Abstract

Recently, a novel heterozygous missense mutation c.T1421G (p. L474R) in the PODXL gene encoding podocalyxin was identified in an autosomal dominant focal segmental glomerulosclerosis (AD-FSGS) pedigree. However, this PODXL mutation appeared not to impair podocalyxin function, and it is necessary to identify new PODXL mutations and determine their causative role for FSGS. In the present study, we report the identification of a heterozygous nonsense PODXL mutation (c.C976T; p. Arg326X) in a Chinese pedigree featured by proteinuria and renal insufficiency with AD inheritance by whole exome sequencing (WES). Total mRNA and PODXL protein abundance were decreased in available peripheral blood cell samples of two affected patients undergoing hemodialysis, compared with those in healthy controls and hemodialysis controls without PODXL mutation. We identified another novel PODXL heterozygous nonsense mutation (c.C1133G; p.Ser378X) in a British–Indian pedigree of AD-FSGS by WES. In vitro study showed that, human embryonic kidney 293T cells transfected with the pEGFP-PODXL-Arg326X or pEGFP-PODXL-Ser378X plasmid expressed significantly lower mRNA and PODXL protein compared with cells transfected with the wild-type plasmid. Blocking nonsense-mediated mRNA decay (NMD) significantly restored the amount of mutant mRNA and PODXL proteins, which indicated that the pathogenic effect of PODXL nonsense mutations is likely due to NMD, resulting in podocalyxin deficiency. Functional consequences caused by the PODXL nonsense mutations were inferred by siRNA knockdown in cultured podocytes and podocalyxin down-regulation by siRNA resulted in decreased RhoA and ezrin activities, cell migration and stress fiber formation. Our results provided new data implicating heterozygous PODXL nonsense mutations in the development of FSGS.

Introduction

Podocytes are terminally differentiated pericyte-like cells with a complex architecture that have an important role in the structure and function of the glomerular filtration barrier [1]. Primary podocytopathy, caused by podocyte injury or loss, can manifest with proteinuria, nephrotic syndrome and/or renal impairment in children and adults and is frequently associated with histological appearances of focal segmental glomerulosclerosis (FSGS) [2]. A minority of podocytopathy is due to genetic mutations, which affect proteins that are expressed in a variety of locations within a podocyte. Currently, mutations that cause podocyte structural and functional alterations have been found in more than 30 genes in human. These genes can mainly be divided into the following categories [3]: (1) podocyte slit diaphragm-associated molecules (e.g. NPHS1, NPHS2, CD2AP), (2) actin-binding complexes within the foot process actin network (e.g. ACTN4, INF2, ANLN, ARHGAP24), (3) actin regulation by RHO/RAC/CDC42 (e.g. ARHGAP24, ARHGDIA, KANK1/2), (4) podocyte transcription factors (e.g. WT1, LMX1B, PAX2), (5) mitochondrial function (e.g. COQ2, COQ6, PDSS2), (6) cell/extracellular matrix (e.g. COL4A3/4/5, ITGB4, LAMB2), (7) cell signaling (e.g. PLCE1, TRPC6), (8) nuclear pore complex proteins (e.g. NUP93, NUP203, NUP107), (9) cell membrane-associated proteins (e.g. PTPRO, EMP2) and (10) lysosome (e.g. SCARB2).

Inherited podocytopathy can present in infancy, childhood or adulthood [4]. Autosomal dominant (AD)-FSGS is characterized by later onset of disease and many patients exhibit proteinuria without nephrotic syndrome [5–7]. Recently, a novel heterozygous missense mutation c.T1421G (p.L474R) of the PODXL gene was identified in a Caucasian AD-FSGS pedigree by whole exome sequencing (WES) [8]. PODXL encodes podocalyxin, which plays an important role in the maintenance of structural integrity through the negative charge of the podocalyxin extracellular domain and by forming the podocalyxin/ezrin/actin complex required for the normal function of podocytes [9]. However, functional experiments showed that the p.L474R PODXL mutant did not alter PODXL protein stability, cell surface expression and subcellular localization, as well as binding to ezrin. Besides, this mutation of PODXL was not found 176 probands of AD-FSGS pedigrees. These results make the inference about a causal role of the L474R mutation in disease difficult. In the present study, we report heterozygous nonsense PODXL mutations in two unrelated pedigrees: c.C976T (p. Arg326X) in a Chinese pedigree that was associated with proteinuria and renal insufficiency, and c.C1133G (p. Ser378X) in a British–Indian AD-FSGS pedigree. We also provide evidence with in vitro study showing that the heterozygous nonsense PODXL mutations may be causative in AD-FSGS.

Methods

Subjects

Patients from the two pedigrees were enrolled in the present study to investigate the causative genes of the glomerular disease. The present study was approved by the Ethical Committee of Xin Hua Hospital affiliated to Shanghai Jiao Tong University School of Medicine (ref XHEC-D-2014-002) and West London Research Committee (ref 06-Q0406), respectively. Blood and urine samples from all subjects were collected with informed and written consent (including consent to publish) in accordance with the Declaration of Helsinki.

Whole exome sequencing and Sanger sequencing validation

DNA was extracted from blood samples using QIAamp DNA midi kit (Qiagen). Whole exome captures were performed on the probands of the two pedigrees using SureSelect Human All Exon 50 Mb Kit (Agilent). Paired-end sequencing (76 bp) was performed using the Genome Analyzer IIx (Illumina). Reads were aligned to the reference genome (GRCh37) using Novoalign (Novocraft, 2010). Variants were called using The Genome Analysis Toolkit (GATK) and annotated with ANNOVAR. To identify candidate variants, we adopted a filtering process to focus on nonsynonymous novel deleterious variants using a combination of filtering against SNP databases (dbSNP and the 1000 Genomes Project), in-house exome data and prioritized according to genes association with renal phenotypes such as renal impairment, proteinuria, microscopic hematuria and abnormal kidney development after searching publicly available databases including PubMed, the Online Mendelian Inheritance in Man (OMIM) and the Human Gene Mutation Database (HGMD). Mapping and coverage data are summarized in Supplementary Table S1. Candidate variants identified by WES were validated by Sanger sequencing using an ABI PRISM 3730xl Genetic Analyzer (Applied Biosytems). Co-segregation of candidate variants was tested in all family members whose DNA sample was available.

