Repurposing antimycotic ciclopirox olamine as a promising anti-ischemic stroke agent.

Ischemic stroke is a severe disorder resulting from acute cerebral thrombosis. Here we demonstrated that post-ischemic treatment with ciclopirox olamine (CPX), a potent antifungal clinical drug, alleviated brain infarction, neurological deficits and brain edema in a classic rat model of ischemic stroke. Single dose post-ischemic administration of CPX provided a long-lasting neuroprotective effect, which can be further enhanced by multiple doses administration of CPX. CPX also effectively reversed ischemia-induced neuronal loss, glial activation as well as blood-brain barrier (BBB) damage. Employing quantitative phosphoproteomic analysis, 130 phosphosites in 122 proteins were identified to be significantly regulated by CPX treatment in oxygen glucose deprivation (OGD)-exposed SH-SY5Y cells, which revealed that phosphokinases and cell cycle-related phosphoproteins were largely influenced. Subsequently, we demonstrated that CPX markedly enhanced the AKT (protein kinase B, PKB/AKT) and GSK3β (glycogen synthase kinase 3β) phosphorylation in OGD-exposed SH-SY5Y cells, and regulated the cell cycle progression and nitric oxide (NO) release in lipopolysaccharide (LPS)-induced BV-2 cells, which may contribute to its ameliorative effects against ischemia-associated neuronal death and microglial inflammation. Our study suggests that CPX could be a promising compound to reduce multiple ischemic injuries; however, further studies will be needed to clarify the molecular mechanisms involved.

Introduction

Ischemic stroke, an acute disturbance of brain function, is a leading contributor to morbidity and mortality globally. Currently, ischemic stroke has remained essentially an unmet medical need. As a matter of fact, there have been limited advances for the clinical treatment of ischemic stroke, except the recombinant tissue plasminogen activator, which has been proven to be restricted to a small portion of ischemic stroke patients due to a narrow time window and the risk of hemorrhagic transformation.

Alternatively, neuroprotective research has gained significant progress and remains as a promising option for the treatment of stroke, although the road toward achieving satisfactory neuroprotection in stroke patients is still extremely challenging. First, the history of acute ischemic stroke has largely developed over the past 30 years, from a simple vascular disease to a neurovascular disease involving a triad consisting of neuron, microglia, astrocyte, pericyte and endothelial cells, and was lately expanded to be a vascular neural network disease even include the venous system, which further enhances our understanding of the underlying complicated mechanisms of ischemic stroke and the way to conquer it. Second, various drugs with promising multifaceted therapeutic targets have been demonstrated to show promising preclinical or clinical effects against stroke, such as dl-3-n-butylphthalide and statins1, 2, 3, 4, 5, 6. Third, the success of endovascular thrombectomy provides new hope for ischemic stroke therapy and has created great opportunity for better estimating the effectiveness of neuroprotective agents7, 8. Given such challenges and opportunities, it is of great importance and necessity to discover novel and potent neuroprotective agents against ischemic stroke.

Currently, translation of therapies that target post-ischemic deleterious mechanisms is still facing a difficult bottleneck, as massive evidences support the idea that ischemic stroke is a severe disease with multi-facet pathological changes, which includes impaired energy metabolism, altered ion homeostasis, free radical production, and aberrant immune response9, 10. Due to lack of efficient therapeutic target, phenotypic screening is currently expected as one of the most popular strategies for discovery of drugs against the complexity of multifactorial diseases including ischemic stroke11, 12, which may also offer additional opportunity to identify novel multipotent neuroprotective agents13, 14. Here, employing a well-accepted in vitro phenotypic cellular model for ischemic stroke4, 15, we screened 778 drugs approved by the United States Food and Drug administration (Supporting Information Fig. S1A) and luckily identified three drugs with significant protective effects against oxygen glucose deprivation (OGD)-stimulated cytotoxicity in SH-SY5Y cells (Supporting Information Table S1), including the target compound of the present study, ciclopirox olamine (CPX, Fig. S1B).

Accumulating evidence clearly indicates that CPX is a well-known antifungal clinical drug, which is lately demonstrated to possess potential therapeutic effects against other diseases, including cancer, diabetes, and HIV infection. However, to our knowledge, no study was systematically conducted on assessing the effects of CPX on ischemic brain injury, except some evidences indicating its advantage effects on other cardiovascular diseases and neuroprotective effects in vitro17, 18, 19, 20, 21. Here in this study, the off-patent synthetic antimicrobial CPX is used for the very first time to influence the pathophysiology of brain ischemia in a middle cerebral artery occlusion (MCAO) rat model. CPX with multiple doses post-ischemic administration exhibited superior anti-ischemic efficacy, as compare to single dose post-ischemic administration of CPX and positive control, edaravone. CPX acted as potent regulator of protein phosphorylation, especially to kinases and cell cycle proteins in neuronal cells, and was also able to directly ameliorate ischemic insults-induced neuronal damage and glial inflammation.

