TrkC Is Essential for Nephron Function and Trans-Activates Igf1R Signaling

Significance Statement

Identifying signaling pathways necessary for healthy glomeruli and nephrons could inform development of targeted therapies for glomerular diseases, which are largely lacking. Neurotrophic tyrosine kinase receptor 3 (TrkC) activity during nephron development and maintenance is part of one such pathway. Mice with deficiency or heterozygous or homozygous overexpression of TrkC in the nephron exhibit an expression level–dependent phenotype concomitant with albuminuria, podocyte foot process effacement, mesangial hypertrophy, and aspects of FSGS. In podocyte culture, activation of TrkC by its specific ligand results in transactivation of the Igf1 receptor (Igf1R), linking TrkC signal transduction to Igf-related signaling. These data identify TrkC as essential for maintaining glomerular integrity and to crosstalk with Igf-related signal transduction.

Abstract

Background

Injury to kidney podocytes often results in chronic glomerular disease and consecutive nephron malfunction. For most glomerular diseases, targeted therapies are lacking. Thus, it is important to identify novel signaling pathways contributing to glomerular disease. Neurotrophic tyrosine kinase receptor 3 (TrkC) is expressed in podocytes and the protein transmits signals to the podocyte actin cytoskeleton.

Methods

Nephron-specific TrkC knockout (TrkC-KO) and nephron-specific TrkC-overexpressing (TrkC-OE) mice were generated to dissect the role of TrkC in nephron development and maintenance.

Results

Both TrkC-KO and TrkC-OE mice exhibited enlarged glomeruli, mesangial proliferation, basement membrane thickening, albuminuria, podocyte loss, and aspects of FSGS during aging. Igf1 receptor (Igf1R)–associated gene expression was dysregulated in TrkC-KO mouse glomeruli. Phosphoproteins associated with insulin, erb-b2 receptor tyrosine kinase (Erbb), and Toll-like receptor signaling were enriched in lysates of podocytes treated with the TrkC ligand neurotrophin-3 (Nt-3). Activation of TrkC by Nt-3 resulted in phosphorylation of the Igf1R on activating tyrosine residues in podocytes. Igf1R phosphorylation was increased in TrkC-OE mouse kidneys while it was decreased in TrkC-KO kidneys. Furthermore, TrkC expression was elevated in glomerular tissue of patients with diabetic kidney disease compared with control glomerular tissue.

Conclusions

Our results show that TrkC is essential for maintaining glomerular integrity. Furthermore, TrkC modulates Igf-related signaling in podocytes.

Nephrons as functional units of the kidney each consist of a glomerulus and a tubular system. Because podocytes play an important role in building and maintaining the kidney filter of the glomerulus, it is not surprising that injury to podocytes often results in glomerular disease such as FSGS, hypertensive nephropathy, or diabetic nephropathy (DN) and consecutive nephron malfunction.^1,2 These diseases usually progress to CKD and in some cases result in ESKD. Podocytes extend primary processes that divide into foot processes that wrap the glomerular capillaries.³ With their intercellular contacts called slit diaphragms, podocytes form an essential part of the kidney filter.³ Damage to podocytes results in retraction and spreading of their foot processes followed by breakdown of slit diaphragms and changes of their adhesive properties leading to podocyte detachment and loss.² Podocyte loss is a principal cause that prompts glomerular disease progression.⁴ For most glomerular diseases, therapies targeting specific pathogenetic mechanisms are still lacking. Thus, it is attractive to explore signaling pathways potentially contributing to the progression of glomerular disease to identify novel drug targets for treatment.

Podocytes are among few cell types outside of the neuronal compartment that also express neurotrophic tyrosine kinase receptor 3 (TrkC).⁵ Conventional TrkC knockout mice exhibit albuminuria in the first days of the postnatal period and die due to neuronal defects several days after birth.^5,6 TrkC belongs to the Trk family of receptor tyrosine kinases that also includes TrkA, TrkB, and a structurally distinct receptor p75NTR.⁷ TrkC is activated by a specific neurotrophin—neurotrophin-3 (Nt-3)—which shows little potential to activate TrkA or TrkB.⁷ Trk receptors are mainly expressed in the nervous system and promote neuronal survival, synaptic plasticity, and neurite growth.⁸ Upon ligand binding, TrkC can form homodimers, which leads to transautophosphorylation on tyrosine residues in its cytoplasmic domain. By this, signaling intermediaries such as Src homologous and collagen-like (Shc) adapter protein or phospholipase C-γ (PLC-γ) can be recruited to the cytoplasmic domain of TrkC. This initiates AKT serine/threonine kinase (Akt) or extracellular signal–regulated kinase (Erk) activation or the release of inositol phosphates with consecutive activation of protein kinase C (PKC).^9–13 In podocytes, TrkC signals to the actin cytoskeleton via Erk-mediated Wiskott–Aldrich syndrome family verprolin-homologous protein 2 (WAVE2) activation thereby resulting in podocyte migration in cell culture.¹⁴

