Molecular and functional characterization of urine-derived podocytes from patients with Alport syndrome

Abstract

Alport syndrome (AS) is a genetic disorder involving mutations in the genes encoding collagen IV α3, α4 or α5 chains, resulting in the impairment of glomerular basement membrane. Podocytes are responsible for production and correct assembly of collagen IV isoforms; however, data on the phenotypic characteristics of human AS podocytes and their functional alterations are currently limited. The evident loss of viable podocytes into the urine of patients with active glomerular disease enables their isolation in a non-invasive way. Here we isolated, immortalized, and subcloned podocytes from the urine of three different AS patients for molecular and functional characterization. AS podocytes expressed a typical podocyte signature and showed a collagen IV profile reflecting each patient's mutation. Furthermore, RNA-sequencing analysis revealed 348 genes differentially expressed in AS podocytes compared with control podocytes. Gene Ontology analysis underlined the enrichment in genes involved in cell motility, adhesion, survival, and angiogenesis. In parallel, AS podocytes displayed reduced motility. Finally, a functional permeability assay, using a podocyte-glomerular endothelial cell co-culture system, was established and AS podocyte co-cultures showed a significantly higher permeability of albumin compared to control podocyte co-cultures, in both static and dynamic conditions under continuous perfusion. In conclusion, our data provide a molecular characterization of immortalized AS podocytes, highlighting alterations in several biological processes related to extracellular matrix remodelling. Moreover, we have established an in vitro model to reproduce the altered podocyte permeability observed in patients with AS. © 2020 The Authors. The Journal of Pathology published by John Wiley & Sons Ltd on behalf of Pathological Society of Great Britain and Ireland..

Keywords: Alport syndrome; co-culture; collagen IV; genetic defects; glomerular endothelial cells; permeability; urine-derived podocytes.

Figure 1

Morphology and marker expression of cultured AS podocytes. (A) Representative light microscopy images of undifferentiated (upper panels) and differentiated (lower panels) podocytes from normal urine (U‐pod), normal kidney (K‐pod), and AS patient urine. Original magnification: ×20; scale bar: 20 μm. (B) Western blot analysis (representative images and quantification) of nephrin, podocin, synaptopodin, and WT1 of differentiated control (U‐pod and K‐pod) and AS podocytes. Data are expressed as mean ± SD of band intensity normalized to vinculin or GAPDH of three different experiments. *p < 0.05. (C) VEGFA ELISA performed on cell lysates and cell supernatants obtained from differentiated AS podocytes, compared with control (K‐pod). Data are mean ± SD of three different experiments. *p < 0.05. (D) RT‐qPCR for synaptopodin (SYNPO), podocalyxin (PODXL), and VEGFA transcripts. Data are shown as relative quantification, normalized to GAPDH and to control podocytes (U‐pod). Three clones for each AS patient were used and data are expressed as mean ± SD of three different experiments. GECs were used as a negative control. *p < 0.05 AS patient podocytes versus U‐pod; ^# p < 0.05 AS patient podocytes versus kidney‐derived podocytes (K‐pod); ^$ p < 0.05 and ^$$$ p < 0.0001 AS patient podocytes versus GECs (negative control). (E) RT‐qPCR for podocyte marker transcripts (SYNPO, PODXL, VEGFA) and collagen IV isoforms (COL4A3, COL4A5) of AS patient 1 podocytes before and after differentiation. Data shown as relative quantification, normalized to GAPDH and to undifferentiated AS patient 1 podocytes, expressed as mean ± SD. ***p < 0.0001.

