Address correspondence and reprint requests to Yizheng Wang, Laboratory of neural signal transduction, Institute of Neuroscience, SIBS, Chinese Academy of Science. 320 Yue-Yang Road, Shanghai, 200031 China. E-mail: firstname.lastname@example.org
Excitotoxicity induced by NMDA receptor-mediated intracellular Ca2+ ([Ca2+]i) overload is a major cause of delayed neuronal death in cerebral ischemia. Transient receptor potential canonical (TRPC) 6 protects neurons from ischemic brain damage. However, the mechanisms by which TRPC6 protects neurons are largely unknown. Here, we reported that TRPC6 suppressed the [Ca2+]i elevation induced by NMDA and protected neurons from excitotoxicity. Over-expressing or down-regulating TRPC6 suppressed or aggravated Ca2+ overload under excitotoxicity, respectively. TRPC6 protected cultured neurons from damage caused by NMDA toxicity or oxygen glucose deprivation (OGD). Moreover, the infarct volume in TRPC6 transgenic (Tg) mice was smaller than that in wild-type (WT) littermates. The TRPC6 Tg mice had better behavior performance and lower mortality than their WT littermates. Thus, TRPC6 inhibited NMDA receptor-triggered neurotoxicity and protected neurons from ischemic brain damage. Increase in TRPC6 activity could be a potential strategy for stroke prevention and therapy.
terminal deoxynucleotidyl transferase dUTP nick end labeling
Excitotoxicity elicited by excitatory amino acid releasing during ischemia-reperfusion process is a major factor inducing ischemic insults in cerebral stroke (Dirnagl et al. 1999; Lo et al. 2003). Overactivation of NMDA receptors is a key event in excitotoxicity that acutely elevates the [Ca2+]i concentration of neurons in a short time (Sattler et al. 1998; Arundine and Tymianski 2004). This sudden Ca2+ overload in neurons activates several endonucleases, kinases, and proteases that induce severe damage to cell functions and constructions, and finally leads to cell apoptosis or necrosis (Szydlowska and Tymianski 2010). Furthermore, a wealth of evidence indicates that suppressing the Ca2+ overload by blocking NMDA receptors can be neuroprotective in various experiment models. However, directly inhibiting NMDA receptors by antagonists for the treatment of stroke could induce severe adverse effects in the central nervous system (CNS) (Smith 2003). Finding endogenous pathways that block NMDA toxicity might provide new approaches for stroke therapy.
TRPC proteins belong to the TRP superfamily. They form Ca2+-permeable non-selective cation channels in various cell types in mammals. TRPC family has seven members falling into four subgroups based on sequence similarity: TRPC1, TRPC2, TRPC3/6/7, and TRPC4/5. Homo- or hetero-tetramers of these subunits form functional channels (Clapham 2003; Vazquez et al. 2004). TRPC channels have high expression levels in all the regions of the CNS (Riccio et al. 2002) and have various roles, including growth cone guidance (Li et al. 2005), fear memory (Riccio et al. 2009), and neuronal development (Jia et al. 2007; Tai et al. 2008; Zhou et al. 2008). In TRPC family, TRPC6 is notable for its important roles in neurite outgrowth (Tai et al. 2008), synapse formation (Zhou et al. 2008), and neuronal survival (Jia et al. 2007) during development. Recently, TRPC6 was found to be degraded at the early stage of cerebral ischemia, and maintaining its protein level could be protective against ischemic insults (Du et al. 2010). However, the mechanisms by which TRPC6 protects neurons in ischemia are still unclear.
In this study, we found that TRPC6 suppressed the [Ca2+]i elevation in neurons stimulated by NMDA, and that over-expression of TRPC6 protected neurons from NMDA toxicity and ischemic insults. Our results suggest that up-regulating TRPC6 levels could be a neuroprotective strategy against excitotoxicity and ischemic brain damage.
Materials and methods
Cell culture, transfection, and oxygen–glucose deprivation
Rat primary cortical or hippocampal cultures (from 17-day-old embryonic SD rats) were prepared as reported previously (Tao et al. 2006). The hippocampal neurons for Ca2+ imaging were transfected following the Ca2+-phosphate protocol (Jiang and Chen 2006). The cortical neurons were transfected using Nucleofector Kit (Amaxa Biosystems, Cologne, Germany) according to the manufacturer's instructions. For the oxygen–glucose deprivation (OGD), the culture medium (10 DIV) was replaced by glucose-free Earle's balanced salt solution. Then, the neurons were purged by nitrogen gas (5% CO2 and 95% N2) for 10 min and incubated for 2 h in an anaerobic chamber (Forma Scientific, Marietta, OH, USA) filled with 5% CO2 and 95% N2 at 37°C. The OGD was terminated by changing the cultures back to the normal culture condition.
