Brain Research Bulletin 62 (2003) 173–178
Chronic psychosocial stress down-regulates central cytokines mRNA Alessandro Bartolomucci a,b,∗ , Paola Palanza a , Stefano Parmigiani a , Tiziana Pederzani a , Elodie Merlot c , Pierre J. Neveu c , Robert Dantzer c a
Dipartimento di Biologia Evolutiva e Funzionale, Università di Parma, Parco area delle Scienze 11/A, 43100 Parma, Italy b Istituto di Psicologia, Università di Milano, Milano, Italy c Laboratory of Integrative Neurobiology, U394, INRA-INSERM, Bordeaux, France Received 23 January 2003; received in revised form 7 August 2003; accepted 17 September 2003
Abstract Brain cytokines have been implicated in brain plasticity and mood alterations. We present here the first evidence of a chronic stress-induced modulation of central cytokines, in absence of experimentally induced inflammatory processes. Several brain areas were extracted from stressed and control mice and cytokines mRNA analyzed with semi-quantitative RT-PCR. Mice subjected to chronic psychosocial stress showed decreased interleukin (IL)-1 mRNA levels in the hippocampus, decreased IL-1Receptor antagonist in the striatum and pituitary, decreased tumor necrosis factor (TNF)-␣ in the striatum and hippocampus, and decreased glucocorticoid receptor (GR) in the striatum and hippocampus compared to group housed sibling mice. An independent group of mice subjected to chronic psychosocial stress also showed increased plasma corticosterone. These findings may open new perspectives for understanding the pathophysiological basis of chronic stress-induced disorders. © 2003 Elsevier Inc. All rights reserved. Keywords: Hippocampus; Striatum; TNF-␣; IL-1; Corticosterone; Glucocorticoid receptors
1. Introduction Cytokines are potent, multifunctional, pleiotropic proteins that have been initially characterized in the context of cellular activation and cellular communication in the immune system. The first direct evidence of the presence and activity of cytokines in the brain was obtained two decades ago and regarded interleukin (IL)-1 in humans and tumor necrosis factor (TNF)-␣ in the rat brain [8,35]. Cytokines are present in several brain structures and cellular types, e.g. glia, neurons, and macrophages, in form of proteins and their receptors [15,29,30,34,35]. Besides their well characterized role in mediating the central components of the host innate immune response, convergent findings point to a direct or indirect role for cytokines in stress-related disorders, including depression [10,19]: (i) cytokines administered to patients and laboratory animals induce symptoms of depression [9,17]; (ii) exposure to acute stressors can induce the expression of cytokines at the periphery and in the brain (e.g. [26]); (iii) somatic disorders with an inflammatory component are as∗ Corresponding author. Tel.: +39-0521-905636; fax: +39-0521-905657. E-mail address:
[email protected] (A. Bartolomucci).
0361-9230/$ – see front matter © 2003 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2003.09.009
sociated with a high prevalence of depressive disorders and cytokines imbalance [21]. Because of their potent activatory effect on the HPA axis, proinflammatory cytokines could contribute to the elevated levels of glucocorticoids that are observed in depressed patients [16,21]. Reciprocally, activation of glucocorticoid receptors leads to down regulation of peripheral and central expression of proinflammatory cytokines, via modulation of AP-1 and NFkB transcription [36]. Accordingly, it has been suggested that glucocorticoids and central/peripheral cytokines are finely balanced to maintain homeostasis with reciprocal inhibitory/activatory pathways being present [8,29,36]. This complex balance appears critical and, if not maintained, it can have clinical consequences in the form for instance of neuronal death or reduced brain plasticity [1,19,21,31]. Conventional models of stress in laboratory rodents, however, show little face validity with the chronic human psychopathologies. Models of social stress based on the naturalistic events of interactions between conspecifics offer a realistic tool to induce chronic stress in animal models [18]. In recent years we developed a naturalistic and ethologically-oriented model of chronic psychosocial stress, which is based on a natural behavior of male mice, i.e. acquiring and defending a territory. When forcing two adult males in a chronic sensory contact and
174
A. Bartolomucci et al. / Brain Research Bulletin 62 (2003) 173–178
with a brief daily agonistic encounter, mice seem to live under a constant threat and activation [2,3,5,6]. From previous investigations it becomes evident that social status and territory ownership are factors determining the vulnerability to chronic stress exposure. This seem true for behavioral, autonomic, metabolic and immune functions, but not for the regulation of the HPA axis. Indeed all animals under chronic stress showed high corticosterone both in basal condition and after dexamethasone challenge. Given the key regulatory role of corticosterone on cytokines expression [36], our aim was to investigate whether subjecting male mice to such a naturalistic stressor would result in a modulation of the central cytokines network (IL-1, IL-1Ra, TNF-␣, IL-6, IL-10) and glucocorticoids receptors (GR) in basal conditions (e.g. [37]), i.e. in absence of any further acute stressor such as restraint stress or lypopolisaccaride (LPS) injection (e.g. [26]).
