DOING RESEARCH IN THE INTRAMURAL PROGRAM
OF THE NATIONAL INSTITUTES OF HEALTH

JULIUS AXELROD*

Introduction

In trying to describe the contributions of intramural research at the
National Institutes of Health (NIH) to international collaboration, I
thought it appropriate to show the role of many foreign visitors in our
research on catecholamine neurotransmitters. This personal tale illus-
trating how research is done at NIH doubtless could be repeated in
different forms by many others who also have benefited from the splen-
did cooperative arrangements at NIH.

I joined NIH in 1950, just as the Intramural Program of the National
Heart Institute was started under the direction of James Shannon. While
working in the National Heart Institute, I obtained a Ph.D. at George
Washington University. After getting a Ph.D. in 1955 at the age of 42, I
was invited by the National Institute of Mental Health (NIMH) to set up
a section on pharmacology. I accepted with anticipation and some ap-
prehension. My impression of neuroscience was that it was mainly con-
cerned with electrophysiology and brain anatomy and involved the use
of complicated electronic equipment. I believed that an investigator in
neuroscience had to be an especially gifted experimentalist and theorist.
Ed Evarts, who was then head of the laboratory in which my section
belonged, assured me that it was not necessary to work on the brain or
mental illness; I could study anything I pleased.

I began a research problem with which I felt comfortable and that
might be related to the mission of the NIMH. Several years earlier, I had
worked on the metabolism of drugs. At the NIMH my initial problem
was concerned with the metabolism of LSD and narcotic drugs [1, 2].
One of the advantages of doing research in Bethesda is the variety of

*Section on Pharmacology, Laboratory of Cell Biology, National Institute of Mental
Health, Bethesda, Maryland 20205.

© 1986 by The University of Chicago. All rights reserved.
003 1-5982/86/2932/$0 1.00

Perspectives in Biology and Medicine, 29, 3, Part 2 - Spring 1986 | S131
physical and intellectual resources available. Bob Bowman of the Na-
tional Heart Institute was in the process of building a new kind of spec-
trophotometer. He was kind enough to let me use one of his experimen-
tal models, which enabled me to develop an exquisitely sensitive assay
for LSD and to study its metabolism and distribution in the brain and
other tissues. I also turned my attention to investigating the metabolism
of narcotic drugs. An enzyme in liver microsomes was found that N-
demethylated narcotic drugs [2]. This enzyme was profoundly affected
during the development of tolerance to narcotic drugs [3]. On the basis
of these findings I proposed a theory of tolerance that stimulated a great
deal of critical reaction, mostly negative. This theory predicted the pres-
ence of opiate receptors and their down-regulation after continous in-
teraction with narcotic drugs.

- The Fate of Noradrenaline at the Nerve Terminal

I felt somewhat guilty about not working directly with the nervous
system, although the administrators of the NIMH were supportive of
the research I was doing. An Opportunity to begin investigation on a
problem more relevant to the mission of the NIMH occurred in an
unexpected way. In 1956, Seymour Kety, who was then the head of my
laboratory, reported at a seminar on a preliminary observation made by
two Canadian psychiatrists, A. Hoffer and H. Osmond, that struck my
interest. They found that when adrenaline was left exposed to the air it
was converted to adrenochrome. When adrenochrome was ingested it
produced hallucinations. Osmond and Hoffer then proposed that
schizophrenia might be produced by an abnormal metabolism of adren-
aline to adrenochrome. In view of these observations, I thought it would
be a good idea to study the metabolism of adrenaline. In 1956, it
was believed that adrenaline and other catecholamines were metab-
olized and inactivated by monoamine oxidase. I spent 4 frustrating
months trying to find an enzyme that converts adrenaline to adreno.
chrome.

In March 1957, I received the Federation Proceedings, and I noticed an
abstract by Armstrong and McMillan [4]. They reported that in patients
with adrenaline-forming tumors (pheochromocytoma), large amounts
of 3-methoxy-4-hydroxymandelic acid were excreted in the urine. From
the structure of this compound it appeared to me that it might be a
metabolic product of either adrenaline or noradrenaline by deamination
and O-methylation. The possibility that catecholamines could be O-
methylated was intriguing and could be examined experimentally.

