Address correspondence and reprint requests to Guang-Yi Zhang, Research Center for Biochemistry and Molecular Biology, Xuzhou Medical College, 84 West Huai-hai Road, Xuzhou, Jiangsu 221002, China. E-mail: firstname.lastname@example.org
We investigated the expression and subcellular localization of the multidomain protein POSH (plenty of SH3s) by immunohistochemistry and western blot analysis, as well as its role in the selective activation of mixed-lineage kinases (MLKs) 3, MAP kinase kinase (MKK) 4, c-Jun N-terminal kinases (JNKs) and the c-Jun signalling cascade in the rat hippocampal CA1 region following cerebral ischaemia. Our results indicated that the cytosol immunoreactivity of POSH was strong in the CA1-CA3 pyramidal cell but weak in the DG granule cell of the rat hippocampus both in sham control and after reperfusion. Co-immunoprecipitation experiments showed that the interactions of MLK3, MKK4 and phospho-JNKs with POSH were persistently enhanced during the early (30 min) and the later reperfusion period (from 1 to 3 days) compared with sham controls. Consistently, MLK3–MKK4–JNK activation was rapidly increased with peaks both at 30 min and 3 days of reperfusion. Intracerebroventricular infusion of POSH antisense oligodeoxynucleotides (AS-ODNs) not only significantly reduced the protein level of POSH, markedly decreased its interactions with MLK3, MKK4 and phospho-JNKs, but also attenuated the activation of the JNK signalling pathway. In addition, infusion of POSH AS-ODNs significantly increased the neuronal density in the CA1 region at 5 days of reperfusion. Our results suggest that POSH might serve as a scaffold mediating JNK signalling activation in the hippocampal CA1 region following cerebral ischaemia, and POSH AS-ODNs exerts its protective effects on ischaemic injury through a mechanism of inhibition of the MLK3–MKK4–JNK signalling pathway, involving c-Jun and caspase 3 activation.
sodium dodecyl sulfate – polyacrylamide gel electrophoresis
Transient, severe global ischaemia in animals leads to selective and delayed neuronal death, particularly of pyramidal neurons in the hippocampal CA1, and the neuronal death occurs within 2–4 days after the initiation of reperfusion following ischaemia in rats and gerbils (Pulsinelli et al. 1982; Tanaka et al. 2000). The c-Jun N-terminal kinases (JNKs) activation and subsequent c-Jun transcriptional activity induced by ischaemic stress are closely associated with post-ischaemic neuronal cell death (Irving and Bamford 2002). JNKs are members of the family of the mitogen-activated protein kinase (MAPK) pathway that is activated in response to many extracellular stimuli and different forms of environmental stress. The MAPK pathway is organized as a cascade of at least three kinases: an MAPK kinase kinase (MAPKKK) first phosphorylates a dual-specificity protein kinase (MAPKK), which in turn phosphorylates the MAPK. In neurons, the mixed lineage kinase (MLK) family of kinases serves as the major MAPKKKs and phosphorylates MAP kinase kinase (MKK) 4 and MKK7, which in turn phosphorylate the JNKs (Wang et al. 2004). Our previous reports have demonstrated that JNK signalling was biphasically activated after reperfusion and inhibition of JNKs activation with JNKs antisense oligodeoxynucleotides (AS-ODNs) significantly inhibited post-ischaemic neuronal death, including cell apoptosis (Gu et al. 2001a). While the functions of the JNKs under ischaemic conditions are not completely understood, there is increasing evidence that JNKs are potent effectors of apoptosis following cerebral ischaemia (Zheng et al. 2003). The activation of caspase 3 and the inducible transcription factor c-Jun by N-terminal phosphorylation is a central event in JNK-mediated apoptosis (Gelderblom et al. 2004). But the exact signalling pathway leading to JNK activation from the upstream activator has not been fully understood in this ischaemic model, and little is known as to how JNK signalling is organized.
