THE TRCORPDRATEON 0F CARBON
DIGXTDE AND ACETATE ENTO NTCOTINE

Thesis for the Degree of Pb“ D.
MICHIGAN STATE UNIVERSITY
HORST RONALD ZTELKE
1968

THESIS

L135: “(1"

MlClllga 1 State
UlllVCL’SiC)’

 

This is to certify that the

thesis entitled
THE INCORPORATION OF CARBON

DIOXIDE AND ACETATE INTO NICOTINE

presented by

HORST RONALD ZIELKE

has been accepted towards fulfillment
of the requirements for

_P*1_-_II-_ degree in _Bi_0¢em_is t ry

Major proffessor

Date August 1, 1968

0-169

    
 

BINDING av V
W“ 5 SUNS'
8m BWEPY NC.

I I-DIDV RIM P"

  

ABSTRACT
THE INCORPORATION OF CARBON DIOXIDE AND ACETATE INTO NICOTINE
by Horst Ronald Zielke

Glutamate, a proposed precursor of the pyrrolidine ring
of nicotine, was isolated and degraded after metabolism of
lLI'COZ for 6 hours by N, rustica and N, glutinosa and after
metabolism of acetate-Z-luc for 2 hours by N, rustica. The
140 distribution in glutamate from.N, glutinosa after 6 hour
14002 metabolism was 20.2%, 14.3%. 13.9%, 24.4% and 26.6%,
in C-l through 0-5, reSpectively. The 14c distribution in
nicotine under the same conditions was pyridine ring, 77.8%;
0-2!, 4.3%; 0-31, 4.4%; c—4t, 4.7%; 0-5!, 4.1% and methyl
carbon, 5.0%. The labeling pattern of the pyrrolidine ring
is in agreement with the symmetrical intermediate pathway
proposed for the formation of the pyrrolidine ring of nico-
tine from glutamate. Similar results were obtained after 6

14

hour C02 incorporation into nicotine by N, rustica. After
3 hour 14002 metabolism by N, glutinosa, 90.4% of the radio-
activity was in pyridine ring, 5.4% in the methyl group and
the remaining radioactivity was distributed equally among
the carbons of the pyrrolidine ring.

Acetate-Z—lu

C labeled glutamate in N. rustica as anti-
cipated if acetate were metabolized via the tricarboxylic
acid cycle: 1.5%. 5.4%, 4.6%, 81.0%, 2.8% in C-l through
C-5, reSpectively. The distribution of 14C in nicotine was

pyridine ring, 64.4%; C-Z', 2.6%; C-3'. 16.4%: C~4'. 15-6%:

Horst Ronald Zielke
C-5', 1.8% and methyl carbon, 1.6%. These results support
the symmetrical intermediate hypothesis for the formation
of the pyrrolidine ring of nicotine proposed on the basis
of previous precursor eXperiments.
The 1“'0 distribution in the pyridine ring of nicotine
14

after 3 hour C02 metabolism by N, glutinosa was 10.0%,
9.2%, 25.0%, 24.2% and 26.7% in carbons 2 through 6, reSpec-
tively. After 6 hour 14C02 metabolism by N, rustica, the

1”C distribution in the pyridine ring of nicotine was 13.1%,
14.0%, 23.4%, 21.7% and 23.0% in carbons 2 through 6,
reSpectively. It is postulated that the above labeling pat-
tern is the result of unequal labeling of glyceraldehyde and
aSpartic acid, the proposed precursors of the 5-carbon chain
of the pyridine ring. It is proposed that glyceraldehyde

and aSpartic acid are labeled unequally due to different
dilution rates and/or slower incorporation of label into the
internal carbons of aSpartic acid. After 2 hour metabolism
of acetate-Z-lhc by N, rustica, carbons 2 and 3 of the pyri-
dine ring are labeled equally and contain approximately 88%
of the total radioactivity of the pyridine ring. This label-
ing pattern is consistent with the proposal that acetate-2-
140 is incorporated into the pyridine ring of nicotine via

aSpartic acid.

THE INCORPORATION OF CARBON DIOXIDE AND ACETATE INTO NICOTINE

By

Horst Ronald Zielke

A THESIS

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

DOCTOR OF PHILOSOPHY

Department of Biochemistry

1968

ACKNOWLEDGMENT

I wish to eXpress my gratitude to Dr. Richard
U. Byerrum for his encouragement and guidance during
this study.

I am grateful to Dr. N. E. Tolbert for his
advice on C02 fixation and to C. Michael Reinke, NSF
Undergraduate Research Participant, for his assis-
tance during the summer of 1967. The cooperation of
Drs. O'Neal and Koeppe and Miss Burns in the gluta-
mate degradation is appreciated.

The financial assistance from the National
Institutes of Health is acknowledged.

11

TABLE OF CONTENTS

INTRODUCTION 0 O O O O O O O O O O 0 O O O 0 O O 0

EXPERIMENTAL PROCEDURE . . . . . . . . . . . . . .

RESULTS AND DISCUSSION . . . . . . . . . . . . . .

Specificity of the Degradation . . . . . . .

Incorporation of Label into Glutamate and
Aspartate O O O I O O O O O O O O O O O O

Incorporation of Label into the Pyrrolidine

Bingoooeeeeoooooooooooo

Study of a Hypothetical Non-Symmetrical
Pathway . . . . . . . . . . . . . . . . .

Incorporation of Label into the Pyridine
Bing-00000000000000...

SUM IVIAR Y O O O O O O O O 0 O O O O O O 0 0 O O O 0

REFERENCES 0 O O O O O O O O O O O O O O O O O O O

APPEINDIX O O O O O O O O O O O O O O 0 O O O O O 0

111

Page

11

16
16

21

25

33

39

44

45

49

LIST OF TABLES

Table Page

1. Degradation of Specifically Labeled Dimethyl
Gly01neoeooooo00000000000019

2. Distribution of 140 in Glutamate and ASpartate . 22

3. Distribution of 140 in the Pyrrolidine Ring
of Nicotine After 14002 Incorporation . . . . 26

4. Distribution of 140 in the Pyrrolidine Ring
of Nicotine Fr m.N, rustica.After 2 Hours
of Acetate-2-1 C Metabolism . . . . . . . . . 31

5. Incorporation of Small Metabolites Into
NicOtine00.000000000000000 36

6. Distribution of 14c in the Pyridine Ring of
Nicotine After 1“cog Incorporation . . . . . 4o

7. Distribution of 140 in the Pyridine of Nicotine

{£0m.N, rustica After 2 Hours of Acetate-2-
CM8t8-501ism00000000000000. “’3

iv

LIST OF FIGURES

Figure Page

1. Symmetrical Pathway for Pyrrolidine Ring
Biosynthesis via the Mesomeric Anion of
A l-PyrrOIj-ne O O O O O O O O O O O O O O O O 2

2. Symmetrical Pathway for Pyrrolidine Biosyn-
thesis via Putrescine . . . . . . . . . . . . 3

3. Non-Symmetrical Pathway for Hygrine Biosyn-
thesis From Ornithine . . . . . . . . . . . . 5

4. Pyrrolidine Ring Degradation Scheme From
Liebman §£,§; . . . . . . . . . . . . . . . . 17

5. Proposed Randomization of Glutamate in the
Formation of thePyrrolidinfi Ring of
Nicotine After 6 Hours of 002 Metabo-
118m by E. rUStica O O O O O O O O O O O O O 28

6. A Hypothetical Non-Symmetrical Pathway for
Pyrrolidine Ring Biosynthesis . . . . . . . . 34

INTRODUCTION
The Pyrrolidine Ring of Nicotine

The biosynthetic pathway for the alkaloid nicotine
has been extensively studied through precursor eXperiments.
These studies have shown that glutamic acid-2-1uC (1, 2)
and ornithine-Z—luc (3, 4) are incorporated into the pyr-
rolidine ring of nicotine via a symmetrical intermediate in
Nicotiana rustica var. humilus and Nicotiana tabacum. Wu,

14

Griffith and Byerrum (5) and Wu and Byerrum (6) using C

labeled acetate, glycerol, propionate and aspartate demon-
strated that the distribution of 14C in the pyrrolidine
ring of nicotine was in general agreement with the eXpected
labeling if these compounds were converted to glutamate by

way of the tricarboxylic acid cycle.' Leete (7) postulated

that ornithine-Z-luc (I) was incorporated into the pyrroli-

dine ring of nicotine (II) via glutamic semialdehyde-Z-luC

(III). A1-pyrroline-5-carboxylic acid-534C (IV), A1-
14

pyrroline-S-carboxylate anion-5- C (V), mesomeric anion of
Al-pyrroline-Z-luc (VI) and Al-pyrroline-Z-luc (VII) as

shown in Figure 1.

