Inhibition of Long non-coding RNA zinc finger antisense 1 improves functional recovery and angiogenesis after focal cerebral ischemia via microRNA-144-5p/fibroblast growth factor 7 axis.

Long non-coding RNA zinc finger antisense 1 (ZFAS1) has been probed in cerebral ischemia, while the regulatory mechanism of ZFAS1 in focal cerebral ischemia (FCI) via binding to microRNA (miR)-144-5p remains rarely explored. This study aims to decipher the function of ZFAS1 on FCI via sponging miR-144-5p to modulate fibroblast growth factor 7 (FGF7). The focal cerebral ischemia rat model was established by occlusion of the middle cerebral artery (MCAO) Lentivirus vectors altering ZFAS1, miR-144-5p or FGF7 expression were injected into rats before MCAO. Then, ZFAS1, miR-144-5p, and FGF7 levels were detected, the inflammatory factor level, oxidative stress level, angiogenesis, neurological function injury and neuronal apoptosis were assessed. The binding relations among ZFAS1, miR-144-5p and FGF7 were validated. ZFAS1 and FGF7 expression was elevated, while miR-144-5p expression was reduced in FCI rats. Decreased ZFAS1 or FGF7 or enriched miR-144-5p repressed the inflammatory response, oxidative stress, neuronal apoptosis, while it improved angiogenesis, and neurological function recovery; while up-regulated ZFAS1 exerted opposite effects. The augmented miR-144-5p or silenced FGF7 reversed the effects of enriched ZFAS1. ZFAS1 sponged miR-144-5p that targeted FGF7. Inhibition of lncRNA ZFAS1 improves functional recovery and angiogenesis after FCI via miR-144-5p/FGF7 axis. This study provides novel therapeutic targets for FCI treatment.

Introduction

Cerebral ischemia (CI) is a pathophysiological condition during which the oxygenated cerebral blood flow is insufficient to meet cerebral metabolic demand [1]. While in focal CI (FCI), cell death saliently emerges in the ischemic focus and may further spread to the penumbra, afflicting all cellular elements, including both neurons and supportive cells [2]. CI often induces brain tissue damage, such as neuronal death and cerebral infarction, resulting in high morbidity and mortality worldwide [3]. The major treatment modalities applied in clinics include thrombolysis, tenecteplase, antiplatelets, anticoagulants; moreover, nanoparticle-mediated targeting of organ ischemia injury contributes to improving diagnosis and treatment by targeted delivery. Nevertheless, the translation from animal models to human models and even from early phase to phase III testing still encounter obstacles for practice [4,5]. Therefore, novel therapeutic targets still necessitate extensive exploration.

Long non-coding RNAs (LncRNAs) are highly expressed in the brain and function as the modulators of physiological and pathophysiological processes, including cerebral ischemic injury, neurodegeneration, neural development, and plasticity [6]. Multiple lncRNAs have been revealed to mediate cerebral ischemia progression. For instance, lncRNA metastasis-associated lung adenocarcinoma transcript 1 can facilitate cerebral ischemia-reperfusion injury, acting as a promising therapeutic target for cerebral ischemic stroke treatment [7]. Maternally expressed 8 could regulate angiogenesis, and its knockout effectively relieves CI after ischemic stroke [8]. As a newly identified lncRNA, long non-coding RNA zinc finger antisense 1 (ZFAS1) exhibits a high level in cardiac ischemia/reperfusion injury [9]. Nevertheless, the role of ZFAS1 in FCI remained largely unclear. In addition, it was predicted that ZFAS1 had a binding relationship with microRNA (miR)-144-5p through the bioinformatics website. Numerous miRs in the central nervous system are associated with the nerve lesions and dysfunction after cerebral ischemia [10]. For instance, miR-144 makes a contribution to enhancing preconditioning-mediated neuroprotection against cerebral ischemic stroke [11]. Low-expressed miR-144 has also been validated to exist in rats with myocardial ischemia/reperfusion injury, and miR-144 elevation reduces myocardial infarct size and apoptosis [12]. Similarly, miR-144-3p up-regulation also exerts protective effects on alleviating myocardial ischemia/reperfusion injury in vivo and in vitro via suppressing inflammatory signals [13]. As for miR-144-5p, it has been unveiled to display a low level of spinal cord ischemia [14], yet the function of miR-144-5p on CI was obscure. Moreover, as predicted by the bioinformatics website, there was a targeting relation between miR-144-5p and fibroblast growth factor 7 (FGF7). FGF system affects the pathophysiology of multiple disorders, diseases and cancers, as well as injuries and regeneration [15]. FGF2 expression is elevated in the peri-infarct region after FCI [16], and basic FGF could induce hypoxia and acidification [17]. However, the impacts of FGF7 on FCI were not fully explored.