Materials and methods

Drug administration

CPX (Sigma–Aldrich, St. Louis, MO, USA) was dissolved in dimethyl sulfoxide (DMSO) at 10 mmol/L as stock solution and diluted before drug administration. In the in vivo experiments, CPX powder was dissolved by saline and diluted to the concentrations for intravenous administration. CPX was administered intravenously immediately after withdrawing the monofilament. For the long-term effects of CPX evaluation, single dose of 3 mg/kg of CPX was administrated immediately after withdrawing the monofilament, while multiple doses were first administrated same as the single dose, then continuously administrated every 24 h for 7 days. For the OGD study, cells were incubated with indicated concentrations of CPX when exposed to OGD.

Cell culture, OGD insults and cell viability assay

Details are described in the Supporting Information.

Animals

Male SD (Sprague–Dawley) rats (7–8 weeks old) weighing 230–300 g were obtained from the Animal Center of Shanghai Institute of Materia Medica (Shanghai, China). All the experimental procedures approved by the animal Care and Use Committee of Shanghai Institute of Materia Medica (Shanghai, China). Efforts were made to minimize the pain of animals and reduce the number of experimental animals. Reporting of this work complies with ARRIVE (animal research: reporting of in vivo experiments) guidelines. Animals were housed under 12 h day/night cycle with freely accessible to food and water. A total of 471 rats were included in the study. Data were analyzed on 307 rats, and 164 rats were excluded from the experiment due to death or failure of MCAO surgery. The animal numbers of individual experiments in the detail were listed in Supporting Information Table S2. The animals were randomly assigned into different groups before surgery.

Middle cerebral artery occlusion (MCAO)

The surgery was performed as previously described. Briefly, rats were anesthetized with intraperitoneally administration of 400 mg/kg chloral hydrate (except the cohort study for the effect of CPX administrated 4.5 h after MCAO was anesthetized by isoflurane), then a surgical monofilament nylon suture (Sunbio Biotech Co. Ltd., Beijing, China) was inlet into the left internal carotid artery (ICA) through the external carotid artery (ECA) stump to block the origin of the middle cerebral artery (MCA). The suture was placed there for 2 h resulting in ischemic injury and then removed to induce reperfusion of the MCA. The rats of sham group underwent the same procedures except occlusion of the MCA. The animals were returned to cages after fully awaken.

Neurological function assessment

Neurological function was evaluated at 24 h after reperfusion according to the guidelines of the modified neurological severity scores (mNSS), which was used for assessment of neurological deficits including motor, visual, tactile and reflex responses and proprioception (scale: 0–18, no neurological deficits was scored as 0; maximal neurological deficits was scored as 18). The whole assessment was performed by an experimenter who was blind to the grouping.

Measurement of infarct volume

Infarct volume was measured immediately following neurological function assessment with the methods of TTC (triphenyltetrazolium chloride; Sinophar Chemical Reagent Co. Ltd., Shanghai, China) staining. Briefly, the experimental animals were fully anaesthetized and sacrificed. The brains were quickly frozen for 5 min, and sliced into six coronal sections, and then incubated with 1% TTC solution for 15 min at 37 °C water-bath away from light, then fixed with freshly prepared paraformaldehyde solution (4%) for 24 h. The images of brain sections were taken by digital camera and analyzed using ImagePro Plus software (Media Cybernetics, Silver Spring, MD, USA). The infarct volume was calculated as a percentage according to Eq. (1):

Immunofluorescence

Rats were euthanized and perfused with 4% paraformaldehyde at 24 h after reperfusion. The brains were fixed in 4% paraformaldehyde overnight, then immersed in 20% and 30% sucrose for 3 days for dehydration, and then frozen in the Cryostat Microtome (SM2000R, LEICA, Nussloch, Germany). The frozen brains were sliced into 30 μm-thick sections at the position of bregma approximating 0.5 to −0.5 mm using Cryostat Microtome (Leica, Buffalo Grove, IL, USA). The sectioned tissues were placed on slides. After blocking, the sections were incubated with indicated primary antibodies: anti-GFAP (glial fibrillary acidic protein) (1:100, Cell Signaling Technology, Beverly, MA, USA) or anti-IBA-1 (ionized calcium binding adaptor molecule-1) (1:200, Wako, Richmond, VA, USA) at 4 °C overnight. For the CD31 (platelet endothelial cell adhesion molecule-1, PECAM-1/CD31) and ZO-1 (zonula occludens 1) immunofluorescence, MCAO rats were sacrificed at 72 h after reperfusion and perfused with saline. After fixation with acetone, the brains were cryo-sliced, and the sections were treated with the primary antibodies: anti-CD31 (1:50, BD Biosciences, San Jose, CA, USA), or anti-ZO-1 (1:50, Invitrogen, Rockford, IL, USA) at 4 °C overnight. Sections were then treated with the secondary antibodies for 2 h at room temperature followed by 10 min incubation with DAPI (4′,6-diamidino-2-phenylindole dihydrochloride). After incubation and rinsing, the ipsilateral cerebral cortexes were visualized with Leica SP8 fluorescence microscope (LEICA, Nussloch, Germany).

