STUDIES ON THE BSGSYNTHES'IS OF THE
WRQNNE RING OF NECONNE

Thesis For Hm Dagny: a} DB. D;
ifiICHlGifl'SYAYE UNE¥£RSE?Y
Kenneth P. Heilman

1962

”lb \1 I“
'rTHESIs

. a Q-..
CD ~ 01/ Mllf‘hlgtm

Universinl
--_ 2

v,
.v" ‘4

 

 

 

 

 

 

MICHIGAN STATE UNIVERSITY
DEPARTMENT OF CHEMISTRY
EAST LANSING, MICHIGAN

ABSTRACT

STUDIES ON THE BIOSYNTHESIS OF THE
PYRIDINE RING OF NICOTINE

by Kenneth P. Hellman

The pyridine ring of nicotine is known to arise from the pyridine
ring of nicotinic acid in tobacco plant metabolism. The $3 wsyn-
thesis of the pyridine ring is not known, however, since it has been
shown that nicotinic acid does not arise from tryptophan in tobacco as
it does in other living organisms. Carbon atoms from various smaller
molecular weight metabolites have been shown to be precursors of the
pyridine ring. ’These include the methyl carbon of acetate, the a-carbon
of propionate, the methylene carbons of succinate, all the carbons of
glycerol, and several others. The elucidation of the exact role of these
substances in pyridine biosynthesis must await the development of
appropriate methods of chemical degradation of the pyridine ring.

The purpose of the present study is twofold: first to attempt to
clarify the role of propionic acid--and similar three carbon units such
as fi-alanine--in pyridine biosynthesis, and to determine whether or
not these compounds are immediate, 3-carbon unit precursors of the
pyridine ring; and secondly, to aid in determining the role of these and
other precursors by finding chemical methods of degradation of the
pyridine ring of nicotine. Of particular interest are the two carbons
adjacent to the nitrogen atom-i. e. , the 2 and 6 carbons--since certain
theories of propionate and glycerol incorporation involve these positions.

To accomplish the first objective, several possible pathways of

chemical degradation were investigated. The first was the synthesis

Kenneth P. Hellman

of Z-aminonicotine and 6-aminonicotine by the action of sodamide on
nicotine. Attempts to replace these amino groups with phenyl rings by
various methods all failed. Also unsuccessful were attempts to
phenylate these aminonicotines at the unsubstituted u-carbon using
phenyllithium .

2-Phenylnicotine and 6-phenylnicotine were finally prepared by
the reaction of nicotine with phenyllithium. However, all attempts to
oxidize these two derivatives to benzoic acid with permanganate as
originally planned (the carboxyl carbons of these benzoic acids would
represent the desired positions of the pyridine ring) yielded only the
respective phenylnicotinic acids and not benzoic acid. Thus, although a
start was made, the degradative objectives of this study were not
accomplished.

The second part of the study involved hydroponic feeding of tobacco
plants with propionate and B-alanine labeled with carbon-14 at various
positions, determining the extent of incorporation of radioactivity into
nicotine isolated from these plants, and then, by appropriate methods
of degradation, determining the pattern of labeling in various parts of
the nicotine molecule, the pyridine moiety in particular.

Results of this study showed that propionate-3-C”, fi-alanine-Z-C”,
and [i-alanine-3-CHr were all significantly incorporated into nicotine,
whereas propionate-l-Cl4 was not. Degradation studies demonstrated
that neither propionate-3-C14 nor B-alanine-3-C” gave rise to radio-
activity in the pyridine ring, compared to previous studies by other
workers which indicated that, after feeding propionate-Z-C” and
fi-alanine-Z-C”, approximately fifty percent of the nicotine radioactivity

was located in the pyridine ring.

Kenneth P. Hellman

These results were interpreted to indicate that, although pro-
pionate and fi-alanine are apparently metabolically related in tobacco
as in other organisms, neither is an immediate precursor of the
pyridine ring of nicotine. Present evidence seems to indicate that
these compounds may be metabolized to acetate, probably by B-oxi-
dation or a similar pathway, prior to incorporation into the pyridine

ring.

STUDIES ON THE BIOSYNTHESIS OF THE
PYRIDINE RING OF NICOTINE

By

Kenneth P. Hellman

A THESIS

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

DOCT OR OF PHILOSOPHY

Department of Chemistry

1962

\

ACKNOWLEDGMENTS

The execution and presentation of this study could not have
been satisfactorily completed without the guidance and patience of
Dr. Richard U. Byerrum and the untiring aid and encouragement
of Dr. Thomas ..Gr'iffith.. To both is extended the author's sincer-
est gratitude.

The author also wishes to express his appreciation to the
National Institutes of Health for their continued financial aid during
the course of this project.

>:< ::< >:< >1: )1: )k :1: >:< >§< >',< >:< >{c >1: >1: 3::

ii

VITA

The author was born July 28, 1934, in Baltimore, Maryland,
and received his secondary education at the Baltimore City College.
He attended Drew University where he received the Bachelor of Arts
degree in 1956, majoring in chemistry.

He entered graduate school at Michigan State University in
1956, and spent two years as a Special Graduate Research Assistant
in the Department of Agricultural Chemistry, receiving the Master
of Science degree in chemistry in 1959.

In 1958, the author transferred to the chemistry department
where he served one year as a graduate teaching assistant, one year
as a Special Graduate Research Assistant under a National Institutes
of Health grant, and the remainder of his graduate work as a National

Institutes of Health Predoctoral Fellow.

iii

TABLE OF CONT ENTS

INTRODUCTION................... ......

EXPERIMENTAL AND RESULTS .

Preparation of Plants . . . . . ..............
Feeding of Plants . . . . . . . . . . .........
Isolation and Purification of Nicotine. . .......
Determination of Radioactivity ..............
Degradation of Radioactive Nicotine . . ...... .
Attempted Degradation of the Pyridine Ring of Nicotine .
Preparation of Sodium Amide ...............
Amination of Nicotine. . . . .....

Attempted Diazotization and Coupling of 2- Aminonicotine
Attempted Phenylation of Aminonicotines. . .
Phenylation of Nicotine . . . . . . . . . . . . ......
Oxidation of Phenylnicotines ......

DISCUSSION. ........ ......

Attempted Degradation of the Pyridine Ring of Nicotine .

Attempted Phenylation of Aminonicotines . . ......

Phenylation of Nicotine .

Feeding Experiments . . . . . ..............
SUMMARY ..... . . ...... . . . . ...........
LITERATURE CITED . . . . . . . . .......

iv

Page

11
12
15
19
2.0
21
24
27
29
31

34
35
37
38
39
46

49

LIST OF TA BLES

TABLE Page
I. Administration of Labeled Precursors of Tobacco
Plants...........................10

II. Incorporation into Nicotine of C14 from Several
Precursors ..... . . . .......... . . . . . . 14

III. Results of a Partial Degradation of Radioactive
Nicotine . . ..... . . . 18

INT RODU CT ION

INT RODU CT ION

The pyridine ring occurs in a number of biologically important
compounds--e. g. , the vitamins pyridoxine and niacin, the pyridine
nucleotides, and a number of alkaloids. This ubiquitous occurrence in
nature of compounds containing the pyridine nucleus gives rise to the
question of its origin, 1. e. , the compounds and enzymatic reactions
which comprise the biosynthetic pathways leading to these pyridine-
cont'aining substances. Knowledge of these pathways is important for
an understanding of both the metabolic interactions in the organisms
in which they occur and the functions which some of these substances
serve.

It has been shown that the pyridine moiety of nicotinic acid arises
from the indole nucleus of the amino acid tryptophan in animals (1, 2)

and in molds such as Neurospora crassa (3, 4). The details of this

 

conversion have been worked out and the intermediates characterized.

Dawson, cit a1. (5), using tritium and carbon-14 labeled nicotinic
acid, demonstrated the conversion of nicotinic acid to nicotine by
excised tobacco root cultures. However, it has been shown that in
higher plants including tobacco (6, 7, 8), and in certain bacteria (9),
nicotinic acid is not derived from tryptophan. Thus in the tobacco plant,
nicotinic acid, and therefore nicotine, is not synthesized by the pathway
involving tryptOphan as it is in animals, but the existence of an alternate
route must be assumed.

The problem of the £13 nogsynthesis of the pyridine ring of
nicotine, and of nicotinic acid in organisms which do not use the
tryptophan pathway, is one which has been investigated by several groups

of workers. In 1959, in this laboratory, it was shown that in tobacco

plants fed acetate-Z—C”, approximately forty percent of the radio-
activity incorporated into nicotine was located in the pyridine ring (10, 11).
Later studies showed that propionateu-Z‘a-C14 and glycerolol, 3-CM also
gave rise to nicotine which was labeled extensively in the pyridine ring
(12); in addition, about fifty percent of the radioactivity in the pyridine
ring was in the 3-position after feeding acetate-Z-C”, and about forty-
percent was in the 3-position after feeding propionate-Z-C”. More
recently, the same workers indicated that after feeding glycerol-Z-C”
to tobacco plants, nearly sixty percent of the radioactivity of the nicotine
was located in the pyridine ring (13).

Ortega and Brown have obtained similar results with Escherichia

 

£911, a bacterium which, like tobacco, does not synthesize its nicotinic
acid from tryptophan. In this organism, glycerol-1, 3-Cl4 and succinate-
2, 3-Cl4 give rise to nicotinic acid labeled predominantly in the pyridine
ring, whereas succinate-l, 4-C14 gives rise mainly to carboxyl labelled
nicotinic acid (14, 15). These workers concluded that in E? 5.9.1.1, the
carbon skeleton of nicotinic acid is derived from glycerol- or a 3-carbon
compound metabolically related to glycerol- and a 4-carbon dicarboxylic
acid; and these compounds condense so that one of the carboxyl groups of
the dicarboxylic acid becomes the nicotinic acid carboxyl. They also
suggested that tobacco plants synthesize the pyridine portion of nicotine
by a mechanism similar to that utilized by bacteria for the synthesis

of the pyridine ring of nicotinic acid.

The biosynthesis of ricinine, a toxic alkaloid produced by the
castor plant and having the structure N-methyl-3-cyano-4-methoxy-2-
pyridone, has been studied recently by Waller and Henderson (16, 17).
Their studies on the incorporation of several metabolites revealed that
succinate, propionate, and B-alanine were incorporated into ricinine,
in that order of efficiency. In addition, they found that after feeding

succinate-l, 4-Cl4, seventy-five percent of the activity in the ricinine

was in the nitrile group and twenty-five percent in the pyridone ring,
whereas with succinate-Z, 3—C”, more than ninety-eight percent was

in the pyridone ring. These results suggest the same conclusions that
were made regarding nicotinic acid in E. _c_:_o_1_i---that the interior two
carbon atoms of a 4-carbon dicarboxylic acid are incorporated into the
pyridine ring, and that one of the carboxyl carbons becomes a sub-
stituent, either the carboxyl group of nicotinic acid or the nitrile group
of ricinine.

