The 12-kDa FK506-binding Protein Involvement in Neuroprotection Provided by Dantrolene in Ischemic/Reperfused Rat Cerebral Cortex

Summary

Purpose:
To investigate any neuroprotection provided by dantrolene and the mechanism involved in its action.

Materials and methods:
A middle cerebral artery occlusion (MCAO) model was followed and the expressions of 12-kDa FK506-binding protein (FKBP12) and ryanodine receptor were assessed by fluorescent double staining. Infarct volume was assessed by triphenyltetrazolium chloride (TTC) and toluidine blue staining.

Results:
The study showed that FKBP12 was upregulated in the dantrolene group compared with the control group at each corresponding time point; RyR co-expressed with FKBP12, although its staining only increased slightly from 1 h to 24 h. We also found that treatment with dantrolene 1 h and 4 h after MCAO did not reduce infarct volumes (p >0.05); however, treatment with dantro­lene 24 h after MCAO reduced infarct volumes and had a significant effect on the infarct volume (p <0.05), indicating that inhibition of ryanodine receptor (RyR) leads to cytopro­tection.

Conclusion:
FKBP12 may play an important role in the processes of neuronal survival or death following cerebral ischemia, and dantrolene may exert its neuroprotective effects by upregulating FKBP12 and thereby decreasing infarct volume.

Key words:
FKBP12 – ryanodine receptor – dantrolene – cerebral ischemia

Introduction

Ischemic brain injuries resulting from either global or focal decreases in perfusion are among the most common and important causes of disability and death worldwide. Abundant preclinical studies have identified a wide range of mechanisms of ischemic brain injury, many of which may offer potential targets for neuroprotective strategies [1,2].

Calcium, in particular intracellular Ca²⁺, plays a central role in neuronal function and injury. During cerebral ischemia, cyto­plasmic Ca²⁺ levels rise and activate several Ca²⁺-dependent pathways that participate in cell survival and/or cell death [3,4]. Increased Ca²⁺ entry through plasma membrane channels during cerebral ischemia is well documented [4–6]. In contrast, the effects of ischemia on endoplasmic reticulum (ER) Ca²⁺ release have received less attention [7,8]. The ER contains at least two types of releasable Ca²⁺ stores, one of which is released by inositol 1,4,5-trisphosphate (IP₃) via the IP₃ receptor (IP₃-induced Ca²⁺ release; IICR) and the other by Ca²⁺ itself in a process referred to as Ca²⁺-induced Ca²⁺ release (CICR), via the ryanodine receptor (RyR) [9]. An in vitro study has indicated that approximately two-thirds of the elevated intracellular Ca2+ during ischemia may be derived from intracellular Ca²⁺ stores [10]. The endo­plasmic reticulum thus appears to play a critical role in the regulation of free cytosolic Ca²⁺ concentration and related cellular activity under ischemic conditions.

Dantrolene, an inhibitor of the ryanodine receptor, has provided neuroprotection in multiple in vitro models and some in vivo models of neural injury, but the mechanism of dantrolene action on RyR remains controversial [11,12]. We there­fore designed this study to investigate changes in RyR and the 12-kDa FK506-binding protein (FKBP12), a ubiquitous protein that copurifies with RyR [13,14], in dantrolene-treated and vehicle-treated rats suffering from focal cerebral ischemia, in order to disclose the mechanism of the protective effect of dantrolene.

Materials and methods

Animals and protocols

Adult male Wistar rats (body weight, 250–280 g, Shanghai SLAC Laboratory Animal Co., Ltd, Shanghai, China) were divided randomly into treatment (n = 18) and control (n = 18) groups. Dimethyl sulfoxide (DMSO) was administered intra­cerebroventricularly in the control group immediately after middle cerebral artery occlusion (MCAO) surgery, and dantrolene dissolved in DMSO was administered to the dantrolene group in the same way.

