Introduction
The erbB receptor tyrosine kinase (RTK) family, including the epidermal growth factor receptor (EGFR), erbB2 (or Her2/neu), erbB3, and erbB4, is arguably the most important receptor family in the context of development and tumorigenesis [
1,
2]. Amplification and/or overexpression of e
rbB2 occur in approximately 25 to 30% of invasive breast cancers and are significantly associated with a worse prognosis in breast cancer patients [
3,
4]. Numerous studies indicate that increased treatment resistance and enhanced metastatic potential are two of the major mechanisms by which erbB2 contributes to breast carcinogenesis [
5,
6]. Most metastatic breast cancers show expression for either EGFR or erbB2, and less often for both [
7]. In contrast, co-expression of erbB2 and erbB3 frequently occurs in breast cancers [
8] and breast cancer cell lines [
9]. The erbB3 receptor is unique among the four erbB family members. Unlike EGFR, erbB2, and erbB4, it lacks kinase activity [
10,
11] or possesses weak kinase activity [
12]. However, erbB3 has been shown to serve as a critical co-receptor of erbB2, and its expression is a rate-limiting factor for erbB2-mediated breast cancer cell survival and proliferation [
13,
14]. We and others have also observed an elevated expression of the endogenous mouse erbB3 in the mammary tumors derived from
erbB2/
neu-transgenic mice [
15,
16]; and the increased erbB3 forms physical and functional interactions with the transgene-encoded erbB2 to promote mammary tumorigenesis [
15]. In fact, the erbB2/erbB3 heterodimer has been identified as the most potent form of all erbB receptor complexes to activate the oncogenic signaling, such as PI-3 K/protein kinase B (Akt), mitogen-activated protein kinase kinase (MEK)/mitogen-activated protein kinase (MAPK), and/or janus kinase (Jak)/signal transducer and activator of transcription (Stat) pathways, and/or Src kinase, in breast cancers [
17‐
19].
Mechanistic studies implicate the function of erbB3 as a major cause of treatment failure in human cancers [
20]. Therapeutic targeting of erbB3 is being investigated. Currently, no erbB3-targeted therapy has been approved for cancer treatment. Several erbB3 blocking antibodies (Abs) that prevent ligand-induced activation of erbB3, such as MM-121/SAR256212 (which is being co-developed by Merrimack Pharmaceuticals, Cambridge, MA, USA and Sanofi, Cambridge, MA, USA), MM-111 (Merrimack Pharmaceuticals) and U3-1287/AMG 888 (Amgen Inc., Thousand Oaks, CA, USA) are actively under preclinical and clinical studies [
21] and show significant antitumor activity in preclinical studies [
22‐
25]. MM-121 is a fully humanized anti-erbB3 monoclonal IgG2 Ab. It inhibits ligand-induced activation of erbB3, and exerts antitumor activity in human cancer models [
23,
24]. Nonetheless, MM-121’s therapeutic potential against erbB2-overexpressing breast cancers that are resistant to chemotherapy and/or the erbB2-targeted therapy has not been explored.
We have reported that activation of erbB3, mainly through PI-3 K/Akt signaling, plays a critical role in erbB2-mediated therapeutic resistance to tamoxifen [
26] and paclitaxel [
27]. The erbB3 receptor also interacts with both erbB2 and the insulin-like growth factor-1 receptor (IGF-1R) to form a heterotrimeric complex, which activates the PI-3 K/Akt signaling and Src kinase and subsequently results in resistance to erbB2-targeted therapy, trastuzumab (Herceptin) [
28]. Since activation of the PI-3 K/Akt signaling is the major determinant of treatment resistance [
29], we investigated,
in vitro and
in vivo, whether the anti-erbB3 Ab MM-121 would be able to overcome resistance and enhance the efficacy of chemotherapy or trastuzumab against erbB2-overexpressing breast cancer models via inhibition of the erbB3/PI-3 K/Akt signaling. In the current report, we sought to determine the antitumor activity of MM-121 in combination with paclitaxel against erbB2-overexpressing breast cancer using both
in vitro and
in vivo models. Our previous studies indicated that elevated expression of erbB3 led to paclitaxel resistance in erbB2-overexpressing breast cancer cells via PI-3 K/Akt signaling-dependent upregulation of Survivin [
27]. Thus, we have focused on studying whether inactivation of erbB3 signaling with MM-121 may specifically downregulate Survivin, and subsequently re-sensitize the otherwise resistant breast cancer cells to paclitaxel-mediated anti-proliferative/anti-survival effects and apoptosis.
