Breast tumor progression induced by loss of BTG2 expression is inhibited by targeted therapy with the ErbB/HER inhibitor lapatinib

Article metrics

Abstract

The B-cell translocation gene-2 (BTG2), a p53-inducible gene, is suppressed in mammary epithelial cells during gestation and lactation. In human breast cancer, decreased BTG2 expression correlates with high tumor grade and size, p53 status, blood and lymph vessel invasion, local and metastatic recurrence and decrease in overall survival, suggesting that suppression of BTG2 has a critical role in disease progression. To analyze the role of BTG2 in breast cancer progression, BTG2 expression was knocked down in mammary epithelial cells. Suppression of BTG2 enhances the motility of cells in vitro and tumor growth and metastasis in vivo. The effects of BTG2 knockdown are mediated through stabilization of the human epidermal growth factor receptor (HER) ligands neuregulin and epiregulin and activation of the HER2 and HER3 receptors, leading to elevated AKT phosphorylation. Suppression of HER activation using the tyrosine kinase inhibitor lapatinib abrogates the effects of BTG2 knockdown, including the increased cell migration observed in vitro and the enhancement of tumorigenesis and metastasis in vivo. These results link BTG2-dependent effects on tumor progression to ErbB receptor signaling, and raise the possibility that targeted inhibition of this pathway may be relevant in the treatment of breast cancers that have reduced BTG2 expression.

Introduction

The B-cell translocation gene-2 (BTG2) is a member of the BTG/Tob antiproliferative protein family, which contains two highly conserved domains known as BTG boxes A and B separated by a spacer sequence of 20 to 25 nonconserved amino acids. BTG1 and BTG2 also contain an additional short stretch of amino acids designated as box C located between positions 118 and 128 (Winkler, 2009). The conserved regions serve as protein–protein interaction modules of the BTG/Tob family. BTG2 inhibits cellular proliferation by impairing G1 to S progression through suppression of cyclin D and cyclin E, which results in reduced cyclin-dependent kinase activity (Lim et al., 1998; Guardavaccaro et al., 2000).

The epithelia of many tissues including the mammary gland express BTG2 (Melamed et al., 2002). In the normal mammary epithelium, BTG2 levels decline during pregnancy and lactation but recover upon the weaning of pups. Estrogen and progestin, which modulate mammary gland proliferation and differentiation during pregnancy, suppress BTG2 (Kawakubo et al., 2006) and are likely to be responsible for pregnancy-associated decline in BTG2 expression in the mammary gland.

In estrogen receptor-positive breast tumors, loss of BTG2 protein expression was associated with increasing tumor grade and size and overexpression of the cyclin D1 protein (Kawakubo et al., 2006), indicating that BTG2 may have a role in the etiology of the disease. Identification of BTG2 as a transcriptional target of p53 (Cortes et al., 2000), and as a key mediator of p53-dependent cellular transformation of mouse and human fibroblasts transduced with oncogenic Ras (Boiko et al., 2006), suggests that it has an important role in inhibiting tumorigenesis. This is consistent with the repression of BTG2 expression in several human tumors including the breast, thymus, prostate and kidney (Lim et al., 1995; Ficazzola et al., 2001; Kawakubo et al., 2004, 2006; Struckmann et al., 2004). Moreover, expression of BTG2 in a mouse model for medulloblastoma suppressed the formation of tumors (Farioli-Vecchioli et al., 2007). Although BTG2 has been shown to interact with transcription factors (Prevot et al., 2000), promote mRNA deadenylation and turnover (Mauxion et al., 2008, 2009) and modulate signaling through its interaction with signaling molecules (Park et al., 2004; Hong et al., 2005), the precise mechanism by which its loss contributes to tumor progression is not well understood.

Activation of various tyrosine kinase receptors, including the human epidermal growth factor receptors (HERs), and increased expression of their respective ligands has been implicated in the autocrine and paracrine regulation of breast cancer growth and metastasis (Tripathy and Benz, 1992). Amplification and overexpression of epidermal growth factor receptor (EGFR) and HER2 are observed in 6 and 25% of human breast cancers, respectively, and a role for the kinase-defective HER3 in these tumors is becoming increasingly evident (Hsieh and Moasser, 2007). Identification of pathways that facilitate the activation of HERs can help realize the full potential of the current therapies that specifically target the HER pathway.

In this paper we demonstrate that decreased BTG2 expression in human breast tumors promotes disease progression. Consistent with the clinical features of human breast cancers associated with decreased BTG2 expression, knockdown of BTG2 in mammary epithelial cells promotes cell migration through increase in the expression of the HER family ligands and activation of the HER pathway. Although loss of BTG2 by itself does not transform cells, it cooperates with oncogenic H-RasV12 to promote tumor growth and distal metastasis in vivo. Lapatinib is a dual tyrosine kinase inhibitor that blocks both EGFR and HER2, and is already successfully used in the treatment of metastatic breast carcinoma (Geyer et al., 2006; Johnston et al., 2009). In this study we show that inhibition of the ErbB/HER receptor pathway with lapatinib suppresses the tumor-promoting effects of BTG2 repression. These results suggest that therapies that target the ErbB/HER pathway may be relevant in breast carcinomas with decreased BTG2 expression.

Results

Decreased BTG2 expression in breast tumors is associated with tumorigenic characteristics

We had previously demonstrated that decreased expression of BTG2 protein in estrogen receptor-positive breast tumor was associated with high tumor grade and size, and overexpression of cyclin D1, whose expression is suppressed by BTG2 (Guardavaccaro et al., 2000). Analysis of the Oncomine data set to determine whether the expression of BTG2 mRNA in breast tumors correlated with other tumor characteristics demonstrated that suppression of BTG2 mRNA expression in tumors correlated with the presence of metastatic events (van de Vijver et al., 2002; van ‘t Veer et al., 2002; Yu et al., 2006; Desmedt et al., 2007; Loi et al., 2007, 2008; Schmidt et al., 2008) and blood (van ‘t Veer et al., 2002) and lymphatic vessel invasion (Figure 1a) (Yu et al., 2006). Moreover, several studies revealed a strong correlation between decreased BTG2 expression and recurrence (Figure 1b). These results were derived from microarray analysis of mRNA isolated from tissue samples (van de Vijver et al., 2002; Sorlie et al., 2003; Sotiriou et al., 2003; Ma et al., 2004; Wang et al., 2005; Loi et al., 2007, 2008), laser capture microdissected tumor cells (Ma et al., 2004; Boersma et al., 2008; Finak et al., 2008), biopsied material (Huang et al., 2003), surgical specimens (Loi et al., 2007; Schmidt et al., 2008) and estrogen receptor-positive tamoxifen-treated primary breast cancer samples (Ma et al., 2004; Loi et al., 2007, 2008). Consistent with these observations, several studies demonstrate that decreased expression of BTG2 in breast cancer correlates with decreased overall survival (Figure 1c) (van de Vijver et al., 2002; Sorlie et al., 2003; Sotiriou et al., 2003; Pawitan et al., 2005; Bild et al., 2006; Boersma et al., 2008).

