The Transmodulation of HER2 and EGFR by Substance P in Breast Cancer Cells Requires c-Src and Metalloproteinase Activation

Background Substance P (SP) is a pleiotropic cytokine/neuropeptide that enhances breast cancer (BC) aggressiveness by transactivating tyrosine kinase receptors like EGFR and HER2. We previously showed that SP and its cognate receptor NK-1 (SP/NK1-R) signaling modulates the basal phosphorylation of HER2 and EGFR in BC, increasing aggressiveness and drug resistance. In order to elucidate the mechanisms responsible for NK-1R-mediated HER2 and EGFR transactivation, we investigated the involvement of c-Src (a ligand-independent mediator) and of metalloproteinases (ligand-dependent mediators) in HER2/EGFR activation. Results and Discussion Overexpression of NK-1R in MDA-MB-231 and its chemical inhibition in SK-BR-3, BT-474 and MDA-MB-468 BC cells significantly modulated c-Src activation, suggesting that this protein is a mediator of NK-1R signaling. In addition, the c-Src inhibitor 4-(4’-phenoxyanilino)-6,7-dimethoxyquinazoline prevented SP-induced activation of HER2. On the other hand, SP-dependent phosphorylation of HER2 and EGFR decreased substantially in the presence of the MMP inhibitor 1–10, phenanthroline monohydrate, and the dual inhibition of both c-Src and MMP almost abolished the activation of HER2 and EGFR. Moreover, the use of these inhibitors demonstrated that this Src and MMP-dependent signaling is important to the cell viability and migration capacity of HER2+ and EGFR+ cell lines. Conclusion Our results indicate that the transactivation of HER2 and EGFR by the pro-inflammatory cytokine/neuropeptide SP in BC cells is a c-Src and MMP-dependent process.


Introduction
The cellular and non-cellular components of the tumor microenvironment shape tumor evolution [1]. Among the components of the tumor microenvironment, the nervous system and the neuropeptides secreted by non-neuronal (i.e., by modulating immune cells) and neuronal cells appear to have a direct and indirect effects on tumor progression [2]. This is the case of neurokinin 1 receptor (NK-1R) (TACR1 gene) and its preferential ligand substance P (SP) (TAC1 gene), a pro-inflammatory cytokine and neuropeptide that belongs to the family of tachykinins [3,4]. This family consists of SP, neurokinin A (NKA) and neurokinin B (NKB), encoded by the TAC1 (SP and NKA) or TAC3 (NKB) genes [5], and the recently discovered hemokinins and endokinins encoded by the TAC4 gene [5][6][7]. Specifically, NK-1R is a G-protein coupled receptor (GPCR) which, together with SP, is expressed in the central nervous, gastrointestinal, and immune systems, and is involved in cellular responses such as pain transmission, paracrine and endocrine secretion, vasodilation, angiogenesis and modulation of cell proliferation [5,[8][9][10][11]. SP not only signals through NK-1R; it can also bind (with lower affinity) to additional tachykinin receptors like neurokinin 2 receptor (NK-2R) and neurokinin 3 receptor (NK-3R) encoded by the TACR2 and the TACR3 gene respectively [5,12].
Despite their physiological functions, G proteins can also activate pathways related to cellular proliferation and survival in several types of cancer cell through secondary messengers and receptors, as in the case of NK-1R [13][14][15]. This receptor is expressed on the cell surface of many cancer cell types like breast [16][17][18][19], pancreatic [20], colon [21,22], and laryngeal cancer cells [23], glioblastoma [22], acute lymphoblastic leukemia [5,24], and melanoma [5]. NK-1R signaling can activate tyrosine kinase receptors (RTKs) like EGFR and HER2 [25][26][27]. The RTK family shares a similar structure, and the receptors belonging to the ErbB family (EGFR, HER2, HER3, and HER4) are driver oncogenes in different types of cancer [28,29]. Several reports have shown the involvement of the non-receptor protein tyrosine kinase c-Src and metalloproteinases (MMPs) in the GPCR-mediated activation of ErbB receptors [30][31][32]. Activated c-Src can bind to the cytoplasmic tail of EGFR and HER2 and phosphorylate tyrosine residues; therefore, c-Src activation may lead to the triggering of ErbB receptors in a ligandindependent manner [30,31]. The signal transduction by G-proteins may also enhance ligandmediated EGFR activation by stimulating MMPs synthesis and secretion and favoring the shedding of membrane-anchored ligands [14,33].
The interaction of GPCRs and RTKs has a prominent role in various physiological processes [13,34,35], but it is also involved in pathologic conditions since its deregulation can drive tumorigenic processes [14]. We previously identified SP as a key modulator of the steady state of HER2 and EGFR, with the functional consequence of enhanced tumor aggressiveness and tumor progression, and alterations in the cellular responses to apoptotic stimuli [27]. In the present study, we aimed to identify the mechanisms involved in the transactivation of HER2 and EGFR by SP in BC cells. Focusing on the involvement of ligand-independent and dependent mediators, we conclude that the transmodulation of HER2 and EGFR in response to SP is a c-Src and MMP-dependent mechanism.

