THE BIOACTIVATION 0F 'SLAFRAMINE

Thesis for the Degree of M. S.
MICHIGAN STATE UNIVERSITY
THOMAS E. SPiKE
1969

w
—v———

LTBIMFLY

THES‘S
Michigan Scam
University

 

ABSTRACT
THE BIOACTIVATION OF SLAFRAMINE
By

Thomas E. Spike

Bioactivation of slaframine by rat liver microsomes has
been demonstrated by its ability to stimulate contraction of
the guinea pig ileum. The enzymatic bioactivation process
has been shown to require reduced nicotinamide dinucleotide
phosphate (NADPH) but not oxygen and is induced by pretreating
the animals with phenobarbital. The production of the active
factor has also been accomplished nonenzymatically with
various flavins in the presence of light. The active factor
causes a prolonged contraction of the ileum by acting directly
on the acetylcholine receptor with an apparent high affinity
for the receptor. Its action is prevented by prior applica-
tion of atropine, but is not reversed by atropine once
administered.

The metabolite has not been identified but its proper—
ties have been studied. Even though EPR has not shown the
presence of any free radicals, it is believed that the active
metabolite is either an N-oxide or loss of one electron by
the tertiary amine, both of which would give the nitrogen a
positive charge. The active compound has been shown to be

present at very low levels and is quite unstable as heat and

Thomas E. Spike

pH changes. The profound effects that are observed in animals
treated with the active compound are described. In addition,
some of the structural requirements for the observed response

have been determined.

THE BIOACTIVATION OF SLAFRAMINE

By

Thomas E. Spike

A THESIS

Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of

MASTER OF SCIENCE

Department of Biochemistry

1969

(,1 5.9735722
/c7 223 {0?

ACKNOWLEDGEMENTS

The author wishes to express appreciation to Dr. Steven
D. Aust under whose guidance this research was conducted.
The author also wishes to thank Dr. John E. Wilson, Dr.
Clarence H. Suelter, Patricia Darby, and the members of

Dr. Aust's laboratory for their suggestions and assistance.

11

TABLE OF CONTENTS

ACKNOWLEDGEMENTS

LIST OF TABLES

LIST OF FIGURES . . . . . . w. . . . . .
LIST OF ABBREVIATIONS . . . . . . . . .. .
INTRODUCTION AND REVIEW OF LITERATURE '. . . . .
METHODS . . . . . . . . . . . . . . .
Chemicals . . . . . . . . .
Gas-Liquid Chromatography . . . . . . . .
Thin-Layer Chromatography . . . . . . . .
Preparation of Slaframine-3H Acetate . . . . .
Animals . . . . . . . . . .
Preparation of Microsomes . . . . . . . .
Aminopyrine Demethylase Assay . . . . . . .
Hexobarbital Hydroxylation Assay . . . .
Guinea Pig Ileum Bioassay . . . . .
Partition Chromatography Assay for Acetate . . .
Perfusion Solutions . . . . _. . . . . .
RESULTS . . . . . . . . . . . . . . .

Attempts to Isolate a Metabolite of Slaframine

Metabolism of 3H(Acetate)-Slaframine by the
Perfused Rat Liver . . . .

Metabolism of Slaframine by Rat Liver Microsomes

Localization of the Enzyme Responsible for

the Metabolism of Slaframine . . . . . .
Stability of the Active Metabolite . . . .
Cofactor Requirements . . . . . .
The Non-Enzymatic Production of the

Slaframine Metabolite . . . .
Stability of the Metabolite Produced by

FMN and Light . . . . . . . . .

111

Page
ii

vi
viii

12

12
l6
16

18
25
25
26
27
32
33
35

37
37

142
A5

53
56
57
58

68

Page

Pharmacological Effects of Activated Slaframine . . 72
Biological Activity of Structurally

Related Analogues . . . . 75

EPR Studies of the Metabolite Produced by FMN . . 76

DISCUSSION . . . . . . . . . . . . . . . 77

SUMMARY . . . . . . . . . . . . . . . . 91

iv

LIST OF TABLES

Table

1. Thin Layer Chromatography of Slaframine
and Its Derivatives . . . . . . . .

2. Activation of Slaframine by Rat Liver

Fractions . . . . . . . . . . .
3. Effect of Various Cofactors on the Microsomal

Activation of Slaframine . . . . . .

Page

. l7

. . 56

. . 58

Figure
l.
2.
3.

10.

ll.

12.

13.

1“.

150

LIST OF FIGURES

Mass Spectrum of l-acetoxy-octahydroindolizine .
TLC of 3H-(acetate)-Slaframine . . . .
TLC of 3H-(acetate)-Slaframine . . . . . .

Activity vs Time for Aminopyrine Demethylase
Assay with Rat Liver Microsomes . . . .

Velocity vs Protein Concentration for
Aminopyrine Demethylase Assay by Rat
Liver Microsomes . . . . . . . . . .

Metabolism of Hexobarbital by the Perfused
Rat Liver O O O O O O O O O O

Metabolism of 3H-(acetate)- Slaframine by
the Perfused Rat Liver . . . . .

Metabolism of Slaframine by Microsomes . .

Dose vs Response for the Metabolite Produced
by Microsomes . . . . . . . .

Time Course Assay of the Production of
Metabolite by Rat Liver Microsomes . .

Ability of Various Subcellular Fractions
to Metabolize Slaframine . . . . .

Dependence of the Nonenzymatic Production
of the Metabolite with FAD upon Time . . .

Dependence of the Production of Metabolite
on the Concentration of FMN . . . . . . .

Dose versus Response for the Nonenzymatically
Produced Metabolite . . . . . . . . .

Reciprocal Plot of Dose vs Response for
Slaframine in Live Rats . . . . . . . .

vi

Page
1A
21
23

28

30

no

"3
us

“9

51

54

6O

62

6M

66

Figure
16.

17.

Decay of the Active Metabolite in the Dark

The Pharmacological Effects of the Active
Metabolite of Slaframine . . . . .

vii

Page
70

73

LIST OF ABBREVIATIONS

NADP = nicotinamide dinucleotide phosphate

NADPH = nicotinamide dinucleotide phosphate, reduced
NAD = nicotinamide dinucleotide

NADH = nicotinamide dinucleotide, reduced

FAD = flavin adenine dinucleotide

FMN = flavin mononucleotide

THF tetrahydrofolic acid

GLC

gas-liquid chromatography
TLC = thin-layer chromatography

DPEA 2,N-dichloro (6-pheny1phenoxy) ethylamine

QD = optical density

i.p. = intraperitoneally

i.v. intraveineously

viii

INTRODUCTION AND REVIEW OF LITERATURE

During the early 1950's, there were a number of reports
throughout the Midwest of excessiVe salivation by cattle
which had been fed certain legumes (l). Cattle eating these
forages would refuse further feed after one to three feedings.
This presented a significant economic problem due to the
decrease in production as well as the cost of replacing the
forages. In 1958, when the reports reaching the Agricultural
Experiment Station in Illinois became so numerous, Byers and
Broquist initiated a study of these "slobber forages" (2, 3).
The most severe outbreak of this problem was reported in 1959
in Missouri where animals displayed additional symptoms ’
including diarrhea, bloat, stiff Joints and, occasionally,
death (A).

Smalley 92 a1. (5) and Crump 33 a1. (6) observed that
the forages were usually red clover and, a1though they did not
visibly display any mold, proved upon microscopic examination
to be infested with the heavy mycelium of a dark brown-
colored fungus. They, furthermore, showed the fungus to be
Rhizoctonia leguminicola.

Aust gt a1. (7) were able to grow pure cultures of

Rhizoctonia leguminicola on cold water extracts of red clover

hay and showed the production of the salivation factor by the
fungus. Salivation factor activity was assayed by injecting
extracts of the mature mycelium into guinea pigs via intra-
peritoneal injections and following the degree of salivation
which ensued. They were successful in crystallizing the
picrate salt of this factor from ethanol and named it
"Slaframine." The structure and absolute stereochemistry
were determined from information obtained by nuclear magnetic
resonance, infrared, and mass spectroscopy. The alkaloid was
found to be l-acetoxy-6-aminoocta—hydroindolizine (8, 9).

The biosynthesis of Slaframine by Rhizoctonia legumini-
cola was studied with various luC-precursors (7). Only lysine
and serine were found to be incorporated. Portions of the
biosynthetic pathway have since been determined (10).

It has been suggested that cystic fibrosis syndrome may
be caused by nonfunctioning of cholinergic nerve fibers, since
the glandular sites involved are predominately supplied by
fibers coming from the parasympathetic nervous system. Clini-
cal symptoms of cystic fibrosis are malfunction of mucous—
secreting membranes resulting in irritation followed by a
toughening of these membranes to a fibrous tissue. Sufferers
of the disease lack sufficient pancreatic enzymes to utilize
dietary protein and, thus, suffer from malnutrition. The
exocrine glands are stimulated by cholinergic drugs, and the
use of these compounds has, therefore, been studied with

regard to therapy. However, the side effects of cholinergic

drugs makes their use almost impossible. The specificity
with which Slaframine stimulates exocrine gland function
without affecting other vital body functions, such as heart
rate (11), eliminates the problems associated with the use
of other parasympathomimetic agents for the treatment of
cystic fibrosis. The administration of Slaframine to animals
results in sustained secretory activity by exocrine glands
(12). For example, studies on the pancreas following admin-
istration of Slaframine have shown that Slaframine increases
the activity of the digestive enzymes and maintains a high
level of secretion for prolonged periods of time (11).

In all studies with the compound $2.21X2: there was
invariably a substantial delay before the onset of salivation,
and no activity was seen with $2.!ifii2 test systems (12).

This suggested that the compound might have to be metabolized
to an active form (12). Evidence that the liver was the site
of activation of Slaframine was obtained by injecting the com-
pound directly into the portal vein which resulted in a faster
response than when it was given into the inferior vena cava
(13). Furthermore, when Slaframine was given into the inferior
vena cava, all activity could be prevented by first ligating
the portal vein (13). Evidence that the bioactivation pro-
cess was being accomplished by the drug-metabolizing enzymes
of the liver was obtained by pretreating animals with com-
pounds which are known to induce drug-metabolizing activity

in liver microsomes and, thus, decrease the lag time.

Likewise, the delay could be increased by giving known
inhibitors of drug-metabolizing systems. Goats, which are
excellent metabolizers of xenobiotics, have a much shorter
period of delay prior to the onset of salivation than does a
calf which has much less ability to alter drugs (12, 13).

The biotransformation of most drugs occurs mainly in
the liver, but may also take place in plasma, kidney, and
other tissues. Drugs are eliminated from the body either
unchanged or as metabolites. Generally, the more polar com-
pounds are excreted unchanged. The less polar, lipid-soluble
compounds must be transformed before elimination can take
place. There are two main types of transformations accom-
plished in the liver, nonsynthetic and synthetic. {Nonsynthetic
reactions include oxidations, reductions, or hydrolysis and
may result in activation, change in activity, or inactivation
of the parent drug. Synthetic reactions involve conjugations
between the drug or its metabolite and an endogenous substrate
that is usually a carbohydrate, an amino acid, or derivatives
of these. Synthetic reactions almost invariably result in
inactivation and excretion of the parent drug.