 

 

 

 

—————+> -—+———> * ———->
*f—-cooH // '*-——COOH ‘\N/’_—COOH
H2N NH2 0 NHZ
I III IV
ii- (3009 ‘5—9 Us + mat-
\ “%V’ N/ \N’
V VI VII VIII
\N 633
II FIGURE 1

This hypothesis is in agreement with Krampl (8) and
Krampl, Zielke and Byerrum (9) who observed that Ae-pyrroline-
5-carboxylic acid-s-luc labeled equally carbons 2' and 5' of
the pyrrolidine ring. However, recent work has indicated
that a different biOSynthetic pathway is Operative. Leete,
Gros and Gilbertson (10) observed with sterile root cultures
of N, tabacum that only the é-amino group of 15N—1abe1ed
ornithine-Z-luc was incorporated into the pyrrolidine ring.
This eliminated glutamic semialdehyde and its cyclic form
Z51-pyrroline-5-carboxylic acid, as an intermediate between
ornithine and the pyrrolidine ring. The participation of
other intermediates shown in Figure 1 was further questioned
by the eXperiments of Mizusaki §£_§l (11) and Leete (12).

4-(N-methyl)-aminobutyraldehyde (N-methyl pyrroline), a

3
compound rapidly labeled after metabolism of ornithine-Z-luc
by aseptic root cultures of N, rustica, was found to be an
efficient precursor of nicotine (11). 4-(N-methyl)-amino-
butyraldehyde was also shown to be rapidly labeled by
putrescine-1,4-140 or methionine-140H3 (11). Leete (12)
observed after feeding N-methyl All-pyrrolinium-Z-luc chloride
to N. tabacum that 18% of the N-methyl- Al-pyrrolinium-Z-lu'c
chloride was incorporated into the pyrrolidine ring of nico-
tine without undergoing tautomerism. Only carbon 2' of the
pyrrolidine ring was labeled. The pathway in Figure 2
represents the hypothetical biosynthetic pathway proposed by

Leete (12) and Mizusaki 33,3; (11).

flown :flifl

1 ‘ l
D; .fl
_ H
©fi©t©

h,

HN-CH

HN-CH
3

FIGURE 2

4
At presence no evidence is available to determine at which
point methylation occurs in the sequence shown in Figure 2.
Although ornithine is incorporated symmetrically into
the pyrrolidine ring of nicotine, present evidence is to the
contrary for several other alkaloids that contain the N-methyl

pyrrolidine ring.

 

 

 

 

 

‘9 0 CH OH
COO \g’ u I 2
-CH3 0-C-CH-C6H5
9 J
N
/\
CH3 CH3 Hyoscyamine
Stachydrine
\N > | | ‘3 I I
-CH O
3 CH -C-CH
__D/ N 2 2 N
I 3
CH3 CH3
Hygrine - Cuscohygrine
14

Essery gfiflgi (13), using ornithine-Z- C, obtained stachydrine
from alfalfa plants labeled only at carbon 2. Robertson and
Marion (14) observed that carboxyl-luc proline was incorpor-
ated into stachydrine. labeling only the carboxyl group.
Presumably proline is an immediate precursor of stachydrine.
Hyoscyamine, a tropane alkaloid from Datura stramonium

(Jimson weed) incorporated label from ornithine-Z-luC into
only one of the bridgehead carbons next to the nitrogen atom
(15). Unpublished data from Liebish, Shutte and Mothes quoted
by Mothes and Shfitte (16) indicates that putrescine-1,4-140

labeled both bridgehead carbons. However, since ornithine is

incorporated unsymmetrically, it is unlikely that putrescine

5
is a normal intermediate of the pathway through which ornithine
is incorporated.
Hygrine, which is structurally related to the tropane

skeleton, has not been studied with 14

C-labeled precursors,
however, Anet §£_§; (17) have been able to increase the amount
of hygrine synthesized by administrating 4-(N-methyl)-amino-
butyraldehyde. This has led Mothes and Shutte (16) to the
opinion that the biosynthetic pathway for hygrine is similar

to that for tropane alkaloids:

non-symmefrica;

jathway ‘— 9/
COOH / N
H N H2 0 HN-CH3 CH3

 

_’ ('JOOH -——> has >= 0
//r_ e> L\N,/L'CH L_/| 3
90011 5H3 >=o

9H2
0=C~CH

 

 

 

 

3
FIGURE 3

Clarke and Mann (18) using diamine oxidase isolated from pea
cotyledons, coupled with catalase, observed that norhygrine
was formed after the addition of putrescine and acetoacetate.
Tuppy and Faltaous (19) observed hygrine and cuscohygrine
formation after adding N-methylputrescine and acetoacetate or
acetonedicarboxylic acid to diamine oxidase and catalase
preparations.

.Before discussing the mechanisms by which glutamate or

ornithine may be incorporated either symmetrically or non-

6
symmetrically into the pyrrolidine ring, it is necessary to
discuss the pathway for putrescine biosynthesis in plants.
Smith and Garraway (20), using barley seedlings, obtained
evidence which indicated that ornithine was converted to
putrescine via citrulline, arginosuccinic acid, arginine,
agmatine and N-carbamyl putrescine. Direct conversion of
ornithine to putrescine does not proceed at a measurable
rate under their conditions. This is unlike the multiple
pathways of putrescine biosynthesis studied by Morris and
Pardee in N, 32;; (21). Morris and Pardee observed direct
conversion of ornithine to putrescine as well as conversion
of ornithine to putrescine via arginine and agmatine.

In theory, glutamate or ornithine could be incorpor-
ated into the alkaloids mentioned previously via 4-(N-methyl)-
aminobutyraldehyde by a symmetrical or non-symmetrical path-
way depending upon the site of methylation or oxidation of
the precursors for 4-(N-methyl)-aminobutyraldehyde. If
ornithine is converted to 4-(N-methyl)-aminobutyraldehyde
via putrescine, the labeling pattern in the alkaloid will
show symmetry. If, however, methylation or oxidation occurs
prior to the step in which putrescine forms, non-symmetry will
result. Another unsymmetrical pathways for ornithine incor-
poration can be proposed if ornithine is incorporated via
£51-pyrroline-5-carboxylic acid and the attachment of the
side chain occurs before decarboxylation of the 7A}-pyrroline-
5-carboxylic acid yields the mesomeric anion (22). These
possibilities have not yet been explored for alkaloids in

7
which unsymmetrical ornithine incorporation occurs.

Recently the eXperiments leading to the pathway pro-
posed in Figure 2 for the symmetrical labeling of the pyr-
rolidine ring of nicotine from ornithine have been criticized
by Liebman g§_§;_(23). An important assumption in the pre-
cursor eXperiments is that the exogenously supplied metabo-
lites are incorporated in a manner indistinguishable from
endogenous metaboliteS. These authors initiated a series of
eXperiments (23-26) in which nicotine was isolated and

degraded after 1”

002 incorporation by N, glutinosa. Liebman
g: 5; observed that carbon 2' of the pyrrolidine ring con-
tained less radioactivity after 6 hour 14CO2 incorporation
than either carbons 3', 4' or 5'. Ornithine-Z-luc, unlike
14C02, was incorporated via a symmetrical intermediate into
the pyrrolidine ring of nicotine by N, glutinosa (23). From
these observations Liebman.g§|a; postulated that the symmet-
rical labeling pattern observed from precursor feedings may
be the result of a minor or aberrant pathway since the 1“C02

eXposures represent normal growth conditions.
The Pyridine Ring of Nicotine

The biosynthesis of the pyridine ring of nicotine has
been extensively studied. Previous work in this laboratory
has indicated that carbons 4, 5 and 6 of the pyridine ringof
nicotine are derived from glyceraldehyde or glyceraldehyde-3-
phosphate while carbons 2 and 3 of the pyridine ring are

derived from carbons 2 and 3 of aSpartic acid. After 4 hour

8

1&0 incorporation by N, rustica, 57% of the

aSpartic acid-3-
pyridine ring label was in carbon 3 and 38% in carbon 2 (27).
In a similar 4 hour glyceraldehyde-3-1uC incorporation study,
55% of the radioactivity of the pyridine ring was in carbon
4 (28). Fleeker and Byerrum (28) postulated a Schiff base
formation between glyceraldehyde and aSpartic acid as an
early step in the pyridine ring formation. Further incor-
14C 14C

poration eXperiments with acetate-Z- , succinate-2,3-.
1

and, malate-3-14C (27) and glycerol- ”C (30) support this
hypothesis. In the above studies with radioactive acetate,
succinate and malate, carbons 2 and 3 of the pyridine ring

of nicotine contained a large proportion of the pyridine

ring label. This labeling pattern is eXplained if these com-
pounds are incorporated into the pyridine ring of nicotine
via aSpartic acid. Fleeker and Byerrum (30), using glycerol-
2-14C and glycerol-1,3-1uC, observed that the carbon atoms of
glycerol were incorporated $2.2222 into positions 4, 5 and 6
of the pyridine ring of nicotine. Over 98% of the 14C in

2-14C was in carbon

the pyridine ring derived from glycerol-
5 of the pyridine ring after a 4 hour metabolic period.
Glycerol-1,3-1uC labeled equally carbons 4 and 6 of the
pyridine ring. Carbons 4 plus 6 contained 68% of the pyridine
ring label, whereas carbon 5 had about 2% of the radioactivity.
Since subsequent studies indicated that glyceraldehyde is
incorporated with less dilution than glycerol (28), it is
assumed that glycerol is incorporated into the pyridine ring

of nicotine after oxidation to glyceraldehyde.

