Journal of Cerebral Blood Flow and Metabolism 20:956-966 © 2000 The International Society for Cerebral Blood Flow and Metabolism Published by Lippincott Williams & Wilkins, Inc., Philadelphia

Ischemia-Induced Interleukin-6 as a Potential Endogenous

Neuroprotective Cytokine Against NMDA Receptor-Mediated

Excitoxicity in the Brain

Carine Ali, Olivier Nicole, Fabian Docagne, Sylvain Lesne, Eric T. MacKenzie, Andre Nouvelot,

Alain Buisson, and Denis Vivien

Universite de Caen, UMR-CNRS 6551, IFR 47, Caen Cedex, France

Summary: In the brain, the expression of the pleiotropic cy­tokine interleukin-6 (IL-6) is enhanced in various chronic or acute central nervous system disorders. However, the signifi­cance of IL-6 production in such neuropathologic states re­mains controversial. The present study investigated the role of IL-6 after cerebral ischemia. First, the authors showed that focal cerebral ischemia in rats early up-regulated the expression of IL-6 mRNA, without affecting the transcription of its recep­tors (IL-6Ra and gpI30). Similarly, the striatal injection of N-methyl-D-aspartate (NMDA) in rats, a paradigm of excito­toxic injury, activated the expression of IL-6 mRNA. The in­volvement of glutamatergic receptor activation was further in­vestigated by incubating cortical neurons with NMDA or a -amino-3-hydroxy -5-meth y 1-4-isoxazolepropionate (AMP A). NMDA and ionomycin (a calcium ionophore) up-regulated IL-6 mRNA, suggesting that neurons may produce IL-6 in

Increasing evidence suggests that, in addition to apop­totic and excitotoxic pathways (Choi, 1996), the immune response to tissue damage may also significantly influ­ence the neuronal outcome after ischemic brain injury, For example, the striatal injection of the proinflamma­tory cytokine interleukin-l � (IL-l �) worsens ischemic brain lesions (Stroemer and Rothwell, 1998), whereas the intracerebral injection of the immunosuppresive cy­

tokine transforming growth factor-� (TGF-�) is neuro­protective in experimental cerebral ischemia (Prehn et aI., 1993), Although the involvement of immunel inflammatory cytokines, such as IL-1�, TNF-a or TGF­

�, in the ischemic neuronal death has been extensively

Received December 16, 1999; final revision received March 8, 2000; accepted March 10, 2000.

Supported by grants from the French Ministry of Higher Education, Research and Technology, and the CEA.

Address correspondence and reprint requests to Drs. A. Buisson, and D. Vivien, Laboratoire de Neurosciences, UMR-CNRS 6551, Bd. H. Becquerel, BP 5229, 14074 CAEN Cedex, France.

956

response to the calcium influx mediated through NMDA re­ceptors. The potential role of IL-6 during ischemic/excitotoxic insults was then studied by testing the effect of IL-6 against apoptotic or excitotoxic challenges in cortical cultures. IL-6 did not prevent serum deprivation- or staurosporine-induced apop­totic neuronal death, or AMPAlkainate-mediated excitotoxic­ity. However, in both mixed and pure neuronal cultures, IL-6 dose-dependently protected neurons against NMDA toxicity. This effect was blocked by a competitive inhibitor of IL-6. Overall, the results suggest that the up-regUlation of IL-6 in­duced by cerebral ischemia could represent an endogenous neu­roprotective mechanism against NMDA receptor-mediated in­jury. Key Words: Interleukin-6-Focal cerebral ischemia­Neuroprotection-Excitotoxicity-Primary cortical neuronal culture.

studied (Loddick et aI., 1997; Feuerstein et aI., 1998;

DeGraba, 1998; Buisson et aI., 1998), the role of IL-6

remains to be elucidated,

First identified as a B-cell stimulating factor (Hirano

et aI., 1986), IL-6 belongs to a subfamily of related he­

matopoietic cytokines, including leukemia inhibitory

factor, ciliary neurotrophic factor, oncostatin-M, cardio­

trophin-1 and IL-11, To exert its biological effects, IL-6

first binds to either the soluble or the membrane­associated receptor IL-6Ra. This complex then recruits the signal transducer gp 130, a transmembrane glycopro­

tein common to the IL-6 related cytokines, as an obliga­tory component for signal transduction. The subsequent dimerization of gp130 activates the Jak/STAT and MAPK pathways, allowing the transcriptional modula­tion of target genes (Akira, 1997; Heinrich et aI., 1998;

Hirano, 1998). The expression of IL-6 in the brain increases in nu­

merous neurological disorders, such as Alzheimer's dis­ease, Parkinson's disease, meningitis, trauma, and stroke

IL-6 PROTECTS NEURONS AGAINST NMDA TOXICITY 957

(Gruol and Nelson, 1997; Benveniste, 1998). IL-6 has been reported to be up-regulated after focal permanent

ischemia in rats (Wang et aI., 1995) and transient global ischemia in gerbils (Saito et aI., 1996). However, con­

flicting data have been reported with regard to the po­tential role of IL-6 production during both acute and chronic insults to the brain. The intracerebral injection of IL-6 in rats reduces the infarcted volume after focal per­manent cerebral ischemia (Loddick et aI., 1998) and pro­tects striatal cholinergic neurons against N-methyl-D­aspartate (NMDA)-induced toxicity (Tou1mond et aI., 1992). However, transgenic mice that overexpress IL-6

in astrocytes exhibit learning disabilities and prominent

neurodegeneration (Campbell, 1998). In vitro, IL-6 has been shown to exert a neurotrophic activity in forebrain neurons (Kushima et aI., 1992) and to protect PC12 cells against an hypoxia-reoxygenation paradigm (Maeda et aI., 1994). While some data indicate that IL-6 reduces the glutamate-induced neuronal death in hippocampal cul­tures (Yamada and Hatanaka, 1994), other authors have

evidenced either no benefit of IL-6 against excitotoxicity in cortical neurons (Toulmond et aI., 1992) or an en­hancement of NMDA toxicity in cerebellar granule neu­

rons chronically exposed to IL-6 (Qiu et aI., 1998). The aim of the present investigation was to assess the

possible involvement of IL-6 in cerebral ischemia. First, the expression of IL-6 and its receptors after permanent or transient focal cerebral ischemia in rats was examined.

Then, the expression of IL-6 and its signalling receptors after the exposure of brain tissue to NMDA both in vivo

and in vitro was investigated. Finally, the authors char­

acterized the effect of IL-6 against the neuronal death induced by either excitotoxic or apoptotic paradigms in murine cortical cultures.

