Presence of a reduced opioid response in interleukin-6 knock out mice

European Journal of Neuroscience, Vol. 11, pp. 1501–1507, 1999 © European Neuroscience Association

Presence of a reduced opioid response in interleukin-6knock out mice

Mauro Bianchi,1 Roberto Maggi,2 Federica Pimpinelli,2 Tiziana Rubino,3 Daniela Parolaro,3 Valeria Poli,4Gennaro Ciliberto,5 Alberto E. Panerai1 and Paola Sacerdote1

1Department of Pharmacology, 2Department of Endocrinology and 3Institute of Pharmacology, University of Milan, Italy4Department of Biochemistry, University of Dundee, Scotland5IRBM ‘P. Angeletti’, Rome, Italy

Keywords: β-endorphin, interleukin-6, mouse, nociception, opioids, tolerance, transgenic

Abstract

Cytokines are known to influence neuronal functions. The purpose of this study was to investigate the putative role of the cytokineinterleukin-6 (IL-6) in the pathways involved in opioid-mediated responses, by using IL-6-deficient mice. We reported that with athermal stimulus IL-6-knock-out (IL-6KO) mice presented nociceptive thresholds similar to those measured in their controls.However, they showed a reduced analgesic response both to the restraint stress and to the administration of low doses of morphine.Hypothalamic levels of the opioid peptide β-endorphin were significantly higher in IL-6KO mice than they were in their controls.The development of tolerance to the analgesic effect of morphine was more rapid in IL-6-deficient mice than in wild-type controls.Binding experiments showed that the number of opioid receptors in the midbrain, but not in the hypothalamus, decreased in IL-6KO mice. Autoradiographic binding analysis revealed that the density of µ receptors diminished while the δ-opioid receptors didnot. These results suggest that IL-6 is necessary for a correct development of neuronal mechanisms involved in the response toboth endogenous and exogenous opiates.

Introduction

It has recently become accepted that cytokines are common mediatorsof both the immune and the nervous system (Besedovsky & DelRey, 1996).

Interleukin-6 (IL-6) is a multifunctional cytokine, implicated in theregulation of multiple aspects of immune responses, haemopoiesisand inflammation (Van Snick, 1990; Kishimotoet al., 1992). Inaddition to its essential role in the functions of the immune system,it is present in the nervous system, where it exerts a variety of effects(Merril, 1992; Mehler & Kessler, 1997). Low amounts of IL-6(Besedovsky & Del Rey, 1996; Gruol & Nelson, 1997; Marzet al.,1998) and IL-6 receptors (Sawadaet al., 1993) have been found tobe constitutively expressed in different cell types in the CNS, andIL-6 levels are elevated in neurodegenerative disorders such asAlzheimer disease (Hu¨ll et al., 1996). Moreover, this cytokine isinvolved in the modulation of several CNS functions; IL-6 is able toalter neuronal development and responses to NMDA in culturedrat cerebellar granule neurons (Qiuet al., 1995); it affects themonoaminergic system in the brain (Zalcmanet al., 1994) and wepreviously showed that it can affect learning and memory in themouse (Bianchiet al., 1997b) as well as sleep and food-intake(Rothwell, 1991).

Although other important proinflammatory cytokines such asinterleukin-1 (IL-1) and tumour necrosis factor-a (TNF-α) have beeninvolved in the modulation of nociceptive pathways in the peripheraland central nervous system (Bianchiet al., 1992; Ferreiraet al., 1993;

Correspondence: Dr Paola Sacerdote, Department of Pharmacology, ViaVanvitelli 32, 20129 Milano, Italy. E-mail: [email protected]

Received 8 June 1998, revised 30 November 1998, accepted 3 December 1998

Oka et al., 1993), little is known on the possible role of IL-6 in thisrespect. We thought it would be of interest to study the possibleinvolvement of this cytokine in nociception, by using a recentlygenerated strain of mice carrying a null mutation in the IL-6 gene,and therefore defective in IL-6 production (Poliet al., 1994). Thesemice represent a valid model for an investigation of the role of IL-6in several physiological and pathological processes (Kopfet al., 1994;Chai et al., 1996; Cressmanet al., 1996). It has been demonstratedthat mice deficient in IL-6 present several alterations in both peripher-ally and centrally mediated responses; they show defective leucocyterecruitment in subcutaneous air pouches injected with inflammatorysubstances (Romanoet al., 1997), a reduction in the synthesis of liveracute phase proteins in response to local inflammation (Fattoriet al.,1994) and a complete resistance to turpentine-induced fever, anorexiaand cachexia (Kozaket al., 1997). In these mice Xuet al. (1997)noticed altered nociceptive responses to peripheral inflammation. Inour present study homozygous IL-6-knock-out (IL-6KO) mice andtheir appropriate controls were used to evaluate the putative effect ofIL-6 on the pathways involved in the modulation of pain sensationsand in the responses to exogenous and endogenous opioids.

Part of the data reported here have already been presented inabstract form (Bianchiet al., 1997c).

