Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Neuroscience. Author manuscript; available in PMC 2010 August 1.
Published in final edited form as:
PMCID: PMC2905034



Isoflurane preconditioning neuroprotection in experimental stroke is male-specific. The role of androgens in the ischemic sensitivity of isoflurane preconditioned male brain and whether androgen effects are androgen receptor dependent were assessed. Male C57BL/6 mice were implanted with flutamide (androgen receptor antagonist), or castrated and implanted with testosterone, dihydrotestosterone, flutamide, letrozole (aromatase inhibitor), or vehicle 7–13 days before preconditioning. P450 estrogen aromatase wild-type and knockout mice were also evaluated. All mice were preconditioned for 4 h with 0% (sham preconditioning) or 1% isoflurane (isoflurane preconditioning) and recovered for 24 h. Mice then underwent 2 h of middle cerebral artery occlusion and were evaluated 22 h later for infarct volume. For neurobehavioral outcomes, sham and isoflurane preconditioned castrated male±dihydrotestosterone groups underwent 1 h of middle cerebral artery occlusion followed by 9 days of reperfusion. Isoflurane preconditioning neuroprotection relative to infarct volume outcomes were testosterone and dihydrotestosterone dose-specific and androgen receptor-dependent. Relative to long-term neurobehavioral outcomes, front paw sensorimotor function improved in isoflurane preconditioned mice regardless of androgen status while androgen replacement independently improved sensorimotor function. In contrast, isoflurane preconditioning improved cognitive function in castrates lacking endogenous androgens, but this improvement was absent in androgen replaced mice. Our findings suggest that androgen availability during isoflurane preconditioning may influence infarct volume and neurobehavioral outcomes in male mice following experimental stroke.

Keywords: androgens, testosterone, dihydrotestosterone, androgen receptor, neuroprotection, preconditioning

Prior brain exposure to minor insults, chemicals, or pharmacological agents can “precondition” or increase the brain’s tolerance to future, more injurious events. Anesthetics in particular may play an important and unique role as brain preconditioning agents because of their clinical use and relevance during the perioperative period for cardiovascular procedures and neurosurgical interventions. Perioperative stroke is a serious complication that can occur during or following procedures like carotid endarterectomy or coronary artery bypass grafting (Selim, 2007). Clinically, there is much interest in pharmacologically preconditioning the human brain with anesthetics and its impact on perioperative stroke, as anesthetic agents are known to affect ischemic outcome in experimental stroke (Kitano et al., 2007a). However, it is currently unclear how anesthetic preconditioning may influence perioperative stroke outcomes in men versus women as experimental brain studies examining anesthetic preconditioning and its neuroprotective mechanisms have used primarily male animals, have not indicated sex, or have not stratified results relative to sex (Wang et al., 2008b).

Few studies have specifically addressed sex differences in the preconditioned brain’s response to ischemia or other forms of brain injury (Wang et al., 2008b). Our prior work has shown that isoflurane preconditioning neuroprotection in experimental focal stroke is male-specific, with the neuroprotective benefits of isoflurane preconditioning being lost in female mice and present in ovariectomized mice lacking female sex steroids (Kitano et al., 2007b; Wang et al., 2008a). Even fewer studies have addressed the role of individual sex steroids in the brain’s response to preconditioning, anesthetic or otherwise, before ischemia or other forms of brain injury. We have previously shown that estradiol attenuates isoflurane preconditioning neuroprotection in ovariectomized females (Wang et al., 2008a). However, no work has been done exploring the role of androgens in any type of experimental brain preconditioning or clinically in perioperative stroke.

Testosterone and dihydrotestosterone (DHT) are the primary androgens seen in rodents and men. Testosterone can be converted via 5α-reductase to the more potent androgen and pure androgen receptor (AR) agonist, DHT or it can be aromatized to 17β-estradiol via P450 estrogen aromatase. The few available data with androgens and human stroke have focused on testosterone levels in men after stroke that were either unaffected (Taggart et al., 1980) or decreased, presumably as a stress response (Jeppesen et al., 1996). Although androgens have been reported in some studies as having harmful effects on ischemic rodent brain injury (Cheng et al., 2007; Hawk et al., 1998; Li et al., 2009; Uchida et al., 2009; Yang et al., 2002, 2005), several studies would suggest that androgens may not be universally detrimental in experimental cerebral ischemia (Hammond et al., 2001; Li et al., 2009; Pan et al., 2005; Toung et al., 1998; Uchida et al., 2009). Androgens may even be required for ischemic preconditioning cardioprotection in experimental myocardial ischemia models (Liu et al., 2006; Song et al., 2003; Tsang et al., 2007) and may mediate their effects via the AR (Liu et al., 2006; Tsang et al., 2007). Little work has been done evaluating androgens and the AR in preconditioned ischemic male brain.

Using a mouse model of isoflurane preconditioning and experimental stroke, we evaluated the role of testosterone and DHT in male-specific isoflurane preconditioning neuroprotection. We also determined if these androgens act via an AR-dependent mechanism in isoflurane preconditioned ischemic male brain.


Experiments were carried out in accordance with the National Institutes of Health guidelines for research animal care and approved by the Oregon Health and Science University Animal Care and Use Committee.

