LPS-induced Fos expression in oxytocin and vasopressin neurons of the rat hypothalamus

Brain Research 858 Ž2000. 9–18 www.elsevier.comrlocaterbres

Research report

LPS-induced Fos expression in oxytocin and vasopressin neurons of the rat hypothalamus Wataru Matsunaga a , Seiji Miyata a

a, )

, Akira Takamata b , Hiromitsu Bun a , Toshihiro Nakashima a , Toshikazu Kiyohara a

Department of Applied Biology, Kyoto Institute of Technology, Matsugasaki, Sakyo-ku, Kyoto 606-8585, Japan b Department of Physiology, Kyoto Prefectural UniÕersity of Medicine, Kamigyo-ku, Kyoto 602-0841, Japan Accepted 23 November 1999

Abstract The aim of this study was to examine the involvement of the hypothalamic oxytocin ŽOXT. and vasopressin ŽAVP. neurons in acute phase reaction using quantitative dual-labeled immunostaining with Fos and either OXT and AVP in several hypothalamic regions. Administration of low dose Ž5 mgrkg. and high dose Ž125 mgrkg. of LPS induced intense nuclear Fos immunoreactivity in many OXT and AVP neurons in all the observed hypothalamic regions. The percentage of Fos-positive nuclei in OXT magnocellular neurons was higher than that of AVP magnocellular neurons in the supraoptic nucleus ŽSON., the magnocellular neurons in the paraventricular nucleus ŽmagPVN., rostral SON ŽrSON., and nucleus circularis ŽNC., whose axons terminate at the posterior pituitary for peripheral release. The percentage of Fos-positive nuclei in AVP parvocellular neurons in the paraventricular nucleus ŽparPVN. was higher than that of OXT parvocellular neurons, whose axons terminate within the brain for central release. Moreover, the percentage of Fos-positive nuclei in AVP magnocellular neurons of the SON and rSON was significantly higher than that of the magPVN and NC when animals were given LPS via intraperitoneal Ži.p..-injection. This regional heterogeneity was not observed in OXT magnocellular neurons of i.p.-injected rats or in either OXT or AVP magnocellular neurons of intravenous Ži.v..-injected rats. The present data suggest that LPS-induced peripheral release of AVP and OXT is due to the activation of the magnocellular neurons in the SON, magPVN, NC, and rSON, and the central release of those hormones is in part derived from the activation of parvocellular neurons in the PVN. It is also suggested that the activation of AVP magnocellular neurons is heterogeneous among the four hypothalamic regions, but that of OXT magnocellular neurons is homogenous among these brain regions in response to LPS administration. q 2000 Elsevier Science B.V. All rights reserved. Keywords: Fever; Hypothalamus; Acute-phase response; IL-1b; Magnocellular neuron

1. Introduction The production of cytokines within macrophages and lymphocytes by contact between animals and exogenous pyrogen represents an essential feature of the early events of immune activation that characterize the acute-phase responses w4x. Lipopolysaccharide ŽLPS. has been widely used as an endotoxin to mimic some of the acute-phase responses. Systemic LPS administration triggers the synthesis and release of cytokines such as interleukin-1b ŽIL-1b ., interleukin-6 ŽIL-6., and tumor necrosis factor-a ŽTNF-a . w13,15,19x. These responses include various

) Corresponding [email protected]

author.

Fax:

q81-75-724-7760;

e-m ail:

physiological reactions, and well-known response is an elevation of body temperature. Oxytocin ŽOXT. and arginine vasopressin ŽAVP. are neurohypophysial hormones and are synthesized mainly in the magnocellular neurons of the supraoptic nucleus ŽSON. and paraventricular nucleus ŽPVN. and partially in the nucleus circularis ŽNC. and rostral supraoptic nucleus ŽrSON. in the hypothalamus w44x. It is well known that OXT contributes to milk ejection, uterus contraction, increase of insulin and glucagon secretion, and AVP has functional significance in body fluid homeostasis and maintenance of arterial pressure w6x. It is noticeable that peripheral and central levels of neurohypophysial hormones are regulated independently and have different physiological roles w2,45x. Magnocellular neurons in the SON, PVN, NC, and rSON terminate at the posterior pituitary to release the hormones into portal blood vessels

0006-8993r00r$ - see front matter q 2000 Elsevier Science B.V. All rights reserved. PII: S 0 0 0 6 - 8 9 9 3 Ž 9 9 . 0 2 4 1 8 - X