These transformations have been shown to be carried out
by certain enzymes which are associated with the endoplasmic
reticulum of a number of tissues particularly the liver
(1A). This enzyme system has been classified as a mixed-
function oxidase system according to the terminology of Mason

(15) because of its requirement for NADPH and molecular oxygen.

The highest level of enzymic activity is found in the
lipoprotein membrane fragments, particularly those of the
smooth endoplasmic reticulum (16). Synthesis of the enzymes
appears to occur in rough reticulum, but this, when saturated
with enzyme, appears to lose its ribosomes to become smooth
reticulum (17). The amount of enzymes present can be increased
by chronic administration of various compounds. Recent studies
have shown that benzpyrene hydroxylase activity is increased
by phenothiazines and polycyclic hydrocarbons in organ and
tissue cultures, indicating the lack of hormonal control of
the induction process (18, 19). Many foreign compounds
stimulate their own metabolism of other drugs. Among these
are phenylbutazone, chlorcyclizine, probenecid, tOlbutamide,
hexobarbital, pentobarbital, phenobarbital, aminopyrine,
meprobamate, glutethimide, chlorpromazine, chlordiazepoxide,
DDT, methoxyflurane, 3,H-benzpyrene and 9,10-dimethyl-1,2- A
benzanthracene (20). Many chemicals in our environment also
stimulate the metabolism of drugs and other foreign substances.
These include insecticides, cigarette smoke, and some poly—
cyclic hydrocarbons found in polluted city air. Treatment of
animals with such inducing agents increases the apparent
concentration of the cytochrome P-USO, the oxygen-activating
component of the mixed-function oxidases (21, 22).

Just as the mixed-function oxidase system can be induced
by various compounds it can also be inhibited. Inhibition can

occur either competitively or noncompetitively. The microsomal

enzyme system is rather nonspecific, and frequently one drug
will competitively inhibit the metabolism of another which
bears no structural resemblance to the inhibitor (23). This
is difficult to understand in view of the common concept of
substrate specificity found in other enzyme systems. Many
drugs which are effective inhibitors in ig_xitgg studies

fail to have an inhibitory effect in vivg (2h). Conceivably,
the failure of these drugs to inhibit in X$XQ could be due to
failure of these drugs to reach effective concentrations at
the metabolic site (2N). Support to this idea has been
obtained by using an isolated liver perfusion system. Ethyl-
morphine, codeine, morphine, and diphenylphenoxyvalerate
(SKF-525A) will inhibit the metabolism of hexobarbital in the
perfused liver, whereas only the first two of these are effec-
tive in XlXQ inhibitors of its metabolism (2“).

Some compounds have been shown to preferentially
inhibit the metabolism of certain drugs while having little
or no effect on the metabolism of others (25). An example
of this is 2,N-dichloro(6-phenylphenoxy)ethylamine (DPEA).
Various steroids have also been shown to be alternative sub-
strates for a common microsomal mixed-function oxidase and,
thus, competitively inhibit the alteration of certain drugs
such as ethylmorphine and hexobarbital (26). They have
also been shown to be less potent inhibitors of
chlorpromazine oxidation and inhibition was not competitive

(27). Carbon monoxide, which binds to cytochrome P-hSO with

a greater affinity than molecular oxygen will inhibit, non-
competitively, the metabolism of those drugs which use P-hSO
in their terminal oxidation step (28).

The hepatic mixed-function oxidase system has proved
too labile for solubilization or separation of its components,
but a hemoprotein known as the CO-binding pigment or cyto—
chrome P—HSO (32) has been shown to be involved in the termi-
nal oxidation step of the metabolism of drugs. P-NSO was
discovered by Klingenberg (29) and Garfinkel (30) and was
partially characterized by Omura and Sato (31).

In its reduced form, cytochrome P-450 has an affinity
for carbon monoxide, although the normal ligand is oxygen
(1“, 32). P—U5O is found in the microsomes of liver and the
mitochondria of adrenal cortex, but brain and skeletal muscle
are devoid of this cytochrome (17). When foreign compounds
are added to hepatic microsomes, certain spectral changes
occur, indicating that the compounds have interacted with a
microsomal pigment, probably cytochrome P-NSO. Two types of
spectral change occur, one exemplified by that induced by
phenobarbital, aminopyrine, or the inhibitor, SKF-525A and
the other seen with such substrates as aniline or the inhib-
itor, DPEA (33). This suggests that there are two different
cytochromes with different substrate specificities and is in
agreement with an observation that pretreatment of rats with

methylcholanthrene produces a hepatic microsomal cytochrome

(Pl-NSC) which appears to contain only one of two components
of the normal cytochrome (P-ASO) (35).

A similar system, the steroid hydroxylating system of
beef adrenal cortex mitochondria, has been solubilized and
separated into a flavoprotein NADPH-diaphorase, a non-heme
iron protein, and cytochrome P-ASO (33). The electron trans-
port system involved in the liver microsomal system has con-
sequently been formulated (on the basis of what is known

about the adrenal system) as follows (3U):

Puso (Fe+2)-CO
co Tl by 02
Puso \N‘W

NADPH .pro-
>(:: :3 (1:8an) (me A) 131502 (eF +2)'o
pr
NADP+ . PA50.¢’//// RCH3
teinp (Fe+3)
RCH OH

2
fp = flavoprotein

The second area of interest in the bioactivation of
Slaframine is its site of action, the cholinergic nervous
system. The action of cholinergic, or parasympathomimetic,
drugs is to augment or duplicate the effects of stimulating
a parasympathetic nerve. This class of drugs consists of
those agents which directly stimulate effector cells, such
as pilocarpine, arecoline, muscarine, and certain choline

esters, and those which inhibit acetylcholinesterase and

thus permit the acetylcholine released to persist in its
action (eserine, neostigmine, and many others).

The relationship between chemical structure and biolo-
gical activity is particularly fascinating with regard to
cholinergic drugs. The chemical grouping which is common to
drugs having direct effect on cholinergic receptors consists
of a nitrogen atom to which three or four methyl groups are
attached. As in the case of ammonium or quaternary ammonium
ions, such a chemical grouping carries a net positive charge.
Because phosphorus and arsenic have nuclear properties which
are similar to nitrogen, the tetramethylphosphonium and
tetramethylarsonium ions also have cholinergic properties.

Choline is the simplest methonium compound which occurs
naturally and is a member of the B vitamin family. It can
act at cholinergic sites but very large, unphysiological
quantities are required. The acetate ester of choline is the
normal chemical mediator of cholinergic nerves. It has been
estimated that this substance is effective at cholinergic
receptors at concentrations as low as 10-9 molar. This makes
it one of the most potent physiological substances known.

Acetylcholine is rapidly hydrolyzed by acetylcholin-
esterase. The physiological role of acetylcholinesterase is
to terminate the transmitter action of acetylcholine at the
junctions of various cholinergic nerve endings with their
effector organs. Drugs that inhibit or inactivate acetylcho-

linesterase are called "anticholinesterases." They cause

10

acetylcholine to accumulate at cholinergic sites and, thus,
are capable of producing effects equivalent to continuous
stimulation of cholinergic fibers.

Physostigmine, also called "eserine" (an alkaloid
obtained from the seed of Physostigma venenosum) is perhaps
the most common anticholinesterase. Its pharmacoligical
properties were investigated by Christionson (36), Fraser
(37), and Argyll-Robertson (38). The elucidation of the
chemical basis of the activity of eserine was accomplished by
Stedman (39). Binding studies have shown that even though
eserine binds to the active site of acetylcholinesterase
reversibly, only a negligible amount of the inhibitor is
released from the enzyme due to the extremely slow hydrolysis
of its ester moiety by the cholinesterase (HO).

Atropine is‘a highly selective antagonist of cholinergic
agents on smooth and cardiac muscle and exocrine gland cells.
Th1; antagonism is so selective for cholinergic agents that
atropine blockade of the action of other types of drugs has
been taken as evidence that they act indirectly through
cholinergic mechanisms (H0).

Atropine, which binds reversibly with specific acetyl-
choline receptors, has a much higher affinity for the receptor
molecule than acetylcholine. Thus, the effects of acetyl-
choline can readily be reversed by atropine (40).

Studies on Slaframine thus far have indicated its

potential value as both a medicinal and research tool. The

ll

bioactivation of such an unusual compound is intriguing in
itself, but the understanding of the complete mechanism by
which it acts so selectively is the ultimate goal. The com-
prehension of the mode of action necessitates knowing what
the active species is and how to handle it. It was therefore
the intent of this thesis to isolate and identify the active
form of Slaframine and to determine some of its chemical
properties. Localization and characterization of the enzyme
system responsible for the bioactivation process were also

intended.

METHODS

Chemicals

Aminopyrine and hexobarbital were purchased from K and

. -... ,5

K Laboratories, Inc., Plainview, N. Y. Phenobarbital was
purchaSed from Merck and Co., Inc., Rahway, N. J. NADPH,
NADP+, FAD, FMN, riboflavin, NADH, NAD+, tetrahydrofolic
acid, biopterin, and folic acid were all purchased from Sigma g;
Co., St. Louis, Mo. Tritiated acetic anhydride was purchased

from Amersham/Searle Corporation, Des Plaines, Illinois.

Carbon monoxide was obtained from Matheson Co., Inc., Joliet,

Illinois. Acetylcholine iodide, eserine sulfate, pilocarpine

HCl, atropine sulfate (monohydrate) were all purchased from

Cal Bio Chem, Los Angeles, California. Slaframine was iso-

lated from Rhizoctoria leguminicola in our laboratory by

the method of Aust, gt al.(7). Slaframine was repeatedly

recrystallized as the dipicrate before converting to the

dicitrate for use. Conversion to the dicitrate was accom-

plished by dissolving the dipicrate in .01 N HCl and extracting

the picric acid out with diethyl ether. The pH was then raised

to 10 and the solution was extracted with two volumes of

purified chloroform. The chloroform extract was dried over

sodium sulfate and removed under vacuo. The residue was

12

13

dissolved in a minimum amount of dry diethyl ether and added
to a saturated solution of citric acid in diethyl ether.
The precipitate was washed with diethyl ether and dried under
vacuo. 8-Aminooctahydroindolizine and l—hydroxyoctatydro-
indolizine were compliments of Robert Gardener, Department
of Chemistry, University of Illinois, Urbana, Illinois.

l-Acetoxyoctahydroindolizine was synthesized by reflux-
ing 50 mg of l—hydroxyoctahydroindolizine in 10 ml of acetic
anhydride and 10 ul of pyridine for one hour. The reaction
mixture was then mixed with 50 ml of 0.1 N HCl and extracted
with an equal volume of chloroform. The pH was then raised
to 10 with sodium carbonate and the solution was extracted
twice with equal volumes of chloroform. The chloroform was
dried and removed under vacuo. Gas chromatography of the
product showed one peak and the structure of l-acetoxyoctahy-
droindolizine was confirmed by mass spectrometry (Figure l).

N-Acetylslaframine was synthesized by the following
method. The crude pH 10 chloroform extract from the isolation
of slaframine was dried over sodium sulfate and used as the
source of slaframine. An excess amount of acetic anhydride
was added, and the solution was refluxed for ten minutes. The
unreacted acetic anhydride was removed by distillation under
vacuum. Pure N-acetylslaframine was obtained by sublimation
of the residue.

