Cell cycle progression in glioblastoma cells is unaffected by pathophysiological levels of hypoxia

Cell culture and hypoxic treatment

All reagents were purchased from Life Technologies, unless otherwise stated. U87 cells (ATCC, HTB-14), U251 cells (CLS, 300385) and D566 cells (a kind gift from Professor DD Bigner, Duke University Medical Centre, USA) were maintained in MEM supplemented with 1% sodium-pyruvate and 10% foetal bovine serum (FBS). U251 and D566 cells were supplemented with 1% non-essential amino acids (NEAA). HeLa cells (ECACC, 93021013) were maintained in MEM plus 10% FBS and 1% NEAA. All cells were maintained at 37 °C in 5% CO₂. For flow cytometry experiments, 1 × 10⁵ cells were seeded in 6 cm tissue culture dishes (Corning). For hypoxic experiments, cells were incubated in a Don Whitley H35 Hypoxystation (1% O₂) or a New Brunswick Galaxy 48R hypoxic incubator (0.1% and 8% O₂). A media change was performed after 48 h.

Tumour dissection and primary culture

Samples of primary GBM tumours were received from patients undergoing craniotomy and resection. All patients gave informed written consent to donate their tissue to the Walton Research Tissue Bank, Walton Centre NHS Foundation Trust, which has full approval of the National Research Ethics Service (11/WNo03/2). Primary cell culture was carried out in accordance with the approved guidelines. Tumour samples were transported in MEM plus 1% penicillin-streptomycin (pen-strep), mechanically dissected, and transferred into dissociation medium (10% trypsin 10X and 1% DNase [Sigma] in MEM plus 1% pen-strep). Samples were incubated for 15–30 min at 37 °C and triturated every 5 min. The trypsin reaction was stopped by adding growth medium (MEM plus 20% FBS, 1% sodium-pyruvate and 1% pen-strep). Cells were centrifuged for 5 min at 3,000 rpm, resuspended in growth medium and seeded into a 75 cm² tissue culture flask.

Flow cytometry

For the viability analyses, adherent cells were washed with phosphate buffered saline (PBS), trypsinised and suspended in Hank’s balanced saline solution (HBSS) together with the floating fraction of cells. The cell suspension was pipetted into a 96 well plate, stained with FITC Annexin V (1:500) and incubated for 15 min in the dark at room temperature (RT). Propidium iodide (PI) was added at a final concentration of 4 µg/mL immediately prior to analyses. Samples were analysed using a Guava EasyCyte Flow Cytometer and cell viability was established using GuavaSoft software (Millipore). For the cell cycle analyses, cells were washed with PBS, trypsinised and resuspended in PBS with 0.1% TritonX-100. PI and ribonuclease A were added at a final concentration of 10 µg/mL. Cells were analysed by flow cytometry and the percentage of cells in G₀/G₁, S and G₂/M phases was established using ModFit LT (Verity Software House).

Western blotting

30–40 µg protein was resolved on a 10% SDS-polyacrylamide gel, transferred onto a nitrocellulose membrane, and probed with primary antibodies against HIF-1α (BD Biosciences 610959, 1:1000), β-actin (Abcam ab8226, 1:1000) and p21 (Santa Cruz sc-396, 1:500) at 4 °C overnight. Membranes were incubated with either an anti-mouse (1:5000) or anti-rabbit (1:3000) horseradish peroxidase-linked secondary antibody (Cell Signalling) for 1 h at RT. Amersham ECL Prime Western Blotting Detection Reagent (GE Healthcare) was used according to the manufacturer’s instructions prior to detection using a G:BOX gel imaging system (Syngene, UK).

Real time reverse transcription polymerase chain reaction (qRT-PCR)

RNA was extracted using a HP RNA isolation kit (Roche) according to the manufacturer’s instructions. Reverse transcription of 1 µg mRNA to cDNA was carried out using a SuperScript VILO cDNA synthesis kit. Real time polymerase chain reaction (qPCR) was performed in triplicate using 10 µl LightCycler^® 480 SYBR Green I Master (Roche), 6 µl DNase/RNase-free H₂O, 2 µl cDNA and 0.5 µM of each primer. Primers used were as follows: cyclophilin A forward: GCTTTGGGTCCAGGAATGG, reverse: GTTGTCCACAGTCAGCAATGGT; p21 forward: AGCTGCCGAAGTCAGTTCCTT, reverse: GTTCTGACATGGCGCCTCCT; p27 forward: TCCGGCTAACTCTGAGGACA, reverse: GAAGAATCGTCGGTTGCAGG; E2F1 forward: CCGCCATCCAGGAAAAGGTG, reverse: GTGATGTCATAGATGCGCC. qPCR was performed using a LightCycler 480 (Roche); for parameters see Table S1. Results were analysed using LightCycler 480 software (version 1.5.0.39, Roche). The values for target genes were normalised to cyclophillin A and expressed as fold change from control values (20% O₂).

Statistical analyses

Statistical significance was determined by ANOVA using SPSS statistical software (IBM).