TUNEL (terminal deoxynucleotidyl transferase dUTP nick end labeling) staining

Cell death was determined by using the In Situ Cell Death Detection Kit (Roche Diagnostics, Mannheim, Germany) according to protocol provided by the manufacturers. Briefly, brain sections were incubated with TUNEL reaction mixture away from light. Then, the sections were incubated with DAPI. After the last wash, images were taken by fluorescence microscopy. The number of TUNEL-positive cells and total cells were calculated in each field. Apoptotic rate was shown as the average percentage of TUNEL-positive cells to the total cells in the three fields selected for each section from ipsilateral cortex.

Evans blue assay

The disruption of blood–brain barrier (BBB) was evaluated by measuring the Evans blue content. MCAO or sham rats at 72 h after surgery were injected with 2% Evans blue (4 mL/kg, Sigma–Aldrich). The rats were euthanized 2 h after Evans blue injection and perfused with saline to fully remove the intravascular dye. Ipsilateral and contralateral hemispheres (without olfactory bulbs or cerebellum) were separated and their weights were recorded. Then the brains were homogenized in 1 mL 50% trichloroacetic acid and centrifuged (10,000×g, 20 min). The concentrations of Evans blue in the supernatants were measured at 620 nm absorbance and then determined with standard curves. The amounts of Evans blue were expressed as μg/g of wet tissue weight.

Rotarod test

The motor coordination of the rats was assessed with the accelerating rotarod (IITC Life Science, CA, USA) as described previously with minor modification. All rats were habituated for 2 days before surgery. The sham or MCAO rats were placed on the accelerating rotarod, and the speed was increased linearly from 2 r.p.m to 20 r.p.m. within 120 s. The latency of the rats to fall from the rotating drum was recorded. Rotarod test was performed at 1, 3, 5, and 7 days post-surgery.

ELISA for TNF-α, IL-1β and IL-6

The cortexes of the 3 brain sections were collected and sonicated by ultrasound homogenization in ice-cold lysis buffer. The supernatants were collected by centrifugation (14,000×g, 15 min). The levels of tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β) and interleukin-6 (IL-6) were measured by the ELISA kits (R&D Biosystems, Minneapolis, MN, USA). The amounts of the inflammatory cytokines were normalized to the protein level of the supernatants.

Phosphoproteomic study

Details are described in the Supporting Information.

Bioinformatics analysis

The phosphoproteomic data analysis was mainly conducted using Excel and R software. The t-test P value and fold change between “OGD” and “OGD+CPX” of each site were calculated and the volcano plot was completed in Excel. Sites with a P value < 0.05 and a minimal fold change of >1.2 or <0.83 were determined to be significantly regulated. Kinases for substrate phosphosites were predicted by NetworKIN 3.0. Only up-regulated phosphosites with a NetworKIN score >3 were reserved for dotplot representation. The gene ontology (GO) biological process enrichment analysis was performed using the STRING database and the result was plotted in R.

Western blot analysis

SH-SY5Y cells were lysed on ice for 30 min with radioimmunoprecipitation assay (RIPA) lysis buffer, then were centrifuged 12,000×g for 15 min. After the centrifugation, the supernatants were collected for further concentration assay and Western blot analysis. For the tissue sample preparation, cortex portions of the 3 brain slices were homogenized in the RIPA lysis buffer using an ultrasound sonicator, and then lysed on ice for 30 min. The supernatants were then collected by centrifugation (12,000×g, 15 min). To detect the level of occludin, the cortex of brain was collected at 72 h after reperfusion and processed as above described. The protein concentrations were determined by bicinchoninic acid Protein Assay Kit (Thermo Scientific, Rockford, IL, USA). Other details are described in the Supporting Information.

Lipopolysaccharide (LPS) stimulation and nitric oxide (NO) assay

The accumulated nitrite released into culture media was measured by Griess reagent (Sigma–Aldrich), which was an indication of microglial production of NO. Briefly, BV-2 cells were seeded into 96-wells and cultured for 24 h to fully attach on the plate. The cells were incubated with different concentrations of CPX for 2 h before stimulation with 100 ng/mL LPS (Sigma–Aldrich). After 24 h stimulation, the culture media was collected and centrifuged to discard the cell debris (500×g, 5 min). Then, cell cultured medium was fully mixed with Griess reagent (50 μL:50 μL) following by 15 min incubation at room temperature away from light. The 540 nm absorbance was measured on the Beckman DTX 800 Multimode Detector (Beckman, Coulter, Fullerton, CA, USA), and a series of concentrations of sodium nitrite was used as a standard for quantification.

