HPK1-IN-2

Suramin Induces and Enhances Apoptosis in a Model of Hyperoxia-induced Oligodendrocyte Injury

SIMONE STARKa,b,, ALEXANDRA SCHÜLLERa,, MARCO SIFRINGERa, BETTINA GERSTNERa,
FELIX BREHMERa, SVEN WEBERa, RODICA ALTMANNa, MICHAEL OBLADENa, CHRISTOPH BÜHRERc and URSULA FELDERHOFF-MUESERa,*

A Department of Neonatology, Campus Virchow-Klinikum, Charité, Universitätsmedizin Berlin, D-13353 Berlin, Germany;
B Department of Pediatrics, Catholic Children’s Hospital Wilhelmstift, D-22149 Hamburg, Germany; and
C Department of Neonatology, Basel University Children’s Hospital, Basel, Switzerland. [email protected] equal contribution

Abstract
Recent evidence suggests oxygen as a powerful trigger for cell death in the immature white mat- ter, leading to periventricular leukomalacia (PVL) as a cause of adverse neurological outcome in survivors of preterm birth. This oligodendrocyte (OL) death is associated with oxidative stress, upregulation of apoptotic signaling factors (i.e., Fas, caspase-3) and decreased amounts of neu- rotrophins. In search of neuroprotective strate- gies we investigated whether the polysulfonated urea derivative suramin, recently identified as a potent inhibitor of Fas signaling, affords neuro- protection in an in vitro model of hyperoxia- induced injury to immature oligodendrocytes.
Immature OLs (OLN-93) were subjected to 80% hyperoxia (48 h) in the presence or absence of suramin (0, 30, 60, 120 M). Cell death was assessed by flow cytometry (Annexin V, caspase-3 activity assay) and immunohistochemistry for activated caspase-3. Immunoblotting for the death receptor Fas, cleaved caspase-8 and the phosphorylated isoform of the serine-threonin kinase Akt (pAkt) was performed. Suramin led to OL apoptosis and potentiated hyperoxia-induced injury in a dose-dependent manner. Immunoblotting revealed increased Fas and caspase-8 expression by suramin treatment.
This effect was significantly enhanced when suramin was combined with hyperoxia. Furthermore, pAkt levels decreased following suramin exposure, indicating interference with neurotrophin-dependent growth factor signaling. These data indicate that suramin causes apop- totic cell death and aggravates hyperoxia-induced cell death in immature OLs. Its mechanism of action includes an increase of previously described hyperoxia-induced expression of pro-apoptotic factors and deprivation of growth factor dependent signaling components.

Keywords: Hyperoxia; Suramin; Periventricular leu- komalacia; White matter

INTRODUCTION
Survivors of preterm birth often suffer from adverse motor and cognitive development, which can only partly be attributed to major neurologic insults (Hintz et al., 2005; Marlow et al., 2005). Periventricular leukomalacia (PVL) in preterm infants underlies the development of cerebral palsy (CP) and its incidence peaks at 23-32 weeks gesta- tion. During this developmental period extensive oligodendrocyte (OL) migration and maturation take place. Various insults during this critical period of brain maturation such as hypoxia-ischemia and inflammation can affect oligodendrocyte develop- ment and result in disturbed myelination (Kaur et al., 2006; Wang et al., 2006).
Epidemiological evidence suggests that oxygen, which is widely used in neonatal intensive care, may be associated with the development of CP (Collins et al., 2001). Recent experimental studies have identified hyperoxia as a powerful trigger for widespread apoptotic neuronal death in the devel- oping rodent brain. Hyperoxia increased the amount of degenerating cells in various brain regions of seven-day-old Wistar rats and C57/BL6 mice such as the caudate nucleus, the cortex, white matter tracts and the periventricular region. Cell death is associated with upregulation of members of the apoptotic cascade, oxidative stress and decreased expression of neurotrophins and neurotrophin- dependent pathways such as the phosphorylated isoforms of the serine-threonin kinase Akt (p-Akt, protein kinase B), and members of the mitogen- activated protein kinase (MAPK) extracellular sig- nal related protein kinase ERK1/2 (p-ERK1/2) pathway, which mediate intracellular signaling fol- lowing activation of receptor tyrosin kinases (Trk) by growth factors (Felderhoff-Mueser et al., 2004). Furthermore hyperoxia triggers caspase-1 depen- dent pro-inflammatory pathways, microvascular degeneration, diminished brain mass and cerebral functional deficits (Felderhoff-Mueser et al., 2005; Sirinyan et al., 2006).
Recently, our group demonstrated that exposure to hyperoxia induced apoptosis in immature OLs and pre-oligodendroglial cells (Gerstner et al., 2006). This cell death was associated with altered development of white matter structures and upregu- lation of Fas death receptor, a major component of the extrinsic apoptotic signal transduction system (Gerstner et al., 2007). To limit the neurotoxic effects of hyperoxia in the developing brain, neuro- protective strategies are highly warranted. Since apoptosis has been identified as the key pathogenic factor in hyperoxia-induced brain injury, pharma- cological intervention in the apoptotic signal trans- duction machinery seems a feasible therapeutic approach.
Suramin is a symmetrical polysulfonated naphthylamine derivative of urea and is effective in prophylaxis and treatment of early stages of human trypanosomiasis and onchocerciasis in adults and also in young children (Rolland and Thylefors, 1982; Voogd et al., 1993). Due to its anti-reverse transcriptase and anti-proliferative activity, it is used for treatment of human acquired immunodefi- ciency syndrome (AIDS), (Mitsuya et al., 1984). It has also been shown to be effective as an anti-can- cer agent (La Rocca et al., 1990). Previous studies suggest anti-apoptotic actions of this compound in fulminant hepatic failure: Suramin inhibits apopto- sis in hepatoma cells through the death receptors CD95/Fas by reduced activation of the downstream death-inducing signaling complex (DISC), and by inhibition of initiator caspases. Futhermore, Fas induced apoptotic liver damage was attenuated by suramin (Eichhorst et al., 2004). Suramin sup- presses NF-B activity from macrophages by inhi- bition of TNF- and IL-6 production and shows therapeutic effects on acute liver damage (Goto et al., 2006). In addition, a neuroprotective effect of suramin has been described in rat cerebellar granule cells, mouse neuroblastoma cells and in a model of cerebral ischemia (Bezvenyuk et al., 2000; Kharlamov et al., 2002).
In search for a potential neuroprotectant for the developing white matter, we examined in the present study the effect of suramin on hyperoxia-induced apoptotic signaling in immature oligodendrocytes (OLN-93).

