Development of Molecular Probes for Biomedical Imaging of Cancer and Neurological Disease

TITLE PAGE

DEVELOPMENT OF MOLECULAR PROBES FOR BIOMEDICAL IMAGING OF CANCER AND NEUROLOGICAL DISEASE

ALLISON G. CONDIE

Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Thesis Advisor: Yanming Wang Co-advisor: Stanton L. Gerson

Department of Chemistry

CASE WESTERN RESERVE UNIVERSITY

August, 2015

COMMITTEE APPROVAL

CASE WESTERN RESERVE UNIVERSITY SCHOOL OF GRADUATE STUDIES

We hereby approve the thesis/dissertation of

Allison G. Condie

candidate for the degree of Doctor of Philosophy.*

Committee Chair Robert G. Salomon

Committee Member Yanming Wang

Committee Member Stanton Gerson

Committee Member Clemens Burda

Committee Member Mary Barkley

Date of Defense May 1, 2015

*We also certify that written approval has been obtained for any proprietary material contained therein.

Page | ii

TABLE OF CONTENTS

Title Page ...... i

Committee Approval ...... ii

Table of Contents ...... iii

List of Tables ...... ix

List of Figures ...... x

List of Schemes ...... xxi

Acknowledgements ...... xxii

Symbols and Abbreviations ...... xxiii

Abstract ...... xxvii

Chapter 1. Introduction ...... 1

1.1 Review of molecular imaging ...... 1

1.1.1 Overview of molecular imaging ...... 1

1.1.2 Magnetic resonance imaging ...... 1

1.1.3 Positron emission tomography ...... 3

1.1.4 Optical imaging ...... 5

1.2 Review of myelin imaging ...... 8

1.2.1 Motivation and basis for disease ...... 8

1.2.2 Histological or in vitro imaging agents ...... 9

Page | iii

1.2.3 In vivo imaging agents ...... 10

1.3 Review of cancer imaging ...... 11

1.3.1 Motivation and basis for disease ...... 11

1.3.2 DNA damage and repair ...... 12

1.3.3 BER and AP sites ...... 13

1.3.4 Existing agents for imaging and detection ...... 15

Chapter 2. Myelin Imaging with CIC ...... 16

2.1 Hypothesis and methodology ...... 16

2.2 CIC synthesis and fluorescent characterization ...... 17

2.3 In vitro and ex vivo histology ...... 21

2.4 In vivo visualization of CNS nerves ...... 24

2.5 Conclusions ...... 30

2.6 Materials and methods ...... 31

2.6.1 General methods ...... 31

2.6.2 Synthesis ...... 32

2.6.3 Fluorescence characterization ...... 38

2.6.4 General animal methods ...... 40

2.6.5 Tissue staining ...... 40

2.6.6 Microscopy ...... 42

Chapter 3. Probes for Cancer Imaging Through AP Site Detection ...... 44

Page | iv

3.1 Hypothesis and methodology ...... 44

3.2 Synthesis of compounds for AP site detection ...... 45

3.3 Fluorescent characterization of probes ...... 51

3.4 Discussion and conclusions ...... 53

3.5 Materials and methods ...... 54

3.5.1 General methods ...... 54

3.5.2 Synthesis ...... 54

3.5.3 Fluorescence quantum yield measurements ...... 72

Chapter 4. Fluorescently-tagged DNA Oligomer SSB Assay for Probe Evaluation ...... 73

4.1 Hypothesis and methodology ...... 73

4.2 Evaluation of potential UDG inhibition ...... 76

4.3 Evaluation of potential APE inhibition ...... 78

4.4 Compound screening by dose-response ...... 81

4.5 Comparison to ARP and MX ...... 82

4.6 Identification of a transient band ...... 83

4.7 Evaluation of probe-AP site lesion repair: endonuclease survey ...... 86

4.8 Discussion and conclusions ...... 88

4.9 Materials and methods ...... 89

4.9.1 General methods ...... 89

4.9.2 Reactions to establish assay validity ...... 91

Page | v

4.9.3 Evaluation of AP site binding probes ...... 94

4.9.4 Evaluation of BER ...... 95

Chapter 5. Genomic DNA Binding Assay for Probe Development ...... 97

5.1 Hypotheses and methodology ...... 97

5.2 Blocking with MX ...... 100

5.3 Binding time course evaluation ...... 101

5.4 Competition with MX and ARP ...... 102

5.5 Measurement of cellular response to DNA damaging drugs ...... 104

5.5.1 Measurement of FUDR response ...... 104

5.5.2 Measurement of MMS response ...... 107

5.6 Discussion and conclusions ...... 108

5.7 Materials and methods ...... 109

5.7.1 General methods ...... 109

5.7.2 Heat/acid treated genomic DNA ...... 109

5.7.3 FUDR and MMS treatment of DLD1 cells ...... 112

Chapter 6. Microscopy and in Vivo Imaging of Cancer ...... 115

6.1 Microscopy ...... 115

6.1.1 Hypothesis and methodology ...... 115

6.1.2 Optimization of fixative ...... 116

6.1.3 Nonspecific binding with FEt2 ...... 117

Page | vi

6.1.4 FUDR dose response with Cy5MX ...... 119

6.1.5 Potential pitfalls in microscopy ...... 121

6.2 In vivo imaging in mouse xenograft models ...... 123

6.2.1 Hypothesis and methodology ...... 123

6.2.2 Preliminary evaluation of AuNP tumor targeting ...... 124

6.2.3 Direct administration of probe ...... 130

6.2.4 Preliminary evaluation of AuNP-conjugated Cy7MX ...... 133

6.2.5 Post mortem detection of AP site probe in isolated DNA ...... 136

6.2.6 Potential pitfalls in animal imaging ...... 137

6.3 Discussion and conclusions ...... 138

6.4 Materials and methods ...... 140

6.4.1 General methods ...... 140

6.4.2 Microscopy ...... 141

6.4.3 In vivo imaging ...... 145

Chapter 7. Future Directions and Conclusions ...... 148

7.1 Considerations for probe design ...... 148

7.1.1 Solubility ...... 148

7.1.2 Stability ...... 149

7.1.3 Biological compatibility...... 150

7.2 Considerations for cellular imaging ...... 151

Page | vii

7.2.1 Repair of probe-AP site lesion ...... 151

7.2.2 Off-target aldehyde binding ...... 151

7.2.3 Nonspecific binding ...... 152

7.3 Considerations for in vivo imaging ...... 153

7.3.1 AuNPs for tumor targeting and distribution ...... 153

7.3.2 DNA extraction from xenograft model ...... 154

7.3.3 In vivo imaging and biodistribution ...... 154

7.4 Conclusions ...... 155

7.4.1 Myelin imaging ...... 155

7.4.2 Cancer imaging ...... 156

References ...... 159

Page | viii

LIST OF TABLES

Table 1. Selected summary of lesions repaired by BER and their corresponding glycosylases...... 14

Table 2. Dielectric constants (relative permittivity) of solvents used in CIC fluorescence measurements compared to blood and nerve cell membranes...... 20

Table 3. Summary of absorption and emission wavelengths for AP site binding probes.

...... 52

Table 4. Fluorescence and absorbance properties of Cy7MX in various solvents...... 52

Table 5. Stock concentration and initial % DMSO of probes evaluated by SSB activity assay...... 94

Table 6. Methods for fixative comparison...... 142

Page | ix

LIST OF FIGURES

Figure 1. (A) The net magnetization (M, red arrow) of precessing nuclei aligns to a strong external magnetic field (B0, green arrow) as parallel alignment is lower energy than antiparallel alignment. (B) A 90° pulse causes some nuclei to flip to antiparallel alignment and for the nuclei to precess in phase. (C) T1 relaxation increases the net magnetization vector along the z-axis as the original distribution of nuclei in the parallel and antiparallel states is restored. (D) T2 relaxation occurs as nuclei dephase and the net magnetization vector decreases in the x/y plane...... 2

Figure 2. Schematic of PET imaging shows a radioactive nucleus (1) decaying by positron (e+) emission. (2) Annihilation occurs when positron and its antiparticle, the electron (e-), meet and releases two Ȗ photons in opposite directions, i.e. 180° apart. (3)

PET detectors must read two simultaneous events in opposite directions, i.e. 180° apart. 4

Figure 3. Jablonski diagrams illustrating quantum events leading to (A) typical one- photon excitation fluorescence and (B) two-photon excitation fluorescence (see §2.4, page 24, for a detailed explanation of two-photon imaging). The absorption photon, hȞA, excites an electron to a higher electronic state. Non-radiative internal conversion decreases the energy through the vibrational levels. Fluorescence occurs as a photon, hȞF, is released when the excited electron returns to the ground electronic state...... 6

Figure 4. Calculated absorption spectrum of tissue shows a minimum absorption in the

NIR range. Figure modified from ref.18 ...... 7

Figure 5. An axon ensheathed by lamellar myelin produced by oligodendrocytes...... 9

Page | x

Figure 6. Clinically used DNA-targeting cancer chemotherapies...... 12

Figure 7. The short patch BER pathway following UDG removal of uracil in DNA. .... 15

Figure 8. Normalized fluorescent spectra of CIC in various solvents.114 ...... 21

Figure 9. Microscope images of normal myelin and myelin-deficient 20 μm axial spinal cord sections stained with CIC and Black-Gold. (a-b) In vitro, wild-type sections with a)

CIC and b) Black–Gold; (c-d) in vitro, Thy1-YFP (normal myelin) sections with c) CIC and d) Black–Gold; (e-f) in vitro, shiverer (myelin-deficient) sections with e) CIC and f)

Black–Gold; (g-h) in vitro, l-Į-lysophosphatidyl choline (LPC)-induced focal demyelination (indicated with a white arrow) with g) CIC and h) Black–Gold; (i-j) ex vivo, wild-type sections with i) CIC and j) Black–Gold. Scale bar: 200 μm.114 ...... 22

Ballistic fluorescence resulting from (A) is removed by the pinhole and does not reach the PMT detector. (E) Scattered fluorescence resulting from (B) is filtered by the pinhole in confocal but contributes to the detected signal in two-photon. (F) Ballistic fluorescence from (B) is detected in confocal and two-photon. Figure modified from reference.16 ...... 26

Page | xi

CIC fluorescence. This box is magnified in d) with 1) axon bodies, 2) CIC fluorescence

(myelin sheaths), and 3) overlay. Swiss–Webster mouse with CIC in green and Texas red-labeled dextran in red e) less than 5 min and f) 2.5 h post CIC injection. Again, the

CIC channel shows a striated pattern (box with dashed lines) consistent with axons surrounded by myelin as in panels c and d. Shiverer mouse model with CIC signal in green and Texas red-labeled dextran in red g) less than 15 min and h) 2.5 h post CIC injection. A lower CIC signal is present in the shiverer mouse spinal cord when compared with the same time points in the wild-type myelin models as in panels c and f. Scale bar:

150 μm (b,c, e–g), 30 μm (d).114 ...... 27

Webster (blue striped bars) and shiverer (yellow solid bars) after data were normalized to the Texas-red signal in the blood vessels of each animal. * Indicates significance (p

<0.05).114 ...... 29

Page | xii

Figure 13. Detection channels for two-photon microscopy. The two-photon microscope signal was divided into four visible light detection channels as shown. The channels are referred to as red (565-605 nm), yellow (520-550 nm), green (467-499 nm), and blue

(<455 nm) throughout the text. Texas red-dextran was observed in the red channel. YFP was observed in the yellow channel with strong signals bleeding into the red. CIC was observed predominantly in the green channel. Autofluorescence from collagen was observed in the blue channel...... 44

Figure 14. Path A: the short patch BER pathway following UDG removal of uracil in

DNA. Path B: the proposed mechanism for interception of the reactive aldehyde present in the AP site with an aminooxy-tagged probe, which blocks further repair by APE and prevents the single strand break (SSB)...... 45

A shows that treatment with APE creates a SSB. Path B shows that an AP site-binding probe blocks the natural substrate of APE and its SSB activity. Denaturing PAGE resolves the fluorescently labeled 40mer and 16mer. The unlabeled 23mer and 40mer antisense strand are not detected by fluorescence imaging...... 74

Figure 16. (A) 5’-HEX labeled dsDNA 40mer used for fluorescence-based cutting assay;

(B) SSB assay visualized by denaturing gel electrophoresis to separate intact 40mer from

16mer. HEX labeled DNA is shown in green and Cy7MX is in red. (C) Analysis of the

Page | xiii gel electrophoresis controls with SSB activity % = (Fl 16mer)/(Fl 16mer + Fl 40 mer) x

100. (D) Analysis of gel electrophoresis with Cy7MX or vehicle control...... 75

Figure 17. Evaluation of Cy7MX inhibition of UDG under assay conditions. Time course represents incubation time of U:A DNA before APE addition with UDG alone

(blue circles), 10 minutes UDG pretreatment then 7 (red triangles), and UDG and 7 without pretreatment (green diamonds). Little difference is observed between the concurrent and tandem additions of Cy7MX and UDG, indicating the probe does not inhibit the enzyme under these conditions...... 77

Figure 18. APE shows mild inhibition by Cy7MX when incubated together before addition of AP-DNA. The SSB activity assay used a 60 min APE incubation after treatment of AP-DNA with Cy7MX, when no inhibition of APE activity by Cy7MX is detected. APE is inhibited about 50% after 3 h incubation. The baseline reduction in cutting seen at 10, 30, and 60 minutes is due to competition of APE and Cy7MX for the

AP-DNA substrate...... 79

Figure 19. APE shows inhibition with increasing dose of Cy7MX when incubated together in the presence of THF:A DNA for 60 min. APE inhibition is detectable at ca.

500 pmol Cy7MX...... 80

Figure 20. SSB activity assay is used to compare the dose-response of AP site binding probes in 5 pmol of AP DNA. Mean and standard deviation of triplicate samples are shown for each concentration...... 82

Page | xiv

Figure 21. Comparison of the SSB activity of U:A DNA (5 pmol) treated with UDG and

Cy7MX, ARP, MX, or vehicle control as a function of incubation time prior to APE addition. (A) ARP (1 nmol), MX (1 nmol), and Cy7MX (1 nmol) with UDG (5 units) and APE (10 units); (B) ARP (200 nmol), MX (200 nmol), and Cy7MX (2 nmol) with

UDG (5 units) and APE (1 unit). Bar represent the average of three samples and error bars are the standard deviation...... 83

Figure 22. Control experiments to identify unknown band in SSB activity assay...... 84

Figure 23. Illustration of the Cy5MX concentration dependent competition with APE.

At low Cy5MX concentrations, APE SSB activity dominates. At high Cy5MX concentrations, APE SSB activity is inhibited by Cy5MX binding to the 40mer before the enzyme reacts...... 85

Figure 24. The unknown band is identified as a 23-mer containing an AP site-bound

Cy5MX after APE SSB activity. HEX and Alexa 532 fluorophores are shown in green,

Cy5MX is shown in red...... 86

Figure 25. Evaluation of potential AP site-probe repair by several endonucleases after 1 h incubation. APE, endonuclease V, and vehicle control showed no repair activity.

Endonucleases III and IV showed SSB activity while retaining the probe on the DNA.

Endonuclease VIII removed the probe from the AP site and caused a SSB. Fluorescent tags on DNA are shown in green, Cy5MX is shown in red...... 88

Figure 26. Schematic of methodologies used in genomic DNA assays. (A-C) Heat and acid treatment of calf thymus DNA for a time period followed by purification by ethanol

Page | xv precipitation are used to (A) report on quantity of AP sites and MX blocking, (B) establish a DNA binding time course, and (C) measure ARP or MX competition. (D)

Chemotherapy treatment of cancer cells followed by DNA isolation and exogenous enzyme treatment...... 99

Figure 27. Calf thymus AP DNA was pretreated with MX (50 mM, +MX) or vehicle (50 mM, -MX) followed by incubation with Cy7MX (0.05 mM). Fluorescence levels show a linear response to Cy7MX in the absence of MX. No fluorescence response was observed with MX pretreatment, indicating MX and Cy7MX share a common binding site. Data points are the average of three samples and error bars represent the standard deviation...... 101

Figure 28. A time course of Cy7MX binding to calf thymus DNA treated with heat and acid for 45 min. was measured at 37 °C. A 2.6 min. reaction half-life was calculated from the curve fitting using the programed model DoseResp nonlinear fit in Origin 9.1.

Data points are the average of three samples and error bars represent the standard deviation...... 102

[Cy7MX]. Based on curve fittings using the programed model ExpDec2 nonlinear fit in

Origin 9.1, ED50 values were calculated to be 2600-fold excess for ARP and 3000-fold excess for MX. For (A), data points are the average of three samples and error bars

Page | xvi represent the standard deviation. For (B), data points are the average of five samples and error bars represent the standard error of the mean...... 104

UDG). Data bars are the average of three samples and error bars represent the standard deviation. *p < 0.001 ...... 106

Figure 31. Detection of AP sites in DNA isolated from DLD1 WT cells after 3 h continuous MMS exposure. Purified DNA was treated with Cy7MX (25 μM). Data bars are the average of three samples and error bars represent the standard deviation. *p <

0.001...... 108

Figure 32. (A) Confocal microscope images with 40x objective of cells fixed in various conditions then stained with F422 (green) and DAPI (blue). Bar = 50 ȝm. (B)

Quantification of the F422 fluorescence intensity normalized to the DAPI signal. Bars represent the average F422 signal of each nucleus and error bars represent the standard deviation of the average...... 117

Figure 33. Quantification of nuclear fluorescence by F422 and FEt2 following a TMZ time course. The maximum AP site formation is observed at 1 h TMZ treatment time.

FEt2 fluorescence indicates nonspecific fluorescence accounts for ca. 50% of the basal

F422 signal. Bars represent the average F422 or FEt2 signal of each nucleus and error bars represent the standard deviation of the average...... 119

Page | xvii

Figure 34. (A) Representative epifluorescent images with a 40x objective of KD DLD1 cells with 1000 nM or without FUDR treatment stained with Cy5MX (red) and DAPI

(blue). Bar = 50 ȝm. (B) Quantification of Cy5MX fluorescence intensity in regions of interest defined by the DAPI signal and normalized to the FUDR untreated KD + UDG signal intensity. Bars represent the average Cy5MX nuclear fluorescence from four fields-of-view and the error bars represent the standard deviation of the averages...... 120

Figure 35. Schematic of concurrent, continuous FUDR and Cy7MX treatment in living cells. DNA was extracted and the fluorescence of Cy7MX was measured on the purified

DNA...... 122

Figure 36. Fluorescence measurement of DNA extracted from living cells treated concurrently with Cy7MX and FUDR. Bars represent the averages of four samples and error bars represent the standard deviation of the averages...... 123

Figure 37. Mice imaged in a prone position following tail vein administration of free or

AuNP-conjugated IR780. Xenograft tumors of DLD1 WT and T98G cells were implanted on the left and right flanks, respectively...... 126

Figure 38. Mice imaged in a supine position following tail vein administration of free or

AuNP-conjugated IR780...... 127

Figure 39. Biodistribution of IR 780 administered directly or conjugated to AuNPs 24 h after administration. (A) Fluorescent imaging of Bo, bone; Br, brain; C, colon; DLD1,

DLD1 xenograft tumor; H, heart; K, kidneys; Li, liver; Lu, lungs; M, muscle; Sp, spleen;

St, stomach; T98G, T98G xenograft tumor. (B) Quantification of fluorescence from

Page | xviii biodistibution normalized to the liver signal. For all conditions, n=1 mouse and error bars represent standard deviation of fluorescence intensity within the ROI...... 129

Figure 40. Mice imaged in a prone position following tail vein administration of

Cy7MX. Xenograft tumors of DLD1 WT and KD cells were implanted on the left and right flanks, respectively...... 131

Figure 41. Mice imaged in a supine position following tail vein administration of

Cy7MX...... 131

Figure 42. Biodistribution of Cy7MX administered directly 16 days after treatment. (A)

Fluorescent imaging of Bo, bone; Br, brain; C, colon; H, heart; K, kidneys; KD, KD tumor; Li, liver; Lu, lungs; M, muscle; Sp, spleen; St, stomach; WT, WT tumor. (B)

Quantification of fluorescence from biodistribution. For all conditions, n=1 mouse and error bars represent standard deviation of fluorescence intensity within the ROI...... 132

Figure 43. Mice imaged in a prone position following tail vein administration of AuNPs loaded with Cy7MX. Xenograft tumors of DLD1 WT and KD cells were implanted on the left and right flanks, respectively...... 134

Figure 44. Mice imaged in a supine position following tail vein administration of AuNPs loaded with Cy7MX...... 134

Figure 45. Biodistribution of AuNP-conjugated Cy7MX 11 days after treatment. (A)

Fluorescent imaging of Bo, bone; Br, brain; C, colon; H, heart; K, kidneys; KD, KD tumor; Li, liver; Lu, lungs; M, muscle; Sp, spleen; St, stomach; WT, WT tumor. (B)

Page | xix

Quantification of fluorescence from biodistibution. For all conditions, n=1 mouse and error bars represent standard deviation of fluorescence intensity within the ROI...... 135

Figure 46. Fluorescence from filtrate following organ digestion. For this preliminary experiment, n=1 for each sample and the bar represent the ...... 137

Figure 47. Comparison of ARP-based AP site quantification kit (Dojindo) and proposed

Cy7MX-based kit. The Cy7MX offers a simple, direct route with less potential for error by manipulation of DNA concentration...... 157

Page | xx

LIST OF SCHEMES

Scheme 1. Similarity scheme of myelin-targeted imaging probes from our laboratory. . 17

Scheme 2. Original synthesis of CIC...... 18

Scheme 3. Improved synthesis of CIC...... 19

Scheme 4. Novel fluorescent probes for AP site detection...... 46

Scheme 5. Syntheses of NpCMX and ACMX...... 47

Scheme 6. Synthesis of MCMX...... 48

Scheme 7. Synthesis of Cy7MX...... 49

Scheme 8. Synthesis of Cy5MX ...... 50

Scheme 9. Synthesis of dansylMX...... 51

Scheme 10. Structure of F422 and synthesis of control compound FEt2...... 51

Page | xxi

ACKNOWLEDGEMENTS

I would like to thank my advisors, Dr. Yanming Wang and Dr. Stanton Gerson, for their leadership, expertise, and training and for giving me the opportunity to study with them. Thanks to my committee members, Dr. Robert Salomon, Dr. Clemens Burda,

Dr. Irene Lee, and Dr. Mark Barkley, for their experience and guidance during my time at

Case Western Reserve University.

For their help, encouragement, and kindness, I would like to thank my past and present research group members, Chunying Wu, Junqing Zhu, Yuguo Li, George Bakale,

Changning Wang, Eduardo Somoza, Nicholas Tomko, Yulan Qing, Hua Fung, John

Kenyon, Amar Desai, Lachelle Weeks, Yan Yan, and Larissa Guimaraes.

Funding for this project was made possible through the Case Comprehensive

Cancer Center and the Cancer Pharmacology Training Program. Special thanks go to Dr.

Noa Noy for managing the grant and for her thoughtful advice. I would also like to thank my fellow trainees.

Thanks to the faculty and staff of the CCCC, the Case Center for Imaging

Research, and the chemistry department.

Finally, I want to thank my parents, sisters, brothers-in-law, and all my extended aunts, uncles, cousins, and in-laws. Without their endless love and support, I would never have made it here today. Most of all I want to thank my former coworker, ceaseless mentor, constant companion, and dear husband, Luca. I would double the length of this thesis if I tried to explain all he has meant to me since the fateful day he helped me troubleshoot a UV spectrometer eight years ago (data not shown).

Page | xxii

SYMBOLS AND ABBREVIATIONS

aka: also known as

AP: apurinic/apyrimidinic

APE: AP endonuclease

APS: ammonium persulfate

ARP: aldehyde reactive probe a.u.: arbitrary units

AuNP: gold nanoparticles

BBB: blood-brain barrier

BMB: 1,4-bis(p-aminostyryl)-2-methoxy benzene

CIC: Case Imaging Compound

CNS: central nervous system

Cy5: pentamethine cyanine

Cy7: heptamethine cyanine

DAPI: 4',6-diamidino-2-phenylindole, dihydrochloride

DBU: 1,8-diazabicyclo[5.4.0]undec-7-ene

DCM: dichloromethane

DMEM: Dulbecco's Modified Eagle's medium

DMF: dimethylformamide

DMSO: dimethylsulfoxide

DNA: deoxyribonucleic acid

DSB: double strand break

Page | xxiii dsDNA: double stranded DNA

DTT: dithiothreitol

EAE: experimental autoimmune encephalomyelitis

EDC: N-(3-dimethylaminopropyl)-Nƍ-ethylcarbodiimide

EDTA: ethylenediaminetetraacetate

ESI: electrospray ionization

EtOAc: ethyl acetate

EtOH: ethanol

FMT: fluorescence mediated tomography

ĭ: fluorescence quantum yield

FUDR: 5-fluoro-2ƍ-deoxyuridine (aka floxuridine)

FUra: 5-fluorodeoxyuracil

Gd-DODAS: gadolinium (E)-2,2ƍ,2Ǝ-(10-(2-((3-(4-((4-(4-(methylamino)styryl)-

phenoxy)methyl)-1H-1,2,3-triazol-1-yl)propyl)amino)-2-oxoethyl)-1,4,7,10-

tetraazacyclododecane-1,4,7-triyl)triacetic acetate

HEX: hexachlorofluorescein

HPLC: high pressure liquid chromatography

HR: homologous repair

HRMS: high resolution mass spectrometry

HOBt: 1-hydroxy-1H-benzotriazole

ICG: indocyanine green or IR-125

KD: knock down (in given gene expression)

KO: knock out (in given gene expression)

Page | xxiv

Lig III: DNA ligase III

LPC: lysophosphatidyl choline

MBP: myelin basic protein

MeCN: acetonitrile

MeDAS: N-methyl-4,4'-diaminostilbene

MeOH: methanol

MIC: Myelin Imaging Compound

MMR: mismatch repair

MMS: methyl methanesulfonate

MRI: magnetic resonance imaging

MS: multiple sclerosis

MX: methoxyamine

NER: nucleotide excision repair

NHEJ: nonhomologous end joining

NIR: near infrared

NMR: nuclear magnetic resonance

Norm. Fl.: normalized fluorescence

NP: nanoparticle

PAGE: polyacrylamide gel electrophoresis

PBS: phosphate buffered saline

PBST: phosphate buffered saline + 0.1 x triton-X

PET: positron emission tomography

PFA: paraformaldehyde

Page | xxv

Pol ȕ: DNA polymerase ȕ

PNS: peripheral nervous system

ROI: region of interest

ROS: reactive oxygen species

RT: room temperature, 22-25 °C

SAR: structure activity relationship

§: section

SSB: single strand break ssDNA: single stranded DNA

TBE: tris-borate-EDTA buffer

TE: tris-EDTA buffer

TEMED: tetramethylethylenediamine

TFA: trifluoroacetic acid

THF: tetrahydrofuran

TLC: thin layer chromatography

TMS: tetramethylsilane

TMZ: temozolomide

UDG: uracil DNA glycosylase

UV: ultraviolet

WM: white matter

WT: wild type

XRCC1: X-ray repair cross-complementing protein 1

YFP: yellow fluorescent protein

Page | xxvi

Development of Molecular Probes for Biomedical Imaging of Cancer and Neurological

Disease

ABSTRACT

Abstract

ALLISON G. CONDIE

Cancer and neurological disease afflict millions of people in the United States and worldwide. In addition to the heavy toll they take on society, these diseases have devastating bodily consequences. Molecular imaging is a noninvasive means to understand the bodily phenomena driving these diseases and to recognize hopes for treatment. Optical fluorescence imaging is a type of molecular imaging that uses light to probe biological processes and states. Fluorescent contrast agents designed for specific molecular targets report on the status of disease and response to therapy.

Myelin is an insulating sheath surrounding axons that aids in nervous signal transduction. Pathologies in myelin are associated with many neurological diseases, most prominently multiple sclerosis. A contrast agent has been evaluated for its ability to bind to myelin in the spinal cord and report on the quantity of myelin sheaths in vitro and in vivo. A novel synthesis of the agent, fluorescent characterization, and application to histological staining is also described.

Page | xxvii

Many chemotherapeutic treatments of cancer damage DNA leading to cell death.

To maintain genomic fidelity, cells have evolved multiple pathways to repair DNA damage. These pathways are active in cancer cells and can limit the therapeutic efficacy of DNA damaging drugs. One notable repair pathway is base excision repair, which excises chemically damaged bases. Optical imaging probes have been designed, synthesized, and characterized for their ability to bind to the first intermediate of the base excision repair pathway, the abasic (or AP) site. An assay has been developed to evaluate AP site-targeted probes. Their ability to report on physiologically relevant quantities of AP sites has been established in tissue culture. While this work focuses on

AP sites in cancer, base excision repair is also relevant to neurological diseases including

Alzheimer’s and Parkinson’s diseases.

The optical imaging probes targeted to myelin and AP sites have potential preclinical application in new drug discovery. They can also be further developed to monitor response to therapy either in cell culture or animal models of disease. The probes could also be modified for combination use with additional imaging modalities for application in a clinical setting.

Page | xxviii

CHAPTER 1. INTRODUCTION

1.1 Review of molecular imaging

1.1.1 Overview of molecular imaging

Medical imaging noninvasively visualizes the interior of an organism. Molecular imaging, a branch of medical imaging, identifies phenomena related to molecular and cellular level events, including biological processes and disease. The “molecular” aspect of imaging is a broad term that encompasses molecules present in the cell, extrinsic contrast agents, or even gene expression products.1 Clinical goals of molecular imaging include improved disease diagnosis, early disease detection, and monitoring disease response to therapy.1 Molecular imaging modalities use an array of detection strategies.

Of particular note are magnetic resonance imaging (MRI), positron emission tomography

(PET), and optical imaging, which all have numerous variations and contrast agents.

1.1.2 Magnetic resonance imaging

MRI detects perturbations in magnetic fields of nuclei in the presence of strong, external static and pulsed magnetic fields. The ubiquity of water protons in biological systems makes them the most extensively examined MRI nuclei. Two important parameters that affect the obtained signal are the T1 (spin-lattice) and T2 (spin-spin) relaxation of the nuclei (Figure 1). These parameters are dependent on the environment and the external magnetic field strength.2 Contrast agents, which modify relaxation time or magnetization, are divided into two general classes, T1 and T2 agents. Gadolinium (III)

Page | 1

and other lanthanide metal-based agents are principally used as T1 agents and iron oxide

3 particles are T2 agents. Practical implications are that T1 agents brighten an image while

4 T2 agents darken an image. ABC

M T B0 1

pulse D T 2

MRI has excellent spatial resolution, on the order of 25-100 μm.5 Another

advantage is that it provides anatomical information and three dimensional images.

Imaging can be performed without contrast agents, but contrast agents provide

information otherwise not accessible. There is no limit on the depth of penetration. MRI

is suitable and currently used in both clinical and preclinical settings.

One major disadvantage of MRI imaging is that its low sensitivity requires large

concentrations of contrast agent, from 10-3 to 10-5 M.5 From a clinical perspective,

persons with kidney dysfunction do not tolerate the required large doses of lanthanide

metals, and acute kidney injury has been observed in some patients owing to exposure to

a contrast agent.6 MRI contrast agents are also sensitive to field strength and will give

Page | 2

varying response based on magnet size.7 While pulse sequences can be designed to evaluate multiple phenomena, akin to having multichannel imaging, clinical scan time is precious and the scan rate is limited by the T1 relaxation time. MRI is fairly costly as scanners require refilling of cryogenic liquid nitrogen and helium, special facilities to minimize magnetic interference, and isolation from iron and other magnetic metal tools.

1.1.3 Positron emission tomography

PET imaging uses positron emitting radioisotopes in its contrast agents. When positrons collide with electrons, they annihilate releasing gamma radiation.1 Special detectors on PET instruments detect the simultaneous radiation from the annihilation on either side of the sample specimen, separated by 180° (Figure 2).8 Contrast agents may

11 18 contain short lived isotopes such as C (t1/2 = 20 min.) and F (t1/2 = 110 min.) or

64 1 relatively long lived isotopes such as Cu (t1/2 = 13 d). The specificity of PET contrast agents comes from the molecule that carries the radiolabel. For example, the most

18 18 9 commonly used clinical PET agent is 2-deoxy-2-( F)fluoro-D-glucose, or [ F]FDG.

This glucose analog accumulates in tumors, which have high rates of glucose metabolism, leading to the selectivity of the imaging agent.

Page | 3

3. γ radiation detection

2. annihilation e+ e–

1. decay by e+ emission 3. γ radiation detection

As PET detects two simultaneous gamma photon emissions, the background noise is miniscule, and the sensitivity is very high. Tiny amounts of contrast agent (10í11 to

10í12 M)5 are needed for imaging and do not present safety risks to patients from the radioactivity. Unlike MRI agents, PET agents behave uniformly across detection platforms. PET imaging has no penetration restrictions and is used both clinically and preclinically. As a tomographic technique, PET provides three dimensional images.

PET imaging does not provide anatomical information necessitating the use of a second imaging modality for co-registration. The resolution limits are on the order of 1-2 mm.5 Multichannel imaging is not possible with PET and a second scan cannot be conducted until the first isotope has decayed sufficiently. This is a disadvantage, especially for long lived isotopes. Special dedicated facilities are required for “hot” Page | 4

syntheses of PET agents. While the radiation exposure risk is small for patients, the risk to the pharmacist or researcher who prepares the agents is much higher. Cyclotrons are required to produce radioisotopes10 and these facilities require high security and safety precautions. Thus, PET imaging can be quite costly and limited to institutions and hospitals with adequate resources.

1.1.4 Optical imaging

Optical imaging uses visible or near-visible light for detection. Most often this light comes from light emitting contrast agents, but not strictly so. Bioluminescence imaging uses an enzyme such as luciferase to generate light in a sample following a biological event.11-12 Bioluminescence requires no incident light. Conversely, fluorescence imaging uses incident light to excite an electron in a fluorophore. When the electron relaxes to its ground state, light of a lower energy is emitted, as expressed in the

Jablonski diagram (Figure 3).13 The quantum yield is efficiency of energy dispersion through fluorescence versus other processes, such as emission of heat, expressed as the percentage of photons emitted from the photons absorbed. The difference between the excitation and emission wavelengths is the eponymous Stokes shift.14 This emitted light has a characteristic wavelength and can be detected and quantified. Fluorescence mediated tomography (FMT)15 and confocal or two-photon16 imaging provide three dimensional images, but optical imaging may also be two dimensional.

Page | 5

ABNon-radiative Non-radiative transition transition 3 3 2 2 S1 S1 1 1 0 0

Fluorescence hνF Fluorescence hνF Energy Energy

hνA hνA Absorption Absorption 3 3 2 2 S0 S0 1 1 0 0

Figure 3. Jablonski diagrams illustrating quantum events leading to (A) typical one-photon excitation fluorescence and (B) two-photon excitation fluorescence (see §2.4, page 24, for a detailed explanation of two-photon imaging). The absorption photon, hȞA, excites an electron to a higher electronic state. Non-radiative internal conversion decreases the energy through the vibrational levels. Fluorescence occurs as a photon, hȞF, is released when the excited electron returns to the ground electronic state.

The most important advantage of optical imaging is that it is multichannel.

Several processes can be analyzed simultaneously by using multiple, different colored fluorophores.17 Another advantage is that optical imaging is relatively inexpensive and does not have the strict facility requirements of PET and MRI.5

The major disadvantage to optical imaging is that there is a limit to the depth of penetration into tissue.18 Tissue strongly absorbs ultraviolet (UV) and visible light but not near infrared (NIR) light (Figure 4), which can penetrate on the order of centimeters.18

Light scattering also occurs in tissue19 but methods have been developed to remove scattered light from detection.16 For preclinical small animal imaging, NIR fluorescence is preferable to UV-vis due to strong autofluorescence in the UV-vis range in tissues and fluids.20 Clinical applications of NIR imaging are limited but have been applied when tissue penetration issues can be circumvented such as in image guided surgery.21-22

Page | 6

100

10 ) -1

0.1 Absorption coefficient (cm Absorption coefficient

0.01 400 450 500 550 600 650 700 750 800 850 900 Wavelength (nm)

Figure 4. Calculated absorption spectrum of tissue shows a minimum absorption in the NIR range. Figure modified from ref.18

Fluorescent contrast agents for optical imaging range from small molecules to proteins and antibodies.17, 23-24 The fluorophore itself can be the active agent (see Chapter

2, page 16) or the fluorophore can be a tag on a targeting moiety.17 This versatility makes fluorescent probes excellent multimodal imaging candidates. A fluorophore can be tethered to a lanthanide chelating complex for use as an MRI agent,25 or can be modified to accommodate a radiolabel, such as a [11C] methyl group, for PET imaging.26

The fluorescence provides an added feature of being a histological marker following in vivo imaging.

