Neuropeptide gastrin-releasing peptide induces PI3K/reactive oxygen species–dependent migration in lung adenocarcinoma cells

Abstract

Nerve fibers and neurotransmitters have increasingly been shown to have a role in tumor progression. Gastrin-releasing peptide is a neuropeptide linked to tumor aggressiveness, acting as an autocrine tumor growth factor by binding to its receptor, gastrin-releasing peptide receptor, expressed by many tumors. Although neuropeptides have been previously linked to tumor cell proliferation, more recent studies have uncovered roles for neuropeptides in chemotaxis and metastasis. Understanding the precise roles of such peptides in cancer is crucial to optimizing targeted therapy design. We have previously described that gastrin-releasing peptide acts directly as a chemotactic factor for neutrophils, dependent on PI3K, ERK, and p38. In this study, we investigated roles for gastrin-releasing peptide in lung adenocarcinoma. We asked if gastrin-releasing peptide would act as a proliferative and/or chemotactic stimulus for gastrin-releasing peptide receptor–expressing tumor cells. In A549 cells, a non-small cell lung carcinoma line, the treatment with gastrin-releasing peptide leads to activation of AKT and ERK1/2, and production of reactive oxygen species. Gastrin-releasing peptide induced migration of A549 cells, dependent on gastrin-releasing peptide receptor and PI3K, but not ERK. However, no proliferation was observed in these cells in response to gastrin-releasing peptide, and gastrin-releasing peptide did not promote resistance to treatment with a chemotherapy drug. Our results suggest that, similar to what happens in neutrophils, gastrin-releasing peptide is a migratory, rather than a proliferative, stimulus, for non–small cell lung carcinoma cells, indicating a putative role for gastrin-releasing peptide and gastrin-releasing peptide receptor in metastasis.

Introduction

The nervous system influences biological processes beyond the roles originally described, namely, physiologic and motor regulation. Cancer innervation has been shown to benefit tumor growth and its dissemination.^1–3 Recent associations between sympathetic neural fibers and tumors have been described. The neuropeptide Y is expressed in many tumors and linked to increased vascularization, cancer promotion, and progression.^4,5 Substance P activates oncogenes and is associated with DNA damage. ⁶ Moreover, different studies have linked neural-related factors to the modulation of metastasis, as well as extravasation, colonization, and degradation of base membranes.^7–12

The neuropeptide GRP (gastrin-releasing peptide) has various physiological roles, mainly acting as a neurotransmitter. ¹³ GRP is known to act by interacting with its preferential receptor, gastrin-releasing peptide receptor (GRPR) ¹⁴ which is overexpressed in lung, prostate, breast, stomach, pancreas, and colorectal tumors. ¹⁵ As a result, GRP is usually thought of as a mitogenic factor in cancer; however, little is known about its effects on tumorigenesis, metastasis, and the pathways involved in mediating this process. Interestingly, a GRPR antagonist, RC-3095, was able to induce regression in several types of tumor cells in vitro and in vivo, including glioblastoma,^16,17 small cell lung cancer (SCLC), ¹⁸ pancreatic cancer, ¹⁹ and ovarian cancer. ²⁰

GRP also acts during fetal lung development^21,22 and as an autocrine and paracrine growth factor for other human tissues^23,24 and tumors.^13,15,25 GRP serum levels are elevated in many malignancies especially in SCLC. Pulmonary neuroendocrine cells (PNECs) are the main source of GRP in the lung. ²⁶ In asthma models, GRP is overexpressed in the lung, acting as an immunomodulatory molecule. ²⁷ Furthermore, GRP acts as a chemotactic molecule for neutrophils. ²⁸ To study the relative contribution of GRP to tumor proliferation, survival, and migration, we used the non–small cell lung carcinoma (NSCLC) cell line A549, a type of lung adenocarcinoma cell, the highest mortality related to malignant diseases in the world. Our results report that GRP is a migratory, rather than a mitogenic or pro-survival, stimulus for NSCLC cells.

