Introduction
Neuropeptide Y (NPY) is a 36 amino acid peptide amide named for its terminal tyrosine [
1,
2]. NPY is produced in the central and peripheral nervous system and is the most abundant neuropeptide in the brain and spinal cord [
3]. NPY acts on six G-protein coupled receptors (GPCRs) named NPY receptors (NPYRs). Most research has focused on NPY1R, NPY2R, and NPY5R because they are the most highly expressed and functionally relevant NPYR subtypes in humans [
3,
4]. NPY1R and NPY5R expression is high in several types of tumors, such as ovarian, prostate, breast, and neural crest relative to normal tissue [
3,
5‐
9]. Breast cancer in particular has come to the forefront because of its high frequency of NPYR overexpression and density compared with all other NPYR-positive tumors [
9]. This characteristic has been exploited to develop chemically modified analogs of NPY that are being explored in breast cancer imaging and diagnosis [
10‐
12]. When examining NPYR expression in breast carcinomas, only 24% are positive for NPY2R compared to 85% for NPY1R. Furthermore, breast cancer cell lines such as MDA-MB-231 and MCF7 have elevated levels of NPY1R and NPY5R [
3,
4]. NPY1R and NPY5R stimulation promotes cellular proliferation, migration, and angiogenesis in breast cancer models [
4,
13,
14]. Since breast tissue is highly innervated by the sympathetic nervous system and provided with a large supply of NPY ligand, it can be the perfect storm for the constitutive signaling of this pathway. Therefore, employing NPYR antagonists in the context of the tumor microenvironment could be a viable strategy in breast cancer therapy.
Hypoxia is a common feature of the tumor microenvironment that can lead to chemo- and radiation therapy resistance and promoting metastasis [
15]. As solid tumors grow, their leaky vasculature provide an insufficient supply of oxygen to often hyper-consuming cancer cells. The cellular response to hypoxia is primarily driven by the hypoxia inducible transcription factors (HIFs) that induce an array of genes including vascular endothelial growth factor and angiopoietin-2 [
16]. HIF-1 is a key regulator for glycolysis and pH regulation whereas HIF-2 is involved in proliferation and differentiation [
17,
18]. Recently, our group demonstrated that NPY1R and NPY5R mRNA abundance is induced by the HIFs, sensitizing hypoxic cells to NPY-stimulated motility and proliferation in MCF7 and MDA-MB-231 breast cancer cells [
19]. Simultaneously, signaling through the mitogen activated protein kinase (MAPK)/ERK pathway was induced more rapidly and potently upon NPY5R stimulation in hypoxic cells relative to normoxic cells. This caused hypoxic breast cancer cells to proliferate and migrate more than their normoxic counterparts. Therefore, hypoxia contributes to NPYR hyperactivity by increasing receptor production.
Recently, modulation of NPYRs has been implicated as a treatment for a wide range of diseases such as obesity, mood disorders, pain, and cancers [
20]. The overexpression of NPY and its role in cancer progression could translate into cancer therapeutics, specifically through the use of NPYR antagonists, which have shown promising results for other diseases [
21]. Preliminary studies on the use of NPYR antagonists as cancer treatments has also shown potential, making this a promising area of research. Further evidence on NPYR antagonism in the progression of cancer, especially in the context of hypoxic cell vulnerability, would shed valuable insight for the future of this potential therapeutic strategy.
