Human kidney samples (n = 8) were collected from nephrectomy procedures conducted to address renal dysfunction triggered by ureteral stone obstruction (Supplementary Table 1). We obtained these kidney specimens from Wuhan Third Hospital and Huangshi Central Hospital. Normal control samples (n = 8) were gathered from individuals with healthy kidneys who underwent nephrectomy due to either tumor-related or trauma-induced ruptured kidneys, without any concurrent kidney diseases. All procedures were performed adhering to the ethical principles outlined in the Helsinki Declaration. The Ethics Committee of Wuhan Third Hospital (WuSanYiLun KY2023-057) and Ethics Committee of Huangshi Central Hospital (2024-17) authorized these investigations. In this study, no form of compensation was provided to any human research participants.
STAP2 knockout mice and genotypeBreeding pairs of STAP2 null C57BL/6J mice (STAP2−/−; age, 6-8-week) were purchased from Cyagen Biosciences Inc. that was headquartered in Suzhou (China) and subsequently housed in the Center of Experimental Animals at Wuhan Third Hospital. We maintain these mice particularly under specific pathogen-free conditions. For the induction of renal fibrosis models, male STAP-2−/− C57BL/6J mice (age, 8-10-week) were specifically utilized. Genotyping of these mice was carried out through genomic DNA extracted from their tails via polymerase chain reaction (PCR). The primer sequences employed for this genotyping procedure are accessible in Table 1. During genotyping, the presence of a singular 724 bp band was suggestive of the wild-type genotype, whereas the presence of only a 528 bp band signified the homozygous (STAP2−/−) genotype. In the case of heterozygous mice (STAP2+/−), both the 724 and 528 bp bands were observed, as displayed in Supplementary Fig. 1A.
Table 1 PCR primers for mouse genotyping used in this studyAnimal modelsWe established murine models of renal fibrosis through three different methods: UUO, IRI, and cisplatin-induced renal interstitial fibrosis (RIF). For the UUO-induced RIF model, mice were anesthetized utilizing a face mask delivering 30% isoflurane. A right abdominal incision was made to expose the right ureter, which was either fully ligated 1 cm below the renal pelvis using 5.0 silk ligature (for 7 days ligated group) or manipulated without ligation (for the sham group). After 7 days, murine kidneys and serum were collected. In the IRI-induced RIF model, mice were anesthetized through a face mask delivering 30% isoflurane. The right kidney pedicle was exposed through a dorsal incision on the right side, and an atraumatic Schwartz micro-vessel clamp was applied to the right renal pedicle for 45 min. The removal of the clamp was subsequently performed to facilitate reperfusion, allowing the tissues to restore their normal coloration. Kidneys and serum were collected after 8 weeks. The cisplatin-induced RIF model involved injecting cisplatin at a dose of 8 mg/kg body weight four times (once a week). Kidneys and serum were collected on day 28.
Animals were divided into 4 groups randomly (n = 6 per group): (1) wild-type sham-operated, (2) wild-type renal fibrosis, (3) STAP2–/– sham-operated, and (4) STAP2–/– renal fibrosis. These experiments were carried out adhering to the rules of Experimental Animal Ethics Committee of Wuhan Third Hospital (WuSanYiShiLun SY2023-003).
Cell cultivation and treatmentThe China Center for Type Culture Collection (Wuhan, China) provided the normal human kidney proximal tubule epithelial cell line (HK-2), which was grown in a DMEM/F12 medium (#G4610-500ML, Servicebio, Wuhan, China) supplemented with 10% fetal bovine serum (FBS, Gibco, USA). Initially, we seeded HK-2 cells into six-well plates and allowed to incubate overnight in the complete medium. Subsequently, these cells underwent a transformative process with plasmids or siRNA particularly in the OPTI MEM medium (#Gnm22600-1, Genom, Zhejiang, China). Plasmids expressing Flag-tagged STAP2, HA-tagged HSP27 and siRNA were constructed by Gene Create (Wuhan, China). After a 24-hour incubation period, the cells were incubated in a complete medium supplemented with 10% FBS. To visualize the cells, a microscope was utilized, while the transfection rate was precisely indicated through the quantitative real-time transcription PCR (qRT-PCR) and Western blotting (WB). In an endeavor to simulate the in vitro assay of renal fibrosis, we seeded 1.5 × 106 cells into six-well plates, and the overnight incubation was performed in the complete medium. Following 12-hour starvation period without 10% FBS, we treated HK-2 cells (confluence, 60-70%) utilizing LPS (#L2880, Sigma, USA) at an ultimate concentration of 50 ng/ml. Subsequently, the cells were cultivated for additional assay after 24 h.
