Kidney Iniernational, Vol. JO (1976), p.
12—24
Hypothalamic neurons secreting vasopressin and neurophysin
EARL A. ZIMMERMAN AND ALAN 0. ROBINSON
Department of Neurology, College of Physicians and Surgeons, Columbia University, New York, New York, and Department of
Medicine, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania
Vasopressin is synthesized in the magnocellular
system of the hypothalamus in clusters of cells which
form the supraoptic nucleus and the paraventricular
nucleus. The hormone is synthesized and packaged in
neurosecretory granules with an intragranular pro-
tein, neurophysin. The demonstration of axon flow of
neurosecretory granules from the perikarya in the
hypothalamus to the posterior lobe of the pituitary
and subsequent release into the blood has been an
important historical chapter in our understanding of
neurosecretion. Isolation of neurophysins from sev-
eral species and development of antisera to these
peptides as well as antibodies to vasopressin have
provided new tools to re-examine this system. In
several species, the data indicate a specific neurophy-
sin for vasopressin and a different neurophysin for
oxytocin. It is now well established that neurophysins
are secreted with hormones and this has provided a
cogent argument that exocytosis is a major form of
neurosecretion of vasopressin. Assay of neurophysin
in plasma can be used to study vasopressin release.
Immunohistochemical studies using antibodies to
vasopressin and neurophysin demonstrated that the
magnocellular system is more diffusely distributed
throughout the hypothalamus than was previously
appreciated. In addition, vasopressin and neurophy-
sin are formed in both the supraoptic and para-
ventricular neurons and there are three pathways of
secretion. The major pathway is the supraoptico-hy-
pophyseal tract to the posterior lobe. The second
pathway is to the external zone of the median emi-
nence for secretion into the hypophyseal portal
blood. The third pathway is to the third ventricle for
secretion into cerebral spinal fluid. Vasopressin in the
external zone is greatly increased by the absence of
adrenal cortical steroids which suggests that vaso-
pressin may play a role in the hypothalamic anterior
© 1976, by the International Society of Nephrology.
12
pituitary adrenal axis. The cerebral spinal fluid path-
way may be important in man if vasopressin is found
to have the memory consolidating effects which have
been investigated in other animals.
Historical background. Extracts of the posterior
pituitary were first identified to have biologic activity
in the last half of the nineteenth century when va-
sodepressor as well as vasopressor activity was de-
scribed [I]. By 1913 the antidiuretic effect of vaso-
pressin was established and its therapeutic benefit in
patients with central diabetes insipidus was demon-
strated [2,3]. In spite of early studies which suggested
that the posterior pituitary did not have the appear-
ance of a gland which was actively synthesizing hor-
mones and in spite of the known neural connection
with the hypothalamus [4], it was not appreciated
that the neurohypophysis was part of a hypothalamic
unit until the classic studies of Bargmann and Schar-
rer [5] and Scharrer and Scharrer [6]. These inves-
tigators applied the chromealum-hematoxylin stain
to the hypothalamus, a technique originally described
by Gomori to stain the beta cells of the pancreas [7].
This dye and aldehyde fuchsin have a special affinity
for the sulfhydryl groups of the Van Dyke protein
and allowed these workers to trace the posterior pi-
tuitary neurosecretory material along axons in the
pituitary stalk to cell bodies in the supraoptic nucleus
(SON) and paraventricular nucleus (PVN) of the
hypothalamus [6,8]. Furthermore, after transection
of the pituitary stalk, Gomori-positive material ac-
cumulated above the transection and was lost from
the posterior pituitary [5]. This clearly established
that the posterior pituitary hormones were synthe-
sized in the hypothalamic nuclei and transported by
axonal flow to the posterior pituitary. The posterior
pituitary, which was previously considered the site of
synthesis of the hormones, was shown to be a locus
of storage and release. It is also of interest that Barg-
mann and Scharrer [5] described and diagrammed a
close connection between the supraoptico-hypophy-
 Secretion of vasopressin and neurophysin
sial tract, i.e. the tracts to the posterior pituitary, and
the long portal vessels which drained to the adenohy-
pophysis. Studies with the Gomori stain led them to
conclude that axons from this tract projected to the
zona externa of the median eminence. In addition,
they mentioned projections from the tract to the third
ventricle. Although it became generally accepted that
the neurohypophysis was a hypothalamic gland
and that the median eminence played a key neurose-
cretory role in the regulation of the adenohypophysis,
the possible interrelationship between these two or-
gans was not emphasized. As will be described below,
newer techniques have confirmed the suggestion that
neurohypophysial peptides are secreted into the cere-
brospinal fluid (CSF) of the third ventricle and into
the long portal vessels of the adenohypophysis.
At about the time that the Gomori-positive path-
ways to the neurohypophysis were being char-
acterized by anatomists, biochemists, pharmacol-
ogists, and physiologists were identifying the active
principles of the gland. Bioassay data from posterior
pituitary extracts had shown a variety of actions:
vasopressor, antidiuretic, galactogenic and utero-
tonic. The studies of Acher, Chauvet and Olivry [9]
demonstrated that mild treatment (changes in pH
and electrodialysis) could reversibly separate two bi-
ologically active hormones, vasopressin and oxyto-
cm, from a carrier protein termed neurophysin [9].
The loosely combined neurophysin and the peptide
hormones formed the Van Dyke protein. Amino acid
studies of vasopressin by du Vigneaud proved that
the vasopressor and antidiuretic actions of the poste-
rior pituitary existed in a peptide which was separate
from the peptide with oxytocic activity [10,11]. As
bioassays were developed it was shown that oxytocin
and vasopressin were concentrated in extracts of the
SON and the PVN confirming the presence of hor-
mone peptides throughout the system which had
been described using Gomori stains [12]. The identi-
fication of two individual hormones led to studies of
the sites of synthesis and the concept emerged that
the supraoptic nucleus mainly formed vasopressin
and the paraventricular nucleus, oxytocin. These re-
sults of bioassay studies of extracts of the nuclei of
the camel [13] and sheep [14,15] were supported by
ablation studies in other animals which suggested
that diabetes insipidus was produced by lesions in the
supraoptic nucleus and not by destruction of the
paraventricular nucleuS [16]. More recent data ob-
tained by innunological methods described below
demonstrate that this interpretation was an over-
simplication. Both hormones are found in both nu-
clei, but axons of the PVN travel through the SON
enroute to the posterior pituitary and, therefore, le-
13
sions in the SON can destroy most of the supraoptic
and paraventricular system.
In recent years, availability of antisera to the neu-
rophysins and vasopressin and their application in
immunocytochemical techniques and radioimmuno-
assays marked a turning point in the long history
of fruitful research of the hypothalamic pathways
secreting vasopressin. The peptides have been quan-
titated in specific brain regions, localized in the
exact cells of formation [17], and measured in blood
[18].
Immunohistology of hypothalamic cells. The first
localization studies of neurohypophyseal peptides
were reported by Livett et a! in 1970 using immu-
nofluorescence of neurophysin [19,20]. It was soon
learned that antiserum to the neurophysins from one
species cross-reacted with the neurophysins of other
species. Anti-ovine, -porcine and -bovine and -human
neurophysins were employed in immunofluorescent
and immunoperoxidase localization techniques to
study the hypothalamus of a wide variety of mam-
mals from mouse to man [2 1-24]. These studies con-
firmed that neurophysins are a major component of
Gomori-positive material in the hypothalamus. They
directly demonstrated that neurophysins are present
throughout the magnocellular neuron from cell body
of synthesis (Fig. 1), along its axon of transport
and in its terminal in the posterior pituitary gland
(Fig. 2). The immunocytochemical procedures proved
more sensitive than the Gomori stains in demonstrat-
ing neurosecretory material. While the major sources
of neurophysin and vasopressin and oxytocin are in
the SON and PVN, many magnocellular neurons
outside the strict confines of the SON and the PVN
were found to contain neurophysin. It was shown
Fig. 1. Supraoptic neuron in a human hypothalamus stained for
neurophysin using antiserum to human estrogen-stimulated neu-
rophysin in an immunoperoxidase technique. Neurosecretory neu-
rons in this region tend to be bipolar, while those in the para-
ventricular area often have more than two processes (see Fig. 2).