Experimental groups

Experiments were carried out on male C57BL/6 mice (Charles River Laboratories, Wilmington, MA, USA), 8–14 weeks of age and weighing 20–25 g. The following experimental groups underwent sham or isoflurane preconditioning and were analyzed for infarct volumes at 22 h reperfusion following 2 h of middle cerebral artery occlusion (MCAO) (Fig. 1): gonadally intact males ±5 mg flutamide (n=39); castrated males ±5 mg flutamide (n=37); castrated males+testosterone (1.5 or 5 mg) ±5 mg flutamide (n=58); castrated males+DHT (0.5 or 1.5 mg) ±5 mg flutamide (n=57); castrated males+1.5 mg testosterone+vehicle (n=22); and castrated males+1.5 mg testosterone +1 mg/kg/day letrozole (n=23). Infarct volumes at 22 h reperfusion after 2 h MCAO were also determined in sham and isoflurane preconditioned P450 estrogen aromatase wild-type (n=24) and knockout (n=24) male mice (8–14 weeks of age, 20–25 g). Based on infarct outcomes, the lowest neuroprotective DHT dose (0.5 mg) was used to assess neurobehavioral outcomes. Sham and isoflurane preconditioning castrated males ±0.5 mg DHT (n=75) subjected to 1 h of MCAO were therefore evaluated for long-term neurobehavioral outcomes during the 9 day period following MCAO. All mice were maintained on a 12/12-h light/dark cycles and permitted ad libitum access to food and water.

Fig. 1
Schematic overview of the time line and experimental design for castration, hormone or drug administrations, preconditioning and ischemia. Abbreviations: ArKO, P450 estrogen aromatase knockout; ArWT, P450 estrogen aromatase wild-type; DHT, dihydrotestosterone; ...

Hormone and drug treatments

Under isoflurane (Hospira, Lake Forest, IL, USA) anesthesia (induction 4%; maintenance 1.5%), male mice were castrated 7–8 days before preconditioning to allow endogenous serum sex steroid hormone levels to decrease and exogenous serum sex steroid hormone levels in implanted mice to come to steady state levels. At the time of castration, select groups of castrated mice were implanted s.c. with 21-day release pellets (Innovative Research of America, Sarasota, FL, USA) containing testosterone (1.5 or 5 mg), DHT (0.5 or 1.5 mg), and/or the AR antagonist, flutamide (5 mg). Selected groups of gonadally intact male mice were also treated with flutamide (5 mg). We have previously confirmed the antagonist properties of flutamide at equimolar or higher doses as concurrently administered testosterone doses in ischemic male brain (Uchida et al., 2009).

In experimentally naive gonadally intact C57BL/6 males, we have previously reported a range of 0.1–21.7 ng/mL for total testosterone, 0.0–55.1 pg/mL for free testosterone and 0.0–3.0 ng/mL for DHT (Uchida et al., 2009). Serum androgen levels following experimental stroke in gonadally intact males, untreated castrates, and castrates treated with the testosterone (1.5, 5 mg) or DHT (0.5, 1.5 mg) implant doses used in this study have been previously characterized by our group (Uchida et al., 2009). In testosterone treated castrates, we have shown that total and free testosterone levels increased relative to testosterone dose and were within the broad physiological range for naive gonadally intact male mice described above (Uchida et al., 2009). We have also shown that DHT treatment in castrates had no effect on total or free testosterone levels as compared to untreated castrates but increased serum DHT levels in a dose-dependent manner, with values within the physiological range described above for naive gonadally intact males (Uchida et al., 2009).

Letrozole (1 mg/kg/day; gift from Novartis Pharma AG, Basel, Switzerland), an aromatase inhibitor that is able to cross the blood-brain barrier (Hume and Wynne-Edwards, 2006), and its associated vehicle (propylene glycol; SIGMA, St. Louis, MO, USA) were dispensed via s.c. implantation of osmotic pumps (Model 1002, Alzet Osmotic Pumps, Durect Corporation, Cupertino, CA, USA) concurrently with castration and implantation of 1.5 mg testosterone pellets in male C57BL/6 mice 13 days before preconditioning. The letrozole dose selected and the timing of its administration prior to preconditioning for the proposed studies is based on biologically effective doses used in rodent studies involving CNS effects (Hume and Wynne-Edwards, 2006; Park et al., 2009).

P450 estrogen aromatase knockout (ArKO) mice

Generation of P450 estrogen aromatase knockout (ArKO) mice is as previously described (Fisher et al., 1998). This strain originated on a C57BL/6J; 129SvEv background and has been backcrossed to C57BL/6J for at least 10 generations. In ArKO mice, exon IX of the Cyp 19 gene is disrupted by insertion of a neo cassette using homologous recombination. Homozygous ArKO mice produce an abnormal transcript that generates a nonfunctional protein. Consequently, estrogen aromatase activity is lacking in all cell types and estrogen synthesis is blocked. Breeding colony was maintained utilizing heterozygous breeding trios (two females, one male). Genotyping was done by polymerase chain reaction (PCR) amplification of tail DNA as previously described (Liu et al., 2008).

Isoflurane preconditioning model

Isoflurane preconditioning model is as described previously (Kitano et al., 2007b; Wang et al., 2008a). Mice were placed for 4 h in an air-tight, temperature-controlled chamber flushed with 1.0% isoflurane (isoflurane preconditioning) or 0% isoflurane (sham preconditioning) in oxygen-enriched air. We have previously shown that physiological systemic variables (glucose, pH, partial arterial pressure carbondioxide [PaCO2]) evaluated after 1 and 4 h of isoflurane preconditioning showed no major perturbations imposed by isoflurane preconditioning other than mild hypotension and partial arterial pressure oxygen (PaO2) values consistent with animals exposed to oxygen-enriched air (Kitano et al., 2007b). Twenty-four hours after preconditioning, mice underwent transient focal cerebral ischemia.