Figure 1
figure1

Suppression of BTG2 is associated with metastasis, recurrence and decreased overall survival. (a) Correlation between BTG2 expression and clinical characteristics associated with cancer progression. Tumors with lower BTG2 expression demonstrate increased propensity to metastasis (van de Vijver et al., 2002), and increased lymphatic and blood vessel invasion (van ‘t Veer et al., 2002; Yu et al., 2006). Seven different data sets demonstrated correlation between decreased BTG2 mRNA expression and metastasis (van de Vijver et al., 2002; van ‘t Veer et al., 2002; Yu et al., 2006; Desmedt et al., 2007; Loi et al., 2007, 2008; Schmidt et al., 2008). Box plots from the studies indicated are shown. (b) Correlation between BTG2 expression and recurrence of breast cancer. Lower BTG2 expression in breast cancer correlates with local and metastatic recurrence at 5 years (van de Vijver et al., 2002). Similar results were obtained in RNA isolated from laser capture microdissected (LCM) breast cancer samples (Ma et al., 2004), and RNA isolated from estrogen receptor-positive tamoxifen (TAM)-treated primary breast cancer samples (Loi et al., 2007). Twelve different data sets demonstrate similar correlation between decreased BTG2 mRNA expression and recurrence (van de Vijver et al., 2002; Huang et al., 2003; Sorlie et al., 2003; Sotiriou et al., 2003, 2006; Ma et al., 2004; Wang et al., 2005; Loi et al., 2007, 2008; Boersma et al., 2008; Finak et al., 2008; Schmidt et al., 2008). Box plots from the studies indicated are shown. (c) Correlation between BTG2 expression and overall survival. Lower BTG2 expression in breast cancer correlates with decreased survival (van de Vijver et al., 2002). Six different data sets demonstrate similar correlation between decreased BTG2 mRNA expression and survival (van de Vijver et al., 2002; Sorlie et al., 2003; Sotiriou et al., 2003; Pawitan et al., 2005; Bild et al., 2006; Boersma et al., 2008). Box plot from the study indicated is shown. For all three panels, the y axis shows normalized expression units or Z-scores. All data are log transformed, median centered and the 25th–75th percentiles are indicated by the closed box. The number of breast carcinoma samples present in each group is shown within parenthesis. LR, local recurrence in 5 years; MR, metastatic recurrence in 5 years; Neg, negative; NR, no recurrence in 5 years; NV, no value; Pos, positive. * above and below the boxes show sample maximum and minimum values.

Knockdown of BTG2 expression increases cell motility

As decreased BTG2 expression in breast tumors demonstrated a strong correlation with disease progression and survival, we decided to explore the functional consequence of decreased BTG2 expression in breast cancer. To this end, the expression of BTG2 was knocked down in MCF10A cells using two short hairpins (sh) against the coding sequence of BTG2; the expression of the protein was greatly reduced by shBTG2 (Figure 2a, left). Decreased BTG2 expression reduced proliferation (Supplementary Figure 1) but increased cell motility in wound healing assays, and increased migration and invasion of cells in Boyden chamber assays (Figure 2a, right panels). To avoid clonal selection effects, all experiments were performed with uncloned pools of cells. Restoration of BTG2 expression in shBTG2-MCF10A cells reversed this phenotype (Figure 2b). Similarly, suppression of BTG2 in human mammary epithelial cells (HMECs) was also associated with increased cellular motility (Figure 2c). Conversely, expression of BTG2 in the highly invasive human breast cancer cell line MDA-MB-231 with nearly undetectable levels of endogenous BTG2 transcripts suppressed the migratory properties of these cells (Figure 2d). These findings demonstrate that expression of BTG2 suppresses cell motility.

Figure 2
figure2

Suppression of BTG2 expression increases cell motility. (a, left) Expression of BTG2 protein in MCF10A cells infected with shGFP or shBTG2 lentivirus. Parental MCF10A cells were used as controls. Knockdown of BTG2 increases cell motility. (Middle) shGFP- and shBTG2-infected MCF10A cells were grown to confluence and the cultures were viewed at 0 and 24 h of wounding. (Right) Migration and invasion of shGFP- and shBTG2-infected MCF10A cells were assayed using Transwell units. The mean was derived from cell counts of nine fields. *P<0.001. (b) Reintroduction of BTG2 reverses the migration and invasion of cells. shBTG2-MCF10A cells were infected with a lentiviral BTG2 construct. Upper panel demonstrates the restoration of BTG2 protein expression and the lower panels demonstrate that it decreases invasion and migration of cells. The mean was derived from cell counts of nine fields. *P<0.001. (c) BTG2 expression was knocked down in HMECs with shBTG2. Upper panel demonstrates the decrease in BTG2 protein levels in cells, and the lower panels demonstrate the increase in migration and invasion. The mean was derived from cell counts of nine fields. *P<0.001. (d) Expression of BTG2 suppresses the motility of MDA-MB-231 cells. (Upper panel) BTG2 mRNA expression in cells transfected with lentiviral vector or BTG2. (Lower panel) Migration and invasion of the cells was evaluated using Transwell assays. The mean was derived from cell counts of nine fields. *P<0.001.

Suppression of BTG2 leads to activation of the HER pathway

We then set out to identify the mechanism responsible for enhanced motility of cells with decreased BTG2 expression. As activation of HER has been implicated in enhancement of cell motility (Hiscox et al., 2006; Lo et al., 2007; Sabbah et al., 2008), we tested the activation of HERs in cells infected with shBTG2. Although the expression of total EGFR, HER2 and HER3 expression was not altered, phosphorylation of HER2 and HER3 as well as AKT was induced in shBTG2-MCF10A cells either in the presence or absence of EGF in the medium (Figure 3a, left). Similar results were obtained with HMECs infected with shBTG2 (Figure 3a, right). The expression of NRG1β (neuregulin1β), a ligand that binds to both HER3 and HER4 receptors, was increased in cells with suppressed BTG2 expression (Figure 3b, left) because of enhanced stability of the transcript in BTG2 knockdown cells (Figure 3b, right); NRG1β mRNA decays faster in control cells compared with that observed in BTG2-repressed cells. Restoration of BTG2 expression in shBTG2-MCF10A cells led to a reduction in HER2, HER3 and AKT phosphorylation and NRG1β expression (Figure 3c). Consistent with this finding, analysis of breast cancer data sets (Oncomine; www.oncomine.org) demonstrated an inverse correlation between BTG2 mRNA and NRG1β expression (P=0.005086961, r=−0.8037, number of patients with relapse=10 (Desmedt et al., 2007); P=0.006373995, r=−0.1851, n=216 (van de Vijver et al., 2002)). To determine whether upregulation of NRG1β has a role in increased migration of the cells, NRG1β expression in BTG2 knockdown cells was suppressed with small interfering RNA against NRG1. As shown in Figure 3d, suppression of NRG1β expression decreased AKT phosphorylation and the invasive properties of these cells.

