Citation
Valproic Acid-Dextran Conjugate

Material Information

Title:
Valproic Acid-Dextran Conjugate Synthesis, Characterization and in Vitro Hydrolysis Studies
Creator:
Hassan, A. M. Mahbub
Place of Publication:
[Gainesville, Fla.]
Publisher:
University of Florida
Publication Date:
Language:
english
Physical Description:
1 online resource (142 p.)

Thesis/Dissertation Information

Degree:
Doctorate ( Ph.D.)
Degree Grantor:
University of Florida
Degree Disciplines:
Pharmaceutical Sciences
Pharmaceutics
Committee Chair:
Hochhaus, Guenther
Committee Co-Chair:
Hughes, Jeffrey
Committee Members:
Derendorf, Hartmut C
Batich, Christopher D
Graduation Date:
12/15/2012

Subjects

Subjects / Keywords:
Dextrans ( jstor )
Epilepsy ( jstor )
Glycols ( jstor )
Hydrolysis ( jstor )
Phosphates ( jstor )
Prodrugs ( jstor )
Protons ( jstor )
Rats ( jstor )
Seizures ( jstor )
Solubility ( jstor )
Pharmaceutics -- Dissertations, Academic -- UF
vpa
Genre:
Electronic Thesis or Dissertation
bibliography ( marcgt )
theses ( marcgt )
government publication (state, provincial, terriorial, dependent) ( marcgt )
Pharmaceutical Sciences thesis, Ph.D.

Notes

Abstract:
The objective of this study was to synthesize valproic acid-dextran conjugate which would be able to release the drug via chemical hydrolysis over an extended period and also to synthesize valproic acid-linker-dextran conjugate which would provide in situ activation to release the drug via enzymatic hydrolysis. Valproic acid (VPA) was conjugated directly to dextrans of molecular weights (Mw) 10,000 and 70,000 using carbonyldiimidazole. To synthesize valproic acid-linker-dextran conjugate, VPA was initially conjugated to linkers (triethylene glycol or pentaethylene glycol) using dicyclohexylcarbodiimide and subsequently, conjugated to dextran (Mw 1,000) using carbonyldiimidazole and dimethylaminopyridine. Syntheses of all the conjugates were confirmed by NMR analysis. The degree of substitution and solubility of the conjugates in water were studied. In vitro hydrolysis studies were conducted for valproic acid-dextran conjugates in phosphate buffered solution (pH 7.5) and in bicarbonate buffer (pH 10.22) at 37 °C and 47 °C. In vitro hydrolysis studies of VPA-linker-dextran conjugates were studied in phosphate buffered solution (pH 7.5) in presence of rat brain fraction (RBF) (S9 fraction) and porcine liver esterase (PLE) at 37 °C. Both pH and temperature were shown to affect the rates of hydrolysis of the VPA-dextran conjugates. Half-lives of hydrolysis in phosphate buffer for the conjugates VPA-dextran (Mw 10,000) and VPA-dextran (Mw 70,000) were 62 days and 57 days at 37 °C respectively. The VPA-linker-dextran conjugates were found to be hydrolyzed by RBF and PLE at a different rate. Modelling of the data of hydrolysis of VPA-linker-dextran conjugates was done and different rate constants for the hydrolysis were estimated from the best-fit model. The half-lives of hydrolysis in presence of RBF for the conjugates VPA-triethyleneglycol-dextran and VPA-pentaethyleneglycol-dextran were 28 hr and 44 hr respectively, whereas in presence of PLE these were 18 min and 21 min respectively. In conclusion, VPA-dextran conjugates were shown to release the drug in vitro for an extended period of time. The VPA-linker-dextran conjugates appeared to be very good substrates for PLE, whereas poor substrates for the enzymes in RBF. VPA-linker-dextran conjugates were shown to exhibit in situ activation to release the drug via enzymatic action. VPA-pentaethyleneglycol-dextran conjugate demonstrated a concentration-dependent cytotoxicity. ( en )
General Note:
In the series University of Florida Digital Collections.
General Note:
Includes vita.
Bibliography:
Includes bibliographical references.
Source of Description:
Description based on online resource; title from PDF title page.
Source of Description:
This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Thesis:
Thesis (Ph.D.)--University of Florida, 2012.
Local:
Adviser: Hochhaus, Guenther.
Local:
Co-adviser: Hughes, Jeffrey.
Electronic Access:
RESTRICTED TO UF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE UNTIL 2014-12-31
Statement of Responsibility:
by A. M. Mahbub Hassan.

Record Information

Source Institution:
UFRGP
Rights Management:
Applicable rights reserved.
Embargo Date:
12/31/2014
Classification:
LD1780 2012 ( lcc )

Downloads

This item has the following downloads:


Full Text

PAGE 1

1 VALPROIC ACID DEXTRAN CONJUGATE: SYNTHESIS, CHARACTERIZATION AND IN VITRO HYDROLYSIS STUDIES By A M MAHBUB HASSAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF TH E REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 A M Mahbub Hassan

PAGE 3

3 To my parents, my loving wife and my daughter

PAGE 4

4 ACKNOWLEDGMENTS First, I would like to thank and express my sincere gratitude to my men tor Dr. Guenther Hochhaus for accepting me as a graduate student and giving me the opportunity to work i n his lab. I would also like to thank Dr. Jeffrey Hughes the co chair of my supervisory committee, for giving me the opportunity to start my gradua te research in his lab. I would also like to thank other members of my supervisory committee Dr. Hartmut Derendorf, and Dr. Christopher Batich for their guidance and support during the course of my study. I also like to acknowledge the Interns, Tanja Sch aeuble, Vanessa Rolle, Catharina Clauss, Elsa Marie Treutlein and S tefanie Schmitt for their work and contribution to my research. Special thanks and appreciation to James Rocca, for his help and support during the NMR experiments. I would also like to tha nk to Dr Sihong Song for letting me use the instruments in his lab for my experiments. Thanks to Dr. Lu for helping me with cytotoxicity experiments. I would like to thank my parents for inspiring me in my graduate study. Finally, I would like to express m y gratitude to my wife for her unrelenting support throughout the course of my study. Without her help my d issertation would not have been possible.

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGMENTS ................................ ................................ ................................ .. 4 LIST OF TABLES ................................ ................................ ................................ ............ 7 LIST OF FIGURES ................................ ................................ ................................ .......... 8 LIST OF ABBREVIATIONS ................................ ................................ ........................... 11 ABSTRACT ................................ ................................ ................................ ................... 14 CHAPTER 1 INTRODUCTION AND BACKGROUND ................................ ................................ 16 Epilepsy and Temporal lobe Epilepsy (TLE) ................................ ........................... 16 Etiology and Pathophysiology ................................ ................................ .......... 17 Initial Precip itating Injury (IPI) and Development of TLE ................................ .. 19 Valproic Acid (VPA) and Importance of Neuroprotection ................................ ........ 21 Dextran Conjugates as Sust ained Release Technology ................................ ......... 23 Direct Attachment of Ligand for Chemical Hydrolysis ................................ ....... 24 Attachment via a Linker for Enzyme assisted Hyd rolysis ................................ 25 Goals for Improvement ................................ ................................ ........................... 25 Prolongation in Activity ................................ ................................ ..................... 26 In situ Activation ................................ ................................ ............................... 27 Reduction in Side Effects ................................ ................................ ................. 27 Research Objectives ................................ ................................ ............................... 28 2 SYNTHESIS OF VALPROIC ACID DEXTRAN CONJUGATES ............................. 32 Introduction ................................ ................................ ................................ ............. 32 Materials and Methods ................................ ................................ ............................ 33 Chemicals and Instruments ................................ ................................ .............. 33 Synthesis of The Conjugates ................................ ................................ ............ 35 Characterization of The Conjugates By IR and NMR Spectroscopy ................. 36 Reversed Phase HPLC Conditions ................................ ................................ .. 36 Degree of Substitution of The Conjugates ................................ ........................ 37 Aqueous Solubility of The Conjugates ................................ .............................. 38 In vitro Hydrolysis Study ................................ ................................ ................... 38 Statistical Analysis ................................ ................................ ............................ 40 Results and Discussion ................................ ................................ ........................... 40 Synthesis of Dextran Valproic Acid Conjugates and Spectral Analysis ............ 40 Degree of Substitution and Solubility of The Conjugates ................................ 42 In vitro Hydrolysis Studies ................................ ................................ ................ 44 Conclusion ................................ ................................ ................................ .............. 46

PAGE 6

6 3 SYNTHESIS OF VALPROIC ACID LINKER DEXTRAN CONJUGATES: DESIGNING A PRODRUG FOR IN SITU ACTIVATION ................................ ........ 56 Introduction ................................ ................................ ................................ ............. 56 Materials and Methods ................................ ................................ ............................ 57 Chemicals and Instruments ................................ ................................ .............. 57 Synthesis of The Conjug ates ................................ ................................ ............ 59 Characterization of The Conjugates by NMR Spectroscopy ............................. 62 Aqueous Solubility of The Conjugates ................................ .............................. 63 Degree of Substitution of The Conjugates ................................ ........................ 63 Results and Discussion ................................ ................................ ........................... 64 Spectral Analysis ................................ ................................ .............................. 64 Solubility Analysis ................................ ................................ ............................. 70 Degree of Substitution ................................ ................................ ...................... 71 Conclusion ................................ ................................ ................................ .............. 71 4 SYNTHESIS OF VALPROIC ACID LINKER DEXTRAN CONJUGATES: OVERCOMING THE SOLUBILITY ISSUE ................................ ............................. 78 Introduction ................................ ................................ ................................ ............. 78 Materials and Methods ................................ ................................ ............................ 79 Chemicals and Instruments ................................ ................................ .............. 79 Synthesis of The Conjugates ................................ ................................ ............ 81 Characterization of The Conjugates by NMR Spectroscopy ............................. 8 4 Degree of Substitution of The Conjugates ................................ ........................ 85 Aqueous Solubility of The Conjugates ................................ .............................. 86 In vitro Hydrolysis Study with Porcine Liver Esterase (PLE) ............................. 86 I n vitro Hydrolysis Study with Rat Brain Fraction (RBF) ................................ ... 87 Cytotoxicity Test ................................ ................................ ............................... 88 Statistical Analysis ................................ ................................ ............................ 89 Results and Discussion ................................ ................................ ........................... 89 Synthesis of Dextran linker Valproic Acid Conjugates and Spectral Analysis .. 89 Degree of Substitution and Solubility of The Conjugates ................................ 94 In vitro Hydrolysis Studies ................................ ................................ ................ 96 Cytotoxicity Test ................................ ................................ ............................. 102 Conclusion ................................ ................................ ................................ ............ 105 5 SUMMARY OF RESULTS AND FUTURE WORK ................................ ................ 123 APPENDIX A SOLUBILI ZING TESTS FOR VPA HXD D70 AND VPA HXD D10 ...................... 126 B SOLUBILIZING TESTS FOR VPA TEG D10 ................................ ........................ 127 LIST OF REFERENCES ................................ ................................ ............................. 128 BIOGRAPHICAL SKETCH ................................ ................................ .......................... 142

PAGE 7

7 LIST OF TABLES Table page 2 1 Degree of substitution of conjugates VPA D10 and VPA D70 from hydrolysis study ................................ ................................ ................................ .................. 47 2 2 Aqueous solubility of the conjugates VPA D10 and VP A D70 at room temperature ................................ ................................ ................................ ....... 47 2 3 Rate consta nts and t 1/2 values of VPA D10 and VPA D70 conjugates in phosphate and bicarbonate buffer at 37 C and 47 C ................................ ...... 47 3 1 Different solubilization approaches ................................ ................................ ..... 73 3 2 Degree of substitution of conjugates VPA HXD D70, VPA HXD D10 and VPA TEG D10 from NMR study. ................................ ................................ ........ 73 4 1 Degree of substitution of conjugates VPA TEG D1 and VPA PEG D1 from hydrolysis study ................................ ................................ ............................... 106 4 2 Aqueous solubility of the conjugates VPA TEG D1 and VPA P EG D1 at room temperature ................................ ................................ ................................ ..... 106 4 3 Rate c onstants for hydrolysis of VPA TEG D1 and VPA PEG D1 in phosphate buffer solution in presence of porcine liver esterase at 37 C ........ 106 4 4 Rate constants for hydrolysis of VPA T EG D1 and VPA PEG D1 in phosphate buffer solution in pres ence of rat brain fraction at 37 C ................ 106 4 5 Half lives (t 1/2 ) for hydrolysis of VPA TEG D1 and VPA PEG D1 in phosphate buffer solution in presence of po rcine liver esterase at 37 C .......................... 106 4 6 Half lives (t 1/2 ) for hydrolysis of VPA TEG D1 and VPA PEG D1 in phosphate buffer solution in pres ence of rat brain fraction at 37 C ................................ .. 106

PAGE 8

8 LIST OF FIGURES Figure page 1 1 The various aspects of epileptogenesis after the initial brain insult ................... 30 1 2 (1 6) linked main chain of dextran ................................ ............................... 31 2 1 Synthesis of valproic acid dextran conjugate. ................................ .................... 48 2 2 FT IR spectra of VP A. ................................ ................................ ........................ 49 2 3 FT IR spectra of dextran. ................................ ................................ .................... 49 2 4 FT IR spectra of the conjugate VPA D10. ................................ .......................... 50 2 5 FT IR spectra of the conjugate VPA D70. ................................ .......................... 50 2 6 1 H NMR spectra of VPA, Dextran and VPA D10 ................................ ............... 51 2 7 1 H NMR sp ectra of VPA, Dextran and VPA D70 ................................ ............... 51 2 8 Degree of substitution of the conju gates VPA D10 and VPA D70 ..................... 52 2 9 Aqueous solubility of the conjugates VPA D10 and VP A D70 at room temperature ................................ ................................ ................................ ....... 53 2 10 In vitro hydrolysis of conjugate VPA D10 in phosphate buffer at 37 C and at 47 C ................................ ................................ ................................ ................. 54 2 11 In vitro hydrolysis of conjugate VPA D10 in bicarbonate buffer at 37 C and at 47 C ................................ ................................ ................................ ............. 54 2 12 In vitro hydrolysis of conjugate VPA D70 in phosphate buffer at 37 C and at 47 C ................................ ................................ ................................ ................. 55 2 13 In vitro hydrolysis of conjugate VPA D70 in bicarbonate buffer at 37 C and at 47 C ................................ ................................ ................................ ............. 55 3 1 Design of the s ynthesis of valproic acid linker dextran conjugate. ..................... 74 3 2 1 H NMR spectra of VPA, HXD and VPA HXD ................................ ................... 75 3 3 1 H NMR spectra of VPA HX D, Dextran and VPA HXD D70 .............................. 75 3 4 1 H NMR spectra of VPA, HXD and VPA HXD ................................ ................... 76 3 5 1 H NMR spectra of VPA HXD, Dextran and VPA HXD D 10 .............................. 76

PAGE 9

9 3 6 1 H NMR spectra of VPA, TEG and VPA TEG ................................ ................... 77 3 7 1 H NMR spectra of VPA TEG, Dextran and VPA TEG D10 .............................. 77 4 1 Synthesis of valproic acid triethylene glycol dextran conjugate ........................ 107 4 2 Synthesis of valproic acid pentaethylene glycol dextran conjugate .................. 108 4 3 1 H NMR spectra of VPA, TEG, and VPA TEG ................................ ................ 109 4 4 1 H NMR spectra of VPA TEG, Dextran and VPA TEG D1 .............................. 109 4 5 1 H NMR spectra of VPA, PEG and VPA PEG ................................ ................. 110 4 6 1 H NMR spectra of VPA PEG, Dextran and VPA PEG D1 .............................. 110 4 7 Degree of substitution of conjugates VPA TEG D1 and VPA PEG D1 ........... 111 4 8 Aqueous solubility of the conjugates VPA TEG D1 and VPA P EG D1 at room temperature ................................ ................................ ................................ ..... 112 4 9 Hydrolysis scheme of valproic acid linker dextran conjugate .......................... 113 4 10 Hydrolysis scheme of valproic acid linker dext ran conjugate .......................... 113 4 11 Hydrolysis scheme of valproic acid linker dextran conjugate .......................... 114 4 12 Hydrolysis scheme of valproic acid linker dextran conjugat e. .......................... 114 4 13 Hydrolysis of the conjugate VPA TEG D1 in PBS in the presence of porcine liver esterase at 37 C ................................ ................................ ..................... 115 4 14 Hydrolysis of the conjugate VPA PEG D1 in PBS in the presence of porcine liver esterase at 37 C ................................ ................................ ..................... 115 4 15 Model fit for the hydrolysis of VPA TEG D1 conjugate according to hydrolysis scheme A. ................................ ................................ ................................ ......... 116 4 16 Model fit for the hydrolysis of VPA PEG D1 conjugate according to hydrolysis scheme A. ................................ ................................ ................................ ......... 116 4 17 Model fit for the hydrolysis of VPA TEG D1 conjugate according to hydrolysis scheme B. ................................ ................................ ................................ ......... 117 4 18 Model fit for the hydrolysis of VPA PEG D1 conjugate according to hydrolysis scheme B. ................................ ................................ ................................ ......... 117 4 19 Model fit for the hydrolysis of VPA TEG D1 conjugate according to hydrolysis scheme C. ................................ ................................ ................................ ........ 118

PAGE 10

10 4 20 Model fit for the hydrolysis of VPA PEG D1 conjugate according to hydr olysis scheme C. ................................ ................................ ................................ ........ 118 4 21 Effect of VPA PEG D1 conjugate on cell viab ility ................................ ............ 119 4 22 Effect of VPA PEG D1 conjugate on ce ll viabil ity ................................ ............ 120 4 23 Effect of VPA on cell viabi lity ................................ ................................ ........... 121 4 24 Effect of dextran on cell vi ability ................................ ................................ ...... 122

PAGE 11

11 LIST OF ABBREVIATION S AED Antiepileptic drug AGU Anhydroglucose unit AMRIS Advanced magnetic resonance imaging and spectroscopy BDNF Brain derived neurotrophic factor CA1 Cornu ammonis 1 CA3 Cornu ammonis3 cAMP Cyclic adenosine monophosphate CDI Car bo nyldiimidazole CED Convection enhanced delivery CNS Central nervous system CREB cAMP responsive element binding protein CT Computed tomography DCC Dicyclohexylcarbodiimide DCM Dichloromethane DG Dentate gyrus DH Dentate hilus DMAP Dimethylaminopyridin e DMEM Dulbecco's Modified Eagle Medium DMF Dimethylformamide DMSO Dime thyl sulfoxide DS Degree of substitution DSGL Dentate supragranular layer EEG Electroencephalogram FT IR Fourier transform infra red

PAGE 12

12 GABA Gamma amino butyric acid GAD Glutamic acid decar boxylase GCL Granule cell layer HPLC High performance liquid chromatography HXD 1,6 Hexanedio l ICV Intracerebroventricular IR Infra red IPI Initial precipitating injury KA Kainic acid kD Kilo Dalton MAP Mitogen activated protein kinase MaTLE Mass associate d temporal lobe epilepsy MHz Mega hertz MRI Magnetic resonance imaging MTLE Mesial temporal lobe epilepsy MTT 3 (4, 5 dimethylthiazolyl 2) 2, 5 diphenyltetrazolium bromide NMR Nuclear magnetic resonance PBS Phosphate buffered saline PEG Pentaethylene glycol PLE Porcine liver esterase PTLE Paradoxical temporal lobe epilepsy RBF Rat brain fraction RT Room temperature SE Status epilepticus SEM Standard error of the mean

PAGE 13

13 SLM Stratum lacunosum moleculare SR Stratum radiatum SRMS Spontaneous recurrent motor seizure s TEG Triethylene glycol THF Tetrahydrofuran TLE Temporal lobe epilepsy TEA Triethylamine VPA Valproic acid

PAGE 14

14 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements f or the Degree of Doctor of Philosophy VALPROIC ACID DEXTRAN CONJUGATE: SYNTHESIS, CHARACTERIZATION AND IN VITRO HYDROLYSIS STUDIES By A M Mahbub Hassan December 2012 Chair: G enther Hochhaus Co C hair: Jeffrey Hughes Major: Pharmaceutical Sciences The objective of this study was to synthesize v alproic acid dextran conjugate which would be able to release t he drug via chemical hydrolysis over a n extended period and also to synthesize valproic acid linker dextran conjugate which would provide in situ acti vation to release the drug via enzymatic hydrolysis Valproic acid (VPA) was conjugated directly to dextrans of molecular w eights (Mw) 10,000 and 70,000 using carbonyldiimidazole. To synthesize valproic acid linker dextran conjugate, VPA w as initially conj ugated to linker s (triethylene glycol or pentaethylene glycol) using dicyclohexylcarbodiimide and s ubsequently, conjugated to dextran (Mw 1,000) using carbonyldiimidazole and dimethylaminopyridine Syntheses of all the conjugates were confirmed by NMR anal ysis. The d egree of substitution and solubility of the conjugates in water were studied. In vitro h ydrolysis studies were conducted for valproic acid dextran conjugates in phosphate buffered solution ( pH 7.5 ) and in bicarbonate buffer (pH 10.22 ) at 37 C a nd 47 C In vitro hydrolysis studies of VPA linker d extran conjugates were studied in phosphate buffered solution ( pH 7.5 ) in presence of rat brain fraction (RBF) (S9 fraction)

PAGE 15

15 and porcine liver esterase (PLE) at 37 C. Both pH and temperature were shown to affect the rate s of hydr olysis of the VPA dextran conjugates. Half liv es of hydrolysis in phosphate buffer for the conjugate s VPA dextran (Mw 10,000) and VP A dextran (Mw 70,000) were 62 day s and 57 days at 37 C respectively The VPA linker dextran conj ugates were found to be hydrolyzed by RBF and PLE at a different rate. Modelling of the data of hydrolysis of VPA linker dextran conjugates was done and different rate constants for the hydrolysis were estimated from the best fit model. The half lives of h ydrolysis in presence of RBF for the conjugates VPA triethyleneglycol dextran and VPA pentaethyleneglycol dextran were 28 hr and 44 hr respectively, whereas in presenc e of PLE these were 18 min and 21 min respectively In conclusion, VPA dextran conjugates were shown to release the drug i n vitro for an extended period of time The VPA linker dextran conjugates appeared to be very good substrates for PLE, whereas poor substrates for the enzymes in RBF. VPA linker dextran conjugates were shown to exhibit in s itu activation to release the drug via enzymatic action. VPA pentaethyleneglycol dextran conjugate demonstrated a concentration dependent cytotoxicity.

