Background
According to global statistics, breast cancer (BC) is the most common cancer in the female population [
1,
2]. BC is a highly heterogeneous disease, which has been classified into distinct subtypes according to clinicopathologic features and molecular profile [
3,
4]. The molecular taxonomy is mandatory for prognosis, diagnosis, and prediction of therapy response [
5]. Estrogen receptor-
α (ER-
α), progesterone receptor (PR), and epidermal growth factor receptor 2 (ERBB2, HER-2/neu) are the most commonly used biomarkers for BC, and their status determines the main BC subtypes.
ERBB2-positive (ERBB2+) breast carcinomas comprise around 20% of BC cases, are defined by its amplification or overexpression, and indicate poor clinical prognosis [
5]. ERBB2 is a member of the ERBB receptor tyrosine kinase (RTK) family, which consists of four members including EGFR, ERBB3, and ERBB4 [
6]. Although ERBB2 lacks a ligand-binding domain, it forms homodimers and heterodimers, exhibiting robust signaling activity in ERBB2 (+) BC subset [
7]. Over the past decades, powerful ERBB2-targeted agents have been developed. Trastuzumab was the first recombinant antibody against ERBB2 approved for the treatment of ERBB2 (+) BC [
8]. Although trastuzumab is remarkably efficient, the percentage of relapsed breast cancer patients is quite worrying, highlighting the need of exploring alternative therapeutic targets in this disease [
9]. The developed resistance relies on various mechanisms. Trastuzumab-based combinations are a strategy aiming to overcome the resistance and increase treatment efficiency.
The receptor activator of nuclear factor (NF)-κB, RANK, and its ligand, RANKL, have been recently implicated in BC initiation, progression, and metastasis [
10‐
13]. Particularly, the RANKL/RANK axis has been reported to play a crucial role in progesterone-driven mammary tumorigenesis [
10,
13]. The RANK pathway has been also associated with tumor initiation in
BRCA1-mutation carriers, introducing RANK (+) luminal progenitor cells as the main target population in this BC subtype [
14]. Ithimakin et al. demonstrated that RANKL stimulation increased ERBB2 expression in luminal breast cancer stem cells (CSCs), regulating their self-renewal [
15]. Probably RANK-mediated IKKα [inhibitor of NF-κB (IκB) kinase subunit α]-NF-κB activation is associated with this regulation [
16] and the increase of ERBB2 expression, which further activates NF-κB, generating a positive feedback loop [
17].
Interestingly, scientific data have implicated the RANK/RANKL pathway in ERBB2 (+) BC tumorigenesis. RANK signaling, probably through IKKα, has been reported to enhance spontaneous mammary tumorigenesis and metastatic potential in ERBB2-overexpressed BC models [
10,
11]. Additionally, a selective pharmacological RANKL inhibitor could decrease tumor development and metastasis in ERBB2-positive context [
10,
11]. Denosumab, a monoclonal antibody against RANKL approved for the treatment of osteoporosis and skeletal-related events [
18], has been proposed to partially block the expansion of CSCs occurring in response to induced ERBB2 expression [
15].
In this study, we examined the consistency and validity of our proposed theoretical model regarding the interplay between ERBB2 and RANK pathways in BC [
19]. In this vein, we provide evidence, for the first time, of a physical interaction between RANK receptor and ERBB family members in BC cells. We show that RANK/ERBB2 dimer formation is related to ERBB2 expression and is disturbed by the presence of RANKL and ERBB2 inhibitors in ERBB2 (+) BC cells. The dimer disruption is accompanied by weakened NF-κB signaling and lower rates of cell proliferation and metastatic potential.