Transient focal cerebral ischemia

Preconditioned male mice were subjected to 1 or 2 h of reversible MCAO via the intraluminal filament technique as previously described (Kitano et al., 2007b; Wang et al., 2008a). Mice were anesthetized with isoflurane (induction 4%; maintenance 1.5%) during surgery until unilateral MCAO was achieved and then allowed to awaken during ischemia. Towards the end of ischemia, mice were briefly reanesthetized with isoflurane and reperfusion was initiated by intraluminal filament withdrawal. Rectal temperatures in anesthetized mice were monitored and maintained at 36.4±0.7 °C throughout MCAO surgery and during induction of reperfusion. Anesthetized mice were warmed as needed using an infrared heating lamp (115 VAC, 50/60 Hz, Cole-Parmer Instrument Company, Vernon Hills, IL, USA). Cortical blood flow was monitored throughout MCAO surgery, initial induction of MCAO, end-ischemia, and initiation of reperfusion by laser Doppler flowmetry (LDF; model MBF3D, Moor Instruments Ltd., Oxford, England). The effectiveness of vascular occlusion was confirmed by sustained reduction in LDF signal while restoration of blood flow following release of vascular occlusion was assessed by increases in LDF signal towards baseline levels. Animals were excluded if intra-ischemic LDF was greater than 25% pre-ischemic baseline. All animals undergoing cerebral ischemia regardless of preconditioning status had comparable exposure times (26.5±0.4 min, n=316) as well as dose relative to isoflurane delivered during MCAO and reperfusion.

For infarct volume outcomes, animals were euthanized after 22 h of reperfusion, brains removed, and blood obtained for hormone analysis. Infarct volume was estimated using 2,3,5-triphe-nyltetrazolium chloride (SIGMA) staining of 2-mm thick coronal sections (five slices total) (Bederson et al., 1986). Both sides of each stained coronal slice were photographed via a digital camera (see Fig. 2A for examples of coronal images) and then evaluated by digital image analysis (SigmaScan Pro, Jandel, San Rafael, CA, USA). Infarcted area was integrated across sections, and volume expressed as % of contralateral structure (cortex, caudate putamen). Animals used for neurobehavioral evaluations were euthanized after 9 days of reperfusion.

Fig. 2
Cortical and caudate putamen infarction volumes (% contralateral structure) determined by 2,3,5-triphenyltetrazolium chloride (TTC) staining in male C57BL/6 mice preconditioned for 4 h with 0% (Sham PC) or 1% isoflurane (IsoPC). Preconditioning occurred ...

Femoral arterial catheters were placed in separate groups of sham and isoflurane preconditioned castrated males (n=9) and castrates treated with 0.5 mg DHT, in the presence or absence of 5 mg flutamide (n=20) subjected to 2 h of MCAO in order to determine mean arterial blood pressure, blood gases (pH, PaO2, PaCO2), and blood glucose values at 1 h of ischemia and at 10 min of reperfusion. Animals were anesthetized with 1.5% isoflurane throughout ischemia and reperfusion. Mice were euthanized after the final blood sample was taken 10 min into the reperfusion period.

Hormone assay

Serum hormone levels (total and free testosterone, DHT, 17β-estradiol) were measured in duplicate by radioimmunoassay (Diagnostic Products Corp., Los Angeles, CA, USA). The lower limits of detection for total and free testosterone, DHT, and estradiol were 0.01 ng/mL, 0.5 pg/mL, 0.01 ng/mL, and 10 pg/mL, respectively. When calculating mean hormone levels per treatment group, a value of zero was assigned to hormone values less than the lower limits of detection.

Neurobehavioral evaluations

We selected neurobehavioral tests that have consistently provided reliable outcome evaluations across experimental focal cerebral ischemia studies in mice in our laboratories (Craft et al., 2006; Li et al., 2004; Uchida et al., 2009). All mice undergoing behavioral testing were single-housed on a 12/12-h light/dark cycle. All neurobehavioral assessments were performed during the second half of the light cycle between 1200 and 1800 h. The observer who performed all neurobehavioral assessments was blinded to treatment group. All equipment was cleaned with 10% ethanol between trials.

Neurological deficit score

A neurological deficit score was recorded in each mouse 1 and 3 days after MCAO (Day 1 and Day 3) as previously described (Li et al., 2004; Uchida et al., 2009). The graded scoring systems ranged from 0 to 2 to 0 to 5 depending on the behavior assessed, with 0 indicating no deficit and 2–5 indicating the most impaired. The behaviors assessed included consciousness (0–3), interaction (0–2), eye appearance (0–2), breathing (0–2), food/water intake (0–2), ability to grab wire top (0–2), motor function (0–5), and activity (0–2). The final neurological deficit score was a summation of these graded scoring systems and ranged from 0 (least impaired) to 20 (most impaired).

Cylinder test

We performed the cylinder test (Schallert et al., 2000; Uchida et al., 2009) to analyze forelimb use bias. Each mouse was placed in a transparent cylinder measuring 9 cm in diameter and 15 cm in height. The cylinder was wide enough for the mouse to move freely and small enough to encourage rearing behavior. Four separate video cameras were placed around the cylinder at 90° intervals in order to record rearing behavior from all angles. We recorded a maximum of two paw touches for one rearing event. If the mouse touched the cylinder wall with the same paw two times, we designated these motions as “independent.” If the mouse touched the cylinder wall the first time with one paw and a second time with the other paw, we designated these motions as “independent” and “both,” respectively. A total of 20 forelimb touches was recorded during the 10-min test and was found to be sufficient to determine bias without habituating the mouse to the apparatus. Mice with fewer than 20 forelimb touches in this time period were excluded from neurobehavioral analysis. The final score was calculated as the percentage of total touches that used the impaired forelimb. Baseline paw preference was assessed one day before MCAO just prior to preconditioning. Subsequent assessments were made at 3 and 7 days after MCAO.

Locomotor activity in the open field

To evaluate spontaneous locomotor activity, we used the open field protocol (Craft et al., 2006). Mice were placed individually into a 41 cm (W) ×41 cm (D) ×38 cm (H) plastic enclosure equipped with video cameras mounted above to record movement in four arenas simultaneously. The video was analyzed off-line using Noldus software (Ethovision 2.3, Noldus, Leesburg, VA, USA). Total distance traveled and mean velocity were assessed at baseline, one day prior to MCAO and 5 days following MCAO.

