Isobaric tag for relative and absolute quantitation based quantitative proteomics reveals unique urinary protein profiles in patients with preeclampsia

Abstract Preeclampsia (PE) is one of the most significant pregnancy‐related hypertensive disorders. Currently, there are no useful markers to predict the onset of the condition in pregnant women. To provide further insights into the pathogenesis of PE and identify biomarkers of the condition, we used isobaric tags for relative and absolute quantitation (iTRAQ) proteomics coupled with 2‐D LC‐MS/MS, to analyze urinary protein profiles from 7 PE patients and 7 normotensive pregnant women. A total of 294 proteins were abnormally expressed in PE patients. Of these, 233 were significantly down‐regulated and 61 proteins were significantly up‐regulated. Bioinformatics analysis using the Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) database, found that the most differentially expressed proteins (DEPs) were involved in coagulation and complement pathways, the renin‐angiotensin system and cell adhesion molecules (CAMs) pathways. We further validated three of the DEPs, including serotransferrin (TF) and complement factor B (CFB) by immunoblottingand serum paraoxonase/arylesterase 1 (PON1) by ELISA using 14 pairs of urine samples from PE patients and normal pregnant women. Taken together, our results provide the basis for further understanding the pathogenesis of PE and identifying predictive biomarkers.

| 5823 F I G U R E 1 Functional analyses of preeclampsia differentially expressed proteins. Differentially expressed proteins (DEPs) enriched for different cellular components (A), biological processes (B) and molecular functions (C), according to GO annotation. KEGG pathway enrichment analysis of DEPs for significantly enriched pathways (D) pre-existing diabetes, multiple (twin) pregnancy, nulliparity and family history have been associated with PE. 6 Nonetheless, it still remains extraordinarily difficult for clinicians to predict patients who will develop the condition in pregnancy. Current research has focused on mapping the changes in the levels of soluble blood vascular factors in response to disease activity. 7 Using the IGY 14 depletion kit (Thermos Fisher), fresh midstream urine samples were first depleted of high molecular weight proteins including immunoglobulin and albumin, which are known to interfere with iTRAQ analysis. 12 A total of 100 μg of deleted protein extract from each sample were then trypsinized and the resulting peptides subjected to iTRAQ and MS analysis. The Protein Prophet algorithm was used to assign the probability that each peptide was related to a particular protein. Normalization of the relative peptide intensities of protein data was carried out iteratively (across samples and spectra) using ANOVA, as previously We found that the seven women with severe PE consistently showed differences in the urine proteomic fingerprint compared to their seven matching controls. Among a total of 985 proteins identified, 294 (30%) were differentially expressed proteins (DEPs) in PE patients (>1.5 fold change, P < 0.05). These 294 DEPs were further analysed by Blast2GO to characterize the relevant biological processes, cellular components and molecular functions associated with PE ( Figure 1). We found that the vast majority of DEPs were located in the extracellular space and thus were secreted proteins ( Figure 1A). Gene Ontology analysis for biological processes revealed that the main functional categories of these proteins were complement activation, platelet degranulation, receptor-mediated endocytosis and negative regulation of endopeptidase activity ( Figure 1B). According to molecular function analysis, these proteins were mainly involved in serine-type endopeptidase activity, serine-type endopeptidase inhibitor activity and calcium ion binding ( Figure 1C). KEGG pathway analysis found that the DEPs were involved in coagulation and complement pathways, the renin-angiotensin system and cell adhesion molecule interactions ( Figure 1D). The challenge now is to determine whether any of these DEPs, or new unknown proteins, appear early in the urine of patients at risk for developing PE. One strategy would be to enroll pregnant patients with known risk factors for PE into a new study whereby urine samples are taken weekly from gestational age of 16 weeks before symptoms first appear 4 and use iTRAQ proteomics to analyse significant protein changes. A larger set of normotensive pregnant patients at the same gestational age could be used as the control group. Prospective analysis would reveal any changes in the normal levels of urine proteins and therefore any significant early markers could be retrospectively identified from those patients that do develop clinical PE. By this approach, it may be possible for the first time to identify the key protein changes in urine with a strong association for development of PE, leading to the design of simple assays to identify at risk pregnancies. This would then allow careful monitoring of these pregnancies with administration of the appropriate therapies to reduce symptoms, leading to better management of a term pregnancy.

ACK N OWLED G EM ENTS
This study was funded by grants from The National Natural Science Foundation of China (81471514) awarded to Yijun Song, and the National Natural Science Foundation of China (81501337) awarded to Xiya Zhou. The authors sincerely thank department of central laboratory for technical assistance. We thank Ning Chen and Hanxing Cheng for providing critical advice.

CO N FLI C T O F I NTE R E S T
DSC is an employee of Berry Genomics Corporation. The other authors confirm that there is no conflict of interest.

AUTH O R CO NTR I B UTI O N S
WD and YS collected the samples, designed and conducted the experiment. WD and DSC wrote the paper. YS revised the manuscript.
BQ, XC, SL and ZW analysed the data. XZ and JL provided critical review and advice.