Introduction

The incidence of preterm birth has increased over the past 20 years in most countries [1, 2]. Despite recent advances in perinatal medicine, severe diseases related to premature birth, including periventricular leukomalacia (PVL), bronchopulmonary dysplasia (BPD), necrotizing enterocolitis (NEC), and retinopathy of prematurity (ROP), remain major causes of mortality and morbidity, which represent a heavy burden for families and society [3]. Therefore, it is an urgent and significant task to develop new safe and effective treatments to improve the prognosis of these diseases in premature infants.

In the past several decades, the development of mesenchymal stem cell (MSC) therapy and its continuous advancement have gained extensive attention. MSCs are multipotent progenitor cells, which can be raised from different tissues, for instance, adipose tissue, umbilical cord, and bone marrow [4, 5]. Human umbilical cord MSCs (hUC-MSCs) are easily accessible and can be harvested from donors without risks or damage [6]. Additionally, the therapeutic application of MSCs is not limited by the aging-like nature of adult tissues such as bone marrow and adipose tissue [7, 8]. Mechanically, MSCs function in vivo via direct differentiation or paracrine action. The therapeutic potential of MSC engraftment has been proved in premature infant diseases, and early clinical trials in preterm neonates with BPD (NCT01297205 [9], NCT01632475 [10]) and severe intraventricular hemorrhage (NCT02274428 [11]) have been conducted. A myriad of bioactive factors are readily available in the conditioned medium (CM) of MSCs and the medium can mediate multiple known functions of MSCs, such as angiogenesis, anti-fibrosis, and antiinflammatory effects [12]. Extracellular vesicles such as exosomes have been isolated from CM, and they have been shown to contain microRNAs and proteins, which partially mediated the effects of MSC [13,14,15]. Many studies have established that the secretome from MSCs can reduce organ damage in animal models of PVL, BPD, NEC, and ROP [16,17,18,19,20]. Other studies have also shown that soluble factors such as heme oxygenase-1 (HO-1) and erythropoietin (EPO) may be principally responsible for the ability of MSC-CM to ameliorate inflammation, angiogenesis, fibrosis, and so on [21, 22]. More types of MSC secreted factors and regulatory mechanisms still need to be established.

Peptides, a type of compound with two or more amino acids connected by peptide bonds, have been shown to play important roles in the treatment of diseases. Glucagon-like peptide-1 (GLP-1), a well-known peptide hormone secreted from the L cells of the duodenum, colon, terminal ileum, and rectal mucosa, has been used in the clinical treatment of type 2 diabetes [23]. Extrinsic calcitonin gene-related peptide (CGRP) could suppress apoptosis, oxidative stress, and ROS production in hyperoxia-induced alveolar epithelium type II (AECII) cells [24]. WKYMVm hexapeptide could attenuate hyperoxia-induced lung injuries in newborn mice [25]. Additionally, peptides from human milk such as PDC213, β-casein 197, and Casein201 exhibited obvious antimicrobial effects on the common pathogenic bacterial species S. aureus and Y. enterocolitica in neonatal intensive care units [26,27,28]. These studies indicated that peptides may hold great promise for the treatment of premature infant diseases. However, the secreted peptidomic profile from hUC-MSCs has not been fully characterized.

In the present study, we compared the secreted peptides from preterm and term hUC-MSCs using the tandem mass tag (TMT) labeling method with liquid chromatography (LC)-tandem mass spectrometry (MS/MS) analysis. Moreover, using ingenuity pathway analysis (IPA) software, we predicted that the differentially expressed peptides are associated with developmental disorder, inflammatory response, and organismal injury. And we preliminarily investigated the antiinflammatory effect of differentially expressed peptides on human lung epithelial cells. This study helps to broaden the knowledge of the hUC-MSC secretome at the peptide level and provided potential clues for the treatment of respiratory diseases in premature infants.