Page | 7

1.2 Review of myelin imaging

1.2.1 Motivation and basis for disease

Myelination is a process conserved through vertebrate species. The myelin sheath is a membrane produced by oligodendrocytes and Schwann cells in the central and peripheral nervous systems, respectively, to provide an insulating layer around nerve cells (Figure 5).27 Myelin assists in signal transduction through nerves. Abnormal formation of myelin (dysmyelination) is common in leukodystrophies28 including

Canavan and Krabbe’s diseases.29 Destruction of myelin (demyelination) is characteristic of many diseases including multiple sclerosis (MS),30 transverse myelitis,31 neuromyelitis optica (Devic’s disease),32 and Guillain–Barré syndrome.33 MS is the largest public health concern of these demyelinating diseases and affects approximately 2.5 million people worldwide.34 Demyelination can result from physical injury to the spinal cord35 and chemical injury such as Lhermitte’s sign caused by the cisplatin family of drugs.36-37

Page | 8

Lamellae

Figure 5. An axon ensheathed by lamellar myelin produced by oligodendrocytes.*

1.2.2 Histological or in vitro imaging agents

Histological agents have been developed to report on myelin status in tissue sections.38 Many of these agents contain heavy metals including copper in Luxol Fast

Blue39-40 and gold in Black-Gold41-42 and Black-Gold II.43 Sudan Black is a nonfluorescent chromophore that stains dark brown.44 The fluorescent myelin probe

Fluoromyelin45 is commercialized by LifeTechnologies as a histological myelin stain and also has been used in live cell imaging of myelin.46 While histological agents are important tools for the analysis of tissue sections, they are limited to post mortem use and cannot be used as diagnostic tools on a living patient. For this, in vivo imaging agents are needed.

* The copyright holder of this figure has released this work into the public domain [online] at http://commons.wikimedia.org/wiki/File:Neuron_with_oligodendrocyte_and_myelin_sheath.svg (accessed Jan 2015). Page | 9

1.2.3 In vivo imaging agents

Myelination has been characterized in vivo by several imaging techniques including MRI (with and without contrast), PET, and optical imaging (vide infra). Cheng and coworkers have used Coherent anti-Stokes Raman scattering to analyze the lysophosphatidyl choline (LPC)-induced focal demyelination model.47

MRI plays a key role in standard-of-care diagnostic tools for myelin diseases, particularly in MS48-49 and the leukodystrophies.50 However, the use of MRI as a sole diagnostic test for MS has not been approved by the FDA because the MRI signal arises from tissue water content, which could be caused by nonspecific edema and inflammation in addition to MS symptoms. Therefore, our group has designed myelin- specific MRI contrast agents termed Myelin Imaging Compound (MIC)25, 51 and Gd-

DODAS (short for gadolinium (E)-2,2ƍ,2Ǝ-(10-(2-((3-(4-((4-(4-(methylamino)styryl)- phenoxy)methyl)-1H-1,2,3-triazol-1-yl)propyl)amino)-2-oxoethyl)-1,4,7,10- tetraazacyclododecane-1,4,7-triyl)triacetic acetate)52 based on coumarin and stilbene myelin-binding moieties, respectively, tethered to a Gd(III) chelator. These agents display promising properties of imaging myelin but cannot permeate the intact blood- brain barrier (BBB).

PET, although used clinically, is not used routinely for diseases of myelin.

Positron emission does not occur in the body without introduction of a radiotracer; therefore, PET imaging is highly sensitive. Our group developed several 11C radiolabeled myelin-binding compounds based on a stilbene core: N-methyl-4,4'-diaminostilbene

(MeDAS),53 Case Imaging Compound (CIC),26 and 1,4-bis(p-aminostyryl)-2-methoxy benzene (BMB).54 These compounds showed excellent brain imaging responses.

Page | 10

The myelin-binding cores in the probes developed by our group have an added benefit of intrinsic fluorescence (see Scheme 1, page 17). These compounds have shown utility as histological stains of tissue sections as well as imaging agents.25-26, 51-54 With this in mind, our group examined several fluorescent probes for myelin binding potential based on stilbene,55-56 coumarin,57-58 and cyanine59 cores. Fluorescence provides a relatively inexpensive and facile tool for preclinical imaging.

1.3 Review of cancer imaging

1.3.1 Motivation and basis for disease

Cancer continues to be one of the great public health concerns in the United States with nearly 1.7 million estimated new cases and 600,000 estimated deaths in 2014 for the

USA alone.60 Cancer as a set of diseases is characterized by several hallmarks including resisting cell death, sustaining proliferative signaling, evading growth suppressors, activating invasion and metastasis, enabling replicative immortality, and inducing angiogenesis, among others.61

Medical science has a three-pronged approach to treating cancer: surgery, radiation, and chemotherapy. Surgery aims to remove the tumor from the body while radiation and chemotherapy kill cancer cells in situ. Chemotherapy also includes pathway controlling agents that alter pathological activities of specific biochemical pathways. Certain cancers respond exceedingly well to one prong while others are nearly unaffected. For example, the DNA cross-linking chemotherapeutic cisplatin, in a cocktail with etoposide and bleomycin, approaches a curative rate greater than 90% in low risk stage III testicular cancer,62 whereas the disease relapse rate in small cell lung cancer

Page | 11

following cisplatin/carboplatin therapy is greater than 95%.63 Like cisplatin, many anticancer drugs target DNA to cause damage leading to cell death.64 Three important classes of DNA-targeting cancer chemotherapies include DNA cross-linking agents, alkylating agents, and antimetabolites. Some examples of these agents are shown in

Figure 6. cross-linkers antimetabolites NH 2 NH2 O H2 O O N O NH3 N N N Pt Cl NH3 Pt O O NH3 Pt O N N O O N N F O H HO F O O Cl NH3 2 P HO O F NH HO O OH carboplatin cisplatin oxaliplatin N O OH F H OH gemcitabine fluorouracil fludarabine (5FU)

O O alkylating agents OH Cl N F O Cl O H2N NH H2N HO OH HN N HO N N O N O N Cl NH H H HO N N CO H N N O Cl P O N 2 O O NH N O O HO O N CO2H OH temozolomide chlorambucil cyclophosphamide streptozotocin pemetrexed floxuridine (TMZ) (FUDR)

Figure 6. Clinically used DNA-targeting cancer chemotherapies.

1.3.2 DNA damage and repair

DNA is constantly bombarded with reactive species that could potentially mutate or damage the information it stores. Reactive oxygen species (ROS), environmental carcinogens, and UV radiation from sunlight all damage DNA. Several DNA damage repair pathways have evolved to maintain the fidelity of DNA.65-68 There is some redundancy between these pathways and within the individual pathways as well. These pathways preserve the integrity of DNA in normal tissue but in cancer cells their activity can lead to drug resistance of DNA-targeting chemotherapies.

DNA repair pathways include homologous repair (HR), nonhomologous end joining (NHEJ), nucleotide excision repair (NER), mismatch repair (MMR), and base excision repair (BER). HR and NHEJ are mechanisms to repair catastrophic DNA

Page | 12

double strand breaks (DSB) and interstrand links.69-70 NER removes damaged bases in the form of oligonucleotides and is particularly relevant to DNA damage caused by ultraviolet (UV) radiation and cisplatin.71-72 MMR primarily targets, but is not limited to, replication errors in the forms or mismatches, insertions, and deletions.66, 73 BER repairs damaged DNA bases and is particularly relevant to exogenous chemical damage such as that caused by chemotherapy.74-76 BER repairs direct chemical modification of nucleotides or accumulation of aberrant bases, such as uracil.77

1.3.3 BER and AP sites

Glycosylases initiate the BER pathway by hydrolyzing the glycosidic bond to release the damaged base and the free sugar. The types of lesions repaired through base excision repair and their corresponding glycosylases are summarized in Table 1.

Page | 13

Table 1. Selected summary of lesions repaired by BER and their corresponding glycosylases.

Glycosylase Important lesions

FPG (formamidopyrimidine [fapy]-DNA glycosylase) fapy, 8-oxoG78-79

MPG (methylpurine DNA glycosylase; aka AAG, ANPG) 3-MeA, 7-MeG80-81

NEIL1* (Nei-like) fapy, Tg, 5-OH-C, DHT82-83

NTH1† (endonuclease III homologue) fapy, Tg, DHU84-85

OGG1‡ (8-oxoguanine DNA glycosylase) 8-oxoG:C, fapy85-86

SMUG1 (single-strand-selective monofunctional uracil-DNA U, 5-X-U (X=OH, hm, glycosylase) f)87

UDG (uracil DNA glycosylase; aka UNG) U, FUra88-91

Abbreviations in this table: 8-oxoG (8-oxoguanine), A (adenine), DHT (dihydrothymine), DHU (dihydrouracil), f (formyl), fapy (formamidopyrimidine), FUra (5-fluorouracil), G (guanine), hm (hydroxymethyl), Me (methyl), OH (hydroxyl), T (thymine), Tg (thymine glycol), U (uracil)

Some glycosylases also have lyase activity, leading to a single stand break (SSB) in the DNA backbone,66 whereas monofunctional glycosylases only cleave the glycosidic bond. Uracil DNA glycosylase (UDG) was the first glycosylase to be isolated and has been the most extensively studied.88 For this reason, and because of the facile introduction of uracil into synthetic and cellular DNA, this enzyme was selected as the model for the studies presented herein.

The free sugar produced by the action of monofunctional glycosylases is termed an abasic or AP (apurinic/apyrimidinic) site.68 The AP site is the substrate for AP endonuclease (APE), which nicks the sugar-phosphate backbone 5’ to the AP site creating a single strand break (SSB). The SSB signals the rest of the short patch BER

* Homolog of E. coli Endo VIII † Homolog of E. coli Endo III ‡ Homolog of E. coli FPG Page | 14

repair cascade recruiting DNA polymerase ȕ (Pol ȕ) to remove the sugar from the AP site then replace the proper nucleotide, and X-ray repair cross-complementing protein 1

(XRCC1) and DNA ligase III (Lig III) to ligate the sugar-phosphate backbone and complete the repair (Figure 7).67-68

5' 3' 5' 3' 5' 3' 5' 3'

HO O O O O O O P O NH O P O O O P O NH O O O P O O N O OH OH N O O UDG O APEO O Polβ/XRCC1/Lig3 O O O O O O P O O P O A O P O O P O O O O O

Figure 7. The short patch BER pathway following UDG removal of uracil in DNA.

1.3.4 Existing agents for imaging and detection

Several research groups have developed tools to detect AP sites based on methods including fluorescence,92-93 nanopore ion detection,94 mass spectrometry,95-96 atomic force microscopy,97 electrochemistry,98 and electron paramagnetic resonance.99 Some of these techniques employ a variety of molecular probes targeted to the AP site through various chemical features of the lesions.100 Several DNA base analogs have been synthesized to detect AP sites based on changes in fluorescence.92, 101-102 These probes are covalently incorporated at sites complementing AP sites in oligomer chains. Other probes rely on interactions with bases flanking or opposing the AP site.93, 103-104 The AP site contains a reactive aldehyde that can react with strong nucleophiles; in the case of the alkoxyamino derivatives, this produces a covalent oxime ether. Several probes have been developed that rely on this chemistry.105-106 Among them, aldehyde reactive probe (ARP), based on biotin tethered to an alkoxyamine, detects AP sites in a colorimetric streptavidin-horseradish peroxidase in vitro assay.107 Ames and coworkers have

Page | 15

successfully measured AP sites in DNA extracted from living cells treated with ARP,108 but the reliance on the streptavidin assay limits its use beyond purified DNA. A derivative of APR with improved binding has recently been introduced as well.109

Our research group has studied the simplest oxime ether-forming probe, methoxyamine (MX), and has shown that it potentiates the effect of temozolomide

(TMZ) in colon cancer cells.110-111 A carbon-11 was introduced into MX to functionalize it as a probe. This radiolabel permitted MX to be used as a positron emission tomography (PET) agent for in vivo imaging of DNA damage and repair.106 While this agent holds great potential for use in a clinical setting, its application in a preclinical setting may be impeded by requirements of an onsite cyclotron for production of the short-lived 11C isotope, designated radiochemistry facilities, and PET scanners. To this end, fluorescent AP site probes that could be used in optical imaging or microscopy present a cost-effective solution.

CHAPTER 2. MYELIN IMAGING WITH CIC

2.1 Hypothesis and methodology

Our group has developed several probes for myelin visualization. Many of these probes contain a stilbene core with electron donating groups on the para position of the aromatic rings (Scheme 1). These probes have been used for PET, MRI, and optical imaging of the brain. However, many demyelinating diseases also present with lesions in the spinal cord, which are a major cause of motor disability in patients. We hypothesized that our agents could translate from brain to spinal cord imaging and allow visualization

Page | 16

and quantification of myelin status in vivo. Due to the intrinsic fluorescence of the probes, optical imaging using two-photon microscopy with the probe CIC was selected as the proof-of-principle method. A novel synthesis of CIC was developed to provide improved yields and facile purification. The fluorescent properties of CIC were fully characterized to ensure its candidacy as a two-photon probe. In vitro and ex vivo histology established that CIC binds to myelin in the spinal cord and this binding can be detected. Finally, two-photon microscopy was used to visualize and quantify CIC binding to normal and pathological myelin.

H2N H2N F

O H2N

O O O O

O O

NH2 N NH NH NH2

CMCFIC MeDAS CIC BDB

Scheme 1. Similarity scheme of myelin-targeted imaging probes from our laboratory.

2.2 CIC synthesis and fluorescent characterization

The first synthesis of CIC (Scheme 2) has been previously reported.26 The novel synthetic route for CIC shown in Scheme 3 (page 19) is an improvement over the original synthesis, which required purification of a complex mixture of methylated products and afforded an overall yield of 13%.

Page | 17

Scheme 2. Original synthesis of CIC.

The synthesis shown in Scheme 3 has a 36% overall yield and simple purification.

Briefly, a Horner-Wadsworth-Emmons reaction coupled bisaldehyde 6 and phosphonate

7 to give stilbene 8 in 73% yield. Phosphonate 9 was first protected as carbamate 10 then monomethylated with iodomethane to give 11 in 90% yield over two steps. The fluorophore scaffold was completed with a second Horner-Wadsworth-Emmons reaction between aldehyde 8 and 11 to afford 12. Deprotection of the carbamate gave advanced intermediate 13, which was converted to CIC by a tin(II)-mediated nitro group reduction.

Page | 18

OMe OMe OEt O O P NaH, DMF + OEt O O r.t., 16 h O N 2 73% OMe OMe O2N 6 78

OEt OEt OEt P (Boc)2O, THF P MeI, NaH, THF P OEt OEt OEt O Boc O Boc O H2O, r.t., 16 h 0 °C to r.t., 16 h H2N N N 96% H 94% 91011

8,NaH,DMF r.t., 16 h 87%

NO2 NO2 OMe OMe

TFA, CH2Cl2

r.t., 4 h 96% OMe Boc OMe N N H 13 12

SnCl2, EtOH, EtOAc 70 °C, 16 h 65%

NH2 OMe

OMe N H CIC

Scheme 3. Improved synthesis of CIC.

The fluorescent properties of CIC were examined to determine its suitability for two-photon microscopy. In order to model the dielectric environments of blood and myelin, a series of solvents with relative permittivity constants ranging from highly polar to highly nonpolar were used to measure CIC fluorescence (Table 2). This is an important consideration because CIC is given intravenously into the polar blood

Page | 19

environment but extravasates into the nonpolar nerve tissue. The fluorescence response of CIC measured in these solvents (Figure 8) indicated that CIC undergoes bathochromic solvatochromism. The excitation maxima ranged from 448-467 nm and the emission maxima from 466-517 nm. In microscopy applications with tunable detectors, this solvatochromism could be exploited to exclude CIC signal from blood. For the two- photon imaging studies, the 467-499 nm detection filter was adequate to detect CIC regardless of dielectric environment.

Table 2. Dielectric constants (relative permittivity) of solvents used in CIC fluorescence measurements compared to blood and nerve cell membranes.

112 Environment Dielectric Constant (HR)

Blood 58*

Nerve cell membrane 8.5113

DMSO 47.2†

iPrOH 20.2†

Hexanes 1.87-1.89†

THF 7.52‡

Water 80.1†

# CH2Cl2 8.93

EtOAc 6.08†

Except where otherwise noted, data taken from ref112 *37 °C, †20 °C, ‡22 °C; #25 °C

Page | 20

Figure 8. Normalized fluorescent spectra of CIC in various solvents.114

The fluorescence quantum yield (ĭ) of CIC was measured relative to fluorescein

(ĭ=0.91 in 0.1 M NaOH, 20 °C) in nonpolar, polar aprotic, and polar protic solvents:

ĭEtOAc = 0.64, ĭMeCN = 0.45, ĭMeOH = 0.32, andĭsaline = 0.05. The two-photon cross section (G) of CIC was determined to be (90 ± 20) x 10-50 cm4 s/photon in methanol. This value compares favorably to well-characterized two-photon probes such as fluorescein and rhodamine B, which have two-photon cross sections of (210 ± 55) x 10-50 cm4 s/photon and (38 ± 9.7) x 10-50 cm4 s/photon, respectively.115

2.3 In vitro and ex vivo histology

Myelin-rich nerve tissue appears white and is associated with the white matter in the central nervous system (CNS). Conversely, the gray matter in the CNS takes its name from its gray color resulting from low quantities of myelin. When sectioned axially, the gray matter in the spinal cord forms a butterfly-like shape of dark tissue surrounded by a white ring. A myelin-targeted fluorescent probe, such as CIC, will stain the myelin-rich

Page | 21

white matter and appear bright in an image while the gray matter will remain unilluminated.

To confirm that CIC stains myelin selectively, axial spinal cord sections were taken from an untreated wild type (WT) mouse (i.e. normal myelin levels and distribution). Adjacent sections were stained in vitro with either CIC or Black-Gold II

(Figure 9 A-B), a common and well-characterized myelin histological stain.41-43 CIC appears blue in the stained image with unstained areas appearing dark. Black-Gold II appears red when bound to myelin. In this normal myelin model, the CIC staining pattern is consistent with the Black-Gold II stain.

Figure 9. Microscope images of normal myelin and myelin-deficient 20 μm axial spinal cord sections stained with CIC and Black-Gold. (a-b) In vitro, wild-type sections with a) CIC and b) Black–Gold; (c-d) in vitro, Thy1-YFP (normal myelin) sections with c) CIC and d) Black–Gold; (e-f) in vitro, shiverer (myelin-deficient) sections with e) CIC and f) Black–Gold; (g-h) in vitro, l- Į-lysophosphatidyl choline (LPC)-induced focal demyelination (indicated with a white arrow) with g) CIC and h) Black–Gold; (i-j) ex vivo, wild-type sections with i) CIC and j) Black–Gold. Scale bar: 200 μm.114

A second normal-myelin mouse was also examined. This mouse, Thy1-YFP, is genetically modified to express yellow fluorescent protein (YFP) in the axons. Axons are ensheathed in myelin and the YFP label provides an excellent anatomical marker for two- Page | 22

photon imaging. Axial spinal cord sections from the Thy1-YFP model were stained with

CIC and Black-Gold II in vitro to verify that this model behaves similarly to the WT model. As shown in Figure 9 C-D, the myelin staining in Thy1-YFP mouse model is consistent with the WT mouse.

The dysmyelinated shiverer mouse is a model for the leukodystrophy family of diseases.116 The shiverer mouse takes its name from its characteristic tremors, or shivers, that increase in severity and progress to seizures and shortened lifespan.117 Myelin may be present in the spinal cord of the shiverer mouse, but the myelin sheaths contain structural abnormalities and are thin or even absent.118 The phenotype arises from a partial deletion of the myelin basic protein (MBP) gene, which is a major protein in myelin necessary to compact the structure.119-120 Thus, the shiverer mouse is a model for pathological hypomyelination.

Lysophosphatidyl choline (LPC) is a neurotoxin that degrades myelin to generate a focally demyelinated mouse model in vivo. LPC demyelinates both CNS and PNS nerves in a dose-dependent manner.121-123 Following subperineural injection, demyelination can be observed after 30 minutes and is complete after 96 hours.121-123

LPC does not damage axons or Schwann cells.121 PNS nerves begin remyelination within

14 days of treatment.121, 123 The focal lesions induced by LPC model the characteristic demyelinated lesions in MS patients.124

Shiverer axial spinal cord sections were treated in vitro with CIC and Black-Gold

II (Figure 9 E-F). Lack of contrast between white and gray matter in both images is consistent with the global hypomyelination in the model (the fluorescence image was captured with a longer exposure than other images in order to visualize the section).

Page | 23

Conversely, the focal demyelination of the LPC-treated mouse shows high contrast between white and gray matter with in vitro CIC and Black-Gold II staining (Figure 9 G-

H, page 22). The demyelinated lesion (indicated with an arrow) appears in the white matter between the symmetric “wings” of the gray matter butterfly. Brightfield Black-

Gold II imaging confirmed that the lesion was present in the intact tissue and not caused by a tear or other damage in the tissue section itself. Therefore, CIC staining can differentiate pathological myelination from normal, healthy myelination by in vitro tissue staining.

While CIC showed excellent myelin staining in vitro, ex vivo staining is necessary to demonstrate that the probe has adequate pharmacological properties for spinal cord imaging. The bioavailability of the CIC is irrelevant when given via a tail vein injection (bioavailability, the fraction of drug absorbed into systemic circulation, is

100% by definition for any intravenous injection).125 However, the rates of metabolism and excretion may compete with the extravasation of CIC into the nerve tissues.

Therefore, a WT mouse was treated with CIC and sacrificed after two hours. The spinal cord was sectioned axially. Adjacent sections were stained with Back-Gold II post mortem. Fluorescent microscopy revealed that CIC did reach the spinal cord under these conditions and the staining pattern was consistent with Black-Gold II and in vitro WT staining (Figure 9, page 22). Thus, CIC has pharmacological properties amenable to in vivo imaging.

2.4 In vivo visualization of CNS nerves

Two-photon fluorescence microscopy excites fluorophores with two low energy

(NIR) photons instead of a single high energy (UV or visible) photon that is the

Page | 24

excitation source in confocal (“one-photon”) microscopy. Denk and coworkers first introduced two-photon microscopy touting “unprecedented capabilities for three- dimensional, spatially resolved photochemistry.”126 Indeed, the two-photon system confers distinct advantages over traditional confocal microscopy: NIR light can penetrate deeper into tissues than UV-vis, low energy excitation decreases photodamage to biological samples, and there is less interference from incident and scattered light than traditional confocal microscopy.126-127

In confocal microscopy, incident photons may scatter and excite fluorophores outside of the focal area (Figure 10A). The emission photons from these off-target fluorophores are removed by the use of a pinhole (Figure 10D). However, the confocal pinhole also removes emission photons that have scattered after exciting the focal area

(Figure 10E), thereby sacrificing sensitivity for improved resolution.16 Conversely, two- photon microscopy requires two quantum events occurring in quick succession making the probability of an excitation outside the focal volume negligible (Figure 10A versus

B). No pinhole is required to remove scattered incident photons and all excitation photons can be collected (Figure 10E and F).16

Page | 25

Confocal Two-photon PMT

Pinhole

F dichroic mirror F

E E

B C B B D

A A

Figure 10. Comparison of two-photon and confocal microscope. Confocal excitation is higher energy than emission (blue v. green lines). Two-photon excitation is lower energy than emission (red v. green lines). (A) Scattered excitation photons within specimen: in confocal excite the fluorophore but in two-photon do not. (B) One ballistic (not scattered) photon excites the fluorophore in confocal while two ballistic photons excite in two-photon. (C) Scattered photons out of the specimen do not induce fluorescence. (D) Ballistic fluorescence resulting from (A) is removed by the pinhole and does not reach the PMT detector. (E) Scattered fluorescence resulting from (B) is filtered by the pinhole in confocal but contributes to the detected signal in two-photon. (F) Ballistic fluorescence from (B) is detected in confocal and two-photon. Figure modified from reference.16

Due to the advantages of two-photon over confocal microscopy for in vivo imaging, live animal spinal cord imaging studies were performed using the former technique. Mice were deeply anesthetized and a laminectomy exposed the spinal cord between the T11 and T12 thoracic vertebrae. Clamps secured the spinal column to minimize movement due to respiration (Figure 11 A).

Page | 26

Figure 11. Summary of two-photon imaging set-up and imaging at 15 minutes and 2.5 hours after CIC treatment. a) Schematic of the two-photon set-up of T12 mouse spinal cord imaging. This region of the spinal cord, shown in b–h, contains vasculature and nervous tissue. Thy1-YFP mouse with YFP-labeled axons in yellow and CIC in green b) less than 15 min and c) 2.5 h after CIC administration. The square (solid lines) in c) encloses a region where the axon bodies, observed in the YFP channel, are surrounded by CIC fluorescence. This box is magnified in d) with 1) axon bodies, 2) CIC fluorescence (myelin sheaths), and 3) overlay. Swiss–Webster mouse with CIC in green and Texas red-labeled dextran in red e) less than 5 min and f) 2.5 h post CIC injection. Again, the CIC channel shows a striated pattern (box with dashed lines) consistent with axons surrounded by myelin as in panels c and d. Shiverer mouse model with CIC signal in green and Texas red-labeled dextran in red g) less than 15 min and h) 2.5 h post CIC injection. A lower CIC signal is present in the shiverer mouse spinal cord when compared with the same time points in the wild-type myelin models as in panels c and f. Scale bar: 150 μm (b,c, e–g), 30 μm (d).114

Normal myelin Thy1-YFP mice were imaged before CIC injection for a background measurement and to ensure the fluorescence of the YFP did not bleed into the CIC detection channel. An image collected 15 minutes after CIC injection showed the probe extravasating from the blood vessels into the nerve tissue (Figure 11 B). After

2.5 hours, CIC cleared completely from systemic circulation and had localized in the nerve tissue (Figure 11 C). YFP-labeled axons appear in a striated pattern as nerves run

Page | 27

in long, parallel branches (Figure 11 D1). These striations are visible as dark shadows in the CIC image, consistent with axons ensheathed in myelin (Figure 11 D2).

To ensure these results were not an artifact of fluorescence imaging, a Swiss

Webster WT normal myelin mouse model was examined as well. Without the intrinsic fluorescence marker the YFP labeled axons provided, an external fluorescent dye was injected to allow an anatomical registration and a proper focus. A Texas red-labeled dextran was used as this dye does not permeate the blood vessels and persists in this WT mouse for the duration of the imaging study. A background image was collected to verify the absence of autofluorescence. An image capturing the bolus in the vein was collected within five minutes of CIC injection (Figure 11 E). In this image, the overlapping CIC (green) and Texas red (red) appear yellow. After 2.5 hours, CIC cleared from systemic circulation and had localized in the nerve tissue (Figure 11 F) as it had for the Thy1-YFP mouse. The same pattern of dark striations appeared in the nerve tissue

(dashed box in Figure 11 F) confirming the hypothesis that this pattern is due to the structure and shape of myelin and is not an artifact of imaging.

Having demonstrated that CIC binds to myelin and can be visualized in vivo, the next step was to determine whether it could distinguish a pathological from a normal myelin state. The shiverer mouse model was chosen for this study because of its consistent hypomyelination throughout the CNS. The EAE model, arguably the best experimental model for MS, was discounted because the unpredictable location of spinal cord lesions. As with the Thy1-YFP and WT mice, Texas red-labeled dextran and CIC were visible in the vasculature of the shiverer mouse 15 minutes after CIC injection

(Figure 11 G). The CIC extravasated from the blood vessels into the surrounding nerve

Page | 28

tissue. However, without myelin to retain the probe, after 2.5 h the CIC was absent in the nerve tissue in the shiverer mouse (Figure 11 H).

To quantify the binding of CIC in the two-photon imaging study, regions of interest (ROIs) were drawn around the vasculature and nerve tissue in images. The average of the absolute CIC fluorescence of these ROIs was plotted as a function of time for three mouse models: Swiss Webster (WT), C57BL/6J (WT), and shiverer (Figure 12

A, B, and C, respectively). CIC passed from the vasculature into the nerve tissue faster in the Swiss Webster and shiverer models than in the C57BL/6J model. However, both WT models retained CIC in the nerve tissue but the probe cleared quickly from the shiverer model. These data confirm that myelin-binding CIC was retained in the normal myelin

WT models but not the myelin-deficient shiverer model.

Figure 12. Comparison of intensity of CIC signal in wild-type versus shiverer mice. The intensities of the fluorescence of CIC in blood vessels (red circles) are compared to nerve tissue, i.e. myelin (green triangles), for (A) Swiss Webster, (B) C57BL/6J, and (C) shiverer models. (D) The CIC signal was compared in nerve tissue over time for Swiss–Webster (blue striped bars) and shiverer (yellow solid bars) after data were normalized to the Texas-red signal in the blood vessels of each animal. * Indicates significance (p <0.05).114

Page | 29

The fluorescence of CIC was then compared between the Swiss Webster WT and shiverer models using the Texas red-labeled dextran as an internal standard by which relative fluorescence could be quantified. Swiss Webster and shiverer models were used because the rates of CIC extravasation and Texas red-labeled dextran clearance were similar while the C57BL/6J model was omitted because its corresponding rates varied considerably. Normalization of the CIC signal to the Texas red-labeled dextran revealed that the retention of CIC in the nerve tissue was significantly higher in the WT than the myelin-deficient model (Figure 12 D). These data support the conclusion that CIC binds to myelin and can distinguish pathological and healthy myelination in vivo.

2.5 Conclusions

This research marks the first application of a myelin-binding probe from our research group to spinal cord imaging. CIC was developed as a PET radiotracer for detecting demyelination in the brain. Since this work successfully showing CIC detection of myelin in the spinal cord was completed, other researchers in Dr. Wang's group have begun studying [11C]-MeDAS and [11C]-CIC as radiotracers for PET spinal cord imaging.128 Further, others in Dr. Wang's research group have begun toxicology studies of CIC and aims to begin imaging studies in nonhuman primates.

Myelin pathology in the spinal cord is a major cause of disability in individuals afflicted with demyelinating and dysmyelinating diseases. The work presented here could benefit preclinical development of therapies targeted at these diseases or their phenotypes. Fluorescent imaging of CIC in animal models could report rapidly, easily, and cheaply on the efficacy of new therapies.

Page | 30

2.6 Materials and methods

2.6.1 General methods

Chemicals were used as received without further purification. Unless otherwise indicated, reactions were performed open to the air. For inert atmosphere reactions, glassware was dried in an oven at 130 °C and purged with a dry argon atmosphere prior to use. Reactions were monitored by TLC and plates visualized by a dual short wave/long wave UV lamp. Column flash chromatography was performed using 230-400 mesh silica gel (Fisher). NMR spectra were recorded on Varian Inova 400 spectrometer at room temperature. Chemical shifts for 1H NMR were reported as į, parts per million,

13 and referenced to CHCl3 at 7.26 ppm. Chemical shifts for C NMR were reported as į, parts per million, and referenced to the center line of the CDCl3 triplet at 77.0 ppm. The abbreviations s, br s, d, dd, ddd, br d, t, td, q, m, and br m stand for the resonance multiplicity singlet, broad singlet, doublet, doublet of doublets, doublet of doublet of doublets, broad doublet, triplet, triplet of doublets, quartet, multiplet, and broad multiplet, respectively. Mass spectra were recorded at the Mass Spectrometry Facility at the

Department of Chemistry at Boston University in Boston, MA on a Waters Q-Tof API-

US with ESI high resolution mass spectrometer.

Page | 31

2.6.2 Synthesis

tert-butyl (4-((diethoxyphosphoryl)methyl)phenyl)carbamate (10). To a 100 mL round bottom flask fitted with a magnetic stir bar were added diethyl 4- aminobenzylphosphonate (9, 2.500 g, 10.28 mmol), di-tert-butyl dicarbonate (2.240 g,

10.27 mmol), THF (15 mL), and water (6.0 mL). The reaction was stirred at room temperature open to air overnight. The THF was removed in vacuo and the resulting residue was diluted with water and extracted with EtOAc three times. The organic layers were combined and washed twice with water and once with brine. The organic layer was dried over MgSO4, filtered, and concentrated in vacuo. The resulting white solid (3.393

1 g, 96%) was used without further purification. Rf = 0.26 (MeOH/CH2Cl2 1:19); H-NMR

(CDCl3, 400 MHz) į = 7.30 (br d, J = 8.4 Hz, 2H), 7.17 (m, 2H), 6.87 (br s, 1H), 4.04-

2 3 4 3.91 (m, 4H), 3.07 (d, JH,P = 21.2 Hz, 2H), 1.49 (s, 9H), 1.22 (td, JH,H = 6.8 Hz, JH,P =

13 0.4 Hz, 6H); C-NMR (CDCl3, 100 MHz) į = 152.8, 137.4, 130.1, 125.5, 118.4, 80.3,

1 62.0, 32.9 (d, JC,P = 138 Hz, 1C), 28.8, 16.3.

OEt MeI, NaH OEt O P THF O P OEt OEt O O 0°tor.t. O N O N H 94% 10 11 tert-butyl (4-((diethoxyphosphoryl)methyl)phenyl)(methyl)carbamate (11). To an oven dried 100 mL round bottom flask purged with argon and fitted with a magnetic stir bar was added sodium hydride (0.589 g, 14.7 mmol, 60% dispersion in mineral oil). The

Page | 32

sodium hydride was washed three times with hexanes (6 mL) and the solvent was discarded. Compound 10 was added under argon and the mixture was suspended in dry

THF (25 mL). The reaction was cooled to 0 °C and methyl iodide (1.20 mL, 19.3 mmol,

2.28 g/mL) was added slowly drop wise. The reaction was stirred under argon and gradually warmed to room temperature overnight. The reaction was quenched with water and the THF was removed in vacuo. The residue was dissolved in EtOAc and water and the aqueous layer was extracted three times with EtOAc. The organic layers were combined and washed twice with water and once with brine. The organic layer was dried over MgSO4, filtered, and concentrated to give crude 11 with some 10 remaining (10 and

1 11 have such similar Rf values that the reaction is more easily monitored by H-NMR than TLC). The crude 11 was subjected to the identical reaction conditions: NaH (0.589 g), MeI (1.20 mL), and dry THF (25 mL). The reaction was worked up exactly as before

1 to give pure 11 as a yellow oil (3.279 g, 94%). Rf = 0.26 (MeOH/CH2Cl2 1:19); H-NMR

(CDCl3, 400 MHz) į = 7.24 (m, 2H), 7.16 (br d, J = 8.0 Hz, 2H), 4.07 – 3.94 (m, 4H),

3 13 3.22 (s, 3H), 3.11 (d, JH,P = 21.6 Hz, 2H), 1.42 (s, 9H), 1.23 (t, J = 7.2 Hz, 6H); C-

NMR (CDCl3, 100 MHz) į = 154.5, 142.4, 129.7, 128.4, 125.4, 80.1, 61.9, 37.1, 33.0 (d,

2 JC,P = 138 Hz, 1C), 28.1, 16.3.

(E)-2,5-dimethoxy-4-(4-nitrostyryl)benzaldehyde (8). To an oven dried 100 mL round bottom flask fitted with a magnetic stir bar was added 2,5-dimethoxybenzene-1,4-

Page | 33

dicarboxaldehyde (6, 402 mg, 2.07 mmol). The flask was purged with argon and dry

DMF (24 mL) was added. The solution was stirred under argon at room temperature. To an oven dried 10 mL round bottom flask fitted with a magnetic stir bar was added diethyl

4-nitrobenzylphosphonate (7, 470 μL, 2.13 mmol, 1.239 g/mL). The flask was purged with argon. Sodium hydride (98.0 mg, 2.45 mmol, 60% in mineral oil) was rinsed with hexanes (1.5 mL) three times. The sodium hydride was added to 7 in dry DMF (5.0 mL).

This sodium hydride mixture was allowed to stir at room temperature under argon for one hour. The sodium hydride mixture was then transferred via syringe to the solution of 6.

The 10 mL flask was rinsed with dry DMF (5 mL) and this rinse was added to the mixture of 6. The reaction was stirred overnight at room temperature under argon.

