Clodronate inhibits the secretion of proinflammatory cytokines and NO by isolated microglial cells and reduces the number of proliferating glial cells in excitotoxically injured organotypic hippocampal slice cultures

Experimental Neurology 189 (2004) 241 – 251 www.elsevier.com/locate/yexnr

Clodronate inhibits the secretion of proinflammatory cytokines and NO by isolated microglial cells and reduces the number of proliferating glial cells in excitotoxically injured organotypic hippocampal slice cultures Faramarz Dehghani a, Ariane Conrad a,b, Angelika Kohl a,b, Horst-Werner Korf a, Nils P. Hailer c,* a

Institute of Anatomy II, Johann Wolfgang Goethe-University, Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany b University Hospital for Orthopaedic Surgery Friedrichsheim, Marienburgstr. 2, D-60528 Frankfurt am Main, Germany c Institute of Surgical Sciences, Department of Orthopedics, Karolinska Institute at Karolinska Hospital, SE-171 76 Stockholm, Sweden Received 12 January 2004; revised 20 April 2004; accepted 3 June 2004 Available online 30 July 2004

Abstract Treatment of excitotoxically injured organotypic hippocampal slice cultures (OHSC) with clodronate is known to result in the inhibition of microglial activation. We hypothesized that this is due to direct effects of clodronate on microglial cells, and investigated microglial proliferation in OHSC, and cytokine and NO secretion in isolated microglial cells. N-methyl-d-aspartate (NMDA) lesioning of OHSC resulted in a massive increase in the number of proliferating, bromo-desoxy-uridine (BrdU)-labeled cells that was reduced to control levels after treatment with clodronate (0.1, 1, 10 Ag/ml). Triple-labeling revealed that clodronate abrogated the proliferation of both glial fibrillary acidic protein (GFAP)-labeled astrocytes and Griffonia simplicifolia isolectin B4 (IB4)-labeled microglial cells. Furthermore, isolated microglial cells were treated with clodronate after stimulation with lipopolysaccharide (LPS) or macrophage colony stimulating factor (M-CSF). Clodronate (0.01, 0.1, 1 Ag/ml) significantly down-regulated the LPS-stimulated microglial secretion of tumor necrosis factor (TNF)-a, Interleukin (IL)1h and NO, but not of IL-6. In contrast, clodronate significantly reduced the microglial IL-6-release induced by M-CSF, indicating different intracellular pathways. The number and morphology of isolated microglial cells did not change significantly after treatment with clodronate. In summary, the number of proliferating microglial cells and astrocytes after excitotoxic injury is reduced to control levels after treatment with clodronate. Furthermore, clodronate inhibits microglial secretion of proinflammatory cytokines and NO. Clodronate could therefore prove to be a useful tool in the investigation of interactions between damaged neurons and microglial cells. D 2004 Elsevier Inc. All rights reserved. Keywords: Microglial cell; Neuron; Clodronate

Introduction In the adult, unlesioned CNS resting microglial cells express very low levels of major histocompatibility complex (MHC), adhesion molecules, and proinflammatory cytokines (Hailer et al., 1998; Wilms et al., 1997; Vincent et al., 1997). Pathological events induce rapid activation of these formerly resting cells, reflected by proliferation, a characteristic change in morphology, and changes in the profile of cytokine secretion (Hailer et al., 1997; Kreutzberg, 1996; Raivich et al., 1994). Adhesion molecules, comple-

* Corresponding author. Fax: +46-8-517-7-46-99. E-mail address: [email protected] (N.P. Hailer). 0014-4886/$ - see front matter D 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.expneurol.2004.06.010

ment receptors, and MHC-class-II molecules are also upregulated on activated microglial cells (Beyer et al., 2000; Guo et al., 2000; Schmitt et al., 2000). Furthermore, activated microglial cells release neurotoxic substances such as free radicals, nitric oxide (NO), proinflammatory cytokines, arachidonic acid derivatives, and excitatory amino acids. The orchestra of these agents is believed to enlarge the extent of an ischemic or traumatic injury, causing the destruction of neurons that were primarily unaffected by the lesion (Bartholdi and Schwab, 1997; Dusart and Schwab, 1994; Minghetti and Levi, 1998). Affecting the activation state of macrophages and microglial cells by pharmacological compounds has therefore been suggested as a therapeutic option to reduce the amount of secondary neuronal damage following brain and spinal cord lesions.

242

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

The bisphosphonate clodronate is clinically established in the treatment of severe hypercalcemia in hyperparathyroidism or malignancy and in the treatment of osteoporosis and Paget’s disease (Russell and Rogers, 1999). Bisphosphonates inhibit bone resorption by inducing apoptosis of osteoclasts (Benford et al., 2001); however, clodronate and other bisphosphonates exert their effects not only on osteoclasts but also on other cells of the monocytic lineage. For example, peripheral macrophage functions, such as NO secretion, are suppressed (Makkonen et al., 1996), apoptotic death is induced in peritoneal macrophages (van Rooijen et al., 1996), and perivascular cells in the CNS are abolished (Polfliet et al., 2001). We have previously demonstrated that the number of microglial cells in excitotoxically lesioned organotypic hippocampal slice cultures (OHSC) is severely reduced after clodronate treatment (Kohl et al., 2003). We therefore hypothesized that clodronate exerts direct effects on microglial cells, inhibiting microglial proliferation and their secretion of proinflammatory cytokines and NO. The present study was designed to investigate the effects of clodronate on the proliferation of glial cells in excitotoxically lesioned OHSC, on the secretion of interleukin (IL)-1h, IL-6, tumor necrosis factor (TNF)-a, and NO by isolated microglial cells, and on the number and morphology of microglial cells in both culture systems.

Materials and methods Organotypic hippocampal slice cultures (OHSC) OHSC were obtained from 8-day-old Wistar rats after decapitation and dissection of the brains under sterile conditions. The frontal pole and the cerebellum were removed and the brains were placed in minimal essential medium (MEM, Gibco BRL Life Technologies, Eggenstein, Germany), containing 1% glutamine (Gibco) at 4jC. Using a sliding vibratome (Vibratome 1000 Classic, St. Louis, MO, USA), approximately 1 mm of the ventral surface was removed, and 350 Am slices were prepared. Six to eight OHSC were obtained from each brain and used for further experiments. OHSC were transferred into cell culture inserts (Becton Dickinson, Franklin Lakes, NJ, USA; pore size 0.4 Am) that were placed in 6-well culture dishes (Becton Dickinson), containing 1 ml culture medium per well. Culture medium consisted of 48% MEM, 25% Hank’s balanced salt solution (HBSS; Gibco), 25% normal horse serum (NHS; Gibco), 2% glutamine, 1 Ag/ml insulin (Boehringer, Mannheim, Germany), 1.2 mg/ml glucose (Braun, Melsungen, Germany), 0.1 mg/ml streptomycin (Sigma, Deisenhofen, Germany), 100 U/ml penicillin (Sigma), and 0.8 Ag/ ml ascorbic acid (Sigma), pH = 7.4. The culture dishes were incubated at 35jC in a fully humidified atmosphere with 5% CO2.