MATERIAL AND METHODS

Oligodendrocyte Culture
OLN-93 is a permanent oligodendrocyte cell line derived from spontaneously transformed cells in primary rat brain glial culture (Richter-Landsberg and Heinrich, 1996), that was kindly provided by Dr. C. Richter-Landsberg (Institute of Molecular Neurobiology, Oldenburg, Germany). The morpho- logical and immunocytochemical properties of OLN-93 cells equalize 5- to 10-day-old (postnatal age) cultured rat brain immature oligodendrocytes and express the classical stage specific surface markers (O4+O1+MBP-). These cells resemble the intermediate stage between pre-oligodendrocytes (O4+O1-MBP-) and mature oligodendrocytes (O4+O1+MBP+). OLN-93 cells also express AMPA receptors (Gerstner et al., 2005) similar to those found in primary OL cultures.
Cells were cultured in Dulbecco’s Modified Eagle Medium (DMEM, Life Technologies, Inc, Rockville, MD, USA), including 3.7 g/L NaHCO3, 25 mM HEPES (Sigma-Aldrich, Steinheim, Germany), 4.5g/l D-Glucose, 4.4 g/l NaCl, containing heat inacti- vated 10% FCS (complete medium), (Biochrom, Berlin, Germany) and 1% human serum albumin (Aventis Behring, Marburg, Germany). At 37°C monolayers were cultured in flasks or flat-bottom well microtiter plates in a humidified, 5% CO2 atmosphere with medium replenishment every two to three days. For hyperoxia experiments, cells were kept in a humidified, 80% oxygen/15% air/5% CO2 atmosphere at 37°C for a defined period of time. Control plates were kept under normoxic 5% CO2 conditions at 37°C.

Suramin Treatment Protocol
OLs were treated with 80% hyperoxia for 48 h in the presence or absence of suramin (Sigma-Aldrich), diluted in cell culture media. Cells were pretreated with suramin (30, 60 and 120 M) for 24 h, fol- lowed by exposure to 21% or 80% oxygen for 48 h. Controls were left untreated and were either kept in room air or in 80% oxygen for 48 h.

Colorimetric Viability Assay
The MT (3-(4,5-dimethylthiazol-2-yl)-5-(3- carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tet- razolium) assay (CellTiter 96® AQuenous One Solution Cell Proliferation Assay, Promega, Madison, USA) was used to estimate cell viability. The yellow dye MTT tetrazolium compound is bioreduced by cells into a colored formazan prod- uct that is soluble in tissue culture medium. This conversion is presumably accomplished by NADPH or NADH produced by dehydrogenase enzymes in metabolically active cells, but not by dead cells. Cells were seeded in flat-bottom 96-well microtiter plates (5,000 cells/well) and developed at 37°C. Then, OL were maintained under normoxic (21% O2) conditions or exposed to hyperoxic (80% O2) conditions for 48 h. Finally, 10 l (to 50 l culture medium) MTS tetrazolium compound was added to the wells for 2 h followed by recording absorbance at 490 nm with a 96-well microplate reader (Bio- Rad, Munich, Germany). The quantity of the formazan product generated is directly proportional to the number of living cells in culture. A 1 mM stock solution of the proteinkinase inhibitor staurosporine (SSP, Sigma-Aldrich) was prepared in dimethylsul- foxide (0.1%) and used in a final concentration of 1 M as a positive control for apoptotic cell death.

