Phospho-proteomic analysis of the effect of merecidin inhibiting the human lung adenocarcinoma A549 cells

Xiuqing Wang (  xiuqingwang1979@163.com ) Ningxia Medical University https://orcid.org/0000-0002-0510-9123 Qinqin Jia Department of Laboratory Medicine college of Clinical Medicine ,Ningxia MedicalUniversity ,Yinchuan,China Qiannan Zhang Ningxia Medical College: Ningxia Medical University Tingting Yang Department of Laboratory ,Yinchuan Maternal and Child Health care Hospital ,56 Wenhua Street ,Xinqing District,Yinchuan City ,Ningxia Province Jinxuan Song Department of laboratory Medicine ,College of Clinical Medicine ,Ningxia Medical University,Yinchuan, Yarong Wang Derpartment of Pathology,Institute of Hematology and Blood Diseases Hospital,Chinese Academy of Medical sciences Tianjin


Background
Lung cancer has a very high incidence and rate of mortality 1 worldwide, accounting for more than 16.7%of the cancer-related deaths 2 . Small-cell lung cancer (SCLC) represents 13-20% of lung cancer [3][4][5] . However, since 1997, there has not been any improvement in the response rates to computed tomography (CT) or survival of SCLC patients 6, 7 . The 5-year survival rate is less than 20% 7 . Therefore, anti-cancer drugs targeting lung cancer are urgently required. Protein phosphorylation plays a signi cant role in many cell processes, such as proliferation, differentiation, migration, autophagy, and apoptosis.
Protein phosphorylation is one of the most important covalent modi cations in organisms. The simple, exible, and reversible properties of phosphorylation, as well as the availability of phosphate donor ATP, make phosphorylation the most common regulatory mechanism in eukaryotic cells 8 . Protein phosphorylation 9 is a reversible process, almost regulates cell proliferation, development, differentiation, and the regulation of the cytoskeleton, apoptosis, autophagy 10, 11 , and neural activity. An increasing number of studies have shown that proteomics analysis 12 of tumor samples is a comprehensive and dynamic method. The combination of these methods with molecular networks in cancer development elucidates the pathogenesis of tumors with respect to diagnosis, treatment, and prognosis of the disease. Several studies [13][14][15] have analyzed the phosphorylated proteomics data of cancer cells.
LL-37 is the only antimicrobial peptide found in the human body, which is expressed in testis, skin 16 , nasal mucosa, saliva 17 , and gastrointestinal tract 18 . It is a multifunctional peptide 19 with antibacterial and anti-cancer properties and is also involved in the regulation of adaptive immunity 20,21 . LL-37 alpha-helix structure combined with cell membrane lipid induces the formation of cell membrane channel 22 or increases the permeability of cell membrane 23,24 , leading to cell death 25,26 . Ren et al. found that the treatment of colon cancer cells with LL-37 exposes it to anionic phosphatidylserine (PS) and DNA fragmentation, suggesting apoptosis 27 . Furthermore, it plays an anti-cancer role in gastric cancer 28 , acute myeloid leukemia 29 , and lymphocytic leukemia 30 , and also upregulates the mRNA expression and protein level of VEGF-A and the phosphorylated levels of ERK1/2 and NF-kB p65 in HPL cells 31 . Merecidin (also known as 17BIPHE2) is a cutting peptide of antibacterial peptide LL-37. It exerts lower hemolytic activity and less toxicity to the human body, and is superior to its parent LL-37 32 in terms of immune regulation and inhibition of various bacteria.Our previous study demonstrated that merecidin promotes lung adenocarcinoma A549 cell apoptosis [33][34][35] and inhibits the formation of Staphylococcus aureus and Pseudomonas aeruginosa bio lm 36, 37 ; however, the underlying molecular mechanism is not yet fully understood.

Cell culture
The human lung adenocarcinoma cancer cell line A549 was obtained from the A liated Hospital of the Ningxia Medical University and cultured in in Roswell Park Memorial Institute (RPMI)-1640 medium (HyClone) containing 10% fetal bovine serum (FBS, Gibco, USA) and 100 U/mL penicillin and streptomycin (Gibco) at 37°C in a cell culture incubator with 5% CO 2 .