Cell cycle analysis

The PI (propidium iodide, Sigma–Aldrich) staining was used to detect cell cycle. After 24 h stimulation with LPS, the cells were fixed with 70% ethanol at −20 °C overnight, following by incubation with RNase A (Sigma–Aldrich), and then stained with PI on ice for 15 min. Cell cycle distribution was then analyzed by flow cytometry (BD Biosciences, San Jose, CA, USA).

PathScan® intracellular signaling array analysis

The expressions of phosphorylated proteins by serine/threonine kinases after treatment with CPX were analyzed by the usage of PathScan® intracellular signaling array kit (Cell Signaling Technology). Briefly, after OGD injury and cultured under normal conditions for 15 min, SH-SY5Y cells were harvested and lysed with the cell lysis buffer on ice. The experiment was performed in strict accordance with the instructions of manufacturer and digital images were taken by the LI-COR Odyssey Imaging system (LI-COR, Lincoln, NE, USA). Then, the RFU intensity of each dot was analyzed by the Image studio software. Finally, a heat map of the average RFU intensity was drawn by the using “pheatmap” package in R software.

Statistical analysis

Data were presented as mean ± SD. Student's t-test was used to determine the statistical significance between two groups. One-way ANOVA (analysis of variance) followed by Turkey's tests were used to analyze the differences of multiple comparisons. The neurological scores were analyzed by Kruskal–Wallis non-parametric test followed by Dunn's multiple comparisons test. For Rotarod test data, two-way repeated measures ANOVA was performed, and then the difference between treatment groups at each time point were analyzed by Bonferroni-corrected post hoc comparisons. Statistical significance was established at P < 0.05.

Results

CPX treatment alleviated brain infarction, neurological deficits and brain edema induced by MCAO

As an extension of in vitro protective effects of CPX on SH-SY5Y cells suffering from OGD (Supporting Information Fig. S1), the in vivo effects of CPX against brain ischemic injury were examined in MCAO rats. Quantitative assessment of TTC-stained brain slices indicated that CPX treatment immediately after reperfusion reduced the infarction volume compared with vehicle group as shown by TTC staining (Fig. 1A and B). CPX (1, 3 and 10 mg/kg) significantly reduced the infarct volume by 13.12%, 22.45% and 11.92% (Fig. 1B). Administration of edaravone (10 mg/kg) also reduced the infarct area by 17.07%. Consistent with the decrease of infarct volume, neurological deficits of MCAO rats were also obviously alleviated by treatment with CPX at 3 and 10 mg/kg (Fig. 1C). Thus, 3 mg/kg of CPX has optimal efficacy against the brain ischemic injury, which was chosen as the appropriate dose for further study. Consistent with above results, CPX at 3 mg/kg dosage also alleviated brain edema shown as reduced brain water content (Supporting Information Fig. S2). Moreover, in order to evaluate the potential therapeutic time window of CPX for ischemic stroke treatment, we further assessed the effect of CPX against ischemic injury when administered at 4.5 h after MCAO surgery. To our delight, CPX administrated at 4.5 h after MCAO surgery (under isoflurane anesthetization) still markedly ameliorated MCAO-induced neurological deficits (Fig. 1D). We further analyzed the long-term efficacy of multiple doses of CPX administrated at 4.5 h, 1 and 3 days after MCAO surgery (under isoflurane anesthetization). As shown in Supporting Information Fig. S3A, CPX administration at 3 mg/kg significantly alleviated MCAO-induced neurological dysfunction [time: F (3,31) = 5.962, P = 0.0025, treatment: F (3,31) = 9.544, P = 0.0042, interaction: F (3,31) = 1.887, P = 0.1523]. In addition, CPX treatment also had a tendency to decrease the mortality rate of the rats suffering from MCAO (log-rank test, P = 0.1281) (Fig. S4B). Above results demonstrated that 4.5 h post-ischemia administration of CPX still protected the animals against the ischemia/reperfusion-induced injuries.

Figure 1

CPX reduced infarct volume and neurological deficits of MCAO rats. (A) Representative TTC staining. (B) Quantification of infarct volumes, n = 16–20 (except sham group n = 5). (C) Effects of CPX on the neurological deficits, n = 16–20 (except sham group n = 5). (D) Effect of CPX administration at 4.5 h after MCAO on neurological deficits tested 24 h after MCAO surgery (Sham, n = 3; Veh: MCAO rats treated with vehicle, n = 7; CPX: MCAO rats treated with 3 mg/kg CPX, n = 6; MCAO surgery was performed under isoflurane anesthetization). ##P < 0.01, ###P < 0.001 versus sham group; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 versus vehicle-treated MCAO group. Veh: vehicle; Eda: edaravone.

