Functionalization and Synthesis Of

GAYATRI SHRIKHANDE

DIFUNCTIONAL FOLATE-TARGETED POLYMERIC CONJUGATES

FOR POTENTIAL DIAGNOSTIC APPLICATIONS

A Dissertation

Presented to

The Graduate Faculty of The University of Akron

In Partial Fulfillment

of the Requirements for the Degree

Doctor of Philosophy

Gayatri Shrikhande

December, 2019 FUNCTIONALIZATION AND SYNTHESIS OF

DIFUNCTIONAL FOLATE-TARGETED POLYMERIC CONJUGATES

FOR POTENTIAL DIAGNOSTIC APPLICATIONS.

Gayatri Shrikhande

Dissertation

Approved: Accepted:

______Advisor Department Chair Dr. Jie Zheng Dr. H Michael Cheung

______Committee Member Interim Dean of the College Dr. Bi-min Zhang Newby Dr. Craig Menzemer

______Committee Member Dean of the Graduate School Dr. Ge Zhang Dr. Chand Midha

______Committee Member Date Dr. Chrys Wesdemiotis

______Committee Member Dr. Mark Soucek

ii ABSTRACT

The aim of this research was to synthesize polymer-diagnostic agent conjugates with two folate functionalities for potential diagnostic applications. Conjugates with fluorescein (FL) as an imaging agent, poly(ethylene glycol) (PEG) as a hydrophilic linker and two folic acid (FA) as targeting agents were synthesized by chemo-enzymatic method using Candida antarctica lipase B (CALB) catalyst for the multivalent targeting of folate receptors (FRs) overexpressed on cancer cells. In this dissertation work, imaging agent FL was first acrylated using acryloyl chloride (AcrCl) in the presence of triethylamine (TEA) to precisely synthesize fluorescein o-acrylate (FL-A, yield: 52.49%) and fluorescein o,o’- diacrylate (FL-DA, yield: 63.9%). A kinetic study of FL-DA synthesis was conducted using a Nuclear Magnetic Resonance (NMR)-750 MHz spectrophotometer instrument which demonstrated the formation FL-DA in 13 seconds reaction time. Hence, our FL-DA synthesis was extremely fast and easy to purify using silica gel. Acrylate moieties of FL-

DA and FL-A allow the CALB-catalyzed Michael addition of thiol and amine to develop the conjugates.

First, FL-A with single acrylate moiety was reacted with PEG-diamine (H2N-PEG-

NH2) by the CALB-catalyzed Michael addition. However, the nucleophilic secondary amine of FL-NH-PEG-NH-FL interfered with the acrylation of ‘OH’ of FL-A. Hence, a new synthetic strategy was developed where H2N-PEG-NH2 was replaced by dithiol- functionalized PEG (HS-PEG-SH, Mn =899 g/mol, Đ=1.00, Mn =1160 g/mol, Đ=1.14 and

iii Mn =2200 g/mol, Đ=1.09) and tetraethylene glycol (HS-TEG-SH, FW= 370.48 g/mol) which were synthesized in Dr. Puskas’ lab to avoid the interference of the amine group in the acrylation reaction. Michael addition between FL-A and HS-TEG-SH, by CALB- catalysis was extremely fast and completed in 1 minute at 52℃. Reaction between FL-A and HS-PEG-SH without CALB catalysis did not go to completion even after 18 hours at

52℃ but completed in 2 minutes when CALB was added. CALB catalysis was found to be extremely useful to synthesize FL-TEG-FL and FL-PEG-FL compounds. Further acrylation of the ‘OH’ of FL-A, followed by the attachment of FA-SH by the Michael addition resulted in successful development of the difunctional FL-PEG conjugate with different molecular weights. PEGylation of FL using brominated PEG was also attempted by lithium chemistry to develop the FL-PEG-FL conjugate which can be further acrylated for the attachment of FA-SH. However, due to the presence of impurities during the

PEGylation, no further reactions were performed.

Next, tetrafunctional FL compounds were synthesized by the aza-Michael addition reaction between FL-DA and secondary amines by studying the effect of CALB-catalysis.

These multifunctional FL compounds can be used as precursors to develop polymeric conjugates. The aza-Michael addition reactions of small molecules, such as diethanolamine

(DEA) and diallylamine (DAA) to FL-DA were completed in 1 minute at room temperature without CALB-catalysis in dimethyl sulfoxide (DMSO), while the reactions were extremely slow in chloroform. CALB catalysis was found to be extremely useful with large sterically hindered molecules, such as diethyl iminodiacetate (DIDA) as the reaction time was three times less than the reaction without CALB catalysis. The aza-Michael addition product of DEA with FL-DA (tetrahydroxy-functionalized FL) was highly unstable, it

iv hydrolyzed back to FL due to neighboring group (OH) participation. Hence to avoid the hydrolysis, new dihydroxy-functionalized secondary amines with longer alkyl chains were synthesized by ultraviolet (UV)-mediated thiol-ene click reaction between DAA and 4- mercapto-1-butanol and with 9-mercapto-1-nonanol. CALB-catalyzed Michael addition of these newly synthesized dihydroxy secondary amines to FL-DA resulted in successful synthesis of tetrahydroxy functionalized FL.

Finally, a scale-up of FL-DA synthesis is discussed in this dissertation, where FL-

DA was successfully scaled up in a 5-liter capacity jacketed reactor with overhead stirrer and a pure product with 54.32% yield was obtained. The cost to synthesize 1 g of FL-DA in the lab was 41.29 USD.

The structures of all products were confirmed by Proton NMR (1H-NMR), Carbon-

13 NMR (13C-NMR) and Matrix-assisted Laser Desorption/Ionization Time of Flight

(MALDI-ToF MS).

v DEDICATION

This dissertation is dedicated to my family: My mother, Swati Shrikhande; my father, Sanjay Shrikhande; my brother, Gaurang Shrikhande and my love, my husband,

Abhishek Deshpande. This dissertation would not have been possible without their warm love, continued patience and endless support.

vi ACKNOWLEDGEMENTS

I am grateful to Dr. Judit Puskas for providing me this great research opportunity, valuable guidance and support during Ph.D. I am thankful to Dr. Jie Zheng for his continued help and suggestions throughout my Ph.D. I would also like to thank My Ph.D. committee Dr. Chrys Wesdemiotis, Dr. Ge (Christie) Zhang, Dr. Mark Soucek and Dr. Bi- min Zhang Newby for providing constructive feedback and motivation through the course of this research. I would like to express deep gratitude to Dr. Sanghamitra Sen for teaching me various chemistry skills and for her guidance. Her teaching helped me grow as a researcher. Financial support by the Breast Cancer Innovation Foundation, Akron is greatly appreciated. I am very delightful to work with my good friend Prajakatta Mulay with whom

I had a great collaboration in research. I also would like to thank Dr. Aditya Jindal, Andrew

McClain, Dr. Jozsef Kantor, Dr. Gabor Kaszas and Dr. Carin Helfer for their help.

I acknowledge Dr. Wesdemiotis’ research group at The University of Akron and

The Ohio State University Mass Spectrometry Facility for the mass spectrometry analysis of the samples. I would like to thank our collaborator Cleveland Clinic for the in vitro and in vivo studies. Dr. Venkat Dudipala’s guidance towards Nuclear Magnetic Resonance

(NMR) technique is valuable. I am thankful to Dr. Richard Elliot for his help with the

Molecular Orbital PACkage (MOPAC) simulation software during the scale-up experiment. Last but not the least I thank all past members of the Dr. Puskas’ group for their help and wonderful memories and all my friends, family for their love and support.

vii TABLE OF CONTENTS

Page

LIST OF TABLES ...... xiii

LIST OF FIGURES ...... xiv

CHAPTER

I. INTRODUCTION ...... 1

II. BACKGROUND ...... 5

2.1. Folate-targeted cancer diagnosis and treatment ...... 5

2.1.1. Cancer statistics ...... 5

2.1.2. Motivation for breast-cancer focus ...... 5

2.1.3. Folate-targeting ...... 6

2.1.3.1. FA-receptor mediated endocytosis (RME) ...... 10

2.1.3.2. Clinical trials ...... 12

2.1.3.3. Multivalent targeting ...... 14

2.2. Modular approach to FA-targeted diagnostic devices ...... 17

2.2.1. Cleveland Clinic comparative trials ...... 19

2.2.2. New synthetic strategies ...... 24

2.3. Components of the new strategies ...... 28

2.3.1. Fluorescein (FL)...... 28

2.3.2. Poly(ethylene glycol) (PEG) linker ...... 32

2.3.2.1. PEG-diamine (H2N-PEG-NH2) ...... 33

viii 2.3.2.2. PEG-dithiol (HS-PEG-SH) ...... 35

2.3.3. Thiol-functionalized folic acid (FA-γ-SH) ...... 37

2.4. Construction of the conjugates ...... 39

2.4.1. Acrylation of FL ...... 39

2.4.2. Michael addition ...... 40

2.4.2.1. “Traditional” Michael addition ...... 40

2.4.2.2. CALB- catalyzed Michael addition...... 43

2.5. Considerations for reaction scale-up ...... 49

2.5.1. Calculations of the enthalpy of reactions ...... 49

III. EXPERIMENTAL ...... 52

3.1. Materials ...... 52

3.2. Procedures ...... 54

3.2.1. Syntheses of fluorescein o-acrylate (FL-A) and fluorescein o,o’- diacrylate (FL-DA) ...... 54

3.2.1.1. Kinetic studies ...... 54

3.2.1.2. Synthesis of fluorescein o-acrylate (FL-A) ...... 55

3.2.1.3. Synthesis of fluorescein o,o’-diacrylate (FL-DA)...... 56

3.2.2. Synthesis of two-functional folate-targeted fluorescein (FL)-based diagnostic nanodevices ...... 57

3.2.2.1. Practice: Synthesis of the Cleveland Clinic trial compound FA-FL-FA ...... 57

3.2.2.2. Strategy 2 : 2 FL-A + H2N-PEG-NH2 + 2 AcrCl + 2 FA- SH ...... 57

3.2.2.3. Strategy 3: 2 FL-A + HS-(TEG or PEG)-SH + 2 AcrCl + 2 FA-SH ...... 59

3.2.2.4. Strategy 4: PEGylation of fluorescin (also known as dihydrofluorescein) via lithium technology ...... 65 ix 3.2.3. Synthesis of tetrafunctional fluoresceins (FL) ...... 67

3.2.3.1. Synthesis of tetraallyl-functionalized FL ...... 67

3.2.3.2. Synthesis of tetraacetate-functionalized FL ...... 69

3.2.3.3. Synthesis of tetrahydroxy-functionalized FL ...... 71

3.2.4. Scale-up of fluorescein o,o’-diacrylate (FL-DA) synthesis ...... 74

3.2.4.1. Experimental study to calculate the enthalpy of acrylation reaction ...... 74

3.2.4.2. Scale-up of FL-DA synthesis ...... 75

3.3. Characterization of products ...... 78

3.3.1. Nuclear Magnetic Resonance (NMR) spectroscopy ...... 78

3.3.2. Matrix Assisted Laser Desorption/Ionization Time-of-Flight Mass Spectrometry (MALDI-ToF MS) ...... 79

3.3.3. Electrospray Ionization Mass Spectrometry (ESI-MS) ...... 79

3.3.4. Thin Layer Chromatography (TLC) ...... 79

3.3.5. Column chromatography ...... 80

IV. RESULTS AND DISCUSSIONS ...... 81

4.1. Syntheses of fluorescein o-acrylate (FL-A) and fluorescein o,o’-diacrylate (FL-DA) ...... 81

4.1.1. Kinetic studies ...... 81

4.1.2. Synthesis of fluorescein o-acrylate (FL-A)...... 90

4.1.3. Synthesis of fluorescein o,o’-diacrylate (FL-DA) ...... 96

4.2. Synthesis of two-functional folate-targeted fluorescein (FL)-based diagnostic nanodevices ...... 104

4.2.1. Objectives ...... 104

4.2.2. Syntheses of the Cleveland Clinic trial compounds FA-FL-NH- PEG-NH-FL-FA and FA-FL-FA ...... 107

x 4.2.3. Strategy 2: 2 FL-A + H2N-PEG-NH2 + 2 AcrCl + 2 FA-SH ...... 110

4.2.3.1. Synthesis of FL-NH-PEG2000 -NH-FL ...... 111

4.2.3.2. Synthesis of Acryl-FL-NH-PEG2000 -NH-FL-Acryl ... 115

4.2.4. Strategy 3: 2 FL-A + HS-(TEG or PEG)-SH + 2 AcrCl + 2 FA-SH 117

4.2.4.1. Synthesis of FL-S-TEG-S-FL ...... 119

4.2.4.2. Synthesis of FL-S-PEG-S-FL...... 127

4.2.4.3. Synthesis of Acryl-FL-S-TEG-S-FL-Acryl ...... 140

4.2.4.4. Synthesis of Acryl-FL-S-PEG-S-FL-Acryl compounds 143

4.3. Strategy 4: PEGylation of fluorescin via lithium technology ...... 152

4.3.1. Synthesis of Br-PEG-OMe ...... 155

4.3.2. Lithiation followed by PEGylation of fluorescin...... 158

4.4. Synthesis of tetrafunctional fluoresceins (FL) ...... 169

4.4.1. Synthesis of tetraallyl-functionalized FL ...... 170

4.4.1.1. Reaction in DMSO, with CALB catalysis ...... 170

4.4.1.2. Reaction in DMSO, without CALB Catalysis ...... 173

4.4.1.3. Reaction in chloroform, with CALB catalysis ...... 179

4.4.1.4. Reaction in chloroform, without CALB catalysis ...... 180

4.4.2. Synthesis of tetraacetate-functionalized FL ...... 181

4.4.2.1. Reaction in DMSO, with CALB catalysis ...... 182

4.4.2.2. Reaction in DMSO, without CALB catalysis ...... 188

4.4.3. Synthesis of tetrahydroxy-functionalized FL...... 190

4.4.3.1. Reaction between DEA and FL-DA...... 190

xi 4.4.3.2. UV-mediated reaction between diallylamine (DAA) and 4-mercapto 1-butanol to synthesize di-butanol-based amine compound...... 200

4.4.3.3. CALB-catalyzed Michael addition reaction between FL- DA and di-nonanol-based amine compound...... 205

4.5. Scale-up of fluorescein o,o’-diacrylate (FL-DA) synthesis ...... 208

4.5.1. Calculations of the enthalpy of acrylation reaction ...... 208

4.5.1.1. Experimental study to measure the enthalpy of acrylation reaction...... 213

4.5.2. Scale-up of FL-DA synthesis ...... 217

4.5.3. Costing of the scale-up...... 225

V. CONCLUSION ...... 230

REFERENCES ...... 234

APPENDIX A: LIST OF ABBREVIATIONS ...... 244

xii LIST OF TABLES

Table Page

33 Table 2.1 % expression of FR- α on various cancer tumors...... 8

Table 2.2 Compounds and their heat of formations...... 50

Table 4.1 Compounds and their heat of formations...... 210

92 Table 4.2 Group contribution values to calculate ΔHf of FL and FL-DA...... 211

Table 4.3 Specific heat capacities of the reactants...... 214

Table 4.4 Specific heat capacities of the reactants...... 215

Table 4.5 Temperature profile during the coolant circulation through the jacket before the addition of AcrCl...... 219

Table 4.6 Temperature profile during the addition of AcrCl...... 220

Table 4.7 Costs of the chemicals...... 226

Table 4.8 Cost of the glassware...... 227

Table 4.9 Cost of the scale-up...... 228

xiii LIST OF FIGURES

Figure Page

Figure 1.1 Synthetic strategy for the the proposed difunctional folate-targeted FL-PEG conjugate...... 3

Figure 2.1 Vitamins that can be used to selectively target cancer cells.27 ...... 7

Figure 2.2 Structure of FA tautomers and in vivo reduced forms.37 ...... 9

Figure 2.3 Mechanism of receptor-mediated endocytosis of folate-targeted SMDC and

PDC.39 ...... 12

Figure 2.4 Structure of EC17.42 ...... 13

Figure 2.5 (Left) Adverse effects of free drug on mice (Loss of weight, hair). (Right) No side effects on mice dosed with dendrimer transported folate-targeted drug.50 ...... 15

Figure 2.6 Comparison of binding of monovalent and divalent FA-PEG-FL conjugates to

FRs...... 16

Figure 2.7 Modular representation of the folate-targeted FL-PEG conjugate...... 17

Figure 2.8 Synthesis of FA-FL-NH-PEG-NH-FL-FA...... 18

Figure 2.9 . FA-FL-FA...... 19

Figure 2.10 FL-FA...... 19

Figure 2.11 Comparison of MD-MB-231 and MD-MB-468 TNBC cells for FR-density

(Unpublished data)...... 20

xiv Figure 2.12 (A) FL-FA uptake and cytotoxicity in MDA-MB-231 TNBC cells. 2 hours incubation followed by propidium iodide staining and process in BD LSR II flow cytometry. Data analyzed in Flow Jo 10 analytical software, N= 4. (B) Confocal microscope image (Unpublished data)...... 21

Figure 2.13 (A) FL-FA-FL uptake and cytotoxicity in MDA-MB-231 TNBC cells. 2 hours incubation followed by propidium iodide staining and process in BD LSR II flow cytometry. Data analyzed in Flow Jo 10 analytical software, N= 4; (B) Confocal microscopy image (Unpublished data)...... 22

Figure 2.14 .(A) FA-FL-PEG2000-FL-FA uptake and cytotoxicity in MDA-MB-231 cells.

2 hours incubation followed by propidium iodide staining and process in BD LSR II flow cytometry. Data analyzed in Flow Jo 10 analytical software, N= 4. (B) Confocal microscopy image (Unpublished data)...... 23

Figure 2.15 Intra-arterial injection of the PEG-based compound in liver cancer rat model.17 ...... 24

Figure 2.16 Synthetic strategies to be investigated...... 27

Figure 2.17 Structure of FL...... 28

Figure 2.18 Tautomeric forms of FL in dynamic equilibrium.60 ...... 29

Figure 2.19 Structures of FL at various pH values.65 ...... 31

Figure 2.20 Structure of the PEG...... 32

Figure 2.21 PEG-diamine...... 34

Figure 2.22 PEG-diamine synthesized by CALB-catalysis...... 35

Figure 2.23 Mechanism of aza-Michael addition between diethylamine and ethyl acrylate in THF.84 ...... 42

xv Figure 2.24 Reaction mechanism of CALB-catalyzed Michael addition between pyrrolidine and acrylonitrile.55 ...... 44

Figure 3.1 Experimental setup to calculate the enthalpy of the acrylation reaction ...... 75

1 Figure 4.1 H-NMR of AcrCl in CD2Cl2...... 82

Figure 4.2 1H-NMR of FL in DMSO d6...... 84

Figure 4.3 1H-NMR of FL in DMSO d6...... 85

1 Figure 4.4 (A) H-NMR of FL-TEA (1: 2.20) in CD2Cl2 (B) Inset is the enlarged resonance of proton (13) of FL...... 87

1 Figure 4.5 (A). H-NMR of product FL-DA in CD2Cl2 (B) Inset is the enlarged resonances of the product...... 89

Figure 4.6 TLC of the FL-A product before column chromatography with 80 v% ether,

20 v% hexane solvent system. Spots from left to right: FL, FL-A (Sigma-Aldrich), FL-

DA (synthesized in the lab), FL-A product (after precipitation and drying)...... 91

Figure 4.7 TLC with 80% ether, 20% hexane solvent system Spots from left to right: FL,

FL-A (Sigma-Aldrich), FL-DA (synthesized in the lab), FL-A product (column fraction).

...... 92

Figure 4.8 (A) 1H-NMR of pure FL-A in DMSO d6 (B) Enlarged resonances of the product (C) Enlarged resonances of 1H-NMR of FL-A from Sigma-Aldrich...... 94

Figure 4.9 13C-NMR of FL-A product after column purification in DMSO d6...... 95

Figure 4.10 1H-NMR of TEA hydrochloride salt in DMSO d6...... 97

Figure 4.11 1H NMR of FL-DA (A) Before silica gel filtration, (B) After silica gel filtration (C) Inset is the enlarged spectrum of 5.8 ppm to 8.5 ppm...... 99

13 Figure 4.12 C NMR of FL-DA in CDCl3...... 100

xvi Figure 4.13 (A) ESI-Mass spectrum of FL-DA. (B) HRMS ESI-Mass spectrum of FL-

DA...... 102

Figure 4.14 Synthetic strategies for the two-functional FA-targeted FL-based diagnostic nanodevices...... 107

Figure 4.15 1H NMR spectrum of compound FA-FL-FA in DMSO d6...... 109

1 Figure 4.16 H-NMR of H2N-PEG-NH2 (Sigma-Aldrich, Mn=2,000 g/mol, Đ ≤1.2) in

CDCl3 ...... 111

1 Figure 4.17 H-NMR of FL-NH-PEG2000-NH-FL in DMSO d6 ...... 113

13 Figure 4.18 C-NMR of FL-NH-PEG2000-NH-FL in DMSO-d6 ...... 114

Figure 4.19 1H-NMR of the product of acrylation of FL-NH-PEG-NH-FL in DMSO d6

...... 116

Figure 4.20 Possible products of the acrylation of FL-NH-PEG-NH-FL ...... 117

1 Figure 4.21 H-NMR of HS-TEG-SH in CDCl3 ...... 119

1 1 Figure 4.22 A. H-NMR of FL-S-TEG-S-FL in CDCl3 (1-minute raw sample), B. H-

NMR of FL-S-TEG-S-FL in CDCl3 after workup...... 122

Figure 4.23 13C NMR of FL-S-TEG-S-FL in DMSO-d6 after workup...... 123

Figure 4.24 (1) MALDI- ToF analysis of FL-S-TEG-S-FL (2) Enlarged mass spectrum of the side-product FL-S-TEG-SH...... 126

1 Figure 4.25 H-NMR of HS-TEG-SH in CDCl3...... 128

Figure 4.26 Kinetic study of Michael addition between HS-PEG1000-SH and FL-A in

DMSO d6...... 130

1 Figure 4.27 H-NMR of FL-S-PEG1000-S-FL in DMSO d6 after workup...... 132

13 Figure 4.28 CNMR of FL-S-PEG1000-S-FL in DMSO d6 after workup...... 133 xvii 13 Figure 4.29 CNMR of HS-PEG2000-SH in CDCl3...... 135

1 Figure 4.30 H-NMR of FL-S-PEG2000-S-FL in CDCl3 after workup...... 136

13 Figure 4.31 C-NMR of FL-S-PEG2000-S-FL in DMSO d6 after workup...... 137

Figure 4.32 Reaction mechanism of CALB-catalyzed Michael addition of HS-TEG-SH with FL-A.55, 86 ...... 139

Figure 4.33 1H-NMR of Acryl-FL-S-TEG-S-FL-Acryl in DMSO d6...... 141

Figure 4.34 13C-NMR of Acryl-FL-S-TEG-S-FL-Acryl in DMSO d6...... 142

1 Figure 4.35 H-NMR of Acryl-FL-S-dPEG899-S-FL-Acryl in DMSO d6 ...... 144

1 Figure 4.36 H-NMR of Acryl-FL-S-PEG1000-S-FL-Acryl in DMSO d6...... 146

13 Figure 4.37 C-NMR of Acryl-FL-S-PEG1000-S-FL-Acryl in DMSO d6...... 147

1 Figure 4.38 H-NMR of Acryl-FL-S-PEG2000-S-FL-Acryl in DMSO d6...... 149

13 Figure 4.39 C-NMR of Acryl-FL-S-PEG2000-S-FL-Acryl in DMSO d6 ...... 150

Figure 4.40 Structure of fluorescin...... 152

Figure 4.41 (A) 1H-NMR of fluorescin in DMSO d6 (B) Inset is the enlarged resonances.

...... 153

Figure 4.42 1H-NMR of FL in DMSO d6...... 154

1 Figure 4.43 . H-NMR of PEG-OMe in CDCl3 ...... 156

1 Figure 4.44 H-NMR of EBV in CDCl3 ...... 157

1 Figure 4.45 H-NMR of Br-PEG-OMe in CDCl3 ...... 158

Figure 4.46 Fluorescin solution turning red during the dropwise addition of n-BuLi. ... 159

Figure 4.47 Reaction mixture after the complete addition of Br-PEG-OMe compound to the lithiated fluorescin...... 159

xviii Figure 4.48 1H-NMR of 15-minute sample of PEGylation reaction in DMSO d6 ...... 161

Figure 4.49 1H-NMR of 2-hour sample of PEGylation reaction in DMSO d6 ...... 162

Figure 4.50 (1) MALDI-ToF mass spectrometry of the product (2) Enlarged distribution

A (Lithiated PEG-OMe) (3) Enlarged distribution B (PEGylated fluorescin). (4) Enlarged distribution C (Lithiated Br-PEG-OMe) ...... 167

1 Figure 4.51 H NMR of the DAA in DMSO d6 ...... 170

Figure 4.52 1H-NMR of the tetraallyl-functionalized FL in DMSO d6 (Reaction with

CALB catalysis) ...... 172

Figure 4.53 1H-NMR of tetraallyl-functionalized FL in DMSO d6 (Reaction without

CALB catalysis) ...... 174

Figure 4.54 1H-NMR of the pure tetraallyl-functionalized FL...... 175

13 Figure 4.55 . C NMR of the pure tetraallyl-functionalized FL in CDCl3...... 176

Figure 4.56 ESI-Mass spectrum of the tetraallyl-functionalized FL ...... 178

Figure 4.57 1H NMR of the 10-minute sample of Michael addition between DAA and

FL-DA in CDCl3 with CALB (Reaction with CALB catalysis) ...... 179

Figure 4.58 1H NMR of the 10-minute sample of Michael addition between DAA and

FL-DA in CDCl3 (Reaction without CALB catalysis) ...... 181

Figure 4.59 1H-NMR of the DIDA in DMSO d6...... 182

Figure 4.60 1H NMR of the 2 hours 45 minute and 9 hour samples of the Michael addition between DIDA and FL-DA (using CALB)...... 184

Figure 4.61 1H-NMR of the tetraacetate-functionalized FL ...... 185

Figure 4.62 13C NMR of the tetraacetate-functionalized FL ...... 186

Figure 4.63 ESI-Mass spectrum of the tetraacetate-functionalized FL ...... 187 xix Figure 4.64 1H NMR of the reaction samples of Michael addition between DIDA and FL-

DA with CALB catalysis in DMSO d6...... 189

Figure 4.65 1H NMR of the DEA in DMSO d6 ...... 191

Figure 4.66 1H NMR of the tetrahydroxy-functionalized FL in DMSO d6 (Reaction with

CALB catalysis) ...... 192

Figure 4.67 1H NMR of the tetrahydroxy-functionalized FL in DMSO d6 (Reaction without CALB catalysis)...... 194

Figure 4.68 ESI-Mass spectrum of the product of Michael addition between FL-DA and

DEA after workup and drying...... 195

Figure 4.69 Proposed mechanism for the reaction between FL-DA and DEA...... 197

Figure 4.70 1H NMR of product of Michael addition between FL-DA and DEA after work up and drying...... 198

1 Figure 4.71 H NMR spectrum of Michael addition of DEA to FL-DA in CDCl3 at room temperature (Reaction with CALB catalysis) ...... 200

Figure 4.72 1H-NMR of the 4- mercapto 1-butanol ...... 201

Figure 4.73 1H-NMR of the di-butanol-based amine compound in DMSO d6 ...... 202

Figure 4.74 13C-NMR of the di-butanol-based amine compound in DMSO d6 ...... 203

Figure 4.75 ESI - Mass spectrometry of di-butanol-based amine compound...... 204

Figure 4.76 1H-NMR of the tetrahydroxy-functionalized FL (2) ...... 206

Figure 4.77 5L capacity jacketed glass reactor for the scale-up...... 208

Figure 4.78 Experimental setup to calculate the enthalpy of the acrylation reaction. .... 213

Figure 4.79 Time (s) vs temperature (℃) (inside the reactor) plot during the scale-up. 221

xx Figure 4.80 Reaction mixture during the dropwise addition of AcrCl...... 222

Figure 4.81 Reaction mixture after the complete addition of AcrCl (No stirring)...... 222

Figure 4.82 1H-NMR spectrum of FL-DA before purification...... 223

Figure 4.83 Brown colored product solution...... 224

Figure 4.84 TEA hydrochloride salt inside the reactor...... 224

Figure 4.85 TEA hydrochloride salt precipitated out after the addition of THF...... 224

Figure 4.86 1H-NMR spectrum of the purified FL-DA...... 225

xxi CHAPTER I

INTRODUCTION

This dissertation will discuss the synthesis of difunctional poly(ethylene glycol)

(PEG)-based folate-targeted nanodevices for potential cancer diagnosis following in vitro screening by the Cleveland Clinic. Although cancer mortality rates have declined during the last 25 years, still it is the second leading cause of death in the US with highest impact on public health (deaths due to heart disease are number one).1 With the motivation to fight against cancer, our group works on a drug delivery platform using polymer science as a primary tool.

For effective cancer diagnosis and therapy, it is necessary to selectively target cancer cells and avoid the collateral damage to healthy cells. Folic Acid (FA) or vitamin

B9 is a good choice to target malignancies, as vitamin receptors such as folate receptors

(FRs) are overexpressed on cancer cells and FA binds efficiently to FRs. FR-α, an isoform of FR is overexpressed on malignant cells of ovarian, testicular, non-small cell lung, breast, endometrial, epithelial, renal, and colorectal cancer, and nonfunctional pituitary adenomas.2,3 Hence FR-α has been an important and widely studied biomarker for cancer diagnosis and therapy for over 30 years.4-10

The concept of cellular internalization of FR-targeted biomacromolecules was first coined by Leamon and Low in1991.4 The ability of cancer cells to internalize folate- conjugated proteins such as Bovine Serum Albumin (BSA) and Immunoglobulin G (IgG)

1 by the mechanism known as Receptor Mediated Endocytosis (RME) was confirmed in vitro in KB (human nasopharyngeal cell line) cancer cells. Further, the efficiency of targeting more than one FR with a single polymer-based therapeutic/diagnostic device was established by the scientists from the University of Michigan in 2007.11 FR targeting with

~2.6 FA on a poly(amidoamine) (PAMAM) dendrimer increased the binding avidity by

2500 times over free folic acid (FA).11 Binding avidity was maximum at about 5 FA groups per dendrimer. However, the improvement in binding avidity from 1 FA to 2 FA was much higher than from 2 to 5 FA.11 In this thesis we are focused on 2 FA-based conjugates for targeted cancer diagnosis.

Our group synthesized conjugates with fluorescein (FL) as an imaging agent, FA as a targeting agent and PEG as a linker. We chose FL as a diagnostic agent because it is inexpensive, exhibits fluorescence in the visible range,12 and is being widely used as an imaging agent for in vitro studies due to its high quantum efficiency.12-14 We have developed a method for thiol-functionalization of FA (FA-SH) exclusively at the γ- carboxylic acid position15 which can be used as a targeting agent. The incorporation of

PEG is important as it is a non-toxic polymer which increases water solubility and circulation time of the conjugates inside the body.16

The Cleveland Clinic had conducted in vitro study of our three compounds: FA-

FL-NH-PEG-NH-FL-FA, difunctional FA-FL-FA and monofunctional FL-FA in triple negative breast cancer (TNBC) cell lines. FA-FL-NH-PEG-NH-FL-FA was found to be the best with excellent solubility, cellular uptake and no cytotoxicity.17 Based on the excellent preliminary in vitro results of the PEG-based targeted device with two FA groups

2 on the same molecule, we designed a more effective synthetic strategy to develop difunctional folate-targeted PEG-based conjugates as shown in Figure 1.1.

Figure 1.1 Synthetic strategy for the the proposed difunctional folate-targeted FL-PEG conjugate.

We have specifically synthesized FA-FL-dPEG1000-FL-FA where dPEG is a

monodisperse polymer, FA-FL-PEG2000-FL-FA and FA-FL-PEG1000-FL-FA by chemo- enzymatic processes which were developed by our group. The fluorescein o,o’-diacrylate

(FL-DA) and PEG-diamine (H2N-PEG-NH2) linker in FA-FL-NH-PEG-NH-FL-FA are replaced by fluorescein o-acrylate (FL-A) and PEG-dithiol (HS-PEG-SH) linker, respectively, in the newly synthesized compounds. We have been working on enzyme catalyzed polymer functionalization using Candida antarctica lipase B (CALB).18-22 In this

3 project, CALB-catalyzed Michael addition was conducted to attach HS-PEG-SH to 2 equivalents of FL-A. Next, Acryl-FL-PEG-FL-Acryl was synthesized. FA-SH developed by our group efficiently undergoes CALB-catalyzed Michael addition with the acrylate groups15of Acryl-FL-PEG-FL-Acryl made by a chemo-enzymatic process (Figure 1.1).

These compounds will be screened by the Cleveland Clinic.

This dissertation is focused on the functionalization of FL which is a precursor of the conjugates. First, successful acrylation of FL using acryloyl chloride is discussed. FL-A and

FL-DA were precisely synthesized. Further, I have scaled up the acrylation process to a ~100 g scale for the synthesis of FL-DA. Subsequently I have studied and successfully conducted

CALB-catalyzed Michael addition for the attachment of HS-PEG-SH to FL. The effect of

CALB catalysis on the Michael addition of small molecules as well as functionalized PEGs was evaluated. Multifunctional FL compounds were also synthesized by Michael addition.

In this dissertation, I will discuss my participation in making the compounds required to synthesize the three PEG-based conjugates listed above by chemo-enzymatic processes, working closely together with my classmate Prajakatta Mulay.

4 CHAPTER II

BACKGROUND

2.1. Folate-targeted cancer diagnosis and treatment

2.1.1. Cancer statistics

Cancer is a global health challenge in which uncontrolled cell division takes place due to cellular abnormality23 and it is one of the primary reasons of mortalities in the US.

In 2019, approximately 1,762,450 cancer cases will be diagnosed, and total 606,880 cancer mortalities are estimated.24 In men, cancer deaths are mainly due to lung, prostate, and colorectal cancer while in women, deaths due to lung, breast, and colorectal cancer are the greatest.25 Therefore it is important to seek new treatments for cancers.

2.1.2. Motivation for breast-cancer focus

In the USA, breast cancer is the second leading cause of cancer death among women and one in eight women develops breast cancer.26 30% of newly diagnosed cancer cases in women are breast cancer25 and a new patient is diagnosed every two minutes.26

Early detection of breast cancer is crucial for early treatment which increases the chance of survival of the cancer patient. For efficient breast cancer diagnosis and treatment, it is necessary to selectively target cancer cells which saves healthy cells from the toxic side effects of the drugs.27 This research has the potential to contribute to the fight against this global challenge with our novel folate-targeted drug delivery agents synthesized by

5 chemoenzymatic methods. Based on molecular subtypes, breast cancer is grouped into four types: luminal A, luminal B, Human Epidermal Growth Factor Receptor 2 (HER2)- enriched and triple-negative breast cancer (TNBC).28 Luminal A type has estrogen (ER) and progesterone (PR) receptors overexpressed on the cancer cells (ER and/or PR positive) and no overexpression of HER2.28 Also, it has a low level of the Ki-67 protein (<14%) which is associated with cell proliferation rate. Luminal B breast cancer is ER and/or PR positive and it is either positive or negative for HER2, but due to high Ki-67 level, it is a faster growing cancer than luminal A.28 In the case of HER2-enriched cancers (10-15% of breast cancers), there is an amplified overexpression of HER2 while ER and PR receptors are absent.28 Triple-negative/basal-like breast cancer (TNBC, 15% of breast cancers) 29,30 does not have hormone receptors or overexpressed HER2, but 86% of TNBC cells have overexpressed vitamin receptors such as folate receptors (FRs).29,30 Therefore the development of folate-targeted diagnostic and therapeutic devices seemed to be important, especially for TNBC.

