Water-soluble variant of human Lynx1 induces cell cycle arrest and apoptosis in lung cancer cells via modulation of α7 nicotinic acetylcholine receptors

Lynx1 is the first three-finger prototoxin found in the mammalian central nervous system. It is a GPI-anchored protein modulating nicotinic acetylcholine receptors (nAChRs) in the brain. Besides the brain, the Lynx1 protein was found in the lung and kidney. Endogenous Lynx1 controls the nicotine-induced up-regulation of the expression of α7 type nAChRs in lung adenocarcinoma A549 cells as well as the cell growth. Here, we analyzed the Lynx1 expression in the set of human epithelial cells. The Lynx1 expression both at the mRNA and protein level was detected in normal oral keratinocytes, and lung, colon, epidermal, and breast cancer cells, but not in embryonic kidney cells. Co-localization of Lynx1 with α7-nAChRs was revealed in a cell membrane for lung adenocarcinoma A549 and colon carcinoma HT-29 cells, but not for breast adenocarcinoma MCF-7 and epidermoid carcinoma A431 cells. The recombinant water-soluble variant of Lynx1 without a GPI-anchor (ws-Lynx1) inhibited the growth of A549 cells causing cell cycle arrest via modulation of α7-nAChRs and activation of different intracellular signaling cascades, including PKC/IP3, MAP/ERK, p38, and JNK pathways. A549 cells treatment with ws-Lynx1 resulted in phosphorylation of the proapoptotic tumor suppressor protein p53 and different kinases participated in the regulation of gene transcription, cell growth, adhesion, and differentiation. Externalization of phosphatidylserine, an early apoptosis marker, observed by flow cytometry, confirmed the induction of apoptosis in A549 cells upon the ws-Lynx1 treatment. Our data revealed the ability of ws-Lynx1 to regulate homeostasis of epithelial cancer cells.


Introduction
Nicotinic acetylcholine receptors (nAChRs) are ligand-gated ion channels responsible for signal transduction in the central and peripheral nervous systems and in the neuromuscular PLOS  Six adult male Wistar rats (230-283 g) were obtained from the Puschino Breeding Center (Moscow region, Russia) and housed 2 animals per cage. Animals were kept on a 12:12-hour light:dark cycle provided with a standard rodent diet and water ad libitum. The animals were acclimatized for at least 7 days after arrival before experiments began. They were anesthetized with isoflurane and then decapitated. The brain was dissected and stored at -70˚C. For the primary culture of cortical neurons newborn Wistar rats were used. The study was performed in accordance with the guidelines set forth by the European Communities Council Directive of November 24, 1986 (86/609/EEC). The study was approved by the institutional commission of the Shemyakin-Ovchinnikov Institute of Bioorganic Chemistry RAS for the control of the maintenance and use of animals (protocol #222 from 13 February 2018).

Cell cultivation
Human epidermoid carcinoma cells A431 (ATCC, USA), human breast adenocarcinoma cells MCF-7 (ATCC, USA), human lung adenocarcinoma A549 cells (ATCC, USA), and HEK-293T cells (Institute of Cytology RAS, Russia) were grown (37˚C, 5% CO 2 ) in a DME medium with phenol red (PanEco, Russia), 10% fetal calf serum (Thermo Fisher Scientific, USA) and 2 mM L-glutamine (PanEco), abbreviated below as the complete medium. Human colon carcinoma HT-29 cells (ATCC) were grown in a RPMI-1640 medium, supplemented with 10% fetal calf serum and 2 mM L-glutamine. Human immortalized oral Het-1A keratinocytes (ATCC) were cultivated in a BEB medium (Lonza, Switzerland). Before cell subculturing and performing experiments the culture flasks and plates were pre-coated with a mixture of 0.01 g/ L fibronectin (Sigma-Aldrich), 0.03 g/L bovine collagen type I (Sigma-Aldrich) and 0.01 g/L bovine serum albumin (Sigma-Aldrich) dissolved in the culture medium. All types of the cells were subcultured twice per week. The primary culture of cortical neurons were obtained as previously described [31]. Neurons were cultured in a Neurobasal-A medium (Gibco, USA) supplemented with Glutamax (Life Technologies, USA) and B27 (Gibco) (so called complete medium) on polylysine-coated cover glasses in 24-well plates for two days, then Cytarabin (20 μM) (Sigma-Aldrich) was added to inhibit growth of glial cells and the culture was cultivated for 12 days with media change every 4 days and then taken to experiments.

