Application of drug design methods for inducing nerve growth factor biosynthesis

MISSING IMAGE

Material Information

Title:
Application of drug design methods for inducing nerve growth factor biosynthesis
Physical Description:
xiii, 160 leaves : ill. ; 29 cm.
Language:
English
Creator:
Kourounakis, Angeliki, 1967-
Publication Date:

Subjects

Subjects / Keywords:
Nerve Growth Factors -- biosynthesis   ( mesh )
Nerve Growth Factors -- pharmacology   ( mesh )
Drug Design   ( mesh )
Drug Evaluation   ( mesh )
Structure-Activity Relationship   ( mesh )
Drug Delivery Systems   ( mesh )
Oxidative Stress -- physiology   ( mesh )
Catechols -- pharmacology   ( mesh )
Catechols -- analogs & derivatives   ( mesh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1995.
Bibliography:
Includes bibliographical references (leaves 144-159).
Statement of Responsibility:
by Angeliki Kourounakis.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
oclc - 50083630
ocm50083630
System ID:
AA00011179:00001

Full Text








APPLICATION OF DRUG DESIGN METHODS FOR INDUCING NERVE
GROWTH FACTOR BIOSYNTHESIS












By


ANGELIKI KOUROUNAKIS


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


1995

























To my father, Dr. Panos N. Kourounakis, and mother, Dr. Lygeri P. Hadjipetrou-
Kourounakis, for their dedication to science.

















"... and in the tomb of an ancient king finds a treasure of gold and

precious jewels. He and his crew load down their ship almost to sinking with all

this treasure, but as they sail he discovers that not only his crew, but that also he

himself is now thinking of settling down, of building estates and villas on the Nile,

of leading the good, comfortable, bourgeoise life. He commands, therefore, that

all this treasure, even the smallest gold coin, be cast overboard, and he exclaims:

"If could choose what gods

to carry on all my ships,

I'd choose both War and Hunger,

that fierce and fruitful pair."

For Odysseus knows well that new horizons are never gained by satisfied

and comfortable men, but by those who are always at war -- with themselves; who

are always hungry -- to explore new regions of thought."


(Kimon Friar, "The spiritual Odyssey of Nikos Kazantzakis")












ACKNOWLEDGMENTS


I would like to thank my advisor Dr. N. Bodor for his excellent guidance,
inspiring input throughout the project, and constant encouragement, as well as
for the opportunity to work on such a uniquely broad spectrum of diverse areas
within the field of pharmacochemistry.
I wish to thank Dr. J. Simpkins for his advice and major contribution by
providing his laboratory facilities for part of this work.
I most gratefully appreciate the valuable help and advice received by Dr.
E. Wu, Dr. L. Prokai, Mrs. N. de Fiebre, Dr. S. Singh, Mrs. E. Simpson, Dr. M-J.
Huang, Dr. H. Farag, and Dr. M. Badawi. I also wish to thank Dr. E. Meyer and
Dr. W. Millard for the use of their labs.
I would like to extend my thanks to members of the Center: Laurie
Johnston, Joan Martignago, Julie Berger, and Kathy Eberst for their helpful
assistance, as well as all friends and colleagues who made this tenure pleasant.
Finally I would like to thank the members of my supervisory committee,
Drs. A.R. Katritzky, M.O. James, and G. Hochhaus, for their interest in the project.












TABLE OF CONTENTS


Page
ACKNOWLEDGMENTS............................................................................................ iv

LIST O F TA BLES....................................................................................................... viii

LIST O F FIG U RES....................................................................................................... ix

A B STRA C T .............................................................................................................. xii

CHAPTERS

1 INTRODUCTION AND BACKGROUND ......................................................... 1

A lzheim er's D isease............................................................................................... 1
Current Treatments for Alzheimer's Disease and Research Areas ..............4...
Nerve Growth Factor and the Treatment of Alzheimer's Disease............... 6
Regulation of NGF Biosynthesis .................................................................. 10
N G F B iosynthesis ...................................................................................... 10
Neurotransmitter Receptor Regulation of NGF Biosynthesis ................11
Steroid Regulation of NGF Biosynthesis in the CNS.......................... 14
Mechanism(s) of NGF Induction................................................................... 15
Potential for Treating Neurodegenerative Diseases with NGF-
Inducing Compounds ............................................................................... 18
BBB and the Redox-Based Chemical Delivery System of Drugs to
the B rain ..................................................................................................... 19
Drug Design Based on Isosteric Replacement or QSAR........................... 21
B ioisosterism .............................................................................................. 22
Quantitative Structure Activity Relationships......................................25
Computer Aided Drug Design and Computational Chemistry..............26

2 CURRENT STUDY AND ITS OBJECTIVES .............................................. 29

Brain-Enhanced Delivery of Potential Neurotrophomodulators..............29
Synthesis and In Vitro Evaluation of Catechol Isosters and
Derivative as NGF-Inducers ................................................................ 32.....3 2
Quantitative Structure Activity Relationships of Catechol
Derivatives on NGF Secretion in L-M Cells........................ 34

3 EXPERIMENTAL AND RESULTS.............................................................. 36

Synthesis.............................................................................................................. 36
Synthesis of 4-Methylcatechol Chemical Delivery Systems...............37







Synthesis of the Redox Analog of Dopamine....................................... 46
Synthesis of the Pyridinium Catechol Derivative and a
Dimethoxy CDS...................................................................................50
Synthesis of 2-Hydroxymethyl-p-Cresol............................................... 53
Synthesis of 5-Methyl- 1-Hydroxy-2-Pyridone..................................... 56
Synthesis of 3- ((N-Methyl- 1,4-Dihydronicotinoyloxy)Methyl) }-
4-(N-Methyl- 1,4-Dihydronicotinoyloxy)Toluene.......................... 57
Synthesis of 3-Hydroxy-4-(N-Methyl-1,4-Dihydro-
nicotinoyloxy)Toluene ....................................................................... 60
Synthesis of 3-Hydroxy-4- {(N-Methyl-1,4-Dihydro-
nicotinoyloxy)Methoxy Toluene .................................................... 63
In Vitro Stability Studies................................................................................. 66
Analytical Method .................................................................................... 66
Stability in Buffers.....................................................................................66
Stability in Biological Media ...................................................................67
In Vivo Distribution Study ............................................................................. 73
Analytical Method .................................................................................... 73
Experimental Procedure ........................................................................... 73
In Vivo NGF-Stimulatory Activity ................................................................78
RNA Isolation ............................................................................................79
Blotting of Total RNA .............................................................................. 80
Preparation of the NGF Probe ................................................................. 81
Radiolabeling of NGF Probe and Hybridization of Nylon
Membrane............................................................................................. 82
Statistical Analysis and Results ............................................................... 84
In Vitro NGFStimulatory Activity ................................................................. 87
Cell Cultures............................................................................................... 87
Determination of NGF level ..................................................................... 88
Nerve Growth Factor ELISA ................................................................... 89
Cellular NGF mRNA Levels ..................................................................... 90
Quantitative Structure Activity Relationships of Catechol
Derivatives (a Theoretical Study)............................................................ 99

4 DISCUSSION ................................................................................................. 114

Synthesis ............................................................................................................ 114
Mono Esterification of 4-Methylcatechol........................................... 114
"Zincke" Type Reactions ....................................................................... 116
Synthesis of 2-Hydroxymethyl-p-Cresol (2HC)................................. 116
Synthesis of 3-Hydroxy-4-(N-Methyl-Nicotinoyloxy) Toluene
Iodide (18S and Attempted Synthesis of (25)............................... 119
In Vitro Stability studies ............................................................................... 120
In Vitro Stability in Buffers.................................................................... 120
Stability in Biological Media................................................................. 121
In Vivo Distribution Studies......................................................................... 125
In Vivo (CNS) NGF-Stimulatory Activity .................................................. 127
In Vitro NGF-Stimulatory Activity .............................................................. 128
Q SA R ................................................................................................................. 131







5 CO N CLU SIO N S ............................................................................................... 135


REFEREN CE S .......................................................................................................... 144

BIO G RA PH ICA L SK ETC H ........................................................................................ 160












LIST OF TABLES


Table page

1. Compounds that induce NGF in cell culture..................................................... 16

2. Compounds that induce NGF in vivo................................................................ 17

3. Effect of treatments on the NGF content in the medium of the C6

glioma cell culture.................................................................................................. 94

4. Effect of treatments on the NGF content in the medium of the L-M cell
culture. ....................................................................................................................... 9 4

5. The set of 23 catechol derivatives and values of their descriptors ........ 104-107

6. The correlation matrix................................................................................. 108-109













LIST OF FIGURES


Figure page

1. Representation of the drug-targetor complex that promotes CNS
retention and accelerated peripheral elimination........................................... 20

2. Catechol and bioisosteric groups..................................................................... 24

3. A) Scheme of 4MC CDSs. B) Metabolism of the CDSs.............................30

4. A) Redox analogue of dopamine and its metabolite. B) 4MC,
isosters and derivative. ....................................................................................... 33

5. Synthetic scheme for the preparation of the 4-methylcatechol
C D S s........................................................................................................................3 8

6. 1H NMR spectrum of 3-hydroxy-4-nicotinoyloxytoluene (A)....................39

7. Expansion of the aromatic region of the 1H NMR spectrum of
3-hydroxy-4-nicotinoyloxytoluene (A)..........................................................40

8. Ultra Violet Spectrum of Dihydro compound J. (Typical UV spectrum
obtained by all dihydro compounds synthesized in this study)..................47

9. Synthetic scheme for the preparation of the redox analog of
dopam ine (4). ...................................................................................................... 49

10. A) Synthesis of the pyridinium catechol derivative (5).
B) Synthesis of the methoxy CDS (7) ............................................................ 51

11. Synthetic scheme for the preparation of A) 2-hydroxymethyl-
p-cresol (2HC) and B) 5-methyl-l-hydroxy-2-pyridone (5MHP)...............55

12. Synthetic scheme for the preparation of 3- {(N-methyl-1,4-dihydro-
nicotinoyloxy)methyl} 4-(N-methyl-1,4-dihydronicotinoyloxy)
toluene (13). ......................................................................................................... 58

13. Synthetic schemes for the preparation of 3-hydroxy-4-
(N-methyl-1,4-dihydronicotinoyloxy)toluene (19)........................................ 61

14. Synthetic scheme for the preparation of 3-hydroxy-4-
{ (N-methyl- 1,4-dihydronicotinoyloxy)methoxy } toluene (25).................... 64








15. pH profile of 4-methylcatechol CDS (E) and its quaternary
m etabolite (D )...................................................................................................... 68

16. pH profiles of CDS 4 (A), and of its quaternary metabolite 3 (B)................ 69

17. In vitro stability in rat brain 20% homogenate of CDSs (E, H, and J)
and their quaternary metabolites (D, G, and I)................................................70

18. In vitro stability in rat blood of CDSs (E, H, and J) and their
quaternary metabolites (D, G, and I). ............................................................... 71

19. In vitro stability of (4) and (3) in various biological media...........................72

20. A) In vivo brain concentration of the CDS J, its quaternary
metabolite I, and final metabolite, 4-methylcatechol (4MC).
B) In vivo blood/brain distribution of quaternary metabolite I..................74

21. In vivo concentrations of CDS 4, its metabolite 3, and final
metabolite 5, in A) rat brain, and B) rat liver...................................................75

22. In vivo concentrations of CDS 4, its metabolite 3, and final
m etabolite 5, in rat blood................................................................................... 76

23. Blood/brain distribution of the pyridinium catechol final
m etabolite 5 ......................................................................................................... 77

24. Effect of 4-MC-CDS (J) on Hippocampal NGF mRNA levels....................85

25. Effect of 4-MC-CDS (J) on Frontal Cortical NGF mRNA levels..............86

26. Schematic representation of the principle of the NGF ELISA..................... 91

27. Effect of 4MC, 5MHP, and 2HC on NGF content in the culture
medium of A) the C6 glioma and B) L-M cells...............................................93

28. Effect of catechol derivative (5) on NGF content in the culture
medium of the A) C6 glioma and B) L-M cells...............................................95

29. Effect of the redox analogue of dopamine (4) and its quaternary
metabolite (3) on NGF content in the culture medium of the L-M cells........96

30. Effect of 4-methylcatechol, isosters 5MHP and 2HC, and catechol
derivative (5) on intracellular NGF content in the C6 glioma cells.............97

31. Effect of confluency of C6 cells on NGF biosynthesis with or
without an NGF inducer (4M C)....................................................................... 98

32. The two low-energy conformations of 4-alkylcatechols............................ 100








33. "Calculated" activity values derived from equation 7 fitted close to
the "experim ental" values................................................................................ 112

34. "Experimental" activity values of the catechol derivatives
plotted against their "calculated" activity values. ...................................... 113

35. Mono esterification of 4-methylcatechol leads to a mixture
of the para and meta isomers........................................................................... 115

36. Reaction of the Zincke type reagent with the amine group of
dopamine and related compounds is believed to go through the
intermediate dianil which cyclizes to the desired pyridinium product........ 117

37. A) Representation of the [3,3] sigmatropic RAR of the intermediate
substituted phenyl benzenboronate. B) Ortho-bridged polyphenol
derivatives. C) Uncatalysed reaction, in the presence of an ether
additive (DME), for the C-ortho site-specific monohydroxymethylation
of p-cresol with formaldehyde........................................................................ 118

38. Degradation and metabolic pathways of 4-methylcatechol CDSs
and their quaternary metabolites in vitro and/or in vivo ............................ 122

39. Degradation and metabolic pathways of the redox dopamine
analogue (4) and its quaternary metabolite (3) in vitro and/or in vivo........ 123

40. A) Oxidation of a catechol derivative to the quinone.
B) A probable mechanism of oxidation of a catechol derivative
to the corresponding quinone ........................................................................... 133

41. Hypothetical energetic diagram for the oxidation of the catechol (A)
via the transition state (B) to the ortho quinone (E) ................................... 134

42. Oxidants and antioxidants believed and proposed to play a role
in the redox reactions of catechol derivatives in cells.
Autoxidation of catechols proceeds via 1-electron transfer reaction
producing radical intermediates. Potential antioxidative or
radical scavenging properties are also depicted. Possible
pathways of glutathione depletion................................................................ 139












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


APPLICATION OF DRUG DESIGN METHODS FOR INDUCING NERVE
GROWTH FACTOR BIOSYNTHESIS

By

Angeliki Kourounakis

August, 1995



Chairman: Nicholas S. Bodor
Major Department: Medicinal Chemistry


The clinical potential of nerve growth factor (NGF) as a therapeutic agent

for Alzheimer's disease is undermined by its inability to cross the blood brain

barrier. The alternative of up-regulating NGF in the central nervous system (CNS)

by NGF-inducing compounds seems more promising, and delivery of such agents

to the brain is therefore of great interest.

An integrated approach, using drug design methods (site-specific delivery,

isosteric substitution, and study of quantitative structure activity relationships),

was undertaken to 1) effectively stimulate NGF biosynthesis in the CNS by

peripheral administration of a brain-targeted chemical delivery system (CDS) of a

known inducer, 4-methylcatechol (4MC), 2) produce novel NGF-inducing







compounds, and 3) elucidate the mechanism of action of catechol derivatives as

NGF-inducers.

Brain-targeted CDSs for 4MC based on the dihydropyridine <-->

pyridinium redox reaction were designed and synthesised, as well as a novel

brain-targeted catechol derivative. In vitro stability studies and in vivo

distribution studies in rats successfully demonstrated the feasibility of sustained

delivery of 4MC and derivative to the rat brain after peripheral administration of

the corresponding dihydropyridine chemical delivery systems, and rapid

peripheral elimination to assure minimal peripheral side effects of the bioactive

compounds. Moreover, the 4MC-CDS evoked a 1.8-fold increase in NGF mRNA

in the hippocampal region of the rat brain.

