Chemokine receptor CXCR3 promotes growth of glioma

Abstract

Human glioblastoma multiforme (GBM) is the most common primary brain tumor in adults. The poor prognosis and minimally successful treatments of GBM indicates a need to identify new therapeutic targets. In this study, we examined the role of CXCR3 in glioma progression using the GL261 murine model of malignant glioma. Intracranial GL261 tumors express CXCL9 and CXCL10 in vivo. Glioma-bearing CXCR3-deficient mice had significantly shorter median survival time and reduced numbers of tumor-infiltrated natural killer and natural killer T cells as compared with tumor-bearing wild-type (WT) mice. In contrast, pharmacological antagonism of CXCR3 with NBI-74330 prolonged median survival times of both tumor-bearing WT and CXCR3-deficient mice when compared with vehicle-treated groups. NBI-74330 treatment did not impact tumor infiltration of lymphocytes and microglia. A small percentage of GL261 cells were identified as CXCR3(+), which was similar to the expression of CXCR3 in several grade IV human glioma cell lines (A172, T98G, U87, U118 and U138). When cultured as gliomaspheres (GS), the human and murine lines increased CXCR3 expression; CXCR3 expression was also found in a primary human GBM-derived GS. Additionally, CXCR3 isoform A was expressed by all lines, whereas CXCR3-B was detected in T98G-, U118- and U138-GS cells. CXCL9 or CXCL10 induced in vitro glioma cell growth in GL261- and U87-GS as well as inhibited cell loss in U138-GS cells and this effect was antagonized by NBI-74330. The results suggest that CXCR3 antagonism exerts a direct anti-glioma effect and this receptor may be a potential therapeutic target for treating human GBM.

Fig. 1.

CXCL9 and CXCL10 are expressed in GL261 glioma cells and/or tumors. (A) RT–PCR identified CXCL10 mRNA in GL261 cells in vitro. GAPDH was used as a control. (B) CXCL10 ELISA showing CXCL10 protein secretion by GL261 cells in vitro at 24 and 48 h (C) CXCL9 and CXCL10 are expressed in intracranial GL261 tumors in vivo as determined by in situ hybridization analysis. Two representative sections, depicting expression of each chemokine, are shown. The colour version of this figure can be found at www.carcin.oxfordjournals.org.

Fig. 2.

GL261 tumor-bearing CXCR3-deficient mice have decreased survival rates and tumor-infiltrated NK and NKT cells. (A) Kaplan–Meier survival analysis indicated that CXCR3-deficient mice (n = 10) have shorter life span than WT mice (n = 10, P < 0.0001). Filled squares: CXCR3-deficient mice; filled diamonds: WT mice. (B) Numbers of tumor-infiltrated CD4⁺, CD8⁺, Foxp3⁺, Ly49G2⁺ and CD11b⁺ cells were evaluated by immunohistochemistry. Gliomas from CXCR3-deficient mice had a significant reduction of Ly49G2⁺ (NK and NKT) cells in the tumor as compared with WT mice (**P < 0.01). WT: wild-type; KO: CXCR3-deficient mice. (C) Intratumoral expression of CXCL9 and CXCL10 mRNA was not altered by host CXCR3 deficiency. Shown are representative sections from WT and CXCR3-deficient glioma-bearing mice subjected to in situ hybridization analysis. The colour version of this figure can be found at www.carcin.oxfordjournals.org.

Fig. 3.