Real-time PCR
To extract total mRNA from the whole rat brain extract, 1 ml of ExtractRNA reagent (Evrogen, Russia) was added to the frozen rat brain, and then the brain was rapidly homogenized. Total mRNA from cultured cells was also extracted using ExtractRNA reagent according to manufacturer's instructions. Then mRNA from the rat brain or cultured cells was treated by DNAse I (Sigma-Aldrich) and purified using CleanRNA Standard kit (Evrogen). Total cDNA was synthesized using MMLV RT kit (Evrogen) in accordance to the manufacturer's protocol. After that real-time PCR was performed with the primers described in the S1 Table, and ready-to-use qPCR mix with SYBR Green I and reference ROX fluorescent dyes (Evrogen). Briefly, 12.5 μL of the diluted cDNA sample produced from 1 μg of total RNA was added to 7.5 μL of the PCR master mix and 5 μL of 5× qPCR mix-HS SYBRHigh ROX (Evrogen). Negative controls contained all the components of the PCR mixture except cDNA and gave no signal. All PCR reactions were performed using DT96 Prime real-time detection thermal cycler (DNA Technology, Russia). Data was analyzed by theΔCt method [32]. Expression level of the nAChR subunits was normalized to the expression level of the β-actin house-keeping gene.
The cells stained only by the secondary antibodies are presented in S1 and S2 Figs. Specificity of anti-Lynx1 antibodies was analyzed using ws-Lynx1. Briefly, 6.7 nM of antibodies were incubated with 1 μM ws-Lynx1 for 1 h, and then A549 cells were stained with this mixture and the primary mouse anti-α7-nAChR antibodies. Then, the cells were stained by the secondary anti-rabbit AlexaFluor 488 labeled IgG and anti-mouse Alexa-594-labelled IgG as described above (S3 Fig).

Growth of A549 cells
To study the effect of ws-Lynx1 on the A549 cell growth, the cells were seeded in 96-well cell culture plates in the complete medium (0.5×10 4 cells/well) and grown for 24 h. Thereafter ws-Lynx1 from the 100% DMSO stock solution was diluted in the complete medium and added to the cells at different concentrations for further incubation during 24, 48 or 72 h. Every 24 h the cell medium was aspirated and replaced by the fresh complete medium, containing pre-dissolved ws-Lynx1 in the same concentration as in the first day. The final DMSO concentration did not exceed 0.1%. The added DMSO did not influence the cell growth as was checked in additional experiments.
Cell growth was assayed using the WST-1 colorimetric test as described elsewhere [33]. WST-1 (water soluble tetrazolium salt 1, Santa Cruz) and 1-m-PMS (1-methoxy-5-methylphenazinium methyl sulfate, Santa Cruz) were added to the cells in concentrations of 0.25 mM and 5 μM, respectively, for 2 h, and the formation of colored product was measured in the 450 nm spectral range using microplate reader Bio-Rad 680 (Bio-Rad, USA).

Cell cycle arrest in A549 cells
Cells we seeded in 6-well culture plates (12.5×10 3 cells per well) and incubated with 1 μM ws-Lynx1 for 24, 48, or 72 h as described above. Then the cells were detached from the wells by trypsine, washed with EBSS, and fixed in ice-cold 70% ethanol for 4 h. After fixation, the cells were washed twice by EBSS, and DNA was extracted by 5 min incubation with the DNA extraction buffer (200 mM Na 2 HPO 4 with 0.004% Triton X-100, pH 7.8). Then the cells were washed by EBSS, resuspended in the DNA staining solution (EBSS, 50 mg/ml propidium iodide, 0.2 mg/ml DNase free RNAse), and analyzed by FACSCalibur flow cytometer (Becton Dickinson, USA). The data were analyzed using Flowing 2.5.1 software (Turku Centre for Biotechnology, Finland).

Silencing of α7-nAChR
To block the expression of native α7 receptors, A549 cells were transfected with siRNA to α7-nAChRs (siRNA duplex was formed by GGAAGCUUUACAAGGAGCUGGUCAA and UUGACCAGCUCCUUGUAAAGCUUCC synthetic oligonucleotides (Synthol, Russia), [34]). Cells were seeded in a 6-well culture plates (1×10 5 cells per well) and grown for 24 h. Then siRNA (1 μg per well) was diluted in 100 μl transfection buffer (Pan-Biotech, Germany), incubated for 5 min and mixed with 15 μl pre-diluted PanFect A-plus transfection reagent (Pan-Biotech, Germany). The final mixture was incubated for 30 min and added to the A549 cells. The cells were incubated in CO 2 -incubator during 4 h and the cell media was replaced by the fresh one. After the 48 h incubation, the cells were detached by Versene solution and divided onto the two parts. The first part was incubated with TRITC-labelled α-bungarotoxin (Sigma, USA), and the expression of functional α7-nAChRs on the cell membrane was analyzed by flow cytometry. The second part of the cells was seeded on 96-well culture plates (5×10 4 cells per well) and incubated with 1 μM ws-Lynx1 for 72 h as described above. Cell viability was analyzed by the WST-1 assay.