Materials and methods

Cell culture

Human alveolar epithelial adenocarcinoma (A549) were obtained from ATCC (CCL-185) and maintained in Dulbecco’s Modified Eagles Medium (DMEM; Gibco™ Cat. No. 31600-034) supplemented with 10% fetal bovine serum (FBS; Gibco Cat. No. 16000-044) and 0.1 mg/mL of penicillin–streptomycin antibiotic (Gibco Cat. No.15140122). All cells were incubated at 37°C and 5% CO₂ in a humidified incubator with medium being replaced every 48 h.

Trypan blue exclusion assay

Cells (5 × 10⁴) were starved (0.5% FBS) for 24 h then treated with GRP (50 and 100 nM; Sigma-Aldrich Cat. No. G8022) with 10% FBS, or medium alone for 24 h at 37°C and 5% CO₂, cell proliferation was determined by counting cells that excluded Trypan Blue (Sigma-Aldrich Cat. No. T8154) with a hemocytometer.

Viability assays

A549 cells (1 × 10⁴) were plated on 96-well culture plates until reached 50% confluence and later starved (0.5% FBS) for 24 h. For the proliferation assay, cells were treated with GRP (10 nM and 100 nM) or medium alone for 24 h at 37°C under 5% CO₂. For the viability assay, cells were pretreated with GRP (10, 50, 100, and 500 nM) for 15 min in the incubator. Cisplatin (CDDP; Fauldcispla 10 mg/10 mL from Libbs Farmacêutica) 20 µg/mL was then added to each well. Cells were incubated for 24 and 48 h. Proliferation and viability were determined by the 3-(4,5-dimethylthi-azol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay in each treatment group described above. Briefly, MTT (5 mg/mL; Sigma-Aldrich Cat. No. M2128) solution was added to each well. Cells were incubated at 37°C for 2 h in a humidified incubator with 5% CO₂. The formed formazan crystals were solubilized by the addition of dimethyl sulfoxide (DMSO). The optical density was measured at a wavelength of 570 nm with an ELISA plate reader (Antros Zenyth 340r). In addition, cells were treated as described above and stained by the Fixable Viability Dye (eFluor® 780; eBioscience Cat. No. 65-0865) following eBioscience Viability Staining Protocol C. Samples were acquired by BD FACSCantoII™ and analyzed using Flowjo v10.

Ki67 expression

A549 cells (5 × 10⁴) were plated on 24-well culture plates until reached 50% confluence. Cells were starved (0.5% FBS) for 24 h then treated with GRP (50 and 100 nM) with 10% FBS, or medium alone for 24 h at 37°C and 5% CO₂. Cells were detached from the plate using trypsin (Sigma-Aldrich Cat. No. T4049), centrifuged, and stained for viability (Fixable Viability Dye eFluor 780) and Ki67 expression (anti-Ki67 eFluor® 710) following eBioscience protocol for Staining Intracellular Antigens for Flow Cytometry (Protocol B) using Foxp3/Transcription Factor Staining Buffer Set (eBioscience Cat. No. 00-5523). Samples were acquired by BD FACSCantoII and analyzed using Flowjo v10.

Propidium iodide cell cycle analysis

A549 cells (5 × 10⁴) were plated on 24-well culture plates until reached 50% confluence. Cells were starved (0.5% FBS) for 24 h then treated with GRP (50 and 100 nM) or RC-3095 (1 µM) with 10% FBS, or medium alone for 24 h at 37°C and 5% CO₂. Cells were detached from the plate and fixed with 70% ethanol for 1 h on ice. Then, cells were washed, centrifuged twice, and 50 µL of RNase A stock solution (10 mg/mL; Sigma-Aldrich Cat. No. R4875) was added to each tube for 15 min at 37°C. After incubation, DNA was stained with 10 µg of Propidium Iodide (PI) Staining Solution (BD Pharmingen™ Cat. No. 556463) for 30 min at room temperature. Samples were acquired by BD FACSCantoII and analyzed using Flowjo v10.

BrdU assay

A549 cells (5 × 10⁴) were plated on 24-well culture plates until reached 50% confluence. Cells were starved (0.5% FBS) for 24 h then treated with GRP (50 and 100 nM) with 10% FBS, or medium alone for 24 h at 37°C and 5% CO₂. Cell cycle analysis was performed using the APC BrdU Flow Kit (BD Pharmingen™ Cat. No. 552598) and protocol was performed according to the manufacturer’s instruction manual. Data were acquired by BD FACSCantoII and analyzed using Flowjo v10.

