Induction of anti-glioma natural killer cell response following multiple low-dose intracerebral CpG therapy

Abstract

Purpose: Stimulation of toll-like receptor-9 by CpG oligodeoxynucleotides (CpG-ODN) has been shown to counteract the immunosuppressive microenvironment and to inhibit tumor growth in glioma models. These studies, however, have used high doses of CpG-ODN, which can induce toxicity in a clinical setting. The goal of this study was to evaluate the antitumor efficacy of multiple low-dose intratumoral CpG-ODN in a glioma model.

Experimental design: Mice bearing 4-day-old intracranial GL261 gliomas received a single or multiple (two or four) intratumoral injections of CpG-ODN (3 microg) every 4 days. Tumor growth was measured by bioluminescent imaging, brain histology, and animal survival. Flow cytometry and cytotoxicity assays were used to assess anti-glioma immune response.

Results: Two and four intracranial injections of low-dose CpG-ODN, but not a single injection, eradicated gliomas in 70% of mice. Moreover, surviving animals exhibited durable tumor-free remission (> 3 months) and were protected from intracranial rechallenge with GL261 gliomas, showing the capacity for long-term antitumor immunity. Although most inflammatory cells seemed to increase, activated natural killer (NK) cells (i.e., NK(+)CD107a(+)) were more frequent than CD8(+)CD107a(+) in the brains of rechallenged CpG-ODN-treated animals and showed a stronger in vitro cytotoxicity against GL261 target cells. Leukocyte depletion studies confirmed that NK cells played an important role in the initial CpG-ODN antitumor response, but both CD8 and NK cells were equally important in long-term immunity against gliomas.

Conclusions: These findings suggest that multiple low-dose intratumoral injections of CpG-ODN can eradicate intracranial gliomas possibly through mechanisms involving NK-mediated effector function.

Figure 1

Multiple low-dose CpG oligonucleotide (CpG-ODN) injections improve survival of glioma-bearing mice. Mice bearing four-day old intracranial GL261-ffLuc tumors were treated with single (x1), two (x2) or four (x4) intratumoral injections of low-dose CpG-ODN (3μg/injection), control ODN (3μg/injection), or phosphate buffered saline (PBS) every four days (arrows). (A) Intracranial tumor burden was assessed by biophotonic imaging of mice at day 7, 14 and 21-post tumor injection. (B) Kaplan-Meier analysis demonstrates improved survival for mice that received four injections of CpG-ODN but not control ODN. (C) Survival was also improved in mice receiving two injections of CpG-ODN but not control ODN. n=7 mice/group, P values determined by Student’s T test. Data is representative of two separate experiments.

Figure 2

Effect of CpG-ODN on CD8 and NK cells. Mice bearing four day-old intracranial GL261 tumors were treated with intratumoral CpG-ODN (3μg/injection), control ODN (3μg/injection), and PBS once (one inj.) or four (four inj.) times as described in Fig. 1. Twenty-four hr after the last injection, brain, spleen, and blood were harvested and analyzed by flow cytometry for the percentage of CD8 (top panel) and NK (bottom panel) cells. P values for significant differences between treatment groups are provided above each bar graph. Significant differences between one and four injections for each treatment is represented as *: p<0.05, **: p<0.001, ***: p<0.0001. n=4 mice for each treatment group. Data is representative of two separate experiments.

Figure 3

Effect of CpG-ODN on regulatory T cells (Treg) and myeloid-derived suppressor cells (MDSC, Gr-1). Mice bearing four day-old intracranial GL261 tumors were treated with CpG-ODN (3μg/injection), control ODN (3μg/injection), and PBS once (one inj.) or four (four inj.) times as described in Fig. 1. Twenty-four hr after the last injection, brain, spleen, and blood were harvested and analyzed by flow cytometry for Treg and MDSCs by flow cytometry. P values for significant differences between treatment groups are provided above each bar graph. Significant differences between one and four injections for each treatment is represented as *: p<0.05, **: p<0.001, ***: p<0.0001. n=4 mice for each treatment group. Data is representative of two separate experiments.

Figure 4

Effect of tumor rechallenge on CD8 and NK cell activation. Normal mice and CpG-ODN-treated mice (4 injections) that had survived for at least three months after initial tumor inoculation were rechallenged with an intracranial injection of GL261 cells. Brain and blood samples from normal (naïve) (N), GL261-challenged naïve (T), CpG-ODN-treated survived (CS), and GL261-rechallenged CpG-ODN-treated survived mice (CR) were harvested and compared by flow cytometry for (A) IFNγ production, (B) activation of CD8 and NK cells (CD107a expression), and (C) CD8 and NK cytotoxicity activity against GL261 target cells. ns: not significant, *: p<0.05; **: p<0.001. n=4 mice for each treatment group. Data is representative of two separate experiments.

Figure 5

Anti-glioma effect of low-dose CpG-ODN is primarily mediated by NK cells. naïve mice were depleted of CD8 or NK cells by i.p. injection of relevant mAb or control IgG (200 μg/injection) one day prior to tumor implantation and each CpG-ODN treatment (3μg/injection, arrows). (A) Intracranial tumor burden was assessed by biophotonic imaging of mice at day 14 and 21 post tumor implantation. NK-depleted mice exhibited more rapid tumor growth. (B) Kaplan-Meier analysis demonstrates lower survival rate for mice that were depleted of NK cells (p=0.348). n=6 mice for each group. Data is representative of two separate experiments.

Figure 6

CpG-ODN-treated mice develop immunity against tumor rechallenge. Normal naïve mice and CpG-ODN-treated GL261-bearing mice that had survived for at least three months after the initial tumor inoculation were rechallenged with an intracranial injection of (A) GL261 glioma (1x10⁵) or (B) B16 F10 melanoma (1x10⁵) cells. Tumor growth was assessed by measuring animal survival (A and B) while NK cytotoxicity assay was measured by incubating isolated brain and blood NK cells with GL261 target cells. *: p<0.05, **: p<0.001, ***: p<0.0001. n=7 mice/group for survival experiments and 4 for cytotoxicity assays. Data is representative of two separate experiments.