PAGE 16

16 CHAPTER 1 INTRODUCTION AND BAC KGROUND Epilepsy and Temporal lobe Epilepsy (TLE) Epilepsy is a neurolog ical disease which is defined as a chronic seizure disorder or group of disorders that usually recur in an unprovoked and unpredictable manner ( 1 4 ) The te to attack It is not a specific disease, rather a group of disorder having heterogeneous symptoms ( 5 ) Seizure is defined by an excessive, hypersynchronous discharge of cortical neuronal activity, which can be measured by electroencephalogram (EEG) ( 6 ) A seizure prod uces abnormal electrical activity in the brain with characteristic changes on the EEG ( 7 ) C ortical area of the brain has different parts and each part has it s own function T he clinical manifestation of a seizure depends on the site of the cortical area involved the extent of irritability of the area affected and the intensity of the impulse generated Seizure activity usually comprises of three major phases prodrome, ictal and post ictal phase ( 6 ) A prodrome is the initial phase typically represented by changes in behavior or mood ( 5 6 ) A prodrome may include an aura which is a subjective sensation of an unusual smell or flashing light ( 5 6 ) The ictal pha se is when the actual seizure happens The post ictal phase immediately follows the seizure (ictal phase). During this phase, t he patient typically experiences lethargy and confusion ( 5 6 ) Depending on which areas of the brain involved, e pileptic seizures are further sub divided into partial or focal seizures and generalized seizures ( 8 ) Partial seizures are those where abnormalities are restricted to a partic ular locus within the brain and generalized seizures are those when the brain is diffusely and bilaterally involved a nd

PAGE 17

17 which can result in abnormal activity throughout the brain ( 9 10 ) Partial seizures are further subdivided where patient retains where consciousness is impaired ( 11 ) Epilepsy is one of the most fr equent ne urological disorders ( 12 ) M ore than 2 million Americans and 50 million people worldwide are affected by the disease ( 13 16 ) The probability of seizure onset is the maximum during the first year of life which decreases subsequently each decade after the first year until age 60 ( 6 ) Majority of the patients (a pproximately 70% ) h as only one seizure type whereas the remaining patients have two or more seizure types ( 6 ) Temporal lobe epilepsy (TLE) is characterized by partial seizures where seizures originate from the temporal lobe of the br ain ( 17 18 ) event generated predominantly by mesial temporal limbic str uctures which usually involves symptoms referable to this area, such as a sensation of epigastric rising, emotional changes (most commonly fear) and occasionally olfactory or gustatory ( 11 ) Temporal lobe epilepsy is one of the most frequent types of drug resistant epilepsy ( 14 16 ) Although t he exact proportion of the drug resistant epilepsy patients varies in the literature, approximately ~ 30% of epilepsy patients appear to be pharmacoresistant, despite administration of two to three first line antiepileptic drugs ( AEDs ) in an optimally monitored regimen ( 19 ) Seizures appear to be medically intractable in about 40% of patients with TLE ( 20 ) E tiology and Pathophysiology The etiology of epilepsy is a major factor which will determine the clinical course and prognosis of the disease ( 21 ) Epilepsy is a multifactorial disease in a majority of the cases. E pilepsy is often found to be the result of genetic ac quired and provoking

PAGE 18

18 factors in combination Because of that, assignment of any such cases to any single etiology is very difficult and to an extent arbitrary However, the case is assigned to a predominant cause which is happened to be identified in most of the cases Idiopathic epilepsy has no identifiable cause with any gross neuroanatomic or neuropathologic abnormality and primarily genetic in origin. Sympt omatic epilepsy is associated with gross anatomic or pathologic abnormalities secondary to an acq uired or genetic cause ( 21 ) These causes include tuberous sclerosis, hippocampal sclerosis, glioma, non accidental head injury in infa nts, viral and bacterial meningitis, cerebral infarctions cerebral hemorrhage, etc. ( 22 ) Cryptogenic epilepsy has no id entifiable cause and presumed symptomatic in nature. Temporal lobe epilepsy is frequently associated with hippocampal sclerosis and some other distinctive histopathological features in the surrounding area. These features are described as follows Hippocam pal neuronal cell loss and reorganization of synapse after seizures Hippocampal cell loss and shrinkage have been observed later in the lives of patients with TLE, after an initial precipitating injury ( 23 ) Acute seizures often cause considerable bilateral neurodegeneration in the dentate hilus CA1 and CA3 subfields, which eventually leads to massive abnormal sprouting of mossy fibers into the dentate supraganular laye r ( 24 ) Degeneration of GABA ergic interneurons after seizures D egeneration of fractions of GABA ergic interneurons has been implicated in epileptogenesis and TLE. Usually there is a balance between excitation and inhibition in the neuronal network. U nder normal conditions GABA ergic hippocampal interneurons provide inhibitory inputs

PAGE 19

19 to the neuronal circuit which sets a threshold for the excitation of pyramidal cell s and prevent the principal excitatory hippocampal neurons from becoming hyper excitable Thus, it is likely that loss of fractions of GABA ergic interneurons may lead to reduced inhibitory input to principal hippocampal neurons which ultimately leads to t he persistent hyper excitability in TLE ( 25 28 ) Dentate neurogenesis a nd temporal lobe epilepsy D entate gyrus exhibits abnormal neuronal circuitry extended in to the dentate hilus Hippocampal injury induced by k ainic acid increa ses neurogenesis in the adult dentate gyrus ( 29 ) during the first few weeks after injury ( 30 ) which is due to the release of multiple mitogenic factors from dying neurons, deafferented granule cells and reactive glia ( 31 33 ) Some of the newly born neu rons move into the dentate hilus abnormally shortly after the injury ( 34 37 ) possibly because of overproduction of new neurons in the dentate subgranular zone lack of space in the granule cell layer and loss of expression of reelin ( 38 ) These ectopic neurons contribute to a lower seizure t hreshold, frequently exhibit spontaneous epileptiform activity and maintain recurrent seizures in epileptic rats ( 39 40 ) Addit ionally, dentate neurogenesis declines drastically in the chronically injured hippocampus after developing chronic TLE characterized by spontaneous recurrent motor seizures ( 41 ) which may contribute to increased seizure susceptibility of the dentate gyrus Consequently, declined neurogenesis could at least partially be linked to hippocampal dependent learning and memory deficits ( 42 44 ) Initial Precipitating Injury (IPI) and Development of TLE Several studies have shown individuals affected with TLE typically have an initial precipitating injury (IPI) such as the status epilepticus (SE), head trauma, encephalitis or childhood febrile seizures ( 45 47 ) Chronic TLE is characterized by spontaneous

PAGE 20

20 recurrent motor seizures o riginating from temporal lobe foci and lea r ning and memory impairments ( 48 49 ) The development of the chro nic TLE is usually preceded by a latent period of several years after the initial injury Also, numerous studies indicate that TLE is frequently associated with hippocampal sclerosis which is characterized by atrophy, induration, neuronal loss and astrogl ial proliferation in the hippocampus ( 50 ) Human studies suggested that the h ippocampal sclerosis could be the likely cause to initiate or contribute to the development of most TLEs ( 51 ) However, not a ll TLE patients exhibiting hippocampal damage have a hi story of initial insult. It is still deba table whether the damage found in hippocampus is the cause or consequ ence of TLE. However, t he process by which normal neuronal cells become hyper excitable ove r the course of time generating spontaneous recurrent motor seizures and learning A variety of neurochemical modulators and multiple mechanisms have been implicated in the initiation of e pi leptogenesis after an IPI. F igure 1 1 outlines major epi leptogenic processes that ensue after an IPI and lead to the development of spontaneous recurrent motor seizures Two mechanisms have been hypothesized to account for at least contribute to the epileptogenesis after brain injury. These are 1) selective GABA ergic interneuron loss and consequent reduction in GABA mediated inhibition (but not complete failure) ( 52 ) an d 2) a xon sprouting by many of the remaining neurons subsequent to the death of glutamatergic principal neurons (and neurogenesis in some ar eas), and consequent formation of new recurrent glutamatergic excitatory circuits ( 52 ) Following the initial injury, there is a sequence of events take place in the injured

PAGE 21

21 area over months to years before the occurrence of spontaneous recurrent seizures. The latency period between the IPI and the development of s pontaneous recurrent motor seizures provides a window of opportunity for initiating an effective treatme nt. can be a very promising therapy in the prevention of TLE. Valproic Acid (VPA) and Importance of Neuroprotection Usually, epileptogenesis is originated after certain brain insults called in itial precipitating injury such as head trauma, cerebrovascular disease, brain tumors, neurosurgical procedures, neurodegenerative con ditions, status epilepticus and complex febrile seizu res. Hippocampal neuronal degeneration and hippocampal sclerosis have also been implicated in the pathogenesis of chronic TLE and found in most of the animal models of TLE and also in chronic epilep tic patients of TLE. Valproic acid and its sodium salt (so dium valproate) are among the most prescribed antiepileptic drugs. The antiepileptic activity of VPA has been ascribed to multiple effects, which result in the reduction of neuronal excitability ( 53 ) These effects include increase of GABA ergic activity ( 54 58 ) negative regulation of NMDA receptor mediated glutamatergic excita tory activity ( 59 62 ) and negative regulation of voltage gated Na + channels ( 63 64 ) Among other conventional antiepileptic drugs, early ad ministration of VPA after status epilepticus had shown a great promise in the preventi on of chronic epilepsy. Although, the efficacy o f VPA in preventing the development of chronic epilepsy was found to be varied and depended on the model employed. In a kainic acid model of TLE, repetitive treatment wit h VPA for 40 days, starting 24 hours after the onset of status epilepticus considerab ly blocked the development of spontaneous seizures, deficits in emotional responses or spatial learning and protected

PAGE 22

22 ag a inst neurodegeneration in C A1 and CA3 subfields and the dentate hilus for 40 days after the status epilepticus ( 65 ) Brandt et. al. ( 66 ) studied the efficacy of VPA treatment in a rat model of status epilepticus induced b y prolonged electrical stimulation of the basal amygda la The study showed that VPA t reatment beginning at 4 hours after status epilepticus and ending at 4 weeks after status epilepticus failed to prevent the occurrence of spontaneous recurrent motor seizures but prevented completely the neuronal damage in t he hippocampal formation including the dentate hilus and the behavioral impair ments to some extent ( 66 ) In another study using kainic acid model of TLE showed th at prolonged administration of VPA starting shortly after status epilepticus blocked the seizure induced abnormal neurogenesis in the dentate gyrus and protected against seizure induced cognitive impairment in a hippocampus dependent learning task ( 67 ) On th e contrary, short term infusion of VPA through the microdialysis probe did not prevent hippocamp al neurodegeneration induced by potassium channel blocker 4 aminopyridine ( 68 ) In another study conducted by Temkin et. al. ( 69 ) showed that VPA therapy for one month and six months did not prevent the late seizures following traumatic brain injury. T he precise mechanism underlying the beneficial effects of VPA is unknown. VPA has been found to regulate a number of cell survival fa ctors including cyclic adenosine monophosphate (cAMP) responsive element binding protein (CREB), brain de rived neurotrophic factor bcl 2, and mitogen activat ed protein kinase which may be attributed to its neuroprotective and neurotrophic effects ( 70 ) However, the beneficial ef fects of VPA administration in decreasing status epilepticus induced hippocampal

PAGE 23

23 neurodegeneration, abnormal neuroge nesis and learning impairments have been accepted in general ( 12 ) So, to valida te the efficacy of VPA as a neuroprotective agent and as an antiepileptogenic drug, long term studies in multiple models of chr onic epilepsy need to be conducted. T he ultimate goal of any prophylactic drug treatment af ter an initial injury such as status epilepticus is to prevent the developme nt of spontaneous recurrent motor seizures. If spontaneous recurrent motor seizures canno t be prevented, alternate goals would be to make these seizures less frequent, less severe and less resistant to AED treatment ( 71 ) Also, any beneficial effect on the neuronal damage or on the learning and memory deficits associated with epilepsy would be desirable. To this effect, a long acting formulation of VPA will be of utmost importance, which will aid in conducting such long term experiment. Dextran Conjugates as Sustained Release Technology Dextrans are glucose polymer s with predominantly (1 6) linkages with branching points in 2, 3 and 4 positions ( 72 ) Figure 1 2 shows typical structure of dextran with branching points in different positions. Dex trans have been used clinically over 50 years as plasma volume expander s and as antithrombolytic agents ( 73 ) Dextrans have been investigated as potential macromolecular carriers delivering drugs, proteins, targeting moieties and imaging a gents Dextran conjugates were reported to increase the circulation time of drugs and proteins and to extend the duration of action ( 74 76 ) The use of d extra n conjugates has been proposed for targeting drugs to specific sites of action via active or passive targeting approaches ( 77 78 ) Dextrans have also been used to increase the stability of therapeutic agents ( 79 ) and to reduce side effects including antigenicity of enzyme or proteins ( 80 81 )

PAGE 24

24 D extran molecules possess several suitable properties which make it a s uitable macromolecular carrier. Clinical dextrans have been used over many years without serious side e ffects Parenteral dextran therapy did induce dextran induced anaphylactoid reactions on rare occasion ( 82 86 ) T his reaction has been elim inated essentially using monovalent hapten dextran preparation ( 87 88 ) Dextran conjugates can be constructed either by linking t he active drug moiety directly to the polymer backbone through covalent bonding, or by linking the drug molecule to the polymer backbone through the use of a suitable spacer arm. These conjugates can be characteri zed with relative ease ( 89 ) The conjugates with dextran retain water solubility up to a high percentage of degree of substitution (15 20% w/w) ( 89 ) Dextrans can be autoclaved and are rela tively stable to wards further chemical manipulation ( 89 ) D extran protein conjugates may retain the pharmacologic activity of the protein to some extent; most of the dextran conjugates with small drug molecule s act as prodrugs, releasing the active drug in vivo ( 90 ) Direct Attachment of Ligand for Chemical Hydrolysis Dextrans possess a large number of hydroxyl groups available for conjugation of drug molecules This single type functional group in turn limits the number of drugs which might be attached direc tly to the carrier. However, d extran prodrugs in which the drug is linked directly to the polymer chain, act as a depot releasing the drug in a predictable manner ( 91 92 ) The rate of hydrolysis of the macromolecular dextran prodrug to regenerate the parent drug is mostly pH dependent because the bulky dextran molecule renders the hydrolytic group inaccessible to enzymatic action ( 89 93 )

PAGE 25

25 Attachment via a Linker for Enzyme assisted Hydrolysis Drug molecules can be linked to dextran molecules via a linker group. Incorporation of a linker group can provide some advantages. Firstly, the terminal functional group of the spacer arm can be varied, which will allow formation of covalent bonds with drugs through a variety of chem ical bonds ( 94 100 ) Secondly, steric hindrance of enzyme activation of the liganded drug might be circumvented by augmenting the distance between the drug and th e dextran matrix ( 89 101 ) So, b y a proper selection of th e linker molecule (t he length and the chemical str ucture) the linker drug construct can be designed in such a way to be cleaved selectively by the enzymes of the target tissue ( 102 ) A localized effect can b e achieved by this approach. Moreover sequentially labile macromolecular dex tran prodrugs can be designed which will allow regeneration of low molecular weight prodrug (linker drug derivative) upon pH d ependent hydrolysis. This linker drug derivative will be activated later at the target tissue by enzymatic action. The macromolecular prodrug will also provide a depot effect in such a way Goals for Improvement T he efficacy of VP A administration for about one month after an initial precipitating injury like status epilepticus was studied in several experiments in animal models ( 65 68 ) Valproic acid was administered in these studies by intra per itoneal injection to rats two to three times a day to maintain therapeutic blood levels of the drug The drug was also administered at a dose (e.g. 200 600 mg/kg) which gave rise to adverse effects in animals, e.g. sedat ion and ataxia which made it less suitable for prolon ged experiments

PAGE 26

26 However, the benefic ial effects of VPA administration for about one month in preventing chronic epilep sy observed in some of these studies may or may not persi st once the VPA administration is termina ted ( 12 ) Also, administration of VPA longer than one month may result in a different outcome. These studies also showed variable re sults about the efficacy of prol onged VPA administration, which depended on the specific model of chronic epilepsy used in the study ( 12 ) To validate the eff icacy of long term administration of VPA for preventing the development of injury induced chro nic epilepsy l ong term studies in multiple models of chronic epilepsy need to be conducted in future ( 12 ) The outcome of animal studies will also pave the way for future studies in human. To conduct such long term studies, a long acting formulation of valproic acid will be of param ount importance. Development of such a long acting formulation of VPA will seek to improve upon the following areas. Prolongation in Activity In animal mo dels, the anticonvulsant activity of valproic acid depends on several factors. Among these, the animal sp ecies used in the experiment, the method of seizure induction, the seizure type, the route of administration of the drug and the time interval between drug administration and seizure induction ( 53 ) have been identified as important factors VPA has a short half life in most animal species. Because of th is, most pronounced effects are observed within 2 to 5 minu tes after parenteral injection ( 103 104 ) Thus, t he d uration of activity of VPA in most laboratory animal species is only short, which necessitates the use of high d oses of VPA to suppress long lasting or repeatedly occurring seizures in animal models ( 53 ) The development of a sustained release formulation of valproic acid would be a pre requisite for any long term animal and human experiment. Any such formulation will

PAGE 27

27 be able to p rolong the activity by releasing the drug slowly by pH dependent chemical hydrolysis ; or site directed hydrolysis by enzymatic action Such a conjugate will eliminate the need for multiple daily administrations, which will be very convenient to conduct lengthy animal and human experiments In situ A ctivation Valproic acid enters the brain through the blood brain barrier by passive diffusion and by carrier mediated facilitated diffusion involving medium and long chain fatty acid selective anion exchanger ( 105 ) The drug leaves the brain into the blood by a probenecid sensitive active transport system ( 105 ) A convection enhanced delivery system can deliver a mac romolecular prodrug of VPA at or near the site of action remote from the blood brain barrier. The conjugate will then be distributed into the surrounding area through convection and diffusion. A properly designed macromolecular prodrug of VPA can deliver the drug at or near the site of action either by chemical or enzymatic action. The released drug will be absorbed into the surrounding cellular structure. This c an limit the efflu x of the drug out of the brain and will offer a significant advantage over co nventional formulation of VPA. Reduction in S ide Effects Valproic acid exerts a number of side effects when used as a neuroprotective agent in a number of animal studies. The drug was administered intraperitoneally multiple times a day to the animals short ly after status epilepticus which produced increased incidences of ataxia and somnolence in the study animals. To conduct such studies for months would be extremely difficult by using conventional immediate relea se formulations of VPA So, there is a need for a long acting formulation of VPA which will

PAGE 28

28 minimize the frequency of administration, release the drug in a predictable way and reduce side effects due to the high drug levels. Research Objectives The primary objective s of our study were to synthesiz e valproic acid dextran conjugate with or without the spacer and study i n vitro hydrolysis of the conjugates to find out the suitability of any of the conjugates for a sustained release formulation and for in situ activation by enzymatic action The ove ral l study is divided into two specific aims. These are described as follows Specific Aim I: To synthesize valproic acid dextran conjugate and study i n vitro hydrolysis of the conjugate. Macromolecular pro drugs with dextran can be used to release drug slowly upon hydrolysis over a long period. In most cases regeneration rates of the parent drug depend mainly on pH and temperature of the hydrolysis media. Molecular weights of carrier dextran molecule and degree of substitution of the conjugates will also affec t the rate of hydrolysis of the conjugates. The primary objective of this specific aim is to synthesize valproic acid dextran conjugate using different molecular weights of dextran characterize the conjugate by FT IR and NMR spectroscopy study degree of substitution and solubility of the conjugates and study the effect of pH and temperature on the rates of hydrolysis of the conjugate s i n vitro Specific Aim II : To synthesize valproic acid linker dextran conjugate and study i n vitro hydrolysis of the conju gate in the presence of esterase enzyme. Macromolecular prodrugs where drug is attached direct ly to the dextran molecules appeared to be resistant towards enzymatic hydrolysis by various esterase enzymes. The bonds formed between drug and dextran seemed to be inaccessible by the active sites of the enzymes due to the steric hindrance offered by the dextran molecule s The

PAGE 29

29 primary objective of this aim is to synthesize valproic acid dextran conjugate using a linker to aid in enzymatic hydrolysis by the estera se enzyme, characterize the co njugate by NMR spectroscopy, study degree of substitution and solubility of the conjugate, study i n vitro hydrolysis of the conjugate in the presence of esterase enzyme and study the toxicity of the conjugate on viable cells These aims are described in detail in three d ifferent chapters. Specific aim I is discussed in Chapter 2. Specific aim II is discus sed in Chapter 3 and Chapter 4.

PAGE 30

30 Figure 1 1. The various aspects of epileptogenesis after the initial brain insult and t he evolution of the initial precipitating injury into chronic epilepsy and learning and memory deficits ( 12 ) (Reproduced with permission from Elsevier Limited, The Boulevard, Langford Lane, Kidlington, Oxford, UK).

PAGE 31

31 Figure 1 2. (1 6) linked main chain of dextran with branching points in 2 3 and 4 positions.

PAGE 32

32 CHAPTER 2 SYNTHESIS OF VALPROI C ACID DEXTRAN CONJUGATES Introduction Valproic acid is a simple branched chain fatty acid and a major and well established first line antiepileptic drug (AED). Valproic acid exhibits a wide spectrum of activity against a broad range of seizure disorders ( 106 ) Valproic acid is effective against partial and generalized seizures. In add ition to these, valproic acid has demonstrated efficacy in the treatment of very refractory syndromes, such as, Lennox Gastout synd rome and West syndrome. Recently, there is a profound interest in searching for therapeutic option which will help to prevent the development of epilepsy that is, to prevent epileptogenesis. Valproic acid has been found to be neuroprotective in several an imal models of temporal lobe epilepsy (TLE) and in models of ischemia ( 107 108 ) It has appeared from those stu dies that the efficacy of valproic acid in preventing the development of chronic epilepsy varies depending on the models employed ( 12 ) T o validate the efficacy of valproic acid as a neuroprotective age nt and as an antiepileptogenic drug, long term studies in multiple models of chronic epilepsy need to be conducted ( 12 ) For these studies v alproic acid needed to be administered multiple times a day throughout the study period which was met wit h several challenges. A dministration of drug caused multiple side effects in the study animals, such as, sedation and ataxia shortly after status epilepticus ( 66 ) So, prolonged administration of valproic acid will defini tely cause increased mortality and morbidity of the study animals and will make a long term study difficult to conduct This warrants the development of a prolonged release formulation of valproic acid, which will obviate the necessity of

PAGE 33

33 multiple daily ad ministrations for a longer period and thus, eliminate or minimize the incidence of side effects in the study animals. A dextran valpro ic acid prodrug would be a very attractive formulation for those long term studies. Dextrans are glucose polymers which h ave been used more than 50 years as plasma volume expanders ( 73 ) Dextrans contain a large number of hydroxyl groups which can be easily conjugated to valproic acid by either direct attachment or through a linker. Dextran conjugates with molecular weights up to 70 kD have been routinely used for anterograde and retrograde pathway tracing studies of the nervous system ( 109 111 ) Hence, convection enhanced delivery of the dextran valproic acid prodrug (with 10 kD or 70 kD dextran) is likely to cause enhanced permeation and retention of the conjugates in the brain tissue. Also, such a localized delivery will require lower amount of conjugate needed to achieve therapeutic level in the brain in comparison to conventional formulation of valproic acid. This will minimize fluctuation in the therapeutic levels, as it needs to be admini ster e d less frequently and cause fewer incidences of side effects. In this regard, our aims were to synthesize valproic acid dextran conjugate using dextrans with Mw 10 kD and 70 kD characterize the conjugates by FT IR and NMR spectro scopy, study degree o f substitution and solubility of t he conjugates and also study i n vitro hydrolysis of the conjugates at different pH values and temperature s. Materials and Methods Chemicals and Instruments Dextran from Leuconostoc mesenteroides (average Mw 64 76 kD ), d e xtran from Leuconostoc mesenteroides (average Mw 9 11 kD ), d imethyl sulfoxide (DMSO)

