Citation
Applications of hydroxyalkyl derivatives of a pyridinium salt-dihydropyridine redox system for drug delivery to the brain

Material Information

Title:
Applications of hydroxyalkyl derivatives of a pyridinium salt-dihydropyridine redox system for drug delivery to the brain
Creator:
Phelan, Michael James
Publication Date:
Language:
English
Physical Description:
x, 119 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Blood ( jstor )
Dihydropyridines ( jstor )
Drug carriers ( jstor )
Esters ( jstor )
Ethers ( jstor )
Nitrogen ( jstor )
Protons ( jstor )
Pyridines ( jstor )
Rats ( jstor )
Solvents ( jstor )
Blood-Brain Barrier -- drug effects ( mesh )
Brain -- drug effects ( mesh )
Dissertations, Academic -- Medicinal Chemistry -- UF ( mesh )
Medicinal Chemistry thesis Ph.D ( mesh )
Pyridinium Compounds -- pharmacokinetics ( mesh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph.D.)--University of Florida, 1987.
Bibliography:
Bibliography: leaves 114-118.
General Note:
Typescript.
General Note:
Vita.
Statement of Responsibility:
by Michael James Phelan.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. This item may be protected by copyright but is made available here under a claim of fair use (17 U.S.C. §107) for non-profit research and educational purposes. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
030217980 ( ALEPH )
17945271 ( OCLC )

Downloads

This item has the following downloads:


Full Text
APPLICATIONS OF HYDROXYALKYL DERIVATIVES
OF A PYRIDINIUM SALT-DIHYDROPYRIDINE REDOX SYSTEM
FOR DRUG DELIVERY TO THE BRAIN
By
MICHAEL JAMES PHELAN
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
1987




FOR MY PARENTS
John (Johnny) and Nadia Phelan
The two people who have loved me every day of my life, and taught me more than any school ever could.
Education is that which remains when one has forgotten everything learned in school.
Albert Einstein (1879-1955)




ACKNOWLEDGMENTS
First of all I would like to express my sincere thanks to Dr. Nicholas Bodor for his kindness, generosity and patience. I appreciate the opportunity to have worked under his direction. It has been a remarkable experience. I would also like to thank Mrs. Sheryl Bodor for her kindness. In addition, I would like to thank each of the members of my committee for their help along the way.
I want to thank the members of Dr. Bodor's group for their help and friendship over the years. I have had the great pleasure of having worked with a wonderful group of people. I must especially mention Dr. Efraim Shek for his help and guidance, especially during difficult times. Those others who have shown real friendship will not be forgotten. I make special reference to Gabrielle Brouillette, my love; Vasu Venkatraghavan, my friend and confidant; and Mrs. Jirina Vlasak, a truly good person and one of the nicest I have ever met.
To all of the others who have especially extended their kindness, David Winwood, Toshio Nakamura, M. Masaki, L. J. Chang, Cindy Jordon, Joan Martignago, Laurie Johnston, Jill McCornack, Pascal Druzgala, Thorsteinn Loftsson, Hartmut Derendorf and Kerry Estes, Emil Pop, Teruo Murakami, Whei Mei Wu, K. S. and Nirmala Raghavan, James Stephens, and Roy
iii




Little, I thank you for making things better. I would also like to thank Bob Perchalski for his curative advice and help with the analytical portion of my work, and Dr. Katovich for getting me started on and loaning the equipment needed for the antipyretic activity study. I wish to express my appreciation to Marcus Brewster for his early help during the good times. I thank Anna Marie Martin for her ability to read the illegible during her superb typing of this manuscript.
I also need to acknowledge my best friends whose caring has helped sustain me through the years. I would especially like to thank my best and oldest friends, Tony Caporaletti and Rocco Caponi, my best friends from U of A, Joe Snyder and Vasu Venkatraghavan, and last but not least, my family, especially my parents, my brother Jack and my grandmother for never wavering.
iv




TABLE OF CONTENTS
Page
ACKNOWLEDGMENTS ........................ .. ........... iii
KEY TO ABBREVIATIONS ............ o ..... ..... o ..... .. vi
ABSTRACTo...... .. .. . .oo....... . .. . ............ viii
CHAPTERS
I INTRODUCTION .......... o ...... o ...... ... ... 1
The Blood-Brain Barrier ....2......... 2
Brain Specific Drug Delivery.................3
Chemistry of the Drug Delivery System........ 6 Epilepsy and Valproic Acid .....o........... 13
Nonsteroidal Anti-inflammatory Agents..... 18 Objectives.............................. 22
II EXPERIMENTAL ....... ...... ... .. ....... ...... 24
Materials and Methods .................... 24
High Pressure Liquid Chromatography Systems. 54 Chemical Stability ........................o. 56
In Vitro Studies .... o .................. o ... 57
In Vivo Studies ............................ 60
III RESULTS AND DISCUSSION.............. ........ 65
Synthesis ..........
Chemical Stability...................... .. 78
In Vitro Studies............................ 81
In Vivo Studies ....o........................ 89
IV CONCLUSION ........ o. ..... .... ........ 111
REFERENCES ...... -........... .. ... .. ........ 114
BIOGRAPHICAL SKETCH ....... ..o... .... ..... .......... 119
v




KEY TO ABBREVIATIONS BBB blood-brain barrier bp boiling point CDS chemical delivery system CNS central nervous system C carrier C degrees centigrade D drug
DCC dicyclohexylcarbodiimide DHC-D dihydropyridine carrier-drug DH-CDS dihydropyridine-chemical delivery system DMAP dimethylaminopyridine DMSO dimethylsulfoxide ED50 dose effective in 50% of experimental units g gram
GABA gamma aminobutyric acid GAD glutamic acid decarboxylase GI gastrointestinal h hour
IHNMR proton nuclear magnetic resonance HPLC high pressure liquid chromatography icy intracerebroventricular iv intravenous kg kilogram
vi




LD50 dose lethal in 50% of experimental units lit. literature M molar mg milligram min minute mL milliliter mm millimeter mM millimolar mp melting point MW molecular weight NAD+ nicotinamide adenine dinucleotide, oxidized form NADH nicotinamide adenine dinucleotide, reduced form nm nanometer QC+ quaternary carrier QC+-D quaternary carrier-drug r correlation coefficient rpm revolutions per minute SEM standard error of the mean sec second TI therapeutic index tl/2 half-life UV ultraviolet vs versus w/v weight per volume uL microliter > greater than
vii




Abstract of Dissertation Presented to the Graduate
School of the University of Florida in Partial
Fulfillment of the Requirements for the Degree of Doctor of Philosophy
APPLICATIONS OF HYDROXYALKYL DERIVATIVES
OF A PYRIDINIUM SALT-DIHYDROPYRIDINE REDOX SYSTEM
FOR DRUG DELIVERY TO THE BRAIN By
Michael James Phelan
May 1987
Chairman: Nicholas S. Bodor Major Department: Medicinal Chemistry
The design of drugs whose site of action is in the brain is often complicated by the presence of the bloodbrain barrier. This barrier is a physical entity which restricts all but the most lipophilic of compounds from entering or leaving the brain. To overcome this problem a system has been developed utilizing pyridinium saltdihydropyridine redox carriers. The lipophilic dihydropyridine is designed to penetrate the blood-brain barrier and then enzymatically oxidize to the hydrophilic pyridinium salt. This charged molecule allows for a "lock in" effect and subsequent sustained release of the parent drug in the brain. The more hydrophilic form can also be rapidly eliminated from the periphery. These characteristics should result in a reduction or elimination
viii




of both central and peripheral toxicities associated with the parent drug.
The carriers developed in this work contained a
nicotinamide moiety derivatized with a hydroxyalkyl group which varied in length and position. This group allowed drugs containing a carboxylic acid functional group to be coupled to the carrier via an ester linkage. Various chain lengths and positions were used in an effort to prepare the most advantageous carrier possible.
Naproxen, indomethacin, and valproic acid were each
used to prepare one or more chemical delivery systems. The rate of disappearance was measured for each naproxen and indomethacin dihydropyridine compound along with its corresponding pyridinium salt, under a variety of in vitro conditions.
Three of the most promising delivery systems and
naproxen itself were each administered to rats to study their in vivo distribution. These experiments proved the carriers' ability to deliver and sustain the release of either parent drug. The investigation also showed a dramatic increase in brain/blood and brain/tissue ratios after giving one naproxen dihydropyridine delivery system when compared to the ratios that were measured after administering naproxen itself. This represented a significant improvement in the distribution of the drug, in accord with the system design. The sustained effect of the dihydropyridine compound significantly improved naproxen's
ix




antipyretic activity in a rat model, when compared to an equimilar dose of naproxen itself.
x




CHAPTER I
INTRODUCTION
Water is an essential "element" of life. It has many
unusual and amazing properties. One example of this, becoming less dense as it freezes, is what allows life to survive in freezing water. The exposed top freezes first, and the resulting ice forms an insulating layer protecting the water and the life below.
Water is also an excellent solvent. It can dissolve a vast number and wide variety of compounds and elements. It is the basis of bioorganic reactions, the solvent of choice! Water is the major component of the blood, and so plays a leading role in distributing the nutrients of life.
Water is also capable of dissolving and in the case of blood, distributing many unwanted products. This allows for the spread of a toxic substance throughout ones circulation. However, one is not entirely without protection against these attacks. The blood-brain barrier is designed to filter the circulating blood flow. In this way, it can protect the brain from a number of harmful substances. Unfortunately, it can also vastly complicate drug design.
1




2
The Blood-Brain Barrier
The term blood-brain barrier was coined in 1921 by the Russian, Lina Saomonoona Stern.1 It was used to describe the impediment to acidic dyes which stained most organs, but not the brain, when given by the circulation.1,2 This observation led to the concept of a membrane wall-like barrier, which existed as a structural blockade interposed between the blood and the brain. This term was later used to explain the restriction from the brain of many water soluble substances, even some of low molecular weight. The basis of this barrier has been shown to be due to the physical aspects of the brain capillary system.3'4
These cerebral capillaries differ from general systemic capillaries in several ways. Endothelial cells of systemic capillaries are joined by relatively loose junctions which allow the passage of many compounds into the extracellular fluid surrounding these areas. In contrast, the endothelial cells which comprise central nervous system capillaries are joined by tight intercellular junctions, which form a continuous cellular layer between the blood and the brain.5-8 Since little passage of water-soluble compounds is possible, and endothelial cell membranes present a significant impediment to transcapillary movement, the result is an atypical barrier which has developed as protection for the CNS.
Movement of solute across the blood brain barrier (BBB) is consistent with the theory of simple diffusion through an




3
aporous lipid membrane.9 Pores between cells do not exist; therefore, bulk flow cannot take place across the BBB. However, the cell membranes do consist of a lipid bilayer having globular proteins asymmetrically distributed. Substances which are lipid soluble will cross the membranes in both directions by simple diffusion. Therefore, the rate at which a compound enters the brain is normally related to its lipid solubility.10'11
The molecular weight of a compound can also be of great importance when considering transport of drugs through the BBB. Larger molecules with MW>500 tend to diffuse very slowly through biological membranes.11'12 Many small watersoluble compounds, however, are thought to pass the BBB by carrier-mediated transport.13'14
In order to deliver therapeutically useful drugs to the brain, one must often circumvent the function of the bloodbrain barrier. A novel and effective way to do this has been brought to light by Bodor et al.15 It involves the site-specific, sustained release of drugs to the brain. Drugs which are normally partially or completely barred from the brain can be delivered using a chemical delivery system (CDS). This system employs carriers containing a pyridine ring moiety, capable of participating in a pyridinium saltdihydropyridine redox-type reaction.
Brain Specific Drug Delivery
The effect of a drug is usually the result of the drug's interaction with macromolecules in some specific




4
target tissue or cell type. Along with the compound's desired therapeutic effect, it will also bring about certain toxic consequences depending on concentration. Often, the specific site of toxicity is not associated with the target tissue. Therefore, if the distribution of a drug could be more precisely controlled, the therapeutic index (TI = LD50/ED50) would likely increase.
The method for brain specific delivery of drugs developed by Bodor has shown great promise of achieving this goal. The first compound used to evaluate this system was phenethylamine. Phenethylamine (D) was chemically coupled with nicotinic acid (C). It was then quaternized to the pyridinium salt (QC+-D), followed by reduction to the dihydropyridine form (DHC-D) using sodium dithionite.
Following iv administration, the lipophilic chemical delivery system (DHC-D) distributes rapidly in both the blood and the brain as illustrated in Figure 1-1. Oxidation back to the original quaternary salt (QC+-D) is then thought to occur via the NAD+-NADH redox enzyme system.15 This more hydrophilic form can be rapidly eliminated from the body resulting in a lower circulating level of the parent drug. In the brain, however, the polar drug-carrier combination (QC+-D) is locked in by the blood-brain barrier. This compound then serves as a source for the sustained release of the parent drug in the brain. The slow release of the drug (D) depends on the rate of enzyme hydrolysis. This is affected by the structure of both the carrier as well as the




5
QC+-D
Reduction
DHC-D
Brain Periphery
DH -D DHC-D
kox kox
QC+-D QC+-D
kcleavage kcleavage
koutl QC+ D QC+ kout2
II
BBB kout3 kout4
Figure 1-1. Schematic representation of the proposed
pyridinium salt-dihydropyridine drug delivery
system




6
parent compound. Once the drug-carrier combination is hydrolyzed, the remaining carrier portion, due to its lower molecular weight, should be actively transported out of the brain. This system should provide for better delivery of a compound to the brain, while offering the advantages of sustained release and reduced toxicity.
Chemistry of The Drug Delivery System Pyridinium Salts
When a nitrogen atom in a heterocyclic ring possesses a lone pair of electrons, those electrons can form a bond between that nitrogen atom and a carbon atom of suitable polarizability. In this case, the nitrogen is in a quaternary form. The attacked molecule must be one that can release an anion during the quaternization.
This type of reaction can be viewed in two ways. The first is to see the reaction as a nucleophilic replacement of the halogen or other similar leaving group, by attack of the lone pair of the ring nitrogen. Alternatively, the quaternization can be looked at as an electrophilic attack on the ring, which usually takes place only at a nitrogen atom.
Therefore, the availability of the electron pair, as influenced by the ring substituents and the steric factors involved, can dramatically affect the rate of quaternization. The solvent used and the nature of the electrophile involved in the reaction are also important factors in predicting the course of a given reaction.




7
Alkyl halides are by far the most common reagents for
the formation of heterocyclic quaternary salts and of these, iodoalkanes are most often reported.16 Primary halides, as expected, react faster than secondary compounds,17 and tertiary halides normally result in elimination, giving the corresponding acid and an alkene.18
The solvent used in the formation of heterocyclic salts can also be quite important. Many solvents of varying polarities have been used, including an excess of the quaternizing agent itself. The solvent's polarity affects the rate of the reaction, as well as the products ability to precipitate from the reaction mixture. The influence of a solvent on the problem of isolating a quaternary salt, once formed, is at times a major one. Water is often held very tightly by the desired product. Non-hydric solvents such as benzene usually cause such reactions to be quite slow. Therefore, the best solvent would seem to be non-hydric, with a reasonably high dielectric constant.
The most important quaternary pyridine derivatives
occurring naturally are those of nicotinamide, which have a coenzyme function. In metabolic oxidations, these coenzymes accept hydrogen directly from a variety of oxidizable substrates and transfer it to other acceptors. In this way, they are key metabolic catalysts. The 1-methylderivative of nicotinic acid is found in plants and is known as trigonelline.19 Another pyridinium compound of biological importance, picolinaldoxine methiodide (PAM), evolved as an




8
antidote for organophosphate poisoning. This bifunctional molecule was designed to serve as an antidote by binding to one enzyme binding site and, at the same time, providing a nucleophilic group to displace the bound phosphate. Dihydropyridines
The partial reduction of pyridinium salts to dihydro derivatives of known structure can be successfully carried out in a limited number of cases. In part, the difficulty may lie in the readiness of the partly reduced structure to oxidize, polymerize, or be further reduced unless stabilizing groups are present. Of all the chemical methods thus far applied, reduction by sodium dithionite to dihydro derivatives as studied by Karrer et al. is probably the most important.20 With dithionite, a variety of nicotinamide quaternary salts have been converted to 1,4-dihydro products.15,20,21
Here, the carbamoyl group exerts a stabilizing influence on the product. Earlier studies of 1,2- and 1,4-dihydropyridines sought to avoid the stability difficulties and also limit the opportunities for tautomerization by using highly substituted derivatives.20 The first use of sodium dithionite to convert NAD+ into NADH led to the preparation of a number of dihydropyridines by this method. Reduction of 3-substituted or 3,5-disubstituted pyridinium compounds by sodium dithionite in mildly basic solution yielded the corresponding 1,4-dihydro-pyridines.15,21 Some early reports of 1,2-dihydropyridines22 have since been shown to actually be the 1,4-isomer.23




9
Many 1-substituted 1,4-dihydropyridines have been prepared by this method, where substitution in the 1-position has been in the form of alkyl,15'21'24-27 benzyl,28-31 alkoxymethyl,28 2-hydroxyethyl,32 or a sugar residue.22,28 A number of 1-alkyl-3-cyano-1,4-dihydropyridines have also been synthesized.33,34
Various 1,4-dihydropyridines with substituents in the 3-position have been synthesized, with X = CH=NNHC6H5, COCH3, C02H, CO2R, CON(CH3)2, CONHC6H5, 4-methyl-2thiazolyl, benzoyl, and 2-benzthiazolyl.30'33'35 However, dithionite did not produce any isolable dihydro products from pyridinium salts when X was hydrogen or alkyl.23
The introduction of a methyl group in the 2, 4, or 6position of a pyridinium salt with an electron-withdrawing substituent in the 3-position did give the expected 1,4dihydropyridine.28'36'37 This was also the case with a number of pyridinium salts with electron-withdrawing groups in both the 3- and 5-positions.38,39
The mechanism of dithionite reduction has been the
subject of some controversy. The prevailing idea shows that the reaction proceeds via a sulfinate intermediate which is stable enough in alkaline solution to have been isolated.31 Upon protonation, the salt is converted into the unstable acid which rapidly decomposes to the 1,4-dihydronicotinamide compound.
When dithionite reduction was carried out in D20, the monodeuterated dihydro form of the compound was




10
obtaned31
obtained.1 Repeated oxidation followed by dithionite reduction of this product in D20 gave the pure dideuterated derivative. A number of deuterated dihydropyridines have been prepared by this method.30'31'34
The 1,2- and 1,4-dihydropyridines are the most stable due, presumably, to the involvement of the nitrogen lone pair in the w electron system. These are the isomers with the most sp2-hybridized centers.
Some systematic work has been carried out to determine the effects of substituents on the stability of dihydropyridines. The parent 1,4-dihydropyridine was described as a very unstable substance in air.40 Electron-withdrawing substituents capable of resonance interaction in the 3and/or 5-positions were shown to stabilize dihydropyridines by extending the conjugation.41 Substitution in the 3- and 5-positions with conjugating groups results in lowered energies and transfer of electronic charge to the substituents.42 This then results in an appropriate decrease in reactivity. Substituents in the 3- and 5-positions which donate electrons by resonance have a destabilizing effect. Alkyl substitution on nitrogen has the same general effect on stability, but a glucosyl substituent on nitrogen has a tremendous stabilizing influence.43
A considerable amount of work has been done to determine the stability of NAD+ and NADH analogs in aqueous solution. The lability of the pyridinium ring has been shown to




result from nucleophilic attack in the 2- or 4position.44'45 The 1,4-dihydropyridine decomposition appears to result from protonation at C-5, and subsequent attack by water or other necleophile at C-6. A further cause of instability of some of these compounds is the lability of a carboxamido or a carbalkoxy group toward hydrolysis.46 Therefore, the three principal sources of instability in model compounds are amide and ester hydrolysis, nucleophilic attack on the pyridinium ring, and acid-catalyzed hydration of the dihydropyridine ring system. The first two of these reactions are favored under basic conditions and the last is, of course, a concern under acidic conditions.
Two esters, 1-carbomethoxymethyl-3-carbamoyl- and 1carboisopropoxymethyl-3-carbamoyl pyridinium ions underwent rapid decomposition in alkaline solutions. At pH 9.2 the methyl ester's half-life was approximately 3 minutes and the isopropyl ester's half life was about six times as long.47 Therefore, in order that not more than 10% of the compound decomposes in 24 hours, the pH must not exceed 5.7 and 6.5 respectively, if the rate of hydrolysis is proportional to the hydroxide concentration over this range.
In the case of compounds which possess strongly electron-withdrawing groups in both the 1 and 3-positions, the pyridinium ring is the first site of attack when the pH is raised. This can take the form of an unwanted decomposition but this is also often the case in a reduction of a




12
pyridinium salt to its dihydro analog, as we have seen in the case of dithionite.
The decomposition of 1,4-dihydropyridines in the presence of aqueous acids has been studied by many groups in the past.44'48-50 It is thought to involve several successive steps. The primary acid decomposition reaction is a twostep process resulting in the hydration of the 5,6-double bond.49'51 The first step, which is normally ratedetermining in aqueous media, is protonation at C-5, followed by fast nucleophilic attack at C-6. Under these conditions, the nucleophile would be hydroxide ion or water. Decomposition of 1,4-dihydropyridines can be followed by monitoring their Xmax in the UV region around 350 nm. A study of the acetic acid-catalyzed rate constants for the hydration of l-alkyl-3-carbamoyl-l,4-dihydropyridines showed a good correlation (r = 0.991) with the a* values for the 1substituents.52,53
Dihydropyridines containing more electron-donating
groups are shown to be highly acid-labile, while pyridinium ions containing electron-withdrawing groups become more susceptible to attack in basic solution.47 This therefore, results in a loss of overlap between the stable pH regions for the two isomers.
The mechanism of oxidation of 1,4-dihydropyridines is still a point of some controversy. If the factors controlling the oxidation of these compounds can be understood, then one can affect the stability of the dihydro compound in




13
order to allow its oxidation to proceed at a desirable rate. The initial view put forward by Abeles et al. was that the mechanism concerned the movement of a proton with two associated electrons i.e., a direct hydride migration.54 Later, this mechanism was modified to an initial electron transfer followed by a hydrogen radical migration. This hypothesis in turn was replaced by an electron-proton-electron transfer mechanism.55,56 However, the most recent findings indicate a dependence on the specific oxidizing agent. The conclusion in the case of enzymatic-type oxidations was that the original theory of a concerted hydride transfer was correct.57
Epilepsy and Valproic Acid
Epilepsy is among the most common of chronic
neurological disorders. Convulsive symptoms or seizures also occur during or as sequels to many of the other diseases that affect the brain. For the majority of epileptic patients long-term drug therapy represents the only practical form of treatment.58 The goal of treatment is, of course, the prevention of recurrent seizures.
One of the most widely used drugs for the treatment of epilepsy is the relatively new valproic acid. Valproic acid is a branched chain carboxylic acid. It is unique among anti-epileptic drugs in that it has neither a nitrogen atom nor a ring moiety. It has a pKa of 4.95 and a molecular weight of 144.59 It occurs as a slightly oily, clear liquid that is very soluble in both water and organic solvents.




14
Valproic acid was first demonstrated to possess anticonvulsant effects by Meunier et al..60 The sodium salt of valproic acid was first marketed as "Depakine" in France in 1967. Since that time valproate has been licensed in much of Europe. It was authorized for use in epilepsy therapy in the United States in 1978.
Valproic acid is now increasingly used in the treatment of primary generalized seizures, especially those of the absence type.61 Several controlled trials of valproate as monotherapy for specific syndromes have been reported in a 1981 review.62 In all seven of the cases reported, valproate was at least as effective as the "optimal" drug in each case. This surely indicates that valproate has a very wide spectrum of anticonvulsant activity, possibly greater than that of any other drug in use.
A larger number of trials have shown that seizure
control in absence attacks is improved when valproate is added to previous therapy.63'64 Valproate has also proven quite useful in patients failing to respond to other therapies.65 About one-third of the patients showed better than 75% reduction in seizures. Successful treatment of status epilepticus, refractory to diazepam or barbiturates has also been reported.66
Mechanism of Action of Valproic Acid
Since valproic acid was first discovered to have
clinical utility in the management of epileptic seizures, its mechanism of action has been studied rather




15
extensively. Although a number of various "facts" have been uncovered, their full implication and precise role in the mechanism of action or actions is not always known. Two main types of anticonvulsant action are suggested by experimental and clinical studies of valproate. One is a direct pharmacological effect related to the plasma and subsequent brain concentration of the drug. This is best illustrated acutely, after high doses of valproate. The other type of action is indirect and presumably relates to active metabolite concentrations remaining in the brain, or to adaptive or other changes brought about by valproate or its metabolites. Possible changes involving membranes, receptors, or enzymes as a whole have been mentioned in the literature.
Almost all of the biochemical and neurophysiological data concern the direct effects of the drug. Several hypotheses for its mechanism of action have been presented;67'68 however, there is not a clear choice among them. This is because some of our basic knowledge is deficient and because the required measurements, both in vitro and in vivo, are sensitive and difficult to perform.
The theory that has received the most attention by far is that GABAergic inhibition is enhanced through an action on the synthesis or further metabolism of GABA. However, doubts have been raised by observations showing poor correlations between increases in brain GABA and anticonvulsant action.62 This has usually been seen after




16
acute doses of the drug. Anti-seizure action is observed but no increase in GABA is detected shortly after administration of valproate. In addition, some in vivo determinations of GABA turnover indicate a reduced synthesis of GABA, even in the presence of unchanged or increased GABA levels, which seems to indicate a reduced synaptic release of GABA. This is further strengthened by the fact that no study had been able to demonstrate any increase in GABA release after valproate administration.62 However, another study showed an increase in overall brain GABA concentration as well as an elevation of glutamic acid decarboxylase (GAD) activity, following administration of valproic acid.69 The enzyme GAD catalyzes the synthesis of GABA and is thought to be the rate-limiting enzyme in determining GABA levels in the brain.70
Other studies have put forward a hypothesis that
valproate itself acts on post-synaptic receptor sites in order to mimic or enhance the inhibitory action of GABA. Unfortunately, there are no clear experimental demonstrations showing enhanced efficacy of physiologically stimulated GABAergic inhibition after systemic valproate administration.71 This is not the case after systemic administration of benzodiazapines. Valproic Acid and the BBB
In the mouse, valproic acid is rapidly absorbed after oral administration. Maximum blood serum levels occurred between 5 and 30 minutes after treatment.72 Approximately




17
35 to 45% of the dose is absorbed via this route of administration. Peak brain concentrations occur at the same time as peak serum levels but only reach about 20% of the serum level. Brain valproate concentrations of 27%69 and 10%73 of plasma levels have also been reported by different groups, but all three studies involved mice. The group finding the initial brain-to-plasma ratio of 27% did so after ip injection. Both of the other two studies involved oral administration.
The extreme rapidity with which valproate enters the brain suggests a possible contribution due to active transport62'69 because valproic acid is predominantly present in its ionized form at physiological pH. This charged species should certainly not pass the BBB by passive diffusion. Active transport of valproic acid from brain to plasma may also be functional in the mouse, as has been shown previously for the dog.74
Uptake of valproic acid into the brain of the cat when given by iv injection has also been studied.75 The plasma level for each given cat was seen to remain rather stable for the entire 90 minute experiment. The brain concentration, however, decreased rapidly to relatively low levels. The brain:plasma ratio declined from a high of 0.72 to 0.11 during the 90 minute period. Toxicity
In general, valproic acid does not suffer from major toxicity problems. However, it is often prescribed in




18
combination with other anticonvulsants. This tends to complicate the delineation of adverse reactions attributable solely to valproate. The side effects most commonly reported are not life-threatening. They include such things as nausea, vomiting, disorientation and fatigue.
A few reports have appeared in the literature for each of several more severe effects. These effects have been life-threatening, and are not always well understood. Hepatic failure and hyperammonemia are the most common causes of mortality associated with valproate treatment. The cases of hyperammonemia are most often found in patients with hepatic failure.76 A reduction in platelet count has also been seen in patients taking valproic acid. The reductions seen in one study caused one-third of the subjects to drop below the normal range.77 However, the drop was not low enough after two months to cause any bleeding abnormalities. Other reported problems associated with this drug have been red cell aplasia78 and pancreatitis.79 In all of these reported cases the adverse effects associated with valproate therapy were shown to be dose (rather than exposure length) related. These effects could be reversed either by reduction or cessation of valproate in the vast majority of cases.
Nonsteroidal Anti-inflammatory Agents
This class of compounds covers a wide variety of drugs, both in chemical structure and in pharmaceutical application. One group, the acidic nonsteroidal anti-inflammatory




19
agents, has been found to have anti-inflammatory, analgesic, and often antipyretic activity. This has led to their usage in a number of clinical applications.
Indomethacin and naproxen are two of the most potent of these compounds. They have been extensively studied in the literature, both for their own activity and as reference standards to judge newly discovered compounds of this class. They have shown activity against a multitude of inflammatory diseases, especially rheumatoid arthritis and gout. They are effective in a number of migraine headache syndromes.80,81 They are also quite useful in lowering the fever associated with a variety of pyrogens, including cancers resulting in neoplastic fever.82
Indomethacin and naproxen are both arylacetic acids.
They are off-white crystalline solids with molecular weights of 358 and 230, and pKa's of 4.5 and 4.2 respectively. In aqueous solution they exist primarily in their ionized form, and so do not effectively penetrate the BBB. After administration these compounds normally achieve a blood/brain ratio of twenty to one. Clinically, these drugs have been used most often for their anti-inflammatory and analgesic properties in the treatment of rheumatoid and other types of arthritic conditions. Mechanism of Action
Traditionally nonsteroidal anti-inflammatory compounds were thought to exert their anti-inflammatory action through peripheral mediation of prostaglandin synthesis.




20
Prostaglandins appear to sensitize pain receptors to mechanical or chemical stimulation. However, as early as 1961 there have been reports in the literature that these drugs accomplish their inhibitory role, at least in part by an action on the central nervous system.83
It now seems clear that these two compounds work principally, if not entirely, by the inhibition of prostaglandin synthesis. They prevent the production of prostaglandins in body tissues by inhibiting cyclooxygenase, the enzyme that catalyzes the formation of prostaglandin precursors from arachidonic acid.84 This is true for their anti-inflammatory action, as well as the analgesic and antipyretic effects associated with indomethacin and naproxen.
The analgesic properties of these compounds have been shown only in the presence of inflammation. This finding has led to a distinction between nonsteroidal antiinflammatory drugs and the narcotic analgesics which increase the pain threshold for both inflamed and normal tissues.85 However, it has been shown that these acidic nonsteroidal compounds do inhibit prostaglandin synthesis at the central level as well as the peripheral level. In addition, simultaneous administration of one of these drugs by systemic and by icv injection to rats with inflammation elicits a synergistic effect rather than a simple addition.86-87
This central mechanism of action is also responsible
for indomethacin and naproxen's antipyretic activity. These




21
compounds work only in the presence of fever and so do not alter body temperature in afebrile animals. It has also been shown that acidic nonsteroidal anti-inflammatory drugs produce their antipyretic action by inhibiting prostaglandin generation within the hypothalamus.88 Toxicity
The most common adverse effects associated with nonsteroidal anti-inflammatories are those of GI disturbances. These include nausea, with or without vomiting, indigestion, heartburn, and abdominal pain. Indomethacin and naproxen are also capable of reactivating latent peptic ulcers. In the case of indomethacin this can extend to intestinal lesions as well. These drugs may also cause such problems in patients with no previous history of ulcers.84 However, these effects can be minimized by administering the compounds with food or antacid.
Indomethacin and naproxen have been found to produce various CNS disturbances. Headache, dizziness, lightheadedness, fatigue, insomnia and depression are the most common of these problems. Indomethacin, however, can also effect more severe CNS reactions. Although much less frequently, such things as psychic disturbances with psychotic episodes, hallucinations, nightmares, anxiety and coma have been reported.84 Psychotic episodes as well as the GI effects are particularly likely to occur in geriatric patients.




22
There are also a number of less frequently reported, but often more severe cases of peripheral toxicities. In extreme instances these problems have even resulted in death. Various types of adverse hematologic effects have occurred in patients receiving indomethacin or naproxen. These include such conditions as thrombocytopenia and granulocytopenia89 as well as a variety of anemias.90 In addition, nonsteroidal anti-inflammatory drugs have been shown to cause nephrotoxicity, including renal failure91 and hepatic effects such as jaundice and hepatitis.84 These are especially common in patients with previously impaired renal or liver function.
The majority of toxicities associated with these
compounds have been shown to be dose dependent. Often the effects are reversible. Normally, by lowering the dose or cessation of therapy many of these problems can be addressed.
Objectives
This project is designed to develop a number of
hydroxyalkyl chemical delivery systems. These carriers are made in order to expand the applicability of the drug delivery system developed by Bodor.15 They allow for the delivery of drugs having a carboxylic acid functional group. This is accomplished by attaching the drug via an ester linkage, to a carrier which contains a pyridine ring capable of participating in a pyridinium salt-dihydropyridine redox-type reaction. The hydroxyalkyl substituent




23
can vary in chain length and position on the basic nicotinamide structure. In this way, one may be able to fine tune the design of an ideal carrier.
The second part of this work is to attach various drugs to the carriers. These drugs are chosen so that their usefullness would be enhanced by increased brain delivery. At the same time, the drugs should show reduced toxicity due to a decrease in the circulating levels of the parent compounds.
Once synthesis is completed both buffer and in vitro stability should be tested. This gives a basis for comparison in order to determine which chemical delivery system or systems should be best suited for in vivo distribution studies. The distribution of the parent drug is indicative of the success or failure of a given delivery system to increase brain penetration of the active compound. However, the activity of the drug once delivered is the final and possibly the most significant determination.




