Development of Liquid Chromatography-Mass Spectrometric

DEVELOPMENT OF LIQUID CHROMATOGRAPHY-MASS SPECTROMETRIC

ASSAYS AND SAMPLE PREPARATION METHODS FOR THE BIOLOGICAL

SAMPLE ANALYSIS

SUJATHA CHILAKALA

Master of Science in Pharmaceutical Sciences University of Greenwich, UK July 2007

submitted in partial fulfillment of requirements for the degree DOCTOR OF PHILOSOPHY IN CLINICAL-BIOANALYTICAL CHEMISTRY at CLEVELAND STATE UNIVERSITY November 2017

We hereby approve this dissertation for Sujatha Chilakala Candidate for the Doctor of Philosophy in Clinical-Bioanalytical Chemistry degree for the Department of Chemistry and the CLEVELAND STATE UNIVERSITY College of Graduate Studies

______Dissertation Chairperson, Dr. Yan Xu ______Department & Date

______Dissertation Committee Member, Dr. Bin Su ______Department & Date

______Dissertation Committee Member, Dr. Aimin Zhou ______Department & Date

______Dissertation Committee Member, Dr. David Anderson ______Department & Date

______Dissertation Committee Member, Dr. Petru Fodor ______Department & Date

Student’s Date of Defense: November 27th, 2017

Dedicated to my loving husband Amar and to my dearest son Nushyanth

Acknowledgement

This thesis would not have been possible without the support of many people. First and foremost, I would like to thank my adviser, Dr. Yan Xu, for his intellectual supervision, continuous support, endless patience, motivation and encouragement throughout my graduate studies. I am very grateful for the time he spent on helping me with every aspect of this research, for his invaluable suggestions, careful guidance and kindness. I am fortunate to have Dr. Xu, a knowledgeable and excellent professor as my adviser.

I am pleased to thank my committee members, Dr. David Anderson, Dr. Aimin Zhou, Dr.

Bin Su and Dr. Petru S Fodor, for their support, sound advice and fruitful comments on my research. I also would like to thank our collaborator from Cleveland clinic Dr. Yogen

Saunthararajah and his group for providing us samples, support and suggestions. I thank Dr. Xiang

Zhou for his continuous support with instrument training and handling problems. A special thanks to Dr. Mundell, for all his help and support during my duties as a TA.

I would also thank Richelle Emery, Michelle Jones and Janet in the Chemistry office for all their administrative support over the years. I would also like to express my sincere thanks to all of my colleagues and friends for their encouragement and invaluable scientific discussions. A big thanks to all fellow graduate students and other staff at Cleveland State University.

Finally, my special thanks go to my family. Words fail me when I try to express my gratitude towards my family. I am deeply and forever indebted to my husband for his everlasting love, support and encouragement, for giving of himself beyond the call of duty. I would also like to thank my parents, mother-in-law, my siblings for their unconditional love and support. I owe all of my achievements to my family, especially to my husband and my dearest son.

DEVELOPMENT OF LIQUID CHROMATOGRAPHY-MASS SPECTROMETRIC

ASSAYS AND SAMPLE PREPARATION METHODS FOR THE BIOLOGICAL

SAMPLE ANALYSIS

SUJATHA CHILAKALA

ABSTRACT

The area of biosample analysis encompass a very broad range of assays which support the clinical and nonclinical studies. Biosample analysis is used to provide a quantitative or qualitative measure of the active drug and/or its metabolite(s) in the biological matrix for the purpose of pharmacokinetics, toxicokinetics, bioequivalence, and exposure–response (pharmacokinetics /pharmacodynamics) studies. Due to the significance of pharmacological analysis, sensitive, reproducible and robust analytical methods are critically needed for pharmacological studies of the biosamples. A bioanalytical method mainly contains two components I) Sample preparation II) detection of the compound. Therefore, the main aims of this thesis, development of quantitative and qualitative analytical methods for the target compounds using LC-MS(/MS) and development of accelerated sample preparation for high throughput sample analysis for

DNA and proteins.

In this dissertation, a brief review on the method rationale, workflow of the method development, sample preparation methods, instrumentations and analytical method validation, are discussed in Chapter 1. Also, research projects were discussed and the techniques used in the experiments for this thesis were reviewed. As so, chapter II and III were mainly focused on the accelerated sample preparation methods for the high throughput sample analysis of DNA and proteins respectively, where the sample preparation time was significantly reduced from hours to minutes, which are suitable for qualitative and quantitative analysis of DNA and proteins. In Chapter IV, a systematic study on the structural characterization of the model glycoprotein Human IgG was described. In chapter V successful development of LC-MS method was developed for the determination of Oxygen -18 isotope enrichment in the phosphate samples in the positional isotope exchange reactions to study the reversibility of certain enzymatic reactions was described. Successful development and validation of a new and sensitive analytical LC-

MS/MS method for the determination and quantitation of incorporation rates of decitabine, an anti-cancer drug which can be applied to determine the sensitivity and responsiveness in patients treated with decitabine was described in Chapter VI.

TABLE OF CONTENTS

ABSTRACT………………………………...………………………………………...... V

LIST OF TABLES...... XV

LIST OF FIGURES...... XVI

CHAPTER I. INTRODUCTION TO BIOLOGICAL SAMPLE ANALYSIS AND

ANALYTICAL METHOD DEVELOPMENT

1.1 General introduction and research objectives ……...………………….…..…….....1

1.2 Analytical methods for biosample analysis………………………………..……….6

1.2.1 Mass spectrometry………………..……………………………..……….7

1.2.1.1 Ionization methods………………………………………..……...10

1.2.1.2 Mass analyzers……....…………………………………...………13

1.2.1.3 Modes of detection …………………………………………....…20

1.3 Method validation. …………………………….……………………………...…..21

1.4 Conclusion ...………………..……………………………………………….....….26

1.5 References….…………….………………………………………………… ..…….27

VII

CHAPTER II. MICROWAVE-ASSISTED ENZYMATIC HYDROLYSIS OF DNA

FOR MASS SPECTROMETRIC ANALYSIS: A NEW APPROACH TO DNA

HYDROLYSIS IN MINUTES……………………………………………………....29

2.1 Introduction ……………………………….……...………………………………30

2.2 Experimental………………..……………………………………….……….…...32

2.2.1 Chemicals and materials…...………………....…...………...... ………….....32

2.2.2 Solutions…………………………………………………………………….33

2.2.3 Instrumentation………………..…………………………………………….33

2.2.3.1 Microwave digestion system ……………. …………………………33

2.2.3.2 LC-MS/MS system………..…………………………………………34

2.2.3.3 Cell culture and DNA extraction……...…… …………………….....35

2.2.4 Tetra-enzyme digestion mix ………… ………………………...………… 36

2.2.5 DNA digestion………………………………….…………………………..36

2.2.5.1 Microwave-assisted digestion…….………...………..…….……….37

2.2.5.2 Conventional digestion………..……………………………………39

2.2.6 Extraction of dNs from enzyme digest…………………….……………….39

2.2.7 Application to other DNA samples ……..………………….....…… …….39

2.3 Results and Discussion……………………………..………………………...... 40

VIII

2.3.1 Microwave Technology………..……………………………………..……...40

2.3.2 Optimization of parameters for microwave digestion ……..………...….…....41

2.3.3 Tetra–enzyme digestion mix…………………………………..……………...41

2.3.4 Microwave assisted hydrolysis…………….…..……………………………..42

2.3.5 Application of the method to the digestion of DNA extracted from cell lines…………………………………………………………………………...... 49

2.4 Conclusion…………………………………………………………….…………...51

2.5 References……………………………………………………………………….....52

CHAPTER III. AN ACCELERATED PROTEIN SAMPLE PREPARATION

METHOD AND SELECTIVE EXTRACTION OF PEPTIDES FROM PLASMA

FOR LC-MS-BASED PROTEOMICS……………………………………………….55

3.1 Introduction ………………………………………….……...………………….....56

3.2 Experimental…………………………..……………………………………….…...61

3.2.1 Chemicals and materials…………………….…………...... …..………...…..61

3.2.2 Optimization of protein reduction and alkylation time……………………...61

3.2.3 Instrumentation……………………..………………………………………...65

3.2.3.1 Chromatography separation and mass spectrometric conditions……..65

3.2.3.2 Microwave digestion system ………..………………………………..66

3.2.4 Data processing………..………………………...………………….………..66

3.2.5 Peptide extraction…………………………………………………………….67

3.2.5.1 Preparation of spiked plasma samples………………………………..67

3.2.5.2 Peptide extraction protocols….………………………………………68

3.2.6 Instrumentation………..……………………………………………………..69

3.2.7 Data processing……………………………………………………………….70

3.3 Results and Discussion…..…………………..……………………….…………….70

3.3.1 Optimization of sample preparation time…………………….……………..70

3.3.1.2 Application of optimized method on mixture of proteins and serum proteins.77

3.3.3 Comparison of extraction methods……………………………………………81

3.4 Conclusion………………..………………………..……………………………….83

3.5 References…………………..………………………………………………………84

CHAPTER IV. STRUCTURAL CHARACTERIZATION OF GLYCOPROTEIN

BY MASS SPECTOMETRY…………………………………………………………..88

4.1 Introduction …………………………………………….……...……………………89

4.2 Experimental……………………………..……………………………………..…...91

4.2.1 Chemicals and materials…………………………………...... …..…………....91

4.2.2 Preparation of glycoprotein for LC-MS and GC-MS analysis………..……….92

4.2.2.1 Reduction, alkylation and trypsin digestion of glycoproteins…...……..94

4.2.2.2 Deglycosylation of glycoprotein using endoglycosidases……...………94

4.2.2.3 Purification of released N-glycans..……..…..………………………..95

4.2.2.4 Exoglycosidase digestion …………….……..……………….……….95

4.2.2.5 Permethylation of glycans…………………………………………….96

4.2.2.6 Purification of permethylated glycans………………………………...96

4.2.2.7 Preparation of Partially Methylated Aditol Acetates…………………..97

4.2.3 Instrumentation.………………………………………………………………..98

4.2.3.1LC-MS/MS analysis of glycopeptides and peptides…………………....98

4.2.3.2 LC-MS/MS analysis of permethylated glycans…………………………99

4.2.3.3 GC-MS analysis of PMAAs……………………………………………100

4.2.4 Bioinformatics.……………………………………………………………………101

4.3 Results and Discussion……………..…..……………………………………..…….102

4.3.1 Glycoprotein analysis of HumanIgG…...…………………….……………...103

4.3.2 Release and analyss of N-glycans……………………………………………105

4.3.3 Analysis of PMAAs using GC-MS…………………………………………..108

4.4 Conclusion…………………………………………..………………………………111

4.5 References………………………..……..…………………………………………..112

CHAPTER V. ANALYSIS OF OXYGEN-18 LABELED PHOSPHATE TO STUDY

POSITIONAL ISOTOPE EXPERIMENTS USING QTOF LC-MS………………114

5.1 Introduction ………………………………………….……...……………………..115

5.2 Experimental………………………..………………………………………….….117

5.2.1 Chemicals and materials………………...……………….....…..…………....117

5.2.2 Sample preparation…….….…………………………………………………117

5.2.3 Instrumentation……….….………………………………………...………...120

5.2.3.1 LC-MS chromatography separation and mass spectrometric

conditions…………………………………………………………..…..120

5.2.3.2 GC-MS chromatography separation and mass spectrometric

conditions……………………………………………………………….121

5.2.4 Isotope enrichment calculation………………………..……………………..122

5.2.5 Positional isotope exchange with Lon protease………………………………123

5.3 Results …………………………….……….……………………………………….123

5.3.1 Method development ………………………………………………………123

5.3.1.1 LC-QTOF-MS method …………………………………………….123

5.3.1.2 GC-MS method …………..…………….…………………….…….128

5.3.2 Application………………………………………………………………...130

XII

5.3.2.1 Positional isotope exchange with Lon protease…...……..……….…130

5.4 Discussion…….….……….……………………………………………..…………135

5.5 Conclusion………..…………………………………………………….……..…...136

5.6 References…………..……………………………………………………………...137

CHAPTER VI. QUANTIFYING DECITABINE INCORPORATION AND

HYPOMETHYLATION EFFECT OF DECITABINE IN VITRO AND IN VIVO

STUDIES USING QUANTITATIVE LC-MS/MS………………………………….140

6.1 Introduction …………………………………………….……...……………….….141

6.2 Experimental………………………………..………………………………….….145

6.2.1 Chemicals and materials…………………………………….....…..……...... 145

6.2.2 In vitro studies …………………..………………………………………...... 145

6.2.3 In vivo animal studies………………………………………..………………146

6.2.4 Clinical studies………………………………….………..…….…………...147

6.2.5 DNA isolation and hydrolysis……………..………..…….………………....148

6.2.6 LC-MS/MS system………………………………………………………….149

6.2.7 Quantitation of decitabine, dG, dC and 5mdC ……………………...……….150

6.3 Results………………………………………………………………..…………....151

6.3.1 Enzymatic hydrolysis of DNA……….……………………………………...151

XIII

6.3.2 Measurement of decitabine, dG, dC and 5mdC by LC-MS/MS……….…....154

6.3.3Decitabine incorporation vs. hypomethylation of DNA in leukemia cell lines.162

6.3.4 Decitabine incorporation vs. hypomethylation of DNA in animal studies…167

6.3.5 Decitabine incorporation vs. hypomethylation of DNA in clinical studies…169

6.4 Discussion…………………………………………………………………………..171

6.5 Conclusion………..……………………………………………………….………..172

6.6 References…………………………………………………………………….…....174

XIV

LIST OF TABLES

Table 2.1 Various experimental conditions and microwave settings tested for DNA digestion ………………………………………………………………………………...38

Table 2.2 Microwave-assisted enzymatic hydrolysis yields of dNs released from calf thymus DNA, DNA from HL-60 and U-937 cell lines………………………………….50

Table 3.1 Sample preparation protocols tested for microwave digestion methods……...64

Table 3.2 Protein sample preparation and digestion protocols………………………….73

Table 3.3 Comparison of protein sequence coverages (%) for tryptic digest of various

protein samples prepared using different protocols and analyzed by MASCOT

search engine………………………………………………………………….....78

Table 6.1 Accuracy and precision of decitabine, dC, 5mdC and dG calibrators over six

validation batches in digestion matrix…………………………………………..159

Table 6.2 Intra- and inter-run precision and accuracy.……………………...…………160

Table 6.3 Matrix effect and recovery of decitabine from enzyme digestion matrix.…...161

Table 6.4 The effects of decitabine dosage and treatment time on decitabine incorporation

in DNA and DNA hypomethylation…………………………..………………...164

`LIST OF FIGURES

Figure 1.1 Schematic of the mass spectrometry components ….………….……………9

Figure 1.2 Formation of charged gaseous phase ions in ESI …………………….…….12

Figure 1.3 Schematic of a Quadrupole Mass Analyzer ..………………………….…...14

Figure 1.4 The quadrupole ion trap and Schematic of Q-TRAP instrument ….……..….16

Figure 1.5 Schematic of Q-TOF instrument ………………...…………………….…...19

Figure 2.1 LC-MS/MS of DNs relased after enzyme digestion…………….……….....43

Figure 2.2 Representative MRM chromatograms of enzymatic digested products of CT

DNA………………………………..……………………………………….…...44

Figure 2.3 Comparison of hydrolytic yields of dC by various experimental conditions...46

Figure 3.1 Reduction of proteins using DTT, Alkylation of proteins using IAA………..60

Figure 3.2 Time study analysis to optimize the reduction and alkylation of proteins……72

Figure 3.3 Total ion chromatograms and mass spectra of tryptic digest of BSA samples

prepared by different sample reduction and alkylation protocols………..…..…..75

Figure 3.4 Comparison of the number of MS/MS spectra identified for tryptic digest of

BSA samples prepared using different protocols analyzed by MASCOT search

engine…………………………………………………..………………………...76

Figure 3.5 Total ion chromatograms and number f proteins identified from tryptic digest of

serum samples……………..……………………………………………………..80

XVI

Figure 3.6 Extracted ion chromatogram and mass spectra of Stapled peptide extracted from

plasma……………………………………………………………………………82

Figure 4.1 Schematic workflow for the structural analysis if glycoprotein HIgG………93

Figure 4.2 Representative mass spectra of glycoprotein……………………………….104

Figure 4.3 Schematic representation of permethylation of glycans for LC-MS/MS

analysis…………………………………………………………………………106

Figure 4.4 LC-QTOF_MS spectrum of permethylated glycans released from HIgG….107

Figure 4.5 Schematic representation of PMAA derivatization of individual glycan released

by exoglycosidases…………………………………………………………….109

Figure 4.6 GC-MS chromatogram and spectra of PMAAs and Linkage analysis of

glycans………………………………………………………………….……...110

Figure 5.1 Chemical structures of inorganic phosphate and Trimethylsilyl phosphate.119

Figure 5.2 Representative LC-MS total ion chromatogram and mass spectra of inorganic

Phosphate………………………………………………………………….……126

Figure 5.3 Calibration curve established by phosphoric acid standard solutions………127

Figure 5.4 Representative selected ion chromatograms of TMSP for determination of

16O/18O ratios in inorganic phosphate by using GC-MS………………….…….129

XVII

Figure 5.5 Isotope enrichment for the incorporation of 18O into the inorganic phosphate

released from the hydrolysis of ATPase in the presence of isotopically enriched

water……………………………………………………………………………132

Figure 5.6 Time study analysis of isotope enrichment for the incorporation of 18O into

inorganic phosphate from the hydrolysis of ATP by Lon protease…………….134

Figure 6.1 Schematic representation of decitabine incorporation into DNA……………144

Figure 6.2 Schematic representation of DNA hydrolysis by a one-step tetra-enzyme digestion system that includes DNase1, NP1, PDE1 and ALP……………………..…..153

Figure 6.3 Fragmentation spectra of Decitabine, IS, 5mdC, dC, and dG……………….155

Figure 6.4 Representative MRM chromatograms of Representative chromatograms of the

calibration standard spiked with decitabine, 5mdC, 2dC, dG and IS………..….156

Figure 6.5 Calibration curves of decitabine, 5mdC, 2dC, dG and IS……………….…..157

Figure 6.6 Decitabine incorporation and correlation with DNA demethylation in HL-60

and U937 cells……………………………………………………………….….163

Figure 6.7 Time-dependent decitabine incorporation and correlation with DNA

demethylation in MOLM-13 parent and resistant cells………………………....166

Figure 6.8 Decitabine incorporation and DNA demethylation in vivo studies………....168

Figure 6.9 Decitabine incorporation and DNA methylation in clinical study………..…170

XVIII

Chapter I

Introduction to biological sample analysis and analytical method development

1.1 General introduction and research objectives

Biosample analysis is used to provide a quantitative or qualitative measure of the active drug and/or its metabolite(s) in the biological from biological matrix for the purpose of pharmacokinetics, toxicokinetics, bioequivalence, and exposure–response

(pharmacokinetics /pharmacodynamics) studies. Biosample analysis also applies to drugs used for illicit purposes, forensic investigations, anti-doping testing in sports, and environmental concerns. The insight of how drugs distribute, transform, and are removed from the biological system is integral in our understanding of how to effectively treat diseases. The area of biosample analysis can encompass a very broad range of assays which support the clinical and nonclinical studies. The analytical investigation can give information on the toxicity, minimum effective dose, and

pharmacokinetics (metabolism and elimination properties) of drugs and chemicals while furthering drug discovery and development. Pharmacokinetic- pharmacodynamic relationships are playing an increasingly important role in decisions on the rational development and use of new drugs and can provide a detailed knowledge of the mechanism of the drug and a better understanding of the molecular targets on which they act. Due to the significance of pharmacological analysis, sensitive, reproducible and robust analytical methods are critically needed for pharmacological studies of the biosamples. This has driven improvements in technology and analytical methods (1,2). Many factors influence the development of robust bioanalytical methods for analyzing samples from clinical and nonclinical studies which includes the matrices of interest, the range over which analytes need to be measured, the number and structures of the analytes, the physicochemical properties of the analytes, and their stability in the biological matrices from the time of sample draw to analysis also needs to be measured.

A bioanalytical method mainly contains two components I) Sample preparation

II) detection of the compound. Because biological samples are extremely complex matrices comprised of many components that can interfere with good separations and/or good mass spectrometer signals, sample preparation is an important aspect of bioanalytical estimation. Sample pretreatment is required for achieving sufficient sensitivity and selectivity, whereas the time should be kept to a minimum in order to obtain adequate speed. Therefore, there is a clear trend towards integration of sample pretreatment with the separation and the detection (3). Numerous sample preparation techniques have been developed for the bioanalytical purpose. Analytical

determination of biosamples and drugs in biological samples may be accomplished using a variety of techniques and instrumentation like Gas Chromatography(GC) (4),

High-Performance Liquid Chromatography (HPLC) (5), and Capillary Electrophoresis.

The most widespread of all techniques is HPLC, which is complemented or hyphenated with mass spectrometry, spectrophotometry or others.

In this thesis, development of accelerated sample preparation methods for high throughput sample analysis and qualitative and quantitative analytical methods for the target compounds using LC-MS(/MS) are the central themes. The current chapter describes the background and scope of the thesis. Also, the targeted compounds, i.e. decitabine, DNA and proteins were discussed and the instrumentation used in the experiments for this thesis were reviewed.

As so, chapter II is focused on the accelerated sample preparation method for the high throughput sample analysis of DNA. Enzymatic hydrolysis of DNA to deoxyribonucleoside is a common requirement for LC-MS/MS analysis. In most of these enzymatic hydrolysis protocols, sample undergoes sequential digestion and pH adjustments which are time-consuming and laborious with the enzyme incubation times lasting for 6 h to 17 h, which is time-consuming and is a limiting factor in terms of throughput. Thus, development of a fast and highly efficient DNA hydrolysis method can be an excellent feature in the bioanalytical industry of DNA research. The main goal of this work is to conduct a systematic study to develop an accelerated one- step, microwave-assisted enzymatic digestion of DNA with comparable efficiency as conventional digestion methods.

Chapter III describes the development of high-throughput sample preparation methods and peptide extraction methods for proteomic analysis. Proteins are the important components signaling pathways. Endogenous proteins have been implicated as potential biomarkers for a variety of diseases. Exogenously delivered biotherapeutics are also gaining popularity, and thus their stability and pharmacokinetic properties in vivo must be investigated. The analysis of proteins and peptides are, therefore, an important analytical pursuit for a variety of scientific areas. Although high-throughput and automated LC-MS methods are available for protein characterization, sample preparation oftentimes is the rate-limiting step for many proteomic workflows. In this study, we demonstrated an accelerated method for protein sample preparation by combined use of ultrasound and microwave technologies where the entire sample preparation took less than 30 min to complete. Also, various peptide extraction methods were investigated to achieve maximum selectivity and recovery of spiked exogenous peptides from the biological matrix and to perform stability studies on the peptides in serum/plasma.

Chapter IV is focused on the qualitative analysis of the glycoprotein to elucidate the structure of glycoprotein and glycans in the Human IgG glycoprotein.

Glycosylation is one of the most important post translational modifications of proteins.

Aberrant glycosylation has been correlated to a number of disease conditions including cancer, Alzheimer’s, arthiritis etc. Glycoprotein analysis can provide valuable information about aberrations in the glycosylation which can be correlated the etiology of the disease state and can be used as potential biomarkers. Currently, liquid chromatography interfaced to mass spectrometry is the most widely used techniques

for the characterization of glycans and glycopeptides for characterization of glycoprotein. Development of reliable methods is required for correlation of these changes with the disease condition. In this study, Human IgG was used as a model glycoprotein to develop sample preparation and analysis methods for complete structural characterization of glycoprotein and glycans.

Chapter V describes successful development and optimization of a new analytical HPLC-MS method for the determination of Oxygen -18 isotope enrichment in the phosphate samples to study the reversibility of certain enzymatic reactions. The positional isotope exchange (PIX) technique has proven to be a valuable tool for both the identification of reaction intermediates and reaction reversibility and in the elucidation of the mechanistic pathways of many enzyme-catalyzed reactions. The general approach involves reactions being conducted in an isotope-enriched buffer solution so that the extent of the incorporation of the isotopes into the intermediates or products generated by the enzymatic reactions is used to evaluate the kinetics of the reaction or the extent of the reversibility process. In oxygen transfer PIX reactions, labelled oxygen (18O) is exchanged with the oxygen bonded to an inorganic phosphate, and the measurement of the 18O/16O ratio is used to determine the reaction mechanism and kinetics. Gas chromatography–mass spectrometry (GC-MS) of a derivatised form of the product is the most commonly used method for detecting 18O/16O isotope enrichment. The main disadvantage of using GC-MS is that a volatile derivative of the isotope labeled compound needs to be prepared for analysis, and the process is time- consuming. The focus of this study to develop a method for the determination of oxygen-18 labeled phosphate so that the positional isotope experiments using sensitive

and rapid liquid chromatography–QTOF–mass spectrometry (LC-QTOF-MS) experiments can be carried out.

Chapter VI describes successful development and validation of a new and sensitive analytical LC-MS/MS method for the determination and quantitation of incorporation rates of decitabine. Decitabine is a DNA hypomethylating drug, and is one of the two drugs approved for treatment of all subtypes of the myeloid malignancy myelodysplastic syndromes (MDS). The mode of action of decitabine strictly depends on the incorporation of the drug into DNA, however, a practical method to measure the incorporation is un-available. This work proposes a fully validated, sensitive and specific LC-MS/MS method to simultaneously measure decitabine incorporation and

DNA hypomethylation which may be useful to study the target engagement of the anti- cancer drug decitabine. DNA incorporated decitabine along with other nucleosides were released using enzymatic digestion and were separated by a reverse-phase C18 column and quantified by mass spectrometry, with a lower limit of quantitation at 1.00 nM. Dose and time-course studies of decitabine DNA incorporation and hypomethylation were conducted using human myeloid leukemia cell lines in vitro and using clinical samples in vivo.

1.2 Analytical methods for biosample analysis

Nowadays, the use of liquid chromatography-(tandem) mass spectrometry (LC-

MS(/MS)) for both qualitative and for the absolute quantitation of target compounds is widespread and probably one of the most applied methodologies in modern bio-

analytical work. Especially for small molecular mass compounds, there are many well- described approaches, and quality assurance guidelines with regard to their development, routine application, and validation. In order to provide accurate and reliable data, LC-MS/MS method usually includes four elements: Biological sample extraction in order to exclude matrix effects, liquid chromatography (LC) separation of the analyte(s) from other interferences compounds, mass spectrometry detection of the analyte(s), and analytical method validation or method performance evaluation.

Throughout the work, Q-TRAP MS was used both small molecule quantitation, QTOF-

MS was used for qualitative analysis of protein and quantitative analysis of phosphate.

