Neuroprotective effect of aminoguanidine on iron-induced neurotoxicity

Brain Research Bulletin 76 (2008) 57–62

Research report

Neuroprotective effect of aminoguanidine on iron-induced neurotoxicity ¨ M. Omer Bostanci ∗ , Faruk Ba˘girici Department of Physiology, Faculty of Medicine, Ondokuz Mayis University, 55139 Samsun, Turkey Received 23 June 2007; received in revised form 4 November 2007; accepted 20 November 2007 Available online 17 December 2007

Abstract Iron is a commonly used metal to induce neuronal hyperactivity and oxidative stress. Iron levels rise in the brain in some neurodegenerative disorders such as Parkinson’s and Alzheimer’s diseases. A body of evidence indicates a link between neuronal death and nitric oxide. The present study was performed to investigate whether nitric oxide produced by inducible nitric oxide synthase is involved in iron-induced neuron death. For this purpose rats were divided into four groups: control, iron, aminoguanidine and iron + aminoguanidine. Animals in iron and iron + aminoguanidine groups received intracerebroventricular FeCl3 injection (200 mM, 2.5 ␮l). Rats belonging to control and aminoguanidine groups received the same amount of saline into the cerebral ventricles. All animals were kept alive for 10 days following the operation and animals in aminoguanidine and iron + aminoguanidine groups received intraperitoneal aminoguanidine injections once a day (100 mg/kg day) during this period. After 10 days, rats were perfused intracardially under deep urethane anesthesia. Removed brains were processed using the standard histological techniques. The total numbers of neurons in hippocampus of all rats were estimated with the unbiased stereological techniques. It was found that aminoguanidine decreased mean neuron loss from 43.4% to 20.3%. Results of the present study suggest that aminoguanidine may attenuate the neurotoxic effects of iron by inhibiting inducible nitric oxide synthase. © 2007 Elsevier Inc. All rights reserved. Keywords: Aminoguanidine; Cell death; Hippocampus; Iron; Stereology

1. Introduction Iron plays an important role in maintaining normal brain function and it is the most abundant transitional metal in the brain [27]. Iron level increases in the brain in some neurodegenerative disorders including Hallervorden-Spatz Syndrome, Parkinson’s and Alzheimer’s diseases [1,33,40]. Iron is also thought to be involved in the mechanism of posttraumatic epileptogenesis [46]. Subpial injections of iron salts lead to formation of brain edema, cavity lesions and recurrent epileptiform discharges as well as inducing the free radical formation [45]. In previous stereological studies, it has been demonstrated that intracortically or intracerebroventiculary administered iron caused hippocampal cell loss [5,6]. Hippocampus is known to be involved in learning and memory processes. Previous studies indicate that in Alzheimer’s disease, iron levels in amygdala, pyriform cortex, hippocampus and olfactory regions were shown to be elevated significantly [10,36].

∗

Corresponding author. Tel.: +90 362 3121919x3065; fax: +90 362 4576041. ¨ Bostanci). E-mail address: [email protected] (M.O.

0361-9230/$ – see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.brainresbull.2007.11.011

Nitric oxide (NO) is one of the substances that are involved in iron-induced cell death [24,34]. NO has been suggested to be a cell-to-cell signaling molecule which regulates guanylyl cyclase [12], aconitase, and iron regulatory protein [31]. In the central nervous system, NO plays a multifaceted role as a neurotransmitter and a regulator of cerebral blood flow, and is also involved in brain development, memory formation and behavior via regulation of synaptic plasticity [11]. The production of NO in excess has been implicated in the induction of cytotoxicity in several types of cell [9,35,38]. However, some studies found that NO is a neuroprotective substance [24,43]. NO is produced by three different nitric oxide synthases (NOS): endothelial (eNOS), neuronal (nNOS) and inducible nitric oxide synthase (iNOS). Inducible NOS has not been found in intact brain, but following inflammation, it produces massive amount of NO [19,29]. In the previous studies, it has been shown that aminoguanidine, generally known as a selective inhibitor of iNOS [14,23], has positive effects on ischemic outcomes [8,9,18] and zinc-induced neurotoxicity [13]. In the present study, we aimed to evaluate the effects of aminoguanidine in neurotoxicity produced by iron injection into brain. Hippocampus was chosen to investigate the effects of