The IGFs Igf1 and Igf2 are structurally related to proinsulin and regulate proliferation, differentiation, survival, insulin-like metabolic processes, and many known actions of growth hormone (GH).^15,16 They interact with two transmembrane receptors—the Igf1 and -2 receptors (Igf1R and Igf2R)—which possess intrinsic tyrosine kinase activity.¹⁶ Most known biologic effects of Igf1 and Igf2 are thought to be mediated by binding to the Igf1R.¹⁶ The action of Igf1 and -2 is modulated by a family of six Igf-binding proteins (Igfbp).¹⁷ Igfbp mainly inhibit Igf actions.¹⁷ Igf1 and -2 are expressed by many cell types and are important for normal pre- and postnatal kidney development.^18,19 Podocytes express Igf1R and Igf2R, several glucose transporters, and insulin receptor (InsR).^20–25 Expression of dominant-negative Igf1R in podocytes leads to foot process effacement and proteinuria in mice.²⁶ Furthermore, the Igf-signaling axis was described to be dysregulated in several kidney diseases such as proteinuric CKD, DN, AKI, and polycystic kidneys.^27–29

Here, we show that nephron-specific TrkC knockout (TrkC-KO) and TrkC-overexpressing (TrkC-OE) mice exhibit enlarged glomeruli with mesangial proliferation, basement membrane thickening, albuminuria, podocyte loss during aging, and aspects of FSGS. On a molecular level, we demonstrate that Igf1R signaling-associated and metabolic gene expression is regulated in TrkC-KO mouse glomeruli. Phospho-proteome analysis of podocyte lysates illustrates that activated TrkC induces signaling via insulin-, erb-b2 receptor tyrosine kinase (Erbb)–, and Toll-like receptor (TLR)–related pathways. We show that Nt-3–dependent TrkC activation in cultured podocytes and overexpression of TrkC in the nephron in mice lead to phosphorylation of activating tyrosine residues of the Igf1R, whereas Igf1R phosphorylation is decreased in TrkC-KO mice kidneys. Furthermore, expression of TrkC is elevated in glomerular tissue of patients with diabetic nephropathy (DN).

Methods

Generation of Nephron-Specific TrkC Knockout Mice

TrkC flox mice were imported from Jackson Laboratories on a C57BL/6 background [B6.129P2(SJL)-Ntrk3tm1Ddg/J; stock number: 022364].³⁰ These mice carry a cassette with LoxP sites flanking a point mutation that results in an amino acid exchange in position 617 from phenylalanine to alanine targeting the TrkC exon 14. This point mutation does not induce a phenotype. We crossed these mice on a C57BL/6 background to mice that express Cre-recombinase by the Six2-promoter to generate mice that are deficient of TrkC in the parietal and visceral epithelium of the glomerulus and the proximal and distal tubulus.³¹ These mice will be called nephron-specific TrkC knockout mice (TrkC-KO).

Cloning of the Mouse ES Cell–Targeting Plasmid

Human TrkC (hTrkC) was amplified from a human podocyte cDNA library with the primers (forward) CAC​CGG​GCG​CGC​CAT​GGA​TGT​CTC​TCT​TTG​CCC​A and (reverse) TTA​ATT​AAC​TAG​CCA​AGA​ATG​TCC​AGG​TAG​ATT​G, cloned into the pENTR/D-TOPO cloning vector (Thermo Fisher Scientific), and further introduced with AscI-sites into the plasmid pSERCA that enables the insertion of a stop-cassette flanked by two LoxP sites into the ROSA26 gene locus as previously described.^32,33 Functionality of the targeting vector expressing hTrkC after Cre recombinase activity was validated in vitro by cotransfecting HEK293T cells with the targeting vector and the Cre expression plasmid pOG231 followed by immunoblotting for TrkC. pOG231 was a gift from Geoff Wahl (plasmid #17736; Addgene).

ES Cell Transfection and Selection of Targeted Clones

CV19 ES cells (passage 12 [129Sv x C57BL/6J]) were expanded in HEPES-buffered DMEM supplemented with 15% FBS (PAA), nonessential amino acids, l-glutamine, β-mercaptoethanol, 1000 U of recombinant leukemia inhibitory factor (MERCK Millipore) per ml, and antibiotics (penicillin [100 U/ml] and streptomycin [100 µg/ml]). For electroporation, 2×10⁷ cells were resuspended in 0.8 ml of Capecchi buffer (20 mM HEPES [pH 7.4], 173 mM NaCl, 5 mM KCl, 0.7 mM Na₂HPO₄, 6 mM dextrose, 0.1 mM β-mercaptoethanol). The targeting vector pSERCA_hTrkC_GFP (100 µg) was linearized with SgfI, and electroporated together with the ROSA26_ZFN (L6 and R4) coding plasmids (25 µg each) at 25 µF and 400 V in 0.8-mm electroporation cuvettes (Gene Pulser; Bio-Rad). After electroporation, cells were cultivated for 10 minutes at room temperature and plated onto ten 100-mm-diameter culture dishes containing a γ-irradiated monolayer of mouse primary G418-resistant fibroblast feeder cells. Thirty-two hours later, 350 µg of G418 (Invitrogen) per ml was added to the culture medium. The medium was replaced every day, and colonies were picked and analyzed 8 days after plating.