Cell lines and reagents used in the study
The following cell lines were purchased from American Type Culture Collection and were cultured in accordance with the instructions: MDA-MB-453, BT-474, SK-BR-3, MDA-MB-231, and MDA-MB-468. The cultures were incubated at 37°C in a humidified 5% CO 2 atmosphere and the cells were serum starved overnight before experiments, unless otherwise specified. For some proliferation experiments, cells were grown in a complete growth medium plus fetal bovine serum (FBS), as specified in the methods section. The authenticity of all the cell lines used in this study was validated by single locus short tandem repeats (STR) typing (Bio-Synthesis, Inc.).

Time-course studies
To determine the effects of SP treatment on c-Src activation, cells were seeded in 100 mm culture dishes, grown until 80% confluence, serum starved for 24 hours, and then treated at the indicated times with 100 nM of SP. After each treatment, the cells were washed twice in cold PBS and rapidly frozen until protein extraction. To determine the effects of SP in the presence of c-Src, inhibitor cells grown until 80% confluence were serum starved for 4 hours, and treated for 20 hours with c-Src inhibitor 4-(4 0 -phenoxyanilino)-6,7-dimethoxyquinazoline (1μM). To determine the effects of SP in the presence of the MMP inhibitor 1-10, phenanthroline monohydrate, cells were serum starved for 16 hours and treated with phenanthroline monohydrate (7μM) for 4 hours. Finally, to determine the effects of SP in the presence of c-Src and MMP inhibitors, cells were serum starved for 4 hours and treated with c-Src inhibitor 4-(4 0 -Phenoxyanilino)-6,7-dimethoxyquinazoline for 16 hours. At this point phenanthroline monohydrate was added to the cells, and both inhibitors were incubated for 4 additional hours. The control group without treatment was serum starved for 24 hours. Subsequently, the cells were treated with SP 100 nM for 6, 10 and 15 min. After the treatment, the cells were washed twice in cold PBS, and rapidly frozen until protein extraction. The experiments with each cell line were repeated at least twice to ensure the reproducibility of the data. In all cases, the corresponding dose of DMSO or MetOH (never above 0.1% v/v) was added to the control points.

Inhibition of NK-1R with L-733,060 antagonist
To inhibit NK-1R signaling, cells cultured until 70% confluence were serum starved for 5 hours and then treated with 20 μM (SK-BR-3 and BT-474) and 30 μM (MDA-MB-453) of NK-1R antagonist L-733,060 during 24h. For the simultaneous inhibition of the three receptors NK-1R, NK-2R and NK-3R cells were also treated with MEN 10376 (30 μM, NK-2R antagonist) and SB 218795 (20 μM, NK-3R antagonist). After the treatment, the cells were washed twice in cold PBS, and rapidly frozen until protein extraction. The experiments with each cell line were repeated at least three times to ensure the reproducibility of the data, and all quantitative measurements were generated from three or more replicates. The statistical significance of the data was analyzed by t-test (two-tailed). P values < 0.05 were considered statistically significant.