Plasmid construction

Full-length human PODXL cDNA was subcloned into the MSC site (HindIII & BamHI) of the green fluorescence pEGFP-C1 vector to produce a recombinant plasmid, pEGFP-PODXL. Mutant pEGFP-PODXL plasmids with the p.Arg326X or p.Ser378X mutation were constructed using the Quikchange site-directed mutagenesis kit (QuikChange II Mutagenesis Kit, Qiagen). The primer sequences used to introduce the mutations were 5′-TCAACTACCCACTGATACCCCAAAACAC-3′ for p.Arg326X and 5′-GAAATTGATCTGACTGATATGCCGAGC-3’ for p.Ser378X. The resulting two mutant plasmids pEGFP-PODXL-Arg326X and pEGFP-PODXL-Ser378X were sequenced to verify the incorporation of the required base changes.

Cell culture and transfection

Human embryonic kidney (HEK) 293T cells were cultured in Dulbecco’s Modified Eagle’s Medium supplemented with 10% fetal bovine serum (FBS), 100 U/ml penicillin and 100 U/ml streptomycin at 37°C and 5% CO2. HEK 293T cells were transfected with the wild-type pEGFP-PODXL plasmid or mutant pEGFP-PODXL plasmids using Lipofectamine 2000 (Invitrogen). The expression of plasmids was examined by a fluorescence microscope 48 h after transfection. To inhibit NMD, wild-type and mutant pEGFP-PODXL plasmids-transfected cells were treated with 20 µg/ml of cycloheximide for 8 h, and total RNA were analyzed by qRT-PCR. HEK 293T cells were sequentially transfected with UPF1 small interfering RNA (siRNA) (5′-GACTCTGGTAATGAGGATTTA -3′) and wild-type pEGFP-PODXL or mutant pEGFP-PODXL plasmids using Lipofectamine 2000 (Invitrogen) for 48 h, and total RNA and proteins were analyzed by qRT-PCR and Western blot analysis.

The conditionally immortalized mouse podocytes (MPC5) were cultured in RPMI-1640 with 10% FBS and supplemented with recombinant mouse interferon-γ at 33°C. After differentiating at 37°C for 10 to 14 days without interferon-γ, MPC5 podocytes were transfected with siRNA targeting PODXL (5′-CCACTACACACAAACCATT-3′) along with a siRNA control (RIBOBIO Biotech.) using Lipofectamine RNAiMAX (Invitrogen). Cells were harvested for various experiments 48–72 h after transfection.

RNA extraction and quantitative RT-PCR assays

Peripheral blood mononuclear cells (PBMCs) were isolated from the whole blood of two affected patients (II-2 and II-3) in the Chinese pedigree, 6 maintenance hemodialysis patients without PODXL mutation and 6 healthy controls without PODXL mutation. RNA isolations from PBMCs, HEK 293T cells and cultured MPC5 podocytes after transfection were performed using RNeasy Mini RNA Isolation Kit (Qiagen). Then, 1 µg of total RNA was reverse-transcribed in a volume of 20 µl using random and oligo dT primers under standard conditions using the PrimeScript RT Kit (TaKaRa). For qRT-PCR assay, we used SYBR Premix Ex Taq (TaKaRa) to assess the mRNA level of PODXL in PBMCs, HEK293 cells and cultured MPC5 podocytes. The comparative ΔΔCT method of relative quantification was used to calculate for differences in gene expression using the software for ABI Prism 7500 sequence detection system (Applied Biosystems). The mRNA levels were normalized to glyceraldehyde-3-phosphate dehydrogenase (GAPDH).

Western blot analysis

Cell lysates from PBMCs, HEK 293T cells and cultured podocytes were prepared in RIPA extraction buffer supplemented with protease inhibitor cocktail. About 20–60 µg protein was separated by 7.5–12% SDS-polyacrylamide gel electrophoresis (SDS-PAGE) and transferred to 0.22 µm PVDF membrane. After 1 h of incubation at room temperature in 5% dry milk powder, the membrane was incubated with the primary antibody against GFP (Cell Signaling Technology, 1:2000), podocalyxin (Proteintech 1:500 for detection of podocalyxin in PBMCs and in HEK 293T cells; Novus 1:1000 for detection of podocalyxin in podocytes), anti-phosphorylated Thr567 of ezrin (Cell Signaling Technology, 1:1000), ezrin (Abcam, 1:1000), UPF1 (AB clonal, 1: 1000), β-actin (Cell Signaling Technology, 1:1000) or GAPDH (Cell Signaling Technology, 1:1000), followed by incubation with the appropriate secondary antibodies. After immunoblotting, an ECL Plus Western Blotting System (GE Health) was used for detection.

RhoA G-LISA assay

For the measurement of RhoA activity, a luminescence-based RhoA activity assay (G-LISA; Cytoskeleton) was used following the manufacturer’s protocol. Briefly, cell lysates were prepared from cultured podocytes in which podocalyxin was knocked down and from control podocytes. The lysates were added to a Rho G-LISA plate coated with Rho-GTP-binding protein, followed by adding anti-RhoA primary antibody (diluted 1:250) and horseradish peroxidase (HRP)-labeled secondary antibody (1:250). After HRP detection reagent was then added, the luminescence signal of RhoA activity was detected using a multi-detection microplate reader (BioTek).

Immunofluorescence

Mouse podocytes growing on type I collagen-coated (Sigma-Aldrich) glass slides were pretreated with different conditions and then fixed in 4% paraformaldehyde for 10 min followed by being permeabilized in 0.2% Triton X-100 for 10 min. Rhodamine-labeled phalloidin (Beyotime) was used to stain F-actin in the podocytes, and the resulting images were examined by an inverted fluorescent microscope (Zeiss). The rhodamine-stained areas of the actin fibers were quantified using ImageJ software (NIH). The mean podocyte actin content per pixel and the total actin content per cell were calculated and expressed as AU.

Wound-healing assays

Mouse podocyte pretreated with different conditions were allowed to grow to confluence on type I collagen-coated (Sigma-Aldrich) culture dishes. Cell monolayers were washed and scratch wounds were applied using a 1000 µl pipet tip. Podocytes were imaged using microscope at time 0 immediately after wound creation. Cells were then returned to growth restrictive conditions for 24 h before final imaging of wound healing.