DNA Southern Blot Analysis

Positively targeted ES cell clones were analyzed using the Southern blot method. Approximately 5 µg of genomic DNA was digested with HindIII, fractionated on 0.8% agarose gels, and transferred to GeneScreen nylon membranes (NEN DuPont). The membranes were hybridized with ³²P-labeled DNA fragments containing sequences 5′ to the targeted homology (HR probe subcloned in pCRII-TOPO vector) and washed with (final concentrations) 0.5× SSPE (1× SSPE is 0.18 M NaCl, 10 mM NaH₂PO₄, and 1mM EDTA [pH 7.7]) and 0.5% SDS at 65°C. After first screening, DNAs isolated from positively targeted ES cells were proven using HindIII, BamHI, EcoRI, and EcoRV digestion with the help of the HR probe and the probe corresponding to the neomycin resistance gene.

Blastocyst Injection

Correctly targeted ES cells from clones (3H1 and 3F6) were injected into 3.5-day B6D2F1 blastocysts. Routinely, we injected 12–14 ES cells into one blastocele. After injection, embryos were kept in KSOM medium and subsequently transferred into the uteri of 2.5-day pseudopregnant CD-1 foster mice, or into the infundibuli of oviducts of 0.5-day pseudopregnant CD-1 mice. Foster mice carried pups to term. Chimeras were identified by their agouti coat color contribution. For the germ-line transmission, high-percentage male chimaeras were crossed to C57BL/6J female mice. Heterozygous agouti offspring (ROSA26_hTrkc +/−) were confirmed by Southern blot analysis for the presence of the targeted allele. The neomycin/STOP cassette (Neo) flanked by two LoxP sites was removed after crossing with CRE-expressing mice. Mice were kept in specific pathogen–free animal facilities.

Establishing hTrkC Knock-In Mice

The mouse line was established by breeding male with female C57Bl/6J ROSA26_ hTrkC knock-in transgenic mice to produce heterozygous mice. Subsequently, heterozygous mice were interbred to homozygosity. Pups were weaned at 19–23 days after birth, and females were kept separately from males. The mice were housed in standard IVC cages. General health checks were performed regularly to ensure that any findings were not the result of deteriorating physical conditions of the animals.

Genotyping

All animal experiments were performed in compliance with the guidelines for the welfare of experimental animals issued by the federal government of Germany/North Rhine-Westphalia (84–02.04.2014.A469; 84–02.04.2014.A467). Standard techniques were employed for genotyping using the following primers: TrkC knockout (fwd TAAACCTTTGGT; rev ATCCCCATGTCT); TrkC overexpression (fwd CCT​AAA​GAA​GAG​GCT​GTG​CTT​TGG; rev wt GAG​CGG​GAG​AAA​TGG​ATA​TG; rev transgene GCG​AAG​AGT​TTG​TCC​TCA​ACC); and Cre recombinase (fwd GAC​CAG​GTT​CGT​TCA​CTC​A; rev TAGCGCCGTAAATCAA).

Urine Analysis for Albumin by SDS–PAGE

First, 5 µl of collected spot or metabolism cage urine was mixed with 5 µl of 2× Laemmli and separated by SDS–PAGE. Gels were stained with Coomassie for 45 minutes and destained using standard methods.

Measurements of Blood Glucose

Blood glucose levels of 5-month-old control and nephron-specific TrkC-KO mice were measured with an Ascensia ELITE set (Bayer, Germany) following the manufacturer’s instructions.

Histologic Evaluation of Mouse Kidney Sections

Mice were perfused through the right ventricle with Sterofundin Ecoflac Plus infusion solution (B. Braun Melsungen AG, Germany). The left kidney was removed and transferred into 0.5 M sucrose solution (see Immunofluorescence Analysis of Mouse Kidney Sections). Perfusion was then continued with 4% paraformaldehyde (PFA) for 3–5 minutes. The right kidney was removed, transferred to 4% PFA for 4 hours, and embedded in paraffin using standard techniques. Then, 2-µm-thick sections were stained with periodic acid–Schiff. Sections were digitalized with a Leica SCN400 slide scanner and analyzed using Aperio Image Scope (Leica Microsystems IR GmbH).