Assessment of Cell Death by Flow Cytometry

Annexin V Assay
A key feature of apoptosis is the loss of plasma membrane integrity. In healthy cells, membrane phospholipids are distributed asymmetrically between the inner and outer leaflets of the plas- ma membrane. Phosphatidylserine, an amino- phospholipid, is normally present in the inner leaflet of the plasma membrane. Early in apopto- sis, before loss of membrane integrity, phos- phatidylserine translocates from the inner to the outer leaflet, exposing it to the external cellular environment at the surface of the plasma mem- brane. Phosphatidylserine translocation can be assessed by Ca2+-dependent high-affinity bind- ing to Annexin V. Recombinant human Annexin V FITC (Bender Medsystems, Vienna, Austria) was used to detect apoptosis determined by flow cytometry (FL-1 channel) using a fluorescence- activated cell sorter (FACScan, Becton Dickinson, Heidelberg, Germany). Discrimination between apoptotic and necrotic events was performed with propidium iodide (50 g/ml, Bender Medsystems) as a second fluorescent marker. Cells undergoing necrosis showed subsequent propidium iodide uptake by a strong signal in the FL-2 channel of the flow cytometer. Data were analyzed by Win MDI 2.8 software (Windows Multiple Document Interface for Flow Cytometry; http://facs.scripps.edu/software.html).

Activated Caspase-3 Assay
The activation of caspase-3 has been shown to play a central role in mediating the effector stage of apoptosis by initiating the process of DNA frag- mentation. Caspase-3 in living cells was deter- mined by the Caspase-3 Activity Assay (Oncogene, San Diego, CA, USA) using a FITC-labelled monoclonal antibody directed against the activat- ed (cleaved) form of caspase-3 in apoptotic cells. FITC, as the fluorescent marker, allows for direct detection of caspase-3 in apoptotic cells by flow cytometry. Samples were evaluated by using the FL-1 channel of the flow cytometer and analyzed by Win MDI 2.8 software.

Assessment of Cell Death by Caspase-3 Immunohistochemistry
FITC-labelled activated caspase-3 antibody (Oncogene) was applied to visualize the binding to apoptotic OL cells in cultures grown in 8-well chamber slides. Cells were washed with PBS, then incubated with 4% PFA in PBS for 15 min and washed with TBS followed by incubation with caspase-3 antibody for 1 h. Cells were washed with TBS and then mounted on coverslips using Vectashield Mounting Medium with DAPI (Vector Laboratories, Biozol, Eching, Germany). Cells were visualized with an epifluorescent microscope (Axiovision, Zeiss, Jena, Germany).

Immunoblotting for Fas, Caspase-8 and Akt/pAkt
Fas is a cell surface type I-membrane glycopro- tein. Binding to its ligand leads to apoptosis of Fas expressing cells (Suda and Nagata, 1994). Caspase-8 takes advantage of the specificity of the enzymes for cleavage of aspartate residues in a particular peptide sequence, present in the sub- strate IETD. Caspase-8 then activates downstream caspases, such as caspase-3, which execute the cell (Fischer et al., 2003). Akt is a serine/threonine kinase that plays a crucial role in cell survival and apoptosis (Kaplan and Miller, 2000). RIPA buffer with 1% NP40 (Fluka, Buchs, Switzerland), 0.5% sodium deoxycholate (Fluka), 0.1% SDS (Serva, Heidelberg, Germany), 1 mM EDTA (Serva), 1 mM EGTA (Serva), 1 mM Na3VO4 (Sigma-Aldrich), 20 mM NaF (Fluka), 0.5 mM DTT (Sigma-Aldrich), 1 mM PMSF (Sigma-Aldrich), and protease inhibitor cocktail (Fluka) in PBS pH 7.4 was used to lyse the cells, lysates were collected, and sonicated for 12 sec. With the Bio-Rad Dc Protein assay (Bio-Rad) protein concentrations were determined. Protein extracts resulted, 20 g of them were heat dena- turated in Laemmli sample loading buffer, sepa- rated by 10% sodium dodecyl sulfate polyacryl- amide gel electrophoresis, and electrotransferred onto a nitrocellulose membrane. To prevent non- specific protein binding, the membrane was treated with 5% nonfat dry milk (Fluka) in Tris- buffered saline/0.1% Tween 20 (Serva) for 2 h at RT. Staining the membranes with Ponceau S solution (Fluka), an equal loading and transfer of proteins was confirmed. Overnight the mem- brane was incubated at 4°C with a mouse mono- clonal anti-Fas (1:1000; Transduction Laboratories, Lexington, KY, USA), a rabbit polyclonal anti-caspase-8 (1:200; Chemicon, Temecula, CA, USA), a rabbit polyclonal anti- Akt (1:1000; Cell Signaling Technology, Beverly, MA, USA) or a rabbit polyclonal anti-phospho- Akt (1:1000, Cell Signaling) antibody. In a sec- ond incubation step horseradish peroxidase- linked anti-mouse (1:10000; DAKO, Glostrup, Denmark) or anti-rabbit (1:2000; Amersham Biosciences, Bucks, UK) antibody was applied. Blots were progressed by stripping, as mem- branes were incubated with stripping buffer (100 mM -mercaptoethanol, 2% SDS, 62.5 mM Tris- HCl, pH 6.7) at 50°C for 30 min, then washed, blocked, and reprobed overnight at 4°C with mouse anti--actin monoclonal antibody (1:8000; Sigma-Aldrich). By enhanced chemilumines- cence (ECL; Amersham Biosciences) positive signals were visualized and serial exposures were made to radiographic film (Hyperfilm ECL; Amersham Biosciences). Densitometric analysis of blots was performed using the image analysis program BioDocAnalyze (Whatman Biometra, Göttingen, Germany).