9

Similar mechanisms are indicated for the formation of
the pyridine or pyridone rings of anabasine (31, 32) and
ricinine (33-35) in other higher plants. Trypthophan is not
a precursor of the pyridine ring of nicotine (36). The bio-
synthesis of nicotinic acid by Escherichia 22}; (37) and
Nycobacterium tuberculosis (38) also proceeds via a 3-carbon
compound condensing with aSpartic acid. However Isquith
and Moat (39), using partially purified extracts of

Clostridium butylicum, concluded that acetate and formate,

 

rather than glycerol, participated in the formation of the
pyridine ring of nicotinic acid in their system. Scott and

Mattey (40) observed that formate-d“)+

C was readily incorpor-
ated into the pyridine ring by'gg, butylicum and that 90%
of the label was located in carbon 6. Using extracts of
El. butylicum Scott gtnal (41) observed formation of radio-
active N-formyl-L—aSpartate from aSpartate and formate-lac
with more than 90% of the radioactivity located in the formyl
carbon. When formate and aSpartate in their incubation mix-
ture were replaced by N-formyl-L-aSpartate, the amount of
nicotinic acid synthesized was increased three fold. Previous
studies in this laboratory indicated that when formate-14C

was metabolized by N, rustica, over 90% of the radioactivity
incorporated into the nicotine resided in the methyl group
(42). This indicates that formate has different roles in
these two systems,

The mechanism by which the precursors of the pyrroli-

dine and pyridine rings combine to form nicotine is not known.

10

However, Dawson 22.2}.(43) observed that label from nicotinic
acid tritiated at carbon 6 was less readily incorporated
into nicotine than tritium at the other carbons. This implies
that a 1,6-dihydropyridine intermediate may be involved in the
condensation reaction.

The work described in this thesis was initiated to
determine the validity of precursor eXperiments by comparing.
the labeling pattern of the pyrrolidine and pyridine rings

of nicotine after 14 14

CO2 and acetate-Z- C incorporation by
N. rustica and N, glutinosa. In addition glutamate, a pre-
cursor of the pyrrolidine ring, and aSpartate, a precursor
of the pyridine ring, were isolated in the same eXperiments
and the labeling pattern in each determined. A hypothetical
non-symmetrical pathway for the biosynthesis of the pyrroli-
dine ring of nicotine involving sarcasine (N-methyl glycine)
was evaluated and rejected. The degradation of the amino
acids was performed with the cooperation of Robert M. O'Neal,
Linda C. Burns and Roger E. Koeppe from the Department of

Biochemistry, Oklahoma State University, Stillwater, Oklahoma.

EXPERIMENTAL PROCEDURE

Plants were grown in a greenhouse under conditions of
controlled light and temperature. For studies of acetate-2-
lLI'C incorporation, 337 N, rustica plants with hydroponically
regenerated roots were used as previously described (42).
The acetate, purchased from.Tracerlab Corporation, had a
Specific activity of 8.56 mC/mmole. Purity was determined
by paper chromatography and radioautography. Ten to 40 no

14C in 1 ml of water was administered

of sodium acetate-2—
per plant and additional water was then added as required.
Light of 300 foot-candles intensity was supplied by two 30-
watt fluorescent bulbs and two 150-watt incandescent bulbs.
After 2 hours the roots were rinsed with distilled water,
the plants cut into pieces with scissors and blended for 1
minute with boiling water in a large stainless steel blender.
The nicotine was isolated from the blended material as the
dipicrate (42). The nicotine dipicrate was transferred to a
steam distillation apparatus and the solution was made basic
and steam distilled. The nicotine was extracted from the
steam distillate with ether. After drying, the ether was
removed by evaporation to yield free nicotine.

When the isolation of glutamate and aSpartate was

desired, the plants were blended in 90% boiling ethanol.

The blended material was refluxed on a steam bath for 30

11

12
minutes and the mixture was filtered, refluxed for 1 hour
with fresh 80% ethanol and filtered again. The volume of
the filtrate was reduced, the solution made strongly alkaline
with KOH and extracted 4 times with ether. Water and HCl
were added to the ether and the mixture was flash evaporated
to remove the ether. The nicotine was then isolated from
this solution. The alkaline aqueous solution containing the
amino acids was neutralized with HClOu, the mixture was
cooled overnight, and the KClOu which formed was removed by
centrifugation. The supernatent was treated with charcoal
to remove colored substances. ASpartate and glutamate were

isolated, degraded and assayed for 1“

(44).

C as previously described

Plants used for 14

002 eXperiments were grown in flats
until 2 weeks prior to the feeding study. At this time, 3
to 5 plants ranging in height from 20 to 25 cm were trans-
planted to round enamel pans containing 8011. Five days
before the feeding eXperiment, the plants were topped. The

14002 was administered to the plants in a 21.5 liter desic-

cator in which the 1”

002 was released by adding 50% lactic
acid to Ba14003. Two m0 of 14002 was generated per plant.
The BaluCO3 was purchased from New England Nuclear Corpora-
tion. After the 1["002 had been generated, the plants were
allowed to metabolize the labeled gas in the sealed desicca-
tor for one hour after which time air was flushed through
the desiccator into saturated Ba(OH)2. Essentially all the

14CO2 had been fixed by the plants since nearly all the

13

12002 in the air used to flush the

BaCO3 formed resulted from
feeding chamber. The desiccator top was removed after 1/2
hour of flushing and the plants maintained at normal CO2
tension for the remainder of the exPeriment. The temperature
in the desiccator rose from 25° to 27° during the first 1%
hours of the eXperiment. The feeding chamber was illuminated
with water-cooled light from two 300-watt Ken-Rad flood
lights. The flood lights provided a light intensity of 2400
foot-candles when measured with a Western illumination meter,
Model 756. Total metabolic periods lasted 3 or 6 hours at
which time the exPeriments were terminated and the amino acids
and nicotine isolated as described previously. Nicotine from
N, glutinosa was further purified by distillation through a
Widmer column as described by Smith (45). Three N, rustica

plants were used for the 6-hour 1“

CO2 eXperiment. Eight
N, glutinosa plants were used for the 6-hour eXperiment and
13 plants for the 3-hour eXperiment.

The degradation of the pyrrolidine ring to obtain car-
bons 3', 4', 5' and the methyl carbon was accomplished by the
method of Liebman, Mundy and Rapoport (23), with minor modi-
fications. The pyridine ring of nicotinic acid, obtained as
a byproduct from the Liebman.§tha; (23) degradation of nico-
tine, was degraded by the method of Jackanicz and Byerrum
(27) with one modification. The N-methyl-Z-pyridone, dis-
solved in 100 ml of 95% ethanol, was reduced to N-methyl-Z-
piperidone with 250 mg of 10% palladium on charcoal in a

Paar low-pressure hydrogenation apparatus under 50 psi of

14
hydrogen for 6 hours at 75° (46). Nicotinic acid was decar-
boxylated to yield C-2' of the pyrrolidine ring. Decarboxy-
lation was accomplished by heating 100 mg of nicotinic acid
mixed with 375 mg of freshly prepared copper chromite catalyst
(47) in a Wood's metal bath preheated to 270°. Two traps were
connected to the decarboxylation apparatus. The first con-

tained 4% H010 in isopropyl alcohol and the second saturated

Ll.

Ba(0H)2. CO free-nitrogen was used to flush the pyridine and

2
C02 from the decarboxylation apparatus. The reaction was
usually completed in 1 hour yielding 50 mg of pyridine
perchlorate and 120 mg of BaCOB. The pyridine perchlorate
was recrystallized from ethanol and ether. Scott recently
published a similar method for the decarboxylation of nico-
tinic acid (48).

