Abstract Background Women who develop diabetes during pregnancy are at higher risk of preterm birth. Here, we identified differentially expressed proteins (DEPs) in the serum of umbilical cord blood samples obtained from preterm neonates delivered by women with gestational diabetes to provide therapeutic targets for clinical drug development. Materials and Methods Umbilical cord blood was collected after delivery of preterm neonates by women with gestational diabetes and after delivery of healthy neonates by women without diabetes. DEPs in the serum samples were identified using liquid chromatography–tandem mass spectrometry. Gene Ontology (GO), cluster analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) were used to determine the biological functions associated with these DEPs. Enzyme linked immunosorbent assay was used to confirm the key DEPs. Results We found that 21 proteins were significantly upregulated, and 51 proteins were significantly downregulated in 72 DEPs in serum samples. GO analyses showed that the DEPs were mainly associated with the GO terms cellular process, biological regulation, cellular anatomical entity, and binding. KEGG signaling pathway analysis indicated that most of the upregulated DEPs were associated with the complement and coagulation cascades, Staphylococcus aureus infection, pertussis, HIF-1 signaling pathway and PPAR signaling pathway and that most of the downregulated DEPs were associated with the complement and coagulation cascades, dilated cardiomyopathy, pathways in cancer, Chagas disease, and hypertrophic cardiomyopathy. The results of KEGG pathway annotation and enrichment analyses indicated that changes in the complement and coagulation cascades may be importantly associated with preterm delivery of neonates by women with gestational diabetes. The key DEPs were confirmed by enzyme linked immunosorbent assay. Conclusion Our proteomics and bioinformatics analyses identified several key proteins and the complement and coagulation cascades pathway that warrant further investigation as potential novel therapeutic targets in preterm delivery among women with gestational diabetes. Keywords: label-free LC-MS/MS, proteomics, gestational diabetes, preterm birth, complement and coagulation cascades Introduction The number of people with diabetes increases every year, including among pregnant women.[42]^1 An increasing number of women in China have received a diagnosis of type 1 or type 2 diabetes, often before they become pregnant.[43]^2^,[44]^3 A reported prevalence of gestational diabetes is 11.91% in China in 2018, but another study showed that the prevalence of gestational diabetes is various in different cities of China.[45]^4^,[46]^5 One potential reason for an increase in gestational diabetes is that women are waiting to become pregnant at older ages (>35 years).[47]^6 Women who develop diabetes during pregnancy are at higher risk of developing complications related to the perinatal period and to diabetes.[48]^1^,[49]^7 Many pregnancies are unplanned and may occur when diabetes is not well controlled. This leads to higher risk of preterm birth, cesarean delivery, eclampsia, preeclampsia, or complications from advanced diabetes.[50]^1^,[51]^7–9 To discover novel therapeutic targets for clinical drug development and to better control the impact of gestational diabetes on the fetus and the occurrence of diabetes complications, two studies in China have used proteomics techniques to identify metabolic profiling of blood serum from gestational diabetes patients and animal model to find novel biomarkers for precision or personalized diagnosis for gestational diabetes.[52]^10^,[53]^11 In addition, another study using proteomics techniques to identify plasma protein changes of gestational diabetes mellitus at 16‐18 weeks.[54]^12 However, it is still necessary to further understand the mechanisms underlying the development of gestational diabetes and the effect of gestational diabetes on fetus. Preterm birth, accounting for 11% of pregnancies worldwide, is the leading cause of perinatal death.