Transmission Electron Microscopy

Kidneys were fixed in 2.5% glutaraldehyde with Sorensen’s phosphate buffer, postfixed with 1% osmiumtetroxide, dehydrated, and embedded in Epon. Ultrathin sections (60 nm) were cut (Leica UC6 ultramicrotome, Vienna, Austria) and counterstained with uranyl acetate incubated in lead citrate solution. Samples were analyzed with an EM 208 S transmission electron microscope (Philipps, Amsterdam, The Netherlands).

Immunofluorescence Analysis of Mouse Kidney Sections

Kidneys of sodium chloride–perfused control, TrkC-KO, TrkC-OE_het (heterozygous), or TrkC-OE_hom (homozygous) mice were kept in 0.5 M sucrose solution overnight at 4°C and were then embedded in Tissue-Tek O.C.T. (Sakura Finetek), immediately frozen in super-cooled 2-methylbutane for 1 minute, and then stored at −80°C. Cryo-sections (5 µm) were obtained by a Leica Cm3050 S cryostat and then fixed with 4% PFA at room temperature, or in the case of PCNA staining fixed with methanol for 5 minutes at −20°C, quenched for 20 minutes (100 mM glycine, 100 mM NH₄Cl in PBS), and pretreated with saponine (30 µg/ml) for 5 minutes. Cryo-sections were blocked with 5% BSA and 5% goat serum in PBS for 30 minutes. Primary antibodies against Nephrin (#BP5030, 1:100; Acris), PCNA (#13110, 1:500; Cell Signaling Technologies), Wt-1 (#rb-9267-P1, 1:50; Thermo Fisher), and Zonula occludens-1 (ZO-1) ( #61–7300; 1:100; Thermo Fisher) were diluted in 1% BSA in PBS, applied on the sections, and incubated overnight at room temperature in a humidity chamber. The cryo-sections were then washed with PBS three times and secondary antibodies (anti-goat Alexa Fluor⁵⁹⁴, anti-rabbit Alexa Fluor⁵⁹⁴, anti-goat Alexa Fluor⁶⁴⁷, or anti-rabbit Alexa Fluor⁶⁴⁷ [Invitrogen]) diluted 1:500 in 1% BSA in PBS were applied for 45 minutes at room temperature. Alexa Fluor⁵⁹⁴ Phalloidin (Thermo Fisher) was used to visualize actin. Images were acquired using an Observer Z1 microscope with Apotome, Axiocam MRm (Zeiss), and EC Plan Apochromat 63×/1.40 Oil M27 objective.

Evaluation of Podocyte Number and Glomerular Volume

Podocyte number and glomerular volume were measured using a two-thickness method.³⁴ In short, sections were stained with Wt-1 and Phalloidin (please see the section above). At least 20 glomeruli per animal of three different control or TrkC-KO mice were evaluated using ImageJ. Wt-1–positive cells were counted and then normalized to the average glomerular volume, which was calculated using the Weibel formula.

Quantification of Glomerular Basement Membrane Thickness and Slit Diaphragms per Micrometer of Basement Membrane

For the quantification of slit diaphragms per micrometer of basement membrane, at least 300 µm of basement membrane per group (at least 100 µm of basement membrane per animal) was measured, employing images of ultrathin sections acquired by electron microscopy. Slit diaphragm were counted manually, and the length of the basement membrane was measured with ImageJ. Three different animals per group were evaluated.

The thickness of the glomerular basement membrane was measured three times per micrometer along a loop. Altogether, 4 mm of basement membrane was evaluated per animal, employing electron micrograph images of ultrathin sections.³⁵ For this, ImageJ was used and four animals per group were analyzed.

Glomerular Lysates for Quantitative PCR

Five-month-old TrkC-KO or control mice were anesthetized using a mixture of ketamine (100 mg/kg body wt) and xylazine (10 mg/kg body wt) and perfused with Dynabeads M-450 Tosylactivated (234155; Invitrogen) as specified by the manufacturer’s instructions. Isolated kidneys were minced in HBSS and digested with collagenase type II (#17101–015, 1 mg/ml; Gibco), and glomeruli were sieved twice with a 100 µm cell strainer, washed with PBS, purified with a magnetic column following the manufacturer's instructions and lysed in RNA lysis buffer for RNA isolation.