Dimethyl glycine labeled with 14C in carbon 1 or 2
or the methyl group was synthesized by the method of Bowman

140, glycine-Z-luc or for-

14

and Stroud (49) using glycine-1-
maldehyde-luC. The formaldehyde- C was obtained by heating
uC-paraformaldehyde in aqueous solution at 1100 overnight.
In determining the Specific activity of a compound by
either scintillation or planchet counter, an average of 5
replicate samples were counted per compound. All compounds,
except BaCOB, were recrystallized to constant Specific activ-
ity and 2 to 8 mg were counted on a Packard Tri-Carb Scintil-
lation Spectrometer, Model 3310. All counts were corrected

for background and efficiency of the counter. The efficiency

of the counter was approximately 92% as determined with a

15
benzoic acid-7-14C standard purchased from New England Nuclear
Corporation. Carbon dioxide from carbons 2' and 4' of the
pyrrolidine ring was counted as BaCO3 on a Nuclear Chicago
Model C115 low background automatic sample changer with a
Model 8703 decade scaler. All counts were corrected for
background, self-absorption and efficiency of the counter.
BaCO3 standards for the low background counter were prepared
by total oxidation of a compound of known specific activity
to CO2 (19), trapping the 002 in saturated Ba(OH)2 and count-
ing the BaCOB. The efficiency of the low background counter
was approximately 12%.

RESULTS AND DISCUSSION
Specificity of the Degradation:

The main obstacle to overcome in the study of the
pyrrolidine ring of nicotine is the degradation procedure.
At present there are 4 general degradative schemes in com-
mon use (1, 6, 12, 23). The procedure of Lamberts and
Byerrum (1) required large amounts of starting material
while that of Wu and Byerrum (6) and Leete (12) failed to
yield the individual carbons of the pyrrolidine ring in a
unique manner. The procedure of Liebman gt a; (23) required
the least quantity of starting material and yielded the
individual carbons of the pyrrolidine ring. The degradative
scheme is shown in Figure 4. After eXploring 3 of the above
degradative schemes (1, 6, 23), the scheme of Liebman 22.3;
was decided upon. The Specificity of the oxidation of
dimethyl glycine by lead tetraacetate to 002, formaldehyde

14c-dimethyl

and dimethyl amine was tested by degrading
glycine labeled in the various carbon atoms. The results
are listed in Table 1. When dimethyl glycine was labeled
either in carbon 1 or 2, more than 99% of the radioactivity
was recovered in the eXpected degradation product. About
0.1% was found as a contaminant in other products. The
degradation of dimethyl glycine-methyl-luc indicated that
the compound was not uniquely labeled. Only 90.5% of the

16

1?

Ammv ”mm mm. Hammad 89H.“ oaosom nodpoowhmom wmam onEHoaHmm

3: mmDUHm

 

N m m m
5 8 z . :8 T o r
.0: amomorcocm rzx

/ n
o r
o r. o +
«so
ill 8w: afllul
. :0 00A n 3490.".
+
nooom o «8..
c «£98
mxo mro
/ \ oxen
z
o nro nro m8: nro mro m:
/z\ ozoo .m ,2\ com... a
o n
.58 + as Jamel m. s. imam... m rlsfl
_ n e
o o o co...
r o _
nroo
oroo oxoo m
:0. \z
z

 

19

 

 

 

 

 

 

 

 

m .
mo
m.oe one.sms e.n oma.ma m.a oam.m eeo.amm m .wz-
s.
ea.o mam.a s.oo soo.osa ma.o oma ooo.m:e immcu
eo.o com no.0 mm s.oo ocm.e:c ooa.omo uooom
m za\sao a za\sco a SS\sao ASS\saov assessed apnea co
oaeaocdm waspadpm soapaoom
mmo/ m
z. . mo- iooom
mmox

 

 

 

 

muopndo adsoaeaosH on» ma scaeapodoaocm

 

 

 

cascade ooaondq saacoacaooam co coapdocswcn .H mqmda

20

radioactivity was recovered in the N-methyl derivative while
7.2% of the label was found in the other two derivatives.
If the label in the latter two derivatives had been due to
contamination from 1“C in the N-methyl groups, the label in
the N-methyl derivative would have equalled 100% since
specific activity is SXpressed as dpm.per mmole. The total
percent of radioactivity recovered in the degradation prod-
ucts of dimethyl glycine-methyl-luc was 97.7%. It was con-
cluded from this study that the degradation of dimethyl
glycine was highly specific.

Since the work of Liebman g§_§l_(23) indicated low

14

incorporation of label into C-2' from CO the accurate

2.
determination of radioactivity in C-2' was critical.
Previous methods for decarboxylation of nicotinic acid
involved heating nicotinic acid with calcium oxide at 3900
or higher. However, previous studies have shown that this
results in breakdown of the pyridine ring to yield 002
(8, 25). When nicotinic acid with 98% of the radioactivity
located in the carboxyl group was decarboxylated with CaO,
only 75 to 85% of the total radioactivity was recovered in
the BaCO3 (8). After several decarboxylation
methods proved to be ineffectual, the decarboxylation of
nicotinic acid was successfully accomplished by heating nico-
tinic acid with copper chromite.

Nicotinic acid-7-1uc purchased from New England

Nuclear Corporation and nicotinic acid-6-1uC purchased from

Nuclear Chicago Corporation was decarboxylated with copper

21
chromite to verfity that the BaCOB formed was solely due to
the carboxyl group of nicotinic acid. In each case 7 suc-
cessive BaCO3 samples were collected and counted. BaCO3
samples 2 through 6 from nicotinic acid-7-luc contained from
94 to 98% of the eXpected radioactivity while sample 1 con-
tained 69% and sample 7 contained 84%. Pyridine perchlorate
contained 0.1%. Pyridine perchlorate from nicotinic acid-6-
14C contained 99.6% of the eXpected radioactivity. BaCO3
samples 2 through 6 contained 0.6 to 1.3% of the original
radioactivity in nicotinic acid-6-14C. Samples 1 and 7
contained 7.8 and 7.2% reSpectively. The first and last
samples, which contained a limited amount of C02 from break-
down of the pyridine ring, were not used in determining the
Specific activity of carbon 2'. Since the pyridine ring
contains proportionally a greater amount of radioactivity

than the pyrrolidine ring, the SXperimental Specific activ-
ity obtained for C-2' is the upper limit of the true value.

Incorporation of Label into Glutamate and ASpartate:

The data obtained in the degradation of glutamate and
14

 

aSpartate from.N, rustica and N, glutinosa fed C02,
acetate-l-luc and acetate-2-14C are presented in Table 2.

Glutamate and aSpartate were not isolated in the 3 hour
1”C02 SXperiment. After 6 hour 14002 incorporation by both
Species, carbons 2 and 3, as well as carbons 4 and 5, of
glutamate were labeled equally. Approximately 30% of the

label was found in carbons 2 and 3 and about 50% in carbons

 

22

 

 

 

 

 

 

 

 

 

 

o.ooa
m.oo
m.mm 3.55
o
m.mm o
m.b . . o H.ma
o.o m m o an o.: . e 0 late
can 3 . a m m; 3 heroes
0 H N000 We N U QOH¥m5zH cl
m.os s.mm m. m m.m N.H Saimiopdpcod. z
s. cm 3.3m m.ma . H o a i doapusw em
duo mm m.sm m m m SH m.om SH m cpdpcod. cod
Ugh.“ I o N o 0 PWSH cl
0 one H 0 SH .60 0 ma 0.:H . No 2
SH aboard R cmwwmmmm o a mic m SH on: o m odd dmoSHpsa am
He duo i o
i HMO m U NIO HlU 03H dOdeHpH om
Hpsnancnmn o a. cane com ossoaao
o
mmdoogm

 

 

dd

 

opopamdmd

opdadpdao

 

 

 

 

 