Therefore it is worthwhile to introduce certain aspects used in our work in both the fields of LC-MS/MS and quality assurance of an analytical method. Instrumentation used in the research work is described in more details in the flowing sections.

1.2.1 Mass spectrometry

The detector is the part of the chromatographic system that responds to the presence of the solutes in the mobile phase. In HPLC there are a number of different detectors used today including Refractive index, UV/Vis, Mass Spectrometry (MS) and

Fluorescence. Less common but important detectors are conductivity and Evaporative light scattering. Since mass spectrometry is the most widely used detection system in this thesis, I will focus my discussion on MS detection in this chapter.

Mass spectrometry (MS) combined with the separation power of chromatography has revolutionized the way chemical analysis is done today. LC-MS

is an extremely versatile instrumental technique. With a selection of LC–MS interfaces now available, a wide range of analytes, from low-molecular-weight drugs and metabolites (100 000 Da), may be studied (6,7). The mass spectrometer provides the most definitive identification of all of the HPLC detectors. It allows for the molecular weight of the analyte to be determined and with certain mass spectrometers, the structure can be elucidated also. Also, the high selectivity of the mass spectrometer often provides identification capability on chromatographically unresolved or partially resolved components. This selectivity allows the use of isotopically labeled analytes as internal standards and this, coupled with high sensitivity, allows very accurate and precise quantitative determinations to be carried out (8,9).

Mass Spectrometer (MS) separates gaseous ions based on their mass to charge ratios (m/z). MS is composed of five parts as illustrated in Figure 1.1:

• Sample inlet system

• Ionization technique

• Mass analyzer

• Ion detector

• Data system

Figure 1.1 Schematic of the mass spectrometry components (Reference 9).

Components eluting from the chromatographic column are introduced into the mass spectrometer by means of a specialized ion source. Interfacing the two techniques is not straightforward as the solutes leaving an LC column are dissolved in mobile phase at atmospheric pressure, whereas the MS is set up to detect gas phase ions in a vacuum. To ease this transition from the liquid to the gas phase, a number of ion source interfaces have been developed.

1.2.1.1 Ionization methods

Ion sources used depends on the type of the molecules to be ionize (10). There are many ionization methods available today (Fig. 1.1), but this work will focus on the most commonly used electrospray ionization (ESI) method.

ESI is a soft ionization technique that has become the most popular ionization technique. The electrospray is created by putting a high voltage on a flow of liquid at atmospheric pressure; sometimes this is assisted by a concurrent flow of gas.

Electrospray ionization is an electrophoretic process and consists of a step that includes charged droplets formation and conversion of these droplets to charged gas phase ions.

Due to the presence of an electric field and higher temperature on the sample inlet, there is increased rate of liquid evaporation. This results in the high-charged density on the droplet followed by increased repulsions as well as surface instability (Rayleigh instability), resulting in the formation of charged gaseous phase ions (Taylor cone)

(Fig. 1.2). For larger molecules, the ions may contain multiple charges, allowing the detection of very large molecules on analyzers that have limited mass to charge (m/z))

ratio ranges (10,11). Because of the natural use of a flowing liquid, it is easily adapted to liquid chromatography (LC).

Electrospray ionization holds many advantages over other ionization techniques, specifically high reproducibility, operation at atmospheric pressure and minimum fragmentation of the analyte in the sample inlet. The ESI can also be operated over a fairly large range of flow rates, making it the most suitable ionization technique to interface to HPLC. Depending on the types of the analyte, ESI can be operated either in positive or negative ion mode. Positive ion formation occurs with the analyte abstracting a proton from the acidic mobile phase components yielding an

[M+H] + ion. Similarly, negative ion formation is induced by the basic mobile phase components abstracting a proton from the analyte and yielding an [M-H]- ion.

Figure 1.2 Formation of charged gaseous phase ions in ESI (reference 10).

1.2.1.2 Mass analyzers

A mass analyzer measures gas phase molecules with respect to their mass-to- charge ratio (m/z), where the charge is produced by the addition or loss of a proton(s), cation(s), anions(s) or electron(s). The addition of charge allows the molecules to be affected by electric fields thus allowing its mass measurement. There are a number of types of mass analyzers available, and the choice depends on the information required from the ionized analytes (Fig. 1.1). Based on mass range limit, analysis speed, mass accuracy, and resolution, there are different types of mass analyzers are used. In this thesis, Q-TRAP (triple quadrupole linear ion trap mass spectrometer) and Q-TOF

(quadrupole–time-of-flight tandem mass spectrometer) were used.

Q-TRAP mass spectroscopy

Quadrupole mass analyzers have been important in the mass analysis for many decades because they are relatively inexpensive, rugged, and have been implemented in a wide variety of instrumental configurations including triple quadrupole instruments which can do MS/MS experiments. A quadrupole mass analyzer consists of four parallel rods that have fixed DC and alternating RF potentials applied to them (Fig.

1.3). Ions produced in the source are focused and passed along the middle of the quadrupoles. The motion of these ions will depend on the electric fields so that only ions of a particular mass to charge ratio (m/z) will have a stable trajectory and thus pass through to the detector (10,11). Varying the RF brings ions of different m/z into focus on the detector and thus builds up a mass spectrum.

Figure 1.3 Schematic of a Quadrupole Mass Analyzer (reference 11)

Ion trap analyzers work similarly to the quadrupole but have a 3D geometry

(Fig. 1.4A). Ions from the source are introduced to the ion trap and form stable 3D trajectories. Similar to the quadrupole, ions with specific m/z can be selected, but unlike the quadrupole, the ions are trapped and expelled to the detector (11). A disadvantage to using the ion trap is the formation of the “ion cloud” inside the trap itself. Ions not only respond to the RF oscillations of the ring and endcap electrodes, but they repel each other as well. To overcome this, extra inert gas (N2, He2) is added to collide with and offer collisional cooling. On cooling, the ions remain stable and avoid expanding their trajectories to collide with the sides of the trap.

The ion trap is essentially a modified quadrupole with additional lenses at the beginning and end tuned to the same polarity as the analysis (positive for positive ions, and negative for negative ions). The like polarities repel each other thus trapping the ions (12). The multiplexing of mass analyzers can increase their functionality and sensitivity. Aligning two quadrupoles in sequence with a collision cell in between allows the first quadrupole (Q1) to filter the molecular ion of choice, send it to the collision cell for fragmentation by collision-induced dissociation (CAD), and finally, all fragments or the most abundant (or unique) fragment ion may be scanned by the second quadrupole (Q2) (Fig.1.4B). A common mass spectrometric experimental setup in quantitative analysis has Q1 set to the molecular ion of choice, fragmentation by

CAD, and Q3 selection of a fragment ion. But just as the ions are able to be trapped, they are able to be fragmented inside the trap. These fragments oscillate inside and can be fragmented again before ejection and analysis. This Multiple Reaction Monitoring

(MRM, also denoted MS/MS) is very sensitive and highly specific.

Figure 1.4A The quadrupole ion trap (reference 10).

Figure 1.4B Schematic of Q-TRAP instrument.

Q-TOF mass spectroscopy

Time of Flight (TOF) analyzers separate and distinguish ions of differing m/z based on their flight times through a field-free vacuum (11). The TOF analyzer is the simplest type of mass analyzer. TOF systems require a pulsed ion source, a flight tube, and an ion detector. This system relies on the fact that if all of the ions produced in the source of a mass spectrometer, by whatever technique, are given the same kinetic energy then the velocity of each will be inversely proportional to the square root of its mass. As a consequence, the time taken for them to traverse a field-free region (the flight tube of the mass spectrometer) will be related in the same way to the m/z of the ion. A complete mass spectrum is obtained simply by allowing sufficient time for all of the ions of interest to reach the detector. Unlike quadrupole mass analyzers where scanning RF and DC voltages guide ions of differing m/z through an electromagnetic region to the detector, TOF analyzers distinguish each m/z present in discrete packets of ions as they arrive at the detector. By varying the conditions in the mass analyzer, e.g. magnetic field, quadrupole field, etc., ions of different m/z values are brought to the detector and a corresponding mass spectrum obtained (Fig. 1.5).

In the time-of-flight instrument, it is essential that ions of all m/z ratios present in the source are transferred, simultaneously and instantaneously, into the mass analyzer at a known time so that their times of flight, and thus their m/z ratios, may be determined accurately. Were ions to be introduced continuously it would be impossible to determine exactly when each began its passage through the flight tube and therefore to calculate its m/z ratio. A complete mass spectrum at a specific time is therefore obtained and when this has been recorded, a matter of milliseconds later, a further set of ions can be transferred

from the source. This is sometimes referred to as a ‘pulsed’ source. Fast scanning, only limited by the time it takes the heaviest ion to travel from the source to the detector, is possible and any distortion of ion intensity brought about by changes in analyte concentration during the scanning process is removed. The mass of an ion is related to the time it takes that ion to travel through the flight tube to the detector. Recently developed

TOF systems have demonstrated high sensitivity and high resolution. Another attribute of

TOF is the ‘scan speed’ benefits. Because TOF scan times for complete mass spectra are so fast, chromatographic techniques that separate complex mixtures in short intervals can be interfaced with this technique. The other great advantage of TOF is the virtually unlimited mass range when sampling in the time domain (11).

The instrument consists of three quadrupoles, Q0, Q1 and Q2, followed by a reflecting TOF mass analyzer with orthogonal injection of ions. Q1 selected certain parent ions. Q0 and Q2 are not real mass filters. They are used to focus ions, and Q2 is also a collision cell. The fragments generated by CID can then be further resolved and analyzed by TOF. Due to its high mass resolution and the wide m/z analysis range, Q-TOF is often utilized in the unknown compound identification and analysis of large molecules such as peptides, proteins, and oligonucleotides.

Figure 1.5 Schematic of Q-TOF instrument (reference 11)

1.2.1.3 Modes of Detection

Typically, the mass spectrometer is set to scan a specific mass range. This mass scan can be wide as in the full scan analysis or can be very narrow as in selected ion monitoring (12). A single mass scan can take anywhere from 10 ms to 1 s depending on the type of scan. Many scans are acquired during an LC-MS analysis. LC-MS data is represented by adding up the ion current in the individual mass scans and plotting the ‘total’ ion current as an intensity against time. Different modes of acquiring LC-MS data are available including full scan acquisition resulting in the typical total ion current plot, selected ion monitoring and selected reaction monitoring or multiple reaction monitoring.

Selected reaction monitoring and multiple reaction monitoring are essentially identical techniques from different manufacturers.

The total ion current (TIC) is a plot of the total number of ions in each MS scan plotted as an intensity point against time. In the TIC plot, ions of every mass over the chosen range are plotted. As many compounds have the same m/z, it can be difficult finding the compound of interest. A specific mass can later be selectively extracted, but sensitivity is not as good as the next technique of selected ion monitoring. TIC plots are often overlaid onto UV or other plots, and this can give useful information about the compounds being studied.

With selected ion monitoring (SIM) the mass spectrometer is set to scan over a very small mass range, typically one mass unit. This is obviously most useful when the actual mass is known. Compounds with the selected mass only are detected and plotted.

However, some compounds have the same mass and in ESI where there are multiple charged species the likelihood is high for compounds to have the same m/z value. As the

mass spectrometer can dwell for a longer time over a smaller mass range, this makes SIM more sensitive that TIC.

Selected reaction monitoring (SRM) and multiple reaction monitoring (MRM) are the most common methods used for performing mass spectrometric quantitation.

SRM/MRM creates a unique fragment (product) ion that can be monitored and quantified in the midst of complicated matrices, thus enabling more confirmatory identification. SRM plots usually contain only a single peak rendering it ideal for sensitive and specific quantitation. In SRM/MRM mode, two stages of mass filtering are employed on a triple quadrupole mass spectrometer. In the first stage, an ion of interest (the precursor) is preselected in Q1 and induced to fragment by collisional excitation with a neutral gas in a pressurized collision cell (q2). In the second stage, instead of obtaining full scan MS/MS where all the possible fragment ions derived from the precursor are mass analyzed in Q3, only a small number of sequence-specific fragment ions (transition ions) are mass analyzed in Q3. This targeted MS analysis using MRM enhances the lower detection limit as compared to full scan MS/MS analysis by allowing rapid and continuous monitoring of the specific ions of interest. Particular precursor ions and product ions are selected for detection based on their unique pathways; MRM provides a great improvement in signal to noise ratio.

1.3 Method validation

All analytical assays must undergo precise and systematic validation before implementation into routine use. Validation determines the functionality of the assay, the validity of the results and whether the analytical method is suitable for the intended purpose. Although there are many different standards for method validation, the most

commonly accepted one in the pharmaceutical industry in the United States is the

“Guidance for Industry Bioanalytical Method Validation” proposed by FDA (13,14).

According to FDA, several factors about the method and the analyte have to be considered when validating a bioanalytical method. Analytical validation of a method includes tests to confirm assay selectivity, sensitivity, recovery, calibration curve, accuracy, precision, and stability. Besides, as the matrix of the biological samples and, even the matrix of the post-extract samples, may affect the signal of the analyte, evaluation of matrix effect has also become a must in the method validation. Each of these parameters should be investigated carefully before implementation of the assay.

Selectivity studies how specific the bioanalytical method is toward the analyte.

Selectivity is the ability of an analytical method to differentiate and quantify the analyte in the presence of other components in the sample. As such, selectivity measures the

“trueness” of the detection signal and describes the plausibility that the MS(/MS) signal is not only produced by the component of interest but also by other components. The interferences coming from the same type of blank biological samples, yet from at least six different sources have to be tested. The signal of the interference caused by nonspecific responses should be evaluated at the level of the lower limit of quantification (LLOQ). In another word, the non-specific response of the detector should not affect the accurate quantification of the LLOQ.

Sensitivity of an analytical procedure is the lowest amount of analyte in a sample that can be quantitatively measured with suitable accuracy and precision. A method’s sensitivity is described by the LLOQ. The sensitivity of a method should be tested by

analyzing at least 5 replicates of the sample at the LLOQ concentration on at least one of the validation days where the accuracy and precision should be lesser than or equal to 20%.

Recovery is a measure of yield after sample preparation and thus is an evaluation of the effectiveness of the extraction methods. Recovery can be estimated by adding a known amount of the analyte of interest to the sample and calculating the recovery after sample preparation. It can be obtained by comparing the signals of the analyte in the extracted sample and in the pure, authentic analyte solution spiked in the post-extraction blank matrix. To determine the recovery of the extraction method accurately, at least three concentration levels (i.e., a low, a medium, and a high) throughout the calibration range should be evaluated. At each concentration level, three replicates of the samples should be prepared. The mean value and the standard deviation of the measurements should be recorded and reported.

Calibration curve: A standard curve of calibrators is created to determine the concentration range of the analysis, and three quality control

(QC) samples. The calibrators must be prepared in the same biological matrix as the real samples. Usually, the calibrators should be prepared by spiking known amount of analyte into the biological matrix. For a linear calibration curve, at least six non-zero calibrators should be included in the curve. A double-blank sample with no addition of analyte or the internal standard (IS) should be included. Besides, a zero sample, with no addition of analyte, yet containing the same concentration of IS as all the non-zero calibrators, should also be included. The accuracy and precision of each calibrator should be evaluated. The accuracy of each calibrator is determined by the percent error of the calculated concentration (equation 1.1).

[퐴푛푎푙푦푡푒]푚푒푎푠.−[퐴푛푎푙푦푡푒]푛표푚푖. 퐴푐푐푢푟푎푐푦 = X 100 (1.1) [퐴푛푎푙푦푡푒]푛표푚푖.

Here (Analyte)meas. represents the measured concentration of the analyte from the calibration curve. (Analyte)nomi. indicates the nominal concentration of the analyte.

The inter-assay precision of each calibrator is determined by the coefficient of variation (CV%) from several measurements toward the same calibrator (equation 1.2).

푆푡푎푛푑푎푟푑 푑푒푣푖푎푡푖표푛 퐶푉% = X 100 (1.2) 푀푒푎푛

Sometimes, the same calibration curve is repeated for several times on different days, so that the inter-assay precision of the calibrators can be determined. By dividing the standard deviation of the calculated values with the average of the calculated values, the inter-assay precision can be obtained. For any point on the calibration curve, except the LLOQ, the accuracy and precision should be within ±15%. For LLOQ, the values should not exceed ±20%.

Accuracy of a method describes the closeness of mean test results obtained by the method to the true value (concentration) of the analyte. As such, accuracy must be determined by replicate analysis of samples containing known amounts of the analyte. The mean value should be within 15%, except at the lower limit of quantitation (LLOQ), where it should not deviate by more than 20%.

Precision of an analytical method describes the closeness of individual measures of multiple analyses of a single concentration level. Again, a minimum of 5 analyses per concentration level in the desired analytical range is recommended, and the result should not exceed 15% (20% at the LLOQ) in terms of the coefficient of variation.

To evaluate the accuracy and precision of a method, a set of quality control (QC) samples are needed. These QC samples usually include at least three concentration levels

(low, medium, and high). For each concentration level, five parallel replicates should be prepared. Since the accuracy and precision at the LLOQ should also be evaluated, if the low concentration QC samples (LQCs) are not the same with the LLOQ, another five replicates of LLOQ should be prepared.

During real sample analysis, there is a possibility that some of the real samples possess concentrations either higher or lower than the upper or lower limit of quantification. In these occasions, diluted or concentrated quality control samples are also needed for the accuracy and precision studies.

Stability of an analyte in a biological fluid is a function of the storage conditions and the chemical properties of the analyte. It is extremely important to perform a stability study of the analyte in biological fluids in order to obtain information concerning the conditions and times of sample storage so that sample integrity before assay is assured.

Chemical compounds can decompose prior to chromatographic investigations, for example, during the preparation of the sample solutions, extraction, cleanup, phase transfer or storage of prepared vials (in refrigerators or in an automatic sampler). Method development should investigate the stability of the analytes and standards under these circumstances. The stability should hereby be determined during sample collection and

handling, after short (benchtop) and long-term storage, and after going through different freeze and thaw cycles post-preparative stability, and stock solution stability. The short- term, long-term, and freeze and thaw stabilities illustrate the stability of the analyte in the biological matrix. The post-preparative stability analyzes the stability of the analyte in the post-extraction sample matrix. The stock solution stability measures the stability of the analyte in high concentration pure stock solutions.

1.4 Conclusion

In this chapter, a brief discussion has been made on the importance of biosample analysis. Then the targeted analytes/biosamples with different unsolved problems were introduced; LC-MS/MS methodologies on how to solve these problems were proposed. To emphasize the problem-solving ability of LC-MS/MS in the pharmacological studies of the biosample, sample preparation methods, biological sample extraction, the underlying theories of the HPLC separation techniques along with the mass spectrometry detection were outlined and discussed. Also, bioanalytical method validation has been reviewed briefly.

1.5 References

1. Vegvari A, Marko-Varga G. Clinical protein science and bioanalytical mass

spectrometry with an emphasis on lung cancer. Chem Re.v 2010; 110:3278-3298.

2. Khojasteh SC, Wong H, Hop C. Drug metabolism and pharmacokinetics quick

guide, 1st ed: Springer. 2011.

3. Backes D. Strategy for the development of quantitative analytical procedures, In

Principles and Practice of Bioanalysis, Venn RF, Taylor & Francis, NY., 2000;

342-358.

4. Koh, H. L.; Yau, W. P.; Ong, P. S.; Hegde, A. Current Trends in Modern

Pharmaceutical Analysis for Drug Discovery. Drug Discov Today., 2003; 8, 889.

5. Ermer, J. and M. Vogel, Applications of hyphenated LC-MS techniques in

pharmaceutical analysis. Biomedical chromatography: BMC., 2000; 14[6]: p. 373-

83.

6. Ardrey, R.E., Interface Technology. Liquid Chromatography - Mass Spectrometry:

An Introduction. John Wiley & Sons, Ltd, 2003; pp. 75-127.

7. Chen, G., Zhang, L. and Pramanik, B.N., LC/MS: Theory, Instrumentation, and

Applications to Small Molecules. HPLC for Pharmaceutical Scientists. John Wiley

& Sons, Inc. 2006; pp. 281-346.

8. Sparkman, O. D. Mass Spectrometry Desk Reference. J. Chem. Educ., 2001; 78,

168. Ashcroft, A. E. Ionization Methods in Organic Mass Spectrometry; Royal

Society of Chemistry: North Yorkshire, UK, 1997.

9. Herbert, C. G.; Johnstone, R. A. W. Mass Spectrometry Basics; CRC: Boca Raton,

FL, 2003.

10. Busch, K. L.; Glish, G. L.; McLuckey, S. A. Mass Spectrometry: Techniques and

Applications of Tandem Mass Spectrometry; VCH: New York, NY., 1988.

11. GATES, P., Quadrupole Mass Analysis. Available:

http://www.chm.bris.ac.uk/ms/newversion/quad-ms.htm., 2009.

12. Cotter, R. J. Time-of-flight Mass Spectrometry: Instrumentation and Applications

in Biological Research; American Chemical Society: Washington, DC., 1997.

13. Surendra, B; Anthony, D. Key Elements of Bioanalytical Method Validation for

Small Molecules. The AAPS Journal., 2007; 9, 109-114.

14. Bashaw ED, DeSilva B, Rose MJ, Wang Y-MC, Shukla C. Bioanalytical Method

Validation: Concepts, Expectations and Challenges in Small Molecule and

Macromolecule—A Report of PITTCON.,2013; Symposium. The AAPS Journal.

2014;16(3):586-591. doi:10.1208/s12248-014-9597-4.

Chapter II

Microwave-assisted enzymatic hydrolysis of DNA for mass spectrometric analysis: A new approach to DNA hydrolysis in minutes

Abstract

Though high-throughput methods are available for the analysis of DNA, enzymatic hydrolysis oftentimes is a rate limiting step. Conventional enzymatic hydrolysis of DNA using a multi-enzyme mixture will take a minimum of 6-17 h to complete the hydrolysis.

An enzymatic digestion of a DNA can be accelerated and made more efficient through microwave-assisted digestion. In this paper, a systematic study was conducted to explore the efficiency of microwave-assisted enzymatic digestion of DNA using non-restriction enzymes as compared with the conventional method. In addition, the optimum experimental parameters for the digestion such as microwave irradiation power, digestion time and reaction temperature were studied. It was determined that efficient digestion of

DNA was attained in 30 min with a performance similar to that obtained by the conventional overnight digestion procedure. Thus, this new application of microwave technology to DNA enzymatic digestion will hasten the application of DNA analysis in biological and clinical research.

2.1 Introduction

Enzymatic hydrolysis of DNA is a routine sample preparation method for the analysis of nucleosides, DNA adducts, and antimetabolites as nucleoside analogs incorporated into DNA. The most commonly used enzymes are bovine pancreatic deoxyribonuclease I (DNase I), micrococcal nuclease (MN), nuclease P1 (NP1), phosphodiesterase I (PDE I), phosphodiesterase II (PDE II), and alkaline phosphatase

(ALP). If a nucleotide or adduct is to be released from the DNA backbone, DNase plus

NP1 and/or PDE I or MN plus PDE II (1-3) are commonly used. Both these enzyme combinations completely hydrolyze DNA and thus release the nucleotides or adducts as free small molecules. Furthermore, ALP is sometimes employed in the digestion system for the removal of phosphate groups to simplify liquid chromatography (LC) method development.

In conventional methods of digestion, DNA samples are incubated in a water bath for durations ranging from a few hours to overnight at temperatures optimum for the enzyme or enzyme mixture to exhibit maximum activity. Most DNA hydrolysis enzymes are employed sequentially; therefore, the samples can be incubated for different periods depending on the enzyme employed for that step (4). Furthermore, the incubation time

depends on the stability and concentrations of the enzyme and substrate. Overall, conventional digestion methods are very time consuming, which is a limiting factor in terms of throughput. Thus, the development of a fast and highly efficient one-step DNA hydrolysis method would be of enormous benefit to DNA research and the bioanalytical industry.

In recent years, microwave-assisted digestion has emerged as an alternative to conventional enzymatic digestion, shortening the duration of the digestion process to several minutes (5, 6). For example, Giorgio et al. reported microwave-assisted chemical hydrolysis of DNA using aqueous formic acid or hydrochloric acid at high temperatures of

140–160°C (7), while acceleration of the enzyme hydrolysis of DNA using microwave technology with restriction endonucleases was reported by Jhingan (U.S. Pat. No.

5,350,686) and Rakha Hari Das (EP1521826 A1).

Thus, microwave-assisted enzymatic digestion of DNA is a promising approach for rapid and efficient hydrolysis of DNA. However, to our knowledge, no accelerated enzymatic method has been developed for the complete hydrolysis of DNA to the single- nucleoside level using non-restriction enzymes. In the present work, using calf-thymus

DNA (CT DNA) as a model, a very fast and efficient one-step, multi-enzyme digestion of

DNA using microwave technology was developed for the first time. Considering the stability of non-restriction enzymes and DNA under high temperature and power conditions, different DNA samples were hydrolyzed using non-restriction enzymes and low power settings. The resultant deoxyribonucleosides (dNs) were analyzed using reverse-phase high-performance liquid chromatography-tandem mass spectrometry

(HPLC-MS/MS), and the results were compared with those obtained by conventional digestion.

2.2 Experimental

2.2.1 Chemicals and materials

Bis(2-hydroxyethyl) amino-tris (hydroxymethyl) methane (BIS-TRIS), deoxyribonuclease I type II (DNase I), nuclease P1 (NP1), bovine ALP, 2′-deoxycytidine

(dC), 2′-deoxyguanosine (dG), 2′-deoxythymidime (dT), 2′-deoxyadenosine (dA), CT

DNA, and formic acid were obtained from Sigma Aldrich (St. Louis, MO, USA). Snake venom phosphodiesterase I (PDE I) was obtained from Worthington Biochemical

15 (Lakewood, NJ, USA). 2'-Deoxycytidine N3 (96–98%), a stable heavy isotope of 2’- deoxycytidine, was purchased from Synthèse AptoChem (Montreal, Quebec, Canada).

NaCl, ZnCl2, phosphate buffered saline (PBS, 10X, pH 7.4), phenol saturated with Tris buffer (pH 6.6), and HPLC-grade methanol, acetonitrile, and chloroform were purchased from Fisher Scientific (Fair Lawn, NJ, USA). Sodium dodecyl sulfate (SDS) solution

(10%) was obtained from Bio-Rad Laboratories (Hercules, CA, USA). RPMI-1640 medium with L-glutamine was purchased from Mediatech (Manassas, VA, USA). Fetal bovine serum was purchased from GE Healthcare Life Sciences (Logan, UT, US).

RiboShredder™ RNase Blend was purchased from Epicentre (Madison, WI, USA).

Proteinase K was obtained from Qiagen (Valencia, CA, USA).