Methods
Antibodies and reagents
The following antibodies were used for Western blot (WB), immunohistochemistry (IHC), immunofluorescence (IF), and proximity ligation assay (PLA, Duolink). For WB: RANK antibody (H-300) (sc-9072), RANKL antibody (N-19) (sc-7628), IKKα antibody (3G12) (#11930), phospho-IKKα (phospho S176 + S180) antibody (ab17943), NF-κB p65 antibody (F-6) (sc-8008), phospho-NF-κB p65 (Ser536) antibody (93H1) (#3033), IκBα antibody (Η-4) (sc-1643), phospho-IκBα antibody (B-9) (sc-8404), anti-actin antibody, clone C4 MAB1501 (Millipore, MA). For IHC: RANK antibody (H-300) (sc-9072). For IF and PLA: RANK antibody (H-300) (sc-9072), RANKL antibody (N-19) (sc-7628), EGFR antibody (ab30), Neu antibody (3B5) (sc-33684), HER3 antibody (2F9) (ab91084), HER4 antibody (L20) (sc-31149). The secondary antibodies were employed: WB: goat anti-mouse IgG, HRPconjugate (12–349, Millipore), goat anti-rabbit IgG, HRPconjugate (12–348, Millipore). For IF: CF 543 Donkey Anti-Rabbit IgG (H + L), Highly Cross-Adsorbed (#20308, Biotinum). The following reagents were used in this study: recombinant human RANKL (390-TN-010, R&D Systems) dissolved in 0.1% BSA in PBS at 50 ng/ml, trastuzumab (Herceptin 150 mg, Roche) dissolved in a medium at 100 μg/ml, pertuzumab (Perjeta 420 mg, Roche) dissolved in a medium at 100 μg/ml, and denosumab (XGEVA 120 mg, AMGEN) dissolved in a medium at 100 μg/ml. The DUOLINK In Situ Ligation Kit was purchased from Sigma-Aldrich (DUO92002-Duolink In Situ PLA Probe Anti-Rabbit PLUS, DUO92004—Duolink In Situ PLA Probe Anti-Mouse MINUS, DUO92006—Duolink In Situ PLA Probe Anti-Goat MINUS, and DUO92007—Duolink In Situ Detection Reagents Orange). Nuclei were stained with NucBlue™ Fixed Cell ReadyProbes™ Reagent (R37606, Thermo Fisher Scientific), and coverslips were mounted with ProLong™ Diamond Antifade Mountant (P36965, Thermo Fisher Scientific).
Cell culture and cell culture
SKBR3, MCF7, BT-474, and MDA-MB-453 cell lines are commonly used in breast cancer research. The cell lines were authenticated at the Laboratory of Genetics of the Biomedical Research Foundation of the Academy of Athens (Athens, Greece) by inverted DAPI banding karyotyping method. SKBR3 and BT-474 are characterized by ERBB2 overexpression. MCF7 and BT-474 are ER positive, and SKBR3 and MDA-MB-453 ER negative. The MCF10A human mammary epithelial cell line was used as normal cells. SKBR3 and MCF7 were cultured in DMEM, l-glutamine (Gibco, Life Technologies) and BT-474 and MDA-MB-453 in RPMI 1640 medium GlutaMAX (Gibco, Life Technologies), supplemented with 10% fetal bovine serum, FBS (Gibco, Life Technologies) and 1% penicillin-streptomycin. MCF10A cells were cultured in DMEM/F-12 (Gibco, Life Technologies) supplemented with 5% horse serum, 100 ng/ml cholera toxin, 20 ng/ml epidermal growth factor (EGF), 0.01 mg/ml insulin, 500 ng/ml hydrocortisone, and 1% penicillin-streptomycin. Cell cultures were incubated at 37 °C in a humidified atmosphere containing 5% CO2–95% air.
Reverse transcription (RT)-PCR, semi-quantitative PCR
Total RNA was extracted from cultured cells using the RNeasy Mini Kit (Qiagen, CA, USA) according to the manufacturer’s instructions. cDNA was synthesized using the PrimeScript RT reagent kit-Perfect Real Time (Takara Bio, Japan) according to the manufacturer’s protocol.