M

23
4 and 5. This labeling pattern correSponds well with that
obtained by Burns £3 E; (50) for glutamate isolated from
N, rustica leaves eXposed to 14002 for 3 and 18 minutes.
These authors found that after the 3 minute 14C02 eXposure,
carbon 1 of glutamate contained about 9%, carbons 2 and 3
each contained about 2% and carbons 4 and 5 each contained
about 43% of the radioactivity. After 18 minutes the dif-
ference in radioactivity of the glutamate carbons diminished.
Burns 23 a; postulated that glyoxylate condensed with
oxalacetate to form oxalmalate, y-hydroxy-d-ketoglutarate,
d—ketoglutarate and finally glutamic acid. This is the same
pathway proposed by Sekizawa gt a; (51) for Acetobacter
suboxydans, an organism with a largely nonfunctional tricar-
boxylic acid cycle. Whether this pathway is operative in
higher plants is questionable. However, v-hydroxy-a—keto-
glutarate seems to be present in plants (23). In addition,
the labeling pattern observed by Burns 33 g; is in agreement
with the proposed pathway for glutamate biosynthesis. If
carbons 4 and 5 of glutamate are derived from glycolate via
glyoxylate it is eXpected that these carbons would be rapidly
and equally labeled (54). Carbons 2 and 3 of glutamate
would originate from the internal carbons of oxalacetate and,
therefore, would also be equally labeled. In short term
eXperiments carbon 1 of glutamate, derived from carbon 1 of
oxalacetate, would be more radioactive than carbons 2 and 3
of glutamate if the carboxyl groups of oxalacetate are more

heavily labeled. Thus Burns 23 a; (50) observed that after

24

a 3 minute 14CO SXposure, 80% of the radioactivity of

2
aSpartic acid was located in the two carboxyl groups. In

Table 2 it is observed that the percent of radioactivity of

14

aSpartate in the carboxyl group after 6 hour CO2 metabo-

lism is approximately 50%. This would indicate nearly uni-

form labeling of aSpartate and, presumably, oxalacetate.

14
CO2

incorporation shows equal labeling of carbons 1, 2 and 3,

Although glutamate isolated from N, rustica after

this is not the case for N, glutinosa where carbon 1 of
glutamate contained 20% and carbons 2 and 3 about 14% each.
This difference is not great enough to rule out the hypothe-
sis of Burns gt.§;,

After 1 and 2 hours of acetate-Z-luc metabolism by

N. rustica carbon 4 of glutamate contained 90.2% and 81.0%

4

of the total 1 C, rSSpectively. ASpartate carboxyl groups

contained 6.8% after 1 hour and 25.8% after a 2-hour metab-
olism period. Carbon 5 of glutamate contained 77.4% and

carbon 1 contained 19.9% of the label after N, rustica

14

metabolized acetate-i- C for 6 hours. ASpartate in this

exPeriment had 100.0% of the radioactivity in the carboxyl

groups. Glutamate and aSpartate were labeled as SXpected,

14

if it is assumed that acetate-i- C.and acetate-2-140 are

metabolized to form these amino acids via the tricarboxylic

14

acid cycle. Relatively little randomization of C is

apparent.

25
Incorporation of Label into the Pyrrolidine Ring of Nicotine:

Table 3 shows the distribution 0 C in the pyrroli-

dine ring of nicotine after 14CO2 incorporation. The number-

ing systems for the pyridine and pyrrolidine rings of nico-

14

tine and the C-distribution pattern for nicotine from N,

14
CO2

|5 4\\3 4.3
63
N

77.8%

glutinosa after 6 hours of incorporation is Shown below.

 

 

 

Nearly uniform labeling of the pyrrolidine ring was observed

in all 1“‘00

2 eXperiments. If glutamate were incorporated
into the pyrrolidine ring of nicotine via the pathway involv-
ing a symmetrical intermediate proposed by Leete (Figure 2),
carbon 1 would be lost and the remaining carbons randomized
as shown in Figure 5. From the figure it is apparent that
when carbons 4 and 5, as well as carbons 2 and 3 of glutamate
are labeled equally, all the carbons of the pyrrolidine ring
will be labeled equally. When this prodiction is compared
with the results shown in Table 3, it is observed that there
is no essential difference in labeling among any of the car-

bons of the pyrrolidine ring after 1“

C02 incorporated for 3
or 6 hours by either N, rustica or N, glutinosa.

Liebman, Mundy and Rapoport obtained ratios of C-5'

 

 

 

 

 

 

m.s om H «.ma 0mm u m.ma cams w mmo..m..s Hum
0mm.o oom.sm oom.oe .csaoaawassecaanuz.z
m.m can H m.ma omoa u m.om chm H mmo..m..s..m Hom.cSHsch
060.5 oam.as oms.am umuHsSDcsHouz.z
s.s on H s.ma cam H o.am can H mmo..m..s..m oddsdacum
omm.s osH.om onm.mm nasnecauzuaaoNSomuz
s.oo omoa H m.ss ommn H 0.:s oooa H
oom.:w onm.mmm oeo.bom Ham cpsaoHSoaca csaeaaam
o.a on H m.: eon u e.e eon u m
owe.a cab.ma om:.ma .N oosm
,o m.No one a o.mm comm w m.os coma H
2 ooo.om omm.apm oom.mam .m.sam oaoo oasabooaz
o.ooa or: H o.ooa oaem u o.ooa Donna M mmo..m..s
omm.mo omo.mms omo.som ..m..m.csam ceapooassncaasonscmuz
R oaoaa\ado R oaoaa\adu & oHoaa\ado muonnoo moHQBSm
OWHS m 0mm“ 0 II

Smocapsam am

.m
cmozapsam .m

 

doapmsn .z

 

 

 

 

 

ScapegoanooSH mound

hopes orapooaz co scam ceaoaaosnam on» ca 0

ea

90 noapfindhpmam .m mqmda

27

opooaadoh no

.moaaadm

moahom a mo came 0:» Scam oozaahopoo mm coma on» aonm Scandaboo uncondpmn

.onaoahmd .Hmm ma

poms Soapoabonnpd ones

 

es

 

 

5.:

:.:

 

omm H
one am

can H
omm.ma

omm H
omm.om

one u
oHo.mH

 

m.o

m.:

m.:

 

com H
oom.sm

com H
omH.mH

omm H
oom.ma
one H
oso.oa

 

mmo

.m
.3

.m

modaSGOMHSm
uncommonoaohn

-mraascoaamiz.z

onoooaau mozzooamaaom

moowm

odes oAONSom

 

28

 

.ooapmSH am an SmHH

ionwpoa moo do mason o gonad endpooac mo mafia oSHcaHOHsz

3H
oSp mo Scandaaom on» ma opdadpzaw do soapmnaaoosdh oomOQOHm

.m mmDOHm

 

 

 

:8 838.....2305Eo
oczoez 5353522622 1.28.2-5-..
mro mmo nro
z :2 0
—SN .3 _N
\ «N _N .N. Ml N. _N /.x.N. _N
.88 .sroNAIll so. 0N $qu so 8 so .8
332:5 32:50
so .3

er er rooo

Mimfl marl... SEN

3m .mN $5.0. $QDN

 

flan.

osomotagrzozuz

 

nrw

:2

SE. “A”: S
AlilsmoN son.

828.228 25220

oxen .3
o 1000

.1 at

Jemdm Rafi.

 

mzo

ATMISJ so. Mull.

$m. 0N arm .0.

 

22:23.0

30.2
I000
:000
new NN

2N:

$0.3

$mfiN arm“.

30

label to C-2' label of 3.1, 4.3 and 1.6 from three 6 hour
14002 incorporation experiments with.N, glutinosa (23). The
ratios of C-5' radioactivity to C-2' radioactivity in the
present study with N, glutinosa were 0.95 and 0.94 after 6
and 3 hour periods, reSpectively. With N, rustica the ratio
was 1.05 after 6 hour l[+002 incorporation. These values are
essentially equal to unity. If an unsymmetrical pathway
existed, non-symmetry should be more evident in a shorter
period of metabolism after the initial uptake of a labeled

precursor because of less randomization of the 1”c.

However,
the ratios of C-5' label to C-2' label after 3 and 6 hour
metabolism of 1LL’COZ by N, glutinosa were the same. At
Shorter times a greater percentage of the 14C was found in
the pyridine ring as previously observed (24).

The distribution of 140 in the pyrrolidine ring of

140 for 2 hours is

nicotine as N, rustica fed acetate-2-
shown in Table 4. The label in C-2' and C-5' as well as in
C-3' and C-4' is the same within eXperimental error; 2.6%

140 in the nicotine molecule were found in

and 1.8% of the
C-2' and C-5' reSpectively. C-3' contained 16.4% and C-4'
15.6% of the total radioactivity in nicotine. Good correla-
tion existed among the counting data for N-Benzoyl—N-methyl-
B-alanine, N,N-dimethyl-B-alanine°HCl, N,N-dimethylglycine-
HCl and the sum of the individual carbons composing these
compounds.

If the glutamate from the 2 hour acetate-Z-luc eXperi-

ment were randomized in the manner shown in Figure 5 before

31

mo moHHom c

.moagaom opsoHHdoH

.HO $08 03». 80H.“ Umfldamfimd m6 S605 03». 309.“ SOHPGHPOU UHGUHHNPmD.