Kinase phosphorylation upon ws-Lynx1 application
To investigate the Lynx1 influence on the cellular signal transduction in A549 cells, we used a Proteome Profiler antibody array (ARY003B, R&D Systems, USA), which allowed a simultaneous detection of phosphorylation of different kinases. The cells were seeded in T25 cell culture flasks and incubated with 1 μM ws-Lynx1 (from the 100% DMSO stock) or an equal DMSO amount in control flasks. The cells were grown for 72 h as described above. After the 72 h incubation the cells were detached by scrapping and lysed by 400 μl of Lysis buffer, provided with the Proteome Profiler antibody array. After the lysate centrifugation, the supernatants were collected, and the human phospho-kinase array was performed according to the manufacturer protocol. Briefly, the array membranes with immobilized capture antibodies were blocked, incubated with the cell lysates overnight and then consequently incubated with biotinylated detection antibodies and streptavidin-HRP. The HRP signal was detected by ECL substrate using Bio-Rad Versa-Doc 4000 imager (Bio-Rad). The spots intensity was quantified by Image J software (NIH).

Study of the A549 cells apoptosis by flow cytometry
To investigate apoptosis in A549 cells, we used Annexin V for detection of the phosphatidylserine externalization, one of the early apoptosis markers. Briefly, A549 cells were seeded on a 35-mm Petri dish (1×10 5 cells/dish) and incubated with 1 μM ws-Lynx1 for 72 h as described above. After incubation, the cells were detached by the Versene solution and washed in PBS.
Then, the cells were incubated with Annexin V conjugated to phycoerythrin (PE) (A35111, Thermo Fisher Scientific) and washed by PBS. Finally, the cells were analyzed on BD FACSCalibur flow cytometer (Becton Dickinson). The data were analyzed using Flowing 2.5.1 software.

Expression of Lynx1 and nAChR genes in human non-neuronal cells
In our work, we used several lines of epithelial cells. The rat brain extract was used as a positive control. Using real-time PCR, we found the different expression profile of the α3, α4, α7, α9, and β2 nAChR subunits genes in the studied cells (Fig 1).
In general, our data agree well with the previous reports, although some new details were revealed. Similarly to the known localization of different nAChRs subunits in the mammal brain [35], we observed the expression of all studied subunits genes (except α9) in the rat brain extract. The gene expression of the α7, α9, and β4 nAChR subunits [36] and the protein expression of α9 nAChR subunit [37] in breast adenocarcinoma MCF-7 cells were previously reported. The Human Protein Atlas (HPA) database also describes expression of the α3 gene in this cell line, while our study also detected the expression of α4 and β2 genes. Previously, the analysis of protein expression in lung adenocarcinoma A549 cells revealed the presence of the α3, α4, α7, α9, β2, and β4 nAChR subunits [38]. Contrary to that, we did not observe the expression of the α9 nAChR subunit. Colorectal carcinoma HT-29 cells were demonstrated the expression of the α4, α5, α7, and β1 genes [39], while we also found the expression of the α9 nAChR subunit. It was previously shown that epidermoid carcinoma A431 cells express the α7 nAChR gene [40], and the expression of the α3 gene was reported in the HPA database, while we additionally detected the expression of the α4, α9, and β2 genes. We found the gene expression of the α3, α4, α7, α9 and β2 nAChR subunits in Het-1A cells that corresponds to the previous study [3]. Similarly to the data from the HPA database, in embryonic kidney HEK-293T cells we observed the expression only of the α3 subunit. It should be mentioned, that the expression of nAChR subunits at the mRNA level, could not correlate with the expression of the functional receptors composed from these subunits on a cellular membrane. Thus, the observed gene expression of the nAChR subunits does not indicate the real expression of the corresponding receptors.
All the cells studied, except HEK-293T, demonstrated the significant level of the LYNX1 expression (Fig 1). It contradicts the previous data about the LYNX1 expression in the kidney [9]. Probably, the LYNX1 expression depends on the stage of the development and rises after birth. In agreement with this hypothesis, no Lynx1 expression was found in the rat cortex and hippocampus up to the 14th postnatal day [20], and the significant increase of the Lynx1 mRNA level in the mouse cerebellum was observed from P0 to P20 days [41]. Similarly, LYNX1 is absent in the fetal breast, skin, and colon tissues, but is expressed in adult tissues (68 FANTOM5 RNA-seq dataset, https://www.ebi.ac.uk/gxa).
According to the HPA database, the LYNX1 expression was reported in epidermoid carcinoma A431 and lung adenocarcinoma A549 cells, and in HACAT immortalized keratinocytes, however it was not detected in breast adenocarcinoma MCF-7 cells. Probably, these discrepancies in the LYNX1 expression according to the different sources are due to the different approaches and lab conditions used for mRNA extraction and detection.