Clonogenic assay

This assay was performed according to Franken et al. ²⁹ Briefly, 100 cells (A549) were plated on six-well plates and let adhered for approximately 4 h. Cells were washed, then treated with GRP (50 and 100 nM) or medium alone (both 10% FBS) for 15 days in 37°C and 5% CO₂. On the 15th day, cells were washed three times with cold phosphate-buffered saline (PBS) 1× and fixed with freezer-cold methanol for 10 min. Colonies were stained with 1 mL of crystal violet for 10 min, and the excess were washed with running water. Only colonies with 50 cells or more were counted.

Expression of phospho-Akt, phospho-Erk1/2, and phospho-P38

A549 cells (5 × 10⁴) were plated on 24-well culture plates until reached 50% confluence. Cells were starved (0.5% FBS) for 24 h and then treated with GRP 50 nM in DMEM 0% FBS or medium alone for 5 min in 37°C and 5% CO₂. The expression of phospho-AKT, phospho-ERK1/2, and phospho-p38 was measured by flow cytometry following BD Bioscience Phosflow Protocol for Adherent Cells using trypsin to detach cells. Briefly, cells were fixed in pre-warmed Phosflow Buffer I (BD Phosflow™ Cat. No. 557870) for 10 min at 37°C. After washing, permeabilization was performed with Phosflow Perm Buffer II (BD Phosflow™ Cat. No. 558050) for 30 min on ice. Cells were washed twice and stained with phycoerithryn (PE) anti-AKT (pT308; BD Phosflow™ Cat. No. 558275), PE-Cy™7 anti-p38 (pT180/pY182; BD Phosflow™ Cat. No. 560241), and FITC anti-ERK1/2 (BD Phosflow™ Cat. No. 612592) for 30 min on ice. Data were acquired by BD FACSCantoII and analyzed using Flowjo v.10.

Intracellular reactive oxygen species generation assay

For the determination of intracellular reactive oxygen species (ROS) generation, two different probes were used: 5-(and-6)-chloromethyl-2′,7′-dichlorodihydroflurescein diacetate, acetyl ester (CM-H₂DCFDA; Molecular Probes™ Cat. No. C6827) and CellROX® Deep Red Reagent (Molecular Probes Cat. No. C10422). Both are nonfluorescent while in a reduced state and, upon oxidation, yield intracellular fluorescent compounds. For the first probe, cells (1 × 10⁵) were plated on 24-well culture plates for 24 h. Medium was removed and cells were stimulated for 1 h at 37°C and 5% CO₂ with GRP (50 and 100 nM) and H₂0₂ (0.03%) in 500 µL of serum-free DMEM. Afterward, cells were detached, centrifuged at 300g for 5 min, and incubated in a 2.5 µM CM-H₂DCFDA solution for 1 h at 37°C and 5% CO₂. Cytosolic ROS production was measured by FACSCantoII (Beckon Dickinson) and analyzed by FlowJo v10. As for the later, 10³ A549 cells were plated on a cover slide for 24 h in DMEM 10% FBS. Cells were treated equally to the previous method. CellROX® Deep Red Reagent was added at a final concentration of 5 µM to cells and then incubated for 30 min at 37°C. Confocal images were taken in a Zeiss LSM 5 Exciter microscope.

Wound healing assay

A549 cells (2 × 10⁶) were seeded on 6-well plates DMEM 10% FBS for 24 h or until reached 90% confluence at 37°C under 5% CO₂. Subsequently, cells were starved (0.5% FBS) for 24 h. Then, using a p200 pipette tip, a straight line was scraped into the well to mimic an injury. Cells were washed three times with PBS to remove the debris and treated with GRP 50 nM in 10% FBS, or medium alone and allowed to migrate for 48 h. Cultures were photographed at day zero, 24 and 48 h via an inverted optical microscope with a capturing digital image system (Olympus IX51), always in the same field. Photographs were further analyzed by Adobe® Photoshop® CS5.

