LncRNA-SNHG7 interferes with miR-34a to de-sensitize gastric cancer cells to cisplatin

2.1 Patient samples

This study was approved by the Medical Ethics Committee of the Fourth Medical Center of PLA General Hospital (number 2020KY012-KS001). Human gastric cancer tissue samples used for this study came from 30 gastric cancer and 30 normal gastric tissues. Samples were collected from the Department of General Surgery, the Fourth Medical Center of PLA General Hospital from January 2017 to June 2018. All patients received no radiotherapy and/or chemotherapy before surgery. Upon dissection, tissues were stored immediately at liquid nitrogen and then transferred to - 80 ∘ C until use. Written informed consent was obtained from all patients.

2.2 Cell culture and reagents

The human gastric epithelial cell line and gastric cancer cell lines were obtained from the American Type Culture Collection (ATCC). Cells were cultured in RPMI 1640 medium (Thermo Fisher Scientific, Inc., Carlsbad, CA, USA) with 10% fetal bovine serum (FBS), 100 units/ml penicillin, and 100 μ g/ml streptomycin (Thermo Fisher Scientific, Inc., Carlsbad, CA, USA) at 37 ∘ C under a humidified atmosphere with 5% CO 2 . The establishment of cisplatin resistance gastric cancer cells from HGC27 was performed according to previous descriptions [18]. Rabbit monoclonal anti-LDHA (#3582), anti-Hexokinase 2 (#2867); anti-PKM 2 (#4053); and anti- β -actin (#4970) antibodies were purchased from Cell Signaling Technology (Danvers, MA, USA). Cisplatin were purchased from Sigma-Aldrich (Shanghai, China).

2.3 Prediction of lncRNA-miRNA and miRNA-mRNA interactions

The interaction between lnRNA-SNHG7 and miR-34a was predicted by starBase of ENCORI (http://star base.sysu.edu.cn/) according to previous reports [19]. The binding of miR-34a on 3’UTR of LDHA was predicted from the TargetScan.org.

2.4 Plasmid, shRNA and miRNAs transfection

The gastric cancer cells were cultured to about 80% confluence in 24 or 96-well plates then cells using were transfected with the indicated shRNA, miRNAs, or plasmid DNA using Lipofectamine 2000 (Thermo Fisher Scientific, Inc., Carlsbad, CA, USA) according to the manufacturer’s instructions. The shRNA-SNHG7 or negative control was conducted by RiboBio (Guangzhou, China). The miR-34a precursor, miR-34a inhibitors or negative control were purchased from GenePharma (Shanghai, China) and transfected at 50 nM. The overexpression plasmid of LDHA was purchased from Origen.com and transfected at 2 ug. After 48 hours transfection, the cells were harvested for downstream experiments.

2.5 RNA isolation and qRT-PCR

The total RNA was extracted from the gastric tumor tissues or adjacent normal gastric tissues by the Trizol method according to the manufacturer’s instructions. First-strand cDNA was synthesized by reversely transcribing RNA into complementary DNA from 500 ng total RNA using a SuperScript First-Standard Synthesis System (Invitrogen, Carlsbad, CA, USA) following the manufacturer’s protocol. qRT-PCR experiments were performed using the SYBR Green qPCR Master Mix (ThermoFisher Scientific, Shanghai, China). The fold changes were analyzed using the 2 - Δ ⁢ Δ ⁢ Ct method. Experiments were performed in triplicate and repeated three times.

2.6 Luciferase reporter assays

The luciferase assay was performed according to previous descriptions [20]. 5 × 10 4 gastric cancer cells were grown in a 24-well plate, and co-transfected with pMIR vectors containing wild-type or mutated 3’UTR of LDHA reporter plasmid and miR-34a precursor or negative control miRNAs by lipofectamine 2000. Forty-eight hours after transfection, luciferase activity was measured using the Dual Luciferase Reporter Assay System (Promega, Madison, WI, USA). Experiments were performed in triplicate and repeated three times.