Cell death assay
Cell death was evaluated by propidium iodide (PI) or terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) staining. For PI staining, cells were washed three times with phosphate-buffered saline (PBS) and stained with PI solution (3 μg/mL in PBS) for 2 min at 37°C. Then, the solution was removed and 4% paraformaldehyde (PFA) was added to fix the cells. The cell image was taken using a fluorescence microscope. The percentage of PI-positive cells was then determined. For TUNEL staining, the in situ cell death detection kit (Roche, Basel, Switzerland) was used according to the manufacturer's instructions. In brief, cells were washed with PBS and fixed with 4% PFA. After permeabilization with 0.2% triton X-100, cells were incubated with 50 μL of reaction mixture for TUNEL enzymatic reaction at 37°C for 1 h in the dark. After rinsing the cells with PBS for three times, the percentage of TUNEL-positive cells was determined.
The measurement of [Ca2+]i was done as described previously (Zhu et al. 1996). Briefly, hippocampal neurons loaded with 2 μM Fura-2 AM in HEPES-buffered saline [HBSS: (in mM) 120 NaCl, 5.3 KCl, 0.8 MgSO4, 1.8 CaCl2, 11.1 glucose, and 20 HEPES, pH7.4] for 35 min at 37°C were washed with HBSS twice for 20 min. The F340/F380 ratios were detected under the eclipse Te2000-e microscope (Nikon Corporation, Tokyo, Japan) with dual excitations at 340 and 380 nm, and detection of fluorescent emissions at 500 nm. Fluorescent images in green fluorescent protein (GFP)-positive neurons were collected every 6 s for ratiometric Ca2+ measurements. The [Ca2+]i levels were determined by R340/380nm using Metamorph software (GE_MDS, Rochester, NY, USA). The baseline was determined after exposure to HBSS for 8 min. NMDA (with 10 μM Glycine) or AMPA was diluted in HBSS and applied by surface perfusion.
Western blot analysis
The tissue samples were homogenized in the buffer [(in mM) 10 Tris-HCl pH7.4, 150 NaCl, 5 EDTA, 1% Triton-X100, 1 Na orthovanadate, 50 NaF, 5 dithiothreitol (DTT), and protease inhibitor cocktail (Sigma, St Louis, MO, USA)] as reported previously (Ashwal et al. 1998). The extracts (100 μg) were electrophoresed in sodium dodecyl sulfate/7% polyacrylamide gel electrophoresis , and transferred to polyvinyldifluoridine membranes. The membranes were incubated with the primary antibodies followed by the anti-rabbit or -mouse secondary antibodies (Amersham Biosciences, Piscataway, NJ, USA). The protein bands were visualized using the Amersham ECL system. The band density was evaluated using ImageQuant.
Generation of TRPC6 transgenic mice
The forebrain-specific TRPC6 Tg mice were generated using the CaMKIIα promoter to drive expression of mouse TRPC6. The plasmid containing an 8.5-kb genomic fragment, including the upstream of CaMKIIα initiation site, was inserted with the cDNA containing a 3-kb coding region for mouse TRPC6. The transgene (CaMKIIα promoter and TRPC6 cDNA with a SV40 polyA sequence) was injected into the fertilized oocytes of C57BL6 × FBN background. Three independent founder lines were generated that were characterized with high levels of TRPC6 in forebrain by immunoblotting and immunohistochemistry (Zhou et al. 2008). Genotypes of all offspring were also analyzed by RT-PCR. Mouse DNAs were amplified using the primers for TRPC6 transgene:
Mouse middle cerebral artery occlusion (MCAO) and evaluation of cerebral infarct size
All of the animal experimental protocols were approved by the Institutional Animal Care and Use Committee of Institute of Neuroscience, Shanghai, China. Male mice (25–30 g) with congenic C57BL6 background were anesthetized with 10% chloral hydrate, (body temperature 37°C) and focal cerebral ischemia was induced by MCAO (the suture method) for 120 min. After reperfusion for different periods, mouse brains were rapidly removed, coronally sectioned at 1.5-mm intervals (five for one brain), and stained with 2% 2,3,5-triphenyltetrazolium chloride (TTC, in 0.9% saline). The mouse rectal temperature was measured and 400 μL of arterial blood was taken for determining PaO2, PaCO2, and pH before MCAO and after 15-min reperfusion. Infarct size was evaluated by digital planimetry of the brain slices using the ImagePro software (Media Cybernetics, Inc. Bethesda, MD, USA) and normalized for edema. Infarct volume for each brain was calculated according to I% = (volume of contralateral − normal volume of ipsilateral)/volume of contralateral. Regional cerebral blood flow (rCBF) was measured by the Laser-Doppler flowmetry (Moor instruments Ltd, Axminster, UK) with a probe fixed on the skull throughout the ischemia until 10 min after reperfusion (Herrmann et al. 2005).