For semi-quantitative PCR, the produced cDNA was amplified with specific primer pairs for—RANK-encoding—TNFRSF11A (annealing 60 °C, forward CCCGTTGCAGCTCAACAAG, reverse GCATTTGTCCGTGGAGGAA) and—RANKL-encoding—TNFSF11 (annealing 60 °C, forward AGCAGAGAAAGCGATGGT, reverse GGGTATGAGAACTTGGGATT) genes (38 cycles) as well as with actin gene primer pairs (28 cycles) using KAPA 2G Multiplex Mastermix (KK5801, Sigma-Aldrich) according to the manufacturer’s instructions. PCR-amplified fragments were analyzed after their separation in agarose gels using image analysis software (ImageJ; La Jolla, CA) and normalized to actin gene levels.
Western blot analysis
Protein extraction was performed using ice-cold RIPA buffer (Thermo Fisher Scientific). Bradford assay (Bio-Rad) was used to assess protein concentration in the extracts. Proteins were resolved by electrophoresis in SDS–polyacrylamide gels with several densities (10%, 12%, and 15%) depending on the molecular weight of each protein. Subsequently, they were transferred to a nitrocellulose membrane (Macherey–Nagel, Germany). Membranes were blocked for 1 h at room temperature in Tris-buffered saline with Tween-20 (TBS-T) with 5% nonfat milk. Then, membranes were incubated with primary antibodies overnight at 4OC (dilutions were 1:250 for antibodies against RANKL, IκBα, and p-IκBα; 1:500 for antibodies against p65 and RANK; 1:1000 for antibodies against IKKα, p-IKKα, and p-p65; and 1:2000 for antibody against actin). After incubation with HRP-conjugated secondary antibodies, the detection of the immunoreactive bands was performed with the Clarity Western ECL Substrate (Bio-Rad). Relative protein amounts were evaluated by a densitometry analysis using ImageJ software (La Jolla, CA, USA) and normalized to the corresponding actin levels.
Cell proliferation assay
The assessment of breast cancer cell proliferation was performed with the XTT Cell Proliferation Assay Kit (10010200, Cayman Chemical, USA). Cells were seeded in a 96-well plate at a density of 103–105 cells/well in a culture medium. Cells were starved in phenol red-free medium supplemented with 5% charcoal stripped serum (CSS) for 24 h prior the treatments. Then, cells were cultured in a 100-μl starvation medium with or without the tested compounds in a CO2 incubator at 37 °C for variable time points. Afterwards, 10 μl of XTT Mixture was added to each well and mixed gently for 1 min on an orbital shaker. The cells were incubated for 2 h at 37 °C in a CO2 incubator. The absorbance of each sample was measured using a microplate reader at 450 nm.
Migration assay
Breast cancer cells were seeded in 6-well plates and maintained in a CO2 incubator at 37 °C. The seeding density was adjusted appropriately for each cell line in order to form a confluent monolayer. The cell monolayer was scratched in a straight line with a sterile 200-μl pipet tip. The debris was removed by washing the cells once with PBS, and then it was replaced with a medium containing the tested compounds. The plates were placed under a phase-contrast, computer-assisted microscope, and the first image of the scratch was photographed at × 10 magnification. Reference points were made. The plates were placed in an incubator for 24 and 48 h. After completion of the incubation, plates were placed under a microscope, having reference points to align the photographed region, and images of the scratch were acquired. Images for each sample at 0, 24, and 48 h were analyzed quantitatively by using the TScratch software (Wimasis image analysis platform).
Clonogenic assay
Breast cancer cells were seeded in 6-well plates, at an appropriate seeding density (~ 103 cells/well). Cells were allowed to attach to the wells and then were treated. Plates were placed in a CO2 incubator at 37 °C for 10–15 days, until control cells formed sufficiently large colonies. Cells were then fixed with a solution containing 1 acetic acid:7 methanol and stained with 0.5% crystal violet in methanol for 15 min. Plates were carefully immersed in a tank with tap water and left to dry. Then, they were scanned, and the relative capacity to produce colonies was evaluated by a densitometry analysis using ImageJ software (La Jolla, CA, USA).