Reaction products to neurophysin are distributed throughout the
cytoplasma of the cell but not in its nucleus (unstained oval area in
the cell body). From Zimmerman EA: The distribution of the
neurophysin-vasopressin system in the mammalian hypothalamus.
Israel J Med Sci. in press
14
Fig. 2. Diagram of the mammalian hypothalamus and pi-
tuitary gland, sagiltal view, depicting pathways secreting
vasopressin (VP) and neurophysin (NP). The hormone and
carrier protein are formed in magnocellular perikarya
in the supraoptic and paraventricular nuclei, transported
in granules along their axons and secreted at three sites:
(a) posterior pituitary gland (systemic circulation), (b)
zona externa of the median eminence (hypophysial portal
circulation), (c) third ventricle (cerebrospinal fluid). Parvi-
cellular neurons in the suprachiasmatic nucleus (SCN) in
rat and mouse also contain VP and NP.
earlier by Gomori stains that accessory cells between
the SON and the PVN contained neurosecretory ma-
terial [25], but the immunocytochemical studies re-
vealed that the fields of the SON and the PVN over-
lapped to a greater extent than previously appreci-
ated and that cells belonging to the neurophysin sys-
tem were even scattered beyond the anterior hypo-
thalamus [17]. Some cells were found anteriorly in
the preoptic area as far as the lamina terminalis,
dorsally near the ventral-anterior thalamus, and cau-
daIly on the lateral tuberal floor at the level of the
anterior median eminence. Application of antisera to
vasopressin and oxytocin has more recently revealed
that the hormones are also located in neurophysin-
positive cells which are distributed throughout this
larger system. It, therefore, is no longer possible to
think of vasopressin being exclusively formed in a
particular nucleus such as the SON [17].
Vasopressin, oxytocin, and the neurophysins have
been found all along the beaded axonal pathways in
the hypothalamus, in the main tract of the zona in-
terna of the median eminence, and along the pituitary
stalk to the storage and secretory terminals in the
posterior pituitary [17] (Fig. 2). Immunoelectron mi-
croscopic studies have shown that vasopressin and
neurophysin are contained in the large electron-dense
granules all along the fiber [26-28]. Although these
ultrastructural studies have not adequately demon-
strated the hormones in the granules in the perikarya,
neurophysin has been localized in these granules [27,
28]. Inability to find hormone in these granules prob-
ably does not represent the absence of vasopressin in
Zimmerman and Robertson
Paraventricular
neurons
Th Pineal
VP, NP Third ventricle
Supraoptic
neuron —
S.C.N.
Tan ycyte
Optic chiasm
Sup. hypophysial A Mamrnillary body
Portal capillaries
in zona externa of
median eminence system
Short VP, NP
Anterior' portal
pituitary vein
the cell bodies and subsequent elaboration of hor-
mone from a precursor along the axons as suggested
by some authors [26]. The hormones have been ade-
quately demonstrated in the perikarya by light micro-
scopic immunohistochemical methods and they are
likely to be located in granules. It is likely that the
inability to recognize hormones in the perikarya with
electron microscopy (EM) is due to problems in tis-
sue processing for immunoelectron microscopy. With
light microscopy, a perinuclear concentration of hor-
mones and neurophysins is commonly seen, probably
representing their accumulation in granules formed
in the Golgi apparatus in this region. The hormones
and neurophysins are also found in the dendrites
which are particularly elaborate in the PVN. A func-
tion for peptides in these processes is unknown, but it
may simply reflect circulating of granules throughout
the neurons after formation in the Golgi apparatus.
The dendrites may serve as a storage area for the
hormones in the perikarya.
Axonal pathways. The immunohistochemical stud-
ies have confirmed previous anatomical observations
concerning the fiber pathways. Neurophysin- and
vasopressin-containing granules are concentrated in
the beaded segments of the fibers, fill the large dilata-
tions (Herring bodies) and are concentrated at the
axon terminals in the posterior pituitary gland [25-
28]. Vasopressin terminals in posterior pituitary ap-
pear to have a general distribution while oxytocin
fibers may have a more restricted pattern particularly
in the more dorsal-lateral regions as seen in the rat
[29].
 Secretion of vasopressin and neurophysin
One of the most striking and initially unexpected
findings of the immunocytochemical work on neu-
rophysin and vasopressin was the presence of fiber
terminals around portal capillaries [21, 30-32]. The
source of the vasopressin-containing fibers to the por-
tal bed is presently uncertain. Vandesande, Dierickx
and DeMey [33] suggested that they may come from
the suprachiasmatic nucleus (SCN) in the rat. Fibers
projecting from this nucleus, however, are difficult to
trace. Vasopressin and neurophysin have not been
found in the SCN of the monkey (Antunes J, Carmel
PL, Zimmerman EA: unpublished data) or guinea pig
(Silverman AJ: personal communication) where vaso-
pressin is present in the magnocellular nuclei of the
hypothalamus and is plentiful in the zona externa. It
is, therefore, evident that the axons carrying vaso-
pressin to the portal bed are likely to come from cells
belonging to the SON or PVN or associated regions.
We have recently found evidence in monkeys that a
significant projection to the zona externa originates
in the ipsilateral PVN [34]. Lesions in one side of the
PVN resulted in depletion of the immunostaining of
vasopressin and its neurophysin in the zona externa
of the median eminence on the same side. Whether
these projections are direct, as shown in the diagram
(Fig. 2), or collaterals of fibers to the posterior pi-
tuitary is not certain.
There may be yet another pathway for vasopressin
which involves the cerebrospinal fluid (CSF) of the
ventricular system. The presence of vasopressin in
CSF has been known for some time, and experiments
by Vorheer et al suggested that the hormone was
secreted into CSF in the dog [35]. In 1973 we re-
ported neurophysin in the CSF of man and monkey
[36]. Since radio-labeled neurophysin administered
systemically did not pass into the CSF of dogs, direct
secretion into the CSF was indicated [36]. The source
of secretion of vasopressin and neurophysin into CSF
is not fully established. More recent transmission and
scanning electron microscopic studies in mammals
demonstrated a variety of free nerve endings in the
floor of the third ventricle, particularly at the level of
the median eminence and anteriorly near the or-
ganum vasculosum of the lamina terminalis [37].
Some of these endings contain large granules typical
of those containing vasopressin in the posterior pi-
tuitary gland. Others contain smaller granules which
may contain releasing factors or catecholamines. By
immunoelectron microscopy, neurophysin and vaso-
pressin have now been visualized inside large gran-
ules in axons which lie in the floor of the third ven-
tricle [38] and it now seems fairly certain that this is
at least one site of axonal secretion into the ventricu-
lar system. The secretory terminal to the third yen-
15
tricle in Fig. 2 is represented as a branch from a
paraventricular axon. Whether such a terminal is di-
rect or a collateral is not known, but the origin is
thought to be the SON or the PVN.
The function of the secretory pathway of vasopres-
sin and neurophysin into CSF is not fully under-
stood. The presence of immunoreactive neurophysin
found in tanycytes in some studies of rats [32] and
monkeys [36] suggested that these specialized epen-
dymal cells in the infundibular recess of the third
ventricle may take up neurophysin and possibly vaso-
pressin from CSF and transport it to portal blood.
This interpretation was based on the current concept
that the major function of tanycytes is to transport
rather than synthesize neuropeptides [39-41]. The an-
atomical configuration of tanycytes su.ggests a trans-
porting role from CSF to portal blood. The large cell
bodies have microvilli lining the ventricle and mul-
tiple branching processes, many of which end on
portal capillaries [39, 40]. The remarkable capacity of
these cells to transport exogenous materials such as
gonadotropin-releasing hormone from the third ven-
tricle to portal blood has been amply demonstrated
[41]. The presence of immunoreactive neurophysin
and gonadotropin-releasing hormone seen in some
preparations suggests that these cells may normally
transport such materials [32]. Whether tanycytes may
also transport materials in the opposite direction,
from portal blood into CSF, has not been estab-
lished. Convincing, consistent evidence for vasopres-
sin in tanycytes is presently lacking, and the role of
these cells in neuroendocrine function is obscure. Fu-
ture research concerning tanycytes, however, may es-
tablish that they are important neuroendocrine cells.
