Vasopressin Gene Expression in the
Rodent Hypothalamus:
Transcriptional and
Posttranscriptional Responses to
Physiological Stimulation
David Murphy and David Carter
Neuropeptide Laboratory
Institute of Molecular and Cell Biology
National University of Singapore
Singapore 0511, Republic of Singapore
The neuropeptide vasopressin (VP) is expressed in
the supraoptic nucleus, a discrete group of neurons
in the hypothalamus that respond to osmotic stimuli.
In the rat the pattern of expression of VP mRNA
changes in two ways as a consequence of the physiological stimulation of these neurons. Firstly, there
is an accumulation of VP mRNA, and secondly, the
poly(A) tail of the VP mRNA increases in length. We
asked whether the increase in VP mRNA level is a
consequence of transcriptional or posttranscriptional mechanisms. We present evidence from nuclear run-on assays that increases in the transcription of the rat VP gene are sufficient to account for
the accumulation of VP mRNA observed in chronically stimulated animals. However, we note that in
acutely stimulated animals there are rapid and relatively large increases in VP gene transcription that
do not correlate with increases in the VP mRNA
level, but coincide with the appearance of a homogeneous class of VP mRNAs with elongated poly(A)
tails. We suggest that immediately after the onset
of an acute osmotic stimulus, there is a rapid destruction of preexisting VP mRNAs and their replacement with new transcripts bearing longer poly(A)
tails. We have also addressed the question of the
function of the elongated VP mRNA poly(A) tail. It is
unlikely that the poly(A) tail extension is involved in
RNA stability; the transcriptional changes observed
are sufficient to account for the increase in VP mRNA
level, and we show that in the mouse similar increases in VP mRNA level are observed without
concomitant changes in poly(A) tail length. We did
not observe a change in the polysome distribution
of the VP mRNA after osmotic stimulation. The elongated poly(A) tail of the VP mRNA may be involved
in translational regulation or intracellular compartmentalization. (Molecular Endocrinology 4: 10511059, 1990)
INTRODUCTION
The brain peptide vasopressin (VP) is a major component of neuroendocrine systems that regulate salt and
water balance in mammals. Circulating VP, released
from stores in posterior pituitary nerve terminals, promotes water conservation by stimulating reabsorbtion
of water by kidney collecting ducts in dehydrated animals.
The VP gene is expressed in two classes of neurons
in the mammalian hypothalamus: magnocellular neurons, which make up the supraoptic nucleus (SON) and
a discrete magnocellular portion of paraventricular nucleus (PVN), and parvocellular neurons, found principally in the suprachiasmatic nucleus (SCN) and the
parvocellular part of the PVN. Magnocellular cells of the
SON and PVN innervate the posterior pituitary. Vasopressin is synthesised as part of a precursor preprohormone that is cleaved and processed during its passage from the neuronal cell bodies in the SON and PVN
to the magnocellular axon terminals in the posterior
lobe of the pituitary (1, 2). The mature VP peptide is
stored in posterior lobe axon terminals until it is released
into the peripheral circulation in response to an appropriate physiological stimulus, resulting in depletion of
stored hormone (3). Presumably to compensate for the
loss of stored material and to anticipate further demand,
there is a concomitant increase in VP precursor synthesis in the hypothalamic magnocellular neurons (4, 5).
The physiological regulation of VP synthesis in magnocellular neurons is not well understood. At the level
of expression of VP mRNA in rat hypothalamus, two
effects have been observed after an osmotic stimulus.
Firstly, there is an increase in the abundance of VP
RNA (6). It has not been determined whether this increase is a consequence of an increase in the transcription of the VP gene or an increase in the stability of the
VP RNA. Secondly, the poly(A) tail of the VP mRNA
increases dramatically in length (7-9). The function of
the poly(A) tail of eukaryotic mRNAs, and how it ulti-
0888-8809/90/1051-1059S02.00/0
Molecular Endocrinology
Copyright © 1990 by The Endocrine Society
1051
Vol 4 No. 7
MOL ENDO-1990
1052
mately affects gene expression, is still not understood.
However, there is support for the hypotheses that
poly(A), possibly through its interaction with a specific
poly(A)-binding protein, is involved in RNA stability (10,
11) and/or the control of translation (12, 13). Thus, we
can speculate that the increase in the length of the VP
mRNA poly(A) tail after an osmotic stimulus could result
in an increase in synthesis of the VP preprohormone
through two possible mechanisms. The larger poly(A)
tail could stabilize the VP mRNA, resulting in an increase
in the available pool of template for translation, and this
could explain the observed increase in steady state
levels of VP RNA observed in osmotically stimulated
rats. Alternatively, the increased length of the poly(A)
tract could directly regulate the translation or compartmentalization of VP mRNA.
We have exploited three experimental models to
investigate the regulation of VP mRNA after hyperosmotic stimulation. These models are 1) substitution of
drinking water with a solution of 2% (wt/vol) NaCI for 5
days (salt loading); this is a chronic hyperosmotic stimulus of VP secretion; 2) complete fluid deprivation for
periods of up 3 days (dehydration); dehydration is a
preferable stimulus to salt loading for investigating the
early events of an osmotic stimulus, as the former is
not dependent upon drinking behavior (9); and 3) the ip
injection of hypertonic (1.5 M) saline; this is an acute
hyperosmotic isovolemic stimulus of VP secretion (15).