Lynx1 is co-localized with α7-nAChRs in epithelial cells
To prove the Lynx1 and α7 nAChR subunit expression at a protein level, we used confocal fluorescent microscopy. In agreement with the observed gene expression patterns, no Lynx1 and α7-nAChR at the protein level were detected in HEK-293T cells, while in other studied cells the expression of both proteins was observed (Figs 2 and 3). For the best of our knowledge, this is the first report about the Lynx1 expression in colon and breast cancer cells as well as in Het-1A oral keratinocytes.
To study the possible target of Lynx1, we analyzed its co-localization with α7-nAChRs. Primary culture of cortical neurons was used as a positive control. In neurons, α7-nAChRs were expressed in the soma and dendrites, while Lynx1 was observed mainly in the soma (Fig 3). Evident membrane co-localization of Lynx1 and α7-nAChRs was observed only in A549 and HT-29 cells (Fig 3). Similar co-localization of Lynx1 and α7-nAChRs was previously demonstrated in developing lung tissue [24] as well as in adult lung epithelium [26]. Surprisingly, in MCF-7 and A431 cells Lynx1 and α7-nAChRs demonstrated different expression pattern with accumulation in intracellular punctuates and cell membrane, respectively (Fig 3). Our results point on a widespread Lynx1 expression in cells of different origin and indicate that Lynx1 could act not only as the neuromodulator in the nervous system but also could be implicated in physiological processes in epithelial cells.

Ws-Lynx1 down-regulates the viability of lung cancer cells and abolishes nicotine-induced cell growth stimulation
α7-nAChR is the main receptor which mediate the proliferative effects of nicotine in cancer cells [6]. Activation by nicotine could increase the α7-nAChR expression and stimulate growth and migration of cancer cells [6,26,42]. Nicotine is known to enhance expression of hypoxiarelated molecules such as HIF-1a in A549 cells stimulating secretion of different growth factors and their resistance to chemotherapy [42]. Also, activation of α7-nAChRs in endothelial cells by nicotine leads to tumor vascularization through the release of VEGF and FGF [42]. Another aspect of nicotine action on tumor development is its ability to induce self-renewal of cancer stem cells by a simultaneous increase of stress neurotransmitters production and decrease of GABA production [43]. On the other hand, endogenous Lynx1 suppresses the nicotineinduced up-regulation of the α7-nAChR expression [26,44]. Moreover, the up-regulation of the Lynx1 expression in lung carcinoma A549 cells by lentiviral transduction decreases cancer cell proliferation [26]. Based on these observations, we decided to evaluate the ability of the recombinant water-soluble variant of Lynx1 (ws-Lynx1) to control the growth of A549 cells.
Prolonged 72 h incubation of A549 cells with ws-Lynx1 resulted in the pronounced concentration-dependent inhibition of the cell growth with EC 50 of 1.9 ± 0.3 nM (Fig 4A). The number of viable cells was reduced up to 66.7 ± 1.3% relative to the control. Shorter 48 h administration resulted in the weak but significant reduction of the number of viable cells, while no significant effect was observed after 24 h incubation (Fig 4A). It should be mentioned, that similar antiproliferative activity on A549 cells for the recombinant analogue of the other human Ly6/uPAR protein,-SLURP-1, also acting on α7-nAChRs was recently reported [18]. However, recombinant SLURP-1 demonstrated the significant antiproliferative effect already after 24 h incubation [18]. This revealed differences in the mechanisms of the Lynx1 and SLURP-1 action. Probably, the faster SLURP-1 effect is connected with the secretion of endogenous SLURP-1 from the intracellular depot in response to the recombinant SLURP-1 application. This results in the significant increase of the SLURP-1 concentration in the cell medium after several hours [18]. Contrary, endogenous Lynx1 is a membrane-tethered protein, and its concentration in the cell medium cannot be enhanced by secretion. The delayed effect of ws-Lynx1 on the cell viability implies the presence of a multi-step response, which could involve (1) changes in the expression of auto-paracrine regulators like SLURPs, growth factors, cytokines (like TNFα) etc; and (2) activation of the respective receptors and intracellular signaling cascades, which control cell growth and proliferation. Moreover, as it will be shown below, ws-Lynx1 simulates proapoptotic signaling cascades, while no signs of apoptosis were previously found in the SLURP-1 study [18].
We also tested whether ws-Lynx1 could affect the nicotine-induced A549 cell growth ( Fig  4B). Incubation of A549 cells during 72 hours with 10 nM nicotine significantly increased the A549 cell proliferation (~118% of viable cells relative to the control). Contrary, co-application of nicotine with 1 μM ws-Lynx1 canceled this effect, and reduced the number of viable cells up to~79% relative to the control (Fig 4B). The data obtained are in line with the previous report about the action of membrane-tethered Lynx1 in A549 cells [26]. Thus, we can conclude that the water-soluble variant of Lynx1 suppresses the nicotine-induced cancer cell growth in vitro. , and #### (p<0.0001) indicate the significant difference from the control (100%) by a two-tailed one sample t-test; && (p<0.01) indicates the significant difference between the "Nicotine" and "ws-Lynx1 + Nicotine" groups by a two-tailed t-test; � (p<0.05), �� (p<0.01), ��� (p<0.001), and ���� (p<0.0001) indicate the significant difference between the "ws-Lynx1" and "ws-Lynx1 + agonist/inhibitor" groups by One-way ANOVA followed by a Dunnet's post hoc test. The control level (100% of viable cells) corresponds to the untreated cells. https://doi.org/10.1371/journal.pone.0217339.g004 Water-soluble variant of human Lynx1 induces cell cycle arrest and apoptosis in lung cancer cells