In light of previous findings, we hypothesized that ZFAS1 might contribute to promoting angiogenesis and functional recovery after FCI via sponge miR-144-5p to target FGF7. This study was conducted to provide a theoretical basis for the effects of the ZFAS1/miR-144-5p/FGF7 axis on angiogenesis and functional recovery after FCI in rats.

Material and methods

Ethics statement

All animal experiments performed in this study were approved by the Ethics Committee in Nanning Second People’s Hospital. All animal experiments were conducted by following per under the Guidelines for the Care and Use of Laboratory Animals issued by the National Institutes of Health.

Laboratory animals

Male Sprague-Dawley rats aged 8–10 weeks and weighed about 250 g were purchased from the Laboratory Animal Center of Guangxi Medical University (Guangxi, China). Before the experiment, the rats were subjected to adaptive feeding for 1 week (supplied with food, water, and 12-h light and night cycle at standard humidity and temperature).

Lentivirus administration

A small hairpin RNA (shRNA) targeting ZFAS1(si-ZFAS1), the full-length cDNA for ZFAS1, miR-144-5p sequence, and FGF7 shRNA sequence (sh-FGF7) were cloned into pLV-EF1a-GFP-Puro plasmid (Biosettia, CA, USA). The lentivirus was packaged and titrated by Shanghai GenePharma Co. Ltd. (Shanghai, China).

Rats were anesthetized with pentobarbital sodium (50 mg/kg) and then fixed onto a stereotaxic frame. Then, by using a 12.5 mm needle that was attached to a 30 cm polyethylene tube matched to a 5-μL Hamilton syringe, 5 μL lentiviral solution (1.54 × 1010 TU/mL) was injected directly into the rat's right lateral cerebral ventricle (Bregma coordinates: 1.5 mm posterior, 1.8 mm lateral) at a rate of 0.5 μL/minute. The needle was kept for 5 minutes. The MCAO rat model was established 14 days after injection [18].

Establishment and grouping of FCI rat model

Rats were subjected to MCAO to establish the FCI model 14 days after the injection of lentivirus solution. First, rats were anesthetized with isoflurane (initial concentration was 3%, then maintained between 1% and 1.5%) in an environment containing N2O and O2 (3:1). The rat rectal temperature was maintained at 37 ± 0.5°C. Then, the skin in the middle of the rat neck was cut open, and the common carotid artery (CCA), internal carotid artery (ICA) and external carotid artery (ECA) were carefully isolated. The middle cerebral artery (MCA) was blocked using nylon sutures that had been previously infiltrated with heparin. The sutures were inserted from CCA into ICA and then to the origin of the MCA. The occlusion lasted for 1 hour; then, the sutures were slowly removed for reperfusion. Thus, the FCI rat model (MCAO group) was established. MCAO was successfully modeled if laser Doppler monitoring (PF5001, Perimed AB, Sweden) confirmed a more than 80% reduction in cerebral blood flow. In addition, the sham-operated group (Sham group) was set as the control group. Only the right middle cerebral artery of rats in the control group was separated, and the incision was not sutured [19].

The rat grouping (18 rats in each group) was classified as follows: sham group, MCAO group, LV-si-negative control (NC) group (injected with lentivirus containing silenced ZFAS1 NC before MACO), LV-si-ZFAS1 group (injected with lentivirus containing silenced ZFAS1 before MACO), LV-miR-NC group (injected with lentivirus containing miR-NC before MACO), LV-miR-144-5p group (injected with lentivirus encoding miR-144-5p before MACO), LV-sh-NC group (injected with lentivirus containing silenced FGF7 NC before MACO), LV-sh-FGF7 group (injected with silenced FGF7 lentivirus before MACO), LV-overexpression(oe)-NC (injected with high-expressed ZFAS1 NC lentivirus before MACO), LV-ZFAS1 group (injected with highexpressed ZFAS1 lentivirus before MACO), LV-ZFAS1 + miR-144-5p group (injected with high-expressed ZFAS1 lentivirus and high-expressed miR-144-5p lentivirus before MACO), and LV-ZFAS1 + sh-FGF7 group (injected with high-expressed ZFAS1 lentivirus and silenced FGF7 lentivirus before MACO)  .