CHAPTER II
EXPERIMENTAL
Materials and Methods
Salts and nondeuterated solvents were obtained from
Fisher Scientific. All salts were of reagent grade. Solvents used for high pressure liquid chromatography were of spectral grade. Water was purified in this laboratory using a Sybron Barnstead Nanopure II deionizer and filter system. All other nondeuterated solvents were of either reagent or spectral grade. Naproxen and indomethacin were obtained from Sigma Company, in the purest grade available. The remaining chemicals and reagents and deuterated solvents were obtained through Aldrich Chemical Company, unless otherwise stated. These products were of at least reagent quality and were used without further purification except where stated.
Melting points were uncorrected and were determined using an Electrothermal melting point apparatus, equipped with the manufacturer's calibrated thermometer. Ultraviolet spectroscopy was performed on a Hewlett Packard 8451A diode array spectrophotometer. Proton nuclear magnetic resonance spectra were obtained using a Varian T60A, EM 360A, or EM 390 spectrometer. Chemical shifts are reported in parts per million units, downfield from tetramethylsilane used as an
24




25
internal standard. Elemental analyses were performed by Atlantic Microlab Inc., Atlanta, Georgia.
High pressure liquid chromatography was carried out on one of the following systems: 1) Perkin-Elmer Series 4 chromatographic pump, ISS-100 auto sampler, LCI-100 integrator, and a Kratos Model 757 UV/visible, variable wavelength detector; 2) LDC/Milton Roy constaMetric III G metering pump, Perkin-Elmer LCI-100 integrator, and a Kratos Model 757 UV/visible, variable wavelength detector; 3) Kontron System 600 pump and auto sampler, Perkin-Elmer LCI-100 integrator, and a Kratos Model 757 UV/visible, variable wavelength detector; 4) Waters Associates Model 510 pump, Kontron MSI 660 auto sampler, Hewlett Packard Model 3390A integrator, and a Kratos Model 757 UV/visible, variable wavelength detector. The column used for this chromatographic work was either a Toyasoto 25 cm, 5 um particle size, ODSC18 reversed phase column or an ASI 25 cm, 10 um particle size, C8 reversed phase column. The Toyasoto column was protected with a guard column packed with Whatman pellicular ODS-C18 media. The centrifuge used to spin down tissue homogenates was a Dynac Centrifuge having a maximum spin rate of 3000 rpm. Chromatographic samples (in vitro) were centrifuged using a Beckman Microfuge 12 capable of a 10,000 rpm spin rate.




26
Synthesis
3-Carbamoyl-l-(2-hydroxyethyl)pyridinium iodide (1)
The compound 2-iodoethanol (2.58 g, 15.0 mmol) was
dissolved in acetone (20 mL). Nicotinamide (1.83 g, 15.0 mmol) was added and the mixture was heated to reflux, where upon complete dissolution of the nicotinamide was obtained. The reflux was stopped after 8 hours and the precipitated product filtered and dried. The 2.20 g of crude product were twice recrystallized from ethanol resulting in 2.00 g of compound (1). The product was obtained in an overall yield of 45%; mp 126-127C.
UV(CH3OH): 224 and 268 nm.
1H NMR (D20) 6: 9.3-9.5 (bs, 1H, pyridine H-2); 8.99.3 [m (9.0-9.3, bd, 1H, pyridine H-6), (8.9-9.2, m, 1H, pyridine H-41; 8.1-8.5 (m, 1H, pyridine H-5); 4.7-5.0 (m, 2H, N-CH2); 4.0-4.3 (bt, 2H, O-CH2).
Analysis: (C8HIN202)
Calculated: C, 32.67; H, 3.77; I, 43.15; N, 9.52; Found: C, 32.63, H, 3.78; I, 43.07; N, 9.47. 1-Hydroxy-3-iodopropane (2)
An acetone solution of sodium iodide (30 g, 0.20 mol), and 1-chloro-3-hydroxypropane (14.2 g, 0.15 mol) was stirred at 60C for one day. The precipitated sodium chloride was removed by filtration. The solvent was removed under reduced pressure and the oily residue was vacuum distilled. The first fraction, bp 64-70C, was shown by TLC, chloroform:methanol (9:1) to contain only compound 2. The




27
purified product was obtained in a yield of 19.8 g or 71.0% overall.
1H NMR (CCI4) 6: 4.2-4.5 (m, 1H, OH); 3.4-3.8 (t, 2H, CH2-I); 3.0-3.4 (t, 2H, CH2-O); 1.7-2.3 (p, 2H, CH2). 3-Carbamoyl-l-(3-hydroxypropyl)pyridinium iodide (3)
Previously prepared 1-hydroxy-3-iodopropane, compound 2 (5.58 g, 30.0 mmol) was combined with nicotinamide (3.66 g, 30 mmol) in acetone (50 mL). The mixture was refluxed overnight, the solvent was removed, and the oily residue was stirred with ether until the formation of a yellow powder was observed. A small amount of unquaternized nicotinamide was removed by its crystallization from methanol. The solvent was removed and the product was recrystallized from a combination of ethanol:2-propanol (9:1). The crystalline needles of the final product (4.00 g) were collected by vacuum filtration under nitrogen, in an overall yield of 43%; mp 112.0-112.5C.
UV (CH3OH): 224 and 268 nm.
1H NMR (d6-DMSO) 6: 9.5-9.7 (bs, 1H, pyridine H-2);
9.2-9.5 (bd, IH, pyridine H-6); 8.9-9.2 (bd, 1H, pyridine H4); 7.9-8.7 [m(7.9-8.2, bs, 1H, NH), (8.0-8.6, m, 1H, pyridine H-5), (8.4-8.7, bs, 1H, NH)I; 4.8-5.0 (t, 2H, N-CH2);
4.5-4.8 (m, 1H, OH); 3.6-3.9 (t, 2H, O-CH2); 2.1-2.6 (p, 2H, CH2).
Analysis:(C9Hl31N202)
Calculated: C, 35.09; H, 4.25; I, 41.19; N, 9.09; Found: C, 35.05; H, 4.29; I, 41.07; N, 9.06.




28
3-Carbamoyl-l-(3-hydroxypropyl)pyridinium bromide (4)
To a flask containing 3-bromopropanol (13.9 g, 0.10
mol) and nicotinamide (12.2 g, 0.10 mol), 50 mL of acetone was added. This mixture was heated to reflux, thus allowing complete dissolution of the nicotinamide. After 6 h the mixture was allowed to cool to room temperature. Upon cooling the product crystallized and the off-white solid was filtered, washed with ether and dried under nitrogen. It was recrystallized from a mixture of 2-propanol:ethanol (3:1), and the final product weighed 12.2 g. This gave an overall yield of 46.7%; mp 124-126C.
UV (CH3OH): 222 and 268 rm.
1H NMR (d6-DMSO) 6: 9.7-9.9 (bs, 1H, pyridine H-2);
9.3-9.6 (d, 1H, pyridine H-6); 9.0-9.3 (m, 1H, pyridine H4); 8.6-8.9 (bs, 1H, NH); 8.1-8.6 (m, 2H, pyridine H-5 and NH); 4.7-5.2 (t, 2H, N-CH2); 4.3-4.6 (bs, 1H, OH); 3.4-3.7 (t, 2H, O-CH2); 2.0-2.5 (p, 2H, CH2).
Analysis:(C9Hl3BrN202)
Calculated: C, 41.40; H, 5.02; Br, 30.60; N, 10.73; Found: C,41.26; H, 5.07; Br, 30.66; N, 10.69. 3-[(2-Hydroxyethyl)carbamoylpyridine (5)
A neat mixture of 2-aminoethanol (6.1 g, 0.10 mol) and ethyl nicotinate (15.1 g, 0.10 mol) was refluxed overnight. As the mixture was cooled to room temperature, the product precipitated as a crystalline solid. It was filtered, washed with ether and then recrystallized from 2propanol/ether. The final product was collected by vacuum




29
filtration and washed with ether. The dried, white compound weighed 10.7 g, resulting in a 64.5% yield; mp 88.5-89.5C (lit. value 92C).
UV (CH3OH): 222 nm.
1H NMR (d6-DMSO) 6: 9.0-9.2 (bs, IH, pyridine H-2);
8.5-8.9 (m, 2H, pyridine H-6 and NH); 8.2-8.4 (m, 1H, pyridine H-4); 7.4-7.7 (m, 1H, pyridine H-5); 4.8-5.0 (t, 1H, OH); 3.3-3.8 (m, 4H, (CH2)2).
Analysis:(C8H10N202)
Calculated: C, 57.82; H, 6.07; N, 16.86; Found: C, 57.73; H, 6.11; N, 16.82. 3-[(2-Hydroxyethyl)carbamoyl]-l-methylpyridinium iodide (6)
Compound 5 (1.00 g, 6.02 mmol) was dissolved in acetone and refluxed overnight with methyl iodide (1.70 g, 12.0 mmol). The yellow precipitate was collected by vacuum filtration. The material was recrystallized from ethanol/ether, and 1.60 g of the cream colored crystalline plates were recovered. This resulted in an 86.5% yield; mp 111-1130C.
UV (CH3OH): 222 and 268 mu.
IH NMR (d6-DMSO) 6: 9.4-9.5 (s, 1H, pyridine H-2);
9.1-9.3 (d, 1H, pyridine H-6); 8.8-9.1 (m, 2H, NH and pyridine H-4); 8.2-8.5 (m, 1H, pyridine H-5); 4.6-4.9 (bt, 1H, OH); 4.4-4.6 (s, 3H, CH3); 3.3-3.8 (m, 4H, CH2-N and CH2-O).
Analysis:(CgHl31N202)
Calculated: C, 35.09; H, 4.25; I, 41.19; N, 9.09; Found: C, 35.14; H, 4.25; I, 41.11; N, 9.06.




30
3-[(3-Hydroxypropyl)carbamoylpyridine (7)
A mixture of ethyl nicotinate (15.12 g, 0.10 mol) and 3-aminopropanol (8.2 g, 0.11 mol) was refluxed in toluene (50 mL). An azeotrope of ethanol/toluene was removed under reduced pressure every 12 h and an equal amount of fresh toluene was added. The reaction was continued for 2 days to increase yield. The solvent was completely removed and the thick oily residue was distilled using a Kugelrohr. The first 3 mL fraction was discarded. The remaining compound was collected, bp 145-155*C at 0.1 mm pressure. This gave a light yellow, viscous oil that solidified into a wax-like substance in the freezer. The compound weighed 16.0 g giving an 88.8% yield.
UV (CH3OH): 226 nm.
1H NMR (CDCI3) 6: 9.0-9.2 (s, 1H, pyridine H-2); 8.68.8 (d, IH, pyridine H-6); 8.2-8.5 (bt, IH, NH); 8.1-8.3 (d, 1H, pyridine H-4); 7.3-7.5 (m, 1H, pyridine H-5); 4.8-5.0 (s, IH, OH); 3.4-3.9 (m, 4H, CH2-O and CH2-N); 1.6-2.1 (p, 2H, CH2).
Analysis: (C9HI2N202 -1/4 H20)
Calculated: C, 58.52; H, 6.82; N, 15.16; Found: C, 58.46; H, 6.94; N, 15.12. 3-[(3-Hydroxypropyl)carbamoyll-l-methylpyridinium iodide (8)
Compound 7 (1.0 g, 3.1 mmol) was quaternized using methyl iodide (2 mL, 30 mmol) in acetone (40 mL) at reflux. The reaction was continued overnight and the acetone and excess methyl iodide were removed under reduced




31
pressure. The thick oily residue was stirred for several hours with anhydrous ether. The ether was decanted and a fresh portion was again added with continued stirring. This resulted in a yellow-brown solid powder which was filtered and dried. The crude material was crystallized from 2-propanol. The yellow crystalline product was filtered, washed with ether and air dried. It weighed 1.65 g. The product was obtained in overall yield of 94.6%; mp 112-1130C.
UV (CH3OH): 222 and 268 nm.
1H NMR (d6-DMSO) 6: 9.3-9.5 (bs, 1H, pyridine H-2);
8.8-9.3 (m, 3H, pyridine H-6 and H-4, and NH); 4.3-4.6 (s, 3H, CH3); 3.2-3.7 (m, 4H, CH2-N and CH2-O); 1.5-2.0 (p, 2H, CH2).
Analysis:(CloH15IN202)
Calculated: C, 37.29; H, 4.69; I, 39.39; N, 8.69; Found: C, 37.35; H, 4.70; I, 39.27, N, 8.68. 2-Propylpentanoyl chloride (9)
Valproic acid (4.32 g, 30.0 mmol) was stirred at 00C, while thionyl chloride (3.60 g, 30.0 mmol) was added dropwise. When the addition was completed, the mixture was allowed to come to room temperature. The flask was then warmed in a water bath for 30 min at 50*C. Dry benzene (2 x 50 mL) was added and then removed under reduced pressure. This compound was used in subsequent steps without further purification.




32
2-Propylpentanoic acid, ester with 2-iodoethanol (10)
The product from the previous reaction, compound 9, was stirred at 0C while 2-iodoethanol (5.16 g, 30.0 mmol) was slowly added. The neat mixture was heated to 100C for 10 min, in a water bath. The reaction mixture was stirred for an additional 10 min at room temperature. It was then dissolved in ether (50 mL) and washed successively with water (30 mL), 5% aqueous sodium hydroxide (2 x 30 mL), and again with water (2 x 30 mL). The organic layer was separated and dried with sodium sulfate. The solvent was removed under reduced pressure giving 6.0 g of a light yellow liquid product resulting in an overall yield of 67% starting from valproic acid. Test with silver nitrate gave a bright yellow precipitate.
UV (CH3OH): 216 and 250 nm.
1H NMR (neat) 6: 4.2-4.5 (t, 2H, CH2-O); 3.1-3.5 (t, 2H, CH2-I); 2.1-2.7 (m, 1H, CH); 1.1-1.9 (m, 8H, propyl CH2is); 0.6-1.1 (m, 6H, CH3).
Analysis:(C10HlgIO2)
Calculated: C, 40.28; H, 6.42; I, 42.56; Found: C, 40.37; H, 6.43; I, 42.45.
(+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 3-bromopropanol (11)
A large excess of thionyl chloride (9.6 g, 80 mmol), freshly distilled from linseed oil, was added at once to naproxen (2.3 g, 10 mmol). The addition was carried out with stirring at 00C. The mixture was slowly warmed to reflux and then heating was continued for an additional 30




33
min. Dry benzene (3 x 30 mL) was added and then removed under reduced pressure to remove the excess thionyl chloride. The residue was dissolved in a minimum amount of dry benzene and 3-bromo-l-propanol (1.4 g, 10 mmol) was slowly added with stirring at room temperature. The solution was refluxed for 40 min and then allowed to cool to room temperature. Transparent crystals formed in the dark brown residue. These were vacuum filtered under nitrogen and washed with cold anhydrous ether. The compound was dried under nitrogen and recrystallized from acetonitrile. The compound weighed 2.1 g, giving an overall yield of 60% from naproxen; mp 52-54C.
UV (CH3OH): 226, 264 and 332 nm.
1H NMR (CDCI3) 6: 7.0-7.7 (m, 6H, naphthalene
protons); 4.0-4.3 (t, 2H, O-CH2); 3.6-3.9 (m, 4H, CH and 0CH3); 3.1-3.3 (t, 2H, Br-CH2); 1.8-2.2 (p, 2H, CH2); 1.4-1.6 (d, 3H, CH3).
Analysis:(Cl7H19BrO3)
Calculated: C, 58.13; H, 5.45; Found: C, 58,09; H, 5.53.
(+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 2-bromoethanol (12)
Naproxen (2.3 g, 10 mmol) was esterified with 2-bromoethanol (1.4 g, 11 mmol), using DCC (2.3 g, 11 mmol) and DMAP (120 mg, 1.0 mmol). The reaction was carried out in acetonitrile at room temperature. The reaction mixture was stirred for 24 h, at which time the precipitated DCU was filtered. The DCU weighed 2.3 g after it was dried. The




34
solvent was removed from the filtrate under reduced pressure, giving an oily residue. This was stirred with anhydrous ether until most of the product dissolved. The remaining solid was filtered and discarded. The ether was removed under reduced pressure giving a dry white solid. A small amount of DCU was still present by NMR. The compound was recrystallized from 2-propanol. The white crystalline product was vacuum filtered and dried in a vacuum desiccator. The final product weighed 3.00 g giving an overall yield of 89.0%; mp 61-63C.
UV (CH3OH): 224, 264 and 332 nm.
IH NMR (CDCl3) 6: 7.0-7.9 (m, 6H, naphthalene protons); 4.2-4.6 (t, 2H, CH2-O); 3.6-4.1 (m, 4H, CH and CH3-O); 3.23.6 (t, 2H, CH2-Br); 1.4-1.8 (d, 3H, CH3).
Analysis:(Cl6Hl7BrO3)
Calculated: C, 56.99; H, 5.08; Br, 23.70; Found: C, 57.09; H, 5.09; Br, 23.60.
(+)-6-Methoxy-c-methyl-2-naphthaleneacetic acid, ester with 2-iodoethanol (13)
Naproxen (4.6 g, 20 mmol) was esterified with 2-iodoethanol (3.4 g, 20 mmol), using DCC (4.5 g, 22 mmol) and DMAP (240 mg, 2.0 mmol) in a minimum amount of acetonitrile (260 mL). The reaction was stirred overnight at room temperature. The precipitated DCU (4.4 g) was filtered, washed with cold acetonitrile (10 mL) and air dried. The reaction solvent was removed under reduced pressure, yielding an oily residue. The product was dissolved in ether (300 mL) and decanted from any undissolved material. The ether was dried




35
with magnesium sulfate and then filtered. The ether was removed under reduced pressure giving a thick, light colored oil which crystallized on standing at room temperature. The light yellow product was recrystallized twice from 2-propanol giving a nearly white final product. The final compound weighed 5.0 g, giving a 65% overall yield; mp 51-52*C.
UV (CH3OH): 224, 264 and 332 nm.
1H NMR (CDCI3) 6: 7.0-7.8 (m, 6H, naphthalene
protons); 4.1-4.5 (t, 2H, CH2-O); 3.6-4.1 (m, 4H, CH and CH3-O); 3.0-3.4 (t, 2H, CH2-I); 1.4-1.7 (d, 3H, CH3).
Analysis:(Cl6Hl7IO3)
Calculated: C, 50.02; H, 4.46; I, 33.03; Found: C, 50.14; H, 4.46; I, 32.96.
(+)-6-methoxy-a-methyl-2-napthaleneacetic acid, ester with chloromethanol (14)
Previously prepared chloromethylchlorosulfate92 (1.9 g, 12 mmol) in methylene chloride (5 mL) was added dropwise to a mixture of naproxen (2.3 g, 10 mmol), sodium bicarbonate (3.2 g, 3.8 mmol), tetrabutylammonium hydrogen sulfate (0.68 g, 2.0 mmol) in water and methylene chloride (10 mL each). The biphasic mixture was stirred at room temperature for one hour after the addition of the chloromethylchlorosulfate. The reaction was followed by TLC, hexanes:chloroform (3:2); Rf=0.35 for the product. The two phases were separated and the aqueous layer was extracted with methylene chloride (10 mL). The organic layers were combined and washed with water (20 mL), and dried over magnesium sulfate. The methylene chloride was removed under reduced pressure giving a light




36
yellow oil which solidified on standing. The white solid was dissolved in hot petroleum ether and filtered to remove any unreacted naproxen. The solvent was removed under reduced pressure and the solid product was recrystallized from 2-propanol. The final product weighed 1.9 g, giving a 67% yield. Compound 14 was recovered as beautiful white crystalline plates; mp 67-68C.
UV (CH3OH): 224, 266 and 332 nm.
1H NMR (CDCI3) 6: 7.0-7.8 (m, 6H, naphthalene
protons); 5.5-5.8 (s, 2H, CH2); 3.7-4.1 (m, 4H, CH and CH30); 1.4-1.7 (d, 3H, CH3).
Analysis:(Cl5Hl5Cl03)
Calculated: C, 64.64; H, 5.42; Cl, 12.72; Found: C, 64.75; H, 5.45; Cl, 12.65.
(+)-6-Methoxy-s-methyl-2-naphthaleneacetic acid, ester with N-(2-hydroxyethyl)nicotinamide (15)
Naproxen (2.30 g, 10.0 mmol) was coupled with compound
5 (1.71 g, 10.0 mmol) using DCC (2.30 g, 11.0 mmol) and DMAP (122 mg, 1.00 mmol) in acetonitrile (150 mL). The reaction was stirred at room temperature for 48 h. The precipitated DCU was filtered, rinsed with acetonitrile and dried to a weight of 2.3 g. The solvent was removed under reduced pressure and the residual clear oil was stirred with anhydrous ether. The resulting white solid was vacuum filtered, washed with ether and air dried. The crude product weighed 2.80 g. The compound was recrystallized from 2-propanol. The final product was filtered, washed with 0.5% aqueous sodium bicarbonate, water, and finally




37
with ether. The compound was dried in a desiccator over P205. The recrystallized material weighed 2.40 g resulting in an overall yield of 63.4%; mp 79-82C.
UV (CH3OH): 226, 264 and 332 nm.
1H NMR (CDCl3) 6: 8.8-9.0 (bs, 1H, pyridine H-2); 8.58.8 (d, 1H, pyridine H-6); 7.0-8.0 (m, 9H, pyridine H-4 and H-5, naphthalene protons, and NH); 4.1-4.4 (t, 2H, CH2-O);
3.6-4.0 (m, 4H, CH and CH3-O); 3.4-3.7 (t, 2H, CH2-N); 1.41.7 (d, 3H, CH3).
Analysis:(C22H22N204)
Calculated: C, 69.83; H, 5.86; N, 7.40; Found: C, 69.92; H, 5.88; N, 7.39.
(+)-6-Methoxy-a-methyi-2-naphthalene etic acid, ester with N-(3-hydroxypropyl)nicotinamide (16)"
A reaction of naproxen (2.30 g, 10.0 mmol) and compound 7 (1.80 g, 10.0 mmol) was carried out in acetonitrile, using DCC (2.26 g, 11.0 mmol) and DMAP (122 mg, 1.00 mmol) as a coupling agent. After 24 hours the reaction mixture was filtered and the collected DCU was washed with acetonitrile and air dried. The DCU weighed 2.1 g. The solvent was removed from the filtrate under reduced pressure, giving an oily residue. By TLC the oil showed a small amount of a DCU adduct but almost none of the unreacted acid. Column chromatography was done using Mallinckrodt silica gel (100-200 mesh, 60A special) and a mobile phase of chloroform: tetrahydrofuran (4:1). The product was a clear oil which crystallized overnight in a vacuum desiccator. The white




38
crystalline material was analytically pure. The compound weighed 2.60 g, giving a 66.7% yield; mp 72-75*C.
UV (CH3OH): 224, 264 and 332 nm.
1H NMR (CDCI3) 6: 8.9-9.1 (bs, 1H, pyridine H-2); 8.58.8 (d, 1H, pyridine H-6); 7.9-8.2 (d, 1H, pyridine H-4);
7.5-7.8 (m, 3H, naphthalene protons); 7.0-7.5 (m, 5H, pyridine H-5, NH and naphthalene protons); 4.0-4.3 (t, 2H, CH20); 3.6-4.0 (m, 4H, CH3-0 and CH); 3.1-3.5 (q, 2H, CH2-N);
1.6-2.1 (p, 2H, CH2); 1.4-1.7 (d, 3H, CH3).
Analysis:(C23H24N204)
Calculated: C, 70.39; H, 6.16; N, 7.14; Found: C, 70.46; H, 6.18; N, 7.09. 1-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-3Zgcetic acid, ester with N-(2-hydroxyethyl)nicotinamide (17)"
A reaction of indomethacin (1.79 g, 5.00 mmol) and
compound 5 (0.830 g, 5.00 mmol) was carried out, using DCC (1.10 g, 5.50 mmol) as the coupling agent and acetonitrile as the solvent. The first two reactants were dissolved completely and the solution was then cooled to 00C. The DCC was added and the mixture was stirred overnight. The reaction was allowed to continue for 48 h. The precipitated DCU (1.2 g) was removed by vacuum filtration. The solvent was removed from the filtrate under reduced pressure leaving an oily residue. The product was solidified by stirring with anhydrous ether. It was filtered, air dried and recrystallized from ethanol/ether. The final product was vacuum filtered, washed with ether, and air dried. The product weighed 1.65 g, giving a 65.2% yield; mp 123-25*C.




39
UV (CH3OH): 222 and 320 nm.
IH NMR (CDCI3) 6: 8.7-8.9 (bs, 1H, pyridine H-2); 8.68.8 (d, 1H, pyridine H-6); 7.7-8.0 (bd, 1H, pyridine H-4); 7.2-7.7 (m, 5H, phenyl protons and pyridine H-5); 6.4-7.0 (m, 4H, indole protons and NH); 4.2-4.4 (t, 2H, CH2-O); 3.53.9 (m, 7H, CH3-O, CH2-N, CH2); 2.2-2.4 (s, 3H, CH3).
Analysis: (C27H24CIN30512 H20)
Calculated: C, 62.98; H, 4.89; N, 8.16; Found: C, 63.27; H, 4.91; N, 8.49. l-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid, ester with N-(3-hydroxypropyl)nicotinamide (18)
A reaction of indomethacin (3.58 g, 10.0 mmol) and
compound 7 (1.80 g, 10.0 mmol) with DCU (2.26 g, 11.0 mmol) and DMAP (0.12 g, 1.0 mmol) was carried out in acetonitrile at room temperature. The reaction mixture was stirred for 24 h and then filtered to remove the precipitated DCU. The solvent was removed under reduced pressure after 2.3 g of DCU were recovered (94% of theoretical amount). The product was a white solid which was recrystallized from 2propanol/ether, using the mixed solvent technique. The compound was filtered, washed with ether and air dried. The NMR showed some traces of DCU. Therefore, the material was recrystallized from 2-propanol only. The final product was filtered, washed with ether and dried in a vacuum desiccator. The compound weighed 3.60 g giving an overall yield of 69.2%; mp 122-123C.




40
UV (CH3OH): 222 and 320 nm.
1H NMR (CDCl3) 6: 8.9-9.2 (bs, 1H, pyridine H-2); 8.68.8 (bd, 1H, pyridine H-6); 7.9-8.3 (bd, 1H, pyridine H-4); 7.2-7.9 (m, 6H, phenyl protons, pyridine H-5, and NH); 6.57.1 (m, 3H, indole protons); 4.1-4.5 (t, 2H, CH2-O); 3.2-4.0 (m, 7H, CH3-O, CH2-N, and CH2-CO); 2.2-2.5 (s, 3H, CH3).
Analysis:(C28H26C1N305)
Calculated: C, 64.68; H, 5.04; Cl, 6.82; N, 8.08; Found: C, 64.55; H, 5.10; Cl, 6.79; N, 8.03. 3-Carbamoyl-l-(2-hydroxyethyl)pyridinium iodide, ester with 2-propylpentanoic acid (19)
A mixture of nicotinamide (1.22 g, 10.0 mmol) and compound 10 (3.28 g, 11.0 mmol) was dissolved in dimethylformamide (50 mL). The solution was heated to reflux for
3 h and then cooled. The solvent was removed under reduced pressure, and the brown oily residue was stirred with ether (60 mL) for 30 min, giving a yellow powder. The ether was decanted and a fresh portion of ether (50 mL) was added. The crude product was vacuum filtered under nitrogen. It was then recrystallized from 2-propanol/ether, giving 3.50 g of light, cream colored crystalline plates. The final product was obtained in 83.3% yield; mp 113-114C.
UV (CH3OH): 222 and 268 nm.
1H NMR (d6-DMSO) 6: 9.5-9.7 (bs, 1H, pyridine H-2);
9.2-9.5 (bd, 1H, pyridine H-6); 8.9-9.2 (bd, 1H, pyridine H4); 8.0-8.7 [m(8.0-8.3, bs, 1H, NH), (8.1-8.5, m, 1H, pyridine H-5), (8.4-8.7, bs, 1H, NH)]; 4.9-5.3 (m, 2H, CH2-N);
4.3-4.8 (m, 2H, CH2-O); 2.0-2.8 (m, 1H, CH); 0.9-1.7 (m, 8H, propyl CH21s); 0.4-1.1 (m, 6H, CH3)-




41
Analysis:(CI6H25IN203)
Calculated: C, 45.73; H, 6.00; I, 30.19; N, 6.66; Found: C, 45.70; H, 6.04; I, 30.10; N, 6.65. 3-Carbamoyl-l-hydroxymethypyridinium chloride, ester with
(+)-6-methoxy-a-methyl-2-naphthaleneacetic acid (20)
Nicotinamide (0.24 g, 2.0 mmol) was dissolved in acetone. The chloromethylester of naproxen, compound 14 (0.56 g, 2.0 mmol + 1% excess), was then added and the solution was stirred at reflux for 48 h. The precipitated white product was vacuum filtered, washed with ether and dried. The dried compound weighed 410 mg. The mother liquor was reduced to dryness and recrystallized from 2-propanol giving a second crop of crystals weighing 80 mg. The yield was
0.49 g or 61% overall; mp 227-229*C.
UV (CH3OH): 226, 268 and 332 nm.
1H NMR (d6-DMSO/CD3OD) 6: 9.5-9.6 (bs, IH, pyridine H2); 9.0-9.3 (m, 2H, pyridine H-6 and H-4); 8.1-8.4 (m, 1H, pyridine H-5); 7.6-7.9 (m, 4H, naphthalene protons and NH);
7.1-7.5 (m, 4H, naphthalene protons and NH); 6.5-6.6 (s, 2H, CH2); 3.9-4.3 (q, 1H, CH); 3.8-3.9 (s, 3H, CH3-O); 1.5-1.6 (d, 3H, CH3).
Analysis:(C21H21CiN204)
Calculated: C, 62.92; H, 5.28; Cl, 8.84; N, 6.99; Found: C, 62.73; H, 5.30; Cl, 8.94; N, 6.94. 3-Carbamoyl-l-(3-hydroxypropyl)pyridinium bromide, ester with (+)-6-methoxy-a-methyl-2-naphthaleneacetic acid (21)
Naproxen (2.5 g, 11 mmol) and compound 4 (2.6 g, 10 mmol) were dissolved in a minimum amount of dimethyl-




42
formamide (200 mL), to which DMAP (130 mg, 1.1 mmol) and DCC (2.3 g, 11 mmol) were added. The solution was stirred at room temperature for 2 days. The precipitated DCU was vacuum filtered, washed and air dried. The collected DCU weighed 2 g, which represented 80% of the theoretical yield. The dimethylformamide was removed from the filtrate under reduced pressure. This gave an oily residue which solidified on standing at room temperature. The solid was powdered and washed well with anhydrous ether. The offwhite powder was filtered, washed with an additional portion of ether and air dried. It was recrystallized from ethanol, and upon partial cooling the mother liquor was decanted from the reddish brown solid that first appeared. Additional cooling of the mother liquor, along with brief scratching, gave a large quantity of fluffy tan crystals. These were once again recrystallized from ethanol. The final product was filtered, washed with ether and dried in a vacuum desiccator. The compound weighed 4.0 g resulting in an 85% yield; mp 152-154C.
UV (CH3OH): 224, 266 and 332 nm.
1H NMR (d6-DMSO) 6: 9.5-9.8 (bs, 1H, pyridine H-2);
9.1-9.4 (bd, 1H, pyridine H-6); 8.9-9.2 (bd, 1H, pyridine H4); 8.5-8.8 (bs, 1H, NH); 8.0-8.5 (m, 2H, pyridine H-5 and NH); 7.0-8.0 (m, 6H, naphthalene protons); 4.6-5.0 (t, 2H, CH2-N); 4.0-4.4 (t, 2H, CH2-O); 3.6-4.0 (m, 4H, CH and CH30); 2.1-2.6 (m, 2H, CH2); 1.3-1.6 (d, 3H, CH3).




43
Analysis:(C23H25BrN204)
Calculated: C, 58.36; H, 5.32; Br, 16.88; N, 5.92; Found: C, 58.23; H, 5.34; Br, 16.95; N, 5.86. 3-[(2-Hydroxyethyl)carbamoyl]-l-methylpyridinium iodide, ester with (+)-6-methoxy-a-methyl-2-naphthaleneacetic acid
(22)
The quaternization of the naproxen ester, compound 15 (1.0 g, 2.6 mmol), was carried out using methyl iodide (2.3 g, 16 mmol) in acetone (45 mL). The solution was heated to reflux for 20 h. Methyl iodide (1.1 g, 8.0 mmol) was again added to the reaction flask. The precipitated product was filtered after an additional 4 h of reaction time. The offwhite powder was dried. The material weighed 2.2 g and was found to be analytically pure without recrystallization. The solvent was removed from the acetone filtrate and the residue was solidified with anhydrous ether. The resulting dark yellow powder was dissolved in water and washed with ether (4 x 30 mL). The water was then removed under vacuum giving 0.2 g of a lighter yellow powder, although still much more highly colored than the precipitated product. The overall yield of the reaction was 93%; mp 169-170*C.
UV (CH3OH): 226, 266 and 332 nm.
1H NMR (d6-DMSO) 6: 9.3-9.5 (bs, 1H, pyridine H-2);
9.0-9.3 (m, 2H, pyridine H-6 and NH); 8.7-9.0 (bd, 1H, pyridine H-4); 8.1-8.5 (m, 1H, pyridine H-5); 7.0-7.9 (m, 6H, naphthalene protons); 4.3-4.5 (s, 3H, CH3-N); 4.1-4.4 (+, 2H, CH2-O); 3.4-4.1 (m, 6H, CH, CH3-O and CH2-N); 1.4-1.7 (d, 3H, CH3).