2.7 Measurements of glucose metabolism rate

The glucose uptake and lactate product measurements were performed using the glucose test kit (Applygen Technologies, Beijing, China) and the L-lactate assay kit (BioVision, Milpitas, CA, USA) according to the manufacturer’s instruction. Results were normalized according to the cell number by dividing the readings from treated cells to that of control cells. Experiments were performed in triplicate and repeated three times.

2.8 Cell viability assay

The cell viability assay was performed by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) method (Sigma-Aldrich, Shanghai, China) according to the manufacture’s instruction. Briefly, cells (5 × 10 3 cells/well) were seeded into 96-well plates. The next day cells were washed by PBS, 0.5 mg/ml MTT solution was added into cells and cultured at 37 ∘ C incubator for 2 hours. Cells were then incubated with 100 ul DMSO for 2 hours at 37 ∘ C. The optical density (OD) of formazan concentrations was determined at 540 nm. Results were normalized by cell numbers.

2.9 Western blot assay

Total proteins were extracted from gastric cancer tissues or cells using the RIPA buffer with protease inhibitor cocktail (Bio-Rad, Hercules, CA, USA) according to the instructions of the manufacturer. After 15 min incubation with lysis buffer on ice, samples were centrifuged at 10,000 g for 10 min at 4 ∘ C. Equal amount protein from each sample with loading buffer were separated by 10% of sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto PVDF membrane (BD Biosciences, San Diego, CA, USA). The membranes were blocked by 5% bovine serum albumin (BSA) then were incubated with the specific primary antibodies rabbit monoclonal anti-Hexokinase2, anti-LDHA and anti-PKM2 at 1:1000 for measurements of glycolysis rate at 4 ∘ C for overnight. Membranes were incubated with the diluted horseradish peroxidase (HRP)-labeled secondary antibody (goat anti-rabbit IgG) (1:3000) for 1 hour at room temperature. The membranes were then visualized with the enhanced electrochemiluminescence (Solarbio, Beijing, China) and photographed.

2.10 Statistical analysis

Results were analyzed by the Prism 6.0 software package (GraphPad Software, Inc.). All experiments were performed three times. Paired student t -tests or unpaired t -tests tests were used to compare data from two groups. One-way ANOVA method was used for analyzing data from multiple groups. 0.05 was considered statistically significant.

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

OPEN IN VIEWER

3.1 LncRNA-SNHG7 is upregulated in gastric cancer and positively correlates with cisplatin resistance

To investigate the clinical relevance of SNHG7 during the progression of gastric cancer, we collected the gastric tumor tissues and their corresponding adjacent normal tissues ( n = 30). Results indicated that the SNHG7 levels were significantly increased in tumor tissues compared with gastric normal tissues (Fig. 1A). Meanwhile, the SNHG7 was significantly upregulated in five gastric cancer cells compared with normal gastric epithelial cells (Fig. 1B), suggesting an oncogenic role of SNHG7 in gastric cancer. Since the acquired chemo-resistance frequently occurs in gastric cancers and the mechanisms are not well understood. We first assessed the effects of SNHG7 on cisplatin sensitivity of gastric cancer cells. SNHG7 was overexpressed in HGC27 and AGS cells by transfection of SNHG7 overexpression plasmid (Fig. 1C and D). As we expected, gastric cancer cells with SNHG7 overexpression showed increased cell viability under cisplatin treatments (Fig. 1E and F). To further elucidate the role of lncRNA-SNHG7 in cisplatin resistance of human gastric cancer, we established a cisplatin resistant cell line using the HGC27 cell line [18]. As shown in Fig. 2A, parental HGC27 cells, which have a cisplatin IC50 at 8.32 μ M, displayed reduced cell viabilities with cisplatin treatments. But the IC50 of cisplatin resistant cells (23.72 μ M) was significantly higher compared with that of parental cells (Fig. 2A). Furthermore, we observed the cisplatin resistant cells displayed significantly increased lncRNA-SNHG7 expression (Fig. 2B). To investigate the direct roles of SNHG7 in cisplatin sensitivity, we effectively knocked down the SNHG7 expressions by transfecting SNHG7 shRNA into HGC27 cisplatin resistant cells (Fig. 2C). Cell viability results demonstrated cisplatin resistant gastric cancer cells with SNHG7 inhibition were more sensitive to cisplatin (Fig. 2D). The cisplatin IC50 of cells with SNHG7 knockdown was significantly lower than that of control cells. Taken together, the above results demonstrated SNHG7 is positively associated with cisplatin resistance in gastric cancer.