2.1.3. Folate-targeting

Vitamins are required for the cell division, cell growth and hence for human survival.27 Due to cellular abnormalities, cancer cells need certain vitamins, such as vitamin

B12 (Cobalamin), vitamin B2 (Riboflavin), vitamin B9 (Folic Acid, FA), and vitamin B7

(Biotin) for the growth of the tumor via rapid cell division (Figure 2.1). Hence, receptors that bind these vitamins are overexpressed on cancer cells.27

Figure 2.1 Vitamins that can be used to selectively target cancer cells.27

Vitamin B9 or FA has widely been researched to target the overexpressed FRs on cancer cells. FR has four major isoforms: FR- α, FR-β, FR-γ and FR-δ, containing 220–

237 amino acids.31 FR- α and FR-β are membrane bound while FR-γ, and FR-δ do not have membrane support and are present in a minute amount in human tissues.31 FR- α is a 38 kDa protein,32 primarily expressed on the membranes of ovarian, lung, breast, endometrial, renal, and colorectal cancer cells. In this project, we are targeting FR-α to develop delivery agents with FA as a targeting ligand. Table 2.1 lists the % expression of FR-α on various cancer tumors.33 FRs are present on the cell surface in the form of clusters.34 There is a variation in the FR expression and density among different tumor masses of the same patient and among the same tumor type in different patients.3 O’Shannessy et al. reported the expression of FR- α in breast cancer cells by immunohistochemical evaluation.35

Table 2.1 % expression of FR- α on various cancer tumors.33

Cancer New cases per year % FR FR- expressing

(Solid Tumors) (USA) cancers/year

Breast 176300 48 84624

Lung 171600 78 133848

Uterus 37400 90 33660

Colorectal 129400 32 41408

Ovarian 25200 90 22680

Kidney 30000 75 22500

Head/neck 39750 52 20670

Brain and CNS 16800 90 15120

Gastric 28700 38 10146

Pancreatic 28600 13 3718

Endocrine 19800 14 2772

Testicular 7400 17 1258

Total 708950 392404

FR-α overexpression was reported in 30% of stage I-stage III invasive ductal breast cancers, while healthy breast tissues showed very restricted expression of FR- α. In case of stage IV metastatic TNBC cells, 70-80% of the cells were shown to overexpress the FR- α receptors.35

FA present in dietary supplements has two tautomeric forms (Structures I and II,

Figure 2.2), while the reduced forms such as 7,8-dihydro FA (DHF), 5,6,7,8-tetrahydro FA

(THF) and 5‐methyltetrahydrofolate (5‐MTHF) (Structures III, IV, V, Figure 2.2) are naturally present in food.36,37 When FA is taken as a supplement, first it reduces to DHF, followed by further reduction to THF which is then enzymatically converts into 5-MTHF, the main active form of FA in the body.37

Figure 2.2 Structure of FA tautomers and in vivo reduced forms.37

Transport of the FA and the folate inside body is governed by two mechanisms.

One mode of transport is via Reduced Folate Carriers (RFCs). RFCs have high affinity for reduced folates (Structures III, IV,V, Figure 2.2.) but much lower affinity for the oxidized forms (Structures I and II, Figure 2.2).38 The second mechanism takes place via FRs

(receptor-mediated endocytosis RME).38 FR can bind to 5-MTHF (Structure V, Figure 2.2),

FA (Structure I, II, Figure 2.2.) and to FR-targeted conjugates with high affinity. FR has

10 times higher affinity to FA than the RFCs,38 thus FA-targeting relies mainly on receptor- mediated endocytosis via the isoform FR-α.

2.1.3.1. FA-receptor mediated endocytosis (RME)

The concept of non-destructive receptor-mediated endocytosis of folate-conjugated biomacromolecules was introduced by Leamon and Low in 1991.4 They conjugated folic acid to biomacromolecules such as bovine pancreatic ribonuclease A (RNase A), horseradish peroxidase (HRP), bovine serum albumin (BSA) and bovine immunoglobulin

4 G (IgG) via -NH2 groups using the activated ester method yielding amide covalent bonds.

The reaction was stochastic, so an average of 1-10 FA was conjugated to the biomacromolecules.4 When folate-conjugated IgG labeled with fluorescein isothiocyanate

(FITC) biomarker was incubated with KB cells (human nasopharyngeal cancer cell line), binding with FRs took place within 5 minutes and internalization via endocytosis was gradual for hours. No binding was observed for IgG without folate targeting agent.4 This confirmed that covalently conjugated folate is required for the FR-mediated endocytosis pathway. Folate-BSA labeled with 125I tightly bound to KB cells in 2 hours and the uptake was similar to that of free FA, with a maximum of 4 x 106 molecules of 125I -BSA-folate uptake from every mole of 6 x 1023 molecules. Autoradiography confirmed that there was no degradation of BSA into peptides or fragments during the endocytosis.4

Later, further studies used the pathway of receptor-mediated endocytosis for drug or diagnostic agent delivery by targeting FR-α overexpressed on cancer cells using Small

Molecule-Drug or Diagnostic agent Conjugates (SMDC) or Polymeric Drug Conjugates

(PDC).3-10 It was shown that after the internalization of an SMDC or PDC, it enters the vesicle and early endosomes, followed by entering the Compartment of Uncoupling of

Receptor and Ligand (CURL).39 CURL has acidic pH which leads to the hydrolysis and dissociates acid- sensitive bonds (e.g. ester bonds) in the conjugate. Amide bonds are more stable, but they have frequently been used both in SMDC and PDC. Bond breaking results in diffusion of the polymer-drug or polymer-diagnostic conjugate into the cellular cytoplasm or the nucleus.39 The cytoplasm has a reducing environment, capable of breaking disulfide bonds so attaching the drug via disulfide linkage is advantageous. The liberated FR and FA enter the recycling endosome which transports them back to the cell membrane surface. In some cases, some of the SMDC or PDC pass from early endosome to late endosome instead of passing through the CURL and enter the lysosome. The lysosome also has acidic pH so it can also dissociate the acid-sensitive groups.39 The current understanding of the mechanism of receptor-mediated endocytosis is shown in

Figure 2.3. The recycling rate of FR from the cellular surface to the endosomes and again back to the cellular surface is reported to be 8–12 h.3

Figure 2.3 Mechanism of receptor-mediated endocytosis of folate-targeted SMDC and PDC.39

2.1.3.2. Clinical trials

Since Low’s discovery, thousands of studies were conducted with SMDCs and

PDCs. However, currently there is only one folate-targeted fluorescein (FL)-based imaging agent (EC17) in human clinical trials.40 EC17 is an SMDC with one folate-targeting ligand41 and sodium salt of FL with a peptide linker to increase water solubility42 as shown in Figure 2.4.

Figure 2.4 Structure of EC17.42

EC17 was tested to efficiently image lung adenocarcinoma cancer cells

(ClinicalTrials.gov Identifier Number, NCT01778920), Occult Ovarian Cancer

(NCT02000778), and renal nodules (NCT01778933). The intraoperative imaging potential of EC17 was also assessed for breast cancer (NCT01994369).43 At 500 nm wavelength, tumor-specific fluorescence was observed up to 5 hours after the administration of EC17.

However, autofluorescence of collagen in benign tissues was also detected which resulted in 23% false positive lesions in ovarian cancer. Similarly, autofluorescence interfered with tumor detection of breast cancer. Thus EC17 did not fulfill all the imaging requirements due to the autofluorescence and lack of enough penetration into the tissues.44 In case of imaging of renal nodules, autofluorescence from the endogenous tissues caused a major noise.45 A clinical trial was conducted for renal cell carcinoma (RCC) (NCT00485563) but was terminated due to changes in treatment paradigm.46 EC17 was also used for intraoperative primary hyperparathyroidism imaging to detect FR-α overexpressed glands

(NCT01996072).47 This study is completed. However, no results have been published.

2.1.3.3. Multivalent targeting

In addition to the autofluorescence problem, it is possible that the trials with EC17 were not successful because of the single FA targeting agent in the molecule. Kim et al. showed that when 10µM of free FA was injected along with a 99mTc-PEG-folate conjugate, competitive inhibition in uptake of the conjugate was observed, demonstrating that the uptake of FA-targeted conjugates can be competitively inhibited by free FA.48 Low’s original work had multivalent targeting.

Scientists from the University of Michigan synthesized generation 5 poly(amidoamine) (PAMAM) dendrimer which was covalently bonded to ~5 FA targeting ligands, ~5 Methotrexate (MTX) drug, and fluorescent dye (either FITC or 6 -

Carboxytetramethylrhodamine (6-TAMRA)) for imaging. Free amine groups of the

PAMAM dendrimer were acetylated to prevent toxicity and non-specific interactions.

Conjugate G5-(FITC)5-(FA)5-(MTX)5 was injected into mice bearing tumors with overexpressed FRs along with the free MTX drug to check the therapeutic efficacy.49

When mice were evaluated at various points of time, the FR-targeted dendrimer had significantly lower toxicity than free MTX. The FR-targeted dendrimer exhibited 10- fold higher efficacy when compared with the free drug at an equal cumulative dose.49

Tumor volume reduction by a factor of two was achieved with the targeted dendrimer at the equivalent dose (5.0 mg/kg), 1.1×10-5 mol/kg of free MTX. Mice had to be treated with free MTX drug of 21.7 mg/kg (4.77×10-5 mol/kg) to reach a similar reduction, but they lost all their hair, appeared sick and died within 30 days, while the mice treated with the drug conjugated to the dendrimer retained all their hair and did not lose weight (Figure 2.5).49

One mouse was completely cured with the folate-targeted therapeutic dendrimer at the 39th day of the trial.50 When targeted and non-targeted dendrimers were compared, it was observed that the non-targeted dendrimer was excreted from the body though the kidneys, while the targeted dendrimer accumulated in the tumor.50

Figure 2.5 (Left) Adverse effects of free drug on mice (Loss of weight, hair). (Right) No side effects on mice dosed with dendrimer transported folate-targeted drug.50

Image reprinted with permission from Clinical and Translational Science 2009,

The efficiency of multivalent targeting was explained as follows. The dissociation constant KD is defined as the ratio of koff, the dissociation rate constant and kon, the

51 association rate constant of the conjugate (KD= koff /kon). When the number of targeting ligands increases, the rate constant of dissociation (koff) decreases due to attachment of more than one targeting ligand to the clustered FR receptors on the cell surface. KD is a

51,52 measure of binding affinity of the ligand to the receptor. Smaller values of KD represent higher binding affinity between the ligands and receptors, resulting in more tightly bound

52 -6 compound. The KD of free FA was measured to be 5 ×10 M. For the dendrimer with an

1 -9 average of 2.7 FA (measured by H-NMR), KD =2 × 10 M was obtained, which is a 2500- 15 fold higher binding avidity as compared to free FA.11 Waddell et al. proposed that the multivalent binding with two or more folate interactions is irreversible and nanoparticles that bind to the cell surface via multiple bonds are permanently affixed.53

Based on the literature, our group designed poly(ethylene glycol) (PEG)-based diagnostic agents with two FA targeting groups (Figure 2.6) to test in cancer cells. We used

FL as an imaging agent for cell culture screening where tissue autofluorescence is not a problem. Three compounds were synthesized and tested by the Cleveland Clinic: FA-FL-

NH-PEG-NH-FL-FA, difunctional FA-FL-FA and monofunctional FL-FA in TNBC cell lines which will be discussed in the next section.

Figure 2.6 Comparison of binding of monovalent and divalent FA-PEG-FL conjugates to FRs.

16 2.2. Modular approach to FA-targeted diagnostic devices

FL was chosen as a diagnostic agent because it is inexpensive, exhibits fluorescence in the visible range,12 and is being widely used as an imaging agent.13.,14 PEG was chosen because it is a hydrophilic polymer which is widely used in biomedical application.16

PEGylation increases the solubility of hydrophobic drugs in the body and increases circulation time.16 Our group has developed a novel method to synthesize thiol- functionalized FA (FA-γ-SH) 15 which was used to make the conjugates. Figure 2.7. shows the concept of our modular approach. The linker system between each module is a pH sensitive ester bond which can be hydrolyzed under acidic condition inside the endosome and lysosome.54

Figure 2.7 Modular representation of the folate-targeted FL-PEG conjugate.

One of the main features of this project is enzyme-catalyzed transesterification and

Michael addition reactions for the synthesis of the conjugates. Candida antarctica lipase

B (CALB) which belongs to α/β-hydrolase family was used in this research. CALB has a

17 molecular weight 33,273 DA and is made up of 317 amino acids.55 The catalytic activity of CALB in transesterification is due to the presence of three amino acids Serine105-

Histidine 224-Asparate187 which form the catalytic triad,19 while in Michael addition

Serine105 is not involved. The lid-like structure of CALB protects the active sites of the enzyme and opens when it is exposed to the suitable system which results in its catalytic activity.55 CALB offers high selectivity, catalyst recyclability and reactions can be operated under mild conditions. As CALB offers greener synthesis approach, this work is focused on chemo-enzymatic synthesis.

FA-FL-NH-PEG-NH-FL-FA, which is the PEG-based folate-targeted difunctional conjugate, was synthesized by CALB-catalyzed Michael addition between PEG-diamine

(H2N-PEG-NH2) and FL o,o’-diacrylate (FL-DA) at 50℃ in dimethyl sulfoxide (DMSO), followed by CALB-catalyzed Michael addition of FA-γ-SH (Figure 2.8, Strategy 1).17

Figure 2.8 Synthesis of FA-FL-NH-PEG-NH-FL-FA. 18

FA-FL-FA (Figure 2.9) and FL-FA (without PEG) (Figure 2.10) were also synthesized for comparative purposes by CALB-catalyzed Michael addition of FA-γ-SH to FL o-acrylate (FL-A) and FL-DA, respectively.17 A comparative study was conducted by the Cleveland Clinic on all the compounds.

Figure 2.9 . FA-FL-FA.

Figure 2.10 FL-FA.

2.2.1. Cleveland Clinic comparative trials

To demonstrate the cytotoxicity and bioavailability of the compounds, human breast cancer cell lines MDA_MB_231 (epithelial cells_HTB-26_adenocarcinoma_patient ethnicity Caucasian) (MB_231) and MDA_MB_468 (epithelial cells HTB

132_adenocarcinoma_patient ethnicity African-American) (MB_468) were used.

Immunocytochemistry was performed which showed differential expression of FR on the cells (Figure 2.11). Blue color represents cell nuclei while green color represents fluorescent secondary antibodies attached to the primary FR-specific antibody.

It can be seen in Figure 2.11 that the MDA_MB_231 cell line showed higher level of FR expression than MDA_MB_468. Three different doses of each compound at 5, 10 and 30 µg/ml were used to assess bioavailability and cytotoxicity. Two hours after incubating the cells, they were harvested for flow cytometry analyses.

Figure 2.11 Comparison of MD-MB-231 and MD-MB-468 TNBC cells for FR-density (Unpublished data).

First cell uptake and cytotoxicity of FL-FA was studied by flow cytometry and compound localization by confocal microscopy as shown in Figure 2.12. In Figure 2.12B, blue color represents cell nuclei while green color represents FA-FL. FL-FA which had one FA targeting agent and there was no dose dependent uptake. Due to the small size of the molecule with one FA, uptake of the compound was deemed to be possible through passive diffusion. Toxicity was observed at the highest dose of 30 µg/mL.

20 100.00 FL Dead 90.00 80.00 70.00 60.00 50.00 40.00 30.00 % Cell Population % Cell 20.00 10.00 0.00 0 5 10 30 Concentration (µg/ml)

A B

Figure 2.12 (A) FL-FA uptake and cytotoxicity in MDA-MB-231 TNBC cells. 2 hours incubation followed by propidium iodide staining and process in BD LSR II flow cytometry. Data analyzed in Flow Jo 10 analytical software, N= 4. (B) Confocal microscope image (Unpublished data).

Similarly, cytotoxicity and cell uptake data were collected for FA-FL-FA as shown in Figure 2.13. This compound had 2 FA for binding with overexpressed FR and showed higher specificity for binding than FL-FA. Strong dose dependent uptake was observed for this compound.

21 100.00 FL Dead 90.00 80.00 70.00 60.00 50.00 40.00

30.00 % Cell Population Cell % 20.00 10.00 0.00 0 5 10 30 Concentration (µg/ml)

A B

Finally, FA-FL-PEG2000-FL-FA was studied for cell uptake and cytotoxicity as shown in Figure 2.14. In Figure 2.14B, green color represents FA-FL-PEG2000-FL-FA. Due to the strong fluorescence of the internalized compound, no blue colored cell nuclei are visible in any cells. Dose dependent increased uptake was observed for this compound. At

10 μg/mL dose, 60% cell uptake was observed while at 30 μg/mL dose, cellular uptake of this compound was >80%.

100.00 FL Dead 90.00 80.00 70.00 60.00 50.00 40.00

30.00 % Cell Cell % Population 20.00 10.00 0.00 0 5 10 30 Concentration (μg/mL)

A B

Figure 2.14 .(A) FA-FL-PEG2000-FL-FA uptake and cytotoxicity in MDA-MB-231 cells. 2 hours incubation followed by propidium iodide staining and process in BD LSR II flow cytometry. Data analyzed in Flow Jo 10 analytical software, N= 4. (B) Confocal microscopy image (Unpublished data).

While flow cytometry gives uptake data, the location of the compounds cannot be distinguished by this method (Unpublished data). Confocal and Z-stack imaging analysis confirmed that the compounds were internalized, as shown by fluorescence green inside the cells (Figures 2.12 – 2.14). Cells untreated with the FL-containing compounds were used as controls and flow cytometry did not detect any fluorescence for controls and they appeared black in confocal microscopy.

In vivo analysis of the difunctional PEG-based compound was conducted for a liver cancer rat model by the Cleveland Clinic.17 Intra-arterial (IA) and intravenous (IV) injections were given to Sprague–Dawley rats. However, background noise was obtained due to autofluorescence. When the compound was directly injected into an artery (local administration) which is feeding the liver of the rat, localized fluorescence intensity in the liver was obtained due to accumulation of the compound as shown in Figure 2.15. IA

23 approach was found to be better than the IV approach for the drug delivery as drug as directly released in the region of interest.17

Figure 2.15 Intra-arterial injection of the PEG-based compound in liver cancer rat model.17

Image reprinted with permission from J. Biomed. Mater. Res. Part A. 2019, 107 (11),

2.2.2. New synthetic strategies

Based on these results, we set out to design new synthetic strategies as shown in

Figure 2.16, because we were concerned about side reactions in Strategy 1. Three more strategies are investigated in this dissertation. In strategy 2, FL-DA was replaced by FL-A.

FL-A improved the selectivity of Michael addition of H2N-PEG-NH2, followed by the acrylation using acryloyl chloride (AcrCl) to attach the FA-γ-SH targeting agent via

Michael addition. In strategy 3, H2N-PEG-NH2 is replaced by PEG-dithiol (HS-PEG-SH) and tetraethylene glycol-dithiol (HS-TEG-SH) to avoid the interference of the amine group

24 in the acrylation reaction. PEGylation of FL by lithiation chemistry was also conducted using n-butyl lithium as described in this dissertation.

26 Figure 2.16 Synthetic strategies to be investigated.

Next the components of the new strategies will be discussed in more detail.

27 2.3. Components of the new strategies

2.3.1. Fluorescein (FL)

FL was selected as the diagnostic agent for cell culture studies (Figure 2.17).

Figure 2.17 Structure of FL.

FL is widely used as a precursor for various fluorescence probes, dyes and drugs.12-

14,56-58 For example, FL-derivatives were used as contrasting agents to provide real-time, high-resolution images in gastrointestinal endoscopy, ophthalmology, neurosurgery and dermatology.13,14 The high photostability and fluorescence quantum efficiency of FL-based compounds in aqueous media make them favorable as fluorescent probes.57-59 FL has high molar absorptivity at 490 nm and both the excitation (490 nm) and emission (530 nm) wavelengths of FL are in the visible region.12 FL has an 11-minute plasma half-life and it is easily excreted from the body within 48 hours after administration.14 After injecting FL,

70-80% binds strongly to the plasma and stays in the vessels which makes it useful to visualize cell morphology and subcellular details.14

FL exists in tautomeric forms which are in dynamic equilibrium with each other in solution as shown in Figure 2.18.60 Hydrogen bonding with the solvent plays a crucial role in the tautomeric equilibria. For example, in DMSO the concentration of the lactone form is three orders of magnitude higher than the concentration of the quinoid form.60

28 Figure 2.18 Tautomeric forms of FL in dynamic equilibrium.60

The crystalline form of each tautomer of FL has a specific color. The quinoid form is brick-red in color, the zwitterionic crystals are yellow while the pure lactone form of FL is colorless.61 However, the FL lactone form is usually crystallized in solvents such as acetone, methanol, and 1,4-dioxane and exhibits color due to the presence of the solvate form. The color of the FL lactone solvates depends on the particle size and the solvent.61

Literature reports that FL lactone may show a yellow color due to the partial loss of solvent molecules from the crystal surface and condensation of water molecules leading to the formation of a yellow zwitterionic surface layer.61 The sodium salt of FL is an orange-red powder known as ‘Uranine.’62 It is water soluble and shows a bright green color in aqueous alkaline medium.62 Dihydro-FL or fluorescin (CAS: 518-44-5) is a reduced nonfluorescent probe while the quinoid FL is an oxidized probe (Scheme 2.1.).63 Fluorescin was used to develop a fluorescent probe for intracellular reactive oxygen species (ROS) as it oxidizes to the quinoid form resulting in fluorescence.64

29 Scheme 2.1 Structure of dihydro-FL and quinoid FL.63

FL exhibits different forms at various pH in aqueous solution as shown in Figure

2.19. and hence fluorescence of FL is highly pH dependent.65 When the pH of the aqueous solution is strongly acidic, the cation is a predominant form. At pH 2-4, FL mainly appears as neutral species. At pH 4.3-6.4, the mono-ionic form is prominent, and at pH above 6.4 the dianion is the most prevalent. Fluorescence intensity increases when the pH of the solution increases above 6.5 ppm.65

30 Figure 2.19 Structures of FL at various pH values.65

FL often is used with an isothiocyanate functionality (FITC) for convenient conjugation with amine-functionalized compounds. Flavins and flavoproteins are endogenous fluorophores which interfere with the fluorescence of the imaging agent.66

These endogenous fluorophores absorb in the 450-500 nm range and hence cause background tissue fluorescence.66 While FL is very convenient for in vitro testing, in vivo

31 it would be advantageous for the imaging agent to have a long wavelength emission to filter out background fluorescence.67 For future works Doxorubicin (DOX) is a good choice since it is a theranostic agent – both diagnostic and therapeutic agent.68

2.3.2. Poly(ethylene glycol) (PEG) linker

We selected PEG as a linker between the diagnostic agent and targeting molecule

FA (Figure 2.20).

Figure 2.20 Structure of the PEG.

PEG is being used in variety of pharmaceutical products approved by the Food and

Drug Administration (FDA).16 PEG is widely used in many drugs and hygiene products.

However, a few hyper allergy- anaphylaxis cases related to PEG have been reported. In

2016, 7935 anaphylactic cases were registered among which only three were due to PEG in Europe.69 Hence, it is necessary to create the awareness about PEG allergy before its administration.

The hydroxyl end groups of PEG are easy to modify by various reactions, such as transesterification and esterification. Hydrophilicity of PEG enhances the aqueous solubility of hydrophobic diagnostic agents and drugs attached to the conjugate. Increased circulation of PEG conjugates inside the body reduces the amount of the drug required and its administration frequency. PEG with Mn < 20kDa is excreted via the kidneys; literature reports 85% excretion in 12 hours after the intravenous injection of 1 g of 1kDA PEG.16

Tumor cells have abnormal transport of fluids and lack lymphatic drainage which results in the so-called Enhanced Permeability and Retention (EPR) effect.70 Hence, macromolecules, such as PEGylated diagnostic/drug conjugates are retained better inside tumors. This EPR driven delivery is observed in PEG-based conjugates and hence leads to achieve a 10 to 50 times higher concentration of drug at the tumor site than the normal tissues within 1-2 days.70 As rapidly dividing tumor cells have higher receptor mediated endocytosis than normal cells, effective diagnosis can be achieved by targeting folate- mediated endocytosis, followed by the EPR effect due to the PEG.16,70 Low polydispersity index of PEG is an important factor which assures homogeneity of the residence time of the drug delivery vehicle.16 Discrete PEG (dPEG) has a specific, single molecular weight distribution (Đ= 1.00).71 We elected to include dPEG in our studies to allow us to make a conjugate with precise molecular weight. This will be the first time dPEG will be used to produce a divalent FA-targeted diagnostic agent.

In strategies 2-4 we will use PEG-diamine and thiol-functionalized PEG, so these will be discussed in more detail.

2.3.2.1. PEG-diamine (H2N-PEG-NH2)

H2N-PEG-NH2 (CAS: 24991-53-5, Figure 2.21) is available commercially. H2N-

PEG-NH2 synthesis is reported in a literature by different synthetic routes such as first protecting the hydroxy group using p-toluenesulfonyl (Ts) chloride, followed by reaction of TsO-PEG-OTs with 28% ammonia which results in 95% conversion and 75% of reaction yield.72

Figure 2.21 PEG-diamine.

H2N-PEG-NH2 synthesis is mainly conducted using NH3 as reported in many studies.73-76 However, formation of the secondary amine is a major problem with this

76 synthetic route. H2N-PEG-NH2 can also be synthesized by converting PEG to halide followed by replacing the halide with azide and reducing the azide group to primary amine.

This method does not involve formation of secondary amine.76 Mongondry et al. reacted

PEG with phthalimide in the presence of triphenylphosphine and diisopropylazodicarboxylate in tetrahydrofuran (THF) at room temperature.77 Phthalimido-

PEG product was then reacted with hydrazine hydrate (80%) and 75-85% yield of H2N-

77 76 PEG-NH2 is reported. This process does not produce secondary amine byproduct.

Literature reported the direct esterification between amino acids and PEG with Mn

= 10,000 g/mol using 1,3-dicyclohexylcarbodiimide (DCC) (Scheme 2.2).76,78 The free

‘NH2’ group of the amino acid are protected with tert-Butyloxycarbonyl (Boc) followed by esterification in dichloromethane. This method was further used to develop a pentapeptide.76, 78

Scheme 2.2 Direct esterification of PEG with amino acid using DCC.78

Acid-catalyzed esterification of PEG (Mn = 6000 g/mol) with amino acids in benzene was also reported with quantitative yield. Reaction time of 2-3 days was required, and no protection of the amino groups was necessary for this reaction.78

Mulay et al. developed a new method (to be published) to synthesize H2N-PEG-

NH2 (Figure 2.22) by CALB-catalyzed esterification between Boc-protected β-alanine and

PEG. This method will be used in our group in the future to synthesize PEG-diamine.

Figure 2.22 PEG-diamine synthesized by CALB-catalysis.

2.3.2.2. PEG-dithiol (HS-PEG-SH)

CALB-catalyzed transesterification of methyl 3-mercaptopropionate (MP-SH) with TEG and PEG (Mn = 1000 g/mol, Ð = 1.14 and Mn = 2050 g/mol, Ð = 1.09) was reported by Mulay et al.22 The reactions were conducted in bulk without using any solvent

(Scheme 2.3).22 Product formation was found to occur in a step-wise manner with TEG- monothiol (HS-TEG-OH) obtained in 15 minutes (yield: 93%), followed by TEG-dithiol 35

(HS-TEG-SH) in 8 hours (yield: 88%). Fresh additional CALB and MP-SH were added to form HS-TEG-SH with 100% conversion. In case of PEGs (Mn= 1000 g/mol and 2050 g/mol), monothiol (HS-PEG-OH) formed in 8 and 16 hours, respectively, with ~100% reaction yield.22 HS-PEG-SH formation needed more than 24 hours with fresh additional

CALB and MP-SH. Reaction yield of 85% and 94% were obtained for PEGs (Mn= 1000 g/mol and 2050 g/mol).22 PEG-dithiol synthesized using this new method will be used in the new synthetic schemes discussed in this thesis.

Scheme 2.3 CALB-catalyzed transesterification of MP-SH with TEG.22

2.3.3. Thiol-functionalized folic acid (FA-γ-SH)

In the literature FA-conjugation with polymer invariably involves the so-called activated ester method in which DCC and 1-ethyl-3- (3-

(dimethylamino)propyl)carbodiimide (EDC) are used as condensing agents with or without activators such as N-hydroxysuccinimide (NHS) and hydroxybenzotriazole (HOBt).15

However, the activated ester method results in the formation of mixture containing both α and γ conjugation. Purification and characterization of the desired FA-γ-conjugated polymer from the mixture is difficult and tedious. For example, high performance liquid chromatography (HPLC) was reported to separate the desired γ-substituted FA attached to

15 1 G5-PAMAM dendrimer. HPLC of a dendrimer with ~2.7 FA (measured by H NMR) showed the stochasticity and broad distribution of the FA on dendrimer as only 13% of the sample had 3 FA,16% had no FA, 22% contained only 1 FA and the remaining had more than 3 ligands.11 Puskas et al. reported the synthesis of FA-γ-SH that did not require HPLC separation. First FA was lithiated exclusively at the γ-COOH position using n-BuLi to form the metal salt. FA-γ-Li was then reacted with a bromine-functionalized disulfide compound

(Br-S-S-Br), forming FA-S-S-FA. Breaking the disulfide bond using dithiothreitol (DTT) in dimethylformamide (DMF) resulted in the formation of FA-γ-SH (Scheme 2.4).15 FA-

γ-SH made using this new method will be employed in our synthetic strategies.

Scheme 2.4 Synthesis scheme of FA-γ-SH.15, 79

2.4. Construction of the conjugates

2.4.1. Acrylation of FL

Acrylation of FL to synthesize FL-DA and FL-A is reported in the literature using

AcrCl and triethylamine (TEA) in dichloromethane.56,57, 80-82 Wang B. reported the FL-A synthesis by reacting FL with 1.22 equivalent of AcrCl at 0℃ in presence of 3.97 equivalent of TEA and stirring the reaction mixture at room temperature overnight. The

FL-A product was purified by column chromatography with CHCl3/ethanol (30/1) solvent mixture and 60% yield was achieved.57 Wang H. et al. synthesized FL-A (with 2 equivalents of TEA, 1.5 equivalents of AcrCl) and FL-DA (with 4 equivalents of TEA and

4 equivalents of AcrCl) with yields of 78% and 75%.56 Both reactions were performed by stirring the reaction mixtures overnight at room temperature after the addition of AcrCl.

He purified both FL-A and FL-DA by silica gel column chromatography with the solvent

56 system of CH2Cl2:CH3OH (100:0.5) for FL-A and CH2Cl2: CH3OH(100:0.2) for FL-DA.

Lu et al. synthesized FL-DA with 3 equivalents of TEA and 3 equivalents of AcrCl in dichloromethane and stirring the reaction mixture overnight at room temperature.82

Purification by silica gel column chromatography with CH2Cl2/ CH3OH (100/1) to obtain

45% yield is reported.82 Ma et al. acrylated FITC with AcrCl and TEA in dichloromethane and stirred the reaction mixture for 2 hours at room temperature.80 Column chromatography on silica gel by ethyl acetate/hexane=1:5 yielded 70% of FITC-diacrylate.80 The literature reports yields 75% with FL quinoid81 and 45% with FL lactone.82 Against this background

I developed effective methods to synthesize FL-A and FL-DA that will be discussed in this thesis.

2.4.2. Michael addition

Michael addition was used to attach PEG-diamine and PEG-dithiol to acrylated FL, and FA-γ-SH to Acryl-FL-PEG-FL-Acryl. Hence, this reaction will be discussed in more detail.

2.4.2.1. “Traditional” Michael addition

Michael addition is a conjugate addition-type reaction between an α, β-unsaturated electrophile and a nucleophile substrate to generate carbon-carbon or carbon-heteroatom bonds with amines, thiols, and phosphines.83 For aza-Michael addition reactions with amines, a base catalyst may not be required as the nucleophilic amines also act as a base.

Secondary amines are more basic than primary amines, and act as a stronger nucleophile depending on their steric hindrance and electronic environment.84 Desmet et al. studied the mechanism of aza-Michael addition reaction between diethylamine and ethyl acrylate in tetrahydrofuran (THF) (Scheme 2.5).84

Scheme 2.5 Aza Michael addition reaction between diethylamine and ethyl acrylate in THF.84

The mechanism of aza-Michael addition between the diethylamine and ethyl acrylate is reported. In the first step, the amine nucleophile attacks the β carbon leading to the transfer of a proton and generates a zwitterionic intermediate. In the next step, proton transfer takes place to the carbon in the α position of the carbonyl group (1,2-addition)

40 leading to the formation of intermediate (II) and (III). There are two possible mechanisms of proton transfer during 1,2-addition: (1) amine assisted 1,2-nucleophilic addition forming intermediate (II), which is a six-membered ring structure to give the keto-product in the final step. However, due to the substitution on the secondary amine, more steric hindrance occurs in this step than with a primary amine. (2) Direct 1,2-nucleophilic addition that is not assisted by an amine and forms intermediate (III), which is a four membered ring to give the keto-product in the final step. The amine-assisted mechanism is thermodynamically more favored than that of the direct addition.84 In this mechanism, keto- enol tautomerism of the final product is shown.84

Figure 2.23 Mechanism of aza-Michael addition between diethylamine and ethyl acrylate in THF.84

Thiol-ene reactions are known to proceed through a Michael addition pathway.