Members of the JNK-interacting protein (JIP) family are a set of scaffold elements for the JNK/c-Jun pathway that form a specific module of the MLK/MKK7/JNK kinase cascade that appear to play required roles for neuronal death in certain apoptotic paradigms (Yasuda et al. 1999). Besides JIP, the multidomain protein POSH (plenty of SH3s) has been reported to act as a scaffold for the specific signalling pathway of neuronal death (Xu et al. 2003). POSH is widely expressed in different tissues, including the brain. POSH is a novel 93-kDa protein containing four Src-homology-3 (SH3) domains (amino acids 139–190, 198–254, 457–511 and 838–892); the first two SH3 domains activate the JNK pathway; the second two SH3 domains promote NF-κB translocation and apoptosis (Tapon et al. 1998). POSH interacts directly with the constitutively active form of Rac1 (but not Cdc42Hs, Ras or Rho) in a yeast two-hybrid screen and acts as a scaffold protein that contributes to Rac1-dependent activation of JNK and apoptotic death (Tapon et al. 1998). Recently, Xu et al. (2003) reported that POSH could directly bind MLKs (MLK1-3, DLK) and complex with MKK4/7 and JNK1/2, and POSH AS-ODNs or POSH siRNA suppress the neuronal apoptotic death promoted by NGF deprivation. These findings clearly indicate that POSH serves as a required component for a multiprotein complex that links activated Rac with the JNK kinase cascade and facilitates the formation of a functional JNK signalling module that culminates in phosphorylation of c-Jun and promotes neuronal cell death. Another report demonstrates that Akt2 preferentially binds the third SH3 domain of POSH in mammalian cells (Figueroa et al. 2003). A POSH mutant that is unable to bind Akt2 exhibits increased binding to MLK3, which is accompanied by increased activation of the JNK signalling pathway. In addition, Akt2 binding to the POSH scaffold promotes the disassembly of the complex through Akt-mediated phosphorylation of the MLK3. These results suggest that POSH plays multiple cellular roles in signalling transduction pathways, and that the SH3 domains of POSH serve as a scaffolding function, directly binding MLKs, that contain SH3 binding sites.
It has been shown that an MLKs inhibitor, CEP-1347, inhibits the activation of the JNK pathway and the cell death in many cell culture and animal models of neuronal death (Saporito et al. 2002; Wang et al. 2004). In our recent study, inhibiting the MLK3 pathway by K252a, an analogue of CEP-1347, significantly prevented ischaemic injury in the hippocampal CA1 region, involving the death effector of caspase 3 (Pan et al. 2005). However, the mechanisms involved in the regulation of MLK3 activity remain unclear. This study aimed to clarify the distribution and possible role of POSH protein in response to cerebral ischaemia. We tested the expression and interactions of POSH with JNK cascade components to get further insight into the probable mechanism in the regulation of JNK signalling in the hippocampal CA1 subfield. We prevented activation of JNKs, c-Jun and caspase 3 by inhibiting POSH using AS-ODNs. Moreover, blocking the endogenous expression of POSH before ischaemia significantly reduced global ischaemic brain injury. Taken together, these data unravel a critical role for POSH in the control of the JNK activity in hippocampal CA1 pyramidal cells.
Materials and methods
Anti-phospho-MKK4 (no. 9151), anti-phospho-MLK3 (Thr277/Ser281, no. 2811) and anti-cleaved caspase 3 (no. 9661) antibodies were from Cell Signaling Technology (Beverly, MA, USA). Antibodies to POSH (sc-8280), MLK3 (sc-13072), active JNKs (recognizes JNK1, JNK2 and JNK3, sc-6254), MKK4 (sc-964), and mouse monoclonal anti-phospho-c-Jun (sc-822) antibody were from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Nitrocellulose filter was from Amersham Pharmacia (Little Chalfont, UK). BCIP (5-bromo-4-chloro-3-indolyl-phosphate) and NBT (nitro blue tetrazolium) were from Promega (Madison, WI, USA). Vectastain ABC Elite kit and 3,3-diaminobenzidine peroxidase substrate kit was from Vector Laboratories (Burlingame, CA, USA). All the other chemicals were from Sigma (St Louis, MO, USA) unless indicated otherwise.