OHSC were divided into different experimental groups and treated according to the following protocols: Group A (CTL): unlesioned OHSC were incubated in control medium from 0 days in vitro (div) until 9 div. Group B (NMDA): OHSC were incubated in control medium for 6 days, then lesioned with 50 AM NMDA for 4 h and cultured in control medium for 3 further days. Group C (NMDA + CLO): OHSC were lesioned with NMDA after 6 divisions for 4 h in the presence of clodronate (Hoffmann La Roche Ltd, Lo¨rrach, Germany) in three different concentrations (0.1, 1, 10 Ag/ml) and incubated in clodronatecontaining medium until fixation after 9 div 16 h before fixation OHSC in all experimental groups were incubated with BrdU (0.01 mM). After 9 div, all OHSC were fixed with 4% paraformaldehyde containing 15% picric acid and 0.1% glutaraldehyde in 0.1 M phosphate buffer (PB) for 15 min, washed with 0.1 M PB and postfixed without glutaraldehyde for 1 h. Finally, OHSC were washed with 0.1M PB for 1 h and carefully removed from the cell culture membranes. For conventional histochemistry, OHSC were stored in 0.8 M sucrose solution containing 0.1% NaN3 for 2 days, sectioned (14 Am) on a Microm HM 560 cryostat (Microm, Walldorf, Germany) and mounted on gelatine-coated glass slides. After endogenous peroxidase quenching with methanol and H2O2 (0.45%), OHSC were incubated with 1 N HCl for 1 h. OHSC were then washed with phosphate-buffered saline (PBS) and incubated with NHS (diluted 1:20 in PBS containing 0.03% Triton) for 30 min. To detect BrdU+ cells, the primary antibody anti-BrdU (DAKO, Glostrup, Denmark) was used (diluted 1:100 in PBS-Triton with 5% bovine serum albumin [BSA]). Anti-mouse-IgG (Sigma) was used as a secondary antibody diluted 1:100 in PBS-Triton with 5% BSA, and OHSC were then incubated with avidin– biotin complex (diluted 1:100 in PBS-Triton) and 3,3’-diaminobenzidine (DAB; Sigma) as a chromogene. After washing, the sections were counterstained with hematoxylin and coverslipped with Entellan (Merck, Darmstadt, Germany). The number of BrdU+ cells was determined in the dentate gyrus (DG) of the different experimental groups (CTL: n = 13; NMDA: n = 9; NMDA + CLO 0.1 Ag/ml: n = 7; NMDA + CLO 1 Ag/ml: n = 7; NMDA + CLO 10 Ag/ml: n = 4) by counting BrdU+ cells in bright field microscopy, where n represents the number of animals used to obtain OHSC from. Six to eight slices were obtained from each rat. The mean number of BrdU+ cells in the DG of OHSC in conventionally labeled OHSC was calculated for each group and statistical analysis was performed. The one-way ANOVA test was used to determine whether differences between experimental groups were statistically significant (P < 0.05 was considered statistically significant). After finding significant differences in the one-way ANOVA test Dunnett’s test for multiple comparisons was performed. Unlesioned OHSC and lesioned OHSC treated with clodronate were compared to OHSC treated with NMDA alone.

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

For triple-labeling sections were incubated with the monoclonal anti-BrdU-antibody (diluted 1:100 in PBS-Triton with 5% BSA; DAKO Diagnostika GmbH, Hamburg, Germany) for 16 h followed by a secondary goat-anti-mouse Cy5-conjugated antibody (diluted 1:100 in PBS-Triton X100) for 1 h. They were then incubated with a polyclonal rabbit anti-GFAP-antibody (diluted 1:500 in PBS-Triton with 5% BSA; DAKO Diagnostika GmbH) for 12 h, washed and incubated with a secondary goat –anti-rabbit antibody (Alexa 546; diluted 1:100 in PBS-Triton with 5% BSA; Molecular Probes Europe, Leiden, The Netherlands) for 1 h. The triple-labeling was completed by staining of microglial cells with FITC-conjugated Griffonia simplicifolia isolectin B4 (FITC-IB4 diluted 1:20 in PBS-Triton with 5% BSA, Sigma) for 1 h. Sections were then washed with PBS and aqua destillata for 10 min each, mounted with DAKO fluorescent mounting medium and analyzed by confocal laser scanning microscopy in multitracking mode. Monochromatic light at 488, 543, and 633 nm with a dichroic beam splitter (FT UV/488/543/633) was used to visualize microglial cells (FITC-IB4; excitation 488 nm, emission bandpass filter 505 –530 nm), astrocytes (GFAP; excitation 633 nm, emission LP 650 nm), and proliferating cells (BrdU; excitation 543 nm, emission BP 585 –615 nm). Isolated microglial cell cultures Microglial cell cultures were prepared from p0-2 Wistar rat brains. After decapitation, the brains were removed under sterile conditions and placed in HBSS (with Ca2+ and Mg2+; Gibco BRL Life Technologies) containing atocopherol (Sigma). The meninges of the brains were removed, the cerebral hemispheres washed with HBSS without Ca2+ or Mg2+ and then incubated for 5 min at 4jC with a mixture of HBSS with Ca2+ and Mg2+ containing trypsin (4 mg/ml; Boehringer) and DNase (0.5 mg/ ml; Worthington, Bedford MA, USA). This suspension was centrifuged at 133  g for 10 min at 4jC and the cell pellets were resuspended in 1 ml culture medium, consisting of Dulbecco’s modified Eagle’s medium (DMEM, Gibco), supplemented by 10% fetal bovine serum (FBS, Gibco), 1% glutamine (Boehringer), 100 U/ml penicillin and 0.1 mg/ml streptomycin (Sigma). The cell suspension was transferred into tissue culture flasks (75 cm2; Falcon, Heidelberg, Germany) and incubated at 37jC in a fully humidified atmosphere containing 5% CO2. After 7 div the microglial cells were isolated from the astrocytic mono-