The rats were anesthetized with an intra­peritoneal injection of chloral hydrate (10%, 0.35 ml/100 g), and two burr holes with a diameter of 2 mm for measuring regional cerebral blood flow (rCBF) and administration of drug were carefully bored in the skull using an electric dental drill to avoid traumatic brain injury. The locations of the burr holes for measuring rCBF were 3 mm dorsal and 4 mm lateral (right) to the bregma, locations in the upper part of the middle cerebral artery (MCA) area. The burr holes for dantrolene or vehicle administration were 0.8 mm dorsal and 1.4 mm lateral (left) to the bregma. Dura was preserved. The animals were allowed to recover under ambient conditions. At 24 h after the drilling, the rats were anesthetized again and subjected to MCAO surgery. The right MCA was occluded by insertion of a processed nylon thread (F-24, Beijing Shadong Co., Ltd, Beijing, China) through the common carotid artery, as previously described ­elsewhere [15,16]. Body temperature was maintained at approximately 37 °C during the procedure by means of a heat pad and lamp. After 90 min of transient MCAO, CBF was restored by withdrawal of the nylon thread. The rCBF of the right fronto­parietal cortex was measured with laser Doppler flowmetry (PowerLab data acquisition systems 16/30, ADInstruments pty ltd, Bella Vista, NSW, Australia). Immediately after MCAO, a micro-injector was slowly inserted into the lateral ventricles (3.6 mm from the cortical surface). The rats received an injection of 4 μl of dantrolene (total 20 μg, dissolved in DMSO at a dilution of 5 μg/μl, cat. no. D9175, Sigma) or DMSO (cat. no. D8779, Sigma) slowly over 4 min by micro-injection pump (Shanghai Alcott Biotech Co., Ltd, Shanghai, China). The dose of the drug was based on previous reports [16,17].

Fluorescent double staining

At 1 h, 4 h, or 24 h after reperfusion, deeply anesthetized rats were sacrificed by decapitation, and their brains rapidly frozen in solid carbon dioxide immediately after removal. Coronal microtome sections were made at the level of the striatum on a cryostat (Leica CM1100, Germany) at –15 °C, at a thickness of 10 μm, and collected on glass slides. For immunofluorescence analysis, the coronal sections were fixed in 4% paraformaldehyde in phosphate-buffered saline (PBS) for 20 min, blocked with 5% bovine serum albumin in 1 × TBST for 60 min, and then incubated with primary antibody against ryanodine receptor (mouse monoclonal, cat. no. ab2868, Abcom)/NeuN (mouse monoclonal, cat. no. MAB377, Chemicon) and FKBP12 (rabbit polyclonal, cat. no. SC-28814, Santa Cruz Biotechnology) at a dilution of 1 : 100 overnight at 4 °C. Negative-control sections were treated simultaneously without the primary antibodies. After washes in TBST, the sections were incubated with secondary antibodies (1 : 500 dilution) for 60 minutes: rhodamine-conjugated goat anti-mouse IgG antibody (ab6786, abcom) for RyR//NeuN and FITC-conjugated donkey anti­--rabbit IgG (ab6798, abcom) antibody for FKBP12. Finally, the sections were observed under a fluorescence microscope (Olympus 1 × 71), and fluorescence was observed at an excitation of 495 nm for FITC and 550 nm for rhodamine.

All experimental protocols used in this work were approved by the Animal Committee of Harbin Medical University, China.

Infarct Volume Determination

Infarct volume was assessed by triphenyl­tetrazolium chloride (TTC) and toluidine blue staining. Infarct volume was assessed with vital TTC staining. Animals were sacrificed by decapitation, the brains dissected out, and 2- mm thick coronal slices were cut using a brain slicer. The slices were ­immersed in 1% TTC in 0.25 M phosphate buffer (pH 8.5) for 20 min in the dark; tissue sections were stored in a 10% formal­dehyde solution. Consecutive slices of 20- mm thickness were cut and stained with toluidine blue. For both staining methods, the infarct areas were assessed with a computerized image analyzer, and the infarct volume was calculated.

Data Analysis

Data are expressed as mean of n values ± S.E.M. and were compared using one-way analysis of variance (ANOVA) ­followed by Fisher’s least significant difference (FLSD) test (Systat,SPSS).

Results

Immunofluorescence staining of FKBP12

FKBP12 exhibited a heterogeneous distribution throughout the rat brain, with primarily neuronal localization (Fig. 1). Following ischemia, neuronal FKBP12 ­immunoreactivity in the ischemic core of the cortex reduced at 4 h and remained very low at 24 h in both dantrolene and control groups, but the sparsely stained areas of the ischemic core were much smaller in the dantrolene group at 24 h than the control group, which may indicate less severe ischemic injury. Additionally, in the ischemic penumbra, the farther away from the ischemic core, the brighter the FKBP12 fluorescence proved to be.

In the ischemic penumbra of control groups, immunofluorescence for FKBP12 was weakly detectable in a few cells at 1 h of reperfusion, displaying punctuate staining in the cytoplasm. Four hours after reperfusion, FKBP12 immunostaining was slightly brighter than at 1 h. The staining appeared as points localized in the cytoplasm that could distinguish the cell clearly from the background. And then, at 24 h, FKBP12 became strongly stained, and both the cytoplasm and ­nuclei of numerous cells were filled with fluorescence, although the staining in the cytoplasm was somewhat brighter than in the nucleus.