Novel object recognition test

Mice were placed individually into a 41 cm (W) ×41 cm (D) ×38 cm (H) plastic enclosure and allowed to habituate to the arena before novel object recognition testing. During the sample session 6 days after MCAO, two identical objects were placed in opposite corners of the arena, approximately 1 in. from the wall. Time spent exploring each object during the sample session was hand scored with stopwatches and each mouse was removed from the arena after accumulating 38 s of exploration time on either of the sample objects (maximum of 10 min total allotted), in agreement with previous studies (Hammond et al., 2004). After a 24 h delay, mice were again placed in the arena for a test session (day 7 after MCAO), with one object replaced with a novel object (location of novel object randomized). A 5 min test session was performed and the time exploring each object was recorded. Novel object preference ratio is the time spent exploring the new object divided by the total exploration time during the test session. All objects used in this study were characterized previously in our laboratory to ensure that mice prefer each object equally.


Values are expressed as mean±standard error of the mean (SEM). Differences in infarct volumes in each isoflurane preconditioning group as compared to its corresponding sham preconditioning group were determined with a Student’s t-test in order to examine the effect of isoflurane preconditioning relative to sham preconditioning under different experimental conditions (castration, hormone or drug administration, presence or absence of aromatase). Differences in serum hormone levels between male P450 estrogen aromatase wild-type (ArWT) and ArKO mice and between testosterone replaced castrated males treated with letrozole or vehicle were determined by a Student’s t-test. Mean arterial blood pressure, blood gases (pH, PaO2, PaCO2), and blood glucose were subjected to two-way analysis of variance (ANOVA) with post hoc Newman–Keuls test. Behavioral outcomes for neurological deficit scores and the paw preference test were analyzed by one-way ANOVA repeated measurement with post hoc Newman–Keuls test. Behavioral outcomes for the novel object recognition test were analyzed by one-way ANOVA with post hoc Newman–Keuls test. Statistical significance was P<0.05. Statistical analyses were performed using SigmaStat Statistical Software, Version 3.1 (SPSS, Inc., Chicago, IL, USA).


Mortality and exclusions in infarct volume outcome studies

Overall mortality was 13 mice out of a total of 284 mice, with mortality ranging from 0 to 2 mice within the experimental groups. Overall number of mice excluded due to intra-ischemic LDF greater than 25% pre-ischemic baseline was 13 mice out of a total of 284 mice, with exclusions ranging from 0 to 3 mice within the experimental groups. At initiation of reperfusion, all experimental groups reperfused to greater than 50% pre-ischemic baseline (data not shown).

Infarct volumes in preconditioned males, castrates, and androgen-replaced castrates

Isoflurane preconditioning significantly decreased infarct volumes in intact male mice as compared to sham preconditioned males (Fig. 2). In contrast, reduction of endogenous androgen levels via castration prevented the decrease in infarct volumes in isoflurane versus sham preconditioned mice (Fig. 2). Isoflurane preconditioning neuroprotection was restored in castrates treated with low dose (1.5 mg) testosterone (Fig. 3A) or (0.5 mg) DHT (Fig. 3B). In castrates treated with high dose (5 mg) testosterone (Fig. 3A), isoflurane preconditioning neuroprotection was only reinstated in the caudate putamen. Isoflurane preconditioning neuroprotection was not re-established in castrates treated with high dose (1.5 mg) DHT (Fig. 3B).

Fig. 3
Cortical and caudate putamen infarction volumes (% contralateral structure) determined by 2,3,5-triphenyltetrazolium chloride staining in male C57BL/6 mice preconditioned for 4 h with 0% (Sham PC) or 1% isoflurane (IsoPC). Preconditioning occurred 24 ...

Infarct volumes in flutamide-treated groups and in mice with aromatase deletion or inhibition

The protective effects of isoflurane preconditioning on infarct volumes in intact male mice were blocked when males were treated with flutamide, an AR antagonist (Fig. 4A). However, when endogenous androgen levels were reduced by castration, flutamide had no effect on infarct volumes regardless of preconditioning status (Fig. 4A). The protective effects of isoflurane preconditioning in castrates treated with low-dose testosterone (1.5 mg) or DHT (0.5 mg) were attenuated by concurrent exposure to flutamide (Fig. 4B). The protective effects of isoflurane preconditioning on infarct volumes were not altered in gonadally intact males by genetic deletion of aromatase versus wild-type (Fig. 5A) or by pharmacologic inhibition of aromatase with letrozole versus vehicle in castrates treated with low dose (1.5 mg) testosterone (Fig. 5B).

Fig. 4
Cortical and caudate putamen infarction volumes (% contralateral structure) determined by 2,3,5-triphenyltetrazolium chloride staining in male C57BL/6 mice preconditioned for 4 h with 0% (Sham PC) or 1% isoflurane (IsoPC). Preconditioning occurred 24 ...
Fig. 5
Cortical and caudate putamen infarction volumes (% contralateral structure) determined by 2,3,5-triphenyltetrazolium chloride staining in male mice preconditioned for 4 h with 0% (Sham PC) or 1% isoflurane (IsoPC). Preconditioning occurred 24 h before ...

Baseline and intra-ischemic arterial blood pressure and blood gases

Overall mortality was three mice out of a total of 29 mice, with mortality ranging from 0 to 1 mouse within the experimental groups. No mice were excluded due to intra-ischemic LDF greater than 25% pre-ischemic baseline. At initiation of reperfusion, all experimental groups reperfused to greater than 50% pre-ischemic baseline (data not shown). Mean arterial blood pressure, blood gases (pH, PaO2, PaCO2), and blood glucose were comparable between castrates, castrates treated with 0.5 mg DHT, and castrates treated with 0.5 mg DHT and 5 mg flutamide sham and isoflurane preconditioning groups at each time point evaluated (Suppl. Table 1). There were no differences between ischemic and reperfusion values within each experimental group (Suppl. Table 1).

