Neuroprotective Effects of the Sonic Hedgehog Signaling Pathway in Ischemic Injury through Promotion of Synaptic and Neuronal Health

Cerebral ischemia is a common cerebrovascular condition which often induces neuronal apoptosis, leading to brain damage. The sonic hedgehog (Shh) signaling pathway has been reported to be involved in ischemic stroke, but the underlying mechanisms have not been fully elucidated. In the present study, we demonstrated that expressions of Shh, Ptch, and Gli-1 were significantly downregulated at 24 h following oxygen-glucose deprivation (OGD) injury in neurons in vitro, effects which were associated with increasing numbers of apoptotic cells and reactive oxygen species generation. In addition, expressions of synaptic proteins (neuroligin and neurexin) were significantly downregulated at 8 h following OGD, also associated with concomitant neuronal apoptosis. Treatment with purmorphamine, a Shh agonist, increased Gli-1 in the nucleus of neurons and protected against OGD injury, whereas the Shh inhibitor, cyclopamine, produced the opposite effects. Activation of Shh signals promoted CREB and Akt phosphorylation; upregulated the expressions of BDNF, neuroligin, and neurexin; and decreased NF-κB phosphorylation following OGD. Notably, this activation of Shh signals was accompanied by improved neurobehavioral responses along with attenuations in edema and apoptosis at 48 h postischemic insult in rats. Taken together, these results demonstrate that activation of the Shh signaling pathway played a neuroprotective role in response to ischemic exposure via promotion of synaptic and neuronal health.


Introduction
Ischemic strokes are one of the leading causes of long-lasting disability in humans [1]. Cell death quickly results in neurons starved of oxygen and nutrients, and the subsequent excitotoxicity, oxidative stress, neuroinflammation, and apoptosis produce loss of structural and functional integrity of the brain [2]. Accordingly, novel therapeutic strategies for the restoration of central nervous system (CNS) integrity and promotion of functional recovery for patients with cerebral ischemia are urgently needed.
The sonic hedgehog (Shh) signaling pathway plays an important role in neurogenesis and neural patterning during development of the CNS [3]. When Shh binds to its receptor, Patched (Ptch), it depresses the G protein-coupled receptor Smoothened (Smo), leading to the activation of gliomaassociated oncogene homolog 1 (Gli-1). Activated Gli-1 meditates the expression of many target genes that regulate cell growth, survival, and differentiation of cells, including neurons [4]. The Shh signaling pathway has been reported to be involved in ischemic stroke [5], as Shh expression was found to be upregulated in neurons during ischemia/hypoxia [6]. Moreover, inhibition of the Shh pathway exacerbated ischemic damage in rats, effects which were correlated with a downregulation in the expressions of Gli1 and Ptch [7]. Purmorphamine (PUR), a small molecular agonist of the Shh coreceptor, Smo, exerted protective effects in the middle cerebral artery occlusion (MCAO) model [8]. PUR, which has also been shown to promote blood-brain barrier formation, plays a crucial role in activation of the endogenous anti-inflammatory system within the CNS [9], and recent findings have indicated that PUR protects cortical neurons and restores neurological deficits after ischemic stroke in rats [10]. It has been reported by us that PUR exerted neuroprotection against subarachnoid hemorrhage-induced injury in adult rats [11] and hypoxia-ischemia in neonatal mice [12].
Despite this relatively substantial survey of information on Shh, details regarding the role of Shh signaling in ischemic stroke have not been fully elucidated. As one attempts to rectify this deficit, in this report, we focused on expression levels and effects of the Shh signaling pathway following oxygenglucose deprivation (OGD) injury. In this way, it will be possible to examine some of the underlying neuroprotective mechanisms of the Shh signaling pathway in ischemic injury.   [13]. Briefly, PND1 mice were used to harvest cortical neurons which were then cultured in serum-free neurobasal medium with 2% B27 and 1% penicillin-streptomycin. Cells were cultured in poly-D-lysine-coated twelve-well plates for 7 days and then treated with PUR (20 μM) with/without Cyc (10 μM).