MATERIALS AND METHODS

Polymerase chain reaction and reverse transcriptase system kits were purchased from Promega (Charbonnieres, France). Dulbecco's Eagle's minimal essential medium, Poly-D-Iysine, cytosine f3-D-arabinoside (AraC), horse serum, fetal calf se­rum, Tween 20, ionomycin (A23187), staurosporine, Extravi­din, MAP2 and glial fibrillary acidic protein (GFAP) antibodies and the synthetic IL-6 competitive inhibitor were obtained from Sigma Chemical Co. (L'Isle D' Abeau, France). NMDA, a-amino-3-hydroxy -5-methy 1-4-isoxazolepropionate (AMP A), 6-cyano-7 -nitroquinoxaline-2,3-dione, kainate and (+)5-methyl-IO, II-dihydro-5H-dibenzo(a,d)cyclohepten-5, 10-imine maleate (MK-80l) were from Tocris (Bristol, U.K.). Laminin was from Life Technologies (Cergy Pontoise, France). Mouse recombinant IL-6 (specific activity: 105 U/f.Lg) was purchased from R&D Systems (Oxford, U.K.). Antibodies raised against IL-6Ra and gpl30 were from Santa Cruz (Heidelberg, Germany).

In vivo procedures All experiments were performed on male Sprague-Dawley

rats weighing 280 to 320g (CERJ, France). Anesthesia, induced by halothane 4% and maintained at 1-1.5%, was administrated through a vaporizer in a gas mixture of 30% O2:70% N20.

Middle cerebral artery occlusion. All of the rats used for ischemia procedures were intubated and a femoral artery was cannulated for continuous arterial pressure monitoring and gas analysis. Body temperature was maintained at 37.5°C ± 0.2°C with a heating pad. Focal cerebral ischemia was induced by intraluminal middle cerebral artery occlusion (MCAO) (Long a et ai, 1989). Briefly, a midline neck incision was made and the right common carotid artery was exposed. After coagulation of its branches, the external carotid artery was severed distally. A nylon thread (0.18mm in diameter) having a distal cylinder (3mm long and 0.38mm in diameter) of thermofusible glue (3M:3764; Radiospares composantes, Paris, France) was in­serted in the lumen of the external carotid artery and pushed into the internal carotid artery up to the origin of MCA. In the case of transient focal ischemia, the nylon thread was removed and cut 90 minutes later, so as to restore blood flow in the MCA. Rats were killed 3, 6, 24 and n hours after MCAO, and cortices were isolated for reverse transcription polymerase chain reaction (RT-PCR) analysis.

Excitotoxic lesions. Rats received a unilateral injection (3 f.LL) into the left striatum (coordinates 0.2 mm posterior, 3 mm lateral, 5.5 mm ventral to the bregma; Paxinos and Watson, 1986) with either vehicle (phosphate-buffered saline [PBS], pH 7.4) or NMDA (75 nmol in PBS, pH 7.4). Five minutes after insertion of the needle, the solution was injected over the next 6 minutes using a Hamilton syringe pump (Carnegie, Stock­holm, Sweden) at a rate of 0.5 f.LLlmin. The needle was re­moved 5 minutes later. Rats were killed after I, 6 and 24 hours and the striata were collected for RT-PCR analysis.

Cell culture. Mixed cortical neuron and glial cell cultures were prepared from fetal OFI mice at 15 to 16 days gestation (Rose et aI., 1993). After removing the meninges, isolated ce­rebral cortices were mechanically dissociated in media stock (MS, minimal essential medium with 25 mmollL glucose). The cell suspension was adjusted with MS supplemented with 2 mmollL glutamine, 5% fetal calf serum, and 5% horse serum, before plating on a preformed confluent layer of astrocytes. The latter (containing no neurons) was derived from 1- to 3-day-old mice following the same procedure, with the exception that the culture medium was MS supplemented with 2 mmollL gluta­mine, 10% fetal calf serum and 10% horse serum.

Near pure neuronal cultures containing <5% glial cells were obtained as described above, but were plated on a poly-D­lysine and laminin layer.

All cultures were kept at 37°C in a humidified 5% CO2 atmosphere. Non- neuronal cells proliferation was halted by adding 10 f.LmollL AraC after 3 days in vitro for near pure neuronal cultures and 7 days in vitro for mixed cultures. Three days before the experiments, culture media were exchanged for MS +10 f.LmollL glycine.

Semiquantitative reverse transcription polymerase

chain reaction Total RNAs were isolated from rats cerebral cortices and

striata or murine cell cultures through the use of the RNaxel kit (Eurobio, Les Ulis, France), or RNeasy kit (Qiagen. Courta­boeuf, France), respectively. One f.Lg of total RNAs was reverse transcribed into cDNA using poly(dT) oligonucleotides that allow a specific targeting of poly(A +)mRNA. For semiquanti­tative experiments with f3-actin as a housekeeping gene, an aliquot of the cDNA library (I f.LL) was amplified by PCR with specific sets of oligonucleotides (Table I). Amplification con­ditions using a Thermolyne thermocycler (Amplitron II; Barn­stead, Dubuque, lA, U.S.A.) were the following: 95°C-30 sec­onds, 55°C-30 seconds, and noc-I minute for f3-actin, IL-6 and gp 130; and 95°C-30 seconds, 58°C-30 seconds, and

J Cereb Blood Flow Metab, Vol. 20, No.6, 2000

958 C. ALI ET AL.

TABLE 1. peR oligonucleotides used for IL-6, IL-6Ra, gp J 30, and j3-actin

Gene Oligonucleotide sequence (5'-3') Product size

IL-6 ATGAAGTTCCTCTCTGCAAGA 638 bps CACTAGGTTTGCCGAGTAGAT

IL-6 Ra CCTGCTTCCGGAAGAACCCC 320 bps TGATACCACAAGGTTGGCAG

gp 130 TGAAGCTGTCTTAGCGTGGG 405 bps GGTGACCACTGGGCAATATG

I3-Actin GTGGGCCGCTCTAGGCACAA 540 bps CTCTTTGATGTCACGCACGATTTC

PCR product sizes are expressed in base pairs (bps). For each set of primers, upper and lower sequences respectively stand for sense and antisense probes. PCR, polymerase chain reaction.

nOc-I minute for IL-6Ra. The number of cycles was chosen corresponding to 50% of the saturation curve. Amplified prod­ucts were separated by agarose gel electrophoresis and visual­ized by ethidium bromide staining. Analysis was based on vi­sual inspection of the agarose gel electrophoresis.

Immunocytochemistry Murine cortical cultures were fixed with 4% paraformalde­

hyde and incubated overnight with the primary antibody raised against either glial fibrillary acidic protein, microtubul associ­ated protein-2 (GFAP, 1:5000 and MAP2, I :200 in PBS plus I % bovine serum albumin plus 0.1 % Tween 20) or IL-6 recep­tors (lL-6Ra and gp 130, I :200 in PBS plus I % bovine serum albumin). Cells were then washed and incubated for I hour with the appropriate secondary biotin-conjugated antibody. An­tibody-antigen complexes were amplified with avidin (Vec­tastain ABC kit; Vector Laboratories, Burlingame, California) and revealed by H202-peroxydase reaction.

SDS·PAGE and Western Blotting Murine cultured cortical neurons and astrocytes were lysed

in Tris-NaCI-Triton buffer and centrifuged for 5 minutes (2500 g) to obtain whole cell extracts. Then, sodium dodecyl sulfate­polyacrylamide gel electrophoresis was performed before im­mobilization on a polyvinyl difluoride membrane. Western blotting were exposed for I hour to the primary antibodies (I: 1500), I hour to the appropriate secondary biotin-conjugated antibody (I: 1000), and for I hour to Extravidin (I :2000) before revelation using a chemiluminescence kit (NEN, Paris, France).

