Evaluation of potential anti-metastatic and antioxidative abilities of natural peptides derived from Tecoma stans (L.) Juss. ex Kunth in A549 cells

Materials

The chemicals, materials, and equipment follow: Flower of Tecoma stans (L.) Juss. ex Kunth was collected in April 2021 in Pathuntani, Thailand (5.7800° N, 101.0372° E). For quality control (QC), the samples were collected in three biological samplings and conducted in independent extraction processes. An ultracentrifugal mill model ZM-200 was purchased from the Retsch Co. (Haan, Germany). A rotatory evaporator (model: R300) was purchased from Buchi Co. (Flawil, Switzerland). Synergy H1 UV-Vis spectrometer microplate reader was purchased from BioTek Co. (Winooski, VT, USA). Vivaspin^® 20 Molecular weights cut off (MWCO) 3 kDa, dithiothreitol (DTT), and iodoacetamide (IAA) were purchased from GE Healthcare Co. (Amersham, UK).

Human non-small cell lung cancer cell line, A549 (ATCC: CLL-185); human hepatocarcinoma cell line, HepG2 (ATCC: HB-8065); human cervical adenocarcinoma cell line, HeLa (ATCC: CCL-2); human melanoma cell line, SK-MEL-28 (ATCC: HTB-72) and human breast cancer cell line, MCF-7 (ATCC: HTB-22) were purchased from the American Type Culture Collection (Manassas, VA, USA). Immortalized human keratinocyte cell line, HaCat (CLS: 300493) was purchased from Cell Lines Service. The biochemical reagents, including Dulbecco’s modified Eagle’s medium (DMEM), fetal bovine serum (FBS), and penicillin-streptomycin, were purchased from Gibco (Grand Island, NY, USA). The dimethyl sulfoxide (DMSO), methylthiazolyldiphenyl-tetrazolium bromide (MTT), 1,1-diphenyl-2-picrylhydrazyl (DPPH), ascorbic acid (As), gallic acid (GA), 2,2′-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) (ABTS), K₂S₂O₈, FeCl₃, 2,4,6-tris(2-pyridyl)-s-triazine (TPTZ), and FeSO₄ were obtained from Sigma Aldrich Co. (St. Louis, MO, USA). A cellular ROS detection kit, 2′–7′ dichlorofluorescein diacetate (DCFH-DA), was purchased from Abcam Co. (Cambridge, UK). The transwell insert (6.5-mm diameter polyvinylpyrrolidone-free polycarbonate filter of 8-µm pore size) and Matrigel were purchased from Corning Inc. (Corning, NY, USA).

Trypsin, used for sequencing, was purchased from Promega Co. (Madison, WI, USA). Rapigest SF surfactant was purchased from Waters Co. (Milford, MA, USA). Halt protease inhibitor cocktail and benchtop centrifuge (Sorvall™ Legend™ Micro 17) were purchased from Thermo Scientific Co. (Waltham, MA, USA). Solvents for LC-MS/MS, including LC-MS waters and acetonitrile (ultra LC-MS), were purchased from J.T. Baker (Fisher Scientific, Loughborough, UK). All other biochemical reagents were purchased from Sigma Aldrich Co.