[55]^13 The causes undergirding preterm birth are various and complex and include gestational diabetes, periodontal health, preeclampsia, intrauterine infection, chorioamnionitis, and multiple births.[56]^14^,[57]^15 In China, a study reported that the prevalence of preterm birth of gestational diabetes is 13.41%.[58]^16 Prevention of preterm birth is considered a public health priority because it has the potential to reduce morbidity and mortality in infants and children. Therefore, the discovery of new therapeutic targets and treatment strategies is important for preventing the preterm birth. Proteomics is a robust method used to detect and quantify serum proteins that are expressed differentially in persons with a condition of interest. In the present study, we used label-free proteomics techniques to detect the differentially expressed proteins (DEPs) in neonatal cord blood from women with gestational diabetes who delivered preterm neonates and from women without diabetes who delivered healthy full-term neonates. Using bioinformatics analyses, we identified the functional roles of the various DEPs and their related signaling pathways to find those DEPs that may be key players in the occurrence and development of preterm birth associated with gestational diabetes. Those DEPs warrant further study as potential novel therapeutic targets for the treatment of the preterm birth. Materials and Methods Clinical Information and Sample Collection All our experiments comply with the Declaration of Helsinki. Neonatal cord blood samples were collected from four women with gestational diabetes who delivered preterm neonates and from four women without diabetes who delivered healthy full-term neonates, all of whom had been admitted to the Anhui Province Maternity and Child Health Hospital. All pregnant women have complete clinical medical records and regular pregnancy examination without hereditary diseases. The delivery was successful. In healthy control group, inclusion criteria were: (1) ≥37 weeks gestation; (2) no disease history before or during pregnancy. Exclusion criteria were: (1) <37 weeks gestation; (2) disease history before or during pregnancy; (3) hereditary disease history. In diabetes group, inclusion criteria were: (1) <37 weeks gestation; (2) no history of diabetes before pregnancy; (3) pregnant with diabetes mellitus and using insulin to control blood glucose. Exclusion criteria were: (1) ≥37 weeks gestation; (2) diabetes history before pregnancy; (3) pregnant with diabetes mellitus and blood glucose was not controlled in the standard reference range; (4) hereditary disease history. Gestational diabetes mellitus diagnosis was conducted according to the World Health Organization (WHO) criteria.[59]^17 All women with gestational diabetes were diagnosed at the middle of pregnancy and treated during pregnancy with sufficient insulin to maintain blood glucose levels within the standard reference range. This study was approved by the Medical Ethics Committee of Anhui Province Maternity and Child Health Hospital (approval number YYLL2020-2020xkj235-02-01), and all participants provided written informed consent. The study was carried out from 03/2021 to 12/2021. Inclusion criteria were: (1) all women conformed to the clinical diagnostic criteria; (2) regular pregnancy examination; (3) successful delivery; (4) no hereditary diseases of coagulation system; (5) complete clinical medical records; and (6) pregnant women and family members gave informed consent to participate in this study. Exclusion criteria were: (1) women were diagnosed with diabetes mellitus or impaired glucose regulation before pregnancy; (2) women with other pregnancy-related diseases; (3) heart, liver or kidney dysfunction; (4) hematological diseases; (5) other diseases that may affect blood glucose; and (6) mental illness or retardation. After the neonate was delivered, the blood samples from the umbilical cord were collected and placed at room temperature for 1 h prior to being centrifuged at 3000 × g for 15 min. The obtained serum was stored at −80°C for use in the following experiments. Protein Preparation and Gel Analysis High-abundance proteins in the plasma were removed using a ProteoMiner Protein