Figure 3
figure3

Activation of the HER pathway increases cell motility of BTG2-repressed cells. (a, left) HER and AKT phosphorylation was examined in proteins isolated from shGFP- and shBTG2-MCF10A cells grown in the presence and absence of EGF. Antibodies against phospho- and total EGFR, HER2, HER3 and AKT were used. (Right) Proteins isolated from HMECs infected with shGFP and shBTG2 were similarly analyzed. The bar graphs to the right of the blots show the fold change in phospho-protein expression in BTG2 knockdown cells calculated by setting the ratio of band intensities of phospho-protein/total protein in control cells at 1. (b, left) NRG1β mRNA expression in shGFP- and shBTG2-MCF10A cells was analyzed by quantitative PCR (qPCR). (Right) qPCR of NRG1β mRNA in untreated and actinomycin D-treated BTG2 knockdown cells demonstrates that NRG1β mRNA is more stable in BTG2 knockdown cells. (c) Restoration of BTG2 in shBTG2-MCF10A cells reduces the phosphorylation of HER2, HER3 and AKT and the expression of NRG1β mRNA. The bar graph in the middle shows the fold change in phospho-protein expression in BTG2-expressing cells, calculated by setting the ratio of band intensities of phospho-protein/total protein in control cells at 1. (d) Suppression of NRG1β expression decreases the AKT phosphorylation and migration in BTG2 knockdown cells. NRG1β expression in shBTG2-MCF10A cells was suppressed with small interfering RNA (siRNA) against NRG; the expression of NRG1β (left), pAKT (middle) and cell migration (right) are shown. The bar graph below the protein blot shows the fold change in phospho-protein expression in NRG-suppressed cells, calculated by setting the ratio of band intensities of phospho-protein/total protein in control cells at 1. (e, left) Epiregulin mRNA expression in shGFP- and shBTG2-MCF10A cells was analyzed by qPCR. (Middle) Restoration of BTG2 in shBTG2-MCF10A cells suppresses epiregulin expression. (Right) qPCR of epiregulin mRNA in untreated and actinomycin D-treated BTG2 knockdown cells demonstrates that epiregulin mRNA is more stable in BTG2 knockdown cells. Lower panels show the suppression of epiregulin in shBTG2-MCF10A cells with siRNA against epiregulin (left) and the decrease in pAKT (middle) and cell migration (right). The bar graph below the protein blot shows the fold change in phospho-protein expression in epiregulin-suppressed cells, calculated by setting the ratio of band intensities of phospho-protein/total protein in control cells at 1.

In addition to NRG1β, epiregulin, another HER ligand that can signal through all combinations of HERs except for homodimeric HER2 (Shelly et al., 1998), was also induced through transcript stabilization upon knockdown of BTG2. Suppression of epiregulin in BTG2-suppressed cells also led to decreased phosphorylation of AKT in these cells and cell migration (Figure 3e).

BTG2-suppressed cells were more resistant to treatment with doxorubicin, a chemotherapeutic drug, but were sensitive to ErbB/HER kinase inhibitor, lapatinib, and the HER2 inhibitor, herceptin (Supplementary Figure 2). Moreover, treatment of shBTG2-MCF10A cells with lapatinib mitigated cellular migration and invasion (Figure 4a left). Lapatinib at 0.5 μM concentration decreased the migration and invasion of BTG2 knockdown cells by 3–4-fold but did not affect control cells. At a higher concentration (2 μM), lapatinib decreased the migration and invasion of shBTG2 cells by 20- and 10-fold, respectively, whereas that of control cells was decreased by just 2-fold. These results demonstrate that the migratory properties of cells with decreased BTG2 are more sensitive to lapatinib treatment. LY294002, a phosphatidylinositol 3-kinase inhibitor that suppresses downstream HER signaling, also suppressed the migration and invasion of BTG2-repressed cells (Figure 4a right). Suppression of cellular motility by lapatinib and LY294002 was associated with suppression of AKT phosphorylation (Figure 4b). Expression of BTG2 in the triple-negative (negative for estrogen receptor, progestin receptor and HER2) breast cancer cell line MDA-MB-231 led to a decline in cellular invasion (Figure 2d), and was associated with reduced HER3 and AKT phosphorylation (Figure 4c). Taken together, the combination of loss-of-function and gain-of-function model systems suggests that BTG2 expression modulates cellular migration and invasion through activation of HER–AKT axis.

Figure 4
figure4

Lapatinib suppresses the migration and invasion of BTG2-repressed cells. (a, left) BTG2 knockdown MCF10A cells were treated with lapatinib for 24 h and the migration and invasion of cells was monitored using Transwell assays. The mean was derived from cell counts of nine fields. *Comparison between untreated groups; **, *** comparison between untreated and lapatinib-treated shBTG2-MCF10A cells. P<0.001 in all cases. (Right) shBTG2-MCF10A cells were treated with 10 μM LY294002 for 24 h and the migration and invasion of cells was monitored using Transwell assays. The mean was derived from cell counts of nine fields. *P<0.001. (b) Suppression of AKT phosphorylation following lapatinib and LY294002 treatment. shBTG2-MCF10A cells were serum starved overnight and stimulated with 30 ng/ml EGF for 20 min in the presence of either vehicle and 10 μM LY294002 (left panel) or lapatinib (right panel). Proteins were analyzed by western blot. The bar graphs below each blot show the change in phospho-protein expression, calculated as the ratio of phospho-protein/total protein band intensities. (c) BTG2 expression in MDA-MB-231 cells suppresses HER3 and AKT phosphorylation. Triple-negative MDA-MB-231 cells were infected with a lentivirus expressing BTG2 and proteins were analyzed with antibodies against phospho- and total EGFR, HER3 and AKT. The bar graph on the right shows the fold change in phospho-protein expression. Fold change was calculated by setting the ratio of phospho-protein/total protein band intensities in control cells at 1.

Loss of BTG2 promotes breast tumor growth and distal metastasis

In order to determine whether increase in motility exhibited in vitro by BTG2 knockdown cells would confer an enhanced metastatic phenotype in vivo, tumorigenesis studies were carried out in mice. Loss of BTG2 by itself was unable to transform MCF10A cells and was not sufficient to induce the formation of tumors when injected into nude mice (data not shown). However, consistent with the report that repression of BTG2 cooperates with oncogenic Ras to transform wild-type p53 containing primary mouse fibroblasts (Boiko et al., 2006), suppression of BTG2 cooperated with activated G12V H-Ras allele (H-RasV12) to transform MCF10A cells (see below). The ShBTG2/ H-RasV12 cells were more proliferative (Supplementary Figure 1) and invasive compared with cells expressing H-RasV12 alone (Figure 5a, left). Matrigel/collagen three-dimensional assay to assess the ability of cells to invade through collagen demonstrated that BTG2 knockdown cells expressing both H-RASV12 showed a prominent invasive phenotype, with multiple cellular protrusions extending into the Matrigel compared with cells expressing H-RASV12 alone (Figure 5a, right). Western blot analysis indicated that the levels of phospho-HER2 and HER3 and phospho-AKT were elevated in these cells compared with cells infected with H-RasV12 alone (Figure 5b).

