- Email: [email protected]
Neurotensin induces Fos and Zif268 expression in limbic nuclei of the rat brain
Pergamon
PII:
Neuroscience Vol. 75, No. 4, pp. 1141–1151, 1996 Copyright ? 1996 IBRO. Published by Elsevier Science Ltd Printed in Great Britain 0306–4522/96 $15.00+0.00 S0306-4522(96)00210-2
NEUROTENSIN INDUCES FOS AND ZIF268 EXPRESSION IN LIMBIC NUCLEI OF THE RAT BRAIN P. D. LAMBERT,* T. D. ELY, R. E. GROSS and C. D. KILTS Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, P.O. Drawer AF, Atlanta, GA 30322, U.S.A. Abstract––The endogenous tridecapeptide neurotensin exerts a wide range of behavioral, electrophysiological and neurochemical effects when administered directly into the brain. These effects are thought to result from the activation of distinct populations of neurotensin receptors distributed throughout the central nervous system. We have mapped the sites of functional change in the rat brain associated with the central administration of neurotensin using the induction of the nuclear protein products of the immediate early genes c-fos and zif268 as markers of cellular activation. The administration of neurotensin into the lateral ventricle of rats produced an increase in the number of nuclei positive for Fos and Zif268 immunoreactivity in the central and basolateral nuclei of the amygdala and the paraventricular and supraoptic nuclei of the hypothalamus. Neurotensin also produced an increase in serum corticosterone concentration and decrease in body temperature. The intraperitoneal administration of SR48692, a non-peptide neurotensin receptor antagonist, blocked the neurotensin-induced corticosterone secretion and significantly reduced the number of neurotensin-induced Fos-positive and Zif268-positive neurons in the amygdaloid complex. A significant positive correlation was found between the number of Fos-positive nuclei in the central or basolateral nucleus of the amygdala and the serum corticosterone concentration. A significant positive correlation was also found between the number of Zif-positive cells in the paraventricular nucleus of the hypothalamus and change in body temperature following treatment. Our findings indicate that the central role of neurotensin in increasing serum corticosterone involves the induction of Fos in the central and basolateral nuclei of the amygdala. In contrast, the neurotensininduced hypothermia, which was unaffected by pretreatment with SR48692, involves Zif induction in the paraventricular nucleus of the hypothalamus. These data support further the existence of central neurotensin receptor subtypes which may regulate distinct immediate early genes. Copyright ? 1996 IBRO. Published by Elsevier Science Ltd. Key words: neurotensin, Fos, Zif268, amygdala, paraventricular nucleus of the hypothalamus, corticosterone.
Neurotensin is a tridecapeptide which was first isolated and sequenced from bovine hypothalamus.5 The peptide and its receptors are distributed throughout the CNS, and neurotensin fulfills many criteria supporting its role as a neurotransmitter or neuromodulator.19,31 The diverse physiological effects of neurotensin when administered directly into the CNS are well documented,23,24 although the neural systems and neurotensin postreceptor mechanisms underlying these effects are not well understood. Neurotensin administered intracerebroventricularly (i.c.v.) has been shown to increase serum corticosterone concentrations markedly, seemingly via a dopamine-independent mechanism as this effect is *To whom correspondence should be addressed. Abbreviations: aCSF, artificial cerebrospinal fluid; Bl, basolateral amygdaloid nucleus; BSA, bovine serum albumin; Ce, central amygdaloid nucleus; CRF, corticotropinreleasing factor; DAB, diaminobenzidine; HPA, hypothalamic–pituitary–adrenal; IEG, immediate early gene; KPBS, potassium phosphate-buffered saline; NPY, neuropeptide Y; PBS, phosphate-buffered saline; PVN, paraventricular nucleus of the hypothalamus; SON, supraoptic nucleus of the hypothalamus.
not antagonized by the dopamine receptor antagonist haloperidol.13 The neurotensin-induced increase in serum corticosterone is reversed by central pretreatment with the selective neurotensin receptor antagonist SR48450 and is thought to be mediated at the level of the paraventricular nucleus of the hypothalamus (PVN).25 A marked neurotensin-induced hypothermia is also found in many species, including the rat.4 It remains unclear as to which area of the brain mediates this effect of neurotensin, although local injection studies have suggested that the floor of the fourth ventricle is the only site in which injection of neurotensin reliably produces hypothermia.17 The aim of this study was to use immunohistochemical detection of the proteins encoded by the immediate early genes (IEGs) c-fos and zif268 (also known as krox 24, egr-1 and NGFI-A) to identify areas of the brain that are activated following i.c.v. neurotensin. Physiological and pharmacological stimulation can elicit rapid and transient increases in the neuronal content of several transcription regulatory factors (for review see Ref. 16). Initial studies focused on the Fos/Jun leucine ‘zipper’ family of transcription factors;6,15,28 however, several proteins
1141
1142
P. D. Lambert et al.
belonging to the zinc finger or Zif family of transcription regulatory factors, including Zif268, are also induced by neuronal stimulation.9 The target genes and precise roles for c-fos or zif268 expression in neurons are not known; however, their expression is generally regarded as a reliable marker of receptordependent neuronal activation.16 Many of the central effects of neurotensin involve an interaction with dopamine systems and, like c-fos, zif268 expression and Zif268 induction are markedly increased in the rat brain following activation of dopamine neuronal systems.8,12 Intracerebroventricular infusions of neuropeptides result in anatomically differentiated patterns of c-fos expression and Fos induction in the basal forebrain and brainstem that represent functional maps of the neural substrates of neuropeptide actions or effects.2,18,21 Transcription factors are known to bind to DNA and are thought to couple extracellular signals to long-term responses by altering gene expression. The longer term effects of a peptide within the brain may depend on the pattern of IEG expression within a particular brain area. The DNA binding sites for Fos and Zif268 have been mapped1,7 and are distinct, suggesting that expression of each IEG would regulate the transcriptional activity of a distinct set of target genes. Therefore, it was of interest to compare the effects of central neurotensin administration on the distribution of Fos- and Zif268immunopositive cells in the rat brain. The effects of the selective non-peptide neurotensin receptor antagonist SR48692 on neurotensin-induced Fos and Zif268 expression, hypothermia and corticosterone secretion were compared to relate specific central effects of neurotensin to discrete sites in the brain. EXPERIMENTAL PROCEDURES
Materials Male rats (Sprague–Dawley, Charles River), weighing 290–310 g, were housed singly under a controlled 12-h light–dark cycle (lights on at 09.00) in a temperature- and humidity-controlled environment. Food and water were available ad libitum. Neurotensin was obtained from Peninsula Laboratories (San Carlos, CA, U.S.A.). Neurotensin(1–11) was obtained from Sigma Chemical Co. (St Louis, MO, U.S.A.). All peptides were dissolved in artificial cerebrospinal fluid (aCSF, NaCl 126 mM, NaHCO3 27 mM, KCl 2 mM, KH2PO4 0.5 mM, CaCl2 1 mM, MgCl2 0.8 mM, Na2SO4 0.5 mM, dextrose 6 mM) on the day of each study. SR48692 was obtained as a generous gift from Dr Danielle Gully (Sanofi Recherche, France) and dissolved in dimethylsulfoxide diluted (60%/40% v/v) in saline. Surgical preparation Rats were anaesthetized with xylazine (10 mg/kg i.p., Rompun; Miles Agricultural Division. Shawnee Mission, KA, U.S.A.) and ketamine (100 mg/kg i.p., Ketaset; Aveco Co., Fort Dodge, IA, U.S.A.). The head was fixed in a stereotaxic frame with the incisor bar set 3 mm below horizontal zero. A guide cannula (22 gauge) was unilaterally implanted with the tip positioned 1 mm above the left lateral ventricle at the following stereotaxic coordinates (from bregma, AP +0.8, L 1.4, V "3.0). Two stainless steel
screws were attached to the skull and the cannula was fixed in place using dental cement applied around the screws and the cannula. The incision was closed around the cannula using glue, and a stainless-steel flush-fitting stylet (26 gauge) was inserted into the cannula. Animals were allowed at least five days to recover before being used in the experimental procedure. Infusions Intracerebroventricular infusions were made in groups (n = 5) of conscious rats between 10.00 and 12.00. The rats were handled daily for five days before the i.c.v. infusions and were allowed to remain in their home cage during the infusion process. The stylet was removed and a stainlesssteel injection cannula (26 gauge) projecting 1 mm below the tip of the guide cannulae was inserted. The injection cannulae was connected (PE 20 polyethylene tubing; Clay Adams) to a 10-µl Hamilton syringe driven by a Hamilton Apparatus 22 infusion pump. The syringe and tubing were filled with aCSF. A small air bubble was drawn into the distal end of the tubing followed by a small volume of aCSF, neurotensin (0.075, 0.75, 7.5 mM) or neurotensin1–11 (7.5 mM) solution. The solution was infused at a rate of 1 µl/min for 2 min and the infusion cannulae was left in place for an additional minute before being removed and replaced by the stylet. Movement of the air bubble in the tubing was a reliable indicator of a successful infusion. Two additional groups of rats were pretreated with SR48692 (0.3 µmol/kg, i.p.) 30 min before i.c.v. infusion of aCSF or neurotensin (15 nmol). Rats were left in their home cage for 90 min following an infusion before being anaesthetized with pentobarbital (100 mg/kg i.p., Nembutal sodium solution, Abbott Laboratories, Chicago, IL, U.S.A.). Cardiac