243

layer and then seeded into 24-well tissue culture plates (Falcon) onto poly-L-lysin-coated coverslips. To examine the morphology of microglial cells and their secretion of the cytokines TNF-a and IL-1h, microglial cell cultures were divided into four different groups and treated for 48 h according the following protocols (1 ml medium per well). Group A (CTL): microglial cells were incubated in control medium, consisting of DMEM supplemented with 2% FBS, 1% glutamine and penicillin/ streptomycin as above. Group B (LPS): microglial cells were stimulated with lipopolysaccharide (LPS, 10 ng/ml, Sigma). Group C (Clodronate): microglial cells were treated with culture medium as above with clodronate in five different concentrations (0.01, 0.1, 1, 10, or 100 Ag/ml). Group D (Clodronate + LPS): microglial cells were incubated with culture medium containing both clodronate in concentrations of either 0.01, 0.1, 1, 10, or 100 Ag/ml and LPS (10 ng/ml). To investigate the secretion of IL-6, microglial cells were stimulated with LPS (10 ng/ml) or macrophagecolony stimulating factor (M-CSF, 100 ng/ml, Sigma) for 48 h. The ability of clodronate to suppress IL-6 secretion was investigated using microglial cells stimulated with LPS or M-CSF and treated with clodronate (1 Ag/ml). After 48 h, the supernatants of all groups were collected and stored at 20jC. To investigate NO secretion, microglial cells were seeded into poly-L-lysin-coated 96-well tissue culture plates (Falcon) and treated for 72 h according to the protocol described above. After 72 h, the supernatants were collected as described above. To investigate their proliferative response, microglial cells were stimulated with LPS (10 ng/ml) or M-CSF (50, 75 or 100 ng/ml) for 48 h. BrdU was added 16 h before fixation and the ability of LPS or M-CSF to induce proliferation was measured by counting BrdU+ nuclei in three fields per well. Microglial cells from all experimental groups were fixed with 4% paraformaldehyde in PB for 10 min and washed with 0.1 M PBS. They were subsequently stained with hematoxylin for 4– 8 min, washed with aqua destillata, and then stained with eosin for 3 –5 min. The coverslips were mounted with Entellan (Merck) and cell counts were performed in three representative fields (magnification: 200 fold) in each coverslip to ensure that the measurements of cytokines or NO were made from equal cell numbers (Table 1).

Table 1 Mean number of isolated microglial cells per area

Mean Standard deviation P value vs. CTL a

CTL

LPS

CLO (1)a

CLO (0.1)a

CLO (0.01)a

LPS + CLO (1)a

LPS + CLO (0.1)a

LPS + CLO (0.01)a

27.06 F5.256 ./.

30.32 F8.526 >0.05

23.91 F8.171 >0.05

21.52 F7.216 >0.05

27.11 F 12.82 >0.05

27.85 F 5.982 >0.05

32.60 F12.28 >0.05

30.66 F10.12 >0.05

Numbers in parentheses represent clodronate concentrations in Ag/ml.

244

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

Analysis of microglial secretion of IL-1b, IL-6, and TNF-a The concentrations of TNF-a, IL-1h, and IL-6 in the supernatants from microglial cells were determined by commercially available ELISA kits (R&D Systems GmbH, Wiesbaden-Nordenstadt, Germany). The optical density was determined at 450 nm using a microplate reader (ELISA CERES 900, Bio-Tek Instruments, Inc., Houston, TX) and the mean concentration of each cytokine after LPS stimulation was normalized to 100% for each experimental group. Analysis of NO secretion NO production in microglial cells was examined by the Griess reaction. NO levels were directly analyzed in unfrozen and undiluted supernatants from microglial cells: 50 Al supernatant was incubated for 10 min at room temperature with 50 Al Griess reagent (acetic acid 286 g/ l, sulfanilic acid 55 g/l, naphtylamine 0.31 g/l; Merck), optical density was measured at 550 nm using a microplate reader (ELISA CERES 900, Bio-Tek Instruments, Inc.), the nitrite concentration was determined from a potassium nitrite standard curve, and results were expressed as normalized mean NO concentrations (set to 100%) after LPS stimulation. Quantitative morphometry of microglial cells Endogenous peroxidase quenching was performed by incubation with 30% H2O2 and methanol for 10 min. Subsequently, the cells were preincubated with NHS (1:20 in PBS/Triton; Gibco) for 30 min and then incubated with biotinylated Griffonia simplicifolia isolectin B4 (IB4, 1:50 in PBS/Triton; Sigma) for 1 h at room temperature. After renewed washing, the microglial cells were incubated with ExtraAvidin-Peroxidase (1:100 in PBS/Triton; Sigma) for 1 h at room temperature and treated with 3,3V-diaminobenzidine (DAB, Sigma) and 0.05% H2O2 for 2– 5 min. The cells were washed in Tris buffer for 10 min and the coverslips mounted in Kaiser’s glycerol gelatine (Merck). The cells were examined with a Zeiss Axiovert 35 (Zeiss, Go¨ttingen, Germany). Three representative visual fields per coverslip (magnification: 200 fold) were analyzed, and the index of ramification (IR) was calculated by computerassisted morphometry (Heppner et al., 1998). IR as the degree of microglial ramification was defined by the ratio of cell area and convex area (IR = cell area/convex area), where ‘cell area’ represents the area covered by the cell, and where ‘convex area’ is the area defined by the cells’ most prominent processes. Amoeboid cells with few cytoplasmic processes display nearly identical values in their ‘cell area’ and ‘convex area’, thus their calculated IR is close to 1. In contrast, strongly ramified cells have a large convex area and a rather small cell area; thus the ratio of these two numbers results in smaller IR values. Data were expressed as mean values (F standard error of the mean [SEM]) derived from at least three independent experiments. Statistical analyses were performed us-

ing one-way ANOVA test followed by Dunnett’s test for multiple comparisons (Graph Pad, San Diego, CA). Results with P < 0.05 were considered significant.