We have previously shown (14, 16) that both dehydration and ip hypertonic saline injection result in an extremely rapid (within 1 - 2 h) increase in the poly(A) tail
length of the VP mRNA. However, dehydration does
not result in an increase in the abundance of the VP
mRNA until 48 h after withdrawal of fluid (17, 18).
We have sought to assess the relative contributions
of transcription, RNA stability, and translational control
to the physiological regulation of the VP gene during
hyperosmotic stimuli in two rodent species: rat and
mouse. We show that in both species a chronic osmotic
stimulus results in an increase in hypothalamic VP
mRNA abundance. However, only in the rat is there an
increase in VP RNA poly(A) tail length. We have used
nuclear run-on transcription assays to directly measure
changes in the level of transcription of the rat VP gene
as a consequence of the chronic osmotic stimuli of salt
loading for 5 days and 3 days of dehydration. Both
stimuli resulted in increases in the level of VP gene
transcription sufficient to account for the concomitant
increases in VP mRNA abundance. However, when we
investigated the early transcriptional events that follow
the onset of dehydration or the ip injection of hypertonic
saline we were surprised to observe relatively large
increases in the rate of VP gene transcription that do
not correlate with changes in steady state levels of VP
mRNA. Finally, we show by polysome analysis that the
change in poly(A) tail length has no effect on the number
of ribosomes associated with the VP mRNA.
RESULTS
Changes in the Pattern of Expression of VP mRNA
in the Rat SON and Mouse Hypothalamus after
Osmotic Stimuli
We have compared the changes in the pattern of
expression of VP mRNA in the rat SON and the mouse
hypothalamus after the onset of two osmotic stimuli:
salt loading and dehydration. Five days of salt loading
result in modest but reproducible increases in the
steady state levels of both rat SON and mouse hypothalamic VP mRNA. This is illustrated by typical Northern blot results in Fig. 1b and is quantitatively summarized in Fig. 1a. Figure 1, a and b, also illustrate that
there is an clear increase in the size of rat SON VP
transcript after 5 days of salt loading. Its migration
through the gel relative to the a-tubulin mRNA is decreased by 16% in salt-loaded animals compared to
that in control animals. This is a consequence of an
increase in the length of the poly(A) tail (7-9); when the
poly(A) tract is enzymatically removed, the RNAs from
control and stimulated animals comigrate (data not
shown). However, there is no such increase in the size
of murine VP mRNA after 5 days of salt loading (Fig. 1,
a and b). Similarly, there is no increase in the length of
the murine hypothalamic VP mRNA poly(A) tail after
dehydration; a Northern blot of a time course of dehydration over a period of 2 days is presented in Fig. 2.
A progressive increase in VP mRNA abundance is
observed, but no increase in mRNA size can be identified. In the rat SON subjected to an identical stimulus,
the change in the abundance of VP RNA is temporally
distinct from the increase in the length of the VP RNA
poly(A) tail (14). An increase in the steady state level of
VP RNA was not observed until day 2, but the increase
in poly(A) tail length can be seen as early as 2 h after
the withdrawal of drinking water (14,18). This contrasts
to the situation in the mouse hypothalamus, where
there is a more rapid accumulation of VP RNA (an
increase is observed by 12 h after water removal), and
no increase in poly(A) tail length is apparent over the
whole duration of the stimulus (Fig. 2).
Changes in Transcription of the VP Gene after
Chronic Hyperosmotic Stimuli
We have used nuclear run-on transcription assays to
assess the level of transcription of the VP gene in the
rat SON after 5 days of salt loading and 3 days of
dehydration (Fig. 3; tabulated in Table 1). After a 5-day
salt loading, transcription of the VP gene increases 2fold (Fig. 3, Exp 1). In this experiment, SON were
collected from 45 rats for each group. This increase in
transcription is consistent with the similarly modest
increase in steady state levels of VP mRNA seen in the
rat SON after 5 days of salt loading (Fig. 1a). Three
days of dehydration also resulted in a 2-fold increase in
VP gene transcription (Fig. 3, Exp 2; eight animals per
group). Again, this is consistent with the 2-fold increase
VP Gene Regulation
3
1053
VP mRNA ABUNDANCE
% Control animals
•I
C
2
12
C
3 4
2
MIGRATION OF VP RNA
••if
% Control animals
RAT
Fig. 2. Time Course of the Accumulation of VP RNA in the
Hypothalamus of Mice Subjected to Dehydration
Northern blotting was performed on RNA (20 ^g/track)
extracted from hypothalami of mice subjected to dehydration
for 0.5 days (track 2), 1 day (track 3), and 2 days (track 4) and
compared to that from control mice (track 1).
EXPERIMENT 1
PUC12
Fig. 1. Changes in the Abundance and Poly(A) Tail Length of
Rat SON and Mouse Hypothalamic VP mRNA after Salt Loading and Dehydration
a, Quantitative analysis of VP mRNA abundance and poly(A)
tail length in rat SON and mouse hypothalamus before (C) and
after (2) salt loading. The level of VP RNA was determined
relative to the level of an a-tubulin internal control and is
expressed as a percentage of the mean of the control groups
(mean ± SE). The length of the poly(A) tail is expressed as a
function of the migration of the VP RNA (V) compared to the
tubulin RNA (T). The migration (V - T) is expressed as a
percentage of the mean of the control groups (mean ± SE).