blood samples were collected and rats were transcardially perfused, first with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde in PBS. Brains were removed, placed in 4% paraformaldehyde for 4 h and then stored in 20% sucrose before being sliced (40-µm thick) on a sliding microtome. Coronal brain slices were stored in cryoprotectant solution (50% NaH2PO4, 30% ethylene glycol, 20% glycerol) at 4)C. Body temperature and serum corticosterone determinations Rectal temperatures were measured using a Model BAT-12 rectal temperature probe (Physitemp Instruments, NJ, U.S.A.) immediately before, and 90 min after, i.c.v. drug infusions. Before being perfused, rats were anaesthetized with pentobarbital and 2–3 ml of blood was rapidly collected by cardiac puncture. The blood was centrifuged at 600 g for 10 min and the serum was removed and stored at "20)C. Serum corticosterone concentration was determined by radioimmunoassay using a commercially available kit (ICN Pharmaceuticals, Costa Mesa, CA, U.S.A.). Immunohistochemical detection of Fos and Zif268 The detection by immunohistochemistry of Fos or Zif268 was conducted using free-floating coronal brain slices in 35 # 10 mm mesh-bottom polystyrene wells. Adjacent sections from each brain were stained for Fos and Zif268. Tissue was rinsed in 10 mM KH2PO4, 40 mM K2HPO4, 0.9% NaCl, pH 7.3 (KPBS) to remove the cryoprotectant. Sections were incubated for 20 min in KPBS containing 4% normal goat serum, 0.4% Triton X-100, and 1% bovine serum albumin (BSA) to block any non-specific reaction sites. The slices were then placed in a 1 : 2500 dilution of Fos antisera (affinity-purified rabbit polyclonal (Ab2), Oncogene Science, Uniondale, NY, U.S.A.) or a 1 : 1500 dilution of Zif268 antisera (affinity-purified rabbit polyclonal (sc-110), Santa Cruz Biotechnology, Santa Cruz, CA, U.S.A.) in KPBS containing 0.4% Triton X-100, 1% normal goat serum, and 0.25% BSA. The reaction mixture was agitated by orbital shaking for 60 h at 4)C. The tissue was
Neurotensin induces Fos and Zif268 in rat brain allowed to warm to room temperature for 1 h and was then rinsed 10 times with KPBS containing 0.02% Triton X-100, and 0.25% BSA. The tissue was then incubated with a 1 : 1500 dilution of biotinylated secondary antibody (goat anti-rabbit, Vector Laboratories, Burlingame, CA, U.S.A.) for 1 h, rinsed with KPBS containing 0.25% BSA and then incubated for 1 h with avidin–biotin–peroxidase complex (Vector Labs). After rinsing with KPBS, 175 mM sodium acetate and 10 mM imidazole, the slices were developed by nickel intensification of a diaminobenzidine (DAB) reaction product. Sections were then mounted on gelatin-coated slides, air-dried and counter-stained with Neutral Red before coverslips were applied. The slides were coded and individuals quantifying immunopositive nuclei were blind to the treatment conditions for the subjects. The immunohistochemical reaction product for Fos-like or Zif268-like immunoreactivity was localized to darkly stained cell nuclei. Quantification of immunopositive nuclei was accomplished using an image analysis system (Nikon Microphot-SA connected to a Macintosh Quadra PC) and software (Image, National Institutes of Mental Health). The boundaries of each brain nuclear group or subdivision were delineated in the Neutral Red-stained sections, and the total area (µm2) determined. The total number of cell nuclei within these boundaries that contained immunostaining of sufficient density to distinguish whole nuclei from background staining was counted manually. Immunopositive cells were expressed per unit area (105 µm2). The area determinations of each brain nuclear group or subdivision for each animal were compared to ensure minimal variation between subjects and to confirm further that similar anterior/posterior coordinates for each area sampled were being analysed between subjects. Statistical analysis of group differences for an effect of neurotensin were assessed by a univariate analysis of variance with post-hoc analysis using Fisher’s test for multiple comparisons. Correlations between data sets were assessed by a linear regression analysis followed by analysis of variance.
RESULTS
Distribution of neurotensin-induced Fos-positive and Zif-positive cells The unilateral injection of neurotensin into the lateral ventricle produced a significant increase in the number of Fos-positive cells, compared to the injection of aCSF, in the central nucleus of the amygdala (Ce; F3,27 = 9.84, P < 0.001), basolateral nucleus of the amygdala (Bl; F3,28 = 3.68, P < 0.03), PVN (F3,31 = 9.14, P < 0.001) and supraoptic nucleus of the hypothalamus (SON; F3,30 = 6.41, P < 0.002) (Fig. 1). The neurotensin-induced increase in Fospositive cells was dose dependent in the Ce and SON (Fig. 1); doses of neurotensin as low as 1.5 nmol produced a significant Fos induction. In contrast, neither 0.15 nmol nor 1.5 nmol doses of neurotensin resulted in significant Fos induction in the PVN. A highly significant increase in the number of Fospositive cells was seen in the Ce, Bl, PVN and SON at the highest dose of neurotensin administered (15 nmol), with the greatest number of neurotensininduced positive cells in the PVN (108 & 16 per 105 µm2, P < 0.001) (Fig. 1). A significant increase in the number of Zif-positive cells was seen in the Ce (F3,15 = 3.47, P < 0.05) and PVN (F3,15 = 11.93, P < 0.001) following injection of
1143
neurotensin compared to aCSF (Fig. 2). The magnitude of the increases in Zif-positive cells was comparable to that in Fos-positive cells, with the greatest number of neurotensin-induced Zif-positive cells in the PVN (189 & 62 per 105 µm2, P < 0.01). The distribution of neurotensin-induced Fos- or Zif-positive cells in the PVN was localized within a diagonal field that spanned the magnocellular and parvocellular subdivisions (Fig. 3). Neurotensininduced Fos or Zif induction in the Ce was localized to the lateral subdivision (Fig. 4). All neurotensininduced increases in Fos or Zif were bilateral, and as no significant difference was seen between the ipsilateral and contralateral areas, the data from each side were pooled for analysis. Fos induction by neurotensin was not observed in other amygdaloid nuclei, subdivisions of the nucleus accumbens, ventral tegmental area, substantia nigra, bed nucleus of the stria terminalis or parabrachial nucleus. In contrast to neurotensin, Fos or Zif induction was not observed in any brain areas examined following i.c.v. injection of the N-terminal fragment of neurotensin, neurotensin1–11 (15 nmol).
Serum corticosterone and body temperature responses to intracerebroventricular neurotensin A dose-dependent increase in serum corticosterone, compared to aCSF-injected controls, was seen 90 min after the i.c.v. administration of neurotensin (F3,16 = 4.93, P < 0.02) (Table 1). The highest dose of neurotensin tested (15 nmol) produced a greater than two-fold increase in the serum corticosterone concentration compared to aCSF (P < 0.01). A significant hypothermia was also induced by neurotensin (F3,19 = 10.41, P < 0.001); a 2)C drop in body temperature from baseline values was produced by 15 nmol of neurotensin (Table 1). Central injection of neurotensin1–11 (15 nmol) had no effect on serum corticosterone concentration or body temperature (Table 1).
Effect of peripherally administered SR48692 on the response of Fos, Zif268, serum corticosterone and body temperature to intracerebroventricular neurotensin An i.p. injection of this dose of SR48692 (0.3 µmol/ kg), 30 min before the i.c.v. injection of neurotensin (15 nmol), significantly attenuated the neurotensininduced increase in Fos-positive cells in the Ce (P < 0.01) (Fig 3, 4) and Bl (P < 0.05) (Fig. 2). Conversely, there was no effect of pretreatment with SR48692 on neurotensin-induced Fos induction in the PVN (P = 0.27) or SON (P = 0.34) (Fig. 2). Pretreatment with SR48692 produced a significant reduction in the number of neurotensin-induced Zif268-positive cells in the Ce (P < 0.01) and PVN (P < 0.05) (Fig. 2). The neurotensin-induced increase in serum corticosterone concentration was blocked
1144
P. D. Lambert et al.
Fig. 1. Effect of i.c.v. neurotensin on the number of Fos-immunoreactive cells in the Ce, Bl, PVN and SON. The graph shows the mean (& S.E.M.) number of Fos-immunopositive cells per unit area (105 µm2) within the Ce, Bl, PVN and SON following i.c.v. injection of aCSF (0), neurotensin (0.15, 1.5, 15 nmol) or neurotensin1–11 (1–11, 15 nmol). The top panel shows data for the Ce ( , left y-axis) and the Bl ( , right y-axis). The lower panel shows data for the PVN ( , left y-axis) and the SON ( , right y-axis). Significant differences from aCSF treatment are shown by *P < 0.05, **P < 0.01. Table 1. Effect of neurotensin, neurotensin1–11 and SR48692 on body temperature and serum corticosterone Body temperature ()C)
aCSF Neurotensin 0.15 Neurotensin 1.50 Neurotensin 15.0 SR SR + 15.0 Neurotensin1–11
t=0 37.4 & 0.15 37.7 & 0.18 37.6 & 0.31 37.8 & 0.23 37.6 & 0.15 38.2 & 0.16 37.3 & 0.14
t = 90 37.6 & 0.12 37.4 & 0.17 36.9 & 0.35 35.6 & 0.32** 37.4 & 0.21 36.3 & 0.47** 37.2 & 0.11
Corticosterone (ng/ml) t = 90 167.5 & 32.6 197.0 & 27.8 256.2 & 32.8† 389.7 & 41.3†† 111.2 & 28.7 171.8 & 85.2 137.3 & 25.4
Mean (& S.E.M.) body temperature and serum corticosterone concentration prior to (t = 0) and 90 min following (t = 90) injection of aCSF, neurotensin (0.15, 1.5, 15 nmol), SR48692 (SR), SR48692 30 min before i.c.v. injection of neurotensin (SR + 15.0) and neurotensin1–11 (15 nmol). Values represent the results of four to nine separate determinations. Significant differences between t = 0 and t = 90 are indicated by *P < 0.05, **P < 0.01. Significant differences in corticosterone concentration from aCSF treatment are indicated by †P < 0.05, ††P < 0.01.
by SR48692 pretreatment (Table 1). In contrast, neurotensin-induced hypothermia was unaffected by SR48692 (Table 1). SR48692 administration alone
did not affect basal corticosterone concentration, body temperature (Table 1) or levels of brain Fos or Zif268 expression.
