The Role of Platelet-Derived Growth Factor in Focal Segmental Glomerulosclerosis

Abstract

Significance Statement

We investigated the role of the profibrotic PDGF in the development and progression of FSGS in a murine model resembling human FSGS. Injured podocytes expressed PDGF-B, inducing parietal epithelial cell activation, proliferation, and a profibrotic switch–driving FSGS. Therapeutic inhibition of PDGF-B significantly reduced proteinuria and FSGS, suggesting that inhibition of the PDGF signaling pathway might be a potential novel treatment for patients with FSGS.

Background

FSGS is the final common pathway to nephron loss in most forms of severe or progressive glomerular injury. Although podocyte injury initiates FSGS, parietal epithelial cells (PECs) are the main effectors. Because PDGF takes part in fibrotic processes, we hypothesized that the ligand PDGF-B and its receptor PDGFR-β participate in the origin and progression of FSGS.

Methods

We challenged Thy1.1 transgenic mice, which express Thy1.1 in the podocytes, with anti-Thy1.1 antibody to study the progression of FSGS. We investigated the role of PDGF in FSGS using challenged Thy1.1 mice, 5/6 nephrectomized mice, Col4^−/− (Alport) mice, patient kidney biopsies, and primary murine PECs, and challenged Thy1.1 mice treated with neutralizing anti–PDGF-B antibody therapy.

Results

The unchallenged Thy1.1 mice developed only mild spontaneous FSGS, whereas challenged mice developed progressive FSGS accompanied by a decline in kidney function. PEC activation, proliferation, and profibrotic phenotypic switch drove the FSGS. During disease, PDGF-B was upregulated in podocytes, whereas PDGFR-β was upregulated in PECs from both mice and patients with FSGS. Short- and long-term treatment with PDGF-B neutralizing antibody improved kidney function and reduced FSGS, PEC proliferation, and profibrotic activation. In vitro, stimulation of primary murine PECs with PDGF-B recapitulated in vivo findings with PEC activation and proliferation, which was inhibited by PDGF-B antibody or imatinib.

Conclusion

PDGF-B–PDGFR-β molecular crosstalk between podocytes and PECs drives glomerulosclerosis and the progression of FSGS.

Both primary and secondary FSGS show similar morphologic appearance, and in pathology diagnostics the term FSGS is used to describe this specific pathology in various kidney disease.^1,2 Only half of FSGS patients is responsive to the current therapies, and both safe and effective therapies are still lacking.^3,4 Podocyte injury was proposed as the first major step in the development of FSGS. However, the parietal epithelial cell (PEC) was suggested as the main effector cell that drives glomerulosclerosis in human FSGS and experimental models.^5,6 PEC activation is thought to be an essential early step in FSGS development. Following their activation, PECs actively deposit extracellular matrix in experimental and human FSGS.^7,8 Although many studies focused on the initiating mechanisms of FSGS, including podocyte injury and PECs activation, the molecular pathways mediating the final step of the fibrotic switch of PECs remain poorly understood.

PDGF receptor β (PDGFR-β) is a tyrosine kinase receptor that after binding of its ligands PDGF-B and PDGF-D induces downstream signaling, triggering cell proliferation, migration, and differentiation.⁹ Adult renal mesenchymal cells express PDGFR-β, and its activation drives their pathologic proliferation and phenotypic switch toward myofibroblasts. PDGF-D is dispensable for normal renal development and physiologic function.¹⁰ However, podocyte-specific overexpression of PDGF-D in mice caused crescentic and mesangioproliferative GN.¹¹ Treatment with a PDGF-B aptamer or PDGF-D monoclonal antibody (CR002) in rat models of mesangioproliferative GN potently reduced glomerular damage and improved renal function.^12,13 However, the role of PDGF in FSGS remained unclear.

Here, we comprehensively analyzed an FSGS model (i.e., Thy1.1 mice),^14–17 focusing on profibrotic PEC phenotype and the regulation and functional role of PDGF in driving the progression of FSGS.

Methods

Study Design and Ethics

Human kidney formalin-fixed, paraffin-embedded biopsy samples were obtained from the archive of the Institute of Pathology and processed anonymously. Samples came from patients with minimal change disease (MCD; n=3) or secondary FSGS (n=4). The study was approved by the local review board (EK 042-17) and was conducted in line with the Declaration of Helsinki.

All animal experiments were performed according to the guidelines for care and use of laboratory animals and were approved by the local authorities and review boards (Landesministerium für Natur-, Umwelt-, und Verbraucherschutz Nordrhein Westfalen). Mice were held in a 12-hour light-dark cycle room with constant temperature and humidity in the animal facility of RWTH Aachen University Hospital. They were free to access standard rodent chow and drinking water. All efforts were made to minimize animal suffering.

In Vivo Mouse Experiments

Thy1.1 Transgenic Mice

To study the spontaneous phenotype of Thy1.1 mice, 35-week-old female (n=6) and male (n=3) mice were analyzed. Female (n=5) and male (n=5) wild-type littermates served as healthy controls.

Thy1.1 FSGS Model

To study FSGS progression the time course after anti-Thy1.1 antibody injection was analyzed. For this, 8- to 12-week-old Thy1.1 transgenic mice on a C57BL/6J background were used. After injection of 1 mg anti-Thy1.1 antibody (19XE5) in 0.1 ml 0.9% saline solution into the tail vein, mice were euthanized, and the kidneys were removed for analysis on day 1, day 3, day 7, day 14, day 21, and day 100. The day 100 group included five male and five female mice; at all other time points, three female and three male mice were analyzed. Thy1.1 mice with an injection of 0.1 ml saline into the tail vein were used as control group (day 0), including two female and two male mice.

5/6 Nephrectomy FSGS Model

Ten- to 12-week-old male mice on an Sv129 background underwent surgical 5/6 nephrectomy (Nx) under ketamine/xylazine anesthesia. The right kidney and two-thirds of the left kidney were removed via two flank incisions in a single-stage surgery (n=5). Mice received analgesia for the first 72 hours after surgery. Three control animals underwent sham surgery. All mice were euthanized after 12 weeks.