N-Acety1-O—deacetyl slaframine was synthesized by heating a

solution of N-acetylslaframine in 2 N NaOH in a boiling water

1”

Figure 1.--Mass Spectrum of 1—acetoxy-octahydro-
indolizine. Mass Spectrum obtained on an LKB 9000
combination Gas Chromatograph mass Spectrometer of
1-acetoxyoctahydroindolizine dissolved in chloroform.
The column was packed with 2% OV-l and the column temp-
erature was 150°. The ion source temperature was 290°,
the filament current was 60 uamps, the electron energy
was 70 eV, and the accelerating voltage was 3500 volts.

15

oh“

omH

om

om

OH

 

ON

O¢

om

om

OOH

ZFIOHU

16

bath for ten minutes and extracting the product with chloro-
form. Deacetylslaframine was prepared by the same method

used to prepare N-acetyl—O-deacetyls1aframine.

Gas-liquid Chromatography

The majority of gas chromatography was done on a Barber
Colman Model 5000 Gas Chromatograph. Columns were six feet
long and had an inside diameter of 5mm. Column packings used
were OV-l and OV-l7 both 3%, pretested, and on a solid support
of chromsorb Q (100/120 mesh). Column temperatures used
ranged between 150° and 185°. At 185°, slaframine had a
retention time of 1.65 minutes. N-Acetylslaframine, N-acetyl—

O—deacetylsalframine, and deacetylslaframine had retention

times of 9.8 minutes, 6.3 minutes and 1.1 minutes, respectively.

Slaframine had a retention time of one minute on an OV—l

volumn at a column temperature of 175°.

Thin-Layer Chromatography
A11 thin—layer chromatography was done on precoated

Silica Gel F25“ plates obtained from Brinkmann Instruments,

Inc., Westbury, L. 1., N. Y. Samples were applied to the
plates in the basic form dissolved in chloroform. A list of

the various solvent systems and the observed RF values are

shown in Table l.

l7

.chozncHz no pcowmom m.mmnoocowmpa ssz oomoHo>oo one: mpoama

 

 

oonOhohn EchoEEw I <
HocmoonQI: I m
Hocmnpoe I z
EpOMOAOHno I 0
mm. m NH o
He. m oH o
mm. m 0H b
N». H oH o
m. H oH oH
ma. mm. no. mH. mo.o o N
am. mm. m s 0H
mm. mm. a m oH
mm. om. a m o
no. mH. H H
ocHEmpmmHm ocHEmnomHm ocHEmpMMHm ocHEmmmmHm <mmm <RNH m z o
IHmpoomooIo IHmpoomIz Iquoomoo
IHmpooMIz
m Eopmmm uco>Hom
3m

 

.mo>Hpm>HpoQ mpH one ocHEMLMMHm no anamamoumsonno poqu CHQBII.H mqm<9

18
Preparation of Slaframine-3H Acetate

Slaframine Free Base

A mixture of 310 mg of slaframine dipicrate, four ml
6 N hydrochloric acid and 16 ml water was stirred at room
temperature for two hours. The aqueous mixture was extracted
with diethyl ether until the ether extracts were cblorless.
The pH was adjusted to 10 with 10% sodium hydroxide and the
alkaline solution was extracted with three 20 m1 portions of
chloroform. The chloroform extracts were combined, dried,
and evaporated to yield a residue which was used directly in

the next step.

N-Carbobenzoxy-Slaframine ‘

A mixture of slaframine free base (from 200 mg of
slaframine dipicrate), carbobenzoxy chloride (60 mg), and
sodium carbonate (50 mg), in 20 ml water was prepared and
stirred at room temperature for three hours. A pH of 9 was
maintained by the addition of 2 M sodium carbonate. The
aqueous mixture was made acidic with l N hydrochloric acid
and then extracted with two 20 ml portions of benzene. The
aqueous layer was made basic with 10% sodium hydroxide and
extracted with five 20 m1 portions of chloroform. The com-
bined, dried chloroform extracts were evaporated to yield

81.6 mg of N-Carbobenzoxyslaframine as a viscous oil.

l9

N-Carbobenzoxy-O-Deacetyls1aframine

A mixture of 81 mg of N-Carbobenzoxyslaframine and 60
mg potassium carbonate in 10 m1 of methanol was stirred at
room temperature for seven hours. The methanol was removed
under vacuum and the residue treated with 15 m1 of chloroform.
A small amount of sodium sulfate was added to the chloroform
solution and the inorganic salts were removed by filtration.
The chloroform was evaporated to yield ”6 mg of a white
solid, which could be recrystallized from ether to give

needles. M.P. 156 - 157°.

Acetylation of Carbobenzoxy:O-deacetylslaframine

The acetylation process was carried out in vacuo (P<1u)
in the laboratory of Dr. R. Neistrom at the University of
Illinois. Twenty-nine mg of the alcohol was placed in a
break seal vial and connected to the vacuum line. Twenty-five
millicuries of tritiated acetic anhydride (specific activity
H000 mc/m mole) in a break seal vial was connected to the
vacuum line.

The tritiated acetic anhydride was transferred to the
carbobenzoxy-O-deacetyls1aframine which had been cooled to
77° K with liquid nitrogen.

The vial which contained the radioactive acetic anhy-
dride was removed and a second container, the contents of
which was 20 ul of acetic anhydride and two ul pyridine,

was connected to the vacuum line. The contents of this

u‘r

‘I'f-‘v -.f . .... .

2O

vessel were transferred to the 77° K vial containing the
alcohol and tritiated acetic anhydride. This reaction vessel
was sealed, removed from the vacuum line, and allowed to
react for 24 hours.

The break vial was broken and the acetic acid was
removed by vacuum using a potassium hydroxide trap. Benzene
(one ml) was added and the reaction mixture was 1yophilized
to dryness. The residue was dissolved in 10 ml water made
alkaline (pH 10) with sodium carbonate and extracted with
five 20 ml portions of chloroform. The combined, dried

chloroform extracts were evaporated until all chloroform had

been removed.

Removal of N—Carbobenzoxy group from
bL-Carbobenzgxyslaframine .

A mixture of N-Carbobenzoxyslaframine and hydrogen
taromide in glacial acetic acid were stirred at room temperature
fkor one hour. The mixture was poured into ice water and the
Eflfl was adjusted to 10 with solid sodium carbonate. The alka-
Iline solution was extracted with two 20 m1 portions of
cfluloroform. The compound was then converted to the dipicrate
Enid recrystallized to constant specific activity before con-
‘verting to the dicitrate for use. The resulting labeled
compound had a specific activity of 350 mC/m mole. Thin layer
(fluromatography employing two different solvent systems re-
Vealed only one peak upon counting 5mm sections which had

been scraped from the plate (Figures 2 and'3). The silica

-1 ‘1: —-——4—$Ia
.‘I

. ...‘.I---l -. _ --l .
o

21

Figure 2.——TLC of 3H-(acetate)-slaframine. The
slaframine was spotted on a TLC plate and developed in
a propanol:chloroform:ammonium hydroxide (6:7:0.05)
solvent system. The plate was scraped in 5mm sections
and counted in a scintillation counter.

 

22

 

 

 

 

 

_.

Amoco

Eau

23

Figure 3.--TLC of 3H-(acetate)-Slaframine. The
slaframine was spotted on a TLC plate and developed
in a chloroform: methanol (1:1) solvent system. The
plate was scraped in 5mm sections and counted in a
scintillation counter.

 

 

—-—. ‘- - _ u;

211

EU

 

 

 

EM... Samba
+ .12

o —x

I A menu
t 10¢

 

25

gel was suspended in 15 m1 of scintillator containing A1

2

w/w Cab-o-sil in toluene, 9.5% PPO, and 0.03% POPOP and

counted in a scintillation counter.

Animals
All rats used were of the Holtzman strain and purchased

from Spartan Research Animals, Inc., Haslett, Michigan.

I ’A‘i‘.’
I

Animals used as a source of microsomes were male rats weigh-
ing 300-“00 grams. Phenobarbital induction was accomplished

by including 0.1% phenobarbital in the drinking water; the

1"...4 .- .-._‘~. J

pH adjusted to 7 with sodium hydroxide. Animals used to
obtain livers for the liver perfusion experiments were 500—
600 g rats of either sex induced with phenobarbital. Blood
donors for the liver perfusion experiments were 500-600 grams
rats of either sex and blood was taken at two week intervals.

Blood was taken by heart puncture after ether anesthesia.

Preparation of Microsomes

The animals were exsanguished and the livers perfused
with 10 ml of ice cold 1.15% KCl via the portal vein. The
livers were removed and placed in 1.15% KCl on ice. The
tissue was blotted dry with filter paper, weighed, and minced
by chopping with a scissors. The minced tissue was homogenized
in four volumes of 1.15% KCl with five strokes in a Potter-
Elvehjen homogenizer equipped with a motor driven Teflon pestle
The homogenate was centrifuged at 10,000 g for 20 minutes and

the precipitate containing the nuclear and mitochondrial

26

fractions discarded. The microsomal fraction was isolated

as a pellet by centrifuging the 10,000 g supernatant fraction
at 105,000 g for 90 minutes in a Spinco Model L ultracentri-
fuge. The supernatant fraction was discarded and the .
microsomes were resuspended in 0.05 M Tris-H01 buffer (pH 7.5)
containing 50% glycerol. In experiments in which the rats
were not starved prior to being sacrificed, the microsomal
fraction was carefully separated from the glycogen by loosen-
ing the pellet in a small volume of buffer with a swirling
action. The protein concentration was assayed by the method
of Lowry (42). All operations were performed at 0-5°. The
microsomes were either used immediately or stored frozen under

N at -15°. These microsomes retained their full animopyrine

2
demethylase activity for several weeks providing they were

kept anaerobic.

Aminopyrine Demethylase Assay

CH CH

CH | 3 I 3
3 \ / N-CH3 CH3\ /N-H
l I microsomes ‘
./ N N
o ’\ / o//\ / + CH 0
N , N 2
NADPH
o

27

Reaction mixtures were incubated at 37° under air in
a Dubnoff metabolic shaker and contained microsomes (0.7
mg/ml), MgCl2 (7mM), NADPH (0.5mM), Tris-H01 (0.05 M pH 7.5)
and Aminopyrine (20mM)a

The N-demethylase activity was assayed by measuring the
rate at which formaldehyde was produced using the method of
Nash (A3). One ml aliquots were removed from the incubation
mixtures and diluted into one m1 of 10% trichloroacetic acid
(TCA). After allowing time for protein coagulation (about
five minutes) two m1 of Nash reagent (2 M NH “02H 30 2, 0.05 M
CH3C00H; 0.02 M 2,A-pentanedione) were added and the mixtures
were heated at 50° for 10 minutes. The assay mixtures were
centrifuged at 1000 g to Sediment the protein and the O.D.
of the supernatant fraction was determined at ”12 mp using a
Coleman Jr. Spectrophotometer equipped with a flow cell. The

-1 -l

extinction coefficient used was 7.08 O.D. ml of assay uM

of HCHO.‘ Linearity of activity with time (Figure A) and pro-

tein (Figure 5) was shown.

Hexobarbital Hydroxylation Assay

microsomes

,0 O

CH3 CH

()5?