Long-term effects of CPX on the motor dysfunction and brain infarction of MCAO rats

The motor functional recovery was further observed to assess the long-lasting anti-ischemic effect of CPX and compare the effects of single dose and multiple doses administration strategy in rats subjected to MCAO. Rotarod assay was performed, and the latency to fall off from the rod was recorded. Strategies of single dose administration of CPX immediately after reperfusion with or without multiple doses of continuous administration of CPX for 7 days were adopted to study the long-term effects of CPX on motor functional recovery (Fig. 2A). In the rotarod test, the latencies to fall off from the rotating rod on days 1, 3, 5, 7 after MCAO surgery were significantly decreased (Fig. 2B) [time: F (4,96) = 80.53, P < 0.0001, treatment: F (1,24) = 231.2, P < 0.0001, interaction: F (4,96) = 63.32, P < 0.0001]. The single dose and multiple doses administration of CPX both significantly alleviated motor dysfunction of the MCAO rats on days 1, 3, 5 and 7 after MCAO surgery (Fig. 2B) [single dose: time: F (4,108) = 105.6, P < 0.0001, treatment: F (1,27) = 18.53, P = 0.0002, interaction: F (4,108) = 3.892, P = 0.0054; multiple dose: time: F (4,104) = 99.73, P < 0.0001, treatment: F (1,26) = 38.85, P < 0.0001, interaction: F (4,104) = 11.66, P < 0.0001]. Moreover, edaravone treatment also ameliorated motor dysfunction of MCAO rats on days 1, 3, 5 and 7 [time: F (4,84) = 74.32, P < 0.0001, treatment: F (1,21) = 22.31, P = 0.0001, interaction: F (4,84) = 5.950, P = 0.0003]. Comparing with that of the single dose group, a significant recovery of motor function was observed in the multiple doses group (Fig. 2B). At the 7th day after MCAO surgery, the latency to fall off from rotarod tended to meet that of the sham group. In consistent to the alleviative effect of CPX on MCAO-induced motor functional impairment, post-surgery administration of CPX (3 mg/kg, multiple doses) also significantly reduced the brain infarction volume at 7 days after MCAO surgery (Fig. 2C and D). Moreover, the final survival rates of CPX administration at single dose and multiple doses administration were 75% and 74%, respectively, which were higher than that of vehicle-treated MCAO group (56%), however, there are no statistical significances between different groups (Supporting Information Fig. S4 and Table S3).

Figure 2

CPX treatment improved motor function of rats at 1, 3, 5 and 7 days after MCAO. (A) Experimental scheme of the single dose and multiple doses administration of CPX. (B) Effects of CPX treatment on the latency fall off the Rotarod of MCAO rats, n = 9–15. (C) Representative TTC staining of multiple doses administration of CPX at 7 days after MCAO. (D) Quantification of infarct volumes of multiple doses administration of CPX at 7 days after MCAO, n = 5–6. ####P < 0.0001 versus sham group; *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001 versus vehicle-treated MCAO group; $P < 0.05 versus single dose group. Veh: vehicle.

CPX alleviated neuronal loss and neuronal apoptosis in MCAO rats

To clarify whether CPX treatment protected the neurons against ischemic injury, immunofluorescence and western blotting were conducted. In the MCAO alone group, a significant loss of neurons was observed in the ipsilateral cortex as shown by the reduced number of NeuN+ cells and decreased level of NeuN compared to that of the sham group. Administration of CPX significantly reversed the MCAO-induced loss of neurons (Fig. 3A and B and Supporting Information Fig. S5). Similarly, TUNEL-positive cells were found in the ipsilateral cortex of vehicle-treated MCAO rats, which was seldomly observed in the sham group. In contrast, treatment with CPX significantly reduced the percentage of TUNEL-positive cells compared to that of the vehicle-treated MCAO group (Fig. 3C and D).

Figure 3

CPX treatment decreased the loss of neurons and apoptosis of neuronal cells in the ipsilateral cortex of MCAO rats. (A) Representative immunostaining for the NeuN in the ipsilateral cortex, scale bar = 100 μm. (B) Western blot for the expression of NeuN, n = 5. (C) Representative images of TUNEL staining. Scale bar = 100 μm. (D) Quantitative analysis of TUNEL-positive cells, n = 5–6. ###P < 0.001 versus sham group, *P < 0.05, ***P < 0.001 versus vehicle-treated MCAO group. M-V: MCAO rats treated with vehicle; M-CPX: MCAO rats treated with 3 mg/kg CPX.

CPX inhibited inflammatory responses in MCAO rats

The effects of CPX on the glial activation and inflammatory cytokines were measured in the ischemic brain. First, we found obvious enhancement in the number of IBA-1 and GFAP positive cells in the ipsilateral cortex of the MCAO rats, suggesting significant glial activations. In contrast, the aberrant change on GFAP positive cells was significantly suppressed by CPX treatment, similarly, CPX treatment had a trend to reduce the number of IBA-1 positive cells (P = 0.06 vs. vehicle-treated MCAO group) (Fig. 4A, B, and Supporting Information Fig. S6). In consistent with the alternations of glial cells, ischemic injury led to a great increase of the TNF-α, IL-1β and IL-6 levels in the ipsilateral cortex of the MCAO rats, whereas the CPX treatment markedly decreased the levels of TNF-α, IL-1β and IL-6 (Fig. 4C). Furthermore, MCAO-induced overexpression of the inflammatory mediator COX-2 was also significantly suppressed by treatment with CPX (Fig. 4C).