A transgenic mouse line with podocyte-specific deletion of Pdgfb was generated by crossbreeding podocin-Cre mice¹⁸ with B6.129P2-Pdgfb^tm2Cbet/J mice (https://www.jax.org/strain/017622). This mouse line did not exhibit a spontaneous kidney phenotype as judged by periodic acid–Schiff (PAS) stain and serum creatinine level (data not shown). Ten- to 12-week-old male mice on a mixed C57BL6-Sv129 background underwent 5/6Nx under ketamine/xylazine anesthesia as described above. Animals were distributed in three groups: Cre-positive and Pdgfb^flox/flox with 5/6Nx (5/6^−/−, n=6), Cre-positive and Pdgfb^+/+ with 5/6Nx (5/6 WT [wild-type], n=5), and Cre-positive and Pdgfb^flox/flox with sham surgery (sham^−/−). All mice were euthanized after 12 weeks.

Col4^−/− (Alport) Mice

Col4^−/− (Alport) mice were bred on an Sv129 background. Three male mice were euthanized at an age of 8 weeks (n=3), and genotypes were double-checked.

Anti–PDGF-B Antibody Treatment for Thy1.1 FSGS Model

To study the role of PDGF-B in FSGS, anti–PDGF-B antibody (MOR8457; Pfizer, New York, NY) (10 mg/kg) was intraperitoneally injected into 8- to 12-week-old Thy1.1 mice after one dose (1 mg in 0.1 ml 0.9% saline) of anti-Thy1.1 antibody. The specificity of MOR8457 and its antifibrotic efficacy in mice was shown previously.^19,20 Mice receiving isotype-matched irrelevant IgG injection (10 mg/kg) served as controls. In order to use all of the mice from breeding, both female and male mice were used for the therapy experiments. Two experiments were performed: experiments of short-term therapy (day 7) were conducted with female mice (11 mice were injected with anti–PDGF-B antibody and nine mice were injected with IgG) and experiments with long-term therapy (day 21) were conducted with male mice (nine mice received anti–PDGF-B antibody and seven mice received IgG).

Assessments

In all of the in vivo experiments, kidney function was evaluated by measurement of serum creatinine, serum urea, urinary creatinine, and urinary protein after a 12- to 16-hour urine collection in metabolic cages. Analyses were performed using the Vitros 350 Chemistry Analyzer (Ortho Clinical Diagnostics, Unterschleißheim, Germany) after calibration with the Performance Verifier I and II. Urine albumin was measured by ELISA Albuwell M according to the manufacturer’s instructions (Ethos Biosciences, Logan Township, NJ).Three male mice did not produce any urine at day 1 for the Thy1.1 FSGS model. Blood pressure was measured by a noninvasive tail-cuff method (CODA; Kent Scientific Corporation, Torrington, CT). After intraperitoneal anesthesia with ketamine (100 mg/kg body weight) and xylazine (10 mg/kg body weight), mice were exsanguinated by cutting the inferior vena cava and perfusion with physiologic saline via the heart left ventricle. Body weight and kidney weight were measured before euthanization and after perfusion, respectively. Both kidneys were removed and immediately processed for histologic or molecular analyses.

Histology, Immunohistochemistry, Immunofluorescence, and Fluorescence In Situ Hybridization

Mice kidney tissues for histopathologic analyses were fixed in methyl Carnoy’s solution and embedded in paraffin. Next 1-µm sections were stained with PAS and counterstained with hematoxylin for overall assessment of kidney damage and for quantification of sclerotic glomeruli. All the glomeruli in the sections were counted. Glomeruli in approximately one-third of the inner cortex were counted to analyze the glomeruli in the juxtamedullary area. All histomorphologic analyses were performed in a blinded manner. In the Thy1.1 time course, dots within the DAPI⁺ area (i.e., nuclear mRNA signals or dots) were analyzed. Nuclear mRNA dots of PDGFB in podocytes and PDGFRB in PECs were manually counted. Nuclear mRNA dots of PDGFB in podocytes and PDGFRB in PECs were manually analyzed. Only podocytes located in the outer layer of glomerular tuft in sections were counted. Ten glomeruli were randomly selected per mouse. Nuclear mRNA dots of PDGFRB in PECs were counted, and five (n=2 biopsies only contained five glomeruli) or eight glomeruli (n=5) were randomly selected per human biopsy. The fused dots were counted as three dots for all of the cases.

Immunohistochemistry, immunofluorescence, and fluorescence in situ hybridization (FISH) were performed as described previously.^21–23 In brief, for FISH the RNAscope Multiplex Fluorescent Reagent Kit v2 (ACDBio, Minneapolis, MN) was used in the experiment. After deparaffinization, the freshly cut formalin-fixed samples were rehydrated. Antigen retrieval was performed in a steamer at 99°C. After incubation with Protease Plus, the Amp1, Amp2, and Amp3 were incubated in sequence. The probes were incubated for 2 hours followed with Opals fluorophores for 30 minutes. All primary and secondary antibodies are listed in Table 1 and probes in Supplemental Table 1.

Table 1.

Primary and secondary antibodies for immunohistochemistry or immunofluorescence staining

Target	Host	Supplier	Catalog No.
α-SMA	Mouse	Dako (Santa Clara, CA)	M0851
CD44	Rat	BD Biosciences (Franklin Lakes, NJ)	553131
Collagen III	Goat	SouthernBiotech (Birmingham, AL)	1330–01
Collagen IV	Goat	SouthernBiotech (Birmingham, AL)	1340–01
F4/80	Rat	Bio-Rad (Hercules, CA)	MCA497G
GAPDH	Mouse	Novus Biologicals (Littleton, CO)	NB300–221
Ki67	Rabbit	Thermo Fisher Scientific (Rockford, IL)	13–5698–82
Ki67	Mouse	Dako (Santa Clara, CA)	M7240
LKIV69	Single chain	Custom made7,33 and kindly provided by T.H. van Kuppevelt, Radboud University Medical Center, Nijmegen, The Netherlands
PDGFR-β	Rabbit	Cell Signaling Technology (Danvers, MA)	3169
PDGFR-β	Rabbit	Abcam (Cambridge, UK)	ab32570
Podocin	Rabbit	Sigma-Aldrich (St. Louis, MO)	P0372
WT-1	Rabbit	Abcam (Cambridge, UK)	ab89901
Alexa 488 anti-rabbit	Goat	Thermo Fisher Scientific (Rockford, IL)	A11008
Alexa 555 anti-rat	Goat	Thermo Fisher Scientific (Rockford, IL)	A21434
Biotinylated anti-goat	Rabbit	Vector (Burlingame, CA)	BA-5000
Biotinylated anti-mouse	Horse	Vector (Burlingame, CA)	BA-2001
Biotinylated anti-mouse IgG2A	Goat	Antibodies online (Aachen, Germany)	ABIN1118481
Biotinylated anti-rabbit	Goat	Vector (Burlingame, CA)	BA-1000
Biotinylated anti-rat	Rabbit	Vector (Burlingame, CA)	BA-4001
Biotinylated anti-rat	Goat	Jackson/Dianova (West Grove, PA)	112–065–167
Biotinylated anti-VSV tag	Rabbit	Abcam (Cambridge, UK)	ab34774
IRD680RD anti-mouse	Donkey	LI-COR (Lincoln, NE)	C70419–08
IRD800CW anti-rat	Goat	LI-COR (Lincoln, NE)	80425–05

α-SMA, α skeletal muscle actin; CD44, cluster of differentiation 44; F4/80, EGF-like module-containing mucin-like hormone receptor-like 1; PDGFR-β, PDGF receptor β; WT-1, Wilms’ tumor suppressor gene 1.

