Primary human brain microvascular endothelial cells were purchased from Cell Systems Corp., (cat. no. ACBRI 376; Kirkland, WA, USA) and cultured in CSC Complete Medium, which includes 10% fetal bovine serum (cat. no. 4Z0-500; Cell Systems Corp.), in a humidified atmosphere (relative humidity >95%) containing 5% CO₂ at 37°C. To avoid undue depletion of glucose in the culture medium, experiments were performed 24 h after the final medium renewal. To simulate ischemia-like conditions in vitro, cells were subjected to OGD as previously described (12). Cells were exposed to OGD for 4 h by replacing the culture medium with a glucose- and serum-free medium that had been equilibrated in an anaerobic atmosphere (<0.1% O₂, 5% CO₂ and 95% N₂) at 37°C inside a hypoxic workstation (H35; Don Whitley Scientific Ltd., Shipley, UK). The pH of the medium remained constant throughout the experiments. As reoxygenation consisted of a return to the initial culture conditions after OGD, cells were withdrawn from the hypoxic workstation and the OGD medium was replaced with glucose- and serum-containing culture medium. Cells grown under normal culture conditions were used as a control.

Human MALAT1 lentiviral vector (cat. no. LV212703) and blank control lentivector (cat. no. LV587) were purchased from Applied Biological Materials Inc., (Richmond, BC, Canada). MALAT1 and blank control lentivectors were transfected with the packaging vectors (psPAX2 and pMD2.G) into 293T cells by calcium chloride to produce the lentivirus. Primary human brain microvascular endothelial cells were subsequently transduced with the blank control or MALAT1-expressing lentivirus. Human MALAT1 siRNA lentivirus (cat. no. iV012699) and scrambled siRNA lentivirus (cat. no. LVP015-G) were purchased from Applied Biological Materials Inc., and were directly used to transduce primary human brain microvascular endothelial cells. Six hours after lentiviral transduction, cells were subjected to OGD-R.

Total RNA from cultured cells was isolated using miRNeasy kits (Qiagen China Co., Ltd., Shanghai, China) according to the manufacturer's protocol, followed by purification with TURBO DNA-free System (Ambion; Thermo Fisher Scientific, Inc., Waltham, MA, USA). To measure lncRNAs, 1,000 ng total RNA was reverse transcribed using MulV reverse transcriptase and random hexamer primers (both Thermo Fisher Scientific, Inc.) in a 20-µl reaction. cDNA was used as a template for RT-qPCR using Fast SYBR Green and an Applied Biosystems StepOnePlus machine (both Applied Biosystems; Thermo Fisher scientific, Inc.). Human ribosomal P0 (RPLP0) mRNA was sequenced and used as a reference gene for normalization. Primer sequences were as follows: MALAT1, forward 5′-GTGATGCGAGTTGTTCTCCG-3′ and reverse 5′-CTGGCTGCCTCAATGCCTAC-3′; and RPLP0, forward 5′-TCGACAATGGCAGCATCTAC-3′ and reverse 5′-ATCCGTCTCCACAGACAAGG-3′. The PCR reaction mixture contined 12.5 µl of SYBR Green Master Mix, 400 ng of template DNA, forward and reverse primers (0.25 µM each), and 12 µl of nuclease-free water. PCR amplification conditions were as follows: 20 sec at 94°C, followed by 40 cycles of 3 sec at 95°C and 30 sec at 62°C, and a final extension at 62°C for 10 mins. Analysis of relative gene expression levels was performed using the following formula: 2^−ΔΔCq with ΔΔCT=Cq_{(target gene)}-Cq_(control) (13).

Primary human brain microvascular endothelial cells were cultured at 9×10⁴ cells/well in 96-well tissue culture plates and subjected to OGD-R. Cell apoptosis was measured at 6, 9, 12, 24 and 36 h post-reoxygenation with a microplate reader-based TiterTACS in situ apoptosis detection kit (cat. no. 4822-96-K; R&D Systems, Inc., Minneapolis, MN, USA) according to the manufacturer's protocol. Each experiment was performed in duplicate and repeated thrice.

Cells were subjected to a TUNEL assay using an ApopTag InSitu apoptosis detection kit (cat. no. S7111; EMD Millipore, Billerica, MA, USA) according to the manufacturer's protocol. TUNEL-positive/apoptosis rate was presented as the percentage of total cells. Images were captured at ×40 magnification with a fluorescence/phase contrast microscope (FSX100; Olympus Corp., Tokyo, Japan). Three independent experiments were performed.

Activity of caspase 3 was determined using a colorimetric caspase 3 assay kit (cat. no. ab39401) from Abcam (Cambridge, MA, USA). The assays were performed in 96-well plates by incubating 20 µl cell lysate protein/sample in 70 µl reaction buffer [1% NP-40, 20 mM Tris-HCl (pH 7.5), 137 mM Nad and 10% glycerol] containing 10 µl caspase-3 substrate (2 mM). Lysates were then incubated at 37°C for 6 h, after which the samples were assayed using a spectrophotometer (cat. no. 1702525; Bio-Rad Laboratories, Hercules, CA, USA) at 405 nm.