58

¨ Bostanci, F. Ba˘girici / Brain Research Bulletin 76 (2008) 57–62 M.O.

aminoguanidine in neurotoxicity since it is an important region involved in learning and memory processes. These effects were assessed by the latest and unbiased stereological techniques. This is the first stereological study to demonstrate the effects of aminoguanidine on iron-induced neurotoxicity. 2. Materials and methods

After the cerebra and cerebella were separated physically, brains were processed using the standard histological techniques (alcohol and chloroform series) and embedded in paraplast embedding media. Serial tissue sections were obtained using a rotary microtome (Leica RM 2135) in horizontal plane with a section thickness of 40 ␮m and stained with cresyl violet. Approval of Ethical Committee of Ondokuz Mayis University was obtained prior to experiments and all animal work was performed according to the Experimental Animal Care Rules of European Community Council.

2.1. Animals

2.3. Stereological analysis

Four-month-old male Wistar rats weighing 220 ± 30 g were used. Animals were divided into four groups: control (n = 10), iron (n = 10), aminoguanidine (n = 10) and iron + aminoguanidine (n = 10). All animals were obtained from Experimental Research Center of Ondokuz Mayis University. Rats were kept in heat-regulated rooms, on a 12-h:12-h day/night cycle, and given as much food and water as needed.

The cytoarchitectonic characteristics of the regio superior (CA1) and regio inferior (CA2 and CA3) were identified using the criteria of West et al. [44]. Although these pyramidal cell layers are well delineated at the central levels of the septotemporal axis of the hippocampus, delineation becomes more complicated as one proceeds towards the septal and temporal poles. In the preparations used, the characteristic features of the CA1, CA2 and CA3 layers were not found to be sufficient for consistent definition at some peripheral levels. Since unambiguous definition of profile boundaries at all levels is a prerequisite to obtain a reliable quantitative data, we considered these subdivisions as a single layer (Fig. 1). In the present study, 14––17 sections were sampled in a systematic random fashion (ssf: 1/7) out of a total of 130 horizontal sections per individual hippocampi. First sections were chosen randomly from the first set of seven sections containing the hippocampus and then the consecutive samples were selected with a fixed interval of seven sections. Hippocampal pyramidal neurons were counted using the optical fractionator counting method which is a combination of fractionator sampling scheme and disector counting technique [44]. All counting and analysis were performed using a modified computerassisted stereological analysis system. Areas for cell counting were determined and delineated using CAST Grid stereological analysis software (Olympus, Denmark). Cell counts were done using a sampling scheme optimized for a total of approximately 500 cell counts per individual. Determined pyramidal sectional areas were scanned automatically using consecutive steps with 250 ␮m × 250 ␮m x–y size. Every step in this scanning was individually analyzed with optical disector probes using 100× oil-objectives. During optical disector application, an unbiased counting frame comprising the 15% of the total step area was used for particle sampling and counting. Thus the area sampling fraction (asf) is determined as 445 ␮m2 /62,500 ␮m2 . The last sampling level in optical fractionator applications is the thickness sampling stage. Optical disector counting requires a virtual vertical scanning of the section of interest in order to count stacks of particles. According to previous pilot studies, a fixed disector height of 10 ␮m was predetermined and used throughout the study. This height is formed by virtual movement of a section plane (the focal plane) through the section thickness. Generally a narrow upper “guard zone” passed before the actual optic disector counting in order to avoid the possible irregularities of the sectional surface. Here we left a 5 ␮m upper guard zone, applied particle counting through a 10 ␮m disector height and mea-