Natural peptide enrichment and purification

Surface sterilization of the flower of T. stans was done using 0.1% Tween-80 solution and washed five times with deionized water at room temperature (25–26 °C). After sterilization, the T. stans was dried (10 g) using a hot-air oven at 40 °C for 2 days on an aluminum foil and then grounded into small particle size by the ultra-centrifugal mill at 8,000g for 3 s. The grounded T. stans was extracted using high pressure and temperature in deionized water (121 °C at 15 psi for 20 min) at 1:2 (w/v) ratio. The extracted T. stans was left to room temperature, then added absolute ethanol at 1:1 (v/v) ratio. The supernatant was filtrated through the MWCO 3 kDa. The natural peptides fraction was collected. The natural peptide enrichment and purification using solid-phase extraction (SPE) cartridges were conducted. The SPE cartridges were pre-conditioned using 20 ml acetonitrile and equilibrated with 15 ml deionized water. The flow-through from MWCO was loaded on the equilibrated SPE and eluted with 20% acetonitrile. An eluent was then evaporated under a vacuum using the rotatory evaporator. For investigating the extract quality using LC-MS/MS analysis, Orbitrap HF hybrid mass spectrometer combined with an UltiMate 3000 LC system was used as described previously (Treesubsuntorn et al., 2021). Briefly, the peptides were first desalted one-line on a reverse-phase C18 PepMap 100 trapping column before being resolved onto a C18 PepMap™ 100 capillary column with a 60 min gradient of CH₃CN, 0.1% formic acid, at a flow rate of 300 nl/min. Peptides were analyzed by applying a data-dependent Top10 method consisting of a scan cycle initiated by a full scan of peptide ions in the Orbitrap analyzer, followed by high-energy collisional dissociation (NCE = 28) and MS/MS scans on the 15 most abundant precursor ions. Full scan mass spectra were acquired from m/z 200 to 2,000 with an AGC target set at 3 × 10⁶ ions and a resolution of 120k. MS/MS scan was initiated when the AGC target reached 10⁵ ions and a resolution of 30k. Ion selection was performed applying a dynamic exclusion window of 15 s. The peptide peak picking using PPD algorithm with signal/noise ratio >1 by an automatic peak detection tools within Thermo Scientific FreeStyle software. The natural peptides profile was illustrated in total ion counts, parent masses (MS), and daughter masses (MS/MS) profiles. The raw mass spectra were analyzed by PeakX studio 10.0 program. De novo peptide sequencing of the highest peptide ion intensity was performed with default parameters. LC-MS run was analyzed with no specific digestion enzyme. Mass error tolerance for MS and MS/MS was 10 ppm and 0.01 Da, respectively. CID-Fragmentation series (b, y, b-H₂O, y-H₂O, and y-NH₃) were used to predict peptide sequence (Krobthong & Yingchutrakul, 2020). The acceptable de novo peptide sequences were achieved by filtering average local confidence to ≥95%.

Cell viability assay

The cytotoxic effect of natural peptides from the flower of T. stans was determined by MTT assay. A549, HepG2, HeLa, SK-MEL-28, MCF-7, and HaCat cells were plated in a 96-well plate at a density of 1 × 10⁴ cells and incubated for 24 h. After incubation, the cells were treated with natural peptides from T. stans at different concentrations (2.0, 1.0, 0.5, 0.25, 0.125 and 0.0625 ng/ml for cancer cells and 25.0, 12.5, 6.25, 3.125, 1.563, 0.781 and 0.391 ng/ml for immortalized cells) or negative control (0.01% DMSO) for 24 h. After treatment, 0.5 mg/ml of MTT solution was added to each well, and the plate was further incubated for 2 h at 37 °C. The MTT assay medium was removed, and 100 µl of DMSO was added to each well. The cell cytotoxicity was measured using absorbance at 570 nm and transformed to the percentage of the cell viability. The inhibitory concentration (IC₅₀) was analyzed using the non-linear sigmoidal model of GraphPad Prism 8.0.1 software (San Diego, CA, USA). Cell viability assay was performed with three independent experiments. The dosage of natural peptides from T. stans at the IC₅₀ was used in the intracellular ROS evaluation and proteomic analysis.

Migration and invasion assays

The effect of natural peptides from the flower of T. stans on cell migration and invasion was assayed by transwell chambers. A549 cells were seeded in a 24-well plate and treated with or without the lowest concentration (0.0625 ng/ml) of natural peptides for 24 h. After treatment, cells were harvested, and 1.0 × 10⁵ cells were suspended in 200 µl of the serum-free medium before plating into the upper chamber of transwell, which was uncoated (migration assay) or pre-coated (invasion assay) with 30 µg of Matrigel. Cells were allowed to migrate or invade through transwell to the lower chamber containing 750 µl of complete medium with 10% FBS, which was used as a chemoattractant at 37 °C for 24 h. After incubation, the cells in the upper surface of the membrane were removed with a cotton swab, and the chamber was washed twice with 1× PBS. The cells that had migrated across the membrane or invaded through the Matrigel were fixed with 3.7% (v/v) formaldehyde in 1× PBS for 20 min. The cells were stained with 0.5% crystal violet in 25% (v/v) methanol overnight, followed by two washes with tap water. Finally, the migrated or invaded cells on the undersurface of the membrane were counted and photographed under a microscope. The data are presented as the percentage of migration or invasion through the uncoated or pre-coated Matrigel membrane, respectively.