Here we investigate the effect of antagonizing NPY1R and NPY5R isoforms on MAPK signaling, cell migration, cell proliferation and invasion, in 2D and 3D models of hypoxic and normoxic MCF7 and MDA-MB-231 breast cancer cell lines. These cell lines are used here as a continuation of our previous study showing they are more sensitive to NPY stimulation in hypoxia [
19]. Further, these cell lines are each models of two different cancer subtypes that can provide insight into potential genetic differences in their susceptibility to NPYR antagonists. MCF7 is an estrogen receptor-positive model of the luminal A subtype and MDA-MB-231 is a triple-negative breast cancer (TNBC) basal-like subtype [
22]. MAPK signaling was more greatly reduced in hypoxia in both cell lines when isoform-specific agonists were used in combination with antagonists. Only hypoxic MDA-MB-231 cell proliferation could be antagonized by NPYR inhibitors. Cell migration in MDA-MB-231 cells was only antagonized in normoxia, while hypoxia improved the effect of the antagonists in MCF7 only when stimulated with the general NPY agonist. Cell invasion was mostly repressed by antagonizing NPY5R in both cell lines, but hypoxia improved the effect of the Y5 antagonist when MCF7 cells were stimulated with the general NPY agonist. Spheroid growth, but not invasion, was repressed in MDA-MB-231 with NPYR antagonists, while MCF7 spheroid growth and invasion were both repressed specifically with the NPY5R antagonist. In human breast tumor tissue, we show that high NPY5R levels correlated with advanced stage cancer, metastasis, and poorly differentiated cells. Further, higher NPY1R levels correlated with poor patient outcomes such as death and progression-free survival. We show that antagonizing the NPYRs increased their own mRNA abundance in hypoxic spheroids. We observed some differences between cell lines and in response to oxygen, highlighting that more studies are required to decipher the complex signaling dynamics of the NPYRs in the tumor microenvironment. This study should help inform the future development of NPYR antagonists in breast cancer therapy and patient-based treatment plans based on NPYR levels.
Methods
Cell culture and reagents
MDA-MB-231 and MCF7 cells were obtained from the American Type Culture Collection and maintained in Dulbecco’s Modified Eagle Medium (DMEM) supplemented with 10% fetal bovine serum, as suggested. They were maintained mycoplasma free in a humidified chamber (5% CO2, and 37˚C). Cells were introduced to and maintained in hypoxia by incubating them in a HypOxystation H35 workstation (HypOxygen) at 1% O2, 5% CO2, and 37˚C. Cells were treated with NPYR agonists (Tocris):1 × 10− 9 M NPY (Cat#1153), 1 × 10− 8 M NPY1R-specific (Cat#1176), or 1 × 10− 8 M NPY5R-specific (Cat#1365), and NPYR antagonists (Tocris): 1 × 10− 6 M NPY1R (BIBP 3226, Cat #2707), 1 × 10− 5 M NPY5R (L-152, Cat# 1382) For spheroid experiments, BIBP 3226 and L-152 were used at 1 × 10− 5 M and 1 × 10− 4 M, respectively. Cells were pre-incubated with antagonists for 30 min prior to agonist exposure.
pERK assay
7,250 MDA-MB-231 or 8,700 MCF7 cells were seeded in a black bottomed 96-well plate and exposed to serum-reduced media for 24 h in normoxia or hypoxia. Media was then replaced with serum-reduced media (0.5% FBS) with or without antagonist. After 30 min, additional serum-reduced media was added with or without agonist. Cells were then lysed after a period of 5, 15, and 30 min. Cells were then fixed and pERK/ERK was measured using fluorescence-based ELISA according to manufacture instructions (BioAssay Systems, EERK-100).
Cell migration assay
Transwell migration assays were used to determine cell chemotaxis under given pharmacological conditions. Cells were given serum reduced (0.5% FBS) media for 24 h, then seeded at a density of 75,000 (MDA-MB-231) or 130,000 cells (MCF7) in the upper chamber of 12 well inserts with 8 μm pores (BD Biosciences). These cells were seeded in serum-reduced media in the presence or absence of antagonist. After 30 min, serum-reduced media with or without agonist was added to the bottom chamber. Cells were then exposed to 22 h (MDA-MB-231) or 24 h (MCF7) of normoxia or hypoxia. The near serum-starvation and end-point of ≤ 24 h was done to reduce the potential of cell proliferation to contribute to cell migration and invasion. Non-migrated cells were removed from the upper side of the inserts with a cotton swab and migrated cells were fixed in methanol and stained with Hoechst. Membranes were excised, mounted on slides, and imaged on a Nikon Eclipse Ti Microscope. Migrated cells were quantified using Fiji for ImageJ. A threshold was first determined that accurately highlighted migrated cells. A watershed mask was then applied to segregate adjacent cells. Parameters of circularity and size were applied, and number of particles was counted by ImageJ. Number of cells migrated were then compared to controls.