Transwell migration assay

Transwell cell culture inserts (Corning) were coated with type I collagen (Sigma-Aldrich), rinsed once with PBS and placed in RPMI medium in the lower compartment. About 1 × 104 ml−1 control and PODXL knockdown podocytes were seeded in the inserts and then allowed to migrate for 24 h while being incubated at 37°C. Nonmigratory cells were removed from the upper surface of the membrane and the migrated cells were fixed with 4% paraformaldehyde and stained with hematoxylin. The number of migrating cells was counted using phase contrast microscopy (Leica).

Statistical analysis

Continuous variables that were normally distributed were expressed as the mean ± SD. Group differences were assessed by ANOVA with post-test correction (Bonferroni–Holm) or Student’s t-test, as appropriate. A P value <0.05 was considered statistically significant.

Results

Case study of the Chinese pedigree with proteinuria and renal impairment

Figure 1A presents the pedigree of Chinese Han origin with proteinuria and renal insufficiency segregated as a dominant trait. The proband (III-1) was a 39-year-old male who first developed proteinuria (urinary albumin: creatinine ratio, ACR 100 mg/mmol) at the age of 25 with intact renal function, normal serum albumin level and normal blood pressure. He was found to have a declined renal function and then underwent renal biopsy at the age of 39. He received treatment with benazepril but his proteinuria persisted. Renal ultrasound examination revealed reduced renal cortical thickness. Serum creatinine (Scr) was 350 µmol/l. Two family members of III-1 (II-2 and II-3) had also developed proteinuria at the age of their 20s and 30s but received no treatment. Both of them progressed to end-stage renal disease at the age of 55 (II-3) and 72 (II-2), respectively. The III-2 (II-3’s son) was 33 years old and had mild proteinuria (urinary ACR of 30 to 50 mg/mmol) and intact renal function at the time when he was recruited in the present study. The II-2 and III-1 had mild microscopic hematuria. III-2 declined a renal biopsy. A summary of the renal phenotypes of the affected patients in the pedigree is presented in Table 1.

Family pedigrees, sequencing results and decreased PODXL mRNA and protein levels in peripheral blood from the affected individuals with PODXL p. Arg326X mutation

Figure 1
Family pedigrees, sequencing results and decreased PODXL mRNA and protein levels in peripheral blood from the affected individuals with PODXL p. Arg326X mutation

(A) The pedigree of Chinese Han origin with proteinuria and renal insufficiency segregated as a dominant trait. (B) Sanger sequencing of exon 4 of PODXL demonstrated a heterozygous C-to-T transition, which is predicted to result in a p.Arg326X mutation in the proband patient (III-1). (C) The bar graph showing the relative PODXL mRNA abundances as determined by quantitative PCR in PBMCs of six unaffected healthy controls (NC: 1.0 ± 0.25), six maintenance hemodialysis (MHD: 1.09 ± 0.13) patients without nonsense PODXL mutations and two affected patients (II-2: 0.30 ± 0.02; II-3: 0.55 ± 0.03). GAPDH (glyceraldehyde-3-phosphate dehydrogenase) was used for normalization. The data represented the mean ± SD of three independent experiments. * significantly different from NC (ANOVA with post-test correction, P<0.05), #significantly different from MHD (ANOVA with post-test correction, P<0.05). (D) Podocalyxin protein levels in PBMCs of two representative unaffected healthy controls (NC1 and NC2), two representative maintenance hemodialysis patients without nonsense PODXL mutations (MHD1 and MHD2) and two affected patients (II-2 and II-3) carrying the PODXL p.Arg326X mutation. Cell lysates from PBMCs were immunoblotted with anti-podocalyxin (upper panel) and with anti-GAPDH (lower panel) as a control for protein loading. (E) An Indian pedigree of biopsy proven AD-FSGS. (F) Sanger sequencing of exon 6 of PODXL of the proband patient (I-1) in the Indian pedigree demonstrated a heterozygous C-to-G transition, which is predicted to result in a p.Ser378X mutation.

Figure 1
Family pedigrees, sequencing results and decreased PODXL mRNA and protein levels in peripheral blood from the affected individuals with PODXL p. Arg326X mutation

(A) The pedigree of Chinese Han origin with proteinuria and renal insufficiency segregated as a dominant trait. (B) Sanger sequencing of exon 4 of PODXL demonstrated a heterozygous C-to-T transition, which is predicted to result in a p.Arg326X mutation in the proband patient (III-1). (C) The bar graph showing the relative PODXL mRNA abundances as determined by quantitative PCR in PBMCs of six unaffected healthy controls (NC: 1.0 ± 0.25), six maintenance hemodialysis (MHD: 1.09 ± 0.13) patients without nonsense PODXL mutations and two affected patients (II-2: 0.30 ± 0.02; II-3: 0.55 ± 0.03). GAPDH (glyceraldehyde-3-phosphate dehydrogenase) was used for normalization. The data represented the mean ± SD of three independent experiments. * significantly different from NC (ANOVA with post-test correction, P<0.05), #significantly different from MHD (ANOVA with post-test correction, P<0.05). (D) Podocalyxin protein levels in PBMCs of two representative unaffected healthy controls (NC1 and NC2), two representative maintenance hemodialysis patients without nonsense PODXL mutations (MHD1 and MHD2) and two affected patients (II-2 and II-3) carrying the PODXL p.Arg326X mutation. Cell lysates from PBMCs were immunoblotted with anti-podocalyxin (upper panel) and with anti-GAPDH (lower panel) as a control for protein loading. (E) An Indian pedigree of biopsy proven AD-FSGS. (F) Sanger sequencing of exon 6 of PODXL of the proband patient (I-1) in the Indian pedigree demonstrated a heterozygous C-to-G transition, which is predicted to result in a p.Ser378X mutation.