(G) are same age 3-sibling-group-housed male mice (see [3,4] for further details). These animals were re-housed in group of 3 (from pre-existing groups of 4–7 animals per cage as described above) the same day in which the chronic stress procedure started and were treated exactly as the animals under stress as far as environmental factors are considered. Within each group, the hierarchical status of the animals was determined during three 60-min periods of continuous observation on the first day and in the days in which the bedding was changed (days 7 and 14). We identified the mouse chasing and biting all the others as dominant, and the animals displaying upright posture, flight behavior and squeaking vocalization as the subordinates. Only the dominant and one of the two subordinate mice (randomly chosen) were used for experimental measurements. Two independent studies were performed, one investigating the central mRNA levels (controls: n = 10, stress: n = 24) and the other the plasma levels of corticosterone (controls: n = 14, stress: n = 28).
2. Materials and methods 2.1. Animals The subjects were adult, males Swiss CD-1 mice originally obtained from Charles River Italia (Calco, Italy). The mice were born and reared at 22 ± 2 ◦ C, with lights ‘on’ at 7.00 and ‘off’ at 19.00. After weaning they were housed in same sex-sibling groups (4–7 per group). Food and water were available ad libitum. Experiment was conducted in accordance with the European Communities Council Directive of 24 November 1986 (86/609/EEC) and approved by Italian Institute of Health. 2.2. Procedure The procedure was scheduled according to Bartolomucci et al. [3]. Briefly, an experimental animal was introduced as intruder in the cage of a resident male. After the interaction, the two animals were divided by means of a perforated polystyrene–metal partition allowing a continuous visual–olfactory–acoustic but not physical interaction. All the animals had food and water ad libitum. The perforated partition was removed daily (for a total of 21 days). The interaction lasted a maximum of 10 min, and to prevent injuries the interaction was interrupted as soon as fighting escalated. All the animals subjected to the above-mentioned procedure are included in the stress group (S). During the interaction the offensive and defensive behavior of the animals were recorded. The chasing and biting animal was defined as dominant, while the mouse displaying upright posture, flight behavior and squeaking vocalization was the subordinate. Under present conditions, four groups of animals were categorized: residents becoming dominants (RD), intruders becoming dominants (InD), residents becoming subordinates (RS) and intruders becoming subordinates (InS). Based on previous observations showing no sign (immuno-endocrine and behavioral) of stress in group house siblings, controls
2.3. Cytokines- and glucocorticoid receptor-encoding transcripts Mice were sacrificed on the morning (between 9:00 and 10:00 h) of the 22nd day. Brains were manually dissected on sterile Petri-dishes containing dry ice. Hippocampus, hypothalamus, striatum and pituitary were quickly removed, placed in mRNAase free tubes, and immediately frozen. Tubes were stored at −80 ◦ C until assayed. Tissues were homogenized in Tri Reagent (Sigma, St. Louis, MO, USA) and mRNA extraction was performed on supernatants. Reverse transcription (RT) was carried out in a final volume of 50 l from 2 g of denatured template RNA incubated at 42 ◦ C for 3 h in the presence of 2.5 M