Cycling Quiescence in Temozolomide Resistant Glioblastoma Cells Is Partly Explained by microRNA-93 and -193-Mediated Decrease of Cyclin D

Abstract

Glioblastoma multiforme (GBM) is a fatal malignancy of the central nervous system, commonly associated with chemoresistance. The alkylating agent Temozolomide (TMZ) is the front-line chemotherapeutic agent and has undergone intense studies on resistance. These studies reported on mismatch repair gene upregulation, ABC-targeted drug efflux, and cell cycle alterations. The mechanism by which TMZ induces cell cycle arrest has not been well-established. TMZ-resistant GBM cells have been linked to microRNA (miRNA) and exosomes. A cell cycle miRNA array identified distinct miRNAs only in exosomes from TMZ-resistant GBM cell lines and primary spheres. We narrowed the miRs to miR-93 and -193 and showed in computational analyses that they could target Cyclin D1. Since Cyclin D1 is a major regulator of cell cycle progression, we performed cause-effect studies and showed a blunting effects of miR-93 and -193 in Cyclin D1 expression. These two miRs also decreased cell cycling quiescence and induced resistance to TMZ. Taken together, our data provide a mechanism by which GBM cells can exhibit TMZ-induced resistance through miRNA targeting of Cyclin D1. The data provide a number of therapeutic approaches to reverse chemoresistance at the miRNA, exosomal and cell cycle points.

Keywords: Cyclin D; cell cycle; chemoresistance; glioblastoma; microRNA.

FIGURE 1

miRNA profile in TMZ resistant GBM cells (U87 and T98G). (A) Exosomes were collected from vehicle- and TMZ-treated GBM cells and then analyzed for CD63 and CD81 by western blot. The membrane was stripped and reprobed for β-actin. (B) The carton (top) demonstrates how exosomes were immunoprecipitated with microbeads conjugated to anti-CD63. The microbeads were incubated with exosomes from vehicle- or TMZ-resistant GBM cells. After this, the beads were incubated with anti-CD81-PE and anti-CD63-FITC. Control beads were incubated with isotype control. The beads were analyzed by flow cytometry: red, negative/isotype control, blue untreated, yellow TMZ-treated). (C) Additional analyses of the exosomes were done by NTA. A represented histogram is shown demonstrating the average size of 100 nm. (D) The miRNAs from the arrays in TMZ-resistant cells and naïve (untreated and vehicle treatment) GBM cells. The results are presented as 2^ΔCT (n = 3, ±SD).

FIGURE 2

Validation of miRNAs from TMZ-resistant GBM cells and identifying the target. (A) Real time PCR was performed for the miRNAs shown to be increased in array studies (Figure 1D). Real-time PCR for miR-19, -23a/b, -93, -193 and -373 was performed with RNA from vehicle-treated and TMZ-resistant U87 and T98G cells. The values for vehicle were assigned 1 to calculate fold changes in the TMZ-resistant GBM cells (n = 4 ± SD). (B) Real time PCR was performed for the miRNAs studied in “A” using RNA from low-passage BT164 (TMZ resistant) and BT145 (TMZ naïve). The results are presented as the ratio of normalized Ct values of BT164/BT145 (n = 4, ±SD). ND = none detected. (C) Ingenuity^® miRNA analyses identified specific miRNAs as targets for CCND1 (Cyclin D1). (D) The miRNAs, identified in “C,” are shown for predictive binding sites on the 3′UTR of cyclin D1 (CCND1).

FIGURE 3

Decrease in Cyclin D1 in TMZ resistant GBM cells and cycling quiescence. (A) Real time PCR was performed with RNA from vehicle- and TMZ-treated U87 and T98G for Cyclin D1 mRNA. The normalized values for vehicle treatment are assigned 1 and the treated cells are presented as fold change (±SD, n = 4). (B) Western blots were performed with cell extracts from vehicle and TMZ-treated U87 and T98G for Cyclin D1. The blot was stripped and reprobed for β-actin. The normalized band densities are presented as the mean ± SD of three different experiments. (C) U87 and T98G cells were labeled with 7-AAD in three independent experiments. The figure shows representative histograms. ^∗ p < 0.05 vs. vehicle treatment.

FIGURE 4

Cyclin D1 in TMZ-resistant, Dicer knockdown GBM cells. U87 and T98G cells were transfected with Dicer siRNA or control siRNA. After this, the cells were treated with 200 μM TMZ for 72 h and then analyzed for Cyclin D1 mRNA by real time PCR (A) and western blot (B). Normalized band densities for three western blots (±SD) are shown in the graph of “A.” ^∗p < 0.05 vs. control siRNA.

FIGURE 5

Cyclin D1 expression in miR-93/-193 GBM transfectants treated with TMZ. (A) Real time PCR for control miR, miR-93 and miR193 with RNA from U87 and T98G, transfected with the respective pre-miRs. The results are presented as the mean (±SD, n = 3) fold change of control. The controls for both primers are set at values of 1. (B) The transfectants in “A” were studied for cyclin D1 and the results are presented as mean fold change (±SD, n = 3) over the transfectants with control pre-miR. The control pre-miR was assigned values of 1. (C) The transfectants in “A” were treated with 200 μM TMZ. After 72 h, the viable cells were studied for Cyclin D1. An additional control with untransfected cells was included in the blot. The mean (±SD, n = 3) normalized band densities are shown in the graphs. ^∗ p < 0.05 vs. control miR.

FIGURE 6

Cell cycle phase of TMZ-treated GBM cells, transfected with pre-miRs. U87 (A and C) and T98G (B and D) cells were transfected with control pre-miRs, pre-miR-93 or pre-miR-193. The cells were treated with vehicle or 200 μM TMZ. After 72 h the cells were analyzed for cell cycle by propidium iodide (miR-93 and -193) treatment and 7-AAD for control miR. The results represent three independent experiments.

FIGURE 7

Active caspase 3 in TMZ and pre-miR-93 and -193 transfected GBM cells. (A) U87 and T98G cells were untransfected for transfected with control pre-miR, pre-miR-93 or -193. The cells were treated with vehicle or 200 μM TMZ for 72 h. Whole cell extracts were analyzed in western blots for total caspase 3 and cleaved caspase 3. The normalized band densities were presented as the % caspase activity ((cleaved caspase/total + cleaved caspase) × 100%) (lower graphs). (B) U87 and T98G were transfected with control pre-miR, pre-miR-93 or -193. The cells were treated with vehicle or 200 M TMZ for 72 h. The cells were labeled intracellularly for cleaved caspase 3 (open histogram) and total caspase (solid histogram).