To investigate the possible role of POSH in response to ischaemic injury, 10 nmol of end-phosphorothioated POSH AS-ODNs in 10 μL TE buffer [10 mm Tris-HCl (PH 8.0), 1 mm EDTA] were administrated to the rats every 24 h for 3 days by means of unilateral intracerebroventricular (i.c.v.) infusion. The same dose of missenses (MS) or vehicle (TE) was used as controls. The sequences for POSH AS-ODNs used in this study were 5′-GGCAGACTCATCCATCTTGAG-3′ from Xu et al. (2003), and missenses were 5′-ATGGACCGCACATGGCTTACT-3′. For i.c.v. injection, the rats were placed on ear bars of a stereotaxic instrument under anaesthesia. Drug infusion was performed using a stepper-motorized microsyringe (Stoelting, Wood Dale, IL, USA) at a rate of 1 μL/min through a pre-implanted cannula in the left cerebral ventricle (from the bregma: anteroposterior, −0.8 mm; lateral, 1.5 mm; depth, 3.5 mm).
Induction of forebrain ischaemia
Adult male Sprague–Dawley rats weighing 250–300 g were used (Shanghai Experimental Animal Center, Chinese Academy of Science; the surgical procedures were approved by the Centre). Brain ischaemia was induced by four-vessel occlusion, as described previously (Pulsinelli and Brierley 1979; Gu et al. 2001a). Briefly, under anaesthesia with chloral hydrate (350 mg/kg, i.p.), vertebral arteries were electrocauterized and common carotid arteries were exposed. Rats were allowed to recover for 24 h and fasted overnight. Ischaemia was induced by occluding the common arteries with aneurysm clips. Rats which lost their righting reflex within 30 s and whose pupils were dilated and unresponsive to light during ischaemia were selected for the experiments. Rats with seizures were discarded. An EEG was monitored to ensure isoelectricity within 30 s after carotid artery occlusion. Carotid artery blood flow was restored by releasing the clips. Rectal temperature was maintained at about 37°C during and 2 h after ischaemia. Sham controls were performed using the same surgical exposure procedures, except that the arteries were not occluded.
For brain tissue preparation, rats were killed under anaesthesia at 10 min, 30 min, 6 h, 1 and 3 days after 15 min of global cerebral ischaemia. The hippocampal CA1 and CA3/DG regions were microdissected from both sides of the hippocampal fissure at 0°C and quickly frozen in liquid nitrogen. Tissues were homogenized in ice-cold homogenization medium consisting of 50 mm HEPES (pH 7.4), 150 mm NaCl, 12 mmβ-glycerophosphate, 3 mm dithiotheitol (DTT), 2 mm sodium orthovanadate (Na3VO4), 1 mm EGTA, 1 mm NaF, 1 mm phenylmethylsulfonyl fluoride (PMSF), 1% Triton X-100, and 5 μg/mL each of leupeptin, pepstain A and aprotinin. The homogenates were centrifuged at 15 000 g for 30 min at 4°C, supernatants were collected and stored at −80°C for use. When necessary, cytosol fractions and nuclear fractions were extracted with some modifications to a previously described procedure (Ogita and Yoneda 1994). Briefly, tissue samples were homogenized in 1.5 mL of 10 mm HEPES, pH 7.9, 0.5 mm MgCl2, 10 Mm KCl, 0.1 mm EDTA, 0.1 mm EGTA, 50 mm NaF, 5 mm DTT, 10 mmβ-phosphoglycerol, 1 mm Na3VO4, 1% NP-40, 1 mm benzamidine and enzyme inhibitors: 5 mg/mL PMSF, 5 mg/mL each of pepstatin A, leupeptin, aprotinin, and then were centrifuged for 10 min at 800 g. Supernatants were centrifuged at 15 000 g for 30 min at 4°C. The nuclear pellets were extracted with 20 mm HEPES, pH 7.9, 20% glycerol, 420 mm NaCl, 0.5 mm MgCl2, 1 mm EDTA, 1 mm EGTA, 1 mm DTT and enzyme inhibitors for 30 min at 4°C with constant agitation. After centrifugation for 15 min at 15 000 g, the supernatant was removed and stored at −80°C until use. The protein concentrations were determined by the method of Lowry et al. (1951) with bovine serum albumin (BSA) as standard.