Therefore, there are at least three major pathways
for the secretion of neurophysin and vasopressin: into
the systemic circulation, via the posterior pituitary;
into the hypophysial portal blood, via the external
zone of the median eminence; and into CSF of the
third ventricle. While the latter two routes may be
important for less accepted roles of vasopressin, and
may be important after stalk section, the former is
quantitatively the most important site of release of
the vasopressin which controls water excretion.
Axon transport. Studies of axonal transport of
vasopressin and neurophysins from the SON and
PVN to the posterior lobe have been aided by the
cysteine content of neurophysins and vasopressin and
consequent ability to inject 35S-labeled cysteine into
the sites of synthesis and to follow the appearance of
radioactive material in the posterior lobe. Sachs et al
[42, 43] used these techniques to study the time
course of vasopressin synthesis. They found a one-
hour lag between the injection of 35S-cysteine and
16 Zimmerman and Robertson
appearance of radioactive vasopressin in isolated hy-
pothalamic tissue in vitro [42, 43]. Axonal transport
and turnover of neurophysins has recently been sum-
marized by Norstrom [44]. Norstrom studied rats
subjected to severe hemorrhage, a maneuver which
produces a high turnover of neurophysin. In both
bled and control animals, radioactive cysteine was
low in the posterior lobe for two hours after the
injection of labeled cysteine [44]. Based on these and
other studies, a lag of 1.5 hr was proposed before
newly synthesized neurosecretory material was avail-
able for transport to the posterior lobe. Transport
down the axons is thought to take place primarily by
rapid transport of the peptides in granules [44-46].
Most of the radioactivity injected into the nuclei of
synthesis appears in the posterior lobe after two to
three hours. In the rats, the length of the supraoptico-
hypophysial tract is approximately 4 mm, and by
subtracting the lag to make neurophysin ready for
transport, Norstrom calculated a transport rate of
190 mm/day. Similar results were obtained by Jones
and Pickering [45] and Burford and Pickering [46].
Flow of other substances in these axons may be
by relatively slow (I mm/day). Norstrom detected a
maximal net accumulation of radioactive neurophy-
sin in the posterior lobe 2 to 7.5 days after injection of
cysteine and interpreted this as a second peak con-
sistent with slow transport. Burford and Pickering,
however, concluded that there was no evidence for
a second peak and slow (non-granular) transport of
neurophysins [46].
Once neurophysin and vasopressin appear in the
posterior lobe, there may be heterogeneity of storage.
Douglas has summarized the evidence that supports
the concept of "last in, first out" regarding neurose-
cretory granules in the posterior lobe [47]. He con-
cludes that this is due to anatomic location of the
newly arrived granules. The newly arrived neurose-
cretory granules are closer to the membrane surface
and, therefore, are more readily available for release
with stimulated transport. Some neurophysin and
vasopressin in the posterior lobe appear to have a
slow turnover. Sachs [48] found that radioactive
vasopressin could be released from the posterior
lobe three weeks after interventricular injection of
35S-labeled cysteine, and Burford and Pickering esti-
mated half-lives of neurophysin of approximately 20
days [46]. This has been interpreted as evidence for
the movement of radioactive labeled neurophysin
and vasopressin into a less readily releasible pool and
has been quoted by others as evidence for the arrival
to the posterior lobe of slowly transported neurophy-
sin over a period of days.
With stimulation of release of posterior lobe hor-
mones (e.g. after hemorrhage and during suckling) an
increased turnover of labeled neurosecretory material
from the posterior lobe can be demonstrated, and, in
spite of an absolute decrease of neurophysin content
in the posterior lobe, there continues to be an en-
hanced turnover. Interestingly, however, studies of
transport under these conditions have not demon-
strated an increase in transport time from the nuclei
to the posterior lobe [43-44]. The lag of incorporation
of injected radioactive cysteine into protein and sub-
sequent transport to the posterior lobe in these prepa-
rations is unchanged [44]. These data would indicate
that a greater bulk of neurophysin must be trans-
ported in the axons if the rate of transport is fixed. If
this is true, one might anticipate that increased neu-
rosecretory material might be demonstrated in the
axons by specific immunohistochemical stains of
the hypothalamus. This has been observed. Recent
studies in our laboratories [49, 50] demonstrated an
increased circulating level of neurophysin and va-
sopressin content of the posterior lobe. These are
characteristic findings in increased secretion and turn-
over. In the hypothalamus, an increase of stainable
material in the axon tracts was found. This histo-
logic data is in line with the transport data of Norst-
rom [44].
Whether there exists a nongranular form of neu-
rophysin or vasopressin in the posterior pituitary is
presently being debated. Norstrom found evidence of
free neurophysin in preparations of neurosecretory
granules from the posterior lobe and found that this
free radioactivity increased with the time interval af-
ter labeling [44]. There is also some anatomical evi-
dence to support the idea that the hormones and
neurophysin may exist outside of granules. In a re-
cent immunoelectron microscopic study, extra-
granular neurophysin (but not vasopressin) was
found in axon terminals in the posterior pituitary
gland [28]. Yet in axon terminals in zona externa
where neurophysin was plentiful, extragranular reac-
tion products to neurophysin were confined to gran-
ules. This represented an internal control and it was
considered that the extragranular reaction products
in the posterior lobe were thus real and not due to
diffusion artefacts. The large amounts of oxytocin
and its neurophysin found by light microscopic im-
munocytochemistry in the magnocellular neurons of
homozygous Brattleboro rats with diabetes insipidus
suggested that there may be nongranular transport of
these peptides [51] because these animals are known
to have very few axonal granules [52]. They do ac-
tively secrete oxytocin [53]. An extragranular secre-
tory mechanism, therefore, is not ruled out in either
normal or abnormal animals.
 Secretion of vasopressin and neurophysin
Neurophysin secretion. After Acher, Chauvet and
Olivry [9] had demonstrated in 1956 that the biologi-
cally active hormones could be separated from the
carrier protein neurophysins, there was less interest in
the neurophysins. Interest re-emerged ten years later
largely due to two observations: (a) the suggested
association of a specific neurophysin with oxytocin
and another neurophysin with vasopressin, and (b)
the secretion of neurophysin in peripheral blood. The
biochemical studies of Hollenberg and Hope [54] and
Ginsburg, Jayasena and Thomas [55] indicated that
there were two major neurophysins in the ox and pig.
In vitro evidence was also obtained which suggested
that some granules separated by centrifugation from
the neurohypophysis contained vasopressin and a
particular neurophysin and others, oxytocin and
another neurophysin [5 5-60]. Fawcett, Powell and
Sachs [61] utilized a dog model to study the synthesis
and secretion of vasopressin and neurophysin. By
injecting 35S-cysteine into the third ventricle and then
stimulating release of labeled peptides from the pos-
terior pituitary by hemorrhage, they obtained evi-
dence for the secretion of a larger molecular weight
peptide, neurophysin, as well as vasopressin into sys-
temic blood. The data were interpreted to represent a
simultaneous release of neurophysin with vasopres-
sin. The promise of detecting vasopressin release by
studying neurophysin secretion prompted some in-
vestigators to develop radioimmunoassays for neu-
rophysins.
In 1969, Legros, Franchimont and Hendrick first
demonstrated a neurophysin-like material in the
blood of pregnant women, using a radioimmunoas-
say based on bovine neurophysin [62]. This obser-
vation was soon followed by reports to Cheng and
Friesen [63] and Robinson et al [641 reporting
the measurement of plasma neurophysin in rat,
pig and ox. Specific radioimmunoassays for bovine
neurophysin I and bovine neurophysin II were de-
veloped, and it was found that two neurophysins
were independently secreted into plasma [64]. This
was consistent with the bioassay data suggesting
independent secretion of vasopressin and oxytocin.