Note that an RNA with a longer poly(A) tail will have a shorter
migration. Rats, n = 5, two animals per group; mice, n = 9,
one animal per group, b, Representative autoradiograms of
Northern analysis of rat SON (2 ^g/track) and mouse hypothalamic (25 Mg/track) RNA from control (C) and 2% NaCI-loaded
(2) animals hybridized simultaneously with probes against atubulin (T) and VP RNAs. Mouse and rat RNAs were not
fractionated on the same gel, and therefore, RNA sizes cannot
be compared between species.
VP
Thy-1
'•mi' GFAP
EXPERIMENT 2
Thy-1
PUC12
«•» VP
GFAP
in VP mRNA levels found in these animals (18). Note
that in both experiments there was no increase in the
level of transcription of the Thy-1 or GFAP genes, and
there was little nonspecific association between probe
and the negative control pUC12 DNA.
Transcriptional Responses of the VP Gene to Acute
Stimuli
We have previously shown that in the rat, the magnocellular vasopressinergic neurons of the SON undergo
Fig. 3. Transcriptional Changes in Rat SON Subjected to
Chronic Stimuli
Exp 1: Nuclei from control SON (1) and salt-loaded SON (2)
were subjected to nuclear run-on analysis. In vitro labeled RNA
was hybridized to 5 nq cloned plasmids containing VP genomic
DNA (VP), glial fibrillary acidic protein cDNA (GFAP), and Thy1 genomic DNA. pUC12 DNA was used as a negative control.
Exp 2: Nuclei from control (1) and 3-day dehydrated (2) rat
SON were subjected to nuclear run-on analysis.
MOL ENDO-1990
1054
Vol 4 No. 7
Table I. Changes in the level of transcription in the SON of
the rat VP gene in response to chronic and acute osmotic
stimuli
No. of
nais/
Chronic
5-Day salt loading
3-Day dehydration
Acute
12-h dehydration (A)
12-h dehydration (B)
24-h dehydration (A)
24-h dehydration (B)
ip hypertonic NaCI, 0.5 h
ip hypertonic NaCI, 1 h
% of Control Level
Croup
VP
Thy-1
GFAP
45
8
195
211
121
133
85
61
10
8
10
8
488
245
211
440
8
586
10
249
95
92
87
132
126
77
135
109
60
90
99
77
Nuclear run-on experiments were performed as described in
Materials and Methods. Results are expressed as a percentage of the level of transcription found in parallel determinations
of SON from unstimulated (control) SON.
GFAP
EXPERIMENT 1
Thy-1
PUC12
mm
Vp
4
EXPERIMENT 2
in VP gene transcription 0.5 and 1 h after the ip introduction of hypertonic saline. These data are tabulated
in Table 1.
mm Thy-1
*
mm
PUC12
Vp
*>- GFAP
Fig. 4. Transcriptional Changes in Rat SON Subject to Acute
Stimuli
Exp 1. Run-on analysis was performed on rat SON nuclei
isolated from control (1) animals and animals subjected to the
following: 0.5-day dehydration (3), 1-day dehydration (4), and
1 h of stimulus after an ip injection of hypertonic saline (2).
Exp 2: Run-on analysis was performed on rat SON nuclei
isolated from control (1) animals and animals subjected to the
following: 0.5-day dehydration (3), 1-day dehydration (4), and
0.5 h of stimulus after an ip injection of hypertonic saline (2).
a rapid response to acute stimuli of VP secretion (14,
16, 18). The stimuli examined were dehydration (a
hyperosmotic, hypovolaemic stimulus), ip injection of
hypertonic saline (a hyperosmotic isovolemic stimulus),
and ip injection of polyethylene glycol (an isosmotic
hypovolemic stimulus). Within 2 h of the onset of the
stimuli, an increase in the length of the VP mRNA
poly(A) tail was measurable (14,16,18). An increase in
the level of VP mRNA is not a component of these early
responses; rather, accumulation of VP RNA appears to
be a chronic response in the rat (17, 18). We investigated the transcriptional changes occurring during the
early acute response to dehydration and ip hypertonic
saline (Fig. 4). We observed surprisingly large increases
in VP gene transcription 12 and 24 h after the withdrawal of water. Similarly, we observed large increases
Polysome Distribution of VP mRNA in Control and
Salt-Loaded Rat SON
One possible function of the longer VP mRNA poly(A)
tail in osmotically stimulated rats is in the stimulation of
translation (13). We tested this possibility by determining whether the distribution of the VP mRNA in polysome gradients changed after a salt load. Two experiments were performed. In the first, cytoplasmic SON
extracts from groups of 25 control or salt-loaded animals were fractionated through 5-50% (wt/vol) sucrose
gradients containing Mg 2+ ions, which maintains the
integrity of the polysomes. RNA was extracted from
gradient fractions and subjected to Northern blotting
(Fig. 5a). In a second experiment, cytoplasmic SON
extracts from control or salt-loaded rats were subjected
to two types of sucrose gradient centrifugation. Extracts from control or salt-loaded rats (45 animals/
group) were fractionated on 10-40% (wt/vol) sucrose
gradients containing Mg 2+ ions, while identical extracts
were fractionated on 10-40% (wt/vol) sucrose gradients containing EDTA. The EDTA chelates the divalent cations required for polysome integrity, resulting in
dissociation of the polysomes and the release of mRNA,
which can now be found only in lighter gradient fractions
(19, 20). RNA extracted from these gradients was also
subjected to Northern analysis (Fig. 5b). Figure 5, a and
b (i), illustrate that in both experiments the distribution
of VP mRNA in the gradients of intact polysomes of
control and salt-loaded SON extracts is identical. The
increase in poly(A) tail length does not apparently increase the overall number of ribosomes associated with
the VP mRNA. As anticipated, EDTA has the effect of
completely dissociating the polysomes from control rats
(Fig. 5b; compare i, tracks 1-10, with ii, tracks 1-10).