Ws-Lynx1 induces cell cycle arrest in A549 cells
To elucidate whether ws-Lynx1 treatment induces the mitotic inhibition during cell division, we performed cell cycle analysis. FACS analysis demonstrated that the 24 h incubation of A549 cells with ws-Lynx1 significantly increased the number of cells in the G0/1 phase of cell cycle, while the number of cells in the S phase was significantly decreased (Fig 5A and 5B). Contrarily, we did not observe any difference in the cell proportion in each cell cycle phase after the 48 h incubation with ws-Lynx1, while the 72 h incubation resulted in the increased number of cells in the G2/M phase compared to the control, that accompanied by the decrease of the G0/1 cell population (Fig 5A and 5B). Thus, we can conclude that ws-Lynx1 induces the G0/1 and G2/M cell cycle arrest in A549 cells after 24 and 72 h incubation, respectively. This observation confirms the presence of the multi-step cellular response to the ws-Lynx1 treatment. The primary response to ws-Lynx1 results in the G0/1 cell cycle arrest during the 24 h incubation preventing cells from entering into the S phase, but after 48 h A549 cells could bypass this cell cycle blockade and start DNA replication. However, during the 72 h incubation, the response to secondary messengers (auto-paracrine regulators, growth factors, or cytokines) results in the G2/M cell cycle arrest, so A549 cells could not divide after successful

