Abstract Introduction Antiphospholipid syndrome (APS) is an autoimmune disorder associated with thrombosis and adverse obstetric outcomes. Early diagnosis and intervention can improve pregnancy outcomes to some extent, but current results are unsatisfactory. Exosomes, containing biomacromolecules relevant to reproduction, play essential roles in pregnancy. However, research progress on their involvement in APS remains limited. Objectives This study aims to investigate protein profile changes in plasma exosomes and identify potential biomarkers for obstetric APS. Methods We employed tandem mass tag (TMT) markers to analyze exosome protein profiles from 6 healthy early pregnant women and 6 early-stage APS patients. Quantitative proteomics analysis was conducted using the Maxquant search engine. Results Differential expression analysis identified 51 upregulated and 22 downregulated proteins in plasma exosomes from early pregnant women with APS, such as serpin peptidase inhibitor C1/A1/A7, apolipoprotein 1/2, orosomucoid 1/2 and apolipoprotein H. Kyoto Encyclopedia of Genes and Genomes analysis shows that differentially expressed proteins are enriched in the PPAR signaling pathway and staphylococcus aureus infection pathway. Enrichment analysis indicated associations with glycerolipid biosynthesis, vitamin transport, and negative regulation of very-low-density lipoprotein particle remodeling. Conclusion Our study highlights alterations in the protein profiles of plasma exosomes in APS pregnant patients and proposes potential biomarkers, offering insights for early diagnosis and treatment and improving reproductive outcomes. Keywords: Antiphospholipid syndrome, Exosomes, Proteomics, Pregnancy Highlights * • Proteomics analysis revealed distinct exosome protein profiles in normal pregnant women versus early-stage APS patients. * • SERPINC1/A1/A7, APOA1/2, ORM1/2, and APOH are differential proteins in exosome protein profiles. * • Our study may contribute valuable insights into identifying biomarkers for early diagnosis of APS during pregnancy. 1. Introduction Antiphospholipid syndrome (APS) is a systemic autoimmune disease characterized by arterial or venous thrombosis, decreased platelets, and pathological pregnancies. Current treatment effectiveness remains around 70 % [[31][1], [32][2], [33][3], [34][4]] [[35][1], [36][2], [37][3], [38][4]] [[39][1], [40][2], [41][3], [42][4]]. Early diagnosis of APS poses challenges, adversely affecting obstetric outcomes for patients [[43]2]. Hence, this study aims to identify biomarkers for early diagnosis of APS during pregnancy through proteomics analysis of plasma exosomes. This has implications for exploring precision clinical treatments and also contributes to exploring new directions in APS therapy. This study specifically focuses on obstetric APS (OAPS), characterized by recurrent fetal loss, early fetal death, or preterm delivery before 34 weeks due to preeclampsia, eclampsia, or placental insufficiency [[44]4,[45]5]. Early identification of such cases is crucial for successful pregnancy outcomes [[46]2]. Diagnosis of APS currently relies on the presence of vascular thrombosis, pregnancy complications, and detection of antiphospholipid antibodies (aPLs) [[47]6,[48]7]. However, the lack of standardization in aPLs testing methods across different medical institutions and the heterogeneity of aPLs themselves contribute to poor reproducibility and difficulty in laboratory test standardization [[49]8]. Standard treatments involving prednisolone, low-dose aspirin (LDA), and heparin significantly improve outcomes for APS patients [[50][8], [51][9], [52][10]] [[53][8], [54][9], [55][10]] [[56][8], [57][9], [58][10]]. However, the success rate of pregnancies in treated APS patients ranges from 70 % to 85 %, with a high prevalence of preterm birth, preeclampsia, eclampsia, and peri-thrombotic events [[59]2]. The efficacy of these measures in significantly improving the prognosis of APS patients is often limited. Therefore, there is an urgent need to explore additional biomarkers, especially those reflecting pregnancy morbidity and prothrombotic states. Previous studies suggest that circulating exosomes derived from peripheral blood in APS patients are detrimental to vascular development and can lead to pregnancy complications [[60]11]. Additionally, extracellular vesicles (EVs) have been implicated in coagulation and inflammation [[61]12]. Recent research indicates elevated levels of circulating EVs in APS patients, possibly related to systemic vascular activation [[62]12]. Although direct evidence linking increased EVs levels to thrombosis and obstetric events in APS is lacking, EVs may participate in the pathogenesis of APS by promoting vascular activation and thrombosis [[63]13]. Exosomes, a subtype of EVs ranging from 30 to 200 nm in diameter, have been implicated in various reproductive processes, suggesting significant potential in reproductive disease diagnosis