Additional 7 (32.0 μL, 0.204 mmol) and NaH (13 mg, 0.325 mmol, 60% in mineral oil) were shaken together in dry DMF (3.0 mL) for 30 min and then added to the reaction mixture. Once completed, the reaction was quenched with water (~25 mL) and extracted with ethyl acetate (~30 mL) three times. The organic layers were combined and washed twice with water (~25 mL) and once with brine (~25 mL). The organic layer was dried over MgSO4, filtered, and concentrated to give a red solid. This solid was dissolved in hot CH2Cl2, loaded onto a silica column, and eluted with 9:1 toluene to hexanes to remove the disubstituted side product and give impure 8 as an orange solid. This solid was recrystallized as a solid from boiling CH2Cl2 to give 8 as orange needles, which turned yellow after prolonged drying (476 mg, 73%). Rf = 0.29 (hexanes/CH2Cl2, 3:7; starting material: Rf = 0.32; major side product: R=0.47); R = 0.24 (EtOAc/hexanes, 1:4;

1 starting material: Rf = 0.36; major side product: Rf = 0.24); H-NMR (CDCl3, 400 MHz)

į = 10.45 (s, 1H), 8.24 (ddd, J = 9.2, 4.4, 2.4 Hz, 2H), 7.70 (ddd, J = 9.2, 4.4, 2.4 Hz,

Page | 34

2H), 7.64 (d, J = 16.6 Hz, 1H), 7.37 (s, 1H), 7.27 (d, J = 16.6 Hz, 1H), 7.20 (s, 1H), 3.99

13 (s, 3H), 3.92 (s, 3H); C-NMR (CDCl3, 100 MHz) į = 188.9, 156.4, 151.6, 147.1, 143.5,

132.5, 129.7, 127.3, 127.2, 124.9, 124.1, 109.9, 109.4, 56.2, 56.1.

tert-butyl (4-((E)-2,5-dimethoxy-4-((E)-4-nitrostyryl)styryl)phenyl)(methyl)carbamate

(12). To an oven dried 100 mL round bottom flask fitted with a magnetic stir bar was added 8 (535 mg, 1.71 mmol). The flask was purged with argon and dry DMF (12 mL) was added. The solution was stirred under argon at room temperature. To an oven dried

50 mL round bottom flask fitted with a magnetic stir bar was added 11 (990 mg, 2.73 mmol). The flask was purged with argon. Sodium hydride (173 mg, 3.76 mmol, 60% in mineral oil) was rinsed with hexanes (1.5 mL) three times. The sodium hydride was added to 11 in dry DMF (12 mL). This sodium hydride mixture was allowed to stir at room temperature under argon for one hour. The sodium hydride mixture was then transferred via syringe to the solution of 8. The 50 mL flask was rinsed with dry DMF

(12 mL) and this rinse was added to the mixture of 8. The reaction was stirred overnight at room temperature under argon. The reaction was quenched with water (25 mL) and extracted with ethyl acetate (30 mL) three times. The organic layers were combined and washed with water (25 mL) twice and brine (25 mL) once. The organic layer was dried over MgSO4, filtered, and concentrated to give a bright red solid. This solid was dissolved in a mixture of CH2Cl2 and EtOAc, loaded onto a silica column, and eluted with

Page | 35

hexanes. Once the band of compound was no longer observed to move in the column, the elution solvent was changed to EtOAc:hexanes (1:19) then EtOAc:hexanes (1:9).

Concentration in vacuo gave 12 as a bright red solid (767 mg, 87%). Rf = 0.32

1 (EtOAc/toluene, 1:19); H-NMR (CDCl3, 400 MHz) į = 8.21 (ddd, J = 9.2, 4.0, 2.4 Hz,

2H), 7.67-7.62 (m, 3H), 7.51 (ddd, J = 9.2, 4.0, 2.4 Hz, 2H), 7.44 (d, J = 16.4 Hz, 1H),

7.24 (m, 2H), 7.18-7.09 (m, 4H), 3.95 (s, 3H), 3.93 (s, 3H), 3.28 (s, 3H), 1.47 (s, 9H);

13 C-NMR (CDCl3, 100 MHz) į = 154.6, 151.9, 151.3, 146.4, 144.4, 143.1, 134.5, 128.9,

127.9, 127.8, 126.7, 126.6, 126.1, 125.3, 124.9, 124.0, 122.5, 109.2, 108.7, 80.4, 56.2,

56.1, 37.1, 28.3.

4-((E)-2,5-dimethoxy-4-((E)-4-nitrostyryl)styryl)-N-methylaniline (13). Compound 12

(110 mg, 0.213 mmol) was added to a 20 mL vial and dissolved in CH2Cl2 (10 mL).

Trifluoroacetic acid (540 μL, 0.703 mmol, 1.48 g/mL) was added to the solution of 12.

The reaction was stirred in the dark, open to air, at room temperature for four hours. The reaction was quenched with saturated NaHCO3 until bubbles no longer formed. The reaction mixture was diluted with water. The aqueous layer was extracted three times with CH2Cl2. The organic layers were combined and washed twice with water and once with brine. The organic layer was dried over MgSO4, filtered, and concentrated to crude

13. The crude residue was dissolved in minimal CH2Cl2 and loaded onto a silica column packed with EtOAc and hexanes (1:9). The column was flashed with a mobile phase of

Page | 36

EtoAc and hexanes (1:9 then 1:4). The solvent was evaporated to give pure 13 as a dark

1 red solid (85 mg, 96%). Rf = 0.56 (EtOAc/hexanes, 1:1); H-NMR (CDCl3, 400 MHz) į

= 8.20 (ddd, J = 9.2, 4.4, 2.4 Hz, 2H), 7.68-7.62 (m, 3H), 7.42 (ddd, J = 9.6, 4.6, 2.8 Hz,

2H), 7.28 (d, J = 16.4 Hz, 1H), 7.16-7.05 (m, 4H), 6.61 (ddd, J = 9.6, 4.6, 2.8 Hz, 2H),

13 3.95 (s, 3H), 3.92 (s, 3H), 3.87 (br s, 1H), 2.88 (s, 3H); C-NMR (CDCl3, 100 MHz) į =

152.0, 151.0, 148.9, 146.3, 144.6, 130.0, 129.1, 128.1, 128.0, 127.0, 126.7, 125.7, 124.1,

124.0, 118.4, 112.5, 109.3, 108.3, 56.3, 56.1, 30.7.

4-((E)-4-((E)-4-aminostyryl)-2,5-dimethoxystyryl)-N-methylaniline (CIC). To a 100 mL round bottom flask fitted with a stir bar was added tin (II) chloride (1.41 g, 7.44 mmol). Compound 13 (155 mg, 0.372 mmol) was added to the tin in ethyl acetate (25 mL) and ethanol (10 mL). The mixture was fitted with a water condenser, heated to 70

°C, and stirred overnight open to the air. The solvent was removed in vacuo. The compound was dissolved in EtOAc and aqueous NaOH and extracted with EtOAc. The organic layers were combined and washed with NaOH(aq), water, and brine. The organic layer was dried over MgSO4, filtered, and concentrated to give crude CIC as a brown solid. The crude product was dissolved in minimal EtOAc and purified on silica by flash chromatography with a mobile phase of EtOAc:hexanes (1:4), which was increased to

EtOAc:hexanes (2:3). Concentration gave CIC as a bright yellow-orange solid (94 mg,

1 65%). Rf = 0.19 (EtOAc/toluene, 1:9); H-NMR (CDCl3, 400 MHz) į = 7.41 (ddd, J =

Page | 37

9.6, 4.6, 2.8 Hz, 2H), 7.37 (ddd, J = 9.2, 4.2, 2.4 Hz, 2H), 7.29 (d, J = 16.4 Hz, 1H), 7.27

(d, J = 16.4 Hz, 1H), 7.10 (s, 1H), 7.09 (s, 1H), 7.03 (d, J = 16.4 Hz, 1H), 7.01 (d, J =

16.4 Hz, 1H), 6.68 (ddd, J = 9.2, 4.2, 2.4 Hz, 2H), 6.60 (ddd, J = 9.6, 4.6, 2.8 Hz, 2H),

3.91 (s, 3H), 3.90 (s, 3H), 3.81 (br s, 1H), 3.74 (br s, 2H), 2.87 (s, 3H); 13C-NMR

(CDCl3, 100 MHz) į = 151.2, 151.1, 148.8, 145.9, 128.8, 128.7, 128.5, 127.8, 127.7,

127.3, 126.7, 126.2, 119.6, 118.8, 115.2, 112.4, 108.7, 108.6, 56.4, 56.3, 30.6; HRMS

+ + (ESI) m/z calculated for C25H27N2O2 ([M+1] ) 387.2073, found 387.2075.

2.6.3 Fluorescence characterization

Absorbance and fluorescence spectra were collected on a Cary Eclipse fluorimeter in a quartz cuvette with a 1 cm pathlength.

Fluorescence quantum yield. Fluorescent quantum yields were collected under magic angle conditions at 20 °C and determined by a comparative method to a fluorescein standard (ĭ = 0.91 in 0.1 M NaOH, 20 °C) with refractive index correction.

Data were collected in a quartz cuvette with a 1 cm pathlength. Absorbance maxima for each sample were kept below 0.1 absorbance units to avoid inner filter effects.

Fluorophores were excited at 430 nm and emission spectra integrated from 450 to 700 nm. Solvents used were HPLC grade ethyl acetate, acetonitrile, and methanol; ACS grade extra dry dimethylformamide; and 0.9% saline for irrigation. For the saline solution, CIC was dissolved first in DMF (500 μM) and diluted with saline (1:49 v/v

DMF/saline). Further dilutions and the solution blank were prepared with 1:49 v/v

DMF/saline.

Two-photon cross section. The experimental setup used to measure the one- and two-photon absorption spectra is based on a Quantronix Integra-i/e 3.5 laser described

Page | 38

previously.129-130 The 400 nm excitation wavelength was generated by doubling the frequency of the 800 nm fundamental beam in a nonlinear ȕ-barium borate (BBO) crystal. A neutral density optical filter was used to adjust the 800 or 400 nm excitation pulses to the desired intensity. A depolarizing plate was used to randomize the excitation beam. The broadband probe pulses were split into two fractions, re-collimated, and focused into optical fibers leading into the spectrometer/CCD detection units. One of the two beams was passed through a 2 mm optical path length sample cell (Quartz, Starna

Cells, Inc.) for probing the signal, while the other beam was used as reference for the probe pulse’s spectral composition. Data were processed by LabView-based software.

Absorbance spectra from 50,000 laser shots were collected and averaged.

For one-photon absorption measurements, an excitation of 400 nm was used and methanolic solutions of CIC and the standard, Rhodamine 6G (į = 134 x 10-50 cm4 s/photon),131 were prepared such that their OD at 400 nm was equivalent. For two- photon absorption, an excitation of 800 nm was used and equimolar solutions of CIC and

Rhodamine 6G were prepared in methanol. The one- and two-photon transient absorption spectra were recorded with a delay time of 60 ps at different excitation energies to ensure that the data were within the linear and quadratic regimes, respectively. Sample solutions were continuously stirred during data acquisition.

Transient absorption signals were probed at 435 nm (Rhodamine 6G) and 650 nm (CIC).

These data were used to calculate the two-photon cross section of CIC based on the procedure of Rentzepis and co-workers.131

Page | 39

2.6.4 General animal methods

All animal experiments were performed in accordance with the guidelines approved by the Institutional Animal Care and Use Committee of Case Western Reserve

University (Protocol 2010-0007). C57BL/6J, C3Fe.SWV-Mbpshi/J (shiverer), and B6.Cg-

Tg(Thy1-YFP)16Jrs/J mice were obtained from Jackson Laboratory, Bar Harbor, MN;

ND4 Swiss Webster mice were obtained from Harlan Industries, Indianapolis, IN.

LPC treatment. Focal demyelinated tissue sections were prepared by treatment of an adult C57BL/6J mouse with L-Į-lysophosphatidyl choline (Sigma Aldrich # L4129,

1% in saline). The mouse was deeply anesthetized under isofluorane. The back was depilated and cleaned with isopropanol and betadine. Following sterile procedure, an incision was made in the skin and the tissue removed between the T11 and T12 vertebrae.

LPC was stereotaxically injected into the spinal cord. The wound was sutured and the mouse allowed to wake on a warming pad. After 14 days, the mouse was humanely euthanized and the spinal cord was excised and sectioned as before.

2.6.5 Tissue staining

Preparation of ex vivo tissue samples. For each injection, CIC (1.0 mg ± 0.1 mg) was dissolved in DMSO (80 μL) and Tween 80 (3 drops from 36 gauge needle). Saline

(200 μL) was added slowly dropwise, shaking vigorously after every few drops. The

CIC solution was injected into the tail vein of mice. After two hours, the mice were deeply anaesthetized and transcardially perfused with aqueous solutions of saline and paraformaldehyde (PFA). The spinal cords were removed and stored in PFA at 4 °C in the dark. After four days, the PFA solution was removed and the spines were stored in

30% sucrose in PBS at 4 °C in the dark for one week to fix the tissue.

Page | 40

The muscles and bone surrounding the spine were carefully removed working from the anterior terminus to the posterior. The spinal cords were stored at 4 °C in the dark in 30% sucrose in PBS until sectioning. Prior to sectioning, the cords were rinsed twice with water. Each cord was frozen in O.C.T. Compound (Tissue-Tek, #4583) and cut into 20 μm sections. The spines were sectioned axially. Sections were fixed onto

Fisherbrand Superfrost Plus microscope slides (precleaned, #12-550-15). Slides were stored in the dark at 4 °C until used.

Preparation of in vitro tissue samples. Untreated mice were deeply anaesthetized with isoflurane before transcardial perfusion with aqueous solutions of saline then PFA.

The spinal cords were removed and prepared exactly as before (ex vivo tissue samples).

CIC ex vivo staining. Ex vivo CIC stained sections were protected with

Vectrashield (Vector Laboratories, #H-100), mounted with glass cover slips, and secured with clear nail polish.

CIC in vitro staining. To prepare CIC in vitro treated slides, a solution of CIC was prepared (1.0 mg in 260 μL DMSO, 10 mM). This solution was used to prepare a

100 μM solution diluted with 1 drop of Tween-80 and 1 x PBS buffer. The areas around the sections of interest were circumscribed with an edge pen. The sections were then treated with 0.1% PBST buffer in the dark for 10 min. The PBST solution was removed and the sections were incubated in the dark with 100 μM CIC solution for 30 minutes.

After incubation, the CIC solution was removed and slides were washed three times with

1 x PBS buffer for 5 minutes. The PBS was removed and the sections were dried, protected with Vectrashield, mounted with glass coverslips, and secured with clear nail polish. Slides were stored in the dark at 4 °C until used.

Page | 41

Black-Gold II in vitro and ex vivo staining. A 0.3% solution of Black-Gold II was prepared by dissolving Black-Gold II (150 mg) in NaCl(aq) (0.9%, 50 mL). The solution was warmed to 60 °C. Tissue sections were wetted for 5 minutes with 1x PBS. The PBS was removed and enough Black-Gold II solution was added to cover the sections.

Tissues were incubated at 60 °C and the color change was monitored by eye, approximately 10 min. Sections were washed with 1x PBS (3 x 5 minutes) to remove unbound Black-Gold II.

2.6.6 Microscopy

Epi-fluorescence microscopy. Tissue sections were analysed on a Nikon Eclipse

Inverted research microscope with a 4x Plan Fluor lens (N.A. 0.13). Fluorescent filter sets used were UV-2E/C (DAPI) (ex 340-380, dm 400, ba 435-4850) and B-2E/C (FITC)

(ex 465-495, dm 505, ba 515-555). Due to bleeding of the YFP signal into the FITC detection channel, only the DAPI channel was used to analyze CIC staining in samples from Thy1-YFP mice.

Two-photon fluorescent microscopy. The mouse was deeply anesthetized using isoflurane prior to and during surgery. The eyes of the mouse were lubricated with

Puralube Vet ointment. Following depilation of the back with clippers and the hair remover, Nair, an incision was made into the skin of the mouse. Muscle and fat were carefully pulled aside to expose the T11-T12 discs. A laminectomy of the T11 or T12 was performed to facilitate access to the spinal cord. Metal clamps were inserted into the muscles on either side of the spine grasping the vertebra above and below the exposed spine. The clamps lifted the spine to reduce the movement due to breathing. A well was made around the exposed site using 3M Vetbond and dental acrylic (Lang Dental Jet

Page | 42

Tooth Shade liquid #1404a and Jet Tooth Shade powder #1420). The mouse was given superficial epinephrine as needed to reduce bleeding. The well was then filled with 1 x

PBS buffer. The mouse was positioned on the microscope platform and the objective was placed in the buffer.

The mouse was then given a dose of CIC (2 mg/kg in 1:4 DMSO+3 drops Tween

80/saline) via tail vein or retro-orbital injection. The mouse was repositioned under the microscope and a pump was placed to continuously replenish the 1 x PBS buffer. An image was collected every five minutes for two to three hours. During image acquisition, the temperature of the ambient air around the mouse was held between 36-37 °C and the mouse was kept under isoflurane (1-2%). Images were collected on a Leica SP5 upright microscope equipped with a high frequency (10-12-10-15/second) near infrared

Cameleon laser (700-1040 nm; 16W; Coherent, Inc.) using a 20x water immersion lens

(N.A. 1.0). Filter sets used were cube 1: no filter (BP 440 nm), CFP (483/32) with BS

RSP 445 and cube 2: YFP (535/30), Tritc/DSred (585/40) with BS RSP 560 (Figure 13).

After imaging, mouse models were humanely euthanized with carbon dioxide. Data were processed with Imaris software and quantified using ImageJ.

Page | 43

Figure 13. Detection channels for two-photon microscopy. The two-photon microscope ssignal was divided into four visible light detection channels as shown. The channels are referred to as red (565-605 nm), yellow (520-550 nm), green (467-499 nm), and blue (<455 nm) throughout the text. Texas red-dextran was observed in the red channel. YFP was observed in the yellow channel with strong signals bleeding into the red. CIC was observed predominantly in the green channel. Autofluorescence from collagen was observed in the blue channel.

CHAPTER 3. PROBES FOR CANCER IMAGING THROUGH AP SITE

DETECTION

3.1 Hypothesis and methodology

UDG and other monofunctional glycosyylases detect and remove damaged DDNA bases to initiate BER. The glycosidic cleavage of the base from the sugar produces the reactive AP site intermediate. The AP site is the natural substrate for APE, which binds to the lesion and makes a SSB (Figure 14, path A). The aminooxy moiety has been used in a variety of AP site probes to bind to the AP site and form an oxime ether (see §1.3.4, page 15). We hypothesized that a fluorophore tethered to an aminooxy group would

Page | 44

covalently bind to the AP site and block the action of APE (Figure 14, path B). While

DNA damage and repair is a dynamic process, by blocking the action of APE, the repair process stalls. Covalently bound fluorescent probes would allow detection and quantification of DNA damage in the presence of repair enzymes.

5' 3' 5' 3' 5' 3' 5' 3'

HO O O O O O O P O NH O P O O O P O NH O O O P O O N O OH OH N O O UDG O APEO O Polβ/XRCC1/Lig3 O O O O O O P O O P O A O P O O P O O O O O

III

5' 3' 5' 3' 5' 3' HO

O O O O P O O P O O P O O O O Probe Probe HO H2N-O-Probe O HO APE O HO O N N O O O O P O O P O O P O O O O

I III

Figure 14. Path A: the short patch BER pathway following UDG removal of uracil in DNA. Path B: the proposed mechanism for interception of the reactive aldehyde present in the AP site with an aminooxy-tagged probe, which blocks further repair by APE and prevents the single strand break (SSB).

3.2 Synthesis of compounds for AP site detection

Fluorescent AP site binding probes were intended for three broad applications: detection of damage in purified DNA, in cells, and in small animals. For the two former applications, fluorescence in the visible range is desirable for compatibility with existing microscopes and plate readers as NIR dyes are typically beyond the detection channels of common fluorescent microscopes. Conversely, red and NIR fluorescence, which has

Page | 45

better tissue penetration and less nonspecific absorption than UV or visible dyes, is desirable for small animal imaging.

Several compounds were synthesized based on the visible emitting coumarin, dansyl, and pentamethine cyanine cores as well as the NIR heptamethine cyanine frame

(Scheme 4). To indicate their similar reactivity to MX, the probes are identified with tags ending with MX.

O O O O O O O O O

O O O H2N NH2 H2N H2N NO2

ACMX MCMX NpCMX

N N N N O N

O S O H NH2 N O HN NH O N O NH2 2 H O O O DansylMX Cy5MX Cy7MX

Scheme 4. Novel fluorescent probes for AP site detection.

The synthesis of two coumarin-based probes began with a piperidine-catalyzed

Knoevenagel condensation between 2,4-dihydroxybenzaldehyde (14) and para- nitrophenylacetonitrile (15) to afford 16 in 81% yield. An excess of 1,3-dibromopropane ensured only minor dimer formation in the SN2 reaction with 16 to afford 17 in 83% yield. A second SN2 reaction between 17 and tert-butylhydroxycarbamate gave the Boc- protected precursor 18 in moderate 48% yield. A quantitative deprotection of the Boc group gave final compound NpCMX, termed NpCMX for Nitrophenyl Coumarin MX. A

Page | 46

palladium catalyzed reduction of the nitro group in NpCMX afforded ACMX, termed

ACMX for Aniline Coumarin MX, in 86% yield (Scheme 5).

Scheme 5. Syntheses of NpCMX and ACMX.

A third coumarin-based probe was synthesized starting with an SN2 reaction between N-hydroxyphthalimide, 19, and 3-bromopropan-1-ol to give 20 in 86% yield. A methanesulfonyl chloride promoted dehydration tethered 7-hydroxy-4-methylcoumarin to the phthalimide-protected arm to give 21 in 49% yield over two steps. Hydrazine deprotection afforded MCMX (Methyl Coumarin MX, MCMX) in 20% overall yield

(Scheme 6).

Page | 47

Scheme 6. Synthesis of MCMX.

Using a modified, two-step procedure, which afforded a higher yield than previously reported,132 N-(3-bromopropyl)phthalimide, 22, and and tert-butyl hydroxycarbamate were coupled to give 23 and subsequent deprotection afforded intermediate 24. An EDC and HOBt amide coupling reaction between 24 and 4- hydroxyphenylacetic acid afforded 25 in 80% yield. An SRN1 reaction between cyanine dye 26 and 25 to give 27, followed by Boc deprotection afforded the heptamethine cyanine Cy7MX (Cy7MX) in 39% yield over two steps (Scheme 7).

Page | 48

Scheme 7. Synthesis of Cy7MX.

2,3,3-Trimethyl-3H-indole (28) and 1-bromopropane were reacted to generate the ammonium salt 29 in 61% yield. Two Knoevenagel condensations between two equivalents of 29 and one molar equivalent of 29 generated the zwitterion 31. An amide coupling between 31 and 24 afforded 32 in 89% yield. Boc deprotection of 32 afforded the pentamethine cyanine probe, Cy5MX (Cy5MX, Scheme 8).

Page | 49

Scheme 8. Synthesis of Cy5MX

A one step substitution reaction between dansyl chloride (33) and, O,O'-(propane-

1,3-diyl)bis(hydroxylamine) hydrochloride (34) afforded dansylMX (DansylMX) in 17% yield (Scheme 9). The yield of DansylMX suffered as a mixture of the title compound, the dimer, and DMF coupled products resulted. The structures of DMF coupled products were confirmed by 1H NMR.

Page | 50

Scheme 9. Synthesis of dansylMX.

A fluorescein derivative called F422 (F422, Marker Gene product #M1036,

Scheme 10) has been used in our labs as a fluorescent probe for AP sites.133 In order to model nonspecific probe binding using F422, a control compound was synthesized, termed FEt2 (FEt2). A one step reaction starting from fluorescein isothiocyanate (35) and diethylamine afforded FEt2 in quantitative yield (Scheme 10).

Scheme 10. Structure of F422 and synthesis of control compound FEt2.

3.3 Fluorescent characterization of probes

To characterize the fluorescent properties, the absorption and emission spectra were collected for all final compounds. Table 3 summarizes the maxima of absorption and emission for these compounds in ethanol. These data were used to optimize detection channels.

Page | 51

Table 3. Summary of absorption and emission wavelengths for AP site binding probes.

Probe Ȝabs max (nm) Ȝem max (nm) Stokes shift (nm)

ACMX 370 450 80

MCMX 325 375 50

NpCMX 365 550 185

DansylMX 340 540 200

Cy5MX 640 660 20

Cy7MX 770 796 20

Further characterization was conducted for Cy7MX (Table 4), Cy7MX, because it was chosen for extensive investigation in biological studies. The absorption and emission spectra were collected in polar protic (water and ethanol), polar aprotic

(acetonitrile), and relatively nonpolar (chloroform) solvents. No solvatochromism was observed for Cy7MX. The extinction coefficient and the fluorescence quantum yields for

Cy7MX compared favorably to other heptamethine cyanine dyes.134-135

Table 4. Fluorescence and absorbance properties of Cy7MX in various solvents.

-1 -1 Solvent İ (M cm ) Ȝabs (nm) Ȝem (nm) ĭ (%)

H2O 140,000 767 790 6.3

EtOH 127,000 770 796 36.3

MeCN 154,000 767 789 45.9

CHCl3 Not determined 770 791 32.6

Page | 52

3.4 Discussion and conclusions

The compounds are synthesized using common intermediates, such as 24 and 18, and similar reaction conditions. This allows for modular synthesis providing easy modification of compounds to adjust properties and develop a library based on these leads. For example, could be homologated to introduce variations of Cy5MX and

Cy7MX. In addition, starting material 26 could be substituted with other heptamethine dyes to improve solubility or bioavailability.

The novel AP site binding probes described in this chapter have fluorescent emissions spanning UV, visible, and NIR. This is advantageous in optical imaging where several channels (i.e. emission wavelengths) can be simultaneously detected. A suitable

AP site binding probe can be selected to complement a costain, such as the ubiquitous

DAPI for nuclear staining, with no overlap. However, the compounds must first be screened for their AP site binding efficacy. Details of this screening can be found in

Chapter 4 (page 73).

AP sites are indicative of DNA damage and repair. The focus of this work has centered on the application of AP site probes in cancer. However, BER is also a major repair pathway in response to damage caused by reactive oxygen species (ROS) with the glycosylases recognizing several forms of oxidative damage (Table 1, page 14). DNA damage by ROS and subsequent BER enzyme repair are implicated in rheumatoid arthritis,136 Alzheimer’s disease,137 Parkinson’s disease,138 and aging.139 X-ray radiation has been shown to produce AP sites as well.140 Therefore, while experiments discussed in chapters five and six (pages 97 and 115, respectively) primarily focus on cancer, the probes and methods investigated have broader application to many diseases.

Page | 53

3.5 Materials and methods

3.5.1 General methods

General methods are reported in §2.6.1 on page 31 and as follows:. NMR spectra were recorded on a Varian Mercury 300 MHz, a Varian Inova 400, and a Varian Inova

600 MHz spectrometer at room temperature, except where otherwise noted. Chemical

1 shifts for H NMR were reported as į, parts per million, and referenced to CHCl3 at 7.26 ppm or TMS at 0 ppm. Chemical shifts for 13C NMR were reported as į and referenced to the center line of the CDCl3 triplet at 77.0 ppm or CH2Cl2 pentet at 54.0 ppm. Low resolution mass spectra were recorded on a ThermoScientific LCQ Advantage or a

Finnigan LCQ Deca mass spectrometer. Fluorescence data were collected on a Varian

Cary Eclipse fluorescence spectrophotometer in a quartz cuvette with a 1 cm pathlength or on a Tecan Infinite M200 scanner. Absorbance data were collected on a Varian Cary

50 Bio UV-Visible Spectrophotometer in a quartz cuvette with a 1 cm pathlength.

3.5.2 Synthesis

7-hydroxy-3-(4-nitrophenyl)-2H-chromen-2-one (16).141 para-Nitrophenylacetonitrile

(15, 597 mg, 3.68 mmol) was added to a 25 mL round bottom flask fitted with a magnetic stir bar. A water reflux condenser was fitted to the round bottom. Ethanol (6.0 mL) was added and the solid dissolved upon gentle warming. 2,4-dihydroxybenzaldehyde (14,

500 mg, 3.62 mmol) was added and allowed to dissolve before piperidine (20.0 μL, 0.202

Page | 54

mmol, 0.862 g/mL) was added. The reaction was stirred at reflux for 4 h. The reaction was cooled to r.t. then filtered. The red solid was suspended in glacial acetic acid (7.5 mL) and stirred at reflux for 16 h. The reaction was cooled to r.t., poured into ice water, and filtered. The solid was collected to afford 16 as a yellow solid (827 mg, 81%)

1 without further purification. Rf = 0.26 (DCM/PhMe/EtOAc, 5:4:1); H NMR (600 MHz, d6-DMSO): į = 10.81 (br s, 1H), 8.39 (s, 1H), 8.29 (d, J = 8.9 Hz, 2H), 8.01 (d, J = 8.9

Hz, 2H), 7.65 (d, J = 8.5 Hz, 1H), 6.85 (dd, J = 8.5, 2.0 Hz, 1H), 6.78 (d, J = 2.0 Hz, 1H);

13 C NMR (150 MHz, d6-DMSO): į = 162.1, 159.5, 155.3, 146.5, 143.1, 141.8, 130.6,

129.2, 123.2, 119.7, 113.7, 111.7, 101.7.

7-(3-bromopropoxy)-3-(4-nitrophenyl)-2H-chromen-2-one (17).142 An oven-dried 25 mL round bottom flask fitted with a magnetic stir bar was cooled under a stream of argon. Coumarin 16 (200 mg, 0.708 mmol) and K2CO3 (203 mg, 1.47 mmol) were added. The flask was evacuated and filled with argon four times. The solids were suspended in dry acetone (9.8 mL) and dibromopropane (600 μL, 5.88 mmol, 1.977 g/mL) was added quickly and all at once. A water reflux condenser was attached and the reaction was stirred at reflux under argon shielded from light for 24 h. The reaction mixture was filtered and the filtrate was concentrated in vacuo. The crude residue was extracted from water with DCM (3x). The combined organic layers were washed with water (2x) and brine, dried over MgSO4, filtered, and concentrated to afford 17 without need for further purification as a yellow solid (236 mg, 83%). Rf = 0.71 (EtOAc/hexanes,

Page | 55

1 1:1); H NMR (600 MHz, d6-DMSO): į = 8.43 (s, 1H), 8.30 (d, J = 8.8 Hz, 2H), 8.02 (d,

J = 8.8 Hz, 2H), 7.74 (d, J = 8.6 Hz, 1H), 7.09 (d, J = 2.2 Hz, 1H), 7.04 (dd, J = 8.6, 2.2

Hz, 1H), 4.22 (t, J = 6.1 Hz, 2H), 3.69 (t, J = 6.4 Hz, 2H), 2.29 (tt, J = 6.4, 6.1 Hz, 2H);

13 C NMR (125 MHz, d6-DMSO): į = 162.1, 159.4, 155.2, 146.7, 142.8, 141.5, 130.3,

129.3, 123.2, 120.9, 113.2, 112.9, 100.8, 66.2, 31.5, 30.9.

tert-butyl (3-((3-(4-nitrophenyl)-2-oxo-2H-chromen-7-yl)oxy)propoxy)carbamate

(18). An oven-dried 50 mL round bottom flask fitted with a magnetic stir bar was cooled under argon. Sodium hydride (399 mg, 16.6 mmol) and tert-butylhydroxycarbamate

(1.97 g, 14.8 mmol) were added and the flask was purged with argon 5 times. Dry DMF

(25 mL) was added and the reaction was stirred at r.t. under argon for 70 minutes. To an oven-dried 100 mL round bottom flask fitted with a magnetic was added coumarin 17

(1.00 g, 2.47 mmol). The flask was purged with argon five times before dry DMF (12.5 mL) was added. The sodium hydride mixture was added to the coumarin solution dropwise. The 50 mL round bottom flask was rinsed with dry DMF (2 x 6 mL) and the rinsate added to the reaction mixture. The reaction was stirred at room temperature under argon in the dark for 3 h. The reaction was quenched with water (125 mL) then 10% HCl

(15 mL). The reaction mixture was extracted with EtOAc (1 x 50 mL, 2 x 75 mL). The combined organic layer were washed with water (2 x 50 mL) and brine (50 mL), dried over MgSO4, filtered, and concentrated. The crude residue was dissolved in DCM and

Page | 56

dry loaded onto silica. The crude residue was purified by silica gel chromatography with a mobile phase of 12:7:1 DCM/hexanes/EtOAc gradually increasing to 6:3:1

DCM/hexanes/EtOAc. Concentration afforded 18 as a yellow powder (549 mg, 48%).

1 Rf = 0.15 (EtOAc/hexanes, 3:7); H NMR (400 MHz, CDCl3): į = 8.29 (d, J = 8.8 Hz,

2H), 7.91 (d, J = 8.8 Hz, 2H), 7.89 (s, 1H), 7.48 (d, J = 8.4 Hz, 1H), 7.15 (br s, 1H), 6.92

(dd, J = 8.4, 2.4 Hz, 1H), 6.89 (d, J = 2.4 Hz, 1H), 4.22 (t, J = 6.0 Hz, 2H), 4.07 (t, J = 6.0

13 Hz, 2H), 2.17 (tt, J = 6.0, 6.0 Hz, 2H), 1.49 (s, 9H); C NMR (125 MHz, d6-DMSO): į

= 162.3, 159.4, 156.1, 155.2, 146.7, 142.9, 141.6, 130.3, 129.3, 123.3, 120.9, 113.2,

112.7, 100.7, 79.5, 71.7, 68.9, 28.0, 27.3.

7-(3-(aminooxy)propoxy)-3-(4-nitrophenyl)-2H-chromen-2-one (NpCMX). To a 100 mL round bottom flask fitted with a magnetic stirbar was added 18 (339 mg, 0.742 mmol) and DCM (50.0 mL). After the coumarin dissolved completely, TFA (1.0 mL,

13.0 mmol, 1.48 g/mL) was added dropwise while stirring. The reaction was stirred at r.t. in the dark for 23 h. Saturated aqueous NaHCO3 (10 mL) was added to quench the reaction. The reaction was then extracted with 40 mL water. The reaction was extracted again with fresh DCM (2 x 15 mL). The combined organic layers were washed with water (2 x 40 mL) and brine (1 x 40 mL). The crude reaction mixture was dried over

MgSO4, filtered and concentrated to afford NpCMX without further purification as a

1 yellow solid (268 mg, quant.). Rf = 0.29 (EtOAc/hexanes, 3:1); H NMR (400 MHz,

CDCl3): į = 8.29 (d, J = 8.8 Hz, 2H), 7.91 (m, 3H), 7.48 (d, J = 8.6 Hz, 1H), 6.91 (dd, Page | 57

J = 8.6, 2.3 Hz, 1H), 6.88 (d, J = 2.3 Hz, 1H), 5.43 (br s, 2H), 4.15 (t, J = 6.3 Hz, 2H),

13 3.87 (t, J = 6.1 Hz, 2H), 2.15 (tt, J = 6.3, 6.1 Hz, 2H); C NMR (100 MHz, CDCl3): į =

162.9, 160.2, 147.4, 141.8, 141.5, 129.4, 129.2, 123.6, 122.2, 113.7, 112.8, 101.0, 71.9,

65.6, 28.1.

7-(3-(aminooxy)propoxy)-3-(4-aminophenyl)-2H-chromen-2-one (ACMX, ACMX).

Palladium on carbon (10 wt%, 15 mg, 0.014 mmol) and coumarin NpCMX were added to a 100 mL round bottom flask fitted with a magnetic stir bar. The flask was purged with argon three times. The coumarin was dissolved in DMF (12 mL) then EtOH (200 proof,

12 mL) was added to the reaction mixture. The flask was purged twice with hydrogen gas (1 atm). The reaction was stirred in the dark at r.t. for 2.5 h in the dark. The reaction mixture was diluted with EtOAc and filtered through a plug of celite. The filtrate was concentrated under vacuum then diluted with water (75 mL). The crude reaction mixture was extracted with EtOAc (1 x 50 mL, 2 x 25 mL). The combined organic layers were washed with water (2 x 50 mL) and brine (1 x 50 mL). The organic layer was dried over

MgSO4, filtered, and concentrated. The crude residue was diluted in DCM then dry loaded on silica. The product was purified by silica gel chromatography with a mobile phase of 75:25:2 EtOAc/hexanes/Et3N gradually increasing to 50:1 EtOAc/Et3N.

Concentration afforded ACMX as a yellow solid (78.8 mg, 86%). Rf = 0.14

1 (EtOAc/hexanes/Et3N, 75:25:2); H NMR (600 MHz, d6-DMSO): į = 8.00 (s, 1H), 7.62 Page | 58

(d, J = 8.6 HZ, 1H), 7.44 (d, J = 8.6 Hz, 2H), 6.96 (d, J = 2.3 Hz, 1H), 6.93 (dd, J = 8.6,

2.3 Hz, 1H), 6.60 (d, J = 8.6 Hz, 2H), 5.38 (s, 2H), 4.66 (t, J = 5.3 Hz, 1H), 4.13 (t, J =

6.3 Hz, 2H), 3.56 (dt, J = 6.1, 5.3 Hz, 2H), 1.88 (tt, J = 6.3, 6.1 Hz, 2H); 13C NMR (125

MHz, d6-DMSO): į = 160.9, 160.2, 154.0, 149.0, 137.0, 128.9, 123.5, 123.4, 121.7,

113.3, 113.2, 112.6, 100.4, 68.9, 57.0, 31.8.