Immunohistochemistry
The study included 20 archival BC tissue samples provided by the Department of Pathology, Medical School, National and Kapodistrian University of Athens. The tissue samples had been already evaluated for ERBB2 and ER expression. Immunohistochemistry was performed on FFPE sections cut at 5-μm thickness. Tissues were dried at 65 °C for 20 min, deparaffinized in xylene, and then rehydrated in an ethanol series. After washing with distilled H2O (dH2O), tissue sections were microwave heated with 10 mM citrate buffer (pH 6.0) for antigen retrieval for 25 min. To remove the endogenous peroxidase activity, sections were then treated with freshly prepared 3% hydrogen peroxide in methanol in the dark, for 10 min at room temperature. Non-specific antibody binding was blocked using 5% normal goat serum (NGS) for 1 h Sections were incubated with the primary antibody overnight at 4 °C (dilution was 1:100 for the antibody against RANK). Sections were then incubated at room temperature with biotinylated linking reagent (#20775, Merck, Darmstadt, Germany) for 10 min, followed by incubation with peroxidase-conjugated streptavidin label (#20774, Merck, Darmstadt, Germany) for 10 min. The sections were developed with diaminobenzidine and counterstained with hematoxylin (Sigma-Aldrich). Sections were dehydrated in ethanol buffers of 70, 80, 96, and 100% concentration and were mounted onto glass coverslips. PBS was used as the negative control instead of the primary antibody. Immunostaining was evaluated by the pathologist ST.
Proximity ligation assay (PLA) and immunofluorescence (IF)
Duolink® Proximity Ligation Assay (PLA) allows in situ detection of protein-protein interactions. The PLA sensitivity is based on the fact that amplified signal is generated only when proteins of interest are in close proximity (< 40 nm). Breast cancer cells were cultured on 12-well chamber slides (81,201, ibidi). After reaching ~ 70% confluency, cells were starved for 24 h and treated accordingly for 1 h. Cells were fixed with 100% methanol for 5 min at − 20 °C.
For IF analysis, cells were blocked with 5% BSA in PBS for 1 h at room temperature, were stained for RANK (1:200) and RANKL (1:200) antibodies, and then, were incubated with fluorescent secondary antibodies (1:2000).
The PLA protocol was applied according to the manufacturer’s instructions (Sigma-Aldrich). Cells were blocked with Duolink® Blocking Solution and stained for RANK (1:200), EFGR (1:1000), HER2 (1:200), HER3 (1:500), and HER4 (1:200) antibodies. Cells were then incubated with the PLA probes diluted 1:5 in Duolink® Antibody Diluent. Subsequently, ligation and amplification steps were performed. All incubations were performed in a humidified chamber at 37 °C.
In both techniques, cells were counterstained with DAPI for nuclei staining and glass coverslips were mounted on the slides. Fluorescence images were acquired using a Leica TCS SP8 confocal microscope (Leica Microsystems, Wetzlar, Germany). At least ten random fields of view were selected and images were taken. Data analysis was performed using Duolink® Image Tool Software (Sigma-Aldrich), developed for objective quantification of PLA signals (Smal et al. 2010). Representative images for each condition are shown.
Statistical analysis
All experiments were performed at least three times, and representative results of one experiment are shown. The data are presented as mean ± SD and analyzed by one-way ANOVA. GraphPad Prism 6 software was employed for these analyses. All statistical tests were two-sided. p values less than 0.05 were considered statistically significant.
Discussion
RANK and RANKL are expressed on various cell types and through their signaling can modulate vital cellular functions. Among the cellular systems that RANK axis has been reported to be implicated is mammary tumorigenesis. Interestingly, there are scientific reports revealing the involvement of RANK signaling in each stage of breast carcinogenesis [
10‐
12,
14]. Concrete data support RANK role in hormone [
10,
13] and
BRCA1 [
14]-associated BCs. Additionally, an implication in ERBB2-positive breast carcinomas is indicated [
10,
11,
21]. In this study, we explored and revealed a novel role of RANK as physical partner of ERBB family members in RANK-expressing BC cells, capable to modulate NF-κB signaling pathway resulting in the regulation of proliferation and survival of ERBB2-positive, ER-negative BC cells.