However, although the vast majority of VP mRNA appears in light fractions, the extracts from salt-loaded
rats contain a class of VP mRNA associated with heavy
cellular components that are apparently resistant to
EDTA dissociation (Fig. 5bii, tracks 16-20). That the
heavy polysomes are intact and quantitatively recovered was shown by reprobing the Northern filter in
Fig. 5bi with a probe directed against the c-fos mRNA.
c-fos mRNA was associated only with the heaviest
gradient fractions from both control and salt-loaded rats
(data not shown).
DISCUSSION
As a consequence of a chronic osmotic stimulus,
expression of VP RNA in rat SON is subject to two
sorts of change that are potentially important in the
adaptive response of the animal. Firstly, there is an
increase in the abundance of the VP mRNA (6). Sec-
1055
VP Gene Regulation
50% 5%
50%
a
Total RNA
1 2 3 4 5 6 7 8 9 10 1112 13 14 15 16 17 18 19 20
b i)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
ii)
CONTROL
SALT LOADED
Fig. 5. Polysomal Distribution of VP RNA from SON of Control and Salt-Loaded Rats
Distribution of VP mRNA in sucrose gradients from control (tracks 1-10) and salt-loaded rats (tracks 11-20). a, Five to 50%
sucrose gradients containing Mg z+ ; b, 10-40% sucrose gradients containing i) Mg2+ and ii) EDTA. Heavier fractions (from the
bottom of the gradient) are on the right of the Northern autoradiograms.
ondly the poly(A) tail of the VP mRNA increases in
length (7-9).
The increased abundance of VP mRNA could be
achieved by either increased transcription of the VP
gene or increased stability of the VP mRNA. Further,
an increase in the stability of the VP mRNA could be a
consequence of the increase in poly(A) tail length (10,
11). This appears to be unlikely, however, as we have
now described five lines of evidence that point to these
two effects being separately regulated and functionally
distinct. Firstly, the accumulation of VP mRNA in the
osmotically stimulated SON is dependent upon seroton-
MOL ENDO-1990
1056
inergic functions, whereas the increase in VP mRNA
poly(A) tail length appears to be independent of serotonin (9). Secondly, while there is no change in the
amount of VP mRNA in the parvocellular cells of the
SCN after an osmotic stimulation, the poly(A) tail length
of the SCN VP mRNA increases in parallel with that of
the magnocellular cells (18). This indicates that the
systems up-regulating the VP mRNA level are unique
to the magnocellular neurons, while the increase in VP
mRNA poly(A) tail is a consequence of the activation of
a pathway common to all hypothalamic vasopressinergic cells. Thirdly, the two modes of regulation are
temporally distinct; after the onset of an acute osmotic
stimulus, such as complete water deprivation, the increase in SON VP mRNA poly(A) tail length is extremely
rapid (measurable within 2 h of the onset of the stimulus), whereas the increase in the SON VP mRNA level
is not measurable until more than 24 h after the withdrawal of drinking water (17, 18). Fourthly, the two
modes of regulation are developmentally distinct; the
systems activating the osmotically stimulated increase
in VP mRNA abundance are effective even before birth
(21), the mechanisms that regulate VP mRNA poly(A)
tail length are not mature until postnatal day 11 (KumFai Chooi, D. C , and D. M., manuscript in preparation).
Fifthly, the increase in VP mRNA poly(A) tail length is
species specific. There is no such effect in the mouse,
although there is an accumulation of VP mRNA after an
osmotic stimulus similar to that in the rat. To summarize
these data, we have described circumstances where
an increase in VP mRNA abundance can occur without
an increase in VP mRNA poly(A) tail length, and circumstances where an increase in poly(A) tail length does
not result in an increase in VP mRNA abundance. Thus,
the osmotically induced accumulation of VP mRNA
must either be a consequence of increased transcription
or the activation of stability systems not involving the
poly(A) tail.
We have used nuclear run-on assays to measure
changes in VP gene transcription in the rat SON after
osmotic stimuli. The increases in transcription observed
after the chronic stimuli of 5 days of salt loading and 3
days of dehydration correlate well with the observed
changes in steady state VP mRNA levels after these
stimuli. Thus, it is not necessary to invoke a stability
mechanism to account for the accumulation of VP
mRNA in the chronically stimulated SON.