Bioisosters of 4MC were synthesised and evaluated for in vitro NGF-

inducing activity in two cell lines. Isosteric substitution renders the molecule

inactive in the L-M cell line, raising the interesting question of why the specific

catechol moiety is indispensable for activity. To obtain a quantitative structure

activity relationship, AMI quantum mechanical calculations were performed on a

set of 23 catechol derivatives, and the derived molecular descriptors were

correlated with activity, which was shown to be associated with variables related

to the oxidation of the catechols. It is therefore proposed that oxidative stress,

from the autoxidation of catechols to the quinones, via perturbation in cellular

homeostasis of antioxidant enzymes (e.g., glutathione peroxidase), triggers NGF

biosynthesis and release.













CHAPTER 1
INTRODUCTION AND BACKGROUND



Alzheimer's Disease



In developed nations, the leading cause of senile dementia -- loss of

memory and reason in the elderly -- is Alzheimer's disease (AD), which constitutes

the most common neurodegenerative disease with an overall 20% likelihood of

incidence over the age of 65. Although the aetiology of this progressive

neurodegenerative disorder is presently unknown, it is characterized histologically

by the following: (1) A decline in the number (cell death) or atrophy of neurons in

the limbic system, including the hippocampus, which is central to learning,

memory, and emotion. Large neurons shrink in parts of the hippocampus and the

cerebral cortex while cell bodies and axons degenerate in certain acetylcholine-

secreting neurons that project from the basal forebrain to the hippocampus and

diverse areas of the cortex. (2) Presence of neurofibrillary tangles. The internal

architecture of certain neuronic cells, for example of the hippocampus, undergoes

alteration and their cytoplasm fills with bundles of helically wound protein

filaments known as neurofibrillary tangles. (3) Widespread presence of senile

plaques. The extracellular spaces of the hippocampus, cerebral cortex, and other








brain regions accumulate aggregates of beta-amyloid protein which constitutes the

deposits of amyloid plaques, that are dramatically increased in AD.

Factors that are implemented in the development and progress of AD, apart

from the age-linked changes in the brain, are head trauma, thyroid dysfunction, or

immune system dysfunctions. Many theories exist on the aetiology of the diverse

structural alterations that occur in the aging brain. One, for example, could be that

defects slowly accrue in the DNA (gene dysfunction), proteins, or lipids of the

neuronic cells, while the genetic and environmental factors contributing to this are

closely intertwined; for example, damage to DNA lowers the quantity and quality

of critical cell proteins such as enzymes involved in the machinery designed to

excise and repair faulty nuclear DNA, or enzymes that protect from the initial

damage (e.g., free radical scavengers). On the other hand, chemical modification

(e.g., free radical oxidation) of cellular proteins may in turn alter cellular

functions, while substantial decline with age in enzymes that inactivate free

radicals, such as superoxide dismutase (SOD) and catalase, potentiate the

oxidative damage on proteins. Thus, whether the damage accrued in DNA is

resulting in increased oxidation of enzymes or whether the oxidation of enzymes

occurs first and leads to accumulation of DNA deficits is an interesting question,

although probably both sequences occur (Selkoe, 1992).

There is now increasing evidence indicating that probably most diseases at

some point during their course involve free radical reactions and tissue injury.

Both acute and chronic degenerative diseases are thought to involve free radical

reactions in tissue injury which may be involved in multiple sites and at different








stages of the disease (Packer, 1992). The brain, and nervous system, is especially

prone to radical damage. Cells of the central and peripheral nervous system,

which are relatively unique in the body being postmitotic and irreplaceable, are

relatively vulnerable to oxidative stress (denoted as a disturbance in the

prooxidant-antioxidant balance in favor of the former), since (1) the brain

consumes approximately 20% of total body oxygen, (2) the membrane lipids are

very rich in polyunsaturated fatty acid side chains (susceptible to lipid

peroxidation), (3) the brain is poor in catalase activity and has only moderate

amounts of SOD and glutathione peroxidase (GSH-Px), and (4) some areas in the

brain (e.g., substantial nigra) are rich in iron and iron ions can stimulate free radical

reactions) (Halliwell, 1992).

Consequences of oxidative stress in the molecular level include lipid

peroxidation, DNA and protein damage, rises in intracellular free Ca2+

(particularly damaging to neurons and thought to be the primary causative event in

mediating necrotic neuronal death), and depletion of glutathione (GSH).

Oxidation of free fatty acids produces free radicals that may contribute to neuronal

damage via inflammatory reactions or may act as mediators of excitotoxicity and

further enhance glutamate release. Excessive concentrations of excitatory amino

acids (EAAs) kill neurons and elevated concentrations of EAA are found in many

models of neurological disorders such as stroke, epilepsy, and head/spinal cord

trauma (Bigge and Boxer, 1994).

The concept that oxidative stress is important in the pathogenesis of

Parkinson's disease is being well substantiated in recent years (Adams and








Odunze, 1991, Jenner et al., 1992), while an increasing body of experimental data

support the hypothesis that an imbalance in the regulation of oxygen-derived free

radical production may result in brain damage and possibly contribute to the

pathogenesis of AD neuropathology (Cohen, 1985, Halliwell and Gutteridge,

1989, Sinet and Ceballos-Picot, 1992).



Current Treatments for Alzheimer's Disease and Research Areas



The "cholinergic hypothesis" (cholinergic deficit) has led to many attempts

to provide cholinergic replacement therapy. Various drugs have been developed

for this purpose that show, however, only moderate effectiveness. These include

cholinergic drugs (muscarinic cholinomimetics) and acetylcholinesterase inhibitors

such as Tacrine (THA), Amiridin, E2020, and Velnacrine (Takeda et al., 1995,

Cassidy et al., 1994, Iversen and Hargreaves, 1994).

Since cholinergic transmission can be stimulated indirectly via the y-amino-

butyric acid (GABA) system, which appears to exert inhibitory control on

acetylcholine (ACh) neurons through GABA-(A) receptors, intervention with

specific GABA-(A) receptor antagonists should theoretically be applicable as a

therapy in AD (Krogsgaard-Larsen, 1993 and 1994, Turner, 1994).

Glutamic acid (Glu) receptor antagonists are of interest as neuroprotective

drugs since hyperactivity of central Glu neurons may be one of the primary causes

of degeneration of ACh neurons in AD. However, because normal function of

central ACh neurons appears to be dependent on stimulation by Glu neurons and








the observed loss of Glu neurons in the progression of AD resulting in

hypoactivity of EAA neuronal pathways has recently been associated with learning

and memory deficits, drugs capable of both protecting and activating EAA

receptors may be of therapeutic interest (Krogsgaard-Larsen, 1993 and 1994;

Stone, 1994).

Nootropic agents such as idebenone and piracetame are also currently being

used although their mechanism of action is not yet well elucidated (Schindler,

1994).

Another area of interest in recent years has been the "beta-amyloid theory"

of AD (Cordell, 1994). For example, a compound that is a selective inhibitor of

the enzyme that processes the amyloid precursor protein to yield beta-amyloid has

been reported. The potential use of anti-inflammatory drugs (indomethacin has

shown some promising results [Rogers, 1993]) is also of interest since the senile

plaque is also a site of a local inflammatory response in the brain. Intervention by

inhibiting possible processes involved in oxidative stress and the use of

antioxidants like vitamin E is also under investigation (Kagan et al. 1992) (e.g.,

among the numerous factors involved in this aspect, aluminium ions, which have

been implicated in AD, as well as calcium ions, may induce molecular structural

changes and degenerative processes [Szabo and Kretzshemar, 1994]).

A more effective strategy, however, would be to slow or stop the

progression of the neurodegenerative changes that accompany the development of

senile dementia rather than to try to restore function to an already damaged brain.

One such approach is the administration of neurotrophic factors.








Nerve Growth Factor and the Treatment of Alzheimer's Disease



Survival, maintenance, and differentiation of neurons in the peripheral and

central nervous system are affected by agents -- most of them polypeptides --

called neurotrophic factors. One of the best such characterized trophic factor is

nerve growth factor (NGF) which is produced by glial and neuronal cells, both of

which also express NGF receptors. Although NGF was discovered and even

sequenced several decades ago, it is only in the last years that the role of this

protein, and a set of closely related factors that all together constitute the

neurotrophin family (including brain derived neurotrophic factor, BDNF,

neurotrophin-3, NT-3, and neurotrophin-4, NT-4), is slowly being more fully

understood. Nevertheless, the range of NGF actions and a mechanistic

explanation for its role have not been established yet. For example, apart from its

crucial role in the regulation of the developing sympathetic and sensory systems,

in promoting survival of injured cholinergic neurons of the basal forebrain, and in

regulating neurite outgrowth and neurotransmitter synthesizing enzymes, a

possible role for NGF in the regulation of oxidant-antioxidant balance has been

recently proposed (Perez-Polo et al., 1990) as part of a molecular explanation for

the known effects of NGF on neuronal survival during development, after injury,

and in the aged CNS. Also, recent studies indicate that NGF protects against

excitotoxicity by preventing the excessive elevation of Ca2+ (stabilizing Ca2+

homeostasis) and oxidative damage via an increase in the levels of detoxifying








enzymes in the brain (Jackson et al., 1990 a and b, Cheng et al., 1991, Mattson et

al., 1993 a and b, Zhang et al., 1993).

In the CNS, high expression of NGF and/or NGF receptor genes is shown

in the regions containing cholinergic neurons or their fibres. Cholinergic neurons

of the mammalian basal forebrain innervate the hippocampus, the neocortex, and

other structures in which the produced NGF can be retrogradely transported to the

basal forebrain cell bodies and exert a trophic effect, activating the expression of

specific proteins like choline acetyltransferase (ChAT) -- the key enzyme for

regulation of acetylcholine synthesis. NGF's activity may normally support the

viability and function of these neurons (Hagg et al., 1988, Lapchack, 1993).

The cholinergic system in the human brain is important for memory and

other cognitive functions. Furthermore, studies on aged rats have demonstrated

the involvement of the forebrain cholinergic system in age-related learning

impairments (Harbaugh, 1989).

A common feature of neurodegenerative diseases is selective dysfunction

and death of specific neuronic populations in the brain. In AD, the degeneration of

the basal forebrain cholinergic neurons along with their axons results in a loss of

cortical and hippocampal ChAT activity which is associated with the progressive

loss of cognitive function. Attempts to restore brain cholinergic function by

administration of acetylcholine precursors, cholinesterase inhibitors, or cholinergic

agonists are limited in success. This suggests that successful treatment may be

possible only by pharmacological agents that act on some very fundamental

neurochemical defect that is responsible for the observed impairments.








A strategy for therapeutic intervention might include administration of a

specific trophic factor to prevent or retard neuron loss. The knowledge of sites of

synthesis of NGF in the nervous system as well as distribution of NGF receptors

suggests a possible future clinical use of NGF to protect and repair in the nervous

system. A wealth of evidence shows that treatment with NGF (intracerebral

infusion) effectively attenuates lesion-induced cholinergic deficits and cognitive

impairments in animal models (Fisher et al., 1987). In addition, recent clinical

study with chronic NGF treatment (infusion of NGF in the brain) in an

Alzheimer's patient showed promising results such as increased blood flow,

[11C]nicotine uptake, and binding in the cerebral cortex, in addition to the

normalized EEG patterns and improved performance in word recognition tests

(Olson, 1993).

However, as already perceived, NGF's use in CNS disorder therapy is

limited not only by the difficulty of human NGF production on a large scale, but

most importantly by its inability to cross the blood brain barrier (BBB). Thus

several different strategies have been considered in order to increase the

neurotrophic effect of NGF on its target sites. These include (1) direct

intracerebral infusion of the NGF protein; (2) slow-release biodegradable

implants -- instead of chronic infusion, incorporation of NGF in biodegradable

polymer capsules or microspheres would provide an implantable slow release NGF

source; (3) carrier-mediated transport across the BBB -- coupling of NGF, for

example, to an antibody to the transferring receptor (CNS blood vessels are rich in

this receptor) would theoretically concentrate NGF in brain vs. periphery








following i.v. injection; (4) grafting NGF-producing cells -- intracerebral

transplantation of cells capable of NGF synthesis for example Schwann cells or

cells that have been genetically engineered to produce and secrete large amounts

of NGF; (5) direct gene transfer to the brain -- transfection of nondividing

neuronal or glial cells with NGF producing genes; (6) developing NGF receptor

agonists -- increasing knowledge of the tertiary structure of NGF and receptor

binding and activation domains enhances the possibility of developing low

molecular weight agonists with NGF-like effects capable of passing the BBB.

An interesting alternative to the previous methods, which are limited in

application and efficacy, is the use of pharmacological agents that control

endogenous NGF production (that enhance NGF biosynthesis and release in the

CNS), an approach that has been the focus of several studies in the past few years.

Delivery to the brain of such neurotrophomodulatory agents would not only make

NGF available to neurons without brain surgery (and consequently BBB damage),

but also would produce NGF in specific CNS cell populations by targeting

selected transmitter receptor subtypes, and thus this strategy would have the

related advantage that the needed extra supply of NGF would be produced at

naturally occurring NGF-dependent sites in the brain. Although at present the

mechanisms that regulate NGF biosynthesis are still unknown, the inducibility of

NGF has been demonstrated in a number of in vitro and in vivo systems by a

variety of agents. Compounds with structural and functional diversity are known

to induce NGF, such as cytokines (e.g., interleukin-1, -4, and -5) (Friedmann et al.,

1990 and 1992, Yoshida et al., 1992, Steiner et al., 1991, Vige et al., 1991,








Awatsuji et al., 1993), 1,25-dihydroxyvitamin D3 (Wion et al., 1991, Saporito et

al., 1993 b), neurotransmitters (e.g., epinephrine, dopamine) (De Bernardi et al.,

1991), and steroids (e.g., dexamethasone, 17-b-estradiol) (Mocchetti, 1991).



Regulation of NGF Biosynthesis



NGF Biosynthesis



Usually NGF refers to the P3-NGF which is a protein dimer consisting of

two identical monomers, held together by noncovalent bonds, each consisting of a

118-amino-acid residue. The biologically active protein is generated by cleavage

from a larger precursor that is encoded by a specific mRNA. Changes in the

biosynthesis of NGF can be estimated by determining the content of the specific

mRNA encoding for the precursor and combining it with the amount of the

biologically active peptide. (If NGF content increases without a change in NGF

mRNA, it can be presumed that NGF accumulates because utilization, catabolism,

or axonal transport are decreased and not because of increased biosynthesis). It

should also be noted that the increase in NGF mRNA elicited by the variety of

agents that will be mentioned below is not due to non-specific change in total

mRNA synthesis in the cells, since the mRNA for the stable structural protein

cyclophilin, or for other proteins like proenkephalin and calmodulin, is unchanged.








Neurotransmitter Receptor Regulation of NGF Biosynthesis



BAR activation

In the rat C-6 glioma cell line (a CNS-derived glial tumour) (Schwartz et al., 1977,

Schwartz and Costa, 1988, Dal Toso et al., 1987 and 1988, Follessa and

Mocchetti, 1992), as well as in pure cultures of type-1 astrocytes from rat cerebral

cortex (that express high levels of BARs) (Schwartz et al. 1990) and also in

fibroblast cultures, it has been shown that P-adrenergic receptor (BAR) activation

increases NGF mRNA and protein, which leads to increased secretion of NGF

from these cells into the culture medium. The BAR-agonist isoproterenol (ISO)

leads to an increase in NGF mRNA and NGF content within 2-3 hs. This effect is

blocked by the P-blocker propranolol but not by the a-blocker phentolamine. The

a-agonist phenylephrin has no effect on NGF mRNA level.

