PLK1 inhibitor facilitates the suppressing effect of temozolomide on human brain glioma stem cells

Abstract Glioblastoma is the most frequent and most aggressive brain tumour in adults. Temozolomide is an oral chemotherapy drug and one of the major components of chemotherapy regimens used as a treatment of some brain cancers. We examined the tolerance of stem cells isolated from glioma cell line U87 and U251 to temozolomide (TMZ) and explored the effect of PLK1 (Polo like kinase 1) protein expression on TMZ sensibility. In our results, the inhibitory effects of TMZ on glioma cells U87, U251 and its stem cells were confirmed to be dose dependent and time dependent. Compared with glioma cells, the glioma stem cells showed a greater degree of tolerance. As the concentration of TMZ increased, the expression of PLK1 protein increased in U87 cells, CD133+ U87 stem cells and CD133‐ U87 cells. The increase range of PLK1 protein was large in CD133+ U87 stem cells and small in CD133‐ U87 cells. TMZ treatment in cells with low PLK1 protein expression efficiently suppressed the cell proliferation and sphere formation, while G2/M arrest was strongly induced. What's more, TMZ and PLK1 inhibitor synergize to inhibit glioma growth in vivo. In conclusion, our results suggest that down‐regulation of PLK1 protein enhanced the inhibition of TMZ on glioma stem cells, suggesting its clinical value to adverse TMZ resistance in glioma treatment.

cancer stem-like properties was suppressed, such as SOX2 protein expression. 10 Hao et al demonstrated that resveratrol combination with TMZ had significant efficacy on glioblastoma progression. 11 Some studies also explored the effects of TMZ on glioblastoma cells activities. For example, miR-146b-5p enhanced cell duplication and impeded cell apoptosis through mediating TMZ resistance in glioblastoma cells. 12 The mechanism of TMZ interacting with other molecules in glioma cells is crucial to the therapies of glioma.
Polo-like kinase 1 protein (PLK1) was reported to highly express in various tumours. By overriding the G2-M DNA damage and spindle checkpoints, overexpression of PLK1 can promote chromosome aneuploidy and instability. 13 Chemical inhibitors or knockdown of PLK1 decreased medulloblastoma cells growth. 13 Robin et al illuminated that PLK1 was promoted in CD133-positive cells and combined inhibition of PLK1 and BRAF resulted in significantly greater pro-apoptotica and anti-proliferative effects than those achieved by monotherapy. 5 Koncar et al researched the interaction of TMZ and PLK1 in glioma, and reported that combination treatment of TMZ and a PLK1 inhibitor BI2536 caused significant cancer shrinkage and tumour regression in in vivo experiments, while TMZ or BI2536 alone had little effect on tumour growth. 14 The influence of TMZ and PLK1 on glioma cellular activities needs to be further studied.
In this study, we evaluated the effects of PLK1 on glioblastoma and the synergistic inhibition effect of PLK1 inhibitor combined with TMZ on human brain glioma stem cells in vitro and vivo. Our study suggested that PLK1 inhibitors may be a novel therapies target for glioma treatment.

| U87 and U251 CD133-positive cells isolation and culture
The human glioblastoma cell line U87 and U251 was obtained commercially from ATCC and were cultured in Dulbecco's modified Eagle's medium (DMEM, Invitrogen, Carlsbad, CA) supplemented with 10% bovine serum and 100 μg/mL streptomycin. For the isolation, U87 and U251 cells were suspended at FcR reagents were added for blocking. Microbeads cultured with CD133 antibody (ab19892, Abcam, Cambridge, MA) were then added, and the mixture was cultured at 37°C for 1 hour. Cells collected was recognized as CD133-fractions while cells obtained after removing the magnetic holder was diagnosed as CD133 + cells, also as glioma stem cells. Glioma stem cells were cultured in a serum-free DMEM-F12 medium (Invitrogen) supplemented with 10 ng/mL basic fibroblast growth factor (bFGF, Invitrogen), 20 mg/mL epidermal growth factor (EGF, Invitrogen) and 2% B27 (Invitrogen) under 5% CO 2 at 37°C.

| Cell viability assay
The human brain glioma U87 cells, U251 cells and stem cells were correspondingly seeded in 10% FBS cell culture medium and stem cell culture medium. The culture medium was moved and DMEM-F12 (Hyclone, Logan, UT, USA) was added. TMZ (human glioma U87, CD133-U87, U251 and CD133-U251 cells: serum medium + TMZ, human glioma U87 and U251 stem cells: serum-free medium + TMZ) were added to 96-well plates at stated concentration. Cells were collected at incubation for 24, 48 and 72 hours and then treated with CCK-8 (10 μL/well, Beyotime, Shanghai, China) reagent for another 5 hours. The absorbance at 450 nm was measured by an automatic enzyme-linked immune detector (Multiskan MK3, Thermo Labsystems, Helsinki, Finland).

| Flow cytometry assay
Monolayer cultured human glioma CD133-U87 and CD133-U251  GAPDH was considered as the internal reference.

