Experimental Neurology 265 (2015) 142–151

Contents lists available at ScienceDirect

Experimental Neurology journal homepage: www.elsevier.com/locate/yexnr

Regular Article

Fullerenols and glucosamine fullerenes reduce infarct volume and cerebral inflammation after ischemic stroke in normotensive and hypertensive rats☆ Felix Fluri a,b,⁎,1, Dan Grünstein c,d, Ertugrul Cam a, Udo Ungethuem e, Florian Hatz b, Juliane Schäfer f, Samuel Samnick g, Ina Israel g, Christoph Kleinschnitz h, Guillermo Orts-Gil c,d, Holger Moch i, Thomas Zeis j, Nicole Schaeren-Wiemers j, Peter Seeberger c,d a

Department of Neurology, University Hospital Zürich, 8091 Zürich, Switzerland Department of Neurology, University Hospital Basel, 4031 Basel, Switzerland c Department of Biomolecular Systems, Max Planck Institute of Colloids and Interfaces, 14476 Potsdam, Germany d Institute for Chemistry and Biology, Freie Universität Berlin, 14195 Berlin, Germany e Department of Surgery, Swiss Hepato-Pancreatico-Biliary Center, University Hospital Zürich, 8091 Zürich, Switzerland f Basel Institute for Clinical Epidemiology and Biostatistics, University Hospital Basel, 4031 Basel, Switzerland g Department of Nuclear Medicine, Interdisciplinary PET Center, University Hospital Würzburg, 97080 Würzburg, Germany h Department of Neurology, University Hospital Würzburg, 97080 Würzburg, Germany i Institute for Surgical Pathology, Department of Pathology and Laboratory Medicine, University Hospital Zürich, 8091 Zürich, Switzerland j Neurobiology Laboratory, Department of Biomedicine, University Hospital Basel, University of Basel, 4031 Basel, Switzerland b

a r t i c l e

i n f o

Article history: Received 18 December 2014 Accepted 15 January 2015 Available online 24 January 2015 Keywords: Ischemic stroke Animal experiments Neuroprotective agents Fullerene Inflammation

a b s t r a c t Cerebral inflammation plays a crucial role in the pathophysiology of ischemic stroke and is involved in all stages of the ischemic cascade. Fullerene derivatives, such as fullerenol (OH-F) are radical scavengers acting as neuroprotective agents while glucosamine (GlcN) attenuates cerebral inflammation after stroke. We created novel glucosamine–fullerene conjugates (GlcN-F) to combine their protective effects and compared them to OH-F regarding stroke-induced cerebral inflammation and cellular damage. Fullerene derivatives or vehicle was administered intravenously in normotensive Wistar-Kyoto (WKY) rats and spontaneously hypertensive rats (SHR) immediately after transient middle cerebral artery occlusion (tMCAO). Infarct size was determined at day 5 and neurological outcome at days 1 and 5 after tMCAO. CD68- and NeuN-staining were performed to determine immunoreactivity and neuronal survival respectively. Cytokine and toll like receptor 4 (TLR-4) expression was assessed using quantitative real-time PCR. Magnetic resonance imaging revealed a significant reduction of infarct volume in both, WKY and SHR that were treated with fullerene derivatives. Treated rats showed an amelioration of neurological symptoms as both OH-F and GlcN-F prevented neuronal loss in the perilesional area. Cerebral immunoreactivity was reduced in treated WKY and SHR. Expression of IL-1β and TLR-4 was attenuated in OH-F-treated WKY rats. In conclusion, OH-F and GlcN-F lead to a reduction of cellular damage and inflammation after stroke, rendering these compounds attractive therapeutics for stroke. © 2015 Elsevier Inc. All rights reserved.

Introduction

Abbreviations: OH-F, fullerenol; GlcN, glucosamine; GlcN-F, glucosamine fullerene; WKY, Wistar-Kyoto-rats; SHR, spontaneously hypertensive rats; tMCAO, transient middle cerebral artery occlusion; TLR-4, toll-like receptor 4; IL-1 β, interleukin 1 β; TNF-α, tumor necrosis factor-α; NF-κB, nuclear factor kappaB; STAIR, Stroke Therapy Academic Industry Roundtable; MRI, magnetic resonance imaging; qRT PCR, quantitative real-time PCR. ☆ The institution in which the work was performed: Department of Neurology, University Hospital Zürich, Switzerland. ⁎ Corresponding author at: Department of Neurology, University Clinic of Wuerzburg, Josef-Schneider Strasse 11, 97080 Wuerzburg, Germany. Fax: +49 931 201 23255. E-mail address: felix.fl[email protected] (F. Fluri). 1 Present address: Department of Neurology, University Hospital Würzburg, 97080 Wuerzburg, Germany.

http://dx.doi.org/10.1016/j.expneurol.2015.01.005 0014-4886/© 2015 Elsevier Inc. All rights reserved.

Inflammation has been recognized as a key contributor to the pathophysiology of ischemic stroke (Moskowitz et al., 2010). There is growing evidence that inflammatory processes are involved in all stages of the ischemic cascade, from the early intravascular events after arterial occlusion to the late regenerative alterations leading to brain damage and tissue repair (Iadecola and Anrather, 2011). Immediately after interruption of blood supply, reactive oxygen species (ROS) trigger the coagulation cascade and lead to the activation of complement (i.e., complement C3), platelet and endothelial cells (Eltzschig and Carmeliet, 2011; Peerschke et al., 2010; Song et al., 2006). Proinflammatory mediators such as cytokines and chemokines are rapidly