PAGE 34

34 anhydrous (99.9 + % ) were obtained from Sigma Aldrich (St Louis, MO). Valproic acid (2 p ropylpentanoic acid) was purchased from MP Biomedicals (Solon, OH). Carbonyl diimidazole (CDI) was purchased from Acros Organics (part of Thermo Fisher Scientific, New Jersey, US). Potassium phosphate monobasic, potassium phosphate dibasic, sodium phosphate dibasic, c itric acid (99%), tert b utyl methyl ether were all purchased from Sigma Aldrich (St. Louis, MO). Isopropyl alcohol (2 p ropanol, HPLC grade ), acetonitrile (HPLC grade), ethyl acetate (HPLC grade), methanol (HPLC grade), t riethylamine (HPLC grade) were obtained from Fisher Scientific (Pittsburgh, PA). Hydrochloric acid (2 N), ortho phosphoric acid (85%), sodium bicarbonate, sodium carbonate, s o dium hydroxide and m olecular sieves (4 ) were all purchased from Fisher Scientific (Pittsburgh, PA). Dimethyl sulfoxide d 6 (99.96 atom % D) was purchased from Sigma Aldrich (St. Louis MO). All the buffers used in the experiments were freshly prepared. The HPLC instrument used was from Waters (Milford, MA) and equipped with Waters 717 Plus Autosampler, Waters 2587 Du Waters 1525 Binary HPLC Pump. NMR spectra were obtained with a Bruker, Avance II 600 Spectrometer with 5 mm TXI CryoProbe sample head, operating at 600 MHz. FTIR spectra were recorded on a Thermo Nicolet NEXUS 670 FTIR Spectrom eter (Thermo Scientific, part of Thermo Fisher Scientific, Pittsburgh, PA). Ultrasonic bath (FS110H) and h omogenizer (Power Gen 125) used were from Fish er Scientific (Pittsburgh, PA). Centrifuge (Eppendorf Centrifuge 5415D) was from Eppendorf (Hauppauge, N Y). Another centrifuge (Sorvall Legend RT) was from Thermo Scientific (part of Thermo Fisher Scientific, Pittsburgh, PA). Rotary eva porators

PAGE 35

35 (Rotavapor RE121 and B chi 461 water bath) were from B chi (Switzerland). Magnetic stirrer (Bell Stir Multi Stir 9) was from Bellco Glass (Vineland, NJ). The incubator used for the i n vitro hydrolysis study was a CO 2 water jacketed incubator (NU 2600) from Nuaire (Plymouth, MN). Synthesis of The Conjugates The conjugate of valproic acid dextran was synthesized accordin g to the scheme presented in the Figure 2 1 ( 112 115 ) V alp roic acid (6 mmol) was added to 10 ml o f dried DMSO, followed by addition of 6.6 mmol (1:1.1 ratio) of c arbon yl diimidazole (CDI), while stirring on a magnet ic stirrer. The resulting solution was stirred continuously for 24 hours Argon gas was used to displace the air on top of the reaction mixture Also 3.0 g of d ext ran (Mw 10 kD, or 70 kD) w as added to 30 ml (for 10 kD dextran) and 60 ml (for 70 kD dextran) of anhydrous DMSO in a separate conic al flask and the mixture was heated at 70C for 30 minutes, while stirring on a magnetic stirrer, to help dissolve dextran in DMSO. Subsequently, 6 ml of triethyla mine (TEA) was added and the solution w as stirred for 10 minutes. Approximately 20 g of dr ied molecular sieves (4 ) was added to this s olution and the mixture was allowed to stand for 24 hours at room temperature. T he two solutions were added together aft er 24 hours in a spinner flask under the positive pressure of Argon gas, and about 8 g of dried molecular sieves was added to this soluti on. The reaction mixture was left for 24 hours at 70C under continuous stirring for conjugation to occur. Finally, con jugates were precipitated by the additi on of ethyl acetate After prec ipitation, the supernatant was discarded and the conjugates were dissolved in DMSO and again precipitated with the addition of ethyl acetate Repeat the process

PAGE 36

36 once again. At the end, t he supernatant was discarded and the conjugates were dispersed in methanol. Finally, the conjugates were dried in a rota ry e vap orator for 12 hours to remove the excess solvent Characterization of The Conjugates By IR and NMR Spectroscopy Infrared (IR) Cha racterization. The FTIR spectra from the VPA dextran conjugates were recorded on a Thermo Nicolet NEXUS 670 FTIR. All samples were characterized, using KBr pellets made with the different conjugates. NMR Characterization. 1 H NMR spectra were obtained with a Bruker, Avance II 600 Spectrometer with 5 mm TXI Cryo Probe sample head and Magnex 14.1 T/54 mm AS Magnet operating at 600 MHz. The sample tube size was 5 mm with a sample con centration of 50 mg/0.5 ml for valproic a cid, 10 mg/0.5 ml for the dextran 10 k D 7 mg/0.5 ml for the dextran 70 kD as control substances, 2 0 mg/0.5 ml for conjugate VPA D10 and 16 mg/0.5 ml for conjugate VPA D70. DMSO d 6 (99.96%) was used as a solvent. All of the chemical shifts are reported in parts per million (ppm) with tetrameth ylsilane as an internal standard. All t he samples were measured at 37 C. Reversed Phase HPLC Conditions All the s amples were analyzed by reversed phase HPLC ( High Performance Liquid Chromatography ) technique The instrument was equipped with a Wate rs 717 P lus Autosampler connected with a Waters 1525 Binary HPLC Pump. The quanti fication of VPA was carried out by a n established procedure ( 116 ) The mobile phase consisted of phosphate buffer: acetonitrile: isopropanol (60:25:15). The phosphate buffer used in the mobile phase had molarity of 0.05M and p H 3.0.

PAGE 37

37 The injec tion volume of all the samples was 50 l and the flow ra te was set to 0.8 ml /min The samples were detected at an wavelength of 220 nm and the run time was 10 minutes The used column was a Symmetry C18 (5.5 m, 4.6 x 50 m m). The peak area was quantified w ith Breeze TM software Calibration curves were generated with standard solutions in an appropriate concentration range during the analytical run for each series of samples Standard solutions were always freshly prepared before analysis Degree of Substitu tion of The Conjugates Degree of substitution (DS%) was defined as the drug content in mg per 100 mg of the dextran conjugate expressed as percentage. To determine degree of substitution 10 15 mg of t he VPA dextran conjugate was dissolved in 10 ml of NaOH soluti on (0.5 N). The solution was kept in the refrigerator for complet e hydrolysis and samples were collected at different tim e points for the next 48 hours ( 113 117 ) For the sample collection 500 l of the solution was collected into 1 ml HPLC vials and 500 l of HCl solution (0.5 N) was added to the v ials to neutralize the base All the experiments wer e done in triplicate. The released valproic acid was analyzed by established HPLC method ( 118 ) Then to calculate the degree of substitution, the total amount of drug per conjugate sample had to be calculated using the equation below. Then the percentage of degree of substitution (DS%) was calculated according to the following equatio n

PAGE 38

38 Aqueous Solubility of The Conjugates To determine aqueous solubility of the conjugate at room temperature approxima tely 50 mg of conjugate was added to 1.5 ml of double distil led water. The mixture was then sonicated in a bath sonicator for 60 minutes to dissolve as much conjugate as possible After that the mixture was centrifuged for 10 minutes at 13,200 rpm to precipitate the un dissolved conjugate. Then, 1 ml of the supernatant was added to 10 ml of NaOH solution (0.5 N) and the solution was kept in the refrigerator for 48 hours to perform complete hydrolysis ( 119 ) At different time points, 500 l of each sa mple was collected into 1 ml HPLC vials and 500 l of HCl solution (0.5 N) was added to each vial to neutralize the base All t he experiment s were done in triplicate. The collect ed samples were stored in the refrigerator before being anal yzed by HPLC Then to calculate the solubility of the conjugate (mg/ ml ), the total amount of drug per conjugate sample dissolved in 1 ml of water had to be calcu lated using the equation below. Then the actual solubility of the conjugate (mg/ ml ) was calculated according to the following equation In vitro Hydrolysis Study In vitro hydr olysis study was performed w ith the conjugate prepared with dextran (Mw 10 kD) (VPA D 10) in three different buffer solutions citric acid/Na 2 HPO 4 buffer (p H 5.0), phosphate buffer (pH 7.5) and NaHCO 3 /Na 2 CO 3 buffer (p H 10.22) at two different temperatures of 37 C and 47 C. In vitro hydrolysis study was also performed

PAGE 39

39 with the conjugate prepared with dextran (Mw 7 0 kD ) (VPA D 7 0) in phosphate buffer and bicarbonate buffer at both temperatures. Hydrolysis of the conjugates at different pH values and temperatures would help us to deter mine the effect of these two factors on the rates of hydrolysis of the conjugates. These studies would also help us to understand the stability of the conjugates at different pH values and temperatures. In vitro Hydrolysis with Conjugate VPA D 1 0 Approxim ately 40 mg of the conjugate was dissolved in 50 ml of each buffer solution. The hydrolysis was conduc ted at 37 C and 47 C in an incubator under continuous stirring of the samples with a magnetic stirrer running at 100 rpm Samples of 1 ml were collected at different time intervals and 1 ml of respective buffer solution was added to the sample solution to keep the volume constant. For conjugate dissolved in citrate buffer at both 37 C and 47 C, samples were collected after 2, 4, 6, 8, 10, 12 and 14 days Samples were collected after 1, 3, 5, 7, 9, 11 and 13 days for conjugate dissolved in phosph ate buffer at 3 7 C, and after 1, 2, 3, 4, 5, 6 and 7 days at 4 7 C Time intervals of sample collection for the conjugate dissolved in bicarbonate buffer at 37 C were after 6, 12, 24, 36, 48, 60, 72, 82, 97 and 106 hours and after 4, 8, 12, 15, 24, 27, 31 and 35 hours at 47 C. All the experiments were done in triplicate for each buffer solution and at each temperature. All the samples were kept in the freezer at 20 C before being analyzed by HPLC. In vitro Hyd rolysis with Conjugate VPA D 70 Approximately 25 mg of the conjugate was dissolved in 50 ml of each buffer solution. The hydrolysis was carried out in an incubator under continuous stirring of the samples with a magnetic stirrer running at 100 rpm and at temperatures of 37 C and 47 C. Samples of 1 ml were collected at

PAGE 40

40 different time intervals and 1 ml of respective buffer solution was added back to the sample solution to replenish the volume Samples wer e collected after 1, 3, 5, 7, 9, 11 and 13 days for conjugate dissolved in phosphate buffer at 37 C, and after 1, 2, 3, 4, 5, 6 7 and 8 days at 47 C. Time intervals of sample collection for the conjugate dissolved in bicarbonate buffer at 37 C were aft er 6, 12, 24, 36, 48, 60, 72, 82, 97 and 106 hours and after 4, 8, 12, 15, 24, 27, 31 and 35 hours at 47 C. All the experiments were done in triplicate for each buffer solution and at each temperature. All the samples were kept in the freezer at 20 C be fore being analyzed by HPLC. Statistical Analysis One multiple comparison test was used to compare the rate constants of i n vitro hydrolysis of the conjugates. P value <0.05 was considered significant. Results and Discussion Synthesis of Dextran Valproic A cid Conjugates and Spectral Analysis The dextran valproic acid conjugates were synthesized according to the procedures mentioned for other drugs in several papers with slight modifications ( 77 113 120 ) Two types of the conjugates were synthesized using two different mol ecular weights of dextran. Conjugate VPA D10 was synthesized using dextran average molecular weight 10 kD and VPA D70 was synthesized using dextran average molecular weight 70 kD. The synthesis was carried out usi carbonyldiimidazole as a coupling a gent. The carbonyldiimidazole method is very efficient and particularly suitable for the conversion of the biopolymers like dextran. It allows the use of DMSO which is a good

PAGE 41

41 solvent for dextran. Also, DMSO does not get oxidized a nd decomposed with the use of carbonyldiimidazole The reaction by products are only CO 2 and imidazole which are non toxic. The imidazole is freely soluble in water and can be easily removed. The actual synthetic scheme was carried out in two steps. First, valproic acid was transfo rmed with carbonyldiimidazole to give the corresponding imidazolide and second, the imidazolide was incorporated with dextran to form the ester in the presen ce of a base triethylamine The c onjugates were precipitated by the addition of ethyl acetate The yields after purification were 53.83% for VPA D10 conjugate and 54.01% for VPA D70 conjugate. Structures of the valproic acid dextran conjugate and the formation of the ester bond were confirmed by performing FT IR and 1 H NMR spectroscopy. The FT IR spectr um of valproic acid showed a strong absorption band at 1706 cm 1 (C=O stretching) (Figure 2 2) The spectra of the two different dextrans (Mw 10 kD and 70 kD) looked very similar (only one shown) and had bold bands around 3426 cm 1 (O H stretching) ( 121 ) (Figure 2 3) Valproic acid dextran conjugates showed the shift of peak from 1706 cm 1 to 1733 cm 1 (C=O stretching) (Figure 2 4 and Figure 2 5) indicated the formation of ester bond between valproic aci d and dextran molecules. The appearance of bold band around 3427 cm 1 (unreacted residual O H stretching of dextran molecules) (Figure 2 4 and Figure 2 5) in the spectra of valproic acid dextran conjugates supported further the formation of the conjugate b etween valproic acid and dextran molecules. The 1 H ppm, these broad bands belonged to the protons of anhydroglucose unit (AGU) and

PAGE 42

42 protons of the hydroxyl groups of the dextran molecul es ( 122 126 ) except for the band at = 3.3 (Figure 2 6 and Figure 2 7 ) Literature showed that this band at = 3.3 belonged to the protons of the water mol ecules that were present in the dextran samples. T he bands for the protons of valproic acid appeared in There was also one band present rom the carboxylic acid group, which was present prominently in the spectrum of va lproic acid. This peak was very small or near absent in the NMR spectra of the conjugates (Figure 2 6 and Figure 2 7 ) was present in all the spectra and represented part of the DMSO used as solvent which was not deuterated completely (e .g. DMSO d 5 ). Due to the broad bands for the conjugates a nd the fact that there were bands for both the valproic acid and the dextran, present in the spectra and the absence of the carboxyl provided strong indic ation for the formation of the conjugate between valproic acid and dextran. Degree of Substitution and Solubility of The Conjugates The synthesized conjugate VPA D 10 had a degree of substitution % of 11.35 0.32 and VPA D70 had a degree of substitution % of 21.02 0.27 (Table 2 1 ) The values of degree of substitution of the two conjugates were quite satisf actory in comparison with the values of degree of substitution of other dextran conjugates found in the literature ( 92 113 127 128 ) D egree of subst itution % of VPA D10 was lower than that of VPA D70 (Figure 2 8 ). It was evident that v alproic acid was conjug ated to a higher extent to the d extran Mw 70 kD co mpared to the d extran Mw 10 kD. Actually, t his was in contrast to the findings of the study cond ucted by Varshosaz et al ( 113 ) They synthesized budesonide dextran conjugates with different molecular

PAGE 43

43 weights of dextran (10 kD, 70 kD and 500 kD) in the pr esence of dimethylaminopyridine using succinate spacer. They con cluded that degree of substitution % was decreased as the molecular weights of the dextran used in the conjugate increased. They hypothesized that this might be due to the steric hindrance of the bulky structure of the dextran molecules preventing the effective interaction between hydroxyl groups and carboxylic group of budesonide hemisuccinate. Additionally, they pointed out that in a solvent like DMSO, dextran molecules are highly folded and only the surface hydroxyl groups could be available for the reaction ( 129 ) In our case, the both conjugates were synthesized in parallel following the same procedure except for the volume o f DMSO used to dissolve the dextran. Because, previous attempt to synthesize VPA D10 conjugate using 60 ml of DMSO resulted in no precipitation of the conjugate upon addition of several organic solvents. By lowering the volume of DMSO to dissolve the dextr an molecules, precipitation of the VPA D10 conjugate was successful upon addition of ethyl acetate. or the conjugate VPA D10, 30 ml of DMSO was used to dissolve the dextran whereas for the conjugate VPA D70, 60 ml of DMSO was used to dissolve the dextran. So, the use of lower volume of DMSO made the dextran solution more concentrated in the synthesis of VPA D10 compared to that in the synthesis of VPA D70. Concentrated dextran solution might have made less dextran molecules available for conjug ation with activated valproic acid molecul es and resulted in lower degree of substitution % of the conjugate VPA D 10 compared to that of VPA D70. Aqueous solubility of the conju gates VPA D10 and VPA D70 at room temperature was 1.03 0.05 mg/ml and 0.63 0.03 mg/ml respectively (Table 2 2 )

PAGE 44

44 In vitro Hydrolysis Stud ies In vitro hydrolysis was performed using conjugate VPA D 10 in citrate buffer, phosphate buffer and bicarbona te buffer at 37 C and 47 C Hydrolysis of the conjugate VPA D 70 was performed in phosphate buffer and bicarbon ate buffer at both temperatures, but not in citrate buffer. This was because of the fact that conjugate VPA D10 was found stable and not hydrolyzed at all in citrate buffer at both temperatures (data not shown). The conjugate VPA D10 showed negligible, or no hydrolysis in acidic me dium (pH 5.0). The rate s of hydrolysis were slow for both conjugate s in phosphate buffer (pH 7.5) and much faster in basic bicarbonate buffer (pH 10.22). Rate constants for hydrolysis and R 2 were calc ulated and compa red for each conjugate. By plotting the results of hydrolysis of both conjugates in phosph ate buffer (pH 7.5) at the temperatures of 37C and 47C it was evident that data for the hydrolysis of both conjugates could be adequately described by both zero and first order kinetic models Carstensen ( 130 ) showed that if degradation was less than 15%, it would be difficult to distinguish a first order reaction from a zero order reaction. But when percent of hydrolysis of both conjugates was higher in case of bicarbon ate buffer (p H 10.22) at both te mperatures; first order kinetic model gave better fit to the data compared to zero order model. So, it could be concluded that hydrolysis of the conjugates followed first order reaction kinetics in bicarbonate buffer at 37C and 47C and it could be assumed that hydrolysis of the conjugates followed first order reaction kinetics in phospha te buffer at both temperatures. First order plots of hydrolysis of both conjugates in phosphate buffer and bicarbonate buffer at 37 C and 47 C were presented in Figure 2 10, Figure 2 11,

PAGE 45

45 Figure 2 12 and Figure 2 13. Estimated observed rate constants (from the slopes of the plots) and calculated t 1/2 values for the hydrolysis of the conjugates were presented in the Table 2 3 By comparing al l the rate constants at different pH values and temperatures several findings were identified. These are summarized as follows Different pH values of different buffers seemed to have more pronounced effect on different rates of hydrolysis of the conjugat es (k 1 vs. k 3 P< 0.001; k 2 vs. k 4 P< 0.001; k 5 vs. k 7 P< 0.001 and k 6 vs. k 8 P< 0.001). T emperature seemed to have more pronounced effect on the r ates of hydrolysis at a higher pH 10.22 (k 3 vs. k 4 P< 0.001 and k 7 vs. k 8 P< 0.001). Degree of substitut ion and solubility of the conjugates can affect the rates of hydrolysis of the conjugates in an opposite direction. However, d egree of substitution and solubility seemed to have more pronounced effect on the rates of hydrolysis at a higher pH 10.22 at both temperatures (k 1 vs. k 5 P> 0.05, k 2 vs. k 6 P> 0.05, k 3 vs. k 7 P< 0.001 and k 4 vs. k 8 P< 0.001 ), but did not seem to have any noticeable effect on the rates at pH 7.5 at both temperatures. Johansen et al. ( 131 ) reported similar findings in their study where they studied kinetics and mecha nism of hydrolysis of O benzoyl dextran conjugates in aqueous buffer and human plasma. They reported at pH 7.4, almost identical rates of hydrolysis were observed for conjugates with different molecular weights (10 110 kD) and degree of substitution (3.4 15. 8% w/w). Weibel et. al. ( 132 ) also reported that varying degree s of substitution of benzyl carbonate esters of dextran had no influence on the reaction rate s at pH 7.47

PAGE 46

46 The half lives of VPA D10 and VPA D70 conjugates in phosphate buffer (pH 7.5) at 37 C wer e approximately 62 days and 57 days respectively (P> 0.05) and at 47 C were approximately 28 days for both conjugates (P> 0.05). The two conjugates seemed to have longer half lives than those of some other dextran conjugates found in the literature ( 127 128 133 ) This might be due to less solubility and hi gher degree of substitution of both VPA D10 and VPA D70 conjugates and also due to higher pKa value of valproic acid. However, lo nger half lives of VPA D10 and VPA D70 conjugates made them attractive candidates for sustained release formulation s of valproi c acid. Conclusion The synthesized valproic acid dextran conjugates were shown to be hydrolyzed i n vitro over approximately 2 months period at pH 7.5 and 37 C and could be potential candidates for a new sustained release formulation of valproic acid. Furt her in vivo studies need to be conducted to optimize the formulation. Temperature and pH were also shown to affect the rate s of hydrolysis of the conjugates i n vitro where pH had a more pronounced effect than temperature on the rate s of hydrolysis of the conjugates. Degree of substitution and solubility of the conjugates had a more pronounced effect on the rate s of hydrolysis of the con jugates at pH 10.22 at 37 C and 47 C

PAGE 47

47 Table 2 1. Degree of substitution (DS%) of conjugates VPA D10 and VPA D70 from h ydrolysis study (mean sem, n=3). Conjugate code Dextran Mw DS % from hydrolysis study VPA D10 10 kD 11.35 0.32 VPA D70 70 kD 21. 02 0.27 Table 2 2. Aqueous solubility of the conju gates VPA D10 and VPA D70 at room temperature (mean sem, n=3). Co njugate code Dextran M w Solubility (mg/ml) VPA D10 10 kD 1.03 0.05 VPA D70 70 kD 0.63 0.03 Table 2 3. Rate constants and t 1/2 values of VPA D10 and VPA D70 conjugates in phosphate and bicarbonate buffer at 37 C and 47 C (mean sem, n=3). Conjug ate Buffer Temperature Symbol Rate constant, k obs (1/day) t 1/2 (day) VPA D10 Phosphate buffer (pH 7.5) 37 C k 1 0.0112 0.0014 62 47 C k 2 0.0249 0.0018 28 Bicarbonate buffer (pH 10.22) 37 C k 3 0.6230 0.0 370 1. 1 47 C k 4 1.120 0. 1095 0.6 2 VPA D70 Phosphate buffer (pH 7.5) 37 C k 5 0.0121 0.0014 57 47 C k 6 0.0248 0.0010 28 Bicarbonate buffer (pH 10.22) 37 C k 7 0.1532 0.0035 4.5 47 C k 8 0.4207 0.0 190 1. 6

PAGE 48

48 Figure 2 1. Synthesis of valproic acid dextran conjugate.

PAGE 49

49 Figure 2 2. FT IR spectra of VPA. Figure 2 3. FT IR spectra of dextran

PAGE 50

50 Figure 2 4. FT IR spectra of the conjugate VPA D10. Figure 2 5. FT IR spectra of the conjugate VPA D70.

PAGE 51

51 Figure 2 6 1 H NMR spectra of VPA (A), Dextran (B) and VPA D1 0 (C ). Figure 2 7 1 H NMR spectra of VPA (A), Dextran (B) and VPA D70 (C ).