Figure 5
figure5

Knockdown of BTG2 promotes distal metastasis. (a, left) Invasion of shLuc, shBTG2, shLuc+ H-RasV12 and shBTG2+ H-RasV12-expressing MCF10A cells was evaluated. (Right) The invasive feature of shLuc+ H-RasV12 and shBTG2+ H-RasV12-expressing MCF10A cells was monitored in three-dimensional (3D) collagen/matrigel culture. A representative figure and the mean number of cells with invasive features (±s.d.; n=3; P=0.0094) are shown. (b) Expression of phospho- and total EGFR, HER2, HER3 and AKT expression in shLuc- and shBTG2-infected MCF10A cells expressing activated H-RasV12. The bar graph on the right shows the fold change in phospho-protein expression. Fold change was calculated by setting the ratio of phospho-protein/total protein band intensities in control cells at 1. (c, left) Changes in the mean tumor volume of shLuc+ H-RasV12 and shBTG2+ H-RasV12-expressing MCF10A cells injected into the mammary fat pad of mice (n=10 mice per group). (Right) The number of GFP-positive metastatic nodules in the lungs and livers of mice harboring shLuc+ H-RasV12 and shBTG2+ H-RasV12-expressing MCF10A tumors was measured (n=5 fields; n=10 mice). P-values are shown.

We then investigated the ability of the shBTG2/ H-RasV12 cells to form tumors in immune-compromised mice. H-Ras mutations are rare in human breast cancer, but activation of the Ras pathway is an essential characteristic of receptor tyrosine kinase activation, and as such is a valuable model for growth factor receptor-mediated oncogenesis. H-RasV12-expressing MCF10A cells infected with shLuc or shBTG2 formed palpable tumors in mice. At 4 weeks after inoculation, the mean volumes of tumors formed by cells expressing activated H-RasV12/shBTG2 and H-RasV12 alone were not significantly different (Figures 5c and 6a), whereas at 6 weeks the tumors formed by H-RasV12/shBTG2 cells were significantly larger than those formed by H-RasV12 cells (Figure 6a, P=0.035). Interestingly, in both experiments, the number of metastatic nodules present in the lung and liver were significantly higher in H-RasV12/shBTG2 tumor-bearing mice compared with the H-RasV12 mice (Figures 5c and 6c), suggesting that loss of BTG2 promotes distal metastasis.

Figure 6
figure6

Lapatinib treatment inhibits tumor progression induced by the loss of BTG2. (a) Mice harboring shLuc- and shBTG2-H-RasV12 tumors were treated with vehicle or lapatinib. Treatment was begun after 2 weeks of tumor growth. Tumor volumes were measured in all four groups (n=8 mice per group). (b) H-RasV12/shBTG2- MCF10A cells were serum starved overnight and subsequently stimulated with 30 ng/ml EGF for 20 min and either dimethylsulfoxide (DMSO) or various concentrations of lapatinib. Proteins were analyzed by western blot. (c) The number of metastatic nodules in the lungs and livers of mice harboring shLuc+ H-RasV12 and shBTG2+ H-RasV12-expressing MCF10A tumors treated with vehicle or lapatinib was measured (n=5 fields; n=8 mice). P-values are shown (* comparison between untreated groups; ** comparison between untreated and lapatinib-treated H-Ras/shBTG2 group). (d) Lapatinib inhibits proliferation and induces apoptosis in H-RasV12/shBTG2-MCF10A cells. Tissue sections from the four groups of tumors were immunostained with proliferation cell nuclear antigen (PCNA) and activated caspase-3 to evaluate proliferation and apoptosis, respectively. P-values are shown (* comparison between untreated and lapatinib-treated H-Ras/shBTG2 groups).

Lapatinib suppresses tumor progression induced by loss of BTG2

As suppression of BTG2 increases the HER–AKT signaling axis and cell motility, we tested whether treatment of MCF10A-H-RasV12/shBTG2 tumor-bearing mice with lapatinib, an inhibitor of HER activation, would interfere with tumor progression in vivo. As shown in Figure 6a, growth was not significantly different between H-RasV12 and H-RasV12/shBTG2-expressing tumors for up to 4 weeks. However, at 6 weeks the H-RasV12/shBTG2 tumors were significantly larger than tumors expressing H-RasV12 alone. Treatment of mice with lapatinib, which decreases the phosphorylation of EGFR, HER2, HER3 and AKT in shBTG2/H-RasV12 cells (Figure 6b), strongly suppressed the growth of shBTG2-MCF10A/H-RasV12 tumors (P=0.035) but did not significantly affect the growth of H-RasV12-expressing tumors (P=0.469). As seen in the previous experiment (Figure 5c), the number of metastatic nodules in the lung and liver was higher in mice bearing shBTG2/H-RasV12 tumors compared with those bearing H-RasV12 tumors, and lapatinib suppressed the number of metastatic nodules formed in the lung and liver of animals harboring shBTG2/H-RasV12 tumors but did not significantly decrease the fewer metastatic nodules observed in control mice (Figure 6c). Examination of untreated and lapatinib-treated tumor tissues from both groups of mice suggests that lapatinib significantly decreases proliferation and enhances apoptosis in shBTG2/H-RasV12 tumors, but not in tumors expressing H-RasV12 alone (Figure 6d). These results suggest that growth and metastatic potential of BTG2-suppressed tumors are sensitive to inhibition of the HER pathway.

Discussion

The expression of BTG2 is suppressed in several types of cancers including breast cancer. We had previously shown that loss of BTG2 protein in breast cancer is associated with higher tumor grade and size and overexpression of the cyclin D1 protein (Kawakubo et al., 2004, 2006). Analysis of several Oncomine microarray data sets corroborated these findings and revealed a strong correlation between decreased BTG2 expression in breast cancer and characteristics of tumor progression. Decreased BTG2 expression in the tumor correlated with increased lymphatic and blood vessel invasion (van ‘t Veer et al., 2002; Yu et al., 2006), and the presence of metastatic events at 1, 3 and 5 years of follow-up (van de Vijver et al., 2002; van ‘t Veer et al., 2002; Desmedt et al., 2007; Loi et al., 2007, 2008; Schmidt et al., 2008). Reduced BTG2 expression in breast tumor samples was also associated with both local and distal recurrence (van de Vijver et al., 2002; Huang et al., 2003; Sorlie et al., 2003; Sotiriou et al., 2003, 2006; Ma et al., 2004; Wang et al., 2005; Finak et al., 2008), and recurrence following tamoxifen treatment (Loi et al., 2007, 2008). Moreover, patients with decreased BTG2 exhibited poor overall survival (van de Vijver et al., 2002; Sorlie et al., 2003; Sotiriou et al., 2003, 2006; Pawitan et al., 2005; Bild et al., 2006; Boersma et al., 2008).