random hexamers, 64 IU recombinant RNAse inhibitor (Promega, France), 2 mM dNTP, 1× RT buffer and 500 IU Superscript II reverse transcriptase (Life Technologies, France). Seven pairs of specific primers were used for polymerase chain reaction (PCR) amplification of cDNA. PCR was performed in a 50 l final volume containing 5 l RT product, 1.5 mM MgCl2 , 50 M dNTP, 0.4 M specific sense and antisense primers, 1× Taq polymerase buffer, 2 IU Taq polymerase (Promega, France) and 1 Ci [␣-32P]dCTP (3000 Ci/mmol, Amersham, Buckinghamshire, UK). PCR amplification (Mastercycler, Eppendorf, Hamburg, Germany) consisted of 24–32 cycles of 1 min denaturation at 94 ◦ C, 1 min annealing at 60 or 62 ◦ C, 1 min of elongation at 72 ◦ C. The sequence, the length of amplicons, the temperatures of annealing and the number of cycles corresponding to each primer pair are described in Table 1. Fifteen microliters of each PCR product were separated by electrophoresis on a 10% arylamide/bisacrylamide gel. Fragment size corresponded to that predicted by the location of the primers. To confirm the specificity of the amplified fragments, PCR products were digested with specific restriction enzymes, leading to restriction fragments separated by gel electrophoresis whose length matched with the expected ones. Gels were exposed for 4 h on a radiosensitive phosphor
A. Bartolomucci et al. / Brain Research Bulletin 62 (2003) 173–178
175
Table 1 RT-PCR parameters: sequence, length of amplicons, temperatures of annealing, and number of cycles corresponding to each primer pair mRNA
Primers (5 –3 )
Length (pb)
Temperature (◦ C)
Cycles
2-Microglobulin
Up CATGAGTGAGCTACAGTGGGGAACA Down GGGCTCGGCCATACTGTCAT Up CCTGCAGCTGGAGAGTGTGGA Down CCCATCAGAGGCAAGGAGGAA Up CAGTTCCACCCTGGGAAGGT Down GAGCGGATGAAGGTAAAGCG Up CGGAGAGGAGACTTCACAGAGGA Down GGAGAGCATTGGAAATTGGGG Up ATGCAGGACTTTAAGGGTTACTTG Down TAGACACCTTGGTCTTGGAGCTTA Up GCGGTGCCTATGTCTCAGCC Down TGAGGAGCACGTAGTCGGGG Up CAGCCTGGCACCAACGGT Down GGCTTCTGATCCTGCTGCTGG
289
60
28
400
60
30
453
62
30
462
62
30
254
60
32
361
60
30
527
60
24
IL-1 IL-1Ra IL-6 IL-10 TNF-␣ GR
screen, and scanned by laser densitometry (200 m pixel; Phosphoimager SI, Molecular Dynamics). The radioactivity was quantified by peak surface analysis using ImageQuant software (Molecular Dynamics). Results were expressed as the ratio (%) of the radiolabeling incorporated in the specific PCR product to 2-microglobulin PCR product obtained from the same reverse transcription experiment. 2.4. Corticosterone assay Mice were sacrificed on the morning (between 9:00 and 10:00 h) of the 22nd day. Trunk blood was collected in heparinized tubes, centrifuged and frozen at −20 ◦ C until analyzed. The measurements were done in duplicate by a commercially available radioimmunoassay kit (Amersham, USA). The sensitivity was 0.06 ng per tube. The intra-assay coefficient of variation was 3.4%. In order to avoid possible problems with inter-assay variability, all samples were run in a single assay. 2.5. Data analysis Cytokines and GR were analyzed by means of two-way ANOVAs (between factor treatment (two levels) and within factor structures (four levels)). Four planned comparisons were a priori defined for each ANOVA, devised to compare stressed and control mice in each brain structure. Missing values were substituted using the mean value of the appropriate experimental group, thus results must be considered conservative for non-differences between the groups. Corticosterone level was compared with independent Student’s t-test.
