XRN2 Is Required for Cell Motility and Invasion in Glioblastomas

Abstract

One of the major obstacles in treating brain cancers, particularly glioblastoma multiforme, is the occurrence of secondary tumor lesions that arise in areas of the brain and are inoperable while obtaining resistance to current therapeutic agents. Thus, gaining a better understanding of the cellular factors that regulate glioblastoma multiforme cellular movement is imperative. In our study, we demonstrate that the 5'-3' exoribonuclease XRN2 is important to the invasive nature of glioblastoma. A loss of XRN2 decreases cellular speed, displacement, and movement through a matrix of established glioblastoma multiforme cell lines. Additionally, a loss of XRN2 abolishes tumor formation in orthotopic mouse xenograft implanted with G55 glioblastoma multiforme cells. One reason for these observations is that loss of XRN2 disrupts the expression profile of several cellular factors that are important for tumor invasion in glioblastoma multiforme cells. Importantly, XRN2 mRNA and protein levels are elevated in glioblastoma multiforme patient samples. Elevation in XRN2 mRNA also correlates with poor overall patient survival. These data demonstrate that XRN2 is an important cellular factor regulating one of the major obstacles in treating glioblastomas and is a potential molecular target that can greatly enhance patient survival.

Keywords: XRN2; cell motility; invasion; tumor progression.

Figure 1

XRN2 expression confers with glioblastoma disease and poor patient survival. (A) Log2 XRN2 mRNA expression levels in normal (Berchtold, 172 samples), GBM-1 (Pfister, 46 samples), GBM-2 (Loeffler, 70 samples), and GBM-3 (Hegi, 84 samples) databases. Graph generated from R2: Genomics Analysis and Visualization Platform. (B) Representative XRN2 immunohistochemistry staining of normal and glioblastoma patient samples. Brown signal is XRN2. Blue signal is hematoxylin (nuclei). Scale bar is 100 micron. (C) Quantification of XRN2 signal from immunohistochemistry staining of normal and GBM patient samples. A signal of less than 2% is considered negative, while a signal of 2% or greater is considered positive. (D) Glioma patient survival outlook based on XRN2 mRNA levels (Kawaguchi, 50 patients). Blue line denotes patients with high expression of XRN2. Red line denotes patients with low expression of XRN2. Graph generated from R2: Genomics Analysis and Visualization Platform. Statistical analysis was performed using the default setting provided by platform.

Figure 2

XRN2 is required for GBM cell motility. (A) Image stills from live-cell imaging of U87-GFP cells. Arrows track the movement of two representative cells over time. Tracking was over 6 h with images taken at 30-min intervals. (B) Positional tracks of U87 H2B-GFP cells during the 6 h live-cell imaging. White marks the position of the cells at the beginning of time to red, the position of the cells at the end of imaging. (C) Quantification of U87-GFP tracking. Changes in speed and displacement upon XRN2 knockdown by siRNA are shown. Scale bar is 100 micron. ** p-value ≤ 0.01, **** p-value ≤ 0.0001. The Students t-test was used for statistical analysis.

Figure 3

XRN2 is required for invasion through a matrix. (A) Diagram of inverted vertical invasion. Day 1—cells are plated at near confluent levels. Day 2—extracellular matrix is applied to the confluent cells. Day 4—cultures are fixed, stained, and imaged for invasion analysis. (B) Representative ZX image of invading G55 cells (white). Red arrows denote invading cells. (C) Quantification of up-invasion of G55 cells with the listed shRNAs. * = p-value ≤ 0.05, ** = p-value ≤ 0.01. The Students t-test was used for statistical analysis.

Figure 4

Loss of XRN2 results in decreased tumor volume in vivo. (A) Magnetic resonance imaging was used to detect control (G55-shluc) and XRN2-deficient (G55-shXRN2) tumors in mouse brains. (B) Quantitation of tumor volumes obtained from control (G55-shluc) and two unique XRN2-deficient (G55-shXRN2 and G55-shXRN2 #2) G55 cell lines. * = p-value ≤ 0.05, ** = p-value ≤ 0.01. The Students t-test was used for statistical analysis.

Figure 5

XRN2-mediated transcriptome landscape. (A) qPCR of XRN2 expression of samples used in the RNA-Seq in B. (B) RNA-Seq heat map of LN229 and U251 transfected with siCont or siXRN2. Heat map generated from transcripts with a log 1.3 or greater change and a p-value of 0.05 or better. (C) Venn diagram of overlapping transcripts from B. (D) Ingenuity pathway analysis of the overlapping transcripts in C.

Figure 6

Loss of VCL or XRN2 results in a similar decrease in the speed or displacement in GBM cells. Quantification of U251-GFP tracking. Changes in (A) speed and (B) displacement upon control, VCL or XRN2 knockdown by siRNA are shown. ** p-value ≤ 0.01, *** p-value ≤ 0.001. The Students t-test was used for statistical analysis.