Figure 1.
Isolation of gefitinib-resistant persisters (GRPs) derived from NSCLC cell lines harboring an activating EGFR mutation.
First, 2×105 PC9 cells were plated in 10-cm plates and allowed to adhere for 24 h (A). Cells were then treated with 1 µM gefitinib. Fresh media-containing gefitinib was replaced every 3 days. Seven days after gefitinib exposure, a small fraction of viable cells remained (B), resumed proliferating 18 days later (C), and continued to proliferate even 30 days later (D). A. Representative microscopic image of parental PC9 cells. B, C, D. Representative images of PC9 GRPs exposed to 1 µM gefitinib for 7, 18, and 30 days were photographed, respectively.
Figure 2.
GRPs were highly enriched for gene expression of stemness, IGF1, and IGF1R.
A. Quantitative RT-PCR was performed with primers specific for CD133, Oct4, Sox2, Nanog, CXCR4, and ALDH1A1, which are stemness genes in PC9 or HCC827 parental cells and GRPs. B. Quantitative RT-PCR was performed with primers specific for IGF1 and IGF1R in PC9 or HCC827 parental cells and GRPs. Data were normalized to actin expression. *p<0.01, **p<0.001, ***p<0.0001.
Figure 3.
Sphere-forming ability of GRPs was upregulated and hypoxia increased this capacity.
PC9 parental cells, GRPs, and hypoxic GRPs were prepared at densities of 2.5×103 cells per 2 mL per well in serum-free media supplemented with growth factors and seeded into 6-well ultra-low attachment plates. PC9 parental cells and GRPs were incubated under normoxic conditions, while PC9 hypoxic GRPs were grown under hypoxic conditions. Culture medium was fed every 3 days. The number and size of spheres were recorded and immunofluorescence was performed 7 days after the start of the culture period. Spheres were fixed and incubated with primary antibodies against CD133, Oct4, or phosphorylated IGF1R (pIGF1R), and then with secondary antibody labeled with Alexa Fluor 594 goat anti-mouse IgG (red) or Alexa Fluor 488 goat anti-rabbit IgG (green). Cell nuclei were stained with DAPI (blue). Images were obtained on an Axioplan 2 imaging system with AxioVision software. A. The number of spheres was significantly increased in PC9 GRPs compared to in parental cells, and was further increased in PC9 hypoxic GRPs. *p<0.01, # p<0.05. B. Sphere size of PC9 hypoxic GRPs was significantly greater than that of parental cells. # p<0.05. C. Immunofluorescent images of control cells of PC9 (left) and spheres of GRPs (right) for CD133, Oct4, and pIGF1R.
Table 1.
Tumor incidence of parental PC9 cells and normoxic and hypoxic GRPs transplanted into NOG mice.
Figure 4.
Population of CD133- and Oct4-positive GRPs were increased under hypoxic conditions.
A, B. PC9 or HCC827 cells, growing on Lab-Tek chamber slides with or without 1 or 2 µM gefitinib and under normoxic or hypoxic conditions for 72 h, fixed, and incubated with the primary antibodies against CD133 (A) and Oct4 (B), and then with secondary antibody labeled with Alexa Fluor 594 goat anti-mouse IgG (red). Cell nuclei were stained with DAPI (blue). Images were obtained on an Axioplan 2 imaging system with AxioVision software. Images used to compare parental cells, GRPs, and hypoxic GRPs were acquired using the same instrument settings and exposure times and were processed similarly. The numbers of CD133 and Oct4-positive cells were counted, and the ratio of these cells was calculated in five fields in each experiment. **p<0.001, *p<0.01, # p<0.05.
Figure 5.
IGF1R was phosphorylated on hypoxic GRPs, and knockdown of IGF1 decreased the population of CD133- and Oct4-positive hypoxic GRPs.
A. Quantitative RT-PCR was performed with primers specific for IGF1 in PC9 or HCC827 parental cells, GRPs, and hypoxic GRPs. Data were normalized to actin expression. **p<0.001. B. PC9 cells, grown on Lab-Tek chamber slides with or without 1 µM gefitinib and under normoxic or hypoxic conditions for 72 h, fixed, and incubated with the primary antibodies against phosphorylated IGF1R (pIGF1R) and then with secondary antibodies labeled with Alexa Fluor 488 goat anti-rabbit IgG (green). Cell nuclei were stained with DAPI (blue). Images were obtained using an Axioplan 2 system imaging with AxioVision software. Images used to compare PC9 parental cells, GRPs, and hypoxic GRPs were acquired with the same instrument settings and exposure times, and were processed similarly. The number of pIGF1R-positive cells was counted, and the ratio of these cells was calculated in five fields in each experiment. **p<0.001. C. IGF1 expression was knocked down in PC9 or HCC827 hypoxic GRPs by using small interfering RNA (siRNA) in Lab-Tek chamber slides. Immunofluorescence staining for pIGF1R, CD133, or Oct4 was then performed. Two specific siRNAs and one non-specific control were used. The numbers of pIGF1R-, CD133- and Oct4-positive cells were counted, and the ratio of these cells was calculated in five fields for each experiment. **p<0.001, *p<0.01.
Figure 6.
Hypoxia regulates IGF1 expression through HIF1α, and the inhibition of HIF1α or IGF1R decreased CD133- and Oct4-positive GRPs under hypoxia.
A. HIF1α expression was suppressed in PC9 or HCC827 hypoxic GRPs by treatment with 50 µM, 100 µM, and 200 µM YC-1 in Lab-Tek chamber slides. Immunofluorescence staining for IGF1, phosphorylated IGF1R (pIGF1R), CD133, and Oct4 was then performed. The numbers of IGF1-, pIGF1R-, CD133-, and Oct4-positive cells were counted, and the ratio of these cells was calculated in five fields from each experiment. **p<0.001. B. PC9 or HCC827 cells were incubated with 1 or 2 µM gefitinib in the presence or absence of 0.01 µM, 0.1 µM, or 1 µM of the IGF1R inhibitor AEW541 under hypoxic conditions for 72 h in Lab-Tek chamber slides. Immunofluorescence staining for CD133 and Oct4 was then performed. The numbers of CD133- and Oct4-positive cells were counted, and the ratio was calculated in five fields for each experiment. **p<0.001. C. PC9 or HCC827 cells were plated in 10-cm plates and allowed to adhere for 24 h. Cells were then incubated with 1 or 2 µM gefitinib in the presence or absence of 0.01 µM, 0.1 µM, or 1 µM of AEW541 under hypoxic conditions for 18 or 11 days. The numbers of colonies were then counted.