ROS were measured using a dichlorofluorescin diacetate (DCFDA) Cellular Reactive Oxygen Species Detection Assay kit (cat. no. ab113851; Abcam, Cambridge, UK) according to the manufacturer's protocol. Cells were plated at 9×10⁴ cells/well in black 96-well plates and incubated with 25 µM DCFDA for 45 min. Fluorescence was detected using a Victor3 1420 Multilabel Counter (PerkinElmer Instruments, Shanghai, China).

PI3K activity was determined with a PI3K activity ELISA kit (cat. no. K-1000s; Echelon Biosciences Inc., Salt Lake City, UT, USA) according to the manufacturer's protocol (12). To functionally assess PI3K activity, PI3K was isolated by immunoprecipitation using a rabbit polyclonal anti-human PI3K antibody (cat. no. 06-195; EMD Millipore) to the p85 adapter subunit. The ability with which the co-precipitated catalytic p110 catalytic subunit was able to convert a standard PIP2 to PIP3 in a kinase reaction was subsequently assessed by measuring the PIP3, generated using an ELISA kit. Each experiment was repeated performed in duplicate, and was repeated three times.

Cells were lysed in a solution containing hypotonic buffer containing 0.5% Nonidet-P40 and a protease inhibitor cocktail (Sigma-Aldrich; Merck Millipore, Darmstadt, Germany) by sonication three times for 3 sec on ice. Following centrifugation at 2,000 × g for 15 min at 4°C, the resulting supernatant was used for protein concentration determination by the Coomassie blue method (Thermo Fisher Scientific, Inc.) and for subsequent steps. From each sample, 5 µg protein for each sample were separated by 10% SDS-polyacrylamide gel electrophoresis and were subsequently blotted onto a polyvinylidene difluoride microporous membrane (EMD Millipore). Membranes were blocked with 5% skim milk powder in Tris-buffered saline with tween 20 (TBS-T) for 2 h and incubated for 1 h at room temperature with rabbit polyclonal anti-human phosphorylated Akt (Ser 473) antibody (cat. no. sc-101629) or mouse monoclonal anti-human Akt antibody (cat. no. sc-81434) at a dilution of 1:1,000, and were subsequently washed three times in TBS-T (cat. no. SRE0031; Sigma-Aldrich; Merck Millipore) for 5 min and revealed using bovine anti-rabbit (sc-2370) or anti-mouse (sc-2371; all Santa Cruz Biotechnology, Inc., Dallas, TX, USA) secondary antibody (both 1:5,000) for 1 h at room temperature. Peroxidase was revealed with an enhanced chemiluminescence kit (GE Healthcare Life Sciences, Shanghai, China). Three independent experiments were performed.

Statistical analyses were performed using SPSS for Windows 19.0 (IBM SPSS, Armonk, NY, USA). All values were expressed as the mean ± standard deviation. Comparisons of the means among multiple groups were performed with one-way analysis of variance followed by post-hoc pairwise comparisons using Tukey's tests. A two-tailed P<0.05 was considered to indicate a statistically significant difference.

In the present study, primary human brain microvascular endothelial cells exposed to OGD-R were employed as an in vitro model of cerebral vascular endothelial cell I/R injury, as previously described (1). As shown in Fig. 1, human brain microvascular endothelial cells were cultured under OGD-R, and the expression level of MALAT1 and apoptosis in the cells were measured at 6, 9, 12, 24 and 36 h post-reoxygenation. As shown in Fig. 1A, using cells under normal culture conditions for normalization, the expression level of MALAT1 in cells under OGD-R significantly increased (P<0.05) after reoxygenation, peaking at 1.5-fold that of cells under normal culture conditions at 9 h post-reoxygenation, and significantly decreased (P<0.05) below the normal level, reaching 0.27-fold that of cells under normal culture conditions at 36 h post-reoxygenation. The apoptosis rate of cells under normal culture conditions did not exhibit any significant change over time (Fig. 1B). Furthermore, no significant change was identified in the apoptotic rate of cells exposed to OGD-R at 6 h (9.8±2.7%) and 9 h (11.5±3.7%) post-reoxygenation, compared with the cells under normal culture conditions. However, the apoptotic rate of cells exposed to OGD-R markedly increased to 29.1±7.2% at 12 h, reaching 42.6±9.1% at 36 h post-oxygenation (Fig. 1B). The contrasting trends of the expression levels of MALAT1 and cell apoptosis in human brain microvascular endothelial cells following OGD-R suggested that MALAT1 may inhibit OGD-R-induced apoptosis in brain vascular endothelial cells.