Immunoprecipitation and western blotting
For immunoprecipitation, tissue homogenates (each containing 400 μg of proteins) were diluted 4-fold with HEPES buffer containing 50 mm HEPES (pH 7.4), 150 mm NaCl, 10% glycerol, 1% Triton X-100, and 1 mm each of EGTA, EDTA, PMSF and Na3VO4. Samples were pre-incubated for 1 h with 20 μL protein A/G then centrifuged to remove any protein adhered non-specifically to the protein A/G. The supernatant was incubated with 2 μg proper antibody or species-relevant non-specific IgG (n.s. IgG) for 4 h at 4°C. After the addition of protein A/G-sepharose, the mixture was incubated at 4°C for an additional 2 h. Samples were triple washed with HEPES buffer and eluted by sodium dodecyl sulfate – polyacrylamide gel electrophoresis (SDS–PAGE) loading buffer then boiled at 100°C for 5 min.
Western blot analysis was carried out on 10% SDS–PAGE according to the method previously described (Gu et al. 2001a). Proteins were electrotransferred onto nitrocellulose filter (pore size, 0.45 μm). After blocking for 2 h in phsophate-buffered saline (PBS) with 0.1% Tween 20 (PBST) and 3% BSA, the membranes were incubated overnight with primary antibody in PBST containing 3% BSA. Detection was carried out by the use of proper alkaline phosphatase conjugated IgG (1 : 20 000) and developed with NBT/BCIP assay kit (Promega). After immunoblotting, the density of the bands was scanned using an image analyzer (LabWorks Software, UVP Upland, CA, USA).
Rats were anaesthetized with chloral hydrate and underwent transcardial perfusion with 0.9% saline followed by 4% paraformaldehyde in 0.1 m PBS. Brains were removed, post-fixed overnight in paraformaldehyde, processed and embedded in paraffin. Coronal brain sections (6-μm thick) were cut on a microtome (Leica RM2155, Nussloch, Germany). Sections were de-paraffinized in xylene and rehydrated in a gradient of ethanol and distilled water. High-temperature antigen retrieval was performed in 1 mm citrate buffer. To block endogenous peroxidase activity, sections were incubated for 2 h in a solution containing 0.01% sodium azide and 0.1% H2O2 in PBS. To reduce non-specific staining, sections were incubated for 2 h in a blocking solution containing 1% BSA, 2% normal goat serum, 0.3% Triton X-100 and 5% non-fat dry milk in PBS. The sections were then incubated in the primary antibodies (anti-phospho-MLK3 or anti-POSH, 1 : 200 diluted in 0.1% BSA, 0.3% Triton X-100 and 1% normal goat serum in PBS) overnight at 4°C. Alternate sections from each brain were incubated without primary antibody as negative controls. After washes three times in PBS, the sections were incubated for 2 h in biotinylated goat anti-rabbit or donkey anti-goat secondary antibody (1 : 200 dilution; Vector Laboratories) made up in 0.1% BSA, 0.3% Triton X-100 and 1% normal goat serum in PBS. Sections were washed and incubated with avidin-conjugated horseradish peroxidase (diluted 1 : 200 in 0.3% Triton X-100) for 1 h at 37°C, followed by avidin–biotin–peroxidase (ABC, Vector Laboratories) at 25°C. The sections were examined by light microscopy.
Rats were perfusion fixed with 10% paraformaldehyde 3 days after ischaemia under anaesthesia. Paraffin sections (6-μm thick) were prepared and stained with Cresyl violet. The sections were examined with a light microscope and the neuronal density of the hippocampal CA1 pyramidal cells was expressed as the number of cells per 1-mm length of the CA1 pyramidal cells counted under a light microscope (× 400).
Four or five independent animals were sampled at each time point for western blotting study and histology examination. Semiquantitative analysis of the bands was performed with the Image J analysis software (Version 1.30v; Wayne Rasband, NIH, USA). All values are expressed as the means ± SD. Statistical analysis of the results was carried out by one-way analysis of variance (anova) followed by the least significant difference (LSD) test or Newman–Keuls test. Differences of p < 0.05 were considered significant.