10

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

w51,53x. However, parvocellular neurons in the PVN ŽparPVN. extend their axonal terminals to various brain regions to control ACTH secretion, or autonomic responses w51x. In addition to the physiological functions, OXT and AVP might be involved in acute-phase and febrile responses. Intravenous Ži.v.. and intraperitoneal Ži.p.. administration of LPS have been shown to increase plasma levels of OXT and AVP w16,21–24x. Administration of IL-1b also has been shown to evoke the release of both neurohypophysial hormones w37x. Intracerebroventricular injection of IL-1b also increases systemic OXT and AVP, which are accompanied by a transient rise in OXT and AVP release from the SON w28x. IL-1b is further shown to induce secretion of OXT and AVP from the isolated rat neurohypophysis w5x. It is also shown that IL-1b directly depolar-

izes resting membrane potentials in most SON neurons in vitro w30x. Peripheral administration of VI-receptor antagonist has been shown to attenuate the rise in body temperature with pyrogen administration w34x, and AVP increases body temperature by peripheral vasoconstriction w14x. These results suggest that peripheral OXT and AVP have some physiological significance with regard to acute-phase andror febrile responses. In addition to peripheral functions, OXT and AVP have been shown to have important roles in controlling febrile responses. Microdialysis and radioimmunoassay demonstrate that OXT and AVP levels in the ventral septal area ŽVSA. are increased by systemic administration of IL-1b w27,28,59,60x. AVP is well known to have an antipyretic action w25x. Electrical stimulation of the PVN or bed nucleus of the stria terminalis ŽBST. suppresses fever,

Fig. 1. Light micrographs showing immunohistochemical dual-labeling for Fos ŽA and C. and either OXT ŽB. or AVP ŽD. in the SON of rats, which were fixed 2 h after i.p. injection of 125 mgrkg LPS. Arrows indicate representative neurons that are positive for both Fos and OXT ŽA and B. or Fos and AVP ŽC and D.. Scale bar s 100 mm.

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

indicating the projections of AVP neurons to the VSA w48,58x. OXT has an antipyretic action for an endotoxin-induced fever, possibly via modulation of the endogenous AVP antipyretic mechanism w36,43x. In contrast, microinjection of oxytocin into the preoptic area or cerebral ventricle has been reported to induce hyperthermia, possibly by increasing prostaglandin synthesis w11,31,33x.

11

The c-fos is one of the immediate early genes, and its protein synthesis is rapidly increased in the neuron, which is stimulated by metabolic activation w35,50x. Therefore, c-fos has been used as a marker of metabolic activation within brain neurons w18x. Evidence has accumulated that systemic administration of LPS or IL-1b causes a profound expression of c-fos genes or protein in many brain

Fig. 2. Light micrographs showing immunohistochemical dual-labeling for Fos ŽA and C. and either OXT ŽB. or AVP ŽD. in the rostral magnocellular division of the PVN ŽmagPVN. of rats, which were fixed 2 h after i.p. injection of 125 mgrkg LPS. Arrows indicate representative neurons that are positive for both Fos and OXT ŽA and B. or Fos and AVP ŽC and D.. Intense Fos immunoreactivity was observed in the medial division of the parPVN Žarrowheads., where the size of Fos-positive nuclei is smaller than those in the magPVN Žarrows.. Scale bar s 100 mm.

12

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

regions, suggesting that a complex neuronal circuit is involved in triggering the activity of neuroendocrinologi-

cal, immunological and autonomic responses w7–10, 17,26,40,46,49,54,57x. However, LPS-induced Fos expres-

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

sion in OXT and AVP neurons is not fully understood, although a few preliminary studies have found that LPS or IL-1b induces fos gene or protein in OXT and AVP neurons w46,49x. Therefore, the present study was designed to investigate the effects of LPS administration on Fos expression in immunologically identified OXT and AVP neurons among the rat hypothalamus. The experiments were performed with quantitative dual-labeling of Fos and neuropeptides and designed to clarify the following: Ž1. regional differences of Fos expression among the SON, magPVN, parPVN, NC, and rSON; Ž2. differences of Fos expression between OXT and AVP neurons; Ž3. differences of Fos expression with administration routes between via i.p.- and i.v.-injection.

2. Materials and methods 2.1. Animals Wistar male rats Ž9–10 weeks old. were used in the present experiments. The rats were housed under temperature-controlled Ž24 " 18C., light-controlled Ž14:10 h light:dark cycle. conditions, with food and water available ad libitum. A total of 37 rats were used in this experiment Ži.p.-injection, n s 24; i.v.-injection, n s 13.. For i.v.-injection of LPS, the rats were anesthetized with Nembutal Ž50 mgrkg. and a catheter Žplastic tube, OD s 1 mm, ATOM, Tokyo, Japan. was implanted into left jugular vein. The free end of the catheter was passed under the skin, exteriorized at the level of neck, and plugged with a sterile wire stylet. The potency of the catheter was maintained by filling with heparinized saline. The rats were housed individually in cages for at least 1 week after surgery. All experimental protocols were performed according to the guideline for animal care of the Japan Society of Physiology.