.oSHoHHha .Hhm mH cons SOHppooHQQS mass

 

 

w.H cm H can mmo odHadsOdem

locoNnonoaoanimrHHSHoaHaiz.z
m.a oN H ode .m odoocsNo cosscoHosaom
o.ma one H omm.m .s moocm
s.oa oo H omm.m .m oaoo oaoNscm
m.oa oNN H ooo.o mmo..m..s Hom.osHoaHNHssecaanuz.z
o.sm omH H ooN.NH mmo..m..s..m Hum.osHsdenmiHNseoSHnsz.z
m.Sm ooa H oom.NH mmo..m..s..m osHsNHououaanpcSnzaHNoNsomuz
s.so oNs H oNN.mN ham cassoHsoNoa osaoaasm
o.N oHN H omo .N moodm
a.oo om: H oao.mN .N.asm odor cacapooaz
o.ooa noHN H osm.nm mmo..m..s..m..N.casm czapooasopcaHNoNSomuz

R oHoaa\ado . mgonhmo magadm

 

 

 

 

 

 

ooHpmsn am Scam oSHHoon do

 

 

 

 

Ill.)

aoHHonccoz o Himucpspood No mason N nosed
msam canoaaons N one ca 01: No consonaneoao .s Hands

32
incorporation into the pyrrolidine ring, C-2' label would
equal C-5' label and C-3' would equal C-4'. The observed
ratio of C-5' radioactivity to C-2' radioactivity shown in
Table 4 was 0.69. However, because of the lower Specific
activity of nicotine synthesized in this SXperiment and the
low incorporation of lLAC into the pyrrolidine ring, these
ratios are not as accurate as in the previous eXperiment.
The ratio of C-4' label to C-3' label is more reliable since
C-3' and C-4' had a higher incorporation of 140. This ratio
equaled 0.95. Wu and Byerrum observed equal labeling of C-2'
and C-5' as well as C-3' and C-4' after 48 hour acetate-2-
140 incorporation by N, rustica (6).

140 incor-

It is predicted from the 2 hour acetate-2-
poration into glutamate that the ratio of C-3' label to C-2'
label or the C-4' to C-5' ratios should equal 10.4. Since
only 95.3% of the radioactivity of glutamate was recovered
after the glutamate degradation, the value of 10.4 may not
be entirely accurate. The experimental ratios obtained
from the degradation data in Table 4 for the C-3' label to
C-2' label and the C-4' to C-5' and 6.3 and 8.7 reSpectively.
The ratio of C-3' label to C-5' label is probably the most
accurate Since both C-3' and C-5' were determined by scintil-
lation counting and did not involve possible errors due to
decarboxylation or planchet counting. This ratio is 9.1 and
is in good agreement with the predicted value of 10.4.

The results in this study are consistent with a path-

way involving a symmetrical intermediate in the biosynthesis

33
of the pyrrolidine ring of nicotine and with previous data

obtained from precursor eXperiments. Species differences
were not observed. The reason for the discrepancy between
these results and those obtained by Liebman 23 g; are not
apparent unless they are related to the relatively low
Specific activity of nicotine degraded by those investiga-
tors. Although the scheme presented in Figure 2 for the
biosynthesis of the pyrrolidine ring of nicotine is in
agreement with the majority of data now available, the pos-

sibility of modifications cannot be eliminated.
Study of a Hypothetical Non-Symmetrical Pathway:

In initially considering various alternatives for the
biosynthesis of the pyrrolidine ring, a pathway was proposed
that would allow non-symmetrical labeling of the pyrrolidine
ring. The most consistent feature of the earlier 14C02
incorporation studies by Rapoport's group with.N, glutinosa
(23-26) was the equal labeling of carbons 4' and 5' and the
low label in carbon 2'. Carbon 3' had a Specific activity
Similar to C-4' and C-5'. The proposed pathway was designed
to meet several requirements: (1) it should eXplain the
labeling pattern observed by Liebman 32 a; (23), (2) the
reactions should utilize physiological compounds via prob-
able reaction mechanisms, (3) the pathway Should involve
relatively few and simple compounds, and (4) it should be
feasible to test the hypothetical pathway. On this basis

it was proposed that acetate and sarcosine (N-methyl glycine)

34
condensed to form the N-methyl pyrrolidine ring of nicotine.

The hypothetical pathway is shown in Figure 6.

 

 

 

.x.
CH C H
l 3 + '00 * 0 *-
HOOC /CH2 -———e> -——+>r——+> L
HN HOOC H00
I
CH3 é OH
H3 3
* . at
-———+> .;::::i’ ____g>
./’ L\\
HC \‘N' N
u NH ' ) I
O l \\ CH
FIGURE 6

Hess and Tolbert (54) have shown that serine, a pre-
cursor of choline and therefore sarcosine, is equally labeled
in all three carbon atoms 11 seconds after eXposure of N,
tabacum leaves to 1'L‘COE. This observation would eXplain
equal labeling in carbon atoms 1 and 2 of sarcosine. Byerrum,
Sato and Ball (22) observed the formation of labeled nicotine,
betaine and dimethyl glycine after feeding choline-methyl-luc
to N, rustica. Although no mention was made of labeled
sarcosine, the possibility existed that the methyl group
labeling the nicotine was part of sarcosine rather than an
"active C-i fragment."

In order to SXplain the low labeling of C42' observed
by Liebman SENS; (23), the carboxyl group of acetate has to

be less rapidly labeled by 14C02 than the methyl group.

35

Unfortunately the biosynthesis of acetate from 1”

COZ in higher
plants has not yet been studied in detail. If acetate arises
from phOSphoenolpyruvic acid or from a phOSphoclastic cleavage
of thiaminepyrophOSphate glycolaldehyde in the photosynthetic
carbon reduction cycle (55), it is difficult to envision

unequal labeling of the acetate carbons from 14

C02. However,
in the present hypothesis, it is assumed that acetate, or
perhaps some other compound condensing with sarcosine, may
possess the necessary labeling pattern. It was postulated
that the reaction of the methyl group of acetate and the car-
boxyl group of sarcosine was analogous to the reaction of
acetyl-CoA with the carbonyl group of oxalacetate.

To test the proposed pathway, the dilution factors
(the Specific activity of the compound fed divided by the

Specific activity of nicotine) for acetate-1 and 2—140,

choline-1,2-1uC and sarcosine-l and methyl-l“

C were compared.
Nicotine samples were partially degraded if they contained
sufficient radioactivity. The results are shown in Table 5.
Seventy-eight percent of the radioactivity incorporated into
nicotine after 6 hours of choline-1,2-1uc metabolism was
located in the pyrrolidine ring and the N-methyl group. The
label in carbon 2. was 7%. This indicates that the majority
of the label is located in carbons 3', 4' and 5' plus the
N-methyl group. This is in agreement with the proposed path-
way. However, Byerrum, Hamill and Ball (56) observed that

carbon atom 2 of glycine almost exclusively labeled the

N-methyl group of nicotine. Therefore, the possibility

.thproo OHHHoodm ..4.m mH moms mnoHppooandd

 

36

 

omNs in- as- in- mow w am.o on.m ooa a as.m N owHuHNSpca
i :Hmoonmm
in- it- in: in- oa w H.mN omo.m oa w No.a o o

m 0H iHionHmoonm

in- omN.a ooe ooo.a mon w NH.N oNN.m mod N mo.s o o a

iN.HiosaHomo

in: in- is: is: moH w m.oH ooo.mH oHoH swam.a o oeHancpmpooa

oom oaN.mN omm oao.mN mod N an.m oom.mm oaoa w oo.a .mae N oanNucedpood
oHoaa oHoaa

\ado oHoaa\amo \ago oaoaa\ado oaoaa\amo oHoaa\amo
are made .N owes Noeocm oeaeooaz ocm escoaeoo cane oom ossoSSoo
ceaoaaam oHsHpooaz eoHDSHHm _No .<.m Hp .<.m . .

 

 

 

 

 

 

 

 

 

 

oSHpoon ounH mopHHonmpoz Hanan mo SOHpoHomHoonH

.m mqmda

3?

exists that choline was metabolized to glycine and then
incorporated into the methyl group. Byerrum gg‘gl (56)
observed no measurable amount of glycine-l-luC incorporation
after a 7 day metabolic period. Unfortunately, low incor-
poration of choline-1,2-14C into nicotine prevented the iso-
lation of the N—methyl group.