3.2 miR-34a is downregulated in gastric cancer and negatively correlate with cisplatin resistance

We next to evaluate the molecular mechanisms for the SNHG7-mediated cisplatin resistance. Previous studies have demonstrated lncRNAs can act as miRNA sponges to suppress their regulatory effects on targets mRNAs [19]. Moreover, miR-34a has been reported to be negatively correlated with cisplatin resistance in osteosarcoma [21], suggesting miR-34a might be a downstream effector of SNHG7, which involves in cisplatin resistance. Expectedly, the miR-34a expressions were significantly downregulated in tumor tissues compared with gastric normal tissues (Fig. 3A). Similar trend was observed that miR-34a was suppressed in gastric cancer cells compared with normal gastric epithelial cells (Fig. 3B). Consistently, miR-34a was found downregulated in cisplatin resistant HGC27 cells compared with parental cells (Fig. 3C). Thus, we hypothesized miR-34a could sensitizes gastric cancer cells to chemotherapy. Results in Fig. 3D illustrated overexpression of miR-34a significantly increased the sensitivity of cisplatin resistant cells (Fig. 3D). The IC50 of miR-34a overexpressing cells was 7.62 μ M, which is around 3 folds less than that of cisplatin resistant cells. In summary, the above results demonstrated miR-34a has suppressive roles in gastric cancer cells and could sensitize cisplatin resistant cells.

3.3 SNHG7 suppresses miR-34a expression via sponging

Accumulating evidence indicated that lncRNAs contain motifs which are complementary to miRNAs sequences to negatively regulate miRNAs. A putative complementary sequence was predicted for miR-34a in SNHG7 Starbase (http://starbase.sysu.edu.cn/index.php) (Fig. 4A), suggesting SNHG7 could inhibit miR-34a expression via direct binding to seeding region of miR-34a. Meanwhile, we found a negative correlation between SNHG7 and miR-34a in gastric tumor specimen (Fig. 4B). Consistently, overexpression of SNHG7 in gastric cancer cells significantly inhibited miR-34a expressions (Fig. 4C). On the other way, knocking down SNHG7 obviously upregulated miR-34a expressions (Fig. 4D). These results revealed SNHG7 suppresses miR-34a expression by sponging it.

3.4 Cisplatin resistant gastric cancer cells showed elevated glycolysis rate

Accumulating evidence suggests that malignant cancer cells demand increased glucose metabolism via a higher anaerobic glycolysis rate, a phenomenon was known as “the Warburg effect” [15]. Moreover, the acquired chemoresistance of cancer cells were frequently associated with higher glycolysis rate [16]. To investigate the molecular mechanism for the SNHG7/miR-34a-meidated cisplatin resistance, we compared the glycolysis rates of HGC27 parental and cisplatin resistant cells. Interestingly, the glucose uptake and lactate product, two glycolysis speed limit steps [15], as well as glycolysis key enzymes were apparently increased in cisplatin resistant cells (Fig. 5A and B), indicating the dysregulated cellular glycolysis rate was tightly associated with cisplatin resistance. Consistent results from Fig. 5C and D demonstrated overexpression of SNHG7 facilitated glucose uptake and lactate product of gastric cancer cells. On the contrary, overexpression of miR-34a significantly suppressed glycolysis rates of HGC27 and AGS cells (Fig. 5E and F). These results suggested targeting the SNHG7-promoted cellular glycolysis rate might contribute to anti-chemoresistance of gastric tumor.