2-(3-hydroxypropoxy)isoindoline-1,3-dione (20).143 To an oven-dried 100 mL round bottom flask fitted with a magnetic stir bar and cooled under a stream of argon was added

N-hydroxyphthalimide (19, 2.00 g, 12.3 mmol), 3-bromo-1-propanol (3.20 mL, 36.6 mmol, 1.59 g/mL) and sodium acetate (3.05 g, 37.2 mmol). The flask was evacuated and purged with argon. Dry DMSO (36.0 mL) was added and the reaction was warmed to 70

°C. The reaction was stirred overnight under argon. The reaction was cooled to r.t. and quenched with water. The crude reaction mixture was extracted with DCM (3x). The combined organic layers were washed with water (2x) and brine, dried over MgSO4, filtered, and concentrated. The crude residue was loaded neat onto a silica gel column and purified with a mobile phase of 9:1 DCM/EtOAc. Concentration afford 20 as a white

1 solid (2.32 g, 86%). Rf = 0.22 (DCM/EtOAc, 9:1); H NMR (400 MHz, CDCl3): į =

7.87-7.82 (m, 2H), 7.78-7.74 (m, 2H), 4.38 (t, J = 6.0 Hz, 2H), 3.94 (dt, J = 6.0, 6.0 Hz,

2H), 2.58 (t, J = 6.0 Hz, 1H), 2.01 (tt, J = 6.0, 6.0 Hz, 2H).

Page | 59

2-(3-((4-methyl-2-oxo-2H-chromen-7-yl)oxy)propoxy)isoindoline-1,3-dione (21).144

To an oven-dried 100 mL round bottom flask fitted with a magnetic stir bar and cooled under a stream of nitrogen was added N-(3-hydroxypropoxy)phthalamide (20, 1.00 g,

4.53 mmol). The flask was purged with argon. The solid was dissolved in dry pyridine

(50 mL, 620 mmol, 0.978 g/mL). Methanesulfonyl chloride (950 μL, 12.3 mmol, 1.48 g/mL) was added and the reaction stirred at r.t. under argon for 2 h. Water was added to quench the reaction and the mixture was quickly extracted with EtOAC (2x). The combined organic layers were washed with water and brine, dried over MgSO4, filtered, and concentrated. 7-Hydroxy-4-methylcoumarin (519 mg, 2.95 mmol) and potassium carbonate were added. The mixture was suspended in dry DMF and heated to 50 °C.

The reaction was stirred under argon overnight. The reaction was cooled and extracted with water and EtOAc (3x). The combined organic layers were washed with water (2x) and brine, dried over MgSO4, filtered, and concentrated. The crude residue was purified by silica gel chromatography with a mobile phase of 9:1 DCM/EtOAc. Concentration afforded 21 as a white solid (553 mg, 49%, 60% based on recovered starting material).

1 Rf = 0.47 (DCM/EtOAc, 9:1); H NMR (400 MHz, CDCl3): į = 7.87-7.82 (m, 2H), 7.78-

7.74 (m, 2H), 7.51 (d, J = 8.8 Hz, 1H), 6.90 (dd, J = 8.8, 2.4 Hz, 1H), 6.87 (d, J = 2.4 Hz,

1H), 6.14 (q, J = 1.2 Hz, 1H), 4.43 (t, J = 6.2 Hz, 2H), 4.33 (t, J = 6.2 Hz, 2H), 2.40 (d,

Page | 60

13 J = 1.2 Hz, 3H), 2.3 (tt, J = 6.2, 6.2 Hz, 2H); C NMR (100 MHz, CDCl3): į = 161.8,

161.3, 155.2, 152.5, 134.7, 128.8, 125.5, 123.6, 113.7, 112.4, 112, 101.7, 73.7, 66.3, 37.1,

28.1, 18.7.

7-(3-(aminooxy)propoxy)-4-methyl-2H-chromen-2-one (MCMX,). Phthalamide 21

(165 mg, 0.435 mmol) was added to a 100 mL round bottom flask fitted with a magnetic stir bar. The solid was dissolved in DCM (20 mL) and EtOH (20 mL). Hydrazine monohydrate (530 μL, 10.9 mmol, 1.032 g/mL) was added all at once. The reaction was stirred at r.t. loosely capped (to retard evaporation) overnight while shielded from light.

The solvents were removed in vacuo. The crude residue was suspended in water and extracted with DCM (3x). The combined organic layers were washed with water (2x) and brine, dried over MgSO4, filtered and concentrated to afford MCMX as a colorless

1 solid (75.8 mg, 70%). Rf = 0.41 (DCM/hexanes/Et3N, 40:20:1); H NMR (600 MHz, d6-

DMSO): į = 7.67 (d, J = 9.4 Hz, 1H), 6.96 (m, 2H), 6.20 (d, J = 0.8 Hz, 1H), 4.13 (t, J =

6.4 Hz, 2H), 3.67 (t, J = 6.2 Hz, 2H), 2.39 (s, 3H), 1.98 (tt, J = 6.4, 6.2 Hz, 2H); 13C NMR

(125 MHz, d6-DMSO): į = 161.6, 160.0, 154.6, 153.3, 126.4, 113.0, 112.3, 111.0, 101.1,

71.0, 65.4, 27.6, 18.0.

Page | 61

tert-butyl 3-(1,3-dioxoisoindolin-2-yl)propoxycarbamate (23).132 22 (10.1 g, 37.5 mmol) and tert-butyl hydroxycarbamate (9.99 g, 75.0 mmol) were added to a dry 250 mL round bottom flask fitted with a stir bar. The flask was sealed with a rubber septum then the atmosphere was evacuated and refilled with argon 5 times. The reagents were dissolved in anhydrous DCM (60.0 mL), added via syringe. DBU was then added via a syringe and the reaction was stirred under argon at room temperature. After five hours, the reaction was quenched with 10% citric acid (50 mL) and extracted with DCM (3 x 30 mL). The combined organic layers were washed with 10% citric acid (2 x 50 mL), water

(50 mL), then brine (50 mL). The organic layer was dried over MgSO4, filtered, and concentrated. The crude residue was diluted in a trace amount of DCM and purified by silica gel chromatography with a mobile phase of pure DCM then gradually increasing polarity to 9:1 DCM/EtOAc. Concentration gave 23 as a white solid (8.14 g, 68%). Rf =

1 0.45 (DCM/EtOAc, 9:1); H NMR (400 MHz, CDCl3): į = 7.85-7.81 (m, 2H), 7.73-7.68

(m, 2H), 7.34 (br s, 1H), 3.91 (t, J = 6.4 Hz, 2H), 3.81 (t, J = 6.7 Hz, 2H), 1.99 (tt, J = 6.7,

13 6.4 Hz, 2H), 1.46 (s, 9H); C NMR (100 MHz, CDCl3): į = 168.4, 156.8, 133.9, 132.0,

123.2, 81.7, 73.9, 35.0, 28.2, 27.2.

Page | 62

tert-butyl 3-aminopropoxycarbamate (24). 24 was prepared following modification of a literature procedure.132 Briefly, 23 (1.95 g, 6.10 mmol) was added to a 250 mL round bottom flask fitted with a magnetic stir bar. Methanol (100 mL) was added and the mixture was stirred until the solid was completely dissolved. Hydrazine hydrate (6.0 mL,

124 mmol, 1.032 g/mL) was added all at once while rapidly stirring. The reaction was stirred overnight at room temperature. The next morning, a white precipitate had formed.

The methanol was removed in vacuo. The remaining residue was suspended in CHCl3 and filtered. The solid was washed several times with CHCl3 before the filtrate was transferred to a separatory funnel, diluted with water, and extracted. The water was extracted twice more with fresh CHCl3. The organic layers were combined and washed twice with water and once with brine. The organic layer was dried over Na2SO4, filtered, and concentrated to afford 24 as a pale yellow oil (1.02 g, 88%), which was used without

1 further purification. H NMR (400 MHz, CDCl3): į = 3.95 (t, J = 6.1 Hz, 2H), 2.85 (t, J

13 = 6.5 Hz, 2H), 1.77 (tt, J = 6.5, 6.1 Hz, 2H), 1.47 (s, 9H); C NMR (100 MHz, CDCl3):

į = 156.8, 80.8, 74.3, 38.8, 31.0, 28.0.

Page | 63

tert-butyl 3-(2-(4-hydroxyphenyl)acetamido)propoxycarbamate (25). 4- hydroxyphenylacetic acid (350 mg, 2.30 mmol), 24 (416 mg, 2.19 mmol), EDC·HCl (629 mg, 3.28 mmol), and HOBt·H2O (503 mg, 3.28 mmol) were added to an oven-dried 50 mL round bottom flask fitted with a magnetic stir bar. The solids were dissolved in dry

DMF (20.0 mL) and the reaction was stirred at room temperature under argon for 24 h.

The reaction was then diluted with water (100 mL) and EtOAc (50 mL) and extracted.

The aqueous layer was extracted twice more with EtOAc (25 mL). The combined organic layers were washed with water (2 x 40 mL) and brine (40 mL) then dried over

MgSO4, filtered, and concentrated. The crude residue was diluted in a trace amount of

DCM and purified by silica gel chromatography with a mobile phase of pure DCM then gradually increasing polarity to 3:2 DCM/MeCN. This product was then further purified by diluting it in EtOAc (25 mL) and washing with sat. NaHCO3 (3 x 25 mL) and brine (1 x 25 mL). The organic layer was dried over MgSO4, filtered, and concentrated to afford

1 pure 25 as a white solid (527 mg, 74%). Rf = 0.23 (DCM/MeCN, 3:2); H NMR (400

MHz, CDCl3): į = 8.64 (br s, 1H), 8.21 (br s, 1H), 7.34 (br s, 1H), 7.07 (d, J = 8.6 Hz,

2H), 6.74 (d, J = 8.6 Hz; 2H), 3.84 (t, J = 5.6 Hz, 2H), 3.46 (s, 2H), 3.35 (dt, J = 6.0, 12

13 Hz, 2H), 1.71 (tt, J = 6.0, 5.6 Hz, 2H), 1.47 (s, 9H); C NMR (100 MHz, CDCl3): į =

172.8, 157.3, 155.7, 130.0, 125.7, 115.6, 81.6, 74.6, 42.4, 37.0, 28.0, 26.8.

Page | 64

2-((E)-2-((E)-3-((E)-2-(3,3-dimethyl-1-propylindolin-2-ylidene)ethylidene)-2-(4-(2,2- dimethyl-4,11-dioxo-3,6-dioxa-5,10-diazadodecan-12-yl)phenoxy)cyclohex-1-en-1- yl)vinyl)-3,3-dimethyl-1-propyl-3H-indol-1-ium chloride (27). NaH (14.1 mg, 0.588 mmol) and 25 (129 mg, 0.398 mmol) were added to an oven-dried 25 mL round bottom flask fitted with a magnetic stir bar. Dry DMF (3.0 mL) was added and the mixture was stirred under argon at room temperature for 30 min. Meanwhile, IR-780 iodide (26,

230.1 mg, 0.345 mmol) was added to a 15 mL oven-dried heart-shaped flask fitted with a magnetic stir bar. DMF (5.0 mL) was added and 26 was stirred under argon at room temperature shielded from light. After 30 min, the solution of 26 was transferred to the

NaH mixture via syringe. The heart-shaped flask was rinsed with dry DMF (4 x 2 mL) and the rinsate was added to the reaction mixture via syringe. The reaction was stirred at room temperature under argon in the dark for 5 h. The reaction was quenched with water

(100 mL) and 10% NH4Cl (aqueous, 50 mL). The aqueous mixture was extracted with

DCM (1 x 100 mL, 1 x 25 mL) until the aqueous layer remained colorless. The combined organic layers were washed with water (2 x 50 mL) and brine (1 x 50 mL), dried over MgSO4, filtered, and concentrated. The crude residue was diluted in a minimal amount of eluent and purified by silica gel chromatography with a mobile phase of 5:20:175 MeOH/MeCN/DCM gradually increasing to 10:15:75 MeOH/MeCN/DCM.

Page | 65

Impure fractions were concentrated and this chromatographic method was repeated one time. Pure fraction were combined and concentrated to afford 27 as an emerald solid

1 (163 mg, 55%). Rf = 0.15 (MeOH/MeCN/DCM, 1:3:16); H NMR (400 MHz, CDCl3): į

= 8.94 (br s, 1H), 8.78 (br t, J = 5.6 Hz, 1H), 7.94 (d, J = 14.0 Hz, 2H), 7.55 (d, J = 8.4

Hz, 2H), 7.35-7.28 (m, 4H), 7.19 (dd, J = 7.6, 7.2 Hz, 2H), 7.05 (d, J = 8.0 Hz; 2H), 6.95

(d, J = 8.8 Hz, 2H), 5.95 (d, J = 14.0 Hz, 2H), 3.96 (t, J = 7.2 Hz, 4H), 3.92 (t, J = 5.6 Hz,

2H), 3.62 (s, 2H), 3.28 (dt, J = 6.0, 6.0 Hz, 2H), 2.67 (dd, J = 6.0, 5.6 Hz, 4H), 2.04 (dd, J

= 6.0, 5.6 Hz, 2H), 1.86 (tq, J = 7.6, 7.2Hz, 4H), 1.70 (tt, J = 6.0, 5.6Hz, 2H), 1.45 (s,

13 9H), 1.32 (s, 12H), 1.04 (t, J = 7.6 Hz, 6H); C NMR (75 MHz, CDCl3): į = 172.1,

171.6, 164.8, 158.3, 156.8, 142.3, 141.9, 140.8, 131.3, 131.2, 128.3, 124.9, 122.1, 121.7,

114.1, 110.2, 99.2, 80.3, 73.5, 48.9, 45.6, 42.3, 35.7, 28.1, 27.6, 27.2, 24.1, 20.9, 20.5,

+ + 11.4; MS-ESI: m/z [M] calcd for C52H67N4O5 : 827.51, found: 827.47.

2-((E)-2-((E)-2-(4-(2-((3-(aminooxy)propyl)amino)-2-oxoethyl)phenoxy)-3-((E)-2-

(3,3-dimethyl-1-propylindolin-2-ylidene)ethylidene)cyclohex-1-en-1-yl)vinyl)-3,3- dimethyl-1-propyl-3H-indol-1-ium 2,2,2-trifluoroacetate (Cy7MX). 27 (23.0 mg,

0.027 mmol) was dissolved in DCM (1.0 mL). TFA (1.0 mL, 13.0 mmol, 1.48 g/mL) was added and the solution immediately turned from dark green to dark red. The reaction

Page | 66

was stirred for 1 h in the dark. The solvents were removed by rotary evaporation to afford pure Cy7MX as an emerald green solid (16 mg, 71%). Rf = 0.21 (MeOH/DCM,

1 1:9); H NMR (600 MHz, CDCl3): į = 8.93-8.20 (br s, 2H), 8.0-7.87 (m, 3H), 7.37-7.26

(m, 6H), 7.19 (dd, J = 7.8, 7.2 Hz, 2H), 7.02 (d, J = 7.8 Hz, 2H), 6.97(J = 8.4 Hz, 2H),

5.93 (bs d, J = 12.6 Hz, 2H), 4.09 (br s, 2H), 3.91 (bs t, J = 7.2 Hz, 4H), 3.48 (br s, 2H),

3.23 (br m, 2H), 2.65 (br s, 4H), 2.02 (br m, 2H), 1.83 (tq, J = 7.2, 7.2, 4H), 1.67 (br s,

13 2H), 1.30 (s, 12H), 1.02 (t, J = 7.2 Hz, 6H); C NMR (150 MHz, CD2Cl2, -10°C): į =

174.2, 172.4, 164.7, 161.0, 159.2, 142.5, 141.3, 131.2, 129.6, 128.7, 125.3, 122.6, 122.1,

115.1, 110.7, 99.8, 72.2, 49.3, 46.0, 41.9, 35.8, 27.7, 27.6, 24.4, 21.3, 20.9, 11.7; UV/Vis

í1 í1 í1 í1 (EtOH): Ȝmax (İ)=705 nm (29200 L·mol ·cm ), 770 nm (127000 L·mol ·cm );

+ + HRMS-ESI: m/z [M] calcd for C47H59N4O3 : 727.4587, found: 727.4615.

2,3,3-trimethyl-1-propyl-3H-indol-1-ium bromide (29).145-146 Sodium iodide (476 mg,

3.18 mmol) was added to a 120 mL pressure vessel fitted with a magnetic stir bar then suspended in MeCN (40 mL). 2,3,3-trimethyl-3H-indole (28, 5.0 mL, 31.2 mmol, 0.992 g/mL) and 1-bromopropane (14.2 mL, 156 mmol, 1.353 g/mL) were added. The vessel was sealed and heated to 85 °C while stirring for 21 h. A yellow precipitate quickly formed and the supernatant slowly darkened overnight. The reaction mixture was then transferred to a 500 mL round bottomed flask with some absolute ethanol and the solvents were removed in vacuo. The crude residue was rinsed several times with diethyl

Page | 67

ether then recrystallized in acetone. The crystals were filtered and washed three times with diethyl ether to yield 29 as lavender crystals (5.36 g, 61%). Rf = 0.41 (DCM/MeOH,

1 9:1); H NMR (400 MHz, CD3OD): G 7.94-7.90 (m, 1H), 7.82-7.77 (m, 1H), 7.68-7.64

(m, 1H), 4.81 (s, 3H), 4.54 (dd, J = 7.6, 7.6 Hz, 2H), 2.03 (tq, J = 7.6, 7.6 Hz, 2H), 1.63

13 (s, 6H), 1.11 (t, J = 7.6 Hz, 3H); C NMR (100 mHz, CD3OD): G 197.9, 143.4, 142.5,

131.1, 130.5, 124.6, 116.6, 67.0, 56.0, 50.7, 22.9, 22.5, 11.3.

6-((1E,3Z)-1-(3,3-dimethyl-1-propyl-3H-indol-1-ium-2-yl)-5-((E)-3,3-dimethyl-1- propylindolin-2-ylidene)penta-1,3-dien-3-yl)nicotinate (31). To an oven-dried 48 mL pressure vessel fitted with a magnetic stir bar was added 29 (292 mg, 1.03 mmol), 30

(101 mg, 0.523 mmol), and sodium acetate (130 mg, 1.58 mmol). The solids were suspended in acetic anhydride (4.9 mL, 51.9 mmol, 1.082 g/mL). The vessel was tightly sealed and heated to 110 °C, while stirring, shielded from light for 2 h. The reaction was cooled to r.t. and quenched with water (50 mL). The reaction mixture was extracted with

DCM (3 x 25 mL). The combined organic layers were washed with water (2 x 25 mL) and brine (25 mL), dried over MgSO4, filtered, and concentrated. The crude residue was purified by silica gel chromatography with a mobile phase of 100% DCM increasing to

19:1 DCM/MeOH then 9:1 DCM/MeOH. Concentration afforded 31 as a dark blue solid

1 (50.3 mg, 36%). Rf = 0.29 (DCM/MeOH, 9:1); H NMR (400 MHz, CDCl3): į = 9.61 (s,

Page | 68

1H), 8.83 (br d, J = 6.0 Hz, 1H), 8.02 (d, J = 14.0 Hz, 2H), 7.42-7.24 (m, 7H), 7.08 (d, J =

8.0 Hz, 2H), 5.88 (d, J = 14.0 Hz, 2H), 3.72 (t, J = 6.6 Hz, 4H), 1.78 (s, 12H), 1.71 (tq, J

13 = 7.2, 6.6 Hz, 4H), 0.84 (t, J = 7.2 Hz, 6H); C NMR (100 MHz, CDCl3): į = 173.7,

170.9, 155.7, 152.5, 152.0, 141.4, 140.9, 138.5, 132.1, 130.4, 128.2, 125.0, 124.3, 122.0,

110.5, 100.8, 49.2, 45.1, 27.7, 20.3, 11.0.

2-((1E,3Z)-3-(5-((3-(((tert-butoxycarbonyl)amino)oxy)propyl)carbamoyl)pyridin-2- yl)-5-((E)-3,3-dimethyl-1-propylindolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-

1-propyl-3H-indol-1-ium chloride (32). To an oven dried 15 mL round bottom flask fitted with a magnetic stir bar was added 31 (53.0 mg, 0.095 mmol), 24 (22.1 mg, 0.116 mmol), EDC hydrochloride (27.7 mg, 0.144 mmol), and HOBt monohydrate (23.5 mg,

0.153 mmol). The solids were dissolved in dry DMF (5.0 mL). The reaction was stirred under argon at r.t. shielded from light overnight. The reaction was diluted with water (25 mL) and extracted with DCM (3 x 25 mL). The combined organic layers were washed with water (3 x 25 mL) and brine (25 mL), dried over Na2SO4, filtered, and concentrated.

The crude residue was purified by silica gel chromatography with a mobile phase of

16:3:1 DCM/MeCN/MeOH. The column and collected fractions were shielded from light during purification. Concentration afforded 32 as a dark blue solid (64.4 mg, 89%).

1 Rf = 0.31 (DCM/MeOH, 9:1); H NMR (400 MHz, CDCl3): G 9.66 (br t, J = 5.6 Hz,

Page | 69

1H); 9.45 (d, J = 1.6 Hz, 1H), 9.20 (br s, 1H), 9.02 (dd, J = 8.0, 1.6 Hz, 1H), 8.03 (d, J =

14.2 Hz, 2H), 7.47 (d, J = 8.0 Hz, 1H), 7.42-7.25 (m, 6H), 7.09 (d, J = 7.6 Hz, 2H), 5.90

(d, J = 14.2 Hz, 2H), 4.11 (t, J = 5.6 Hz, 2H), 3.80 (br t, J = 7.0 Hz, 4H), 3.73 (dt, J = 6.0,

5.6 Hz, 2H), 2.04 (tt, J = 6.0, 5.6, 2H), 1.79 (s, 12H), 1.74 (tq, J = 7.2, 7.0 Hz, 4H), 1.47

13 (s, 9H), 0.86 (t, J = 7.2 Hz, 6H); C NMR (150 mHz, CDCl3): į = 173.7, 165.8, 157.0,

155.7, 152.1, 150.8, 142.0, 141.0, 137.2, 132.3, 129.9, 128.8, 125.6, 125.0, 122.2, 111.0,

101.5, 80.6, 74.4, 49.4, 45.9, 37.1, 28.3, 28.2, 27.6, 20.8, 11.5.

2-((1E,3Z)-3-(5-((3-(aminooxy)propyl)carbamoyl)pyridin-2-yl)-5-((E)-3,3-dimethyl-

1-propylindolin-2-ylidene)penta-1,3-dien-1-yl)-3,3-dimethyl-1-propyl-3H-indol-1- ium 2,2,2-trifluoroacetate (Cy5MX, Cy5MX). To a 4 dram vial fitted with a magnetic stir bar was added 32 (18.1 mg, 0.024 mmol). The solid was dissolved in DCM (1.0 mL) and trifluoroacetic acid (1.0 mL, 13 mmol, 1.48 g/mL) was added. The reaction was shielded from light and stirred at r.t. for 1 h. The reaction was concentrated in vacuo to afford Cy5MX as dark blue solid (18.0 mg, quantitative) without further purification. Rf

1 = 0.066 (DCM/MeCN/MeOH, 16:3:1); H NMR (500 MHz, CDCl3): į = 9.51 (br s, 1H),

9.06 (br s, 1H), 8.76 (br s, 1H), 8.08-8.05 (m, 4H), 7.42 (m, 4H), 7.33 (m, 2H), 7.18 (m,

2H), 5.90 (br s, 2H), 4.34 (br s, 2H), 3.90 (m, 4H), 3.68 (br s, 2H), 2.05 (br s, 2H), 1.73

13 (asymmetric br s, 16H), 0.87 (br s, 6H); C NMR (125 mHz, CDCl3): į = 176.1, 163.2,

Page | 70

160.5 (q, J = 39 Hz), 152.2, 144.4, 143.2, 141.3, 131.2, 128.9, 126.8, 122.3, 118.7, 116.4,

114.5 (q, J = 284 Hz), 114.2, 111.9, 111.7, 99.9, 73.3, 50.2, 46.0, 37.1, 27.4, 27.1, 20.8,

10.8,

N-(3-(aminooxy)propoxy)-5-(dimethylamino)naphthalene-1-sulfonamide

(DansylMX). To an oven-dried 25 mL round bottom flask fitted with a magnetic stir bar was added 34 (245 mg, 1.37 mmol). The flask was evacuated and purged with argon.

The solid was suspended in dry DMF (4.0 mL). To an oven-dried 5 mL pear-shaped flask was added dansyl chloride (33, 403 mg, 1.49 mmol). The solid was dissolved in dry

DMF (2.0 mL). The dansyl chloride solution was added to the 25 mL round bottom flask. The pear-shaped flask was rinsed with dry DMF (5 x 2mL) and the rinsate was added to the reaction mixture. The reaction was stirred under argon at r.t. overnight. The crude reaction mixture was diluted with water and extracted with EtOAc (3x). The combined organic layers were washed with water (2x) and brine, dried over MgSO4, filtered, and concentrated. The crude residue was purified by silica gel chromatography with a mobile phase of 7:3 DCM/EtOAc. Concentration afforded DansylMX as a yellow

1 solid (79 mg, 17%). Rf = 0.26 (DCM/EtOAc, 7:3); H NMR (400 MHz, CDCl3): į =

8.57 (d, J = 8.8 Hz, 1H), 8.31 (d, J = 7.2 Hz, 1H), 8.27 (d, J = 8.4 Hz, 1H), 7.56-7.51 (m,

2H), 7.16 (d, J = 7.6 Hz, 1H), 3.90 (t, J = 6.4 Hz, 2H), 3.38 (t, J = 6.4 Hz, 2H), 2.86 (s,

Page | 71

13 6H), 1.74 (tt, J = 6.4, 6.4 Hz, 2H); C NMR (100 MHz, CDCl3): į = 152.0, 132.1, 131.6,

131.5, 130.0, 129.7, 128.8, 123.2, 118.3, 115.3, 74.2, 72.2, 45.4, 27.0.

5-(3,3-diethylthioureido)-2-(6-hydroxy-3-oxo-3H-xanthen-9-yl)benzoic acid (FEt2).

Fluorescein isothiocyanate (35, 143 mg, 0.367 mmol) was added to a 25 mL pear-shaped flask and dissolved in methanol (20 mL). Diethylamine (200 μL, 1.93 mmol, 0.707 g/mL) was added all at once. The reaction was stirred at r.t. overnight shielded from light. The solvent was removed in vacuo to afford FEt2 (FEt2) as an orange powder

1 (196 mg, quant.). Rf = 0.12 (DCM/MeOH/AcOH, 95:5:2); H NMR (300 MHz,

CD3OD): į = 7.95 (s, 1H), 7.61 (d, J = 8.1 Hz, 1H), 7.19-7.14 (m, 3H), 6.55-6.51 (m,

13 4H), 3.87 (q, J = 6.9 Hz, 4H), 1.31 (t, J = 6.9 Hz, 6H); C NMR (75 MHz, CD3OD): į =

181.8, 181.4, 161.1, 160.1, 143.1, 141.1, 132.7, 131.5, 130.6, 128.9, 128.7, 123.7, 113.7,

104.3, 46.7, 13.1.

3.5.3 Fluorescence quantum yield measurements

Fluorescent quantum yields were determined by a comparative method to an indocyanine green (ICG) standard (ĭ = 0.132 in EtOH)134 with refractive index correction according to Equation 1:147

Page | 72

ܨ ͳെͳͲି஺ೃ ݊ଶ ߔൌߔோ ൬ ൰ ቆ ି஺ ቇቆ ଶቇ ܨோ ͳെͳͲ ݊ோ

Equation 1 where ĭ is the quantum yield; F is the integrated fluorescence intensity (area); A is the absorbance; n is the refractive index of the solvent (ethanol, 1.3611; acetonitrile, 1.3442; water, 1.333; and chloroform, 1.4459);148 and R is the reference sample, ICG in EtOH.

Sample and reference fluorophores were excited at 700 nm and emission spectra integrated from 705 to 1000 nm and absorption measurements were collected at 700 nm.

Data were collected in a quartz cuvette with a 1 cm pathlength. Absorbance maxima for each sample were kept at or below 0.2 absorbance units to avoid inner filter effects.

Solvents used were spectroscopic grade ethanol and HPLC grade acetonitrile, HPLC grade chloroform, and HPLC grade water. For the water solution, Cy7MX was first dissolved first in MeCN and diluted at least 200 fold.

CHAPTER 4. FLUORESCENTLY-TAGGED DNA OLIGOMER SSB ASSAY FOR

PROBE EVALUATION

4.1 Hypothesis and methodology

The AP site binding probes were designed to bind to DNA in the complex environment of the living cell. Initial evaluation of the probes was conducted on a model system that removed many complications of the cell such as dynamic repair, metabolism, membrane permeability, etc. The initial screen was designed to ensure that the probes

Page | 73

could bind to AP sites before slowly introducing additional components, which could be potential confounding variables.

The probe screening assay was developed using a fluorescently-tagged double stranded DNA (dsDNA) oligomer. This 40mer was synthesized containing a fluorophore on the 5’ end of the sense strand. The sense strand also contained a deoxyuridine (U) as the 17th base from the 5’ end. The antisense strand was unlabeled. Treatment of this oligomer with UDG was expected to produce an AP site from the uracil nearly quantitatively. Addition of APE would create a single strand break (SSB) 5’ of the AP site (Figure 14, complex II, page 45). Denaturing polyacrylamide gel electrophoresis

(PAGE) would separate the DNA into single stranded DNA (ssDNA). Fluorescence imaging of the gel would reveal the 16mer truncated sense strand fragment, but the remaining 23mer sense strand and the antisense strand, without fluorescent tags, would be invisible to fluorescence detection (Figure 15, path A).

U UDG APE Probe A ss 40mer U:X ds40mer AP DNA

+Probe B ss 16mer

Denaturing Gel Electrophoresis Probe Probe APE = fluorophore

Figure 15. Schematic of the fluorescently-tagged DNA oligomer SSB activity assay for probe evaluation. A dsDNA 40mer labeled with a 5’ fluorophore and a uracil on the sense strand is treated with UDG to make a 1:1 stoichiometric AP site per oligomer. Path A shows that treatment with APE creates a SSB. Path B shows that an AP site-binding probe blocks the natural substrate of APE and its SSB activity. Denaturing PAGE resolves the fluorescently labeled 40mer and 16mer. The unlabeled 23mer and 40mer antisense strand are not detected by fluorescence imaging.

I hypothesized that treatment of the AP-DNA with an AP site binding probe could mask the natural substrate for APE and block its SSB activity. In this case, denaturing

Page | 74

PAGE would separate the intact sense strand from the antisense strand and only the former would be detected by fluorescence imaging (Figure 15, path B). A mixture of intact 40mer and truncated 16mer could be resolved as two separate bands in the denaturing PAGE to indicate the degree of SSB activity. If the hypothesis is correct, this

SSB activity assay could be used as a tool to screen AP site binding probes.

Initial control experiments were conducted to support the SSB activity assay hypothesis. Hexachlorofluorescein (HEX) and a pentamethine cyanine dye known as

Cy5 were chosen as 5’ fluorescent labels because of their complementary absorption and emission with red/NIR and UV-vis AP site-binding probes, respectively. The structure of the 40mer dsDNA labeled with HEX is shown in Figure 16A. Compound Cy7MX was used in assay validation studies.

A)C)D) Cy7MX Cy7MX X=UX=U or T YY=A,=A, C, G, or T B)

40mer SSB Activity (%) SSB Activity (%) 16mer

X:Y U:A U:A U:A U:A U:A U:A T:A U:T U:T U:G U:G U:C U:C UDG -+- -+++++++++ APE - -+-+++++++++ Cy7MX ---+-+-+-+-+-

Figure 16. (A) 5’-HEX labeled dsDNA 40mer used for fluorescence-based cutting assay; (B) SSB assay visualized by denaturing gel electrophoresis to separate intact 40mer from 16mer. HEX labeled DNA is shown in green and Cy7MX is in red. (C) Analysis of the gel electrophoresis controls with SSB activity % = (Fl 16mer)/(Fl 16mer + Fl 40 mer) x 100. (D) Analysis of gel electrophoresis with Cy7MX or vehicle control.

As the U:A base pair is most likely to occur when dUTP is incorporated in place of dTTP,77 we used this as our primary substrate. Omission of UDG and/or APE resulted in no detectable SSB activity, as expected. The presence of both enzymes in the absence

Page | 75

of Cy7MX afforded nearly quantitative SSB activity, regardless of the base pairing the

AP site. Substitution of the U:A base pair for the T:A base pair prevented SSB activity, as expected, which confirmed the specificity of UDG for uracil in these conditions.

Adding Cy7MX to UDG treated U-DNA prior to treatment with APE blocked nearly all of the enzyme’s SSB activity, again regardless of the base pair opposite the AP site.

Colocalization of the HEX fluorescent image with the Cy7MX image revealed that the probe bound only to the 40mer strand of DNA containing an AP site. Nonspecific binding of the probe to DNA was not observed within the detection limits of the instrument. These results are shown in Figure 16 B.

One advantage to fluorescence imaging is that it is quantitative provided the image contains no saturated pixels. Imaging settings were kept consistent and no saturation was observed in this or subsequent experiments. Therefore, the SSB activity was quantified based on the intensity of the 16mer versus the sum of the total fluorescence. Quantification of the results in Figure 16 B confirm the visual observations of minimal SSB activity in the absence of uracil, APE, or UDG and nearly quantitative

SSB activity in the U:A system treated with APE and UDG (Figure 16 C). The base pair opposite the AP site had no significant effect on enzyme activity and Cy7MX blocked nearly 90% of the SSB activity (Figure 16 D). This finding support the hypothesis and validity of the assay to screen AP site binding probes.

4.2 Evaluation of potential UDG inhibition

The enzymes in this assay could be potential confounding variables as neither the enzymes nor the probes are removed during the reaction. Inhibition of UDG activity by the probe, either by direct enzyme inhibition or blocking enzyme access to the uracil

Page | 76

substrate, was examined first. Previous studies had shown that a 10 minute UDG incubation time was sufficient to quantitatively convert uracil bases to AP sites. DNA was treated with UDG for 10 minutes before Cy7MX was added to the reaction mixture

[UDG (10 min) + Cy7MX in Figure 17]. These results were compared to Cy7MX added just before UDG (UDG + Cy7MX in Figure 17). The trends of the two lines were consistent within experimental error indicating that Cy7MX had little inhibitory effect on the activity of UDG. The variation in the early time points, namely at two and ten minutes probe incubation time, may be due to experimental error arising from manual manipulation; however, there could also be some inhibition that is detachable on the minute time scale but that does not contribute substantially over an hour-long period.

The 60 minute probe incubation time was used for all subsequent experiments, except where otherwise noted, as a compromise of expediency and variability.

Cy7MX Cy7MX SSB Activity (%)

Figure 17. Evaluation of Cy7MX inhibition of UDG under assay conditions. Time course represents incubation time of U:A DNA before APE addition with UDG alone (blue circles), 10 minutes UDG pretreatment then 7 (red triangles), and UDG and 7 without pretreatment (green diamonds). Little difference is observed between the concurrent and tandem additions of Cy7MX and UDG, indicating the probe does not inhibit the enzyme under these conditions.

Page | 77

4.3 Evaluation of potential APE inhibition

The potential inhibition of APE by Cy7MX was examined. APE and Cy7MX are expected to compete for the AP site substrate. This competition appears as a reduction in

SSB activity relative to a control. To distinguish the competition from possible inhibition, a time course of APE and Cy7MX incubated together before addition of AP

DNA was measured. APE was combined with either Cy7MX or a vehicle control and incubated at 37 °C for 10, 30, 60, 90, 120, and 180 minutes. These solutions were added to U:A DNA that had been pretreated for 1 h with UDG. The DNA was treated with the

APE mixture for 1 hour before analysis. If Cy7MX were not inhibiting APE, the SSB activity would be constant over time, albeit with less SSB activity than the control due to competition. Conversely, a change in SSB activity as a function of APE and Cy7MX incubation time would indicate some degree of inhibition.

As shown in Figure 18, the vehicle control showed that APE did not lose activity in the experimental conditions over time. Cy7MX did show some ability to inhibit APE over time with a decrease in SSB activity apparent at 90 minutes. The APE incubation time used in the SSB activity assay was 60 minutes. Therefore, although Cy7MX does inhibit APE over time, it does not do so on the time scale of the experiment. The APE inhibition is not expected to have a negative effect on the assay.