These reactions are catalyzed by either a base or a nucleophile, resulting in an anti-

Markovnikov addition product similarly to a thiol-ene radical addition.85

2.4.2.2. CALB- catalyzed Michael addition

Literature reports that CALB-catalyzed Michael addition of small molecules involves the catalytic diad of His224 and Asp187, and Ser103 is not involved.55

Scheme 2.6 shows the CALB catalyzed Michael addition of pyrrolidine to acrylonitrile86 and Figure 2.24 shows our rendition of the mechanism of the reaction.55 The top (dark shaded) portion of the enzyme which is a “carbonyl pocket” consisting of Thr40 and Gln106. It was suggested that the carbonyl pocket activates the nitrile group of the acrylonitrile (Michael acceptor) forming enzyme-substrate complex 1 (ESC-1). Pyrrolidine

(Michael donor), which is a nucleophile, then enters the cycle and forms the complex ESC-

2 with the oxyanion hole and His224 - Asp187 pair.55 At the last step, transfer of the proton, catalyzed by His224- Asp187 pair, to the α-carbon of acrylonitrile leads to the formation of ESC-3 followed by the release of the product.55

Scheme 2.6 CALB-catalyzed Michael addition of pyrrolidine to acrylonitrile.55, 86

Figure 2.24 Reaction mechanism of CALB-catalyzed Michael addition between pyrrolidine and acrylonitrile.55

2.4.2.2.1. CALB-catalyzed Michael addition of amines

Dhake et al. studied the Michael addition of primary and secondary amines to acrylates and found that CALB-catalysis was most efficient.87 Other lipases yielded 10% to 36% of the desired product while CALB yielded 58 % of the product.87

Puskas et al. reported the Michael addition of small molecules such as thymine to vinyl acrylate and to vinyl methacrylate in DMSO. The reactions were conducted at 50℃ 44 for 24 hours in presence of CALB. The reaction with vinyl acrylate resulted in100% conversion while 63% conversion was obtained with vinyl methacrylate due to the hindrance at α-carbon group of vinyl methacrylate.88

Our group introduced CALB-catalyzed Michael addition for the functionalization

88 of low-molecular weight polymers (Mn < 10,000 g/mol) and patented the process. We investigated Michael addition of amine-functionalized polymers to acrylates and methacrylates. CALB-catalyzed Michael addition of aminoethoxy PEG-monomethyl ether to methacrylate-functionalized PEG monomethyl ether was successfully in DMSO at 50℃ for 72 hours (Scheme 2.7). This method was used to prepare block copolymers of PEG and polyisobutylene (PIB).88

Scheme 2.7 CALB-catalyzed Michael addition of aminoethoxy PEG monomethyl ether to methacrylate-functionalized PEG monomethyl ether.88

The synthesis of a 2nd generation PEG-based dendrimer by CALB-catalyzed

Michael addition is also reported by our group.88 Reaction of ω-amino terminated PEG methyl ether (Mn = 2000 g/mol, Đ= 1.05) to 1,3,5-Triacryloylhexahydro-1,3,5-triazine is reported in DMSO at 50℃ (Scheme 2.8). The reaction was completed in 24 hours and the

45 product obtained had two free acrylate groups which can be used further for Michael addition reactions with suitable substrates.88 Attachment of PEG to all three acryl group was conducted in a same manner with 2 more equivalents of the ω-amino terminated PEG

nd methyl ether to yield first generation PEG, PEG-(NH2)2 and further to 2 generation PEG,

88 PEG-(NH2)4.

Scheme 2.8 CALB-catalyzed Michael addition to synthesize PEG-based dendrimer.88

Multifunctional PEGs were synthesized by CALB-catalyzed Michael addition of 2 equivalents of diethanolamine to PEG-diacrylate in presence of CALB at 50℃ in DMSO

(Scheme 2.9). This reaction resulted in 99.5% yield of (OH)2-PEG-(OH)2 product which was further reacted with esters by transesterification.18

Scheme 2.9 CALB-catalyzed Michael addition of diethanolamine to PEG-diacrylate.18

2.4.2.2.2. CALB-catalyzed Michael addition of thiols

With the exception of publications from our group, I could not find any references for CALB-catalyzed Michael addition of thiols to α, β-unsaturated esters.

Our group reported the CALB-catalyzed Michael addition of FA-γ-SH to acrylate-

PEG-FL conjugate at 50℃ in DMSO and 80% conversion of the desired product was obtained after purification.15 We will be using CALB catalysis in this research for the

Michael addition of FA-γ-SH to develop the conjugates described in Strategies 2-4.

2.5. Considerations for reaction scale-up

This thesis will also present scale-up of the synthesis of FL-DA. Scale-up is one of the most important steps of the process development. Reaction and product isolation which is easier at the lab scale can become unsafe and difficult on a large scale.89 Hence, scale- up is an important study which involves applications of concepts of heat transfer, thermodynamics and kinetics for the safe and successful synthesis of the pure product.89

The rate of heat transfer to or from the reaction mixture depend on the physical properties of reactants and coolant used in exothermic reactions.90 It also depends on vessel geometry, agitator speed, size and type of the reactor, agitator and jacket.90 Trivedi et al. explained the perspective for the scale-up and case study on factors for the scale-up synthesis of active pharmaceutical ingredients.91As discussed in the literature, a scale-up study involves heat transfer calculations such as estimation of the enthalpy of the reaction, cooling requirements, determining exothermic behavior and rate of heat transfer by heat balance. In this dissertation, these factors were considered for the scale-up of the synthesis of FL-DA in a 5L capacity jacketed reactor.91

2.5.1. Calculations of the enthalpy of reactions

The reaction to be scaled up is shown in scheme 2.10. Excess AcrCl and TEA were used to make sure the reaction was complete. The side products are salts (TEA.HCl and

AcrTEACl).

Scheme 2.10 Reaction scheme for FL-DA synthesis.

The enthalpy of a chemical reaction can be calculated using Hess’s law:92

∆퐻푟푒푎푐푡푖표푛(kJ/mol) = ∑ ∆퐻푓 푝푟표푑푢푐푡 − ∑ ∆퐻푓 푟푒푎푐푡푎푛푡

Where

∆퐻푓 푝푟표푑푢푐푡푠 = 퐸푛푡ℎ푎푙푝푦 표푓 푝푟표푑푢푐푡 푓표푟푚푎푡𝑖표푛 (kJ/mol)

∆퐻푓 푟푒푎푐푡푎푛푡 = 퐸푛푡ℎ푎푙푝푦 표푓 푟푒푎푐푡푎푛푡 푓표푟푚푎푡𝑖표푛 (kJ/mol)

Hf values of the reactants and products can be obtained from the literature as shown in

Table 2.2.

Table 2.2 Compounds and their heat of formations.

Compound ΔHf (kJ/mol)

Acryloyl Chloride93 -108

TEA 94 -169

TEA hydrochloride salt95 -388

Unfortunately, enthalpies for FL-A, FL-DA and the side product quaternary ammonium salt could not be found in the literature. For this latter we used the value of the

TEA-HCl salt. The values for FL-A and FL-DA were calculated at 298.15 °K using

92 Joback’s group contribution method. This method sums up the ΔH (kJ/mol) groups in the

92 compound in question as follows: ΔHf = 68.29 + Total ΔH

For example, for FL the ΔH values of (ring), (ring) , (ring) and

(ring), (ring), (ring) are summed up, while for FL-DA, (ring), (ring) ,

(ring and non-ring) and (ring), (non-ring), (ring), (non-ring),

(non-ring), (non-ring) are summed up. Joback method considers only additive contribution of functional groups and no interactive contribution is considered.92

For FL and FL-DA ΔH values were also obtained using the MOPAC (Molecular

Orbital PACkage) simulation software. This software first defines geometry of the molecule such as intra-atomic distance, angle of rotation between the bonds and internal coordinate derivatives are obtained which represent energy variations in the molecular system. After drawing the structure, optimized geometry of the molecule in cartesian coordinates is obtained and the heat of formation values are computed.96 The calculations will be presented in the Results section of this dissertation.

CHAPTER III

EXPERIMENTAL

3.1. Materials

Fluorescein (FL), fluorescein o-acrylate (FL-A, 95%), poly(ethylene glycol) diamine (H2N-PEG-NH2, Mn= 2000, Đ ≤1.2), triethylamine (TEA, purity ≥99%), acryloyl chloride (AcrCl, purity ≥97%), diallylamine (DAA, purity 99%), diethyl iminodiacetate

(DIDA, purity 98%), diethanolamine (DEA, purity 99%), 4-mercapto-1-butanol (Purity

95%), 2,2-dimethoxy-2-phenylacetophenone (DMPA/Irgacure 651, purity 99%), 9- mercapto-1-nonanol (Purity 96%), calcium hydride (Reagent grade, 95% (gas- volumetric)), ethyl 5-bromovalerate (EBV, 98%), poly(ethylene glycol) methyl ether

(PEG-OMe, , Mn= 750, Đ= 1.459), n-butyllithium (n-BuLi, 2.0 M in cyclohexane), tetraethylene glycol (TEG, 99%), acetonitrile (anhydrous, 99.8%), dichloromethane (ACS reagent grade), Candida antarctica lipase B (CALB, ≥ 5,000 U/g, 33273 Da, , 20 wt.% immobilized on macroporous acrylic resin, Novozyme® 435, Sigma-Aldrich), tetrahydrofuran (THF, ACS reagent grade), n-hexane (Hexane, ACS reagent grade), and anhydrous dimethyl sulfoxide (DMSO, ≥ 99.9%) were purchased from Sigma-Aldrich and used without any further purification.

Other solvents such as anhydrous diethyl ether (95.8%, BHT free ACS Certified), methanol (ACS Certified), and acetone (ACS Certified) were obtained from Fisher

Chemicals. Deuterated solvents, such as dimethyl sulfoxide (DMSO d6, purity 99.9%),

52 chloroform (CDCl3, purity 99.8%), methylene chloride (CD2Cl2, purity 99.9%), and methanol (CD3OD, purity 99.8%) were purchased from Cambridge Isotope Laboratories.

DL-α-tocopherol (Vitamin E, purity 97+%) was obtained from Alpha Aesar and silica gel classic column (63-200 μm) was obtained from Dynamic Adsorption Inc.

Poly(ethylene glycol)s (PEG1000, 푀̅̅̅푛̅ = 1000 g/mol, Ð = 1.14; and PEG2050, 푀̅̅̅푛̅ =

2050 g/mol, Ð = 1.09), and methyl 3-mercaptopropionate (MP-SH, purity 98%) were obtained from Aldrich Chemicals. Discrete poly(ethylene glycol) (dPEG, Ð =1.00) was bought from Quanta Biodesign Limited.

Fluorescin (<95%, CAS: 518-44-5) was bought from TCI America. PEG-dithiol

(HS-PEG-SH, Mn =899 g/mol, Đ=1.00, Mn =1160 g/mol, Đ=1.14 and Mn =2200 g/mol,

Đ=1.09) and TEG-dithiol (HS-TEG-SH, FW= 370.48 g/mol) were synthesized at Dr.

Puskas’ lab by CALB-catalyzed transesterification reaction.

3.2. Procedures

3.2.1. Syntheses of fluorescein o-acrylate (FL-A) and fluorescein o,o’-diacrylate

(FL-DA)

3.2.1.1. Kinetic studies

A kinetic study of FL-DA synthesis was conducted using Nuclear Magnetic

Resonance (NMR) spectroscopy. FL (0.04 g, 1.2037×10-4 mol, 1 equivalent), TEA (0.04

-4 mL, 2.6481×10 mol, 2.20 equivalents), 0.25 mL CD2Cl2, 0.0026 mL 1,4-dioxane (added as an internal) were measured in a 4 mL vial to obtain a homogenous reddish-brown mixture. The reaction mixture was then transferred into the 5 mm NMR tube and covered with a Teflon cap. A cooling bath of acetone and dry ice (temperature -75°C) was prepared and the NMR tube was kept inside the bath for 30 minutes. 0.3 mL solution of AcrCl

-4 (0.0240 g, 0.02 mL, 2.6481×10 mol, 2.20 equivalents) in CD2Cl2 was prepared and cooled to -75°C in the cooling bath for 30 minutes. Next, the entire AcrCl solution was added into the NMR tube at once. As soon as a drop of AcrCl solution was added, the color of the reaction mixture turned to yellow-orange from dark red. The temperature inside the Varian

Mercury 750 MHz spectrometer instrument was set at -20°C and the NMR tube was placed inside the instrument immediately after the addition of the AcrCl solution. The 1H-NMR spectrum was recorded at -20°C. The temperature inside the instrument was gradually increased and 1H-NMR spectra were recorded with a 3 second interval at temperatures of

-20℃, -15℃, -10℃, -5℃, 0℃, 30℃ to assess the progress of the reaction. After recording the NMR spectra, reaction mixture was added in 15 mL of hexane with stirring. 0.04 g

(9.0890 ×10-5 mol), 75.5% of the pure product was obtained.

3.2.1.2. Synthesis of fluorescein o-acrylate (FL-A)

FL-A was synthesized by reacting FL and AcrCl. FL (4.097 g, 0.0123 mol, 1.00 equivalent) and TEA (6 mL, 0.0430 mol, 3.49 equivalents) were added to a 25 mL round bottom flask equipped with a magnetic stirrer and thermocouple. The flask was purged with nitrogen. Next, dichloromethane (25 mL) was added to the flask and stirred to obtain a homogenous reddish-brown mixture. The flask was then kept in an acetonitrile/dry ice bath at -8 °C. Once the reaction mixture reached -2 °C, AcrCl (1.10 mL, 0.0135 mol, 1.09 equivalents) was added dropwise to the reaction mixture using a syringe. The temperature increased to 0 to 2℃ during the addition of AcrCl. After the complete addition of AcrCl, the color of the reaction mixture turned yellow-orange. The flask was then removed from the acetonitrile bath and covered with aluminum foil to avoid exposure to light. It was then stirred at room temperature for 2 minutes. THF (50 mL) was added to the flask and the

TEA salt was precipitated. The reaction mixture was then filtered using a Whatman #2 filter paper and the dichloromethane was evaporated under vacuum using a rotary evaporator. 35 mL of THF was again added to precipitate the residual TEA salt which was then filtered off. The solution was concentrated to 15 mL by evaporating THF. It was then added dropwise into hexane (300 mL) with vigorous stirring. The precipitated solid yellow product was then filtered and dried under vacuum for 48 hours. The product was analyzed by thin layer chromatography (TLC) with the solvent system of 80 v% diethyl ether and

20 v% n-hexane. Pure commercial FL-A (Sigma-Aldrich), the starting material FL (Sigma-

Aldrich), FL-DA (synthesized in the lab) and the newly synthesized FL-A product were spotted on a TLC plate. Column chromatography was then performed with a solvent system

55 of 50 v% hexane and 50 v% diethyl ether solvent mixture and the polarity of the solvent system was gradually increased to 80 v% diethyl ether and 20 v% hexane. All fractions were collected. Each fraction was analyzed by TLC and a fraction of pure FL-A was isolated by evaporating the solvents at reduced pressure. 2.5 g (0.006471 mol) 52.49% of pure product was obtained.

3.2.1.3. Synthesis of fluorescein o,o’-diacrylate (FL-DA)

FL-DA was synthesized by reacting FL and AcrCl. FL (3.0583 g, 0.0092 mol, 1.00 equivalent) and TEA (5.24 mL, 0.0376 mol, 4.08 equivalents) were added to a 25 mL round bottom flask equipped with a magnetic stirrer and thermocouple. The flask was purged with nitrogen. Next, dichloromethane (42 mL) was added to the flask and stirred to obtain a homogenous reddish-brown mixture. The flask was then kept in an acetonitrile/dry ice bath at -8 °C. Once the reaction mixture reached -5 °C, AcrCl (3 mL, 0.0369 mol, 4.01 equivalents) was added dropwise to the reaction mixture using a syringe. Temperature increased to 0℃ during the addition of AcrCl. After the complete addition of AcrCl, the color of the reaction mixture turned yellow. The flask was then removed from the acetonitrile bath and covered with aluminum foil to avoid exposure to light. It was then stirred at room temperature for 2 minutes. THF (60 mL) was added to the flask and the

TEA salt was precipitated. The reaction mixture was then filtered using a Q5 filter paper and the dichloromethane was evaporated under vacuum using a rotary evaporator. 100 mL of THF was added to precipitate the residual TEA salt which was then filtered off. 0.1 g silica gel was added to the solution and the mixture was stirred at room temperature for 15 minutes. Silica gel was then filtered out and the solution was concentrated to 15 mL by 56 evaporating THF. It was then added dropwise to 300 mL of hexane with vigorous stirring.

The precipitated solid yellow product was then filtered and dried under vacuum for 48 hours. 2.5872 g (5.878 mmol), 63.9 % of light-yellow product was obtained.

3.2.2. Synthesis of two-functional folate-targeted fluorescein (FL)-based

diagnostic nanodevices

3.2.2.1. Practice: Synthesis of the Cleveland Clinic trial compound FA-FL-FA

Thiol-functionalized folic acid (FA-SH, 0.2414 g, 0.0004 mol, 2.01 equivalents) was measured in 20 mL vial and 0.6 mL DMSO d6 was added to it. The mixture was then stirred at 50℃ in an oil bath to obtain a homogenous brown colored FA-SH solution. FL

(0.09 g,0.0002 mol, 1 equivalent) was measured in a 25 mL round bottom flask and the

FA-SH solution was added dropwise to it. The reaction mixture was stirred at room temperature to dissolve the FL-DA completely. To this reaction mixture, CALB (0.0340 g resin @ 20 wt.% enzyme, 0.0002 mmol, 4×10-3 mol/L) and a drop of vitamin E were added, and the round bottom flask was heated to 52℃ in an oil bath. The reaction mixture continued to stir overnight at 52℃ and then the next day cooled to room temperature.

Further purification and isolation of the product was not carried out.

3.2.2.2. Strategy 2 : 2 FL-A + H2N-PEG-NH2 + 2 AcrCl + 2 FA-SH

3.2.2.2.1. Synthesis of FL-NH-PEG2000 -NH-FL

H2N-PEG2000-NH2 (Mn= 2200 g/mol, Đ ≤1.2, 0.75 g, 0.4 mmol, 1 equivalent) was dried under vacuum (Schlenk line) at 65 ºC in a 25 mL round bottom flask until bubble

57 formation ceased. When it cooled to room temperature, FL-A (0.2899 g, 0.8 mmol, 2 equivalents) and 1 mL DMSO-d6 were added to the dried H2N-PEG2000-NH2. The reaction mixture was stirred at room temperature for 5 minutes to obtain a yellow colored homogenous solution. CALB (0.0624 g resin @ 20 wt.% enzyme, 0.0004 mmol, 3×10-3 mol/L) and a drop of vitamin E (antioxidant) were then added to it and the reaction mixture was heated in an oil bath at 50°C. After 2 hours, the round bottom flask was taken out of the oil bath and the reaction mixture was diluted with 20 mL THF. CALB was filtered out using Q5 filter paper and the THF was removed using rotary evaporator. The product was then precipitated twice in 150 mL of hexane twice and the light-yellow colored product was dried in the vacuum oven for 48 hours. The molecular weight of the product is 2772 g/mol. 0.9 g (0.3247 mmol); 86.58% of the product was obtained.

3.2.2.2.2. Synthesis of Acryl-FL-NH-PEG2000 -NH-FL-Acryl

FL-NH-PEG2000-NH-FL (0.811 g, 0.3 mmol, 1 equivalent) and TEA (0.09 mL, 0.6 mmol, 2.20 equivalents) were added to a 25 mL round bottom flask equipped with a magnetic stirrer and thermocouple. Next, CD2Cl2 (1.5 mL) was added to the flask and stirred to obtain a homogenous reddish-brown mixture. The flask was then kept in an acetonitrile/dry ice bath at -8 °C. Once the reaction mixture reached -8 °C, AcrCl (0.1 mL,

1.2 mmol, 4.20 equivalents) was added dropwise to the reaction mixture using a syringe for 15 minutes. The temperature of the reaction mixture was maintained between -5 and 0

°C. After the addition of the AcrCl, the color of the reaction mixture became yellow. Five minutes after the complete addition of the AcrCl, the flask was removed from the

58 acetonitrile bath and covered with aluminum foil to avoid exposure to light. It was then stirred at room temperature for 24 hours. THF (20 mL) was added to the flask and the TEA salt was precipitated. The reaction mixture was then filtered using a Q5 filter paper and the solvents were reduced to 5 mL under vacuum using a rotary evaporator. The product was recovered by precipitation in 150 mL of hexane four times followed by drying in the vacuum over for 24 hours. 0.52 g (0.1806 mmol), 68.58 % of the product was obtained.

3.2.2.3. Strategy 3: 2 FL-A + HS-(TEG or PEG)-SH + 2 AcrCl + 2 FA-SH

3.2.2.3.1. Synthesis of FL-S-TEG-S-FL

HS-TEG-SH (0.1 g, 0.2699 mmol, 1.0 equivalent) was dried under vacuum

(Schlenk line) at 65 ºC in a 25 mL round bottom flask until bubble formation ceased. When it cooled to room temperature, FL-A (0.2100 g, 0.5 mmol, 2.01 equivalent) and 1 mL

DMSO-d6 were added to the dried HS-TEG-SH. The reaction mixture was stirred at room temperature for 5 minutes to obtain a yellow colored homogenous solution. CALB

(0.04490 g resin @ 20 wt.% enzyme, 0.0003 mmol, 2.48 × 10-4 mol/L) and vitamin E

(antioxidant) were then added to it and the reaction mixture was heated in an oil bath at

50°C. After one minute, the round bottom flask was taken out of the oil bath and was diluted with 15 mL THF. CALB was filtered out using Q5 filter paper and THF was removed using a rotary evaporator. The product was then precipitated twice in 150 mL of hexane and the light-yellow colored product obtained was dried in the vacuum oven for 48 hours. 0.3 g (0.2624 mol), 97.23% of the product was obtained.

3.2.2.3.2. Synthesis of FL-S-PEG-S-FL.

3.2.2.3.2.1. Kinetic study: Synthesis of FL-S-PEG1000-S-FL

HS-PEG1000-SH (0.3833 g, 0.334 mmol, 1 equivalent) was dried under vacuum

(Schlenk line) at 65 ºC until bubble formation ceased and cooled to room temperature. To the dried HS-PEG1000-SH, FL-A (0.2607 g, 0.675 mmol, 2.02 equivalents) and 1.2 mL

DMSO-d6 were added and stirred at room temperature for 15 minutes to obtain a homogeneous solution. Vitamin E (antioxidant) was then added to the solution and the reactor was placed in an oil bath at 52oC. Samples were taken out at 1, 3, 5, 30 minutes, 2 hours, and 18 hours. and the progress of the reaction was monitored by 1H-NMR. After 18 hours, fresh HS-PEG1000-SH was added to the reaction mixture. Five minutes after this addition, a sample was taken out for 1H-NMR. After the NMR analysis, fresh CALB

(0.0556 g resin @ 20 wt.% enzyme, 0.0003 mmol, 2.78 × 10-4 mol/L) was added to the reaction mixture, still at 52oC, and a sample was taken out for 1H-NMR at two minutes.

3.2.2.3.2.2. Synthesis of FL-S-PEG1000-S-FL

HS-PEG1000-SH (0.3833 g, 0.334 mmol, 1 equivalent) was dried under vacuum

(Schlenk line) at 65 ºC in a 25 mL round bottom flask until bubble formation ceased. When it cooled to room temperature, FL-A (0.2607 g, 0.675 mmol, 2.02 equivalents) and 1.2 mL

DMSO-d6 were added to the dried HS-PEG1000-SH. The reaction mixture was stirred at room temperature for 5 minutes to obtain a yellow colored homogenous solution. CALB

(0.0556 g resin @ 20 wt.% enzyme, 0.0003 mmol, 2.78 × 10-4 mol/L) and vitamin E

(antioxidant) were then added to it and the reaction mixture was heated in an oil bath at

50°C. After one minute, the round bottom flask was taken out of the oil bath and the reaction mixture was diluted with 18 mL THF. CALB was filtered out using Q5 filter paper and THF was removed using rotary evaporator. The product was then precipitated twice in

150 mL of hexane and the light-yellow colored product was dried in the vacuum oven for

48 hours. 0.5022 g (0.3245 mol), 97.13% of the light-yellow colored product was obtained.

3.2.2.3.2.3. Synthesis of FL-S-PEG2000-S-FL

HS-PEG2000-SH (0.4074 g, 0.2 mmol, 1 equivalent) was dried under vacuum

(Schlenk line) at 65 ºC until bubble formation ceased and cooled to room temperature. To the dried HS-PEG2000-SH, FL-A (0.1429 g, 0.4 mmol, 2.01 equivalents) and 1.2 mL

DMSO-d6 were added and stirred at room temperature for 15 minutes to obtain a homogeneous solution. CALB (0.0581 g resin @ 20 wt.% enzyme, 3.49 × 10-4 mmol, 3 ×

10-4 mol/L) and vitamin E (antioxidant) were then added to the solution and the reactor was placed in an oil bath at 52oC. After one minute, the reaction mixture was diluted with 15 mL THF, filtered out using Q5 filter paper and dried using a rotary evaporator. The reaction mixture was then precipitated twice in 100 mL of hexane and the light-yellow colored precipitate was dried in the vacuum oven for further analysis. 0.5 g, (0.167 mmol) 91.02% of the light-yellow product was obtained.

3.2.2.3.3. Synthesis of Acryl-FL-S-TEG-S-FL-Acryl

FL-S-TEG-S-FL (0.7 g, 0.6 mmol, 1 equivalent) and TEA (0.51 mL, 3.7 mmol, 6 equivalents) were added to a 25 mL round bottom flask equipped with a magnetic stirrer

61 and thermocouple. Next, CD2Cl2 (3 mL) was added to the flask and stirred to obtain a homogenous reddish-brown mixture. The flask was then kept in an acetonitrile/dry ice bath at -8°C. Once the reaction mixture reached -8 °C, AcrCl (0.5 mL, 6.1 mmol, 10 equivalents) was added dropwise to the reaction mixture using a syringe for 15 minutes. The temperature of the reaction mixture was maintained between -5 and 0°C. After the addition of the AcrCl, the color of the reaction mixture became yellow. Five minutes after the complete addition of the AcrCl, the flask was removed from the acetonitrile bath and covered with aluminum foil to avoid exposure to light. It was then stirred at room temperature for 24 hours. THF (15 mL) was added to the flask and the TEA salt was precipitated. The reaction mixture was then filtered using a Q5 filter paper and the solvents were reduced to 5 mL under vacuum using a rotary evaporator. The product was recovered by precipitation in 150 mL and 100 mL of hexane followed by drying in the vacuum oven for 48 hours. 0.555 g (0.4435 mmol), 72.43 % of the light-yellow product was obtained.

3.2.2.3.4. Synthesis of Acryl-FL-S-PEG-S-FL-Acryl compounds

3.2.2.3.4.1. Synthesis of Acryl-FL-S-dPEG899 -S-FL-Acryl

FL-S-dPEG899-S-FL (0.7853 g, 0.4 mmol, 1 equivalent) and TEA (0.39 mL, 2.8 mmol, 6.59 equivalent) were added to a 25 mL round bottom flask equipped with a magnetic stirrer and thermocouple. Next, CD2Cl2 (5.10 mL) was added to the flask and stirred to obtain a homogenous reddish-brown mixture. The flask was then kept in an acetonitrile/dry ice bath at -8°C. Once the reaction mixture reached -8°C, AcrCl (0.35 mL,

4.3 mmol, 10 equivalents) was added dropwise to the reaction mixture using a syringe for

15 minutes. The temperature of the reaction mixture was maintained between -5 and 0°C.

After the addition of the AcrCl, the color of the reaction mixture became yellow. Five minutes after the complete addition of the AcrCl, the flask was removed from the acetonitrile bath and covered with aluminum foil to avoid exposure to light. It was then stirred at room temperature for 24 hours. THF (20 mL) was added to the flask and the TEA salt was precipitated. The reaction mixture was then filtered using a Q5 filter paper and the solvents were reduced to 5 mL under vacuum using a rotary evaporator. The product was recovered by precipitation in 150 mL of hexane followed by drying in the vacuum oven for 48 hours. 0.6559 g (0.3297 mmol), 77.6% of the light-yellow product was obtained.

3.2.2.3.4.2. Synthesis of Acryl-FL-S-PEG1000-S-FL-Acryl

FL-S-PEG1000-S-FL (0.4021 g, 0.0003 mmol, 1 equivalent) and TEA (0.22 mL,

0.0122 mmol, 6 equivalents) were added to a 25 mL round bottom flask equipped with a magnetic stirrer and thermocouple. Next, CD2Cl2 (2 mL) was added to the flask and stirred to obtain a homogenous reddish-brown mixture. The flask was then kept in an acetonitrile/dry ice bath at -8 °C. Once the reaction mixture reached -8 °C, AcrCl (0.21 mL, 0.0026 mmol, 10 equivalent) was added dropwise to the reaction mixture using a syringe for 15 minutes. The temperature of the reaction mixture was maintained between -

5 and 0°C. After the addition of the AcrCl, the color of the reaction mixture became yellow.

Five minutes after the complete addition of the AcrCl, the flask was removed from the acetonitrile bath and covered with aluminum foil to avoid exposure to light. It was then stirred at room temperature for 24 hours. THF (15 mL) was added to the flask and the TEA

63 salt was precipitated. The reaction mixture was then filtered using a Q5 filter paper and the dichloromethane was evaporated under vacuum using a rotary evaporator. The product was recovered by precipitation in 150 mL of hexane followed by drying in the vacuum over for

48 hours. 0.33 g (0.1615 mmol), 62.2% of the light-yellow product was obtained.

3.2.2.3.4.3. Synthesis of Acryl-FL-S-PEG2000 -S-FL-Acryl

FL-S-PEG2000-S-FL (410 mg, 0.1 mmol, 1 equivalent) and TEA (0.11 mL, 0.8 mmol, 6 equivalents) were added to a 25 mL round bottom flask equipped with a magnetic stirrer and thermocouple. Next, CD2Cl2 (1 mL) was added to the flask and stirred to obtain a homogenous reddish-brown mixture. The flask was then kept in an acetonitrile/dry ice bath at -8°C. Once the reaction mixture reached -8°C, AcrCl (0.11 mL, 1.4 mmol, 10 equivalent) was added dropwise to the reaction mixture using a syringe for 15 minutes. The temperature of the reaction mixture was maintained between -5 and 0 °C. After the addition of the AcrCl, the color of the reaction mixture became yellow. Five minutes after the complete addition of the AcrCl, the flask was removed from the acetonitrile bath and covered with aluminum foil to avoid exposure to light. It was then stirred at room temperature for 24 hours. THF (15 mL) was added to the flask and the TEA salt was precipitated. The reaction mixture was then filtered using a Q5 filter paper and the dichloromethane was evaporated under vacuum using a rotary evaporator. The product was recovered by precipitation in 150 mL of hexane four times followed by drying in the vacuum over for 24 hours. 0.2605 g (0.0856 mmol), 62.46 % of the light-yellow product was obtained.

3.2.2.4. Strategy 4: PEGylation of fluorescin (also known as dihydrofluorescein)

via lithium technology

3.2.2.4.1. Synthesis of Br-PEG-OMe

PEG-OMe (2.211g, 0.0029 mol, 1 equivalent) was measured in a 25 mL round bottom flask and dried under vacuum (Schlenk line) at 65 ºC until bubble formation ceased.

It was then cooled to room temperature and EBV (3.0818 g, 0.0147 mol, 5 equivalents) was added to it. CALB (0.1 g @ 20 wt.% enzyme, 2×10-4 mol/L) was added to the reaction mixture and heated in an oil bath at 50 °C with 420 Torr vacuum on Schlenk line. After 4 hours, the reaction mixture was diluted with 15 ml of dried THF, filtered over a Q5 filter paper and precipitated three times into 150 mL of hexane to remove excess EBV. The precipitate was then dried in a vacuum oven for 48 hours and 2.2127 g (2.424 mmol),

82.21% of the product was obtained.

3.2.2.4.2. Lithiation followed by PEGylation of fluorescin

3.2.2.4.2.1. Distillation of anhydrous DMSO

Three 25 mL round bottom flasks, one magnetic stirring bar, distillation column and one glass cork were washed and kept in an oven overnight for drying.

12 mL of anhydrous DMSO and calcium hydride (2 g) were measured in a 25 mL round bottom flask with a magnetic stir bar. The distillation was set up with a distillation column, water inlet-outlet, vacuum pump and a collector flask. The flask containing anhydrous

DMSO and calcium hydride was heated and stirred at 50℃ in an oil bath. Full vacuum was applied, and the temperature was gradually increased to 68℃. The initial 4-5 drops of

DMSO distillate were removed as waste. The distillation was then continued for 2 hours and 10 mL pure DMSO was obtained in the collector flask.

3.2.2.4.2.2. Drying and freeze pump thaw of PEG and fluorescin.

Br-PEG-OMe (0.4153 g, 0.455 mmol, 1.1 equivalent) and fluorescin (0.1332 g, 0.4 mol, 1 equivalent) were measured in two separate 25 mL round bottom flasks. Both Br-

PEG-OMe and fluorescin were dried overnight under full vacuum on a Schlenk line at

50℃. After 24 hours, the vacuum was stopped and 1 mL of distilled DMSO (2 mL) was added in each flask. The solution in each flask was stirred under argon at room temperature for 15 minutes. Both solutions of Br-PEG-OMe and fluorescin were degasified using freeze pump thaw technique. Liquid nitrogen was taken in a small dewar and the flask containing fluorescin was placed in it. Once the DMSO was completely frozen, full vacuum was applied for 4-5 minutes. The vacuum was then turned off and the flask was kept in a warm water bath. Once the solution was completely melted, argon was purged into the solution for 4 to 5 minutes. The flask was again placed in the liquid nitrogen dewar. This procedure was repeated 4 times. The same procedure was followed for the Br-PEG-OMe solution.

After freeze pump thaw, both the solutions were stirred under argon at room temperature for 15 minutes.

3.2.2.4.2.3. Lithiation and PEGylation

0.25 mL of n-BuLi (0.0255 g, 0.4 mol, 1 equivalent) was measured using a gas tight syringe in a dry box under nitrogen and added dropwise into fluorescin solution under

66 argon with stirring at room temperature. As soon as the drop of n-BuLi was added, the reaction mixture started turning blood red colored. The reaction was stirred for 15 minutes at room temperature. The n-BuLi syringe was washed with 10 mL of hexane and kept in an oven for drying. Br-PEG-OMe solution was then added dropwise into the reaction mixture. Samples were taken out for 1H-NMR after 15 minutes and 2 hours and precipitated in 150 mL hexane. Hexane was then decanted, and samples were dried in an oven for 24 hours.

3.2.3. Synthesis of tetrafunctional fluoresceins (FL)

3.2.3.1. Synthesis of tetraallyl-functionalized FL

3.2.3.1.1. Reaction in DMSO, with CALB catalysis

FL-DA (0.5049 g, 0.0011 mol, 1.00 equivalent) was dissolved in DMSO d6 (1.00 mL) in a25 mL round bottom flasks. To this solution, DAA (0.2365 g, 0.0024 mol, 2.12 equivalents), and one drop of vitamin E (antioxidant) were added. An immediate color change was observed from yellow to orange in the reaction mixture after the addition of

DAA. Enzyme CALB (0.08 g @ 20 wt. % enzyme, 2.90×10-4 mol/L) was added to the reaction mixture. Reactor was then placed on a stirrer plate and stirred at room temperature.

Sample was taken out at 1 minute and analyzed by 1H-NMR. After one minute, reaction mixture was diluted with dichloromethane (as the product was soluble in dichloromethane and this solvent is easier to evaporate) and CALB was removed from reaction mixture by filtration through Q5 filter paper. A rotary evaporator was then used to remove

67 dichloromethane and excess DAA from the product. The light-orange colored solid product was dried in a vacuum oven for 24 hours to obtain 0.670 g (1.056 mol), 92% of the product.

3.2.3.1.2. Reaction in DMSO, without CALB catalysis

DAA. Reactor was then placed on a stirrer plate and stirred at room temperature. Sample was taken out at 1 minute and analyzed by 1H-NMR. After one minute, reaction mixture was diluted with dichloromethane (as the product was soluble in dichloromethane and this solvent is easier to evaporate) and CALB was removed from reaction mixture by filtration through Q5 filter paper. A rotary evaporator was then used to remove dichloromethane and excess DAA from the product. The light-orange colored solid product was dried in a vacuum oven for 24 hours to obtain 0.655 g (1.001 mol), 89.95% of the product.

3.2.3.1.3. Reaction in chloroform, with CALB catalysis

FL-DA (0.5 g, 0.0011 mol, 1.00 equivalent) was dissolved in CDCl3 (1.00 mL) in a 25 mL round bottomed flask. To this solution, DAA (0.2365 g, 0.0024 mol, 2.12 equivalents), and a drop of vitamin E (antioxidant) were added. An immediate color change was observed from yellow to orange in the reaction mixture after the addition of DAA.

Enzyme CALB (0.08 g @ 20 wt.% enzyme, 2.91×10-4 mol/L) was added. Reactor was then

68 placed on a stirrer plate and stirred at room temperature. Samples were taken out at 1 minute and 10 minute and analyzed by 1H-NMR. The reaction was not further continued after the 1H-NMR analysis.

