A novel interaction between HER2/neu and cyclin E in breast cancer

Abstract

HER2/neu (HER2) and cyclin E are important prognostic indicators in breast cancer. As both are involved in cell cycle regulation we analyzed whether there was a direct interaction between the two. HER2 and cyclin E expression levels were determined in 395 breast cancer patients. Patients with HER2-overexpression and high levels of cyclin E had decreased 5-year disease-specific survival compared with low levels of cyclin E (14% versus 89%, P<0.0001). In vitro studies were performed in which HER2-mediated activity in HER2-overexpressing breast cancer cell lines was downregulated by transfection with HER2 small interfering RNA or treatment with trastuzumab. Cyclin E expression levels were determined by western blot analysis, and functional effects analyzed using kinase assays, MTT assays were used to assess cell viability as a marker of proliferation and fluorescence-activated cell sorting analysis was used to determine cell cycle profiles. Decreased HER2-mediated signaling resulted in decreased expression of cyclin E, particularly the low molecular weight (LMW) isoforms. Decreased HER2 and LMW cyclin E expression had functional consequences, including decreased cyclin E-associated kinase activity and decreased proliferation, because of increased apoptosis and an increased accumulation of cells in the G1 phase. In vivo studies performed in a HER2-overexpressing breast cancer xenograft model confirmed the effects of trastuzumab on cyclin E expression. Given the relationship between HER2 and cyclin E, in vitro clonogenic assays were performed to assess combination therapy targeting both proteins. Isobologram analysis showed a synergistic interaction between the two agents (trastuzumab targeting HER2 and roscovitine targeting cyclin E). Taken together, these studies show that HER2-mediated signaling effects LMW cyclin E expression, which in turn deregulates the cell cycle. LMW cyclin E has prognostic and predictive roles in HER2-overexpressing breast cancer, warranting further study of its potential as a therapeutic target.

Conflict of interest statement

Conflict of Interest

Dr. L. Meijer is a coinventor on the patent on roscovitine licensed to Cyclacel.

Figures

Figure 1

Relationship between HER2 and cyclin E in breast cancer. Tumors from 395 patients (Keyomarsi et al., 2002) were assessed for HER2 and cyclin E levels by western blot. A) Disease-specific survival (DSS) was higher in patients with HER2-negative (n = 262) tumors (P < .0001). B) HER2-overexpresing tumors were stratified by total cyclin E expression. Patients with high total cyclin E had markedly decreased DSS (median DSS, 2 years) (P< .0001). C) Stratification by LMW cyclin E demonstrates decreased median DSS in patients with high LMW cyclin E (median DSS, 2 years) (P < .0001). (Total cyclin E = full-length + LMW cyclin E).

Figure 2

Effect of HER2 downregulation on cyclin E expression. A) HER2-overexpressing MCF-7-HER-18 and SKBr3 breast cancer cells were grown exponentially (MCF-7 cells were used as a HER2 negative control). Lysates from these cells were used to perform western blot analysis probing for HER2 and cyclin E. Western blots were performed using a 7% (HER2) or 10% (cyclin E) SDS-PAGE gel. For a given experiment, gels for HER2 and cyclin E were run using lysates that had been prepared as a single sample that was then aliquoted to the two gels that were run concurrently. B) HER2-overexpressing MCF-7-HER-18 cells were transfected with increasing concentrations of HER2 siRNA. Cells were harvested 48 hours after transfection and western blots performed on lysates obtained from these cells demonstrated decreased HER2 expression which was quantitated using densitometry. Mock transfected (control) and cells transfected with a random sequence, negative silencer siRNA were used as controls. C) After optimization of HER2 siRNA transfection conditions, HER2-overexpressing MCF-7-HER-18 cells were again transfected with HER2 siRNA. Lysates harvested from these cells were used to perform western blot analysis probing for HER2 and cyclin E. Western blot demonstrated a decrease in HER2 expression with a concomitant decrease in cyclin E expression, primarily a decrease in the LMW forms. MCF-7 was used as a HER2-negative control. D) Confocal immunofluorescence microscopy demonstrated knockdown of HER2 (green) in cells transfected with HER2 siRNA, compared to mock transfected or random sequence siRNA transfected controls. (TO-PRO-3-iodide (blue) used for nuclear staining). E) Confocal immunofluorescence microscopy using C-19 antibody against cyclin E (red) showed decreased total cyclin E expression in HER2-overexpressing cells that had decreased HER2 expression after transfection with HER2 siRNA. F) Western blot confirmed HER2 knockdown after transfection with two different HER2 siRNA in MCF-7-HER18 and SKBr3 breast cancer cell lines. Densitometry was performed to quantitate HER2 expression. Western blot with HE-12 antibody against cyclin E revealed decreased LMW cyclin E in HER2 siRNA transfected cells compared to controls confirming the differential effect on full-length versus LMW cyclin E.