Hormone levels

Regardless of preconditioning group, there were no differences between male ArWT and ArKO mice in free (ArWT, 0.5±1.0 pg/mL, n=18; ArKO, 0.6±0.1 pg/mL, n=20) and total (ArWT, 0.5±0.1 ng/mL, n=21; ArKO, 0.5±0.1 ng/mL, n=21) testosterone and DHT (ArWT, 0.2±0.0 mg/mL, n=21; ArKO, 0.3±0.1 mg/mL, n=20) levels. However, estradiol levels were (P<0.05) lower in male ArKO mice (16±2 pg/mL, n=10) as compared to male ArWT mice (28±2 pg/mL, n=15). Estradiol levels were also (P<0.05) lower in testosterone-replaced (1.5 mg) castrates treated with letrozole (3±2 pg/mL, n=17) versus vehicle (14±3 pg/mL, n=15).

Neurobehavioral testing

Neurobehavioral testing was performed over a 9 day period following 1 h of MCAO in sham and isoflurane preconditioned castrated C57BL/6 mice and sham and isoflurane preconditioned castrated C57BL/6 males treated with 0.5 mg DHT. Mortality in each experimental group was as follows: sham preconditioned castrates, 12% (2/17 in total); isoflurane preconditioned castrates, 29% (6/21 in total); sham preconditioned castrates treated with 0.5 mg DHT, 21% (4/19 in total); and isoflurane preconditioned castrates treated with 0.5 mg DHT, 17% (3/18 in total). Only two mice, one from each sham and isoflurane preconditioned groups of castrates treated with 0.5 mg DHT, were excluded as neurobehavioral testing outliers (unable to perform tasks).

Neurological deficit score was assessed at 1 and 3 days after MCAO as an indicator of general health. There were no differences in neurological deficit scores between the experimental groups on each day tested (Fig. 6A). All experimental groups showed similar levels of impairment 1 and 3 days after MCAO, with declining neurological deficit scores over time. Motor dysfunction was assessed on the affected, contralateral side (left paw/forelimb) using the cylinder test. No asymmetries were observed in evaluations done one day before MCAO, just prior to preconditioning (Fig. 6B). At 3 and 7 days following MCAO, isoflurane preconditioning minimized contralateral forelimb impairment regardless of whether castrates were untreated or treated with DHT (Fig. 4B). DHT also reduced forelimb asymmetry at 3 and 7 days following MCAO regardless of preconditioning status (Fig. 6B).

Fig. 6
Neurobehavioral assessments in castrated (CAST) male C57BL/6 mice preconditioned for 4 h with 0% (Sham PC) or 1% isoflurane (IsoPC). Preconditioning occurred 24 h before 1 h of MCAO followed by 9 days of reperfusion. A 0.5 mg s.c. dihydrotestosterone ...

The open field test was performed 1 day before (Pre) and 5 days after MCAO as a control for recovery of motor function, in order to avoid this possible confounding factor during the novel object recognition test. No significant change in total distance moved or velocity of movement was observed during the 30-min open field test, regardless of treatment (data not shown). To determine the effect of MCAO on cognitive function using hippocampal-dependent tasks, the novel object recognition test was performed 6–7 days after MCAO. Only mice with an intact memory will recognize and preferentially explore a new or novel object, thus showing an increased novel object preference ratio. In castrated mice, isoflurane preconditioning prevented while DHT had no effect on the loss of novel object recognition and preference after MCAO observed in sham preconditioned castrates (Fig. 6C). However, the protective effect of isoflurane preconditioning on novel object recognition and preference was blocked in castrates treated with DHT (Fig. 6C).


This study demonstrates four important findings. First, the protective effects of isoflurane preconditioning on infarct volume in ischemic male brain are androgen-dependent. Second, isoflurane preconditioning neuroprotection relative to infarct volume outcomes is testosterone and DHT dose-specific. Third, the effects of testosterone and DHT on infarct outcomes in isoflurane versus sham preconditioned ischemic male brain are likely mediated through the AR. Last, isoflurane preconditioning shows a complex pattern of protection against long-term neurobehavioral deficits. Isoflurane preconditioning improved long-term sensorimotor outcome regardless of androgen status. Similarly, androgen replacement improved sensorimotor outcome regardless of preconditioning status. In contrast, isoflurane preconditioned mice in the absence of endogenous androgens showed improved cognitive outcome that was prevented by androgen replacement.

We have previously reported that neuroprotection due to isoflurane preconditioning is male-specific in ischemic brain (Kitano et al., 2007b). The current study demonstrates that castration ameliorated the protective response to isoflurane preconditioning in ischemic male brain, suggesting that androgens may be important to the male brain’s response to isoflurane preconditioning. Because androgen levels are not constant during life and can begin to decline during the middle years of life (andropause), it seems likely that androgen loss during aging would have a similar effect on the aging male brain’s response to isoflurane preconditioning as castration did in the young male brain. Therefore, future studies will need to address the role of androgens in isoflurane preconditioning neuroprotection in aging ischemic male brain as age can increase perioperative stroke risk following procedures like carotid endarterectomy (Bond et al., 2005; Schaller, 2007).

A potential difficulty in interpreting results from experiments using intact male mice is that endogenous androgen levels can be quite variable depending on time of day and environmental stresses as evidenced by the wide range of androgen values reported for mice (Overpeck et al., 1978; Quimby, 1999; vom Saal et al., 1994). Low endogenous androgen levels may therefore explain why in this study we failed to observe the protective effect of castration described in other rodent ischemic stroke models (Cheng et al., 2007; Hawk et al., 1998; Yang et al., 2002) on infarct volume outcomes in sham preconditioned intact versus castrated male mice. Furthermore, our findings indicate that the neuroprotective response to isoflurane preconditioning is androgen dose-dependent, with lower doses restoring and higher doses failing to fully re-establish isoflurane preconditioning neuroprotection in ischemic brain. Thus, androgens may have a complex dose-response relationship in preconditioned (anesthetic or otherwise) ischemic brain.