Western blot
For protein extraction, cells were lysed in ice-cold radioimmunoprecipitation assay buffer (RIPA) (Tris-HCl 50 mM, pH 7.4; NP-40 1%; Na-deoxycholate, 0.25%; NaCl 150 mM; EDTA 1 mM; PMSF 1 mM; proteinase inhibitors; Na3VO4 1 mM and NaF 1 mM) and sonicated for 10 seconds. After centrifugation (13000 rpm from 5 min) supernatants were quantified for protein content. Equal amounts of proteins were separated by SDS-PAGE and electrophoretically transferred to polyvinylidene difluoride membranes (BioRad Laboratories, CA), blocked with 5% milk in PBS for 1 hour, incubated overnight with the corresponding primary antibodies: phospho-EGFR Tyr1068 ( Chemiluminiscence on membranes was detected after ECL treatment (GE Healthcare Amersham, NJ, Cat# RPN2209) and image capture was performed with a Fujifilm LAS3000 imaging system. The Image Gauge software was used for the densitometric quantification of each protein. Correct Mr was compared with pre-stained protein standards (BioRad Laboratories, CA, Cat# 161-0374). The experiments with each cell line were repeated at least three times to ensure the reproducibility of the data and all quantitative data were generated from three or more replicates. The statistical significance of the data was analyzed by t-test (two-tailed). P values < 0.05 were considered statistically significant. The cells were incubated for 45 minutes at 37°C in a humidified 5% CO 2 atmosphere and then, the plate was read on a Synergy HT Multi-Detection Microplate Reader (BioTek) at 485±10 nm (excitation optical filter) and 530±12,5 nm (emission optical filter). Different doses were assessed in sixtiplicate. In all cases, the corresponding dose of DMSO or MetOH (never above 0.1% v/v) was added to the control points. Assay values for controls were taken as 100% of viability, and the viability at each treatment point were calculated relative to controls by the formula: %Live Cells = (F(530) sam-F(530) min) /F(530) max -F(530) min ) x 100% according to the manufacturer's instructions.

Cell migration assay
Cell migration was assessed in culture cells not greater than 80% confluence and serum starved 24h prior to assay. A total of 1 × 10 6  In all cases, the corresponding dose of DMSO or MetOH (never above 0.1% v/v) was added to the control points. Assay values for controls without inhibitors were taken as 100% of migration, and the viability at each treatment point were calculated relative to their controls.

Statistical analysis
Statistical analysis of the results was performed by ANOVA with Tukey's Multiple Comparison post-hoc test and t-test (two-tailed). Statistical significance was considered since P values less than 0.05.

Results
The neuropeptide/proinflammatory mediator SP activates c-Src in BC cell lines We first investigated whether SP used the c-Src protein as a cell-signaling mediator in BC cells, as previously shown in other cell types [36,37]. First, we checked a panel of BC cell lines under basal conditions without stimulation ( Fig 1A) and we detected different levels of phosphorylated c-Src protein relative to total levels. Second, using time-course studies, we observed that SP treatment induced the phosphorylation of c-Src Tyr416 (indicative of Src activation [38,39]) at the indicated time points (Fig 1B) in all the cell lines, including the HER2 negative cell lines MDA-MB-231 or MCF7. Some lines have more pronounced phosphorilation than others, and this activation is not consistent in all time points used because NK-1R activation by SP is a cyclic activation as we previously described [16]. For these reason, the activation of c-Src Tyr416 within this time frame was consistently observed in all the replicates conducted, although the exact time point and intensity of maximum activation varied. Lower activation of c-Src was found after SP treatment in the MDA-MB-453 cell line and was slightly pronounced in MDA-MB-468 line, probably due to the very low or very high basal levels of c-Src (phosphorylated and total protein) [30], respectively.

The overexpression or inhibition of NK-1R modulates c-Src activity in BC cell lines
We previously reported that the stable transfection of NK-1R into the HER2-negative MDA-MB-231 cell line can be used as a tool to study the mechanism by which SP contributes to the persistent transmodulation of the ERBB receptors [17]. These previous results demonstrated that the overexpression of NK-1R enhanced SP-mediated HER2 activation even in a HER2-negative and NK-1R-low cell line, the main reason we selected that particular cell line , the treatment with SP for 6 or 10 minutes further increased the phosphorylation of c-Src Tyr416 (5.5-and 4.6-fold, respectively, red bar) and in all cases were expressed by ratio of phospho/total protein (Fig 2A).
We next analyzed the effects of NK-1R inhibition on c-Src activation. The HER2+ SK-BR-3, BT-474, MDA-MB-453, and the EGFR+ MDA-MB-468 cell lines were treated with the NK-1R antagonist L-733,060 for 48 hours. NK-1R antagonism significantly reduced c-Src phosphorylation at Tyr416 in SK-BR-3 and BT-474 cell lines while a non-significant trend towards inhibition was observed in the MDA-MB-453 and MDA-MB-468 cell line with the lowest or highest levels of c-Src (phosphorilated and total protein), respectively (Fig 2B), so, it is not surprising to observe fewer changes in cell lines which steady state of c-Src (both, the phosphorylated and total protein) is already low or high. Since SP can also bind with lower affinity to NK-2R and NK-3R receptors, we next investigated the effects of the triple chemical inhibition of NK-1R, NK-2R, and NK-3R with L-733,060, MEN 10376, and SB 218795 inhibitors respectively. We observed that the triple inhibition of SP receptors cause a dramatic downregulation of c-Src phosphorylation (Fig 2C), indicating that c-Src is indeed triggered by tachykinin signaling in BC cells.