There has been some controversy about whether or not this stimulatory

effect of ISO (and of other catecholamines such as norepinephrin, NE, and

dopamine, DA) is due to its chemical structure (catechol part) and, thus, not

mediated by adrenergic receptors. It was suggested that the catechol ring of

catecholamines is essential and the aliphatic side chain enhances their effect while

the terminal amino-residue is not essential.

A series of synthetic analogues of catecholamines, such as 4-alkylcatechols

(Furukawa et al. 1990), increases the content of NGF mRNA and protein in mouse

fibroblast cells, while salbutamol (a 13-agonist with no catechol ring) does not.

This suggests that the stimulatory effect of catecholamines on NGF biosynthesis is








not mediated by catecholaminergic receptors, and that the possibility exists of

developing a variety of synthetic molecules able to enhance endogenous NGF

production. However, another BAR agonist that does not possess the

catecholamine typical structure, clenbuterol, elicits an increase of NGF

biosynthesis in C6 glioma cells similar to that elicited by ISO. This effect is

blocked by the 03-2 antagonist ICI 118,551, suggesting that BAR-2 subtypes may

be more involved in NGF-biosynthesis regulation (Mocchetti, 1991).

Molecular mechanisms (involved in regulation of NGF synthesis)

BAR stimulation activates adenylate cyclase which increases the

intracellular c-AMP level. Thus NGF gene expression can be mediated via the

activation of c-AMP-dependent protein kinase A. This hypothesis was confirmed

by the increase in NGF mRNA content elicited by addition to glioma cells of

dibutyryl-c-AMP (a lipophilic c-AMP analogue), by addition of forskolin which

directly activates the catalytic subunit of adenylate cyclase, and by addition of

PGE to stimulate prostaglandin E receptors that also lead to c-AMP

accumulation. (Moreover, PGEI 's effect, in increasing NGF, is potentiated by 3-

isobutyl-1-methylxanthine, IBMX, a phosphodiesterase inhibitor [Dal Toso et al.,

1988].)

On the other hand, an involvement of protein kinase C in the regulation of

NGF mRNA content has also been suggested since the phorbol ester 12-0-

tetradecanoyl phorbol-13-acetate (TPA), an activator of protein kinase C, induces

NGF gene expression in mouse fibroblast cells (Wion et al. 1990, D'Mello et al.

1990).








Furthermore, NGF gene expression by activation of protein kinase A or C

may involve an intermediate process: the catalytic subunit of the protein kinase

phosphorylates a protein (cytosolic or nuclear) which binds to a specific promoter

region of the NGF gene. Nevertheless, studies have shown that another

intermediate step may be involved: NGF gene expression may be regulated by

nuclear proteins ("third messengers") that are encoded by early-response genes

like that of c-fos (Mocchetti et al. 1989, Heumann et al. 1991), since it has been

shown that BAR stimulation and phorbol esters increase c-fos mRNA and c-fos

protein preceding the increase in NGF mRNA.

Other neurotransmitters

Since BAR-stimulation induces NGF biosynthesis, it can be assumed that

noradrenaline might be the physiologic regulatory stimulus in the normal brain.

However, recent findings also suggest that other neurotransmitters could play a

role in NGF regulation. GABAergic and glutamatergic activity seem to provide an

inhibitory and excitatory input in NGF mRNA expression (Zafra et al. 1990, 1991,

and 1992). Kainic acid, a glutamatergic receptor agonist, enhances NGF mRNA

content in neurons of the hippocampus. So do experimental seizures (by electrical

stimulation or bicuculline injection) in which it is presumed that excitatory amino-

acid transmission overtakes inhibitory GABAergic transmission.

An antagonist of the N-methyl-D-aspartate -- NMDA -- sensitive glutamate

receptor, MK 801, did not block NGF mRNA increase while 6-cyano-7-

nitroquinoxaline-2,3-dione, CNQX (a non NMDA receptor antagonist), prevented

the increase elicited by electrical stimulation suggesting that stimulation of non








NMDA receptors are involved in NGF mRNA expression. In addition, kainic

acid's and bicuculline's induction of NGF mRNA is blocked by diazepam (Zafra

et al. 1990).

Other neurotransmitters such as cholinergic agonists methacholinee,

carbamylcholine and nicotinic acid), serotonin, and histamine fail to change NGF

content.



Steroid Regulation of NGF Biosynthesis in the CNS



Adrenal steroids released during stress increase the content of NGF mRNA

and protein in specific brain structures. In adrenalectomized rats, hippocampal

NGF decreases. Dexamethasone (a synthetic glucocorticoid) and reserpine (which

increases blood corticosterone) both elicit an increase in the amount of cortical

NGF mRNA (Saporito et al., 1993 a, Fabrazzo et al., 1990). In addition, the

previously mentioned bicuculline-seizure-mediated increase in NGF expression

could also be due to the concomitant increased plasma corticosteroids, suggesting

that adrenal steroids could be the common mechanism for induction of NGF by

drugs that elicit a stressful situation.

Other hormones that are capable of inducing NGF expression include

thyroxin and 17-p-estradiol. In the C-6 glioma cell culture, 17-p-estradiol

enhances the de novo synthesis of NGF protein and increases the amount of NGF

secreted by these cells into the surrounding medium (Perez-Polo et al. 1977).








Mechanism(s) of NGF induction



It has become clear that diverse mechanisms exist by which a variety of

agents enhance NGF biosynthesis and secretion, both in vitro and in vivo.

Presently however, there is only preliminary knowledge on how specific

compounds upregulate NGF biosynthesis.

The structural diversity of NGF-inducing compounds but most importantly

the variety of cell cultures used to demonstrate activity in vitro (Table 1) or in vivo

systems (Table 2) provide additional complications since direct comparisons of

unrelated systems is usually not possible. Furthermore, specific compounds have

been shown to have differing NGF-inducing profiles in various cell types.

Catecholamines are potent in stimulating NGF synthesis and secretion in

the C6 glioma cell line and the fibroblastic L-M cell line. In C6 cells their activity

seems to be associated with p-adrenergic receptor stimulation (as discussed

previously) since P-adrenergic antagonists block their effect while other

structurally diverse 3-agonists also appear to be effective.

In the L-M cell line however, the effect of catecholamines seems not to be
mediated by 0-adrenergic receptor activation. Evidence indicates that their effect

is due to the catechol part of their molecule and not mediated by adrenergic
receptors that are present in this cell line, since alpha or beta adrenergic
antagonists do not block their effect. In addition, m- or p-dehydroxy precursors of
catecholamines (4- or 3- aminoalkyl phenols), m-O-methylated metabolites (2-
methoxy-4-aminoalkyl phenols), or non-catechol adrenergic agonists show no









Table 1. Compounds that induce NGF in cell culture.


Cell type


Steroids
Dexamethasone
Aldosterone
1,25-Dihydroxyvitamin D3
Catechols
Dopamine and epinephrine

Dopa and norepinephrine

Isoproterenol



4-methylcatechol


4-alkylcatechols and diacetylated derivatives

Benzoquinone derivatives
1,4-benzoquinone
Idebenone
Vitamin K3
Pyrroloquinoline quinone
Growth factors and cytokines
IL-1p3


IL-4 and 5
Acidic fibroblast growth factor (aFGF)
Basic fibroblast growth factor (bFGF)
Tumor necrosis factor-a (TNF-a)
Epidermal growth factor
TGF-pl
Miscellaneous compounds
Retinoic acid
Prostaglandin El
Prostaglandin E2
Lipopolysaccharide
Phorbol 12-myristate 13-acetate (PMA/TPA)
HTLV tax protein
fos gene product
Serum factor
Mixed gangliosides
Hericenones C, D, and E
Fellutamide A
Propentophylline
Dibutyryl cAMP
Forskolin and cAMP analogs


Kainic acid


Primary hippocampal neurons
Primary hippocampal neurons
Mouse fibroblast L929 cells

Mouse fibroblast L-M cells
Quiescent primary astrocytes
Mouse fibroblast L-M cells
Primary Schwann cells
Mouse fibroblast L-M cells
Quiescent primary astrocytes
C6-2B rat glioma cells
Primary rat Schwann cells
Primary cortical astrocytes
Mouse fibroblast L-M cells
Quiescent primary astrocytes
Mouse fibroblast L929 cells
Mouse fibroblast L-M cells
Quiescent primary astrocytes

Primary mouse astroglial cells
Primary mouse astroglial cells
Primary mouse astroglial cells
Primary mouse astroglial cells

Cultured explants of adult rat sciatic nerve
Primary rat mesangial cells
Primary rat astrocytes
Hippocampal cultures
Mouse astrocytes
Primary rat astrocytes
Primary rat fibroblasts and astrocytes
Primary rat astrocytes
Primary rat astrocytes
Primary rat astrocytes


Mouse fibroblast L929 cells
C6-2B rat glioma cells
Primary rat hippocampal cultures
Primary rat hippocampal cultures
Mouse fibroblast L929 cells
NIH 3T3 cells
Transgenic mice
Mouse fibroblast L929 cells
Primary rat Schwann cells
Primary mouse astroglial cells
Mouse fibroblast L-M cells
Primary mouse astroglial cells
C6-2B rat glioma cells
C6-2B rat glioma cells
Primary Schwann cells
Primary rat hipoocamoal neurons


Compound









Table 2. Compounds that induce NGF in vivo.


Site of induction


Interleukin- 1lp (IL- 1p)


Transforming growth factor-Pfl (TGF-pl1)
Thyroid hormone

Colchicine

Kainic acid


NMDA
Clenbuterol
Isoproterenol
1.25-Dihydroxyvitamin D3
4-Methylcatechol
Dexamethasone


Reserpine
Testosterone
Corticosterone
Yohimbine
Idebenone


Acetvl-L-caritine


Adult rat hippocampus and cortex
Astrocytes of adult rat basal forebrain and
hippocampus
Adult rat hippocampus
Female mouse salivary gland
Aged rat hippocampus
Multiple adult rat brain regions, including
hippocampus, cortex, and basal forebrain
Adult rat hippocampus and cortex
Multiple adult rat brain regions, including
hippocampus, cortex, and basal forebrain
Adult rat hippocampus
Adult rat cortex
Adult rat cortex and hippocampus
Adult rat cortex and hippocampus
Adult rat cortex and hippocampus
Adrenalectomized adult rat cortex and
hippocampus
Adult rat hippocampus
Adrenalectomized adult rat cortex
Mouse submaxillary gland
Mouse submaxillary gland
Adult rat hippocampus
Aged rat frontal and parietal cortex and
hippocampus
Adult rat hippocampus


Compound


~











stimulatory effect on NGF content (Furukawa et al., 1986 a and b, 1987 and 1989).

On the other hand, non-amine catechols like 4-alkylcatechols (1, 2 dihydroxy-4-

alkylbenzenes) demonstrated high activity. A simple catechol derivative such as

4-methylcatechol (4MC) is one of the most potent stimulators of NGF synthesis

and secretion in L-M cells, demonstrating therefore that the only structural

requirement for activity lies within the catechol functional moiety of the molecule,

although the specific mechanism by which catechol or catecholamine analogues

stimulate NGF synthesis in this cell line is still under investigation.



Potential for Treating Neurodegenerative Diseases with NGF-Inducing Compounds



There is substantial evidence demonstrating the feasibility of developing an

NGF-inducing therapeutic. The significant correlation between in vivo and in

vitro results of several compounds is encouraging. It would be important to

understand the, as yet ill-defined, multiple mechanisms of NGF induction for the

development of an NGF-inducing therapeutic, by eliminating compounds that act

via non-specific ways, and by developing a drug with the appropriate specificity

and efficacy by rational (mechanism-based) design.

Potential complications, such as inability of these compounds to penetrate

the BBB after peripheral administration or toxicity/side effects due to up-

regulation of NGF in the periphery, may successfully be overcome by delivering

the compounds that enhance NGF selectively into the brain by the use of an








appropriate brain-targeted chemical-delivery system such as the one based on the

dihydropyridine <--> pyridinium redox reaction.



BBB and the Redox-Based Chemical Delivery System of Drugs to the Brain



The major obstacle in drug delivery to the brain is the BBB, a complex of

morphological and enzymatic components that retard the passage of both large and

small molecules that are not essential for cerebral function while housing specific

transport systems for essential molecules. The tight junctions of the cerebral

capillary endothelial cells exclude hydrophilic compounds and high molecular

weight substances from entering the brain while lipophilic compounds, utilizing

transcellular transport, can readily pass through the phospholipoidal membranes

and enter the brain.

A general method that has proven useful in selectively enhancing drug

delivery to the CNS is the chemical delivery system (CDS) that exploits the

distinct properties of the BBB. This CDS is based upon the dihydropyridine <-->

pyridinium salt redox reaction. In this redox-delivery system, the lipoidal

dihydropyridine moiety (Figure 1) is covalently attached to the drug, thus

increasing its lipophilicity and its BBB permeability. Upon systemic

administration, and after extensive distribution, the dihydropyridine is oxidized to

the charged pyridinium ion in the brain and in systemic tissues by the same means

as the ubiquitous NADPH <--> NADP redox system. The charged pyridinium-

drug moiety is thus retained, or "locked", in the brain since the BBB prevents










Nicotinic Acid + Drug-OH

V o
r O-Druq

1 CH, I
2 NaSO INaHCO,


0
O-Drug


CH3
I


Figure 1. Representation of the drug-targetor complex that promotes CNS
retention and accelerated peripheral elimination.








rapid reequilibration of polar species, and enzymatic hydrolysis of the drug-carrier

complex results in a sustained release of active drug. Meanwhile, in the periphery,

the ionized pyridinium-drug moiety can be rapidly cleared by renal or biliary

excretion due to its increased hydrophilicity (Figure 1). The overall result is that

peripheral (side) effects or toxicity should be reduced by preventing significant

accumulation of the parent active agent while it is specifically delivered to its site

of action (bioreceptors in the CNS). Consequently, apart from improvement of its

therapeutic index, central toxicity is also attenuated since the majority of the active

species is present in the form of an inactive carrier complex. Since it was first

proposed in 1978 by N. S. Bodor, the brain-targeted CDS has been extensively

applied to various pharmacologically active agents (Bodor et al., 1981a and b, and

1983).



Drug Design Based on Isosteric Replacement or OSAR



The Therapeutic Index (T.I.), defined as the ratio of the toxic dose by the

effective dose, is the most important characteristic of a drug. Among the principal

goals of Drug Design is the improvement (increase) of the T.I. of a drug either by

increasing the toxic dose, or in other words reducing the toxicity of the active

substance (an approach which is accomplished with, for e.g., the development of

Site Specific Chemical Delivery Systems, such as the one described previously),

or else by reducing the effective dose of a drug which is the effort to increase

activity or potency of the active substance. One of the approaches towards the








latter is by molecular manipulations of the active compound of known chemical

structure (the "lead"). Lead finding includes the identification of novel

compounds displaying an interesting biological activity and compounds with a

newly discovered biological activity. Among the various methods for lead

optimisation, but also for (further) elucidation of the mechanism of action of a

bioactive substance, are bioisosteric replacement and Quantitative Structure

Activity Relationship (QSAR) methods.



Bioisosterism



"Every physiologically active compound of known structure is a challenge

to the medicinal chemist a challenge either to better it, or merely to equal it.

