Synergistic growth inhibition by Iressa and Rapamycin is modulated by VHL mutations in renal cell carcinoma

Abstract

Epidermal growth factor receptor (EGFR) and tumour growth factor alpha (TGFalpha) are frequently overexpressed in renal cell carcinoma (RCC) yet responses to single-agent EGFR inhibitors are uncommon. Although von Hippel-Lindau (VHL) mutations are predominant, RCC also develops in individuals with tuberous sclerosis (TSC). Tuberous sclerosis mutations activate mammalian target of rapamycin (mTOR) and biochemically resemble VHL alterations. We found that RCC cell lines expressed EGFR mRNA in the near-absence of other ErbB family members. Combined EGFR and mTOR inhibition synergistically impaired growth in a VHL-dependent manner. Iressa blocked ERK1/2 phosphorylation specifically in wt-VHL cells, whereas rapamycin inhibited phospho-RPS6 and 4E-BP1 irrespective of VHL. In contrast, phospho-AKT was resistant to these agents and MYC translation initiation (polysome binding) was similarly unaffected unless AKT was inhibited. Primary RCCs vs cell lines contained similar amounts of phospho-ERK1/2, much higher levels of ErbB-3, less phospho-AKT, and no evidence of phospho-RPS6, suggesting that mTOR activity was reduced. A subset of tumours and cell lines expressed elevated eIF4E in the absence of upstream activation. Despite similar amounts of EGFR mRNA, cell lines (vs tumours) overexpressed EGFR protein. In the paired cell lines, PRC3 and WT8, EGFR protein was elevated post-transcriptionally in the VHL mutant and EGF-stimulated phosphorylation was prolonged. We propose that combined EGFR and mTOR inhibitors may be useful in the subset of RCCs with wt-VHL. However, apparent differences between primary tumours and cell lines require further investigation.

Figures

Figure 1

Differential EGFR levels and ERK phosphorylation in RCC cells. (A) Aliquots (10 μg) of cell lysates were analysed by Western blot for levels of EGFR protein. PRC3, MPR6 and WT8 are derived from 786-O cells by transfection of empty vector (PRC3) or independent wild-type VHL expression constructs (MPR6 and WT8). Tubulin served as a loading control. (B) RCC cell lines WT8VHL-wt and PRC3VHL-mut were grown until 50% confluent, then treated for 2 h with DMSO (lanes 1 and 5), rapamycin (10 nM; lanes 2 and 6), Iressa (10 μM; lanes 3 and 7) or both (lanes 4 and 8). Aliquots were analysed for the indicated total and phospho-proteins. (C) WT8VHL-wt and PRC3VHL-mut cells at 50% confluency were treated for 2 h with decreasing doses of Iressa (10–0.1 μM) and analysed as above. (D) WT8VHL-wt cells at 50% confluency were treated for 20 h with 10 or 100 μM CoCl2, or not, as indicated (Co:). For the last 2 h, selected cultures were treated with rapamycin (10 nM), Iressa (10 μM) or both; control cells were treated with DMSO. Cells were washed twice with PBS, harvested and analysed with the indicated antibodies. One lane of PRC3VHL-mut lysate was included as a positive control for the HIF2α antibody (arrow). This lane was not analysed with the other four antibodies.

Figure 2

Growth inhibition by Iressa and rapamycin in derivatives of 786-O cells. The parental RCC line 786-O and three stably transfected derivatives were tested for the effects of varying doses of Iressa and rapamycin on growth using MTT assays. MTT absorbance values from single and combined agents were converted to the combination index (see Methods).

Figure 3

Phospho-protein analysis in additional RCC cell lines. Six RCC cell lines were grown to 50% confluency, then treated for 2 h with DMSO (lanes 1, 5, 9, 13, 17 and 21), rapamycin (10 nM; lanes 2, 6, 10, 14, 18 and 22), Iressa (10 μM; lanes 3, 7, 11, 15, 19 and 23) or both (lanes 4, 8, 12, 16, 20 and 24). Lysates were analysed as in Figure 1. VHL mutational status is indicated along the bottom.

Figure 4

Total and phospho-EGFR levels in PRC3 and WT8. (A) Protein lysates, as described in Figure 1, were analysed for total EGFR protein. The blot was stripped and reprobed with anti-VHL and anti-HA antibodies. (B) PRC3 and WT8 cultures were grown to 50% confluency, washed with PBS, then starved in 0.1% serum for 2 h. Cells were harvested at the indicated time points (min) after the addition of 10% serum and 50 ng ml−1 EGF. More total protein was loaded in the WT8 lanes to compensate for the lower amounts of EGFR in this cell line. Otherwise, equal protein aliquots were analysed for the phosphorylation levels of tyrosine 845 and 1068 using phospho-specific antibodies.

Figure 5

Polysome analysis of WT8 cells. WT8 cells were grown to 70% confluency, then treated for 2 h with DMSO, rapamycin (10 nM), Iressa (10 μM) or both, and harvested for polysome fractionation. A representative A280 fractionation profile is shown along with the fractions pooled and the results from quantitative RT–PCR analysis.

Figure 6

Analysis of primary tumours. (A) Protein (10 μg) from 12 matched pairs of renal tumours (T) and normal kidney (N) were analysed with the indicated antibodies on two separate sets of blots. Equivalent gels were stained with Coomassie blue (CBB) for loading. The low protein content of tumour #10 indicates that relative Erk phosphorylation was even higher than the band intensity suggests. VHL mutational status is summarised by m=mutant; w=wild type. (B) In total, 30 μg of tumour and normal lysates were analysed for levels of EGFR, phospho-RPS6 and total RPS6. The four cell line lysates were treated with Iressa and rapamycin (I/R) for 2 h or not (−). Densitometric analysis for the left panel was normalised to the tumour with the lowest EGFR signal. Densitometry for the right panel utilised a shorter exposure to remain within the linear response range. Even the longest exposures failed to detect EGFR in T-12. The asterisk denotes a background band. (C) Direct comparisons of phospho-Erk and phospho-Akt levels in a subset of tumours and cell lines on the same filter. (D) WT8VHL-wt cultures starting at 70% confluency (time zero) were grown under standard culture conditions over 48 h and periodically sampled. Phospho-RPS6 levels were downregulated at 100% confluency.