44
Analysis:(C23H25IN204)
Calculated: C, 53.09; H, 4.84; I, 24.39; N, 5.38; Found: C, 52.98; H, 4.85; I, 24.29; N, 5.34. 3-[(3-Hydroxypropyl)carbamoyl]-l-methylpyridinium iodide, ester with (+)-6-methoxy-a-methyl-2-naphthaleneacetic acid
(23)
The ester of naproxen, compound 16 (2.3 g, 5.9 mmol), was dissolved in acetone. Methyl iodide (4.0 g, 29 mmol) was added and the solution was heated to reflux for 48 h. The reaction mixture was cooled and then taken to an oily foaming residue upon removal of the solvent under reduced pressure. It was then dried further on a vacuum pump. The product was dissolved in acetone (2 x 25 mL) and each time the solvent was removed under reduced pressure. The first process yielded a foam and the second gave a powdery yellow solid. This was then completely dried using a vacuum pump. The compound weighed 2.9 g which resulted in a 93% yield; mp 100-1020C. This product required no further purification.
UV (CH3OH): 224, 266 and 332 nm.
1H NMR (d6-DMSO) 6: 9.4-9.5 (s, 1H, pyridine H-2);
8.8-9.3 (m, 3H, pyridine H-6 and H-4 and NH); 8.2-8.4 (m, 1H, pyridine H-5); 7.1-8.0 (m, 6H, naphthalene protons);
4.4-4.5 (s, 3H, CH3-N); 4.1-4.3 (t, 2H, CH2-O); 3.8-4.1 (m,
4H, CH and CH3-O); 3.2-3.5 (t, 2H, CH2-N); 1.7-2.1 (p, 2H, CH2); 1.4-1.7 (d, 3H, CH3).




45
Analysis:(C24H27IN204)
Calculated: C, 53.94; H, 5.09; I, 23.75; N, 5.24; Found: C, 54.00; H, 5.13; I, 23.81; N, 5.22. 3-[(2-Hydroxyethyl)carbamoyl]-l-methylpyridinium iodide, ester with 1-(p-chlorobenzoyl)-5-methoxy-2-methylindole-3acetic acid (24)
The quaternization of compound 17 (0.50 g, 1.0 mmol) was carried out in acetone, using methyl iodide (1.7 g, 12 mmol). The reaction was refluxed overnight. The solvent was removed under reduced pressure and a yellow solid was obtained. The product was recrystallized using ethanol and a very small amount of ether. Small mold-like crystals were obtained which were light yellow in color. The reaction gave 0.43 g or a 66% yield of the purified material; mp 1781790C.
UV (CH3OH): 220 and 320 nm.
1H NMR (d6-DMSO) 6: 9.3-9.5 (bs, 1H, pyridine H-2);
9.1-9.3 (m, 2H, pyridine H-6 and NH); 8.8-9.0 (bd, 1H, pyridine H-4); 8.1-8.4 (m, IH, pyridine H-5); 7.6-7.8 (s, 4H, phenyl protons); 6.6-7.2 (m, 3H, indole protons); 4.4-4.6 (s, 3H, CH3-N); 4.2-4.5 (t, 2H, CH2-O); 3.5-4.0 (m, 7H, CH30, CH2-CO and CH2-N); 2.2-2.3 (s, 3H, CH3).
Analysis:(C28H27C1IN305)
Calculated: C, 51.91; H, 4.20; Cl, 5.47; I, 19.59; N, 6.48; Found: C, 51.80; H, 4.24; Cl, 5.41; I, 19.46; N, 6.42.




46
3-[(3-Hydroxypropyl)carbamoyl]-l-methylpyridinium iodide, ester with l-(p-chlorobenzoyl)-5-methoxy-2-methylindole-3acetic acid (25)
The ester of indomethacin, compound 18 (3.1 g, 6.0 mmol), was dissolved in a minimum of acetone. A twofold excess of methyl iodide (2.5 g, 18 mmol) was added to the solution which was then heated to reflux. The reaction was continued for 24 h. Approximately one-half of the reaction solvent was removed under reduced pressure. The precipitated, light yellow product was filtered, washed with ether and dried in a vacuum desiccator. The compound weighed 3.9 g giving a 97% yield; mp 168-169*C. The compound was analytically pure without recrystallization.
UV (CH3OH): 222 and 320 nm.
1H NMR (d6-DMSO) 6: 9.3-9.5 (bs, 1H, pyridine H-2); 8.8-9.2 (m, 3H, pyridine H-6 and H-4 and NH); 8.2-8.4 (m, 1H, pyridine H-5); 7.5-7.8 (s, 4H, phenyl protons); 6.6-7.2 (m, 3H, indole protons); 4.3-4.5 (s, 3H, CH3-N); 4.0-4.3 (t, 2H, CH2-O); 3.6-4.0 (bs, 5H, CH3-O and CH2-CO); 3.2-3.5 (t, 2H, CH2-N); 2.2-2.4 (s, 3H, CH3); 1.7-2.1 (p, 2H, CH2).
Analysis: (C29H29C1IN305)
Calculated: C, 52.62; H, 4.42; Cl, 5.36; I, 19.17; N, 6.35; Found: C, 52.55; H, 4.46; Cl, 5.29; I, 19.19; N, 6.34. 2-Propylpentanoic acid, ester with 1,4-dihydro-l-(2-hydroxyethyl)nicotinamide (26)
Compound 19 (420 mg, 1.0 mmol) was dissolved in icecold degassed, deionized water (50 mL). Sodium bicarbonate (370 mg, 4.0 mmol) and sodium dithionite (700 mg, 4.0 mmol) were added with stirring. Nitrogen gas was bubbled through




47
the yellow solution about 30 min. The reaction mixture was extracted with ether (6 x 25 mL) until no additional color was transferred to the organic layer. The combined ether extracts were washed with H20 (50 mL) and dried over magnesium sulfate. The solution was decanted away from the drying agent and the solvent was removed under reduced pressure. To the oily residue, ether was added and removed (10 x 5 mL) on a vacuum pump. A foam was obtained which returned to an oil upon exposure to the atmosphere.
UV (CH3OH): 216 and 354 nm.
1H NMR (CDCI3) 6: 6.9-7.1 (bs, IH, dihydropyridine H2); 5.6-6.0 (m, 3H, NH2 and dihydropyridine H-6); 4.5-4.9 (m, 1H, dihydropyridine H-5); 4.0-4.4 (t, 2H, CH2-O); 3.23.6 (m, 2H, CH2-N); 2.9-3.2 (bs, 2H, dihydropyridine H-4);
2.1-2.6 (m, 1H, CH); 0.9-1.7 (m, 8H, propyl CH2's); 0.5-1.1 (m, 6H, CH3).
Analysis: (CI6H26N203"H20)
Calculated: C, 61.51; H, 9.04; N, 8.97 Found: C, 61.96; H, 8.69; N, 8.53;
C, 61.49; H, 8.49; N, 8.80.
(+)-6-Methoxy-G-methyl-2-naphthaleneacetic acid, ester with 1,4-dihydro-l-hydroxymethylnicotinamide (27)
The quaternary ester of naproxen, compound 20 (200 mg, 0.5 mM), was dissolved in a mixture of acetonitrile:2-propanol, 6:1. An excess of 1-benzyl-l,2-dihydronicotinamide94 (200 mg, 0.8 mmol) was added and the reaction was allowed to proceed for one hour at room temperature. The reaction was carried out under nitrogen gas. When the reaction was




48
complete, the solvent was removed under reduced pressure giving an oily orange foam. This was dissolved in methylene chloride and then filtered to remove the quaternary side product. The solvent was again removed giving an oily residue. The product showed traces of contamination with quaternary material, by NMR. The oil was dissolved in chloroform and passed down a short column of neutral alumina. The appropriate fraction was collected and the solvent removed under reduced pressure. The oily residue was triturated with anhydrous ether. The ether was removed using a vacuum pump. This gave a hygroscopic yellow foam as the product. However, NMR showed the final compound had partially hydrolyzed.
UV (CH3OH): 228, 264, 334 and 346 nm.
1H NMR (CDCI3) 6: 7.0-7.8 (m, 7H, naphthalene protons
and dihydropyridine H-2); 5.5-5.9 (m, 1H, dihydropyridine H6); 5.1-5.5 (bs, 2H, NH2); 4.6-5.0 (m, 1H, dihydropyridine H-5); 4.4-4.6 (s, 2H, CH2); 3.8-4.0 (s, 3H, CH3-O); 3.3-3.7
(q, 1H, CH); 3.0-3.3 (bs, 2H, dihydropyridine H-4); 1.4-1.7 (d, 3H, CH3).
Analysis:(C21H22N204"H20)
Calculated: C, 65.61; H, 6.29; N, 7.29; Found: C, 65.50; H, 6.68; N, 7.99.
(+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 1,4-dihydro-l-(3-hydroxypropyl)nicotinamide (28)
The quaternary ester, compound 21 (470 mg, 1.0 mmol), was dissolved in degassed water (70 mL) and sodium dithionite (520 mg, 3.0 mmol) and sodium bicarbonate (420 mg, 5.0




49
mmol) were added at once. A layer of ether was added and the reaction was allowed to proceed under a nitrogen atmosphere for 60 min. The reaction mixture was extracted with methylene chloride (4 x 30 mL). The extracts were combined and washed with water (50 mL), and then dried over magnesium sulfate. The solvent was removed under reduced pressure giving an oily foam. The product was repeatedly redissolved and dried on a vacuum pump until the compound was obtained as a light yellow solid foam. The final product weighed 180 mg resulting in a 46% yield; mp 42-46'C.
UV (CH3OH): 224, 264, 334 and 356 nm.
1H NMR (CDCI3) 6: 6.8-7.8 (m, 7H, naphthalene protons and dihydropyridine H-2); 5.2-5.5 (m, 3H, dihydropyridine H6 and NH2); 4.3-4.7 (m, 1H, dihydropyridine H-5); 3.7-4.3 (m, 6H, CH2-O, CH3-O and CH); 2.7-3.2 (m, 4H, dihydropyridine H-4 and CH2-N); 1.3-2.0 (m, 5H, CH2 and CH3).
Analysis:(C23H26N204"H20)
Calculated: C, 66.97; H, 6.84; N, 6.79; Found: C, 67.02; H, 6.57; N, 6.74.
(+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 1,4-dihydro-N-(2-hydroxyethyl)-l-methylnicotinamide (29)
The quaternary salt compound 22 (780 mg, 1.5 mmol) was dissolved in degassed, deionized water (200 mL) and acetonitrile (10 mL). Sodium dithionite (780 mg, 4.5 mmol) and sodium bicarbonate (630 mg, 7.5 mmol) were combined, and added to the solution at room temperature. The reaction was continued for 1 h, while nitrogen gas was slowly bubbled




50
through the solution. The partially precipitated product was extracted repeatedly with ether (8 x 30 mL). The extracts were combined, washed with water (25 mL) and dried over magnesium sulfate. The drying agent was filtered and the solvent was removed from the filtrate under reduced pressure. The oily residue was dissolved in methylene chloride (3 x 5 mL) and removed under reduced pressure. The resulting foam was rinsed with anhydrous ether (3 mL) and the solvent was removed under vacuum. The final product weighed 390 mg, giving a 66% yield. The hygroscopic solid foam was stored under nitrogen at -100*C.
UV (CH30H): 224, 264, 334 and 358 nm.
1H NMR (CDCl3) 6: 7.0-7.9 (m, 6H, naphthalene
protons); 6.7-7.0 (bs, 1H, dihydropyridine H-2); 5.1-5.7 (m, 2H, dihydropyridine H-6 and NH); 4.2-4.6 (m, 1H, dihydropyridine H-5); 4.0-4.3 (t, 2H, CH2-O); 3.6-4.0 (m, 4H, CH3-O and CH); 3.3-3.7 (t, 2H, CH2-N); 2.6-3.0 (s, 3H, CH3-N); 2.4-2.6 (bs, 2H, dihydropyridine H-4); 1.3-1.7 (d, 3H, CH3).
Analysis:(C23H26N204"H20)
Calculated: C, 66.97; H, 6.84; N, 6.79; Found: C, 67.11; H, 6.69; N, 6.63.
(+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 1,4-dihydro-N-(3-hydroxypropyl)-l-methylnicotinamide (30)
Compound 23 (530 mg, 1.0 mmol) was dissolved in degassed water (150 mL). This was washed with ether (50 mL), separated and the organic phase was discarded. The aqueous layer was transferred to a flask and sodium dithionite (520




51
mg, 3.0 mmol) and sodium bicarbonate (420 mg, 5.0 mmol) were added at once. The reaction was run under a nitrogen atmosphere, at room temperature. The reaction was continued for 30 min, and then extracted with methylene chloride (5 x 40 mL) until no more yellow color transferred into the organic layer. The methylene chloride was combined and was back extracted with water (40 mL) and then dried with magnesium sulfate. The crude product was an orange oil. This showed traces of the unreduced starting material by both TLC and NMR. The oil was dissolved in CHC13 and quickly passed down a short column of neutral alumina. Chloroform was also used as the eluting solvent. The fraction which contained the dihydro compound was collected and the solvent removed under reduced pressure. The product was then dried on a vacuum pump yielding a yellow hygroscopic foam. The product weighed 300 mg giving an overall yield of 73%.
UV (CH3OH): 224, 264, 334 and 354 nm.
1H NMR (CDCI3) 6: 7.0-7.9 (m, 6H, naphthalene
protons); 6.9-7.1 (bs, IH, dihydropyridine H-2); 5.3-5.8 (m, 2H, dihydropyridine H-6 and NH); 4.4-4.8 (m, 1H, dihydropyridine H-5); 4.0-4.3 (t, 2H, CH2-O); 3.6-4.0 (m, 4H, CH3-O and CH); 3.1-3.4 (t, 2H, CH2-N); 2.8-3.1 (bs, 2H, dihydropyridine H-4); 2.8-3.0 (s, 3H, CH3-N); 1.4-2.0 (m, 5H, CH2 and CH3).




52
Analysis:(C24H28N204" H20)
Calculated: C, 69.04; H, 7.00; N, 6.71; Found: C, 69.40; H, 7.23; N, 6.39. 1-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid, ester with 1,4-dihydro-N-(2-hydroxyethyl)-l-methylnicotinamide (31)
The indomethacin quaternary carrier, compound 24 (140 mg, 0.22 mmol), was dissolved in a minimum amount of water:acetonitrile (8:2). The water had been bubbled with nitrogen for 20 min previous to its use. Sodium bicarbonate (91 mg, 1.1 mmol) and sodium dithionite (110 mg, 0.65 mmol) were added to the solution while stirring at 00C. The solution was then allowed to warm to room temperature. The reaction was continued for about 1 h. Some of the product had precipitated during the reaction. This was dissolved in ethyl ether. The water layer was extracted several times with ether until no more yellow color transferred to the organic layer. The ether portions were combined and dried with magnesium sulfate, filtered and the ether was removed under reduced pressure. The resulting oil was dissolved in acetone and the solvent was removed (2 x 10 mL) under reduced pressure to form a dry foam. The final product weighed 92 mg. The yield was 82%; mp 60-65*C.
UV (CH3OH): 220 and 332 nm.
1H NMR (CDCI3) 6: 7.3-7.9 (q, 4H, phenyl protons); 6.5-7.1 (m, 4H, indole protons and dihydropyridine H-2);
5.5-5.8 (m, 1H, dihydropyridine H-6); 5.2-5.4 (bs, 1H, NH); 4.4-4.8 (m, 1H, dihydropyridine H-5); 4.1-4.4 (t, 2H, CH2-




53
0); 3.8-4.0 (s, 3H, CH3-O); 3.6-3.8 (s, 2H, CH2-CO); 3.4-3.7
(m, 2H, CH2-N); 2.8-3.0 (s, 3H, CH3-N); 2.6-2.9 (bs, 2H, dihydropyridine H-4); 2.3-2.5 (s, 3H, CH3).
Analysis:(C28H28C1N305"2H20)
Calculated: C, 60.27; H, 5.78; Cl, 6.35; N, 7.53; Found: C, 60.14; H, 5.73; Cl, 6.11; N, 7.24. l-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid, ester with 1,4-dihydro-N-(3-hydroxypropyl)-l-methylnicotinamide (32)
The quaternary ester, compound 25 (660 mg, 1.0 mmol), was dissolved in degassed, deionized water (80 mL) and washed once with ether (50 mL). The organic layer was separated and acetonitrile (5 mL) was added to the aqueous phase in order to dissolve any remaining starting material. A mixture of sodium dithionite (520 mg, 3.0 mmol) and sodium bicarbonate (420 mg, 5.0 mmol) was added to the solution. Ether (30 mL) was added and the reaction was allowed to proceed for 60 min under a nitrogen atmosphere. The reaction mixture was extracted with methylene chloride (4 x 40 mL), the extracts combined, and then washed with degassed water (40 mL). The organic phase was separated and dried using magnesium sulfate. The drying agent was gravity filtered and the solvent removed under reduced pressure. The dark orange oil was redissolved in methylene chloride (3 x 5 mL) and the solvent removed under reduced pressure, and then dried using a vacuum pump. The product was a yellow-orange solid foam. This crude material showed traces of quaternary starting material which were removed by dissolving in




54
chloroform and passing it down a short column of neutral alumina. Chloroform was also used as the eluting solvent. The purified product was recovered, the solvent removed and the resulting oil was dissolved in methylene chloride. Again the solvent was removed, under reduced pressure and the resulting foam was triturated with anhydrous ether. This was removed using a vacuum pump, giving 300 mg of a dry, light yellow foam. The final product gave an overall yield of 56%; mp 52-56*C.
UV (CH3OH): 212 and 332 nm.
1H NMR (CDCI3) 6: 7.3-7.8 (q, 4H, phenyl protons); 6.5-7.1 (m, 4H, indole protons and dihydropyridine H-2);
5.2-5.9 (m, 2H, dihydropyridine H-6 and NH); 4.5-4.8 (m, 1H, dihydropyridine H-5); 3.9-4.3 (t, 2H, CH2-O); 3.7-3.9 (s, 3H, CH3-O); 3.5-3.8 (s, 2H, CH2-CO); 3.2-3.5 (t, 2H, CH2-N); 2.9-3.2 (bs, 2H, dihydropyridine H-4); 2.8-3.0 (s, 3H, CH3N); 2.2-2.5 (s, 3H, CH3); 1.6-2.2 (p, 2H, CH2).
Analysis:(C29H30Cl'3/4H2ON305)
Calculated: C, 63.38; H, 5.78; Cl, 6.45; N, 7.65; Found: C, 63.42; H, 5.78; Cl, 6.53; N, 7.65.
High Pressure Liquid Chromatography Systems
Mobile phase systems were developed in order to carry out all in vitro and in vivo analyses. The quaternary compounds (20-25) were analyzed utilizing a mobile phase that was one-half organic and one-half aqueous. The organic portion was entirely acetonitrile. However, the aqueous portion was composed of both monobasic potassium phosphate




55
(50 mM) and water. The ratio was adjusted for each quaternary compound in order to have its retention time between 5 and 6 min. The amount of potassium phosphate used varied between 15 and 50% of the entire mobile phase.
Naproxen and indomethacin could be analyzed for using these same systems. Their retention times were not greatly affected by the changes in buffer concentration, as long as the acetonitrile:aqueous ratio remained constant. This allowed both the quaternary compound and the free parent drug to be analyzed for simultaneously.
The dihydropyridine compounds (27-32) required analysis using alternate mobile phase systems. The compounds were chromatographed using various ratios of acetonitrile: water. Ratios as high as 80% organic were used in order to obtain retention times in the 4 to 6 min range. All of these analyses, involving the quaternary compounds, the free parent drugs and the dihydropyridine-CDS were carried out using a Toyasota column (see Materials and Methods).
The hydrophilic nature of the hydroxyalkylpyridinium carriers did not allow them to be retained in any of the previously mentioned systems. A reversed phase C-8 column (see Materials and Methods) was chosen for the work involving the pyridinium carriers. A mobile phase of acetonitrile, water, and monobasic potassium phosphate (50 mM); (40:50:10) was used. This gave retention times of 4 to
6 min. The flow rate used for all of the HPLC systems was
1.0 mL/min.




56
Chemical Stability
Stability of Quaternary Drug-Carrier Compounds (20-25) in pH
7.4 Phosphate Buffer
Phosphate buffer (pH 7.4) was equilibrated in a water bath at 37*C. A 5 x 10-3M stock solution was prepared for each compound, using DMSO as the solvent. Ten microliters of the stock solution were added for each milliliter of buffer used in a given experiment. This resulted in a 5 x 10-5M solution. Aliquots of 100 pL were withdrawn at various time intervals and pipetted into 400 uL portions of ice-cold acetonitrile. The sample were centrifuged for 3 min at 10,000 rpm. The rate of ester hydrolysis was determined by measuring the disappearance of the quaternary compound and the appearance of the free drug after hydrolysis. This was accomplished using high pressure liquid chromatography.
Stability of Dihydropyridine CDS Compounds (27-32) in pH 7.4 Phosphate Buffer
Phosphate buffer was equilibrated in a water bath at 37*C. A 5 x 10-3M stock solution in DMSO was freshly prepared for each compound before use. Ten microliters of the stock solution were added per milliliter of buffer used in each experiment. This resulted in a 5 x 10-5M solution. Aliquots of 100 UL were withdrawn at various time points and pipetted into 400 pL portions of ice-cold acetonitrile. The samples were centrifuged for 3 min at 10,000 rpm. The rate of disappearance was measured using high pressure liquid chromatography. Appearance of the quaternary form of each




57
compound and the free, parent drug was monitored using a second HPLC system.
In Vitro Studies
Stability of Quaternary Drug-Carrier Compounds (20-25) in 100% Whole Human Blood and 100% Whole Rat Blood
Blood was withdrawn from a volunteer shortly before beginning each experiment. It was placed in heparinized tubes and stored on ice until needed. The blood was then incubated at 37*C. A 5 x 10-3M stock solution in DMSO was prepared for each compound before use. Ten microliters of the stock solution were added per milliliter of blood used in a given experiment. Aliquots of 100 UL were withdrawn at various time intervals and pipetted into 400 UL of ice-cold acetonitrile containing 5% DMSO by volume. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was used to determine the rate of ester hydrolysis by high pressure liquid chromatography. The same procedure was used for fresh rat blood that was withdrawn into a heparinized syringe via heart puncture. The blood was collected just prior to the start of each experiment.
Stability of Dihydropyridine CDS Compounds (27-32) in 100% Whole Human Blood and 100% Whole Rat Blood
Blood was withdrawn from a volunteer shortly before beginning each experiment. It was placed in heparinized tubes and stored on ice until needed. The blood was incubated at 37*C. A 5 x 10-3M stock solution in DMSO was freshly prepared. Ten microliters of the stock solution




58
were added for each milliliter of blood used in a given experiment. Aliquots of 200 uL were withdrawn at various time points and pipetted into 800 pL of ice-cold acetonitrile containing 5% DMSO by volume. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was divided into two equal portions. The first was used to follow the rate of disappearance of the dihydropyridine compound. Later the second portion that had been kept at 00C was used to measure the appearance of the quaternary form of the compound and the free, parent drug. These samples were analyzed using high pressure liquid chromatography. The dihydro compound was monitored using one system and a second HPLC system was required to follow the quaternary compound and the parent drug. The same experimental procedure was used for fresh rat blood that was collected via heart puncture into a heparinized syringe. The blood was collected just prior to beginning each experiment.
Stability of Quaternary Drug-Carrier Compounds (20-25) in 20% Rat Brain Homogenate and 20% Rat Kidney Homogenate
One gram of freshly obtained rat brain was homogenized with 4 mL of phosphate buffer (pH 7.4). The homogenate was centrifuged at 3,000 rpm for 5 min. The supernatant was removed and incubated in a water bath at 370C. A 5 x 10-3M stock solution in DMSO was prepared for each compound before its use. Ten microliters of the stock solution were added per milliliter of homogenate used in a particular experiment. Aliquots of 100 UL were withdrawn at various time




59
intervals and pipetted into 400 UL of ice-cold acetonitrile. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was used to measure the rate of ester hydrolysis by high pressure liquid chromatography. The same procedure was used for freshly obtained rat kidney.
Stability of Dihydropyridine CDS Compounds (27-32) in 20% Rat Brain Homogenate
One gram of freshly obtained rat brain was homogenized with 4 mL of phosphate buffer (pH 7.4). The homogenate was centrifuged at 3,000 rpm for 5 min. The supernatant was removed and incubated in a water bath at 370C. A 5 x 10-3M stock solution in DMSO was prepared for each compound just before its use. Ten microliters of the stock solution was added for each milliliter of homogenate used in a given experiment. Aliquots of 200 UL were withdrawn at various time points and pipetted into 800 uL of ice-cold acetonitrile. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was divided into two equal portions. The first was used to measure the rate of disappearance of the dihydropyridine compound. This was done by high pressure liquid chromatography. The second portion which had been kept at 00C was used to monitor the appearance of the quaternary compound and the parent drug. This was done via a second HPLC system.




60
In Vivo Studies
Preliminary In Vivo Distribution of Compound 31 in Brain and Blood
Male Sprague-Dawley rats weighing 220 to 270 g were anesthetized with InovarO (40 UL) intramuscularly. The dihydropyridine (20 mg/kg, 67 mg/mL) in DMSO was given by intrajugular injection using an infusion pump fitted with a glass syringe and a 27 gauge butterfly needle. The animals were sacrificed at various time intervals by decapitation. Trunk blood was collected into heparinized tubes and the brains were removed. Brains were homogenized in 1 mL of water and then extracted with 4 mL of cold acetonitrile. One milliliter of blood was also extracted with 4 mL of acetonitrile. The samples were placed in a freezer for 2 h in order to separate the organic layer. This was used for analysis by HPLC after being filtered through 0.45 Um Millipore filters.
In Vivo Distribution of Compound 28
Male Sprague-Dawley rats weighing 255 to 280 g were placed in a restraining cage. The dihydropyridine (25 mg/kg, 50 mg/mL) in DMSO was given by tail vein injection. The animals were sacrificed by decapitation. Trunk blood was collected into heparinized tubes. The animals were opened along the midline, and brain, lung, liver, kidney, testis, and fat tissues (1.0 g each) were weighed and then quickly frozen using dry ice. The tissues were homogenized with 1.0 mL of water, using a ground glass pestle and tube. The samples were extracted with 4.0 mL of




61
acetonitrile by homogenizing again. The separation of the organic layer was expedited by adding 1.0 mL of saturated aqueous sodium chloride solution and homogenizing the mixture once more. The blood samples (1 mL) were extracted and separated in the same manner; however, vigorous shaking was used rather than homogenization.
All samples were placed in a freezer for 2 h in order to complete the separation of the organic phase. The acetonitrile layer was removed and filtered through a 0.45 Um Millipore filter. The samples were then analyzed using high pressure liquid chromatography. The quaternary compound and the free parent drug were both monitored by ultraviolet detection at 264 nm. The mobile phase consisted of acetonitrile, water and aqueous, monobasic potassium phosphate (50 mM); (50:15:35). An attempt was made to detect the dihydropyridine compound using a mobile phase of acetonitrile and water (80:20). However, no measurable amounts remained by the first time point (30 min).
Controls were run by injecting DMSO (0.5 mL/kg) as a blank. The animals were sacrificed and tissues were collected in an identical fashion. Standard curves were constructed by adding known amounts of both the quaternary compound 21 and the parent drug, naproxen to blood and tissue blanks. In this way the experimental results could be quantitated and the extent of recovery examined.




62
In Vivo Distribution of Compound 29 vs an Equimolar Dose of Naproxen
Male Sprague-Dawley rats weighing 300*40 g were placed in a retaining cage. The dihydropyridine (25 mg/kg, 50 mg/mL) or an equimolar amount of naproxen (15 mg/kg, 29 mg/mL) was given by tail vein injection. In both cases DMSO was used as a vehicle. The animals were sacrificed by decapitation. Trunk blood was collected into heparinized tubes. The brains were removed from the skull. The body was opened along the midline, and lung, liver, kidney, testis and fat tissues (1.0 g each) were removed, weighed and immediately frozen using dry ice. The tissues were homogenized with 1.0 mL of water, using a ground glass pestle and tube. The samples were homogenized again after adding 4.0 mL of ice-cold acetonitrile. An additional 1.0 mL of saturated aqueous sodium chloride solution was added and the samples were homogenized once more. The blood samples (1.0 mL), which had been kept at 0C, were prepared in a similar manner; however, vigorous shaking was used rather than homogenization. Controls were run by injecting DMSO (0.5 mL/kg) as a blank. The animals were sacrificed and the tissues were collected and prepared in the same manner. Standard curves were constructed by adding known amounts of both the quaternary compound 22 and the parent drug, naproxen to blood and tissue homogenate blanks. In this way the experimental results could be quantitated and complete recovery verified.




63
All samples were placed in a freezer for 2 h in order to completely separate the organic phase. The acetonitrile layer was removed and filtered through a 0.45 um Millipore filter. The samples were then analyzed using high pressure liquid chromatography. The quaternary compound and the free parent drug were monitored using ultraviolet detection at 264 nm. The mobile phase consisted of acetonitrile, water, and aqueous monobasic potassium phosphate (50 mM); (50:25:25). The dihydropyridine compound was detected by ultraviolet absorption at 356 nm, using a mobile phase of acetonitrile and water (75:25). An attempt was also made to detect levels of the free quaternized carrier compound 6 in both the brain and the blood. Extraction was also tested using sample blanks as in the case of the quaternary compound and the free drug. The hydrophilic carrier was analyzed using a C-8 reverse phase column (see Materials and Methods) and a mobile phase consisting of acetonitrile, water, and monobasic potassium phosphate (50 mM); (40:50:10). The compound was detected by UV at 268 nm. Antipyretic Activity of Compound 29 vs an Equimolar Dose of Naproxen
Male Sprague-Dawley rats weighing 235-285 g were
used. Rectal temperature was measured twice a day by means of a telethermometer (Yellow Springs Instrument Co., Model 46 TUC) equipped with a thermistor probe (Model 402). Normal temperatures were recorded over a three day period. These temperatures were then averaged to determine the prefever value for each animal.




64
Pyrexia was induced by a subcutaneous injection of 15
mL/kg of a 15% (w/v) suspension of active dry yeast in sterile saline (0.9%; Abbot Laboratories). Five days later the animals were placed in a retaining cage. The dihydropyridine (0.36 mg/kg, 0.90 mg/mL) or an equimolar dose of naproxen (0.20 mg/kg, 0.50 mg/mL) was given by tail vein injection. In each case DMSO was used as a vehicle. Controls were run by injecting DMSO (0.40 mL/kg) as a blank. Rectal temperatures were measured 1 h before, and at one-hour intervals for 4 h after drug administration using the same thermistor probe.




CHAPTER III
RESULTS AND DISCUSSION
Synthesis
Pyridinium Carriers
The preparation of various pyridinium salt carriers was the first step in the synthetic pathway. Two basic types of carriers were made. One group consisted of compounds that contained a hydroxyalkyl functional group attached to the pyridine ring nitrogen of nicotinamide. The other was comprised of carriers in which the hydroxyalkyl group was attached to the amide nitrogen of nicotinamide.
The first group was easily prepared using nicotinamide and a haloalcohol. The quaternization reactions were run in acetone in order to facilitate the precipitation of the product as it was formed. Carriers were made using 2-iodoethanol, 3-iodopropanol and 3-bromopropanol. The 3-iodopropanol was prepared using the Finkelstein reaction. The products were purified by recrystallization when necessary.
The second type of carrier required a different synthetic approach. Initially, nicotinoyl chloride hydrochloride was prepared, purified, and dried. It was then stirred with a 50% excess of 2-aminoethanol or 3-aminopropanol. Unfortunately in both cases the hydrochloride salt of the aminoalcohol was formed.
65




66
A second attempt using the ethyl ester of nicotinic
acid with an equimolar amount of 2-aminoethanol was successful. This reaction was run neat, at reflux temperature. The product crystallized upon cooling to room temperature and was easily purified. However, the 3-aminopropanol derivative was isolated as a thick oil, which required vacuum distillation using a Kugelrohr in order to obtain a pure product. These two carriers were then quaternized to form the corresponding pyridinium salts. These reactions were run using methyl iodide as the quaternizing agent and acetone as the solvent. The products were light yellow crystalline solids. These carriers could also be used before quaternization in order to prepare various esters. However, the 1-substituted pyridinium carriers were not used in esterification reactions directly, in most cases. This was due to their very low solubility in the majority of organic solvents.
Drug Esters
The three drugs that were used in this work were valproic acid, indomethacin, and naproxen. These compounds were attached to the previously mentioned carriers via an ester linkage. In some cases the drugs were attached directly to the hydroxyalkyl carrier while in others they were esterified using a haloalcohol. The resulting haloalkyl ester was then used to quaternize the nicotinamide carrier.
The N-substituted (amide nitrogen) hydroxyalkyl carriers were reacted with each of the three drugs.




67
Valproic acid was reacted with the hydroxyethyl carrier (compound 5) using DCC as the coupling agent. However, the desired product was not obtained. This may have been due to the reaction stopping after the initial attack of the acid on DCC. The stability of this adduct as well as the nucleophilicity of the acid and then the alcohol are all determining factors in this reaction going to completion.
In the case of indomethacin this coupling was successfully carried out using DCC. The reaction was run in acetonitrile at room temperature. However, in the case of naproxen this reaction was again shown to be less than ideal. In this case the product was not the desired ester. In order for this reaction to be successful a catalytic amount of p-toluenesulfonic acid or DMAP was required. Under these conditions the product could be obtained in a 63% yield.
In regard to the hydroxypropyl carrier (compound 7),
the reaction with either indomethacin or naproxen was not of value when DCC alone was used as the coupling agent. This problem also occurred when the acid catalyst zinc chloride or p-toluenesulfonic acid was used in combination with DCC. These esters were successfully synthesized using DCC along with DMAP which acts as a nucleophilic catalyst.
The preparation of haloalkyl esters of these three compounds was also attempted using a variety of techniques. The acid chloride of each drug was seen as a possible first step in the synthesis of these esters.