Neurotensin induces Fos and Zif268 in rat brain
1145
Fig. 2. Effect of pretreatment with SR48692 on the number of neurotensin-induced Fos- or Zif268immunoreactive cells in the Ce, Bl and PVN. The graphs show the mean (& S.E.M.) number of ( ) Fosor ( ) Zif268-immunopositive cells per unit area (105 µm2) within the Ce, Bl, and PVN following i.c.v. injection of aCSF (0), neurotensin (0.15 and 15 nmol), or SR48692 (0.3 µmol/kg, i.p.) 30 min before neurotensin (15 nmol, SR+5). Significant differences from aCSF treatment are shown by *P < 0.05, ** P < 0.01. Significant differences from neurotensin (15 nmol) treatment are shown by †P<0.05, ††P < 0.01.
Relationships between responses to neurotensin The relationships between the number of Fospositive or Zif-positive cells in each brain region studied and serum corticosterone concentrations or
body temperature were assessed by regression analysis. A significant positive correlation was found between the number of Fos-positive cells in the Ce or Bl and serum corticosterone concentration (Ce
1146
P. D. Lambert et al.
Fig. 3. Photomicrographs of representative brain sections showing the effect of neurotensin on the number of Fos-immunopositive and Zif268-immunopositive cells in the PVN. Photomicrographs were taken at # 10 magnification of representative adjacent 40-µm coronal brain slices from two rats following i.c.v. injection of aCSF and stained for (a) Zif268 or (b) Fos; or i.c.v. injection of neurotensin (15 nmol) and stained for (c) Zif268 or (d) Fos. 3V, third ventricle; PaP, parvocellular PVN; MaP, magnocellular PVN.
R = 0.80, P < 0.0001; Bl R = 0.71, P < 0.0001) (Fig. 5). A smaller correlation was found between the number of Zif-positive cells in the Ce and corticosterone (R = 0.64, P < 0.05), with no correlation in the Bl (R = 0.07, P = 0.84). However, a significant correlation was found between the number of Zifpositive cells in the PVN and the serum corticosterone concentration (R = 0.84, P < 0.002), compared to the smaller correlation observed between the number of Fos-positive cells in the PVN and corticosterone (R = 0.51, P < 0.02). No significant correlation was seen between the number of Fos-positive cells in any brain area examined and the change in body temperature. Interestingly, a significant positive correlation was found between the number of Zif-positive cells in the PVN and the change in body temperature following treatment (R = 0.78,
P < 0.002) (Fig. 5). A positive correlation was also found between the number of Fos-positive cells in the Ce and in the Bl (Fos: R = 0.81, P < 0.0001). This correlation was much less for Zif-positive cells (R = 0.53, P = 0.06). In contrast, a smaller correlation was seen between Fos induction in the Ce and in the PVN (R = 0.57, P < 0.003) compared to the robust correlation between Zif induction in the Ce and in the PVN (R = 0.92, P < 0.001). DISCUSSION
The complex pharmacological effects seen after i.c.v. injection of neurotensin have been well documented. However, the discrete sites in the brain involved in mediating these effects are unknown. The findings of this study indicate that specific brain
Neurotensin induces Fos and Zif268 in rat brain
Fig. 4. Photomicrographs of representative brain sections showing the effect of pretreatment with SR48692 on the number of neurotensin-induced Fos-immunopositive cells in the Ce. Photomicrographs were taken at # 10 magnification of representative 40-µm coronal brain slices from rats, following (a) i.c.v. injection of aCSF, (b) i.c.v. injection of neurotensin (15 nmol), and (c) SR48692 (0.3 µmol/kg, i.p.) 30 min before i.c.v. injection of neurotensin. st, stria terminalis.
1147
1148
P. D. Lambert et al.
Fig. 5. Correlation between the number of Fos-immunopositive cells in the Ce and serum corticosterone concentration and the number of Zif268-immunopositive cells in the PVN and the change in body temperature. The serum corticosterone concentration (ng/ml, x-axis) 90 min following i.c.v. injection is plotted against the number of Fos-immunopositive cells per unit area (105 µm2, y-axis) within the Ce (top panel). The decrease in body temperature ()C, x-axis) 90 min following i.c.v. injection compared with before injection is plotted against the number of Zif268-immunopositive cells per unit area (105 µm2, y-axis) within the PVN (bottom panel). The data shown are from all treatment groups. The line of best fit and the R and P-values obtained by correlation analysis are shown for each area.
nuclei within the amygdala and hypothalamus are selectively activated by i.c.v. neurotensin. The effects of i.c.v. administration of neurotensin on the rat amygdaloid complex consisted of a selective increase in the number of Fos- or Zif-immunopositive cells in the lateral subdivision of the Ce and throughout the Bl; other nuclei of the amygdaloid complex showed no increase in Fos or Zif268 following i.c.v. neurotensin. We propose that the selective induction of Fos or Zif268 in the Ce and Bl following central neurotensin injection reflects a transsynaptic activation. It seems unlikely that neurotensin administered into the ventricular system gains access to the amygdala by diffusion. The localization of dense neurotensin receptors on nerve terminals in Ce, as suggested by the mismatch between neurotensin binding sites and
neurotensin mRNA,25 further suggests an indirect effect of neurotensin on Ce function. A significant positive correlation between the number of Fosimmunopositive cells in the Ce or Bl and circulating corticosterone concentrations was observed. This result supports a role for the Ce and Bl in mediating the effect of central neurotensin to increase circulating corticosterone concentrations. Previous studies investigating a role for the amygdaloid complex in the control of corticosterone secretion have found a significant and differential role for discrete nuclei in regulating basal and stress-induced hypothalamic– pituitary–adrenal (HPA) axis activity.3,11,32 Electrical stimulation of the amygdalofugal systems of Ce or Bl results in a decrease in plasma corticosterone concentration,11 whereas neurotensin administration leads
Neurotensin induces Fos and Zif268 in rat brain
to an increase in serum corticosterone. It would therefore seem that Fos induction in the Ce and Bl by neurotensin reflects receptor activation rather than spike activity in neurons involved in HPA axis activity. Our data did not demonstrate a Fos or Zif268 induction in the Ce or Bl which correlated well with the neurotensin-induced hypothermic response. The parvocellular and magnocellular subdivisions of the PVN were activated by central infusion of neurotensin, with the greatest induction of Fos or Zif268 being observed in the dorsal parvocellular region. The effect of i.c.v. neurotensin on Fos and Zif268 immunostaining in the PVN may be the result of a direct action on neurotensin receptors in the PVN.25 This PVN activation is similar to but distinct from that seen after i.c.v. infusion of other peptides such as neuropeptide Y (NPY) and corticotropinreleasing peptide (CRF). NPY also induces the largest number of Fos-positive cells in the dorsal parvocellular region;21 however, induction of Fos in the medial parvocellular region was much less following i.c.v. NPY than seen here with neurotensin. CRF increases the level of Fos in the parvocellular region of the PVN but does not similarly affect the magnocellular region.2,26 These collective results suggest that distinct populations of neurons in the PVN may be activated by different peptides. These patterns of PVN activation may underlie, in part, the distinct physiological effect(s) of peptides. We found no significant correlation between the effect of neurotensin on the number of Fos-positive cells in the PVN and serum corticosterone concentrations or body temperature. The i.c.v. infusion of neurotensin is also known to reduce food intake14 and, considering that the induction of Fos in the PVN following i.c.v. neurotensin was similar to that seen after i.c.v. infusion of NPY and CRF, both of which have profound effects on feeding via an action in the PVN,20,22 it may be that the actions of neurotensin within the PVN, via an induction of Fos, are associated with its effect on food intake. However, we did find a correlation between the number of Zif-positive cells in the PVN following treatment and both serum corticosterone concentration and change in body temperature. These data indicate that neurotensin activates receptors coupled to different IEGs which may be associated with distinct effects. In the case of neurotensin, induction of Zif268 in the PVN may lead to hypothermia, whereas induction of Fos may be involved in mediating increases in serum corticosterone. It is unclear from our data whether such effects of neurotensin in the PVN are via neurotensin receptors or whether the induction of Fos and Zif268 occurs within the same cells. An effect of i.c.v. neurotensin on Fos or Zif268 induction in structures comprising the basal ganglia– thalamocortical motor circuits was not observed, thus precluding the use of these measures to define the neural mechanisms for the locomotor decreasing effects of central neurotensin.23 This may indicate
1149
that the action of neurotensin to produce hypolocomotion involves neurons in which neurotensin does not regulate c-fos or zif268, or neuronal inhibition which cannot be reliably detected by monitoring Fos or Zif268 immunoreactivity. These data also provide further evidence for the existence of subtypes of central neurotensin receptors. Data already exist indicating that SR48692 does not attenuate all of the in vitro or in vivo effects of neurotensin.10,27 Our data suggest that SR48692 antagonizes the IEG response to neurotensin and is selective for a subtype of central neurotensin receptors responsible for mediating the effect of neurotensin on corticosterone release, but ineffective at blocking the neurotensin receptor involved in producing hypothermia. In agreement with previous reports, the central neurotensin receptors in the rat involved in mediating these two physiological effects appear to be distinct and can be differentiated using SR48692. In summary, comparison of the induction of Fos and Zif268 in response to i.c.v. neurotensin showed a similar magnitude and distribution of effect. Therefore, these data suggest that, in general, brain nuclei that are activated following i.c.v. neurotensin express both c-fos and zif268. Interestingly, pretreatment with SR48692 reduced the number of neurotensininduced Zif268-positive cells in the PVN but had no effect on the number of neurotensin-induced Fospositive cells. A discrete population of neurons may exist within the PVN that are stimulated by i.c.v. neurotensin, via an SR48692-sensitive mechanism, and show induction of Zif268 but not Fos. Our data suggest that the functional consequences of an activation of Zif268 in this population of neurons may be to mediate hypothermia. The regulation of synapsin I gene expression by Zif268 has recently been reported.30 Synapsin I is a neuron-specific protein involved in neurotransmitter release and synaptic plasticity29 which, therefore, may be involved in the action of neurotensin to regulate neurotransmitter release in the PVN (or amygdala) via an induction of Zif268.
