Long noncoding RNA MRUL promotes ABCB1 expression in multidrug-resistant gastric cancer cell sublines

Abstract

Multidrug resistance (MDR) is the most common cause of chemotherapy failure in gastric cancer (GC) treatment; however, the underlying molecular mechanisms remain elusive. Long noncoding RNAs (lncRNAs) can be involved in carcinogenesis, but the effects of lncRNAs on MDR are poorly understood. We show here that the lncRNA MRUL (MDR-related and upregulated lncRNA), located 400 kb downstream of ABCB1 (ATP-binding cassette, subfamily B, member 1), was significantly upregulated in two multidrug-resistant GC cell sublines, SGC7901/ADR and SGC7901/VCR. Furthermore, the relative expression levels of MRUL in GC tissues were negatively correlated with in vitro growth inhibition rates of GC specimens treated with chemotherapeutic drugs and indicated a poor prognosis for GC patients. MRUL knockdown in SGC7901/ADR and SGC7901/VCR cells led to increased rates of apoptosis, increased accumulation, and reduced doxorubicin (Adriamycin [ADR]) release in the presence of ADR or vincristine. Moreover, MRUL depletion reduced ABCB1 mRNA levels in a dose- and time-dependent manner. Heterologous luciferase reporter assays demonstrated that MRUL might positively affect ABCB1 expression in an orientation- and position-independent manner. Our findings indicate that MRUL promotes ABCB1 expression and is a potential target to reverse the MDR phenotype of GC MDR cell sublines.

FIG 1

Microarray screening of related lncRNAs using SGC7901/ADR and SGC7901/VCR sublines. (A) SGC7901/ADR and SGC7901/VCR sublines were induced by the parental gastric adenocarcinoma cell line SGC7901 using ADR and vincristine, respectively. (B) lncRNA microarrays covered 18,534 lncRNAs and were used to identify MDR-related lncRNAs. When expression profiles were compared between SGC7901/VCR and SGC7901 and between SGC7901/ADR and SGC7901, 1,418 and 1,497 differentially expressed lncRNAs were identified, respectively; of these, SGC7901/VCR and SGC7901/ADR shared 625 differentially expressed lncRNAs (lncRNA-Cs). GES-1 is the immortalized normal gastric epithelial cell line. (C) lncRNA-Cs (fold change, ≥4.0). (D) lncRNA-Cs (fold change, ≥4.0) surrounded by drug resistance-related genes reported in multiple tumors and the corresponding drug resistance-related genes.

FIG 2

lncRNA MRUL expression levels showed a negative linear correlation with chemosensitivity to gastric cancer (GC) cell lines and tissue specimens. (A) MRUL relative expression levels in SGC7901/VCR, SGC7901/ADR, and SGC7901 cells. (B-a) MRUL relative expression levels of clinical samples compared with those of paired noncancerous tissues. (B-b) P-gp expression of 40 primary GC tissues and paired noncancerous tissues. Relative IHC scores of P-gp were positively correlated with MRUL expression levels. T, tumor; N, normal. (C) Forty GC tissues were divided into MRUL high-expressor and low-expressor groups according to the fold change median. (D) Inhibition rates of MRUL high- and low-expressor groups. (E) Negative correlation between inhibition rates and MRUL expression levels. (F) Patient life spans in MRUL low- and high-expressor groups. Data represent means plus standard deviations of three replicates (***, P < 0.001; **, P < 0.01).

FIG 2

lncRNA MRUL expression levels showed a negative linear correlation with chemosensitivity to gastric cancer (GC) cell lines and tissue specimens. (A) MRUL relative expression levels in SGC7901/VCR, SGC7901/ADR, and SGC7901 cells. (B-a) MRUL relative expression levels of clinical samples compared with those of paired noncancerous tissues. (B-b) P-gp expression of 40 primary GC tissues and paired noncancerous tissues. Relative IHC scores of P-gp were positively correlated with MRUL expression levels. T, tumor; N, normal. (C) Forty GC tissues were divided into MRUL high-expressor and low-expressor groups according to the fold change median. (D) Inhibition rates of MRUL high- and low-expressor groups. (E) Negative correlation between inhibition rates and MRUL expression levels. (F) Patient life spans in MRUL low- and high-expressor groups. Data represent means plus standard deviations of three replicates (***, P < 0.001; **, P < 0.01).

FIG 3

lncRNA MRUL depletion leads to increased chemosensitivity of MDR GC sublines to P-gp related drugs. (A) MRUL relative expression levels. Following the transfection of SGC7901/ADR and SGC7901/VCR cells with siMRUL for 3 days, RT-PCR was used to assess MRUL expression levels and GAPDH was used as the internal control. (B and C) Survival rates of SGC7901/ADR and SGC7901/VCR cells treated with siMRUL or siNC in the presence of ADR (B-a), VCR (B-b), CDDP (C-a), or 5-FU (C-b). The MTT assay was used to examine cell survival rates at the designated time. (D) Plating colony formation assay of SGC7901/ADR and SGC7901/VCR cells treated with siMRUL or siNC in the presence of ADR or VCR. (E) Apoptosis rates of SGC7901/ADR and SGC7901/VCR cells treated with siMRUL or siNC in the presence of ADR, CDDP, or VCR. Data represent means plus standard deviations of three replicates (***, P < 0.001).

FIG 4

MRUL depletion effects on drug accumulation and apoptosis of MDR GC sublines. (A) ADR accumulation and retention of SGC7901/ADR and SGC7901/VCR cells treated with siMRUL or siNC. (B) Effects of MRUL depletion on apoptosis of SGC7901/ADR and SGC7901/VCR cells. (C) Effects of MRUL depletion on PRL23, RPS13, JNK1, and CPP32 expression of SGC7901/ADR and SGC7901/VCR cells. Data represent means plus standard deviations of three replicates (#, P < 0.01).

FIG 5

MRUL knockdown reduced P-gp expression. (A) P-gp expression examined by Western blotting at the designated time. (B) Effects of MRUL knockdown using different concentrations of siMRUL on MRUL and ABCB1 (P-gp) transcription in SGC7901/ADR cells. (C) Effects of MRUL depletion using 100 nM siMRUL on MRUL and ABCB1 (P-gp) transcription in SGC7901/VCR cells for the indicated times. (D) Effects of MRUL upregulation on MRUL and ABCB1 (P-gp) transcription in SGC7901/ADR cells. (E) Effects of P-gp depletion using different concentrations of siP-gp on MRUL and ABCB1 (P-gp) transcription. (F) Effects of P-gp depletion using 100 nM siP-gp on MRUL and ABCB1 (P-gp) transcription for the indicated times. Data represent means plus standard deviations of three replicates (***, P < 0.001; **, P < 0.01; *, P < 0.05).

FIG 6

lncRNA MRUL plays an enhancer-like role in the activation of a heterologous reporter. (A) Schematic diagram of MRUL and ABCB1 (P-gp) gene locus. (B) Schematic diagram of constructs. SV40, simian virus 40. (C) Enhancement role of MRUL inserts in the heterologous luciferase reporter in MDR GC sublines. (D) Depletion of MRUL using siMRUL reduced markedly the enhancement role of MRUL inserts in the heterologous luciferase reporter. (E) Reversal of MRUL inserts did not affect the enhancement role of MRUL. (F) Alteration of position of MRUL inserts did not affect the enhancement role of MRUL, either. Data represent means + standard deviations of three replicates (***, P < 0.001; **, P < 0.01).