Using nuclear run-on assays we have also examined
the early transcriptional response of the VP gene to
acute stimuli. We subjected rats to two acute stimuli,
complete water deprivation and ip injection of hypertonic saline. Both of these stimuli result in a rapid
increase in the length of the VP mRNA poly(A) tail; a
change is measurable within 1 - 2 h of the onset of the
stimulus. However, no increase in VP mRNA abundance was observed at these early time points and,
indeed, it cannot be detected until after 2 days of
dehydration (18). We were surprised, therefore, to observe rapid and relatively large increases in VP gene
transcription 12 and 24 h after water deprivation and
Vol 4 No. 7
0.5 and 1 h after hypertonic saline injection. This early
activation of VP gene transcription does not correlate
with increased VP mRNA abundance. Indeed, in dehydration experiments we have observed decreases in
VP mRNA levels during the first 16 h of the stimulus
(14). In the case of ip hypertonic saline injection experiments, we have consistently observed a decrease in
VP mRNA abundance 1 h after application of the stimulus (14). Although these early transcriptional activation
events do not correlate with increases in VP mRNA
abundance, they do coincide with the increases in VP
mRNA poly(A) tail length (14, 16, 18). The poly(A) tail
length changes in a population of VP mRNAs are very
homogenous; all of the molecules in a population have
the same size poly(A) tail, and separate populations of
elongated and unelongated mRNAs do not exist (7-9,
14,16,18). Two possible mechanisms, therefore, could
mediate the increase in poly(A) tail length. Firstly, a
cytoplasmic mechanism may act upon all VP mRNAs,
both preexisting and newly synthesized. Such a mechanism could involve a cytoplasmic poly(A) polymerase
(22), or it may be the result of an increased protection
of the 3' end of the VP transcript from nucleolytic attack
as a consequence of an increased number of ribosomes
associated with a translationally stimulated VP mRNA.
This latter possibility would appear unlikely, as our data
indicate that the size of the VP mRNA polysomes does
not change with salt loading. The second mechanism
would involve the rapid degradation of preexisting VP
RNAs with shorter poly(A) tails and their replacement
with newly synthesised molecules with longer poly(A)
tails. We have shown that there is a rapid increase in
VP gene transcription, but a decrease in VP mRNA
steady state levels, after application of an acute stimulus. We, therefore, suggest that the preexisting population of VP mRNA molecules is subject to degradation
after an acute stimulus, to be rapidly replaced, through
increased VP gene transcription, by a new population
of molecules bearing an elongated poly(A) tail. The
longer poly(A) tail could be added in at the primary site
of polyadenylation in the nucleus or in the cytoplasm
by the action of a cytoplasmic poly(A) polymerase.
Alternatively, the cytoplasmic shortening of the poly(A)
tail of VP mRNA recently transported from the nucleus
may be prevented after salt loading (13).
The precise role of the 3' poly(A) tract of eukaryotic
mRNAs in the regulation of gene expression remains
to be determined. Similarly, we still cannot attribute a
function to the physiologically induced increase in length
of the hypothalamic VP mRNA poly(A) tail that follows
an osmotic stimulus. There is evidence to suggest that
the poly(A) tail is involved in the stability of mRNAs (10,
11). However, given the transcriptional changes observed during a chronic osmotic stimulation, it is not
necessary to invoke a stability mechanism to explain
the concomitant increases in VP mRNA abundance.
However, poly(A) removal, which precedes mRNA degradation, is thought to be a translation-dependent process (12). An increase in the translation of VP mRNAs
after an osmotic stimulus might, therefore, lead to a
1057
VP Gene Regulation
decrease (rather than an increase) in the pool of translatable template were it not for a compensatory increase
in poly(A) tail length. Recent evidence from yeast (13)
has further implicated the poly(A) tail in association with
the poly(A)-binding protein in the control of translation
initiation. It is, thus, conceivable that the increase in
poly(A) tail length may directly stimulate translation of
the VP mRNA and consequent accumulation of VP
precursor. We tested this hypothesis by analyzing the
polysome distribution of the VP mRNA in extracts of
control and salt-loaded SON. If the VP mRNA is subject
to an increased rate of translation initiation as a consequence of an increase in the length of its poly(A) tail,
then this may result in an increase in the number of
ribosomes bound. The VP mRNA was found to be
associated mainly with heavy polysomal fractions in
control animals, and this distribution did not change
with salt loading. The number of ribosomes associated
with the VP mRNA populations in the SON of control
and salt-loaded rats is identical. However, it should be
noted that the VP mRNA is small and is subject to a
high rate of translation even in control animals. Given
that a small message can only accommodate a limited
number of ribosomes, this may represent a physical
limit. Thus, the rates of translation initiation, elongation,
and termination, none of which can be measured by
the polysome analysis described here if the physical
size of the RNA is limiting ribosome number, may be
influenced by the length of the poly(A) tail. The influence
of poly(A) tail changes on translation of the VP mRNA
could possibly be analyzed using in vitro translation
systems prepared from SON. Additionally, we cannot
rule out the possibility that the increased poly(A) tail
length has effects on systems other than those affecting
stability and translation. For example, the longer poly(A)
tail could be involved in cellular compartmentalization.