RNA isolation and qRT-PCR analysisThe process of isolating total RNA was performed using murine kidneys and HK-2 cells through a specialized RNA isolation kit (#R6934–01, Omega Bio-Tek, USA). Thereafter, we conducted cDNA synthesis employing the TOYOBO ReverTra Ace qPCR RT kit, adhering diligently to the protocols outlined by the manufacturer. After this intricate process, qRT-PCR was executed utilizing the Bio-Rad CFX Manager system. In this molecular investigation, indicators of fibrosis, involving collagen type I (Col-I) and fibronectin (FN), were precisely identified. The relative changes of mRNA level were determined by 2-△△ CT method with ACTB as control. The primer sequences utilized are accessible in Table 2.
Table 2 Primers for qRT-PCR used in this studyWestern blotting (WB)Murine kidneys and HK-2 cells underwent lysis through lysis buffer, enriched with a protease inhibitor cocktail and phosphatase inhibitors, which were purchased from Servicebio. The quantification of protein content was performed through a BCA Protein Assay kit purchased from Servicebio. Subsequently, 20 µg of each protein specimen was precisely separated via 6%, 10%, or 12% SDS-PAGE, respectively, and the transfer particularly onto polyvinylidene difluoride membranes was thereafter carried out. We blocked the membranes through 5% non-fat milk for a duration of 1 h, followed by an overnight incubation with a series of primary antibodies precisely detailed in Supplementary Table 2. This nocturnal incubation occurred at a temperature of 4 °C to ensure precision. The resulting blots were elegantly scanned utilizing a two-color infrared imaging system (Bio-Rad, USA) and subjected to analysis employing both ImageLab and ImageJ software, in which the latter was developed by the National Institutes of Health (Bethesda, USA).
Histology and immunohistochemistry assayParaffin-embedded murine renal sections, which cut at a thickness of four micrometers, underwent staining with hematoxylin-eosin (HE) on the basis of the instructions documented by the manufacturer. This precise process aimed to figure out the structural integrity of the kidney and fibrotic injuries. The extent of tubular injury was scored from 0 to 4 based on the percentage of the cortico-medullary region affected, as documented previously [8]: 0 indicating no damage, 1 for less than 25% damaged area, 2 for 25–50% damaged area, 3 for 50–75% damaged area, and 4 for more than 75% damaged area.
We investigated interstitial fibrosis and glomerulosclerosis through Masson staining, Sirius red staining, and immunohistochemistry targeting α-smooth muscle actin (α-SMA), FN, and Col-I in the renal tissue. The renal fibrotic-positive area was quantified by analyzing the percentage of colored pixels in the total field encompassing the cortico-medullary region. Six randomly selected fields were analyzed on each slide, specifically at 200× or 400× magnification.
We conducted histological and immunohistochemical (IHC) staining procedures adhering to protocols outlined by manufacturers in advance. Deparaffinizing each slide was performed, and antigen retrieval was thereafter executed. Once an overnight incubation was implemented with antibodies meticulously detailed in Supplementary Table 2 (temperature, four centigrade degree), each tissue specimen was subjected to incubation with a horseradish peroxidase-conjugated secondary antibody and subsequently stained using 3,3-diaminobenzidine tetrahydrochloride (DAB, #HP191902) purchased from Servicebio. Each slide’s positive regions were captured under a TE2000-U microscope (Nikon Eclipse, Japan) at magnifications of 200× in six randomly selected fields. The analysis of resulting images was performed through ImageJ software.
Immunofluorescence stainingFrozen renal sections, precisely sliced at a 5-µm thickness, were dedicated to the subsequent staining of ECM-related proteins. We immersed these sections in a protein-blocking solution before undergoing a sequential incubation process with primary antibodies targeting Col-I (#GB11022-3), FN (#GB13091), and E-cadherin (E-ca) (#GB13083), which all purchased from Servicebio. Following this precise incubation, the treatment of sections was performed with appropriate secondary antibodies, accelerating the binding process. Additionally, 25 µL of DAPI solution was delicately applied to the samples and allowed to incubate for 5 min. Utilizing the capabilities of a fluorescence microscope equipped with a digital camera, the intricate patterns of fluorescence were precisely visualized.
To quantify the average fluorescence intensity, a rigorous analysis was performed through ImageJ software, and the average fluorescence intensity was computed through dividing the total fluorescence intensity of a specific region by its corresponding area. To ensure a comprehensive assessment, at least six distinct, non-overlapping regions were analyzed per sample.
Immunoprecipitation (IP) and mass spectrometry (MS) analysisHK-2 cells underwent lysis utilizing the pre-cooled IP lysis buffer. Following centrifugation, the collection of resultants was performed, and overnight incubated separately with anti-FLAG (#F1804, Sigma) or anti-HA (#T0050, Affinity, China) antibodies particularly at a temperature of 4 °C. Concurrently, protein A/G-agarose beads were suspended in lysis buffer encompassing a protease inhibitor and left to being 4-hour incubated particularly at 4 °C. The rinsing of beads was conducted four times, and the expression levels of the eluted proteins were analyzed via WB.