The histologic sections were scanned with a whole-slide scanner (AT2; Leica Biosystems, Wetzlar, Germany). For quantitative glomerular analyses, all of the glomerular cross-sections on each slide were photographed at ×400 magnification with ImageScope (Leica Biosystems) to quantify the stained area of glomerular tuft with ImageJ (National Institutes of Health, Bethesda, MD). Glomeruli with cluster of differentiation 44 (CD44)-positive staining in PECs were counted. CD44 score was evaluated according to the total positive staining in PECs as follows: 0=no staining, 1=0–1/4 of Bowman’s capsule, 2=1/4–1/2 of Bowman’s capsule, 3=1/2–3/4 of Bowman’s capsule, 4=3/4–4/4 of Bowman’s capsule. Glomeruli with α-SMA–, collagen IV-, Ki67-, and PDGFR-β–positive staining in Bowman’s space were counted. For analysis of the tubulointerstitium, 20–30 pictures from the cortex were randomly taken at ×200 magnification with ImageScope, and then the positive area was quantified by ImageJ, or the whole positive areas in the cortex in the section were semiautomatically analyzed by Leica ImageScope software. Immunofluorescence and FISH pictures were taken with digital fluorescence microscopes (BZ-9000; Keyence, Japan or Axio Imager 2 microscope; Zeiss, Jena, Germany).

Transmission Electron Microscopy

The renal cortex of the kidney specimens was cut into 1×1×1-mm pieces, and the tissue, processed for electron microscopy as described before.²⁴ In brief, the tissue was embedded in epon resin, cut to ultrathin sections, stained with uranyl acetate and lead citrate, and observed with a Hitachi HT7800 transmission electron microscope at 100 kV.

In Vitro Cell Culture Experiments

Mouse primary PECs were isolated as previously described.²⁵ PECs were grown in RPMI 1640 medium supplemented with 10% FCS and 1% penicillin/streptomycin. Cells in the ninth to 15th passage were stimulated for 6 or 24 hours in 0.1% FCS growth media with 20 ng/ml PDGF-B (Sigma-Aldrich, Steinheim, Germany). Cell proteins were extracted by applying protein lysis buffer that was mixed with protease and phosphatase inhibitor cocktail (Thermo Fisher Scientific, Rockford, IL). After PECs were stimulated with 20 ng/ml PDGF-B, PDGF-B plus 400 ng/ml anti–PDGF-B antibody (MOR8457; Pfizer ), and 10 µM imatinib (Sigma-Aldrich) for 24 hours in 0.1% FCS growth medium, bromodeoxyuridine incorporation proliferation assay was performed according to the manufacturer’s instructions (Roche, Mannheim, Germany).

RNA Extraction and Analysis

Total RNA was isolated from cell lysate using the RNeasy Mini Kit (Qiagen, Hilden, Germany). RNA purity determination, cDNA synthesis, and RT-PCR were performed as described.¹³ The real-time RT-PCR was conducted using an ABI Prism 7300 sequence detector (Applied Biosystems, Weiterstadt, Germany), and data were normalized to glyceraldehyde 3-phosphate dehydrogenase (GAPDH). Sequences of primers are listed in Supplemental Table 2.

Western Blot

Western blot analyses of lysates from PECs were performed as previously described.²⁶ In brief, after SDS-PAGE, the samples were transferred on a 0.45-µm nitrocellulose membrane. Nonspecific binding sites were blocked with 3% (wt/vol) BSA diluted in Tris-buffered saline solution. Primary and secondary antibodies used are listed in Table 1. GAPDH was used as internal loading control for normalization. The membranes were exposed using the LI-COR fluorescence imaging system (LI-COR Biosciences, Lincoln, NE), and finally results were quantified with ImageJ software.

Statistical Analyses

All statistical analyses were performed with GraphPad Prism 8.3.0 (La Jolla, CA). All data are presented as mean±SD. Two-tailed unpaired t test was applied for the comparison of two groups. One-way ANOVA with Tukey correction was used for the comparison of more than two groups. Statistical significance was defined as P<0.05.

Results

Unchallenged Thy1.1 Transgenic Mice Developed Mild Spontaneous Phenotype

In the analyses of the spontaneous phenotype of Thy1.1 transgenic mice, compared with wild-type mice, no changes in blood count and renal function were detected at the age of 35 weeks (Supplemental Figure 1). The histopathologic examination revealed development of FSGS in the Thy1.1 mice (Supplemental Figure 2, A and B), with juxtamedullary glomeruli being particularly affected (Supplemental Figure 2, A′ and B′). No changes of collagen IV and podocin deposition in the tuft were detected either in the whole cortex or specifically the juxtamedullary area (Supplemental Figure 2, C, C′, D, and D′).

CD44 specifically marks activated PECs invading the glomerular tuft during glomerular scarring in humans and animals^6,7,27–29 and is functionally involved in the pathogenesis of experimental crescentic GN and collapsing FSGS.^30,31 When analyzing all cortical glomeruli, no changes in CD44 were found (Supplemental Figure 2E), whereas the number of juxtamedullary glomeruli with CD44-positive PECs was significantly increased in Thy1.1 mice (Supplemental Figure 2E′). This suggested that the juxtamedullary glomeruli are particularly sensitive to the development of FSGS; however, given that only a few glomeruli were affected, this did not translate into functional impairment. When comparing males and females, no differences in blood pressure, body weight, kidney and body weight ratio, or kidney function were detected in 35-week-old Thy1.1 mice (Supplemental Table 3). However, male Thy1.1 mice showed significantly higher collagen IV expression in the tuft and increased numbers of sclerotic juxtamedullary glomeruli (Supplemental Tables 4 and 5).

