth is calculated at each of the drug concentrations levels. Percentage growth inhibition is calculated as:

[(Ti-Tz)/(C-Tz)] x 100 for concentrations for which Ti>/=Tz

[(Ti-Tz)/Tz] x 100 for concentrations for which Ti

Three dose response parameters are calculated for each experimental agent. Growth inhibition of

50% (GI50) is calculated from [(Ti-Tz)/(C-Tz)] x 100 = 50, which is the drug concentration resulting in a 50% reduction in the net protein increase (as measured by SRB staining) in control cells during the drug incubation. The drug concentration resulting in total growth inhibition

(TGI) is calculated from Ti = Tz. The LC50 (concentration of drug resulting in a 50% reduction

146

in the measured protein at the end of the drug treatment as compared to that at the beginning) indicating a net loss of cells following treatment is calculated from [(Ti-Tz)/Tz] x 100 = -50.

Values are calculated for each of these three parameters if the level of activity is reached; however, if the effect is not reached or is exceeded, the value for that parameter is expressed as greater or less than the maximum or minimum concentration tested. The full results of 3a screening are outlined below.

147

148

DNA synthesis termination studies

An Eppendorf Mastercycler personal thermocycler was used with the method listed below:

• 1 T = 95 C 4:00 minutes

• 2 T = 52 C 1:00 minutes

• 3 T = 72 C 1:00 minutes

• 4 T = 95 C 1:00 minutes

• 5 Go to 2 Repeat 10

• 6 hold at 4 C

149

Spectra (1H, 13C, DEPT 13C, HRMS), HPLC profiling, and MTT assay data in MCF7 breast cancer cells (by default) and in fibroblast cells (specified)

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

]

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

207

208

209

210

211

212

213

214

215

216

217

218

219

220

221

222

223

224

225

226

227

228

229

230

231

232

233

234

235

236

237

238

239

240

241

242

243

244

245

246

247

248

249

250

251

252

253

254

255

256

257

258

259

260

261

262

263

264

265

266

267

268

269

270

271

272

273

274

275

276

277

278

279

280

281

282

283

284

285

286

287

288

289

290

291

292

293

294

295

296

297

298

299

300

Barker, S. and R. Norris (1983). "Nucleophilic substitution reactions in m-nitrobenzylic substrates." Australian Journal of Chemistry 36(1): 81-95. Bickelhaupt, F., C. Jongsma, P. de Koe, R. Lourens, N. R. Mast, G. L. van Mourik, H. Vermeer and R. J. M. Weustink (1976). "Synthesis and application of precursors of hetero-anthracenes: 2- halo- and 2,2′-dihalodiphenylmethanes and methine substituted derivatives." Tetrahedron 32(15): 1921-1930. Hartmann, R. W. (1986). "Influence of alkyl chain ramification on estradiol receptor binding affinity and intrinsic activity of 1,2-dialkylated 1,2-bis(4- or 3-hydroxyphenyl)ethane estrogens and antiestrogens." Journal of Medicinal Chemistry 29(9): 1668-1674. Liao, Y.-X. and Q.-S. Hu (2011). "CuCl/Bipyridine-Catalyzed Addition Reactions of Arylboroxines with Aldehydes, α,β-Unsaturated Ketones, and N-Tosyl Aldimines." The Journal of Organic Chemistry 76(18): 7602-7607. Litosh, V. A., W. Wu, B. P. Stupi, J. Wang, S. E. Morris, M. N. Hersh and M. L. Metzker (2011). "Improved nucleotide selectivity and termination of 3'-OH unblocked reversible terminators by molecular tuning of 2-nitrobenzyl alkylated HOMedU triphosphates." Nucleic Acids Res 39(6): e39. Maruoka, K., S. Saito, A. B. Concepcion and H. Yamamoto (1993). "Chemoselective functionalization of more hindered aldehyde carbonyls with the methylaluminum bis(2,6- diphenylphenoxide)/alkyllithium system." Journal of the American Chemical Society 115(3): 1183-1184. Nagaradja, E., F. Chevallier, T. Roisnel, V. Jouikov and F. Mongin (2012). "Deprotonative metalation of aromatic compounds using mixed lithium–iron combinations." Tetrahedron 68(14): 3063-3073. Newman, A. H., S. Izenwasser, M. J. Robarge and R. H. Kline (1999). "CoMFA study of novel phenyl ring-substituted 3alpha-(diphenylmethoxy)tropane analogues at the dopamine transporter." J Med Chem 42(18): 3502-3509. Pajouhesh, H., Z. P. Feng, Y. Ding, L. Zhang, H. Pajouhesh, J. L. Morrison, F. Belardetti, E. Tringham, E. Simonson, T. W. Vanderah, F. Porreca, G. W. Zamponi, L. A. Mitscher and T. P. Snutch (2010). "Structure-activity relationships of diphenylpiperazine N-type calcium channel inhibitors." Bioorg Med Chem Lett 20(4): 1378-1383. Roberts, D. D. and E. W. Hall (1988). "Solvent effects on the rates of solvolysis of pinacolyl derivatives." The Journal of Organic Chemistry 53(11): 2573-2579. Seyferth, D., R. C. Hui and W. L. Wang (1993). "Direct, low temperature, in situ nucleophilic aroylation with aroyllithium reagents." The Journal of Organic Chemistry 58(21): 5843-5845. Smith, W. B. and O. C. Ho (1990). "Application of the isoamyl nitrite-diiodomethane route to aryl iodides." The Journal of Organic Chemistry 55(8): 2543-2545. Storz, M. P., C. K. Maurer, C. Zimmer, N. Wagner, C. Brengel, J. C. de Jong, S. Lucas, M. Müsken, S. Häussler, A. Steinbach and R. W. Hartmann (2012). "Validation of PqsD as an Anti- biofilm Target in Pseudomonas aeruginosa by Development of Small-Molecule Inhibitors." Journal of the American Chemical Society 134(39): 16143-16146.

301