Isolation of Nephron Segments for Quantitative PCR

Renal tubules were isolated from collagenase-digested cortical fragments of kidneys isolated from wild-type mice by a modification of previously described methods.^36,37 Renal cortices were dissected visually in ice-cold dissection solution (Eagle’s MEM containing 5 mmol/L glycine and 1% BSA). The fragments were transferred into warm MEM containing 1 mg/ml type 2 collagenase, 5 mM glycine, and 48 μg/ml soybean trypsin inhibitor and incubated at 37°C to digest. Every 5 minutes, supernatant was poured off the sedimented tissue into a new tube and fresh MEM-collagenase medium was added to the sediment. Then, the digestion was continued. The supernatant overlying the tubule segments was carefully removed and replaced by ice-cold, enzyme-free dissection solution containing 1% BSA to inactivate collagenases and proteases. For tubule selection, tubule-rich suspensions were gently pipetted into a dissection dish containing MEM with 0.05% BSA at 4°C. Proximal tubule (PT), thick ascending loop (TAL), and cortical collecting duct (CCD) segments were sorted from the central group of tubule segments as described before.^38,39 Total mRNA was isolated using the GeneElute Kit (Sigma). cDNA was prepared employing SuperScript-III First-Strand Synthesis SuperMix (Invitrogen, Karlsruhe, Germany). Gene expression profiles were analyzed by real-time PCR using SYBR Select Master Mix for CFX (Thermo Fisher, Waltham, MA) on a CFX Realtime Detection System (Biorad, Munich, Germany) as duplicates. Specific primer pairs were used as listed in Supplemental Table 1. Relative gene expression values were evaluated with the 2-ΔCt method using Gapdh as housekeeping gene.⁴⁰

Quantitative PCR of TrkC-KO Mouse Glomerular Lysates

Total RNA was isolated from mouse glomeruli using the Mammalian Total RNA Miniprep Kit (Sigma Aldrich) following the manufacturer’s instructions. cDNA synthesis was achieved using the Superscript III reverse transcriptase (Invitrogen). Quantitative real-time PCR was performed with SYBR Green PCR Master Mix (Thermo Fisher) using a CFX384 Touch (Bio-Rad, Munich, Germany) and CFX Manager v3.1 software. Each cDNA sample was measured as a triplicate and ΔCt-values of genes of interest were normalized to the murine housekeeping gene Gapdh. Primers used for rt-PCR are listed in Supplemental Table 1.

Mass Spectrometry Analysis

Proliferating murine podocytes inducibly expressing TrkC (K8 TrkC)²¹ were cultured on 15-cm (Ø) dishes and starved in serum-free medium 24 hours before stimulation with Nt-3. TrkC expression was induced by adding doxycycline (125 ng/ml) to the cell culture medium 20 hours before incubation with Nt-3 (50 ng/ml) for 10 or 60 minutes. Cells were washed twice with ice-cold PBS and lysed in 99°C prewarmed lysis buffer: 6 M guanidinium hydrochloride, 5 mM TCEP [tris(2-carboxyethyl)phosphine], 10 mM chloroacetamide, 100 mM Tris HCl (pH 8.5). Samples were incubated for 10 minutes at 99°C and sonicated for 2 minutes (1 s on, 1 s off, 50% amplitude). Proteins were first digested with LysC for 3 hours at 37°C, followed by sixfold dilution with 50 mM Tris-HCl and a second digest phase with trypsin overnight at 37°C. Samples were then desalted using SepPak cartridges, lyophilized, dissolved in 1 ml of buffer A (10 mM NH₄OH, pH 10.2), and subjected to offline, concatenated high-pH reversed-phase fractionation using an XBridge C18 BEC reversed-phase column (4.6×250 mm) and a multilinear gradient running from 0% to 25% B in 90 minutes, from 25% to 40% B in 5 minutes, and with a wash-out phase at 90% B for 10 minutes before re-equilibration at starting conditions (A: 10 mM NH₄OH; B:10 mM NH₄OH, 90% acetonitrile; flow rate 0.75 ml/min). After lyophilization, phosphopeptides were then enriched using TiO₂ beads from the resulting 12 peptide pools and three flowthrough fractions, the last one being extracted twice.

