Supplementary Figure 3: G-CSF inhibition reduces primary tumour growth and lung metastasis, an effect that is partially overcome by exogenous MDSCs.

(a) P53N tumour bearing mice were treated with G-CSF neutralizing antibody. Relative G-CSF levels in blood after treatment were measured by ELISA. n = 5 and 3 animals in IgG and anti-G-CSF groups, respectively. Two experiments were performed with consistent results, one representative experiment is shown. (b) 4T1 tumour cells were lentivirus-transduced with G-CSF shRNA (Control: GIPZ vector transduced). Efficiency in reducing G-CSF levels in vivo (left panel) and in vitro (right panel) was evaluated by G-CSF ELISA. Left: n = 10 and 17 animals for Control and shG-CSF groups, respectively. Right: cells cultured in 3 different wells were used as technical replicates. Two independent experiments were conducted with representative results shown. (c) MDSC numbers in blood of P53N-C tumour bearing mice after G-CSF-neutralizing antibody regimen compared to IgG control antibody-treated tumour-bearing mice. n = 3, 4, and 7 animals for the three groups, respectively P value is determined by non-parametric Wilcoxon test. (d) 4T1 cells transduced with vector control lentivirus (GIPZ, control) or with G-CSF shRNA expressing lentivirus (shG-CSF) were orthotopically injected to mammary gland. MDSC numbers in blood were quantified. n = 9 and 7 animals for the two groups, respectively. (e) No direct role of G-CSF in 4T1 tumour cell survival. Cell viability was assessed by WST-1 assay in presence of G-CSF neutralizing antibody or G-CSF shRNA compared to controls ‘n.s.’: no significant difference. One experiment was performed in which cells of different groups were cultured in 5 wells each. Error bars indicate s.e.m., and P values are calculated by two-tailed Student’s t tests unless otherwise noted. Statistics source data of Supplementary Fig. 3b–d are provided in Supplementary Table 4.