Lumiflavin increases the sensitivity of ovarian cancer stem-like cells to cisplatin by interfering with riboflavin

Abstract

Here, we used lumiflavin, an inhibitor of riboflavin, as a new potential therapeutic chemosensitizer to ovarian cancer stem-like cells (CSCs). This study demonstrates that the enrichment of riboflavin in CSCs is an important cause of its resistance to chemotherapy. Lumiflavin can effectively reduce the riboflavin enrichment in CSCs and sensitize the effect of cisplatin Diamminedichloroplatinum (DDP) on CSCs. In this study, CSCs of human ovarian cancer cell lines HO8910 were separated using a magnetic bead (CD133+). We also show the overexpression of the mRNA and protein of riboflavin transporter 2 and the high content of riboflavin in CSCs compared to non-CSCs (NON-CSCs). Moreover, CSCs were less sensitive to DDP than NON-CSCs, whereas, the synergistic effect of lumiflavin and DDP on CSCs was more sensitive than NON-CSCs. Further research showed that lumiflavin had synergistic effects with DDP on CSCs in increasing mitochondrial function damage and apoptosis rates and decreasing clonic function. In addition, we found that the combination of DDP and lumiflavin therapy in vivo has a synergistic cytotoxic effect on an ovarian cancer nude mice model by enhancing the DNA-damage response and increasing the apoptotic protein expression. Notably, the effect of lumiflavin is associated with reduced riboflavin concentration, and riboflavin could reverse the effect of DDP in vitro and in vivo. Accordingly, we conclude that lumiflavin interfered with the riboflavin metabolic pathways, resulting in a significant increase in tumour sensitivity to DDP therapy. Our study suggests that lumiflavin may be a novel treatment alternative for ovarian cancer and its recurrence.

Keywords: cancer stem-like cells; cisplatin; lumiflavin; ovarian cancer; riboflavin.

Figure 1

Chemical structure of lumiflavin and riboflavin

Figure 2

Difference in the cancer stem‐cell surface markers, riboflavin content, expression levels of riboflavin transporter 2 (RFT2) gene and protein, sensitivity to DDP, and the synergistic effect of lumiflavin and DDP between cancer stem‐like cells (CSCs) and NON‐CSCs. CSCs of human ovarian cancer cell lines HO8910 were isolated through magnetic bead sorting (CD133+). Co‐expression of CD133+/CD117+, CD44+/CD177, CD44+/CD24− were detected through flow cytometry, riboflavin contents were detected using ultra‐high performance liquid chromatography‐tandem mass spectrometry, and expression levels of the RFT2 mRNA and protein were examined using RT‐PCR, Western blot analysis and immunofluorescence methods. CSCs and NON‐CSCs were treated with different concentrations of DDP and with lumiflavin combined with DDP for 48 h. Cell viability was detected using CCK‐8 methods. A, Flow detection results of CD133+/CD117+, CD44+/CD177, CD44+/CD24− cells ratio of ovarian cancer cell lines HO8910 before and after by magnetic bead (CD133+) separation. B, Statistical analysis graph of A (n = 3). C, Electrophoretograms of RFT2 proteins. D, Graph of relative protein expression as determined by Western blot analysis (n = 3). E, Statistics of the gene expression of RFT2 as determined by RT‐PCR (n = 3). F, Immunofluorescent images of the RFT2 protein (red dots) and DAPI (blue dots) of CSCs and NON‐CSCs. G1‐G3: NON‐CSCs; G4‐G6: CSCs. G, Quantification of RFT2 dots per field. In total, 30 fields from three cell images were analysed per group. H, Statistical graph of riboflavin contents of CSCs and NON‐CSCs (n = 3). (I,J) Statistical maps of cell activities (n = 6). *P < 0.05 between groups; **P < 0.01 between groups. Mean ± SD