3.2.3.1.4. Reaction in chloroform, without CALB Catalysis

Reactor was then placed on a stirrer plate and stirred at room temperature. Samples were taken out at 1 minute and 10 minute and analyzed by 1H-NMR. The reaction was not further continued after the 1H-NMR analysis.

3.2.3.2. Synthesis of tetraacetate-functionalized FL

3.2.3.2.1. Reaction in DMSO, with CALB catalysis

FL-DA (1.5057 g, 0.0034 mol, 1.00 equivalent), DIDA (1.60 mL, 0.0089 mol, 2.55 equivalents) and DMSO d6 (1.7 mL) were measured into a 15 mL round bottom flask. One drop of vitamin E (antioxidant) was added to it and the reaction mixture was stirred under nitrogen for 5 minutes at room temperature. The reactor was sealed and placed into an oil bath at 50ºC after the addition of CALB (0.1078 g @ 20 wt.% enzyme, 2×10-4 mol/L).

Aliquots were taken out at 2 hour 45 minutes and 9 hours and the extent of the reaction was measured using 1H-NMR spectroscopy. The reaction mixture was then taken out of

69 the oil bath after 9 hours and cooled to room temperature. After dilution with THF (the product was not soluble in dichloromethane) CALB was filtered out using Q5 filter paper.

The filtrate was reduced by a rotary evaporator and precipitated from 150 mL of hexane two times to remove DIDA. The remainder of the DIDA was removed by heating the product to 85 ºC under 0.8 torr vacuum for 2 hours. 1.9 g (2.323 mmol), 67.8% of the dark red colored product was obtained.

3.2.3.2.2. Reaction in DMSO, without CALB catalysis

FL-DA (1.5057 g, 0.0034 mol, 1.00 equivalent), DIDA (1.60 mL, 0.0089 mol, 2.55 equivalents) and DMSO d6 (1.7 mL) were measured into a round bottom flask. One drop of vitamin E (antioxidant) was added to it and the reaction mixture was stirred under nitrogen for 5 minutes at room temperature. The reactor was sealed and placed into an oil bath at 50ºC. After putting the reactor at the 50℃ temperature oil bath, aliquots were taken out at 5, 30 minutes, 1 hour 30 minutes and 27 hours and the extent of the reaction was measured using 1H-NMR spectroscopy. The reaction mixture was then taken out of the oil bath after 27 hours and cooled to room temperature. After dilution with THF (the product was not soluble in dichloromethane) CALB was filtered out using a Q5 filter paper. The filtrate was reduced by a rotary evaporator and precipitated from 150 mL hexane two times to remove DIDA. The remainder of the DIDA was removed by heating the product to 85

ºC under 0.8 torr vacuum for 2 hours. 1.69 g (2.063 mmol), 60.3 % of the dark red colored product was obtained.

3.2.3.3. Synthesis of tetrahydroxy-functionalized FL

3.2.3.3.1. Reaction of FL-DA with DEA

3.2.3.3.1.1. Reaction in DMSO, with CALB catalysis

FL-DA (0.5000 g, 0.0011 mmol, 1.00 equivalent) was dissolved in DMSO d6 (1.00 mL) in a25 mL round bottom flasks. To this solution, DEA (0.2410 g, 0.0023 mmol, 2.02 equivalents), one drop of vitamin E and enzyme CALB (0.140 g @ 20 wt.% enzyme,

2.73×10-4 mol/L) were added. Reactor was then placed on a stirrer plate and stirred at room temperature. Samples was taken out at 1 minute and analyzed by 1H-NMR. Reaction mixture was then diluted with THF. CALB was removed from one of the reactors by filtration through a Q5 filter paper. The product was recovered by precipitation in 100 mL of diethyl ether.

3.2.3.3.1.2. Reaction in DMSO, without CALB catalysis

NMR. Reaction mixture was then diluted with THF. CALB was removed from one of the reactors by filtration through a Q5 filter paper. The product was recovered by precipitation in 100 mL of diethyl ether.

3.2.3.3.1.3. Reaction in chloroform, with CALB catalysis

FL-DA (0.5000 g, 0.0011 mmol, 1.00 equivalent) was dissolved in CDCl3 (1.5 mL) in two 25 mL round bottom flasks. To this solution, DEA (0.2410 g, 0.0023 mmol, 2.02 equivalents), and one drop of vitamin E were added. In one of the reactors, enzyme CALB

(0.140 g @ 20 wt.% enzyme, 2.73×10-4 mol/L) was added. Both the reactors were then placed on a stirrer plate and stirred at room temperature. Samples were taken out at 1 minute, 5 minute and analyzed by 1H-NMR. However, the reaction was not continued further after 1H-NMR analysis.

3.2.3.3.2. UV-mediated reaction between diallylamine (DAA) and 4-mercapto 1-

butanol to synthesize di-butanol-based amine compound.

DAA (1.1003 g, 0.0113 mol, 1 equivalent) and 4-mercapto 1-butanol (2.6628 g,

0.0251 mol, 2.21 equivalents) were measured in a 20 mL vial reactor and Irgacure 651 photo initiator (0.5980 g, 0.0023 moles, 0.21 equivalents) was added to it. To this solution,

1 mL CDCl3 was added and the reaction mixture was stirred at room temperature for 20 minutes to obtain a homogenous solution.

The UV machine was switched on 20 minutes before the reaction and the power was set to 75 mW. The reactor was placed in an ice bath and the distance between the UV source and the bottom of the reactor was set to 3.2 inches. The reaction mixture was illuminated with UV light for 1 hour 30 minutes. A hazy yellow colored product was obtained. It was then precipitated in 150 mL of hexane. Impurities were removed by column chromatography with a solvent system of 20% methanol-80%THF and the pure

72 product was eluted out with 100% methanol at the end. The product was dried in a vacuum oven for 48 hours and 1.1 g (0.0036 mol), 31.4% of the product was obtained.

3.2.3.3.3. CALB catalyzed Michael addition reaction between FL-DA and di-

nonanol-based amine compound

Di-nonanol-based amine compound which was synthesized by UV mediated thiol- ene click reaction between DAA and 9- mercapto-1-nonanol (using Irgacure-651) was reacted with FL-DA. FL-DA (0.2 g, 0.0005 mol, 1 equivalent) and di-nonanol-based amine compound (0.4492 g, 0.001 mol, and 2.2 equivalents) were measured in a round bottom flask and 1.5 mL CDCl3 was added to it. One drop of vitamin E (antioxidant) was added and the reaction mixture was purged with nitrogen for 10 minutes. As amine (b) is insoluble in CDCl3 at room temperature, the reaction mixture was heated in an oil bath at 80℃ for

15 minutes till a dark yellow homogenous solution was obtained. The round bottom flask was then taken out of the oil bath and CALB (0.0680 g @ 20 wt. % enzyme, 2.7×10-4 mol/L) was added to it. The temperature of the oil bath was the reduced to 65°C and the reaction mixture was heated for 24 hours. The reaction mixture was then cooled to room temperature. After dilution with THF (as the product is soluble in THF), CALB was removed from the reaction mixture by filtration through Q5 filter paper. The volume of

THF was reduced under vacuum using a rotary evaporator and the product was recovered by precipitation from 150 mL hexane. Impurities were removed by washing the column with dichloromethane multiple times and the pure product was then eluted out with 100%

73 methanol. Product was dried in a vacuum oven for 48 hours and 0.029 g (0.02164 mmol),

33.89 % of the orange product was obtained.

3.2.4. Scale-up of fluorescein o,o’-diacrylate (FL-DA) synthesis

3.2.4.1. Experimental study to calculate the enthalpy of acrylation reaction

In a 15 mL round bottom flask, FL (2.04 g, 0.0061 moles, 1 equivalent) and TEA

(2.0314 g, 2.80 mL, 3.27 equivalents) were measured and 20 mL of dichloromethane was added to it. The reaction mixture was stirred at room temperature for 15 minutes to obtain the homogenous dark red-brown solution. Styrofoam cup was half stuffed with insulating cloth and the reactor flask was kept in it. A thermocouple was placed inside the reactor which was closed with a rubber septum and tied with Teflon tape as shown in Figure 3.1.

AcrCl (1.8938 g, 1.70 mL, 3.41 equivalents) was measured using syringe and added entirely at once into the reaction mixture through the rubber septum at the room temperature. Change in temperature before and after the addition of AcrCl was monitored with the thermocouple.

Figure 3.1 Experimental setup to calculate the enthalpy of the acrylation reaction

3.2.4.2. Scale-up of FL-DA synthesis

FL (98 g, 0.2949 mol, 1 equivalent) and TEA (160 mL, 1.1471 mol, 3.89 equivalents) were added to a 5L capacity jacketed cylindrical reactor with overhead stirrer

(Figure 3.2). A thermocouple was attached to the inlet and outlet of the reactor jacket by adhesive tape. First, the reactor was purged with nitrogen. Next, 1250 mL of dichloromethane was added to the reactor and stirred at 120 rpm for 35 minutes to obtain a homogenous reddish-brown mixture.

A cooling bath was prepared using 12 L of aqueous ethylene glycol that was cooled using an electric coil. The coolant was circulated through the jacket with a flow rate of 8

L/min to reduce the temperature of the stirring solution of FL. The temperature of the coolant in the cooling bath before circulating it through the jacket was reduced to -16.6°C. 75

The temperature profile was recorded during the coolant circulation at the following locations: temperature (i) at the inlet (T1), (ii) at the outlet of reactor jacket (T2), (iii) inside the reactor and (iv) inside the cooling bath. All the temperatures were recorded as soon as the coolant circulation was started.

AcrCl (90 mL, 1.1077 mol, 3.76 equivalents) was measured using a measuring cylinder and added into a 200 mL addition funnel. When temperature inside the reactor was reduced to -10℃, AcrCl was added dropwise into the reaction mixture and the mass flow rate of AcrCl was calculated. During the addition of AcrCl, temperature inside the reactor increased to -3℃ and the color of the reaction mixture turned yellow orange. The temperature profile of the coolant and the reaction mixture was recorded during the addition of AcrCl. The reaction mixture continued to be stirred for 3 more by circulating the coolant. After 3 hours, stirring and coolant flow were turned off.

Next day, two phases were obtained in the reactor, yellow layer of TEA hydrochloride salt on the surface of a brown colored product solution. The product was then taken out from the bottom of the reactor and collected in a round bottom flask. After removing the product, yellow solid phase was taken out by washing the reactor with 250 mL dichloromethane and 500 mL THF. 250 mL THF was then added into the brown colored solution of the product and resulted in the precipitation of TEA hydrochloride salt.

Salt was filtered out using a Buchner funnel and filter paper followed by washing the filter paper with 250 mL THF. TEA hydrochloride salt started to reprecipitate in the filtrate which was again filtered out. 250 mL THF was again added and the same procedure was repeated three more times until no more salt was precipitated. Silica gel (138 g) was added

76 into the product solution and stirred for 40 minutes using a magnetic bar. Silica gel was then filtered using a Buchner funnel and filter paper. Silica on the filter paper was washed repeatedly with a total 1 L THF (FL-DA is soluble in THF) and the solution was reduced to 20 mL using a rotary evaporator under reduced pressure. The product was then precipitated in a total of 8 L of hexane. This resulted in a light-yellow powder which was filtered out using a Buchner funnel and filter paper. The product was then dried in a vacuum oven for 48 hours and 70.5 g (0.160 mol), 54.32% of the product was obtained.

3.3. Characterization of products

Characterization of products for this research was conducted using various spectroscopic, chromatographic and analytical techniques. NMR spectroscopy, Matrix assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-ToF MS),

Electrospray ionization mass spectrometry (ESI-MS), Thin layer chromatography (TLC),

Column chromatography were utilized in this research.

3.3.1. Nuclear Magnetic Resonance (NMR) spectroscopy

A Varian Mercury 300 MHz spectrometer was used to record the 1H-NMR spectra in CDCl3/ CD2Cl2/CD3OD/d6-DMSO at 40 mg/mL concentration with the following parameters: 2 seconds relaxation time, 64 scans, and a 45° half angle. A Varian Mercury

13 500 MHz spectrometer was used to record the C-NMR spectra in CDCl3/d6-DMSO at

100 mg/mL concentration with the following parameters: 2 seconds relaxation time, 10,000 scans, and a 90° half angle. A Varian Mercury 750 MHz spectrometer was used to study the synthesis of FL-DA in CD2Cl2 with 3 seconds relaxation delay. 5 mm NMR tubes were used for all the analysis. The resonances of non-deuterated solvents were used as an internal reference for 1H and 13C-NMR. Chloroform at δ = 7.27 ppm and δ = 77.23 ppm, dimethyl sulfoxide at δ = 2.5 ppm and δ = 39.51 ppm, dichloromethane at δ = 5.33 ppm and δ =

54.24 ppm, methanol at δ = 3.34 ppm and δ = 49.86 ppm were set as a reference for 1H and

13C-NMR respectively.

3.3.2. Matrix Assisted Laser Desorption/Ionization Time-of-Flight Mass

Spectrometry (MALDI-ToF MS)

The reported MALDI-ToF-MS spectra were recorded at the University of Akron on a Bruker Ultra-Flex III MALDI-ToF/ToF mass spectrometer (Bruker, Billerica, MA) equipped with a Nd:YAG laser emitting at 355 nm. The instrument was operated in positive ion mode. Samples were dissolved in THF to a final concentration of 10 mg mL-1. Trans-

2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene] malononitrile (DCTB) (20 mg mL-1) served as matrix and sodium trifluoroacetate (NaTFA) (10 mg mL-1) as cationizing agent.

The latter two were prepared and mixed in the ratio 10:1 (v/v), respectively. Matrix:salt and sample solutions were applied onto the MALDI-ToF-MS target plate using the sandwich method. Bruker's FlexAnalysis software was used for analysis.

3.3.3. Electrospray Ionization Mass Spectrometry (ESI-MS)

ESI-MS were performed at The Ohio State Mass Spectrometry Facility on a Bruker

Amazon Ion Trap. The samples were prepared with Methanol:H2O 1:1 with 0.1% formic acid. It was a direct infusion ESI with dry gas flow of 4.0 L/min (temperature 180℃).

Nebulizer pressure of 7.3 PSI, capillary voltage of 4.5 kV and capillary current was 2500 nA.

3.3.4. Thin Layer Chromatography (TLC)

TLC was conducted by spotting the samples on aluminum-based silica gel plates and using solvents for the elution. The plates were then analyzed using either phosphomolybdic acid staining reagent or UV light (λ = 254 nm, λ = 365 nm).

3.3.5. Column chromatography

Silica gel column chromatography was carried out for the purification of some of the products. A silica slurry was prepared by stirring silica gel (Silica gel classic column,

63-200 μm, Dynamic Adsorption Inc) in hexane and pouring into the column. Hexane was then drained out of the column and the level of the solvent was adjusted just above the surface of the silica gel to avoid drying. Crude product was dissolved in a minimum amount of solvent and loaded on the top of the silica gel in the column. Solvent was again drained out to pack the column and the level of the solvent was adjusted just above the silica gel.

Appropriate solvent was then used as an eluent and fractions were collected. Each fraction was analyzed by TLC. Fractions of the pure product were combined and dried under reduced pressure on a rotary evaporator.

CHAPTER IV

RESULTS AND DISCUSSIONS

4.1. Syntheses of fluorescein o-acrylate (FL-A) and fluorescein o,o’-diacrylate

(FL-DA)

My first assignment was to study the kinetics of the acrylation of fluorescein (FL)

(Scheme 4.1) in order to be able to synthesize FL-A and FL-DA effectively.

Scheme 4.1 Synthesis of FL-DA using AcrCl and triethylamine (TEA).

4.1.1. Kinetic studies

The kinetic study of the reaction of FL with AcrCl (2.20 equivalents, 0.4164 mol/L), in the presence of the base TEA (2.20 equivalents, 0.4164 mol/L) was performed in NMR tube using a Varian Mercury 750 MHz spectrometer in deuterated methylene

1 chloride (CD2Cl2) with acquisition time of 3 seconds. First the H-NMR of the individual components were tested. The 1H-NMR spectrum of the starting material AcrCl was recorded in CD2Cl2 at room temperature (25°C) using a 300 MHz instrument as shown in

Figure 4.1. It shows that the Integral ratio of protons (1): (1'): (2) is 1: (1.02):0.90. No other signals appeared in the spectrum. Complex coupling of protons takes place for AcrCl which

81 results in unequal splitting of vinyl signals. Proton (1) is in trans configuration with proton

(2) and hence splits into a doublet with J =16.69 Hz. Also, each doublet signal of (1) splits again into a doublet (not visible on the 300 MHz NMR instrument) due to the geminal coupling with proton (1’) resulting in doublet of doublet. The signal due to proton (2) splits into a doublet due to trans-proton (1) (J=16.69 Hz) and cis-proton (1’) (J= 10.25 Hz).97,98

Similarly, proton signal (1’) split into a doublet of doublet due to geminal coupling with

(1) (Not visible on NMR 300 MHz instrument) and cis-coupling with (2) (J=10.25 Hz).

Hence non-equivalent proton coupling leads to the unequal splitting of vinyl signals97,98

(Figure 4.1).

1 Figure 4.1 H-NMR of AcrCl in CD2Cl2.

1 [AcrCl (300 MHz, H-NMR, CD2Cl2): δ 6.64 ppm ((1), dd, J= 16.69 Hz, 1H), δ 6.42-6.29

((2), dd, 1H), δ 6.20 ((1’), dd, J= 10.25 Hz, 1H)]

As discussed in the background, FL exists in tautomeric forms which are in dynamic equilibrium with each other in solution (Scheme 4.2)60 and hydrogen bonding with the solvent is important in the tautomeric equilibria of FL. It is reported that the concentration of the lactone form of FL is three orders of magnitude higher in dimethyl sulfoxide (DMSO) than the concentration of the quinoid form.60

Scheme 4.2 Tautomeric forms of FL in dynamic equilibrium.60

The 1H-NMR of precursor FL which was bought from Sigma-Aldrich was first checked in DMSO d6 on NMR-300MHz instrument as shown in Figure 4.2. The integral ratios relative to the aromatic proton (13) selected as a reference to all the other protons of

FL (5, 6, 9, 14, 15, 16, 20, 22, 23) are (1:1:1:1:1:1:1:1:1) as expected from the structure.

As reported in the literature, lactone form of FL is predominant in DMSO d6.60 Other tautomers of FL such as quinoid and zwitterionic are present in a minute amount,60 which could not be detected on NMR 300 MHz spectrometer.

Figure 4.2 1H-NMR of FL in DMSO d6.

[FL (300 MHz, 1H-NMR, DMSO d6): δ 10.10 ((OH), s, 2H), 8.01 ((13), dt, J=7.61Hz, 1H),

δ 7.84-7.69 ((15,16), m, 2H), δ 7.28 ((14),d, J=7.61 Hz, 1H), δ 6.70 ((6,23),d, J= 1.17Hz,

1.46 Hz, 2H), δ 6.56 ((9,20,22,5), m, J=1.46 Hz, 4H)]

The 13C-NMR spectrum in Figure 4.3 shows that the resonance due to carbon (10) is present at 83.31 ppm which confirms the predominant lactone FL. For the quinoid FL, resonance at 185 ppm and 134 ppm should be present corresponding to ‘carbonyl carbon

(21)’ and ‘central carbon (10)’ which could not be detected by the 13C-NMR 500 MHz due to extremely low amount.

Figure 4.3 1H-NMR of FL in DMSO d6.

[FL (500 MHz, 13C-NMR, DMSO d6): δ 168.91 (11), 159.86 (21,4), 152.73 (8,19), 152.13

(17), 135.74 (15), 130.24 (6,23), 129.23 (12), 126.44 (14), 124.80 (13), 124.23 (16),

,112.85 (5,22), 109.89 (18,7), 102.52 (9,20) , 83.31 (10)]

The kinetic study of acrylation reaction was conducted in solvent CD2Cl2. Hence,

1 H-NMR of the starting material FL was checked in CD2Cl2 using the NMR-750 MHz instrument as shown in Figure 4.4 for the precise comparison of chemical shifts with the product and to be able to monitor the progress of acrylation. FL is soluble in CD2Cl2 in the

99 1 presence of TEA. Hence, the H-NMR spectrum of FL with TEA in CD2Cl2 (Figure 4.4)

85 was first obtained at room temperature (25 °C). Figure 4.4B shows that signal (13) due to the aromatic proton of FL is not an exact doublet due to the presence of minor amount of tautomer quinoid/zwitterionic form detected by the 750 MHz instrument. Signals corresponding to the methylene and methyl protons of TEA can be seen at 2.60 ppm (b) and 1.08 ppm (a) (Figure 4.4A). To check the effect of temperature on splitting of signals, the 1H-NMR of the FL-TEA was also checked at 0 °C, -5 °C and -20 °C. Identical spectra of FL-TEA were obtained at all temperatures. From the 1H-NMR spectrum in Figure 4.4A, the lactone form of FL is found to be a dominant in CD2Cl2 and TEA.

1 Figure 4.4 (A) H-NMR of FL-TEA (1: 2.20) in CD2Cl2 (B) Inset is the enlarged resonance of proton (13) of FL.

1 [FL-TEA (750 MHz, H-NMR, CD2Cl2) : δ 8.09 ((13), m, 1H), δ 7.57-7.53 ((15,16), m,

2H), δ 7.17 ((14), m , 1H), δ 6.92 ((6,23), J= 9.17 Hz, m, 2H), 6.56 ((9,20), d, J= 2.20

Hz, 2H), δ 6.50 ((22,5), d , J= 2.20 Hz, 2H), δ 2.61-2.57 ((b), m , 6H), δ 1.01- 0.98 ((a), t

, J= 7.34, 7.15 Hz, 9H)]

After the 1H-NMR analysis of the starting materials, a kinetic study was conducted by preparing a reaction mixture in an NMR tube at -75°C in an acetone-dry ice bath. After the addition of a drop of AcrCl (2.20 equivalents, 0.4164 mol/L concentration), immediate 87 color change was observed from dark red to yellow-orange at -75°C. The NMR tube was then immediately placed inside the NMR cavity which was maintained at -20°C. The temperature inside the instrument was gradually increased and 1H-NMR spectra were recorded with 3 second intervals and at temperatures -20, -15, -10, -5, 0, 30℃. Identical spectra were obtained at all temperatures. Figure 4.5A shows 1H-NMR spectrum of the reaction mixture at 0°C. Resonance (1”’) is a newly generated signal after the reaction of

FL with AcrCl. It is observed that the resonance due to proton (1’”) moved from 6.20 ppm

(1’) (Figure 4.1) to 6.05 ppm (Figure 4.5B) when it is compared to the 1H-NMR spectrum of AcrCl (Figure 4.1). The integral ratio of the newly generated methylene proton (1’”) to aromatic reference proton (13) is (1:2) which confirms conversion of FL to FL-DA. Signals

(1”) and (2) are not integrated with respect to (13) as the overlap of the resonance takes place at 6.64 and 6.42-6.29 due to excess AcrCl present in the sample (Figure 4.1, Figure

4.5).

1 Figure 4.5 (A). H-NMR of product FL-DA in CD2Cl2 (B) Inset is the enlarged resonances of the product.

From the 1H-NMR analysis, it can be concluded that the reaction of FL with AcrCl is extremely fast and it starts immediately as soon as the drop of AcrCl is added.

Considering the time to put the sample inside the instrument (10 seconds) and total acquisition time (3 seconds), the reaction takes approximately 13 seconds even at low temperatures such as -75°C. The immediate color change after the addition of a drop of

AcrCl also supports the conclusion that the acrylation reaction of FL with AcrCl is an extremely fast reaction even at low temperatures such as -75°C.

4.1.2. Synthesis of fluorescein o-acrylate (FL-A)

Based on the kinetic studies, FL-A was attempted to be synthesized using a slight excess over one molar equivalent of AcrCl to FL. AcrCl (1.09 equivalents, 0.4545 mol/L) was added dropwise to the homogenous solution of FL and TEA (3.49 equivalents, 1.4506 mol/L) in dichloromethane (DCM) and the temperature was maintained between -5 to 0°C during the addition. The product was recovered by precipitation in n-hexane.

The product was analyzed by thin layer chromatography (TLC) with the solvent system of 80 v% diethyl ether and 20 v% n-hexane. (Figure 4.6) Pure commercial FL-A

(Sigma-Aldrich), the starting material FL (Sigma-Aldrich), FL-DA (synthesized in the lab) and the newly synthesized FL-A product were spotted on a TLC plate. Distinct yellow spots of each compound were observed on the TLC plate under UV light at 365 nm. As the order of polarity for the three compounds is FL>FL-A>FL-DA, the distance travelled by the three compounds on the TLC plate by elution is FL-DA>FL-A>FL. The spot of TEA hydrochloride salt was not visible under UV light at 254 nm and 365 nm wavelength

(Figure 4.6).

FL-DA FL-DA FL-A (Sigma-Aldrich) FL-A product FL Unreacted FL

Figure 4.6 TLC of the FL-A product before column chromatography with 80 v% ether, 20 v% hexane solvent system. Spots from left to right: FL, FL-A (Sigma-Aldrich), FL-DA (synthesized in the lab), FL- A product (after precipitation and drying).

The product showed unreacted FL, FL-A and FL-DA. To remove the by-products such as TEA hydrochloride salt, FL-DA and unreacted FL, column chromatography was performed with a solvent system of 50 v% hexane and 50 v% diethyl ether solvent mixture.

The first fraction was FL-DA. The polarity of the solvent system was gradually increased to 80 v% diethyl ether and 20 v% hexane and fraction 2 of pure FL-A was obtained. 2.5 g

(0.006471 mol) of pure product with 52.49% yield was obtained after column chromatography. The TLC of the pure FL-A product is shown in Figure 4.7. It is seen that the distance travelled by the product spot is the same as that of FL-A from Sigma-Aldrich and no other spots were observed.

FL-DA FL-A (Sigma-Aldrich) Pure FL-A product FL

Figure 4.7 TLC with 80% ether, 20% hexane solvent system Spots from left to right: FL, FL-A (Sigma-Aldrich), FL-DA (synthesized in the lab), FL- A product (column fraction).

The pure FL-A product was further analyzed by 1H-NMR (Figure 4.8). The signals due to the methine (2) and methylene protons (1’’’, 1”) of the acrylate group appear at

1 6.35-6.47 ppm (-CH=) and 6.17, 6.52 ppm (=CH2). H-NMR spectrum of the pure FL-A product is identical to the 1H-NMR spectrum of commercial FL-A from Sigma-Aldrich with methine (2) and methylene protons (1’’’, 1”) at 6.35-6.47 ppm (-CH=) and 6.17, 6.52 ppm (=CH2) (Figure 4.8B, 4.8C). The integration ratio of the reference aromatic proton

(13) to the newly appeared methine and methylene protons (2,1”’,1”) is (1:1:1:1) confirms the structure of the product. No signals corresponding to the protons of TEA hydrochloride salt are present at δ 3.08, 1.22 ppm. (Figure 4.8A)

Figure 4.8 (A) 1H-NMR of pure FL-A in DMSO d6 (B) Enlarged resonances of the product (C) Enlarged resonances of 1H-NMR of FL-A from Sigma-Aldrich.

[FL-A (300 MHz, 1H-NMR, DMSO d6): δ 8.08- 8.0 ((13),d, J=7.32Hz,1H), δ7.88-7.71

((15,16), m,2H), δ 7.39- 7.33 ((20),d, J=7.61Hz,1H), δ7.32-7.30 ((14),d, J= 2.34Hz, 1H),

δ7.02- 6.94 ((9), dd, J=8.78Hz, 8.49 Hz, 1H), δ6.89-6.81 ((22),d, J=8.78 Hz, 1H) , δ 6.77-

6.71 ((5),s,1H), δ 6.66 -6.58 ((23,6), s, 2H),δ 6.52 ((1”), d, J=1.46Hz,1H), δ 6.48-6.35

((2),m,1H), δ6.17 ((1”’),dd, J=10.25Hz, 10.25Hz,1H)]

13C-NMR analysis of the product (Figure 4.9) shows that the signals corresponding to the carbons of the acrylate group appear at 135.6 (1), 130.13 ppm (2), 163.53 (3), 150.69

(4) which further confirmed the structure of the product.

Figure 4.9 13C-NMR of FL-A product after column purification in DMSO d6.

13 [ FL-A (500 MHz, C-NMR, CDCl3): δ 168.32 (11), 163.53 (3), 159.53 (21), 152.08 (19),

151.37 (8), 151 (17), 150.69 (4), 135.6 (1), 133.92 (15), 130.13 (2), 128.86 (12), 127.15

(6), 125.62 (14), 124.61 (13), 123.86 (23), 117.82 (16), 116.58 (5), 112.91 (9), 110.13 (7),

108.92 (22), 102.08 (18,20), 81.79 (10)]

2.5 g of pure FL-A was obtained with an experimental yield of 52.49%. As discussed in the background, Wang B. et al. reported the FL-A synthesis by reacting FL with 1.22 equivalent of AcrCl at 0℃ in presence of 3.97 equivalent of TEA and stirring the reaction mixture at room temperature overnight. The FL-A product was purified by

95 column chromatography with CHCl3/ethanol (30/1) solvent mixture and 60% yield was achieved.57 Wang H. et al. synthesized FL-A (with 2 equivalents of TEA, 1.5 equivalents of AcrCl) with yield of 78%.56

4.1.3. Synthesis of fluorescein o,o’-diacrylate (FL-DA)

Based on the kinetic study, we modified the reported synthesis of FL-DA.56,57,80-82

Our improved procedure takes 13 seconds to complete after the dropwise addition of AcrCl

(4.01 equivalents, 0.7080 mol/L) to the reaction mixture while the reported reaction time with FL isothiocyanate (FITC) is 2 hours.80 Our method avoids tedious column chromatography procedure reported in earlier studies. The literature reports column

56 chromatography with various solvent systems: CH2Cl2:CH3OH(100:0.2),

82 80 CH2Cl2/CH3OH (100/1) and ethyl acetate/hexane (1:5) for FITC. We purified the product by stirring silica gel (silica gel classic column, 63-200um) in the THF solution of the product for 30 minutes, followed by filtration and washing the silica gel with THF. Our synthesis is faster, easier and 0.9 g (0.002045 mol), 63.9 % light yellow colored pure FL-

DA was obtained. The literature reports yield of 70% with FITC,80 75% with FL quinoid,81 and 45% yield with FL lactone.82

Figure 4.10 shows the 1H-NMR spectrum of the TEA hydrochloride salt which was filtered out from the reaction mixture. Methyl and methylene proton signals (a) and (b) appeared at δ 1.22 - 1.15 ppm and δ 3.08-2.97 ppm in DMSO d6. The ratio of the integrals of proton signals (a) : (b) is 9:6 as expected from the structure.

Figure 4.10 1H-NMR of TEA hydrochloride salt in DMSO d6.

[TEA hydrochloride salt (300 MHz, 1H NMR, DMSO d6): δ 3.08 ((b), m, 6H), δ 1.22 ((a), t, J= 7.32 Hz, 9H)]

Figure 4.11A shows the 1H-NMR spectrum of FL-DA in DMSO d6 after filtering.

There are still strong TEA salt signals present. Methyl and methylene proton signals (a) and (b) were still present at δ 1.22 - 1.15 ppm (a) and δ 3.08-2.97 ppm (b) after filtration.

1 Figure 4.11B shows the H-NMR spectrum of FL-DA product in CDCl3 after silica gel purification. No signals appeared at δ 1.22-1.15 ppm and δ 3.08-2.97 ppm corresponding to the TEA hydrochloride salt. Signals corresponding to the methylene and methine protons of the acrylate functionality were observed at 6.70 ppm (1), 6.30 ppm (2) and 6.10 ppm

(1’). The ratios of the integrals of the methylene and methine protons (1), (2), (1’) to that of the aromatic proton of FL (13) is (1:2:2:2) (Figure 4.11C), verifying the structure of the

FL-DA

Figure 4.11 1H NMR of FL-DA (A) Before silica gel filtration, (B) After silica gel filtration (C) Inset is the enlarged spectrum of 5.8 ppm to 8.5 ppm.

1 [FL-DA (500 MHz, H NMR, CDCl3): δ 8.07 ((13), dt, J =0.88,7.61 Hz, 1 ,H), δ 7.75-

7.68 ((15, 16), m, 2H), δ 7.25 ((14), d, J =1.17 Hz, 1H), δ 7.20 ((9,20), s, 2H), δ 6.85

((22, 5, 23, 6), s, 4 H), δ 6.70 ((1”), m, 2H), δ 6.30-6.25 ((2), m, 2H), δ 6.10 ((1’”), m, 2

H), δ 3.08-2.97 ((b), m, 6H), δ 1.22 - 1.15 ((a), t, J= 7.32 Hz, 9H)].

The 13C-NMR of the FL-DA is shown in Figure 4.12. The peaks corresponding to the acrylate carbons are observed at 133.47 ppm (2) and 127.42 ppm (1). No carbon signals corresponding to FL-A and TEA hydrochloride salt were observed.

13 Figure 4.12 C NMR of FL-DA in CDCl3.

13 [FL-DA (500 MHz, C NMR, CDCl3): δ 168.97 (11), 163.68 (3), 152.70 (17), 151.87

(8,19), 151.40 (4,21), 135.27 (15), 133.47 (2), 130.05 (12), 128.74 (23,6), 127.42 (1), 126

(14), 125.94 (13), 125.12 (16), 123.97 (22,5), 117.57 (18), 116.42 (7), 110.18 (20,9), 81.61

(10)]

The High Resolution Mass Spectrometry (HRMS) Electrospray ionization (ESI) in

Figure 4.13A and 4.13B confirmed the formation of FL-DA with peaks at m/z = 441.0969,

100

463.07908 and 479.05314. The corresponding calculated monoisotopic masses of the

+ peaks are 441.0969 (error 75 ppb) [440.09 (C26H16O7, FL-DA) + 1.01 (H )], 463.07908

+ [440.09 (C26H16O7, FL-DA) + 22.99 (Na) ] and 479.05314 [440.09 (C26H16O7, FL-DA) +

38.96 (K)+]. Peaks corresponding to the mono-functional FL (m/z = 386.07900) are not detected in the sample (100% conversion) as seen in Figure 4.13A. In Figure 4.13, ‘M’ represents the desired product FLDA.

(A)

101

(B)

Figure 4.13 (A) ESI-Mass spectrum of FL-DA. (B) HRMS ESI-Mass spectrum of FL- DA.

In conclusion, FL was successfully acrylated using AcrCl in presence of TEA with precision synthesis of FL-DA (yield 63.9%). Synthetic procedure is improved than the previously reported literature56,57,80-82 as explained before. Kinetic study of FL-DA synthesis using NMR-750MHz instrument demonstrated the formation of FL-DA in 13 seconds. Purity of FL-DA product was also confirmed by ESI mass spectrometry.

Similarly, pure FL-A was synthesized with yield of 52.49% and structure of the product was confirmed by 1H-NMR and 13C-NMR. FL-DA and FL-A are important precursors in

102 the development of polymeric conjugate as the acrylate moiety allows enzyme-catalyzed

Michael addition of thiol and amine.

103

4.2. Synthesis of two-functional folate-targeted fluorescein (FL)-based

diagnostic nanodevices

4.2.1. Objectives

As discussed in the Background section, four synthetic strategies were developed to synthesize two-functional folate-targeted FL-poly(ethylene glycol) (PEG) conjugates.