InS. Since no difference emerged between the four categories, all the animals under stress were pooled and compared to control group-housed mice (IL-1 F(3, 20) = 0.6, NS; IL-1Ra F(3, 20) = 0.2, NS; IL-6 F(3, 15) = 0.3, NS; IL-10 F(3, 15) = 1.5, NS; TNF-␣ F(3, 20) = 0.1, NS; GR F(3, 20) = 0.8, NS). There was a clear alteration of central cytokines and GR mRNA content that was mainly evident in the striatum and hippocampus, but not in the hypothalamus (see Fig. 1A–F). The expression of IL-1 (ANOVA: treatment, F(1, 32) = 3.71, P = 0.062; structures F(3, 96) = 87.24, P < 0.0001) was reduced only in the hippocampus of the stressed group (P = 0.051) while that of its natural antagonist IL-1Ra (ANOVA: treatment, F(1, 32) = 7.7, P < 0.01; structures F(3, 96) = 7.54, P < 0.001) was reduced in the pituitary (P < 0.05) and striatum (P < 0.05). TNF-␣ expression (ANOVA: treatment, F(1, 32) = 2.66, P = 0.11; structures F(3, 96) = 21.76, P < 0.0001; treatment × structures, F(3, 96) = 4.18, P < 0.01) was clearly reduced in the striatum (P < 0.05) and hippocampus (P < 0.05) of chronically stressed mice. There was also a reduction in GR mRNA (ANOVA: treatment, F(1, 32) = 3.46, P = 0.071; structures F(3, 96) = 77.16, P < 0.0001; treatment × structures, F(3, 96) = 3.14, P < 0.05) in the striatum (P = 0.052) and hippocampus (P < 0.05). Finally, IL-6 was not affected by chronic stress and the same applied to IL-10. Plasma levels of corticosterone were increased in an independent group of chronically stressed animals compared to control group-housed siblings (t39 = 2.76; P < 0.01; control: 3.7 ± 0.8 ng/ml, stress: 15.3 ± 3.3 ng/ml).
4. Discussion 3. Results A preliminary analysis (two-way ANOVA for each variable, see below) was conducted to compare the four categories of mice under chronic stress, i.e. RD, RS, InD and
The results of the present study provide the first evidence of a modulation of the cytokines network by chronic social stress. In contrast, in a recent report rats intermittently exposed to a natural predator odors showed no change in a similar panel of central cytokines mRNA when compared
176
A. Bartolomucci et al. / Brain Research Bulletin 62 (2003) 173–178
IL-1 BETA
IL-1 receptor antagonist
Control Stress
1,4
60
1,2
#
1
40
IL- 1Ra/B2
IL-1beta/B2
50
30 20
* *
0,8 0,6 0,4
10
0,2
0
(A)
Pituitary
Striatum
Hypothalamus Hippocampus
0
(B)
Pituitary
Striatum
TNF-ALFA
IL-6 16
30
*
25
14
*
20
12 IL-6/B2
TNF ALFA/B2
Hypothalamus Hippocampus
15
10 8 6
10 4 5
2
0
0
(C)
Pituitary
Striatum
Hypothalamus Hippocampus
Pituitary
(D)
Striatum
Hypothalamus Hippocampus
GR receptors
IL- 10 200
3,5
#
3
160
*
GR/B2
IL-10/B2
2,5 2 1,5
120 80
1 40 0,5 0
0
(E)
Pituitary
Striatum
Hypothalamus Hippocampus
(F)
Pituitary
Striatum
Hypothalamus Hippocampus
Fig. 1. Cytokines and glucocorticoid receptor mRNAs are down-regulated under chronic psychosocial stress. (A) Interleukin-1 (IL-1), (B) IL-1Receptor antagonist (IL-1Ra), (C) interleukin-6 (IL-6), (D) tumor necrosis factor-␣ (TNF-␣), (E) interleukin-10 (IL-10), (F) glucocorticoid receptors (GR). Levels of cytokine and GR transcripts are expressed as the ratio (%) of the radiolabeling incorporated in the specific PCR product to 2-microglobulin PCR product obtained from the same reverse transcription experiment. ∗ P < 0.05; # P < 0.06.