Legros, Reynaert and Peeters have demonstrated a
clear-cut increase in bovine neurophysin I with no
change in bovine neurophysin II during milking [65].
Bovine neurophysin I went from values of 1.8 to 2.7
ng/ml plasma before milking to 2.2 to 6.1 ng/ml
after milking, while bovine neurophysin II values were
unchanged. Changes in bovine neurophysin were de-
scribed with hemorrhage, and there was a specific
increase in bovine neurophysin II in five animals
subjected to hemorrhage. Bovine neurophysin II
went from values of less than 1 ng/ml to peak values
17
greater than 30 to 50 ng/ml. Bovine neurophysin I
showed some slight increase in some animals, but all
values were less than 5 ng/ml [66]. Therefore, it
seems likely, at least in the cow, that there are two
major neurophysins and that one of these neurophy-
sins, bovine neurophysin I, is associated with oxyto-
cm and another neurophysin, bovine neurophysin II,
is associated with vasopressin. Furthermore, the data
demonstrate that these neurophysins are released at
the same times that the hormones are secreted, and
that the release of one neurophysin can be independ-
ent of the release of the other.
There are also data in the rat which would support
the concept of a specific neurophysin for vasopressin.
These dat are based on the study of neurophysins in
the horn ozygous Brattleboro rat with diabetes insip-
idus (DI rat) as compared to the normal rat. Using
analytical disc gel electrophoresis to identify neu-
rophysins, three bands of neurophysin peptides were
identified in the normal rat [67-69]. In the DI rat, one
of these peptides is lacking and it has been suggested
that this peptide is the normal rat neurophysin which
is associated with vasopressin. Immunocytochemical
studies of the DI rat have also demonstrated that the
missing neurophysin is absent in hypothalamic cells
which are deficient in vasopressin and that the neu-
rophysin present in these rats is present in oxytocin
cells [51].
Two neurophysins have been demonstrated in
the human and specific assays have been developed
[70]. Human posterior pituitaries were extracted in
acid and the crude neurophysins were separated by
salt precipitation and column chromatography. This
crude human neurophysin consisted of four peptide
bands by analytic disc gel electrophoresis. Rabbits
were injected with isolated human neurophysin and
antibodies were harvested. When these antibodies
were used to test cross-reactivity between the various
neurophysin bands, only two specific antigens were
recognized in the four peptide bands [70]. Therefore,
the immunologic behavior of the human neurophy-
sins is again consistent with two major neurophy-
sins. Having developed specific radioimmunoassays,
neurophysins were measured in unextracted human
plasma and the two neurophysins were found to be
independently secreted. One assay measured a neu-
rophysin which was secreted in response to estro-
gen administration, estrogen-stimulated neurophysin
(ESN). The other assay measured a neurophysin
which was secreted in response to cigarette smok-
ing, nicotine-stimulated neurophysin (NSN). Figure
3 demonstrates the plasma neurophysin responses of
five normal male subjects given 5 mg of diethylstil-
bestrol for two days, and Fig. 4 demonstrates the
18
zU)
U)
1 2 3 4 5tDESf
6 7 8 9 10
Days
Fig. 3. Changes in plasma ESN, top, and NSN. bottom, in five nor-
mal male subjects given 5 mg of DES on days 5 and 6. Reprinted
with permission from [70).
plasma neurophysin responses of 13 normal subjects
who smoked two cigarettes. The independent secre-
tion of these two neurophysins is clearly demon-
strated.
Nicotine in cigarette smoke is thought to stimulate
vasopressin release [71], and vasopressin and NSN
have been simultaneously assayed in 21 nicotine stim-
ulation tests in 14 volunteers (Fig. 5). In 14 tests
there was a significant rise in plasma neurophysin
and in the same 14 there was a significant elevation
of plasma vasopressin. The peak plasma neurophysin
and vasopressin values in each subject showed a high
degree of correlation (Fig. 3, r = 0.832). Further-
more, inhibition of response was demonstrated by the
prior administration of 90 ml of whiskey. Ethanol
blocked release of both vasopressin and neurophysin
in three subjects who had previously responded to
nicotine stimulation. NSN is also stimulated by hy-
potension and surgical stess [73-74]. Neurophysin
has not been observed to be elevated after overnight
Zimmerman and Robertson
dehydration but is modestly elevated in some subjects
after the administration of hypertonic saline as in the
standard Hickey-Hare procedure [74]. There appears,
therefore, to be an association between NSN release
and vasopressin release during strong stimuli for
vasopressin secretion, but with milder stimuli little or
no measurable difference in plasma neurophysin is
detected. Whether this is because the mechanism of
hormone release is different under these two condi-
tions or because changes in neurophysin are below
the lowest detectable limit during mild stimuli has not
been determined.
Human neurophysin assays for ESN and NSN
have been demonstrated to measure similar neuro-
physins in rhesus monkeys [75]. In the monkey, a
specific elevation of NSN in response to hemorrhage
was noted with NSN mean values rising from 1.3
ng/ml 0.17 SENt to 12.2 ng/ml 5.8 SEM while ESN
was unchanged. There is also a response of ESN to
estrogen in monkeys. Preliminary data from our lab-
zU)
w
zU)
2 Cigarettes
—40 20 0 +20 40 60 80 100 120
Mm
Fig. 4. Changes in plasma ESN, top, and NSN, bottom, in 13 normal
subjects who smoked two cigarettes from 0 to 12 mm. The (*)
indicates women on oral contraceptives containing estrogen. In
one subject there was a rise in plasma ESN to 2.6 ng/mI at 30 mm.
This was the subject whose NSN rose to 11.5 ng/ml. Reprinted
with permission from 1701.
 Secretion of vasopressin and neurophysin
oratories suggests that ESN may be a measure of
oxytocin secretion the way NSN may be a monitor of
vasopressin release in man and monkey. ESN secre-
tion in monkeys has been reviewed elsewhere [75].
The observations of simultaneous release of vaso-
pressin and neurophysin into the blood stream have
provided a cogent argument that vasopressin secre-
tion is accomplished by exocytosis of the contents
of neurosecretory granules. Any diffusion process
would be unlikely to allow rapid diffusion of the large
molecular weight neurophysin (approximately 10,000
mol wt) into the blood stream and, in the human,
cigarette smoking causes the simultaneous appear-
ance of vasopressin and neurophysin in blood. The
molar ratio of neurophysin to vasopressin in plasma
under these conditions as compared to this ratio in
neurosecretory granules suggests, however, that ex-
ocytosis may not be the only mechanism for vaso-
pressin secretion. While initial studies suggested that
neurophysin could bind up to 4 molecules of vaso-
pressin [76], most investigators now feel that there is
a one-to-one ratio between neurophysin and vaso-
pressin in the neurohypophysis. Hope and Pickup
calculated that this was the ratio in bovine posterior
pituitary based on bioassay concentration of vaso-
pressin and intensity of staining of neurophysin on
starch gel electrophoresis [77]. We found a molar
12.0
r = 0.832
P <0.0005
10.0
8.0
C
-c
6.0
. confined to the plasma volume whereas vasopressin is
0 The difference in the half-life of vasopressin and neu-
2 rophysin and the different volumes of distribution
4.0
. S
S
2.0
0 40 80 120 160
Vasopressin, pg/mi
Fig. 5. Correlation of vasopressin and neurophysin peak values after
cigarette smoking. Reprinted with permission from [72].
19
ratio of vasopressin to neurophysin of 0.65 in the
SON and the PVN in the posterior lobe of the ox
[78]. However, investigators have found a relative
molar excess of neurophysin in plasma when com-
pared to vasopressin. Differences in basal concentra-
tions have been explained by the longer half-life of
neurophysin. In the goat, at the onset of the appear-
ance of vasopressin in the circulation, the ratio of
vasopressin and neurophysin is close to a one-to-one
molar ratio [79]. Subsequently, the neurophysin con-
centration in plasma increases dramatically in com-
parison to the vasopressin level and the ultimate ratio
has at least a 10-molar excess of neurophysin [79].