The rotarod assay
The mice were placed on the rotarod set (Ugo Basile, Italy/Panlab Apparatus, Comerio (VA), Spain) as reported previously (Hamm et al. 1994). The rotarod was accelerated uniformly from 2.688*10−4 to 2.688*10−2g. over 5 min. The mouse latency to fall was recorded. The mice were trained three times per day for 3 days and on the morning of the experimental day to reduce the variability between subjects. We examined the motor function before surgery and 1, 7, 14, 21, and 28 day after ischemia/reperfusion.
Whole-cell patch-clamp recordings of hippocampal neurons under current clamp were carried out at room temperature (22–25°C) using an Axopatch 700A patch-clamp amplifier (Axon Instruments, Union city, CA, USA) to assess the effects of TRPC6 over-expression on NMDA/AMPA stimulation-mediated membrane potential. Data acquisition was achieved using a DigiData 1322A with pClamp 9.0 software (Axon Instruments, Union city, CA, USA). The acquisition rate was 10 kHz and signals were filtered at 5 kHz. Patch electrodes were pulled with a Flaming/Brown micropipette puller (Sutter Instruments, Novato, CA, USA) and fire polished. The recording electrodes had a resistance of 4–6 MΩ when filled with the internal solution. The pipette solution [(in mM): 120 potassium gluconate, 20 KCl, 2 MgCl2, 10 HEPES, 10 EGTA, 2 Na2ATP, pH7.3] and the external solution [(in mM): 135 NaCl, 5.4 KCl, 1.8 CaCl2, 1.3 MgCl2, 10 glucose, 10 HEPES, 0.0002 TTX, pH 7.4] were used. NMDA/AMPA (with 1 μM Glycine), 100 or 300 μM, was prepared in extracellular solutions and applied to neurons using the eight-channel focal perfusion system (ALA Scientific Instruments, Westbury, NY, USA). Neurons were bathed constantly in extracellular solutions between drug applications. Drug solution exchange was accomplished by electronic control.
Antibodies and reagents
Rabbit anti-TRPC1, 3, 4, and 5 antibodies were from Alomone Labs (Alomone Labs, Jerusalem, Israel), anti-TRPC6 antibodies from Millipore (Millipore, Temecula, CA, USA) and Sigma (Sigma), anti-CD31 antibodies from Millipore, mouse anti-GFP, rabbit anti-GluR1, GluR2/3, and mouse anti-postsynaptic density protein 95 (PSD95) antibodies from Upstate (Upstate Biotechnology, Inc., Lake Placid, NY, USA), anti-α-tubulin from Sigma. Unless stated, all chemicals were from Sigma. The TRPC6 was cloned into the pcDNA3.1 and pCAGGS–IRES–GFP vectors. The pCAGGS–IRES–GFP vector directed target gene expression together with GFP.
All data were presented as mean ± SEM. The means between two groups were analyzed by either paired or unpaired t-tests. One-way anova followed by a post hoc least-significant difference test was performed for statistical comparison of several groups. Statistically significant differences were defined as p <0.05. The rotarod test was calculated using repeated-measures anova. The probability of survival was calculated using the non-parametric Kaplan–Meier method.
