Hypoxia-mediated mitochondria apoptosis inhibition induces temozolomide treatment resistance through miR-26a/Bad/Bax axis

TMZ is a kind of alkylating agent to cause DNA damage and cell apoptosis. The phosphorylation of histone family member X at serine 139 (γH2AX) is widely used as a molecular marker for DNA double strands break (DSB)²⁵. Interestingly, our results showed that TMZ treatment induced the expression levels of γH2AX, while hypoxia reduced the augment of γH2AX in TMZ treatment U87MG cells relative to the control cells (Fig. 1d). Furthermore, Immunofluorescence staining of γH2AX foci showed that the number of γH2AX foci was increased in the TMZ-treated U87MG cells and was attenuated under hypoxic condition compared with normoxic condition, suggesting that hypoxia helped inhibition of DNA damage in U87MG cells (Fig. 1g). Thus, our data indicate that hypoxic microenvironment protects the glioma cells from apoptosis to TMZ treatment though protecting mitochondria function and decreasing apoptosis and DNA damage.

To further evaluate the role of miR-26a/Bax/Bad axis on TMZ resistance in vivo, the cells combination of miR-26a or Bax /Bad were implanted intracranially in each mouse and treated by TMZ at day 14 after implantation and representative pseudocolor bioluminescence images of intracranial mice were conducted. The results showed that Bax/Bad overexpressiondecreased activities of bioluminescence and inhibited tumor growth, further confirmed that Bax/Bad overexpression promotes TMZ sensitivity of glioma cells in vivo (Fig. 5f). T2-weighted MRI and hematoxylin and eosin staining confirmed that mice bearing Bax/Bad overexpression produced smaller tumor than miR-26a (Fig. 5g). We also found that Bax/Bad overexpression tumors showed a higher apoptotic index as demonstrated by active caspase-3 staining compared with miR-26a group (Fig. 5h). Kaplan–Meier curves were drawn to measure mouse survival with tumor burden induced by U87/miR-26a with or without Bax/Bad overexpression, and U87MG/miR-26a with Bax/Bad overexpression showed a higher survival rate (Fig. 5i). Taken together, our study first demonstrates that miR-26a/Bax/Bad axis contributes to hypoxic-mediated TMZ resistance in vitro and in vivo.

U87MG and HEK293T cells were obtained from American Type Culture Collection and were cultured in Dulbecco’s Modified Eagle Medium (Gibco, Grand Island, NY, USA) with 10% FBS. Temozolomide (TMZ) was purchased from Selleck Chemicals (Shanghai, China). Echinomycin was purchased from Sigma (St. Louis, MO, USA). Bevacizumab was purchased from Roche (Welwyn Garden City, UK). See Supplementary Tablefor detail information of antibodies. 1

Lentivirus expressing miR-26a, miR-NC (negative control) and anti-miR-26a inhibitor were purchased from Genepharma (Shanghai, China). Stable cell lines U87/miR-NC, U87/miR-26a, and U87/anti-miR-26a were established by infecting U87 cells with lentivirus, then selected using puromycin (Sigma, MI, USA).

To confirm miR-26a effect on TMZ treatment, we planted 3000 cells per well in 96-well plates. The absorption value was measured by CCK8 kit (Dojindo Laboratories, Kumamoto, Japan) at different time points indicated. Data were obtained from three separate experiments with six replications each time.

For colony formation assays, cells were placed in 24-well plates with 200 cells per well. The medium was changed every 4 days. Cells were fixed with methanol 14 or 28 days later, stained with 0.05% crystal violet, photographed, and counted. We did triplicate for each group. And experiments were independently repeated three times.

Cell-Light™ EdU Apollo®488 In Vitro Imaging Kit (Ribobio, Guangzhou, China) was performed as manufracture instruction to determine cells proliferation. Cell nuclei were stained for 30 min by Hoechst 33342 at 5 µg/mL. After incorporation, a fluorescent dye making possible fluorescent visualization of proliferating cells was used that reacted specifically with EdU. Microscopic observation was performed under a fluorescence microscope (ZEISS, Germany).