LC-MS/MS analysis of a total of 18 fractions per treatment was performed employing a Q-Exactive HF mass spectrometer (Thermo Scientific), online coupled via a Nanoflex ion source and a home-packed coated silica emitter tip (20-cm length, 75-µm ID, 360-µm OD; C18-AQ 1.9 µm, Dr. Maisch, Ammerbuch, Germany) to an EASY nLC nanoflow HPLC pump (Thermo Scientific). Online reversed-phase chromatography conditions were: 3% to 30% B in 60 minutes; 30% to 60% B in 15 minutes; and 60% to 90% B in 2 minutes followed by a 3-minute washout phase at 90% B and re-equilibration at starting conditions (A: 0.1% formic acid; B: 0.1% formic acid, 80% acetonitrile; flow rate 300 nl/min). The Q-Exactive HF mass spectrometer was operated in data-dependent acquisition mode (spray voltage 2.1 kV; column temperature maintained at 45°C using a PRSO-V1 column oven [Sonation, Biberach, Germany]). MS1 scan resolution was set to 60,000 at m/z 200 and the mass range to m/z 300–1750. AGC target value was 3×10⁶ with a maximum fill time of 100 ms. Fragmentation of peptides was achieved by higher-energy collisional dissociation using a top-12 method. Dynamic exclusion of previously identified peptides was allowed and set to 20 seconds; singly charged peptides and peptides assigned with a charge of 6 or more were excluded from the analysis. Data were recorded with Xcalibur software (Thermo Scientific). MS Data Analysis Raw MS files were processed using the MaxQuant computational platform (version 1.5.5.1).⁴¹ Identification of peptides and proteins was enabled by the built-in Andromeda search engine by querying the concatenated forward and reverse mouse Uniprot database (UP000000589_10090.fasta, version from 12/2015) including common laboratory contaminants. The allowed initial mass deviation was set to 7 and 20 ppm in the search for precursor and fragment ions, respectively. Trypsin with full enzyme specificity and only peptides with a minimum length of six amino acids were selected. The minimal score for a correctly identified modified peptide was set to 40. A maximum of two missed cleavages was allowed; the “match between runs” option was turned on. Carbamidomethylation (Cys) was set as a fixed modification, whereas oxidation (M), N-acetylation at the protein terminus, and phosphorylation at S, T, or Y residues were defined as variable modifications. For peptide and protein identifications, a minimum false discovery rate of 1% was required. The mass spectrometry proteomics data have been deposited to the ProteomeXchange Consortium via the PRIDE partner repository with the dataset identifier “PXD017960.”⁴²

Protein assignments derived from the extracted unique peptides in each sample were evaluated with the functional annotation tool of DAVID Bioinformatics Resources 6.8^43,44 regarding enrichment of signaling pathways and biologic processes. Venn diagrams of overlapping or unique proteins were visualized with Venny 2.1 (https://bioinfogp.cnb.csic.es/tools/venny/).

Cell Culture

Murine immortalized podocytes were cultured as described elsewhere.⁴⁵ Proliferating podocytes were cultured at a permissive temperature of 33°C. Murine wild-type podocytes (K8 wt) and podocytes inducibly expressing TrkC (K8 TrkC) were cultured on six-well plates and TrkC expression was induced by adding doxycycline to the cell culture medium for 16 hours (125 ng/ml) before stimulation with insulin (#I9278; Sigma Aldrich) or Nt-3 (#450–03; PeproTech). Cells were cultured in serum-free medium for 3 hours before incubation with Nt-3 (50 ng/ml) or Insulin (1.7 nmol/ml) for 10 minutes. HEK293T cells were cultured in standard DMEM (Thermo Fisher Scientific) containing 10% FCS and 1% antibiotics (penicillin/streptomycin) and glutamine at 37°C. Transient transfection of HEK293T cells was performed using the calcium-phosphate method.

Immunoblot Analysis

Murine immortalized podocytes or HEK293T cells were lysed in ice-cold radioimmunoprecipitation assay buffer (RIPA buffer; 20 mM Tris-HCl pH 7.4, 150 mM NaCl, 5 mM EDTA, 0.1% [w/v] SDS, 0.1% [w/v] deoxycholate, 1% [w/v] Triton X-100) containing protease (Roche) and phosphatase (Sigma Aldrich) inhibitors for 30 minutes on ice. Lysates were mixed every 5 minutes and were also mechanically lysed by passing through a 27-gauge syringe needle, and then centrifuged for 15 minutes at 4°C and >14,000 × g. Supernatant was mixed with 2× Laemmli buffer (4% SDS, 5% 2-ME, 10% glycerol, 0.002% bromophenol blue, and 0.0625 M Tris-HCl [pH 6.8]) and boiled at 95°C for 5 minutes. Western blotting was performed as described previously.¹⁴ Primary antibodies were diluted 1:1000 in 5% BSA in TBST. The following antibodies were used: Akt (#9272), IGF-I Receptor β (D23H3) XP (#9750), InsR β (4B8) (#3025), p44/42 MAPK (Erk1/2) (#9102), phospho-Akt (Ser⁴⁷³) (#9271), Phospho-IGF-I Receptor β (Tyr^1135/1136)/InsR β (Tyr^1150/1151) (19H7) (#3024), Phospho-IGF-I Receptor β (Tyr¹¹³¹)/InsR β (Tyr¹¹⁴⁶) (#3021), phospho-p44/42 MAPK (Erk1/2) (Thr²⁰²/Tyr²⁰⁴) (#9101), phospho-TrkA (Tyr⁴⁹⁰)/TrkB (Tyr⁵¹⁶) (C35G9) (#4619) (used for detection of phospho-TrkC Tyr⁵¹⁶ due to identical epitopes of TrkB and TrkC) and TrkC (C44H5) (#3376) were purchased from Cell Signaling Technologies, TrkC (7H6) (#sc-517245) from Santa Cruz, α-Actinin 4 (#ALX210–356) from Enzo Life Sciences, and β-Tubulin (#T8328) from Sigma Aldrich.