Methods
Reagents and antibodies
MM-121 was from Merrimack Pharmaceuticals, Inc.. Paclitaxel (Ben Venue Labs, Inc., Bedford, OH, USA) was obtained from University of Colorado Hospital pharmacy. Antibodies used for western blots were as follows: erbB2 (EMD Chemicals, Inc., Gibbstown, NJ, USA); erbB3 and P-erbB2 (Tyr1248) (LabVision Corp., Fremont, CA, USA); P-erbB3 (Tyr1289), caspase-8 (1C12), and caspase-3 (8G10), P-MAPK (Thr202/Tyr204), MAPK, P-Akt (Ser473), Akt, Survivin (6E4), Bcl-xL (Cell Signaling Technology, Inc., Beverly, MA, USA); Mcl-1 (clone 22) (BD Biosciences, San Jose, CA, USA); poly(ADP-ribose) polymerase (PARP) (BIOMOL Research Laboratories Inc., Plymouth Meeting, PA, USA); and β-actin (Sigma-Aldrich Co., St. Louis, MO, USA). All other reagents were purchased from Sigma unless otherwise specified.
Cells and cell culture
Human breast cancer cell lines MCF-7, MDA-MB-231, SKBR3, and BT474 were obtained from the American Type Culture Collection (Manassas, VA, USA). The SKBR3.B3.1 and SKBR3.B3.2 cells are two
erbB3-transfected stable clones, and the SKBR3.neo1 is an empty vector-transfected clone of SKBR3 cells [
27]. The trastuzumab-resistant sublines BT474-HR20 and SKBR3-pool2, derived from BT474 and SKBR3, respectively, were described previously [
28]. All cell lines were maintained in DMEM/F-12 (1:1) (Sigma) containing 10% FBS (Sigma), and cultured in a 37°C humidified atmosphere containing 95% air and 5% CO
2 and split twice a week.
Cell proliferation assay
The CellTiter96 AQ nonradioactive cell proliferation kit (Thermo Fisher Scientific Inc., Waltham, MA, USA) was used to determine cell viability as previously described [
27,
28,
30]. Briefly, cells were plated onto 96-well plates for 24 h, and then grown in either DMEM/F12 medium with 0.5% FBS as control, or the same medium containing different concentrations of paclitaxel in the presence or absence of MM-121, and then incubated for another 72 h. After reading all wells at 490 nm with a microplate reader, the percentages of surviving cells from each group relative to controls, defined as 100% survival, were determined by reduction of 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS).
Specific knockdown of Survivin expression with a lentiviral system
Lentiviral production and specific knockdown of Survivin expression with an shRNA strategy were carried out as described previously [
27]. In brief, the lentivirus-containing either control shRNA or Survivin specific shRNA (Survivin-S3 or Survivin-S5) were produced in 293 T cells following the standard procedure. The virus in conditioned medium were harvested, aliquoted, and stored in a -80°C freezer. Prior to infection, the lentivirus-containing media were thawed completely at room temperature, and mixed with the same amount of fresh medium containing polybrene (8 μg/ml). The culture media of the candidate breast cancer cells were then replaced with the lentivirus-containing media. After 24 h, the virus-infected cells were selected with puromycin (1 μg/ml) for 48 h, and then subjected to required experiments.