In vitro toxicity experiments Excitotoxicity. Slowly triggered excitotoxicity was induced

at 37°C in mixed neuron-glia or pure neuronal cortical cultures (DIV 14) by a 24-hour exposure to 12.5 f,LmollL NMDA, 10 f,LmollL AMP A or 50 f,LmollL kainate in MS supplemented with glycine (Choi, 1992). MK-80l (10 f,LmollL) was always added concurrently with AMPA or kainate to block secondary activation of NMDA receptors. Excitotoxins were applied alone or concurrently with mouse recombinant IL-6 (mrIL-6).

Neuronal death was estimated by examining the cultures under phase-contrast microscopy and quantitated by measuring lactate dehydrogenase (LDH) release from damaged cells into the bathing medium one day after the onset of exposure to the appropriate excitotoxin (Koh and Choi, 1987). The LDH level corresponding to complete neuronal death (without glial death) was determined in parallel dishes exposed to 200 f,Lmol/L NMDA. Background LDH levels were determined in sister cultures subjected to sham wash and subtracted from experi­mental values to yield the signal specific for experimentally­induced injury.

J Cereb Blood Flow Me/ab. Vol. 20. No.6. 2000

Apoptosis. Serum deprivation was initiated by transferring near pure neuronal cultures (DIV 7) for 24 hours into growth medium lacking serum (Martin et aI., 1988) in the presence or absence of mrIL-6. Secondary NMDA receptor activation was blocked by the addition of MK-80l to the bathing medium. Neuronal death was assessed by phase-contrast cell counting after staining with 0.4% trypan blue dye. The percentage of neuronal death was estimated as the ratio of trypan blue posi­tive neurons after serum deprivation to the total neurons labeled with the same dye after paraformaldehyde permeation.

Staurosporine exposure was performed at 37°C for 24 hours by transferring mixed cortical cell cultures in MS supplemented with glycine, containing 200 nmollL of staurosporine (Koh et aI., 1995) with or without mrIL-6. MK-801 was added concur­rently to staurosporine to block secondary NMDA receptor activation. Neuronal death was assessed by the measurement of LDH release.

Statistical analysis Results are expressed as mean ± SD. Statistical analysis

consisted in one-way variance analysis, followed by Bonfer­ronni-Dunn's test.

RESULTS

Specificity of the PCR products

Oligonucleotide probes were designed from the pub­

lished cDNA sequence of IL-6 (Chiu et ai., 1988), IL-

6Ra (Sugita et ai., 1990) and gp 130 (Saito et ai., 1992).

Each set of oligonucleotides used for semiquantitative

RT-PCR studies gave the PCR product of the expected

size (638 bps for IL-6, 320 bps for IL-6Ra and 405 bps

for gp130), as illustrated in Fig. 1. The specificity of the

PCR products was confirmed by restriction digest con­

trols. IL-6 and IL-6Ra PCR products were digested us­

ing Alul endonuclease, whereas the gp130 PCR product

was submitted to EcoRI digestion. Fig. I shows that

restriction enzymes generated the expected fragments of

365, 166, and 107 bps for IL-6 (Fig. 1A); 155 and 111

bps for IL-6Ra (two expected fragments of 46 and 8 bps

were not visualized because of their small size) (Fig.

IB); 215 and 190 bps for gp130 (Fig. IC).

As predicted by the high degree of homology between

the rat and the mouse cDNA sequences encoding for IL-6 and its receptors, similar results were obtained with cDNAs derived from rat cerebral cortex (data not shown).

Transcriptional expression of IL·6 and its receptors

after permanent or transient MCAO in rats

Focal cerebral ischemia was induced by transient (90 minutes) or permanent intraluminal occlusion of the

middle cerebral artery. In both cases, MCAO induced,

after 24 hours a severe lesion affecting both the striatum and the cortex, and representing approximately 40% ( =

200 mm3) of the total volume of the hemisphere.

IL-6 PROTECTS NEURONS AGAINST NMDA TOXICITY 959

IL-6

�638 - 365

200 bps-_ 166

100 bps- - 107 A ladder ND AluI

IL-6Ra.

400 bps-�320

200 bps-- 1 5 5

100 bps - - 1 I 1 B ladder ND AluI

gp130

400 bps-

200 bps-

100 bps-

�405 _215 -190

C ladder ND EcoRI FIG. 1. Specificity of the polymerase chain reaction (PCR) prod­ucts. Total RNAs were extracted from murine cerebral cortex and reverse transcribed using a poly(dT) primer. PCR amplifica­tion was then performed using specific sets of primers (Table 1). Agarose gel electrophoresis of the IL-6 PCR product digested or not digested by Alul (A), the IL-6Ra PCR product digested or not di­gested by Alul (8), and the gp130 PCR product digested or not digested by EcoRI (C). NO, not digested.

Total RNAs from ipsi- and contra-lateral cortices were harvested at the indicated times after permanent or tran­sient MCAO. RNA samples (l J..lg) were then reverse transcribed with poly-dT oligonucleotides. An aliquot of the cDNA libraries (l J..ll) was amplified by peR with specific sets of oligonucleotides for [3-actin, IL-6, IL-6Ra or gp130. The expression pattern of IL-6 and its

related receptors in a model of permanent cerebral isch­

emia was then studied. As shown in Fig. 2A, all PCR reactions showed the same level of the chosen house­keeping gene ([3-actin). The IL-6 cDNA product was not detected in ipsi- and contra- lateral cortices obtained 6 hours after the intervention in sham rats. However, in the injured cortex the expression of IL-6 mRNA enhanced at 3 hours, peaked at 24 hours, and returned to baseline levels after 3 days. In contrast to IL-6, the expression of IL-6Ra and gp130 receptors mRNAs remained un­changed. The expression of IL-6 mRNA in a model of 90

minutes transient cerebral ischemia displayed a pattern slightly delayed when compared to the model of perma­nent cerebral ischemia (Fig. 2B).

Permanent ischemia

3h 6h 24h 72h sham 6h

IL-6

IL-6Ra ... C I

gp130 ...

P.actin ...

A C I C I C I C I Transient ischemia

P.actin ...

B

3h 6h 24h 72h

C I C I C I C J FIG. 2. Transcriptional expression of IL-6 and its receptors after permanent or transient middle cerebral artery (MCAD) occlusion in rats. (A) Total RNAs were extracted from rat contra-(C) and ipsi-(I) lateral cortices at the times indicated after permanent middle cerebral artery occulsion and analyzed by reverse tran­scription polymerase chain reaction (RT-PCR) for IL-6, IL-6Ra, gp130 and l3-actin. (8) Total RNAs were isolated from rats con­tra-(C) and ipsi-(I) lateral cortices at the times indicated after transient MCAD (90 minutes) and analyzed by RT-PCR for IL-6 and l3-actin. Sham, sham operated animal. Presented electro­phoresis are representative of three independent experiments.