Table 1
Clinical characteristics of patients with PODXL nonsense mutations
Characteristics Chinese pedigree with p. Arg294X Indian pedigree p. Ser346X 
 II-2 (Male) II-3 (Male) III-1 (Male) III-2 (Male) I-1 (Male) II-1 (Male) II-3 (Female) 
 Onset Last follow-up Onset Last follow-up Onset Last follow- up Onset Last follo-up Onset Last follow- up Onset Last follow- up Onset Last follow- up 
Age (y) 30 72 22 69 25 39 33 35 56 62 20 26 26 31 
Urine ACR (mg/mmol) 30–100 50 <30 Anuria 100 150 30–50 50–100 313 351 308 118 101 80 
Scr (μmol/L) NR ESRD NR ESRD NR 350 NR NR 182 208 78 69 52 46 
Hematuria (/HP) 2–3 3–5 5–8 Anuria 0–2 1–3 
Hypertension No Yes No Yes No Yes No No No No No No No No 
Renal biopsy findings N/A N/A N/A N/A N/A N/A N/A N/A FSGS N/A FSGS N/A FSGS N/A 
Characteristics Chinese pedigree with p. Arg294X Indian pedigree p. Ser346X 
 II-2 (Male) II-3 (Male) III-1 (Male) III-2 (Male) I-1 (Male) II-1 (Male) II-3 (Female) 
 Onset Last follow-up Onset Last follow-up Onset Last follow- up Onset Last follo-up Onset Last follow- up Onset Last follow- up Onset Last follow- up 
Age (y) 30 72 22 69 25 39 33 35 56 62 20 26 26 31 
Urine ACR (mg/mmol) 30–100 50 <30 Anuria 100 150 30–50 50–100 313 351 308 118 101 80 
Scr (μmol/L) NR ESRD NR ESRD NR 350 NR NR 182 208 78 69 52 46 
Hematuria (/HP) 2–3 3–5 5–8 Anuria 0–2 1–3 
Hypertension No Yes No Yes No Yes No No No No No No No No 
Renal biopsy findings N/A N/A N/A N/A N/A N/A N/A N/A FSGS N/A FSGS N/A FSGS N/A 

Abbreviations: ACR, albumin:creatinine ratio (normal range <15 mg/mmol); ESRD, end-stage renal disease; FSGS, focal segmental glomerulosclerosis; N/A, not applicable; NR, within normal range; Scr, serum creatinine (normal range 35–97μmol/l).

WES was performed on III-1 and no abnormality in the genes known to be associated with FSGS was identified except a PODXL heterozygous nonsense variant of c.C976T (p. Arg326X) in exon 4. The PODXL variant (rs367825197) is extremely rare. It is not found in the 1000 genomes database and has an allele frequency of only 0.0001 in the NHLBI Exome Sequencing Project. Sanger sequencing (Figure 1B) confirmed this PODXL variant and its co-segregation with disease.

Because the patients’ kidney tissues were unavailable, we examined PODXL mRNA and protein expression levels in PBMCs isolated from available peripheral blood samples of the two affected patients (II-2 and II-3) currently undergoing maintenance hemodialysis, as podocalyxin were reported to be expressed on hematopoietic cells in PBMCs [10]. Using CD34+CD45 as the marker of hematopoietic cells [11], flow cytometry showed that the percentage of CD34+CD45 cells were not significantly altered in PBMCs of II-2 and II-3 compared with those in the PBMCs of healthy controls (n=6) and maintenance hemodialysis patients (n=6) without PODXL mutation (Supplementary Table S2 and Supplementary Figure S1). Real-time PCR revealed a considerable reduction in PODXL mRNA expression in PBMCs of II-2 and II-3 compared with that of the healthy and hemodialysis controls (Figure 1C). Western blot analysis revealed decreased level of the PODXL core protein band (∼55 kDa) in II-2 and II-3 (Figure 1D) compared with that in the healthy and hemodialysis controls.

Case study of the autosomal dominant FSGS pedigree of Indian origin

After the identification of PODXL p. Arg326X heterozygous mutation in the Chinese pedigree, we reviewed WES data of FSGS pedigrees in the UCL Centre for Nephrology and identified a novel PODXL heterozygous nonsense mutation (c. C1133G; p. Ser378X) in exon 6 in the proband (I-1) of an AD-FSGS Indian pedigree (Figure 1E). Sanger sequencing (Figure 1F) confirmed this PODXL variant and its co-segregation with disease in the pedigree. Moreover, this variant was not present in healthy relatives or unrelated healthy controls. PBMC samples were unavailable from the Indian pedigree. A summary of the renal phenotypes of the affected patients in the Indian pedigree is presented in Table 1.

PODXL mutations (p.Arg326X and p.Ser378X) resulting in podocalyxin deficiency by triggering nonsense-mediated decay

To further investigate the effects of PODXL nonsense mutations (p. Arg326X and p. Ser378X), plasmids expressing mutant pEGFP-PODXL plasmids were transfected into HEK 293T cells at the same dose. Forty-eight hours after transfection, cells transfected with pEGFP-PODXL-Arg326X or pEGFP-PODXL-Ser378X both revealed much weaker fluorescence signals than did those transfected with wild-type plasmid (Figure 2A). Under the same experimental settings, immunoblot analysis using antibodies against GFP showed that compared to those transfected with the wild-type plasmid expressing PODXL protein (∼160 kDa), pEGFP-PODXL protein levels were greatly decreased in mutant-plasmid-transfected cell lysates (∼75 kDa for Arg326X; ∼100kDa for Ser378X) (Figure 2B). The mRNAs transcribed from the mutant plasmids were both significantly decreased (Figure 2C) and were rescued after the treatment with NMD inhibitor cycloheximide (Figure 2D). To confirm whether mutant mRNA is degraded via NMD, we used RNA-silencing to inhibit the core NMD factor UPF1 to suppress the NMD pathway and the result showed that in cells transfected with the mutant pEGFP-PODXL plasmids, down-regulation of UPF1 by co-transfected with UPF1-siRNA significantly increased mutant pEGFP-PODXL mRNAs (Figure 2E) and protein levels (Figure 2F).