Page | 78

Cy7MX SSB Activity (%) Cy7MX

A second method was also used to detect APE inhibition. Tetrahydrofuran (THF) is an analog to the AP site that APE recognizes as a substrate but the AP site probes will not bind to THF. A dsDNA oligomer with a THF:A base pair instead of a U:A base pair was used to evaluate if Cy7MX could inhibit APE directly without a background competition. A dose-response of Cy7MX with a constant [APE] was observed for 1 h.

The data indicate that Cy7MX has some inhibitory activity on the ability of APE to excise THF beginning at ~500 pmol. Except where otherwise noted, experiments were conducted using 1000 pmol of probe, and this corresponds to a 25% reduction in SSB activity for this system (Figure 19). The experiment suggests that 500 pmol would have been a better dose than 1000 nmol.

Page | 79

Figure 19. APE shows inhibition with increasing dose of Cy7MX when incubated together in the presence of THF:A DNA for 60 min. APE inhibition is detectable at ca. 500 pmol Cy7MX.

However, one should note that the THF:A substrate behaved differently than the

U:A substrate. In the absence of Cy7MX, the 90% SSB activity in the THF substrate is ca. 5% lower than what was observed in the AP site model (see Figure 16, page 75;

Figure 17, page 77; and Figure 18, page 79). In addition, for the THF substrate with

1000 pmol of Cy7MX, a 65% SSB activity is observed (a 25% reduction from the baseline, Figure 19), whereas under the same conditions for the dU:A substrate, there was only a 10% reduction from baseline (see Figure 18, page 79). The competition of

Cy7MX with APE would be expected to make the U:A SSB activity lower than the

THF:A under similar conditions. These differences suggest that APE has different affinities for the two substrates and the extent of APE inhibition by Cy7MX may be contingent on the affinity.

Page | 80

4.4 Compound screening by dose-response

The SSB assay was proposed as a quick, facile tool to screen AP site binding probes. Therefore, the assay was performed on novel AP site binding probes (Scheme 4, page 46) and the commercially available AP site binding probe, F422 (Scheme 10, page

51). ACMX (ACMX) was not studied due to poor aqueous solubility. As shown in

Figure 20, the coumarin-based probes, MCMX and NpCMX, have similar dose-response profiles and are superior to the other probes assayed in dose-response and overall efficacy. The cyanine-based dyes, Cy5MX and Cy7MX, behaved similarly to each other and had acceptable dose-response and overall efficacy. The dansyl-based compound,

DansylMX, and F422 showed unacceptable dose-response with an overall reduction in

SSB activity of only ca. 60% and 90%, respectively. The reduced response of F422 and

DansylMX may be due in part to the compounds being fairly electron rich near the aminooxy binding moiety. This could lead to a Coulombic repulsion with the DNA phosphate backbone.

Page | 81

SSB Activity (%)

Figure 20. SSB activity assay is used to compare the dose-response of AP site binding probes in 5 pmol of AP DNA. Mean and standard deviation of triplicate samples are shown for each concentration.

4.5 Comparison to ARP and MX

The SSB activity assay was used to compare new AP site binding probes to the previously developed ARP107 and MX.110-111 Using the same conditions as the control experiments (1 nmol probe, 5 pmol DNA, 1X APE), Cy7MX time course (before APE addition) showed remarkably superior AP site binding than ARP and MX. The decrease in SSB activity caused by ARP and MX was only slightly better than the vehicle control

(Figure 21A). To verify that this result is not an artifact of the assay, reaction conditions were modified to tease out ARP and MX time courses of binding. To this end, the molar equivalents of ARP and MX were increased 100-fold relative to Cy7MX. As the APE is expected to compete with the probes for the AP site, its concentration was decreased by a factor of ten as well. Using these modified conditions, MX showed a pronounced improvement in the time course and had an overall lower SSB activity than Cy7MX.

ARP also showed a steady decrease in SSB activity over time, but never reached the

Page | 82

levels of Cy7MX or MX (Figure 21B). From these data, one may conclude that the initial results were not an artifact of the assay and that Cy7MX shows superior AP site binding ability. This conclusion is further supported in a subsequent experiment described in §5.4 (page 102). AB

Cy7MX SSB Activity (%)

Figure 21. Comparison of the SSB activity of U:A DNA (5 pmol) treated with UDG and Cy7MX, ARP, MX, or vehicle control as a function of incubation time prior to APE addition. (A) ARP (1 nmol), MX (1 nmol), and Cy7MX (1 nmol) with UDG (5 units) and APE (10 units); (B) ARP (200 nmol), MX (200 nmol), and Cy7MX (2 nmol) with UDG (5 units) and APE (1 unit). Bar represent the average of three samples and error bars are the standard deviation.

This study does observe the reaction over time. However, one must be careful in drawing too specific a determination of the reaction kinetics as there may be some degree of APE inhibition by Cy7MX. A better measure of the reaction kinetics can be performed in genomic DNA, which was done for Cy7MX in §5.3 (page 101).

4.6 Identification of a transient band

Under the conditions of the control experiments, only two major bands, corresponding to the expected 40- and 16-mer ssDNA, were observed in the SSB activity assay (Figure 16 B, page 75). However, during dose-response experiments an unknown

Page | 83

band of intermediate molecular weight was observed containing the AP site binding probe Cy5MX but lacking the 3’ HEX label. More surprisingly, this band was present at low concentrations of Cy5MX but disappeared at high concentrations. Control experiments eliminated an enzyme-Cy5MX complex, a nonspecific DNA-Cy5MX interaction, and fluorescence channel bleeding as possible species (Figure 22).

Figure 22. Control experiments to identify unknown band in SSB activity assay.

The apparent molecular weight of the unknown band was consistent with the truncated 23mer that remains after the SSB. This band does not contain a HEX label for fluorescence detection. Examination of the BER pathway reveals that the AP site persists after the action of APE (Figure 14, page 45, complex II). Therefore, the band likely results from binding of the probe to the product following the action of APE. The proposed hypothesis for the concentration-dependent appearance and disappearance of this band is based on competition between the probe and APE for the AP site. While

APE can act only on the unaltered AP site substrate (Figure 14, complex I, page 45) and not the probe-AP site lesion (complex III), the probe can bind either complex I or II.

Therefore, it is predicted that at low probe concentrations, the competition favors APE

SSB and the probe binds predominately to the truncated 23mer. At high probe

Page | 84

concentrations, the competition favors the probe and APE cannot cause a SSB (Figure

23). Initial control experiments serendipitously used a probe concentration that exceeded the critical competition limit.

100% 100%

% APE % Cy5MX SSB 40mer activity binding

[Cy5MX]

Figure 23. Illustration of the Cy5MX concentration dependent competition with APE. At low Cy5MX concentrations, APE SSB activity dominates. At high Cy5MX concentrations, APE SSB activity is inhibited by Cy5MX binding to the 40mer before the enzyme reacts.

To show that the molecular weight of the unknown band was consistent with the

23mer fragment, DNA was employed that contained the fluorescent tag on the 3’ end of the uracil containing sense strand. As shown in Figure 24, the proposed 23mer-probe conjugate (red, columns 2-4 and 9) corresponds to the 3’ labeled 23mer (green, columns

5-8, 10, and 12). Note that this band has a molecular weight equal to the sum of the

23mer and a fluorophore (the probe and 3’ label have similar molecular weights). The resolving power of the gel was enough to separate the singly labeled 23mer (probe or 3’ label) from the doubly labeled 23mer (probe and 3’ label), which appears as the lower of the two yellow bands in Figure 24 columns 6-8 and 10.

Page | 85

Figure 24. The unknown band is identified as a 23-mer containing an AP site-bound Cy5MX after APE SSB activity. HEX and Alexa 532 fluorophores are shown in green, Cy5MX is shown in red.

Support for the competition hypothesis was gathered by controlling the order of addition of APE and probe to the DNA reaction mixture. DNA allowed to react completely with APE before addition of Cy5MX showed only 16mer and 23mer strrands

(Figure 24, columns 9-10). When APE was added after Cy5MX, they competed for the

AP site substrate in a dose-dependent manner with APE SSB activity decreasing with increasing probe (Figure 24 columns 1-4 and 5-8). These data support the competition hypothesis giving rise to a band that is present on the gel at intermediate probe concentrations. Further, these observations support the specific binding mechanism of the probes.

4.7 Evaluation of probe-AP site lesion repair: endonuclease survey

The SSB activity assay was developed using APE, the predominate endonuclease in the BER pathway. The assay has shown that the probe blocks the activity of APE by masking its native substrate, as APE does not recognize the AP site-probe lesion. By consequence, the probe also blocks further BER. The persistence of the probbe is important when considering a dynamic, living system, such as in cells or mouse models.

Page | 86

Dogliotti and coworkers have demonstrated a precedence in MX-AP site lesion repair by endonucleases III and IV.149 Therefore, the ability of various endonucleases to recognize and repair the AP site-probe lesion was examined.

A small survey of endonucleases was used in this experiment. Endonuclease III and VIII were chosen for their similarity to mammalian glycosylases NTH1 and NEIL1

(Table 1, page 14). Endonuclease V has been observed by others to cleave the 5’ fluorescent label from an oligonucleotide.* Thus, both 3’ and 5’-labeled DNA was used in this experiment. For this study, DNA was treated with UDG then Cy5MX was allowed to react completely before addition of the glycosylase. This ensured that the bands observed were products of AP site-probe lesion repair and not endonuclease-probe competition as discussed in §4.6 (page 83).

As expected, a vehicle control showed no repair of the AP site-Cy5MX lesion.

Consistent with previous SSB activity assay results, APE did not show any repair activity either. Endonuclease III showed some SSB activity, but did not cleave the probe from the DNA. Endonuclease IV behaved similarly to endonuclease III and showed complete

SSB activity but no cleavage of the probe from the DNA. Removal of the probe from the

DNA is presumably necessary for complete BER. In addition, endonuclease IV removed the 3’ fluorophore but not the 5’ fluorophore. Endonuclease V had no SSB activity, but was observed to cleave the 5’ fluorophore while leaving the 3’ label intact. Only endonuclease VIII was observed to remove Cy5MX from the DNA backbone, create a

SSB, and provide a substrate for further BER (Figure 25).

* Marks, K. and Landry D., New England Biolabs, Inc. unpublished observation, from https://www.neb.com/products/m0305-endonuclease-v [accessed January 27, 2015]. Page | 87

Figure 25. Evaluation of potential AP site-probe repair by several endonucleases after 1 h incubation. APE, endonuclease V, and vehicle control showed no repair activity. Endonucleases III and IV showed SSB activity while retaining the probe on the DNA. Endonuclease VIII removed the probe from the AP site and caused a SSB. Fluorescent tags on DNA are shown in green, Cy5MX is shown in red.

This experiment demonstrates that the AP site-probe lesion may be susceptible to repair in cancer cells. Cancer cells have different protein expression levels and thus may respond differently to probes in situ. A Western blot analysis of endonuclease VIII, for example, may give a researcher insight into how effectively these probes would work in his or her system of interest. This test examined the repair of the AP site-Cy5MX lesion as a proof-of-principle experiment. Other AP site probes may have different susceptibilities and should be examined separately.

4.8 Discussion and conclusions

The SSB activity assay was developed to screen potential AP site binding probes based on the hypothesis that an AP site-probe lesion is not repaired by APE. Two compounds, DansylMX and F422, were shown to be less effective than other probes

Page | 88

based on this assay. One reason for their failure may be that APE does in fact recognize the specific lesions they form with AP sites. These probes may also be inhibiting the activity of either UDG or APE in the assay reaction. Potential inhibition was not examined using these compounds, but should be done before eliminating them from the

AP site probe pool. Despite the limitation of finding false negatives (and neither of these probes have been confirmed as true or false negatives), the SSB activity assay represents a useful tool for compound screening.

In addition to compound screening, the assay can be modified to illuminate other chemical processes. For example, the use of the assay to screen AP site-probe repair (in

§4.7, page 86) could save an experimentalist considerable time by allowing analysis of cellular phenomena in vitro. While beyond the scope of this project, the assay could be easily modified to use a probe to evaluate inhibitors targeted to UDG, APE, or other glycosylases.

This assay used a uracil lesion and UDG as the substrate-enzyme combination.

The assay can be modified to examine other common lesions as well. 8-OxoG base substitutions and hOGG1 are both commercially available and of particular interest due to the high toxicity of 8-oxoG lesions. The toxicity of 8-oxoG has been implicated in diseases beyond cancer including Alzheimer’s and Parkinson’s diseases.137-138

4.9 Materials and methods

4.9.1 General methods

SSB activity assay procedures. SSB activity assays were performed on a 40-mer duplex

DNA synthesized by Integrated DNA Technologies with the sequences:

Page | 89

5’-[HEX] TCCTGGGTGACAAAGCXAAACACTGTCTCCAAAAAAAATT

3’-AGGACCCACTGTTTCGYTTTGTGACAGAGGTTTTTTTTAA

5’-TCCTGGGTGACAAAGCUAAACACTGTCTCCAAAAAAAATT-[Alexa532]

3’-AGGACCCACTGTTTCGATTTGTGACAGAGGTTTTTTTTAA

where X = uracil, thymine, or THF and Y = adenine, cytosine, guanine, or thymine.

DNA was diluted in dH2O to 500 nM and 5 pmol (10 ȝL) aliquots were used in each sample. APE (10,000 Units/mL) and UDG (5,000 Units/mL) enzymes and corresponding buffers were purchased from New England BioLabs. UDG storage buffer (10 mM Tris-

HCl, 50 mM KCl, 1 mM DTT, 0.1 mM EDTA, 0.1 mg/ml BSA, 50% Glycerol, pH 7.4) and APE storage buffer (10 mM Tris-HCl, 50 mM NaCl, 1 mM DTT, 0.05 mM EDTA,

200 ȝg/ml BSA, 50% glycerol, pH 8.0) were prepared according to formulation provided by New England BioLabs to use as blanks, where necessary. The enzymes were not heat inactivated as this was observed to give rise to artifacts. Reaction products were resolved on denaturing 20% polyacrylamide gels (5.3 g urea, 5.0 mL 40% acrylamide, 2.3 mL 5X

TBE buffer, 200 ȝL 10% APS, and 20 ȝL TEMED) at 300 V for 30-45 min in the dark, as observed by the progression of the loading dye. 5X TBE buffer was prepared with tris base (54 g), boric acid (27.5 g), and EDTA (4.65 g) diluted to 1 L in water. Loading dye

(300 ȝL 10M NaOH, 20 mg bromophenol blue, 9.7 mL formamide) was added to samples to aid loading and visualization of gel progression.

Gels were imaged based on the DNA tag (HEX or Alexa 532) on Typhoon Trio +

Variable Mode Imager (Amersham Biosciences) in fluorescent mode with 532 nm

Page | 90

excitation and 555 nm emission with a 20 nm band pass, PMT set to 400 V, and pixel size resolution of 100 ȝm. Fluorescence of the AP site probe Cy5MX was imaged with

633 nm excitation and 670 nm emission with a 30 nm band pass, PMT set to 500 V, and pixel size resolution of 100 ȝm. Gel data were analyzed using ImageQuant software

(Amersham Biosciences). DNA cutting was defined as the DNA label fluorescence intensity of the 16-mer strand divided by the sum of the fluorescence intensities of the

16- and 40-mer strands. The fluorescence of Cy7MX on the gels was imaged on a

Syngene G:Box Chemi XT4 scanner.

The image in Figure 16 (page 75) was modified in Adobe Photoshop from the original in the following ways: 1) green and red photo filters were applied to corresponding monochrome images for clarity; 2) the HEX image from the Typhoon scanner (green) was scaled to same size as the Cy7MX image from the Syngene scanner

(red) to account for different image resolutions and to facilitate coregistration; 3) a levels adjustment filter was applied uniformly to the red image to increase signal-to-noise and improve contrast; 4) the images were cropped to the region of interest; and 5) a "screen" blending mode was applied to the green HEX layer to allow the red layer to be observed beneath it without changing opacity settings. No quantitative measurements were taken from the modified images.

4.9.2 Reactions to establish assay validity

SSB activity assay control reactions. Samples were prepared in triplicate. To a 0.6 mL

Eppendorf tube were added HEX-labeled dsDNA (10 ȝL, 5 pmol), 10X UDG reaction buffer (1 ȝL), 10X APE reaction buffer (1 ȝL), H2O (5 ȝL), Cy7MX or vehicle control (2

ȝL, 1 nmol; vehicle = 1% DMSO in H2O), and UDG or UDG storge buffer (1 ȝL, 5

Page | 91

Units). Samples were incubated at 37 °C for 1 h in the dark. Then APE (1 ȝL, 10 Units) or APE storage buffer (1 ȝL) was added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (7.5 ȝL) was added to each sample then 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

SSB activity assay with base pairs. Reaction samples were prepared in triplicate; control samples were prepared singly. To a 0.6 mL Eppendorf tube were added HEX-labeled dsDNA (10 ȝL, 5 pmol), 10X UDG reaction buffer (1 ȝL), 10X APE reaction buffer (1

ȝL), H2O (5 ȝL), Cy7MX or vehicle control (2 ȝL, 1 nmol; vehicle = 1% DMSO in

H2O), and UDG (1 ȝL, 5 Units). Samples were incubated at 37 °C for 10 min, 30 min, 1 h, 2 h, or 3 h in the dark. Then APE (1 ȝL, 10 Units) was added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (5 ȝL) was added to each sample then 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

SSB activity assay UDG inhibition.

Method UDG (10 min) + Cy7MX: All samples were prepared in triplicate. To a

0.6 mL Eppendorf tube were added HEX-labeled dsDNA (U:A, 10 ȝL, 5 pmol), 10X

UDG reaction buffer (1 ȝL), 10X APE reaction buffer (1 ȝL), H2O (5 ȝL), and UDG (1

ȝL, 5 Units). Samples were incubated at 37 °C for 10 min. Compound Cy7MX (2 ȝL, 1 nmol) was added and the samples were incubated at 37 °C in the dark for 2 min, 10 min,

30 min, 60 min, or 180 min. Then APE (1 ȝL, 10 Units) was added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (5 ȝL) was added to each sample then 5 ȝL of each sample was loaded onto a 1.0 mm thick, 15-well gel.

Page | 92

Method UDG + Cy7MX: All samples were prepared in triplicate. To a 0.6 mL

Eppendorf tube were added HEX-labeled dsDNA (U:A, 10 ȝL, 5 pmol), 10X UDG reaction buffer (1 ȝL), 10X APE reaction buffer (1 ȝL), H2O (5 ȝL), Cy7MX (2 ȝL, 1 nmol), and UDG (1 ȝL, 5 Units). Samples were incubated at 37 °C in the dark for 2 min,

10 min, 30 min, 60 min, or 180 min. Then APE (1 ȝL, 10 Units) was added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (5 ȝL) was added to each sample then 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

Method UDG: As in method UDG+Cy7MX except substitute DMSO (2 ȝL) for

Cy7MX.

SSB activity assay APE inhibition – with competition. APE (2 ȝL, 20 Units) and Cy7MX or vehicle (4 ȝL, 2 nmol; vehicle = 1% DMSO in H2O) were mixed and incubated at 37

°C for 0.5 h, 1 h, 1.5 h, 2 h, or 3 h in the dark. Meanwhile, five samples of each DNA reaction were prepared. To a 0.6 mL Eppendorf tube were added HEX-labeled dsDNA

(U:A, 10 ȝL, 5 pmol), 10X UDG reaction buffer (1 ȝL), 10X APE reaction buffer (1 ȝL),

H2O (5 ȝL), and UDG (1 ȝL, 5 Units). Samples were incubated at 37 °C for 1 h in the dark. Then APE/Cy7MX mixtures (3 ȝL) were added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (5 ȝL) was added to each sample then 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

SSB activity assay APE inhibition – without competition. Three samples of each DNA reaction were prepared. To a 0.6 mL Eppendorf tube were added HEX-labeled dsDNA

(THF:A, 10 ȝL, 5 pmol), 10X APE reaction buffer (2 ȝL), H2O (5 ȝL), and Cy7MX (2

Page | 93

ȝL, 10, 50, 200, 500, 1000, 2000, or 5000 pmol) or a vehicle control. APE (1 ȝL, 10

Units) or APE storage buffer (1 ȝL) was added and aamples were incubated at 37 °C for

1 h in the dark. Loading dye (5 ȝL) was added to each sample then 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

4.9.3 Evaluation of AP site binding probes

Compound screening. Stock solutions (Table 5) were diluted in water to the following quantity of probe/sample (nmol): 10, 5, 2, 1, 0.5, 0.2, 0.05, 0.01, and 0, unless where otherwise noted. Samples were prepared in triplicate. To a 0.6 mL Eppendorf tube were added HEX-labeled dsDNA (U:A, 10 ȝL, 5 pmol), 10X UDG reaction buffer (1 ȝL), 10X

APE reaction buffer (1 ȝL), H2O (5 ȝL), probe or vehicle control (2 ȝL), and UDG (1 ȝL,

5 Units). Samples were incubated at 37 °C for 1 h in the dark. Then APE (1 ȝL, 10

Units) was added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (5 ȝL) was added to each sample then 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

Table 5. Stock concentration and initial % DMSO of probes evaluated by SSB activity assay.

Probe Stock [nmol] % DMSO

F422 10 10

Cy7MX 10 0.5

ACMX 5 1

NpCMX 10 10

MCMX 10 1

Cy5MX 10 1

DansylMX 5 1

Page | 94

SSB activity assay ARP and MX comparison.

Method 1 (see Figure 21 A, page 83): Stock solutions of probes were prepared at

50 mM. Compound Cy7MX was dissolved in DMSO, MX was dissolved in H2O (pH 7), and ARP was dissolved in H2O. Stock solutions were further diluted with H2O to 500

ȝM. Water was used as a blank. All samples were prepared in triplicate. To a 0.6 mL

Eppendorf tube were added HEX-labeled dsDNA (U:A, 10 ȝL, 5 pmol), 10X UDG reaction buffer (1 ȝL), 10X APE reaction buffer (1 ȝL), H2O (5 ȝL), probe or blank (2

ȝL, 1 nmol), and UDG (1 ȝL, 5 Units). Samples were incubated at 37 °C for 10 min, 30 min, 1 h, 2 h, or 3 h in the dark. Then APE (1 ȝL, 10 Units) was added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (5 ȝL) was added to each sample then 5 ȝL of each sample was loaded onto a 1.0 mm thick, 15-well gel.

Method 2 (see Figure 21 B, page 83): Stock solutions of 50 mM ARP and MX were used without further dilution. Compound Cy7MX was diluted to 500 ȝM. APE was diluted 10-fold in APE storage buffer. Water was used as a blank. All samples were prepared in triplicate. To a 0.6 mL Eppendorf tube were added HEX-labeled dsDNA

(U:A, 10 ȝL, 5 pmol), 10X UDG reaction buffer (1 ȝL), 10X APE reaction buffer (1 ȝL),

H2O (5 ȝL), probe or blank (4 ȝL; 2 nmol 7, 200 nmol ARP and MX), and UDG (1 ȝL, 5

Units). Samples were incubated at 37 °C for 10 min, 30 min, 1 h, 2 h, or 3 h in the dark.

Then APE (1 ȝL, 1 Unit) was added and samples were again incubated at 37 °C for 1 h in the dark. Loading dye (5 ȝL) was added to each sample then 5 ȝL of each sample was loaded onto a 1.0 mm thick, 15-well gel.

4.9.4 Evaluation of BER

Identification of transient band.

Page | 95

Controls. A stock solution of compound Cy5MX (500 μM) was freshly prepared

in 1% aqueous DMSO and stored in the dark. Samples were prepared in triplicate. To

each sample was added HEX-labeled dsDNA (U:A, 10 μL, 5 pmol) or a vehicle blank,

water (5 μL), 10X UDG reaction buffer (1 μL), 10X APE reaction buffer (1 μL), Cy5MX

(2 μL, 1 nmol) or a vehicle control, and UDG (1 μL, 5 Units) or UDG storage buffer (1

μL). Samples were incubated 1 h at 37 °C in the dark. APE (1 μL, 10 Units) or APE storage buffer (1 μL) were added to each sample. Samples were incubated at 37 °C for 1 h. Loading dye (5 μL) was added and 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

Dose response. Stock solutions of compound Cy5MX (25, 100, 250, and 500

μM) were prepared in 1% aqueous DMSO and stored in the dark. Samples were prepared in triplicate.

For addition of APE after the probe samples, to each was added HEX- or Alexa- labeled dsDNA (U:A, 10 μL, 5 pmol), water (5 μL), 10X UDG reaction buffer (1 μL),

10X APE reaction buffer (1 μL), Cy5MX (2 μL; 0.05, 0.2, 0.5, or 1 nmol) or a vehicle

control, and UDG (1 μL, 5 Units). Samples were incubated 1 h at 37 °C in the dark.

APE (1 μL, 10 Units) was added to each sample. Samples were incubated at 37 °C for 1

For addition of APE before the probe samples, to each was added HEX- or Alexa-

labeled dsDNA (U:A, 10 μL, 5 pmol), water (5 μL), 10X UDG reaction buffer (1 μL),

10X APE reaction buffer (1 μL), APE (1 ȝL, 10 Units), and UDG (1 μL, 5 Units).

Samples were incubated 1 h at 37 °C in the dark. Cy5MX (2 μL, 1 nmol) was added to

each sample. Samples were incubated at 37 °C for 1 h.

Page | 96

To all samples, loading dye (5 μL) was added and 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

Endonuclease survey. Endonuclease III (#M0268S), endonuclease IV (#M0304S), endonuclease V (#M0305S), endonuclease VIII (#M0299S), APE I (#M0282L), and their corresponding 10X reaction buffers were purchased from New England Biolabs. A stock solution of compound Cy5MX (500 μM) was freshly prepared in 1% aqueous DMSO and stored in the dark. Samples were prepared in triplicate. To each sample was added HEX- or Alexa-labeled dsDNA (U:A, 10 μL, 5 pmol), water (5 μL), 10X UDG reaction buffer

(1 μL), and UDG (1 μL, 5 Units). Samples were incubated 30 minutes at 37 °C.

Compound Cy5MX (2 μL, 1 nmol) was added to all samples before incubating 1 h at 37

°C in the dark. Corresponding 10X reaction buffers (1 μL) were added using the APE reaction buffer in the vehicle control. Endonuclease (1 μL, 10 Units) or vehicle (1 μL,

APE storage buffer) were added to each sample. Samples were incubated at 37 °C for 1 h. Loading dye (2.5 μL) was added to 10 μL aliquots of sample before 10 ȝL of each sample was loaded onto a 1.0 mm thick, 10-well gel.

CHAPTER 5. GENOMIC DNA BINDING ASSAY FOR PROBE DEVELOPMENT

5.1 Hypotheses and methodology

The SSB activity assay relies on an AP site model that contains a defined number of AP sites per total number of bases. This is a useful model system for comparing multiple probes in a reliably quantitative environment. The shortcoming of the SSB

Page | 97

assay is that 1 AP site/80 bases is such a large concentration of AP sites that it would not be expected in nature. Also, this assay has not been used to establish if the probes can report on quantities of AP sites. To model physiologically relevant sizes of DNA and quantities of AP sites, the probes were examined in genomic DNA. Two methods were used: in vitro generation of AP sites in calf thymus DNA and in chemotherapy treated cancer cell lines.

AP sites were generated in calf thymus DNA by heating at 70 °C in citrate buffer

(pH 5.0) following a procedure described by Lindahl and Nyberg.150 This treatment generates a number of AP sites that increases linearly with time. I hypothesized that if the probes can report on AP site quantity, then a linear fluorescence response is predicted provided the concentration of the probe is within the dynamic range of fluorescence emission. Further, this fluorescence response could be used to establish specific binding to AP sites in genomic DNA by MX blocking; the DNA binding time course; and ARP and MX competition with probes for AP sites (Figure 26A).

Page | 98

A Cold Cold Heat EtOH Conditions EtOH pH 5 -Cy7MX i. 1. ± MX 2. Cy7MX free -MXfree 45 min or ii. Cy7MX, 30 s - 60 min -ARP 15min free iii. Cy7MX + (ARP or MX) B intervals Cold FUDR Proteinase K PhOH 1. ± UDG EtOH

RNase A CHCl3 2. Cy7MX -Cy7MXfree WT or KD 24, 48 DLD1 or 72 h

C Cold MMS Proteinase K PhOH Cy7MX EtOH

RNase A CHCl3 -Cy7MXfree WT DLD1 3 h

Figure 26. Schematic of methodologies used in genomic DNA assays. (A) Heat and acid treatment of calf thymus DNA for a time period followed by purification by ethanol precipitation are used to (i) report on quantity of AP sites and MX blocking, (ii) establish a DNA binding time course, and (iii) measure ARP or MX competition. (B) Chemotherapy treatment of cancer cells followed by DNA isolation and exogenous enzyme treatment. (C) Chemotherapy treatment of cancer cells followed by DNA isolation without exogenous enzyme treatment.

5-Fluoro-2ƍ-deoxyuridine (FUDR or floxuridine) is a cancer chemotherapeutic that induces DNA damage in two ways: first, by inhibiting thymidylate synthase and decreasing the thymine pool leading to uracil misincorporation;151-152 and second, by cellular activation then direct incorporation of 5-fluorodeoxyuracil (FUra) into the DNA chain.153-154 Both uracil and FUra lesions are repaired by UDG (see Table 1, page 14).89-

91, 153 I hypothesized that following FUDR treatment a UDG deficient cell line would accumulate lesions while a UDG normal cell line would repair them effectively, and that the AP site binding probe Cy7MX could report on the quantity of AP sites in DNA isolated from FUDR treated colon cancer cells following exogenous UDG treatment. AP

Page | 99

sites would correspond to the quantity of uracil and FUra in the cell at a moment in time

(Figure 26 B). This experiment would demonstrate that the sensitivity of Cy7MX is sufficient to visualize physiologically generated quantities of lesions over time.

As opposed to the antimetabolite, FUDR, methyl methanesulfonate (MMS) directly damages DNA by methylating bases. These methylated bases are excised by

BER to form an AP site. I hypothesized that Cy7MX could report on AP sites in extracted cellular DNA following MMS treatment of a normal cell line. The quantity of

AP sites would correspond to the quantity of converted methylated bases (Figure 26 C).

This experiment would demonstrate that the sensitivity of Cy7MX is sufficient to visualize physiologically generated AP sites directly at a single point in time, as opposed to the lesions accumulated over time observed in the FUDR experiment.

5.2 Blocking with MX

Calf thymus DNA was treated for 15 minute increments at pH 5.0 and 70 °C over

90 minutes. After purification by ethanol precipitation, MX or an equimolar NaCl vehicle control was added to AP DNA solutions. After a 30 minute incubation, Cy7MX was added to each sample without removing MX. This ensured that MX was allowed to completely bind the AP sites without competing with Cy7MX. Ethanol precipitation was used to remove free Cy7MX and MX before analysis (Figure 26 Ai).

In the absence of MX, a linear fluorescence response of Cy7MX was observed with increasing numbers of AP sites (Figure 27). This supports the hypothesis that

Cy7MX can report on AP site quantity. Also, the results show that the fluorescence response is within the dynamic range of the probe for the quantities of AP sites examined.

Page | 100

Figure 27. Calf thymus AP DNA was pretreated with MX (50 mM, +MX) or vehicle (50 mM, - MX) followed by incubation with Cy7MX (0.05 mM). Fluorescence levels show a linear response to Cy7MX in the absence of MX. No fluorescence response was observed with MX pretreatment, indicating MX and Cy7MX share a common binding site. Data points are the average of three samples and error bars represent the standard deviation.

Pretreatment with MX followed by Cy7MX produced no florescence response relative to the number of AP sites (Figure 27). This confirms that Cy7MX and MX share a binding site. A low level of fluorescence was observed above the background, which indicated that Cy7MX may have some nonspecific binding to DNA. However, this level was low and not observed with other detection methods (see Figure 16, page 75).

5.3 Binding time course evaluation

The sluggish enzymatic activity of APE and viscosity of solutions were problematic for accurate time course measurement in the SSB assay. AP sites produced by heat and acid treated DNA proved a better system than the oligomer assay to measure

DNA binding kinetics because of the ability to stop the reaction by precipitating the DNA with ethanol. Using a single, 45 minutes heat and acid treatment time, the time course of

Cy7MX binding was measured between 0.5 and 60 minutes (schematic shown in Figure

Page | 101

26 Aii, page 99, data shown in Figure 28). The reaction half-life (t1/2 = 2.6 minutes) was calculated. The reaction was complete within ten minutes. This binding time course is acceptable based on the time scale of cellular BER (see §5.5.1, page 104 for example).

5.4 Competition with MX and ARP

Probe Cy7MX showed superior inhibition of SSB activity in the oligomer assay compared to ARP and MX (see Figure 21, page 83). To independently verify these results, a competition was conducted using a single, 45 minute heat and acid treatment time point of calf thymus DNA. Solutions of Cy7MX and either APR or MX were prepared and rigorously kept in the dark. This was necessary as a solution of Cy7MX and MX was observed to decompose rapidly when exposed to light. The concentration of

Cy7MX was maintained at 25 ȝM and 1 ȝM while varying the molar concentrations of

MX and ARP, respectively. ARP has lower solubility than MX and this limited the

Page | 102

maximum concentration of Cy7MX in the assay. The probe solutions were added to AP

DNA and free probes were removed after one hour by ethanol precipitation of the DNA

(Figure 26 Aiii, page 99).

The fluorescence competition results are shown in Figure 29 A (MX) and B

(ARP). The signal-to-noise was higher for the ARP than the MX competition because of the lower concentration of Cy7MX used, and this noise is apparent in the higher fluorescence minimum and larger error bars in the ARP study than in the MX study.

Based on the fittings, the 50% effective dose (ED50) of Cy7MX was calculated to be

3000-fold excess for MX and 2600-fold excess for ARP. These results support the superior AP site binding of Cy7MX. Presumably, the increased hydrophobicity of

Cy7MX over MX and ARP as well as a mild Coulombic interaction may favor an association of Cy7MX with DNA over bulk solution and contribute to this drastic difference.

Page | 103

A) B)

Cy7MX Cy7MX

Figure 29. Competition studies between Cy7MX and (A) MX or (B) ARP for AP sites in calf thymus DNA treated with heat and acid for 45 min are shown. The molar ratio of probe to Cy7MX was increased by increasing [ARP] or [MX] and maintaining a constant [Cy7MX]. Based on curve fittings using the programed model ExpDec2 nonlinear fit in Origin 9.1, ED50 values were calculated to be 2600-fold excess for ARP and 3000-fold excess for MX. For (A), data points are the average of three samples and error bars represent the standard deviation. For (B), data points are the average of five samples and error bars represent the standard error of the mean.

The enhanced potency of Cy7MX over ARP and MX could make the probe useful

in developing an assay kit. Based on these data, the assay would require three orders of

magnitude less moles of Cy7MX than ARP or MX, which is still substantial after

adjusting for different molecular weights. An additional application could take advantage

of the differential reactivity with DNA between Cy7MX and MX to use the latter

compound to block non-DNA aldehydes in the cell.

5.5 Measurement of cellular response to DNA damaging drugs

5.5.1 Measurement of FUDR response

The model AP site binding probe Cy7MX has been shown to bind AP sites

specifically and to report on the quantity of artificially induced AP sites. To understand

if the probe is sensitive to physiologically produced AP sites, the probe must be

Page | 104

examined in DNA taken from cells. The complexity of the cell – from membrane permeability, metabolism, and dynamics of DNA repair processes – is too large a jump from purified calf thymus DNA and introduces many new variables. Therefore, an experiment was envisioned in which cells would be treated with DNA damaging drugs and the DNA extracted for in vitro analysis.

As BER is a dynamic process, this method would be limited to the single point in time of DNA extraction and would not reflect the cumulative extent of repair over the treatment course. In addition, AP sites are weak points in DNA and are prone to shearing by physical forces. This means that standard methods of DNA extraction could introduce artifacts. Uracil and FUra lesions are more stable to isolation methods than AP sites.

When treated with FUDR, a UDG deficient cell line would accumulate uracil and FUra lesions over time while a matched UDG normal cell line undergoing BER would remove these lesions. DNA isolated from these cell lines could be treated with exogenous UDG in vitro to convert the accumulated dU/FUra lesions to AP sites (Figure 26, page 99).

This method avoids the isolation of AP DNA.

A UDG knockdown (KD) and wild type control (WT) were prepared in DLD1 colon cancer cells. The cell lines were treated with various concentrations FUDR for 24,

48, or 72 h. After treatment, the DNA was extracted from the cells and treated with UDG or a vehicle control. AP sites were detected with Cy7MX. As shown in Figure 30, both cell lines with and without UDG displayed similar fluorescence responses without FUDR treatment at all three time points. As the FUDR dose and treatment time increased, the

KD + UDG treatment displayed increasing fluorescence intensity with a nearly 9-fold increase at 72 h and 1000 nM relative to the 72 h untreated sample. The KD – UDG

Page | 105

treatment group did not show any fluorescence response to increasing FUDR concentration or treatment time, indicating that the observed signal in the KD + UDG was due to AP sites formed at dU/FUra sites following exogenous UDG treatment.