Materials and Methods
To mimic the ischemic condition in vitro, cells were subjected to OGD. Briefly, cells were cultured at 37°C under 95% nitrogen and 5% CO 2 for 6 h in glucose-free DMEM (125 mM NaCl, 2.8 mM KCl, 1.5 mM MgCl 2 , 0.05mM MgSO 4 , 2 mM CaCl 2 , 0.83 mM NaH 2 PO 4 , 24mM NaHCO 3 , and 2 mM HEPES). After OGD, primary neurons were placed in the original neurobasal medium with PUR (20 μM) with/without Cyc (10 μM) and returned to the incubator under normoxic conditions for the times indicated. Control cells were maintained under normal conditions without OGD. Cell apoptosis was assessed by TUNEL staining according to the manufacturer's protocol.

PC12
Cells. PC12 cells were cultured in DMEM containing 5% FBS and 1% penicillin/streptomycin. For MSCexosome treatment, PC12 cells were seeded with FBS-free culture medium in 12-well plates and treated with exosomes (100 μg/mL) from transfected and untransfected MSCs at the times indicated.
To mimic ischemic condition in vitro, PC12 cells were exposed to OGD for 6 h. PC12 cells were then placed in the original medium with or without PUR (20 μM) and with/without Cyc (10 μM) and returned to the incubator under normoxic conditions for the times indicated.

Determination of Apoptosis by Flow Cytometric Analysis.
Apoptosis of PC12 cells was assessed with use of the annexin V-FITC/PI Double Labeling Apoptosis Detection Kit as previously described. The percent of annexin V-positive cells determined over 10,000 acquired events was analyzed with use of a FACS flow cytometer C6 (BD Biosciences, San Jose, CA, USA). All assays were performed in triplicate, and each experiment was repeated three times.
2.4. TUNEL Staining. Cellular death was determined with use of TUNEL staining according to the instructions provided and counterstained with DAPI. The number of TUNELpositive cells was measured in six randomly selected microscopic fields at 200x magnification within the lesion area of each group as described above (N = 4 mice/group).