CONCLUSIONS
The detection of Fos and Zif268 protein was used to identify sites of neural activation following neurotensin administration. Focal areas of the brain were activated and were confined to the amygdaloid complex and hypothalamus. Our data indicate a role for the Ce and Bl in the mechanism of action of neurotensin to stimulate corticosterone secretion via an induction of Fos and a role for the PVN in neurotensin-induced hypothermia via an induction of Zif268. These data also further support the existence of subtypes of central neurotensin receptors.
1150
P. D. Lambert et al. REFERENCES
1. Abate C. and Curran T. (1990) Encounters with Fos and Jun on the road to AP-1. Semin. Cancer Biol. 1, 19–26. 2. Andreae L. C. and Herbert J. (1993) Expression of c-fos in restricted areas of the basal forebrain and brainstem following single or combined intraventricular infusions of vasopressin and corticotropin-releasing factor. Neuroscience 53, 735–748. 3. Beaulieu S., Di Paolo T., Cote J. and Barden N. (1987) Participation of the central amygdaloid nucleus in the response of adrenocorticotropin secretion to immobilization stress: opposing roles of the noradrenergic and dopaminergic systems. Neuroendocrinology 45, 37–46. 4. Bissette G., Nemeroff C. B., Loosen P. T., Prange A. J. and Lipton M. A. (1976) Hypothermia and cold tolerance induced by the intracisternal administration of the hypothalamic peptide neurotensin. Nature 262, 607–609. 5. Carraway R. and Leeman S. E. (1973) The isolation of a new hypotensive peptide, neurotensin, from bovine hypothalami. J. biol. Chem. 248, 6854–6861. 6. Ceccatelli S., Villar M., Goldstein M. and Hokfelt T. (1989) Expression of c-Fos immunoreactivity in transmittercharacterised neurons after stress. Proc. natn. Acad. Sci. U.S.A. 86, 9569–9573. 7. Christy B. and Nathans D. (1989) DNA binding site of the growth factor inducible protein Zif268. Proc. natn Acad. Sci. U.S.A. 86, 8737–8741. 8. Cole A. J., Bhat R. V., Patt C., Worley P. F. and Baraban J. M. (1992) D1 dopamine receptor activation of multiple transcription factor genes in rat striatum. J. Neurochem 58, 1420–1426. 9. Dragunow M., Yamada K., Bilkey D. K. and Lawlor P. (1992) Induction of immediate-early gene proteins in dentate granule cells and somatostatin interneurons after hippocampal seizures. Molec. Brain Res. 13, 119–126. 10. Dubuc I., Costentin J., Terranova J. P., Barnouin M. C., Soubrie P., Le Fur G., Rostene W. and Kitabgi P. (1994) The nonpeptide neurotensin antagonist, SR48692, used as a tool to reveal putative neurotensin receptor subtypes. Br. J. Pharmac. 112, 352–354. 11. Dunn J. D. and Whitener J. (1986) Plasma corticosterone responses to electrical stimulation of the amygdaloid complex: cytoarchitectural specificity. Neuroendocrinology 42, 211–217. 12. Graybiel A. M., Moratalla R. and Robertson H. A. (1990) Amphetamine and cocaine induce drug-specific activation of the c-fos gene in striosome-matrix compartments and limbic subdivisions of the striatum. Proc. natn Acad. Sci. U.S.A. 87, 6912–6916. 13. Gudelsky G. A., Berry S. A. and Meltzer H. Y. (1989) Neurotensin activates tuberoinfundibular dopamine neurons and increases serum corticosterone concentrations in the rat. Neuroendocrinology 49, 604–609. 14. Hawkins M. F. (1986) Central nervous system neurotensin and feeding. Physiol. Behav. 36, 1–8. 15. Hughes P. and Dragunow M. (1994) Activation of pirenzapine-sensitive muscarinic receptors induces a specific pattern of immediate early gene expression in rat brain neurons. Brain Res. 24, 166–178. 16. Hughes P. and Dragunow M. (1995) Induction of immediate-early genes and the control of neurotransmitter-regulated gene expression within the nervous system. Pharmac. Rev. 47, 133–178. 17. Kalivas P. W., Nemeroff C. B. and Prange A. J. (1982) Neuroanatomical sites of action of neurotensin. Ann. N.Y. Acad. Sci. 400, 307–316. 18. Kamegai J., Minami S. and Wakabayashi I. (1993) Barrel rotation evoked by intracerebroventricular injection of somatostatin and arginine-vasopressin is accompanied by the induction of c-fos gene expression in the granular cells of rat cerebellum. Molec. Brain Res. 18, 115–120. 19. Kitabgi P., Checler F., Mazella J. and Vincent J. P. (1985) Pharmacology and biochemistry of neurotensin receptors. Rev. clin. Basic Pharmac. 5, 397–486. 20. Krahn D. D., Gosnell B. A., Levine A. S. and Morley J. E. (1988) Behavioral effects of corticotropin-releasing factor: localization and characterization of central effects. Brain Res. 443, 63–69. 21. Lambert P. D., Phillips P. J., Wilding J. P. H., Bloom S. R. and Herbert J. (1995) c-fos expression in the paraventricular nucleus of the hypothalamus following intracerebroventricular infusions of neuropeptide Y. Brain Res. 670, 59–65. 22. Leibowitz S. F., Sladek C., Spencer L. and Tempel D. (1988) Neuropeptide Y, epinephrine and norepinephrine in the paraventricular nucleus: stimulation of feeding and the release of corticosterone, vasopressin and glucose. Brain Res. Bull. 21, 905–912. 23. Nemeroff C. B., Bissette G., Prange A. J.Jr, Loosen P. T., Barlow T. S. and Lipton M. A. (1977) Neurotensin: central nervous system effects of a hypothalamic peptide. Brain Res. 128, 485–496. 24. Nemeroff C. B., Levant B., Myers B. and Bissette G. (1992) Neurotensin, antipsychotic drugs,and schizophrenia. Basic and clinical studies. Ann. N.Y. Acad. Sci. 668, 146–156. 25. Nicot A., Berod A., Gully D., Rowe W., Quirion R., de Kloet R. and Rostene W. (1994) Blockade of neurotensin binding in the rat hypothalamus and of the central action of neurotensin on the hypothalamic–pituitary–adrenal axis with non-peptide receptor antagonists. Neuroendocrinology 59, 572–578. 26. Parkes D., Rivest S., Lee S., Rivier C. and Vale W. (1993) Corticotropin-releasing factor activates c-fos, NGFI-B, and corticotropin-releasing factor gene expression within the paraventricular nucleus of the rat hypothalamus. Molec. Endocr. 7, 1357–1367. 27. Pinnock R. D. and Woodruff G. N. (1994) The non-peptide neurotensin receptor antagonist SR48692 is not a potent antagonist of neurotensin(8-13) responses of rat substantia nigra neurones in vitro. Neurosci. Lett. 172, 175–178. 28. Rea M. A. (1989) Light increases Fos-related protein immunoreactivity in the rat suprachiasmatic nuclei. Brain Res. Bull. 23, 577–581. 29. Rosahl T. W., Spillane D., Missler M., Herz J., Selig D. K., Wolff J. R., Hammer R. E., Malenka R. C. and Sudhof T. C. (1995) Essential functions of synapsins I and II in synaptic vesicle regulation. Nature 375, 488–493. 30. Thiel G., Schoch S. and Petersohn D. (1994) Regulation of synapsin I gene expression by the zinc finger transcription factor M268/egr-1. J. biol. Chem. 269, 15294–15301. 31. Uhl G. R., Kuhar M. J. and Snyder S. H. (1977) Neurotensin: immunohistochemical localization in the central nervous system. Proc. natn Acad. Sci. U.S.A. 4, 4059–4063.
Neurotensin induces Fos and Zif268 in rat brain 32.