Statistical Analysis
Values are presented as mean ± SEM. The sig- nificance of differences among groups was deter- mined by analysis of variance (two-way ANOVA) followed by Bonferroni post-tests. For all tests, p <0.05 was considered significant. RESULTS Previously, we have shown that exposure of the OLN-93 cell line to 80% oxygen for 48 h reduces cell viability to 60%, and after 96 h almost no cells were viable. OLs exhibited the morphologi- cal features of apoptosis, including nuclear con- densation (Gerstner et al., 2006). Suramin Treatment Potentiates Hyperoxia-induced OL Death Cells were pre-incubated with different concentra- tions of suramin for 24 h and either exposed to 80% oxygen for 48 h or kept in room air. Cell viability as detected by MTS assay decreased upon treat- ment with increasing concentrations of suramin (30 M, 60 M, 120 M Suramin). There was a potentiation of OL death when suramin was combined with 80% hyperoxia (FIG. 1A). During early apoptosis phosphatidylserine is externalized and becomes accessible for Annexin V binding. When OLs were exposed to various con- centrations of suramin (30, 60, 120 M) for 72 h under room air conditions, FITC-labelled Annexin V binding was increased in a dose-dependent man- ner compared to untreated controls. Exposure to hyperoxia for 48 h after preincubation with three different suramin doses for 24 h further enhanced the amount of Annexin V labelled cells (FIG. 1B). A FITC-labelled monoclonal antibody specific to the activated form of the effector caspase-3 assessed the activation of the enzyme. OLs, treated with 120 M suramin displayed a 2.5-fold increase of activated caspase-3 compared to controls. The apop- totic death rate of OLs without suramin was 8-10% in controls maintained in 21% oxygen, as published previously (Gerstner et al., 2006). When OLs were exposed to 80% oxygen for 48 h after 24 h pre- incubation with suramin (120 M), the number of (B) Representative transmission light phase contrast photomicrographs of native immature oligodendrocyte cell cultures. Photomicrographs were taken by Olympus digital camera. Untreated control cells do not show signs of apop- tosis (Panel a), whereas treated cells grow with less density and show plasma membrane blebbing, and nuclear con- densation. Panel a: Untreated control OLN-93 cells were kept in 21% O2 for 72 h. Panel b: Untreated OLN-93 cells exposed to 80% O2 for 48 h. Panel c / Panel d: OLN-93 cells incubated with suramin (representative image with 60 M suramin) for 24 h and kept under room air conditions. OLN-93 cells pre-treated with 60 M suramin and kept in an 80% O2 environment for 48 h showing signs of apoptosis such as complete loss of processes, cell shrinkage (black arrowheads), plasma membrane blebbing (white arrowheads), and nuclear condensation (grey arrowhead). cells which stained positive for the activated form of caspase-3 was increased 1.6-fold compared to cells treated with suramin and 48 h 21% oxygen, and 3.1-fold compared to OLs kept in room air (FIG. 1C, 1D). Representative transmission light phase contrast photomicrographs and immunofluorescence photo- micrographs of OLs stained for activated caspase-3 showed few positive cells when untreated OLs were kept under room air conditions, representing the apoptotic cell death rate in controls. Exposure to hyperoxia for 48 h displayed increased numbers of apoptotic cells staining positive for activated cas- pase-3. Upon pretreatment with suramin (60 M) followed by a 48 h incubation under room air conditions the number of apoptotic and activated cas- pase-3 positive cells increased. A combination of suramin treatment and exposure to hyperoxia fur- ther reduced the amount of living cells and increased the numbers of cleaved caspase-3 positive cells (FIG. 2A, 2B). Suramin Treatment Increases Fas Death Receptor Expression Fas can function as a death receptor by inducing apoptosis (Felderhoff-Mueser et al., 2000). Previously, we showed an increase in Fas expres- sion following 80% oxygen exposure in the oligo- dendroglia cell line at different time points (6, 12 h) by immunoblotting (Gerstner et al., 2007). M which is increased by a combination of suramin and 80% O2. White bars represent cells exposed to 21% O2; black bars represent exposure to 80% O2. The densito- metric data represent the density ratio of respective bands to the corresponding -actin band (means and SEM). Data are normalized to levels of control cells (control=100%). Significance *p <0.05, ***p <0.001 (ANOVA). (B) Representative of a series of Western blots demonstrating a strong increase in the protein lev- els of Fas protein upon treatment with either suramin or oxygen or a combination of both. M) for 24 h and consecutively kept in either room air or 80% O2 environment for 48 h. (A) Western blot analysis by densitometry of