The transactivation of HER2 and EGFR by SP is dependent on c-Src and MMPs
To investigate the role of c-Src in SP-mediated HER2 and EGFR activation [30] we next performed time-course studies with SP in the presence of the c-Src inhibitor 4-(4 0 -phenoxyanilino)-6,7-dimethoxyquinazoline [40]. Inhibition of c-Src activity in the HER2 positive SK-BR-3 cell line significantly blocked SP-induced phosphorylation of HER2 Tyr1248 compared to control cells (Fig 3A and 3B). HER2 transactivation by SP was also substantially inhibited in the presence of the MMP inhibitor 1-10, phenanthroline monohydrate, and almost completely abolished after the inhibition of both pathways, suggesting that the transactivation of HER2 by SP in BC cells is a c-Src and MMP-dependent process (Fig 3A and 3B). SP signaling activates the mitogen-activated protein kinase (MAPK) pathway [10,26,41]; therefore, the phosphorylation of p42/44 MAPK was used to control of NK-1R downstream activation. For this reason, the phosphorylation of p42/44 MAPK was not always reduced in the presence of c-Src and MMP inhibitors (Fig 3A and 3C), since the activation of the MAPK pathway can be triggered by both NK-1R and ERBB signaling [14,37].
To determine whether the transmodulation of EGFR by SP was also dependent on c-Src and MMPs in BC cells, we performed similar experiments in the EGFR positive cell line MDA-MB-468. In the control situation, addition of SP increased EGFR phosphorylation at 6 min (1.17-fold), 10 min (1.45-fold) and particularly at 15 min (2.61-fold). No increase occurred in the presence of the inhibitors (alone or in combination) under SP treatment, as we observed in the western blot and densitrometic quantification of phospho EGFR/EGFR ratio (Fig 4A  and 4B). In particular, c-Src inhibition significantly decreased the capability of SP to induce EGFR phosphorylation. Similarly, MMP inhibition affected the phosphorylation of EGFR induced by SP, as did the concomitant inhibition of both c-Src and MMPs, especially with both c-Src and MMPs inhibitor treatment at 15 min point (0.46 fold decrease) compared with point 0 (Fig 4B, right diagram).We also observed that c-Src inhibition significantly decreased the capability of SP to induce EGFR phosphorylation. Similarly, MMP inhibition affected the phosphorylation of EGFR induced by SP, as did the concomitant inhibition of both c-Src and MMPs, especially with both c-Src and MMPs inhibitor treatment at 15 min point (0.46 fold decrease) compared with the point 0, right diagram ( Fig 4B). As before, the inhibition of c-Src and MMPs in this cell line did not block MAPK signaling due to its modulation by both NK-1R and RTKs (Fig 4A and 4C).
Taken together, these data demonstrate that SP-mediated HER2 and EGFR activation is a c-Src and MMP-dependent process in BC cells.
The inhibition of c-Src, MMPs and NK-1R decreases cell viability and migration of breast cancer cells To study the role of c-Src, MMPs, and NK-1R in cell viability and migration capacities, we treated the HER2+ SK-BR-3 and EGFR+ MDA-MB-468 cell lines with NK-1R antagonist L-733,060, c-Src inhibitor 4-(4 0 -phenoxyanilino)-6,7-dimethoxyquinazoline and MMP inhibitor 1-10, phenanthroline monohydrate. Cell viability was significantly decreased, above all under NK-1R antagonist in both cell lines (Fig 5A). Inhibition of c-Src and MMPs activity also significantly decreased cell viability and was almost completely abolished after the combination of each drug with NK-1R antagonist and after triple inhibition.
Of particular note, the migration rate of MDA-MB-468 significantly decreased under c-Src inhibitor, L-733-060 antagonist and after the combination of c-Src and MMPs inhibitor ( Fig 5B); however, only L-733-060 significantly decreased the migration rate of SK-BR-3 cells. This finding suggests that the cells' migration capacity was partially mediated through c-Src and NK-1R signaling in MDA-MB-468 and mainly by NK-1R signaling in SK-BR-3 cells (Fig 5B).