The demonstration of the incorporation of these metabolites into
the pyridine ring of nicotine and other naturally occurring pyridine com-
pounds raises the question of the mechanism of this incorporation.

In order to speculate on the mechanism, it is first necessary to know
the orientation of the incorporated molecules with respect to the pyridine
ring itself. For example, we may know with reasonable certainty that
the l and 3 positions of glycerol are incorporated into the pyridine ring
of nicotine, but we do not know into which specific positions of the
pyridine ring these particular carbon atoms of glycerol are converted.
To obtain this information it is necessary to develop a method of
chemical degradation which will allow the isolation of the various carbon
atoms of the pyridine ring unambiguously. The great stability of the
pyridine nucleus to chemical attack, either oxidation or substitution,
renders the problem of degradation somewhat difficult.

Studies previously reported from this laboratory have postulated
that the pyridine ring of nicotine might be formed dc: n_o_\Lo_from the
carbons of glycerol and propionate (18, 19). It was further suggested
that the Z and 3 carbons of propionate might give rise to the carbons in
positions 3 and 2, respectively, of the pyridine ring, whereas glycerol
might contribute its carbons to positions 4, 5, and 6 of the pyridine ring.
Indeed, in aforementioned reports (12, 13), it was demonstrated that

glycerol and propionate-Z-C14 are incorporated into the pyridine ring of

nicotine, and that fifty percent of the activity in the pyridine ring from
propionate-Z-C14 occurs in the 3 position. It was argued that the
remaining fifty percent of the pyridine activity might reside in the 2
position, since one of the major pathways of propionate metabolism
involves conversion to succinate (202%. Thus, if propionate is in equi-
librium with succinate in tobacco, the radioactivity in propionate-Z-Cl4
would tend to be randomized between positions 2 and 3. In such a case,
one might predict that the remaining fifty percent of the activity of the
pyridine ring after feeding propionate-Z-C” would reside in the
2 position of the pyridine ring. In addition, one would predict that pro-
pionate-3--C14 would give rise to results identical to those with
propionate-Z-CM.

On the other hand, if speculation that the carbons of the pyridine
nucleus arise from glycerol and propionate is correct, and if, as it
has been partially demonstrated previously, carbons Z and 3 of the
pyridine ring arise from propionate, it is likely that carbons 4, 5, and
6 come from glycerol. In addition, it would seem, according to this
hypothesis, that glycerol-1, 3-C14 would give rise to nicotine labeled
in the 6 position of the pyridine ring, whereas glycerol-Z-C” would give
rise to nicotine labeled, not in the 6 position of the pyridine moiety,
but principally in the 5 position. Since both glycerol-Z-Cl4 and glycerol-
1, 3--C14 have been shown to be efficient precursors of the pyridine ring
of nicotine, acceptance or rejection of the stated hypothesis depends
on the determination of the location of the radioactivity in the nicotine
pyridine ring after feeding each of the labeled glycerol precursors.
Similarly, proof of the other part of the hypothesis--the role of propionate
as a precursor of carbons 2 and 3--also depends on the answer to
several questions: (1) whether propionate-3-Cu‘ and propionate--l--Cl4
are also efficient precursors of the pyridine ring of nicotine; and, if

so, (2) the location of the radioactivity in the pyridine ring after feeding

these metabolites. It would also be interesting to know if there are any
other low molecular weight precursors of the pyridine ring, and, if so,

the relationship between these and the previously demonstrated precursors,
propionate and glycerol.

Therefore, the purpose in undertaking this study was to attempt
to resolve two general problems: (1) to determine more specifically
the role of propionate in the biosynthesis of the pyridine ring of nicotine
and to test other suspected precursors related to propionate, such as
B—alanine; and (Z) to devise techniques of chemical degradation of the
pyridine ring of nicotine which will allow the isolation of its specific
carbon atoms, especially those at positions 2 and 6. The first part of
the problem was to be approached by determining the extent of incorpora-
tion of radioactivity into the pyridine ring of nicotine from appropriately
labeled propionate and B-alanine. This might be done using methods of
degradation already devised. It was hoped that the second part of the
problem could be resolved by investigating sequences of organic reactions
which might permit isolation of carbon atoms 2 and/or 6 unambiguously
from the pyridine ring of nicotine.

Successful completion of either or both of these investigations
could be important steps in the solution of the problem of the biosynthesis
of pyridine compounds in general, and could aid in the clarification of
the role of that functionally unknown group of natural products, the

alkaloids .

EXPERIMENTAL AND RESULTS

EXPERIMENTAL AND RESULTS

The preparation, feeding and harvesting of the plants as described
in the following sections are a result of several years of experience in
carrying out feeding experiments with tobacco plants in this laboratory.
The procedures represent some methods used in other laboratories as
well as original variations devised by the author's predecessors in studies
on tobacco. These methods have been found over a period of time to be

the best and most successful under the prevailing laboratory conditions.

Preparation of Plants

 

The tobacco plants used in this study were Nicotiana rustica, var.

 

humilus, a strain with high nicotine content. The seeds were planted in
the greenhouse in flats containing Vermiculite, a commercially available
micaceous material used for support. The plants were cultivated as
described by Griffith (20). After two weeks the new seedlings were
transplanted to insure adequate room for growth. They were fed weekly
with a solution containing 1.0 g. MgSO4. 7HZO, 8. 3 g. Ca(NO3)7_.4HzO,
and O. 2 g. KZHPO4 and watered every other day.

When the plants reached a height of 12-15 cm. , they were removed
from the flats and the roots rinsed with tap water. The roots were then
clipped to approximately 1 cm. from the stem, since it had been shown
previously that the rate of nicotine synthesis is proportional to the rate
of growth of the tobacco roots (21). To allow new roots to form, the
plants were placed in 125 ml. Erlenmeyer flasks containing 50 ml. of an
aerated inorganic nutrient solution. This solution was composed of 2
parts water to 1 part of a stock solution containing 2.0 g. Ca(NO3).4HzO;
756 mg. MgSO4- 7HZO; 500 mg. KCl; 500 mg. (NH4)ZSO4; 5.6 mg. FeC13.
6HZO; and 50 mg. KHzPO4 in 2 liters of distilled water. The plants were

8

held in the flask by means of a cotton plug, and the flasks were wrapped
in shields of black construction paper to promote root growth and pre-
vent contamination of the roots with photosynthetic microorganisms (22).
The roots were allowed to grow under ordinary laboratory lighting
conditions, with daily addition of a few m1. of aerated water, for a
period of one to two weeks. By this time a healthy, voluminous root
system had usually developed, and at this stage the plants were ready

for the feeding of the radioactive materials.

Feeding of Plants

 

Healthy appearing plants with extensive new root systems were
selected and prepared for feeding as follows. The plants were removed
from the flasks, and the roots were rinsed with distilled water and then
blotted dry with tissue paper. The plants were then transferred to clean
125 ml. Erlenmeyer flasks containing the solution of radioactive material.
These solutions were prepared by dissolving the radioactive compound
in the required amount of distilled water, and adding non- radioactive
material when necessary to make the sample up to the standard concen-
tration. The feeding data are summarized in Table I.

In each experiment, three 10 ul. aliquots of the feeding solution
were taken, deposited on aluminum planchets, dried, and assayed to
determine the amount of radioactivity fed per plant.

In all cases except with B—alanine-3-C”, of which there was an
amount sufficient to feed only seven plants, twenty plants were fed in
duplicate runs of ten plants in each.

The roots of the plants were agitated in the bottom of the flask so
that there was maximum contact with the feeding solution. The flask
was fitted with a cotton plug and shielded with black paper as before,

then placed in a hood and exposed to artificial light supplied by two 36

10

Table 1. Administration of Labeled Precursors to Tobacco Plants.

 

 

 

Per Plant
Radio-
Ml. activity Amount fed Number of
Precursorl fed fedgic.) (moles x 105) plants
Propionate- 1 -Cl4 2. 0 5 2. 44 20
Propionate-3-C” 2. o 5 2. 44 25
fi-Alanine-Z-C” 2. o 5 2. 44 20
B-Alanine—3-C” 1. o 1 1. 22 7

 

1The radioactive propionate samples were purchased from Volk
Radiochemical C. , Skokie, Illinois; (3-alanine—2-C14 was purchased
from Research Specialties Co. , Richmond, California. (i-Alanine-3-Cl4
was synthesized by another worker in this laboratory from ethylchlor-
acetate and KCMN obtained from Volk, using the procedure of Fritzson
and Eldyarn (23).

inch, 30 watt fluorescent tubes and one 100 watt incandescent bulb placed
about 14 inches above the plants. This arrangement eXposed the plants
to a light intensity of approximately 200 foot-candles. Under these con-
ditions, the 2 ml. of feeding solution was observed to be completely
absorbed in a period of 3-4 hours. The plants were allowed to absorb
two additional 2 ml. portions of water to dryness, to insure maximum
uptake of the radioactive material. To each flask was then added 25 m1.
of the inorganic nutrient solution described previously. The plants were
given a few ml. of aerated water daily; the period of illumination was

12 hours per day. They were then left to grow under these conditions

for a period of 7 days before harvesting.

ll

Isolation and Purification of Nicotine

 

On the seventh day after the administration of the radioactive
compounds the plants were harvested and their nicotine isolated. This
was accomplished by removing the plants from the nutrient solution and
thoroughly rinsing the roots with water. Several 10 pl. aliquots were
taken from the remaining nutrient solution and plated on aluminum
planchets for counting to determine the amount of residual radioactivity,
if any, not absorbed by the plants. In no case was the activity of these
aliquots more than 20 counts per minute above background. Since this
amount of activity is negligible compared to the amount fed, it was con-
cluded that the radioactive compounds were completely absorbed by the
plants probably in the initial 3-4 hours.

The excess water was removed from the roots by blotting with
paper toweling, and the plants were cut into small pieces and dried under
a heat lamp at 800C. for 6 hours. The dried plant material was finely
ground in a mortar, weighed, and transferred to a micro-Kjeldahl flask
containing an amount of calcium oxide equal to 20 percent of the weight
of the plant material. Several drops of G. E. antifoam solution was
added, and the nicotine was separated by steam distilling the mixture.
The distillate was collected in 1-3 ml. of 6N HCl. The distillation was
continued until there was no precipitate with silicotungstic acid, a com-
pound which forms a white insoluble precipitate with nicotine and other

alkaloids (24). The distillate was concentrated in vacuo to a small

 

volume in a flash evaporator. The resulting solution was made basic
to litmus by the addition of 6N NaOH and then purified by azeotropic
distillation through a Widmer column (25), collecting the distillate in a
small volume of 6N HCl, until it gave no precipitate with silicotungstic
acid. This distillate was reduced to dryness in the flash evaporator.