Transwell chemotaxis assay

Protocol described by Shi et al. ³⁰ was followed with minor modifications for the Transwell® system (Corning Cat. No. 3464). Briefly, A549 cells (2 × 10⁶) were seeded on a six-well plate and starved (0.5% FBS) for 24 h at 37°C under 5% CO₂. Cells were then suspended in DMEM 10% FBS (2 × 10⁵ cells/200 µL) and added to top wells of 8 µm-Transwell® inserts. GRP was added to the bottom wells at 50 nM containing 10% FBS, or medium alone with 10% FBS. Cells were incubated for 8 h at 37°C under 5% CO₂. To evaluate the involvement of ROS, ERK, PI3K, and GRPR on GRP-induced cell migration, cells were pretreated in suspension with selective inhibitors N-acetyl-l-cysteine (NAC; Fluimucil®), PD98059 (Cell Signaling Technology® Cat. No. 9900), LY294002 (Cell Signaling Technology® Cat. No. 9901), and RC-3095 (provided by Cristália Ltda), respectively, at 37°C under 5% CO₂ for 1 h. At last, inserts were removed from the plate; cells on the bottom were fixed in the membrane and stained with 4′,6-diamidino-2-phenylindole (DAPI; 1:50; Molecular Probes Cat. No. D1306) for 5 min. Using a scalpel, membranes were removed and placed on a glass slide in a mounting medium and covered with a coverslip. Slides were examined at 200× magnification using an optical microscope Olympus BX41 and six random fields were captured and further analyzed using ImageJ software (NIH) to perform cell count.

Statistical analysis

Data are presented as mean ± standard error of mean (SEM). Results were analyzed using GraphPad Prism 6. Statistical differences among the experimental groups were evaluated by analysis of variance (ANOVA) with Tukey correction or with Student’s t test. The level of significance was set at p < 0.05.

Results

GRP activates signaling pathways in NSCLC

GRPR is the most commonly expressed GRP receptor in patients with lung cancer, ³¹ which is of particular interest considering that adenocarcinoma expresses higher levels of this receptor than all other histological types. ³² To further investigate NSCLC responses to GRP in in vitro models, we selected the A549 cell line for its high expression of GRPR. ³³ This tumor metastasizes primarily to the liver as well as in bone, adrenal glands, and brain, and yet, little is known about the biology of this disease. Previous studies investigating the effect of GRP on NSCLC cell lines have shown that GRP stimulates AKT pathway. ³³ GRP also induces rapid activation of p44/42 MAPK in lung cancer cells through epidermal growth factor receptor (EGFR). ³⁴ When we stimulated cells with GRP and evaluated AKT, p38, and ERK1/2 phosphorylation, we observed not only AKT activation (Figure 1(a)) but also ERK1/2 (Figure 1(b)). No changes were seen in p38 phosphorylation (Figure 1(c)). Because these proteins are involved in several cell-signaling pathways that regulate not only proliferation but also survival and migration, we sought to investigate these outcomes in A549 cells following GRP treatment.

Figure 1.

GRP induces AKT/MAPKs phosphorylation. A549 cells were stimulated with GRP (50 nM) for 5 min and (a) AKT, (b) ERK1/2, or (c) p38 activation patterns were monitored by Phosflow analysis. Number of counted cells and expression of pAKT, pERK1/2, and p-p38 is shown in the histograms. Isotype control (tinted histogram) and medium only (black histogram line) were used as negative controls. Data are representative from two independent experiments performed in triplicates.

GRP does not stimulate proliferation or survival of NSCLC cells

To evaluate the effect of GRP on lung cancer cell proliferation, we exposed cells to different concentrations of GRP, and analyzed proliferation/survival by Trypan Blue exclusion and MTT assays. No differences in cell counts were observed compared to medium only, in any of the concentrations used, 24 (Figure 2(a)) or 48 h (Figure 2(b)), or in the MTT assay (Figure 2(c)). In addition, a clonogenic assay did not show any increase in colony formation after treatment with GRP (50 and 100 nM; Figure 2(d)). Cells were also analyzed for expression of proliferation marker Ki67 (Figure 2(e) and (f)), cell cycle accumulation (Figure 2(g)), and the ability to incorporate bromodeoxyuridine (BrdU; Figure 2(i)). Results of Ki67 and cell cycle assays were similar to the previous assays.

Figure 2.