2.2. Operation Animals were kept away from food for 12 h prior to surgery and all animals were weighed just before the surgical operation. Anesthesia was induced by intraperitoneal (i.p.) injection of ketamine hydrochloride (100 mg/kg). Animals were fixed to a stereotaxic apparatus and a rostro-caudal incision of 2 cm length was made using an electric cutter (Ellman Surgitron). For the intracerebroventricular (i.c.v.) injection, Bregma line was exposed clearly and a hole with a diameter of 1 mm was drilled. The coordinates for hole were: 0.6 mm posterior to Bregma, 2.0 mm lateral to the midline. Iron was given through this hole to a depth of 4.2 mm using a Hamilton (Type 701N) syringe. These coordinates were determined according to rat brain atlas of Paxinos and Watson [32]. Rats in the control and aminoguanidine groups received 2.5 ␮l saline while rats in iron and iron + aminoguanidine groups received 200 mM (2.5 ␮l, i.c.v.) FeCl3 [5,45]. Then, incisions were sutured and incision area was cleaned using 10% povidon iodide just prior the placement of the animals to their cages. All animals were allowed to survive for 10 days following the surgery. Only the rats belonging to aminoguanidine and iron + aminoguanidine groups received additional i.p. aminoguanidine treatment as 100 mg/kg day for 10 days. The first dose of aminoguanidine was administered within the first 5 min following the surgical operation. Rats belonging to other groups received same amount of i.p. saline injection for 10-day survival period. After the survival period, all animals perfused intracardially under deep urethane anesthesia with 10% formaldehyde and saline, buffered for pH 7.2. After the completion of the perfusion process all animals were decapitated and brains were removed immediately. Brains of the three animals from each group were used for the determination of tissue iron levels. For this purpose, removed brains were weighed and placed into plastic dishes. They were stored in frozen (−70 ◦ C) prior to analysis. Brains for the stereological analysis soaked in the same fixative (buffered 10% formaldehyde) at room temperature for 2 weeks.

Fig. 1. A cresyl violet stained section through the hippocampus of a Wistar rat. To the right is a diagram of the layers in the hippocampal formation. Pyramidal cell counting was performed in the CA1, CA2 and CA3 regions. g: Granule cell layer of the dentate gyrus, h: hilus of the dentate gyrus, ri: regio inferior (CA3 and CA2), rs: regio superior (CA1), s: subiculum.

¨ Bostanci, F. Ba˘girici / Brain Research Bulletin 76 (2008) 57–62 M.O. sured the section thickness. All such measurements were done using a digital microcator (Heidenhain, Germany), incorporated in the stereological analysis system. Thus the final sampling stage, generally called the thickness sampling fraction (tsf) was calculated by [disector height]/[mean section thickness]. Average section thickness was estimated for each section by measuring the thickness of every 10th field of counting with a random start and by averaging the measured thickness values for each section. The total average of section thickness was 23.60 ± 1.74 ␮m among all animals. After completing a throughout sampling for all sampled sections, properly sampled pyramidal cells were counted as disector particles (Q− ). Total number of hippocampal pyramidal neurons (N) was then calculated using the following formulation: N=

1 1 1  ssf

×

asf

×

tsf

×

Q−

Data obtained from counting are expressed as mean ± S.E.M. and checked for distribution using Kolmogorov–Smirnov test. Hemispheric differences in terms of neuron numbers were tested using paired T-test and differences between groups were tested using post hoc tukey honestly significant difference test. A value of p < 0.05 was considered statistically significant.

2.4. Determination of tissue iron levels Iron levels of the brain tissue were determined by dry ashing method using an atomic absorption apparatus (PerkinElmer 2280, FLAME). Homogenized brain tissue was put in porcelain crucible and placed into a furnace at 100 ◦ C for 2 days. Then, the temperature of the furnace was raised to 450 ◦ C (50 ◦ C/h). Brain tissue was burned for about 8 h, until a white (or grayish) ash residue was obtained. Ashes were then dissolved in 3 ml 3N HCI. The solution was filtered with Whatman-49 filter paper then transferred to a 25 ml volumetric flask. The ash solution was then diluted with 0.36 N HCI. During the process, iron contamination from the environment was avoided. Each such solution was measured using 248.3-nm wavelength with the atomic absorption spectrophotometer. One-way analysis of variance (ANOVA) was employed to compare differences between groups.