Materials and methods

Patients and samples

Umbilical cords were obtained from six infants without genetic or structural anomalies delivered at 27–41 weeks of gestation with parental written consent. Cases involving maternal diabetes, pre-eclampsia, eclampsia, intrauterine growth retardation (IUGR), or infectious diseases were excluded, because these factors may influence cell proliferation, cytokine expression, and other functions [29,30,31,32,33]. This study was approved by the Ethics Committee of Changzhou Maternal and Child Health Care Hospital (approval number 2019126) and conducted in accordance with the approved guidelines.

Preparation of hUC-MSCs and CM

Human umbilical cords were collected after preterm (n = 3) or full term (n = 3) deliveries hUC-MSCs were obtained by the tissue explants adherent method, as previously reported [17, 34]. Each umbilical cord (about 10 cm) was washed in phosphate-buffered saline (PBS; Gibco, Grand Island, CA, USA) and with 1% penicillin/streptomycin (P/S; Gibco) to remove residual blood from the vein and arteries. After the cord was cut longitudinally, and the arteries and vein were removed, Wharton’s jelly was finely dissected into small pieces. The pieces were individually placed on 100-mm2 tissue culture dishes with Dulbecco’s modified Eagle medium/nutrient mixture F-12 (DMEM/F-12; Gibco) containing 10% fetal bovine serum (FBS; Gibco), and 1% P/S (Gibco) and incubated for 10–12 days at 37 °C with 5% CO2. The medium was subsequently exchanged every 2–3 days. The cultures were passaged when they reached 80–90% confluency after carefully removing the umbilical cord tissues. hUC-MSCs at passage 3 were cultured to 80–90% confluence in a T75 culture flask (about 106 cells). The complete medium was replaced with serum-free DMEM/F-12 medium (5 ml) to avoid peptides contamination from FBS. The collected serum-free medium was centrifuged for 10 min at 300×g at 4 °C to remove cell debris, and protease inhibitor (Roche, Basel, Switzerland) was added. Lastly, the mixture was snap-frozen in liquid nitrogen and stored at − 80 °C until used.

Peptide extraction and purification

Before peptide extraction, the protein integrity of the CM samples was appraised. The collected CM samples were concentrated by centrifugation under vacuum (LaboGene, Allerød, Denmark), boiled in sodium dodecyl sulfate (SDS)-sample buffer at 95 °C for 10 min and then subjected to 12% SDS-polyacrylamide gel electrophoresis (PAGE). The SDS-PAGE gel was stained using a Pierce Silver Stain Kit (Thermo Fisher Scientific, Waltham, MA, USA) following the manufacturer’s protocol. Thereafter, the samples were filtered through an ultrafiltration tube (Amicon Ultra-15, Millipore, MA, USA) with a molecular weight cutoff (MWCO) of 10 kDa to acquire the filtered liquid containing the peptides. The protein concentration of the supernatant was also measured using the Bradford protein assay [35].

TMT labeling and LC-MS/MS analysis

The peptides from preterm and term hUC-MSC CM were reduced with 10 mM DL-dithiothreitol (DTT; Promega, WI, USA) for 1 h at 56 °C and alkylated with 55 mM iodoacetamide (Promega) for 1 h in the dark at RT. Thereafter, precooled acetone was added, and the peptides were precipitated over 3 h at − 20 °C. After centrifuging for 20 min at 20,000×g and 4 °C, the precipitate was dissolved in 300 μl of the following buffer: 50% triethylamine borane (Sigma) and 0.1% SDS (Sigma). Next, the peptide solution was desalted using a Strata-X C18 column (Phenomenex, Torrance, CA, USA) and dried and labeled with TMT reagent (TMT 6-plex Label Reagent; Thermo Fisher Scientific) for 1 h [36]. Next, the preterm and term samples were mixed at a 1:1 ratio on the basis of the total peptide amount. Analysis of labeled peptides was performed on a Q Exactive Orbitrap LC-MS/MS system (Thermo Fisher Scientific). Qualitative and relative quantitative analyses of the detected peptides were performed using the SWISSPROT_human database and Mascot software (version 2.3.01). Peptides with absolute fold change ≥ 1.5 and P value < 0.05 were considered differentially expressed.