The most likely methyl donor for the methylation of a hydroxy group
was S-adenosylmethionine. I then incubated adrenaline with a rat liver
homogenate, ATP, and methionine. The latter compounds should be

$132 | Julius Axelrod « Intramural Program of the NIH
converted to S-adenosylmethionine in the liver. I was delighted to ob-
serve that adrenaline was rapidly metabolized under these conditions.
When either ATP or methionine was omitted, there was little metabo-
lism of the catecholamine. Similar results were obtained when S-
adenosylmethionine was used. These were crucial experiments and
demonstrated a new pathway for the metabolism of catecholamines by
O-methylation. The most likely site of methylation would be on the
hydroxyl group in a position meta to the side chain, to form O-methy]
adrenaline. I called Bernard Witkop, a bioorganic chemist at NIH, and
asked if he or his associates could synthesize the O-methylated metabo-
lite of adrenaline. Three days later, Sino Senoh, a visiting scientist from
Osaka, synthesized meta O-methyl adrenaline, which we named
metanephrine. This is an example of the exceptional resources available
at NIH. Metanephrine was identified after incubating liver with S-
adenosylmethionine and adrenaline. This compound was also found to
be present in rat urine, and amounts increased after adrenaline injec-
tion. The responsible enzyme was isolated and purified and shown to O-
methylate other catecholamines, such as noradrenaline, dopamine, and
dopa [5]. Because of these properties, this enzyme was named catechol-
O-methytransferase (COMT). O-Methylated catecholamines _ nor-
metanephrine and 3-methoxytyramine were soon found to be present in
the brain [6]. I felt that I was a neuroscientist at last. As a result of the
discovery of COMT, the pathway for the metabolism of catecholamines
was described as involving monoamine oxidase and COMT.

It was believed in 1957 that neurotransmitters were inactivated by the
enzyme monoamine oxidase. After inhibiting both monoamine oxidase
and COMT, however, the physiological actions of noradrenaline rapidly
disappeared after the injection of noradrenaline. If it was not enzymes,
what was the mechanism for terminating the actions of catecholamines?
The answer to this important question came with the help of two visiting
scientists to my laboratory, Gordon Whitby from Cambridge University
and Georg Hertting from the University of Vienna.

When Whitby arrived in Bethesda in 1959, I suggested that he exam-
ine the tissue distribution of [>H]noradrenaline in cats. Dr. Kety commis-
sioned New England Nuclear Corporation to synthesize [?H]nor-
adrenaline so that he could examine the metabolism of this amine in
schizophrenic and normal subjects. He gave me a few microcuries of
{*H]noradrenaline for the animal studies. Parenthetically, I note that
Kety and co-workers found no difference in the metabolism of (?H]nor-
adrenaline between schizophrenics and normal subjects. We observed
‘that after the injection of [*H]noradrenaline in cats, the catecholamine
persisted unchanged in tissues rich in the sympathetic nerves, heart,
spleen, salivary gland, and adrenal medulla long after its physiological
effects had dissipated [7]. It became apparent that [*H]noradrenaline

Perspectives in Biology and Medicine, 29, 3, Part 2 « Spring 1986 | $133
was selectively retained in sympathetic nerves and glands that concen-
trate catecholamines.

George Hertting, who came as a visiting scientist while these experi-
ments were going on, suggested an elegant approach to establish that the
site of uptake of catecholamine neurotransmitters was the sympathetic
nerves. First, the superior cervical ganglia of a cat was removed unilat-
erally, causing the sympathetic nerves in the salivary gland and eye
muscles to degenerate on one side only. A week later, [*H]noradrenaline
was injected and its concentration determined in structures innervated
by nerves arising from the superior cervical ganglia. There was little or
no uptake of (*H]noradrenaline in denervated structures as compared
to innervated tissue [8]. This experiment provided strong evidence for a
new mechanism for the inactivation of a neurotransmitter—reuptake
into sympathetic nerve terminals. We were then in a position to label the
sympathetic nerve selectively by injecting [*H]noradrenaline and then to
examine the physiological disposition and metabolism of this neuro-
transmitter during and after nerve stimulation. In a series of experi-
ments, Hertting [9], Sune Rosell [10], a visiting scientist from the
Karolinska Institute, and I showed that noradrenaline is taken up in the
sympathetic nerves and released on stimulation. In another experiment
in collaboration with Keith Richardson, a British anatomist who was
spending some time at NIH, we directly demonstrated by radioauto-
graphic studies and electron microscopy that [*H]noradrenaline is taken
up and stored in vesicles of sympathetic nerves [11]. It was with the
collaboration of a number of visiting scientists that we were able to
describe a model for the fate of the neurotransmitter noradrenaline in
sympathetic nerve terminals [12] that has stood the test of time. This
model made it possible to explain the actions of psychoactive drugs such
as amphetamine, cocaine, and antidepressants.