To determine whether decreased BTG2 expression in breast cancer is functionally critical for the process of breast cancer metastasis, we modulated the expression of BTG2 in the non-tumorigenic MCF10A and HMEC cells, and in the highly invasive MDA-MB-231 cells, which have undetectable levels of BTG2 mRNA. Consistent with the data obtained in the human breast cancer microarray data sets, suppression of BTG2 increased cell motility vitro, and enhanced tumor growth and distal metastasis in vivo, indicating that decreased BTG2 expression in breast tumors functionally contributes to disease progression.

Mechanistically, our data demonstrate that BTG2 suppresses the activation of the HER pathway. Phosphorylation of HER2, HER3 and AKT was elevated upon suppression of BTG2. Although HER2 can homo- or heterodimerize with EGFR or HER3, HER2/HER3 heterodimers are the most potent mitogenic and transforming complex (Hynes et al., 2001), and heterodimers containing HER3 have been implicated in enhanced cell migration and invasiveness (Sithanandam et al., 2005; Xue et al., 2006). Activation of HER pathway in shBTG2 cells was coincident with the increase in NRG1β, a ligand that binds to both HER3 and HER4 receptors (Zhang et al., 1997) and epiregulin, a component of the 18-gene breast-to-lung metastatic gene expression signature (Minn et al., 2005). Epiregulin binds to various HER receptor combinations with lower affinity than HER ligands that have more stringent selectivity, but produces a mitogenic signal more potent than EGF (Shelly et al., 1998). BTG2-induced polyA degradation has been shown to result in mRNA decay (Mauxion et al., 2008); consistent with this finding, the transcripts encoding NRG1β and epiregulin were stabilized in BTG2 knockdown cells. Thus, it is likely that loss of BTG2 modifies the tumor microenvironment leading to paracrine and autocrine activation of the HER pathway. In fact, analysis of breast cancer data sets demonstrates an inverse correlation between BTG2 mRNA and these HER ligands ((van de Vijver et al., 2002; Desmedt et al., 2007; Finak et al., 2008). Interestingly, neuregulin was able to reverse BTG2-mediated cell cycle arrest (Supplementary Figure 3). These results, taken together with the correlation between decreased BTG2 expression and HER2 receptor positivity (Ma et al., 2004; Zhao et al., 2004; Bild et al., 2006; Saal et al., 2007; Julka et al., 2008) and loss of PTEN (Saal et al., 2007) in human breast cancers, suggest that suppression of BTG2 facilitates the activation of the HER2/HER3 pathway. A similar trend has been reported in gliomas; BTG2 expression is downregulated in high-grade gliomas addicted to PDGF-B (platelet-derived growth factor β polypeptide; Calzolari et al., 2008).

Activation of receptor tyrosine kinases promotes the growth of several tumors including breast cancer, and therapies that target activated receptor tyrosine kinases are becoming increasingly useful clinically. Up to 25–30% of breast cancer patients exhibit amplification and overexpression of HER2. Lapatinib is a small-molecule competitive tyrosine kinase inhibitor that binds reversibly in the kinase domains of EGFR and HER2. Recently, cells with increased HER3 expression were also found to be sensitive to lapatinib, suggesting that lapatinib may be targeting additional members of the ErbB/HER family (Konecny et al., 2008). Our results indicate that lapatinib inhibits the growth of H-RasV12/shBTG2 tumors by decreasing proliferation and inducing apoptosis. Tumors expressing H-RasV12 alone were palpable but consistent with their undetectable levels of phospho-HER2, -HER3 and AKT, were not sensitive to lapatinib.

Although HER2 gene copy number and protein level are the best predictable markers for lapatinib responsiveness (Esteva et al., 2010), just 20% of untreated HER2-amplified metastatic patients respond to first-line lapatinib treatment in a phase II trial (Gomez et al., 2008), suggesting that additional factors besides HER2 overexpression may be modulating responses to receptor tyrosine kinase inhibitors. In fact, lapatinib was recently shown to restore endocrine sensitivity in breast cancer cells with acquired resistance because of modest adaptive upregulation of HER2 (Leary et al., 2010). Moreover, lack of p53 and phosphorylated HER3 have been associated with better lapatinib responses in HER2-overexpressing patients; HER2-overexpressing tumors also exhibit increased expression of heregulin/neuregulin (Johnston et al., 2008). Our results for the first time link BTG2-dependent effects on cell transformation to ErbB receptor signaling, and raise the possibility that inhibition of this pathway may be relevant in the treatment of breast cancers that have reduced BTG2 expression.

In addition to suppressing the growth of the primary tumor, lapatinib also suppressed distal metastasis induced by loss of BTG2. HER2 activation has been implicated in intravasation of tumor cells in mouse models of breast cancer (Kedrin et al., 2009), and lapatinib has been shown to inhibit the brain metastasis of MDA-MB-231 cells injected into the left cardiac ventricle of mice (Gril et al., 2008). Whether lapatinib treatment by inhibiting HER2/HER3 phosphorylation prevents the seeding of shBTG2/ H-RasV12 tumor cells into the blood remains to be determined.

In conclusion, suppression of BTG2 in human breast cancer has a critical role in disease progression through activation of the HER pathway, and inhibitors of the HER pathway may be useful in the clinical management of these tumors.

Materials and methods

Cell culture

Culture conditions used for the maintenance of cell lines are described (Kawakubo et al., 2004). For three-dimensional cultures, cells were cultured in growth factor-reduced reconstituted basement membrane (Matrigel; BD Biosciences, Research Triangle Park, NC, USA) as described (Debnath et al., 2003). To evaluate the effect of BTG2 expression in cells, MDA-MB-231 cells were infected with BTG2-expressing lentiviral constructs as described (Hayashida et al., 2010).

Knockdown of BTG2 in MCF10A cells

Lentiviruses carrying the sequences 5′-IndexTermCAAGAACTACGTGATGGCAGT-3′ and 5′-IndexTermCGTGAGCGAGCAGAGGCTTAA-3′ targeting the coding sequence of BTG2 were obtained from the RNAi Consortium shRNA Library (Moffat et al., 2006), the Broad Institute, and were used to knockdown BTG2 expression in MCF10A and HMECs as described in Hayashida et al. (2010).