Discussion
Tachykinins are pro-inflammatory mediators/neuropeptides that contribute to tumor progression by modulating the properties of both cancer and stromal cells. In previous work, we showed that SP contributes to BC progression by modulating the activity of oncogenic receptors like HER2 and EGFR, thus influencing tumor responses to targeted therapies designed to inhibit these receptors [16]. In the present study, we show that SP triggers HER2 and EGFR activation by activating c-Src and MMPs.
The modulation of the steady state of RTKs like HER2 and EGFR by neuropeptides such as SP can influence the clinical response of a tumor [17]. Although the oncogenic addiction to RTKs is therapeutically exploited for BC treatment, the transmodulation of RTKs by SP and other neuropeptides and pro-inflammatory mediators [42,43] can influence the cancer cell response to RTK inhibitors since it serves as a mechanism for RTK activation in a ligand-independent way [14]. The protein tyrosine kinase c-Src can directly phosphorylate Tyr residues in the kinase domain HER2 [30,32] and the cytoplasmic tail of EGFR [31], allowing the formation of stable homo-or heterocomplexes with other receptors or the binding of scaffold proteins and the activation of signal transduction. In addition, activated RTKs will reciprocally activate c-Src, thereby creating a positive regulatory loop. This overactivation may contribute to the permanent signaling through the RTKs and the maintenance of multiple signaling pathways downstream of the receptor [44]. Then, the transactivation of these receptors by c-Srcdependent mechanisms may contribute to the persistence of RTK-related signaling pathways even in the presence of tyrosine kinase inhibitors or antibodies against extracellular domains of these receptors (Fig 6).
It is known that the c-Src protein is overexpressed in 70% of BC tumors, and that in most of them c-Src is co-expressed with at least one ErbB family member [45]. The finding that the basal activation of HER2 and EGFR depends, in part, on the activity of other additional signaling pathways suggests that these instigator pathways might be used for therapeutic purposes to deregulate the activation of RTKs. We observed that overexpression of NK-1R in a BC cell line increases c-Src phosphorylation at Tyr416 more than 6-fold under the stimulus of SP, in addition to increasing HER2 phosphorylation. On the other hand, chemical inhibition of NK-1R decreases c-Src phosphorylation at Y416 in the BT-474 and SK-BR-3 cell lines and the combination of NK-1R, NK-2R and NK-3R chemical inhibitors strongly decreases c-Src phosphorylation at Y416 in SK-BR-3 (cell line expressing all 3 tachykinin receptors). Thus, the use of c-Src and MMP inhibitors allowed us to demonstrate that the SP-mediated transactivation of HER2 or EGFR depends, in part, on c-Src and MMP signaling pathways in BC cell lines. Moreover, the use of these inhibitors demonstrated that this Src and MMP-dependent signaling is important to the cell viability and migration capacity of HER2+ and EGFR+ cell lines, being more pronounced using NK1-R antagonist, L-733,060 alone or in combination. These results suggest an oncogenic addiction to NK-1R signaling in breast cancer cells, where c-Src and MMPs play an important role, probably due to the transactivation mechanism-dependent process of HER2 and EGFR.
Therefore, the c-Src protein may be crucial not only in the ligand-independent transactivation of RTKs, but probably also in MMP maintenance and activation by triggering cleavage of membrane-anchored ligands. These ligands, once released, would bind to receptors as EGFR [46,47] which could homodimerize or heterodimerize with HER2 as the preferred heterodimerization partner.
In summary, we have shown that c-Src and MMPs are involved in HER2 and EGFR transactivation processes through NK-1R in BC. Therefore, a simultaneous blockade of ERBB receptors and other instigators of c-Src/MMP-induced MAPK activation such as NK-1R may improve treatment responses against the ERBB family of receptors.