Whole-Kidney Lysates and Immunoprecipitation

For whole-kidney lysates, 13±0.1 mg of perfused C57BL/6, TrkC-KO, or TrkC-OE mouse kidney was homogenized in 500 µl of ice-cold IP buffer (20 mM Tris-HCl pH 7.4, 20 mM NaCl, 1 mM EDTA, 50 mM NaF, 10 mM Na₄P₂O₇, 1% Triton-X-100) containing protease and phosphatase inhibitors, then mechanically lysed by passing first through a 20-gauge syringe needle (five times) and then through a 27-gauge syringe needle (five times), and rested on ice for 30 minutes. Lysates were then centrifuged at 14,000 × g for 20 minutes at 4°C, and supernatants were used for immunoprecipitation or western blot analysis.

For immunoprecipitation, TrkC antibody was bound to Protein G Sepharose 4 Fast Flow beads (GE Healthcare Life Sciences) for 1 hour at 4°C while rotating before immunoprecipitation using 1 µl of antibody per 20 µl of beads. Beads were then washed with IP buffer and whole-kidney lysates were loaded on the beads; an input of 5% of the loaded lysates was kept. Immunoprecipitation was achieved overnight with rotating at 4°C; thereafter, the beads were washed three times with dilution buffer (10 mM Tris-HCl pH 7.5, 300 mM NaCl, 0.5 mM EDTA) and were then taken up in Laemmli buffer. Input and beads were boiled for 5 minutes at 95°C and then used for western blot.

Microarray Analysis of Human Kidney Biopsy Samples

Human renal biopsy specimens and Affymetrix microarray expression data were obtained within the framework of the European Renal cDNA Bank–Kröner-Fresenius Biopsy Bank after informed consent and with approval of the local ethics committees.⁴⁶ After renal biopsy, the tissue was transferred to RNase inhibitor and microdissected into glomeruli and tubulointerstitium and further processed as described before.⁴⁷ Fragmentation, hybridization, staining, and imaging were performed according to the Affymetrix Expression Analysis Technical Manual (Affymetrix). Published datasets of glomerular samples were used in this study (GSE99340, LD data from: GSE32591, GSE37463). Analysis included datasets from patients with DN (n=14), minimal change disease (MCD) (n=14), FSGS (n=23), and membranous nephropathy (n=21). Pretransplantation kidney biopsy samples from living donors (n=42) were used as control renal tissue. CEL file normalization was performed with the Robust Multichip Average method using RMAExpress (Version 1.0.5) and the human Entrez‐Gene custom CDF annotation from Brain Array version 18 (http://brainarray.mbni.med.umich.edu/Brainarray/default.asp). To identify differentially expressed genes, the Significance Analysis of Microarrays (SAM) method was applied using the SAM function in Multiple Experiment Viewer (TiGR MeV, version 4.9).⁴⁸ A Q value below 5% was considered to be statistically significant.

Statistical Analyses

Statistics and graphs were created using GraphPad PRISM software v5. All data are shown as mean with SEM. At least three independent experiments were performed and analyzed using a t test (two-tailed, unpaired) or two-way ANOVA with Bonferroni post hoc analysis, as specified.

Results

TrkC Is Expressed in Glomerular Podocytes

To test whether TrkC is expressed in glomerular cells, we immunoprecipitated TrkC from mouse glomerular lysates (Supplemental Figure 1A). Additionally, we performed immunofluorescence analysis of mouse kidney sections with antibodies specific for TrkC and nephrin or podocin as slit diaphragm markers and ZO-1 as a junctional marker (Figure 1A). TrkC appears to be expressed in podocytes in a partly similar pattern as the slit diaphragm proteins nephrin and podocin.

Figure 1.

TrkC is not necessary for developing slit diaphragms. (A) Immunofluorescence analysis was performed after staining mouse glomerular sections with antibodies specific for TrkC (green) and Nephrin (red), Podocin (red), or ZO-1 (red). Scale bar, 10 µm. Details to the right, 20 µm in total. n=3. (B) Immunoblot of whole-kidney lysates of three control and three nephron-specific TrkC knockout mice (TrkC-KO) with antibody specific for TrkC or β-Tubulin. (C) Densitometric quantification of the immunoblot bands in (B) is shown as a ratio of TrkC/β-Tubulin normalized to the control. Shown are means and SEM. *P<0.05 by unpaired two-tailed t test. (D) SDS–PAGE analysis of 5 µl of urine of representative 3-month-old control and TrkC-KO mice. As controls, 1 and 5 µg of BSA were loaded. (E) PAS staining of kidney sections of 3-month-old control and nephron-specific TrkC knockout mice (TrkC-KO). Scale bars, 25 µm. n=3. (F) TEM analysis of 3-month-old representative control and TrkC-KO mice. Scale bars, 2 µm (left column), 1 µm (right column). (G) Shown is the body weight of aging control and TrkC-KO mice from birth up to 15 months of age. n=3–9.