Ws-Lynx1 controls the growth of A549 cells by modulation of α7-nAChRs and activation of PKC/IP3 pathway and other signaling cascades
To study the mechanisms underlying the ws-Lynx1 action, we investigated its effect on the growth of A549 cells in the presence of inhibitors of different cell surface receptors: bicuculline (antagonist of the GABA A receptors), gefitinib (inhibitor of the epidermal growth factor receptor EGFR), AG 825 (inhibitor of the ErbB2 receptor), timolol (inhibitor of the β-adrenergic receptors (β-AR)), mecamylamine (Mec, the nonspecific non-competitive antagonist of nAChRs), and methyllycaconitine (MLA, the specific α7-nAChRs antagonist). We used 1 μM ws-Lynx1 concentration, at which the maximal reduction of the cell growth was observed, and inhibitor concentrations equivalent or larger than IC 50 values for the respective receptor. All the inhibitors (except bicuculline) alone did not significantly influence the viability of A549 cells upon 72 h incubation. Co-incubation of the cells with ws-Lynx1 and bicuculline, gefitinib, AG 825, or timolol did not diminish the ws-Lynx1 effect on the A549 cell growth (Fig 4B), while co-application with Mec or MLA canceled the ws-Lynx1 effect restoring the number of viable cells to the control level. Probably, the ws-Lynx1 action on the A549 cell growth is mediated by α7-nAChR.
To demonstrate that the antiproliferative action of ws-Lynx1 is connected with α7-nAChRs, we blocked expression of this receptor by siRNA. Flow cytometry showed that 48 h transfection of A549 cells with siRNA led to more than two-fold decrease of the functional α7-nAChRs expression on the cell surface (Fig 6A and 6B). The WST-1 assay showed that knockdown of the α7-nAChR gene completely cancelled the ws-Lynx1 antiproliferative effect (Fig 6C). We also tested the ws-Lynx1 R38A mutant [45] and the recombinant analogue of neurotoxin II from Naja oxiana [46], the proteins which have no activity against α7-nAChRs. As expected, both of these proteins did not reduce the A549 cell growth (Fig 6C). Thus, we can conclude that the ws-Lynx1 antiproliferative effect on A549 cells is a result of the protein interaction with α7-nAChRs.
The observed difference between the effects of nicotine (agonist of nAChRs), MLA (specific α7 antagonist), and ws-Lynx1 on the A549 cell viability (Fig 4B) revealed the difference in the mechanisms of action of these compounds. It seems like ws-Lynx1 acts on nAChRs in A549 cells in the unusual manner, not like an agonist or antagonist. This is indeed the case, the original study of ws-Lynx1 on α7-nAChRs expressed in Xenopus oocytes revealed that ws-Lynx1 by itself is unable to evoke currents through the receptor channel, but it potentiates ACh-evoked currents at �1 μM concentrations and suppresses them at higher concentrations [28]. Probably, ws-Lynx1 can modulate effects of agonist on the signal transduction through the α7 receptor (as shown here for nicotine), but cannot overcome the effects of antagonists, which block the receptor.
The signaling through α7-nAChRs could activate intracellular responses by two pathways. The canonical ionotropic pathway is connected with the Ca +2 influx through the receptor channel and subsequent activation of the Ca 2+ /calmodulin-dependent protein kinase (CaMK) [47]. The alternative metabotropic pathway is not associated with the ion flow through the receptor channel, but involves activation of the phospholipase C (PLC) and the IP 3 receptors [48][49][50]. PLC hydrolyzes the membrane phospholipid phosphatidylinositol 4,5-bisphosphate with production of the two second messengers IP 3 and 1,2-diacylglycerol (DAG). IP 3 causes the Ca 2+ release from the ER due to the interaction with the IP 3 receptors [51] and both Ca 2+ and DAG in turn activates the protein kinase C (PKC) [52]. Activation of both these pathways results in the phosphorylation of the MAP/ERK kinase that transmits the signal to the nucleus activating different transcriptional factors. Previously it was described, that ws-Lynx1 inhibits the nicotine-induced ERK phosphorylation in the striatal brain slices [53]. In addition, PKC could activate signaling through the c-Jun N-terminal kinase (JNK). The influence of ws-Lynx1 on the JNK activation in the primary cortical neurons was demonstrated recently [54].
The observed cancelation of the ws-Lynx1 effect by the inhibitors of the IP 3 receptors, PKC, and MEK/ERK reveals the involvement of the PLC-dependent signaling with further activation of the transcriptional factors NFκB and STAT5, but not STAT3. The dependence of the ws-Lynx1 effect on JNK, p38 MAP kinase, and PI3K indicates that other signaling pathways could also be involved. It should be noted that in contrast to other cell types, where ERK and JNK are considered as oncogenic kinases, the activation of MEK/ERK [55], JNK and p38 MAPK [56] signaling pathways is required for triggering of cytostatic or apoptotic effects in A549 cells.