Enrichment Kit (Bio-Rad). The protein concentrations were determined using the Bradford method. From each serum sample, 35 μg of total protein was mixed with 5 × sample buffer, boiled, and loaded onto 10% gels for sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE). The proteins were electrophoresed at a constant current of 14 mA for 90 min. Coomassie blue was used to stain the resultant protein bands. Enzymatic Hydrolysis of Proteins and High-Performance Liquid Chromatography with Tandem Mass Spectrometry (LC-MS/MS) Analysis A flowchart of the study analyses is provided in [60]Supplementary Figure 1. Dithiothreitol (5 µL of 1 M) was added to samples containing 200 µg of each protein. Iodoacetamide (20 µL of 1 M) was added to the samples following incubation at 37°C for 1 h. After another incubation in the dark at room temperature for 1 h, the samples were centrifuged at 14,000 × g for 20 min at 4°C. The precipitate was collected, and 100 µL of UA buffer (8 M urea, 100 mM Tris-HCl, pH 8.0) was added. After ultracentrifugation, the resulting pellet was digested with trypsin according to the (mass) protein to enzyme ratio of 50:1 for 12–16 h at 37°C. Next, we used nanoliter flow rate, high-performance liquid chromatography with a liquid phase system to separate the samples. Before adding samples, the chromatographic column was balanced with 95% phase A solution (0.1% formic acid in water). The mobile phases included phase A and phase B (0.1% formic acid in acetonitrile). The phase B gradient parameters were 0–2 min for a linear increase from 6% to 9%; 2–10 min for an increase to 13%; 10–50 min for an increase to 26%; 50–70 min for an increase to 38%; and 70–71 min for an increase to 100%, which was maintained for 71–78 min. We used capillary LC-MS/MS with an Orbitrap Fusion Mass Spectrometer (Thermo Scientific) to separate and analyze all samples. Database Retrieval and Bioinformatics Analyses We used Proteome Discoverer, version 2.4 (Thermo Fisher Scientific) software to analyze the LC-MS/MS data. Briefly, human proteome databases containing reviewed UniProt sequences were used to identify peptides with the SEQUST search engine. We used P values < 0.05 and 1.5-fold changes (FC) to define DEPs between the two groups of samples. A FC ≥ 1.5 and P < 0.05 represented upregulated proteins, whereas a FC ≤ 0.667 and P < 0.05 represented downregulated proteins. If the FC was between 0.667 and 1.5 or if P > 0.05, this was defined as no change detected in protein expression levels between the two groups. For bioinformatics analyses, we obtained gene ontology (GO) results of these DEPs using Metascape, which is a web-based resource ([61]http://metascape.org). Through this analysis, DEPs were mapped to the terms in the database, and the number of proteins per term was determined. In addition, the hypergeometric test was used to identify GO entries that were significantly enriched.[62]^18^,[63]^19 The signaling pathways were analyzed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) Orthology-Based Annotation System (KOBAS) online analysis tool ([64]http://kobas.cbi.pku.edu.cn/),[65]^20 through which we identified those signaling pathways that were significantly enriched with DEPs. Enzyme Linked Immunosorbent Assay (ELISA) ELISA experiment was used to confirm the key proteins identified by LC-MS/MS. The kit was purchased from Shanghai Enzyme-linked Biotechnology Co., Ltd. (Shanghai, China). After we collected the blood samples, optical density (OD) values were measured at a wavelength of 450 nm using a microplate reader. The protein concentrations of the blood samples were calculated using a standard curve. Statistical Analysis We performed two-tailed Mann–Whitney and Fisher exact tests with SigmaPlot software. Unless otherwise stated, values are expressed as means ± SEM. A two-sided value of P < 0.05 was considered statistically significant. Results Participants in LC-MS/MS Test No significant differences were found in mother age and weight gain during pregnancy between the four healthy women and the four women with diabetes ([66]Table 1). However, significant differences were detected for the gestational