Similar data were observed by Forsling et al with
hemorrhage in the rat where the molar ratio of vaso-
pressin and neurophysin was 2 to 1 at the onset of
vasopressin release, but within a few minutes there
was a threefold molar excess of neurophysin [80]. In
the studies of neurophysin and vasopressin release in
the human in response to smoking, there was a seven-
fold molar excess of neurophysin noted even at the
time of peak vasopressin, which was usually the same
blood sample when peak neurophysin was detected
[72]. This molar excess of neurophysin during the
basal and stimulated state was thought to be due to
the difference in half-life of vasopressin and neu-
rophysin. The half-life of vasopressin is usually re-
ported to be under 2 mm whereas the half-life of
neurophysin has been reported as 3.4 mm [78], 4.5
mm by Fawcett, Powell and Sachs [61], and 18 mm
by Robinson et al [64]. While the differences in dis-
appearance from plasma may account for the differ-
ences in the molar ratios of vasopressin and neu-
rophysin in plasma, the possibility of different pools
of neurophysin and vasopressin (e.g. extragranular)
in the posterior lobe with differing molar ratios also
must be considered possible. Part of the difference in
disappearance of the two peptides is probably due to
volume of distribution. Neurophysin appears to be
diffused throughout the extracellular fluid space [80].
suggests that these two peptides do not circulate in
the bound state. This also is suggested by the low
binding affinity of neurophysin for vasopressin [81]
and the ability of physiologic concentrations of cal-
cium to inhibit this binding [55].
Localization of hormone specific neurons in the hy-
pothalamus. Specific antibodies to individual neuro-
physins which may represent oxytocin-neurophysin
and vasopressin-neurophysin prompted studies to de-
termine if the general neurohypophysial units de-
scribed earlier could be further dissected. By radio-
20 Zimmerman and Robertson
immunoassay of neurophysin and bioassay of va-
sopressin and oxytocin it was shown that both
neurophysins and their associated hormones were lo-
cated in extracts of both the SON and the PVN in the
ox [78]. The ratios of vasopressin to neurophysin II
and oxytocin to neurophysin I were the same in both
nuclei confirming the association of each of these
hormones with a particular neurophysin [78]. There
is now immunocytochemical evidence that the hor-
mones and respective neurophysins are located in
specific cells in both nuclei in this species [82]. In the
ox [82] and rat [83] there are immunocytochemical
data that vasopressin and oxytocin are contained in
separate perikarya in the SON and the PVN support-
ing the earlier suggestion based on studies of DI rats
that the hormones are formed in separate magno-
cellular neurons [29]. In the homozygous genotype
Brattleboro rats which have no vasopressin, vaso-
pressin and neurophysin are completely missing from
ventral SON and the more medial PVN where vaso-
pressin and neurophysin are present in the normal rat
[51]. Oxytocin and its neurophysin are retained in the
other regions of these nuclei in normal and vasopres-
sin-deficient rats [51]. These data confirm the associa-
tion of specific neurophysins with each of the hor-
mones in the rat and also support the separate neuron
theory for the formation of oxytocin and vasopressin
[29, 82, 83]. In the human hypothalamus, vasopressin
is found in NSN-containing cells and oxytocin in
ESN-forming cells in the SON and PVN [17, 24].
While it is possible to stain groups of cells which
predominantly make oxytocin or vasopressin, in our
laboratory immunocytochemical studies of normal
rat and man suggest that the cellular division of labor
may be incomplete. Although vasopressin without ox-
ytocin was found in ventral SON and in parts of the
PVN, immunoreactive vasopressin was also found in
oxytocin-reactive cells in dorsal SON and other parts
of PVN [17, 51]. LeClerc and Pelletier in an im-
munoelectron microscopic study of the posterior pi-
tuitary of normal rat found that all terminals were
positive for vasopressin, suggesting that vasopressin
may be contained in oxytocin-secreting neurons [26].
Despite elaborate controls, however, cross-reactivity
of antisera to vasopressin with oxytocin may still be a
part of the problem and further studies are needed to
prove or disprove the completeness of the one-cell-
one-hormone theory.
Neurophysin and other hormones. In any case it is
now clear that many vasopressin-forming cells, as
well as those secreting oxytocin, are found in both the
SON and the PVN, and in addition in the many
"accessory" magnocellular nuclei. Whether these
"accessory" neurophysin-containing cells could con-
tam other peptides such as vasotocin, thyrotropin
releasing hormone (TRH), somatostatin, corticotro-
pin releasing factor (CR F) or gonadotropin releasing
hormone is not certain. Vasotocin is only present in
the pineal gland in the adult [841 and TRH and
somatostatin are widespread in brain while neuro-
physin is primarily restricted to the hypothalamus
and nearby areas [17]. Immunoreactive neurophysin
has been reported in the arcuate nucleus of ox [85]
and man [24] where perhaps it could be associated
with gonadotropin releasing hormone. More recent
control studies in this laboratory in human hypo-
thalamus indicate that this is probably an artefact of
the immunoperoxidase technique due to intrinsic per-
oxidase activity in arcuate cells. Vasopressin and ox-
ytocin have not been found in the arcuate nucleus
[17].
It has also been suggested that a neurophysin may
be associated with CRF [86], but our recent studies of
adrenalectomized rats provide evidence that this is
not the case. We found a great increase in neuro-
physin around portal vessels of the median eminence
after adrenalectomy, but we have shown that this is a
route of vasopressin secretion. In the normal rat after
adrenalectomy the increase in neurophysin was ac-
companied by an increase in vasopressin in the same
areas of the median eminence. In addition we have
studied the Brattleboro rat which has no vasopressin
but which does have CRF. In these rats there is no
increase in neurophysin in the zona externa of the
median eminence after adrenalectomy [50, 51]. It is
the authors' impression at present that all the neu-
rophysin-positive cells in the brain contain vasopres-
sin and/or oxytocin, and the association of a neu-
rophysin-like peptide with a releasing factor remains
speculative.
The large amount of vasopressin in hypophysial
portal blood which apparently reaches the anterior
pituitary gland reopens the question of the role of
vasopressin in anterior pituitary function. The phys-
iological meaning of the direct secretion of vasopres-
sin into the circulation of the anterior pituitary is not
fully appreciated but it seems possible that there is a
function of vasopressin in anterior pituitary secretion
involving adrenocorticotropic hormone (ACTH) and
adrenal function, perhaps during stress. This is not a
new idea [87, 88]. Since the very early work with
Gomori stains this possibility has been discussed [5].
Most, although not all [89], ultrastructural studies
reported few of the typical neurosecretory granules of
the vasopressin system in the zona externa, but it was
also found that the systemic doses of vasopressin
required to release ACTH far exceeded the amount
needed to produce an antidiuretic effect [90]. How-
 Secretion of vasopressin and neurophysin
ever, when neurophysin and vasopressin were meas-
ured in blood from individual portal veins of mon-
keys, the concentrations of vasopressin (13,500 pg/mI
or about 5 mU/ml) [30] were close to the predicted
values of Goldman and Lindner [90] and were more
than enough to stimulate ACTH release. Vasopressin
and neurophysin have now been demonstrated in the
zona externa in a number of mammalian species by
immunocytochemistry [17], but not in man (Defen-
dini R, Zimmerman EA, Robinson AG: unpublished
study of autopsy material fixed routinely in forma-
un). Better preservation of axons with other fixatives,
however, may lead to their demonstration in man in
the future. Oxytocin-containing fibers also project to
the zona externa, but they appear to be far fewer in
number than those containing vasopressin [51], and
adrenalectomy does not appear to increase the immu-
nostaining of oxytocin or its neurophysin. The au-
thors are not suggesting that vasopressin is CRF.
Other substance(s) in the hypothalamus different
from and more effective than vasopressin in releas-
ing ACTH [91, 92] have been described and termed
CRF. We are compelled, however, by the extensive
evidence of secretion into portal blood to consider
that vasopressin may act with CRF or in addition to
CRF in stimulating ACTH [93, 94].
