Design, synthesis and evaluation of excessively charged nucleophiles as potential reactivators of aged phosphorylated ac...

MISSING IMAGE

Material Information

Title:
Design, synthesis and evaluation of excessively charged nucleophiles as potential reactivators of aged phosphorylated acetylcholinesterases, and N-bromo imides as universal decontaminants
Physical Description:
xvi, 246 leaves : ill. ; 28 cm.
Language:
English
Creator:
Radhakrishnan, B
Publication Date:

Subjects

Subjects / Keywords:
Acetylcholinesterase   ( lcsh )
Organophosphorus compounds   ( lcsh )
Antidotes   ( lcsh )
Genre:
bibliography   ( marcgt )
theses   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1989.
Bibliography:
Includes bibliographical references (leaves 237-245).
Statement of Responsibility:
by B. Radhakrishnan.
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:
aleph - 001540000
notis - AHF3469
oclc - 22453914
System ID:
AA00003355:00001

Full Text












DESIGN, SYNTHESIS AND EVAUAICON OF EXCESSIVELY CHARGED
NUCLFHDLES AS POTENTIAL REACITVATORS OF AGED
FROSPHORYATED ACEIYLCHOLUNESTASES, AND N-BRMD
IMIDES AS UNIVERSAL DECDNAMIUNANIS







By

B. RADAKRISHNAN


A DISSERITTION PRESENTED TO 1TE GRAMIATE SCHOOL
OF THE UNIVERSITY OF FLRIDA
IN PARTIAL FULIFIMENT OF THE REQUIREMENTS FOR THE
DEGREE OF DOCIOR OF PHIISOPHY

UNIVERSITY OF FIDRIDA


1989




































To Father and Mother

































Good reasons must, of force, give place to better.



Shakespeare














ACKNOWLEDGMENS


I would like to express my deepest gratitude to Dr. Nicholas Bodor

for the invaluable guidance, generous support, and the unique

opportunity to work directly with him and make this interdisciplinary

program come true. I would like to express my sincere appreciation to

Dr. Alan Katritzky for his advice, help, and for having me as the

"adopted son" of his group. In addition, I would like to thank the

members of my graduate committee, Dr. Hartmut Derendorf, Dr. John

Selling and Dr. Kenneth Wagener, for their advice and encouragement.

My special thanks are extended to Dr. Dave Winwood, Dr. Marcus

Brewster and Dr. Gene Brown for their immediate guidance and helpful

hints during my work at the Center for Drug Design and Delivery (CDDD)

laboratory at Alachua.

I am grateful to all my colleagues and members of Dr. Bodor's

group and the Center for Drug Design and Delivery at the College of

Pharmacy and members of Dr. Katritzky's group in the chemistry

department. I sincerely thank everyone at the CDDD Labs for their

friendliness and for providing all facilities during the stages of my

"experimental era."

I would also like to express my deepest thanks to Laurie Johnson,

Joan Martinago and Julie Drigger for their indisputable part in my work.














I would like to extend my thanks personally and collectively to all

those in the chemistry and medicinal c neistry departments who have

provided their support throughout my stay at the University of Florida.

Finally, I wish to thank all my friends for their moral support in

the pursuit of this degree.


































v














TABLE OF OCt1EMIS

age

ACKNOWLED0GE14S.................................................. iv

LIST OF TABLES.................................................... vii

LIST OF FIGURES.............................................. ix

KEY TO ABBREVIATIONS......................................... xiii

ABSTRACT .................... ................................... xiv

CHAPTERS

I INTRODUCTICON..................... .................... 1

Pharmacology of Acetylcholinesterase and
Organophosphate Poisoning....................... 1
Specific Aims (Reactivators of Aged AChE)............. 22
Universal Decontaminant (UD)......................... 33
Specific Aims (N-Haloamine Decontaminants)............ 45

II RESULTS AND DISCUSSION................ .............. 56

Synthesis of ECNRs and Phosphate Models.............. 56
Synthesis of N-Bramo Canpounds and their Derivatives.. 84
Physical-Chemical Studies of ECNR COmpounds ........... 99
Evaluation of N-Brmo Compounds........................ 149

III EXPERIMENTAL................. ........................ 170

Materials............................................ 170
Methods......................................... ...... 171
Compounds....................... .................. 172
Physical-Chemical Studies................................. 219

IV SUMMARY AND CONCLUSIONS.............................. 233

REFERENCES. ........ .......... ... .................... ............ 237

BIOGRAPHICAL SKEI.......................................................................... 246














LUST OF TABLES


Table BMe
1-1 Half lives of aging of some organophosphate
derivatives of acetylcholinesterase................... 13

2-1 Apparent ionization constants of ECNR and reference
ccpounds........................................... 100

2-2 Critical micelle concentration of sane ECNR
and reference ccapounds at 350C........................ 102

2-3 Kinetic data for the hydrolysis of Paraoxon at various
concentration of 14 in aqueous buffer pH 9.5 at 350C.. 106

2-4 Kinetic data for the hydrolysis of octyl methyl 4-nitrophenyl
phosphate by 14 in aqueous buffer pH 9.3 (0.03M)
at 350C............................................... 110

2-5 Hydrolysis of Paraoxon by hydroxamic acid 15 (non-micellar
analogue 14) in pH 9.5 (0.03M) at 350C................ 112

2-6 Kinetic data of the reaction of oxime ECNRs with
Paraaxon and octyl methyl PNPP at 350C
in pH 9.4 (0.03M) ................................... 114

2-7 Effect of CEAB on the rate of reaction of 14 with Paraoxon
at pH 9.3............................................. 117

2-8 The reactivity of hydroxamic acid 14 with Paraoxon at
various pH buffers (0.03M) at 350C................... 119

2-9 Kinetic data for the reaction of bis 4-nitrcphenyl
phosphate (32) by hydroxamic acid 14 in aqueous buffer
pH 9.3 (0.03M) at 350C.............................. 134

2-10 Kinetic data for the biphasic reaction of 14 with bis
2,4-dinitrphenyl phosphate (33) in aqueous
buffer pH 9.4, 8.0 and 7.0 at 350C ................... 128

2-11 Kinetic data for the hydrolysis of dodecyl, octyl, butyl and
benzyl 4-nitrophenyl phosphates by 14 in aqueous buffer
pH 9.5 at 350C....................................... 132



vii





I____________















2-12 Ocmparison of ECNR reactivity with other reported systems
against ethyl 4-nitrqphenyl phosphate (2)............ 137

2-13 Kinetic data for the hydrolysis of ethyl 4-nitrcphenyl
phosphate (29) by hydroxamic acid 14 in aqueous buffer pH
9.5 at 350C......................................... 138

2-14 Kinetic data for the hydrolysis of 4-nitrqphenyl
phenylphosphonate by M at 350C in pH 9.5 (0.03M) ..... 143

2-15 Equilibrium constants of the reversible reaction between N-
braminating agent and succinimide to form free
oxazolidinone and N-bramosuccinimide (NBS)........... 152

2-16 Stability of NHS (47), NE Succn (49), tetrabrano dimethyl
glyoouril (54(b)) and NBO (43) (reference) in various
buffer solution and in solid form ..................... 156

2-17 Stability of tetrabrano dimethylglycouril 54(b) in
hydroxyprypyl P-cyclodextrin (HPCD) formulation....... 158

2-18 An estimation of release of active bramine species (OBr")
in solution of 3-brano axazolidinane system as a
function of the solution pH .......................... 162

2-19 Hydrolysis of Paraoxon by NBMOS (47) as a function of pH at
30C.................................................. 166

2-20 Hydrolysis of Paraoxon by NEOSucc n (49) as a function of pH
at 30C.................................... 167

2-21 Hydrolysis of Paraoxon as the function of the concentration of
NR3S (47) in pH 11.17 at 350C ....................... 168












viii














IMST OF FIGURES


Figure Pae

1-1 Organctosphorus anticholinesterase agents................ 2

1-2 Normal functioning of acetylcholinesterase.................. 5

1-3 Inhibition of acetyldcolinesterase by an
organphsphorus derivative........................... 7

1-4 Reactivation of inhibited AChE by a conventional
oxime (2-PAM) ......................................... 10

1-5 Aging of the poisoned (phosphorylated) AChE................. 12

1-6 Diester anion 2-PAM complex.............................. 21

1-7 General structure of hydroxamic acid ECNRs.................. 24

1-8 Retrosynthetic scheme for hydroxamic acid ECNRs ............. 25

1-9 General structure of oxime ECNRs .......................... 26

1-10 Retrosynthetic scheme for oxime ECNRs....................... 27

1-11 General structure of aged enzyme models and
organrxphsphorus substrates .......................... 28

1-12 Retrosynthetic scheme for the phosphate substrates........... 30

1-13 Reaction of hypochlorite anion with Scanan.................. 39

1-14 Reaction of mustard agent with hypbrcnous acid............. 40

1-15 Retrosynthetic scheme for the derivatives of
4-hydroxy-4-methyl-2-oxazolidinnes ................... 48

1-16 Retrosynthetic scheme for dimethylglycouril systems ......... 51

2-1 General synthetic scheme for hydroxamic acid ECNRs.......... 57








2-2 Direct synthesis of N- (-diethylamincethyl)-O-benzyl
hydroxylamine ........................................ 60

2-3 The tH NMR spectrum (300 MHz) of .6 in Cl,3 ................ 64

2-4 General synthetic schees for oxime ECNR compounds .......... 67

2-5 Proposed pathway of decarboxylation reactions of the
l-carbxynethylpyridinium-2-alddme ....................... 68

2-6 Alternative synthetic scheme to pyridinium ECNR: Via
quaternization with brarmesters of aminoalcohols...... 70

2-7 Preparation of bramoacylesters of quaternary aminoalcohols.. 72

2-8 'he l NMR spectrum (300 MHz) of 16b in EMO-d
trace Hp .................. ........ .................. 74

2-9 The H NMR spectrum (300 MHz) of 18(b) in EMSO-
trace Hl............................................. 75

2-10 he 'H NMR spectrum (300 MHz) of 19b) in MsO-d,
trace p.......................................... .... 76

2-11 General synthetic sequence for esters of 4-nitrophenyl
phosphoric acids..................................... 79

2-12 Diester hydrogen phosphate synthesis.................. ...... 81

2-13 The 1H NMR spectrum (90 MHz) of n-butyl 4-nitrcphenyl
hydrogen phosphate in CDC13............................ 83

2-14 Synthetic scheme of (4-brcmo4-methyl-2-cxazolidincne-4-yl)
methyl sulfate...................o.................... 88

2-15 Brnination of (4-methyl-2-cxazolidin~oe-4-yl)methyl sulfate 89

2-16 Synthetic scheme for polar derivatives of 4-hydroxymethyl-4-
methyl-2-oxazolidinone................................ 91

2-17 The 'C chemical shift assignment of
4- (-diethylamincethoxymethyl-4-methyl-2-oxazolidinone
(50) and model carpounds.............................. 93

2-18 he 1C NMR spectrum (74.5 MHz, MSO-d) of 4-hydroxymethyl-
4-imethyl-2-oxazolidinane (AA) ............... ........ 94

2-19 The 1C MR spectrum (74.45 MHz, CD13)
4- (-dimethylaminetxymethyl-4ethyl-42-cxazolidinone
(5Q) ............................................... 95


x








2-20 Synthetic scheme to dimethylglycouril (4) and attempted
mocnsubstitution reactions............................ 97

2-21 The structure of ECUR ccmpounds............................. 107

2-22 Hydrolysis of Paraoxon by hydrxamic acid 14. in pH 9.3 at
350C......................................................... 108

2-23 Hydrolysis of octyl methyl 4-nitrcphenyl phosphate by
hydroxamic acid 14 in pH 9.4 at 350C.................. 108

2-24 Hartley model of micelle .................................... 109


2-25 Relative reactivity of xime ECNRs against Paraoxn. ........ 115

2-26 Hydrolytic rate of Paraxon by hydroxamic acid 14 under
non first-order rate conditions....................... 121

2-27 First-order rate constants for the hydrolysis of bis
4-nitrophenyl phosphate (M5) at various amount 14
in pH 9.5 at 35C................................... 125

2-28 Hydrolysis of dodedecyl 4-nitrqphenyl phosphate (2&) by
14 in aqueous buffer pH 9.5 at 350C................. 133

2-29 Hydrolysis of ethyl 4-nitrcphenyl phosphate (29) by 14...... 139

2-30 Relative reactivity of ECNR against aged enzyme
odels............................................... 140

2-31 Hydrolysis of ethyl 4-nitrphenyl phosphate by 14 under
non first-order rate conditions........................ 141

2-32 Hydrolysis of dodecyl 4-nitrqphenyl phosphate by 14
under non first-order rate conditions ................ 141

2-33 Hydrolysis of phenyl 4-nitr phenylphosphonate by hydroxamic
acid 14 in pH 9.6 at 350C ............................. 144

2-34 A molecular model representation of the mode of action of 14
with orgamphosphorus monoanion ....................... 147

2-35 Hydrolysis of Paraoxon at various concentration of NMOS (7)
in pH 11.2 at 350C .................................... 168







xi


I














KEY TO ABBREVIATIONS


CMC:

CIAB:

DOD:

But:

Et:

PNPP:

NP Phospn:

2,4-ENPP:

HA:

BzHA:

r.t:

th:

K12:

k:

At:



r:

accn:

Exct:

EBiss:

X:

log:


critical micelle concentration

Hexadecyltrimethylammonium brcnide (co-surfactant)

n-dodecy group

n-butyl group

ethyl group

p-nitxrphenylphosphate moiety

p-nitrophenylphosphonate moiety

2,4-dinitrophenylphosphate moiety

imidazolium hydroxamic acid

benzyl protected imidazolium hydrxamic acid

room temperature

half life, based on log of activity vs time

apparent bimolecular rate constant

pseudo first-order rate constant

end point absorption

absorption at time t

correlation coefficient

reaction rate acceleration based on buffer hydrolysis

excitation

emission

wavelength

log~0


xii
















M: mole per liter

Bis: disubstituted

mono: mcnosubstituted

min: minutes

sec: seconds

ri: nanometer

UV: ultraviolet

i4: microliter

ECQR: excessively charged nucleophilic reagent


xiii














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 Thilosophy

DESIGN, SYNIESIS AND EVALUATION OF EXCESSIVELY CHARGED
NUCLEOHIUES AS POTENTIAL REACTIVATCRS OF AGED
H6SPH~YYIATED ACEY lIJNESTERA~F SES, AND
N-BRC IMIDES AS UNIVERSAL DECXIMINAMNS

By

B. Radhakrishnan

December 1989

Chairman: Alan R. Katritzky
Cochairman: Nicholas Bodor
Major Department: Chemistry

Acetylcholinesterase (AChE) enzyme has a major physiological

function in humans in ensuring that neurotransmission is of finite

duration by cleaving neurotransmitter acetylcholine at the nerve

junctions. The organophosKhorus nerve agents inhibit this process by

irreversibly phosphonylating or phosphorylating the enzyme, which

results in elevated levels of acetylcholine, causing convulsion,

muscular paralysis and death.