Cell invasion assay
Cells were given serum reduced (0.5% FBS) media for 24 h, then seeded at a density of 40,000 (MDA-MB-231) or 69,000 (MCF7) in the upper chamber of 24 well transwell inserts with 8 μm pores. For MDA-MB-231 cells, inserts (Corning) were rehydrated in cell culture media in a humidified chamber for 2 h, then transferred to a companion plate using sterile forceps. For MCF7 cells, inserts were coated in 0.2 mg/mL growth factor reduced Matrigel Matrix (BD Bioscience) and left to set in a humidified chamber for 2 h. Protocol for the cell migration assay was then followed for both cell lines.
Cell proliferation assay
BrdU-ELISA assays were used to examine cell proliferation with pharmacological treatment. 10,000 cells were seeded in a 96-well plate and the next day full media was replaced with serum-reduced media for 24 h. Antagonists were applied for 30 min following the addition of agonist or vehicle control in serum-reduced medium. Cells were incubated in normoxia or hypoxia for 20 h followed by an additional 4 h incubation with 1X BrdU substrate (Abcam, ab126556). Cells were then fixed and BrdU incorporation was measured according to manufacturer instructions.
10,000 cells were plated in round-bottom U shaped 96-well plates (Corning). The plates were then spun in a circular motion to promote aggregation of cells into a single spheroid per well. Spheroids were then grown for 72 h. Once spheroids reached a size of 1-1.2 mm in diameter, antagonist and agonist treatments were applied. Following 24 h of treatment, a minimum of 24 MDA-MB-231 or 16 MCF7 spheroids were collected per condition for RNA extractions and qPCR analysis. Images of spheroids were captured on a Nikon Eclipse Ti Microscope after treatments at 0 and 24 h. Surface area of spheroid was measured using ImageJ Fiji software to assess spheroid growth. Spheroids were cultured in normoxia, but produce a hypoxic microenvironment that make them valuable tools to study the tumor microenvironment. We detected the degree of spheroid hypoxia by using CAIX mRNA levels as a hypoxia marker in an RT-qPCR of spheroid lysates compared to a normoxic monolayer.
Spheroid invasion assay
10,000 cells were plated in flat bottomed 96-well plates (Thermo Scientific) coated with 1.5% low melting agarose (Fisher Scientific). MDA-MB-231 spheroids were grown on Spheroid Formation ECM (Cultrex). The plates were spun in a circular motion to promote aggregation of cells in the middle of each well into single spheroids. Spheroids were then grown for 72 h before embedding in vehicle control or antagonist-enriched ECM (Cultrex). After 30 min, agonists were applied on top of solidified ECM in serum-reduced media. Images were captured of spheroids after initial treatment and at intervals of 0 h, 24 h, 48 h, and 96 h to assess treatment impact on invasion into surrounding ECM on a Nikon Eclipse Ti Microscope. The invasive protrusions into the ECM were measured using ImageJ Fiji software.
Immunofluorescence
Formalin-fixed paraffin-embedded human tumour tissue samples with a thickness of 5 μm were obtained from the Ontario Tumour Bank (46 invasive ductal breast carcinoma cases and 10 adjacent normal breast tissue), which is supported by the Ontario Institute for Cancer Research through funding provided by the Government of Ontario. Samples were chosen based on receptor status such that 24 were negative for the estrogen receptor (ER), progesterone receptor (PR), and the human epidermal growth factor receptor 2 (HER2), and 22 were positive for the estrogen receptor to reflect the cell lines used in this study. Sample pathological T (tumour), grade, clinical M (metastases), and patient outcome were also considered. Samples on microscope slides were rehydrated using xylene and serial dilutions of ethanol followed by a rinse in deionized water and a wash with 1XTBS. Antigen retrieval was then preformed using pre-heated antigen retrieval buffer (10 mM Tris, 1 mM EDTA, 0.05% Tween 20, pH 9.0) at 85 °C for 30 min in a water bath. Slides were permeabilized using TBS + 0.025% Triton X-100 then blocked in 10% goat serum with 1% BSA in TBS for 2 h at room temperature. CAIX at 1/35 (Novus, NBP1-51691) was combined with either NPY1R at 1/100 (Abcam, ab91262) or NPY5R at 1/400 (Abcam, ab133757) at 4 °C overnight. Primary antibody was omitted for the negative controls. Slides were then counterstained with 1/100 Alexa Fluor 555 (Life Technologies) and 1/250 Alexa Fluor 488 (Cell Signaling Technologies) for one hour followed by 1/50,000 Hoechst (Cell Signaling Technologies) for 8 min. Slides were then mounted using ProLong Gold (ThermoScientific) and images were captured on a Nikon Eclipse Ti Microscope. CAIX, NPY1R, and NPY5R expression and colocalization were quantified using ImageJ Fiji software. Deconvolution was performed using the Iterative Deconvolution plugin with 16 iterations followed by colocalization analyses using the JACoP plugin to calculate fractional overlap between CAIX and NPY1R/NPY5R using Manders’ Colocalization Coefficients after thresholding. The same threshold was used to calculate the percent positive pixels for NPY1R, NPY5R, and CAIX. Full and informed patient consent was obtained, and the project was approved by the University of Guelph Research Ethics Board Committee and Ontario Institute for Cancer Research Ethics Committee.