2.2.2 Solutions

The stock standard solution of dC, dG, dT and dA (4.00 mM) were prepared in water individually, and their working solutions of 2.00 μM were prepared by diluting each individual stock standard solution with deionized water. The mixed calibrators of the above deoxyribonucleosides (1.00, 4.00, 20.0, 40.0, 200, 400 and 1.00 x 103 nM) and the mixed quality controls (QCs) (3.00, 30.0 and 900 nM) were prepared in deionized water by mixing and serial dilution of their working standard solutions. The stock internal

15 standard solution of 2'-deoxycytidine N3 (2.00 mM) was prepared in deionized water and

15 the working internal standard solution of 2’-deoxycytidine N3 (20.0 µM) was prepared by diluting the stock internal standard solution with deionized water. CT DNA was dissolved in deionized water to prepare 1.00 mg/mL solution.

The enzyme working solutions were prepared as follows: DNase I was dissolved with 0.9% NaCl solution to 10.0 mg/mL (ca. 20000 U/mL); NP1 was dissolved in 1.00 mM ZnCl2 solution to 1.00 mg/mL (ca. 200 U/mL); PDE I was dissolved in deionized water to 100 U/mL.

2.2.3 Instrumentation

2.2.3.1 Microwave digestion system

CEM Discover (model 908005) System (CEM, Mathews, NC, USA) was used for

DNA digestion. The system was an open vessel apparatus that had a single mode cavity with both power and temperature controls. The temperature was controlled through a fiber-

optic-temperature probe placed in the sample solution. The microwave system has a maximum microwave power of 300 W and a magnetron frequency of 2455 MHz.

2.2.3.2 LC-MS/MS system

The LC-MS/MS system consisted of a Shimadzu SIL-20AC autosampler

(Shimadzu, Columbia, MD, USA), and an AB Sciex QTrap 5500 tandem mass spectrometer (AB Sciex, Foster City, CA, USA). The system was controlled by AB Sciex

Analyst® (version 1.6.1) software for its operation, data acquisition and processing. The

LC system included a system controller (CBM-20A), two binary pumps (LC-20AD), a temperature-controlled autosampler (SIL 20AHT) and an online degasser (DGU20A3), and the mass spectrometer came with a Turbo IonSpray source.

The chromatographic separation was carried out isocratically under ambient temperature on a reversed phase Atlantis T3 C18 column (50 x 2.1 mm i.d.; 3 µm particle size) from Thermo Fisher Scientific (Waltham, MA, USA) using a mobile phase containing

0.1% formic acid aqueous solution and methanol (87.5:12.5, v/v) at a flow rate of 0.200 mL/min. Prior to initial sample analysis, the column was equilibrated with at least 20 column volumes of the mobile phase. For each run, 10.0 μL of sample was injected into the system by the autosampler set at 4.0 °C. The column eluate was introduced into the mass spectrometer through a turbo-ion spray probe that was operated under the positive ionization mode.

The mass spectrometer was tuned by infusion of each analyte at 500 ng/mL in 0.1

% formic acid in 50% methanol and 50% water to optimize both compound-dependent and

source-dependent parameters. The optimized instrument settings were as follows: curtain

(CUR), 40 psi; collision activated dissociation (CAD) gas, medium; nebulizer gas (GS1),

40 psi; turbo heater gas (GS2), 45 psi; turbo-ion-spray voltage (IS), +5200; source temperature (TEM), 300°C; declustering potential (DP), 35 V; entrance potential (EP), 7

V; collision energy (CE), 15 V; collision cell exit potential (CXP), 18 V; and mass resolutions (Q1 and Q3), were set as unit. Calibration curves were established by the multiple-reaction-monitoring (MRM) mode with the mass transitions of m/z 231 >115 for the IS, m/z 268 > 152 for dG, m/z 228 > 112 for dC, m/z 252 > 136 for dA, and m/z 243 >

127 for dT, respectively.

2.2.3.3 Cell culture and DNA extraction

HL-60 and U937 cell lines were obtained from American Type Culture Collection

(Rockville, MD, USA). HL-60 and U937 were cultured in RPMI-1640 medium supplemented with 10% (v/v) fetal bovine serum in a humidified 5% CO2 incubator at

37°C. After culturing, the cells were removed from the medium by centrifugation at 1,500

× g at 4°C for 5 min and washed with 5.00 mL of 1X PBS. The cell pellets were collected and stored at −20°C until DNA extraction.

Cells were lysed by adding 2.00 mL of TE buffer (containing 10.0 mM Tris and

1.00 mM EDTA at pH 8.0) and 240 μL of 10% SDS solution to each cell pellet with gentle mixing. The lysates were then incubated with 25.0 μL of proteinase K for 1 h at 37°C.

Thereafter, each sample was transferred to a phase lock gel tube (5 Prime, Gaithersburg,

MD, USA) and the DNA was extracted using standard phenol/chloroform extraction (8).

RNA interference was removed by adding 2.00 µL of RiboShredder™ RNase Blend (1

U/µL) to each 100 µL of DNA sample (1.00 mg/mL). After 30 min of incubation at 37°C, the sample was mixed with 1.00 mL of pre-chilled ethanol (−20°C) and precipitated at

−20°C overnight. After centrifugation (15,000 × g for 15 min), each sample was washed with 1.00 mL of pre-cooled 70% ethanol (−20°C) twice. Each DNA sample was then air- dried and reconstituted in 100 µL of 5.00 mM BIS-TRIS buffer (pH 7.0). To measure the concentration and purity of the DNA, optical density was measured at 260 and 280 nm by

UV spectrophotometry.

2.2.4 Tetra-enzyme digestion mix

The tetra-enzyme cocktail used for the one-step, multi-enzyme digestion was prepared by mixing the enzyme solutions as follows: 10.0 μL of DNase I (20,000 U/mL),

15.0 μL of NP1 (200 U/mL), 40.0 μL of PDE I (100 U/mL), and 0.50 μL of ALP (40,000

U/mL) (9). DNase I is an endonuclease that splits phosphodiester bonds, typically yielding oligonucleotides with a free 3’-end hydroxyl group and a free 5’-end phosphate group (10).

PDE I is a 3’ to 5’ exonuclease. It can successively hydrolyze DNA or RNA from the 5’ end to the 3’ end to the single-nucleotide level with 5’-phosphates as the typical digestion product (11). NP1 is a 5’ to 3’ exonuclease and can completely hydrolyze single-strand

DNA to its single nucleotides with a phosphate group on the 5’ position (12). ALP is a hydrolase that hydrolyzes the phosphate group from various molecules, including nucleotides and proteins (13).

2.2.5 DNA digestion

2.2.5.1 Microwave-assisted digestion

DNA samples were prepared by diluting 1.00 µL of CT DNA stock solution (1.00 mg/mL) in 50.0 μL of 5.00 mM BIS-TRIS buffer (pH 7.0). The experimental design for the microwave digestion is summarized in Table 2.1. Three different sample series (I, II, and III) were digested for various durations at different microwave powers and temperatures. Sample series I was incubated in a boiling water bath for 15 min and chilled in an ice bath for 5 min to denature ds-DNA to ss-DNA. It was then digested by adding

2.00 μL of the digestion buffer but without adding the tetra-enzyme mix. Sample series II was digested by adding 2.00 μL of the tetra-enzyme mix with the enzyme mix but without prior incubation in a boiling water bath. Sample series III was incubated in a boiling water bath for 15 min and chilled in an ice bath for 5 min to denature ds-DNA to ss-DNA. Then,

2.00 μL of the tetra-enzyme mix was added.

Each DNA sample in sample series I, II, and III was digested following the eight sets of parameters outlined in Table 3.1. Each set of samples was digested in triplicate for all experimental conditions. The resultant dNs from the digest were extracted as will be described later.

Table 2.1. Various experimental conditions and microwave settings tested for DNA digestion.

Set Microwave parameters Sample series

1 10 W, 37 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

2 25 W, 37 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

3 50 W, 37 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

4 100 W, 37 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

5 10 W, 55 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

6 25 W, 55 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

7 50 W, 55 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

8 100 W, 55 ℃ (5, 10, 15, 20, 30, 45, and 60 min) I, II, III

Note: I: Negative controls (no tetra-enzyme mix was added to DNA samples). II: Tetra-enzyme mix was added to DNA samples. III: DNA samples were heated in boiling water for 15 min and chilled in ice/water bath for 5 min. Tetra-enzyme mix was then added to DNA samples.

2.2.5.2 Conventional digestion

A 1.00 µL aliquot of the CT DNA stock solution (1.00 mg/mL) was diluted in 50.0

μL of 5.00 mM BIS-TRIS buffer (pH 7.0). For samples in series I and III, ds-DNA was denatured to ss-DNA by incubating in a boiling water bath for 15 min and chilling in an ice bath for 5 min, which is a routine for conventional digestion. To sample series I, 2.00

μL of the digestion buffer was added, and to sample series II and III, the tetra-enzyme mix was added. All three series were digested conventionally in triplicate by incubating at 37°C in a water bath for 17 h (4). The resultant dNs from the digest were extracted as described below.

2.2.6 Extraction of dNs from enzyme digests

After digestion, the vials were removed from the microwave or water bath and the samples were prepared for MS analysis. First, 5.00 μL of the internal standard (IS) 2'-

15 deoxycytidine N3 (2.00 µM in the LC mobile phase) was added to a digested sample with vortex-mixing for 30 s. The sample was then deproteinized by adding 450 μL of HPLC- grade acetonitrile with vortex mixing for 30 s and then centrifuged at 15,000 × g for 10 min. Then, 450 μL of the supernatant was pipetted into a 1.50 mL microcentrifuge tube and evaporated to dryness at 30°C for 60 min in a TurboVap® LV Evaporator (Zymark,

Hopkinton, MA, USA) under a pressurized stream of nitrogen gas. Finally, each dried sample was reconstituted in 1000 μL deionized water for HPLC-MS/MS analysis.

2.2.7 Application of microwave-assisted digestion to other DNA samples

DNA extracted from HL-60 and U937 was also used to test the digestion efficiency of the microwave method using the optimized power, time, and temperature conditions.

Samples containing 1.00 µg of DNA diluted in 50.0 μL of 5.00 mM BIS-TRIS buffer (pH

7.0) were digested by both conventional and microwave methods. The digested products were extracted and analyzed using LC-MS/MS.

2.3. Results and Discussion

2.3.1 Microwave Technology

Microwave technology has proven to be a better alternative to conventional heating owing to its high heating rate, accelerated digestion (by molecular agitation and enzyme diffusion), and especially for its reaction selectivity, which can be achieved by tuning the microwave parameters based on the nature of the molecules. Microwave heating is based on the ability of a material to absorb high-frequency electromagnetic energy (microwaves) and convert it to heat, thus it heats the sample through direct activation (14, 15). The electric field interacts with any molecule that has a dipole or that is ionic, and as the molecules move, they generate heat, leading to the rapid increase in temperature commonly associated with microwave irradiation. Therefore, irradiation can allow for shorter reaction times than those available with conventional enzymatic hydrolysis alone. Every solvent and reagent absorbs microwave energy differently as they have different degrees of polarity. Thus, they will be affected differently by changing the microwave field, allowing optimization of parameters based on the type of reactants (16, 17).

2.3.2 Optimization of parameters for microwave digestion

Microwave digestion parameters were optimized in terms of microwave power, irradiation time, and reaction temperature. The amount of power applied to the reaction is very important as, with an increase in the power, more energy is produced and transferred to the sample, where it induces molecular agitation and sample diffusion. The power applied keeps the reaction mixture at its maximum attainable temperature. Hydrolysis time is also an important factor in the choice of digestion method. The effects of different microwave incubation times also depend on the microwave power and temperature applied.

The hydrolysis temperature should be selected based on the stability of the enzyme because each enzyme has a specific spatial structure and will be irreversibly denatured if the reaction temperature is too high, leading to loss or decrease of activity. However, if the temperature is too low, the enzyme activity is reduced due to the low probability of collision.

2.3.3 Tetra-enzyme digestion mix

An enzyme cocktail using the four enzymes was developed to achieve digestion in one-step instead of through a multi-step sequential digestion process. During the development of this tetra-enzyme mix, the significance of each enzyme was evaluated by digesting the DNA with various combinations of the enzymes. The digestion time was optimized by digesting the DNA with the enzyme cocktail for 5, 10, 15, 20, or 25 h. From kinetic curves (9), it was found that the optimized digestion time is between 15 and 20 h when following conventional digestion methods. Consequently, a total digestion time of

17 h was adopted due to its compatibility with an 8 h working schedule. In all the

experiments, the digestion efficiencies of the tetra-enzyme system on DNA adducts through sequential and cocktail digestions were compared (9).

2.3.4 Microwave-assisted hydrolysis

DNA was digested with one-step tetra-enzyme cocktail in a microwave using different experimental parameters, as given in Table 2.1. During the enzymatic release of

DNA adducts, all the normal DNA units are released as dNs (Figure 2.1). The released dNs can serve as indicators to reflect the digestion efficiency of the enzymes on a specific

DNA sample. The hydrolysis products of DNA (i.e., the dNs) were separated using the chromatographic conditions described in the experimental section. To compare the digestion efficiencies of the protocols described above, the peak area for each type of dN resulting from microwave digestion (i.e., dA, dC, dG, or dT) was compared with that of the corresponding dN from conventional digestion (Figure 2.2).

3.00E+06 dC dT

2.00E+06

Intensitty, cps Intensitty, 1.00E+06 dG

0.00E+00 0 1 2 3 4 5 6 7 8 9

Time, min

Figure 2.1 LC-MS/MS of dNs released after enzyme digestion.

Figure 2.2. Representative MRM chromatograms of enzymatic digested products of CT

DNA. A) dNs released from conventional digestion at 37 oC temperature for 17 h; B) dNs released from microwave digestion at 55 oC temperature and 50 W power for 30 min.

The peak area of each dN released from CT DNA by conventional digestion was defined as 100% (Figure 2.3 (A, I)). In the first four experiments (Table 2.1, sets 1–4), the temperature was fixed at that used in conventional digestion. Keeping the temperature constant (37°C), the digestion efficiency was assessed for various durations at different microwave powers. As shown in Figure 2.3 (B, I–IV) for sample series III, the digestion efficiency increases with increasing microwave power up to 50 W and reaches a plateau of only 82.0% at 30 min. Sample series II also shows a similar pattern with a hydrolysis yield of 79.6%. Hydrolysis yields of only 31.5% and 47.0% are observed at 10 W and 25 W, respectively, when incubated for 60 min. The non-enzymatic hydrolysis in sample series I has a very low yield of ≤10.0%, and the hydrolysis increases with time but is independent of microwave power. At 100 W, very low hydrolysis yields (≤10.0%), which are comparable to those of non-enzymatic digestion, are observed for all three-sample series.

More experiments were conducted to assess whether microwave digestion at higher temperatures gives better hydrolysis yields. All the enzymes used in the tetra-enzyme digestion mix are stable at temperatures below 70°C. The enzyme nuclease P1 has an optimal reaction temperature of 70°C, but temperatures below 60°C are recommended for longer incubation periods. DNase I activity is affected only if it is heated beyond 68°C for

5 h. The optimum reaction temperature for PDE I is between 37 and 60°C, and decreased activity is observed at temperatures above 70°C. Although ALP has an optimum reaction temperature of 50°C, the enzyme will denature only if subjected to temperatures above approximately 72°C.

Figure 2.3 Comparison of dC hydrolysis yields under different experimental conditions.

(A) DNA digested at 37°C or 55°C in a water bath; (B) DNA digested at 37°C using microwave irradiation at various powers (10, 25, 50, and 100 W); (C) DNA digested at

55°C using microwave irradiation at various powers (10, 25, 50, and 100 W). Each column represents the mean ± SD (n = 6). The red bars indicate sample series I, the gray bar indicates sample series II, and the yellow bar indicates sample series III.

Considering the reaction temperatures and stabilities of the enzymes used in the digestion mix, another four sets of experiments were conducted maintaining the temperature at 55°C, and digestion efficiency was assessed at different microwave powers for different incubation times (Table 2.1, sets 5–8). For sample series II and III, hydrolysis yields of 96.5% are obtained when digested at a power of 50 W for 30 min, where the yield plateaus. Yields of only 55.0% and 69.5% are observed for an irradiation of 10 W and 25

W, respectively, when digested for 60 min (Figure 2.3 (C, I–IV)). Samples digested at 100

W present hydrolysis yields of only ≤10.0%. Furthermore, the yield for non-enzymatic hydrolysis of series I does not increase with power, remaining below 10.0%. When we observed the enzyme kinetics for samples II and III at a temperature of 55°C and a power of 55 W for 5 to 60 min, we observed that the hydrolysis yield does not change after 30 min, giving an optimum digestion time of 30–60 min.

Sample series II (in which ds-DNA was not denatured to ss-DNA prior to digestion) shows a similar digestion yield as sample series III (in which ds-DNA was denatured to ss-

DNA prior to digestion) under all experimental conditions in microwave-assisted digestion. However, under conventional digestion conditions, sample series II shows only

47.4% hydrolysis compared to 96.5% for microwave-assisted digestion (Figure 2.3 (A, I)).

In conventional digestion methods, DNA is first incubated in boiling water for 15 min and chilled on ice before adding the enzyme digestion mix to break the double-stranded DNA into a single strand. This is because NP1 can only hydrolyze single-strand DNA completely to single nucleotides, and thus double-stranded DNA is not a suitable substrate for NP1. In microwave-assisted hydrolysis, microwave irradiation promotes the reaction by thermal energy, inducing ionic diffusion, and by enhancing dipole rotation, which helps in breaking

the double-stranded DNA into single strands. Moreover, DNase I is thought to increase the digestion speed of NP1 and PDE I by generating free ends on the DNA. The oligonucleotide generated by DNase I weakens the hydrogen bonds between the double strands and releases single-strand DNA, which is a better substrate for NP1. These two factors might contribute to the high hydrolysis yield of series II in microwave digestion.

At higher power (100 W), we observed very low hydrolysis yields. The localized energy uptake under high microwave power conditions may denature the enzymes by molecular agitation, resulting in low hydrolysis yields. The non-restriction multi-enzyme system used here works via a sequential mechanism and the enzymes are all stable up to

70°C. Thus, for this enzyme system, using a lower power setting with increased incubation time will help the individual enzymes work with optimal activity without enzyme degradation. From all the experiments in the negative control (series I), low hydrolysis yields (≤10.0%) are observed, indicating that microwave irradiation can only catalyze the hydrolysis of DNA in the presence of enzymes with temperature, microwave power, and digestion times optimized. In conventional digestion, series I presents much lower hydrolysis yields (≤5.00%) compared to those of microwave digestion (Figure 2.3 (A, I)).

Finally, to evaluate whether the hydrolysis is due to the temperature effect alone or the effects of both temperature and microwave irradiation combined, three more experiments were performed by digesting the DNA using the same temperature-time profiles (37°C for 30 min and 55°C for 30 min and 17 h) in a water bath. The results presented in Figure 2.3 (A, II–IV) indicate that digestion efficiency is affected by both temperature and microwave irradiation. When DNA is digested at 55°C for 30 min without any microwave irradiation, the hydrolysis yield is ≤30.5%, though both series II and III

present similar yields. Furthermore, a significant decrease in enzyme activity is observed due to longer incubation time (17 h) at 55°C, resulting in low digestion yields (≤69.2%).

Based on these findings, the optimum conditions for microwave-assisted digestion of DNA (tetra-enzyme mix digestion at 55°C and 50 W for 30 min) were identified. The similarity in hydrolysis yields between conventional and microwave-assisted digestion methods proves that microwave digestion gives similar quantitative results as those of the traditional methods. Moreover, we used an isotope-labeled IS during all the sample preparations, and no significant degradation of the IS is observed when comparing the IS peak areas before and after digestion.

2.5 Application of the method to the digestion of DNA extracted from cell lines

To assess the utility of this method for high-speed and efficient DNA extraction,

DNA extracted from HL-60 and U937 cell lines were subjected to tetra-enzyme digestion, either following the conventional overnight approach or by microwave-assisted digestion under the optimized conditions (i.e., 50 W and 55°C for 30 min). The dNs released from the digested DNA were analyzed using LC-MS/MS and the intensities of the peaks were compared to those obtained with the conventional method. As shown in Table 2.2, the digestion yields for each dN are comparable to those of conventional digestion.

Table 2.2 Microwave-assisted enzymatic hydrolysis yields of dNs released from Calf thymus DNA, DNA from HL-60 and U-937 cell lines

Hydrolysis Yield (%) (n = 6) Cell Line dA dC dT dG

HL-60 91 ± 5 95 ± 5 94 ± 4 96 ± 7

U-937 90 ± 4 93 ± 3 90 ± 3 93± 2

CT DNA 94 ± 3 96 ± 1 94 ± 3 92 ± 2

2.4 Conclusion

Using non-restriction enzymes at a microwave power of 50 W and a temperature of 55°C, the highest yield for the hydrolysis of DNA was obtained in 30 min with a performance similar to that obtained by the conventional overnight digestion procedure, thereby increasing sample throughput. Thus, microwave-assisted enzymatic digestion of

DNA was demonstrated to be at least 34-times faster than that obtained using the traditional digestion method. This method is simple, reproducible, automated, and does not require additional expertise or costly equipment. Furthermore, although this method involves microwave irradiation, it does not cause degradation of the enzymes used. However, for different DNA adducts, the digestion efficiency of each enzyme may vary. Thus, to achieve the best digestion effect for a specific DNA adduct, the enzyme system and microwave- assisted digestion conditions need to be optimized individually. In conclusion, we have developed a novel and promising strategy for the hydrolysis of DNA using microwave irradiation and non-restriction enzymes that appears to be suitable for qualitative and quantitative DNA analysis.

2.5 References

1. Wang M, Yu N, Chen L, Villalta PW, Hochalter JB, Hecht SS. Identification of an

acetaldehyde adduct in human liver DNA and quantitation as N2-

ethyldeoxyguanosine. Chem. Res. Toxicol., 2006; 319-324.

2. Mingyao Wang,Guang Cheng,Shana J. Sturla,Yongli Shi,Edward J. McIntee,Peter W.

Villalta,Pramod Upadhyaya, Stephen S. Hecht. Identification of adducts formed by

pyridyloxobutylation of deoxyguanosine and DNA by 4-(acetoxymethylnitrosamino)-

1-(3-pyridyl)-1-butanone, a chemically activated form of tobacco specific carcinogens.

Chem. Res. Toxicol., 2003; 616-626.

3. Singh R, Teichert F, Verschoyle RD, Kaur B, Vives M, Sharma RA, Steward

WP, Gescher AJ, Farmer PB. Farmer. Simultaneous determination of 8-oxo-2'-

deoxyguanosine and 8-oxo-2'-deoxyadenosine in DNA using online column-switching

liquid chromatography/tandem mass spectrometry. Rapid Commun. Mass Spectrom.,

2009; 151-160.

4. Crain, PF. Preparation and enzymatic hydrolysis of DNA and RNA for mass

spectrometry. Meth. Enzymol., 1990; 782-790.

5. Pramanik, B. N., Mirza, U. A., Ing, Y. H., Liu, Y.-H., Bartner, P. L., Weber, P. C., &

Bose, A. K Microwave-enhanced enzyme reaction for protein mapping by mass

spectrometry: A new approach to protein digestion in minutes. Protein Sci., 2002; 2676-

2687.

6. Juan H, Chang S, Huang H, Chen S. A new application of microwave technology to

proteomics. Proteomics., 2005; 840-842.

7. Marrubini G, Fattorini P, Previderé C et alMarrubini G, Fattorini P, Previderé C, Goi

S, Sorçaburu Cigliero S, Grignani P, Serra M, Biesuz R, Massolini G. Experimental

design applied to the optimization of microwave-assisted DNA hydrolysis. J.

Chromatogr. A., 2012; 8-16.

8. Sambrook J, Russell DW. Purification of nucleic acids by extraction with phenol:

chloroform. Cold Spring Harb Protoc., 4455; 2006.

9. Li, L. Development of quantitative LC-MS/MS methods for the pharmacological

studies of anti-cancer drugs. Cleveland State University, Cleveland, Ohio, USA.,2011;

53-67.

10. Matsuda M, Ogoshi H. Specificity of DNase I: Estimation of nucleosides present at the

5'-phosphate terminus of a limit digest of DNA by DNase I. J. Biochem., 1966; 230-

235.

11. Falcone JM, Box HC. Selective hydrolysis of damaged DNA by nuclease

p1. Biochimica et Biophysica Acta - Protein Structure and Molecular

Enzymology.,1997; 267-275.

12. Laskowski M. Venom Exonuclease. In: The Enzymes, 3rd Ed. Boyer P (Ed). Academic

Press, New York, USA, Vol. 4.,1971.

13. Weissig H, Schildge A, Hoylaerts MF, Iqbal M, Millán JL. Cloning and expression of

the bovine intestinal alkaline phosphatase gene: biochemical characterization of the

recombinant enzyme. Biochem. J., 1993; 503-508.

14. Lin S, Wu C, Sun M, Sun C, Ho Y. Articles: Microwave-assisted enzyme-catalyzed

reactions in various solvent systems. J. Am. Soc. Mass Spectrom., 2005; 581-588.

15. Adam D. Microwave chemistry: Out of the kitchen. Nature., 2003; 571-572.

16. Chen S. The studies of microwave effects on the chemical reactions. J. Chin. Chem.

Soc., 1997; 169-182.

17. Langa F, de la Cruz, P, de la Hoz A, Diaz-Ortiz A, Diez-Barra, E. Microwave

irradiation: more than just a method for accelerating reactions. Contemp. Org. Synth.,.

1997; 373-386.

Chapter III

An Accelerated Protein Sample Preparation Method and

Selective Extraction of peptides from Plasma for LC-MS-

Based Proteomics

Abstract

Although high-throughput and automated LC-MS methods are available for protein and peptide characterization, sample preparation oftentimes is the rate-limiting step for many proteomic workflows. Sample preparation for proteomics by LC-MS usually entails two sequential steps: (a) denaturation, reduction and alkylation of proteins; and (b) digestion of proteins to peptides by protease such as trypsin, where the resultant peptides fall in the preferred mass range with basic residue at the carboxyl terminus suitable for LC-

MS analysis and produce information-rich mass spectra for protein sequencing, identification, and quantitation. For instance, reduction and alkylation of proteins usually take ca. 2h and digestion of proteins takes >16 h.

To reduce the time consumed in protein sample preparation, new technologies

(e.g., infrared-, ultrasound- and microwave-assisted enzyme digestion) have emerged.

However, these new approaches focus only on the enzyme digestion step and still need about 2 h for reduction and alkylation of proteins. Peptides have a significant role in the inflammatory response and cancer biology making them important biomarker candidates.

However, extraction of peptides efficacy for profiling, stability and bioavailability studies remains a key technological challenge due to the complexity of the plasma matrix and low concentration of peptides compared to the abundant proteome in plasma.