With the observation that antidepressant drugs block the reuptake of
noradrenaline in peripheral sympathetic nerves, it appeared that this
process could explain the mechanism of action of drugs that relieve
mental depression. It was a visiting postdoctoral fellow, Jacques Glo-
winski from Paris, who made it possible to examine this problem.
Glowinski devised a technique to introduce [*H]noradrenaline into the
brain by injecting it into the lateral ventricle. In collaboration with a
visiting scientist from England, Hans Weil-Malherbe, we had previously
shown that (®H]noradrenaline does not cross the blood-brain barrier. To
study the action of antidepressant drugs on catecholamines, it was neces-
sary to inject (*H]noradrenaline directly into the brain. We found that
radioactive noradrenaline injected into the lateral ventricle was distrib-
uted in different areas of the brain in about the same proportions as
were the endogenous amines. After labeling the noradrenergic
neurones with [*H]noradrenaline, Glowinski and I found that pretreat-

S134 | Judius Axelrod - Intramural Program of the NIH
ment of rats with clinically effective antidepressant tricyclic drugs
blocked the neuronal uptake of [?H]noradrenaline {13}. Tricyclic com-
pounds related in structure to antidepressant drugs but with little or no
clinical therapeutic activity did not block the reuptake of [?H]nor-
adrenaline. This, together with the observation that monoamine oxidase
inhibitors have antidepressant actions and that reserpine, a depletor of
noradrenaline and other amines, sometimes causes depression, led to
the formulation of the catecholamine hypothesis of depression [14]. Al-
though this hypothesis has not entirely held up, it has provided a pro-
ductive framework for new experimental approaches to the study of
depression.

Regulation of Catecholamines

Catecholamine neurotransmitters in sympathetic nerves are in a state
of flux, continually being synthesized, released, metabolized, and recap-
tured. In spite of these dynamic changes, the concentration of catechol-
amines in tissues remains at a constant level. This is due to a variety of
adaptive mechanisms that change the biosynthesis, release, and response
of catecholamines. One such mechanism was discovered with the help of
Hans Thoenen, a visiting scientist from Basel. Thoenen asked me if he
could spend a sabbatical year in my laboratory. Before that, Thoenen
and his co-workers had found that the injection of 6-hydroxydopamine
destroyed sympathetic nerve terminals. I wrote to Thoenen that we
would be happy to have him join our laboratory, especially if he brought
some 6-hydroxydopamine with him. ,

When Thoenen arrived in Bethesda, the first experiment we did was
to study the localization of the catecholamine biosynthetic enzyme, ty-
rosine hydroxylase. In this experiment we exploited the ability of 6-
hydroxydopamine to destroy sympathetic nerve terminals selectively.
‘Tyrosine hydroxylase disappeared after the destruction of sympathetic
nerves with 6-hydroxydopamine, indicating the selective localization of
tyrosine hydroxylase in noradrenergic nerves [15]. An unexpected
finding was a marked increase in tyrosine hydroxylase in the adrenal
medulla after treatment with 6-hydroxydopamine [15]. It was known
that 6-hydroxydopamine also causes an increased firing of sympathetic
nerves, and we speculated that the increased tyrosine hydroxylase was
due to increased sympathetic nerve activity. We then injected a number
of drugs that caused increased sympathetic nerve activity; these drugs
also increased tyrosine hydroxylase activity in the adrenal medulla and
sympathetic ganglia. When nerves to the adrenal gland were cut, none
of the drugs that caused heightened sympathetic nerve activity could
elevate tyrosine hydroxylase, indicating that induction of this enzyme
was a transsynaptic event [16]. Subsequent experiments showed that

Perspectives in Biology and Medicine, 29, 3, Part 2 - Spring 1986 | $135
increased nerve activity induced the synthesis of new tyrosine hydroxy-
lase molecules. Similar results were obtained with another biosynthetic
catecholamine enzyme, dopamine-B-hydroxylase.