Figure 4

Glial cells activation and inflammatory cytokines production were reduced in the CPX treated MCAO rats. Immunostaining for the IBA-1 (A) and GFAP (B) in the ipsilateral cortex, scale bar = 100 μm. (C) The levels of TNF-1α, IL-6, IL-1β (n = 7–10), and COX-2 (n = 4) in the ipsilateral cortex measured by ELISA or Western blotting. ##P < 0.01, ###P < 0.001, ####P < 0.0001 versus sham group, *P < 0 0.05 versus vehicle-treated MCAO group. M-V: MCAO rats treated with vehicle; M-CPX: MCAO rats treated with 3 mg/kg CPX.

CPX ameliorated BBB damage in MCAO rats

The effect of CPX on BBB permeability after ischemic injury was further determined by Evans blue assay. We observed a marked increase in Evans blue staining in the ipsilateral brain of MCAO rats. However, CPX treatment significantly reduced the amount of Evans blue extravasation in the brain of MCAO rats (Fig. 5A and B). Furthermore, we also assessed the alterations of tight junction proteins after ischemic injury by immunofluorescence and western blotting. Double immunofluorescence results showed that the intensities of CD31 (endothelial marker) and ZO-1 were decreased after ischemic injury, while CPX treatment alleviated the downregulated intensity (Fig. 5C). Similarly, the level of occludin in the MCAO rats was also decreased after ischemic injury, and CPX treatment significantly alleviated the loss of occludin protein (Fig. 5D).

Figure 5

CPX attenuated the BBB disruption and the loss of tight junction protein expression in the brain of rats at 72 h after MCAO. (A) Representative Evans blue extravasation image in the brain of MCAO rats. (B) Quantification of the Evans blue extravasation, n = 4–7. (C) Representative image of the double immunofluorescence of CD31 and ZO-1, n = 4, scale bar = 30 μm. (D) Western-blot of occludin, n = 4. #P < 0.05, ##P < 0.01 versus sham group, **P < 0.01 versus vehicle-treated MCAO group. M-V: MCAO rats treated with vehicle; M-CPX: MCAO rats treated with 3 mg/kg CPX.

CPX robustly regulated protein phosphorylation in the OGD-exposed SH-SY5Y cells

As our preliminary results suggested that CPX had a very quick and robust neuroprotective effect against OGD insults, we therefore utilized mass spectrometry (MS)-based phosphoproteomic approach to study the potential molecular mechanism of CPX in SH-SY5Y cells. A multiplexed sample labeling method TMT10plex followed by Ti4+-IMAC-based phosphopeptides enrichment was employed to quantitatively analyze the phosphorylation changes in “OGD” and “OGD+CPX” compared to “CON” (Supporting Information Fig. S7). Totally, 10,459 phosphosites were identified, of which 7036 were highly reliable phosphosites (site localization probability >0.75) with quantification values. By comparing “OGD+CPX” to “OGD”, 130 phosphosites (including 84 up-regulated and 46 down-regulated) in 122 proteins were determined to be significantly regulated by CPX (Fig. 6A). By kinase-substrate prediction analysis of the up-regulated phosphosites, multiple serine/threonine kinases, including cyclin dependent kinase 1/2 (CDK1/2), creatine kinase 1 (CK1), mitogen-activated protein kinase 3 (MAPK3) and PKBα (AKT), seem to be involved in CPX regulation (Fig. 6B). To further investigate the key signaling pathways affected by CPX, we performed the GO biological process enrichment analysis using the regulated phosphoproteins. Cell cycle and chromosome organization are among the top affected processes (Fig. 6C), as also suggested by the potential activation of cell cycle related kinases CDK1/2 (Fig. 6B).

Figure 6

Analysis results from phosphoproteomic study in OGD-exposed SH-SY5Y cells. Cells were collected after OGD exposure followed by another 15 min culture in normal condition. (A) 130 differently regulated phosphosites represented by the volcano plot. It is displayed by –log10 (P-value) versus log2 of the fold change of “OGD + CPX” to “OGD”. Red points: up-regulated sites; blue points: down-regulated sites. (B) The top-scored substrate-kinase prediction result. (C) GO biological process enrichment analysis of changed phosphoproteins. (D) Heat-map of the average RFU intensity of the antibody array. Western blots of phosphorylated AKT (E) and GSK-3β (F) in the OGD-exposed SH-SY5Y cells. ##P < 0.01 versus control group, ***P < 0.001 versus OGD group, n = 4.

