Epidermal growth factor receptor (EGFR) is transcriptionally induced by the Y-box binding protein-1 (YB-1) and can be inhibited with Iressa in basal-like breast cancer, providing a potential target for therapy

Abstract

Introduction: Basal-like breast cancers (BLBCs) are very aggressive, and present serious clinical challenges as there are currently no targeted therapies available. We determined the regulatory role of Y-box binding protein-1 (YB-1) on epidermal growth factor receptor (EGFR) overexpression in BLBC, and the therapeutic potential of inhibiting EGFR. We pursued this in light of our recent work showing that YB-1 induces the expression of EGFR, a new BLBC marker.

Methods: Primary tumour tissues were evaluated for YB1 protein expression by immunostaining tissue microarrays, while copy number changes were assessed by comparative genomic hybridization (CGH). The ability of YB-1 to regulate EGFR was evaluated using luciferase reporter, chromatin immunoprecipitation (ChIP) and gel shift assays. The impact of Iressa on monolayer cell growth was measured using an ArrayScan VTI high-throughput analyser to count cell number, and colony formation in soft agar was used to measure anchorage-independent growth.

Results: YB-1 (27/37 or 73% of cases, P = 3.899 x 10(-4)) and EGFR (20/37 or 57.1% of cases, P = 9.206 x 10(-12)) are expressed in most cases of BLBC. However, they are not typically amplified in primary BLBC, suggesting overexpression owing to transcriptional activation. In support of this, we demonstrate that YB-1 promotes EGFR reporter activity. YB-1 specifically binds the EGFR promoter at two different YB-1-responsive elements (YREs) located at -940 and -968 using ChIP and gel shift assays in a manner that is dependent on the phosphorylation of S102 on YB-1. Inhibiting EGFR with Iressa suppressed the growth of SUM149 cells by approximately 40% in monolayer, independent of mutations in the receptor. More importantly anchorage-independent growth of BLBC cell lines was inhibited with combinations of Iressa and YB-1 suppression.

Conclusion: We have identified for the first time a causal link for the expression of EGFR in BLBC through the induction by YB-1 where it binds specifically to two distinguished YREs. Finally, inhibition of EGFR in combination with suppression of YB-1 presents a potential opportunity for therapy in BLBC.

Figure 1

Epidermal growth factor receptor (EGFR) and Y-box binding protein 1 (YB-1) are detected in basal-like breast cancer specimens on a tumour tissue microarray. (a) EGFR-negative staining (40×). (b) Brown cells indicate EGFR positivity (40×), a segment of the core is magnified at 200×. (c) YB-1-negative staining (40×). (d) Brown staining indicates YB-1 positivity (40×), which is detected in both the nucleus and cytoplasm (arrowheads 200×).

Figure 2

Basal-like breast tumours do not exhibit amplifications for epidermal growth factor receptor (EGFR) or Y-box binding protein 1 (YB-1). Primary breast tumours were evaluated for genetic amplifications using SMRT array CGH. DNA was isolated from ten primary basal-like breast tumours and genomic profiles were generated by submegabase resolution tiling array comparative genomic hybridisation. There was no obvious gain of copy number on chromosomes 1 or 7, representing the loci for YB-1 and EGFR, respectively. The exception to this trend was BLC9, where there was a large amplicon on chromosome 7. The lung adenocarcinoma cell line HCC827 was included as a positive control of EGFR amplification.

Figure 3

Y-box binding protein 1 (YB-1) regulates the expression of epidermal growth factor receptor (EGFR) in basal-like breast cancer cells. (a) The levels of YB-1 and EGFR proteins were compared between immortalized breast epithelial cells, 184 htert, SUM149 and HCC1937 basal-like breast cancer cells. Actin was evaluated as a control for equal protein input. (b) 184 htert cells were transfected with an EGFR promoter (1 kb) luciferase construct (pER1), a control renilla plasmid (pRL-TK) and either flag-EV or flag-YB-1 or flag-YB-1(A102). Luciferase and renilla activity were measured after 24 hours. YB-1 induced EGFR promoter activity by 1.5-fold (P = 0.04, N = 6), whereas the A102 mutant did not. (c) SUM149 cells were treated with YB-1 small interfering RNA (siRNA) (5 nM) for 48 h. The cells were then transfected with the EGFR reporter for 24 h and compared with the empty vector. Loss of YB-1 expression resulted in a 78% decrease in EGFR reporter activity (P = 4.53 × 10^-5, N = 6). Inset: evidence that siRNA targeting YB-1 causes a decrease in expression of the protein. Actin was used as a loading control. (d) The same experiment was repeated using HCC1937 cells treated with 20 nM YB-1 siRNA for 48 h. Loss of YB-1 expression resulted in a 77% reduction in EGFR promoter activity (P = 5.98 × 10^-7, N = 6).