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

generated (Iadecola and Anrather, 2011). As the ischemic cascade progresses, cell death leads to a new phase of inflammatory response (Iadecola and Anrather, 2011). Ischemic cell death leads to activation of toll-like receptor 4 (TLR-4) that in turn activates pathways linked to the transcription of many pro-inflammatory gene encoding cytokines such as interleukin-1β (IL-1β) and tumor necrosis factor-α (TNF-α). Hence, modulation of inflammatory processes after stroke might exert a protective effect on ischemic brain tissue and therefore might be a promising new stroke therapy. C60 fullerenes, comprised of 60 carbon atoms configured as a hollow sphere, are candidate therapeutic agents because they exert multiple effects on the ischemic cascade (Chiang et al., 1995; Jin et al., 2000) and therefore, fullerene-derivatives might influence cerebral ischemia (Lin et al., 2002). When appropriately modified, fullerenes can pass the blood–brain barrier (Yamago et al., 1995). Fullerenols are hydroxylated fullerenes that are neuroprotective because they scavenge free radicals (Fig. S1 in the online-only data supplement) (Chiang et al., 1995). Furthermore, fullerenols inhibit glutamate channels resulting in a reduction in glutamate-induced intracellular calcium (Jin et al., 2000), preventing cell death. Another neuroprotective agent is glucosamine (GlcN) which reduces the immune response after ischemic stroke by inhibiting nuclear factor kappaB (NF-κB) (Hwang et al., 2010). Since the Stroke Therapy Academic Industry Roundtable (STAIR) committee recommends investigating neuroprotective drugs in animals with a cerebrovascular risk factor (Fisher et al., 2009) spontaneously hypertensive rats (SHR) were considered as an appropriate rat strain to investigate the effect of fullerene derivatives; additionally, this strain allows for comparing drug effects to “healthy” i.e., normotensive Wistar-Kyoto (WKY) rats (Liu et al., 2009). Here we show that intravenously administered polyhydroxylated fullerene reduces infarct size and inhibits cerebral inflammation in normotensive and hypertensive rats after ischemic stroke. In addition, glucosamine–fullerene conjugates potentiate the neuroprotective effects of fullerene. Methods Animals A total number of 43 male SHR and 24 male WKY rats (Charles River, Sulzfeld, Germany), weighing 250 to 300 g, were used throughout this study. Animal experimentation was approved by the Veterinäramt Zürich (approval number 148/2009) and is adherent to the NIH Guide for the Care and Use of Laboratory Animals. Weight of each animal was measured before surgery and decapitation (Table I in the onlineonly data supplement). Rats were anesthetized with 2.5% isoflurane in 70% N20/30% O2 during the surgery and were monitored for physiological parameters that could affect stroke outcome (Tables II and III in the online-only data supplement). Transient middle cerebral artery occlusion (tMCAO) Temporary focal cerebral ischemia was induced by a 60 min occlusion of the left middle cerebral artery according to the method of Longa et al. (1989). Thereafter, the animals were allowed to recover from anesthesia. Rats were euthanized 5 days after tMCAO for histological analyses. Animals without infarction, detected 24 h after tMCAO, were excluded from the study. We performed surgery and evaluation of all readout parameters while being blinded to the experimental groups. Application of OH-F and GlcN-F in animals OH-F 0.5 mg/kg was administered intravenously in WKY (n = 10) and SHR (n = 10) immediately after reperfusion and compared with controls of each strain. To achieve a greater reduction of

143

infarction in SHR, 1.0 mg/kg (n = 7) and 2.0 mg/kg (n = 7) were administered in animals of this strain. GlcN-F was injected in SHR at doses of 0.5 mg/kg and 5.0 mg/kg. Neurological score Neurological evaluation was performed for all animals one day and five days after tMCAO induction. Each animal was assigned only one score, indicating the worst symptom even if the animal showed others, i.e., less severe symptoms. All animals (untreated WKY rats: n = 10; WKY rats treated with 0.5 mg/kg OH-F: n = 10; untreated SHR: n = 10; SHR treated with 0.5 mg/kg OH-F: n = 10; SHR treated with 5.0 mg/kg GlcN-F: n = 10) were scored as follows (Gerriets et al., 2004): 0 = no neurologic deficits; 1 = contralateral forelimb flexion; 2 = inconstant contralateral circling or contralateral circling after tail pull; 3 = spontaneously contralateral circling; 4 = falling to the right; and 5 = no spontaneous walking with depressed level of consciousness. Cerebral MRI and measurement of infarct volume Five days after tMCAO, MRI was conducted with a 4.7 T/16 cm Bruker Pharma Scan tomograph (Bruker Bio Spin AG, Fällanden/ Switzerland). T2-weighted (T2w) spin-echo imaging was used to map lesion and hemispheric volumes. Lesion size was determined on T2w using ImageJ Analysis Software 1.45 s (National Institutes of Health, USA; http://rsb.info.nih.gov/ij/). Immunohistochemistry Coronal cryosections (8 μm thick) were cut at 400 μm intervals using a cryostat (Hyrax C60, Zeiss, Switzerland) and mounted on Superfrost Plus slides (Menzel, Braunschweig/Germany). Sections were stained with antibodies directed against NeuN (1:500, Millipore, Zug/ Switzerland), CD68 (1:250, Serotec Ltd, Düsseldorf/Germany) and MAP-2 (1:2000 Abcam, ab32454, Abcam plc,Cabrindge/UK). CD68+ cell density was measured on ten fields of view in three consecutive sections of the perilesional zone of the caudate putamen using an Olympus microscope (Olympus, BX61, Volketswil/Switzerland). NeuN+ cell count was quantified on five fields of view in three consecutive sections. Quantitative real-time PCR Quantitative real-time PCR (qRT PCR) was performed on an Applied Biosystems Fast 7500 system (Applied Biosystems, Life Technologies, Zug/Switzerland). Using TaqMan gene expression assays, qRT-PCR of interleukin 1β (IL-1β), tumor necrosis factor α (TNF-α), complement C3 and toll-like receptor 4 (TLR-4) was performed according to the manufacturer's protocol (Applied Biosystems, Life Technologies, Zug/ Switzerland). Statistical analyses We used multiple linear regression analysis to estimate associations between infarct volume and both treatment with 0.5 mg/kg OH-F (yes/ no) and strain (SHR/WKY) as independent variables. A key question for this study was whether treatment with 0.5 mg/kg OH-F confers a smaller reduction in infarct volume among SHR compared to WKY rats. To address this question, we added to our model a covariate for the interaction between treatment with 0.5 mg/kg OH-F and strain. The estimate for this interaction shows how the association between infarct volume and treatment changes at higher blood pressure values. Prior to model fitting, we logarithm-transformed the infarct volume to achieve a multiplicative model for infarct volume when transformed back to the original units. Note that the geometric mean is found as the anti-log of

