Blockade of brain type II interleukin-1 receptors potentiates IL1β-induced anorexia in mice

Neuroscience Letters 246 (1998) 101–104

Blockade of brain type II interleukin-1 receptors potentiates IL1b-induced anorexia in mice Sandrine Cremona*, Sophie Laye´, Robert Dantzer, Patricia Parnet Inserm U394, Neurobiologie Inte´grative, rue Camille Saint-Sae¨ns, 33077 Bordeaux Cedex, France Received 15 December 1997; received in revised form 9 March 1998; accepted 16 March 1998

Abstract Interleukin-1b (IL1b) peripheral activities are mediated by type I IL1 receptors (IL1RI), whereas type II IL1 receptors (IL1RII) act as ‘decoy’ targets. To study the functionality of IL1RII in the brain, mice were treated with an intracerebroventricular injection of a neutralising MoAb directed against IL1RII (4E2, 1 mg) followed by recombinant rat IL1b at a dose (2 ng) that produced a moderate but significant decrease of food intake 1 h 30 min after injection. The administration of 4E2 to IL-1b treated mice significantly potentiated IL1b-induced decrease in food intake without altering hypothermia. The effects of IL1b were abrogated in the positive control group treated with IL1ra (2 mg, i.c.v). These results suggest that brain IL1RII down-regulate the effects of IL1b on its cell targets in the brain.  1998 Elsevier Science Ireland Ltd.

Keywords: IL1b; IL1 receptors; Food intake; Rectal temperature; Brain; Mouse

The prototypical pro-inflammatory cytokine interleukin1b (IL1b) is expressed, with its specific receptors, in the central nervous system of sick animals. Glial cells are the main cellular source [3] of IL1b in the mouse brain. IL1b exerts its effects on two types of receptors that are the product of two different genes [19]. The type I (IL1RI) and type II (IL1RII) IL1 receptors are predominantly located in the choroid plexus and the dentate gyrus of the hippocampus of mouse brain [2,16]. Their cellular localisation has been shown on astrocytes [18] and neurons [2,16]. Binding of IL1 to its two receptors is blocked by an endogenous antagonist [1], known as the IL1 receptor antagonist (IL1ra). IL1ra is expressed in the mouse brain in response to a peripheral immune activation [8]. Injection of recombinant IL1b in the brain induces changes in body temperature, activation of the hypothalamo-pituitary-adrenal axis, anorexia, sleepiness and decreased exploring behaviours [7]. The fact that these central effects of IL1b are abrogated by intracerebroventricular (i.c.v.) administration of IL1ra [7] confirms that they are mediated by brain IL1 receptors.

* Corresponding author. Tel.: +33 5 57573714; fax: +33 5 56989029; e-mail: [email protected]

However, the respective role of the two IL1 receptor subtypes remains unclear. In the immune system, the binding of IL-1b to IL1RI activates signalling pathways that lead to cellular [10] and physiological [15] responses. IL1RII are described as cell surface receptors that do not transduce a signal. Moreover, IL1RII can be shed from the cell surface and antagonise IL1 activity by sequestering active ligands [5]. In the brain, the ability of IL1RI to mediate IL1b biological action has been assessed by experiments using specific receptor neutralising treatments. In rat, i.c.v. pre-treatment with an antisense to IL1RI abrogated the anorexia induced by i.c.v. administration of IL1b [20]. In mice, i.c.v. treatment with a neutralising monoclonal antibody (MoAb) directed against IL1RI (35F5) abrogated the decrease of social exploration induced by peripheral or central administration of IL1b [6]. The biological function of brain IL1RII is still controversial. I.c.v. administration of a neutralising MoAb against IL1RII (ALVA42) abrogated the fever induced by i.c.v. injection of IL1b in rat [13]. The same antibody inhibited the IL1b-induced PGE2 release by rat hypothalamus explants [14]. However, the ability of ALVA42 to recognise specifically IL1RII has been questioned [9]. Based on the hypothesis that IL1RII act as ‘decoy’ recep-

0304-3940/98/$19.00  1998 Elsevier Science Ireland Ltd. All rights reserved PII S0304- 3940(98) 00238- 9