1151
Van de Kar L. D., Piechowski R. A., Rittenhouse P. A. and Gray T. S. (1991) Amygdaloid lesions: differential effect on conditioned stress and immobilization-induced increases in corticosterone and renin secretion. Neuroendocrinology 54, 89–95. (Accepted 2 April 1996)
PII:
Neuroscience Vol. 75, No. 4, pp. 1141–1151, 1996 Copyright ? 1996 IBRO. Published by Elsevier Science Ltd Printed in Great Britain 0306–4522/96 $15.00+0.00 S0306-4522(96)00210-2
NEUROTENSIN INDUCES FOS AND ZIF268 EXPRESSION IN LIMBIC NUCLEI OF THE RAT BRAIN P. D. LAMBERT,* T. D. ELY, R. E. GROSS and C. D. KILTS Department of Psychiatry and Behavioral Sciences, Emory University School of Medicine, P.O. Drawer AF, Atlanta, GA 30322, U.S.A. Abstract––The endogenous tridecapeptide neurotensin exerts a wide range of behavioral, electrophysiological and neurochemical effects when administered directly into the brain. These effects are thought to result from the activation of distinct populations of neurotensin receptors distributed throughout the central nervous system. We have mapped the sites of functional change in the rat brain associated with the central administration of neurotensin using the induction of the nuclear protein products of the immediate early genes c-fos and zif268 as markers of cellular activation. The administration of neurotensin into the lateral ventricle of rats produced an increase in the number of nuclei positive for Fos and Zif268 immunoreactivity in the central and basolateral nuclei of the amygdala and the paraventricular and supraoptic nuclei of the hypothalamus. Neurotensin also produced an increase in serum corticosterone concentration and decrease in body temperature. The intraperitoneal administration of SR48692, a non-peptide neurotensin receptor antagonist, blocked the neurotensin-induced corticosterone secretion and significantly reduced the number of neurotensin-induced Fos-positive and Zif268-positive neurons in the amygdaloid complex. A significant positive correlation was found between the number of Fos-positive nuclei in the central or basolateral nucleus of the amygdala and the serum corticosterone concentration. A significant positive correlation was also found between the number of Zif-positive cells in the paraventricular nucleus of the hypothalamus and change in body temperature following treatment. Our findings indicate that the central role of neurotensin in increasing serum corticosterone involves the induction of Fos in the central and basolateral nuclei of the amygdala. In contrast, the neurotensininduced hypothermia, which was unaffected by pretreatment with SR48692, involves Zif induction in the paraventricular nucleus of the hypothalamus. These data support further the existence of central neurotensin receptor subtypes which may regulate distinct immediate early genes. Copyright ? 1996 IBRO. Published by Elsevier Science Ltd. Key words: neurotensin, Fos, Zif268, amygdala, paraventricular nucleus of the hypothalamus, corticosterone.
Neurotensin is a tridecapeptide which was first isolated and sequenced from bovine hypothalamus.5 The peptide and its receptors are distributed throughout the CNS, and neurotensin fulfills many criteria supporting its role as a neurotransmitter or neuromodulator.19,31 The diverse physiological effects of neurotensin when administered directly into the CNS are well documented,23,24 although the neural systems and neurotensin postreceptor mechanisms underlying these effects are not well understood. Neurotensin administered intracerebroventricularly (i.c.v.) has been shown to increase serum corticosterone concentrations markedly, seemingly via a dopamine-independent mechanism as this effect is *To whom correspondence should be addressed. Abbreviations: aCSF, artificial cerebrospinal fluid; Bl, basolateral amygdaloid nucleus; BSA, bovine serum albumin; Ce, central amygdaloid nucleus; CRF, corticotropinreleasing factor; DAB, diaminobenzidine; HPA, hypothalamic–pituitary–adrenal; IEG, immediate early gene; KPBS, potassium phosphate-buffered saline; NPY, neuropeptide Y; PBS, phosphate-buffered saline; PVN, paraventricular nucleus of the hypothalamus; SON, supraoptic nucleus of the hypothalamus.
not antagonized by the dopamine receptor antagonist haloperidol.13 The neurotensin-induced increase in serum corticosterone is reversed by central pretreatment with the selective neurotensin receptor antagonist SR48450 and is thought to be mediated at the level of the paraventricular nucleus of the hypothalamus (PVN).25 A marked neurotensin-induced hypothermia is also found in many species, including the rat.4 It remains unclear as to which area of the brain mediates this effect of neurotensin, although local injection studies have suggested that the floor of the fourth ventricle is the only site in which injection of neurotensin reliably produces hypothermia.17 The aim of this study was to use immunohistochemical detection of the proteins encoded by the immediate early genes (IEGs) c-fos and zif268 (also known as krox 24, egr-1 and NGFI-A) to identify areas of the brain that are activated following i.c.v. neurotensin. Physiological and pharmacological stimulation can elicit rapid and transient increases in the neuronal content of several transcription regulatory factors (for review see Ref. 16). Initial studies focused on the Fos/Jun leucine ‘zipper’ family of transcription factors;6,15,28 however, several proteins
1141
1142
P. D. Lambert et al.
belonging to the zinc finger or Zif family of transcription regulatory factors, including Zif268, are also induced by neuronal stimulation.9 The target genes and precise roles for c-fos or zif268 expression in neurons are not known; however, their expression is generally regarded as a reliable marker of receptordependent neuronal activation.16 Many of the central effects of neurotensin involve an interaction with dopamine systems and, like c-fos, zif268 expression and Zif268 induction are markedly increased in the rat brain following activation of dopamine neuronal systems.8,12 Intracerebroventricular infusions of neuropeptides result in anatomically differentiated patterns of c-fos expression and Fos induction in the basal forebrain and brainstem that represent functional maps of the neural substrates of neuropeptide actions or effects.2,18,21 Transcription factors are known to bind to DNA and are thought to couple extracellular signals to long-term responses by altering gene expression. The longer term effects of a peptide within the brain may depend on the pattern of IEG expression within a particular brain area. The DNA binding sites for Fos and Zif268 have been mapped1,7 and are distinct, suggesting that expression of each IEG would regulate the transcriptional activity of a distinct set of target genes. Therefore, it was of interest to compare the effects of central neurotensin administration on the distribution of Fos- and Zif268immunopositive cells in the rat brain. The effects of the selective non-peptide neurotensin receptor antagonist SR48692 on neurotensin-induced Fos and Zif268 expression, hypothermia and corticosterone secretion were compared to relate specific central effects of neurotensin to discrete sites in the brain. EXPERIMENTAL PROCEDURES
Materials Male rats (Sprague–Dawley, Charles River), weighing 290–310 g, were housed singly under a controlled 12-h light–dark cycle (lights on at 09.00) in a temperature- and humidity-controlled environment. Food and water were available ad libitum. Neurotensin was obtained from Peninsula Laboratories (San Carlos, CA, U.S.A.). Neurotensin(1–11) was obtained from Sigma Chemical Co. (St Louis, MO, U.S.A.). All peptides were dissolved in artificial cerebrospinal fluid (aCSF, NaCl 126 mM, NaHCO3 27 mM, KCl 2 mM, KH2PO4 0.5 mM, CaCl2 1 mM, MgCl2 0.8 mM, Na2SO4 0.5 mM, dextrose 6 mM) on the day of each study. SR48692 was obtained as a generous gift from Dr Danielle Gully (Sanofi Recherche, France) and dissolved in dimethylsulfoxide diluted (60%/40% v/v) in saline. Surgical preparation Rats were anaesthetized with xylazine (10 mg/kg i.p., Rompun; Miles Agricultural Division. Shawnee Mission, KA, U.S.A.) and ketamine (100 mg/kg i.p., Ketaset; Aveco Co., Fort Dodge, IA, U.S.A.). The head was fixed in a stereotaxic frame with the incisor bar set 3 mm below horizontal zero. A guide cannula (22 gauge) was unilaterally implanted with the tip positioned 1 mm above the left lateral ventricle at the following stereotaxic coordinates (from bregma, AP +0.8, L 1.4, V "3.0). Two stainless steel
screws were attached to the skull and the cannula was fixed in place using dental cement applied around the screws and the cannula. The incision was closed around the cannula using glue, and a stainless-steel flush-fitting stylet (26 gauge) was inserted into the cannula. Animals were allowed at least five days to recover before being used in the experimental procedure. Infusions Intracerebroventricular infusions were made in groups (n = 5) of conscious rats between 10.00 and 12.00. The rats were handled daily for five days before the i.c.v. infusions and were allowed to remain in their home cage during the infusion process. The stylet was removed and a stainlesssteel injection cannula (26 gauge) projecting 1 mm below the tip of the guide cannulae was inserted. The injection cannulae was connected (PE 20 polyethylene tubing; Clay Adams) to a 10-µl Hamilton syringe driven by a Hamilton Apparatus 22 infusion pump. The syringe and tubing were filled with aCSF. A small air bubble was drawn into the distal end of the tubing followed by a small volume of aCSF, neurotensin (0.075, 0.75, 7.5 mM) or neurotensin1–11 (7.5 mM) solution. The solution was infused at a rate of 1 µl/min for 2 min and the infusion cannulae was left in place for an additional minute before being removed and replaced by the stylet. Movement of the air bubble in the tubing was a reliable indicator of a successful infusion. Two additional groups of rats were pretreated with SR48692 (0.3 µmol/kg, i.p.) 30 min before i.c.v. infusion of aCSF or neurotensin (15 nmol). Rats were left in their home cage for 90 min following an infusion before being anaesthetized with