2.5. Chemicals Aminoguanidine, FeCl3 ·6H2 O and cresyl violet obtained from Sigma Chemical Co. (St. Louis, Mo, USA); entellan, xylene and acetic acid obtained from Merck (Darmstadt, Germany); absolute alcohol, formaldehyde, chloroform and others were obtained from Aklar Chemistry (Ankara, Turkey).

59

Table 1 Total pyramidal cell numbers in left and right hippocampus Regions

Groups

Total cell number (N ± S.E.M)

Left Hipp

Control Iron AG Iron + AG

640,156 361,026 637,294 517,178

± ± ± ±

Right Hipp

Control Iron AG Iron + AG

651,994 370,093 643,563 512,260

± ± ± ±

CV

CE

13,158 10,303 19,676 14,592

0.0521 0.0843 0.0715 0.0463

0.042 0.053 0.061 0.056

18,186 15,530 14,744 12,287

0.0524 0.0921 0.0671 0.0367

0.055 0.035 0.043 0.048

S.E.M., standard error of the mean; CV, coefficient of variation; CE, coefficient of error; AG, aminoguanidine; Hipp, hippocampus.

(Fig. 3). Rats in iron + aminoguanidine group have 19.2% lower hippocampal cell numbers with respect to controls (p < 0.001). Treatment of aminoguanidine alone did not significantly change left hippocampal neuron numbers with respect to control values (p > 0.05). Comparison between iron and iron + aminoguanidine rats revealed that aminoguanidine significantly attenuates the iron-induced neuron loss from 43.6% to 19.2% and protect hippocampal neurons against iron toxicity (p < 0.001). Estimated mean total neuron number of right hippocampi was found to be 651,994 ± 18,186 in control group (Table 1). Iron group displays a marked decrease (43.2%) in hippocampal neuron number when compared to controls (p < 0.001). Neuron loss in iron + aminoguanidine group with respect to controls was also significant and was 21.4% (p < 0.001). Aminoguanidine appears to attenuate the iron-induced neuron loss from 43.2% to 21.4% and seems to have a significant neuroprotective effect against the iron-induced neurotoxicity (p < 0.001, Fig. 2). Right hippocampal neuron numbers also were not significantly changed by the administration of aminoguanidine alone with respect to control values (p > 0.05). There were no significant differences between left and right hippocampal neuron numbers in each group (p > 0.05).

3. Results Iron amount in the brain was determined to control whether iron level was increased artificially by the injection. Total iron levels in rat brain tissue were found to be 108 ± 12 ␮g in control rats (n = 3), 176 ± 15 ␮g in iron group (n = 3), 111 ± 16 ␮g in aminoguanidine group (n = 3) and 170 ± 12 ␮g in iron + aminoguanidine group (n = 3). Difference between iron and non-iron applied animals were statistically significant (p < 0.001). Pyramidal cells were counted separately in left and right hippocampi, since iron was injected in left cerebral ventricle. Estimated mean total neuron number of left hippocampi was found to be 640,156 ± 13,158 in control group (Table 1). After comparing iron group with controls, rats in iron group were appeared to have 43.6% less number of neurons than control group and this difference was statistically significant (p < 0.001, Fig. 2). Representative light micrographs from the same levels of left hippocampal sections show that cell numbers in iron + aminoguanidine group were more than iron group

Fig. 2. The effects of iron, aminoguanidine and iron + aminoguanidine on left and right hippocampal pyramidal cell number. Error bars represent ± S.E.M. (*) p < 0.001 compared to control and (#) p < 0.001 compared to iron group. AG, aminoguanidine.

60

¨ Bostanci, F. Ba˘girici / Brain Research Bulletin 76 (2008) 57–62 M.O.

Fig. 3. Representative photomicrographs from the same level of regio inferior pyramidal neurons. (A) Control. (B) Aminoguanidine alone did not affect hippocampal neuron number. (C) Iron (200 mM) decreased hippocampal pyramidal cell number. (D) Aminoguanidine treatment (100 mg/kg) reversed iron-induced neurotoxicity. Scale bar: 2 ␮m.