Western blot analysis

Expression of the BTG2 protein was detected using a rabbit anti-BTG2 antibody generated against the KLH-peptide conjugate. The peptide VSEQRLKVFSGALQC spans amino acids 30 to 43 of the BTG2 protein and does not share any homology with other members of the BTG/Tob family. Antibodies against cleaved caspase-3, phospho-AKT, phospho-HER3, total HER3, phospho-HER2, total HER2 (Cell Signaling Technology, Danvers, MA, USA), phospho-EGFR (Abcam, Cambridge, MA, USA) and total EGFR were also used for western blot analyses. The western protocols used are as previously described (Ha et al., 2000). Band intensities of western blots were measured by using the software Quantity One version 4.6.7 (Bio-Rad, Hercules, CA, USA). The fold change in phospho-protein expression in BTG2-knockdown or -expressing cells was calculated by setting the ratio of band intensities of the phospho-protein/total protein in control cells at 1.

RNA analysis

Total RNA from cells was extracted and converted to complementary DNA. Quantitative PCR to detect the expression of BTG2, COX2, MMP1, MMP2, epiregulin and NRG1β was done using the following gene-specific primers.

BTG2

Forward 5′-IndexTermGCGAGCAGAGGCTTAAGGTCTTC-3′

Reverse 5′-IndexTermATGCGAATGCAGCGGTAGC-3′.

COX2

Forward 5′-IndexTermATATGTTCTCCTGCCTACTGGAA-3′

Reverse 5′-IndexTermGCCCTTCACGTTATTGCAGATG-3′.

MMP1

Forward 5′-IndexTermACACATCTGACCTACAGGATTGA-3′

Reverse 5′-IndexTermGTGTGACATTACTCCAGAGTTGG-3′.

MMP2

Forward 5′-IndexTermGCCCCAGACAGGTGATCTTG-3′

Reverse 5′-IndexTermGCTTGCGAGGGAAGAAGTTGT-3′.

NRG1β

Forward 5′-IndexTermCATCACTGGCTGATTCTGGA-3′

Reverse 5′-IndexTermTTCCACGATGGTGATATTGG-3′.

Epiregulin

Forward 5′-IndexTermCTGGGTTTCCATCTTCTACAGG-3′

Reverse 5′-IndexTermCGTGGATTGTCTTCTGTCTGAACT-3′.

Primers specific for GAPDH were used as control.

Sequence of the small interfering RNA against NRG1: 5′-IndexTermGGGAGUGCUUCAUGGUGAAAGACCU-3′.

Sequence of the small interfering RNA against epiregulin: 5′-IndexTermUGAAUGGCUAUUGUUUGCAUGGACA-3′.

Cell migration and invasion assays

In vitro chemo-migration and -invasion assays were performed as described (Hayashida et al., 2010).

Tumorigenicity in mice

All animals were cared for and experiments were performed in this facility under AAALAS guidelines using protocols approved by the institutional review board and the institutional animal care and use committee of the Massachusetts General Hospital. MCF10A cells expressing activated G12V H-RasV12-expressing lentivirus containing a GFP cassette and either shBTG2 or shLuc were generated using standard protocols. Xenografts were established by injecting 5 × 106 cells in 100 μl of phosphate-buffered saline into the mammary fat pad of 6-week-old female Swiss nu/nu mice. Each group consisted of 10 mice. Tumor volumes were measured at regular intervals, and volume was calculated as L × W2 (length=L and width=W). At the end of the experiment, lungs and livers of the animals were examined for the presence of metastasis.

To determine the effect of lapatinib on tumor growth, MCF10A cells expressing activated G12V H-Ras-expressing lentivirus containing a GFP cassette and either shBTG2 or shLuc were injected into the mammary fat pads of mice as described above. Each group contained 16 mice. After 2 weeks of tumor growth, 8 animals from each group were treated with 75 mg/kg lapatinib given by oral gavage once a day. Lapatinib was dissolved in 0.5% hydroxypropyl methyl cellulose and 0.1% Tween-80. The remaining eight animals from each group were treated with 0.5% hydroxypropyl methyl cellulose and 0.1% Tween-80 (Vehicle). Animals were continued to be treated for an additional 4 weeks. Tumors were measured once a week, and volumes calculated as length × width × width. At the end of the experiment, lungs and livers of the animals were examined for the presence of metastasis.

Oncomine microarray data sets

The Oncomine database was searched for data sets in which significant differences in BTG2 mRNA was present in breast cancers grouped according to clinical and prognostic parameters (gene rank top 1%; t-test P-value 0.0001). Several data sets were identified and a representative box plot from Oncomine (www.oncomine.org) is shown to illustrate the difference in BTG2 mRNA levels within the groups. All data are log transformed, median centered and the 25th–75th percentiles are indicated by the closed box.

References

  1. Bild AH, Yao G, Chang JT, Wang Q, Potti A, Chasse D et al. (2006). Oncogenic pathway signatures in human cancers as a guide to targeted therapies. Nature 439: 353–357.

  2. Boersma BJ, Reimers M, Yi M, Ludwig JA, Luke BT, Stephens RM et al. (2008). A stromal gene signature associated with inflammatory breast cancer. Int J Cancer 122: 1324–1332.

  3. Boiko AD, Porteous S, Razorenova OV, Krivokrysenko VI, Williams BR, Gudkov AV . (2006). A systematic search for downstream mediators of tumor suppressor function of p53 reveals a major role of BTG2 in suppression of Ras-induced transformation. Genes Dev 20: 236–252.

  4. Calzolari F, Appolloni I, Tutucci E, Caviglia S, Terrile M, Corte G et al. (2008). Tumor progression and oncogene addiction in a PDGF-B-induced model of gliomagenesis. Neoplasia 10: 1373–1382, following 1382.

  5. Cortes U, Moyret-Lalle C, Falette N, Duriez C, Ghissassi FE, Barnas C et al. (2000). BTG gene expression in the p53-dependent and -independent cellular response to DNA damage. Mol Carcinog 27: 57–64.

  6. Debnath J, Muthuswamy SK, Brugge JS . (2003). Morphogenesis and oncogenesis of MCF-10A mammary epithelial acini grown in three-dimensional basement membrane cultures. Methods 30: 256–268.

  7. Desmedt C, Piette F, Loi S, Wang Y, Lallemand F, Haibe-Kains B et al. (2007). Strong time dependence of the 76-gene prognostic signature for node-negative breast cancer patients in the TRANSBIG multicenter independent validation series. Clin Cancer Res 13: 3207–3214.

  8. Esteva FJ, Yu D, Hung MC, Hortobagyi GN . (2010). Molecular predictors of response to trastuzumab and lapatinib in breast cancer. Nat Rev Clin Oncol 7: 98–107.

  9. Farioli-Vecchioli S, Tanori M, Micheli L, Mancuso M, Leonardi L, Saran A et al. (2007). Inhibition of medulloblastoma tumorigenesis by the antiproliferative and pro-differentiative gene PC3. FASEB J 21: 2215–2225.