TRPC6 inhibited the NMDA-induced Ca2+ influx in neurons
We initially examined whether TRPC6 could affect the Ca2+ influx in neurons induced by NMDA. Hippocampal neurons transfected with TRPC6 was subjected to 100 μM NMDA stimulation for 15 min, a condition known to induce the neuronal cell death (Dawson et al. 1991), and the [Ca2+]i elevation was recorded by Ca2+ imaging. We found that [Ca2+]i elevation induced by NMDA in the neurons transfected with TRPC6 was much smaller than that in the neurons transfected with empty vectors, and this TRPC6-mediated inhibition of the [Ca2+]i elevation was found during the entire experiment period (Fig. 1a, p = 0.014 vs. vector). Expressing the dominant-negative TRPC6 (DNC6) blocked TRPC6 inhibition of [Ca2+]i elevation induced by NMDA (Fig. 1a, p = 0.021 vs. WT-TRPC6). Moreover, when neurons were stimulated with 300 μM NMDA, the TRPC6 inhibitory effect on the [Ca2+]i elevation induced remained evident (Figure S1a). In contrast, over-expressing TRPC6 did not suppress the [Ca2+]i elevation induced by AMPA (Figure S1b) and did not affect the membrane potential induced by NMDA (Figure S2). Furthermore, in the neuronal cultures transfected with TRPC6, the protein levels of TRPC3, NR1, NR2A, and NR2B were not changed (Fig. 1b), indicating that TRPC6 suppressed the Ca2+ influx not by altering the expression of NMDA receptor subunits. These results suggested that TRPC6 specifically suppressed NMDA receptor activities. To show whether the endogenous TRPC6 also inhibits the NMDA receptor activities, we studied the effect of knockdown of TRPC6 on the [Ca2+]i elevation induced by NMDA. Down-regulation of TRPC6 by siRNA against TRPC6 (C6-i) was evident and specific as C6-i did not affect the expression of TRPC3, NR1, NR2A, and NR2B (Fig. 2a). As shown in Fig. 2b, down-regulation of TRPC6 greatly enhanced the [Ca2+]i elevation induced by NMDA (p = 0.0005).
We then explored the possible mechanism of TRPC6 inhibition of NMDA receptor activities. As TRPC6 did not alter the expression of NMDA subunits (Fig. 1b) and TRPC6 can activate calcineurin (Kuwahara et al. 2006), a phosphatase known to inhibit the NMDA currents (Lieberman and Mody 1994), we examined the effect of FK506, an agent known to specifically inhibit calcineurin (Wu et al. 2004), on TRPC6 inhibition of the NMDA-induced [Ca2+]i elevation. As shown in Fig. 2c, in the presence of FK506, TRPC6 could not inhibit the NMDA-mediated [Ca2+]i elevation. Together, these results suggested that TRPC6 suppressed the NMDA-mediated [Ca2+]i elevation in the neurons likely through a calcineurin-dependent mechanism.
TRPC6 protected neurons against NMDA toxicity and modeled ischemia
We then asked whether TRPC6 could protect neurons against NMDA toxicity and ischemic insults. The neurons transfected with TRPC6 or empty vectors were exposed to 100 μM NMDA for 30 min and the cell death was assessed 24 h later. The transfection rate for TRPC6 or vector groups was about 46.72 ± 2.51% or 47.70 ± 1.30%, respectively (Figure S3a and b). As shown in Fig. 3a, the numbers of apoptotic cells in TRPC6-over-expressing cultures (30.77 ± 6.02%) were much less than those in the control cultures (67.19 ± 9.76%). We then subjected the cultured neurons to OGD, a condition mimicking ischemia in cultures to induce the NMDA-dependent cell death (Goldberg and Choi 1993; Lei et al. 2006) (Figure S3c), and examined the effect of TRPC6 expression on cell survival. As shown in Fig. 3b, neuronal cell death was greatly increased in OGD (30.49 ± 1.34%). Over-expressing TRPC6 (13.79 ± 2.39%, p = 0.0001 vs. control) markedly protected neurons from the injury induced by OGD. Meanwhile, the numbers of GFP-positive neurons after OGD in TRPC6-over-expressing cultures (66.74 ± 4.49%) were much more than those in the control cultures (51.33 ± 2.58%), indicating that neurons transfected with TRPC6 tended to survival (Figure S3b).Together, all these results suggested that up-regulating TRPC6 made neurons resistant to excitotoxicity and modeled ischemic insults.
TRPC6 transgenic mice presented smaller infarct size and better behavior performance after ischemia
We then examined whether TRPC6 could protect neurons from ischemic brain damage in mouse MCAO, a well-known model of focal ischemia (Longa et al. 1989; Merchenthaler et al. 2003). Mice subjected to MCAO for 2 h and followed by different periods of reperfusion showed a stable increase in the infarct volume (Figure S4a). Intraperitoneal injection of memantine, known as a NMDA receptor inhibitor, dramatically decreased the infarct volume (Figure S4b). These results indicated that inhibition of NMDA toxicity was protective in our MCAO model.
To provide direct evidence that TRPC6 is indeed neuroprotective in ischemia, we made transgenic mice in which over-expression of TRPC6 was driven under the CaMKIIα promoter, known to post-natally control gene expression in the forebrain neurons (Mayford et al. 1996). Three independent founder mice were obtained (Fig. 4a), which did not show obvious developmental defects and had normal physiological exhibition (Table S1). Immunoblot experiments confirmed that the TRPC6 protein levels in the forebrain of Tg mice were indeed enhanced relative to those in the forebrain of the WT littermates (Zhou et al. 2008). In contrast, the protein levels of TRPC3, TRPC4, TRPC5, NR2A, GluR2/3, and PSD95 were unaltered (Figure S4c).