Materials and methods

Animals

In order to obtain IL-6-deficient mice and to abolish IL-6 function,the sequence coding for the amino-terminal half of the protein waseliminated. Embryonic stem (ES) cell clones (129 type, from the EScell line CCE, see Robertsonet al., 1986) carrying the IL-6 mutation

1502 M. Bianchiet al.

were injected into blastocysts of C57BL6 mice and these weretransplanted into the uteri of F1 (CBA3 C57BL6) foster mothers.Male chimeras were mated to MFI strain females and agouti offspring(representing germline transmission of the ES genome), were screenedfor the presence of the targeted IL-6 locus using Southern blotanalysis of ECO-RI-digested tail DNA using the probe 59. Femaleoffspring heterozygous for the mutation were bred once with miceof the 129/SV/EV strain, from which the CCE ES cells were derived.The resulting heterozygous offspring were bred together to generatehomozygous mice (IL-6 –/–) for the mutation. These procedures havebeen described in greater detail (Poliet al., 1994; Lattanzioet al.,1997). Wild-type (WT) littermates (IL-61/1) were used as controls.The homozygous IL-6 –/– mice and the WT IL-61/1 mice used inthese experiments were bred in our Department. The mice werehoused in a temperature-controlled room, with food and wateradlibitum and under a 12 h light : 12 h dark cycle. Only male micewere used in the study. The animals used were aged from 45 to60 days. No differences were observed in the growth curves betweenIL-6KO and WT mice. In each experimental group we used eightanimals of the same age. Experiments involving animals wereperformed in accordance with the NIH guidelines for care and useof laboratory animals. The animal work was done in the Departmentof Pharmacology laboratories, and approved by the departmentalethical committee.

Drugs

Morphine hydrochloride (SIFAC, Como, Italy) was dissolved in asolution of 0.9% NaCl and administered s.c.

Evaluation of nociceptive thresholds

The hot-plate test (Eddyet al., 1950; Bianchiet al., 1996a) was usedto assess nociceptive thresholds in untreated mice and to evaluate theanalgesic effects induced by restraint stress and acute or chronicadministration of morphine. The apparatus (Basile, Comerio, Como,Italy) was set at a temperature of 546 0.5 °C; the cut-off time was60 s. The time between the placement of the mouse on the hot plateand the moment when it licked both fore paws was recorded as theresponse latency. In order to prevent tissue damage and adaptation,only one hot-plate response was measured per time point.

Experimental design in morphine studies

In the hot-plate test, the nociceptive thresholds were measured 60 minafter the subcutaneous administration of morphine at doses of 0.625,1.25 and 2.5 mg/kg.

In order to evaluate the development of tolerance to the analgesiceffect of morphine, the drug was injected twice a day at 09.00 h and19.00 h. Morphine was administered s.c. at a dose of 2.5 mg/kg for11 days. Nociceptive thresholds were evaluated with the hot-platetest on alternate days, immediately before, and 60 min after, the first(09.00 h) morphine administration. Data are expressed as the algebraicdifference between the test latency and the basal latency, positiveresults thereby indicating analgesia.

Restraint stress and evaluation of stress-induced analgesia

In order to induce stress, mice were physically restrained for 8 h ina 50-mL polypropylene centrifuge tube in which holes were drilled;animals were not squeezed or compressed, they could breath freelyand did not show pain-related behaviours. It has been demonstratedthat this procedure activates a stress-induced analgesia (Porro & Carli,1988). Control animals were left in their home cages. Nociceptive

© 1999 European Neuroscience Association,European Journal of Neuroscience, 11, 1501–1507

thresholds were measured with the hot-plate test immediately beforethe beginning of restraint and 5 min after the end of stress.

Binding studies with brain membranes

Mice were killed by cervical dislocation, and midbrain and hypothal-amic tissues were collected as previously described (Bianchiet al.,1997a), frozen on dry ice and stored at –70 °C until the receptorbinding assay. Unless otherwise specified, all reagents used camefrom Sigma, St. Louis, Mo, USA. Brain tissues were homogenizedin 10 volumes of 0.32M sucrose. The homogenates were centrifugedat 1400g for 10 min; the resulting supernatants were decanted andpreincubated for 30 min at 37 °C to eliminate the endogenous ligandthat might interfere with the assay and then further centrifuged for30 min at 40 000g at 4 °C. The pellets obtained (crude membranepreparations) were resuspended and homogenized in assay buffer(Tris-HCl 50 mM, pH 7.4) and immediately used for the bindingassay. The proteins in each crude membrane preparation was deter-mined using a micro method (Bradford, 1976) using human serumalbumin as a standard. The receptor binding assay was performedusing [15, 15-(n)-3H-diprenorphine] (3H-DIP, specific activity 31 Ci/mmol, Amersham Italia, Milano, Italy) as an opioid ligand. This drugbinds the three major classes of opioid receptors (µ, δ and κ) withsimilar affinities (Wolleman, 1990). The conditions of the receptorbinding assay were optimized as previously described (Maggiet al.,1996). Aliquots from a pool of three tissue membrane preparations(150µg of protein for the hypothalami and 700µg for midbrain)were incubated in the presence of labelled ligands (from 0.1 to100 nM). Incubations were carried out at 25 °C for 30 min. Thecontent of the tubes were then individually filtered through WhatmanGF/B filters (Whatman International Ltd, Maidstone, England) pre-soaked in assay buffer saturated with isoamyl alcohol. Each filterwas washed twice with 5 mL ice-cold assay buffer and counted in a7-mL Instagel scintillation cocktail. Nonspecific binding was assessedby addition of 1µM of unlabelled naltrexone (Sifac, Como, Italy).All the samples were assayed in duplicate, and three independentsaturation curves were carried out for each experimental group.