The findings from our experiments with DHT (a non-aromatizable direct AR agonist), flutamide (anti-androgen with high AR specificity and ability to cross the blood-brain barrier) (Neri et al., 1972; Singh et al., 2000), and P450 estrogen aromatase deletion or inhibition with letrozole suggest that the effects of androgens on infarct outcomes in isoflurane preconditioned brains are mediated via the AR. The results from these experiments also suggest that testosterone is predominantly binding either directly or indirectly through its conversion via 5α-reductase to DHT to the AR in preconditioned ischemic brain rather than being chiefly aromatized to 17β-estradiol by P450 estrogen aromatase. Since the AR was implicated by our results in mediating the effects of androgens in isoflurane preconditioned ischemic brain, we evaluated physiological parameters during ischemia and early reperfusion in preconditioned castrates and castrates treated with 0.5 mg DHT in the presence or absence of 5 mg flutamide. Our findings relative to infarct volume are not explained by differences in physiological parameters among treatment groups involving AR manipulations.

This study did not examine which mechanisms may be involved downstream of ARs such as transcriptional regulation of target genes through AR binding. Several possible target genes have already been identified that are induced by DHT in ischemic cortex and that could alter cell death outcomes through their involvement with inflammation, cell signaling, and blood-brain barrier regulation (Cheng et al., 2007). AR-independent mechanisms might also explain the observed androgen effects on ischemic outcomes in isoflurane preconditioned male brain. For example, we have shown that isoflurane preconditioning alone in males enhances Akt activation, a non-genomic neuronal survival-signaling pathway (Brunet et al., 2001; Dudek et al., 1997), but does not alter Akt activation in females at 24 h following preconditioning (Kitano et al., 2007b). Current literature suggests that testosterone enhances Akt signaling in heart (Bai et al., 2005). Therefore, it is possible that androgens may be beneficial in isoflurane preconditioned ischemic brain through effects on Akt activation. Regional differences in the mechanisms downstream or independent of ARs through which androgens exert their dose-specific effects in preconditioned ischemic brain may also explain why high dose testosterone restored isoflurane preconditioning in ischemic caudate putamen but not in cortex. Future studies from our laboratories will be done to determine which AR-dependent and independent mechanisms underlying the effects of androgens on ischemic outcomes in isoflurane preconditioned male brain are involved.

The lack of success in translating promising experimental rodent data into clinical trials has led to guidelines recommending evaluation of ischemia-induced functional deficits in addition to histological outcomes (Fisher et al., 2009). Therefore, we used several behavioral tasks to assess functional outcome during the first week following MCAO to extend our initial finding that isoflurane preconditioning protects the male brain in an androgen-dependent manner. For these longer term experiments, we used a significantly less severe insult (60 versus 120 min ischemia) in order to improve survival and recovery of gross sensorimotor deficits as cognitive function tests would be complicated by alterations in gross motor activity. Our data using a neurological deficit score to evaluate general sensorimotor performance and overall health did not show differences between treatment groups. We evaluated contralateral forelimb impairment using the cylinder test because we have previously demonstrated that this task detects impairment for several weeks after MCAO in mice (Li et al., 2004; Uchida et al., 2009). We observed that isoflurane preconditioning resulted in a decreased sensorimotor deficit regardless of androgen status, being equally effective in castrates and DHT replaced mice. Similarly, low dose DHT replacement in sham preconditioned mice improved long term sensorimotor outcome as measured by paw preference. Therefore, isoflurane preconditioning or low dose DHT appears to improve sensorimotor deficits at early and later time points following cerebral ischemia. We conclude from these data that subtle sensorimotor testing provides insight into “neuroprotection” that is not apparent in histological analysis of infarct volume following MCAO.

It is becoming increasingly apparent that studying cognitive and sensorimotor deficits following cerebral ischemia provides additional insight into outcomes and efficacy of experimental treatments. We used the novel object recognition test to assess long-term cognitive outcome. In the present study, isoflurane preconditioning prevented the decline in memory retention observed following MCAO in castrated male mice. Our findings that isoflurane preconditioning improves post-ischemic cognitive function are in agreement with isoflurane preconditioning studies in rat using the same cognitive test (McAuliffe et al., 2009). However, two studies have observed no effect of isoflurane preconditioning on post-ischemic cognitive outcomes when other tests like the Morris water maze, Y maze, and social recognition (McAuliffe et al., 2007; Zhao et al., 2007) were used to evaluate cognitive function. Interestingly, we observed that low dose DHT blocked the protective effect of isoflurane preconditioning on novel object recognition and preference. This finding seems at odds with the observation that isoflurane preconditioned DHT-replaced castrates showed reduced infarct volumes. However, infarct volumes may not necessarily predict long-term functional outcomes as infarct volume determinations only discriminate between live and dead neurons but does not provide information on the functional status of the remaining neurons. Distribution of regional cell death may also have a greater impact on neurobehavioral functional outcomes than infarct size (DeVries et al., 2001), indicating the need to perform more detailed histopathology in the future to better correlate histological and functional outcomes in isoflurane preconditioned brain following ischemia. In addition, this finding is somewhat surprising in conjunction with the improved sensitormotor function observed in these experiments, indicating complex interactions between androgen signaling and isoflurane preconditioning during the 9 day period following transient focal cerebral ischemia.


In summary, we have demonstrated that male-specific ischemic outcomes in isoflurane preconditioned brain are androgen dose- and AR-dependent. Few studies have examined androgens as mediators in the ischemic response to anesthetic or other types of brain preconditioning. More research is needed to further delineate the role of androgens and ARs in the brain’s response to anesthetic preconditioning relative to histological versus neurobehavioral outcomes as well as determine which pathways are involved downstream of ARs or independently of ARs. Clinically, our findings suggest that isoflurane anesthesia and androgen availability during “at-risk” cardiovascular surgical procedures may influence perioperative stroke outcomes in men.