13

fever occurred after a transient body temperature decrease w8,9x. The rats were deeply anesthetized with Nembutal, and were perfused with 4% paraformaldehyde ŽPFA. in 0.1 M phosphate buffer 2, 4, 6 and 24 h after LPS injection. The brains were dissected out and postfixed with 4% PFA in 0.1 M phosphate buffer for 2 days at 48C. Fixed brains were immersed in 25% sucrose in phosphate-buffered saline ŽPBS. for cryoprotectant. 2.3. Immunohistochemistry Frozen sections were cut with a cryostat ŽCRYO CUT II, American Optical, Buffalo, USA., at a thickness of 20 mm. Free-floating sections were pretreated with 1% H 2 O 2 in PBS for 20 min, treated with 5% normal goat serum in PBS containing 0.3% Triton X-100 ŽPBST. overnight at 48C, and incubated with polyclonal anti-Fos antibody ŽSanta Cruz Biotechnology, Santa Cruz, CA. diluted 1:6000 in PBST containing 1% normal goat serum for 2 days at 48C. After incubation with the primary rabbit antibody, the sections were incubated with biotinylated goat anti-rabbit IgG Ždilution 1:200, Vector Labs, Burlingame, CA. for 2 h, followed by ABC elite kit solution Ždilution 1:300, Vector Lags. for 2 h. Visualization of antibodies was performed with 0.02% 3,3X diaminobenzidine ŽDAB. and 0.01% H 2 O 2 in 0.05 M Tris HCl buffer ŽpH 7.4.. After the DAB reaction, the sections were preincubated with 5% normal goat serum in PBST for 1 h and incubated for 2 days at 48C with either anti-OXT-neurophysin ŽPS 38. or anti-AVP-neurophysin ŽPS 41, both diluted 1:20. antibodies, which were generously provided by Dr. Gainer w3x. The sections were then incubated with goat anti-mouse IgG conjugated with fluorescein isothiocyanate ŽFITC; Kirkegaard and Perry Labs, Gaithersburg, USA, dilution 1:50. for 2 h, mounted on gelatin-coated glass slide, and sealed with fluorescence mounting medium, Vectashild ŽVector Labs.. 2.4. QuantitatiÕe analysis

2.2. LPS treatment Rats were injected with LPS from E. coli Ž055:type B5, DIFCO, Detroit, USA. or pyrogen-free saline. Each animal was given one of two doses of LPS Ž5 or 125 mgrkg body weight. via either i.p.- or i.v.-injection. The injection dose of LPS was determined by referring to the previous studies. Injection of 5 mgrkg of LPS causes body temperature elevation, and when 125 mgrkg of LPS was injected,

Sections containing the SON, PVN, NC and rSON were observed with an IMT-2 fluorescence microscope ŽOlympus, Tokyo, Japan.. These regions were identified with the rat brain stereotaxic atlas w42x. Magnocellular neurons in the PVN ŽmagPVN. and parPVN were identified according to the previous study w51x, and the posterior magnocellular divisions and the lateral parvocellular divisions were chosen for analysis of the percentage of Fos-positive nu-

Fig. 3. Light micrographs showing immunohistochemical dual-labeling for Fos ŽA, C, E, I, G and K. and either OXT ŽB, F and H. or AVP ŽD, J and L. in the lateral division of the parPVN ŽA–D., NC ŽE–H. and rSON ŽI–L. of rats, which were fixed 2 h after i.p. injection of 125 mgrkg LPS. In the lateral division of the parPVN, Fos immunoreactivity was not observed in OXT-positive neurons ŽA and B., but a number of Fos-positive nuclei were observed in AVP-positive neurons ŽC and D, arrows.. In the NC, many Fos-positive nuclei were observed in OXT-positive neurons ŽE and F, arrows. or AVP-positive neurons ŽG and H, arrows.. In the rSON, Fos-positive nuclei were observed in OXT-positive neurons ŽI and J, arrows. or AVP-positive neurons ŽK and L, arrows.. Scale bar s 50 mm.

14

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

clei in the magPVN and parPVN, respectively. The quantitative analysis was performed on at least four sections in each animal. The percentage of Fos-positive nuclei given for OXT and AVP neurons was the group mean " SE derived from animal means. The percentage of Fos-positive nuclei in the OXT- or AVP-positive neurons in each brain region was compared between the control and LPS-injected groups, between OXT- and AVP-positive neuron groups, or between i.p.and i.v.-injected groups with a one-way analysis of variance ŽANOVA. and Fisher’s test. The null hypothesis was rejected at the 5% level confidence.

3. Results We performed dual-labeling immunohistochemistry with Fos and either OXT or AVP in the hypothalamus of LPSand saline-injected rats. Rats were given either high-dose Ž125 mgrkg b.w.. or low-dose LPS Ž5 mgrkg b.w.. via i.p.- or i.v.-injection. Fig. 1 shows the Fos immunoreactivity in OXT- and AVP-positive neurons of the SON in high-dose LPS-injected rats. Many Fos-positive nuclei were observed in OXT- ŽFig. 1A, B. and AVP-positive neurons ŽFig. 1C, D.. Fos immunoreactivity was also seen in the magPVN of high-dose LPS-injected rats and a large number of OXTŽFig. 2A, B. and AVP-positive neurons ŽFig. 2C, D. contained Fos immunoreactivity in their nuclei. There were many Fos-positive nuclei in the medial division in the parPVN of high-dose LPS-injected rats ŽFig. 2A, C.. Fos immunoreactivity was not frequently observed in OXTpositive neurons of the lateral division in the parPVN ŽFig. 3A and B., but some AVP-positive neurons contained Fos immunoreactivity in the lateral parPVN ŽFig. 3C and D.. In the NC ŽFig. 3E–H. and rSON, ŽFig. 3J–L., a number of Fos-positive nuclei were observed in OXT- and AVPpositive neurons. Fig. 4 shows the time course of the percentages of Fos-positive nuclei of AVP- and OXT-positive neurons of the SON ŽFig. 4A., magPVN ŽFig. 4B. and lateral division in the parPVN ŽFig. 4C. of high-dose LPS-injected rats. The percentages of Fos-positive nuclei in both AVP- and OXT-positive neurons reached a maximum level at 2 h after LPS injection. Therefore, rats were anesthetized and fixed 2 h after LPS or saline injection and used for the subsequent quantitative analysis. The quantitative analysis was performed to obtain the percentage of Fos-positive nuclei in OXT- and AVP-positive neurons as shown in Fig. 5: to compare Fos-positive percentage among the SON, magPVN, lateral division in the parPVN, NC and rSON Žbrain regions.; between via i.p.- and i.v.-injection Žinjection route.; and between OXT and AVP neurons Žneuronal phenotype.. The results of the quantitative analysis showed that: Ž1. The percentage of Fos-positive nuclei in LPS-injected groups was signifi-