Autoradiographic binding study

Mice were killed by cervical dislocation and brains were rapidlyremoved and frozen by lowering them gently into liquid nitrogen.They were stored at –80 °C until processing. Brains were broughtto –16 °C in a cryostat and a series of 12-µm-thick serial sectionswere collected on gelatine-coated slides. The sections were brieflydried at 30 °C and stored at –80 °C until they were processed forreceptor binding autoradiography (Eghbaliet al., 1987). The slideswere incubated for 1 h at room temperature with 4.5 nM [3H]DAMGOfor µ-opioid receptor (52.5 Ci/mM, NEN Life Science Products Italy,Cinisello Balsamo, Italy) and 7 nM [3H]DPDPE forδ receptor (36 Ci/mM, NEN Life Science Products Italy, Cinisello Balsamo, Italy) inbinding buffer (50 mM tris-HCl, pH 7.4, 10µM bacitracin). Adjacentsections were incubated in parallel with 2.5µM naloxone to assessnonspecific binding. Sections were rinsed three times for 5 min at4 °C in a 50 mM Tris-HCl (pH 7.4) buffer. After washing, the sectionswere dipped briefly in water and dried under a cool stream of air.Autoradiograms were made by exposing the dried sections in X-raycassettes for 30 days to a tritium-sensitive film (Hyperfilm-3H;Amersham Italia, Milano, Italy). Autoradiograms were developedwith a Kodak D19 developer (25 °C, 4 min), fixed in Kodak Unifix(8 min) and rinsed with water (15 min).

The intensity of the receptor binding signal was assessed bymeasuring the grey levels of the autoradiographic films with an image

Opioid responses in IL-6-deficient mice 1503

analysis system consisting of a solid state video camera (Hamamatsu,Tokyo, Japan) connected to an Apple Macintosh II personal computer.We used the public domain 1.47 software (NIH, Bethesda, MD,USA). Each brain section was traced with a mouse cursor control,and the light transmittance was determined as grey level. The greylevel of densitometric measurements calculated after the subtractionof the film background density was established within the linearrange, determined using tritium standards (3H Microscales, AmershamItalia, Milano, Italy).

Measurement of β-endorphin (BE) and substance P (SP)

For the measurement of tissue concentrations of peptide, micewere killed by microwave irradiation in order to prevent enzymaticdegradation and the hypothalamus was dissected out as previouslydescribed (Bianchiet al., 1997a). The tissue was resuspended in1 mL 0.1N acetic acid, homogenized and centrifuged at 10 000g.Supernatants were collected for radioimmunoassay, and pellets wereused for protein evaluation. Protein concentration was measured usingLowry’s method (Lowry et al., 1951). The BE measurement wasperformed using a radioimmunoassay according to the method previ-ously described by Paneraiet al. (1988). The BE antiserum wasobtained against synthetic BE (1–27) (Sigma, St Louis, Mo, USA)and it is directed towards the C-terminal of the peptide. The 3-[125]

iodotyrosyl-27 BE was purchased from Amersham Italia, Milano, Italy.SP concentrations were measured using radioimmunoassay with

antiserum and methods previously described and validated (Paneraiet al., 1988; Bianchiet al., 1996b). The antibody was raised in rabbitagainst synthetic SP (Sigma, St Louis, Mo, USA), and it was directedtowards the C-terminal of the peptide. The125I-SP was purchasedfrom Amersham Italia, Milano, Italy.

Statistical analysis

Behavioural data were analysed using onewayANOVA followed bythe Bonferroni’s t-test for multiple comparisons. The statisticalanalysis of results concerning BE and SP concentrations wereperformed by means of Student’st-test. The data obtained in thereceptor binding experiments were analysed using the LIGANDprogram (Munson & Rodbard, 1980). Statistical comparisons betweenreceptor binding results were made with theF-test for correlatedparameters using the LIGAND program. Autoradiographic densitieswere compared by means of Student’st-test.

Results

Both behavioural and biochemical data indicate that IL-6KO micediffered markedly from control animals in several test situationsdesigned to assess the functional state of the opioid system.