Detection of inflammatory factor level

Rats were anesthetized with pentobarbital sodium (150 mg/kg) and euthanized through decapitation. Their right brain was rapidly isolated on ice as experimental samples. The tissue homogenate was obtained by ultrasonic treatment and then centrifuged at 2000 × g at 4°C for 20 minutes. The kits used for the detection of interleukin (IL)-1β, tumor necrosis factor-α (TNF-α), IL-6 and IL-10were purchased from Shanghai mlbio biotechnology Co., Ltd. (Shanghai, China). The absorbance at 450 nm was examined on a microplate reader and the contents of IL-1β, TNF-α, IL-6 and IL-10 were calculated [20].

Detection of oxidative stress response

The obtained brain tissue homogenate was centrifuged to obtain the supernatant. The protein concentration in the supernatant was determined by the Bradford method, and the activity of superoxide dismutase (SOD) was determined by xanthine oxidase and that of malondialdehyde (MDA) was assessed by thiobarbituric acid. The SOD assay kit and MDA assay kit were purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China) [21].

Evaluation of angiogenesis

The cerebral cortex was isolated from brain tissues, and the rest tissues were discarded. The cerebral cortex was placed in cold phosphate-buffered saline (PBS; Gibco, CA, USA), cut into 1 mm3 chunk and homogenized by ultrasonic treatment. The homogenate was centrifuged for 10 minutes at 500 g at 4°C, and the precipitate was resuspended in 20% bovine serum albumin (Sigma-Aldrich, CA, USA) and centrifuged for 20 minutes at 1000 g at 4°C to observe the microvessels in the bottom. Microvessels were collected into new tubes and washed with PBS. Then, microvessels were detached with 0.1% collagenase II /dispase and DNase I (1000 IU/mL; Gibco) for 45 minutes. The microvessels were centrifuged at 500 g at 4°C for 10 minutes, and the precipitate was resuspended in 10 mL MCDB 131 medium (Gibco), which was pre-added with microvascular growth medium (Invitrogen, CA, USA). The cultured cells were brain microvascular endothelial cells (BMVECs) and the cell suspension was seeded onto a 6-well plate and cultured in humidified air at 37°C (5% CO2, 95% O2). The 4th–6th passage cells with confluence over 95% were screened for tube formation.

Then, 150 μL thawed Matrigel (Corning, NY, USA) was used to coat each well of the 48-well plate. BMVECs (5 × 104 cells/100 μL) were suspended in the above MCDB 131 medium containing a microvascular growth medium, then seeded onto the cell plate coated with Matrigel and incubated for 24 hours. Images were captured from random visual fields by ImageJ software and tube length was captured and assessed [22].

Neurobehavioral score

Rat neurobehavioral scores were assessed using the Rotarod Test (RRT) and modified neurological severity score (mNSS). For RRT, prior to MCAO, rats were consecutively trained for 3 days with 3 times a day and 30-minute intervals each time. The duration of the rotating rod from 0 to 40 rpm was 4 minutes, and the mean value of the maximum duration was set as the baseline. Three tests were performed on the 1st, 7th and 14th after MCAO. The best score on that day was recorded [23].

For mNSS, four tests were performed on the 0th, 1st, 7th and 14th after MCAO. Tests were performed by trained researchers who knew nothing about the experimental conditions. The scoring criteria for mNSS were described in the literature [24]. Balance and reflex and abnormal motion assessment were included. Score 0 meant normal, while scores 18 represented the maximal defect level [25].

Hematoxylin-eosin (HE) staining

The brain tissue obtained above was fixed with 4% paraformaldehyde for 2 hours, then immersed in distilled water for 4 hours, dehydrated with gradient ethanol, and then permeabilized with xylene. Finally, the tissues were embedded with paraffin and cut into 5 μm slices. After being treated with polylysine, the slices were dewaxed and stained with hematoxylin for 5 minutes. After staining, the slices were differentiated with 1% hydrochloric acid alcohol for 20 seconds, returned to blue with 1% ammonia water for 30 seconds and counter-stained with eosin for 5 minutes. Finally, the slices were cleaned with xylene, sealed with neutral glue, and observed under an optical microscope [26].

Transferase-mediated deoxyuridine triphosphate-biotin nick end labeling (TUNEL) staining

The paraffin sections were stained by an in situ cell death detection kit (Roche, Mannheim, Germany). The nuclei were stained with DAPI (blue), and the apoptotic cells appeared red. The apoptotic cells were randomly counted in five visual fields in the rat cerebral cortex by a fluorescence microscope.

Dual luciferase reporter gene assay

The luciferase reporters containing ZFAS1 and FGF7 3ʹ untranslated region were designed and synthesized by Guangzhou RiboBio Co., Ltd (Guangzhou, China). The binding sites of miR-144-5p sequences were included in the luciferase reporters. HEK293T cells (American Type Culture Collection, MD, USA) were cultured until the cell confluence reached 40 ~ 60% and then seeded. The luciferase reporter gene vectors were transfected into HEK293T cells with miR-144-5p-mimic and its NC-mimic using Lipofectamine 2000 transfection reagent (Invitrogen). After 48-h transfection, luciferase activity was examined by dual-luciferase reporter gene detection system (Promega, Madison, WI, USA) [27].