Protein levels and subcellular localization of POSH after ischaemic insults
To explore the cellular distribution and the expression manner of POSH, we first examined POSH protein by immunohistochemistry observation in the rat hippocampus. Results showed that weak POSH immunoreactivity was seen in the hippocampal DG region, while strong POSH immunoreactivity was seen in the CA1, CA2, and CA3 regions (Fig. 1). As to the subcellular localization of POSH, note that the immunoreactivity was presented mainly in the cytosol but not in the nucleus. The POSH immunoreactivity in cytosol extracts of whole hippocampus was further examined by western blotting at several time points (sham, 10 min, 30 min, 6 h, 1 and 3 days) after 15 min of ischaemia (Fig. 1g). The time course of POSH in western blotting analysis was similar to that in immunohistochemistry; POSH expression did not change significantly.
Ischaemia and reperfusion induced the selective activation of the MLK3/MKK4/JNK cascade and increased the interactions of MLK3, MKK4 and p-JNKs with POSH in the hippocampal CA1 region
It has been well characterized that the hippocampus, particularly the CA1 subfield, is one of the regions most vulnerable to global cerebral ischaemia. As shown in Fig. 2(a and b), and in our previous work (Gu et al. 2001a; Pan et al. 2005), the MLK3/MKK4/JNK cascade was selectively activated in the CA1 region but not in the CA3/DG subfields. To determine whether POSH is involved in the JNK signalling activation in the hippocampal CA1 region induced by ischaemic reperfusion, we examined the biochemical ability of POSH to interact with MLK3, MKK4 and phospho-JNKs (p-JNKs) at various time points (sham, 10 min, 30 min, 6 h, 1 and 3 days) of reperfusion after 15 min of ischaemia. The sample proteins from the hippocampal CA1 regions were immunoprecipitated with antibody against POSH then immunoblotted with antibodies against POSH, MLK3, MKK4 and p-JNKs, respectively. We found that transient brain ischaemia and reperfusion induced rapid and sustained increases in the interactions between POSH and MLK3, MKK4 and p-JNKs, as shown in Fig. 2(c and e), and the interactions were markedly enhanced at 30 min and 1 day of reperfusion, indicating that MLK3, MKK4 and p-JNKs formed a complex with POSH after ischaemic reperfusion. In addition, POSH expression in the CA1 subfield remained unchanged. In reciprocal co-immunoprecipitation experiments, homogenates from the hippocampal CA1 regions at 30 min of reperfusion were subjected to immunoprecipitation with antibodies against POSH, MLK3, MKK4, p-JNKs or non-specific IgGs and the immunocomplexes were probed for the presence of POSH with POSH specific antiserum. As shown in Fig. 2(d), the results confirmed the interaction of POSH with the JNK cascade components, while non-specific IgGs as controls had negligible effects, testifying to their specificity. These results indicated that, as a scaffold protein, POSH was involved in the JNK pathway by interacting with at least MLK3, MKK4 and JNKs in the hippocampal CA1 region following reperfusion.
POSH antisense oligodeoxynucleotide treatment significantly inhibited POSH expression in the hippocampal CA1 region
To investigate the possible relationship between JNK signal activation and POSH protein, we initially examined the alteration of POSH expression after i.c.v. injection of the POSH AS-ODNs using immunohistochemistry. The results showed that POSH AS-ODNs markedly inhibited its protein expression when compared with scrambled ODNs in the rat hippocampal CA1 region both before ischaemia and after reperfusion of 30 min and 1 day (Figs 3a–f). Further western blot analysis using the samples from the hippocampal CA1 region of POSH AS-ODNs treatment rats confirmed the inhibitory effects shown in immunohistochemical analysis (Figs 3g and h). The reduction in POSH protein was not significantly different between the ipsilateral and contralateral CA1 regions (data not shown). It has previously shown that POSH AS-ODNs specifically depress POSH expression in PC12 cell line and sympathetic neurons (Xu et al. 2003), our present results confirmed that POSH AS-ODNs targeted POSH mRNA in vivo.
POSH antisense oligodeoxynucleotide treatment significantly inhibited the increased interactions of MLK3, MKK4 and p-JNKs with POSH and the selective activation JNK cascade in the hippocampal CA1 region
To further examine the contribution of POSH to ischaemia-induced activation of JNK signalling in the hippocampal CA1 region, rats were treated with POSH AS-ODNs and the samples from the hippocampal CA1 region were co-immunoprecipitated with an anti-POSH antibody. The immunoreactivity of POSH-associated MLK3, MKK4, and p-JNKs were examined by western blot analysis. Figure 4(a and b) showed that there was a significant decrease in the expression of MLK3, MKK4 and p-JNK precipitated by POSH at 30 min of reperfusion after AS-ODNs infusion. Additionally, reciprocal immunoprecipitation experiments confirmed the decreased interaction of POSH with these JNK cascade components (data not shown).