pentobarbital (100 mg/kg i.p., Nembutal sodium solution, Abbott Laboratories, Chicago, IL, U.S.A.). Cardiac blood samples were collected and rats were transcardially perfused, first with phosphate-buffered saline (PBS) followed by 4% paraformaldehyde in PBS. Brains were removed, placed in 4% paraformaldehyde for 4 h and then stored in 20% sucrose before being sliced (40-µm thick) on a sliding microtome. Coronal brain slices were stored in cryoprotectant solution (50% NaH2PO4, 30% ethylene glycol, 20% glycerol) at 4)C. Body temperature and serum corticosterone determinations Rectal temperatures were measured using a Model BAT-12 rectal temperature probe (Physitemp Instruments, NJ, U.S.A.) immediately before, and 90 min after, i.c.v. drug infusions. Before being perfused, rats were anaesthetized with pentobarbital and 2–3 ml of blood was rapidly collected by cardiac puncture. The blood was centrifuged at 600 g for 10 min and the serum was removed and stored at "20)C. Serum corticosterone concentration was determined by radioimmunoassay using a commercially available kit (ICN Pharmaceuticals, Costa Mesa, CA, U.S.A.). Immunohistochemical detection of Fos and Zif268 The detection by immunohistochemistry of Fos or Zif268 was conducted using free-floating coronal brain slices in 35 # 10 mm mesh-bottom polystyrene wells. Adjacent sections from each brain were stained for Fos and Zif268. Tissue was rinsed in 10 mM KH2PO4, 40 mM K2HPO4, 0.9% NaCl, pH 7.3 (KPBS) to remove the cryoprotectant. Sections were incubated for 20 min in KPBS containing 4% normal goat serum, 0.4% Triton X-100, and 1% bovine serum albumin (BSA) to block any non-specific reaction sites. The slices were then placed in a 1 : 2500 dilution of Fos antisera (affinity-purified rabbit polyclonal (Ab2), Oncogene Science, Uniondale, NY, U.S.A.) or a 1 : 1500 dilution of Zif268 antisera (affinity-purified rabbit polyclonal (sc-110), Santa Cruz Biotechnology, Santa Cruz, CA, U.S.A.) in KPBS containing 0.4% Triton X-100, 1% normal goat serum, and 0.25% BSA. The reaction mixture was agitated by orbital shaking for 60 h at 4)C. The tissue was
Neurotensin induces Fos and Zif268 in rat brain allowed to warm to room temperature for 1 h and was then rinsed 10 times with KPBS containing 0.02% Triton X-100, and 0.25% BSA. The tissue was then incubated with a 1 : 1500 dilution of biotinylated secondary antibody (goat anti-rabbit, Vector Laboratories, Burlingame, CA, U.S.A.) for 1 h, rinsed with KPBS containing 0.25% BSA and then incubated for 1 h with avidin–biotin–peroxidase complex (Vector Labs). After rinsing with KPBS, 175 mM sodium acetate and 10 mM imidazole, the slices were developed by nickel intensification of a diaminobenzidine (DAB) reaction product. Sections were then mounted on gelatin-coated slides, air-dried and counter-stained with Neutral Red before coverslips were applied. The slides were coded and individuals quantifying immunopositive nuclei were blind to the treatment conditions for the subjects. The immunohistochemical reaction product for Fos-like or Zif268-like immunoreactivity was localized to darkly stained cell nuclei. Quantification of immunopositive nuclei was accomplished using an image analysis system (Nikon Microphot-SA connected to a Macintosh Quadra PC) and software (Image, National Institutes of Mental Health). The boundaries of each brain nuclear group or subdivision were delineated in the Neutral Red-stained sections, and the total area (µm2) determined. The total number of cell nuclei within these boundaries that contained immunostaining of sufficient density to distinguish whole nuclei from background staining was counted manually. Immunopositive cells were expressed per unit area (105 µm2). The area determinations of each brain nuclear group or subdivision for each animal were compared to ensure minimal variation between subjects and to confirm further that similar anterior/posterior coordinates for each area sampled were being analysed between subjects. Statistical analysis of group differences for an effect of neurotensin were assessed by a univariate analysis of variance with post-hoc analysis using Fisher’s test for multiple comparisons. Correlations between data sets were assessed by a linear regression analysis followed by analysis of variance.
RESULTS
Distribution of neurotensin-induced Fos-positive and Zif-positive cells The unilateral injection of neurotensin into the lateral ventricle produced a significant increase in the number of Fos-positive cells, compared to the injection of aCSF, in the central nucleus of the amygdala (Ce; F3,27 = 9.84, P < 0.001), basolateral nucleus of the amygdala (Bl; F3,28 = 3.68, P < 0.03), PVN (F3,31 = 9.14, P < 0.001) and supraoptic nucleus of the hypothalamus (SON; F3,30 = 6.41, P < 0.002) (Fig. 1). The neurotensin-induced increase in Fospositive cells was dose dependent in the Ce and SON (Fig. 1); doses of neurotensin as low as 1.5 nmol produced a significant Fos induction. In contrast, neither 0.15 nmol nor 1.5 nmol doses of neurotensin resulted in significant Fos induction in the PVN. A highly significant increase in the number of Fospositive cells was seen in the Ce, Bl, PVN and SON at the highest dose of neurotensin administered (15 nmol), with the greatest number of neurotensininduced positive cells in the PVN (108 & 16 per 105 µm2, P < 0.001) (Fig. 1). A significant increase in the number of Zif-positive cells was seen in the Ce (F3,15 = 3.47, P < 0.05) and PVN (F3,15 = 11.93, P < 0.001) following injection of
1143
neurotensin compared to aCSF (Fig. 2). The magnitude of the increases in Zif-positive cells was comparable to that in Fos-positive cells, with the greatest number of neurotensin-induced Zif-positive cells in the PVN (189 & 62 per 105 µm2, P < 0.01). The distribution of neurotensin-induced Fos- or Zif-positive cells in the PVN was localized within a diagonal field that spanned the magnocellular and parvocellular subdivisions (Fig. 3). Neurotensininduced Fos or Zif induction in the Ce was localized to the lateral subdivision (Fig. 4). All neurotensininduced increases in Fos or Zif were bilateral, and as no significant difference was seen between the ipsilateral and contralateral areas, the data from each side were pooled for analysis. Fos induction by neurotensin was not observed in other amygdaloid nuclei, subdivisions of the nucleus accumbens, ventral tegmental area, substantia nigra, bed nucleus of the stria terminalis or parabrachial nucleus. In contrast to neurotensin, Fos or Zif induction was not observed in any brain areas examined following i.c.v. injection of the N-terminal fragment of neurotensin, neurotensin1–11 (15 nmol).
Serum corticosterone and body temperature responses to intracerebroventricular neurotensin A dose-dependent increase in serum corticosterone, compared to aCSF-injected controls, was seen 90 min after the i.c.v. administration of neurotensin (F3,16 = 4.93, P < 0.02) (Table 1). The highest dose of neurotensin tested (15 nmol) produced a greater than two-fold increase in the serum corticosterone concentration compared to aCSF (P < 0.01). A significant hypothermia was also induced by neurotensin (F3,19 = 10.41, P < 0.001); a 2)C drop in body temperature from baseline values was produced by 15 nmol of neurotensin (Table 1). Central injection of neurotensin1–11 (15 nmol) had no effect on serum corticosterone concentration or body temperature (Table 1).
Effect of peripherally administered SR48692 on the response of Fos, Zif268, serum corticosterone and body temperature to intracerebroventricular neurotensin An i.p. injection of this dose of SR48692 (0.3 µmol/ kg), 30 min before the i.c.v. injection of neurotensin (15 nmol), significantly attenuated the neurotensininduced increase in Fos-positive cells in the Ce (P < 0.01) (Fig 3, 4) and Bl (P < 0.05) (Fig. 2). Conversely, there was no effect of pretreatment with SR48692 on neurotensin-induced Fos induction in the PVN (P = 0.27) or SON (P = 0.34) (Fig. 2). Pretreatment with SR48692 produced a significant reduction in the number of neurotensin-induced Zif268-positive cells in the Ce (P < 0.01) and PVN (P < 0.05) (Fig. 2). The neurotensin-induced increase in serum corticosterone concentration was blocked
1144
P. D. Lambert et al.
Fig. 1. Effect of i.c.v. neurotensin on the number of Fos-immunoreactive cells in the Ce, Bl, PVN and SON. The graph shows the mean (& S.E.M.) number of Fos-immunopositive cells per unit area (105 µm2) within the Ce, Bl, PVN and SON following i.c.v. injection of aCSF (0), neurotensin (0.15, 1.5, 15 nmol) or neurotensin1–11 (1–11, 15 nmol). The top panel shows data for the Ce ( , left y-axis) and the Bl ( , right y-axis). The lower panel shows data for the PVN ( , left y-axis) and the SON ( , right y-axis). Significant differences from aCSF treatment are shown by *P < 0.05, **P < 0.01. Table 1. Effect of neurotensin, neurotensin1–11 and SR48692 on body temperature and serum corticosterone Body temperature ()C)
aCSF Neurotensin 0.15 Neurotensin 1.50 Neurotensin 15.0 SR SR + 15.0 Neurotensin1–11
t=0 37.4 & 0.15 37.7 & 0.18 37.6 & 0.31 37.8 & 0.23 37.6 & 0.15 38.2 & 0.16 37.3 & 0.14
t = 90 37.6 & 0.12 37.4 & 0.17 36.9 & 0.35 35.6 & 0.32** 37.4 & 0.21 36.3 & 0.47** 37.2 & 0.11
Corticosterone (ng/ml) t = 90 167.5 & 32.6 197.0 & 27.8 256.2 & 32.8† 389.7 & 41.3†† 111.2 & 28.7 171.8 & 85.2 137.3 & 25.4
Mean (& S.E.M.) body temperature and serum corticosterone concentration prior to (t = 0) and 90 min following (t = 90) injection of aCSF, neurotensin (0.15, 1.5, 15 nmol), SR48692 (SR), SR48692 30 min before i.c.v. injection of neurotensin (SR + 15.0) and neurotensin1–11 (15 nmol). Values represent the results of four to nine separate determinations. Significant differences between t = 0 and t = 90 are indicated by *P < 0.05, **P < 0.01. Significant differences in corticosterone concentration from aCSF treatment are indicated by †P < 0.05, ††P < 0.01.
by SR48692 pretreatment (Table 1). In contrast, neurotensin-induced hypothermia was unaffected by SR48692 (Table 1). SR48692 administration alone
did not affect basal corticosterone concentration, body temperature (Table 1) or levels of brain Fos or Zif268 expression.