RNA immunoprecipitation (RIP) assay

The binding relation among ZFAS1, and miR-144-5p was validated using the RIP test kit (Millipore, MA, USA). Cells were washed with pre-cooled PBS, and then lysed on ice with an equal volume of radio-immunoprecipitation assay cell lysis buffer (Beyotime, Shanghai, China) , then centrifuged . Cell extracts were divided into two parts for inputs and co-precipitation, respectively. During co-precipitation with antibodies, 50 μL magnetic beads were resuspended in 100 μL RIP wash buffer, and incubated with  anti-Ago2 (Abcam, MA, USA) and anti-immunoglobulin G (IgG) (1:5000; 7074; Cell Signaling Technology, MA, USA), respectively. After that, the magnetic bead-antibody complex was washed and resuspended in 900 μL RIP wash buffer followed by overnight incubation at 4°C with 100 μL cell extract. The purified complex was collected, processed with protease K, and followed by reverse transcription quantitative polymerase chain reaction (RT-qPCR) [28].

RT-qPCR

Trizol reagent (Invitrogen) was used to extract total RNA. Then, ZFAS1 and FGF7 were reversely transcribed using the RevertAid first chain complementary DNA (cDNA) synthesis kit (Thermo Fisher Scientific, Massachusetts, USA), respectively. The miRcute Plus miRNA First-Strand cDNA Kit and miRcute Plus miRNA qPCR Kit (SYBR Green) (Tiangen) were used to detect miRNA expression based on RT-qPCR method [29]. RT-qPCR analysis was performed in ABI Step one plus real-time PCR system (Applied Biosystems, CA, USA) using TB Green Advantage qPCR premix solution (Takara, Shiga, Japan). The relative expression of ZFAS1, miR-144-5p and FGF7 was calculated by 2−ΔΔCt method, and the specific primer sequences were listed in Supplementary Table 1.

Western blot assay

Rat brain tissue was homogenized in Laemmli lysis buffer containing protease inhibitors (Sigma-Aldrich). The protein concentration in the sample solution was assessed using the PierceTM bicinchoninic acid protein quantification kit (Thermo Fisher Scientific), and the sample was stored at −80°C. An equal amount of sample  was taken for 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis. After electrophoresis, the sample was transferred to polyvinylidene fluoride (PVDF) membrane (Millipore) and sealed for 2 hours with 5% skimmed milk powder. PVDF membrane was incubated with anti-FGF7 (1:1000; Abcam), anti-β-actin (1:5000; Abcam) and IgG (1:2000; Abcam) that coupled with horseradish peroxidase. After incubation, the protein blots were observed via the enhanced chemiluminescence system (GE Healthcare, MA, USA). The gray value was quantified via Image J software (National Institutes of Health, MA, USA) [30,31].

Statistical analysis

SPSS 21.0 software (IBM, NY, USA) was used to analyze the data, and the results were expressed as mean ± standard deviation. Each experiment was repeated at least 3 times. t-test was used for comparison between two groups, and one-way analysis of variance (ANOVA) and Tukey’s post test were adopted for comparison among multiple groups. Duplicate analysis of variance was used for data comparisons of different timepoints in groups. Data were considered to be statistically significant when P < 0.05.

Results

This research hypothesized that ZFAS1 might contribute to promoting angiogenesis and functional recovery after FCI via sponge miR-144-5p to target FGF7. To validate the hypothesis, the FCI rat model was established. Thereafter, the expression of ZFAS1, miR-144-5p and FGF7 was detected in the FCI rats, in which ZFAS1 and FGF7 were elevated, while miR-144-5p was depleted. Thereafter, it was manifested that the inhibition of ZFAS1 or FGF7, or overexpression of miR-144-5p promotes angiogenesis and neurological function recovery after FCI, with reduced levels of inflammatory response and oxidative stress. Up-regulation of ZFAS1 promoted the development of inhibited angiogenesis and neurological function recovery, thus promoting FCI development, while up-regulation of miR-144-5p or down-regulation of FGF7 reversed this effect of elevated ZFAS1. ZFAS1 sponged miR-144-5p that targeted FGF7. This study was designed to provide a theoretical basis for the effects of ZFAS1/miR-144-5p/FGF7 axis on angiogenesis and neurological function recovery after FCI in rats.