Bioinformatics analysis

The molecular weight (MW) and isoelectric point (PI) of the identified peptides were calculated using the online tool PI/MW (http://web.expasy.org/compute.pi/). A Gene Ontology (GO) analysis (http://www.blast2go.com/b2ghome) was carried out to explore the possible cellular components, biological processes, and molecular functions related to the precursor proteins. Diseases and regulator effects networks analysis of the differentially expressed peptides and their precursors were performed using ingenuity pathway analysis (IPA) software v7.1 (Ingenuity Systems, Mountain View, CA, USA) [37]. The UniProt database (http://www.uniprot.org/) was used to detect the predominant subcellular locations of the precursors of the differentially expressed peptides. The Open Targets Platform database (http://www.targetvalidation.org/) was applied to study the diseases associated with the protein precursors [38].

Synthetic peptides

All the peptides used in this study were synthesized by GenScript (Nanjing, Jiangsu, China) through the solid-phase method. The purity of each peptide was 95% detected by HPLC-MS method. All the used peptides were preserved in freeze-drying at − 20 °C until immediately dissolved in aseptic water for cell treatment in vitro.

Cell culture

Two peptides TGAKIKLVGT and AAAAGPANVH were selected with high fold change, and we investigated the effects on human lung epithelial cells A549 (Meiyan, Shanghai, China) stimulated by hydrogen peroxide (H2O2; Kelong, Chengdu, Sichuan, China). A549 cell is a human lung carcinoma cell line from an adult with similar characteristics of human alveolar basal epithelial cells, and was often used for researches in BPD. A549 cells were cultured in Dulbecco’s modified Eagle medium (Gibco) with 10% FBS and 1% P/S. The A549 cells were exposed to 1 mM H2O2 with or without peptides (1 μM, 10 μM, and 100 μM) for 24 h in serum-free DMEM with 1% P/S. The sample size was 3 biological independent samples per group.

Quantitative real-time polymerase chain reaction (QRT-PCR)

Total RNA was extracted with Trizol reagent (Invitrogen; Thermo Fisher Scientific Inc., Shanghai, China) according to the manufacturer’s instructions. The RNA concentration was detected by ultraviolet spectrophotometer and the appropriate OD value at 260 to 280 nm was 1.8 to 2.0. Then, qRT-PCR was carried out to measure the gene expression levels of inflammatory factors: tumor necrosis factor-α (TNF-α), interleukin-1β (IL-1β), and interleukin-6 (IL-6) mRNAs in A549 cells using SYBR Green qPCR method (Thermo Fisher Scientific, Waltham, MA). Primer sequences were the following: human TNF-α: forward, 5′ CCTCTCTCTAATCAGCCCTCTG 3′, reverse, 5′ GAGGACCTGGGAGTAGATGAG 3′; human IL-1β: forward, 5′ AGCTACGAATCTCCGACCAC 3′, reverse, 5′ CGTTATCCCATGTGTCGAAGAA 3′; human IL-6: forward, 5′ ACTCACCTCTTCAGAACGAATTG 3′, reverse, 5′ CCATCTTTGGAAGGTTCAGGTTG 3′; and human β-actin: forward, 5′ AGCGAGCATCCCCCAAAGTT 3′, reverse, 5′ GGGCACGAAGGCTCATCATT 3′. To calculate fold change in the expression of these genes, ΔCt = Ct of individual genes − Ct of β-actin was first obtained. ΔΔCt = ΔCt of treated groups − ΔCt of control groups was then obtained. Fold change was calculated as 2−ΔΔCt, with control groups as 1.0-fold.

Western blot analysis

Lysates from cultured cells were similarly prepared using a cell scraper. Homogenates were clarified by centrifugation (10,000×g, 4 °C, 10 min). Protein concentration was quantified with a BCA protein assay (Beyotime, Shanghai, China). Cell extracts resolved on a 10% or 12% reducing SDS-PAGE gel were transferred to a nitrocellulose membrane according to molecular weight. Blots were probed with the following antibodies: rabbit anti-TNF-α (1:2000; Proteintech, Wuhan, China), anti-IL-1β (1:1000; Proteintech, Wuhan, China), and mouse anti-IL-6 (1:5000; Proteintech, Wuhan, China), whereas mouse GAPDH (1:5000; Proteintech, Wuhan, China) served as a loading control.