  10. Ficazzola MA, Fraiman M, Gitlin J, Woo K, Melamed J, Rubin MA et al. (2001). Antiproliferative B cell translocation gene 2 protein is down-regulated post-transcriptionally as an early event in prostate carcinogenesis. Carcinogenesis 22: 1271–1279.

  11. Finak G, Bertos N, Pepin F, Sadekova S, Souleimanova M, Zhao H et al. (2008). Stromal gene expression predicts clinical outcome in breast cancer. Nat Med 14: 518–527.

  12. Geyer CE, Forster J, Lindquist D, Chan S, Romieu CG, Pienkowski T et al. (2006). Lapatinib plus capecitabine for HER2-positive advanced breast cancer. N Engl J Med 355: 2733–2743.

  13. Gomez HL, Doval DC, Chavez MA, Ang PC, Aziz Z, Nag S et al. (2008). Efficacy and safety of lapatinib as first-line therapy for ErbB2-amplified locally advanced or metastatic breast cancer. J Clin Oncol 26: 2999–3005.

  14. Gril B, Palmieri D, Bronder JL, Herring JM, Vega-Valle E, Feigenbaum L et al. (2008). Effect of lapatinib on the outgrowth of metastatic breast cancer cells to the brain. J Natl Cancer Inst 100: 1092–1103.

  15. Guardavaccaro D, Corrente G, Covone F, Micheli L, D‘Agnano I, Starace G et al. (2000). Arrest of G(1)-S progression by the p53-inducible gene PC3 is Rb dependent and relies on the inhibition of cyclin D1 transcription. Mol Cell Biol 20: 1797–1815.

  16. Ha TU, Segev DL, Barbie D, Masiakos PT, Tran TT, Dombkowski D et al. (2000). Mullerian inhibiting substance inhibits ovarian cell growth through an Rb-independent mechanism. J Biol Chem 275: 37101–37109.

  17. Hayashida T, Takahashi F, Chiba N, Brachtel E, Takahashi M, Godin-Heymann N et al. (2010). HOXB9, a gene overexpressed in breast cancer, promotes tumorigenicity and lung metastasis. Proc Natl Acad Sci USA 107: 1100–1105.

  18. Hiscox S, Jiang WG, Obermeier K, Taylor K, Morgan L, Burmi R et al. (2006). Tamoxifen resistance in MCF7 cells promotes EMT-like behaviour and involves modulation of beta-catenin phosphorylation. Int J Cancer 118: 290–301.

  19. Hong JW, Ryu MS, Lim IK . (2005). Phosphorylation of serine 147 of tis21/BTG2/pc3 by p-Erk1/2 induces Pin-1 binding in cytoplasm and cell death. J Biol Chem 280: 21256–21263.

  20. Hsieh AC, Moasser MM . (2007). Targeting HER proteins in cancer therapy and the role of the non-target HER3. Br J Cancer 97: 453–457.

  21. Huang E, Cheng SH, Dressman H, Pittman J, Tsou MH, Horng CF et al. (2003). Gene expression predictors of breast cancer outcomes. Lancet 361: 1590–1596.

  22. Hynes NE, Horsch K, Olayioye MA, Badache A . (2001). The ErbB receptor tyrosine family as signal integrators. Endocr Relat Cancer 8: 151–159.

  23. Johnston S, Pippen Jr J, Pivot X, Lichinitser M, Sadeghi S, Dieras V et al. (2009). Lapatinib combined with letrozole versus letrozole and placebo as first-line therapy for postmenopausal hormone receptor-positive metastatic breast cancer. J Clin Oncol 27: 5538–5546.

  24. Johnston S, Trudeau M, Kaufman B, Boussen H, Blackwell K, LoRusso P et al. (2008). Phase II study of predictive biomarker profiles for response targeting human epidermal growth factor receptor 2 (HER-2) in advanced inflammatory breast cancer with lapatinib monotherapy. J Clin Oncol 26: 1066–1072.

  25. Julka PK, Chacko RT, Nag S, Parshad R, Nair A, Oh DS et al. (2008). A phase II study of sequential neoadjuvant gemcitabine plus doxorubicin followed by gemcitabine plus cisplatin in patients with operable breast cancer: prediction of response using molecular profiling. Br J Cancer 98: 1327–1335.

  26. Kawakubo H, Brachtel E, Hayashida T, Yeo G, Kish J, Muzikansky A et al. (2006). Loss of B-cell translocation gene-2 in estrogen receptor-positive breast carcinoma is associated with tumor grade and overexpression of cyclin d1 protein. Cancer Res 66: 7075–7082.

  27. Kawakubo H, Carey JL, Brachtel E, Gupta V, Green JE, Walden PD et al. (2004). Expression of the NF-kappaB-responsive gene BTG2 is aberrantly regulated in breast cancer. Oncogene 23: 8310–8319.

  28. Kedrin D, Wyckoff J, Boimel PJ, Coniglio SJ, Hynes NE, Arteaga CL et al. (2009). ERBB1 and ERBB2 have distinct functions in tumor cell invasion and intravasation. Clin Cancer Res 15: 3733–3739.

  29. Konecny GE, Venkatesan N, Yang G, Dering J, Ginther C, Finn R et al. (2008). Activity of lapatinib a novel HER2 and EGFR dual kinase inhibitor in human endometrial cancer cells. Br J Cancer 98: 1076–1084.

  30. Leary AF, Drury S, Detre S, Pancholi S, Lykkesfeldt AE, Martin LA et al. (2010). Lapatinib restores hormone sensitivity with differential effects on estrogen receptor signaling in cell models of human epidermal growth factor receptor 2-negative breast cancer with acquired endocrine resistance. Clin Cancer Res 16: 1486–1497.

  31. Lim IK, Lee MS, Lee SH, Kim NK, Jou I, Seo JS et al. (1995). Differential expression of TIS21 and TIS1 genes in the various organs of Balb/c mice, thymic carcinoma tissues and human cancer cell lines. J Cancer Res Clin Oncol 121: 279–284.

  32. Lim IK, Lee MS, Ryu MS, Park TJ, Fujiki H, Eguchi H et al. (1998). Induction of growth inhibition of 293 cells by downregulation of the cyclin E and cyclin-dependent kinase 4 proteins due to overexpression of TIS21. Mol Carcinog 23: 25–35.

  33. Lo HW, Hsu SC, Xia W, Cao X, Shih JY, Wei Y et al. (2007). Epidermal growth factor receptor cooperates with signal transducer and activator of transcription 3 to induce epithelial-mesenchymal transition in cancer cells via up-regulation of TWIST gene expression. Cancer Res 67: 9066–9076.

  34. Loi S, Haibe-Kains B, Desmedt C, Lallemand F, Tutt AM, Gillet C et al. (2007). Definition of clinically distinct molecular subtypes in estrogen receptor-positive breast carcinomas through genomic grade. J Clin Oncol 25: 1239–1246.

