LC-MS/MS based metabolomics and proteomics reveal candidate biomarkers and molecular mechanism of early IgA nephropathy

Study population

Samples of IgAN plasma were obtained from Zhejiang Provincial People’s Hospital at the time of diagnostic renal biopsies performed. Patients with a pathologically confirmed IgAN diagnosis and not treated with hormone immunosuppressants were included. People without kidney dysfunction constructed the healthy control group. Fifty-three individuals (33 patients with IgAN and 20 healthy controls) were enrolled for proteomic and metabolomic analyses. And 19 people (9 patients with IgAN and 10 healthy controls) were tested by ELISA in the validation set. The ethics committee approved this study for clinical studies at Zhejiang Provincial People’s Hospital. For plasma sampling, whole blood was collected in an EDTA tube and centrifugated at 1000 g for 10 min. Plasma was collected after centrifugation and kept at −80℃ until processed.

Sample preparation for MS proteomic analysis

Samples of plasma were processed according to the previously described procedure with modifications [22]. Briefly, 100 μg of the protein was purified by methanol/chloroform precipitation. After centrifugation, the supernatant was removed, and the protein precipitate was dried. Afterward, the sample was resuspended in 50 mM NH4HCO3 solution and reduced with dithiothreitol (0.25 M in 50 mM NH4HCO3) for 30 min at 56 °C, then alkylated with 0.3 M chloroacetamide for 30 min at 37℃ in darkness. After alkylation, protein samples were diluted by adding 40 μL 100 mM NH4HCO3 and digested by trypsin at 37℃ for 16 h (enzyme to protein ratio of 1:100). The digestion was terminated by adding 20 μL 0.1% formic acid. After that, the samples were then purified using Ziptip (Merck Millipore, MA) C18 cartridges according to standard procedures. Mixed peptides for quality control were prepared by mixing equal volumes of each sample.

Nanoflow liquid chromatography-tandem mass spectrometry analysis of proteomics

An Orbitrap 480 (Thermo Fisher Scientific, Dreieich, Germany) coupled with an EASY-nLC 1200 nanoflow liquid chromatography system (Thermo Fisher Scientific, Dreieich, Germany) was used to obtain mass spectral data. Then peptides were separated on an Acclaim Pep Map 100 C18 column (250 mm × 75 μm, 2 μm-C18) (Thermo Scientific Inc.). A binary mobile phase system was used, consisting of solvent A containing 0.1% formic acid in water and solvent B containing 80% acetonitrile in 0.1% formic acid/water at a flow rate of 300 nl/min. Elution for 65 min was performed by the following gradient: 0–1 min, 3%-5% B; 1–52 min, 8%—40% B; 52–56 min, 40%-95% B; 56–65 min, 95% B. Each peptide sample was ionized by an NSI source (Thermo Fisher Scientific, Dreieich, Germany). Data-independent acquisition (DIA) mode was used to detect the peptide ions. The spray voltage was set at static, and the ion transfer tube temperature to 320 °C. The full scan range was set to 325–1500 m/z, MS1 resolution to 120,000, MS1 automatic gain control (AGC targets) to 300, and MS1 maximum injection time to “Auto”. Fragmentation used higher‐energy collision disassociation (HCD) with a normalized collision energy of 30. The scan was equally divided into 48 continuous isolation windows for MS/MS scan, window overlap was 1 m/z. Precursors were prevented from being repeatedly sequenced by setting the expected peak width to 30 s. MS/MS resolution was set to 30,000, MS/MS AGC target to “Standard”, and MS/MS maximum injection time to “Auto”.

Protein identification from the MS data

For the construction of the spectral library, the mixed peptides were fractionated and analyzed by LC–MS/MS in DDA mode. The DDA raw data were processed on the Protein Discoverer (Thermo Fisher Scientific, Dreieich, Germany) software with database UP000005640 downloaded from Uniprot. The result was imported into Skyline, and the spectral library was constructed. For protein quantification, chromatograms were extracted from RAW files acquired in DIA mode, and the target FASTA file was using database UP000005640. MSstats-Input file was exported and further processed using the R package MSstats for protein quantification and group comparison.

Functional analysis of proteins

Gene set enrichment analysis (GSEA) was performed using the Kyoto Encyclopedia of Genes and Genomes (KEGG) subset of canonical pathways gene set in the Molecular Signatures Database (MSigDB). The KEGG database (http://www.genome.jp/kegg/) was used to map differentially expressed proteins (DEPs) pathway mapping. The clusterprofiler R package was used to annotate DEPs according to Gene Ontology (GO) database. A protein–protein interaction (PPI) network was constructed using the STRING database (http://string-db.org) and drawn in Cytoscape 3.9.0.