After both transient and permanent focal ischemia, a transient increase in IL-6 mRNA levels was observed in the contralateral cortex. As previously suggested, this increase could result from the physical deformation of

the contralateral hemisphere induced by the swelling in the injured ischemic side (Lad dick et a!., 1998).

JL-6 ...

IL-6 Ra ...

gp130 ...

P.actin ... c

Ih

I c

6h 24h PBS3h

I c I c I

FIG. 3. Dverexpression of IL-6 mRNA after the intrastriatal in­jection of N-methyl-D-aspartate (NMOA). Total RNAs were har­vested from the contra-(C) and ipsi-(I) lateral striata of rats at the times indicated after unilateral injection of NMOA or vehicule (PBS) and then analyzed by reverse transcription polymerase chain reaction for IL-6, IL-6Ra, gp130 and l3-actin.

J Cereh Blood Flo\\' Me/ab. Vol. 20. No.6. 2000

960 C. ALI ET AL.

A

IL-6 -+

IL-6Ra-+

gpJ30 -+

f3-actin -+

AMPA

CNQX

AMPA C O.Sh Ih 3h

+ + +

Ih 3h

+ +

+ +

B

NMDA

MK-801

NMDA C O.Sh Ih 3h

+ + +

Ih 3h

+ +

+ +

FIG. 4. The application of N-methyl-D-aspartate (NMDA), but not a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate (AMPA), induces an overexpression of IL-6 in cultured cortical neurons. Pure cortical neurons cultures were incubated for the indicated times either with AMPA (50 IJmol/L) plus MK-801 (10 IJmol/L) alone or in combination with 6-cyano-7-nitroquinoxaline-2,3-dione (30 IJmol/L) (A), or with NMDA alone (50 IJmol/L) or in combination with MK-801 (10 IJmollL) (8). Total RNAs were then isolated and submitted to reverse transcription polymerase chain reaction for IL-6, IL-6Ra, gp130 and [3-actin mRNAs. Presented electrophoresis are representative of three independent experiments.

Overexpression of IL-6 mRNA after the

intrastriatal injection of NMDA

Histologic analysis revealed that, whereas the striatal injection of vehicle (PBS) alone did not induce any sig­nificant lesion, rats injected with NMDA (75 nmol/L in PBS) exhibited a lesion volume of 27.0 ± 2.5 mm3 (mean ± SD, n = 8) (Ruocco et aL, 1999).

Total RNAs (l I-lg) harvested from the control and injected rat striata were submitted to reverse transcrip­tion using poly-dT oligonucleotides and peR was per­formed for IL-6, IL-6Ra, gp130, and l3-actin mRNAs, as described above. All peR reactions showed the same level of expression of the chosen housekeeping gene (13-actin) (Fig. 3).

Figure 3 shows that, in contrast to the transcription of both IL-6 related receptors (IL-6Ra and gp130), the ex­pression of IL-6 mRNA after striatal injection of NMDA displayed a marked overexpression after 6 hours, that

returned to control levels after 24 hours. As observed after focal ischemia (Fig. 2), a minor increase in IL-6

mRNA was detected in the contralateral striatum at 1 and 6 hours. After 3 hours, the injection of the vehicule (PBS) into the striatum induced only minor modifica­tions in IL-6 mRNA levels.

NMDA treatment induces the overexpression of

IL-6 mRNA in cultured cortical neurons

As both ischemic and excitotoxic insults to the brain

induce an early up-regulation of IL-6 expression in vivo,

the question arose whether the activation of ionotropic glutamatergic receptors in cultured cortical neurons may

J Cereb Blood Flow Metah, Vol. 20, No.6, 2000

also induce an increase of this cytokine. By using semi­quantitative RT -peR, the authors studied the transcrip­tional expression of IL-6 in cultured neurons after expo­sure to AMPA or NMDA. Near pure neuronal cultures (DIV 14), obtained from the cortex of mice were exposed

for a short period (less than 3 hours) to high concentra­tions (50 I-lmollL) of glutamatergic agonists (the AMPA

exposure was performed in the presence of 10 I-lmollL MK-801 to prevent secondary NMDA receptor activa­tion), These applications induced an acute swelling of neuronal bodies but did not alter the membrane per­meability, as estimated by the LDH release into the bath­ing medium (data not shown). As shown in Fig, 4A,

C O.Sh Ih 3h

IL-6 -+-

f3-actin -+-

/onomycin + + + FIG. 5. lonomycin treatment induces an overexpression of IL-6 mRNA in cultured cortical neurons. Pure cortical neuron cultures were incubated for the indicated times with ionomycin (5 IJmol/L) and MK-801 (10 IJmoI/L). Total RNAs were then isolated and submitted to reverse transcription polymerase chain reaction for IL-6 and [3-actin mRNAs. Presented electrophoresis is represen­tative of three independent experiments.

IL-6 PROTECTS NEURONS AGAINST NMDA TOXICITY 961

A Neurons Astrocytes B Neurons Astrocytes MW 84kDa

IL-6Ra-+

- 210kDa gpJ30

-+ _�ji1i gp130

-127kDa FIG. 6. Expression of IL-6 related re­ceptors in cultured cortical neurons and astrocytes. The cellular expres­sion of IL-6Ra and gp130 was studied on murine astrocytic or neuronal cor­tical cultures at the mRNA (A) and protein (8 and C) levels by reverse transcription polymerase chain reac­tion, Western blotting and immunocy­tochemistry, respectively. Cellular marker: glial fibrillary acidic protein, for cultured astrocytes; and MAP-2, for cultured neurons. MW, molecular weight. Scale bar = 100 j.Jm.

c IL-6Ra gpJ30 cellular marker

Neurons

Astrocytes

AMP A exposure did not obviously modify the transcrip­tional activity of IL-6 and its related receptors (IL-6RQ' and gp 130) in cultured cortical neurons. The expression of IL-6RQ' and gp130 mRNA was equally unchanged

80

O �--�-U��--L-�-L--���� JL-6 50 50 (ng/mL) - SD Chx -STP'Chx

FIG. 7. IL-6 does not protect neurons against apoptotic cell death. Neuronal death (%) was assessed after 24 hours by trypan blue dye staining (mean ± standard deviation, n = 10), after serum deprivation (SO) in pure neuronal cultures (OIV 7), or by lactate dehydrogenase release (mean ± standard deviation, n = 14), after exposure of mixed cortical cultures (OIV 14) to stau­rosporine 200 nmol/L (STP). SO and STP exposure were per­formed in the presence (shaded bars) or absence (open bars) of mrlL-6 and with or without cycloheximide (Chx) at 1 j.Jg/mL. In both apoptotic paradigms, MK-801 was systematically added to block the secondary activation of N-methyl-D-aspartate recep­tors. 'significantly different from SO or STP by analysis of vari­ance, followed by Bonferronni-Ounn's test (P < 0.01).

after NMDA exposure. However, NMDA treatment en­hanced the transcription of IL-6 in a time-dependent manner. This neuronal overexpression of IL-6 was spe­cifically induced by NMDA, because it was completely blocked in the presence of the NMDA antagonist, MK-

801 (Fig. 4B). After 3 hours, exposure to MK-80 l ( lOf,LmolfL) had no effect on IL-6 transcription as com-

1

.- 80 'te '-' -= � 60 � "CS -; § 40 .. = �

Z 20

IL_60 (ng/mL) -kainate-FIG. 8. IL-6 fails to modify a-amino-3-hydroxy-5-methyl-4-isoxazolepropionate- (AMPA) or kainate-induced excitoxicity in neurons. Neuronal death (%) was estimated by lactate dehydro­genase release (mean ± SO, n = 12), after a 24-hour exposure to AMPA (10 j.Jmol/L) or kainate (50 j.Jmol/L) in mixed neuron-glia cortical cultures in the presence (shaded bars) or not (open bars) of mrlL-6 (50 ng/mL). Secondary activation of N-methyl-D­aspartate receptors was blocked by the co-application of MK-801 (10 j.JmoI/L).