68
Valproic acid and naproxen were each treated with an excess of thionyl chloride in order to prepare their corresponding acid chlorides. Unfortunately, indomethacin decomposed under these conditions.
The acid chlorides of valproic acid and naproxen could be used to achieve our goal. The compounds were stirred with one of several haloalcohols. The valproic acid chloride was reacted with 2-iodoethanol or 3-iodopropanol. In each case the reaction was run neat and the product was purified by removing the excess thionyl chloride.
The esterification of naproxen could also be accomplished using a similar method. In this case esters were made using 2-iodoethanol, 2-bromoethanol and 3-bromopropanol. In each case the acid chloride was reacted neat and the reaction mixture was warmed slightly. The products were isolated as crystalline solids which could be recrystallized from 2-propanol. These esters could also be prepared directly from naproxen and a haloalcohol. This improved synthetic method was carried out using a combination of DCC and DMAP. This gave a one-step esterification which resulted in greater or equal yields of the desired product.
One additional naproxen ester was prepared by an altogether different method. Naproxen was dissolved in a biphasic mixture of water and methylene chloride, along with sodium bicarbonate and the phase-transfer catalyst tetrabutylammonium hydrogen sulfate. Chloromethylchlorosulfate92 was added dropwise with stirring. The mixture was then kept




69
at room temperature for one hour after the addition was complete. The phases were separated and the product was isolated from the organic layer. The white solid was dissolved in petroleum ether and filtered to remove any unreacted naproxen. The solvent was removed under reduced pressure and the compound was recrystallized from 2-propanol. The final product was recovered as beautiful white crystalline plates.
The final ester of naproxen was prepared using a carrier already in its quaternized form. Therefore, it is discussed in the following section. Drug-Pyridinium Carrier Combinations
The drug-pyridinium carrier combinations were each
prepared by way of a multistep synthesis. In most cases, the final reaction involved the quaternization of the pyridine ring nitrogen. This was done in order to avoid the problems associated with the pyridinium salts' low solubility in most organic solvents. The sole exception to this procedure was the reaction of naproxen with the 1-(3hydroxypropyl)pyridinium bromide carrier (compound 4) to form the corresponding ester (compound 21).
Initially, attempts were made to synthesize this product by reacting nicotinamide with a haloalkyl ester of naproxen. This method gave an elimination, resulting in the isolation of the corresponding hydrogen halide salt of nicotinamide. This type of reaction was attempted with the 2iodoethyl, 2-bromoethyl, and 3-bromopropyl esters. In each




70
case the reaction with nicotinamide led to elimination rather than substitution. The reaction was originally run in acetone at reflux temperature. However, replacing acetone with acetonitrile, nitromethane, or dimethylformamide did not seem to influence the course of the reaction. The temperature was reduced as low as 40C, but this still did not give the desired quaternary compound. Therefore, esterification of the hydroxyalkyl quaternary carrier was pursued. The successful procedure involved dissolving the quaternary carrier (compound 4) in a minimum amount of dimethylformamide. Naproxen was then coupled to the pyridinium salt in the presence of DCC and DMAP. The majority of the DCU that was formed during the reaction precipitated and was filtered away. The solvent was removed under reduced pressure, and the resulting solid was twice recrystallized from ethanol. This gave the final product as a fluffy, tan crystalline solid.
The syntheses of the remaining drug-pyridinium carrier combinations required quaternization of the pyridine nitrogen as the final step. In the two additional cases in which naproxen was linked via the 1-position of the nicotinamide carrier, the appropriate haloalkyl ester of naproxen was directly substituted onto nicotinamide. The 2-iodoethyl ester of valproic acid was used in the synthesis of compound 19. The reaction was run in dimethylformamide at reflux temperature. The product was isolated and recrystallized from 2-propanol/ether using the mixed solvents technique.




71
The other synthesis of this general type involved the chloromethyl ester of naproxen. This reaction was run using acetone at reflux temperature. This allowed for the majority of the product to precipitate as it was formed during the reaction.
The N-substituted (amide nitrogen) esters of indomethacin and naproxen were each quaternized using an excess of methyl iodide. This was a straightforward reaction using acetone as the solvent in each case. The product either partially or completely precipitated from the reaction solution. It was then further purified by recrystallization if necessary. Typical proton NMR spectra are seen in Figures 3-1 and 3-2.
Dihydropyridine-Chemical Delivery Systems
The reduction of the quaternary drug-carrier combinations was effected in order to form their corresponding 1,4dihydropyridines. In the majority of cases the dihydropyridines (compounds 26, 28-32) were synthesized using sodium dithionite as the reducing agent (see Figures 3-3 and 34). The reactions were run in degassed, deionized ice-cold water. The water was degassed to remove any dissolved oxygen in order to prevent the reoxydation of the dihydro compound once formed in solution. The reactions were bubbled with nitrogen gas to again displace any oxygen in the reaction atmosphere. Sodium dithionite was used in excess as a mild reducing agent. Sodium bicarbonate was used to raise the pH of the reaction mixture in order to




Figure 3-1. 1H NMR spectrum of compound 22.




Figure 3-2. H NMR spectrum of compound 24.




110H(CH2)2OH 4" CHOCH DCCpTs N C 2 6H CH3CN
j NH(CH2)20R +* CH33 ACETONE > NH(CH220R IH3
~H (CHj2OR NaHCO /Na SNOq(CH2OR CH3 CH3
Figure 3-3. Synthetic reaction sequence for compound 29, where R
represents the naproxen ester.




HCH22OH 3
Q r 6--ODcc/CH3CN N C2-0
+I
+H2 20R + CH3'[ ACETONE y H(.OR
r2A 2
3
Figure 3-4. Synthetic reaction sequence for compound 31, where R tJeen
represents the indomethacin ester.




76
stabilize the product as it was formed. Once the reaction was complete the product was isolated by extracting the aqueous reaction mixture with either methylene chloride or diethyl ether. The organic extracts were combined and dried with magnesium sulfate. The solvent was removed and the dihydro compound was dried using a vacuum pump. The product was obtained as a yellow solid foam. These hygroscopic compounds were somewhat difficult to handle and elemental analysis always showed the presence of some water. When these analyses were run in duplicate the second measurement always showed additional water present. This indicated that these compounds drew water quite readily from the humid atmosphere. A typical proton NMR spectrum is seen in Figure 3-5.
The one exception to this procedure was made in the
preparation of the dihydro form of the 1-hydroxymethylnicotinamide ester of naproxen (compound 27). The compound's quaternary form (compound 20) is quite unstable in basic medium. Therefore, it was impossible to reduce this compound in the presence of aqueous sodium bicarbonate. An alternate method of reduction was found that could successfully reduce the quaternary salt to the desired product. A more active (less stable) 1,2-dihydronicotinamide94 was used as the reducing agent. This redox reaction was run in a mixture of acetonitrile and 2-propanol at room temperature. The solvent was removed and the residue was taken up in methylene chloride in order to remove the quaternary side




Figure 3-5. H NMR spectrum of compound 29.




78
product by filtration. The solvent was removed under reduced pressure and the resulting oil was dried on a vacuum pump. An attempt was made to remove the final traces of quaternary salt contamination by dissolving the product in chloroform and quickly passing it down a short column of neutral alumina. This, however, led to decomposition of the product.
Chemical Stability
The testing of the chemical stability of the indomethacin and naproxen quaternary drug-carrier combinations and dihydropyridine-CDSs was the first step in evaluating the likelihood of their success in the delivery of the parent drug to the brain. Each of these compounds' rate of disappearance was measured at 37*C in phosphate buffer at pH
7.4. This work was carried out to determine the relative stability of each compound, and to determine if possible the various reaction products.
The pyridinium salts were quite stable with the expected exception of the 1-hydroxymethylnicotinamide ester of naproxen. This soft quaternary salt was by far the least stable of the group. This compound had a half-life of 8.6 minutes as compared to one or more days for the other quaternary compounds.
The general stability trends followed the anticipated patterns for both the dihydropyridines and the pyridinium salts. As the ester linkage was moved further away from the electron-withdrawing pyridinium ring it became more stable




79
to chemical hydrolysis. This tendency resulted in a corresponding decrease in the rate of disappearance of these compounds. In the case of the dihydropyridine compounds however, the order of stability was reversed. The electronwithdrawing effect of the ester moiety has a stabilizing influence on the 1,4-dihydropyridine ring. Therefore, as the ester linkage was moved further away from the ring the rate of disappearance of these compounds increased. These effects were also reflected in the half-lives of all of the compounds seen in Table 3-1.
The products of decomposition were also studied when possible. In the case of the quaternary compounds, hydrolysis of the ester bond was the principal instability. The free drug was detected in increasing amounts as the pyridinium salt concentration decreased following a pseudo-first order rate of disappearance.
The dihydropyridine compounds produced a more complicated variety of reaction products. The corresponding quaternary compound could be detected in most cases even when the buffer was degassed before use. This could have been due to a small amount of atmospheric oxygen interacting at the buffer surface or redissolving into the buffer as each experiment proceeded. The exception again involved the 1hydroxymethylnicotinamide ester of naproxen (compound 20). This compound was not found in detectable amounts due to it being much less stable than its corresponding dihydropyridine form. Therefore, if the oxidation product was formed




80
Table 3-1. Half-life of disappearance and correlation
coefficient for each dihydropyridine and quaternary pyridinium compound, in pH 7.4 phosphate buffer. Half-life (min); or h
hour. Results are from one run per compound,
and six to fifteen sample measurements per run.
Compounds Dihydro Quat
31 and 24 32 26ha (0.97) (0.92)
32 and 25 24 35ha (0.99) (0.998)
29 and 22 16 58ha (0.97) (0.75)
30 and 23 9.0 550ha (0.99) (0.88)
28 and 21 52 750ha (0.998) (0.98)
27 and 20 5.1h 8.6 (0.98) (0.998)
aparent drug was detectable, but not within first hour of each experiment.




81
it would decompose at a much greater rate thus preventing its detection.
The parent drug was also detectable in small amounts. This phenomenon could be due to ester hydrolysis either before or after oxidation of each dihydro compound. The experimental results indicated that the quaternary compound was not the primary source of this freed drug. This hypothesis is based on the relatively low levels of the quaternary compounds which were present, as well as their great stability in most cases.
The major decomposition product in pH 7.4 buffer is an addition product. In an aqueous medium this reaction normally results in the addition of water across the 5,6double bond of the dihydropyridine ring. The mechanism involves protonation at the 5-position, which is usually rate determining. This step is then followed by nucleophilic attack at the 6-position by hydroxide ion or water. However, the addition product could not be detected using either type of high pressure liquid chromatography system as attempted. This result was most likely due to the compound's increased polarity and/or its transient nature.
In Vitro Studies
Stability in Human and Rat Blood
The stability of a drug in the blood of the patient is always an important factor. In order to deliver drugs to the brain using the chemical delivery system described, this factor becomes one of the criteria for success. A given




82
dihydropyridine compound must penetrate into the brain before oxidation takes place. Then, once oxidation has occurred, the resulting quaternary ester must also be reasonably stable in the bloodstream. This stability would allow for its excretion before significant amounts of the parent drug are released in the periphery. In order to measure the stability of these compounds, each of the naproxen and indomethacin pyridinium esters and its corresponding dihydro-CDS was tested. These experiments followed the disappearance of each compound in both human and rat blood under in vitro conditions (Table 3-2).
The most surprising finding of this particular work was the quaternary compounds' relative stability in rat blood when compared to human blood. The only exception was the hydroxypropyl carrier ester of indomethacin (compound 25), for which the rate of disappearance in rat blood was twice that found for human blood. Normally, one would assume this type of greater enzymatic activity in rat blood to be the rule rather than the exception. This phenomenon was apparently due to some type of specificity involving one or more of the enzymes contained in human blood.
The dihydropyridine compounds, however, behaved in a more expected manner. These compounds for the most part were more stable in human blood than in rat blood. So much so, that each dihydro-CDS in which the drug ester was attached through the amide nitrogen of nicotinamide was even more stable in human blood than in pH 7.4 phosphate




83
Table 3-2. Half-life of disappearance and correlation
coefficient for each dihydropyridine and
quaternary pyridinium compound, in 100% whole
rat and 100% whole human blood. Half-life
(min); or h = hour. Results are from one to
three runs per compound and six to fifteen
sample measurements per run.
Rat Human Rat Human Compound Dihydro Dihydro Quat Quat
31 and 24 5.2 60 84 15
(0.98) (0.96) (0.97) (0.995)
32 and 25 24 27 18 37
(0.999) (0.98) (0.999) (0.993)
29 and 22 5.5 61 15ha 5.4
(0.98) (0.98) (0.91) (0.995)
30 and 23 5.1 52 6.2ha 1.8
(0.99) (0.99) (0.95) (0.97)
28 and 21 10 30 stableb 3.2
(0.97) (0.995) (0.998)
27 and 20 3.7h 2.7h 0.4 0.2
(0.93) (0.96) (0.98) (0.999) aExperiment was followed for < one half-life. bCompound very stable; parent drug was not detectable over course of experiment.




84
buffer. The only dihydropyridine compound that was not more stable in human blood compared to rat blood was the dihydro form of the 1-hydroxymethylnicotinamide ester of naproxen (compound 27).
This dihydro compound was also unusual in that it was more stable than its corresponding pyridinium salt in both human and rat blood. Normally, the dihydropyridines were less stable than their quaternary forms in rat blood, but more stable than the corresponding quaternary forms in human blood. The only exception to both of these patterns involved the hydroxypropyl carrier ester of indomethacin (compounds 25 and 32).
The major products of decomposition for each of the
compounds tested were the same in both human and rat blood. In each case, hydrolysis of the naproxen or indomethacin ester was the major enzymatic process involved. This was true for the pyridinium salts and at least in the case of rat blood the dihydropyridines. In human blood it was not always possible to determine if the dihydropyridines were oxidized before hydrolysis, due to the relative instability of most of the quaternary compounds in this medium. However, in each case except for the 1-hydroxymethylnicotinamide ester of naproxen, the quaternary ester compound (oxidation product) was detected along with the parent drug. This result occurred after the addition of each of the dihydropyridines to either human or rat blood, except as previously mentioned.




85
Stability in 20% Brain and Kidney Homogenate
In order to perform the desired testing in brain or kidney homogenate fresh rat tissue was used. The fresh organ was obtained and homogenized in pH 7.4 phosphate buffer (4 mL/g). The homogenate was centrifuged at 3,000 rpm, and the supernatant was removed and stored briefly on ice. When the experiment was begun the homogenate was incubated in a water bath at 37C.
The dihydropyridine compounds were generally much less stable in brain homogenate than were their corresponding pyridinium salts (Figure 3-6 and Table 3-3). The exception again involved the 1-hydroxymethylnicotinamide ester of naproxen (compounds 20 and 27). In this case the quaternary compound was rather unstable in brain or kidney homogenate due to the ester linkage being quite close to the electronwithdrawing positively charged ring. However, the five additional pyridinium compounds were rather stable in both brain or kidney homogenate. The stability in brain homogenate was normally somewhat lower than that measured in kidney homogenate. This finding was obtained even in the case of compound 20, but not in the case of the Nsubstituted (amide nitrogen) hydroxyethylnicotinamide carrier ester of naproxen (compound 22).
The dihydropyridine compounds were readily oxidized in rat brain homogenate. The half-life of disappearance for these delivery systems was generally one to five minutes. The only exception again came in the form of compound 27.




86
10.0
5.0
1.0
0 0.5
0
0.1 L- 1
5 10 15 20 25 30 35 40 45 Time(min) Figure 3-6. In vitro results of the dihydropyridine
(compound 31) A in 20% rat brain homogenate.
Increases in the quaternary compound (24) 0
and freed indomethacin S are also seen.




87
Table 3-3. Half-life of disappearance and correlation
coefficient for each dihydropyridine and
quaternary pyridinium compound, in 20% rat brain
homogenate, and for each quaternary pyridinium
compound in 20% rat kidney homogenate. Halflife (min); or h = hour. Results are from one to three runs per compound and six to fifteen
sample measurements per run.
Compound Dihydro Quat Quat
31 and 24 3.5 llha 35ha (0.999) (0.97) (0.97)
32 and 25 1.9 stablec 16hb (0.99) (0.89)
29 and 22 1.2 5.8hb stablec (0.99) (0.77)
30 and 23 4.5 6.8ha 9.7ha (0.998) (0.94) (0.79)
28 and 21 1.6 stablec stablec (0.98)
27 and 20 37 1.9 2.3 (0.97) (0.998) (0.997)
aExperiment was followed for < one half-life. bparent drug was detectable, but not within first hour of each experiment.
Ccompound very stable; parent drug was not detectable over course of experiment.




88
This compound's half-life was thirty seven minutes in brain homogenate.
The major product of decomposition that was detected for compound 27 was another unique exception. The parent drug was detected after adding the dihydropyridine (compound 27) to the brain homogenate, but none of the respective quaternary compound could be measured as in the five other cases. This result was due to the quaternary compound being much less stable than its corresponding dihydropyridine. Therefore, if oxidation was taking place its product would decompose at a faster rate than the rate at which it was formed.
The five remaining dihydropyridines were oxidized to
give the pyridinium salt as the major decomposition product in brain homogenate. Lower levels of the parent drug were also detected in each case.
The proposed system of delivering drugs to the brain,
has many requirements to obtain ideal results. The dihydroCDS should oxidize rather rapidly once it enters the brain. This process allows the drug-carrier complex to acquire its "locked-in" effect. Then, slow hydrolysis of the ester bond would yield the parent drug in a sustained release manner. In the kidney the pyridinium ester should be as stable as possible in order to facilitate the excretion of the drug-carrier combination. This would help prevent the release of the parent drug in the periphery.




89
The results of the in vitro testing were used to compare and evaluate each chemical delivery system. The results were then contrasted with the ideal system. In this way one has a steppingstone to access the likelihood of success in an in vivo experiment.
In Vivo Studies
Preliminary Distribution of Compound 31
The N-substituted (amide nitrogen) hydroxyethylnicotinamide dihydro carrier ester of indomethacin (compound 31) was the first chemical delivery system in the series to be synthesized. In order to assess the new carriers' ability to deliver a drug to the brain this first compound (31) was used in a preliminary in vivo distribution experiment. Qualitative measurements of both the quaternary drug-carrier combination and the freed parent drug were made. The experiment was conducted in male rats and the drug was administered by iv injection.
The results of this experiment showed good penetration of the chemical delivery system into the brain. Low levels of the parent drug were also detected in the brain over the four hour experiment (Table 3-4). In the blood the quaternary drug-carrier combination was found in about one-half of the concentration in the brain five minutes after injection. This compound appeared to be excreted reasonably well from the periphery and after four hours its concentration was just above the detectable limit.




90
Table 3-4. Results from in vivo distribution of compound
31, 20 mg/kg (values in peak heightSEM). Two
animals per time point were injected with the
dihydropyridine compound.
QUAT INDOMETHACIN
BRAIN
5 min 9.30.8 1.80.6
1 hr.* 5.8 2.5
4 hr. 1.30.4 1.20.3
BLOOD
5 min 4.42.0 8.60.5 4 hr. 0.30.3 2.80.8
*one animal




Full Text

PAGE 1

APPLICATIONS OF HYDROXYALKYL DERIVATIVES OF A PYRIDINIUM SALT-DIHYDROPYRIDINE REDOX SYSTEM FOR DRUG DELIVERY TO THE BRAIN By MICHAEL JAMES PHELAN A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 1987

PAGE 2

FOR MY PARENTS John (Johnny) and Nadia Phelan The two people who have loved me every day of my life, and taught me more than any school ever could. Education is that which remains when one has forgotten everything learned in school. Albert Einstein (1879-1955)

PAGE 3

ACKNOWLEDGMENTS First of all I would like to express my sincere thanks to Dr. Nicholas Bodor for his kindness, generosity and patience. I appreciate the opportunity to have worked under his direction. It has been a remarkable experience. I would also like to thank Mrs. Sheryl Bodor for her kindness. In addition, I would like to thank each of the members of my committee for their help along the way. I want to thank the members of Dr. Bodor's group for their help and friendship over the years. I have had the great pleasure of having worked with a wonderful group of people. I must especially mention Dr. Efraim Shek for his help and guidance, especially during difficult times. Those others who have shown real friendship will not be forgotten. I make special reference to Gabrielle Brouillette, my love; Vasu Venkatraghavan, my friend and confidant; and Mrs. Jirina Vlasak, a truly good person and one of the nicest I have ever met. To all of the others who have especially extended their kindness, David Winwood, Toshio Nakamura, M. Masaki, L. J. Chang, Cindy Jordon, Joan Martignago, Laurie Johnston, Jill Mccornack, Pascal Druzgala, Thorsteinn Loftsson, Hartmut Derendorf and Kerry Estes, Emil Pop, Teruo Murakami, Whei Mei Wu, K. s. and Nirmala Raghavan, James Stephens, and Roy iii

PAGE 4

Little, I thank you for making things better. I would also like to thank Bob Perchalski for his curative advice and help with the analytical portion of my work, and Dr. Katovich for getting me started on and loaning the equipment needed for the antipyretic activity study. I wish to express my appreciation to Marcus Brewster for his early help during the good times. I thank Anna Marie Martin for her ability to read the illegible during her superb typing of this manuscript. I also need to acknowledge my best friends whose caring has helped sustain me through the years. I would especially like to thank my best and oldest friends, Tony Caporaletti and Rocco Caponi, my best friends from U of A, Joe Snyder and Vasu Venkatraghavan, and last but not least, my family, especially my parents, my brother Jack and my grandmother for never wavering. iv

PAGE 5

TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . KEY TO ABBREVIATIONS. Page iii vi ABSTRACT. . . . . . . . . . . .viii CHAPTERS I II III IV INTRODUCTION. The Blood-Brain Barrier Brain Specific Drug Delivery Chemistry of the Drug Delivery System. Epilepsy and Valproic Acid Nonsteroidal Anti-inflammatory Agents Objectives EXPERIMENTAL. Materials and Methods. Synthesis .................................. High Pressure Liquid Chromatography Systems. Chemical Stability In Vitro Studies In Vivo Studies RESULTS AND DISCUSSION Synthes is .......................... Chemical Stability In Vitro Studies In Vivo Studies CONCLUSION. . . . . . . . . REFERENCES 1 2 3 6 13 18 22 24 24 26 54 56 57 60 65 65 78 81 89 111 114 BIOGRAPHICAL SKETCH 119 V

PAGE 6

BBB bp CDS CNS C oc D DCC DHC-D DH-CDS DMAP DMSO ED50 g GABA GAD GI h 1HNMR HPLC icv iv kg KEY TO ABBREVIATIONS blood-brain barrier boiling point chemical delivery system central nervous system carrier degrees centigrade drug dicyclohexylcarbodiimide dihydropyridine carrier-drug dihydropyridine-chemical delivery system dimethylaminopyridine dimethylsulfoxide dose effective in 50% of experimental units gram gamma aminobutyric acid glutamic acid decarboxylase gastrointestinal hour proton nuclear magnetic resonance high pressure liquid chromatography intracerebroventricular intravenous kilogram vi

PAGE 7

LD50 lit. M mg min mL mm mM mp MW NAD+ NADH nm r rpm SEM sec TI tl/2 UV VS w/v L > dose lethal in 50% of experimental units literature molar milligram minute milliliter millimeter millimolar melting point molecular weight nicotinamide adenin e dinucleot ide, oxidized form nicotinamide adenine dinucleotide, reduced form nanometer quaternary carrier quaternary carrier-drug correlation coefficient revolutions per minute standard error of the mean second therapeutic index half-life ultraviolet versus weight per volume microliter greater than vii

PAGE 8

Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy APPLICATIONS OF HYDROXYALKYL DERIVATIVES OF A PYRIDINIUM SALT-DIHYDROPYRIDINE REDOX SYSTEM FOR DRUG DELIVERY TO THE BRAIN By Michael James Phelan May 1987 Chairman: Nicholas s. Bodor Major Department: Medicinal Chemistry The design of drugs whose site of action is in the brain is often complicated by the presence of the bloodbrain barrier. This barrier is a physical entity which restricts all but the most lipophilic of compounds from entering or leaving the brain. To overcome this problem a system has been developed utilizing pyridinium saltdihydropyridine redox carriers. The lipophilic dihydropyridine is designed to penetrate the blood-brain barrier and then enzymatically oxidize to the hydrophilic pyridinium salt. This charged molecule allows for a "lock in effect and subsequent sustained release of the parent drug in the brain. The more hydrophilic form can also be rapidly eliminated from the periphery. These characteristics should result in a reduction or elimination viii

PAGE 9

of both central and peripheral toxicities associated with the parent drug. The carriers developed in this work contained a nicotinamide moiety derivatized with a hydroxyalkyl group which varied in length and position. This group allowed drugs containing a carboxylic acid functional group to be coupled to the carrier via an ester linkage. Various chain lengths and positions were used in an effort to prepare the most advantageous carrier possible. Naproxen, indomethacin, and valproic acid were each used to prepare one or more chemical delivery systems. The rate of disappearance was measured for each naproxen and indomethacin dihydropyridine compound along with its corresponding pyridinium salt, under a variety of in vitro conditions. Three of the most promising delivery systems and naproxen itself were each administered to rats to study their in vivo distribution. These experiments proved the carriers' ability to deliver and sustain the release of either parent drug. The investigation also showed a dramatic increase in brain/blood and brain/tissue ratios after giving one naproxen dihydropyridine delivery system when compared to the ratios that were measured after administering naproxen itself. This represented a significant improvement in the distribution of the drug, in accord with the system design. The sustained effect of the dihydropyridine compound significantly improved naproxen's ix

PAGE 10

antipyretic activity in a rat model, when compared to an equimilar dose of naproxen itself. X

PAGE 11

CHAPTER I INTRODUCTION Water is an essential "element" of life. It has many unusual and amazing properties. One example of this, becoming less dense as it freezes, is what allows life to survive in freezing water. The exposed top freezes first, and the resulting ice forms an instilating layer protecting the water and the life below. Water is also an excellent solvent. It can dissolve a vast number and wide variety of compounds and elements. It is the basis of bioorganic reactions, the solvent of choice! Water is the major component of the blood, and so plays a leading role in distributing the nutrients of life. Water is also capable of dissolving and in the case of blood, distributing many unwanted products. This allows for the spread of a toxic substance throughout ones circulation. However, one is not entirely without protection against these attacks. The blood-brain barrier is designed to filter the circulating blood flow. In this way, it can protect the brain from a number of harmful substances. Unfortunately, it can also vastly complicate drug design. 1

PAGE 12

2 The Blood-Brain Barrier The term blood-brain barrier was coined in 1921 by the Russian, Lina Saomonoona Stern.1 It was used to describe the impediment to acidic dyes which stained most organs, but not the brain, when given by the circulation.1 2 This observation led to the concept of a membrane wall-like barrier, which existed as a structural blockade interposed between the blood and the brain. This term was later used to explain the restriction from the brain of many water soluble substances, even some of low molecular weight. The basis of this barrier has been shown to be due to the physical aspects of the brain capillary system.3,4 These cerebral capillaries differ from general systemic capillaries in several ways. Endothelial cells of systemic capillaries are joined by relatively loose junctions which allow the passage of many compounds into the extracellular fluid surrounding these areas. In contrast, the endothelial cells which comprise central nervous system capillaries are joined by tight intercellular junctions, which form a continuous cellular layer between the blood and the brain.5 8 Since little passage of water-soluble compounds is possible, and endothelial cell membranes present a significant impediment to transcapillary movement, the result is an atypical barrier which has developed as protection for the CNS. Movement of solute across the blood brain barrier (BBB) is consistent with the theory of simple diffusion through an

PAGE 13

3 aporous lipid membrane.9 Pores between cells do not exist; therefore, bulk flow cannot take place across the BBB. However, the cell membranes do consist of a lipid bilayer having globular proteins asymmetrically distributed. Substances which are lipid soluble will cross the membranes in both directions by simple diffusion. Therefore, the rate at which a compound enters the brain is normally related to its lipid solubility.lO,ll The molecular weight of a compound can also be of great importance when considering transport of drugs through the BBB. Larger molecules with MW>SOO tend to diffuse very slowly through biological membranes.11,12 Many small watersoluble compounds, however, are thought to pass the BBB by carrier-mediated transport.13,14 In order to deliver therapeutically useful drugs to the brain, one must often circumvent the function of the bloodbrain barrier. A novel and effective way to do this has been brought to light by Bodor et al.15 It involves the site-specific, sustained release of drugs to the brain. Drugs which are normally partially or completely barred from the brain can be delivered using a chemical delivery system (CDS). This system employs carriers containing a pyridine ring moiety, capable of participating in a pyridinium saltdihydropyridine redox-type reaction. Brain Specific Drug Delivery The effect of a drug is usually the result of the drug's interaction with macromolecules in some specific

PAGE 14

4 target tissue or cell type. Along with the compound's desired therapeutic effect, it will also bring about certain toxic consequences depending on concentration. Often, the specific site of toxicity is not associated with the target tissue. Therefore, if the distribution of a drug could be more precisely controlled, the therapeutic index (TI= LD50/Eo50) would likely increase. The method for brain specific delivery of drugs developed by Bodor has shown great promise of achieving this goal. The first compound used to evaluate this system was phenethylamine. Phenethylamine (D) was chemically coupled with nicotinic acid (C). It was then quaternized to the pyridinium salt (QC+-D), followed by reduction to the dihydropyridine form (DHC-D) using sodium dithionite. Following iv administration, the lipophilic chemical delivery system (DHC-D) distributes rapidly in both the blood and the brain as illustrated in Figure 1-1. Oxidation back to the original quaternary salt (QC+-o) is then thought to occur via the NAD+-NADH redox enzyme system.15 This more hydrophilic form can be rapidly eliminated from the body resulting in a lower circulating level of the parent drug. In the brain, however, the polar drug-carrier combination (QC+-D) is locked in by the blood-brain barrier. This compound then serves as a source for the sustained release of the parent drug in the brain. The slow release of the drug (D) depends on the rate of enzyme hydrolysis. This is affected by the structure of both the carrier as well as the

PAGE 15

Brain DH -D kcleavage BBB oc+-o Reduction DHC-D D 5 Periphery DHC-D oc+-o Figure 1-1. Schematic representation of the proposed pyridinium salt-dihydropyridine drug delivery system

PAGE 16

6 parent compound. Once the drug-carrier combination is hydrolyzed, the remaining carrier portion, due to its lower molecular weight, should be actively transported out of the brain. This system should provide for better delivery of a compound to the brain, while offering the advantages of sustained release and reduced toxicity. Chemistry of The Drug Delivery system Pyridinium Salts When a nitrogen atom in a heterocyclic ring possesses a lone pair of electrons, those electrons can form a bond between that nitrogen atom and a carbon atom of suitable polarizability. In this case, the nitrogen is in a quaternary form. The attacked molecule must be one that can release an anion during the quaternization. This type of reaction can be viewed in two ways. The first is to see the reaction as a nucleophilic replacement of the halogen or other similar leaving group, by attack of the lone pair of the ring nitrogen. Alternatively, the quaternization can be looked at as an electrophilic attack on the ring, which usually takes place only at a nitrogen atom. Therefore, the availability of the electron pair, as influenced by the ring substituents and the steric factors involved, can dramatically affect the rate of quaternization. The solvent used and the nature of the electrophile involved in the reaction are also important factors in predicting the course of a given reaction.