3

‘Pons~.a_c .l.‘_ah -'- ..-.— . . .. 1

28

Figure H.--Activity vs Time for Aminopyrine
Demethylase Assay with Rat Liver Microsomes. The
metabolizing ability.of the microsomes used in all
experiments is shown. The reaction is linear for
approximately 15 minutes. See text for incubation
conditions. "

29

2......
us:

 

 

 

but-.lul.... .HI. .. RAUL. .IIIIILr

30

Figure 5.--Velocity vs Protein Concentration for
Aminopyrine Demethylase Assay by Rat Liver Microsomes.
See text for incubation conditions.

31

_E
\OE
H2525

 

 

 

:_E\¢<IE

32

The hydroxylation of hexobarbital by the perfused liver
was assayed by the method of Brodie and Cooper (AA). This .
assay involves the disappearance of substrate and is developed
on the principle that the more polar metabolite can be ex-
tracted out of a nonopolar solvent such as heptane into a
basic water solution, while the less polar parent drug

remains in the organic solvent.

‘. " T T' “2‘1

Two m1 samples of the perfusate were added directly to
one ml of 6 N HCl and extracted with five m1 heptane in a.

glass stoppered centrifuge tube. The samples were centrifuged

”IT—u““" ‘""-“"’““. "_"
E
A

in a clinical centrifuge at 3000 RPM for five minutes. A A
ml aliquot of the organic layer was extracted with two ml of
0.5 N NaOH and the optical density of the organic layer mea-
sured on a Beckman DB spectrophotometer at 2A0 mu. The

extinction coefficient used for Hexobarbital was lO-BM-lcm'l.

Guinea Pig Ileum Bioassay

About 250 g guinea pigs which had been starved approxi-
mately 2A hours, were sacrificed and the ileum was removed
immediately, washed, and placed in ice cold Tyrodes solution.
Pieces of mid-ileum about 3 cm long were suspended in a verti-
cal 2 x 15 cm organ bath. The organ baths were built into a
constant temperature bath maintained at 37°. The organ bath
was equipped with two way valve at the bottom such that the
sections of ileum could be washed quickly by draining and
refilling the bath with fresh Tyrodes solution pre-equilibrated

at 37°. The sections of ileum were anchored at one end and

33

fastened to a force-displacement transducer at the other end.
A small capillary tube bubbled air into the organ bath at all
times. The organ bath was always filled to the same level
which was measured to contain 20 ml.

Treatments were made by injecting the test substances

directly into the bathing medium via a syringe. The air

I-Grl
‘.

bubbling through the bath served to quickly mix the solutions.
Contractions were recorded through a force-displacement trans-

ducer on a Sargent model SRL recorder. I

Partition Chromatography
Assay for Acetate

1"“ “

Silicic acid (100 mesh) was purchased from Mallinckdrodt
Chemical Company. The fines were removed by making a slurry
in H20 and decanting off the material unsettled after three
minutes. The silicic acid was dried at 100° for 2A hours and
stored in a glass stoppered bottle until used. A one cm
diameter glass column filled With benzene was packed with 10
gm of silicic acid to which 6.5 m1 of 0.5 N H280“ had been
added.

Solvents were equilibrated with 0.5 N H280“ by shaking
in a separatory funnel and then passing the solvent through
Whatman #1 filter paper to remove excess sulfuric acid.

The sample (about one ml) was adjusted to pH 2 with
0.5 N H280“ and absorbed to l - 2 g of silicic acid, before
applying to the column. The elution pattern used was 50 ml

of benzene, 100 m1 of chloroform, and 150 ml chloroform

3A

containing 1% tertiary-butyl alcohol. Fractions of five ml

were collected from the column and titrated with 0.0163 N

alcoholic KOH to the phenolphthalein end point. The samples

were then evaporated to dryness and counted in a scintilla-

tion counter. The scintillator was dioxane containing 5%

naphthalene, 0.7% 2,5—diphenyloxazole (PPO), and 0.05% a.
l,A-bis [2-(5-phenyloxazolyl)J-Benzene (POPOP).

Precipitation of Protein with
Zinc Hydroxide

 

 

Since the lability of the metabolite was unknown, a

ring”... _:.u~,-~; II.

mild procedure for deproteinizing blood and microsomal
incubations was required. The procedure was essentially that
of Somogyi (A5) in which the protein_solution was diluted in
seven volumes of distilled water. One volume of a 10% solu-
tion of ZnSOuz7 H20 was added and the solution was thoroughly
mixed. One volume of 0.5 N NaOH was added slowly with contin-
uous mixing. The flocculant solution was then filtered
through Whatman #1 filter paper. The pH did not go below 5.0
and was titrated back to pH 7.0 or 8.0 with the 0.5 N NaOH.

A clear solution resulted which did not form emulsions when

extracted with chloroform.

Liver Perfusion Experiment

t Isolated rat liver perfusions were performed employing
a recirculating system similar to that of Van Harken 33 El-
(A6). Male rats were anesthetized with diethyl ether and the

liver exposed by an abdominal incision. The portal vein and

35

bile duct were cannulated, ligated and the liver carefully
removed and attached to the perfusion apparatus as quickly
as possible. The apparatus (A7) consisted of a stirred
blood reservoir connected to a peristaltic pump which cir—
culated the blood to an oxygenator. The blood then flowed
through the oxygenator, which had a mixture of oxygen: carbon
dioxide (95:1) passing through to the portal vein cannula.
The pressure head was regulated by adjusting the height of
the oxygenator above the liver. Excess blood flowed, via a
bypass, directly to the reservoir.

The perfusate was usually heparinized blood obtained
by heart puncture from blood donor rats. The blood was di-
luted 1:1 with 0.9% sodium chloride. The temperature of the
entire system was maintained at 37°. The perfusion system
was operated before attaching the liver for 30 minutes to
oxygenate the blood. Other perfusion solutions are listed

below.

Perfusion Solutions

 

Tyrode Solution

 

Solution A g/lO 1 EM

NaCl 80.0 138.00
Glucose 10.0 5.50
KCl 2.0 2.66
Ca012.2 H20 2.7 1.8A
Mg012.6 H20 1.0 .A9

36

 

Solution B
NaHCO3 10.0 19.00
NaH2POu 0.5 .36

These two solutions were kept at 0.A° until needed, at

which time they were combined in equal volumes.

Krebs-Bicarbonate Ringer Solution

 

 

 

51.1 MI.
NaCl 7.076 122.0
KCl .222 3.0
MgSou .1AA 1.2
CaCl2 .191 1.3
KH2POA .05A 0.A
Glucose 1.80 10.0
NaHCO3 2.10 25.0
Krebs-Henseleit Solution
Amount Order of Addition Solution M
770.0 (1/2) A/(1/2)7 0.90% NaCl 0.15A
30.8 1 1.15% KCl 0.15A
23.1 2 1.62% CaC12.2H20 0.110
7.7 5 2.11% KHZPOu' 0.15A
7.7 3 3.82% MgSOu.7H20 0.15A
161.7 6 0.15A

1.30% Ncho3

All solutions were made in triple distilled water and
kept at 0-A° until needed. The solutions were combined in

the order given to prevent precipitation of any salts.

RESULTS

Attempts to Isolate a
Metabolite of Slaframine

 

The following experiments were conducted in attempts
to observe a product of slaframine which might be, or be
derived from, an active metabolite. The first experiment
involved examination of the urine of animals given slaframine.
Urine was collected from control and slaframine (2 mg/kg, i.p.)
treated animals for A8 hours, adjusted to pH 10 with sodium
carbonate and extracted three times with two volumes of
chloroform. The extract was dried over sodium sulfate and
concentrated to 0.1 ml. It was then analyzed by thin-layer
chromatography (TLC) using Dragendorff's reagent to detect
tertiary amines. No metabolites of slaframine other than
deacetylslaframine could be detected by these methods even
when all extractions were done at A°. The experiment was

luC-(ring)-slaframine (1A mc/Mole), but, upon

repeated with
analysis by TLC and scintillation counting, all counts
appeared to correspond to slaframine.

Attempts to find a metabolite in the blood of animals

given slaframine were also unsuccessful. Blood samples were

taken at the peak of salivation by heart puncture. Protein

37

o _ .

"!‘;:‘—_‘—_f 7:?— “‘3” r‘ -

38

was precipitated by 10% TCA, saturating with ammonium sul-
fate, or with zinc hydroxide and the solution extracted at
pH 10 with chloroform. Analysis of the extract by TLC and
scintillation counting failed to show a metabolite of
slaframine.

Attempts to produce a metabolite of slaframine by
incubations with crude liver homogenates, liver slices, and
microsomes were not sucessful. The pH 10 chloroform
extracts of these incubations were analyzed by TLC and gas
chromatography without any evidence of a metabolite with
either radioactive slaframine or with unlabeled slaframine.

It was concluded that the metabolite was present at such
low levels that it could not be detected by these methods, or
that it was being destroyed by the methods used to precipitate
the protein or during extraction into chloroform at pH 10.

3H-(acetate)-s1a—

Therefore some very high specific activity
framine (350 mc/m mole) was administered (2 mg/kg, i.p.) to
rats and the blood and liver were removed and analyzed by
TLC without prior treatment, except for homogenization and
centrifugation of the liver protein. Five millimeter sections
were subsequently scraped, put in scintillation vials and
counted. These experiments also failed to reveal any radio-
active metabolites.

At this time it was decided to use the perfused liver in
an attempt to show a metabolite of slaframine. The ability of

livers to metabolize drugs was determined under various con-

ditions with hexobarbital as a model substrate.

in“;‘il . . "

39

Forty—seven m1 of whole blood was obtained by heart
puncture from five male rats (500 g). The whole blood was
diluted to 70 ml with 0.85% sodium chloride. A phenobarbital
induced, 500 g, male rat served as the liver donor (22.5 g).
The portal vein and bile duct were canulated and the liver
was connected to the perfusion system as described previously.

After a thirty minute equilibration period, 20 mg of
hexobarbital (dissolved in ten ml of 0.5 N sodium hydroxide
and neutralized with ten ml of 0.5 N hydrochloric acid) was
added to the perfusion system. Aliquots of two ml were taken
at 5,10,20,30, and A5 minutes and assayed for the disappearance »9
of hexobarbital by the procedure described in Methods.

In one experiment the bile duct canula was removed and
the experiment was repeated with a second addition of 20 mg
of hexobarbital. At the end of two hours the bile duct was
recanulated and Krebs—Hanseleit solution was substituted for
blood in the perfusion system. A fresh liver was also perfused
with Krebs-Hanseleit solution to eliminate the possibility
that the liver had degenerated in the first experiment and
thus account for the observed results. The results of this
experiment were, however, the same as the first, showing that
the metabolizing ability of the liver decreased when Krebs-

Hanseleit solution was used in the place of blood in the per-
fusion fluid. This is probably because Krebs-Hanseleit
solution is less able to carry oxygen to the cells. The

results of these experiments are shown in Figure 6. Curve A

I WIN, AFFILIII LEI-”w"

A0

Figure 6.-—Metabolism of Hexobarbital by the
Perfused Rat Liver. The effects of various perfusion
fluids on the ability of the perfused liver to metabolize
hexobarbital. A- Krebs-Habseleit Solution; B- Blood
diluted 1:1 with saline, bile duct ligated; C- Blood
diluted 1:1 with saline, bile duct cannulated. The amount
of hexobarbital remaining in the perfusate was measured
and plotted as the percent of the amount present at five
minutes on semilogarithmic paper.