Lack of proliferative effect of GRP on A549 cells. (a and b) A549 cells were treated with GRP 50 or 100 nM in 10% of FBS medium for (a) 24 or (b) 48 h. Cell count was performed using a hemocytometer and Trypan Blue exclusion. Total number (mean ± SEM) of counted cells per group is shown. (c) MTT proliferation assay was performed with treated-cells in two concentrations of GRP (10 and 100 nM) in 10% of FBS for 24 h. Data (mean ± SEM) are expressed as percentage of an untreated control. (d) Clonogenic assay of A549 cells treated with GRP (50 or 100 nM). Number of colonies per well was counted 15 days after stimuli. Data (mean ± SEM) are representative from three independent experiments performed in triplicates. (e and f) Ki67 expression (% of Ki67 + cells) of A549 treated with GRP (50 and 100 nM) in 10% FBS medium after (e) 24 or (f) 48 h. (g) PI–cell cycle analysis of A549 treated with GRP (50 and 100 nM). Percentage of G0/G1 (white bars), S (gray bars), and G2/M (black bars) cell cycle stages are shown for cells cultured for 24 h. (h and i) BrdU incorporation assay of A549 treated with GRP (50 and 100 nM). (h) Representative FACS plots show gate strategy for BrdU/7-AAD staining. (i) Percentage of G0/G1 (white bars), S (gray bars), and G2/M (black bars) cell cycle stages are shown for cells cultured for 24 h. Data (mean ± SEM) are representative from three independent experiments performed in triplicates. Statistics: one-way ANOVA with Tukey post-test or Student’s t test (p < 0.05).

Because previous studies have seen a decrease in GRP efficiency in the presence of serum, ³⁵ all experiments were performed with either 0.5% or 10% of serum content during stimulation (data on 0.5% FBS experiments are presented on Supplementary Figure 1). As can be observed, results were not different when performed in 0.5% or 10% of FBS during experiments (Figure 2 and Supplementary Figure 1). Especially, proliferation, which can be affected by growth factors present in serum, did not differ in these two culture conditions. Altogether, our results indicated that GRP does not stimulate A549 cell proliferation.

To investigate whether the signaling pathways activated by GRP were related to improving survival in NSCLC, we pretreated A549 cells with GRP and incubated them with cisplatin (CDDP) for 24 and 48 h. No difference was observed in cell survival by MTT and viability staining assays in any of the GRP concentrations tested in 24 (Figure 3(a) and (c)) or 48 h (Figure 3(b) and (d)). These findings indicate that although GRP activates proteins involved in survival pathways, it is does not prevent cells from death by CDDP.

Figure 3.

Pre-treatment with GRP does not increase cell survival upon challenge with cisplatin. (a and b) MTT survival assay of A549 cells treated with GRP (10, 50, 100, and 500 nM) for 15 min prior to CDDP (20 µg/mL) for (a) 24 or (b) 48 h. Data (mean ± SEM) are representative from two independent experiments performed in quadruplicates and are expressed as percentage of an untreated control. (c and d) Cells cultured in 10% FBS medium were then treated with GRP (50 and 100 nM) for (c) 24 or (d) 48 h and stained by the Fixable Viability Dye (1:1000) for 30 min. Cell viability was analyzed by flow cytometry. Data (mean ± SEM) are representative from two independent experiments performed in triplicates. Statistics: one-way ANOVA with Tukey post-test (*p < 0.05).

GRP induces ROS production

Elevated rates of ROS have been detected in almost all cancers, where they promote many aspects of tumor development and progression. ROS activate proteins involved in cell proliferation, migration, and survival. ³⁶ Here, we report for the first time that GRP stimulates ROS production on A549 cells (Figure 4(a) and (b)) when incubated with both CellROX Deep Red reagent dye and CM-H₂DCFDA probe, respectively.

Figure 4.

GRP induces ROS production in A549 cells. (a) 10³ cells were stimulated with GRP (50 and 100 nM) for 1 h and incubated with 5 µM of CellROX Deep Red Reagent for 30 min. Confocal images were taken in a Zeiss LSM 5 Exciter microscope. Data are representative of one experiment. (b) Cells were stimulated with GRP 50 and 100 nM for 1 h and incubated with 2.5 µM of CM-H2DCFDA for 1 h. ROS generation was analyzed by flow cytometry using FACSCantoII. Data are representative of two independent experiments performed in triplicates with similar results. Statistics: Student’s t test, ***p < 0.001.