Nociceptive thresholds and analgesic responses to acuteadministration of morphine

Assessment of nociceptive thresholds showed that IL-6KO miceexhibited a sensitivity to a painful stimulus similar to that observedin control animals (Fig. 1). However, different responses to theadministration of morphine were observed in WT and IL-6KO mice.At the very low dose of 0.625 mg/kg, a significant antinociceptiveresponse was elicited only in the WT animals (onewayANOVA,P , 0.05 versus basal, pretreatment values), while the hot-platelatencies of IL-6KO mice did not differ before or after the morphinetreatment. At the dose of 1.25 mg/kg morphine induced a significantanalgesic effect in both WT and IL-6KO mice (onewayANOVA,P , 0.05 versus basal values), but the response was significantly


FIG. 1. Analgesic responses to acute morphine administration in WT and IL-6-deficient (IL-6KO) mice. Nociceptive thresholds were measured using thehot-plate test immediately before (basal) and 60 min after morphine treatment.The results are expressed in seconds (cut-off time, 60 s). Each bar representsthe mean6 SEM from eight animals. Statistical significances were determinedusing the one-wayANOVA followed by Bonferroni’s t-test for multiplecomparisons; *P , 0.05 versus basal values;1P , 0.05 versus WT animals.

lower in IL-6KO mice (onewayANOVA P , 0.05) than in theircontrols (Fig. 1). The analgesic effect induced by the administrationof morphine at the dose of 2.5 mg/kg was similar in IL-6KO andWT mice.

Morphine tolerance

The daily injections of morphine (2.5 mg/kg twice s.c.) elicitedtolerance to its analgesic effect in the hot-plate test as shown inFig. 2. The results are expressed in seconds as the algebraic differencebetween the test (609 after morphine) and the basal (i.e. pretreatment)latencies. We considered that tolerance was achieved when thechallenge with morphine did not elicit any significant (P , 0.05)enhancement of nociceptive thresholds in comparison with saline-treated animals. However, whereas in WT mice tolerance was evidentafter 11 days of treatment, in IL-6KO mice it developed earlier; itwas already evident after 5 days of treatment with morphine.

Stress-induced analgesia

Consistent with previous reports (Porro & Carli, 1988; Glavinet al., 1994), restraint stress produced a significant enhancement ofnociceptive thresholds to a thermal stimulus in WT animals. In fact,Fig. 3 shows that 5 min after the end of the stress, a significantincrease of nociceptive thresholds is present in comparison with basalthresholds, before the beginning of restrain treatment (onewayANOVA,P , 0.05). This figure also shows that, in contrast, the exposure torestraint stress did not produce any analgesic response in IL-6KO mice.

BE and SP hypothalamic levels

The hypothalamic levels of the opioid peptide BE were significantlyhigher in IL-6-deficient mice than in their controls: 46.396 9.85 and


FIG. 2. Analgesic responses to acute administration of saline or morphine(2.5 mg/kg s.c.) in WT and IL-6-deficient (IL-6KO) mice chronically treatedwith the opioid drug, or saline, twice daily. Nociceptive thresholds weremeasured using the hot-plate test immediately before and 60 min after anacute treatment with morphine. The results are expressed in seconds as thealgebraic difference between the test (609) and the basal latency. Each pointrepresents the mean6 SEM from eight animals. Statistical significances aredetermined using the one-wayANOVA followed by Bonferroni’s t-test formultiple comparisons; *P , 0.05 versus saline-treated group.

6.766 2.2 ng/mg protein, respectively (mean6 SEM, P , 0.001,Student’st-test).

It is interesting to note that the hypothalamic concentrations of thenonopioid peptide SP did not differ in IL-6KO and WT mice:1.216 0.017 and 1.256 0.02 ng/mg protein, respectively(mean6 SEM).

Brain opioid receptors

As described in Fig. 4, the analysis of the saturation curves showeda significant (P , 0.001,F-test for correlated parameters using theLIGAND program) decrease of the number of opioid binding sites inthe midbrain (Bmax5 0.326 0.06 and 0.216 0.04 pmol/mg protein,mean6 SD, in WT and IL-6KO mice, respectively) with no significantmodifications of the binding affinity (Kd5 0.916 0.21 and1.246 0.34 nM, respectively).

The number of opioid binding sites in the hypothalamus did notdiffer significantly in WT and IL-6KO mice (Bmax5 1.196 0.54 and


FIG. 3. Analgesia induced by the exposure to restraint stress (8 h) in WT andIL-6-deficient (IL-6KO) mice. Nociceptive thresholds were measured usingthe hot-plate test immediately before (basal) and 5 min after the end of thestress section (stress). Each bar represents the mean6 SEM from eightanimals. Statistical significances were determined using one-wayANOVA

followed by the Bonferroni’st-test for multiple comparisons; *P , 0.01 versusbasal values and IL-6KO group.

FIG. 4. Scatchard plot of saturation curves of3[H]diprenorphine to midbraincrude membrane preparations of WT (d) and IL-6KO (s) mice.

1.146 0.52 pmol/mg protein, respectively). Similarly, no significantmodifications of the opioid binding affinity were observed in thehypothalamus of WT and IL-6KO mice (Kd5 2.566 0.99 and1.966 0.91 nM, respectively).