According to above results of phosphoproteomics study and the massive literatures about the critical roles of protein phosphorylation in the pathophysiological of ischemic stroke26, 27, 28, 29, it is suggested that CPX treatment may largely influence the protein serine/threonine kinase activity. Therefore, we further explored the detail protein phosphorylations regulated by serine/threonine kinases employing the antibody array kit. The intensity of the antibody array (Supporting Information Fig. S8) was summarized as the heat-map (Fig. 6D and Supporting Information Table S4). The phosphorylations of AKT (Ser473), mammalian target of rapamycin (mTOR) (Ser2448), p70S6K (70-kDa ribosomal protein S6 kinase) (Thr389) and GSK-3β (Ser9) were increased after treatment with CPX in the OGD-stimulated SH-SY5Y cells (Fig. 6D, and Table S4). Employing Western blotting analysis, CPX treatment was verified to significantly enhance the level of phosphorylated AKT in the OGD-exposed SH-SY5Y cells (Fig. 6E). Consistent with the changes of AKT, the phosphorylation of GSK-3β, a down-stream kinase of AKT, was also robustly increased after CPX treatment (Fig. 6F). However, the expression of phosphorylated mTOR and its down-stream p70S6K did not change among control, vehicle-treated, and CPX-treated groups in the Western blotting study (Supporting Information Fig. S9). To further determine the relationship between regulation of AKT phosphorylation and protective effect of CPX on neuronal damage, we employed AKT inhibitor, MK2206 in the current study. As shown in Supporting Information Fig. S10, MK2206 at 2 μmol/L significantly suppressed the ameliorative effect of CPX against OGD-induced reduction in cell viability.

CPX inhibited the cell cycle progression and NO production in LPS-stimulated BV-2 cells

As interfering cell cycle progression was increasing known to play beneficial roles in both neuronal cells and glial cells (reviewed by30, 31), together with our results from phosphoprotein omics analysis, cell cycle analysis was conducted in the LPS-stimulated BV-2 microglia by the flow cytometry of PI staining. After CPX treatment, the LPS-stimulated BV-2 cells were enriched with the S phase, but significantly decreased in the G2 phase, compare with that of the LPS group (Fig. 7A and B). Moreover, LPS induced a prominent release of NO in the medium (P < 0.001 versus control group) of BV-2 cells, which implies the activation of microglia cells, while CPX treatment significantly reduced the LPS-induced overproduction of NO in the BV-2 cells (Fig. 7C).

Figure 7

CPX treatment prevented the cell cycle progression and activation of LPS-stimulated BV-2 cells. Representative plots (A) and statistical results (B) of cell cycle analysis by PI staining detected by Flow cytometry. (C) The concentrations of nitrite in the medium after LPS stimulation for 24 h. ###P < 0.001 versus control group, *P < 0.05, **P < 0.01, ***P < 0.001 versus LPS group, n = 4.

Discussion

Current clinical drugs for the treatment of ischemic stroke are known to inadequately address the underlying complex mechanistic pathogenesis of the disease9, 32. Here, we systematically verified that anti-fungus drug, CPX, exhibits great potential against ischemic cerebral injury, using a classical cerebral ischemia model (MCAO rat model). Our results showed that (i) post-ischemic CPX treatment reduced the brain infarction, neurological deficits and brain edema of MCAO rats; (ii) multiple doses post-ischemic administration of CPX further enhanced the long-lasting anti-ischemic effect; (iii) CPX significantly regulated phosphorylation of multiple proteins, especially AKT and GSK3β phosphorylation, under ischemic insult in neuronal cells; (iv) CPX inhibited cell cycle progression and inflammatory response in microglial cells.

In our in vivo study, we observed that CPX potently ameliorated cerebral ischemia/reperfusion-induced multi-facet damages in the ischemia brain areas, which may extend its clinical usage upon further investigation. As neuronal death (including apoptosis), glial inflammation, and BBB damage in the ischemic brain area are well-accepted evident pathological changes of brain ischemia, the robust effects of CPX against these aberrant changes may indicative of the multi-potent capability against ischemic injury. More interestingly, the improved anti-ischemic effect of CPX shown by multiple doses administration further suggest the possible involvement of different pharmacological mechanisms, which also indicates that better efficacy may be achieved through optimization of the drug administration. In order to assess the effect of CPX on MCAO-induced brain injuries without reperfusion, we administered CPX immediately after MCAO surgery, and assessed the changes of infarction volume and neurological function 24 h after MCAO surgery. To a pity, our data revealed that the CPX didn't confer any protection against the MCAO injury without reperfusion (Supporting Information Fig. S11), indicating the distribution of CPX in the ischemic brain area might be necessary for achieving potent anti-ischemic effects as CPX administration without reperfusion may not arrive in the ischemic brain area. Above promising pharmacological effects of CPX administered after reperfusion drive us to further elucidate the molecular mechanisms of CPX.

