Plasma microRNA panel is a novel biomarker for focal segmental glomerulosclerosis and associated with podocyte apoptosis

Abstract

Focal segmental glomerulosclerosis (FSGS) is a frequent glomerular disease, and is the common cause of nephrotic syndrome. However, there is no validated diagnostic blood biomarker for FSGS. Here, we performed a real-time PCR-based high-throughput miRNA profiling to identify the plasma signature for FSGS. We found four miRNAs (miR-17, miR-451, miR-106a, and miR-19b) were significantly downregulated in the plasma of FSGS patients (n = 97) compared with healthy controls (n = 124) in the training, validation, and blinded-test phases. The miRNA panel produced an AUC value of 0.82, and was associated with FSGS severity and histologic classification. A three-miRNA panel, including miR-17, miR-451, and miR-106a was related to FSGS remission. Furthermore, the downregulation of plasma-miRNA signature was not detected in disease controls (n = 119) such as IgA nephropathy (IgAN), mesangial proliferative glomerulonephritis (MSPGN), and membranous nephropathy (MN), and the miRNA panel discriminated between FSGS and disease controls. Pathway analysis showed that the four-miRNA panel may cooperatively regulate the pathways involved in the development of FSGS, such as apoptosis. We identified that phosphatase and tensin homolog (PTEN), Bcl-2-like protein 11 (BCL2L11), and chemokine (C-X-C motif) ligand 14 (CXCL14) were targets of miR-106a in human podocyte. Additionally, miR-106a overexpression suppressed podocyte apoptosis in vitro and the downregulation of four-miRNA panel probably resulted in the enhanced apoptosis in podocyte during FSGS development. Taken together, our data show that the plasma-miRNA panel is a potential independent diagnostic and prognostic factor for FSGS. Above miRNAs are involved in FSGS pathogenesis through regulating podocyte apoptosis.

Fig. 1. A diagram of sample collection and study design.

A schematic of the study outlining the independent patients and samples used in discovery, training, validation, and blinded-test phases of the identification of plasma-miRNA panel for FSGS

Fig. 2. Hierarchical clustering analysis of differentially expressed miRNAs in plasma of FSGS patients in the screening phase.

miRNA profiling in plasma from five FSGS patients and five healthy controls was performed by using a real-time PCR-based high-throughput miRNA array

Fig. 3. Expression of plasma-miRNA panel for FSGS diagnosis in the training phase.

a Expression of miR-17, miR-451, miR-106a, and miR-19b in plasma of FSGS (n = 24) and healthy controls (n = 35). P-values were calculated using the Mann–Whitney test. b ROC analysis of individual miRNAs for the diagnosis of FSGS. c ROC analysis of four-miRNA panel for the diagnosis of FSGS. Logistic regression demonstrated that a linear combination of values for miR-17, miR-451, miR-106a, and miR-19b produced the best model for FSGS diagnosis. *P < 0.05, **P < 0.01

Fig. 4. Expression of plasma-miRNA panel for FSGS diagnosis in the validation phase.

a Expression of miR-17, miR-451, miR-106a, and miR-19b in plasma of FSGS (n = 50) and healthy controls (n = 68). b ROC analysis of individual miRNAs for the diagnosis of FSGS. c ROC analysis of four-miRNA panel for the diagnosis of FSGS in validation set. d ROC curves of the four-miRNA panel generated by analyzing all 221 samples from training, validation, and blinded-test sets. e ROC curve of serum creatinine for the diagnosis of FSGS in the training, validation and blinded-test phases. *P < 0.05, ***P < 0.001

Fig. 5. Association of the miRNA biomarker panel with FSGS severity and remission.

a The expression of four-miRNA panel in FSGS patients with different grades of disease (CKD 1 (n = 27) vs. CKD 2–4 (n = 47)). b The expression of four-miRNA panel in FSGS patients with different subtypes of histologic classification. c The expression of four-miRNA panel between in FSGS patients with proteinuria (n = 56) and in FSGS patients with complete remission stage (urinary protein <400 mg/24 h after treatment) (n = 18). d ROC analysis of individual miRNAs for discriminating FSGS remission. e ROC analysis of three-miRNA panel including miR-17, miR-451, and miR-106a for discriminating FSGS remission. *P < 0.05, **P < 0.01

Fig. 6. Disease specificity of downregulation of four-miRNA panel in FSGS.

a-d The expression of miR-17, miR-451, miR-106a, and miR-19b between in FSGS patients (n = 74) and in other chronic kidney diseases including 69 IgAN patients, 24 MSPGN patients, and 26 MN patients. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 7. Prediction of pathways co-regulated by four-miRNA panel and identification of targets of miR-106a in human podocyte during FSGS development.

a The relative mRNA expression of target genes in miR-106a or miR-control transfected human podocytes. b The protein expression analysis of target genes by western blotting in miR-106a or miR-control transfected human podocytes. c Wild-type or mutant binding sites of CXCL14 3′-UTR for miR-106a. d HEK293 cells were transiently co-transfected with luciferase report vectors containing wild-type or mutant CXCL14 3′-UTR, and either miR-106a mimics or miR-control. Luciferase activities were normalized to the activity of Renilla luciferase. e, f The relative expression of four-miRNA panel and target genes in 20 set of FFPE FSGS tissue specimens and FFPE normal renal tissues. g Immunohistological staining of BCL2L11, PTEN, and CXCL14 in FSGS renal biopsies and normal renal tissues. Scale bars = 20 μm. *P < 0.05, **P < 0.01, ***P < 0.001

Fig. 8. Overexpression of miR-106a suppresses the apoptosis of human podocytes in vitro.

a Target genes and biological pathways for validated four-miRNA panel were identified using prediction algorithm (MICRORNA.ORG) and KEGG pathway enrichment. b Human podocyte cells were transfected with miR-106a or miR-control. The apoptotic rates of these cells were assessed by annexin V/7AAD staining. One representative flow cytometry analysis is shown. The bar graph shows the mean ± SD of three independent experiments. c Apoptosis in situ in FSGS renal biopsies and normal renal tissues were measured by TUNEL. Podocytes are marked by an arrowhead. Scale bars = 20 μm. *P < 0.05