Because POSH acts upstream of the MLK family and directly binds MLK3, the coronal brain sections in immunohistochemistry were then stained with anti-p-MLK3 antibody to examine the effect of AS-ODNs on MLK3 activation. As shown in Fig. 4(c and d), strong immunoreactivity of p-MLK3 induced by reperfusion was observed in the hippocampal CA1 region compared with sham control, which was localized mainly in the cytosol but not in the nucleus (Fig. 4e). Treatment of POSH AS-ODNs markedly decreased the selective activation MLK3 in CA1 region compared with missense POSH at 30 min of reperfusion (Figs 4f and g). These observations showed that knock-down of POSH protein could at least down-regulate the selective activation of MLK3, one important partner of POSH, in the rat hippocampal CA1 region. Further western blot analysis using the samples from the hippocampal CA1 region showed that the activation of MLK3, MKK4 and JNKs were simultaneously decreased at 30 min and 3 days of reperfusion, when their activation was still maintained in the rats treated with missense POSH (Figs 4h and i). Together with the aforementioned immunocytochemical findings, these data strongly suggest that POSH appears to function as a scaffold in the multiprotein complex of MLK3–MKK4–JNK that links upstream stimulti and downstream elements of the JNK signalling cascade, by which stress signalling is transmitted to JNK substrates such as c-Jun.
POSH knock-down affected the activation of c-Jun and caspase 3 induced by ischaemia reperfusion
We next addressed whether POSH knock-down inhibited the activation of c-Jun, a nuclear substrate of JNK, during reperfusion. After intracerebroventricular administration of POSH AS-ODNs, nuclear extracts from the CA1 regions were subjected to immunoblotting analysis with anti-p–c-Jun–Ser63 antibody. As shown in Fig. 5(a and c), c-Jun was activated after reperfusion and reached a high level at 6 h of reperfusion, POSH AS-ODNs prevented the increased c-Jun activation induced by ischaemia. In cerebral ischaemia studies, caspase 3 has been shown to play a key role in neuronal death after ischaemia. Our previous study demonstrated that the active cleaved subunits of caspase 3 protease increased dramatically from 1 day after ischaemia and reached its peak level at 3 days of reperfusion (Zhang et al. 2002). Interestingly, knock-down of POSH protein also attenuated the ischaemia-induced increases in activated caspase 3 in the post-ischaemic CA1 region (Figs 5b and d).
Antisense oligodeoxynucleotide inhibition of POSH expression is neuroprotective after cerebral ischaemia in the rat hippocampal CA1 region
To further investigate the potential protective effects against ischaemic injury, the sections from control animals (TE only) and animals treated with POSH AS-ODNs, or POSH missense ODNs in the aforementioned groups, were performed. Cresyl violet staining was used to examine the surviving cells of CA1 pyramidal neurons. Normal cells showed round and pale stained nuclei. The shrunken cells with pyknotic nuclei after reperfusion were counted as dead cells. As shown in Fig. 6, transient cerebral ischaemia followed by 5 days of reperfusion induced severe cell death (Figs 6b, f and i). However, administration of POSH AS-ODNs obviously limited the neuronal degeneration (Figs 6d, h and i). The observed effects on cell viability were similar on the contralateral side to the injection which was not quantified. In contrast, treatment with the non-specific ODNs did not improve the neuronal density (Figs 6c, g and i), suggesting that antisense action was specific for POSH. The results indicate that POSH AS-ODNs is capable of protecting against neuronal injury induced by reperfusion after ischaemia.