4. Discussion In the present study, we investigated the effects of aminoguanidine, an iNOS inhibitor, on iron-induced hippocampal cell loss in rats. Total number of neurons in hippocampus was determined by optical fractionator technique, which is a combination of systematic random sampling and the optical disector counting method [44]. According to our findings, aminoguanidine significantly decreases total (left and right) hippocampal cell loss from 43.4% to 20.3% and protects neurons against iron toxicity. This is the first stereological study that demonstrates the neuroprotective effects of aminoguanidine on iron-induced neurotoxicity. Stereological methods are known to efficiently prevent sources of bias from individuals and from the methodology itself, and hence allow us to determine the number of neurons in a certain brain region in an unbiased and efficient manner. In modern stereological cell counting approaches, each and every cell filling the region of interest have an equal probability of being a part of the final sample. This ensures the unbiased sampling and unbiased estimation of total neuron number [37]. Therefore, we have preferred this method to determine the hippocampal neuron number. Iron plays an important role in maintaining normal brain function. Iron overload and enhanced hydroxyl radical formation have been implicated as the causative factors of some neurodegenerative disorders. However, a key question – why do iron levels increase abnormally in the brain? – has not been answered. Iron is a frequently used metal in order to induce lipid peroxidation and cellular damage. Ferro (Fe2+ ) and ferric (Fe3+ ) iron produce hydroxyl radicals via Fenton and Haber–Weis reactions

and cause cellular damage [2,7]. In the present study, iron levels in the brain were elevated artificially to induce oxidative stress. According to our results, there were no statistically differences between left and right hippocampal neuron numbers in iron-treated groups. Presumably, this finding may depend on intracerebroventricular injection of iron. In this way, iron molecules may have dispersed to whole brain in an equal manner. Iron as well has been used to induce epileptic activity. Meyerhoff et al. have applied iron chloride (600 mM in a volume of 10 ␮l) intracortically to induce epilepsy [26]. In our study, however, none of the animals displayed seizure behavior during the survival period. We have injected 2.5 ␮l of 200 mM iron chloride (i.c.v.) solution to induce free radical formation. Hence our dose of iron may not be sufficient to trigger a detectable level of epileptiform activity. In addition, the route of iron application in our experimental design might be another factor for absence of epileptic behavior. Subpial injections of iron salt solutions lead to free radical formation in the rat brain [45]. Free radicals are the products of routine energy metabolism and hence produced continuously during the oxidation–reduction reactions in the cell. Oxidative stress is the major cause of damage associated with elevated NO [4,15], resulting largely from the formation of peroxynitrite (ONOO− ) [3,4,28]. Reaction of superoxide (O• 2 − ) with NO forms the short-lived ONOO− (half life <1 s) which is far more reactive and damaging than its precursors [3]. This reaction is extremely rapid, therefore ONOO− forms whenever O• 2 − and NO are produced simultaneously. NO also causes oxidative stress by other pathways. NO reacts with oxygen to produce reactive oxides of nitrogen (NOx ), which may form preferentially within lipid membranes [25], however the amount

¨ Bostanci, F. Ba˘girici / Brain Research Bulletin 76 (2008) 57–62 M.O.