J Cereb Blood Flow Me/ab, Vol. 20. No.6. 2000

962 C. ALI ET AL.

pared to control cultures (data not shown). The authors thus postulated that this induction of IL-6 mRNA could

result from the calcium influx mediated through the ac­tivation of NMDA receptors. The authors tested this hy­

pothesis by exposing near pure neuronal cultures to the calcium ionophore, ionomycin (5j-LmollL in the presence of lOj-LmollL MK-801 to prevent excessive NMDA re­ceptor activation). The addition of ionomycin induced an up-regulation of IL-6 mRNA over the same interval of time as seen after the application of NMDA (Fig. 5).

Expression of IL-6 related receptors in cultured

cortical neurons and astrocytes

The cellular expression of IL-6 related receptors in

murine cortical cultures was determined at both the mRNA and protein levels. RT-PCR studies revealed that both cultured cortical neurons (DIV 14) and astrocytes exhibited mRNAs for IL-6Ra and gp130 (Fig. 6A). Western blotting analysis performed with an antibody raised against IL-6Ra (80 kDa) or gp130 (130 kDa) re­

vealed the presence of a single band at the expected molecular weight (Fig. 6B); thus demonstrating that neu­rons and astrocytes in culture express both IL-6 related receptors, as described in the rodent adult brain (Schobitz et aI., 1993; Gadient and Otten, 1994; Watanabe et aI., 1996). The expression of IL-6Ra and gp 130 in cortical

A

1 NS 2 Control

3 NMDA

B 100

1L-6 -

(nglmL)

astrocytes and neurons was confirmed by immunocyto­

chemistry (Fig. 6C).

IL-6 is neuroprotective against NMDA-induced

neuronal death

To test whether IL-6 might alter neuronal outcome

after an ischemic insult, the authors determined the ef­fects of mouse recombinant IL-6 (mrIL-6) on murine

cortical cell cultures exposed to apoptotic or necrotic

paradigms.

To induce apoptosis, pure cortical neuronal cultures

were transferred after 7 days in vitro (DIV 7) to a serum­

deficient medium supplemented with MK-801 (Martin et

aI., 1988), that provokes the degeneration of approxi­mately 50% of neurons over 24 hours. This treatment resulted in three features typical of apoptosis: (1) a gradual shrinkage of the cell body was exhibited by neu­rons; (2) death was blocked by the addition of cyclohex­imide (Chx), a protein synthesis inhibitor; and (3) death was accompanied by the appearance of DNA fragmen­tation (data not shown). The addition of mrIL-6 to the bathing medium (50 ng/ml) failed to modify the extent of neuronal degeneration (Fig. 7) (n = 10). This lack of

effect was not due to the absence of IL-6 receptors in 7-day-old cultured neurons, because both IL-6Ra and

5 25 50 --NMDA--

c 100

80 -� = -

.; 60 *

� Q,j "C

-; � 40 I-= Q,j

Z 20

O'--'--I-� 50 lL-6 - 50 50

(nglmL) -NMDA-FIG. 9. IL-6 is neuroprotective against N-methyl-D-aspartate- (NMDA) induced neuronal death. (A) Micrograph analysis of the neuronal immunostaining for MAP-2 after NMOA application in the presence or absence of mrlL-6 (A1) nonspecific staining of control cultures; (A2) MAP-2 staining in sham-washed cultures; (A3) MAP-2 staining in NMOA-treated cells; (A4) MAP-2 staining in NMOA-treated cells co-incubated with IL-6 50 ng/mL). Scale bar = 100 IJm. (8) Neuronal death (%) was estimated by lactate dehydrogenase (LOH) release (mean ± SO, n = 13), after a 24-hour exposure to NMOA (12.5 IJ mol/L) in mixed neuron-glia cortical cultures in the presence (shaded bars) or absence (open bars) of mrIL-6. ·significantly different from NMOA alone by analysis of variance (ANOVA), followed by Bonfer­ronni-Ounn's test (P < 0.001). (C) Neuronal death (%) was estimated by LOH release (mean ± SO, n = 16), after a 24-hour exposure to NMOA (12.5 IJmol/L) in near pure cortical neuron cultures in the presence (shaded bars) or absence (open bars) of mrlL-6 (50 ng/mL). ·significantly different from NMOA alone by ANOVA, followed by Bonferronni-Ounn's test (P < 0.0001).

J Cereb Blood Flow Metab. Vol. 20. No.6, 2000

IL-6 PROTECTS NEURONS AGAINST NMDA TOXICITY 963

gp130 (mRNAs and proteins) are expressed in these cul­tures (data not shown).

In addition, neuronal apoptosis was also induced by exposing mixed neuron-glia cultures (DIV 14) to stau­

rosporine, a nonspecific protein kinase inhibitor (Koh et aI., 1995). Although cycloheximide blocked neuronal death, the addition of mrIL-6 (50 ng/mL) was without effect (Fig. 7) (n = 14).

Thereafter, the authors determined the influence of IL-6 on excitotoxic necrosis, a process morphologically

distinct from apoptosis and characterized by a prominent and early cell swelling. Necrosis was induced over 24 hours by the following glutamatergic agonists: NMDA

(l2.5/LmoIlL), AMPA (lO/LmoI/L) and kainate (50/Lmoll L) (Choi, 1992). The exposure of mixed cortical neuron­glia cultures (DIV 14) to these agonists produced an acute swelling of neuronal bodies, followed 24 hours later by widespread neuronal degeneration, whereas the glia remained intact. Although mrIL-6 (from 5 to 50 ng/mL) was found ineffective against AMPA- or kain­ate-induced toxicity (Fig. 8) Cn = 12), it significantly reduced NMDA-induced necrotic neuronal death in

mixed neuron-glia cultures, as illustrated by the large

restoration of MAP-2 immunostaining organization (Fig. 9A). The quantification of the neuroprotective effect of IL-6 revealed that it was dose-dependent (Fig. 9B) (n =

13; P < 0.001). The potential role of astrocytes in the neuroprotective activity of IL-6 was investigated by ap­plying NMDA C12.5/Lmol/L) for 24 hours in near pure neuronal cultures (DIV 14). In these cultures, mrIL-6 (50 ng/mL) was able to display a neuroprotective activity against NMDA-induced necrotic death (Fig. 9C) (n =

16; p < 0.0001). To characterize the effect of mrIL-6 on NMDA-induced necrotic neuronal death, the authors tested a polypeptide fragment corresponding to amino acids 88 to 121 of human IL-6 (IL-6 88-121). This pep­tide has previously been characterized as a competitive inhibitor for IL-6 binding to IL-6Ra (Ekida et aI., 1992). Although IL-6 88-121 (2 /Lg/mL) displayed no toxicity in mixed cortical cultures, it prevented the neuroprotec­tive effect of IL-6 against NMDA toxicity (10 /LmollL), as revealed by the increase in the LDH release in the bathing medium 24 hours after the application of the excitotoxin (Fig. 10) (n = 10; P < 0.0001).