A1

)3
I.

 

Eco

 

ms. _ .—
ON— 0— p 00— 00 on OK 00 on O? 00 ON 0—
. _ — _ — _ _ H — _ a _
I U
P..././
.I I .x I
U C . cl... . I. I1
IIIIIIIII o. ’
n I I ./ H
< I I.
.....i
...
O ’0 J
1..

 

O—

 

OO—

0/0 1v1IeIIvaoxaI-I

A2

was obtained with Krebs-Hanseleit solution as the perfusate.
Curve B was obtained with the bile duct ligated while Curve

C was obtained with the bile duct canulated using diluted
blood as the perfusate. The liver, perfused with blood, was
able to remove roughly half of the hexobarbital from the per-

fusate in 10 minutes, while roughly one hour was required to

 

remove the same amount of hexobarbital from the Krebs-Hanseleit 1
I
perfusate. ;
T
I
Metabolism of 3H(Acetate)—Slaframine *

by the Perfused Rat Liver

 

‘P‘Aaa- -.; A.

From the results of the previous experiment it was
decided to use a liver perfused with blood without the bile
duct cannulated for the following experiments. After a 30
minute equilibration period 1 mM of the dicitrate salt of
3H-(acetate) slaframine (specific activity 350 mc/m mole)
was added to the perfusion system. The total volume of the
perfusate was 75 ml. One ml aliquots were removed at 2,10,
20,30,A0,60,90, and 120 minutes and placed on ice. The pH
of the blood was raised to 10 with solid sodium carbonate and
extracted with 10 ml of chloroform. Ten ul aliquots of the
chloroform fractions were transferred to scintillation vials
and dried under nitrogen. One-hundred ul aliquots of the
water fractions were added to scintillation vials. All sam-
ples were counted in 15 m1 of scintillation fluid which
contained 5% naphthalene, 0.7% PPO, and 0.05% POPOP in
dioxane. An increase in the fraction of total counts occurred

in the water fraction with time (Figure 7).

Figure 7.--Metabolism of 3H-(acetate)—Slaframine
by the Perfused Rat Liver. One ml samples were with-
drawn from the perfusion fluid reservoir at various
time intervals. The pH was raised to 10 with solid
sodium carbonate and the solution extracted with 10 ml
of chloroform. Aliquots of the water and chloroform
fractions were then counted by liquid scintillation
( , water fraction; ——I , chloroform
fraction).

 

 

 

AA

ON— oo—
_ _

LA

ow

on

 

 

On

00—

 

thDOU
._<._.O._. X.

A5

The radioactivity remaining in the water fraction was
shown, upon subsequent analysis, to correspond to acetic
acid by chromatography on silicic acid (see Methods).

The radioactivity in the chloroform extract was shown
to be attributable solely to the slaframine by TLC developed
with Dragendorff's reagent. The presence of deacetylslafra-
mine in the chloroform could not be demonstrated since the
amount present was too small to detect visually following
spraying with Dragendorff's reagent or ninhydrin.

It was realized at this time that perhaps there was no
metabolite distinguishable by chemical methods, but that one
might be seen physiologically. However, an extremely sensi-
tive bioassay assay would be necessary. The guinea pig ileum,
which is a very sensitive and well known bioassay for
cholinergic drugs was used to search for physiologically
active metabolite of slaframine. As mentioned above, the
pharmacological action of slaframine indicated it was acting
through the cholinergic nervous system.

Metabolism of Slaframine by
Rat Liver Microsomes

Portions of a microsomal incubation mixture containing
microsomes, slaframine and NADPH in Tyrodes solution were
added to the organ bath containing the ileum. The complete
incubation mixture caused a marked contraction of the ileum
(Figure 8,A) which persisted much longer (about five minutes)
than ones initiated by acetylcholine and could not be reversed

by washing.

A6

Figure 8.--Metabolism of Slaframine by Microsomes.
Microsomal incubations were tested for their ability to
contraction of the guinea pig ileum. A, Complete incu-
bation mixture; B, minus slaframine; C, minus slaframine
and microsomes; D, minus NADPH; E, slaframine alone;

F, Tyrodes only. Upper arrows indicate washings and
lower arrows additions.

A7

Fifi... .I..iI.I IIII-.v..-..I..I. 51E.

Tlfiqdlll 3203:

TIME

 

5 min

A8

Various controls were run to gain confirmation that the
observed reaction was attributable to the active metabolite
of slaframine and required microsomes and NADPH. Figure 8,B
displays the results of an injection of the incubation mixture
minus slaframine. NADPH and Tyrodes solution gave similar
results (Figure 8,C). A requirement for NADPH is shown in
Figure 8,D while slaframine alone (dissolved in Tyrodes
solution) failed to elicit a response in the ileum (Figure
8,B). Figure 8,F displays the lack of reaction by adding
solely Tyrodes solution.

Because of variations in ileum responses all assays
were corrected by comparing the response to acetylcholine.

By this method a dose response curve for the metabolite pro—
duced by liver microsomes could be obtained (Figure 9).

The production of the metabolite was shown to be a
function of time removing one ml aliquots from an incubation
at minute intervals and placing them on ice until assayed with
the ileum. Keeping the samples on ice was necessitated by
the fact that the metabolite causes a prolonged contraction
of the ileum and subsequent assays must await the return of
the ileum to a resting state. In addition acetylcholine had
to be given to test the sensitivity of the ileum between
each assay. In each experiment there seemed to be some of
the metabolite present at zero time. This phenomenon will be
explained in a later section. The reaction did seem to be
linear for about five minutes (Figure 10) after which the

amount of the metabolite present began to decrease. A

A9

Figure 9.-—Dose vs Response for the Metabolite
Produced by Microsomes. Standard incubation mixtures
were incubated for three minutes at 37° and various
volumes of the mixture were injected into the organ
bath. Results were recorded as percent of the response
to 0.1 pg of acetylcholine.

5O

 

 

 

 

14H 1 I
e e e s

asuoasau SNI'IOHD'IAHDV :IO %

1.0

DOSE

(ml)

51

Figure 10.—-Time Course Assay of the Production
of Metabolite by Rat Liver Microsomes. Standard
incubation mixtures (Methods) were incubated for various
time intervals and 0.5 ml of the mixture were introduced
into the ileum chamber. Results were recorded as percent
of the response realized by 0.1 ug of acetylcholine.

52

 

 

-0
V
0 ~53
o
0 ~51
o
_o
F
o
l 1
.. §

JSNOJSJU 3NI'IOHD'IAHDV 1° °/°

TIME

reg-33:. .

53

definite increase in activity was realized when incubation
mixtures of phenobarbital induced microsomes were compared to

control microsomes.

Localization of the Enzyme Responsible
for the Metabolism of Slaframine

 

The livers (28.7 g) from two female rats (A00 g each)
were homogenized as described in Methods. Slow speed centri-
fugation (1000g) was used to remove the nuclear fraction and
cellular debris. The resulting supernatant fraction was used
as a crude homogenate and assayed for its ability to metabolize
slaframine to an active form capable of contracting the guinea
pig ileum. The mitochondrial fraction was prepared by centri-
fuging the 1000g supernatant fraction at 10,000g for 20
minutes. The mitochondrial fraction was washed in Tyrodes
solution, recentrifuged and then resuspended in Tyrodes for
use in the incubations. Further subcellular fractionation
was then accomplished as described in Methods for isolation
of microsomes. Protein concentration and metabolizing ability
were assayed on each resulting fraction. The incubation mix-
ture consisted of one ml of the liver fraction and the incu-
bation mixture used for the aminopyrine demethylation assay
except that slaframine (0.2 mg) replaced aminopyrine.

Activity (Figure 11) was recorded as response obtained from
injecting 0.2 ml of the incubation mixture into the ileum
bath, compared to the response obtained from 0.1 ug of acetyl-
choline just previously injected, per milligram of protein in

the incubation mixture (Table 2).

5A

Figure ll.-—Ability of Various Subcellular
Fractions to Metabolize Slaframine. NADPH (0.A mM)
and slaframine (0.2mM) were added to all incubations.
One ml of each fraction was used as the protein source.
A, crude homogenate (1000g supernatant fraction);

B, 10,000g supernatant fraction; C, mitochondrial
fraction (10,000g pellet); D, microsomal fraction
(105,000g pellet). Upper arrows indicate washings
and lower arrows additions. Time marks are 5 minutes.

55

 

H _
J02

.._u< a t

2.32: w

3

a»

,3
.7. a

a

a

a

S
ta

 

 

 

 

 

 

 

BSNOJSBU

(W3)

56

TABLE 2.-—Activation of Slaframine by Rat Liver Fractions.

 

 

Fraction Activity*
Crude Homogenate (1000 g Supernatant fraction) .132/mg
10,000 g Supernatant fraction .025/mg
10,000 g pellet .03A/mg
105,000 g Supernatant fraction O/mg
105,000 g pellet .098/mg

 

*Activity = cm response due to metabolite % cm response due
to 0.1 pg acetylcholine per mg protein in incubation
mixture.

Stability of the Active Metabolite

Experiments preliminary to the isolation of the active
metabolite of slaframine (changes in pH, boiling, etc.) con-
sistently resulted in loss of activity. Attempts to depro-
teinize incubation mixtures, which contained the active
metabolite, by placing in a boiling water bath for two minutes,
resulted in the destruction of the metabolite. The metabolite
was not bound to the protein and could be separated from the
microsomes by centrifugation at 105,000g, however, and could
be detected in the supernatant fraction at levels equivalent
to those observed before centrifuging. Resuspension of the
centrifuged protein and subsequent incubation with slaframine
and NADPH also gave an equivalent response by the ileum. The
conventional methods of precipitating proteins by ionic means
could not be used since the ileum is very sensitive to ion

concentration. No reaction by the ileum was observed after

57

raising or lowering the pH of the microsomal incubation to

10 or 2, respectively.

Cofactor Requirements
Since the production of the metabolite appeared to
cease by the end of ten minutes, it was thought that some
additional cofactor might be required. Therefore a number of en
cofactors were tried in an attempt to increase the production I
of the metabolite. To standard incubation mixtures (2X10-u M

slaframine, 7 mM MgCl 0.A mM NADPH in 5 ml of Tyrodes

2,
solution) 1 mg of various cofactors were added. Controls con—
sisted of identical incubation mixtures without slaframine.
Incubations were carried out at 37° for three minutes in a ,
Dubnoff metabolic shaker. None of the controls elicited any
response by the ileum. All injections into the ileum chamber
were 0.2 m1. A marked increase in activity was observed with
all flavins (Table 3). FMN gave the largest increase in
activity and FAD the least. NADH gave a small increase, while
biopterin, tetrahydrofolic acid, folic acid, NAD+ and NADP+V
had no effect on activity.

The fact that all three flavins tried gave enhanced
activity suggested the possibility that the enhancement was
not due to a cofactor requirement of an enzyme but rather a
reaction of the flavin itself. Further support for this
hypothesis was the fact that time assays which contained any
of the three flavins did not show a decrease in activity even

after one hour incubations. A definite decrease in activity

58

TABLE 3.--Effect of Various Cofactors on the Microsomal
Activation of Slaframine.