The evidence for the secretion of vasopressin into
the ventricular system may have meaning in reference
to the experimental demonstration that CSF carries
hormone to extrahypothalamic sites for action in
memory consolidation in rats [95-97]. Administra-
tion of small amounts of antiserum to vasopressin,
but not antiserum to oxytocin, into the CSF of rats
interfered with memory consolidation [95-97]. Ho-
mozygous Brattleboro rats lacking vasopressin also
are deficient in memory functions which can be
corrected by the administration of vasopressin [97].
When vasopressin and its analogues are studied in
reference to learning and memory in man, they may
be discovered to have clinical importance.
The recent unexpected finding of vasopressin and
neurophysin but not oxytocin in parvicellular neu-
rons in the dorsomedial portion of the SCN (Fig. 2)
of the mouse [17] and rat [17, 98, 99] is of interest.
The SCN is the only site of direct retinal projections
to the hypothalamus, and lesions here result in loss of
the diurnal adrenal rhythms [100]. It has been sugges-
ted that the vasopressin in this system may have a
transmitter-like role in stimulating neurons secreting
coricotropin releasing factor (CRF) [17]. Others have
suggested that the SCN may send terminals to the
portal bed, hence to the anterior pituitary where the
hormone may affect ACTH directly [33]. A better
understanding of the role of vasopressin in the SCN
21
awaits elucidation of its efferent projections. Vaso-
pressin and neurophysin have not been found in the
SCN of monkey and man [17] and the nucleus may
have different roles in different mammals.
In summary, antibodies to vasopressin, oxytocin
and specific neurophysins have provided new insights
into the anatomy and physiology of the neurohy-
pophysis. Continued investigations with these tools
are likely to provide more information and may lead
to unsuspected functions for neurohypophysial pep-
tides.
Acknowledgments
This work was supported by The International In-
stitute for the Study of Human Reproduction, NIH
Grant AM 16166; Clinical Research Unit Grant of
the University of Pittsburgh FR56; Population Coun-
cil Grant M7432; Drs. Zimmerman and Robinson
are recipients of NIH Research Career Development
Awards NSI 1008 and AM00093. The authors dedi-
cate this work to Professor Wolfgang Bargmann on
his seventieth birthday in honor of his significant and
lifelong contributions to our understanding of neuro-
secretion. We thank our co-workers mentioned in the
text for allowing us to quote their work in progress
and for their ideas, Ms. Karen Bodnar for typing the
manuscript, and Mr. Robert Demarest and Mr. Al
Lamme for assistance with the illustrations.
Reprint requests to Dr. Earl A. Zimmerman, New York Neurolog-
ical Institute, 710 West 168th Street, New York, New York 10032,
U.S.A.
References
1. HELLER H: History of neurohypophysial research, in Hand-
book of Physiology, edited by Knobil E, Sawyer WH, Wash-
ington, D.C., American Physiological Society, 1974, vol. iv,
p. 103.