PAGE 52

52 Figure 2 8. Degree of substitution (DS%) of the conjugates VPA D10 and VPA D70 (mean sem, n=3 ).

PAGE 53

53 Figure 2 9. Aqueous solubility of the conjugates VPA D10 and V PA D70 at room temperature (mean sem, n=3 ).

PAGE 54

54 Figure 2 10. In vitro hydrolysis of conjugate VPA D10 in phosphate buffer at 37 C and at 47 C (mean sem, n=3). Figure 2 11. In vitro hydrolysis of conjugate VPA D10 in bicarbonate buffer at 37 C an d at 47 C (mean sem, n=3)

PAGE 55

55 Figure 2 12. In vitro hydrolysis of conjugate VPA D70 in phosphate buffer at 37 C and at 47 C (mean sem, n=3) Figure 2 13. In vitro hydrolysis of conjugate VPA D70 in bicarbonate buffer at 37 C and at 47 C (mean sem n=3)

PAGE 56

56 CHAPTER 3 SYNTHESIS OF VALPROIC ACID LINKER DEX TRAN CONJUGATES: DES IGNING A PRODRUG FOR IN SITU ACTIVATION Introduction Hydrolysis of dextran prodrugs where drug is covalently bonded to the dextran molecules appeared to be pH and temperatur e dependent. This is evident from our study described in Chapter 2. However, this type of prodrugs also appeared to be stable towards enzymatic hydrolysis by various esterases ( 98 134 ) Hydrolysis of VPA D10 conjugate in presence of porcine liver esterase demonstrated no hydrolysis of the conjugate. It has been postulated that bulky molecular str ucture of dextr an molecules make the hydrolytic center inaccessible providin g steric hindrance and thus make these prodrugs resistant to enzymatic hydrolysis ( ( 89 98 134 ) Using a spacer arm to form a bond between the drug and the dextran molecule would be an approach to aid in the hydrolysis of the conjugate by esterase en zymes. Aft er convection enhanced delivery, this type of macromolecular prodrug might act as a local depot which will release the drug by enzymatic action. Alternatively, this macromolecular prodrug might also release the low molecular weight prodrug by che mical hydrolysis which in turn will be diffused away and activated at the target site by enzymatic action. So, synthesizing another class of dextran prodrugs with valproic acid having a spacer arm linking together might provide additional benefit of in si tu activation and thereby reducing even more drug induced adverse effects. This study aimed at synthesizing dextran valproic acid prodrug connected through a spacer arm suitable for enzymatic hydrolysis. There are two major human carboxylesterases (CESs) a human CES1 family isozyme (hCE1) and a hum an CES2 family isozyme (hCE2) ( 135 ) Human CE1 is highly expressed in the liver, while hCE2 is

PAGE 57

57 present in the small intestine, colon, kidney, li ver, heart, brain and testis ( 135 136 ) Also, a third brain specific carboxylesterase was isolated in 1999 ( 137 ) The substrate specificity of hCE1 and h CE2 is significantly different. In general, hCE1 mainly hydrolyzes a substrate with a small alcohol group and large acy l group, while hCE2 hydrolyzes a substrate with a large alcohol group and small acyl group ( 138 ) Webb reported the reactivity and affinity of 33 esters, where th e lengths of both acyl and alkyl chain varied from C1 to C9, with respect to horse liver carboxylesterase ( 139 ) He concluded that both the affinity and reactivity increased as e ither chain was lengthened to about C4 to C6 ( 140 ) A further increase in the alkyl chain length resu lted in a decrease in both the affinity and reactivity. Similarly, branched chain substrates exhibited higher affinities but lower reactivities ( 141 ) Considering these, 1,6 hexanediol was chosen as a linker for the synthesis. Triethylene glycol was chosen as a relatively more polar alternative to hexanediol, if solubility becomes an issue for the conjugates. Dextr ans having Mw 70 kD and 10 kD were selected to be the macromolecular carrier. The rationale for selecting dextrans with different molecular weights was explained in the previous chapter. So, the aims of the study were to synthesize valproic acid dextran co njugates connected through either hexanediol or triethylene glycol as a linker, to charac terize the conjugates with NMR spectroscopy, and to study degree of substitution and aqueous solubility of the conjugates Materials and Methods Chemicals and Instrume nts Dextran from Leuconostoc mesenteroides (average M w 64 76 kD ) d extran from Leuconostoc mesenteroides (average M w 9 11 kD ) d icyclohexylcarbodiimide

PAGE 58

58 (DCC), dimethylaminopyridine (DMAP), d imethyl sulfoxide (DMSO) anhydrous (99.9+%) were purchase d from Sigma Aldrich (St Louis, MO). Valproic acid (2 p ropylpentanoic Carbonyldiimidazole (CDI) was purchased from Acros Organics (part of Thermo Fisher Scientific, New Jersey, US). Potassium phosp hate monobasic, p otas sium phosphate dibasic, sodium phosphate dibasic, s odium hydrogen sulfate, 1,6 h exanediol, t riethyl ene glycol t etrahydrofuran (THF) (2 hydroxypropyl) cyclodextrin and m olecular sieves (3) were all purchased from Sigma Aldrich (St. Louis, MO). I sopropy l alcohol (2 p ropanol, HPLC grade), 1 butanol, d imethylformamide (DMF), acetonitrile (HPLC grade), ethyl acetate (HPLC grade), d i chloromethane (HPLC grade) and m ethanol (HPLC grade) were obtained from Fisher Scientific (Pittsburgh, PA) Tween 80 and PEG 300 were also purchased from Fisher Scientific (Pittsburgh, PA). Hydrochloric acid (2N), ortho phosphoric acid (85%), magnesium sulfate, sodium bicarbonate, sodium carbonate, and s odium hydroxide were all purchased from Fisher Scientific (Pittsburgh, PA). Dimethyl sulfoxide d 6 (99.96 atom % D) was purchased from Sigma Aldrich (St. Louis, MO). The HPLC instrument used was from Waters (Milford, MA) and equipped with ters 1525 Binary HPLC Pump. NMR spectra were obtained with a Bruker, Avance II 600 Spectrometer with 5 mm TXI CryoProbe samplehead, operating at 600 MHz Ultrasonic bath (FS110H) and h omogenizer (Power Gen 125) used were from Fisher Scientific (Pittsburgh, PA). Centrifuge (Eppendorf Centrifuge 5415D) was from Eppendorf (Hauppauge, NY). Another centrifuge (Sorvall Legend RT) was from Thermo Scientific

PAGE 59

59 (part of Thermo Fisher Scientific, Pittsburgh, PA). Rotary evaporators (Rotavapor RE121 and Bchi 461 water bath ) were from Bchi (Switzerland). Synthesis of The Conjugates Three types of conjugates were synthesized using two different linkers and using different molecular weights of dextran The purpose was to synthesize different valproic acid linker dextran c onjugate s for in situ activation. The design of the synthesis is described in Figure 3 1. The overall synthetic procedures for all the conjugates were mostly the same with minor difference in the extraction procedure. Total synthetic procedure is broken do wn into four steps. Step 1: synthesis of valproic acid linker conjugate, step 2: extraction of the valproic acid linker conjugate, step 3: synthesis of valproic acid linker dextran conjugate, and step 4: extraction of the valproic acid linker dextran conju gate. These steps are described as follows. Stpe 1: Synthesis of valproic acid linker conjugate Valproic acid was esterified with hexanediol using Steglich esterification ( 142 ) About 46.43 mmol of th e linker ( 5.484 g for 1,6 hexanediol and 6.196 ml for triethylene glycol) was added to 20 to 40 ml of dried acetonitrile in a conical flask and dissolved it with the help of magnetic stirrer (sol A). Then, 20 ml dried dicholoromethane (DCM) (dried over mol ecular sieves) was taken in a 125 ml conical flask and sol A was added slowly to DCM in this flask while shaking (sol B). Now, 1848 L (1664 mg, or, 11.6 mmol) of valproic acid, 432.2 mg (3.49mmol) of dimethyl aminopyridine (DMAP) were added and the flask was purged with argon gas (sol C). Now, 2631.4 mg (12.76 mmol) of dicyclohexyl carbodiimide (DCC) was dissolved in 10 ml of dried DCM and this solution was added to sol C drop wise for 5 minutes. The

PAGE 60

60 flask was purged with argon gas again. After that the r eaction was continued at room temperature for 24 hours. Step 2: Extraction of valproic acid linker conjugate After 24 hours, the VPA linker conjugate was extracted from the reaction mixture. The reaction generated urea which was precipitated in the react ion mixture. This urea was filtered out from the reaction mixture by using conventional paper filter. Then, the reaction mixture was evaporated in a rotary evaporator under high vacuum at 55 C for about 15 20 minutes and the reaction mixture was concentra ted to a small amount of oily liquid (about 5 ml ). About 10 ml of dried DCM was added to dissolve the oily liquid and transferred it into a separatory funnel. 10 ml of 5% NaHSO 4 solution was added to the funnel and the funnel was shaken vigorously for 5 to 10 minutes to have unreacted DMAP dissolved in the aqueous layer. The upper aqueous layer was removed and discarded. The wash with 5% NaHSO 4 solution was repeated one time to ensure complete removal of DMAP from the reaction mixture. Now, 10 ml of 5% NaH CO 3 solution was added to the funnel and funnel was also shaken vigorously for 5 to 10 minutes to remove excess acid from the reaction mixture. The upper aqueous layer was also removed and discarded. The wash ing step with 5% NaHCO 3 solution wa s repeated on e more time for completeness. At the end of the extraction, the organic phase (with some precipitated conjugates) was collected in a conical flask and 80 ml of DCM was added to dissolve the precipitated conjugate completely. Then, 5 g of dry MgSO 4 was adde d and was allowed to stand for 30 minutes to remove water from the organic solvent. After that, the salt was filtered out completely with a conventional paper filter and the organic

PAGE 61

61 liquid was transferred in a rotary evaporator and dried under high vacuum at 50 C to complete dryness for about 3 4 hours. Step 3: Synthesis of the valproic acid linker dextran conjugate The extracted valproic acid linker conjugate was dissolved in THF in an argon atmosphere in a conical flask. Then the appropriate amount of c arbonyldiimidazole (CDI) was added to it for the activation of the conjugate (the number of millimoles of CDI necessary for the activation was calculated by multiplying the number of millimoles of valproic acid linker conjugate with 1.1). This solution was then stirred for 27 hours at room temperature. Afterwards the solvent was evaporated in a rotary evaporator at 50 C under reduced pressure yielding viscous oil. This product was used without further purification. For each conjugate about 1.5 g of freeze dried Dextran (M W 70 k or, 10k ) was dissolved in 10 ml of dried DMSO and 1 g of molecular sieves 3 was added to this solution in an argon atmosphere. Activated valproic acid linker conjugate was also dissolved in 10 ml of dried DMSO and was added to dext ran solution under argon atmosphere. At last dimethylaminopyridine (DMAP) was added to the flask as nucleophilic catalyst to induce coupling of activated conjugate to dextran in an argon atmosphere. The number of millimoles of DMAP needed was calculated by multiplying the number of millimoles of valproic acid linker conjugate with 1.5. The flask was then placed on a magnetic stirrer and kept on running at 50 C for 4 days. Step 4: Extraction of the valproic acid linker dextran conjugate Ethyl acetate was u sed to precipitate va lproic acid linker dextran conjugate 5 ml of the reaction medium was transferred into 4 centrifugal tubes and to each tube 25 ml of organic solvent was added to precipitate the conjugate. The mixtures were then centrifuged at

PAGE 62

62 8000 rpm for 10 minutes. After that the supernatant was discarded and the conjugate was dissolved in 6 ml of DMSO with the help of a homogenizer. Afte rwards the conjugate was precipitated again by adding 25 ml of organic solvent and then, centrifuged. These steps were repeated 3 more times. Finally, 15 ml of methanol was added to each tube to disperse the conjugate and all the dispersion was transferred into a round bottom flask and dried in a rotary evaporator for 15 hours to remove all the organic solvents. The o verall synthetic scheme is described in Figure 3 1 Characterization of The Conjugates by NMR Spectroscopy The 1 H NMR study was performed in AMRIS (Advanced Magnetic Resonance Imaging and Spectroscopy) facility at McKnight Brain Institute in the University of Florida. Two different Bruker Avance Consoles were used to obtain the spectra: Bruker Avance II 600 Console with 5 mm TXI CryoProbe sample head and Magnex 14.1 T/54 mm AS Magnet operating at 600 MHz and Bruker Avance 500 Console with 5 mm BBO Probe an d Magnex 11.75 T/54 mm Magnet operating at 500 MHz The samp le tube size was 5 mm. All of the chemical shifts are reported in parts per million (ppm) with tetramethylsilane as an internal standard. T he three synthesized conjugates were analyzed at a tempe rature of 37C. Furthermore the structures of two intermediate conjugates and the two linkers were determined to confirm the synthesis of the conjugates and to facilitate interpretations of all the spectra Conjugates VPA HXD D70 and VPA HXD D10; the two i ntermediate conjugates VPA HXD and VPA TEG ; and the linkers HXD and TEG were characterized by Bruker Avance II 600 Console. Only the

PAGE 63

63 conjugate VPA TEG D10 was characterized by Bruker Avance 500 Console. DMSO D 6 (99.9%) was used as solvent to dissolve the p roducts. Aqueous Solubility of The Conjugates The conjugate needed to be dissolved in aqueous phosphate buffer solutions to conduct hydrolysis study in presence of different enzyme solutions. However, all of the synthesized conjugates appeared to be insol uble in aqueous solutions, which made it difficult to carry out any further study Thus, several solubilizing tests were performed with various organic solvents, surfactants or other organic compounds that might enhance their solubility. These approaches a re summarized in the Table 3 1. The details of all the solubilizing tests are attached in the appendices A and B Degree of Substitution of The Conjugates Degree of substitution (DS%) is defined as the drug content in mg per 100 mg of the dextran conjugate expressed as percentage. Basic hydrolysis method was used t o determine degree of substitution of the conjugates. Since the conjugates synthesized with spacer are completely insoluble in aqueous buffer solutions, the degree of substitution of the conjugates were determined by 1 H NMR spectroscopy in DMSO D 6 The degree of substitution of the conjugates can be calculated by using the following equation ( 143 ) : A (VPA) is the area of the integrated peaks of the protons of valproic acid and A (anomeric H) is the area of the anomeric proton of the dextran monomeric unit. These two areas

PAGE 64

64 the valproic acid was divided by 15 and the area of the anomeric proton needed to be divided by 1. Results and Discussion Spectral Analysis Analysis of 1 H NMR spectra of VPA HXD The Figure 3 2 and Figure 3 4 showed 1 H NMR spectra of valpr oic acid (A), 1,6 hexanediol (B) and valproic acid hexanediol (VPA HXD ) (C). The 1 H NMR spectrum showed the bands for the protons of spectrum of valproic acid. Three peaks ap belonged to the methylene protons of hexanediol The peak for two hydroxyl proton s 2 5. The conjugate VPA HXD had the peak for six methyl protons of valproic acid methylen e protons of valproic acid and eight methylene valproic acid appeared k for four methylene protons of and the peak for one h ydroxyl proton at 4. 2 5 DMSO (e.g. DMSO d5). The peak s at or around the conjugate. The peak for carboxyl proton from valp ro ic acid was not pr esent in the spectrum of VPA HXD which indicated the formation of ester bond between valp roic acid and hexanediol The fact that there were bands for both valp roic acid and hexanediol p resent in the spectra of VPA HXD and the absence of the carboxyl proton from v alproic acid at or

PAGE 65

65 provided strong indication for the formation of the conjugate between valp roic acid and hexanediol Analysis of 1 H NMR spectra of VPA HXD D70 conjugate The Figure 3 3 showed 1 H NMR spe ctra of valproic acid hexanediol (VPA H XD ) (A), dextran (B) and valproic acid hexanediol dextran conjugate (VPA HXD D70 ) (C). The 1 H NMR spectrum of dextran showed belonged to the protons of anhydroglucose unit (AGU) and protons of th e hydroxyl groups of the dextran molecules ( 144 ) except for the band at = 3.3. This band at = 3.3 belonged to the protons of the water molecules that were present in the dextran samples. The conjugate VPA HXD had the peak for six methyl protons of valproic acid methylene protons of valproic acid and eight methylene and 1.6. The methine proton of valproic acid appeared The pea k for four methylene protons of and the peak for one hydroxyl pro ton at 4.25 DMSO (e.g. DMSO d5). The peak s at or around the conjugate. The peak for carboxyl proton from valpro ic acid was not pr esent in the spectrum of VPA HXD. In the spectrum of VPA HXD D70 the peak from six methyl protons from valproic acid appeared at 3. All the methylene protons of valproic acid and eight e methine proton of valproic acid usually appeared

PAGE 66

66 peak from DMSO protons of AGU of dextran molecules (except for the anomeric proton), methylene protons at C 6 of AGU and from the four methyle ne protons of hexanediol These peaks were overlapping with each other. The peaks from three hydroxyl protons and one presence of hydroxyl proton from VPA HXD conjugate 2 5 could not be identified, which indicated strongly the formation of carbonic ester bond between hexanediol and the dextran molecules and this also indicated the formation of the conjugate VPA HXD D70. Analy sis of 1 H NMR spectra of VPA HXD D 1 0 conjugate The Figure 3 5 showed 1 H NMR spectra of valproic acid hexanediol (VPA HXD) (A), dextran (B) and valproic acid hexanediol dextran conjugate (VPA HXD D10) (C). The 1 H NMR spectrum of dextran showed chemical shi belonged to the protons of anhydroglucose unit (AGU) and protons of the hydroxyl groups of the dextran molecules except for the band at = 3.3. This band at = 3.3 belonged to the protons of the water molec ules that were present in the dextran samples. The conjugate VPA HXD had the peak for six methyl protons of valproic acid methylene protons of valproic acid and eight methylene nd 1.6. The methine proton of valproic acid appeared The pea k for four methylene protons of and the peak for one hydroxyl proton at 4.25 o partially deuterated

PAGE 67

67 DMSO (e.g. DMSO d5). The peak s at or around the conjugate. The peak for carboxyl proton from valpro ic acid was not pr esent in the spectrum of VPA HXD. In the spectrum of VPA HXD D10, the peak fr om six methyl protons from valproic acid appeared at 3. All the methylene protons of valproic acid and eight proton of valproic acid usually appeared und to merged with the large peak from DMSO to ring protons of AGU of dextran molecules (except for the anomeric proton), methylene protons at C 6 of AGU and from the four methylene protons of hexa nediol. presumably from the water present in the conjugate. The peaks from three hydroxyl protons and one anomeric proton of AGU of The presence or absence of hydroxyl proton from VPA because the peaks from three hydroxyl protons and one anomeric proton of AGU of ks were not sharp and very hard to separate one from the other. There were also three peaks present at the spectrum of VPA HXD D10, these were were presumably from the protons of imidazole, which were present with the conjugates as impurity. The fact that there were bands for both valproic acid and dextran present in the spectra and the fact that valproic acid could not be conjugated to the dextran in this

PAGE 68

68 case without hexanediol conjugated to the dextran, provided indication for the formation of the conjugate VPA HXD D10 Analysis of 1 H NMR spectra of VPA TEG The Figure 3 6 showed 1 H NMR spectra of valproic acid (A), triethylene glycol (B) and valproic acid triethylene glycol (VPA TEG) (C). The 1 H NMR spectrum sho wed the bands for the protons of valproic 12 from the carboxylic acid group, which was present prominently in the spectrum of valproic acid. Three peaks appeared in between methylene protons of triethylene glycol. The hydr The conjugate VPA TEG had all the methyl, methylene and methine protons otons of triethyl ene glycol and hydroxyl proton at 4.5 (e.g. DMSO ate. The peak for carboxyl proton from valpro ic acid was not present in the spectrum of VPA TEG, which indicated the formation of ester bond between valproi c acid and triethylene glycol. The fact that there were bands for both valproic acid and triethylene glycol present in the spectra of VPA TEG and the absence of the carboxyl proton from v alproic acid at provided strong indication for the formation of the conjugate between valproic acid and triethylene glycol. Analysis of 1 H NMR spectra of VPA TEG D1 0 conjugate The Figure 3 7 showed 1 H NMR spectra of valproic acid triethyle ne glycol (VPA TEG) (A), dextran (B)

PAGE 69

69 and valproic acid triethylene glycol dextran conjugate (VPA TEG D1 0 ) (C). The 1 H NMR spectrum of dextran showed bands belonged to the protons of anhydroglucose unit (AGU) and protons of the hydroxyl groups of the dextran molecules except for the band at = 3.3. This band at = 3.3 belonged to the protons of the water molecules that were present in the dextran samples. T he bands for the protons of VPA TEG conjugate a ppeared in 2.3 glycol) and at 4.5 (hydroxyl proton from triethylene glycol). was present in all the spectra and represented part o f the DMSO used as solvent which was not deuterated completely (e.g. DMSO d 5 ). In the spectrum of VPA TEG D1 0 protons from valproic acid appeared in between AGU of dextran mo lecules (except for the anomeric proton), methylene protons at C 6 of AGU and all the methylene protons of triethylene glycol. These peaks were overlapping with each other. The presence could be due to hydroxyl proton from VPA TEG indicated that there might be some VPA TEG present in the sample which might not have conjugated to the dextran molecules and remained physically mixed with the final conjugate VPA TEG D10. The fact that there were bands for both valproic acid and dextran present in the spectra and the fact that valproic acid could not be conjugated to the dextran in this case without triethylene glycol conjugated to the dextran, provided indication for the formation of the conjugate VPA TEG D1 0

PAGE 70

70 Solubility Analysis The con jugate VPA HXD D70 was found to be insoluble in aqueous buffer solutions. So, no further study could be done with this conjugate. Subsequently, conjugates VPA HXD D10 and VPA TEG D10 were synthesized using lower molecule of dextran (10 kD) and a more polar linker triethylene glycol compared to 1,6 hexanediol. The two conjugates did not result in any improvement in aqueous solubility and also posed the same problem. It has been hypothesized that incorporation of hydrophobic drug or drugs with limited water s olubility like valproic acid into the dextran chain might result in intra molecular and inter molecular hydrophobic interaction and ultimately result in a very compact coil structure of the conjugate molecules, leading to water insolubility ( 145 ) So, several solubilization approaches were taken to dissolve these conjugates in aqueous media (Table 3 1). The details of these te sts have been described in the Appendix A and Appendix B All the conjugates were soluble in DMSO and resulted in clear solutions. Upon addition of a very small amount of aqueous buffer to these solutions resulted in the precipitation of the conjugates. U se of (2 hydroxypropyl) cyclodextrin and PEG 300 failed to solubilize these conjugates. Only one approach was successful in solubilizing these conjugates. Dissolving the conjugate in DMSO followed by the addition of Tween 80 seemed to hold the conjugate in solution upon addition of aqueous buffer solutions. Several attempts were made to conduct hydrolysis studies with these conjugate solutions in presence of porcine liver esterase. Tween 80 itself was an ester and was also being hydrolyzed by the enzyme simultaneously, with the conjugates, which made the analysis of the active drug content extremely difficult. Several extraction attempts were made to extract active drug from this mixture after hydrolysis and proved to be

PAGE 71

71 inconsistent. So, different approa ches were taken to synthesize new conjugates to improve the aqueous solubility so that additional studies could be carried out with relative ease. Degree of Substitution The degree of substitution of the two conjugates synthesized, VPA HXD D70 and VPA HXD D10, were determined by using 1 H NMR spectra of the conjugates. The degree of substitution of the conjugate VPA TEG D10 could not be determined with reasonable certainty and hence omitted. The values of the degree of su bstitution of the conjugates were pre sented in Table 3 2. The values of the degree of substitution of the conjugates VPA HXD D70 and VPA HXD D10 were 0.19 and 0.20 respectively. Conclusion The synthesis of the three different conjugates VPA HXD D70, VPA HXD D10 and VPA TEG D10, using differe nt linkers has been achieved However, the insolubility of these conjugates in aqueous buffer solutions made it very difficult to conduct any further experiment. Several so lubilizing approaches were tried to dissolve these conjugates in aqueous buffers. Us ing DMSO and Tween 80 to solubilize these conjugates found to be successful. Although use of Tween 80 complicated the analysis of drug content even more in the hydrolysis media. This study helped us to understand the physico chemical properties of these t ypes of conjugates and directed us towards the synthesis of the new types of conjugates with better aqueous solubility. By reducing the molecular weights of dextran used in the conjugate from 70 kD to 10 kD did not improve the aqueous solubility of the con jugate. By changing from hexanediol to a more polar linker triethylene glycol, did not improve

PAGE 72

72 the solubility either. So, reducing the molecular weights of the dextrans even further to synthesize the new conjugates can be tried to improve the solubility of the conjugates.