144

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

the mean of logarithm-transformed values. For our analysis, we report estimated geometric mean ratios with 95% confidence intervals. The effect on infarct volume with the administration of higher doses of OH-F (1.0 and 2.0 mg/kg, respectively) and of GlcN-F, as well as comparisons between treatment groups in regard to neurological deficits, anti-inflammatory properties, immunoreactivity, neuronal survival and cell proliferation were addressed with descriptive and exploratory statistics. For immunostaining, we calculated the mean of all cell counts/ area within one animal to obtain a single summary measure per animal. P-values were calculated using the U-test. A level of P b 0.05 was considered significant. We used R version 3.0.1 (R Foundation for Statistical Computing, Vienna, Austria) and the R add-on package lattice version 0.20-21 (Sarkar, 2008) for our analyses and graphics, respectively. Results Infarct volume is reduced after OH-F and GlcN-F treatments in WKY rats and SHR Initial studies on neuroprotective agents in experimental stroke should demonstrate positive effects in “healthy” animals (Fisher et al., 2009). Therefore, we assessed the effect of fullerenol (OH-F) [C60(OH)34 -36] on infarct volume in WKY rats without additional

comorbidities, such as arterial hypertension. A total of 20 WKY rats underwent transient middle cerebral artery occlusion (tMCAO) for 60 min. Thereafter, 0.5 mg/kg OH-F in 2.5 mL NaCl (0.9% in water) was administered to 10 WKY rats within 5 min after recirculation. Another 10 WKY rats received 2.5 mL NaCl (0.9% in water) within 5 min of recirculation after tMCAO, and served as a control group. Infarct volume was quantified non-invasively with T2w MRI five days after tMCAO. In control WKY rats, infarction was detected in the left caudate putamen and partly in the cortex, whereas in six of ten OH-F-treated WKY rats, infarction was restricted to the left caudate putamen (Fig. 1). The median infarct volume of control and OH-F-treated WKY rats was 104 mm3 (interquartile range [IQR] 95, 140) and 36 mm3 (IQR 26, 42) respectively (Figs. 1 and 2A). Unlike this animal model, stroke patients suffer from different comorbidities, of which arterial hypertension is the most important. Therefore, we performed the same experiment in hypertensive rats, i.e., SHR, treating the animals either with 0.5 mg/kg OH-F (n = 10) or 0.9% NaCl (n = 10, one died during the tMCAO) within 5 min of recirculation after tMCAO. Weight-matched SHR were chosen because they are selectively bred from the WKY strain. In both the control and OH-F-treated SHR, infarction was observed in the left caudate putamen and the cortex using MRI (T2w) five days after tMCAO. The median infarct volume of the control and OH-F-treated SHR was 267 mm3 (IQR 266, 284) and 213 mm3 (IQR 180, 220) respectively (Figs. 1 and 2B).

Fig. 1. Infarct volume in normotensive WKY and SHR was measured using T2 weighted (T2w) MRI five days after tMCAO. (A) Representative coronal slice of a rat brain; cortex and caudate putamen are marked as blue and green areas respectively. (B) T2w images of an infarction (boundary marked by red tracks) in control WKY rats (1) and WKY rats treated with 0.5 mg/kg OH-F (2), control SHR (3), SHR treated with either 0.5 mg/kg OH-F (4) or 5.0 mg/kg GlcN-F (5). Infarction was detected in the left caudate putamen and partly in the cortex of all untreated and treated SHR and untreated WKY rats, whereas in six of ten OH-F-treated WKY rats infarction was restricted to the left caudate putamen only. Infarct area was traced manually on 17 consecutive coronal brain sections (1 mm thick) using image analysis software (Image J 64, National Institutes of Health, USA) and infarct volume was calculated by summing up the areas from all slices.

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

145

Fig. 2. Infarct volume in WKY and SHR treated with either OH-F or GlcN-F and compared to untreated animals. (A) Box plots of the distribution of infarct volume (mm3) among control WKY rats (n = 10) and WKY rats treated with 0.5 mg/kg OH-F (n = 10), (B) box plots of the distribution of infarct volume among control SHR (n = 9), SHR treated with 0.5 mg/kg OH-F (n = 10), 1.0 mg/kg OH-F (n = 5) and 2.0 mg/kg OH-F (n = 4) as well as SHR treated with GlcN-F at doses of 0.5 mg/kg (n = 7) and 5.0 mg/kg (n = 8) respectively. The black dot indicates the median infarct volume of one group.