For dual-luciferase assay, Bad or Bax 3′-UTR region containing putative miR-26a matching sites (wide type, WT) and corresponding mutant sites (Mut) were cloned into the SacI and HindIII restriction enzyme sites of pMIR-REPORTER vector (Ambion, CA, USA), immediately downstream of luciferase stop codon. Then we transfected the Wild-type constructs and mutant ones into the cells and co-transfected luciferase reporter plasmids containing either wild-type (WT) or mutant-type (Mut) of Bad or Bax -3′UTR regions, pGL4.74 luciferase reporter vector (Ambion) and miR-26a mimics or miR-NC mimics, anti-miR-26a inhibitor or anti-miR-NC inhibitor (Thermofisher, US) using jetPRIME into U87 cells. Firefly and Renilla luciferase activities were confirmed using a dual-luciferase assay kit (Promega, WI, USA) at 24 h after transfection. Primers used for reporter constructions were shown in Supplemental Table S2.

Total RNAs of cells were extracted using TRIzol reagent (Takara, Dalian, China) from cultured cell or fresh tissues following the manufacturer’s instructions. qPCR was conducted to test expression of miR-26a and mRNA. U6 and GAPDH was used as an endogenous control. qPCR using SYBR Premix DimerEraser (Takara, Dalian, China) was performed on the 7900HT system. Relative expression levels of the target gene were determined by the comparative cycle threshold method (2^−ΔΔCt). Primers used for qPCR were shown in Supplemental Table S1.

RIPA buffer (150 mM NaCl, 50 mM Tris, pH 7.4, 1% Triton X-100, 1 mM EDTA) was used to lyse cells on ice for 30 min. The lysates were centrifugated at 13,000 rpm 4 °C for 15 min, and the supernatants were collected and protein concentrations were determined using BCA assay (Beyotime Institute of Biotechnology, China). Samples were separated by SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to PVDF membranes in transfer buffer. Membranes were blocked with 5% nonfat milk in 1 × PBS containing 0.05% Tween-20 and incubated overnight with primary antibodies at 4 °C. The protein bands were detected with appropriate horseradish peroxidase conjugated secondary antibodies and reacted with the SuperSignal West Pico Chemiluminescent Substrate Kits (Thermo Scientific, MA, USA). Antibodies used for western blotting were shown in Supplemental Table S2

Cells were treated with TMZ as indicated. ANXA5 and PI was stained by the Alexa Fluor 488Annexin V Dead Cell Apoptosis Kit (BD Pharmingen) as protocol. The cells were analyzed by flow cytometry (BD Bioscience, FACSCalibur, Singapore), using 488 nm excitation and measuring the fluorescence emission at 530 and 585 nm. FlowJo software were used to analyze data. Three experiments were performed in triplicate.

ChIP assays were performed using the Simple ChIP PlusEnzymatic Chromatin IP Kit (Agarose Beads) (Cell Signaling Technology, USA). Cells were cross-linked by addition of 37% paraformaldehyde at 37 ℃ for 20 min and quenched by glycine. Then cells were rinsed with PBS buffer. The extracted chromatin was digested and fragmented into 150 to 900 bp, then DNA was enriched from the cell lysates by specific antibodies immunoprecipitating HIF-1α or IgG overnight at 4 °C, and collected by incubation with Protein A/G Agarose/Salmon Sperm DNA (50% Slurry) for 2 h. The purified DNA was used as a template for qPCR. Fold enrichment was analyzed according to Ct as 2-Δ(ΔCt), where ΔCt = CtIP-CtInput and Δ(ΔCt) = ΔCtantibody-ΔCtIgG. Sequences of Primers used in this study are listed in Supplementary Table. S2

U87MG cells transfected with miR-NC or miR-26a were fixed with 1% formaldehyde, followed by chromatin fragmentation. Cells were lysed in NETN buffer and incubated with Dyna beads Protein A (Invitrogen) together with anti-pan Ago antibody (Millipore) or IgG control for immunoprecipitation. After proteinase K digestion and extraction, the immunoprecipitated RNA was acquired. Ethanol with glycogen was used to precipitate RNA.

The mitochondrial membrane potential was analyzed by fluorescence microscopy using the 5,50,6,60-tetrachloro-1,10,3,30-tetraethylbenzimidazol carbocyanineiodide probe. After treatment as indicated, we mixed JC-1 staining solution (Beyotime, China) and culture medium at the ratio of 5 μL per millilite and incubated the samples in cell culturing incubator for 20 min avoided from light. After twice washing, cells were checked using a fluorescence microscope (Zeiss, Germany).

