Intermittent hypoxia regulates stem-like characteristics and differentiation of neuroblastoma cells

Abstract

Background: Neuroblastomas are the most common extracranial solid tumors in children. Neuroblastomas are derived from immature cells of the sympathetic nervous system and are characterized by clinical and biological heterogeneity. Hypoxia has been linked to tumor progression and increased malignancy. Intermittent hypoxia or repeated episodes of hypoxia followed by re-oxygenation is a common phenomenon in solid tumors including neuroblastoma and it has a significant influence on the outcome of therapies. The present study focuses on how intermittent hypoxia modulates the stem-like properties and differentiation in neuroblastoma cells.

Methods and findings: Cell survival was assessed by clonogenic assay and cell differentiation was determined by morphological characterization. Hypoxia-inducible genes were analyzed by real-time PCR and Western blotting. Immunofluorescence, real-time PCR and Western blotting were utilized to study stem cell markers. Analysis of neural crest/sympathetic nervous system (SNS) markers and neuronal differentiation markers were done by real-time PCR and Western blotting, respectively. Intermittent hypoxia stimulated the levels of HIF-1α and HIF-2 α proteins and enhanced stem-like properties of neuroblastoma cells. In intermittent hypoxia-conditioned cells, downregulation of SNS marker genes and upregulation of genes expressed in the neural crest were observed. Intermittent hypoxia suppressed the retinoic acid-induced differentiation of neuroblastoma cells.

Conclusions: Our results suggest that intermittent hypoxia enhances stem-like characteristics and suppresses differentiation propensities in neuroblastoma cells.

Figure 1. Regulation of HIF-1α expression in intermittent hypoxia-conditioned human neuroblastoma cells.

Cells were grown under normoxic conditions (N) or exposed to 1% O₂ for 24 h (H). Intermittent hypoxia-conditioned cells were derived from NB1691 cells that were exposed to 10 cycles of hypoxia (1% O₂, 24 h) and reoxygenation. (A) Real-time PCR. Total RNA was extracted from neuroblastoma cells using Trizol and cDNA was generated by reverse transcription. Real-time PCRs were done using primers specific to HIF-1α and β-actin. **P<0.01 hypoxia or intermittent hypoxia versus normoxia. (B) Cell extracts were assessed for HIF-1α and β-actin by immunoblotting. (C) Immunofluorescence. Cells were fixed in ice-cold methanol for 20 min at −20°C. Then, cells were labeled with HIF-1α antibodies and Alexa-488 antimouse-conjugated antibodies. Photomicrographs were taken using Olympus fluorescence microscope. Nuclei were stained with DAPI (bar, 100 µm). (D) Cell lysates were probed for HIF-2α and β-actin by western blotting.

Figure 2. Effects of intermittent hypoxia on VEGF, a hypoxia-response gene and cell survival.

(A) Real-time PCR. Total RNA was extracted from normoxic (N), and intermittent hypoxia (IH) conditioned neuroblastoma cells using Trizol and cDNA was generated by reverse transcription. Real-time PCRs were done to measure VEGF gene transcript. **P<0.01, intermittent hypoxia versus normoxia. (B) Clonogenic assay. Cells were trypsinized, plated into 100-mm dishes, and incubated at 37°C in a humidified incubator containing 5% CO2. After 15 days, cells were stained with crystal violet and colonies having >50 cells were counted as surviving colonies. **P<0.01, intermittent hypoxia versus normoxia.

Figure 3. Effects of intermittent hypoxia on stem-like characteristics.