As shown by multivariable analysis, SHR had an increased infarct volume compared with WKY rats (geometric mean ratio of 2.46; 95% confidence interval [CI] 1.81, 3.34; P b 0.001). Treatment with 0.5 mg/kg OH-F had a smaller effect on infarct volume reduction in SHR compared with WKY rats (geometric mean ratio for interaction 2.33; 95% CI 1.54, 3.53; P b 0.001). Relative to their untreated controls, treatment with 0.5 mg/kg OH-F was associated with a significant reduction in infarct volume for both WKY rats (geometric mean ratio of 0.32; 95% CI 0.23, 0.42; P b 0.001) and SHR (geometric mean ratio of 0.74; 95% CI 0.55, 0.98; P = 0.038). Treatment with 0.5 mg/kg OH-F led to a reduction in infarct volume of 68% (95% CI 58%, 77%; P b 0.001) in WKY rats and of 26% (95% CI 2%, 45%; P = 0.038) in SHR. Importantly, untreated SHR had a 146% increase in infarct volume (95% CI 81%, 234%; P b 0.001) compared to untreated WKY rats, supporting the notion that increased blood pressure has a negative impact on brain damage after ischemic stroke (McCabe et al., 2009). To investigate whether a further reduction of infarct volume in the SHR model could be achieved by increasing the OH-F dose, two additional doses of OH-F, namely 1.0 mg/kg (n = 7) and 2.0 mg/kg (n = 7), were intravenously administered within 5 min after reperfusion and compared with the effect of 0.5 mg/kg OH-F on infarct volume in SHR. Five days after tMCAO, median infarct volume in SHR treated with 1.0 mg/kg (n = 5) and with 2.0 mg/kg (n = 5; exclusion of one animal without detectable infarct) was 214 mm3 (IQR 163, 227) and 200 mm3 (IQR 173, 217) respectively (Fig. 2B). In summary, higher OH-F doses had no impact on infarct volume. However, adverse reactions such as nasal and periocular overproduction of porphyrine and writhing and stretching of the trunk were more severe in the SHR treated with either 1.0 or 2.0 mg/kg OH-F than in animals receiving the lower dose. Additionally, two SHR treated with 1.0 mg/kg and two treated with 2.0 mg/kg died within 48 h, whereas all ten SHR treated with 0.5 mg/kg survived. Based on these observations, all further experiments were conducted with 0.5 mg/kg OH-F. Treatment with OH-F, known as a strong radical scavenger (Chiang et al., 1995), had a limited effect on reducing the infarct volume in SHR, suggesting that cerebral inflammation after ischemic stroke may be crucial for infarct evolution. Anti-inflammatory properties have, so far, not been associated with OH-F, therefore we conjugated fullerene to glucosamine (GlcN), which exerts anti-inflammatory effects (Hwang et al., 2010). Glucosamine–fullerene (GlcN-F) [C60(GlcN)12]

was intravenously administered at a dose of 0.5 mg/kg in SHR (n = 7) within 5 min after a tMCAO of 60 min. However, 0.5 mg/kg GlcN contains only about 10% of fullerene compared with the same dose of OH-F and we hypothesized that fullerene may contribute to the overall neuroprotective effect in combination with GlcN. Therefore, a second series of experiments was conducted in SHR (n = 8) with 5.0 mg/kg GlcN-F, which contains an equivalent amount of fullerene as 0.5 mg/kg, enabling a better comparison of the two molecules. In all SHR treated with either 0.5 mg/kg or 5.0 mg/kg GlcN-F, infarction was observed both in the left caudate putamen and partly in the cortex (Fig. 1). Median infarct volume in SHR treated with 0.5 and 5.0 mg/kg GlcN-F was 211 mm3 (IQR 207, 232) and 172 mm3 (IQR 160, 191) respectively (Fig. 2B). Thus, GlcN-F-treatment resulted in greater reduction of infarct volume than OH-F-treatment at the two different doses we tested. GlcN-F is detectable in the ischemic brain area The ability of neuroprotective agents to cross the blood–brain barrier (BBB), i.e., to reach the ischemic area where they should act is crucial for successful neuroprotection. Passing the BBB depends on hydrophilic and lipophilic properties of a molecule. NXY-059 for example did not pass the BBB easily probably because of its high solubility in water (Kuroda et al., 1999) which might have contributed to the lack of neuroprotection in the clinical trials. Therefore, we aimed to detect GlcN-F in the ischemic area as early as 2 h after induction of cerebral ischemia. One hour after tMCAO, GlcN-F was administered intravenously and 1 h later, brains were harvested and immediately frozen. Coronal sections (10 μm thick) were cut and exposed against a phosphor image plate (Biostep, Jahnsdorf, Germany) overnight. We found an accumulation of the 68Ga-labeled GlcN-F (68Ga-GlcN-F) in the ischemic area of the brain, indicating that the compound has passed the BBB (Fig. 3). Neurological symptoms improve after treatment with OH-F and GlcN-F Clinical trials judge drug efficacy by using neurological and/or functional outcomes rather than infarct volume. Therefore, we assessed therapeutic efficacy by scoring the WKY rats treated with 0.5 mg/kg OH-F, the corresponding control WKY rats, the SHR treated with either

146

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

Fig. 3. Visualization of 68Ga-labeled GlcN-F accumulation in the ischemic brain area 2 h after induction of tMCAO (60 min) by ex-vivo autoradiography, confirming the ability of GlcN-F to cross the blood–brain barrier and to reach the ischemic area for neuroprotection. (A) Representative MAP-2 staining of a coronal brain section. The infarcted area in the caudate putamen remains unstained; (B) the same brain section as depicted in A, merged with autoradiography of this brain area. (C) Scaled up caudate putamen, encompassing the infarcted region (unstained area); (D) the corresponding autoradiographic image of the caudate putamen impressively demonstrates accumulation of the 68Ga-labeled GlcN-F in the infracted area. (E) Fused autoradiography/caudate putamen images.