PAGE 73

73 Tab le 3 1. Different solubilization approaches Solubilizing agents Example Organic solvents DMSO 1 Butanol DMF Ethyl acetate Surfactants Tween 80 PEG 300 Cyclodextrins (2 hydroxypropyl) cyclodextrin Aqueous media 0.01M phosphate buffer Distilled water Table 3 2 Degree of substitution (DS) of conjugates VPA HXD D70, VPA HXD D10 and VPA TEG D10 from NMR study. Conjugate code Dextran Mw Linker DS from NMR study (mole fraction) V PA HXD D70 70 kD 1,6 Hexanediol 0.19 VPA HXD D10 10 kD 1,6 Hexanediol 0.20 VPA TEG D10 10 kD Triethylene glycol Not detected

PAGE 74

74 Figure 3 1. Design of the synthesis of valproic acid linker dextran conjugate.

PAGE 75

75 Figure 3 2 1 H NMR spectra of VPA (A) HXD (B) and VPA HXD (C ) Figure 3 3 1 H NMR spectra of VPA HXD (A), Dextran (B) and VPA HXD D70 (C )

PAGE 76

76 Figure 3 4 1 H NMR spectra of VPA (A), HXD (B) and VPA HXD (C ) Figure 3 5 1 H NMR spectra of VPA HXD (A), Dextran (B) and VPA HXD D1 0 (C )

PAGE 77

77 Figure 3 6 1 H NMR spectra of VPA (A), TEG (B) and VPA TEG (C ) Figure 3 7 1 H NMR spectra of VPA TEG (A), Dextran (B) and VPA TEG D10 (C )

PAGE 78

78 CHAPTER 4 SYNTHESIS OF VALPROIC ACID LINKER DEXTRAN CONJUGATES: OVERCOMING THE SOLUBILITY ISSUE Introducti on The conjugates synthesized using hexanediol or triethylene glycol as a linker and using dextran with Mw either 70 kD or 10 kD resulted in conjugates which are practically insoluble in water or aqueous buffer. Valproic acid is a branched, short chain fat ty acid. Conjugation of valproic acid to dextran through non polar linker 1,6 hexanediol, or through triethylene glycol resulted in a hydrophobic interaction between the conjugate molecules, which in turn resulted in a compact coil stru cture of the conjuga te. This interaction might have made these conjugates practical ly insoluble in aqueous buffer. Various solubilizing approaches were taken to dissolve these conjugates in aqueous buffer. None of the approaches was successful, except for the simultaneous use of organic solvent DMSO a nd the non ionic surfactant Tween 80. But, this combination imposed additional difficulties in the analysis of the active drug during i n vitro hydrolysis studies. Also, use of DMSO and Tween 80 to dissolve the se conjugate s would c omplicate future animal studies even further D extrans with 70 kD or 10 kD were selected in earlier studies as macromolecular carrier But, insolubility made these conjugates unsuitable for any further use. Considering all of these dextran with Mw 1 kD w as chosen to be the carrier to see if the newly synthesized conjugates will have some degrees of aqueous solubility so that further studies can be carried out. About the spacer arm, relative ly more polar triethylene glycol was chosen over 1,6 hexanediol to improve the aqueous solubility of the final conjugate. Pentaethylene glycol was chosen as an alternative to triethylene

PAGE 79

79 glycol, which will lengthen the distance between valproic acid and dextran even further, and thus may aid more in the enzymati c hydroly sis of the conjugates. So, the aims of this study were to synthesize valproic acid conjugated to dextran 1 kD using either triethylene glycol or pentaethylene glycol as a linker, characterize the conjugates with NMR spectroscopy, study degree of substituti on and solubility of the conjugates, study i n vitro hydrolysis of the conjugates in presence of rat brain fraction and porcine liver esterase and study the toxicity of the conjugate on viable cells. Materials and M ethods Chemicals and Instruments Dextran T 1 (Mw 1 kD ) was purchased from Pharmacosmos (Holbaek, Denmark). dicyclohexylcarbodiimide (DCC), d imethylaminopyridine (DMAP), d imethyl sulfoxide (DMSO) anhydrous (99.9+%) were obtained from Sigma Aldrich (St Louis, MO). Valproic acid (2 p ropylpentano ic acid) was purchased from MP Biomedicals Carbonyldiimidazole (CDI) was purchased from Acros Organics (part of Thermo Fisher Scientific, New Jersey, US). Potassium phosphate monobasic, potassium phosphate dibasic, s odium phosphate dibasi c, sodium hydrogen sulfate, triethylene glycol, pentaethylene glycol, t etrahydrofuran, tert b utyl methyl ether and m olecular sieves (3) were all purchased from Sigma Aldrich (St. Louis, MO). Isopropyl alcohol (2 propanol, HPLC grade), a cetonitrile (HPL C grade), e thyl acetate (HPLC grade), d ichloromethane (HPLC grade) and m ethanol (HPLC grade) were obtained from Fisher Scientific (Pittsburgh, PA). Hydrochloric acid (2N), ortho p hosphoric acid (85%), m agnesium sulfate, sodium bicarbonate, s odium carbonate, and s odium h ydroxide were all purchased from Fisher Scientific (Pittsburgh, PA). Dimethyl sulfoxide d 6 (99.96 atom % D) was purchased from Sigma Aldrich (St. Louis, MO). All

PAGE 80

80 the buffers used in the experiments were freshly prepared. Enzyme porcine l iver e s terase (PLE) was purchased from Sigma Aldrich (St. Louis, MO). Rat brain f raction (RBF ) was prepared from r at brain according to the procedure described below and frozen at 80 C before use. Preparation of rat b rain f raction (S9 fraction) ( 146 147 ) The S9 fra ction of rat brain was use d in hydrolysis studies of the conjugates, because it contained both cytosolic and microsomal enzyme fractions ( 148 ) The fraction from the rat brain was prepared according to the following procedure. R at brain (2.2 g) was collected from a male Sprague Dawley rat weighing in between 300 350 grams according to the protocol approved by IACUC (Institutional Animal Care and Use Committee). The brain was add ed to 11 ml of 1X PBS solution pH 7.4 in a polypropylene centrifuge tube. During the transfer of the brain from a different facility (lab), the tube was kept on ice in a n ice cooler. Then, the brain was homogenized with the help of a homogenizer (Power Gen 125) for about 15 minutes keeping the tube imm ersed into ice. T he tube was centrifuged at approximately 9000 g for about 17 minutes. 500 l aliquot of the clear supernatant was collected into several Eppendorf tubes. After that the se tubes were stored in the freeze r at 80 C. HEK 293 cells were obtained from ATCC. medium was purchased from Cellgro (Man assas, VA). Dimethyl sulfoxide (DMSO) was purchased from Fisher S cientific (Pittsburgh, PA). MTT (3 (4,5 dimethylthiazol 2 yl) 2,5 diphenyltetrazolium bromide) was purchased from Calbiochem (EMD Chemicals Inc., Gibbstown, NJ).

PAGE 81

81 The HPLC instrument used was from Waters (Milford, MA) and equipped with Waters 717 Plus Autosampler, Waters 2587 1525 Binary HPLC Pump. NMR spectra were obtained with a Bruker, Avance II 600 Spectrometer with 5 mm TXI CryoProbe samplehead, operating at 600 MHz. Ultrasonic bath ( FS110H) and h omogenizer (Power Gen 125) used were fr om Fisher Scientific (Pittsburgh, PA). Centrifuge (Eppendorf Centrifuge 5415D) was from Eppendorf (Hauppauge, NY). Another centrifuge (Sorvall Legend RT) was from Thermo Scientific (part of Thermo Fisher Scientific, Pittsburgh, PA). Rotary evaporators (Rot avapor RE121 and Bchi 461 water bath) were from Bchi (Switzerland). Magnetic stirrer (Bell Stir Multi Stir 9) was from Bellco Glass (Vineland, NJ). The incubator used for the i n vitro hydrolysis study was a CO 2 water jacketed incubator (NU 2600) from Nua ire (Plymouth, MN). The microplate reader (MRX TM model) used in MTT study was from Dynex Technologies (Chantilly, VA). Synthesis of The Conjugates Two types of conjugates were synthesized using two different linkers. One was valproic acid conjugated to dex tran (Mw 1kD) via triethylene glycol linker and the other was valproic acid conjugated to dextran (Mw 1kD) via pentaethylene glycol linker. The overall synthetic procedure s for the both con jugates were presented in Figure 4 1 and Figure 4 2. These two proc edures were mostly the same (described earlier in Chapter 3) with minor difference in the extraction procedure. Total synthetic procedure is broken down into four ste ps. Step 1: synthesis of valproic acid linker conjugate, step 2: extraction of the valproi c acid linker conjugate, step 3: synthesis of valproic acid linker dextran conjugate, and step 4: extraction of the valproic acid linker dextran conjugate. These steps are described as follows.

PAGE 82

82 Stpe 1: Synthesis of valproic acid linker conjugate About 46. 43 mmol of the linker (6.196 ml for triethylene glycol and 9.814 ml for pentaethylene glycol ) was added to 20 ml dried acetonitrile in a conical flask and dissolved it with the help of magnetic stirrer (sol A). Then, 20 ml dried dicholoromethane (DCM) (dri ed over molecular sieves) was taken in a 125 ml conical flask and sol A was added slowly to DCM in this flask while shaking (sol B) Now, 1848 L (1664 mg, or, 11.6 mmol) of valproic acid, 432.2 mg (3.49mmol) of dimethyl aminopyridine (DMAP) were added and the flask was purged with argon gas (sol C) Now, 2631.4 mg (12.76 mmol) of dicyclohexyl carbodiimide (DCC) was dissolved in 10 ml of dried DCM and this solution was added to sol C drop wise for 5 minutes. The flask was p urge d with argon gas again. After that the reaction was continued at room temperature for 24 hours. Step 2: Extraction of valproic acid linker conjugate After 24 hours, the VPA linker conjugate was extracted from the reaction mixture. The reaction generated urea which was precipitated i n the reaction mixture. This urea was filtered out from the reaction mixture by using conventional paper filter. Then, the reaction mixture was evaporated in a rota ry e vap orator under high vacuum at 55 C for about 15 20 minutes and the reaction mixture wa s concentrate d to a small amount of oily liquid (about 5 ml ). A bout 10 ml of dried DCM was added to dissolve the oily liquid and transfer red it into a separatory funnel. 10 ml of 5% NaHSO 4 solution was added to the funnel and the funnel was shake n vigorous ly for 5 to 10 minutes to have unreacted DMAP dissolved in the aqueous layer. T he upper aqueous layer was removed and discarded.

PAGE 83

83 The wash with 5% NaHSO 4 solution was repeated one time to ensure complete removal of DMAP from the reaction mixture. Now, 10 m l of 5% NaHCO 3 solution was added to the funnel and funnel was also shake n vigorously for 5 to 10 minutes to remove excess acid fr om the reaction mixture. T he upper aqueous layer was also removed and discarded. The wash with 5% NaHCO 3 solution was repeated one more time for the completeness. At the end of extraction, the organic phase (with some precipitated conjugates) was collected in a conical flask and 80 ml of DCM was added to dissolve the precipitated conjugate completely. Then, 5 g of dry MgSO 4 was a dded and was allowed to stand for 30 minutes to remove water from the organic s olvent. After that, the salt was filtered out completely with a conventional paper filter and the organic liquid was transferred in a rota ry e vap orator and dried under high vacu um at 50 C to complete dryness for about 3 4 hours. Step 3: Synthesis of the valproic acid linker dextran conjugate The extracted valproic acid linker conjugate was dissolved in THF in an argon atmosphere in a conical flask. Then the appropriate amount o f carbonyldiimidazole ( CDI ) was added to it for the activation of the conjugate (the number of milli moles of CDI necessary for the activation was calculated by multiplying the number of milli moles of valproic acid linker conjugate with 1.1) This solution was then stirred fo r 27 hours at room temperature. Afterwards the solvent was evaporated in a rota ry e vap orator at 50 C under reduced pressure yielding viscous oil This product was used without further purification. For each conjugate about 1.5 g of free ze dried Dextran (MW 1k) was dissolved in 10 ml of dried DMSO and 1 g of molecular sieves 3 was added to this s olution in an argon

PAGE 84

84 atmosphere. Activated valproic acid linker c onjugate was also d issolved in 10 ml of dried DMSO and was added to dextran solu tion under argon atmosphere. At last dimethylaminopyridine ( DMAP ) was a dded to the flask as nucleophilic catalyst for the coupling of activated conjugate to dextran in an argon atmosphere. The number of millimoles of DMAP needed was calculated by multiplyi ng the number of milli moles of valproic acid linker conjugate with 1.5. The flask was then placed on a magnetic stirrer and kept on running at 50 C for 4 days. Step 4: Extraction of the valproic acid linker dextran conjugate Two different organic solvent s were used to extract the two different conjugates. Ethyl acetate was used to precipitate va lproic acid triethylene glycol dextran conjugate and t butyl methyl ether was used to precipitate valproic acid pentaethylene glycol dextran conjugate. 5 ml of the reaction medium was transferred into 4 centrifugal tubes and to each tube 25 ml of organic solvent was added to p recipitate the conjugate. T he mixtures were then centrifuged at 8000 rpm for 10 minutes. After that the supernatant was discarded and the conj ugate was dissolved in 6 ml of DMSO with the help of a homogenizer. Afterwards the conjugate was again precipitated by adding 25 ml of organic solvent and centrifu ged. These steps were repeated 3 more times. Finally, 15 ml of m ethanol was added to each tub e to disperse the conjugate and all the dispersion was transfer red into a round bottom flask and dried in a rotary evaporator for 15 hours to remove all the organic solvents. The two synthetic schemes are described in Figure 4 1 and Figure 4 2. Characteriz ation of The Conjugates by NMR Spectroscopy 1 H NMR spec tra were obtained with a Bruker Avance II 600 Spectrometer with 5 mm TXI CryoProbe sample head and Magnex 14.1 T/54 mm AS Magnet operating at

PAGE 85

85 600 MHz. The sample tube size was 5 mm The concentration s of the samples were 20 mg/0.7 ml for valproic a cid, 20 mg/0.7 ml for both triethylene glycol and pentaethylene glycol, 37 mg/0.7 ml for dextran (Mw 1k), 30 mg/0.7 ml for VPA TEG conjugate, 23 mg/0.7 ml for VPA PEG conjugate, 35 mg/0.7 ml for VPA TEG D1 co njugate and 34 mg/0.7 ml for VPA PEG D1 conjugate. DMSO d 6 (99.96%) was used as a solvent for the samples All of the chemical shifts are reported in parts per million (ppm) with tetramethylsilane as an internal standard. All the samples were measured at 3 7 C. Degree of Substitution of The Conjugates Degree of substitution (DS%) is defined as the drug content in mg per 100 mg of the dextran conjugate expressed as percentage. To determine degree of substitution, 10 15 mg of the conjugate was dissolved in 10 ml of NaOH solution (0.5 N). The solution was kept in the refrigerator for complete hydrolysis and samples were collected at different time points for the next 48 hours. For the sample collection, 500 l of the solution was collected into 1 ml HPLC vials a nd 500 l of HCl solution (0.5 N) was added to the v ials to neutralize the base All the experiments were done in triplicate. The released valproic acid was analyzed by established HPLC method Then to calculate the degree of substitution, the total amount o f drug per conjugate sample had to be calculated using the equation below. Then the percentage of degree of substitution (DS%) was calculated according to the following equation

PAGE 86

86 Aqueous Solubility of The Conjugates To determine t he aqueous solub ility of the conjugates at room temperature approximately 70 mg of VPA TEG D1 conjugate was added to 1.5 ml of double distilled water. For the conjugate VPA PEG D1, approximately 20 mg was added to 1.5 ml of double distilled water. The mix tures were then sonicated in a bath sonicator for 60 minutes to dissolve as much conjugate as possible. After that the mixtures were centrifuged for 10 minutes at 13,200 rpm to precipitate the un dissolved conjugates. Then, 1 ml of the supernatant was adde d to 10 ml of NaOH solution (0.5 N) and the solution was kept in the refrigerator for 72 hours to perform complete hydrolysis. At different time points, 500 l of each sample was collected into 1 ml HPLC vials and 500 l of HCl solution (0.5 N) was added t o each vial to neutralize the base All the experiments were done in triplicate. The collected samples were stored in the refrigerator before being analyzed by HPLC. Then to calculate the solubility of the conjugate (mg/ ml ), the total amount of drug per co njugate sample dissolved in 1 ml of water had to be calculated using the equation below. Then the ac tual solubility of the conjugate (mg/ ml ) was calculated according to the following equation In vitro Hydrolysis Study with Porcine Liver Esterase (PLE) A certain amount of conjugate (approximately 80 mg for VPA TEG D1 conjugate and 16 mg for VPA PEG D1 conjugate) was weighed in glass vials. 5 ml of preheated PBS solution w as added to each glass vial to dissol ve the conjugate in buffer solution.

PAGE 87

87 G lass vials were placed in the incubator at 37 C and keep the magnetic stirrer running at 100 rpm. Af ter that, 1 ml of porcine liver esterase (10 U/ ml ) was added to each glass vial. All the experiment s were done in triplicate. S imilar experiments were done in parallel with both conjugates using PBS only ( without the addition of porcine liver esterase ) These served as negative control s 500 l of the reaction solution from eac h glass vial was removed periodically and collected into eppendorf tubes already containing 500 l of acetonitrile to stop the enzyme catalyzed hydrolysis of the conjugate ( 149 150 ) 500 l of fresh PBS solution was added back to each vial after each sample coll ection to replenish the volume. T hese tubes were then centrifuged at 13,200 rpm for 5 minutes. After th at 800 l of supernatant was collected from each tube in 1 ml HPLC glass vials and kept in the refrigerator before being analyzed by HPLC. T ime intervals for sample collection were at 5 minutes, 10 minutes, 15 minutes, 20 minutes, 25 minutes and 30 minutes Addition of PBS solution after each sample collection diluted both the enzymes and the substrate conjugate present in the hydrolysis media. This dilution factor was considered when calculating the concentration of VPA released in the media at different tim e intervals. In vitro Hydrolysis Study with Rat Brain Fraction ( RBF ) A certain amount of conjugate (approximately 80 mg for VPA TEG D1 conjugate and 16 mg for VPA PEG D1 conjugate) was weighed in glass vials. 5 ml of preheated (37 C) PBS solution was adde d to each glass vial to dissolve the conjugate in buffer solution. Glass vials were placed in the incubator at 37 C with the magnetic stirrer running at 100 rpm. Subsequently 1 ml of rat brain fraction was added to each glass vial. All the experiments we re done in triplicate.

PAGE 88

88 S imilar experiments were done in parallel with both conjugates using PBS only ( without the addition of rat brain fraction ) These served as negative controls. 500 l of the reaction solution from each glass vial was removed periodica lly and collected into eppendorf tubes already containing 500 l of acetonitrile. 500 l of fresh PBS solution was added back to each vial after each sample coll ection to replenish the volume. These tubes were then centrifuged at 13,200 rpm for 5 minutes. After that 800 l of supernatant was collected from each tube in 1 ml HPLC glass vials and kept in the refrigerator before being analyzed by HPLC. Time intervals for sample collection were at 17 hours, 24 hours, 41 hours, 48 hours, 65 hours and 72 hours. A ddition of PBS solution after each sample collection diluted both the enzymes and the substrate conjugate present in the hydrolysis media. This dilution factor was considered when calculating the concentration of VPA released in the media at different time intervals. Cytotoxicity Test The MTT assay was performed to test the cytotoxicity of the conjugate VPA PEG D1 at three different concentrations. 0.05x10 6 HEK cells (Human Embryonic Kidney cells) were seeded to each well of a 24 well plate in DMEM for 24 h ours After 24 hours, the medium was removed and replaced with DMEM containing the conjugate VPA TEG D1 at the following concentrations 17.33 mg/ml, 3.47 mg/ml and 0.69 mg/ml The medium DMEM containing no conjugate was also added as a control. After incu bation for 24 hours, the medium was replaced with DMEM medium and 20 l of filtered MTT solution (5 mg/ml in 1X PBS ) was added to each well and incubated at 37C for 4 hours. After that medium was removed and 200 l of DMSO was added to each well and in cub ated at 37 C for 5 minutes Then 200 l solution from each well fr om each plate was transferred to a 96 well plate and absorbance was measured by a plate reader at a

PAGE 89

89 wavelength of 550 nm. The percent cell viability was calculated according to the followi ng equation The experiment was repeated the same way adding two more controls in addition to the medium control. Th e concentrations of the conjugate were reduced for this experiment. The two new controls we re valproic acid and dextran. The following concentrations were used for the conjugate VPA TEG D1 4 mg/ml, 1mg/ml and 0.25 mg/ml. The concentrations of valproic aci d were used as follows 0.27 mg/ml, 0.068 mg/ml and 0.017 mg/ml. The following concentrations were used for dextran 3.301 mg/ml, 0.825 mg/ml and 0.206 mg/ml. The medium DMEM containing no conjugate was also added as a control. Statistical Analysis Statist ical significance of the difference in the percent cell viability between each concentration of the conjugate VPA TEG D1 valproic acid and dextran was tested using one value <0.05 was considered si gnificant. Results and Discussion Synthesis of Dextran linker Valproic A cid Conjugates and Spectral Analysis The dextran linker valpr oic acid conjugates were synthesized according to the procedures mentioned in several papers for other compounds with sligh t modifications ( 151 ) Two types of conjugates were synthesized by using two different linkers. Conjugate VPA TEG D1 was synthesized using triethylene glycol as a linker and VPA

PAGE 90

90 PEG D1 was synthesized using pentaethylene glycol as a linker. Dextrans with an average molecular weight of 1 kD were used in the synthesis of b oth conjugates. Initially, valproic acid linker conjugate was synthesized usi ng dicyclohexylcarbodiimide as a coupling agent in the pr esence of dimethylaminopyridine D icyclohexylcarbodiimide gives quantitative esterification even in the presence of moistu re. Then, this conjugate was extracted from the reaction media using the extraction procedure described in Chapter 3 The yield% of extraction for VPA TEG was 49.31% whereas the same yield% for VPA PEG was 93.71%. The reason behind the low yield% of VPA TE G could be explained in terms of solubility. Triethylene glycol is 100% soluble in water and the resulting conjugate VPA TEG was likely to be more soluble in aqueous media than the conjugate VPA PEG. Because of that, significant amount of VPA TEG was lost during the multiple washing steps during the extraction, which resulted in a lower extraction yield% of VPA TEG. The valproic acid linker conjugates were then acti vated using carbonyldiimidazole in t he solvent tetrahydrofuran and then subsequently coupled to dextran in DMSO in the presence of dimethylamin opyridine Finally, the conjugate VPA TEG D1 was precipitated using ethyl acetate. The VPA PEG D1 conjugate was not precipitated upon addition of ethyl acetate, methanol, acetonitrile or dichloromethane. Th e conjugate was precipitated at last upon addition of tert butyl methyl ether The conjugation yield was 92.99% for VPA TEG D1 and 49.20% for VPA PEG D1. Analysis of 1 H NMR spectra of VPA TEG The Figure 4 3 showed 1 H NMR spectra of valproic acid (A), trie thylene glycol (B) and valproic acid triethylene glycol (VPA TEG) (C). The 1 H NMR spectrum showed t he bands for the prot ons of valproic