Quantification of apoptosis
An apoptotic ELISA kit (Roche Diagnostics Corp., Indianapolis, IN, USA) was used to quantitatively measure cytoplasmic histone-associated DNA fragments (mononucleosomes and oligonucleosomes) as previously reported [
27,
28,
30]. This enzyme immunoassay was performed according to the manufacturer’s instructions.
Western blot analysis
Protein expression levels were determined by western blot analysis as described previously [
27,
28,
30,
31]. Equal amounts of total cell lysates were boiled in Laemmli SDS-sample buffer, resolved by SDS-PAGE, transferred to nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA, USA), and probed with the primary antibodies described in the figure legends. After the blots were incubated with horseradish peroxidase-labeled secondary antibody (Jackson ImmunoResearch Laboratories, West Grove, PA, USA), the signals were detected using the enhanced chemiluminescence reagents (GE Healthcare Bio-Sciences Corp., Piscataway, NJ, USA).
Immunohistochemistry (IHC)
Five-micron-thick paraffin sections were deparaffinized, antigens unmasked and immunohistochemically stained for Ki-67 (Thermo Fisher Scientific; rabbit monoclonal SP6; cat# RM-9106-SO; dilution 1:500 in Tris-buffered saline and Tween 20 (TBST) + 1% BSA w/v), cleaved caspase-3 (Cell Signaling Technology; rabbit polyclonal; cat#: 9661, 1:1,000 in TBST + 1% BSA w/v), Survivin (Epitomics, Burlingame, CA, USA; rabbit monoclonal EP2880Y; cat# 2463; dilution 1:100 in TBST + 1% BSA w/v), erbB2 (EMD Chemicals; mouse monoclonal 96G; cat#OP14T; dilution 1:500 in TBST + 1% BSA w/v), and erbB3 (Spring Bioscience, Pleasanton, CA, USA; rabbit monoclonal SP71; cat# M3710; dilution 1:200 in TBST + 1% BSA w/v). The specificity of all antibodies has been confirmed by both positive and negative controls. For erbB2 and erbB3, SKBR3 cells were used as a positive control. For Survivin, the endometrial cancer tissues originating from ovary were used a positive control. For Ki-67 and cleaved caspase-3, human tonsil tissues were used a positive control. For the negative controls, in addition to use the same cells/tissues without addition of the primary antibodies, we also performed the IHC assays using the tissues or cell lines that are known to have no expression of the antigens (Additional file
1).
Ki-67, cleaved caspase-3 and Survivin antigens were revealed in pH 9.5 BORG solution (Biocare Medical, Concord, CA) for 5 minutes at 125°C (22 psi; decloaking chamber, Biocare). ErbB2 required modest retrieval in 10 mmol/L sodium citrate for 5 minutes at 125°C in the decloaking chamber. ErbB3 required retrieval in Cell Conditioner 1 (standard retrieval time, Ventana Medical Systems, Tucson, AZ, USA). Immunodetection of Ki-67, cleaved caspase-3 and erbB2 was performed on the NexES stainer (Ventana) at an operating temperature of 37°C. Ki-67 and cleaved caspase-3 antibodies were incubated for 32 minutes and detected with a modified I-VIEW DAB detection kit (Ventana). The I-VIEW secondary antibody and enzyme were replaced with a species-specific secondary antibody (biotinylated goat anti-rabbit; 1:75; cat# 111-065-144; Jackson ImmunoResearch; 8 minutes) and streptavidin-horseradish (SA-HRP; 1:50; cat# SA-5004; DAKO Cytomation, Carpinteria, CA, USA; 8 minutes). Survivin was optimized under ambient conditions with the Rabbit ImmPress polymer detection system (Vector Laboratories, Burlingame, CA, USA; cat# MP-7401). ErbB2 was incubated for 32 minutes and detected with the standard I-VIEW detection. ErbB3 was incubated for 32 minutes and detected with a modified I-VIEW DAB kit in which the secondary antibody was replaced with Rabbit ImmPress (Vector Laboratories; cat# MP-7401; 8 minutes at 37°C) and enzyme was replaced with Rabbit ImmPress (diluted 1:1 in PBS, pH 7.6; 8 minutes at 37°C). Sections were sequentially blocked for 10 minutes in 3% hydrogen peroxide (v/v) and 30 minutes in Rodent Block M (Biocare; cat# RBM961), followed by primary antibody incubation for 30 minutes and 30 minutes in polymer. Antibody complexes were visualized with IP Flex DAB (Biocare; cat# IPK5010 G80; 4.5%). All sections were counterstained in Mayer’s hematoxylin for 2 minutes, nuclei blued in 1% ammonium hydroxide (v/v), dehydrated in graded alcohols, cleared in xylene and coverglass-mounted using synthetic resin.