Inhibited AChE can be reactivated by conventional nucleohilic

antidotes such as 2-PAM and Obidoxime. However, the inhibited enzyme

"ages" from a reactivatable phosphate triester state to a

nonreactivatable phosphate diester anion or an analog thereof, which is

extremely inactive towards any nucleophiles due to anionic repulsion.

So far, no antidotes have been shown to reactivate the aged enzyme.


xiv










A series of multicationic nuclecphiles, namely, excessively

charged nuclecphilic reagents (ECNR), same having surface activity, have

been designed, synthesized and investigated for the hypothesized mode of

action: (a) neutralization of repulsive charge and (b) simultaneous

nucleophilic attack against the aged enzyme models such as phosphate

diester and phosphnate monoanions.

The ECR carrying hydroxamate anion showed hydrolytic cleavage of

the monoanions with 40-to 2,600-fold rate enhancement at pH 9.5. The

reaction appears not to be influenced by micellar action but takes place

by the hypothesized mode. The ECNR carrying oximate anions were less

effective against the aged enzyme models. Both types of ECNR showed

rapid acceleration of hydrolysis of organophosphate triesters. Some of

the ECUR should have in vivo applications.

Derontamination by hydrolytic cleavage using positive halogen

sources is one effective method to decontaminate nerve gas and mustard

agents in field or under various environmental conditions. Sane stable,

water soluble, "soft" brcminating N-brcmo 2-oxazolidinone systems, as

well as an N-brmno glycouril system with increased molar positive

branine, were designed, synthesized and investigated for use as

universal decontaminants. The water-soluble 2-oxazolidinone system

showed high stability and activity against both mustard and nerve gas

(organophoshate) agent models. The glycouril system is exceptionally

stable as a solid.











All synthetic strategies and methods to target molecules and the

structural characterization of products and side products as well as

physical-chemical studies and kinetic processes are discussed.














CHAPTER I
INTRODUCTION


This thesis will examine the design, synthetic methods, and

evaluation strategies that are relevant to successful development of

potential reactivators (antidotes) of organophosphate blocked "aged"

acetylcholinesterases (AChE) and a new class of decontaminants of

organophosphates.

Pharmacolovg of Acetvlcholinesterase and Oranophosphate Poisoning

The organophosphate anticholinesterase agents (nerve agents) are

highly toxic derivatives of phosphonic and phosphoric acid.

Representatives of this class of compounds have widespread use in

agriculture (e.g., parathion, malathion), pharmacology and medicine

(e.g., DFP, paraoxon), and are of concern to the military because of

their application as chemical weapons in war (e.g., tabun [GA], sarin

[VD] and [VX]). The structures of these ccmpounds and rat ID, values

are given in Figure 1-1.1

All of these organophosphates elicit their primary effects by

phosphorylating or phosphonylating the serine hydroxyl group at the

active site of the enzyme acetylcholinesterase, causing essentially

irreversible inhibition of the enzyme. Differences in their ultimate

pharmacological effects and their lethality (toxicity) are derived from

their relative ability to penetrate into the central nervous system

(CNS), their rate of reaction with AChE, and their selectivity for








2

reaction with the enzyme at specific loci, and their behavior (e.g.,

aging) once attached at the active site on the enzyme.


(CH3)2N 0


02H5 0 Z N


LD50


Tabun (GD)
0.20


CH3
t- Bu--O 0

H P F
CH3


i C3H70 O


CH3 F

Sarin (GB)
0.14






CHsOs
C2H50 0S
C SCH2CH2-N(i C3H7)2
CH3


Soman (GD)
0.10


EtO /

EtO Q/ NO2


Paraoxon

0.50


LDso(rat) mg/kg, Route sc


Figure 1-1: Organophosphorus anticholinesterase agents.


LD50


VX
0.02


- C3H70 0


i-C3H70


LD50


DFP

3.00




-7


3

Normal Function and Inhibition of AChE

Acetylcholine is found as a metabolite in high concentrations in

nervous tissue and in motor nerve tracts, as an active neurotransmitter

substance. When a nerve cell enervates a muscle fibre, the nerve-cell

ending at the muscle junction forms a morphologically distinct and

functionally discrete contact called a synapse. This is a gap

(typically of about 500 A) between the presynaptic nerve-ending membrane

and postsynaptic membrane of the muscle-cell membrane. This interaction

transmits a depolarization signal to the muscle membrane in a manner as

yet imperfectly understood.

The physiological role of acetylcholinesterase is to ensure the

nerve impulse is of finite duration by lowering the acetylcholine

concentration in the cleft via hydrolysis. This enzymatic cleavage will

lower the concentration of free acetylcholine in the synaptic cleft, and

the resulting perturbation of equilibrium will induce dissociation of

the acetylcholine-acetylcholine receptor complex, turning off the signal

to the muscle cell. Active acetylcholinesterase is crucial for normal

neuronuscular function, and inhibition of this enzyme produces tetanic

shock, with eventual muscle paralysis. In severe cases asphyxiation can

result. Not surprisingly, this enzyme is a target for nerve agents and

insecticides such as Saman and diisopropylphosphofluoridate (DFP).

AChE is loosely associated with postsynaptic membrane.2 As a

catalytic entity, AChE is remarkably efficient, with a turn over number

(K.t) of 25,000 sec' and will cleave a substrate molecule once every 40

psec. The active site of AChE as depicted in Figure 1-2, is considered

to consist basically of two subsites: an anionic site, possibly










represented by a glutamate ion, and an esteratic site, located about 2.5

A from the anionic site which has been shown to incorporate a serine

moiety and may contain a histidine residue (p, 7.2) and a tyrosine

residue (pK 9.3).3 A hydrcphcbic area at the active site is also

indicated.

In normal functioning, acetylcholine, after acting at a

cholinergic receptor, complexes reversibly with the active site of the

enzyme. In a fast step, the acetyl group is transferred from the

acetylcholine molecule to a serine hydroxyl group, forming acetylated

enzyme and releasing choline. This is followed by a rate limiting

hydrolysis of the acetate ester group, presumably acid catalyzed by the

imidazolium ion and possibly also aided by nucleophilic action of the

tyrosine hydroxyl, releasing acetate anion and providing reactivated

enzyme available to hydrolyze another molecule of acetylcholine. The

released choline is transported back into the nerve ending for

reconversion to acetylcholine and storage.

It is recognized that, although the active site has been fairly

well elucidated, acetylcholinesterase is an enzyme of highly complex

structure that has by no means been fully characterized. The enzyme

appears normally to consist of several subunits, and enzyme obtained

from different species and different tissues (e.g., erythrocyte vs.

brain) can show significant quantitative differences in sensitivity to

organcphosphorus agents and ability to be reactivated after inhibition.

















H 0

0 C-O


ENZYME SURFACE


Hist Ser Tyr


ESTERATIC SITE


ANIONIC SITE


Step 1


Enzyme- OH


O
II +
+ CH3C-OCH2CH2N(CH3)3
Acetylcholine


Step 2

Enzyme-O-CCH3



Step 3

H2O Enzyme-OH
Rate-limiting


+
HOCH2CH2N(CH3)3 (Choline)


+ H30+


Figure 1-2: Normal functioning of acetylcholinesterase.


Complex


CH3 O-










The organophosphorus anticholinestrases react rapidly with the

enzyme; the nerve agents react particularly rapidly. A lethal dose of

Scman, for example, will have completely reacted within minutes of

administration to an animal. A hypothetical formulation of the process

is depicted in Figure 1-3. The more potent organrphosphorus agents

react so rapidly with enzyme (in seconds), initial kinetic studies

indicated the reaction to be a simple bimolecular process without

intervention of a binding complex of the reversible Michaelis-Menton

type. Since the reaction is stoichicmetric, one molecule of inhibitor

abolishing the functioning of one enzyme active site, the extreme

toxicity of the organophosphorus agents can be understood when one

considers that acetylcholinesterase is present in the body in only small

amounts.

The organophosphorus esters act by acylating the serine hydroxyl

group at the active site of AChE, causing essentially irreversible

inhibition of the enzyme with consequent elevation of acetylcholine

levels. Acetylcholine accumulates in the peripheral and central nervous

system (CNS). Ultimately, at sufficiently high level there is

depression of the respiratory center in the brain abetted by peripheral

neuromuscular blockage, causing respiratory paralysis and death.

Generally more lipophilic, organophosphorus esters show high

neurotoxicity as the result rapid penetration to the CNS system.

However, the acute toxicity effects of the organophosphorus

anticholinesterase agents are elicited by the elevated acetylcholine

levels resulting from inhibition of the enzyme.











In sharp contrast to the rapid hydrolysis of the acetylated

enzyme, hydrolysis of phosphorylated or phosphonylated enzyme is

extremely slow, with a half-life typically of the order of hours to

days.3 Thus the enzyme is effectively inactivated and prevented from

carrying out its normal function of hydrolyzing acetylcholine.


( F


(e.g., Soman)


Hist Ser Tyr
ESTERATIC SITE


Glu

ANIONIC SITE





Modified Ser side chain
,-*'


+X" + H





INHIBITED ENZYME


Hist Ser Tyr
ESTERATIC SITE


Glu

ANIONIC SITE


Figure. 1-3: Inhibition of acetylcholinesterase by an organcphosphorus
derivative.


^
*











Reactivators (Antidotes) of AChE

Despite considerable effort taken in design of antidotes

(particularly by the military ever since the development of Tabun in

Germany in 1937) no fully satisfactory therapeutic or prophylactic

agents have yet been devised.2 One of the most effective reactivation

nuclecphiles yet found is pyridine-aldoxime methiodide, or 2-PAM. It

illustrates rational drug design based on a knowledge of enzyme

mechanism. The military has considerable interest in prophylactic as

well as therapeutic agents, the former intended for protection of

personnel in the event that nerve agent attack appears imminent.

Design of antidotes has focused on essentially two ways to counter

the toxicity of organophosphorus derivatives, namely (1) blocking the

cholinergic effects of the elevated acetylcholine levels, and (2)

reducing those levels. So far no single compound acting in either of

these ways has proved completely satisfactory. Efforts to achieve

reduction of the elevated levels of acetylcholine induced by the

organophosphorus nerve agents have largely focused on the development of

nucleophilic agents that can reactivate inhibited AChE by cleaving the

bond attaching the phosphoryl group to the serine hydroxyl at the active

site.

A large number of nucleophilic agents were prepared and tested in

the early 1950s; two groups, hydroxamic acids and oximes, emerged as

promising reactivators. With respect to hydroxamic acid, the effective

nucleophile is hydroxamate anion, at physiological pH. However, the

oximes appeared to offer more promise than hydroxamic acids. The

prototype agent for this purpose is an oxime derivative, 2-










pyridinealdoxime methiodide (1, 2-PAM) or methyl methylsulfonate (P2S),

designed some years ago by Wilson et al.4 Since then structure-

activity relationship among the huge number of analogues of 2-PAM that

have been prepared have been studied, Obidoxine C(,Toxogonin) has been

found to be somewhat more effective in reactivating AChE in animals.








CH-NOH NOHC -CH,2CHfN CH=NOH

1H3




2-PAM (1) Obidoxime (2)


A hypothetical picture of the way in which 2-PAM may reactivate

inhibited enzyme is presented in Figure 1-4. ihe pyridinium ion

complexes with carboxylate anion at the anionic site, anchoring the

oximino anion in position to displace the phosphorus ester group from

the electrostatic site. The tyrosine and the histamine residues

possibly assist in the displacement. It should be noted that

reactivation with 2-PAM requires the anti-isamer of 2-PAM; the anti-syn

interconversion is a very facile process that may occur in the oximate

form at the enzyme level.5























INHIBITED ENZYME


Hist Ser Tyr Glu


ESTERATIC SITE


+N



N
H


ANIONIC SITE

0

0- p-
i-
I .
+ R


H

I


ESTERATIC SITE


ANIONIC SITE


R,
p- OH
I.
R


NEC'


Figure 1-4: Reactivation of inhibited AChE by a conventional oxime
(2-PAM).


N+
CH3


I I I








11

2-PAM chloride, in conjunction with atropine, has generally been

adopted in the United States for treatment of poisoning by

organcphosphorus esters, whereas Obidoxime is preferred in same European

countries and the methasulfonate salt of 2-PAM (P2S) in others.