There are numerous ways of attacking such a problem. One of the methods which

has been used frequently, very often with success, is that of isosteric replacement"

(H.L. Friedman).

The concept of isosterism was first introduced by Langmuir (1919) who

expanded the principle of similarity in properties of the elements in the same

group of the periodic table (due to the same electronic configuration) to molecules

or groups of atoms which possess the same number and arrangement of electrons.

Isosters have similar stereochemical and electronic properties. Isoelectronic

groups have similar electronic structure for e.g. same total charge and charge

distribution, while isosymmetric have similar stereochemistry. In related bioactive

compounds chemically more-or-less equivalent groups (isosteric) can be








distinguished in which the equivalence may include, apart from steric or electronic

properties, properties such as lipo or hydrophility. A more biological approach in

this field was given by Friedman who extended Langmuir's older concept of

isosterism by introducing the principal of bioisosterism. Bioisosteric groups are

chemical groups which are considered to be equivalent and therefore

interchangeable as far as their contribution to biological activity is concerned.

Thus, compounds are termed "bioisosteric" if they fit the broadest definition for

isosters and have the same type of biological activity.

A rational approach of lead optimisation by molecular modifications is the

replacement of components/substituents on the active molecule with others of

similar electronic configuration and/or stereochemistry by applying Friedman's

rules of bioisosterism.

Several bioisosteric groups of the catechol moiety are known as shown in

Figure 2. Since this moiety, contained in several P-adrenergic compounds

catecholaminess), is sensitive to metabolism by COMT (catechol-O-

methyltransferase) (Guldberg et al 1975), analogues were sought that would not be

substrates for this enzyme. Thus, structures that would substitute for the catechol

function were developed, basically in order to discover drugs with higher 3-

adrenergic activity/selectivity and longer duration of action.














<[OH
NH-SO2-CH3


K-


XOHO
Xo

X = 0, NR
H

CXN


Figure 2. Catechol and bioisosteric groups.








Quantitative Structure Activity Relationships



Based on the simple yet fundamental reasoning that some form of chemical

interaction, between the molecules of a drug and part of the molecules composing

the biological object, is responsible for the biological effect of a compound, one

can expect that chemical properties of the drug are decisive for its biological

potentialities, and thus, that structure activity relationships (SAR) are inherent in

drug action.

The study of the relationship between molecular structure and activity could

be considered to have started with Mendeleev and the periodic classification of the

elements and has continued on to quantum chemistry and the wide development of

computers.

Assuming that the chemical structure is determinant of a drug's action, then

certain quantitative properties of the molecule should be correlated with the

quantitative expression of the biological activity. The idea that chemical structure

of drugs could be correlated mathematically with their biological effect, was first

expressed in the late 19th century by Crum-Brown and Fazer. The semiempirical

approach to SAR by Meyer and Overton, the development of Hammett's a

constants for the electronic effect of substituents on rates and equilibria of organic

reactions, the definition by Taft of the steric parameter Es, and the development of

hydrophobic parameters, 7t andf, by Hansch and Rekker, using the partition

coefficients from the octanol/water model system were the sources that gave the

tools for handling the problem of SAR on a quantitative basis (Hansch, 1971).








Further progress in Quantitative Structure Activity Relationships (QSAR), has

been indispensably connected with the use of the rapidly developed computers and

computational chemistry.

QSAR is more than a means for optimizing activity: it is based on the

assumption that the relative importance of physicochemical properties for the

biological activity of compounds can be described in numerical terms, thus

objectively rationalising the interaction of drugs with macromolecular systems.

The most important physicochemical parameters related to activity of compounds

are electronic (e.g. dipole moments, quantum chemical parameters), hydrophobic

(e.g. log P), and steric parameters derived from linear free energy relationships or

from geometric considerations. Estimation of such molecular properties on a

theoretical basis has been facilitated by the development and advancement of

computational chemistry.



Computer Aided Drug Design and Computational Chemistry



Progress during the past few years in the field of Computer Aided Drug

Design has rendered techniques that can extend QSAR by representing electronic

and steric molecular properties graphically in 3D and quantifying them. Among

the various methods for evaluation of molecular properties on a theoretical basis

are force-field or empirical potential energy methods (molecular mechanics) and

molecular orbital calculations (quantum mechanics).








Molecular Mechanics (M.M.) is a method to calculate energies based on the

assumption that the total energy of a molecule is the sum of the individual

contributions of the electrical and mechanical energy. The internal energy of a

molecule can be expressed as a potential function which has a minimum

corresponding to the equilibrium geometry, and is a sum of a number of potential

functions related to bond stretching, valence angle bending, torsional angle

changes, non-bonded (van der Waals) interactions, electrostatic (coulombic)

interactions and hydrogen bonding. Thus, M.M. uses a "classical elastic force-

field" to make predictions about the equilibrium structure of a molecule (stable

conformations). Force-fields used most frequently are MM2, AMBER and

CHARMM, although no method can guarantee finding the absolute lowest energy

--the global minimum-- since energy minimization will stop at the first local

minima encountered without realising that more stable minima may be accessible.

In Quantum Mechanical Calculations, molecular energies are calculated by

using the Schroedinger equation with the Molecular Orbital (M.O.) formalism,

which can provide greater accuracy along with the ability to model electronic

effects not treated by molecular mechanics. The Schroedinger equation of a given

molecular system can be solved either with no approximations at all (ab initio) or

with the introduction of some approximations semiempiricall). The GAUSSIAN

and HONDO series are typical ab initio programs. Semiempirical methods use

several approximations including CNDO, INDO and NDDO. Dewar has used

these approximations to develop a number of semiempirical methods. In the

approximations used, various terms were parameterized using experimental data,








in order to generate molecular orbital methods that provided "chemical accuracy".

Thus, MNDO and AMI (Austin Model 1) were developed based on the NDDO

approximation. They have several advantages over the other models, and, AMI,

which is also faster, is considered to represent, in its present form, the best that

can be achieved using the NDDO approximation as a basis.

Quantum chemical calculations can provide detailed insight into the

electronic nature of the molecular structures. Properties reproduced by this

method are for e.g. charge distribution, dipole moments, electronic density,

HOMO (highest occupied molecular orbital) energy, LUMO (lowest unoccupied

molecular orbital) energy, transition energy, 7r electron energy, resonance or

delocalization energy, electron density, net charge, heat of formation, molecular

geometries. Furthermore, predictive models for partition coefficients (LogP) have

been developed in recent years based on M.O. calculations. Most of these

parameters along with others have shown at times to correlate with the activity of

drugs, rendering quantum chemical calculations to be of great value in QSAR

studies.













CHAPTER 2
CURRENT STUDY AND ITS OBJECTIVES



Brain-Enhanced Delivery of Potential Neurotrophomodulators



4-methylcatechol (4MC) is a potent NGF stimulator both in vitro (mouse

fibroblast L929 cells and L-M cells) (Carswell et al., 1992, Furukawa et al., 1991)

and in vivo, in the rat peripheral nervous system (Kaechi et al., 1993, Hanaoka et

al., 1994) and in the rat brain after intracerebroventricular injection (Saporito et al.,

1993). Peripheral administration of 4MC, as in the case of most

pharmacologically active or not agents, excludes it from entering the brain, since

the morphological and enzymatic components of the BBB not only prohibit large

molecules such as peptides (e.g. NGF) from entering the brain but also small

molecules that lack adequate lipophilicity. Efficient and sustained delivery of

4MC to the brain could be achieved by means of an appropriate brain-targeted

chemical delivery system (CDS) such as the one based on the dihydropyridine

<--> pyridinium salt redox reaction described previously.

Several CDSs were designed as potential brain selective targetry forms for

4-methylcatechol (Figure 3, A), in which one of the catecholic hydroxyl groups is

esterified with the dihydropyridine targetor while the other is usually masked with

an other lipophilic moiety or esterified with a second targetor.














CH3


(25) CH3


CH3 O0CO CH

oco
N(
(E) CH3


CH3
N
CH3 COO



() CH3


CH3


I
,OCO- C--
I


(H)


OH3


CH3 l OH


N

(19) CH3


CH3, O-CO-R CH3~ O-CO-R CH3 O-CO-R CH3 OH

OCO H3OCO OH H3 4MC


CDS 6H3 CH3 4MC


Figure 3. A) Scheme of 4MC CDSs. B) Metabolism of the CDSs.








Furthermore, a redox analogue (4) of dopamine was designed, in which the

amino group of dopamine is derivatized in the dihydropyridine ring, in order for

the analogue to function as a CDS for the 4-alkyl-catechol moiety (Figure 4, A).

The hydroxyl groups are protected in this case with lipophilic functions while the

targetor is placed on the side alkyl chain of the molecule. The derivative is

designed in a way that in vivo metabolism (in the brain) will liberate the 4-

substituted catechol moiety which is expected not only to have NGF stimulatory

activity due to its catechol part but also to "lock" in its site of action, the brain.

Specific objectives are:

1) Synthesis of the above mentioned CDSs.

2) Development of a suitable analytical system to evaluate the stability of

the CDSs and metabolites in various buffer systems and biological media in order

to determine the feasibility of delivery of 4MC and its derivative by this method.

pH-profiles of representative CDS's and their quaternary metabolites as well as

half-lives in rat blood, brain and liver homogenate are determined.

3) Evaluation of the in vivo distribution of the CDSs and their metabolites

in order to determine whether the CDS is accomplishing its purpose, the selective

and sustained delivery of the compounds of interest to the brain. In all cases, the

dihydro moiety is expected to be oxidized (in vivo) after entering the brain and

subsequent hydrolysis of the esters will liberate 4-MC (Figure 3, B) or the

pyridinium catechol derivative (Figure 4, A). In vivo studies (in rat) of CDS J and

4 are performed to demonstrate oxidation of the CDSs, rapid peripheral








elimination of the pyridinium salt, and sustained central delivery of final

compounds metabolitess) of interest.

4) Determination of the in vivo NGF-stimulatory activity of the 4MC-CDS

J by pharmacological testing in suitable animal models. NGF stimulatory activity

in the rat hippocampus and cerebral cortex, brain regions where NGF is

predominantly expressed, is estimated by measuring the NGF mRNA content in

these regions after treating rats with a single iv injection of the compound; the

dose is based on the brain concentration observed in the distribution studies as

well as the concentrations of 4-methylcatechol that stimulate NGF secretion in

vitro/in vivo.



Synthesis and In Vitro Evaluation of Catechol Isosters and Derivative as NGF-

Inducers



In order to produce novel NGF-stimulating compounds, while also

attempting to elucidate the mechanism of action of the catechol derivatives on

NGF induction, two isosters of 4MC were designed, synthesized and evaluated for

in vitro activity. The functional catechol moiety of 4MC was substituted by two

isosteric/isoelectronic moieties, chemical groups that are considered to be

equivalent and therefore interchangeable as far as their contribution to biological

activity is concerned. Since the only functional moiety that can be responsible for

activity in the molecule of 4MC is the catechol moiety, biologically active











NH2d OH


dopamine


CONH2

N OOCOC(CH3)3

-^OCOC(CH3)3

(4)

Sin vivo
(in the brain)

CONH2

() OH
(5) OH


B.




Figure 4. A) Redox analogue of dopamine and its metabolite. B) 4MC,
isosters and derivative.


CH3 OH


OH
4MC


'OH


2HC


+


.OH


'OH


5MHP








compounds are expected to be derived by substituting it with the aforementioned

functions.

Thus, two compounds are synthesized, 5-methyl-1-hydroxy-2-pyridone

(5MHP) and 2-hydroxymethyl-p-cresol (2HC) (Figure 4, B), and their biological

activity as NGF-stimulators was evaluated in vitro in two cell lines. Furthermore,

in the case that the isosters prove to be active in vitro, appropriate CDSs are

intended to be prepared.

In addition, the pyridinium catechol derivative 5, being the final metabolite

of 4, is separately synthesised and also evaluated for in vitro NGF stimulatory

activity, in order to establish that this novel catechol derivative is indeed active as

other catechols and therefore to establish the usefulness of 4 as a CDS for the

NGF-inducing catechol moiety.



Ouantitative Structure Activity Relationships of Catechol Derivatives on NGF

Secretion in L-M Cells



Some preliminary structure activity relationships of the catechol analogues

suggested that the catechol ring is essential for the stimulatory effect on NGF

synthesis. From the structure-effect of the aliphatic side chain it was deduced that

P3-hydroxylation decreased, N-substitution (non-bulky) enhanced, while a-

carboxylation decreased the stimulation effect on NGF synthesis (Furukawa et al.

1991). Also, shortening of the chain length progressively reduced activity except

in the case of alkyl side chains without any substituents, where the opposite was








observed. The question, therefore, that arises is why the catechol moiety is

responsible for activity and, furthermore, what specific characteristics of that

moiety, as modified by varying substitution on the ring, influence that activity.

In order to study the effects of the side chain on the catechol moiety of the

molecules and to correlate characteristics of the molecular structure with activity, a

set of 23 substituted catechols, with activities indicated in literature data, are

characterized by various electronic, steric, and thermodynamic factors derived

from the semi-empirical AMI method. By correlating activity with the calculated

descriptors, using simple or multiple regression analysis, a quantitative structure-

activity relationship of catechol derivatives on NGF secretion in L-M cells could

be obtained.













CHAPTER 3
EXPERIMENTAL AND RESULTS



Synthesis



All chemicals used were reagent grade obtained from Aldrich Company and

solvents from Fisher Scientific. All melting points were recorded using Fisher-

Johns melting point apparatus and are uncorrected. NMR data were recorded with

Varian T-90, QE-300, or Varian Unity-300 spectrometer and are reported in parts

per million (8) relative to tetramethylsilane. The Elemental analyses were carried

out either at Atlantic Microlab. Inc., Atlanta, Ga, or at the Analytical Services of

the Department of Chemistry of the University of Florida. FAB or Electron

Ionization mass spectrometry was performed with Kratos MFC 500. Ultra Violet

spectra were obtained with a Perkin Elmer UV/VIS spectrometer Lambda II. Thin

Layer Chromatography was carried out using Merck DC-aluminium foil Plates

coated to a thickness of 0.2 mm with silica gel 60 containing Florescent (254)

indicator, or aluminum oxide 60 F neutral (Type E). The High Performance

Liquid Chromatography (HPLC) system consisted of a SP 8810 precision isocratic

pump, SP4290 injector, Waters RCM C-18 column, SP 8450 UV/visible detector

and SP4290 integrator. The mobile phase consisted of acetonitrile-water or

methanol-water in different proportions.








Synthesis of 4-Methylcatechol Chemical Delivery Systems


The synthetic scheme followed for the preparation of the 4-methylcatechol

CDSs is represented in figure 5. Synthesis of the respective intermediates and

final compounds is described below.

Synthesis of 3-hydroxy-4-nicotinoyloxytoluene (A) and 3.4-nicotinoyloxytoluene


4-Methylcatechol (34.7 mmols) was dissolved in 90 ml pyridine and 52.1

mmols of nicotinoyl chloride hydrochloride was added. After heating under reflux

for 48 h, the pyridine was distilled under vacuum at 60 OC, and methylene chloride

added. The mixture was washed with ice-cold water and dried over Na2SO4. The

products (A) and (B) were separated by flash chromatography (silica gel, 2%

methanol in chloroform). Yield: 12.6 mmols (2.88 g) of (A) and 8.9 mmols (2.97

g) of (B).

(A): White solid, m.p.: 176-178 oC.