Pathophysiological hypoxia has minimal effects on GBM cell proliferation and survival, while long-term exposure to severe hypoxia results in cell death

To investigate the effect of different oxygen levels on cell survival, three GBM cell lines (D566, U87 and U251 cells) were exposed to atmospheric (20% O₂), physiological (8% O₂) and pathophysiological (1% O₂) levels of oxygen for up to 96 h. Cell proliferation and survival were measured using Annexin V/PI staining and cell counting using flow cytometry. D566 and U87 cell proliferation was not affected at any time, while U251 cells only displayed a reduction in cell number after 96 h in 1% O₂ (Fig. 1A). The viability of all three cell lines was not affected by the different levels of oxygen (Fig. 1B). These results suggest that pathophysiological levels of hypoxia do not significantly affect cell proliferation or survival.

We investigated whether severe hypoxia (0.1% O₂) could have a more pronounced effect on cell fate. In both D566 and U251 cells, a significant reduction in cell number was evident after 48 h in 0.1% O₂, while a significant decrease in U87 cell number could only be observed after 96 h (Fig. 2A). A reduction in cell number can be due to a reduction in proliferation or an increase in cell death. The latter was the case for U251 and U87 cells, which showed a significant decrease in viability after 48 h exposure to 0.1% O₂ (Fig. 2B). In both U251 cells (Fig. 2C) and U87 cells (Fig. S1), cell death was predominantly due to necrosis. In contrast, the viability of D566 cells was unaffected by exposure to 0.1% O₂, suggesting that the reduction in D566 cell number may be due to cell cycle arrest.

Pathophysiological hypoxia has no effect on cell cycle distribution, while severe hypoxia induces a transient cell cycle arrest in D566 cells only

As a more direct assessment of the effect of hypoxia on cell proliferation, cells were exposed to 0.1%, 1% or 20% O₂ prior to cell cycle analyses by flow cytometry. There was no difference in cell cycle distribution between cells exposed to 1% O₂ and 20% O₂ in any of the cell lines tested (Fig. 3A). In U87 and U251 cells there were no significant differences in cell cycle distribution between cells exposed to 0.1% and 20% O₂ (Fig. 3A). However, after 48 h in 0.1% O₂ D566 cells displayed a 13% increase (p = .002) in the proportion of cells in G₁ phase. Interestingly, this cell cycle arrest was only transient and was lost in chronic hypoxia (≥72 h) (Fig. 3A, Fig. S2). To confirm that resistance to hypoxia-induced cell cycle arrest is not an acquired characteristic of cell lines maintained in culture, we conducted the same experiment with primary GBM cells isolated from tumour samples obtained at resection (designated GBM03 and GBM04). We saw no evidence of cell cycle arrest in 1% or 0.1% O₂ (Fig. 3B). These results show that only severe hypoxia has the potential to induce cell cycle arrest in GBM cells and, furthermore, this arrest is only transient.

As GBM cells are known to harbour mutations in a number of genes involved in cell cycle control (Catalogue of somatic mutation in cancer (COSMIC), http://cancer.sanger.ac.uk/cosmic), we verified that these cells were capable of undergoing cell cycle arrest by exposing them to 10 µM etoposide for 24 h. All three cell lines displayed a profound S phase arrest (Fig. S3). To investigate whether the observed cell cycle response was specific to GBM cells or common to other cancer cells, we exposed HeLa cells to 0.1%, 1% and 20% O₂ for up to 72 h. As with the GBM cells, HeLa cell cycle distribution was unaffected by exposure to 1% O₂. An increase in cells in G₁ phase was observed following 48 h in 0.1% O₂; however, this increase was not significant and cell cycle distribution returned to normal after 72 h, as previously observed in D566 cells (Fig. 3B).

The effect of hypoxia on the expression of HIF-1α and cell cycle regulatory proteins

We confirmed the ability of GBM cells to respond to hypoxia by measuring HIF-1α stabilisation using western blot (Fig. 4A). In all three cell lines, HIF-1α was stabilised in hypoxia, with increased protein levels in 0.1% compared to 1% O₂ (Fig. 4B). To confirm that hypoxia does not have a significant effect on cell cycle regulation, mRNA levels of the cyclin-dependent kinase inhibitors p21, p27 and the well described transcription factor involved in G₁/S transition, E2F1, were measured by qRT-PCR following up to 96 h in 0.1% O₂ (Fig. 4C). No significant changes in p21, p27, or E2F1 expression were observed in any of the cell lines. D566 cells showed a non-significant increase in p21 expression at 24 h, with expression levels peaking at 48 h and decreasing after 72 h, consistent with the temporary G₁ phase accumulation observed in D566 cells exposed to 0.1% O₂ (Fig. 3A). Western blotting was used to investigate whether this transient increase in p21 mRNA was translated at the protein level. Only a small increase in protein was detected after 24 h exposure to 0.1% O₂ (Fig. 4D). Furthermore, this increase was negligible compared to the amount of p21 detected following treatment with etoposide, a potent inducer of cell cycle arrest (see Fig. S3).