PAGE 17

7 Alkyl halides are by far the most common reagents for the formation of heterocyclic quaternary salts and of these, iodoalkanes are most often reported.16 Primary halides, as expected, react faster than secondary compounds,17 and tertiary halides normally result in elimination, giving the corresponding acid and an alkene.18 The solvent used in the formation of heterocyclic salts can also be quite important. Many solvents of varying polarities have been used, including an excess of the quaternizing agent itself. The solvent's polarity affects the rate of the reaction, as well as the products ability to precipitate from the reaction mixture. The influence of a solvent on the problem of isolating a quaternary salt, once formed, is at times a major one. Water is often held very tightly by the desired product. Non-hydric solvents such as benzene usually cause such reactions to be quite slow. Therefore, the best solvent would seem to be non-hydric, with a reasonably high dielectric constant. The most important quaternary pyridine derivatives occurring naturally are those of nicotinamide, which have a coenzyme function. In metabolic oxidations, these coenzymes accept hydrogen directly from a variety of oxidizable substrates and transfer it to other acceptors. In this way, they are key metabolic catalysts. The 1-methylderivative of nicotinic acid is found in plants and is known as trigonelline.19 Another pyridinium compound of biological importance, picolinaldoxine methiodide (PAM), evolved as an

PAGE 18

8 antidote for organophosphate poisoning. This bifunctional molecule was designed to serve as an antidote by binding to one enzyme binding site and, at the same time, providing a nucleophilic group to displace the bound phosphate. Dihydropyridines The partial reduction of pyridinium salts to dihydro derivatives of known structure can be successfully carried out in a limited number of cases. In part, the difficulty may lie in the readiness of the partly reduced structure to oxidize, polymerize, or be further reduced unless stabilizing groups are present. Of all the chemical methods thus far applied, reduction by sodium dithionite to dihydro derivatives as studied by Karrer et al. is probably the most important.20 With dithionite, a variety of nicotinamide quaternary salts have been converted to 1,4-dihydro products.15,20, 21 Here, the carbamoyl group exerts a stabilizing influ-ence on the product. Earlier studies of 1,2-and 1,4-dihydropyridines sought to avoid the stability difficulties and also limit the opportunities for tautomerization by using highly substituted derivatives.20 The first use of sodium dithionite to convert NAD+ into NADH led to the preparation of a number of dihydropyridines by this method. Reduction of 3-substituted or 3,5-disubstituted pyridinium compounds by sodium dithionite in mildly basic solution yielded the corresponding l,4-dihydro-pyridines.15,21 Some early reports of l,2-dihydropyridines22 have since been shown to actually be the 1,4-isomer.23

PAGE 19

9 Many !-substituted 1,4-dihydropyridines have been prepared by this method, where substitution in the !-position has been in the form of alkyl,15, 21 ,24-27 benzyl,28-31 alkoxyrnethyl,28 2-hydroxyethyl,32 or a sugar residue.22,28 A number of l-alkyl-3-cyano-1,4-dihydropyridines have also been synthesized.33,34 Various 1,4-dihydropyridines with substituents in the 3-position have been synthesized, with X = CH=NNHC6 H 5 COCH3, C02H, C02R, CON(CH3)2, CONHC6Hs, 4-methyl-2-thiazolyl, benzoyl, and 2-benzthiazolyl.30,33 ,35 However, dithionite did not produce any isolable dihydro products from pyridinium salts when X was hydrogen or alkyl.23 The introduction of a methyl group in the 2, 4, or 6-position of a pyridinium salt with an electron-withdrawing substituent in the 3-position did give the expected l,4-dihydropyridine.28,36,37 This was also the case with a number of pyridinium salts with electron-withdrawing groups in both the 3-and 5-positions.38,39 The mechanism of dithionite reduction has been the subject of some controversy. The prevailing idea shows that the reaction proceeds via a sulfinate intermediate which is stable enough in alkaline solution to have been iso lated.31 Upon protonation, the salt is converted into the unstable acid which rapidly decomposes to the 1,4-dihydronicotinamide compound. When dithionite reduction was carried out in o2o, the monodeuterated dihydro form of the compound was

PAGE 20

10 obtained.31 Repeated oxidation followed by dithionite reduction of this product in o2o gave the pure dideuterated derivative. A number of deuterated dihydropyridines have been prepared by this method.30,31,34 The 1,2-and 1,4-dihydropyridines are the most stable due, presumably, to the involvement of the nitrogen lone pair in thew electron system. These are the isomers with the most sp2-hybridized centers. Some systematic work has been carried out to determine the effects of substituents on the stability of dihydropyridines. The parent l,4~dihydropyridine was described as a very unstable substance in air.40 Electron-withdrawing substituents capable of resonance interaction in the 3-and/or 5-positions were shown to stabilize dihydropyridines by extending the conjugation.41 Substitution in the 3-and 5-positions with conjugating groups results in lowered energies and transfer of electronic charge to the substituents.42 This then results in an appropriate decrease in reactivity. Substituents in the 3-and 5-positions which donate electrons by resonance have a destabilizing effect. Alkyl substitution on nitrogen has the same general effect on stability, but a glucosyl substituent on nitrogen has a tremendous stabilizing influence.43 A considerable amount of work has been done to determine the stability of NAD+ and NADH analogs in aqueous solution. The !ability of the pyridinium ring has been shown to

PAGE 21

11 result from nucleophilic attack in the 2or 4position.44,45 The 1,4-dihydropyridine decomposition appears to result from protonation at C-5, and subsequent attack by water or other necleophile at C-6. A further cause of instability of some of these compounds is the !ability of a carboxamido or a carbalkoxy group toward hydrolysis.46 Therefore, the three principal sources of instability in model compounds are amide and ester hydrolysis, nucleophilic attack on the pyridinium ring, and acid-catalyzed hydration of the dihydropyridine ring system. The first two of these reactions are favored under basic conditions and the last is, of course, a concern under acidic conditions. Two esters, l-carbomethoxymethyl-3-carbamoyl-and lcarboisopropoxymethyl-3-carbamoyl pyridinium ions underwent rapid decomposition in alkaline solutions. At pH 9.2 the methyl ester's half-life was approximately 3 minutes and the isopropyl ester's half life was about six times as long.47 Therefore, in order that not more than 10% of the compound decomposes in 24 hours, the pH must not exceed 5.7 and 6.5 respectively, if the rate of hydrolysis is proportional to the hydroxide concentration over this range. In the case of compounds which possess strongly electron-withdrawing groups in both the 1 and 3-positions, the pyridinium ring is the first site of attack when the pH is raised. This can take the form of an unwanted decomposition but this is also often the case in a reduction of a

PAGE 22

12 pyridinium salt to its dihydro analog, as we have seen in the case of dithionite. The decomposition of 1,4-dihydropyridines in the presence of aqueous acids has been studied by many groups in the past.44,48-50 It is thought to involve several successive steps. The primary acid decomposition reaction is a twostep process resulting in the hydration of the 5,6-double bond.49 ,51 The first step, which is normally ratedetermining in aqueous media, is protonation at C-5, followed by fast nucleophilic attack at C-6. Under these conditions, the nucleophile would be hydroxide ion or water. Decomposition of 1,4-dihydropyridines can be followed by monitoring their Amax in the UV region around 350 nm. A study of the acetic acid-catalyzed rate constants for the hydration of l-alkyl-3-carbamoyl-1,4-dihydropyridines showed a good correlation (r = 0.991) with the o* values for the lsubstituents.52,53 Dihydropyridines containing more electron-donating groups are shown to be highly acid-labile, while pyridinium ions containing electron-withdrawing groups become more susceptible to attack in basic solution.47 This therefore, results in a loss of overlap between the stable pH regions for the two isomers. The mechanism of oxidation of 1,4-dihydropyridines is still a point of some controversy. If the factors controlling the oxidation of these compounds can be understood, then one can affect the stability of the dihydro compound in

PAGE 23

13 order to allow its oxidation to proceed at a desirable rate. The initial view put forward by Abeles et al. was that the mechanism concerned the movement of a proton with two associated electrons i.e., a direct hydride migration.54 Later, this mechanism was modified to an initial electron transfer followed by a hydrogen radical migration. This hypothesis in turn was replaced by an electron-proton-electron transfer mechanism.55, 56 However, the most recent findings indicate a dependence on the specific oxidizing agent. The conclusion in the case of enzymatic-type oxidations was that the original theory of a concerted hydride transfer was correct.57 Epilepsy and Valproic Acid Epilepsy is among the most common of chronic neurological disorders. Convulsive symptoms or seizures also occur during or as sequels to many of the other diseases that affect the brain. For the majority of epileptic patients long-term drug therapy represents the only practical form of treatment.58 The goal of treatment is, of course, the prevention of recurrent seizures. One of the most widely used drugs for the treatment of epilepsy is the relatively new valproic acid. Valproic acid is a branched chain carboxylic acid. It is unique among anti-epileptic drugs in that it has neither a nitrogen atom nor a ring moiety. It has a pKa of 4.95 and a molecular weight of 144.59 It occurs as a slightly oily, clear liquid that is very soluble in both water and organic solvents.

PAGE 24

14 Valproic acid was first demonstrated to possess anticonvulsant effects by Meunier et al 60 The sodium salt of valproic acid was first marketed as "Depakine" in France in 1967. Since that time valproate has been licensed in much of Europe. It was authorized for use in epilepsy therapy in the United States in 1978. Valproic acid is now increasingly used in the treatment of primary generalized seizures, especially those of the absence type.61 several controlled trials of valproate as monotherapy for specific syndromes have been reported in a 1981 review.62 In all seven of the cases reported, valproate was at least as effective as the "optimal" drug in each case. This surely indicates that valproate has a very wide spectrum of anticonvulsant activity, possibly greater than that of any other drug in use. A larger number of trials have shown that seizure control in absence attacks is improved when valproate is added to previous therapy.63,64 Valproate has also proven quite useful in patients failing to respond to other therapies.65 About one-third of the patients showed better than 75% red~ction in seizures. Successful treatment of status epilepticus, refractory to diazepam or barbiturates has also been reported.66 Mechanism of Action of Valproic Acid Since valproic acid was first discovered to have clinical utility in the management of epileptic seizures, its mechanism of action has been studied rather

PAGE 25

15 extensively. Although a number of various facts" have been uncovered, their full implication and precise role in the mechanism of action or actions is not always known. Two main types of anticonvulsant action are suggested by experimental and clinical studies of valproate. One is a direct pharmacological effect related to the plasma and subsequent brain concentration of the drug. This is best illustrated acutely, after high doses of valproate. The other type of action is indirect and presumably relates to active metabolite concentrations remaining in the brain, or to adaptive or other changes brought about by valproate or its metabolites. Possible changes involving membranes, receptors, or enzymes as a whole have been mentioned in the literature. Almost all of the biochemical and neurophysiological data concern the direct effects of the drug. Several hypotheses for its mechanism of action have been presented;67,68 however, there is not a clear choice among them. This is because some of our basic knowledge is deficient and because the required measurements, both in vitro and in vivo, are sensitive and difficult to perform. The theory that has received the most attention by far is that GABAergic inhibition is enhanced through an action on the synthesis or further metabolism of GABA. However, doubts have been raised by observations showing poor correlations between increases in brain GABA and anticonvulsant action.62 This has usually been seen after

PAGE 26

16 acute doses of the drug. Anti-seizure action is observed but no increase.in GABA is detected shortly after administration of valproate. In addition, some in vivo determinations of GABA turnover indicate a reduced synthesis of GABA, even in the presence of unchanged or increased GABA levels, which seems to indicate a reduced synaptic release of GABA. This is further strengthened by the fact that no study had been able to demonstrate any increase in GABA release after valproate administration.62 However, another study showed an increase in overall brain GABA concentration as well as an elevation of glutamic acid decarboxylase (GAD) activity, following administration of valproic acid.69 The enzyme GAD catalyzes the synthesis of GABA and is thought to be the rate-limiting enzyme in determining GABA levels in the brain.70 Other studies have put forward a hypothesis that valproate itself acts on post-synaptic receptor sites in order to mimic or enhance the inhibitory action of GABA. Unfortunately, there are no clear experimental demonstrations showing enhanced efficacy of physiologically stimulated GABAergic inhibition after systemic valproate administration. 71 This is not the case after systemic administration of benzodiazapines. Valproic Acid and the BBB In the mouse, valproic acid is rapidly absorbed after oral administration. Maximum blood serum levels occurred between 5 and 30 minutes after treatment.72 Approximately

PAGE 27

17 35 to 45% of the dose is absorbed via this route of administration. Peak brain concentrations occur at the same time as peak serum levels but only reach about 20% of the serum level. Brain valproate concentrations of 27%69 and 10%73 of plasma levels have also been reported by different groups, but all three studies involved mice. The group finding the initial brain-to-plasma ratio of 27% did so after ip injection. Both of the other two studies involved oral administration. The extreme rapidity with which valproate enters the brain suggests a possible contribution due to active transport62,69 because valproic acid is predominantly present in its ionized form at physiological pH. This charged species should certainly not pass the BBB by passive diffusion. Active transport of valproic acid from brain to plasma may also be functional in the mouse, as has been shown previously for the dog.74 Uptake of valproic acid into the brain of the cat when given by iv injection has also been studied. 75 The plasma level for each given cat was seen to remain rather stable for the entire 90 minute experiment. The brain concentration, however, decreased rapidly to relatively low levels. The brain:plasma ratio declined from a high of 0.72 to 0.11 during the 90 minute period. Toxicity In general, valproic acid does not suffer from major toxicity problems. However, it is often prescribed in

PAGE 28

18 combination with other anticonvulsants. This tends to complicate the delineation of adverse reactions attributable solely to valproate. The side effects most commonly reported are not life-threatening. They include such things as nausea, vomiting, disorientation and fatigue. A few reports have appeared in the literature for each of several more severe effects. These effects have been life-threatening, and are not always well understood. Hepatic failure and hyperammonemia are the most common causes of mortality associated with valproate treatment. The cases of hyperammonemia are most often found in patients with hepatic failure. 76 A reduction in platelet count has also been seen in patients taking valproic acid. The reductions seen in one study caused one-third of the subjects to drop below the normal range. 77 However, the drop was not low enough after two months to cause any bleeding abnormalities. Other reported problems associated with this drug have been red cell aplasia78 and pancreatitis. 79 In all of these reported cases the adverse effects associated with valproate therapy were shown to be dose (rather than exposure length) related. These effects could be reversed either by reduction or cessation of valproate in the vast majority of cases. Nonsteroidal Anti-inflammatory Agents This class of compounds covers a wide variety of drugs, both in chemical structure and in pharmaceutical application. One group, the acidic nonsteroidal anti-inflammatory

PAGE 29

19 agents, has been found to have anti-inflammatory, analgesic, and often antipyretic activity. This has led to their usage in a number of clinical applications. Indomethacin and naproxen are two of the most potent of these compounds. They have been extensively studied in the literature, both for their own activity and as reference standards to judge newly discovered compounds of this class. They have shown activity against a multitude of inflammatory diseases, especially rheumatoid arthritis and gout. They are effective in a number of migraine headache syndromes.80,81 They are also quite useful in lowering the fever associated with a variety of pyrogens, including cancers resulting in neoplastic fever.82 Indomethacin and naproxen are both arylacetic acids. They are off-white crystalline solids with molecular weights of 358 and 230, and pKa's of 4.5 and 4.2 respectively. In aqueous solution they exist primarily in their ionized form, and so do not effectively penetrate the BBB. After administration these compounds normally achieve a blood/brain ratio of twenty to one. Clinically, these drugs have been used most often for their anti-inflammatory and analgesic properties in the treatment of rheumatoid and other types of arthritic conditions. Mechanism of Action Traditionally nonsteroidal anti-inflammatory compounds were thought to exert their anti-inflammatory action through peripheral mediation of prostaglandin synthesis.

PAGE 30

20 Prostaglandins appear to sensitize pain receptors to mechanical or chemical stimulation. However, as early as 1961 there have been reports in the literature that these drugs accomplish their inhibitory role, at least in part by an action on the central nervous system.8 3 It now seems clear that these two compounds work principally, if not entirely, by the inhibition of prostaglandin synthesis. They prevent the production of prostaglandins in body tissues by inhibiting cyclooxygenase, the enzyme that catalyzes the formation of prostaglandin precursors from arachidonic acid.84 This is true for their anti-inflammatory action, as well as the analgesic and antipyretic effects associated with indomethacin and naproxen. The analgesic properties of these compounds have been shown only in the presence of inflammation. This finding has led to a distinction between nonsteroidal antiinflammatory drugs and the narcotic analgesics which increase the pain threshold for both inflamed and normal tissues.8 5 However, it has been shown that these acidic nonsteroidal compounds do inhibit prostaglandin synthesis at the central level as well as the peripheral level. In addition, simultaneous administration of one of these drugs by systemic and by icv injection to rats with inflammation elicits a synergistic effect rather than a simple addition.86-87 This central mechanism of action is also responsible for indomethacin and naproxen's antipyretic activity. These

PAGE 31

21 compounds work only in the presence of fever and so do not alter body temperature in afebrile animals. It has also been shown that acidic nonsteroidal anti-inflammatory drugs produce their antipyretic action by inhibiting prostaglandin generation within the hypothalamus.88 Toxicity The most common adverse effects associated with nonsteroidal anti-inflammatories are those of GI disturbances. These include nausea, with or without vomiting, indigestion, heartburn, and abdominal pain. Indomethacin and naproxen are also capable of reactivating latent peptic ulcers. In the case of indomethacin this can extend to intestinal lesions as well. These drugs may also cause such problems in patients with no previous history of ulcers.84 However, these effects can be minimized by administering the compounds with food or antacid. Indomethacin and naproxen have been found to produce various CNS disturbances. Headache, dizziness, lightheadedness, fatigue, insomnia and depression are the most common of these problems. Indomethacin, however, can also effect more severe CNS reactions. Although much less frequently, such things as psychic disturbances with psychotic episodes, hallucinations, nightmares, anxiety and coma have been reported.84 Psychotic episodes as well as the GI effects are particularly likely to occur in geriatric patients.

PAGE 32

22 There are also a number of less frequently reported, but often more severe cases of peripheral toxicities. In extreme instances these problems have even resulted in death. Various types of adverse hematologic effects have occurred in patients receiving indomethacin or naproxen. These include such conditions as thrombocytopenia and granulocytopenia89 as well as a variety of anemias.90 In addition, nonsteroidal anti-inflammatory drugs have been shown to cause nephrotoxicity, including renal failure91 and hepatic effects such as jaundice and hepatitis.84 These are especially common in patients with previously impaired renal or liver function. The majority of toxicities associated with these compounds have been shown to be dose dependent. Often the effects are reversible. Normally, by lowering the dose or cessation of therapy many of these problems can be addressed. Objectives This project is designed to develop a number of hydroxyalkyl chemical delivery systems. These carriers are made in order to expand the applicability of the drug delivery system developed by Bodor.15 They allow for the delivery of drugs having a carboxylic acid functional group. This is accomplished by attaching the drug via an ester linkage, to a carrier which contains a pyridine ring capable of participating in a pyridinium salt-dihydropyridine redox-type reaction. The hydroxyalkyl substituent

PAGE 33

23 can vary in chain length and position on the basic nicotinamide structure. In this way, one may be able to fine tune the design of an ideal carrier. The second part of this work is to attach various drugs to the carriers. These drugs are chosen so that their usefullness would be enhanced by increased brain delivery. At the same time, the drugs should show reduced toxicity due to a decrease in the circulating levels of the parent compounds. Once synthesis is completed both buffer and in vitro stability should be tested. This gives a basis for comparison in order to determine which chemical delivery system or systems should be best suited for in vivo distribution studies. The distribution of the parent drug is indicative of the success or failure of a given delivery system to increase brain penetration of the active compound. However, the activity of the drug once delivered is the final and possibly the most significant determination.

PAGE 34

CHAPTER II EXPERIMENTAL Materials and Methods Salts and nondeuterated solvents were obtained from Fisher Scientific. All salts were of reagent grade. Solvents used for high pressure liquid chromatography were of spectral grade. Water was purified in this laboratory using a Sybron Barnstead Nanopure II deionizer and filter system. All other nondeuterated solvents were of either reagent or spectral grade. Naproxen and indomethacin were obtained from Sigma Company, in the purest grade available. The remaining chemicals and reagents and deuterated solvents were obtained through Aldrich Chemical Company, unless otherwise stated. These products were of at least reagent quality and were used without further purification except where stated. Melting points were uncorrected and were determined using an Electrothermal melting point apparatus, equipped with the manufacturer's calibrated thermometer. Ultraviolet spectroscopy was performed on a Hewlett Packard 8451A diode array spectrophotometer. Proton nuclear magnetic resonance spectra were obtained using a Varian T60A, EM 360A, or EM 390 spectrometer. Chemical shifts are reported in parts per million units, downfield from tetramethylsilane used as an 24

PAGE 35

internal standard. Elemental analyses were performed by Atlantic Microlab Inc., Atlanta, Georgia. 25 High pressure liquid chromatography was carried out on one of the following systems: 1) Perkin-Elmer Series 4 chromatographic pump, ISS-100 auto sampler, LCI-100 integrator, and a Kratos Model 757 UV/visible, variable wavelength detector; 2) LDC/Milton Roy constaMetric III G metering pump, Perkin-Elmer LCI-100 integrator, and a Kratos Model 757 UV/visible, variable wavelength detector; 3) Kontron System 600 pump and auto sampler, Perkin-Elmer LCI-100 integrator, and a Kratos Model 757 UV/visible, variable wavelength detector; 4) Waters Associates Model 510 pump, Kontron MSI 660 auto sampler, Hewlett Packard Model 3390A integrator, and a Kratos Model 757 UV/visible, variable wavelength detector. The column used for this chromatographic work was either a Toyasoto 25 cm, 5 um particle size, 0DSc18 reversed phase column or an ASI 25 cm, 10 um particle size, c 8 reversed phase column. The Toyasoto column was protected with a guard column packed with Whatman pellicular 0Ds-c18 media. The centrifuge used to spin down tissue homogenates was a Dynac Centrifuge having a maximum spin rate of 3000 rpm. Chromatographic samples (in vitro) were centrifuged using a Beckman Microfuge 12 capable of a 10,000 rpm spin rate.

PAGE 36

Synthesis 3-Carbamoyl-1-(2-hydroxyethyl)pyridinium iodide (_!) 26 The compound 2-iodoethanol (2.58 g, 15.0 mmol) was dissolved in acetone (20 mL). Nicotinamide (1.83 g, 15.0 mmol) was added and the mixture was heated to reflux, where upon complete dissolution of the nicotinamide was obtained. The reflux was stopped after 8 hours and the precipitated product filtered and dried. The 2.20 g of crude product were twice recrystallized from ethanol resulting in 2.00 g of compound (_!) The product was obtained in an overall yield of 45%; mp 126-127C. UV(CH 30H): 224 and 268 nm. 1 H NMR (o2o) 6: 9.3-9.5 (bs, lH, pyridine H-2); 8.9-9.3 [m (9.0-9.3, bd, lH, pyridine H-6), (8.9-9.2, m, lH, pyridine H-4]; 8.1-8.5 (m, lH, pyridine H-5); 4.7-5.0 (m, 2H, N-CH2); 4.0-4.3 (bt, 2H, O-CH2). Analysis: (c8tt11rN 2o2 ) Calculated: C, 32.67; H, 3. 77; I, 43.15; N, 9.52; Found: C, 32.63, H, 3.78; I, 43.07; N, 9.47. l-Hydroxy-3-iodopropane (1_) An acetone solution of sodium iodide (30 g, 0.20 mo!), and l-chloro-3-hydroxypropane (14.2 g, 0.15 mo!) was stirred at 60C for one day. The precipitated sodium chloride was removed by filtration. The solvent was removed under reduced pressure and the oily residue was vacuum distilled. The first fraction, bp 64-70C, was shown by TLC, chloroform:methanol (9:1) to contain only compound 1_. The

PAGE 37

27 purified product was obtained in a yield of 19.8 g or 71.0% overall. 1 H NMR (CC1 4 ) 15: 4.2-4.5 (m, lH, OH); 3.4-3.8 (t, 2H, CH2-I); 3.0-3.4 (t, 2H, CH2-0); 1.7-2.3 (p, 2H, CH2). 3-Carbamoyl-1-(3-hydroxypropyl)pyridinium iodide(]_) Previously prepared l-hydroxy-3-iodopropane, compound 2 (5.58 g, 30.0 mmol) was combined with nicotinamide (3.66 g, 30 mmol) in acetone (50 mL). The mixture was refluxed overnight, the solvent was removed, and the oily residue was stirred with ether until the formation of a yellow powder was observed. A small amount of unquaternized nicotinamide was removed by its crystallization from methanol. The solvent was removed and the product was recrystallized from a combination of ethanol:2-propanol (9:1). The crystalline needles of the final product (4.00 g) were collected by vacuum filtration under nitrogen, in an overall yield of 43%; mp 112.0-112.5C. UV (CH30H): 224 and 268 nm. 1 H NMR (d6 -DMS0) 15: 9.5-9.7 (bs, lH, pyridine H-2); 9.2-9.5 (bd, lH, pyridine H-6); 8.9-9.2 (bd, lH, pyridine H4); 7.9-8.7 [m(7.9-8.2, bs, lH, NH), (8.0-8.6, m, lH, pyridine H-5), (8.4-8.7, bs, lH, NH)]; 4.8-5.0 (t, 2H, N-CH 2); 4.5-4.8 (m, lH, OH); 3.6-3.9 (t, 2H, O-CH 2); 2.1-2.6 (p, 2H, CH2). Analysis:(C9H13IN202) Calculated: C, 35.09; H, 4.25; I, 41.19; N, 9.09; Found: C, 35.05; H, 4.29; I, 41.07; N, 9.06.

PAGE 38

28 3-Carbamoyl-1-(3-hydroxypropyl)pyridinium bromide (_!) To a flask containing 3-bromopropanol (13.9 g, 0.10 mol) and nicotinamide (12.2 g, 0.10 mol), 50 mL of acetone was added. This mixture was heated to reflux, thus allowing complete dissolution of the nicotinamide. After 6 h the mixture was allowed to cool to room temperature. Upon cooling the product crystallized and the off-white solid was filtered, washed with ether and dried under nitrogen. It was recrystallized from a mixture of 2-propanol:ethanol (3:1), and the final product weighed 12.2 g. This gave an overall yield of 46.7%; mp 124-126C. UV (CH30H): 222 and 268 nm. 1 H NMR (d6 -DMS0) o: 9.7-9.9 (bs, lH, pyridine H-2); 9.3-9.6 (d, lH, pyridine H-6); 9.0-9.3 (m, lH, pyridine H-4); 8.6-8.9 (bs, lH, NH); 8.1-8.6 (m, 2H, pyridine H-5 and NH); 4.7-5.2 (t, 2H, N-CH2); 4.3-4.6 (bs, lH, OH); 3.4-3.7 (t, 2H, O-CH 2); 2.0-2.5 (p, 2H, CH2) Analysis:(C9 B13BrN 202) Calculated: C, 41.40; H, 5.02; Br, 30.60; N, 10.73; Found: C,41.26; H, 5.07; Br, 30.66; N, 10.69. 3-[(2-Hydroxyethyl)carbamoylpyridine (~) A neat mixture of 2-aminoethanol (6.1 g, 0.10 mol) and ethyl nicotinate (15.1 g, 0.10 mol) was refluxed overnight. As the mixture was cooled to room temperature, the product precipitated as a crystalline solid. It was filtered, washed with ether and then recrystallized from 2-propanol/ether. The final product was collected by vacuum

PAGE 39

29 filtration and washed with ether. The dried, white compound weighed 10.7 g, resulting in a 64.5% yield; mp 88.5-89.5C (lit. value 92C). UV (CH30H): 222 nm. 1 H NMR (d6-DMS0) 6: 9.0-9.2 (bs, lH, pyridine H-2); 8.5-8.9 (m, 2H, pyridine H-6 and NH); 8.2-8.4 (m, lH, pyridine H-4); 7.4-7.7 (m, lH, pyridine H-5); 4.8-5.0 (t, lH, OH); 3.3-3.8 (m, 4H, (CH2)2) Analysis:(C8 H10N 2o2 ) Calculated: C, 57.82; H, 6.07; N, 16.86; Found: C, 57. 73; H, 6.11; N, 16.82. 3-[(2-Hydroxyethyl)carbamoyl)-1-methylpyridinium iodide (__) Compound 2. (1.00 g, 6.02 mmol) was dissolved in acetone and refluxed overnight with methyl iodide (1.70 g, 12.0 mmol). The yellow precipitate was collected by vacuum filtration. The material was recrystallized from ethanol/ether, and 1.60 g of the cream colored crystalline plates were recovered. This resulted in an 86.5% yield; mp lll-113C. UV (CH30H): 222 and 268 nm. 1 H NMR (d6 -DMS0) 6: 9.4-9.5 (s, lH, pyridine H-2); 9.1-9.3 (d, lH, pyridine H-6); 8.8-9.1 (m, 2H, NH and pyridine H-4); 8.2-8.5 (m, lH, pyridine H-5); 4.6-4.9 (bt, lH, OH); 4.4-4.6 (s, 3H, CH3); 3.3-3.8 (m, 4H, CH2-N and CH2-o). Analysis:(C9H13IN202) Calculated: C, 35.09; H, 4.25; I, 41.19; N, 9.09; Found: c, 35.14; H, 4.25; I, 41.11; N, 9.06.

PAGE 40

30 3-[(3-Hydroxypropyl)carbamoylpyridine (1.) A mixture of ethyl nicotinate (15.12 g, 0.10 mol) and 3-aminopropanol (8.2 g, 0.11 mol) was refluxed in toluene (50 mL). An azeotrope of ethanol/toluene was removed under reduced pressure every 12 hand an equal amount of fresh toluene was added. The reaction was continued for 2 days to increase yield. The solvent was completely removed and the thick oily residue was distilled using a Kugelrohr. The first 3 mL fraction was discarded. The remaining compound was collected, bp 145-155C at 0.1 mm pressure. This gave a light yellow, viscous oil that solidified into a wax-like substance in the freezer. The compound weighed 16.0 g giving an 88.8% yield. UV (CH30H): 226 nm. 1 H NMR (CDC1 3 ) o: 9.0-9.2 (s, lH, pyridine H-2}; 8.6-8.8 (d, lH, pyridine H-6); 8.2-8.5 (bt, lH, NH}; 8.1-8.3 (d, lH, pyridine H-4}; 7.3-7.5 (m, lH, pyridine H-5}; 4.8-5.0 (s, lH, OH}; 3.4-3.9 (m, 4H, CH2-o and CH2-N}; 1.6-2.1 (p, 2H, CH2}. Analysis: ( c9 H12N 2o2 /4 H 2 0} Calculated: C, 58.52; H, 6.82; N, 15.16; Found: C, 58.46; H, 6.94; N, 15.12. 3-[(3-Hydroxypropyl}carbamoyl]-1-methylpyridinium iodide (~_) Compound 2. (1.0 g, 3.1 mmol) was quaternized using methyl iodide (2 rnL, 30 mmol} in acetone (40 mL) at reflux. The reaction was continued overnight and the acetone and excess methyl iodide were removed under reduced

PAGE 41

31 pressure. The thick oily residue was stirred for several hours with anhydrous ether. The ether was decanted and a fresh portion was again added with continued stirring. This resulted in a yellow-brown solid powder which was filtered and dried. The crude material was crystallized from 2-propanol. The yellow crystalline product was filtered, washed with ether and air dried. It weighed 1.65 g. The product was obtained in overall yield of 94.6%; mp 112-113C. UV {CH30H): 222 and 268 nm. 1 H NMR {d 6 -DMS0) 6: 9.3-9.5 {bs, lH, pyridine H-2); 8.8-9.3 {m, 3H, pyridine H-6 and H-4, and NH); 4.3-4.6 {s, 3H, CH3); 3.2-3.7 {m, 4H, CH2-N and CH2-0); 1.5-2.0 {p, 2H, CH2). Analysis:{c10H15IN 202) Calculated: C, 37.29; H, 4.69; I, 39.39; N, 8.69; Found: C, 37.35; H, 4.70; I, 39.27, N, 8.68. 2-Propylpentanoyl chloride {..2_) Valproic acid {4.32 g, 30.0 mmol) was stirred at 0C, while thionyl chloride {3.60 g, 30.0 mmol) was added dropwise. When the addition was completed, the mixture was allowed to come to room temperature. The flask was then warmed in a water bath for 30 min at 50C. Dry benzene (2 x 50 mL) was added and then removed under reduced pressure. This compound was used in subsequent steps without further purification.