0.5 mg/kg OH-F or 5.0 mg/kg GlcN-F, and the control SHR on days one and five after recirculation (Fig. 4). A six-point scoring system (modified according to Gerriets) was used (Gerriets et al., 2004). All rats received a single score indicating the worst symptom. For all untreated control WKY rats, infarction after tMCAO led to impairment of motor function with a range of severity. Weak impairments were contralateral forelimb flexion (score 1), and contralateral circling after tail pull (score 2). More severe symptoms were spontaneously contralateral circling (score 3) and falling to the right side (score 4). These behaviors were observed in nearly all untreated WKY rats, and in some cases, animals were no longer able to walk spontaneously (score 5). Overall, WKY rats treated with 0.5 mg/kg OH-F developed less severe neurological deficits (score 0 and 1). After one day, only two of the treated WKY rats exhibited impairment of motor function (score: 1), and this was resolved after five days (score 0; Fig. 4A). Initially, tMCAO in untreated control SHR had a similar outcome to untreated WKY rats; all animals had a clinical score between 1 and 4 after the first day. However, after five days spontaneous amelioration of the neurological deficits was observed. Previously, rodents have been reported to spontaneously recover from motor dysfunction (Reglodi et al., 2003), probably due to recruitment of the undamaged hemisphere (Biernaskie et al., 2005). Treatment with 0.5 mg/kg OH-F and with 5.0 mg/kg GlcN-F revealed a general improvement of motor function in SHR with only residual motor deficits after five days (Fig. 4B). Therefore, in this animal model, OH-F and GlcN-F-treatments

lead to a positive outcome after ischemic stroke, indicated by reduced lesion volume and amelioration of motor deficits.

OH-F and GlcN-F ameliorate post-stroke cerebral inflammation The cellular immune response in the infarcted area is characterized by activated microglia and infiltrating macrophages. Both cell types present CD68 and therefore can be visualized by anti-CD68immunostaining. To investigate the anti-inflammatory properties of OH-F and GlcN-F, we measured the degree of CD68-immunoreactivity in the affected caudate putamen of untreated and treated SHR and WKY rats, using fresh frozen acetone-fixed coronal brain sections (8 μm thick). Since infarction was confined to the caudate putamen in all WKY rats, we did not assess CD68-immunoreactivity in the cortex. CD68-immunoreactivity was expressed as the ratio of the area of CD68-immunostained cells to the non-stained area per field of view. In OH-F-treated WKY rats, CD68-immunoreactivity was reduced by 44% compared to control animals (P = 0.030; Figs. 5A and C). Similarly, SHR treated with 0.5 mg/kg OH-F displayed a reduction in CD68immunoreactivity of 24% compared with control animals (P = 0.15; Figs. 5B and D). GlcN-F treatment of SHR, however, led to a strong and significant (40%) reduction in CD68-immunoreactivity (P = 0.008; Figs. 5B and D), confirming that this molecule is a more effective antiinflammatory agent.

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

147

Fig. 4. Neurological scores of WKY rats and SHR, treated with either OH-F or GlcN-F compared to controls at day 1 (A) and day 5 (B) after tMCAO. A 6-point score was used (modified according to Gerriets): (Gerriets et al., 2004) 0 = no neurological deficits, 1 = contralateral forelimb flexion, 2 = contralateral circling after tail pull, 3 = spontaneously contralateral circling, 4 = falling to the right; 5 = no spontaneous walking. % gradient indicates the % of the experimental group population graded with a score.

Effects of OH-F and GlcN-F on the expression of inflammatory molecules following cerebral ischemia were measured using qRT PCR to track changes in the expression of IL-1β, TNF-α, complement C3 and TLR-4. In WKY rats, treatment with OH-F resulted in a 49.2% reduction in expression of IL-1β (P = 0.032; Fig. 6A, Table IV in the

online-only data supplement), and to a 31.8% decrease in expression of TLR-4 (P = 0.035; Fig. 6B, Table IV in the online-only data supplement) as well as C3-expression (54.8%, P = 0.115; Fig. 6C, Table VI in the online-only data supplement), both relative to controls. TNF-αexpression was unaffected (Fig. 6D, Table IV in the online-only data

148

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

Fig. 5. CD68-immunoreactivity of the caudate putamen. (A) CD68-immunoreactivity in WKY rats and (B) in SHR. Control animals are compared with those treated with either 0.5 mg/kg OH-F (WKY rats and SHR) or 5.0 mg/kg GlcN-F (SHR). CD68-immunoreactivity was expressed as the ratio of the area of CD68-immunostained cells to the non-stained area per field of view. Ten regions of interest were measured in three consecutive brain sections. Coronal brain sections (8 μm thick) of WKY rats (C) and SHR (D) treated with either 0.5 mg/kg OH-F or 5.0 mg/kg GlcN-F (magnification × 100). CD68 immunoreactivity was defined as the ratio of the area of CD68-immunostained cells to the non-stained area per field of view (FOV) in three consecutive brain sections. Length of the scale bar: 200 μm.

supplement). In SHR, treatment with either 0.5 mg/kg OH-F or 5.0 mg/kg GlcN-F did not affect the expression of IL-1β, TNF-α, complement C3 or TLR-4 (Figs. 6A–D). OH-F-and GlcN-F-treatments result in enhanced neuronal survival To determine whether OH-F and GlcN-F protect neurons from death after ischemic stroke, the neuronal marker neuron-specific nuclear protein (NeuN) was immunostained in samples of brain tissue. Specifically, three consecutive coronal brain tissue sections localized at the stereotaxic level of the Bregma (i.e., where the caudate putamen is also localized) were analyzed; in each of these three sections, five regions of interest were selected. In OH-F-treated WKY rats, 47.6% more NeuN+ cells were detected in the subcortical perilesional area (i.e., penumbra) compared with control WKY rats (P = 0.002; Figs. 7A and C). In SHR, GlcN-F-treatment resulted in 46.4% more NeuN+ cells in the subcortical perilesional area than in control SHR (P = 0.008; Figs. 7B and D). In control and OH-F treated SHR, no difference concerning NeuN+ cells was observed (6.7%; P = 0.67; Figs. 7B and D). Discussion In the present study, the effects of hydroxylated and glucosamineconjugated fullerenes on cerebral infarction and on post-ischemic cerebral inflammation have been investigated for the first time in normo- and hypertensive rats. Intravenously administered OH-F