age, birth weight of neonates and body mass index (BMI) before pregnancy ([67]Table 1). The results showed that the gestational age of neonates delivered by women with gestational diabetes was significantly younger, and the birth weight of the preterm neonates delivered by women with gestational diabetes was significantly lower than neonates delivered by healthy women, and BMI before pregnancy of women with gestational diabetes was significantly higher than that in healthy women. Table 1. Demographic Characteristics of Participants Group Identification No. Age of Mother (Years) Gestational Age (Days) Neonate Birth Weight (g) Healthy control 1 32 280 3090 2 33 287 3550 3 31 286 3100 4 35 280 3360 Diabetes 1 35 238 2380 2 31 257 2510 3 32 229 1900 4 31 258 2220 [68]Open in a new tab SDS-PAGE Proteins in the serum of neonatal cord blood obtained from four women with gestational diabetes who delivered preterm neonates and from four women without diabetes who delivered full-term neonates were separated by SDS-PAGE. The molecular weights of the total proteins ranged from 25 to 180 kDa in all eight samples. The proteins were effectively separated in similar patterns without degradation, indicating that the samples were of sufficient quality to be used in the following experiments. Compared with the bands of the original serum (Lane E in [69]Supplementary Figure 2), the high-abundance proteins were significantly decreased, whereas the bands of other molecular weight regions were significantly increased. The total protein content met the requirements for use in the present study ([70]Supplementary Figure 2). LC-MS/MS Spectrum Analysis to Identify DEPs LC-MS/MS was used to detect the proteins. A total of 450 proteins were identified among all serum samples ([71]Supplementary Tables 1 and [72]2). Among them, 72 serum proteins from neonate cord blood were differentially expressed between preterm and full-term neonates, with 21 upregulated proteins and 51 downregulated proteins ([73]Figure 1A; [74]Table 2). We used cluster analyses to find that the protein expression pattern of the DEPs differed significantly between the two groups ([75]Figure 1B). Figure 1. [76]Figure 1 [77]Open in a new tab Profile of the differentially expressed proteins. (A) Volcano plot. Red dots represent proteins with a significant fold change (FC) > 1.5; green dots, proteins with a significant FC < 0.667; black dots, proteins with no significant change. (B) Heatmap showing the upregulation and downregulation of different proteins observed by cluster analysis. Each band in the image represents a protein, each column is a sample (1–4, full-term infant; 5–8, preterm infant), and colors represent various expression levels. Table 2. Differentially Expressed Proteins in Serum from Umbilical Cord Blood of Full-Term Neonates Delivered by Healthy Women vs Preterm Neonates Delivered by Women with Gestational Diabetes Change Protein Accession Number Protein Symbol Protein Description FC Increased A0A0G2JPR0 C4A C4a anaphylatoxin 51.43 Increased [78]P0DJI8 SAA1 Serum amyloid A-1 protein 33.22 Increased [79]Q15848 ADIPOQ Adiponectin 29.42 Increased [80]P13727 PRG2 Bone marrow proteoglycan 24.09 Increased [81]P55058 PLTP Phospholipid transfer protein 12.87 Increased [82]P07988 SFTPB Pulmonary surfactant-associated protein B 7.18 Increased [83]Q9BTY2 FUCA2 Plasma alpha-L-fucosidase 6.41 Increased F8WDD7 ARPC4 Actin-related protein 2/3 complex subunit 4 5.28 Increased C9JVG0 TF Serotransferrin (Fragment) 4.99 Increased A0A3B3IS80 ALDOB Fructose-bisphosphate aldolase 3.94 Increased B4DLR2 FAP Prolyl endopeptidase FAP 3.56 Increased [84]P08185 SERPINA6 Corticosteroid-binding globulin 3.27 Increased [85]Q9UBQ6 EXTL2 Exostosin-like 2 2.97 Increased [86]P02771 AFP Alpha-fetoprotein 2.84 Increased [87]P06681 C2 Complement C2 2.81 Increased [88]P13645 KRT10 Keratin, type I cytoskeletal 10 2.76 Increased [89]P02760 AMBP Protein AMBP 2.65 Increased [90]P29622 SERPINA4 Kallistatin 2.55 Increased [91]P05546 SERPIND1 Heparin cofactor 2 1.79 Increased [92]P02679 FGG Fibrinogen gamma chain 1.53 Increased [93]P01011 SERPINA3 Alpha-1-antichymotrypsin 1.53 Decreased [94]P14151 