Ws-Lynx1 regulates phosphorylation of kinases and transcription factors which control the cell growth
To further disclose signaling cascades triggered by ws-Lynx1 in A549 cells, we studied the phosphorylation of the set of 45 kinases and regulatory proteins upon the 72 h cell incubation with 1 μM of ws-Lynx1 (Fig 7). In agreement with the results of inhibitory analysis (Fig 4C), we observed the significant increase in the ERK1/2 phosphorylation at the positions T202, Y204, T185, and Y187 upon the ws-Lynx1 treatment (Fig 7). Contrary to ERK, we did not detect an increase in the phosphorylation level of JNK1/2/3 (T183, Y185, T221, Y223) and p38α (T180, Y182) (Fig 7), although the inhibitory analysis revealed the participation of these kinases in the ws-Lynx1-induced cell growth suppression (Fig 4C). Probably, the phosphorylation level of JNK and p38α does not change upon the ws-Lynx1 treatment and is already sufficient to activate downstream players of their cascades. In line with this hypothesis, the constant JNK and p38 activity was previously shown in A549 cells [57,58].
As described above, the cell cycle analysis demonstrated that ws-Lynx1 causes two types of the cell cycle arrest-initial G0/1 arrest and G2/M arrest after 24 and 72 h incubations, respectively. Thus, the prolonged ws-Lynx1 application could result in the secondary response(s) related with the activation of different membrane receptors, including growth factors receptors, integrins, G-protein-coupled receptors (GPCRs), T-cell receptors etc. The intracellular signaling cascades triggered by these receptors are mediated by different kinases from the Src family (Fig 8). Indeed, we observed the increased phosphorylation of the platelet-derived growth factor receptor β (PDGFRβ) at the position Y751 and changes in the phosphorylation level of different kinases from the Src family (Fig 7). The significant increase in the phosphorylation of Lyn(Y397), Fyn(Y420), FAK(Y397), and Lck(Y394) and decrease in the Yes(Y426) phosphorylation were observed upon the ws-Lynx1 treatment (Fig 7). In addition, the enhanced phosphorylation of the Src(Y419), Fgr(Y412), and Hck(Y411) kinases was also detected, but did not reach statistical significance. Some of the Src kinases are involved in control of cell growth, proliferation, and adhesion, and are considered as oncogenic. For example, the Fyn and Fgr kinases were significantly down-regulated in patients with lung carcinoma compared to normal cells [59]. The Fgr silencing enhances the phosphorylation of the protein kinase B (AKT) at the Ser473 position, resulting in the AKT stabilization in ovarian carcinoma [60]. The Yes kinase promotes the growth of basal-like breast cancers [61]. Thus, the reverse phosphorylation pattern observed under the ws-Lynx1 treatment (Fig 7) could be responsible for the suppression of cancer cell growth. The FAK and Fyn kinases participate in the regulation of adhesion of colon cancer cells [62].
The growth factor receptors could also be responsible for the observed changes in the AKT phosphorylation (Figs 7 and 8). This signaling cascade could be mediated by PI3K, whose inhibition abolished the ws-Lynx1 effect ( Fig 4C). Contrary, the involvement of the ERK signaling in the AKT activation is unlikely, as we did not observe changes in the phosphorylation of mitogen and the stress activated protein kinase MSK1/2 (S376, S360), which transfers the signal between ERK and AKT (Fig 7). The application of ws-Lynx1 increased the AKT phosphorylation at the T308 position and decreased the phosphorylation of Ser473 (Fig 7). The AKT phosphorylation in these positions causes activation and stabilization of active AKT, respectively [63]. So we cannot say exactly, if the ws-Lynx1 treatment enhanced or not the AKT activity in A549 cells. At the same time, we observed the significant increase in the phosphorylation of the cyclin-dependent kinase inhibitor p27(T198), which could be connected with the AKT activation. This regulator of the cell cycle is considered to be a tumor suppressor and is often inactivated via impaired synthesis, accelerated degradation, or by mislocalization. The p27 phosphorylation induces the p27 mislocalization from the nucleus to the cytoplasm where it undergoes the proteasomal degradation [64]. In line with this, the silencing of p27 was shown to reduce the cell proliferation and induce apoptosis in A549 cells [65]. So the ws-Lynx1-induced p27 mislocalization and further degradation in the cytoplasm may also be involved in control of the A549 cell growth. The AKT activation could also suppress the glycogen synthase kinases-3α/β (GSK3-α/β),-a participant of the Wnt-signaling cascade. Indeed, we observed the moderate (non significant) decrease of the phosphorylation of GSK3-α/β (S9, S21) and βcatenin,-a protein involved in the Wnt-signaling downstream to GSK3-α/β (Fig 9). So, we can speculate, that the ws-Lynx1 effects are not connected with the Wnt-signaling. Another kinase, which phosphorylation level was increased and which can be activated by AKT, is the WNK1 kinase (Fig 7).
The activation of the transcriptional factor NFκB, which according to the inhibitory analysis participates in the ws-Lynx1 effect on the A549 cell growth (Fig 4C), may be mediated by the activation of ERK1/2 and/or AKT (Fig 8). In line with the results of the inhibitory analysis, we observed the significant enhancement of the phosphorylation of the transcription factors STAT5a(Y694) and STAT6(Y641), but not of STAT3(Y705, S727) (Fig 7). Although STAT5 could be activated by the Src kinases, the simultaneous activation of the several STATs is probably connected with the activation of the cytokine receptors and downstream JAK kinases ( Fig  8). This signaling pathway could be responsible for some of the effects induced by ws-Lynx1. For example, the dephosphorylated STAT6 drives the constitutive cyclooxygenase-2 (COX2) expression and cytotoxicity resistance in lung cancer cells [66], so the ws-Lynx1-induced STAT6 phosphorylation could diminish the COX2 level and increase the sensitivity of A549 cells to apoptosis. Despite the quite speculative nature of the proposed ws-Lynx1 action model, we can conclude that ws-Lynx1 controls the growth of A549 cells by activation of different intracellular signaling pathways either immediately after initial interaction with α7-nAChR or by inducing secondary responses associated with secretion of different signaling factors.