Neither WT +/– UDG showed a fluorescence response to FUDR treatment, supporting the assertion that the DLD1 WT cells are actively undergoing BER and AP sites, dU, and

FUra lesions are not present in sufficient concentrations for detection at single points in time. A) B) C) * * * * * *

[FUDR] [FUDR] [FUDR]

Figure 30. Detection of AP sites in DNA isolated from DLD1 UDG knockdown (KD) and control (WT) cells after (A) 24 h, (B) 48 h, and (C) 72 h of continuous FUDR exposure. Purified DNA was treated in vitro with UDG (+ UDG) or vehicle control (-UDG). Data bars are the average of three samples and error bars represent the standard deviation. *p < 0.001

From these results, one may conclude that the probe Cy7MX is sensitive to AP site quantity at physiologically relevant concentrations. Further, DLD1 cells use BER to respond to FUDR damage. Presumably, uracil and FUra lesions are formed in both KD and WT cell lines at similar rates. The UDG WT cell line can actively repair these lesions but BER stalls in the UDG KD cell lines. These results suggest that DLD1 cells treated with FUDR will produce AP sites in sufficient quantities for fluorescence detection.

Page | 106

5.5.2 Measurement of MMS response

FUDR and DLD1 KD cells were able to elucidate the accumulated effect of a

DNA damaging drug over time. However, UDG KD cell lines are not expected in nature.

Measurement of the accumulated effect of a drug may not be possible for all purposes.

Thus, the ability of Cy7MX to measure AP sites at a single time point was explored.

MMS is a methylating agent used to induce AP sites in cells. An MMS dose- response in DLD1 WT cells was performed for a 3 h treatment time. The WT cell line has normal BER and is expected to actively undergo repair in response to the drug.

Treatment time and dosages were taken from the literature.155-156 DNA was then extracted by phenol-chloroform. Purified DNA was incubated with Cy7MX for 1 h. Fluorescence was measured following ethanol precipitation purification.

The data indicate that Cy7MX was sensitive to the quantity of AP sites induced by mM concentrations of MMS (Figure 31). This indicates that Cy7MX can detect physiological concentrations of AP sites at a single time point. MMS methylates DNA and these lesions may be repaired by a glycosylase (see Table 1), or AP sites may form from spontaneous depurination as the positively charged base is a good leaving group.

Page | 107

5.6 Discussion and conclusions

To be useful in live cell or animal imaging, AP site binding probes must be sensitive to physiologically relevant quantities of AP sites. If the fluorescence of AP site- probe lesions is less than the detection limit of an instrument, then the probes will not be able to distinguish AP sites in DNA. From these studies, the model probe Cy7MX has been shown to be sensitive to lesions produced in DLD1 cells treated with FUDR.

Many considerations may limit the use of probes in cells or animals, including binding rate, metabolism, access to binding sites, and BER kinetics. In addition, cells do not have the same responses to the same drugs leaving the experimentalist wondering if his or her model cell line is generating AP sites. Therefore, demonstrating that a particular cell line responds to a particular drug is crucial before developing experiments

Page | 108

to characterize AP site binding. Only after a probe has been established as effective in detecting AP sites in known situations can it be employed in an unknown situation.

5.7 Materials and methods

5.7.1 General methods

DNA concentrations were measured on a Nanodrop ND-1000 spectrometer.

Fluorescence data were collected on a Tecan Infinite M200 scanner in a Corning 96 well black, clear bottom plate.

5.7.2 Heat/acid treated genomic DNA

Calf thymus DNA was purchased from Sigma and was reconstituted overnight at

4 °C in either H2O or 500 mM MX (to eliminate basal AP sites) with gentle shaking (1.5 mL water or solution per 5 mg DNA). For samples reconstituted in MX solution, an ethanol precipitation was performed twice before heat and acid treatment (vide infra) then reconstituted in water. The DNA-water solution was aliquoted to 1.5 mL Eppendorf tubes (360 uL) and 10X Citrate buffer (1M NaCl, 100 ȝM monosodium phosphate, 100

ȝM monosodium citrate, pH 5.0) was added to a final concentration of 1X. For t = 0 min samples, ice-cold 100% EtOH (1.0 mL) was added immediately and the Eppendorf tubes stored at -20 °C. Other samples were placed in a 70 °C heating block for 15-90 minutes in 15 minute increments. Samples were removed from heat, precipitated in ice-cold

100% EtOH (1.0 mL), and then chilled at -20 °C for at least 20 minutes. Samples were centrifuged at 12,000 x G for 10 min. at 4 °C. The supernatant was discarded. The process was repeated once beginning with the addition of EtOH (1.0 mL). The DNA

Page | 109

pellet was resuspended in 700 ȝL H2O or TE buffer. Aliquots of 80 ȝL were put in 1.5 mL Eppendorf tubes. Any samples not used immediately were stored at -80 °C.

Genomic DNA time course of Cy7MX binding. A 100 mM stock solution of Cy7MX in

DMSO was diluted to 25 ȝM in H2O. Heat/acid treated DNA (t = 45 min, 80 ȝL) were warmed to 37 °C. In triplicate, Cy7MX (20 ȝL) was added to the DNA pipetting up and down to mix. Samples were kept at 37 °C in the dark for the following incubation times

(min): 0.5, 1, 2, 5, 10, 15, 30, and 60. Immediately after the incubation time, ice-cold

EtOH (1.0 mL) was added. Samples were quickly inverted to mix then stored in the dark at -78 °C in a dry ice-EtOH bath. The DNA was then purified by EtOH precipitation three times as described in the heat/acid treatment of genomic DNA. The purified DNA pellets were resuspended in 150 ȝL H2O. Aliquots (125 ȝL) of the samples were added to a black, clear-bottom 96 well plate (Corning) and analyzed with 750 nm excitation and

800 nm emission. Fluorescence was normalized to DNA concentration.

MX and Cy7MX genomic DNA competition assay. Solutions of MX and Cy7MX were prepared in H2O varying [MX] from 0-1.9 M and maintaining [Cy7MX] at 25 ȝM.

CAUTION: Solutions were kept vigorously in the dark as MX was observed to rapidly

(on the minute time scale) decompose Cy7MX in the presence of light. The ratios of

MX: Cy7MX used in this study were: 0, 0.01, 0.1, 0.5, 1, 5, 10, 50, 100, 250, 500, 1000,

2500, 5000, 10000, 25000, 50000, and 75000. For visual clarity, the data for the ratios

0.01, 0.1, 5, 10, 50, 250, and 500 were omitted from the graph in Figure 29 (page 104) as this did not affect the analysis. The following samples were prepared in triplicate: 20 ȝL

Page | 110

of MX/Cy7MX solution was added to 80 ȝL aliquots of heat/acid treated DNA (t = 45 min). Samples were incubated for 1 h at 37 °C in the dark. Ice-cold EtOH (1.0 mL) was added and the DNA was purified by EtOH precipitation three times as described in the heat/acid treatment of genomic DNA section. DNA was resuspended in 150 ȝL H2O.

Aliquots (125 ȝL) of the samples were added to a black, clear-bottom 96 well plate

(Corning) and analyzed with 750 nm excitation and 800 nm emission. Fluorescence was normalized to DNA concentration.

ARP and Cy7MX genomic DNA competition assay. As before except solutions of ARP and Cy7MX were prepared in H2O varying [ARP] from 0-75 mM and maintaining

[Cy7MX] at 1 ȝM. The ratios of ARP: Cy7MX used in this study were: 0, 0.1, 1, 10,

100, 500, 1000, 5000, 10000, 25000, 50000, and 75000.

MX and Cy7MX genomic DNA blocking assay. To 80 ȝL aliquots of DNA was added either 10 ȝL of MX (500 mM, pH 7) or vehicle control (500 mM NaCl). TE buffer was used as a -DNA control. Samples were incubated at 37 °C for 30 min. Then, either 10

ȝL of Cy7MX (500 ȝM) or 10 ȝL of 1%DMSO in water was added to samples. Samples were incubated at 37 °C in the dark for 1 h before DNA was purified by ethanol precipitation three times as described in the heat/acid treatment of genomic DNA. The purified DNA pellets were resuspended in 150 ȝL H2O. Aliquots (125 ȝL) of the samples were added to a black, clear-bottom 96 well plate (Corning) and analyzed with

700 nm excitation and 790 nm emission. Fluorescence was normalized to DNA concentration.

Page | 111

5.7.3 FUDR and MMS treatment of DLD1 cells

Preparation of DLD1 cell lines

UDG directed shRNA clones and scrambled targeted control shRNA clones were purchased from Sigma-Aldrich. According to manufacturer’s instructions from

Lipofectamine 2000 (Invitrogen), HEK293 cells were transfected to produce lentiviral particles that were used to infect DLD1 cells. Forty-eight hours after transfection, DLD1 cells were diluted for passage and selected with puromycin. The UDG knockdown levels were verified for RT-PCR and Western blot analysis.

Treatment of DLD1 cells with FUDR.

DLD1 shUDG (KD) or shControl (WT) cells were plated in Falcon brand 100x20 mm cell culture dishes in 10 mL medium (DMEM supplemented with 10% heat inactivated

FBS, penicillin/streptomycin, and nonessential amino acids) and incubated at least 16 h at

37 °C and 5% CO2 to ensure adhesion. To allow for cell proliferation, the following approximate numbers of cells were plated for each time point given in parenthesis: 4 million (24 h), 1 million (48 h), and 0.3 million (72 h). Two mL of medium were removed from each plate and replaced with 2 mL of FUDR solution at a final concentration of 0, 10, 200, or 1000 nM in 10 mL. For each time point, six plates of each

FUDR treatment group were prepared for each cell line. Cells were incubated at 37 °C and 5% CO2 with continuous FUDR exposure for 24, 48, or 72 h.

Page | 112

supernatant was removed and discarded. Cell pellets were suspended in TE buffer (2 mL) then treated with 10% SDS (240 μL) and RNase A (10 μL, 20 mg/mL purchased from Invitrogen) for at least 15 minutes at 37 °C. Then, proteinase K (10 μL, 20 mg/mL purchased from Invitrogen) was added and cell lysates were incubated for a least 15 minutes at 37 °C. Cell lysates were transferred to Phase Lock Gel Light 15 mL conical tubes purchased from 5Prime. Saturated phenol (2 mL, pH 6.6) was added to the cell lysates and the mixture was shaken vigorously. Chloroform (0.5 mL) was then added and the cell lysate mixtures shaken vigorously. The organic and aqueous phases were separated by centrifuging the gel tubes at 2,000 rpm for 10 minutes. After a second round of phenol-chloroform extraction and centrifugation, pure chloroform (2 mL) was added to the cell lysates, shaken, and centrifuged at 2,000 rpm for 10 minutes. The aqueous layer containing the isolated DNA was decanted into a clean 15 mL conical tube and precipitated with 100% EtOH (5 mL) and 3M sodium acetate (100 μL) by gentle rocking at 4 °C for at least 30 minutes. DNA was isolated by centrifuging at 3,000 rpm for 10 minutes. The DNA pellets were washed once with 70% EtOH (1.5 mL) and centrifuged at 3,000 rpm for 10 minutes.

Pure DNA pellets were suspended in 200 μL 1X UDG buffer. Samples were treated in triplicate with either UDG (1 μL, 5 units) or UDG storage buffer (see SSB activity assay procedures) and incubated at 37 °C for 1 h. After UDG incubation, 10 μL of each solution was removed and set aside for analysis with the ARP assay (Dojindo). A solution of Cy7MX (10 μL, final = 25 μM) was added to each sample, keeping the stock solution and all treated samples vigorously in the dark until the final analysis. Samples were incubated with Cy7MX for 1 h at 37 °C in the dark. After incubation, ice cold

Page | 113

100% EtOH (1 mL) and 3M sodium acetate (5 μL) were added to each sample. Samples were inverted to mix then chilled for at least 20 minutes at -20 °C. Samples were centrifuged at 12,000 G and 4 °C for 10 minutes. The supernatant was discarded. This

EtOH wash was repeated twice with 70% EtOH (1 mL) and no sodium acetate. DNA pellets were resuspended in H2O (minimum 150 μL). DNA concentrations were measured and adjusted to a maximum of 300 μg/mL. Aliquots (125 ȝL) of the samples were added to a black, clear-bottom 96 well plate (Corning) and analyzed with 760 nm excitation and emission scan of 790-847 nm with a 3 nm step size. Integrated fluorescence intensities were adjusted to DNA concentration and normalized to the

FUDR untreated KD +UDG sample.

Treatment of DLD1 cells with MMS.

DLD1 shControl (WT) cells at a density of 3 million cells per dish were plated in

Falcon brand 100x20 mm cell culture dishes in 10 mL medium (DMEM supplemented with 10% heat inactivated FBS, penicillin/streptomycin, and nonessential amino acids) and incubated 16 h at 37 °C and 5% CO2 to ensure adhesion. Solutions of MMS were prepared in serum free DMEM medium at concentrations of 0.05, 0.25, 1, 4, and 10 mM.

Medium was removed from the cells and replaced with the MMS or a serum free medium vehicle. Three plates were prepared per treatment group. Cells were incubated at 37 °C and 5% CO2 with continuous MMS exposure for 3 h. At the time points, cells were collected and DNA isolated as described above (Treatment of DLD1 cells with FUDR).

Pure DNA pellets were suspended in 190 μL water. A solution of Cy7MX (10

μL, final = 25 μM) was added to each sample, keeping the stock solution and all treated

Page | 114

samples vigorously in the dark until the final analysis. Samples were incubated with

Cy7MX for 1 h at 37 °C in the dark. After incubation, DNA was purified and analyzed as described above.

CHAPTER 6. MICROSCOPY AND IN VIVO IMAGING OF CANCER

6.1 Microscopy

6.1.1 Hypothesis and methodology

Microscopy is a useful tool to visualize fluorescent dyes in a complicated and potentially dynamic setting. Cells can either be treated while they are living or be fixed to stop dynamic repair. Fixed cells can also be permeabilized to allow free access of drugs through cell membranes. Having demonstrated that the AP site binding probes can detect and report on quantities of AP sites in purified DNA, the next step is to introduce the complexity of the cell. Aldehydes other than AP sites are present in cells, such as those formed by reactive oxygen species (ROS) interacting with lipid bilayer membranes.

These aldehydes contribute to off-target probe binding that could lead to high background and reduced quantities of probe available for AP site detection.

I hypothesized that AP site binding probes could be used to qualitatively visualize

AP sites in fixed, permeabilized cells. To develop a method for AP site detection, first a compatible method for fixation was identified. The next step was to evaluate nonspecific binding using a pair of probes with the same fluorescent scaffold but one of which was expected to be unable to bind AP sites. A preliminary study extending the results of

Page | 115

DLD1 cells treated with FUDR and exogenous UDG (see §5.5.1, page 104) was performed to determine if cells had a similar response as purified DNA.

6.1.2 Optimization of fixative

Fixation is a process of chemically preserving cells or histological samples in a life-like manner that resists decay.157 The most popular method for fixation of cells is paraformaldehyde (PFA). This agent creates crosslink between proteins and any unbound agent can be washed away. However, these protein crosslinks might be susceptible to displacement with a nucelophilic aminooxy group, which is present in all the AP site binding probes. Therefore, five conditions were surveyed to determine which cause the least nonspecific background signal: PFA, methanol, acetone, ethanol/acetic acid (19:1 v/v), and methanol/acetone (1:1 v/v).157-158 DAPI (4',6-diamidino-2- phenylindole, dihydrochloride) was used as a costain to locate the nucleus and as a standard to normalize the fluorescent signal. The AP site binding probe F422 (see

Scheme 10, page 51 for structure) was chosen for this study due to the brightness of the fluorescein core and its good aqueous solubility.

From the microscope images (Figure 32A), acetone and PFA show very bright fluorescence in otherwise untreated cells. Methanol, ethanol/acetic acid (19:1 v/v), and methanol/acetone (1:1 v/v) showed more modest nonspecific fluorescence.

Quantification of these signals (Figure 32B) supported these observations. In addition to high nonspecific probe binding, acetone makes a poor choice for fixative in the molecular biology setting where commonly used plasticware is rapidly degraded by the organic solvent. Despite relatively low nonspecific binding, the methanol/acetone mixture was rejected for this reason as well. Either alcoholic system showed equally adequate

Page | 116

nonspecific binding, but methanol was chosen for future studies due to the unpleasant odor of acetic acid.

Figure 32. (A) Confocal microscope images with 40x objective of cells fixed in various conditions then stained with F422 (green) and DAPI (blue). Bar = 50 ȝm. (B) Quantification of the F422 fluorescence intensity normalized to the DAPI signal. Bars represent the average F422 signal of each nucleus and error bars represent the standard deviation of the average.

6.1.3 Nonspecific binding with FEt2

Two signals contribute to background when visualizing AP site-targeted probes.

The first is off-target binding to aldehydes not associated with AP sites. This binding occurs through a reaction of the aldehyde with the aminooxy moiety on the probe. The term “off-target” is more appropriate than “nonspecific” in this case as the nucleophilic

Page | 117

addition to aldehydes is desired, but only those aldehydes in AP sites are targeted. The second signal comes from true nonspecific binding to chemical functional groups or cellular structures other than AP sites in DNA. Presumably, nonspecific binding is due mainly to the large fluorophore. Therefore, an analogue of F422, FEt2 (see Scheme 10, page 51 for structure), was prepared maintaining the fluorophore and the thiourea but substituting the hydrazine with two ethyl groups. This maintains the large fluorophore, backbone electronics, and steric size without AP site binding ability. The ROI of the AP site binding probes is confined to the nucleus. Nonspecific binding to other cellular structures such as the mitochondria and cellular membrane can be removed by drawing

ROIs around the DAPI nuclear stain. Compound FEt2 will thus report on general nonspecific DNA interactions including intercalation, minor groove binding, and backbone binding.

TMZ is a methylating chemotherapeutic drug used clinically for treatment of glioma.159 TMZ induced DNA damage is repaired by BER.75, 160 A glioma cell line,

T98G, was plated on glass cover slips and was treated with TMZ with continuous exposure for 0.5-24 h. At time points, cells were fixed in methanol then were stained with DAPI and either FEt2 or F422. Quantification of the F422 fluorescence in the nuclear ROIs delineated by DAPI showed a maximum number of AP sites after 1 hour

TMZ treatment (Figure 33). The basal signal was regained within 7 hours. These basal levels are higher than the nonspecific signal because AP sites are formed spontaneously in the cell as well. The low 4 hour time point could be due to upregulation of repair enzymes in response to damage reducing both exogenously and endogenously generated

AP sites.

Page | 118

Figure 33. Quantification of nuclear fluorescence by F422 and FEt2 following a TMZ time course. The maximum AP site formation is observed at 1 h TMZ treatment time. FEt2 fluorescence indicates nonspecific fluorescence accounts for ca. 50% of the basal F422 signal. Bars represent the average F422 or FEt2 signal of each nucleus and error bars represent the standard deviation of the average.

Figure 33 also indicated that the nonspecific binding of FEt2 is ca. 50% of the basal signal and 30% of the maximum signal. This nonspecific binding is considerable.

These results suggest that the fluorescein/thiourea scaffold is unsuitable to this particular application. While fluorescein is a common fluorophore in microscopy, this experiment demands a larger quantity of substrate in order to increase the ratio of specific to nonspecific binding. This result is not altogether disappointing as it provides a starting point that the novel AP site binding probes can improve upon.

6.1.4 FUDR dose response with Cy5MX

Previous results indicated that DLD1 KD cells accumulated a nearly 10-fold increase in Cy7MX fluorescence intensity in extracted DNA following 72 h exposure to

1000 nM FUDR and exogenous UDG treatment (Figure 30C, page 106). A similar experiment was envisioned in cells for microscopy. Cells were treated with 0, 200, or

Page | 119

1000 nM of FUDR continuously for 72 h. Cells were fixed in methanol before exogenous treatment with UDG. The amount of UDG was increased from 5 to 25 units and the incubation time extended from 1 h to 16 h in this experiment. The UDG solution was removed before staining with Cy5MX and DAPI. Representative images are shown in Figure 34A. In the FUDR untreated KD +UDG cells, Cy5MX is observed approximately uniformly throughout the cell, but the same cells with 1000 nM FUDR show some localization in the nucleus. Quantification of these data (Figure 34B) reveals that this effect is small with only a 2-fold increase in fluorescence.

Figure 34. (A) Representative epifluorescent images with a 40x objective of KD DLD1 cells with 1000 nM or without FUDR treatment stained with Cy5MX (red) and DAPI (blue). Bar = 50 ȝm. (B) Quantification of Cy5MX fluorescence intensity in regions of interest defined by the DAPI signal and normalized to the FUDR untreated KD + UDG signal intensity. Bars represent the average Cy5MX nuclear fluorescence from four fields-of-view and the error bars represent the standard deviation of the averages.

This experiment was expected to give results similar to those shown in Figure

30C (page 106). Methanol fixation removes lipids from cells,161 which makes it unlikely that the probe is trapped in the membranes. Further, methanol permeabilizes cells allowing probes unhindered entry into the cells. This leaves two explanations for the

Page | 120

differing results in the cell study: the probe is not sufficiently binding to the AP sites or the quantity of AP sites is below the limit of detection. Other aldehydes in the cell could be absorbing the probe leading to a low effective dose and not sufficient probes remaining to bind AP sites.

Even with a 10-fold increase in number, the total quantity of AP sites may be below the detection limits of the microscope. While the previous study showed that

UDG can generate AP sites in isolated DNA, it did not show that this could be done while other cell structures are present. Thus, the AP sites may not be forming reliably with the UDG treatment used here. The enzyme conditions could be optimized by examination of various concentrations and incubation times to ensure all uracil/FUra lesions are converted to AP sites and that AP sites are not degrading during the incubation time, respectively. Subsequent cell staining performed by Shriya Srinivasan in Dr. Burda's research group indicate that staining is improved when the quantity of

UDG is increased 10-fold.

6.1.5 Potential pitfalls in microscopy

Fixation and permeabilizeation allow probes to pass freely into the cells. Living cells include additional challenges such as membrane permeability and pumps to actively remove foreign species. For these AP site-targeting probes to be useful as imaging agents in living cells, either in culture or in xenograft models, the probes must bind to AP sites in situ.

Data from the extracted DNA experiment support the formation of AP sites in

DLD1 WT cells, which are actively repaired. Conversely, DLD1 KD cells accumulate lesions that are not repaired by BER and do not form intermediate AP sites. An

Page | 121

experiment was conducted treating DLD1 WT and KD cells with continuous exposure for

72 h with FUDR (0, 200, or 1000 nM) and Cy7MX (Figure 35). After the time points, free probe was removed. Following centrifugation, cell pellets were distinctly green, indicating that Cy7MX was entering the cells. DNA was then extracted from these cells and the fluorescence measured.

FUDR PhOH Cold Cy7MX Proteinase K CHCl3 EtOH RNase A

-Cy7MXfree 72 h

Figure 35. Schematic of concurrent, continuous FUDR and Cy7MX treatment in living cells. DNA was extracted and the fluorescence of Cy7MX was measured on the purified DNA.

No dose response was observed in either cell line (Figure 36). There are several possible explanations exist for these findings. First, the intense color in the cell pellets could mean that the probe is binding strongly to the cell membrane, effectively reducing the dosage from 25 μM to the equilibrium between membrane and cytoplasm. Lipid bilayer membranes contain negatively charged heads and hydrophobic tails. Compound

Cy7MX is positively charged but is also quite hydrophobic. Therefore, both Coulombic and nonpolar interactions favor association of the probe with the membranes over the aqueous cytoplams. A second possible explanation is that a large quantity of aldehydes is present in the cell cytoplasm, which act as a sponge for the probe. This, too, would reduce the effective concentration of the probe available to bind AP sites. Thirdly, previous experiments with the SSB activity assay (see §4.7, page 86) indicate that the probe-AP site lesion can be repaired by endonucleases in the cell. In this case, the SSB activity assay could be employed to test this hypothesis. A probe-labeled oligomer would

Page | 122

be treated with a cell extract mixture and the removal of the probe, or a SSB event, would be monitored by fluorescence to determine if repair can occur.

6.2 In vivo imaging in mouse xenograft models

6.2.1 Hypothesis and methodology

In addition to all of the complications of cell culture, probes used in animal models must contend with organs specialized for metabolism and excretion, bioavailability and biodistribution, and finding a balance between effective and toxic doses. I proposed preliminary in vivo studies of the NIR probe, Cy7MX, to identify potential areas of concern for future work. I hypothesized that in vivo imaging with

Cy7MX could identify major organs responsible for metabolism and clearance, a qualitative clearance lifetime, biodistribution, and an effective imaging dose. These data,

Page | 123

combined with the data from microscopy, will help develop second generation probes with improved imaging properties.

Gold nanoparticles (AuNPs) have been shown to improve tumor targeting through the enhanced permeability and retention effect.162 Further, AuNPs are amphiphilic.

Nonpolar molecules will noncovalently absorb into hydrophobic pockets of the particles while the probe-AuNP conjugate remains soluble in aqueous environments. We hypothesized that AuNPs could improve the solubility of Cy7MX and help localize biodistribution to the tumor region.

Nude mice, which are immune compromised and lack body fur, were used in this study. The former trait makes these mice ideal for xenograft implantation as a fully functional immune system would attack foreign cells. The latter trait makes the model great for fluorescent imaging applications as fur has autofluorescence and causes light scattering. Autofluorescence was further minimized by feeding the mice a special, alfalfa-free diet for at least one week prior to imaging. DLD1 (WT or KD) colon cancer or T98G glioma tumors were implanted in the flanks of mice. Tumors were allowed to grow and once they became palpable, imaging studies were performed. Free or AuNP- conjugated Cy7MX and IR-780 iodide (26 in Scheme 7, page 49), a control compound, were administered. Intravenous (i.v.) and intraperitoneal (i.p.) injections were performed to determine the better route of administration. Biodistribution and clearance were also evaluated.

6.2.2 Preliminary evaluation of AuNP tumor targeting

Compound 26 (Scheme 7, page 49) was used as a control compound for imaging studies. In collaboration with Dr. Clemens Burda and Shriya Srinivasan, PEGylated

Page | 124

AuNPs were prepared. Compound 26 was used either free in solution or conjugated to the AuNPs at equimolar concentrations. Free and conjugated 26 were administered i.v. or i.p. at a dose of 0.13 mg/kg and the mice were imaged over 24 hours. Two tumor cell lines, T98G and DLD1 WT, were examined to see if the uptake into one cell line was stronger or equal to the other.

Despite the equimolar dosages, the AuNP-conjugated mice were much brighter than the free 26 (Figure 37 and Figure 38). This may be due to the hydrophobic environment of the AuNP shielding the dye from water or dissolved oxygen that could decompose the probe over time. Thus, probe solutions were freshly prepared and rigorously kept in the dark in all subsequent experiments to minimize decomposition.

Page | 125

Figure 37. Mice imaged in a prone position following tail vein administration of free or AuNP- conjugated IR-780. Xenograft tumors of DLD1 WT and T98G cells were implanted on the left and right flanks, respectively.

Page | 126

Figure 38. Mice imaged in a supine position following tail vein administration of free or AuNP- conjugated IR-780.

No fluorescence was observed in background images indicating that all fluorescence signal after dye administration was due to the dye alone. Two hours after administration, the i.v. treated mice showed the dye in systemic circulation while the i.p. treated mice showed the dye mostly confined to the gut (Figure 37 and Figure 38).

Within 24 h, the dye had localized in the tumor regions when mice were viewed in a prone position. As expected, the AuNP-conjugated dye was localized in the tumor.

Surprisingly, uptake of the free dye was observed in T98G tumors as well (Figure 37).

Regardless of i.p. or i.v. injection, considerable dye was retained at the site of injection (the tail vein entry point has been cropped from the images for clarity). The i.p. injection leads to considerable nonspecific signal in the region of the body where the

Page | 127

most important metabolic and excretory organs reside. The i.v. injection, conversely, causes nonspecific signal in the tail, which has been cropped from the image for scaling purposes. Thus, the i.v. injection was chosen as a better method than i.p.

After 24 h, mice were sacrificed and the organs were imaged (Figure 39A). To account for different organ sizes, these images were quantified based on the mean fluorescence in the ROI. All organs were normalized to the liver fluorescence, to account for and scale to the different fluorescence intensities between animal samples and to facilitate the comparison, as the liver had consistently high fluorescence,. Due to the preliminary nature of this experiment, only one mouse was used per condition.

Quantification of these data (Figure 39 B) indicate strong uptake of the free dye by the liver and digestive organs. The large uptake in the T98G tumor in the i.p. sample of the free dye appears to be an artifact and would require a larger sample size to confirm or refute. AuNP-conjugated dye had the most uptake into the liver and tumor regions. No major difference was observed between i.v. and i.p. administration routes.

Page | 128

Figure 39. Biodistribution of IR-780 administered directly or conjugated to AuNPs 24 h after administration. (A) Fluorescent imaging of Bo, bone; Br, brain; C, colon; DLD1, DLD1 xenograft tumor; H, heart; K, kidneys; Li, liver; Lu, lungs; M, muscle; Sp, spleen; St, stomach; T98G, T98G xenograft tumor. (B) Quantification of fluorescence from biodistibution normalized to the liver signal. For all conditions, n=1 mouse and error bars represent standard deviation of fluorescence intensity within the ROI.

From these data, one may conclude that AuNPs improve the delivery of a model

NIR dye to the tumors. Liver metabolism appears to be the predominant metabolism and elimination pathway for the control dye. The T98G tumors appear to have slightly higher

Page | 129

uptake than the DLD1 for the free dye, although this should be confirmed with a sample size larger than one, but the two cell lines had consistent uptake from the AuNP- conjugated samples. The dosage of the free dye should be increased in future studies to ensure a strong fluorescence signal for detection.

6.2.3 Direct administration of probe

Having shown considerable liver metabolism of the control dye, the next step was to evaluate the metabolism of the AP site-targeted probe, Cy7MX. Along with metabolism, the clearance of the compound was also observed over a 12 day period. The free dye was given i.v. to mouse models containing one DLD1 KD and one DLD1 WT tumor on either flank. The dosage was increased from 0.13 mg/kg to 0.40 mg/kg. Mice were treated with FUDR or a vehicle control to determine if differential uptake due to

BER could be observed. Based on previous studies, DLD1 WT tumors treated with

FUDR were expected to undergo BER and create AP sites. FUDR untreated cells were expected to have baseline repair and FUDR treated KD cells were expected to accumulate damage but not have active BER. The Cy7MX fluorescence intensity was predicted to be directly proportional to the extent of BER.

Imaging of probe clearance over 12 days showed that the probe persisted through the entire course of the experiment and did not return to baseline fluorescence levels

(Figure 40 and Figure 41). No free probe accumulated in tumors regardless of tumor or treatment type (Figure 40). Within 2 h, most of the fluorescence signal was observed in the liver and digestive tract (Figure 41). After 24 h, the initial brightness decreased but signal was still observed in the liver throughout the study.

Page | 130

Figure 40. Mice imaged in a prone position following tail vein administration of Cy7MX. Xenograft tumors of DLD1 WT and KD cells were implanted on the left and right flanks, respectively.

Figure 41. Mice imaged in a supine position following tail vein administration of Cy7MX.

Page | 131

After observing the clearance for 12 d, the mice were sacrificed and the organs removed. Residual fluorescence was imaged and quantified (Figure 42). The liver and spleen showed the maximum fluorescence. Very little fluorescence was observed in the tumors. This result is not unexpected as the FUDR treatment was stopped on day 5 due to the treated mice losing too much weight.

Figure 42. Biodistribution of Cy7MX administered directly 16 days after treatment. (A) Fluorescent imaging of Bo, bone; Br, brain; C, colon; H, heart; K, kidneys; KD, KD tumor; Li, liver; Lu, lungs; M, muscle; Sp, spleen; St, stomach; WT, WT tumor. (B) Quantification of fluorescence from biodistribution. For all conditions, n=1 mouse and error bars represent standard deviation of fluorescence intensity within the ROI.

These data indicate that, as with the IR-780 iodide control dye, Cy7MX is metabolized predominantly through the liver. Persistence of the probe in the mouse body after 16 days indicate that the clearance rate is low and will need to be addressed in second generation compounds. The dosage of 0.40 mg/kg is adequate for visualization.

No probe uptake was observed in tumors. Perhaps AuNP conjugation may benefit imaging by assisting probe availability to the tumor region through decreased distribution. Further, as the free probe accumulates in the tumor, AuNPs may shield the probe from the first pass effects.

Page | 132

6.2.4 Preliminary evaluation of AuNP-conjugated Cy7MX

Gold nanoparticles were loaded with 100 equivalents of Cy7MX and administered to mice i.v. at a probe dose of 0.40 mg/kg. As before in the free Cy7MX study, DLD1

KD and WT tumors were implanted on either side of the mice. Mice were treated with

FUDR or vehicle control until treated mice became too underweight to continue. The purpose of this study was to monitor the clearance of the probe over time while determining if the AuNP improved tumor delivery.

Fluorescence data reveal that the clearance rate was slow with baseline fluorescence not regained after 12 d (Figure 43 and Figure 44). Very little tumor uptake was observed above the baseline with no correlation to drug treatment or cell line (Figure

43). After two hours, the probe was observed in the liver and digestive organs and persisted in the liver through the course of the study (Figure 44).

Page | 133

Figure 44. Mice imaged in a supine position following tail vein administration of AuNPs loaded with Cy7MX. Page | 134

After 12 days, the mice were sacrificed and the organs were imaged to determine biodistribution. As before, the maximum fluorescence was observed in the liver, but much less fluorescence was observed in the spleen with AuNP-conjugated probe than with free probe (Figure 45). For this result to be significant, however, the experiment would need to be repeated on a larger than one mouse per treatment group sample size.

Uptake in the other organs was relatively low. Low tumor uptake was not surprising as

FUDR treatment was stopped after the first day of imaging when treated mice began to show too much weight loss.

Figure 45. Biodistribution of AuNP-conjugated Cy7MX 11 days after treatment. (A) Fluorescent imaging of Bo, bone; Br, brain; C, colon; H, heart; K, kidneys; KD, KD tumor; Li, liver; Lu, lungs; M, muscle; Sp, spleen; St, stomach; WT, WT tumor. (B) Quantification of fluorescence from biodistibution. For all conditions, n=1 mouse and error bars represent standard deviation of fluorescence intensity within the ROI.

These results were inconsistent with the previous imaging using IR-780 iodide.

One explanation could be that Cy7MX was not conjugated as deeply in the AuNP as the control dye had been. This could have been due to the sample preparation. Additional studies to optimize the absorption of the probe into the hydrophobic pocket of the AuNP are currently in progess by our collaborator, Dr. Clemens Burda, and his students.

Page | 135

6.2.5 Post mortem detection of AP site probe in isolated DNA

As previously described in §6.1.3 (page 117), nonspecific binding could also be a major source of error in these experiments. AuNPs were expected to deliver the probe to the tumor, but that is not a guarantee that the observed fluorescence signal is due to AP site binding. Therefore, after fluorescence imaging with free and AuNP-conjugated

Cy7MX indicated the level of compound within each organ, the organs were digested and the DNA was extracted. The fluorescence was measured in both the isolated DNA

(specific binding) and in the remaining filtrate after DNA was extracted (nonspecific and off-target binding). To facilitate sample preparation and ensure reproducibility, a commercial kit using a silica column was used to isolate DNA instead of the phenol- chloroform method described in Chapter 5 (page 97). The filtrate refers to that portion of the cell extract that passes through the column that retains DNA. Two control samples, free Cy7MX and DNA-bound Cy7MX, were included.

Fluorescence measurements of isolated DNA showed no signal above the baseline for all samples, including Cy7MX bound to DNA (data not shown). One reason for this was very low yields of DNA, with an average final concentration of 35 ± 20 μg/mL of

DNA. Analysis of the filtrate confirmed DNA yields were low (Figure 46). The

DNA+Cy7MX sample that had been prepurified to remove free probe showed considerable fluorescence in the filtrate. The initial concentration of DNA in the control sample was 300 μg/mL. Adjusting for volume changes, this result indicates that ca. 90% of the DNA was able to pass through the silica column without retention.

Page | 136

Figure 46. Fluorescence from filtrate following organ digestion. For this preliminary experiment, n=1 for each sample and the bar represent the

From these results, the silica column extraction can be eliminated as a tool to quantify specific fluorescence for this work due to lack of retention of DNA on the column. The specific origin of the fluorescence from the liver, spleen, and other organs cannot be determined from this result due to the low recovery of DNA and the contamination of the filtrate with DNA. This is an important experiment to demonstrate the nature of the probe association with each organ and must be optimized for future experiments. Phenol-chloroform extraction gave adequate DNA yields (see §5.7.3, page

112). and is effective in removing free probe, which has a strong affinity for halogenated solvents.