with unexposed controls [28]. The implication of the present findings stands in the observation that under a naturalistic paradigm of chronic psychological challenge, and in the absence of any inflammatory stimulus due for instance to fighting injuries or LPS injection, the hippocampal and striatal cytokine network is altered. Glucocorticoids down reg-
ulate immune-mediated central cytokines mRNA expression [11–13]. GR have been similarly found to be down regulated under chronic stress and repeated LPS administration [14,24]. The increase in plasma corticosterone levels is therefore certainly responsible for both the down regulation of GR expression and cytokine transcripts in the brain
A. Bartolomucci et al. / Brain Research Bulletin 62 (2003) 173–178
[20,25,32]. Based on the present knowledge it is not possible yet to draw a definite conclusion on the functional role of a down-regulation of IL-1, IL-1Ra and TNF-␣ mRNA constitutive levels. However, recent evidences point to a direct role for central pro-inflammatory and anti-inflammatory cytokines as neuro-modulators [27,37]. In addition, Gemma et al. [11] showed serotoninergic activation to stimulate the HPA axis and to decrease hippocampal IL-1␣ and IL-1 mRNA transcript. These evidences, along with the involvement of these cytokines in synaptic plasticity [7] and long term potentiation in the hippocampus [33] together with the known impairment of hippocampal functions under conditions of chronic stress [22,23] point to the possibility of a pathophysiologic role of the down regulation of brain cytokines in decreased brain plasticity. Accordingly, several predictions follow our results. Hippocampal and striatal cells from the stressed mice should display altered synaptic functions [7]. Moreover, if increased circulating corticosterone is the key factor inducing a down-regulation of central cytokines, the removal of its main source (by adrenalectomy) and the maintenance of stable levels of corticosterone by a corticosterone clamp should prevent the decrease in cytokine mRNAs [13]. Further studies have to address these and related issues.
[7]
[8] [9]
[10]
[11]
[12]
[13]
[14]
[15]
Acknowledgements [16]
Supported by Italian MURST (COFIN 2000) and University of Parma (FIN 2000-01) and DGA (agreement no. 00.34.060.00.470.75.01). Authors wish to thank all the colleagues of the U394-INSERM of Bordeaux, and in particular J. Arsaut, G. Hebert, J.P. Konsman, E. Moze, for their kind help provided in several phases of the present study. V. Vascelli is acknowledged for technical assistance.