Neurological function injury induced by FCI in rats

Experiments were conducted in FCI rat models. The outcome of detection of cytokines suggested that the content of IL-1β, TNF-α, IL-6, and MDA was elevated in the brain tissue of MCAO rats, while that of anti-inflammatory factor IL-10 was reduced and SOD activity was impaired (Figure 1a-c).

Figure 1.

Neurological function injury induced by FCI in rats. (a), inflammatory factor level in MCAO rats were detected by ELISA; (b/c), SOD and MDA levels in MCAO rats were detected by ELISA; (d), the tube formation ability was assessed; (e/f), the neurological function injury was tested by rotarod test and mNSS score; (g), the neuronal injury in cerebral cortex was examined by HE staining; (h), apoptosis rate in cerebral cortex was detected by TUNEL staining. The data were presented as mean ± standard deviation. n = 6; # P < 0.05 vs. the Sham group.

Evaluation of angiogenesis revealed a decrease in the number of branches as well as the length of tubes of MCAO rats (Figure 1d).

The neurobehavioral score indicated that the time on the rotarod was shortened, while the mNSS score was high after MCAO modeling, implying that the neurological function was damaged in MCAO rats (Figure 1e).

In HE staining, it was observed that the neurons in the cerebral cortex of sham-operated rats were orderly arranged with complete structure and uniform staining. However, the neurons in the ischemic cerebral cortex exhibited enlarged peripheral space, sparse arrangement, deformation, degeneration and necrosis, and even the disappearance of nucleoli in MCAO rats (figure 1f). The result of TUNEL staining revealed the high apoptosis rate of neurons in the cerebral cortex of MCAO rats (Figure 1g).

These outcomes above unveiled that MCAO induced neurological function injury.

Down-regulated ZFAS1 promotes angiogenesis and neurological function recovery in FCI rats

A previous study has shown that ZFAS1 exhibits a high level of myocardial infarction; while silenced ZFAS1 can partially eliminate a portion of cardiomyocyte apoptosis induced by ischemia [32]. We first detected ZFAS1 expression in sham-operated rats and MCAO rats through RT-qPCR, which implied that ZFAS1 was highly expressed in MCAO rats (Figure 2a).

Figure 2.

Down-regulated ZFAS1 promotes angiogenesis and neurological function recovery in FCI rats. (a/b), ZFAS1 expression level was detected by RT-qPCR; (c), inflammatory factor level after the down-regulation of ZFAS1 was examined by ELISA; (d/e), SOD and MDA levels after the down-regulation of ZFAS1 were examined by ELISA; (f), the tube formation ability after the down-regulation of ZFAS1 was assessed; (g/h), the neurological function injury after the down-regulation of ZFAS1 was tested by rotarod test and mNSS score; (i), the neuronal injury in cerebral cortexafter the down-regulation of ZFAS1 was examined by HE staining; (j), apoptosis rate in cerebral cortex after the down-regulation of ZFAS1 was detected by TUNEL staining. The data were presented as mean ± standard deviation.n = 6; * P < 0.05 vs. the Sham group; # P < 0.05 vs. the LV-si-NC group.

To investigate the effect of ZFAS1 on angiogenesis and neurological function recovery in FCI rats, lentivirus vectors LV-si-NC and LV-si-ZFAS1 were injected into FCI rats, and the result of RT-qPCR indicated that ZFAS1 expression was ablated after injection of LV-si-ZFAS1 (Figure 2b).

In MCAO rats with down-regulated ZFAS1, it was observed that IL-1β, TNF-α, IL-6 and MDA contents were decreased, while IL-10 content and the activity of SOD were elevated; the number of branches and the length of tube formation were extended; time on the rotarod was prolonged and mNSS score was lowered; the neurons in rat cerebral cortex were orderly arranged, and the apoptosis rate of the rat cerebral cortical cells was depleted (Figure 2c-j).

These outcomes suggested that the down-regulation of ZFAS1 facilitated angiogenesis and neurological function recovery in FCI rats.

ZFAS1 binds to miR-144-5p

It has been unraveled that ZFAS1 is mainly distributed in the cytoplasm and can function as ceRNA to sponge miR [33,34]. The presence of binding sites between ZFAS1 and miR-144-5p was predicted by the bioinformatics website Starbase (http://starbase.sysu.edu.cn) (Figure 3a). Thereafter, dual-luciferase reporter gene assay and RIP assay were carried out to further validate the binding relation. It turned out that the luciferase activity of ZFAS1-WT was impaired after transfected with miR-144-5p-mimic in dual-luciferase reporter gene assay (Figure 3b). RIP experiment showed that the enrichment of ZFAS1 and miR-144-5p was augmented under Ago2 treatment (Figure 3c). By using RT-qPCR, it was found that in MCAO rats injected with LV-si-ZFAS1, miR-144-5p expression was enhanced (Figure 3d).