Challenged Thy1.1 Mice Developed Progressive Decline of Kidney Function and Interstitial Fibrosis

Because the spontaneous phenotype in younger mice was not sufficient and the high age in older mice was not feasible to effectively study FSGS progression, we next characterized FSGS development in Thy1.1 mice after the injection of anti-Thy1.1 antibody. In the time course study, serum urea was significantly increased at day 1 (Figure 1A). Proteinuria was significantly higher at day 1 and day 7 and then decreased back to normal level at day 21 and day 100 (Figure 1B). Creatinine clearance was decreased at all of the time points after disease induction (Figure 1C).

Figure 1.

Decline of kidney function and glomerulosclerosis during Thy1.1 FSGS time course. Female and male mice were injected with anti-Thy1.1 antibody and euthanized on days 0 (n=4), 1 (n=6), 3 (n=6), 7 (n=6), 14 (n=6), 21 (n=6), and 100 (n=10). There was no urine for three male mice at day 1. Kidney function was decreased after one dose of anti-Thy1.1 antibody shown by increased serum urine (A), increased 24-hour proteinuria (B), and decreased creatinine clearance (C). Podocytes were damaged after injection of anti-Thy1.1 antibody and podocin expression in tuft was significantly decreased during FSGS (D). (D′–D′′′′) Representative pictures from immunohistochemistry staining for podocin. (E) Segmentally and globally sclerotic glomeruli were significantly increased. Representative pictures showing a normal glomerulus at day 0 (E′) and PEC proliferation on day 7 (E′′, arrow) which led to formation of segmental scars (E′′′, arrow) and global glomerulosclerosis (E′′′′). (F) Collagen IV deposition in the glomerular tuft was significantly upregulated during the disease. (F′–F′′′′) Representative pictures for collagen IV staining during the time course. Scale bars represent 20 µm. *P<0.05, **P<0.01, $P<0.001, #P<0.0001, versus day 0. Red dots indicate female mice; blue dots indicate male mice.

We next analyzed whether the specific glomerular injury in this model would translate to interstitial fibrosis and inflammation. Collagen III and collagen IV deposition in the cortex increased from day 7 until day 100, becoming significant at day 14 for collagen IV and day 100 for collagen III (Supplemental Figure 3, A–A′′′′ and B–B′′′′). The development of interstitial fibrosis was also confirmed by PDGFR-β expression analysis, which was significantly upregulated at day 100 (Supplemental Figure 3, C–C′′′′). Interstitial inflammation analyzed by F4/80 staining, as a marker of macrophages, increased at later time points (Supplemental Figure 3, D–D′′′′).

When comparing males and females, the majority of parameters at most time points were similar (apart from a few parameters which differed only on day 1 or day 100; Supplemental Tables 6 and 7).

Podocyte Injury and FSGS Developed in Challenged Thy1.1 Mice

The mice also developed hypertension which reached significance on day 100 (Supplemental Figure 4A). To assess podocyte injury, we analyzed the expression of the podocyte marker podocin using immunohistochemistry. This showed a discontinuous spotty staining pattern at day 1 when the podocytes got injured (Supplemental Figure 4B). Podocin expression was markedly decreased in glomerular tufts from day 7 to day 100, and the podocin-positive glomerular area decreased at day 7 and day 21 (Figure 1, D–D′′′′). At later time points, some glomeruli with complete loss of podocin were observed (not shown), and the number of glomeruli with podocin was also reduced from day 7 to day 100 (Supplemental Figure 4C). We also observed PEC vacuolization at day 1 and hypertrophy at day 3 (Supplemental Figure 4, D and D′). Accumulation of PECs was detected in the Bowman’s space and adhesions formed between Bowman’s space and glomerular tufts from day 7 (Figure 1, E′ and E′′). On day 7 and later time points, the injury evolved to segmental and global glomerulosclerosis (Figure 1, E′′′ and E′′′′). The number of sclerotic glomeruli was increased at day 14, day 21, and day 100 compared with day 0 (Figure 1E). Although in the unchallenged Thy1.1 mice the affected glomeruli were mainly in the juxtamedullary area, in the challenged mice the glomerular damage was diffuse and all cortical glomeruli were affected. Glomerular collagen IV deposition was progressively increased, reaching significance at day 14, day 21, and day 100 (Figure 1, F–F′′′′). Additionally, quantification of the number of glomeruli with collagen IV within the Bowman’s space was increased at day 14, day 21, and day 100 (Supplemental Figure 4E). No changes in podocin expression in the tuft and glomerulosclerosis were found between males and females during the time course (Supplemental Table 8).

Ultrastructural analysis confirmed regular podocyte foot processes lining the basement membrane at day 0, whereas they were fused from day 1 onward (Supplemental Figure 5, A–E). PECs became hypertrophic on day 3 and started to migrate and form tip lesions to the glomerular tuft on day 7 (Supplemental Figure 5, F–J). We also observed PEC vacuolization from day 1 onward (Supplemental Figure 5, K and L) and podocyte stress (irregular cell surface and small black dots in Bowman’s space; Supplemental Figure 5, K and M).

Taken together, in the accelerated FSGS model after anti-Thy1.1 antibody injection, prominent and progressive FSGS development was associated with pathologic renal function parameters, rendering the model suitable to study FSGS development.

Phenotype of PECs during FSGS Development in Thy1.1 Mice

We next analyzed the pathologic processes that PECs are undergoing during FSGS development. Using CD44 staining, PECs showed signs of activation from day 3 (Supplemental Figure 6A) and migrated to the glomerular tuft from day 7. Scoring of CD44 expression in PECs was significantly increased from day 7 until day 100 (Figure 2, A–A′′′′). The number of glomeruli with CD44 expression in PECs increased from day 7 (Supplemental Figure 6B). Costaining of CD44 and Ki67 suggested that the activated PECs proliferated during the disease (Supplemental Figure 6C). This was further confirmed by costaining of Ki67 and PAS, especially at day 3 and day 7. The number of Ki67-positive PECs per glomerulus was increased at day 3 and day 7, and then decreased at later time points but was still higher when compared with day 0 (Figure 2, B–B′′′′). Quantification of glomeruli with Ki67 expression in PECs showed the same trend as Ki67-positive cells per glomerulus (Supplemental Figure 6D). Most Ki67-positive PECs were found in glomeruli at day 3 and day 7 compared with other time points (Supplemental Figure 6E). The phenotype of PECs changes during the disease as also shown by the prominent increases in expression of α-SMA, a marker of mesenchymal transition and “myofibroblast-like” phenotype (Figure 2, C–C′′′′). In line with α-SMA, PDGFR-β, a marker of mesenchymal cells,³² was markedly upregulated in the cells occupying Bowman’s space. A significantly increased number of glomeruli with PDGFR-β in Bowman’s space was found at day 14, day 21, and day 100 compared with day 0 (Figure 2, D–D′′′′). LKIV69 was shown previously to be a specific marker for the PEC-derived matrix.³³ In this study, LKIV69-positive staining was detected in Bowman’s space from day 3 to day 100, suggesting that PECs were involved in the FSGS development in the Thy1.1 model (Supplemental Figure 7). When comparing males and females, no differences were found apart from the number of glomeruli with CD44-positive PECs on day 100 and PEC proliferation in Ki67 at day 14, both of which were higher in males (Supplemental Tables 9 and 10).