102

S. Cremona et al. / Neuroscience Letters 246 (1998) 101–104

tors [5], we predicted that blockade of brain IL1RII should potentiate IL1b central action. To test this hypothesis, we assessed the effect of the blockade of brain IL1RII with a neutralising MoAb (4E2) on IL-1b-induced a moderate decrease in food intake (1 h 30 min post-injection) and hypothermia, and compared it with the effect of IL1ra, used as positive control. Adult male mice of the CD-1 (ICR) BR strain were obtained from Charles River (France) at 5 weeks of age. They were housed in polypropylene cages (42 × 22 × 17 cm) in groups of 10 until surgery, with food and water freely available, at controlled ambient temperature (22 ± 2°C) and under a 12:12 h light-dark cycle (lights off at 0900 h). Experiments were conducted when mice were 8–9 weeks old and weighed 25–35 g. A total of 47 mice were used for all the experiments. Mice were anaesthetised with a mixture of ketamine (12.2 mg/kg) and xylazine (1.8 mg/kg; 10 ml/kg body weight, i.p). Implantation of a guide cannula for i.c.v. injections was carried out when mice were 6-weeks old. A 23-gauge, 7 mm length, stainless steel guide cannula was stereotaxically inserted over the lateral ventricle at the following coordinates computed from bregma: −0.6 mm antero-posterior, ±1.5 mm lateral, −2 mm vertical. Following surgery animals were allowed a 2-week recovery period and individually housed in polypropylene cages (24 × 14 × 13 cm) [6]. Food intake was measured with a calibrated food pellet (Extralabo, Provins, France) placed in the animal cage the day before the injections. Its weight (precision 0.01 g) was measured just before the injections (occurring 1 h 30 min after lights off) and 1 h 30 min later. Rectal temperature (precision 0.01°C) was measured immediately after food intake using a thermocouple rectal probe (Thermalerte TH5; Physitemp, Clifton, NJ, USA). The IL1 receptor antagonist (recombinant human IL1ra; Synergen, Boulder, CO, USA), was diluted in apyrogenic physiological saline (NaCl 0.9%). The rat anti-mouse IL1RII MoAb (4E2) (Pharmingen, San Diego, CA, USA) was diluted in phosphate-buffered saline (PBS; 20 mM sodium phosphate, pH

7.4; 0.25 M NaCl). The binding specificity of 4E2 to IL1RII have been previously demonstrated [10,15]. Rat IgG (Sigma Immuno Chemicals, St. Louis, MO, USA) diluted in PBS served as a control treatment. IL1b (recombinant rat IL1b; NIBSC, Potters Bar, UK) was diluted in 0.1% bovine serum albumin (BSA) solution (A-8806, Sigma). Fresh solutions were prepared on every day of the test. For the control experiment, mice received successively an i.c.v. injection of IL1ra (2 mg/mouse per 1 ml) or NaCl 0.9% followed by IL1b (2 ng/mouse per 1 ml) or BSA 0.1%. To assess the role of IL1RII, mice received successively an i.c.v. injection of 4E2 (1 mg/mouse per 2 ml) or rat IgG (1 mg/mouse) followed by IL1b (1 ng/mouse per 1 ml) or BSA 0.1%. All protocols were approved by the Animal Care and Use Committee of the French Minister of Agriculture. Food intake (g) and rectal temperature (D°C) are expressed as mean ± SEM. Data were analysed according to a one-way analysis of variance and post-hoc comparisons of individual group means were carried out with the least significant difference (LSD) test. Preliminary experiments (data not shown) revealed that IL1b (2 ng, i.c.v.) induced a significant decrease in food intake between 1 h 30 min and 6 h after treatment (F(1,10) = 13.7; P , 0.01). The peak effect occurred at 3 h (0.105 ± 0.066 g). The decrease in food intake at 1 h 30 min (0.512 ± 0.087 g) was moderate but significant and was selected as the dependent variable. IL1b induced at the same time a significant hypothermic response (F(1,10) = 24.1; P , 0.001). As expected, IL1ra (2 mg, i.c.v.) prevented the decrease in food intake (Fig. 1A; group effect: F(3,20) = 4.02, P , 0.05) and in rectal temperature (Fig. 1B; group effect: F(3,20) = 3.31, P , 0.05) induced by IL1b (2 ng, i.c.v). Only mice that were treated with NaCl before IL1b treatment displayed a significant reduction in food intake (NaClBSA vs. NaCl-IL1b, P , 0.01) and a significant hypothermia (NaCl-BSA vs. NaCl-IL1b, P , 0.05). In the second experiment (Fig. 2), IL1b (2 ng, i.c.v.) produced a significant decrease in food intake (Fig. 2A;

Fig. 1. IL1ra (2 mg/mouse, i.c.v.) abrogates the decrease in food intake (A) and hypothermia (B) induced by IL1b (2 ng/mouse, i.c.v). Mice were injected with physiological saline (NaCl, left) or IL1ra (right) and 0.1% BSA (open columns) or IL1b (close columns). Food intake (g) and hypothermia (D°C) are expressed as mean ± SEM. Significantly different from respective control group (LSD test): **P , 0.01; *P , 0.05.