Sample preparation for MS metabolomics analysis

The plasma sample (50 μL) was mixed with ketoprofen (internal standard, IS1) purchased by Makclin (Shanghai, China), Sulfamerazine (internal standard, IS2) purchased by Aladdin (Shanghai, China), 2-chloro-L-phenylalanine (internal standard, IS3) purchased by Makclin (Shanghai, China), and 150 μL of methanol (MS grade). The mixture was kept at − 20 ℃ for 30 min and centrifuged at 14,000 g at 4 ℃  for 20 min. The supernatant was further dried in a vacuum centrifugal concentrator. The quality control (QC) sample was prepared by mixing an equal aliquot from each plasma sample. All samples were redissolved by methanol in water (v:v = 2:1) before MS detection.

Untargeted metabolomics assay

Ultra Performance Liquid Chromatography (UPLC) (Thermo Fisher Scientific, Dreieich, Germany) with Waters ACQUITY UPLC T3 C18 Column (2.1 mm × 100 mm, 1.8 um) was used to perform the separation at a flow rate of 0.30 mL/min. The injection volume was 8 μL, and the auto-sampler temperature was 4 °C. Solvent A (0.1% formic acid in water) and solvent B (0.1% formic acid in acetonitrile) were applied as mobile phases. Within 20 min, the analytes were separated in gradient condition: 0–1 min, 1% B; 1–15 min, 2% ~ 98% B; 15–18 min, 98% B; 18–20 min, 98% ~ 2% B. A polarity-switching mass spectrometer operated by Thermo Scientific Orbitrap 480 was used to detect metabolites with the spray voltage of 3.5 kV and -2.5 kV in positive and negative modes, respectively. Samples were ionized by electrospray ionization (ESI) (Thermo Fisher Scientific, Dreieich, Germany). The spray voltage was set at static, the ion transfer tube temperature to 300 °C, and the capillary temperature was 350 °C. Precursors were prevented from being repeatedly sequenced by setting the expected peak width to 15 s. The full scan range was set to 70–1050 m/z, MS1 resolution to 60,000, MS1 AGC targets to “Standard”, and MS1 maximum injection time to “Auto”. Fragmentation used HCD with a normalized collision energy of 30, 50, 150. MS/MS resolution was set to 15,000, MS/MS AGC target to “Standard”, and MS/MS maximum injection time to “Auto”.

Metabolomics data processing

The acquired data were processed with Compound Discoverer 3.1 software (Thermo Fisher Scientific, Dreieich, Germany), including retention time correction, peak identification, peak extraction, peak integration, and peak alignment. Data were annotated by online databases KEGG, HMDB, mzCloud, and Chemspider. MetaboAnalyst (https://www.metaboanalyst.ca) was used to perform the principal component analysis. KEGG pathway databases and HMDB (https://hmdb.ca) were used to identify the top metabolic pathways associated with IgAN.

Machine learning analysis

The LASSO regression analysis was performed in R using the glmnet package and binary logistic regression to construct the diagnosis model. To further assess the performance of the diagnosis model, a tenfold cross-validation was conducted. The ROC curve was used to evaluate the model diagnostic performance. And the diagnosis model was further verified by a support vector machine (SVM) and random forest.

Validation of potential markers by ELISA assays

To further validate the reliability of the combination biomarkers, including PRKAR2A, IL6ST, SOS1, and palmitoleic acid, were selected for targeted analysis by ELISA assays. The independent batch of samples, including 9 plasma samples from the IgAN group and 10 from healthy people, were used for the assays. They were measured by a human PRKAR2A ELISA kit (PRKAR2A_AE1652A-96 T, Chuangshi, Tianjin, China), human Gp130/IL6ST ELISA Kit (EK0367, BOSTER, California, USA), human SOS1 ELISA (EH12498-96 T, FineTest, Wuhan, China) kit and human palmitoleic acid ELISA kit (CB15581-Hu, COIBO, Shanghai, China), respectively.

Statistics analysis

Statistical analyses were conducted with the program R 4.1.0. Two-way unpaired Student's t-tests were used to compare two groups. Differences were considered significant at a p < 0.05.

留言 (0)

沒有登入
gif