Figure 2.

PECs were activated, migrated, and proliferated and exhibited a phenotypic change during Thy1.1 FSGS time course. During the time course CD44 expression per glomerulus was significantly increased (A). The immunohistochemistry staining showed no expression of CD44 protein in PECs in control mice, whereas PECs were activated and migrated to tuft from day 7 until later time points (A′–A′′′′, arrows). (B) Ki67-positive PECs significantly increased at day 3 and day 7. (B–B′′′′) There was almost no proliferation of PECs at day 0, whereas during the disease proliferation of PECs (arrows) occurred. (C and D) Glomeruli with α-SMA and PDGFR-β expression in Bowman’s space were significantly increased during FSGS. (C′–C′′′′ and D′–D′′′′) Immunohistochemistry pictures for α-SMA and PDGFR-β showing upregulated proteins in Bowman’s space (arrows). Scale bars represent 20 µm. *P<0.05, **P<0.01, #P<0.0001, versus day 0. Red dots indicate female mice, blue dots indicate male mice.

Taken together, a prominent PEC phenotypic switch accompanies FSGS development with an initial activation phase, followed by a proliferative phase, and finally leading to a profibrotic (“execution”) phase (Supplemental Figure 8). PDGFR-β is highly upregulated in Bowman’s space during the FSGS development.

Expression of Pdgfb and Pdgfrb in Thy1.1 Mice and Human FSGS

In view of the upregulation of PDGFR-β in PECs, we next analyzed the expression of its ligand Pdgfb using FISH. Expression of Pdgfb in podocytes was upregulated at day 3, day 7, and day 14 (Figure 3, A–D and A′–D′), and there were less changes at day 100 compared with day 0 (Supplemental Figure 9, A and C). We counted the nuclear mRNA dots of Pdgfb in podocytes locating in the outer layer of glomerular tuft in the sections at early time points (i.e., day 0, day 3, and day 7), and results showed they were significantly increased at day 3 and day 7 (Supplemental Figure 9D). To confirm the results of immunohistochemistry, we also analyzed Pdgfrb, which was expressed in PECs at all of the time points and was increased from day 7 onward (Figure 3, A–D and A′′–D′′ and Supplemental Figure 9, A and B). The quantification of Pdgfrb nuclear mRNA dots in PECs was significantly increased at day 7 and day 14 (Supplemental Figure 9E).

Figure 3.

Upregulation of Pdgfb in podocytes and Pdgfrb in PECs in Thy1.1 mice and human FSGS. FISH pictures of glomeruli for Pdgfb (red) and Pdgfrb (green) for the time course (n=3 in each group) (A–D). Pdgfb in podocytes was upregulated after podocyte injury (A′–D′, arrows) and Pdgfrb was upregulated on PECs (A′′–D′′, arrows) during the disease. Biopsies from MCD patients (n=3) and secondary FSGS patients (n=4) were analyzed for expression of PDGFB (red) and PDGFRB (green) by FISH (E and F). In the slides from MCD patients very limited expression of PDGFB in podocytes (E and E′, arrow) and PDGFRB in PECs (E and E′′, arrow) was observed. In FSGS they were both upregulated in podocytes or in PECs, respectively (F–F′′, arrows). (G) Quantification of PDGFRB mRNA nuclear dots in PECs was significantly increased in FSGS patients compared with MCD patients, and dots of the same color represents the same patients. Scale bars represent 10 and 20 µm (mouse) or 10 and 50 µm (human), respectively. #P<0.0001, MCD versus FSGS.

In biopsy samples from patients with secondary FSGS, the mRNA expression of PDGFB was upregulated in podocytes (Figure 3, E, E′, F, and F′) and PDGFRB was upregulated in PECs (Figure 3, E, E′′, F, and F′′) compared with glomeruli from MCD patients. The quantification of nuclear mRNA dots of PDGFRB in PECs was significantly increased in FSGS patients compared with MCD patients (Figure 3G).

Taken together, PDGFB was expressed by damaged podocytes and PDGFR-β was upregulated on PECs, providing a potential paracrine molecular signaling network in FSGS.

Increased Expression of Pdgf in PECs in 5/6Nx and Col4^−/− Mice

To verify our findings from Thy1.1 mice, two other models developing sclerotic lesions were analyzed: 5/6Nx mice and Alport mice. The 5/6Nx mice developed FSGS as shown by increased number of sclerotic glomeruli (Figure 4, A–A′′), CD44 expression in PECs (Figure 4, B–B′′), the number of glomeruli with collagen IV (Figure 4, C–C′′), and PDGFR-β expression (Figure 4, D–D′′) compared with sham control. The mRNA expression of Pdgfb in podocytes was upregulated in 5/6Nx compared with sham mice (Figure 4, E, G, G′, H, and H′). The expression of Pdgfrb in PECs was slightly but not significantly increased (Figure 4, F, G, G′′, H, and H′′; Supplemental Table 11). In Col4^−/− mice, the mRNA expression of Pdgfb in podocytes and Pdgfrb in PECs was also increased (Supplemental Figure 10).

Figure 4.