Figure 3

Effect of HER2 downregulation on cyclin E-associated kinase activity and cell cycle profiles. A) HER2-overexpressing cells were transfected with HER2 siRNA, random sequence siRNA or mock transfected. Protein lysates (250 µg) were immunoprecipitated with anti-cyclin E antibody and protein G-sepharose beads using Histone H1 as a substrate. Bands corresponding to Histone H1 phosphorylation were quantitated through phosphoimaging. Cells transfected with HER2 siRNA had decreased cyclin E-associated kinase activity compared to controls. This was more pronounced in SKBr3 cells which have lower levels of LMW cyclin E than MCF-7-HER-18 cells. B) Cell viability was assessed with MTT assay. Cells were transfected with HER2 siRNA, random sequence siRNA or mock transfected. Following transfection with HER2 siRNA the number of tumor cells was decreased compared to control cells, suggesting that HER2 knockdown resulted in decreased proliferation. Error bars represent standard deviation. C) Propopidium iodine staining was performed to determine the percentage of cells undergoing apoptosis following transfection with HER2 siRNA. Experiments were repeated in triplicate, and mean percentage of cells in the sub-G1 phase for mock transfected controls versus siRNA-transfected MCF-7-HER-18 and SKBr3 cells is shown. Error bars represent the standard error of the mean. There was a significant increase in the percentage of cells in sub-G1 phase after transfection with HER2 siRNA consistent with HER2 knockdown causing an increase in apoptosis. D) . Cell cycle profiles were determined using FACS analysis. Experiments were repeated in triplicate, and the mean percentage of cells in the G1 phase for mock transfected controls versus siRNA-transfected MCF-7-HER-18 and SKBr3 cells is shown. Error bars represent the standard error of the mean. There was a significant increase in the percentage of MCF-7-HER-18 cells in G1 phase after transfection with both HER2 siRNA sequences. For SKBr3 cells, there was a significant increase in the percentage of cells in G1 phase after transfection with HER2 siRNA #1.

Figure 4

Effect of decreased HER2-mediated cell signaling after trastuzumab treatment. A) HER2-overexpressing cells were treated with trastuzumab for 48 hours. Confocal immunofluorescence microscopy revealed decreased total cyclin E (red) expression as quantified by Image-Pro Plus Software (TO-PRO-3-iodide (blue) used for nuclear staining). Error bars represent the standard deviation. B) Western blot performed following trastuzumab treatment showed decreased LMW cyclin E and cyclin D1 expression in trastuzumab-treated cells. There was no difference in p21 or p27 expression. Western blots were performed using separate 10% SDS-PAGE gels for each antibody. All gels were run concurrently using lysates that were prepared as a single sample and aliquoted to the individual gels. C) Cyclin E-associated kinase activity was decreased in trastuzumab-treated cells compared to controls. This was more pronounced in SKBr3 cells (MCF-7 used as HER2 negative control). D) Following trastuzumab treatment for 48 hours, confocal immunofluorescence microscopy revealed increased nuclear localization of both p21 (red, left) and p27 (red, right) (TO-PRO-3-iodide (blue) used for nuclear staining).

Figure 5

Synergistic effect of trastuzumab and roscovitine in breast cancer cell lines overexpressing HER2 and cyclin E. A) High throughput clonogenic assays were used to compare the cytotoxic effects of trastuzumab alone, roscovitine alone, and the combination in SKBr3 and BT474 breast cancer cells (X-axis: roscovitine µM; Y-axis: trastuzumab µg/ml; Z-axis: fraction non-viable cells). B) Isobologram analysis showed a synergistic interaction between the two agents. Isobologram analysis and graphs were obtained using CalcuSyn software, which performs drug dose-effect calculation using the median effect method. Experiments were performed in triplicate and representative data are shown.

Figure 6

In vivo effects of trastuzumab therapy. MCF-7-HER-18 cells were injected into the mammary fat pads of nude mice. When tumors reached 100 mm3, intraperitoneal injections of trastuzumab or PBS were given twice weekly for 3 weeks. A) Tumors from trastuzumab-treated mice showed decreased p-HER2 staining (2 versus 1) and a concomitant decrease in cyclin E expression (4 versus 3). B) Lysates were obtained from tumors from 13 mice (6 treated with PBS and 7 treated with trastuzumab). Lane 7 contains lysate that was likely subcutaneous tissue, not tumor, as evidenced by the low actin expression. For the remaining tumors, western blot for cyclin E expression revealed that those obtained from trastuzumab-treated mice demonstrated decreased expression of cyclin E, particularly the LMW forms when compared to PBS-treated control mice.