Results Clodronate reduces the number of proliferating cells in excitotoxically lesioned OHSC DAB-stained preparations of unlesioned OHSC contained only small numbers of BrdU+ nuclei in different layers of the DG (mean cell number in the DG: 18.83; Fig. 1). BrdU+ nuclei displayed varying morphologies, some were fairly large, indicating that they belonged to astrocytes, others were round and small, indicating that they belonged to microglial cells. BrdU+ cells were scattered across the entire OHSC, but were preferentially observed in the DG and the hilar area. Hematoxylin counterstaining showed an intact cytoarchitecture of the hippocampal formation, and the stratification of the DG remained intact, indicating excellent neuronal survival (Fig. 2A). In contrast, NMDA lesioning of OHSC after 6 div induced a massive increase in the number of BrdU+ nuclei in the entire OHSC, most pronounced in the DG (mean number of BrdU+ nuclei in the DG: 68.7) and the pyramidal cell layer of the cornu ammonis. Compared to unlesioned OHSC, the number of BrdU+ nuclei was increased by a factor of 3.7 (Fig. 1). As in unlesioned OHSC, the morphology of BrdU+ nuclei in lesioned OHSC varied from small to fairly large. Furthermore, hematoxylin staining showed a massive destruction of the regular hippocampal cytoarchitecture in lesioned OHSC. Numerous intensely stained, often

Fig. 1. Quantitative analysis of the numbers of proliferating cells in OHSC: the mean number of BrdU+ nuclei in OHSC lesioned by application of NMDA was significantly higher than in control OHSC. A dose-dependent decrease in the number of BrdU+ nuclei in NMDAlesioned OHSC was induced after additional treatment with clodronate (0.1, 1, 10 Ag/ml).

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

245

Fig. 2. Visualization of proliferating BrdU+ nuclei in the dentate gyrus (DG) of OHSC: counterstaining with hematoxylin in bright field microscopy (A, C, E) and triple-labeling for proliferating cells (BrdU, blue), microglial cells (IB4, green), and astrocytes (GFAP, red) using confocal laser scanning microscopy (B, D, F). (A, B) Control OHSC (CTL). Control OHSC show an intact cytoarchitecture of the dentate gyrus (DG) and the Cornu ammonis (CA), only very few BrdU+ nuclei are found in the DG. The majority of BrdU+ nuclei are in the non-cellular layers of the DG (A). In control OHSC, few IB+4 microglial cells and a moderate number of GFAP+ astrocytes are detected in the DG. Triple-labeling reveals that the majority of BrdU+ nuclei belongs to astrocytes, shown by double-labeling for BrdU and GFAP (B). (C, D) Lesioned OHSC (NMDA). Bright field microscopy shows massive neuronal damage in the DG and a strong increase in the number of BrdU+ nuclei (C). Lesioning of OHSC results in an increase in the numbers of IB+4 microglial cells and GFAP+ astrocytes (D). The numbers of IB+4 microglial cells and GFAP+ astrocytes with BrdU+ nuclei increase dramatically. (E, F) Lesioned OHSC treated with clodronate (10Ag/ml) (NMDA + CLO). The number of BrdU+ nuclei in the DG of lesioned OHSC treated with clodronate is significantly reduced when compared to untreated, NMDA-lesioned OHSC (E). After clodronate treatment only very few BrdU+ nuclei or IB+4 microglial cells are found in the DG, the number of GFAP+ astrocytes is also reduced. Arrows point to BrdU+ nuclei; arrowheads show the borders between the granule cell layer and the inner molecular layer of the DG; scale bars: 100 Am; frames in overviews indicate the area where the magnified picture was taken.

condensed nuclei were observed in the granule cell layer of the DG, indicating severe damage to neurons after excitotoxic lesioning (Fig. 2C).

Treatment of lesioned OHSC with clodronate resulted in a robust inhibition of the glial proliferative response. The number of BrdU+ nuclei decreased dramatically in a dose-

246

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

dependent manner after treatment of lesioned OHSC with clodronate. Treatment with 10 Ag/ml clodronate resulted in 15.7 BrdU+ nuclei per DG, treatment with 1 or 0.1 Ag/ml clodronate resulted in 29.1 or 37.1 BrdU+ nuclei per DG, respectively (Fig. 1). Moreover, the number of proliferating cells in the entire OHSC was reduced. Hematoxylin staining showed many condensed nuclei obviously belonging to damaged cells in the cellular layer of the DG, and the cytoarchitecture of the DG was profoundly deranged (Fig. 2E). Triple-labeling and confocal laser scanning microscopy confirmed that (i) NMDA-lesioned OHSC displayed a massive increase in the number of BrdU+ nuclei and (ii) treatment with clodronate during and after lesioning induced a dose-dependent decrease in the number of BrdU+ nuclei. In control OHSC, the majority of BrdU+ nuclei belonged to astrocytes labeled by antibodies directed against GFAP. Staining with IB4 showed a very low number of IB4+ microglial cells, only few of which were BrdU+ (Fig. 2B). In lesioned OHSC, the majority of BrdU+ nuclei were double-labeled with GFAP, some BrdU+ cells showed a positive reaction for IB4 (Fig. 2D). Lesioned OHSC that were additionally treated with 10 Ag/ml clodronate showed very few cells that were both IB4+ and BrdU+, confirming the observations that microglial proliferation was prevented. Clodronate treatment also reduced the number of GFAP+ astrocytes nearly to control values and abolished BrdU+ labeling of these cells (Fig. 2F). Clodronate down-regulates microglial secretion of the proinflammatory cytokines TNF-a, IL-1b and IL-6, and NO Microglial cells treated with high concentrations of clodronate (100 or 10 lg/ml) showed signs of apoptotic cell death with condensed nuclei and vacuolization of the cytoplasm. Therefore, these groups were excluded from further investigation on cytokine and NO secretion. Microglial cells incubated with control medium for 48 h did not produce detectable amounts of the proinflammatory cytokine TNF-a, whereas microglial cells stimulated with LPS (10 ng/ml) secreted large amounts of TNF-a. (Fig. 3). Cultures treated with clodronate only (0.01, 0.1, or 1 lg/ ml) did not secrete measurable amounts of the investigated cytokine. In contrast, treatment of LPS-stimulated cells with clodronate (0.01, 0.1, or 1 lg/ml) significantly down-regulated the secretion of TNF-a when compared to microglial cells that were treated with LPS alone (P < 0.05, Fig. 3). As the number of IB4+ microglial cells did not vary significantly between the various experimental groups (Tab. 1), it is unlikely that the reduced cytokine secretion results from a reduction in the total number of cells per well. Microglial cells incubated with control medium secreted only small amounts of IL-1h, whereas cells that were stimulated with LPS released large amounts of of this cytokine. Microglial cells treated with clodronate (0.01, 0.1, or 1 Ag/ml) alone released only small amounts of IL-1h. In contrast,