To explore the effect of MALAT1 on OGD-R-induced apoptosis of brain vascular endothelial cells, human brain microvascular endothelial cells were transduced with human MALAT1 lentivirus or human MALAT1 siRNA lentivirus to overexpress or knockdown MALAT1, respectively. Cells transduced with the blank control lentivirus (VC) or the scrambled siRNA lentivirus (SC) were used as controls. To investigate the potential role of PI3K/Akt survival signaling in the effects induced by MALAT1, cells overexpressing MALAT1 were treated with a PI3K inhibitor, Wortmannin. As shown in Fig. 2, the VC and SC controls exposed to OGD-R exhibited similar changes in the expression level of MALAT1 as nontransduced cells in Fig. 1A, indicating that transduction of the VC or SC did not significantly affect the expression level of MALAT1. In the presence or absence of OGD-R, lentiviral transduction of MALAT1 and MALAT1-siRNA resulted in a significant increase and decrease (both P<0.05) in the expression levels of MALAT1, when compared with the controls, respectively (Fig. 2). No significant change was detected in the expression levels of MALAT1 over time in either of the experimental groups in the absence of OGD-R (Fig. 2A). In the presence of OGD-R, all experimental groups exhibited a significant increase in MALAT1 expression levels by 9 h post-reoxygenation, followed by a continuous decrease (P<0.05) (Fig. 2B). Inhibition of PI3K signaling by Wortmannin showed no significant effect on the expression level of MALAT1 in the presence or absence of OGD-R (Fig. 2).

As shown in Fig. 3A, in the absence of OGD-R, there was no significant change detected in the cell apoptosis rate over time in each experimental group, compared with the controls. Similarly, the overexpression and knockdown of MALAT1 and inhibition of PI3K signaling by Wortmannin also exhibited no significant effect on cell apoptosis. These results suggested that MALAT1 had no significant effect on human brain vascular endothelial cell apoptosis under normal conditions. In the presence of OGD-R, however, overexpression of MALAT1 significantly decreased cell apoptosis compared with the controls (P<0.05); this effect was abolished by Wortmannin (Fig. 3B). However, knockdown of MALAT1 significantly increased cell apoptosis compared with cells transfected with the blank control lentivirus (P<0.05), up to ~90% at 36 h post-reoxygenation (Fig. 3B). TUNEL staining at 36 h post-reoxygenation corroborated these findings (Fig. 4).

Consistent with the cell apoptosis data, in the absence of OGD-R, there was no significant change identified between the groups or over time in the activity of caspase 3 (Fig. 5A), a critical caspase in the apoptotic pathway (14). In the presence of OGD-R, however, overexpression of MALAT1 significantly decreased caspase 3 activity compared with the controls (P<0.05). This effect was abolished by Wortmannin; however, knockdown of MALAT1 markedly increased the caspase 3 activity compared with the controls (Fig. 5B).

OGD-R has been shown to induce cellular overproduction of ROS, which leads to cell apoptosis (15–18). Therefore, the effect of MALAT1 on ROS production in human brain microvascular endothelial cells in the presence and absence of OGD-R was investigated. As shown in Fig. 6, in the absence of OGD-R, the ROS content in all experimental groups remained similar to the levels exhibited by cells under normal culture conditions (designated as 1) over time. However, in the presence of OGD-R, the ROS content in all experimental groups increases ~2.5-fold, compared with the cells under normal culture conditions, with no significant differences identified between the groups or over time.

These findings suggested that MALAT1 may protect human brain vascular endothelial cells from OGD-R-induced apoptosis, not by reducing OGD-R-induced ROS overproduction, but by a PI3K-dependent survival mechanism. Therefore, the effect of MALAT1 on PI3K/Akt survival signaling, which reportedly protects cerebral endothelial cells from hypoxia/reoxygenation-induced apoptosis (17), was subsequently examined in human brain microvascular endothelial cells in the presence and absence of OGD-R.

Overexpression and knockdown of MALAT1 had no significant effect on PI3K activity in the absence of OGD-R compared with the controls, except when Wortmannin completely abolished PI3K activity (Fig. 7A). There was no significant change detected in PI3K activity over time in each experimental group. However, in the presence of OGD-R, overexpression of MALAT1 markedly increased PI3K activity in human brain microvascular endothelial cells compared with the controls; this effect was completely abolished by Wortmannin (Fig. 7B). Knockdown of MALAT1 significantly decreased (P<0.05) PI3K activity compared with the controls (Fig. 7B). Similar data trends were observed with the activation of the phosphorylation of Akt at serine 473 (17) in human brain microvascular endothelial cells in the presence and absence of OGD-R, respectively (Fig. 8). These findings indicated that MALAT1 may protect human brain vascular endothelial cells from OGD-R-induced apoptosis via the PI3K/Akt survival pathway.