  35. Loi S, Haibe-Kains B, Desmedt C, Wirapati P, Lallemand F, Tutt AM et al. (2008). Predicting prognosis using molecular profiling in estrogen receptor-positive breast cancer treated with tamoxifen. BMC Genomics 9: 239.

  36. Ma XJ, Wang Z, Ryan PD, Isakoff SJ, Barmettler A, Fuller A et al. (2004). A two-gene expression ratio predicts clinical outcome in breast cancer patients treated with tamoxifen. Cancer Cell 5: 607–616.

  37. Mauxion F, Chen CY, Seraphin B, Shyu AB . (2009). BTG/TOB factors impact deadenylases. Trends Biochem Sci 34: 640–647.

  38. Mauxion F, Faux C, Seraphin B . (2008). The BTG2 protein is a general activator of mRNA deadenylation. EMBO J 27: 1039–1048.

  39. Melamed J, Kernizan S, Walden PD . (2002). Expression of B-cell translocation gene 2 protein in normal human tissues. Tissue Cell 34: 28–32.

  40. Minn AJ, Gupta GP, Siegel PM, Bos PD, Shu W, Giri DD et al. (2005). Genes that mediate breast cancer metastasis to lung. Nature 436: 518–524.

  41. Moffat J, Grueneberg DA, Yang X, Kim SY, Kloepfer AM, Hinkle G et al. (2006). A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 124: 1283–1298.

  42. Park S, Lee YJ, Lee HJ, Seki T, Hong KH, Park J et al. (2004). B-cell translocation gene 2 (Btg2) regulates vertebral patterning by modulating bone morphogenetic protein/smad signaling. Mol Cell Biol 24: 10256–10262.

  43. Pawitan Y, Bjohle J, Amler L, Borg AL, Egyhazi S, Hall P et al. (2005). Gene expression profiling spares early breast cancer patients from adjuvant therapy: derived and validated in two population-based cohorts. Breast Cancer Res 7: R953–R964.

  44. Prevot D, Voeltzel T, Birot AM, Morel AP, Rostan MC, Magaud JP et al. (2000). The leukemia-associated protein Btg1 and the p53-regulated protein Btg2 interact with the homeoprotein Hoxb9 and enhance its transcriptional activation. J Biol Chem 275: 147–153.

  45. Saal LH, Johansson P, Holm K, Gruvberger-Saal SK, She QB, Maurer M et al. (2007). Poor prognosis in carcinoma is associated with a gene expression signature of aberrant PTEN tumor suppressor pathway activity. Proc Natl Acad Sci USA 104: 7564–7569.

  46. Sabbah M, Emami S, Redeuilh G, Julien S, Prevost G, Zimber A et al. (2008). Molecular signature and therapeutic perspective of the epithelial-to-mesenchymal transitions in epithelial cancers. Drug Resist Updat 11: 123–151.

  47. Schmidt M, Bohm D, von Torne C, Steiner E, Puhl A, Pilch H et al. (2008). The humoral immune system has a key prognostic impact in node-negative breast cancer. Cancer Res 68: 5405–5413.

  48. Shelly M, Pinkas-Kramarski R, Guarino BC, Waterman H, Wang LM, Lyass L et al. (1998). Epiregulin is a potent pan-ErbB ligand that preferentially activates heterodimeric receptor complexes. J Biol Chem 273: 10496–10505.

  49. Sithanandam G, Fornwald LW, Fields J, Anderson LM . (2005). Inactivation of ErbB3 by siRNA promotes apoptosis and attenuates growth and invasiveness of human lung adenocarcinoma cell line A549. Oncogene 24: 1847–1859.

  50. Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A et al. (2003). Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci USA 100: 8418–8423.

  51. Sotiriou C, Neo SY, McShane LM, Korn EL, Long PM, Jazaeri A et al. (2003). Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci USA 100: 10393–10398.

  52. Sotiriou C, Wirapati P, Loi S, Harris A, Fox S, Smeds J et al. (2006). Gene expression profiling in breast cancer: understanding the molecular basis of histologic grade to improve prognosis. J Natl Cancer Inst 98: 262–272.

  53. Struckmann K, Schraml P, Simon R, Elmenhorst K, Mirlacher M, Kononen J et al. (2004). Impaired expression of the cell cycle regulator BTG2 is common in clear cell renal cell carcinoma. Cancer Res 64: 1632–1638.

  54. Tripathy D, Benz CC . (1992). Activated oncogenes and putative tumor suppressor genes involved in human breast cancers. Cancer Treat Res 63: 15–60.

  55. van ‘t Veer LJ, Dai H, van de Vijver MJ, He YD, Hart AA, Mao M et al. (2002). Gene expression profiling predicts clinical outcome of breast cancer. Nature 415: 530–536.

  56. van de Vijver MJ, He YD, van‘t Veer LJ, Dai H, Hart AA, Voskuil DW et al. (2002). A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 347: 1999–2009.

  57. Wang Y, Klijn JG, Zhang Y, Sieuwerts AM, Look MP, Yang F et al. (2005). Gene-expression profiles to predict distant metastasis of lymph-node-negative primary breast cancer. Lancet 365: 671–679.

  58. Winkler GS . (2009). The mammalian anti-proliferative BTG/Tob protein family. J Cell Physiol 222: 66–72.

  59. Xue C, Liang F, Mahmood R, Vuolo M, Wyckoff J, Qian H et al. (2006). ErbB3-dependent motility and intravasation in breast cancer metastasis. Cancer Res 66: 1418–1426.

  60. Yu K, Ganesan K, Miller LD, Tan P . (2006). A modular analysis of breast cancer reveals a novel low-grade molecular signature in estrogen receptor-positive tumors. Clin Cancer Res 12: 3288–3296.

  61. Zhang D, Sliwkowski MX, Mark M, Frantz G, Akita R, Sun Y et al. (1997). Neuregulin-3 (NRG3): a novel neural tissue-enriched protein that binds and activates ErbB4. Proc Natl Acad Sci USA 94: 9562–9567.

  62. Zhao H, Langerod A, Ji Y, Nowels KW, Nesland JM, Tibshirani R et al. (2004). Different gene expression patterns in invasive lobular and ductal carcinomas of the breast. Mol Biol Cell 15: 2523–2536.

Download references

Acknowledgements

We thank Drs Isselbacher, Haber, Engelman and Zou for critically reading the paper. This work was supported by the NIH/NCI Grant CA89138 (to SM), the Susan G Komen for the Cure Grants PDF0600282 and KG090412 (to SM) and ESSCO breast cancer research grant (to SM).

Author information

Correspondence to S Maheswaran.

Ethics declarations

Competing interests

The authors declare no conflict of interest.

Additional information

Supplementary Information accompanies the paper on the Oncogene website

Supplementary information

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • breast cancer
  • BTG2
  • invasion
  • metastasis
  • ErbB/HER pathway

Further reading