However, a major drawback of all available nuclecphilic reactivators is

that they cannot reactivate aced enzyme,

Acina of Inhibited Acetvlcholinesterase

It is generally accepted that after an organaphasphate reacts with

the AChE, a process of aging of the inhibited AChE will take place.6

This aging is due to the transformation of the inhibited enzyme from a

form which can be relatively easily reactivated by appropriate

nucleophiles, to one which cannot be dephosphorylated at all. The time

required for aging depends very much on the type of organophosphates

involved. Thus AChE inhibited by DFP is transformed into the non-

reactivatable form in several hours. In contrast, aging due to Saman

occurs in 2 minutes (Table 1-1). The aging process of Scman inhibited

AChE is depicted in Figure 5. It was suggested 7 that the rate

determining step in the aging reaction of an alkylphosphorylated AChE is

the fission of one of the C-O bonds. This theory was substantiated by

Michael et al.8 The loss of the pinacolyl group, probably by an acid-

catalyzed process, and presence of the imidazolium ion may play critical

role in initiating the process.9 his aging process transforms the

inhibited enzyme from a form which can be relatively easily reactivated

by appropriate nucleophiles, to one which cannot be dephosphorylated at

all.






















H..
! ,


INHIBITED ENZYME


Hist Ser Tyr

ESTERATIC SITE


Glu

ANIONIC SITE


Hist Ser Tyr

ESTERATIC SITE


Glu

ANIONIC SITE


and other products



AGED ENZYME*


Figure 1-5: Aging of the poisoned (phosphorylated) AChE.








13

Chemically, it is evident that according to the aging process the

main change is essentially the transformation of a neutral phosphate

ester (neutral triester) into a negatively charged phosphate ester

(diester anion). The diester anion is significantly more stable than

the original neutral triester, for the following reasons: the first is

an electrostatic effect, a direct repulsion between the diester anion

and the attacking OH" hydrolysiss) or other nuclecphilic anion (acetate,

phosphate, carbonate, oximate, hydroxamate, thiosulfate, etc.). The

second factor is the decreased reactivity of the phosphorus because of

the decrease in its partial positive change. As a consequence of these

effects, the second-order rate constants for attack by anionic reagents

(nucleophilic reagents) are from 2,000 to 7,000 times smaller for the

diester anions than for the triester.'0



Table 1-1: Half lives of aging of same organophosphate
derivatives of acetylcholinesterase.


AChE-Nerve agent tr

DFP 4h

GA (Tabun) 14h

GB (Sarin) 3h

GD (Sanan) 2-6 min








14

The anions of simple phosphate diesters are generally exceedingly

unreactive.11 Since the pKa of the dialkyl hydrogen phosphate is

between 1 and 2, the species present under most physiological conditions

after the aging takes place is the anion, and consequently the first

problem to be solved is the acceleration of hydrolytic cleavage of

phosphate diester anions.


RO ,O K1
P
RO OR FAST
Triester


o-\ o0
P
OR' OR
Diester anion


K2 o Po
SP O-
I O
SLOW OR'
Monoester dianion


S 2,000 7,200 times


RO .O
'P
O R1

Neutral Triester form


(Inhibited Enzyme)


-R
fast


reactivatable


,..uNu (repulsion)

O0 O
0' R,
PV
O" "R1

Diester anionic form


(Aged Enzyme)

nonreactivatable


Eqn 1-1


Eqn 1-2








15

Various efforts have been made to circumvent the problem of

reactivation of aged enzymes. Among these are (a) the design of agents

capable of realkylating the enzyme bound phosphate anion, converting it

back to neutral triester and thus making it again susceptible to

nucleophilic reactivators such as 2-PAM (to date, no such alkylating

agent has been found); (b) investigation of agents to slow the rate of

aging in vitro (certain amines and ammonium salts, particularly

bispyridinium salts, have been found capable of retarding the rate of

aging in vitro)12; (c) investigation of prophylactic treatment of AChE

temporarily with a reversible inhibitor, thus temporarily preventing a

subsequent dose of organophosphate from gaining access to the

physiologically important enzyme13; and (d) development of reactivators

with the concept that a suitably constructed nucleophilic reagent might

be capable of displacing the anionic phosphorus ester group.

This work is specifically aimed on approach (d) to develop

conditions and specific new compounds which will solve one major problem

related to organophosphate poisoning; effective reactivation of aged

blocked AChE.

At physiological pH hydroxamic acid and oximes are largely

ionized, and the anions are effective nucleophiles. In addition to the

pi of the reactivators, other physicochemical properties of the

reactivating agent, including steric and electronic factors and

lipophilic hydrophobic balance, can influence relative reactivating

ability.


1










Cationic Nucleophiles

It was shown by several authors 14 that the active forms of the

common nucleophilic reagents are the corresponding anions (hydroxamate,

oximate, imidazole anion, thiolate, alkoxide, etc.) which have increased

reactivity in cationic micelles. Within the past two decades, a number

of micellar systems such as I IQ have been investigated in which

functional groups have been incorporated into micelle forming molecules.

It was proved by Bodor et al.15-17 that a molecular combination of anionic

nucleophile and a cationic surfactant results in a highly effective

accelerator of hydrolysis of various organophosphates. Thus, the

micellar n-dcdeceyl-3-(hydroxyiminrmethyl)pyridinium salt (3) or the

corresponding 1-n-heptyl (or nonyl or hexadecyl) compounds (A) were

found to be much more effective nucleophilic reagents for the reaction

with well known organophosphates such as Paraoxon diethyll p-nitrophenyl

phosphate) and O-ethyl S-2-diisopropylaminoethylmethyl phosphonothiolate

(VX) than were nonmicellar analogs of l-alkyl-3-(hydroxyimino methyl)

pyridinium salts.

CH=NOH CH=NOH


N+ N+

(CH2)11CH3
a 4
X= I (a) R =C7H15 (a)
= I (b) = C9H9 (b)
= Br (C) = C16H33(C)
= CH3SO3 (d)
= CH3PhSO3 (e)








17

qEstein et al.15 have shown that 3(a) brought the half life of the

hydrolysis of paraoxon to 17.5 minutes from 10,500 minutes at slightly

basic pH (9.3). They also showed that hydrolysis of VX can be

significantly accelerated by 3. The various hamologs of 3 such as 4(a-c)

have also shown dramatic acceleration of hydrolysis of Paraoxon when

used above their critical micelle concentration (CMC).


C N-OH
R
HN N+HCI-

R = dodecyl

5


CH3

C16H33-N+-CH2CH2OH
CH3


CH3

nC16H33- N+-CHCH20H

CH2CH2OH


CH3(CH2)- N OH
CH3

R= CH3 or CH2N(CH3)2

6



C16H33-N+
N JNH


CH3
S, N'--CH2CH2OH

N H nC16H33










Kunitake et al.18 introduced a chymotrypsin model bifunctional

nucleophilic reagent containing an imidazole moiety substituted at the

4-position of the ring with an alkyl hydroxamic acid (5). In mildly

basic conditions this reagent caused dramatic, enzyme like acceleration

of hydrolysis of p-nitrophenyl acetate. It was also shown that

imidazole containing quaternized choline type analogues catalyze

phosphate hydrolysis more than does simple quaternized choline.

Hydroxamic acid micellar nucleophiles such as 6 have shown good

acceleration of hydrolysis of organophosphates at mild alkaline

conditions (as specified for oximes).19 Bodor and Kaminski17 as well as

Moss et al.m have found that N-hexadecyldimethylanmonium ethanol

bromide 7 accelerates micellar cleavage of organophosphates at basic pH

(10.5 to 12). Bifunctional analogues such as 9 showed7 reduced

acceleration of hydrolysis of organophosphate compared to monofunctional

7. This is understandable if we consider that anions are the reactive

species at alkaline pH; this would tend to retard reactivity due to

repulsion of anionic species in the case of 9.

In vitro work21 with a series of 2-PAM analogues at pH 8.0,

showed that the reactivity of the benzyl derivative (11(f)) is by far

the highest, while the dodecyl derivative (11(e) is most potent among

the alkyl derivatives against Tabun inhibited AChE.











R = CH3 (a)
= C2H5 (b)
SC3H7 (c)
N+ CH=NOH = i-C3H7 (d)
S= C12H25 (e)
R = benzyl (f)
11


CH3
N R = O(CH2)nCH3 (a)
CH=NOH= OCH2Pyridiniumaldoxime (b)
\ / = (CH2)nCH3 (C)
N = unsaturated chain (d)
CH2R
12


Very recently, Bedford et al.2 have shown that some 3-substituted

(hydroxyimino) imidazolium salts (2I(aL) were effective in providing in

vivo and in vitro reactivation of both Saman (GD) and ethyl p-

nitrophenyl methyl phosphonate inhibited AChE (non-aged enzyme). Their

findings suggest that the imidazolium ring system is a likely target for

structural "fine tuning" to improve activity of poisoned AChE in vivo.

Pretreatment of enzymes with quaternary ammonium compounds

including pyridinium oximes, has been shown12 to have a retarding effect

on enzyme aging due either to binding or to action on the acetylcholine

(cholinergic) receptors. For example, compound 13 and its derivatives

decrease aging of AChE by 4 fold.












CH3 N+ N -CH3

NH-C-NH
0
13





Sane quaternary amines which are allosteric effectors of AChE can

retard aging (due to their binding points on the enzyme, distant from

the active site), but this effect has no practical value as large

concentrations of these effectors have to be used.3 It is also found

that imidazole and alkyl substituted imidazoles retard aging, presumably

by binding to the anionic site on the enzyme. However, it might be

possible that the undissociated form of the oximes accelerates the aging

of the tabun inhibited AChE.

To date, no significant results have been reported for

reactivation of aged AChE. The hydrolysis accelerating effect of

quaternary ammonium compounds are marginal, a maximum 5-fold rate

enhancement being achieved. Monoanion phosphonate and phosphates are

about 5,000 times less susceptible to hydrolysis than are the triesters.

It is clear that the problem of reactivation of aged uninhibited

AChE cannot be solved until conditions for acceleration of hydrolysis of

the phosphate diester anions are found.

There are two things that an effective accelerator of phosphate

diester monoanions must accomplish:

a. partial or complete neutralization of the negative charge,








21

thus approaching the case of a phosphate triester

hydrolysis; and

b. acceleration of the hydrolysis by an effective nucleophilic

attack on the phosphorus.

Epstein et al.. postulated a "charge effect" theory, making

uncharged pyridine aldoxime a very ineffective nucleophilic reagent for

reactivation of inhibited AChE. It was also pointed out by Gray1 that

the electrostatic binding of the quaternary ammonium group to the enzyme

anionic site would have the quaternary ammonium group carrying the

nucleophile in a favorable position for reacting with the phosphorus

ester groups thus facilitating reactivation.




.H
O N=C

O i
EZ'
OR



Figure 1-6: Diester anion -2-PAM complex.



A 2-PAM model may bird to the aged AChE model by a possible strong

interaction between the positive charge of 2-PAM (anti) and the negative

charge of the phosphate (Figure 1-6). However, no reaction is possible

because of the reduced nucleophilicity of the oxime position due to the

loss of charge effect exerted by the pyridinium positive charge.










One conclusion emerges from these observations: the need for

"excessive positive charge" in the molecular combination for

reactivating phosphate diester anions. An advantage presented by such a

combination is that excessive positive charge should directly or

indirectly facilitate the nucleophilic attack of the nucleophile

(oximate, hydroxamate, etc.) without reducing its nuclephilicity. In

solutions, such a combination effect was achievedf5'1617 only in the case

of neutral triesters, by mixtures of positively charged micelles and

specific nucleophilic reagents. In vivo, however, a molecular

combination of the "excessive positive charge" and effective nucleophile

group, covalently bonded in the same molecule have to be used, primarily

due to the differential transport problems typical for complex in vivo

systems.

Specific Aims (Reactivators of Aged AChE)

This thesis suggests a novel and more flexible solution to the

problem of reactivating aged inhibited AChE: the use of type

nucleophilic reactivators having multiple positive charges for

acceleration of the hydrolysis of the phosphate diester monoanion (or

phosphonate monoanion). A systematic study of the reaction of properly

designed "excessively charged nucleophilic reagents" (ECNR) carrying

nucleophilic heads such as oximate or hydroxamate ions with simple

phosphate diester anions and diester anions simulating an aged

phosphorylated AChE will be used to evaluate the viability of this

approach. The basic concept of this study is then to synthesize charged

and multicharged functional nucleophilic reagents. For the in vitro

modeling, same of the compounds in addition will have surface active










properties and will simultaneously bind to the aged AChE, neutralize the

negative charge of the phosphate moiety and serve as nucleophilic

reagents for attacking the phosphorus. This will result in

dephosphorylation and thereby reactivation of the enzyme.

It is obvious that the success of the program is based on

execution of clear design and synthesis of effective nucleophiles, and a

detailed basic study of the kinetics and mechanism of the hydrolysis of

diester anions. Only after the problem of diester monoanion hydrolysis

is solved on the model compounds can one go to in vitro and then in vivo

reactivation studies. The study is also aimed at examining the

hydrolysis of neutral phosphate triesters so as to include the whole

spectrum of phosphate esters and the accelerating effect of the various

surfactant type nucleophilic reagents.

Design of Model Substrates and Reagents

A strong association between the nucleophile and the inhibited

enzyme is an important factor in enzyme reactivation. It appears from

all reactivators and antidotes so far studied (in vivo and in vitro),

that structural functions such as pyridinium and imidazolium rings are

involved in the binding of the nucleophile to the enzyme site that is in

close proxmity to the phosphorylated (inhibited) serine residue. It is

reported' that the positively charged ring anchors on to the negatively

charged carboxylate anion of the glutamic acid side chain at the active

site. Therefore the inclusion of either imidazolium or pyridinium ring

systems in the design of ECNR is highly desirable for structural "fine

tuning" to improve maximum activity.