Intermittent hypoxia facilitates expression of stem-like characteristics. (A, B) Real-time PCR analysis was performed in normoxic (N), and intermittent hypoxia (IH) conditioned neuroblastoma cells using primers specific to Oct-4 and CD133, and normalized to β-actin transcripts. **P<0.01, intermittent hypoxia versus normoxia. (C) Immunofluorescence analysis of CD133 expression. Cells were fixed and labeled with CD133 antibodies and Alexa-488 antimouse-conjugated antibodies. Photomicrographs were taken using Olympus fluorescence microscope. Nuclei were stained with DAPI (bar, 100 µm). (D) Flow cytometry. Cells were incubated with CD133/1-PE antibodies according to the manufacturer's instructions to determine the surface expression of CD133. After washing, flow cytometry was done using FACScan. IgG-PE antibody was used as a control. A representative flow cytometry analysis is shown. The graph represents the results of experiment done in triplicate.

Figure 4. Effects of intermittent hypoxia on neural crest /SNS markers.

Upregulation of markers for neural crest genes. (A) Western Blotting. Cell lysates of normoxic (N) and intermittent hypoxia (IH) conditioned neuroblastoma cells were analyzed by western blotting for the levels of c-Kit and TH. Real-time PCR. PCR analysis was performed in normoxic (N) and intermittent hypoxia (IH) conditioned neuroblastoma cells using primers specific to Notch-1 (B), ID2 (C) and HES-1(D) gene transcripts. **P<0.01, intermittent hypoxia versus normoxia. Downregulation of SNS markers. Real-time PCR. PCR analysis was performed in normoxic (N) and intermittent hypoxia (IH) conditioned neuroblastoma cells using primers specific to NPY (E), HASH-1(F) and dHAND (G). **P<0.01, intermittent hypoxia versus normoxia.

Figure 5. Effect of retinoic acid on human neuroblastoma cells.

(A) Normoxic (N) and intermittent hypoxia (IH) conditioned neuroblastoma cells were treated with 5 µm retinoic acid for 24 h and cell morphology was examined. Phase-contrast images were taken under bright field using an Olympus CKX41 inverted microscope (bar, 50 µm). (B) Graphic illustration of quantification of neurite lengths of normoxic and intermittent-hypoxia conditioned neuroblastoma cells treated with 5 µm retinoic acid. *P<0.05 **P<0.01, retinoic acid-treated versus untreated. (C) Immunofluorescence. Cells were fixed and incubated with primary antibodies for NF-M or HIF-1α. Then cells were washed in PBS and incubated with secondary antibodies, Alexa Fluor 488-conjugated anti-mouse IgG (HIF-1α) or Alexa Fluor 594-conjugated anti-rabbit IgG (NF-M) (bar, 100 µm). (D) Dual Immunofluorescence. Cells were treated with 10 µM retinoic acid for 24 h fixed and incubated with primary antibodies for NF-M or HIF-1α. Then cells were washed in PBS and incubated with secondary antibodies, Alexa Fluor 488-conjugated anti-mouse IgG or Alexa Fluor 594-conjugated anti-rabbit IgG. Nuclei were stained with DAPI. Photomicrographs were taken using Olympus fluorescence microscope (bar, 100 µm). (E) Western blotting: Cells were treated with 10 µM retinoic acid for 24 h. Cell lysates were analyzed for the levels of HIF-1α, NF-M and Neu N proteins by western blotting. β-actin served as loading control.

Figure 6. Effect of HIF-1α siRNA on differentiation of human neuroblastoma cells.

(A) Normoxic (N) and intermittent hypoxia (IH) conditioned neuroblastoma cells were treated with non-targeted control (NTC) or HIF-1α siRNA smart pool for 36 h and phase-contrast images were taken under bright field using an Olympus CKX41 inverted microscope (bar, 50 µm). (B) Graph illustrates quantification of neurite lengths of cells treated with NTC or HIF-1α siRNA. **p<0.01, intermittent hypoxia conditioned cells treated with HIF-1α siRNA versus NTC or untreated. (C) Western blotting. Cells were treated with NTC or HIF-1α siRNA smart pool. After 36 h, cells were lysed and cell extracts were subjected to western blotting analysis for HIF-1α, NF-M and Neu N.