In the ischemic penumbra of the dantrolene group, the fluorescent staining of FKBP12 increased from 1 h to 24 h post-reperfusion, a trend similar to that in the control group, except that more fluorescent cells and brighter fluorescence were found at each time point compared with the control group; that is to say, compared with the control groups, the expression of FKBP12 was up-regulated in the dantrolene group at 1 h, 4 h and 24 h of reperfusion. The results of the study for FKBP12 are summarized in Fig. 2.

Immunofluorescence staining of RyR

Fluorescent double-staining experiments showed that RyR was co-expressed with FKBP12 (Fig. 1), but the values of IOD (integrated optical density) and IOD/area of RyR were not identical with FKBP12. The immunoreactivities for RyR were weakly detectable at 1 h of reperfusion and became progressively stronger at 4 and 24 h after reperfusion in the ischemic peripheral region in the DMSO group. ­Further, in the dantrolene group the increases in staining were intensified (but still moderate) at 24 h post-reperfusion, whereas the changes at 4 h were not so conspicuous. As a result, at 1 h of reperfusion in both groups, the IOD and IOD/area of FKBP12 were a little higher than RyR. But as the immunoactivity of FKBP12 increased, the difference increased. So although RyR increased in a similar manner to that of FKBP12, the variation curve of RyR was flatter for the latter.

Infarct volume determination

The mean infarct volume in control treated rats (n = 18) was assessed by TTC stai­ning. Treatment with dantrolene 1h and 4 h after MCAO did not reduce infarct volumes (n = 6) and no significance emerged (p >0.05). Treatment with dantrolene 24 h after MCAO reduced infarct volumes by 38% and had a significant effect on the ­infarct volume (p <0.05), indicating that inhibition of ryanodine receptor (RyR) leads to cytoprotection in this model (Fig. 3).

Discussion

Multiple in vitro models and some in vivo models of neural injury have verified the neuropreotective effect of dantro­lene [12]. Furthermore, one of our previous reports had suggested that dantro­lene significantly decreases infarct volume and provides neuroprotective effect in rats after transient MCAO by reducing ER stress-mediated apoptotic signal pathway activation in the ischemic area [16], but the mechanism of the neuroprotective effect remains obscure. In the current study, we found that treatment with dantrolene 24 h after MCAO reduced infarct volumes, while treatment with dantrolene 1 h, 4h after MCAO did not reduce infarct volumes, and we suggest that the possible mechanism lies in the expression of RyR and that levels of the 12-kDa FK506-binding protein (FKBP12) do not increase.

It has been demonstrated that the ­single RyR channel from ischemic and control brains displays the same three responses to Ca²⁺, characterized by low, moderate, or high maximum activity. ­Relative to controls, RyR channels from ischemic brains display increased frequency of high-activity response and lower frequency of low-activity response [8]. Studies have also revealed that FKBP12 is responsible for stabilizing the closed conformation of RyR in various tissues [14,18–21], and that disruption of the FKBP12-RyR complex enhanced Ca²⁺ leakage through RyR and gave rise to an irreversible increase in intracellular calcium concentration and decrease in calcium concentration in the endoplasmic reticulum [14,20,22], both of which are pathogenic [12,13]. A rise in intracellular Ca²⁺ leads to activation of catabolic enzymes, mitochondrial disturbances, free radical production and immediate early gene responses. A fall in Ca²⁺ in the ER evokes an ER stress res­ponse, which mediates a great number of ER-dependent functional disturbances [23–25].

As far as we can establish, this is the first record that dantrolene reduces infarct volume and that FKBP12 is involved in the process. The preliminary data need to be thoroughly discussed and further study definitely undertaken. In conclusion, the current study showed that FKBP12 plays an important role in the process of neuronal survival and death following cerebral ischemia, and that dantrolene reduces ­infarct volume, possibly by means of ­exerting a neuroprotective effect by upregulating FKBP12.

Acknowledgments

We would like to thank Prof. Wen-Zhi LI for providing the laser Doppler flowmetry for measuring regional cerebral blood flow. This work was supported by 211 Project Special Fund for Key Laboratory of Neuro-immunology, Science and Technology Projects for Returned Overseas Chinese Scholars of Heilongjiang Province (LC06C26), and the China Postdoctoral Science Foundation.

Wei-Zhi Wang, MD.
Department of Neurology
Second Affiliated Hospital of Harbin Medical University
Harbin 150086, China
e-mail: wwzguo@yeah.net