activated caspase-8, demon- strating a significant increase upon treatment with suramin at a concentration of 30 M, which is even more increased by a combination of suramin and 80% O2. White bars represent cells exposed to 21% O2; black bars represent exposure to 80% O2. Densitometric data represent the density ratio of respective bands to the cor- responding -actin band (means and SEM). Data are normalized to levels of control cells (control=100%). Significance *p <0.05 (ANOVA). (B) Representative of a series of Western blots demonstrating a strong increase in the protein levels of activated caspase-8 protein upon treatment with either suramin or O2 or a combination of both. In the present study, immunoblotting with spe- cific antibodies against the death receptor Fas revealed an increased expression when treated with suramin (30 M, 60 M or 120 M) for 72 h in 21% oxygen. Pretreatment with suramin for 24 h in 21% oxygen and additional exposure to 80% hyper- oxia for 48 h significantly aggravated this upregula- tion (FIG. 3A, 3B). Suramin Causes Activation of the Initiator Caspase-8 Caspase-8 is an initiator caspase acting downstream of Fas cell surface receptor activation. Caspase-8 is recruited to the DISC upon Fas receptor activation (Kruidering and Evan, 2000; Curtin and Cotter, 2003). Immunoblotting with specific antibodies against cleaved caspase-8 revealed an increased expression after treatment with suramin (30, 60 or 120 M) for 72 h in 21% oxygen. Pretreatment with suramin for 24 h and additional exposure to 80% hyperoxia for 48 h significantly increased this expression pattern (FIG. 4A, 4B). Suramin Potentiates Hyperoxia-induced Downregulation of Growth Factor Signaling Dependent Pathways Upon activation by growth factors such as nerve growth factor (NGF), Akt, a downstream kinase of phosphoinositide 3-kinase (PI3K), is a strong pro- motor of survival signals in the central nervous system (CNS). Targets of Akt include Bad, an inhibitor of the Bcl-2 anti-apoptotic protein; pro- caspase-9, whose cleavage into the pro-apoptotic caspase-9 is inhibited (Kaplan and Miller, 2000). To investigate the impact of treatment with suramin on levels of the phosphorylated isoform of Akt, we performed Western blot analysis on protein samples obtained from OLN-93 cultures kept in an 80% oxygen environment. Immunoblotting with specific antibodies against the phosphorylated active isoform of Akt (pAkt) revealed a dose-dependent decreased expression after treatment with 30 M, 60 M or 120 M suramin for 72 h in 21% oxygen. Additional expo- sure to 80% hyperoxia for 48 h and treatment with suramin for 24 h significantly potentiated oxygen- induced downregulation of pAkt in a dose-depen- dent manner. The non-phosphorylated isoform of Akt remained unaffected (FIG. 5A, 5B). (B) Representative of a series of Western blots demonstrating a reduction in the protein levels of pAkt protein upon treatment with either suramin or O2, or a combina- tion of both, whereas Akt (phosphorylation independent) levels remain unaffected. DISCUSSION In the present study we have investigated the effect of suramin on hyperoxia-induced apoptosis in the immature oligodendrocyte cell line OLN-93. We show that suramin leads to OL cell death and potentiates hyperoxia-induced apoptosis in a dose- dependent manner. Suramin has been widely used as a treatment for various diseases including trypanosomiasis, O. volvulus and HIV infections. Its efficacy as a che- motherapeutic agent has been shown for different types of cancer such as prostate cancer. Interference with growth factor receptor function (Gorospe et al., 1996), induction of lysosomal storage defects (Gritli-Linde et al., 1994), inhibition of enzymes involved in cell growth and proliferation such as protein kinase C may contribute to the cytostatic activity of the compound. Beside its antineoplastic activity, suramin prevents apoptosis in fulminant hepatic failure by inhibition of pro-inflammatory cytokines and suppression of NF-B activity (Goto et al., 2006). Suramin has also a marked antiviral activity capable of inhibiting reverse transcriptase of a number of retroviruses (Mitsuya et al., 1984). However, the mechanisms underly- ing the various effects of suramin treatment are poorly understood. The experiments presented here demonstrate that suramin has pro-apoptotic effects on immature oligodendrocytes, especially when these cells are exposed to increased concentrations of oxygen. This is in contrast to results obtained in rat cere- bellar granular neurons undergoing apoptosis in response to glutamate (Bezvenyuk et al. 2000) or dequalinium (Chan and Lin-Shiau, 2000), mouse neuroblastoma cells treated with staurosporine (Bezvenyuk et al., 2000), or human hepatic or T cell lines following Fas ligation (Eichhorst et al., 2004). The latter investigation