Mutant PODXL mRNA (p. Arg326X and p. Ser378X) are degraded via NMD

Figure 2
Mutant PODXL mRNA (p. Arg326X and p. Ser378X) are degraded via NMD

(A) HEK 293T cells transfected with wild-type pEGFP-PODXL or mutant plasmids (pEGFP-PODXL-Arg326X; pEGFP-PODXL-Ser378X) under the same conditions showed that the mutant constructs exhibited lower fluorescence yield; scale bar = 100 µm (40× magnification). (B) Immunoblot analysis using antibodies against GFP showed EGFP-fusion products (∼75 kDa for Arg326X; ∼100 kDa for Ser378X) with considerably decreased protein levels in mutant-construct transfected HEK 293T cells lysate than the EGFP-fusion product (∼160 kDa) in the cells transfected with the wild-type pEGFP-PODXL plasmid. (C) pEGFP-PODXL mRNA expression in HEK 293T cells transfected with mutant plasmids (pEGFP-PODXL-Arg326X or pEGFP-PODXL-Ser378X) was significantly decreased compared with that in cells transfected with the wild-type pEGFP-PODXL plasmid (WT: 1.0 ± 0.05; Arg326X: 0.11 ± 0.006; Ser378X: 0.11 ± 0.007). *significantly different from the wild-type plasmid (ANOVA with post-test correction, P<0.05). (D) Treatment of cells with cycloheximide (CHX) significantly increased pEGFP-PODXL mRNAs transcribed from the two mutant plasmids (Arg326X without CHX: 0.05 ± 0.008, Arg326X with CHX: 0.32 ± 0.02; Ser378X without CHX: 0.07 ± 0.007, Ser378X without CHX: 0.35 ± 0.01), but had no effect on the pEGFP-PODXL mRNA level transcribed from the wild-type plasmid (WT without CHX: 1.0 ± 0.03; WT with CHX: 1.02 ± 0.05). *significantly different from those without cycloheximide treatment (Student’s t-test, P<0.05). (E) Inhibition of UPF1 in HEK 293T cells by RNA interference resulted in significantly increased mRNA of mutant pEGFP-PODXL harboring Arg326X or Ser378X mutation (Arg326X with si-ctrl: 0.1 ± 0.009, Arg326X with si-UPF1: 0.27 ± 0.02; Ser378X with si-ctrl: 0.08 ± 0.02, Ser378X with si-UPF1: 0.27±0.02), whereas UPF1 siRNA had no effect on pEGFP-PODXL mRNA levels (WT with si-ctrl: 1.0 ± 0.03, WT with si-UPF1: 0.91 ± 0.04). *significantly different from those with si-ctrl treatment (Student’s t-test, P<0.05). (F) The protein levels of UPF1 and wild-type and mutant pEGFP-PODXL detected by Western blot analysis after HEK 293T cells were sequentially transfected with si-UPF1 or si-ctrl and wild-type or mutant pEGFP-PODXL plasmids.

Figure 2
Mutant PODXL mRNA (p. Arg326X and p. Ser378X) are degraded via NMD

(A) HEK 293T cells transfected with wild-type pEGFP-PODXL or mutant plasmids (pEGFP-PODXL-Arg326X; pEGFP-PODXL-Ser378X) under the same conditions showed that the mutant constructs exhibited lower fluorescence yield; scale bar = 100 µm (40× magnification). (B) Immunoblot analysis using antibodies against GFP showed EGFP-fusion products (∼75 kDa for Arg326X; ∼100 kDa for Ser378X) with considerably decreased protein levels in mutant-construct transfected HEK 293T cells lysate than the EGFP-fusion product (∼160 kDa) in the cells transfected with the wild-type pEGFP-PODXL plasmid. (C) pEGFP-PODXL mRNA expression in HEK 293T cells transfected with mutant plasmids (pEGFP-PODXL-Arg326X or pEGFP-PODXL-Ser378X) was significantly decreased compared with that in cells transfected with the wild-type pEGFP-PODXL plasmid (WT: 1.0 ± 0.05; Arg326X: 0.11 ± 0.006; Ser378X: 0.11 ± 0.007). *significantly different from the wild-type plasmid (ANOVA with post-test correction, P<0.05). (D) Treatment of cells with cycloheximide (CHX) significantly increased pEGFP-PODXL mRNAs transcribed from the two mutant plasmids (Arg326X without CHX: 0.05 ± 0.008, Arg326X with CHX: 0.32 ± 0.02; Ser378X without CHX: 0.07 ± 0.007, Ser378X without CHX: 0.35 ± 0.01), but had no effect on the pEGFP-PODXL mRNA level transcribed from the wild-type plasmid (WT without CHX: 1.0 ± 0.03; WT with CHX: 1.02 ± 0.05). *significantly different from those without cycloheximide treatment (Student’s t-test, P<0.05). (E) Inhibition of UPF1 in HEK 293T cells by RNA interference resulted in significantly increased mRNA of mutant pEGFP-PODXL harboring Arg326X or Ser378X mutation (Arg326X with si-ctrl: 0.1 ± 0.009, Arg326X with si-UPF1: 0.27 ± 0.02; Ser378X with si-ctrl: 0.08 ± 0.02, Ser378X with si-UPF1: 0.27±0.02), whereas UPF1 siRNA had no effect on pEGFP-PODXL mRNA levels (WT with si-ctrl: 1.0 ± 0.03, WT with si-UPF1: 0.91 ± 0.04). *significantly different from those with si-ctrl treatment (Student’s t-test, P<0.05). (F) The protein levels of UPF1 and wild-type and mutant pEGFP-PODXL detected by Western blot analysis after HEK 293T cells were sequentially transfected with si-UPF1 or si-ctrl and wild-type or mutant pEGFP-PODXL plasmids.

Podocalyxin down-regulation leads to decreased activation of RhoA and ezrin, podocyte migration and F-actin stress fibers formation

Based on the observation that the two nonsense PODXL mutations are targeted by NMD and lead to podocalyxin deficiency, we performed PODXL knockdown to down-regulate podocalyxin expression in mouse cultured podocytes in order to investigate the molecular and cellular consequences of PODXL insufficiency. Knockdown of podocalyxin by siRNA in the podocytes led to decreased activity of ezrin (Figure 3A,B) and RhoA (Figure 3C). In the wound-healing assay and transwell migration assay (Figure 4A–D), we found that podocytes showed decreased migration when podocalyxin was down-regulated. Knockdown of podocalyxin also led to significantly reduced F-actin stress fibers in the podocytes (Figure 5A,B).

Effects of PODXL knockdown on activities of RhoA and ezrin in cultured podocyte

Figure 3
Effects of PODXL knockdown on activities of RhoA and ezrin in cultured podocyte

(A) PODXL siRNA transfection led to significant reduction of phosphorylated ezrin (p-ezrin) in the podocytes compared with controls. β-Actin was included as a loading control. (B) Bar graph represented relative p-ezrin protein levels normalized against β-actin and data represented the mean ± SD of three independent experiments (Mock: 1; NC-siRNA: 0.93 ± 0.04; PODXL-siRNA: 0.31 ± 0.03). *significantly different from mock transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05). (C) PODXL siRNA induced significant reduction of activated RhoA and the bar graph showed the relative RhoA G-LISA values in mock vector transfection group (0.43 ± 0.03), NC-siRNA transfection group (0.42 ± 0.03) and PODXL-siRNA group (0.16 ± 0.02). *significantly different from mock transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05).