PAGE 91

91 acid appeared A carboxylic acid proton band was present This band was prominently present in the spectrum of valproic acid. Three peaks appeared belong ing to the methylene protons of triethylene glycol. Th The conjugate VPA TEG had all the methyl, methylene and methine protons of valproic acid appearing bet peaks for methylene protons of triethylene glycol appe aring 4.2 and hydroxyl proton at 4.5 DMSO (e.g. DMSO d5) was due to the presence of water in the conjugate. The peak for the carboxyl proton from valpro ic acid was not present in the spectrum of VPA TEG, which indicated the formation of an ester bond between valproi c aci d and triethylene glycol. The fact that there were bands for both valproic acid (except for the carboxyl proton) and triethylene glycol present in the spectra of VPA TEG and the absence of the carboxyl proton from v provided strong indication for the formation of the conjugate be tween valproic acid and triethylene glycol. Analysis of 1 H NMR spectra of VPA TEG D1 conjugate The Figure 4 4 showed 1 H NMR spectra of valproic acid triethylen e glycol (VPA TEG) (A), dextran (B) and valproic acid triethylene glycol dextran conjugate (VPA TEG D1) (C). The 1 H NMR spectrum of dextran showed bands belonged to the protons of anhydroglucose unit ( AGU) and protons of the hydroxyl groups of the dextran molecules except for the band at = 3.3. This band at

PAGE 92

92 = 3.3 belonged to the protons of the water molecules that were present in the dextran samples. T he bands for the protons of VPA TEG conjugate app eared 2.3 (protons from valproic acid) and (protons from triethylene glycol) and at 4.5 (hydroxyl proton from triethylene glycol) was present in all the spectra and represented part of the DM SO used as solvent which was not deuterated completely (e.g. DMSO d 5 ). In the spectrum of VPA TEG D1, proton s from valproic acid appeared between 0.8 and 2.3 Peaks present anhydroglucose unit ( AGU ) of dext ran molecules (except for the anomeric proton), methylene protons at C 6 of AGU and all the methylene protons of triethylene glycol. These peaks were overlapping with each other. The presence or absence of hydroxyl ould not be identified, because the peaks from three hydroxyl protons and one anomeric proton of AGU of dextran molecules appeared Th e fact that there were bands for both valproic acid and dextran present in the spectra and the fact that valproic acid could not be conjugated to the dextran in this case without triethylene glycol conjugated to the dextran, provided indication for the formation of the conjugate VPA TEG D1. Analysis of 1 H NMR spectra of VPA P EG The Figure 4 5 showed 1 H NMR spectra of valproic acid (A), pentaethylene glycol (B) and valproic acid pentaethylene glycol (VPA PEG) (C). The 1 H NMR spectrum showed the bands for the protons of valproic acid appeared

PAGE 93

93 to the methylene protons of pent aethylene glycol. The hydr 4.5. The conjugate VPA P EG had all the methyl, methylene and methine protons (protons of valproic acid) methylene protons of pentaethylene glycol appe and hydroxyl proton at 4.5 partially deuterated DMSO (e.g. DMSO water in the conjugate. The peak for carboxyl proton from valpro ic acid was not present in the spectrum of VPA P EG, which indicated the formation of ester bond betwe en valproic acid and pentaethylene glycol. The fact that there were bands for both valproic acid and pentaethylene glycol present in the spectra of VPA P EG and the absence of the carboxyl proton from v provided strong indication for the formation of the conjugate between valproic acid and penta ethylene glycol. Analysis of 1 H NMR spectra of VPA P EG D1 conjugate The Figure 4 6 showed 1 H NMR spectra of valproic acid penta ethylene glycol (VPA P EG) (A), dextran (B) and valproic acid penta ethylene glycol dextran conjugate (VP A P EG D1) (C). The 1 H NMR spectrum of dextran showed pm, these broad bands belonged to the protons of anhydroglucose unit (AGU) and protons of the hydroxyl groups of the dextran molecules except for the band at = 3.3. This band at

PAGE 94

94 = 3.3 belonged to the protons of the water molecules that were present in t he dextran samples. T he bands for the protons of VPA P EG conjugate appeared in 2.3 (protons from valproic aci (protons from penta ethylene glycol) and at 4.5 (hydroxyl proton from penta ethylene glycol). The was present in all the spectra and represented part of the DMSO used as solvent which was not deuterated completely (e.g. DMSO d 5 ). In the spectrum of VPA P EG D1, protons from valproic acid appeared in between Peaks appe AGU of dextran molecules (except for the anomeric proton), methylene protons at C 6 of AGU and all the methylene protons of penta ethylene glycol. These peaks were overlapping with each other. The presen ce or abs ence of hydroxyl proton from penta hydroxyl protons and one anomeric proton of AGU of dextran molecules appeared in and 5. The fact that there were bands f or both valproic acid and dextran present in the spectra and the fact that valproic acid could not be conjugated to the dextran in this case without penta ethylene glycol conjugated to the dextran, provided indication for the formation of the conjugate VPA P EG D1. Degree of Substitution and Solubility of The Conjugates The synthesized conjugate VPA TEG D1 had a degre e of substitution (DS%) of 1.15 0.02 and VPA PEG D1 had a degree of substitution of 6.91 0.14 (Table 4 1). The degree of substitution of the conjugate VPA TEG D1 was quite low compared to that of the conjugate V PA PEG D1 (Figure 4 7 ). The degree of substitution of the

PAGE 95

95 conjugate VPA PEG D1 was quite comparable to the values of the degree of substitution of other dextran conjugates found in seve ral other studies ( 77 95 102 ) Solubility is one of the most essential physicochemical properties that should be considered in selecting a potential drug candidate and should be examined as early as possible in the process of drug discovery. If the drug is not soluble under physiological conditions, develo pment of the formulation of the drug and conduct any future study would be very difficult. The solubility issue will also cause severe problems concerning absorption, per meability and bioavailability. Earlier attempts for the synthesis of valproic acid lin ker dextran conjugates faced with this solubility issue. By reducing the m olecular weights of dextran and using more polar linker molecules resulted in conj ugates VPA TEG D1 and VPA PEG D1 which are soluble in physiological pH and temperature. Aqueous solu bility of the conjugates VPA TEG D1 and VPA PEG D1 at room temperature was 36.80 0.55 mg/ml and 4.96 0.05 mg/ml respectively (Table 4 2 ). Like valproic acid dextran conjugates synthesized before, t he solubility was also found to be in inverse proportio n with degree of substitution % of the conjugat es. The conjugate with higher degree of substitution % had less aqueous solubility so was the case with conjugates VPA TEG D1 and VPA PEG D1. The low degree of substitution % of VPA TEG D1 conjugate may cert ainly limit the possibility of using this conjugate as a potent ial candidate for a conjugate formulation. But, th e conjugate VPA PEG D1 with a degree of substitution % of 6.9 06 could be a potential candidate for such a formulation. In animal models, drug l evels of valproic acid

PAGE 96

96 needed to achieve antiepileptic activity was in micrograms level. The valproic acid level needed for antiepileptogenic activity is yet to be det ermined. However, s everal studies used the antiepileptic dosage of valproic acid to study antiepileptogenic property of valproic acid ( 65 66 ) In fact, i t could be the same as, lower or higher than that needed to provide antiepileptic activity. Moreover, t here are no clearance studies available (in animals or human) for these conjugates either in the systemic circulation or in the brain. So, the degree of substitution % of the conjugate VPA PEG D1 m ay or may not be enough depending on the clearance of the conjugate in the system of interest. But, the degree of substitution % of the conjugate can certainly be improved by optimizing the synthetic procedure i f needed So, the conjugate VPA PEG D1 certa inly holds the promise to be a candidate for a conjugate formulation. In vitro Hydrolysis Studies The hydrolysis of the conjugates VPA TEG D1 and VPA PEG D1 was investigated in aqueous phosphate buffer solution (pH 7.5) at 37 C in presence of rat brain fr action and porcine liver esterase. Porcine liver esterase was cho sen because it was highly p urified, characterized and relatively stable ( 152 154 ) It also c ontained a group of isozymes with a broad range of substrate specificity ( 155 ) Rat brain fraction was chosen to evaluate the hydrolytic activity of rat brain esterases against our synthesi zed conjugates. Because, ultimately brain would be the target site of action for the synthesized conjugates. B oth conjugates possessed two types of ester groups w hich were susceptible to hydrolytic cleavage. The two ester groups were carboxyl ester and car bonate ester type. Three different schemes were proposed for the hydrolysis of VPA TEG D1 and VPA PEG D1 conjugate s, which were described in Figure 4 9 (scheme A) Figure 4 10

PAGE 97

97 (scheme B) and Figure 4 11 (scheme C). T he rate constants k 1 k 2 k 3 and k 4 wer e assumed to be pseudo first order rate constants for the depicted reactions, where both k 1 and k 3 appeared to be enzyme catalyzed, k 2 appeared to be pH dependent and k 4 assumed to be either pH dependent or enzyme catalyzed Letting k obs represent the over all first order rate constant for the hydrolysis of the conjugates: (1) The time dependency of the three compounds in the reaction mixtur e ( A, B and C in the Figure 4 9 ) could be represented by the following differential equations ( 156 ) : Also, the time dependency of the three compounds A, B and C in the Figure 4 10 could be represented by the following differential equations: (7) Also, the time dependency of the three compound s A, B and C in the Figure 4 11 could be represented by the following differential equations: (8)

PAGE 98

98 All these hydrol ysis models were used to fit the data obtained from the hydrolysis of VPA TEG D1 and VPA PEG D1 conjugates in different hydrolysis media. Hydrolysis of VPA linker Dextran conjugate in presence of Porcine Liver Esterase (PLE) The hydrolysis of the conjugat es VPA TEG D1 and VPA PEG D1 was conducted in aqueous phosphate buffer solutions in presence of porcine liver esterase (PLE) and compared against the hydrolysis of the both conjugates in phosphate buffer solutions as a control. The hydrolysis was carried o ut at 37 C to make the study relevant to physiological condition. T he hydrolysis was carried out for only 30 minutes because preliminary studies with PLE showed very rapid hydrolysis of the conjugates. S amples were collected at 5 minutes, 10 minutes, 15 minutes, 20 minutes, 25 minutes and 30 minutes interval s General hydrolytic scheme of VPA TEG D1 conjugate and VPA PEG D1 conjugate is described in Figure 4 9. Both conjugates appeared to be hydrolyzed by PLE at a very fast rate In the control samples, the presence of low molecular weight conjugates VPA TEG and VPA PEG was detected at a very small amount, which was actually physically mixed with the macromolecular conjugates. Presence of VPA was not detected at all in the control samples. This indicate d that the hydrolysis pathway from A to C to B was not significant for both conjugates during the time course of hydrolysis. The both conjugates were mostly hydrolyzed according to the pathway from A to B by the enzymatic action of PLE The data was fitted w ith linear regression accordingly for the both conjug ates with a very high value of R 2 0.9750 for the conjugate VPA TEG D1 (Figure 4 13) and 0.9724 for the conjugate VPA PEG D1 (Figure 4 14) The estimated k obs values and

PAGE 99

99 corresponding half lives (t 1/2 ) fo r the hydrolysis of both conjugates were presented in the Table 4 3 and Table 4 5 respectively The k obs and t 1/2 values of the conjugate VPA TEG D1 were 0.0381 min 1 and 18 minutes respectively, whereas k obs and t 1/2 values of the conjugate VPA PEG D1 wer e 0.0334 min 1 and 21 minutes respectively. The both conjugates seemed to be very good substrates for the enzymes present in porcine liver V alproic acid conjugated to dextran (10 kD) without any spacer did not get hydrolyzed in aqueous buffer solution in the presence of PLE (data not shown) So, the linkers used in the conjugates VPA TEG D1 and VPA PEG D1 might have provided the space that made ester bond between valproic acid and the linker accessible by the enzyme and made these conjugates good substrat es for the enzymes present in PLE. In addition to that low molecular weight dextran (1 kD) used in the conjugates might have provided less steric hindrance to the enzyme activity. Studies also reported that PLE appeared to have a very wide range of substra te specificity, because it contained several isozymes of differing substrate specificity ( 154 ) Hydrolysis of VPA linker D extran conjugate in the presence of Rat Brain Fraction (RBF ) The hy drolysis of the conjugates VPA TEG D1 and VPA PEG D1 was investigated in aqueous phosphate buffer solutions in presence of the S9 fraction of rat brain and compared against the hydrolysis of the both conjugates in phosphate buffer solutions as a control. T he hydrolysis was carried out at physiological temperature of 37 C. Initially the hydrolysis was carried out for only 30 minutes assuming the hydrolysis would be very fast. No detectable extent of hydrolysis was observed. Then, duration of the study was l engthened to 3 days. Samples were collected at 17 hours, 24 hours, 41 hours, 48 hours, 65 hours and 72 hours intervals

PAGE 100

100 The presence of VPA and VPA TEG or VPA PEG was detected in the hydrolysis media of both conjugates in presence of RBF whereas only eith er VPA TEG or VPA PEG was detected in the media of controls. Presence o f VPA was not detected at all in the media of controls. This indicated that VPA was released in the media by the action of enzymes present in RBF following one of the two possible (from A to B, or C to B) or, bo th hydrolytic pathways (Fig ure 4 9 Figure 4 10 and Figure 4 11 ). The presence of VPA TEG or VPA PEG in the control samples indicated that the hydrolysis of VPA TEG D1 or VPA PEG D1 conjugate to the corresponding low molecular we ight conjugate (from A to C) was mostly chemical hydrolysis. Both conjugates seemed to be hydrolyzed by the esterase enzymes present in rat brain fraction although at a very slow rate To quantify the amount of VPA PEG released in the hydrolysis media, st andard curve was constructed for this low molecular weight prodrug To quantify the amount of VPA TEG released in the hydrolysis media, the standard curve for VPA PEG was used, because of insufficient quantity of VPA TEG available to construct such a stand ard curve. The assumption was made that both compounds would have the same molar absorptivity. Three different hydrolysis models were fitted to the data obtained from the hydrolysis of both conjugates VPA TEG D1 and VPA PEG D1 in phosphate buffered saline solution in the presence of rat brain fraction at 37 C Fitting of the models was accomplished using Scientist 3.0 software. Model selection criteria and AIC (Akaike Information Criterion) were used to compare against the models. Based on these criteria hydrolysis scheme described in Figure 4 10 seemed to be the most likely model to describe the hydrolysis of both conjugates in phosphate buffered solution in the

PAGE 101

101 presence of rat brain fraction. Different rate constants were estimated by fitting the model to the hydrolysis data and presented in Table 4 4 After analyzing different rate constants, i t appeared that hydrolysis pathway for both conjugat es from C to B was mostly insignificant which simplified the model described in Figure 4 10, leading to the model described in Figure 4 12 Both conjugates appeared to be hydrolyzed f ollowing the pathway from A to B and A to C C onversio n from A to C appeared to follow chemical hydrolysis whereas conversion from A to B appeared to be catalyzed by enzyme action. We also assumed non specific degradation of compound C (VPA TEG or VPA PEG) via chemical or enzymatic pathway. Degradation via this pathway might have resulted in random chain scission of triethylene glycol or pentaethylene glycol chain ( 157 ) and generated different molecular weight fragments from VPA TEG or VPA PEG, which might not have detected by our analytical technique The estimated k obs values and corresponding half lives (t 1/2 ) for the hydrolysis of both conju ga tes were presented in the Table 4 6 The k obs and t 1/2 values of the conj ugate VPA TEG D1 were 0.0247 hr 1 and 28 hr respectively, whereas k obs and t 1/2 values of the conjugate VPA PEG D1 were 0.0157 hr 1 and 44 hr respectively. B oth conjugates did not s eem to be very good substrates for the enzymes present in rat brain fraction It also could be d ue to the low activity of esterase enzymes in the rat brain fraction ( 136 146 ) The fraction prepared from the rat brain wa s crude and was not purified, characterized and quantified for the enzymes present Although the activity of the rat brain fraction was tested u sing para nitrophenyl acetate as a substrate. This substrate was hydrolyzed in rat brain fraction although at a much slower rate This

PAGE 102

102 indicated the presence of esterase enzyme in rat brain fraction at a small quantity. Also, enzymes present in rat brain fraction could be highly substrate specific compared to the enzymes present in the liver. Cytotoxicity T est One treatment group of three different concentrations of the conjugate VPA PEG D1 was tested against control (DMEM only ). The concentrations were a s follows T1 = 17.33mg/ ml T2 = 3.47 mg/ ml and T3 = 0.69mg/ ml. Figure 4 21 showed the effect of the conjugate on cell viability. Cytotoxic effect of the conjugate at three different concentrations were measured against control (=100% cell viability) Only sample T1 showed very low cell viability (11.51 0.63 %, P< 0.001) and seemed to cytotoxic. Sample T2 did not show any significant difference against control (P> 0.05), although sample T3 showed difference that was significant against control (P< 0.05). On the contrary, sample T3 did not show any significant difference against the sample T2 (P> 0.05). Moreover, sample T1 showed significant difference against both samples T2 and T3 (P< 0.001). It could be concluded that both of the samples T2 and T3 were l east likely to be cytotoxic. Sample T1 appeared to be cytotoxic or antiproliferative to the cell line. According to the lit erature dextran by itself did not appear to be cytotoxic ( 158 ) .T he apparent toxicity of the highest concentration of the conjugate could be due to the HDAC (histone deacetyl ase) inhibitory effect of valproic acid, released from the conjugate at higher concentration. Studies showed that valproic acid inhibited HDACs efficiently in multiple settings at therapeutically relevant levels ( 159 ) Valproic acid has also been shown to induce differentiation and cell death of neuroblastoma cells and this effect has been attributed to its possible HDACs inhibitory activity ( 160 ) Toxicity could

PAGE 103

103 also result from the activity of the low molecular weight conjugate of valproic acid with the linker which might have released into the media from the macrom olecular conjugate. In the repeat experiment, three different concentrations of the conjugate VPA PEG D1 were tested against control (DM EM only). These concentrations we re as follows A 1 = 4 mg/ ml A 2 = 1 mg/ ml and A 3 = 0.25 mg/ ml Cytotoxicity of valproic acid and dextran was also tested at three different concentrations. Figure 4 22 showed the effect of the conjugate on cell viability. Cytotoxic effect of the conjugate at three different concentrations were measured against control (=100% cell viability) Sample A 1 showed very low cell viability (29.19 1.31 %, P< 0.00 0 1) and seemed to be cytotoxic. Sample A 2 did not show any significant difference against control (P> 0.05) Sample A 3 showed significant difference against control (P< 0.0 0 5) and it appeare d to have positive effect on the cell proliferation The sample A1 showed significant difference against A2 (P< 0.001) and against A3 (P< 0.0001). Moreover, sample A2 showed significant difference against the sample A3 (P< 0.0001). It could be concluded th at sample A 1 appeared to be cytotoxic or antiproliferative to the cell line. Sample A2 did not appear to be cytotoxic to a significant extent in comparison to the control. On the contrary, sample A3 appeared to have positive effect on cell proliferation, o r it might have increased the activity of mitochondrial oxido reductase enzyme systems which might have resulted in increased conversion of MTT into formazan. It appeared from the study that conjugate VPA PEG D1 seemed to have concentration dependent effect on cell viability, or on the activity of mitochondrial oxidoreductase en zyme systems.

PAGE 104

104 Figure 4 23 showed the effect of valproic acid on cell viability. Three different concentrations of valproic acid were tested, which were as follows B 1 = 0.27 mg/ ml, B 2 = 0.068 mg/ ml and B 3 = 0.0.017 mg/ ml. These concentrations of valproic acid did not seem to have any effect on cell viability in comparison to the control. Figure 4 24 showed the effect of dextran on cell viability at three different concentrations. These concentrations were as follows C 1 = 3.301 mg/ ml, C 2 = 0.825 mg/ ml and C 3 = 0.206 mg/ ml. The samples C1 and C2 did not seem to have any significant effect on cell viability in comparison to the control. On the contrary, t he sample C3 seemed to have positi ve effect on cell proliferati on in comparison to the control (P< 0.05). The effects of the samples C1 and C2 appeared to be significantly different compared to the effect of the sample C3 (C1 vs. C3, P< 0.005; C2 vs. C3, P< 0.05). It also appeared from the study that dextran seemed to have concentration dependent effect ( 161 ) on cell viability, or on the activity of mitochondrial oxidoreductase enzyme systems. Based on the two studies, it can be concluded that t he effect of the conjugate VPA PEG D1 seemed to have a concentration dependent effect on cell viability. Although, i t did not appear to be cytotoxic to the dividing cells at 3.47 mg/ml concentration in one study but it appeared to be cytotoxic at a concentration of 4 mg/ml in another study. The conjug ate appeared to exhibit a little toxicity at a concentration of 0.69 mg/ml in one study, whereas exhibited positive effect on cell viability at 0.25 mg/ml in another study. So, t he toxicity study with the conjugate needs to be repeated to resolve these app arent contradictory results between the two studies This would also eliminate the possibility of any inadvertent experimental error that might h ave happened during these two experiments

PAGE 105

105 Conclusion The aqueous solubility of the valproic acid linker dextran conjugates w as shown to be improved by using lower molecular weight (1 kD) dextran and also by using polar linkers such as triethyl ene glycol and pentaethylene glycol. The two conjugates VPA TEG D1 and VPA PEG D1 appeared to be very good substrates for po rcine liver esterase s but apparently poor substrates for the enzymes in rat brain fraction Although used rat brain fraction was not purified to concentrate the enzymes and characterized for the types of enzymes present. However, the study offered a proof of principle that val proic acid linker dextran conjugates could be designed and synthesized in such a way which would provide enzyme mediated cleavage of the conjugate to release the drug Further investigation and optimization are however necessary. Th e conjugate VPA PEG D1 demonstrated a concentration dependent toxic effect on the viability of human embryonic kidney cells The results from the two studies appeare d contradictory to some extent. The dosage of the conjugate VPA PEG D1 for a prospective a nim al study is yet to be determined Clearance study of the conjugate needs to be conducted to determine such a dosage. Hence, t he cytotoxicity study would be worth repeating to establish a toxicity level of the conjugate before conducting an animal experimen t

PAGE 106

106 Table 4 1. D egree of substitution (DS%) of conjugates VPA TEG D1 and VPA PEG D1 from hyd rolysis study (mean sem ), n=3). Conjugate code Dextran Mw Linker DS % from hydrolysis study VPA TEG D1 1 kD Triethylene glycol 1.15 0.02 VPA PEG D1 1 kD Pent aethylene glycol 6.91 0.14 Table 4 2. Aqueous solubility of the conjugates VPA TEG D1 and VPA PEG D1 at room temperature (mean sem n=3). Conjugate code Dextran M w Linker Solubility (mg/ ml ) VPA TEG D1 1 kD Triethylene glycol 36.80 0.5 5 VPA PEG D 1 1 kD Pentaethylene glycol 4.96 0.0 5 Table 4 3. Rate constants for hydrolysis of VPA TEG D1 and VPA PEG D1 in aqueous phosphate buffer solution in presence of porcine liver esterase (PLE) at 37 C (mean sem n=3). Conjugate code k 1 (min 1 ) k 2 (min 1 ) k 3 (min 1 ) k obs (= k 1 + k 2 ) (min 1 ) VPA TEG D1 0.0 38 1 0.0015 0.0381 0 .0015 VPA PEG D1 0.0 334 0.0014 0.0334 0 .0014 Table 4 4 Rate constants for hydrolysis of VPA TEG D1 and VPA PEG D1 in aqueous phosphate buffer solution in prese nce of rat brain fraction (RBF ) at 37 C ( mean sem, n=3). Conjugate code k 1 (hr 1 ) k 2 (hr 1 ) k 3 (hr 1 ) k 4 (hr 1 ) VPA TEG D1 0.0025 0.0005 0.0222 0.0026 ~ 0 0.0301 0 .00 33 VPA PEG D1 0.0029 0.0009 0.0128 0.0010 ~ 0 0.0149 0 .0025 Table 4 5. Half lives (t 1/2 ) for the hydrolysis of VPA TEG D1 and VPA PEG D1 in aqueous phosphate buffer solution in presence of porcine liver esterase (PLE) at 37 C (mean sem n=3). Conjugate code k obs (min 1 ) t 1 /2 (min) VPA TEG D1 0. 0381 0.0015 18 VPA PEG D1 0. 0334 0.0014 2 1 Table 4 6 Half lives (t 1/2 ) for the hydrolysis of VPA TEG D1 and VPA PEG D1 in aqueous phosphate buffer solution in presence of rat brain fraction (RBF ) at 37 C (mean sem n=3). Conjugate code k obs (hr 1 ) t 1 /2 (hr ) VPA TEG D1 0. 0247 28 VPA PEG D1 0.0 157 44

PAGE 107

107 Figure 4 1. Synthesis of valproic acid triethylene glycol dextran conjugate

PAGE 108

108 Figure 4 2. Synthesis of valproic acid pentaethylene glycol dextran conjugate

PAGE 109

109 Figure 4 3 1 H NMR spectra of VPA (A), TEG (B), and VPA TEG (C ) Figure 4 4 1 H NMR spectra of VPA TEG (A), Dextran (B) and VPA TEG D1 (C )

PAGE 110

110 Figure 4 5 1 H NMR spectra of VPA (A), PEG (B) and VPA PE G (C ) Figure 4 6 1 H NMR spectra of VPA PEG (A ), Dextran (B) and VPA PEG D1 (C )

PAGE 111

111 Figure 4 7 Degree of substitution (DS%) of the conjugates VPA TEG D 1 and VPA PEG D 1 (mean sem n=3).

PAGE 112

112 Fi gure 4 8 Aqueous solubility of the conjugates VPA TEG D1 and VPA PEG D1 at room temperature (mean sem n=3).

PAGE 113

113 Figure 4 9. Hydrolysis scheme of valproic acid linker dextran conjugate (scheme A) Figure 4 10. Hydrolysis scheme of valproic acid linker dextran conjugate (scheme B)

PAGE 114

114 Figure 4 11. Hydrolysis scheme of valproic acid linker dextran conjugate (scheme C) Figure 4 12. Hydr olysis scheme of valproic acid linker dextran conjugate.

