Topoisomerase IIβ mediates the resistance of glioblastoma stem cells to replication stress-inducing drugs

Abstract

Background: Glioblastoma stem cells (GSC) have been extensively recognized as a plausible cause of glioblastoma resistance to therapy and recurrence resulting in high glioblastoma mortality. Abnormalities in the DNA repair pathways might be responsible for the inability of the currently used chemotherapeutics to eliminate the (GSC) subpopulation.

Methods: In this work, we compared the expression of sixty DNA repair related genes between primary glioblastoma cell cultures and the glioblastoma enriched stem cell primary cultures. MTT test was used to analyze the effect of selected drugs and immunofluorescence to evaluate the load of DNA damage.

Results: We found several differentially expressed genes and we identified topoisomerase IIβ (Top2β) as the gene with highest up-regulation in GSC. Also among the tested cell lines the expression of Top2β was the highest in NCH421k cells, a well-characterized glioblastoma cell line with all the stemness characteristics. On the other hand, Top2β expression markedly decreased upon the induction of differentiation by all trans-retinoic acid. Depletion of Top2β increased the sensitivity of NCH421k cells to replication stress inducing drugs, such as cisplatin, methyl-methanesulfonate, hydrogen peroxide, and temozolomide. Consistently, we found an increased load of DNA damage and increased Chk1 activation upon Top2β depletion in NCH421k cells.

Conclusion: We suggest that Top2β may represent a new target for gene therapy in glioblastoma. In addition, the other genes that we found to be up-regulated in GSC versus glioblastoma primary cells should be further investigated as glioblastoma theranostics.

Keywords: Drug resistance; Glioblastoma stem cells; Glioma; Replication stress; Theranostic markers; Topoisomerase IIβ.

Fig. 1

Expression of Top2β in different cell types. a Expression of Top2β at mRNA level in established GBM cell lines U87, U251, and T98G, established GSC cell line NCH421k (i.e. NCH) and ATRA-differentiated NCH421k (i.e. NCH-A), normal human neural stem cells (NSC) and normal human astrocytes, measured by RT-PCR, expression normalized to β-actin, average ± SD is presented. b Top2β protein expression in established GBM and GSC cell lines as above, as determined using Western Blotting, where β-actin was used as loading control. The numbers below represent the quantitative expression of Top2β normalized to β-actin

Fig. 2

Correlation of Top2β expression and differentiation. The expression of Top2β in NCH421k cells was measured at the indicated time points after ATRA-treatment; GFAP was used as a marker of differentiation, β-actin was used as a loading control. Below the quantification of Top2β expression normalized to actin

Fig. 3

Top2β level dependent sensitivity to replication stress inducing drugs. Survival of U251 and NCH421k cells after exposure to increasing concentrations of selected drugs was determined by MTT assay as described in “Methods”. Drugs used were H₂O₂, CisPt, etoposide, CPT, MMS and TMZ. Green triangles U251 cells, blue diamond NCH421k control cells (siLuc), red squares NCH421k with Top2β down-regulation; average of three independent experiments ± SD is presented, asterisk indicates significance at p < 0.05. In the inset WB showing the efficiency of Top2β depletion in NCH421k cells

Fig. 4

Sensitivity of primary GSC to MMS and CisPt. Top2β was depleted in primary GSC cultures from 2 different patients (sample 1 and 2). Cells were treated with 200 µM MMS or 50 µM CisPt and survival was determined by MTT assay. Average of 2 independent experiments ± SD is presented

Fig. 5

Increased load of DNA damage in Top2β depleted cells. a DNA damage was induced using 200 μM methyl-methanesulphonate (MMS) or 50 μM cisplatin (CisPt) in control (Luc) and Top2β down-regulated NCH421k cells. Cells were stained for yH2AX (red) and counterstained with Toto3 (blue). Representative images are shown, scale bar 10 µm. b Number of foci per cell was quantified as described in “Methods” and is presented as the average of three experiments ± sd. c Chk1 activation after MMS and CisPt treatment, tubulin is a loading control. Below quantification of pChk1 normalized to tubulin