The white solid residue of nicotine hydrochloride was dissolved in a small

12

amount of methanol, and an equal volume of a saturated solution of
picric acid in methanol was added. The yellow precipitate of nicotine
dipicrate was allowed to settle and then was isolated by filtration.

The solid was recrystallized from water, resulting in needles of yellow
nicotine dipicrate melting at 225—2700" 2 Yields were generally on

the order of 8-12 mg. of nicotine dipicrate per plant.

Dete rmination of Radioactivity

 

All measurements of radioactivity were made with either of two
instruments: a Nuclear-Chicago Model 192A scaler with Model D-47
proportional gas flow counter and a Nuclear-Chicago Model 192x sealer
(both from Nuclear-Chicago Corporation, Chicago 10, Illinois) with a
Tracerlab Model SC-16 proportional gas flow counter (Tracerlab, Inc. ,
Boston 10, Massachusetts). Which instrument was used was determined
by availability at the time of counting, but the same counter was used
throughout a given experiment to eliminate errors caused by differences
in counting efficiency between the two counting apparatus. The pro-
portional counting gas, 90 percent argon and 10 percent methane, was
a product of Tracerlab, Inc. The counting system was determined to
be 40 percent efficient using a sodium carbonate-Cl4 B-ray standard
supplied by the National Bureau of Standards.

The radioactive feeding solutions were counted by plating 10 111.
of the solution on an aluminum planchet and evaporating the water with
a heat lamp. The compound was then assumed to be in a layer of infinite
thinness, so that no self-absorption corrections were applied. The other
samples to be counted, with the exception of barium carbonate, were

ground in a small agate mortar. Approximately 30-40 mg. of the ground

 

lAll melting points were determined by the capillary method.
A stem correction is not included.

zRecorded melting point is 224—250C (26)°

13

material was then placed in an aluminum planchet, distributed as evenly
as possible, and the surface then made as smooth as possible by tamping
with a flat surface. Barium carbonate samples were plated directly
from suspension by vacuum filtration through a stainless steel funnel
onto a Whatman No. 2 filter paper disc having an area of 2. 83 sq. cm.
This formed a disc containing 30-40 mg. barium carbonate which was
then dried at 1050C for at least one hour. The dried barium carbonate
was then transferred to an aluminum planchet, weighed, and counted.

To eliminate possible errors caused by self-absorption differences
between samples of different compounds, all compounds, after counting,
were oxidized to carbon dioxide, which was then counted as barium
carbonate. This combustion was accomplished in the following manner:
approximately 10 mg. of the radioactive material was mixed in a Pyrex
combustion tube with a small amount of a solid mixture containing 2
parts potassium iodate to one part potassium dichromate. The tube was
connected to a series of three traps containing saturated barium hydroxide,
and the system was flushed with nitrogen. Two ml. of an oxidizing mix-
ture (66 parts fuming H2804z33 parts conc. H3PO4:1 part K103) was added
to the tube. The mixture was heated, slowly at first, and then vigorously
for one minute after iodine fumes were observed. The heat was then
removed and the nitrogen stream continued until all the carbon dioxide
had precipitated as barium carbonate in the traps. This barium carbonate
was then plated by filtration, dried, weighed, and counted as previously
described. Unless otherwise designated, all counting results are reported
as barium carbonate obtained from combustion. All reported specific
activities were corrected for background, for efficiency of the counting
system, and for self absorption as suggested by Yankwich, 3t a_._l.(27).

The incorporation of radioactivity from the four carbon-l4 labeled

compounds into nicotine is given in Table II.

14

Table II. Incorporation into Nicotine of C” from Several Precursors.

 

 

Nic otine dipicrate

 

 

Specific Number

activity of of Specific '
Compound Fed precursor plants Yield activity Dilutiona

cpm/mMolxlO“5 mg. cpmeMol
Propionate-l-C” 3. 28 10 100 3. 27x104 10030
Propionate-l-C” 3,28 10 150 4.27x104 7681
Propionate-3-Cl4 2.65 10 99 2.99x105 888
Propionate-3-CM 2. 65 10 75 3. 49x105 764
Propionate-3-C14 15. 4 5 57 2. 22x106 692
B-Alanine-Z-C” 3. 34 10 64 1. 25x106 267
B-Alanine~2~Cl4 3. 34 10 118 0. 61x106 547
6-Alanine-3-c” 2.09 7 64 2.69x105 777

 

*
Dilution was calculated by dividing the specific activity of the pre-
cursor fed by the specific activity of the nicotine dipicrate.

Comparison of the calculated dilutions for the four metabolites
shows that when these compounds were fed under essentially identical
conditions, B-alanine-Z-C” was incorporated into nicotine to the greatest
extent, fi-alanine-3-C” and propionate—3-C“ to a somewhat lesser ex-
tent, and propionate-LC” only very slightly. The twofold difference
between the nicotine dipicrate specific activities in the two fi-alanine-Z-
C” experiments is attributed to a possible difference in condition of the
tobacco plants used. It was observed that, in the case where the lower
specific activity (greater dilution) was obtained, the plants were larger
in size and had a greater dry weight than the other plants; and, possibly
more important, the yield of nicotine dipicrate in the former case was

nearly twice as great as in the latter. This suggests that the larger

15

plants might have been at a later stage of development when fed, and
the nicotine synthesized from the radioactive precursor therefore diluted
with previously synthesized, endogenous nicotine.

Thus it was demonstrated that, of the four compounds tested,
three-propionate-3-C14, B-alanine-Z-CM, and [3-alanine«--3-Cl4 were
precursors of nicotine in tobacco, while the radioactive carbon from
propionate-1--C14 was utilized to such a slight extent that its incorporation

into nicotine was considered insignificant.

Degradation of Radioactive Nicotine

 

The demonstration that several compounds were precursors of the
nicotine molecule next led to experimentation to determine which of these
compounds were precursors of the pyridine moiety of nicotine. The
chemical degradation employed established procedures (2, 28) and is
based on a permanganate oxidation of nicotine originally described by
Laiblin (29). A recent study of this procedure demonstrated that the
products of the oxidation were as follows: nicotinic acid (40-50% of theory);
potassium carbonate, representing carbons 3, 4, 5, and the N-methyl
carbon of the pyrrolidine ring (30-45%); ammonia (90%); and small
amounts of other unidentified products (28).

The weighed radioactive nicotine dipicrate was suspended in water
and the solution made basic to litmus with 6M NaOH. The solution was
then azeotropically distilled through a Widmer column (30) until no more
nicotine appeared in the distillate, as determined by testing with silico-
tungstic acid. To insure yields of oxidation products which would be
adequate for counting, the radioactive nicotine was usually diluted with
non-radioactive nicotine at this stage. Dilutions were desired which
would yield 300-500 mg. of nicotine for degradation, but in no case were

dilutions made which would result in nicotine dipicrate having less than

16

200 counts per minute per planchet. An aliquot of this solution contain-
ing approximately 20 mg. of nicotine was taken in order to determine

the activity of the diluted nicotine. This was done by acidifying the
solution with 3M HCl, evaporating it to dryness, taking the residue up

in a small amount of methanol, and adding pic ric acid-saturated
methanol to precipitate the diluted nicotine dipicrate. The dipicrate

was then recrystallized from water and counted as previously described,
first as nicotine dipicrate and then as barium carbonate after combustion.

The remaining solution was then treated with a 4 percent solution
of potassium permanganate, adding 5 ml. portions of the permanganate
at 3-5 minute intervals with swirling. Addition was continued until a
weight of potassium permanganate equal to six times the weight of the
nicotine present had been added. The solution was then allowed to stand
overnight on a steam bath, to facilitate completion of the reaction and
flocculation of the manganese dioxide formed.

After at least 10 hours on the steam bath, the solution was allowed
to cool to room temperature and then filtered through a fine sintered
glass funnel. The residue was washed repeatedly with hot water. The
washings and filtrate were combined and taken to dryness on a flash
evaporator. The white residue from the evaporation was taken up in
30 ml. of water, and then, in an air-tight system, the solution was
treated gradually with excess 6M nitric acid. The evolved carbon dioxide
was absorbed in carbonate-free 0. 2M sodium hydroxide. The closed
system was allowed to stand for approximately 45 minutes. Barium
carbonate was then precipitated from the basic solution by the addition
of excess saturated barium chloride, and the precipitate was then plated
by filtration, dried, and counted. This barium carbonate contains carbon
representing carbons 3, 4, 5, and the methyl carbon from the pyrrolidine

moiety of nicotine .

17

The acid fraction from which the carbon dioxide had been evolved
was then evaporated to dryness, taken up in water, re-evaporated, and
this process repeated three times. The third time, before evaporation,
the aqueous solution was made basic to litmus with concentrated
ammonium hydroxide, and then evaporated. The white residue was
dissolved in 20 ml. of water, and to this neutral solution was added an
excess of 0. 1M silver nitrate solution. The mixture was allowed to
stand one-half hour at room temperature to allow coagulation of the cloudy
white precipitate of silver nicotinate. The precipitate was then removed
by filtration and washed with water. A slurry of the silver salt in 25 ml.
of hot water was prepared and hydrogen sulfide was bubbled through
this suSpension. The resulting mixture was heated on a steam bath
and then filtered under reduced pressure followed by a wash of the black
silver sulfide with hot water. The filtrate and washings were taken to
dryness. The whitish residue of nicotinic acid was transferred to a sub-
limation apparatus and twice purified by vacuum sublimation at 1 mm.
Hg and loo-120°C. The sublimate was then recrystallized from ethanol.
The white flakes of nicotinic acid, melting at 234-350C, l were plated and
counted. A 5-10 mg. sample of the nicotinic acid was usually oxidized
to carbon dioxide and counted as barium carbonate.

After counting, the nicotinic acid (25-50 mg.) was decarboxylated
by dry distillation of its calcium salt. This was accomplished by heating
nicotinic acid with an excess of calcium oxide slowly to 2000C and then
rapidly to 400°C (31). During the next 25 minutes the distillate of
pyridine was collected in picric acid-saturated methanol as the pyridine
picrate. This was recrystallized from water to yield yellow needles
having a melting point of 164-165OC.Z This product was counted, first

as pyridine picrate and then as barium carbonate after combustion.

 

1Reported melting point is 2320C (22).

ZReported melting point is 1670C.

18

The residue, containing the carboxyl carbon of nicotinic acid as
calcium carbonate, was acidified with 6M hydrochloric acid and the
liberated carbon dioxide was swept in a stream of nitrogen into a
saturated solution of barium hydroxide. The resulting barium carbonate,
representing carbon 2 of the pyrrolidine ring of nicotine, was plated and
counted as previously described.