[17]
[18]
[19]
References [20] [1] W.P. Arend, The balance between IL-1 and IL-1Ra in disease, Cytokine Growth Factor Rev. 13 (2002) 323–340. [2] A. Bartolomucci, P. Palanza, T. Costoli, E. Savani, G. Laviola, S. Parmigiani, A. Sgoifo, Chronic psychosocial stress persistently alters autonomic function and physical activity in mice, Physiol. Behav. 80 (2003) 57–67. [3] A. Bartolomucci, P. Palanza, L. Gaspani, E. Limiroli, A.E. Panerai, G. Ceresini, M.D. Poli, S. Parmigiani, Social status in mice: behavioral, endocrine and immune change are context dependent, Physiol. Behav. 73 (2001) 401–410. [4] A. Bartolomucci, P. Palanza, P. Sacerdote, G. Ceresini, A. Chirieleison, A.E. Panerai, S. Parmigiani, Individual housing induces altered immuno-endocrine responses to psychological stress in male mice, Psychoneuroendocrinology 28 (2003) 540–558. [5] A. Bartolomucci, T. Peterzani, P. Sacerdote, A.E. Panerai, S. Parmigiani, P. Palanza, Behavioral and physiological characterization of male mice under chronic psychosocial stress. Psychoneuroendocrinology, in press. [6] A. Bartolomucci, P. Sacerdote, A.E. Panerai, T. Peterzani, P. Palanza, S. Parmigiani, Chronic psychosocial stress induced down-regulation
[21] [22] [23] [24]
[25]
[26]
177
of immunity depends upon individual factors, J. Neuroimmunol. 141 (2003) 58–64. E.C. Beattie, D. Stellwagen, W. Morishita, J.C. Bresnahan, B.K. Ha, M. von Zastrow, M.S. Beattie, R.C. Malenka, Control of synaptic strenght by glial TNF-␣, Science 259 (2002) 2282–2285. H.O. Besedovsky, A. Del Rey, Immuno-neuro-endocrine interactions: facts and hypotheses, Endocr. Rev. 17 (1996) 64–102. L. Capuron, A. Ravaud, P.J. Neveu, A.H. Miller, M. Maes, R. Dantzer, Association between decreased serum tryptophan concentrations and depressive symptoms in cancer patients undergoing cytokine therapy, Mol. Psychiatry 7 (2002) 468–473. R. Dantzer, E.W. Wollman, L. Vitkovic, R. Yirmiya, Cytokines, stress, and depression. Conclusion and perspectives, Adv. Exp. Med. Biol. 461 (1999) 317–329. C. Gemma, L. Imeri, M.R. Opp, Serotonergic activation stimulates the pituitary–adrenal axis and alters interleukin-1 mRNA expression in rat brain, Psychoneuroendocrinology 28 (2003) 875–884. E. Goujon, P. Parnet, S. Laye, C. Combe, R. Dantzer, Adrenalectomy enhances pro-inflammatory cytokines gene expression in the spleen, pituitary and brain of mice in response to lypopolisaccaride, Mol. Brain Res. 36 (1996) 53–62. E. Goujon, P. Parnet, S. Laye, C. Combe, K.W. Kelley, R. Dantzer, Stress downregulates lipopolisaccaride-induced expression of proinflammatory cytokines in the spleen, pituitary and brain of mice, Brain Behav. Immun. 9 (1995) 292–303. V. Grinevich, X.-M. Ma, J.P. Herman, D. Jezova, I. Akmayev, G. Aguilera, Effect of repeated lipopolisaccaride administration on tissue cytokine expression and HPA axis activity in rats, J. Neuroendocrinol. 13 (2001) 711–723. S.H. Haas, K. Schauenstein, Neuroimmunomodulation via limbic structures—the neuroanatomy of psychoneuroimmunology, Prog. Neurobiol. 51 (1997) 195–222. F. Holsboer, The corticosteroid receptor hypothesis of depression, Neuropsychopharmacology 23 (2000) 477–501. J.P. Konsman, P. Parnet, R. Dantzer, Cytokines-induced sickness behavior: mechanisms and implications, Trends Neurosci. 25 (2002) 154–159. J.M. Koolhaas, S.F. de Boer, A.J.H. de Ruiter, P. Meerlo, A. Sgoifo, Social stress in rats and mice, Acta Physiol. Scand. 161 (1997) 69– 72. J. Licinio, M.-L. Wong, The role of inflammatory mediators in the biology of major depression: central nervous system cytokines modulate the biological substrate of depressive symptoms, regulate stress-responsive system, and contribute to neurotoxicity and neuroprotection, Mol. Psychiatry 4 (1999) 317–327. A.C.E. Linthorst, J.M.H.M. Reul, Inflammation and brain function under basal conditions and during long-term elevation of brain corticotropin-releasing hormone level, Adv. Exp. Med. Biol. 461 (1999) 129–152. M. Maes, Major depression and activation of the inflammatory response system, Adv. Exp. Med. Biol. 461 (1999) 25–46. B.S. McEwen, Stress and hippocampal plasticity, Annu. Rev. Neurosci. 22 (1999) 105–122. B.S. McEwen, R.M. Sapolsky, Stress and cognitive function, Curr. Opin. Neurobiol. 5 (1995) 205–216. U. Meyer, M. van Kampen, E. Isovich, G. Flugge, E. Fuchs, Chronic psychosocial stress regulates the expression of both GR and MR mRNA in the hippocampal formation of tree shrews, Hippocampus 11 (2001) 329–336. A.H. Miller, C.M. Pariante, B.D. Pearce, Effects of cytokines on glucocorticoid receptor expression and function, Adv. Exp. Med. Biol. 461 (1999) 107–116. K.T. Nguyen, T. Deak, S.M. Owens, T. Kohno, M. Fleshner, L.R. Watkins, S.F. Maier, Exposure to acute stress induces brain interleukine-1beta protein in the rat, J. Neurosci. 18 (1998) 2239– 2246.
