The role of granulocyte macrophage colony stimulating factor (GM-CSF) in radiation-induced tumor cell migration

Abstract

Recently it has been observed in preclinical models that that radiation enhances the recruitment of circulating tumor cells to primary tumors, and results in tumor regrowth after treatment. This process may have implications for clinical radiotherapy, which improves control of a number of tumor types but which, despite continued dose escalation and aggressive fractionation, is unable to fully prevent local recurrences. By irradiating a single tumor within an animal bearing multiple lesions, we observed an increase in tumor cell migration to irradiated and unirradiated sites, suggesting a systemic component to this process. Previous work has identified the cytokine GM-CSF, produced by tumor cells following irradiation, as a key effector of this process. We evaluated the ability of systemic injections of a PEGylated form of GM-CSF to stimulate tumor cell migration. While increases in invasion and migration were observed for tumor cells in a transwell assay, we found that daily injections of PEG-GM-CSF to tumor-bearing animals did not increase migration of cells to tumors, despite the anticipated changes in circulating levels of granulocytes and monocytes produced by this treatment. Combination of PEG-GM-CSF treatment with radiation also did not increase tumor cell migration. These findings suggest that clinical use of GM-CSF to treat neutropenia in cancer patients will not have negative effects on the aggressiveness of residual cancer cells. However, further work is needed to characterize the mechanism by which GM-CSF facilitates systemic recruitment of trafficking tumor cells to tumors.

Keywords: Cancer; GM-CSF; Metastasis; Radiation therapy.

Figure 1

Radiation stimulates migration and invasion of tumor cells. (A) 4T1 cells cultured in vitro migrate preferentially toward supernatant (SN) from 4T1 cells spiked with recombinant GM-CSF. SN harvested from 4T1 cells irradiated to a dose of 20 Gy also stimulates migration, while inhibition of GM-CSF through shRNA or neutralizing antibodies blocks it. (B) Irradiation of 4T1 tumors in mice promotes recruitment of metastatic luciferase-labeled 4T1 cells. This process is inhibited when GM-CSF is genetically inhibited in the irradiated tumors. * denotes P < 0.05. Adapted from [2].

Figure 2

Radiation promotes recruitment of migrating tumor cells in a systemic manner. (A) A donor-recipient system was created in which two unlabeled 4T1 tumors were grown in contralateral mammary fat pads of nude mice. One was focally irradiated to a dose of 20 Gy, immediately after which luciferase-expressing 4T1 cells were injected intravenously through the tail vein. (B) Bioluminescence images of irradiated recipient tumors (Rad), as well as the contralateral tumors from the irradiated mice (Shield). (C) Ex vivo bioluminescence images of recipient tumors harvested from untreated mice 10 days post-treatment. (D) Quantification of the bioluminescence signals seen in (B) and (C). * denotes P < 0.05.

Figure 3

GM-CSF as well as PEG-GM-CSF induce invasion of 4T1 cells in vitro. Conditioned media from 4T1 cells was used as an attractant in a transwell invasion assay. Either GM-CSF or PEG-GM-CSF was added to this media, and significant increases were obsereved in invasion, as visualized by DAPI staining of the transwell membrane. * denotes P < 0.05, ** denotes P < 0.01.

Figure 4

Systemic administration of PEG-GM-CSF does not increase recruitment of migrating tumor cells to tumors. (A) Blood half life of GM-CSF after a single intravenous injection of 0.05 mg/kg of PEG-GM-CSF. (B) Ex vivo bioluminescence measured in shGFP or shGM-CSF tumors in mice treated with daily injections of 20 pg PEG-GM-CSF. (C) Plasma levels of GM-CSF measured 3 and 10 days after the initiation of treatment. (D) Counts for blood cells before (Control) and at specified times following the initiation of PEG-GM-CSF treatment.

Figure 5

Addition of PEG-GM-CSF to radiotherapy does not increase recruitment of migrating tumor cells. (A) The model system from Figure 2A was used, with the addition of daily intravenous injections of 20 pg PEG-GM-CSF beginning immediately after radiation treatment. (B) Ex vivo bioluminescence measured in tumors for the specified treatment groups. (C) Tumor volumes measured by calipers for treated tumors. Counts of white blood cells, red blood cells (D), neutrophils (E), monocytes (F), and eosinophils (G) measured 10 days after the initiation of treatment for the specified groups.