In vitro radical scavenging assay of the natural peptides by using DPPH, ABTS and FRAP assays

The evaluation of the relative antioxidant capacity of natural peptides derived from the flower of T. stans was carried out using DPPH, ABTS and FRAP (Ferric reducing antioxidant power) assays, as mentioned previously (Krobthong et al., 2021). An antioxidant capacity was determined at the fixed concentration of 1 μg/μl of the natural peptides. For DPPH scavenging activity, the 10 µl of natural peptide solution was mixed with 190 µl of 0.2 mM DPPH in the absolute ethanol. The absorbance was measured at 517 nm. DPPH antioxidant efficiency used ascorbic acid (As) as the standard. The DPPH scavenging activity was expressed as the ascorbic acid equivalent antioxidant capacity (AsEAC).

For ABTS scavenging activity, ABTS radical solution (7 mM ABTS solution with 2.45 mM potassium persulfate) was diluted in 5 mM phosphate buffer saline solution, pH 7.2. The 10 µl of natural peptide solution was mixed with 190 µl ABTS solution. The experiment was incubated at room temperature in the dark for 15 min. The absorbance of the mixture was measured at 734 nm. ABTS antioxidant efficiency used gallic acid (GA) as the standard. The ABTS scavenging activity was expressed as the gallic acid equivalent antioxidant capacity (GAEAC).

For FRAP scavenging activity, the FRAP working solution (10 mM TPTZ, 20 mM FeCl₃, and 300 mM sodium acetate) was prepared before the experiment. The 10 µl of natural peptide solution was mixed with 190 µl of the FRAP working solution. The reaction was incubated at room temperature in the dark for 25 min. The absorbance was measured at 593 nm. FRAP efficiency used FeSO₄ as the standard. The FRAP scavenging activity was expressed as FeSO₄ molar equivalent.

Intracellular ROS determination

For intracellular ROS production, A549 cells were stimulated with a low dose of LPS that does not affect cell viability (Baek et al., 2020; Seo et al., 2015). The cells were pretreated with or without the IC₅₀ dosage of natural peptides for 6 h, followed by 0.5 µg/ml LPS for 24 h. ROS produced within the intracellular compartment was evaluated by the cellular ROS detection kit (Abcam, Cambridge, United Kingdom) using fluorescence-based microplate platform. The working ROS staining solution was prepared by 4 μl of stock staining solution was freshly diluted with 2 ml of an assay buffer. The intracellular ROS was detected by adding 100 μl of working ROS staining solution into the A549 cells and incubating at 37 °C, 5% CO₂ for 1 h. The fluorescence intensities for DCFH-DA were evaluated using λ_excited at 520 nm and measured λ_emitted at 605 nm using the microplate reader (n = 3). The intracellular ROS was expressed as relative ROS abundant intensity normalized to the fluorescence of the negative control or untreated cells.

Sample preparation for proteomic analysis and LC-MS/MS setting

The effect of the natural peptides from the flower of T. stans on A549 cells was examined using a proteomic approach. The treated cells were extracted using 0.2% SDS, 20 mM DTT, 10 mM HEPES-KOH, pH 8.0 with protease inhibitor cocktail, followed by protein cleaned-up by acetone precipitation (1:5) at −20 °C for 16 h. The protein pellet was resolubilized by 0.15% RapidGest SF in 20 mM ammonium bicarbonate. The amount of 2 µg protein was reduced (5 mM of DTT at 72 °C for 30 min) and alkylated (25 mM IAA at RT in the dark for 30 min) prior to the digestion using trypsin at 37 °C for 6 h. The tryptic peptides were dried and reconstituted in 0.1% formic acid before being subjected to the LC-MS/MS.

Data acquisition in LC-MS runs was operated in the data-dependent acquisition mode. The protein expression was quantified by Progenesis^® QI for proteomics as described previously (Krobthong et al., 2021). The raw MS files were identified against the Uniprot human database (Homo sapiens retrieved 29 June 2021). The parameters were set to; MS tolerance, 20 ppm; MS/MS tolerance, 60 ppm; minimum fragment ion matches/peptide, 3; minimum fragment ion matches/protein, 3; minimum peptide ions/protein, 1; digest enzyme, trypsin; cysteine carbamidomethylation as fixed modification and methionine oxidation, asparagine and glutamine deamination as variable modifications.