Synthesis and characterization of merecidin
The LL-37 peptide contains 37 amino acids, with two leucine residues at the N-terminus. The minimum sequence responsible for the antimicrobial effect of LL-37 was deduced from peptide activity assays and was found to be the 17-29 or 18-29 amino acids (amino acid sequence: GBKRLVQRLKDBLRNLV, L is D-type amino acid, B is biphenyl phenylalanine, and the molecular weight is 2 561.866 38, 39 ). Solid-phase chemistry was used to synthesize the peptides with ≥ 95% purity. ddH 2 O solution was used in all the experiments.

Protein extraction
After treatment with 9 mmol/L merecidin for 6 h, A549 cells were sonicated three times on ice using a high-intensity ultrasonic processor (Scientz,Shanghai,China) in lysis buffer (8 M urea, 1% protease inhibitor cocktail). For PTM (Protein Post-translational Modi cation) experiments, inhibitors were added to the lysis buffer (3 µM TSA and 50 mM NAM for acetylation). The remaining debris was removed by centrifugation at 12,000 g at 4°C for 10 min. Finally, the supernatant was collected, and the protein concentration was determined using the BCA assay (Beyotime, Shanghai, China), according to the manufacturer's instructions.

Trypsin digestion of the proteins
For digestion, the protein solution was reduced with 5 mM dithiothreitol for 30 min at 56°C and alkylated with 11 mM iodoacetamide for 15 min at room temperature in the dark. Subsequently, the protein sample was diluted by adding 100 mM TEAB(Tetraethyl ammonium bromide) to urea < 2M. Finally, trypsin was added at 1:50 trypsin-to-protein mass ratio for the rst digestion overnight and 1:100 trypsin-to-protein mass ratio for 4h digestion.

TMT/iTRAQ labeling
After trypsin digestion, the peptide was desalted by Strata X C18 SPE column, vacuum-dried, reconstituted in 0.5 M TEAB, and processed according to the manufacturer's protocol for TMT kit/iTRAQ kit (Sigma, USA). Brie y, 1 U of TMT/iTRAQ reagent was thawed and reconstituted in acetonitrile. The peptide mixtures were incubated for 2 h at room temperature and pooled, desalted, and dried by vacuum centrifugation.

High-performance liquid chromatography (HPLC) fractionation
The tryptic peptides were fractionated by high pH reverse-phase HPLC using Thermo Betasil C18 column (5 µm particles, 10 mm ID (Identity document), 250 mm length). Brie y, the peptides were rst separated with a gradient of 8-32% acetonitrile (pH 9.0) into 60 fractions over 60 min and then combined into 6 fractions and dried by vacuum centrifugation.

IMAC-enriched phosphorylated peptide
The peptides were dissolved in the enrichment buffer solution (50% acetonitrile/6% tri uoroacetic acid), the supernatant was transferred to the pre-washed IMAC material (AbD Serotec, UK), and incubated for xxx h. Then, a buffer solution consisting of 50% acetonitrile, 6% tri uoroacetic acid, and 30% acetonitrile/ 0.1% tri uoroacetic acid was used for washing the resin. Finally, 10% ammonia water was used to elute the modi ed peptide section, and the eluent was collected and vacuum-frozen. After draining, desalination was carried out according to C18 ZipTips kit (Millipore, Bedford, USA).