PAGE 42

32 2-Propylpentanoic acid, ester with 2-iodoethanol (10) The product from the previous reaction, compound 1._, was stirred at 0C while 2-iodoethanol (5.16 g, 30.0 mmol) was slowly added. The neat mixture was heated to 100C for 10 min, in a water bath. The reaction mixture was stirred for an additional 10 min at room temperature. It was then dissolved in ether (50 mL) and washed successively with water (30 mL), 5% aqueous sodium hydroxide (2 x 30 mL), and again with water (2 x 30 mL). The organic layer was separated and dried with sodium sulfate. The solvent was removed under reduced pressure giving 6.0 g of a light yellow liquid product resulting in an overall yield of 67% starting from valproic acid. Test with silver nitrate gave a bright yellow precipitate. UV (CH30H): 216 and 250 nm. 1 H NMR (neat) 6: 4.2-4.5 (t, 2H, CH2-0); 3.1-3.5 (t, 2H, CH2-I); 2.1-2. 7 (m, lH, CH); 1.1-1.9 (m, 8H, propyl CH21s); 0.6-1.1 (m, 6H, CH3). Analysis:(c10H19I02) Calculated: C, 40.28; H, 6.42; I, 42.56; Found: C, 40.37; H, 6.43; I, 42.45. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 3-bromopropanol (11) A large excess of thionyl chloride (9.6 g, 80 mmol), freshly distilled from linseed oil, was added at once to naproxen (2.3 g, 10 mmol). The addition was carried out with stirring at 0C. The mixture was slowly warmed to reflux and then heating was continued for an additional 30

PAGE 43

33 min. Dry benzene (3 x 30 mL) was added and then removed under reduced pressure to remove the excess thionyl chloride. The residue was dissolved in a minimum amount of dry benzene and 3-bromo-1-propanol (1.4 g, 10 mmol) was slowly added with stirring at room temperature. The solution was refluxed for 40 min and then allowed to cool to room temperature. Transparent crystals formed in the dark brown residue. These were vacuum filtered under nitrogen and washed with cold anhydrous ether. The compound was dried under nitrogen and recrystallized from acetonitrile. The compound weighed 2.1 g, giving an overall yield of 60% from naproxen; mp 52-54C. UV (CH30H): 226, 264 and 332 nm. 1 H NMR (CDC1 3 ) o: 7.0-7.7 (m, 6H, naphthalene protons); 4.0-4.3 (t, 2H, o-cH2); 3.6-3.9 (m, 4H, CH and OCH3); 3.1-3.3 (t, 2H, Br-CH2); 1.8-2.2 (p, 2H, CH2); 1.4-1.6 (d, 3H, CH3). Analysis:(c17H19Br03) Calculated: C, 58.13; H, 5.45; Found: c, 58,09; H, 5.53. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 2-bromoethanol (12) Naproxen (2.3 g, 10 mmol) was esterified with 2-bromoethanol (1.4 g, 11 mmol), using DCC (2.3 g, 11 mmol) and DMAP (120 mg, 1.0 mrnol). The reaction was carried out in acetonitrile at room temperature. The reaction mixture was stirred for 24 h, at which time the precipitated DCU was filtered. The DCU weighed 2.3 g after it was dried. The

PAGE 44

34 solvent was removed from the filtrate under reduced pressure, giving an oily residue. This was stirred with anhydrous ether until most of the product dissolved. The remaining solid was filtered and discarded. The ether was removed under reduced pressure giving a dry white solid. A small amount of DCU was still present by NMR. The compound was recrystallized from 2-propanol. The white crystalline product was vacuum filtered and dried in a vacuum desiccator. The final product weighed 3.00 g giving an overall yield of 89.0%; mp 61-63C. UV (CH30H): 224, 264 and 332 nm. 1 H NMR (CDC1 3 ) 6: 7.0-7.9 (m, 6H, naphthalene protons); 4.2-4.6 (t, 2H, CH2-o); 3.6-4.1 (m, 4H, CH and CH3-o); 3.2-3.6 (t, 2H, CH2-Br); 1.4-1.8 (d, 3H, CH3). Analysis:(c16H17Br03) Calculated: C, 56.99; H, 5.08; Br, 23.70; Found: C, 57.09; H, 5.09; Br, 23.60. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 2-iodoethanol (13) Naproxen (4.6 g, 20 mmol) was esterified with 2-iodoethanol (3.4 g, 20 mmol), using DCC (4.5 g, 22 mmol) and DMAP (240 mg, 2.0 mmol) in a minimum amount of acetonitrile (260 mL). The reaction was stirred overnight at room temperature. The precipitated DCU (4.4 g) was filtered, washed with cold acetonitrile (10 mL) and air dried. The reaction solvent was removed under reduced pressure, yielding an oily residue. The product was dissolved in ether (300 mL) and decanted from any undissolved material. The ether was dried

PAGE 45

35 with magnesium sulfate and then filtered. The ether was removed under reduced pressure giving a thick, light colored oil which crystallized on standing at room temperature. The light yellow product was recrystallized twice from 2-propanol giving a nearly white final product. The final compound weighed 5.0 g, giving a 65% overall yield; mp 51-52C. UV (CH30H): 224, 264 and 332 nm. 1 H NMR (CDC1 3 ) 6: 7.0-7.8 (m, 6H, naphthalene protons); 4.1-4.5 (t, 2H, CH2-o); 3.6-4.1 (m, 4H, CH and CH3-0); 3.0-3.4 (t, 2H, CH2-I); 1.4-1.7 (d, 3H, CH3). Analysis:(c16H17I03) Calculated: C, 50.02; H, 4.46; I, 33.03; Found: C, 50.14; H, 4.46; I, 32.96. (+)-6-methoxy-a-methyl-2-napthaleneacetic acid, ester with chloromethanol (14) Previously prepared chloromethylchlorosulfate92 (1.9 g, 12 mrnol) in methylene chloride (5 mL) was added dropwise to a mixture of naproxen (2.3 g, 10 mmol), sodium bicarbonate (3.2 g, 3.8 mmol), tetrabutylammonium hydrogen sulfate (0.68 g, 2.0 mmol) in water and methylene chloride (10 mL each). The biphasic mixture was stirred at room temperature for one hour after the addition of the chloromethylchlorosulfate. The reaction was followed by TLC, hexanes:chloroform (3:2); Rf=0.35 for the product. The two phases were separated and the aqueous layer was extracted with methylene chloride (10 mL). The organic layers were combined and washed with water (20 mL), and dried over magnesium sulfate. The methylene chloride was removed under reduced pressure giving a light

PAGE 46

36 yellow oil which solidified on standing. The white solid was dissolved in hot petroleum ether and filtered to remove any unreacted naproxen. The solvent was removed under reduced pressure and the solid product was recrystallized from 2-propanol. The final product weighed 1.9 g, giving a 67% yield. Compound~ was recovered as beautiful white crystalline plates; mp 67-68C. UV (CH30H): 224, 266 and 332 nm. 1 H NMR (CDC1 3 ) cS: 7.0-7.8 (m, 6H, naphthalene protons); 5.5-5.8 (s, 2H, CH2); 3. 7-4.1 (m, 4H, CH and CH3 0 ) ; 1 4-1 7 ( d 3 H CH 3 ) Analysis:(c15H 15Cl03) Calculated: C, 64.64; H, 5.42; Cl, 12.72; Found: C, 64.75; H, 5.45; Cl, 12.65. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with N-(2-hydroxyethyl)nicotinamide (15) Naproxen (2.30 g, 10.0 mrnol) was coupled with compound 5 (1.71 g, 10.0 mrnol) using DCC (2.30 g, 11.0 mrnol) and DMAP (122 mg, 1.00 mrnol) in acetonitrile (150 mL). The reaction was stirred at room temperature for 48 h. The precipitated DCU was filtered, rinsed with acetonitrile and dried to a weight of 2.3 g. The solvent was removed under reduced pressure and the residual clear oil was stirred with anhydrous ether. The resulting white solid was vacuum filtered, washed with ether and air dried. The crude product weighed 2.80 g. The compound was recrystallized from 2-propanol. The final product was filtered, washed with 0.5% aqueous sodium bicarbonate, water, and finally

PAGE 47

37 with ether. The compound was dried in a desiccator over P 2o5 The recrystallized material weighed 2.40 g resulting in an overall yield of 63.4%; mp 79-82C. UV (CH30H): 226, 264 and 332 nm. 1e NMR (CDCl3) 6: 8.8-9.0 (bs, lH, pyridine H-2); 8.5-8.8 (d, lH, pyridine H-6); 7.0-8.0 (m, 9H, pyridine H-4 and H-5, naphthalene protons, and NH); 4.1-4.4 (t, 2H, CH2-o); 3.6-4.0 (m, 4H, CH and CH3-o); 3.4-3.7 (t, 2H, CH2-N); 1.4-1.7 (d, 3H, CH3). Analysis:(C22H22N204) Calculated: C, 69.83; H, 5.86; N, 7.40; Found: C, 69.92; H, 5.88; N, 7.39. (+)-6-Methoxy-a-methyl-2-naphthalene~etic acid, ester with N-(3-hydroxypropyl)nicotinamide (16) A reaction of naproxen (2.30 g, 10.0 mmol) and compound 7 (1.80 g, 10.0 mmol) was carried out in acetonitrile, using DCC (2.26 g, 11.0 mmol) and DMAP (122 mg, 1.00 mmol) as a coupling agent. After 24 hours the reaction mixture was filtered and the collected DCU was washed with acetonitrile and air dried. The DCU weighed 2.1 g. The solvent was removed from the filtrate under reduced pressure, giving an oily residue. By TLC the oil showed a small amount of a DCU adduct but almost none of the unreacted acid. Column chromatography was done using Mallinckrodt silica gel (100-200 mesh, 60A special) and a mobile phase of chloroform: tetrahydrofuran (4:1). The product was a clear oil which crystallized overnight in a vacuum desiccator. The white

PAGE 48

38 crystalline material was analytically pure. The compound weighed 2.60 g, giving a 66.7% yield; mp 72-75C. UV (CH30H): 224, 264 and 332 nm. 1 H NMR (CDC1 3 ) o: 8.9-9.1 (bs, lH, pyridine H-2); 8.5-8.8 (d, lH, pyridine H-6); 7.9-8.2 (d, lH, pyridine H-4); 7.5-7.8 (m, 3H, naphthalene protons); 7.0-7.5 (m, SH, pyridine H-5, NH and naphthalene protons); 4.0-4.3 (t, 2H, CH2 0); 3.6-4.0 (m, 4H, CH3-o and CH); 3.1-3.5 (q, 2H, CH2-N); 1.6-2.1 (p, 2H, CH2); 1.4-1. 7 (d, 3H, CH3). Analysis:(C2 3H24N2D4) Calculated: C, 70.39; H, 6.16; N, 7.14; Found: C, 70.46; H, 6.18; N, 7.09. 1-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-39~cetic acid, ester with N-(2-hydroxyethyl)nicotinamide (17) A reaction of indomethacin (1.79 g, 5.00 mmol) and compound 2. (0.830 g, 5.00 mmol) was carried out, using DCC (1.10 g, 5.50 mmol) as the coupling agent and acetonitrile as the solvent. The first two reactants were dissolved completely and the solution was then cooled to 0C. The DCC was added and the mixture was stirred overnight. The reaction was allowed to continue for 48 h. The precipitated DCU (1.2 g) was removed by vacuum filtration. The solvent was removed from the filtrate under reduced pressure leaving an oily residue. The product was solidified by stirring with anhydrous ether. It was filtered, air dried and recrystallized from ethanol/ether. The final product was vacuum filtered, washed with ether, and air dried. The product weighed 1.65 g, giving a 65.2% yield; mp 123-25C.

PAGE 49

39 UV (CH30H): 222 and 320 nm. 1 H NMR (CDC1 3 ) 6: 8.7-8.9 (bs, lH, pyridine H-2); 8.6-8.8 (d, lH, pyridine H-6); 7. 7-8.0 (bd, lH, pyridine H-4); 7.2-7.7 (m, SH, phenyl protons and pyridine H-5); 6.4-7.0 (m, 4H, indole protons and NH); 4.2-4.4 (t, 2H, CH2-o); 3.5-3.9 (m, 7H, CH3-o, CH2-N, CH2); 2.2-2.4 (s, 3H, CH3). Analysis: ( c2 7 H24clN305 l/2 H20) Calculated: C, 62.98; H, 4.89; N, 8.16; Found: C, 63.27; H, 4.91; N, 8.49. 1-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid, ester with N-(3-hydroxypropyl)nicotinamide (18) A reaction of indomethacin (3.58 g, 10.0 mmol) and compound.]_ (1.80 g, 10.0 mmol) with DCU (2.26 g, 11.0 mmol) and DMAP (0.12 g, 1.0 mmol) was carried out in acetonitrile at room temperature. The reaction mixture was stirred for 24 hand then filtered to remove the precipitated DCU. The solvent was removed under reduced pressure after 2.3 g of DCU were recovered (94% of theoretical amount). The product was a white solid which was recrystallized from 2-propanol/ether, using the mixed solvent technique. The compound was filtered, washed with ether and air dried. The NMR showed some traces of DCU. Therefore, the material was recrystallized from 2-propanol only. The final product was filtered, washed with ether and dried in a vacuum desiccator. The compound weighed 3.60 g giving an overall yield of 69.2%; mp 122-123C.

PAGE 50

40 UV (CH30H): 222 and 320 nm. 1 H NMR (CDC1 3 ) 6: 8.9-9.2 (bs, lH, pyridine H-2); 8.6-8.8 (bd, lH, pyridine H-6); 7.9-8.3 (bd, lH, pyridine H-4); 7.2-7.9 (m, 6H, phenyl protons, pyridine H-5, and NH); 6.5-7.1 (m, 3H, indole protons); 4.1-4.5 (t, 2H, CH2-o); 3.2-4.0 (m, 7H, CH3-0, CH2-N, and CH2-CO); 2.2-2.5 (s, 3H, CH3). Analysis:(c28H26clN305) Calculated: C, 64.68; H, 5.04; Cl, 6.82; N, 8.08; Found: C, 64.55; H, 5.10; Cl, 6.79; N, 8.03. 3-Carbamoyl-1-(2-hydroxyethyl)pyridiniurn iodide, ester with 2-propylpentanoic acid (19) A mixture of nicotinamide (1.22 g, 10.0 rnrnol) and compound 10 (3.28 g, 11.0 rnrnol) was dissolved in dimethylformamide (50 mL). The solution was heated to reflux for 3 hand then cooled. The solvent was removed under reduced pressure, and the brown oily residue was stirred with ether (60 mL) for 30 min, giving a yellow powder. The ether was decanted and a fresh portion of ether (50 mL) was added. The crude product was vacuum filtered under nitrogen. It was then recrystallized from 2-propanol/ether, giving 3.50 g of light, cream colored crystalline plates. The final product was obtained in 83.3% yield; mp 113-114C. UV (CH30H): 222 and 268 nm. 1 H NMR (d6 -DMS0) 6: 9.5-9.7 (bs, lH, pyridine H-2); 9.2-9.5 (bd, lH, pyridine H-6); 8.9-9.2 (bd, lH, pyridine H-4); 8.0-8. 7 [m(8.0-8.3, bs, lH, NH), (8.1-8.5, m, lH, pyridine H-5), (8.4-8.7, bs, lH, NH)]; 4.9-5.3 (m, 2H, CH2-N); 4.3-4.8 (m, 28, CH2-0); 2.0-2.8 (m, lH, CH); 0.9-1.7 (m, 8H, propyl CH21s); 0.4-1.1 (m, 6H, CH3).

PAGE 51

41 Calculated: C, 45.73: H, 6.00: I, 30.19: N, 6.66: Found: C, 45.70: H, 6.04: I, 30.10: N, 6.65. 3-Carbamoyl-1-hydroxymethypyridinium chloride, ester with (+)-6-methoxy-a-methyl-2-naphthaleneacetic acid (20) Nicotinamide (0.24 g, 2.0 rnmol) was dissolved in acetone. The chloromethylester of naproxen, compound 14 (0.56 g, 2.0 mmol + 1% excess), was then added and the solution was stirred at reflux for 48 h. The precipitated white product was vacuum filtered, washed with ether and dried. The dried compound weighed 410 mg. The mother liquor was reduced to dryness and recrystallized from 2-propanol giving a second crop of crystals weighing 80 mg. The yield was 0.49 g or 61% overall: mp 227-229C. UV (CH30H): 226, 268 and 332 nm. 1 H NMR (d6 -DMSO/CD 3 0D) 6: 9.5-9.6 (bs, lH, pyridine H2): 9.0-9.3 (m, 2H, pyridine H-6 and H-4): 8.1-8.4 (m, lH, pyridine H-5): 7.6-7.9 (m, 4H, naphthalene protons and NH): 7.1-7.5 (m, 4H, naphthalene protons and NH): 6.5-6.6 (s, 2H, CH2): 3.9-4.3 (q, lH, CH): 3.8-3.9 (s, 3H, CH3-0): 1.5-1.6 ( d 3 H CH 3 ) Analysis:(c21H21ClN204) Calculated: C, 62.92: H, 5.28: Cl, 8.84: N, 6.99: Found: C, 62.73: H, 5.30: Cl, 8.94: N, 6.94. 3-Carbamoyl-1-(3-hydroxypropyl)pyridinium bromide, ester with (+)-6-methoxy-a-methyl-2-naphthaleneacetic acid (21) Naproxen (2.5 g, 11 mmol) and compound_! (2.6 g, 10 mmol) were dissolved in a minimum amount of dimethyl-

PAGE 52

42 formamide (200 mL), to which DMAP (130 mg, 1.1 mmol) and DCC (2.3 g, 11 mmol) were added. The solution was stirred at room temperature for 2 days. The precipitated DCU was vacuum filtered, washed and air dried. The collected DCU weighed 2 g, which represented 80% of the theoretical yield. The dimethylformamide was removed from the filtrate under reduced pressure. This gave an oily residue which solidified on standing at room temperature. The solid was powdered and washed well with anhydrous ether. The offwhite powder was filtered, washed with an additional portion of ether and air dried. It was recrystallized from ethanol, and upon partial cooling the mother liquor was decanted from the reddish brown solid that first appeared. Additional cooling of the mother liquor, along with brief scratching, gave a large quantity of fluffy tan crystals. These were once again recrystallized from ethanol. The final product was filtered, washed with ether and dried in a vacuum desiccator. The compound weighed 4.0 g resulting in an 85% yield: mp 152-154C. UV (CH30H): 224, 266 and 332 nm. 1 H NMR (d6-DMS0) o: 9.5-9.8 (bs, lH, pyridine H-2): 9.1-9.4 (bd, lH, pyridine H-6): 8.9-9.2 (bd, lH, pyridine H4): 8.5-8.8 (bs, lH, NH): 8.0-8.5 (m, 2H, pyridine H-5 and NH): 7.0-8.0 (m, 6H, naphthalene protons): 4.6-5.0 (t, 2H, CH2-N): 4.0-4.4 (t, 2H, CH2-o): 3.6-4.0 (m, 4H, CH and CH3 -0): 2.1-2.6 (m, 2H, CH2): 1.3-1.6 (d, 3H, CH3).

PAGE 53

43 Calculated: C, 58.36; H, 5.32; Br, 16.88; N, 5.92; Found: C, 58.23; H, 5.34; Br, 16.95; N, 5.86. 3-[(2-Hydroxyethyl)carbamoyl)-1-methylpyridinium iodide, ester with (+)-6-methoxy-a-methyl-2-naphthaleneacetic acid (22) The quaternization of the naproxen ester, compound 15 (1.0 g, 2.6 mmol), was carried out using methyl iodide (2.3 g, 16 mmol) in acetone (45 mL). The solution was heated to reflux for 20 h. Methyl iodide (1.1 g, 8.0 mmol) was again added to the reaction flask. The precipitated product was filtered after an additional 4 h of reaction time. The offwhite powder was dried. The material weighed 2.2 g and was found to be analytically pure without recrystallization. The solvent was removed from the acetone filtrate and the residue was solidified with anhydrous ether. The resulting dark yellow powder was dissolved in water and washed with ether (4 x 30 mL). The water was then removed under vacuum giving 0.2 g of a lighter yellow powder, although still much more highly colored than the precipitated product. The overall yield of the reaction was 93%; mp 169-170C. UV (CH30H): 226, 266 and 332 nm. 1 H NMR (d6 -DMS0) 6: 9.3-9.5 (bs, lH, pyridine H-2); 9.0-9.3 (m, 2H, pyridine H-6 and NH); 8.7-9.0 (bd, lH, pyridine H-4); 8.1-8.5 (m, lH, pyridine H-5); 7.0-7.9 (m, 6H, naphthalene protons); 4.3-4.5 (s, 3H, CH3-N); 4.1-4.4 (+, 2H, CH2-0); 3.4-4.1 (m, 6H, CH, CH3-o and CH2-N); 1.4-1.7 ( d, 3 H, CH 3)

PAGE 54

44 Calculated: c, 53.09; H, 4.84; I, 24.39; N, 5.38; Found: C, 52.98; H, 4.85; I, 24.29; N, 5.34. 3-[(3-Hydroxypropyl)carbamoyl)-1-methylpyridinium iodide, ester with (+)-6-methoxy-a-methyl-2-naphthaleneacetic acid (23) The ester of naproxen, compound 16 (2.3 g, 5.9 mmol), was dissolved in acetone. Methyl iodide (4.0 g, 29 mmol) was added and the solution was heated to reflux for 48 h. The reaction mixture was cooled and then taken to an oily foaming residue upon removal of the solvent under reduced pressure. It was then dried further on a vacuum pump. The product was dissolved in acetone (2 x 25 mL) and each time the solvent was removed under reduced pressure. The first process yielded a foam and the second gave a powdery yellow solid. This was then completely dried using a vacuum pump. The compound weighed 2.9 g which resulted in a 93% yield; mp 100-102C. This product required no further purification. UV (CH30H): 224, 266 and 332 nm. 1 H NMR (d6-DMS0) 6: 9.4-9.5 (s, lH, pyridine H-2); 8.8-9.3 (m, 3H, pyridine H-6 and H-4 and NH); 8.2-8.4 (m, lH, pyridine H-5); 7.1-8.0 (m, 6H, naphthalene protons); 4.4-4.5 (s, 3H, CH3-N); 4.1-4.3 (t, 2H, CH2-o); 3.8-4.1 (m, 4H, CH and CH3-0); 3.2-3.5 (t, 2H, CH2-N); 1.7-2.1 (p, 2H, CH2); 1.4-1.7 (d, 3H, CH3).

PAGE 55

45 Calculated: C, 53.94; H, 5.09; I, 23.75; N, 5.24; Found: C, 54.00; H, 5.13; I, 23.81; N, 5.22. 3-[(2-Hydroxyethyl)carbamoyl]-1-methylpyridinium iodide, ester with l-(p-chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid (24) The quaternization of compound 17 (0.50 g, 1.0 mmol) was carried out in acetone, using methyl iodide (1.7 g, 12 mmol). The reaction was refluxed overnight. The solvent was removed under reduced pressure and a yellow solid was obtained. The product was recrystallized using ethanol and a very small amount of ether. Small mold-like crystals were obtained which were light yellow in color. The reaction gave 0.43 g or a 66% yield of the purified material; mp 178-1 79 C. UV (CH30H): 220 and 320 nm. 1 H NMR (d6 -DMS0) o: 9.3-9.5 (bs, lH, pyridine H-2); 9.1-9.3 (m, 2H, pyridine H-6 and NH); 8.8-9.0 (bd, lH, pyridine H-4); 8.1-8.4 (m, lH, pyridine H-5); 7.6-7.8 (s, 4H, phenyl protons); 6.6-7.2 (m, 3H, indole protons); 4.4-4.6 (s, 3H, CH3-N); 4.2-4.5 (t, 2H, CH2-0); 3.5-4.0 (m, 7H, CH3 -0, CH2-co and CH2-N); 2.2-2.3 (s, 3H, CH3). Analysis:(c28 H27clIN305) Calculated: C, 51.91; H, 4.20; Cl, 5.47; I, 19.59; N, 6.48; Found: C, 51.80; H, 4.24; Cl, 5.41; I, 19.46; N, 6.42.

PAGE 56

46 3-[(3-Hydroxypropyl)carbamoyl)-1-methylpyridinium iodide, ester with l-(p-chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid (25) The ester of indomethacin, compound 18 (3.1 g, 6.0 mmol), was dissolved in a minimum of acetone. A twofold excess of methyl iodide (2.5 g, 18 mmol) was added to the solution which was then heated to reflux. The reaction was continued for 24 h. Approximately one-half of the reaction solvent was removed under reduced pressure. The precipitated, light yellow product was filtered, washed with ether and dried in a vacuum desiccator. The compound weighed 3.9 g giving a 97% yield; mp 168-169C. The compound was analytically pure without recrystallization. UV (CH30H): 222 and 320 nm. 1 H NMR (d6-DMS0) cS: 9.3-9.5 (bs, lH, pyridine H-2); 8.8-9.2 (m, 3H, pyridine H-6 and H-4 and NH); 8.2-8.4 (m, lH, pyridine H-5); 7.5-7.8 (s, 4H, phenyl protons); 6.6-7.2 (m, 3H, indole protons); 4.3-4.5 (s, 3H, CH3-N); 4.0-4.3 (t, 2H, CH2-0); 3.6-4.0 (bs, SH, CH3-o and CH2-CO); 3.2-3.5 (t, 2H, CH2-N); 2.2-2.4 (s, 3H, CH3); 1. 7-2.1 (p, 2H, CH2). Analysis: (c29 H 29c1IN305 ) Calculated: C, 52.62; H, 4.42; Cl, 5.36; I, 19.17; N, 6.35; Found: C, 52.55; H, 4.46; Cl, 5.29; I, 19.19; N, 6.34. 2-Propylpentanoic acid, ester with l,4-dihydro-1-(2-hydroxyethyl)nicotinamide (26) Compound 19 (420 mg, 1.0 mmol) was dissolved in icecold degassed, deionized water (50 mL). Sodium bicarbonate (370 mg, 4.0 mmol) and sodium dithionite (700 mg, 4.0 mmol) were added with stirring. Nitrogen gas was bubbled through

PAGE 57

47 the yellow solution about 30 min. The reaction mixture was extracted with ether (6 x 25 mL) until no additional color was transferred to the organic layer. The combined ether extracts were washed with H 2o (50 mL) and dried over magnesium sulfate. The solution was decanted away from the drying agent and the solvent was removed under reduced pressure. To the oily residue, ether was added and removed (10 x 5 mL) on a vacuum pump. A foam was obtained which returned to an oil upon exposure to the atmosphere. UV (CH30H): 216 and 354 nm. 1 H NMR (CDC13 ) o: 6.9-7.1 (bs, lH, dihydropyridine H-2); 5.6-6.0 (m, 3H, NH2 and dihydropyridine H-6); 4.5-4.9 (m, lH, dihydropyridine H-5); 4.0-4.4 (t, 2H, CH2-o); 3.2-3.6 (m, 2H, CH2-N); 2.9-3.2 (bs, 2H, dihydropyridine H-4); 2.1-2.6 (m, lH, CH); 0.9-1.7 (m, 8H, propyl CH21s); 0.5-1.1 (m, 6H, CH3). Analysis: (c16H26N 2o3H20) Calculated: C, 61.51; H, 9.04; N, 8.97 Found: C, 61.96; H, 8.69; N, 8.53; c, 61.49; H, a.49; N, a.so. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 1,4-dihydro-l-hydroxymethylnicotinamide (27) The quaternary ester of naproxen, compound 20 (200 mg, 0.5 mM), was dissolved in a mixture of acetonitrile:2-propanol, 6:1. An excess of l-benzyl-l,2-dihydronicotinamide94 (200 mg, 0.8 mmol) was added and the reaction was allowed to proceed for one hour at room temperature. The reaction was carried out under nitrogen gas. When the reaction was

PAGE 58

48 complete, the solvent was removed under reduced pressure giving an oily orange foam. This was dissolved in methylene chloride and then filtered to remove the quaternary side product. The solvent was again removed giving an oily residue. The product showed traces of contamination with quaternary material, by NMR. The oil was dissolved in chloroform and passed down a short column of neutral alumina. The appropriate fraction was collected and the solvent removed under reduced pressure. The oily residue was triturated with anhydrous ether. The ether was removed using a vacuum pump. This gave a hygroscopic yellow foam as the product. However, NMR showed the final compound had partially hydrolyzed. UV (CH30H): 228, 264, 334 and 346 nm. 1 H NMR (CDC1 3 ) 6: 7.0-7.8 (m, 7H, naphthalene protons and dihydropyridine H-2); 5.5-5.9 (m, lH, dihydropyridine H-6); 5.1-5.5 (bs, 2H, NH2); 4.6-5.0 (m, lH, dihydropyridine H-5); 4.4-4.6 (s, 2H, CH2); 3.8-4.0 (s, 3H, CH3-o); 3.3-3.7 (q, lH, CH); 3.0-3.3 (bs, 2H, dihydropyridine H-4); 1.4-1.7 (d, 3H, CH3). Analysis:(c21 H 22 N 204H20) Calculated: C, 65.61; H, 6.29; N, 7.29; Found: C, 65.50; H, 6.68; N, 7.99. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with l,4-dihydro-1-(3-hydroxypropyl)nicotinamide (28) The quaternary ester, compound 21 (470 mg, 1.0 mmol), was dissolved in degassed water (70 mL) and sodium dithionite (520 mg, 3.0 mmol) and sodium bicarbonate (420 mg, 5.0

PAGE 59

49 mmol) were added at once. A layer of ether was added and the reaction was allowed to proceed under a nitrogen atmosphere for 60 min. The reaction mixture was extracted with methylene chloride (4 x 30 mL). The extracts were combined and washed with water (50 mL), and then dried over magnesium sulfate. The solvent was removed under reduced pressure giving an oily foam. The product was repeatedly redissolved and dried on a vacuum pump until the compound was obtained as a light yellow solid foam. The final product weighed 180 mg resulting in a 46% yield; mp 42-46C. UV (CH30H): 224, 264, 334 and 356 nm. 1 H NMR (CDC1 3 ) 6: 6.8-7.8 (m, 7H, naphthalene protons and dihydropyridine H-2); 5.2-5.5 (m, 3H, dihydropyridine H6 and NH2); 4.3-4.7 (m, lH, dihydropyridine H-5); 3.7-4.3 (m, 6H, CH2-o, CH3-o and CH); 2. 7-3.2 (m, 4H, dihydropyridine H-4 and CH2-N); 1.3-2.0 (m, SH, CH2 and Calculated: C, 66.97; H, 6.84; N, 6.79; Found: C, 67.02; H, 6.57; N, 6.74. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 1,4-dihydro-N-(2-hydroxyethyl)-l-methylnicotinamide (29) The quaternary salt compound 22 (780 mg, 1.5 mmol) was dissolved in degassed, deionized water (200 mL) and acetonitrile (10 mL). Sodium dithionite (780 mg, 4.5 mmol) and sodium bicarbonate (630 mg, 7.5 mmol) were combined, and added to the solution at room temperature. The reaction was continued for 1 h, while nitrogen gas was slowly bubbled

PAGE 60

50 through the solution. The partially precipitated product was extracted repeatedly with ether (8 x 30 mL). The extracts were combined, washed with water (25 mL) and dried over magnesium sulfate. The drying agent was filtered and the solvent was removed from the filtrate under reduced pressure. The oily residue was dissolved in methylene chloride (3 x 5 mL) and removed under reduced pressure. The resulting foam was rinsed with anhydrous ether (3 mL) and the solvent was removed under vacuum. The final product weighed 390 mg, giving a 66% yield. The hygroscopic solid foam was stored under nitrogen at -100C. UV (CH30H): 224, 264, 334 and 358 nm. 1 H NMR (CDC1 3 ) o: 7.0-7.9 (m, 6H, naphthalene protons); 6.7-7.0 (bs, lH, dihydropyridine H-2); 5.1-5.7 (m, 2H, dihydropyridine H-6 and NH); 4.2-4.6 (m, lH, dihydropyridine H-5); 4.0-4.3 (t, 2H, CH2-o); 3.6-4.0 (m, 4H, CH3-o and CH); 3.3-3.7 (t, 2H, CH2-N): 2.6-3.0 (s, 3H, CH3-N); 2.4-2.6 (bs, 2H, dihydropyridine H-4); 1.3-1. 7 (d, 3H, CH3). Analysis:(c23H26N204H20) Calculated: C, 66.97; H, 6.84; N, 6. 79; Found: C, 67.11; H, 6.69; N, 6.63. (+)-6-Methoxy-a-methyl-2-naphthaleneacetic acid, ester with 1,4-dihydro-N-(3-hydroxypropyl)-l-methylnicotinamide (30) Compound 23 (530 mg, 1.0 mmol) was dissolved in degassed water (150 mL). This was washed with ether (50 mL), separated and the organic phase was discarded. The aqueous layer was transferred to a flask and sodium dithionite (520

PAGE 61

51 mg, 3.0 mmol) and sodium bicarbonate (420 mg, 5.0 mmol) were added at once. The reaction was run under a nitrogen atmosphere, at room temperature. The reaction was continued for 30 min, and then extracted with methylene chloride (5 x 40 mL) until no more yellow color transferred into the organic layer. The methylene chloride was combined and was back extracted with water (40 mL) and then dried with magnesium sulfate. The crude product was an orange oil. This showed traces of the unreduced starting material by both TLC and NMR. The oil was dissolved in CHC13 and quickly passed down a short column of neutral alumina. Chloroform was also used as the eluting solvent. The fraction which contained the dihydro compound was collected and the solvent removed under reduced pressure. The product was then dried on a vacuum pump yielding a yellow hygroscopic foam. The product weighed 300 mg giving an overall yield of 73%. UV (CH30H): 224, 264, 334 and 354 nm. 1 H NMR (CDC1 3 ) o: 7.0-7.9 (m, 6H, naphthalene protons); 6.9-7.1 (bs, lH, dihydropyridine H-2); 5.3-5.8 (m, 2H, dihydropyridine H-6 and NH); 4.4-4.8 (m, lH, dihydropyridine H-5); 4.0-4.3 (t, 2H, CH2-o); 3.6-4.0 (m, 4H, CH3-o and CH); 3.1-3.4 (t, 2H, CH2-N); 2.8-3.l (bs, 2H, dihydropyridine H-4); 2.8-3.0 (s, 3H, CH3-N); 1.4-2.0 (m, SH, CH2 and CH3 )

PAGE 62

52 Calculated: C, 69.04: H, 7.00: N, 6.71: Found: C, 69.40: H, 7.23: N, 6.39. l-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid, ester with 1,4-dihydro-N-(2-hydroxyethyl)-l-methylnicotinamide (31) The indomethacin quaternary carrier, compound 24 (140 mg, 0.22 mmol), was dissolved in a minimum amount of water:acetonitrile (8:2). The water had been bubbled with nitrogen for 20 min previous to its use. Sodium bicarbonate (91 mg, 1.1 mmol) and sodium dithionite (110 mg, 0.65 mmol) were added to the solution while stirring at 0C. The solution was then allowed to warm to room temperature. The reaction was continued for about 1 h. Some of the product had precipitated during the reaction. This was dissolved in ethyl ether. The water layer was extracted several times with ether until no more yellow color transferred to the organic layer. The ether portions were combined and dried with magnesium sulfate, filtered and the ether was removed under reduced pressure. The resulting oil was dissolved in acetone and the solvent was removed (2 x 10 mL) under reduced pressure to form a dry foam. The final product weighed 92 mg. The yield was 82%: mp 60-65C. UV (CH30H): 220 and 332 nm. 1 H NMR (CDC1 3 ) cS: 7.3-7.9 (q, 4H, phenyl protons): 6.5-7.1 (m, 4H, indole protons and dihydropyridine H-2): 5.5-5.8 (m, lH, dihydropyridine H-6): 5.2-5.4 (bs, lH, NH): 4.4-4.8 (m, lH, dihydropyridine H-5): 4.1-4.4 (t, 2H, CH2 -

PAGE 63

53 O); 3.8-4.0 (s, 3H, CH3-o); 3.6-3.8 (s, 2H, CH2-co); 3.4-3.7 (m, 2H, CH2-N); 2.8-3.0 (s, 3H, CH3-N); 2.6-2.9 (bs, 2H, dihydropyridine H-4); 2.3-2.5 (s, 3H, CH3). Analysis:(c28 H28clN3 o 5H2o) Calculated: C, 60.27; H, 5.78; Cl, 6.35; N, 7.53; Found: C, 60.14; H, 5.73; Cl, 6.11; N, 7.24. 1-(p-Chlorobenzoyl)-5-methoxy-2-methylindole-3-acetic acid, ester with 1,4-dihydro-N-(3-hydroxypropyl)-1-methylnicotinamide (32) The quaternary ester, compound 25 (660 mg, 1.0 mmol), was dissolved in degassed, deionized water (80 mL) and washed once with ether (50 mL). The organic layer was separated and acetonitrile (5 mL) was added to the aqueous phase in order to dissolve any remaining starting material. A mixture of sodium dithionite (520 mg, 3.0 mmol) and sodium bicarbonate (420 mg, 5.0 mmol) was added to the solution. Ether (30 mL) was added and the reaction was allowed to proceed for 60 min under a nitrogen atmosphere. The reaction mixture was extracted with methylene chloride (4 x 40 mL), the extracts combined, and then washed with degassed water (40 mL). The organic phase was separated and dried using magnesium sulfate. The drying agent was gravity filtered and the solvent removed under reduced pressure. The dark orange oil was redissolved in methylene chloride (3 x 5 mL) and the solvent removed under reduced pressure, and then dried using a vacuum pump. The product was a yellow-orange solid foam. This crude material showed traces of quaternary starting material which were removed by dissolving in

PAGE 64

54 chloroform and passing it down a short column of neutral alumina. Chloroform was also used as the eluting solvent. The purified product was recovered, the solvent removed and the resulting oil was dissolved in methylene chloride. Again the solvent was removed, under reduced pressure and the resulting foam was triturated with anhydrous ether. This was removed using a vacuum pump, giving 300 mg of a dry, light yellow foam. The final product gave an overall yield of 56%; mp 52-56C. UV (CH30H): 212 and 332 nm. 1 H NMR (CDC1 3 ) o: 7.3-7.8 (q, 4H, phenyl protons); 6.5-7.1 (m, 4H, indole protons and dihydropyridine H-2); 5.2-5.9 (m, 2H, dihydropyridine H-6 and NH); 4.5-4.8 (m, lH, dihydropyridine H-5); 3.9-4.3 (t, 2H, CH2-o); 3.7-3.9 (s, 3H, CH3-0); 3.5-3.8 (s, 2H, CH2-CO); 3.2-3.5 (t, 2H, CH2-N); 2.9-3.2 (bs, 2H, dihydropyridine H-4); 2.8-3.0 (s, 3H, CH3 -N); 2.2-2.5 (s, 3H, CH3); 1.6-2.2 (p, 2H, CH2). Analysis:(c29H30cl/4H2oN3 o 5 ) Calculated: C, 63.38; H, 5.78; Cl, 6.45; N, 7.65; Found: C, 63.42; H, 5.78; Cl, 6.53; N, 7.65. High Pressure Liquid Chromatography systems Mobile phase systems were developed in order to carry out all in vitro and in vivo analyses. The quaternary compounds (20-25) were analyzed utilizing a mobile phase that was one-half organic and one-half aqueous. The organic portion was entirely acetonitrile. However, the aqueous portion was composed of both monobasic potassium phosphate

PAGE 65

55 (50 mM) and water. The ratio was adjusted for each quaternary compound in order to have its retention time between 5 and 6 min. The amount of potassium phosphate used varied between 15 and 50% of the entire mobile phase. Naproxen and indomethacin could be analyzed for using these same systems. Their retention times were not greatly affected by the changes in buffer concentration, as long as the acetonitrile:aqueous ratio remained constant. This allowed both the quaternary compound and the free parent drug to be analyzed for simultaneously. The dihydropyridine compounds (27-32) required analysis using alternate mobile phase systems. The compounds were chromatographed using various ratios of acetonitrile: water. Ratios as high as 80% organic were used in order to obtain retention times in the 4 to 6 min range. All of these analyses, involving the quaternary compounds, the free parent drugs and the dihydropyridine-CDS were carried out using a Toyasota column (see Materials and Methods). The hydrophilic nature of the hydroxyalkylpyridinium carriers did not allow them to be retained in any of the previously mentioned systems. A reversed phase C-8 column (see Materials and Methods) was chosen for the work involving the pyridinium carriers. A mobile phase of acetonitrile, water, and monobasic potassium phosphate (50 mM): (40:50:10) was used. This gave retention times of 4 to 6 min. The flow rate used for all of the HPLC systems was 1.0 mL/min.