Figure 3.

ZFAS1 binds to miR-144-5p. (a), the binding sites between ZFAS1 and miR-144-5p were predicted by bioinformatic website; (b), the binding relation between ZFAS1 and miR-144-5p was validated through dual luciferase reporter gene assay; (c), the enrichment level of ZFAS1 and miR-144-5p was detected by RIP assay; (d), miR-144-5p expression after the down-regulation of ZFAS1 was examined by RT-qPCR. The data were presented as mean ± standard deviation. N = 3; # P < 0.05 vs. the LV-si-NC group.

These results showed that ZFAS1 had binding relation with miR-144-5p.

Up-regulation of miR-144-5p facilitates angiogenesis and neurological function recovery in FCI rats

MiR-144-5p has been verified to be low-expressed in renal ischemia-reperfusion injury and silenced miR-144-5p accelerates apoptosis [35]. In light of this, we first examined miR-144-5p expression in sham-operated rats and MCAO rats, and the results suggested that miR-144-5p displayed low level in MCAO rats (Figure 4a).

Figure 4.

Up-regulation of miR-144-5p facilitates angiogenesis and neurological function recovery in FCI rats. (a/b), miR-144-5p expression was detected by RT-qPCR; (c), inflammatory factor level after the up-regulation of miR-144-5p was examined by ELISA; (d/e), SOD and MDA levels after the up-regulation of miR-144-5p were examined by ELISA; (f), the tube formation abilityafter the up-regulation of miR-144-5p was assessed; (g/h), the neurological function injury after the up-regulation of miR-144-5p was tested by rotarod test and mNSS score; (i), the neuronal injury in cerebral cortex after the up-regulation of miR-144-5p was examined by HE staining; (j), apoptosis rate in cerebral cortex after the up-regulation of miR-144-5p was detected by TUNEL staining. The data were presented as mean ± standard deviation. n = 6. * P < 0.05 vs. the Sham group; # P < 0.05 vs. the LV-miR-NC group.

Afterward, LV-miR-NC and LV-miR-144-5p were injected into MCAO rats, and the outcome of RT-qPCR revealed that miR-144-5p expression was amplified after injection with LV-miR-144-5p (Figure 4b). In MCAO rats with augmented miR-144-5p, the contents of IL-1β, TNF-α, IL-6 and MDA were decreased, while IL-10 content was elevated and SOD activity was facilitated (Figure 4c-e); number of branches and length of tubes were increased (Figure 4f); time on rotarod was prolonged and mNSS score was reduced (Figure 4g, h); the deformed, denatured and necrotic neurons with sparse arrangement were depleted in the cerebral cortex (Figure 4i); the apoptosis rate of cerebral cortical cells of rats was also ablated (Figure 4j).

The results uncovered that the amplification of miR-144-5p improved angiogenesis and neurological function recovery in FCI rats.

MiR-144-5p targets FGF7

It has been reported that the high expression of FGF7 in Alzheimer’s disease can promote the occurrence of the inflammatory response of neurons and improve the apoptosis rate [36]. In this study, the results of RT-qPCR and Western blot assay indicated that FGF7 was enriched in MCAO rats (Figure 5a-b)

Figure 5.

MiR-144-5p targets FGF7. (a/b), FGF7 expression in MCAO rats was detected by RT-qPCR and Western blot assay; (c), the targeting relation between miR-144-5p and FGF7 was predicated by TargetScan; (d), the targeting relation between miR-144-5p and FGF7 was validated by dual luciferase reporter gene assay; (e-h), FGF7 expression after the down-regulation of ZFAS1 or up-regulation of miR-144-5p was assessed by RT-qPCR and Western blot assay. The data were presented as mean ± standard deviation. N = 3; n = 6; * P < 0.05 vs. the Sham group; # P < 0.05 vs. the LV-si-NC group; % P < 0.05 vs. the LV-miR-NC group.

The targeting relation between miR-144-5p and FGF7 was predicted by the bioinformatics website TargetScan (http://www.targetscan.org/vert_72/) (Figure 5c). Similarly, the dual-luciferase reporter gene assay manifested that the luciferase activity was inhibited after co-transfection with FGF7-WT and miR-144-5p-mimic (Figure 5d).

LV-si-NC, LV-si-ZFAS1, LV-miR-NC and LV-miR-144-5p were then injected into MCAO rats, it came out that FGF7 expression was constrained after the silencing of ZFAS1 or the amplification of miR-144-5p (Figure 5e-h).