In the first part of the study, we demonstrated an accelerated method for protein sample preparation by combined use of ultrasound and microwave technologies. The entire sample preparation took less than 30 min to complete using bovine serum albumin as a model protein and trypsin as a protease. In this work, the ultrasonic water bath was used for protein reduction and alkylation steps (5 min each) and followed by microwave- assisted enzyme digestion (15 min). And, in the second part of the study, we evaluated commonly employed peptide extraction methodologies to evaluate the recovery and stability of a spiked synthetic peptide. The resultant peptides from both methods were separated and analyzed by Agilent 6540 Accurate-Mass Q-TOF LC/MS system. Protein sequencing and identification were accomplished using MASCOT database and synthetic peptide sequence was confirmed by Agilent Bioconfirm software.

3.1 Introduction

Part 1: In the proteomic analysis of complex protein mixture by mass spectrometry, bottom-up proteomics is the most widely used method. In this bottom-up proteomics approach, the total intact proteins are proteolytically cleaved by enzymatic digestion and

the resultant peptides were separated on a LC column and analyzed using mass spectrometry (1,2). Enzymatic digestion of protein essentially includes denaturation, reduction, and alkylation prior to trypsin digestion. Reduction is achieved by incubating the protein with DTT, where the disulfide bonds between the cysteine within are protein are broken down followed by alkylation where the reactive sulfhydryl groups are acetylated with iodoacetamide to prevent the reformation of disulfide bonds (Figure 1.1) (3,4). Thus, reduction and alkylation of protein are essential to prevent aggregation that negatively affects proteolytic digestion. Also, completeness of the reduction and alkylation is important to avoid any variations in the protein analysis.

Although high-throughput and automated LC-MS methods are available for protein characterization, sample preparation oftentimes is the rate-limiting step for many proteomic workflows. The conventional trypsin digestion workflow includes reduction of protein using DTT for 1 h and alkylation with IAA for 45 min followed by trypsin digestion at 37 oC for 17 h (5). As the digestion step is the most time-consuming step, various accelerated digestion techniques like heating, infrared radiation, microwave assisted digestion, microspin columns, ultrasonic assisted digestion, digestion at high pressure, etc. were developed that reduced the digestion time from hours to minutes (6-9).

Also, few methods were developed to accelerate the sample preparation step. In methods like microwave assisted enzymatic digestion, protein samples were only reduced in boiled water for 5 min and not alkylated (10), but the complete reduction of disulfide bonds and alkylation of –SH groups on cysteine in proteins is necessary to expose LYS and ARG to trypsin for total digestion, especially in proteins that have more cysteines.

This accelerated sample preparation method can be applied only to simplified samples of

proteins only (11). Also for the digestion of samples like glycoproteins using microwave, conventional sample preparation methods were only used (12). Moreover, in samples like secreted proteins, the disulfide bonds inhibit proteolytic digestion by maintaining the tertiary structures and proteins from biological fluids like plasma and serum are bound with protease inhibitors which make the denaturation, reduction and alkylation steps compulsory for complete digestion (13). And methods like ultrasonic assisted enzymatic digestion used sonoreactor for reduction, alkylation and for trypsin digestion of proteins, but requires specialized equipment (14). To our knowledge none of these accelerated sample preparations steps were tested with the overnight conventional digestion method.

Part 2: Biomolecules like proteins and peptides have gained increased interest as therapeutics during recent years. More than 7000 naturally occurring peptides have been identified, and found to play key roles in human physiology as growth factors, signal transmitters etc (15,16). Thus, peptide therapeutics are often considered as specialized biopharmaceuticals since they are analogs to, or are endogenous compounds, and can be well tolerated by the body. More than 60 peptide-based drugs are currently on the market and more than 400 in preclinical and clinical trials. (17). Analysis of stability and pharmacokinetic studies are of great value for in the development of peptide therapeutics.

LC-MS technology is becoming widely used for sensitive and quantitative detection of proteins and peptides. However, there are major challenges in the analysis of plasma peptides. First, large amounts of plasma proteins like albumin and salts will greatly deteriorate the performance of LC systems or interfere with the signal of peptides in direct

MS assays (18). Second, the concentration of therapeutic peptides and endogenous peptide biomarkers in plasma are present in very low concentrations (19). Third, some peptides are

inclined to bind to proteins like albumin (20). Therefore, for peptides enrichment and depletion of abundant proteins and salts, peptide extraction is an essential step before LC separation and MS identification.

Herein, in the first part of the study, we have developed a hyphenated technique using an ultrasonic water bath for reduction and alkylation steps and microwave technology for the trypsin digestion. Also, we hyphenated the accelerated sample preparation step using an ultrasonic water bath with conventional overnight trypsin digestion methods. In the present work, we demonstrate that protein denaturation, reduction, and alkylation can be achieved within 10 min using an ultrasonic water bath that is commonly available in most of the laboratories and compared the completeness of digestion of proteins and glycoproteins in both microwave assisted and overnight digestion methods. The resultant peptides were separated and analyzed using Agilent QTOF 6540 mass spectrometer followed by MASCOT database search to identify the proteins and compare the sequence coverages from different sample preparation and digestion methods. For this study, we have used bovine serum albumin (BSA) and a glycoprotein- human IgG as model proteins and also applied the methods to a complex mixture of proteins derived from human serum.

In the second part of this study, five widely used extraction methods (ACN precipitation, chloroform - methanol extraction, acetone precipitation, acidified acetone precipitation, precipitation using trifluoracetic acid (TFA)) were assessed to extract a spiked synthetic peptide (18Mer Stapled peptide : SEITKQKED(Staple-)SHRHYC(-Staple)) to study the stability of the peptide in the human plasma. With this peptide as target analyte, these methods were evaluated on the basis of recoveries and reproducibility with human plasma as the study matrix.

Figure 3.1 Reduction of proteins using DTT, Alkylation of proteins using IAA

(Adopted from reference 14)

3.2 Experimental

3.2.1 Chemicals and materials

Bovine serum albumin (BSA), Human IgG, iodoacetamide (IAA), ammonium bicarbonate, acetic acid, trichloroacetic acid (TCA) and formic acid were ordered from

Sigma Aldrich (St. Louis, MO, USA). Dithiothreitol (DTT), sequencing grade modified trypsin was purchased from Promega (Madison, WI, USA). Trifluoroacetic acid (TFA) and

C18 ZipTips were obtained from EMD Millipore (Massachusetts, MA, USA). HPLC grade

Acetonitrile and water were purchased from Fischer Scientific (Hanover park, IL, USA).

Hydrochloric acid (HCl) was ordered from VWR (West Chester, PA, USA). Different lots of human serums were purchased from Innovative Research (Novi, MI, USA).

PART 1

3.2.2 Optimization of protein reduction and alkylation time

1.00 mg of BSA and Human IgG were weighed separately and dissolved in 1.00 mL of 100 mM ammonium bicarbonate solution to make a 1.00 mg/mL solution of each.

Pooled human serum proteins were obtained by acetone precipitation. Briefly, to 50.0 µL of human serum 1.00 mL of cold acetone was added and kept at -80 oC overnight. Then centrifuged at 4 oC at 13000 rpm for 30 min. The supernatant is discarded, and the precipitated proteins were air dried and dissolved in water. Protein concentration was measured using Bradford reagent.

During the experiment, two sets of positive and negative controls were used. In one set, samples were reduced, alkylated and digested with trypsin using the conventional methods and in the second set samples were reduced, alkylated using conventional methods and digested with trypsin in the microwave. For all the experiments BSA was used as a model protein. In the conventional methods of digestion: BSA digested with trypsin for

17 h at 37oC in a water bath without any pretreatment was used as a negative control. In the positive control, reduction was achieved through the addition of DTT to a final concentration of 10mM and incubated at room temperature for 1 h and alkylation by adding

IAA to a final concentration of 55mM and incubated for 45 min in the dark followed by trypsin digestion at 37oC for 17 h in a water bath. In the microwave digestion methods:

BSA digested with trypsin for 15 min at 50 W power and 55oC in the microwave without any pretreatment was used as a negative control. In the positive control, reduction was achieved through the addition of DTT to a final concentration of 10mM and incubated at room temperature for 1 h and alkylation by adding IAA to a final concentration of 55mM and incubated for 45 min in dark followed by digestion with trypsin for 15 min at 50 W power and 55oC in the microwave.

To optimize the reduction time and alkylation time using the ultrasonic water bath,

BSA was used as a model protein and various conditions were tested as given in Table 3.1.

10.0 µg of protein was digested in triplicate using the digestion protocols listed in Table

5.1. In the protocols 1-5, samples were reduced by DTT at a final concentration of 10.0 mM and incubated at room temperature for 0, 2 ,5 ,10 ,15 min and alkylated by adding IAA to a final concentration of 55.0mM and incubated for 0, 2 ,5 ,10 ,15 min respectively in an ultrasonic water bath at full power. Then samples were digested with trypsin at a ratio of

1:50 using CEM microwave system for 15 min at 55oC and 50 W power. The enzyme digestion was quenched by adding 10.0 µL of 5% formic acid, desalted using C18 ZipTips and eluted in 0.1% FA for further analysis using LC-MS/MS. The efficiency was determined by comparing the results with conventional methods. After the parameters had been optimized, the most effective sample preparation method was further tested on Human

IgG, a mixture of BSA and Human IgG and proteins from pooled human serum.

Table 3.1 Sample preparation protocols tested for microwave digestion methods

Protein Preparation Time (min) Protocol # Microwave Reduction a Alkylation a digestion*

1 0 0 15

2 2 2 15

3 5 5 15

4 10 10 15

5 15 15 15

a Protein samples reduced and alkylated in an ultrasonic water bath at 37 oC at 143 W ultrasonic power

* All samples were digested using CEM discovery microwave at 50 W power and 55 oC temperature.

3.2.3 Instrumentation

3.2.3.1 Chromatography separation and mass spectrometric conditions

Protein digests were analyzed using an Agilent infinity 1290 UPLC system interfaced to an Agilent QTOF 6540 mass spectrometer (Santa Clara, CA, USA). The

Agilent infinity 1290 UPLC is composed of a solvent reservoir, a degasser, a binary pump

(G4220A), a thermostat (G1330B), an autosampler (G4226A), and a thermostated column compartment (G1316C). The separation of peptides was carried out on an Agilent Zobrax

300SB - C18 capillary column (1 X 150mm; 3 u particle size) maintained at 50 0C using mobile phases A (0.1% FA in water) and B (0.1% FA in Acetonitrile). The elution gradient was: 5% B at 0 min, 0-25% B from 0-19 min, 25-45% B from 19-34 min, 45-60% B from

34-57 min, 60-5% B from 57-60 min, with a total run time of 60 min including equilibration. at a flow rate of 0.050 ml/min. A 3.00 µL (ca. 50 ng protein equivalent) amount of sample was injected into the system by the autosampler set at 4 °C for analysis.

Prior to sample analysis, the column was first equilibrated with at least 25 column volumes of the mobile phase at a flow rate of 0.050 mL/min.

The LC elute was introduced into the Quadrupole Time of flight (Q-TOF 6540) mass spectrometer equipped with an orthogonal ESI interface (Agilent Jet Stream, AJS operated under positive mode and the data was acquired in a data-dependent manner. The instrument was controlled by a PC running Mass Hunter Workstation (version B.05.01) software from Agilent Technologies (Santa Clara, CA, USA). MS operating parameters are as follows: capillary voltage, 2250V; nebulizer pressure, 40 psi; drying gas flow rate,

10 L/min; gas temperature, 350 oC; sheath gas flow, 11 L/min; sheath gas temperature, 325

0C; skimmer voltage, 65 V; fragmentor voltage, 150V. Both centroid and profile mode

were used in data collection with an extended dynamic range of 2 GHz. TOF MS accurate mass spectra were recorded at isolation width of 1.3 m/z across the range 300-1800 m/z at

5 spectra/s and MS/MS spectra were recorded across a range of 50 to 1800 m/z at 3 spectra/s. For MS/MS analysis, peptides were subjected to fragmentation using collision energy with a slope of 3 and offset +2, and the preferred charge states were set at 2, 3, >3 and unknown. TOF was calibrated on daily basis and real-time calibration was performed by continuous infusion of reference ions (API-TOF reference mass solution kit containing

100.0 mM ammonium trifluoroacetate in acetonitrile/water (90:10); 5.0 mM purine in acetonitrile/water (90:10); 2.5 mM hexakis (1H, 1H, 3H-tetrafluoropropoxy)phosphazine in acetonitrile/water (90:10)) [Catalog# G1969-85001] obtained from Agilent

Technologies (Santa Clara, CA, USA).

3.2.3.2 Microwave system

Microwave system used for trypsin digestion was purchased from CEM Discovery

(Mathews, NC, USA). The apparatus is an open vessel system that contains a single mode cavity with power and temperature control. The temperature is controlled through a fiber optic temperature probe.

3.2.4 Data processing

The LC-MS/MS data acquired using Agilent Mass hunter workstation (.d files) were processed in Mass Hunter Qualitative software (Agilent Technologies, Santa Clara,

CA, USA) for database analysis. The datasets were subjected to compound using find compounds by auto MS/MS, selecting the positive MS/MS TIC threshold to 100000.

Reference ion mass 922.0097 was excluded. Both MS/MS and extracted ion

chromatograms were selected for extraction. Now all the compounds identified by find compounds by auto MS/MS were selected and exported as Mascot generic format (MGF).

While exporting all spectra and extracted spectra were selected, and absolute height of ≥

500 counts was set as peak filter. Isotope grouping was performed based on the peptide model with a peak spacing tolerance of 0.0025 ± 7 ppm. The peptide MFG was searched with Mascot Search Engine – MS/MS Ion Search against the Swiss-Prot Database for protein identification. Cysteine carbamidomethylation was selected as fixed modification.

The mass value was monoisotopic. Peptide mass tolerance was set to 50 ppm, and the

MS/MS tolerance was at 0.6 Da. No restriction was placed on molecular weight range and taxonomy. Two miss-cleavage were tolerated. ESI-QUAD-TOF was selected as the instrument. A protein was regarded identified when it had a significant Mascot probability score (scores greater than 60 correspond to P < 0.05), and at least two matched peptides.

PART 2

3.2.5 Peptide extraction

3.2.5.1 Preparation of spiked plasma samples

Stock Solution of Stapledpeptide was prepared in DMSO at a concentration of 500

µM and stored at -20 oC. Stock solution was added to 100 µL human plasma at a target concentration of 10.0 µM. A total of 3 sets were prepared 1) Human plasma without

Stapled peptide, incubated for 0, 1 and 10 h. 2) Human plasma with 10.0 µM Stapled peptide, incubated for 0 and 1 h. 3) Human plasma with 10.0 µM Stapled peptide,

Incubated for 10 h. Similarly, Stapled peptide was added to 100 µL ddH2O in three sets

and incubated one for 0, 1 and 10 h as controls. Also, a working solution of 10.0 µM

Stapled peptide solution in ddH2O was prepared freshly as a control. Each set of samples were prepared in triplicate.

3.2.5.2 Peptide extraction protocols

The following five peptide-extraction methods were evaluated to perform the stability study of the synthetic Stapled peptide in human plasma.

i. Acetonitrile extraction method

A volume of 400 μL of acetonitrile was mixed with a volume of 100 μL of plasma.

Samples were vortexed briefly and were centrifuged at 15,000 X g for 15 min at 4 oC.

The supernatant was transferred to new 1.5 mL Eppendorf tube and dried under a

stream of nitrogen gas. Dried samples were reconstituted in 100 µL of 0.1% formic

acid in water solution.

ii. Chloroform - Methanol extraction method

900 μL of chloroform - methanol (1:4 v/v) was mixed with a volume of 100 μL of

plasma. Samples were vortexed and incubated for 1 hr at -20 oC. Then samples were

centrifuged at 15,000 X g for 15 min at 4 oC. The supernatant was transferred to new

1.5 mL Eppendorf tube and dried under a stream of nitrogen gas. Dried samples were

reconstituted in 100 µL of 0.1% formic acid in water solution. iii. Cold acetone extraction method

Briefly, 900 μL of cold acetone (-80 oC) was mixed with a volume of 100 μL of

plasma. Samples were vortexed and incubated for 1 hr at -80 oC. Then samples were

centrifuged at 15,000 X g for 15 min at 4 oC. The supernatant was transferred to new

1.5 mL Eppendorf tube and dried under a stream of nitrogen gas. Dried samples were

reconstituted in 100 µL of 0.1% formic acid in water solution. iv. Acidified – acetone methanol extraction method

100 μL of sample was mixed with 400 μL of acidified-acetone: methanol

(50:50) (Acidified acetone was prepared by adding 1.00 μL concentrated HCl to 12.0

mL of acetone) and kept at -20 oC for overnight. Samples were vortexed and incubated

for 1 hr at -80 oC. Then samples were centrifuged at 15,000 X g for 15 min at 4 oC. The

supernatant was transferred to new 1.5 mL Eppendorf tube and dried under a stream of

nitrogen gas. Dried samples were reconstituted in 100 µL of 0.1% formic acid in water

solution.

v. Aqueous acid precipitation:

100 μL of sample was mixed with 10% TFA or TCA solution. Samples were

vortexed and incubated for 1 hr at -80 oC. Then samples were centrifuged at 15,000 X

g for 15 min at 4 oC. The supernatant was transferred to new 1.5 mL Eppendorf tube

and dried under a stream of nitrogen gas. Dried samples were reconstituted in 100 µL

of 0.1% formic acid in water solution.

3.2.6 Instrumentation

Extracted peptide samples were analyzed using an Agilent infinity 1290 UPLC system interfaced to an Agilent QTOF 6540 mass spectrometer as described in section

3.2.3.1.

3.2.7 Data processing

Data was processed using Agilent Qualitative analysis and Bioconfirm software,

Expasy findPept and manual analysis to analyze amino acid sequence.

3.3 Results and Discussion

PART 1

3.3.1 Optimization of sample preparation time

To optimize the sonication time in an ultrasonic water bath for each reduction and alkylation step, two sets of BSA samples in triplicate were reduced and alkylated for 0, 2,

5 10 and 15 mins followed by 15 min microwave trypsin digestion for one set and 17 h conventional trypsin digestion for the other set. The number of peptides matched to the database and the protein sequence coverage was compared with the samples prepared using the conventional method of 1 h reduction and 45 min alkylation followed by 17 h trypsin digestion or 15 min microwave digestion. BSA was identified from the tryptic digest of samples at all the sonication times studied. But, the number of peptide matches and protein sequence coverage was less for 0 and 2 min time study and no difference was observed with rest of the sonication time studies (Fig. 3.2). From the samples reduced and alkylated for 0 and 2 min, the LC-MS/MS results showed significantly lower number of MS/MS spectra, which indicates that there is only partial enzymatic digestion. This may due to the incomplete reduction of disulfide bonds which might be hindering the Lys and Arg residues and not cleaved by trypsin. And samples reduced and alkylated for 5, 10 or 15 min the number of peptides identified from the MS/MS data and the percent sequence coverage

was high and equal in all three samples. This data suggest that 5 to 15 min time is optimum for each reduction and alkylation steps. Thus, to make the procedure more robust, we selected 5 min as the optimum time for further studies.

In order to compare the sample preparation methodology using an ultrasonic water bath on glycoproteins and complex mixture of protein samples, we digested Human IgG, a simple mixture of BSA and Human IgG and complex mixture of protein samples from human serum using three different protocols mentioned in Table.3.2.

35 30 25 20 15 10 5

0 No. of MS/MS spectra identifiedspectra MS/MS of No. 0 3 6 9 12 15 Time, min

Figure 3.2 Time study analysis to optimize the reduction and alkylation of proteins. The line chart represents the number of MS/MS spectra identified by MASCOT when BSA was reduced and alkylated each for 0, 25,10,15 min in the ultrasound water bath

Table 3.2 Protein sample preparation and digestion protocols

Protein Preparation Time (min) Total Time Microwave Reduction Alkylation (Min) digestion*

Current a 60 45 15 120 Method

b Our Method 5 5 15 25

c Control 0 0 15 15

a. Protein samples reduced and alkylated at 25 oC in a heating block for 1 h and 45 min respectively b. Protein samples reduced and alkylated in an ultrasonic water bath at 37oC at 143 W ultrasonic power for 5 min each. c. No reduction and alkylation were done on protein samples. * All samples were digested using CEM discovery microwave at 50 W power and 55 oC temperature

Figure 3.3A shows the total ion chromatograms(TIC) of a tryptic digest of BSA samples prepared using the three different protocols s mentioned in Table 3.1. As shown, samples prepared using the conventional procedure and those prepared using 5 min reduction and alkylation each in the ultrasonic water bath have the same TIC with similar intensities. Also, the MS spectra of the peptides from the tryptic digest of the both samples are also similar (Fig. 3.3B) despite the differences in sample preparation time. The number of identified MS/MS spectra for a protein can be used as the measure of that protein’s abundance (21-23). In this study, the MS/MS spectra number from protein samples was used to evaluate the efficiency of each sample preparation protocol. For the same sample amount, the more MS/MS spectra identified, the higher the efficiency of the protocol was.

From the data, the number of MS/MS spectra identified in both samples showed the same trend, thus suggesting that reduction and alkylation of proteins can be achieved in 5 min for each using ultrasonic water bath and showed that similar peptide recover efficiency with that of conventional sample preparation methods could be attained (Fig. 3.4). As can be seen in the Figure 3.4, the percentage protein sequence coverage and the number of peptide matches were similar for all both methods. However, the MS/MS spectral number for samples prepared without any reduction or alkylation was much less than that of the other two samples, indicating reduced digestion efficiency.

Figure 3.3 Total ion chromatograms and mass spectra of tryptic digest of BSA samples prepared by different sample reduction and alkylation protocols. A) Comparison of chromatograms of digestion products for BSA; B) Comparison of average mass spectra of digestion products for BSA.

Note: Blue lines represent samples prepared using current method: Protein samples reduced and alkylated at 25 oC in a heating block for 1 h and 45 min respectively. Red lines represent samples prepared using our method: Protein samples reduced and alkylated in an ultrasonic water bath at 37oC at 143 W ultrasonic power for 5 min each. Yellow lines represent control samples: No reduction and alkylation were done on protein samples.

spectrum 10

5 Number identifiedMS/MS of Number 0 A B C Sample #

Figure 3.4 Comparison of the number of MS/MS spectra identified for tryptic digest of

BSA samples prepared using different protocols analyzed by MASCOT search engine.

A. Protein samples reduced and alkylated at 25 oC in a heating block for 1 h and 45 min respectively. B. Protein samples reduced and alkylated in an ultrasonic water bath at 37oC at 143 W ultrasonic power for 5 min each. C. No reduction and alkylation were done on protein samples.

3.3.1.2 Application of optimized method on mixture of proteins and serum proteins

Based on the BSA results the most convenient and fastest protocol is to reduce and alkylate the proteins in ultrasonic water bath for 5 min each and to digest using the microwave protocol. A protein mixture of BSA and hIgG, pure hIgG were used to retest the sample preparation efficiency of the protocol described above. As shown in Table 3.3, no significant difference observed in the number of MS/MS spectra identified and the percent sequence coverage between samples prepared by the conventional method and by using an ultrasonic water bath. However, the MS/MS spectral number for samples prepared without any reduction or alkylation was much less than that of the other two samples, indicating reduced digestion efficiency.

Table 3.3 Comparison of protein sequence coverages (%) for tryptic digest of various protein samples prepared using different protocols and analyzed by MASCOT search engine

Protein Sequence Coverage (%)

Protein BSA & hIgG BSA hIgG (1:1)

Current method a 57 49 46 & 33

Our Method b 57 49 46 & 33

Control c 12 8 5 & 3

Further experiments with protein samples from human serum were used to prove high throughput sample preparation efficiency for complex protein mixtures. Pooled human serum proteins were obtained by acetone precipitation. Protein mixture was reduced and alkylated in an ultrasonic water bath for 5 min each, digested with trypsin, and resolved by LC-ESI-MS. The number of proteins identified with our method versus the standard protocol was approximately same. The number of identified MS/MS spectra obtained with our protocol was same as that with the standard protocol. From the above results, we concluded that the sample preparation protocol using ultrasonic water bath could obtain digestion efficiency similar to that of the standard protocol (Figure 3.5A &

B).

180

120

60 Number of proteins identified of proteins Number 0 A B

Figure 3.5 Total ion chromatograms and number f proteins identified from tryptic

digest of serum samples prepared by different sample reduction and alkylation protocols.

A) Comparison of chromatograms of digestion products for serum proteins; B)

Comparison of number of protein identified from each method.

PART 2

3.3.3 Comparison of extraction methods

Extraction efficiency of the methods was assessed by comparing the signal of spiked peptide peak area with that of peak area of peptide in solvent. Also, interference from other endogenous proteins and peptides was also monitored during the LC-MS analysis. Methanol, acetone and acetonitrile extraction methods have almost completely removed all of the plasma proteins. But the recovery of the peptide is < 50% when compared the peak area of the peptide which may be due to the co-precipitation properties.

Aqueous acid precipitating using TFA or TCA also removed plasma proteins, but the recovery is found to be less than < 20 %. Neutralization followed by solid phase extraction using C18 Ziptips was performed as such high concentrations of acid must be removed before LC-MS analysis. This whole sample preparation process might be the reason for low recovery of the peptide. Precipitation efficiency in any extraction method varies with salt concentration. The protein yield from an any precipitation in a complex biological matrix correlates with protein concentration, dielectric strength of the solution, and intrinsic protein charge

Out of all five extraction methods, extraction using acidified – acetone: methanol gave good recovery (> 86%) with complete depletion of plasma proteins (Figure. 3.6). .

After, determining the extraction efficiencies, stability of the peptide in plasma was analyzed over time. It was found that the spiked Stapled peptide was completely degraded when peptide was incubated for 10 hr in plasma (Figure 3.6).