The results of this complete degradation scheme carried out on
the nicotine obtained from tobacco plants fed propionate-3-C“, and

T3-alanine-3-C14 are shown in Table III.

Table III. Results of a Partial Degradation of Radioactive Nicotine.

 

Propionate-3-Cl4 (3-Alanine-3-Cl4
Compound Spec. act. Spec. act.
cipm/melo“4 Percent cpm/melO'“4 Percent

 

 

 

Nicotine dipicrate 7. 00 100 1. 32 100
Nicotinic acid 3. 58 51.1 0. 97 73. 5
Barium carbonate* 2. 96 42. 3 0. 61 46. 3
Barium carbonate** 2. 73 39.0 0.44 33. 5
Pyridine picrate 0. 16 2. 3 0 0

 

*
Barium carbonate from the 2-., 3-, 4-, and N-methyl carbons of the

pyrrolidine ring. The specific activity was multiplied by 4.

3::
From the carboxyl group of nicotinic acid--carbon 2 of the
pyrrolidine ring.

The figures reported in Table III represent an average of the results
from at least two experiments in each case.

Although (i-alanine-Z-C14 was incorporated into nicotine to an extent
sufficient to warrant investigation by the described degradation procedure,

this procedure was not carried out. The reason for this is that Dawson

l9

and co-workers1 had already accomplished this degradation. Although
the detailed figures are not available, the over-all distribution pattern
was similar to that obtained with propionate-Z-CM, showing that approxi-
mately 41 percent of the total nicotine activity was located in the
pyridine ring. Subsequent work by Griffithz demonstrated that of the
activity in the pyridine ring, some 40 percent was located in the
3-position. The significance of these data will be commented on as part

of the discussion.

Attempted Degradation of. the Pyridine Ring of Nicotine

 

The chemical stability of the pyridine ring is a widely recognized
phenomenon among organic chemists. Generally speaking, the pyridine
ring is more resistant to oxidative disruption than the benzene ring (32).
The chemical degradation of the pyridine ring of nicotine attempted in
this project is based on the discovery by Tschitschibabin that, although
the phenylpyridines are oxidized by acid permanganate to the pyridine-
carboxylic acids, with alkaline potassium permanganate it is the pyridine
ring which is attacked. 2-Phenylpyridine, for example, gives a 63
percent yield of benzoic acid on oxidation with alkaline permanganate (33).
Apparently, under proper conditions, phenylpyridines can be oxidized
in a manner which will destroy the pyridine rather than the benzene
moiety. Thus it was hoped that if a phenyl substituent could be introduced
into the pyridine ring of nicotine at the positions of particular interest
to this study--namely, the 2 and 6 positions--these phenyl nicotines
could be oxidized to benzoic acid using Tschitschibabin's alkaline per-
manganate method. The carboxyl carbon of this benzoic acid would

represent the carbon atom of the pyridine ring at which the phenyl

 

1’ 2Personal communication; unpublished data.

20

substituent was originally located. Thus, this approach involved two
distinct problems: 1) obtaining phenyl substituents at particular positions
of the pyridine ring of nicotine, and 2) oxidizing these derivatives to
benzoic acid.

To accomplish the first objective it was determined to take
advantage of another well-known property of pyridine, the relative ease
with which it is attacked by nucleophilic agents at the 2 position, also
discovered by Tschitschibabin (34). Thus Tschitschibabin had shown
that nicotine, on treatment with sodamide, could be converted to a mixture
of 2- and 6-aminonicotine in equal amounts, in approximately 60 percent
yield over-all (35). Separation of these two isomers was effected by
taking advantage of their different solubilities in water. . Once the 2-
and/or 6-aminonicotines were obtained, an attempt could then be made
to replace these amino substituents with phenyl groups. Several potential
methods for doing this are described in the literature, analogous methods
in which amino substituents on benzene rings are replaced by phenyl or
pyridine groups. The most likely of these was a method of preparing
unsymmetrical biaryls by the diazo reaction or by the nitrosoacetyl-
amine reaction, described by Bachmann and Hoffman (36), or variations
of the above reactions, including those devised by Elks, e_t a_._l. (37).

Thus, the original general plan for degradation of the pyridine
ring of nicotine involved the amination of nicotine to form the 2- and
6-amino derivatives; replacement of these amino groups with phenyl

groups; and oxidation of the phenyl derivatives to benzoic acid.

Preparation of Sodium Amide (38)

 

To a mechanically stirred mixture of 25 mg. of finely powdered
ferric nitrate hexahydrate in 50 ml. of liquid ammonia was added 0. 1 g.

of clean sodium metal. Dry air was bubbled through the solution until

21

the blue color was discharged. The bubbling of air was then discontinued,
and the remainder of 2. 3 g. (0. 1M) of sodium was added in small pieces.
A reaction began at once, and in 30 minutes, the blue color was replaced

by gray, indicating completion of the reaction.

Amination of Nicotine

 

While stirring the above reaction mixture, about 100 ml. of
anhydrous xylene was added over a period of 30 minutes, simultaneously
allowing the excess liquid ammonia to evaporate. When the addition of
the xylene was completed and all the ammonia had evaporated, the stirring
assembly was removed, a condenser fitted with a calcium chloride drying
tube put in its place, and 8. 2 ml. (0. 05M) of freshly distilled nicotine
was added. The reaction vessel was then heated slowly to 135°C in an
oil bath, bringing the xylene in the mixture to reflux. The reaction was
allowed to proceed at this temperature for 10 to 12 hours.

After refluxing, the reaction mixture, now dark brown in color,
was allowed to cool to room temperature, and any excess sodamide

was very carefully decomposed by the addition of ice water. An equal

 

volume of 6N hydrochloric acid was added and the product extracted from
the xylene into the lower aqueous layer as the hydrochloride. The aqueous
layer was separated and neutralized with 6M sodium hydroxide. The
neutral, or slightly basic, solution was saturated with sodium carbonate
and then repeatedly extracted with ethyl ether. The ether fractions were
combined and the ether removed by evaporation on a steam bath. The
residue was a viscous dark oily liquid, with a pungent pyridine-like

odor. Approximately 5 volumes of water was added to this residue with
shaking, and the mixture was allowed to stand overnight in a refrigerator.
A white solid, insoluble in the water, formed. This solid was removed

by filtration and recrystalized from ligroin and twice from water, using

22

Norite to decolorize the product. The product was 2. 5 g. (28 percent
of theoretical) of white plates of 2—aminonicotine melting at 123-124OC.l
This product displayed the following properties:

a. an elemental analysis gave the following results for

CIOHISN3:2
Calculated: C, 67.76; H, 8.53; N, 23.71
Found: C, 67.54; H, 8.58; N, 23.53

b. formed a picrate in picric acid saturated methanol which, on

recrystallization from water, melted at 225«-227OC.3

c. with the nickel-salicylaldehyde reagent yielded a delayed,

copious white precipitate, indicative of primary aromatic

amines (40).

d. the infrared spectrum, determined in carbon tetrachloride

on a Perkin-Elmer Model 21 instrument, was similar to that

of nicotine (41), with additional peaks at 2. 90, 3. 05, and

6. 25 u, all three of which are indicative of primary aromatic

amines.
These results indicate that the first product isolated from the amination
of nicotine was the same as that obtained by Tschitschibabin, 2-amino-
nicotine.

The aqueous residue from which the crude 2-aminonicotine had
been precipitated, and which presumably contained 6-aminonicotine,
was saturated with potassium carbonate and repeatedly extracted with
ether. The ether was removed by evaporation, leaving a dark oily
residue. . Numerous attempts to purify this residue by treatment with

Norite and crystallization from various solvents failed to produce any

 

o lTschitschibabin reported melting points of 1250C (35) and 124-
125 C (39).

ZElemental analyses were carried out by Spang Microanalytical
Laboratory, Ann Arbor, Michigan.

3Tschitschibabin reported 223-2250C (39).

23

crystalline material like the 6-aminonicotine described by Tschitschibabin.
[This difficulty in crystallizing 6-aminonicotine was noted by Tschitschibabin
(35).] Therefore, the 6-aminonicotine was isolated as the hydrochloride
by dissolving the residue in absolute ethanol, adding a few drops of concen-
trated hydrochloric acid, and allowing the mixture to stand in a refrigerator.
After several hours, the appearance of white crystals was noted. When
recrystallized from absolute ethanol, these crystals were in the form of
prisms melting at 172-173OC. The yield was 2.0 g., or 23 percent of
theoretical.

C10H15N3. 2HC1 -- Calculated: C, 48.00; H, 6.86; N, 16.80.

Found: C, 47.61; H, 7.09; N, 16.34.

Since a melting point for 6-aminonicotine hydrochloride has not
been reported, and since many of the properties of 2- and 6-aminonicotine
are similar-elemental analysis, picrate melting point, infrared spectrum--
the latter product was converted to a derivate to ascertain that it was
indeed the hydrochloride of 6-aminonicotine and not that of 2-aminonicotine
which had been isolated. Thus, the following diazotization was carried
out: to a solution of 250 mg. of 6-aminonicotine hydrochloride in 2.0 ml.
of 5 percent sulfuric acid was added a solution of 0. 83 g. of sodium
nitrite in 1 ml. of water dropwise with shaking. The solution was then
heated to 500C, neutralized with sodium carbonate and evaporated to
dryness. The solid residue was extracted with boiling ligroin. On
evaporation of the ligroin, an oil remained which solidified on cooling.
Recrystallization from ligroin and treatment with Norite gave white
crystalline prisms of 6-oxynicotine melting at 104-1050C. Tschitschibabin
reported the melting point of 6-oxynicotine as 103. 5-104OC (35), and that
of 2-oxynicotine as 121-1230C (42). Thus the second water-soluble
product isolated as the hydrochloride from the amination of nicotine was

clearly 6 ~aminonicotine .

24

Attempted Diazotization and Couplingof 2-Aminonicotine

 

The general method employed was that described by Bachmann and
Hoffman (43). It involves replacing the amino group of aromatic amines
by aryl groups with the formation of unsymmetrical biaryls. This is
accomplished by reaction of the aryldiazo hydroxide or acetate, obtain-
able from the diazotized amine, with aromatic nuclei. A typical reaction
was carried out as follows: to 0. 55 g. (3. 11 mM) of 2-aminonicotine
was added, with stirring, enough concentrated hydrochloric acid (0. 26 ml.)
to form the hydrochloride. Then 0. 21 ml. (3. 73 mM) concentrated
sulfuric acid in 5 ml. water was added and the mixture cooled to 00C in
an ice-salt bath. After cooling, 0. 236 g. (3.42 mM) sodium nitrite in
1 ml. water was added dropwise over a period of 15-20 minutes. The
temperature was maintained at O-SOC. After addition was complete,
stirring was continued for one-half hour. At the end of this time, 12 m1.
benzene was added to the reaction and the temperature allowed to rise to
100C to keep the benzene from freezing. Then 1. 7 ml. of 6M sodium
hydroxide was added dropwise until the aqueous layer remained alkaline
(approximately pH 10 by Fischer Alkacid test paper). As each drop of
base was added, a white precipitate formed in the aqueous layer, and
this was dispelled-apparently extracted into the benzene layer--upon
continued stirring. When addition of the base was complete, the mixture
was stirred for 2 additional hours in the cold and then shaken for 40
hours at room temperature. At the end of this time, the benzene layer
was removed, washed three times with water, and the benzene removed
on the flash evaporator. When the tan solid residue was recrystallized
from ligroin, it displayed a melting point of 123-124OC. It formed a
picrate melting at 223-2240C, and gave rise to infrared and ultraviolet
spectra identical to those of 2-aminonicotine. Therefore, the reaction
was unsuccessful, and approximately 60-70 percent of the starting

material, 2—aminonicotine, was recovered.