Hydroxamic acid ECNR

It has been shown that micellar type hydroxamates are very effective

nucleophilic accelerators of organophosphates and carboxy ester

hydrolysis. Kunitake et al.m designed scme bifunctional surfactant

catalysts of the imidazole type (5). These cationic surface active

nucleophilic reagents effectively catalyzed hydrolysis of p-nitrophenyl

acetate and p-nitrophenyl hexanoates. On the other hand, Bodor, et

al.W found tertiary amine containing hydroxamates such as 5 to be very

effective accelerators of Paraoxon hydrolysis. It is noteworthy that

hydrmxylamines were used as antidotes for organophosphate poisoning

prior to the introduction of 2-PAM. In vitro, hydroxyl amines show high

nuclecphilicity since they are largely ionized at physiological pH

(7.4). Based on this information, it is suggested that a class of ECNR

can be designed and synthesized with the general structure as Figure 1-

7.

0
OH
_--_ N

N N-' 2X"
R, NR2 +
1 N Et
R3


R1 = H CH3 long alkyl group

Ra = CH3, C2H5 CH, C715 12H25, etc.

R3 = CH3, C2H5, or long alkyl group, etc.

X = halogen or CH3SO3 ,etc.

Figure 1-7: General structure of hydroxamic acid ECRs.








25

The hydroxamic acid ECNRs may be synthesized starting from N and O

protected hydroxylamine followed by careful N-alkylation1 to link the

imidazole moiety and the alkyl diethylamino group. Finally, alkylation

of the nitrogens with the appropriate alkyl halides to create the

positive segments and the selective deprotection of the benzyl group

either by catalytic hydrogenation or hydrolysis should give the target

hydroxamic acid ECNRs (Figure 1-8). Alternatively, the imidazole

containing acids can be replaced by pyridine carboxylic acids.

O
/OH

S2X X =Br; R =H; R2 = R3 = C12H25
N N, 2X +
R, R2 N Et
R3
X =Br; R,= H; R2 = R = CH3



ia
N .0-Block

SN-Et

NH O-Block
N N,,









O X NRCHO
N^N >N-E'







R,/ N N NH2OH

Figure 1-8: Retrosynthetic scheme for hydroxamic acid ECNRs.















Ns Ro N '- or H
N+ R,
I
H2 n R1 = CH3, C2H5, (CH2)11CH3, etc.
O=C.

ON R X= halogen or CH3SO3
R \ R






Figure 1-9: General structure of oxime ECNRs.

Oxime ECNRs

Pyridiniumoximes with the general structure as in Figure 1-9 suggested

as the second type of ECNR. The structural features of this ECNR were

essentially based on the oxime antidotes for organophosphate poisoning

such as 2-PAM (A) and Obidoxime (B). It was proven by Epstein et al.15

that 1-dodecyl and 1-heptyl-3-pyridiniumaldoximes are effective

accelerators of hydrolysis of same neutral organophosphates. Dejon and

Wolring19 found that N-long chain alkyl 2-pyridiniumaldoxime effectively

reactivated Tabun inhibited AChE in vivo. Further, quaternary

amonium compounds such as pyridinium oximes and pyridinium salts have

been shown to slow the rate of aging of Saman and Tabun inhibited

ACE12. Pyridinium-3-aldoximes usually are good in vitro nucleophilic

models for the hydrolytic studies of organophosphates and reactivation

of inhibited AChE, where as 4 and 2-oximes are generally the more potent

reactivators of AChE in vivo.














N --CH=NOH


CH2
O=C
O ,
(NR
R \ 1


SCH=NOH

N+
CH2
O-C


R \


N CH=NOY

HO
N* + N /
N
CH2 R


O=C
OR2






I- CH=NOH

N +


XCH2COOX


Figure 1-10: Retrosynthetic scheme for oxime ECNRs


OXIME

3

3


3

4


R




N'R,
-NR



H


Nt R


Ri

(CH2),,CH3


(CH2)3CH3


(CH2),,CH3


(CH2)11CH3








28

A logical retrosynthetic scheme for oxime ECNR is depicted in

Figure 1-10, starting from the benzyl ether of the corresponding oxime,

followed by N-alkylation of the pyridine moiety with either an a-

haloacid or an a-haloacid ester of the corresponding aminoalcohol. 1-

Carboxymetyl pyridinium oxime could be coupled with the aminoalcohol

moiety by several methods and quaternization of the amine moieties can

be achieved as the final steps. Alternatively, the pyridinium acetic

acid ester function could be replaced either by transesterification with

the appropriate aminoalcohol or with an amide function by reacting with

an aminoamine; R, and 1R can be either short, medium or long alkyl

chains, or combinations; the aminoalcohol may carry a mono or bis amine

function. Finally, deprotection of the O-benzyl oxime by catalytic

hydrogenation would yield the target ccmpounds. If deprotection by

hydrogenolysis presents any problem, alternative routes can be found,

such as trifluoroacetic acid hydrolysis.


OR
OR o o I

02 O- I0- O-PP --
II o


R = O-alkyl, O-aryl, etc. R and R, = OCH3, O(CH2)nCH3

= akyl, aryl = phenyl, benzyl, etc.


Aged enzyme models Tricsters




Figure 1-11: General structure of aged enzyme models and
organphosphorus substrates.










The use of various p-nitrophenyl alkyl phosphate or phosphonate

monoanions as aged enzyme models can simulate the environment of various

organophosphate inhibited aged AChE. Reaction of ECNR with these

monoanian substrates will be monitored by observing the extent of

product (p-nitrophenolate) formation.

The triesters (23-2J) can be obtained by literature methods 2*

from R-O-P(O)C1l. The diester monoanions (26-30) can be synthesized,

generally by dealkylation of methyl triesters or by a direct approach

(Figure 1-12). Diester anions, some carrying phospholipid analogues

instead of the p-nitrophenyl moiety, can be obtained by methods

described by Loew and Laceye. Strict safety measures must be exercised

in the handling of all organophosphate triesters carrying the p-

nitrophenyl group (analogues of Paraoxon). A decontamination

procedure2 that is used for Paraoxon would be used for the waste

disposal of these toxic triesters.

ECNRs with long chain alkyl groups (e.g., heptyl, octyl, dodecyl)

on the hydrophilic ammonium head groups will behave as cationic

nucleophilic surfactants. Rapid rate acceleration at concentrations

above their CMC may be attributed to micellar properties. The micelle-

substrate binding constants and electrostatic interaction between the

positive groups of micelle with the negative P-0 moiety at the Stern

layer9 should provide an excellent orientation for the nucleophile to

access the electrophilic phosphorus atom. This would demonstrate the

importance of both electrostatic and hydrophilic interaction. The nature

of the micelle surface in micelle-substrate binding as well as the

utility of this system as a good in vitro model for studying the







30

specific interactions involved in reagent to aged enzyme model binding
may thus be illustrated.


OR

02 O--P -- OX
II
0
X = lutidine, Na, etc.









OR
02 \O-- P -OR1
II
0 O


RO-P -CI
II
0


CI

02- -II
0


Figure 1-12: Retrosynthetic scheme for the phosphate substrates.


J/










Physical Chemical Studies of ECNR

Kinetics

The kinetic studies of the hydrolysis of diester monoanions and

neutral triesters by ECNR under various conditions would verify the

efficacy of the design concept and its potential application for the

reactivation of aged AChE. The activities of ECNR (-10-3 M) with

phosphate substrates would be made at pseudo first order conditions.

More detailed studies will be made under various conditions. These

will include: examining reaction rates below and above critical micelle

concentration (CMC), at various temperatures, at various pH's, in the

presence of co-surfactants and at various concentrations. The effect of

the ratio of reactants on the reaction rates should provide more

mechanistic details about the reaction. The study of hydrolysis of

neutral triesters by these ECNRs should clarify the accelerating effects

of multicationic nucleophiles.

The kinetic measurements of hydrolysis of phosphate substrates

carrying 4-nitrophenyl or 2,4-dinitrophenyl groups would be determined

spectrophotanetrically by following the rate of appearance of 4-

nitrophenolate or 2,4-dinitrophenolate ions by monitoring UV absorbance

at either 400 or 350 nm in slightly alkaline medium. It is noteworthy

that hydroxide catalysis under slightly alkaline pH is likely to be

negligible.14"15 Under these circumstances interpretation of the results

is less ambiguous.

Ionization constant

The effective nucleophile in the reaction is the anion of the

oxime or hydroxamic acid under consideration, which indicates the role








32

of ionization constants of ECNRs in the rate of hydrolysis. Thus, by

promoting more of the ionized species without reducing its intrinsic

reactivity, one could achieve a substantial rate enhancent over that

observed with less ionizable cationic nuclephiles. It also has been

reported1'2'3 that there is a pK requirement for optimal efficacy.

Oximes having pKa value <7.5 or >8.0 were found to be ineffective as

reactivators in vivo. Apparently, nucleophiles with high pKI would have

only a small amount of the anion available at physiological pH to act

as a nucleophile. On the other hand, if the pKI is too low the anion

will certainly be available, but will also be too weakly nucleophilic

for efficacy. So the pK values of the ECNR help to account for the

active nucleophile concentration at various pHs, and provide an indirect

assessment of their nucleophilicity in in vivo conditions. The pK

measurement of ECNRs can be made spectrophotametrically by methods

described by Epstein et al.15.

Critical micelle concentration

The CMC measurements of those ECNR carrying long hydrophobic

chains would be indicative of their micellar effect on the reaction

rate. A dramatic acceleration of hydrolysis will result as the

concentration of ECNR reaches the CMC2. In general, low values of CMC

parallel the enhanced reactivity of multicationic nucleophiles. These

CMC measurements can be made using either a spectroflurometer15 or

surface tension apparatus6.












Universal Decontaminant (UD)

Besides the struggle to find suitable antidotes for

organophosphate poisoning it is a challenge to chemists to devise

methods of destroying scme of the most noxious compounds known to man,

compounds which a more sane world would never produce. The structures

of several of these compounds are shown in Figure 1-1 and the

physiological effects of these compounds are cited in the beginning of

the Chapter. Another class of chemicals that have similar military

application as the organophosphates are the mustard agents.




CH3N (CH2CH2CI)2 CICH2CH2SCH2CH3

nitrogen mustard sulfur (half) mustard (35}



CICH2CH2SCH2CH2SCH2CH2CI

(HD)

Mustard agents are primarily vessicants, attacking eyes and lungs.

Blisters are formed by either liquid or vapor contact. Mustards are

insidious in their action; symptoms usually do not appear until several

hours after exposure. Although the sulfur and nitrogen mustards have

limited solubility in water at neutral pH, the small quantity which

dissolves is extremely reactive. The reaction proceeds via

intramolecular formation of cyclic sulfonium or imnmonium intermediates

as illustrated in equation 1-3.












CH2

CICH2CH2ZCH2CH2CI CICH2CHZ\ Cl
CH2 Eqn 1-3

Z=SorNR





This intermediate is highly electrophilic, and will attack

cpounpdsu containing a variety of functional groups such as primary,

secondary and tertiary amino nitrogen atams, carboxylate ions, and

sulfhydryl groups. With nitrogen mustards, the inmonium ion, which

apparently forms even in the absence of any solvent, readily attacks

another molecule to form a dimer. For this reason, the nitrogen

mustards are less stable than sulfur-mustard in long term storage. The

physiological effects of sulfur and nitrogen mustards are similar. They

act first as cell irritants and finally as a cell poison on all tissue

surfaces contacted. The local action of the mustards results in

conjunctivitis (inflammation of the eyes): erythema (redness of the

skin), which may be followed by blistering or ulceration; and

inflammatory reaction of the respiratory system. Injuries produced by

mustard heal much more slowly and are more liable to infection than

burns of similar intensity caused by physical means or by other

chemicals.

Although chemical agents such as organophosphorus nerve agents and

mustard agents have been in use for many years, only a handful of

reagents have been proposed, synthesized and tested as decontaminants








35

with same degree of success.3,31 Potentially significant applications of

such reagents and methods are in the cleanup of chemical spills such as

pesticides. In military circles applications include the

decontamination of equipment and personnel dermall application) that

have been exposed to chemical warfare agents, and disposal of chemical

weapons.

During the past several years there have been numerous

developments in this field. Decontamination of potential threat

materials by rapid hydrolysis has been the major focus of such work.

Because of the possibility of the use of nerve agents in military

situation, particular attention has been focused upon finding an

effective system to selectively hydrolyze organophosphorus compounds and

sulfur mustard agents without causing any serious human toxicity.

A large number of organophosphorus and mustard compouds contain a

labile halogen atom or a good leaving group such as halogen and an

electrophilic center (e.g., S, P=0) susceptible to attack by OH

resulting in displacement of the labile group.31b The net effect is

detoxification, since the reaction products are much less dangerous and

more readily disposed of safely.

Chloramine

Bleach hypochloritee) and organic bleach (N-haloamines) have been

commonly utilized as decontaminants in field conditions. One such

specially developed bleach is supertropical bleach ("STB') 3', which

contains about 30 % chlorine and chloramine (chloramine T, 36)J bleach.

"STB"3 is used in the form of a slurry. However, it is only about 50%

effective against organophosphates other than VX. This reagent suffers








36

problems such as instability in aqueous media, and is corrosive to

personnel and machinery. Chloramine T (N-bromo-N-

sodiotoluenesulfonamide hydrate), is a crystalline salt, a strong

oxidant in both acidic and alkaline media, and reacts readily with

mustard (CICHC) yS to give the relatively innocuous sulfimide

product. A mixture of a solution containing monoethanolamine,

hexylene glycol, histidine and chloramine T has been used as a non-

flammable, less corrosive and low toxic decontaminant for the disposal

of Soman (GD) and dichlorodimethylvinyl phosphate. Although chloramine

T is a strong oxidant, it is rapidly destroyed in alkaline solution.