6.2.6 Potential pitfalls in animal imaging

Animal imaging is at the mercy of pharmacokinetics and pharmacodynamics. For successful imaging, these limitations must be explored thoroughly but always under

Page | 137

practical time and ethical animal usage restraints. However, a probe optimized for cell culture staining will not necessarily translate to an optimized in vivo imaging agent.

Therefore, conducting simultaneous microscopy and xenograft studies may provide the most information for developing second generation AP site binding probes.

While individual strains of mice have been developed to maintain genetic similarity, different strains will inevitably vary. An example of this variance was observed for myelin imaging in the three strains of mice used in §2.4 (page 24). Hence, the optimized imaging parameters in nude mice may not apply to other mouse strains or mammalian species.

To ensure scientific validity of the data, the sample size of each treatment group will have to be at minimum three animals, with a statistics determining the appropriate number based on initial data. Biodistribution and clearance data were evaluated separately here, but ideally they would be measured together by sacrificing mice at various time points. The two largest potential pitfalls in animal imaging would be either using too few mice for scientific validity or using too many mice for ethical animal guidelines.

6.3 Discussion and conclusions

Histological stains are used to visualize elements of cells and tissues. Fluorescent

AP site binding probes could be used as such histological stain to evaluate the quantities of AP sites or lesions that can be converted to AP sites (such as uracil following UDG treatment) in cell culture and tissue samples. In cell culture, the BER response of individual cell lines to new or existing drugs could be monitored to help evaluate the drugs and understand differential cytotoxicity. In tissue samples, such as those obtained

Page | 138

from clinical biopsy, a patient’s individual response could be determined based on the degree of BER. A patient sample showing high BER would indicate that the patient is not receiving therapeutic benefit of the drug and a new treatment regimen should be considered.

Likewise, new drug development in animal models can be facilitated with AP site binding probes. Xenograft models of well characterized cell lines could be treated concurrently with a new, DNA damaging drug candidate and the AP site binding probe.

The probe could report on the efficacy of the drug based on the extent of BER.

Applications are not limited to cancer treatments but to a host of diseases and processes associated with oxidative damage. Other studies indicate that the binding of the probe to the intermediate AP site can block further BER and produce a therapeutic benefit.111, 163

These experiments presented here are preliminary studies to identify complications of applying AP site probes to imaging DNA repair in situ. In particular, nonspecific binding not only contributes to a high background, but also consumes the probe so that less is available to maximize signal to noise. The F422/FEt2 staining presents a simple method to evaluate nonspecific binding in the cell environment and can be used to screen probes.

A report by Frangioni and coworkers suggests that liver metabolism is extensive for cyanine dyes with positive charges. They observed nonspecific organ uptake for dyes with net charges but not for neutral zwitterions. Their research suggests that to minimize liver metabolism and nonspecific organ uptake, the hydrophobic cyanine core should be shielded with charged functional groups but have a net zero charge.164

Page | 139

The in vivo fate of cyanine-based Cy7MX was consistent with the findings of

Frangioni and cowrokers. Liver metabolism was extensive and the probe was distributed throughout the body. Structure-activity relationship studies to develop second generation

AP site binding probes should consider the observation of the influence of charge on the in vivo behavior of the cyanine-based probes.

6.4 Materials and methods

6.4.1 General methods

DLD1 cells (ATCC #CCL-221) were obtained from the laboratory of Sanford

Markowitz at Case Western Reserve University. T98G cells were obtained from ATCC

(#CRL-1690). Cells were grown in Dulbecco's Modified Eagle's medium (DMEM) supplemented with heat inactivated fetal bovine serum, penicillin (100 I.U./mL), streptomycin (100 ȝg/mL), and nonessential amino acids (Gibco # 11140-050) at 37 °C and 5% CO2.

DLD1 KD cells were prepared by Yan Yan according to the following procedure:

UDG directed shRNA clones and scrambled targeted control shRNA clones were purchased from Sigma-Aldrich. According to manufacturer’s instructions from

Page | 140

6.4.2 Microscopy

Optimization of fixative. T98G cells were plated at a density of 0.1 x 106 cell/well in a 6 well plate fitted with sterile microscope glass coverslips. Cells were grown in 1.5 mL medium supplemented with 0.5% v/v DMSO (vehicle for drug treatments) overnight.

The media was removed and the cells were washed with PBS. Cells were fixed according to Table 6. Methanol and acetone were removed and samples were allowed to dry for 20 min. PFA and ethanol samples were removed and slides rinsed with PBS (2 x

2 mL, 5 min. r.t.). PFA and ethanol were rinsed with MeOH (-20 °C, 2 mL, 5 min.) then dried for 20 min. All samples were permeabilized with 0.1% Triton-X in PBS (2 mL, 10 min. r.t.) then washed with PBS (2 x 2 mL, 5 min. r.t.). Cells were incubated with F422

(1 mL, 5 ȝM in PBS, 1 h, r.t. in the dark) then washed with PBS (2 x 2 mL, 5 min. r.t. in the dark). Cells were costained with DAPI (1 mL, 10 ȝM in PBS, 5 min. r.t. in the dark) and then washed with PBS (2 x 2 mL, 5 min. r.t. in the dark). Coverslips were dried and mounted on slides with Vectrashield hard set (Vector Laboratories #H-1400). Slides were imaged on Zeiss LSM 510 confocal microscope with a Plan-Neofluor 40x/1.3 numerical aperture oil immersion DIC corrected lens. F422 was excited at 488 nm and collected from 500-550 nm. DAPI was excited at 750 nm (two-photon) and collected from 435-485 nm. Images were acquired at 2048x2048 pixels. Fluorescence intensities were measured in ImageJ in regions of interest defined by the DAPI signal. The F422 fluorescence intensity was normalized to the DAPI signal.

Page | 141

Table 6. Methods for fixative comparison.

Fixative Time (min.) Temperature °C

PFA (4% w/v) 10 25

AcOH/EtOH (1:19 v/v) 10 25

MeOH 5 -20

Acetone 5 -20

MeOH/acetone (1:1 v/v) 5 -20

Abbreviations in this table: PFA, paraformaldehyde; AcOH, acetic acid; EtOH, ethanol; MeOH, methanol.

Nonspecific binding with FEt2. T98G cells were plated at a density of 0.1 x 106 cell/well in a 6 well plate fitted with sterile microscope glass coverslips. Temozolomide (TMZ) was added to the medium at a final concentration of 350 ȝM TMZ and 0.5% v/v DMSO.

Cells were incubated with TMZ for 0, 0.5, 1, 2, 4, 7, 16, and 24 h. At the end of the time points, the medium was removed and cells were washed with PBS (2 mL). Cells were fixed in MeOH (2 mL, -20 °C, 5 min.). The MeOH was removed and samples were dried for 20 min at r.t. Cells were permeabilized with 0.1% Triton-X (1 mL, 10 min.) then washed with PBS (2 x 1 mL, 5 min.). Cells were incubated with either F422 or FEt2 (1 mL, 100 ȝM in 1% DMSO v/v in PBS) for 1 h at r.t. in the dark. The staining solution was removed and cells were washed with PBS (2 x 1 mL, 5 min. r.t. in the dark).

Samples were costained with DAPI (1 mL, 10 ȝM, 5 min.) then washed with PBS (2 x 1 mL). Coverslips were dried and mounted on slides with Vectrashield hard set (Vector

Laboratories #H-1400). Slides were imaged on a Nikon Eclipse TE2000-S epifluroescent microscope with a 40x CFI Plan Fluor 0.60 numerical aperture, air immersion lens.

DAPI was visualized with a DAPI filter set (ex. 340-380, DM 400, BA 435-485) and

Page | 142

F422 and FEt2 were visualized with a FITC filter set (ex. 465-495, DM 505, BA 515-

555). Fluorescence intensities were measured in ImageJ in regions of interest defined by the DAPI signal.

FUDR dose response. DLD1 WT and KD cells were plated at a density of 0.02 x 106 cell/well in a 6 well plate fitted with sterile microscope glass coverslips. Cells were grown in 2 mL medium for 24 hours. The medium was removed and replaced with medium supplemented with 0, 200, or 1000 nM FUDR. Cells were grown for 72 hours with continuous FUDR exposure. The medium was removed and cells were washed with

PBS (2 x 2 mL). Cells were fixed in MeOH (2 mL, -20 °C, 10 min.) then permeabilized in 0.1% Triton-X (1 mL, 10 min.). The cells were washed briefly in PBS (2 x 2 mL) then for 1 h (1 x 2 mL). 1X UDG reaction buffer (2 mL) was added to all the samples then

UDG (5 ȝL, 25 Units) or UDG storage buffer (5 ȝL) were added. Then samples were mixed gently at r.t. for 1 h then incubated at 37 °C for 16 h. The enzyme or vehicle solutions were removed and replaced with Cy5MX (2 mL, 25 ȝM, in MeOH). Cells were incubated with gentle stirring at r.t. for 1 h in the dark. The probe solution was removed and cells were washed with MeOH (2 x 2 mL, 5 min.). Samples were costained with

DAPI (2 mL, 10 ȝM in water) for 5 min. at r.t. in the dark. The costain solution was removed and cells were washed with water (2 x 2 mL, 5 min.) in the dark. Coverslips were dried and mounted on slides with Vectrashield hard set (Vector Laboratories #H-

1400). Slides were imaged on a Nikon Eclipse TE2000-S epifluroescent microscope.

DAPI was visualized with a DAPI filter set (ex. 340-380, DM 400, BA 435-485) and

Cy5MX was visualized with a Texas red filter set (ex. 540-580, DM 595, BA 600-660).

Page | 143

Four images were acquired per slide in separate, non-overlapping regions. Fluorescence intensities were measured in ImageJ in regions of interest defined by the DAPI signal.

Concurrent FUDR and Cy7MX treatment in living cells. DLD1 WT and KD cells were plated at a density of 2.0 x 106 cell/plate in a 10 cm dishes with 12 plates/cell line. Cells were grown in 5 mL medium overnight. To each plate was added solutions of FUDR (2 mL, final concentrations 0, 200 and 1000 nM) and Cy7MX (2 mL, final concentration 5

μM) in complete medium. Samples were prepared in quadruplicate. Cells were incubated at 37 °C and 5% CO2 for 72 hours in the dark with continuous FUDR and

Cy7MX exposure.

At the time points, the media was removed and cells were rinsed with PBS. Cells were trypsinized with 0.25% trypsin (1 mL) and transferred to 15 mL conical tubes in 5 mL PBS. The conical tubes were centrifuged at 1,700 rpm to pellet the cells. The supernatant was removed and discarded. Cell pellets were suspended in TE buffer (2 mL) then treated with 10% SDS (240 μL) and RNase A (10 μL, 20 mg/mL purchased from Invitrogen) for at least 15 minutes at 37 °C. Then, proteinase K (10 μL, 20 mg/mL purchased from Invitrogen) was added and cell lysates were incubated for a least 15 minutes at 37 °C. Cell lysates were transferred to Phase Lock Gel Light 15 mL conical tubes purchased from 5Prime. Saturated phenol (2 mL, pH 6.6) was added to the cell lysates and the mixture was shaken vigorously. Chloroform (0.5 mL) was then added and the cell lysate mixtures shaken vigorously. The organic and aqueous phases were separated by centrifuging the gel tubes at 2,000 rpm for 10 minutes. After a second round of phenol-chloroform extraction and centrifugation, pure chloroform (2 mL) was

Page | 144

added to the cell lysates, shaken, and centrifuged at 2,000 rpm for 10 minutes. The aqueous layer containing the isolated DNA was decanted into a clean 15 mL conical tube and precipitated with 100% EtOH (5 mL) and 3M sodium acetate (100 μL) by gentle rocking at 4 °C for at least 30 minutes. DNA was isolated by centrifuging at 3,000 rpm for 10 minutes. The DNA pellets were washed once with 70% EtOH (1.5 mL) and centrifuged at 3,000 rpm for 10 minutes.

DNA pellets were resuspended in H2O (150 μL). DNA concentrations were measured and adjusted to a maximum of 300 μg/mL. Aliquots (150 ȝL) of the samples were added to a black, clear-bottom 96 well plate (Corning) and analyzed with 760 nm excitation and emission scan of 790-850 nm with a 3 nm step size. Integrated fluorescence intensities were adjusted to DNA concentration and normalized to the average.

6.4.3 In vivo imaging

All animal experiments were performed in accordance with the guidelines approved by the Institutional Animal Care and Use Committee of Case Western Reserve

University (Protocol 2010-0007). NCR nude mice were obtained from AA7 at Case

Western Reserve University, Cleveland, OH. Mice were fed a special alfalfa free diet for at least one week before imaging to minimize autofluorescence. Imaging was performed on a Xenogen IVIS 200 with fluorescent and photographic channels. Fluorescent settings were 1 s exposure, high resolution small binning, f/stop 2, high fluoro lamp level, and

ICG excitation and emission filter sets. Photographic setting were 0.2 s exposure, medium binning, and f/stop 8. The field of view was set to 18.4 cm. The subject height was1.5 cm. Mice were anesthetized with 2% isoflurane for induction and 1-1.5% for

Page | 145

maintenance. Mice were sacrificed by CO2 before organs were removed for biodistribution measurements. AuNPs were prepared by Shriya Srinivasan. DNA concentrations were measured on a Nanodrop ND-1000 spectrometer. Fluorescence data were collected on a Tecan Infinite M200 scanner in a Corning 96 well black, clear bottom plate.

Direct and AuNP-conjugated control dye. Subcutaneous flank tumors were prepared by injecting 15-20 million T98G and DLD1 WT cells on the right and left sides of nude mice (relative to the prone position of the head), respectively. IR-780 iodide (Sigma

#425311) was used as a control compound. A background image was collected immediately before the dye was dosed either alone or conjugated to AuNPs at 0.13 mg/kg either i.p. or i.v. The ratio of dye to AuNP was 1:100. Images were collected every 3 minutes (x4), every 10 minutes (x7), every 15 minutes (x5), then at 24 h. Mice were sacrificed and the following organs were imaged: bone, brain, colon, DLD1 tumor, heart, kidneys, liver, lungs, muscle, spleen, stomach, and T98G tumor.

Direct administration of AP site probe. Subcutaneous flank tumors were prepared by injecting 15-20 million DLD1 KD or WT cells on the right and left sides of nude mice

(relative to the prone position of the head), respectively. Mice were treated i.p. with

FUDR (25 mg/mL in sterile saline) at a dose of 200 mg/kg on days -2, 1, 2, and 5. A background image was collected immediately before Cy7MX (0.1 mg/mL in 1% DMSO in sterile saline) was administered i.v. on day 1 at a dose of 0.4 mg/kg. Images were collected at times relative to Cy7MX treatment: >10 min, 30 min, 1 h, 2 h, 8 h, 24 h, 26

Page | 146

h, 31 h, 48 h, 5 d, 6 d, 7 d, 8 d, 9 d, 12 d, and 16 d. Mice were sacrificed and the following organs were imaged: bone, brain, colon, heart, kidneys, KD tumor, liver, lungs, muscle, spleen, stomach, and WT tumor.

AuNP-conjugated AP site probe. Subcutaneous flank tumors were prepared by injecting

15-20 million DLD1 KD or WT cells on the right and left sides of nude mice (relative to the prone position of the head), respectively. Mice were treated i.p. with FUDR (25 mg/mL in sterile saline) at a dose of 200 mg/kg on days -7, -5, -3, and 1. A background image was collected immediately before AuNP-conjugated Cy7MX (1 μM AuNP/100

μM Cy7MX in sterile saline) was administered i.v. on day 1 at a dose of 0.4 mg/kg.

Images were collected at times relative to Cy7MX treatment: >10 min, 30 min, 1 h, 2 h,

23 h, 47 h, 71 h, 4 d, 7 d, and 11 d. Mice were sacrificed and the following organs were imaged: bone, brain, colon, heart, kidneys, KD tumor, liver, lungs, muscle, spleen, stomach, and WT tumor.

Post mortem detection of AP site probe in isolated DNA. Tissues from direct and AuNP- conjugated Cy7MX imaging were used. Four control samples were prepared: blank,

Cy7MX only (20 nmol), purified DNA only (300 mg/mL), and purified DNA bound to

Cy7MX (300 mg/mL). DNA was isolated using the Invitrogen PureLink Genomic DNA mini kit (#K1820-20) following manufacturer’s instructions. Digestion buffer volume was increased to 360 μL. Samples were heated to 55 °C and vortexed briefly every 30 minutes for 5.5 h then left undisturbed for 16 h. Filtrate from the DNA binding step (i.e. digested cell components and probe not bound to DNA) was saved for analysis. Samples

Page | 147

were eluted from spin columns with 2 x 100 μL elution buffer. DNA concentrations were measured and found to be 35 ± 20 μg/mL (including for control samples).

CHAPTER 7. FUTURE DIRECTIONS AND CONCLUSIONS

7.1 Considerations for probe design

The probes described in this text are first generation contrast agents. These lead agents provide a framework for subsequent generation probes, which would be modified to improve specific properties of the agents. Based on the observations in this work, three properties that should be considered in designing new chemical entities include aqueous solubility, probe stability, and biological compatibility.

7.1.1 Solubility

Extended aromatic conjugation is common for organic fluorophores, especially as emission wavelengths increase, making these molecules quite hydrophobic. Sulfonate, carboxylate, and ammonium groups are routinely added to improve aqueous solubility of organic fluorophores at the cost of decreased membrane permeability. These chemical changes to the molecule also influence pharmacological and chemical properties.

CIC and the coumarin-based AP site probes all have very poor aqueous solubility.

Chemical modifications to CIC to improve water solubility may limit its myelin binding ability as myelin is lipophilic. Rather than risk sacrificing binding ability, vehicle formulations to increase solubility of the drug should be considered. For imaging studies,

DMSO and surfactant Tween® 80 were used to maintain CIC in aqueous solution for i.v.

Page | 148

administration. This formulation was adequate for the purposes of preliminary studies but should be optimized in future work. Several review articles detail formulations to improve solubility and CIC may benefit in particular with the use of cyclodextrins, micelles, or surfactants.165-168 In addition, for human use, an oral formulation might be preferred that would require additional measures to enhance solubility in gastrointestinal fluids,169 for example by conjugation with cholic acids.170

The cyanine-based AP site probes were freely soluble in aqueous solution up to

500 μM (higher concentrations were not examined) when diluted from a 100 mM stock in DMSO. If increased solubility is desired, then addition of anionic groups is recommended as this technique may also serve to decrease nonspecific binding and improve stability.

7.1.2 Stability

Fluorophores are notorious for photobleaching, a type of degradation induced by the absorption of photon energy. Cyanine dyes undergo a “suicide” photobleaching process wherein the excited state sensitizes singlet oxygen, which in turn attacks the polymethine backbone and destroys the fluorescent core.171 Some fluorophores are more susceptible to photobleaching than others; for example, CIC is much slower to decompose than MeDAS (see Scheme 1, page 17) under similar conditions. Chemical modifications can help to reduce photobleaching. In the case of cyanine dyes, increasing the rigidity or the steric bulk around the polymethine chain both decrease photobleaching.171 Anionic sulfonate groups attached by an alkyl chain to the heterocyclic nitrogen atoms improve stability by adding steric bulk and preventing aggregation.171 Observations of Cy7MX in the presence of MX and light showed the

Page | 149

nucleophilic alkoxyamine rapidly degraded the fluorophore. Addition of anionic sulfonate groups to the cyanine-based AP site binding probes may reduce their sensitivity to nucleophilic attack by increasing electron density around the periphery of the molecules.

7.1.3 Biological compatibility

Pharmacokinetics is a chief concern of biological compatibility. For CIC and other myelin imaging agents, this includes permeability through the BBB and the blood- to-spinal cord barrier. Altering the fluorophore to improve water solubility may decrease its distribution to the brain. Water-octanol partition (LogP) calculations to measure the ratio of hydrophilicity to lipophilicity of a molecule should be evaluated for every proposed myelin imaging agent to ensure a potential to cross these barriers.172 The LogP, and the other criteria in Lipinski’s rule of five, are excellent guidelines and should be taken into consideration when designing new drug-like compounds.173

Another aspect of medicinal chemistry critical to improving second generation probes is structure activity relationship (SAR). After a lead compound is detected, for example by screening AP site probes in the SSB activity assay, a library of derivative compounds can be prepared and screened for biological activity. Derivatives of Cy7MX should evaluate the addition of anions, sulfonates and carboxylates, to help reduce membrane binding and nonspecific DNA interactions. The linker connecting the fluorescent scaffold and the AP site binding group could be modified to improve photostability (as described above) or to alter the absorption and emission. Substituents on the linker and fluorescent backbone could be surveyed to optimize pharmacologic properties, i.e. lowering metabolism rates and products. Incorporation of radiolabels can

Page | 150

be used to identify and characterize products of metabolism. Structure activity relationships should be used to optimize the imaging properties and minimize toxicity.

7.2 Considerations for cellular imaging

7.2.1 Repair of probe-AP site lesion

Fluorescent AP site binding probes may be used in both living and dead cells. In living cells, the dynamics of DNA repair may complicate imaging. Based on the results of the endonuclease survey in §4.7 (page 86) and the reports of endonucleases III and IV repairing MX-AP site lesions,149 certain enzymes can recognize and repair the probe-AP site lesions. A simple experiment to evaluate this potential in a given cell line involves preparing a probe labeled oligomer, treating it with a cell extract, and evaluating it with the SSB activity assay. The protein expression in cell extracts may be influenced by drug treatment, with the repair enzymes being upregulated in response to DNA damage. Thus, the extracts should be taken from drug treated and untreated cells. If many cell lines are examined and varying degrees of repair detected, Western blot analysis of the likely enzymes involved (e.g. endonucleases III, IV, and VIII) could be performed to determine a correlation between a given protein and probe-lesion repair.

7.2.2 Off-target aldehyde binding

Considerable off-target and nonspecific binding was observed in microscopy studies of AP site binding probes. The nonspecific binding is best addressed by chemical modifications as described in §7.1 (page 148) to decrease binding to structures such as cell membranes and mitochondria. Off-target binding to other aldehydes in the cell is a more subtle issue than nonspecific binding. Chemical modification to increase the

Page | 151

affinity for DNA over cytosolic aldehydes could be one route. Although, Cy7MX displayed much higher potency than MX or ARP in vitro, this potency did not translate into cells. One method then to decrease off-target binding would be to block cytosolic aldehydes with a less selective agent before treatment with the more selective AP site probe. For example, in the FUDR treatment of DLD1 cells presented in §6.1.4 (page

119), cells could be pretreated with MX before addition of exogenous UDG and Cy5MX, which would bind MX to non-AP site aldehydes before converting uracil/FUra lesions to

AP sites. For situations where lesions are continuously present, the differential reactivity of MX and Cy5MX could be exploited. The probes would compete for all the aldehydes, but the cyanine-based probe could have sufficient affinity for DNA over other aldehydic species so as to tip the scales in favor of MX binding cytosolic aldehydes and Cy5MX binding AP sites. The doses, timing, and probe pairing would require optimization for this technique to be useful.

7.2.3 Nonspecific binding

Chemical modification can reduce nonspecific binding, but without understanding the culprit, most modifications will be at random. Logical modification requires understanding the nature of the nonspecific binding. One way to elucidate the entities the probe binds is to treat cells with the probe then use differential centrifugation to separate cell components. The fluorescence of each component can be measured to show where the probe localizes. In addition, RNA can be isolated from cells using a method similar to the DNA isolation described to quantify probe binding therein.

Page | 152

7.3 Considerations for in vivo imaging

A probe that is highly selective for AP sites may still not give optimal results in live animal imaging. This is because the probe can only be as specific as the DNA damaging chemotherapy, which often damages healthy and cancerous cells indiscriminately. The effectiveness of the probe is dependent on the effectiveness of the drug. The type of imaging performed must be considered and proper controls employed.

If AP site probes are being used to evaluate new drug candidates in xenograft models, a negative control with no drug treatment is crucial. A positive control known to contain AP sites would control for experimental errors including dosage and imaging parameters. A positive control might be a previously established cell line and drug combination. Until this control is established, however, another method should be used.

This could be direct injection of a methylating drug, such as methyl methanesulfonate

(MMS) into the tumor. MMS methylates DNA without activation and direct injection may avoid biodistribution and toxicity.

7.3.1 AuNPs for tumor targeting and distribution

Despite the poor specificity of DNA damaging drugs and the presence of AP sites throughout the body, selective delivery of the imaging probes can be achieved using

AuNPs.162 Preliminary studies indicate that the AuNPs accumulate in tumor tissues and successfully deliver a control compound. However, the nanoparticles appear to drop the

AP site probe payload quickly. The complexation reaction could be optimized to ensure the probe is conjugated deep within the PEG chains. Likewise, the structure of the arms could be modified to increase residence time of the probe.

Page | 153

7.3.2 DNA extraction from xenograft model

Accumulation of the AP site imaging probe in the tumor does not necessarily indicate that the probe is binding to AP sites, especially if a delivery tool such as AuNPs is employed. The relative fluorescence of a chemotherapy treated and an untreated control could demonstrate that the probe is binding to AP sites. However, this method may give false negatives if the basal AP site level is much higher than what is induced by treatment; the quantity of AP sites following treatment is too low for detection; or the cell rapidly repairs the probe-AP site lesions. The latter may be examined using the method described in §7.2.1 (page 151). The specific binding to AP DNA could be determined post mortem by extracting the DNA and measuring fluorescence in the purified sample.

The fluorescence in the non-DNA cell components should also be measured.

7.3.3 In vivo imaging and biodistribution

SAR studies are not limited to in vitro experiments. Despite thorough research and excellent advanced planning, many compounds do not behave as predicted when used in vivo (hence the huge failure rate for drugs in clinical trials). If a compound is observed to have unfavorable pharmacokinetics in vivo, SARs should be used to improve the imaging properties. Modification should derive from the behavior of the probe. For instance, extensive accumulation in the liver may indicate the compound is readily metabolized. SAR to decrease derivatization by the cytochrome P450 enzymes would be appropriate. Conversely, if the probe does not extravasate from the blood vessels, SAR to minimize serum protein binding would benefit the imaging. Biodistribution of the fluorescence can provide an estimate of the fate of the probe.

Page | 154

7.4 Conclusions

7.4.1 Myelin imaging

CIC is a fluorescent myelin probe that can be used in optical imaging. While it absorbs and emits in the UV-visible range, some of the limitations of imaging in this range are circumvented by the use of two-photon fluorescence microscopy, which uses

NIR incident light to excite the fluorophore. The UV-visible emission allows the probe to be an effective histological agent for both in vitro and ex vivo tissue staining.

The resolution of fluorescent microscopy is sufficient to visualize subcellular structures. CIC can be employed as a stain for microscopy to visualize myelination on a cellular level. This could potentially provide insights into the causes and treatments of demyelinating and dysmyelinating diseases. The in vivo imaging application of CIC suggests the probe can penetrate cell membranes for use in live cell imaging.

Optical imaging of CIC is not intended for clinical use in human patients as most hospitals do not routinely use UV-visible optical imaging for diagnosis due to limitations of light penetration and scattering. Conversely, [11C] CIC is intended for clinical use in patients, particularly in MS and other demyelinating diseases, to improve diagnosis, to monitor response to therapy, and to evaluate disease progression. The work presented in chapter 2 represents the first demonstration that CIC can be used to image spinal cord myelin. Following the optical imaging studies shown herein, the study of [11C] CIC as a

PET tracer in spinal cords is ongoing.

One preclinical application for CIC imaging is the evaluation of new drug candidates in animal models. Several models of demyelinating disease are available.

Often, the extent of disease progression is measured using a scale based on macroscopic

Page | 155

symptoms. While this method is well established, subjective evaluation and unintentional bias cannot be dissociated from the measure. CIC imaging would allow an additional parameter from the molecular level to be considered in disease scoring. This could demonstrate if the macroscopic and molecular phenotypes are consistent as well.

7.4.2 Cancer imaging

Several novel probes for AP site detection have been developed, with an emphasis on characterizing the cyanine-based probes. UV and visible emitting probes have potential application as histological agents. However, studies presented in §6.1.2 (page

116) suggest the probes are incompatible with formalin fixation, which is typically used in a clinical setting. Thus, UV-visible probes would have limited application in fixed post mortem tissue samples. The UV-visible probes would, however, be applicable to analysis of tissue culture samples or frozen patient samples. They could be used to evaluate the mechanisms of DNA damaging drug toxicity or resistance by quantifying AP sites or pre-AP site lesions.

Red and NIR cyanine-based AP site probes have the same potential as histological agents as the UV-visible probes with the added benefit of being useful agents for in vivo optical imaging in animals. In this case, human tumor xenografts can be treated with a chemotherapeutic drug and the efficacy of the drug can be evaluated based on the extent of BER detected. The preliminary results from chapter 6 suggest lead compounds

Cy7MX and Cy5MX will require derivitization to improve their suitability for in vivo imaging.

As fluorescence is quantitative, an AP site quantification kit could be developed based on these probes. Several kits based on ARP have been commercialized but they

Page | 156

suffer from low reproducibility, multiple manipulations that may introduce error, and

time sensitivity for analysis. Conversely, AP site analysis by Cy7MX has less error (see

§4.5, page 82), fewer steps and manipulations, and a broader window for analysis than

the ARP kit (Figure 47).

Purified AP DNA

Adjust [DNA] to 100 μg/mL Incubate with ARP*** Incubate with Cy7MX

Purify DNA**** Purify DNA

Add binding solution* Adjust [DNA] (optional) Measure fluorescence within 2 days

Wash x 5

Add HRP-streptavidin*

Wash x 5

Add substrate solution*

Measure absorbance within 1 h * Transfer or dilution after [DNA] adjustment

Figure 47. Comparison of ARP-based AP site quantification kit (Dojindo) and proposed Cy7MX-based kit. The Cy7MX offers a simple, direct route with less potential for error by manipulation of DNA concentration.

While the AP site probes are fluorescent and were envisioned as contrast agents

for optical imaging and microscopy, they are not limited to this application. The probes

could be modified to accommodate a positron emitting radiolabel. Thus, their efficacy of

AP site binding could be established using a “cold” probe and in vivo detection with

potential clinical translation could be performed using a “hot” probe. Ex vivo fluorescent

imaging could confirm the PET results.

The use of these probes as bimodal imaging agents with MRI is less

straightforward. MRI requires large quantities of contrast agent for detection. The

Page | 157

number of AP sites a cell can support is likely to be too low for sufficient accumulation of MRI contrast probe. Further, lanthanide chelates are too large and hydrophilic to penetrate intact cell membranes making them unlikely to reach the nucleus. This obstacle for bimodal MRI-optical AP site imaging is large but not insurmountable. Gliomas are detected clinically with a gadolium-based contrast agent that takes advantage of the BBB breakdown in the tumor region. TMZ, a methylating chemotherapeutic, is used clinically to treat gliomas and is well known to be repaired by BER. Thus, a contrast agent containing a MRI contrast agent with a labile tether to an AP site probe could be used to locate gliomas and report on their response to drug.

AP site directed probes covalently bind to the intermediate in BER. This mechanism halts the repair process as APE, the next repair enzyme, does not recognize the probe-AP site lesion as a substrate. Thus, the AP site probes have the potential to be both diagnostic and therapeutic agents. This idea coined the term “theranostics.” This area of clinical application is gaining popularity and these probes have the capability to contribute to the field.

AP sites are indicative of DNA damage and repair. The focus of this work has centered on the application of AP site probes in cancer. However, BER is a major repair pathway in response to damage caused by reactive oxygen species (ROS) with the glycosylases recognizing several forms of oxidative damage (Table 1, page 14). DNA damage by ROS and subsequent BER enzyme repair are implicated in rheumatoid arthritis,136 Alzheimer’s disease,137 Parkinson’s disease,138 and aging.139 X-ray radiation has been shown to produce AP sites as well.140 Therefore, while experiments primarily

Page | 158

focused on cancer, the investigated probes and methods have broader application to many diseases.

REFERENCES

(1) Cassidy, P. J.; Radda, G. K., Molecular imaging perspectives. J. R. Soc. Interface 2005, 2 (3), 133-144. (2) Brown, R. W.; Cheng, Y.-C. N.; Haacke, E. M.; Thompson, M. R.; Venkatesan, R., Magnetic resonance imaging: Physical principles and sequence design. 2 ed.; John Wiley & Sons: Hoboken, New Jersey, 2014. (3) The chemistry of contrast agents in medical magnetic resonance imaging. 2 ed.; Merbach, A.; Helm, L.; Tóth, É., Eds. John Wiley & Sons: Chichester, West Sussex, UK, 2013. (4) Artemov, D., Molecular magnetic resonance imaging with targeted contrast agents. J. Cell. Biochem. 2003, 90 (3), 518–524. (5) Massoud, T. F.; Gambhir, S. S., Molecular imaging in living subjects: Seeing fundamental biological processes in a new light. Gene Dev. 2003, 17 (5), 545- 580. (6) Chien, C.-C.; Wang, H.-Y.; Wang, J.-J.; Kan, W.-C.; Chien, T.-W.; Lin, C.-Y.; Su, S.-B., Risk of acute kidney injury after exposure to gadolinium-based contrast in patients with renal impairment. Ren. Fail. 2011, 33 (8), 758-764. (7) Rohrer, M.; Bauer, H.; Mintorovitch, J.; Requardt, M.; Weinmann, H.-J., Comparison of magnetic properties of mri contrast media solutions at different magnetic field strengths. Invest. Radiol. 2005, 40 (11), 715-724. (8) Ollinger, J. M.; Fessler, J. A., Positron-emission tomography. IEEE Sig. Process Mag. 1997, 14 (1), 43-55. (9) Som, P.; Atkins, H. L.; Bandoypadhyay, D.; Fowler, J. S.; MacGregor, R. R.; Matsui, K.; Oster, Z. H.; Sacker, D. F.; Shiue, C. Y.; Turner, H.; Wan, C.-N.; Wolf, A. P.; Zabinski, S. V., A fluorinated glucose analog, 2-fluoro-2-deoxy-d- glucose (f-18): Nontoxic tracer for rapid tumor detection. J. Nucl. Med. 1980, 21 (7), 670-675. (10) Strijckmans, K., The isochronous cyclotron: Principles and recent developments. Comput. Med. Imaging Graph. 2001, 25 (2), 69-78. (11) Contag, P. R.; Olomu, I. N.; Stevenson, D. K.; Contag, C. H., Bioluminescent indicators in living mammals. Nat. Med. 1998, 4 (2), 245 - 247. (12) Bhaumik, S.; Gambhir, S. S., Optical imaging of renilla luciferase reporter gene expression in living mice. Proc. Natl. Acad. Sci. U.S.A. 2001, 99 (1), 377-382. (13) Lakowicz, J. R., Principles of fluorescence spectroscopy. 3 ed.; Springer: New York, 2006; p 954. (14) Stokes, G. G., On the change of refrangibility of light. Phil. Trans. R. Soc. Lond. 1852, 142, 463-562.