and treatment [[64]14,[65]15]. In addition, exosomes proteins serve as important biomarkers for reproductive system diseases such as endometrial cancer and ovarian cancer [[66]16]. Utilizing exosomes proteins as biomarkers could greatly enhance our understanding of reproductive dysfunction and related diseases. In this study, we conducted proteomics analysis of exosomes derived from APS and non-APS pregnant women. The results of this study can provide reference for early diagnostic biomarkers research of OAPS. 2. Materials and methods 2.1. Sample collection In this study, we collected peripheral blood plasma from normal pregnant women and APS patients in early pregnancy (start of the last menstrual period through the end of 12 weeks of gestation) to extract exosomes for the identification of their protein characteristics. All subjects had their peripheral blood collected in a fasting state in the morning. Subsequently, the blood samples were quickly transferred into heparin-containing anticoagulation tubes and mixed by vortexing or inverting 5 to 10 times. 2.1.1. Participants From July 2018 to December 2018, our study enrolled a total of 12 pregnant women in the early stages of pregnancy: 6 healthy individuals (NC group) and 6 with APS (APS group), all receiving care at the First Maternal and Infant Health Hospital Affiliated with Tongji University ([67]Table 1). Table 1. Clinical characteristics of the APS group and NC group. (If the content of antibody was below 2 in the laboratory tests, then the result was negative. For analyzing more conveniently, these negative results were regarded as 2.)In the APS group, all patients received the standard treatment regimen, which includes: (1) Prednisone (produced by Shanghai Xinyi Pharmaceutical Co., Ltd.) taken orally, starting from the pre-pregnancy period, once daily. The initial dose is 5 mg, with a general medication dose ranging from 5 to 15 mg, continuing until the end of pregnancy; (2) Aspirin (produced by Shanghai Xinyi Pharmaceutical Co., Ltd.) taken orally, starting from the confirmation of pregnancy, once daily. The initial dose is 25 mg, with a general medication dose ranging from 25 to 75 mg, continuing until three days before delivery; (3) Low molecular weight heparin (produced by Hangzhou Jiuyuan Gene Engineering Co., Ltd.) administered via intramuscular injection, starting from the confirmation of pregnancy, three times daily. The initial dose is 5000 U/d, with a general medication dose of 5000U/d, continuing until the 35th week of pregnancy. Regarding outcomes, all patients in the control group had full-term vaginal deliveries, while in the APS group, there was one full-term delivery, one delivery at 32 weeks, one case of premature rupture of membranes, one early miscarriage, and two cases where outcome data were missing. NC group (Mean ± SEM) APS group (Mean ± SEM) P Count 6 6 Age (year) 28.33 ± 1.764 27.83 ± 0.703 >0.05 Gestational age (day) 42.33 ± 2.951 41.00 ± 2.309 >0.05 Times of abortion 0 2.167 ± 0.167 <0.05 Antiphospholipid antibody Anti-apoh IgM antibody (AU/ml) 2 52.92 ± 4.565 <0.05 Anti-apoh IgG antibody (AU/ml) 2 2 Anticardiolipin antibody IgM antibody (MPL/ml) 2 2 Anticardiolipin antibody IgG antibody (GPL/ml) 2 2 Anti-lupus anticoagulant(S) negative positive [68]Open in a new tab 2.1.2. Inclusion and exclusion criteria for the APS group For the diagnosis of patients with OAPS, we followed the criteria set forth in the "American Society for Reproductive Medicine Technical Committee and the Society for Reproductive Medicine Consensus," where clinical criteria are consistent with a history of pathological pregnancy and are confirmed through laboratory tests as positive for β2-glycoprotein 1 (β2-GP1) antibodies. Pathological pregnancy includes: (1) one or more unexplained cases where a morphologically normal fetus is confirmed by ultrasound or direct fetal examination to be dead in utero after more than 10 weeks of pregnancy; (2) one or more cases where a morphologically normal fetus is delivered prematurely before 34 weeks of pregnancy due to severe preeclampsia or severe placental dysfunction; (3) two or more consecutive unexplained miscarriages within 10 weeks of pregnancy. Meeting any one of these criteria suffices. Laboratory criteria: The presence of medium or high titers of IgG or IgM anti-β2-GP1 antibodies in the blood on two or more occasions at least 12 weeks apart [[69]17,[70]18]. Pregnant women diagnosed with OAPS were included in the APS group. The participants ranged from 20 to 40 years old, were between 6 and 8 weeks gestational age. Those with comorbidities such as uterine fibroids, ovarian cysts, endocrine disorders, and infections were excluded from the study. Patients with other autoimmune diseases, thrombotic disorders, and other serious comorbidities are also excluded. 