formed in vivo is uncertain [47]. NO enhances the oxidative stress caused by hydrogen peroxide [21]. The iron liberated from ferritin by NO may contribute significantly to elevated oxidative stress [16]. It is known that, under physiological conditions, iNOS is absent from mammalian cells. However, in pathological conditions iNOS is activated at the transcriptional level by proinflammatory stimuli such as bacterial lipopolysaccharide or cytokines, including tumor necrosis factor-[alpha], interferon-[gamma] and interleukin-1 [41,42]. Another possible mechanism for iNOS expression involves the generation of hemin from methemoglobin during hemolysis of subarachnoid blood [39]. It has been shown that aminoguanidines have neuroprotective effects on optic nerve atrophy [20] and zinc-induced hippocampal cell loss [13]. Furthermore, widespread neuronal loss in the CA1 region of the hippocampus after bilateral occlusion of the carotid arteries in rats was diminished by application of aminoguanidine [30] and this substance decreased both the duration and the severity of experimental encephalomyelitis in rats [50]. These findings are supported by the results of the present study that the inhibition of iNOS significantly decreased ironinduced hippocampal cell loss. In the previous study, it has been shown that the administration of aminoguanidine did not substantially modify arterial pressure, arterial blood gases, pH, hematocrit, plasma glucose or rectal temperature [49]. Iadecola et al. proposed the possibility that aminoguanidine acts via mechanisms independent of iNOS inhibition [17]. Aminoguanidine is known to act on a variety of other cellular metabolic pathways and some of these actions may be relevant in biochemical events [8]. Aminoguanidine inhibits polyamine metabolizing enzymes polyamine oxidase and diamine oxidase which evidently are targets for the neuroprotective effects of aminoguanidine [22]. Free radical scavenging is another mechanism by which aminoguanidine may lead to neuroprotection [48]. In summary, results of the present study clearly demonstrate that treatment with aminoguanidine, a selective iNOS inhibitor, significantly attenuates the iron-induced cell loss in the rat hippocampus. Further studies are needed to reveal the possible molecular mechanisms underlying the neuroprotective effects of aminoguanidine. Conflict of interest statement There is no conflict of interest. References [1] P. Aisen, M. Wessling-Resnick, E.A. Leibold, Iron metabolism, Curr. Opin. Chem. Biol. 3 (1999) 200–206. [2] S.D. Aust, L.A. Morehouse, C.E. Thomas, Role of metals in oxygen radicals reactions, J. Free Rad. Biol. Med. 1 (1985) 3–25. [3] J.S. Beckman, T.W. Beckman, J. Chen, P.A. Marshall, B.A. Freeman, Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide, Proc. Natl. Acad. Sci. USA 87 (1990) 1620–1624. [4] J.S. Beckman, W.H. Koppenol, Nitric oxide, superoxide and peroxynitrite: the good, the bad and the ugly, Am. J. Physiol. 271 (1996) 1424–1437.