S. Cremona et al. / Neuroscience Letters 246 (1998) 101–104

103

Fig. 2. Blockade of the IL1RII with 4E2 (1 mg/mouse, i.c.v.) potentiates the effects of IL1b (2 ng/mouse, i.c.v.) on food intake (A) but not on rectal temperature (B). Mice were injected with IgG (left) or 4E2 (right) and BSA (open columns) or IL1b (close columns). Food intake (g) and hypothermia (D°C) are expressed as mean ± SEM. Significantly different from respective control group (LSD test): ***P , 0.001; *P , 0.05.

group effect: F(3,18) = 17.4, P , 0.001) whatever the first treatment (LSD: IgG P , 0.001 and 4E2 P , 0.001). Moreover, 4E2 potentiated the IL1b effect. Food intake of IgG-IL1b mice was significantly higher than food intake of 4E2-IL1b mice (LSD: P , 0.05) whereas there was no significant difference between IgG-BSA and 4E2-BSA mice. The IL1b-induced hypothermia (Fig. 2B; F(3,18) = 4.49, P , 0.01) was not altered by the previous administration of 4E2. The present study was designed to assess the capacity of constitutive expressed IL1RII to regulate acute IL1b central effects. We observed that the specific blockade of brain IL1RII with a neutralising MoAb (4E2) potentiated the suppressing effect of IL1b on food intake without altering the IL1b-induced hypothermia. These results demonstrate, for the first time, a possible down-regulatory action of endogenous IL1RII on the effects of IL1b in the brain. Previous studies on biological function of IL1RII in the brain suggested that these receptors mediate the pyrogenic activity of IL1b in rats [13,14]. However, the validity of these experiments has been questioned since the MoAb used (ALVA42) does not appear to be specific to IL1RII and recognise the HLA-DR alpha and beta chains [9]. Hypothermia is a common response to IL1b when mice are tested at ambient temperature below their thermoneutral zone (30–31°C) [11]. In the first experiment, IL1ra abrogated this effect, which confirms the role of brain IL1 receptors in the mediation of IL1b-induced changes in body temperature [7]. In the second experiment, 4E2 did not modify the hypothermic response to IL1b (2 ng, i.c.v.), and it was not more effective when IL-1b was injected at a subthreshold dose (1 ng, i.c.v.; data not shown). This result suggests that brain IL1RI are involved in this IL1b central action, an interpretation that is consistent with previous results obtained on IL1RI KO mice [12]. The main result of this study is the potentiating action of 4E2 on the decrease in food intake induced by IL1b. In comparison to their respective control group, mice treated with 4E2-IL1b ate three times less, while mice treated with IgG-IL1b ate two times less. This is an important result

since it demonstrates a negative regulatory function of brain IL1RII on IL1b central action. The possibility that IL1RII function as decoy targets has been first proposed by Colotta et al. [5], and it has received support mainly from in vitro transfection experiments in keratinocytes. The over-expression of cell surface IL1RII and of a soluble form in the culture medium impaired the IL1b responsiveness of these cells [4]. In vivo, transgenic mice over-expressing IL1RII on keratinocytes were protected from the acute cutaneous vascular leakage and chronic inflammation induced by IL1 [17]. The physiological relevance of these results is somewhat limited by the artificial over-expression of IL1RII. The findings of the present study, which investigates the role of endogenously expressed IL1RII, support the concept that IL1RII can be considered, like IL1ra, as an endogenous system that down-regulates IL1 action in the brain. IL1RII would negatively regulate the IL1 system by binding IL1 and preventing signal transduction via IL1RI [4,5]. This study revealed the nature of the interaction between brain IL1RII and exogenously injected IL1b. In previous experiments [6], we demonstrated that brain IL1RI mediates the decreased social exploration induced not only by exogenous injection of IL1b but also by endogenous production of brain IL1 (a and b) in response to systemic IL1b. It can therefore be predicted that blockade of brain IL1RII should potentiate the response to endogenously produced IL1, but this remains to be tested. This study was supported by INSERM, INRA, DRET and NIH (MH-51569 and DK-49311). We wish to thank Dr. R.J. Vannice for his generous gift of recombinant human IL1 receptor antagonist (IL1ra). Recombinant rat IL1b was obtained through the BIOMED I Concerted Action ‘Cytokines in the Brain’. S.C. is financially supported by the Conseil Re´gional d’Aquitaine (France) and S.L. by the IPSEN foundation. [1] Arend, W.P., Interleukin-1 receptor antagonist, J. Clin. Invest., 88 (1991) 1445–1451.