Recently, “omics” approaches have increasingly recognized to possess great potential to illustrate an exciting new field of target discovery. Among which phosphoprotein omics has received great attention as which enables high-throughput identification of phosphorylation events on metabolic enzymes, and it has been well employed to understand the pathophysiological changes of multiple diseases and pharmacological mechanisms of active compounds34, 35, 36, 37. Our phosphoproteomic results identified 130 significantly regulated phosphosites upon CPX treatment, and the following GO-analysis indicated that many key molecular functions and biological processes were involved. We therefore chose protein serine/threonine kinases and cell cycle phase transition signaling to further study the molecular mechanisms of CPX, as numerous studies have demonstrated that they are both important cascades involved in ischemia/reperfusion28, 30, 31, 38, 39. Interestingly, we found CPX treatment further enhanced the OGD-induced increase in AKT and GSK3β phosphorylation. Moreover, the neuroprotective effect of CPX against OGD-induced cytotoxicity was markedly weakened after treatment with AKT inhibitor, suggesting that triggering AKT pathway may contribute to the protection of CPX against neuronal damage under ischemic insult. As activation of AKT signaling confer neuroprotection to reduce ischemic damage38, 40, 41, together with the evidences that GSK-3β inhibition play beneficial role on improving outcomes after ischemic stroke and agents that potently inhibit GSK-3β activation were found to alleviate the deleterious consequences of stroke43, 44, our finding that targeting AKT-GSK-3β pathway by CPX could effectively contribute, at least partially, to its potent anti-ischemic effects.

Moreover, considerable evidences demonstrated that aberrant cell cycle reactivation exhibited in both glial cells and neurons following ischemia30, 31. From one aspect, inappropriate cell cycle activation could induce proliferation of astrocyte and microglia and cause inflammatory responses including overproduction of inflammatory cytokines and glial scar formation. From another aspect, delicately control of cell cycle progression is important for neuronal functions as accumulating evidence suggest neuronal cells retain the ability during various neurological injuries including stroke. It is suggested that stroke-stimulated neuronal apoptosis is often associated to the dysregulation of cell cycle. Abnormal phosphorylation of proteins was also found to be involved in the neuronal damage associated with cell cycle activation46, 47, 48. To our knowledge, we for the first time demonstrated the in vivo and in vitro effects of CPX on microglial inflammation. We currently observed that CPX effectively regulated the cell cycle progression of LPS-exposed BV-2 cells, which may contribute to the anti-inflammatory effects of CPX in the cellular model. It is interesting that LPS failed to regulate the cell cycle (Fig. 7) and promote the proliferation (Supporting Information Fig. S12) of BV-2 cells, which suggest that the in vivo and in vitro ischemia-associated microglial inflammation are different. Even though, the regulative effect of CPX on cell cycle and cell proliferation may still contribute to its inhibitory effect on inflammatory responses. However, further studies will be necessary to make this conclusion. Similar to the result in the BV-2 cells, we also observed the significant influence of CPX on the cell cycle progression in OGD-exposed SH-SY5Y cells (Supporting Information Fig. S13). Therefore, together with our phosphoproteomic results about the largely influenced phosphorylation sites associated with mitotic nuclear division and mitotic cell cycle phase transition induced by CPX treatment, the regulative effects of CPX in cell cycle progression in both neuronal and glial cells may contribute to the neuronal damage and microglial inflammation under ischemic insult.

Besides that, other pharmacological mechanisms such as angiogenesis and neurogenesis may also partly interpret the anti-ischemic effect of CPX. First, previous studies have demonstrated that CPX treatment could functionally activate hypoxia-inducible factor 1 and elevate the vascular endothelial growth factor level, which may consequently lead to angiogenesis. Second, it is reported that CPX could increase neurite outgrowth in neuronal cells in the presence of nerve growth factor. As accumulating evidence have already demonstrated the beneficial roles of angiogenesis and neurogenesis against cerebral ischemic damage50, 51, 52, which might additionally interpret the robust in vivo effects of CPX against brain ischemia in rats. Moreover, our findings about the superior long-lasting anti-ischemic effect of multiple doses post-ischemic administration of CPX as compared to single dose post-ischemic administration of CPX may give additional support to the involvement of angiogenesis and neurogenesis.

Despite the uncertainty on the precise mode of action, our in vivo and in vitro multifactorial anti-ischemic results of CPX together with its previously reported good safety and pharmacokinetic profiles16, 53, 54 may shed new light on the discovery of potent clinical drug for the therapy of ischemic stroke. However, further systematic studies will be necessary to further illustrate the full potential of bench to bedside research and translate to clinical usage.

Conclusions

In summary, our data systematically demonstrated that post-treatment with the anti-fungal drug ciclopirox olamine protected the rat brain against MCAO-induced injuries including brain infarction, neurological dysfunction, apoptosis, inflammation as well as BBB disruption. Moreover, CPX regulated protein phosphorylation under ischemic insults in neuronal cells, especially AKT and GSK-3β phosphorylation, and directly modulated cell cycle and inflammatory responses of microglial cells, which may contribute to its in vivo effects on ischemic neuronal damage and microglial inflammation.