1H-NMR (DMSO-d6) 5 (ppm) (Figures 6 and 7): 9.61 (bs, 1H, for Ph-OH)

9.21 (t, 1H, for pyrid. H-2, J2,4-6=l 1.2-0.8 Hz) 8.87 (dt, 1H, for pyrid. H-6, J6,5=4.5

Hz, J6,4-2=2.1, 0.8 Hz) 8.43 (dt, 1H, for pyrid. H-4 J4,5=8.1 Hz, J4,6-2=2.1, 1.5 Hz)

7.63 (dd, 1H, for pyrid. H-5, J5,6=4.8 Hz, J5,4=7.5 Hz) 7.02 (d, for Ph-H-5 para

esterified molecule J5,6=8.1 Hz) 6.97 (s, for Ph-H-2' meta esterified isomer)

6.925 (d, for Ph-H-5', J5.,6,=9 Hz) 6.85 (d for Ph-H-6', J6',5'=8.1 Hz) 6.77 (s, for

Ph-H-2) 6.64 (dd for Ph-H-6, J6,5=8.1 Hz, J6,2=1.2 Hz) (Total integration for all

phenylic protons, range 7.02-6.64 ppm, is 3 H. The integration ratio of the









CH3 OH
--OH


- sCOCI

" HCI


CH3 OH
O(A) O
(A) N


CI-CO-CH(CH3)2


CH3 COC
+ CH3Co

(B) N

CH3I


CH3


CH3I


CH3I


CH3


(D) N (G) N 1-
NaHCO3 CH3 NaHCO3 CH3
Na2S204 Na2S204

CH3 OP CO-CH(CH3)2 CH3 .. CO-C(CH3)3
ocof OCO 1I )
(E) N (H) N
CH3 CH3


Figure 5. Synthetic scheme for the preparation of the 4-methylcatechol


CDSs.
























I




Figure 6. 1H NMR spectrum of 3-hydroxy-4-nicotinoyloxytoluene (A).














(A)
CH3 QOH

aa isOCOmer

Para isomer I


w
II

'-I

i~J\ \ I


CH3 OCOf

Mea isOHmer
Meta isomer


'a
'a
'a I I -r ________
n _____________






/ / _
s.- -' ~


Figure 7. Expansion of the aromatic region of the IH-NMR spectrum of 3-
hydroxy-4-nicotinoyloxytoluene (A).


_ I







specific H peaks of the phenyl ring are: H2-H2'= 6:4, H6-H6' = 6:4, and H5-H5'=

6:4.) 2.23 (d, total 3H, for Ph-CH3, ratio of two isomers -ratio of the 2 peaks- 6:4).

Elemental analysis for C13H1103N: Theory: C 68.12, H 4.80, N 6.11.

Found: C 68.02, H 4.82, N 6.03.

(MW=229) El (Electron Ionization) Mass 230[M+H]+,100. HPLC: Single

peak of retention time 1.77 min (60% acetonitrile in water) and 1.76 min (80%

methanol in water).

(B): White solid, m.p.: 91-92 oC

1H-NMR (CDCI3) 8 (ppm): 9.2 (t, 2H, for 2 pyrid. H-2, J2,4-6=1.8 Hz)

8.72 (dt, 2H, for 2 pyrid. H-6, J6,5=5.1 Hz, J6,4-2=1.8, 1.2 Hz) 8.25 (dt, 2H, for 2

pyrid. H-4, J4,5=8.1 Hz, J4,2-6=1.8 Hz) 7.3 (dd, 2H, for 2 pyrid. H-5, J5,6=5.1 Hz,

J5,4=7.8 Hz) 7.29 (s, 1H, for Ph-H-2) 7.2 (d, for Ph-H-5, J5,6=7.5 Hz ) 7.12 (d,

for Ph-H-6, J6,5=8.7 Hz) 2.38 (s, 3H, for Ph-CH3).

Elemental analysis for C19HI404N2: Theory: C 68.26, H 4.19, N 8.38

Found: C 68.35, H 4.19, N 8.34. (MW= 334) El Mass 335 [M+H]+,98, 357

[M+Na]+,100.

Synthesis of 3-isobutyryloxy-4-nicotinoyloxytoluene (C) :

In a suspension of 9 mmols (2 g) of monoester (A) in CHCI3, 2 ml (18

mmols) of isobutyryl chloride in CHCI3 was added dropwise (1:2) and small

amount of pyridine. The mixture was refluxed for 48 hs. After evaporating the

solvents, the product (viscous liquid) was purified by column chromatography

(silica gel, 10% methanol in chloroform). Yield: 60%.








1H-NMR (CDCl3) 8 (ppm): 9.36 (d, 1H, for pyrid. H-2, J2,4=1.7 Hz) 8.85

(dt, 1H, for pyrid. H-6, J6,5=4.8 Hz, J6,4=1.3 Hz) 8.42 (dt, 1H, for pyrid. H-4,

J4,5=8 Hz, J4,2=1.95 Hz) 7.46 (qt, 1H, for pyrid. H-5, J5,4=7.9 Hz, J5,6=4.86 Hz,

J5,2=1 Hz ) 7.19-7.04 (m, 3H, for 3 Ph-H) 2.68 (septet of d, 1H, J=7 Hz, J=2.58

Hz) 2.38 (s, 3H, for 3 Ph-H) 1.15-1.12 (dd, 6H, for i-propyl 2 CH3, J=1.95 Hz,

J=7 Hz).

Elemental analysis for C17H1704N: Theory: C 68.22, H 5.68, N 4.68

Found: C 68.57, H 5.92, N 4.63.

Synthesis of 3-isobutyryloxy-4-(N-methyvlnicotinoyvloxy)toluene iodide (D) :

In a solution of (C) in anhydrous acetone, an excess of methyl iodide was

added. The mixture was heated at 50 OC for 8h. After distilling part of the

solvent, ether was added and the semi solid precipitate was triturated to give a

yellow crystalline mass which was filtered and washed with ether. Yield: 71%.

Bright yellow solid, m.p.: 131-133 oC.

1H-NMR (DMSO-d6) 5 (ppm): 9.7 (s, 1H, for pyridinium H-2) 9.2 (d, IH,

for pyrid. H-6) 9.0 (d, 1H, for pyrid. H-4) 8.2 (t, 1H, for pyrid. H-5) 7.2 (m, 3H,

for 3Ph-H) 4.4 (s, 3H, for N-CH3) 2.8-2.6 (m, 1H, for i-propyl CH) 2.3 (s,3H, for

Ph-CH3) 1.1 (d, 6H, for i-propyl 2 CH3).

Elemental analysis for Cl8H2004NI: Theory: C 48.97, H 4.53, N 3.17, I1

28.79. Found: C 49.02, H 4.61, N 3.13, 128.83. (MW= 441) El Mass 314[M-

I]+,100.








Synthesis of 3-isobutyryloxy-4-(N-methyl- 1,4-dihydronicotinoyloxy)toluene (E) :

Compound (D) (2 mmols) was dissolved in deaerated water, peroxide-free

ether was added and the mixture cooled. 4 mmoles of NaHCO3 was added slowly

followed by 4 mmols Na2S204. The reaction mixture was stirred in an ice bath,

under N2, for 2 hs. The layers were separated, washed and the combined ether

solutions dried and evaporated to give a yellow viscous liquid. It reduces rapidly

methanolic silver nitrate.

1H-NMR (CDCI3) 5 (ppm): 7-6.8 (m, 3H, for Ph-H) 6.7 (dd, IH, for

dihydropyr. H-2) 5.65 (dt, 1H, dihydropyr. H-6) 4.85 (m, 1H, dihydropyr. H-5)

3.15 (bs, 1H, dihydropyr. H-4) 2.95 (s, 3H, for N-CH3) 2.8 (m, 1H, for i-propyl

CH) 2.3 (3H, for Ph-CH3) 1.3 (m, 6H, for i-propyl 2 CH3).

(MW=315) El Mass 338[M+Nal+,100. UV (in MeOH) Xmax 270, 345

nm.

Synthesis of 3-pivaloyloxy-4-nicotinoyloxytoluene (F) :

Compound (A) (2.8 mmols) was dissolved in 15 ml CHCl3 and excess

(0.012 mols) of pivaloyl chloride in 10 ml CHC13 was added dropwise. After the

addition of 5 ml pyridine the reaction mixture was stirred at room temperature for

10 hs. The solvents were evaporated and toluene added. The white undissolved

solid was filtered off and the filtrate was condensed and purified by flash

chromatography (silica gel, 2% methanol in chloroform) to yield a viscous liquid.

Yield: 68.7%.

1H-NMR (CDCI3) 8 (ppm): 9.36 (d, 1H, for pyrid. H-2, J2,4=2) 8.86 (dt,

1H, for pyrid. H-6, J6,5=4.5 Hz, J6,4=2 Hz) 8.42 (dt, 1H, for pyrid. H-4, J4,5=8 Hz,








J4,2-6=2 Hz) 7.47 (dd, 1H, for pyrid. H-5, J5,4=8 Hz, J5,6=4.5 Hz) 7.2-7.0 2 ( m,
3H, for 3 Ph-H) 2.4 (s, 3H, for Ph-CH3) 1.2 (s, 9H, for t-butyl 3 CH3).

(MW=313) El Mass 336[M+Na]+,100, 314[M+H]+,75.

Synthesis of 3-pivalovyloxy-4-(N-methyinicotinoyloxy)toluene iodide (G):

Compound (F) (1.6 mmols) was dissolved in 20 ml anhydrous acetone and

5ml iodomethane was added. The mixture was refluxed for 24 hs. Ether was

added to get a yellow precipitate which was filtered off and recrystallized by

acetone/ether. Yield: 50%.

Bright yellow solid, m.p.:140-142 oC.

IH-NMR (DMSO-d6) 8 (ppm): 9.7 (s, 1H, for pyrid. H-2) 9.3 (d, 1H, for

pyrid. H-6, J6,5=4.8 Hz) 9.14 (d, 1H, for pyrid. H-4, J4,5=6.4 Hz) 8.35 (t, 1H, for

pyrid. H-5, J5,4-6=7.6 Hz) 7.4-7.2 (m, 3H, for 3 Ph-H) 4.5 (s, 3H, for N-CH3) 2.4

(s, 3H, for Ph-CH3) 1.15 (s, 9H, for t-butyl 3 CH3).

Elemental analysis for C19H2204NI: Theory: C 50.10, H 4.83, N 3.07, I

27.91. Found: C 50.21, H 4.86, N 3.04.

(MW=455) El Mass 328[M-I]+,100.

Synthesis of 3-pivaloyloxy-4-(N-methyl-1,.4-dihydronicotinoyloxy)toluene (H) :

Compound (G) (0.13 mmols) was dissolved in 8 ml deaerated water and 5

ml peroxide-free ether was added. NaHCO3 (0.52 mmols) was slowly added while

stirring in an ice bath and then 0.52 mmols of Na2S204. The reaction was stirred

for 1 h under N2 and then the layers were separated, and washed. The ether layers

were dried over Na2SO4, then filtered and evaporated to yield a yellow viscous

liquid. Yield: 30%. The product reduces instantly methanolic silver nitrate.








1H-NMR (CDCI3) 5 (ppm): 7-6.8 (3H, for 3 Ph-H) 6.65 (1H, for dihydro

H-2) 5.65 (d, 1H, for dihydro H-6 ) 4.84 (1H, for dihydro H-5, J5,4=8 Hz) 3.15

(m, 2H, for pyrid. 2 H-4) 2.42 (s, 3H, for N-CH3) 2.35 (s, 3H, for Ph-CH3) 1.4

(s, 9H, for t-butyl 3 CH3).

(MW=329) El Mass 352[M+Na]+,100. UV (MeOH) Xmax 268, 345 nm.

Synthesis of 3.4-(N-methylnicotinoyloxy)toluene iodide (1) :

To 0.65g (1.9 mmols) of nicotinoyl diester of 4-methylcatechol (B)

dissolved in MeOH, an excess of iodomethane was added and the mixture was

heated at 40 OC for 2hs. The solvent was evaporated, acetone was added, and

stirred for I h, to give a yellow precipitate which was filtered off and recrystallized

from methanol/ether to give 0.74g ( 61.6 % yield) of the diquatemary salt.

Yellow solid, m.p.: 204-206 OC.

1H-NMR (DMSO-d6) 8 (ppm): 9.7 (2H) 9.2 (d ,2H) 9.1 (d, 2H) 8.4 (m,

2H) 7.5 (m, 3H) 4.5 (s, 6H) 2.5 (s, 3H).

Elemental analysis for C21H2004N212: Theory: C 40.77, H 3.23, N 4.53, I

41.10. Found: C 40.83, H 3.25, N 4.56,1 41.13. (MW=618)

Synthesis of 3.4-(N-methyl-l,4-dihydronicotinoyloxy)toluene () :

Compound (I) (1.2 mmols) was dissolved in deaerated water, ethyl acetate

was added and the mixture cooled in an ice-bath. NaHCO3 (9.6 mmols) was

added very slowly followed by 9.6 mmols of Na2S204. The whole procedure was

conducted under N2. The mixture was stirred for I h 45 min. The two layers were

separated, washed and the combined ethyl acetate extracts were dried over








Na2SO4. After filtration and evaporation it yielded a yellow liquid that reduces

instantly methanolic silver nitrate.

1H-NMR (CDC13) 5 (ppm): 6.95-6.67 (m, 5H, for 3 Ph-H and 2

dihyropyridine H-2) 5.7 (d, 2H, for 2 dihydropyridine H-6) 4.9 (m, 2H, for 2

dihydropyridine H-5) 3.2 (bs, 4H, for 2 dihydropyridine H-4) 3.02 (s, 3H, for Ph-

CH3) 2.3 (d, 6H, for 2 N-CH3). (MW=366) El Mass 389[M+Na]+,100. UV

(MeOH) Xmax 280 and 353 nm (Figure 8).



Synthesis of the Redox Analog of Dopamine



The synthetic scheme for this preparation is depicted in Figure 9 and

respective synthesis of intermediates and final product described below.

Synthesis of 3-carbamoyl-l-(2.4-dinitro)phenylpyridinium chloride ('Zincke-type

reagent') (1) :

Nicotinamide (0.065 mols) and 0.098 mols of 2,4-dinitro-chloro-benzene

were mixed and heated in a 90 oC oil bath for 75 min. The melt was dissolved in

MeOH and ether was added to precipitate a yellow solid. Using the same system

the product was re-precipitated 3 times and then dissolved in water and treated

with charcoal. The product, 3-carbamoyl-1-(2,4-dinitro)phenylpyridinium

chloride (or 'Zincke reagent') (1), was freeze-dried to a pale foamy solid. Yield:

33%. m.p.: 141-142 oC.

1H-NMR (DMSO-d6) 5 (ppm): 9.8 (bs, 1H) 9.4 (d, 1H) 9.1 (d, 1H) 8.9-

8.7 (m, 3H) 8.4-8.2 (m, 2H) 8.1 (bs, 1H).















X: USER892;
inf:


400.8 258. nm; pts 376; int 8.40; ord 8.1054 a.8157 a
18:47:22 34 i22o


1.0 1


A.ft

0.4


8.2 J-


/
/


'1


328
Ti


368


Figure 8. Ultra Violet Spectrum of Dihydro compound J. (Typical UV
spectrum obtained by all dihydro compounds synthesized in this study)








Synthesis of di-O-pivaloyl dopamine (trifluoroacetate) (2) :

In a suspension of 4g of dopamine hydrochloride in 20 ml trifluoro acetic

acid (TFA), 10 ml pivaloyl chloride was added dropwise. After stirring for 45 hs

the TFA was distilled and the product was isolated by column chromatography

(silica gel, 10 % methanol in chloroform). Yield: 92%.

