Differential proteomic analysis of plasma-derived exosomes as diagnostic biomarkers for chronic HBV-related liver disease

Hepatitis B virus (HBV) infection is still a major public health problem worldwide. We aimed to identify new, non-invasive biomarkers for the early diagnosis of chronic HBV-related diseases, reveal alterations in the progression of chronic hepatitis B (CHB), liver cirrhosis (LC), and hepatocellular carcinoma (HCC). Here, exosomes were isolated and characterized through size exclusion chromatography and nanoparticle tracking analysis. Profiles of differentially expressed proteins (DEPs) were analyzed through liquid chromatography-tandem mass spectrometry (LC–MS/MS), Gene Ontology, and Kyoto Encyclopedia of Genes and Genomes analyses. Results showed that the DEPs, including CO9, LBP, SVEP1, and VWF levels in extracellular vesicles (EVs) were significantly higher in CHB than in healthy controls (HCs). VWF expression levels in EVs were significantly lower in CHB than in those with LC. KV311 expression levels in EVs were significantly higher, whereas LBP levels were significantly lower in patients with CHB than in those with HCC. All biomarkers seemed to exhibit a high diagnostic capacity for HBV-related liver disease. Patients with HBV-induced chronic liver disease exhibit characteristic protein profiles in their EVs. Thus, serum exosomes may be used as novel, liquid biopsy biomarkers to provide useful clinical information for the diagnosis of HBV-related liver diseases at different stages.

. Characterization of plasma-derived exosomes from patients with chronic hepatitis B virus infection (CHB), liver cirrhosis (LC), and hepatocellular carcinoma (HCC) and healthy controls (HCs). The size and morphological distribution of exosomes are shown, and were obtained using nanoparticle tracking analysis (NTA) (A) and transmission electron microscopy (TEM) (B). The samples were positive for the exosome markers CD9, TSG101, and CD63 and negative for calnexin (C).
Best candidate liquid biopsy biomarkers for diagnosing CHB, LC, and HCC. Of the differentially expressed proteins, CO9, LBP, SVEP1, and VWF were the best candidate liquid biopsy biomarkers for diagnosing CHB, as their levels were increased in serum EVs (sEVs) and presented higher area under the curve (AUC) values of 0.925 (1.000/0.929), 0.950 (0.964/0.964), 0.900 (1.000/1.000), and 0.875 (1.000/0.964), respectively, in patients with LC and HCC compared with those of HCs (Fig. 5a-c). Of these, VWF had a better diagnostic efficiency for CHB and LC, with AUC values of 0.914 (Fig. 5d), whereas CO9 had a better diagnostic efficiency for CHB and HCC, with AUC values of 0.857 (Fig. 5e). KV311 was found to be a good biomarker, as its level also increased in sEVs and presented higher AUC values of 0.857 for the diagnosis of LC in comparison with those of HCC (Fig. 5f). Table 1 shows a panel of biomarkers that distinguish HCC from LC and CHB, as well as from healthy controls. The lower left corner shows groups with differential expression and the number of different proteins that were screened among each group. The dots on the right show proteins common or specific to the corresponding group. The histogram shows the number of specific or common proteins. www.nature.com/scientificreports/ Target analysis of differentially expressed proteins in sEV-enriched fractions of CHB, LC, HCC, and HCs groups. We performed Gene Ontology (GO) enrichment of differentially expressed proteins identified in sEVs. Our findings revealed that the selected differentially expressed proteins were enriched in the biological process pathways of cell death, regulation of complement activation, and leukocyte chemotaxis in the inflammatory response; cellular components of the extracellular region and blood microparticle, and the membrane attack complex; and molecular function terms of lipoteichoic acid binding, immunoglobulin binding, and lipopeptide binding; and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways of complement and coagulation cascades (Fig. 6a-d) 16,17 .