QEKD

SEITKQEKD

Fragment Fragment

peptide human plasma at plasma human 10h peptide

peptide extracted from Plasma. from peptideextracted

apled apled

SHR

plasma

in human plasmahuman in

human

Fragment Staple Fragment

SHRHYC SHR

- -

peptide solvent in peptide (10 uM)

Control human plasma Controlhuman

Control Control

SEITKQEKD Staple

Stapled

QEK

peptide

Stapled

SHRHYC

Extracted ion chromatogram of mass and chromatogram ion spectra Extracted

Extracted chromatogramion of Stapled Extracted chromatogramion of Human blank plasma Extracted chromatogramion of spectraMass of spectraMass Humanof blank plasma spectraMass peptideof fragment spectraMass peptideof fragment spectraMass peptideof fragmentStaple spectraMass of fragment

eptide eptide

A) B) C) D) E) F) G) H) I)

Fragment Fragment

Peptide in human plasma at at plasma human in Peptide 0h Figure 3.6 Figure

3.4 Conclusion

In this study, we developed a fast and efficient protocol for protein preparation for digestion useful for both conventional and microwave digestion of protein. The sample preparation protocol developed enabled protein mixtures to be prepared in 10 min compared to 105 min for digestion and showed protein efficacy similar to that using standard protocols. For large scale clinical proteomic research, one of the important tasks is to prepare and digest tens or hundreds of protein samples. Present standard sample preparation and digestion protocols are labor-intensive and time-consuming. Though the digestion times were reduced significantly with the use of microwave techniques, protein reduction and alkylation processes are still time-consuming. The sample preparation protocol we present here simplifies the sample preparation procedure and will be helpful for the application of proteomics to biological, clinical research. Sample preparation for peptide analysis is an important part of MS. There seem to be significant differences between the peptide extraction ability of the five tested methods, with acidified – acetone: methanol extraction showed to have the highest precipitation activity and peptide recovery amongst all. Further testing and investigation into possible method optimization to possibly extract endogenous peptides may be useful.

3.5 References

1. Cañas, B.; Piñeiro, C.; Calvo, E.; López-Ferrer, D.; Gallardo, J. M. Trends in

sample preparation for classical and second-generation proteomics. Journal of

Chromatography A 2007, 1153, 235-258.

2. Capelo, J. L.; Carreira, R.; Diniz, M.; Fernandes, L.; Galesio, M.; Lodeiro, C.;

Santos, H. M.; Vale, G. Overview on modern approaches to speed up protein

identification workflows relying on enzymatic cleavage and mass spectrometry-

based techniques. Anal. Chim. Acta 2009, 650, 151-159.

3. López-Ferrer, D.; Cañas, B.; Vázquez, J.; Lodeiro, C.; Rial-Otero, R.; Moura, I.;

Capelo, J. L. Sample treatment for protein identification by mass spectrometry-

based techniques. Trends in Analytical Chemistry 2006, 25, 996-1005.

4. Nelson, D. N. and M. M. Cox, Eds. Lehninger: Principles of Biochemistry,

Freeman and Company, 2008.

5. Vukovic, J., et al."Improving off-line accelerated tryptic digestion: Towards fast-

lane proteolysis of complex biological samples." Journal of Chromatography A,

2008 1195(1-2): 34-43.

6. Hustoft, H. K., et al. “Critical assessment of accelerating trypsination methods.” J.

Pharmaceut. Biomed. Anal. 2011, 10.1016/j.jpba.2011.08.013.

7. Havliš, J., et al. "Fast-Response Proteomics by Accelerated In-Gel Digestion of

Proteins." Anal. Chem.2003, 75(6): 1300-1306.

8. Carreira, R. J., et al. "Indirect ultrasonication for protein quantification and peptide

mass mapping through mass spectrometry-based techniques." Talanta 2010, 82(2):

587593.

9. Rial-Otero, R.; Carreira, R. J.; Cordeiro, F. M.; Moro, A. J.; Santos, H. M.; Vale,

G.; Moura, I.; Capelo, J. L. Ultrasonic assisted protein enzymatic digestion for fast

protein identification by matrix-assisted laser desorption/ionization time-of-flight

mass spectrometry. Journal of Chromatography A 2007, 1166, 101-107.

10. Sun, W.; Gao, S. F.; Wang, L. F.; Chen, Y. F.; Wu, S. F.; Wang, X. F.; Zheng, D.

F.; Gao, Y. “Microwave-assisted protein preparation and enzymatic digestion in

proteomics”. Molecular & cellular proteomics: MCP JID - 101125647 0621.

11. Lill, J. R., et al. "Microwave-assisted proteomics." Mass Spectrometry Reviews

2007, 26(5): 657-671.

12. Segu, Z. M.; Hammad, L. A.; Mechref, Y. Rapid and efficient glycoprotein

identification through microwave-assisted enzymatic digestion. Rapid

Communications in Mass Spectrometry 2010, 24, 3461-3468.

13. Hale, J. E., et al. "A simplified procedure for the reduction and alkylation of

cysteine residues in proteins prior to proteolytic digestion and mass spectral

analysis." Analytical Biochemistry, 2004, 333(1): 174-181.

14. Hanne Kolsrud Hustoft, Helle Malerod, Steven Ray Wilson, Leon Reubsaet, Elsa

Lundanes and Tyge Greibrokk .” A Critical Review of Trypsin Digestion for LC-

MS Based Proteomics, Integrative Proteomics”, Dr. Hon-Chiu Leung (Ed.), 2012;

ISBN: 978-953-51-0070-6.

15. H. Buchwald, et al.Effects on GLP-1, PYY, and leptin by direct stimulation of

terminal ileum and cecum in humans: implications for ileal transposition. Surg.

Obes. Relat. Dis. 2014.

16. C. Giordano, et al.Neuroactive peptides as putative mediators of antiepileptic

ketogenic diets. Front. Neurol., 5 2014, p. 63.

17. A.A. Kaspar, J.M. ReichertFuture directions for peptide therapeutics development.

Drug Discov. Today, 18 2013, pp. 807-817.

18. Sadek PC, Carr PW, Bowers LD, Haddad LC. A radiochemical study of irreversible

adsorption of proteins on reversed-phase chromatographic packing materials. Anal

Biochem., 1986, 153(2):359–371.

19. Gillette MA, Mani DR, Carr SA. Place of pattern in proteomic biomarker

discovery. J Proteome Res., 2005, 4(4):1143–1154.

20. Anderson NL. The Human Plasma Proteome: History, Character, and Diagnostic

Prospects. Mol Cell Proteomics., 2002, 1(11): 845–867.

21. Sun, W., Wu, S. Z., Wang, X. R., Zheng, D. X., and Gao, Y. H. A systematical

analysis of tryptic peptide identification with reverse phase liquid chromatography

and electrospray ion trap mass spectrometry. Genomics Proteomics Bioinformatics.

2004; 2, 174–183 32.

22. Sun, W., Wang, X. R., Wu, S. Z., Zheng, D. X., and Gao, Y. H. An analysis of

protein abundance suppression in proteomics research with five protein digestion

mixtures. Eur. J. Mass Spectrom.2005; 11, 575–580 33.

23. Sadygov, R. G., and Yates, J. R., III A model for random sampling and estimation

of relative protein abundance in shotgun proteomics. Anal. Chem. 2004; 76, 4193–

4201.

24. Granvogl B et al, Sample preparation by in-gel digestion for mass spectrometry-

based proteomics. Anal Bioanal Chem. 2007; 389(4):991-1002.

Chapter IV

Structural Characterization of Glycoprotein by Mass

Spectrometry

Abstract

A common post-translational modification of proteins is glycosylation. By glycosylation, a given protein may contain multiple forms of glycan moieties at different locations. Glycan moieties are involved in signal transduction, extracellular and cell-cell recognition, and molecular trafficking, etc. Structural characterization of proteins and their post-translational modifications is particularly important for the study of human diseases such as heart disease, cancer and diabetes. To correlate the functional features of a glycoprotein with its structure (both temporally and spatially), a complete structural analysis of glycan is desirable. However, the complexity of carbohydrate structures and their derivatives impose a real challenge to such study.

Human IgG is a glycoprotein and its physiologic significance and function are mainly regulated by the N-glycans that attach to each heavy chain. In this study, we used human IgG as a model protein to establish methods of sample preparation, analysis and bioinformatics for complete structural elucidation. In our experiments, human IgG was first prepared by trypsin digestion, enzyme deglycosylation, chemical derivatization

(permethylation,), exoglycosidase digestion, and PMAA derivatization (partially methylated aditol acetates) reaction, then analyzed using LC-MS, LC-MS/MS and GC-

MS, and characterized by bioinformatics. The glycan composition, linkage and configuration obtained will be presented.

4.1 Introduction

Glycosylation is one of the most importnt protein posttranslational modifications and is estimated that more than 50 % of human proteins are glycosylated (1). Glycans are involved in involved in many biological processes and especially its functional role in cellular interaction with respect to adhesion, cell growth, and signaling is prone to be affected in cancer progression, invasion, and metastasis (2•7). Several cancer-associated alterations in protein glycosylation have been reported: (1) increased branching of N- glycans, (2) higher density of O-glycans, and (3) incomplete synthesis of glycans. More particularly, an increased or induced expression of N-acetylglucosamine transferase V resulting in N-glycan structures with β1–6GlcNAc antennae (8-10), and the expression of

(sialyl) Tn-antigens 23 as aberrant O-glycosylation have been reported. Recent studies in prostate cancer have also shown changes in fucosylation associated with progression of the disease (11). Thus, it is of great interest to identify and validate glycoproteins with altered

glycosylation whose function may reveal information of critical events in cancer progression. The other most common types of glycan aberrations include changes in branching and increases or decreases in fucosylation, mannosylation, or sialylation (8).

There are two major types of protein glycosylation, O-glycosylation, where glycans attach to the oxygen of Ser or Thr, and N-glycosylation, where glycans attach to the amide group of Asn in the consensus sequence Asn-XXX-Ser/Thr (XXX is not Proline). To release O-glycans from the peptide backbone, harsh chemical reactions with anhydrous hydrazine were used as no specific enzyme is no available. Whereas, N-glycans can be easily released with peptide N-glycosidase F (PNGase F). N-glycans share a common pentasaccharide core composed of two N-acetylglucosamines (GlcNAc) and three mannoses (Man). There are three types of N-glycans: a) high-mannose type (where only mannoses are added to the core), b) complex-type (which contains N-acetyllactosamine

(Gal-GlcNAc) in the antennal region), and c) hybrid type (which carries both mannose and

N-acetyllactosamine).

Mass spectrometry has been widely used for both the qualitative and quantitative study of N-glycans. However, the use of mass spectrometry for intact glycans remains challenging because glycans have low ionization efficiency, and components such as neuraminic acid (NeuAc) are labile and vulnerable to in-source or post-source fragmentation. Several derivatization techniques such as reductive amidation, methyl esterification, and permethylation have been introduced to address these issues.

Permethylation, which converts the free hydroxyl groups in glycans to methoxy groups, is widely used because of its simplicity and robustness (9). The conversion greatly enhances the ionization efficiency and sensitivity of mass spectrometric analysis. Permethylation

stabilizes sialic acids, enabling the simultaneous analysis of neutral and acidic glycans in positive ion mode. Furthermore, permethylated glycans produce more informative fragments in CID MS/MS, facilitating the structural determination of glycans (10). A multi-institutional comparative study revealed that MALDI-TOF MS analysis of permethylated glycans was highly reproducible and yielded satisfying quantitation results

(11). Fucosylation occurs at the maturation stage of N-glycosylation where fucoses attach to the core GlcNAc via an 1•6 linkage or the subterminal GlcNAc via an 1•3 or 1•4 linkage

(12). Fucosylation levels of several serum glycoproteins have been found to change in various types of cancers (7, 13, 14). Fucosylation of human alphafetoprotein has been found to be elevated in liver cancer patients and is used in clinical screening for liver cancer

(15).

4.2 Experimental

4.2.1 Chemicals and materials

Human IgG (HIgG), iodoacetamide (IAA), ammonium bicarbonate, acetic acid, formic acid, sodium borodeuteride 98 atom% D(NaBD4), sodium hydroxide and dimethylsulfoxide were ordered from Sigma Aldrich (St. Louis, MO, USA). Dithiothreitol

(DTT), sequencing grade modified trypsin was purchased from Promega (Madison, WI,

USA). Trifluoroacetic acid (TFA) and C18 ZipTips were obtained from EMD Millipore

(Massachusetts, MA, USA). HPLC grade Acetonitrile, methanol, chloroform and water were purchased from Fischer Scientific (Hanover park, IL, USA). PNGase F, deglycosylation enzyme mix and α 1-2 mannosidase were purchased from New England

Biolabs (Ipswich, MA, USA). Myo-inositol and iodomethane(ICH3) were obtained from

VWR (Randor, PA, USA).

4.2.2 Preparation of glycoprotein for LC-MS/MS and GC-MS analysis

Structural characterization of the intact glycoprotein involves the digestion of glycoprotein into glycopeptide, identification of the peptide, the potential glycosylation site as well as the glycan that is attached. Followed by the enzymatic release of glycans from the protein and identification of glycan and linkage analysis. Liquid chromatography/mass spectrometry and tandem mass spectrometry (LC-MS/MS) is the technique that is most commonly used for the characterization of glycopeptide and Gas chromatography/mass spectrometry (GC-MS) is the most common technique for identification of glycans and the linkage analysis. Figure 4.1 illustrates the schematic work flow followed for structural analysis of HIgG.

Figure 4.1 Schematic work flow for the structural analysis of glycoprotein HIgG

4.2.2.1 Reduction, alkylation and trypsin digestion of glycoprotein

1.00 mg of HIgG were weighed separately and dissolved in 1.00 mL of 100 mM ammonium bicarbonate solution to make a 1.00 mg/mL solution. 40.0 µg of sample was aliquoted into a 1.5 mL micro centrifuge tube. Glycoprotein was then reduced by the adding 1.00 M DTT to a final concentration of 10.0mM and incubated at room temperature for 1 h and alkylated by adding 100 mM IAA to a final concentration of 55.0 mM and incubated for 45 min in the dark. Excess IAA was then neutralized by adding 10.0 µL of

1.00 M DTT and incubated for 45 min. Sample was then digested by adding trypsin and incubated at 37oC for 17 h in a water bath (enzyme-to-substrate ratio of 1:20 (wt/wt)). The enzyme digestion was quenched by adding 10.0 µL of 5% formic acid. A portion of the digestate (~20.0 µg) was desalted using C18 ZipTips and eluted in 0.1% FA for further analysis of glycopeptides using LC-MS/MS. Samples were lyophilized and stored at -20 oC until further analysis. Remaining portion of the digestate (~20.0 µg) was lyophilized for glycan analysis.

4.2.2.2 Deglycosylation: Release of N-Glycans from glycopeptides using endoglycosidase

Intact N-glycans attached to the HIgG were removed using endoglycosidase enzyme as following: Freeze dried glycopeptide was reconstituted in 200 μL of freshly prepared 50 mM ammonium hydrogen carbonate solution. Intact N-glycans were released by adding 3.00 μL of the PNGase F enzyme (1 U / μL) solution and incubated at 37 °C for

16 h in a water bath. Samples were lyophilized and stored at -20 oC until further purification of released N-glycans.

4.2.2.3 Purification of released N-glycans

Samples were dissolved in 200 µl of 5% acetic acid. Sep-Pak C18 column

(Phenomenex, Torrance, CA, USA) was conditioned with 5.00 ml of methanol and then with 10.0 ml of 5% acetic acid. Then, sample was loaded on to the column and eluted stepwise with 3.00 ml of 5% acetic acid in water and 4.00 ml of 5% acetic acid in 80% acetonitrile and collected each fraction. N-glycans were eluted with 5% acetic acid. The fraction eluted with 5% acetic acid in 80% acetonitrile contains deglycosylated peptides.

Collected intact N-glycan sample was aliquoted into three different vials and lyophilized and stored at -20 oC for further LC-MS/MS analysis, exoglycosylation and permethylation of glycans.

4.2.2.4 Exoglycosidase digestion: release of monosaccharides

To define the monosaccharides and their anomeric configurations, and to confirm sequences glycans were treated with exoglycosidases. An aliquot sample of released and freeze-dried intact N-glycans was dissolved in 200 μL 50.0 mM ammonium acetate buffer.

Monosaccharide were released by adding 50.0 μL of exodeglysoyaltion mix equivalent to

50 mU of α-Sialidase, 10 mU of β-Galactosidase, 310 mU of α-N-Acetylglucosaminidase,

40 mU α-Fucosidase and 525 mU α-Mannosidase) and incubated at 37 °C for 48 h in a water bath. Enzyme reaction was quenched by heating the solution to 100°C for 10 min.

Sample were freeze-dried and stored at -20 °C until permethylation.

4.2.2.5 Permethylation of glycans

Permethylation stabilizes the relatively labile sialic acids and fucoses, and significantly improves the sensitivity of glycans. Permethylation is the most important type of derivatization used in LC-MS analysis of glycans and GC-MS analysis for linkage analysis. Both the intact N-glycans and released monosaccharides were permethylated in a vacuum vessel saturated with an argon atmosphere as follows: 500 µl of DMSO was added to the lyophilized sample. Then approximately 25.0 mg of NaOH (around 10 pellets of NaOH were crushed in a dry mortar) was added to the sample. 300 µl of ICH3 was then added and flushed the tube with a stream of argon. The sample tube was vortexed vigorously and the reaction mixture was placed in an ultrasonic bath for 90 min at room temperature. Reaction was quenched by adding 1.00 mL of 5% acetic acid at 4 °C and vortexed vigorously. Finally, permethylated glycans were extracted by adding 600 µl of chloroform. Sample tubes were vortexed at high speed for a minute, and allowed the mixture to settle into two layers at 4 °C. The lower chloroform phase was transferred into a new glass tube. Chloroform extraction was repeated twice and the collected chloroform phases were combined. Then, chloroform phase was washed eight times with 1 volume of water at 4 °C and discarded aqueous phases. Chloroform phase was dried down under a stream of nitrogen in a hood and stored at -20 °C until further purification.

4.2.2.6 Purification of permethylated of glycans

Dried samples from above step were dissolved in 200 µL of methanol. Sep-Pak

C18 columns were used to clean up the salts. Columns were conditioned with 5.00 mL of

methanol and then with 10.0 mL of water. Sample was loaded on to the column and eluted sequentially with 15.0 mL of water, 2.00 mL of 10% ACN and 3.00 mL of 80% ACN.

Permethylated glycans eluted in 80% ACN fraction were collected and evaporated under a stream of nitrogen in a hood. Samples were aliquoted into two vials and were freeze-dried and stored at -20 °C until further analysis.

4.2.2.7 Preparation of Partially Methylated Aditol Acetates (PMAAs)

The permethylated glycan are further derivatized for the GC-MS analysis to analyze the linkage for structural conclusions. Permethylated glycans were derivatized to partially methylated aditol acetates as follows: Permethylated glycans were dissolved in 300 µL of

4 M TFA in a glass tube and incubated at 100 °C for 4 h. Samples were dried under a stream of nitrogen. The dried samples were then dissolved in 200 µL of reduction solution

(2 M NH4OH solution containing 4.00 mg of NaBD4 per mL) and incubated at room temperature overnight. Reaction was quenched by adding concentrated acetic acid dropwise until no fizzing was observed (approximately three drops). Samples were dried down under a stream of nitrogen. Borate salts were removed by repeated washing and evaporation with 500 µL methanol containing 5% acetic acid under a stream of nitrogen.

Sample washing was repeated for five times and the mixture was evaporated to a complete dryness under a stream of nitrogen.

Samples were then peracetylated by adding 20.0 µL of pyridine and 200 µL of acetic anhydride and incubated at 100 °C for 2 h. Samples were evaporated to dryness under a stream of nitrogen. Dried samples were then dissolved in 400 µL of chloroform

and 1.00 mL of water, vortexed vigorously, and left at room temperature for the mixture to settle into two layers. The lower chloroform phase was transferred into a new glass tube.

Another 400 µL of chloroform was added into the first tube, mixed vigorously, and allowed the mixture to settle into two layers. The lower chloroform phase was combined with the previous chloroform phase. Chloroform extraction step was repeated once. Chloroform phase was washed ten times into the new glass tube with 1 volume of water and discard aqueous phases. Chloroform phase was dried down under a stream of nitrogen. Dried

PMAAs were stored at -20 °C until further GC-MS analysis.

4.2.3 Instrumentation

4.2.3.1 LC-MS/MS analysis of glycopeptides and peptides

Protein digests reconstituted in 50:50:0.1% ACN: water: formic acid were analyzed using a Prominence UFLC system (Shimadzu, Columbia, MD, USA) for analyte separation and a QSTAR Elite tandem mass spectrometer (AB Sciex, Foster City, CA, USA). The system was controlled by AB Sciex Analyst® software (version 1.6.1) for its operation, data acquisition, and processing. The UFLC system included a system controller (CBM-

20A), two binary pumps (LC-20AD), a temperature-controlled autosampler (SIL 20AHT) and an online degasser (DGU20A3).

The separation of peptides was carried out on an Aries peptide - C18 column (3 X

250mm; 6 u particle size) (Phenomenex, Torrance, CA, USA) maintained at room temperature using mobile phases A (0.1% FA in water) and B (0.1% FA in Acetonitrile).

For each analysis, 10.0 μL of sample was injected into the system by autosampler set at

4°C. The elution gradient was: 5% B at 0 min, 0-25% B from 0-29 min, 25-45% B from

29-54 min, 45-60% B from 54-77 min, 60-5% B from 77-85 min, with a total run time of

80 min including equilibration. at a flow rate of 0.200 ml/min. The LC elute was introduced into the mass spectrometer operated under positive mode and the data was acquired in a data-dependent manner. Prior to sample analysis, the column was first equilibrated with at least 25 column volumes of the mobile phase at a flow rate of 0.200 mL/min.

The optimized instrument parameters were as follows: curtain (CUR), 30 psi; collision activated dissociation (CAD) gas, medium; nebulizer gas (GS1), 30 psi; turbo heater gas (GS2), 10 psi; turbo ion spray voltage (IS), +5500 V; source temperature (TEM),

40 °C; declustering potential (DP), 80 V; entrance potential (EP), 230V; collision energy

(CE), 15 V; collision cell exit potential (CXP), 18 V; mass resolutions (Q1 and Q3), 1 unit.

The data-dependent acquisition mode was used for analyte quantitation. Protein identification was done using MASCOT search.

4.2.3.2 LC-MS/MS analysis of permethylated glycans

Permethylated glycans reconstituted in 50:50:0.1% ACN: water: formic acid was analyzed using a Prominence UFLC system (Shimadzu, Columbia, MD, USA) for analyte separation and a QSTAR Elite tandem mass spectrometer (AB Sciex, Foster City, CA,

USA). The system was controlled by AB Sciex Analyst® software (version 1.6.1) for its operation, data acquisition, and processing. The UFLC system included a system

controller (CBM-20A), two binary pumps (LC-20AD), a temperature-controlled autosampler (SIL 20AHT) and an online degasser (DGU20A3).

The separation of glycans was carried out on an Hypercarb - C18 column (2.1 X

100mm; 3 u particle size) (Thermo Scientific, Waltham, MA, USA) maintained at room temperature using mobile phases A (0.1% FA in water) and B (0.1% FA in Acetonitrile).

For each analysis, 10.0 μL of sample was injected into the system by autosampler set at

4°C. The elution gradient was: 5% B at 0 min, 0-25% B from 0-19 min, 25-45% B from

19-24 min, 45-60% B from 24-47 min, 60-5% B from 47-60 min, with a total run time of

60 min including equilibration. at a flow rate of 0.200 ml/min. The LC elute was introduced into the mass spectrometer operated under positive mode and the data was acquired in a data-dependent manner. Prior to sample analysis, the column was first equilibrated with at least 25 column volumes of the mobile phase at a flow rate of 0.200 mL/min.

The optimized instrument parameters were as follows: curtain (CUR), 30 psi; collision activated dissociation (CAD) gas, medium; nebulizer gas (GS1), 30 psi; turbo heater gas (GS2), 5 psi; turbo ion spray voltage (IS), +5500 V; source temperature (TEM),

40 °C; declustering potential (DP), 80 V; entrance potential (EP), 230V; collision energy

(CE), 15 V; collision cell exit potential (CXP), 10 V; mass resolutions (Q1 and Q3), 1 unit.

Glycan bioinformatics was performed using SimGlycan® software (Premier Biosoft, Palo

Alto, CA, USA).

4.2.3.3 GC-MS analysis of PMAAs

100

The GC-MS system used for the linkage analysis in this work consisted of a Varian gas chromatography interfaced with a Varian Saturn 2100T ion trap mass spectrometer

(Agilent technologies, Santa Clara, CA, USA). GC separation was carried out on Agilent

VF5-MS GC column (30 m x 0.25mm x 0.2 µm) with helium as a carrier gas at a constant flow rate of 1.00 mL/min. For each run, 1.00 µL sample was injected with a split ratio of

1:10, and the injector temperature was maintained at 250 o C. The temperature profile started at 130°C, then increased by 2 °C/min to 240 °C and held at 240 °C for 5 min.

Electron impact (EI) ionization mode was used for MS detection. The optimized mass spectrometry operating parameters are as follows: target TIC, 11000 counts; maximum ionization time, 25000 µs; pre-scan ionization time, 100 µs. Mass spectra were obtained in the mass range of m/z 100 to 1200.

4.2.4 Bioinformatics

For the protein identification, the acquired auto MS/MS were selected and exported as Mascot generic format (MGF). While exporting all spectra and extracted spectra were selected, and absolute height of ≥ 500 counts was set as peak filter. The peptide MFG was searched with Mascot Search Engine – MS/MS Ion Search against the Swiss-Prot

Database for protein identification. Cysteine carbamidomethylation was selected as fixed modification. The mass value was monoisotopic. Peptide mass tolerance was set to 50 ppm, and the MS/MS tolerance was at 0.6 Da. No restriction was placed on molecular weight range and taxonomy. Two miss-cleavage were tolerated. ESI-QUAD-TOF was selected as the instrument. A protein was regarded identified when it had a significant

101

Mascot probability score (scores greater than 60 correspond to P < 0.05), and at least two matched peptides.

For Glycan bioinformatic analysis SimGlycan was used for identification and annotation. SimGlycan is a high throughput data processing tool for characterizing glycans from MS and MS/MS data from various vendors. A high throughput search was performed on the data files with relative threshold intensity of 5% and with an error tolerance of 5 ppm for precursor ions and 10 ppm for MS/MS fragments. Glycans were identified by searching against SimGlycan database server and the match scores were assigned based on the number of MS/MS fragments matched with the database library.

For linkage analysis, peaks are identified their MS fragmentation pattern. This analysis indicates which residues are terminal and how each monosaccharide is substituted, and the occurrence of branching points.

4.3 Results and Discussion

In our work, we sought to develop LC-MS and GC-MS assay combined with enzyme digestion and derivatization to evaluate the glycoprotein and glycan structure of Human

IgG. The work flow of this study is outlined in Figure 4.2. Briefly, Human IgG was digested using trypsin followed byLC-QTOF- MS peptide analysis to verify protein identification using MASCOT database search. The Human IgG glycopeptides were then deglycosylated using the PNgase F and permethylated. Permethylated glycans were subject to mass spectrometric analysis for structural elucidation. SimGlycan was used for the database search. Permethylated glycans were further subjected to endoglycosylation to release each

102

glycan followed by derivatization to PMAAs. PMAAs were analysed on the GC-MS to perform the glycan linkage analysis.