GRP induces cell migration

It is known that morphogenesis is an important step for cellular motility during the development of invasive nature of many tumors. Recent data suggest that GRP serves primarily as a morphogen and not as a mitogen as usually viewed.^35,37 In order to evaluate the migration potential of GRP on the A549 cell line, we used two established migration assays. In the wound healing assay, we found a significant difference when treating cells with GRP in 24 and 48 h after the wound compared to non-treated cells (Figure 5(a) and (b)). However, some argue this assay does not evaluate migration alone and the effect seen could be due to cell proliferation, which was dismissed in our previous experiments (Figure 2). To support our data, we also performed the transwell assay. We found that GRP 50 nM was able to induce cell migration (Figure 5(c) and (d)). To investigate if GRP-induced cell migration is dependent on ROS, PI3K, ERK, or GRPR activation, cells were pretreated with selective inhibitors, or RC-3095. Pretreating A549 cells with NAC (Figure 5(e)), PI3K inhibitor, or RC-3095 (Figure 5(f)) significantly decreased the migration toward GRP. To discard a possible cytotoxic role for the GRPR antagonist, RC-3095, which could bias our result, we performed some toxicity tests. RC-3095 at 10 µM had no significant cytotoxic effect or in any lower concentration tested (not shown). Altogether, these results demonstrate that the rapid activation of AKT/MAPK pathways via GRPR induces cell migration, rather than proliferation or survival, in A549 cells.

Figure 5.

GRP promotes A549 cell migration. (a and b) A549 cells stimulated with GRP (50 nM) in 10% FBS medium were subject to wound healing assay and stained with hematoxylin (blue area). (a) Representative images at 0 (left column), 24 (middle column), or 48 h (right column) after treatment. Microscopy images with 200× magnification. (b) Image quantification of the wounded area size. Migration is expressed as percentage of coverage of the starting wounded area (set as 0%). Nine pictures per sample were taken and analyzed. Data (mean + SEM) representative of three independent experiments performed in triplicates. Statistics: Student’s t test, *p < 0.05, ***p < 0.001. (c)–(f) A549 cells were starved and placed in the upper side of transwell migration chambers and medium alone (as control) or medium plus GRP 50 nM with 10% FBS were added to the lower chamber. Transwell membranes were stained with DAPI and six pictures per sample were taken. (c) Representative microscopy images (at 200× of magnification) of transwell migration assay (quantified in d). (e and f) Cells were pretreated with selective inhibitors of ERK, PI3K, GRPR, and ROS (PD98059, LY294002, RC-3095, and NAC, respectively). (d)–(f) Quantification of cell number per field of each group is shown. Data (mean + SEM) representative of two independent experiments performed in triplicates. Statistics: one-way ANOVA with Tukey post-test (p < 0.05).

Discussion

Tumors often resemble integrated organs, being composed by different cells, which work coordinately. Although both central and sympathetic nerve elements are present in different types of tumors, ³⁸ little is known regarding the roles they play. Sympathetic nerve fibers that reach tumors can release neurotransmitters and neuropeptides locally, ¹² activating signaling pathways that can facilitate tumor growth.^7,8

There are two basic types of lung cancer. SCLCs comprise about 15%–20% of all cases, while NSCLC is related to 80%–85% of lung cancer cases. Based on histology, NSCLCs are divided in squamous cell carcinoma, large cell carcinoma, or adenocarcinoma, the latter being the most common type of lung cancer. ³⁹ SCLCs express high levels of GRP, ⁴⁰ which has been shown to influence tumor aggressiveness.^25,41 However, NSCLC does not usually produce GRP^42,43 and NSCLC patients do not have high plasma levels of GRP.^44,45 The cell line selected for this study (A549) is derived from NSCLC adenocarcinoma, and we did not find any GRP production in A549 supernatants (not shown). However, we found a significant ROS- and PI3K-dependent effect for GRP over A549 migratory activity mediated by GRPR.