25

This reaction, the conversion of Z-aminonicotine to Z-phenyl-
nicotine, was attempted eight times in all, making use of variations
described by Bachmann and Hoffman and also by Elks, e_t a}. , in subse-
quent work on analogous reactions (37). These variations involved
running the diazotization reaction at higher temperatures, using hydro-
chloric acid in place of sulfuric acid in the medium, replacing sodium
hydroxide with sodium acetate, and longer reaction periods. In no case
was there any evidence that the reaction was successful. Each time,
standard procedures for isolating a nitrogen containing, basic product
such as 2-phenylnicotine yielded only a large percentage of the starting
material. After eight attempts, this general reaction procedure was
abandoned.

The next attempt to replace the amino group with a benzene ring
was again a diazotization procedure (44). In general, the method in-
volves diazotization of two different aromatic amines at low temperatures,
mixing them in the presence of cuprous ion and then raising the tempera-
ture of the mixture. Depending on the nature of the amines, they may
couple with the liberation of nitrogen, presumably by a free radical
mechanism. This method was applied to 2-aminonicotine in the following
manner: cuprous chloride was purified by washing it first with a 1-20
sulfuric acid to water mixture and then with glacial acetic acid. It was
then dried at 1050C and stored in the dark until used. Crude aniline was
purified by distillation from zinc dust. The subsequent steps were
carried out in a cold room at a temperature of 2-4OC, the reagents all
having been stored at this temperature previously.

A. Cuprous chloride (0.72 g.) was suspended in 6 ml. water and

1. 26 ml. of 6N sodium hydroxide was added dropwise with
stirring. The resulting orange hydrate was washed twice

with water by decantation. After a third washing, the hydrate

C.

26

was filtered and dried by suction. This hydrate was suspended
in a solution of 1.4 ml. concentrated ammonium hydroxide and

1. 5 ml. water. The mixture assumed a blue color.

. 2-Aminonicotine (250 mg. , 1.4 mM) was dissolved in 3.6 ml.

water and 1. 2 ml. 6N hydrochloric acid. To this was added
slowly with stirring 115. 8 mg. sodium nitrite (1. 7 mM). The
yellowish solution was kept in the dark for one-half hour while
part C was carried out.

Aniline (0. 24 ml. , 2. 7 mM) was dissolved in 4 ml. water and
O. 21 g. sodium nitrite (3.0 mM) was added. To this mixture
was added--dropwise, with shaking--0. 93 ml. concentrated

hydrochloric acid. The solution turned bright orange in color.

. The two diazonium salt solutions were mixed together and

added dropwise to the well-stirred, cold, alkaline copper sus-
pension. The blue color changed to dark brown, accompanied
by considerable frothing. When addition was completed, the
mixture was stirred for fifteen additional minutes, then removed
from the cold room and heated to boiling. On heating, the mix-
ture first became yellow, then green, and had a phenol—like
odor. After standing overnight in the dark, the mixture was
filtered through a fine sintered glass funnel. The green filtrate
had a pH of 4. 7 as determined on a Beckman Model H-2 glass
electrode pH meter. This acid filtrate was extracted with

ethyl ether for 24 hours using a liquid-liquid extractor. The
yellow ether layer was reduced to dryness and the viscous brown
liquid residue was dissolved in chloroform. This chloroform
solution, when treated with a formaldehyde-sulfuric acid
reagent, gave rise to a reddish-brown ring at the interface, a

positive test for biphenyls (45).

27

The original aqueous solution was made alkaline with 6N sodium
hydroxide (approximately pH 10) and again extracted with ether in the
liquid- liquid extractor for 24 hours. The ether layer was reduced to
dryness. The tan residue, after recrystallization from water, had a
melting point of 121-1220C and all the properties-picrate, u. v.
spectrum--of the starting material, 2-aminonicotine.

This method was attempted a second time, this time attempting
to diazotize the 2-aminonicotine at room temperature before mixing
with the diazotized aniline and copper solutions. Again the only isolable
products were the starting material and traces of biphenyl. , It was
apparent that the failure of this method, and presumably the failure of
the Bachmann and Hoffman method as well, was caused by the fact
that 2-aminonicotine was not diazotizable at all or only slightly so.

This will be further discussed in a later section.

Attempted Phenylation of Aminonicotines

 

It was at this point in the investigation, when the probability of
replacing the amino group in 2- or 6-aminonicotine with a phenyl group
seemed increasingly unlikely, that a paper appeared which suggested a
different approach for obtaining 2- and 6-phenylnicotine. In studies on
the orientation of the entering phenyl substituent in the addition of phenyl-
lithium to various pyridine derivatives, Abramovitch, e_t a_t_l. (46, 47)
prepared a mixture of 2- and 6-phenylnicotines by the reaction of nicotine
with phenyllithium. It was not deemed practicable to use this reaction
in the present study because the procedure used by Abramovitch to
separate the mixture of 2- and 6-pheny1nicotines involved a method of
preparative vapor phase chromatography not available at the time this
work was in progress. Thus separation of these isomers might prove

difficult; and, in addition, this separation might be unnecessary if

28

Abramovitch's method could be applied to Zuaminonicotine and 6-amino-
nicotine to yield 2-amino-6-phenylnicotine and 2~phenyle6«aminonicotine,
respectively. Therefore, the following attempts were made to prepare
the above compounds, using reaction conditions similar to those of
Abramovitch.

Ten mg. of lithium metal was finely cut and suspended in 5 m1.
of anhydrous ethyl ether under a stream of dry nitrogen. To this was
added 0. 12 ml. of anhydrous bromobenzene in 5 ml. of dry ether, slowly
and with stirring. When the reaction was complete, 131 mg. of 6-amino-
nicotine in dry ether was added slowly. The 6-aminonicotine was
obtained from 158 mg. of the hydrochloride by treating with excess 6M
sodium hydroxide, extracting the alkaline solution repeatedly with ether,
and drying the ether solution over anhydrous magnesium sulfate. The
ether in the reaction mixture was evaporated, simultaneously replacing
it with anhydrous toluene. The mixture was then heated to boiling,
continuing the stirring and nitrogen stream. Refluxing was continued
for 8 hours, after which the mixture was allowed to cool and ice water

added cautiously, to decompose any excess phenyllithium. The water

 

layer was acidified, the layers separated, and the toluene subsequently
extracted with dilute acid. The combined acid washings were made
alkaline by the addition of solid sodium hydroxide and repeatedly extracted
with ether. The ether layers were combined, dried over magnesium
sulfate, and the ether removed by evaporation. The residue, a dark
oily substance, was identified by its ultraviolet absorption spectrum and
the melting point of its hydrochloride to be the starting material,
6-aminonicotine.

This reaction was attempted 5 additional times, twice using
Z-aminonicotine as the starting material, and changing such variables
as the amount of materials, the solvent--to xylene, (allowing a higher

reaction temperature), and the drying agent to reduce adsorption of

Z9
6-aminonicotine. The only substances which were isolated from these
reaction mixtures in each case were the starting materials. The signifi-

cance of this result will be discussed in a later section.

Phenylation of Nicotine

 

At this point, it was decided to approach the problem by direct
phenylation of nicotine with phenyllithium using the exact procedure of
Abramovitch, St a_l. (46), with the hope that a method could be found to
separate the 2- and 6-phenyl isomers which were formed. The instru-
ments for the gas chromatographic procedure were not at hand, but it
was hoped that the isomers would be separable by other means--fractional
distillation, fractional crystallization of the picrates, etc. In addition,
another possibility was suggested by Abramovitch's work. He had shown
that both 2- and 6-pheny1nicotine could be oxidized with neutral per-
manganate to the corresponding phenylnicotinic acids. The resulting
acids appeared to differ enough in properties that separation of a mixture
of these, obtained by permanganate oxidation of the mixture of 2- and
6—phenylnicotines, might be feasible.

The procedure used was similar to that previously described in
attempts to phenylate 2-aminonicotine and was the same as that described
by Abramovitch (46). One g. of clean lithium was allowed to react with
12 m1. of bromobenzene in 100 ml. ether under anhydrous conditions.

To this mixture was added dropwise 12 g. (11. 9 ml.) of nicotine in 30
m1. of ether, with stirring under nitrogen. The solution turned from
a cloudy white to orange and finally to a deep reddish brown. The ether
was removed by evaporation with the simultaneous addition of 50 ml. of
dry toluene. The toluene solution was kept at reflux for 8 hours. After
this time, the dark brown mixture was allowed to cool and ice water was

carefully added. There was some mild bubbling, indicating an excess of

30

phenyllithium. The layers were separated and the aqueous layer
extracted several times with ether. The ether extracts were combined
with the toluene layer and this combined organic mixture was extracted
with 3N hydrochloric acid. The acid layer which had assumed a dark
green color, was made strongly alkaline with solid sodium hydroxide
and then extracted repeatedly with ether. The resulting dark red ether
solution was dried over potassium hydroxide pellets after which the
ether was removed by evaporation. The residue, a dark, pungent,
viscous liquid, was distilled under reduced pressure, collecting the

following fractions:

(i) 47-53OCJO. 5-0.4mm 4.0 g. colorless liquid
(ii) 119-135 C/0.3mm 3.5 g. yellow liquid
(iii) residue 4.0 g. deep red resin

Later repetitions of this reaction showed that fraction (ii), if the
conditions of temperature and pressure were carefully controlled, might
be further separated into two fractions. According to Abramovitch,
the first of these was presumably 2-phenylnicotine, and the second 6-
phenylnicotine. Sample data from one of these distillations showed the

following fractions:

(1) 45-48°C 0. 3mm 4.0 g. colorless liquid
(ii) 120-125 C/0.3mm 1.5 g. yellow liquid
(iii) 130-135°c/0.3mrn 2.0 g. yellow liquid
(iv) residue 4.0 g. brown resin

Fractions (ii) and (iii) were identified as 2- and 6-phenylnicotine,
respectively, by their ultraviolet absorbtion spectra, determined in
ethanol on a Cary Model 11 recording spectrophotometer. These spectra
were identical to those reported by Abramovitch (46).