has elegantly dem- onstrated that the protective effects of suramin are highly specific for the cell type and apoptosis- inducing agent. While suramin does not prevent apoptosis induced by dexamethasone or gamma- irradiation, suramin has been shown to protect Jurkat and HepG2 cells against apoptosis induced by Fas ligation. These cells display only limited caspase-8 cleavage in response to Fas ligation and require subsequent cleavage of Bid and mitochondrial activation to ultimately activate caspase-3 (Scaffidi et al., 1998). In contrast, B-lymphoblastoid SKW6.4 cells that have strong activation of cas- pase-8 and direct activation of caspase-3 in response to Fas ligation, are not protected by suramin. Increased concentrations of suramin even induce apoptosis in LNCaP prostate cancer cells, particularly when combined with Fas liga- tion (Eichhorst et al., 2004). We found that in immature OLN93 oligodendrocytes, suramin treatment directly increased Fas expression, cas- pase-8 cleavage and apoptotic death. These effects were aggravated by exposure to elevated oxygen which in itself increases Fas expression, cas- pase-8 cleavage, and apoptotic cell death. Furthermore, pAkt levels decreased following suramin exposure, indicating interference with neurotrophin-dependent growth factor signaling. Thus, there is a notable discrepancy between the antiapoptotic effect of suramin in T lymphoid and hepatic cell lines, the lack of effect in B lympho- blastoid cell lines, and the proapoptotic action of suramin in prostate carcinoma cells or immature oligodendroglia cells. This cell type specificity has been claimed in part on the caspase-9-mediat- ed need for mitochondrial amplification of the death receptor signal in those cells that are res- cued by suramin from Fas ligation-induced apop- tosis (Eichhorst et al., 2004) and on the differen- tial dependence on growth factors for cell sur- vival. Indeed, pAkt levels decreased following suramin exposure in OLN93 immature oligoden- drocytes, indicating interference with neurotro- phin-dependent growth factor signaling. However, the discrepancy between the proapoptotic and anti-apoptotic action of suramin at present defies a definitive explanation, and elucidation of the contribution of the proposed mechanisms awaits future experimental endeavours. The background for our studies on the effect of suramin on hyperoxia-induced oligodendrocyte apoptosis was the observation that hyperoxia leads to cell death via Fas receptor activation and involvement of caspase-8 and the effector cas- pase-3 (Gerstner et al., 2006; 2007). A drug such as suramin targeting this pathway at several steps within the apoptotic machinery seemed feasible for testing in our experimental setting. However, the pro-apoptotic findings of the present study in immature oligodendrocytes are in contrast to the anti-apoptotic findings reported for liver cells and in fulminant liver damage. In our model suramin acted via direct enhancement of Fas death receptor expression and upregulation of the activated (cleaved) form of caspase-8. Fas and cleaved caspase-8 were upregulated by suramin treatment alone and largely enhanced by a combi- nation of hyperoxia and suramin. The toxic effect of suramin was still present when doses were used, that were over a third lower than applied in other in vitro studies demonstrating a neuropro- tective effect of this compound (Chan and Lin- Shiau, 2000). Direct involvement of death receptors, particu- larly that of Fas, has been shown in another experimental setting of chemotherapeutic drug- induced apoptosis in mouse thymocytes in vivo (Eichhorst et al., 2001). These findings combined with the pro-apoptotic results of our study points towards a key role of Fas death receptor itself in the pro- or anti-apoptotic action of suramin, depending on the experimental setting. In previous studies investigating oxygen toxicity in the immature CNS we presented evidence that hyperoxia-exposure leads to changes in gene expression and phosphorylation of proteins that control neuronal survival during development. Hyperoxia depressed synthesis of BDNF, NGF, NT-3 and NT-4 and reduced levels of the active forms of Ras, ERK1/2 and Akt in a time depen- dent fashion in several brain regions such as thala- mus, cortex and also in white matter (Felderhoff- Mueser et al., 2004; Gerstner et al., 2007). In addition the vulnerability of developing oligoden- drocytes was associated with a decrease of the survival promoting pathways mediated by pAkt and pERK1/2. In our present experimental setting suramin further potentiated the growth factor sig- naling impairment caused by hyperoxia treatment in the immature oligodendrocyte lineage. A vari- ety of growth factors can be targeted by suramin, which is known to be a potent antagonist of angio- genesis by inhibition of