Neurotensin induces Fos and Zif268 in rat brain
1145
Fig. 2. Effect of pretreatment with SR48692 on the number of neurotensin-induced Fos- or Zif268immunoreactive cells in the Ce, Bl and PVN. The graphs show the mean (& S.E.M.) number of ( ) Fosor ( ) Zif268-immunopositive cells per unit area (105 µm2) within the Ce, Bl, and PVN following i.c.v. injection of aCSF (0), neurotensin (0.15 and 15 nmol), or SR48692 (0.3 µmol/kg, i.p.) 30 min before neurotensin (15 nmol, SR+5). Significant differences from aCSF treatment are shown by *P < 0.05, ** P < 0.01. Significant differences from neurotensin (15 nmol) treatment are shown by †P<0.05, ††P < 0.01.
Relationships between responses to neurotensin The relationships between the number of Fospositive or Zif-positive cells in each brain region studied and serum corticosterone concentrations or
body temperature were assessed by regression analysis. A significant positive correlation was found between the number of Fos-positive cells in the Ce or Bl and serum corticosterone concentration (Ce
1146
P. D. Lambert et al.
Fig. 3. Photomicrographs of representative brain sections showing the effect of neurotensin on the number of Fos-immunopositive and Zif268-immunopositive cells in the PVN. Photomicrographs were taken at # 10 magnification of representative adjacent 40-µm coronal brain slices from two rats following i.c.v. injection of aCSF and stained for (a) Zif268 or (b) Fos; or i.c.v. injection of neurotensin (15 nmol) and stained for (c) Zif268 or (d) Fos. 3V, third ventricle; PaP, parvocellular PVN; MaP, magnocellular PVN.
R = 0.80, P < 0.0001; Bl R = 0.71, P < 0.0001) (Fig. 5). A smaller correlation was found between the number of Zif-positive cells in the Ce and corticosterone (R = 0.64, P < 0.05), with no correlation in the Bl (R = 0.07, P = 0.84). However, a significant correlation was found between the number of Zifpositive cells in the PVN and the serum corticosterone concentration (R = 0.84, P < 0.002), compared to the smaller correlation observed between the number of Fos-positive cells in the PVN and corticosterone (R = 0.51, P < 0.02). No significant correlation was seen between the number of Fos-positive cells in any brain area examined and the change in body temperature. Interestingly, a significant positive correlation was found between the number of Zif-positive cells in the PVN and the change in body temperature following treatment (R = 0.78,
P < 0.002) (Fig. 5). A positive correlation was also found between the number of Fos-positive cells in the Ce and in the Bl (Fos: R = 0.81, P < 0.0001). This correlation was much less for Zif-positive cells (R = 0.53, P = 0.06). In contrast, a smaller correlation was seen between Fos induction in the Ce and in the PVN (R = 0.57, P < 0.003) compared to the robust correlation between Zif induction in the Ce and in the PVN (R = 0.92, P < 0.001). DISCUSSION
The complex pharmacological effects seen after i.c.v. injection of neurotensin have been well documented. However, the discrete sites in the brain involved in mediating these effects are unknown. The findings of this study indicate that specific brain
Neurotensin induces Fos and Zif268 in rat brain
Fig. 4. Photomicrographs of representative brain sections showing the effect of pretreatment with SR48692 on the number of neurotensin-induced Fos-immunopositive cells in the Ce. Photomicrographs were taken at # 10 magnification of representative 40-µm coronal brain slices from rats, following (a) i.c.v. injection of aCSF, (b) i.c.v. injection of neurotensin (15 nmol), and (c) SR48692 (0.3 µmol/kg, i.p.) 30 min before i.c.v. injection of neurotensin. st, stria terminalis.
1147
1148
P. D. Lambert et al.
Fig. 5. Correlation between the number of Fos-immunopositive cells in the Ce and serum corticosterone concentration and the number of Zif268-immunopositive cells in the PVN and the change in body temperature. The serum corticosterone concentration (ng/ml, x-axis) 90 min following i.c.v. injection is plotted against the number of Fos-immunopositive cells per unit area (105 µm2, y-axis) within the Ce (top panel). The decrease in body temperature ()C, x-axis) 90 min following i.c.v. injection compared with before injection is plotted against the number of Zif268-immunopositive cells per unit area (105 µm2, y-axis) within the PVN (bottom panel). The data shown are from all treatment groups. The line of best fit and the R and P-values obtained by correlation analysis are shown for each area.
nuclei within the amygdala and hypothalamus are selectively activated by i.c.v. neurotensin. The effects of i.c.v. administration of neurotensin on the rat amygdaloid complex consisted of a selective increase in the number of Fos- or Zif-immunopositive cells in the lateral subdivision of the Ce and throughout the Bl; other nuclei of the amygdaloid complex showed no increase in Fos or Zif268 following i.c.v. neurotensin. We propose that the selective induction of Fos or Zif268 in the Ce and Bl following central neurotensin injection reflects a transsynaptic activation. It seems unlikely that neurotensin administered into the ventricular system gains access to the amygdala by diffusion. The localization of dense neurotensin receptors on nerve terminals in Ce, as suggested by the mismatch between neurotensin binding sites and
neurotensin mRNA,25 further suggests an indirect effect of neurotensin on Ce function. A significant positive correlation between the number of Fosimmunopositive cells in the Ce or Bl and circulating corticosterone concentrations was observed. This result supports a role for the Ce and Bl in mediating the effect of central neurotensin to increase circulating corticosterone concentrations. Previous studies investigating a role for the amygdaloid complex in the control of corticosterone secretion have found a significant and differential role for discrete nuclei in regulating basal and stress-induced hypothalamic– pituitary–adrenal (HPA) axis activity.3,11,32 Electrical stimulation of the amygdalofugal systems of Ce or Bl results in a decrease in plasma corticosterone concentration,11 whereas neurotensin administration leads
Neurotensin induces Fos and Zif268 in rat brain
to an increase in serum corticosterone. It would therefore seem that Fos induction in the Ce and Bl by neurotensin reflects receptor activation rather than spike activity in neurons involved in HPA axis activity. Our data did not demonstrate a Fos or Zif268 induction in the Ce or Bl which correlated well with the neurotensin-induced hypothermic response. The parvocellular and magnocellular subdivisions of the PVN were activated by central infusion of neurotensin, with the greatest induction of Fos or Zif268 being observed in the dorsal parvocellular region. The effect of i.c.v. neurotensin on Fos and Zif268 immunostaining in the PVN may be the result of a direct action on neurotensin receptors in the PVN.25 This PVN activation is similar to but distinct from that seen after i.c.v. infusion of other peptides such as neuropeptide Y (NPY) and corticotropinreleasing peptide (CRF). NPY also induces the largest number of Fos-positive cells in the dorsal parvocellular region;21 however, induction of Fos in the medial parvocellular region was much less following i.c.v. NPY than seen here with neurotensin. CRF increases the level of Fos in the parvocellular region of the PVN but does not similarly affect the magnocellular region.2,26 These collective results suggest that distinct populations of neurons in the PVN may be activated by different peptides. These patterns of PVN activation may underlie, in part, the distinct physiological effect(s) of peptides. We found no significant correlation between the effect of neurotensin on the number of Fos-positive cells in the PVN and serum corticosterone concentrations or body temperature. The i.c.v. infusion of neurotensin is also known to reduce food intake14 and, considering that the induction of Fos in the PVN following i.c.v. neurotensin was similar to that seen after i.c.v. infusion of NPY and CRF, both of which have profound effects on feeding via an action in the PVN,20,22 it may be that the actions of neurotensin within the PVN, via an induction of Fos, are associated with its effect on food intake. However, we did find a correlation between the number of Zif-positive cells in the PVN following treatment and both serum corticosterone concentration and change in body temperature. These data indicate that neurotensin activates receptors coupled to different IEGs which may be associated with distinct effects. In the case of neurotensin, induction of Zif268 in the PVN may lead to hypothermia, whereas induction of Fos may be involved in mediating increases in serum corticosterone. It is unclear from our data whether such effects of neurotensin in the PVN are via neurotensin receptors or whether the induction of Fos and Zif268 occurs within the same cells. An effect of i.c.v. neurotensin on Fos or Zif268 induction in structures comprising the basal ganglia– thalamocortical motor circuits was not observed, thus precluding the use of these measures to define the neural mechanisms for the locomotor decreasing effects of central neurotensin.23 This may indicate
1149
that the action of neurotensin to produce hypolocomotion involves neurons in which neurotensin does not regulate c-fos or zif268, or neuronal inhibition which cannot be reliably detected by monitoring Fos or Zif268 immunoreactivity. These data also provide further evidence for the existence of subtypes of central neurotensin receptors. Data already exist indicating that SR48692 does not attenuate all of the in vitro or in vivo effects of neurotensin.10,27 Our data suggest that SR48692 antagonizes the IEG response to neurotensin and is selective for a subtype of central neurotensin receptors responsible for mediating the effect of neurotensin on corticosterone release, but ineffective at blocking the neurotensin receptor involved in producing hypothermia. In agreement with previous reports, the central neurotensin receptors in the rat involved in mediating these two physiological effects appear to be distinct and can be differentiated using SR48692. In summary, comparison of the induction of Fos and Zif268 in response to i.c.v. neurotensin showed a similar magnitude and distribution of effect. Therefore, these data suggest that, in general, brain nuclei that are activated following i.c.v. neurotensin express both c-fos and zif268. Interestingly, pretreatment with SR48692 reduced the number of neurotensininduced Zif268-positive cells in the PVN but had no effect on the number of neurotensin-induced Fospositive cells. A discrete population of neurons may exist within the PVN that are stimulated by i.c.v. neurotensin, via an SR48692-sensitive mechanism, and show induction of Zif268 but not Fos. Our data suggest that the functional consequences of an activation of Zif268 in this population of neurons may be to mediate hypothermia. The regulation of synapsin I gene expression by Zif268 has recently been reported.30 Synapsin I is a neuron-specific protein involved in neurotransmitter release and synaptic plasticity29 which, therefore, may be involved in the action of neurotensin to regulate neurotransmitter release in the PVN (or amygdala) via an induction of Zif268.