DHE Analysis.
For determinations of ROS production in primary neurons, cells were stained with 10 μM DHE for 30 min. After rinsing and mounting, images were captured with the use of fluorescent microscopy (BX51; Olympus, Tokyo, Japan). DHE-staining results were pixilated and quantified with the use of the Image-Pro Plus image analysis system.
2.6. Western Blot. Tissues were homogenized with RIPA containing PMSF and protease/phosphatase inhibitors following centrifugation at 4°C at 13,800 × g for 10 min. Then, 5× loading buffer was added to the protein supernatant and total protein was quantified using a BCA assay kit CWBIO (Haimen, Jiangsu, China). Equal amounts of protein were separated by SDS-PAGE and then transferred to PVDF membranes. After blocking in 5% nonfat milk for 2 h, blots were probed using the following primary antibodies: Shh, Gli-1, Patch, p-CREB, CREB, BDNF, p-Akt, Akt, NF-κB, p-NF-κB, and β-actin at 4°C overnight. Secondary antibodies 2 Neural Plasticity were then incubated with the membranes at 37°C for 60 min. The chemiluminescent signal was developed with use of ECL kit reagents (MILLIPORE, USA) and then detected with the use of the Tanon Imaging System (Tanon-4600). Densities of protein bands were semiquantified using ImageJ (National Institutes of Health, Bethesda, MD, USA). 2.8. Immunofluorescent Staining. For immunofluorescent staining, cells were incubated with the primary antibody (Gli-1, 1 : 100) overnight at 4°C and with the secondary antibody at 37°C for 30 min on the following day. The nucleus was stained with DAPI for 10 min. Images were obtained with fluorescent microscopy (OLYMPUS-BX51, Olympus Corporation, Japan), and analyses of these images were performed using the Image-Pro Plus 6.0 software (Media Cybernetics, MD, USA). 3 Neural Plasticity executed to minimize the pain experienced by the animals in these protocols.
MCAO was based on the modified Longa method as previously described [14]. Briefly, for MACO model, rats were anesthetized with isoflurane. The right common carotid artery (CCA), external carotid artery (ECA), and internal carotid artery (ICA) were isolated followed by clamping of the ICA and CCA with microartery clips. The proximal portion of the ECA was ligated using a 5-0 polyester suture and severed at 3.0 mm from the bifurcation of the CCA. The ICA was then completely dissociated, and microsurgical scissors were used to incise a small opening in the arterial wall at 3 mm from the arterial bifurcation at the proximal end of the ECA. A thread embolus was inserted into the ECA parallel with that of the ICA, and the clamp on the ICA was then removed. After achieving microresistance, advancement of the embolus was stopped and the ECA was then tightened with a 5-0 polyester suture. The sham group animals underwent similar surgical procedures without applying the occlusion. The brain infarct was assessed with use of 2,3,5,-triphenyltetrazolium chloride (TTC) staining at 2 days following injury. Neurological functions and brain water content were assessed at 2 days following injury.
2.10. Brain Tissue Water Content Determination. Brain tissues were quickly removed and weighed on an analytical balance with an accuracy of 0.01 mg (wet weight). The hemispheres were then dried in an oven at 105°C for 24 h to obtain the dry weight content [15]. The formula for brain water content was brain water content ð%Þ = ½ðwet weight − dry weightÞ/wet weight × 100%.  Neural Plasticity 1, unable to extend the contralateral forelimb and failure to straighten limb; 2, contralateral forelimb flexion and walking in a circle; 3, leans slightly to the contralateral side and walking in a circle toward the contralateral side; and 4, walking in a circle toward the contralateral side. Animals with scores of 1, 2, or 3 points were selected for the experiment.
2.12. TUNEL Analyses. TUNEL analyses of the brain section were determined as described above. Then, the brain sections were performed using the Image-Pro Plus 6.0 software by an investigator blinded as to experimental group assignments. The brain slices in the region containing the infarct lesion (between -1.60 and -2.00 mm from the bregma) were chosen to undergo TUNEL staining. All the slices of each group used in every independent experiment have the similar anatomical positions. The positive cells were counted within randomly selected peri-infarct areas which limited within 300 μm to the infarct.

Statistical Analyses.
Results are expressed as mean ± SD. Correlation analysis between the expressions of neuroligin/neurexin and TUNEL counts in vitro was performed with Pearson correlation test. Unless otherwise indicated, other data were analyzed using one-way ANOVAs followed by Tukey's test or Dunnett's test for post hoc comparisons using Prism software. A p value < 0.05 was required for results to be considered statistically significant.  (Figure 1(b)). These effects of PUR on the Shh pathway following OGD exposure were blocked by Cyc pretreatment (Figure 1).