It has recently been demonstrated that the VP mRNA
in hypothalamic cell bodies of the chronically osmotically
stimulated Brattleboro rat (1, 2) has a different intracellular distribution than that in normal animals (23). In the
latter, VP mRNA is found throughout the cytoplasm,
but in the Brattleboro rat, the VP mRNA is mainly
confined to the cytoplasmic periphery. The VP mRNA
poly(A) tail of the Brattleboro rat is elongated compared
to that of normal animals (24) as a consequence chronic
dehydration. The poly(A) tail extension might, therefore,
act as an intracellular localization signal. Similarly, we
have previously described the presence of VP RNA in
the posterior lobe of the pituitary. The abundance of
this RNA increases dramatically with a hyperosmotic
stimulus (25). We still do not know if this RNA is made
locally in posterior pituitary astrocytes (pituicytes) or if
it is transported down the pituitary stalk to posterior
pituitary nerve terminals from vasopressinergic perikarya in the hypothalamus (24). If the posterior lobe VP
RNA proves to be derived from the hypothalamus, then
perhaps the increase in the length of the poly(A) tail
forms part of a signal or facilitatory mechanism for the
transportation of this RNA. However, this is unlikely, as
the mouse, too, has VP RNA in its posterior pituitary,
which also increases in abundance with salt loading
(25). However, the VP mRNA poly(A) tail in the osmotically stimulated mouse hypothalamus does not increase in length. Another possible function of the long
poly(A) tail in osmotically stimulated rats is that it may
act as a signal for the storage of VP RNA. We have
noticed that a class of heavy VP mRNA-containing
fractions is present in EDTA-containing sucrose gradients of salt-loaded, but not control, extracts. Such
conditions dissociate polysomes, so the heavy fractions
may contain storage, or perhaps even transport, particles.
Studies in this laboratory are now focussing on identifying the VP gene sequences responsible for mediating the transcriptional and posttranscriptional changes
that occur as a consequence of an osmotic stimulus.
Further, we are attempting to identify protein factors
that interact with these sequences and ascertain the
role of intracellular signalling systems, such as cAMP
(26), in the control of the activity of these factors.
MATERIALS AND METHODS
Animals
Groups of male Sprague-Dawley rats (200-300 g) and male
CBA/J mice (8-10 weeks of age) were obtained from the
Laboratory Animal Centre (National University of Singapore)
and were maintained in constant conditions of lighting (lights
on, 0600-1800 h) and temperature. Food was freely available.
Physiological stimuli consisted of the following: 1) salt loading:
drinking water was replaced with 2% NaCI (wt/vol) at 0800 h
on day 0, and samples were taken at 0800 h 5 days later; 2)
dehydration: for the 12 h time point, water was withdrawn at
2000 h, and samples were taken at 0800 h on the following
morning. For the 24 h and 3 day time points, water was
withdrawn at 0800 h, and samples were taken at 0800 h, 24
and 72 h later; and 3) ip hypertonic saline injection (1.5 M; 1.8
ml/100 g): a single injection was administered at 0700 h, with
samples taken 0.5 and 1 h later. Rats and mice were killed by
decapitation. Brains were removed and rinsed in ice-cold PBS.
Rat SON and mouse hypothalami were dissected and either
frozen on dry ice for subsequent RNA analysis or placed into
ice-cold lysis buffer for immediate processing for nuclear runon or polysome analysis.
RNA Analysis
Total cellular RNA was isolated and analyzed on Northern
blots, as previously described (25). Northern filters were
probed either simultaneously or separately with 32P-labeled
oligonucleotides directed against the VP RNA (27) and the <xtubulin RNA (designed by New England Nuclear-DuPont, Boston, MA). Both oligonucleotide probes were synthesized by
Dr. Ben Li (National University of Singapore). Autoradiograms
were scanned using an LKB Ultroscan XL laser densitometer
(Rockville, MD). Levels of VP RNA are expressed as a fraction
of the a-tubulin internal control. The size of the VP RNA is
expressed as a function of its migration through the gel relative
to the migration of the tubulin RNA [i.e. (migration of VP RNA)
- (migration of tubulin RNA)]. Results are expressed as a
percentage of the mean of the control group ± SE.
Nuclear Run-On Analysis of Transcription
Nuclei were isolated from control and osmotically stimulated
rat SON, as previously described (28, 29). Briefly, dissected
Vol 4 No. 7
MOL ENDO-1990
1058
SON were collected into ice-cold lysis buffer [10 mM HEPES,
pH 7.9; 10 mM NaCI; 3 mM MgCI2; 0.5% (vol/vol) Nonidet-P40;
and 50 U/ml human placental ribonuclease inhibitor] and homogenized with 15 strokes of a Dounce homogenizer (Kontes,
Vineland, NJ). After a 3-min incubation on ice, the SON were
subjected to 10 further strokes. The homogenate was centrifuged at 500 x g for 10 min, and the nuclear pellet was
resuspended in 100 ^l storage buffer [50 mM HEPES, pH 7.9;
40% (vol/vol) glycerol; 5 mM MgCI2; 0.25 mM EDTA; and 500
U/ml placental RNase inhibitor], snap-frozen in liquid nitrogen,
and stored at - 8 0 C until required. The supernatant was used
for polysome analysis (see below). For the in vitro transcription
reaction, 100 /J SON nuclei were thawed on ice, and [32P]UTP
was incorporated into nascent RNA chains by the addition of
20 MI 10 x transcription buffer (1.5 M NaCI; 25 mM MgCI2; 50
mM Mg acetate; 10 mM dithiothreitol; 1.25 mM EDTA, pH 8.0;