NSCLC are higher GRPR expressers ³² than SCLCs, indicating that this receptor could have an important role in this tumor’s biology. One possibility is that nerve fibers infiltrating the tumor secrete GRP, influencing metabolic processes that contribute to tumor progression. ⁷ GRP is widely expressed in the brain, spinal cord, and nerve fibers. ⁴⁶ Moreover, many studies have shown that GRP/GRPR are involved in the mediation of itch, fear memory consolidation, anxiety, and stress.^47–49 The relationship between chronic stress and cancer is well established in the literature.^9,11 Therefore, chronic stress may influence cancer cells by the release of GRP, further activating the sympathetic nerve system and the hypothalamic–pituitary–adrenal axis through the release of stress hormones. ³⁸

Another possibility is that GRP is provided to tumors by PNECs. PNECs are resident lung cells and have been suggested to be the cell of origin for some SCLCs.^50,51 Lung adenocarcinoma is the most frequent type of cancer in smokers. ⁵² GRP expression is increased in PNECs during hypoxia ⁵³ and smoking,^54,55 providing a candidate link for stress-induced lung cancers. Therefore, it is possible that NSCLCs use GRP produced by PNECs to stimulate migration, promoting tumor aggressiveness. Nicotine and other tobacco toxins activate PNECs; ⁵⁶ thus, it could be hypothesized that GRP is connected to lung cancer promotion in smokers.

In our study, we found that both AKT and ERK1/2 were activated after stimulation with GRP. These proteins are involved not only in cell proliferation and migration but also in survival. Previous studies investigating the effect of GRP on NSCLC cell lines showed that GRP stimulates AKT activation primarily via c-Src activation, followed by extracellular release of the EGFR ligand amphiregulin, leading to the activation of EGFR and PI3K. Also, GRP induces rapid activation of p44/42 MAPK in lung cancer cells through EGFR. ³⁴ Nevertheless, no previous study has reported ERK1/2 activation on a non-small cell lung carcinoma cell line. Although AKT and ERK1/2 were activated following GRP treatment, we did not observe any increase on cell proliferation in any of the assays performed, and it was not influenced by serum content (0.5% or 10% FBS). These results are consistent with other studies, which reported that even without the presence of serum in the medium, GRPR activation has an inconsistent, and only modest, effect on tumor growth, leading to the hypothesis that the GRP acts as a weak mitogen on some tumors. ³⁵

Liu et al. ³³ found that GRP rescues A549 cells exposed to gefitinib, an EGFR inhibitor. In order to verify this survival effect, we investigated if GRP could rescue cells from cytotoxic effects of cisplatin, which is still one of the main treatments in NSCLC cases. ⁵⁷ However, no difference was observed, possibly due to the mechanism involved in cisplatin-mediated cytotoxicity, which induces apoptosis differently from EGFR inhibitors. Even though GRP activated survival signals, it was not enough to protect cells from CDDP. Consequently, in tumor cells that do not constitutively produce GRP, even though GRP may be provided paracrinally by local nerve cells, that signal alone might not be enough to protect from drug-induced cell death.

Early studies suggested that GRP could play a role on NSCLC migration. ⁵⁸ Jensen ³⁵ proposed that GRP would rather act as a morphogen. More recent evidence indicated that Caco-2 and HT-29 colon cancer cell lines had increased cell motility when treated with bombesin. ³⁷ A different study demonstrated that motility of the same cell lines was mediated by GRP via upregulation of ICAM-1 via FAK. ⁵⁹ Others have observed a morphological change in BE(2)-C neuroblastomas with high expression of GRPR, and GRPR knockdown inhibited the metastatic potential both in vitro and in vivo. ⁶⁰ Our results are in agreement with these findings, further elucidating the mechanisms underlying the migratory stimulation potential of GRP for lung cancer cells.

Finally, this is the first study to demonstrate that the treatment with GRP induces ROS generation in lung adenocarcinoma cells, which can be related to PI3K/AKT activation that leads to cell migration. ROS are known to participate directly or as a second messenger in promoting lung carcinogenesis, ⁶¹ facilitating mutagenesis, cell proliferation, survival, differentiation, and motility.^36,62 Future studies will determine the relative contribution of this pathway during in vivo progression and metastatic activity of NSCLC.