For the comprehensive shotgun proteomic analysis of the eluted proteins, an initial step involved the partial separation of protein mixtures to approximately 1.5 cm using 10% SDS-PAGE. The region containing the proteins of the appropriate size was excised and subjected to in-gel tryptic digestion to extract peptides. These recovered peptides were thereafter analyzed through data-dependent liquid chromatography in conjunction with tandem mass spectrometry (MS/MS) on the Q Exactive™ liquid chromatography-mass spectrometry system attained from Thermo, USA.
In this intricate process, we injected peptide samples into an automatic sampler, followed by separation through a C18 trapping column (3 μm, 75 μm × 20 mm, 100 Å). Subsequently, they were eluted onto an analytical column (50 μm × 150 mm, 2 μm particle size, 100 Å pore size) attained from Thermo for further separation. The data acquisition encompassed the precise collection of both intact mass data from MS and fragmentation pattern data from MS/MS of the peptides. This was achieved through an advanced data-dependent acquisition (DDA) strategy, incorporating dynamic exclusion techniques to enhance and maximize the depth of coverage.
The resulting data were thereafter subjected to identification through Protein Discoverer (V2.3), employing the Percolator algorithm. For this analysis, the UniProt human proteome reference database (uniprot-Human-20230111.fasta) served as the foundation. The results were filtered based on criteria with a cutoff value of ≥ 0.05 for Maximum Delta Cn and Maximum Rank for peptide-spectrum matches (PSMs), removing entries retrieved from the decoy database and contaminating proteins.
RNA sequencing (RNA-seq) analysisFirstly, the total RNA from LPS-induced siNC or siSTAP2 HK-2 cells was extracted. Secondly using the Fragment Analyzer, RNA quality is accurately evaluated by generating a digital profile of RNA quality through RNA concentration measurement, 28 S/18S detection, and RIN/RQN assessment. Once qualified, we diluted the RNA to a concentration of 300 ng/µL for subsequent library preparation and sequencing, adhering precisely to instructions outlined by the manufacturer. Following sequencing, the sequencing data were filtered with SOAPnuke (v1.5.6) by: (1) removing reads containing sequencing adapter; (2) removing reads with more than 20% low-quality bases (base quality ≤ 15); and (3) removing reads with over 5% unknown bases (‘N’). Filtered reads were subsequently obtained and stored in FASTQ format. The aligning of sequenced reads was thereafter performed through HISAT2 (v2.1.0), and the resulting mapping data were converted and sorted utilizing Samtools (v1.4). The TPM value, computed by RSEM (v1.3.1), was utilized for precise quantification of gene expression level. To visualize the gene expression patterns across diverse samples, we plotted a heatmap through the pheatmap R package (v1.0.8). Subsequently, differential expression analysis was carried out via DESeq2 (v1.4.5) with a stringent significance threshold of Q value ≤ 0.05 (or FDR ≤ 0.001). For further functional insights, the clusterProfiler R package was used to perform enrichment analysis of the annotated differentially expressed genes using Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG). With a strict criterion of Q value ≤ 0.05, the pathways and enriched terms’ significance levels were presented.
Protein-binding site predictionUtilizing in silico methodologies, we identified putative functional domains modulating the interaction of STAP2 with HSP27. We retrieved AlphaFold PDB files STAP2_HUMAN (ID: AF-Q9UGK3-F1) and HSPB1_HUMAN (ID: AF-P04792-F1) from Uniprot to predict structures as receptor proteins and ligand proteins, respectively. Protein-protein docking was performed using HDOCK server to identify binding modes with the potential to mediate the interaction between STAP2 and HSPB1 [9]. Based on the Docking score and confidence score, the top 10 most probable binding models were developed. A greater negative docking score could be suggestive of an escalated probability of binding model. If the confidence score was higher than 0.7, it indicated that the probability of binding of the two molecules was high. If the confidence score was between 0.5 and 0.7, it was considered that the two molecules could bind together. If the confidence score was below 0.5, it indicated that the possibility of binding of two protein molecules was low (Supplementary Table 3). The top scoring model was analyzed for protein-ligand non-covalent associations via PLIP(https://projects.biotec.tu-dresden.de/plip-web/plip/index) [10].Structural visualizations were implemented using PyMOL 2.5.4 software.
Statistical analysisWe used ImageJ to quantify and standardize the results of WB, IHC, and IF and analyzed all data attained from independent assays through the GraphPad Prism software developed by GraphPad Software 8.0 (San Diego, CA). All data were expressed as mean ± SEM (standard error of the mean). The examination of two groups was performed through the two-tailed Student’s t-test, and multiple groups was performed through the one- or two-way analysis of variance (ANOVA) followed by Tukey’s test. The linear relationship between the variables was represented by linear regression, while the coefficient (r) and P-value were evaluated using Spearman correlation analysis. For all studies, differences were considered significant at P < 0.05.
留言 (0)