5 mM ATP; 5 mM GTP; 5 mM CTP; 20 mM creatine phosphate;
30 U/ml creatine phosphokinase; and 5 mg/ml heparin), nucleoside 5'-diphosphate kinase to 12 U/ml, RNase inhibitor to
500 U/ml, between 100-250 MCi [32P]UTP, and water to a final
volume of 200 n\. After incubation at room temperature for 1
h, vanadylribonucleoside complex (Bethesda Research Laboratories, Gaithersburg, MD) was added to 2 mM along with
600 U RNase-free pancreatic DNase-l (Amersham, Arlington
Heights, IL). Digestion proceeded at room temperature for 30
min. The digest was incubated at 37 C for 1 h after the addition
of 25 MI 10 x PK buffer [10 mM Tris-HCI, pH 7.4; 50 mM EDTA;
10% (wt/vol) sodium dodecyl sulfate (SDS); and 1.6 mg/ml
(wt/vol) proteinase-K]. RNA was then isolated by extraction
with phenol-chloroform-isoamyl alcohol (24:24:1) and ethanol
precipitation. The RNA pellet was resuspended in 45 M' icecold 0.2 M NaOH and incubated on ice for 10 min. After
neutralization with 5 fi\ 2.4 M HEPES, the labeled RNA was
passed through a Sephadex G-50 column that had previously
been equilibrated in RNase-free water. The specific activity of
the RNA synthesized was determined by scintillation counting,
and equal ammounts of radioactive RNA from control and saltloaded SON were used to probe 5 ng specific cloned DNAs
fixed to a nylon matrix. The following plasmid clones were
used: mouse glial fibrillary acidic protein cDNA (30), mouse
Thy-1 genomic DNA (31), and rat VP genomic DNA (Phang
Chiew Hun, D. M., unpublished). The rat VP genomic subclone,
cloned into pUC18, corresponds to the 3.8-kilobase (kb) Hin6\\\
fragment previously described (32). This clone contains the
entire exon and intron sequences of the VP gene encompassing approximately 1.8 kb. Homology with the rat oxytocin gene
is mainly confined to the short (~200-basepair) neurophysincoding region (6). To prepare the slot blot filter, 5 ng DNA
were boiled for 10 min in 0.3 M NaOH. The denatured and
nicked DNA was quenched on ice and, after the addition of
NH4 acetate to 3 M, applied to Hybond-N (Amersham) using a
slot blot applicator. The filter was rinsed in 2 x SSPE (0.3 M
NaCI; 20 mM Na phosphate, pH 6.8; and 2 mM EDTA), air
dried, and baked at 80 C for 1 h. The DNA was covalently
cross-linked to the matrix by exposure to a 312-nm UV light
transilluminator for 2 min. The filter was prehybridized in 0.5
M Na phosphate (pH 6.8), 7% (wt/vol) SDS, 15% (vol/vol)
formamide (Fluka), and 1 mM EDTA (hybridization buffer) at
65 C for 5 min. Hybridization proceeded at 65 C in as small a
volume of hybridization buffer as possible (1-2 ml, depending
upon the size of the filter) for at least 60 h. After hybridization,
the filters were washed twice for 10 min at room temperature
and twice at 65 C for 10 min in 50 mM Na phosphate (pH 6.8)
containing 0.1% (wt/vol) SDS, rinsed well in 2 x SSPE, then
washed for 1 h in 10 ^g/ml (wt/vol) RNase-A in 2 x SSPE.
After two final 15-min washes in 50 mM Na phosphate (pH
6.8)-1% (wt/vol) SDS at 65 C, the filter was exposed to x-ray
film. Autoradiograms were scanned using an LKB Ultroscan
XL laser densitometer (Rockville, MD).
Polysome Analysis
This was performed by an adaptation of previously described
protocols (19, 20). The cytoplasmic extracts derived from the
isolation of nuclei (see above) were made 1 -fold concentrated
in either HKM (20 mM HEPES, pH 7.6; 100 mM KCI; and 20
mM MgCI2) or HKE (20 mM HEPES, pH 7.6; 100 mM KCI; and
10 mM EDTA) and 2 mM with vanadylribonuceoside complex
(Bethesda Research Laboratories). Extracts were then centrifuged at 10,000 x g for 10 min. The HKM- or HKE-containing
supernatants were then loaded onto HKM or HKE sucrose
gradients, respectively, as previously described (19, 20). Two
sorts of sucrose gradients were compared: 1) 5-50% (wt/vol),
and 2) 10-40% (wt/vol) with a 60% (wt/vol) cushion. After
centrifugation at 25,000 x g for 4 h in a Beckman SW28 rotor
(Palo Alto, CA), 10 gradient fractions were collected from the
bottom of the centrifuge tube into 2 vol 100% cold ethanol.
The heaviest polysome fractions are, therefore, present in the
first fraction collected. The precipitated nucleic acids were
collected by centrifugation. RNA was prepared from the pellet
as previously described (25). RNA was analyzed by Northern
blotting (25).
Acknowledgments
Received March 26,1990. Accepted April 17, 1990.
Address requests for reprints to: Dr. David Murphy, Neuropeptide Laboratory, Institute of Molecular and Cell Biology,
National University of Singapore, Singapore 0511, Republic of
Singapore.
REFERENCES
1. Richter D, Schmale H 1984 Vasopressin-expression in
normal and diabetes insipidus (Brattleboro) rats. TINS
7:317-319
2. Ivell R 1987 Vasopressinergic and oxytocinergic cells:
models in neuropeptide gene expression. In: Turner AJ
(ed) Neuropeptides and Their Peptidases. Ellis Horwood,
Chichester, pp 31-64
3. Carter D, Lightman S 1985 Neuroendocrine control of
vasopressin secretion. In: Baylis PH, Padfield PL (eds)
The Posterior Pituitary. Marcel Dekker, New York, pp 53118
4. Gainer H, Same Y, Brownstein MJ 1977 Biosynthesis and
axonal transport of rat neurohypophyseal proteins and
peptides. J Cell Biol 73:366-381
5. Brownstein MJ, Russell JT, Gainer 1980 Synthesis and
release of posterior pituitary hormones. Science 207:373387
6. Richter D 1989 Molecular events in expression of vasopressin and oxytocin and their cognate receptors. Am J
Physiol255:F207-F219
7. Carrazana EJ, Pasieka KB, Majzoub JA 1988 The vasopressin mRNA poly(A) tail is unusually long and increases
during stimulation of vasopressin gene expression in vivo.