2.1.3. Inclusion and exclusion criteria for the normal control (NC) group Those within the age range of 20–40 years, who have confirmed intrauterine pregnancy through B-ultrasound and are in their gestational period between 6 and 8 weeks without any history of spontaneous abortion, preeclampsia, eclampsia or preterm birth were included in the NC group. Those with uterine fibroids, ovarian cysts, endocrine abnormalities, infections, and other acute or chronic illnesses, as well as those who tested positive for antiphospholipid antibodies were excluded from the study. 2.1.4. Extraction of exosomes Immediately proceed with the extraction of exosomes following sample collection. Initially, the plasma samples were briefly centrifuged at 3000 g for 5 min at 4 °C to remove any large cell fragments or debris. The supernatant was then subjected to further centrifugation at 10000g for 60 min at 4 °C (Avanti J-26S XP) in order to extract the small vesicles. To remove small cellular debris, the supernatant was collected again and mixed with phosphate-buffered saline (PBS, Biological Industries, Israel). The mixture was then centrifuged at 10000 g for 120 min at 4 °C using an OptimaXE-90 ultracentrifuge. Finally, the supernatant was discarded and the remaining pellet was resuspended in PBS before being stored at −80 °C. 2.2. Protein extraction and quantification Samples were removed from −80 °C, and each group of samples was respectively added to four volumes of lysis buffer (8 M urea, 100 mM Tris-HCl, pH 8.0, 1 % protease inhibitor) and lysed using an ultrasonic cell disruptor JY92-IIN (Ningbo Xinzhi) at 20 % power, with 3s sonication in an ice-water bath followed by a 2s pause, for a total sonication time of 3 min. Centrifugation was performed at 4 °C, 20,000 g for 10 min, and the supernatant was collected and mixed with trichloroacetic acid to a final concentration of 20 %, then left to stand at 4 °C for 2 h. Afterward, centrifugation at 4 °C, 12,000 g for 3 min was done, the supernatant was discarded, and the precipitate was washed three times with pre-cooled acetone. Finally, the precipitate was redissolved in 8 M urea. Protein concentration was determined using a bicinchoninic acid (BCA) assay kit (Beyotime P0010) as follows: Take 5 μL of the protein sample and determine the protein concentration using a BCA assay kit. The method is as follows: (1) Add standard samples at volumes of 0 μL, 5 μL, 10 μL, 15 μL, and 20 μL to the wells of an enzyme label plate, fill up to 20 μL with sample diluent, and run three replicates for each; (2) Add 5 μL of the protein sample to be tested to the wells of the enzyme label plate, fill up to 20 μL with sample diluent, and run three replicates for each; (3) Add 200 μL of BCA working solution to each well and incubate at 37 °C for 30 min; (4) Measure A570 with an enzyme label meter (the optimal absorption wavelength is 562 nm, but other wavelengths between 540 and 595 nm can also be applied); (5) Calculate the protein concentration of the samples based on the standard curve and the volume of sample used. 2.3. Trypsin digestion The protein solution was reduced with 5 mM dithiothreitol at 56 °C for 30 min, followed by alkylation with 11 mM iodoacetamide in the dark at room temperature for 15 min prior to digestion. Once the urea concentration dropped below 2 M, the protein sample was diluted with 100 mM tetraethyl ammonium bromide (TEAB). Subsequently, trypsin was added at a ratio of 1:50 for the initial overnight digestion and a ratio of 1:100 for a second 4-h digestion. 2.4. Proteomics analysis In this part, samples were processed for total-protein tandem mass tag (TMT) labeling by Jingjie PTM BioLab (Hangzhou, China). 2.4.1. TMT labeling Peptides digested by trypsin were desalted using Strata X C18 (Phenomenex) and then lyophilized under vacuum. The peptides were dissolved in 0.5 M TEAB and labeled according to the instructions of the TMT reagent kit. The procedure is as follows: After thawing, the labeling reagent was dissolved in acetonitrile, mixed with the peptides, and then incubated at room temperature for 2 h. After labeling, the peptides were desalted and lyophilized under vacuum. 2.4.2. High performance liquid chromatography (HPLC) Fractionation Peptides were fractionated using high pH reverse HPLC, with the chromatography column being Agilent 300Extend C18 (5 μm particle size, 4.6 mm inner diameter, 250 mm length). The operation is as follows: The gradient for peptide fractionation was from 8 % to 32 % acetonitrile, pH 9, over 60 min to separate 60 components, which were then combined into 9 fractions. These combined fractions were lyophilized under vacuum for further processing. 