PAGE 66

56 Chemical Stability Stability of Quaternary Drug-Carrier Compounds (20-25) in pH 7.4 Phosphate Buffer Phosphate buffer (pH 7.4) was equilibrated in a water bath at 37C. A 5 x 10-3 M stock solution was prepared for each compound, using DMSO as the solvent. Ten microliters of the stock solution were added for each milliliter of buffer used in a given experiment. This resulted in a 5 x 10-5 M solution. Aliquots of 100 L were withdrawn at various time intervals and pipetted into 400 uL portions of ice-cold acetonitrile. The sample were centrifuged for 3 min at 10,000 rpm. The rate of ester hydrolysis was determined by measuring the disappearance of the quaternary compound and the appearance of the free drug after hydrolysis. This was accomplished using high pressure liquid chromatography. Stability of Dihydropyridine CDS Compounds (27-32) in pH 7.4 Phosphate Buffer Phosphate buffer was equilibrated in a water bath at 37C. A 5 x 10-3 M stock solution in DMSO was freshly prepared for each compound before use. Ten microliters of the stock solution were added per milliliter of buffer used in each experiment. This resulted in a 5 x 10-SM solution. Aliquots of 100 L were withdrawn at various time points and pipetted into 400 L portions of ice-cold acetonitrile. The samples were centrifuged for 3 min at 10,000 rpm. The rate of disappearance was measured using high pressure liquid chromatography. Appearance of the quaternary form of each

PAGE 67

compound and the free, parent drug was monitored using a second HPLC system. In Vitro Studies 57 Stability of Quaternary Drug-Carrier Compounds (20-25) in 100% Whole Human Blood and 100% Whole Rat Blood Blood was withdrawn from a volunteer shortly before beginning each experiment. It was placed in heparinized tubes and stored on ice until needed. The blood was then incubated at 37C. A 5 x 10-3 M stock solution in DMSO was prepared for each compound before use. Ten microliters of the stock solution were added per milliliter of blood used in a given experiment. Aliquots of 100 L were withdrawn at various time intervals and pipetted into 400 L of ice-cold acetonitrile containing 5% DMSO by volume. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was used to determine the rate of ester hydrolysis by high pressure liquid chromatography. The same procedure was used for fresh rat blood that was withdrawn into a heparinized syringe via heart puncture. The blood was collected just prior to the start of each experiment. Stability of Dihydropyridine CDS Compounds (27-32) in 100% Whole Human Blood and 100% Whole Rat Blood Blood was withdrawn from a volunteer shortly before beginning each experiment. It was placed in heparinized tubes and stored on ice until needed. The blood was incubated at 37C. A 5 x 10-3 M stock solution in DMSO was freshly prepared. Ten microliters of the stock solution

PAGE 68

58 were added for each milliliter of blood used in a given experiment. Aliquots of 200 L were withdrawn at various time points and pipetted into 800 L of ice-cold acetonitrile containing 5% DMSO by volume. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was divided into two equal portions. The first was used to follow the rate of disappearance of the dihydropyridine compound. Later the second portion that had been kept at 0C was used to measure the appearance of the quaternary form of the compound and the free, parent drug. These samples were analyzed using high pressure liquid chromatography. The dihydro compound was monitored using one system and a second HPLC system was required to follow the quaternary compound and the parent drug. The same experimental procedure was used for fresh rat blood that was collected via heart puncture into a heparinized syringe. The blood was collected just prior to beginning each experiment. Stability of Quaternary Drug-Carrier Compounds (20-25) in 20% Rat Brain Homogenate and 20% Rat Kidney Homogenate One gram of freshly obtained rat brain was homogenized with 4 mL of phosphate buffer (pH 7.4). The homogenate was centrifuged at 3,000 rpm for 5 min. The supernatant was removed and incubated in a water bath at 37C. A 5 x 10-3 M stock solution in DMSO was prepared for each compound before its use. Ten microliters of the stock solution were added per milliliter of homogenate used in a particular experiment. Aliquots of 100 L were withdrawn at various time

PAGE 69

59 intervals and pipetted into 400 L of ice-cold acetonitrile. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was used to measure the rate of ester hydrolysis by high pressure liquid chromatography. The same procedure was used for freshly obtained rat kidney. Stability of Dihydropyridine CDS Compounds {27-32) in 20% Rat Brain Homogenate One gram of freshly obtained rat brain was homogenized with 4 mL of phosphate buffer {pH 7.4). The homogenate was centrifuged at 3,000 rpm for 5 min. The supernatant was removed and incubated in a water bath at 37C. A 5 x 10-3 M stock solution in DMSO was prepared for each compound just before its use. Ten microliters of the stock solution was added for each milliliter of homogenate used in a given experiment. Aliquots of 200 L were withdrawn at various time points and pipetted into 800 L of ice-cold acetonitrile. The samples were vortexed for 5 sec and centrifuged at 10,000 rpm for 5 min. The supernatant was divided into two equal portions. The first was used to measure the rate of disappearance of the dihydropyridine compound. This was done by high pressure liquid chromatography. The second portion which had been kept at 0C was used to monitor the appearance of the quaternary compound and the parent drug. This was done via a second HPLC system.

PAGE 70

60 In Vivo Studies Preliminary In Vivo Distribution of Compound 31 in Brain and Blood Male Sprague-Dawley rats weighing 220 to 270 g were anesthetized with Inovar (40 L) intramuscularly. The dihydropyridine (20 mg/kg, 67 mg/mL) in DMSO was given by intrajugular injection using an infusion pump fitted with a glass syringe and a 27 gauge butterfly needle. The animals were sacrificed at various time intervals by decapitation. Trunk blood was collected into heparinized tubes and the brains were removed. Brains were homogenized in 1 mL of water and then extracted with 4 mL of cold acetonitrile. One milliliter of blood was also extracted with 4 mL of acetonitrile. The samples were placed in a freezer for 2 h in order to separate the organic layer. This was used for analysis by HPLC after being filtered through 0.45 m Millipore filters. In Vivo Distribution of Compound 28 Male Sprague-Dawley rats weighing 255 to 280 g were placed in a restraining cage. The dihydropyridine (25 mg/kg, 50 mg/mL) in DMSO was given by tail vein injection. The animals were sacrificed by decapitation. Trunk blood was collected into heparinized tubes. The animals were opened along the midline, and brain, lung, liver, kidney, testis, and fat tissues (1.0 g each) were weighed and then quickly frozen using dry ice. The tissues were homogenized with 1.0 mL of water, using a ground glass pestle and tube. The samples were extracted with 4.0 mL of

PAGE 71

61 acetonitrile by homogenizing again. The separation of the organic layer was expedited by adding 1.0 mL of saturated aqueous sodium chloride solution and homogenizing the mixture once more. The blood samples (1 mL) were extracted and separated in the same manner; however, vigorous shaking was used rather than homogenization. All samples were placed in a freezer for 2 h in order to complete the separation of the organic phase. The acetonitrile layer was removed and filtered through a 0.45 m Millipore filter. The samples were then analyzed using high pressure liquid chromatography. The quaternary compound and the free parent drug were both monitored by ultraviolet detection at 264 run. The mobile phase consisted of acetonitrile, water and aqueous, monobasic potassium phosphate (50 mM); (50:15:35). An attempt was made to detect the dihydropyridine compound using a mobile phase of acetonitrile and water (80:20). However, no measurable amounts remained by the first time point (30 min). Controls were run by injecting DMSO (0.5 mL/kg) as a blank. The animals were sacrificed and tissues were collected in an identical fashion. Standard curves were constructed by adding known amounts of both the quaternary compound 21 and the parent drug, naproxen to blood and tissue blanks. In this way the experimental results could be quantitated and the extent of recovery examined.

PAGE 72

62 In Vivo Distribution of Compound 29 vs an Eguimolar Dose of Naproxen Male Sprague-Dawley rats weighing 300 g were placed in a retaining cage. The dihydropyridine (25 mg/kg, 50 mg/mL) or an equimolar amount of naproxen (15 mg/kg, 29 mg/mL) was given by tail vein injection. In both cases DMSO was used as a vehicle. The animals were sacrificed by decapitation. Trunk blood was collected into heparinized tubes. The brains were removed from the skull. The body was opened along the midline, and lung, liver, kidney, testis and fat tissues (1.0 g each) were removed, weighed and immediately frozen using dry ice. The tissues were homogenized with 1.0 mL of water, using a ground glass pestle and tube. The samples were homogenized again after adding 4.0 mL of ice-cold acetonitrile. An additional 1.0 mL of saturated aqueous sodium chloride solution was added and the samples were homogenized once more. The blood samples (1.0 mL), which had been kept at 0C, were prepared in a similar manner; however, vigorous shaking was used rather than homogenization. Controls were run by injecting DMSO (0.5 mL/kg) as a blank. The animals were sacrificed and the tissues were collected and prepared in the same manner. Standard curves were constructed by adding known amounts of both the quaternary compound 22 and the parent drug, naproxen to blood and tissue homogenate blanks. In this way the experimental results could be quantitated and complete recovery verified.

PAGE 73

63 All samples were placed in a freezer for 2 h in order to completely separate the organic phase. The acetonitrile layer was removed and filtered through a 0.45 m Millipore filter. The samples were then analyzed using high pressure liquid chromatography. The quaternary compound and the free parent drug were monitored using ultraviolet detection at 264 nm. The mobile phase consisted of acetonitrile, water, and aqueous monobasic potassium phosphate (50 mM); (50:25:25). The dihydropyridine compound was detected by ultraviolet absorption at 356 nm, using a mobile phase of acetonitrile and water (75:25). An attempt was also made to detect levels of the free quaternized carrier compound i_ in both the brain and the blood. Extraction was also tested using sample blanks as in the case of the quaternary compound and the free drug. The hydrophilic carrier was analyzed using a C-8 reverse phase column (see Materials and Methods) and a mobile phase consisting of acetonitrile, water, and monobasic potassium phosphate (50 mM); (40:50:10). The compound was detected by UV at 268 nm. Antipyretic Activity of Compound 29 vs an Eguimolar Dose of Naproxen Male Sprague-Dawley rats weighing 235-285 g were used. Rectal temperature was measured twice a day by means of a telethermometer (Yellow Springs Instrument Co., Model 46 TUC) equipped with a thermistor probe (Model 402). Normal temperatures were recorded over a three day period. These temperatures were then averaged to determine the prefever value for each animal.

PAGE 74

64 Pyrexia was induced by a subcutaneous injection of 15 mL/kg of a 15% (w/v) suspension of active dry yeast in sterile saline (0.9%; Abbot Laboratories). Five days later the animals were placed in a retaining cage. The dihydropyridine (0.36 mg/kg, 0.90 mg/mL) or an equimolar dose of naproxen (0.20 mg/kg, 0.50 mg/mL) was given by tail vein injection. In each case DMSO was used as a vehicle. Controls were run by injecting DMSO (0.40 mL/kg) as a blank. Rectal temperatures were measured 1 h before, and at one-hour intervals for 4 h after drug administration using the same thermistor probe.

PAGE 75

Pyridinium Carriers CHAPTER III RESULTS AND DISCUSSION Synthesis The preparation of various pyridinium salt carriers was the first step in the synthetic pathway. Two basic types of carriers were made. One group consisted of compounds that contained a hydroxyalkyl functional group attached to the pyridine ring nitrogen of nicotinamide. The other was comprised of carriers in which the hydroxyalkyl group was attached to the amide nitrogen of nicotinamide. The first group was easily prepared using nicotinamide and a haloalcohol. The quaternization reactions were run in acetone in order to facilitate the precipitation of the product as it was formed. Carriers were made using 2-iodoethanol, 3-iodopropanol and 3-bromopropanol. The 3-iodopropanol was prepared using the Finkelstein reaction. The products were purified by recrystallization when necessary. The second type of carrier required a different synthetic approach. Initially, nicotinoyl chloride hydrochloride was prepared, purified, and dried. It was then stirred with a 50% excess of 2-aminoethanol or 3-aminopropanol. Unfortunately in both cases the hydrochloride salt of the aminoalcohol was formed. 65

PAGE 76

66 A second attempt using the ethyl ester of nicotinic acid with an equimolar amount of 2-aminoethanol was successful. This reaction was run neat, at reflux temperature. The product crystallized upon cooling to room temperature and was easily purified. However, the 3-aminopropanol derivative was isolated as a thick oil, which required vacuum distillation using a Kugelrohr in order to obtain a pure product. These two carriers were then quaternized to form the corresponding pyridinium salts. These reactions were run using methyl iodide as the quaternizing agent and acetone as the solvent. The products were light yellow crystalline solids. These carriers could also be used before quaternization in order to prepare various esters. However, the !-substituted pyridinium carriers were not used in esterification reactions directly, in most cases. This was due to their very low solubility in the majority of organic solvents. Drug Esters The three drugs that were used in this work were valproic acid, indomethacin, and naproxen. These compounds were attached to the previously mentioned carriers via an ester linkage. In some cases the drugs were attached directly to the hydroxyalkyl carrier while in others they were esterified using a haloalcohol. The resulting haloalkyl ester was then used to quaternize the nicotinamide carrier. The N-substituted (amide nitrogen) hydroxyalkyl carriers were reacted with each of the three drugs.

PAGE 77

67 Valproic acid was reacted with the hydroxyethyl carrier (compound j_) using DCC as the coupling agent. However, the desired product was not obtained. This may have been due to the reaction stopping after the initial attack of the acid on DCC. The stability of this adduct as well as the nucleophilicity of the acid and then the alcohol are all determining factors in this reaction going to completion. In the case of indomethacin this coupling was successfully carried out using DCC. The reaction was run in acetonitrile at room temperature. However, in the case of naproxen this reaction was again shown to be less than ideal. In this case the product was not the desired ester. In order for this reaction to be successful a catalytic amount of p-toluenesulfonic acid or DMAP was required. Under these conditions the product could be obtained in a 63% yield. In regard to the hydroxypropyl carrier (compound.]_), the reaction with either indomethacin or naproxen was not of value when DCC alone was used as the coupling agent. This problem also occurred when the acid catalyst zinc chloride or p-toluenesulfonic acid was used in combination with DCC. These esters were successfully synthesized using DCC along with DMAP which acts as a nucleophilic catalyst. The preparation of haloalkyl esters of these three compounds was also attempted using a variety of techniques. The acid chloride of each drug was seen as a possible first step in the synthesis of these esters.

PAGE 78

68 Valproic acid and naproxen were each treated with an excess of thionyl chloride in order to prepare their corresponding acid chlorides. Unfortunately, indomethacin decomposed under these conditions. The acid chlorides of valproic acid and naproxen could be used to achieve our goal. The compounds were stirred with one of several haloalcohols. The valproic acid chloride was reacted with 2-iodoethanol or 3-iodopropanol. In each case the reaction was run neat and the product was purified by removing the excess thionyl chloride. The esterification of naproxen could also be accomplished using a similar method. In this case esters were made using 2-iodoethanol, 2-bromoethanol and 3-bromopropanol. In each case the acid chloride was reacted neat and the reaction mixture was warmed slightly. The products were isolated as crystalline solids which could be recrystallized from 2-propanol. These esters could also be prepared directly from naproxen and a haloalcohol. This improved synthetic method was carried out using a combination of DCC and DMAP. This gave a one-step esterification which resulted in greater or equal yields of the desired product. One additional naproxen ester was prepared by an altogether different method. Naproxen was dissolved in a biphasic mixture of water and methylene chloride, along with sodium bicarbonate and the phase-transfer catalyst tetrabutylammonium hydrogen sulfate. Chloromethylchlorosulfate92 was added dropwise with stirring. The mixture was then kept

PAGE 79

69 at room temperature for one hour after the addition was complete. The phases were separated and the product was isolated from the organic layer. The white solid was dissolved in petroleum ether and filtered to remove any unreacted naproxen. The solvent was removed under reduced pressure and the compound was recrystallized from 2-propanol. The final product was recovered as beautiful white crystalline plates. The final ester of naproxen was prepared using a carrier already in its quaternized form. Therefore, it is discussed in the following section. Drug-Pyridinium Carrier Combinations The drug-pyridinium carrier combinations were each prepared by way of a multistep synthesis. In most cases, the final reaction involved the quaternization of the pyridine ring nitrogen. This was done in order to avoid the problems associated with the pyridinium salts' low solubility in most organic solvents. The sole exception to this procedure was the reaction of naproxen with the 1-(3-hydroxypropyl)pyridinium bromide carrier (compound_!) to form the corresponding ester (compound...!.). Initially, attempts were made to synthesize this product by reacting nicotinamide with a haloalkyl ester of naproxen. This method gave an elimination, resulting in the isolation of the corresponding hydrogen halide salt of nicotinamide. This type of reaction was attempted with the 2-iodoethyl, 2-bromoethyl, and 3-bromopropyl esters. In each

PAGE 80

70 case the reaction with nicotinamide led to elimination rather than substitution. The reaction was originally run in acetone at reflux temperature. However, replacing acetone with acetonitrile, nitromethane, or dimethylformamide did not seem to influence the course of the reaction. The temperature was reduced as low as 40C, but this still did not give the desired quaternary compound. Therefore, esterification of the hydroxyalkyl quaternary carrier was pursued. The successful procedure involved dissolving the quaternary carrier (compound .. !.> in a minimum amount of dimethylformamide. Naproxen was then coupled to the pyridinium salt in the presence of DCC and DMAP. The majority of the DCU that was formed during the reaction precipitated and was filtered away. The solvent was removed under reduced pressure, and the resulting solid was twice recrystallized from ethanol. This gave the final product as a fluffy, tan crystalline solid. The syntheses of the remaining drug-pyridinium carrier combinations required quaternization of the pyridine nitrogen as the final step. In the two additional cases in which naproxen was linked via the !-position of the nicotinamide carrier, the appropriate haloalkyl ester of naproxen was directly substituted onto nicotinamide. The 2-iodoethyl ester of valproic acid was used in the synthesis of compound 19. The reaction was run in dimethylformamide at reflux temperature. The product was isolated and recrystallized from 2-propanol/ether using the mixed solvents technique.

PAGE 81

71 The other synthesis of this general type involved the chloromethyl ester of naproxen. This reaction was run using acetone at reflux temperature. This allowed for the majority of the product to precipitate as it was formed during the reaction. The N-substituted (amide nitrogen) esters of indomethacin and naproxen were each quaternized using an excess of methyl iodide. This was a straightforward reaction using acetone as the solvent in each case. The product either partially or completely precipitated from the reaction solution. It was then further purified by recrystallization if necessary. Typical proton NMR spectra are seen in Figures 3-1 and 3-2. Dihydropyridine-Chemical Delivery Systems The reduction of the quaternary drug-carrier combinations was effected in order to form their corresponding 1,4-dihydropyridines. In the majority of cases the dihydropyridines (compounds 26, 28-32) were synthesized using sodium ---dithionite as the reducing agent (see Figures 3-3 and 3-4). The reactions were run in degassed, deionized ice-cold water. The water was degassed to remove any dissolved oxygen in order to prevent the reoxydation of the dihydro compound once formed in solution. The reactions were bubbled with nitrogen gas to again displace any oxygen in the reaction atmosphere. Sodium dithionite was used in excess as a mild reducing agent. Sodium bicarbonate was used to raise the pH of the reaction mixture in order to

PAGE 82

Figure 3-1. 1 H NMR spectrum of compound 22. ....J N

PAGE 83

Figure 3-2. 1 H NMR spectrum of compound 24. .....J w

PAGE 84

Figure 3-3. Synthetic reaction sequence for compound 29, where R represents the naproxen ester.

PAGE 85

Figure 3-4. ACETONE > NH{~}:OR 2 H-E%t0R 2 Synthetic reaction sequence for compound 31, where R represents the indomethacin ester.

PAGE 86

76 stabilize the product as it was formed. Once the reaction was complete the product was isolated by extracting the aqueous reaction mixture with either methylene chloride or diethyl ether. The organic extracts were combined and dried with magnesium sulfate. The solvent was removed and the dihydro compound was dried using a vacuum pump. The product was obtained as a yellow solid foam. These hygroscopic compounds were somewhat difficult to handle and elemental analysis always showed the presence of some water. When these analyses were run in duplicate the second measurement always showed additional water present. This indicated that these compounds drew water quite readily from the humid atmosphere. A typical proton NMR spectrum is seen in Figure 3-5. The one exception to this procedure was made in the preparation of the dihydro form of the 1-hydroxymethylnicotinamide ester of naproxen {compound 27). The compound's quaternary form {compound 20) is quite unstable in basic medium. Therefore, it was impossible to reduce this compound in the presence of aqueous sodium bicarbonate. An alternate method of reduction was found that could successfully reduce the quaternary salt to the desired product. A more active {less stable) l,2-dihydronicotinamide94 was used as the reducing agent. This redox reaction was run in a mixture of acetonitrile and 2-propanol at room temperature. The solvent was removed and the residue was taken up in methylene chloride in order to remove the quaternary side

PAGE 87

Figure 3-5. 1 H NMR spectrum of compound 29.

PAGE 88

78 product by filtration. The solvent was removed under reduced pressure and the resulting oil was dried on a vacuum pump. An attempt was made to remove the final traces of quaternary salt contamination by dissolving the product in chloroform and quickly passing it down a short column of neutral alumina. This, however, led to decomposition of the product. Chemical Stability The testing of the chemical stability of the indomethacin and naproxen quaternary drug-carrier combinations and dihydropyridine-Coss was the first step in evaluating the likelihood of their success in the delivery of the parent drug to the brain. Each of these compounds' rate of disappearance was measured at 37C in phosphate buffer at pH 7.4. This work was carried out to determine the relative stability of each compound, and to determine if possible the various reaction products. The pyridinium salts were quite stable with the expected exception of the 1-hydroxymethylnicotinamide ester of naproxen. This soft quaternary salt was by far the least stable of the group. This compound had a half-life of 8.6 minutes as compared to one or more days for the other quaternary compounds. The general stability trends followed the anticipated patterns for both the dihydropyridines and the pyridinium salts. As the ester linkage was moved further away from the electron-withdrawing pyridinium ring it became more stable

PAGE 89

79 to chemical hydrolysis. This tendency resulted in a corresponding decrease in the rate of disappearance of these compounds. In the case of the dihydropyridine compounds however, the order of stability was reversed. The electronwithdrawing effect of the ester moiety has a stabilizing influence on the 1,4-dihydropyridine ring. Therefore, as the ester linkage was moved further away from the ring the rate of disappearance of these compounds increased. These effects were also reflected in the half-lives of all of the compounds seen in Table 3-1. The products of decomposition were also studied when possible. In the case of the quaternary compounds, hydrolysis of the ester bond was the principal instability. The free drug was detected in increasing amounts as the pyridinium salt concentration decreased following a pseudo-first order rate of disappearance. The dihydropyridine compounds produced a more complicated variety of reaction products. The corresponding quaternary compound could be detected in most cases even when the buffer was degassed before use. This could have been due to a small amount of atmospheric oxygen interacting at the buffer surface or redissolving into the buffer as each experiment proceeded. The exception again involved the 1-hydroxymethylnicotinamide ester of naproxen (compound 20). This compound was not found in detectable amounts due to it being much less stable than its corresponding dihydropyridine form. Therefore, if the oxidation product was formed

PAGE 90

80 Table 3-1. Half-life of disappearance and correlation coefficient for each dihydropyridine and quaternary pyridinium compound, in pH 7.4 phosphate buffer. Half-life (min); or h = hour. Results are from one run per compound, and six to fifteen sample measurements per run. Comeounds Dihydro Quat 31 and 24 32 26ha (0.97) (0.92) 32 and 25 24 35ha (0.99) (0.998) 29 and 22 16 58ha (0.97) (0.75) 30 and 23 9.0 550ha (0.99) (0.88) 28 and 21 52 750ha (0.998) (0.98) 27 and 20 5.lh 8.6 (0.98) (0.998) aParent drug was detectable, but not within first hour of each experiment.

PAGE 91

81 it would decompose at a much greater rate thus preventing its detection. The parent drug was also detectable in small amounts. This phenomenon could be due to ester hydrolysis either before or after oxidation of each dihydro compound. The experimental results indicated that the quaternary compound was not the primary source of this freed drug. This hypothesis is based on the relatively low levels of the quaternary compounds which were present, as well as their great stability in most cases. The major decomposition product in pH 7.4 buffer is an addition product. In an aqueous medium this reaction normally results in the addition of water across the 5,6-double bond of the dihydropyridine ring. The mechanism involves protonation at the 5-position, which is usually rate determining. This step is then followed by nucleophilic attack at the 6-position by hydroxide ion or water. However, the addition product could not be detected using either type of high pressure liquid chromatography system as attempted. This result was most likely due to the compound's increased polarity and/or its transient nature. In Vitro Studies Stability in Human and Rat Blood The stability of a drug in the blood of the patient is always an important factor. In order to deliver drugs to the brain using the chemical delivery system described, this factor becomes one of the criteria for success. A given

PAGE 92

82 dihydropyridine compound must penetrate into the brain before oxidation takes place. Then, once oxidation has occurred, the resulting quaternary ester must also be reasonably stable in the bloodstream. This stability would allow for its excretion before significant amounts of the parent drug are released in the periphery. In order to measure the stability of these compounds, each of the naproxen and indomethacin pyridinium esters and its corresponding dihydro-CDS was tested. These experiments followed the disappearance of each compound in both human and rat blood under in vitro conditions (Table 3-2). The most surprising finding of this particular work was the quaternary compounds' relative stability in rat blood when compared to human blood. The only exception was the hydroxypropyl carrier ester of indomethacin (compound 25), for which the rate of disappearance in rat blood was twice that found for human blood. Normally, one would assume this type of greater enzymatic activity in rat blood to be the rule rather than the exception. This phenomenon was apparently due to some type of specificity involving one or more of the enzymes contained in human blood. The dihydropyridine compounds, however, behaved in a more expected manner. These compounds for the most part were more stable in human blood than in rat blood. So much so, that each dihydro-CDS in which the drug ester was attached through the amide nitrogen of nicotinamide was even more stable in human blood than in pH 7.4 phosphate

PAGE 93

83 Table 3-2. Half-life of disappearance and correlation coefficient for each dihydropyridine and quaternary pyridinium compound, in 100% whole rat and 100% whole human blood. Half-life (min); or h = hour. Results are from one to three runs per compound and six to fifteen sample measurements per run. Compound 31 and 24 32 and 25 29 and 22 30 and 23 28 and 21 27 and 20 Rat Dihydro 5.2 (0.98) 24 (0.999) 5.5 (0.98) 5.1 (0.99) 10 (0.97) 3. 7h (0.93) Human Dihydro 60 (0.96) 27 (0.98) 61 (0.98) 52 (0.99) 30 (0.995) 2.7h (0.96) Rat Quat 84 (0.97) 18 (0.999) 15ha (0.91) 6.2ha (0.95) stableb 0.4 (0.98) Human Quat 15 (0.995) 37 (0.993) 5.4 (0.995) 1.8 (0.97) 3.2 (0.998) 0.2 (0.999) aExperiment was followed for< one half-life. bcompound very stable; parent drug was not detectable over course of experiment.

PAGE 94

84 buffer. The only dihydropyridine compound that was not more stable in human blood compared to rat blood was the dihydro form of the 1-hydroxymethylnicotinamide ester of naproxen (compound 27). This dihydro compound was also unusual in that it was more stable than its corresponding pyridinium salt in both human and rat blood. Normally, the dihydropyridines were less stable than their quaternary forms in rat blood, but more stable than the corresponding quaternary forms in human blood. The only exception to both of these patterns involved the hydroxypropyl carrier ester of indomethacin (compounds 25 and 32). The major products of decomposition for each of the compounds tested were the same in both human and rat blood. In each case, hydrolysis of the naproxen or indomethacin ester was the major enzymatic process involved. This was true for the pyridinium salts and at least in the case of rat blood the dihydropyridines. In human blood it was not always possible to determine if the dihydropyridines were oxidized before hydrolysis, due to the relative instability of most of the quaternary compounds in this medium. However, in each case except for the 1-hydroxymethylnicotinamide ester of naproxen, the quaternary ester compound (oxidation product) was detected along with the parent drug. This result occurred after the addition of each of the dihydropyridines to either human or rat blood, except as previously mentioned.

PAGE 95

85 Stability in 20% Brain and Kidney Homogenate In order to perform the desired testing in brain or kidney homogenate fresh rat tissue was used. The fresh organ was obtained and homogenized in pH 7.4 phosphate buffer (4 mL/g). The homogenate was centrifuged at 3,000 rpm, and the supernatant was removed and stored briefly on ice. When the experiment was begun the homogenate was incubated in a water bath at 37C. The dihydropyridine compounds were generally much less stable in brain homogenate than were their corresponding pyridinium salts (Figure 3-6 and Table 3-3). The exception again involved the 1-hydroxymethylnicotinamide ester of naproxen (compounds 20 and 27). In this case the quaternary compound was rather unstable in brain or kidney homogenate due to the ester linkage being quite close to the electronwithdrawing positively charged ring. However, the five additional pyridinium compounds were rather stable in both brain or kidney homogenate. The stability in brain homogenate was normally somewhat lower than that measured in kidney homogenate. This finding was obtained even in the case of compound 20, but not in the case of the Nsubstituted (amide nitrogen) hydroxyethylnicotinamide carrier ester of naproxen (compound 22). The dihydropyridine compounds were readily oxidized in rat brain homogenate. The half-life of disappearance for these delivery systems was generally one to five minutes. The only exception again came in the form of compound 27.

PAGE 96

10.0 5.0 1.0 ,....., .., .c 0, ,-C1I IO 0.5 C1I 0.. '--' O> 0 0.1 Figure 3-6. 5 86 10 15 20 25 30 35 40 45 Time(mfo) In vitro results of the dihydropyridine (compound 31) A in 20% rat brain homogenate. Increases in the quaternary compound (24) 0 and freed indomethacin are also seen.

PAGE 97

87 Table 3-3. Half-life of disappearance and correlation coefficient for each dihydropyridine and quaternary pyridinium compound, in 20% rat brain homogenate, and for each quaternary pyridinium compound in 20% rat kidney homogenate. Halflife (min); or h = hour. Results are from one to three runs per compound and six to fifteen sample measurements per run. Com2ound Dihydro Quat Quat 31 and 24 3.5 llha 35ha (0.999) (0.97) (0.97) 32 and 25 1.9 stablec 16hb (0.99) (0.89) 29 and 22 1.2 5.8hb stablec (0.99) (0. 77) 30 and 23 4.5 6.8ha 9.7ha (0.998) (0.94) ( 0. 79) 28 and 21 1.6 stablec stablec (0.98) 27 and 20 37 1.9 2.3 (0.97) (0.998) (0.997) < one half-life. but not within first hour of aExperiment was followed for bParent drug was detectable, each experiment. ccompound very stable; parent drug was not detectable over course of experiment.

PAGE 98

88 This compound's half-life was thirty seven minutes in brain homogenate. The major product of decomposition that was detected for compound 27 was another unique exception. The parent drug was detected after adding the dihydropyridine (compound 27) to the brain homogenate, but none of the respective quaternary compound could be measured as in the five other cases. This result was due to the quaternary compound being much less stable than its corresponding dihydropyridine. Therefore, if oxidation was taking place its product would decompose at a faster rate than the rate at which it was formed. The five remaining dihydropyridines were oxidized to give the pyridinium salt as the major decomposition product in brain homogenate. Lower levels of the parent drug were also detected in each case. The proposed system of delivering drugs to the brain, has many requirements to obtain ideal results. The dihydroCDS should oxidize rather rapidly once it enters the brain. This process allows the drug-carrier complex to acquire its "locked-in" effect. Then, slow hydrolysis of the ester bond would yield the parent drug in a sustained release manner. In the kidney the pyridinium ester should be as stable as possible in order to facilitate the excretion of the drug-carrier combination. This would help prevent the release of the parent drug in the periphery.