Further, handling of chloramine T is hazardous because of the risk of

explosion, which constitutes a major limitation to its use in field

conditions.

Another type of reagent used by the military is agent "DS2"'13, a

general purpose decontaminant. It consists of diethylenetriamine,

ethylene glycol moncmethyl ether and 2 % sodium hydroxide. "DS2" reacts

with both the nerve agents and blister agents (mustard agents). Its

reaction is based on hydrolysis at the electrophilic centers of the

toxic reagents. The major limitations of the use of DS2 are: (a) Highly

toxic vapors, (b) It is a combustible liquid (possible hazard in field

conditions), (c) Protection must be worn during use, and (d) It is

corrosive to metals.

The decomposition rates of organophosphate, phosphonate and

phosphinate esters under various conditions are clearly of considerable

importance, and effective methods for their detoxification have

attracted the attention of numerous research groups over the past








37

several years.1535'36 A number of these studies were aimed at micellar

compounds bearing functional groups capable of catalyzing the hydrolysis

of organophosphorus compounds. Epstein et al.15 investigated the use of

various oxime type nucleophilic reagents, primarily the long N-alkyl

chain pyridinium oximes and obtained high micellar acceleration of

hydrolysis of Paraoxon and O-ethyl-S-2-diisopropylaminoethyl

methylphosphonothiolate. It has been shown by Bodor and co-workers18

that sane long chain N-alkylhydroxamic acids in the presence of

cetyltrimethylamnonium bromide (CIAB) significantly increase the

hydrolysis of Paraoxon.19 Same cupric ion complexes of N-alkylated

N,N,N-trimethylenediamine (""metallanicelles") were shown by Menger and

co-workers3 to have a huge rate of acceleration on the hydrolysis of p-

nitrophenyl diphenylphosphate (PNPDPP). Recently Katritzky and co-

workers6 developed a micellar system which consists of 2-iodoso or 2-

iodcbenzoic acid derivatives having long alkyl chains, and showed the

potency of this micellar system against some organophosphate triesters

in the presence of cetyltrimethylammonium chloride (CTAC). The

catalysis by the iodobenzoic acid has been speculated as being due to

the formation of anionic iodobenzoic acid species acting as the

nucleophile.

The effort under this heading of the thesis was directed towards

finding a universal decontaminant by a novel approach. This

decontaminant should effectively decontaminate toxic organophosphates

and mustard agents and itself revert to a non-toxic compound. In

addition, this reagent should suit field conditions by being non-

corrosive to machinery and personnel, be useful as a potential








38

prophylactic for use against agent exposure, easy to handle, non-

explosive, etc. A formulation of this reagent should also have a

potential for use in vivo dermall application). An effective

decontamination system such as this would be of substantial value for

both civilian and military use.

It is known that bleach (HOCd) reacts rapidly with

organophosphorus compounds and mustard agents. In both cases the

reactive species hypochlorite ion (OC1') is responsible. It has been

suggested2 that the rapid reaction with organophosphate is due to the

polarized nature of 0'-C1s, which is therefore attracted to the p= ',

group, so that a bifunctional attack is made.









Xa+............ O
11------OR
II OR
o--------'----- P

R R X = CI orBr






Hypochlorous acid is also known to take part in many other

reactions, such as oxidation of a alcohol to ketone aldehydee). In

alkaline medium OC1- ion is notably more nucleophilic than OH- ion, and

attacks electrophilic centers in the reaction medium. It is noteworthy








39

that HOC1 has a pK around 7.4; therefore in aqueous medium there will

be adequate nucleophile concentration (OX" ion) available for reaction.

Epstein et al.38a) showed hypochlorite ion as one of the most effective

catalysts for the hydrolysis of Sarin and O,O,S-

triethylphosphonothiolate under mild alkaline conditions. Epstein also

showed that these reactions involve a direct substitution of the

fluorine atom by hypochlorite anion. This process determines the rate

of overall reaction since it is followed by a rapid hydrolysis of the

intermediate. The effectiveness of OC1' over OCH was greater than could

be accounted for simply by the increased nucleophilic power bestowed by

the Cl upon the 0.



0_ 0
O II
II P
OC1- + i Pr-O F iPr-O F OCI
MOMe
OMe
Soman

O
o II
II P,
0OI OMe
OMe




Figure 1-13: Reaction of hypochlorite anion with Soman.

The mechanism of detoxification of mustard agent by hypcbraous

acid appearsm to be as described Figure 1-14. The oxidation of S to

S-O dramatically reduces its alkylating potency. Further hydrolysis of

the product results in caoplete destruction of the functional moieties.













/ / fast
HOBr + S BrS+ + HO f products




+ 1 fas-
B/ 1 BrS+ + 20H fast products
BrO" + SBrS 2-






Figure 1-14: Reaction of mustard agent with hypobrcmous acid.m



N-Haloamines have been used as positive halogen (X') sources in

various reactions. They have a labile N'-X& bond that produces reactive

O-X' (hypohalide ion) under alkaline conditions. Due to their increased

stability as positive halogen sources they are better decontaminant

formulations than is conventional bleach (e.g., NaOC1, or Ca(OCI)2).

Although N-haloamines have been used in the decontamination

process,3133 none have proven satisfactory as a safe universal

decontaminant for organophosphates and mustard agents. Consequently,

they have no practical value for in vivo or in vitro use.




N + OH- N + OX
X H
Eqn 1-4


X = CI or Br










N-Haloamines

It has been shown by Bodor et al." that some novel N-chloramines

are effective as a non-toxic and stable source of positive chlorine

(Cl) and same of these ccmpourds are effective as antimicrobial

agents. N-Brcmosuccinimide (NBS)37' 4 is known as a stable and versatile

"positive" bromine reagent among the number of the N-halo-cczpounds.

Only a few N-halo derivatives of amides and amino acids have been shown

to be useful "positive" halogen sources. This has been attributed to

the almost non-polar nature of their N-halogen bonds as well as the

favorable geometric arrangements which exist between the N-halogen bond

and the carbonyl functions. In general, the lower the polarity of the

N-X bond, the more stable the N-haloamine. The haloamine bond has

properties different from those of covalent C-X bonds. For many

properties, it is convenient to regard chlorine (or brcmine) when bonded

to a nitrogen as "positive source".


0 CH3- /CH3
C
N-X X-N COOR
Y X = CI, Br, I
S X BrCY = halogen, H
X = Br, CI
R = alkyl moiety

37 38

In the effort to find a stable and non-corrosive reagent Bodor et

al. designed and synthesized various chloroamides based on N-halogen

bonds of low "chlorine potential". The chlorine potential is defined as

pk~,, pk, = -log k, where k is the equilibrium constant41 for the

hydrolysis of N-chloro compounds to yield hypochlorous acid (Eqn 1-5).








42

Direct determination of the equilibrium constant (kI) values is only

possible in a few cases since the equilibria are usually so far to the

left that direct measurement is not practicable. In such cases it is

more convenient to measure the equilibrium constant, K. for chlorine

exchange reactions and relate the results directly to hypochlorous acid.

The K. value is then used as an experimental measure of the polarity of

the N-C1 bond. A high K. value is probably due to the polarization of

the N-Cl bond in the N-chloro derivative, and changes. This

polarization changes from N'----Cl+ for derivatives of strong bases to

N'*----Cl- for derivatives of weak bases.




N-CI + H20 N-H + HOCI
/ /
HOCI CI+ + OH"
Eqn 1-5

The former type of polarization would destabilize the molecule

with respect to separation of Cl1, whereas the latter type polarization

would stabilize the molecule. Bodor et al.42 studied the polarization

of N-C1 bonds of structurally different N-chloramines by photoelectron

spectroscopy and compared the polarization with the basicities of the

parent amines. Their investigation showed that the polarity of N-C1

bonds corresponds to the basicity of the parent amine and chlorine

potential of the N-chloramine.

A number of "low chlorine" potential N-chloramines and N-

chloramides containing low polar N-C1 bonds have been reported by

Kaminski et al.3,43 The chlorination of a-amino acids and the








43

corresponding N-chloro derivatives were investigated for use as low

chlorine potential ccmp~unds and positive chlorine. Their results

elucidated the factors that significantly influence the stability of

these N-chloramines and other low chlorine potential N-chloramines in

general. N-Chloramines of a-aminoacids, including those even the N-

methyl derivatives (39 and 4Q) containing a-hydrogens, decompose rapidly

via an elimination (dehydrchalogenation) pathway. This decomposition is

sometimes explosive (AQ). Decomposition pathways through

decarboxylation were also observed with carboxylates.4



O ,CI

CH3 -N-CH2CO2H CH3N-CH2CO2CH3
I I C/N



39 40 411
On the basis of these observations, 1,4-dichloro-,2,25,5-

tetramethyl-3,6-piperazinedione (41) was designed and studied.'3 The

incorporation of geminal dimethyl substitution at the a-carbon and a

cyclic amide (lactam) linkage resulted in greater stability. In

addition to the structural requirements, a synthetic amino acid was

chosen as the precursor, which should ideally exhibit an inherently low

toxicity. These types of low chlorine potential N-chloramines are of

interest as "soft" positive chlorine sources for use in vivo (eg., skin)

These N-chlorinated a-amino butyric acids and their derivatives

exhibited a high degree of antimicrobial activity by inhibiting the

bacterial growth due to inhibition ENA, RNA and protein synthesis










without readily being deactivated by extraneous reaction.4

Based on the low polarity of the nitrogen-chlorine bond and the

soft nature of the N-chloramines, the N-halo-4,4-dimethyl-2-

oxazolidinone system (43) was developed as a cyclic derivative of 2-

amino-2-methyl-l-propanol, a compound of low toxicity.44' The ideal

proximity between the amino and hydroxyl groups in 2-amino-2-methyl-l-

propanol lead to the investigation of the 2-oxazolidinone system as a

precursor for a low chlorine soft N-chloramine. This system contains

specific structural requirement, essential for optimizing the stability

and reactivity of the nitrogen-chlorine bond; that is, the gem-dimethyl

substitution at the carbon adjacent to the nitrogen-halogen bond.

R,


SX = CI, Br, H
0 NX R= R2= CH3

0
X = H (421
X = Br (4, NBO)



Bodor and Kaminski45 examined a series of N-bramo derivatives of

the 2-oxazolidinone system and tested their ability to serve as

versatile brominating agents with respect to N-bromosuccinimide

(NBS,37). Out of the many 4,4-dialkyl derivatives, 3-brcmo-4,4-

dimethyl-2-oxazolidinone (NBO, 43) proved to be by far the most stable.

The polarity of N-bromine bond in NBO is low, primarily as a result of

the electron donating inductive effect of the adjacent methyl groups,

and resonance effects of the oxygen atom adjacent to the carbonyl group.









45

NBO was also found to be equivalent or better than NBS as a braoinating

and oxidizing agent. The term "bromine potential" in analogy to the

"chlorine potential" was used to explain the low bromine nature of NBO

with respect to NBS. The bromine potential was defined as the

equilibrium constant (Ks,) for the reversible reaction between NBO and

NBS.

Bodor and Kaminski4 determined the equilibrium constant of this

process and related that to relative polarity of the N-Br bond with

respect to that of NBS. A value less than 1.00 is indicative of a low

polarity N-Br bond with respect to NBS. Low bromine compound such as

NBO showed a value k, = 0.34. Although NBO satisfies all criteria of

an effective soft low bromine potential N-brcmoamine, its low water

solubility (0.5%) seriously hinders its use in aqueous medium.

Specific Aims (N-Haloamine Decontaminants)

Non-corrosive N-bramoamines that are structurally similar to the

2-oxazolidinone system with additional features to increase aqueous

solubility without reducing the stability and reactivity would serve as

"soft" Universal Decontaminants for organophosphate and mustard agents.

Sane of the specific compounds should exhibit micellar properties and

same would be designed to yield increased active branine/per mole. In

addition to the chemical use of these compounds, some should possess

potential in vivo dermall) application. A detailed investigation based

on clear design and synthesis of N-bramoamines, examination of their

stability under various conditions, and kinetic study of hydrolysis of

organophosphate and mustard models is essential.










Design

2-Oxazolidinone decontaminant system

Based on available information it is envisaged that a careful

functionalization of 4,4-dimethyl-2-oxazolidinone should lead to the

"fine tuning" of the 2-oxazolidinone system for a Universal

Decontaminant (UD). Functionalizing one of the methyl groups of the gem

dimethyl a-carbon would not detract from the stability requirement of

having a dialkyl-substituted carbon a to the N-brcmine bond.


CH3 R X
CH2OR H H (44), Br(45)

O SO3H H (4),Br (Z)

0
CH2CO2H H (4.), BR(49)

(CH2),NEt2 H (5Q)

The electron donating inductive effect of the 4,4-dimethyl groups

in providing a less polar N-Br bond is unlikely to be altered by

hydroxymethylation of one of the methyl groups. However, the hydroxyl

group would certainly increase the polarity of the molecule. The

aqueous solubility of the 4-hydroxymethyl-4-methyl-2-oxazolidinone (44)

and its N-bramo derivative can be manipulated by derivatizing the

hydroxyl group with carefully selected functional groups. For example,

conversion of (44) to a sulfonate (46) should dramatically increase the

aqueous solubility of 46 and its N-brcamamine (47). A sodium salt of 47

would be an ideal candidate, with all requirements for a Universal

Decontaminant. Another "soft" analogue will be the dicarboxylic acid

ester of (48). The extension of chain length of 49 by more than about








47

six carbons would add lipophilicity to the molecule and thereby provide

a unique solubility property. The attachment of an alkylamine moiety to

the oxazolidione system was intended to provide a structural frame (50)

that can be transformed to various quaternary forms (51 and 52), and

same of the quaternary systems with long alkyl chains are expected to

show micellar properties.