Upregulation of Pdgfb in podocytes and Pdgfrb in PECs in 5/6Nx mice. Three male mice were used as sham control and five male mice underwent 5/6Nx. (A–A′′) The number of sclerotic glomeruli was significantly higher in the 5/6Nx group compared with the sham group. (B–B′′) The number of glomeruli with CD44-positive staining in PECs was significantly increased in 5/6Nx mice. (C–C′′) Expression of collagen IV in Bowman’s space was enhanced in the 5/6Nx group and the number of glomeruli with collagen IV deposition in Bowman’s space was significantly increased in comparison to the sham group. (D–D′′) The number of glomeruli with PDGFR-β in Bowman’s space was also increased in the 5/6Nx group. (E) Nuclear dots of Pdgfb mRNA (red) per podocyte was significantly increased in the 5/6Nx group in comparison with controls. (F) The mRNA expression of Pdgfrb (green) in PECs was also increased by trend in the 5/6Nx group. There was only low mRNA expression of Pdgfb (G and G′, arrow) in podocytes and Pdgfrb (G and G′′, arrow) in PECs in the control group. (H–H′′′) The mRNA expression of Pdgfb in podocytes (arrow) and Pdgfrb in PECs (arrow) was increased in 5/6Nx mice. Scale bars represent 10 or 20 µm. *P<0.05, **P<0.01, $P<0.001, sham group versus 5/6Nx group. Red dots indicate female mice. BM, Bowman’s.

In a transgenic mouse line with podocyte-specific Pdgfb deletion, CD44 expression in PECs was significantly reduced after 5/6Nx compared with wild-type littermates with normal Pdgfb expression (Supplemental Figure 11).

Therapeutic Effects of Anti–PDGF-B Antibody Treatment in Thy1.1 Mice

To analyze the functional role of PDGF in FSGS, we treated Thy1.1 mice with PDGF-B neutralizing antibodies (treatment group) or isotype-matched IgG (control group). To analyze the effects on early phases of PEC activation and proliferation, we analyzed the mice with the accelerated model on day 7 (Figure 5A). Serum urea was significantly reduced by the therapy (Figure 5B), whereas the urine albumin-creatinine ratio (Supplemental Figure 12A) and proteinuria were not significantly changed, albeit they were slightly lower in the treated mice (P=0.062; Figure 5C). Creatinine clearance improved after treatment with the anti–PDGF-B antibody (Figure 5D). No difference in blood pressure was detected between the two groups (Figure 5E), consistent with normal blood pressure found at this time point in the time course experiment.

Figure 5.

Effects of anti–PDGF-B antibody therapy in Thy1.1 mice on day 7. Female mice were injected with anti–PDGF-B antibody (“Anti-B”) or IgG control (“IgG”) after one dose of anti-Thy1.1 antibody (A). Serum urea was significantly reduced (B). (C) Proteinuria was slightly reduced but could not obtain significance (P=0.062). (D) Kidney function was improved revealed by creatinine clearance. (E) Systolic blood pressure did not show any difference between the two groups. (F–F′′) The number of sclerotic glomeruli was decreased in the anti–PDGF-B group. (G–G′′) The number of WT-1–positive cells did not differ between the two groups, whereas podocin expression in the tuft tended to be increased (H–H′′). PEC activation (I–I′′) and proliferation (J–J′′) were significantly reduced by the short time administration of anti–PDGF-B antibody. Collagen IV in Bowman’s space was decreased (K–K′′). (L–L′′) The number of glomeruli with PDGFR-β in Bowman’s space was also reduced after treating the mice with anti–PDGF-B antibody. Arrows point to the PECs or Bowman’s space. Scale bars represent 20 or 50 µm. *P<0.05, **P<0.01, anti-B versus IgG. Red dots indicate female mice. BM, Bowman’s.

The number of sclerotic glomeruli was significantly decreased in the anti–PDGF-B group (Figure 5, F–F′′). The number of WT-1–positive cells in tuft did not differ (Figure 5, G–G′′), and podocin expression in the tuft tended to be increased after anti–PDGF-B antibody treatment (Figure 5, H–H′′). The activation and proliferation of PECs, analyzed as CD44 expression (Figure 5, I–I′′) and Ki67-positive PECs (Figure 5, J–J′′), were both significantly decreased in the anti–PDGF-B group. The number of glomeruli with CD44 or Ki67-positive PECs was also significantly reduced after anti–PDGF-B treatment (Supplemental Figure 12, B and C). The number of glomeruli with collagen IV within Bowman’s space was also significantly decreased in the treatment group (Figure 5, K–K′′). The number of glomeruli with PDGFR-β (Figure 5, L–L′′) and α-SMA (Supplemental Figure 12D) expression in Bowman’s space was markedly decreased in the treated mice (Figure 5, L–L′′).

Next, we treated male Thy1.1 mice with the PDGF-B neutralizing antibodies until day 21 to analyze the effects on FSGS progression (Figure 6A). Compared with the control group injected with isotype-matched IgG, serum urea did not show differences (Figure 6B), whereas albuminuria and proteinuria were significantly reduced in the treatment group (Figure 6C, Supplemental Figure 12E). No differences in creatinine clearance and blood pressure were detected (Figure 6, D and E). The number of sclerotic glomeruli was decreased by the treatment (Figure 6, F–F′′). No differences in WT-1–positive cells (Figure 6, G–G′′) nor podocin expression in the glomerular tuft were detected at this later disease stage (Figure 6, H–H′′). CD44 expression in PECs per glomerulus (Figure 6, I–I′′) and glomeruli with CD44-positive PECs (Supplemental Figure 12F) were significantly decreased in the anti–PDGF-B group. In line with the day 7 experiment, the number of Ki67-positive PECs per glomerulus (Figure 6, J–J′′), the number of Ki67-positive glomeruli (Supplemental Figure 12G), and the number of glomeruli with collagen IV (Figure 6, K–K′′), PDGFR-β (Figure 6, L–L′′), and α-SMA (Supplemental Figure 12H) in Bowman’s space were all significantly reduced by the treatment.

Figure 6.

Effects of anti–PDGF-B antibody therapy in Thy1.1 mice on day 21. Male mice were injected with anti–PDGF-B antibody (“Anti-B”) or IgG control (“IgG”) until day 21 (A). No differences in serum urea were detected between the two groups (B). Proteinuria was reduced (C), whereas no differences in creatinine clearance (D) and blood pressure (E) were detected after 21 days’ of treatment with anti–PDGF-B antibody. Glomerular sclerosis was significantly reduced (F) as shown in the PAS staining (F′ and F′′). (G–G′′) The number of WT-1–positive cells in tuft did not differ between the two groups. (H–H′′) Quantification and representative images showed no differences in podocin expression. CD44 expression (I) or Ki67⁺ PECs per glomeruli (J) were significantly decreased, accompanied by the reduced number of glomeruli with collagen IV (K) and PDGFR-β (L) in Bowman’s space after 21 days of anti–PDGF-B antibody therapy. (I′–L′ and I′′–L′′) Representative pictures for staining of CD44, Ki67+PAS, collagen IV, and PDGFR-β, respectively. Arrows point to the PECs or Bowman’s space. Scale bars represent 20 or 50 µm. *P<0.05, ** P<0.01, $P<0.001, anti-B versus IgG. Blue dots indicate female mice. BM, Bowman’s.