Figure 3

Synergistic effects of lumiflavin and DDP on mitochondrial membrane potential, reactive oxygen species (ROS) contents, apoptosis rate, and clonality of cancer stem‐like cells (CSCs). CSCs were seeded in 6‐well plates (1 × 10⁶ cells per well). Cells were randomly divided into the control group, lumiflavin (20 μmol/L) group, DDP (20 μmol/L) group, DDP + lumiflavin (20 μmol/L) group and DDP (20 μmol/L) + riboflavin (20 μmol/L) group. After treatment for 48 hours, the membrane potential, ROS contents and apoptosis rate were detected using a flow cytometer. CSCs were seeded in 6‐well plates (2 × 10³/well) and held for 2 weeks. The colony were fixed and then stained with crystal violet to evaluate the colony formation. A, Flow cytometry scatter plot of the potential of the mitochondrial membrane (Δψ _m) of cells as detected through flow cytometry by using a JC‐1 probe. B, Statistical analysis graph of (A). C, Flow cytometry of ROS levels as detected using the DCFH‐DA kit. D, Statistical analysis graph of (C). E, Flow cytometry of cell apoptosis as detected using Annexin V‐FITC. F, Statistical analysis graph of (E). G, Images of colony formation assay. H, Statistical analysis graph of (G). *P < 0.05 compared to control group; **P < 0.01 compared to control group. ^# P < 0.05 compared to DDP group; ^## P < 0.01 compared to DDP group. Mean ± SD (n = 6)

Figure 4

Synergistic effects of lumiflavin and DDP on the mouse xenograft model of ovarian cancer stem‐like cells (CSCs). CSCs of HO8910 (5 × 10⁶) cells were subcutaneously injected under the oxter skin of female BALB/c nude mice. When the tumours reached palpable size (100 mm³), mice were intraperitoneally injected for 25 days with saline control, riboflavin (4 mg/kg, once a day), lumiflavin (8 mg/kg, once a day), DDP (5 mg/kg, once a week), DDP + lumiflavin and DDP + riboflavin. Tumour sizes and weight were measured at the end of the experiment. Enzyme activities of GSH‐PX and superoxide dismutase (SOD) and the malondialdehyde (MDA) contents were tested using the corresponding kits, following the manufacturer's instructions. A, Photograph of tumours. B, Tumour growth curve of nude mice. C, Statistical graph of tumour weight. D, Statistical graph of enzyme activity of GSH‐PX. E, Statistical graph of MDA contents in tumour. F, Statistical graph of enzyme activity of SOD. *P < 0.05 compared to control group; **P < 0.01 compared to control group. ^# P < 0.05 compared to DDP group; ^## P < 0.01 compared to DDP group. Mean ± SD (n = 5)

Figure 5

Synergistic effects of lumiflavin and DDP on the down‐regulation of BCL‐2 expression and up‐regulation of the cleaved caspase 3, Bax, and Bad expression in the tumour cells of mice with ovarian cancer stem‐like cells (CSCs) by using the xenograft model. CSCs of HO8910 (5 × 10⁶) cells were subcutaneously injected under the oxter skin of female BALB/c nude mice. When the tumours reached a palpable size (100 mm³), mice were intraperitoneally injected for 25 days with saline control, riboflavin (4 mg/kg, once a day), lumiflavin (8 mg/kg, once a day), DDP (5 mg/kg, once a week), DDP + lumiflavin and DDP + riboflavin. Protein expression of the cleaved caspase 3, Bax, Bad, and BCL‐2 in tumour tissues were analysed through Western blot analysis. A, Electrophoretograms of cleaved caspase 3, Bax, Bad and BCL‐2 proteins. B, Graph of relative protein expression as determined by Western blot analysis. ^* P < 0.05 compared to control group; **P < 0.01 compared to control group. ^# P < 0.05 compared to DDP group; ^## P < 0.01 compared to DDP group. Mean ± SD (n = 5)

Figure 6

Synergistic effects of lumiflavin and DDP on ovarian cancer CSCs involved in riboflavin. CSCs were seeded in 6‐well plates (1 × 10⁶ cells per well), and then treated for 48 hours as follows: control group, DDP (20 μmol/L) group, and DDP + lumiflavin (20 μmol/L) group. CSCs were subcutaneously injected under the oxter skin of female BALB/c nude mice. Mice were intraperitoneally injected after 25 days with saline control, DDP (5 mg/kg, once a week) and DDP + lumiflavin (8 mg/kg, once a day). The contents of riboflavin in CSCs and tumour tissues were measured using ultra‐high performance liquid chromatography‐tandem mass spectrometry (A) Statistical figure of the CSCs contents of riboflavin. B, Statistical figure of tumour tissue contents of riboflavin. *P < 0.05 compared to control group; **P < 0.01 compared to control group. ^## P < 0.01 compared to DDP group. Mean ± SD (n = 5)