Enzyme-linked immunosorbent assay (ELISA)

ELISA was also performed to detect the expression levels of inflammatory factors in A549 cells by ELISA kits for human TNF-α, IL-1β, and IL-6 (4A Biotech, Beijing, China), according to the manufacturer’s instructions. Optical density was measured at 450 nm using an ELISA microplate reader. No significant cross-reactivity or interference was observed.

Statistical analysis

Student’s t test or one-way ANOVA was employed for statistical comparisons. The results of the bioinformatics analysis were visualized using GraphPad Prism 5/7 software. The statistical significances were calculated as P values, and P < 0.05 was considered statistically significant.

Results

Isolation and characterization of MSCs derived from human umbilical cords

The range of gestational ages (GA) was 31–32 weeks of gestation for the preterm umbilical cords and 40–41 weeks for the term cords (Table 1). The hUC-MSCs (obtained by the tissue explants adherent method) had a typical fibroblast phenotype (Fig. S1A). Both preterm and term hUC-MSCs were positive for CD29, CD73, and CD105 staining and negative for CD31, CD34, and HLA-DR staining (Fig. S1B). Furthermore, these isolated cells had the potential to differentiate into adipocytes and chondrocytes (Fig. S1A). These results confirmed that MSCs from human umbilical cords were successfully isolated, without significant differences in morphology, expression of cell surface markers or differentiation capacities between the preterm and term groups.

Table 1 Cohort clinical date of hUC-MSC samples

Identification of differentially expressed peptides in hUC-MSC CM from preterm and term infants

We verified the protein integrity of the hUC-MSC CM by silver staining (Fig. S2). The peptides from preterm and term hUC-MSC CM were directly analyzed by the TMT labeling method combined with LC-MS/MS. We identified a total of 3099 peptides in hUC-MSC CM from both groups. A total of 131 peptides were observed to be significantly differentially expressed (absolute fold change ≥ 1.5, P value < 0.05) in the hUC-MSC CM from the preterm group compared with the term group, comprising 37 upregulated peptides (Fig. 1a) and 94 downregulated peptides (Fig. 1b). The top 20 upregulated and top 20 downregulated peptides are shown with their precursor proteins in Fig. 1c and d. All the differentially expressed peptides are shown in Table 2.

Fig. 1
figure 1

Differentially expressed peptides in hUC-MSC conditioned medium (CM) from preterm infants compared with term infants. a, b Upregulated and downregulated peptides visualized using heatmaps (n = 3 per group, P1–3 represent preterm infants and T1–3 represent term infants). c, d Top 20 upregulated and top 20 downregulated peptides visualized using heatmaps

Table 2 Differentially expressed peptides in hUC-MSC conditioned medium from preterm and term infants

Basic characteristics of the differentially expressed peptides in hUC-MSC CM from preterm and term infants

The MW and PI of the differentially expressed peptides were analyzed. The MW of most peptides ranged from 500 to 700 Da (Fig. 2a), and the PI ranged from 3 to 11 (Fig. 2b). We also investigated the distribution of the MW relative to the PI (Fig. 2c). Peptides are cleaved from their precursor proteins by specific enzymes [37], so we analyzed the cleavage sites at the N- and C-terminals of the identified peptides. Lysine (K) was the most common N-terminal amino acid (accounting for 13.7% of the peptides), while asparagine (N) was the most common C-terminal amino acid (accounting for 16.0% of the peptides) (Fig. 2d).