2.4.3. Liquid chromatography-mass spectrometry (LC-MS) analysis The peptides were dissolved in liquid chromatography mobile phase A (0.1 % (v/v) formic acid in water) and separated using the EASY-nLC 1000 ultra high performance liquid system. Mobile phase A consisted of 0.1 % formic acid and 2 % acetonitrile in water; Mobile phase B consisted of 0.1 % formic acid and 90 % acetonitrile in water. The liquid chromatography gradient was set as follows: from 0 to 26 min, 9%–25 % B; from 26 to 34 min, 25%–36 % B; from 34 to 37 min, 36%–80 % B; from 37 to 40 min, 80 % B, with the flow rate maintained at 700 nL/min. After separation by the ultra high performance liquid system, the peptides were ionized in the NSI ion source and then analyzed by the Orbitrap FusionTM mass spectrometer. The ion source voltage was set at 2.0 kV, and both the precursor ions and their secondary fragments were detected and analyzed using the high-resolution Orbitrap. The scan range for the primary mass spectrometry was set to 350–1550 m/z, with a resolution of 60,000; the secondary mass spectrometry scan range started at 100 m/z, with an Orbitrap resolution of 15,000. The data acquisition mode used was a data-dependent scan (DDA) program, where after a primary scan, the top 20 precursor ions with the highest signal strength were selected to enter the HCD collision cell for fragmentation with 35 % fragmentation energy, followed by sequential secondary mass spectrometry analysis. To improve the effective utilization rate of mass spectrometry, the automatic gain control (AGC) was set to 5E4, the signal threshold was set to 5000 ions/s, the maximum injection time was set to 200 ms, and the dynamic exclusion time for tandem mass spectrometry scans was set to 30 s to avoid repeated scanning of parent ions. 2.4.4. Database search The secondary mass spectrometry data was searched using Maxquant (v1.5.2.8). The search parameters were set as follows: the database was SwissProt Human (20130 sequences), with a decoy database added to calculate the false positive rate (FDR) caused by random matches, and a common contaminant database was included to eliminate the impact of contaminant proteins in the identification results; the enzymatic digestion was set to Trypsin/P; the number of missed cleavage sites was set to 2; the minimum length of peptides was set to 7 amino acid residues; the maximum number of modifications per peptide was set to 5; the mass tolerance for the first and main search of precursor ions was set to 20 ppm and 5 ppm, respectively, and the mass tolerance for secondary fragment ions was set to 0.02 Da. Cysteine alkylation was set as a fixed modification, with variable modifications being methionine oxidation and protein N-terminal acetylation. The quantification method was set to TMT-6plex, with both protein identification and PSM identification FDRs set at 1 % [[71]19]. 2.4.5. Mass spectrometry quality control First, we examined the mass deviation (mass error) of all identified peptides. The mass errors were centered around zero and concentrated within a range of less than 10 ppm, indicating that the mass accuracy meets the requirements. Secondly, the majority of peptide lengths were distributed between 8 and 20 amino acid residues, consistent with the pattern of peptides digested by trypsin, indicating that the sample preparation met the standard. 2.5. Bioinformatics analysis methods and statistical analysis 2.5.1. Protein annotation methods Gene Ontology (GO) annotations at the proteomics level come from the UniProt-GOA database ([72]www.ebi.ac.uk/GOA/). Initially, protein IDs are converted to UniProt IDs, which are then used to match GO IDs, and based on these GO IDs, relevant information is retrieved from the UniProt-GOA database. If the UniProt-GOA database does not contain information on the queried protein, we use a protein sequence-based algorithm software, InterProScan, to predict the GO functions of the protein. The proteins are then classified according to cellular components, molecular functions, or biological processes. For project data, the software InterProScan and the corresponding InterPro domain database ([73]http://www.ebi.ac.uk/interpro/) are used to annotate the protein domains of the identified proteins. The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway database is used for annotating protein pathways: firstly, the submitted proteins are annotated using the KEGG online service tool KEGG Automatic Annotation Server (KAAS), then the annotated proteins are matched to corresponding pathways in the database through KEGG mapper [[74]20]. The software wolfpsort, which predicts subcellular localization, is used to annotate the subcellular localization of the submitted proteins. 2.5.2. Protein function enrichment GO annotations of proteins are divided into three main categories: biological processes, cellular components, and molecular functions. Fisher's exact test is used to test the differential expression proteins against the background of identified proteins. A GO enrichment test P-value of less than 0.05 is considered significant. The KEGG database is used for pathway enrichment analysis. Fisher's exact test is used to test the differential expression proteins against the background of identified proteins. A pathway enrichment test P-value of less than 0.05 is considered significant. Finally, these pathways are categorized according to the KEGG website's pathway hierarchy classification method. The InterPro database (which provides resources for the analysis of protein families, domains, and functional sites) is used to analyze the enrichment of functional domains in differentially expressed proteins. Fisher's exact test is used to test the differential expression proteins against the background of identified proteins. An enrichment test P-value of less than 0.05 for domain units is considered significant. 