Figure 3
Effects of PODXL knockdown on activities of RhoA and ezrin in cultured podocyte

(A) PODXL siRNA transfection led to significant reduction of phosphorylated ezrin (p-ezrin) in the podocytes compared with controls. β-Actin was included as a loading control. (B) Bar graph represented relative p-ezrin protein levels normalized against β-actin and data represented the mean ± SD of three independent experiments (Mock: 1; NC-siRNA: 0.93 ± 0.04; PODXL-siRNA: 0.31 ± 0.03). *significantly different from mock transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05). (C) PODXL siRNA induced significant reduction of activated RhoA and the bar graph showed the relative RhoA G-LISA values in mock vector transfection group (0.43 ± 0.03), NC-siRNA transfection group (0.42 ± 0.03) and PODXL-siRNA group (0.16 ± 0.02). *significantly different from mock transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05).

Effects of PODXL knockdown on podocyte mobility in cultured podocytes

Figure 4
Effects of PODXL knockdown on podocyte mobility in cultured podocytes

The wound-healing assay (A) and the transwell migration assay (B) were performed. PODXL siRNA treatment significantly slowed podocyte migration after 24 h compared with the podocytes in the control groups. Magnification: wound-healing assay ×10; transwell migration assay ×40. Bar graphs showed wound closure area in the scratches in the groups of cells that were treated differently after 24 h (Mock: 0.74 ± 0.02; NC-siRNA: 0.73 ± 0.03; PODXL-siRNA: 0.22 ± 0.02) (C) and the numbers of migrating podocytes in the groups treated differently (Mock: 87.67 ± 4.51; NC-siRNA: 85.67 ± 7.02; PODXL-siRNA: 20.33 ± 4.16) (D). Data represent the mean ± SD of three independent experiments. *significantly different from mock transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05).

Figure 4
Effects of PODXL knockdown on podocyte mobility in cultured podocytes

The wound-healing assay (A) and the transwell migration assay (B) were performed. PODXL siRNA treatment significantly slowed podocyte migration after 24 h compared with the podocytes in the control groups. Magnification: wound-healing assay ×10; transwell migration assay ×40. Bar graphs showed wound closure area in the scratches in the groups of cells that were treated differently after 24 h (Mock: 0.74 ± 0.02; NC-siRNA: 0.73 ± 0.03; PODXL-siRNA: 0.22 ± 0.02) (C) and the numbers of migrating podocytes in the groups treated differently (Mock: 87.67 ± 4.51; NC-siRNA: 85.67 ± 7.02; PODXL-siRNA: 20.33 ± 4.16) (D). Data represent the mean ± SD of three independent experiments. *significantly different from mock transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05).

Effects of PODXL knockdown on F-actin stress fibers formation in cultured podocytes

Figure 5
Effects of PODXL knockdown on F-actin stress fibers formation in cultured podocytes

(A) Representative results of phalloidin staining showing that PODXL siRNA induced significant reduction of F-actin fibers in cultured podocytes compared with controls; magnification: 40×. (B) Quantification of the results in (A) (n=3) (Mock: 3117 ± 246.8; NC-siRNA: 3027 ± 293.3; PODXL-siRNA: 1371 ± 140.8). *significantly different from mock vector transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05).

Figure 5
Effects of PODXL knockdown on F-actin stress fibers formation in cultured podocytes

(A) Representative results of phalloidin staining showing that PODXL siRNA induced significant reduction of F-actin fibers in cultured podocytes compared with controls; magnification: 40×. (B) Quantification of the results in (A) (n=3) (Mock: 3117 ± 246.8; NC-siRNA: 3027 ± 293.3; PODXL-siRNA: 1371 ± 140.8). *significantly different from mock vector transfection group (ANOVA with post-test correction, P<0.05), #significantly different from NC-siRNA transfection group (ANOVA with post-test correction, P<0.05).

Discussion

In the present study, we report the identification of the heterozygous nonsense PODXL mutations (Arg326X) that may underlie the proteinuria and renal insufficiency clinically suggestive of AD-FSGS in the Chinese pedigree. In addition, we identified another novel heterozygous PODXL nonsense mutation (Ser378X) in an unrelated British–Indian pedigree of AD-FSGS, further supporting that nonsense mutations in the PODXL gene contribute to FSGS. We demonstrated that PODXL mRNA harboring the nonsense PODXL mutations identified in the two pedigrees was degraded through the NMD pathway, leading to reduced production of PODXL protein. The loss-of-function nature of the two nonsense mutations was supported by the observation that podocalyxin knockdown led to decreased activity of RhoA and ezrin, decreased migration and altered cytoskeletal structure of podocytes.

Podocalyxin is a negatively charged, O-glycosylated and sialylated type I transmembrane protein [12] that is expressed primarily in podocytes and peritubular capillaries but absent in tubules [13] in adult human kidney. Due to the essentiality of PODXL for the formation of the filtration barrier and the signals that mediate cytoskeletal organization of the foot processes, alterations in podocalyxin expression and function can induce podocyte injury and podocytopathies. Podocalyxin-deficient (PODXL−/−) mice died within 24 h of birth due to anuric renal failure that was caused by the complete loss of foot processes (FP) and slit diaphragm (SD) [14]. During the preparation of this manuscript, a study came out and reported compound heterozygous PODXL mutations that led to complete loss of PODXL protein expression in a patient [15]. This patient presented with congenital nephrotic syndrome, omphalocele and microcoria, phenocopying the PODXL−/− mice. In human pluripotent stem cell-derived podocyte-like cells, knockout of podocalyxin caused defects in the organization of junctions and the spacing of adjacent podocyte-like cells [13].