Fig. 4. Time course change in the percentages of Fos-positive nuclei in AVP-positive Ž`. and OXT-positive neurons Žv . in the SON ŽA., magPVN ŽB., lateral division of the parPVN ŽC. after i.p. injection of 125 mgrkg LPS. The percentage of Fos-positive neurons reached a maximum level at 2 h after LPS injection. Data are mean"S.E. of four rats. U p- 0.05, UU p- 0.01 from 0 point control.

cantly higher than that in control groups among all observed regions ŽFig. 5A–E.. Ž2. The percentage of Fospositive nuclei of OXT magnocellular neurons was significantly or tended to be higher than that of AVP magnocellular neurons in the SON, magPVN, NC, and rSON ŽFig. 5A, B, D, E.. Ž3. The percentage of Fos-positive nuclei in

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

15

Fig. 5. Quantitative analysis of the percentage of Fos-positive nuclei in AVP-positive or OXT-positive neurons in the SON ŽA., magPVN ŽB., parPVN ŽC., NC ŽD. and rSON ŽE.. Rats were received with saline Žcontrol., 5 mgrkg of LPS ŽLPS5. or 125 mgrkg of LPS ŽLPS125., and were killed 2 h after injection. Data are mean " S.E. of four to five rats. UU p - 0.01 from control; ap - 0.05 or aap - 0.01 between AVP- and OXT-positive neurons.

AVP parvocellular neurons was significantly higher than that of OXT parvocellular neurons in high-dose groups ŽFig. 5C.. Ž4. Regional heterogeneity of Fos-positive percentage was observed in AVP magnocellular neurons of i.p.-injected groups ŽFig. 5A, B, D, E.. The percentages of

Fos-positive nuclei in AVP magnocellular neurons of the magPVN Ž23.91 " 7.40%. and NC Ž13.59 " 2.85%. were significantly lower Ž p - 0.01. than those in the SON Ž50.39 " 5.98%. and rSON Ž58.42 " 8.72%. in high-dose groups ŽFig. 5A, B, D, E.. Ž5. However, the percentage of

16

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

Fos-positive nuclei in magnocellular neurons of i.v.-injected groups was not significantly different among the SON, magPVN, NC, and rSON ŽFig. 5A, B, D, E.: 49.20–57.24% of AVP neurons and 60.74–70.25% of OXT neurons were Fos-positive in high-dose groups, and 5.19–11.82% of AVP neurons and 36.14–40.22% of OXT neurons were Fos-positive in low-dose groups. The Fospositive percentage of OXT neurons in high-dose i.p.-injected groups was 68.66–83.65% and did not significantly differ among the SON, magPVN, NC, and rSON.

4. Discussion It has been well demonstrated that Fos is induced by various physiological stimuli and that Fos immunostaining is useful for the study of neural metabolic activation w18,50x. A lot of evidence has accumulated to suggest that LPS induces Fos expression in many brain regions including the hypothalamus, thalamus, and brain stem w7– 10,17,26,40,46,49,54,57x. However, the LPS-induced Fos expression in hypothalamic OXT and AVP neurons is not fully understood, although a few preliminary studies have reported that LPS induces Fos expression in OXT and AVP neurons w46,49x. In the present experiment, we found that administration of low-dose Ž5 mgrkg. and high-dose Ž125 mgrkg. LPS induced intense nuclear Fos immunoreactivity in many OXT and AVP neurons. The percentage of Fos-positive nuclei in OXT magnocellular neurons was higher than that in AVP magnocellular neurons of the SON, magPVN, NC and rSON, whose axons terminate at the posterior pituitary, suggesting that LPS stimulates magnocellular OXT and AVP neurons in these regions for peripheral release. However, the percentage of Fos-positive nuclei in AVP parvocellular neurons of the parPVN was higher than that of OXT parvocellular neurons. The central release of those hormones might be related to the activation of parvocellular neurons in the PVN. Moreover, the percentages of Fos-positive nuclei in AVP magnocellular neurons of the SON and rSON were significantly higher than those of the magPVN and NC when animals were given LPS via i.p.-injection. This regional heterogeneity was not found in OXT neurons of i.p.-injected rats or in either OXT or AVP neurons of i.v.-injected rats. Therefore, the activation of AVP magnocellular neurons by i.p.-injection is heterogeneous among the four hypothalamic regions, but that of OXT magnocellular neurons is homogenous among these brain regions. In the present experiment, LPS-induced Fos immunoreactivity was observed in many OXT and AVP magnocellular neurons of the SON, magPVN, NC, and rSON. The percentage of Fos-positive nuclei in OXT magnocellular neurons was significantly higher or tended to be higher than that in AVP magnocellular neurons among these four regions. Magnocellular neurons are known to localize in