PAGE 115

115 Figure 4 13 Hydrolysis of the conjugate VPA TEG D1 in PBS in the presence of porcine liver esterase (PLE) at 37 C (mean sem, n=3). Figure 4 14 Hydrolysis of the conjugate VPA PEG D1 in PBS in the presence of porcine liver esterase (PLE) at 37 C (mean sem, n=3).

PAGE 116

116 Figure 4 15 Model fit for the h ydrolysis of VPA TEG D1 conjugate according to hydrolysis scheme A. Figure 4 16 Model fit for the hydrolysis of VPA PEG D1 conjugate a ccording to hydrolysis scheme A.

PAGE 117

117 Figure 4 1 7 Model fit for the hydrolysis of VPA TEG D1 conjugate according to hydrolysis scheme B. Figure 4 1 8 Model fit for the hydrolysis of VPA PEG D1 conjugate according to hydrolysis scheme B

PAGE 118

118 Figure 4 1 9 Model fit for the hydrolysis of VPA TEG D1 conjugate according to hydrolysis scheme C. Figure 4 20 Model fit for the hydrolysis of VPA PEG D1 conjugate according to hydrolysis scheme C.

PAGE 119

119 Figure 4 21 Effect of VPA PEG D1 conjugate on cell viabilit y (T1 = 17.33 mg/ml, T2 = 3.47 mg/ml, T3 = 0.69 mg/ml, mean sem n = 4, P<0.05, ** P<0.001).

PAGE 120

120 Figure 4 22 Effect of VPA PEG D1 conjugate on cell viability (A1 = 4 mg/ml, A2 = 1 mg/ml, A3 = 0.25 mg/ml, mean sem n = 6 P<0.0 05 ** P<0.00 1 and ** P< 0.00 0 1 ).

PAGE 121

121 Figure 4 23 Effect of VPA on cell viability (B1 = 0.27 mg/ml, B2 = 0.068 mg/ml, B3 = 0.017 mg/ml, mean sem, n = 6).

PAGE 122

122 Figure 4 24 Effect of dextran on cell viability (C1 = 3.301 mg/ml, C2 = 0.825 mg/ml, C3 = 0.206 mg/ml, mean sem, n = 6, P<0.05, and ** P<0.005).

PAGE 123

123 CHAPTER 5 SUMMARY OF RESULTS A ND FUTURE WORK The central aim s of this study were to synthesize valproic acid dextran conjugate which would be able to deliver drug slowly over a prolonged period of time and also to synt hesize valproic acid dextran conjugate using a spacer to provide enzyme mediated cleavage of the conjugate to deliver the drug at or near the site of action. Subsequent to these aims, the effect s o f pH, temperature degree of substitution and solubility of the conjugates on the rate s of hydrolysis of the conjugates were also studied The valproic acid dextran conjugates were synthesized using two different molecular weights of dextran (10 kD and 70 kD). The conjugates were characterized by FT IR and NMR spe ctroscopy. The synthesized conjugates were shown to be hydrolyzed i n vitro over approximately 2 months period at pH 7.5 and 37 C Temperatu re and pH were shown to affect the rate s of hydrolysis of the conjugates i n vitro However, pH seemed to have a more pronounced effect than temperature on the rate s of hydrolysis of the conjugates. Degree of substitution and solubility of the conjugates had a more pronounced effect on the rate s of hydrolysis at higher pH of 10.22 at 37 C and 47 C than at pH 7.5. The c onjugates of valproic acid with dextran connected through a spacer were also synthesized. Several initial conjugates resulted in insoluble conjugates. By using lower molecular weights of dextran and relatively more polar linker in the conjugates, the probl em had been overcome. The two conjugates VPA TEG D1 and VPA PEG D1 were synthesized using dextran (molecular weight 1 kD) and using triethyl ene glycol and pentae thylene glycol as a linker, which resulted in water soluble conjugates.

PAGE 124

124 The hydrolysis of the two conjugates was conducted in aqueous phosphate buffer solution in presence of rat brain fraction and porcine liver esterase at 37 C. The two conj ugates appeared to be very good substrates for porcine liver esterase s but apparently poor substrates for the enzymes in rat brain fraction Although used rat brain fraction was not purified and characterized for the types of enzymes present. The synthesized valproic acid dextran conjugates could be potential candidates for a sustained release formulation of v alproic acid. Also, hydrolysis studies with valproic acid linker dextran conjugates provided a proof of concept that such conjugates could be designed and synthesized to provide enzyme mediated cleavage of the conjugate to deliver the drug at or near the s ite of action. In future, two areas of research with these types of conjugates would be worth pursuing. First, to show that valproic acid dextran conjugates is effective in preventing the development of epilepsy, or at least, effective in treating epilepti c seizures for an extended period of time. Second, once the efficacy has been demonstrated, new formulation of these conjugates can be developed for better ease of administration. Treatment Objective Both types of conjugates can be tested in animal models of epile psy by injecting via convection enhanced delivery to see if these conjugates are able to deliver the drug for an extended period of time and be able to p rotect the neurons from death after an initial precipitating injury and ultimately prevent the development of spontaneous recurrent motor seizures. Formulation Objective Nanoparticles can be made using dextran conjugates with or without linkers by o/w emulsion or solvent diffusion methods ( 162 163 ) or nanostructured lipid carrier with valproic acid ( 164 ) The size and surface properties of

PAGE 125

125 t hese nanopa rticles can then be optimized for longer circulation time in blood and increased permeability through the blood brain barrier Ultimately, the goal would be to test whether these nanoparticles made from valproic acid dextran conjugates would be able to pre vent neuronal death and ultimately prevent the development of spontaneous recurrent motor seizures by delivering the drug for a prolonged period of time. Even in the absence of a neuroprotective effect, developing nanoparticles made from these conjugates w hich will be able to treat acute epileptic seizures for an extended period of time would be a significant addition to the current epileptic treatment modalities.

PAGE 126

126 APPENDIX A SOLUBILIZING TESTS FOR VPA HXD D70 AND VPA HXD D10 Conjugate VPA HXD D70 Amoun t of conjugate (mg) Volume of DMSO (ml) Volume of buffer (p H 7.4, 0.1 M) (ml) Method and observations Test 1 22.2 25 25 Conjugate was first dissolved in 5 ml DMSO and the n 25 ml of phosphate buffer was added. Since it precipitated, 20 more ml of DMSO was added not soluble Test 2 11.8 10 5 5 ml of buffer was added to the conjugate already dissolved in 10 ml of DMSO precipitation Conjugate VPA HXD D10 Amount of conjugate (mg) Cyclodextrins (mg) Surfactants Volume of buffer (p H 7.4, 0.1 M) or dH 2 O M ethod and observations Test 1 10.5 (2 hydroxypropyl) cyclodextrin: 202.4 mg dH 2 O: 10 ml (2 hydroxypropyl) cyclodextrin and conjugate were added to dH 2 O not soluble Test 2 10.1 Tween 80: dH 2 O: 10 ml Tween was first mixed with dH 2 O and then the conjugate was added not solu ble

PAGE 127

127 APPENDIX B SOLUBILIZING TESTS FOR VPA TEG D10 Conjugate VPA TEG D10 Amount of conjugate (mg) Volume of DMSO (ml) Surfactants Volume of buffer (p H 8, 0.1 M) (ml) Method and observations Test 1 28.8 5 45 The conjugate was dissolved in DMSO and then buffer was added to this solution precipitation Test 2 12.3 2.5 Tween 80: 1 ml 21.5 The conjugate was dissolved in DMSO. Then tween was mixed with it and finally buffer was added clear solution Test 3 5 2 Tween 80: 2 ml 16 For method see test 2 clear solution Test 4 24.8 2 Tween 80: 1 ml 20 For method see test 2 clear solution Test 5 21.3 1 Tween 80: 19 For method see test 2 cloudy solution Test 6 19.8 1 Tween 80: 19 For method see test 2 cloudy solution Test 7 20.4 1 Tween 80: 19 For method see test 2 clear solution Test 8 19.4 1 PEG 300: 1 ml 18 The conjugate was dissolved in DMSO. Then PEG was mixed with it and finally buffer was added precipitation

PAGE 128

128 LIST OF REFERENCES 1. J. Go mez Alonso, C. Andrade, and A. Koukoulis. On the definition of epileptic seizures and epilepsy. Epilepsia 46:1699 1700; author reply 1701 1692 (2005). 2. J. Gomez Alonsoand B.G. Giraldez. [Epilepsy: a new definition for an old disease]. Rev Neurol 45:12 6 127 (2007). 3. H. Jaseja. Definition of epilepsy: significance of its revision on clinical neurophysiological basis to improve prognosis and quality of life of patients with epilepsy. Med Hypotheses 72:756 757 (2009). 4. Z. Servit. [Epileptic seizure. Its pathogenetic mechanisms and its definition from the viewpoint of comparative physiology]. Cesk Fysiol 18:231 241 (1970). 5. J. Engeland T.A. Pedley. Epilepsy : a comprehensive textbook, Wolters Kluwer Health/Lippincott Williams & Wilkins, Philadelph ia, 2008. 6. L. Shargel. Comprehensive pharmacy review, Lippincott Williams & Wilkins, Baltimore, 2010. 7. R. D'Ambrosio, S. Hakimian, T. Stewart, D.R. Verley, J.S. Fender, C.L. Eastman, A.H. Sheerin, P. Gupta, R. Diaz Arrastia, J. Ojemann, and J.W. Mill er. Functional definition of seizure provides new insight into post traumatic epileptogenesis. Brain 132:2805 2821 (2009). 8. E. Faught. Seizure classification. Neurology 44:1555 (1994). 9. B.G. Katzung. Basic & clinical pharmacology, Lange Medical Boo ks/McGraw Hill, New York, 2004. 10. P.J. Coppleand J.B. Isom. Classification and treatment of seizure disorders in children. J Lancet 87:167 173 (1967). 11. J. Engel, Jr. Introduction to temporal lobe epilepsy. Epilepsy Res 26:141 150 (1996). 12. M.M. Acharya, B. Hattiangady, and A.K. Shetty. Progress in neuroprotective strategies for preventing epilepsy. Prog Neurobiol 84:363 404 (2008). 13. T.W. Strine, Kobau, R., Chapman, D.P., Thurman, D.J., Price, P., Balluz, L.S. Psychological distress, comorbi dities, and health behaviors among U.S. adults with seizures: results from the 2002 National Health Interview Survey. Epilepsia 46:1133 1139 (2005).

PAGE 129

129 14. J. Engel, Jr. Intractable epilepsy: definition and neurobiology. Epilepsia 42 Suppl 6:3 (2001). 15. B. Litt, R. Esteller, J. Echauz, M. D'Alessandro, R. Shor, T. Henry, P. Pennell, C. Epstein, R. Bakay, M. Dichter, and G. Vachtsevanos. Epileptic seizures may begin hours in advance of clinical onset: a report of five patients. Neuron 30:51 64 (2001). 1 6. M.J. McKeownand J.O. McNamara. When do epileptic seizures really begin? Neuron 30:1 3 (2001). 17. M. Manford, Y.M. Hart, J.W. Sander, and S.D. Shorvon. National General Practice Study of Epilepsy (NGPSE): partial seizure patterns in a general populati on. Neurology 42:1911 1917 (1992). 18. B.J. Wilderand R.P. Schmidt. Current classification of epilepsies. Guide to seizure type and characteristics. Postgrad Med 77:188 194, 199 (1985). 19. G. Regestaand P. Tanganelli. Clinical aspects and biological b ases of drug resistant epilepsies. Epilepsy Res 34:109 122 (1999). 20. N.C. de Lanerolleand T.S. Lee. New facets of the neuropathology and molecular profile of human temporal lobe epilepsy. Epilepsy Behav 7:190 203 (2005). 21. S.D. Shorvon. The etiolog ic classification of epilepsy. Epilepsia 52:1052 1057 (2011). 22. S.D. Shorvon, F. Andermann, and R. Guerrini. The causes of epilepsy : common and uncommon causes in adults and children, Cambridge University Press, Cambridge ; New York, 2011. 23. R.S. S loviter. Status epilepticus induced neuronal injury and network reorganization. Epilepsia 40 Suppl 1:S34 39; discussion S40 31 (1999). 24. M.S. Rao, B. Hattiangady, D.S. Reddy, and A.K. Shetty. Hippocampal neurodegeneration, spontaneous seizures, and mos sy fiber sprouting in the F344 rat model of temporal lobe epilepsy. J Neurosci Res 83:1088 1105 (2006). 25. Y. Perez, F. Morin, C. Beaulieu, and J.C. Lacaille. Axonal sprouting of CA1 pyramidal cells in hyperexcitable hippocampal slices of kainate treate d rats. Eur J Neurosci 8:736 748 (1996). 26. S. Williams, P. Vachon, and J.C. Lacaille. Monosynaptic GABA mediated inhibitory postsynaptic potentials in CA1 pyramidal cells of hyperexcitable hippocampal slices from kainic acid treated rats. Neuroscience 52:541 554 (1993).

PAGE 130

130 27. J.E. Franck, D.D. Kunkel, D.G. Baskin, and P.A. Schwartzkroin. Inhibition in kainate lesioned hyperexcitable hippocampi: physiologic, autoradiographic, and immunocytochemical observations. J Neurosci 8:1991 2002 (1988). 28. S.M. C ornishand H.V. Wheal. Long term loss of paired pulse inhibition in the kainic acid lesioned hippocampus of the rat. Neuroscience 28:563 571 (1989). 29. W.P. Grayand L.E. Sundstrom. Kainic acid increases the proliferation of granule cell progenitors in th e dentate gyrus of the adult rat. Brain Res 790:52 59 (1998). 30. J.M. Parentand D.H. Lowenstein. Mossy fiber reorganization in the epileptic hippocampus. Curr Opin Neurol 10:103 109 (1997). 31. A.K. Shetty, V. Zaman, and G.A. Shetty. Hippocampal neuro trophin levels in a kainate model of temporal lobe epilepsy: a lack of correlation between brain derived neurotrophic factor content and progression of aberrant dentate mossy fiber sprouting. J Neurochem 87:147 159 (2003). 32. A.K. Shetty. Progenitor cel ls from the CA3 region of the embryonic day 19 rat hippocampus generate region specific neuronal phenotypes in vitro. Hippocampus 14:595 614 (2004). 33. D.H. Lowenstein, M.S. Seren, and F.M. Longo. Prolonged increases in neurotrophic activity associated with kainate induced hippocampal synaptic reorganization. Neuroscience 56:597 604 (1993). 34. J.M. Parent, T.W. Yu, R.T. Leibowitz, D.H. Geschwind, R.S. Sloviter, and D.H. Lowenstein. Dentate granule cell neurogenesis is increased by seizures and contrib utes to aberrant network reorganization in the adult rat hippocampus. J Neurosci 17:3727 3738 (1997). 35. H.E. Scharfman, J.H. Goodman, and A.L. Sollas. Granule like neurons at the hilar/CA3 border after status epilepticus and their synchrony with area C A3 pyramidal cells: functional implications of seizure induced neurogenesis. J Neurosci 20:6144 6158 (2000). 36. J.M. Parent. The role of seizure induced neurogenesis in epileptogenesis and brain repair. Epilepsy Res 50:179 189 (2002). 37. J.M. Parent. Injury induced neurogenesis in the adult mammalian brain. Neuroscientist 9:261 272 (2003). 38. C. Gong, T.W. Wang, H.S. Huang, and J.M. Parent. Reelin regulates neuronal progenitor migration in intact and epileptic hippocampus. J Neurosci 27:1803 1811 (2007).

PAGE 131

131 39. H.E. Scharfman. Insight into Molecular Mechanisms of Catamenial Epilepsy. Epilepsy Curr 3:86 88 (2003). 40. H.E. Scharfman. Does the Development of a GABAergic Phenotype by Hippocampal Dentate Gyrus Granule Cells Contribute to Epileptogenesis Epilepsy Curr 2:63 (2002). 41. B. Hattiangady, M.S. Rao, and A.K. Shetty. Chronic temporal lobe epilepsy is associated with severely declined dentate neurogenesis in the adult hippocampus. Neurobiol Dis 17:473 490 (2004). 42. S. Oddo, P. Solis, D. Co nsalvo, B. Giagante, W. Silva, L. D'Alessio, E. Centurion, P. Saidon, and S. Kochen. Mesial temporal lobe epilepsy and hippocampal sclerosis: cognitive function assessment in Hispanic patients. Epilepsy Behav 4:717 722 (2003). 43. A. Alessio, B.P. Damasc eno, C.H. Camargo, E. Kobayashi, C.A. Guerreiro, and F. Cendes. Differences in memory performance and other clinical characteristics in patients with mesial temporal lobe epilepsy with and without hippocampal atrophy. Epilepsy Behav 5:22 27 (2004). 44. L .M. Brown Croyts, P.W. Caton, D.T. Radecki, and S.L. McPherson. Phenobarbital pre treatment prevents kainic acid induced impairments in acquisition learning. Life Sci 67:643 650 (2000). 45. A.S. Harvey, S.F. Berkovic, J.A. Wrennall, and I.J. Hopkins. Tem poral lobe epilepsy in childhood: clinical, EEG, and neuroimaging findings and syndrome classification in a cohort with new onset seizures. Neurology 49:960 968 (1997). 46. P.D. Fisher, E.F. Sperber, and S.L. Moshe. Hippocampal sclerosis revisited. Brain Dev 20:563 573 (1998). 47. F. Cendes. Febrile seizures and mesial temporal sclerosis. Curr Opin Neurol 17:161 164 (2004). 48. O. Devinsky. Diagnosis and treatment of temporal lobe epilepsy. Rev Neurol Dis 1:2 9 (2004). 49. J. Detour, H. Schroeder, D. Desor, and A. Nehlig. A 5 month period of epilepsy impairs spatial memory, decreases anxiety, but spares object recognition in the lithium pilocarpine model in adult rats. Epilepsia 46:499 508 (2005). 50. P. Gloor. Mesial temporal sclerosis: historical background and an overview from a modern perspective. In: Luders H, editor. Epilepsy surgery New York: Raven Press:689 703 (1991).

PAGE 132

132 51. J. Engel, Jr. Excitation and inhibition in epilepsy. Can J Neurol Sci 23:167 174 (1996). 52. Y. Ben Ariand F.E. Dudek. Primary and secondary mechanisms of epileptogenesis in the temporal lobe: there is a before and an after. Epilepsy Curr 10:118 125 (2010). 53. W. Loscher. Valproate: a reappraisal of its pharmacodynamic properties and mecha nisms of action. Prog Neurobiol 58:31 59 (1999). 54. M.K. Tickuand W.C. Davis. Effect of valproic acid on [3H]diazepam and [3H]dihydropicrotoxinin binding sites at the benzodiazepine GABA receptor ionophore complex. Brain Res 223:218 222 (1981). 55. W. Loscherand M. Vetter. Drug induced changes in GABA content of nerve endings in 11 rat brain regions. Correlation to pharmacological effects. Neurosci Lett 47:325 331 (1984). 56. A. Rohlfs, C. Rundfeldt, R. Koch, and W. Loscher. A comparison of the effec ts of valproate and its major active metabolite E 2 en valproate on single unit activity of substantia nigra pars reticulata neurons in rats. J Pharmacol Exp Ther 277:1305 1314 (1996). 57. Y.F. Zhang, J.W. Gibbs, 3rd, and D.A. Coulter. Anticonvulsant dru g effects on spontaneous thalamocortical rhythms in vitro: valproic acid, clonazepam, and alpha methyl alpha phenylsuccinimide. Epilepsy Res 23:37 53 (1996). 58. C.U. Johannessenand S.I. Johannessen. Valproate: past, present, and future. CNS Drug Rev 9: 199 216 (2003). 59. M.L. Zeise, S. Kasparow, and W. Zieglgansberger. Valproate suppresses N methyl D aspartate evoked, transient depolarizations in the rat neocortex in vitro. Brain Res 544:345 348 (1991). 60. P.W. Gean, C.C. Huang, C.R. Hung, and J.J. Tsai. Valproic acid suppresses the synaptic response mediated by the NMDA receptors in rat amygdalar slices. Brain Res Bull 33:333 336 (1994). 61. G.Y. Ko, L.M. Brown Croyts, and T.J. Teyler. The effects of anticonvulsant drugs on NMDA EPSP, AMPA EPSP, a nd GABA IPSP in the rat hippocampus. Brain Res Bull 42:297 302 (1997). 62. G. Gobbiand L. Janiri. Sodium and magnesium valproate in vivo modulate glutamatergic and GABAergic synapses in the medial prefrontal cortex. Psychopharmacology (Berl) 185:255 26 2 (2006).