HPLC/MS/MS analysis
The puri ed peptide was solubilized in liquid chromatography mobile phase A (0.1% formic acid solution) and then separated by easy-NLC 1000 ULTRA high-performance liquid phase system. Mobile phase A was an aqueous solution containing 0.1% formic acid and 2% acetonitrile, while mobile phase B was an aqueous solution containing 0.1% formic acid and 90% acetonitrile. Liquid phase gradient settings were as follows: 0-40 min, 4-22% B; 52 min, 22-35% B; 52-56 min, 35-80% B; 56-60 min, 80% B, ow rate maintained at 400 nL/min. The eluted peptides were separated by an ULTRA high-performance liquid system, injected into an NSI ion source for ionization, and then analyzed by Q Exactive Plus mass spectrometry. The ion source voltage was set at 2.0 kV, and the peptide fragments and secondary fragments were detected and analyzed using high-resolution Orbitrap. The level of mass spectrum scanning range was set to 350-1800 m/z, scanning resolution was set to 70,000, the scanning range of the secondary mass spectrometry was xed at 100 m/z, and the resolution of the secondary scanning was set at 35,000. A data-dependent acquisition (DDA) program was used to select the rst 10 peptide parent ions with the highest signal intensity into the HCD collision pool after the rst level scan; 31% energy was used for fragmentation. Secondary mass spectrometry was performed sequentially to improve the effective utilization rate of mass spectrometry. The automatic gain control (AGC) was set at 1E5, the signal threshold was set at 20000 ion/S, the maximum injection time was set at 100 ms, and the dynamic elimination time of tandem mass spectrometry scanning was set at 30 s to avoid repeated scan of the parent ion.

Bioinformatics analysis of phosphorylated proteins
The original mass spectrometry data were retrieved using the database Maxquant (V1.5.2.8, China). The retrieval parameter settings were as follows: Swissprot_Human (20422 sequences) was used as the database, and the negative library was added to calculate the positive error rate (FER) caused by random matching to eliminate the in uence of contaminated proteins in identi cation results. The enzyme digestion method was set as trypsin/P, the maximum number of leakage sites of the enzyme was set as 2, the minimum length of the peptide segment was set as 7 amino acid residues, and the maximum number of modi cations of the peptide segment was set as 5. The mass error tolerance of the primary parent ion of the rst search and main search is set as 20×10 − 6 and 5×10 − 6 , respectively, and the mass error tolerance of the secondary fragment ion is 0.02 Da. The alkylation of cysteine was xed, while the variable modi cation was methionine oxidation, n-terminal acetylation, deamidation (NQ), and phosphorylation of serine, threonine, and tyrosine. The quantitative method was set as TMT-6plex, and the Protein identi cation and PSM(Anti-PSM/Proteasome 20S) identi cation false discovery rate (FDR) was set at 1%. The differences between phosphorylated proteins were detected via a volcanic diagram to characterize these genes through the GO (Fisher's exact test), function, KEGG( Kyoto Encyclopedia of Genes and Genomes) pathways, and analysis of the protein structure domain in cell biology and cell and molecular biology function analysis. Furthermore, Clusters of Orthologous Groups of proteins(COG/KOG) database function classi cation statistics and Cytoscape were employed to build the protein network.

Statistical analysis
All values were expressed as mean ± standard deviation calculated from three independent experiments. The data were analyzed with a one-way analysis of variance (ANOVA), followed by Tukey's multiple comparison posthoc test (SPSS 24.0 software). P < 0.05 and P < 0.01 indicated statistical signi cance.

Phosphorylation proteomics results
A total of 10320 phosphorylated modi cation sites located on 3,089 proteins were identi ed, among which, 9094 provided quantitative information. The identi cation data were ltered by 0.75 criterion to identify 8,317 phosphorylation modi cation sites on 2,900 proteins; of these 7,937 sites for 2,787 proteins contained quantitative information (Fig. 1A). According to the fold-change > or < 1.5, 246 upregulated phosphorylated proteins and 559 downregulated phosphorylated proteins, corresponding to 485 and 860 phosphorylated sites were identi ed, respectively (Fig. 1B). The distribution of proteins is shown (Fig. 1C) in the volcanic diagram. Red represents an upregulated protein, blue represents downregulated protein, and the difference in protein expression (drug/mock) occurred within 0.3-3 fold.

