Stable luciferase expression does not alter immunologic or in vivo growth properties of GL261 murine glioma cells

Abstract

Background: GL261 cells are murine glioma cells that demonstrate proliferation, invasion, and angiogenesis when implanted in syngeneic C57BL/6 mice, providing a highly useful immunocompetent animal model of glioblastoma. Modification of tumor cells for luciferase expression enables non-invasive monitoring of orthotopic tumor growth, and has proven useful for studying glioblastoma response to novel therapeutics. However, tumor modification for luciferase has the potential for evoking host immune response against otherwise syngeneic tumor cells, thereby mitigating the tumor cells' value for tumor immunology and immunotherapy studies.

Methods: GL261 cells were infected with lentivirus containing a gene encoding firefly luciferase (GL261.luc). In vitro proliferation of parental (unmodified) GL261 and GL261.luc was measured on days 0, 1, 2, 4, and 7 following plating, and the expression of 82 mouse cytokines and chemokines were analyzed by RT-PCR array. Cell lines were also evaluated for differences in invasion and migration in modified Boyden chambers. GL261 and GL261.luc cells were then implanted intracranially in C57BL/6 mice, with GL261.luc tumor growth monitored by quantitative bioluminescence imaging, and all mice were followed for survival to compare relative malignancy of tumor cells.

Results: No difference in proliferation was indicated for GL261 vs. GL261.luc cells (p>0.05). Of the 82 genes examined by RT-PCR array, seven (9%) exhibited statistically significant change after luciferase modification. Of these, only three changed by greater than 2-fold: BMP-2, IL-13, and TGF-β2. No difference in invasion (p=0.67) or migration (p=0.26) was evident between modified vs. unmodified cells. GL261.luc cell luminescence was detectable in the brains of C57BL/6 mice at day 5 post-implantation, and tumor bioluminescence increased exponentially to day 19. Median overall survival was 20.2 days versus 19.7 days for mice receiving implantation with GL261 and GL261.luc, respectively (p=0.62). Histopathologic analysis revealed no morphological difference between tumors, and immunohistochemical analysis showed no significant difference for staining of CD3, Ki67, or CD31 (p>0.05 for all).

Conclusions: Luciferase expression in GL261 murine glioma cells does not affect GL261 proliferation, invasion, cytokine expression, or in vivo growth. Luciferase modification increases their utility for studying tumor immunology and immunotherapeutic approaches for treating glioblastoma.

Figure 1

Luciferase activity in GL261.luc cells. U87MG cells, previously modified with the same lentivirus used here for GL261 modification [14], were used as a control for assessing luciferase activity in transduced GL261 cells. U87MG.luc cells were plated at densities of 1x10⁶, 0.5x10⁶, 0.25x10⁶, 0.1x10⁶, 0.05x10⁶, and 0.025x10⁶ cells/well, from left to right. GL261 (negative control) and GL261.luc cells were plated at densities of 1x10⁶, 0.5x10⁶, and 0.25x10⁶ cells/well.

Figure 2

Luciferase expression does not affect proliferation Luciferase expression by GL261.luc cells does not cause a difference in proliferation, in vitro , as demonstrated by ATP-based viability assay. The graph shows fold increase, relative to day 0, as determined by comparing the luminescence at the specified time point to the luminescence at time 0.

Figure 3

Differences in gene expression between GL261 and GL261.luc. Volcano plot (A) demonstrating the differences in gene expression, for 82 cytokine and chemokine genes, between GL261.luc and unmodified GL261 cells. The horizontal grey line indicates the level at which a significance difference (P = 0.05) exists. Points above that line represent genes whose expression is significantly different between the two cells. The vertical grey lines indicate a fold change of 2: either increased or decreased by a factor of 2 in GL261.luc cells. Points outside of the vertical lines represent genes which are more than 2-fold changed. A bar graph (B) depicts the extent of change for the seven genes whose expression is significantly different between GL261.luc and GL261.

Figure 4

Luciferase expression does not alter invasion or migration. The bar graphs showing no difference in in vitro invasion (A) and migration (B) between GL261.luc and GL261 cells. Error bars indicate standard error between triplicates.

Figure 5

Luciferase expression does not affect overall survival. Kaplan-Meier survival analysis (A) demonstrates no difference in overall survival for mice implanted intracranially with either GL261.luc (dashed line) or GL261 cells (solid line). One animal in the GL261.luc group was censored due to procedural-related death. In vivo quantitative bioluminescence imaging (B) demonstrates non-invasive monitoring of intracranial tumor growth in animals implanted with GL261.luc cells.

Figure 6

Histopatholgic and immunohistochemical analysis of tumor specimens from GL261 and GL261.luc bearing animals. Hematoxylin and Eosin (H&E) staining was used for conventional morphologic analysis of tumor. Ki-67 staining was for examining tumor cell proliferation, whereas C3 and CD31 staining was for T-cell infiltration and tumor neovascularization, respectively. Magnification is 200 X.

Figure 7

In vitro luciferase expression of GL261.luc over 21 days. GL261.luc cells were maintained in culture over 21 days. At day 0, 7, 14, and 21, cells were seeded in 6-well plates at a defined density and were evaluated for bioluminescence. GL261 and U87.luc cells were used as negative and positive controls, respectively.