Knock down of HIF-1alpha in glioma cells reduces migration in vitro and invasion in vivo and impairs their ability to form tumor spheres

Abstract

Background: Glioblastoma (GBM) is the most common and malignant primary intracranial human neoplasm. GBMs are characterized by the presence of extensive areas of necrosis and hypoxia. Hypoxia and its master regulator, hypoxia inducible factor 1 (HIF-1) play a key role in glioma invasion.

Results: To further elucidate the functional role of HIF-1alpha in glioma cell migration in vitro and in invasion in vivo, we used a shRNA approach to knock down HIF-1alpha expression complemented with genome-wide expression profiling, performed in both normoxic and hypoxic conditions. Our data show that knock down of HIF-1alpha in glioma cells significantly impairs their migration in vitro as well as their ability to invade into the brain parenchyma in vivo. Next, we assessed the role that HIF-1alpha plays in maintaining the characteristics of cancer stem cells (CSCs). By using the tumor sphere forming assay, we demonstrate that HIF-1alpha plays a role in the survival and self-renewal potential of CSCs. Finally, expression profiling experiments in glioma cells provided detailed insight into a broad range of specific biological pathways and processes downstream of HIF-1alpha. We discuss the role of these processes in the migratory and invasive properties, as well as the stem cell biology of glioblastomas

Conclusions: Our data show that knock down of HIF-1alpha in human and murine glioma cells impairs their migration in vitro and their invasion in vivo. In addition, our data suggest that HIF-1alpha plays a role in the survival and self-renewal potential of CSCs and identify genes that might further elucidate the role of HIF-1alpha in tumor migration, invasion and stem cell biology.

Figure 1

Knock down of HIF-1α reduces hypoxia-induced HIF-1α expression and hypoxia-induced migration of glioma cells. (A) Western blot showing the expression level of HIF-1α under normoxic (Norm) and hypoxic (Hyp) conditions in human (LN308 and U87MG), and murine (GL261) glioma cell lines, after HIF-1α shRNA lentivirus-mediated infection (shHIF). β-Actin was used as loading control. A representative result of two independent experiments is shown. (B) Knock down of HIF-1α reduces hypoxia-induced migration of glioma cells. Boyden chamber assays showing migration of human (LN308 and U87MG) and murine (GL261) glioma cell lines infected with HIF-1 shRNA (shHIF) or the control cells. Two independent experiments, done in duplicate, were performed. Each graph represents the number of cells per field counted after 16 h of migration in either a normoxic (Norm) or a hypoxic (Hyp) environment (three asterisks, p < 0.0001).

Figure 2

Knock down of HIF-1α reduces invasion of murine glioma cells in vivo. Representative tumors derived from GFP-tagged GL261 shHIF-1 and control cells showing tumor sections stained for GFP, H&E and HIF-1α. (A) The bar graph represents the average depth of invasion (μm) at the edge of tumors derived from GL261 control and GL261 shHIF cells (N = 7/group) (asterisk denotes p = 0.03). (B) As shown by GFP staining and confirmed by H&E, control tumors derived from cells expressing HIF-1α displayed an invasive edge compared with tumors derived from GL261 shHIF-1 cells with marginal invasion. HIF-1α immunohistochemistry confirms lack of HIF-1 expression in HIF-1α knockdown tumors and its expression in the controls. Open arrowheads point to HIF-1α expressing cells at the infiltrative margin in the control tumors. Scale bar represents 50 μm.

Figure 3

Knock down of HIF-1α reduces the ability of murine glioma cells to form tumor spheres. Primary tumor sphere forming capacity of GL261 shHIF cells in comparison with control cells. Three independent experiments were performed. (A) Bar graph represents the % of the total number of tumor spheres in shHIF cells versus the control. Knockdown of HIF-1α significantly reduces the ability of glioma cells to form tumor spheres. The shHIF cells formed 60% fewer tumor spheres compared with control cells (asterisk denotes p = 0.02). (B) Bar graph represents the average diameter (μm) of tumor spheres in shHIF cells versus the control. Lack of HIF-1α significantly reduces the average diameter of the spheres formed (167.4 ± 4.1 μm) compared with the size of those derived from control cells (240.6 ± 2.8 μm) (three asterisks denotes p < 0.0001). (C) Representative images of tumor spheres derived from either control or shHIF cells.

Figure 4

Profile of genes that are upregulated in response to hypoxia in GL261 control cells and show no change or downregulation in GL261 shHIF cells, using as normalization baseline the GL261 control cells in normoxia. For each cell line (control and shHIF) and condition: Norm (Normoxia) and Hyp (Hypoxia) we show the results from two independent experiments used in the analysis (R1 and R2). (A) Box plot analysis shows a five-number summary (the smallest observation, lower quartile, median, upper quartile, and largest observation) for the 149 significantly modulated Affymetrix Probe sets, found by Pavlidis Template Matching (p < 0.05), based on two individual experiments (R1 and R2). (B) Heat map showing the 149 Affymetrix probe sets including known HIF-1α targets that are upregulated in response to hypoxia in GL261 control cells and show no change or downregulation in GL261 shHIF cells. Red and blue colors denote increased and decreased mRNA abundance, respectively.