Discussion
Chronic HBV-related liver diseases, such as CHB, LC, and HCC, are highly prevalent in China; however, early and effective diagnostic markers that can be obtained non-invasively are lacking. Recently, exosomes have attracted increasing attention in chronic disease research 18,19 and few studies have investigated the involvement of exosomes in the pathogenesis of the abovementioned diseases. In this study, we conducted a proteomic analysis of the isolated and purified serum exosomes from patients of CHB, LC and HCC, while using healthy individuals as controls. Our preliminary results provided a panel of differentially expressed proteins as promising biomarker candidates for the early and differential diagnosis of HBV-related liver diseases.
Considering the stated clinical difficulty in the differential diagnosis of CHB, LC, and HCC, we analyzed the differentially expressed protein profiles of plasma-derived exosomes of patients with the abovementioned diseases and the HCs group. The clustergram was drawn by the R package Pheatmap. The ordinate represents proteins, and the abscissa represents samples. Clustering was performed using the values of log 10 (label-free quantification + 1, LFQ + 1) and then clustering. On the left, lines in black show the hierarchical clustering of proteins. Colored bars at the top represent CHB, LC, HCC and HCs; the corresponding samples are shown at the bottom. Red and green represent proteins with high and low expression, respectively. (D) A clustergram of differentially expressed proteins for pairwise comparison between disease groups (CHB, LC, and HCC). The clustergram was drawn by the R package Pheatmap. The ordinate represents different proteins, and the abscissa represents different samples. Clustering was performed using the values of log 10 (LFQ + 1). Red and green represent highly and lowly expressed proteins, respectively. www.nature.com/scientificreports/ and HCs. The highest number of differentially expressed proteins was observed between LC and HCs, whereas the lowest was observed between CHB and HCC. Exosomes have a rich and complex protein composition, which may be involved in regulating a variety of biological processes. Therefore, exosomes may be considered as a new, specific, sensitive, and non-invasive diagnostic marker for HBV-related liver disease based on the differentially expressed proteins detected. Label-free proteomics has substantially advanced the capacity for discovering new candidate biomarkers. We analyzed the proteomic profiles of patients with CHB, LC, and HCC and HCs using sEVs and identified five biomarkers, namely CO9, LBP, SVEP1, VWF, and KV311. CO9 plays a crucial role in the innate and adaptive immune response, as a component of the membrane attack complex which forms pores in the plasma membrane of target cells 20 . LBP, a glycolipid found in the outer membrane of all gram-negative bacteria, binds to a portion of bacterial lipopolysaccharide. LBP plays a significant role in the innate immune response 21 . SVEP1 plays a crucial role in regulating intercellular adhesion 22 . VWF, a multimeric glycoprotein synthesized primarily by endothelial cells 23 , is involved in platelet adhesion and aggregation and other hemostatic mechanisms 24 . KV311 is involved in antigen recognition via the V region of immunoglobulin light chain variable domains 25 .
Our results indicated that the expression of CO9, LBP, SVEP1, and VWF in the sEVs of CHB, LC, and HCC was significantly higher than that in HCs. Furthermore, these markers were associated with excellent diagnostic accuracy for CHB, LC, and HCC. The AUC values of CO9, LBP, SVEP1, and VWF for HCC were 0.929, 0.964, 1.000, and 0.964, respectively, whereas the AUC values of AFP for HCC were reported to be 0.830 in a systematic review and meta-analysis 26 . Obviously, the AUC values of these novel biomarker candidates were significantly higher than AFP's for HCC. Moreover, LBP expression in the sEVs from the HCC group was higher than that in the CHB group. These findings are consistent with those of the study by Cai et al., who reported that serum AFP combined with LBP, which is highly expressed in HCC, may predict poor prognosis after curative resection for HCC 27 . Chen et al. showed that in the virus-associated tumor microenvironment, LBP may significantly realize anti-tumor effects through the NF-κB-associated signaling pathway 28 . Li et al. showed that LBP promotes the anti-proliferative and apoptotic effects of 125I, which provides a solid foundation for better combination therapy 29 . Furthermore, LBP could maintain high levels of T cells in tumor tissue, tumor-draining lymph nodes, and peripheral blood. LBP can inhibit the increase of Treg populations and the production of interleukin-10 (IL-10) and transforming growth factor-beta 1 (TGF-β1) in the serum, mitigate the exhaustion of T cells, promote the infiltration of CD8 + T cells in tumor tissue, and maintain the cytotoxicity of lymphocytes 30 . LBP may, therefore, play an important role in the diagnosis and treatment of HCC.
KV311 and CO9 may be used as complements of LBP for the diagnosis and evaluation of HCC. We found that KV311 expression in the sEVs of the HCC group was significantly lower than that in the CHB and LC groups. Thus, KV311 seems to have a better diagnostic efficiency for HCC than LC, with AUC values of 0.857, whereas CO9 seems to have a better diagnostic efficiency for HCC than CHB, with AUC values of 0.857. KV311, which   38 . Our research shows that SVEP1 may be not a promising biomarker for the diagnosis of HBV-related liver diseases at different stages, because its expression is just higher than the HCs group. SVEP1 expression may be one of universal markers for HCC and liver fibrosis, but it is not classify HBV-related liver diseases at different stages.
Our findings revealed that VWF may be a predictor of cirrhosis. VWF expression in the sEVs of patients with LC was higher than that in patients with CHB. VWF seems to have a better diagnostic efficiency for LC than CHB, with an AUC value of 0.914. GO enrichment of differentially expressed proteins revealed that the main function of VWF is immunoglobulin binding, and KEGG enrichment analysis of differentially expressed proteins showed that VWF is mainly involved in complement and coagulation cascades. Wu et al. found that an increase in VWF level in liver tissues may cause an increase in VWF level in the plasma. In addition, as a noninvasive biomarker, VWF can predict portal hypertension and esophageal varices in patients with HBV infection and cirrhosis 39 . VWF was reported to be associated with the progression of LC in chronic hepatitis and was negatively correlated with platelet count, prothrombin time, and albumin level 40 . Moreover, LC was www.nature.com/scientificreports/ reported to be involved in angiogenesis 41 , whereas VWF suppresses angiogenesis [42][43][44] . VWF may therefore be related to the progression of liver fibrosis.
To the best of our knowledge, this is the first report profiling differentially expressed proteins in plasmaderived exosomes of patients with CHB, LC, and HCC in comparison to HCs. Moreover, we identified novel, non-invasive, and specific biomarkers for diagnosing the progression of CHB. Our results demonstrate that LBP, KV311, and CO9 could be used in combination as indicators for the diagnosis and treatment evaluation of HCC. SVEP1 and VWF may be promising biomarkers for auxiliary diagnosis of HCC and liver fibrosis. Our preliminary results provide insights into the value of exosomes as candidate biomarkers for predicting the progression of HBV-related liver diseases.