Mol Cell Biol 8:2267-2274
8. Zingg HH, Lefebvre DL, Almazan G 1988 Regulation of
poly(A) tail size of vasopressin mRNA. J Biol Chem
263:11041-11043
9. Carter D, Murphy D 1989 Independent regulation of neuropeptide mRNA level and poly(A) tail length. J Biol Chem
264:6601-6603
10. Bernstein P, Ross J 1989 Poly(A) binding protein and the
regulation of mRNA stability. TIBS 14:373-377
11. Bernstein P, Peltz SW, Ross J 1989 The poly(A)-poly(A)binding protein complex is a major determinant of mRNA
stability in vitro. Mol Cell Biol 9:659-670
12. Wilson T, Treisman R 1988 Removal of poly(A) and consequent degradation of c-fos mRNA facilitated by 3' AUrich sequences. Nature 336:396-399
13. Sachs AB, Davis RW 1989 The poly(A) binding protein is
VP Gene Regulation
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
required for poly(A) shortening and 60S ribosomal subunit-dependent translation initiation. Cell 58:857-867
Carter D, Murphy D, Rapid changes in poly(A) tail length
of vasopressin and oxytocin mRNA's form a common
early component of neurohypophyseal peptide gene activation following physiological stimulation. Neuroendocrinology, in press
Dunn FL, Brennann TJ, Nelson AE, Robertson GL 1973
The role of blood osmolality and volume in regulating
vasopressin secretion in the rat. J Clin Invest 52:32123219
Carter D, Murphy D 1989 Activation of vasopressin gene
expression in fluid deprived rats. J Physiol 416:33P
Majzoub JA 1985 Vasopressin biosynthesis. In: Schrier
RW (ed) Vasopressin. Raven Press, New York, pp 465477
Carter D, Murphy D 1990 Vasopressin mRNA in parvocellular neurons of the rat suprachiasmatic nucleus exhibits increased poly(A) tail length following water deprivation. Neurosci Lett 109:180-185
Kleene KC, Distel RJ, Hecht NB 1984 Translational regulation and deadenylation of protamine mRNA during spermatogenesis in the mouse. Dev Biol 105:71-79
Kew D, Jin DF, Kim F, Laddis T, Kilpatrick DL 1989
Translational status of proenkephalin mRNA in the rat
reproductive system. Mol Endocrinol 3:1191-1196
Reppert SM, Uhl GR 1988 The vasopressin gene is expressed prior to regulation in the supraoptic nuclei of fetal
rats. Brain Res 456:392-396
Diez J, Brawerman G 1974 Elongation of the polyadenylate segment of messenger RNA in the cytoplasm of
mammalian cells. Proc Natl Acad Sci USA 71:4091-4095
Guitteny A-F, Fouque B, Teole R, Bloch B 1989 Vaso-
1059
24.
25.
26.
27.
28.
29.
30.
31.
32.
pressin gene expression in the normal and Brattleboro
rat: a histological analysis in semi-thin sections with biotinylated probes. J Histochem Cytochem 37:1479-1487
Levy A, Carter D, Lightman S, Murphy D, The origin and
regulation of posterior pituitary vasopressin RNA. J Neuroendocrinol, in press
Murphy D, Levy A, Lightman S, Carter D 1989 Vasopressin RNA in the neural lobe of the pituitary: dramatic
accumulation in response to salt loading. Proc Natl Acad
Sci USA 86:9002-9005
Carter D, Murphy D 1989 Cyclic nucleotide dynamics in
the rat hypothalamus during osmotic stimulation: in vivo
and in vitro studies. Brain Res 487:350-356
Lightman SL, Young III WS 1987 Vasopressin, oxytocin,
dynorphin, enkephalin and corticotrophin-releasing factor
mRNA stimulation in the rat. J Physiol 394:23-39
Mason IJ, Murphy D, Munke M, Franke U, Elliot RW,
Hogan BLM 1986 Developmental and transformationsensitive expression of the SPARC gene on mouse chromosome 11. EMBO J 5:1831-1837
Pardy K, Carter D, Murphy D 1990 Dopaminergic mediation of physiological changes in proopiomelanocortin
mRNA expression in the neurointermediate lobe of the rat
pituitary. Endocrinology 2960-2964
Lewis SA, Balcarek JM, Krek V, Shelanski M, Cowan NJ
1984 Sequence of a cDNA clone encoding mouse glial
fibrillary acidic protein: structural conservation of intermediate filaments. Proc Natl Acad Sci USA 81:2743-2746
Giguere V, Isobe K-l, Grosveld F 1985 Structure of the
murine Thy-1 gene. EMBO J 4:2017-2024
Mohr E, Schmitz E, Richter D 1988 A single rat genomic
fragment encodes both the oxytocin and vasopressin
genes separated by 11 kilobases and orientated in opposite transcriptional directions. Biochemie 70:649-654