Figure 2

Collagen IV isoform expression in AS podocytes. (A) 3D models of the C‐terminal C4 domains for Col4α3 and Col4α5 in AS patients were generated using Phyre2 software, as described in supplementary material, Supplementary materials and methods. Images show the structural modifications of COL4α3 and COL4α5 chains predicted on the basis of the different mutations, compared with normal. In patients AS1 and AS2, frameshift mutations cause the protein to be truncated and with predicted alteration in folding (c.1937dup and c.2777del), or conformation (c.4803del). In patient AS3, a single nucleotide variant (c.2546G>A) leads to minor differences in protein folding. (B) COL4α3 ELISA performed on cell lysates and cell supernatants of differentiated AS podocytes and control kidney‐derived podocytes (K‐pod). (C) COL4α5 ELISA performed on cell lysates and cell supernatants of differentiated AS podocytes compared with control kidney‐derived podocytes (K‐pod). ELISA data are expressed as arbitrary units (A.U.) and are the mean ± SD of three different experiments normalized to control (referred to as 1). *p < 0.05; **p < 0.001; ***p < 0.0001. (D) RT‐qPCR for COL4A1 and COL4A2 genes. Data are shown as relative quantification, normalized to GAPDH and to control kidney‐derived podocytes (K‐pod). Three clones for each AS patient were analysed and data are expressed as mean ± SD. GECs were used as a negative control. ^$ p < 0.05 and ^$$$ p < 0.0001 versus K‐pod; *p < 0.05 versus U‐pod.

Figure 3

RNA‐seq analysis of AS patients. (A) Heatmap of podocyte signatures showing similar levels of expression of podocyte genes in AS podocytes (three clones per patient) and in control podocytes (U‐pod). (B) Pearson similarity plot of podocyte signature showing similar levels of expression of podocyte genes in AS podocytes (three clones per patient) and in control podocytes (U‐pod). (C) Pie chart representation of up‐regulated (red) and down‐regulated (blue) differentially expressed transcripts in AS patient‐ compared with control urine‐derived podocytes. RNA‐seq was performed on three AS patients (AS P1, P2, P3), including three clones for each patient, and urine‐derived podocytes (U‐pod) as a control.

Figure 4

Validation of differentially expressed genes in AS. (A) mRNA expression of genes differentially expressed in AS related to cell motility, survival, and angiogenesis was evaluated by RT‐qPCR. Data are shown as relative quantification, normalized to GAPDH and to control kidney‐derived podocytes (K‐pod). One clone for each AS patient was analysed and the results are expressed as mean ± SD of the three AS patients. (B) Single‐cell tracking assay showing reduced motility in AS podocytes compared with urine‐derived podocytes (U‐pod) and kidney‐derived podocytes (K‐pod). Three independent experiments were performed. *p < 0.05; **p < 0.001. (C) Representative images of the cell motility of control podocytes and AS podocytes..

Figure 5

Static permeability assay in podocyte–GEC co‐cultures. (A) Schematic representation of co‐culture and experimental set‐up: podocytes and GECs were co‐cultured for 48 h before the permeability assay was performed, as described in the Materials and methods section. (B) Immunofluorescence of co‐cultures showing the entire field and cross‐sections (XZ and YZ) of both the upper podocyte layer (left) and the lower GEC layer (right). Cells were stained with phalloidin (green) and nuclear staining was performed with Hoechst dye 33342. Original magnification: ×400. (C) Snapshot of a perspective view of 3D reconstruction of the co‐culture system (see supplementary material, Video S1). (D, E) FITC‐BSA permeability indicating albumin passage from the GEC compartment to the podocyte compartment was measured over 6 h. AS podocytes alone or in co‐culture showed significantly increased filtration compared with control podocytes alone or in co‐culture. Data are expressed as the mean amount of filtered BSA‐FITC of four different experiments using at least three inserts for each condition in each experiment. ***p < 0.0001 insert versus all conditions; ^$$$ p < 0.0001 AS podocytes versus K‐pod; ^### p < 0.0001 AS podocyte co‐culture versus K‐pod co‐culture. U‐pod: urine‐derived podocytes; K‐pod: kidney‐derived podocytes; AS 1‐2‐3 pod: AS 1‐2‐3 patient podocytes.

Figure 6

Millifluidic system and permeability assay in podocyte–GEC co‐cultures. (A) Schematic representation of the millifluidic system, as described in the Materials and methods section. Cells seeded in the LiveBox (LB) were perfused with fluids circulated by peristaltic pumps at 100 μl/min in two independent circuits. (B) Representative image of a chamber of LiveBox. (C) After 48 h of podocyte/GEC co‐culture, the passage of FITC‐BSA from the lower to the upper compartment was measured in the dynamic system over 3 h. The permeability in AS co‐cultures was significantly higher than that in control kidney‐derived co‐cultures. Data are expressed as mean ± SD of three different experiments performed in parallel in control and AS podocytes. *p < 0.05. K‐pod: kidney‐derived podocytes.