Figure 4

Y-box binding protein 1 (YB-1) binds to the epidermal growth factor receptor (EGFR) promoter. (a) Chromatin immunoprecipitation was performed on SUM149 cells. YB-1 binds to the EGFR promoter in the basal-like cells where the 2a loci is the preferred binding site (lane 2). Weak binding was also detected with the 1b primers (lane 1). No binding was observed in the 2b or 3 sites (lanes 3 to 4), nor was there any non-specific binding detected in the IgY negative controls (lanes 5 to 8). Input DNA was diluted fourfold and amplified to demonstrate that the primer produced an expected product (lanes 9 to 12). The no input controls (lanes 13 to 16) are presented to show a lack of non-specific amplifications. (b) Serial ChIP was performed by sequentially pulling down YB-1 and then immunoprecipitating with a phospho-serine antibody. This demonstrated that at least some of the YB-1 is serine phosphorylated when bound to the EGFR 2a site. YB-1 binds to the 2a site (lane 1) as expected. Similarly, the phospho-serine antibody pulls down a complex that can be amplified with the 2a primers (lane 2). Re-ChIP with the YB-1 antibody and subsequently with the phospho-serine antibody also bound to EGFR at the 2a site (lane 3). A phospho-serine YB-1 complex bound to the 2a site on EGFR (lane 3). Species-matched IgG and IgY controls were included to show that the binding was specific (lane 4). The input DNA and no DNA controls were also included (lanes 5 and 6). (c) ChIP was carried out using a phospho-YB-1 antibody (S102), and binding was detected for the EGFR 2a region (lane 4). There was no binding observed when immunoprecipitation was performed using IgG as a control (lane 3). Input DNA was diluted fourfold and amplified to demonstrate that the primer produced an expected product (lanes 5 and 6). Lane 1 is the DNA ladder.

Figure 5

Y-box binding protein 1 (YB-1) binds to specific sites within the epidermal growth factor receptor (EGFR) promoter. (a) Sequence of the EGFR2a oligonucleotide used in the gel shift assays (-979 to -934). Highlighted sequences are the potential YB-1 binding sites. The substitutions made in the two mutants are given under the wild-type sequence. (b) Direct evidence for YB-1 binding to the EGFR promoter using gel shift assays. Nuclear extract from SUM149, MDA-MB-468 or HCC1937 cells were incubated in the presence of the EGFR oligonucleotide spanning -979 to -934. There was no binding in the absence of protein (lanes 1, 5 and 10), whereas the addition of the nuclear extract (lanes 2, 6 and 11) resulted in strong binding that could be inhibited with the unlabelled oligonucleotide (lanes 3, 7 and 12). The addition of a YB-1 antibody caused a supershift (lane 4, 8 and 13) that did not occur when the non-related CREB antibody was used (lanes 9 and 14). (c) Nuclear extracts from 6 primary BLBC samples were pooled and used in a gel shift assay for the EGFR 2a site. Lane 1 contains EGFR2a biotin-labelled oligo only. Binding to the probe is evident in lane 2, which was competed off in lane 3 and supershifted with a YB-1 antibody in lane 4. A CREB antibody was used to demonstrate specificity of the supershift (lane 5). (d) Validation of putative YB-1-responsive elements on the EGFR promoter. SUM149 nuclear extracts were incubated with either wild-type (lane 1) or mutant biotin oligo nucleotides (lanes 3, 4, and 5). A competition reaction was carried out against the wild-type (lane 2). nuclear extract bound to the wild-type sequence (lane 1), but was unable to bind the mutants (lanes 3, 4 and 5).

Figure 6

Inhibiting epidermal growth factor receptor (EGFR) suppresses the growth of basal-like breast cancer cells. (a) Inhibition of EGFR with Iressa (0.25, 0.5, 1 and 2 μM) blocks the growth of basal-like breast cancer cells by up to 40% when the cells were treated for 72 h (0.5 μM P = 0.02, 1 μM P = 0.02, 2 μM P = 0.07). Each experiment was performed in replicates of six on two separate occasions. (b) Anchorage-independent growth was measured by counting colonies formed after 4 weeks exposure to Iressa or vehicle control. Representative images of colonies following each treatment are shown, with average colony number/well shown underneath. The ability to form colonies was completely lost in the presence of concentrations of Iressa as low as 0.25 μM in SUM149 cells. (c) The ability of HCC1937 cells to form colonies was not effected by Iressa alone; however, knockdown of YB-1 significantly reduced the number of colonies (P < 0.001). The addition of Iressa further reduced the number of colonies. This was highly significant at all concentrations (P < 0.001). Statistical analysis carried out using students t-test; *P < 0.05, **P < 0.01.