Compared with the PODXL-null models, partial suppression of podocalyxin expression by morpholino antisense oligonucleotides resulted in poorly developed fine foot processes and hypoplasia of pronephric glomerulus in zebrafish morphants [16]. The PODXL+/− mice with loss of one PODXL allele also displayed abnormal omphalocele during development but had no overt abnormality of FP and SD by electron microscopy [14]. Human diabetic nephropathy [17] and puromucin aminonucleoside-induced nephrotic mice [18] also exhibit reduced expression or sialylation of podocalyxin. In the study reporting the patient with congenital nephrotic syndrome and compound heterozygous PODXL mutations, the PODXL construct with nonsense p.Try341X mutation failed in expressing podocalyxin in contrast with wild-type PODXL [15], suggesting that nonsense PODXL mutations may induce NMD [19]. The above results are consistent with our in vitro study results that cells transfected with mutant pEGFP-PODXL-Arg326X or pEGFP-PODXL-Ser378X constructs resulted in considerably reduction in mutant pEGFP-PODXL mRNAs and PODXL protein levels via NMD in cell lysates. Previous study found that podocalyxin could be released to the extracellular space as intact protein or a soluble cleaved fragment of ectodomain [20–21]. In our in vitro study, full size PODXL protein (∼160 kDa) was detected in the extracellular medium of HEK 293T cells transfected with wild-type pEGFP-PODXL plasmid by Western blot analysis, whereas no PODXL protein was found in the extracellular medium of HEK 293T cells transfected with mutant pEGFP-PODXL plasmids (Supplementary Figure S2), probably due to deficient production of mutant PODXL protein via NMD.

Mechanistically, the basal activity of RhoA has been shown to be required for podocyte integrity [22], likely through its binding to ezrin to maintain the proper distribution of actin filaments and thus the architecture of the foot process [23]. We demonstrated the loss-of-function nature of PODXL nonsense mutations by knockdown of podocalyxin in cultured podocytes. We found that PODXL down-regulation by siRNA led to significant decrease of RhoA activity, down-regulation of ezrin phosphorylation, decrease of podocyte migration and reduction of F-actin stress fibers in the podocytes. These effects of nonsense PODXL mutations are different from that of the missense PODXL mutation found previously in a Caucasian AD-FSGS pedigree in which podocalyxin expression was not altered but podocalyxin dimerization was facilitated [8].

Similar to the Caucasian AD-FSGS pedigree with heterozygous missense PODXL mutation, disease severities among the patients in the two pedigrees of the present study were variable in terms of proteinuria and renal function. For example, the patient of III-2 in the Chinese pedigree had subtle clinical manifestation with only microproteinuria but normal renal function, whereas the renal function of his cousin (III-1) of similar age was already impaired with overt proteinuria. The intrafamilial clinical heterogeneity, e.g. proteinuria onset ranging from 10 to 54 years, might be due to other unknown modifying genes or environmental factors that affect the natural history of disease in different affected family members. Consistently, the mother of the patient with congenital nephrotic syndrome and compound heterozygous PODXL mutations (including the p. Try341X) [15] was apparently healthy although information on proteinuria was not provided.

In summary, we have identified pathogenic nonsense PODXL mutations in two unrelated pedigrees with clinical-suspected or biopsy-proven AD-FSGS and shown their adverse effects on cultured podocytes, which further implicated PODXL mutation in the development of FSGS.

Clinical perspectives

  • A heterozygous missense mutation in the PODXL gene encoding podocalyxin, a major constituent of the glycocalyx of podocytes, was recently identified in AD-FSGS pedigree. However, the causal role of PODXL mutation in FSGS remains to be verified. Our study for the first time reported two heterozygous nonsense PODXL mutations (Arg326X and Ser378X ) in two unrelated pedigrees with clinical-suspected or biopsy-proven AD-FSGS.

  • The nonsense PODXL mutations (Arg326X or Ser378X) were loss-of-function mutations, which led to decreased activity of RhoA and ezrin, decreased migration and altered cytoskeletal structure of podocytes.

  • The results of our study provided new data implicating PODXL mutations in the development of FSGS.

Acknowledgements

The authors express their gratitude to the patients and relatives who participated in this project.

Competing Interests

The authors declare that there are no competing interests associated with the manuscript.

Funding

F.-J.L. is funded by the National Natural Science foundation of China [81500507]; the Clinical Research Study Grant of Xin Hua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine [17CSY03]; the Young Investigator Funding provided by Xin Hua Hospital affiliated to Shanghai Jiao Tong University School of Medicine [2016-2018]; D.P.G. is funded by The Rosetrees and St Peters Trusts. LY is funded by the National Natural Science foundation of China [81800301]; and H.-Q.R. is funded by the Jiangsu Provincial Natural Science Foundation [BK20181150].

Author Contribution

F.-J.L., D.P.G., G.-R.J. and H.-Q.R. were responsible for the concept and design of the study. F.-J.L., D.P.G., F.B. and G.J. were responsible for diagnosis, recruitment, clinical characterization and blood collection for all subjects studied. F.-J.L. and D.P.G. analyzed and interpreted the data. L.Y., X.-Q.H. and H.-Q.R. performed the functional experiments. F.-J.L., H.-Q.R. and D.P.G. wrote the manuscript.