Pale white solid, m.p.: 108-109 OC.

1H-NMR (CDCI3) 5 (ppm): 7.7 (br, 2H) 7 (3H) 3 (m, 4H) 1.3(s 18H).

Elemental analysis for C20H2806NF3 : Theory: C 55.17, H 6.4, N 3.2.

Found: C 54.85, H 6.5, N 3.12.

Synthesis of 1-(3',4'-dipivaloylphenethyl)-3-carbamoylpyridinium trifluoroacetate

M :

To a solution of 4.62 mmols of (1) and 9.25 mmols of (2) in 10 ml

anhydrous MeOH, 1.28 ml tetraethylamine (TEA) (in 6 ml MeOH) was added

dropwise. The mixture was stirred for 30 min and refluxed for 30 min. Upon

cooling, a yellow solid precipitated (by-product) and was removed. The filtrate

was evaporated and dissolved in a small amount of MeOH and ether was added to

precipitate a pale solid, which was recrystallized several times by methanol/ether.

Yield: 52%.

Pale coloured solid, m.p.: 194-198 oC.

1H-NMR (DMSO-d6) 8 (ppm): 9.61 (s, IH) 9.15 (d, 1H) 8.96 (d, 1H)

8.63 (s, 1H, for 1H of CONH2) 8.25 (t, 1H) 8.16 (s. 1H, for 1H of CONH2) 7.2

(m, 3H) 4.9 (t. 2H) 3.15 (m, for H20 and CH2-Ph) 1.3 (d ,18H). (MW=540) El

Mass 427 [M-CF3COOl]+,100.











CONH2
LIN


C1
CSNO2


NO2


.DCONH2

Cl + N
NO2

NO2


HCI. NH2OH

OH


CF3COOH. NH2 OCOC(CH3)3
+ CIOCOC(CH3)3 Ll
(CF3COOH) OCOC(CH3)3
(2)


NH2
NO2

NO2


TEA
(MeOH)

CONH,
C0+ CF3COO
0N OCOC(CH3)3

I ,O C O C ( C H 3 ) 3


NaHCO3
Na2S204
CONH2

C N Ns%.. OCOC(CH3)3

OCOC(CH3)3

(4)


Figure 9. Synthetic scheme for the preparation of the redox analog of
dopamine (4).








Elemental analysis for C26H3107N2F3 : Theory: C 57.70, H 5.74, N 5.18.

Found: C 57.70, H 5.73, N 5.21.

Synthesis of 1-(3'.4'-dipivaloylphenethyl)-3-carbamoyl-1.4 dihydropyridine (4) :

Compound (3) (1.076 mmols) was dissolved in deaerated water, peroxide-

free ether was added and the mixture cooled. NaHCO3 (5 mmols) was added

slowly followed by 5 mmols of Na2S204 and the reaction mixture was stirred in an

ice-bath, under N2, for 3 hs. The two layers were then separated, washed, and the

combined ether solutions dried and evaporated to give the product. Yield: 45.4%.

The product reduces instantly methanolic silver nitrate.

Yellow solid, m.p.: 62-66 oC.

IH-NMR (CDCI3) 6 (ppm): 7.06 (s, 2H) 6.96 (s, 1H) 6.91(s, IH) for ArH

and dihydropyridine H-2, 5.65 (d, 1 H) for dihydropyridine H-6, 5.25 (bs, 2H) for

CONH2, 4.71 (m, 1H) for dihydropyridine H-5, 3.32 (t,2H) 3.13 (s, 2H) for

dihydropyridine 2 H-4, 2.81 (t, 2H) 1.35 (d, 18H). (MW=428) El Mass

451[M+Naj+,100.

UV (in MeOH) Imax 270, 345 nm.

Elemental analysis for C24H3205N2: Theory: C 67.29, H7.48, N 6.54.

Found: C 67.58, H 7.65, N 6.28.



Synthesis of the Pyridinium Catechol Derivative and a Dimethoxy CDS



Synthesis of these products was conducted according to the reactions shown in

Figure 10. Synthesis of the final products and intermediate is described below.










































B.




Figure 10. A) Synthesis of the pyridinium catechol derivative (5). B)
Synthesis of the methoxy CDS (7).








Synthesis of 1-(3'.4'-dihydroxyphenethyl)-3-carbamoylpyridinium chloride (5) :

To a solution of 4.6 mmols of dopamine hydrochloride, in 10 ml methanol,

4.6 mmols of TEA was added followed by 2.3 mmols of the Zincke type reagent

(1) in 3 ml methanol. The mixture was refluxed for I h. Upon cooling the yellow

precipitate was removed and to the filtrate ether was added to precipitate a yellow

solid. m.p.: 234-235 OC

1H-NMR (DMSO-d6) 5 (ppm): 9.3 (s, 1H) 9.05-8.8 (m, 5H) 8.2 (m, 2H)

6.6 (m, 2H) 6.4 (d, 1 H) 4.8 (t, 2H for N-CH2) 3.1 (t, 2H for Ar-CH2).

(MW=294) El Mass 259 IM-Cl-]+,100.

Elemental analysis for Cl4HI503N2CI : Theory: C 57.04, H 5.09, N 9.51

Cl 12.05. Found: C 55.40, H 5.18, N 9.19 Cl 11.59.

Synthesis of 1-(3'.4'-dimethoxyphenethyl)-3-carbamoylpyridinium chloride (6):

To a solution of 6.15 mmols of dimethoxy-phenethylamine in 8 ml dry

methanol, 3.07 mmols of the Zincke type reagent (1) were added and the mixture

stirred for 2 hs at room temperature. After cooling on ice, the yellow precipitate

(by-product) was filtered off. Ether was then added to the solution to give an

orange precipitate which was recrystallized from methanol/ether. Yield: 77%.

Orange coloured solid, m.p.: 235-239 oC

IH-NMR (DMSO-d6) 6 (ppm): 9.64 (s, 1H, for pyridinium H-2) 9.07 (d,

1H, for pyrid. H-6, J6,5=6.3 Hz) 8.97 (d, 1H, pyrid. H-4, J4,5=8.1 Hz) 8.82 (s, 1H,

for 1 of CONH2) 8.19 (dd, 1H, for pyrid. H-5, J5,6=6.3 Hz, J5,4=8.1 Hz) 8.16 (s,

1H, for 1 of CONH2) 6.92 (d, 1H, for phenylic H-2, J2,6=1.8 Hz) 6.82 (d, 1H, for








phenylic H-5, J5,6=8.1 Hz) 6.67 (dd, 1H, for phenylic H-6, J6,5=8.1 Hz, J6,2= 1.8

Hz) 4.84 (t, 2H, for N-CH2) 3.7 (d, 6H for 2 CH30) 3.21 (t, 2H, for Ph-CH2).

(MW=322) El Mass 287[M-Cl-]+,100.

Elemental analysis for Cl6HI903N2CI: Theory: C 59.53 H5.89, N 8.68,

Cl 11.00. Found: C 59.46, H 5.90, N 8.61, Cl 10.92.

Synthesis of 1-(3'.4'-dimethoxyphenethyl)-3-carbamoyl-1,4 dihydropyridine (7):

To a solution of 0.309 mmols of (6) in deaerated water, methylene chloride

was added and the mixture cooled. 5 mmols of NaHCO3 were added slowly

followed by 5 mmols of Na2S204 and the reaction mixture was stirred in cool

under N2, for 3 hs. The two layers were then separated, washed and the combined

organic layers dried over Na2SO4. After evaporation, a yellow semi solid was

obtained. Purity checked by HPLC: single peak using two different systems (1.48

min retention time with 60% acetonitrile-water, flow rate 2ml/min, or 1.6 min with

80% methanol in water)

IH-NMR (CDCl3) 5 (ppm): 7.02 (s, 1H) 6.8 (d, 1H) 6.7 (m, 2H) 5.65 (d,

1H) 5.48 (s, 2H) 4.7 (m, 1H) 3.86 (d, 6H) 3.3 (t, 2H) 3.14 (s, 2H) 2.76 (t, 2H).

(MW=288) UV (in MeOH) Xmax 280, 355 nm.



Synthesis of 2-Hydroxymethyl-p-Cresol



Synthesis of 2-hydroxymethyl-p-cresol(2HC) was performed according to

Figure 11 (A) as follows:








A solution of p-cresol (32 mmols), benzene boronic acid (49 mmols),

propionic acid (16 mmols), and paraformaldehyde (0.23 mols), in dry benzene

(130 ml), was heated under reflux, with azeotropic removal of water using a Dean-

Stark type separator, for 8 hs. The large excess of paraformaldehyde was added

portionwise in intervals of 1.5-2 hs. Evaporation of the solvent, extraction with

dichloromethane, washing with aqueous Na2CO3 and water and evaporation of the

solvent after drying gave a residue which recrystallized from petroleum ether to

the 2-phenyl-4H-1,3,2-dioxaborin derivative 8, Figure 11, A. In the next step, a

mixture of derivative 8 (4.5 mmols), 30% hydrogen peroxide solution (5ml) and

tetrahydrofuran (5mi) was stirred at 0-25 oC for 2.5 hs. The reaction mixture was

poured into ice-water, extracted with ether several times, and the extracts were

washed with sodium hydrogen sulphite solution. After drying and removing the

solvent, the residue was recrystallized from ether/petroleum ether to pure 2HC.

White solid, m.p.: 104-105 oC.

IH-NMR (DMSO-d6) 8 (ppm): 7.05 (b, 1H, for Ph-O-H) 7.01 (dd, 1H, for

C5-H, J6,5 = 8.4 Hz, J3,5 = 1.5 Hz) 6.86 (ds, 1 H, for C3-H, 13,5 = 1.5 Hz) 6.8 (d,

1H, for C6-H, J6.5 = 8.4 Hz) 4.83 (s, 2H, for CH2) 2.25, (s, 3H, for CH3) 1.65 (s,

1H, for CH2-OH).

Elemental analysis for C8H1002: Theory: C, 69.56, H, 7.24. Found: C,

69.43, H, 7.28.













OH

<-OH


OH

C H ,a


+ (CH20)n ,-
C2HsCOOH


THF
H202 CH3


CH3


N OH
OH


CH3

N r


HCI, NaNO2 CH3 n M

NH2 N CI


CH3

nN CI
t


CPBA or
H202


NaOH CH3
IBM--


OH


(10)


5MHP


Figure 11. Synthetic scheme for the preparation of A) 2-hydroxymethyl p-
cresol (2HC) and B) 5-methyl-1-hydroxy-2-pyridone (5MHP).


r








Synthesis of 5-Methyl-I -Hydroxy-2-Pyridone



Synthesis of 5-methyl-l-hydroxy-2-pyridone (5MHP) was performed

according to Figure 11 (B) as follows:

5-Methyl-2-amino pyridine was converted to the 2-chloro derivative (9),

figure 11 B, according to the procedure described in Lister et al., 1968, in 60 %

yield. N-Oxidation of the chloro derivative using 3-chloroperbenzoic acid, gave

the N-oxide (10) which was separated from the accompanying 3-chlorobenzoic

acid by vacuum liquid chromatography, using silica gel as an adsorbent. The N-

oxide was obtained solely and in 70 % yield when the chloro derivative was heated

with 30 % H202 in acetic acid at 76 oC for 40 hs and the product was crystallized

from ethyl acetate (m.p. 70-72 OC). The final product, 5-methyl-l-hydroxy-2-

pyridone was obtained directly (70 % yield) from the N-oxide on alkaline

hydrolysis by boiling with IN NaOH aqueous solution for 24 hs. The product

precipitated from chloroform solution by addition of hexane in crystalline form

and showed one spot on TLC plates of silica gel using several solvent systems.

White solid, m.p.: 155-156 oC.

1H-NMR (DMSO-d6) 8 (ppm): 12 (s, 1H, for N-OH), 7.5 (s, 1H) 7.2 (dd,

1H) 6.5 (d, 1H) for three aromatic protons, 2.1 (s, 3H, for C-CH3).

Elemental analysis for C6H7NO2: Theory: C, 57.6, H, 5.6, N, 11.2. Found

C, 57.5, H, 5.65, N, 11.11.







Synthesis of 3- ((N-Methyl- 1.4-Dihydronicotinoyloxy)Methyl }-4-(N-Methyl- 1.4-

Dihydronicotinoyloxy)Toluene



Synthetic scheme for the preparation of 3-{(N-methyl-1,4-dihydro-

nicotinoyloxy)methyl) -4-(N-methyl-1,4-dihydronicotinoyloxy)toluene (13), is

shown in Figure 12.

Synthesis of 3-nicotinoyloxymethyl 4-nicotinoyloxytoluene (11) :

A mixture of 1.45 mmols 2HC and 4.35 mmols of nicotinoyl chloride

hydrochloride in 10 ml pyridine were refluxed for 3 hs. After evaporating the

pyridine, chloroform was added and the mixture washed twice with water and

dried over MgSO4. The product was purified by flash chromatography (silica gel,

1% methanol in methylene chloride).

Pale white solid, m.p.: 85-86 OC.

1H-NMR (CDCI3) 5 (ppm): 9.27 (d, 1H, for pyrid. H-2, J2,4=1.66 Hz) 9.04

((d, 1 H, for pyrid'.-connected to the 3 position of toluene- H-2', J2',4'= 1.66 Hz)

8.76 (dd, 1H, for pyrid. H-6, J6,5=4.58 Hz, J6,4=1.66 Hz) 8.67 (dd, 1H, for pyrid'.

H-6', J6,5,-=4.58 Hz, J6',4'=1.66 Hz) 8.32 (dt, 1H, for pyrid. H-4, J4,5=8.33 Hz, J4,2-

6=1.66 Hz), 8.11 (dt, 1H, for pyrid'. H-4', J4,,5,=8.33 Hz, J4',2'-6'=1.66 Hz) 7.36 (dd,

1H pyrid. H-5, J5,6=4.58 Hz, J5,4=8.33 Hz), 7.26 (dd, 1H,for pyrid'. H-5',

J5',6'=4.58 Hz, J5',4'=8.33 Hz) 7.3 (d, 1H, for Ph-H-2, J2,6=1.6 Hz) 7.2 (dd, 1H for

Ar-H-5, J5,6=8.5 Hz, J5,2=1 Hz) 7.08 (d, 1H, for Ph-H-6, J6,5=8.5 Hz ) 5.32 (s, 2H,

for Ph-CH2-O) 2.35 (s, 3H, for Ph-CH3).

(MW=348) El Mass 371 [M+Na]+, 100.













CH3 OH


2HC


CH3'


COCI


. HC1


0


;0CH3I

O-CO J CH3
N"


(12)

NaHCO3
Na2S204
jr


CH3.


(13)


O-CO-

1+ N
OH3
CH3



0H

NCO
CH,


Figure 12. Synthetic scheme for the preparation of 3-{(N-methyl-1,4-
dihydronicotinoyloxy)methyl)} -4-(N-methyl- 1,4-dihydronicotinoyloxy)toluene
(13).


(11)








Elemental analysis for C20H16N204: Theory: C 68.97 H 4.6 N 8.05.

Found: C 69.06 H 4.66 N 7.96.

Synthesis of 3- {(N-methylnicotinoyloxy)methyl-4-(N-methylnicotinoyloxy)

toluene iodide (12):

Excess of methyl iodide (3ml) was added to 0.55 mmols of (11) dissolved

in 20 ml dry acetone. The mixture was refluxed overnight after which the yellow

precipitate was filtered and washed with cold acetone. Yield 90%. (MW=632)

Yellow solid, m.p.: 160-163 OC.