61

[5] M.O. Bostanci, F. Bagirici, S. Canan, A calcium channel blocker flunarizine attenuates the neurotoxic effects of iron, Cell. Biol. Toxicol. 22 (2006) 119–125. [6] M.O. Bostanci, F. Bagirici, A. Korkmaz, The neurotoxic effect of iron on pyramidal cell number in rat hippocampus: a stereological study, Neurosci. Res. Commun. 32 (2003) 151–159. [7] J.M. Braughler, L.A. Duncan, R.L. Chase, The involvement of iron in lipid peroxidation, J. Biol. Chem. 261 (1986) 10282–10289. [8] D. Cash, J.S. Beech, R.C. Rayne, P.M.W. Bath, B.S. Meldrum, S.C.R. Williams, Neuroprotective effect of aminoguanidine on transient focal ischemia in the rat brain, Brain Res. 905 (2001) 91–103. [9] V. Danielisova, M. Nemethova, J. Burda, The protective effect of aminoguanidine on cerebral ischemic damage in the rat brain, Physiol. Res. 53 (2004) 533–540. [10] M.A. Deibel, W.D. Ehmann, W.R. Markesbery, Copper, iron and zinc imbalances in severely degenerated brain regions in Alzheimer’s disease: possible relation to oxidative stress, J. Neurol. Sci. 143 (1996) 137–142. [11] J.V. Esplugues, NO as a signalling molecule in the nervous system, Br. J. Pharmacol. 135 (2002) 1079–1095. [12] J. Garthwaite, C.L. Boulton, Nitric oxide signaling in the central nervous system, Annu. Rev. Physiol. 57 (1995) 683–706. [13] F.M. Gokce, F. Bagirici, S. Kaplan, S. Demir, M. Ayyildiz, C. Marangoz, A NOS inhibitor aminoguanidine reduces zinc-induced neuron loss in rat hippocampus, Neurosci. Res. Commun. 33 (2003) 53–62. [14] M.J.D. Griffiths, M. Messent, N.P. Curzen, T.W. Evans, Aminoguanidine selectively decreases cyclic GMP levels produced by inducible nitric oxide synthase, Am. J. Res. Crit. Care Med. 152 (1995) 1599–1604. [15] S.S. Gross, M.S. Wolin, Nitric oxide: pathophysiological mechanisms, Annu. Rev. Physiol. 57 (1995) 737–769. [16] B.H. Halliwell, J.M.C. Gutteridge, Free Radicals in Biology and Medicine, Oxford University Press, Oxford, 1989. [17] C. Iadecola, F. Zhang, R. Casey, H.B. Clark, M.E. Ross, Inducible nitric oxide synthase gene expression in vascular cells after transient focal cerebral ischemia, Stroke 27 (1996) 1373–1380. [18] C. Iadecola, F. Zhang, R. Casey, M. Nagayama, M.E. Ross, Delayed reduction of ischemic brain injury and neurological deficits in mice lacking the inducible nitric oxide synthase gene, J. Neurosci. 17 (1997) 9157–9164. [19] C. Iadecola, F. Zhang, X. Xu, Inhibition of inducible nitric oxide synthase gene expression in brain following focal cerebral ischemia, Am. J. Physiol. 268 (1995) 286–292. [20] M. Inoue, N. Ohgiya, M. Yamamoto, Effect of aminoguanidine on optic nerve involvement in experimental diabetic rats, Brain Res. 800 (1998) 319–322. [21] I. Ioannidis, H. de Groot, Cytotoxicity of nitric oxide in FU5 hepatoma cells: evidence for co-operative action with hydrogen peroxide, Biochem. J. 296 (1993) 341–345. [22] S. Ivanova, G.I. Botchkina, Y. Al-Abed, M. Meistrell, F. Betliwalla, J.M. Dubinsky, C. Iadecola, H. Wang, P.K. Gregersen, J.W. Eaton, K.J. Tracey, Cerebral ischemia enhances polyamine oxidation: identification of enzymatically formed 3-aminopropanal as an endogenous mediator of neuronal and glial cell death, J. Exp. Med. 188 (1998) 327–340. [23] S.L. Leib, Y.S. Kim, S.M. Black, J.H. Tureen, M.G. Tauber, Inducible nitric oxide synthase and the effect of aminoguanidine in experimental neonatal meningitis, J. Infect. Dis. 177 (1998) 692–700. [24] A.M.Y. Lin, Recovery by NO of the iron-attenuated dopamine dynamics in nigrostriatal system of rat brain, Neurosci. Res. 34 (1999) 133–139. [25] X. Liu, M.J.S. Miller, M.S. Joshi, D.D. Thomas, J.R. Lancaster, Accelerated reaction of nitric oxide with oxygen within the hydrophobic interior of biological membranes, Proc. Natl. Acad. Sci. USA 95 (1998) 2175– 2179. [26] J.L. Meyerhoff, J.K. Lee, B.W. Rittase, A.Y. Tsang, D.L. Yourick, Lipoic acid pretreatment attenuates ferric chloride-induced seizures in the rat, Brain Res. 1016 (2004) 139–144. [27] T. Moos, Brain iron homeostasis, Dan. Med. Bull. 49 (2002) 279–301. [28] M.P. Murphy, M.A. Packer, J.L. Scarlett, S.W. Martin, Peroxynitrite: a biologically significant oxidant, Gen. Pharmacol. 31 (1998) 179–186. [29] C. Nathan, Inducible nitric oxide synthase: what difference does it make? J. Clin. Invest. 100 (1997) 2417–2423.

62

¨ Bostanci, F. Ba˘girici / Brain Research Bulletin 76 (2008) 57–62 M.O.