4.3.1 Glycoprotein analysis of HumanIgG

Due to its mass accuracy, selectivity and high throughput mass spectrometry has become the most preferred technique for the analysis of glycopeptides. LC-MS/MS

Analysis of glycopeptides with the intact glycan structure can allow identification, site specific analysis of glycosylation, simultaneous analysis of other peptide sequences, and posttranslational modifications which can give a better understanding of the protein structure at a molecular level. Tryptict digested glycoproteins were subjected to LC-ESI-

QTOF-MS and MS-MS. The tandem MS data was manually inspected for the CID that produced the diagnostic peaks of glycopeptides. Figure 4.2 A & B shows a representative mass spectra and MASCOT peptide view of the glycopeptides. A good coverage (71%) of peptide sequence was obtained when the MS/MS data was searched against MASCOT search engine.

103

+TOF MS: Exp 1, 119.296 min from Sample 1 (igg msms) of igg msms.wiff Max. 952.0 counts. A a=3.57719274498003910e-004, t0=-1.19667517483394480e+001 (Turbo Spray) 824.8389 952 806.8349 500 412.9333 435.9086 636.9133 1236.7745 521.9687 710.8217770.8268 841.8503 974.6940 1113.4426 1157.4306 1258.7208 0 350In400 te450 n500 sity,550 600 650 co700 750 u800 n850 ts 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 m/z, Da +TOF MS: Exp 1, 111.453 min from Sample 1 (igg msms) of igg msms.wiff Max. 935.0 counts. a=3.57719274498003910e-004, t0=-1.19667517483394480e+001 (Turbo Spray)

824.8497 935

806.8349 500 412.9333 435.9165 579.7926 650.8727 725.7574 841.8503 956.6392 974.7057 1062.8178 1236.7480 0 350In400 te450 n500 sity,550 600 650 co700 750 u800 n850 ts 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 m/z, Da +TOF MS: Exp 1, 100.927 min from Sample 1 (igg msms) of igg msms.wiff Max. 320.0 counts. a=3.57719274498003910e-004, t0=-1.19667517483394480e+001 (Turbo Spray)

824.8389 320 478.8515 742.8174 806.8563 436.8683 682.7340 493.9738 357.9453 429.9308 641.7594 713.6254773.6416 935.5140 1012.6157 1072.6857 1154.7164 1236.7877 1337.5624 0 350In400 te450 n500 sity,550 600 650 co700 750 u800 n850 ts 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 m/z, Da +TOF MS: Exp 1, 91.104 min from Sample 1 (igg msms) of igg msms.wiff Max. 788.0 counts. a=3.57719274498003910e-004, t0=-1.19667517483394480e+001 (Turbo Spray)

603.6615 788

500 474.6337 530.6590 586.6583

479.8655 624.5628 436.8762 670.5103 824.8497 898.4465 971.6164 1172.2998 1194.2935 1226.1374

0 In te n sity, co u n ts 350 400 450 500 550 600 650 700 750 800 850 900 950 1000 1050 1100 1150 1200 1250 1300 1350 1400 1450 1500 m/z, Da

B ))

Figure 4.2 Representative A) mass spectra of the glycopeptides analyzed by LC- MS/MS

B) Peptide view of MS/MS spectra of the a glycopeptide.

104

4.3.2 Release and analysis of N-glycans

Mass spectrometry has been widely used for both the qualitative and quantitative analysis of N-glycans. But the low ionization efficiency of the intact glycans is a limiting factor in the analysis of glycans. And glycan components such as neuraminic acid (NeuAc) are labile and vulnerable to in-source or post-source fragmentation resulting in decreased sensitivity. Derivatization techniques such as reductive amidation, methyl esterification, and permethylation have been developed to resolve these issues. In permethylation, the free hydroxyl groups in glycans were converted to methoxy groups, and is widely used because of its simplicity and robustness. Permethylation stabilizes the relatively labile sialic acids and fucoses, and significantly improves the sensitivity and signal-over-noise ratio of glycans (16).

Glycans released from the glycopeptides using PNgase F were permethylated and analyzed using LC-MS/MS (Figure 4.3). The acquired MS/MS data was subjected to database search using SimGlycan software. SimGlycan software is a glycan and glycopeptide structure analysis software that uses mass spectrometric data to predict glycan structures from carbohydrate database. The glycan identification was based on measured accurate mass (within a 10 ppm window around the theoretical m/z of each glycan) and confirmed the glycan structure using MS/MS analysis. Figure 4.4 shows the mass spectrum of the permethylated glycans released from Human IgG. The N-glycan in the IgG comprises a core structure of N-acetylglucosamine (GlcNAc) and mannose, plus additional carbohydrate residues, including fucose, galactose, and bisecting mannose and GlcNAc.

105

DMSO Methyliodide N-Glycans Permethylated Glycans

NaOH

Figure 4.3 Schematic representation of permethlyation of glycans for LC-MS/MS analysis

106

Figure 4.4 LC-QOF-MS spectrum of permethylated glycans released from HIgG

107

4.3.3 Analysis of PMAAs using GC-MS – Linkage analysis

An important step in understanding the detail structure of an oligosaccharide is to determine how the individual monosaccharides are linked to one another. Linkage analysis, also known as methylation analysis provides information on the substitution position of their glyosidic linkages. The permethylated glycans were subjected to exoglycosidase digestion by various exoglycosidase enzymes and each released monosaccharaide was derivatized to PMAAs. Exoglycosidase hydrolysis produces individual permethylated monosaccharides with free hydroxyl groups at the positions that were previously involved in a linkage. Now, the partially methylated monosaccharides are derivatized to produce volatile molecules suitable to GC-MS analysis. The most common strategy involves reduction of the monosaccharides to produce alcohols at C-1 opening the ring structure, followed by acetylation of free hydroxyl groups (Figure 4.5). Individual components of the mixture of partially methylated (where methyl groups mark the hydroxyl groups that were initially free), partially acetylated (where acetyl groups mark hydroxyl groups initially at substituted, linked, or ring-closure positions) monosaccharide alditols can be analyzed using GC-MS. PMAAs are identified by a combination of GC retention times and mass fragmentation pattern (Figure 4.6A). The fragmentation patterns and retention times were compared with the GC retention times with those of known standards. The predominant linkages observed were: GlcNAc β1-4 linked to GlcNAc; GlcNAc β1-4 linked to mannose; bisecting Mannose is α1-3 linked; GlcNAc β1-2 linked to mannose;

Gal is β1-4 linked and terminal (Figure4.6B).

108

Incubated 2M ammonia+ Pyridine+Aceti Purification NaBD4 c anhydride at PMAAs (Partially permethylated overnight and drying glycans 200°C for 2 h. methyalted aditol acetates

Figure 4.5 Schematic representation of Partially Methylated Aditol Acetate derivatization of individual glycan released by exoglysidases.

109

β1-2 β1-4 MAN GlcNac Gal GlcNac GlcNac MAN β1-4 β1-4 MAN GlcNac Gal β1-2 β1-4 FUC

Figure 4.6 A) GC-MS chromatogram and mass spectra of Partially Methylated Aditol

Acetate derivatives of individual glycan released by exoglysidases. B) Linkage analysis

of glycans released from HIgG.

110

4.4 Conclusion

Comprehensive characterization of glycoprotein is critical to understand the structure and composition of the glycoprotein and attached glycans. The development of high mass accuracy and mass resolution MS instruments with versatile fragmentation methods makes the structural characterization of glycoprotein more easily and confidently.

A range of LC-MS and GC-MS methods along with enzymatic hydrolysis and derivatization have been developed for comprehensive characterization of primary structure of glycoprotein, intact glycan structures, and linkage analysis.

111

4.5 References

1. Hakomori S. (1996) Tumor Malignancy Defined by Aberrant Glycosylation and

Sphingo(glyco)lipid Metabolism. Cancer Res., 1996; 5309–5318.

2. Fernandes B., Sagman U., Auger M., Demetrio M., Dennis J. W; β1–6 branched

oligosaccharides as a marker of tumor progression in human breast and colon

neoplasia. Cancer Res., 1991; 718–723.

3. Brockhausen I; Mucin-type O-glycans in human colon and breast cancer:

glycodynamics and functions. EMBO Rep., 2006; 599–604.

4. Meany DL, Zhang Z, Sokoll LJ, Zhang H, Chan DW; Glycoproteomics for prostate

cancer detection: changes in serum PSA glycosylation patterns; J Proteome Res.,

2009; 613-9.

5. Miyoshi E., Moriwaki K., Nakagawa T; Biological Function of Fucosylation in

Cancer Biology. J. Biochem., 2008; 725–729.

6. Byrd J., Bresalier ; Mucins and mucin binding proteins in colorectal cancer. Cancer

Metastasis Rev., 2004; 77–99.

7. Tonetti M, Sturla L, Bisso A, Benatti U, De Flora A. Synthesis of GDP-L-fucose

by the human FX protein. J Biol Chem., 1996; 271:27274–27279.

8. Ohyama C, Smith PL, Angata K, Fukuda MN, Lowe JB, Fukuda M. Molecular

cloning and expression of GDP-D-mannose-4,6-dehydratase, a key enzyme for

fucose metabolism defective in Lec13 cells. J Biol Chem., 1998; 273:14582–14587.

9. Takeda K, Hayakawa Y, Smyth MJ, Kayagaki N, Yamaguchi N, Kakuta S, Iwakura

Y, Yagita H, Okumura K. Involvement of tumor necrosis factor-related apoptosis-

112

inducing ligand in surveillance of tumor metastasis by liver natural killer cells. Nat

Med., 2001; 7:94–100.

10. Morelle, W. & Michalski, J.C. The mass spectrometric analysis of glycoproteins

and their glycan structures. Curr. Anal. Chem., 2005; 1, 27–55.

11. Hakomori S. Cancer-associated glycosphingolipid antigens: Their structure,

organization, and function. Acta Anatomica., 1998; 161:79–90.

12. Pascher I. Molecular arrangements in sphingolipids: Conformation and hydrogen

bonding of ceramide and their implication on membrane stability and

permeability. Biochim Biophys Acta., 1976; 455:433–451.

13. Varki NM, Varki A. Diversity in cell surface sialic acid presentations: implications

for biology and disease. Lab Invest., 2007); 87(9):851–7.

14. Mujoo K, Cheresh DA, Yang HM, Reisfeld RA. Disialoganglioside GD2 on human

neuroblastoma cells: target antigen for monoclonal antibody-mediated cytolysis

and suppression of tumor growth. Cancer Res., 1987; 47(4):1098–104.

15. Misonou Y., Shida K., Korekane H., Seki Y., Noura S., Ohue M., Miyamoto

Y. Comprehensive Clinico-Glycomic Study of 16 Colorectal Cancer Specimens:

Elucidation of Aberrant Glycosylation and Its Mechanistic Causes in Colorectal

Cancer Cells. J. Proteome Res., 2009; 2990–3005.

16. Swetha Pyreddy. Qualitative and Quantitative Analysis of Glycans and

Glycopeptides Released from Model Glycoproteins and Biological Samples using

MRM Mode and ESI LC-MS/MS. Thesis. Texas Tech University; Lubbock, TX.

2012.p. 53-63.

113

Chapter V

Analysis of oxygen-18 labeled phosphate to study positional isotope experiments using QTOF LC-MS

Abstract

A new method was developed for the determination of oxygen-18 labeled phosphate to study positional isotope experiments using sensitive and rapid liquid chromatography – QTOF- mass spectrometry (LC-MS). The accurate mass LC-MS-based method for monitoring of 18O/16O exchange was validated with gas chromatography–mass spectrometry(GC-MS), the current standard methods to study positional isotope experiments. However, current methodologies require derivatization to gas and/or time consuming. Accuracy is demonstrated by good agreement with between GC-MS and LC-

MS methods. Results demonstrate that LC-MS provides a sensitive and robust analytical platform for simultaneous determination of levels and to analyze positional isotope experiments to study the reversibility of ATP hydrolysis by Lon proteases.

114

5.1 Introduction

The positional isotope exchange (PIX) technique has proven to be a valuable tool for both the identification of reaction intermediates and reaction reversibility and in the elucidation of the mechanistic pathways of many enzyme-catalyzed reactions (1,2). The general approach involves reactions being conducted in an isotope-enriched buffer solution so that the extent of the incorporation of the isotopes into the intermediates or products generated by the enzymatic reactions is used to evaluate the kinetics of the reaction or the extent of the reversibility process. PIX techniques have also been used in time course studies to calculate individual rate constants or the ratio of rate constants in a reaction mechanism (3-5). The most commonly used PIX technique for enzyme-catalyzed reactions involves phosphate and the transfer of oxygen. In oxygen transfer PIX reactions, labelled oxygen (18O) is exchanged with the oxygen bonded to an inorganic phosphate, and the measurement of the 18O/16O ratio is used to determine the reaction mechanism and kinetics.

The most commonly used method for detecting 18O/16O isotope enrichment involves gas chromatography–mass spectrometry (GC-MS) of a derivatised form of the product. Nuclear magnetic resonance (NMR) imaging involving 31P and 15N (6-8) has been used to study isotope enrichment utilizing labelled phosphate and nitrogen. The main disadvantage of GC-MS is that a volatile derivative of the isotope labeled compound needs to be prepared for analysis (9,10). However, although 31P NMR imaging offers a direct analysis method, its sensitivity is insufficient and the process is time-consuming (11,12).

Although oxygen labelled isotope exchange reactions have been widely used to monitor the reversibility of enzyme reactions and their kinetics (6), no sensitive and simple analytical method has been developed for the analysis of 18O-labeled phosphate in PIX

115

experiments. In order to address this shortcoming, this study proposes a simple, sensitive, and efficient LC-QTOF-MS method for studying 18O isotope enrichment in inorganic phosphate in PIX experiments. The accurate mass spectroscopy of this method allows for both labeled and unlabeled phosphate ions to be specifically detected, and this can be achieved at resolutions of over 20,000.

In this study, we demonstrate that this new method can be used to detect the reversibility of the adenosine triphosphate (ATP) hydrolysis of Lon proteases. Lon is an

ATP-dependent serine protease. It is known that ATP hydrolysis increases the protein hydrolytic activity of Lon (13,14). Despite being a homo-oligomer, Lon has a set of high- and low-affinity sites for ATP, which suggests that there is a functional non-equivalency in the sub-units of Lon. This half-site reactivity has already been demonstrated by chemical quench flow experiments, which have allowed for the initial turnover of ATP hydrolysis at different ATPase sites in Lon to be monitored. A reactivity of only 50% was observed in these experiments (13,14); this supported the half-site reactivity theory, in which identical

ATPase active sites on each Lon sub-unit possess two different affinities for ATP. In order to confirm that the 50% reactivity is due to the half-site reactivity and not the reversibility of ATP hydrolysis, and in order to better understand how ATP hydrolysis contributes to the enzyme catalysis of peptide bond breakages, we must determine if ATP hydrolysis is reversible in Lon. If it is, then one ATP molecule may be responsible for repeated rounds of protease activity, because the molecule can continuously regenerate. This reversibility can be detected by adding 18O-labeled water to an ATPase reaction mixture, as this allows for PIX to occur within phosphate molecules; we can then monitor the number of labeled

116

oxygen atoms incorporated in the resultant inorganic phosphate by using mass spectrometry (13-15).

The data we obtained using LC-QTOF-MS was compared to data obtained using

GC-MS, and they were found to be in good agreement. Our proposed method is therefore excellent for analyzing the isotope enrichment produced by PIX experiments during the study of enzyme reactions.

5.2 Experimental

5.2.1 Chemicals and materials

Ammonium formate, water with 25 atom% 18O, N-trimethylsilyldiethylamine

(TMSDEA) (95%), dithiothreitol, magnesium acetate, formic acid, and ethylenediaminetetraacetic acid (EDTA) were procured from Sigma-Aldrich (St. Louis,

MO, USA). Phosphoric acid (85.0%) and hydrochloric acid (38.0%) were procured from

Fischer Scientific (Hanover park, IL, USA). Deionized water (resistivity = 18.2 MΩ·cm) was prepared using a Barnstead Model 7148 Nanopure ultrapure water system from

Thermo Scientific (Waltham, MA, USA).

5.2.2 Sample preparation

For the analysis of the isotope enrichment performed by the LC-QTOF-MS method, standard working solutions (10.0, 50.0, 100, 250, 500, and 1.00 × 103 nM) of phosphoric

117

acid were prepared by diluting and mixing appropriate volumes of concentrated phosphoric acid (H3PO4, 85%) (14.62 M) with deionized water.

A stock solution of ammonium formate (0.10 M) was prepared by dissolving an appropriate amount of ammonium formate salt in a known volume of HPLC-grade water and storing it in a refrigerator at 4°C before use. A working solution of ammonium formate

(10.0 mM) was then prepared through a ten-fold dilution of the stock solution with HPLC- grade water; the pH was adjusted to 3.65 using formic acid, if necessary.

For the GC-MS analysis, samples were prepared and derivatized as described in earlier studies (6,13,15); in brief, 10.0 µL of phosphoric acid (ca. 500 nM in methylene chloride) was transferred together with 10.0 µL of TMSDEA into a 1.5 mL glass vial

(VWR, Radnor, PA, USA). After the vial cap had been tightened, a derivatization reaction was allowed to proceed at room temperature for 60 min with occasional shaking. The derivatization of the inorganic phosphate with TMSDEA generated tris(trimethylsilyl)phosphate (TMSP) (Figure 5.1 A&B). The final product was analyzed using GC-MS. The derivatized samples were stored at −20°C until the GC-MS analysis was performed.

118

Figure 5.1 A) Inorganic phosphate for LC-MS analysis. B) Trimethylsilyl phosphate is generated from the reaction of phosphate with trimethylsilyl diethylamine to allow analysis by GC–MS.

119

5.2.3 Instrumentation

5.2.3.1 LC-MS chromatography separation and mass spectrometric conditions

The LC-QTOFMS system used in this work consisted of an Agilent Infinity 1290

UPLC system that was interfaced with an Agilent QTOF 6540 mass spectrometer (Santa

Clara, CA, USA). Agilent Infinity 1290 UPLC is composed of a solvent reservoir, a degasser, a binary pump (G4220A), a thermostat (G1330B), an autosampler (G4226A), and a thermostatted column compartment (G1316C). The quadrupole time-of-flight (Q-

TOF 6540) mass spectrometer was equipped with an orthogonal ESI-interfaced Agilent Jet

Stream (AJS) operated under a negative mode; the data was acquired in a data-dependent manner. The instrument was controlled by a personal computer running the Mass Hunter

Workstation (version B.05.01) software from Agilent Technologies (Santa Clara, CA,

USA).

The LC separation was carried out on Acclaim Trinity P2 (100 × 2.1 mm; 3.6nm particle size) from Thermo Fisher Scientific (Waltham, MA, USA) maintained at 30°C with a mobile phase containing 25% acetonitrile and 75% 10.0 mM ammonium formate with a pH of 3.65 at a flow rate of 0.3 mL/min. Prior to the initial sample analysis, the column was equilibrated with at least 20 column volumes of the mobile phase. For each run, 10.0 μL of the sample was injected into the system by an autosampler that was set to

4.0°C.

The mass spectrometric detection was carried out using the negative electrospray ionization (ESI−) mode, which was tuned for compound-dependent and source-dependent

120

parameters. The optimized mass spectrometry operating parameters were as follows: capillary voltage, 2250V; nebulizer pressure, 40 psi; drying gas flow rate, 10 L/min; gas temperature, 350°C; sheath gas flow, 11 L/min; sheath gas temperature, 325°C; skimmer voltage, 65 V; and fragment voltage, 150 V. Both the centroid and profile modes were used in the data collection with an extended dynamic range of 2 GHz. MS accurate mass spectra were recorded for 90–110 m/z at a speed of 5 spectra/s, The TOF was calibrated on a daily basis, and a real-time calibration was performed by the continuous infusion of reference ion 121.0508.

5.2.3.2 GC-MS chromatography separation and mass spectrometric conditions

The GC-MS system used in this work consisted of a Varian gas chromatograph interfaced with a Varian Saturn 2100T ion trap mass spectrometer (Agilent Technologies,

Santa Clara, CA, USA). The GC separation was carried out using an Agilent VF5-MS GC column (30 m × 0.25 mm × 0.2 nm) in which helium was used as a carrier gas at a constant flow rate of 1.00 mL/min. In each run, a 1.00 µL sample was injected with a split ratio of

1:10, and the injector temperature was maintained at 250°C. The temperature profile was begun at 60°C before being increased to 110°C at a rate of 20°C/min; following this, it was increased to 240°C at a rate of 40°C/min before being held at this temperature for 5 min.

The electron impact (EI) ionization mode was used for the MS detection. The optimized mass spectrometry operating parameters were as follows: target total ion count,

11000 counts; maximum ionization time, 25000 µs; and pre-scan ionization time, 100 µs.

The mass spectra were obtained in the mass range of m/z 100–400. The following extracted

121

ion chromatograms of the TMSP ions (including the parent ion) were monitored:

16 16 18 TMSP O4-CH3 (299), parent ion + 2 m/z: TMSP O3 O-CH3 (301), and parent ion + 4

16 18 m/z: TMSP O2 O2-CH3 (303).

5.2.4 Isotope enrichment calculation

The chromatograph peaks were extracted using the accurate mass of each ion species (i.e., the extracted ion chromatograms), and the isotope enrichment achieved by the experiment was calculated using the following equation (6):

푖푠표푡표푝푒 푝푒푎푘 푠푖푔푛푎푙 푟푒푙푎푡푖푣푒 % 푖푠표푡표푝푒 = ×100 (1) 푠푖푔푛푎푙 표푓 푝푟푖푚푎푟푦 푖푠표푡표푝푒

In order to correctly account for the enrichment due to 18O from the data obtained from the LC-QTOF-MS analysis, the expected natural isotopic abundance was subtracted from the experimental abundance.

For the data obtained using GC-MS, the isotopic spillover was calculated; this was because the derivatization reagent TMSP has a high natural abundance of 29Si and 30Si, which can obscure the interpretation of the 18O incorporation results. This isotopic spillover was calculated as per the method described in a previously published paper (6). In order to correctly account for the enrichment due to 18O, the expected natural isotopic abundance was subtracted from the experimental abundance. The spillover from the species was subtracted from the 29Si and 30Si isotopes in addition to the expected natural abundance of

18O.

122

5.2.5 PIX with the Lon protease

The PIX reactions with the Lon protease were carried out to study reversibility of

ATP hydrolysis by LON proteases in the presence and in the absence of the substrate peptide as per the method described in previously published papers (13,15). All the samples were prepared in triplicate with two control experiments. The first control experiment contains only inorganic phosphate (phosphoric acid) without enzymes being added. In the second control experiment, ATP was incubated in the reaction mixture without enzymes being added. The samples were then prepared for each LC-QTOF-MS and GC-MS analysis separately, as described in the sub-sections above.

5.3. Results

5.3.1. Method development

5.3.1.1. LC-QTOF-MS method

In this work, a QTOF-MS) was used for analyte detection due to its high mass accuracy and resolution. Because the inorganic phosphate readily deprotonates in the acidic mobile phase (pH 3.65), ESI− was chosen as the means of sample introduction. In order to demonstrate the utility of the LC-MS method for the detection of PIX in enzyme-catalyzed reactions, phosphoric acid (100 nM) was used in this method. The predominant ion for the unlabeled sample in the negative ion mode occurs at m/z 96.9896 for the parent ion

− (H2PO4 ) species. Isotope peaks were also observed at parent ion + 2 m/z: 98.9864

16 18 − 16 18 − (H2P O3 O ) and parent ion + 4 m/z: 100.9972 (H2P O2 O2 ).

123

Due to the ionic nature of the analyte, which is difficult to retain in reverse phase

LC columns without the use of an ion pair agent, ion exchange chromatographic columns were considered for the analytical separation. Several columns, including Waters HILIC

(Waters Corporation, Milford, MA, USA) and Thermo Fisher Scientific Acclaim Trinity

P2, were tested. The Acclaim Trinity P2 column was chosen from these, because it provides hydrophilic interaction chromatography (HILIC) and a strong anion-exchange mixed- mode retention mechanism, and it exhibited more symmetrical and sharper peaks with better sensitivity and signal-to-noise ratio than the HILIC column.

The composition of the mobile phase used for the separation was optimized.

Acetonitrile was chosen as the organic solvent due to its greater solvent strength and the higher signal response of analytes in it during mass spectrometry than could be obtained using methanol. Another reason as to why it was preferred over methanol was due to the presence of a carboxylic acid group in the stationary phase of the column. The percentage content of acetonitrile in the mobile phase was also investigated. When a greater amount

(> 40%)of acetonitrile was used, the analyte was not retained in the column. Furthermore, because the pH value of the mobile phase influences the ionization and retention of phosphoric acid, the optimal pH of the mobile phase was investigated. In this study, the baseline resolution of the analyte was achieved with ammonium formate (10.0 mM) at pH

3.65. At this pH value, phosphoric acid (pKa1 = 2.15) was in its deprotonated form.

In this study, the optimized liquid chromatographic separation was obtained using a Thermo Fisher Scientific Acclaim Trinity P2 C18 column and a mobile phase consisting of 25% acetonitrile and 75% 10.0 mM ammonium formate pumped at a flow rate of 0.300 mL min−1. The total run time was 4.00 min, with a retention time of 1.2 min. Figure 5.2 A

124

and B show the representative total ion chromatogram and mass spectra of the sample, respectively. The relative isotope amounts were calculated using Eq. 1. A linear calibration curve was created with phosphoric acid standard solutions ranging from 10.0 to 1.00 × 103 nM (Fig. 5.3).

125

6.25E+04 A 5.00E+04 A

3.75E+04

2.50E+04

Intensity, Cps Intensity, 1.25E+04

0.00E+00 0 1 2 3 4 Time, Min 5.00E+04 B 96.9896 3.75E+04

2.50E+04 Counts

1.25E+04 98.9864 100.9972 0.00E+00 96 98 100 102 m/z

Figure 5.2 Determination of the 18O/16O ratios using LC-QTOF-MS. (a) Representative total ion chromatogram of inorganic phosphate and (b) representative mass spectral resolution of the phosphate isotopologues.

126

3.00E+05 y = 0.0216x + 0.0016 (R² = 0.9997)

2.00E+05

1.00E+05 Intensity, cps Intensity,

0.00E+00 0 500 1000 Concentrtion, nM

Figure 5.3 Calibration curve established using the phosphoric acid standard solutions from

10.0 to 1.00 × 103 nM using the LC-QTOF-MS method.

127

5.3.1.2. GC-MS method

The GC-MS analysis was carried out as per the method described in a previously published paper (13,15). The GC separation was carried out using an Agilent VF5-MS GC column. The isotope enrichment was analyzed using the extracted ion chromatogram (EIC) mode of the ion-trap mass spectrometer. Figure 5.4 shows the representative extracted ion chromatograms of the parent ion = TMSPO4-CH3 (299.0) and parent ion + 2 m/z =

16 18 TMSP O3 O-CH3 (301.0) from the GC-MS analysis.