CONCLUSIONS
The detection of Fos and Zif268 protein was used to identify sites of neural activation following neurotensin administration. Focal areas of the brain were activated and were confined to the amygdaloid complex and hypothalamus. Our data indicate a role for the Ce and Bl in the mechanism of action of neurotensin to stimulate corticosterone secretion via an induction of Fos and a role for the PVN in neurotensin-induced hypothermia via an induction of Zif268. These data also further support the existence of subtypes of central neurotensin receptors.
1150
P. D. Lambert et al. REFERENCES
1. Abate C. and Curran T. (1990) Encounters with Fos and Jun on the road to AP-1. Semin. Cancer Biol. 1, 19–26. 2. Andreae L. C. and Herbert J. (1993) Expression of c-fos in restricted areas of the basal forebrain and brainstem following single or combined intraventricular infusions of vasopressin and corticotropin-releasing factor. Neuroscience 53, 735–748. 3. Beaulieu S., Di Paolo T., Cote J. and Barden N. (1987) Participation of the central amygdaloid nucleus in the response of adrenocorticotropin secretion to immobilization stress: opposing roles of the noradrenergic and dopaminergic systems. Neuroendocrinology 45, 37–46. 4. Bissette G., Nemeroff C. B., Loosen P. T., Prange A. J. and Lipton M. A. (1976) Hypothermia and cold tolerance induced by the intracisternal administration of the hypothalamic peptide neurotensin. Nature 262, 607–609. 5. Carraway R. and Leeman S. E. (1973) The isolation of a new hypotensive peptide, neurotensin, from bovine hypothalami. J. biol. Chem. 248, 6854–6861. 6. Ceccatelli S., Villar M., Goldstein M. and Hokfelt T. (1989) Expression of c-Fos immunoreactivity in transmittercharacterised neurons after stress. Proc. natn. Acad. Sci. U.S.A. 86, 9569–9573. 7. Christy B. and Nathans D. (1989) DNA binding site of the growth factor inducible protein Zif268. Proc. natn Acad. Sci. U.S.A. 86, 8737–8741. 8. Cole A. J., Bhat R. V., Patt C., Worley P. F. and Baraban J. M. (1992) D1 dopamine receptor activation of multiple transcription factor genes in rat striatum. J. Neurochem 58, 1420–1426. 9. Dragunow M., Yamada K., Bilkey D. K. and Lawlor P. (1992) Induction of immediate-early gene proteins in dentate granule cells and somatostatin interneurons after hippocampal seizures. Molec. Brain Res. 13, 119–126. 10. Dubuc I., Costentin J., Terranova J. P., Barnouin M. C., Soubrie P., Le Fur G., Rostene W. and Kitabgi P. (1994) The nonpeptide neurotensin antagonist, SR48692, used as a tool to reveal putative neurotensin receptor subtypes. Br. J. Pharmac. 112, 352–354. 11. Dunn J. D. and Whitener J. (1986) Plasma corticosterone responses to electrical stimulation of the amygdaloid complex: cytoarchitectural specificity. Neuroendocrinology 42, 211–217. 12. Graybiel A. M., Moratalla R. and Robertson H. A. (1990) Amphetamine and cocaine induce drug-specific activation of the c-fos gene in striosome-matrix compartments and limbic subdivisions of the striatum. Proc. natn Acad. Sci. U.S.A. 87, 6912–6916. 13. Gudelsky G. A., Berry S. A. and Meltzer H. Y. (1989) Neurotensin activates tuberoinfundibular dopamine neurons and increases serum corticosterone concentrations in the rat. Neuroendocrinology 49, 604–609. 14. Hawkins M. F. (1986) Central nervous system neurotensin and feeding. Physiol. Behav. 36, 1–8. 15. Hughes P. and Dragunow M. (1994) Activation of pirenzapine-sensitive muscarinic receptors induces a specific pattern of immediate early gene expression in rat brain neurons. Brain Res. 24, 166–178. 16. Hughes P. and Dragunow M. (1995) Induction of immediate-early genes and the control of neurotransmitter-regulated gene expression within the nervous system. Pharmac. Rev. 47, 133–178. 17. Kalivas P. W., Nemeroff C. B. and Prange A. J. (1982) Neuroanatomical sites of action of neurotensin. Ann. N.Y. Acad. Sci. 400, 307–316. 18. Kamegai J., Minami S. and Wakabayashi I. (1993) Barrel rotation evoked by intracerebroventricular injection of somatostatin and arginine-vasopressin is accompanied by the induction of c-fos gene expression in the granular cells of rat cerebellum. Molec. Brain Res. 18, 115–120. 19. Kitabgi P., Checler F., Mazella J. and Vincent J. P. (1985) Pharmacology and biochemistry of neurotensin receptors. Rev. clin. Basic Pharmac. 5, 397–486. 20. Krahn D. D., Gosnell B. A., Levine A. S. and Morley J. E. (1988) Behavioral effects of corticotropin-releasing factor: localization and characterization of central effects. Brain Res. 443, 63–69. 21. Lambert P. D., Phillips P. J., Wilding J. P. H., Bloom S. R. and Herbert J. (1995) c-fos expression in the paraventricular nucleus of the hypothalamus following intracerebroventricular infusions of neuropeptide Y. Brain Res. 670, 59–65. 22. Leibowitz S. F., Sladek C., Spencer L. and Tempel D. (1988) Neuropeptide Y, epinephrine and norepinephrine in the paraventricular nucleus: stimulation of feeding and the release of corticosterone, vasopressin and glucose. Brain Res. Bull. 21, 905–912. 23. Nemeroff C. B., Bissette G., Prange A. J.Jr, Loosen P. T., Barlow T. S. and Lipton M. A. (1977) Neurotensin: central nervous system effects of a hypothalamic peptide. Brain Res. 128, 485–496. 24. Nemeroff C. B., Levant B., Myers B. and Bissette G. (1992) Neurotensin, antipsychotic drugs,and schizophrenia. Basic and clinical studies. Ann. N.Y. Acad. Sci. 668, 146–156. 25. Nicot A., Berod A., Gully D., Rowe W., Quirion R., de Kloet R. and Rostene W. (1994) Blockade of neurotensin binding in the rat hypothalamus and of the central action of neurotensin on the hypothalamic–pituitary–adrenal axis with non-peptide receptor antagonists. Neuroendocrinology 59, 572–578. 26. Parkes D., Rivest S., Lee S., Rivier C. and Vale W. (1993) Corticotropin-releasing factor activates c-fos, NGFI-B, and corticotropin-releasing factor gene expression within the paraventricular nucleus of the rat hypothalamus. Molec. Endocr. 7, 1357–1367. 27. Pinnock R. D. and Woodruff G. N. (1994) The non-peptide neurotensin receptor antagonist SR48692 is not a potent antagonist of neurotensin(8-13) responses of rat substantia nigra neurones in vitro. Neurosci. Lett. 172, 175–178. 28. Rea M. A. (1989) Light increases Fos-related protein immunoreactivity in the rat suprachiasmatic nuclei. Brain Res. Bull. 23, 577–581. 29. Rosahl T. W., Spillane D., Missler M., Herz J., Selig D. K., Wolff J. R., Hammer R. E., Malenka R. C. and Sudhof T. C. (1995) Essential functions of synapsins I and II in synaptic vesicle regulation. Nature 375, 488–493. 30. Thiel G., Schoch S. and Petersohn D. (1994) Regulation of synapsin I gene expression by the zinc finger transcription factor M268/egr-1. J. biol. Chem. 269, 15294–15301. 31. Uhl G. R., Kuhar M. J. and Snyder S. H. (1977) Neurotensin: immunohistochemical localization in the central nervous system. Proc. natn Acad. Sci. U.S.A. 4, 4059–4063.
Neurotensin induces Fos and Zif268 in rat brain 32.
1151
Van de Kar L. D., Piechowski R. A., Rittenhouse P. A. and Gray T. S. (1991) Amygdaloid lesions: differential effect on conditioned stress and immobilization-induced increases in corticosterone and renin secretion. Neuroendocrinology 54, 89–95. (Accepted 2 April 1996)