Significance of urine complement proteins in monitoring lupus activity

Patients and sample collection

First of all, this study protocol was approved by the ethics committee of Beijing Shijitan Hospital, Capital Medical University (sjtkyll-lx-2021(55)). The participants all gave written informed consent of each subject, which was in accordance with the provisions of the Helsinki Declaration. All subjects were adult females. Study participants included 24 patients with diagnosed systemic lupus erythematosus (SLE group) and 12 controls that did not have SLE (NC) with routine physical examination. Participants were recruited from Beijing Shijitan Hospital from September 2020 until September 2021. SLE patients were diagnosed through the use of 1997 American College of Rheumatology classification criteria for systemic lupus erythematosus. According to the SLEDAI-2K scoring system, SLEDAI-2K ≥5 were classified as the active group (SLE-A), and SLEDAI-2K<5 were classified as the stable group (SLE-S). Exclusion criteria included acute or chronic infections, tumors and other autoimmune diseases, and any diagnoses of other severe liver and renal disease. The morning midstream urine samples were collected into sterile polypropylene tubes. Immediately after collection, urine samples were centrifuged at 400 ×g for 5 min to remove cell debris and casts and finally supernatants were divided in aliquots and frozen at −80 °C.

Proteomic analysis and bioinformatics analysis

Base on SDS Polyacrylamide Gel Electrophoresis (SDS-PAGE) combined with nano-upgraded reversed-phase liquid chromatography-tandem mass spectrometry (nanoRPLC-MS/MS) method, we detected and analyzed the urinary proteins of SLE patients and controls without SLE using Thermo Q-Exactive mass spectrometer. MaxQuant software was used to process the original mass spectrometry files, and the significant difference in protein quantification between SLE-A and SLE-S (Fold change = 2.0, P < 0.05) was defined as differentially expressed proteins (DEPs). The KEGG database and the Metascape platform were used to analyze the function and pathway enrichment of DEPs to determine the important pathways and proteins related to SLE disease activity. Protein–protein interaction (PPI) network analysis is performed by using a search tool (STRING) database for searching interacting genes/proteins.

Qualitative identification by western blot

To verify the previous screening results, we selected C9 and MASP2 with the most obvious differences for western blot analysis. The total protein was enriched by a 10 kD ultrafiltration device (Millipore) and concentrated at 6000g for 15 min. The volume of the sample was calculated according to the protein mass (7 µg). After SDS-PAGE gel electrophoresis, the protein was transferred to the PVDF membrane and sealed with skim milk powder for 2 h. The PVDF membranes were incubated overnight at 1:1000 anti-C9 antibody and anti-MASP2 antibody at 4 °C. The 1 × TBST buffer solution was used to rinse the membranes three times, each 20 min. Then, 1:3000 s antibody was added and incubated at room temperature for 90 min. Finally, the PVDF membrane was washed with 1 × TBST buffer and detected by the enhanced chemiluminescence method.

Quantitative identification by parallel reaction monitoring

The expression of the target proteins in urine were quantitatively determined by the parallel reaction monitoring method. Twelve urine samples were selected from each of the active group (SLE-A), the stable group (SLE-S) and the normal controls (NC), and 1 ml of urine was taken from each patient for the experiment. The urinary samples were centrifuged at 2,000 g for 10 min at 4 °C, and urea particles were added to the supernatant with a urea concentration of 8M, shaken until the urea particles were fully dissolved. The volume of the samples were concentrated by ultrafiltration in a 10 K ultrafiltration tubes. After protein quantification by the Bradford method, 60 µg proteins were taken and placed in a centrifuge tubes. A total of 5 µl of 1M DTT (5 mM/L) was added to the solution and mixed, and incubated at 37 °C for 1 h; 20 µl of 1M IAA (20 mM/L) solution was added, and after mixing, the reaction was performed for 1 h at room temperature and avoided light. Trypsin was added to the ultrafiltration tube at 1:50 enzyme-to-substrate ratio, and digested at 37 °C for more than 12 h. The mixed samples were fractionated on a Waters BEH C18 column (5 µm, 4.6 × 250 mm) by high performance liquid chromatography system (HPLC), centrifuged at 14,000 g for 20 min, and the supernatant was taken for separation at a flow rate of 0.7 ml/min. Then, the samples were eluted with 100 µl of mobile phase A (100% ddH2O, 0.1% formic acid) and mobile phase B (100% acetonitrile, 0.1% formic acid) in a gradient dissolution. The solvent gradient was set as follows: 5% B, 0 min; 5–8% B, 5 min; 8–18% B, 30 min; 18–32% B, 27 min; 32–95% B, 6 min; 95–5%, 4 min. Finally, the eluates were monitored at UV 214 nm, collected for a tube per minute and merged into 3 fractions, which are lyophilized and waiting for Q Exactive HF-X mass spectrometer testing.