In addition, the significance of the above proposed
pathway is doubious due to the low incorporation (high dilu-

tion factor) of choline-1,2—14

C into nicotine. If sarcosine
is a direct precursor of the pyrrolidine ring, it is eXpected
that the dilution factors for choline and sarcosine would be
lower than dilution factors for compounds, like acetate, which

140, like choline-1,2-

are more distantly related. Acetate-1-
luc, labeled mainly the pyrrolidine ring of nicotine (5),
however its dilution factor is similar to that of choline-1,2-
140. The dilution factor for sarcosine-i-luC is three times
as great as that for choline-1,2-1uC. In other words, choline-
1,2—luC was incorporated more efficiently than sarcosine-l-
1“'C. This observation, and those of Byerrum g; g; (56) on
glycine-1 and 2-140 incorporation, indicate that the label
from choline-1,2-1“C and sarcosine-i-luc may have been
incorporated via glycine.

The incorporation of sarcosine-methyl-luc for a 2 hour
metabolic period is similar to that for acetate-Z-luc. How-
ever, sarcosine-i-luC after 6 hours has a dilution factor 4

times as large as sarcosine-methyl-luc after 2 hours. If

sarcosine was incorporated l2 toto into the pyrrolidine ring

38

of nicotine, the dilution factor for sarcosine-i-luc and for
sarcosine-methyl-luc Should be equal at the end of any meta-
bolic period. Since this was not observed and since the
overall incorporation was low, it seems likely that the
proposed pathway for sarcosine incorporation contributes
little, if anything, to the normal biosynthetic pathway for
the pyrrolidine ring of nicotine. It is probable that the
label from sarcosine-methyl-luc was incorporated via trans-
methylation. The participation of acetate in the biosyn-
thesis of the pyrrolidine ring as envisioned above has also
been eliminated by the data in the previous section which
indicated that acetate-2-1“c was incorporated symmetrically.

Leete (12) has also postulated a non-symmetrical
pathway for the biosynthesis of the pyrrolidine ring of
nicotine. In Leete's pathway glycolic aldehyde condenses
with acetate to yield 3,4-dihydroxybutyric acid. Dehydra-
tion followed by transamination and N-methylation would
yield 4-(N-methyl)-aminobutyraldehyde. In order for unsym-

metrical labeling to occur, the labeling pattern from 14

002
incorporation into acetate has to differ from the labeling
pattern in glycolic aldehyde. This mechanism, as well as
the mechanism proposed earlier, fail to SXplain unequal
labeling of acetate. If the 14002 incorporation studies in
the previous section are correct, neither of these non-sym-

metrical pathways are necessary to eXplain the observed

data.

39
Incorporation of Label into the Pyridine Ring of Nicotine:

The distribution of 1”C in the pyridine ring of nico-
tine after 1L’COZ incorporation by N, rustica and N, glutinosa
is shown in Table 6. After a 3 hour metabolic period, the
labeling pattern from 1”C02 in the pyridine ring of nicotine
from N, glutinosa is striking. Table 6 Shows that the
Specific activity of carbons 4, 5 or 6 is 2% times as great
as that of carbons 2 or 3. After a 6 hour metabolic period
with N. rustica, 14002 labeled the pyridine ring in a similar
manner, however, greater equilibration had occurred among
the carbons.

Since the evidence presented in the Introduction
indicated that aSpartate and glyceraldehyde condense to form
the pyridine ring,the most obvious eXplanation is to assign
the two groups of equally labeled carbons in the present
eXperiments to two different precursor molecules. The two
precursor molecules would have to meet two requirements.
First, the carbons within each precursor that forms the
pyridine ring would have to be equally labeled. Second, the
Specific activity of the two precursors has to differ at the
time of pyridine ring synthesis. The first requirement is
met by aSpartate and 3-phOSphoglycerate, which is closely
related to the 3-carbon precursor. The two center carbons
of aSpartic acid, which give rise to carbons 2 and 3 of the
pyridine ring, are randomized by cycling through the tricar-
boxylic acid cycle. Hess and Tolbert (54) have shown that

3-phOSphoglycerate is essentially uniformly labeled in

4O

.oHod oHondpcodonHadimiHHSHNZLm op oohdmaoo mm mprHpoSOHodH mo owdpnoonoms

 

 

 

 

 

 

o.mN 23.le NaoN oom.m o cpodopaco 5:8
SN oam.Nm N.sN onum m oedeonwco 333m
e.mm omo.mm o.mm OHm.m : oposopwdo SSHHcm
0.:H owo.HN m.m 05H.N m cannonawo aanmm
H.mH oom.mH 0.0H oom.m N opoaonhdo SSHHdm
m.Ne omm.mm e.mm oom.mH m.m opdpooc SSHoom
m.Nm oom.mm o.m: omo.OH :.m.m oHNSOHmowa asHoom
o.ooH oNH.omH o.ooa osoéN o.m.s.m.N Soc 323ch

sodaacimiifioziz

*& oaoaa\aao *R oaoaa\ano muonnmo oqsogaoo
.m w_l ,m _I
[‘1' (ll
SOHpmHoQHooQH

 

 

 

 

 

 

 

 

 

 

 

 

Nooda Hopw4 SQHPOOHZ mo wCHm DGHdHHhm 0:» ma 03H mo Godpfindhgmdm .0 mflmda

41
leaves from.N, tabacum after 1 minute photosynthesis in
14002. The conversion of 3-phosphoglycerate to 3-phOSpho-
glyceraldehyde would eXplain the uniform labeling of carbons
4, 5, and 6 of the pyridine ring.

The second requirement, that the two precursors have
different Specific activities at the time of pyridine ring
synthesis, is met if the labeled aSpartic acid is diluted
out rapidly before incorporation into the pyridine ring due
to pool size or turnover rate and/or if the rate of incor-

poration of label from 14

C02 into the internal carbons of
aSpartic acid is slower than into glyceraldehyde or a related
3-carbon compound. Since the synthesis of nicotine is Slow
(24) and the central position of aSpartic acid in metabolism
results in a high turnover rate, it is anticipated that the
carbons of the pyridine ring derived from aSpartic acid would
be relatively cold. Like 3-phOSphoglycerate, aSpartic acid
is rapidly labeled by 1”C02 (57). However, Burns §£_§l_(50)
have demonstrated that aSpartic acid is not uniformly labeled

14

in N, rustica leaves after 3 or 18 minute eXposure to C02.
After an 18 minute eXposure, more than 70% of the label was in
the two carboxyl groups of aSpartic acid. Thus the rapid
labeling of aSpartic acid does not truly reflect the rate of
labeling of the carbons that form carbons 2 and 3 of the
pyridine ring of nicotine. Although these considerations
support the hypothesis that glyceraldehyde and aSpartic acid

condense to form the pyridine ring of nicotine, it does not

eliminate the possibility that one, two or more compounds

42
with unequal internal labeling may be reSponsible for the
observed labeling pattern of the pyridine ring.

The labeling pattern of the pyridine ring of nicotine
from N. rustica after 2 hours of acetate-Z-luc metabolism
shown in Table 7 correSponds well with that observed by
Griffith and Byerrum (29). Using a partial degradation of
the pyridine ring, they observed that more than 70% of the
total radioactivity in the pyridine ring was equally distrib-
uted between carbons 2 and 3 after 6 hour acetate-2-1uC
incorporation into nicotine by N, rustica. Anabasine, an
analogue of nicotine, was similarily labeled in the pyridine
ring after administering acetate-Z-luC to N, glauca (31).
This labeling pattern is explained by assuming that acetate-
2-14C is incorporated into the pyridine ring of nicotine via
aSpartic acid. This would also eXplain the low incorpora-
tion of acetate-1-1“c into the pyridine ring (58).

The present 14CO and acetate-Z-luc eXperiments fur-

2
ther collaborate the pathway for pyridine ring biosynthesis
previously proposed on the basis of feeding chemically syn-

thesized intermediates.

43

TABLE 7. Distribution of 140 in the Pyridine Ring of Nico-
2 Hours of Acetate-2-14C

tine From.y, rustica After

 

 

 

Metabolism
Compound Carbons dpm/mmole %*

fl-Methyl-S-amino-

pentanoic acid 2.3.4.5,6 11,170 100.0
Sodium acetate 5.6 1,350 12.1
Sodium propionate 2,3,4 9.630 91.3
Barium carbonate 2 4,990 44.7
Barium carbonate 3 4,780 42.8
Barium carbonate 4 460 4.1
Barium carbonate 5 790 7.1
Barium carbonate 6 530 4.7

 

 

 

 

*Percentage of radioactivity as compared to §¢Methyl-5-amino-

pentanoic acid.