128

1.50E+07 Ion 299.0 Ion 301.0 1.20E+07

9.00E+06

6.00E+06 Intensity, cps Intensity, 3.00E+06

0.00E+00 4 5 6 7 8 Time, Min

Figure 5.4 Representative selected ion chromatograms of TMSP used for the determination of the 18O/16O ratios in inorganic phosphate using GC-MS.

129

5.3.2 Application

5.3.2.1 PIX with the Lon protease

In order to demonstrate the applicability of the LC-MS method, we monitored the isotope enrichment in PIX reactions catalyzed by ATPase in the presence and absence of a peptide substrate

(13,15). In brief, in these reactions ADP is produced by ATP hydrolysis with the Lon protease, which is a rate-limiting step that releases inorganic phosphate as a by-product. In order to check the reversibility of the ATPase reactions in the presence and absence of this peptide substrate, the experiments were conducted in 18O-enriched water so that the number of labeled oxygen atoms incorporated in the resultant inorganic phosphate generated by the ATP hydrolysis could be monitored. Furthermore, in this study, inorganic phosphate was used as a positive control, and the reaction mixture without the enzyme was used as a negative control.

In the LC-QTOF-MS analysis, the lyophilized inorganic phosphate samples were prepared by being dissolved in deionized water. For the GC-MS analysis, the inorganic phosphate product was converted to a volatile derivative by the reaction with TMSDEA; the extent of the incorporation of 18O was thusly determined by GC-MS. In order to ascertain if the ATP hydrolysis was a reversible process, we determined the number of isotopically labeled oxygen atoms incorporated into the phosphate from an enriched reaction mixture by comparing it to the natural isotopic abundance. Furthermore, control experiments were performed by incubating ATP in the reaction mixture without enzymes being added, so that we could ensure that no labeled oxygen was incorporated due to the non-enzymatic hydrolysis during the time course of the experiment.

Additional isotopically labeled oxygen atoms will be detected if the phosphate combines with ADP in the reverse reaction and unlabeled oxygen is lost due to dehydration.

As a result, detection of more than one incorporation of 18O into the inorganic phosphate

130

reveals if the ATP hydrolysis is reversible at the ATPase sites. Figure 5.5 A and B show the results of the isotope enrichment experiments from both the LC-QTOF-MS and GC-

MS analyses, respectively. As can be seen in Figure 5.5, in the presence of the peptide

16 18 − substrate, we detected enrichment in the parent ion + 2 m/z (H2P O3 O ); namely, an

16 18 − enrichment in (H2P O3 O ) of 21 ± 3% over the expected natural abundance. In the absence of the peptide substrate, an enrichment of 2.5 ± 1% in the parent ion + 2 m/z

16 18 − (H2P O3 O ) was observed. The data from the LC-QTOF-MS was consistent with the

GC-MS data, which means that both methods could be used for such studies.

131

24 A

Isotope ,% enrichment Isotope 6

0 97 99 101 m/z

Control - Inorganic Phosphate Control - No enzyme Control- No substrate 24 Sample with substrate

B ,%

Isotope enrichment enrichment Isotope 6

0 299 301 m/z 303

Figure 5.5 Isotope enrichment calculated for the incorporation of 18O into the inorganic phosphate released from the hydrolysis of ATPase in the presence of isotopically enriched water. (a) Enrichment measured using the LC-QTOF-MS method and (b) enrichment measured using the GC-MS method.

132

In order to further analyze the usefulness of the method for studying the time course of peptide hydrolyses, we measured the isotope enrichment in the inorganic phosphate. The

PIX reactions with the Lon protease were carried out for 0, 5, 10, 30, and 60 min (as described in Section 2.5). As shown in Figure 5.6, no enrichment was observed in the control. Much less enrichment was observed in the reaction mixture that did not contain the peptide substrate than the one that did contain it.

133

30 Control - Inorgnic phoshate Control - No enzyme 25 Control - No substrate Sample - With substrate 20

Isotope enrchment, % enrchment, Isotope 5

0 0 10 20 30 40 50 60 Time, Min

Figure 5.6 Time study analysis of isotope enrichment for the incorporation of 18O into the inorganic phosphate from the hydrolysis of ATP by the Lon protease using the LC-QTOF-

MS method.

134

5.4 Discussion

A number of methods like GC-MS and NMR have been utilized for the analysis of

PIX rates in enzyme-catalyzed reactions. Although GC-MS methods are widely used for the analysis of PIX experiments, the main disadvantage of such methods is the need for the chemical derivatization of the samples so as to make them volatile; however, this is time- consuming (i.e., it takes 1 h to do). Furthermore, the derivatization reagent TMSP has a high natural abundance of 29Si and 30Si, which can obscure the interpretation of the results of the incorporation of 18O if left unchecked (14). In order to correctly account for the enrichment due to 18O, the expected isotopic abundance has to be subtracted from the experimental abundance (6). ESI-MS-based methods for the monitoring of 18O/16O exchange in metabolic oligophosphates, such as for the determination of the turnover of phosphoryl in nucleotide triphosphates, which occurs in many different metabolic processes, has been developed (16,17). No LC-QTOF-MS method, however, has been developed that would allow for the PIX rates to be monitored; such a method would allow one to determine the reversibility of ATP hydrolysis by Lon proteases. In this study, therefore, we explored LC-QTOF-MS as a tool for analyzing inorganic phosphate.

When compared the sample preparation required for LC-QTOF-MS analysis to

GC-MS analysis of the same sample is simple and efficient, and this technique could potentially be used to monitor the PIX rates in underivatized samples. Moreover, as no derivatization agent was used for the sample preparation, no interference was expected in the interpretation of the incorporation of 18O. Furthermore, the use of a LC column significantly reduced the matrix effect and made the detection method significantly more sensitive than the direct flow injection analysis of the sample by LC-QTOF-MS. This new

135

method is more sensitive than GC-MS, as its detection limit is 10.0 nM rather than 60.0 nM.

The methodology for monitoring PIX rates using LC-QTOF-MS in enzyme- catalyzed reactions was developed in conjunction with Lon protease. Because Lon protease is known to catalyze ATP hydrolysis, in the exchange reaction in the presence of 18O enriched water and in the incorporation of 18O into inorganic phosphate, this enzyme proved to be suitable for testing for PIX within inorganic phosphate using LC-QTOF-MS.

If ATP hydrolysis is reversible, it can result in the addition of multiple 18O. The experimental isotopic abundance was therefore compared to the expected molecular weight distribution from naturally occurring isotopes; the results of the enzymatic hydrolysis of

ATP in the presence of 18O-labeled water indicate that only one labeled oxygen was incorporated into any single phosphate; this was determined as there was no enrichment of

16 18 − any isotope except for the parent ion + 2 m/z (H2P O3 O ).

5.5 Conclusion

We have shown that the LC-QTOF-MS method that has been developed by this study can be used to measure the 18O content of underivatized samples. The method is fast, simple, and significantly more sensitive than a GC-MS analysis. Sample preparation is simple and needs no derivatization and no isotope spill over to account for. Also, the LC-

MS run time is shorter than the GC-MS method. This technique developed was used for the measurement of PIX rates in enzyme-catalyzed reactions so that the reversibility of the reactions could be monitored.

136

5.6 References

1. A.N. Singh, L.S. Hester, F.M. Raushel, Examination of the mechanism of sucrose

synthetase by positional isotope exchange, J. Biol. Chem. ,1987; 2554–2557.

2. F.C. Kokesh, Y. Kakuda, Evidence for intermediate formation in the mechanism of

potato starch -hosphorylase from exchange of the ester and phosphoryl oxygens of

alpha-D-glucopyranose 1-phosphate, Biochemistry., 1977; 2467–2473.

3. W. Von der Saal, P. M. Anderson, J. J. Villafranca, Mechanistic Investigations of

Escherichia coli Cytidine-5’- triphosphate synthetase. Detection of an intermediate

by positional isotope exchange experiments, J. Biol. Chem., 1985; 14993–14997.

4. I. A Rose, Positional isotope exchange studies of enzyme mechanisms; Advances

in Enzymology and Related Areas of Molecular Biology; John Wiley & Sons, Inc.,

1979; pp 361–395.

5. Y. Xiao, D. Rooker, Q. You, C. Free Meyers, P. Liu, IspG-catalyzed positional

isotopic exchange in methylerythritol cyclodiphosphate of the deoxyxylulose

phosphate pathway: mechanistic implications, Chembiochem., 2011; 527–530.

6. D.D. Hackney, K.E. Stempel, P.D. Boyer, Oxygen-18 probes of enzymic reactions

of phosphate compounds. Methods Enzymol., 1980; 60–83.

7. L.S. Hester, F.M. Raushel, Analysis of ping-pong reaction mechanisms by

positional isotope exchange, J. Biol. Chem.,1987; 12092–12095.

8. F.M. Raushel, L.J Garrard, A positional isotope exchange study of the

argininosuccinate lyase reaction. Biochemistry., 1984; 1791–1795.

137

9. P.P. Dzeja, K.T. Vitkevicius, M.M. Redfield, J.C. Burnett, A. Terzic, Adenylate

kinase-catalyzed phosphotransfer in the myocardium—increased contribution in

heart failure, Circ Res.,1999; 1137–1143.

10. R.J. Zeleznikar, R.A. Heyman, R.M. Graeff, T.F. Walseth, S.M. Dawis, E.A. Butz,

N.D. Goldberg, Evidence for compartmentalized adenylate kinase catalysis serving

a high energy phosphoryl transfer function in rat skeletal muscle, J Biol Chem.,

1990; 300–311

11. D. Pucar, P.P. Dzeja, P. Bast, N. Juranic, S. Macura, A. Terzic, Cellular energetics

in the preconditioned state—protective role for phosphotransfer reactions captured

by 18O-assisted 31P NMR, J. Biol. Chem., 2010; 44812–44819

12. D. Pucar, P.P. Dzeja, P. Bast, R.J. Gumina, C. Drahl, L. Lim, N. Juranic, S. Macura,

A. Terzic, Mapping hypoxia-induced bioenergetic rearrangements and metabolic

signaling by 18O-assisted 31P NMR and 1 H NMR spectroscopy. Mol Cell

Biochem., 2004; 281–289

13. J. Thomas, J. Fishovitz, I. Lee, Utilization of positional isotope exchange

experiments to evaluate reversibility of ATP hydrolysis catalyzed by Escherichia

coli Lon protease, Biochem. Cell Biol., 2010; 119–128.

14. D. Vineyard, J. Patterson-Ward, I. Lee, Single-turnover kinetic experiments

confirm the existence of high- and low-affinity ATPase sites in Escherichia coli lon

protease, Biochemistry., 2006; 4602–4610.

15. J. Fishovitz, Z. Sha, S. Chilakala, I. Cheng, Y. Xu, I. Lee, Utilization of mechanistic

enzymology to evaluate the significance of adp binding to human lon protease.

Front. Mol. Biosci., 2017; 00047.

138

16. R. Alvarez, L.A. Evans, P. Milham, M.A. Wilson, Analysis of oxygen-18 in

orthophosphate by electrospray ionisation mass spectrometry, Int J Mass

Spectrom., 2000; 177–186.

17. E. Nemutlu, N. Juranic, S. Zhang, L. Ward, T. Dutta, K. Nair, A. Terzic, S. Macura,

P. Dzeja, Electron spray ionization mass spectrometry and 2D P NMR for

monitoring O/O isotope exchange and turnover rates of metabolic oligophosphates.

Anal. Bioanal. Chem., 2012; 697–706.

139

Chapter VI

Quantifying Decitabine Incorporation and

Hypomethylation Effect of Decitabine in vitro and in vivo

Studies Using Quantitative LC-MS/MS

Abstract

The DNA hypomethylating drug decitabine and its pro-drug 5-azacytidine are the only two drugs approved for treatment of all subtypes of the myeloid malignancy myelodysplastic syndromes (MDS). The key to drug activity is incorporation into target cell DNA, however, a practical method to measure this incorporation is un-available. Here, we report a sensitive and specific LC-MS/MS method to simultaneously measure decitabine incorporation and DNA hypomethylation. A stable heavy isotope of 2'- deoxycytidine was used as internal standard and one-step multi-enzyme digestion released the DNA bound drug. Enzyme-released decitabine along with other mononucleosides were

140

separated by a reverse-phase C18 column and quantified by mass spectrometry using multiple-reaction-monitoring (MRM) mode, with a lower limit of quantitation at 1.00 nM.

In vitro studies demonstrated dosage and time-dependent incorporation of decitabine into myeloid leukemia cells that correlated with extent of DNA hypomethylation. When applied to clinical samples serially collected from MDS patients treated with decitabine, the method again demonstrated correlation between decitabine DNA-incorporation and DNA hypomethylation, suggesting resistance (failure to hypomethylate DNA) resulted from failure of administered drug to incorporate into DNA of target cells. This novel assay can therefore potentially provide insights into mechanisms underlying sensitivity versus resistance of malignant myeloid cells to mainstay treatment.

6.1 Introduction

Decitabine (5-aza-2’-deoxycytidine) is a nucleoside analog of 2'-deoxycytidine.

Decitabine was introduced clinically four decades ago and was approved for the treatment of patients with myelodysplastic syndrome (MDS) in 2006 in the USA (1-4). The only other drug approved to treat all subtypes of MDS is 5-azacytidine which is reduced to decitabine in cells. After parenteral administration, decitabine undergoes a three-step phosphorylation within cells into its active metabolite, decitabine triphosphate [first, to its monophosphate by deoxycytidine kinase (dCK); then to its diphosphate by deoxycytidine monophosphokinase; and finally to its triphosphate by nucleoside diphosphokinase] which is then directly incorporated by DNA polymerases into DNA during the S-phase of replication (5) (Figure 6.1). Decitabine triphosphate is the primary intracellular metabolite that has the antileukemic effect both in vivo and in vitro. Decitabine and its mono-

141

phosphate can be rapidly metabolized by cytidine deaminase (CDA) and deoxycytidine deaminase (DCTD) respectively into their inactive uridine derivatives, which could be causes of treatment resistance (6). After incorporation into DNA, decitabine inhibits DNA methylation by forming a covalent bond with DNA methyltransferase 1 (DNMT1).

DNMT1 is a corepressor which generates an epigenetic repression (‘off’) mark by the addition of a methyl group to the fifth carbon of 2-deoxycytosine within the 5′-cytosine- guanosine (CpG) dinucleotides of DNA. This epigenetic mark generated by DNMT1 is robustly linked with gene repression and aberrant repression of tumor suppressor genes in malignant transformation (7-9). Thus DNMT1-depletion by the DNA incorporated decitabine can reactivate tumor suppressor genes, leading to terminal differentiation or apoptosis of malignant cells (5, 10).

Target engagement is crucial for therapeutic efficacy of any drug. Since the incorporation of decitabine into DNA is essential for the anti-malignant effect, we postulate that quantitation of decitabine incorporation into cellular DNA can provide critical insights into sensitivity versus resistance of malignant cells to the drug; therefore, it could provide useful information for adjustment of the dosage regime to improve clinical outcomes.

Although decitabine incorporation into DNA has previously been measured using radioisotopic and liquid scintillation counting assays (11), these previous methods had limited sensitivity and used radioactive isotope labeled decitabine, which is not practical for use in clinical settings.

To address this unmet need, here we developed a sensitive LC-MS/MS method to simultaneously quantify DNA incorporation by decitabine and DNA methylation by measuring dG, dC, and 5mdC. The method was applied both in vitro and in vivo. The

142

results demonstrated that this method can readily quantify the incorporation of decitabine into DNA, enabling quantification of the variation in the incorporation and response to treatment in different models and in vivo. This assay could be a useful tool for the purpose of understanding treatment sensitivity versus resistance and provide crucial guidance towards an overall goal of individualizing and optimizing therapy with this exclusive class of agent.

143

Figure 6.1 Schematic representation of mechanism of decitabine incorporation into DNA

144

6.2 Experimental

6.2.1 Chemicals

Decitabine was obtained from Developmental Therapeutics Program of the

National Cancer Institute (Bethesda, MD, USA). Bis(2-hydroxyethyl)amino- tris(hydroxymethyl)methane (BIS-TRIS), deoxyribonuclease I type II (DNase I), nuclease

P1 (NP1), bovine alkaline phosphatase (ALP), 2′-deoxycytidine (dC), 2′-deoxyguanosine

(dG), 2′-deoxythymidime (dT), 2′-deoxyadenosine (dA) and formic acid were obtained from Sigma-Aldrich (St. Louis, MO, USA). Snake venom phosphodiesterase I (PDE I) was obtained from Worthington Biochemical (Lakewood, NJ, USA). 2'-Deoxycytidine

15 N3 (96-98%, a stable heavy isotope of 2’-deoxycytidine) was purchased from Synthèse

AptoChem (Montreal, Quebec, Canada). 5-methyl-2′-deoxycytidine (5mdC) was purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Sodium chloride, zinc chloride, phosphate buffered saline (PBS, 10X, pH 7.4), phenol saturated with Tris buffer

(pH 6.6), HPLC-grade of methanol, acetonitrile and chloroform were purchased from

Fisher Scientific (Fair Lawn, NJ, USA). 10% sodium dodecyl sulfate (SDS) solution was obtained from Bio-Rad Laboratories (Hercules, CA, USA). RPMI-1640 medium with L- glutamine was purchased from Mediatech (Manassas, VA, USA). Fetal bovine serum was purchased from GE Healthcare Life Sciences (Logan, UT, US). RiboShredder™ RNase

Blend was purchased from Epicentre (Madison, WI, USA). Proteinase K was ordered from

Qiagen (Valencia, CA, USA).

6.2.2 In vitro studies

145

Cell lines HL-60 and U937 were obtained from American Type Culture Collection

(Rockville, MD, USA). MOLM-13 cell line was obtained from Leibniz Institute DSMZ -

German Collection of Microorganisms and Cell Cultures (Braunschweig, Germany). HL-

60, U937 and MOLM-13 cell lines were cultured in RPMI-1640 medium supplemented with L-glutamine, 1% Penicillin-Streptomycin and 10% (v/v) fetal bovine serum in a

o humidified 5% CO2 incubator at 37 C. Decitabine solution was freshly prepared for each experiment by dissolving decitabine powder in PBS (1X, pH 7.4) at a concentration of 1.00 mM.

For each 1 × 106 cells of HL-60 and U937 cell lines, the dosages of decitabine added to the culture media are 0.00, 0.100, 0.500, 1.00, or 10.0 μM; and were treated for 24 h and

o 6 48 h at 37 C in a humidified 5% CO2 incubator. MOLM-13 cells (1× 10 cells) was treated with 0.500 μM of decitabine or an equal volume of PBS (1X, pH 7.4) as control for 0, 4,

o 8, and 24 h at 37 C in a humidified 5% CO2 incubator. After treatment, the cells were removed from the medium by centrifugation at 1,500 x g, 4 oC for 5 min; and washed twice with 5.00 mL of PBS (1X, pH 7.4) each. The cell pellets were collected and stored in -20 oC until DNA extraction.

6.2.3 In vivo animal study

The animal study for this work was approved by the Cleveland Clinic Institutional

Animal Care and Use Committees (IACUC). NSG mice were obtained from Jackson

Laboratory (Bar Harbor, ME, USA). Primary acute myelogenous leukemia (AML) cells from patients were transplanted by tail-vein injection (0.4 x 106 cells/mouse) into non-

146

irradiated 6-8 week old NSG mice (n = 4 per/group). Mice were anesthetized with isoflurane before transplantation. On day 15 of inoculation, animal groups were treated subcutaneously with vehicle (PBS 1X, pH 7.4) or decitabine 0.100 mg/kg for three consecutive days per week during the course of study. The control-group animals (median survival of 45 days) were euthanized by an IACUC approved method for signs of distress, and the treatment-group animals were continued with treatment for 90 days. At 90 day of the treatment, animals were euthanized by the IACUC approved method.

White blood cells (WBCs) were extracted from bone marrow using the procedure published in the reference (12). Extracted cells were washed twice with 10.0 mL PBS (1X, pH 7.4) each. In each wash step, the cells were gently vortexed in PBS (1 X, pH 7.4) for

1 min and then centrifuged down at 1,500 × g, 4 °C, for 10 min. After washing, the cell pellet was frozen at -20 oC until DNA extraction.

6.2.4 Clinical study

The five AML patients of this study were enrolled in the clinical trial

(NCT01165996) with written informed consent on a protocol approved by the Institutional

Review Board of Taussig Cancer Institute at the Cleveland Clinic. The decitabine regimen was 0.200 mg/kg administered subcutaneously two consecutive days per week with the post-treatment samples obtained after six weeks of this treatment. 10.0 mL of peripheral blood samples were collected from each patient in heparinized BD Vacutainer® tubes

(Becton, Dickinson and Company, Franklin Lakes, NJ, USA) before and at the end of 6th week of treatment. The blood samples were fractioned by the Ficoll-Paque method (13).

147

Briefly, 10.0 mL heparinized blood was layered on the top of 12.0 mL Ficoll-Paque Plus reagent (GE Healthcare Bio-Sciences, Piscataway, NJ, USA) in a 50.0-mL sterile polypropylene centrifuge tube (RNase/DNase free); then the tube was centrifuged at 300 × g, 4 °C, for 20 min. The peripheral blood mononuclear cells (PBMCs) at the interface between plasma and polymorphonuclear cells were collected and transferred into a clean

15.0-mL centrifuge tube and washed twice with 10.0 mL PBS (1X, pH 7.4) each. In each wash step, the PBMCs were gently vortexed in PBS (1X, pH 7.4) for 1 min and then centrifuged down at 4 °C and 1,500 × g for 10 min. After washing, the cell pellet was collected and stored in -20 oC until DNA extraction.

6.2.5 DNA isolation and hydrolysis

Each cell pellet collected from the in vitro, in vivo and clinical studies was lysed by mixing gently with 2.00 mL of TE buffer (containing 10.0 mM Tris and 1.00 mM EDTA at pH 8.0) and 240 μL of 10% SDS solution for 2 min. Then, the lysate was incubated with

25.0 μL of proteinase K (600 mAU/mL) for 1 h at 37 oC. After the incubation, the sample solution was transferred to a phase lock gel tube (5 Prime, Gaithersburg, MD, USA), and

DNA was extracted using standard phenol/chloroform extraction method (14). The DNA extracted was dissolved in 5.00 mM BIS-TRIS buffer (pH 7.0) to a concentration of 1.00 mg/mL measured by UV spectrophotometry. The co-extracted RNA was removed by adding 2.00 µL of RiboShredder™ RNase Blend (1 U/µL) to each 100 µL of DNA sample

(1.00 mg/mL). After a 30-min incubation at 37 °C, the sample was mixed with 1.00 mL of pre-chilled ethanol (-20 °C) and kept at -20 °C for overnight to precipitate the DNA in the sample. After centrifugation at 15,000 x g for 15 min, the supernatant was discarded,

148

and the DNA pellet was washed with 1.00 mL of pre-cooled 70% ethanol (-20 °C) twice.

Then, the DNA pellet was air-dried and reconstituted with 100 µL of 5.00 mM BIS-TRIS buffer (pH 7.0) for subsequent enzyme digestion.

DNA hydrolysis was performed by denaturing ds-DNA to ss-DNA in boiling water for 30 min followed by a one-step tetra-enzyme digestion (15). The tetra-enzyme cocktail was prepared by mixing the enzyme solutions as follows: 10.0 μL of DNase I (20,000

U/mL), 15.0 μL of NP1 (200 U/mL), 40.0 μL of PDE I (100 U/mL) and 0.50 μL of ALP

(40,000 U/mL). For each 50.0 μL of DNA (0.500 g/L for the measurement of DNA incorporated decitabine; 1.25 ng/L for the measurements of dC, dG, and 5mdC), 4.00 μL of the tetra-enzyme cocktail was added and mixed well; then, the mixture was placed in a

37 °C water bath for overnight incubation to secure a complete digestion of DNA to

15 mononucleosides. After incubation, 5.00 μL of 2'-deoxycytidine N3 (the internal standard, IS, 2.00 µM in the LC mobile phase) was added to a digested sample and vortex- mixed for 30 s. Each DNA digest was deproteinized by 450 L of acetonitrile, then centrifuged at 15,000 x g for 10 min. 450 L of supernatant was collected and evaporated to dryness at 30 °C for 60 min in a TurboVap® LV Evaporator (Zymark, Hopkinton, MA,

USA) under a pressurized stream of nitrogen gas (20 psi). The dried residue was reconstituted in 50.0 μL of the LC mobile phase and subjected to LC-MS/MS analysis.

6.2.6 LC-MS/MS system

The LC-MS/MS instrumentation used for this work consisted of a Prominence

UFLC system (Shimadzu, Columbia, MD, USA) for analyte separation and a QTRAP 5500

149

tandem mass spectrometer (AB Sciex, Foster City, CA, USA) for quantitation. The system was controlled by AB Sciex Analyst® software (version 1.6.1) for its operation, data acquisition, and processing. The UFLC system included a system controller (CBM-20A), two binary pumps (LC-20AD), a temperature-controlled autosampler (SIL 20AHT) and an online degasser (DGU20A3).

6.2.7 Quantitation of decitabine, dG, dC, and 5mdC

The LC separation was carried out on a Hypersil Gold aQ C18 column (50 x 2.1 mm, 3 µm) (Thermo Scientific, Waltham, MA, USA) with a Hypersil Gold aQ C18 guard column (10 × 2.1 mm, 3 μm) using the LC mobile phase consisting of 0.1% formic acid aqueous solution and methanol (87.5:12.5, v/v) at a flow rate of 0.300 mL/min. For each analysis, 10.0 μL of sample was injected into the system by autosampler set at 4°C. The

LC elute was introduced into the mass spectrometer operated under the positive turbo-ion spray ionization mode. The optimized instrument parameters were as follows: curtain

(CUR), 40 psi; collision activated dissociation (CAD) gas, medium; nebulizer gas (GS1),

40 psi; turbo heater gas (GS2), 45 psi; turbo ion spray voltage (IS), +5200 V; source temperature (TEM), 300 °C; declustering potential (DP), 35 V; entrance potential (EP), 7

V; collision energy (CE), 15 V; collision cell exit potential (CXP), 18 V; mass resolutions

(Q1 and Q3), 1 unit. The multiple-reaction-monitoring (MRM) mode was used for analyte quantitation.

Mixed calibrators of decitabine, dC, dG and 5mdC (1.00, 4.00, 20.0, 40.0, 200, 400 and 2.00 x 103 nM) together with single- and double- blanks, and mixed QCs (3.00, 70.0

150

and 1.60 x 103 nM) were prepared in 5.00 mM BIS-TRIS buffer (pH 7.0) and subjected to the same sample preparation procedure as DNA samples. The linear calibration plots for decitabine, dC, dG, and 5mdC were constructed using peak area ratios (y) of each analyte to the IS versus the concentrations of calibrators (x) with 1/x weighting, and the least squares linear regression equations were obtained as the calibration equations for individual analytes.