The Q Exactive HF-X mass spectrometer was operated in the data-dependent acquisition mode and there was a single full-scan mass spectrum in the Orbitrap (350–1,500 m/z, 120,000 resolution) followed by data dependent MS/MS scans in an Ion Routing Multipole at 27% normalized collision energy (HCD). The single-sample was reconstituted in 0.1% FA, and injected into the Q Exactive HF-X mass spectrometer (Thermo Fisher) operating in the Parallel reaction monitoring (PRM) mode. For PRM acquisition, primary mass spectrum (MS1) resolution was set to 120,000, and secondary mass spectrum (MS2) resolution was set to 15,000. The m/z range covered from 350 to 1,500 m/z. The maximum capacity of C-trap was set to 3 × 10⁶, and the maximum injection was 80 ms. The dynamic exclusion range was set to 16 s and allow the mass spectrometer always operating in the parallel ion filling and detection mode. The raw data of mass spectrometry (MS) was generated and processed by Skyline software. The false discovery rate was set to 1% for proteins and peptides. The enzyme specificity was set to trypsin, and a maximum of three missed cleavages were allowed in the database search. Parameter settings for screening differential proteins: difference ratio >1.2 or difference ratio <1/1.2, P value <0.05.

Statistical analysis

Statistical analysis was performed using SPSS 21.0 software and the visualization of the data carried out by using GraphPad Prism 9.0 software. The results of quantitative data were expressed as mean ± standard deviation (SD) or median (interquartile range). The P-values were calculated by Student’s t test and Mann–Whitney U test for two groups, or ANOVA (more than two groups). The correlation analysis was performed to examine the relationship between continuous variables by Spearman’s correlation coefficient. Receiver operating characteristic (ROC) curve analysis and area under the curve (AUC) calculations were used to analyze the diagnostic value of urine candidate DEPs in the disease activity of patients with SLE. In all cases two-tailed P < 0.05 was accepted as statistically significant.

Comparison of complement-related proteins between active group (SLE-A) and stable group (SLE-S)

Screening and identifying the urine protein profile of SLE patients demonstrated that complement-related proteins showed varying degrees of difference between the active group (SLE-A) and stable group (SLE-S). To classify the functional annotations of the identified proteins, pathway analysis was performed by kyoto encyclopedia of genes and genomes (KEGG) database. The most representative pathway was complement and coagulation pathways (map04610). Among the urine protein profile, there were 16 proteins involved in the complement pathway, 14 of these proteins were differentially expressed, eight differentially expressed proteins (DEPs) were up-regulated, and six DEPs were down-regulated (Fig. 1A).