Fig. 3. Effects of clodronate on the microglial secretion of the proinflammatory cytokines tumor necrosis factor (TNF)-a (top) and interleukin (IL)-1h (bottom), determined by ELISA. The mean proinflammatory cytokine concentration in LPS-stimulated cultures is normalized to 100%. (top) Treatment of unstimulated microglial cells with different concentrations of clodronate (0.01, 0.1, or 1 Ag/ml) did not significantly change the release of TNF-a. In contrast, LPS induced a robust increase in microglial TNF-a secretion (mean: 1381 pg/ml), whereas treatment of LPS-stimulated microglial cells with clodronate significantly reduced the release of TNF-a ( P < 0.05 vs. LPS). (bottom) Stimulation of microglial cells with LPS resulted in an increase in IL-1h secretion (mean: 218.8 pg/ml), whereas treatment with clodronate alone did not influence the secretion of IL-1h. Simultaneous incubation of microglial cells with LPS and clodronate (0.01, 0.1, or 1 Ag/ml) significantly reduced the secretion of IL-1h ( P < 0.05 vs. LPS).

simultaneous treatment with LPS and clodronate (0.01, 0.1, or 1 Ag/ml) significantly down-regulated the release of IL-1h when compared to microglial cells that were treated with LPS alone (P < 0.05, Fig. 3B).

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

Cultures incubated in control medium did not secrete measurable amounts of IL-6, whereas stimulation with LPS resulted in a robust increase in IL-6 secretion. IL-6 secretion of microglial cells was also stimulated by MCSF, but did not reach the level of cells stimulated with LPS. Cells incubated with clodronate (1 Ag/ml) secreted no detectable levels of IL-6. The combined treatment with LPS and clodronate (1 Ag/ml) did not induce downregulation of IL-6 secretion when compared to microglial cells that were treated with LPS alone. In contrast, M-CSFinduced secretion of IL-6 was inhibited by clodronate treatment (1 Ag/ml) when compared to treatment with MCSF alone (P < 0.05; Fig. 4). To investigate their proliferative response microglial cells were stimulated with LPS (10 ng/ml) or M-CSF (50, 75, or 100 ng/ml) for 48 h. Neither LPS (0.5%) nor M-CSF (M-CSF 50 ng/ml: 1.2%, M-CSF 75 ng/ml: 0.99%, and M-CSF 100 ng/ml: 0.94%) were able to change the proliferation rate of microglial cells significantly as compared to controls (0.6%) (P > 0.05). To investigate the microglial production of NO, microglial cells were incubated for 72 h with or without LPS. In the supernatants derived from control cultures and in those derived from microglial cells treated with clodronate (0.01, 0.1, or 1 Ag/ml) alone, small amounts of nitrite were found. Treatment with LPS alone induced a burst of nitric oxide production. In contrast, treatment of microglial cells with LPS and clodronate (0.01, 0.1, or 1 Ag/ml) significantly down-regulated NO secretion when compared to microglial cells treated with LPS alone (P < 0.05, Fig. 5).

Fig. 4. Effects of clodronate on the secretion of the proinflammatory cytokine interleukin (IL)-6 determined by ELISA. The mean cytokine concentration of LPS-stimulated cultures (mean: 256.9 pg/ml) was normalized to 100%. Small amounts of IL-6 were secreted in control cultures and in cultures treated with clodronate alone (0.01, 0.1, or 1 Ag/ ml). Treatment with clodronate did not inhibit the LPS-induced increase in IL-6 secretion, whereas the IL-6 release induced by macrophage colonystimulating factor (M-CSF) was significantly inhibited after treatment with clodronate (1 Ag/ml; P < 0.05 vs. LPS).

247

Fig. 5. Effects of clodronate on microglial NO production, determined by the Griess reaction. Clodronate (0.01, 0.1, or 1 Ag/ml) alone did not significantly change the release of NO from microglial cells. Stimulation with LPS strongly induced the release of NO (mean: 330 Ag/ml), and treatment of LPS-stimulated microglial cells with clodronate (0.01, 0.1, or 1 Ag/ml) significantly reduced the release of NO ( P < 0.05 vs. LPS).

Clodronate does not influence the number or morphology of isolated microglial cells The number of IB4+ microglial cells did not differ significantly between the control group and the groups treated with different concentrations of clodronate. Microglial cells incubated with control medium predominantly displayed short processes (Fig. 6A) with an IR of 0.69 (Fig. 7), microglial cells stimulated with LPS had an IR of 0.64 (Fig. 7). Microglial cells treated with clodronate also displayed typical characteristics of activated cells: The incubation with clodronate (0.01, 0.1, or 1 Ag/ml) resulted in an IR around 0.7 (Figs. 6B – D, 7), and the cell morphology did not change significantly when compared to cells treated with LPS. Combined treatment with LPS and clodronate did not result in significant changes in the morphology of microglial cells that again possessed short cytoplasmic processes (Fig. 7). Microglial cells treated with higher concentrations of clodronate (10 or 100 Ag/ml) showed morphological signs of cell death, for example, nuclear condensation or karyorrhexis; thus, these concentrations were not included in the analysis of cytokine and NO secretion (Figs. 6E, F).

Discussion Traumatic and ischemic lesions to the brain and spinal cord induce activation of microglial cells, resulting in spe-

248

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

Fig. 6. Effects of different concentrations of clodronate on the number and morphology of isolated microglial cells. (A) IB4 labeling of microglial cells incubated with control medium. These cells possess an amoeboid morphology with few, plump cytoplasmic processes without signs of apoptosis or necrosis. (B – D) IB4 labeling of microglial cells incubated with low concentrations of clodronate (B: 0.01, C: 0.1, or D: 1 Ag/ml). Microglial cells maintained their amoeboid appearance. No signs of nuclear condensation indicating apoptotic changes were observed. (E – F) IB4 labeling of microglial cells incubated with high concentrations of clodronate (E: 10 or F: 100 Ag/ml). Microglial cells treated with higher concentrations of clodronate displayed an amoeboid morphology with small cell somata and few cytoplasmic processes. Many of the cells showed signs of nuclear condensation indicating apoptotic cell death. Scale bar: 20 Am.