Discussion
In the present study, we demonstrate that activation of Shh signals was beneficial for neurological recovery after MCAO in rats. Moreover, we show that these beneficial effects of Shh signaling in response to ischemic exposure were associated with enhanced neuronal viability, increased neuroligin and neurexin expression along with activated Akt, and decreased NF-κB signaling.
Overall, alterations in Shh signaling have been shown to be related with a number of regenerative responses and post-injury pathophysiologies after trauma within diverse regions of the CNS [17]. For example, the Shh pathway is maximally activated at 72 h in response to brain injury followed by a return to baseline levels at 14 days [18] and cortical Shh protein levels are increased at 1 to 5 days after a cortical stab wound injury [19]. There is one report showing that expressions of Shh, Gli-1, and Ptch1 protein were all upregulated in cortical neurons at 6 h after MACO injury in rats [10], while others have reported that Gli1 and Ptch1 expressions were upregulated at 6, 12, 24, and 48 h postischemic injury [20]. However, a downregulation of Shh expression within the cortex has also been reported in the early stages after experimental subarachnoid hemorrhage [11,21] and hypoxia-ischemia in neonatal mice [12]. In the present study, we found that neuronal expressions of Shh and Gli-1 were increased in the early stages of OGD insult, while the expressions of Shh, Gli-1, and Ptch were decreased at later time points following OGD. Activation of the Shh signaling pathway has been shown to increase Bcl-2, while suppressing Bax expression [22]. Moreover, exogenous Shh treatment reduces infarct volume along with promoting angiogenesis and neuronal survival after MACO injury [23]. Our current results demonstrate that restoring expression of the Shh signaling pathway with PUR played a protective role against neurotoxicity after ischemic exposure. That this neuroprotective effect 9 Neural Plasticity did involve a PUR-induced activation of the Shh signaling pathway was substantiated from results obtained with Cyc treatment, which reversed these beneficial effects of PUR upon ischemic exposure. Thus, we concluded that the upregulation of Shh signaling serves to resist ischemic injury in the early stages thereby exerting its beneficial therapeutic effects in cerebral ischemic stroke, while it is then decreased in the later stages due to the aggravation resulting from the injury.
Results from a previous study have demonstrated that PUR exhibits beneficial effects against stroke insult in rodent models and PUR does not alter the stroke-induced level of Shh signaling [10]. PUR plays an antiapoptotic role in the early stage by targeting neurons and also plays an antiinflammatory role in late-stage inflammation by targeting astrocytes [10]. In addition, we found that PUR can reverse the expression of Shh signaling following OGD exposure in neurons in vitro. One explanation for this discrepancy likely comes from the different cell targets of PUR in the CNS following ischemic injury.
The CNS exhibits a substantial degree of plasticity after injury which then enables it to recover from functional deficits via processes involving neurogenesis, angiogenesis, axonal sprouting, and synaptic formation and remodeling. Activation of Shh receptors in the dendrites of hippocampal neurons has been shown to accelerate axonal elongation and synaptic plasticity after ischemic stroke [24]. The Shh pathway can mediate brain plasticity and functional recovery via plasminogen activator which, in part, explains the functional recovery observed after treatment of stroke with bone marrow stromal cells [25]. Moreover, exogenous Shh treatment increases the levels of BDNF and promotes nerve regeneration after cavernous nerve injury [26]. In line with these findings, our present results showed that activation of Shh signaling with PUR increased p-CREB and BDNF expression in response to OGD exposure, effects which were associated with an upregulation in the synaptic proteins, neuroligin and neurexin. The presynaptic neurexin and postsynaptic neuroligin are synaptic cell-adhesion molecules that connect neurons at synapses and regulate synaptic transmission [27]. Ischemia injury has been reported to increase neurexin-neuroligin1-PSD-95 interactions, which may represent an important target for therapeutic agents directed at the treatment of brain ischemia [28]. In the present study, restoring the expression of Shh signaling with PUR upregulated the expressions of neuroligin and neurexin, along with promoting cell survival and modulation of CREB/BDNF pathways. These data suggest that Shh signaling was able to mitigate ischemic injury via affects upon neurotrophic responses and synaptic plasticity.
Phosphorylated Akt is closely related to diverse cellular functions, such as cellular survival, apoptosis, and metabolism. The hedgehog signaling pathway component of Gli-1 can activate the PI3K-Akt pathway [29]. In line with this finding, we demonstrated that PUR activated Akt following OGD exposure in neurons. NF-κB induces the expressions of inflammation cytokines in astrocytes via the Shh signaling pathway [30]. The Shh ligand in the bone marrow microenvironment was involved in promoting NF-κB activity in multiple myeloma cells [31]. We found that PUR decreased OGD-induced phosphorylated NF-κB, associated with antiapoptotic effects. We speculated that the Akt and NF-κB pathways were responsible for PUR's neuroprotective activity, which needs further study.
Collectively, our data indicate that the Shh signaling pathway plays a significant neuroprotective role against ischemic injury by promoting synaptic and neuronal health.

Data Availability
The raw/processed data required to reproduce these findings cannot be shared at this time as the data also forms part of an ongoing study.