Increasing evidence suggests that JNK is an important kinase mediating the neuronal cell death in response to cerebral ischaemia (Nozaki et al. 2001; Irving and Bamford 2002). Histological evidence of degeneration, exhibiting the hallmarks of apoptosis, is not observed until 2–3 days after induction of ischaemia in rats (Pulsinelli et al. 1982; Chen et al. 1996; Rosenbaum et al. 1998). We have previously shown that, in the hippocampal CA1 region where neurons are particularly vulnerable to ischaemic injury, the JNK signal cascade was biphasically activated at 30 min and then 3 days after reperfusion (Gu et al. 2001a). Cerebral ventricular infusion of JNK1/2 antisense not only decreased JNK1/2 protein expression and the activation level but also significantly decreased CA1 pyramidal cell death and DNA fragmentation. Furthermore, we have found that the important protective mechanisms in ischaemic tolerance preconditioning involve the eliminated diphosphorylation of JNK as well as the inhibited MLK3 activation in the CA1 region (Gu et al. 2000; Yin et al. 2005). These observations suggest that the JNK signalling pathway is activated and involved in the selective cell death in the hippocampal CA1 subfield following global ischaemia in this model. In the current study, knock-down of POSH expression using POSH antisense ODNs caused inhibition of the MLK3/MKK4/JNK pathway and enhanced neuronal survival in the hippocampal CA1 region. The results suggest that the increased interaction between POSH and JNK cascade components in response to ischaemic stress is closely related to the activated JNK signalling as a pathogenic event occurring in the vulnerable hippocampal CA1. Thus, POSH protein selectively participated in the neuronal injury after cerebral ischaemia.
We initially found that the endogenous POSH protein expression was located exclusively in the cytoplasm and was not significantly altered between normal controls and rats within several days of reperfusion. As an explanation for this observation, cytoplasmic POSH might induce the JNK signalling pathway by remodelling the signalling cascade components rather than by changing the protein level in the CA1 pyramidal cells. In addition, the basal levels of POSH may be essential for normal growth and development and for the high level of basal JNK activity in the brain. For example, compound mutants lacking Jnk1 and Jnk2 genes are embryonic lethal and lead to neural tube defects (Kuan et al. 1999). Our results also imply that the highly expressed POSH in the hippocampal CA2/CA3 regions may be correlated with other distinct signals, such that the elevated ERK activity may be more predictive of neuroprotection in this region as recently described (Gu et al. 2001b; Zablocka et al. 2003; Wang et al. 2005). It remains uncertain whether POSH participated in other signalling networks as a scaffold.
As previously reported, MLK3 contains an N-terminal SH3 domain, followed sequentially by a kinase domain, a leucine zipper, a Cdc42/Rac interactive binding (CRIB) motif, and a large proline-rich C-terminal region (Gallo et al. 1994). In resting cells, similar to Src tyrosine kinases, MLK3 is auto-inhibited by binding of its own SH3 domain to an auto-inhibitory sequence, a crucial proline residue, between the zipper and CRIB motif, and silences its activity (Zhang and Gallo 2001). Activated Rac or Cdc42 binds the CRIB motif and relieves the negative regulation of the SH3 domain on the MLK3 catalytic domain and then facilitates the dimerization and autophosphorylation, which is required for proper substrate interaction and subsequent phosphorylation of downstream targets (Leung and Lassam 2001). Additionally, recent findings support a model in which apoptotic stimuli or POSH overexpression induces direct association between POSH and inactive MLKs, and that JNK phosphorylation in response to NGF deprivation, which requires MLK activation, is suppressed by POSH AS-ODNs (Xu et al. 2003). Besides acting as a scaffold to facilitate the activation of the JNK pathway, POSH play multiple cellular roles. For example, Akt2 directly interacts and phosphorylates POSH-binding MLK3 and then disassembles the MLK-MKK-JNK complex bound to POSH and down-regulates the JNK signalling pathway (Figueroa et al. 2003). Therefore, it is reasonable that knock-down of POSH protein might affect its direct association with MLK3 and the possible regulation by Akt, and then lead to the decreased interactions with downstream targets. Indeed, in the present study, western blot and co-immunoprecipitation results showed that POSH protein and its bindings with JNK cascade components were down-regulated by antisense treatment in the CA1 region. Similarly, the extent of p-MLK3, p-MKK4 and p-JNK was markedly altered by POSH AS-ODNs delivery. Thus, we conclude that POSH play a positive regulatory role in mediating MKK4-JNK phosphorylation/activation by permitting MLK3 activation following transient global ischaemia.