fibroblast growth factor (FGF) signaling (Kathir et al., 2006). In SH-SY5Y human neuronal cells, suramin acted as a cyto- static agent by blocking insulin-like growth factor (IGF)-II-dependent cell growth by preventing IGF receptor activation (Sullivan et al., 1997). However, survival promoting properties have also been described. Suramin promotes proliferation of renal epithelial cells and was shown to stimulate outgrowth, scattering, and proliferation of primary cultures of renal proximal tubule cells by enhancing phosphorylation of Akt and ERK1/2 (Zhuang and Schnellmann, 2005), in contrast to results from the present study. There are several controversial reports on either cytoprotective and also cytotoxic actions of suramin on neuronal cells. Suramin was shown to reduce the infarct volume in a model of focal brain ischemia in rats, indicating its potential use for CNS diseases. Its mechanism of action was attributed to its ability to block ionotropic ligand-gated ion channel puri- noceptors (P(2X)), responsible for increased calci- um influx during excitotoxic injury (Kharlamov et al., 2002). Furthermore, suramin abolished dequal- inium (an anticancer drug) neurotoxicity by target- ing the mitochondrial transition pore and restora- tion of the mitochondrial membrane potential (Chan and Lin-Shiau, 2000). However, induction of apop- tosis has been proposed for the cytotoxic effect of suramin in dorsal root ganglion neurons (Gill and Windebank, 1998). In these cells, exposure of cul- tures to suramin led to apoptotic cell death, pre- ceded by a significant increase in intracellular accumulation of ceramide. The accumulation of ceramide following suramin treatment mediated activation and nuclear translocation of NFB, fol- lowed by cyclin D1 protein expression and caspase cleavage (Gill and Windebank, 2000). Ceramide has pleiotropic effects, ranging from induction of cell differentiation to senescence, cell cycle arrest and apoptosis (Taha et al., 2006), which is a reflec- tion of the discrepant results obtained with suramin in various cell types. The present results indicate that in the immature oligodendrocyte lineage OLN-93, suramin induces apoptotic cell death and also enhances the cyto- toxic effect of oxygen toxicity. Its mechanism of action in this experimental setting includes enhancement of pro-apoptotic proteins and impair- ment of growth factor dependent signaling path- ways. The fact that suramin targets a variety of intracellular molecular mechanisms, highly depen- dent on the experimental setting calls for caution to recommending this compound as a neuropro- tectant for white matter injury. References Bezvenyuk Z, T Suuronen, A Salminen and V Solovyan (2000) Protective effect of suramin against cell death in rat cerebel- lar granular neurons and mouse neuroblastoma cells. Neurosci. Lett. 292, 111-114. Chan CF and SY Lin-Shiau (2000) Site of action of suramin and reactive blue 2 in preventing neuronal death induced by dequalinium. J. Neurosci. Res. 62, 692-699. Collins MP, JM Lorenz, JR Jetton and N Paneth (2001) Hypocapnia and other ventilation-related risk factors for cerebral palsy in low birth weight infants. Pediatr. Res. 50, 712-719. Curtin JF and TG Cotter (2003) Live and let die: regulatory mechanisms in Fas-mediated apoptosis. Cell Signal 15, 983-992. Eichhorst ST, S Muerkoster, MA Weigand and PH Krammer (2001) The chemotherapeutic drug 5-fluouracil induces apoptosis in mouse thymocytes in vivo via activation of the CD95(Apo-1/Fas) system. Cancer Res. 61, 243-248. Eichhorst ST, A Krueger, S Muerkoster, SC Fas, A Golks, U Gruetzner, L Schubert, C Opelz, M Bilzer, AL Gerbes and PH Krammer (2004) Suramin inhibits death receptor- induced apoptosis in vitro and fulminant apoptotic liver damage in mice. Nat. Med. 10, 602-609. Felderhoff-Mueser U, DL Taylor, K Greenwood, M Kozma, D Stibenz, UC Joashi, AD Edwards and H Mehmet (2000) Fas/ CD95/APO-1 can function as a death receptor for neuronal cells in vitro and in vivo and is upregulated following cere- bral hypoxic-ischemic injury to the developing rat brain. Brain Pathol. 10, 17-29. Felderhoff-Mueser U, P Bittigau, M Sifringer, B Jarosz, E Korobowicz, L Mahler, T Piening, A Moysich, T Grune, F Thor, R Heumann, C Buhrer and C Ikonomidou (2004) Oxygen causes cell death in the developing brain. Neurobiol. Dis. 17, 273-282. Felderhoff-Mueser U, M Sifringer, O Polley, M Dzietko, B Leineweber, L Mahler, M Baier, P Bittigau, M Obladen, C Ikonomidou and C Buhrer (2005) Caspase-1-processed interleukins in hyperoxia-induced cell death in the develop- ing brain. Ann. Neurol. 57, 50-59. Fischer U, RU Janicke and K Schulze-Osthoff (2003) Many cuts to ruin: a comprehensive update of caspase substrates. Cell Death Differ. 