Taken together, inhibition of PDGF-B reduced activation, proliferation, and profibrotic activity of PECs resulting in reduced formation of FSGS and improved indices of renal function.

PDGF-B Is Involved in PEC Activation and Proliferation In Vitro

To further confirm the specific effects of PDGF-B on PECs, we analyzed the same processes in vitro using primary murine PECs. The expression of Cldn1, a marker of differentiated PEC, was significantly downregulated at 24 hours after stimulation with PDGF-B whereas Cd44 was upregulated after 6 hours (Figure 7, A and B). The protein level of CD44 was significantly increased after 24 hours after stimulation with PDGF-B (Figure 7, C and D). PDGF-B stimulation also induced PEC proliferation, and this proliferation could be completely suppressed by anti–PDGF-B antibody or the small molecule tyrosine kinase inhibitor imatinib, which inhibits PDGFR-β signaling (Figure 7E).

Figure 7.

PDGF-B induced PEC activation and proliferation in vitro. Primary mouse PECs were stimulated with PDGF-B for 6 or 24 hours. The expression of the PEC marker Cldn1 (A) was downregulated and the activation marker Cd44 (B) was upregulated. (C and D) The protein expression level of CD44 was increased after PDGF-B stimulation by Western blot. (E) Bromodeoxyuridine (BrdU) assay showed proliferation of PECs after PDGF-B stimulation that was completely suppressed either by anti–PDGF-B antibody (Anti-B) or imatinib. Unstim. (n = 6), PDGF-B (n = 6). BrdU experiment: Unstim. (n = 24), PDGF-B (n = 32), PDGF-B+Anti–PDGF-B antibody (n = 8), PDGF-B+Imatinib (n = 8). Unstim: unstimulated. *P<0.05, **P<0.01, $P<0.001, #P<0.0001.

Discussion

In this study, we demonstrated that the Thy1.1 mouse is a suitable model to study the development and progression of FSGS. Utilizing this model, we show the involvement of PDGF in FSGS pathogenesis.

The Thy1.1 transgenic mice developed only a mild spontaneous renal phenotype and FSGS, consistent with the original publication of this mouse strain.¹⁴ We found that juxtamedullary glomeruli exhibited stronger damage compared with glomeruli from the rest of the cortex, which could be of potential interest for studying the higher sensitivity of the juxtamedullary glomeruli to FSGS as was also described in other animal models or human autopsies.^34–37 On the other hand, injection of a single dose of anti-Thy1.1 antibody in the Thy1.1 transgenic mice represents a robust model to study FSGS. It is easy to conduct, reproducible with all animals developing the disease with reasonably low variance and both kidneys could be used for analyses. Although the disease severity in females was similar to that in males, male mice tended to be more susceptible to kidney damage as observed before in rats with chronic renal failure and CKD patients.^38,39 We already found a higher glomerular collagen IV deposition in unchallenged male mice. Although we can exclude blood pressure as the culprit for sex-dependent differences we did not analyze hormone levels^40,41 or specific genes such as EGFR.⁴²

We also looked at late stages of the diseases (e.g., day 100), revealing that the FSGS and decline in creatinine clearance persisted and could not be restored. On the other hand, proteinuria and serum urea at day 100 did not significantly differ from sham animals whereas interstitial fibrosis/tubular atrophy and FSGS persisted. This is in line with a recent study showing that current kidney function parameters can overestimate kidney tissue repair in mouse kidney disease models.⁴³

A growing body of evidence suggests that PECs are crucially involved in FSGS formation.^6,7,30 Using a set of analyses, we confirmed and further extend and support this hypothesis by showing that PECs first become activated (activation phase), then proliferate (proliferation phase) followed by a profibrotic phenotype switch contributing to FSGS (profibrotic execution phase). Besides activation and proliferation, the phenotype of PECs changed from epithelial to mesenchymal or myofibroblast-like, characterized by expression of α-SMA and PDGFR-β. Interestingly, this mesenchymal phenotype of PECs was retained for a long time. A similar finding was previously identified in human pauci-immune crescentic GN and FSGS,^44,45 which further supports the relevance of this process and the model to study FSGS. Furthermore, we verified that PECs became activated and underwent a phenotype change expressing PDGFR-β in two other models developing sclerotic lesions, namely 5/6Nx mice and Col4^−/− Alport mice. In line with this, mice with podocyte-specific Pdgfb deletion exhibited significantly less PEC activation after 5/6Nx. In this experiment, however, the kidney injury and FSGS development were very mild in both transgenic mice and WT littermates, most likely because of the known glomerulosclerosis-resistant C57BL/6 genetic background in these mice.⁴⁶

Given the effects of PDGF in mesenchymal cells, we hypothesized that PDGF might mediate the pathologic processes in PECs during FSGS development. We show that PDGF-B is released by injured podocytes, thereby acting as a paracrine stress signal to PECs, which show increased expression of the PDGFR-β. This regulation is similar in the murine model and human biopsies with FSGS. One previous study showed that in crescentic GN PDGF-B and PDGFR-β were both increased in PECs during crescent formation, suggesting an autocrine signaling.⁴⁷ This is consistent with our findings regarding PDGFR-β, but in our model, we have not seen an upregulation of Pdgfb in PECs. Also, in another study focusing on renal interstitial fibroblasts, we have not seen any indication of autocrine effects of PDGF-B,²⁶ which is in line with the majority of reports showing PDGF to be mainly involved in paracrine signaling and cell crosstalk. The upregulation of PDGFR-β in PECs and their phenotypic switch in FSGS are analogous to our previous findings, showing that a specific activation of PDGFR-β in renal mesenchymal cells in mice resulted in their activation and phenotypic switch to myofibroblasts.²⁶

Our interventional studies showed that specific inhibition of PDGF-B effectively reduced FSGS and improved renal function by blocking PEC proliferation, PEC activation, and profibrotic activity, supported by our in vitro studies in PECs. We cannot fully exclude a potential direct effect on podocytes, but it seems rather unlikely because we could neither detect PDGFR-β expression on podocytes nor observe effects on WT-1, podocin expression, or ultrastructural changes after anti–PDGF-B treatment. We used the monoclonal anti-human PDGF-B antibody MOR8457 that was developed for use in patients. However, the antibody can also be used in preclinical testing, because it effectively inhibited PDGF-B in mice in our study and in a previous study in liver fibrosis in multiple drug resistance-2 (MDR2)-null mice.²⁰ Compared with other small molecule tyrosine kinase inhibitors which inhibit PDGFR signaling such as imatinib, the antibody is highly specific whereas the tyrosine kinase inhibitors are not. This might be associated with a better safety profile, as was suggested in a phase 1 trial for anti–PDGF-D antibody (CR002) in humans. Besides, inhibition of PDGF signaling can effectively reduce interstitial fibrosis^10,12,13 (i.e., the common consequence of progressive kidney diseases). This was also the case in our study, thereby suggesting that anti-PDGF might be an effective double-hit approach in CKD.