DISCUSSION

In the present study, we investigated the implication of IL-6 during cerebral ischemia.

Our observation that the expression of the IL-6 gene is activated early in the injured cortex after focal cerebral

ischemia is consistent with previous experimental data (Wang et aI., 1995; Hill et aI., 1999; Clark et aI., 1999). Although we did not address the protein level, the well­documented increase in IL-6 immunoreactivity (Maeda

et aI., 1994; Saito et aI., 1996; Orzylowska et aI., 1999) and bioactivity (Loddick et aI., 1998) strongly supports the idea that the early occurrence of IL-6 mRNA pro­duction (evidenced as early as 3 hours after occlusion)

could influence the outcome of ischemic insults. Two major additional findings presented here are the follow­ing: (1) the over-activation of NMDA receptors elicits an enhancement of IL-6 transcription in neurons; and 2) IL-6 exerts a neuroprotective effect restricted to NMDA­induced excitotoxic necrosis.

Since the glutamatergic ionotropic receptor-mediated excitotoxic cascade is a major contributory factor for ischemic neuronal death (Choi, 1996), we tested whether

the activation of this pathway could reproduce the effect of cerebral ischemia on the expression of IL-6 mRNA. We evidenced that the injection of NMDA resulted in an enhancement of IL-6 transcription in the striatum similar to that provoked at the cortical level by MCAO. These observations suggest that the excitotoxic pathway could

initiate the synthesis of IL-6 in the ischemic brain. This hypothesis is strengthened by the induction of IL-6 re­

ported after the striatal injection of another glutamatergic agonist, quinolinic acid (Schiefer et aI., 1998). However, because increasing evidence suggests that NMDA may

1

80

-� �60 ..= -�

� "C -; 40 = o a.. =

�

Z 20

#

o J--1_�--'-IL-6

(ng/mL) 50 50

-IL-6 88-121-NMDA--

FIG. 10. IL-6 neuroprotective effect against N-methyl-D­aspartate (NMDA) toxicity is abolished by a synthetic competitive inhibitor. Neuronal death (%) was estimated by lactate dehydro­genase release (mean ± SD, n = 10), after a 24-hour exposure to NMDA (1 0 �mol/L) in mixed neuron-glia cortical cultures in the presence (shaded bars) or absence (open bars) of mrIL-6, and with (hatched bars) or without IL-6 88-121 (2 �g/mL). 'signifi­cantly different from NMDA alone by analysis of variance (ANOVA), followed by Bonferronni-Dunn's test (P < 0.0001). #Significantly different from NMDA+IL-6 by ANOVA, followed by Bonferronni-Dunn's test (P < 0.0001).

J Cereb Blood Flow Metab, Vol. 20, No.6, 2000

964 C. ALI ET AL.

induce both excitotoxic and apoptotic mechanisms (Bon­foco et aI., 1995), and to eliminate any effect resulting from tissue lesion elicited by the striatal injection of

NMDA, we developed an in vitro model allowing the

investigation of the transcriptional modulation(s) medi­ated during a severe glutamatergic ionotropic receptor­mediated excitotoxic stress. In our conditions, the expo­sure of pure cultured cortical neurons to high concentra­tions of glutamatergic agonists, for up to 3 hours, induced the characteristic feature of necrosis (acute swelling of the cell body) without linkage of the neuronal membrane. We showed that cortical neurons are able to

overexpress IL-6 mRNA in response to an intense NMDA receptor activation, an effect that is selectively blocked by MK-801. To investigate further the cellular signal(s) responsible for the increased transcription of IL-6, and because NMDA receptor activation is charac­terized by a raise in the intracellular calcium concentra­tion (MacDermott et aI., 1986), we tested the ability of a selective calcium ionophore, ionomycin (A23187) to

mimic the activation of IL-6 mRNA expression induced

by NMDA. In cultured cortical neurons, ionomycin elic­ited a time-dependent activation of IL-6 transcription,

similar to that resulting from NMDA exposure. Thus, the activation of IL-6 expression in neurons may involve a calcium-dependent pathway. Overall, these data demon­strate that although glial cells have been suggested to be the source of IL-6 in the ischemic brain (Maeda et aI., 1994; Schiefer et aI., 1998), neurons are a potential pa­renchymal source of this cytokine. Recent evidence of IL-6 immunoreactive neurons in the ischemic rodent brain (Suzuki et aI., 1999a,b) supports that data. In ad­dition, this is the first demonstration that the calcium ions influx evoked by the over-activation of NMDA receptors could be a critical signal for IL-6 production in the isch­emic brain.

When considering the wide variety of central nervous system pathologies that exhibit a locally increased pro­duction of IL-6, this cytokine may be viewed as a ubiq­uitous alarm signal after tissue injury. However, the in­fluence of this IL-6 induction remains controversial (Tarkowski et aI., 1995; Campbell, 1998; Qiu et aI., 1998; Toulmond et aI., 1992; Loddick et aI., 1998; Mat­suda et aI., 1996). To address the role of IL-6 during

cerebral ischemia, we used cultured cortical neurons and astrocytes that express both the binding, (IL-6Ra) and the transducing (gp130) IL-6 receptors, and are thus able to respond to IL-6. These cultures were submitted to apoptotic and excitotoxic paradigms, reproducing the major pathways leading to ischemic neuronal death. By

using two well characterized apoptotic challenges (serum deprivation and staurosporine exposure), we demon­strated that recombinant IL-6 fails to prevent the neuro­nal cell death, whereas cycloheximide succeeds. This observation can be related to the lack of protective effect

J Cereb Blood Flow Metab. Vol. 20. No.6, 2000

of IL-6 against NO-induced cortical neuronal apoptotic cell death (Toku et aI., 1998); but contrasts with the beneficial activity of this cytokine against serum depri­vation-mediated PC12 cell death (Umegaki et aI., 1996). By using necrotic paradigms, we evidenced that IL-6 does not prevent AMPAlkainate receptor-mediated neu­ronal death, but dose-dependently protects neurons against NMDA receptor-induced necrosis. We further characterized this selective protective activity of IL-6 by showing that it requires the binding of this cytokine to its receptor, because it is abolished by a competitive inhibi­tor. Previous data have suggested that IL-6 could support

neuronal survival by controlling the astrocytic activation (Fattori et aI., 1995; Raivich et aI., 1996) or secretion of neurotrophic factors (Kossmann et aI., 1996, Marz et aI., 1999, Carlson et aI., 1999). Here we showed that astro­cytes are not necessary for IL-6 neuroprotective activity, as evidenced by the maintenance of IL-6 beneficial effect against NMDA exposure in pure neuronal cultures. The idea that IL-6 directly targets neurons opens new insights

into the cellular processes of IL-6-mediated neuroprotec­tion during cerebral ischemia. However, further investi­

gations are required to extend our in vitro findings to the adult brain and to clarify the molecular mechanism(s) that underlies the protective effect of IL-6. In particular, the competitive inhibitor, which prevents the neuropro­tective activity of IL-6 in cortical cultures, represents a powerful tool to evaluate the role of IL-6 during cerebral ischemia.