 

 

Cofactor Response ACH Response Fraction
(cm) (cm) of ACH

THF 0.A 5.3 0.07
FAD 3.6 6.0 .60
FMN 8.5 A.5 1.89
NAD .8 1.5 .53 F“*
Folate 1.0 8.2 .12 .
Biopterin 0.3 6.0 .05 !
Riboflavin 8.0 A.5 1.95 i
NADPH 6.0 7.0 .86 !
THF + NADPH 2.5 6.5 .38
FAD + NADPH 8.2 5.5 1.A9
FMN + NADPH 8.0 1.8 A.A5
NAD + NADPH 3.7 2.5 1.A8 ;,
Folate + NADPH 1.1 2.6 .A2
Biopterin + NADPH 1.5 2.0 .75
NADH + NADPH 5.5 A.5 1.22

 

was observed in incubation mixtures containing microsomes,
NADPH, and slaframine after 20 minutes (Figure 10).

Carbon monoxide inhibition could not be demonstrated
with liver microsomes. Standard incubation mixtures which
were degassed and flushed with N2 and then incubated at 37°
for three minutes caused identical contractions by the ileum
as when incubations were done aerobically.

The Non-Enzymatic Production of
the Slaframine Metabolite

 

 

For an additional control in the cofactor requirement
experiment (Table 3), only FMN (1 mg) and slaframine (.2 mg)

were added to Tyrodes solution with a final volume of five ml

59

and incubated for three minutes. This solution, which was
completely devoid of microsomes or any protein, caused a
contraction by the guinea pig ileum which was equivalent to
that observed when microsomes were included. Riboflavin and
slaframine at similar concentrations have similar results,
however, FAD and slaframine alone caused no contraction.
Incubations of FAD and slaframine when incubated longer,
however, did give activity. The rate of production of the
metabolite with FAD could be followed and was linear for

about A0 minutes (Figure 12). A similar experiment using FMN
reached a maximum at about 15 minutes. The activity with only
FMN (l mg/S m1) and slaframine (0.2 mg/.5 m1) gave roughly 100
fold greater activity than the microsomal incubation mixture.
However, the activity was proportional to the concentration

of FMN (Figure 13). A dose response curve for the FMN pro-
duced metabolite is shown in Figure 1A.

Confirmation that the species responsible for the con-
traction of the ileum was the same as that which caused the
characteristic reactions in the live animal was obtained by
injecting some of the compound intravenously in a small rat
(100 gm). One ml of a five m1 incubation containing 1 mg
FMN and 0.2 mg slaframine was administered via the femural
vein and caused profuse salivation, lacrimation, and continu-
ous defecation within 30 seconds. These symptoms would
normally require nearly 100 times this quantity of slaframine
and would have required a much longer time for an observable

response (Figure 15).

at l .. z. I“.:

. ‘V‘—x—_ —.»—u—‘—~
.1....'IAA' n I -
1
.
.

60

Figure l2.—-Dependence of the Nonenzymatic Pro-
duction of the Metabolite with FAD upon Time. Injections
of 1 ul of a five ml incubation mixture containing 0.2 mg
slaframine, and 0.1 mg FAD. Results are recorded as
percent of the response realized by the prior injection of
0.1 ug of Acetylcholine.

61

 

 

 

 

O
l l l l
e .2. a

BSNOdSJU 3N I'IOHD'IAISDV IO °/¢

4O 60 80 100

20

TIME

”I'm—'5? 1'33??? 73—51 a: 731717.

IL

62

Figure l3.--Dependence of the Production of
Metabolite on the Concentration of FMN. Various levels

of FMN were incubated with 2 x 10"Ll M slaframine. The

incubation mixtures had a final volume of five ml. One
ul of this incubation mixture was injected into the
ileum chamber. Responses were recorded as a percent of
the response realized by a prior administration of 0.1
ug of acetylcholine. ‘

63

 

 

 

iSNOdSilI SNIIOHDMHDV 5° "4,

6A

Figure lA.--Dose versus Response for The Non-
enzymatically Produced Metabolite. Incubation mixtures,
(five ml total volume) containing 0.2 mg of slaframine
and 0.1 mg FMN were mixed in the light. One pl of the
mixture was injected into the ileum chamber. Results
are expressed as percent of response obtained with 0.1
pg of acetylcholine.

65

 

 

 

 

%
3V :IO
OdSBII 3NI'IOHD'IA13

SSN

DOSE
(III)

66

Figure 15.--Reciprocal Plot of Dose vs Response
for Slaframine in Live Rats. Small rats (100 g) were
injected via heart puncture with slaframine and the
time between the injection and the onset of salivation

was measured.

67

 

10L

1 J L
co 0 <1- 0!

(aim) NoIivavs 01 awu

      

 

.1.

[S] (mg/K9)

68

Typical reactions of flavins often involve free radicals
formed by light. The requirement for light was tested in
incubations with FMN and slaframine. No activity was found
when the two compounds were combined in the dark. The same
mixture, however, when exposed to light did cause the usual
response by the guinea pig ileum.

To determine if the flavin carried out the reaction via
a H2O2 intermediate, 8.6 mg of slaframine were added to 0.5 ml
of 30% hydrogen peroxide. The solution was placed in a boil-
ing water bath for ten minutes. Introduction of 10 pl of the
mixture resulted in contraction of the ileum while the same
amount of 30% hydrogen peroxide caused no response.

Microsomal incubations were carried out in the dark to
see if the observed reaction could be attributed entirely to
a light catalyzed reaction with a fravin. The activity ob-
served when incubations were accomplished in the dark was
equivalent to the amount of activity realized from microsomal
incubations in the light. Furthermore when microsomal incu-
bations in the dark were supplemented with FMN, FAD, or
riboflavin, no increase in activity was observed. Thus it
is shown that the microsomal reaction is indeed a separate
reaction from the photochemical reaction.

Stability of the Metabolite
Produced by FMN and Light

 

The metabolite produced by FMN and light was subjected

to the same conditions as had the metabolite produced by

69

microsomes and NADPH. In all cases (acid pH, basic pH,
boiling, etc.) the treatments gave identical results, namely,
loss of cholinergic activity.

When active incubations were stored in the dark in the
refrigerator for about 2A hours, activity was found to be
absent, but could be regained by exposure to light for long
periods of time. The reacquisition of activity proved to I

require longer time periods at night in the absence of sun-

 

light than in the day when both artificial and sunlight were

.I-u.Mr-2I-¢.—_.-

present. Further additions of fresh FMN and/or slaframine

 

had no effect. On the other hand when active incubations
were kept in the light and in the cold for 2A hours, activity
was immediately observable.

The decay of the active metabolite could be followed by
combining FMN and slaframine in the light and then storing in
the dark for various periods of time prior to assaying with
the guinea pig ileum. A logarithmic plot of the rate of
decay yields a straight line indicating a first order rate
constant for the decay process (Figure 16). The apparent
half-life of the metabolite in the dark is about eight minutes.

Attempts to isolate the metabolite by TLC were unsuc-
cessful. Samples were spotted directly onto TLC plates and
developed with two solvent systems. The only Dragendorff

positive spot corresponded to slaframine.

70

Figure l6.--Decay of the Active Metabolite in
the Dark. The first order rate constant is demonstrated
by the straight line of a plot of time versus the log-
arithmic of the activity. The apparent half-life of the
metabolite is about eight minutes.

71

 

 

 

__I_Ih__
0

$20.55. mZ_._OIU._>.EU< ".0 0\o

30 4O

20
TIME

(min)

10

 

72

Pharmacological Effects of
Activated Slaframine

 

Some unusual pharmacological actions of this metabolite
would suggest that this compound is unlike any other cho-
linergic compound. Addition of the metabolite, either pro-
duced by microsomes and NADPH or FMN and light, caused a
rather slow but prolonged contraction (Figure 17,A). The I?“
slow contraction was intermediate between the action of ;
acetylcholine and an anticholinesterase such as eserine.
Further evidence that the compound was not an anticholines- ‘
terase was obtained by the use of atropine. The addition of L
atropine to the organ bath at levels (8.6 pg), double that
required for the reversal of eserine effects, failed to
reverse the effect of the metabolite (Figure 17,B). These
results suggests a very strong binding of the metabblite
with the receptor.

Further experiments with atropine did confirm the fact
that the action of the metabolite was cholinergic. Addition
of atropine before the addition of the metabolite to the
organ bath completely blocked its action, which is consistent
with data obtained in animals when observing salivation in
response to slaframine.

A gradual but nonetheless significant decrease in sensi-
tivity was observed following administration of the active
metabolite. No comparable result was observed when only

acetylcholine was used to initiate contractions.

73

Figure l7.--The Pharmacological Effects of the
Active Metabolite of Slaframine. The active metabolite
of slaframine produced by microsomes and NADPH was
added to the ileum bath alone (A), prior to atropine (B),
and after atropine (C). Upper arrows indicate washings

and lower arrows additions. Time markings are five
minutes.

7A

m2..—

 

.32
..._U<

«a

A:

 

 

«U «n
« ~ «

 

 

 

t.

 

 

 

 

 

 

(m) aswoasaa

75
Biological Activity of Structurally
Related Analogues

The structure-activity relationship of cholinergic
compounds makes the study of compounds which are related
structurally to slaframine interesting. In addition it
should delineate some of the structure requirements for the
unusual activity exhibited by the active metabolite of
slaframine.
l—hydroxy—octahydroindolizine: The guinea pig ileum did not
contract when 1-hydroxy-octahydroindolizine was administered
by itself or when in combination with RMN.
l-acetoxy-octahdroindolizine: This compound causes a con-
traction by the guinea pig ileum whether given in combination
with FMN or by itself. The observed contraction did not
persist as did the response elicited by the active metabolite
of slaframine. After a single wash the ileum relaxes to its
resting state as it does following the administration of
acetylcholine.

8-amino-octahydroindolizine: Of all the analogues tested,

 

this is the only one which had an observable effect on the

live animal. When mice were injected with this amine the
animals seemed to be tranquillized. .No effect was observed

in the ileum, however, when this compound was given alone

or with FMN.

N-acetyl- and N—acetyl-O-deacetyl-S1aframine: These analogues
caused no response in the guinea pig ileum whether administered

by themselves or with FMN. It has previously been shown that

76

deacetylsalframine is inactive in the live animal and thus
one would not expect to see activity with analogues without

the ester group (A7).

EPR Studies of the Metabolite
Produced by FMN

 

EPR spectra were recorded at room temperature with a
Varian X—band spectrometer equipped with an optical trans-
mission cavity. The magnetic field strength was calibrated
with a Varian F-8 proton resonance flux meter, the frequency
of which was monitored by a Hewlett-Packard frequency counter.
The microwave frequency was calibrated with a Silverlab
wavemeter. Light of a Zenon Lamp was focused on the sample
by a quartz lens.

No free radicals were observed in the range where
N-oxides absorb even with maximum sensitivity. The sensitivity
of the instrument was such that a concentration of 10-7M

could be readily detected.

bur~m' '25.! " :Au-c...’ ug-...——.~ -1
. a
:1

DISCUSSION

The data presented in this thesis confirms the indica-
tions obtained by Aust 32 a1. (13) that slaframine, which is
inactive itself, is bioactivated to a very potent and long
lasting parasympathomimetic substance. [Slaframine is acti-
vated by liver microsomes, but can also be produced non;
enzymatically via a photochemical reaction with flavins.