Fig. 2
figure 2

Basic features of the differentially expressed peptides in hUC-MSC CM from preterm and term infants. a, b Molecular weights (MW) and isoelectric points (PI) of the differentially expressed peptides. c Scatter plot of MW versus PI of the differentially expressed peptides. d Distributions of the N- and C-terminals of the differentially expressed peptides

GO analysis and subcellular location analysis of the differentially expressed peptide precursors

Next, molecular functions, cellular components and biological processes of the corresponding precursor proteins were determined by GO analysis to predict the latent functions of the differentially expressed peptides. Binding and catalytic activity were the most highly enriched molecular functions (Fig. 3a). Cell part, organelle part, and intrinsic component of membrane were the most highly enriched cellular components (Fig. 3b). Cellular process, biological regulation, and cellular component organization were the most highly enriched biological processes (Fig. 3c). Furthermore, we categorized the subcellular locations of the precursor proteins of all 131 peptides in accordance with their annotations in the UniProt database. The analysis revealed that the nucleus (25%), plasma membrane (16%), and cytoskeleton (15%) were the predominant subcellular locations of the differentially expressed peptide precursors. About 10% of the precursors were types of proteins that are located in the extracellular region of hUC-MSCs (Fig. 3d).

Fig. 3
figure 3

Gene Ontology (GO) and subcellular location analysis of the differentially expressed peptide precursors. a Molecular functions. b Cellular components. c Biological processes. d Subcellular locations

Diseases and regulator effects networks associated with the differentially expressed peptide precursors

We further evaluated the diseases and regulator effects networks associated with the differentially expressed peptide precursors using IPA software. Disease and functional protein network analysis indicated that several precursor proteins were involved in developmental disorders and inflammatory responses (Fig. 4a, b). More precisely, precursor proteins including Alpha-2A adrenergic receptor (ADRA2A), Protein argonaute-2 (AGO2), Baculoviral IAP repeat-containing protein 6 (BIRC6), Kalirin (KALRN), and Histone-lysine N-methyltransferase 2C (KMT2C) were involved in developmental disorders, and KMT2C, Solute carrier family 2 (SLC2A4), Electrogenic sodium bicarbonate cotransporter 1 (SLC4A4), STIM1L (STIM1), and Rootletin (CROCC) were involved in inflammatory responses. All the putative precursor proteins associated with diseases are shown in Table S1. Furthermore, the regulator effects network analysis showed that some of the protein precursors participated in the networks of cellular development, embryonic development, organismal development, and organismal injury and abnormalities (Fig. 4c, d). For example, AGO2, CROCC, DENN domain-containing protein 2A (DENND2A), Krueppel-like factor 14 (KLF14), and Lon protease homolog (LONP1) were involved in the networks of cellular development, embryonic development, and organismal development. Additionally, Atrophin-1 (ATN1), Collagen alpha-1 (VIII) chain (COL8A1), Protein jagged-2 (JAG2), KMT2C, and Mucin-19 (MUC19) were related to the network of organismal injury and abnormalities. All the precursor proteins involved in regulator effects networks are shown in Table S2.

Fig. 4
figure 4

Diseases and regulator effects networks associated with the differentially expressed peptide precursors. Functional proteins related to a developmental disorder and b inflammatory response. Regulator effects network related to c cellular development, embryonic development, and organismal development and d organismal injury and abnormalities. Precursor proteins, diseases, and functions are shown as nodes, and the biological relationships between nodes are represented as lines with arrows. All lines are supported by at least one literature reference from the ingenuity pathway analysis (IPA) analysis. The intensity of the node color indicates the degree of upregulation (red) or downregulation (green)

Putative bioactive peptides associated with respiratory diseases

It is well known that peptides with biological functions have functions that are related to the functions of their precursor proteins, with domains playing key roles in the biological functions [39, 40]. The UniProt database was used to analyze the peptides and their precursors and the results showed that 25 peptides were located in the functional domains of their corresponding precursors (Table S3). The preceding results suggested that these precursors are mainly associated with inflammatory responses and abnormal organ development, which contribute to premature infant respiratory diseases. Therefore, we focused on peptides and their precursors related to respiratory diseases. Using the Open Targets Platform database, we investigated whether these peptides might play potential roles in respiratory diseases. All told, 17 precursor proteins were found to be closely related to respiratory diseases (association score ≥ 0.5) (Table S4).