The autoradiographic binding analysis ofµ-opioid receptors,reported in Fig. 5, revealed a significant decrease of these opioidreceptors in the grey matter of the midbrain of IL-6KO mice incomparison with WT animals. The densitometric analysis revealed asignificant 20% decrease inµ-opioid receptor density in this area(P , 0.05 Student’st-test) (Fig. 5, lower right panel). A positivesignal forµ-opioid receptors in WT animals was present also in theinterpeduncular nucleus and hippocampus, but no alteration ofµreceptor density in these areas was observed in IL-6KO mice. Finally,


FIG. 5. Autoradiograms of representative coronal sections of the midbrain region in WT and IL-6 knock out mouse brains showingµ-opioid receptors. A cameralucida drawing from the atlas of Paxinos and Watson is shown for reference. The bottom right-hand panel shows the quantification of the opioid receptorlevelin the grey matter of WT and KO mice. Data are the mean6 SEM from six animals. *P , 0.05 versus WT.

no differences inδ-opioid receptor density were detected between WTand KO mice, in cortex, striatum or hippocampus (data not shown).

Discussion

Overall, these data suggest that there is a complex interaction betweenthe IL-6 and the opioid systems. It is now clear that many inflammatorycytokines, such as IL-6, are involved in successive stages of develop-ment of the central and peripheral nervous systems, including expres-sion of neurotransmitters and receptors (Fann & Patterson, 1994;Mehler & Kessler, 1997). Consistently with these data, it has beenreported that IL-6-deficient mice exhibit modifications in someneurobehavioural performances (Chaiet al., 1996; Kozaket al., 1997).In this study we have used IL-6-deficient mice to further investigatethe role of IL-6 in neurobehavioural performance such as painsensitivity, activation of the endogenous analgesic system by stressand responses to morphine administration. We observed that the lackof IL-6 did not have any significant effect on the magnitude of basalnociceptive responses to a thermal stimulus. However, the absenceof IL-6 had an effect on nociceptive thresholds when the mice hadbeen treated with morphine. Indeed, the analgesic effect of this opioiddrug was clearly reduced in IL-6KO mice compared with theircontrols. Moreover, these animals were unable to activate the endo-genous analgesic system as indicated by the complete resistance tothe analgesic effect of stress. Since a role for opioid mechanisms in


the development of analgesia that follows restraint stress has beendemonstrated (Kelly & Franklin, 1987), the absence of stress-inducedanalgesia in IL-6KO mice further suggests an alteration of the opioidsystem in these animals. Earlier reports on IL-6-deficient mice indicatethat, when injected with bacterial endotoxin, these mice exhibit anormal corticosterone response (Kopfet al., 1994). Consistently withthis observation, normal corticosterone responses to restraint stresshave been observed in IL-6KO mice (Manfrediet al., 1998). Thus,not all the stress-induced responses are modified in these animals.

The finding that the lack of IL-6 induces an alteration in theanalgesic responses to morphine administration raises the question asto whether this condition is associated with changes in the synthesisof neuropeptides involved in the modulation of pain sensitivity andin morphine-induced analgesia. Many studies in this field havedemonstrated that the opioid peptide BE is able to enhance nociceptivethresholds and to affect the analgesic response to morphine administra-tion in animals (Meglioet al., 1977; Jacquet, 1978; Paneraiet al.,1987). Interestingly, we observed that, under basal conditions, thelevels of BE, but not those of the non-opioid peptide SP, werestrongly modified in the hypothalamus of IL-6-deficient mice. Theseobservations indicate that the alteration of the opioid system in IL-6-deficient mice is rather specific.

The results of the binding studies also indicate that the opioidreceptors are altered in the brain of these mice. Indeed, we observeda relevant reduction of opioid receptors in the midbrain, a brain area


that is critical for the analgesic action of opioids, and which receivesβ-endorphinergic terminals from the hypothalamus (Basbaum &Fields, 1984). A relationship between the high hypothalamic levelsof BE and the down-regulation of opioid receptors in the midbraincan therefore exist. Moreover, the autoradiography study indicatesthat the decreasing number of opioid receptors depends mainly onthe modification ofµ receptors, to which BE binds with high affinity.The other important class of central receptors involved in nociception,δ receptors, for which BE has only a slight affinity, are not affectedby the lack of IL-6.

The observation that tolerance to the analgesic effect of morphinedevelops earlier in IL-6KO than in WT mice further indicates thatthe opioid system is modified in these animals.

Interestingly, the IL-6KO animals share many characteristics withopioid tolerant mice. In normal mice made tolerant to morphine,decreased analgesic responses to morphine administrations, crosstolerance with stress-induced analgesia, and lower numbers ofµ-opioid receptors have been observed (Lewis et al., 1980; Bhargava& Gulati, 1990). It can therefore be speculated that in mice, the lackof IL-6 induces an ‘opioid tolerant-like status’.

It has been reported that chronic exposure to IL-6 enhances theintracellular calcium responses to glutamate and NMDA stimulationin developing neurons (Hollidayet al., 1995; Qiuet al., 1995). Inparticular, the extracellular calcium influx and release from storeswere larger in IL-6 treated neurons compared with controls. Since ithas been demonstrated that morphine exerts an inhibitory effect oncellular calcium (North, 1986) and that calcium plays an important rolein modulating the analgesic effect of morphine in both experimentalanimals and humans (Cartaet al., 1990; Kuzminet al., 1994; Smith& Stevens, 1995), we can speculate that changes in calcium signallingrelated to the persistent lack of IL-6 could alter neuronal function indeveloping nervous system and therefore the effects of morphine inadult mice. However, the data available at the moment do not allowus to make any definitive conclusions.