the SON, magPVN, NC, and rSON and their axonal terminals project to the posterior pituitary for neurohypophysial hormone release w53x. Therefore, the present results are in agreement with the evidence that elevation of plasma level of OXT and AVP is observed with LPS or pyrogen administration w16,21–24,27,46x, and OXT- and AVP-mRNA increases in the hypothalamus w55x. IL-1b evokes OXT and AVP release in electrically stimulated rat neurohypophysis in vitro w5x and administration of IL-1b evokes the release of OXT and AVP w37x. The Fos expression in OXT and AVP magnocellular neurons in our studies may be related to IL-1b mediated OXT and AVP secretion. Taken together, the present study reveals that peripheral administration of LPS excites OXT and AVP magnocellular neurons in the SON, magPVN, NC, and rSON, leading to release of neurohypophysial hormones into the circulation. The present experiment showed that LPS-induced Fos expression in OXT and AVP magnocellular neurons of hypothalamic regions, although the functional significance of peripheral OXT and AVP is not clearly elucidated. There were two possible explanations for the physiological significance of OXT during acute-phase responses to LPS administration. First, OXT promotes insulin and glucagon secretion w1,20,29,56x. Second, intramuscular or intravenous injection of OXT attenuates LPS-induced fever w36x, and OXT treatment is shown to reduce endogenous pyrogen formation in cultured blood mononuclei w36x. Taken together, we may assume that the increase of systemic OXT results in promotion of insulin and glucagon secretion, antipyretic action andror inhibition of excess production of cytokines. Endotoxin administration causes elevation of systemic AVP w16,21–24x, and systemic injection of AVP increases body temperature w14x. Therefore, AVP may contract blood vessels and cause reduction of heat loss during fever w6,41x. In contrast to magnocellular neurons, the percentage of Fos-positive nuclei in AVP neurons of the lateral division in the parPVN was significantly higher than that in OXT parvocellular neurons. It is well known that AVP has an antipyretic action via the V1-receptor in the central nervous system, and central injection of AVP causes a reduction in the febrile response but no effect on normal body temperature w12,38x. The ventral septal area ŽVSA. contains V1-receptors and is responsible for the antipyretic action of AVP w22x. AVP neurons in the BST that project to the VSA have also been shown to participate in antipyretic action, and AVP is released into the VSA during fever w27,28,60x. The parvocellular neurons of the lateral division in the parPVN project into many brain areas including the septal area w51x, and electrical stimulation of the parPVN has been demonstrated to attenuate pyrogeninduced fever w48x. In this experiment, it was also found that Fos was induced in the medial division of the parPVN, which influences the release of ACTH from the anterior lobe w51x.

W. Matsunaga et al.r Brain Research 858 (2000) 9–18

In this experiment, the percentages of Fos-positive AVP magnocellular neurons in the SON and rSON of i.p.-injected groups were significantly higher than those of the magPVN and NC. These results are consistent with the previous studies that AVP neurons in the PVN showed lower Fos expression than that of the SON in hemorrhaging or lactating rats w32,47x. Taken together, the present results suggest that regional differences of Fos expression are observed exclusively in AVP magnocellular neurons but not in OXT magnocellular neurons on the basis of various physiological stimulations. Signal communication between the magnocellular nuclei is observed in OXT neurons, but not in AVP neurons w39x. Thus, it is probable that excitation of OXT magnocellular neurons in several hypothalamic nuclei are coordinately regulated by each other, but that of AVP neurons is not, and the excitation threshold level of AVP magnocellular neurons in the SON is lower than that of the magPVN. The percentage of Fos-positive nuclei of AVP and OXT magnocellular neurons was not significantly different among the SON, magPVN, rSON and NC of i.v.-injected groups. This result indicates that there is heterogeneity of LPS-induced Fos expression between i.p.- and i.v.-injection. The heterogeneity of Fos expression is thought to be due to the following reasons. The first possibility may be due to the difference of signal transmission pathway. The i.p.-injected LPS stimulates the subdiaphragic vagal afferent nerve, but i.v.-injected LPS is transported to whole body by the blood flow and can contact to the afferent terminals mainly in heart or lung w52x. Therefore, hypothalamic neurons receive afferent inputs from multiple parts of the body following LPS i.v.-injection. The second possibility is the difference of cytokine production by injection route. Cytokines are generally not detectable in plasma until at least 45 min after i.v. LPS injection w15,19x and 90 min after i.p.-injection w13x. This suggests that immunological stimulation of cytokines by i.p.-injected LPS was weaker than that with i.v.-injected LPS. Finally, i.p.-injected LPS is exposed to various degradative factors before it has access to the vagal nerve terminals. Therefore, the stimulation by i.p.-injected LPS becomes weaker than that of i.v.-injected LPS, and as a result, in the magPVN, the percentage of Fos-positive nuclei in AVP neurons of the i.p.-injected group is lower than that of the i.v.-injected group.