TNF-α, IL-1β, and IL-6 in H2O2-treated lung epithelial cells with differentially expressed peptides

Previous researches have reported that lung epithelial cells A549 can be stimulated by H2O2 to induce inflammatory response [41, 42]. Therefore, we explored the effect of differentially expressed peptides on inflammatory response of H2O2-treated A549 cells. As observed, TNF-α, IL-1β, and IL-6 mRNA and protein expression levels were higher in A549 cells from the H2O2 (1 mM) group compared with the control group (Fig. 5), respectively, measured by qRT-PCR, ELISA, and western blot. According to the high fold change, we selected two peptides 508AAAAGPANVH517 derived from Homeobox protein SIX5 (SIX5, absolute fold change: 5.9) and 7118TGAKIKLVGT7127 derived from MUC19 (absolute fold change, 6.3) for future study. Peptide 7118TGAKIKLVGT7127 (MUC19) at 10 and 100 μM significantly attenuated the H2O2-induced increase of TNF-α, IL-1β, and IL-6. Moreover, we observed that the levels of inflammatory cytokines decreased more obviously with the 7118TGAKIKLVGT7127 at the concentration of 100 μM than 10 μM (Fig. 5). Meanwhile TNF-α, IL-1β, and IL-6 were significantly reduced in H2O2-treated A549 cells with peptide 508AAAAGPANVH517 derived from SIX5 at concentration of 100 μM, but not at 1 μM or 10 μM (Fig. 5). Both the two peptides help to suppress inflammatory response in H2O2-treated A549 cells.

Fig. 5
figure 5

Function analysis of differentially expressed peptides in vitro. (a) TNF-α, IL-1β, and IL-6 mRNA expression assessed by qRT-PCR in A549 cells stimulated by H2O2 (1 mM) with or without peptides 7118TGAKIKLVGT7127 (MUC19) and 508AAAAGPANVH517 (SIX5) at concentrations of 1, 10, and 100 μM. b The protein levels of TNF-α, IL-1β, and IL-6 measured by ELISA analyze in H2O2-treated A549 cells with or without peptides. c, d The protein levels of TNF-α, IL-1β, and IL-6 measured by western blot in H2O2-treated A549 cells with or without peptides (n = 3 biological independent samples per group in qRT-PCR and WB, technical replication = 3 in ELISA. *P < 0.05, **P < 0.01, ***P < 0.001)

Discussion

The MSC secretome and its therapeutic effects have been extensively demonstrated in preterm diseases, such as BPD [43] and NEC [19]. Previously, most researchers considered soluble factors and extracellular vesicles as the primary components of the secretome derived from MSCs [44,45,46]. However, the paracrine substances from MSCs are not limited to these biomolecules. Therefore, further studies are required to explore more types of components derived from MSCs and investigate their potential functions.

With the advance of detection technologies, mounting evidence has confirmed the differences in hUC-MSCs between preterm and term groups, which may help to identify the possible regulators or mechanisms underlying MSC function. To compare the global gene expression patterns in hUC-MSCs between these two groups, Iwatani et al. used microarray analysis and revealed that upregulated WNT2B in preterm hUC-MSCs was involved in the control of hUC-MSC proliferation [47]. A very recent study comparing hUC-MSC transcriptomics and proteomics profiles from term and preterm groups showed that Frizzled-2 (FZD2) protein and mRNA expression levels were both higher in preterm hUC-MSCs [48]. Importantly, FZD2 is the receptor of Wnt5a/b and FZD2 mutations influence Wnt signaling, which mediates the epithelial to mesenchymal transition (EMT) during lung development [49]. In addition, by comparing the proteome of microvesicles collected from hUC-MSC CM between preterm and term groups, Bruschi et al. found that 173 proteins were significantly changed, 163 of which were increased in the preterm group [50]. However, there have been no comprehensive comparisons of hUC-MSC CM peptidomic profiles between preterm and term infants. In the present study, we found that 131 peptides derived from 106 precursor proteins were differentially expressed in the preterm hUC-MSC CM compared with the term group by TMT labeling quantification (Fig. 1). Our study provides hUC-MSC CM polypeptide profiles for preterm and term infants. Secreted peptides have been shown to have important biological functions. Neuropeptide Y, a 36-amino acid peptide secreted by the hypothalamus, was found to play key roles in neurodegenerative diseases including modulation of neurogenesis, food intake, and thermogenesis [51]. Mao et al. showed that peptides derived from human beta-defensins are secreted by viable human cryopreserved amniotic membrane and exhibited direct antimicrobial effects against P. aeruginosa [52]. Further studies are needed to understand and explore the functions of secreted peptides from hUC-MSCs. In our study, we used p value to screen differentially expressed peptides between term and preterm groups. Although the false discovery rate (FDR) can detect differentially expressed peptides more accurately than p value, we would get very few peptides by FDR. Additionally, in order to identify more differentially expressed peptides, some studies also applied the p value to the analysis of peptides [53, 54]. Thus, in some condition, it is also appropriate to use the p value.