Motif analysis of phosphorylated proteins
In order to understand the modi ed proteins identi ed in the data and the functions and characteristics of the proteins, the software Protein motif was applied to analyze the regularity of amino acid sequences before and after all phosphorylation modi cation sites in the samples and calculate the regularity trend of amino acid sequences within the region harboring the phosphorylation modi cation sites. Figure 2 shows the motif enrichment heat map of amino acids upstream and downstream of all identi ed phosphorylation modi cation sites; red indicates signi cant enrichment of the amino acids near the modi cation site, while green indicates a signi cant reduction of the amino acid near the modi cation site. The results showed that the downstream amino acids were signi cantly enriched at the phosphorylation modi cation sites. Also, the Gene Ontology (GO) protein annotation, the distribution of biological process, cell composition, and molecular function of proteins corresponding to the differentially modi ed sites in the GO secondary annotation were analyzed (Fig. 3A). Furthermore, the phosphorylated proteins were distributed in the cell process, biological regulation, cell and organelle composition, molecular binding, and catalytic activity. The prediction and classi cation statistics of the subcellular structure of the differentially modi ed proteins are shown in Fig. 3B, indicating that the differentially modi ed proteins are mainly concentrated in the nucleus and cytoplasm .

Functional enrichment analysis of differential phosphorylated proteins
The functional enrichment of differential proteins is mainly carried out at the three levels of GO classi cation, KEGG pathway, and protein domain. For the enrichment test, the P-value showed signi cant enrichment of differentially modi ed proteins (P < 0.05). Differences were detected in the corresponding protein phosphorylation modi cation sites with respect to the biological process (Fig. 3C), cellular component (Fig. 3D), and molecular function (Fig. 3E). These ndings showed signi cant differences in the degradation of proteins involved in cell signal transduction, DNA transfer and translation of the cellular energy metabolism, protein synthesis, and the formation of the cytoskeleton. Proteins corresponding to the differential phosphorylation modi cation sites were classi ed in the protein domain (Fig. 3F). KEGG pathways mainly included metabolism, genetic information processing, environmental information processing, cell processes, human diseases, and drug development. In this study, pathways such as mTOR, ErbB, PI3K/AKT,and AMPK were enriched and monitored (Fig. 3G).

Functional classi cation of different proteins
The Chinese de nition of COG, namely "homologous protein," is compared and analyzed in the database, and the COG/KOG functional classi cation statistics of the differentially modi ed proteins were carried out (Fig. 4). It was found that differential phosphorylated proteins mainly focus on cell signal transduction, RNA transcription, translation processing and modi cation, ribosomes, cytoskeleton proteins, intracellular material transport, protein secretion, and vesicle transport.