2. FARINI F: Diabete insipido ed opoterapia. Gass Osped Clin
34:1135—1139, 1913
3. VON DEN VELDEN R: Die Nierenwirkung von Hypophyse-
nextrakten. Kim Wochenschr 50:1969, 1913
4. CAJAL S, RAYMON Y: Histologie du système nerveux de
l'homme et des verthbrès. Paris, Maloine, 1911
5. BARGMANN W, SCHARREa E: The site of origin of the hor-
mones of the posterior pituitary. Am Sci 39:255—259, 1951
6. SCHARRER E, SCHARRER B: Hormones produced by neurose-
cretory cells. Recent Prog Horm Res 10:183—240, 1954
7. SCHARRER B: Neurosecretion and its role in neuroendocrine
regulation, in Pioneers in Neuroendocrinology. edited by
MEITES J, DONOVAN BT, MCCANN SM, New York, Plenum
Publishing Co., 1975, p. 260
8. SLOPER JC: The experimental and cytopathological in-
vestigation of neurosecretion in the hypothalamus and pi-
tuitary, in The Pituitary Gland, edited by HARRIS GW, D0N0-
22 Zimmerman and Robertson
VAN BT,Berkeley, University of California Press, 1966, vol.3,
p. 131
9. ARCHER R, CHAUVET J, OLIVRY G: Sur l'existence d'une
hormone unique neurohypophysaire: Relation entre l'oxytocin,
Ia vasopressine, et Ia protein de Van Dyke extradites de Ia
neurohypophyse du boeuf. Biochem Biophys Acta (Amster-
dam) 22:428—433, 1956
10. Dv VIGNEAUD V: Hormones of the posterior pituitary gland:
Oxytocin and vasopressin. The Harvey Lecture Series, L: 1—26
(1954-55), 1956
II. TURNER RA, PIERCE JG, Du VIGNEAUD V: The purification
and amino acid content of vasopressin preparations. J Biol
Chem 191:21—28, 1951
12. HELLER H: The hormone content of the vertebrae hypo-
thalamoneurohypophysial system. Br Med Bull 22:227—23 I,
1966
13. ADAMSONS K JR, STANFORD LE, VAN DYKE HB, SCHMIDT-
NIELSON B, SCHMIDT-NIELSEN K: The distribution of oxyto-
cm and vasopressin (antiduretic hormone) in the neurohy-
pophysis of the camel. Endocrinology 58:272—276, 1956
14. LEDERIS K: The distribution of vasopressin and oxytocin in
hypothalamic nuclei. Mem Soc Endocrinol 12:236—277, 1961
15. LEDERIS K: Vasopressin and oxytocin in the mammalian hy-
pothalamus. Gen Comp Endocrinology 1:80—89, 1961
16. OLIVECRONA H: Paraventricular nucleus and pituitary gland.
Ada Physiol Scand fSupp!J 136:1—178, 1957
17. ZIMMERMAN EA: Localization of hypothalamic hormones by
immunocytochemical techniques, in Frontiers in Neuroen-
docrinology, edited by Martini L, Ganong WF, New York,
Raven Press, 1976, vol. 4, p. 25
18. ROBINSON AG, FRANTZ AG: Radioimmunoassay of poste-
rior pituitary peptides: A review. Metabolism 22:1047—1057,
1973
19. ALVAREZ-BUYLLA R, LIVETT BG, Ui-I'ENTHAL LO, MILTON
SI-I, HOPE DB: lmmunohistochemical evidence for the trans-
port of neurophysin in the neurosecretory neurons of the
dog. Acta Physiol Scand /SupplJ 357:5, 1970
20. LIVETT BG, UTTENTHAL LO, HOPE DB: Localization of neu-
rophysin 11 in the hypothalamo-neurohypophysial system of
the pig by immunofuorescence histochemistry. Philos Trans
R Soc Lond (Ser B) 261:371—378, 1971
21. PARRY HB, LIVETT BG: A new hypothalamic pathway to the
median eminence containing neurophysin and its hyper-
trophy in sheep with natural scrapie. Nature 242:63—65, 1973
22. LIvErr BO: Immunochemical studies on the storage and
axonal transport of neurophysins in the hypothalamo-neu-
rohypophyseal system. Ann NY Acad Sci 248:112—113, 1975
23. WATKINS WB: Immunohistochemical demonstration of neu-
rophysin in the hypothalamoneurohypophysial system. mt
Rev Cytol 41:241—284,
(975
24. ZIMMERMAN EA, DEFENDINI R, S0K0L HW, ROBINSON AG:
The distribution of neurophysin-secreting pathways in the
mammalian brain: Light microscopic studies using the immu-
noperoxidase technique. Ann NY A cad Sd 248:92—Ill, 1975
25. LAQUER GL: Neurosecretory pathways between the hypo-
thalamic paraventricular nucleus and the neurohypophysis. J
Comp Neurol 101:543—554, 1954
26. Le Clerc R, Pelletier G: Electron microscope immunohis-
tochemical localization of vasopressin in the hypothalamus
and neurohypophysis of the normal and Brattleboro rat. Am
J Anat 140:583—588, 1974
27. PELLETIER G, LE CLERC R, LA BRIE F, PUVIANI R: Electron
microscopic immunohistochemical localization of neurophy-
sin in the rat hypothalamus and pituitary. Mo! Cell Endocri-
nol 1:157—166, 1974
28. SILVERMAN AJ, ZIMMERMAN EA: Ultrastructural immunocy-
tochemical localization of neurophysin and vasopressin in
the median eminence and posterior pituitary of the guinea
pig. Cell Tissue Res 159:291—301, 1975
29. SOK0L HW, VALTIN H: Evidence for the synthesis of oxytocin
and vasopressin in separate neurons. Nature (Lond) 214:3 14—
316, 1967
30. ZIMMERMAN EA, CARMEL PW, HUSAIN MK, FERIN M, TAN-
NENBAUM M, FRANTZ AG, ROBINSON AG: Vasopressin and
neurophysin: High concentrations in monkey hypophyseal
portal blood. Science 182:925—957, 1973
31. BURLET A, MARCHETTI J, DUHEILLE J: Immunohistochemistry
of vasopressin: Study of the hypothalamo-neurohypophysial
system of normal, dehydrated and hypophysectomized rats,
in Neurosecretion—The Final Neuroendocrine Pathway, VI
International Symposium on Neurosecretion 1973, New York,
Springer-Verlag, 1974, p. 28
32. ZIMMERMAN EA, KOZLOWSKI GP, SCOTT DE: Axonal and
ependymal pathways for the secretion of biologically active
peptides into hypophysial portal blood, in Brain-Endocrine
Interaction II.' The Ventricular System, edited by KNIGGE
KM, SCOTT DE, KOBAYASHI H, MIURA 1, BASEL S,KARGER
AG, 1975, p. 127
33. VANDESANDE F, DIERICKX K, DE MEY J: Identification of
neurophysin producing cells: I. The origin of the neurophy-
sin-like substance-containing nerve fibers in the external re-
gion of the median eminence of the rat. Cell Tissue Res
151:187—200, 1974
34. ANTUNES JL, CARMEL PW, FERIN M, ZIMMERMAN EA: Para-
ventricular nucleus: The origin of vasopressin secreting ter-
minals on hypophysial portal vessels in the monkey (ab-
stract). Endocrinology, in press
35. VORHERR H, BRADBURY WB, HOGHOUGHI M, KLEEMAN CR:
Antidiuretic hormone in cerebrospinal fluid during endogen-
ous and exogenous changes in its blood level. Endocrinology
73:246—250, 1968
36. ROBINSON AG, ZIMMERMAN EA: Cerebrospinal fluid and
ependymal neurophysin. J C/in Invest 52:1260—1267, 1973
37. Scorr DE, KozLowsKl GP, SHERIDAN MN: Scanning dee-
tronmicroscopy in the ultrastructural analysis of the mam-
malian cerebral ventricular system. mt Rev Cytol 37:349—388,
1974
38. GOLDSMITH PC, ZIMMERMAN EA:Ultrastructural localization
of neurophysin and vasopressin in rat median eminence.
Endocrinology 96:A377, 1975
39. KNOWLES F: Ependyma of the third ventricle in relation to
pituitary function. Prog Brain Res 38:255—270, 1972
40. KNIGGE KM, SCOTT DE: Structure and function of the me-
dian eminence. Am J Anat 129:223—228, 1970
41. BEN-JONATHAN N, MICAL RS, PORTER JC: Transport of LRF
from CSF to hypophysial portal and systemic blood and the
release of LH. Endocrinology 95:18—25, 1974
42. SACHS H, TAKABATAKE Y: Evidence for a precursor in vaso-
pressin biosynthesis. Endocrinology 75:939—948, 1964
43. SACHS H, FAWCETT P, TAKABATAKE Y: Biosynthesis and
release of vasopressin and neurophysin. Recent Prog Harm
Res 25:447—452, 1969
44. NORSTROM A: Axonal transport and turnover of neurohy-
pophyseal proteins in the rat. Ann NY Acad Sci 248:46—63,
1975
45. JONES CW, PICKERING BT: Intra-axonal transport and turn-
over of neurohypophysial hormones in the rat. J Physiol
(Lond) 227:553—561, 1972
 Secretion of vasopressin and neurophysin
46. BURFORD GD, PICKERING BT: Intra-axonal transport and
turnover of neurophysin in the rat: A proposal for a possible
origin of the minor neurophysin component. Biochem J
136:1047—1052, 1973
47. DOUGLAS WW: Mechanism of release of neurohypophysial
hormones: Stimulus-secretion coupling, section 7 in Hand-
book of Physiology, edited by KNOBIL E, SAWYER WH,
Washington, D.C., American Physiological Society, 1974,
vol. 4, pp. 19 1—224
48. SACHS H: Biosynthesis and release of vasopressin. Am J Med
42:687—700, 1967
49. SEW SM, HUELLMANTEL AR, STILLMAN M, RECHT L, ROBIN-
SON AG: Neurophysin and vasopressin in the plasma and
hypothalamus of adrenalectomized and normal rats. Endocri-
nology %:A272, 1975
50. STILLMAN MA, RECHT ID, SElF SM, ZIMMERMAN EA: Neu-
rophysin, vasopressin and the hypophysial portal system:
Evidence that adrenalectomy selectively increases the neu-
rophysin associated with vasopressin. Society for Neuro-
science Program Abstracts, A-697, 1975
SI. SOKOL HW, ZIMMERMAN EA, SAWYER WH, ROBINSON AG:
The hypothalamo-neurohypophysial system of the rat: Lo-
calization and quantification of neurophysin by light micro-
scopic immunocytochemistry in normal rats and in Battle-
boro rats deficient in vasopressin and a neurophysin. En-
docrinology, in press
52. KALIMO H, RINNE U: Ultrastructural studies on the hypo-
thalamic-neurohypophysial system in rats with hereditary
hypothalamic diabetes insipidus. Z Zellforsch Mikrosk Anal
134:205—225, 1972
53. VALTIM Hereditary hypothalamic diabetes insipidus in rats
FL:
(Brattleboro strain). Am J Med 42:814—827, 1967
54. HOLLENBURG MD, HOPE DB: The isolation of the native
hormone-binding proteins from bovine pituitary posterior
lobes crystallation of neurophysin I and II as complexes
with (8-arginine)-vasopressin. Biochem J 106:557—564, 1968
55. GINSBURG M, JAYASENA K, THOMAS Pi: The preparation and
properties of porcine neurophysin and the influence of cal-
cium on the hormone-neurophysin complex. J Physiol
1966
184:387—401,
56. PARDOE AU, WEATHERALL M: The intracellular localization
of oxytocic and vasopressor substances in the pituitary gland
of rats. J Physiol (Lond) 127:201—212, 1955
57. LABELLA FS, RIEFFENSTEIN RT, BEAULIEU G: Subcellular
fractionation of bovine posterior pituitary glands by centrifu-
gation. Arch Biochem Biophys 100:399—408, 1963
58. SCHAPIRO 5, STJARNE I: Evidence for the granular local-
ization of posterior pituitary hormones. Nature 189-669—674,
1961
59. PICKUP JC, JOHNSTON CI, NAKAMUR.A S, UTTENTHAL LO,
HOPE DB: Subcellular organization of neurophysins, oxyto-
cm, (8-lysine)-vasopressin and adenosine triphosphatase in
porcine posterior pituitary lobes. Biochem J 132:361—371,
1973
60. GINSBURG M, IRELAND M: The role of neurophysin in the
transport and release of neurohypophysial hormones. Endo.-
crinology 35:289—298, 1966
61. FAWCETT CP, POWELL AE, SACHS H: Biosynthesis and re-
lease of neurophysin. Endocrinology 83:1299—1310, 1968
62. LEGR0S JJ, FRANCH!MONT P, HENDRICK JC: Dosage radio-
immunologique de Ia neurphysine dans Ic serum des femmes
normales et des femmes enceintes. C R Soc Biol (Paris)
163:2773. 1969
63. CHENG KW, FRIESEN HG: Physiological factors regulating
secretion of neurophysin. Metabolism 19:876—890, 1970
23
64. ROBINSON AG, ZIMMERMAN EA, ENGLEMAN EU, FRANTZ
AG: Radioimmunoassay of bovine neurophysin: Specificity
of neurophysin land II. Metabolism 20:1138—I 147, 1971
65. LEGROS JJ, REYNAERT R, PEETERS G: Specific rekase of
bovine neurophysin I during milking and suckling rn the cow.