SELL L-selectin 0.64 Decreased [95]P19823 ITIH2 Inter-alpha-trypsin inhibitor heavy chain H2 0.64 Decreased [96]Q96PD5 PGLYRP2 N-acetylmuramoyl-L-alanine amidase 0.63 Decreased [97]Q12805 EFEMP1 EGF-containing fibulin-like extracellular matrix protein 1 0.63 Decreased [98]P27169 PON1 Serum paraoxonase/arylesterase 1 0.62 Decreased [99]Q9NZP8 C1RL Complement C1r subcomponent-like protein 0.62 Decreased [100]P43652 AFM Afamin 0.60 Decreased [101]P19827 ITIH1 Inter-alpha-trypsin inhibitor heavy chain H1 0.60 Decreased D6RAR4 HGFAC Hepatocyte growth factor activator 0.60 Decreased C9JB55 TF Serotransferrin (Fragment) 0.57 Decreased [102]Q7Z7M0 MEGF8 Multiple epidermal growth factor-like domains protein 8 0.57 Decreased H0YD13 CD44 CD44 antigen 0.56 Decreased [103]P55290 CDH13 Cadherin-13 0.56 Decreased [104]P23470 PTPRG Receptor-type tyrosine-protein phosphatase gamma 0.55 Decreased G3XAP6 COMP Cartilage oligomeric matrix protein 0.55 Decreased [105]P02747 C1QC Complement C1q subcomponent subunit C 0.54 Decreased K7ER74 APOC4-APOC2 Apolipoprotein C-II 0.54 Decreased X6RLJ0 C1QA Complement C1q subcomponent subunit A (Fragment) 0.53 Decreased [106]P41222 PTGDS Prostaglandin-H2 D-isomerase 0.53 Decreased A0A0C4DGZ8 GP1BA Glycoprotein Ib (Platelet), alpha polypeptide 0.53 Decreased [107]Q6UWP8 SBSN Suprabasin 0.51 Decreased [108]P04217 A1BG Alpha-1B-glycoprotein 0.50 Decreased [109]P00748 F12 Coagulation factor XII 0.50 Decreased E7END6 PROC Vitamin K-dependent protein C 0.47 Decreased [110]P02647 APOA1 Apolipoprotein A-I 0.46 Decreased [111]P01042 KNG1 Kininogen-1 0.45 Decreased V9GYM3 APOA2 Apolipoprotein A-II 0.44 Decreased [112]P0DOY3 IGLC3 Immunoglobulin lambda constant 3 0.33 Decreased [113]P68871 HBB Hemoglobin subunit beta 0.31 Decreased [114]P08637 FCGR3A Low affinity immunoglobulin gamma Fc region receptor III-A 0.30 Decreased A0A6Q8PHJ8 CDH2 Cadherin-2 (Fragment) 0.29 Decreased [115]Q06830 PRDX1 Peroxiredoxin-1 0.29 Decreased [116]P23284 PPIB Peptidyl-prolyl cis-trans isomerase B 0.28 Decreased [117]P62942 FKBP1A Peptidyl-prolyl cis-trans isomerase FKBP1A 0.27 Decreased [118]Q99497 PARK7 Parkinson disease protein 7 0.26 Decreased H0Y984 BST1 ADP-ribosyl cyclase/cyclic ADP-ribose hydrolase (Fragment) 0.25 Decreased [119]P18065 IGFBP2 Insulin-like growth factor-binding protein 2 0.24 Decreased [120]P02790 HPX Hemopexin 0.23 Decreased A0A1B0GVK0 FGL2 Fibroleukin 0.22 Decreased A0A2R8YGX3 TPM4 Tropomyosin alpha-4 chain 0.21 Decreased [121]O95467 GNAS Neuroendocrine secretory protein 55 0.19 Decreased A0A0G2JHD9 COL11A2 Collagen alpha-2(XI) chain (Fragment) 0.17 Decreased [122]P21333 FLNA Filamin-A 0.16 Decreased [123]P02775 PPBP Platelet basic protein 0.15 Decreased A0A087X089 CCL14 C-C motif chemokine 0.15 Decreased [124]P08833 IGFBP1 Insulin-like growth factor-binding protein 1 0.15 Decreased [125]P35579 MYH9 Myosin-9 0.14 Decreased [126]P05019 IGF1 Insulin-like growth factor I 0.13 Decreased A0A087WWU8 TPM3 Tropomyosin alpha-3 chain 0.12 Decreased [127]P02042 HBD Hemoglobin subunit delta 0.12 Decreased [128]O75144 ICOSLG ICOS ligand 0.09 [129]Open in a new tab Abbreviation: FC, fold change (ratio of the protein level in the two groups). GO Functional Annotation and Enrichment Analysis GO has three categories indicating how gene functions may be described: biological process, cellular component, and molecular function. GO analysis, a useful method and tool in bioinformatics analyses, has functional annotations indicating the number of DEPs in each category, and GO functional enrichment provides GO functional terms related to these DEPs. Thus, through GO analysis, the distribution of DEPs in these three categories and the involvement of the DEPs in cellular functions were identified. In addition, the most important biochemical metabolic pathways and signal transduction pathways were determined. Our GO functional annotation analysis showed that the highest percentage of DEPs in the biological process classification were related to the term “cellular process” for the upregulated proteins (n = 17 proteins) ([130]Figure 2A), with the top three proteins being SAA1, ADIPOQ and PRG2, and to the term “biological