Ws-Lynx1 induces apoptosis in A549 cells via phosphorylation of the proapoptotic factor p53
The large increase in the phosphorylation of the tumor protein p53, essential for the apoptosis induction [67], was observed upon the A549 cell treatment with ws-Lynx1 (Fig 7). The phosphorylation of p53 at the Ser46 and Ser392 positions is characteristic for the active p38 and ERK1/2 kinases, respectively [68,69]. The p53 phosphorylation could also be controlled by the Lck and Lyn kinases from the Src family [70,71]. The triggering of apoptotic signaling is supported by the large increase in the phosphorylation of the heat shock protein 60 (HSP60, Fig 7). HSP60 initiates events upstream of the cytochrome C-triggered apoptosome activation and participates in the caspase-3 activation [72]. In addition, we observed the decrease in the phosphorylation of the other p53 signaling partner-the AMPKα1 kinase at the position T183, which is responsible for the control of cell proliferation (Fig 7). Thus, the p53 activation by ERK, p38, and/or Src kinases could trigger apoptosis in A549 cells in response to the ws-Lynx1 application.
To prove this, we used the Annexin V-PE conjugate that binds to the externalized phosphatidylserine, an early apoptotic marker. The analysis by flow cytometry revealed that the number of A549 cells with the externalized phosphatidylserine significantly increased upon the 72 h treatment with 1 μM ws-Lynx1 from 5.39 ± 0.6% to 37.44 ± 2.16% as compared to the control (Fig 9). Thus, ws-Lynx1 down-regulates the growth of A549 cells not only by suppression of proliferation via the cell cycle arrest, but also by the apoptosis induction. Notably, we did not observe DNA fragmentation in nuclei of A549 cells stained by Hoechst 33342 after the 72-hour incubation with ws-Lynx1. It should be mentioned, that the assay based on the morphology analysis of nuclei stained by Hoechst dye detects the changes characteristic for the late stages of apoptosis, while the phosphatidylserine externalization is a marker of early apoptosis [73]. In line with it, no significant increase in the p53 phosphorylation at the Ser15 position associated with the DNA damage on the late stages of apoptosis was observed (Fig 7).

Conclusion
nAChRs play an important role in many essential processes and are involved in nicotineinduced carcinogenesis [6]. Previously we demonstrated that the human secreted proteins SLURPs could inhibit the growth of cancer cells of epithelial origin [18,19]. Human Lynx1 was also found to participate in the regulation of the lung cancer cells growth [26,44]. Here, for the first time we described the Lynx1 protein expression in epidermal, colon and breast epithelial cells. Colocalization of Lynx1 with α7-nAChRs in lung and colon cancer cells was observed. We also found that the water-soluble variant of human Lynx1 controls the growth and apoptosis of non-small lung cancer cells. Based on the data obtained, we propose the two-stage mechanism of the ws-Lynx1 action (Fig 7). Ws-Lynx1 interaction with α7-nAChRs activates different intracellular kinase cascades with the subsequent phosphorylation of the proapototic factor p53 and other factors participated in the control of the gene transcription as well as cell growth and apoptosis. Thus, the endogenous Ly6/uPAR proteins like Lynx1 are the promising prototypes for development of new drugs for treatment of epithelial cancers. Anti-Lynx1 antibodies was incubated with ws-Lynx1 for 1 h, and A549 cells were stained either with rabbit anti-Lynx1 and mouse anti-α7-nAChR antibodies or by the mixture of ws-Lynx1 with rabbit anti-Lynx1 and mouse anti-α7-nAChR antibodies. Secondary antibodies were anti-rabbit Alexa-488 labeled IgG (green) and anti-mouse Alexa-594 labelled antibodies (red). Cell nuclei were visualized by Hoechst 33342 (blue). Scale bar 10 μm. (TIF) S1