Bioinformatics analysis

The expression data were normalized and visualized using bioinformatics software. The normalization of relative protein expression data was analyzed using an R package, NormalyzerDE 1.1.13 (Willforss, Chawade & Levander, 2019), in which Quantile-normalization was applied to the relative expression data analysis, after adding “1” to all expression values to avoid errors upon log-transformation. The expression heatmap was obtained using MultiExperiment Viewer (MeV) for self-organizing tree clustering based on the expression data. Volcano plot analysis was obtained using plotting between the magnitude changes of proteins (log₂ fold change) and statistical significance from ANOVA-test (-log ANOVA) in the experimental groups. Proteins having relative abundance with ANOVA < 0.05 between 6 LC-runs (n = 3) are considered differentially expressed.

Statistical analysis

All experiments were carried out with three independent replicates (n = 3), and the results were reported as mean ± standard deviation (SD). One-way ANOVA with Tukey’s post-hoc test was performed using SPSS 18.0 software (SPSS Inc, Chicago, IL, USA) to analyze the statistical significance of differences between experimental groups for cell migration and invasion assays while paired Student’s t-test was used for intracellular ROS analysis, where asterisk (*, **, ***) indicated p-value of <0.05, <0.01 and <0.001, respectively. One-way analysis of variance was performed by Progenesis^® QI for proteomic analysis.

T. stans-derived native peptide isolation and purification

In our study, we initially used the hot water extraction process to release the natural peptides in the flowers of T. stans (Fig. 1A). The natural peptides extracted from the flowers were purified using size-exclusion and reverse-phase approaches. After the purification steps, the eluent fractions were subjected to LC-MS/MS to explore the peptide profile (Fig. 1B). The natural peptides profile analysis revealed 126 peptide ions peaks. The highest peptide ion was eluted at 36.5 min with an intensity = 9.18e⁹ AU. Analysis of natural peptides, the value of peptide ions found in MS1 were verified using ion peaks of MS2 that resulted in peptides sequences with an ion-fragmentation evidence.

The de novo algorithm was used to predict the sequence originating from the peptide ion peak. The LC-MS/MS results contained 4,230 MS/MS spectra of +2 and +3 charged ions. Matching of the peptide spectrum revealed 126 peptides with lengths in the range of 6–44 amino acids and a molecular weight of 818.49–4,451.30 Da. The numbers reported in Fig. 1 correspond to the compounds shown in Table S1. The QC was done using qualitative analysis for different sampling batches. The peptide profile alignment of 3-batches revealed a high level of similarity (>95%).

Cytotoxicity of natural peptides from T. stans in human cancer cell lines

The human non-small cell lung cancer, A549 cells, were undergone a cytotoxicity test. Cell viability was determined by MTT assay, using logarithmically growing A549 cells treated with various concentrations of the natural peptides for 24 h. The natural peptides induced cell cytotoxicity in the A549 cells (cell viability <60%) for concentrations >0.125 ng/ml (Fig. 2A). However, at the lowest tested concentration of 0.0625 ng/ml, the cell viability was 82.6 ± 3.4%. The percentage of A549 cell viability as the action of natural peptide concentrations was plotted in a sigmoidal curve. The sigmoidal curve of percent cell viability vs concentration showed the IC₅₀ of 0.3321 ng/ml in A549 cells. In addition, we also evaluated the cytotoxicity of natural peptides from T. stans in the other human cancer (HepG2, HeLa, SK-MEL-28, and MCF-7) and immortalized keratinocyte (HaCat) cell lines (Figs. 2B–2F). The IC₅₀ of natural peptides from T. stans in HepG2, HeLa, SK-MEL-28, MCF-7, and HaCat cells was 0.5679, 0.1786, 0.5291, 0.2756, and 3.531 ng/ml, respectively. The higher IC₅₀ dosage in the HaCat cells, transformed from normal keratinocytes, indicated that natural peptides from T. stans have selective dose-dependent cytotoxicity for cancers. The IC₅₀ dosage of A549 cells was chosen for intracellular ROS estimation and proteomics.

Effect of natural peptides from T. stans on A549 cell migration and invasion

The anti-migration ability of natural peptides from T. stans in A549 cells was determined by using a transwell chamber assay. Treatment with natural peptides at the lowest concentration of 0.0625 ng/ml markedly suppressed A549 cell migration (Fig. 3A, left panels) and showed a migration rate of less than 10% (p < 0.0001) (Fig. 3B, left panel). The Matrigel-coated transwell chamber assay was also performed to investigate the anti-invasion ability of natural peptides from T. stans in A549 cells. The result showed that natural peptides at 0.0625 ng/ml completely inhibited A549 cell invasion (Fig. 3A, right panels). The percentage of invaded cells of untreated condition was 38% ± 0.92, while A549 cells treated with natural peptides had completely lost their ability to invade through Matrigel-coated membrane (p < 0.0001) (Fig. 3B, right panel). Collectively, these results indicated that T. stans possesses natural peptides acting with both anti-migration and anti-invasion abilities in A549 cells.