178
A. Bartolomucci et al. / Brain Research Bulletin 62 (2003) 173–178
[27] C.R. Pawlak, Y.-J. Ho, R.K.W. Schwarting, A. Bauhofer, Relationship between striatal level of interleukin-2 mRNA and plus-maze behavior in rats, Neurosci. Lett. 341 (2003) 205–208. [28] C.R. Plata-Salaman, S.E. Llyin, N.P. Turrin, D. Gayle, M.C. Flynn, T. Bedard, Z. Merali, H. Anisman, Neither acute nor chronic exposure to a naturalistic (predator) stressor influence the interleukine-1 system, tumor necrosis factor-␣, transforming growth factor-1, and neuropeptide mRNAs in specific brain regions, Brain Res. Bull. 51 (2000) 187–193. [29] C.R. Plata-Salaman, N.P. Turrin, Cytokine interactions and cytokine balance in the brain: relevance to neurology and psychiatry, Mol. Psychiatry 4 (1999) 303–306. [30] S. Rivest, S. Lacroix, L. Vallieres, S. Nadeau, J. Zhang, N. Laflamme, How the blood talk to the brain parenchyma and the paraventricular nucleous of the hypothalamus during systemic inflammatory and infectious stimuli, Proc. Soc. Exp. Biol. Med. 223 (2000) 22–38. [31] N. Rothwell, Cytokines—killers in the brain? J. Physiol. 514.1 (1999) 3–17.
[32] R. Sapolsky, C. Rivier, G. Yamamoto, P. Plotsky, W. Vale, Interleukine-1 stimulates the secretion of hypothalamic corticotropinreleasing factor, Science 238 (1987) 522–524. [33] H. Schneider, F. Pitossi, D. Balschun, A. Wagner, A. del Rey, H.O. Besedovsky, A neuromodulatory role of interleukin-1beta in the hippocampus, Proc. Natl. Acad. Sci. U.S.A. 95 (1998) 7778– 7783. [34] J. Szelenyi, Cytokines and the central nervous system, Brain Res. Bull. 54 (2001) 329–338. [35] L. Vitkovic, J. Bockaert, C. Jaque, “Inflammatory” cytokines: neuromodulators in normal brain? J. Neurochem. 74 (2000) 457–471. [36] J.I. Webster, L. Tonelli, E.M. Sternberg, Neuroendocrine regulation of immunity, Annu. Rev. Immunol. 20 (2002) 125–163. [37] R. Wei, S.J. Listwak, E.M. Sternberg, Lewis hypothalamic cells constitutively and upon stimulation express higher levels of mRNA for pro-inflammatory cytokines and related molecules: comparison with inflammatory resistant Fischer rat hypothalamic cells, J. Neuroimmunol. 135 (2003) 10–28.
