Flutamide Enhances Neuroprotective Effects of Testosterone during Experimental Cerebral Ischemia in Male Rats

Testosterone has been shown to worsen histological and neurological impairment during cerebral ischemia in animal models. Cell culture studies revealed that testosterone is implicated in protecting neural and glial cells against insults, and they started to elucidate testosterone pathways that underlie these protective effects. These studies support the hypothesis that testosterone can be neuroprotective throughout an episode of cerebral ischemia. Therefore, we evaluated the mechanisms underlying the shift between testosterone protective and deleterious effects via block testosterone aromatization and androgen receptors in rats subjected to 60-minute middle cerebral artery occlusion. Fifty rats were divided into five equal groups: gonadally intact male; castrated male; intact male + flutamide; intact male + letrozole; intact male + combination flutamide and letrozole. Our results indicated that castration has the ability to reduce histological damage and to improve neurological score 24 hours after middle cerebral artery occlusion. Moreover, flutamide improved histologic and neurological impairment better than castration. Letrozole induced increases in striatal infarct volume and seizures in gonadally intact rats. Combination of flutamide and letrozole showed that letrozole can reverse beneficial effects of flutamide. In conclusion, it seems that the beneficial effects of flutamide are the prevention of the deleterious effects and enhancement of neuroprotective effects of testosterone during cerebral ischemia.


Introduction
Stroke is a great reason of disability and death throughout the world [1]. Nowadays treatments are not very effective to reduce brain ischemia, whereas size of the infarct area will affect on patient's chance of recovery from a stroke and it will keep growing if treatments are not appropriate [1][2][3][4]. So, for reducing damage we need to �nd more effective treatments. Among risk factors, sex has prominent role in stroke [5][6][7][8]. Epidemiological studies have shown that overall incidence of stroke is higher in men relative to age-matched women in most countries [9][10][11][12]. Present evidence suggests that mechanisms of cell death and neuroprotection are not similar and equal in males and females [9,13,14]. A large part of this difference between sexes is attributed to sex steroids [14][15][16][17][18]. Previous studies demonstrate that estrogen and progesterone give protection against cerebral ischemia by several mechanisms [19][20][21][22][23]. On the other hand, data about androgens are sparse and controversial [9,18,24,25]. Human studies suggested male susceptibility to cerebral ischemia because male sex is a stroke risk factor in humans and low testosterone levels have been associated with risk for stroke and worse outcomes aer stroke in men [9-12, 14, 17]. In animal studies, data are contradictory and show that androgens can protect or make worse cerebral injury [17,[26][27][28][29]. In rodents, testosterone replacement before cerebral ischemia in castrates increments histological injury [17,26], whereas, stressors, such as halothane anesthesia, reduced brain injury, because when they applied before cerebral ischemia have potency to reduce plasma testosterone levels [30]. Moreover, testosterone replacement aer an episode of cerebral ischemia accelerates histological and behavioral recovery in castrated rats [29]. ese contradictory results may be reconciled by the hypothesis that testosterone has deleterious effects throughout an episode of cerebral ischemia but is bene�cial in the recovery period. Differently in vitro studies showed both detrimental and bene�cial effects of testosterone in neuronal cultures during exposure to different models of insult (e.g., oxidative stress, excitotoxicity, serum deprivation, and amyloid toxicity) [26,[31][32][33][34][35][36][37]. One explanation for these inconsistent results of in vitro studies is that two potentially competing mechanisms exert bene�cial and deleterious effects of testosterone during exposing to injury. ese two possible mechanisms for testosterone can be androgen receptor (AR) dependent pathways or aromatization to estrogen and then activation of estrogen pathways [38,39]. However, previous studies have not well determined effects of testosterone on cerebral cortex as well as striatum. None of these studies did whether the relation of testosterone and neurological de�cit during the acute phase of cerebral ischemia has been mediated via AR or cerebral aromatase [9,19,38]. Accordingly, we aimed to investigate protective and/or deleterious effects of testosterone on cerebral ischemia of cortex and striatum as well as their contribution to neurological de�cit in gonadally intact male rats subjected to transient middle cerebral artery occlusion (tMCAO). We used gonadally intact males experience tMCAO as a model of the effects of testosterone reduction during stroke. Moreover, �utamide (androgen receptor antagonist) and letrozole (aromatase inhibitor) were used to inhibit two potential mechanisms effects of testosterone.

