Quercetin inhibits the epithelial-mesenchymal transition and reverses CDK4/6 inhibitor resistance in breast cancer by regulating circHIAT1/miR-19a-3p/CADM2 axis

Abstract

Breast cancer (BC) cells have a high risk of metastasis due to epithelial-mesenchymal transition (EMT). Palbociclib (CDK4/6 inhibitor) is an approved drug for BC treatment. However, the drug resistance and metastasis can impair the treatment outcome of Palbociclib. Understanding the mechanisms of EMT and Palbociclib drug resistance in BC is conducive to the formulation of novel therapeutic strategy. Here, we investigated the role of circHIAT1/miR-19a-3p/CADM2 axis in modulating EMT and Palbociclib resistance in BC. circHIAT1 and CADM2 were down-regulated in BC tissues and cell lines, and miR-19a-3p showed an up-regulation. circHIAT1 could interact with miR-19a-3p and suppress its activity, while miR-19a-3p functioned to negatively regulate CADM2. Forced over-expression of circHIAT1 could impaired the EMT status and migratory ability of BC cells, and this effect was inhibited by miR-19a-3p mimic. In addition, we also generated Palbociclib resistant BC cells, and showed that circHIAT1 and CADM2 were down-regulated in the resistant BC cells while miR-19a-3p showed an up-regulation. Forced circHIAT1 over-expression re-sensitized BC cells to Palbociclib treatment. Quercetin, a bioactive flavonoid, could suppressed the migration and invasion of BC cells, and re-sensitized BC cells to Palbociclib. The anti-cancer effect of quercetin could be attributed to its regulatory effect on circHIAT1/miR-19a-3p/CADM2 axis. In vivo tumorigenesis experiment further revealed that quercetin administration enhanced the anti-cancer effect of Palbociclib, an effect was dependent on the up-regulation of circHIAT1 by quercetin. In summary, this study identified quercetin as a potential anti-cancer compound to reverse Palbociclib resistance and impair EMT in BC cells by targeting circHIAT1/miR-19a-3p/CADM2 axis.

Fig 1. circHIAT1 negatively targets miR-19a-3p in BC.

(A). qRT-PCR analysis of circHIAT1 levels in BC tumor and para-cancerous normal tissues (n = 20 pairs); (B). qRT-PCR analysis of miR-19a-3p levels in BC tumor and para-cancerous normal tissues (n = 20 pairs); (C). Starbase prediction of interacting sites between circHIAT1 and miR-19a-3p; (D). Dual luciferase reporter assay using circHIAT1 WT or MUT reporter in MCF-7 cells; (E) RNA pull-down analysis using biotin-labeled miR-NC (negative control) or miR-19a-3p probe in MCF-7 cells. qRT-PCR was conducted to examine the relative enrichment of circHIAT1. (F). qRT-PCR analysis of circHIAT1 levels in MCF-10A (normal breast epithelial cell) and BC cell lines (MCF-7 and MDA-MB-231); (G). qRT-PCR analysis of miR-19a-3p levels in MCF-10A (normal breast epithelial cells) and BC cells (MCF-7 and MDA-MB-231); (H) qRT-PCR analysis of circHIAT1 levels after the transfection of empty vector or circHIAT1 expression vector; (I). qRT-PCR analysis of miR-19a-3p and an unrelated miRNA (miR-134-5p, not predicted to interact with circHIAT1) expression levels after the transfection of empty vector or circHIAT1 expression vector. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Fig 2. circHIAT1/miR-19a-3p axis regulates CADM2 expression in BC cells.

(A). Starbase prediction of interacting sites between miR-19a-3p and CADM2 mRNA 3’UTR; (B). Dual luciferase reporter assay using CADM2 WT or MUT reporter; (C). RNA pull-down analysis using biotin-labeled miR-NC (negative control) or miR-19a-3p probe in MCF-7 cells. qRT-PCR was conducted to examine the relative enrichment of CADM2 mRNA; (D) Western blot analysis of CADM2 level in MCF-10A (normal breast epithelial cells) and BC cells (MCF-7 and MDA-MB-231). (E-H). BC cells were transfected with empty vector, circHIAT1 expression vector, circHIAT1 expression vector+miR-NC, or circHIAT1 expression vector+miR-19a-3p mimic. qRT-PCR was conducted to analyze the relative expression of (E) circHIAT1, (F) miR-19a-3p, (G) CADM2 mRNA; (H) Western blot analysis of CADM2 protein levels in above conditions; (I-L). BC cells were transfected with si-NC, si-circHIAT1, si-circHIAT1+Inh-NC, or si-circHIAT1+miR-19a-3p inhibitor. qRT-PCR was conducted to analyze the relative expression of (I) circHIAT1, (J) miR-19a-3p, (K) CADM2 mRNA; (L) Western blot analysis of CADM2 protein levels in above conditions. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Fig 3. circHIAT1/miR-19a-3p/CADM2 axis modulate EMT status and the mobility of BC cells.