3.5 miR-34a directly targets glycolysis enzyme LDHA to increase the cisplatin sensitivity of gastric cancer cells

To identify the potential target of miR-34a, we analyzed the miR-34a binding sites on mRNAs through the TargetScan databases. Among candidates, we found the 3’-UTR of human LDHA, a glycolysis key enzyme which catalyzes the convert of pyruvate to lactate, contains a putative miR-34a binding region (Fig. 6A). Consistent results from Spearman’s correlation analysis demonstrated overexpression of miR-34a significantly decreased the protein levels of LDHA (Fig. 6B). Importantly, the reverse correlation between miR-34a and LDHA mRNA expressions was found in gastric tumor specimen (Fig. 6C). To verify whether miR-34a could directly bind to the 3-UTR region of LDHA mRNA, a luciferase reporter assay was performed by co-transfecting control miRNA, miR-34a precursor and original-LDHA 3’UTR or nucleotides mutant LDHA 3’UTR (Fig. 6A) into HGC27 and AGS cells. Luciferase assay showed that miR-34a significantly decreased the luciferase activity of plasmid containing the wild type LDHA 3’-UTR but not the binding site mutant LDHA 3’-UTR. Taken together, the above results strongly demonstrated miR-34a could directly target 3’UTR of LDHA mRNA in gastric cancer cells and tumor. To test whether the miR-34a/LDHA-mediated glycolysis inhibition led to cisplatin sensitization, we transfected HGC27 cisplatin resistant cells with control miRNAs, or miR-34a precursor or miR-34a plus LDHA overexpression plasmid. Western blot results demonstrated co-transfection of miR-34a and LDHA successfully rescued the LDHA protein expressions (Fig. 7A). Furthermore, the glucose uptake (Fig. 7B) and lactate product (Fig. 7C) were significantly recovered in miR-34a and LDHA co-transfected cells. We next asked whether the miR-34a/LDHA axis was response for the miR-34a-mediated cisplatin sensitization by rescue experiments. As we expected, restoration of LDHA expression in miR-34a overexpressed cells contradicted the miR-34a-mediated cisplatin sensitization in gastric cancer cells (Fig. 7D and E). Importantly, recovering LDHA successfully restored the cisplatin resistance of HGC27 cisplatin resistant cells (Fig. 7F).

3.6 The acquired cisplatin resistance is through the SNHG7-miR-34a-glycolysis axis

Our results revealed SNHG7 negatively regulated miR-34a, suggesting that the miR-34a/LDHA binding was associated with the SNHG7-mediated cisplatin resistance. To test this hypothesis, lncRNA-SNHG7 was knocked down in HGC27 cells, or knock down SNHG7 plus anti-miR-34a treatment. We detected inhibition of endogenous SNHG7 significantly induced miR-34a expressions and downregulated LDHA expressions (Fig. 8A and B). Moreover, knockdown SNHG7 plus anti-miR-34a recovered the LDHA protein levels in HGC27 cells (Fig. 8A and B), suggesting SNHG7 could upregulate LDHA through suppressing miR-34a in gastric cancer cells. Furthermore, inhibiting the SNHG7 downregulation-induced miR-34a apparently recovered glucose uptake and lactate product (Fig. 8C and D). Consistent results in Fig. 8E showed GLUT1, HK2 and PKM2 which are glycolysis key enzymes were significantly downregulated by SNHG7 knocking down, however, such downregulation was rescued by further miR-34a inhibition (Fig. 8E). Finally, we transfected HGC27 cisplatin resistant cells with control, sh SNHG7 or sh SNHG7 plus anti-miR-34a, followed by treatments with increased concentrations of cisplatin. Expectedly, SHHG7 knockdown cells showed increased cisplatin sensitivity, however, such sensitization effect was further recovered by miR-34a inhibition, suggesting the SNHG7-mediated cisplatin resistance was through the miR-34a-LDHA axis.