2-phenylnicotine: xmax 264 mp. (e: 8. 50x103).

6-phenylnicotine: imax 249 mu (6 = 17. 90x103); 280 mu (6: 13. 57x103).

The picrates had melting points as follows:

31

2-phenylnicotine picrate: 213n214OC.
6-phenylnicotine picrate: 167~1690C.1

Elemental analysis: C16H18N2
Calculated: C, 80.62; H, 7.63; N, 11.75
Found (Z-phenylnicotine): C, 80.57; H, 7.77; N, 11.69
Found (6wphenylnicotine): C, 80.44; H, 7.86; N, 11.88.

Oxidation of Phenylnicotine s

 

The oxidation of nicotine to nicotinic acid using neutral potassium
permanganate has been previously described. The oxidation of phenyl-
pyridines with neutral or alkaline permanganate to benzoic acid as
described by Tschitschibabin (33) involves the same procedure except
that it is carried out under conditions of reflux. Thus it was assumed
that using Tschitschibabin's procedure on 2- or 6-phenylnicotine should
give benzoic acid as the main product. The reaction was carried out as
follows: purified 6-phenylnicotine (0. 5 grams) was mixed with 10 ml.
water. Then 5. 0 grams of potassium permanganate dissolved in 100 ml.
water was added and the mixture refluxed for 9 hours. The reaction
mixture was allowed to cool and filtered through a medium sintered
glass filter, followed by a wash of the dark residue to manganese dioxide
with hot water until the washings were colorless. The filtrate was purple,
indicating an excess of permanganate. Therefore, the filtrate was
treated with ethanol on a steam bath until the purple color had disappeared,
thus decomposing the excess permanganate. This mixture was again
filtered, the precipitate of manganese dioxide washed with hot water,
and the combined washings and filtrate evaporated to dryness in a flash
evaporator. The residue was dissolved in a small amount of water, and

glacial acetic acid was added until the solution was acid to pH paper.

 

o 1Abramovitch (46) reports melting points of 211-213OC and 166—
167 C for the picrates of 2- and 6-pheny1nicotines, respectively.

32

This addition was accompanied by vigorous effervescence, and when

the solution became acidic, a fine white precipitate had formed in the
yellow solution. After cooling the mixture, the precipitate was removed
by filtration and recrystallized from aqueous ethanol. The white crystal-
line product had a melting point of 230-2310C.l Thus the product was
obviously not benzoic acid (m.p. 121—1220C). This product was identi-
fied as 2-pheny1pyridine—5~carboxylic acid (6-phenylnicotinic acid) from
the following data:

1. melting point

2. mixed melting point with sample of authentic 6-pheny1nicotinic
acid: 229-230°c ’-

3. the ultraviolet spectrum determined in methyl alcohol was
identical to the spectrum of an authentic sample of 6~phenyl-
nicotinic acid:
imax 260 mu (6 = 1. 20x104); 288 mu (6 = 1.71x104)

4. elemental analysis:

Calculated: C, 72.34; H, 4.56; N, 7.03.
Found: C, 72.18; H, 4.61; N, 7.17.

The oxidation of 2-phenylnicotine was carried out under the same
conditions as that of 6-pheny1nicotine. The product of this permanganate
oxidation, a white crystalline solid, was identified as 2-phenylpyridine-3-
carboxylic acid (2-phenylnicotinic acid) from the following data:

1. melting point: 167-1690C 3

2. mixed melting point with an authentic sample of 2-phenyl-

nicotinic acid: 166- 168°C

 

lAbramovitch reports a melting point of 232°C for 6- henylnicotinic
acid (46). Benary and Psille report a melting point of 229 C (48).

Z6-Phenylnicotinic and 2-phenylnicotinic acids prepared from the
corresponding phenyl picolines were graciously supplied by Dr. R. A.
Abramovitch, Univ. of Saskatchewan, Saskatoon, Canada.

3Abramovitch (46) reports a melting (point of 167-1690C for this
acid, and Ishiguro, e_t a_1_l., report 168-169 C (49).

33

3. the ultraviolet absorption spectrum in methanol was identical

to that of an authentic sample of Z—phenylnicotinic acid:

1 : 247 mu (6 = 12.38x103).
max
4. elemental analysis:

Calculated: C, 72.34; H, 4.56; N, ..7.03.
Found: C, 72.20; H, 4.62; N, 7.14.

In neither case of permanganate oxidation of phenylnicotines was
there any evidence of the presence of benzoic acid, indicating that the
conditions employed were not vigorous enough to oxidize either the
pyridine or the benzene ring.

Subsequent attempts to oxidize the pyridine rings of both 2- and
6-phenylnicotinic acids with permanganate under a variety of conditions—-
neutral medium, alkaline medium, longer periods of reflux-~were all
unsuccessful. Procedures designed to isolate organic acids from the
reaction mixtures all yielded high percentages of the starting materials,
and in no case was there any evidence--spectrophotometric or other-
wise-"of the presence of the desired product, benzoic acid.

Therefore, this general method of degradation for the 2 and 6
positions of the pyridine ring of nicotine was abandoned. Possible explana—

tions for the failure of this method are discussed in the following section.

DISCUSSION

34

DISCUSSION

Attempted Degradation of the Pyridine Ring of Nicotine

 

The preparation and characterization of 2- and 6-aminonicotine
require little comment. These compounds were prepared and identified
by Tschitschibabin more than twenty-five years ago, and the same
procedures were employed in this work. The results in this study were
similar to those of Tschitschibabin, even to the difficulty in crystal-
lizing 6-aminonicotine. For this reason, 6-aminonicotine was isolated
as the dihydrochloride, and since the melting point of this compound
has not been reported, its identity was verified by elemental analysis,
picrate melting point, and conversion to a derivative, 6-hydroxynicotine.
The 6-hydroxynicotine was identified by melting point and ultraviolet
absorption spectrum (50). The fact that 6-hydroxynicotine can be pre-
pared by diazotization of 6-aminonicotine--and thesame applies to
Z-hydroxy- and 2-aminonicotine, as shown by Tschitschibabin (42)--
are significant in the following discussion of the coupling attempts.

The reason for the failure of the attempts to replace the amino
groups of 2- and 6-aminonicotine with phenyl groups by the diazoti-
zation method of Bachmann and Hoffman is not obvious. An explanation
is suggested by Barnes (51) when he points out that, while the diazoti-
zation of 3-aminopyridine proceeds normally by standard methods,

2- and 4-aminopyridine can be converted to diazonium salts only with
considerable difficulty. The reason for this is that the mechanism of
diazotization in dilute acid solution involves the formation of an inter-
mediate which is similar to the addition of a second proton to the mono
salt. This is rendered difficult by the fact that one of the contributing

resonance structures in 2- and 4—aminopyridine, but not in

35

36

3-aminopyridine, involves a positive charge situated on the amino
nitrogen. This explains the existence of these compounds in acid solu-
tion as the monohydrochloride only and the fact that they are considered
to be monobasic. Thus it would seem that this reasoning would explain
the failure of 2- and 6-aminonicotine, which have the same 17 electron
system as 2-aminopyridine, to undergo the coupling reactions. Because
the electronic conditions were unfavorable--i. e. , the amino nitrogen
in one of its resonance forms bears a positive charge; diazotization of
the primary amine did not take place. This is supported by the fact
that the only product in most cases was the original amine, indicating
that diazotization did not, in fact, take place.

There are however, observations which refute this hypothesis.
The first is that although products other than the original amines could
not be isolated, the amine was not all recovered, and the difference was
usually greater than can be accounted for by simple mechanical loss.
Something apparently happened to some of the reactant. Secondly, it has
been stated that Tschitschibabin could prepare 2- and 6~hydroxynicotine
by diazotization of the apprOpriate aminonicotine at room temperature.
Indeed, his diazotization of 6-aminonicotine was repeated as part of this
study. However, the yields in some of these reactions were small (on
the order of 10-15 percent) so that it appears that Tschitschibabin also
had difficulty with this conversion. However, since in his work he
mentions no attempt to isolate the starting material, we cannot say with
absolute certainty that he did not succeed in the formation of the
diazonium salt.

We must assume, then, that the failure of the attempts to replace
the amino group of the aminonicotines with a phenyl ring by the various
diazotization procedures was brought about by the difficulty in forming

the diazonium salts of these particular amines.

37

Attempted Phenylation of Ami.nonicotines_

 

The possibility of direct phenylation of the aminonicotines was an
attractive hypothesis because it rendered separation of any isomers
unnecessary and would presumably yield only one product. That is, it
is known that metal alkyls and aryls can substitute in the pyridine ring
by nucleophilic attack at the 2~position. Attack at the 3-position is
mechanistically unlikely, and the formation of 4~substituted pyridines
has not been reported in such reactions. Therefore, with the 2-position
blocked, the only remaining point of attack would be the 6-position, and

vice versa. The fact that this reaction might be carried out is suggested,

 

but not demonstrated, by Abramovitch when he says, "It is clear, that
the addition of 1 mole of phenyllithium to a 2- or 4-substituted pyridine
can only lead to the formation of one compound" (46). Moreover, in an
analogous reaction, when 2-alkylpyridines are treated with alkali amides
in hydrocarbon solvents at elevated temperature, the 2—alkyl-6amino-
pyridines are produced (52). , In addition, treatment of pyridine with an
excess of sodium amide at temperatures near 1700C results in the
introduction of two amino groups, forming 2, 4-diaminopyridines (52).
However, the questionable stability of the aminonicotines at such
temperatures rendered such conditions undesirable.

Thus, the reason for the failure of this method is not clear.
Considerations of the mechanism of nucleophilic attack at the 2-position
of the pyridine ring suggest that the presence of an electron releasing
substituent such as the amino group at the 6-position would, if anything,
enhance the reaction by placing a negative charge on the ring nitrogen,
making it more susceptible to attack by the positively charged lithium.
On the other hand, it is possible that the electron releasing tendency of
the amino group, by placing a negative charge on the ring nitrogen,

renders the bond between this nitrogen and carbon-2 less polar, and

38

therefore discourages nucleophilic attack at the 2-position. At any rate,
for some reason, the presence of an amino group at the 2- or 6-position
renders the 6- or 2-position, respectively, less susceptible to nucleo-

philic attack.

Phenylation of Nicotine

 

The reaction of nicotine with phenyllithium is a straightforward
reaction, the details of which are described by Abramovitch (46). That
it was necessary, in this study, to resort to separation of the mixture
of 2- and 6-phenylnicotine by fractional distillation is somewhat
unfortunate. Spectrophotometric data suggest that this fractionation is
not as clear-cut as one might wish. This problem could be obviated by
the availability of a more effective fractionating system or apparatus for
carrying out the gas chromatographic partition described by Abramovitch.