The compatibility of various cationic surfactants (e.g., 1-n-

dodecyl-3-pyridiniumaldoxime chloride (2-PAD-C1), hexadecyltrimethyl

ammonium bromide (CTAB), 2-n-hexadecyldimethylanmonium-l-ethanol

bromide) with various N-haloamines and their activity in deactivating

chemical agents has been examined.
cationic ionic surfactants are compatible in a premade mixture.

However, this system significantly increased the reactivity of NBO

against chemical agents by formation of Br3" species, which was

subsequently captured by the cationic surfactant micelle. Based on

these findings a covalent attachment of dodecyl dialkylammonium ethanol

ether of 52 should be effective as a cationic micellar N-brcmoamine.



Br
I CH2O-(CH2)n -N Br
2 2 I R,

0 N X R


O R = (CH2)nCH3, CH3 X= H or Br
51-52 R, = CH2CH3 or CH3












tCH20-R

o Ny


O


SCH2OH

0 N,


0


I CH2OH
OH NH2


NH2 NH2

0


CHgOH +
OH NH2


Y = H or protective group

R = (CH2)nCOOH, C(O)O(CH2)nCOOH, P032-, SO32-, etc.


= (CH2)nNR2


Figure 1-15: Retrosynthetic
methyl-2-cxazolidinones.


scheme for the derivatives of 4-hydroxy-4-


LG LG
O










Increased active bromine/molecular weight

A somewhat related brominating agent which has been used in

several synthetic procedures with various degrees of success is 1,3-

dibraco-5,5-dimethylhydantoin47, which yields two positive bromine per

molecule. However, compound (53) suffers the major disadvantage of

containing the a-hydrogens, and hence contains N-bromo bonds of

relatively high polarity. Thus, compounds of the type 54 are the

objectives. A structure such as 54 will afford four active bromides per

molecule, thereby increasing the available reactive species per mole of

reagent by four fold.

0
R Br,1 N, Br

R

BrBr Br-N NBr
0 0


53 54A=)
It is necessary, from a stability standpoint, that there should be

no hydrogens on the carbon a-to the N-bramo band in order to prevent

dehydrdoalogenation. Rapid dehydrohalogenation from more than one part

of the molecule could result in an explosion. Once the active brcmine

content per mole has been increased, the next approach will be the

insertion of solubilizing groups such as hydroxymethyl, carboxylate and

sulfonate moieties on to one of the nitrogen of the glycouril system.

These substituents should dramatically increase the solubility of N-

bramo glycouril system in polar organic solvents and water whilst

retaining three active Bre per molecule.










Synthesis

The 2-oxazolidinone system can be prepared by methods similar to

the one used for protection of carboxyl functions in carboxylic acids.47

First, an appropriate ethanolamine with ideal proximity between amino

group and hydroxyl group has to be found. Condensation of 2-amino-2-

methyl-l,3-propanediol with urea has been reported to give(44) in

excellent yield.48 This hydroxymethylated oxazolidinone could be

converted to its brmno derivative. However, under N-bramination

conditions oxidation may occur at the hydroxymethyl site. Reacting (44)

with a pyridine-sulfur trioxide complex should yield the desired sulfate

derivative. The corresponding sodium salt of the sulfate may be obtained

either by treatment on an ion-exchange column (Na form) or by reacting

with one equivalent of sodium hydroxide.49

Carboxylic acid derivatives of 46 may be obtained by reaction with

a dicarboxylic acid anhydride (e.g., succinic anhydride) or by treatment

with manoacyl chlorides of dicarboxylic acids (e.g., malonyl chloride)

in a suitable solvent. Cyclic anhydrides with various chain lengths

would provide a series of carboxylate derivatives of 44 with various

solubility properties. Reaction of the bis sodium salt (or lithium

salt) of 44 with one equivalent of chloroethyl diethylamine under mild

conditions would provide the corresponding O-ethyl diethylamine

capound. Quaternization of this free amine with an alkylbrcmide such

as dodecyl bromide would yield the precursor for a micellar N-bramo

oxazolidinone system (52).









0




X N N x
0




-H
0






0
O










0 0
Y-R, R


+ electrophiles
(Pyr.S03, ICH2CO2R, CH20, etc.)

X = H or Y (Na, Li, etc.)











+ Urea (or N-substituted urea)


Figure 1-16: Retrosynthetic scheme for the dimethylglycouril systems.








52

Dimethylglycouril (3a, 7a-dimethyltetrahydroimidazo [4,5-d]

imidazole-2,5-dione) (54) and its analogues would be synthesized by

condensation of the appropriate a,a-diketone with urea.50 Using an

equimolar ratio of formaldehyde with (54) it should be possible to

prepare mono N-hydroxymethyl dimethylglycouril under controlled

conditions. Following the success in monchydroxymethylation of

glycouril further derivatization with sulfur trioxide-pyridine complex

should provide mono N-methylsulfate dimethylglycouril, which is expected

to have high water solubility.

The synthetic strategy involves preparation of the monoanion of

dimethylglycouril in situ either by reacting with sodium hydride (NaH)

or lithium diisopropylamide (IDA). Subsequent reaction with appropriate

electrophiles (e.g., XCH2CH2H, BrClO2Et, S0-pyridine) would provide a

flexible solution for the synthesis of monosubstituted glycourils. Use

of electrophiles that carry long alkyl chains would yield a product

that could have emulsifying properties. Methods for the N-bromination

of various amines, amides and amino acids are well illustrated by Bodor,

et al. 345 Generally bromine/aqueous sodium hydroxide solution is used

under cold conditions. Since all designed N-brnoamines are expected to

have lower reactivity (low brcmine potential) than N-brcmosuccinimide,

bramination of the N-H bond could possibly be achieved by reacting with

NBS. Most of the brominated amines are easily isolable as precipitates

from the reaction mixture at -5 to 0C or by solvent extraction.

However, highly water soluble N-brcmo products may be isolated from the

aqueous phase by freeze drying techniques lyophilizationn).








53

Physical Chemical Studies of N-Bromoamines

Bromine content

Active brcmine content of the N-branoamines would be analyzed

principally by an iodametric method using standardized sodium

thiosulfate solution.39'51 Active bromine content can also be determined

spectrophotmetrically by recording the UV absorbance at about 350 nm in

mildly basic solution.





N + OH-1 N + OBr Eqn 1-8
I H
Br



Stability studies

The stability measurement of N-brcmoamines at various conditions

is an integral part of the evaluation process. These measurements will

also render kinetic information about factors influencing the stability

of the compound. The effect of pH of the medium, concentration of the

compound, and temperature will be examined. The decomposition half life

of the compound can be measured by determining the positive bromine

content iodanetrically or spectrophotametrically over a period. The

spectrophotametric method involves monitoring the change of absorption

of the initial N-brcmoamine or the change in hypobromite ion (OBr")

production at 350 nm. It has been pointed out that kinetic rate

constants are independent of methods used to determine bromine

content.4








54

The decomposition of N-haloamines in aqueous solution generally

follows a first order kinetic process.9 Therefore, the apparent rate

constant can be characterized from the reaction half-life.

Characteristic UV absorption spectra would be used to determine product

composition and decomposition products. Stability studies will also be

extended to determine the compatibility of external surfactant N-

brmoamine mixtures.

Bromine potential

The "bromine potential", defined as the equilibrium constant (K)

would be determined spectrophotmnetrically from the reversible reaction

between N-bromoamine and succinimide. The value of K, will be

calculated by using a kinetic equation derived by Kaminski and Bodor45

from the absorbance data at a wavelength where the corresponding free

amine or NBS has no appreciable absorption. The observed equilibrium

constant will relate to the polarity of the N-bramine bond and to the

stability data.

Kinetics

The major objective of the effort in this research is to verify

the effective reactivity of each of the N-bramoamines synthesized or

formulations thereof against organophosphate and mustard model

ccpuTnds. The kinetic result obtained from these experiments should

correspond to the potency of novel N-bramoamines in decontaminating the

chemical agents. The model compound for a phosphate ester to be used in

the kinetic studies is diethyl p-nitrophenyl phosphate (Paraoxon).

Paraoxon is generally more stable towards hydrolysis than is the agent G

(GD (Scman)), while it is somewhat less stable than the agent V (VX).








55

However, due to the advantages offered by the ease with which its

cleavage can be followed by monitoring the appearance of the p-

nitrophenolate ion, it was preferred over the other possible model, DFP.

The model compound for the mustard agent (bis-(2-chloroethyl) -sulfide)

is the corresponding "half mustard" (2-chloroethyl-ethyl sulfide).

The destruction efficacy of N-bromoamines (-10.2 to 10'3 M) against

Paraoxon (-5x10l M) would be determined from the pseudo first order-

rate constant for the hydrolysis at various pHs, at different buffer

concentrations, and at different reagent concentration, at various

temperatures. A rational examination of the kinetic data with the

stability values will allow prediction of the optimum reaction

conditions needed for a maximum efficacy. The reaction rate constants

would be calculated either by half life method or by initial rate method

from the absorption data recorded over at least 4 half lives.

The kinetic studies of the destruction of 2-chloroethyl-

ethylsulfide by N-bromoamines will be made by assaying the mustard agent

concentration in the reaction mixture by a gas chrcmatographic method,

at different intervals. A G.C instrument with a flame ionization

detector will be utilized for such determination.


~














CHAPTER II
RESULTS AND DISCUSSIONS


Synthesis of ECNR and Phosphate Models

Synthesis of Hydroxamic Acid ECNR Compounds

A scheme (Figure 2-1) using the route executed by Kunitake, et

al.18 was initially outlined for the preparation of hydroxamic acid

target compounds. This route begins from benzyl benzohydroxamate, at N

and O protected (acylated) hydroxyl group, and requires N-monoalkylation

and subsequent deprotection of the N-acyl group in order to introduce

the acyl group of interest. Benzyl benzohydroxamate 56 was prepared

by the reaction of the potassium salt of benzohydroxamic acid 55 with

benzyl bromide in the presence of potassium carbonate in 60% yield. The

potassium salt of the hydroxamic acid was obtained from commercially

available benzohydroxamic acid using ethanolic potassium hydroxide.

Using a similar method to that of Kunitake et al.18,5, benzyl

benzohydroxamate was reacted with sodium hydride in DMF to generate the

corresponding anionic salt, and the subsequent attack with 2-

chloroethyl-l-diethylamine gave the N-alkylated benzyl benzohydroxamate.

Owing to the instability of the free amine form of the electrophile, 2-

chloroethyl-l-diethylamine was generated at mild conditions prior to its

use. The N-alkylated product (57) was obtained in widely varying yield

(30 60%), which similar to that observed by Kunitake et al. in same

N-alkylation (or acylation) reactions of benzyl benzohydroxamate.











(1) KOH/ EtOH (1) NaH /DMF or CHCI3
PhCONHOCH2Ph ,-
(2) BrCH2Ph (2) CICH2CH2NEt2


Kunitake's Method

aq. HCI hydrolysis\
\- ^


H-N-OCH
CH2
CH2
NEt2


2Ph


Ph-C-N -OCH2Ph
CH2
CH2
NEt2
57


O
II
". C-OH

NH N


(1) soci2
(2) BzBr


(1) SOC12
(2) BzBr


(1) H2, P
(2) SOCI


Minor product



X H2, Pd-C / H20: EtOH


R = CH2(CH2)o1CH3 (6Z)
SCH3 (f)


O
II
__ -C-CI

NH N+.HCI


d-C 4
2


Et3N (X'S)
CHCI3, 60C


0
O
C- N -OCH2Ph
+ OCH2
NH N CH2
NEt2


Major product 65


0
C-N-OH
CH2
3rH.N + N CH2
R N+Et2
R
R = CH2(CH2)10CH3 (14)
= CH3 15)


Figure 2-1: General synthetic scheme for hydroxamic acid ECNRs.


PhCONHOH
55








58

Kunitake's method 18,5 of removal of the benzoyl group from 57

under hydrolytic conditions to yield N- (p-diethylaminoethyl)-O-benzyl

hydroxyl amine (58) was not successful. This method of removal by

acidolysis under various hydrolytic conditions gave no trace of 58, but

gave a debenzylated product mixture which spectrally (IR, 'HNMR)

corresponded to the N- (3-diethylaminoethyl) benzohydroxamic acid

[FC(O)MNH (OH) Cl2N (ClHCI] It appears that under the hydrolytic

conditions employed the benzyl group is selectively removed from 57

instead of the benzoyl moiety, a result different from that observed 18

with N-dodecyl benzyl benzohydroxamate under similar conditions.

Acylation of 56 with imidazole propionoyl chloride prior to N-

alkylation with chloroethyl diethylamine was an ambitious attempt at not

having to go through 57 in the hydrolytic cleavage of the benzoyl group.

However, acylation of 56 using an adopted literature 18 procedure failed

to yield the corresponding N-acyl-0-benzyl benzchydroxamate, and instead

resulted in an complex product mixture.

A search for alternative routes4'55 to N-substituted hydroxylamines

that avoids hydrolytic reactions led to the finding of a direct and

simple approach (Figure 2-2) to prepare 58. The reaction involved a one

step alkylation of O-benzylhydroxylamine with 2-chloroethyldiethylamine

in the presence of a controlled amount of triethylamine to afford 58 in

moderate yield (= 60%). The compound 58 was isolated as the major

product from the reaction oil by vacuum distillation, and was

characterized by IH NMR and mass spectrum. A high boiling fraction that

was isolated from the reaction oil accounted for less than 5% of the

total yield and was identified as the dialkylated O-benzylhydroxylamine








59

(59) on the basis of its 1H NMR spectrum. Neither of these products (58

and 59) are reported in the literature. Wawzoneck and Kampt 5 have

found that N-alkylation of hydroxylamine always yields a mixture of N,N-

dialkylhydroxylamine and N-alkylhydroxylamine irrespective of the ratio

of the alkyl halide and hydroxylamine employed. The optimum conditions

worked out to provide increased yields of 58 were (a) use of not more

than 2.1 equivalent triethylamine, (b) use of 1.1 molar ratio O-

benzylhydroxylamine/2-chloroethyldiethylamine hydrochloride and (c) mild

reflux conditions.