In vitro antioxidative activity using DPPH, ABTS and FRAP assays

The natural peptides from the flower of T. stans exhibited antioxidative ability. Ten micrograms of natural peptides were used to determine the free radical scavenging capacity relative to ascorbic acid, gallic acid, and FeSO₄. The relative antioxidative abilities were determined in vitro using DPPH, ABTS and FRAP assays. The scavenging activity of 10 µg natural peptides from T. stans was 0.234 ± 0.02 µg of AsEAC for DPPH assay (Fig. 4A), 0.191 ± 0.03 µg of GAEAC for ABTS assay (Fig. 4B), and 1.026 µM of FeSO₄ equivalent for FRAP assays (Fig. 4C), respectively. This result indicated that the natural peptides from the flower of T. stans are the intensive antioxidative peptides.

Natural peptides from T. stans reduced intracellular ROS in LPS-induced A549 cells

We next investigated the antioxidative activity of natural peptides from the flower of T. stans in LPS-induced A549 cells. The relative ROS within the cells was measured by analyzing intracellular ROS with DCFH-DA. The qualitative analysis of relative ROS generation induced by LPS in A549 cells was presented in Fig. 4D. Compared to the control group, the intracellular ROS of the cells pretreated with natural peptides from T. stans at IC₅₀ dosage significantly decreased by about 26.53 ± 2.91% (p < 0.0001) after treatment with LPS. The protective intracellular ROS production indicated that the natural peptides from the flower of T. stans are capable of reducing free radicals in the cells.

Proteomic analysis

Changes in the cellular proteins affected by the natural peptides from the flower of T. stans were studied using label-free quantitative proteomics. A total of 1,604 differentially expressed proteins were successfully identified. The relative expression ratios from both T. stans and control groups were analyzed using the Self Organizing Tree Algorithm (SOTA) to get a holistic view of the proteome expression pattern. After clustering, eleven distinct clusters were constructed based on the expression pattern (Fig. 5A).

The columns in the SOTA dendrogram represent the relative protein expression of the treatment (n = 3) and control (n = 3) groups. Meanwhile, the rows represent the average abundance level of the proteins in each cluster. Green denotes higher expression levels, while red denotes lower expression levels. The analysis revealed that cluster ID two has the highest cluster diversity (3.6533). There were five proteins in this cluster, including acetolactate synthase-like protein (ILVBL_HUMAN), calcium load-activated calcium channel (TMCO1_HUMAN), diacylglycerol kinase-α (DGKA_HUMAN), ectoderm-neural cortex protein 1 (ENC1_HUMAN), and nuclear cap-binding protein subunit 2 (NCBP2_HUMAN).

The differential expression ratio from the quantification of 1,604 proteins ranged from 0.01- to 7.43-fold changes as shown in Data S1. A total of 15 proteins exhibited a two-fold change between the two sample groups, while only seven proteins passed the two-fold change with ANOVA < 0.001 cutoffs. These results are illustrated as a volcano plot (Fig. 5B). Among these seven proteins, two were up-regulated while five were down-regulated. Identities of protein details that showed the differential expression of more than two-fold changes between T. stans and control are listed in Table 1.

According to Table 1 and Fig. 5B, the relative protein expressions of nuclear cap-binding protein subunit 2 (NCBP2_HUMAN), peptidyl-glycine alpha-amidating monooxygenase (AMD_HUMAN), endogenous retrovirus group MER34 member 1 Env polyprotein (MER34_HUMAN), ectoderm-neural cortex protein 1 (ENC1_HUMAN), and cytochrome c oxidase assembly factor 4 homolog (COA4_HUMAN) under the treatment condition were lower than that of the control by 7.430, 3.027, 2.789, 2.679 and 2.366 (log₂) fold changes, respectively. Meanwhile, the relative protein expressions of ankyrin repeat and SOCS box protein 6 (ASB6_HUMAN) and alpha-1B-glycoprotein (A1BG_HUMAN) in the treatment group were higher than that of the control by 2.542 and 2.282 (log₂) fold changes, respectively.