Finally, the failure of the reaction designed to produce benzoic
acid as a product of the oxidation of the phenylnicotines (or phenylnicotinic
acids) by neutral or alkaline permanganate is even more difficult to
explain. There seems to be no alternative to citing the presence of the
carboxyl substituent as the cause of the increased resistance of the
pyridine ring to permanganate oxidation. Apparently, the electron
attracting properties of the carboxyl group in some way make the pyridine
ring even less stable to oxidation than usual, possibly because the
electrons which must be removed from the carbon atoms on oxidation have
been made less available to the oxidizing agent.

The idea has arisen that it might be possible to first decarboxylate
the phenylnicotinic acids, and then oxidize the resulting phenylpyridine
as planned. The main argument against this procedure is the fact that
the yields in the series of reactions originally proposed are low enough
that the addition of another step--especially one as‘ inefficient as the

decarboxylation of an aromatic acid--wou1d make the degradation infeasible,

39

considering the relatively small amounts of starting material available.
There are reactions, however, which might be inserted into this
sequence to render the pyridine ring more labile to oxidation and in
which the yields would be great enough to justify their use. One possible
pathway might involve the successive formation of phenylnicotine iso-
methiodide hydroidide, the nicotone of this, and finally benzoic acid, in
a series of reactions analogous to those described by Griffith, (31: a}. (12).
Although the stated objective of this part of the work-~the isolation
of the 2- and/or 6-carbons of the pyridine ring of nicotine--was not
accomplished, it is hoped that the various nicotine and nicotinic acid
derivatives which were synthesized during the course of this study might

serve as an aid and a guide in further attempts to solve this problem.

Feedirg Experiments

 

Preliminary studies from this laboratory have postulated that the
pyridine ring of nicotine might be formed £13 wfrom the carbons of
glycerol and prOpionate (l8, 19). Subsequent work showed not only that
propionate-Z-C” is a relatively efficient precursor of the pyridine ring
but also that approximately 40 percent of the radioactivity was in the
3-position. This supported the hypothesis that propionate was incorpor-
ated into the pyridine ring by a mechanism in which the 2 and 3 carbons
of propionate would give rise to the 3 and 2 positions respectively of the
pyridine ring. However, the data was not consistent with immediate
incorporation of propionate as a unit, since approximately half of the
activity of the pyridine ring was unaccounted for after feeding propionate-
2—c”.

Extensive studies of propionate metabolism have revealed at least
two major pathways by which propionate can be metabolized. Flavin

e_t a_L_1. (53, 54), have shown that in various animal tissues propionate can

40

be carboxylated to methylmalonate which subsequently rearranges to
succinate. On the other hand, Giovanelli and Stumpf (55) showed that
in peanut mitochondria propionate can be converted by fi-oxidation and
subsequent loss of the original carboxyl group to acetate, which could
then enter the tricarboxylic acid cycle. In either case propionate-Z-CH
would give rise to succinate randomly labeled in the methylene carbons,
as would propionate-3-Cl4 and acetate-Z-C”. Propionate formed in
turn from the labeled succinate, possibly by way of methylmalonate,
would be labeled in carbons 2 and 3', thereby tending to randomize
radioactivity between these positions in the propionate pool. Thus if
propionate-Z-C” were the precursor of carbons 2 and 3 of the pyridine
ring, carbon 2 would contain a quantity of Cl4 similar to carbon 3,
following the randomizing reactions. In addition, the carbons from
these two positions would account for most of the radioactivity of the
pyridine ring. Accordingly, it was suggested, after feeding propionate-
2-C” and location of half the activity of the pyridine ring in the 3-position,
that the other half should reside in the 2-position. Similar results would,
of course, be expected from propionate-LC”. Evidence confirming or
rejecting this hypothesis could come from isolation of the 2-position or
demonstration that propionate-3-C” also gave rise to labeling in the
pyridine ring, half of which was located in the 3-position.

The average incorporation dilution figure of 725 for propionate-3-
C” is evidence that it is a possible precursor of nicotine in tobacco.
Its greater dilution indicates that it is not as good as propionate-Z-C”,
and the difference between these two is reminiscent of the difference
between acetate- l-Cl4 and acetate-Z-C” (12). It should also be noted
that acetate-Z-C“ (450) shows a greater dilution than propionate-2--C”I
(200) and acetate-l-C“I (950) greater than propionate-3-C” (725). If

acetate and propionate are part of a metabolic pathway leading to nicotine,

41

these data could indicate that acetate precedes propionate, but the data
are not clear enough to warrant a definite conclusion.

The dilution data also do not allow the assignment of the role of
immediate precursor to 8-a1anine. The striking similarity of the
dilution values for propionate-2 and [3--alanine-2-C14 and for propionate-3
and (3-alanine-3-Cl4 suggest that they are about the same distance from
nicotine, and, if anything, (S-alanine is even more remote than propionate.

In summary, the dilution data, when compared with the previously
reported data, seem to indicate that the compounds tested might be
divided into three main groups, each containing compounds displaying
similar dilutions: 1) fi-alanine-Z-C”, propionate-Z-C“, and acetate-Z-C”,
the best precursors with an average dilution factor approximating 350;

2) fi-alanine-3-C“, propionate-LC“, and acetate-l-C“ next at about 800;
and 3) propionate-l-C”, which is not significantly incorporated.

The failure of propionate-l-C” to be incorporated into nicotine is
consistent with the hypothesis of propionate incorporation. The demon-
stration of the conversion of nicotinic acid into nicotine has been
mentioned (5). Thus it is reasonable to assume in the present study that
nicotinic acid or a related metabolite serves as an intermediate in the
synthesis of nicotine. . In the conversion to nicotine, nicotinic acid should
be decarboxylated. . Thus, if the 2 and 3 carbons of propionate were
incorporated into the pyridine ring, it is possible that the carboxyl carbon
of propionate would become the carboxyl carbon of nicotinic acid. . If this
be the case, no activity from propionate— l-C” would be found in the
pyridine ring of nicotine. Such a suggestion is borne out by the results
of this study.

The next phase of the study was the degradation of the nicotine
obtained from those substances which were significantly incorporated in
order to determine their incorporation into the pyridine ring. The data

obtained in this work show our original hypothesis of propionate

42

incorporation into the pyridine ring to be in error. The permanganate
oxidation of nicotine and decarboxylation of the resulting nicotine acid

to obtain pyridine showed that after feeding propionate-LC”, less than

3 percent of the radioactivity of the nicotine was located in the pyridine
ring, a striking demonstration of the failure of the 3 carbon of propionate
to be incorporated into the pyridine moiety. Thus it is obvious that
propionate is not a 3-carbon unit precursor of the pyridine ring of nicotine.
The fact that the C14 from the 2-labeled propionate is significantly
assimilated into the pyridine ring indicates that propionate must undergo
some metabolic change involving loss of carbon 3 before the incorporation
of carbon 2.

Results with Iii-alanine are similar to those with propionate and
also serve to refute the original hypothesis. Assuming that the metabolic
relationship which exists between B-alanine and propionate in various
microorganisms and animal and plant tissues (56, 57, 58) also exists in
tobacco, it is not surprising that B-alanine-3-Cl4 also fails to produce
any radioactivity in the pyridine ring of nicotine. If B-alanine and pro-
pionate are in metabolic equilibrium with each other, it would be pre-
dicted that, whereas their incorporation dilutions might be different, the
distribution of C14 in the nicotine molecule would be similar for the two.
The data in Tables II and III support such a prediction.

The close metabolic relationship between propionate and B-alanine
is further demonstrated by the fact that approximately 40 percent of the
activity of the pyridine ring, after feeding B-alanine-Z-C”, was in the
3 position, 1 compared to the figure of 50 percent already stated for
propionate-Z-C“.

At this point, then, the degradation data warrant three general

conclusions: (1) propionate and fi-alanine are closely related metabolically

 

lT. Griffith, unpublished data.

43

in tobacco; (2) neither fi-alanine nor propionate is a 3-carbon unit pre-

c ursor of the pyridine ring of nicotine; (3) incorporation of the ‘carbon
atoms of fi-alanine and propionate into the pyridine ring of nicotine occurs
after a metabolic scheme involving loss of carbon-3.

A clue to what may be happening to both propionate and B-alanine
prior to incorporation into nicotine can be found by comparing the distri-
bution of labeling in nicotine after feeding these two metabolites labeled
in various positions to results obtained with acetate- l-C14 and acetate—

2-c” (ll).

     

acetate-l-C” B-alanine-3-CM propionate-3-Cl4

The striking similarity of distribution with acetate- l-CM‘,
B-alanine-3-C”, and propionate-3-Cl4 is apparent from the diagram in
which the numbers represent the percent of the total carbon- 14 located
at the positions indicated. Not only is the lack of incorporation of
carbon- 14 into the pyridine ring common to all three compounds, but
also the pattern of labeling in the pyrrolidine ring is similar. A simi-
larity of distribution of labeling is also observed with acetate-24Cl4 (l l),
propionate-Z-C” (12) and [i-alanine-Z-CH'.l These similarities could be
explained rather simply by assuming that the two following relationships,
shown to occur in other plants, also apply in tobacco: (1) fi-alanine and

propionate are so related metabolically that carbons 1, 2, and 3 of one

 

1R. F. Dawson, personal communication.

44

are the precursors of the same carbons of the other; (2) either (3-alanine
or propionate is converted to acetate: propionate, by B-oxidation to
malonyl semialdehyde, followed by loss of the original carboxyl as
carbon dioxide and oxidation of the aldehyde group to form acetate;
(3-alanine, by transamination to malonyl semialdehyde, and then to
acetate as described above. Such an hypothesis as this would suggest
that acetate is a closer precursor to the pyridine ring than either pro-
pionate or B-alanine. However, incorporation dilution data, which show
a greater dilution for acetate than for either of the 3 carbon compounds,
are inconsistent with this idea. Short term feeding experiments, now in
progress, may shed further light on this dilemma.

Data from similar studies on the pyridine containing alkaloid
ricinine by Anwar e_t a}. (59), are in agreement with the results of this
work. These studies indicate that about 90 percent of the radioactivity
from acetate—l-CM, propionate-l-C”, and propionate-3-Cl4 was located
in the nitrile carbon of the ricinine, of which the carboxyl of nicotinic acid
is known to be the precursor (6, 16). In addition, approximately 75 per-
cent of the radioactivity from both acetate-Z-Cl4 and propionate-Z-C”
was believed to reside in the pyridone ring. These studies also imply
both a relationship between B-alanine and propionate and the possible
conversion of the 3 carbon of propionate to the carboxyl carbon of acetate.