(A). IHC staining of CADM2, E-cadherin and N-cadherin in BC tumor and para-cancerous tissues. BC cells were transfected with empty vector, circHIAT1 expression vector, circHIAT1 expression vector+miR-NC, circHIAT1 expression vector+miR-19a-3p mimic, or circHIAT1 expression vector+miR-19a-3p mimic+ CADM2 expression vector. (B). Western blot analysis of E-cadherin and N-cadherin in above conditions. (C). CCK-8 proliferation assay (D). Transwell invasion assay and (E). Scratch assay were performed in above groups of BC cells. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Fig 4. Quercetin suppresses migration and invasion of BC cells at sub-toxic dose.

(A). Cytotoxicity of quercetin on MDA-MB-231 and MCF-7 cell lines were assessed by LDH release assay after treating the cells with 0 μM, 10 μM, 20 μM, 40 μM, 80 μM, 160 μM quercetin for 48 hours; (B). At 10 μM of quercetin, CCK-8 assay was performed to determine the anti-proliferation effect of quercetin in BC cells; (C). Transwell invasion assay (D). Scratch assay was performed in BC cells treated with 10 μM of quercetin for 48 hours; (E). Western blot analysis of CADM2, E-cadherin and N-cadherin in BC cells treated with 10 μM of quercetin for 48 hours; (F). qRT-PCR analysis of CADM2, circHIAT1 and miR-19a-3p in BC cells treated with 10 μM of quercetin for 48 hours. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Fig 5. The anti-cancer effect of quercetin depends on circHIAT1/miR-19a-3p axis.

MCF-7 cells were treated with 10 μM quercetin in the presence of control siRNA, circHIAT1 siRNA or circHIAT1 siRNA plus miR-19a-3p inhibitor. (A). qRT-PCR was performed to analyze circHIAT1 and miR-19a-3p expression; (B) Western blot analysis of CADM2, E-cadherin and N-cadherin; (C). CCK-8 proliferation assay; (D). Transwell invasion assay; and (E). Scratch assay in above conditions.*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Fig 6. Quercetin re-sensitizes BC cells to CDK4/6 inhibitor Palbociclib by targeting circHIA.

(A). To investigate the effect of quercetin on Palbociclib drug resistance, we continuously exposed BC cells to increasing doses of Palbociclib. CCK-8 assay showed that drug-resistant MCF7 and MDA-MB-231 cells were able to proliferate in the presence of different doses of Palbociclib. (B). qRT-PCR was performed to analyze circHIAT1, miR-19a-3p and CADM2 expression in parental and drug-resistant cells; (C). Drug-resistant BC cells were transfected with empty vector or circHIAT1 expression vector. CCK-8 assay was performed in the presence of 0.8 μM Palbociclib; (D). LDH cytotoxicity assay was performed in above (C) conditions; (E). To further study whether quercetin impacts on Palbociclib resistance, drug-resistant BC cells were treated with 10 μM quercetin for 24 hours before Palbociclib treatment. Cell proliferation ability of BC cells was assessed under 0.8 μM Palbociclib; (F). LDH cytotoxicity assay was performed in above (E) conditions; (G). CCK-8 proliferation assay was performed in drug-resistant BC cells treated with 10 μM quercetin and/or with circHIAT1 silencing. Palbociclib was applied at 0.8 μM in all conditions; (H). LDH cytotoxicity assay was performed in above (G) conditions. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.

Fig 7. Quercetin boosts the anti-cancer effect of Palbociclib on BC cells in mouse xenograft model.

To evaluate the in vivo effect of quercetin on drug-resistant BC cells, nude mice were injected with drug-sensitive or drug-resistant MCF-7 cells. The drug-resistant group was also administered with quercetin, or quercetin and circHIAT1 siRNA. All the mice were treated with Palbociclib to assess the drug sensitivity. (A). Images of xenograft tumors in each group; (B). Summary of tumor weight; (C). IHC staining of Ki-67 cell proliferation marker in tumor sections; (D). qRT-PCR was performed to analyze circHIAT1, miR-19a-3p and CADM2 expression in the tumor samples of each group. (E). Western blot analysis of N-cadherin and E-cadherin protein levels in the tumor samples of each group. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001.