In vitro experiments in differentiating osteoclasts revealed for the first time a functional and physical association between RANK and EGFR signaling [
23]. Particularly, RANKL stimulation promotes EGFR expression and transactivation. Subsequently, activated EGFR forms a multiprotein complex including RANK, which leads to enhanced RANK downstream signaling, regulating osteoclast differentiation and survival [
23]. Of note, similar results have been observed in primary BC [
24]. They provided preliminary evidence of RANK interaction with EGFR pathways. A positive correlation between RANK (
TNFRSF11A) and
EGFR expression in BC was indicated [
24]. Moreover, RANK and EGFR co-expression is characterized by worse clinical outcome, enhanced downstream pathways, and induced cellular invasiveness in BC cells [
24]. Our findings from proximity ligation assay is an extension of the aforementioned studies revealing for the first time a direct association of RANK with ERBB family members (EGFR/ERBB2/ERBB3/ERBB4) in BC cells.
Dimer formation between RANK and ERBB2 seemed to be positively related to ERBB2 expression; thus, SKBR3 and BT-474, ERBB2-positive cells, exhibit a high number of RANK/ERBB2 dimers. RANKL treatment resulted in increased RANK/ERBB2 dimerization, while treatment with denosumab, trastuzumab, and/or pertuzumab had the opposite effect in ERBB2 (+) BC cells. Our findings are in accordance with previous studies in osteoclasts showing that RANK and EGFR association was enhanced by osteoclast differentiation medium, containing RANKL, and reduced by AG1478, a EGFR kinase inhibitor, treatment [
23]. RANKL stimulation seems to be crucial as its addition and inhibition affect RANK/ERBB2 dimerization pattern. The observation that dual targeting with trastuzumab and pertuzumab is more effective compared to trastuzumab, combined with the inhibitors molecular action [
25], leads to the assumption that ERBB2 associates with RANK as part of active homodimer or heterodimer with other members of ERBB family, forming a multiprotein complex.
Previous studies in osteoclasts and BC cells revealed that EGFR is implicated in augmentation of RANK downstream signaling [
23,
24]. NF-κB signaling is the main RANK downstream pathway. Activated NF-κB has been associated with tumor growth [
16] and drug resistance [
26] in ERBB2-positive breast cancer. NF-κB is activated through canonical and non-canonical signaling pathways. Although IKK
β is considered to be a key mediator in the canonical pathway, data from normal [
20] and malignant [
21] mammary revealed also a role of IKK
α in canonical pathway. IKK
α is necessary for ERBB2-induced breast tumorigenesis and plays an essential role in self-renewal of cancer stem cells [
16]. RANKL upregulated the phosphorylation and nuclear translocation of IKK
α in a ERBB2-induced mammary carcinoma cell line [
11]. Additionally, RANKL stimulation induced NF-κB activation in SKBR3 cells [
13]. In line with that, we observed a significant increase of phospho-IκB
α; however, IKK
α and p65 phosphorylation were slightly increased following treatment of SKBR3 with RANKL. Moreover, studies reported that lapatinib and trastuzumab decreased NF-κB signaling in SKBR3 cells [
21,
27]. Of note, additional to reduced phosphorylation levels of IκB
α and p65, we noticed a significant downregulation of phospho-IKK
α in SKBR3-treated cells with trastuzumab and trastuzumab plus pertuzumab. Merkhofer and colleagues, in an attempt to elucidate the pathway leading to NF-κB activation, proved that ERBB2 requires IKK
α to activate NF-κB through the canonical pathway independently of the PI3K pathway in ERBB2-positive breast cancer cells [
21]. Thus, ERBB2 might hold a regulatory role in IKK
α-mediated canonical pathway. Interestingly, the addition of RANKL to anti-ERBB2 drugs seemed to decrease inhibitors’ efficacy. Whereas when denosumab was added to the combination, RANKL had no further effect on downregulation of NF-κB signaling in SKBR3 cells. This data suggests that ERBB2 cooperate with RANK to regulate IKK
α-mediated canonical signaling.