Figure 4.14 shows all four strategies. Strategy 1 involves fluorescein o,o’-diacrylate (FL-

DA) as a starting material to synthesize compound FA-FL-NH-PEG-NH-FL-FA. In strategy 2 and 3, FL-DA was replaced by fluorescein o-acrylate (FL-A) to improve the selectivity of the Michael addition between acrylate moiety and amine/thiol of PEG. FL-A was initially reacted with PEG-diamine (H2N-PEG-NH2) and later with tetraethylene glycol dithiol (HS-TEG-SH) and PEG-dithiol (HS-PEG-SH) using enzyme Candida antarctica lipase B (CALB). Strategy 4 involves lithiation of the FL followed by the addition of brominated PEG to synthesize the conjugate. Targeting agent thiol- functionalized folic acid (FA-SH) was synthesized as reported by Puskas et al.15 by lithiation of FA at γ-carboxylic acid group, followed by the addition of bromine- functionalized disulfide compound, forming FA-S-S-FA. Disulfide bond of FA-S-S-FA was then reduced using dithiothreitol (DTT) to form the FA-SH. Michael addition between acrylate moiety and FA-SH resulted in the two-functional folate-targeted FL-based diagnostic nanodevices.

104

105

106

Figure 4.14 Synthetic strategies for the two-functional FA-targeted FL-based diagnostic nanodevices.

4.2.2. Syntheses of the Cleveland Clinic trial compounds FA-FL-NH-PEG-NH-

FL-FA and FA-FL-FA

Compound FA-FL-NH-PEG-NH-FL-FA with H2N-PEG-NH2 (Mn=2,000 g/mol, Đ

≤1.2) backbone and two FA-SH targeting agents and compound FA-FL-FA without the

PEG linker (Scheme 4.3) were synthesized using FL-DA which was synthesized by me as discussed in Section 4.1. With compound FA-FL-NH-PEG-NH-FL-FA, excellent in vitro

107 and in vivo results were obtained by the Cleveland Clinic while compound FA-FL-FA was slightly cytotoxic and had poor water solubility as discussed in the Background section.

FA-FL-FA synthesis was later repeated by me for practice using FL-DA. CALB- catalyzed Michael addition of FA-SH (2 equivalent) to FL-DA (1 equivalent) was carried out at 50℃ in solvent DMSO d6. (Scheme 4.3)

Scheme 4.3 Synthetic scheme for compound FA-FL-FA

1H-NMR analysis of the raw sample of compound FA-FL-FA (before purification)

(Figure 4.15) showed that the methylene and methine proton signals corresponding to acrylate functionality of FL-DA which were present at 6.70 ppm (1”), 6.30 ppm (2) and

6.10 ppm (1”’) (Figure 4.11, Figure 4.15), reduced to negligible and demonstrated the formation of the product. New signals due to the successful Michael addition appeared between 2.6-3.00 ppm. However, further purification of the product was not conducted.

108

Figure 4.15 1H NMR spectrum of compound FA-FL-FA in DMSO d6. 109

4.2.3. Strategy 2: 2 FL-A + H2N-PEG-NH2 + 2 AcrCl + 2 FA-SH

PEG-based compound FA-FL-NH-PEG-NH-FL-FA showed excellent in vitro and in vivo results in terms of solubility, specificity for FR, non-toxicity and bioavailability.

Hence, PEG was further used to develop a folate-targeted polymeric conjugates. However, to improve the selectivity of Michael addition of H2N-PEG-NH2 and to avoid possible side products, FL-A was used to synthesize the conjugates instead of FL-DA.

First, commercial H2N-PEG-NH2 (Mn= 2000 g/mol, Đ ≤1.2) was reacted with FL-

A by CALB-catalyzed Michael addition (Scheme 4.4) to obtain FL-NH-PEG-NH-FL and acrylation was conducted using acryloyl chloride (AcrCl) for the attachment of FA-SH.

Scheme 4.4 Synthesis scheme of FL-NH-PEG-NH-FL

110

4.2.3.1. Synthesis of FL-NH-PEG2000 -NH-FL

CALB-catalyzed Michael addition of commercial H2N-PEG-NH2 (Mn= 2200 g/mol, Đ ≤1.2) and FL-A was conducted by me using CALB catalysis. After purification and drying, 0.9 g (0.3247 mmol), 86.58% of the product was obtained.

1 Figure 4.16 shows the H-NMR spectrum of the commercial H2N-PEG-NH2

(Mn=2,000 g/mol, Đ ≤1.2). The integral ratio of resonance (b+c+d+x) to the proton (y) is

186.61:4 which confirms the structure of H2N-PEG-NH2.

1 Figure 4.16 H-NMR of H2N-PEG-NH2 (Sigma-Aldrich, Mn=2,000 g/mol, Đ ≤1.2) in CDCl3

111

1 [H2N-PEG-NH2 (300 MHz, H-NMR, CDCl3): δ 3.64 ppm ((b, c, d, x), 186H), δ7.88-7.71

((y), 4H)]

The 1H-NMR spectrum of the purified product is shown in Figure 4.17. Methine and methylene proton signals due to the acrylate groups which were present at 6.52 ppm

(1”), 6.48-6.35 ppm (2), 6.17 ppm (1”’) (Figure 4.8), disappeared after the reaction. New signals (1', 2', y) appeared at 3.01-2.47 ppm after the successful Michael addition (Figure

4.17). The integral ratio of the newly generated protons (1'), (2’), (y) to the aromatic proton

(13) is 4:4:4:2, which confirms the structure of the product and demonstrates successful

Michael addition.

112

1 Figure 4.17 H-NMR of FL-NH-PEG2000-NH-FL in DMSO d6

1 [FL-NH-PEG2000-NH-FL (300 MHz, H-NMR, DMSO d6): δ 8.04- 7.91 ppm ((13), 2H),

δ 7.83- 7.63 ppm ((15,16), 4H), δ 7.40 -7.20 ((14,6), 4H), δ 6.92- 6.61 ppm ((23,5,9), 6H),

δ 6.59- 6.49 ppm ((22,20), 4H), δ 3.55 ((x, b, c, d), 196H), δ 3.01-2.47 ((1',2',y), 12H)]

Figure 4.18 shows the 13C-NMR spectrum of the product. No signals corresponding to the carbons of the acrylate group were observed at 133.47 and 127.42 ppm. New signals

113 appeared at 45.83 ppm (1'), 34.09 ppm (2'), 76 ppm (X), 45.88 ppm (Y) after the reaction.

13 Figure 4.18 C-NMR of FL-NH-PEG2000-NH-FL in DMSO-d6

13 [FL-NH-PEG2000-NH-FL (300 MHz, C-NMR, DMSO d6) : δ 170.48 ppm (3), δ 168.64

(11), δ 159.48 (21), δ 152.35 (4), δ 152.30 (8), δ 151.84 (17), δ 151.08 (19), δ 135.55 (15),

δ 129.95 (6), δ 128.93 (23), δ 126 (12), δ 125.75 (13), δ 124.73 (14), δ 123.96 (16), δ

117.98 (5), δ 116.5 (9), δ 112.52 (7), δ 109.58 (22), δ 109.07 (18), δ 102.4 (20), δ 81.96

(10), δ 76 (X), δ 69.01(C,D), δ 65 (B), δ 45.88 (Y), δ 45.83 (1'), δ 34.09 (2')]

114

4.2.3.2. Synthesis of Acryl-FL-NH-PEG2000 -NH-FL-Acryl

FL-NH-PEG2000-NH-FL was acrylated in 1.5 mL CD2Cl2 in the presence of TEA using AcrCl. The 1H-NMR spectrum of the product (Figure 4.19) shows that the signals due to the methine and methylene protons of the acrylate group appear at 6.46-6.32 ppm (-

CH=) (25) and 6.54, 6.17 ppm (=CH2) (26, 26’) after the reaction. The integral ratios of the protons (25), (26), (26’) to the aromatic proton of FL (13) is 2:2:2:2.

Signals due to the protons (1'), (2'), (y) which were present at 2.70 ppm and 2.87 ppm (Figure 4.17), disappeared completely after the acrylation. (1'), (2'), (y) are the protons adjacent to the secondary amine groups (N1 and N2) of the PEG. As the secondary amine acts as a nucleophile, AcrCl (4 equivalents) can react with N1 and N2 in the presence of

TEA (2.24 equivalents) (Figure 4.20).

Also, ‘HCl’ liberated as a by-product of the reaction can form a hydrochloride salt with ‘N1’ and ‘N2’. Structures of all the possible products are shown in Figure 4.20.

115

Figure 4.19 1H-NMR of the product of acrylation of FL-NH-PEG-NH-FL in DMSO d6

116

Figure 4.20 Possible products of the acrylation of FL-NH-PEG-NH-FL

To avoid the side products during acrylation and to achieve the selectivity, HS-

PEG-SH was used to develop the conjugate instead of H2N-PEG-NH2 as discussed in the following sections.

4.2.4. Strategy 3: 2 FL-A + HS-(TEG or PEG)-SH + 2 AcrCl + 2 FA-SH

First, HS-TEG-SH, HS-PEG1000-SH, HS-PEG2000-SH and HS-dPEG899-SH were synthesized as reported22 followed by the Michael addition to FL-A. Acrylation was then conducted using AcrCl in the presence of triethylamine (TEA) (Scheme 4.5) for the attachment of targeting agent FA-SH by the CALB-catalyzed Michael addition.

117

Scheme 4.5 Synthesis scheme of Acryl-FL-S-PEG-S-FL-Acryl

118

4.2.4.1. Synthesis of FL-S-TEG-S-FL

Figure 4.21 shows the 1H-NMR spectrum of HS-TEG-SH, which was synthesized by me as reported by the transesterification between methyl 3-mercaptopropionate (MP-

SH) and TEG.22 Methylene protons adjacent to the thiol group are present at 2.75 ppm (h’),

2.64 ppm (g’). The ratio of the integrals of proton signals (h’ + g’) to (b’) (methylene proton next to carbonyl) is (8.47: 3.94). The triplet due to (i) is present at 1.66 ppm with the ratio of integrals (1.90: 3.94) with (b’). This demonstrates the formation of HS-TEG-SH.

1 Figure 4.21 H-NMR of HS-TEG-SH in CDCl3

1 [HS-TEG-SH (300 MHz, H-NMR, CDCl3): δ 4.23 ppm ((b’), 4H), δ 3.68 ppm ((c’), 4H),

δ 3.61 ((d), 8H), δ 2.75 ppm ((h’), 4H), 2.64 ppm ((g’), 4H), δ 1.66 ppm ((i), 2H)] 119

HS-TEG-SH was reacted with FL-A in DMSO-d6 at 52℃ using CALB. To check the progress of the reaction, 1-minute after the addition of CALB, a sample was taken out for 1H-NMR at 52℃. 1H-NMR analysis shows that after 1-minute reaction time, the methylene protons next to the thiol groups in HS-TEG-SH (h’ and g’, see Figure 4.21) shifted from 2.75 ppm (h’), 2.64 ppm (g’) to δ = 2.88 ppm (h”) and 2.71-2.64 ppm (g”), overlapping with the new conjugation protons of (1’) and (2’). The presence of the desired product was detected after 1-minute by 1H-NMR analysis in DMSO d6 (Figure 4.22A).

Methine and methylene proton signals due to the acrylate groups of FL-A at 6.52 ppm (1”),

6.48-6.35 ppm (2), 6.17 ppm (1”’) (Figure 4.8) disappeared. New proton signals (1' and 2') appeared at δ = 2.80 ppm and 2.71-2.64 ppm as shown in Figure 4.22A. Figures 4.22B

1 shows the H-NMR spectrum of the purified product in CDCl3. The relative integrals of proton signals (13): (1'): (2') are in the ratio 2:4:4 (1’ and 2’ overlap with resonance h”) demonstrating the formation of di-substituted product of the Michael addition in one minute.

120

121

1 1 Figure 4.22 A. H-NMR of FL-S-TEG-S-FL in CDCl3 (1-minute raw sample), B. H- NMR of FL-S-TEG-S-FL in CDCl3 after workup.

1 [FL-S-TEG-S-FL (300 MHz, H-NMR, CDCl3): δ8.02- 8.00 ppm ((13), 2H), δ7.71-7.58

((15,16), 4H), δ7.21 -7.11 ((14), 2H), δ7.10 -7.05 ((6), 2H), δ6.82- 6.72 ((23,5,9), 6H), δ

6.66 -6.50 ((22,20), 4H), δ4.32-4.19 ((b’), 4H), δ3.73-3.68 ((c’), 4H), δ3.67-3.62 ((d), 8H),

δ2.95-2.80 ((1',2',h”), 12H), δ2.71-2.64 ((g”), 4H)]

The successful Michael addition is also confirmed by 13C-NMR analysis in DMSO d6 as seen in Figure 4.23. New carbon signals appeared at 34.40 ppm (2') and δ 26.10 ppm

(1') after the reaction. No signals corresponding to the acrylate functionality were observed at 133.47 ppm and 127.42 ppm.

122

Figure 4.23 13C NMR of FL-S-TEG-S-FL in DMSO-d6 after workup.

[FL-S-TEG-S-FL (500 MHz, 13C-NMR, DMSO d6): δ171.54 ppm (3), δ 170.10 (11), δ

168.66 (F’), δ 159.67 (12), δ 152.26 (4), δ 151.75(8), δ 151.54 (17), δ 151.14 (19), δ

135.87 (15), δ 130.38 (21), δ 129.03 (6),δ 125.74 (23), δ 124.85 (13), δ 124.34 (14), δ

123.97 (16), δ 118.05 (5), δ 116.65 (9), δ 113.15 (7), δ 110.28 (22), δ 109.13 (18), δ 102.33

(20), δ 82.04 (10), δ 69.84 (D), δ 68 (C’), δ 64 (B), δ 34.82(G”), δ 34.40 (2'),δ 26.21(H”),

δ 26.10 (1')]

123

The product was further analyzed by Matrix-Assisted Laser Desorption Ionization,

Time of Flight (MALDI-ToF) mass spectrometry as shown in Figure 4.24. A single small distribution is obtained, and formation of the desired product is confirmed with peaks at m/z = 1165.26, 1209.12, 1253.270.

The peak at m/z 1165.26 corresponds to the Na complex of the desired product.

The calculated monoisotopic mass of the peak at m/z = 1165.26 is [m/z =3 ×44.03 (C2H4O repeat unit) + 491.08 (C26H19O8S- end group) + 519.11 (C28H23O8S- end group) + 22.99

(Na+)] is 1165.27.

The peak at m/z 1209.12 corresponds to the Na complex of the 4-mer fraction of the desired product. The calculated monoisotopic mass of this peak is [m/z =3 ×44.03

(C2H4O repeat unit) + 491.08 (C26H19O8S- end group) + 519.11 (C28H23O8S- end group) +

22.99 (Na+)] 1209.3. The peak at m/z 1253.270 corresponds to the Na complex of the 5- mer fraction of the desired product. The calculated monoisotopic mass of this peak is [m/z

=5 ×44.03 (C2H4O repeat unit) + 491.08 (C26H19O8S- end group) + 519.11 (C28H23O8S- end group) + 22.99 (Na+)] 1253.333.

The small peaks at m/z 779.511 from the high-resolution image of the MALDI- spectrum corresponds to the traces of ‘Na’ complex of mono-substituted FL which could not be detected by 1H-NMR. The calculated monoisotopic mass of this peak is [m/z =3

×44.03 (C2H4O repeat unit) + 105 (C3H5O2S- end group) + 519.11 (C28H23O8S- end group)

+ 22.99 (Na+)] 779.37.

The intensity of the desired product FL-S-TEG-S-FL is found to be 111504.87 while that of the mono-substituted FL is 3529.52. Hence based on the absolute intensity,

124 approximately 3% mono-substituted FL is present in the desired product. No traces of the unreacted HS-TEG-SH were found. The other peaks at m/z 691.190, 751.458, 795.192,

858.578, 907.364, 962.619, 1001.606, 1533.374 are present due to the matrix.

125

Figure 4.24 (1) MALDI- ToF analysis of FL-S-TEG-S-FL (2) Enlarged mass spectrum of the side-product FL-S-TEG-SH.

126

4.2.4.2. Synthesis of FL-S-PEG-S-FL.

The reaction of HS-TEG-SH and FL-A completed in 1 minute by CALB catalysis.

In order to investigate, the effect of CALB catalysis on the Michael addition of HS-PEG-

SH to FL-A was investigated by conducting the reaction between HS-PEG1000-SH and FL-

A was investigated without and with CALB catalysis at 52℃.

4.2.4.2.1. Synthesis of FL-S-PEG1000-S-FL

1 Figure 4.25 shows the H-NMR of starting material HS-PEG1000-SH which was synthesized by my colleague. Methylene protons adjacent to the thiol group are present at

2.75 ppm (h’), 2.64 ppm (g’). The internal protons of PEG1000 (c’+ d) were set at 84. The ratio of the integrals of proton signals (h’ + g’) to (b’) (methylene proton next to carbonyl) is (8.47: 3.37). The triplet due to (i) is present at1.66 ppm with ratio of integrals (1.90:

3.37) with (b’). This demonstrates the structure of HS-PEG-SH. The excess MPSH was later removed under vacuum at 60℃ overnight.

127

1 Figure 4.25 H-NMR of HS-TEG-SH in CDCl3.

1 [SH-PEG1000-SH (500 MHz, H-NMR, CDCl3)]: δ 4.21 ppm ((b’), 4H), δ3.61 ppm ((c’, d),

84H), 2.72 ((h’), 4H), δ 2.63 ((g’), 4H), δ1.65 ((i), 2H)]

o HS-PEG1000-SH was reacted with FL-A in DMSO-d6 at 52 C without adding

CALB. The 1H-NMR spectrum in the Figure 4.26 shows that after 1-minute reaction time, methylene protons (h’, g’) next to the SH groups in HS-PEG1000-SH moved from 2.72 ppm

(h’) and 2.63 ppm (g’) and overlapped at 2.87-2.66 ppm (h”, g”). Protons (h”, g”) overlap with the new conjugation protons of (1’) and (2’) at δ = 2.88 ppm and 2.80 ppm but the

128 methine and methylene proton signals of the acrylate groups (6.52 ppm (1”), 6.48-6.35 ppm (2), 6.17 ppm (1”’) (Figure 4.8) were still visible and reduced from 2 to 0.82, 0.64 and

0.82 (Figure 4.26). The integration ratio of protons (1’+2’+h”+g”) to proton (13) and (b’)

(the methylene protons next to the ‘O’ in the HS-PEG-SH) to (13) indicates the progress of the reaction. After 5 minutes reaction time, the ratio started to decrease, and the proton signals due to acrylate started to increase at 6.52 ppm, 6.48-6.35 ppm, 6.17 ppm indicating reversion of the reaction. After 18 hours fresh HS-PEG-SH was added, but no change in the extent of the reaction was observed. Subsequently, CALB (2.78 × 10-4 mol/L) was added to the reaction and a sample was taken after 2 minutes reaction time. Only traces of the acrylate protons were observed at 6.52 ppm, 6.48-6.35 ppm, 6.17 ppm (Figure 4.26).

129

Figure 4.26 Kinetic study of Michael addition between HS-PEG1000-SH and FL-A in DMSO d6. 130

Subsequently, the reaction was repeated in the presence of CALB from the beginning. HS-PEG1000-SH was reacted with FL-A in DMSO d6 using CALB. The workup was performed 2 minutes after the addition of CALB at 52℃. The 1H-NMR spectrum of the product after workup and drying is shown in the Figure 4.27. 1H-NMR spectrum does not show any traces of the signals of the methine and methylene protons of the FL-A at

6.52 ppm (1”), 6.48-6.35 ppm (2), 6.17 ppm (1”’) (Figure 4.8) but new proton signals (1' and 2') are present at 2.88 ppm and 2.80 ppm (Figure 4.26). The relative integrals of (13):

(1'): (2') are in the ratio (2:4:4) (1’ and 2’ overlap with h” and g”) demonstrating the formation of di-substituted product of the Michael addition in two minutes.

131

1 Figure 4.27 H-NMR of FL-S-PEG1000-S-FL in DMSO d6 after workup.

1 [FL-S-PEG1000-S-FL (300 MHz, H-NMR, DMSO d6) : δ8.05-7.99 ppm ((13), 2H), δ7.84-

7.70 ((15,16), 4H), δ7.38 -7.27 ((14), 2H), δ7.24 -7.14 ((6), 2H), 6.94- 6.87 ((23), 2H),

δ6.86- 6.80 ((5), 2H), 6.72 ((9), 2H), 6.62-6.55 ((22,20), 4H), δ4.20-4.10 ((b’), 4H), δ3.61-

3.58 ((c’), 4H), δ3.54-3.45 ((d), 72H), δ2.87-2.66 ((1',2', h”, g”), 16H)]

The successful Michael addition is also confirmed by 13C-NMR analysis in DMSO d6 as shown in Figure 4.28. New carbon signals appeared at 34.17 ppm (2'), 25.98 ppm

132

(1') after the reaction. No signals corresponding to acrylate functionality were observed at

δ 133.47 ppm and 127.42 ppm.

13 Figure 4.28 CNMR of FL-S-PEG1000-S-FL in DMSO d6 after workup.

13 [FL-S-PEG1000-S-FL, (500 MHz, C-NMR, DMSO d6): δ171.24 ppm (3), δ 169.82 (11),

δ 168.40 (F’), δ 159.52 (12), δ 152.06 (4), δ 151.59 (8), δ 151.35 (17), δ 151 (19), δ 135.61

(15), δ 130.16 (21), δ 128.86 (6), δ 125.67 (23), δ 125.55 (13), δ 124.60 (14), δ 123.89

(16), δ 117.85 (5), δ 116.43 (9), δ 112.88 (7), δ 110.04 (22), δ 108.98 (18), δ 102.11 (20),

133

δ 81.75 (10), δ 69.56(D), δ 68.02 (C’), δ 63.29 (B’), δ 34.29 (G”), δ 34.17 (2'), δ 26 (H”),

δ 25.98 (1')]

4.2.4.2.2. Synthesis of FL-S-PEG2000-S-FL

Figure 4.29 shows the 1H-NMR spectrum of the starting material for the next reaction HS-PEG2000-SH, which was synthesized by my colleague. Methylene protons adjacent to thiol group are present at 2.75 ppm (h’), 2.64 ppm (g’). The internal protons of

PEG2000 (c+d) were set at 180. The ratio of the integrals of proton signals (h’ + g’) to (b’)

(methylene proton next to carbonyl) is (180: 3.87). The triplet due to (i) is present at 1.66 ppm with the ratio of integrals (1.91: 3.87) with (b’). This demonstrates the formation of

HS-PEG2000-SH.

134

13 Figure 4.29 CNMR of HS-PEG2000-SH in CDCl3.

1 [SH-PEG2000-SH (500 MHz, H-NMR, CDCl3)]: δ 4.21 ppm ((b’), 4H), δ3.61 ppm ((c’,d’),

180H), 2.72 ((h’), 4H), δ 2.63 ((g’), 4H), δ1.65 ((i), 2H)]

The reaction was repeated in the presence of CALB for PEG2000. HS-PEG2000-SH was reacted with FL-A in DMSO-d6 using CALB. The workup was performed 2 minutes after the addition of CALB at 52℃. The 1H-NMR spectrum of the product after workup and drying is shown in Figure 4.29.

Methylene protons (h’, g’) next to the SH groups in HS-PEG2000-SH moved from

2.72 ppm (h’) and 2.63 ppm (g’) and overlapped at 2.97-2.63 ppm (h”, g”,1’, 2’). The 1H-

NMR spectrum does not show any traces of the signals of the methine and methylene protons of the FL-A at 6.52 ppm (1”), 6.48-6.35 ppm (2), 6.17 ppm (1”’) ((Figure 4.8), but 135 the new proton signals (1' and 2') appeared at 2.88 ppm and 2.80 ppm (Figure 4.30). The relative integrals of (13): (1'): (2') are in the ratio 2:4:4 (1’ and 2’ overlap with h” and g”) demonstrating the formation of di-substituted product of Michael addition in two minutes.

1 Figure 4.30 H-NMR of FL-S-PEG2000-S-FL in CDCl3 after workup.

1 [FL-S-PEG2000-S-FL (300 MHz, H-NMR, CDCl3): δ8.08-7.99 ((13), 2H), δ7.74-7.57

((15,16), 4H), δ7.22 -7.14 ((14), 2H), δ7.09 -7.04 ((6), 2H), 6.88- 6.7 ((9,23,5), 6H), 6.68-

6.50 ((22,20), 4H), δ4.32-4.19 ((b’), 4H), δ3.72-3.68 ((c'), 4H), δ3.68-3.61 ((d), 176H),

δ2.97-2.63 ((1',2', h”, g”), 16H)]

136

The successful Michael addition is also confirmed by 13C-NMR spectrum in DMSO d6 as shown in Figure 4.31. New carbon signals appeared at 34.17 ppm (2') and 25.98 ppm

(1') after the reaction. No signals corresponding to acrylate functionality were observed at

133.47 ppm and 127.42 ppm.

13 Figure 4.31 C-NMR of FL-S-PEG2000-S-FL in DMSO d6 after workup.

13 [FL-S-PEG2000-S-FL (500 MHz, C-NMR, DMSO d6) : δ171.4 ppm (3), δ 169.02 (11), δ

168.40 (F’), δ 159.66 (12), δ 152.8 (4), δ 151.7 (8), δ 151.5 (17), δ 151.2 (19), δ 136.7

137

(15), δ 131.2 (21), δ 129.56 (6), δ 127.01 (23), δ 126.65 (13), δ 125.80 (14), δ 123.40 (16),

δ 118.68 (5), δ 117.83 (9), δ 112.02 (7), δ 110.14 (22), δ 109.88 (18), δ 103.31 (20), δ 82.55

(10), δ 70.24 (D), δ 68.4 (B’), δ 64.05 (C’), δ 34.46 (G”), δ 34.17 (2'), δ 26.35 (H”), δ

26.01 (1')]

In conclusion, we successfully PEGylated FL which can be further acrylated using

AcrCl for the attachment of FA-SH.

CALB-catalyzed Michael addition between HS-TEG-SH, HS-PEG-SH and FL-A was found to be extremely fast and completed in 1-minute and 2-minute respectively.

Reaction of HS-PEG-SH with FL-A without CALB catalysis did not go to completion even after overnight stirring at 52℃ but completed in 2 minutes after the addition of CALB as seen in the kinetic study.

Based on the mechanism of Michael addition discussed in the Background, we have our rendition of the mechanism of Michael addition between HS-TEG-SH and FL-A as illustrated in Figure 4.32.55,86 The top (dark shaded) portion of the enzyme which is a

“carbonyl pocket” consisting of Thr40 and Gln106 activates the carbonyl group of the FL-

A forming enzyme-substrate complex 1 (ESC-1). HS-TEG-SH which is a nucleophile then enters the cycle and forms the complex ESC-2 with oxyanion hole and His224 - Asp187 pair. At the last step, transfer of the proton catalyzed by His224, Asp187 to the α-carbon of FL-A leads to the formation of ESC-3 followed by the release of the product. It is a stepwise mechanism where one side of HS-TEG-SH reacts first followed by the second side and Serine (Ser105) residue of CALB is not involved in the catalytic activity for the

Michael addition reaction.55,86

138

Figure 4.32 Reaction mechanism of CALB-catalyzed Michael addition of HS-TEG-SH with FL-A.55, 86

Linkage of an imaging agent, which is “FL” in this case, to PEG is known as

“PEGylation” and enhances the retention time of the imaging agent in the blood and protects their degradation in cells and tissues.100 The CALB-catalyzed Michael addition of and HS-TEG-SH/HS-PEG-SH to FL-A was extremely fast and efficient with easier purification. Hence, can be called as a ‘click reaction.’ 139

4.2.4.3. Synthesis of Acryl-FL-S-TEG-S-FL-Acryl

Acrylation of all the conjugates was carried out for the attachment of targeting agent ‘FA-SH’ via Michael addition reaction. FL-S-TEG-S-FL was acrylated in CD2Cl2 in the presence of TEA with AcrCl and 0.555 g (0.444 mol), 72.43 % of the product was obtained.

The 1H-NMR spectrum in the Figure 4.33 shows that the resonance due to the aromatic protons (20, 22) moved from 6.66-6.50 ppm to 7.04-6.86 ppm (Figure 4.8) after acrylation, overlapping with signals (5,9). Methine and methylene proton signals due to the acrylate group appeared at 6.45-6.33 ppm (-CH=) (25) and 6.56, 6.17 ppm (=CH2) (26,

26’). The integral ratios of these protons [(25), (26), (26’)] to the aromatic proton of FL

(13) is 2:2:2:2, which demonstrates the formation of the product of acrylation. TEA hydrochloride salt by-product (0.0274 g), which was not removed shows resonances at 1.20 ppm and 3.06 ppm.

140

Figure 4.33 1H-NMR of Acryl-FL-S-TEG-S-FL-Acryl in DMSO d6.

[Acryl-FL-S-TEG-S-FL-Acryl (300 MHz, 1H-NMR, DMSO d6): δ 8.05 ppm ((13), 2H),

δ7.85-7.71 ((15,16), 4H), δ 7.45- 7.36 ((14, 23, 6), 6H), δ 7.04- 6.86 ((22,5,20,9), 8H), δ

6.56 ((26), 2H), δ 6.45-6.33 ((25), 2H), δ 6.17 ((26’), 2H), δ 4.12 ((b’), 4H), δ 3.57 ((c’),

4H), δ 3.48 ((d), 8H), δ 2.96-2.57 ((h”,1', 2', g”), 16H)]

Figure 4.34 shows the 13C-NMR spectrum of the product of the acrylation reaction.

The signals due to the FL-A ring carbons (22), and (20) moved from 110.28 ppm and

102.33 ppm (Figure 4.9) to 118.4 ppm and 110.63 ppm respectively, after the reaction. The

141 signals due to the carbons of the acrylate group appeared at 166.55 (24), 129.5 ppm (25) and 134.38 ppm (26) after the successful reaction.

Figure 4.34 13C-NMR of Acryl-FL-S-TEG-S-FL-Acryl in DMSO d6.

[Acryl-FL-S-TEG-S-FL-Acryl (500 MHz, 13C-NMR, DMSO d6): δ 172.4 ppm (3), δ

171.24 (11), δ 168.43(F’), δ 166.55 (24), δ 152.31 (19), δ 152.11 (8), δ 151.92 (17), δ

150.98(4), δ 150.96(21), δ 136.35 (15), δ 134.38 (26), δ 132.47 (23), δ 129.5 (25), δ 129

(6), δ 128 (12), 125.49 (13), δ 124.7(14), δ 124.4 (16), δ 118.4 (22), δ 118.4 (5), δ 116.6

(18), δ 116.36 (7), δ 110.7(9), δ 110.63 (20), δ 81.15 (10), δ 69.25 (D), δ 68.36 (C’), δ

63.59 (B’), δ 34.48 (G”), δ 34.4 (2'), δ 26.35 (1'), δ 26.21 (H”)]

142

4.2.4.4. Synthesis of Acryl-FL-S-PEG-S-FL-Acryl compounds

FL-S-PEG-S-FL compounds were acrylated to be able to attach the targeting agent

‘FA-SH’ to the conjugates via Michael addition.

4.2.4.4.1. Synthesis of Acryl-FL-S-dPEG899 -S-FL-Acryl.

A discrete PEG (dPEG)-based conjugate was synthesized to obtain a precise molecular weight and lower Đ. FL-S- dPEG899 -S-FL was synthesized by my colleague which I then acrylated to allow the attachment of 2FA-SH targeting agents. FL-S-dPEG-

S-FL was acrylated in CD2Cl2 in the presence of base TEA with AcrCl. The yield was

0.6559 g (0.33 mmol), 77.6%.

The 1H-NMR spectrum in Figure 4.35 shows that the signals due to the methine and methylene protons from the acrylate group appeared at 6.49-6.35 ppm (-CH=) (25) and

6.55, 6.18 ppm (=CH2) (26, 26’). The integral ratios of these protons [(25), (26), (26’)] to the aromatic proton of FL (13) is 2:2:2:2 which demonstrates the formation of the product of acrylation. TEA hydrochloride salt by-product (0.0429 g) which was not removed shows resonances at 1.20 ppm and 3.06 ppm.

143

1 Figure 4.35 H-NMR of Acryl-FL-S-dPEG899-S-FL-Acryl in DMSO d6

[Acryl-FL-S-dPEG-S-FL-Acryl (300 MHz, 1H-NMR, DMSO d6): δ 8.05 ((13), 2H), δ7.86-

7.72 ((15,16), 4H), δ 7.48- 7.21 ((14,23,6), 6H), δ 7.05-6.87 ((22,5,20,9), 8H), δ 6.55 ((27),

2H), δ 6.49-6.35 ((25), 2H), δ 6.18 ((26), 2H), δ 4.13 ((b’), 4H), δ 3.58 ((c’), 4H), δ 3.49

((d), 72H), δ 2.96-2.61 ((h”,1',2',g”), 16H) ]

144

4.2.4.4.2. Synthesis of Acryl-FL-S-PEG1000 -S-FL-Acryl

FL-S-PEG1000-S-FL was acrylated in CD2Cl2 in the presence of TEA with AcrCl.

The yield was 0.33 g (0.162 mmol), 62.2%.

1H-NMR spectrum in Figure 4.36 shows that the resonance due to the aromatic protons (20, 22) moved from 6.66 -6.50 ppm to 7.03-6.83 ppm (Figure 4.8) and overlapped with signals (5,9) after the acrylation. Signals due to the methine and methylene protons of the acrylate group appeared at 6.46-6.33 ppm (-CH=) (25) and 6.55, 6.17 ppm (=CH2) (26,

26’). The integral ratios of these protons [(25), (26), (26’)] to the aromatic proton of FL

(13) is 2:2:2:2 which demonstrates the formation of the product of acrylation. TEA hydrochloride salt by-product (0.0437 g) which was not removed shows resonances at 1.20 ppm and 3.06 ppm.

145

1 Figure 4.36 H-NMR of Acryl-FL-S-PEG1000-S-FL-Acryl in DMSO d6

1 [Acryl-FL-S-PEG1000-S-FL-Acryl (300 MHz, H-NMR, DMSO d6): δ 8.05 ppm ((13),

2H), δ7.86-7.70 ((15,16), 4H), δ 7.45- 7.20 ((14, 23, 6), 6H), δ 7.03-6.83 ((22,5,20,9), 8H),

δ 6.55 ((27), 2H), δ 6.46-6.33 ((25), 2H), δ 6.17 ((26), 2H), δ 4.13 ((b’), 4H), δ 3.58 ((c’),

4H), δ 3.59 ((d), 80H), δ 2.96-2.61 ((h”,1',2',g”), 16H) ]

Figure 4.37 shows the 13C-NMR spectrum of the product of the acrylation reaction.

The signals due to the FL-A ring carbons (22), and (20) moved from 110.28 ppm and

102.33 ppm (Figure 4.9) to 118.7 ppm and 110.33 ppm, respectively after the reaction.

146

Signals due to the carbons corresponding to the acrylate group appeared at 165.65

(24), δ 130.49 (25) and 134.18 (26) after the successful reaction. TEA hydrochloride salt by-product (0.0437 g) which was not removed, shows resonances at 1.20 ppm and 3.06 ppm.

