MiR-34a chemosensitizes bladder cancer cells to cisplatin treatment regardless of p53-Rb pathway status

Abstract

MiR-34a is a downstream effector of p53 that has been shown to target several molecules associated with cell cycle and cell survival pathways. As alterations in these pathways are frequent in muscle invasive transitional cell carcinoma of the bladder (MI-TCC), for example mutation or loss of p53 and Rb, the goal of this study was to determine whether manipulation of miR-34a expression levels could abrogate the effect of these alterations and sensitize bladder cancer cells to chemotherapy. We demonstrate that transfection of T24, TCCSUP and 5637 with pre-miR-34a followed by cisplatin treatment results in a dramatic reduction in clonogenic potential and induction of senescence compared to treatment with cisplatin alone. Molecular analyses identified Cdk6 and sirtuin (SIRT)-1 as being targeted by miR-34a in MI-TCC cells, however, inhibition of Cdk6 and SIRT-1 was not as effective as pre-miR-34a in mediating chemosensitization. Analysis of 27 preneoadjuvant chemotherapy patient samples revealed many of the patients who subsequently did not respond to treatment (based on surgical resection postchemotherapy and 5-year survival data) express lower levels of miR-34a, however, a statistically significant difference between the responder and nonresponder groups was not observed (p = 0.1174). Analysis of eight sets of pre- and postneoadjuvant chemotherapy patient samples determined miR-34a expression increased postchemotherapy in only two of the eight patients. The combined data indicate that elevation of miR-34a expression levels before chemotherapy would be of benefit to MI-TCC patients, particularly in a setting of low miR-34a expression.

Figure 1. Basal miR-34a expression levels correlate with the relative chemo-sensitivity of bladder cancer cell lines to cisplatin treatment

As expected comparison of clonogenic assay (d10 post-treatment) versus WST-1 assay (d3 post-treatment) determined that the effect of cisplatin treatment is most potent after several rounds of cell division; the IC50 values for cisplatin were 3 to 35-fold lower by clonogenic assay compared to WST-1 assay (A and B). Cisplatin concentrations of 0.5uM and 5uM were used for all subsequent long term and short term assays respectively. TCCSUP, the most chemosensitive of the 3 TCC cell lines, expressed the highest endogenous levels of miR-34a (C), a fact that is likely due to methylation of the miR-34a promoter in T24 and 5637, but not in TCCSUP (D) and to TCCSUP expressing wildtype p53 (E). All 3 TCC sublines harbor mutations in the p53-Rb signaling axis; both T24 and 5637 harbor p53 mutations, TCCSUP and 5637 harbor Rb mutations (E). The 5637-resistant subline expressed ~3-fold lower miR-34a compared to the parental 5637-sensitive subline (F). This result further validates a connection between miR-34a expression level and chemosensitivity in TCC cells. (* P < 0.05)

Figure 2. Increased miR-34a expression chemosensitizes TCC cells to cisplatin treatment by reducing clonogenic potential and inducing senescence

Transfection of the T24 and 5637 cell lines with pre-miR-34a prior to cisplatin treatment caused a dramatic reduction in clonogenic potential in all 3 TCC cell lines, and a dramatic increase in cellular senescence in T24, the most chemoresistant of the 3 TCC cell lines (B and C). Morphological changes were also apparent (C). Little or no increase in apoptosis was observed in any of the TCC cell lines transfected with pre-miR-34a prior to cisplatin treatment relative to cisplatin treatment alone (D). (* P < 0.05)

Figure 3. MiR-34a targets Cdk6 and SIRT-1 in TCC cell lines

We chose to focus on targets of miR-34a that are associated with the p53-Rb signaling axis (A) as this pathway is frequently perturbed in MI-TCC and several of these perturbations have been linked to chemoresistance. Pre-miR-34a consistently targeted both SIRT-1 and Cdk6 in all 3 TCC cell lines (B). In T24, miR-34a-mediated targeting of Cdk6 resulted in decreased phosphorylation of Rb. E2F3 expression also decreased, although whether this was due to direct targeting by miR-34a or simply to decreased Rb phosphorylation is unclear. E2F3 mRNA and protein expression did not decrease in TCCSUP or 5637, both of which harbor Rb mutations (B and C). MiR-34a did not target Bcl-2 mRNA or protein expression in any of the 3 TCC cell lines (B and C), and increased miR-34a appeared to be associated with increased Bcl-xL expression.

Figure 4. Decreased miR-34a expression causes chemo-resistance, inhibition of individual effectors of miR-34a has only a minor impact on chemosensitization

Transfection of TCCSUP, the most chemo-sensitive of the 3 TCC cell lines, with anti-miR-34a prior to treatment with cisplatin caused an increase in clonogenic potential (A) and increased Cdk6 and SIRT-1 expression (B), furthering validating a connection between miR-34a and chemosensitivity and establishing Cdk6 and SIRT-1 as important downstream effectors of miR-34a in TCC cells. Down-regulation of Cdk6 (C and D) or SIRT-1 (E), both proven targets of miR-34a in TCC cells, using siRNA (Cdk6 siRNA, 50nM) or pharmacological inhibitors (BML-210, (SIRT-1 inhibitor), 10uM) did not confer the same level of chemo-sensitization as transfection with pre-miR-34a. Combined inhibition of Cdk6 and SIRT-1 further sensitized T24 to cisplatin (F, ~40% reduction in clonogenic potential) but was still less effective than transfection with pre-miR-34a (Figure 2A, ~50–60% reduction in clonogenic potential). (* P < 0.05)

Figure 5. Chemotherapy causes increased expression of miR-34a regardless of p53 status in TCC cell lines

Unexpectedly, cisplatin caused increased miR-34a expression in all 3 TCC cell lines including T24 which are effectively p53 null and 5637 which harbor mutant p53 (A-E). (* P < 0.05)

Figure 6. Analysis of miR-34a expression in MI-TCC samples from patients who respond versus do not respond to chemotherapy

Analysis of miR-34a expression levels in archival tumor samples that had been obtained from TCC patients prior to treatment (27 patients, 15 of whom subsequently responded to treatment (responders) versus 12 who did not (non-responders), response based on surgical resection post-chemotherapy and 5 year survival data, (A)) determined that many of the non-responders express low levels of miR-34a, however, the difference between responders and non-responder groups was not statistically significant, p=0.1174 (B). Similar rates of p53 mutation were observed in the responder and non-responder groups (60% versus 58%, (A)). Methylation miR-34a promoter was more frequent in the responder group (A), however, miR-34a expression levels did not correlate with p53 status (C). Analysis of 8 matched pre- and post-neoadjuvant chemotherapy samples from patients who did not respond to chemotherapy revealed that in most cases miR-34a expression levels do not change dramatically following chemotherapy treatment, an increase in miR-34a was observed in only 2 of 8 patients (D).