These 14 DEPs played different roles in the complement pathway, 10 of these DEPs were key factors of the complement pathway. The expressions of Complement C5 (C5), Complement component C9 (C9), Complement C3 (C3), Complement component C7 (C7), Complement C4-B (C4B), Complement component C8 alpha chain (C8A) and Complement component C8 gamma chain (C8G) were up-regulated between SLE-A and SLE-S, while the expressions of Mannan-binding lectin serine protease 2 (MASP2), Complement component C6 (C6) and Complement factor D (CFD) are down-regulated. The other 4 DEPs were the inhibitory molecules of complement pathway, namely Vitronectin (VTN) whose expression was up-regulated and Complement receptor type 1(CR1), Plasma protease C1 inhibitor (SERPING1) and V-set and immunoglobulin domain-containing protein 4(VSIG4) whose expression were down-regulated (Fig. 1B). The relative abundance of these DEPs among SLE-A, SLE-S and controls that did not have SLE (NC) were shown in Fig. 1C, and detailed information were shown in Table 1. These DEPs may play key roles in disease activity.

Bioinformatics analysis of complement-related proteins

In order to predict the interactions of complement-related proteins, protein–protein interaction (PPI) network was conducted by STRING database. The PPI network involved in the complement pathway were tight and these proteins had strong interactions (Fig. 2A). Based on the Metascape platform, the functional analysis of these proteins revealed that the main enrichment pathways involved in complement and coagulation cascades, complement cascade, complement activation, regulation of complement cascade and other pathways. These findings suggested that these proteins play important roles in the activation and regulation of the complement pathway (Fig. 2B).

Preliminary verification of urine C9 and MASP2 among SLE-A, SLE-S and controls group (NC)

To verify the screening results, we selected the two most different proteins from the DEPs, namely C9 and MASP2, performing western blot analysis. The results showed that the expression of C9 in the urine was significantly increased in most patients with SLE-A compared with SLE-S and NC. But the changes of MASP2 in the three groups were not obvious, and even the detection rate of MASP2 in the urine of individual SLE patients and control groups was low (Fig. S1 ). It may be due to the limitations of the western blot method in the detection of urine proteins, and the sensitivity is not enough. Therefore, we focused on the up-regulated proteins for further verification.

Quantitative verification of up-regulated complement proteins in urine

Parallel reaction monitoring (PRM) technology was used to identified 8 up-regulated proteins related to the complement pathway. The results revealed that the expression of four proteins were consistent with the previous mass spectrometry screening results. The detailed information of the four proteins were shown in Table 2. Compared with the SLE-S and NC, the expression levels of C9, C8A, C4B, and C8G exhibited a significant increase in urine of SLE-A. The levels of these four proteins among the three groups are shown in Fig. 3A.

Differentially expressed proteins (DEPs) were closely associated with lupus activity

Spearman correlation analysis was used to analyze the correlation between four DEPs and disease activity. The results shown that the levels of urine C9 (r = 0.599, P = 0.002), C8A (r = 0.669, P < 0.001), C4B (r = 0.588, P = 0.003), and C8G (r = 0.597, P = 0.002) were significantly positively correlated with SLEDAI-2K. The higher the score of SLEDAI-2K, the greater the concentration of these four proteins (Fig. 3B).

ROC curve analysis was performed to determine the cutoff value of the four DEPs for distinguishing SLE-A from SLE-B. The areas under the curves(AUCs) of urine C9, C8A, C4B and C8G were 0.750 (95% CI [0.554–0.947]), 0.840 (95% CI [0.674–1.00]), 0.757(95% CI [0.556–0.958]) and 0.736 (95% CI [0.526–0.946]), respectively (Fig. 3C). These four proteins all have high discriminatory capacity for disease activity. We evaluated the clinical performance of these four DEPs for SLE disease activity, including sensitivity, specificity, positive predictive value and negative predictive value, using SLEDAI-2K as the reference standard, as shown in Table 3.

Among the four urinary proteins, C8A had the highest judgment value for lupus activity, with an AUCs of 0.840. When the cutoff level was 0.640 ng/ml, the sensitivity, specificity, positive predictive value and negative predictive value were 91.67%, 75%, 78.57% and 90%, respectively.

Statement of Ethics

This study was approved by the ethics committee of Beijing Shijitan Hospital, Capital Medical University. The participants all gave written informed consent of each subject, which was in accordance with the provisions of the Helsinki Declaration.