significantly reduced infarct volume and attenuated cerebral inflammation by reducing the expression of IL-1β and TLR-4 in WKY rats compared to controls. This anti-inflammatory effect of OH-F is a novel finding and – together with its reported scavenging properties (Chiang et al., 1995) – might contribute to the reduction of infarct volume. In contrast, OH-F-treatment (0.5 mg/kg) in SHR reduced the infarct volume less than in WKY rats even after increasing the dose of OH-F to 1.0 or 2.0 mg/kg. This finding is in line with studies where the efficacy in the context of comorbidity was also substantially lower (Howells et al., 2010). The different responses of SHR and WKY rats to OH-F may be the result of a more pronounced microglial response to focal ischemia in hypertensive rats when compared to WKY rats (Marks et al., 2001). Since GlcN suppresses microglia activation and macrophage accumulation (Hwang et al., 2010) in the post-ischemic brain, we synthesized a novel fullerene–glucosamine conjugate to potentiate the effects of each component. In fact, GlcN-F (5.0 mg/kg) reduced the infarct volume in SHR significantly when compared to control SHR. However, GlcN-F-treatment reduced the infarct volume only slightly more than OH-F-treatment in SHR. Anatomical and morphological differences between the cerebral vasculature of these two strains may play key roles in the evolution of ischemic infarction (Jiménez-Altayó et al., 2007). Furthermore, SHR reveal a smaller penumbra when compared to normotensive rats (McCabe et al., 2009) and thus, there is less salvageable ischemic tissue in SHR than in WKY rats. Nevertheless, both OH-F and GlcN-F have a strong neuroprotective effect in SHR when compared to the efficacy of the well-studied

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

149

Fig. 6. Pro-inflammatory marker expression in control WKY rats and SHR and in animals treated with either 0.5 mg/kg OH-F or 5.0 mg/kg GlcN-F, five days after tMCAO. Changes in RNA expression levels were determined by quantitative real time (qRT) PCR. 0.0 means no expression of the pro-inflammatory marker compared to relative standard curves; bars (i.e., mean log fold change) in the negative and positive scale indicate decreased and increased expressions of pro-inflammatory markers compared to relative standard curves (A) IL-1β: Interleukin-1β; (B) TLR-4: Toll-like receptor 4; (C) C3: Complement C3; (D) TNF-α: Tumor necrosis factor-α.

neuroprotective drug phenylbutynitrone (NXY-059) (Macleod et al., 2008; Zhao et al., 2001). NXY-059 significantly reduced infarct volume (48%) in normotensive rats but failed to reduce infarct volume significantly in SHR (Macleod et al., 2008; Zhao et al., 2001). This discrepancy in the preclinical investigation may have been an early indicator of what was found in the clinic: NXY-059 failed in the SAINT II trial (Shuaib et al., 2007). Clinical examination revealed an early amelioration of neurological symptoms in OH-F- and GlcN-F-treated animals compared to controls. Notably, the symptoms improved not only in normotensive but also in hypertensive animals that were treated. This observation underscores the strong therapeutic effect of OH-F and GlcN-F as potential treatments for ischemic stroke and confirms that functional improvement is not necessarily associated with infarct size in animals (Corbett and Nurse, 1998; Kawamata et al., 1996). Inflammatory processes are involved in the early post-ischemic period and enhance secondary tissue damage (Iadecola and Anrather, 2011). We observed that OH-F exerted anti-inflammatory effects in WKY rats as demonstrated by a reduction in CD68-immunoreactivity, either due to activation of fewer microglia and/or less monocyte infiltration from the periphery. While OH-F exhibited no anti-inflammatory effect following cerebral ischemia in SHR, GlcN-F significantly reduced the number of CD68-immunoreactive cells in SHR. Since hypertensive rats typically have a greater degree of cerebral inflammation after cerebral ischemia than normotensive animals (Marks et al., 2001), we conclude that OH-F is a less-effective anti-inflammatory agent than

GlcN-F. The GlcN-moiety is likely the most effective anti-inflammatory component of GlcN-F which supports the recent report that GlcN administration suppressed microglial activation and macrophage accumulation in cerebral ischemia of rats (Hwang et al., 2010). Both, OH-F and GlcN-F suppressed activation of microglia and/or infiltration of macrophages, but only OH-F reduced inflammatory molecules in WKY rats. The reduction of IL-1β, TLR-4, and complement C3 expression in WKY rats after OH-F treatment is likely a result of reduced microglia activation in these animals. Both TLR-4 (Jack et al., 2005) and IL-1β (Allan et al., 2005) are expressed in microglia. A recent study demonstrated, that TLR-4 is involved in cerebral inflammation and brain damage after stroke (Caso et al., 2007); TLR-4-deficient mice had minor infarction and a less inflammatory response after stroke (Caso et al., 2007). In stroke patients, serum levels of TLR-4 were associated to clinical outcome and proposed as a therapeutic target (Brea et al., 2011). Complement C3, which is also synthesized by microglia (Haga et al., 1993), is a critical effector of complementmediated ischemic neurotoxicity (Mocco et al., 2006). C3-knockout mice showed reduced infiltrating granulocytes and decreased oxidant stress levels (Mocco et al., 2006). Altogether, OH-F might affect different aspects of cerebral inflammation. Reduced cerebral inflammation after OH-F and GlcN-F administration may also effectively preserve neurons in the perilesional zone of the caudate putamen. This observation corroborates a report showing that functionalized fullerenes prevent neuronal death (Dugan et al., 1996). Furthermore, our work indicates that OH-F and GlcN-F are not