Blood sample collection. Twenty-four patients with chronic HBV infection were enrolled at the First
Affiliated Hospital of Medical College, Zhejiang University, Hangzhou, China between October 2020 and November 2020. The patients were divided into 3 groups, including 10 with CHB, 7 with LC, and 7 with HCC, according to the practice guidelines established by the American Association for the Study of Liver Diseases 6 and the guideline established by the Chinese Medical Association (Chinese Society of Hepatology and Chinese Society of Infectious Diseases) 45 . An additional four HCs were included in the study. The information of enrolled cases is listed in Table 2. Exclusion criteria included patients with other chronic liver diseases related to drugs, alcohol, or autoimmune diseases. Patients infected with hepatitis A, C, or D virus, or human immunodeficiency virus, were excluded. This study was approved by the Ethics Review Committee of the First Affiliated Hospital, School of Medicine, Zhejiang University. All volunteers provided written informed consent and a questionnaire for participation in the study, in accordance with the Declaration of Helsinki.
NTA. Vesicle suspensions with concentrations (between 1 × 10 7 and 1 × 10 9 /mL) were examined using the ZetaView PMX 110 (Particle Metrix, Meerbusch, Germany), which was equipped with a 405 nm laser, to determine the quantity and size of particles isolated. A video of 60 s duration was taken with a frame rate of 30 frames/s, and article movement was analyzed using the NTA software (ZetaView 8.02.28).
TEM. For the characterization of exosomes, isolated fractions of the exosomes were analyzed using TEM. Briefly, exosome fraction (10 μL) was placed on a copper mesh and incubated at 37 °C for 1 min. The exosomes were contrasted using uranyl acetate solution for 1 min, washed with sterile distilled water, and dried for 2 min under incandescent light. Finally, the copper mesh was observed and photographed using a transmission electron microscope (H-7650, Hitachi Ltd., Tokyo, Japan).  Trypsin treatment. We added trypsin (3 μL, 1 μg/μL) and TEAB buffer (500 μL, 100 mM) to each protein sample, and incubated the mixture at 37 °C overnight. The digested sample was mixed with an equal volume of 1% formic acid and centrifuged at 12,000 × g for 5 min at room temperature. The supernatant was slowly loaded onto the C18 desalting column, washed three times with wash solution (1 mL, 0.1% formic acid and 4% acetonitrile), and eluted two times with elution buffer (0.4 mL, 0.1% formic acid and 75% acetonitrile). The eluates were pooled and lyophilized.
Liquid chromatography-tandem mass spectrometry (LC-MS/MS) analysis. The lyophilized protein powder was dissolved in 0.1% formic acid (10 μL) in water (solvent A), and then injected into a home-made C18 Nano-Trap column (2 cm × 75 μm, 3 μm). Then, using a mobile phase of 0.1% formic in 80% acetonitrile (solvent B), peptides were separated in a homemade analytical column (15 cm × 150 μm, 1.9 μm). Sample elution was performed at a flow rate of 600 nL/min by increasing the concentration of solvent B from 6 to 100% in 60 min. In addition, the separated peptides were analyzed using the Q Exactive HF-X mass spectrometer (Thermo Fisher), with the Nanospray Flex (ESI) ion source and a spray voltage of 2.3 kV. MS raw data were retrieved from the UniProt database (http:// www. unipr ot. org); oxidation of methionine and acetylation of the N-terminus were specified as variable modifications, whereas carbamidomethylation was specified as a fixed modification. The identified protein contained at least 1 unique peptide with a false discovery rate of no more than 0.01. Then, GO analysis was performed to evaluate the potential roles of differentially expressed proteins in biological processes, as cellular components, and in molecular functions. The databases of Clusters of Orthologous Groups and KEGG were used to annotate protein families and pathways.
Statistical analysis. For the detection of differentially expressed proteins, we used the limma package in R 47 to analyze differences between groups of duplicate samples. MaxQuant software was used for protein identification and relative quantification. Swissprot Database was used to annotate proteins. Blast software was used to compare protein sequences with NR (Non-Redundant Protein Sequence) database. According to the corresponding relationship between NR and GO provided by the UCSC database, the results of protein GO annotation were obtained. The enrichment factor was calculated as follows: Differences were considered significant when P ≤ 0.05 and fold change > 1.1. Analysis of differences between groups was performed using t-tests.

Data availability
The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request. www.nature.com/scientificreports/