These discoveries evidenced that miR-144-5p could target FGF7.

Downregulation of FGF7 promotes angiogenesis and neurological functionrecovery in FCI rats

LV-sh-NC and LV-sh-FGF7 were then injected into MCAO rats, and the outcomes of RT-qPCR and Western blot assay reflected that FGF7 was ablated after the injection of LV-sh-FGF7 (Figure 6a, b). It was observed that silenced FGF7 reduced the inflammatory and oxidative stress responses, increased tube formation ability, partially improved the neurological function recovery, and decreased the apoptosis rate. Moreover, the abnormal neurons were reduced and neurons displayed normal arrangement in the cerebral cortex after the deletion of FGF7 (Figure 6c-j).

Figure 6.

Downregulation of FGF7 promotes angiogenesis and neurological function recovery in FCI rats. (a/b), FGF7 expression after the down-regulation of FGF7 was detected by RT-qPCR and Western blot assay; (c), inflammatory factor level and oxidative stress level after the down-regulation of FGF7 were examined by ELISA; (d/e), SOD and MDA levels after the down-regulation of FGF7 were examined by ELISA; (f), the tube formation abilityafter the the down-regulation of FGF7 was assessed; (g/h), the neurological function injury after the down-regulation of FGF7 was tested by rotarod test and mNSS score; (i), the neuronal injury in cerebral cortex after the down-regulation of FGF7 was examined by HE staining ; (j), apoptosis rate in cerebral cortex after the down-regulation of FGF7 was detected by TUNEL staining. The data were presented as mean ± standard deviation. n = 6. # P < 0.05 vs. the LV-sh-NC group.

The results above implied that the ablation of FGF7 facilitated angiogenesis and neurological function recovery in FCI rats.

ZFAS1 inhibits angiogenesis and neurological function recovery in FCI rats via sponging miR-144-5p to target FGF7

Finally, to verify that ZFAS1 could exacerbate FCI via sponging miR-144-5p to target FGF7, rats were injected with LV-oe-NC, LV-ZFAS1, LV-ZFAS1 + miR-144-5p, LV-ZFAS1 + sh-FGF7, respectively. The result of RT-qPCR and Western blot assay implied that the elevated miR-144-5p or silenced FGF7 reversed the effects of enriched ZFAS1 on increasing FGF7 expression (Figure 7a-b).

Figure 7.

ZFAS1 inhibits angiogenesis and neurological function recovery in FCI rats via sponging miR-144-5p to target FGF7. (a/b), FGF7 expression  was detected by RT-qPCR and Western blot assay; (c), inflammatory factor level    were examined by ELISA; (d/e), SOD and MDA levels   were examined by ELISA; (f), the tube formation ability   was assessed; (g/h), the neurological function injury   was tested by rotarod test and mNSS score; (i), the neuronal injury in cerebral cortex   was examined by HE staining; (j), apoptosis rate in cerebral cortex   was detected by TUNEL staining. The data were presented as mean ± standard deviation. n = 6. * P < 0.05 vs. the LV-oe-NC group; # P < 0.05 vs. the LV-ZFAS1 group.

In the same series of experiments, ZFAS1 augmentation exacerbated the degree of brain injury in MCAO rats, constrained angiogenesis and neurological function recovery, while the enriched miR-144-5p or silenced FGF7 reversed the adverse effects of ZFAS1 overexpression (Figure 7c-j).

In conclusion, ZFAS1 restrained angiogenesis and neurological function recovery in FCI rats via sponging miR-144-5p to target FGF7.

Discussion

FCI is termed as the loss or deletion of local cerebral blood flow in a specific vascular territory, such as the middle cerebral artery [37]. This research focused on the regulatory mechanism of ZFAS1 on angiogenesis and neurological function recovery after FCI. Collectively, it was manifested that ZFAS1 could hinder angiogenesis and neurological function recovery in FCI rats via binding to miR-414-5p to target FGF7.