Although it could be expected that the absence of IL-6 could becompensated for by other pro-inflammatory cytokines, such as IL-1and TNF-α, which have been implicated in the modulation ofnociceptive pathways (Bianchiet al., 1992; Ferreiraet al., 1993; Okaet al., 1993), this does not seem to be relevant in our case. In theseknock out mice, IL-1 levels and IL-1 induction by lipopolysaccharidesare normal. Although an increased TNF-α production has beenobserved (Fattoriet al., 1994), TNF has been shown to induce acentral analgesia not related to the opioid system (Bianchiet al.,1992). Thus, its involvement in the altered opioid responses observedin these animals seems unlikely.

The lack of expression of the IL-6 gene might have affected braindevelopment and, as a consequence, some neuronal functions. Wepreviously showed that in these animals aggressive behaviours werealtered (Allevaet al., 1998) and we now report that the lack of IL-6is related to an altered function of the opioid system. The possibilitythat the effect of the IL-6 gene knock-out on brain developmentcould underly these abnormal responses is further suggested by theobservations that in a model of transgenic mice overexpressing IL-6in the brain (Fattoriet al., 1995) we did not observe any alterationof the opioid responses (unpublished results). Moreover, the IL-6-overexpressing mice presented slight behavioural alterations whichcould not be considered opposite to the ones presented by the IL-6KO mice (Alleva et al., 1998). Moreover, we previously reportedthat the exogenous administration of IL-6 to normal mice did notinduce any effect on nociceptive thresholds in the hot-plate test(Bianchi et al., 1997b).

In conclusion, although we are aware that there are some potential


limitations with the use of transgenic mice in behavioural studies(such as the development of compensatory mechanisms or theexistence of redundancy mechanisms), and that further studies in thisdirection are needed in order to understand the basis of the alteredopioid response observed, we suggest this model of mice provides avalid tool for the study of neuronal modifications responsible for aresistance to the effects of opioid drugs and for the development ofthe tolerance to opioids.

Abbreviations

BE, β-endorphin; ES, embryonic stem; IL-1, interleukin-1; IL-6, interleukin-6; IL-6KO, interleukin-6 knock-out; SP, substance P; TNF-α_, tumour necrosisfactor; WT, wild type.

References

Hull, M., Strauss, S., Berger, M., Volk, B. & Bauerm, J. (1996) Theparticipation of IL-6, a stress-inducible cytokine, in the pathogenesis ofAlzheimer’s disease. Behav. Brain Res., 78, 37–41.

Alleva, E., Cirulli, F., Bianchi, M., Bondiolotti, G.P., Chiarotti, F., de Acetis,L. & Panerai, A.E. (1998) Behavioural Characterisation of Interleukin-6overexpressing or deficient mice during agonistic encounters.Eur. J.Neurosci., 10, 3664–3672.

Basbaum, A.I. & Fields, H.L. (1984) Endogenous pain control systems:brainstem spinal pathways and endorphin circuitry.Annu. Rev. Neurosci.,7, 309–338.

Besedovsky, H.O. & Del Rey, A. (1996) Immune neuroendocrine interactions:facts and hypothesis.Endocrine Rev., 17, 64–102.

Bhargava, H.N. & Gulati, A. (1990) Down regulation of spinal cord mu-opioid receptors in morphine tolerant-dependent rat.Eur. J. Pharmacol.,190, 305–311.

Bianchi, M., Clavenna, A., Bondiolotti, G.P., Ferrario, P. & Panerai, A.E.(1997a) GM-CSF affects hypothalamic neurotransmitter levels in mice:involvement of interleukin-1.Neuroreport, 8, 3587–3590.

Bianchi, M., Ferrario, P., Clavenna, A. & Panerai, A.E. (1997b) Interleukin-6 affects scopolamine-induced amnesia, but not brain amino acid levels inmice.Neuroreport, 8, 1775–1778.

Bianchi, M., Maggi, R., Panerai, A.E. & Martini, L. (1996a) RU 486 reducesmorphine-induced analgesia in mice, but steroid receptors are not involved.Pharmacology, 52, 387–391.

Bianchi, M., Sacerdote, P., Poli, V. & Panerai, A.E. (1997c) Interleukin-6knock-out mice exhibit increased hypothalamic beta-endorphin levels, andreduced opioid sensitivity.Soc. Neurosci. Abstr., 23, 269.2.

Bianchi, M., Sacerdote, P., Ricciardi Castagnoli, P., Mantegazza, P. & Panerai,A.E. (1992) Central effects of tumor necrosis factor-(and interleukin-1 (onnociceptive thresholds and spontaneous locomotor activity.Neurosci. Lett.,148, 76–80.

Bianchi, M., Sotgiu, M.L., Manfredi, B. & Sacerdote, P. (1996b) Peripheralmononeuropathy affects hypothalamic and splenocyte beta-endorphin levelsbut not immune function in the rat.Brain Res. Bull., 40, 269–272.