All attempts to isolate a metabolite of slaframine have
been unsuccessful. Both TLC and GLC of basic chloroform
extracts of blood or urine of animals given slaframine
revealed only the presence of slaframine or deacetyl-
slaframine, even when radioactive slaframine was used. Livers
of slaframine treated animals were also homogenized and ex—
tracted with chloroform with similar results, excluding the
possibility that much of the metabolite was remaining in the
liver. These results indicated that the metabolite might be
either unstable or present in very minute quantities, or both.
If the amount of metabolite was substantial but unstable, the
decomposition product must be slaframine itself, since nearly
all of the slaframine given was recovered as the same. The
identical Rf values to authentic slaframine in three TLC

solvent systems of the recovered material was taken as evi;

dence that it was actually slaframine.

77

78

Results from liver perfusions seemed to confirm the
results obtained in live animals, i.e. that no metabolite
could be detected by the previously mentioned chemical
methods. Radioactivity could be nearly completely recovered
when 3H-(acetate)-slaframine was used in the perfusion. All
the radioactivity in the pH 10 chloroform extract was shown
to be slaframine by TLC. A substantial amount of the tritium T
was not extractable by chloroform at pH 10, however, and the I

amount of counts in the water increased with incubation time.

 

The water fractions Were analyzed to determine if the tritium
in the water was a metabolite of slaframine or just the
acetate which had been cleaved. All the counts were shown to
correspond to acetate which would indicate that hydrolysis
was taking place and was time dependent.

The possibility that deacetylslaframine was the active
metabolite of slaframine was rejected because biological
activity could not be demonstrated when deacetylslaframine
was injected into live animals.‘

Aust (A7) had previously shown that slaframine is not
hydrolyzed by a number of cholinesterases. Therefore the
results of this experiment may indicate that slaframine may
be altered to a species which is susceptible to hydrolysis
by some esterases.

Attempts to demonstrate a metabolite of slaframine by
TLC after incubation with various liver preparations (slices,

crude homogenates and microsomes) were without success.

79

Since attempts to demonstrate the presence of an active
metabolite of slaframine by chemical methods were unsuccessful,
attention was shifted to the biological evidence for its
presence. Aust (13) had previously shown that slaframine
acted as a cholinergic drug in live animals. This along with
the probability that the concentration of active metabolite
was very 13w, made the guinea pig ileum an obvious assay.

Microsomal incubations containing slaframine and NADPH
caused a sustained contraction of the guinea pig ileum while
proper controls (Figure 8) failed to give a similar response.
These results were considered conclusive evidence that there
is an active metabolite of slaframine and that it is produced
by liver microsomes.

The transformation could not be prevented by carbon
monoxide inhibition even after repeatedly degassing and

flushing with N followed by gassing with carbon monoxide.

2
Therefore cytochrome P-A50, commonly involved in the terminal
oxidation step in the metabolism of foreign compounds, was
shown to be not involved in the activation of slaframine.
There was an increase in activity when phenobarbital induced
microsomes were used instead of control microsomes. This
indicates that some portions of the microsomal drug metabo-
lizing system might be responsible, but not the entire system.
Another atypical characteristic was that the production of

metabolite reached a maximum after approximately five minutes

of incubation. This suggested that another cofactor may be

80

required for further increases in the production of the meta-
bolite. Another explanation would be that the enzyme concen-
tration was limiting, therefore experiments were designed to
localize the enzyme system responsible for the bioactivation
of slaframine, so that the responsible enzyme could be con-
centrated. Results showed that the greatest activity was in
the microsomal fraction. A considerable amount of activity
was also found in the washed mitochondrial fraction suggest—
ing that the reaction may not be carried out solely by the
mixed-function oxidases of the microsomal fraction; the usual
location for drug transformations.

The search for an additional cofactor resulted in the
finding that a marked stimulation in activity was observed
with added free flavins (Table 3). Stimulation of metabblism
of tertiary amines by added free flavins has previously been
reported (A8). The lack of a requirement for molecular
oxygen and the apparent fast rate constant are also consistent
with reports of the transformation of some tertiary amines to
N-oxides (A8). However, this stimulation, by flavins, was
found to be nonenzymatic, which has not previously been I
reported for the transformation of other tertiary amines.

The nonenzymatic reaction with flavins is thought to
proceed via a free radical since an absolute requirement for
light was demonstrated. Photoreduction of free flavins goes
via rapid dismutation of the semiquinone to the hydroquinone-
level. Since the photoreduction of flavins by tertiary

amines has been reported (A9, 50) it is conceivable that

81

either the half-reduced species of flavins, which is a free
radical, or the process of producing it, is responsible for
the observed reaction. Evidence for the latter mechanism

has been obtained by comparison of the reaction rate with
riboflavin, FMN, and FAD. McCormick (50) has reported that
photoreduction of FMN and riboflavin to the hydroquinone
level is approximately four times faster than the photoreduc-
tion of FAD. Similarly, the activation of slaframine by
riboflavin or FMN was very rapid, however, then FAD was used

as the flavin a much slower rate of production is observed

 

(Figure 12).

If this free radical mechanism is indeed responsible for
the activation of slaframine, slaframine may participate in
the photoreduction of the flavin to the half-reduced, free
radical species. The oxidized (i.e. minus one electron)
slaframine perhaps is the active metabolite. Alternatively,
the hydroquinone free radical of FMN, could react with
slaframine, or with water to form hydrogen peroxide which
could react with slaframine by some mechanism and form the
active metabolite which may be the N-oxide of the tertiary

amine.

Slaframine -flavin (H202 Slaframine ‘
Slaframine (-e-)> (flavin.‘> H2O) CSlaframine N-oxide
Hydrogen peroxide and slaframine, after refluxing for ten min-

utes gave some activity, indicating that the above mechanism

is plausible.

82

The concentration of the active metabolite is unknown,
but is estimated to be very low. In experiments with the

guinea pig ileum, one p1 of an incubation mixture which is

I;

2 x 10' M slaframine is added to the organ chamber which has

8 M final concentration). This con-

centration gave a response approximately equal to 2 x 10'8 M

volume of 20 ml (1 x 10-

acetylcholine. Thus, if slaframine was completely converted
to the active metabolite, the active metabolite would appear
to be roughly as active as acetylcholine. However, acetyl-
choline is readily hydrolyzed by acetylcholinesterase, so

its effective concentration is not known but is probably much
less than 2 x 10'8 M. Slaframine is insensitive to the action
of acetylcholinesterase (A7), so its effective concentration
is the actual concentration. If such comparison can be made
with acetylcholine, then the concentration of the active ‘
metabolite is even less. In all probability the active meta-
bolite is more active than acetylcholine, which makes it
perhaps the most potent physiological substance known at the
present time. The concentration of the free radical, which,
in addition to the problem of stability, could easily be
present at concentrations below the limits of detection by
EPR.

Failure in attempts to increase the amount of metabolite
by increasing the irradiation periods suggests that the
photodecomposition products may prevent the production of the
active metabolite. This also explains why activity cannot be

regained in a solution which has lost activity after setting

 

83

at room temperature in the light, even after additions of
fresh slaframine and FMN.

The prolonged contraction observed by the guinea pig
ileum even after repeated washings suggests that the active
compound has a very high affinity for the receptor. Indeed
the binding of the active metabolite to the receptor may be
irreversible. After the addition of the active metabolite
and repeated washings, the ileum rarely completely relaxes
and the response to acetylcholine diminishes. If the active
metabolite is a free radical and the anionic site of the
receptor had at least a partial negative charge the binding
could be covalent.

Proof that the active metabolite binds to the acetyl-
choline receptor was obtained by experiments with atropine.
Since atropine will block the effects of the metabolite if
given previously, its action must be elicited by means of
the acetylcholine receptor. If it was blocking cholinesterase
which would allow acetylcholine to act, atropine would reverse
its effects if given at the height of contraction. This does
not happen (Figure 17), in fact there is no difference between
application of atropine to the contracted ileum and washing.
Therefore the metabolite must act directly on the receptor.

One should recognize the structural similarity between

acetylcholine and slaframine:

8A

 

0
H
OTCH3
CH2 0
CH H
/ 2 000113
N+ g l
/ \ C“
H3C CH3 L \\b
cHé//’ H2
CH3 ,

The structure-activity relationship is at present only
speculative, but the similarity in structure of slaframine
and acetylcholine suggest that the active metabolite 18‘
bound in the same manner as acetylcholine. The receptor most
likely binds at the ester and tertiary amine positions. It
is reasonable that the stronger binding of the metabolite is
due to the anionic site rather than the esteratic site. A
very plausible explanation of this could be a partial or com-
plete covalent bond is formed between the ring nitrogen of
the metabolite and the receptor site. 3

A possible explanation of the specificity with which the
active metabolite of slaframine acts can be built on the struc-
tural properties of slaframine. If the receptors at various
glands differ in the spatial arrangement between the estera-

tic and anionic sites, only those which have the exact

85

structure to compliment the rigid structure of slaframine
would be susceptible to binding by activated slaframine.

Since acetylcholine is flexible it can accommodate a wide
range of spacial differences between the esteratic and

anionic sites. Therefore acetylcholine stimulates all
cholinergic receptors while the active form of slaframine
stimulates only those receptors which have the correct spatial
arrangement for these binding sites.

An esthetically pleasing possibility for the bioactiva-
tion of slaframine would be to create a positive charge on the
tertiary nitrogen. If slaframine gives up an electron from
its tertiary amine nitrogen, (to a flavin in the presence of
light) the tertiary nitrogen acquires a positive charge.

This charge would be stabilized to some extent by the primary

amine group in the six position of slaframine. This could

give the radical a significant lifetime rather than its probable

immediate destruction in the absence of the primary amine.

The unpaired electron is subsequently available for sharing
with the anionic site of the receptor molecule and could form
a partial covalent bond with the receptor. This would explain
the extreme persistence of action of the metabolite on the
guinea pig ileum despite frequent washings. If this assump-
tion is correct, the active metabolite of slaframine could be
an extremely useful tool in the isolation of the receptors)

for acetylcholine.

86

Failure in the attempts to isolate the active metabolite
in the in XlXQ experiments discussed above could easily be
explained if the active species was merely slaframine minus
an electron. Changes in pH would certainly affect the stab-
ility of such a species so it would be destroyed by the pH
10 chloroform extractions. In addition, the radical would be
more polar and thus would be less readily extractable by
organic solvents. However, these arguments are probably not
necessary, for all attempts to show a product of slaframine
and FMN by TLC under very mild conditions (no pH change or
extraction) were unsuccessful. TLC of the mixture 2A hours
after mixing also failed to show any product of slaframine.

A very likely candidate for the enzyme responsible for
the enzymatic activation of slaframine is NADPH-cytochrome
0 reductase. This FAD containing enzyme has been speculated
as being part of the electron transport chain of the mixed-
function oxidases of liver microsomes, however, this has not
been confirmed. Its candidacy as the enzyme responsible for
activation of slaframine is supported by the finding that
the reaction requires NADPH which can act as the electron
donor to reduce the flavin in the reductase.