| Soft agar colony formation assay
The bottom layer of soft agar (0.9%) was prepared in a six-well plate, and the top layer (0.5%) was prepared with 5 × 10 4 /mL CD133 + stem cells, CD133-glioma cells and U87 or U251 cells in single-cell suspension. The cells were exposed to 40 μmol/L TMZ for 2 weeks and cultured in an incubator at 37°C with 5% CO 2 . Colony formation was observed by microscopy, and colonies of >30 cells were counted under a microscope. The experiments were repeated in triplicate.

| GSC sphere-forming assays
GSC spheres were enzymatically dissociated to single cells and replated in 96-well plates at optimal density (500-1000 cells) in nonadherent conditions. Cells were cultured in serum-free DMEM-F12 medium (Invitrogen) supplemented with 10 ng/mL bFGF, 20 ng/mL EGF and 2% B27. Half of the medium was renewed every other day, and after 6 days, the cells were fixed with 4% formalin. The cells were photographed and spheres larger than 50 μm were counted.

| Haematoxylin and eosin (HE) staining
The tumour tissues obtained from mice were fixed with 10% formaldehyde, embedded in paraffin, cut into 4 μm sections and

| Statistical analysis
Each assay was conducted in triplicate. Statistical analysis was performed using GraphPad Prism 6.0 software (GraphPad Software, La Jolla, CA). Student's t test was applied to compare the differences between two groups, while the differences between multi-samples were analysed by analysis of variance (ANOVA). P value of <0.05 was considered statistically significant. showing strong tolerance ( Figure 1C, P < 0.01). However, there was no significant difference between CD133-U87 cells and U87 cells.

| TMZ suppressed the cell viability and induced cell cycle arrest of glioma cells and glioma stem cells
The results of CCK-8 showed that the inhibitory effect of TMZ on cells increased with time. The inhibitory effect on CD133 + U87 cells was significantly weaker than that on U87 cells, but the inhibitory effect on U87 CD133-cells was no significant difference with that on U87 cells ( Figure 1D, P < 0.01). Soft agar assay suggested that the clone size and quantity in CD133 + U87 cells were more than CD133-U87 cells and U87 cells after exposing to 40 μmol/L TMZ for 2 weeks ( Figure 1G, P < 0.01). Similar results were observed in U251 cells, CD133 + U251 cells and CD133-U251 cells ( Figure 1E,F,H, P < 0.01). These results indicated that glioma stem cells were more tolerant to TMZ than glioma cells.
We further examined the effect of TMZ on the cell cycle (

| PLK1 was involved in TMZ tolerance
We

U87 stem cells and CD133 + U251 stem cells
To further verify the effect of PLK1 on tolerance in glioma stem cells, we inhibited or increased the expression of PLK1 protein in CD133 +

U87 stem cells and CD133 + U251 stem cells. PLK1 inhibitor BI2536
and Volasertib had no significant impact on PLK1 mRNA expression.

| PLK1 inhibitor displayed synergistic activity with TMZ to inhibit CD133 + U87 stem cells growth in vivo
We also examined whether the combination of PLK1 inhibitor and TMZ displays synergistically anti-glioma effects in vivo. PLK1 inhibitor (BI2536 and Volasertib) and TMZ observably inhibited CD133 + U87 stem cells growth in vivo compared with control or Blank groups.
Strikingly, combinatorial treatment with both drugs (TMZ plus BI2536 or TMZ plus Volasertib) resulted in synergistically reduced tumour growth rates ( Figure 6A, P < 0.01  Figure 6C). Western blot results suggested that PLK1 protein was notably down-regulated by TMZ, BI2536 and Volasertib treatment and was lower in TMZ plus BI2536 and TMZ plus Volasertib treatment ( Figure 6D, P < 0.01).
These results indicated that inhibition of PLK1 by BI2536 or Volasertib enhanced sensitivity to TMZ of CD133 + U87 stem cells in vivo.  Relevant studies have revealed the PLK1 was involved in the mechanism of cancer growth. For instance, PLK1 was elevated in glioblastoma multiforme cells and its inhibition suppressed cell growth and induced cell death. 19 In medulloblastoma, down-regulation of PLK1 impaired tumour sphere formation of medulloblastoma cells and induced cell apoptosis. 13 Additionally, Koncar et al demonstrated that PLK1 inhibition enhanced TMZ efficacy in IDH1 mutant gliomas. 14 Additional modalities of TMZ-resistant such as PLK1 expression further complicate the mechanism of glioma drug resistance. Hence, we employed experiments to examine the interaction of PLK1 and TMZ in TMZ-resistant glioma cells, and found that PLK1 expression was impeded with the increase in TMZ concentration.

| DISCUSSION
Furthermore, PLK1 inhibitor or knockdown of PLK1 facilitated the inhibitory effect of TMZ on cell viability and cell cycle.
Although we have elucidated the links between PLK1 expression and TMZ efficacy on U87 and U251 stem cells, the current study also has some unavoidable limitations. For example, other molecules involved in drug-resistant glioma stem cells are worthy to be studied.
In summary, our study proposed the synergistic inhibition effect of PLK1 inhibitor with TMZ on glioblastoma stem-like cells and suggested the critical role of PLK1 in glioma stem cells progression, providing new treatment strategies for gliomas.