104

S. Cremona et al. / Neuroscience Letters 246 (1998) 101–104

[2] Ban, E., Milon, G., Prudhomme, N., Fillion, G. and Haour, F., Receptors for interleukin-1 (a and b) in mouse brain: mapping and neuronal localization in hippocampus, Neuroscience, 43 (1991) 21–30. [3] Benveniste, E.N., Cytokines: influence on glial cell gene expression and function. In J.E. Blalock (Ed.), Neuroimmunoendocrinology, Karger, Basel, 1992, pp. 106–153. [4] Bossu, P., Visconti, U., Ruggiero, P., Macchia, G., Muda, M., Bertini, R., Bizzarri, C., Colagrande, A., Sabbatini, V., Maurizi, G., Delgrosso, E., Tagliabue, A. and Boraschi, D., Transfected type II interleukin-1 receptor impairs responsiveness of human keratinocytes to interleukin-1, Am. J. Pathol., 147 (1995) 1852– 1861. [5] Colotta, F., Re, F., Muzio, M., Bertini, R., Polentarutti, N., Sironi, M., Giri, J.G., Dower, S.K., Sims, J.E. and Mantovani, A., Interleukin-1 type II receptor: a decoy target for IL-1 that is regulated by IL-4, Science, 261 (1993) 472–475. [6] Cremona, S., Goujon, E., Kelley, K.W., Dantzer, R. and Parnet, P., Brain type I but not type II interleukin-1 (IL1) receptors mediate the behavioural effects of IL1b, Am. J. Physiol. (Regul. Integr. Comp. Physiol.), (1998) in press. [7] Dinarello, C.A., The biological properties of interleukin-1, Eur. Cytokine Netw., 5 (1994) 517–531. [8] Gabellec, M.A., Griffais, R., Fillion, G. and Haour, F., Expression of interleukin-1a, interleukin-1b and interleukin-1 receptor antagonist mRNA in mouse brain: regulation by bacterial lipopolyssacharide (LPS) treatment, Mol. Brain Res., 31 (1995) 122–130. [9] Gayle, M.A., Sims, J.E., Dower, S.K. and Slack, J.J., Monoclonal antibody 1994-01 (also known as ALVA42) reported to recognize type II IL-1 receptor is specific for HLA-DR alpha and beta chains, Cytokine, 6 (1994) 83–86. [10] Hestdal, K., Ruscetti, F.W., Chizzonite, R., Ortiz, M., Gooya, J.M., Longo, D.L. and Keller, J.R., Interleukin-1 (IL-1) directly and indirectly promotes hematopoietic cell growth through type I IL-1 receptor, Blood, 84 (1994) 125–132. [11] Kozak, W., Conn, C.A. and Kluger, M.J., Lipopolysaccharide induces fever and depresses locomotor activity in unrestrained

[12]

[13]

[14]

[15]

[16]

[17]

[18]

[19] [20]

mice, Am. J. Physiol. (Regul. Integr. Comp. Physiol. 35), 266 (1994) R125–R135. Leon, L.R., Conn, C.A., Glaccum, M. and Kluger, M.J., IL-1 type I receptor mediates acute phase response to turpentine, but not lipopolyssacharide, in mice. Am. J. Physiol. (Regul. Integr. Comp. Physiol. 40), 271 (1996) R1668–R1675. Luheshi, G., Hopkins, S.J., Lefeuvre, R.A., Dascombe, M.J., Ghiara, P. and Rothwell, N.J., Importance of brain IL-1 type II receptors in fever and thermogenesis in the rat, Am. J. Physiol. (Endocrinol. Metab. 28), 265 (1993) E585–E591. Mirtella, A., Tringali, G., Guerriero, G., Ghiara, P., Parente, L., Preziosi, P. and Navarra, P., Evidence that the interleukin-1binduced prostaglandin E2 release from rat hypothalamus is mediated by type I and type II interleukin-1 receptors, J. Neuroimmunol., 61 (1995) 171–177. Oldenburg, H.S.A., Pruitt, J.H., Cazarus, D.D., Ragy, M.A., Chizzonite, R., Lowry, S.F. and Moldaver, L.L., Interleukin 1 binding to its type I, but not type II receptor, modulates the in vivo acute phase response, Cytokine, 7 (1995) 510–516. Parnet, P., Amindari, S., Wu, C., Brunke-Reese, D., Goujon, E., Weyhenmeyer, J.A., Dantzer, R. and Kelley, K.W., Expression of type I and type II interleukin-1 receptors in mouse brain, Mol. Brain Res., 27 (1994) 63–70. Rauschmayr, T., Groves, R.W. and Kupper, T.S., Keratinocytes expression of the type 2 interleukin 1 receptor mediates local and specific inhibition of interleukin 1-mediated inflammation, Proc. Natl. Acad. Sci. USA, 94 (1997) 5814–5819. Rubio, N., Demonstration of the presence of an interleukin-1 receptor on the surface of murine astrocytes and its regulation by cytokines and Theiler’s virus, Immunology, 82 (1994) 178– 183. Sims, J.E. and Dower, S.K., Interleukin-1 receptors, Eur. Cytokine Netw., 5 (1994) 539–546. Sonti, G., Flynn, M.C. and Plata-Salaman, C.R., Interleukin-1 (IL-1) receptor type I mediates anorexia but not adipsia induced by centrally administered IL-1b, Physiol. Behav., 62 (1997) 1179–1183.