Bradford, M.M. (1976) A rapid and sensitive method for the quantificationof microgram quantities of protein utilizing the principle of protein-dyebinding.Anal. Biochem., 72, 248–254.

Carta, F., Bianchi, M., Argenton, S., Cervi, D., Marolla, G., Tamburini, M.,Breda, M., Fantoni, A. & Panerai, A.E. (1990) Effect of nifedipine onmorphine-induced analgesia.Anesth. Analg., 70, 443–498.

Chai, Z., Gatti, S., Toniatti, C., Poli, V. & Bartfai, T. (1996) Interleukin (IL)26 gene expression in the central nervous system is necessary for feverresponse to lipopolysaccharide or IL-1β: a study on IL-6-deficient mice.J. Exp. Med., 183, 311–316.

Cressman, D.R., Greenbaum, L.E., De Angelis, R.A., Ciliberto, G., Furth,E.E., Poli, V. & Taub, R. (1996) Liver failure and defective hepatocyteregeneration in interleukin-6-deficient mice.Science, 274, 1379–1383.

Eddy, N.B., Touchberry, C.F. & Lieberman, J.E. (1950) Synthetic analgesics.I Methadone isomers and derivatives.J. Pharmacol. Exp. Ther., 98, 121–137.

Eghbali, M., Santoro, C., Paredes, W., Gardner, E.L. & Zukin, R.S. (1987)Visualization of multiple opioid-receptor types in rat striatum after specificmesencephalic lesions.Proc. Natl Acad. Sci. USA, 84, 6582–6586.

Fann, M.J. & Patterson, P.H. (1994) Neuropoietic cytokines and activin Adifferentially regulate the phenotype of cultured sympathetic neurons.Proc.Natl Acad. Sci. USA, 91, 43–47.


Fattori, E., Cappelletti, M., Cota, P., Sellitto, C., Cantoni, L., Carelli, M.,Faggioni, R., Fantuzzi, G., Ghezzi, P. & Poli, V. (1994) Defectiveinflammatory response in IL-6 deficient mice.J. Exp. Med., 180, 1243–1250.

Fattori, E., Lazzaro, D., Musiani, P., Modesti, A., Alonzi, T. & Ciliberto, G.(1995) IL-6 expression in neurons of transgenic mice causes reactiveastrocytosis and increase in ramified microglial cells but no neuronaldamage.Eur. J. Neurosci., 7, 2441–2449.

Ferreira, S.H., Lorenzetti, B.B. & Poole, S. (1993) Bradykinin initiatescytokine mediated inflammatory hyperalgesiaBr. J. Pharmacol., 110,1227–1231.

Glavin, G.B., Par (, W.P., Sandbak, T., Bakke, H. & Murison, R. (1994)Restraint stress in biomedical research: an update.Neurosci. Biobehav. Rev.,18, 223–249.

Gruol, D.L. & Nelson, T.E. (1997) Physiological and pathological roles ofinterleukin-6 in the central nervous system.Mol. Neurobiol., 15, 307–339.

Holliday, J., Parson, K.L., Lee, S.Y., Curry, J.G. & Gruol, D.L. (1995)Cerebellar granule neurons develop elevated calcium responses when treatedwith interleukin-6 in culture.Brain Res., 673, 141–148.

Jacquet, Y.F. (1978) Opiate effects after adrenocorticotropin or beta-endorphininjection in the periaqueductal gray matter of rats.Science, 201, 1032–1034.

Kelly, S. & Franklin, K. (1987) Role of peripheral and central opioid activityin analgesia induced by restraint stress.Life Sci., 41, 789–794.

Kishimoto, T., Akira, S. & Taga, T. (1992) Interleukin-6 and its receptor: aparadigm for cytokines.Science, 258, 593–597.

Kopf, M., Baumann, H., Freer, G., Freudenberg, M., Lambers, M., Klahimoto,T., Zinkernagel, R., Bluethmann, H. & K.(hler, G. (1994) Impaired immuneand acute phase responses in interleukin-6-deficient mice.Nature, 368,339–342.

Kozak, W., Poli, V., Soszynsky, D., Conn, C.A., Leon, L.R. & Kluger, M.J.(1997) Sickness behavior in mice deficient in interleukin-6 during turpentineabscess and influenza pneumonitisAm. J. Physiol., 272, R621–R630.

Kuzmin, A.V., Patkina, N.A. & Zvartau, E.E. (1994) Analgesic and reinforcingeffects of morphine in mice. Influence of Bay K-8644 and nimodipine.Brain Res., 652, 1–8.

Lattanzio, G., Libert, C., Aquilina, M., Cappelletti, M., Ciliberto, G., Musiani,P. & Poli, V. (1997) Defective development of pristane-oil-induced-plasmacytoma in interleukin-6-deficient balb/c mice.Am. J. Pathol., 151,689–696.

Lewis, J.W., Cannon, J.T. & Liebeskind, J.C. (1980) Opioid and non opioidmechanisms of stress-analgesia.Science, 208, 623–625.

Lowry, O.H., Rosenbrough, N.J., Farr, A.L. & Randall, R.J. (1951) Proteinmeasurement with the folin phenol reagent.J. Biol. Chem., 193, 265–268.