The formation of distinct signalling complexes is known for regulating the specificity of signal transduction pathways (Pawson and Scott 1997). β-Arrestin 2 is another scaffold protein, which preferentially enhances JNK3 phosphorylation through the interactions with upstream activators ASK1 and MKK4 (McDonald et al. 2000). The JIP group of scaffold proteins were found to selectively mediate signalling by aggregating components of a MAPK module (including MLK, MKK7 and JNK) and facilitate signal transmission by the protein kinase cascade in vivo (Yasuda et al. 1999). JIP1-deficient mice were found to suppress excitotoxic stress, JNK activation and neuronal cell death in post-ischaemic brain (Im et al. 2003). It suggests that JIP1 plays a pivotal role in regulating JNK activation and neuronal survival in response to ischaemia, but is not essential for JNK activation and early development (Whitmarsh et al. 2001; Im et al. 2003). As previously reported in the PC12 cell, POSH may also act on MKK7 in a manner that cooperates to increase JNK activation (Xu et al. 2003). However, the binding of MKK7 with POSH was weakly detected in vivo in this ischaemic model (data not shown). We speculated that MKK7 might be exclusively organized by JIPs in the efficient route of the MLKs/MKK7/JNK signalling cascade. As reported in our recent findings (Li et al. 2005), activated MKK7 markedly binds to JIP-1 during reperfusion which may play a pivotal role in the activation of the JNK signalling pathway in ischaemic injury. However, the biochemical mechanisms underling the specific scaffolds leading to JNK activation in the ischaemic CA1 region are currently unclear.
Once activated, JNK phosphorylates Ser63 and Ser73 residues of c-Jun and increases activator protein-1 transcription activity (Karin 1995; Gupta et al. 1996). Previous studies indicate that c-Jun was specifically phosphorylated in CA1 at an early stage of reperfusion and was closely associated with ischaemia-induced neuronal apoptosis (Matsuoka et al. 1999). Here, we examined c-Jun phosphorylation and observed that POSH AS-ODNs inhibited the level of p-c-Jun, compared with scrambled ODNs at 6 h of reperfusion. It has been shown that global ischaemia induced a marked increase in activated caspase 3 in CA1, and that JNK signalling also targets the mitochondria and regulates the release of cytochrome c (Chen et al. 1998; Tournier et al. 2000; Muller et al. 2004; Dluzniewska et al. 2005). Caspase 3 is not significantly changed in CA3 and dentate gyrus, indicating that the ischaemia-induced changes in gene expression were cell specific. In our present work, POSH AS-ODNs attenuated the ischaemia-induced increases in activated JNK and decreased the cleavage of caspase 3 at the later reperfusion stage, which suggests that the AS-ODNs may also intervenes at the level of apoptotic signalling cascades that lie downstream of JNK signalling. Meanwhile, cell density as estimated by cresyl violet staining analysis showed more survival cells after treatment with POSH AS-ODNs; this implies that POSH might potentate neurotoxicity through remodelling of the JNK signalling cascade and then initiate subsequent cellular events including inducing the activation of c-Jun and caspase 3 that lead to delayed neuronal death in the hippocampal CA1 subfield after ischaemia. The results of the current study are consistent with the notion that JNK signalling can induce neuronal cell death by both transcriptional induction of death-promoting genes and modulation of the mitochondrial apoptosis pathways (Davis 2000). Taken together, our findings suggest that ischaemia-induced elevation in the multiprotein complex facilitated JNK activation and subsequent cellular events, at least through the MLK3/MKK4/JNK cascade.
Our current in vivo study was carried out to explore the possible role of POSH in cerebral ischaemia. We have demonstrated that POSH was implicated in the POSH/MKK4/JNK signalling cascade through increased associations in the vulnerable CA1 region following transient global ischaemia. Meanwhile, downstream events including c-Jun and caspase 3 were involved in this signalling cascade and closely related to neuronal cell death. These results extended the understanding of the pathological role of JNK signalling in ischaemic stress and confirmed the possibility that targeting JNK signalling may offer an effective treatment for ischaemic stroke. Future studies in animal models will be required to mechanistically assess the upstream region of MLK3 such as Rac and identify other binding partners of POSH in post-ischaemic cell death.
This work was supported by a grant from the Key Project of the National Natural Science Foundation of China (no. 30330190).