Subcellular location analysis of precursor proteins can help to better understand the source and potential functions of peptides. As shown in Fig. 3d, most precursor proteins were annotated as being part of organelles or membranes (56%). Notably, a small fraction of peptides were annotated as being derived from proteins located in the extracellular region, termed secreted proteins from hUC-MSCs. Classically, a large proportion of peptides (such as peptide neurotransmitters) are generated by the proteolysis of macromolecular proteins followed by release into the space outside of cells [55]. However, some other peptides containing one or more cleavage sites do not derive from endosomal processing [56]. These peptides are the N- or C-terminal peptides of their precursor proteins, rather than internal fragments [57]. In addition, the identification in our study of several peptides arising from secreted proteins raised the possibility that some bioactive peptides may be produced by enzymatic hydrolysis of extracellular proteins. These observations provide us with more methods to evaluate the potential functions of differentially expressed peptides in our study; additional studies are needed to ascertain whether these peptides are actually secreted from hUC-MSCs and have biological activities. From the results of IPA analysis, we found that a series of precursor proteins were involved in networks of developmental disorders, inflammatory responses, and organismal injury and abnormalities associated with premature infant diseases and most of these precursor proteins participated in the course of premature infant diseases according to the Open Targets Platform database. These observations raised the possibility that the peptides secreted by hUC-MSCs may have beneficial effects with a similar role of precursor proteins in neonatal respiratory diseases, which are worthy of in-depth functional studies.

As the functions of secretory peptides derived from hUC-MSCs were unclear, we assessed whether the peptides were located in the functional domains of their precursor proteins to analyze their functionality. Using the UniProt database, we discovered that 25 peptides were situated in the functional domains of their corresponding precursors. Peptides derived from Titin (8990ESDSG8994, 26941EGNKDD26946, 6684SSRLECKI6691, and 19103VVHAGGVIRIIAYV19116) and one peptide derived from striated muscle preferentially expressed protein kinase (2669SCTVAVARVPGKLAPPEVPQ2688) were located in Ig-like domains. The Ig-like domain was initially characterized as a structure composed of two sheets of antiparallel β strands [58]. Okano et al. reported that the Ig-like domain contributed to the maintenance of the structure, activity, and stability of metagenome-derived glycoside hydrolase family 9 endoglucanase [59]. Interestingly, the Ig-like domain made endoglucanase Cel9A from Alicyclobacillus acidocaldarius dependent on calcium [60]. Additionally, Ig-like domains have been shown to play various roles in functions [61, 62]. We also found that two peptides derived from MUC19 (961DDFMSSQN968 and 1382QNGIIVI1388) were located in the type D von Willebrand factor (VWFD) domain. This domain plays a substantial role in reducing soluble VWF binding to platelet GpIba and regulates platelet activation and adhesion [63, 64]. It also participates in fertilization, as it binds to sperm proteases [65]. Like the Ig-like domain, the VWFD domain is partially responsible for biological processes and functions. The above motif analysis may provide a new perspective for clarifying the possible roles of these newly identified peptides.