[30] M.J. O’Neill, C. Hicks, M. Ward, J.A. Panetta, Neuroprotective effects of the antioxidant LY231617 and NO synthase inhibitors in global cerebral ischemia, Brain Res. 760 (1997) 170–178. [31] K. Pantopoulos, G. Weiss, M. Hentze, Nitric oxide and the posttranscriptional control of cellular iron traffic, Trends Cell Biol. 4 (1994) 82–86. [32] G. Paxinos, C. Watson, The Rat Brain in the Stereotaxic Coordinates, fourth ed., Academic Press, London, 1998. [33] Z.M. Qian, Q. Wang, Y. Pu, Brain iron and neurological disorders, Chin. Med. J. 110 (1997) 455–458. [34] P. Rauhala, I. Sziraki, C.C. Chiueh, Peroxidation of brain lipids in vitro: nitric oxide versus hydroxyl radicals, Free Rad. Biol. Med. 21 (1996) 391–394. [35] E. Rodriguez-Martin, M.J. Casarejos, S. de Canals, S. Bernardo, M.A. Mena, Thiolic antioxidants protect from nitric oxide-induced toxicity in fetal midbrain cultures, Neuropharmacology 43 (2002) 877–888. [36] D.L. Samudralwer, C.C. Diprete, B.-F. Ni, W.D. Ehmann, W.R. Markesbery, Elemental imbalances in the olfactory pathway in Alzheimer’s disease, J. Neurol. Sci. 130 (1995) 139–145. [37] C. Schmitz, P.R. Hof, Design-based stereology in neuroscience, Neuroscience 130 (2005) 813–831. [38] A.M. Sharifi, S.H. Mousavi, M. Bakhshayesh, F.K. Tehrani, M. Mahmoudiana, S. Oryan, Study of correlation between lead-induced cytotoxicity and nitric oxide production in PC12 cells, Toxicol. Lett. 160 (2005) 43–48. [39] S. Suzuki, N.F. Kassell, K.S. Lee, Hemin activation of an inducible isoform of nitric oxide synthase in vascular smooth-muscle cells, J. Neurosurg. 83 (1995) 862–866. [40] K.F. Swaiman, Hallervorden-Spatz and brain iron metabolism, Arch. Neurol. 48 (1991) 1285–1293. [41] C. Szabo, C. Thiemermann, Regulation of the expression of the inducible isoform of nitric oxide synthase, Adv. Pharmacol. 34 (1995) 113–153.

[42] J.R. Vane, J.A. Mitchell, I. Appleton, A. Tomlinson, D. Bishop-Bailey, J. Croxtall, D.A. Willoughby, Inducible isoforms of cyclooxygenase and nitric oxide synthase in inflammation, Proc. Natl. Acad. Sci. USA 91 (1994) 2046–2050. [43] C. Verrecchia, R.G. Boulu, M. Plotkine, Neuroprotective and deleterious effects of nitric oxide on focal cerebral ischemia-induced neuron death, Adv. Neuroimmunol. 5 (1995) 359–378. [44] M.J. West, L. Slomianka, H.J.G. Gundersen, Unbiased stereological estimation of the total number of neurons in the subdivisions of the rat hippocampus using the optical fractionator, Anat. Record. 231 (1991) 482–497. [45] L.J. Willmore, M. Hiramatsu, H. Kochi, A. Mori, Formation of superoxide radicals after FeCl3 injection into rat isocortex, Brain Res. 277 (1983) 393–396. [46] L.J. Willmore, W.J. Triggs, Iron-induced lipid peroxidation and brain injury responses, Int. J. Dev. Neurosci. 9 (1991) 175–180. [47] D.A. Wink, P.C. Ford, Nitric oxide reactions important to biological systems: a survey of some kinetics investigations, Method 7 (1995) 14–20. [48] G. Yildiz, A.T. Demiryurek, I. Sahin-Erdemli, I. Kanzik, Comparison of antioxidant activities of aminoguanidine, methylguanidine and guanidine by luminol-enhanced chemiluminiscence, Br. J. Pharmacol. 124 (1998) 905–910. [49] F. Zhang, C. Iadecola, Temporal characteristics of the protective effect of aminoguanidine on cerebral ischemic damage, Brain Res. 802 (1998) 104–110. [50] W. Zhao, R.G. Tilton, J.A. Corbett, M.L. McDaniel, T.P. Misko, J.R. Williamson, A.H. Cross, W.F. Hickey, Experimental allergic encephalomyelitis in the rat is inhibited by aminoguanidine, an inhibitor of nitric oxide synthase, J. Neuroimmunol. 64 (1996) 123–133.