Our study has some limitations. We mainly focused on the podocyte and PEC changes during the progression of FSGS in this study. Even though podocytes and PECs are supposed to be the main cell types involved in the FSGS,⁴⁸ other cell types in glomeruli (e.g., mesangial cells and endothelial cells) could also participate in the development of FSGS. One limitation of this model is that genetic manipulations, such as cell-specific gene knockouts, are very challenging or even not feasible, and seem to be strongly dependent on the genetic background of the lines being crossed (unpublished data). However, we also do not know any Cre drivers that specifically target mesangial cells or glomerular endothelium or all PECs.

Although the Thy1.1 model was successfully conducted in females and males, data from Alport and 5/6Nx models come exclusively from male mice. Some groups only comprise a small animal number. Our study design started with anti–PDGF-B antibody treatment 1 day after disease induction (i.e., at a very early disease stage). This showed the involvement of PDGF in FSGS pathogenesis. However, further studies will be required to refine the treatment protocol and full therapeutic potential of the inhibition of PDGF in FSGS.

In conclusion, we provide to our knowledge the first evidence of a novel molecular crosstalk between podocytes and PECs mediated by PDGF-B-PDGFR-β, driving the progression of FSGS. This provides a rationale for more detailed studies addressing the therapeutic potential of PDGF inhibition in patients with FSGS.

Supplementary Material

jasn-34-241-s001.pdf ^{(2.4MB, pdf)}

Disclosures

P. Boor reports consultancy fees from Bristol Myers Squibb. J. Floege reports consultancy fees and honoraria from Amgen, AstraZeneca, Bayer, Boehringer Ingelheim, Calliditas, Chinook, Novo Nordisk, Novartis, Omeros, Travere, Vifor, and Visterra; reports advisory or leadership role with Calliditas, Omeros, and Travere; and is on the speakers bureau for Amgen, AstraZeneca, Novartis, and Vifor. All remaining authors have nothing to disclose.

Funding

This study was funded by the German Federal Ministries of Education and Research (STOP-FSGS-01GM1901A to P. Boor and M.J. Moeller), the German Research Foundation (322900939, 454024652, and 432698239 to P. Boor; 446660122 to M.J. Moeller; and 445703531 to B.M. Klinkhammer, M.J. Moeller, J. Floege, and P. Boor), and the European Research Council (Consolidator Grant AIM.imaging.CKD 101001791 to P. Boor).

Author Contributions

P. Boor, T. Jia, and B.M. Klinkhammer conceptualized the study; P. Boor, B.M. Klinkhammer, J. Floege, and M.J. Moeller were responsible for funding acquisition; P. Boor was responsible for project administration; P. Boor, M.J. Moeller, and B. Smeets were responsible for resources; P. Boor, B.M. Klinkhammer, and J. Floege were responsible for supervision; E.M. Buhl and T. Jia were responsible for data curation; E.M. Buhl, T. Jia, B.M. Klinkhammer, and T. Xu were responsible for investigation and methodology; T. Jia was responsible for formal analysis and visualization; P. Boor, E.M. Buhl, and T. Jia wrote the original draft; and J. Floege, B.M. Klinkhammer, M.J. Moeller, and B. Smeets reviewed and edited the manuscript.

Data Sharing Statement

All data used in this study are available in this article.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/JSN/D683.

Supplemental Figure 1. No changes of blood count and kidney function in 35-week-old Thy1.1 mice.

Supplemental Figure 2. Mild kidney injury in 35-week-old Thy1.1 mice.

Supplemental Figure 3. Tubulointerstitial fibrosis and inflammation in the Thy1.1 FSGS model.

Supplemental Figure 4. Characteristics of the Thy1.1 FSGS model.

Supplemental Figure 5. Ultrastructural analyses of the glomerular damage of the Thy1.1 FSGS model.

Supplemental Figure 6. Activation and proliferation of PECs during FSGS in Thy1.1 mice.

Supplemental Figure 7. PECs produced extracellular matrix during FSGS in Thy1.1 mice.

Supplemental Figure 8. Scheme of FSGS progression in Thy1.1 mice.

Supplemental Figure 9. Glomerular mRNA expression of Pdgfb and Pdgfrb in Thy1.1 FSGS mice.

Supplemental Figure 10. Glomerular mRNA expression of Pdgfb and Pdgfrb in Col4^−/− Alport mice.

Supplemental Figure 11. Deletion of Pdgfb in podocytes reduces PEC activation in 5/6Nx mice.

Supplemental Figure 12. Reduced kidney damage after anti–PDGF-B antibody treatment.

Supplemental Table 1. Probes and opals used for FISH.

Supplemental Table 2. Primers for quantitative RT-PCR.

Supplemental Table 3. Blood pressure, body weight, and kidney weight with blood and urine laboratory parameters of 35-week-old male and female Thy1.1 mice.

Supplemental Table 4. Comparison of parameters in 35-week-old male and female Thy1.1 mice in cortex.

Supplemental Table 5. Comparison of parameters in 35-week-old male and female Thy1.1 mice in juxtamedullary area.

Supplemental Table 6. Body weight, kidney weight, and blood pressure with parameters of male and female mice in Thy1.1 FSGS time course.

Supplemental Table 7. Parameters of protein expression in tubulointerstitium for male and female mice in Thy1.1 FSGS time course.

Supplemental Table 8. Comparison of podocin and glomerulosclerosis between male and female mice in Thy1.1 FSGS time course.

Supplemental Table 9. Comparison of collagen IV and CD44 for male and female in Thy1.1 FSGS time course.

Supplemental Table 10. Comparison of Ki67, α-SMA and PDGFR-β expression in glomeruli for male and female mice during FSGS time course.

Supplemental Table 11. mRNA expression of Pdgfb in podocytes and Pdgfrb in PECs in 5/6Nx mice.