SUMMARY

This study supports the biosynthetic pathways for the
pyrrolidine and pyridine rings of nicotine previously pro-
posed from incorporation eXperiments with chemically synthe-
sized intermediates. The present evidence indicates that
the chemically synthesized intermediates are incorporated
into nicotine through normal physiological pathways. The
14002 and acetate-Z-luc data demonstrate that a symmetrical
intermediate is involved in the biosynthesis of the pyrroli-
dine ring of nicotine.

No evidence was found for a proposed non-symmetrical
pathway involving sarcosine for the biosynthesis of the
pyrrolidine ring.

The 14co2 and acetate-Z-luc data is in agreement with
the theory that two compounds condense to form the pyridine
ring of nicotine. The significance of long term 14C02 incor-
poration eXperiments is apparent from this study. In con-

junction with chemically synthesized precursors, 14

002
incorporation under normal physiological conditions can
provide supporting evidence regarding the probable minimum
number of precursors forming the final product. However,
the immediate precursors giving rise to a complex product

cannot be identified solely from long term 1L’COZ incorpora-
tion.

41+

10.

11.

12.
13.

14.

15.
16.

17.

REFERENCES

LambertS, B. L., and Byerrum! 8' U09 J. B1010 Chem., £29
939 (1958).

Leete, E., J. Amer. Chem. Soc., 82, 2162 (1958).

Leete, E., and Siegfried, K. J., J. Amer. Chem. Soc.. 12.
4529 (1957)-

Lamberts, B. L., Dewey, L. J., and Byerrum, R. U., Biochim.
Biophys. Acta, 23, 22 (1959).

Wu, P. -H. L., Griffith, T., and Byerrum, R. U., J. Biol.
Chem., 232, 887 (1962).

Wu, P.-2. L., and Byerrum, R. U., Biochemistry, 4, 1628
(19 5).

Leete, E., J. Amer. Chem. Soc., 8, 3520 (1956).

Krampl, V., Doctoral dissertation, Michigan State Univer-
sity, 1961.

Krampl, V., Zielke, H. E., and Byerrum, R. U., Unpublished
data.

Leete, E., Gros, E. G., and Gilbertson, T. J., Tetrahedron
Lett., 587 (196“).

Mizusaki, S., Kisaki, T., and Tamaki, E., Plant Physiol.,
22. 93 (1968).

Leete, E., J. Amer. Chem. Soc., 82, 7081 (1967).

Essery, J. M.. McCandin, D. J., and Marion, L., Phytochem-
istry, 1, 209 (1962).

Robertsgn, A. V., and Marion, L., Can. J. Chem., 28, 396
(19 0).

Leete, E., J. Amer. Chem. Soc., 84, 55 (1962).

Mothes, K., and snfitte, H. 3., Angew. Chem. (Intern. Ed.),

Anet( E.,)Hughes, G. K., and Ritchie, E., Nature,162, 289
19 49

45

18.

19.
20,

21.

22.

23.

24.

25.

26.

27o

28.

29,

300

31.

32.

33.

34.
35.
36.

46

Clarke, A. J., and Mann, P. J. G., Biochem. J., 1;, 596
(1959).

Tuppy, H., and Faltaons, M. S., Mh. Chem., 2;, 167 (1960).

Smith, T. A., and Garraway, J. L., Phytochemistry, 8, 23
(1964).

Morris, D. H., and Pardee, A. B., J. Biol. Chem., 241,
3129 (1966). "'

Bassham, J. A. in Bonner, J., and Varner, J. E. (Editors),

Plant Biochemistr , Academic Press, New York and
London, 1965, p. 535.

Liebman, A. A., Mundy y, B. P., and Rapoport, H., J. Amer.
Chem. Soc., 82, 664 (1967).

Alworth, W. L., DeSelms, H. C., and Rapoport, H., J. Amer.
Chem. Soc., 88, 1608 (1964).

Alworth, W. L., Liebman, A. A., and Rapoport, H., J. Amer.
Chem. Soc., 88, 3375 (1964).

Liebman, A. A., Morsingh, F., and Rapoport, H., J. Amer.

Jackanicz, T. M., and Byerrum, R. U., J. Biol. Chem., 241,
1296 (1966).

Fleeker6 J., and Byerrum, R. U., J. Biol. Chem., 242, 3042
(19 7)

Griffith, T., and Byerrum, R. U., Biochem. Biophys. Res.
communOO .1_O_0 293 (1963)-

Fleeker, J., and Byerrum, R. U., J. Biol. Chem., 240,
4099 (1965).

Friedman, A. H., and Leete, E., J. Amer. Chem. Soc., 82.
2141 (1963).

Friedman, A. H., and Leete, E., J. Amer. Chem. Soc., 88,
1224 (1964).

Essery, J., Juby, P. F., Marion, L., and Trumbull, E.,
J. Amer. Chem. Soc.,‘8fl, 4597 (1962).

Juby, P. F., and Marion, L., Can. J. Chem., 3;, 117 (1963).
Waller, G. H., and Yang, K. C., Fed. Proc., 88, 356 (1963).

Henderson, L. M., Someroski, J. F., Rao, D. H., Wu, P.-H.
L., Griffith, T., and Byerrum, R. U., J. Biol. Chem.

:3.
,7.

38.

40.

41.

42.

43.

44.

45.
46.
47.

48.

49.

50-

51.

52.

53-

54.

55.

47

Ortega, M. V., and Brown, G. M., J. Biol. Chem., 285,
2939 (1960).

Albertson, J. N., Jr., and Moat, A. G., J. Bacteriol.,
£2. 540 (1965).

Isquith, A. J., and Moat, A. G., Biochem. BiOphys. Res.
Commun., g, 565 (1966).

Scott, T. A., and Mattey, M., Biochem. J., 107, 606
(1968).

Scott, T. A., Bellion, E., and Mattey, M., Biochem. J.,
102, 23P (1968).

Brown, S. A., and Byerrum, R. U., J. Amer. Chem. Soc.,

11. 1523 (1952).

Dawson, B. F., Christman, D. H., D'Adamo, A., Solt, M. L.,
and Wolf, A. P., J. Amer. Chem. Soc., 88, 2628 (1960).

0'Nea1,623 M., and Koeppe, R. E., J. Neurochem., 18, 835
(19 .

Smith, C. H., Ind. Eng. Chem., 4, 251 (1942).

Fleeker, J., Personal communication.

Lazier, W. A., and Arnold, H. H., in Blatt, A. H., (Editor),
Organic Synthesis, Collective Vol. II, John Wiley and
Sons, Inc., New York, 1943, p. 142.

Scott, T. A., Biochem. J., 102, 87 (1967).

Bowman, R. E., and Stroud, H. H., J. Chem. Soc., 1342
(1950).

Burns, L. C., O'Neal, R. M., and Koeppe, R. E., J. Amer.
Chem. Soc., 82, 3938 (1967).

Sekizawa, Y., Maragoudakis, M. E., King, T. E., and
Cheldelin, V. H., Biochemistry, 2, 2392 (1966).

Payes, B., and Laties, G. G., Biochem. Biophys. Res.
Commun., 12. 179 (1963).

Payes, B., and Laties, G. G., Biochem. Biophys. Res.
Commun., 88, 460 (1963).

Hess, J. L., and Tolbert, N. E., J. Biol. Chem., 241, 5705
(1966).

Bassham, J. A. in Bonner, J., and Varner, J. E. (Editors),
Plant Biochemistry, Academic Press, New York and
London, 1965, p. 897.

56.

57.

58.
59-

60.

48

Byerrum, R. U., Hamill, R. L., and Ball, C. D., J. Biol.
Chem., 210, 645 (1954).

Stepka W., Benson, A. A., and Calvin, M., Science, 108,
304 (i948). —-

Griffith, T., and Byerrum, B. U., Science, 122, 1485 (1959).

Yankwich, P. E., Norris, T. H., and Huston, J., Anal. Chem.,

Van Slyke, D. D., Steele, H., and Plazin, J., J. Biol.
Chem., 122, 769 (1951).

APPENDIX

The formula used in correcting the observed BaCO3

counts to infinite thickness was (59, 60):

= §§—’ where Nk_= fi__E§£Q_
k th std
A = dpm/mg BaCO3
NS = cpm of sample minus background
F = Absorption factor for BaCO3 sample
Nk = Efficiency factor for counter

Fstd = Absorption factor for BaCO3 standard
Nth = dpm/mg BaC03 (Calibrated standard)

Nstd = cpm of standard BaCO3 minus background

49

   

      

IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII

ummgummmmmunmmI1Wham

1293 03169