Materials and Methods
All experiments were approved by Research Ethics Committees at Tehran University of Medical Sciences. All chemicals were obtained from Sigma-Aldrich (America), unless otherwise stated.

Animals and Treatments.
Experiments were carried out on 50 male Wistar rats weighing 280-320 g. e rats were maintained under controlled conditions with temperature at 22-24 ∘ C, relative humidity of 50-60% and a 12-hour lighting cycle and permitted ad libitum access to water and standard lab chow. e experimental animals were divided into �ve groups ( per group) and each rat had 60 minutes tMCAO. Experimental groups were gonadally intact male; castrated male (castrations were operated 7 days before tMCAO); intact male � 10 mg/kg/sc �utamide; intact male � 1 mg/kg/i.p letrozole; intact male � 10 mg �utamide � 1 mg letrozole. Flutamide and letrozole were injected 2 hours before tMCAO.

Determination of Serum Testosterone Concentration in
Intact and Castrated Rats. Blood samples were taken from animals 48 h before and 24 h aer tMCAO. Blood samples were centrifuged for serum separation and testosterone serum levels were determined by radioimmunoassay.

Transient Focal Cerebral
Ischemia. tMCAO was induced by intraluminal �lament method as previously described [40]. Brie�y, animals were anesthetized with ketamine (100 mg/kg, i.p) and xylazine (10 mg/kg, i.p.) Right common carotid artery (CCA) and the external carotid artery (ECA) were exposed and carefully separated from the vagus nerve. tMCAO was produced by advancing a silicone-coated nylon mono�lament through the ECA into the internal carotid artery (ICA) and then into the circle of Willis until a mild increase in resistance was felt. Such light resistance indicates that the tip of the nylon thread reached the origin of the MCA (18-20 mm from CCA bifurcation). en, �lament was �xed in place by fastening a silk suture around the ECA. Surgery was operated in 15 to 20 minutes. Sixty minutes later, the �lament was withdrawn to allow brain tissue reperfusion. Rats were then allowed to recover and were observed according to the experimental protocol.

Measurement of Brain Infarct and Edema Volumes.
Infarct volume evaluation was performed 24 h aer tMCAO. Animals were deeply anesthetized and decapitated. Subsequently, their brains were removed and sectioned coronally into consecutive 2 mm thick slices. e series of slices from each brain was soaked in 2% 2,3,5-triphenyltetrazolium chloride (TTC) solution and incubated for 30 minutes at 37 ∘ C in a water bath. e slices were then transferred to 10% buffered formalin. Normal tissue appears red, whereas infarct tissue appears colorless.
Both sides of the TTC-stained slices were scanned using a �atbed scanner (Scanjet, Hewlett-Packard, USA) connected to a computer. Infarct areas in each slice were measured using Image J soware (NIH). en, infarct volume of each brain was determined by integration of the infracted areas of the slices [44]. Following formula was applied to eliminate the effect of swelling on the infarct volume: corrected infarct volume = infarct size × contralateral hemisphere size/ipsilateral hemisphere size. Cerebral infarct volume was measured as a percentage of the total brain, cortex, and striatum. Edema volume was also calculated by the following formula: edema = (volume of right hemisphere − volume of le hemisphere)/volume of le hemisphere.
2.6. Statistical Analysis. SPSS 11.5 soware (SPSS, Chicago, Illinois) was used for the statistical calculations. One-way analysis of variance (for multigroup comparisons) was used for comparisons in infarct volume, brain edema, and testosterone serum levels followed by Bonferroni tests. Data on brain infarct volume and brain edema are reported as mean ± SD. Neurological de�cit and seizure activity were determined using the Kruskal-Wallis test and are presented as medians and interquartile ranges (25th and 75th percentiles). Differences were considered to be statistically signi�cant when .
Seizure behaviors are shown in Figure 5. At 4 h aer tMCAO the seizure activity was signi�cantly increased in the letrozole group (2 4 ± 96; ) only more than intact group ( 8 ± 78). No obvious seizure activity was observed in all experimental groups at 24 h aer the tMCAO (data not shown).