Tumor xenograft model
Athymic nu/nu mice (Harlan Laboratories, Inc., Indianapolis, IN, USA) were maintained in accordance with the Institutional Animal Care and Use Committee (IACUC) procedures and guidelines approved by University of Colorado Health Sciences Center Animal Care and Use Committee. We suspended 8 × 106 BT474-HR20 cells in 100 μL of PBS mixed with 50% Matrigel (BD Biosciences) and injected these subcutaneously into the flanks of 5-week-old female mice. Tumor formation was assessed by palpation and measured with fine calipers three times a week. Tumor volume was calculated by the formula shown below, where length was the longest axis and width the measurement at a right angle to the length:
Volume = (Length × Width2)/2
This was followed by statistical analysis as we described previously [
32]. When tumors reached approximately 150 mm
3 or 100 mm
3, mice were randomly assigned into four groups (n = 5): 1) control-group mice received intraperitoneal (i.p.) injection of 100 μl PBS; 2) mice received i.p. injection of paclitaxel (15 mg/kg or 7.5 mg/kg) in 100 μl PBS twice a week; 3) mice received i.p. injection of MM-121 (10 mg/kg) in 100 μl PBS twice a week, or 4) mice received i.p. injection of paclitaxel (15 mg/kg or 7.5 mg/kg) and MM-121 (10 mg/kg) in 100 μl PBS twice a week. The animals’ health status was monitored daily for weight loss or for signs of altered motor ability while in their cages. At the end of study, mice were euthanized according to the approved IACUC protocol. Tumors from all animals were excised and embedded in paraffin for IHC analyses.
Statistics
Statistical analyses of the experimental data were performed using either the two-sided t-test or analysis of variance (ANOVA) for each time point followed by post-hoc testing between groups. Significance was set at a P-value <0.05. All statistical analyses were conducted with the software StatView v5.1 from SAS Institute Inc., Cary, NC, USA.
