Tables and Figure Legends for Supplementary Information
Figure S1. Dose response curves for growth inhibition of control and TKI-resistant cultures of EGFR mutant NSCLC cell lines. The indicated control and resistant NSCLC cell lines were submitted to growth assays in the presence of increasing concentrations of BibW2992, gefitinib or AZ8010 using 96-well plates and the CyQUANT assay as described in the Materials and Methods. The individual values are the mean and SEM of three replicate determinations.
Figure S2. Acquisition of a EGFR-T790M mutation in gefitinib-resistant PC9 cells and retained sensitivity to BIBW2992.A) Total RNA was collected from control and gefitinib adapted PC9 cells, reverse transcribed to cDNA and sequenced with PCR primers flanking exon 20 in the tyrosine kinase domain of the EGFR (FP, 5’-TCCTCTTGCTGCTGGTGGTG-3’; RP, 5’-GCGTTCTCCTTTCTCCAGGATG-3’). The appearance of the C to A mutation encoding 790M is indicated by the arrow. B) Cell lysates from control and gefitinib-resistant PC9 cells treated for 2 hrs with the indicated TKIs were submitted to immunoblot analysis for phospho-ERK and total ERK as a loading control. The findings demonstrate retention of BIBW2992-inhibited phospho-ERK in the gefitinib-resistant PC9 cells consistent with the ability of BIBW2992 to block activity of EGFR-T790M mutants. C) The control and gefitinib-resistant PC9 cultures were submitted to a clonogenic growth assay in the presence of the indicated TKIs. The data demonstrate retained growth inhibition by BIBW2992 in gefitinib-resistant PC9 cells.
Figure S3. Increased MET expression in gefitinib-resistant HCC827 cells and acquired sensitivity to crizotinib. A) Control and gefitinib-resistant HCC827 cells were incubated for 2 hours with the indicated TKIs and then immunoblotted for phospho-MET (Y1234/1235), total MET and the a-subunit of NaK-ATPase as a loading control. The findings show increased expression of total MET and phospho-MET in the gefitinib-resistant cells and inhibition of phospho-MET by crizotinib. B) The control and gefitinib-resistant HCC827 cells were submitted to an anchorage-independent growth assay in the presence of the indicated TKIs. The findings demonstrate loss of sensitivity to the ERBB family inhibitor, lapatinib, and gain of sensitivity to the MET inhibitor, crizotinib.
Figure S4. Quantification of immunoblot results of phospho-AKT-T308 in control and gefitinib-resistant HCC4006 and H1650 cells following treatment with TKIs. The immunoblot from Figure 2 was submitted to densitometry and the values for phospho-AKT-T308 were normalized to the total AKT levels and graphed as shown. The findings demonstrate decreased phospho-AKT-T308 levels following EGFR inhibition in control, but not gefitinib-resistant, HCC4006 and H1650 cells. The FGFR-specific inhibitor, AZ8010 did not inhibit AKT phosphorylation in either control or gefitinib-resistant cell lines.
Figure S5. Inhibition of phospho-EGFR by lapatinib and BIBW2992 in control, but not gefitinib-resistant H1650 cells. The indicated cell cultures were treated for 2 hrs with the TKIs and cell extracts were immunoblotted for phospho-EGFR-Y1068 and total EGFR. The findings demonstrate inhibition of phospho-EGFR in control H1650 cells by lapatinib and BIBW2992 and loss of phospho-EGFR levels in gefitinib-resistant H1650 cells.
Figure S6. Sensitivity of phospho-EGFR and phospho-ERK in control and TKI-resistant H1975 and HCC2279 cells. Control and BIBW2992-resistant H1975 cells and control and gefitinib-resistant HCC2279 cells were treated for 2 hours with the indicated TKIs and cell extracts were submitted to immunoblot analyses for phospho-EGFR (Y1068) and phospho-ERK. The filters were stripped and reprobed for total EGFR and ERK as loading controls. The findings demonstrate inhibited phospho-EGFR and phospho-ERK by BIBW2992 in control, but not resistant H1975 cells. Also, the gefitinib-resistant HCC2279 cells show a similar acquired sensitivity to AZ8010and RO4383596 as HCC4006 and H1650 cells in Figure 2.
Figure S7. Selective inhibition of cell proliferation by FGFR1 silencing in gefitinib-resistant HCC4006 cells relative to control cells. Control and gefitinib-resistant HCC4006 cells were transfected with a non-silencing control siRNA or an siRNA targeting FGFR1 as described in the Materials and Methods. Quantitative RT-PCR revealed marked reduction of FGFR1 mRNA by the specific siRNA (left panel), but not FGFR2 mRNA as a negative control (middle panel). Direct measurement of cell numbers following FGFR1 silencing shows inhibition of growth by the FGFR1 siRNA in gefitinib-resistant, but not control, HCC4006 cells. The * indicates p<0.05 by 2-tailed T-test.
Figure S8. Caspase 3 activity in control and gefitinib-resistant lung cancer cell lines following treatment with gefitinib and AZ8010.HCC827, H1650, HCC4006 gefitinib resistant or control cell lines were plated at 80% confluency and treated the following day with DMSO, 0.3 μM gefitinib or 0.3 μM AZ8010 for 72 hours in full media. Caspase-3 activity was measured as described by Matassa et al (J Biol Chem 276: 29719-29728; 2001).