In summary, we demonstrate that during cerebral isch­emia and other excitotoxicity-related injuries, neurons could synthesize and release IL-6 in response to the raise

in the intracellular calcium concentration mediated through NMDA receptor activation, This induction of IL-6 could serve an endogenous autocrine protective mechanism, directly targeting neurons to prevent NMDA receptor-induced degeneration. These experimental ob­servations, together with the clinical evidence of a cere­bral production of IL-6 in stroke patients (Fassbender et aI., 1994, Tarkowski et aI., 1995), raise the possible rel­evance of the IL-6 signaling pathway as an early target for the development of new therapeutic strategies.

Acknowledgments: The authors thank A. Ruocco and S. Roussel for technical assistance.

REFERENCES

Akira S ( 1997) IL-6-regulated transcription factors. Int J Biochem Cell Bioi 29:140 1- 1418

Benveniste EN ( 1998) Cytokine actions in the central nervous system. Cytokine Growth Factor Rev 9:259-275

Bonfoco E, Krain D, Ankarcrona M, Nicotera P, Lipton SA ( 1995) Apoptosis and necrosis: two distinct events induced, respectively, by mild and intense insults with N-methyl-D-aspartate and nitric oxide/superoxide in cortical cell culture. Froc Natl Acad Sci US A 92:7162-7166

IL-6 PROTECTS NEURONS AGAINST NMDA TOXICITY 965

Buisson A, Nicole 0, Docagne F, Sartelet H, MacKenzie ET, Vivien D (1998) Up-regulation of a serine protease inhibitor in astrocytes mediates the neuroprotective activity of transforming growth fac­tor-i3l. FASEB J 12: 1683-1691

Campbell IL ( 1998) Structural and functional impact of the transgenic expression of cytokines in the CNS, Ann N Y Acad Sci 840:83-96

Carlson NG, Wieggel W A, Chen J, Bacchi A, Rogers SW, Gahring LC (1999) Inflammatory cytokines IL-J a, IL-li3, IL-6, and TNF-a impart neuroprotection to an excitotoxin through distinct path­ways, J lmmunol 163:3963-3968

Chiu CP, Moulds C, Coffman RL, Rennick D, Lee F (1988) Multiple biological activities are expressed by a mouse interleukin-6 eDNA clone isolated from bone marrow stromal cells, Proc Natl Acad Sci US A 85:7099-7103

Choi DW (1992) Excitotoxic cell death, J NeurobioI23:1261-1276

Choi DW ( 1996) Ischemia-induced neuronal apoptosis. Curr Opin

Neurobiol 6:667-672

Clark WM, Rinker LG, Lessov NS, Hazel K, Eckenstein F ( 1999) Time course of IL-6 expression in experimental CNS ischemia. Neurol Res 21 :287-292

DeGraba TJ (1998) The role of inflammation after acute stroke: utility of pursuing anti-adhesion molecule therapy. Neurology 5 1:S62-S68

Ekida T, Nishimura C, Masuda S, Itoh S, Shimada I, Arata Y (1992) A receptor-binding peptide from human interleukin-6: isolation and a proton nuclear magnetic resonance study. Biochem Biophys Res Commun 189:211-220

Fassbender K, Rossol S, Kammer T, Daffertshofer M, Wirth S, Doll­man M, Hennerici M ( 1994) Proinflammatory cytokines in serum of patients with acute cerebral ischemia: kinetics of secretion and relation to the extent of brain damage and outcome of disease. J Neurol Sci 122:135-139

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

Feuerstein GZ, Wang X, Barone FC (1998) The role of cytokines in the neuropathology of stroke and neurotrauma. Neuroimmullomodula­

tioll 5: 143-159

Gadient RA, Otten U (1994) Expression of interleukin-6 (IL-6) and interleukin-6 receptor (IL-6R) mRNAs in rat brain during postnatal development. Braill Res 637:10--14

Gruol DL, Nelson TE ( 1997) Physiological and pathological roles of interleukin-6 in the central nervous system. Mol Neurobiol 15: 307-339

Heinrich PC, Behrmann I, Miiller-Newen G, Schaper F, Graeve L ( 1998) Interleukin-6-type cytokine signalling through the gp 130/ Jak/STAT pathway. Biochem J 334:297-314

Hill JK, Gunion-Rinker L, Kulhanek D, Lessov N, Kim S, Clark WM, Dixon MP, Nishi R, Stenzel-Poore MP, Eckenstein FP (1999) Temporal modulation of cytokine expression following focal ce­rebral ischemia in mice. Brain Res 820:45-54

Hirano T ( 1998) Interleukin-6 and its receptor: ten years later. lilt Rev Immunol 16:249-284

Hirano T, Yasukawa K, Harada H, Taga T, Watanabe Y, Matsuda T, Kashiwamura S, Nakajima K, Koyama K, Iwamatsu A, Tsunasawa S, Sakiyama F, Matsui H, Takahara Y, Taniguchi T, Kishimoto T (1986) Complementary DNA for a novel human interleukin (BSF-2) that induces B lymphocytes to produce immunoglobulin. Nature

324:73-76

Koh JY, Choi DW (1987) Quantitative determination of glutamate mediated cortical neuronal injury in cell culture by lactate dehy­drogenase efflux assay. J Neurosci Methods 20:83-90

Koh JY, Wie MB, Gwag BJ, Sensi SL, Canzoniero LM, Demaro J, Cser Nansky C, Choi DW ( 1995) Staurosporine-induced neuronal apoptosis. Exp Neurol 135: 153-159

Kossmann T, Hans V, Imhof H-G, Trentz 0, Morganti-Kossmann MC (1996) Interleukin-6 released in human cerebrospinal fluid follow­ing traumatic brain injury may trigger nerve growth factor produc­tion in astrocytes. Brain Res 713:143-152

Kushima Y, Hama T, Hatanaka H (1992) Interleukin-6 as a neuro­trophic factor for promoting the survival of cultured catecholamin-

ergic neurons in a chemically defined medium from fetal and post­natal rat midbrains. Neurosci Res 13:267-280

Loddick SA, Wong ML, Bongiorno PB, Gold PW, Licinio J, Rothwell NJ (1997) Endogenous interleukin-I receptor antagonist is neuro­protective. Biochem Biophys Res Commun 234:211-215

Loddick SA, Turnbull A V, Rothwell NJ (1998) Cerebral interleukin-6 is neuroprotective during permanent focal cerebral ischemia in the rat. J Cereb Blood Flow Metab 18:176-179