Confirmation that the enhancement in activity obtained
by added free flavins was nonenzymatic came from experiments
employing microsomes in the dark. Microsomes themselves
continue to produce the metabolite in the dark but no increase

in activity is obtained by supplementing the incubation

 

87

mixtures with riboflavin, FMN, or FAD. This clearly points
out that two separate reactions are involved. Both reactions
could very well be carried out by flavins. The microsomal
reaction, as stated above, could involve the flavoprotein,
cytochrome 0 reductase which gets its reducing power from
NADPH thus differing from free flavins which abstract an
electron from an electron souce, like a tertiary amine, in L
the presence of light. I

The transient nature of the active metabolite is

 

evidenced by the fact that its activity in the dark decays
with a half-life of eight minutes.' The rate of decay of the
metabolite is dependent on pH and temperature. Pathways of
decay, however, may be different at low pH than at high pH.
Slaframine is known to be readily hydrolyzed by basic con-
ditions. Hydrolysis of the ester moiety is probably not
significant in these experiments, however, since slaframine
can still be detected by TLC after exposure to pH 10 or

pH 2.

It is difficult to understand why the liver carries out
any reaction with slaframine. The typical drug metabolizing
reactions of the liver make drugs sufficiently polar to be
excreted by the kidneys. Slaframine is quite polar and can
be excreted readily in the urine as was determined by the
live animal experiments. Nearly all of the slaframine could
be recovered from the urine. One cannot rule out the possi-

bility that slaframine is in its active form when the kidney

88

removes it from the blood and then decomposes to slaframine
once it is in the urine.

One also cannot eliminate the possibility that the
hydrolysis of the active compound may be much more easily
accomplished. Thus the actual detoxification of the "toxib"
metabolite could occur by hydrolysis while excretion of the
"non toxic" slaframine is carried out without alteration.
This theory is consistent with results of administration of
3H-(acetate)-slaframine to the perfused liver. The results
of this experiment demonstrate the loss of acetate from
slaframine with time. Slaframine and deacetylslaframine are
also found in the urine of animals given slaframine.

The stability of the metabolite, produced by different
systems, was not found to differ. Activity was lost when
either the microsomal or flavin incubation mixtures were
subjected to acidic conditions (pH 2), basic conditions
(pH 10) or boiling. This was taken as additional evidence
that the metabolites were the same. No difference could be
observed in the pharmacological effects of the enzymatically
produced and the nonenzymatically produced metabolite.

Studies with analogues of slaframine help to delineate
the necessary structural properties for activity. For ex-
ample, all attempts (as previously mentioned) to demonstrate
activity in analogues which lack the O-acetate group have .
been unsuccessful. Thus an absolute requirement for the
acetate group has been demonstrated. l-acetoxy-octahydro-

indolizine showed some biological activity by itself, but

89

could readily be washed off the receptor sites. The observed
action may have been a general response of the acetate ester
because the concentration was unknown. If the observed re-
action of this analogue is specific, it may point out the
necessity of the primary amine for the long lasting effect of
the metabolite. This would strengthen the arguments for a
free radical since the presence of the primary amine would
contribute significantly to the stabilization of a free
radical. Further evidence for this is shown by the lack of
activity when an electron releasing group is attached to the
primary amine as with N—acetylslaframine. This function of
the primary amine gains interest in view of the fact that
both thydroxy-octahydroindolizine and l-acetoxy-octahydro-
indolizine have been implicated as precursors of slaframine
(10). It is interesting to speculate about the purpose of
the enzyme which places the primary amine in the six position
of slaframine during its synthesis by R. leguminicola.

All attempts to find a free radical by EPR have thus
far been unsuccessful, but this does not eliminate the
possibility of its presence. The sensitivity of EPR for
the detection of free radicals is about 10"7 M. It is en-
tirely possible that the concentration of the metabolite is
less than this. It may also be necessary to lower the
temperature of the solution to see the radical. This would
require the use of an organic solvent. The production of

metabolite has not yet been investigated in organic solvents.

‘ _-‘.M.'.I 4“ '-_—4

90

One problem with the use of organic solvent is that they may
have an adverse effect on the guinea pig ileum, which is at

the present time the only assay method known.

SUMMARY

Slaframine is bioactivated to an active compound which
has a stronger affinity for the acetylcholine receptor than
either acetylcholine or atropine.i This active metabolite
can be produced by liver microsomes and NADPH in zitgg. It
can also be produced nonenzymatically by flavins in the
presence of light. The production of the active metabolite
by the photochemical reaction with flavins results in a much
higher (about 200 fold) yield of metabolite than the micro-
somal production. The active compound appears to be effective
at molar concentrations below the minimum level of acetyl-
choline required to elicit a responSe.

Isolation attempts have failed thus far because apparently
only an extremely small amount of the metabolite is present and
it is unstable. Chemical evidence nor the presence of the
metabolite of slaframine has not been successful for the same
reasons.

The only assay at the present time is the guinea pig
ileum. The active form of slaframine has a sustained effect
on the ileum and cannot be reversed by atropine. The loss of
ability of the guinea pig ileum to contract occurs much faster

after exposure to the metabolite than when only exposed to

91

 

92

acetylcholine. This makes the guinea pig assay undesirable
and a chemical assay more appealing.

Certain structural properties of slaframine have been
shown to be important for the observed effect of the meta-
bolite. The acetate ester and the ring nitrogen (three
carbons away) are necessary for activity. The primary amine
group may be important, but conclusive evidence awaits the

absolute identity of the active compound.

10.

ll.
12.

13.
1A.

15.

REFERENCES
Aust, S. D., Broquist, H. P. and Rinehart, K. L., Jr.
Biotechnology and Bioengineering,lQ,A03 (1968).

Byersé J. H. and Broquist, H. P. J. Dairy Sci.,fl§,873
l9 0).

Byers, J. H. and Broquist, H. P. J. Dairy Sci.,flfl,ll79
(1961).

O'Dell, B. L., Regan, W. O. and Beach, T. S. Univ. of
Missouri Research Bu11.,702,(1959).

Smalley, E. B., Nichols, R. E., Crump, M. H. and
Henning, J. N. Phytopath.,52,753 (1962).

Crump, M. H., Smalley, E. B., Henning, J. N. and

Nichols, R. E. J. Am. Vet. Med. Assoc.,1u3,996 (1963).
Aust, s. D. and Broquist, H. P. Nature,20§,20A (1965).

Aust, S. D., Broquist, H. P. and Rinehart, K. L., Jr.
J. Am. Chem. Soc.,§§,2879 (1966).

Gardiner, R. A., Rinehart, K. L., Jr., Snyder, J. J.
and Broquist, H. P. J. Am. Chem. Soc.,22,5630 (1968).

Snydgr, J. J. Ph. D. Thesis, University of Illinois,
19 9. " .

Aust, S. D. Biochem. Pharmacol. (in press).

Aust, S. D. and Salisbury, G. W. Fed. Proc.,gfl,6Al
(1965).

Aust, S. D. Biochem. Pharmacol.,l§,929 (1969).

Mason, H. 3., North, J. C. and Vanneste, M. Fed.

Mason, H. s. Science,l25,1l85 (1957).

93

16.

17.

18.

19.

20.
21.

22.

23.

2A.

25.

26.

27.

28.

29.
30.
31.
32.

33.

9A
Strittmater, P. Biological Oxidations,MInterscience
Publishers, Wiley (1968).

Ernster, L. and Orrenius S. Fed. Proc.,gfl,ll90-9

Wattenberg, L. W., Leong, J. L. and Galbraith, A. R.
Proc. Soc. Exptl. Biol. Med.,l27,A67-69 (1968).

Alfred, L. J. and Gelbain, H. V. Science,1§2975-76
(1967).

Conney, A. H. Pharmacol. Rev.,lg,3l7-66 (1967).

Remmer, H. and Merker, H. H. Ann. N. Y. Acad. Sci.,
125,79-97 (1955).

Gram, T. E., Rogers, L. A. and Founts, J. R. J.
Pharmacol. Exptl. Therap.,lSZ,A35-A5 (1967).

Rubin, A., Tephly, T. R. and Mannering, G. J. Biochem.

Pharmacol., 13 1007 (1965) .

Rubin, A., Tephly, T. R. and Mannering, G. J. Biochem.

Pharmacol., 13, 1053 (196A) .

Stizel, R. E., Tephly, T. R. and Mannering, G. J. Mol.

Pharmacol.,fl,15-l9 (1968).

McMahon, R. E., Mills, J., Culp, H. W., Gibson, W. R.,
Miller, W. M. and Marshall, F. J. J. of Med. Chem.,

g, 207 (1969).

Tephly, T. R. and Mannering, G. J. Mol. Pharmacol.,fl,
lO-lA (1968). ,

Cooper, D. Y., Levin, S., Narasimhula S., Rosenthal, O.

and Estabrook, R. W. Science,lAZ,A00 (1965).
Klingenberg, M. ‘Arch. Biochem.,15,376 (1958).
Garfinkel, D. Arch. Biochem.,11,A93 (1958).

Omura, T. and Sato, R. J. Biol. Chem.,232,2370 (196A).

Omura, T., Sato, R., Cooper, D. Y., Rosenthal, O. and
Estabrook, R. W. Fed. Proc.,gfl,ll8l (1965).

Omura, T., Sanders, E., Estabrook, R. W., Cooper, D. Y.
and6gosenthal, 0. Arch. Biochem. Biophys.,ll7,660
(19 ).

All; 'E‘AI-Jaof .

3A.

35.

36.

37.
38.

39.
A0.

A1.
A2.

A3.
AA.

A5.
A6.

A7.
A8.

A9.

50.

95

Remmer, H., Schenkman, J., Estabrook, R. W., Sasame,
H., Gillette, J., Narasimhulu, 8., Cooper, D. Y.
and Rosenthal, 0. Mol. Pharmacol.,2,187 (1966).

Sladek, N. E. and Mannering, G. J. Biochem. Biophys.
Res. Commun.,gfl,668 (1966).

Christianson, R. Mon. J. Med., Condon, 29,193 (1855).
Fraser, T. R. Edin. Med. J.,2,36 (1863).
Argyll—Robertson, D. Edin. Med. J.,§,815 (1863).
Stedman, E. Biochem. J.,gg,719 (1926).

Wilson, I. B., Hatch, M. A. and Ginsburg, S. J. Biol.
Chem.,235,2312 (1960).

Stowe, C. M. Veterinary Pharmacology.

Lowry, O. H., Rosebrough, N. J., Farr, A. L. and
Randall, R. S. J. Biol. Chem.,l93,265 (1951).

Nash, T. Biochem. J.,§§,Al6 (1953).

Brodie, B. B. and Cooper, J. R. J. Pharmacol. Expt'l
Therap.,llA,A09 (1955). '

Somogyi, M. J. Biol. Chem.,§§,655 (1930).

Van Harken, D. R., Tephly, T. R. and Mannering, G. J.
J. Pharmacol. Expt'l Therap.,lA2,36 (1965).

Aust, S. D. Ph.D. Thesis, U. of Ill. 1965.

Pettit, F. H. and Ziegler, D. M. Biochem. Biophys.
Res. Comm.,l_3,193 (1963).

Merkel, J. R. and Nickerson, W. J. Biochem. Biophys.
Actau,lfl,303 (195A).

McCormick, D. B. Flavins and Flavoproteins, University
Park Press (1968).

IHIIIHIHHIW||||H

3177 5582

l
l
Ill
l
l
Ill
l
l

"I

31

'IIHIAIH