13 Figure 4.37 C-NMR of Acryl-FL-S-PEG1000-S-FL-Acryl in DMSO d6.

13 [Acryl-FL-S-PEG1000-S-FL-Acryl (500 MHz, C-NMR, DMSO d6): δ 173.48 ppm (3), δ

171.24 (11), δ 168.88(F’), δ 165.65 (24), δ 152.21 (19), δ 152.13 (8), δ 151.92 (17), δ

150.88 (4), δ 149.9 (21), δ 136.22 (15), δ 134.18 (26), δ 132.37 (23), δ 130.49 (25), δ 147

128.34 (6), δ 127.5 (12), δ 126.4 (13), 126 (14), δ 124 (16), δ 118.7 (22), δ 118.6 (5), δ

117.4 (18), δ 116.9 (7), δ 111.7 (9), δ 110.33 (20), δ 80.2 (10), δ 68.7 (D), δ 67.99 (C’), δ

64.6 (B), δ 36.4 (G”), δ 36.2 (2'), δ 26.8 (1'), δ 26.2 (H”)]

4.2.4.4.3. Synthesis of Acryl-FL-S-PEG2000 -S-FL-Acryl

FL-S- PEG2000-S-FL was acrylated in CD2Cl2 in presence of TEA with AcrCl and

0.2605 g (0.08561 mmol), 62.46 % of the product was obtained.

The 1H-NMR spectrum in Figure 4.38 shows that the resonance due to the aromatic protons (20, 22) moved from 6.66 -6.50 ppm to 7.07-6.81 ppm (Figure 4.8) after acrylation and overlapped with signals (5,9). Resonances of the methine and methylene protons due to the acrylate group appeared at 6.45-6.33 ppm (-CH=) (25) and 6.56, 6.17 ppm (=CH2)

(26, 26’). The integral ratios of these protons [(25), (26), (26’)] to the aromatic proton of

FL (13) is 2:2:2:2, which demonstrates the formation of the product of acrylation. TEA hydrochloride salt by-product (0.0240 g) which was not removed shows resonances at 1.20 ppm and 3.06 ppm.

148

1 Figure 4.38 H-NMR of Acryl-FL-S-PEG2000-S-FL-Acryl in DMSO d6

1 [Acryl-FL-S-PEG2000-S-FL-Acryl (300 MHz, H-NMR, DMSO d6): δ 8.05 ((13), 2H),

δ7.86-7.70 ((15,16), 4H), δ 7.49- 7.23 ((14, 23, 6), 6H), δ 7.03-6.84 ((22,5,20,9), 8H), δ

6.55 ((27), 2H), δ 6.47-6.34 ((25), 2H), δ 6.18 ((26), 2H), δ 4.12 ((b’), 4H), δ 3.58 ((c’),

4H), δ 3.50 ((d), 176H), δ 2.93-2.61 ((h”,1',2',g”), 16H) ]

Figure 4.39 shows the 13C-NMR spectrum of the product of the acrylation reaction.

The signals due to the FL-A ring carbons (22), and (20) moved from 110.04 ppm and

102.11 ppm (Figure 4.9) to 112.5 ppm and 102.1 ppm respectively, after the reaction.

149

Signals due to the carbons corresponding to the acrylate group appeared at 161 ppm (24),

129.01 ppm (25) and 133 ppm (26).

13 Figure 4.39 C-NMR of Acryl-FL-S-PEG2000-S-FL-Acryl in DMSO d6

13 [Acryl-FL-S-PEG2000-S-FL-Acryl (500 MHz, C-NMR, DMSO d6): δ 172.4 ppm (3), δ

171.44 (11), δ 169.90 (F’), δ 161 (24), δ 152.51 (19), δ 152.23 (8), δ 152.11 (17), δ 151.88

(4), δ 150.9 (21), δ 137.10 (15), δ 133 (26), δ 129.01 (25), δ 128.70 (23), δ 127.9 (6), δ

127.6 (12), 126.2 (13), δ 118.7 (14), δ 117.7 (16), δ 112.5 (22), δ 110.23 (5), δ 110 (18), δ

150

108.2 (7), δ 102.4 (9), δ 102.1 (20), δ 82.2 (10), δ 70.33 (D), δ 64.88 (C’), δ 64.00 (B’), δ

34.21 (G”), δ 34.1 (2'), δ 26.45 (1'), δ 26.8 (H”)]

In conclusion, we successfully synthesized two-functional folate-targeted FL-based diagnostic conjugates using HS-TEG-SH and HS-PEG-SH. FL-TEG/PEG conjugates were then successfully acrylated using AcrCl which allow the attachment of FA-SH via CALB- catalyzed Michael addition. My colleague conducted the final Michael addition of FA-SH to the newly synthesized acrylated conjugates (synthesized by me as discussed) to develop the difunctional folate-targeted diagnostic agents with different molecular weights.

151

4.3. Strategy 4: PEGylation of fluorescin via lithium technology

As discussed in Section 4.1, strategy 4 was developed to obtain the PEGylated dihydro form of fluorescein (FL) which is known as fluorescin using lithium technology.

The PEGylated product will be further acrylated for the attachment of thiol-functionalized folic acid (FA-SH). First, an experimental trial was conducted to check the feasibility of the PEGylation of fluorescin using n-butyllithium (n-BuLi).

n-BuLi is a strong base with pKa 50101 while the ‘COOH’ group of fluorescin has pKa (3.88±0.36)102 and the phenolic ‘OH’ of fluorescin has pKa (6.4).102 ‘COOH’ being more acidic than phenolic ‘OH’, we expected the reaction of ‘COOH’ first with 1:1 molar equivalent of n-BuLi resulting in the formation of lithium salt at ‘COOH.’

Fluorescin which is a reduced nonfluorescent probe63 was bought from TCI

America to lithiate ‘COOH’ of fluorescin to attach the brominated PEG compound. (Figure

4.40)

Figure 4.40 Structure of fluorescin.

Figure 4.41 shows the 1H-NMR spectrum of fluorescin and Figure 4.42 shows the

1H-NMR of fluorescein (lactone dominant as discussed in Section 4.1) in DMSO d6. Proton

(10) appears at 6.13 ppm in Figure 4.41 which is absent in Figure 4.42 due to the closed lactone form. Comparison of Figure 4.41 and 4.42 shows that the resonance (13) of lactone 152

FL moved from 8.01 ppm to 7.72 ppm (13’), resonance (14) moved from 7.28 ppm to 6.89 ppm (14’) and resonances (15, 16) moved from 7.84 -7.69 ppm (15’) to 7.32, 7.20 ppm

(16’). The presence of the proton (10) in Figure 4.41 and differences in the chemical shifts of protons (13',14’,15’,16’) as compared to the Figure 4.42 confirm the structure of fluorescin.

Figure 4.41 (A) 1H-NMR of fluorescin in DMSO d6 (B) Inset is the enlarged resonances.

153

[fluorescin (300 MHz, 1H-NMR, DMSO d6): δ 7.72 ppm ((13’), 1H), δ 7.32 ((15’), 1H), δ

7.20 ((16’), 1H) δ 6.89 (14’), 1H), δ 6.80 ((23,6), 2H), δ 6.57- 6.46 ((20,9), 2H), δ 6.42

((22,5), 2H), δ 6.13 ((10), 1H)]

Figure 4.42 1H-NMR of FL in DMSO d6.

[FL (300 MHz, 1H-NMR, DMSO d6): δ 10.10 ppm ((OH), 2H), 8.01 ((13), 1H), δ 7.84-

7.69 ((15,16), 2H), δ 7.28 ((14), 1H), δ 6.70 ((6,23), 2H), δ 6.56 ((9,20,22,5), 4H)]

Scheme 4.6 shows the synthetic strategy for the PEGylation of fluorescin. First, brominated PEG (Br-PEG-OMe) was synthesized by Candida antarctica lipase B (CALB) catalyzed transesterification of ethyl 5-bromovalerate (EBV) with PEG methyl ether as

79 reported (PEG-OMe, Mn= 750, Đ= 1.459) (Scheme 4.6A). FL was then lithiated by n-

BuLi in anhydrous DMSO at 25℃ (Scheme 4.6B) and reacted with Br-PEG-OMe to obtain the PEGylated FL product. (Scheme 4.6C) 154

Scheme 4.6 Synthesis of PEGylated fluorescin by lithiation technology. (A) Synthesis of Br-PEG-OMe by CALB - catalyzed transesterification of EBV with PEG-OMe. (B) Lithiation using n-BuLi in anhydrous DMSO. (C) Attachment of Br-PEG-OMe to lithiated fluorescin.

4.3.1. Synthesis of Br-PEG-OMe

Figure 4.43 shows the 1H-NMR of the starting material PEG-OMe in deuterated

Chloroform (CDCl3). The integral ratio of signals due to the protons (b+c+d+e) to the methyl protons (a) is 64:3 which yields Mn = 736 g/mol, in excellent agreement with the nominal value. No other signals are present in the spectrum.

155

1 Figure 4.43 . H-NMR of PEG-OMe in CDCl3

1 [PEG-OMe (300 MHz, H-NMR, CDCl3): δ 3.68-3.59 ppm ((b, c, d, e), 64H), δ 3.36 ((a),

3H)]

Figure 4.44 shows the 1H-NMR of EBV, which is the other starting material, in

CDCl3. The integral ratio of protons (h, i, j, y, z) with respect to reference proton (g) is

2:2:2:3:2:2 confirms the structure of EBV.

156

1 Figure 4.44 H-NMR of EBV in CDCl3

1 [EBV (300 MHz, H-NMR, CDCl3): δ 4.15 ppm ((z), 2H), δ 3.4 ((j), 2H), δ 2.32 ((g), 2H),

δ 1.85 ((i), 2H) δ 1.74 ((h), 2H) δ 1.25 ((y), 3H)]

The 1H-NMR spectrum of the product of transesterification (Scheme 4.6A) in

Figure 4.45 shows that the signal (e) is moved from 3.68 ppm (Figure 4.39) to 4.21 ppm

(e’) after the successful reaction. Resonance (y) (Figure 4.44) completely disappeared after the reaction. The integral ratio of protons (g) to the methyl protons (a) is 1.97:2.99 which confirms the structure of the desired product.

157

1 Figure 4.45 H-NMR of Br-PEG-OMe in CDCl3

1 [Br-PEG-OMe (300 MHz, H-NMR, CDCl3): δ 4.21 ppm ((e’), 2H), δ 3.69 ((j), 2H), δ

3.68-3.59 ((b, c, d), 64H), δ 3.36 ((a), 3H), δ 2.40 ((g), 2H), δ 1.95-1.72 ((h, i), 4H)]

4.3.2. Lithiation followed by PEGylation of fluorescin

Br-PEG-OMe was later used to PEGylate the fluorescin by lithiation with n-BuLi in freshly distilled dry DMSO (Scheme 4.6C). The color of the fluorescin solution immediately started turning red during the addition of n-BuLi (Figure 4.46) and turned dark red after the complete addition of n-BuLi (Figure 4.47).

158

Figure 4.46 Fluorescin solution turning red during the dropwise addition of n-BuLi.

After 15 minutes of stirring, dry Br-PEG-OMe was added to the reaction mixture and stirred at room temperature.

Figure 4.47 Reaction mixture after the complete addition of Br-PEG-OMe compound to the lithiated fluorescin.

159

Samples were taken for 1H-NMR analysis after 15 min and 2 hours. The 1H-NMR spectrum of the 15-minute sample shows that the resonance (j) moved from 3.69 ppm to

4.31 ppm (j’) after the reaction (Figure 4.45, Figure 4.48). Resonance (13’) moved from

7.72 ppm to 7.69 ppm (13”) after the reaction (Figure 4.41, Figure 4.48). Similarly, proton

(10) moved from 6.13 ppm to 5.90 ppm (10’) respectively (Figure 4.41, Figure 4.48). The integral ratio of signal (j’) with respect to reference proton (13”) is 1.03: 1.67 demonstrating the progress of the PEGylation reaction. Extra resonances are present between 6 ppm–7 ppm (Figure 4.48). Therefore, the reaction was further continued for a total of 2 hours and the 1H-NMR spectrum of the sample was checked (Figure 4.49). In

Figure 4.49, the chemical shifts of the signals of the 2-hour sample are same as that of 15- minute sample with extra resonances between 6 ppm-7ppm which may be due to the unreacted fluorescin (Figure 4.48). The integral ratio of signal (j’) with respect to reference proton (13”) is (1: 2.01) (Figure 4.49) confirming the completion of the reaction.

160

Figure 4.48 1H-NMR of 15-minute sample of PEGylation reaction in DMSO d6

[PEGylated fluorescin (300 MHz, 1H-NMR, DMSO d6): δ 7.64 ppm ((13”), 1H), δ 7.32

((15’), 1H), δ 7.20 ((16’), 1H) δ 6.93-6.89 (14’,23,6,20,9,22,5), 7H), δ 5.90 ((10), 1H), δ

4.31 ((j’), 2H), δ 4.08 ((e), 2H), δ 3.47 ((b,c,d), 64H), δ 3.19 ((a), 3H), 2.31 ((g), 2H), 1.75

((h), 2H), 1.64 ((i), 2H)]

161

1 Figure 4.49 H-NMR of 2-hour sample of PEGylation reaction in DMSO d6

[PEGylated fluorescin (300 MHz, 1H-NMR, DMSO d6): δ 7.64 ppm ((13”), 1H), δ 7.32

((15’), 1H), δ 7.20 ((16’), 1H) δ 6.93-6.89 (14’,23,6,20,9,22,5), 7H), δ 5.90 ((10), 1H), δ

4.31 ((j’), 2H), δ 4.08 ((e), 2H), δ 3.47 ((b,c,d), 64H), δ 3.19 ((a), 3H), 2.31 ((g), 2H), 1.75

((h), 2H), 1.64 ((i), 2H)]

The product was further analyzed by matrix-assisted laser desorption/ionization, time of flight (MALDI-ToF) mass spectrometry as shown in Figure 4.50. There are a total of three distributions of peaks A, B and C (Figure 4.50(1)) and their enlarged spectra are shown in Figure 4.50(2), 4.50(3) and 4.50(4) with each peak separated by 44 m/z units.

162

In Figure 4.50(2), the peak at m/z 743.675 corresponds to the lithium (Li) complex of the

15-mer fraction of PEG-OMe. The calculated monoisotopic mass of this peak is [m/z =15

×44.03 (C2H4O repeat unit) + 45.03 (C2H4OH- end group) + 31.02 (OCH3- end group) +

7.02 (Li+)] is 743.52. In Figure 4.50(3), the peak at m/z 1159.802 corresponds to the Li complex of the 15-mer fraction of the desired product. The calculated monoisotopic mass of this peak [m/z = (1152.57 (C58H88O23) + 7.02 (Li+)] is 1159.59. In Figure 4.50(4), the small distribution of peaks overlapping both distributions A and B corresponds to the traces of Br-PEG-OMe. The peak at m/z 905.748 corresponds to the Li complex of the 15-mer fraction of Br-PEG-OMe. The calculated monoisotopic mass of this peak [m/z = 898.41

(C38H75BrO18) +7.02 (Li+)] is 905.44.

Based on MALDI-ToF mass spectrometry, the sample contains PEGylated fluorescin along with unreacted Br-PEG-OMe and HO-PEG-OMe. However, the NMR showed the formation of the desired product so fragmentation may have occurred during

MALDI.

163

(1)

164

(2)

165

(3)

166

(4)

Figure 4.50 (1) MALDI-ToF mass spectrometry of the product (2) Enlarged distribution A (Lithiated PEG-OMe) (3) Enlarged distribution B (PEGylated fluorescin). (4) Enlarged distribution C (Lithiated Br-PEG-OMe)

In summary, PEGylation via lithium technology was successfully carried out where fluorescin was lithiated at ‘COOH’ followed by the addition of Br-PEG-OMe into the lithium salt of fluorescin at 25℃. 1H-NMR analysis demonstrated the formation of the desired product, but extra resonances appeared between 6 ppm - 7 ppm corresponding to the aromatic protons of unreacted fluorescin. Mass spectrometry confirmed the presence of the PEGylated product along with unreacted Br-PEG-OMe. The relatively large distribution corresponding to (HO-PEG-OMe) without unreacted fluorescin is not clear, 167 but one option is fragmentation occurring during MALDI. Further acrylation of the product was not performed due to the presence of the mixture.

168

4.4. Synthesis of tetrafunctional fluoresceins (FL)

Tetrafunctional FL compounds were synthesized via aza-Michael addition reaction to develop intermediates for multifunctional polymeric conjugates. First, fluorescein o,o’- diacrylate (FL-DA) was synthesized as discussed in Section 4.1, which allows the Michael addition of amines to the acrylate for the synthesis of tetrafunctional FL compounds. Since

PEGylation of these were not successful so they were not used to make our final compounds, the syntheses of these are presented here and will be published.

In this research, I studied the aza-Michael addition of diallylamine (DAA), diethyl iminodiacetate (DIDA), diethanolamine (DEA) and newly synthesized dihydroxy secondary amines to the FL-DA (Scheme 4.7).

Scheme 4.7 Synthesis of tetrafunctional FL by the aza-Michael addition reaction

The reactions of DAA and DEA were conducted in both deuterated dimethyl sulfoxide (DMSO d6) and in deuterated chloroform (CDCl3) to compare the reaction time.

For the NMR purpose, deuterated solvents were preferred to conduct the reactions. The enzyme Candida antarctica lipase B (CALB) was used in the synthesis.

169

4.4.1. Synthesis of tetraallyl-functionalized FL

Tetraallyl-functionalized FL was synthesized by reacting DAA with FL-DA at room temperature.

4.4.1.1. Reaction in DMSO, with CALB catalysis

Figure 4.51 shows the 1H-NMR spectrum of starting material DAA in DMSO-d6.

Integral ratio of protons (b), (c) and, (c’) to the reference proton (a) is 1.74: 1.97: 1.90:

4.00. This confirms the structure of the starting material DAA.

1 Figure 4.51 H NMR of the DAA in DMSO d6

170

[DAA (500 MHz, 1H-NMR, DMSO d6): δ 5.89-5.78 ppm ((b), 2H), δ 5.13 ((c), 2H), δ 5.02

((c’), 2H), δ 3.13 ((a), 4H)]

Reaction was then conducted between DAA and FL-DA in DMSO d6 using CALB catalysis. Figure 4.52 shows the 1H-NMR of the 1-minute sample of the reaction mixture

(Not purified). Splitting of resonance due to allyl proton into c and c’ (Figure 4.51) is not distinctly visible on NMR-300 instrument (Figure 4.52). Hence it is labeled as a single resonance ‘c’ (Figure 4.52). Methylene and methine protons due to the acrylate which were present at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11) disappeared after the reaction. Resonance (a) moved from 3.13 ppm to 3.29 ppm (a’) after the reaction and new signals (1””), (26’) and (2’), (25’) appeared at 2.92 ppm and 2.72 ppm (Figure 4.51,

Figure 4.52). Relative integral ratio of signals (1””, 26’) and (2’, 25’) with respect to the reference proton (13) is 1:4:4 which demonstrates the completion of Michael addition and confirms the structure of the product.

171

Figure 4.52 1H-NMR of the tetraallyl-functionalized FL in DMSO d6 (Reaction with CALB catalysis)

When Michael addition between DAA and FL-DA was conducted in DMSO d6 using CALB catalysis, 0.670 g (0.001 mol), 92 % of the product was obtained in one minute at room temperature. As the Michael addition with DAA is extremely fast and completes in 1 minute at room temperature using CALB catalysis, feasibility of the reaction was checked without CALB catalysis in DMSO d6.

172

4.4.1.2. Reaction in DMSO, without CALB Catalysis

Reaction was then conducted between DAA and FL-DA in DMSO d6 without

CALB catalysis. Figure 4.53 shows the 1H-NMR of the 1-minute sample of the reaction mixture (Not purified) when reaction was conducted without CALB catalysis. Splitting of resonance due to allyl proton into c and c’ (Figure 4.51) is not distinctly visible on NMR-

300 instrument (Figure 4.53). Hence it is labeled as a single resonance ‘c’ (Figure 4.53).

Methylene and methine protons due to the acrylate which were present at 6.70 ppm (1”),

6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11) disappeared after the reaction (Figure 4.53).

Resonance (a) moved from 3.13 ppm to 3.29 ppm (a’) after the reaction and new signals

(1””), (26’) and (2’), (25’) appeared at 2.92 ppm and 2.72 ppm (Figure 4.51, Figure 4.53).

Relative integral ratio of signals (1””, 26’) and (2’, 25’) with respect to the reference proton

(13) is 1:4:4 which demonstrates the completion of Michael addition and confirms the structure of the product.

173

Figure 4.53 1H-NMR of tetraallyl-functionalized FL in DMSO d6 (Reaction without CALB catalysis)

Figure 4.54 shows the 1H-NMR of the pure product after workup and drying. No methylene and methine protons due to the acrylate functionality are present at 6.70 ppm

(1”), 6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11, Figure 4.54). Integral ratios of

174 signals (1””, 26’) and (2’, 25’) with respect to the reference proton (13) is 1:4:4 which confirms the structure of the product.

Figure 4.54 1H-NMR of the pure tetraallyl-functionalized FL.

1 [Tetraallyl-functionalized FL (300 MHz, H NMR, CDCl3): δ 8.03 ppm ((13), 1H), 7.72-

7.59 ((15, 16), 2H), 7.19 ((14), 1H), 7.10 ((9,20), 2H), 6.82 ((5,6,22,23), 4H), 5.94-5.78

((b), 4H), 5.25-5.12 ((c), 8H), 3.16 ((a’), 8H), 2.92 ((1””, 26’), 4H), 2.72 ((2’, 25’), 4H)].

The structure of the tetraallyl-functionalized FL was further confirmed using 13C-

NMR (Figure 4.55). The 13C-NMR spectrum shows that the carbon signals due to acrylate

175 which were present at 133.47 and 127.42 ppm (Figure 4.12) disappeared after the Michael addition reaction. The successful Michael addition between DAA and the acrylate functionalities of the FL is confirmed by peaks at 48.41 ppm (1””, 26’), 32.69 ppm (2’,25’)

(Figure 4.55).

13 Figure 4.55 . C NMR of the pure tetraallyl-functionalized FL in CDCl3.

13 [Tetraallyl-functionalized FL (500 MHz, C NMR, CDCl3): δ 170.42 ppm (3,24), 169.27

(11), 152.90 (17), 152.31 (8,19), 152.15 (4,21), 135.28 (15), 135.05 (B), 130.02 (12),

128.99 (6,23), 126.20 (14), 125.22 (13), 124.06 (16), 118.18 (C), 117.69 (5,22), 116.38

(7,18), 110.38 (9,20), 81.82 (10), 56.60 (A), 48.41 (1””,26’), 32.69 (2’,25’)] 176

The High Resolution Mass Spectrometry (HRMS) Electrospray ionization (ESI) of the product as shown in Figure 4.56 confirms the formation of tetraallyl- functionalized FL with peaks at m/z = 633.25897, 634.26267, 635.27491. The corresponding calculated monoisotopic masses of these peaks are 635.28 [m/z= 634.27 (C38H38N2O7, tetraallyl-

+ functionalized FL) + 1.01 (H )], 634.27 [m/z= 634.27 (C38H38N2O7, tetraallyl- functionalized FL)], 633.26 [m/z= 634.27 (C38H38N2O7, tetraallyl- functionalized FL) +

1.01 (H+) – 2.02 (2H+)]. Calculated monoisotopic mass for the peak at m/z 631.24334 is

+ + 631.24 [m/z= 634.27 (C38H38N2O7, tetraallyl-functionalized FL) + 1.01 (H ) – 4.04 (2H )].

The other peaks at m/z 636.27850, 637.29068 are due to the presence of 13C and 14C isotope of the product. In Figure 4.56, ‘M’ represents the desired product tetraallyl-functionalized

FL.

177

Figure 4.56 ESI-Mass spectrum of the tetraallyl-functionalized FL

When Michael addition between DAA and FL-DA was conducted in DMSO d6 without CALB catalysis, 0.655 g (0.0010 mol), 89.95% of the product was obtained in one minute at room temperature.

From the 1H-NMR and mass spectrometry analysis of reaction, it was found that the catalyst CALB is not required for the aza-Michael addition between DAA and FL-DA in DMSO d6. This is due to the sufficiently high basicity of DAA (pKa 9.29),103 which makes it a strong nucleophile for the aza-Michael addition. The effect of solvent on reaction was checked by conducting the reaction in CDCl3.

178

4.4.1.3. Reaction in chloroform, with CALB catalysis

Reaction between DAA and FL-DA was then conducted in CDCl3 using CALB

1 catalysis. Figure 4.57 shows the H-NMR of the 10-minute sample of the reaction mixture

(Not purified). Methylene and methine protons due to the acrylate were still present at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11, Figure 4.57). New signals (1””),

(26’) and (2’), (25’) appeared at 2.92 ppm and 2.72 ppm. Michael addition between DAA and FL-DA in CDCl3 with CALB catalysis did not complete even after 10 minutes at room temperature.

Figure 4.57 1H NMR of the 10-minute sample of Michael addition between DAA and FL-DA in CDCl3 with CALB (Reaction with CALB catalysis) 179

Although reaction did not complete in 10-minute in CDCl3, reaction was also checked without CALB catalysis in CDCl3.

4.4.1.4. Reaction in chloroform, without CALB catalysis

Michael addition between DAA and FL-DA in CDCl3 without CALB catalysis did not complete even after 10 minutes at room temperature as shown in Figure 4.58.

Methylene and methine protons due to the acrylate were still present at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11, Figure 4.58). New signals (1””), (26’) and (2’),

(25’) appeared at 2.92 ppm and 2.72 ppm (Figure 4.58).

180

Figure 4.58 1H NMR of the 10-minute sample of Michael addition between DAA and FL-DA in CDCl3 (Reaction without CALB catalysis)

Based on the NMR and mass spectrometry analysis, it is concluded that the

Michael addition of DAA to FL-DA works best in DMSO d6 and completes in 1 minute at room temperature. Catalyst CALB is not required for the reaction.

4.4.2. Synthesis of tetraacetate-functionalized FL

Tetraacetate-functionalized FL was synthesized by reacting DIDA with FL-DA.

DIDA, which is a starting material for the tetraacetate-functionalized FL, has two acetate groups substituted on the secondary amine. It is a sterically hindered molecule with pKa

(4.92).102,104 As the basicity of DIDA is lesser than that of DAA and due to the sterically 181 hindered structure,102,104 aza-Michael addition between DIDA and FL-DA was conducted at 50℃.

4.4.2.1. Reaction in DMSO, with CALB catalysis

First, 1H-NMR of the starting material DIDA was checked in DMSO d6 as shown in Figure 4.59. Integral ratios of protons (a, d) with respect to reference proton (c) is 4:4:6 which confirms the structure of DIDA.

Figure 4.59 1H-NMR of the DIDA in DMSO d6

1 [DIDA (300 MHz, H NMR, CDCl3): δ 4.10 ppm ((c), 4H), δ 3.36 ((a), 4H), δ 1.19 ((d),

6H)]

Progress of the CALB catalyzed Michael addition between DIDA and FL-DA was checked by 1H-NMR analysis in DMSO d6. Figure 4.60 shows the 1H-NMR of the 2 hours

45-minute sample and 9-hour sample of the reaction mixture (before purification). 182

Resonance (a) of DIDA shifted from 3.36 ppm (Figure 4.59) to 3.63 ppm (a’) as the reaction progressed. Methylene and methine protons due to the acrylate functionality which are present at 6.70 ppm (1’’), 6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11) reduced from

2 to 0.50, 0.41 and 0.40 and new signals (1””), (26’) and (2’), (25’) appear at 3.18 ppm and

2.78 ppm (Figure 4.60). Relative integral ratio of protons (1””, 26’) and (2’, 25’) with respect to reference proton (13) should be 1:4:4 after the reaction. For the 2 hours 45- minute sample the ratio is 1: (2.97): (2.60) while for the 9 hours sample the ratio is 1:

(4.07): (4.07), which indicates the completion of the reaction in 9 hours. Methylene and methine protons due to the acrylate appearing at 6.70 ppm (1’’), 6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11) were negligible (0.09) after 9 hours (Figure 4.60). This demonstrates the successful aza-Michael addition reaction in 9 hours with CALB catalysis and confirms the structure of tetraacetate-functionalized FL.

183

Figure 4.60 1H NMR of the 2 hours 45 minute and 9 hour samples of the Michael addition between DIDA and FL-DA (using CALB)

Figure 4.61 shows the 1H-NMR of the pure product after workup and drying. Proton

(a) of the DIDA shifted from 3.36 ppm (Figure 4.59) to 3.63 ppm (a’) after the successful reaction. No methylene and methine protons due to the acrylate are present at 6.70 ppm,

6.30 ppm and 6.10 ppm (Figure 4.11, Figure 4.61) and integral ratios of signals (1””, 26’)

184 and (2’, 25’) with respect to aromatic proton (13) is 1:4:4. This demonstrates the completion of Michael addition and confirms the structure of the product.

Figure 4.61 1H-NMR of the tetraacetate-functionalized FL

1 [Tetraacetate-functionalized FL (300 MHz, H NMR, CDCl3): δ 8.03 ppm ((13), 1H),

7.72-7.59 ((15, 16), m, 2H), 7.19 ((14), 1H), 7.10 ((9,20), s, 2H), 6.82 ((22, 5,23,6) J = 1.5

Hz, 4H), 4.15 ((c), 8H), 3.63 ((a’), 8H), 3.18 ((2’,25’), 4H), 2.78 ((1””, 26’), 4H), 1.27 ((d),

12H)].

Figure 4.62 shows the 13C-NMR spectrum of tetraacetate-functionalized FL. The

13C-NMR spectrum shows that the acrylate carbon resonances at 133.47 ppm and 127.42 ppm (Figure 4.12) disappeared after the Michael addition reaction. The successful Michael addition between DIDA and the acrylate functionalities of the FL is confirmed by

185 resonances at 171.06 ppm (B), 54.94 ppm (A’), 49.91ppm (1””, 26’), 33.77 ppm (2’,25’)

(Figure 4.62).

Figure 4.62 13C NMR of the tetraacetate-functionalized FL

13 [Tetraacetate-functionalized FL (500 MHz, C NMR, CDCl3): δ 171.06 ppm (B), 170.28

(24,3), 169.52 (11), 152.83 (17), 152.10 (8,19), 151.87 (4,21), 135.26 (15), 128.66 (6,23),

125.92 (14), 123.83 (13, 16), 117.66 (5,22), 116.36 (7,18), 110.58 (9,20), 81.67 (10),

60.41 (C), 54.94 (A’), 49.91 (1””, 26’), 33.77 (2’, 25’), 14.06 (D)]

The HRMS-ESI in Figure 4.63 confirms the formation of tetraacetate- functionalized FL. The intense peak at m/z 819.29701 corresponds to the protonated form 186 of the desired product. The other peaks at m/z 820.30017, 821.3035 are due to the presence of 13C and 14C isotope of the product. The corresponding calculated monoisotopic masses of the peaks at m/z 819.29701 and 817.28161 are 819.30 [m/z= (C42H46N2O15, tetraacetate-

+ functionalized FL) + 1.01 (H )], and 817.28 [m/z= (C42H46N2O15, tetraacetate- functionalized FL) - 2.02 (2H+)]

In Figure 4.63, ‘M’ represents the desired product tetraacetate-functionalized FL.

Figure 4.63 ESI-Mass spectrum of the tetraacetate-functionalized FL

When DIDA and FL-DA were reacted in DMSO d6 using CALB catalysis, 1.9 g

(0.002323 mol) 67.8 % of the product was obtained. After the successful Michael addition 187 of DIDA with FL-DA in 9 hours. However, feasibility of the reaction was also checked without CALB catalysis.

4.4.2.2. Reaction in DMSO, without CALB catalysis

Reaction between DIDA and FL-DA was conducted without CALB catalysis.

Progress of the Michael addition was checked by 1H-NMR analysis in DMSO d6. 1H-NMR of the 5 minutes, 30 minutes, 1 hour 30 minutes, and 9 hours samples from the reaction mixture (before purification) (Figure 4.64) show that the proton (a) of DIDA shifted from

3.36 ppm (Figure 4.59) to 3.63 ppm (a’) during the reaction (Figure 4.64). Methylene and methine protons due to the acrylate, which are present at 6.70 ppm (1’’), 6.30 ppm (2) and

6.10 ppm (1”’), gradually reduced from ‘2’ to 0.82, 0.63, and 0.69, respectively (Figure

4.11, Figure 4.64), and new signals (1””), (26’) and (2’), (25’) increased at 3.18 ppm and

2.78 ppm. The sample at 27 hours (before purification) shows that the methylene and methine protons at 6.70 ppm (1’’), 6.30 ppm (2) and 6.10 ppm (1”’) became negligible

(Figure 4.11, Figure 4.64). The relative integral ratio of signals (1””, 26’) and (2’, 25’) with respect to the reference proton (13) is 1:4:4. These results demonstrate the completion of the Michael addition, and confirm the structure of the tetraacetate-functionalized FL.

188

Figure 4.64 1H NMR of the reaction samples of Michael addition between DIDA and FL- DA with CALB catalysis in DMSO d6.

189

The reaction between DIDA and FL-DA without CALB catalysis yielded 1.69 g

(0.002063 mol), 60.3% of product. As the DIDA is less basic than DAA with pKa 4.92102

CALB catalysis is necessary to accelerate the Michael addition. Basicity of DIDA is lesser than DAA due to the presence of an electron withdrawing ester group resulting in a more negative inductive effect than that of DAA. Reaction between DIDA and FL-DA works best with CALB catalysis in DMSO d6.

4.4.3. Synthesis of tetrahydroxy-functionalized FL

Tetrahydroxy-functionalized FL was attempted to synthesize first by reacting DEA with FL-DA.

4.4.3.1. Reaction between DEA and FL-DA

4.4.3.1.1. Reaction in DMSO, with CALB catalysis

First, 1H-NMR of the starting material DEA was checked in DMSO d6 as shown in

Figure 4.65. The integral ratio of proton (b) with respect to proton (a) is 4:4 which confirms the structure of DEA.

190

Figure 4.65 1H NMR of the DEA in DMSO d6

[DEA (300 MHz, 1H NMR, DMSO d6): δ 3.42 ppm ((b), 4H), δ 2.54 ((a), 4H)]

Reaction between DEA and FL-DA was then conducted using CALB catalysis in

DMSO d6. Figure 4.66 shows the 1H-NMR of a 1-minute sample from the reaction mixture

(not purified). 1H-NMR shows that the proton (a) shifted from 2.54 ppm (Figure 4.65) to

2.57 ppm after the reaction (Figure 4.66). Methylene and methine protons due to the acrylate which were present at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’) disappeared after the reaction and new signals (1””), (26’) and (2’), (25’) appeared at 3.28 ppm and

2.87 ppm (Figure 4.11, Figure 4.66). The relative integral ratio of signals (1””, 26’) and

(2’, 25’) with respect to reference proton (13) is 1:4:4 which demonstrates the completion of Michael addition and confirms the structure of the product.

191

Figure 4.66 1H NMR of the tetrahydroxy-functionalized FL in DMSO d6 (Reaction with CALB catalysis)

[Tetrahydroxy-functionalized FL (300 MHz, 1H NMR, DMSO d6): 8.05 ppm ((13), 1H),

7.87-7.70 ((15,16), 2H), 7.40 ((14), 1H), 7.26 ((9,20), 2H), 7.02-6.82 ((5,6,22,23), 4H),

3.60-3.28 ((b’, 26’, 1””), 12H), 2.87 ((2’, 25’), 4H), 2.55 ((a’), 8H)]

As the Michael addition with DEA was completed in 1 minute at room temperature using CALB catalysis, reaction was also checked without CALB catalysis in DMSO d6.