2.5.3. Clustering analysis based on protein function enrichment First, we collect the functional classification information and corresponding enrichment P-value of all protein groups, then select the functional classifications that are significantly enriched (P-value <0.05) in at least one protein group. The selected P-value data matrix is first transformed using a logarithm of -log10, and then the transformed data matrix is subjected to a Z-transform for each functional classification. Finally, the dataset obtained after the Z-transform is subjected to hierarchical clustering (Euclidean distance, average linkage clustering) for unilateral cluster analysis. The clustering relationships are visualized using the heatmap.2 function from the R package gplots. 2.5.4. Statistical analysis and protein-protein interaction (PPI) network Furthermore, we constructed a PPI network of differentially expressed proteins (DEPs) by utilizing the search tool for the retrieval of interacting genes/proteins (STRING) online database (v11.5) [[75]21]. GraphPad Prism 6.0 was utilized for statistical analysis. The clinical information of participants is presented as the mean ± standard error of the mean (SEM). T-test was performed to estimate significant differences between groups, and differences were considered statistically significant when P-value was <0.05. 2.6. Ethics statement The studies involving participants were conducted in accordance with the ethical policies and procedures approved by the Ethics Committee of Shanghai First Maternal and Infant Hospital affiliated with Tongji University (KS22212). Written informed consent was obtained from all participants. 3. Result 3.1. Clinical characteristics of the APS group and NC group The diagnostic criteria of APS were based on the 2006 international consensus statement [[76]2]. Clinical characteristics of the APS group and NC group are presented in [77]Table 1. There was no significant difference (P > 0.05) in age (25–35 years old) and gestational age (6–8 weeks) between participants in the NC and APS groups. 3.2. Proteomics profiling and identification of DEPs in plasm exosomes [78]Fig. 1, A depicts the experimental design and proteomics analytical processes employed in this study. In the course of quantitative proteomics analysis, 293 proteins were identified, out of which 261 had quantifiable information ([79]Supplement Table 1). In this study, a change threshold of 1.2-fold was applied and only proteins with a t-test P-value<0.05 were considered as DEPs. As such, we identified 51 significantly up-regulated and 22 down-regulated DEPs in the APS group compared to the NC group ([80]Fig. 1B–[81]Supplement Table 2). To ensure data quality, we conducted sample repeatability tests and calculated Pearson's correlation coefficients for all samples, which were then used to generate a heatmap ([82]Fig. 1C–[83]Supplement Table 3). Based on the primary analysis, we performed a systematic bioinformatics investigation of quantifiable proteins, encompassing GO annotation, functional classification, functional enrichment and cluster analysis based on functional enrichment. Fig. 1. [84]Fig. 1 [85]Open in a new tab The basic procedure of the study and the identification of DEPs.(A) Overview of the experimental design and proteomics analytical processes. Plasm exosomes were isolated from healthy pregnant women or APS patients, followed by a total-protein TMT labeling and quantitative proteomics analysis to determine the abundance of exosome proteins. The data were subjected to analysis using a series of bioinformatics tools. (B) Heatmap showing 66 effective DEPs in the APS group and NC group, for that ones with censored data were excluded among 73 DEPs. Each protein is represented by a row, and each sample by a column. Varied hues on the heatmap represented varied protein expression levels, with red denoting proteins with high expression and blue denoting proteins with low expression. (C) Heatmap according to Pearson's correlation coefficients among 12 samples.. (Pearson's correlation coefficients, which is a measure of the degree of linear correlation between two sets of data, when it is close to −1, the two sets of data present a negative correlation. On the contrary, when the coefficient tends to be 1, the two sets of data present a positive correlation, and the closer it is to 0, the it is more possible that the two sets of data is no correlation.). (For interpretation of the references to color in this