10, 76-100. Gerstner B, A Gratopp, M Marcinkowski, M Sifringer, M Obladen and C Buhrer (2005) Glutaric acid and its metabo- lites cause apoptosis in immature oligodendrocytes: a novel mechanism of white matter degeneration in glutaryl-CoA dehydrogenase deficiency. Pediatr. Res. 57, 771-776. Gerstner B, C Buhrer, C Rheinlander, O Polley, A Schuller, M Berns, M Obladen and U Felderhoff-Mueser (2006) Maturation-dependent oligodendrocyte apoptosis caused by hyperoxia. J. Neurosci. Res. 84, 306-315. Gerstner B, M Sifringer, M Dzietko, A Schuller, J Lee, S Simons, M Obladen, JJ Volpe, PA Rosenberg and U Felderhoff-Mueser (2007) Estradiol attenuates hyperoxia- induced cell death in the developing white matter. Ann. Neurol. 61, 562-573. Gill JS and AJ Windebank (1998) Suramin induced ceramide accumulation leads to apoptotic cell death in dorsal root ganglion neurons. Cell Death Differ. 5, 876-883. Gill JS and AJ Windebank (2000) Ceramide initiates NFB- mediated caspase activation in neuronal apoptosis. Neurobiol. Dis. 7, 448-461. Gorospe M, X Wang, KZ Guyton and NJ Holbrook (1996) Protective role of p21(Waf1/Cip1) against prostaglandin A2-mediated apoptosis of human colorectal carcinoma cells. Mol. Cell. Biol. 16, 6654-6660. Goto T, S Takeuchi, K Miura, S Ohshima, K Mikami, K Yoneyama, M Sato, T Shibuya, D Watanabe, E Kataoka, D Segawa, A Endo, W Sato, R Yoshino and S Watanabe (2006) Suramin prevents fulminant hepatic failure resulting in reduction of lethality through the suppression of NF-B activity. Cytokine 33, 28-35. Gritli-Linde A, JV Ruch, MP Mark, S Lecolle and M Goldberg (1994) Effects of suramin, a polyanionic drug inducing lyso- somal storage disorders on tooth germs in vitro. Biol. Cell 81, 143-152. Hintz SR, DE Kendrick, BR Vohr, WK Poole and RD Higgins (2005) Changes HPK1-IN-2 in neurodevelopmental outcomes at 18 to 22 months’ corrected age among infants of less than 25 weeks’ gestational age born in 1993-1999. Pediatrics 115, 1645-1651.
Kaplan DR and FD Miller (2000) Neurotrophin signal trans- duction in the nervous system. Curr. Opin. Neurobiol. 10, 381-391.
Kathir KM, TK Kumar and C Yu (2006) Understanding the mechanism of the antimitogenic activity of suramin. Biochemistry 45, 899-906.
Kaur C, V Sivakumar, LS Ang and A Sundaresan (2006) Hypoxic damage to the periventricular white matter in neo- natal brain: role of vascular endothelial growth factor, nitric oxide and excitotoxicity. J. Neurochem. 98, 1200-1216.
Kharlamov A, SC Jones and DK Kim (2002) Suramin reduces infarct volume in a model of focal brain ischemia in rats. Exp. Brain Res. 147, 353-359.
Kruidering M and GI Evan (2000) Caspase-8 in apoptosis: the beginning of “the end”? IUBMB Life 50, 85-90.
La Rocca RV, CA Stein and CE Myers (1990) Suramin: proto- type of a new generation of antitumor compounds. Cancer Cells 2, 106-115.
Marlow N, D Wolke, MA Bracewell and M Samara (2005) Neurologic and developmental disability at six years of age after extremely preterm birth. N. Engl. J. Med. 352, 9-19.
Mitsuya H, M Popovic, R Yarchoan, S Matsushita, RC Gallo and S Broder (1984) Suramin protection of T cells in vitro against infectivity and cytopathic effect of HTLV-III. Science 226, 172-174.
Richter-Landsberg C and M Heinrich (1996) OLN-93: a new permanent oligodendroglia cell line derived from primary rat brain glial cultures. J. Neurosci. Res. 45, 161-173.
Rolland A and B Thylefors (1982) Basis for treatment and prevention of ocular onchocerciasis. Rev. Int. Trach. Pathol. Ocul. Trop. Subtrop. Sante Publique 1-156.
Scaffidi C, Fulda S, Srinivasan A, Friesen C, Li F, Tomaselli KJ, Debatin KM, Krammer PH, Peter ME (1998) Two CD95 (APO-1/Fas) signaling pathways. EMBO J. 17, 1675-1687.
Sirinyan M, F Sennlaub, A Dorfman, P Sapieha, F Gobeil Jr, P Hardy, P Lachapelle and S Chemtob (2006) Hyperoxic expo- sure leads to nitrative stress and ensuing microvascular degeneration and diminished brain mass and function in the immature subject. Stroke 37, 2807-2815.
Suda T and S Nagata (1994) Purification and characterization of the Fas-ligand that induces apoptosis. J. Exp. Med. 179, 873-879.
Sullivan KA, B Kim, M Buzdon and EL Feldman (1997) Suramin disrupts insulin-like growth factor-II (IGF-II) medi- ated autocrine growth in human SH-SY5Y neuroblastoma cells. Brain Res. 744, 199-206.
Taha TA, TD Mullen and LM Obeid (2006) A house divided: ceramide, sphingosine, and sphingosine-1-phosphate in pro- grammed cell death. Biochim. Biophys. Acta 1758, 2027-2036.
Voogd TE, EL Vansterkenburg, J Wilting and LH Janssen (1993) Recent research on the biological activity of suramin. Pharmacol. Rev. 45, 177-203.
Wang X, CI Rousset, H Hagberg and C Mallard (2006) Lipopolysaccharide-induced inflammation and perinatal brain injury. Semin. Fetal Neonatal Med. 11, 343-353.
Zhuang S and RG Schnellmann (2005) Suramin promotes pro- liferation and scattering of renal epithelial cells. J. Pharmacol. Exp. Ther. 314, 383-390.