Discussion
The erbB3 receptor has long been considered as an inactive
pseudokinase[
10,
11], although a recent study suggests that erbB3 has weak kinase activity that can trans-autophosphorylate its intracellular region [
12]. In order to fully transduce cell signaling, however, erbB3 has to be phosphorylated by its interactive partners; of these, erbB2 is the most important one [
33]. It has been well-documented that activation of the erbB3 signaling plays a pivotal role in the development of erbB2-overexpressing breast cancer [
13,
14]. Although a number of strategies targeting erbB2 are being used in the clinic [
34], no erbB3-targeted therapy has been approved for cancer treatment. MM-121 is an erbB3-blocking Ab that is being actively investigated in clinical trials of cancer patients with solid tumors, such as advanced non-small cell lung cancer, colorectal cancer, squamous cell head and neck cancer, and platinum resistant/refractory ovarian cancer [
35]. In breast cancer, the therapeutic potential of MM-121 is being tested in patients with estrogen receptor- and/or progesterone receptor-positive and erbB2-negative breast cancers in combination with the aromatase inhibitor exemestane, and in patients with triple-negative or erbB2-negative breast cancers in combination with paclitaxel. To date, no study has been initiated to test the activity of MM-121 in breast cancer patients with erbB2-overexpressing tumors, including those resistant to paclitaxel. Here we provide experimental evidence indicating that MM-121 inactivates erbB3 (Figures
1B and
4A), significantly enhances paclitaxel-induced anti-proliferative/anti-survival effects (Figures
1A,
4B,
5A, and
6), and apoptosis (Figures
2B,
2C,
4C,
4D, and
6) in erbB2-overexpressing breast cancer cells with either medium (SKBR3, SKBR3.neo1, and BT474-HR20) or high erbB3 (SKBR3.B3.1 and SKBR3.B3.2) expression. Our data suggest that MM-121 is efficacious in all erbB2-overexpressing breast cancer cell lines tested. We believe that MM-121 will exhibit therapeutic potential in all erbB2-positive breast cancer patients (not only those with strong erbB3-expressing tumors) as long as the tumors show active erbB3 signaling.
The concentrations of paclitaxel used in the
in vitro studies (0 to 16 nmol/L or 0 to 32 nmol/L) are much lower than the average steady-state plasma concentrations of paclitaxel (300 to 800 nmol/L) found in patients [
36,
37]. This difference may be explained by the two very distinct systems. In the 2-dimensional cell culture condition, paclitaxel should easily get into the tumor cells which grow as a monolayer. However, due to the complexity of the human body, some substances in the circulation may attenuate the efficacy of paclitaxel. The presence of fibroblast, microphage, immune cells and others in the tumor mass may block or decrease the drug’s entry into the tumor cells. In addition, because of the heterogeneicity of tumors and their 3-dimensional architecture, drug uptake varies in tumor cells. Thus, the steady-state plasma concentration of paclitaxel may not reflect the drug concentration inside the tumor cells. The data obtained from our animal studies provide further support of this point. It has been reported that administration of 10 mg/kg paclitaxel to mice gives rise to a peak plasma concentration of approximately 3 μmol/L, which gradually and near-linearly declines to 0 at 16 h [
38]. This peak concentration of paclitaxel is much higher than that we used in our cell culture studies. However, in the
in vivo studies, although administration of 15 mg/kg paclitaxel to mice had significant inhibitory effects on tumor growth (Additional file
4), we found no such effect when paclitaxel was used at 7.5 mg/kg (Figure
5A). Thus, both our
in vitro and
in vivo data seem to indicate that the ability of MM-121 to enhance the therapeutic efficacy of paclitaxel against erbB2-overexpressing breast cancer may be restricted to the ineffective doses of paclitaxel. Although a recent report shows that as a single agent, neoadjuvant treatment with paclitaxel induces tumor response in 92% of patients with erbB2-overexpressing breast cancer [
39], our findings suggest that the presence of MM-121 may convert the tumors from non-responsive to responsive to the lower doses of paclitaxel with less side effects, and therefore merit translational implications in the clinic.