Longa EZ, Weinstein PR, Carlson S, Cummins R ( 1989) Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20:84-91

MacDermott AB, Mayer ML, Westbrook GL, Smith SJ, Barker JL (1986) NMDA-receptor activation increases cytoplasmic calcium concentration in cultured spinal cord neurons. Nature 321 :519-522

Maeda Y, Matsumoto M, Hori 0, Kuwabara K, Ogawa S, Yan SD, Ohtsuki T, Kinoshita T, Kamada T, Stem DM (1994) Hypoxia! reoxygenation-mediated induction of astrocyte interleukin-6: a paracrine mechanism potentially enhancing neuron survival. J Exp

Med 180:2297-2308 Martin DP, Schmidt RE, Distefano PS, lowry OH, Carter JG, Johnson

EM Jr ( 1988) Inhibitors of protein synthesis and RNA synthesis prevent neuronal death caused by nerve growth factor deprivation. J Cell Physiol 106:829-844

Marz P, Heese K, Dimitriades-Schmutz B, Rose-John S, Otten U (1999) Role of interleukin-6 and soluble IL-6 receptor in region­specific induction of astrocytic differentiation and neurotrophin expression. Glia 26: 191-200

Matsuda S, Wen T-C, Morita F, Otsuka H, Igase K, Yoshimura H, Sakanaka M ( 1996) Interleukin-6 prevents ischemia-induced learn­ing disability and neuronal and synaptic loss in gerbils. Neurosci Lett 204: 109-112

Orzylowska 0, Oderfeld-Nowak B, Zaremba M, Januszewski S, Mossakowski M (1999) Prolonged and concomitant induction of astroglial immunoreactivity of interleukin-I 13 and interleukin-6 in the rat hippocampus after transient global ischemia. Neurosci Lett 263:72-76

Paxinos G, Watson C (1986) The rat brain in stereotaxic coordinates. San Diego, CA: Academic Press

Prehn JH, Backhauss C, Krieglstein J (1993) Transforming growth factor-i3l prevents glutamate neurotoxicity in rat neocortical cul­tures and protects mouse neocortex from ischemic injury in vivo. J Cereb Blood Flow Metab 13:521-525

Qiu Z, Sweeney DD, Netzeband JG, Gruol DL (1998) Chronic inter­leukin-6 alters NMDA receptor-mediated membrane responses and enhances neurotoxicity in developing CNS neurons. J Neurosci

18: I 0445-10456 Raivich G, Bluethmann H, Kreutzberg GW ( 1996) Signaling molecules

and neuroglial activation in the injured central nervous system. Keio J Med 45:239-247

Rose K, Goldberg MP, Choi DW (1993) Cytotoxicity in murine neo­cortical cell culture. In: In vitro biological methods. Methods in toxicology (Tyson CA, Frazier JM, eds), San Diego, CA: Aca­demic Press, pp 46-60

Ruocco A, Docagne F, Nicole 0, Ali C, Chazalviel L, Komesli S, Yablonsky F, Roussel S, MacKenzie ET, Vivien D, Buisson A ( 1999) A TGF-i3 antagonist unmasks the neuroprotective role of this endogenous cytokine in excitotoxic and ischemic brain injury. J Cereb Blood Flow Metab 19: 13145-13153

Saito K, Suyama K, Nishida K, Sei Y, Basile AS (1996) Early increases in TNF-a, IL-6, and IL-l 13 levels following transient cerebral isch­emia in gerbil brain. Neurosci Lett 206:149-152

Saito M, Yoshida K, Hibi M, Taga T, Kishimoto T (1992) Molecular cloning of a murine IL-6 receptor-associated signal transducer, gp130, and its regulated expression ill vivo. J Immullol 148:4066-407 1

Schiefer J, Topper R, Schmidt W, Block F, Heinrich PC, Noth J, Schwarz M (1998) Expression of interleukin-6 in the rat striatum following stereotaxic injection of quinolinic acid. J Neuroimmunol

89:168-176 Schobitz B, De Kloet ER, Sutanto W, Holsboer F (1993) Cellular

localization of interleukin-6 mRNA and interleukin-6 receptor mRNA in rat brain. Eur J Neurosci 5:1426-1435

Stroemer RP, Rothwell NJ ( 1998) Exacerbation of ischemic brain dam-

J Cereb Blood Flow Metab, Vol. 20, No.6. 2000

966 C. ALI ET AL.

age by localized striatal injection of interleukin-113 in the rat. J Cereb Blood Flow Metab 18: 833-839

Sugita T, Totsuka T, Saito M, Yamasaki K, Taga T, Hirano T, Kishi­moto T (1990) Functional murine interleukin-6 receptor with the intracisternal A particle gene product at its cytoplasmic domain. Its possible role in plasmacytomagenesis. J Exp Med 171:2001-2009

Suzuki S, Tanaka K, Nogawa S, Nagata E, Ito D, Dembo T, Fukuuchi Y (1999a) Temporal profile and cellular localization of interleu­kin-6 protein after focal cerebral ischemia in rats. J Cereb Blood Flow Metab 19:1256-1262

Suzuki S, Tanaka K, Nagata E, Ito D, Dembo T, Fukuuchi Y (1999b) Cerebral neurons express interleukin-6 after transient forebrain ischemia in gerbils. Neurosci Lett 262: 117-120

Tarkowski E, Rosengren L, Blomstrand C, Wikkelsii C, Jensen C, Ekholm S, Tarkowski A (1995) Early intrathecal production of interleukin-6 predicts the size of brain lesion in stroke. Stroke

26: 1393-1398 Toku K, Tanaka J, Yano H, Desaki J, Zhang B, Yang L, Ishihara K,

Sakanaka M, Maeda N (1998) Microglial cells prevent nitric ox-

J Cereb Blood Flow Metab, Vol. 20, No. 6, 2000

ide-induced neuronal apoptosis in vitro. J Neurosci Res 53:415-425

Toulmond S, Vige X, Fage D Benavides J (1992) Local infusion of interleukin-6 attenuates the neurotoxic effects of NMDA on rat striatal cholinergic neurons. Neurosci Lett 144:49-52

Umegaki H, Yamada K, Naito M, Kameyama T, Iguchi A, Nabeshima T (1996) Protective effect of interleukin-6 against the death of PC 12 cells caused by serum deprivation or by the addition of a calcium ionophore. Biochem Pharmacal 52:911-916

Wang X, Yue T-L, Young PR, Barone FC, Feuerstein GZ (1995) Expression of interleukin-6, c-fos, and zif268 mRNAs in rat isch­emic cortex. J Cereb Blood Flow Metab 15:166-171

Watanabe D, Yoshimura R, Khalil M, Yoshida K, Kishimoto T, Taga T, Kiyama H (1996) Characteristic localization of gp130 (the sig­nal-transducing receptor component used in common for IL-6/IL­l l1CNTFILIF/OSM) in the rat brain. Eur J Neurosci 8:1630--1640

Yamada M, Hatanaka H (1994) Interleukin-6 protects cultured rat hip­pocampal neurons against glutamate-induced cell death. Brain Res

643:173-180