cific morphological and functional changes: microglial cells undergo amoeboid transformation, proliferate, express adhesion molecules, and secrete proinflammatory cytokines and NO (Bal-Price and Brown, 2001). Due to the neurotoxicity of the latter, it has been suggested that suppression of microglial activation inhibits the development of secondary damage and thus contributes to improved neuronal survival in the aftermath of traumatic or ischemic CNS lesions (Liu et al., 2003; Nicholas et al., 2001; Thery et al., 1994). Several immunosuppressive substances have been investigated for their capacity to inhibit microglial cells, and in some experimental paradigms, an amelioration of CNS structure and function has been described (Bethea et al., 1999; Dehghani et al., 2003; Oudega et al., 1999). Bisphosphonates have been shown to deactivate cells belonging to the monocytic lineage, and we have recently

described that the robust increase in the number of microglial cells following excitotoxic injury in OHSC is virtually abolished after treatment with clodronate (Kohl et al., 2003). Clodronate belongs to the first generation bisphosphonates possessing a chemical structure that is similar to endogenous pyrophosphates and that is characterized by an enzymatic hydrolysis-resistant P – C – P band (Fleisch, 1991). In vivo, clodronate shows a short half-life in plasma and in body fluids (Fleisch, 1989), since approximately 20% of an absorbed oral dose and approximately 40% of an intravenous dose are rapidly bound to bone and then slowly eliminated. Clodronate is mainly renally excreted and around 60 –80% of an intravenous dose leaves the body within 48 h. However, in peripheral tissues, as for example, bone, clodronate is stable and has a half-life of approximately 17 h. The terminal half-life of clodronate in

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

Fig. 7. Quantitative morphometry of microglial cell ramification. Microglial cells incubated with control medium (CTL) were amoeboid with a high index of ramification (IR). Treatment with different concentrations of clodronate (1, 0.1, or 0.01 Ag/ml), LPS (10 ng/ml), or a combination of LPS (10 ng/ml) with clodronate (1, 0.1, or 0.01 Ag/ml) did not significantly change the IR ( P > 0.05).

vivo is determined to be approximately 13 h (Hurst and Noble, 1999; Hoffmann La Roche Ltd., 1998). The present study was designed to investigate whether clodronate modulates the activation of microglial cells by inhibiting proliferation, proinflammatory cytokine secretion, and NO production. This question was addressed both in the complex setting of the excitotoxically lesioned OHSC and in isolated microglial cells. NMDA lesioning alone induced a significant increase in the number of proliferating BrdU+ cells in the dentate gyrus when compared to unlesioned preparations. Lesioned OHSC additionally treated with clodronate displayed a significant inhibition of their glial proliferative response. Triple labeling with BrdU, IB4, and GFAP showed that almost no proliferating microglial cells were present in lesioned OHSC treated with clodronate, and the remaining BrdU+ nuclei were attributed to astrocytes. These data indicate that clodronate exerts its effects predominantly on microglial cells. However, the possibility cannot be ruled out that these effects are induced indirectly via other glial cells. To address the question whether clodronate affects secretion of neurotoxic substances microglial cell cultures were prepared, stimulated with LPS, treated with clodronate, and the secretion of NO and proinflammatory cytokines IL-1h, IL-6 and TNF-a measured. Additionally, the number and the index of ramification of microglial cells were determined. Neither the number nor the morphology of the cells treated with clodronate concentrations of 0.01, 0.1, or 1 Ag/ml (Figs. 6 and 7) varied significantly between different groups.

249

Whereas the production of IL-1h, IL-6, and TNF-a was not affected in unstimulated cells treated with clodronate, the secretion of these cytokines in stimulated preparations simultaneously treated with clodronate was significantly reduced. This is in accordance with previously published data on RAW 264 and J774 macrophages and osteoclasts, showing a decrease in the production of proinflammatory cytokines and NO following clodronate treatment (Makkonen et al., 1999). These findings underscore the anti-inflammatory potential of clodronate as a representative of the halogen-containing bisphosphonates. In contrast to the antiinflammatory properties of halogen-containing bisphosphonates, it has been shown that amino-containing bisphosphonates such as alendronate display pro-inflammatory properties (Rogers et al., 1996). IL-6 release from microglial cells was strongly stimulated by LPS. However, in contrast to the effect of clodronate treatment on IL-1h and TNF-a secretion, the release of IL-6 by microglial cells was not influenced by clodronate after stimulation with LPS. On the other hand, the increase in IL6 secretion induced by M-CSF was significantly reduced after clodronate treatment. This discrepancy indicates that different intracellular pathways are responsible for LPS- and M-CSF-stimulated IL-6 secretion. To evaluate whether the decreased production of IL-6 may be due to a change in cell number or inhibition of proliferation, we treated isolated microglial cell cultures with various concentrations of LPS or M-CSF. In all investigated groups, no significant increase in the number of BrdU+ microglial cells was observed. This is in contrast with data from previous studies demonstrating a mitogenic effect of M-CSF on microglial cells (Lee et al., 1993, 1994; Liu et al., 1994). The lack of a microglial proliferative response in our study may be explained by the observation time of 48 h. Lee et al. reported significant differences in proliferation in their cultures 96 h after stimulation. Furthermore, a review of the literature reveals that most studies dealing with the mitogenic effects of M-CSF or LPS have been performed in vivo, in microglial/astrocytic co-cultures, or in cell lines from other species than rats (Ganter et al., 1992; Kloss et al., 1997; Raivich et al., 1998; Takeuchi et al., 2001; Vincent et al., 2002). It seems that isolated rat microglial cells exposed to M-CSF need additional factors, possibly produced by astrocytes, to proliferate. The mechanisms of action of bisphosphonates have not yet been fully elucidated. Frith et al. (1997), (2001) have shown that, in peritoneal macrophages, clodronate can be metabolized to adenosine 5V-(h,g-dichloromethylene) triphosphate (AppCCl2p), and the intracellular formation of AppCCl2p reduced the number of osteoclasts. Furthermore, the increase in the apoptosis rate of rat peritoneal macrophages induced by liposome-encapsulated clodronate could not be distinguished from that induced by liposome-encapsulated AppCCl2p. It is also believed that clodronate and its metabolites inhibit the DNA-binding activity of NF-nB and suppress the release of proinflammatory cytokines. This

250

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251

mechanism may also be responsible for effects of clodronate on microglial cells described in this study. It is well documented that activation of macrophages and microglial cells is accompanied by the release of a wide variety of proinflammatory cytokines (Elkabes et al., 1996; Galiano et al., 2001; Giulian et al., 1986; Murphy et al., 1998; Sauty et al., 1996), and these changes are mediated by NF-nB (Chen and Wang, 1999; Saccani et al., 2002). In summary, our experiments show that clodronate is a potent inhibitor of microglial cell activation by reducing the secretion of proinflammatory cytokines (TNF-a, IL-1h, and IL-6) and NO. Furthermore, the proliferation of microglial cells and astrocytes after excitotoxic neuronal injury is almost abolished after treatment with clodronate. In consequence, the application of clodronate in brain or spinal cord lesion models enables the investigation of neuron – glia interactions and may help to develop strategies supporting neuronal survival after CNS injuries.