150

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151

Fig. 7. NeuN+ cells (red stained) of the perilesional area of the caudate putamen five days after tMCAO. (A) NeuN+ cells in WKY rats and (B) in SHR; untreated and treated with either OH-F (WKY and SHR) or GlcN-F (SHR). Results are expressed as mean of NeuN+ cells per field of view (FOV). NeuN+ cells in (C) WKY and (D) SHR, untreated control or treated with either OH-F or GlcN-F (magnification ×200). Length of the scale bar: 200 μm. A total of five regions of interest in three consecutive brain sections were determined.

neurotoxic, even during ischemia when cells are vulnerable. Highly soluble derivatives of fullerene such as C60(OH)24 are reported to be less toxic as its pristine form (Sayes et al., 2004). In conclusion, we demonstrate the anti-inflammatory effects of functionalized fullerenes administered following cerebral ischemia. Inflammation constitutes an attractive target for therapeutic intervention as it plays a key role in the pathophysiology of cerebral ischemia by exerting deleterious effects on the progression of tissue damage (del Zoppo, 2009). However, functionalized fullerenes have been shown to interfere in other processes of the ischemic cascade such as oxidative stress and glutamate excitotoxicity rendering functionalized fullerenes, with their multimodal properties, a promising candidate for stroke treatment. It remains to be investigated whether simultaneous administration of OH-F and GlcN-F leads to a stronger infarct reduction in SHR than each compound alone. Efforts to translate these findings to humans will commence following further preclinical evaluations.

Source of funding This research project was funded by grants from the Novartis foundation (F.F.), the Käthe-Zingg-Schwichtenberg-Fonds (KZS04/09) (Schweizerische Akademie der Medizinischen Wissenschaften) (F.F.), the Gottfried und Julia Bangerter-Rhyner-Foundation (F.F., J.S.), the Freiwillige Akademische Gesellschaft Basel (F.F.), the Max-Planck Society, the Körber-Foundation, and the European Union FP7 (CARMUSYS) (P.H.S), the Federal Institute for Materials Research and Testing (G.O.-G.) and the Santésuisse (J.S.)

Disclosures None. Acknowledgments We thank Dr. V. Mountain for critically editing this manuscript. Appendix A. Supplementary data Supplementary data to this article can be found online at http://dx. doi.org/10.1016/j.expneurol.2015.01.005. References Allan, S.M., Tyrrell, P.J., Rothwell, N.J., 2005. Interleukin-1 and neuronal injury. Nat. Rev. Immunol. 5, 629–640. Biernaskie, J., Szymanska, A., Windle, V., Corbett, D., 2005. Bi-hemispheric contribution to functional motor recovery of the affected forelimb following focal ischemic brain injury in rats. Eur. J. Neurosci. 21, 989–999. Brea, D., Blanco, M., Ramos-Cabrer, P., Moldes, O., Arias, S., Pérez-Mato, M., Leira, R., Sobrino, T., Castillo, J., 2011. Toll-like receptors 2 and 4 in ischemic stroke: outcome and therapeutic values. J. Cereb. Blood Flow Metab. 31, 1424–1431. Caso, J.R., Pradillo, J.M., Hurtado, O., Lorenzo, P., Moro, M.A., Lizasoain, I., 2007. Toll-like receptor 4 is involved in brain damage and inflammation after experimental stroke. Circulation 115, 1599–1608. Chiang, L.Y., Lu, F.J., Lin, J.T., 1995. Free radical scavenging activity of water-soluble fullerenols. J. Chem. Soc. Chem. Commun. 12, 1283–1284. Corbett, D., Nurse, S., 1998. The problem of assessing effective neuroprotection in experimental cerebral ischemia. Prog. Neurobiol. 54, 531–548. Del Zoppo, G.J., 2009. Inflammation and the neurovascular unit in the setting of focal cerebral ischemia. Neuroscience 158, 972–982.

F. Fluri et al. / Experimental Neurology 265 (2015) 142–151 Dugan, L.L., Gabrielsen, J.K., Yu, S.P., Lin, T.S., Choi, D.W., 1996. Buckminsterfullerenol free radical scavengers reduce excitotoxic and apoptotic death of cultured cortical neurons. Neurobiol. Dis. 3, 129–135. Eltzschig, H.K., Carmeliet, P., 2011. Hypoxia and inflammation. N. Engl. J. Med. 364, 656–665. Fisher, M., Feuerstein, G., Howells, D.W., Hurn, P.D., Kent, T.A., Savitz, S.I., Lo, E.H., 2009. Update of the stroke therapy academic industry roundtable preclinical recommendations. Stroke 40, 2244–2250. Gerriets, T., Stolz, E., Walberer, M., Müller, C., Rottger, C., Kluge, A., Kaps, M., Fisher, M., Bachmann, G., 2004. Complications and pitfalls in rat stroke models for middle cerebral artery occlusion: a comparison between the suture and the macrosphere model using magnetic resonance angiography. Stroke 35, 2372–2377. Haga, S., Ikeda, K., Sato, M., Ishii, T., 1993. Synthetic Alzheimer amyloid β/A4 peptides enhance production of complement C3 component by cultured microglial cells. Brain Res. 601, 88–94. Howells, D.W., Porritt, M.J., Rewell, S.S.J., O'Collins, V., Sena, E.S., van der Worp, H.B., Traystman, R.J., Macleod, M.R., 2010. Different strokes for different folks: the rich diversity of animal models of focal cerebral ischemia. J. Cereb. Blood Flow Metab. 30, 1412–1431. Hwang, S.-Y., Shin, J.-H., Hwang, J.-S., Kim, S.-Y., Shin, J.-A., Oh, E.-S., Oh, S., Kim, J.-B., Lee, J.-K., Han, I.-O., 2010. Glucosamine exerts a neuroprotective effect via suppression of inflammation in rat brain ischemia/reperfusion injury. Glia 58, 1881–1892. Iadecola, C., Anrather, J., 2011. The immunology of stroke: from mechanisms to translation. Nat. Med. 17, 796–808. Jack, C.S., Arbour, N., Manusow, J., Montgrain, V., Blain, M., McCrea, E., Shapiro, A., Antel, J.P., 2005. TLR signaling tailors innate immune responses in human microglia and astrocytes. J. Immunol. 175, 4320–4330. Jiménez-Altayó, F., Martín, A., Rojas, S., Justicia, C., Briones, A.M., Giraldo, J., Planas, A.M., Vila, E., 2007. Transient middle cerebral artery occlusion causes different structural, mechanical, and myogenic alterations in normotensive and hypertensive rats. Am. J. Physiol. Heart Circ. Physiol. 293, H628–H635. Jin, H., Chen, W.Q., Tang, X.W., Chiang, L.Y., Yang, C.Y., Schloss, J.V., Wu, J.Y., 2000. Polyhydroxylated C(60), fullerenols, as glutamate receptor antagonists and neuroprotective agents. J. Neurosci. Res. 62, 600–607. Kawamata, T., Alexis, N.E., Dietrich, W.D., Finklestein, S.P., 1996. Intracisternal basic fibroblast growth factor (bFGF) enhances behavioral recovery following focal cerebral infarction in the rat. J. Cereb. Blood Flow Metab. 16, 542–547. Kuroda, S., Tsuchidate, R., Smith, M.-L., Maples, K.R., Siesjö, B.K., 1999. Neuroprotective effects of a novel nitrone, NXY-059. After transient focal cerebral ischemia in the rat. J. Cereb. Blood Flow Metab. 19, 778–787. Lin, A.M.-Y., Fang, S.-F., Lin, S.-Z., Chou, C.-K., Luh, T.-Y., Ho, L.-T., 2002. Local carboxyfullerene protects cortical infarction in rat brain. Neurosci. Res. 43, 317–321.