Acknowledgements This work was supported in part by grants for scientific research from the Japan Society for the Promotion of Science ŽNo. 11640663 to S.M.. and The Ministry of Education, Science, Sports and Culture of Japan ŽNo. 10470016 to T.K... We are grateful to Dr. H. Gainer for generous supplies of monoclonal antisera against oxytocinand vasopressin-neurophysin.

17

References w1x N. Altszuler, B. Winkler, C.R. Rosenberg, F.X. Pi-Sunyer, A.R. Fuchs, Role of insulin and glucagon in oxytocin effects on glucose metabolism, Proc. Soc. Exp. Biol. Med. 199 Ž1992. 236–242. w2x A. Argiolas, G.L. Gessa, Central functions of oxytocin, Neurosci. Biobehav. Rev. 15 Ž1991. 217–231. w3x Y. Ben-barak, J.T. Russell, M.H. Whitnall, K. Ozato, H. Gainer, Neurophysin in the hypothalamo–neurohypophysial system: I. Production and characterization of monoclonal antibodies, J. Neurosci. 5 Ž1985. 81–97. w4x T.-Y. Chen, M.-G. Lei, T. Suzuki, D.C. Morrison, Lipopolysaccharide receptors and signal transduction pathways in mononuclear phagocytes, Curr. Top. Microbiol. Immunol. 181 Ž1992. 169–188. w5x J.D. Christensen, E.W. Hansen, B. Fjalland, Influence of interleukin1 beta on the secretion of oxytocin and vasopressin from the isolated rat neurohypophysis, Pharmacol. Toxicol. 67 Ž1990. 81–83. w6x E.T. Cunningham, P.E. Sawchenko, Reflex control of magnocellular vasopressin and oxytocin secretion, Trends Neurosci. 14 Ž1991. 406–411. w7x J.K. Elmquist, M.R. Ackermann, K.B. Register, R.B. Rimler, L.R. Ross, C.D. Jacobson, Induction of Fos-like immmunoreactivity in the rat brain following Pasteurella multocida endotoxin administration, Endocrinology 133 Ž1993. 3054–3057. w8x J.K. Elmquist, C.B. Saper, Activation of neurons projecting to the paraventricular hypothalamic nucleus by intravenous lipopolysaccharide, J. Comp. Neurol. 374 Ž1996. 315–331. w9x J.K. Elmquist, T.E. Scammell, C.D. Jacobson, C.B. Saper, Distribution of Fos-like immunoreactivity in the rat brain following intraveneous lipopolysaccharide administration, J. Comp. Neurol. 371 Ž1996. 85–103. w10x A. Ericsson, K.J. Kovacs, P.E. Sawchenko, A functional anatomical analysis of central pathways subserving the effects of interleukin-1 on stress-related neuroendocrine neurons, J. Neurosci. 14 Ž1994. 897–913. w11x S.E. Fahrbach, J.I. Morrell, J.I. Pfaff, Possible role for endogenous oxytocin in estrogen-facilitated maternal behavior in rats, Neuroendocrinology 40 Ž1985. 526–532. w12x P. Federico, W.L. Veale, Q.J. Pittman, Vasopressin-induced antipyresis in the medial amygdaloid nucleus of conscious rats, Am. J. Physiol. 262 Ž1992. 901–908. w13x K.D. Foster, C.A. Conn, M.J. Kluger, Fever, tumor necrosis factor and interleukin-6 in young, mature and aged, Fischer 344 rats, Am. J. Physiol. 262 Ž1992. 211–215. w14x I.T. Gagalo, E.E. Hac, M.T. Matuszek, P. Rekowski, G. Kupryszewski, K.Z. Korolkiewicz, Thermoregulatory activity of sodium nitroprusside and arginine vasopressin, Gen. Pharmacol. 26 Ž1995. 393–397. w15x L. Givalois, J. Dornand, M. Mekaouche, M.D. Solier, A.F. Bristow, G. Ixart, P. Siaud, I. Assenmacher, G. Barbanel, Temporal cascade of plasma level surges in ACTH, corticosterone, and cytokines in endotoxin-challenged rats, Am. J. Physiol. 266 Ž1994. 164–170. w16x E.W. Hansen, J.D. Christensen, Endotoxin and interleukin-1b induces fever and increased plasma oxytocin in rabbits, Pharmacol. Toxicol. 70 Ž1992. 389–391. w17x A.S. Hare, G. Clarke, S. Tolchard, Bacterial lipopolysaccharideinduced changes in Fos protein expression in the rat brain: correlation with thermoregulatory changes and plasma corticosterone, J. Neuroendocrinol. 7 Ž1995. 791–799. w18x G.E. Hoffman, M.S. Smith, J.G. Verbalis, c-Fos and related immediate early gene products as markers of activity in neuroendocrine systems, Front. Neuroendocrinol. Ž1993. 173–213. w19x L. Jansky, S. Vybiral, D. Pospisilova, J. Roth, J. Dornand, E. Zeisberger, J. Kaminkova, Production of systemic and hypothalamic cytokines during the early phase endotoxin fever, Neuroendocrinology 62 Ž1995. 55–61. w20x M.V. Kaminski Jr., H.A. Neufeld, J.G. Pace, Effect of inflammatory