1H-NMR (DMSO-d6) 5 (ppm): 9.8 (bs, 1H, for pyrid. H-2) 9.47 (bs, 1H.

for pyrid'.-connected to the 3 position of toluene- H-2') 9.3 (d, 1 H, for pyrid. H-6,

J6,5=5.83 Hz) 9.2 (d, 1H, for pyrid'. H-6', J6',5-=5.83 Hz) 9.15 (d, 1H, pyrid. H-4,

J4,5=8.75 Hz) 8.9 (d, 1 H, pyrid'. H-4', J4',5-=8.75 Hz) 8.35 (dd, 1 H, for pyrid.H-5,

J5,6=5.83 Hz, J5,4=8.75 Hz) 8.25 (dd, 1 H, for pyrid. H-5', J5',6'=5.83 Hz, J5',4'=8.75

Hz) 7.56-7.33 (m, 3H, for Ph-H ) 5.53 (s, 2H, for Ph-CH2-O) 4.5 (s, 3H for N-

CH3) 4.44 (s, 3H for N'-CH3) 2.4 (s, 3H, for Ph-CH3).

Elemental analysis for C22H22N20412: Theory: C 41.77 H 3.48 N 4.43 I

40.19. Found: C 41.84 H 3.5 N 4.47 140.26.

Synthesis of 3- ((N-methyl- 1,.4-dihydronicotinoyloxy)methyl 1-4-(N-methyl- 1.4-

dihydronicotinoyloxy)toluene (13) :

Compound (12) (0.16 mmols) was dissolved in deaerated water, methylene

chloride was added and the mixture cooled in an ice-bath. NaHCO3 (1.26 mmols)

was added very slowly followed by 1.26 mmols of Na2S204. The whole

procedure was conducted under N2. The mixture was stirred for I h 45 min. The








two layers were separated, washed and the combined ethyl acetate extracts were

dried over Na2SO4. After filtration and evaporation it yielded a yellow liquid that

reduces instantly methanolic silver nitrate. (MW=380)

UV (CH2C12) Amax 360 nm.



Synthesis of 3-Hydroxy-4-(N-Methyl- 1,.4-Dihydronicotinoyloxy)Toluene



Attempted synthesis of 3-hydroxy-4-(N-methyl-1,4-dihydronicotinoyloxy)

toluene (19) by various routes is shown in Figure 13.

Synthesis of nicotinic anhydride (14) :

Sodium salt of nicotinic acid (0.033 mols) (prepared by adding 0.05 mols of

nicotinic acid in 50 ml aqueous solution of 0.5 mols NaOH and distilled under

reduced pressure) was suspended in 200 ml warm xylene and 0.016 mols of

nicotinoyl chloride hydrochloride was added and the mixture refluxed for 4h. The

precipitate was filtered off and the solvent from the filtrate was evaporated. The

solid product was recrystallised from benzene and stored in dessicator with P205.

Yield: 50%.

White solid, m.p.: 122-126 oC

1H-NMR (DMSO-d6) d (ppm): 9.1 (d, 2H, for 2 pyrid. H-2, J2,4=2Hz)

8.82 (dd, 2H, for 2 pyrid. H-6, J6,5=6 Hz, J6,4=2 Hz) 8.3 (dt, 2H,for 2 pyrid. H-4,

J4,5=8 Hz, J4,2-6=2 Hz) 7.57 (dd, 2H, for 2 pyrid. H-5, J5,6=4.4 Hz, J5,4=8 Hz).











COOH
I + NaOH
N


0 0
II II
,.-C-O-C COClI
(14) N N
HCI
CH31
'


0 0 L
II II
C-O-C.


CH3 (15) CH3


CH3 Q OH
I
(AOCO
(A) N^


N (CH3)2

N


(BOC)20
------k


COONa


(18)

HCI


CH3 O OH
NaHCO3 OCO
Na2S204 n i
H N
--' (19)



CH3


I 0

Ht CH3


CH3I 3
0
O
CH (16OOC(CH3)3

(16) N


(17)


Figure 13. Synthetic schemes for the preparation of 3-hydroxy-4-(N-
methyl-1,4-dihydronicotinoyloxy)toluene (19).








Synthesis of trigonelline anhydride (15) :

To 7 mmols of (14) in 75 ml acetonitrile, 20 mmols of methyl iodide

(excess) were added and refluxed overnight. After cooling the precipitated solid

was filtered and stored in dessicator with P205. Yield: 80%

Yellow solid, m.p.: 218-220 oC (MW=512)

1H-NMR (DMSO-d6) 5 (ppm): 9.51 (s, 2H, for 2 pyrid. H-2) 9.2 (d, 2H,

for 2 pyrid. H-6, J6,5=4.8 Hz) 8.95 (d, 2H, for 2 pyrid. H-4, J4,5=8.4 Hz) 8.26 (dd,

2H, for 2 pyrid. H-5, J5,6=6 Hz, J5,4=8.4 Hz) 4.45 (s, 6H, for 2 -N-CH3).

Synthesis of 3-BOC-4-nicotinoyloxytoluene (16):

To a solution of 0.43 mmols of (A) in chloroform, 1 g (excess) of di-tert-

butyl-pyrocarbonate ( BOC anhydride) were added and the mixture stirred at room

temperature overnight. The product, white viscous liquid, was isolated by flash

chromatography (silica gel, 2% methanol in methylene chloride).

Yield: 70%. (MW=329)

1H-NMR d (CDC13) 6 (ppm): 9.0 (bs, 1 H, for pyrid. H-2), 8.8 (bd, 1 H, for

pyrid. H-6), .37 (dt, 1H, for pyrid. H-4), 7.39 (dd, 1H, for pyrid. H-5), 7.2-7.0

(m, 3H, for Ar-H), 2.3 (s, 3H, Ar-CH3), 1.34 (s, 9H, for t-But)

Synthesis of 3-BOC-4-(N-methylnicotinoyloxy)toluene iodide (17) :

Excess of iodomethane (1 ml) was added to a solution of (14) (0.2 mmols),

in 10 ml dry acetone. The reaction was refluxed for 2 days. The product was

precipitated by addition of ether to the reaction mixture. Yield: 80%. (MW=471)

Yellow solid, m.p.: 130-133 OC








1H-NMR (DMSO-d6) 8 (ppm): 9.78 (bs, 1H, for pyrid. H-I) 9.29 ((bd,

1H, for pyrid. H-6) 9.11 (bd, 1H, for pyrid. H-4) 8.3-8.4 (bt, 1H, for pyrid. H-5)

7.4-7.2 (m, 3H, for Ar-H) 4.5 (s, 3H, for N-CH3) 2.38 (s, 3H, for Ar-CH3) 1.35

(s, 9H, for t-But)

Elemental analysis for C19H22NO5I: Theory: C 48.40 H 4.67 N 2.97 I

26.96. Found C 48.29 H 4.71 N 2.91 1 27.08.

Synthesis of 3-Hydroxy-4-(N-Methylnicotinoyloxy)Toluene Iodide (18):

0.08 mmols of (15) were finely ground and dissolved in 1 ml of 4M HCI in

Dioxane. After stirring for 2 hs, the solvents were evaporated in vacuum and the

viscous liquid obtained was recrystallised by warm acetone/ether. Yield: 32%

Pale yellow solid, m.p.: 197-199 OC

1H-NMR (DMSO-d6) 8 (ppm): 9.98, 9.83 (d, ratio of 2 peaks 7:3, total 1H,

for -OH) 9.78 (bs, 1H, for pyrid. H-I) 9.3 ((bd, 1H, for pyrid. H-6) 9.14 (dt, 1H,

for pyrid. H-4) 8.32 (bt, 1H, for pyrid. H-5) 7.1-6.68 (m, 3H, for Ar-H) 4.5 (s,

3H, for N-CH3) 2.27, 2.25 (d, ratio of 2 peaks 7:3, total 3H, for Ar-CH3).

(MW=371) El Mass 244 [M-I-I+,100.



Synthesis of 3-hydroxy-4- { (N-methyl- 1,.4-dihydronicotinoyloxy)methoxy) toluene



Attempted synthesis of 3-hydroxy-4-{(N-methyl-1,4-dihydronicotinoyloxy)

methoxy)toluene (25) is represented in Figure 14.

Synthesis of chloromethyl chlorosulfate (20) :

Chlorosulfonic acid (200 ml, 3 M) and bromochloromethane (100ml, 1.5M) were


____













+ BrCH2C ---


CI-CH2SO3Cl
(20)


NaHCO3
+ Cl-CH2SO3Cl uNaHCO3
BU4NHSO4
(20)


COOCH2Cl

NJ
(21)


CH3 :, OH
0"- OH


+ MeONa


COOCH2Cl
+ N ACN


(21)


MeOH
----------AND--


CH3 c:P OH
I ONa


(22)


CH3 OH
VONa

(22)



CH3


(25)


CH3 OH O

(23OCH20

(23) NJ


CH3


CH3


(24)


Figure 14. Synthetic scheme for the preparation of 3-hydroxy-4-((N-
methyl-1,4-dihydronicotinoyloxy)methoxy } toluene (25).


C1SO3H


1N7;
N


+ 1
CH,3


--00








boiled for 3 hs. The mixture was poured on ice and extracted with 100ml

methylene chloride, which was washed with water (100ml), dried and evaporated

under vacuum. The product was distilled at 45-50 oC (9-10 mmHg). Yield: 30%.

Synthesis of chloromethyl nicotinate (21) :

To 0.04 mols of nicotinic acid in 40 ml methylene chloride and 40 ml

water, 0.15 mols of NaHCO3, 0.002 mols of Bu4NHSO4, and 0.046 mols of (20) in

10 ml methylene chloride were added. The mixture is stirred for I h and the layers

separated. The organic layer was washed several times with water, and the

product was isolated by column chromatography (neutral A1203, methylene

chloride). The product was kept in solution.

IH-NMR (CDC13) 5 (ppm): 9.19 (bt, 1H, for H-2, J2,4-6=2 Hz) 8.94 (dd,

1 H, for H-6, J6,5=4.5 Hz, J6,2=2 Hz) 8.4 (dt, 1 H, for H-4, J4,5=8.5 Hz, J4,2-6=2 Hz)

7.18 (dd, 1H, H-5, J5,4=8.5 Hz, J5,6=4.5 Hz) 6.2 (s, 2H, for -O-CH2-CI)

Synthesis of 4-methyvl-2-hydroxy sodium phenoxide (22) :

To 5.82 mmols of 4-methylcatechol in 6 ml dry and deaerated methanol,

5.82 mmols of MeONa (4.37 M solution in methanol) were added and the solvent

was immediately distilled under vacuum. (The product which is unstable and

oxidizes in air -dark color appears rapidly- was kept under N2.)

Attempted synthesis of 3-hydroxy-4- f (nicotinoyloxy)methoxy I toluene (23):

Equimolar amounts of (21) and (22) were mixed in acetonitrile. The

mixture was refluxed under N2 for many days. The dark mass obtained did not

contain the desired product (23). 18-crown-ether was also added during the

reaction but did not alter the result.







In Vitro Stability Studies



Analytical Method



A high performance liquid chromatographic (HPLC) method was

developed to assay the CDSs and their metabolites. The HPLC system consisted

of an Autochrom M500 pump, Rheodyne injector, Waters RCM C- 18 column,

Spectroflow 757 absorbance detector and a Fisher Recordall series 5000. The

mobile phase consisted of aqueous phosphate buffer solution (0.037 M, pH 6.5)

and acetonitrile in different proportions. HPLC peak heights were used as a

measure of the concentration of the compounds (assay detection limit 1 .g/ml) and

were plotted against time to evaluate the disappearance rates of the compounds.

The stabilities were determined by measuring the pseudo-first order rate constant

(kobs, min-1) or the half-life (tl/2, min) of disappearance of the compound in the

solution. kobs is determined from the slope of the log of the disappearance curve

(kobs = slope x 2.303) and the t1/2 of the compounds is calculated from the relation

t/2 = 0.693 / kobs.



Stability in Buffers



USP standard phosphate buffer solutions (0.2M) (USP XXI, 1985) in the

pH range of 4.5 to 9.5 were used in this study. Solutions of D, E, 4 and 3 in

buffers were made at a concentration of 1 mg/ml. The solutions were kept at 37oC







and aliquots were taken at frequent time intervals and injected in the HPLC. The

study was repeated 3 times, and the half live at each pH was calculated from the

average of the values obtained. pH profiles of D and E are shown in Figure 15,

while the pH profiles of 4 and 3 are represented in Figure 16.



Stability in Biological Media



A 0.3 ml aliquot of stock solution (3 mg/ml DMSO) of the CDS compounds

or their quaternary metabolites were added respectively to 3 ml of biological

medium (whole heparinized rat blood, 20% rat brain homogenate, or 20% rat liver

homogenate, in isotonic phosphate buffer of pH 7.4), which was kept in a 37 OC

water bath, to obtain a final concentration of 0.33 mg/ml biological media.

Samples of 0.1 ml were taken at appropriate time intervals ( 0, 0.25, 0.5, 0.75, 1, 3,

5, 10, 15, 20, 30, 40, 60, 80, 100, and 120 min) and were mixed with 0.2 ml

acetonitrile containing 5% DMSO. The mixtures were centrifuged and the

supernatants injected in the HPLC. The experiment was repeated once more and

the average half lives calculated. The stabilities (half-lives) of the 4-methyl-

catechol chemical delivery systems (E, H, J) and their quaternary metabolites (D,

G, I) in rat blood and brain homogenate are shown respectively in Figures 17 and

18, while the stability of 4 and 3 in rat blood, liver and brain homogenate is shown

in Figure 19.





68










80

70 E

60 \ "----- D



o --- i ---- i --- i- .^---E
5. 50
E
40
S30

20

10

0 I I I I I
4 5 6 7 8 9
pH











Figure 15. pH profile of 4-methyl catechol CDS (E) and its quaternary
metabolite (D). Each value is the mean of three independent determinations.














120 -

100 -

80 -

60

40

20


4 4.5 5
44.5 5


U









U


- I I I I 7 7.5 8
5.5 6 6.5 7 7.5 8


350.
300 -

,- 250-

- 200 -

150 -
100 -
50 -

0
e


II *


6.5 7


7.5 8


9 9.5


pH

B.





Figure 16. pH profiles of CDS 4 (A) and of its quaternary metabolite 3
(B). Each value is the mean of three independent determinations.


..














250

200

150

100


231


49.5


53.3


50 -
1 9.8 2.1
01 ^4I^Mi


G H

compound


Figure 17. In vitro stability in rat brain 20% homogenate of CDSs (E, H,
and J) and their quaternary metabolites (D, G, and I). Each value is the mean of
two independent determinations. r is the regression coefficient, k is the rate of
disappearance, and tj/2 is the half-life in minutes.


D E G H I J
r 0.997 0.999 0.995 0.986 0.979 0.98
k 0.666 0.071 0.326 0.014 0.003 0.013
t 1/2 (min) 1 9.8 2.1 49.5 231 53.3














57.7


60 ,-


49.5


S40-
E
-- 30


0 4


18.7


I I I I


compound


Figure 18. In vitro stability in rat blood of CDSs (E, H, and J) and their
quaternary metabolites (D, G, and I). Each value is the mean of two independent
determinations. r is the regression coefficient, k is the rate of disappearance, and
ti/2 is the half-life in minutes.