Discussion
Clinical and epidemiological studies indicate that male sex is risk factors for stroke in humans [9,10,12]. is evidence was evaluated by earlier studies that mimic the pathology of human stroke and they proved that quantity of brain tissue injury aer cerebral ischemia is greater in male versus female animals [9,39]. Moreover, for testing effects of testosterone on cerebral ischemia, male rodents were castrated in order to testosterone depletion. In this model, castrated rodents aer MCAO had lower infarct size than gonadally intact male animals, while testosterone administration to produce physiologic testosterone concentration in castrated rodents increases infarct size [17,28,30,45]. In contrast, recent studies displayed that androgens have dose-dependent effects in cerebral ischemia, thereby testosterone in low physiological range gives protection while in high physiological range displays deleterious effects during an episode of MCAO [45,46]. Whereas in real situation, plasma concentration of testosterone will decline slowly 24 h aer cerebral ischemia and it has no constant concentration [26,30,47]. erefore, to be close to the actual conditions of cerebral ischemia, we assessed the effect of testosterone via androgen receptor and aromatization to estrogen on cerebral ischemia by �utamide and letrozole pretreatment in gonadally intact rats that they subjected to tMCAO. So, it is suitable to focus on mechanisms of testosterone effects on stroke. Our results indicated that cortical infarct volume in castrated group was smaller than intact group, but this difference was not signi�cant. While on the contrary, previous studies displayed castration Brain edema (%) * * * F 3: Brain edema expressed as a percentage of the total brain volume in rats treated with castration, �utamide, letrozole, and �utamide in combination with letrozole (mean ± SD). * Signi�cant difference ( ) with Intact.
can signi�cantly reduces cortical infarct volume [28,45]. Flutamide pretreatment signi�cantly reduces cortical infarct volume compared to intact group and even castrated group. But other studies only indicated that �utamide reverses all testosterone effects (bene�cial and deleterious) on infarct volume [45]. Castration signi�cantly reduced striatal and total infarct volume as well as brain edema. Moreover, castration sig-ni�cantly improved neurological score. �lthough there was a similar tendency in cortical infarct volume of castrated rats, there was no signi�cant difference with intact group. In agreement with our results, other studies have demonstrated that castration and subsequent testosterone depletion led to protection or no difference in infarct volume and better neurologic score aer MCAO [17,30,45,48]. On the other hand, they displayed testosterone administration in castrated rodents exacerbates infarct volume and neurologic score in part through amplifying of glutamate neurotoxicity in neuronal cells [19,26,49]. erefore, it suggested that testosterone exposure could be harmful during the cerebral ischemia episode. Results of our study showed that �utamide signi�cantly improved histological damage of rat brain in all regions and even �utamide had better effect than castration on infarct volume. In addition, �utamide confers protection on the brain edema and the neurologic score and although it had no signi�cant difference with castrated group, but it was more effective than castration. On the contrary, most previous studies showed that testosterone confers deleterious effects to both adult castrates and gonadally intact rodents during stroke and these effects are abrogated by administration of the AR antagonist �utamide [45]. Also few recent studies indicated that maintaining plasma testosterone in low physiological range during MCAO induces protective effects in rodents and again this effect abolished by �utamide [19,45].
Similar in vivo studies, both protective (in low physiological concentrations) and deleterious (in high physiological concentrations) effects of testosterone have been showed in glial and neuronal cultures under harmful conditions (e.g., oxidative stress, excitotoxicity, serum deprivation, and amyloid ) and again these effects were abolished by �utamide [19, 33-35, 37, 49, 50]. However, our study indicated �utamide had protective effects on gonadally intact rats during stroke.