Figure S9. Addition of FGFR TKIsincreasesgrowth reduction by gefitinib in parental HCC4006 and H1650 cells. A) HCC4006 parental cells (100,000) were cultured in RPMI-1640 media containing DMSO, 1µM gefitinib, 0.3µM AZD4547 or gefitinib + 0.3µM AZD4547 for 10 days. Colonies were then rinsed with phosphate-buffered saline, stained with 1mL of 6% (vol/vol) gluteraldehyde, 0.5% (wt/vol) crystal violet in H2O and digitally photographed. B) H1650 parental cells were submitted to an anchorage-independent growth assay with increasing concentrations of gefitinib with or without 100nM AZ8010 or increasing concentrations of AZ8010 alone. After 2 weeks, the colonies were stained and quantified as described in Materials and Methods.
Figure S10. Expression of AXL mRNA in control and TKI-resistant NSCLC cell lines. Total RNA from the indicated cultures was submitted to quantitative RT-PCR analysis of AXL mRNA as described in the Materials and Methods. The data reveal increased AXL mRNA expression in gefitinib-resistant cultures of HCC4006, HCC2279, H1650 and PC9 and in BIBW2992-resistant H1975 cells.
Table S1Cell Line / EGFR genotype / Initial Sensitivity / Resistance / Resistance Mechanism
H1650 / E746-A750 del / Gefitinib / Gefitinib Resistant / FGF2 and FGFR1 induction
H1975 / L858R, T790M / BIBW2992 / BIBW2992 resistant / Unknown
HCC827 / E746-A750 del / Gefitinib / Gefitinib resistant / MET amplification
HCC2279 / E746-A750 del / Gefitinib / Gefitinib resistant / FGF2 and FGFR1 induction
HCC4006 / L747-A750 del, P insert / Gefitinib / Gefitinib resistant / FGF2 and FGFR1 induction
HCC4011 / L858R / Gefitinib / Gefitinib Resistant / Unknown
PC9 / E746-A750 del / Gefitinib / Gefitinib resistant / T790M
HCC2935 / E746-S752 del, I insert / Gefitinib / Resistance not obtained / -
Table S1. The panel of EGFR mutant NSCLC cell lines adapted to EGFR targeted therapies and mechanism of resistance. The EGFR mutant cell lines employed in the study are tabulated with their defined EGFR mutations, initial sensitivity to gefitinib or BIBW2992 (for H1975) and their mechanism of acquired resistance defined in this study. In all cases, the cell lines were serially adapted as pooled cultures to TKIs until the cultures could be routinely passaged in 3µM gefitinib or 2µM BIBW2992 (H1975). A resistant culture of HCC2935 was not obtained after two independent attempts.
Table S2IC50, nM
Cell Line / BIBW2992 / gefitinib / AZ8010
HCC4006 / Control / 1.4 / ND / >1000
HCC4006 / Gefitinib-resistant / >1000 / ND / 18.4
H1650 / Control / 9.3 / ND / >1000
H1650 / Gefitinib-resistant / 1500 / ND / 5.2
HCC2279 / Control / 10.4 / ND / 1739
HCC2279 / Gefitinib-resistant / 407 / ND / 26.4
HCC827 / Control / 1.0 / ND / 2536
HCC827 / Gefitinib-resistant / 2078 / ND / >1000
PC9 / Control / ND / 1.7 / 1216
PC9 / Gefitinib-resistant / ND / 273 / 589
HCC4011 / Control / 9.4 / ND / 84
HCC4011 / Gefitinib-resistant / 921 / ND / 33
H1975 / Control / ND / ND / ND
H1975 / BIBW2992-resistant / ND / ND / ND
Table S2. IC50 values for inhibition of proliferation by EGFR and FGFR-specific TKIs. The IC50 values associated with the dose response curves shown in Figure S1 were calculated using the Prism software program.
Table S3HCC4006 / HCC827
Control-1 / Control-2 / Gefitinib resistant-1 / Gefitinib resistant-2 / Control-1 / Control-2 / Gefitinib resistant-1 / Gefitinib resistant-2
Met / 1202 / 1353 / 1435 / 1521 / 3458 / 3579 / 29463 / 12469
FGFR1 / 21 / 68 / 345 / 431 / 9 / 4 / 16 / 14
FGF2 / 92 / 103 / 275 / 200 / 6 / 40 / 92 / 47
EGFR / 1079 / 662 / 770 / 752 / 7686 / 7184 / 2283 / 580
ERBB2 / 127 / 122 / 56 / 57 / 425 / 214 / 804 / 528
AXL / 1064 / 2004 / 2483 / 2844 / 25 / 54 / 27 / 13
CDH-1 / 1518 / 945 / 3 / 3 / 2372 / 1962 / 2965 / 2596
MAPK1 / 240 / 371 / 631 / 661 / 923 / 869 / 743 / 483
MAPK3 / 610 / 328 / 224 / 290 / 360 / 360 / 481 / 631
JUP / 764 / 800 / 76 / 78 / 950 / 901 / 1185 / 768
Vimentin / 2036 / 3292 / 9074 / 9193 / 224 / 445 / 1132 / 1316
CDH2 / 326 / 355 / 762 / 771 / 0 / 0 / 0 / 0
ZEB1 / 4 / 7 / 56 / 58 / 13 / 15 / 17 / 16
ZEB2 / 2 / 1 / 79 / 78 / 0 / 1 / 4 / 1
Slug / 18 / 67 / 76 / 58 / 18 / 72 / 70 / 42
Snail / 12 / 4 / 6 / 5 / 30 / 27 / 4 / 15
Twist / 78 / 72 / 148 / 176 / 0 / 0 / 0 / 0
Table S3. Expression of selected mRNAs determined by RNA-seq analysis. Two independent mRNA samples obtained from the control and gefitinib-resistant cultures of HCC4006 and HCC827 were submitted to RNA-seq analysis. Abundance of selected mRNAs is reported as fragments per kilobase transcript per million reads (FPKM).
1