18

w21x

w22x

w23x

w24x

w25x

w26x

w27x

w28x

w29x w30x

w31x

w32x

w33x w34x

w35x

w36x

w37x

w38x

w39x

w40x

W. Matsunaga et al.r Brain Research 858 (2000) 9–18 and noninflammatory stress on plasma ketone bodies and free fatty acids and on glucagon and insulin in peripheral and portal blood, Inflammation 3 Ž1979. 289–294. N.W. Kasting, Indomethacin an antipyretic drug, prevent the endotoxin-induced and potentiates the hemorrhage-induced oxytocin release into the plasma of the male rat, Can. J. Physiol. Pharmacol. 42 Ž1986. 285–288. N.W. Kasting, Characterization of body temperature, vasopressin, and oxytocin responses to endotoxin in the rat, Can. J. Physiol. Pharmacol. 64 Ž1986. 1575–1578. N.W. Kasting, M.F. Mazurek, J.B. Martin, Endotoxin increases vasopressin release independently of known physiological stimuli, Am. J. Physiol. 248 Ž1985. E420–424. N.W. Kasting, D.B. Carr, J.B. Martin, H. Blume, R. Bergland, Changes in cerebrospinal fluid and plasma vasopressin in the febrile sheep, Can. J. Physiol. Pharmacol. 61 Ž1983. 427–431. G.L. Kovacs, D. De Wied, Hormonally active arginine-vasopressin suppresses endotoxin-induced fever in rats: lack of effect of oxytocin and a behaviorally active vasopressin fragment, Neuroendocrinology 37 Ž1983. 258–261. S. Lacroix, S. Rivest, Functional circuitry in the brain of immunechallenged rats:partial involvement of prostaglandins, J. Comp. Neurol. 387 Ž1997. 307–324. R. Landgraf, T.J. Malkinson, W.L. Veale, K. Lederis, Q.J. Pittman, Vasopressin and oxytocin in rat brain in response to prostaglandin fever, Am. J. Physiol. 259 Ž1990. R1056–1062. R. Landgraf, I. Neumann, F. Holboer, Q.J. Pittman, Interleukin-1 beta stimulates both central and peripheral release of vasopressin and oxytocin in the rat, Eur. J. Neurosci. 7 Ž1995. 592–598. C.H. Lang, C. Debrescu, Interleukin-1 induced increases in glucose utilization are insulin mediated, Life Sci. 45 Ž1989. 2127–2134. Z. Li, K. Inenaga, H. Yamashita, Effects of interleukin-1 beta on membrane potentials of hypothalamic supraoptic neurons in rats in vitro, Ann. N.Y. Acad. Sci. 689 Ž1993. 689–692. M.T. Lin, L.Y. Ho, H.K. Chan, Effects of oxytocin and Ž1-penicillamine, 4-threonine. oxytocin on the thermoregulation in rats, Neuropharmacology 22 Ž1983. 1007–1013. S.H. Lin, S. Miyata, T. Itoh, T. Kawarabayashi, T. Nakashima, T. Kiyohara, Fos expression in the hypothalamic magnocellular neurons of rats during pregnancy, parturition and lactation, Neurosci. Res. 23 Ž1995. 29–34. J.M. Lipton, J.R. Glyn, Central administration of peptide alter thermoregulation in the rabbit, Peptides 1 Ž1980. 15–18. N.G. Milton, E.W. Hillhouse, A.S. Milton, Does endogenous peripheral arginine vasopressin have a role in the febrile responses of conscious rabbits?, J. Physiol. 469 Ž1993. 525–534. J.I. Morgan, T. Curran, Stimulus-transcription coupling in the nervous system: involvement of the inducible proto-oncogenes fos and jun, Annu. Rev. Neurosci. 14 Ž1991. 421–451. P.P. Murzenok, L.F. Kutasaeva, T.I. Zhitkevich, V.N. Gurin, Antipyrogenic properties of oxytocin, Bull. Exp. Biol. Med. 108 Ž1989. 1410–1412. Y. Naito, J. Fukata, K. Shindo, O. Ebisui, N. Murakami, T. Tominaga, Y. Nakai, K. Mori, N.W. Kasting, H. Imura, Effects of interleukins on plasma arginine vasopressin and oxytocin levels in conscious, freely moving rats, Biochem. Biophys. Res. Commun. 174 Ž1991. 1189–1195. A.M. Naylor, W.D. Ruwe, A.F. Kohut, W.L. Veale, Perfusion of vasopressin within the ventral septum of the rabbit suppresses endotoxin fever, Brain. Res. Bull. 15 Ž1985. 209–213. I. Neumann, R. Landgraf, L. Bauce, Q.J. Pittman, Osmotic responsiveness and cross talk involving oxytocin, but not vasopressin or amino acids, between the supraoptic nuclei in virgin and lactating rats, J. Neurosci. 15 Ž1995. 3408–3417. A. Oladehin, C.M. Blatteis, Lipopolysaccharide-induced Fos expres-