Therefore, the conclusions that are drawn from the results of the
present study are these: (1) although (3-alanine and propionate appear to
be related metabolically in tobacco, neither is an immediate 3 carbon
precursor of the pyridine ring of nicotine; and (2) it is possible that both
[B-alanine and propionate are converted to acetate, the carbon atoms of
which are then incorporated into nicotine by an unknown pathway such
that the methyl carbon, but not the carboxyl carbon, is a precursor of

the pyridine ring .

45

Recently, several theories of nicotine pyridine ring biosynthesis
have been proposed (60). In one, Dawson and co-workers suggest that
carbons 2, 3, and 4 of citric acid are precursors of carbons 2, 3, and
4 of the pyridine ring of nicotine. The results of Byerrum and Griffith
with aspartic acid-3-C”, while demonstrating that aSpartic acid is
itself not a direct precursor of the pyridine ring, indicate that it is not
likely that citric acid is either.

Thus, at present, we still do not know which substances are the
immediate precursors of the pyridine ring of nicotine and therefore what
happens to such metabolites as B-alanine, propionate, and acetate prior
to their incorporation. Suitable degradation procedures for the pyridine

ring should prove to be of great help in resolving this dilemma.

SUMMARY

46

1.

SUMMARY

The following procedures for the isolation of carbon from the

2- and 6—positions of the pyridine ring of nicotine were undertaken:

a.

2.

amination of nicotine with sodamide yielded 2- and 6—amino-

nicotine which were separable from each other;

. attempts to replace the above amino groups with a benzene

ring by several procedures were unsuccessful;

. attempts to phenylate the 2- and 6-positions of 6- and Z-amino-

nicotine, reSpectively, using phenyllithium were also

unsucces sful;

. treatment of nicotine with phenyllithium yielded a mixture of

2- and 6-phenylnicotine which were separated by fractional

distillation;

. oxidation of the phenylnicotines with neutral or alkaline per-

manganate yielded only the respective phenylnicotinic acids.

Propionate-l-CM, propionate-3-C”, fl-alanine-Z-C”, and

B-alanine-3-C” were fed to tobacco plants. With the exception (of

propionate- l-Cl‘, all were found to be significantly incorporated into

nicotine isolated from the plants, with B-alanine-Z-C“ serving as the

best precursor; propionate-~3-C14 and (3-alanine-3-C“ were incorporated

with somewhat greater dilution.

3.

The labeling pattern of nicotine from plants fed propionate-3-

C” and (i-alanine-3-Cl4 was determined by appropriate degradation

procedures; the distribution of Cl4 within nicotine was similar for both

compounds, and neither contributed its 3 carbon for synthesis of the

pyridine ring.

47

48

4. It was concluded that B-alanine and propionate are closely
related metabolically, but that neither is an immediate precursor to
the pyridine ring of nicotine. In the light of evidence from other workers,
it is suggested that propionate and B-alanine may be converted to acetate

prior to incorporation into the pyridine ring of nicotine.

LIT ERATUR E CIT ED

49

10.

ll.

12.

13.

14.

LITERATURE CIT ED

. Krehl, W. A., Teply, L. J., Sarma, P. B., and Elvehjem,. C. A.,

Science, 101, 489 (1945).

. Heidelberger, C., Abraham, E. P., and Lepkovsky, S., J. Biol.

Chem., 179, 143, 151 (1949).

. Beadle, G. W.,, Mitchell, H..K., and Nyc, J. F., Proc. Natl.

Acad. Sci. U. s., 3_3_, 155 (1947).

. Partridge, C. W. H., Bonner, D..M., and Yanofsky, C., J. Biol.

Chem., 194, 269 (1952).

. Dawson, R. F., Christman, D. R., Anderson, R. C., Solt, M. L.,

D'Adamo, A. F., and Weiss, U., J. Am. Chem. Soc., _7_8_, 2645
(1956).

. Leete, E., Chem. and Ind., 1270 (1957).
. Grimshaw, J., and Marion, La, Nature, 181, 112 (1958).

. Henderson, L...M., Someroski, J., Rao, D. R., Wu, P. -H. L.,

Griffith, T., and Byerrum, R. U., J. Biol. Chem., 234, 93 (1959).

. Yanofsky, C., J. Bacteriol., _6_8, 577 (1954).

Byerrum, R. U. , and Griffith, T. ,. Federation Proc. , _1_8_, 942
(1959).

Griffith, T., and Byerrum, R. U., Science, 129, 1485 (1959).

Griffith, T., Hellman, K. P., and Byerrum, R..U., J. Biol.
Chem., 235, 800 (1960).

Griffith, T., Hellman, K. P., and Byerrum, R. U., 138th meeting
of Am. Chem..Soc., September 1960, p. 25C.

Ortega, M. V., and Brown, G. M., J. Am. Chem. Soc., _8__1_, 4437
(1959).

50

15.

16.

17.

18.

19.

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

51
Ortega, M. V., and Brown, G. M., J. Biol. Chem., 235, 2939
(1960).

Waller, G. R., and Henderson, L. M., J. Biol. Chem., 236,
1186 (1961).

Waller, G. R., and Henderson, L. M., Biochem. Biophys.
Research Commun., 5, 5 (1961).

Griffith, T., and Byerrum, R. U., Federation Proc., 18, 942,
(1959).

Byerrum, R. U. , and Griffith, T. , Proc. Ninth Internat.
Botanical Congress, 55 (1959).

Griffith, T. , "A Study of Ribose Metabolism in the Tobacco Plant, "
Ph.D. Thesis, Michigan State University, 1958.

Solt, M. L., Dawson, R. F., and Christman, D. R., Plant Physiol.,
15, 887 (1960).

Lamberts, B. L., and Byerrum, R. U., J. Biol. Chem., 233,
939 (1958).

Fritzson, P., and Eldjarn, L., Scand. J. Clin. and Lab. Invest.,
4, 375 (1952). ' ‘

Spies, J. R., Ind- Eng. Chem., Anal. Ed., 2, 46 (1937).
Smith, c. R., Ind. Eng. Chem., Anal. Ed., 34, 251 (1942).

Henry, T. A. , The Plant Alkaloids, The Blakiston Co. ,
Philadelphia, 1949, p. 39.

Yankwich, P- E., Norris, T. H., and Huston, J., Ind., Eng. Chem.,
Anal. Ed. , 12, 439 (1947).

Krampl, V. , "The Role of A'-pyrroline-5-carboxylic Acid in the
Biosynthesis of the Pyrrolidine Ring of Nicotine, Ph.D. Thesis,
Michigan State University, 1961.

Laiblin, R., Ann. Chem. Liebigs., 196, 129 (1879).

Smith, C. R., Ind. Eng. Chem., 34, 251 (1942).

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

52

Huber, C., Ann. Chem. Liebigs., 141, 271 (1867).

Campbell, N. , in Chemistry of Carbon Compounds, Vol. IVA, Rodd,
E. H. (ed.). Elsevier Publishing Co., New York, 1957. pp. 491 ff.

 

Tschitschibabin, A. E., J. Russ. Phys.-Chem. Soc., _3_3, 251
(1901).
Tschitschibabin, A. E., Ber., _3_7_, 1373 (1904).

Tschitschibabin, A. E., and Seide, O. A., J. Russ. Phys.-Chem.
Soc., g, 1216 (1914).

Tschitschibabin, A. E., and Kirssanow, A. W., Ber., _5_7_, 1163
(1924).

Bachmann, W. E., and Hoffman, R. A. , in Organic Reactions,
Vol. II, John Wiley and Sons, Inc. , New York, 1944.

 

Elks, J., Haworth, J. W., and Hey, D. H., J. Chem..Soc., 1284
(1940).

Vaughn, T. H., Vogt, R. R., and Nieuwland, J. A., J. Am. Chem.
Soc., 2_6_, 2120 (1934).

Tschitschibabin, A. E., Chem. Zentr., 1023, 1923 III.

Shriner, R. L., Fuson, R-. C., and Curtin, D. Y., The Systematic
Identification of Organic Compounds, John Wiley and Sons, Inc. ,

 

 

New York, 1956, p. 125.
Eddy, C. R., and Eisner, A., Anal. Chem., 26, 1428 (1954).

Tschitschibabin, A. E., and Buholz, L. A., J. Russ. Phys.-Chem.
Soc. 50, 551 (1918).

Bachmann, W. E., and Hoffman, R. A. , in Organic Reactions,
Vol. II, John Wiley and Sons, Inc. ,. New York, 1944, p. 224.

 

Atkinson, E. R., Morgan, C. R., Warren, H. H., and Manning,
T. J., J. Am. Chem. Soc., 61, 1513 (1945).

Silverman, L., and Bradshaw, W., Anal. Chem., 21, 96 (1955).

Abramovitch, R. A., Seng, G. C., and Notation, A. D., Can. J.
Chem., -3_8_, 761 (1960).

47.

48.

49.

50.

51.

52.

53.

54.

55.

56.

57.

58.

59.

60.

53

Abramovitch, R. A., and Notation, A. D., Can. J. Chem., E,
1445 (1960).

Benary, E., and Psille, H., Ber., _5_:/_, 828 (1924).

Ishiguro, T., Morita, Y., and Ikushima, K., Yakugaku Zasshi,
38, 220 (1958); Chem. Abstr., _5_;Z_, 11846g (1958).

Hochstein, L. 1., and Rittenberg, S. C., J. Biol. Chem., 234,
156 (1959).

Barnes, R. A. , in The Chemistry of Heterocyclic Compounds:
Pyridine and Derivatives, Part I, Klingsberg, E. , (ed. ). . Inter-
science Publishers, Inc., New York, 1960, pp. 70-74.

 

Leffler, M. T., in Organic Reactions, Vol. I, John Wiley and Sons,
Inc., New York, 1942, p. 93f.

 

Flavin, M., Ortiz, P. J., and Ochoa, 8., Nature, 176, 823 (1955).
Flavin, M., and Ochoa, S., J. Biol. Chem., 229, 965 (1957).

Giovanelli, J., and Stumpf, P..K., J. Am. Chem. Soc., 12, 2652
(1957); J. Biol. Chem., 231, 411 (1958).

Stadtman, E. R., Federation Proc., _1_5_, 360 (1956).
Kupiecki, F. P., and Coon, M. J., J. Biol. Chem., 229, 743 (1957).

Giovanelli, J., and Stumpf, P. K., J. Am. Chem. Soc., _7_9, 2652
(1957).

Anwar, R., A. , Griffith, T. , and Byerrum, R. U. , Federation Proc. ,
Q, 374 (1961).

Chem. and Eng, News, 32, 40 (1961).

Grammy UBRAR‘Y

FEB 1 5 ’63

 

 

 

 

IIIIIIIIIIIIIIIIIIIII

5

12

((1);

Tu)