The concentrations of decitabine, dC, dG, and 5mdC in each unknown sample were back calculated by the calibration equations using the peak-area ratios of each analyte in the unknown sample to that of the IS. The accurate mass concentration of DNA was determined using the measured molar concentration of dG by the following equation:

[DNA] (mg/L) = [dG] X 618 (g/mol)/ 0.41 (Equation 1) where [dG] is the measured concentration of dG in mM; 618 (g/mol) is the molar mass of

G/C pair, and 0.41 is the percentage of G/C pair in human DNA (16). The amount of DNA incorporated decitabine was expressed as pmol of decitabine per g of DNA. The percent methylation was calculated using the following equation (17, 18):

% methylation = [5mdC/(dC + 5mdC)] × 100% (Equation 2)

6.3 Results

6.3.1 Enzymatic hydrolysis of DNA

In this work, the DNA samples were hydrolyzed by a one-step tetra-enzyme reaction

(Figure 6.2). This one-step enzyme reaction had been optimized and validated previously

151

to retain 100% digestion efficiency in comparison to an optimized stepwise enzyme reaction (15). Among the four enzymes employed, DNase I is an endonuclease that splits phosphodiester bonds of DNA and yields oligonucleotides with a free 3’-end hydroxyl group and a free 5’-end phosphate group; PDE I is a 3’ to 5’ exonuclease that successively hydrolyzes an oligonucleotide from 5’-end to 3’-end and produce deoxynucleoside 5'- phosphate; NP1 is a 5’ to 3’ exonuclease that acts in opposite direction to PDE I, and completely hydrolyzes an oligonucleotide from 3’-end to 5’-end to produce deoxynucleoside 5'-phosphate; and ALP is a hydrolase that hydrolyzes phosphate groups of deoxynucleotides to deoxynucleosides. Since the electrospray ionization tandem mass spectrometer (ESI-MS/MS) used for this work had much lower limits of quantitation for deoxynucleosides than those of deoxynucleotides; therefore, deoxynucleosides were the sought-after products for the enzyme digestion.

152

Figure 6.2 Schematic representation of DNA hydrolysis by a one-step tetra-enzyme digestion system that includes DNase 1, NP 1, PDE 1, and ALP.

153

6.3.2 Measurement of decitabine, dG, dC and 5mdC by LC-MS/MS

The mass spectrometer was tuned by infusion of a mixture of decitabine, dC, 5mdC, dG and IS at each concentration of 500 ng/mL in the mobile phase to optimize both compound- dependent and source-dependent parameters. Since these deoxynucleoside were more easily to form protonated ions than deprotonated species in the acidic mobile phase, the positive ESI was used for the analytes’ identification and quantification. As shown in

Figure 6.3A, 6.3C, 6.3E, 6.3G and 6.3I, the predominated precursor ions of decitabine, dC,

5mdC, dG and IS were at m/z 229.1, 228.1, 242.0, 268.0 and 231.0, respectively. These precursor ions were broken into product ions by collision with nitrogen gas and produced the predominant product ions at m/z 113.1, 112.1, 126.0, 152.0 and 115.0, respectively

(Figure 6.3B, 6.3D, 6.3F, 6.3H and 6.3J). Therefore, the mass transitions of m/z 229.1 >

113.1, 228.1 > 112.1, 242.0 > 126.0, 268.0 > 152.0 and 231.0 > 115.0 for were chosen for the selective quantitation of decitabine, dC, 5mdC, dG and IS by MRM mode.

As shown in Figure 6.4, decitabine, dC, 5mdC, dG and IS were separated and detected by our LC-MS/MS method in less than 6 min. This method has a lower limit of quantification (LLOQ) of 1.00 nM for each analyte, which was defined by the lowest calibrator of a calibration plot; and a linear calibration range from LLOQ to 2.00 x 103 nM

(Figure 6.5).

154

Figure 6.3 Representative mass spectra of precursor and product ions of decitabine, dC, 5mdC, dG and IS. A) the precursor ion of decitabine; B) the major product ion of decitabine; C) the precursor ion of dC; D) the major product ion of dC; E) the precursor ion of 5mdC; F) the major product ion of 5mdC; G) the precursor ion of dG; H) the major product ion of dG; H) the precursor ion of IS; I) the major product ion of IS.

155

Figure 6.4 Representative MRM chromatograms of decitabine, dC, 5mdC, dG, and IS.

156

Figure 6.5 Calibration plots of decitabine, dC, 5mdC, and dG.

157

Validation of the method was done according to the US-FDA guidance for industry on bioanalytical method validation (19). The inter-day and intra-day accuracy and precision of the three mixed QC concentrations were ≤ ±6% and ≤ 8%, respectively (Table

6.1). The accuracy and precision of mixed calibrators over six calibration plots were ≤

±6% and ≤ 11%, respectively (Table 6.2). These data were well within the acceptance ranges of US-FDA.

The studies of matrix effect and analyte recovery on all analytes were conducted using mixed QC at three concentrations (Table 6.3). For all analytes, the absolute matrix factors ranged 0.933-1.03 and the IS normalized matrix factors ranged 0.955-1.03; and the absolute recoveries ranged 90.3-100% and the IS normalized recoveries ranged 96-102%.

These data indicated there was neither significant matrix effect from enzyme digestion buffer nor significant difference in recoveries of analytes between enzyme digestion buffer and the mobile phase of LC separation, and the preparation of mixed calibrators of decitabine, dC, dG and 5mdC in either enzyme digestion buffer or the LC mobile phase makes no difference in the analytical results. Therefore, the mixed calibrators of dC, dG and 5mdC prepared in the mobile phase were adopted for the method.

158

Table 6.1. Accuracy and precision of decitabine, dC, 5mdC and dG calibrators over six validation batches in digestion matrix

Nominal [Decitabine] (nM) [dC] (nM) [5mdC] (nM) [dG] (nM) [Analyte] [Measured ± [Measured ± [Measured ± %REb CVc [Measured ± SDa] %REb CVc REb CVc %REb CVc (nM) SDa] SDa] SDa] 1.00 0.98 ± 0.09 -2 11 0.97 ± 0.07 -3 7 1.02 ± 0.06 2 6 0.97 ± 0.07 -3 7 4.00 4.1 ± 0.2 3 4 4.1 ± 0.2 2 4 3.9 ± 0.2 3 5 4.1 ± 0.2 3 5 20.0 20.7 ± 0.2 4 1 19.8 ± 0.5 -1 3 22 ± 1 10 5 19.6 ± 0.2 -2 1 40.0 40 ± 2 1 4 41 ± 3 3 7 42 ± 3 6 6 39 ± 1 -3 2 200 205 ± 4 3 2 196 ± 2 -2 1 194 ± 3 -3 8 203 ± 8 2 4 400 386 ± 1 -4 3 377 ± 2 --6 5 407 ± 1 2 2 412 ± 1 3 3

(1.97± 0.02) x (1.98 ± 0.02) x (1.99 ± 0.02) x (2.02 ± 0.02) 2.00 x 103 -1 1 -1 1 -0.4 3 1 1 103 103 103 x103

a Each measured value was calculated based on six measurements from different days. b %RE = {(measured [analyte] − nominal [analyte])/nominal [analyte]} x 100%. c CV = (standard deviation/mean value) x100%

159

Table 6.2 Intra- and inter-assay precision and accuracy

Intra-assaya Inter-assayb Analyte Nominal [QC] (nM) [Measured QC ± SD] (nM) %RE CV [Measured QC ± SD] (nM) %RE CV

3.00 (LQC) 3.2 ± 0.1 5 3 3.2 ± 0.1 6 4

70.0 (MQC) 71 ± 2 3 2 71 ± 1 3 2 Decitabine 1.60 x 103 (HQC) (1.57 ± 0.02) x103 -2 1 1.56 ± 0.01) x 103 -2 2

3.00 (LQC) 2.9 ± 0.1 -2 4 2.9 ± 0.2 -1 6

70.0 (MQC) 69 ± 3 2 4 72 ± 2 4 3 dC 1.60 x 103 (HQC) (1.63 ± 0.02) x103 2 1 (1.65 ± 0.02) x103 3 2

3.00 (LQC) 3.1± 0.1 5 4 3.2 ± 0.3 6 8

70.0 (MQC) 71 ± 4 2 6 72 ± 3 3 4 5mdC 1.60 x 103 (HQC) (1.58 ± 0.02) x103 -1 1 (1.58 ± 0.02) x103 -1 1

3.00 (LQC) 3.2 ± 0.1 6 4 3.1 ± 0.2 3 6

70.0 (MQC) 66 ± 3 -5 4 67 ± 2 -3 3 dG 1.60 x 103 (HQC) (1.61 ± 0.02) x103 1 1 (1.61 ± 0.02) x103 1 2

a Intra-assay precision and accuracy were assessed by five replicate measurements of individual QC at each concentration. b Inter-assay precision and accuracy were assessed by five parallel measurements of five identical QCs at each concentration.

160

Table 6.3 Matrix effect and recovery of decitabine from enzyme digestion matrix (n = 5)

[Decitabine] (nM) [dC] (nM) [5mdC] (nM) [dG] (nM)

1.60 x 3.00 70.0 70.0 1.60 x 103 3.00 70.0 1.60 x 103 3.00 70.0 1.60 x 103 103 3.00 (LQC) (LQC) (MQC) (MQC) (HQC) (LQC) (MQC) (HQC) (LQC) (MQC) (HQC) (HQC)

a 0.959 ± 0.972 ± 1.01 ± 0.964 ± 0.976 ± 1.01 ± 0.933 ± 0.951 ± 0.968 ± 0.943 ± 0.961 ± 0.990 ± MFAna ± SD 0.001 0.001 0.00 0.002 0.002 0.03 0.003 0.006 0.003 0.003 0.007 0.002

b 0.977 ± 0.981 ± 0.98 ± 0.977 ± 0.981 ± 0.98 ± 0.977 ± 0.981 ± 0.977 ± 0.981 ± MFIS ± SD 0.98 ± 0.01 0.98 ± 0.01 0.002 0.001 0.01 0.002 0.001 0.01 0.002 0.001 0.002 0.001

IS 0.982 ± 0.991 ± 1.03 ± 0.987 ± 0.995 ± 1.03 ± 0.955 ± 0.970 ± 0.988 ± 0.965 ± 0.980 ± Normalized 1.01 ± 0.01 0.002 0.001 0.01 0.003 0.002 0.03 0.004 0.006 0.01 0.004 0.007 MFc ± SD

d RAna ± SD 90.3 ± 91.2 ± 0.2 97± 1 100 ± 1 92.6 ± 0.3 98± 2 100 ± 2 95.± 3 98 ± 1 90.7 ± 0.5 96± 4 99 ± 1 (%) 0.4

e RIS ± SD (%) 91± 2 99± 3 99 ± 2 91± 2 99± 3 99 ± 2 91± 2 99± 3 99 ± 2 91± 2 99± 3 99 ± 2

IS 1.00± 0.98 ± 1.01 ± 0.99 ± 1.01 ± 0.99± 0.97 ± Normalized 1.02± 0.0 0.96 ± 0.04 0.99 ± 0.02 1.00± 0.02 1.00 ± 0.02 0.02 0.03 0.02 0.04 0.03 0.02 0.05 Rf ± SD

a MFAna = (mean peak area of analyte in extracted enzyme digestion matrix)/ (mean peak area of analyte in the mobile phase). b MFIS = (mean peak area of IS in extracted enzyme digestion matrix)/ (mean peak area of IS in the mobile phase). c IS Normalized MF = MFAna/MFIS. d RAna = [(mean peak area of analyte in enzyme digestion matrix)/ (mean peak area of analyte in extracted enzyme digestion matrix)] x 100%. e RIS = [(mean peak area of IS in enzyme digestion matrix)/ (mean peak area of IS in extracted enzyme digestion matrix)] x 100%. f IS Normalized R = (RAna/ RIS) x 100%.

161

6.3.3 Decitabine incorporation vs. hypomethylation of DNA in leukemia cell lines

The LC-MS/MS method developed was first tested by human leukemia cell lines

HL-60 and U937 for quantitation of decitabine incorporation in cellular DNA and cell responsiveness to decitabine treatment (i.e., decrease of DNA methylation or DNA hypomethylation). In this study, the effects of drug dose (0.000, 0.100, 0.500, 1.00 and

10.0 μM decitabine) and treatment times ((24- and 48-h incubation) were investigated. The cellular DNAs were isolated from samples of each dose and time point. The enzyme- released decitabine, dC, dG and 5mdC from the cellular DNA were determined by the method developed and percent methylations were calculated.

As shown in Figure 6.6 and summarized in Table 6.4, the amounts of decitabine incorporated in HL-60 and U937 DNAs were directly proportional to the drug dose and the treatment time. The higher dose and the longer treatment time produced larger amounts of decitabine incorporation in DNA and greater degree of DNA hypomethylation. Also, differential sensitivities to decitabine treatment were observed between the two cell lines.

HL-60 cells were more sensitive to the drug treatment in comparison to U937 with larger decitabine incorporation and greater degree of DNA hypomehtylation, which suggested that for U937 cell line, it may need higher decitabine exposure (either dose or time) than

HL-60 cell line to reach the same drug effect.

162

Figure 6.6 Decitabine incorporation correlates with DNA demethylation in HL-60 and

U-937 cells. A) 24-h treatment, and B) 48-h treatment.

163

Table 6.4 The effects of decitabine dosage and treatment time on decitabine incorporation in DNA and DNA hypomethylation

24-h Study 48-h Study

Decitabine incorporation DNA methylation Decitabine incorporation DNA methylation Decitabine (fmol/µg DNA) (fmol/µg DNA) dose (µM)

HL-60 U937 HL-60 U937 HL-60 U937 HL-60 U937

0.000 0.000 0.000 4.86 4.89 0.000 0.000 4.86 4.89 0.100 88.0 46.0 4.21 4.47 96.0 65.0 4.10 4.23 0.500 120 64.0 3.66 4.16 132 95.0 3.45 3.99 1.00 133 94.0 3.16 3.45 155 114 2.77 3.06 10.0 136 96.0 2.88 3.25 172 146 2.04 2.70

164

Decitabine incorporation and hypomethylation effect in a different human leukemia cell line MOLM-13 were also investigated using both a single dose of decitabine (0.500

μM) for the treatment group and an equal volume of vehicle (PBS 1X, pH 7.4) for the control group at various time points (0, 4, 8, and 24 h). As shown in Figure 6.7, the profiles of drug effect in the cell line could be monitored by the time course, and the maximum amount of drug incorporation and hypomethylation effect were observed at 8 h in MOLM-

13 cells.

165

Figure 6.7 Time course study of decitabine incorporation and DNA demethylation in

MOLM-13 cells. A) Decitabine incorporation profiles, and B) DNA demethylation profiles.

166

6.3.4 Decitabine incorporation vs. hypomethylation of DNA in animal study The method developed was further tested by WBCs extracted from bone marrow of NSG mice inoculated with human primary AML cells for quantitative measurement of

DNA-incorporated decitabine and DNA hypomethylation. The experimental details were described in “In vivo animal study” section under “Materials and Methods”. As shown in

Figure 6.8A, decitabine incorporation in four genetically identical mice after the drug treatment ranged 18.4 to 25.9 fmol per µg DNA, and such treatment significantly reduced

DNA methylation in all mice (Figure 6.8B). Because of the responsiveness of the mice to decitabine treatment, the median survival time of the control mice (45 days) was significantly extended for the treated mice (90 days).

167

Figure 6.8 Decitabine incorporation and DNA demethylation in vivo animal study. A)

Decitabine incorporation in the controls and decitabine-treated mice (n = 4), and B) DNA demethylation in the controls and decitabine-treated mice.

168

6.3.5 Decitabine incorporation vs. hypomethylation of DNA in clinical study

Finally, the method developed was applied to PBMCs isolated from peripheral blood samples of five AML patients before and after decitabine treatments. The experimental details were described in “Clinical study” section under “Materials and

Methods”. Among the five patients, four responded well to the decitabine therapy, and one did not. The molecular mechanism of decitabine action and drug resistance in these patients would be depicted by Figure 6.9. For patients #1 to #4 who were sensitive to decitabine treatment, decitabine incorporation and DNA hypomethylation were large and significant (Figures 6.9A and 6.9B); whereas for patient #5 who was resistant to the drug, there were minimum decitabine incorporation (Figure 6.9C), and even higher DNA methylation in posttreatment (Figure 6.9D) suggesting that decitabine incorporation and

DNA methylation data may be used to access patients’ responses to decitabine treatment.

169

Figure 6.9 Decitabine incorporation and DNA demethylation in clinical study. A)

Decitabine incorporation in AML patients who were sensitive to decitabine treatment (n =

4), B) DNA demethylation in AML patients who responded to decitabine treatment (n =

4), C) Decitabine incorporation in AML patient who did not respond to decitabine treatment, and D) DNA demethylation in AML patient who did not respond to decitabine treatment.

170

6.4 Discussion

Target engagement is necessary for any drug to work, and the measurement of target engagement of decitabine (i.e., decitabine incorporation and DNA hypomethylation) is the first step to understand patients’ sensitivity and resistance in decitabine therapy; however, a practical method for this purpose was lacking. The method we developed is practical and very sensitive and specific for measurement of decitabine, 5mdC, and dG, as no interferences from endogenous molecules were observed. Furthermore, this method has been fully validated in enzyme digestion matrix and essentially met the validation criteria established by US-FDA. Such method may help us not only to understand drug action and resistance mechanisms of decitabine, but also to design more effective dosing regimens for patients.

In the in vitro dose and time course studies, no cell death was observed during the treatment and within this range (0.100 to 10.0 μM), decitabine incorporation into DNA of

HL-60 and U937 was found to be dose and time dependent. This may be due to differences in expression of the transporters and enzymes that are in the pathway of decitabine uptake into cells and its conversion into decitabine tri-phosphate, e.g., CDA, ENT1 and DCK (20-

22) (DCK converts decitabine to its monophosphate, and is the rate-limiting enzyme in its conversion to decitabine tri-phosphate, and CDA deaminates decitabine to its uridine counterpart that cannot deplete DNMT1). In the time course study of decitabine incorporation in DNA of MOLM-13 cells, the decitabine incorporation profile that was deviated from a steady-state profile might cause by the decomposition of decitabine in cell culture media. Although decitabine incorporation in DNA and DNMT1 depletion is known as S-phase specific, decitabine enters and accumulates in the cell during the G-phase of

171

cell cycle where decitabine is converted to its mono, di and tri-phosphate by DCK.

Decitabine deamination to its uridine counterpart by CDK in cells limits its availability to incorporation in the DNA (21), which can be a cause for the deviation in the time profile in MOLM-13 cell studies.

Similarly, the differential responses to decitabine treatment in vivo can also result from differential expression of CDA, ENT1 and DCK enzymes at the organism or patient level and within target cells (23-25). In the case of patient drug resistance, the decitabine incorporation was about 9% (3 fmol/g DNA) (Figure 6.9C) in comparison to the drug responsive group (33 ± 11 fmol/g DNA) (Figure 6.9A); furthermore, DNA hypomethylation was not observed - instead DNA methylation increased from 4.22%

(pretreatment) to 5.60% (posttreatment) (5, 10).

In sum, quantification of decitabine incorporation rates and DNA hypomethylation effects can potentially identify at an early point patients with disease that is non-responsive to their decitabine dosing regimens, to guide dosage adjustment or selection of alternative therapies. Further investigation in prospective clinical trials of decitabine incorporation as a pharmacodynamic endpoint that can guide individualization and optimization of therapy for MDS and acute myeloid leukemia patients, is thus warranted.

6.5 CONCLUSION

We have developed a sensitive and specific LC-MS/MS method for quantitative determination of decitabine DNA incorporation and hypomethylation effect both in vitro and in vivo. This method is practical and can be easily adopted in preclinical and clinical

172

studies of decitabine. The preliminary data from decitabine-treated AML patients show that this method not only allows us to quantitatively measure the pharmacokinetic endpoints of decitabine but also provides supporting evidence for drug action mechanisms on patients’ sensitivity and resistance in decitabine therapy. It may be useful for determining suitable decitabine dosing regimens for individual patients to achieve desirable therapy outcomes.

173

6.6 References

1. Saba, H. I. Decitabine in the treatment of myelodysplastic syndromes. Ther Clin

Risk Manag., 2007; 3(5): 807–817.

2. Saunthararajah Y, Sekeres M, Advani A, Mahfouz R, Durkin L, Radivoyevitch T,

Englehaupt R, Juersivich J, Cooper K, Husseinzadeh H, et al. Evaluation of

noncytotoxic DNMT1-depleting therapy in patients with myelodysplastic

syndromes. J Clin Invest., 2015; 125(3):1043-55.

3. McKeage, K. Croom, K.F. Decitabine: In myelodysplastic syndromes. Drugs.,

2006; 66, 951-958.

4. Kantarjian, H., Issa, JP., Rosenfeld, CS., Bennett, JM., Albitar, M., DiPersio,

J., Klimek, V., Slack, J., Saba, H., et al. Decitabine improves patient outcomes in

myelodysplastic syndromes - Resuits of a Phase III randomized study. Cancer.,

2006; 106(8):1794–1803.

5. Momparler, R. L. Pharmacology of 5-Aza-2′-deoxycytidine (decitabine). Semin.

Hematol., 2005; 42, S9-S16.

6. Taichun, Q., Ryan, C., Samih El, A., Jaroslav, J., Xiaodan, W., Jiali, S., Jingmin,

S., Rong, H., Kantarjian, H., and Jean-Pierre J. Issa, et al. Mechanisms of

Resistance to Decitabine in the Myelodysplastic Syndrome., PLOS ONE., 2011;

6(8): e23372.

7. Jones, P. A. and Baylin, S. B. The fundamental role of epigenetic events in cancer.

Nature Reviews Genetics., 2002; 3, 415-428.

8. Herman, J. G. and Baylin, S. B. Gene silencing in cancer in association with

promoter hypermethylation. N. Engl. J. Med., 2003; 349(21):2042-54.

174

9. Baylin, S. B. Mechanisms underlying epigenetically mediated gene silencing in

cancer. Semin. Cancer Biol.,2002; 12, 331-337.

10. Christman, J. K. 5-Azacytidine and 5-aza-2'-deoxycytidine as inhibitors of DNA

methylation: mechanistic studies and their implications for cancer therapy.

Oncogene., 2002; 21, 5483-5495.

11. Öz, S., Raddatz, G., Rius, M., Blagitko-Dorfs, N., Lübbert, M., Maercker, C. and

Lyko, F. Quantitative determination of decitabine incorporation into DNA and its

effect on mutation rates in human cancer cells . Nucleic Acids Res., 2014;

29;42(19).

12. Swamydas, M.,and Lionakis, M. S. Isolation, Purification and Labeling of Mouse

Bone Marrow Neutrophils for Functional Studies and Adoptive Transfer

Experiments. Journal of Visualized Experiments: JoVE, 2013; (77), 50586.

13. Kanof, ME., Smith, PD. and Zola H. Isolation of whole mononuclear cells from

peripheral blood and cord blood. Curr Protoc Immunol. 2001; Chapter 7: Unit 7.1.

14. Sambrook, J. and Russell DW. Purification of nucleic acids by extraction with

phenol: chloroform. Cold Spring Harb Protoc. 2006; 4455.

15. Li, L. Development of quantitative LC-MS/MS methods for the pharmacological

studies of anti-cancer drugs. Ph.D. Thesis; 2011; Cleveland State University,

Cleveland, OH., 53-67.

16. Alexander, E. and Vinogradov. Measurement by Flow Cytometry of Genomic

AT/GC Ratio and Genome Size. Cytometry., 1994; 6:34-40.

17. Liu, Z., Liu, S.Xie, Z., Blum, W., Perrotti, D., Paschka, P., Klisovic, R., Byrd, J.,

Chan, K. K. and Marcucci, G. Characterization of in vitro and in vivo

175

hypomethylating effects of decitabine in acute myeloid leukemia by a rapid,

specific and sensitive LC-MS/MS method . Nucleic Acids Res., 2007; 35(5): e31.

18. Liu, Z., Wu, J., Xie, Z., Liu, S., Fan-Havard, P., Huang, T. H., Plass, C.,Marcucci,

G. and Chan, K. Quantification of Regional DNA Methylation by Liquid

Chromatography/Tandem Mass Spectrometry. Anal Biochem., 2009;

15;391(2):106-13.

19. U.S. Food and Drug Administration (FDA), Center for Drug Evaluation and

Research (CDER), Guidance for Industry: Bioanalytical Method Validation, 2001;

available at:

http://www.fda.gov/downloads/Drugs/guidanceComplianceRegulatoryInformatio

n/Guidances/UCM070107.pdf

20. Hummel-Eisenbeiss, J., Hascher, A., Hals, P. A., Sandvold, M. L., Muller-Tidow,

C., Lyko, F. and Rius, M. The Role of Human Equilibrative Nucleoside

Transporter 1 on the Cellular Transport of the DNA Methyltransferase Inhibitors

5-Azacytidine and CP-4200 in Human Leukemia Cells. Mol. Pharmacol., 2013; 84,

438-450.

21. Saunthararajah, Y. Key clinical observations after 5-azacytidine and decitabine

treatment of myelodysplastic syndromes suggest practical solutions for better

outcomes. Hematology-American Society of Hematology Education Program.,

2013; 511-521.

22. Liu, Z., Byrd, J.C., Chan, K.K., Xiao, J, Marcucci, G and Grever, M.

Characterization of decomposition products and preclinical and low dose clinical

pharmacokinetics of decitabine (5-aza-2'-deoxycytidine) by a new liquid

176

chromatography/tandem mass spectrometry quantification method. Rapid

Communications in Mass Spectrometry., 2006; 20, 1117-1126.

23. Saunthararajah, Y., Ebrahem, Q., Negrotto, S., Mahfouz, R., Link, K., Hu, Z., Gu,

X.,Advani, A., Kalaycio, M., and Copelan, E.; et al. P53 Independent Epigenetic-

Differentiation Treatment in Xenotransplant Models of Acute Myeloid Leukemia.

Leukemia., 2011; 25(11):1739-50.

24. Ebrahem, Q., Mahfouz, R., Ng, K. P. and Saunthararajah, Y. High cytidine

deaminase expression in the liver provides sanctuary for cancer cells from

decitabine treatment effects. Oncotarget., 2011; 3(10):1137-45.

25. Blum, W., Klisovic, RB., Hackanson, B., Liu, Z., Liu, S. and Devine H, et al. Phase

I study of decitabine alone or in combination with valproic acid in acute myeloid

leukemia. J Clin Oncol., 2007; 25(25):3884–3891.

177