Supplementary Material

Table 1


The authors thank Dr. Orhan Oz, Department of Radiology, University of Texas Southwestern Medical Center, for providing breeder pairs heterozygous for the deletion of the gene for P450 estrogen aromatase. The authors would also like to thank Ms. Heather Hoem for her assistance with all manuscript figures and Dr. Rochelle Fu, Department of Public Health and Preventative Medicine, Oregon Health and Science University, for her assistance with statistical analyses. They would also like to acknowledge Dr. Ling Zhang’s parent affiliation of the Department of Neurology at the Affiliated Drum Tower Hospital of Nanjing University Medical School, Nanjing, P R China. This work was supported by grants from NIH/NINDS (NS49210, NS33668, NS20020), from NIH/NINR (NR03521), and from the OHSU Medical Research Foundation (Seed Grant).


analysis of variance
androgen receptor
aromatase knockout
aromatase wild-type
laser Doppler flowmetry
middle cerebral artery occlusion
partial arterial pressure carbon dioxide
polymerase chain reaction
partial arterial pressure oxygen
standard error of the mean


Supplementary data

Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.neuroscience.2010.05.038.


  • Bai CX, Kurokawa J, Tamagawa M, Nakaya H, Furukawa T. Nontranscriptional regulation of cardiac repolarization currents by testosterone. Circulation. 2005;112:1701–1710. [PubMed]
  • Bederson JB, Pitts LH, Germano SM, Nishimura MC, Davis RL, Bartkowski HM. Evaluation of 2,3,5-triphenyltetrazolium chloride as a stain for detection and quantification of experimental cerebral infarction in rats. Stroke. 1986;17:1304–1308. [PubMed]
  • Bond R, Rerkasem K, Cuffe R, Rothwell PM. A systematic review of the associations between age and sex and the operative risks of carotid endarterectomy. Cerebrovasc Dis. 2005;20:69–77. [PubMed]
  • Brunet A, Datta SR, Greenberg ME. Transcription-dependent and -independent control of neuronal survival by the PI3K-Akt signaling pathway. Curr Opin Neurobiol. 2001;11:297–305. [PubMed]
  • Cheng J, Alkayed NJ, Hurn PD. Deleterious effects of dihydrotestosterone on cerebral ischemic injury. J Cereb Blood Flow Metab. 2007;27:1553–1562. [PMC free article] [PubMed]
  • Craft TK, Zhang N, Glasper ER, Hurn PD, Devries AC. Neonatal factors influence adult stroke outcome. Psychoneuroendocrinology. 2006;31:601–613. [PubMed]
  • DeVries AC, Nelson RJ, Traystman RJ, Hurn PD. Cognitive and behavioral assessment in experimental stroke research: will it prove useful? Neurosci Biobehav Rev. 2001;25:325–342. [PubMed]
  • Dudek H, Datta SR, Franke TF, Birnbaum MJ, Yao R, Cooper GM, Segal RA, Kaplan DR, Greenberg ME. Regulation of neuronal survival by the serine–threonine protein kinase Akt. Science. 1997;275:661–665. [PubMed]
  • Fisher CR, Graves KH, Farlow AF, Simpson ER. Characterization of mice deficient in aromatase (ARKO) because of targeted disruption of the cyp19 gene. Proc Natl Acad Sci U S A. 1998;95:6965–6970. [PubMed]
  • Fisher M, Feuerstein G, Howells DW, Hurn PD, Kent TA, Savitz SI, Lo EH. STAIR Group. Update of the stroke therapy academic industry roundtable preclinical recommendations. Stroke. 2009;40:2244–2250. [PMC free article] [PubMed]
  • Hammond J, Le Q, Goodyer C, Gelfand M, Trifiro M, LeBlanc A. Testosterone-mediated neuroprotection through the androgen receptor in human primary neurons. J Neurochem. 2001;77:1319–1326. [PubMed]
  • Hammond RS, Tull LE, Stackman RW. On the delay-dependent involvement of the hippocampus in object recognition memory. Neurobiol Learn Mem. 2004;82:26–34. [PubMed]
  • Hawk T, Zhang YQ, Rajakumar G, Day AL, Simpkins JW. Testosterone increases and estradiol decreases middle cerebral artery occlusion lesion size in male rats. Brain Res. 1998;796:296–298. [PubMed]
  • Hume JM, Wynne-Edwards KE. Paternal responsiveness in biparental dwarf hamsters (Phodopus campbelli) does not require estradiol. Horm Behav. 2006;49:538–544. [PubMed]
  • Jeppesen LL, Jorgensen HS, Nakayama H, Raaschou HO, Olsen TS, Winther K. Decreased serum testosterone in men with acute ischemic stroke. Arterioscler Thromb Vasc Biol. 1996;16:749–754. [PubMed]
  • Kitano H, Kirsch JR, Hurn PD, Murphy SJ. Inhalational anesthetics as neuroprotectants or chemical preconditioning agents in ischemic brain. J Cereb Blood Flow Metab. 2007a;27:1108–1128. [PMC free article] [PubMed]
  • Kitano H, Young JM, Cheng J, Wang L, Hurn PD, Murphy SJ. Gender-specific response to isoflurane preconditioning in focal cerebral ischemia. J Cereb Blood Flow Metab. 2007b;27:1377–1386. [PMC free article] [PubMed]
  • Li X, Blizzard KK, Zhu Z, DeVries AC, Hurn PD, McCullough LD. Chronic behavioral testing after focal ischemia in the mouse: functional recovery and the effects of gender. Exp Neurol. 2004;187:94–104. [PubMed]