We have reported that the underlying mechanism that elevated expression of erbB3 results in paclitaxel resistance in erbB2-overexpressing breast cancer cells is attributed to PI-3 K/Akt-dependent upregulation of Survivin [
27]. Numerous studies indicate that Survivin functions as a critical inhibitor of apoptosis to promote cell survival and proliferation, and regulates mitosis during cell cycle progression [
40]. Survivin is selectively expressed in a variety of human malignancies and its overexpression positively correlates with poor prognosis, tumor recurrence and therapeutic resistance [
40]. A number of different strategies targeting Survivin, including antisense oligonucleotide and pharmacological inhibitors have been developed and are currently under clinical trials for cancer treatment [
40‐
42]. Our data showed that inactivation of erbB3 signaling with MM-121 specifically downregulated Survivin in our
in vitro models, and significantly enhanced paclitaxel-induced cytotoxicity and apoptosis in the otherwise resistant breast cancer cells (Figures
1 and
2). In our mouse model, although treatment with MM-121 alone had no significant effects on Survivin expression, MM-121 did dramatically downregulate Survivin when combined with paclitaxel (Figure
6). It is possible that MM-121 at the dose we used only partially inhibits the PI-3 K/Akt signaling
in vivo, and the inactivation of Akt to such an extent alone could not significantly alter the expression of Survivin. Alternatively, other signaling pathways may also be critical to control Survivin expression in the
in vivo circumstance. A marked reduction of Survivin was only observed with the combinatorial treatment. In addition, we have found that hMP-RM-1, a humanized version of the anti-erbB3 Ab MP-RM-1 [
43], exhibited similar
in vitro activity to specifically downregulate Survivin in erbB2-overexpressing breast cancer cells (data not shown). Thus, inhibition of erbB3 with blocking Ab may be a novel strategy to target Survivin for cancer treatment. As MP-RM-1 inhibits both ligand-dependent and -independent activation of erbB3 [
43], we speculate that hMP-RM-1 might exert more potent antitumor activity than MM-121 against erbB2-overexpressing breast cancer. More detailed studies are needed to confirm our hypothesis.
Elevated expression of Survivin was also observed in the trastuzumab-resistant subline BT474-HR20 and causally led the cells resistant to paclitaxel-induced apoptosis (Additional files
3 and
4). However, we did not find Survivin upregulation in another trastuzumab-resistant subline SKBR3-pool2. These data suggest that elevated expression of Survivin may cause cross-resistance to paclitaxel treatment in some trastuzumab-resistant breast cancers. We believe that profound activation of PI-3 K/Akt signaling, which we observed in BT474-HR20, but not in SKBR3-pool2 cells [
28], is the major mechanism contributing to the upregulation of Survivin. At this moment, it remains unclear whether activation of the PI-3 K/Akt signaling by any means, such as
PIK3CA mutation or phosphatase and tensin homolog (PTEN) deletion in addition to erbB2/erbB3 receptors, would also enhance expression of Survivin in breast cancer cells. The interesting phenomenon of Survivin-mediated cross-resistance to paclitaxel and trastuzumab warrants further investigation.
Activation of the PI-3 K/Akt signaling has been identified as the major determinant of trastuzumab resistance [
29]. In addition to inducing paclitaxel resistance, the erbB2/erbB3/PI-3 K/Akt signaling also results in resistance to hormonal therapy [
44] and other chemotherapy [
45] in breast cancer treatment. Because MM-121 mainly inhibits activation of erbB3 and Akt in erbB2-overexpressing breast cancer cells, it is conceivable to hypothesize that MM-121 may abrogate erbB3 signaling-mediated therapeutic resistance to tamoxifen, trastuzumab, and other chemotherapeutic agents, such as doxorubicin. By taking advantage of the trastuzumab-resistant breast cancer model (BT474-HR20 and SKBR3-pool2), we have discovered that MM-121 is able to overcome trastuzumab resistance and significantly enhance trastuzumab-induced growth inhibition and/or apoptosis
in vitro and
in vivo (submitted separately).
Competing interests
The authors declare that they have no competing interests. MM-121 was kindly provided by Merrimack Pharmaceuticals, Inc. (Cambridge, MA, USA).
Authors’ contributions
The authors’ contributions to this research work are reflected in the order shown, with the exception of BL who supervised the research and finalized the report. SW, JH, BC, and HL carried out all the in vitro experiments and most of the in vivo experiments. SW and BL drafted the manuscript. XY and FL participated in the animal studies. JH, BC, and SE reviewed the IHC slides and counted the positive staining. JT, AT, CKL, and BL conceived of the study, and participated in its design and coordination. All authors read and approved the final manuscript.