Page | 159

(15) Montet, X.; Ntziachristos, V.; Grimm, J.; Weissleder, R., Tomographic fluorescence mapping of tumor targets. Cancer Res. 2005, 65 (14), 6330-6336. (16) Denk, W.; Svoboda, K., Photon upmanship: Why multiphoton imaging is more than a gimmick. Neuron 1997, 18 (3), 351–357. (17) Koyama, Y.; Barrett, T.; Hama, Y.; Ravizzini, G.; Choyke, P. L.; Kobayashi, H., In vivo molecular imaging to diagnose and subtype tumors through receptor- targeted optically labeled monoclonal antibodies. Neoplasia 2007, 9 (12), 1021– 1029. (18) Weissleder, R.; Ntziachristos, V., Shedding light onto live molecular targets. Nat. Med. 2003, 9 (1), 123 - 128. (19) Frangioni, J. V.; Nakayama, A.; Lim, Y. T.; Kim, S.; Stott, N. E.; Bawendi, M. G., Selection of quantum dot wavelengths for biomedical assays and imaging. Mol. Imaging 2003, 2 (1), 50-64. (20) Frangioni, J. V., In vivo near-infrared fluorescence imaging. Curr. Opin. Chem. Biol. 2003, 7 (5), 626-634. (21) Aoki, T.; Yasuda, D.; Shimizu, Y.; Odaira, M.; Niiya, T.; Kusano, T.; Mitamura, K.; Hayashi, K.; Murai, N.; Koizumi, T.; Kato, H.; Enami, Y.; Miwa, M.; Kusano, M., Image-guided liver mapping using fluorescence navigation system with indocyanine green for anatomical hepatic resection. World J. Surg. 2008, 32 (8), 1763-1767. (22) Cutter, J. L.; Cohen, N. T.; Wang, J.; Sloan, A. E.; Cohen, A. R.; Panneerselvam, A.; Schluchter, M.; Blum, G.; Bogyo, M.; Basilion, J. P., Topical application of activity-based probes for visualization of brain tumor tissue. PLoS One 2012, 7 (3), e33060. (23) Licha, K., Contrast agents for optical imaging. In Contrast agents ii: Optical, ultrasound, x-ray and radiopharmaceutical imaging, Krause, W., Ed. Springer: New York, 2002; pp 1-29. (24) Yang, M.; Baranov, E.; Moossa, A. R.; Penman, S.; Hoffman, R. M., Visualizing gene expression by whole-body fluorescence imaging. Proc. Natl. Acad. Sci. U.S.A. 2000, 97 (22), 12278-12282. (25) Frullano, L.; Wang, C.; Miller, R. H.; Wang, Y., A myelin-specific contrast agent for magnetic resonance imaging of myelination. J. Am. Chem. Soc. 2011, 133, 1611–1613. (26) Wang, Y.; Wu, C.; Caprariello, A. V.; Somoza, E.; Zhu, W.; Wang, C.; Miller, R. H., In vivo quantification of myelin changes in the vertebrate nervous system. J. Neurosci. 2009, 29 (46), 14663–14669. (27) Sherman, D. L.; Brophy, P. J., Mechanisms of axon ensheathment and myelin growth. Nat. Rev. Neurosci. 2005, 6 (9), 683-690. (28) Aicardi, J., The inherited leukodystrophies: A clinical overview. J. Inher. Metab. Dis. 1993, 16 (4), 733 - 743. (29) Hatten, H. P., Jr., Dysmyelinating leukodystrophies: “Lack proper myelin”. Pediatr. Radiol. 1991, 21 (7), 477-482. (30) Boccaccio, G. L.; Steinman, L., Multiple sclerosis: From a myelin point of view. J. Neurosci. Res. 1998, 45 (6), 647–654.

Page | 160

(31) Krishnan, C.; Kaplin, A. I.; Pardo, C. A.; Kerr, D. A.; Keswani, S. C., Demyelinating disorders: Update on transverse myelitis. Curr. Neurol. Neurosci. Rep. 2006, 6 (3), 236-243. (32) Lucchinetti, C. F.; Mandler, R. N.; McGavern, D.; Bruck, W.; Gleich, G.; Ransohoff, R. M.; Trebst, C.; Weinshenker, B.; Wingerchuk, D.; Parisi, J. E.; Lassmann, H., A role for humoral mechanisms in the pathogenesis of devic’s neuromyelitis optica. Brain 2002, 125 (7), 1450 –1461. (33) Hughes, R. A. C., Guillain-barré syndrome. Lancet 2005, 366 (9497), 1653–1666. (34) Compston, A.; Coles, A., Multiple sclerosis. Lancet 2002, 359 (9313), 1221– 1231. (35) Waxman, S. G., Demyelination in spinal cord injury. J. Neurol. Sci. 1989, 91 (1- 2), 1–14. (36) Walther, P. J.; Rossitch, E., Jr.; Bullard, D. E., The development of lhermitte's sign during cisplatin chemotherapy: Possible drug-induced toxicity causing spinal cord demyelination. Cancer 1987, 60 (9), 2170–2172. (37) Heinzlef, O.; Lotz, J.-P.; Roullet, E., Severe neuropathy after high dose carboplatin in three patients receiving multidrug chemotherapy. J. Neurol. Neurosurg. Psychiatry 1998, 64, 667-669. (38) Kiernan, J. A., Histochemistry of staining methods for normal and degenerating myelin in the central and peripheral nervous systems. J. Histotechnol. 2007, 30 (2), 87-106. (39) Scholtz, C. L., Quantitative histochemistry of myelin using luxol fast blue mbs Histochem. J. 1977, 9 (6), 759-765. (40) Klüver, H.; Barrera, E., A method for the combined staining of cells and fibers in the nervous system. J. Neuropathol. Exp. Neurol. 1953, 12 (4), 400-403. (41) Schmued, L.; Slikker, W., Jr., Black-gold: A simple, high-resolution histochemical label for normal and pathological myelin in brain tissue sections. Brain Res. 1999, 837 (1-2), 289-297. (42) Savaskan, N. E.; Weinmann, O.; Heimrich, B.; Eyupoglu, I. Y., High resolution neurochemical gold staining method for myelin in peripheral and central nervous system at the light- and electron-microscopic level. Cell Tissue Res. 2009, 337 (2), 213-221. (43) Schmued, L.; Bowyer, J.; Cozart, M.; Heard, D.; Binienda, Z.; Paul, M., Introducing black-gold ii, a highly soluble gold phosphate complex with several unique advantages for the histochemical localization of myelin. Brain Res. 2008, 1229, 210-217. (44) Stilwell, D. L., A sudan black b myelin stain for peripheral nerves. Stain Technol. 1957, 32 (1), 19-23. (45) Kilgore, J. J. Lipophilic dyes and their application for detection of myelin US 20060073541, Apr 6, 2006. (46) Monsma, P. C.; Brown, A., Fluoromyelin (tm) red is a bright, photostable and non-toxic fluorescent stain for live imaging of myelin. J. Neurosci. Meth. 2012, 209 (2), 344-350. (47) Fu, Y.; Wang, H.; Huff, T. B.; Shi, R.; Cheng, J.-X., Coherent anti-stokes raman scattering imaging of myelin degradation reveals a calcium-dependent pathway in lyso-ptdcho-induced demyelination. J. Neurosci. Res. 2007, 85 (13), 2870–2881.

Page | 161

(48) McDonald, W. I.; Compston, A.; Edan, G.; Goodkin, D.; Hartung, H.-P.; Lublin, F. D.; McFarland, H. F.; Paty, D. W.; Polman, C. H.; Reingold, S. C.; Sandberg- Wollheim, M.; Sibley, W.; Thompson, A.; Van Den Noort, S.; Weinshenker, B. Y.; Wolinsky, J. S., Recommended diagnostic criteria for multiple sclerosis: Guidelines from the international panel on the diagnosis of multiple sclerosis. Ann. Neurol. 2001, 50 (1), 121–127. (49) Polman, C. H.; Reingold, S. C.; Edan, G.; Filippi, M.; Hartung, H.-P.; Kappos, L.; Lublin, F. D.; Metz, L. M.; McFarland, H. F.; O'Connor, P. W.; Sandberg- Wollheim, M.; Thompson, A. J.; Weinshenker, B. G.; Wolinsky, J. S., Diagnostic criteria for multiple sclerosis: 2005 revisions to the “mcdonald criteria”. Ann. Neurol. 2005, 58 (6), 840–846. (50) Barkovich, A. J., Magnetic resonance techniques in the assessment of myelin and myelination. J. Inherit. Metab. Dis. 2005, 28 (3), 311 - 343. (51) Frullano, L.; Zhu, J.; Wang, C.; Wu, C.; Miller, R. H.; Wang, Y., Myelin imaging compound (mic) enhanced magnetic resonance imaging of myelination. J. Med. Chem. 2012, 55 (1), 94–105. (52) Frullano, L.; Zhu, J.; Miller, R. H.; Wang, Y., Synthesis and characterization of a novel gadolinium-based contrast agent for magnetic resonance imaging of myelination. J. Med. Chem. 2013, 56 (4), 1629–1640. (53) Wu, C.; Wang, C.; Popescu, D. C.; Zhu, W.; Somoza, E. A.; Zhu, J.; Condie, A. G.; Flask, C. A.; Miller, R. H.; Macklin, W. B.; Wang, Y., A novel pet marker for in vivo quantification of myelination. Bioorg. Med. Chem. 2010, 18 (24), 8592- 8599. (54) Stankoff, B.; Wang, Y.; Bottlaender, M.; Aigrot, M.-S.; Dolle, F.; Wu, C.; Feinstein, D.; Huang, G.-F.; Semah, F.; Mathis, C. A.; Klunk, W.; Gould, R. M.; Lubetzki, C.; Zalc, B., Imaging of cns myelin by positron-emission tomography. Proc. Natl. Acad. Sci. U.S.A. 2006, 103 (24), 9304-9309. (55) Wu, C.; Wei, J.; Tian, D.; Feng, Y.; Miller, R. H.; Wang, Y., Molecular probes for imaging myelinated white matter in cns. J. Med. Chem. 2008, 51 (21), 6682– 6688. (56) Wu, C.; Tian, D.; Feng, Y.; Polak, P.; Wei, J.; Sharp, A.; Stankoff, B.; Lubetzki, C.; Zalc, B.; Mufson, E. J.; Gould, R. M.; Feinstein, D. L.; Wang, Y., A novel fluorescent probe that is brain permeable and selectively binds to myelin. J. Histochem. Cytochem. 2006, 54 (9), 997-1004. (57) Wang, C.; Wu, C.; Zhu, J.; Miller, R. H.; Wang, Y., Design, synthesis, and evaluation of coumarin-based molecular probes for imaging of myelination. J. Med. Chem. 2011, 54 (7), 2331–2340. (58) Wang, C.; Popescu, D. C.; Wu, C.; Zhu, J.; Macklin, W. B.; Wang, Y., In situ fluorescence imaging of myelination. J. Histochem. Cytochem. 2010, 58 (7), 611- 621. (59) Wang, C.; Wu, C.; Popescu, D. C.; Zhu, J.; Macklin, W. B.; Miller, R. H.; Wang, Y., Longitudinal near-infrared imaging of myelination. J. Neurosci. 2011, 31 (7), 2382-2390. (60) Siegel, R.; Ma, J.; Zou, Z.; Jemal, A., Cancer statistics, 2014. CA Cancer J. Clin. 2014, 64 (1), 9–29.

Page | 162

(61) Hanahan, D.; Weinberg, R. A., Hallmarks of cancer: The next generation. Cell 2011, 144 (5), 646–674. (62) Hanna, N. H.; Einhorn, L. H., Testicular cancer — discoveries and updates. N. Engl. J. Med. 2014, 371, 2005-2016. (63) Giaccone, G., Clinical perspectives on platinum resistance. Drugs 2000, 59 (4 Supplement), 9-17. (64) Woods, D.; Turchi, J. J., Chemotherapy induced DNA damage response. Cancer Biol. Ther. 2013, 14 (5), 379-389. (65) Sancar, A., DNA repair in humans. Annu. Rev. Genetics 1995, 29 (1), 69-105. (66) Schärer, O. D., Chemistry and biology of DNA repair. Angew. Chem. Int. Ed. 2003, 42 (26), 2946–2974. (67) Hoeijmakers, J. H. J., Genome maintenance mechanisms for preventing cancer. Nature 2001, 411 (6835), 366-374. (68) Krokan, H. E.; Bjørås, M., Base excision repair. Cold Spring Harb. Perspect. Biol. 2013, 5, a012583. (69) Bunting, S. F.; Nussenzweig, A., End-joining, translocations and cancer. Nat. Rev. Cancer 2013, 13, 443–454. (70) Li, X.; Heyer, W.-D., Homologous recombination in DNA repair and DNA damage tolerance. Cell Res. 2008, 18, 99–113. (71) Friedberg, E. C., How nucleotide excision repair protects against cancer. Nat. Rev. Cancer 2001, 1, 22-33. (72) Kartalou, M.; Essigmann, J. M., Mechanisms of resistance to cisplatin. Mutat. Res. 2001, 478 (1-2), 23-43. (73) Marra, G.; Schär, P., Recognition of DNA alterations by the mismatch repair system. Biochem. J. 1999, 338 (1), 1–13. (74) Wyatt, M. D.; Pittman, D. L., Methylating agents and DNA repair responses:ௗ Methylated bases and sources of strand breaks. Chem. Res. Toxicol. 2006, 19 (12), 1580–1594. (75) Trivedi, R. N.; Almeida, K. H.; Fornsaglio, J. L.; Schamus, S.; Sobol, R. W., The role of base excision repair in the sensitivity and resistance to temozolomide- mediated cell death. Cancer Res. 2005, 65 (14), 6394-6400. (76) Matuo, R.; Sousa, F. G.; Escargueil, A. E.; Soares, D. G.; Grivicich, I.; Saffi, J.; Larsen, A. K.; Henriques, J. A. P., DNA repair pathways involved in repair of lesions induced by 5-fluorouracil and its active metabolite fdump. Biochem. Pharmacol. 2010, 79 (2), 147–153. (77) Krokan, H. E.; Drabløs, F.; Slupphaug, G., Uracil in DNA - occurrence, consequences and repair. Oncogene 2002, 21 (58), 8935-8948. (78) Chetsanga, C. J.; Lindahl, T., Release of 7-methylguanine residues whose imidazole rings have been opened from damaged DNA by a DNA glycosylase from escherichia coli. Nucleic Acids Res. 1979, 6 (11), 3673–3684. (79) Tchou, J.; Kasai, H.; Shibutani, S.; Chung, M. H.; Laval, J.; Grollman, A. P.; Nishimura, S., 8-oxoguanine (8-hydroxyguanine) DNA glycosylase and its substrate specificity. Proc. Natl. Acad. Sci. U.S.A. 1991, 88 (11), 4690–4694. (80) Samson, L.; Derfler, B.; Boosalis, M.; Call, K., Cloning and characterization of a 3-methyladenine DNA glycosylase cdna from human cells whose gene maps to chromosome 16. Proc. Natl. Acad. Sci. U.S.A. 1991, 88 (20), 9127–9131.

Page | 163

(81) O'Connor, T. R.; Laval, J., Human cdna expressing a functional DNA glycosylase excising 3-methyladenine and 7-methylguanine. Biochem. Biophys. Res. Commun. 1991, 176 (3), 1170-1177. (82) Bandaru, V.; Sunkara, S.; Wallace, S. S.; Bond, J. P., A novel human DNA glycosylase that removes oxidative DNA damage and is homologous to escherichia coli endonuclease viii. DNA Repair 2002, 1 (7), 517–529. (83) Hazra, T. K.; Izumi, T.; Boldogh, I.; Imhoff, B.; Kow, Y. W.; Jaruga, P.; Dizdaroglu, M.; Mitra, S., Identification and characterization of a human DNA glycosylase for repair of modified bases in oxidatively damaged DNA. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (6), 3523–3528. (84) Ikeda, S.; Biswas, T.; Roy, R.; Izumi, T.; Boldogh, I.; Kurosky, A.; Sarker, A. H.; Seki, S.; Mitra, S., Purification and characterization of human nth1, a homolog of escherichia coli endonuclease iii: Direct identification of lys-212 as the active nucleophilic residue. J. Biol. Chem. 1998, 273 (34), 21585-21593. (85) Hu, J.; de Souza-Pinto, N. C.; Haraguchi, K.; Hogue, B. A.; Jaruga, P.; Greenberg, M. M.; Dizdaroglu, M.; Bohr, V. A., Repair of formamidopyrimidines in DNA involves different glycosylases: Role of the ogg1, nth1, and neil1 enzymes. J. Biol. Chem. 2005, 280 (49), 40544-40551. (86) Rosenquist, T. A.; Zharkov, D. O.; Grollman, A. P., Cloning and characterization of a mammalian 8-oxoguanine DNAௗglycosylase. Proc. Natl. Acad. Sci. U.S.A. 1997, 94 (14), 7429-7434. (87) Masaoka, A.; Matsubara, M.; Hasegawa, R.; Tanaka, T.; Kurisu, S.; Terato, H.; Ohyama, Y.; Karino, N.; Matsuda, A.; Ide, H., Mammalian 5-formyluracilíDNA glycosylase. 2. Role of smug1 uracilíDNA glycosylase in repair of 5-formyluracil and other oxidized and deaminated base lesions. Biochemistry 2003, 42 (17), 5003–5012. (88) Lindahl, T., An n-glycosidase from escherichia coli that releases free uracil from DNA containing deaminated cytosine residues. Proc. Natl. Acad. Sci. U.S.A. 1974, 71 (9), 3649-3653. (89) Mauro, D. J.; De Riel, J. K.; Tallarida, R. J.; Sirover, M. A., Mechanisms of excision of 5-fluorouracil by uracil DNA glycosylase in normal human cells. Mol. Pharmacol. 1993, 43 (6), 854-857. (90) Ingraham, H. A.; Tseng, B. Y.; Goulian, M., Mechanism for exclusion of 5- fluorouracil from DNA. Cancer Res. 1980, 40 (4), 998-1001. (91) Caradonna, S. J.; Cheng, Y.-C., The role of deoxyuridine triphosphate nucleotidohydrolase, uracil-DNA glycosylase, and DNA polymerase Į in the metabolism of fudr in human tumor cells. Mol. Pharmacol. 1980, 18 (3), 513-520. (92) Lee, S. H.; Wang, S.; Kool, E. T., Templated chemistry for monitoring damage and repair directly in duplex DNA. Chem. Commun. 2012, 48 (65), 8069-8071. (93) Wu, W.; Wang, Y.; Zhou, Y.; Shao, Y.; Zhang, L.; Liu, H., Selective fluorescence lighting-up recognition of DNA abasic site environment possessing guanine context. Sensor. Actuator. B 2015, 206, 449–455. (94) An, N.; Fleming, A. M.; White, H. S.; Burrows, C. J., Crown ether–electrolyte interactions permit nanopore detection of individual DNA abasic sites in single molecules. Proc. Natl. Acad. Sci. U.S.A. 2012, 109 (29), 11504–11509.

Page | 164

(95) Bowler, F. R.; Diaz-Mochon, J. J.; Swift, M. D.; Bradley, M., DNA analysis by dynamic chemistry. Angew. Chem. Int. Ed. 2010, 49 (40), 1809–1812. (96) Chowdhury, G.; Guengerich, F. P., Tandem mass spectrometry-based detection of c4ƍ-oxidized abasic sites at specific positions in DNA fragments. Chem. Res. Toxicol. 2009, 22 (7), 1310–1319. (97) Sun, H. B.; Qian, L.; Yokota, H., Detection of abasic sites on individual DNA molecules using atomic force microscopy. Anal. Chem. 2001, 73 (10), 2229– 2232. (98) Buzzeo, M. C.; Barton, J. K., Redmond red as a redox probe for the DNA- mediated detection of abasic sites. Bioconjugate Chem. 2008, 19 (11), 2110– 2112. (99) Jakobsen, U.; Shelke, S. A.; Vogel, S.; Sigurdsson, S. T., Site-directed spin- labeling of nucleic acids by click chemistry: Detection of abasic sites in duplex DNA by epr spectroscopy. J. Am. Chem. Soc. 2010, 132 (30), 10424–10428. (100) Lhomme, J.; Constant, J.-F.; Demeunynck, M., Abasic DNA structure, reactivity, and recognition. Biopolymers 1999, 52 (2), 65–83. (101) Greco, N. J.; Tor, Y., Simple fluorescent pyrimidine analogues detect the presence of DNA abasic sites. J. Am. Chem. Soc. 2005, 127 (31), 10784–10785. (102) Tanpure, A. A.; Srivatsan, S. G., A microenvironment-sensitive fluorescent pyrimidine ribonucleoside analogue: Synthesis, enzymatic incorporation, and fluorescence detection of a DNA abasic site. Chem.—Eur. J. 2011, 17 (45), 12820–12827. (103) Zeglis, B. M.; Boland, J. A.; Barton, J. K., Targeting abasic sites and single base bulges in DNA with metalloinsertors. J. Am. Chem. Soc. 2008, 130 (24), 7530– 7531. (104) Abulfazl, F. M.; Rokita, S. E., A new solvatochromic fluorophore for exploring nonpolar environments created by biopolymers. Chem. Commun. 2011, 47 (14), 4222-4224. (105) Boturyn, D.; Boudali, A.; Constant, J.-F.; Defrancq, E.; Lhomme, J., Synthesis of fluorescent probes for the detection of abasic sites in DNA. Tetrahedron 1997, 53 (15), 5485-5492. (106) Wang, Y.; Liu, L.; Wu, C.; Bulgar, A.; Somoza, E.; Zhu, W.; Gerson, S. L., Direct detection and quantification of abasic sites for in vivo studies of DNA damage and repair. Nucl. Med. Biol. 2009, 36 (8), 975–983. (107) Kubo, K.; Ide, H.; Wallace, S. S.; Kow, Y. W., A novel sensitive and specific assay for abasic sites, the most commonly produced DNA lesion. Biochemistry 1992, 31 (14), 3703–3708. (108) Atamna, H.; Cheung, I.; Ames, B. N., A method for detecting abasic sites in living cells: Age-dependent changes in base excision repair. Proc. Natl. Acad. Sci. U.S.A. 2000, 97 (2), 686–691. (109) Kojima, N.; Takebayashi, T.; Mikami, A.; Ohtsuka, E.; Komatsu, Y., Construction of highly reactive probes for abasic site detection by introduction of an aromatic and a guanidine residue into an aminooxy group. J. Am. Chem. Soc. 2009, 131 (37), 13208–13209. (110) Taverna, P.; Liu, L.; Hwang, H.-S.; Hanson, A. J.; Kinsella, T. J.; Gerson, S. L., Methoxyamine potentiates DNA single strand breaks and double strand breaks

Page | 165

induced by temozolomide in colon cancer cells. Mutat. Res. 2001, 485 (4), 269- 281. (111) Liu, L.; Nakatsuru, Y.; Gerson, S. L., Base excision repair as a therapeutic target in colon cancer. Clin. Cancer Res. 2002, 8 (9), 2985–2991. (112) Permittivity (dielectric constants) of liquids. In CRC Handbook of Chemistry and Physics, 92 (Internet version) ed.; Haynes, W. M., Ed. CRC Press/Taylor and Francis: Boca Raton, FL, 2012. (113) Northrup, R. B., Introduction to dynamic modeling of neuro-sensory systems. CRC Press: Boca Raton, FL, 2001; p 469. (114) Condie, A. G.; Gerson, S. L.; Miller, R. H.; Wang, Y., Two-photon fluorescent imaging of myelination in the spinal cord. Chem. Med. Chem. 2012, 7 (12), 2194– 2203. (115) Xu, C.; Webb, W. W., Measurement of two-photon excitation cross sections of molecular fluorophores with data from 690 to 1050 nm. J. Opt. Soc. Am. B. 1996, 13 (3), 481–491. (116) Vaurs-Barriere, C.; Bonnet-Dupeyron, M.-N.; Combes, P.; Gauthier-Barichard, F.; Reveles, X. T.; Schiffmann, R.; Bertini, E.; Rodriguez, D.; Vago, P.; Armour, J. A. L.; Saugier-Veber, P.; Frebourg, T.; Leach, R. J.; Boespflug-Tanguy, O., Golli-mbp copy number analysis by fish, qmpsf and maph in 195 patients with hypomyelinating leukodystrophies. Ann. Hum. Genet. 2005, 70 (1), 66–77. (117) Chernoff, G. F., Shiverer: An autosomal recessive mutant mouse with myelin deficiency. J. Hered. 1981, 72 (2), 128. (118) Rosenbluth, J., Central myelin in the mouse mutant shiverer. J. Comp. Neurol. 1980, 194 (3), 639-648. (119) Popko, B.; Puckett, C.; Lai, E.; Shine, H. D.; Readhead, C.; Takahashi, N.; Hunt, S. W.; Sidman, R. L.; Hood, L., Myelin deficient mice: Expression of myelin basic protein and generation of mice with varying levels of myelin. Cell 1987, 48 (4), 713-721. (120) Roach, A.; Takahashi, N.; Pravtcheva, D.; Ruddle, F.; Hood, L., Chromosomal mapping of mouse myelin basic protein gene and structure and transcription of the partially deleted gene in shiverer mutant mice. Cell 1985, 42 (1), 149–155. (121) Hall, S. M.; Gregson, N. A., The in vivo and ultrastructural effects of injection of lysophosphatidyl choline into myelinated peripheral nerve fibres of the adult mouse. J. Cell Sci. 1971, 9 (3), 769-789. (122) Hall, S. M., The effect of injections of lysophosphatidyl choline into white matter of the adult mouse spinal cord. J. Cell Sci. 1972, 10 (2), 535-546. (123) Gregson, N. A.; Hall, S. M., A quantitative analysis of the effects of the intraneural injection of lysophosphatidyl choline. J. Cell Sci. 1973, 13 (1), 257- 277. (124) Compston, A.; Coles, A., Multiple sclerosis. Lancet 2008, 372 (9648), 1502– 1517. (125) Brunton, L.; Lazo, J.; Parker, K., Goodman & gilman's the pharmacological basis of therapeutics. 11 ed.; McGraw-Hill Professional: New York, 2005. (126) Denk, W.; Strickler, J. H.; Webb, W. W., Two-photon laser scanning fluorescence microscopy. Science 1990, 248 (4951), 73-76.

Page | 166

(127) So, P. T. C.; Dong, C. Y.; Masters, B. R.; Berland, K. M., Two-photon excitation fluorescence microscopy. Annu. Rev. Biomed. Eng. 2000, 2, 399-429. (128) Wu, C.; Zhu, J.; Baeslack, J.; Zaremba, A.; Hecker, J.; Kraso, J.; Matthews, P. M.; Miller, R. H.; Wang, Y., Longitudinal positron emission tomography imaging for monitoring myelin repair in the spinal cord. Ann. Neurol. 2013, 74 (5), 688- 698. (129) Reichardt, C.; Vogt, R. A.; Crespo-Hernández, C. E., On the origin of ultrafast nonradiative transitions in nitro-polycyclic aromatic hydrocarbons: Excited-state dynamics in 1-nitronaphthalene. J. Chem. Phys. 2009, 131, 224518. (130) Santos-Pérez, J.; Crespo-Hernández, C. E.; Reichardt, C.; Cabrera, C. R.; Feliciano-Ramos, I.; Arroyo-Ramírez, L.; Meador, M. A., Synthesis, optical characterization, and electrochemical properties of isomeric tetraphenylbenzodifurans containing electron acceptor groups. J. Phys. Chem. A 2011, 115 (17), 4157–4168. (131) Oulianov, D. A.; Tomov, I. V.; Dvornikov, A. S.; Rentzepis, P. M., Observations on the measurement of two-photon absorption cross-section. Opt. Commun. 2001, 191 (3-6), 235-243. (132) Salisbury, C. M.; Maly, D. J.; Ellman, J. A., Peptide microarrays for the determination of protease substrate specificity. J. Am. Chem. Soc. 2002, 124 (50), 14868–14870. (133) Gerson, S. L.; Liu, L.; Xu, Y. Use of fluorescent aldehyde reactive probe in method for measurement of DNA abasic sites and for screening for base excision repair-inhibiting drugs. WO 2008066990, 2008. (134) Rurack, K.; Spieles, M., Fluorescence quantum yields of a series of red and near- infrared dyes emitting at 600í1000 nm. Anal. Chem. 2011, 83 (4), 1232–1242. (135) Licha, K.; Riefke, B.; Ntziachristos, V.; Becker, A.; Chance, B.; Semmler, W., Hydrophilic cyanine dyes as contrast agents for near-infrared tumor imaging: Synthesis, photophysical properties and spectroscopic in vivo characterization. Photochem. Photobiol. 2000, 72 (3), 392–398. (136) Lo, S.-F.; Wan, L.; Huang, C.-M.; Lin, H.-C.; Chen, S.-Y.; Liu, S.-C.; Tsai, F.-J., Genetic polymorphisms of the DNA repair gene ung are associated with the susceptibility of rheumatoid arthritis. Rheumatol. Int. 2012, 32 (12), 3723-3727. (137) Iida, T.; Furuta, A.; Nishioka, K.; Nakabeppu, Y.; Iwaki, T., Expression of 8- oxoguanine DNA glycosylase is reduced and associated with neurofibrillary tangles in alzheimer's disease brain. Acta Neuropathol. 2002, 103 (1), 20-25. (138) Fukae, J.; Takanashi, M.; Kubo, S.-i.; Nishioka, K.-i.; Nakabeppu, Y.; Mori, H.; Mizuno, Y.; Hattori, N., Expression of 8-oxoguanine DNA glycosylase (ogg1) in parkinson’s disease and related neurodegenerative disorders. Acta Neuropathol. 2005, 109 (3), 256-262. (139) Cui, H.; Kong, Y.; Zhang, H., Oxidative stress, mitochondrial dysfunction, and aging. J. Signal Transduct. 2012, 2012, 646354. (140) Shiina, T.; Watanabe, R.; Shiraishi, I.; Suzuki, M.; Sugaya, Y.; Fujii, K.; Yokoya, A., Induction of DNA damage, including abasic sites, in plasmid DNA by carbon ion and x-ray irradiation. Radiat. Environ. Biophys. 2013, 52, 99-112. (141) Solodar, W.; Green, M., 4'-nitro-2-stibenol. J. Org. Chem. 1958, 23 (1), 103-104.

Page | 167

(142) Yee, S. W.; Shah, B.; Simons, C., Synthesis and antimycobacterial activity of 7-o- substituted-4-methyl-2h-2-chromenone derivatives vs mycobacterium tuberculosis. J. Enzym. Inhib. Med. Chem. 2005, 20 (2), 109-113. (143) Wey, S.-J.; Augustyniak, M. E.; Cochran, E. D.; Ellis, J. L.; Fang, X.; Garvey, D. S.; Janero, D. R.; Letts, L. G.; Martino, A. M.; Melim, T. L.; Murty, M. G.; Richardson, S. K.; Schroeder, J. D.; Selig, W. M.; Trocha, A. M.; Wexler, R. S.; Young, D. V.; Zemtseva, I. S.; Zifcak, B. M., Structure-based design, synthesis, and biological evaluation of indomethacin derivatives as cyclooxygenase-2 inhibiting nitric oxide donors. J. Med. Chem. 2007, 50 (25), 6367-6382. (144) Nakane, M.; Ichikawa, S.; Matsuda, A., Triazole-linked dumbbell oligodeoxynucleotides with nf-țb binding ability as potential decoy molecules. J. Org. Chem. 2008, 73 (5), 1842-1851. (145) Nanjunda, R.; Owens, E. A.; Mickelson, L.; Alyabyev, S.; Kilpatrick, N.; Wang, S.; Henary, M.; Wilson, W. D., Halogenated pentamethine cyanine dyes exhibiting high fidelity for g-quadruplex DNA. Bioorg. Med. Chem. 2012, 20 (24), 7002-7011. (146) Samanta, A.; Maiti, K. K.; Soh, K.-S.; Liao, X.; Vendrell, M.; Dinish, U. S.; Yun, S.-W.; Bhuvaneswari, R.; Kim, H.; Rautela, S.; Chung, J.; Olivo, M.; Chang, Y.- T., Ultrasensitive near-infrared raman reporters for sers-based in vivo cancer detection. Angew. Chem. Int. Ed. 2011, 50 (27), 6089–6092. (147) Brouwer, A. M., Standards for photoluminescence quantum yield measurements in solution (iupac technical report)*. Pure Appl. Chem. 2011, 83 (12), 2213-2228. (148) Physical constants of organic compounds. In CRC Handbook of Chemistry and Physics, 94 (Internet version) ed.; Haynes, W. M., Ed. CRC Press/Taylor and Francis: Boca Raton, FL, 2014. (149) Rosa, S.; Fortini, P.; Karran, P.; Bignami, M.; Dogliotti, E., Processing in vitro of an abasic site reacted with methoxyamine: A new assay for the detection of abasic sites formed in vivo. Nucleic Acids Res. 1991, 19 (20), 5569-5574. (150) Lindahl, T.; Nyberg, B., Rate of depurination of native deoxyribonucleic acid. Biochemistry 1972, 11 (19), 3610–3618. (151) Langenbach, R. J.; Danenberg, P. V.; Heidelberger, C., Thymidylate synthetase: Mechanism of inhibition by 5-fluoro-2ƍ-deoxyuridylate. Biochem. Biophys. Res. Commun. 1972, 48 (6), 1565-1571. (152) Santi, D. V.; McHenry, C. S.; Sommer, H., Mechanism of interaction of thymidylate synthetase with 5-fluorodeoxyuridylate. Biochemistry 1974, 13 (3), 471–481. (153) Meyers, M.; Hwang, A.; Wagner, M. W.; Bruening, A. J.; Veigl, M. L.; Sedwick, W. D.; Boothman, D. A., A role for DNA mismatch repair in sensing and responding to fluoropyrimidine damage. Oncogene 2003, 22 (47), 7376–7388. (154) Major, P. P.; Egan, E.; Herrick, D.; Kufe, D. W., 5-fluorouracil incorporation in DNA of human breast carcinoma cells. Cancer Res. 1982, 42 (8), 3005-3009. (155) Gupta, P. K.; Sirover, M. A., Cell cycle regulation of DNA repair in normal and repair deficient human cells. Chem.-Biol. Interact. 1981, 36 (1), 19-31. (156) Bianchi, V.; Zantedeschi, A.; Ronchese, F.; Levis, A. G., Induction of DNA repair synthesis by ultraviolet radiation and methylmethanesulphonate in cultured mouse lymphocytes. Cancer Lett. 1983, 18 (1), 21-27.

Page | 168

(157) Jamur, M. C.; Oliver, C., Cell fixatives for immunostaining. In Immunocytochemical methods and protocols, 3 ed.; Oliver, C.; Jamur, M. C., Eds. Humana Press: New York, 2010; Vol. 588, pp 55-62. (158) Suthipintawong, C.; Leong, A. S.-Y.; Vinyuvat, S., Immunostaining of cell preparations: A comparative evaluation of common fixatives and protocols. Diagn. Cytopathol. 1996 15 (2), 167–174. (159) Darkes, M. J. M.; Plosker, G. L.; Jarvis, B., Temozolomide. Am. J. Cancer 2002, 1 (1), 55-80. (160) Sarkaria, J. N.; Kitange, G. J.; James, C. D.; Plummer, R.; Calvert, H.; Weller, M.; Wick, W., Mechanisms of chemoresistance to alkylating agents in malignant glioma. Clin. Cancer Res. 2008 14 (10), 2900-2908. (161) Ogilvie, R. F., Histopathology. 6 ed.; Williams and WIlkins Co.: Baltimore, 1962; p 514. (162) Cheng, Y.; Meyers, J. D.; Broome, A.-M.; Kenney, M. E.; Basilion, J. P.; Burda, C., Deep penetration of a pdt drug into tumors by noncovalent drug-gold nanoparticle conjugates. J. Am. Chem. Soc. 2011, 133 (8), 2583–2591. (163) Wei, S.; Shalhout, S.; Ahn, Y.-H.; Bhagwat, A. S., A versatile new tool to quantify abasic sites in DNA and inhibit base excision repair. DNA Repair 2015, 27, 9–18. (164) Choi, H. S.; Nasr, K.; Alyabyev, S.; Feith, D.; Lee, J. H.; Kim, S. H.; Ashitate, Y.; Hyun, H.; Patonay, G.; Strekowski, L.; Henary, M.; Frangioni, J. V., Synthesis and in vivo fate of zwitterionic near-infrared fluorophores. Angew. Chem. Int. Ed. 2011, 50 (28), 6258–6263. (165) Strickley, R. G., Solubilizing excipients in oral and injectable formulations. Pharm. Res. 2004, 21 (2), 201-230. (166) Lawrence, M. J., Surfactant systems: Their use in drug delivery. Chem. Soc. Rev. 1994, 23 (6), 417-424. (167) Uekama, K.; Hirayama, F.; Irie, T., Cyclodextrin drug carrier systems. Chem. Rev. 1998, 98 (5), 2045–2076. (168) Savjani, K. T.; Gajjar, A. K.; Savjani, J. K., Drug solubility: Importance and enhancement techniques. ISRN Pharm. 2012, 2012, 195727. (169) Khadka, P.; Ro, J.; Kim, H.; Kim, I.; Kim, J. T.; Kim, H.; Cho, J. M.; Yun, G.; Lee, J., Pharmaceutical particle technologies: An approach to improve drug solubility, dissolution and bioavailability. Asian J. Pharm. Sci. 2014, 9 (6), 304- 316. (170) Kim, S. K.; Vaishali, B.; Lee, E.; Lee, S.; Lee, Y.-k.; Kumar, T. S.; Moon, H. T.; Byun, Y., Oral delivery of chemical conjugates of heparin and deoxycholic acid in aqueous formulation. Thromb. Res. 2006, 117 (4), 419-427. (171) Henary, M.; Mojzych, M., Stability and reactivity of polymethine dyes in solution. Top. Heterocycl. Chem. 2008, 14, 221-238. (172) Leo, A.; Hansch, C.; Elkins, D., Partition coefficients and their uses. Chem. Rev. 1971, 71 (6), 525–616. (173) Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J., Experimental and computational approaches to estimate solubility and permeability in drug discovery and development settings. Adv. Drug Deliver. Rev. 2001, 46 (1-3), 3– 26.

Page | 169