192

4.4.3.1.2. Reaction in DMSO, without CALB catalysis.

Reaction was then conducted without CALB catalysis in DMSO d6. Figure 4.67 shows the 1H-NMR of a 1-minute sample from the reaction mixture (not purified). 1H-

NMR shows that the proton (a) shifted from 2.54 ppm (Figure 4.65) to 2.57 ppm after the reaction (Figure 4.67). Methylene and methine protons due to the acrylate which were present at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’) disappeared after the reaction and new signals (1””), (26’) and (2’), (25’) appeared at 3.28 ppm and 2.87 ppm (Figure

4.11, Figure 4.67). The relative integral ratio of signals (1””, 26’) and (2’, 25’) with respect to reference proton (13) is 1:4:4 which demonstrates the completion of Michael addition and confirms the structure of the product.

193

Figure 4.67 1H NMR of the tetrahydroxy-functionalized FL in DMSO d6 (Reaction without CALB catalysis)

[Tetrahydroxy-functionalized FL (300 MHz, 1H NMR, DMSO d6): 8.05 ppm ((13), 1H),

7.87-7.70 ((15,16), 2H), 7.40 ((14), 1H), 7.26 ((9,20), 2H), 7.02-6.82 ((5,6,22,23), 4H),

3.60-3.28 ((b’, 26’, 1””), 12H), 2.87 ((2’, 25’), 4H), 2.55 ((a’), 8H)]

Workup was performed after 1 minute of reaction time and mass spectrometry analysis of the product was conducted however, the desired product was not detected.

194

The HRMS-ESI in Figure 4.68 shows an intense peak at m/z 333.07571 which corresponds to the protonated form of the starting material FL. The corresponding calculated

+ monoisotopic mass of the peak is 333.08 [m/z= 332.07 (C20H12O5, FL) + 1.01 (H )]. The other peaks at m/z 334.07910, 335.0824 are due to the presence of 13C and 14C isotope.

In Figure 4.68, ‘M’ represents the staring material FL.

Figure 4.68 ESI-Mass spectrum of the product of Michael addition between FL-DA and DEA after workup and drying.

Desired tetrahydroxy-functionalized FL product was formed in 1-minute reaction time as confirmed by the HNMR spectra (Figure 4.66, Figure 4.67). However, after workup 195 and storage it was hydrolyzed back to FL. We concluded that the product is unstable and hydrolyzes to FL after some time. The possible reason is neighboring group participation during the reaction with amine.

Literature reports that when FL-DA is incubated with cysteine in ethanol phosphate buffer, cleavage of an ester bond of FL-DA takes place and FL obtains.56,81 as shown in

Scheme 4.8.

Scheme 4.8 Addition reaction of cysteine with FL-DA in ethanol-phosphate buffer.56, 81

Based on Scheme 4.8, the possible mechanism of reaction between FL-DA and

DEA can be as follows where neighboring group participation due to ‘OH’ takes place and

‘OH’ hydrolyzes the product by attacking the carbonyl carbon of the FL-DA leading to the formation of FL and a heterocyclic ring (oxazepin) as shown in Figure 4.69.

196

Figure 4.69 Proposed mechanism for the reaction between FL-DA and DEA.

Figure 4.70 shows the 1H-NMR of the product after the workup. Methylene and methine protons due to the acrylate are not present at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’) (Figure 4.11, Figure 4.70) and new peaks corresponding to the oxazepin and FL are obtained.

197

Figure 4.70 1H NMR of product of Michael addition between FL-DA and DEA after work up and drying.

[(300 MHz, 1H NMR, DMSO d6): δ 7.95 ppm ((13), 1H), 7.62-7.76 ((15,16), 2H), 7.22

((14), 1H), 6.64 ((6,23), 2H), 6.51 ((5,9,20,22), 4H), 4.19 ((f), 4H), 3.45 ((b, i), 6H), 2.77-

2.50 ((a, c, d, g, h), 12H)]

Michael addition between DEA and FL-DA was extremely fast and completed in 1 minute with as well as without CALB catalysis at room temperature in DMSO d6.

However, the product was unstable under ambient condition and hydrolyzed back to the

FL. We concluded that the DEA with pKa 8.7105 being more basic than DIDA reacts with

198

FL-DA in DMSO d6 in 1 minute at room temperature to form the desired tetrahydroxy FL product. However, due to the neighboring group participation it hydrolyzes back to FL.

The effect of solvent on the reaction was checked by conducting the reaction in CDCl3.

4.4.3.1.3. Reaction in chloroform, with CALB catalysis

Reaction between DAA and FL-DA was conducted in CDCl3, with CALB catalysis. Figure 4.71 shows the 1H-NMR of 5-minute sample from the reaction mixture

(not purified). Methylene and methine protons due to the acrylate at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’9) reduced from ‘2’ to 0.94, 0.83, 0.83 with CALB catalysis

(Figure 4.11, Figure 4.71). Reaction did not complete in 5 minutes with CALB.

199

1 Figure 4.71 H NMR spectrum of Michael addition of DEA to FL-DA in CDCl3 at room temperature (Reaction with CALB catalysis)

Reaction between DEA and FL-DA worked best in DMSO d6 and completed in 1 minute at room temperature. The neighboring group participation can be avoided by synthesizing secondary amines with longer alkyl chains. Hence, new secondary amines with longer alkyl chains and 2-hydroxy groups were synthesized by ultraviolet (UV)- mediated thiol-ene click reactions as discussed in the following section.

4.4.3.2. UV-mediated reaction between diallylamine (DAA) and 4-mercapto 1-

butanol to synthesize di-butanol-based amine compound.

To avoid hydrolysis due to the short alkyl chain of dihydroxy secondary amine, a longer chain alkyl amine was synthesized by UV mediated thiol-ene click reaction between

200

DAA and 4- mercapto 1-butanol using photo initiator 2,2-dimethoxy-2- phenylacetophenone (DMPA).

Figure 4.72 shows the 1H-NMR spectrum of 4- mercapto 1-butanol in DMSO-d6.

Integral ratio of protons (e+f), (d) to the reference proton (g) is (3.87): (1.64): (2.00). This confirms the structure of the starting material 4- mercapto 1-butanol.

Figure 4.72 1H-NMR of the 4- mercapto 1-butanol

[4- mercapto 1-butanol (300 MHz, 1H NMR, DMSO d6): δ 3.38 ppm ((g), 2H), δ 2.45 ppm ((d), 2H), δ 1.54-1.36 ppm ((e, f), 4H)]

1H-NMR of the product after purification (Figure 4.73) shows that the allyl protons of DAA which were present at 5.94-5.78 ppm, 5.25-5.12 ppm (Figure 4.51) disappeared and new signals (b’), (c’) appeared at 1.60 ppm and 2.45 ppm (Figure 4.73). The relative

201 integral ratio of signals (b’), (c’) with to (g) is (1.88) : (2.01) : (2.00), which confirms the completion of thiol-ene click reaction.

Figure 4.73 1H-NMR of the di-butanol-based amine compound in DMSO d6

[Di-butanol-based amine compound (300 MHz, 1H NMR, DMSO d6): δ 3.35 ppm ((g),

4H), δ 2.55-2.49 ((a), 4H), 2.45 ((c’, d), 8H), 1.60 ((b’), 4H), 1.53 - 1.37 ((e,f), 8H))]

Figure 4.74 shows the 13C-NMR spectrum of the di-butanol-based amine compound. The allyl carbon peaks which appear at 135.05, 118.18 ppm are not present in the spectrum (Figure 4.74). The successful thiol-ene click reaction between DAA and 4- mercapto 1-butanol is confirmed by peaks at 31.61 (B’), 30 (C’) ppm.

202

Figure 4.74 13C-NMR of the di-butanol-based amine compound in DMSO d6

[Di-butanol-based amine compound (500 MHz, 13C NMR, DMSO d6): δ 60.71 ppm (G),

48.58 (A), 32.15 (D), 31.61 (B’), 31.5 (F), 30 (C’), 26.26 (E)]

The HRMS-ESI in Figure 4.75 confirms the formation of di-butanol-based amine compound with peaks at m/z = 310.3, 326.3, 342.2. The corresponding calculated monoisotopic masses of these peaks are 310.19 [m/z= 309.18 (C14H31NO2S2, dihydroxy-

+ functionalized secondary amine (1) + 1.01 (H )], 326.18 [m/z= 309.18 (C14H31NO2S2, dihydroxy-functionalized secondary amine (1)) + 1.01 (H+) + 15.99 (O+)], 342.18 [m/z=

+ + (C14H31NO2S2, dihydroxy-functionalized secondary amine (1)) + 1.01 (H ) + 31.99 (2O )].

The other peaks at m/z 311.3, 312.2, 327.3, 328.3, 343.2 are due to the presence of 13C and

14C isotope.

203

In Figure 4.75, ‘M’ represents the desired product dihydroxy-functionalized secondary amine (1).

Figure 4.75 ESI - Mass spectrometry of di-butanol-based amine compound

The reaction yielded 1.1 g (0.0036 mol), 31.4% of the product which was then further reacted with FL-DA by my colleague to obtain the tetrahydroxy-functionalized FL

(1). The product was found to be stable and did not hydrolyze back to FL.

Similarly, UV mediated thiol-ene click reaction between DAA and 9- mercapto-1- nonanol was conducted by my colleague using Irgacure-651. This secondary amine was then reacted with FL-DA by me to obtain the tetrahydroxy-functionalized FL (2) as discussed in the following section.

204

4.4.3.3. CALB-catalyzed Michael addition reaction between FL-DA and di-

nonanol-based amine compound

CALB-catalyzed Michael addition reaction between FL-DA and di-nonanol-based amine compound was conducted. 1H-NMR in Figure 4.76 shows that the methylene and methine protons due to the acrylate are reduced from ‘2’ to ‘0.16’ at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”) after 48 hours stirring and new signals (1””), (26’) and (2’),

(25’) appeared at 4.23-4.05 ppm and 3.08 ppm (Figure 4.11, Figure 4.76). The relative integral ratio of signals (1””+26’) and (2’+25’) with respect to (13) is 1 : (4.44) : (4.16).

205

Figure 4.76 1H-NMR of the tetrahydroxy-functionalized FL (2)

1 1 [Tetrahydroxy-functionalized FL (300 MHz, H NMR, DMSO d6): H NMR (CD3OD, 300

MHz) δ 8.05 ppm ((13), 1H), 7.53- 7.63 ((15,16), 2H), 7.15 ((14), 1H), 6.68 ((9, 20), 2H),

6.64 - 6.48 ((5,22,23,6), 4H), 4.00-4.23 ((1””,26’), 4H), 3.65 – 3.50 ((m), 8H), 2.97

((2’,25’), 8H), 2.62 ((b, d’, e), 24H), 1.98 ((c’), 8H) 1.38 - 1.32 ((f,g,h,i,j,k,l), 56H)]

In summary, tetrafunctional FL compounds were successfully synthesized by aza-

Michael addition reaction between FL-DA and secondary amines with as well as without

CALB-catalysis. For the aza Michael addition reactions without CALB catalysis, basicity

206 of the amine is an important factor. For large molecules, such as DIDA, CALB-catalysis was found to be extremely useful as the reaction time is three times less than that of the reaction time without CALB catalysis.

Reactions with DAA and DEA are extremely fast in DMSO d6 and product was obtained in 1 minute at room temperature with as well as without CALB-catalysis while the reaction in CDCl3 did not complete even after 10 and 5 minutes of the reaction time with DAA and DEA, respectively. CALB catalysis did not have any effect on the reactions with DAA and DEA in DMSO d6 due to their sufficiently high basicity for the Michael addition.

However, the product of the Michael addition between FL-DA and DEA was highly unstable and hydrolyzed to FL due to the shorter alkyl chain of the secondary amine which resulted in neighboring group participation. This problem was resolved by synthesizing longer alkyl chain secondary amines with hydroxy groups, followed by the Michael addition with FL-DA. These newly synthesized tetrahydroxy-functionalized FL were stable and did not hydrolyze to FL. These tetrafunctional-FL compounds are suitable precursors to attach the polymer for diagnostic applications.

207

4.5. Scale-up of fluorescein o,o’-diacrylate (FL-DA) synthesis

As discussed in Section 4.1, FL-DA was first synthesized on a small scale by the acrylation of fluorescein (FL) using acryloyl chloride (AcrCl) in the presence of triethylamine (TEA) in dichloromethane (DCM). Then the synthesis was scaled up utilizing a 5L capacity jacketed glass reactor (Figure 4.77) owned by Enzyme Catalyzed

Polymers LLC.

Figure 4.77 5L capacity jacketed glass reactor for the scale-up.

4.5.1. Calculations of the enthalpy of acrylation reaction

An increase in temperature was observed during the small-scale synthesis. Hence, the enthalpy of the reaction was calculated before the scale-up of FL-DA synthesis. Scheme

4.11 shows the stoichiometric coefficients of FL-DA synthesis. TEA hydrochloride salt generates during the reaction. Also, excess acryloyl chloride reacts with TEA to form quaternary ammonium salt as shown by the Scheme 4.9.

208

Scheme 4.9 Reaction scheme for FL-DA synthesis.

The enthalpy of the reaction is the difference between the enthalpy of product formation and reactant formation as shown in following equation. The enthalpy of a chemical reaction can be calculated using Hess’s law:92

∆퐻푟푒푎푐푡푖표푛(kJ/mol) = ∑ ∆퐻푓 푝푟표푑푢푐푡 − ∑ ∆퐻푓 푟푒푎푐푡푎푛푡

where

∆퐻푓 푝푟표푑푢푐푡푠 = 퐸푛푡ℎ푎푙푝푦 표푓 푝푟표푑푢푐푡 푓표푟푚푎푡𝑖표푛 (kJ/mol)

∆퐻푓 푟푒푎푐푡푎푛푡 = 퐸푛푡ℎ푎푙푝푦 표푓 푟푒푎푐푡푎푛푡 푓표푟푚푎푡𝑖표푛 (kJ/mol)

Hf values of the reactants and products can be obtained from the literature at 298.15

°K as shown in Table 4.1.

209

Table 4.1 Compounds and their heat of formations.

Compound ΔHf (kJ/mol)

AcrCl93 -108

TEA 94 -169

TEA hydrochloride salt 95 -388

As discussed in Background section, enthalpies for FL-A, FL-DA and the side product quaternary ammonium salt could not be found in the literature. For this latter we used the value of the TEA-HCl salt. The values for FL-A and FL-DA were calculated at

298.15 °K using Joback’s group contribution method by considering the contribution of

92 the following groups and their respective ΔHf values as shown in Table 4.2.

210

92 Table 4.2 Group contribution values to calculate ΔHf of FL and FL-DA.

Group Number ΔH Number ΔH Increments of groups in (kJ/mol) of groups in (kJ/mol) FL FL-DA 10 20.90 10 20.90 (ring)

8 371.44 8 371.44 (ring)

1 - 1 -164.50 (ring) 164.50

(ring) 2 -276.32 2 -276.32

(ring) 2 -443.30 - -

1 79.72 1 79.72

(ring)

- - 2 -266.44 (Non ring)

- - 2 75.94 (Non ring)

(Non ring) - - 2 -19.26

(Non ring) - - 2 -264.44

Total ΔH -412.06 -442.96

211

This method sums up the ΔH (kJ/mol) groups in the compound in question as

92 follows: ΔHf = 68.29 + Total ΔH

92 ΔHf = 68.29 + Total ΔH

Substituting the values from Table 4.2,

ΔHf FL = 68.29 - 412.06 = -343.77 kJ/mol

ΔHf FL-DA = 68.29 - 442.96 = -374.67 kJ/mol

∆퐻푟푒푎푐푡푖표푛 = [(-374.67) + 2(-388)+1.76(-388)] - [(-343.77) + 3.89(-169) + 3.76(-108)]

∆퐻푟푒푎푐푡푖표푛 = -426.29 kJ/mol

A reaction that releases the heat or an exothermic reaction has negative ∆퐻푟푒푎푐푡푖표푛 by convention. Using Joback method, the enthalpy of the FL-DA synthesis was calculated as -426.29 kJ/mol at 298.15 °K.

ΔHf FL and ΔHf FL-DA were also calculated by the simulation software- MOPAC

(Molecular Orbital PACkage) with the help of Prof. J. Richard Elliott (Chemical &

Biomolecular Engineering, The University of Akron, Ohio). MOPAC is a program used in computational chemistry which obtained the ΔHf FL and ΔHf FL-DA. The molecular structure of FL and FL-DA were specified which calculated the required values for FL and FL-DA.

However, input files of the program could not be added here. The following heat of formation values were obtained for FL and FL-DA using simulation:

ΔHf FL = -296 kJ/mol

ΔHf FL-DA = -370.6 kJ/mol

Substituting the above ΔHf FL, ΔHf FL-DA values, obtained from simulation and values from Table 4.1 in Hess’s law equation,

212

∆퐻푟푒푎푐푡푖표푛 = [ (-370.6) + 2(-388)+1.76(-388)] - [ (-296) + 3.89(-169) + 3.76(-108)]

= -469.99 kJ/mol.

The enthalpy of FL-DA synthesis using heats of formation determined from

MOPAC was calculated as -469.99 kJ/mol at 298.15 °K. The negative sign represents that the reaction is exothermic. This is in reasonable agreement with the Joback method.

4.5.1.1. Experimental study to measure the enthalpy of acrylation reaction.

Before performing the scale-up experiment, the heat liberated during the reaction was measured by performing a small bench scale experiment with 2.04 g FL in a closed system. The reaction was conducted with as much as insulation possible to approximate adiabatic conditions. A thermocouple was placed inside the reactor which was closed with a rubber septum and tied with teflon tape as shown in Figure 4.78. The reactor covered with an insulating material was placed in a styrofoam cup. (Figure 4.78).

Figure 4.78 Experimental setup to calculate the enthalpy of the acrylation reaction.

213

The reaction was conducted with FL in DCM and TEA. AcrCl was added entirely at once into the reaction mixture at room temperature and the temperature change was monitored with the thermocouple. A rapid temperature rise was observed from 20°C to

46°C in 5 seconds.

Specific heat capacity data (Cp, J/mol. K) of all the reactants and solvent are required for the calculation. Literature reports Cp values of AcrCl, TEA and DCM.

However, the Cp value of FL could not be found in the literature. Hence, the Cp (J/mol. K) of FL was calculated using the atomic contribution method. This method was developed

92 by Hurst and Harrison which calculates the Cp of a solid at 298.15 °K.

Cp FL = C20H12O5

Cp FL = 20(10.89) + 12(7.56) + 5(13.42)

Cp FL = 375.62 J/mol. K

Table 4.3 shows the recipe of the above small-scale experiment with the Cp value of each reactant at 298.15 °K

Table 4.3 Specific heat capacities of the reactants.

Cp at 298.15 MW C n*C Chemical W (g) V(m K p (g/mol) n (mol/L) Eq (J/K) L) (J/mol. K) FL 332.31 2.04 1.28 0.0061 0.2382 1.00 375.62 2.3059 AcrCl106 90.51 1.8938 1.70 0.0209 0.8118 3.41 85.3 1.7848 TEA107 101.19 2.0314 2.80 0.0201 0.7789 3.27 220* 4.4165

DCM108 84.93 26.6 20.00 0.3132 98.58 30.8752

Total 25.78 Total: 39.3824 (J/K)

* Cp of TEA is at 288.15 K

214

(MW: Molecular weight V: Volume, n= Number of moles, C= Concentration, Eq. =

Number of equivalents)

Initial temperature T1: before the addition of AcrCl = 20°C (293.15 °K)

Final temperature T2: after the addition of AcrCl = 46°C (319.15 °K)

o Hence, dT = T2 - T1 = 26

Heat liberated during the reaction, Q = n × Cp × dT = 39.3824 J/K× 26 K = -1023.9421 J

During the acrylation reaction with the recipe in Table 4.3, the amount of heat liberated in

5 seconds during the experiment is 1023.94 J. As the reaction system was not completely insulated, there should be some heat loss to the air.

Using the Joback method, ∆퐻푟푒푎푐푡푖표푛 was calculated as -426.29 kJ/mol. For the reaction with 2.04 g FL (0.0061 moles of FL) with the recipe in Table 4.3, heat liberated is calculated as, Q= -426.29 kJ/mol × 0.0061 mol = -2600.37 J. This is twice as much as the measured value. Similarly, calculations were performed to calculate the maximum heat that can be liberated during the scale-up reaction with 98 g of FL (Table 4.4)

Table 4.4 Specific heat capacities of the reactants.

Cp at 298.1 MW C Chemical W (g) 5K n*C (J/K) (g/mol) V(mL) n (mol/L) Eq p (J/mol .K) FL 332.31 98.000 61.25 0.2949 0.1889 1.00 375.6 110.7724 2 AcrCl 90.51 100.26 90 1.1077 0.7095 3.76 85.3 94.4888 TEA 101.19 116.08 160 1.1471 0.7348 3.89 220* 252.3728 DCM 84.93 1662.5 1250 19.5749 98.58 1929.6980 Total 1561.26 Total: 2387.3319J

215 * Cp of TEA is at 288.15 K

For the scale-up reaction with 98 g FL (0.2949 moles of FL) as given in the Table

4.4., amount of heat liberated is calculated as, Q= -426.29 kJ/mol × 0.2949 mol = -125.71 kJ= 125.713×103 = -125713 J which is the maximum heat that liberates with the 98 g

(0.2949 moles of FL).

To find the dT during the scale-up, amount of heat liberated, Q = n × Cp × dT

125713 = 2387.3319 × dT dT = 52.66℃ (This is the dT when AcrCl is completely added at once)

The increase in temperature of the reaction mixture (dT) = 52.66℃.

If AcrCl is added completely at once, the temperature of the reaction mixture would increase by 52.66℃. As the reaction is exothermic, AcrCl is added dropwise into the reaction mixture. Let us consider that the dropwise addition of 1.1077 moles of AcrCl

(Table 4.4) takes 90 minutes (5400 seconds). Hence, the rate of addition of AcrCl= 1.1077 mol /5400 s = 2.0513 × 10-4 mol/s.

As soon as a drop of AcrCl is added, heat liberates. Hence, rate of heat liberation during the addition of AcrCl,

Q= -426.29 kJ/mol × 2.0513 × 10-4 mol/s.

Q= -87.44 J/s

87.44 J/s heat liberates during the dropwise addition of AcrCl into the reaction mixture. Based on the calculations, cooling is necessary during the scale-up experiment to control the temperature. Aqueous ethylene glycol was used as the coolant. Considering that the reaction mixture will have a temperature of -3℃, the required flow rate of the coolant

216 to maintain the reaction temperature is calculated as follows. The assumption is that all the heat liberated during the addition of AcrCl is completely absorbed by the coolant.

The heat generated per second during the addition of AcrCl = Heat absorbed by the coolant. Q = ṁcoolant × Cp × dT

Where, ṁcoolant is the mass flow rate of the coolant.

During the scale-up, dT of the coolant should be negligible. Considering dT = 0.5℃.

87.44 J/s = ṁcoolant × 3.75 J/(g. ℃)× 0.5℃

ṁcoolant = 46.63 g/s

Density of the coolant: 1090 kg/m3=1090 g/L

Dividing the ṁcoolant by density gives the volumetric flow rate.

46.63 /1090 = 0.04278 L/s = 2.567 L/min

The volumetric flow rate of the coolant required to maintain the dT of 0.5℃ is

2.567 L/min.

4.5.2. Scale-up of FL-DA synthesis

Scale-up was conducted in a 5L capacity jacketed glass reactor. A solution of FL,

TEA and DCM was prepared inside the reactor and stirred to obtain a homogenous solution. Aqueous ethylene glycol (12 L) bath was prepared and cooled using MGW Lauda

RM20 Thermostat. Thermostat operates with the coolant flow rate of 8 L/min. The coolant was circulated with 8 L/min flow rate through the jacket to cool down the stirring solution of FL inside the reactor. The temperature inside the cooling bath thermostat before coolant circulation was first reduced to -16.6°C. Table 4.5 shows the temperature profile during

217 the coolant circulation before starting the reaction. Temperatures were noted as soon as the coolant circulation was started.

218

Table 4.5 Temperature profile during the coolant circulation through the jacket before the addition of AcrCl.

Time (t) TJacket_inlet TJacket_outlet ΔTcoolant TInside_reactor Tinside_cooling_bath Agitator

Hr:min:sec (°C) (°C) (°C) (°C) (°C) Rpm

0:0:10 -8.5 -6 2.5 -4 -11.5

0:1:36 -8.6 -6.2 2.4 -5 -11.6 58 0:4:38 -8.8 -6.4 2.4 -6 -11.9

0:19:58 -9.7 -7.1 2.6 -8 -12.4

0:31.08 -10.1 -7.5 2.6 -9 -12.4

0:38:24 -10.2 -7.8 2.4 -10 -12.5 120

0:43:45 -10.3 -7.9 2.4 -10 -12.6

The ΔT of the coolant between the inlet and the outlet before starting the reaction

averaged ~2.5°C. The mixture inside the reactor reached the desired -10 C in about 40

minutes. AcrCl was then added dropwise into the reaction mixture using an addition funnel

and the following temperature profile was observed during the addition (Table 4.6).

219

Table 4.6 Temperature profile during the addition of AcrCl.

Time (t) TJacket_inlet TJacket_outlet ΔTcoolant TInside_reactor TInside_

Hr:min:sec (°C) (°C) (°C) (°C) Thermostat bath (°C)

0:0:0 -10.3 -7.9 2.4 -10.0 -12.6

0:0:5 -9.7 -7.1 2.6 -10.0 -13.0

0:3:56 -9.7 -6.9 2.8 -8.0 -13.0

0:11:28 -9.7 -6.8 2.9 -7.0 -12.9

0:29:40 -9.8 -6.9 2.9 -6.0 -12.8

0:33:33 -9.3 -6.7 2.6 -5.0 -12.5

0:41:17 -9.1 -6.0 2.6 -4.0 -12.2

0:51:56 -8.9 -6.3 2.6 -3.0 -11.9

1:11:00 -9.0 -6.2 2.8 -3.0 -11.7

1:16:00 -8.9 -6.2 2.7 -3.0 -11.6

1:30:00 -8.9 -6.2 2.7 -3.0 -11.5

The ΔTcoolant increased by about 0.5℃ and remained constant throughout the reaction. The temperature inside the reactor increased by 7℃ from -10℃ to -3℃. Figure

4.79 shows the plot of time (s) vs temperature (℃) inside the reactor. The plot shows that as the time proceeded, the temperature of the reaction mixture gradually increased from -

10℃ to -7℃ in about 10 minutes. At this time, the flow rate of the AcrCl addition was increased to 0.03 mL/s and temperature inside the reactor gradually increased to -3℃ and remained stable. The addition of AcrCl was completed at 90 min (5400 s). The coolant

220 flow rate during the experiment (8 L/min) was higher than the calculated flow rate (2.567

L/min) required to obtain isothermal conditions. Some heat loss through the reactor glass also took place.

Time (t) vs Temperature Inside the reactor (T) Plot t(s)

-1 0 2000 4000 6000

-3

-5

C) o T( -7

-9

-11

Figure 4.79 Time (s) vs temperature (℃) (inside the reactor) plot during the scale-up.

Figure 4.80 shows the reaction mixture during the addition of AcrCl and Figure

4.81 shows the reaction mixture after the addition of AcrCl.

221

Figure 4.80 Reaction mixture during the Figure 4.81 Reaction mixture after the dropwise addition of AcrCl. complete addition of AcrCl (No stirring).

18 hours after the complete addition of AcrCl, a product sample was taken out from the reactor. DCM and AcrCl were evaporated using a rotary evaporator and the 1H-NMR spectrum of the product was recorded in CD2Cl2 (Figure 4.82). The signals corresponding to the methylene and methine protons of the acrylate functionality were observed at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’) as discussed in Section 4.1. The ratios of the integrals of the methylene and methine protons (1”), (2), (1”’) with respect to that of the aromatic proton of FL (13) is (1:2:2:2), verifying the structure of FL-DA. Methyl and methylene proton signals (a) and (b) appeared at δ 1.22 - 1.15 ppm and δ 3.08-2.97 ppm respectively, due to the presence of TEA hydrochloride salt as a by-product. The ratio of the integrals of proton signals (a):(b) is 9:6 as expected from the structure.

222

Figure 4.82 1H-NMR spectrum of FL-DA before purification.

1 [FL-DA (300 MHz, H NMR, CDCl3): δ 8.07 ((13), dt, J =0.88,7.61 Hz, 1 ,H), δ 7.75-

7.68 ((15, 16), m, 2H), δ 7.25 ((14), d, J =1.17 Hz, 1H), δ 7.20 ((9,20), s, 2H), δ 6.85

((22, 5, 23, 6), s, 4 H), δ 6.70 ((1”), m, 2H), δ 6.30-6.25 ((2), m, 2H), δ 6.10 ((1’”), m, 2

H), δ 3.08-2.97 ((b), m, 6H), δ 1.22 - 1.15 ((a), t, J= 7.32 Hz, 9H)].

Product and salt were removed from the bottom of the reactor (Figure 4.83, Figure

4.84). Addition of THF to the brown colored product solution (Figure 4.85) precipitated the TEA hydrochloride salt which was filtered using Q5 filter paper. The remaining TEA hydrochloride salt was removed by stirring the product solution in THF and silica gel

(Silica gel classic column (63-200 μm)) together for 25 minutes followed by filtration of the silica gel.

223

Figure 4.83 Brown colored Figure 4.84 TEA Figure 4.85 TEA product solution. hydrochloride salt inside hydrochloride salt the reactor. precipitated out after the addition of THF.

Figure 4.86 shows the 1H-NMR spectrum of the purified product where no signals corresponding to the methyl (a) and methylene (b) protons of the TEA hydrochloride salt are present at δ 1.22 - 1.15 ppm and δ 3.08-2.97 ppm, respectively. Hence, no TEA hydrochloride salt is present after the purification using silica gel. Signals corresponding to the methylene and methine protons of the acrylate functionality were observed at 6.70 ppm (1”), 6.30 ppm (2) and 6.10 ppm (1”’). The ratios of the integrals of the methylene and methine protons (1”), (2), (1”’) with respect to that of the aromatic proton of FL (13) is (1:2:2:2), verifying the structure of the FL-DA..

224

Figure 4.86 1H-NMR spectrum of the purified FL-DA.

1 [FL-DA (300 MHz, H NMR, CDCl3): δ 8.07 ((13), dt, J =0.88,7.61 Hz, 1 ,H), δ 7.75-

7.68 ((15, 16), m, 2H), δ 7.25 ((14), d, J =1.17 Hz, 1H), δ 7.20 ((9,20), s, 2H), δ 6.85

((22, 5, 23, 6), s, 4 H), δ 6.70 ((1”), m, 2H), δ 6.30-6.25 ((2), m, 2H), δ 6.10 ((1’”), m, 2

H), δ 3.08-2.97 ((b), m, 6H), δ 1.22 - 1.15 ((a), t, J= 7.32 Hz, 9H)].

The amount of light-yellow colored product obtained after purification and drying was 70.5 g (0.1602 mol), 54.32%.

4.5.3. Costing of the scale-up

The cost of the scale-up reaction was calculated. Table 4.7 and Table 4.8. show the costs of the chemicals and glassware respectively. Table 4.9. shows the total costing of the scale-up. 225

Table 4.7 Costs of the chemicals.

Chemical Supplier Purity, Description Quantity, Quantity

(CAS) Cost (USD) used, Cost

FL Millipore Free acid dye content 95 % 100 g, 98 g

(2321-07-5) Sigma $ 31.20 $30.58

AcrCl Millipore ≥97 %, contains~400 ppm 500 g, 100.26 g

(814-68-6) Sigma phenothiazine as stabilizer $ 476 $95.45

TEA Millipore ≥99 % 500 L, 116.08 g

(121-44-8) Sigma $ 51.70 $17.95

DCM University 99.9 %, Fischer Scientific 4 L, 1.25 L

(75-09-2) ChemStore $ 22.29 $6.97

THF University Certified, Fischer Scientific 20 L, 3 L

(109-99-9) ChemStore $ 126.14 $18.92

Silica gel Dynamic Silica Gel Classic Column, 5 kg, 138 g

Adsorbent 60A, 04667-5, 63-200μ $ 242 $6.68

Inc.

Hexane University Certified, Fischer Scientific 20 L, 8 L

(110-54-3) ChemStore $ 48.31 $19.32

Coolant Gas station Peak, 50/50 prediluted 3.78L, 11.34 L

antifreeze + coolant $ 11.53 $34.59

Total Cost (USD) 230.50

226 Table 4.8 Cost of the glassware.

Glassware (Supplier) Quantity Cost

(USD)

Dropping funnel (Sigma-Aldrich) 1 135.00

Filling Funnel (University ChemStore) 1 9.13

Beakers 4 (600 mL) (University ChemStore) 4 3.29×4 = 13.16

Round bottom flask (500 mL) (University ChemStore) 1 18.00

Measuring cylinders (University ChemStore) 2 18 × 2 = 36.00

Buchner funnel (University ChemStore) 1 29.00

Erlenmeyer flask (University ChemStore) 1 9.00

Total Cost (USD) 249.29

227 Table 4.9 Cost of the scale-up.

Lab supplies/utility Quantity Cost

(Supplier) (USD)

Chemicals 230.50

Glassware 249.29

5L reactor Purchase cost: 7800 USD, reactor life: 10 780.00 years, depreciation: 780 USD/year Filter paper Pack of 100 for 18.06 USD, 0.55

(University ChemStore)

Thermocouple 2 (VWR) 64.00 Amprobe TMD-50 1 (Test Equipment Depot) 60.50 Thermocouple Thermometer K Type 1H-NMR 3 samples 24.00

Mass Spectrometry 1 sample 35.00

NMR tube 1 (University ChemStore) 5.50

Lauda Thermostat RMT 20 Purchase cost: 3500 USD, thermostat life: 350.00 15 years, depreciation: 350 USD/year

Electricity, Nitrogen Approximately 8 hrs, 32.00 4 USD /1hours Labor and supervision Labor cost: 22USD/hr ×32 hrs = 704 USD, 800.00 Supervision: 50 USD/hr ×16 hrs=800 USD Supervision Cost 8 hours, 35 USD /1 hr 280.00

Total Cost to produce 70.5 g FL-DA = 2911.34

228 The total cost for the synthesis of 70.5 g of FL-DA at Dr. Puskas' lab was 2911.34

USD. Cost of 1 g of FL-DA product synthesized in the lab is 41.29 USD. Millipore Sigma cost for 1 g of FL-DA is 338 USD.

229 CHAPTER V

CONCLUSION

This dissertation demonstrated the successful functionalization of fluorescein (FL) and synthesis of two functional folate-targeted FL-poly(ethylene glycol) (PEG) conjugates by chemo-enzymatic method for the potential diagnosis of cancer.

In Section 4.1, we have shown the successful acrylation of FL to synthesize the fluorescein o-acrylate (FL-A) and fluorescein o,o’-diacrylate (FL-DA) using acryloyl chloride (AcrCl) and triethylamine (TEA). For the synthesis of FL-DA, kinetic study using

NMR-750 MHz instrument demonstrated formation of the product in 13 seconds with 2.20 equivalents of AcrCl. FL-DA was purified by precipitating the TEA hydrochloride salt with tetrahydrofuran (THF) and stirring the product solution with silica gel (silica gel classic column, 63-200um). 63.9% yield of the pure FL-DA was obtained which is comparable to the reported yield.56,57,80-82 Hence, our modified synthesis of FL-DA takes

13 seconds and avoids tedious column chromatography for the purification. FL-A was synthesized in the same manner with 1.1 equivalents of AcrCl and purified by silica gel column chromatography with solvent system of 50 v% hexane-50 v% diethyl ether. After purification, 52.49% of the pure FL-A was obtained. FL-DA and FL-A both are the main precursors of PEG-based conjugate as the acrylate moiety allows the CALB-catalyzed

Michael addition of thiol and amine. Synthesis of previous FA-FL-HN-PEG-NH-FL-FA compound was improved by replacing the starting material FL-DA by FL-A which

230 improved selectivity of the Michael addition with amine and thiol-functionalized PEG.