Other routes to make 58 such as direct monoalkylation of free

hydroxylamine (HN2OH) were explored. Reaction of NH2OH with 2-

chloroethyldiethylamine gave a product mixture containing largely (>80%)

the dialkylated product. This was not totally unexpected, owing to the

unprotected nature of hydroxylamine and the high reactivity of the

electrophile.54 The synthesis of 0-benzyl-N- (-diethylaminoethyl)-3-(4-

imidazolyl)propenichydroxamate (65), the major synthon to the hydroxamic

acid target molecule, presented enormous difficulties. Addition of the

acyl chloride of urocanic acid (3-(4-imidazolyl)propenoic acid) to 58

in the presence of excess base (triethylamine or pyridine) under various

conditions gave a product mixture containing little or no 65. Use of

sodium hydride to generate the anion of 58 followed by the addition of

acylchloride 60 gave a non-olefinic compound which showed two p-

diethylaminoethyl moieties intact. It is not unreasonable to suppose

that the anionic 58 could add across the double bond of an a,B

unsaturated hydroxamate 5; this may well occur on formation of 65 in

the reaction medium.













(1) Et3N (2equ) H N-OCH2Ph
H N-OCH2Ph --
(2) CICH2CH2NEt2. HCI / CHCI3 CH9
H reflux, 4h CH,
NEt2 60%

+

Et2NCH2CH2-N -OCH2Ph
CH,
CH?
NEt2 <10%








Figure 2-2: Direct synthesis of N- (P-diethylaminoethyl) -O-benzyl
hydroxylamine.


After a series of unsuccessful attempts, a method utilizing

inverse addition of the 3-(4-imidazolyl)propionyl chloride to a well

stirred solution of 58 in the presence of ~ 10 molar excess of

triethylamine or pyridine gave the desired coupling to yield 65 along

with other minor components. The major component (Rf 0.34), the desired

product 65, was isolated after repeated flash chramatography on silica

gel in 15% yield. The structure of 65 was confirmed by mass spectral,

H NMR and C.H.N data. The 'H NMR (300 MHz) spectrum (Figure 2-3) shows

65 to be solely in one isomeric form, and trans identity was established

based on the trans coupling constant57 (J. = 15 HZ) of the alkenic








61

doublets. A component with lower R (0.18) than 65 was also isolated

as the side product during the chromatographic separation, and accounted

for about 5% of the crude yield. This side product was assigned

structure 66 on the basis of spectral data, (M+1) peak at the expected

value of 565. The NMR spectrum showed the absence of akenic proton

signals and accounted for two diethylaminoethyl moieties.

Mechanistically, addition of 58 to 65 in a Michael fashion should yield

66.

Rajendra and Miller6 have utilized Michael type addition of amino

groups to B, -unsaturated 0-acylhydroxamates to form B-lactams with a-

amino substituents. Allowing the reaction mixture to age over 48 hours

has been shown to increase the yield of side product (65) from 5% up to

about 10% of the total yield. On the other hand, use of short reaction

time (less than 4-5 hours) resulted in a low yield of 65 with incomplete

reaction.



NEt2
CH2
CH2
PhCH2O-N O
C N OCH2Ph
CH2
NH N CH2
NEt2




Micheal adduct 66









62

The N-alkylation of 65 with alkyl halides to give mono or

dialkylated derivatives was not as straightforward as expected. The

choice of solvent (or solvent system) was crucial for the completion of

reaction of 65 to its dialkyl alkylated derivative 67. In general, less

polar solvents are expected to favor N-alkylation of more basic nitrogen

(trialkyl amine) over less basic nitrogens aromaticc heterocycles).

However, non-polar solvents such as ethyl acetate and chloroform with

excess or one equivalent of dodecyl bromide gave a mixture of di- and

mono-dodecyl quaternary salts of 65. Reaction in chloroform-

acetonitrile 3:1 gave an amorphous solid of dialkylated 65 (67) in 63%

yield.

Since monoalkylation of 65 could not be achieved without having

same degree of dialkylation, attempts to make the monododecyl salt of 65

were reverted towards synthesis of the didodecyl salt. In a similar

attempt, treating 65 with excess of methyl iodide in chloroform-

acetonitrile 2:1 in a pressure bottle afforded 86% yield of the dimethyl

analogue of 65 (68). Purification of compounds 67 and 68 by solvent

recrystallization was not successful, and the cationic nature of these

compounds precluded any preparative chromatographic purifications. In

spite of these problems, the products were purified by a series of

solvent (petroleum ether, diethyl ether and ethyl acetate) washes. The

pure samples of 67 and 68 showed satisfactory C. H. N. analysis and dH

NMR.

A few options are cited in the literature for the removal of 0-

benzyl groups, for example, hydrogenation over Pd-SrCOO and acidolysis5

selectively remove 0-benzyl group without cleaving N-alkyl groups. A








63

trial hydrogenation of 67 over 2% Pd-SrCOD showed no sign of reaction.

Hydrogenation over 5% Pd-C in absolute ethanol over 3 days at 30 psi

hydrogen pressure gave only partial deprotection.

Finally, removal of the benzyl group and the reduction of the

double bond of 67 and 68 were accomplished successfully by hydrogenation

over 5% Pd-C in ethanol-water 6:8 over 12 hours to give the final

compounds hygroscopicc solids) 14 and 15. It is noteworthy that

simultaneous deprotection and reduction in the final step allowed the

elimination of three additional steps that were originally planned.

These additional steps would have involved conversion of 3-(4-

imidazolyl)propenoic acid (60) to 3-(4-imidazolyl)propionic acid (64)

instead of the direct use of acylchloride of 60 with 58 to yield 65.60

The mass spectral data of the final products were obtained using

fast-atam bombardment (FAB) in a glycerol matrix. The conventional type

of electron ionization methods (e.g., low eV EI) often fail to show

crucial mass spectral data for multiquaternary salts. The FAB

techniques have proven9 to be the method of choice for unveiling

molecular ion peaks of quaternary ammonium compounds. The mass spectrum

of 67 showed a strong peak (M) at the expected value of 680. However,

the mass spectrum of 68 showed no molecular ion peak but instead showed

a peak at 284 corresponding to M-NEEt2(Me). The hydroxamic acids 14 and

15 showed strong (M-l) peaks at the expected values 591 and 283

respectively, corresponding to the dialkylated products.
























































'-4l
cmJ
=a /

z
DoO
z
00
0o








z













O O
OH NOH
___ N __

HBr.N N
N.. (CH2)1CH3 N- NEt N Et
2 1 C 3 N + E t EC H
(CH2)1iCH3 3

14 15

Repeated washings and trituration with solvents of the oily

product yielded a pure sample of 14, which gave satisfactory C.H.N.

analysis. However, the elemental analysis data for the bis dimethyl

hydroxamic acid 15 varied over 1% of the expected values. The 'H NMR

spectrum of 15 showed four non-equivalent methyl peaks of which two

equal intensity singlets at 6 3.0 and 2.90 correspond to the methyl

group of the tetraalkylammonium moiety, and notably the spectrum showed

two single peaks for each imidazole hydrogen. This is an indication for

the existence of 15 in two different ionic (isameric) forms.

It is logical to envision the existence of 15 partly in a

zwitterionic form, where the anionic oxygen forms the counterionic

interaction with the quaternary ammonium moiety through a 5-member ring.

The low pK (-7.6) of the hydroxamic acids would allow part

deprotonation of 15 in neutral medium. Intramolecular 5-member ring type

interaction of anionic oxygen with the ammonium moiety has been

explained by Bodor et al.19 for the loss of reactivity (nucleophilicity)

of same B-amino hydroxamic acids. As a consequence one would expect the

reactivity of 15 to be different than that of 14. The physical chemical

studies of these hydroxamic acids are cited elsewhere in this Chapter.










0 Et Et
N./


H NN CH3 ON.-Et N/0

H Et N Et I
tI CH3-C=
CH3

Zwitterionic 15 B-Amino hydroxamic acid



Synthesis of Oxime ECNR Compounds

The most logical synthetic scheme for oxime ECNR ccipounds was

originally outlined (Figure 2-4) as preparation of l-carboxymethyl

pyridiniumaldoxime and subsequent tranesterification or esterification

with an appropriate amino alcohol followed by quaternization of these

esters with alkyl halides. The O-benzyl protected pyridinealdoximes

were synthesized from the corresponding free aldoxime in 50-60% yield

and were converted to 1-carbonylmethylpyridinium derivatives (71 to 73)

by reacting with the appropriate haloacetate in solvents such as

acetonitrile, ethanol and THF in a pressure bottle. However, under

these conditions the reaction of bramoacetic acid with O-benzyl-2-

aldoxime gave only 5% yield of O-benzyl-l-carboxymethylpyridinium-2-

aldoxime branide (71(b), whereas O-benzylpyridine-3-aldoxime gave an

80% yield of the corresponding carboxymethylpyridinium acid (72). This

explains the reduced nucleophilicity of O-benzylpyridine-2-aldoxime

relative to the O-benzylpyridine-3-aldoxime as the result of increased

delocalization through the 2-oxime moiety and the steric effect of the

benzyl group on the ring heteroatcm.














'- CHNOH () NaOMe/EOH
(2) BrBz


C- CHNOY



Y = Bz


:" CHNOBz

N*
CH2CO2R
71 to 24


transesterfication


esterification


R'OH Or R'OH


Et

Et



R -- NN


, >CHNOBZ

N+
CH2CO2R

RX R = ClIH2

= C4H, CH3, etc


Sdebenzylation




16 to.19


Figure 2-4: General synthetic schemes for oxime ECNR cnpounds.


BrCH2CO2R

R= H
= C2,H









68

Attempts at tranesterification of 1-ethoxy

carbonylmethylpyridinium-o-benzyl-2-aldoxime (1(b)) with 1,3-

bis(dimethylamine)-2-propanol or N,N-diethylethanolamine under various

conditions (ZnCl2 catalyst, p-toluene sulfonic acid etc), resulted in

formation of a salt of l-methyl-2-pyridiniumaldoxime (74) and

decompostion products as identified by H NMR. A probable mechanism is

outlined in Figure 2-5. Katritzky and co-workers61 have realized the

formation of l-methylpyridinium salts on heating t-butoxycarbonylmethyl

pyridinum salts and postulated a decarboxylation pathway followed by the

formation of 1-carboxymethylpyridinium betaine. In addition, the

basicity of the amino alcohols would assist the formation of a betaine

ester of(72(b)) by abstracting proton from the carboxymethyl moiety,

which can lead to the decomposition of 71(b).




Heat
CHNOBz CHNOBz
I Br- N --
CH2CO2Et N y-OH CH2CO2
7 betaine


-CO2





Z& CHNOBz L CHNOBz
N' XI
CH3 CH2
74

Figure 2-5: Proposed pathway of decarboxylation reactions of the
1-carboxymethylpyridinium-2-aldoxime









Alternatively, coupling of the aminoalcohol (1,3-

bis (dimethylamino) -2-propanol or N, N-diethylethanolamine) to the
pyridinium carboxylate moiety was attempted through esterification of 0-
benzyl-l-carboxymethylpyridiniumaldoximes (2- and 3-oximes). This
reaction was attempted in the presence of the condensing agent
dicyclahexylcarbodiimide (DCC) or diisopropylcarbodiimide (DIC) and a
catalyst, like 4-dimethylaminopyridine (EMAP) or hydroxybenzotriazole
(HOBT), these reactions lead to no trace of the expected product (25).
Instead a dicyclohexylurea (DCU) adduct2 of 12 was obtained. None of
the attempts, using different molar ratios of DCC/EMAP or DOC/pyridine,
and various conditions were found to carry the reaction beyond the
formation of the initial DCU adduct. The stable adduct was isolated by
column chrcnatography and found to have structure 76 on the basis of
mass spectrum (M: 477) and 'H NMR. Diisopropylcarbodiimide (DIC), often
used3 as the condensing agent in solid phase peptide synthesis, was
found to have little or no reactivity with l-carboxymethylpyridinium
salts (1-pyridinium carboxylic acids).


CHNOBz ,, CHNOBz


I I
CH2 CH2
C=O C=O

SC-NH\

S 00 C


Initial DaU adduct


N-Acyl urea (76)









70


/
N
HO /

\


RBr/EtOH
RBr OH R = dodecyl, butyl, methyl
reflux

/
N*-R


N'-N R




BrCH2COBr,





YON=CH-
N



,CH NOY R oxime compd
N dodecyl 3-oxime 16 (or 18)
butyl 3-oxime 17
R-N CH dodecyl 4-oxime 19
RN C=0 methyl 3-oxime 2Q
\ Y Y-H (W
R- N --' Bz ()


Figure 2-6 : Alternative synthetic scheme to pyridinium ECNR: Via
quaternization with bramoacylester of aminoalcohols.



Since all routes to 75 via esterification of carboxymethyl

pyridinium salts were unsuccessful, a direct pathway (Figure 2-6) to

target compounds which involved quarternization of pyridinealdoxime with











appropriate bramoacylesters was attempted. The success of this scheme

relied heavily on the preparation of the bromo ester of the amino

alcohol, particularly the 1, 3-bis(dimethylamino) -2-propanol.