However, �utamide is commonly accounted pure antiandrogens without AR agonist activity [51,52]. One explanation for the protective effects of �utamide in our work is that �utamide blocked deleterious effects or mimicked and enhanced neuroprotective effects of testosterone. In agreement with this explanation some studies in neuronal and glial cultures observed that �utamide failed to abolish neuroprotective effect of testosterone [35,38,53]. Furthermore, some works propose that �utamide did not only fail to reverse the effects of testosterone but mimicked or behaved as partial AR agonist for them as well [35,38,53,54].
One interpretation of these data for our result is that �utamide is functioning as androgen agonist in activating cell signaling pathways that contributes to neuroprotection or as androgen antagonist in inhibiting cell signaling pathways that are active in deleterious effect of testosterone, or both of them. Another possible reason is that �utamide blocked AR and thereby increases available testosterone for conversion to estradiol by aromatase enzyme. Estradiol is known as a neuroprotective factor against cerebral ischemia and because aromatase is present in cerebral tissue and can convert testosterone into estradiol, an estrogen signaling pathway is reasonable [19,30,38,45]. Although in our study letrozole increased signi�cantly infarct volume only in striatum than intact group, it had similar tendency in cortex and total infarct volume as well as brain edema. On the other hand, letrozole signi�cantly increased seizure activity.
Brain aromatase is not only involved in the regulation of neuroendocrine events and behaviours connected with reproduction as previously thought, but also has roles in the reaction of brain tissue to damages and regulates synaptic plasticity, synaptic activity, and neurogenesis, via conversion of testosterone to estrogen [19,55].
Other studies indicated that cerebral aromatase expression increased aer brain injury in all damaged brain areas including, cortex, striatum, hippocampus, thalamus, and hypothalamus and exert neuroprotection effects via production of local estrogen [19,55,56]. On the other hand, wildtype rodents that were treated with aromatase inhibitors, as well as aromatase knockout (ArKO) rodents, showed more cortical and striatal damage in tMCAO model than wild-type animals treated with vehicle [19].
Also, an earlier study indicated that letrozole decreased estradiol levels, number of glutamic acid decarboxylase (GAD) positive cells, GAD expression as well as GABA production in hippocampal cultures, suggesting that aromatase by these effects in hippocampal neurons regulate synaptic activity [57]. According to our results, aromatase may play an important role in defense against overexcitation and seizure activity of neurons induced by glutamate aer cerebral ischemia.
Coadministration of letrozole with �utamide showed that letrozole has a potency to reverse the bene�cial effects of �utamide on histological damage and even make it worse than letrozole only level. One interpretation for effect of combination therapy is that letrozole ceases conversion testosterone to estrogen and thereby increases available testosterone to competing with �utamide for AR. On the other hand, the pernicious effect of letrozole could be related to seizures and not be related to testosterone levels. A previous study reported that early epileptic seizures worsen prognosis in human atherothrombotic stroke. is work stated that inhospital mortality was signi�cantly higher in stroke patients with early seizures [58]. ese occurred in the temporal window, where a penumbra of potentially salvageable brain tissue is believed to exist and this may determine a greater 6 ISRN Neurology area of de�nite cerebral infarction, which in turn would be associated with a poorer clinical course [58].

Conclusion
Contrary to the previous in vivo studies we have demonstrated that �utamide not only blocks deleterious effects of testosterone but also it can enhance neuroprotective effect of testosterone. Additional research is required to further describe the effects of �utamide on interaction between testosterone, aromatase, and AR during brain response to ischemia as well as determine which pathways are involved downstream of AR and aromatase. Clinically, our results suggest that �utamide and testosterone availability during cerebral ischemia may have bene�cial effects in men. Finally, this work opens new encouraging perspectives for the protection of the brain from the ischemic injury. A prospective randomized trial of �utamide may be justi�ed in humans.

Con�ict of �nterests
e authors declare that they have no con�ict of interests.