Abbreviations

     
  • AD

    autosomal dominant

  •  
  • FP

    foot processes

  •  
  • FSGS

    focal segmental glomerulosclerosis

  •  
  • GAPDH

    glyceraldehyde-3-phosphate dehydrogenase

  •  
  • HRP

    horseradish peroxidase

  •  
  • PBMC

    peripheral blood mononuclear cell

  •  
  • PCR

    polymerase chain reaction

  •  
  • SD

    slit diaphragm

  •  
  • SDS-PAGE

    SDS-polyacrylamide gel electrophoresis

  •  
  • siRNA

    small interfering RNA

  •  
  • WES

    whole exome sequencing

References

References
1.
Bryer
J.S.
and
Susztak
K.
(
2018
)
Screening drugs for kidney disease: targeting the podocyte
.
Cell Chem. Biol.
25
,
126
127
[PubMed]
2.
Bierzynska
A.
,
Soderquest
K.
and
Koziell
A.
(
2015
)
Genes and podocytes - new insights into mechanisms of podocytopathy
.
Front. Endocrinol.
5
,
226
3.
Schlondorff
J.
(
2015
)
How many Achilles’ heels does a podocyte have? An update on podocyte biology
Nephrol. Dial. Transplant.
30
,
1091
1097
[PubMed]
4.
Pollak
M.R.
(
2014
)
Familial FSGS
.
Adv. Chronic Kidney Dis.
21
,
422
425
[PubMed]
5.
Hao
X.
,
Xie
J.
,
Ma
J.
,
Wang
Z.
,
Zhou
Q.
,
Yang
L.
et al. .
(
2013
)
Increased risk of treatment failure and end-stage renal disease in familial focal segmental glomerular sclerosis
.
Contrib. Nephrol.
181
,
101
108
[PubMed]
6.
Brown
E.J.
,
Schlöndorff
J.S.
,
Becker
D.J.
,
Tsukaguchi
H.
,
Tonna
S.J.
,
Uscinski
A.L.
et al. .
(
2010
)
Mutations in the formin gene INF2 cause focal segmental glomerulosclerosis
.
Nat. Genet.
42
,
72
76
[PubMed]
7.
Barua
M.
,
Stellacci
E.
,
Stella
L.
,
Weins
A.
,
Genovese
G.
,
Muto
V.
et al. .
(
2014
)
Mutations in PAX2 associate with adult-onset FSGS
.
J. Am. Soc. Nephrol.
25
,
1942
1953
[PubMed]
8.
Barua
M.
,
Shieh
E.
,
Schlondorff
J.
,
Genovese
G.
,
Kaplan
B.S.
and
Pollak
M.R.
(
2014
)
Exome sequencing and in vitro studies identified podocalyxin as a candidate gene for focal and segmental glomerulosclerosis
.
Kidney Int.
85
,
124
133
[PubMed]
9.
Schmieder
S.
,
Nagai
M.
,
Orlando
R.A.
,
Takeda
T.
and
Farquhar
M.G.
(
2004
)
Podocalyxin activates RhoA and induces actin reorganization through NHERF1 and Ezrin in MDCK cells
.
J. Am. Soc. Nephrol.
15
,
2289
2298
[PubMed]
10.
Kuhara
M.
,
Yoshino
T.
,
Shiokawa
M.
,
Okabe
T.
,
Mizoguchi
S.
,
Yabuhara
A.
et al. .
(
2009
)
Magnetic separation of huan podocalxin-like protein 1 (hPCLP1)-positively cells from peripheral blood and umbilical cord blood using anti-hPCLP1 monoclonal antibody and protein A expressed on bacterial magnetic particles
.
Cell Struct. Funct.
34
,
23
30
[PubMed]
11.
Kerosuo
L.
,
Juvonen
E.
,
Alitalo
R.
,
Gylling
M.
,
Kerjaschki
D.
and
Miettinen
A.
(
2004
)
Podocalyxin in human haematopoietic cell
.
Br. J. Haematol.
124
,
809
818
[PubMed]
12.
Nielsen
J.S.
and
McNagny
K.M.
(
2009
)
The role of podocalyxin in health and disease
.
J. Am. Soc. Nephrol.
20
,
1669
1676
[PubMed]
13.
Freedman
B.S.
,
Brooks
C.R.
,
Lam
A.Q.
,
Fu
H.
,
Morizane
R.
,
Agrawal
V.
et al. .
(
2015
)
Modelling kidney disease with CRISPR-mutant kidney organoids derived from human pluripotent epiblast spheroids
.
Nat. Commun.
6
,
8715
[PubMed]
14.
Doyonnas
R.
,
Kershaw
D.B.
,
Duhme
C.
,
Merkens
H.
,
Chelliah
S.
,
Graf
T.
et al. .
(
2001
)
Anuria, omphalocele, and perinatal lethality in mice lacking the CD34-related protein podocalyxin
.
J. Exp. Med.
194
,
13
27
[PubMed]
15.
Kang
H.G.
,
Lee
M.
,
Lee
K.B.
,
Hughes
M.
,
Kwon
B.S.
,
Lee
S.
et al. .
(
2017
)
Loss of podocalyxin causes a novel syndromic type of congenital nephrotic syndrome
.
Exp. Mol. Med.
49
,
e414
[PubMed]
16.
Ichimura
K.
,
Powell
R.
,
Nakamura
T.
,
Kurihara
H.
,
Sakai
T.
and
Obara
T.
(
2013
)
Podocalyxin regulates pronephric glomerular development in zebrafish
.
Physiol. Rep.
1
,
e00074
[PubMed]
17.
Koop
K.
,
Eikmans
M.
,
Baelde
H.J.
,
Kawachi
H.
,
De Heer
E.
,
Paul
L.C.
et al. .
(
2003
)
Expression of podocyte-associated molecules in acquired human kidney disease
.
J. Am. Soc. Nephrol.
14
,
2063
2071
[PubMed]
18.
Kerjaschki
D.
,
Vernillo
A.T.
and
Farquhar
M.G.
(
1985
)
Reduced sialylation of podocalyxin-the major sialoprotein of the rat kidney glomerulus-in aminomucleoside nephrosis
.
Am. J. Pathol.
118
,
343
349
[PubMed]
19.
Fang
Y.
,
Bateman
J.F.
,
Mercer
J.F.
and
Lamandé
S.R.
(
2013
)
Nonsense-mediated mRNA decay of collagen-emerging complexity in RNA surveillance mechanisms
.
J. Cell Sci.
126
,
2551
2560
[PubMed]
20.
Fernández
D.
,
Larrucea
S.
,
Nowakowski
A.
,
Pericacho
M.
,
Parrilla
R.
and
Ayuso
M.S.
(
2011
)
Release of podocalyxin into the extracellular space. Role of metalloproteinases
.
Biochim. Biophys. Acta
1813
,
1504
1510
[PubMed]
21.
Larrucea
S.
,
Butta
N.
,
Arias-Salgado
E.G.
,
Alonso-Martin
S.
,
Ayuso
M.S.
and
Parrilla
R.
(
2008
)
Expression of podocalyxin enhances the adherence, migration, and intercellular communication of cells
.
Exp. Cell. Res.
314
,
2004
2015
[PubMed]
22.
Wang
L.
,
Ellis
M.J.
,
Gomez
J.A.
,
Eisner
W.
,
Fennell
W.
,
Howell
D.N.
et al. .
(
2012
)
Mechanisms of the proteinuria induced by Rho GTPases
.
Kidney Int.
81
,
1075
1085
[PubMed]
23.
Mouawad
F.
,
Tsui
H.
and
Takano
T.
(
2013
)
Role of Rho-GTPases and their regulatory proteins in glomerular podocyte function
.
Can. J. Physiol. Pharmocol.
91
,
773
782

Author notes

*

These authors contributed equally to this work.