regulation” for the downregulated proteins (n = 37 proteins) ([131]Figure 2B), with the top three proteins being ICOSLG, HBD and IGF1. For the cellular component classification, the highest percentage of DEPs were related to the term “cellular anatomical entity” for both the upregulated proteins (n = 20 proteins) ([132]Figure 2A)—with the top three proteins being C4A, SAA1 and ADIPOQ—and for the downregulated proteins (n = 46 proteins) ([133]Figure 2B)—with the top three proteins being ICOSLG, HBD and TPM3. For the molecular function classification, the highest percentage of DEPs were related to the term “binding” for both the upregulated proteins (n = 15 proteins) ([134]Figure 2A)—with the top three proteins being SAA1, ADIPOQ and PRG2—and for the downregulated proteins (n = 39 proteins) ([135]Figure 2B)—with the top three proteins being ICOSLG, HBD and TPM3. Figure 2. [136]Figure 2 [137]Open in a new tab Functional classification of gene ontology analysis. (A) Upregulated proteins. (B) Downregulated proteins. Abbreviations: BP, biological process; CC, cellular component; MF, molecular function. In the GO functional enrichment analysis, the top five upregulated GO terms were serine-type endopeptidase inhibitor activity, endopeptidase inhibitor activity, endopeptidase regulator activity, peptidase inhibitor activity, and peptidase regulator activity ([138]Figure 3A). The top five downregulated GO terms were protein stabilization, positive regulation of activated T cell proliferation, regulation of activated T cell proliferation, regulation of protein stability, and Ras protein signal transduction ([139]Figure 3B). Figure 3. [140]Figure 3 [141]Open in a new tab Gene ontology (GO) enrichment for functional terms. (A) Upregulated proteins. (B) Downregulated proteins. The abscissa represents the enrichment factor, showing the percentage of differentially expressed proteins in the GO classification relative to the identified proteins in that classification. The ordinate indicates the GO term description. The size of the bubble represents the number of proteins in the GO classification. The enrichment test P value was obtained using the Fisher exact test; −log10 (P value): the logarithmic conversion of the Fisher exact test P value. KEGG Pathway Annotation and Enrichment Analysis We used KEGG signaling pathways analysis to identify the signaling pathways that were altered in cord blood from preterm neonates delivered by women with gestational diabetes. In the KEGG pathway annotation analysis, we found that the top five upregulated KEGG pathways were the complement and coagulation cascades, Staphylococcus aureus infection, pertussis, HIF-1 signaling pathway, and PPAR signaling pathway ([142]Figure 4). The top five downregulated KEGG pathways were complement and coagulation cascades, dilated cardiomyopathy, pathways in cancer, Chagas disease, and hypertrophic cardiomyopathy ([143]Figure 4). Figure 4. [144]Figure 4 [145]Open in a new tab Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway annotation. KEGG signaling pathways were identified in differentially expressed proteins (DEPs). Abscissa represents the number of DEPs; ordinate, KEGG pathways associated with the DEPs; green bars, upregulated pathways; and red bars, downregulated pathways. The results of our KEGG pathway enrichment analysis indicated seven KEGG signaling pathways were enriched for the DEPs: inflammatory mediator regulation of TRP channels, Chagas disease, adrenergic signaling in cardiomyocytes, long-term depression, PPAR signaling pathway, ovarian steroidogenesis, and longevity regulating pathway ([146]Figure 5). All but one pathway, PPAR signaling, were downregulated. Figure 5. [147]Figure 5 [148]Open in a new tab Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment. KEGG signaling pathways were enriched for the differentially expressed proteins (DEPs). Abscissa: the enrichment factor showing the total proportion of DEPs in the KEGG signaling pathway as a proportion of the identified proteins in the classification. Ordinate: the KEGG term description. Bubble size indicates the number of DEPs in the KEGG pathway. The enrichment