151

Liu, S., Zhen, G., Meloni, B.P., Campbell, K., Winn, H.R., 2009. Rodent stroke model guidelines for preclinical stroke trials (1st edition). J. Exp. Stroke Transl. Med. 2, 2–27. Longa, E.Z., Weinstein, P.R., Carlson, S., Cummins, R., 1989. Reversible middle cerebral artery occlusion without craniectomy in rats. Stroke 20, 84–91. Macleod, M.R., van der Worp, H.B., Sena, E.S., Howells, D.W., Dirnagl, U., Donnan, G.A., 2008. Evidence for the efficacy of NXY-059 in experimental focal cerebral ischaemia is confounded by study quality. Stroke 39, 2824–2829. Marks, L., Carswell, H.V., Peters, E.E., Graham, D.I., Patterson, J., Dominiczak, A.F., Macrae, I.M., 2001. Characterization of the microglial response to cerebral ischemia in the stroke-prone spontaneously hypertensive rat. Hypertension 38, 116–122. McCabe, C., Gallagher, L., Gsell, W., Graham, D., Dominiczak, A.F., Macrae, I.M., 2009. Differences in the evolution of the ischemic penumbra in stroke-prone spontaneously hypertensive and Wistar-Kyoto rats. Stroke 40, 3864–3868. Mocco, J., Mack, W.J., Ducruet, A.F., Sosunov, S.A., Sughrue, M.E., Hassid, B.G., Nair, M.N., Laufer, I., Komotar, R.J., Claire, M., Holland, H., Pinsky, D.J., Connolly, E.S., 2006. Complement component C3 mediates inflammatory injury following focal cerebral ischemia. Circ. Res. 99, 209–217. Moskowitz, M.A., Lo, E.H., Iadecola, C., 2010. The science of stroke: mechanisms in search of treatments. Neuron 67, 181–198. Peerschke, E.I., Yin, W., Ghebrehiwet, B., 2010. Complement activation on platelets: Implications for vascular inflammation and thrombosis. Mol. Immunol. 47, 2170–2175. Reglodi, D., Tamás, A., Lengvári, I., 2003. Examination of sensorimotor performance following middle cerebral artery occlusion in rats. Brain Res. Bull. 59, 459–466. Sarkar, D., 2008. Lattice: Multivariate Data Visualization With R. Springer, New York. Sayes, C.M., Fortner, J.D., Guo, W., Lyon, D., Boyd, A.M., Ausman, K.D., Tao, Y.J., Sitharaman, B., Wilson, L.J., Hughes, J.B., West, J.L., Colvin, V.L., 2004. The differential cytotoxicity of water-soluble fullerenes. Nano Lett. 4, 1881–1887. Shuaib, A., Lees, K.R., Lyden, P., Grotta, J., Davalos, A., Davis, S.M., Diener, H.-C., Ashwood, T., Wasiewski, W.W., Emeribe, U., 2007. NXY-059 for the treatment of acute ischemic stroke. N. Engl. J. Med. 357, 562–571. Song, Y.S., Lee, Y.-S., Narasimhan, P., Chan, P.H., 2006. Reduced oxidative stress promotes NF-κB-mediated neuroprotective gene expression after transient focal cerebral ischemia: lymphocytotrophic cytokines and antiapoptotic factors. J. Cereb. Blood Flow Metab. 27, 764–775. Yamago, S., Tokuyama, H., Nakamura, E., Kikuchi, K., Kananishi, S., Sueki, K., Nakahara, H., Enomoto, S., Ambe, F., 1995. In vivo biological behavior of a water-miscible fullerene: 14C labeling, absorption, distribution, excretion and acute toxicity. Chem. Biol. 2, 385–389. Zhao, Z., Cheng, M., Maples, K.R., Ma, J.Y., Buchan, A.M., 2001. NXY-059, a novel free radical trapping compound, reduces cortical infarction after permanent focal cerebral ischemia in the rat. Brain Res. 909, 46–50.