Initially, this research uncovered that ZFAS1 exhibited a high level in MCAO rats. A similar expression trend of ZFAS1 has also been reported in previous studies. For instance, Wang et al. have elucidated that ZFAS1 is robustly expressed in rats after traumatic brain injury [38]. Rats with ischemic cardiac injuries also display elevated ZFAS1 [39]. To gain the insights of ZFAS1 function in FCI progression, ZFAS1 was up- or down-regulated, and it was disclosed that ZFAS1 silencing could dampen the progression of FCI via improving the neurological function recovery and angiogenesis, and repressing inflammatory response, oxidative stress and apoptosis in FCI rats, while ZFAS1 overexpression exerted opposite effects on FCI progression. In line with our findings, some studies have also reported the similar effects of ZFAS1 on other ischemia-related diseases. For instance, Huang et al. have revealed that ZFAS1 displays an upward trend in the in vitro model of cardiac ischemia/reperfusion injury; while the ischemia/reperfusion-induced cardiomyocyte apoptosis is decelerated after ZFAS1 knockout [9]. As similarly reported by Jiao et al., the deficiency of endogenous ZFAS1 partially abrogates the ischemia-induced apoptosis of cardiomyocytes, while the amplification of ZFAS1 induces impaired cardiac function [32]. Additionally, in mice with epilepticus-induced hippocampal neurons injury, it has been uncovered that ZFAS1 knockdown can inhibit hippocampal neurons apoptosis and autophagy [40]. Our study specifically demonstrated that ZFAS1 deficiency contributed to promoting the activity of SOD yet reducing MDA content. In consistent with the finding, a study of myocardial infarction has reported that ZFAS1 knockdown results in the reduction of MDA content but the activation of SOD activity, thus relieving hypoxia/reoxygenation-induced injury [41]. As for the repressive effects of silenced ZFAS1 on the inflammatory response, Feng et al. have also unveiled that, in traumatic brain injury, the ablation of ZFAS1 effectively retards inflammatory response and facilitates neurological function recovery, thus relieving traumatic brain injury in modeled mice [42]. We then further demonstrated that ZFAS1 overexpression could induce deteriorated FCI. Identically, it has been reported ZFAS1 triggers the aggravation of ischemia-induced cardiac function impairment [39]. Moreover, Yuan et al. have reported that ZFAS1 is associated with bupivacaine-induced neurotoxicity [43]. Compared with such a finding, our research further demonstrated the repressive role of ZFAS1 in FCI development with the combination of miR- miR-144-5p/FGF7 axis, expanding the internal mechanism of ZFAS1 in neurological function.

Thereafter, it was proved that there exists a binding relation between ZFAS1 and miR-144-5p. Then, this study uncovered that miR-144-5p was low-expressed in MCAO rats. Li et al. have manifested that miR-144-5p also exhibits a low level in spinal cord ischemia/reperfusion injury and is involved in damage evolution [15]. After the upregulation of miR-144-5p, the FCI development was decelerated as reflected by attenuated neurological dysfunction, decelerated tube formation and decreased apoptosis rate in MCAO rats. As similarly reported by Xu et al., the augmentation of miR-144-5p results in decreased apoptotic cells in rats with renal ischemia-reperfusion injury [44]. Furthermore, this study also illustrated that miR-144-5p induction could reduce the content of inflammatory factors IL-1β, TNF-α, IL-6 in FCI rats. In line with our finding, Zhou et al. have elucidated that miR-144-5p overexpression saliently hinders the expression of TNF-α, IL-6 and IL-8 in macrophages [45].

Moreover, it was also predicted that miR-144-5p had a targeting relation with FGF7, which was highly expressed in MCAO rats. Furthermore, FGF7 depletion could facilitate neurological function recovery and angiogenesis, yet reduce the content of inflammatory factors and apoptosis cells in MCAO rats. A similar upward expression trend of FGF7 has also been validated by Peng et al. in retinal ganglion cells, and they further unveiled that the repression of FGF7 exerts neuroprotective effects on the activation of glaucoma-related retinal glial cells and apoptosis of ganglion cells [46]. Moreover, as for its effects on the inflammatory response, it has been evidenced that FGF7 depletion can mitigate acute inflammatory nociceptive responses in the dorsal root ganglion neurons after peripheral nerve injury [47]. In addition, FGF7 also displays high expression in all cerebrospinal fluid samples, and the low-expressed FGF7 can ameliorate the Aβ-induced inflammation and apoptosis of SH-SY5Y cells in Alzheimer’s Disease [35].

Conclusion

Collectively, this study manifests that ZFAS1 and FGF7 expression is enriched, while the miR-144-5p level is decreased in MCAO rats. ZFAS1 aggravates the neurological dysfunction and tube formation ability in FCI rats via sponging miR-144-5p to target FGF7. The current discovery makes a contribution to introducing novel therapy targets for FCI by highlighting the importance of ZFAS1/miR-144-5p/FGF7 axis. However, as for clinical practice, there were still many obstacles that remained to be addressed. Affected by the complicated regulatory mechanism of ZFAS1/miR-144-5p/FGF7 axis and the poor practice experiences, how to apply this approach to the FCI treatment in an effective and quantitative way to optimally minimize side effects still requires further investigation, and how to solve such limitation would be a key study point in our future works.

Click here for additional data file.