Maggi, R., Pimpinelli, F., Casulari, L.A., Piva, F. & Martini, L. (1996)Antiprogestins inhibit the binding of opioids to mu receptors in nervousmembrane preparations.Eur. J. Pharmacol., 301, 169–177.

Manfredi, B., Sacerdote, P., Gaspani, L., Poli, V. & Panerai, A.E. (1998) IL-6knock out mice show modified basal immune responses, but normal immuneresponses to stress.Brain Behav. Immun., 12, 201–211.

Marz, P., Heng, G. Jr, Gadient, R.A., Patterson, P.H., Stoyan, T.T., Otten, U.& Rose-John, S. (1998) Sympathetic neurons can produce and respond toIL-6. Proc. Natl Acad. Sci. USA, 95, 3251–3256.

Meglio, M., Hosobuchi, Y., Lohm, H.H., Adams, J.E. & Li, C.H. (1977) Beta-


endorphin: behavioral and analgesic activity in cats. Proc. Natl Acad. Sci.USA, 74, 774–776.

Mehler, M.F. & Kessler, J.A. (1997) Hematolymphopoietic and inflammatorycytokines in neural development.Trends Neurosci., 20, 357–365.

Merrill, J.E. (1992) Tumor necrosis factor alpha, interleukin-1 and relatedcytokines in brain development: normal and pathological.Dev. Neurosci.,14, 1–10.

Munson, P.J. & Rodbard, D. (1980) LIGAND: a versatile computerizedapproach for characterization of ligand-binding system,Anal. Biochem.,107, 220–239.

North, R.A. (1986) Opioid receptor types and membranes on ion channels.Trends Neurosci., 9, 174–176.

Oka, T., Aou, S. & Hori, T. (1993) Intracerebroventricular injection ofInterleukin-1 (induces hyperalgesia in rats.Brain Res., 624, 61–68.

Panerai, A.E., Sacerdote, P., Brini, A., Bianchi, M. & Mantegazza, P. (1988)Central nervous system neuropeptides after peripheral nerve deafferentation.Peptides, 9, 319–324.

Panerai, A.E., Sacerdote, P., Brini, A. & Mantegazza, P. (1987) Analgesiceffect of morphine: a role for (β-endorphin.Neurosci. Lett., 74, 348–352.

Poli, V., Balena, R., Fattori, E., Markatos, A., Yamamoto, M., Tanaka, H.,Ciliberto, G., Rodan, G. & Constantini, F. (1994) IL-6 deficient mice areprotected from bone loss caused by estrogen depletion.EMBO J., 13,1189–1196.

Porro, C.A. & Carli, G. (1988) Immobilization and restraint effects on painreaction in animals.Pain, 32, 289–307.

Qiu, Z., Parsons, K., l. & Gruol, D.L. (1995) Interleukin-6 selectively enhancesthe intracellular calcium response to NMDA in developing CNS neurons.J. Neurosci., 15, 6688–6699.

Robertson, E., Bradley, A., Kuehn, M. & Evans, M. (1986) Germ-linetransmission of genes introduced into cultured pluripotent cells by retroviralvector.Nature, 323, 445–448.

Romano, M., Sironi, M., Toniatti, C., Polentarutti, N., Fruscella, P., Ghezzi,P., Faggioni, R., Luini, W., van Hinsbergh, V., Sozzani, . S., Bussolino, F.,Poli, V., Ciliberto, G. & Mantovani, A. (1997) Role of IL-6 ant its solublereceptor in induction of chemokines and leukocyte recruitment.Immunity,6, 315–325.

Rothwell, N.J. (1991) Functions and mechanism of IL-1 in the brain.TrendsPharmacol. Sci., 12, 430–436.

Sawada, M., Itohm, Y., Suzumura, A. & Marunouchi, T. (1993) Expressionof cytokine receptors in cultured neuronal and glial cells.Neurosci. Lett.,160, 131–134.

Smith, F.L. & Stevens, D.L. (1995) Calcium modulation of morphine analgesia:role of calcium channels and intracellular pool calcium.J. Pharmacol. Exp.Ther., 272, 290–299.

Van Snick, J. (1990) IL-6: an overview.Annu. Rev. Immunol., 8, 253–278.Wolleman, M. (1990) Recent developments in the research of opioid receptor

sub type molecular characterization.J. Neurochem., 54, 1095–1101.Xu, X., Hao, J., Andell-Jonsson, S., Poli, V., Bartfai, T. & Wiesenfeld-Hallin,

Z. (1997) Nociceptive responses in interleukin-6-deficient mice to peripheralinflammation and peripheral section.Cytokine, 9, 1028–1033.

Zalcman, S., Green-Johnson, J.M., Murray, L., Nance, D.M., Dyck, D.,Anisman, H. & Greenberg, A.H. (1994) Cytokine-specific centralmonoamine alterations induced by interleukin-1–2 and 6.Brain Res., 43,40–49.

Presence of a reduced opioid response in interleukin-6 knock out mice

Documents

Transcript of Presence of a reduced opioid response in interleukin-6 knock out mice