RNA extraction and qRT-PCR
RNA was extracted using Trizol (Invitrogen) per manufacturer’s instructions. 2 µg of RNA was reverse transcribed using a high-capacity cDNA reverse transcription kit (Applied Biosystems). Primers used are described in Table
S1. Quantitative PCR was performed using SsoAdvanced Universal SYBR Green Supermix (BioRad). Data was retrieved with CFX manager software (BioRad) and melting curves were examined to confirm the absence of primer dimers. Relative fold change of expression was calculated using the ΔΔCT method, and transcript levels were normalized to endogenous controls RPLP0 and RPL13A and then compared to internal controls.
Statistical analysis
Statistical analyses were performed using GraphPad and data are presented as mean ± SEM. The Bartlett test was used to verify the normal distribution of data. Statistical differences between treatments were then evaluated by one-way ANOVA followed by Tukey’s honestly significant difference (HSD) post-hoc test.
Discussion
NPY1R and NPY5R first gained relevance in breast cancer research because of their high abundance and density compared with all other NPYR-positive tumors [
9]. This characteristic has been exploited to develop chemically modified analogs of NPY that are used in breast cancer imaging and diagnosis [
10‐
12]. We investigated whether such analogs could antagonize NPY1R and NPY5R to influence cancer hallmarks in two different breast cancer cell lines while considering the effects of hypoxia, a major component of the tumor microenvironment. We found that, in general, antagonizing NPY1R and/or NPY5R in hypoxia can more greatly reduce MAPK signaling, cell proliferation, cell migration and invasion, and spheroid growth and invasion. This suggests that inhibiting the NPYRs could be a clinically relevant avenue. However, antagonizing either NPY1R or NPY5R often produced different effects with respect to oxygen or cell line. This reveals that NPYR signaling and function in breast cancer has complexities that need to be discussed and researched further.
We noticed during the MAPK signaling assay that pERK1/2 was not induced by any agonist in any of our treatments, even though such activation was observed in other cell lines [
3,
4,
26]. The MAPK pathway can be constitutively active in breast cancer [
24] and in some cell lines such as MDA-MB-231 [
25], and it is possible that a higher baseline pERK1/2 is difficult to induce further in these cell lines. We would like to note that a previous study from our group could produce an induction of pERK1/2 with NPY. The MCF7 and MDA-MB-231 cells were obtained from the ATCC in both studies but at different times. This could point to genomic variability and clonal evolution that can be observed within cell lines. A thorough analysis of this phenomenon was done in HeLa [
27], but has also been observed in MCF7 and MDA-MB-231 [
28]. These cell lines in both studies display many consistencies such as their agonist-induced proliferation that is more potent in hypoxia (Fig.
2). Even with what appears to be higher baseline levels of pERK1/2, we were still able to repress pERK1/2 levels with NPY1R and NPY5R antagonists in some contexts (Fig.
1). We observed repression of pERK1/2 in normoxia in both cell lines, but this effect was negated by hypoxia (Fig.
1A-B). Hypoxia may influence NPY agonist efficiency through crosstalk between NPY1R and NPY5R that includes heterodimerization and receptor recycling changes [
29,
30]. When we used Y1- and Y5-specific agonists, we observed almost no reductions in pERK1/2 levels in normoxia. In hypoxia, however, both isoform-specific agonists did reduce pERK1/2 levels in all conditions. Isoform-specific NPY agonists are physiologically relevant. For example, there is a hypoxia-induced peptidase DPPIV that cleaves the general NPY agonist into a Y5-specific agonist [
31]. Therefore, hypoxic tumor cells could be more vulnerable to MAPK repression through NPYR antagonists.