NBI-74330 suppresses tumor growth in both WT and CXCR3-deficient mice. (A) Kaplan–Meier survival analysis of glioma-bearing WT mice shows that NBI-74330 prolonged animal survival (n = 8), as compared with vehicle-treated mice (n = 8) (P = 0.0212). Filled squares: NBI-74330 treated; filled diamonds: vehicle treated. (B) Kaplan–Meier survival analysis shows that glioma-bearing CXCR3-deficient mice treated with NBI-74330 (n = 7) had a higher survival rate than vehicle-treated glioma-bearing CXCR3-deficient mice (n = 6, P = 0.0028). Filled squares: NBI-74330 treated; filled diamonds: vehicle treated. (C) Similar numbers of tumor-infiltrated lymphocytes and microglia in GL261 gliomas from NBI-74330- and vehicle-treated WT mice. Numbers of tumor-infiltrated CD4⁺, CD8⁺, Foxp3⁺, Ly49G2⁺ and CD11b⁺ cells were not affected by NBI-74330 when compared with vehicle treatment. (D) In vivo expression of CXCL9 and CXCL10 was not altered by NBI-74330 treatment. Shown are representative sections from vehicle- and NBI-74330-treated glioma-bearing mice subjected to in situ hybridization analysis. The colour version of this figure can be found at www.carcin.oxfordjournals.org.

Fig. 4.

CXCR3 and CXCL10 expression in murine and human glioma cell lines cultured in serum-containing media. (A) RT–PCR identified CXCL10 mRNA in T98G and U87 cells in vitro. GAPDH was used as a control. (B) Representative histograms from fluorescence-activated cell sorting analysis showing CXCR3 expression by GL261, A172, T98G, U87, U118 and U138 cells in vitro. Gray filled area, anti-CXCR3-APC (mouse) or anti-CXCR3-PE (human) staining; blank area: isotype controls. GL261 has the highest CXCR3 expression level among all the glioma cell lines.

Fig. 5.

CXCL10, CXCL11 and CXCR3 expression in GS cells. (A) RT–PCR identified in vitro expression of CXCL10 mRNA in GL261-, T98G-, U87- and U118-GS cells and a patient GBM tissue-derived primary GS (GBM L0). In addition, CXCL11 is expressed by GL261-, T98G-, U87-, U118-GS and GBM L0 cells. GAPDH was used as a control. (B) Representative histograms from fluorescence-activated cell sorting analysis showing CXCR3 expression by GL261-, A172-, T98G-, U87-, U118-, U138-GS and GBM L0 cells. Gray filled area: anti-CXCR3-APC (mouse) or anti-CXCR3-PE (human) staining. Blank area: isotype controls. (C) RT–PCR analysis identified in vitro expression of CXCR3 mRNA isoforms in murine and human GS cells. CXCR3-A was detected in all cells with A172-, T98G- and U87-GS cells expressing the highest levels. CXCR3-B was detected in T98G-, U118- and U138-GS cells. One form of CXCR3 mRNA was detected in GL261-GS cells; only the equivalent of CXCR3-A exists in mouse.

Fig. 6.

NBI-74330 inhibits CXCL9 and CXCL10 stimulation of GS growth or survival in vitro. (A) GL261- and U87-GS cells (2000 cells/ml) were incubated with 1 nM CXCL9 (open squares) or 1 nM CXCL10 (open triangles); U118-, U138-GS cells (5000 cells/ml) and GBM L0 (1000 cells/ml) were incubated with 10 nM CXCL9 (open squares) or 10 nM CXCL10 (open triangles). The control group was cultured in medium without chemokines or growth factors (filled circles). All conditions contained 0.1% dimethyl sulfoxide (NBI-74330 vehicle). Representative results of three individual experiments performed in triplicate are shown. CXCL9 and CXCL10 significantly enhanced GL261- and U87-GS growth at day 6 and 9 (*P < 0.05, **P < 0.01) and prevented U138-GS cell loss at day 9. CXCL9 stimulation significantly increased cell numbers of GBM L0 at day 9 (#P < 0.05); CXCL10-stimulated group was not statistically significant as compared with control. (B) The effect of 1 μM NBI-74330 (black filled bars) and 0.1% dimethyl sulfoxide (open bars) on chemokine- and growth factor-stimulated GS growth. Representative results of three individual experiments performed in triplicate are shown. NBI-74330 attenuated response of GL261-, U87- and U138-GS to CXCL9 and CXCL10 but did not affect either control or growth factor-supplemented groups.