Structures of the products were confirmed by 1H-NMR and 13C-NMR.

In Section 4.2, FL was successfully PEGylated by the Candida antarctica lipase B

(CALB)-catalyzed Michael addition. Michael addition of PEG-diamine (H2N-PEG-NH2,

Mn=2,000 g/mol, Đ ≤1.2) to FL-A yielded 86.58% of FL-NH-PEG-NH-FL. However, to avoid interference of the nucleophilic secondary amine of FL-NH-PEG-NH-FL during the acrylation, H2N-PEG-NH2 was replaced by tetraethylene glycol-dithiol (HS-TEG-SH) and

PEG-dithiol (HS-PEG-SH) which were synthesized by the CALB-catalyzed transesterification of methyl 3-mercaptopropionate with PEG.22 CALB-catalyzed “Click”

Michael addition between FL-A and HS-PEG-SH (Mn =1160 g/mol, Đ=1.14 and Mn =2200 g/mol, Đ=1.09), HS-TEG-SH (FW= 370.48 g/mol) completed in 2 minute and 1 minute respectively at 52℃. Reaction of HS-PEG1000-SH with FL-A without CALB catalysis did not go to completion even after conducting the reaction for 18 hours at 52℃ but completed in 2 minutes when CALB was added. Progress of the reaction was monitored by 1H-NMR, which shows disappearance of the methine and methylene proton signals of the FL-A at

6.52 ppm, 6.48-6.35 ppm, 6.17 ppm. Structures of the products were also confirmed using

13C-NMR.

In Section 4.3, FL was successfully PEGylated by lithium technology where fluorescin or dihydro-FL was lithiated at ‘COOH’ followed by the addition of brominated

PEG compound (Br-PEG-OMe) into the lithium salt of FL at 25℃. 1H-NMR analysis demonstrated formation of the desired product. Mass spectrometry confirmed presence of

231 the PEGylated FL along with the unreacted PEG compounds. Further acrylation of the product was not performed due to the presence of the mixture.

In Section 4.4, tetrafunctional FL compounds were successfully synthesized by the aza-Michael addition reactions between FL-DA and secondary amines such as diallylamine (DAA), diethyl iminodiacetate (DIDA) and newly synthesized dihydroxy secondary amines. Reactions were studied both with and without CALB-catalysis. For the sterically hindered molecule, such as DIDA, CALB-catalysis reduced the reaction time by three times than the reaction without CALB catalysis. Michael addition with DAA was extremely fast in DMSO and product was obtained in 1 minute at room temperature even without CALB-catalysis. The reaction in chloroform with CALB catalysis did not complete even after 10 minutes of the reaction time. The reaction was also conducted in the same manner with diethanolamine (DEA) and same results were obtained in DMSO and chloroform both with and without CALB catalysis. It was found that the CALB catalysis did not have any effect on the reactions with DAA and DEA in DMSO. However, due to the less stability of the product of the Michael addition between FL-DA and DEA, product hydrolyzed back to FL. This was possibly due to the shorter alkyl chain of the DEA, resulting in neighboring group participation. The problem was solved by synthesizing two new longer alkyl chain secondary amines with two hydroxy groups by UV-mediated thiol- ene click reactions between DAA and 4-mercapto-1-butanol and 9-mercapto-1-nonanol respectively. These newly synthesized dihydroxy amines were reacted with FL-DA and the product obtained was stable and did not hydrolyze to FL. All the products were

232 characterized by 1H-NMR, 13C-NMR and mass spectrometry. The tetrafunctional-FL compounds are suitable precursors to develop the multifunctional polymeric conjugates.

In the last Section 4.5, FL-DA synthesis was successfully scaled up in a 5L capacity jacketed reactor to ~ 100 g scale by reacting FL with AcrCl using TEA. Some heat loss through the reactor glass also takes place because of which complete isothermal conditions were not obtained, but it was close (ΔTcoolant: = 0.5℃). Purification of the product was conducted using silica gel and the amount of light-yellow colored product obtained was

70.5 g (0.1602 mol), 54.32%. Cost of 1 g of FL-DA product synthesized at Dr. Puskas’ lab was calculated as 41.29 USD.

In summary, this dissertation primarily discusses the chemistry and functionalization of FL followed by the Michael addition reaction of HS-PEG-SH to successfully develop the difunctional folate-targeted conjugate for the diagnostic applications.

233 REFERENCES

1. TIME, https://time.com/5495804/cancer-death-decline/ (last accessed on August 6, 2019)

2. Lu, Y.; Low, P. S. Folate-mediated Delivery of Macromolecular Anticancer Therapeutic Agents. Adv. Drug Deliv. Rev. 2012, 64 (SUPPL.), 342-352.

3. Yang J., Vlashi E., Low P. Folate-Linked Drugs for the Treatment of Cancer and Inflammatory Diseases. In Water Soluble Vitamins.: Subcellular Biochemistry, Stanger O., Eds.: Springer, Dordrecht, 2012,vol. 56; pp 163-179.

4. Leamon, C. P.; Low, P.S. Delivery of macromolecules into living cells: A method that exploits folate receptor endocytosis. Proc. Nadl. Acad. Sci. USA. Cell Biology. 1991, 88, 5572-5576.

5. Zwicke, G. L.; Mansoori, G. A.; Jeffery, C. J. Utilizing the Folate Receptor for Active Targeting of Cancer Nanotherapeutics. Nano Rev.2012, 3(1), 18496.

6. Liu, F.; Deng, D.; Chen, X.; Qian, Z.; Achilefu, S.; Gu, Y. Folate-Polyethylene Glycol Conjugated Near-Infrared Fluorescence Probe with High Targeting Affinity and Sensitivity for In Vivo Early Tumor Diagnosis. Mol. Imaging Biol. 2010, 12 (6), 595–607.

7. Wang, S.; Luo, J.; Lantrip, D. A.; Waters, D. J.; Mathias, C. J.; Green, M. A.; Fuchs, P. L.; Low, P. S. Design and Synthesis of [111In]DTPA−Folate for Use as a Tumor- Targeted Radiopharmaceutical. Bioconjugate Chem. 1997, 8(5), 673–679.

8. Leamon, C.; Low, P. Selective Targeting of Malignant Cells with Cytotoxin-Folate Conjugates. J. Drug Target. 1994, 2(2), 101–112.

9. Leamon, C.; Reddy, J.; Vlahov, I.; Vetzel, M.; Parker, N.; Nicoson, J.; Xu, L.; Westrick, E. Synthesis and Biological Evaluation of EC72: A New Folate- Targeted Chemotherapeutic. Bioconjugate Chem. 2005, 16 (4), 803-811.

10. Shiokawa, T. Effect of Polyethylene Glycol Linker Chain Length of Folate-Linked Microemulsions Loading Aclacinomycin A on Targeting Ability And Antitumor Effect In Vitro and In Vivo. Clin. Cancer Res. 2005, 11 (5), 2018-2025.

11. Hong, S.; Leroueil, P. R.; Majoros, I. J.; Orr, B. G.; Baker, J. R.; Banaszak Holl, M. M. The Binding Avidity of a Nanoparticle-based Multivalent Targeted Drug Delivery Platform. Chem. Biol. 2007, 14(1), 107–115. 234 12. Achilefu, S.; Jimenez, H. N.; Dorshow, R. B.; Bugaj, J. E.; Webb, E. G.; Wilhelm, R. R.; Rajagopalan, R.; Johler, J.; Erion, J. L. Synthesis, In Vitro Receptor Binding, and In Vivo Evaluation of Fluorescein and Carbocyanine Peptide-based Optical Contrast Agents. J. Med. Chem. 2002, 45, 2003.

13. Robertson, T. A.; Bunel, F.; Roberts, M. S. Fluorescein Derivatives in Intravital Fluorescence Imaging. Cells. 2013, 2, 591.

14. Becker, V., von Delius, S.; Bajbouj, M.; Karagianni, A.; Schmid, R. M.; Meining, A. Intravenous Application of Fluorescein for Confocal Laser Scanning Microscopy: Evaluation of Contrast Dynamics and Image Quality with Increasing Injection-to-Imaging Time. Gastrointest. Endosc. 2008, 68 (2), 319.

15. Puskas, J.E.; Castano, M.; Mulay, P.; Dudipala, V.; Wesdemiotis, C. Method for the Synthesis of γ-PEGylated Folic Acid and its Fluorescein-Labeled derivative. Macromolecules 2018, 51 (22), 9069–9077.

16. Knop, K.; Hoogenboom, R.; Fischer, D.; Schubert, U.S. Poly(ethylene glycol) in Drug Delivery: Pros and Cons as well as Potential Alternatives. Angew. Chem. Int. Ed. 2010, 49, 6288– 6308.

17. Koirala, N.; Das, D.; Fayazzadeh, E.; Sen, S.; Mcclain, A.; Puskas, J. E.; Drazba, J. A.; Mclennan, G. Folic Acid Conjugated Polymeric Drug Delivery Vehicle for Targeted Cancer Detection in Hepatocellular Carcinoma. J. Biomed. Mater. Res. Part A. 2019, 107 (11), 2522–2535.

18. Puskas, J.E; Seo, K.S.; Sen, M.Y. Green Polymer Chemistry: Precision Synthesis of Novel Multifunctional Poly(ethylene glycol)s using Enzymatic Catalysis. Eur. Polym. J. 2011, 47, 524–534.

19. Puskas, J. E.; Sen, M. Y.; Seo, K. S., Green Polymer Chemistry using Nature's Catalysts, Enzymes. J. Polym. Sci. Part A: Polym. Chem. 2009, 47, 2959-297.

20. Puskas, J. E.; Sen, M. Y.; Kasper, J. R., Green Polymer Chemistry: Telechelic Poly(ethylene glycol)s via Enzymatic Catalysis. J. Polym. Sci. Part A: Polym. Chem. 2008, 46, 3024-3028.

21. Sen, M. Y.; Puskas, J. E.; Ummadisetty, S.; Kennedy, J. P., Green Polymer Chemistry: II. Enzymatic Synthesis of Methacrylate-terminated Polyisobutylenes. Macromol. Rapid Commun. 2008, 29, 1598-1602.

22. Mulay, P.; Shrikhande, G.; Puskas, J. E. Synthesis of Mono- and Dithiols of Tetraethylene Glycol and Poly(ethylene glycol)s via Enzyme Catalysis. Catalysts 2019, 9, 228.

23. MedicalNewsToday. https://www.medicalnewstoday.com/articles/323648.php 235 (Accessed August 27, 2019)

24. American Cancer Society. https://www.cancer.org/latest-news/facts-and-figures- 2019.html (Accessed August 27, 2019)

25. Siegel, R.L.; Miller, K.D.; Jemal, A. Cancer Statistics, 2019. CA Cancer J. Clin. 2019, 69, 7–34.

26. National Breast Cancer Foundation Inc. https://www.nationalbreastcancer.org/breast-cancer-facts (Accessed August 27, 2019)

27. Russell-Jones, G.; McTavish, K.; McEwan, J.; Rice, J.; Nowotnik, D. Vitamin- mediated Targeting as a Potential Mechanism to Increase Drug Uptake by Tumours. J. Inorg. Biochem. 2004, 98 (10 SPEC. ISS.), 1625–1633.

28. Goldhirsch, A., Wood W.C.; Coates, A. S.; Gelber, R. D.; Thurlimann, B.; Senn, H-J. Strategies for Subtypes-Dealing with the Diversity of Breast Cancer: Highlights of the St. Gallen International Expert Consensus on the Primary Therapy of Early Breast Cancer 2011. Ann. Oncol. 2011, 22, 1736–1747.

29. Necela, B.M.; Crozier J.A., Andorfer C.A., Lewis-Tuffin L.; Kachergus J.M.; Geiger, X.J.; Kalari, K.R.; Serie, D.J.; Sun, Z.; Moreno-Aspita, A.; O’Shannessy, D.J.; Maltzman, J.D.; McCullough, A.N.; Pockaj, B.A.; Cunliffe, H.E.; Ballman, K.V.; Thompson, A.; Perez, E.A. Folate Receptor-α(FOLR1) Expression and Function in Triple Negative Tumors. PLoS ONE. 2015, 10(4), 1-13.

30. Zhang, Z.; Wang, J.; Tacha, D. E.; Li, P.; Bremer, R. E.; Chen, H.; Wei, B.; Xiao, X.; Da, J.; Skinner, K.; Hicks, D. G.; Bu, H.; Tang, P. Folate Receptor α Associated with Triple-Negative Breast Cancer and Poor Prognosis. Arch. Pathol. Lab. Med. 2014, 138(7), 890–895.

31. Salazar. M. D; Ratnam, M. The Folate Receptor: What Does it Promise in Tissue- targeted Therapeutics? Cancer Metastasis Rev. 2007, 26,141–152.

32. Basal, E.; Eghbali-Fatourechi, G.Z.; Kalli, K.R.; Hartmann, L.C.; Goodman, K.M.; Goode, E.L.; Kamen, B.A.; Low, P.S.; Knutson, K.L. Functional folate receptor alpha is elevated in the blood of ovarian cancer patients. PLoS One. 2009, 4(7), e6292.

33. Low, P. S.; Kularatne, S. A. Folate-targeted Therapeutic and Imaging Agents for Cancer. Curr. Opin. Chem. Biol. 2009, 13 (3), 256–262.

34. Ghaghada, K. B.; Saul, J.; Natarajan, J. V.; Bellamkonda, R. V.; Annapragada, A. V. Folate Targeting of Drug Carriers: A Mathematical Model. Journal of Controlled Release 2005, 104 (1), 113–128. 236 35. O’Shannessy, D. J.; Somers, E. B.; Maltzman, J.; Smale, R.; Fu, Y.-S. Folate Receptor Alpha (FRA) Expression in Breast Cancer: Identification of a New Molecular Subtype and Association with Triple Negative Disease. SpringerPlus. 2012, 1-22.

36. Oregon State University. https://lpi.oregonstate.edu/mic/vitamins/folate (Last accessed on May 03, 2019)

37. Rezk, B. M.; Haenen, G. R.; Vijgh, W. J. V. D.; Bast, A. Tetrahydrofolate and 5- Methyltetrahydrofolate are Folates with High Antioxidant Activity. Identification of the Antioxidant Pharmacophore. FEBS Letters 2003, 555(3), 601–605.

38. Kelemen, L. E. The Role of Folate Receptor α in Cancer Development, Progression and Treatment: Cause, Consequence or Innocent Bystander? Int. J. Cancer 2006, 119 (2), 243–250.

39. Srinivasarao, M.; Low, P.S. Ligand-Targeted Drug Delivery. Chem. Rev. 2017, 117, 12133−12164.

40. Clinical trials, A service of the U.S. National Institutes of Health. https://clinicaltrials.gov/ct2/results?cond=&term=EC17&cntry=&state=&city=&d ist= (last accessed on May 1, 2019)

41. Frigerio, B.; Bizzoni, C.; Jansen, G.; Leamon, C. P.; Peters, G. J.; Low, P. S.; Matherly, L. H.; Figini, M. Folate Receptors and Transporters: Biological Role and Diagnostic/Therapeutic Targets in Cancer and Other Diseases. Journal of Experimental & Clinical Cancer Research 2019, 38 (1), 1-12.

42. Jesus, E.D.; Keating, J.J.; Kularatne, S.A.; Jiang, J.; Judy, R.; Predina, J.; Nie, S.; Low, P.; Singhal, S. Comparison of Folate Receptor Targeted Optical Contrast Agents for Intraoperative Molecular Imaging. Int. J. Mol. Imaging. 2015, 1-10.

43. Clinical trials, A service of the U.S. National Institutes of Health. https://clinicaltrials.gov/ct2/results?term=folate-fluorescein&Search=Search (Accessed April 26, 2019)

44. Tummers, Q. R.; Hoogstins, C. E.; Gaarenstroom, K. N.; Kroon, C. D. D.; Poelgeest, M. I. V.; Vuyk, J.; Bosse, T.; Smit, V. T.; Velde, C. J. V. D.; Cohen, A. F.; Low, P. S.; Burggraaf, J.; Vahrmeijer, A. L. Intraoperative Imaging of Folate Receptor Alpha Positive Ovarian and Breast Cancer using the Tumor Specific Agent EC17. Oncotarget. 2016, 7(22), 32144-32155.

45. Guzzo, T.; Jiang, J.; Keating, J.; DeJesus, E.; Judy, R.; Nie, S.; Low, P.; Lal, P.; Singhal, S. Intraoperative Molecular Diagnostic Imaging Can Identify Renal Cell Carcinoma. J. Urol. 2016, 195 (3), 748-755.

237 46. Clinical trials, A service of the U.S. National Institutes of Health. https://clinicaltrials.gov/ct2/show/NCT00485563 (last accessed on May 1, 2019)

47. Clinical trials, A service of the U.S. National Institutes of Health. https://clinicaltrials.gov/ct2/show/NCT01996072 (last accessed on May 1, 2019)

48. Kim, S.-L.; Jeong, H.-J.; Kim, E.-M.; Lee, C.-M.; Kwon, T.-H.; Sohn, M.-H. Folate Receptor Targeted Imaging Using Poly (Ethylene Glycol)-Folate: In Vitro and In Vivo Studies. J. Korean Med. Sci. 2007, 22(3), 405.

49. Kukoska-Latallo, J. F. Candido, K.A.; Cao,Z.; Nigavekar, S.S.; Majoros, I.J.; Thomas, T.P.; Balogh, L.P.; Khan, M.K.; Baker, J.R. Jr. Nanoparticle Targeting of Anticancer Drug Improves Therapeutic Response in Animal Model of Human Epithelial Cancer. Cancer Res. 2005, 65 (12), 5317–5324.

50. Baker, J. R.; Jr, B. B. W.; Thomas, T. P. Nanotechnology in Clinical and Translational Research. Clinical and Translational Science 2009, 123–135.

51. Attie, A.D.; Raines, R.T. Analysis of Receptor–Ligand Interactions. J. Chem. Educ. 1995, 72(2), 119–124.

52. Malvern Panalytical. https://www.malvernpanalytical.com/en/products/measurement-type/binding- affinity (Accessed April 27, 2019)

53. Waddell, J. N.; Mullen, D. G.; Orr, B. G.; Holl, M. M. B.; Sander, L. Origin of Broad Polydispersion in Functionalized Dendrimers and its Effects on Cancer-Cell Binding Affinity. M. Phys Rev. E. 2010, 82(3), 1-4.

54. Srinivasarao, M.; Galliford, C. V.; Low, P. S. Principles in the Design of Ligand- targeted Cancer Therapeutics and Imaging Agents Nat. Rev. Drug Discov. 2015, 14(3), 203–219.

55. Sen, S.; Puskas, J. Green Polymer Chemistry: Enzyme Catalysis for Polymer Functionalization. Molecules. 2015, 20 (5), 9358–9379.

56. Wang, H.; Zhou, G.; Gai, H.; Chen, X.; A Fluorescein-based Probe with High Selectivity to Cysteine Over Homocysteine and Glutathione. Chem. Commun. 2012, 48, 8341.

57. Wang, B.; Guan X.; Hu, Y.; Su, Z. Synthesis and Photophysical Behavior of a Water-Soluble Fluorescein-bearing Polymer for Fe3+ Ion Sensing J. Polym. Res. 2008, 15, 427.

58. Ueno, T.; Urano, Y.; Setsukinai, K.; Takakusa, H.; Kojima, H.; Kikuchi, K.; Ohkubo, K.; Fukuzumi, S.; Nagano, T. Rational Principles for Modulating 238 Fluorescence Properties of Fluorescein. J. Am. Chem. Soc. 2004, 126, 14079.

59. Sjoback, R.; Nygren J.; Kubista M. Absorption and Fluorescence Properties of Fluorescein Spectrochimica Acta Part A, 1995, 51, L7.

60. Lebed, A.V.; Biryukov, A.V.; Mchedlov-Petrossyan, N. O. A Quantum-Chemical Study of Tautomeric Equilibria of Fluorescein Dyes in DMSO. Chem. Heterocycl. Compd. 2014, 50 (3), 336-348.

61. Arhangelskis, M.; Eddleston, M.D.; Reid, D.G.; Day, G.M.; Bučar. D.; Morris, A.J.; Jones, W. Rationalization of the Color Properties of Fluorescein in the Solid State: A Combined Computational and Experimental Study. Chem. Eur. J. 2016, 22, 10065-10073.

62. Chemicalland21. http://www.chemicalland21.com/specialtychem/finechem/FLUORESCEIN%20S ODIUM.htm (Accessed March 24, 2019)

63. Hempel, S.L.; Buettner, G.R.; O’Malley, Y.Q.; Wessels, D.A.; Flaherty, D.M. Dihydrofluorescein Diacetate is Superior for Detecting Intracellular Oxidants: Comparison with 2’,7’-Dichlorodihydrofluorescein Diacetate, 5(and 6)- CARBOXY-2’,7’-Dichlorodihydrofluorescein Diacetate, and Dihydrorhodamine 123. Free Radic Biol Med. 1999, 27(1-2), 146–159.

64. Negre-Salvayre, A.; Auge, N.; Duval, C.; Robbesyn, F.; Thiers, J-C.; Nazzal, D.; Benoist, H.; Salvayre, R. Detection of Reactive Intracellular Oxygen Species in Cultured Cells using Fluorescent Probes. In Redox Cell Biology and Genetics Part 1; Sen, C.K.; Packer, L. Ed.; Academic Press: An Imprint of Elsevier Science, USA. 2002, pp 62-64.

65. Panchompoo, J.; Aldous, L.; Baker M.; Wallace, M.I.; Compton, R.G. One-step Synthesis of Fluorescein Modified Nano-carbon for Pd(II) Detection via Fluorescence Quenching. Analyst 2012, 137, 2054–2062.

66. Benson, R. C.; Meyer, R. A.; Zaruba, M. E.; Mckhann, G. M. Cellular Autofluorescence - Is It Due to Flavins? J. Histochem. Cytochem. 1979, 27(1), 44– 48.

67. Rich, R.M.; Stankowska, D.L.; Maliwal, B.P.; Sørensen, T.J.; Laursen, B.W.; Krishnamoorthy, R.R., Gryczynski, Z.; Borejdo. J.; Gryczynski, I.; Fudala, R. Elimination of Autofluorescence Background from Fluorescence Tissue Images by Use of Time-gated Detection and the AzaDiOxaTriAngulenium (ADOTA) Fluorophore. Anal Bioanal Chem. 2013, 405(6), 2065–2075.

68. Mohan, P.; Rapoport, N. Doxorubicin as a Molecular Nanotheranostic Agent: Effect of Doxorubicin Encapsulation in Micelles or Nanoemulsions on the 239 Ultrasound-mediated Intracellular Delivery and Nuclear Trafficking. Mol Pharm. 2010, 7(6), 1959–1973.

69. Wylon, K.; Dölle, S.; Worm, M. Polyethylene Glycol as a Cause of Anaphylaxis. Allergy Asthma Clin Immunol. 2016, 12(67),1-3.

70. Iyer, A.K.; Khaled, G.; Fang, J.; Maeda, H. Exploiting the Enhanced Permeability and Retention Effect for Tumor Targeting. Drug Discov. Today. 2006, 11 (17/18), 812-818.

71. Quanta Biodesign Limited. https://www.quantabiodesign.com/what-is-dpeg/ (Accessed May 2, 2019)

72. Abandansari, H. S.; Ghanian, M. H.; Varzideh, F.; Mahmoudi, E.; Rajabi, S.; Taheri, P.; Nabid, M. R.; Baharvand, H. In Situ Formation of Interpenetrating Polymer Network Using Sequential Thermal and Click Crosslinking for Enhanced Retention of Transplanted Cells. Biomaterials 2018, 170, 12–25.

73. Iijima, M.; Ulkoski, D.; Sakuma, S.; Matsukuma, D.; Nishiyama, N.; Otsuka, H.; Scholz, C. Synthesis of PEGylated Poly(Amino Acid) Pentablock Copolymers and Their Self-Assembly. Polymer International 2016, 65 (10), 1132–1141.

74. Liu, C.; Ewert, K. K.; Wonder, E.; Kohl, P.; Li, Y.; Qiao, W.; Safinya, C. R. Reversible Control of Spacing in Charged Lamellar Membrane Hydrogels by Hydrophobically Mediated Tethering with Symmetric and Asymmetric Double- End-Anchored Poly(Ethylene Glycol)s. ACS Applied Materials & Interfaces 2018, 10 (50), 44152–44162.

75. Shibasaki, Y.; Mori, T.; Fujimori, A.; Jikei, M.; Sawada, H.; Oishi, Y. Poly(Amide– Ether) Thermoplastic Elastomers Based on Monodisperse Aromatic Amide Hard Segments as Shape-Memory and Moisture-Responsive Materials. Macromolecules 2018, 51 (23), 9430–9441.

76. Zalipsky, S. Functionalized Poly(Ethylene Glycols) for Preparation of Biologically Relevant Conjugates. Bioconjugate Chemistry 1995, 6 (2), 150–165.

77. Mongondry, P.; Bonnans-Plaisance, C.; Jean, M.; Tassin, J. F. Mild Synthesis of Amino-Poly(Ethylene Glycol)s. Application to Steric Stabilization of Clays. Macromol. Rapid Commun. 2003, 24 (11), 681–685.

78. Gravert, D. J.; Janda, K. D. Organic Synthesis on Soluble Polymer Supports: Liquid-Phase Methodologies. Chem. Rev. 1997, 97 (2), 489–510.

79. Gil, Y.M.C., Green Polymer Chemistry: The Role of Candida antarctica lipase B in Polymer Functionalization. Doctoral Dissertation, The University of Akron, Akron, OH, USA, 2014. 240 80. Ma, D. H.; Kim, D.; Akisawa, T.; Lee, K-H.; Kim, K-T.; Ahn, K. H. An FITC- BODIPY FRET Couple: Application to Selective, Ratiometric Detection and Bioimaging of Cysteine. Chem. Asian J. 2015, 10, 894-902.

81. Wang, L.; Zhou, Q.; Zhu, B.; Yan, L.; Ma, Z.; Du.; Zhang, X. A Colorimetric and Fluorescent Chemodosimeter for Discriminative and Simultaneous Quantification of Cysteine and Homocysteine. Dyes and Pigments. 2012, 95, 275-279.

82. Lu, H.; Zhang, H.; Chen, J.; Zhang, J.; Liu, R.; Sun, H.; Zhao, Y.; Chai, Z.; Hu, Y. A Thiol Fluorescent Probe Reveals the Intricate Modulation of Cysteine's Reactivity by Cu(II). Talanta. 2016, 146, 477–482.

83. Mather, B. D.; Viswanathan, K.; Miller, K. M.; Long, T. E. Michael Addition Reactions in Macromolecular Design for Emerging Technologies. Prog. Polym. Sci. 2006, 31 (5), 487–531.

84. Desmet, G.B.; D’hooge, D.R.; Omurtag, P.S.; Espeel, P.; Marin, G.B.; Du Prez, F.E.; Reyniers, M-R. Quantitative First-Principles Kinetic Modeling of the Aza- Michael Addition to Acrylates in Polar Aprotic Solvents. J. Org. Chem. 2016, 81, 12291−12302.

85. Nair, D. P.; Podgórski, M.; Chatani, S.; Gong, T.; Xi, W.; Fenoli, C. R.; Bowman, C. N. The Thiol-Michael Addition Click Reaction: A Powerful and Widely Used Tool in Materials Chemistry. Chem. Mater. 2013, 26 (1), 724–744.

86. Torre, O.; Alfonso, I.; Gotor, V. Lipase catalysed Michael Addition of Secondary Amines to Acrylonitrile. Chem. Commun. 2004, 1724-1725.

87. Dhake, K.P.; Tambade, P.J.; Singhal, R.S.; Bhanage, B.M. Promiscuous Candida antarctica Lipase B-catalyzed Synthesis of Beta-amino Esters via Aza-Michael Addition of Amines to Acrylates. Tetrahedron Lett. 2010, 51, 4455–4458.

88. Puskas, J. E.; Sen, M. Y. Process of Preparing Functionalized Polymers via Enzymatic Catalysis. US 8,710,156 B2, April 29, 2014.

89. Bequette, B. W. From Pilot Plant to Manufacturing: Effect of Scale‐up on Operation of Jacketed Reactors. In Pharmaceutical Manufacturing Handbook, S. C. Gad., Ed., 2010; pp 1-21.

90. Dream, R.F. Heat transfer in agitated jacketed vessels. Chem. Eng., 1999, 106(1), 90-96.

91. Trivedi, B.; Shah, B.H. Scale up of API. Int. J. Sci. Eng. Technol.2012, 1(2), 190- 196.

92. Perry, R. H.; Green, D.W. Perry's Chemical Engineers' Handbook, 7th ed.; 241 McGraw-Hill: New York, 1999; pp 349, 351.

93. Chemeo, High Quality Chemical Properties. https://www.chemeo.com/cid/58-741- 5/2-Propenoyl%20chloride (Last accessed on January 31, 2017)

94. National Institute of Standard and Technology, US Department of Commerce. http://webbook.nist.gov/cgi/cbook.cgi?ID=C121448&Mask=1#Thermo-Gas (Last accessed on January 31, 2017)

95. Derakhshan, B. M.; Finch, A.; Gates, P.N.; Stephens, M. Standard enthalpies of formation and lattice energies of alkylammonium halides. Part 2. Ethylammonium halides. J. Chem. Soc. Dalton Trans. 1984, 601-603.

96. Dymek, C. J. Jr.; Stewart, J.J.P.; Storch, D.M. MOPAC Workshop Manual. Air Force Systems Command United States AIR Force; 1988.

97. Hanhong. http://www.hanhonggroup.com/nmr/nmr_en/B17462.html (Accessed June 20, 2019).

98. ChemistryLibreTexts.https://chem.libretexts.org/Bookshelves/Organic_Chemistry /Map%3A_Organic_Chemistry_(Bruice)/14%3A_NMR_Spectroscopy/14.10_The _Splitting_of_the_Signals_is_Described_by_the_N___1_Rule. (Accessed March 24, 2019).

99. Pellach, M.; Margel, S. The Encapsulation of an Amphiphile into Polystyrene Microspheres of Narrow Size Distribution. Chem. Cent. J. 2011, 5:78

100. Mishra, P.; Nayak, B.; Dey, R.K. PEGylation in anti-cancer therapy: An overview. Asian J. Pharm. Sci. 2016, 11, 337–348.

101. http://umich.edu/~chemh215/W13HTML/SSG5/ssg5.4/n-Butyllithium.html (Accessed August 30, 2019).

102. Scifinder. https://scifinder.cas.org/scifinder/view/scifinder/scifinderExplore.jsf (Accessed Mar 23, 2019)

103. PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/Diallylamine#section=Top. (Accessed Feb 23, 2019)

104. El-Nahhal, I.M.; Zaggout F.R.; Nassar, M.A. Synthesis, Characterization and Applications of Immobilized Iminodiacetic Acid-Modified Silica. Journal of Sol- Gel Sci and Technol. 2003, 28, 255–265.

105. PubChem. https://pubchem.ncbi.nlm.nih.gov/compound/diethanolamine. (Accessed Feb 23, 2019) 242

106. Compton, D.A.C. Values for the gas-phase thermodynamic functions of conjugated compounds existing as a mixture of conformers. J. Chem. Soc., Perkin Trans. 2. 1977, 1307-1311.

107. Roux, G.; Roberts, D.; Perron, G.; Desnoyers, J.E. Microheterogeneity in aqueous- organic solutions: heat capacities, volumes and expansibilities of some alcohols, aminoalcohol and tertiary amines in water. J. Solution Chem., 1980, 9(9), 629-647.

108. Chorażewski, M.; Troncoso, J.; Jacquemin, J. Thermodynamic Properties of Dichloromethane, Bromochloromethane, and Dibromomethane under Elevated Pressure: Experimental Results and SAFT-VR Mie Predictions. Ind. Eng. Chem. Res. 2015, 54(2), 720–730.

243 APPENDIX A

LIST OF ABBREVIATIONS

• 5,6,7,8-THF : 5,6,7,8-Tetrahydro folic acid

• 5‐MTHF : 5‐Methyltetrahydrofolate

• 6-TAMRA : 6 - Carboxytetramethylrhodamine

• AcrCl : Acryloyl chloride

• Acryl-S-PEG-S-FL : Acrylated fluorescein-labeled poly(ethylene glycol) dithiol

• Ala : Alanine

• BOC : tert-Butyloxycarbonyl

• Br-S-S-Br : Bromine-functionalized disulfide compound

• BSA : Bovine serum albumin

• CALB : Candida antarctica lipase B

• CD2Cl2 : Methylene chloride

• CD3OD : Deuterated methanol

• CDCl3 : Deuterated chloroform

• CURL : Compartment of uncoupling of receptor and ligand

• DAA : Diallylamine

• DCC : 1,3-Dicyclohexylcarbodiimide

• DCM : Dichloromethane

244 • DEA : Diethanolamine

• DHF : 7,8-Dihydro folic acid

• DIDA: Diethyl iminodiacetate

• DMF : Dimethylformamide

• DMPA : 2,2-Dimethoxy-2-phenylacetophenone

• DMSO d6 : Dimethyl sulfoxide

• DOX : Doxorubicin

• dPEG : Discrete poly(ethylene glycol)

• DTT : Dithiothreitol

• EBV : Ethyl 5-bromovalerate

• EDC : 1-Ethyl-3- (3-(dimethylamino)propyl)carbodiimide

• EPR : Enhanced Permeability and Retention

• ER : Estrogen

• ESI : Electrospray ionization

• FA : Folic acid

• FA-SH : Thiol-functionalized FA

• FDA : Food and Drug Administration

• FITC : Fluorescein isothiocyanate

• FL : Fluorescein

• FL-A : Fluorescein o,-acrylate

• FL-DA : Fluorescein o,o’-diacrylate

• FL-S-PEG-S-FL : PEGylated Fluorescein 245 • FR : Folate receptors

• Gln : Glutamine

• HER2 : Human Epidermal Growth Factor Receptor 2

• HOBt : Hydroxybenzotriazole

• HPLC : High performance liquid chromatography

• HRP : Horseradish peroxidase

• HS-PEG-SH : Poly(ethylene glycol) dithiol

• HS-TEG-OH : Tetraethylene glycol-monothiol

• HS-TEG-SH : Tetraethylene glycol-dithiol

• IA : Intra-arterial

• IgG : Immunoglobulin G

• IV : Intravenous

• MALDI : Matrix-assisted laser desorption/ionization

• MOPAC : Molecular Orbital PACkage

• MPSH : Methyl 3-mercaptopropionate

• MTX : Methotrexate

• n-BuLi : n-Butyllithium

• NH2-PEG-NH2 : Poly(ethylene glycol) diamine

• NHS : N-Hydroxysuccinimide

• NMR : Nuclear Magnetic Resonance

• PAMAM : Poly(amidoamine)

• PDC : Polymeric Drug Conjugates 246 • PEG : Poly(ethylene glycol)

• PEG-OMe : Poly(ethylene glycol) methyl ether

• PIB : Polyisobutylene

• PR : Progesterone

• RFCs : Reduced Folate Carriers

• RME: Receptor Mediated Endocytosis

• SMDC : Small Molecule-Drug or Diagnostic agent Conjugates

• TEA : Triethylamine

• TEG : Tetraethylene glycol

• THF : Tetrahydrofuran

• TLC : Thin layer chromatography

• TNBC : Triple-negative breast cancer

• Ts : p-Toluenesulfonyl

• UV : Ultra Violet

247