Based on the above results, we focused on exploring the potential role of differentially expressed peptides in developmental disorders and inflammatory responses. BPD is clearly one of the most common respiratory morbidity in preterm infants, and it may not simply be a consequence of lung immaturity [66]. And inflammation, one key contributor, often initiated by a pulmonary fetal inflammatory response, is aggravated by invasive or non-invasive mechanical ventilation and exposure to hyperoxia [67]. Given the large predominance of inflammation in the pathogenesis of BPD, we focused on the effect of peptides on inflammation response. A549 cells, a human lung carcinoma cell line, are often used for researches in pathological mechanism and treatment of BPD by A549 cells [68, 69]. And in our study, H2O2, known as a strong two-electron oxidant, could induce the production for TNF-α, IL-1β, and IL-6 in human lung epithelial cell line A549 in agreement with previous studies [70]. As known to all, primary human AECII are difficult to isolate and culture and also are mutable and unsuitable for experimental intervention, while A549 cells have similar biological characteristics to AECII cells [69], although embryonic alveolar epithelial cells are more in line with the purpose of the project and can be better used to evaluate the effects of peptides on the characteristics of lung development. However, this cell is difficult to obtain and easy to mutate [69]. Thus, we chose human lung epithelial cells A549 to explore the effects of differentially expressed peptides in vitro. Accordingly, to make a full understanding of these peptides, animal experiments will be carried out to support our researches in the following experiments.

We also observed that 7118TGAKIKLVGT7127 derived from MUC19 reduced the levels of TNF-α, IL-1β, and IL-6 in H2O2-treated A549 cells and the degree of decrease is related to the concentration of peptide as shown in Fig. 5. As a secreted mucin, MUC19 is released to the extracellular medium and has been identified in respiratory, digestive, and reproductive tracts [71]. It has been reported that MUC19 was differentially regulated after exposure to inflammatory cytokines [72]. And one recent study found that MUC19 peptides may enhance vaginal mucous immunity against infections [73]. These results indicated that 7118TGAKIKLVGT7127 derived from MUC19 may also play a role in protecting from inflammatory response stimulated by H2O2 in human lung epithelial cells. In addition, the other peptide 508AAAAGPANVH517 derived from SIX5 also reduced the expression levels of TNF-α, IL-1β, and IL-6 in H2O2-treated A549 cells only at the maximum concentration of 100 μM. It has been acknowledged that SIX5 was correlated with eye development [74] and myotonic dystrophy [75], which are related to embryonic and organismal development. Thus, we put forward one hypothesis that the hUC-MSCs from preterm infants may secrete protective substances such as peptides under stress. These findings indicated that two peptides may play an antiinflammatory role in the process of BPD. And previous studies have shown that the regulation of inflammation-related signal pathways such as p38 mitogen activated protein kinases (p38MAPK) signal pathways and nuclear factor-κB (NF-κB) signal pathway can reduce the level of pro-inflammatory cytokines, thus relieving the pulmonary inflammation of BPD [76, 77]. Combined with previous studies, we consider that the two peptides secreted from hUC-MSCs may decrease the production of pro-inflammatory cytokines through inflammation-related signal pathways and further researches need to be conducted to validate the possible pathway.

Conclusion

As far as we know, no large-scale quantitative peptidomic analysis has been carried out on the secretory components of hUC-MSCs. Our study identified the differentially expressed peptides secreted by preterm and term hUC-MSCs using TMT-based LC-MS/MS technology. Furthermore, bioinformatics analysis of precursors predicted the possible functions of peptides that may be useful in the treatment of premature respiratory diseases in connection with inflammatory responses and developmental disorders. And we first investigated the antiinflammatory effect of the peptides 7118TGAKIKLVGT7127 derived from MUC19 and 508AAAAGPANVH517 derived from SIX5 on human lung epithelial cells. This study expands our knowledge of the hUC-MSC secretome and may provide insights into new therapy for premature respiratory diseases.