PAGE 99

89 The results of the in vitro testing were used to compare and evaluate each chemical delivery system. The results were then contrasted with the ideal system. In this way one has a steppingstone to access the likelihood of success in an in vivo experiment. In Vivo Studies Preliminary Distribution of Compound 31 The N-substituted (amide nitrogen) hydroxyethylnicotinamide dihydro carrier ester of indomethacin (compound 31) was the first chemical delivery system in the series to be synthesized. In order to assess the new carriers' ability to deliver a drug to the brain this first compound (31) was used in a preliminary in vivo distribution experiment. Qualitative measurements of both the quaternary drug-carrier combination and the freed parent drug were made. The experiment was conducted in male rats and the drug was administered by iv injection. The results of this experiment showed good penetration of the chemical delivery system into the brain. Low levels of the parent drug were also detected in the brain over the four hour experiment (Table 3-4). In the blood the quaternary drug-carrier combination was found in about one-half of the concentration in the brain five minutes after injection. This compound appeared to be excreted reasonably well from the periphery and after four hours its concentration was just above the detectable limit.

PAGE 100

90 Table 3-4. Results from in vivo distribution of compound 31, 20 mg/kg (values in peak heightSEM). Two animals per time point were injected with the dihydropyridine compound. QUAT INDOMETHACIN BRAIN 5 min 9.3.8 1.8.6 1 hr.* 5.8 2.5 4 hr. 1. 3. 4 1. 2. 3 BLOOD 5 min 4.4.0 8.6.5 4 hr. 0.3.3 2.8.8 *one animal

PAGE 101

91 Indomethacin, however, was detected in the blood at a level about five times that in the brain after five minutes. This ratio represented a dramatic improvement over the twenty to one ratio found in the literature after indomethacin alone is given. The hydrolysis of the ester in the blood which led to significant amounts of indomethacin release appeared to be this delivery system's only major deviation from an ideal design. A problem that we hoped could be overcome by choosing a delivery system that had been shown to possess a greater stability in rat blood. In Vivo Distribution of Compound 28 In order to determine which chemical delivery system or systems were the most promising, a comparison of in vitro data was made. In this way one can start to assess a particular compound's chances of meeting the required criteria for success. The 1-hydroxyalkylnicotinamide esters of naproxen were quite distinct in their in vitro characteristics. The 1-hydroxymethylnicotinamide ester of naproxen was unusual in almost all respects. The pyridiniurn salt was exceptionally unstable under each in vitro condition tested, while the dihydro-CDS was the most stable of any tested. Unfortunately, this combination did not afford the type of delivery system required. The 1-(3-hydroxypropyl)nicotinamide ester of naproxen (compounds 21 and 28) appeared to be a much better candidate for the proposed drug delivery system. The dihydro form of

PAGE 102

92 this compound was administered by iv injection (tail vein) in rats. Six tissues were collected along with whole blood and quantitatively analyzed for drug concentrations. The analysis was carried out using high pressure liquid chromatography with UV detection. The results were quantified by constructing an individual standard curve for blood and for each tissue type. Each curve consisted of three or four relevant concentrations of the quaternary compound and naproxen. Extraction and recovery were verified and shown to be greater than 90% in each case by comparing the peak height 0 the extracteQ samples with those obtained from standard dilutions. The error in accuracy associated with making replicate injections of one sample varied from 2-10%. The limit of detection was 0.4 g/g for both the quaternary compound (21) and for naproxen. The dihydro compound was not detectable at the first time point (30 min). However, the quaternary drugcarrier combination was easily detected in all tissues samples except liver (Table 3-5). Naproxen was detectable in each sample except the four hour liver measurement (Table 3-6). These findings gave evidence of a rapid metabolism and/or excretion by the liver of each of the three compounds involved. The quaternary ester and the parent drug were both found in significant amounts in the brain. The pyridinium salt showed a relatively slow rate of hydrolysis in the brain, which in turn gave rise to a stable, sustained

PAGE 103

93 Table 3-5. Concentrations of the pyridinium drug-carrier combination, compound 21 (g/g or g/mLSEM), after administration o~a 25 mg/kg dose of the dihydropyridine compound 28. Three animals were used for each time point.~ Tissue 0.5 h 1 h 4 h Brain 28.0.8 26.1.2 16.8.7 Blood 7.6.0 9.4.7 6.0.8 Lung 52. 7.0 75.0.5 17.2.1 Kidney 16.2.8 12.7.1 4.5.8 Liver Testes 0.9.6 2.3.9 1.6. 9 Fat 2.0.7 3.2.9 *below level of detection

PAGE 104

94 Table 3-6. Concentrations of naproxen {g/g or g/mLSEM), after administration of a 25 mg/kg dose of the dihydropyridine compound 28. Three animals were used for each time point.~ Tissue 0.5 h 1 h 4 h Brain 1. 7. 4 1.6.8 1.6.6 Blood 6.3.6 7.4.5 6.0.5 Lung 10.3.3 13.1.8 9.8.0 Kidney 4.5.6 3.4.8 3.2.6 Liver 3.4.0 2. 7 1. 0 Testes 2.1.6 3.4.5 1.6.6 Fat 1. 9. 7 2.6.0 1.1. 5 *below level of detection

PAGE 105

95 concentration of the parent drug. Unfortunately, the highest concentrations of both quaternary compound and naproxen were found in lung tissue. This distribution seemed to cause respiratory depression which was the only acute toxicity discernible during the animal experiments. The lungs of the animals that were sacrificed up to one hour after drug administration were curiously red due to what may have been micro-vascular bleeding. However, this phenomenon reversed itself over the course of the experiment. The four hour animals had recovered from any noticeable sedative effect and the lungs again appeared normal. The quaternary compound seemed to be excreted via the kidney before extremely large amounts of the free drug were released in this tissue. This finding was another which demonstrated the relative success of this delivery system. The low levels of blood profussion to the testes and fat tissues resulted in their having the lowest concentrations of both pyridinium salt and parent drug, compared with the other tissues in which these compounds were present. These results were also quite promising. However, the extremely large concentrations of both compounds found in the lungs were unsettling to the point that the search for the "ideal" delivery system was continued. In Vivo Distribution of Compound 29 vs an Equimolar Dose of Naproxen The N-substituted (amide nitrogen) naproxen carriers were the last type of CDS to be experimented with under in vivo conditions. The N-hydroxyethylnicotinamide ester was

PAGE 106

96 choosen in preference to the N-hydroxypropylnicotinamide ester based on the previously obtained in vitro data. The quaternary compound (compound 22) was somewhat more stable than the hydroxyproyl derivative in human blood as well as rat kidney and rat blood (Tables 3-2 and 3-3). In the brain the pyridinium salt exhibited a half-life of about six hours, which should allow for a sustained release delivery. The DH-CDS (compound 29) was also more stable than the corresponding hydroxypropyl ester in both human and rat blood. However, compound 29 showed a faster rate of oxidation in brain homogenate when compared to the hydroxypropyl compound. This should allow the "locked-in" pyridinium compound to concentrate in the brain and better serve as a prodrug of naproxen. The results of the in vivo comparison of compound 29 and an equimolar dose of naproxen were quite exciting. The analysis was carried out using high pressure liquid chromatography with UV detection (see Figures 3-7 and 3-8). The results were quantified by constructing an individual standard curve for blood and for each tissue type. Each curve consisted of three or four relevant concentrations of the dihydropyridine (compound 29), the quaternary compound(22), the quaternary hydroxyalkyl carrier (compound 2._), and naproxen. Each of the compounds was tested separately. Extraction and recovery were shown to be greater than 90% in the case of the quaternary compound (22) and in the case of naproxen, by comparing the

PAGE 107

r--t-~--------------------------------~z_o~, Figure 3-7. 02 A -u75,z 3, 1)3 The HPLC where A is the quaternary compound (22) at a concentration of 6lg/g and Bis naproxen (3-:-S-g/g) extracted from rat brain. 2.42 ~ -. "'

PAGE 108

-====~;;;;;;;;~::::=B= s~ );;:J = A e.1e Figure 3-8. The HPLC where A is the quaternary compound (22) at a concentration of 38 g/mL and Bis naproxen 1"4'o g/mL) extracted from rat blood. 2. 7.:1

PAGE 109

99 peak height of the extracted samples with those obtained from standard dilutions. The error in accuracy associated with making replicate injections of one sample varied from 2-10%. The limit of detection was 0.4 g/g for the quaternary ester (compound 22), the quaternary carrier (compound 2._), and naproxen. The limit of detection for the dihydropyridine (compound 29) was 0.5 g/g. The concentration of the pyridinium ester (compound 22) was highest ln brain tissue, up to 24 hours after administration (Figure 3-9 and Table 3-7). Initially lung, kidney and blood concentrations were also rather high, but these levels decreased much more rapidly, especially in the lung and kidney. The liver again showed a rapid metabolism and/or excretion of this compound. However, the levels detected in the testes and fat, although extremely low, were more sustained than that found in the liver. The pyridinium ester did successfully serve as a prodrug of naproxen. This drug-carrier combination released naproxen in the brain to give a much higher brain/blood ratio of the parent drug, when compared to administering an equimolar dose of naproxen. When naproxen itself was given, an initial spike concentration was measured in the brain which surpassed the level attained after treatment with the DH-CDS (compound 29). However, the concentration achieved from naproxen fell within 30 minutes to values that were significantly lower than those recorded after giving the dihydropyridine compound (Figure 3-10 and Tables 3-8 and 3-9)

PAGE 110

160 140 \20 Brain 0 Blood \00 Lung I Kidney .. .... 80 0 liver ::ll. "' D Testes 0 0 .... 60 0 Fat ::ll. ,o Time (h) Figure 3-9. Concentrations of the pyridinium drug-carrier combination, compound 22 (g/g or g/mLSEM), after administration of a 25 mg/kg dose of the dihydropyridine compound~-I-' 0 0

PAGE 111

Table 3-7. Concentrations of the pyridinium drug-carrier combination, compound 22 (g/g or g/mLSEM), after administration of a 25 mg/kg dose of the dihydropyridine compound 29. Three to six animals were used for each time point. Tissue 5 min 0.5 h 1 h 4 h 8 h 24 h Brain 169.0.9 134. 7.8 131. 4. 0 56.2.2 2 7 .1. 5 4.5.3 Blood 142.5.3 133.4.9 104.7.2 34.8.2 7.3.2 5.4.3 Lung 157.1.5 11. 9. 6 12.3.2 7. 8 1. 0 10.6.8 Kidney 144.2.8 4 7. 3 .1 41.8.4 6. 7. 3 Liver 36.0.1 10. 6. 2 * Testes 9. 7.1 9.6.5 9.6.9 3.4.3 4.3.3 Fat 8. 2 1. 4 4. 6. 0 5. 3. 0 2.4.2 *below level of detection

PAGE 112

300 260 220 180 140 10 -' E ....... L 80 0 ....... 60 40 20 0 S111tn O.Sh 1 h 4h Sh 24h Smin O.Sh 1h 4h BRAIN D DH-CDS CS:) NAPROXEN 102 Sh 24h Figure 3-10. Concentrations of naproxen (g/g or g/mLSEM), after administration of a 25 mg/kg dose of the dihydropyridine compound 29, or an equimolar amount of naproxen (15 mg]kg) itself. Three to six animals were used for each time point.

PAGE 113

103 Table 3-8. Concentration of naproxen (g/g or g/mLSEM), after: administration of a 25 mg/kg dose of the dihydropyridine compound 29, or an equimolar amount of naproxen (15 mg/kg) itself. Three to six animals were used for each time point. Tissue Brain Blood Lung Kidney Liver Testes Fat Tissue Brain Blood DH-CDS+NAPROXEN 5 min 0.5h 1 h 4 h 8 h 24 h 10.6.1 7.6.1 6.7.4 3.9.6 1. 7.1 50.4.3 41.9.2 41.4.8 26.5.1 28.4.5 8.3.1 41.5.0 29.1.8 30.7.3 31.5.8 27.1.1 39.6.3 19.5.5 20.4.7 14.5.3 9.0.1 31.4.4 9.3.8 9.1.5 6.5.3 11.8.0 3.5.4 5.9.7 6.4.3 4.7.3 3.5.4 4.0.0 3.3.3 3.4.2 3.2.2 6.1.1 5 min 18. 2. 2 0.5 h 4.4.4 NAPROXEN 1 h 4.6.4 4 h 2.8.8 8 h 2.5.1 * 24 h 291.4.9 88.6.9 74.0.4 56.4.7 44.0.7 5.0.0 Lung 259.1.3 69.3.7 59.4.9 39.1.8 43.0.2 4.5.4 Kidney Liver Testes Fat 87.0.7 41.7.5 30.7.6 18.5.0 12.5.5 0.2.2 79.3.9 22.8.2 15.9.2 7.9.2 16.0.3 2.6.3 8.5.9 12.1.0 11.1.4 7.2.2 5.0.4 0.6.6 7. 9. 7 9.6.7 5.1.6 5.9.3 6.9.0 *below level of detection

PAGE 114

104 Table 3-9. Comparison of brain/blood and brain/tissue ratios of naproxen concentration after: administration of a 25 mg/kg dose of the dihydropyridine compound 29, or an equimolar amount of naproxen (15 mg/kg) itself. Three to six animals were used for each time point. Blood Lung Kidney Liver Testes Fat Blood Lung Kidney Liver Testes Fat 5 min 0.21 0.26 0.27 0.34 3.03 2.65 5 min 0.06 0. 07 0.21 0.23 2.14 2.30 0.5 h 0.18 0.26 0.39 0.82 1.29 2.30 0.5 h 0.05 0.06 0.11 0.19 0.36 0.46 DH-CDS+NAPROXEN 1 h 0.16 0.22 0.33 0.74 1.05 1.87 NAPROXEN 1 h 0.06 0.08 0.15 0.29 0.41 0.90 4 h 0.15 0.12 0.27 0.60 0.83 1. 22 4 h 0.05 0. 0 7 0.15 0.35 0.39 0.47 8 h 0.06 0.06 0.19 0.18 0.49 0.28 8 h 0.06 0.06 0.20 0.16 0.50 0.36 *numerator or denominator or both below level of detection 24 h * 24 h *

PAGE 115

105 The brain/tissue ratios of naproxen were also higher, up to eight hours after administration of the DH-CDS compared with those obtained after injecting naproxen. This finding gave good evidence for the contention that the CDS would spare the body from peripheral toxicity while still accomplishing the desired central effect. The elimination of the initial spike concentration of drug in the brain should also reduce any central toxicity associated with a given drug. The problem relating to the accumulation in the lungs of the quaternary compound and the parent drug was also greatly reduced. In this way the goals of the proposed drug delivery system were attained in the case of this chemical delivery system. The concentration of the dihydro-CDS was also measured by HPLC (see Figure 3-11). However, measurable quantities were detected only in the early lung and fat samples. This finding was due to higher enzymatic activity in the other tissues analyzed. The total amount of dihydropyridine delivered to each tissue was only of secondary importance. This unfortunate result is due to enzymatic activity continuing before freezing and during the homogenization of each tissue. The guillotine regrettably does not vanquish enzymatic activity. The detection of the quaternary carrier after cleavage of the parent drug posed another problem. Because of the compound's increased hydrophilicity its retention on a c18, reversed phase column was not accomplished. The compound

PAGE 116

Figure 3-11. The HPLC of the dihydropyridine (compound 29) standard (7.9 g/g) extracted from rat brain blank tissue.

PAGE 117

107 was detected using a c8 column, but due to the extraction technique utilized for this study, only two to three percent of the compound in each sample was recovered. Brain and blood levels were examined in order to determine if the compound was accumulating. The results showed a decrease in concentration in both brain and blood levels over the course of the experiment. This finding indicated that the carrier did not accumulate, but instead was excreted from both the brain and the blood. Antipyretic Activity of Compound 29 vs an Eguimolar Dose of Naproxen The final and most important factor in determining the success or failure of the proposed chemical delivery system was to verify the activity of the parent drug once it was released in the brain. Compound 29 was choosen for study after its superiority was recognized from the previously mentioned distribution studies. Naproxen and indomethacin both belong to the group of drugs known as nonsteroidal anti-inflammatory agents. In addition to their anti-inflammatory activity, they also exhibit analgesic and antipyretic properties. The antipyretic activity was investigated due to its control being centrally mediated in the hypothalamus.86 This experiment was carried out in order to support the hypothesis that once the parent drug was delivered to the brain, it would be free to effect its characteristic activity .The dihydro-CDS was compared to an equimolar dose of naproxen in order to access their relative abilities to

PAGE 118

108 lower fever in the rat. The dosage amount was based on the EDso for naproxen.95 Pyrexia was induced by giving a subcutaneous injection of a suspension of active dry yeast. The dosage given resulted in a l-2C increase in core temperature. The drugs were given by iv injection using DMSO as a vehicle. A control group was also used in which DMSO alone was given. This control was used in order to verify that DMSO did not elicit an antipyretic effect. The temperature of each animal was recorded once per hour for four hours after the antipyretic or control injection. One hour after administration both drugs elicited a similar response. Subsequently the naproxen group showed a leveling off or slight increase in body temperature. However, the group treated with the dihydropyridine delivery system exhibited a continuing decrease in body temperature for two additional hours. This result brought about a statistically significant difference in the effectiveness of the two treatments (Table 3-10}. Three hours after administration of the drugs the CDS showed improved antipyretic activity with a confidence level greater than 95% using student's t-test to compare two treatments.96 Overall, the two treatments gave statistically similar maximal effects. Although the results did suggest that larger test groups may show an improvement in maximal response using the CDS. The present study also showed both treatments were effective in reducing fever, but a sustained activity was brought about using the chemical delivery

PAGE 119

Table 3-10. Results (CSEM) of the antipyretic activity effected after administration of a 0.36 mg/kg dose of the dihydropyridine compound 29, an equimolar dose of naproxen (0.20 mg/kg) itself, or the DMSO vehicle To.40 mL/kg) as a control. Six animals were used for each treatment group. Treatment a Pre-fever Fever AFever Temperature Amaxb -1 h + 1 h + 2 h + 3 h + 4 h DH-CDS 37.2.05 38. 7.26 -0.53.19 -0.67.21 -0.88.18c -0. 8 7. 23 -0.97.20 Naproxen 3 7. 2 .10 38.4.18 -0.60.25 -0.42.15 -0.38.16d -0.50.30 -0.70.24 Control 37.2.10 38.5.14 -0.13.07 -0.15.11 -0.08.10 -0.03.09 -0.28.08 adrugs were administered at time= 0 bmean of the maximum change in fever temperature for the animals in each treatment group at any time over the four hour experiment. cstatistically significant increase in antipyretic effect over naproxen itself or DMSO alone; passes student's t-test with a confidence level> 95% compared to the naproxen group and a confidence level >99.5% compared to the DMSO control group. dstatistically significant increase in antipyretic effect over DMSO control group; passes student's t-test with a confidence level >90%.

PAGE 120

110 system. These findings were in good agreement with those expected from the distribution studies of these two compounds.

PAGE 121

CHAPTER IV CONCLUSION In order to produce an ideal chemical delivery system a number of requirements must be met. The dihydropyridine delivery system must first be synthesized. This was accomplished by a variety of synthetic procedures. Once synthesis was completed, methods of analysis were developed. High pressure liquid chromatography equipped with UV detection was used to separate, detect, and often quantitate the various compounds involved in this study. The stability of each chemical delivery system along with its corresponding pyridinium salt was tested. This chemical and in vitro stability was utilized as an initial indication of the fitness of each CDS for use in the proposed system of delivering drugs to the brain. The comparisons were made with each of the proposed system's objectives in mind. In this way one could attempt to move a step closer to the "ideal" system. Three of the chemical delivery systems that were synthesized and tested were chosen for in vivo rat distribution studies. Naproxen was also administered using the same technique in order to compare the in vivo results. This continuing process of elimination was employed in order to identify the most valuable CDS of the group. The compound that most closely adhered to the "ideal" system. 111

PAGE 122

112 The three carrier systems all improved the delivery of the parent drug to the brain, but two of the compounds produced results which deviated from the "ideal" system. However, the third CDS that was examined fulfilled each of the proposed goals or advantages of the brain delivery system. 1.) A sustained release of the parent drug was achieved in the brain. 2.) The amount of the free drug in the blood and peripheral tissues was greatly reduced, which should decrease any peripheral toxicity associated with the drug. 3.) The peak or spike concentration seen shortly after administration of the parent drug was eliminated using the chemical delivery system. This should reduce any central toxicity associated with the parent drug. 4.) A more consistent level of parent drug was found in the brain after administration of the dihydropyridine compound when compared to the amounts detected after giving an equimolar dose of naproxen alone. This should increase the efficacy of a given dose of the drug. The final experiment used to prove the effective and useful nature of the proposed drug delivery system was designed to test the actual antipyretic activity elicited after administration of the chemical delivery system (compound 29). This was compared to the effect of an equimolar dose of naproxen. The two treatments were both effective in reducing the fever induced by yeast injection into rats. The group receiving the dihydropyridine compound exhibited a slightly greater reduction in fever than did the group

PAGE 123

113 receiving naproxen. The antipyretic effect was also more sustained in the animals treated with the CDS. This phenomenon resulted in a statistically significant improvement in antipyretic activity three hours after administration of the dihydropyridine versus naproxen itself. The one and two hour temperature measurements did not show significant differences in the two treatments. However both treatments were effective when compared to a control group receiving only the DMSO vehicle. This project was designed to develop carriers which would deliver drugs to the brain. The results illustrate the effective and superior manner in which at least one of the developed chemical delivery systems accomplished this goal. In the future, the hydroxyalkyl carriers that were developed for this work could be used to improve the delivery of a variety of drugs to the brain. These drugs could include the three employed in this work or any of the number of other compounds which contain a carboxylic acid functional group.

PAGE 124

REFERENCES 1. Stern, L.S., Gautier, R., Arch. Int. Physiol., 17, 138 (1921). 2. Ehrlich, P., Ther. Monatsh., ..!., 88 (1887). 3. Reese, T.S., Karnovsky, M.J., J. Cell. Biol., 34, 207 (1967). 4. Karnovsky, M.J., J. Cell. Biol., 35, 212 (1967). 5. Patlak, c.s., Paulson, O.B., Microvas. Res., 21, 117 (1981). 6. Fenstermacher, J.D., Blasberg, R.G., Patlak, c.s., Pharmac. Ther., 14, 217 (1981). 7. Levin, E., Exp. Eye Res. Suppl., 191 (1977). 8. van Deurs, B., Int. Rev. Cyt., 65, 117 (1980). 9. Rapoport, s., Ohno, K., Pettigrew, K.D., Brain Res., 172, 354 (1979). 10. Oldendorf, W.H., Ann. Rev. Pharmac., 14, 239 (1974). 11. Levin, V.A., J. Med. Chem., 23, 682 (1980). 12. Diamond, J.M., Wright, E.M., Ann. Rev. Physiol., _l!_, 581 (1969). 13. Oldendorf, W.H., Amer. J. Physiol., 221, 1629 (1971). 14. Ross, S.B., Froden, O., Eur. J. Pharmacol., 13, 46 ( 19 70 ) 15. Bodor, N., Farag, H.H., Brewster, M.E., Science, 214, 13 70 ( 19 81) 16. Katritzky, A.R., Advances in Heterocyclic Chemistry, Vol. 3, Academic Press, New York, 1964. 17. Brown, H.C., Cahn, A., J. Am. Chem. Soc., 77, 1715 (1955). 18. Brown, H.C., Nakagawa, N., J. Am. Chem. Soc., 78, 2197 (1956). 114

PAGE 125

115 19. Marion, L., in The Alkaloids, Manske, R.H.F. and Holmes, H.L., Ed., Vol. 1, Academic Press, New York, 1950. 20. Karrer, P., Schwarzenbach, G., Benz, F., Solrnssen, u., Helv. Chim. Acta, 19, 811 { 1936). 21. Bodor, N., Shek, E., Higuchi, T., Science, 190, 155 { 19 75) 22. Haynes, L.J., Todd, A.R., J. Chern. Soc., 303, (1950). 23. Stout, D.M., Meyers, A.I., Chern. Rev., 82, 223 (1982). 24. Traber, w., Karrer, P., Helv. Chern. Acta, 41, 2066 (1958). 25. Kuhnis, H., Traber, w., Karrer, P., ibid., 40, 751 (1957). 26. Stein, G., Stiassny, G., Nature, 176, 734 (1955). 27. Fox, H.H., Lewis, J.I., Wenner, w., J. Org. Chem., 16, 1259 { 1951). 28. Lovesey, A.C., Ross, w.c.J., J. Chem. Soc. B, 192 (1969). 29. Segal, R., Stein, G., J. Chem. Soc., 5254 (1960). 30. Dittmer, D.C., Trouty, R.A., J. Arn. Chem. Soc., 86, 91 (1964). 31. Caughey, w., Schellenberg, K.A., J. Org. Chem., 31, 19 78 { 19 6 6 ) 32. Bodor, N., Shek, E., Higuchi, T., J. Med. Chem., 19, 102 (1976). 33. Supple, J.H., Nelson, D.A., Lyle, R.E., Tetrahedron Lett., 1645 (1963). 34. Wang, B.J.S., Thorkton, E.R., J. Arner. Chern. Soc., 90, 1216 (1968). 35. Taylor, G.A., J. Chem. Soc., 3332 (1965). 36. Lyle, R.E., Mallett, S.E., Ann. N.Y. Acad. Sci., 145, 83 (1967). 37. Stock, A., Otting, F., Tetrahedron Lett., 4017 (1968). 38. Hanstein, w., Wallenfels, K., Tetrahedron, 23 585 (1967).

PAGE 126

116 39. Sims, A.F.E., Smith, P.W.G., Proc. Chem. Soc., 282 (1958). 40. Cook, N.C., Lyons, J.E., J. Amer. Chem. Soc., 87, 3283 (1965). 41. Eisner, u., Kuthan, J., Chem. Reviews, 72, 1 (1972). 42. Kuthan, J., Musil, L., Chem. Commun., 34, 3173 (1969). 43. Sugiyama, N., Inoue, G., and Ito, K., Bull. Chem. Soc. Jap., 35, 927 (1962). 44. Sund, H., Biological Oxidations, Interscience Press, New York, 1968. 45. Rodkey, F.L., J. Biol. Chem., 213, 777 (1955). 46. Burton, R.M., Kaplan, N.O., Arch. Biochem. Biophys., 101, 139 (1963). 47. Norris, D.J., Stewart, R., Can. J. Chem., 55, 1687 (1977). 48. Oppenheimer, N.J., Kaplan, N.O., Biochemistry, 13, 4675 (1974). 49. Kim, c.s.H., Chaykin, s., Biochemistry, 2., 2339 (1968). 50. Rafter, G.W., Chaykin, s., Krebs, E.G., J. Biol. Chem., 208, 799 (1953). 51. Choi, K.S., Alivisatos, B., Biochemistry, 2., 190 (1968). 52. Charton, M., J. Org. Chem., 29, 1222 (1964). 53. Taft, R.W., J. Am. Chem. Soc., 79, 1045 (1957). 54. Abeles, R.H., Hutton, R.F., Westheimer, F.H., J. Am. Chem. Soc., 79, 712 (1957). 55. Kosower, E.M., Teverstein, A., Burrows, H.D., Swallow, A..J., J. Am. Chem. Soc., 100, 5185 (1978). 56. Postovskii, I.Y., Chupakhin, O.N., Matern, A.I., Khim. Geterot. Soedin., 10, 1299 (1984). 57. Meijer, L.H.P., van Niel, J.C.G., Pandit, U.K., Tetrahedron, 40, 5185 (1984). 58. Richens, A., Drug Treatment of Epilepsy, Henry Kimpton Publishers, London, 1976.

PAGE 127

117 59. Bruni, J., Wilder, J., Arch. of Neurol., 36, 393 ( 19 79) 60. Meunier, G., Carray, G., Meunier, Y., Therapie, 18, 435 (1963). 61. Pinder, R.M., Brogden, R.N., Speight, T.M., Avery, G.S., Drugs, 13, 81 (1977). 62. Chapman, A., Keane, P.E., Meldrum, B.S., Simiand, J., Vernieres, J.C., Prog. Neurobiol., 19, 315 (1982). 63. Coulter, D.L., Wu, H., Allen, R.J., J. Am. Med. Ass., 244, 785 (1980). 64. Jeavons, P.M., Clark, J.E., Makeshwari, M.C., Dev. Med. Child Neurol., 19, 9 (1977). 65. Simon, D., Penry, J.K., Epilepsia, 19,379 (1975). 66. Manhire, A.R., Espir,M., Br. Med. J., l._, 808 (1974). 67. Hackman, J.C., Grayson, v., Davidorff, R.A., Brain Research, 220, 269 (1981). 68. Hammond, E.J., Wilder, B.J., Bruni, J., Life Sci., 29, 2561 ( 1982). 69. Nau, H., Loscher, w., J. Pharmacol. Exp. Ther., 220, 654 (1982). 70. Roberts, E., Biochem. Pharmacol., 23, 2637 (1974). 71. Kerwin, R.W., Olpe, H.R., Schmutz, M., Br. J. Pharmac., 71, 545 (1980). 72. Loscher, w., Esenwein, H., Arzneimittel-Forsch., 28, 782 ( 19 78). 73. Loscher, w., Nau, H., Arch. Int. Pharmacodyn., 257, 20 (1982). 74. Frey, H.H., Loscher, w., Neuropharmacology, 17, 637 ( 19 78) 75. Hammond, E.J., Perchalski, R.J., Villarreal, H.J., Wilder, B.J., Brain Res., 240, 195 (1982). 76. Batshaw, M.L., Brusilow, s.w., Ann. Neurol., ..!..!_, 319 (1982). 77. Neophytides, A.N., Nutt, J.G., Lodish, J.R., Ann. Neurol., 2._, 389 ( 19 79)

PAGE 128

118 78. MacDougall, L.G., J. Am. Med. Ass., 247, 53 (1982). 79. Camfield, P.R., Bagnell, P., Camfield, c.s., Tibbles, J.A.R., Lancet,.!.., 1198 (1979). 80. Mathew, N.T., Headache, 21, 147 (1981). 81. Lindegaard, K., Ovrelid, L., Sjaastad, o., Headache, 20, 96 ( 1980) 82. Chang, J.C., Gross, H.M., J. Clin. Oncology, J_, 552 (1985). 83. Davison, c., Guy, J.L., Levitt, M., Smith, P.K., Analgesics in CNS, 134, 176 (1961). 84. McEvoy, G.K., Drug Information 84, Amer. Soc. of Hospital Pharmacists, Inc., Bethesda, Md., 1983. 85. Winter, C.A., Flataker, L., J. Pharmacol. Exptl. Therap., 150, 165 (196-5). 86. Ferreira, S.H., Lorenzetti, B.B., Correa, F.M.A., Eur. J. Pharmacol., 53, 39 ( 19 78) 87. Okuyama, S., Aihara, H., Japan J. Pharmacol., 35, 95 (1984). 88. Nakamura, H., Yokoyama, Y., Seto, Y., Kadokawa, T., Shimizu, M., Agents and Actions, 15, 606 (1984). 89. Camba, L., Joyner, M.V., Acta Haemat., 71, 350 (1984). 90. Hughes, J.A., Arthritis and Rheum., 26, 1054 (1983). 91. Green, J., Yoffe, B., Barzilai, D., Better, o.s., Isr. J. Med. Sci. 21, 142 ( 19 8 5) 92. Brouillette, D.G., Pop, E., Unpublished, compound was supplied from within Dr. Bodor's laboratory. 93. Casadio, s., Cousse, H., Mouzin, G., French Patent No. 7 7 13 4 78 ( 19 7 7 ) 94. Raghavan, K.S., Unpublished, compound was supplied from within Dr. Bodor's laboratory. 95. Nakamura, H., Yokoyama, Y., Motoyoshi, s., Ishii, K., Imazu, C., Seto, Y., Kadokawa, T., Shimizu, M., Arzneim. Forsch., Jl., 1555 (1983). 96. Schefler, w.c., Statistics for the Biological Sciences, Addison-Wesley Publishing Co., Reading, Mass., 1979.

PAGE 129

BIOGRAPHICAL SKETCH The author lived his first twenty-four years in Cuyahoga Falls, Ohio. He attended and graduated from Walsh Jesuit High School with honors. He earned college credit in both calculus and spanish while at Walsh Jesuit. He attended the University of Akron for a total of six years. The first four were spent obtaining a B.S. in chemistry, and the last two were spent earning a M.S. in bio-organic chemistry. The author continued his education at the University of Florida where he worked to obtain a Ph.D. in the department of medicinal chemistry. 119

PAGE 130

I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Nicholas s. Bodor, Chairman Graduate Research Professor of Medicinal Chemistry I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Alan R. Katritzky Professor of Chemist I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the d&~27);~ Richard H. Hammer Professor of Medicinal Chemistry I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the degree of Doctor of Philosophy. Hartmut Derendorf Assistant Professor of Pharmaceutics I certify that I have read this study and that in my opinion it conforms to acceptable standards of scholarly presentation and is fully adequate, in scope and quality, as a dissertation for the deg~~~ Marcus E. Brewster Assistant Professor of Medicinal Chemistry

PAGE 131

This dissertation was submitted to the Graduate Faculty of the College of Pharmacy and to the Graduate School and was accepted as partial fulfillment of the requirements for the degree of Doctor of Ph~ ~osophl. 1 /] / J 'f May 1987 r~ ,V_ Dean, Colle~ of Pharmacy Dean, Graduate School