Discussion
Protein composition is the giant arm of the post-genetic age, and many researchers have investigated the pathogenesis of cancer through proteomics analysis. A previous study 40 demonstrated that all the proteins might be expressed in the tumor cells and tumor microenvironment. Surprisingly, cluster analysis revealed that diffuse gastric cancer could not only be divided into three molecular subtypes (PX1-3), but also was closely related to the survival prognosis and chemotherapy sensitivity. Sato et al. 41 found that the use of CKAP4 as a biomarker changes the current treatment of lung cancer patients. In addition, the combination of CKAP4 and conventional markers can signi cantly improve diagnostic accuracy.
Although there have been extensive studies on LL-37, the phosphorylation analysis of LL-37 and its derived peptides in cells has not yet been reported. In this study, we implemented a quantitative proteomics analysis to determine how protein phosphorylation systematically advances apoptosis of lung cancer A549 cells treated with antimicrobial peptide merecidin.
Next, we used to quantitative proteomics research strategy via TMT labeling, phosphorylation modi cation enrichment techniques, and highresolution liquid chromatography-mass spectrometry. The localization probability was > 0.75. The data were ltered to identify 8,317 phosphorylation sites on 2,900 proteins, of which 7,937 sites for 2,787 proteins provided quantitative information. We identi ed that 485 sites in the drug group vs mock comparison group were upregulated, and 860 sites were downregulated. The identi ed proteins were annotated regarding GO, protein domains, KEGG pathways, and subcellular structure localization to describe the biological process, cellular component, and molecular function. The protein motif is closely related to the biological functions of proteins, which was used to analyze the regularity of amino acid sequences before and after all phosphorylation modi cation sites in the sample. The data suggested that in the biological process, the protein motif is mainly involved in cell process, biological regulation, and signal transduction. The cell composition mainly involves the cytoplasm, cell membrane, and organelle, and the molecular function involves protein molecule binding, activation, and molecular function regulation. In terms of subcellular composition, the composition of the nucleus, organelles, cytoplasm, and mitochondria are involved. The functional enrichment results showed that differentially expressed proteins are involved in the regulation of the whole process from DNA replication to RNA transcription and translation, as well as protein-peptide chain formation, plait, and domain formation. The enrichment of the KEGG pathway revealed the proteins, such as GnRH, RIG-I-like, Ras, ErbB, PI3K-Akt, mTOR, and AMPK in signaling pathways. In summary, merecidin-treated A549 cells affect the cell functions from the nucleus to the cytoplasm and organelles. In addition, the COG/KOG functional classi cation statistics identi ed cytoskeleton proteins, intracellular material transport, secretion, and vesicle transport.
Historically, we found that the antimicrobial peptide merecidin induced apoptosis in lung cancer A549 cells 34,35 . We also screened an array of proteins from the identi cation protein interaction network (Fig. 5) as follows: MAPK1, also known as ERK2, is associated with tumor proliferation 42 , differentiation, and invasion, and its upregulation is associated with the occurrence and development of various tumors 43 .Mitogen-activated protein kinase 3(MAPK3),knows as ERK1,whereas ERK pathway mediates cellular survival and growth,as well as leads to increased apoptosis 44 the study suggests that MAPK3 has close relationship with cardiomyocyte apoptosis in IRI 45 . MAPK14 gene, known as p38α, is one of the four family members of MAP kinase which exert versatile functions in diverse cellular processes in cancer: cell cycle regulation, proliferation, survival and motility 46 ,as well as MAPK14 (P38) can be involved in a variety of cellular stresses, including internal metabolic stress, DNA damage, external growth factor pathway, cell-matrix interaction and intercellular communication 47,48 .Mesquita et al have showed p38α inhibition Cellular migration was impaired and decreased the gastric cancer cell proliferation by provoking cell cycle arrest and cell death 49 .In this study ,the phosphorylation site T180/Y182 was downregulated by merecidin .Ephrin type-B receptor 2(EPHB2) is associated with the occurrence and development of A variety of diseases.EPHB2 has been shown to be associated with cells apoptosis in many studies 50,51 . 52 Dong et al have suggested that EphB/ephrinB reverse signaling is involved in retinal ganglion cell (RGC) apoptosis in experimental glaucoma.In our study ,the phosphorylation site S776/S575of EphB2 Upregulated. There are also a number of identi ed apoptosis-related proteins that have been published in the previous by our research group 53 .Furthermore, we speculated that merecidin induction of A540 cells enriched the KEGG-identi ed classic apoptosis-related signaling pathways: ErbB 54 , PI3K/AKT 55 , and AMPK-Akt 56 .
We suspected that merecidin might induce autophagy of A549 cells. Based on the literature, we selected the following proteins from the protein interaction network. ATG2B was mainly involved in the formation of autophagosomes and fused with lysosomes before degradation 57 . ATG2A and ATG2B genes are involved in autophagosome formation in HeLa cells 58 . In this study, the phosphorylation was upregulated at sites S240/S255 of ATG2B. ATG9A is involved in autophagosome and cytoplasmic vacuolar transport (Cvt) and is the only transmembrane protein.
Atg9 plays a crucial role in the formation of autophagosomes 59 . After the treatment of A549 cells with merecidin, the phosphorylation at site S656/738/735 is upregulated. ATG13 60 is involved in autophagy and targets the mTOR kinase signaling pathways by controlling the ATG13-ULK1 phosphorylation status and ATG13-ULK1 to regulate autophagy. Human serine/threonine protein kinase ULK1 is one of the major regulatory factors of apoptosis. ULK1/2 induces proliferation 61 , autophagy in ULK complex, and controls autophagosome biogenesis via early signaling pathways 62 . The S738/S735 of ATG13 and S479/S477 of ULK1 was upregulated in this study. RAB7A is a key regulator of lysosomal transport and critical for autophagosome-lysosome fusion 63 . As a vital protein in the FOXO1/RAB7 signaling pathway, RAB7A induces the autophagy by regulating RAB7 upregulation. The phosphorylation at S72 site in RAB7 is also upregulated in this study. MAPK1 64 harbors several signal integration points that participate in a variety of cellular processes, such as proliferation, differentiation, transcription, and development.
The upstream kinase phosphorylation activation MAPK1/ERK2 was applied for the stimulation of the nucleus, and the inhibition of phosphorylation of MAPK1 activates autophagy 65, 66 . The phosphorylation site T185/Y187 of MAPK1 is upregulated after treatment with merecidin. EIF4B is also known as cap-binding protein, one of the vital components of the eIF4F complex. The high expression can lead to tumor proliferation, invasion, and metastasis. Zhang et al. 67 found that inhibition of EIF4B expression enhances autophagic death, and S579 of EIF4B was found to be upregulated in this study. The EGFR signaling pathway regulates the autophagy process; inhibition of the expression of EGFR elevates autophagy, leading to autophagic death 68 . In this study, TSC2 was found to activate autophagy by activating the TSC2/mTOR signaling pathway 69, 70 , and the phosphorylation site S1132/S1452 of TSC2 is downregulated. PAK4 is highly expressed in a variety of tumor tissues and cells and is related to cytological behaviors, such as cell proliferation, cycle, and migration. Low expression of PAK4 can upregulate p53, inhibit mTOR or Akt/mTOR signaling pathway, inhibits cell migration, and activate autophagy 71 , or the signaling pathway 72 can be downregulated, thus activating autophagy 73,74 . The phosphorylation site S103 of PAK4 downregulated after merecidin treatment. AKT1 regulates metabolism, proliferation, cell survival, growth, and angiogenesis, inhibits growth, and induces apoptosis of non-SCLC by directly regulating Akt1/2 29 , mediated by phosphorylation of serine and/or threonine by a series of downstream substrates. In addition, Akt1/Foxo3a 75 and PI3K/AKT/mTOR 76 signaling pathways are associated with autophagy, and the phosphorylation at site S124 of AKT1 is downregulated after merecidin treatment. Additionally, we found that the autophagy-related pathways mTOR and PI3K-Akt were selected from KEEG pathway enrichment data. In conclusion, we hypothesized that merecidin induces autophagy in A549 cells and by mTOR and/or PI3K-Akt signaling pathway.