  • Li Z, Cui S, Zhang Z, Zhou R, Ge Y, Sokabe M, Chen L. DHEA-neuroprotection and -neurotoxicity after transient cerebral ischemia in rats. J Cereb Blood Flow Metab. 2009;29:287–296. [PubMed]
  • Liu J, Tsang S, Wong TM. Testosterone is required for delayed cardioprotection and enhanced heat shock protein 70 expression induced by preconditioning. Endocrinology. 2006;147:4569–4577. [PubMed]
  • Liu M, Oyarzabal E, Yang R, Murphy SJ, Hurn PD. A novel method for assessing sex-specific and genotype-specific response to injury in astrocyte culture. J Neurosci Methods. 2008;171:214–217. [PMC free article] [PubMed]
  • McAuliffe JJ, Joseph B, Vorhees CV. Isoflurane-delayed preconditioning reduces immediate mortality and improves striatal function in adult mice after neonatal hypoxia-ischemia. Anesth Analg. 2007;104:1066–1077. [PubMed]
  • McAuliffe JJ, Loepke AW, Miles L, Joseph B, Hughes E, Vorhees CV. Desflurane, isoflurane, and sevoflurane provide limited neuroprotection against neonatal hypoxia-ischemia in a delayed preconditioning paradigm. Anesthesiology. 2009;111:533–546. [PubMed]
  • Neri R, Florance K, Koziol P, Van Cleae S. A biological profile of a nonsteroidal antiandrogen, SCH 13521 (4′nitro-3′trifluorom-ethyl-ispbutyranilide) Endocrinology. 1972;91:427–437. [PubMed]
  • Overpeck JG, Colson SH, Hohmann JR, Applestine MS, Reilly JF. Concentrations of circulating steroids in normal prepubertal and adult male and female humans, chimpanzees, rhesus monkeys, rats, mice, and hamsters: a literature survey. J Toxicol Environ Health. 1978;4:785–803. [PubMed]
  • Pan Y, Zhang H, Acharya AB, Pin H, Oliver PD, Morley JE. Effect of testosterone on functional recovery in a castrate male rat stroke model. Brain Res. 2005;1043:195–204. [PubMed]
  • Park JH, Bonthuis P, Ding A, Rais S, Rissman EF. Androgen-and estrogen-independent regulation of copulatory behavior following castration in male B6D2F1 mice. Horm Behav. 2009;56:254–263. [PMC free article] [PubMed]
  • Quimby FW. The mouse. In: Loeb WF, Quimby FW, editors. The clinical chemistry of laboratory animals. 2. Philadelphia, PA: Taylor and Francis; 1999. pp. 3–32.
  • Schaller BJ. Influence of age on stroke and preconditioning-induced ischemic tolerance in the brain. Exp Neurol. 2007;205:9–19. [PubMed]
  • Schallert T, Leasure JL, Kolb B. Experience-associated structural events, subependymal cellular proliferative activity, and functional recovery after injury to the central nervous system. J Cereb Blood Flow Metab. 2000;20:1513–1528. [PubMed]
  • Selim M. Perioperative stroke. N Engl J Med. 2007;356:706–713. [PubMed]
  • Singh SM, Gauthier S, Labrie F. Androgen receptor antagonists (anti-androgens): structure-activity relationships. Curr Med Chem. 2000;7:211–247. [PubMed]
  • Song X, Li G, Vaage J, Valen G. Effects of sex, gonadectomy, and oestrogen substitution on ischaemic-preconditoning and ischaemia-reperfusion injury in mice. Acta Physiol Scand. 2003;177:459–466. [PubMed]
  • Taggart H, Sheridan B, Stout RW. Sex hormone levels in younger male stroke survivors. Atherosclerosis. 1980;35:123–125. [PubMed]
  • Toung TJ, Traystman RJ, Hurn PD. Estrogen-mediated neuroprotection after experimental stroke in male rats. Stroke. 1998;29:1666–1670. [PubMed]
  • Tsang S, Liu J, Wong TM. Testosterone and cardioprotection against myocardial ischemia. Cardiovasc Hematol Disord Drug Targets. 2007;7:119–125. [PubMed]
  • Uchida M, Palmateer JM, Herson PS, DeVries AC, Cheng J, Hurn PD. Dose-dependent effects of androgens on outcome after focal cerebral ischemia in adult male mice. J Cereb Blood Flow Metab. 2009;29:1454–1462. [PMC free article] [PubMed]
  • vom Saal FS, Finch CE, Nelson JF. Natural history and mechanisms of reproductive aging in humans, laboratory rodents, and other selected vertebrates. In: Knobil E, Neill JD, Greenwald GS, Markert CL, Pfaff DW, editors. The physiology of reproduction. 2. Vol. 2. New York, NY: Raven Press; 1994. pp. 1213–1314.
  • Wang L, Kitano H, Hurn PD, Murphy SJ. Estradiol attenuates isoflurane preconditioning neuroprotection in ischemic mouse brain. J Cereb Blood Flow Metab. 2008a;28:1824–1834. [PMC free article] [PubMed]
  • Wang L, Traystman RJ, Murphy SJ. Inhalational anesthetics as preconditioning agents in ischemic brain. Curr Opin Pharmacol. 2008b;8:104–110. [PMC free article] [PubMed]
  • Yang SH, Liu R, Wen Y, Perez E, Cutright J, Brun-Zinkernagel AM, Singh M, Day AL, Simpkins JW. Neuroendocrine mechanism for tolerance to cerebral ischemia-reperfusion injury in male rats. J Neurobiol. 2005;62:341–351. [PubMed]
  • Yang SH, Perez E, Cutright J, Liu R, He Z, Day AL, Simpkins JW. Testosterone increased neurotoxicity of glutamate in vitro and ischemia-reperfusion injury in an animal model. J Appl Physiol. 2002;92:195–201. [PubMed]
  • Zhao P, Peng L, Li L, Xu X, Zuo Z. Isoflurane preconditioning improveslong-term neurologic outcome after hypoxic-ischemic brain injury in neonatal rats. Anesthesiology. 2007;107:963–997. [PubMed]