Cells were treated using the procedure as indicated. Then cells were incubated with 50 nM MitoTracker Green to visualize mitochondria. Cells were fixed 30 min later with 37% paraformaldehyde and permeabilized with PBS containing 0.5% Triton X-100 at room temperature. And 1% bovine serum albumin (BSA) in PBST was used to block the samples. Bax and Bad antibodies was used for immunostain overnight at 4 °C. The fluorescence isothiocyanate (FITC)-labeled goat anti-rabbit secondary antibody (Santa Cruz) and tetramethylrhodamine isothiocyanate (TRITC)-labeled goat antimouse secondary antibody (Santa Cruz) were incubated with the samples. Cells were incubated with 10 μM Hoechst 33342 for 10 min to visualize nuclei. Microscopic observation was performed under a fluorescence microscope (ZEISS, Germany).

Paraffin-embedded tissues were sectioned at 5 μm and then deparaffinized in xylenes. After above processes, tissue slides were rehydrated through an graded ethanol dilution series (from 100 to 70%). Slides were submerged in diethylpyrocarbonate-treated water and subjected to Hydrogen Peroxide treatment and Pepsin digestion, refixed in 4% paraformaldehyde, then slides were rinsed with PBS between treatments. Slides were prehybridized in hybridization solution at 38–42 °C for 60 min. Then, 10 pmol probes (Boster, Wuhan, China) complementary to miR-26a was added and hybridized for 20 h at a temperature of 38–42 °C following washes in SSC at the same temperature. After incubating with anti-DIG-HRP Fab fragments conjugated to horseradish peroxidase, the hybridized probes were measured by 3′-3-diaminobenzidine solution, and nuclei were counterstained with Haematoxylin.

Fresh tumor tissues were fixed with Bouin solution for 24–48 h and processed by the paraffin-embedded method. Deparaffinage and dehydration were as same as the processes of in situ hybridization. Following, tissue slides were heat immobilized or pepsin immobilized and incubated with antibody against HIF-1α or Caspase-3 overnight at 4 ℃. After incubated with secondary antibody, the slides were reacted with DAB Histochemistry Kit (Invitrogen, USA).

Female BALB/c nude mice, 4 to 5 weeks old, were purchased from Vital River Animal Center (Beijing, China) and maintained in special pathogen-free (SPF) condition for one week. Animal experimental procedures were in consistent with the Care and Use of Laboratory Animals Guide and approved by the Animal Experimental Ethics Committee of Nanjing Medical University. Cells (4 × 10⁶) mixing with 150 μL of FBS-free DMEM medium were subcutaneously injected into both sides of the posterior flanks of 5 nude mice per group. After implantation for 24 days, mice were killed to dissect the tumor tissues. Total proteins and RNAs were extracted for further detection. Tumors were formalin-fixed, paraffin-embedded, and tumor sections at 5 μm were analyzed by immunohistochemical staining for HIF-1α levels and by hybridization in situ for miR-26a levels.

U87-Luc cells (1 × 10⁵) with indicated virus infection or treatment were resuspended in Matrigel in a total volume of 5 μL and injected intracranially into the striatum of adult nude mice (6-week-old) by a stereotactic device (coordinates: 2 mm anterior, 2 mm lateral, 3 mm depth from the dura). At one week after injection, the tumor-bearing mice were intraperitoneally injected with 20 mg/kg TMZ in saline (final DMSO concentration 25%). To visualize the injected cells, mice were performed intraperitoneal injection with D-luciferin (150 mg/kg) and formated of image under anesthesia with the IVIS Illumina System (Caliper Life Sciences). Picture acquisition was conducted once a week to supervise tumor growth. After 6 or 8 weeks, brains were perfused with 4% paraformaldehyde by cardiac perfusion and further fixed with Bouin solution at 4 °C overnight. Then brains were resected and performed H&E stainings.

Data shown in our study were represented as means ± SD from at least three times independent experiments. Student’s unpaired t-test was conducted for comparison between two groups. Significantly difference were considered when p < 0.05.

Supplement Figure legend supplemental-table1 supplemental-table2 Supplement_Fig1 Supplement_Fig2 Supplement_Fig3 Supplement_Fig4 Supplement_Fig5-1 Supplement_Fig5-2 Supplement_Fig5-3 Supplement_Fig5-4