The problem of self quaternization of free amino acylesters was

avoided by using N-alkylated aminoalcohols (79(a)-(Ic). Attempts to

acylate 79(b) and -79(c) with bramoacetyl bromide in chloroform in the

presence of a base such as pyridine, diiscpropyl ethylamine and

triethylamine gave a inseparable mixture containing the desired

bramester ('H NMR), salt of the base and unreacted alcohol.

While literature procedures on brono acylesters of aminoalcohols

and their quaternized derivatives are scanty, a convenient way to 80-81

was discovered. A reaction of neat brcmoacetyl bromide in the absence

of any solvent or base, with 1,3-bis(alkyldimethyl)-2-propanol at 400C

afforded the desired brcmoester in excellent yield (70-85%). The

driving force for the reaction is the elimination of HBr gas. These

branoesters were characterized from H NMR and mass spectral (FAB) data

and used in the next synthesis without further purification.

In contrast, 1,3-bis(trimethylanmonium)-2-propanol (79(a) failed

to give the corresponding branoester (80(a)) by direct acylation

methods. Attempts to prepare 80(a) using the activated ester methods

and mixed anhydride methods as well as the reaction with brcmoacetic

acid anhydride were unsuccessful. The reduced reactivity for acylation

appears be the result of insolubility of the quaternary alcohol 80(a) in

aprotic solvents.














/

HO f N
S /


RBr /EtOH, reflux
---------^
R =CH3.
= (CH2)3CH,
= (CH2),,CH3


Method A

(1) neat BrCH2COBr

(2) Na2CO3 (solid)


O *Et
II N"-R
BrCH2-C -O0 E
2 Et

R
81 dodecyl


0 N--R
BrCH2-C-O-
^-N+-R
N-
R
8loa methyl

a0BLb butyl

80Lc dodecyl


Figure 2.7 : Preparation of bromoacylester of quaternary aminoalcohols.



The final step in the preparation of oxime ECNR (16-19) involved

quaternization of the bulky bramoesters (80(b), 80(c) and 81) with

protected or unprotected pyridinealdoxime, which initially appeared

impractical but proved to be a facile and general method of preparation

of potentially diverse analogues of pyridinium-l-alkoxycarbonylmethyl

ccxpounds. The reaction of the brcmo acylester of ammonium alcohols

with pyridinealdoxime in 1:1 molar ratio in acetonitrile under positive

argon pressure afforded the target ECNR ccmpounds in 40-55% yield. It

is noteworthy that unsatisfactory (<10%) yields are obtained with


HO- N

N--N+-R


method B


BrCH2COBr /
base, CHC13


92a M)








73

solvents such as EMF and chloroform, and the increase in the reaction

temperature over 400C resulted in partly decomposed products.

The 'H NMR in IMSO-d6 of all ECNRs (16-19) showed the expected

pattern. A down field single peak (6 12.9 to 12.35) for the OH proton

and a single peak (- 6 8.4) for the CH=N proton were seen for each ECMR,

consistent with its existence in one isomeric form. On the basis of

the OH down field signal (6 12.9 to 12.35) the E configuration was

assigned to 16(bJ, 18(b) and 19(b) by analogy with the literature

result.5 The 'H NMR spectra in EMS-d of 16(b), 18(b) and 19() are

given in Figures 2-7, 2-8 and 2-9. Unexpectedly these spectra showed

deuterium exchange of the carbonylmethylene peak (- 6 6.0) on addition

of CaOD in all cases. The enol betaine behavior of the

carbonylmethylene group is the result of neighboring pyridinium and the

carbonyl groups.6 However, pyridiniumaldoximes of simple acylesters

such as 21 ( -octyloxycarbonylmethyl-3-hydroxyiinethylpyridinium

bromide) and 22 (l-dodecyloxycarbonylmethyl-3-

hydroxyiminamethylpyridinium bromide) showed no deuterium exchange of

carbonylmetylene protons. This suggest that the quaternary ammonium

moiety must influence the acidity of the carbonylmethlene group in

compounds 16 to 19.




























A _


U,






0a


I



: o






o1
V1
m
a








'&0


K'







V

0 f


mo V 0


























































zc


USL
-


kII a, oI
'I. *3


L0















L






El

















1-

.- I










































UZ I




II


F

















0




I-



N
r1



r











Further characterization of compounds 16 and 22 were obtained from

their elemental and mass spectral data. Some limitations6 to the use

of the FAB technique in gaining molecular weight information of the

pyridinium salts with long chain tri and dicationic derivatives 16, 18

and 19: the mass spectra showed pattern for intense fragmentation.

Attempts to convert the benzyloxime to the corresponding hydroxamic acid

by hydrogenation over Pd-C gave products that lacked the carboxy ester

portion. This indicates that selective removal of the benzyl group or

hydrogenation of the imine bond can not be performed with the activated

pyridinium ester systems like 16 to 22 without cleaving the ester bond.

Synthesis of Organophosphate Esters

The synthetic schemes for the synthesis of phosphorus esters that

are used as substrates to evaluate the hydrolytic potential of ECNR are

outlined in Figure 2-11. In general all phosphates were designed and

synthesized to carry a hydrolytically labile chro~mphore, such as 4-

nitrophenol or 2,4-dinitrophenol.

In general, organophosphate triesters that are analogues of nerve

agents are not available through commercial sources, and therefore uast

be prepared in the laboratory prior to use. There are references cited

in the literature for the synthesis of 4-nitrophenyl derivatives of

phosphoric acid triester. However, no references were found for the

synthesis of 4-nitrophenyl phosphoric acid esters carrying alkyl chains

larger than the butyl group. According to a method (Figure 2-11, Path

2) described by Loew and Jeffrey7 methyl dichlorophosphate and 4-

nitrophenyl dichlorophosphate was sequentially alkoxylated to give mixed

esters octyl methyl 4-nitrophenyl phosphate (2) and octyl bis(4-








78

nitrophenyl phosphate (24), respectively, in excellent yield. The use

of non polar solvent (ether) and hindered base (2,6-lutidine) are

reasoned to stop the reaction at the mono alkoxylation stage in the

first step and to inhibit nucleophilic displacement of alkoxy groups in

the second step. It is noteworthy that the insolubility problem in the

second step of the reaction was avoided by carrying out the addition of

alkoxide in portions in the dry box.

The triesters isolated were characterized by 'H NMR, UV and mass

spectra (EI), and showed only a trace amount of 4-nitrophenol impurity.

Since this impurity does not affect the kinetics of the triester

hydrolysis, it was decided to use these products for the hydrolytic

study without further purification.

Owing to the highly toxic cholinesterasee inhibitor) nature of the

products 22 and 23 very strict safety measures were assured for handling

and disposing these compounds and their synthetic intermediates. All

reaction residuals and glassware were carefully decontaminated with 2%

sodium hydroxide solution.

The organophosphate diesters are often prepared from

phosphotriesters.68, This method is routinely employed in the

oligonucleotide synthesis which involves dealkylation (demethylation)

and debenzylation of the triester. This conversion requires special

reagents (generally strong nucleophile) such as magnesium-

tetrahydrofuran ccmplexesm, thiophenol6 and lithium chloride1, and this

approach is distinctive and indirect. Only the mono nitrophenyl

phosphoric acid esters are prepared by direct phosphorylation of the

corresponding nitrophenol (or its salt) .n,














0
CH30-P-CI
CI



02 O-C-P- Cl

Path 1
Path 1


Alkoxylation

dry R1OH, lutidine,0C

Argon, non-polar solvent


Path 2


dry ROH, lutidine, 0C
non polar solvent


Cl
R-O-P-OR,
II
0


R2ONa, r.t
Et2O (THF)
(Dry box)


OR2
OO-P-OR1

0
Akyl 4-nitrophenyl triester


Hydrolysis




OsXe
O- -OR,
0
Diester


R1

methyl

octyl


4-PNP dodecyl

4-PNP octyl

4-PNP butyl

4-PNP ethyl

4-PNP benzyl

4-PNP 2,4-DNP
2,4-PNP 2,4-DNP


I

,' Dealkylation

S UCI / Acetone


Triester

R2
octyl

4-PNP


R3
4-PNPP

4-PNP


X = lutidine, H, Na, Li, etc.


Figure 2-11: General synthetic sequence for esters of 4-nitrophenyl
phosphoric acids.


Cl
I
R10-P-OR
O








80

Despite the potential use of 4-nitrophenyl phosphate diesters in

hydrolytic studies of phosphate diesterase enzymes, it is surprising

that there are only a few examples68 in the literature for the direct

preparation of these esters. In general, application of the triester

method to synthesize diester monoanion suffers from three major

setbacks: First, this method involves two or three alkoxylation steps

from the starting phosphorylating agent and in same cases poor yield

due to low percentage yields in the first alkoxylation step. Second,

most all 4-nitrophenyl triesters of phosphoric acids (e.g., Paraoxon)

are highly toxic nerve agents and, therefore, cause problems in the

handling and disposal of the compounds. Third, dealkylation (usually

demethylation) or debenzylation would not be selective with mixed

triesters carrying other labile groups and in same cases would require

use of moisture sensitive reagents.

Turner and Khoraa68 have described phosphorylation of thymidine-

3' to its triester and diester with 4-nitrophenyl phosphodichloridate

(82). However, it appears that direct use of 82 to make long chain or

other mixed 4-nitrophenyl diesters and their salts has not been explored

so far. The direct reaction of 82 with alcohols of interest under

conditions employed for triester synthesis gave the corresponding

monochloridate, which then on differential treatment either with

lutidine or acidic water gave the corresponding diester phosphate. The

4-nitrophenyl monochloridate was readily hydrolyzed with aqueous 2,6-

lutidine to provide the lutidine salt in 50-60% yield. The isolated

salt was found to be >99% pure by 1H NMR and UV. Increased solubility

of the lutidine salts of octyl and dodecyl derivative (26 and 27) in











organic solvents increased their yield slightly over butyl and benzyl

(28 and 29) counterparts. An attempt to alkoxylate 82 with 2,4-

dinitrophenol resulted in the isolation of the crude lutidine salt of

31, which was evident from the 'H NMR, but could not be purified.


OR
02 O-P-CI
0


OR
Lutidine / H20 OR
02 O-P -O-lutidine0
11
0


H+ (mild) / H20 /THF


exchange H+ form


OR
02 O-P -OH
0



Figure 2-12: Diester hydrogen phosphate synthesis.



The neutral hydrogen phosphate form of the diesters were obtained

either by eluting the lutidine salt through the acidic Amberlyst ion

exchange column or directly from the monochloridate by hydrolyzing in

mildly acidic water-THF followed by extraction with an organic solvent

(chloroform). In contrast to the lutidine salt, the neutral form of the

diester is generally a solid, which was purified by recrystallization in

mixed solvents to give pure hydrogen phosphate diesters (26 to 30). The








82

low yield obtained for the ethyl 4-nitrophenyl phosphate (0) is the

result of its increased aqueous solubility. All compounds were found to

be hygroscopic, and particularly, the dodecyl (26) and the octyl (27)

derivatives showed decomposition at 25 *C.

Overall, the direct method employed here for the synthesis of 4-

nitrophenyl phosphate diesters afforded a versatile, convenient and non

hazardous synthetic route to spectroscopically pure diesters. The mass

spectra (EI) gave the molecular ion (M') in all cases except the benzyl

derivative 30 (M-l peak). The C,H,N results were within accepted values

and in most cases accounted for the hydrated forms. The UV spectra of

diesters (26-32) showed X around 294 nm and in same cases absorption

maxima were seen at two wavelengths (274 and 294 nm).

The 'H NMR spectra of compounds 26 to 30 agreed well with the

assigned structures. Figure 2-13 shows a 'H NMR spectrum of n-butyl 4-

nitrophenyl hydrogen phosphate. The 3P H coupling caused a

distinctive double split of P-0-methylene group with a coupling constant

of J 14 Hz.67 The chemical shift of P-O-methylene hydrogens showed a

up field shift (6 4.40 to 4.05) on conversion of the

phosphamonochloridate to the hydrogen phosphate (or lutidine salt). The

3C NMR spectra of these diesters showed distinctive 31P-13C coupling of

P-O-C carbons.










83













1













r













Z, 0
E5


R. o

0






IC
0








84

Several attempts to synthesize bis(2,4-dinitrophenyl) phosphate

(3-) using a method described by Vilkas and coworkers7 were

unsuccessful. The product isolated upon reaction of phosphorous

oxychloride with excess of 2,4-dinitrophenol in 2,6-lutidine gave a salt

that accounted for only one nitrophenyl group ('H NMR) and less or no

33. This salt turned pink on treatment with alkali. Bunton and Farber74

suggested this to be due to the formation of an opened chain derivative

of the dinitrophenyl group. Similar results were encountered when

pyridine was used as the solvent/base. It seemed that the major problem

was in the insolubility of the 2,4-dinitrophenol in the reaction medium

pyridinee or 2,6-lutidine). Therefore, a modified reaction condition, a

hom geneous solution of 2,4-dinitrophenol (recrystallized) in

acetonitrile/2,6-lutidine was reacted with redistilled phosphorous

oxychloride for a short period which on hydrolysis provided the lutidine

salt of 13 in modest yield (40%). The melting point of the

recrystallized product agreed with the literature value.1

Synthesis of N-Bromo Compounds and their Derivatives

Oxazolidinone Systems

The 2-oxazolidinone ring is often used as a protective system for

&-hydroxyamino groups. Scme 2-oxazolidinones are physiologically

active75a) and have a number of industrial uses. Same have uses in

optical resolution processes of active B-adrenergic receptor blocking

agents (e.g., metoprolol75(b).

Preparation of 2-oxazolidinone usually involves reaction of the

appropriate amino alcohol with a carbonyl synthon such as diethyl

carbonate7 (Et2CD), acetone, carbonyl imidazole76 or urea "', or can be