test P value was obtained using the Fisher exact test; −log10 (P value): the logarithmic conversion of the Fisher exact test P value. On the basis of our KEGG pathway annotation and enrichment analyses, we concluded that the complement and coagulation cascades may be importantly associated with preterm delivery of neonates by women with gestational diabetes. The DEPs in the serum obtained from the cord blood of these neonates involved in the complement and coagulation cascades included C4A, C2, SERPIND1, FGG, C1QC, C1QA, F12, PROC, and KNG1 ([149]Figure 6). Figure 6. [150]Figure 6 [151]Open in a new tab Complement and coagulation cascades. The differentially expressed proteins (DEPs), highlighted in a red frame, participating in the complement and coagulation cascades. The upregulated DEPS are C4A (C4, C4a anaphylatoxin), C2 (complement C2), SERPIND1 (HCII, heparin cofactor 2) and FGG (fibrinogen, fibrinogen gamma chain). The downregulated DEPs are C1QC (complement C1q subcomponent subunit C), C1QA (complement C1q subcomponent subunit A), F12 (coagulation factor XII), PROC (PC, vitamin K-dependent protein C) and KNG1 (KNG, kininogen-1). Concentrations of Coagulation-Associated Proteins in DEPs To confirm the DEPs identified by the LC-MS/MS, we examined the concentrations of coagulation-associated proteins in DEPs in neonatal cord blood samples using ELISA experiment and expended samples. Beside the samples used in LC-MS/MS experiment, we further collected 9 neonatal cord blood samples respectively from women with gestational diabetes who delivered preterm neonates and from women without diabetes who delivered full-term neonates. Our data showed that the expressions of SERPIND1 and FGG were significantly upregulated in the cord blood samples of preterm neonates compared to those of full-term neonates. On the contrary, the expressions of F12, KNG1and PROC were significantly downregulated in the cord blood samples of preterm neonates compared to those of full-term neonates ([152]Figure 7). Therefore, these results are consistent with the data from our LC-MS/MS experiment. Figure 7. [153]Figure 7 [154]Open in a new tab Concentrations of coagulation-associated proteins in differentially expressed proteins (DEPs). (A–E) Summary data showing the concentrations of heparin cofactor 2 (SERPIND1, A), fibrinogen gamma chain (FGG, B), coagulation factor XII (F12, C), kininogen-1 (KNG1, D) and vitamin K-dependent protein C (PROC, E) in serum samples from cord blood of healthy women who delivered full-term neonates and women with gestational diabetes who delivered preterm neonates. Among of them, SERPIND1and FGG are upregulated DEPs, but F12, KNG1and PROC are downregulated DEPs. Values are shown as the mean ± SEM (n = 13); *P < 0.05 for full-term neonates vs preterm neonates. Discussion The incidence of diabetes in China continues to increase every year, including among pregnant women. Gestational diabetes is a serious threat to the health of pregnant women and their fetuses. Thus, understanding the pathogenesis and development of gestational diabetes is critical for finding new targets for prevention and treatment. In the present study, we used label-free LC-MS/MS analysis to identify key proteins that were differentially expressed between the serum obtained from neonatal cord blood of healthy women who delivered full-term neonates and of women with gestational diabetes who delivered preterm neonates. Overall, we identified 450 proteins, of which 21 DEPs were upregulated and 51 DEPs were downregulated. Our KEGG pathway annotation and enrichment analyses determined that many of these DEPs were involved in complement and coagulation cascades. Hemorrhagic or thrombotic complications are important factors inducing neonate morbidity and mortality.[155]^21–25 One study reported that the coagulation system is similar in preterm and full-term infants and does not show a rapid change between 30 and 40 weeks of life.[156]^26 Even in extremely preterm neonates (24–31 weeks), the initial clotting tests, including platelets, fibrinogen, K-dependent factors, and factors VIII and V, are within standard references ranges.[157]^27