Cell proliferation could not be inhibited in MCF7 by NPYR antagonists, while it was inhibited in MDA-MB-231 cells only in hypoxia (Fig.
2). Stimulation of estrogen receptor-ɑ (ER-ɑ) via estrogen has been shown to result in upregulated NPY1R expression in MCF7 cells, but not in MDA-MB-231 cell [
32]. Estrogen treatment increases cell proliferation, which can be reduced by addition of NPY. This NPY-induced inhibition of the proliferative effect of estrogen can be rescued by the addition of NPY1R inhibitors [
32]. Further, estrogen treatment decreased NPY secretion via the PI3K and AMPK pathways [
33]. Due to the evidence of crosstalk between ER-ɑ and the NPYR pathway, we may have stimulated proliferation with NPY agonists via ER-α that could not be reversed with NPYR antagonists. In fact, we produced even more proliferation in MCF7 cells treated with both Y1-specific agonist and antagonist (Fig.
2E). Therefore, the reductions in cell proliferation by NPYR antagonists observed in a triple negative cell line such as MDA-MB-231 cells may not translate to ER-positive cancers.
Similar to the observations for MAPK signaling, repression of cell migration in normoxia by the NPYR antagonists was largely reversed by hypoxia except when the Y1 isoforms were specifically stimulated (Fig.
3). Our data suggest that NPYR antagonists could be more successful at repressing cell migration in hypoxic ER + tumors since NPY-induced MCF7 cell migration was only impaired with antagonists in hypoxia (Fig.
3D). Cell invasion was more uniformly repressed by NPYR antagonists despite cell line differences or oxygen availability. This suggests that the NPYRs could induce more invasion-related genes than migration.
We investigated how the NPYR antagonists influence CAIX expression as a possible explanation for the observed reduction in spheroid growth and invasion. CAIX was used as a marker of hypoxia, but it has also been connected to hypoxic invasion through its interactome in breast cancer cells. CAIX associates with α2β1 integrin, CD98hc, and MMP14 at the leading edge where it donates hydrogen ions required for MMP14 catalytic activity and subsequent extracellular degradation [
34]. Indeed, CAIX mRNA levels were reduced only in the presence of the Y5-specific antagonist (Fig.
5C and F). This reduction of CAIX expression could explain the slower spheroid growth and impaired invasion in MCF7 spheroids. MDA-MB-231 spheroids did display some reduction in growth in a Y5-specific manner when stimulated with the general NPY agonist, but invasion was not affected. This could be due to their decreased sensitivity to drugs or the hyperactivity of select pathways discussed earlier. Importantly, we noticed that the antagonists alone sometimes elicited a response or that the antagonists induced the mRNA abundance of the receptors they were antagonizing. This was not entirely surprising since antagonizing one NPYR isoform could influence its dimerization with another isoform or another GPCR altogether. This highlights the importance of performing an antagonist alone control. We did not notice many instances of the antagonist alone eliciting a response, and in fact this control allowed us to reveal a positive feedback loop whereby the antagonist induced the expression of its own receptor or that of another NPYR isoform.
Translating molecular profiles into clinical relevancy is important for bridging the gap between in vitro experiments and patients. In breast tumor tissue compared to normal samples, we show that high NPY5R levels correlate with advanced stage cancer, metastasis, and poorly differentiated cells. Further, higher NPY1R levels correlated with poor patient outcomes such as death and progression-free survival. These data are in agreement with the in vitro experiments showing that spheroid growth and invasion flow mostly through the NPY5R axis. However, it is unclear why NPY1R abundance is more strongly connected to outcome, which currently limits the clinical relevance of NPY5R expression levels in breast cancer.
Research into the complexities of how the NPYR isoforms interact with one another and perhaps other GPCRs in normoxia and hypoxia to influence signaling cascades is still in its infancy. This study highlights that the development of NPYR antagonists in breast cancer therapy and patient-based treatment plans could be a promising avenue to continue pursuing.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.