PAGE 133

133 63. M.J. McLeanand R.L. Macdonald. Sodium valproate, but not ethosuximide, produces use and voltage dependent limitation of high frequency repetitive firing of action potentials of mouse central neurons in cell culture. J Pharmacol Exp Ther 237 :1001 1011 (1986). 64. M.J. Owensand C.B. Nemeroff. Pharmacology of valproate. Psychopharmacol Bull 37 Suppl 2:17 24 (2003). 65. A.R. Bolanos, M. Sarkisian, Y. Yang, A. Hori, S.L. Helmers, M. Mikati, P. Tandon, C.E. Stafstrom, and G.L. Holmes. Compariso n of valproate and phenobarbital treatment after status epilepticus in rats. Neurology 51:41 48 (1998). 66. C. Brandt, A.M. Gastens, M. Sun, M. Hausknecht, and W. Loscher. Treatment with valproate after status epilepticus: effect on neuronal damage, epil eptogenesis, and behavioral alterations in rats. Neuropharmacology 51:789 804 (2006). 67. S. Jessberger, K. Nakashima, G.D. Clemenson, Jr., E. Mejia, E. Mathews, K. Ure, S. Ogawa, C.M. Sinton, F.H. Gage, and J. Hsieh. Epigenetic modulation of seizure ind uced neurogenesis and cognitive decline. J Neurosci 27:5967 5975 (2007). 68. F. Penaand R. Tapia. Seizures and neurodegeneration induced by 4 aminopyridine in rat hippocampus in vivo: role of glutamate and GABA mediated neurotransmission and of ion chan nels. Neuroscience 101:547 561 (2000). 69. N.R. Temkin, S.S. Dikmen, G.D. Anderson, A.J. Wilensky, M.D. Holmes, W. Cohen, D.W. Newell, P. Nelson, A. Awan, and H.R. Winn. Valproate therapy for prevention of posttraumatic seizures: a randomized trial. J Ne urosurg 91:593 600 (1999). 70. H.K. Manji, G.J. Moore, G. Rajkowska, and G. Chen. Neuroplasticity and cellular resilience in mood disorders. Mol Psychiatry 5:578 593 (2000). 71. J.L. Noebelsand H.H. Jasper. Jasper's basic mechanisms of the epilepsies, Oxford University Press, New York, 2012. 72. M. Naessens, A. Cerdobbel, W. Soetaert, and E.J. Vandamme. Leuconostoc dextransucrase and dextran: production, properties and applications. Journal of Chemical Technology and Biotechnoloy 80:845 860 (2005). 7 3. L. Thoren. The dextrans -clinical data. Dev Biol Stand 48:157 167 (1980). 74. F. Markwardt, M. Richter, P. Walsmann, G. Riesener, and M. Paintz. Preparation of dextran bound recombinant hirudin and its pharmacokinetic behaviour. Biomed Biochim Acta 4 9:1103 1108 (1990).

PAGE 134

134 75. F.X. Jin, Y.H. Cheng, and K. Toda. Distribution model for the intact urokinase and urokinases modified by soluble macromolecules in rat and mouse bodies. Radioisotopes 37:441 447 (1988). 76. H. Yura, N. Yoshimura, T. Hamashima, K. Akamatsu, M. Nishikawa, Y. Takakura, and M. Hashida. Synthesis and pharmacokinetics of a novel macromolecular prodrug of Tacrolimus (FK506), FK506 dextran conjugate. J Control Release 57:87 99 (1999). 77. R. Mehvarand D.A. Hoganson. Dextran methylpredni solone succinate as a prodrug of methylprednisolone: immunosuppressive effects after in vivo administration to rats. Pharm Res 17:1402 1407 (2000). 78. X. Zhangand R. Mehvar. Dextran methylprednisolone succinate as a prodrug of methylprednisolone: plasma and tissue disposition. J Pharm Sci 90:2078 2087 (2001). 79. S. Mumtazand B.K. Bachhawat. Enhanced intracellular stability of dextran horse radish peroxidase conjugate: an approach to enzyme replacement therapy. Biochim Biophys Acta 1117:174 178 (1992) 80. S.D. Mikolajczyk, D.L. Meyer, R. Fagnani, M.S. Hagan, K.L. Law, and J.J. Starling. Dextran modification of a Fab' -beta lactamase conjugate modulated by variable pretreatment of Fab' with amine blocking reagents. Bioconjug Chem 7:150 158 (1996). 8 1. J. Marshalland J. Humphreys. Experimental enzyme therapy: suppression of allergic reactions in preimmunized animals by administration of exogenous enzymes in the form of dextran conjugates. Journal of Applied Biochemistry 1:88 94 (1979). 82. T.Y. Kim, S.G. Chong, J.W. Kim, and H. Choe. Anaphylactoid reactions to dextran. Korean Journal of Anesthesiology 13:69 72 (1980). 83. J.R. Maltby. Anaphylactic reaction to dextran. A case report. Br J Anaesth 40:552 553 (1968). 84. G. Briseidand K. Briseid. De xtran induced anaphylactoid reaction in man: altered reactivity of high molecular weight kininogen. Acta Pharmacol Toxicol (Copenh) 47:119 126 (1980). 85. S. Falchiand G.P. Novelli. [Dextran induced anaphylactoid reaction (DIAR). 2 clinical cases]. Miner va Anestesiol 59:543 545 (1993). 86. B. Garty, Y.L. Danon, and M. Nitzan. Anaphylactoid reaction to dextran in a 3 month old infant. Acta Paediatr Scand 73:559 (1984).

PAGE 135

135 87. K.G. Ljungstrom, H. Renck, H. Hedin, W. Richter, and B. Rosberg. Prevention of de xtran induced anaphylactic reactions by hapten inhibition. I. A Scandinavian multicenter study on the effects of 10 ml dextran 1, 15% administered before dextran 70 or dextran 40. Acta Chir Scand 149:341 348 (1983). 88. H. Renck, K.G. Ljungstrom, B. Rosb erg, K.G. Dhuner, and S. Dahl. Prevention of dextran induced anaphylactic reactions by hapten inhibition. II. A comparison of the effects of 20 ml dextran 1, 15%, administered either admixed to or before dextran 70 or dextran 40. Acta Chir Scand 149:349 3 53 (1983). 89. C. Larsen. Dextran prodrugs structure and stability in relation to therapeutic activity. Advance Drug Delivery Reviews 3:103 154 (1989). 90. R. Mehvar. Dextrans for targeted and sustained delivery of therapeutic and imaging agents. J Co ntrol Release 69:1 25 (2000). 91. E. Harboe, C. Larsen, M. Johansen, and H.P. Olesen. Macromolecular prodrugs. XV. Colon targeted delivery -bioavailability of naproxen from orally administered dextran naproxen ester prodrugs varying in molecular size in the pig. Pharm Res 6:919 923 (1989). 92. S. Vyas, P. Trivedi, and S.C. Chaturvedi. Ketorolac dextran conjugates: synthesis, in vitro and in vivo evaluation. Acta Pharm 57:441 450 (2007). 93. C. Larsen, E. Harboe, M. Johansen, and H.P. Olesen. Macromole cular prodrugs. XVI. Colon targeted delivery -comparison of the rate of release of naproxen from dextran ester prodrugs in homogenates of various segments of the pig gastrointestinal (GI) tract. Pharm Res 6:995 999 (1989). 94. C. Larsen. Hydrolysis of de xtran metronidazole monosuccinate ester prodrugs. Evidence for participation of intramolecularly catalyzed hydrolysis of the conjugate metronidazole succinic acid ester by the neighbouring dextran hydroxy groups. Acta Pharm Suec 23:279 288 (1986). 95. A. D. McLeod, D.R. Friend, and T.N. Tozer. Synthesis and chemical stability of glucocorticoid dextran esters: potential prodrugs for colon specific delivery. International Journal of Pharmaceutics 92:105 114 (1993). 96. T. Kojima, M. Hashida, S. Muranishi, and H. Sezaki. Mitomycin C dextran conjugate: a novel high molecular weight pro drug of mitomycin C. J Pharm Pharmacol 32:30 34 (1980). 97. H. Yura, N. Yoshimura, T. Hamashima, T. Oka, Y. Takakura, and M. Hashida. Development of a macromolecular prodrug of FK506: I. Synthesis of FK506 dextran conjugate. Transplant Proc 30:3598 3599 (1998).

PAGE 136

136 98. C. Larsen, P. Kurtzhals, and M. Johansen. Macromolecular prodrugs. IX. The release kinetics of metronidazole from various dextran dicarboxylic acid hemiester conju gates in aqueous buffer, human plasma and in pig liver homogenate. Acta Pharm Suec 25:1 14 (1988). 99. J. Vermeersch, F. Vandoorne, D. Permentier, and E. Schact. Macromolecular prodrugs of metronidazole. 1. Esterification of hydroxyl containing polymers with metronidazole monosuccinate. Bull Soc Chim Belg 94:591 596 (1985). 100. C.Y. Wonand C.C. Chu. Dextran estrone conjugate: Synthesis and in vitro release study. Carbohydrate Polymers 36:327 334 (1998). 101. M. Harada, H. Sakakibara, T. Yano, T. Suzu ki, and S. Okuno. Determinants for the drug release from T 0128, camptothecin analogue carboxymethyl dextran conjugate. J Control Release 69:399 412 (2000). 102. S. Penugonda, A. Kumar, H.K. Agarwal, K. Parang, and R. Mehvar. Synthesis and in vitro chara cterization of novel dextran methylprednisolone conjugates with peptide linkers: effects of linker length on hydrolytic and enzymatic release of methylprednisolone and its peptidyl intermediates. J Pharm Sci 97:2649 2664 (2008). 103. W. Loscher. Basic ph armacology of valproate: a review after 35 years of clinical use for the treatment of epilepsy. CNS Drugs 16:669 694 (2002). 104. C. Hariton, L. Ciesielski, S. Simler, M. Valli, G. Jadot, S. Gobaille, E. Mesdjian, and P. Mandel. Distribution of sodium va lproate and GABA metabolism in CNS of the rat. Biopharm Drug Dispos 5:409 414 (1984). 105. D.D. Shen. Valproate: absorption, distribution and excretion. In: Loscher W, editor. Valproate Basle: Birkhauser:77 90 (1999). 106. R. Davis, D.H. Peters, and D. McTavish. Valproic acid. A reappraisal of its pharmacological properties and clinical efficacy in epilepsy. Drugs 47:332 372 (1994). 107. Y.R. Qian, M.J. Lee, S. Hwang, J.H. Kook, J.K. Kim, and C.S. Bae. Neuroprotection by valproic Acid in mouse models o f permanent and transient focal cerebral ischemia. Korean J Physiol Pharmacol 14:435 440 (2010). 108. J. Biermann, P. Grieshaber, U. Goebel, G. Martin, S. Thanos, S. Di Giovanni, and W.A. Lagreze. Valproic acid mediated neuroprotection and regeneration i n injured retinal ganglion cells. Invest Ophthalmol Vis Sci 51:526 534 (2010).

PAGE 137

137 109. L. Schmued, K. Kyriakidis, and L. Heimer. In vivo anterograde and retrograde axonal transport of the fluorescent rhodamine dextran amine, Fluoro Ruby, within the CNS. Bra in Res 526:127 134 (1990). 110. A. Vercelli, M. Repici, D. Garbossa, and A. Grimaldi. Recent techniques for tracing pathways in the central nervous system of developing and adult mammals. Brain Res Bull 51:11 28 (2000). 111. A. Reiner, C.L. Veenman, L. Medina, Y. Jiao, N. Del Mar, and M.G. Honig. Pathway tracing using biotinylated dextran amines. J Neurosci Methods 103:23 37 (2000). 112. Y.J. Jung, J.S. Lee, H.H. Kim, Y.T. Kim, and Y.M. Kim. Synthesis and properties of dextran 5 aminosalicyclic acid e ster as a potential colon specific prodrug of 5 aminosalicylic acid. Arch Pharm Res 21:179 186 (1998). 113. J. Varshosaz, J. Emami, N. Tavakoli, A. Fassihi, M. Minaiyan, F. Ahmadi, and F. Dorkoosh. Synthesis and evaluation of dextran budesonide conjugate s as colon specific prodrugs for treatment of ulcerative colitis. Int J Pharm 365:69 76 (2009). 114. Y.N. Pang, Y. Zhang, and Z.R. Zhang. Synthesis of an enzyme dependent prodrug and evaluation of its potential for colon targeting. World J Gastroenterol 8:913 917 (2002). 115. J.S. Lee, Y.J. Jung, M.J. Doh, and Y.M. Kim. Synthesis and properties of dextran nalidixic acid ester as a colon specific prodrug of nalidixic acid. Drug Dev Ind Pharm 27:331 336 (2001). 116. H. Amini, M. Javan, and A. Ahmadiani. Development and validation of a sensitive assay of valproic acid in human plasma by high performance liquid chromatography without prior derivatization. Journal of Chromatography B 830:368 371 (2006). 117. R. Mehvar. Simultaneous analysis of dextran met hylprednisolone succinate, methylprednisolone succinate, and methylprednisolone by size exclusion chromatography. J Pharm Biomed Anal 19:785 792 (1999). 118. H. Amini, Javan, M. and Ahmadiani, A. Development and validation of a sensitive assay of valproi c acid in human plasma by high performance liquid chromatography without prior derivatization. Journal of Chromatography B 830:368 371 (2006).

PAGE 138

138 119. J. Varshosaz, J. Emami, F. Ahmadi, N. Tavakoli, M. Minaiyan, A. Fassihi, P. Mahzouni, and F. Dorkoosh. Pre paration of budesonide dextran conjugates using glutarate spacer as a colon targeted drug delivery system: in vitro/in vivo evaluation in induced ulcerative colitis. J Drug Target 19:140 153 (2011). 120. K.L. Rensberger, D.A. Hoganson, and R. Mehvar. Dex tran methylprednisolone succinate as a prodrug of methylprednisolone: in vitro immunosuppressive effects on rat blood and spleen lymphocytes. Int J Pharm 207:71 76 (2000). 121. C. Liu, Q. Lin, Y. Gao, L. Ye, Y. Xing, and T. Xi. Characterization and antit umor activity of polysaccharide from Strongylocentrotus nudus eggs. Carbohydrate Polymers 67:313 318 (2007). 122. R.L. Sidebotham. Dextrans. Advances in Carbohydrate Chemistry and Biochemistry 30:371 444 (1974). 123. F.R. Seymour. Structural analysis o f dextrans containing 2 o D D glucopyranosyl residues at the branch points, by use of 13 C nuclear magnetic resonance spectroscopy and gas liquid chromatography mass spectrometry. Carbohydrate Research 71:231 250 (1979a). 124. F.R. Seymou r. Correlation of the structure of dextran to their 1 H NMR spectra. Carbohydrate Research 74:77 92 (1979b). 125. F.R. Seymourand R.D. Knapp. Structural analysis of dextrans from strains of Leuconostoc and related genera that contain 3 O glucosylated D glucopyranosyl residues at the branched points, or in consecutive, linear position. Carbohydrate Research 81:105 129 (1980). 126. N.W. Cheethamand E. Fiala Beer. Dextran structural details from high field proton NMR spectroscopy. Carbohydrate Polymers 1 4:149 158 (1991). 127. A.S. Williamsand G. Taylor. Synthesis, characterization and releas of cromoglycate from dextran conjugates. International Journal of Pharmaceutics 83:233 239 (1982). 128. A.D. McLeod, D.R. Friend, and T.N. Tozer. Glucocorticoid de xtran conjugates as potential prodrugs for colon specific delivery: hydrolysis in rat gastrointestinal tract contents. J Pharm Sci 83:1284 1288 (1994). 129. E. Catikerand A. Guner. Unperturbed molecular dimensions and the theta temperature of dextran in dimethylsulfoxide (DMSO) solutions. Polymer Bulletin 41:223 230 (1998). 130. J.T. Carstensen. Drug stability : principles and practices, Dekker, New York, 1995.

PAGE 139

139 131. M. Johansenand C. Larsen. Macromolecular prodrugs. II. Influence of variation in molecul ar weight and degree of substitution of O benzoyl dextran conjugates on their physicochemical properties and stability in aqueous buffer and in plasma. International Journal of Pharmaceutics 27:219 231 (1985). 132. H. Weibel, L.S. Nielsen, C. Larsen, and H. Bundgaard. Macromolecular prodrugs. IXX. Kinetics of hydrolysis of benzyl dextran carbonate ester conjugates in aqueous buffer solutions and human plasma. Acta Pharm Nord 3:159 162 (1991). 133. C. Larsen. Dextran prodrugs structure and stability in relation to therapeutic activity. Advance Drug Delivery Reviews 3:103 154 (1989). 134. C. Larsenand M. Johansen. Macromolecular prodrugs XI. Regeneration rates of various NSAID compounds from their corresponding dextran ester prodrugs in aqueous buffer a nd in different biological media. Acta Pharm Nord 2:57 66 (1989). 135. T. Imai. Human carboxylesterase isozymes: catalytic properties and rational drug design. Drug Metab Pharmacokinet 21:173 185 (2006). 136. M. Xie, D. Yang, L. Liu, B. Xue, and B. Ya n. Human and rodent carboxylesterases: immunorelatedness, overlapping substrate specificity, differential sensitivity to serine enzyme inhibitors, and tumor related expression. Drug Metab Dispos 30:541 547 (2002). 137. M. Mori, M. Hosokawa, Y. Ogasawara, E. Tsukada, and K. Chiba. cDNA cloning, characterization and stable expression of novel human brain carboxylesterase. FEBS Lett 458:17 22 (1999). 138. T. Imai, M. Taketani, M. Shii, M. Hosokawa, and K. Chiba. Substrate specificity of carboxylesterase is ozymes and their contribution to hydrolase activity in human liver and small intestine. Drug Metab Dispos 34:1734 1741 (2006). 139. M. Dixonand E.C. Webb. Enzymes, Academic Press, New York,, 1964. 140. L.G. Rajewski, Stinnett, A.A., Stell, V.J., and Top p, E.M. Enzymic and non enzymic hydrolysis of a polymeric prodrug: Hydrocortisone esters of hyaluronic acid. International Journal of Pharmaceutics 82:205 213 (1992). 141. A.J. D'Souzaand E.M. Topp. Release from polymeric prodrugs: linkages and their deg radation. J Pharm Sci 93:1962 1979 (2004). 142. B. Neisesand W. Steglich. Simple method for the esterification of carboxylic acids. Angewandte Chemie 17:522 524 (1978).

PAGE 140

140 143. M. Elomaa, T. Asplund, P. Soininen, R. Laatikainen, S. Peltonen, S. Hyvarinen, and A. Urtti. Determination of the degree of substitution of acetylated starch by hydrolysis, 1 H NMR and TGA/IR. Carbohydrate Polymers 57:261 267 (2004). 144. M.R. Rodrigues. Hydrophobic derivatives of dextran polysaccharide: characterization and propert ies. Journal of Carbohydrate Chemistry 24:733 744 (2005). 145. N.A. Vieira, M.S. Moscardini, V.A. Tiera, and M.J. Tiera. Aggregation behavior of hydrophobically modified dextran in aquous solution: a fluorescence probe study. Carbohydrate Polymers 53:13 7 143 (2002). 146. G. Hoffmann, C. Funk, S. Fowler, M.B. Otteneder, A. Breidenbach, C.R. Rayner, T. Chu, and E.P. Prinssen. Nonclinical pharmacokinetics of oseltamivir and oseltamivir carboxylate in the central nervous system. Antimicrob Agents Chemother 53:4753 4761 (2009). 147. J. Bernsohn, K.D. Barron, P.F. Doolin, A.R. Hess, and M.T. Hedrick. Subcellular localization of rat brain esterases. J Histochem Cytochem 14:455 472 (1966). 148. H. Greimand R. Snyder. Toxicology and risk assessment : a compre hensive introduction, John Wiley & Sons, Chichester, West Sussex ; Hoboken, NJ, 2007. 149. P.C. Seville, C. Simons, G. Taylor, and P.A. Dickinson. Prodrug to probe solution HFA pMDI formulation and pulmonary esterase activity. Int J Pharm 195:13 16 (2000 ). 150. A.P. Morris, K.R. Brain, and C.M. Heard. Synthesis of haloperidol prodrugs and their hydrolysis by porcine liver esterase. Drug Metab Lett 2:275 279 (2008). 151. W.N. van Dijk Wolthuis, S.K. Tsang, J.J. Bosch, and W.E. Hennink. A new class of po lymerizable dextrans with hydrolyzable groups: hydroxyethyl methacrylated dextran with and without oligolactate spacer. Polymer 38:6235 6242 (1997). 152. T. Kawaguchi, Y. Suzuki, N. Nambu, and T. Nagai. Specificity of esterases and structures of prodrug esters. III. Activity of rat tissue homogenates, rat plasma and porcine liver esterase for the hydrolysis of 3',5' bis dicarboxylic acid hemiesters of 5 fluoro 2' deoxyuridine. Chem Pharm Bull (Tokyo) 33:2956 2961 (1985). 153. A.J. Adlerand G.B. Kistiako wsky. Isolation and properties of pig liver esterase. J Biol Chem 236:3240 3245 (1961). 154. E. Heymannand W. Junge. Characterization of the isoenzymes of pig liver esterase. 1. Chemical studies. European Journal of Biochemistry 95:509 518 (1979).

PAGE 141

141 155. E. Heymannand W. Junge. Characterization of the isoenzymes of pig liver esterase. 2. Kinetic studies. European Journal of Biochemistry 95:519 525 (1979). 156. C. Larsenand M. Johansen. Macromolecular prodrugs IV. Kinetics of hydrolysis of metronidazole m onosuccinate dextran ester conjugates in aqueous solution and in plasma sequential release of metronidazole from the conjugates at physiological pH. International Journal of Pharmaceutics 35:39 45 (1987). 157. S. Han, C. Kim, and D. Kwon. Thermal/oxidat ive degradation and stabilization of polyethylene glycol. Polymer 38:317 323 (1997). 158. D. Sgouras. Methods for the evaluation of biocompatibility of soluble synthetic polymers which have potential for biomedical use: 1 Use of the tetrazolium based c olorimetric assay (MTT) as a preliminary screen for evaluation ofin vitro cytotoxicity. Journal of Materials Science: Materials in Medicine 1:61 68 (1990). 159. C.J. Phiel, F. Zhang, E.Y. Huang, M.G. Guenther, M.A. Lazar, and P.S. Klein. Histone deacetyl ase is a direct target of valproic acid, a potent anticonvulsant, mood stabilizer, and teratogen. J Biol Chem 276:36734 36741 (2001). 160. M.T. Stockhausen, J. Sjolund, C. Manetopoulos, and H. Axelson. Effects of the histone deacetylase inhibitor valproi c acid on Notch signalling in human neuroblastoma cells. Br J Cancer 92:751 759 (2005). 161. T.C. Laurent. Enzyme reactions in polymer media. European Journal of Biochemistry 21:498 506 (1971). 162. T. Liebert, S. Hornig, S. Hesse, and T. Heinze. Nanop articles on the basis of highly functionalized dextrans. J Am Chem Soc 127:10484 10485 (2005). 163. A. Aumelas, A. Serrero, A. Durand, E. Dellacherie, and M. Leonard. Nanoparticles of hydrophobically modified dextrans as potential drug carrier systems. C olloids Surf B Biointerfaces 59:74 80 (2007). 164. J. Varshosaz, S. Eskandari, and M. Tabakhian. Production and optimization of valproic acid nanostructured lipid carriers by the Taguchi design. Pharm Dev Technol 15:89 96 (2010).

PAGE 142

142 BIOGRAPHICAL SKETC H A M Mahbub Hassan was born in Tangail, Bangladesh in 1972. He earned his Bachelor of Pharmacy in 1995 and Master of Pharmacy in 1997 from the University of Dhaka, Bangladesh. Then, he joined Beximco Pharmaceuticals Ltd., a leading pharmaceutical compan y in Bangladesh in 1997 and continued to work there until the end of 1999 He attended the graduate program in Long Island University, Brooklyn, New York, USA from January 2000 until July 2003. He started working as a Graduate Pharmacy Intern on August 200 3 in Eckerd Drugs in Lady Lake, FL He became a Registered Pharmacist on January 2004 and continued to wo rk as a Registered Pharmacist for Eckerd Drugs which later beca m e CVS Pharmacy. He joined Ph D program in the Department of Pharmaceutics, University of Florida in Fall 2005. He continued his graduate study while he was working full time as a Registered in 2007 and continued his research under his ment left academia to accept a positio n in a Pharmaceutical Industry. It was a daunting and challenging task to pursue a Ph D while working full time and raising a family. He continued his res earch steadfastly under the mentorship of Dr. Hochhaus and finally graduated on December 2012.