I Endocrinol 60:327-332, 1974
66. LEGRO5 ii, REYNAERT R, PEETERS G: Specific release of
bovine neurophysin II during arterial or venous haemorrhage
in the cow. I Endocrinol 67:297—302, 1975
67. BURI'ORD GD, JONES CW, PICKERING BT: Tentative identi-
fication of a vasopressin-neurophysin and an oxytocin-neu-
rophysin in the rat. Biochem 1 124:809—8 13, 1971
68. SUNDE DA, SOKOL HW; Quantification of rat neuropilysin
by polyacrylamide gel electrophoresis (PAGE): Application
to the rat with hereditary hypothalamic diabetes insipidus.
Ann NY Acad Sci 248:345—364, 1975
69. SElF SM, HALUSZCZAK CH, I-ILJELLMANTEL AB, ROBINSON
AG: Isolation, assay and physiologic secretion of rat neu-
rophysin. (submitted to Endocrinology)
70. ROBINSON AG: Isolation and secretion of individual human
neurophysin. J Clin Invest 55:360—367, 1975
71. DINGMAN iF, BENIRSHKE K, THORN GW: Studies of neu-
rohypophyseal function in man. Am) Med 23:226—238, 1957
72. HUSAIN MK, FRANTZ AG, CIAROCHI FF, ROBINSON AG:
Nicotine stimulated release of neurophysin and vasopressin
in humans.) Cli,, E,ndocrinol Metab 41:113—1117, 1975
73. ROBINSON AG: Radioimmunoassay of neurophysin proteins:
Utilization of specific neurophysin assays to demonstrate
independent secretion of different neurophysins in vivo Ann
NY Acad Sci 248:246—256, 1975
74. ROBINSON AG, HALUSZCZAK C, WILKINS J, HUELLMANTEL
AB, WATSON CG: Independent physiologic control of two
neurophysins in humans. (submitted to I Clin Endocrinol
Metab)
75. RosIr'sot.t AG, FERIN M, ZIMMERMAN EA: Neurophysin se-
cretion in monkeys: Emphasis on the hypothalamic respo,nse
to estrogen and ovarian events. Endocrinology 98:468—475,
1976
76. GINSBURG M, IRELAND M: The preparation of bovine neu-
rophysin an the estimation of its maximum capacity to bind
oxytocin and arginine vasopressin. J Endocrinol 32:187—198,
1965
77. HOPE DB, PICKUP JC: Neurophysins, in Handbook of Phys-
iology, edited by KNOBIL E, SAWYER WH, Washington,
D.C., American Physiological Society, 1974, vol. 4, pp. 173—
189
78. ZIMMERMAN EA, ROBINSON AG, HUSAIN MK, ACOSTA M,
FRANTZ AG, SAWYER WH: Neurohypophysial peptides in
the bovine hypothalamus: The reationship of neurophysin I
to oxytocin, and neurophysin II to vasopressin in supraoptic
and paraventricular regions. Endocrinology 95:931—939, 1974
79. McNEILLY AS, LEGROS JJ, FORSLING ML: Release of oxyto-
cm, vasopressin and neurophysin in the goat. I Endocrinol
52:209—210, 1972
80. FORSLING ML, MARTIN MJ, STURDY JC, BURTON AM: Ob-
servations on the release and clearance of neurophysin and
the neurohypophysial hormones in the rat. I Endocrinol
57:307—3 IS, 1973
81. CHENG KW, FRIESEN HG: A radioimmunoassay for vaso-
pressin binding proteins: Neurophysin. Endocrinology 88:
608, 1971
82. VANDESANDE F, DIERICKX K, DE MEY J: Identification of the
vasopressin-neurophysin II and the oxytocin-neurophysmn I
producing neurons in the bovine hypothalamus. Cell Tissue
Res 156:289—200, 1975
24 Zimmerman and Robertson
83. SWAAB DF, POOL CW, NIJVELDT F: immunofluorescence of
vasopressin and oxytocin in the rat hypothalamo-neurohy-
pophyseal system. J Neural Trans 36:195—215, 1975
84. ROSENBLOOM AA, FISHER DA: Radioimmunoassayable AVT
and AVP in adult mammalian brain tissue: Comparison of
normal and Brattleboro rats. Neuroendocrinology 17:354—
361, 1975
85. DE MEY J, DIERICKX K, VANDESANDE F: Identification of
neurophysin producing cells: 111. Immunohistochemical
demonstration of neurophysin I-producing neurons in the
bovine infundibular nucleus. Cell Tissue Res 161:219—224,
1975
86. WATKINS WB, SCHWABEDAL P, BOCK R: Immunohistochem-
ical demonstration of a CRF-associated neurophysin in the
external zone of the rat median eminence. Cell Tissue Res
152:411—421, 1974
87. KURTZMAN NA, BOONJARERN S: Physiology of antidiuretic
hormone and the interrelationship between the hormone and
the kidney. Nephron 15:167—185, 1975
88. KLEEMAN CR, LEVI J, BETTER 0: Kidney and adrenocortical
hormones. Nephron 15:261—278, 1975
89. KOBAYASHI H, MATsUI T, isHli S: Functional electron micro-
scopy of the hypothalamic median eminence. mt Rev Cytol
29:281—381, 1970
90. GOLDMAN H, LINDNER L: Antidiuretic hormone concentra-
tion in blood perfusing the adenohypophysis. Experientia
18:279—281, 1962
91. SCHALLY A, ARIMURA A, KASTIN A: Hypothalamic regu-
latory hormones. Science 179:341—350, 1973
92. SIRETT NE, PURVES H D: The assay of corticotrophin-releas-
ing factor in ACTH primed "grafted" rats, in Brain-Pi-
tuitary-Adrenal Interrelationships, edited by BR0DI5H A,
REDGATE ES, Basel, Karger, 1973, p. 88
93. HEDGE GA, YATES MB, MARCUS R, YATES FE: Site of action
of vasopressin causing corticotropin release. Endocrinology
79:328—340, 1966
94. YATES FE, RUSSEL SM, DALLMAN MF, HEDGE GA,
MCCANN SM, DHARIWAL APS: Potentiation by vasopressin
of corticotropin release induced by corticotropin-releasing
factor. Endocrinology 88:3—15, 1971
95. VAN WIMERSMA GREIDANUS Ti B, DOOTEROM J, Da WIED D:
Intraventricular administration of anti-vasopressin serum in-
hibits memory consolidation in rats. Life Sci 16:637—644,
1975
96. VAN WIMERSMA GREIDANUS Ti B, BOHUS B, DE WIED D: The
role of vasopressin in memory processes. Prog Brain Res
42:135—141, 1975
97. VAN WIMERSMA GREIDANUS Ti B, DE WIED D: The phys-
iology of the neurohypophysial system and its relation to
memory processes, in Biochemical Correlates in Brain Func-
tion, edited by DAVISON AN, New York, Academic Press,
in press
98. VANDESANDE F, DIERICKX K, DE MEY J: Identification of
vasopressin-neurophysin producing neurons of the rat supra-
chiasmatic nuclei. Cell Tissue Res 156:337—342, 1975
99. SWAAB DF, POOL CW: Specificity of oxytocin and vasopres-
sin immunofluorescence. J Endocrinol 66:263—272, 1975
100. MOORE RY, EICHLER YB: Loss of circadion adrenal corti-
costerone rhythm following suprachiasmatic lesions in the
rat. Th'ain Res 42:201—206, 1972