Acknowledgments This study was supported by the Stiftung Friedrichsheim, the Dr. August Scheidel-Stiftung, the Paul und Ursula Klein-Stiftung, and the Medical Faculty of the Johann Wolfgang Goethe University. The authors gratefully acknowledge the expert technical assistance by Mr. Ch. Ghadban.

References Bal-Price, A., Brown, G.C., 2001. Inflammatory neurodegeneration mediated by nitric oxide from activated glia-inhibiting neuronal respiration, causing glutamate release and excitotoxicity. J. Neurosci. 21, 6480 – 6491. Bartholdi, D., Schwab, M.E., 1997. Expression of pro-inflammatory cytokine and chemokine mRNA upon experimental spinal cord injury in mouse: an in situ hybridization study. Eur. J. Neurosci. 9, 1422 – 1438. Benford, H.L., McGowan, N.W., Helfrich, M.H., Nuttall, M.E., Rogers, M.J., 2001. Visualization of bisphosphonate-induced caspase-3 activity in apoptotic osteoclasts in vitro. Bone 28, 465 – 473. Bethea, J.R., Nagashima, H., Acosta, M.C., Briceno, C., Gomez, F., Marcillo, A.E., Loor, K., Green, J., Dietrich, W.D., 1999. Systemically administered interleukin-10 reduces tumor necrosis factor-alpha production and significantly improves functional recovery following traumatic spinal cord injury in rats. J. Neurotrauma 16, 851 – 863. Beyer, M., Gimsa, U., Eyupoglu, I.Y., Hailer, N.P., Nitsch, R., 2000. Phagocytosis of neuronal or glial debris by microglial cells: upregulation of MHC class II expression and multinuclear giant cell formation in vitro. Glia 31, 262 – 266. Chen, C.C., Wang, J.K., 1999. p38 but not p44/42 mitogen-activated protein kinase is required for nitric oxide synthase induction mediated by lipopolysaccharide in RAW 264.7 macrophages. Mol. Pharmacol. 55, 481 – 488. Dehghani, F., Hischebeth, G.T., Wirjatijasa, F., Kohl, A., Korf, H.W., Hailer, N.P., 2003. The immunosuppressant mycophenolate mofetil attenuates neuronal damage after excitotoxic injury in hippocampal slice cultures. Eur. J. Neurosci. 18, 1061 – 1072. Dusart, I., Schwab, M.E., 1994. Secondary cell death and the inflammatory

reaction after dorsal hemisection of the rat spinal cord. Eur. J. Neurosci. 6, 712 – 724. Elkabes, S., DiCicco-Bloom, E.M., Black, I.B., 1996. Brain microglia/ macrophages express neurotrophins that selectively regulate microglial proliferation and function. J. Neurosci. 16, 2508 – 2521. Fleisch, H., 1989. Bisphosphonates: a new class of drugs in diseases of bone and calcium metabolism. Recent Results Cancer Res. 116, 1 – 28. Fleisch, H., 1991. Bisphosphonates. Pharmacology and use in the treatment of tumour-induced hypercalcaemic and metastatic bone disease. Drugs 42, 919 – 944. Frith, J.C., Monkkonen, J., Blackburn, G.M., Russell, R.G., Rogers, M.J., 1997. Clodronate and liposome-encapsulated clodronate are metabolized to a toxic ATP analog, adenosine 5V-(beta, gamma-dichloromethylene) triphosphate, by mammalian cells in vitro. J. Bone Miner. Res. 12, 1358 – 1367. Frith, J.C., Monkkonen, J., Auriola, S., Monkkonen, H., Rogers, M.J., 2001. The molecular mechanism of action of the antiresorptive and antiinflammatory drug clodronate: evidence for the formation in vivo of a metabolite that inhibits bone resorption and causes osteoclast and macrophage apoptosis. Arthritis Rheum. 44, 2201 – 2210. Galiano, M., Liu, Z.Q., Kalla, R., Bohatschek, M., Koppius, A., Gschwendtner, A., Xu, S., Werner, A., Kloss, C.U., Jones, L.L., Bluethmann, H., Raivich, G., 2001. Interleukin-6 (IL6) and cellular response to facial nerve injury: effects on lymphocyte recruitment, early microglial activation and axonal outgrowth in IL6-deficient mice. Eur. J. Neurosci. 14, 327 – 341. Ganter, S., Northoff, H., Mannel, D., Gebicke-Harter, P.J., 1992. Growth control of cultured microglia. J. Neurosci. Res. 33, 218 – 230. Giulian, D., Baker, T.J., Shih, L.C., Lachman, L.B., 1986. Interleukin 1 of the central nervous system is produced by ameboid microglia. J. Exp. Med. 164, 594 – 604. Guo, H., Tong, N., Turner, T., Epstein, L.G., McDermott, M.P., Kilgannon, P., Gelbard, H.A., 2000. Release of the neuronal glycoprotein ICAM-5 in serum after hypoxic – ischemic injury. Ann. Neurol. 48, 590 – 602. Hailer, N.P., Bechmann, I., Heizmann, S., Nitsch, R., 1997. Adhesion molecule expression on phagocytic microglial cells following anterograde degeneration of perforant path axons. Hippocampus 7, 341 – 349. Hailer, N.P., Heppner, F.L., Haas, D., Nitsch, R., 1998. Astrocytic factors deactivate antigen presenting cells that invade the central nervous system. Brain Pathol. 8, 459 – 474. Heppner, F.L., Roth, K., Nitsch, R., Hailer, N.P., 1998. Vitamin E induces ramification and down-regulation of adhesion molecules in cultured microglial cells. Glia 22, 180 – 188. Hoffmann La Roche Ltd., 1998. Ostac product monograph. Mississauga, Ontario. Hurst, M., Noble, S., 1999. Clodronate: a review of its use in breast cancer. Drugs Aging 15, 143 – 167. Kloss, C.U., Kreutzberg, G.W., Raivich, G., 1997. Proliferation of ramified microglia on an astrocyte monolayer: characterization of stimulatory and inhibitory cytokines. J. Neurosci. Res. 49, 248 – 254. Kohl, A., Dehghani, F., Korf, H., Hailer, N.P., 2003. The bisphosphonate clodronate depletes microglial cells in excitotoxically injured organotypic hippocampal slice cultures. Exp. Neurol. 181, 1 – 11. Kreutzberg, G.W., 1996. Microglia: a sensor for pathological events in the CNS. Trends Neurosci. 19, 312 – 318. Lee, S.C., Liu, W., Roth, P., Dickson, D.W., Berman, J.W., Brosnan, C.F., 1993. Macrophage colony-stimulating factor in human fetal astrocytes and microglia. Differential regulation by cytokines and lipopolysaccharide, and modulation of class II MHC on microglia. J. Immunol. 150, 594 – 604. Lee, S.C., Liu, W., Brosnan, C.F., Dickson, D.W., 1994. GM-CSF promotes proliferation of human fetal and adult microglia in primary cultures. Glia 12, 309 – 318. Liu, W., Brosnan, C.F., Dickson, D.W., Lee, S.C., 1994. Macrophage colony-stimulating factor mediates astrocyte-induced microglial ramification in human fetal central nervous system culture. Am. J. Pathol. 145, 48 – 53. Liu, Y., Qin, L., Li, G., Zhang, W., An, L., Liu, B., Hong, J.S., 2003.