ERBB2 activation of NF-κB via IKK
α promotes invasive phenotype in ERBB2-positive BC cells [
21]. Particularly, NF-κB promotes proliferation and survival, inducing ERBB2-mediated mammary tumorigenesis [
26]. RANK signaling stimulates proliferation and survival of breast tumor cells through induction of cyclin D1 and Bcl-2 protein expression [
28]. IKK
α plays a crucial role in NF-κB-cyclin D1-mediated proliferation of cancer cells [
13]. Accordingly, data from proliferation, apoptosis, and clonogenic assay revealed that RANKL stimulation enhanced proliferation, while survival was quite unaffected in SKBR3 cells. Interestingly, SKBR3 cells exhibited decreased rates of proliferation and enhanced apoptosis when treated with a combination of denosumab with trastuzumab. RANKL addition to the combination diminished the efficacy of the inhibitors. However, denosumab plus dual ERBB2 targeting seemed to be more effective concerning growth inhibition and tolerance to RANK ligand.
RANKL promotes migration, invasion, and metastasis, by activation of NF-κB signaling and subsequent upregulation of Snail and Twist, key regulators of epithelial to mesenchymal transition (EMT) [
29]. RANKL signaling has, also, a vital role in the expansion of BC stem cells (CSCs) [
13], leading probably to a robust metastatic activity. RANKL has been demonstrated to mediate CSC self-renewal in ERBB2-positive cells [
15]. Notably, IKK
α is essential to this process [
16]. In particular, IKK
α has been shown to regulate ERBB2-induced CSC expansion by stimulating the nuclear export of p27, a negative regulator of the G1–S transition [
30]. Our observations about RANKL treatment having no additional effect on migration capacity are in contrast with the aforementioned studies. However, denosumab combined with trastuzumab and/or pertuzumab significantly decreased migration rate in SKBR3 cells. This is in agreement with previous findings showing that RANKL inhibition decreased proliferation and metastases in ERBB2 transgenic mice [
10]. In this line, RANK+/− ERBB2 transgenic mice exhibited reduced metastatic rates compared with homozygotes [
11]. Furthermore, silencing or blocking RANK diminished metastatic rate, while RANKL treatment had the opposite effect in ERBB2-induced mammary carcinomas [
11].
Based on the data described so far, we can assume the importance of RANK signaling in proliferation, survival and metastatic potential of ERBB2 positive BC cells. Also, it is reasonable to conclude that combinatorial targeting of ERBB2 and RANKL could be a more effective approach, which would overcome RANKL effect in suppressing anti-tumor action of anti-ERBB2 agents. There are evidence supporting that NF-κB signaling might also constitute an essential mechanism of resistance to anti-ERBB2 strategies when applied for the BC treatment [
26,
31]. Following this rationale, the combination of anti-ERBB2 agents with proteasome inhibitors (which block IκB degradation and thus NF-κB activation) [
31] and NF-κB inhibitors [
26,
32] has been successfully tested as a novel therapeutic strategy for treating ERBB2-positive BC patients.
RANK expression has been linked with hormone receptor negativity, high pathological grade, and worse clinical outcome [
33]. Moreover, a recently published study revealed an association of RANK and RANKL dual expression with poor clinical outcomes in triple-negative BC (TNBC) [
34]. Interestingly, our research using MDA-MB-453 cells demonstrated that RANKL stimulation and inhibition affected NF-κB signaling, proliferation, and slightly migration, indicating a RANK role in TNBC.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.