w41x

w42x w43x

w44x

w45x w46x

w47x

w48x

w49x

w50x

w51x

w52x

w53x

w54x

w55x

w56x

w57x

w58x

w59x

w60x

sion in hypothalamic nuclei of neonatal rats, Neuroimmunomodulation 2 Ž1995. 282–289. R.F. Parrott, S.V. Vellucci, J.A. Goode, D.M. Lloyd, M.L. Forsling, Interrelated adrenocortical and neurohypophysial responses associated with fever in endotoxin-treated pigs, Am. J. Physiol. 273 Ž1997. 1046–1052. G. Paxinos, C. Watson, The Rat Brain in Stereotaxic Coordinates, Academic Press, New York, 1997. P. Poulin, Q.J. Pittman, Possible involvement of brain oxytocin in modulating vasopressin antipyretic action, Am. J. Physiol. 265 Ž1993. R151–156. C.H. Rhodes, J.I. Morrel, D.W. Pfaff, Immunohistochemical analysis of magnocellular elements in rat hypothalamus: distribution and numbers of cells containing neurophysin, oxytocin and vasopressin, J. Comp. Neurol. 198 Ž1981. 45–61. P. Richard, F. Moos, M.-J. Freund-Mercier, Central effects of oxytocin, Physiol. Rev. 71 Ž1991. 331–370. S. Rivest, N. Laflamme, Neuronal activity and neuropeptide gene transcription in the brains of immune-challenged rats, J. Neuroendocrinol. 7 Ž1995. 501–525. M.M. Robert, A.G. Robinson, M.D. Fitzsimmons, F. Grant, W.S. Lee, G.E. Hoffman, c-fos expression in vasopressin and oxytocin neurons reveals functional heterogeneity within magnocellular neurons, Neuroendocrinology 57 Ž1993. 388–400. W.D. Ruwe, Electrical stimulation of the paraventricular nucleus attenuates pyrogen fever in the rabbit, Brain Res. 588 Ž1992. 181–190. S.M. Sagar, K.J. Price, N.W. Kasting, F. Sharp, Anatomic patterns of Fos immunostaining in rat brain following systemic endotoxin administration, Brain Res. Bull. 36 Ž1995. 381–392. S.M. Sagar, F.R. Sharp, T. Curran, Expression of c-fos protein in brain: metabolic mapping at the cellular level, Science 240 Ž1988. 1328–1331. P.E. Sawchenko, L.W. Swanson, Hypothalamic integration: organization of the paraventricular and supraoptic nuclei, Annu. Rev. Neurosci. 6 Ž1983. 269–324. E. Sehic, C.M. Blatteis, Blockade of lipopolysaccharide-induced fever by subdiaphragmatic vagotomy in gunia pig, Brain Res. 726 Ž1996. 160–166. D.F. Swaab, F. Nijveldt, C.W. Pool, Distribution of oxytocin and vasopressin in the rat supraoptic and paraventricular nucleus, J. Endocrinol. 67 Ž1975. 461–462. N. Tkacs, J. Li, A.M. Strack, Central amygdala Fos expression during hypotensive or febrile, nonhypotensive endotoxemia in conscious rats, J. Comp. Neurol. 379 Ž1997. 592–602. S.V. Vellucci, R.F. Parrott, Expression of mRNAs for vasopressin, oxytocin and corticotrophin releasing hormone in the hypothalamus and of cyclooxygenases-1 and -2 in the cerebral vasculature, of endotoxin-challenged pigs, Neuropeptides 32 Ž1998. 439–446. L.A. Wallin, C.P. Fawcett, C.R. Rosenfeld, Oxytocin stimulates glucagon and insulin secretion in fetal and neonatal sheep, Endocrinology 125 Ž1989. 2289–2296. W. Wan, L. Anz, C.Y. Vriend, C.M. Sorensen, A.H. Breenberg, D.M. Nance, Differential induction of c-fos imunoreactivity in hypothalamus and brain stem nuclei following central and peripheral administration of endotoxin, Brain Res. Bull. 32 Ž1993. 581–587. M.F. Wilkinson, T.F.W. Horn, N.W. Kasting, Q.J. Pittman, Central interleukin-1 stimulation of vasopressin release into the rat brain: activation of an antipyretic pathway, J. Physiol. 481 Ž1994. 641–646. M.F. Wilkinson, N.W. Kasting, Vasopressin release within the ventral septal area of the rat brain during drug-induced antipyresis, Am. J. Physiol. 264 Ž1993. R1133–1138. M.F. Wilkinson, Q.J. Pittman, Alteration of the physiological responses to indomethacin by endotoxin tolerance in the rat: a possible role for central vasopressin, J. Physiol. 479 Ž1994. 441–449.