We then subjected the WT and Tg mice to ischemia and assessed the infarct volume in a double-blinded manner. The infarct volume was 32.72 ± 3.27% in WT mice, whereas it was 18.66 ± 3.95% in Tg mice (Fig. 4b, p = 0.0109). After ischemia, TRPC6 expression levels in Tg mice remained higher than those in WT mice (Fig. 4c, sham: WT = 1.01 ± 0.02, TRPC6 = 1.83 ± 0.09; ischemia: WT = 0.44 ± 0.02, TRPC6 = 1.45 ± 0.08; p < 0.0001). To test whether the protection was associated with a high level of TRPC6 in the neurons, we probed the brain sections with the TRPC6 antibody and the TUNEL labeling. The neurons in Tg mice with relatively high TRPC6 immunoreactivity were TUNEL negative, whereas those in WT littermates with a dramatic loss in TRPC6 immunoreactivity were TUNEL positive (Fig. 4d), indicating that neurons with a high level of TRPC6 were more resistant to ischemic insults. The expression of CD31, an endothelial cell marker used to indicate the structure and density of the vessels, and the relative cerebral blood flow were not different between Tg and WT mice (Figure S4d). We evaluated the behavior performance of both WT and Tg mice using the rotarod assay, a method widely used to test motor coordination and balance of ischemic animals (Hunter et al. 2000; Borsello et al. 2003). As shown in Fig. 4e, Tg mice had better behavior performance than WT mice. In addition, the survival analysis in a double-blinded manner showed that the mortality of Tg mice was much lower than that of WT mice (Fig. 4f). These results suggested that increasing TRPC6 expression reduced infarct size and improved behavior outcome of the ischemic mice.
Here, we showed that TRPC6 inhibited NMDA-induced [Ca2+]i elevation and that increasing TRPC6 levels protected neurons against excitotoxicity and ischemic brain damage. Several lines of evidence support this conclusion. First, over-expressing or down-regulating TRPC6 suppressed or enhanced [Ca2+]i elevation in neurons stimulated by NMDA, respectively. Second, over-expressing TRPC6 rendered neurons resistant to excitotoxicity in modeled ischemia. Finally, TRPC6 Tg mice presented smaller infarct size and better behavior performance than WT mice after ischemia. Our results thus suggest that up-regulating TRPC6 could be a neuroprotective strategy against ischemic brain damage by alleviating excitotoxicity.
It has been reported that TRPC6 is important for neuronal survival during development or under pathological conditions (Jia et al. 2007; Du et al. 2010). Our present studies suggested that TRPC6 inhibited Ca2+ overload induced by NMDA to protect neurons from excitotoxicity. As the important excitatory receptors in the CNS, NMDA receptors can be regulated by various mechanisms. It has been reported that NMDA receptors can be modified by interacting with other receptors (Dalva et al. 2000; Lee et al. 2002). For instance, the dopamine D1 receptor interacts with NR2A subunits to suppress the NMDA current (Lee et al. 2002). Moreover, after cerebral ischemia, both NR2A and NR2B were increasingly phosphorylated (Pei et al. 2000; Liu et al. 2003). As over-expressing TRPC6 did not affect the expression of NMDA receptor subunits, but increased activity of ERK (Jia et al. 2007) and CaMK (Tai et al. 2008), TRPC6 might regulate the phosphorylation of NMDA receptors to affect its activities. In heart, the high level of TRPC6 activates calcineurin, a phosphatase involved in the antiapoptotic effect against ischemic insults (Lieberman and Mody 1994). In neurons, activated calcineurin inhibits the NMDA currents by dephosphorylating NMDA receptors. Consistently, TRPC6 could not inhibit NMDA-induced [Ca2+]i elevation in the presence of FK506. Therefore, it is possible that TRPC6 inhibits the activity of NMDA receptors by increasing its dephosphorylation via calcineurin. In this context, TRPC6 could be another protein that regulates NMDA receptor activity.
In conclusion, TRPC6 can suppress the NMDA-mediated Ca2+ overload and excitotoxicity. Keeping a high level of TRPC6 could be a new strategy for stroke prevention or therapy to suppress the excitotoxic pathways in cerebral ischemia.
This work was supported by a grant (81130081) from NNSF of China. We thank Q. Hu and ZJ. Fan for technical assistance. The authors have no conflict of interest to declare.