Conclusion
In summary, merecidin treatment of lung cancer A549 cells induced apoptosis through signaling pathways such as ErbB, PI3K/AKT, and AMPK-Akt. In addition, the antimicrobial peptide merecidin might induce autophagy in A549 cells. This study laid a solid foundation for further exploration of merecidin in lung cancer A549 cells.

Competing interests
The authors declare no con ict of interest.

Funding
This study was supported by the National Natural Science Foundation of China (No.81760661, No. 81560573) but did not receive any speci c grant from the funding agencies in the public, commercial, or not-for-pro t sectors.
Authors' contributions 1. Jia Qinqin is the writer ,contributor of this paper and executor of the experiment of this study.

Zhang qiannan ,Yang
Tingting is the executor of the experiment.  Functional enrichment analysis of proteins corresponding to differentially modi ed sites. 3A: Statistical distribution of proteins corresponding to the differential phosphorylation modi cation sites in GO secondary classi cation (green represents the distribution of biological processes, red represents the distribution of cell composition, and purple represents the distribution of molecular functions); 3B: Green represents the distribution of biological processes, red represents the distribution of cell composition, and purple represents the distribution of molecular functions; 3C-F: Differential phosphorylation modi cation site corresponding to protein enrichment distribution bubble map (3C: biological processes; 3D: cell composition; 3E: molecular functions; 3F: domain); 3G: KEGG pathway. COG/KOG functional classi cation distribution map of proteins corresponding to differential phosphorylation modi cation sites.

Figure 5
Differential modi ed protein interaction network diagram.