D E G H I J
r 0.99 0.984 0.95 0.972 0.991 0.991

k 0.037 2.23 0.642 0.633 0.012 0.014
t 1/2 (min) 18.7 0.3 1.1 1.1 57.7 49.5


20

10

























12
0.2 0.7


Rat
blood


Rat
liver
(4)


Rat
brain


13 21.6


Rat
blood


Rat
liver
(3)


Figure 19. In vitro stability of (4) and (3) in various biological media. The
first 3 columns refer to the dihydro CDS (4), while the last 3 columns refer to the
quaternary (3). Each value is the mean of two independent determinations. r is the
regression coefficient, k is the rate of disappearance, and t/2 is the half life in
minutes.


Rat
brain


Blood (4) Liver (4) Brain (4) Blood (3) Liver (3) Brain (3)
r 0.986 0.99 0.991 0.995 0.989 0.612
k 3.236 1.0187 0.0574 0.053 0.032 0.003
tl/2 (min) 0.2 0.7 12 13 21.6 229







In Vivo Distribution Study



Analytical Method



The HPLC system consisted of a SP 8810 precision isocratic pump, SP4290

injector, Waters RCM C-18 column, SP 8450 UV/visible detector and SP4290

integrator. The mobile phase consisted of aqueous phosphate buffer solution (0.05

M, pH 6.5) and acetonitrile in different proportions for the detection of the dihydro

compounds, quaternary metabolites and final metabolites. The area under the peak

was used as a measure of the concentration of the compound. Detection was made

at 345 nm for the dihydro compounds and at 254 nm for metabolites.



Experimental procedure



Male Sprague Dawley rats (body weight 200-220 g) were injected

intravenously (tail vein) with compounds J or 4 (50mg/Kg body weight) in 1:1

DMSO and 50% 2-hydroxypropyl-p-cyclodextrin in water. Animals were

sacrificed at the appropriate time points and trunk blood, brain (and liver)

collected. Samples were mixed or homogenised with 2 volumes of acetonitrile

containing 5% DMSO, centrifuged, and supernatants analysed by HPLC.

Calibration curves were used to determine the concentrations of the CDSs

and their metabolites in blood, brain, or liver, respectively. Results of the in vivo

distribution study are represented in Figures 20-23.


















-- 4MC
.-..---- I

-- 4MC


0 10


20 30


time (min)


0.2
0.18
0.16
0.14
0.12
0.1
0.08
0.06
0.04
0.02
0


-*- Blood I

-Brain I


S- I I
0 10 20


time (min)


Figure 20. A) In vivo brain concentration of the CDS J, its quaternary
metabolite I, and final metabolite, 4-methyl catechol (4MC). B) In vivo
blood/brain distribution of quaternary metabolite I. Each value is the mean of two
independent determinations.


0.25

0.2

0.15

0.1

0.05

0












0.25

0.2

0.15

0.1

0.05

0


50 100


time (min)


0.02
0.018
0.016
0.014
0.012
0.01
0.008
0.006
0.004
0.002
0


-U- 4



-.- 5


150


50 100


150


time (min)
B.






Figure 21. In vivo concentrations of CDS 4, its metabolite 3, and final
metabolite 5, in A) rat brain, and B) rat liver. Each value is the mean of two
independent determinations.
















0.08

0.07

0.06

0.05

0.04

0.03

0.02

0.01

0


0 10 20


30 40 50


time (min)













Figure 22. In vivo concentrations of CDS 4, its metabolite 3, and final
metabolite 5, in rat blood. Each value is the mean of two independent
determinations.


3 3

-----5



















0.09 --- blc
0.08 -
0.07 -----br
c 0.06 -
0.05 -
0 0.04-
0.03 -
0.02 -
0.01 -
0 I I
0 20 40 60 80 100

time (min)














Figure 23. Blood/brain distribution of the pyridinium catechol final
metabolite 5. Each value is the mean of two independent determinations.


120








In Vivo NGF-Stimulatory Activity



The effect of the 4-MC-CDS on the biosynthesis of NGF in the specific

brain regions, hippocampus and frontal cortex, was determined by measuring the

levels of NGF mRNA. Levels of NGF mRNA were quantified by the use of the

dot blot technique. In summary, total RNA extracted from fresh tissue samples

was blotted onto a nylon membrane and U.V. crosslinked to avoid further

degradation of RNA. The nylon blot was then hybridized with a [32p]-labelled

NGF probe (771 bp) that recognizes the entire pre-pro sequence of the NGF

mRNA. The extent of hybridization was then quantified with the use of a Betagen

that measures radioactive signal of different intensity. The NGF signal was

normalized by stripping the NGF probe and subsequently rehybridizing the blot

with an actin probe.

For this assay, 3 groups of animals were used. Male Sprague Dawley rats

(body weight 320-360 g) were injected i.v. (tail vein) with: 1) 4MC-CDS (J): 50

mg/Kg body weight (drug vehicle consisted of 1:1 DMSO-50% 2-hydroxypropyl-

P3-cyclodextrin in water, total injected volume: 1.8 ml/Kg), 2) drug vehicle only

(1.8 ml/Kg), and 3) 4MC: dose of 4MC equimolar of the 4MC-CDS, 17 mg/Kg (in

same amount of drug vehicle). After 8 hs animals were sacrificed by decapitation,

and the brains removed and placed on ice-cooled surface. The hippocampus and

frontal cortex were rapidly dissected out and separately placed in volume of lysis

solution (4M guanidinium thiocyanate, 25 mM Na citrate, 0.5% N-lauryl


_ __








sarcosine, and 0.1 M 2-mercaptoethanol, total volume I ml/100 mg tissue) and

immediately homogenized.



RNA Isolation



Total RNA was isolated using the acid guanidinium isothiocyanate method

followed by the phenol/chloroform method of purification and isopropanol

precipitation (Chomczynski, 1987, and Sambrook et al., 1989). All solution used

for the isolation and purification of RNA were DEPC (Diethylpyrocarbonate)

treated in order to eliminate RNAses. Freshly dissected tissue (frontal cortex and

hippocampus) was homogenized in lysis solution (consisting of 4M guanidinium

thiocyanate, 25 mM Na citrate, 0.5 %N-lauryl sarcosine sodium salt, and 0.1M 2-

mercaptoethanol) using a Polytron homogenizer. Following this lysis procedure,

buffer saturated phenol (GIBCO-BRL) and chloroform/isoamyl alcohol (49:1)

were utilized to purify the RNA from protein contaminants: the phenol is used in

order to remove the protein contaminants while the chloroform/isoamyl

purification steps help remove the residual phenol. 2M Na acetate, TRIS buffered

phenol and 49:1 chloroform/isoamyl alcohol was added to the homogenate and

vortexed for Imin. This suspension was subsequently centrifuged at 10,000 x g at

4 oC for 15 min. after which the aqueous top layer was aliquoted into a second

tube. Phenol and chloroform/isoamyl alcohol solution were re-added to the new

tube, vortexed and recentrifuged at 10,000 x g at 4 OC for 15 min. The upper

phase was removed again, aliquoted into a fresh tube, and an equal volume of








isopropanol was added and vortexed. After an overnight storage at -200C, the

isopropanol containing solution was centrifuged at 10,000 x g at 4 OC, upon which

a pellet was formed. The supernatant was discarded and the pellet was dissolved

in a small volume of lysis solution. The isopropanol precipitation was repeated

after which a wash with ice-cold 75% ethanol was performed. The ethanol was

evaporated off and the remaining pellet was resuspended in 0.5% sodium dodecyl

sulphate (SDS). The concentration of the RNA was evaluated spectro-

photometrically at a wavelength of 260 nm. The purity of RNA was assessed by

calculating the ratio of absorbance at 260/280 nm. A ratio of 2 is considered pure

RNA, while samples that had ratios below 1.6 were estimated to be too

contaminated by protein and thus were not used.



Blotting of Total RNA



Two concentrations (5 and 15 jig) of RNA from each sample (to ascertain

linearity of hybridization versus RNA loaded) were filtered though a Nylon

membrane held in a Hybri-Dot manifold (Bethesda Research Laboratories), as

described below: To every 22 gig of RNA sample DEPC water was added to reach

a final volume of 30 pl. Also, plasmid containing the pre-pro-NGF mRNA

sequence was diluted with DEPC water to give a final concentration of 4.5 ng / 30

pl. To these samples, that were kept on ice, formamide (60 p1), formaldehyde (21

pl) and 20 x SSC (sodium chloride 30 M and sodium citrate 0.3 M) (6 il) were

added and incubated at 68 OC for >15min. After assembling and preparing the








Nylon membrane on the Hybri-Dot manifold, 243 .Il of ice cold 20x SSC was

added to each sample, vortexed, and placed on ice until loaded (by suction) on the

Nylon membrane and U.V. crosslinked (using a U.V. Crosslinker). The dot blot

was subsequently stored in a dessicator until the time of hybridization.



Preparation of the NGF Probe



The NGF probe was a gift from Dr. Scott Whittemore (Miami, Fl) and Dr.

Paul Isackson (Mayo Clinic, Jacksonville). The probe was received incorporated

in pBS (Bluescript). The excision of the probe from the plasmid and its

purification was performed by Dr. Sonny Singh as follows: A restriction enzyme

digestion was carried out using the enzymes BAMH I (4U/mg DNA) and ECOR I

(5 U/mg DNA) at 37 OC for 3 hs. At this point the excised probe was separated

from the plasmid by running the sample on a 1 % low melting temperature agarose

gel. After running the gel for 3 hs at 60 V in a cold room, the bands were

photographed and the band corresponding to the NGF probe (by comparison with

the DNA ladder, the separated DNA fragment had an approximate size of 800 b.p.

which is very close to the actual size of the NGF probe, 771 b.p.) was cut out of

the gel and placed in an eppendorf tube. The following day, the probe was

extracted from the gel and purified using the phenol/chloroform/isoamyl alcohol

method described previously. After the ethanol wash, the resulting pellet was

resuspended in 1 x TE (pH=8) (TRIS and EDTA solution). The concentration and








purity of the DNA probe was assessed spectrophotometrically (a ratio A260/A280

nm of 1.8 is considered pure DNA).



Radiolabelling of NGF Probe and Hybridization of Nylon Membrane



The Nylon membrane was placed in a "hybridization" vial with

prehybridization buffer and incubated at 420C for approximately 2 hs. The

prehybridization buffer consisted of 5 x SSPE (solution of NaCI, NaH2PO4.H20

and EDTA), 5 x Denhardts solution (ficoll, polyvinyl pyrrolidone, and BSA), 50%

formamide, 0.5% SDS, and 100 gig/ml denatured salmon sperm DNA. During this

prehybridization time period, the NGF probe (pre-pro NGF) was labelled with [a-
32p]-CTP (Specific activity: 3000 Ci/mM,10 giCi/tl, DuPont, New England

Nuclear, Cambridge, MA) using the Random primer probe labelling kit (Gibco-

BRL). After the labelling procedure was complete, the resulting labelled probe

was separated from the free [32P]-CTP by passing the reaction solution through a

Sephadex-50 column with 1 x STE (NaCI 0.1 M, Tris.HCl 10 mM, and EDTA 10

mM). From the labelled probe fraction, l1gl was added in 4 ml of scintillation

cocktail and radioactivity measured with a scintillation counter. Based on a

precalculated volume of hybridization buffer to be used, the amount of labelled

probe that would give a final activity of 1-2 x 106 cpm/ml of hybridization buffer

was added to fresh prehybridization buffer producing the final hybridization

solution that was maintained at 42 OC until the time of use. Following the

prehybridization of the blot, the prehybridization buffer was discarded and the








nylon membrane was placed in the prepared hybridization buffer and incubated at

42 oC. Hybridization was carried out for 22 hs after which the nylon membrane

was subjected to 4 washes (washes 1 and 2 with 2 x SSC/0.1% SDS at room

temperature and 60 OC respectively, and washes 3 and 4 with 0.2 x SSC/0.1% SDS

at 60 OC). After the washes, the membrane was placed on Whatmans filter paper,

and while still damp, placed in a Betagen densitometer for overnight quantitation

of NGF mRNA signal. The Dot Blot was quantitated by computer image analysis

with the Betascope Blot Analyzer, Model 603 (Betagen Corp., Waltham, MA).

(The data were calculated as the counts of [32P]dCTP-NGF/gg RNA collected

over a 16 h time period). Following this, the NGF probe was stripped from the

membrane by incubation in a solution of ImM Tris.Cl (pH 8.0), 1mM EDTA

(pH8.0), and 0.1 x Denhardt's reagent, for 2 hs at 75 oC. The membrane was then

rinsed with 0.1 x SSPE at room temperature and was ready for rehybridization

with the actin probe. Since the amount of actin mRNA is considered an indicator

of total RNA per sample, the hybridization with the actin probe was done in order

to normalize the amount of NGF mRNA measured per total amount of RNA

loaded for each sample. Thus, an increased NGF signal for a sample would

represent, after normalizing, true stimulation of the specific mRNA synthesis and

could not be attributed to variation of the total RNA loaded for that sample. The

radiolabelling of the actin probe, subsequent hybridization of the membrane with

it, and the quantitation of the actin mRNA signal was performed using the same

procedure described previously for the NGF probe and mRNA.










Statistical Analysis and Results



Results were expressed as counts per min (cpm) for the NGF signal by cpm

for the actin signal (NGF mRNA units/actin mRNA units). Analysis was

performed using the ANOVA followed by Fisher PLSD test for determination of

group differences.

Results showed a 1.76 fold increase in NGF mRNA compared to control in

the rat hippocampus (1.3 fold increase compared to the 4-methylcatechol control)

and a 1.32 fold increase in the frontal cortex (1.22 fold increase compared to 4-

methylcatechol control) (Figures 24 and 25).












Hippocampus


Control 4-MC


4-MC-CDS
(J)


Figure 24. Effect of 4-MC-CDS (J) on Hippocampal NGF mRNA levels.
Each value is the mean S.E. of 4 to 6 animals. P < 0.05 relative to control
according to Fisher PLSD.

















Frontal


1






.**. ^ ."'


Cortex


Control


4-MC


4-MC-CDS
(J)


Figure 25. Effect of 4-MC-CDS (J) on Frontal Cortical NGF mRNA
levels. Each value is the mean S.E. of 4 to 6 animals.


0.4-



0.3-



0.2-



0.1-


.. -.. ........ ....
,. .Oo ..... -........,

...... .......:.- .::
,......-.-..,.'.....'..
..-...-.- ..... .....
............ .

............ 7 7:' ...








In Vitro NGF-Stimulatory Activity



Cell Cultures



Mouse L-M cells were obtained from the American Type Culture

Collection (ATCC) (Rockville, MD) and maintained as monolayer cultures in

medium 199 (Sigma) supplemented with 0.5% bactopeptone (Difco Laboratories,

Detroit, MI), antibiotic free. Stock cultures were grown in 25 cm2 flasks and

subcultured once a week. To study the effect of the compounds, cells were

inoculated in 24-well plates (well surface 2.1 cm2) at a cell density of 4-8 x 104

cells/well and cultured for 3 days in peptone containing medium after which the

medium was changed to medium 199 containing 0.5% bovine serum albumin

(BSA) with or without any compounds, and the cells were cultured for 24 hs.

C6 glioma cell line was also obtained from the ATCC and was grown as

monolayer cultures in RPMI-1640 (Sigma) serum free medium supplemented with

1% L-Glutamine and 0.5% penicillin/streptomycin in a humidified chamber with

95% air and 5% C02 at 37 OC. Incubation with compounds was performed in 25

cm2 flasks (1-1.5 x 106 cells/flask) in the same serum free (SF) medium at 37 oC

for 17 hs.

For determination of the cellular NGF mRNA levels of the C6 glioma cells,

cultures were grown to confluency and subsequently incubated with or without

compounds for 4 hs.