F. Dehghani et al. / Experimental Neurology 189 (2004) 241–251 Dextromethorphan protects dopaminergic neurons against inflammation-mediated degeneration through inhibition of microglial activation. J. Pharmacol. Exp. Ther. 305, 212 – 218. Makkonen, N., Hirvonen, M.R., Teravainen, T., Savolainen, K., Monkkonen, J., 1996. Different effects of three bisphosphonates on nitric oxide production by RAW 264 macrophage-like cells in vitro. J. Pharmacol. Exp. Ther. 277, 1097 – 1102. Makkonen, N., Salminen, A., Rogers, M.J., Frith, J.C., Urtti, A., Azhayeva, E., Monkkonen, J., 1999. Contrasting effects of alendronate and clodronate on RAW 264 macrophages: the role of a bisphosphonate metabolite. Eur. J. Pharm. Sci. 8, 109 – 118. Minghetti, L., Levi, G., 1998. Microglia as effector cells in brain damage and repair: focus on prostanoids and nitric oxide. Prog. Neurobiol. 54, 99 – 125. Murphy Jr., G.M., Yang, L., Cordell, B. 1998. Macrophage colony-stimulating factor augments beta-amyloid-induced interleukin-1, interleukin-6, and nitric oxide production by microglial cells. J. Biol. Chem. 273, 20967 – 20971. Nicholas, R.S., Compston, A., Brown, D.R., 2001. Inhibition of tumour necrosis factor-alpha (TNFalpha)-induced NF-kappaB p52 converts the metabolic effects of microglial-derived TNFalpha on mouse cerebellar neurones to neurotoxicity. J. Neurochem. 76, 1431 – 1438. Oudega, M., Vargas, C.G., Weber, A.B., Kleitman, N., Bunge, M.B., 1999. Long-term effects of methylprednisolone following transection of adult rat spinal cord. Eur. J. Neurosci. 11, 2453 – 2464. Polfliet, M.M., Goede, P.H., Kesteren-Hendrikx, E.M., van Rooijen, N., Dijkstra, C.D., van den Berg, T.K., 2001. A method for the selective depletion of perivascular and meningeal macrophages in the central nervous system. J. Neuroimmunol. 116, 188 – 195. Raivich, G., Moreno-Flores, M.T., Moller, J.C., Kreutzberg, G.W., 1994. Inhibition of posttraumatic microglial proliferation in a genetic model of macrophage colony-stimulating factor deficiency in the mouse. Eur. J. Neurosci. 6, 1615 – 1618. Raivich, G., Haas, S., Werner, A., Klein, M.A., Kloss, C., Kreutzberg, G.W., 1998. Regulation of MCSF receptors on microglia in the normal and injured mouse central nervous system: a quantitative immunofluorescence study using confocal laser microscopy. J. Comp. Neurol. 395, 342 – 358. Rogers, M.J., Chilton, K.M., Coxon, F.P., Lawry, J., Smith, M.O., Suri, S.,

251

Russell, R.G., 1996. Bisphosphonates induce apoptosis in mouse macrophage-like cells in vitro by a nitric oxide-independent mechanism. J. Bone Miner. Res. 11, 1482 – 1491. Russell, R.G., Rogers, M.J., 1999. Bisphosphonates: from the laboratory to the clinic and back again. Bone 25, 97 – 106. Saccani, S., Pantano, S., Natoli, G., 2002. p38-Dependent marking of inflammatory genes for increased NF-kappa B recruitment. Nat. Immunol. 3, 69 – 75. Sauty, A., Pecherstorfer, M., Zimmer-Roth, I., Fioroni, P., Juillerat, L., Markert, M., Ludwig, H., Leuenberger, P., Burckhardt, P., Thiebaud, D., 1996. Interleukin-6 and tumor necrosis factor alpha levels after bisphosphonates treatment in vitro and in patients with malignancy. Bone 18, 133 – 139. Schmitt, A.B., Buss, A., Breuer, S., Brook, G.A., Pech, K., Martin, D., Schoenen, J., Noth, J., Love, S., Schroder, J.M., Kreutzberg, G.W., Nacimiento, W., 2000. Major histocompatibility complex class II expression by activated microglia caudal to lesions of descending tracts in the human spinal cord is not associated with a T cell response. Acta Neuropathol. (Berl) 100, 528 – 536. Takeuchi, A., Miyaishi, O., Kiuchi, K., Isobe, K., 2001. Macrophage colony-stimulating factor is expressed in neuron and microglia after focal brain injury. J. Neurosci. Res. 65, 38 – 44. Thery, C., Dobbertin, A., Mallat, M., 1994. Downregulation of in vitro neurotoxicity of brain macrophages by prostaglandin E2 and a betaadrenergic agonist. Glia 11, 383 – 386. van Rooijen, N., Sanders, A., van den Berg, T.K., 1996. Apoptosis of macrophages induced by liposome-mediated intracellular delivery of clodronate and propamidine. J. Immunol. Methods 193, 93 – 99. Vincent, V.A.M., Tilders, F.J.H., Van Dam, A.-M., 1997. Inhibition of endotoxin-induced nitric oxide synthase production in microglial cells by the presence of astroglial cells: a role for transforming growth factor beta. Glia 19, 190 – 198. Vincent, V.A., Selwood, S.P., Murphy Jr., G.M., 2002. Proinflammatory effects of M-CSF and A beta in hippocampal organotypic cultures. Neurobiol. Aging 23, 349 – 362. Wilms, H., Hartmann, D., Sievers, J., 1997. Ramification of microglia, monocytes and macrophages in vitro: influences of various epithelial and mesenchymal cells and their conditioned media. Cell Tissue Res. 287, 447 – 458.