The National Institutes of Health have also served as a productive
training ground for young Ph.D.’s and M.D.’s who wish to pursue ca-
reers in biomedical research. I found that these postdocs create an open
atmosphere in the lab, and free exchange of ideas has made it possible to
try new approaches to the solution of problems. An example of the value
of the interaction with postdoctoral fellows is the experimental demon-
stration in which the relationship between the adrenal cortex and
medulla was demonstrated. [t has been known that the ratio of adrena-
line and noradrenaline in the adrenal medulla is dependent on the
extent to which the adrenal cortex surrounds the medulla. In those
species in which the adrenal cortex is in close juxtaposition to the
medulla, the main catecholamine is the methylated catecholamine,
adrenaline, while in species where the adrenal cortex is separated from
the medulla, the principle catecholamine is noradrenaline. {n an ex-
change of ideas about the role of the cortex in regulating adrenaline
formation, Richard Wurtman, a postdoctoral fellow in my laboratory,
proposed an experiment to determine how the adrenal cortex could
regulate adrenaline formation. He removed the pituitary from a rat and
then measured the adrenaline-forming enzyme, phenylethanolamine-
N-methyltransferase (PNMT). The pituitary secretes ACTH, which
then stimulates the adrenal cortex to synthesize glucocorticoids. The
ablation of the pituitary gland resulted in a profound decrease in PNMT
{17}. The administration of either glucocorticoids or ACTH to these rats
restored enzyme activity. This experiment clearly demonstrated that
glucocorticoids from the adrenal cortex regulate the adrenaline-forming
enzyme.

These are a few examples of the value of the visiting scientist and
postdoctoral program in promoting high-quality research in the In-
tramural Research Program at NIH.,More than 60 postdoctoral fellows
and visiting scientists have spent time in my laboratory, and more than
50 percent of these postdocs were found overseas. With one or two
exceptions, they all went on to highly productive careers in research.
They all told me that the years that they spent in Bethesda had an
important influence on their subsequent careers in biomedical research.

REFERENCES

1. AxELRop, J.; Bropy, R. O.; Witxop, B.; and Evarts, E. V. Metabolism of
lysergic acid diethylamide. Nature 178:143—144, 1956.

2. AXELROD, J. The enzymatic N-demethylation of narcotic drugs. J. Pharmacol.
Exp. Ther. 117:322-330, 1956.

S136 | Julius Axelrod - Intramural Program of the NIH
11.

12.

- AXELROD, J. Possible mechanism of tolerance to narcotic drugs. Science

124:263-264, 1956.
ARMSTRONG, M. D., and McMiuuan, A. Identification of a major urinary
metabolite of norepinephrine. Fed. Proc. 16:146, 1952.

. AXELROD, J., and Tomcnick, R. Enzymatic O-methylation of epinephrine

and other catechols. J. Biol. Chem. 233:702—705, 1958.

. AXELROD, J.; SENOH, S.; and Witxop, B. O-methylation of catechol amines in

vivo. J]. Biol. Chem. 233:697—701, 1958.
Wuirtsy, L. G.; AXxELRop, jJ.; and Weit-Macuerse, H. The fate of 3H-
norepinephrine in animals. J. Pharmacol. Exp. Ther. 132:193-201, 1961.

. HERTTING, G.; AXELROD, J.; Kopi, I. J. and Wurrsy, L. G. Lack of uptake

of catecholamines after chronic denervation of sympathetic nerves. Nature
189:66, 1961.

HERTTING, G., and AxELRop, J. The fate of tritiated noradrenaline at the
sympathetic nerve-endings. Nature 192:172-173, 1961.

- RosELt, S.; Kopin, I. J.; and AxeLrop, J. Fate of 3H-noradrenaline in skele-

tal muscle before and following sympathetic stimulation. Am. J. Physiol.
205:317-321, 1963.

Wo re, D. E.; Potrer, L. T.; RicHARDSON, K. C.: and AXELROD, J. Lo-
calizing tritiated norepinephrine in sympathetic axons by electron micros-
copy autoradiography. Science 138:440—442, 1969.

AXELROD, J. Noradrenaline: fate and control of its biosynthesis. Science
173:598-606, 1971.

- GLowinski, J., and AxeLrop, J. Inhibition of uptake of tritiated-

noradrenaline in the intact rat brain by imipramine and structurally related
compounds. Nature 204:1318-1319, 1964,

. SCHILDKRAUT, j. J. and Kery, S. S. Biogenic amines and emotion. Science

156:21-30, 1967.

. Mue ter, R. A.; THOENEN, H.; and AXELROD, J. Adrenal tyrosine hydroxy-

lase: compensatory increase in activity after chemical sympathectomy. Sci-
ence 163:468-469, 1969.

. THoenen, H.; MUELLER, R. A.; and AXELROD, J. Transsynaptic induction of

adrenal tyrosine hydroxylase. J. Pharmacol. Exp. Ther. 169:249-254, 1967.

- WurtTMan, R. J., and AXELROD, J. Control of enzymatic synthesis of adrena-

line in the adrenal medulla by adrenal cortical steroids. J. Biol. Chem.
241:2301-2305, 1966. .

Perspectives in Biology and Medicine, 29, 3, Part 2 - Spring 1986 | $137