TrkC Is Not Necessary for Nephron Development

TrkC knockout results in proteinuria and foot process effacement in a conventional knockout mouse model.⁵ As conventional TrkC knockout mice die due to neuronal defects several days after birth, this model is not suitable to study the role of TrkC in the kidney.⁶ Therefore, we generated a knockout in nephrons by crossing mice that express Cre recombinase under the control of the Six2 promoter to mice carrying LoxP sites flanking TrkC exon 14.^30,31 Six2 is expressed during development in the metanephric mesenchyme.³¹ Thus, our mice show an early TrkC knockout in podocytes, parietal epithelial cells, and proximal and distal tubular cells and will be referred to as nephron-specific TrkC knockout mice (TrkC-KO) and respective littermate controls (control). We showed by immunoblot of whole-kidney lysates that TrkC expression was reduced in the kidneys of TrkC-KO mice compared with controls (Figure 1, B and C). TrkC-KO mice were born at the expected Mendelian frequency. They did not exhibit proteinuria at the age of 3 months as shown by SDS–PAGE and Coomassie staining (Figure 1D). Kidneys of 3-month-old TrkC-KO mice did not show histologic alterations in the glomerular or tubular compartment (Figure 1E). Three-month-old TrkC-KO mice had normal slit diaphragm morphology by transmission electron microscopy (TEM) (Figure 1F). This implies that TrkC is not essential for nephron development.

Nephron-Specific TrkC Deficiency Results in Mesangial Proliferation and Aspects of FSGS during Aging

TrkC-KO mice had a normal life span, had normal blood glucose by the age of 5 months, and did not exhibit differences in weight compared with controls while aging (Figure 1G, Supplemental Figure 1B). Histologically, TrkC-KO mice showed mesangial expansion, mesangial hypercellularity, and glomeruli with enlarged volume starting at 3–6 months of age (Figure 2, A and B; Supplemental Figure 1C). We then performed immunofluorescence analysis of TrkC-KO and control kidney sections of 6-month-old mice with antibody specific for proliferating cell nuclear antigen (PCNA) which marks proliferating cells. This showed that TrkC knockout in the nephron resulted in increased proliferation of mesangial cells (Figure 2B). At 12–15 months of age, nephron-specific TrkC-KO mice developed aspects of FSGS (Figure 2C). Histologically, the tubular cell compartment of TrkC-KO mice did not differ from control mice. PCNA staining of glomerular sections of 15-month-old mice also showed increased proliferation in nephron-specific TrkC knockout mice (Figure 2D). Podocyte depletion is an important pathomechanism that contributes to the development of FSGS. To assess this, we determined the podocyte number per glomerular tuft.³⁴ This showed that TrkC-KO mice exhibited a reduced podocyte number per glomerular tuft by 6 months of age (Figure 3A, Supplemental Figure 1D). Thus, TrkC is necessary for maintaining intact glomeruli during aging.

Figure 2.

TrkC deficiency in the nephron results in mesangial proliferation and aspects of FSGS during aging. (A) PAS staining of kidney sections of 6-month-old control and nephron-specific TrkC knockout mice (TrkC-KO). Scale bars, 100 µm (upper panel), 20 µm (middle panel), 20 µm (bottom panel). (B) Immunofluorescence analysis of kidney sections of 6-month-old control and TrkC-KO mice stained with antibodies specific for PCNA (red) and Nephrin (green). Scale bar, 10 µm. (C) PAS staining of kidney sections of 15-month-old control and TrkC-KO mice. Scale bars, 100 µm (upper panel), 20 µm (middle panel), 20 µm (bottom panel). (D) Immunofluorescence analysis of kidney sections of 15-month-old control and TrkC-KO mice stained with antibodies specific for PCNA (green) and Nephrin (red). Scale bar, 10 µm. (A–D) n=3–8.

Figure 3.

TrkC is necessary for maintaining intact filtration slits. (A) The podocyte number per glomerulus was calculated employing a single histologic kidney section method. Absolute values with SEM are shown. **P<0.01 by unpaired two-tailed t test. n=3. (B) SDS–PAGE analysis of 5 µl of urine of representative 15-month-old control and nephron-specific TrkC knockout mice (TrkC-KO). As controls, 1 and 5 µg of BSA were loaded. (C) TEM analysis of 15-month-old representative control and TrkC-KO mice. Scale bars, upper panel, 5 µm; middle panel, 2 µm; lower panel, 1 µm. (D) Quantification of slit diaphragms per micrometer of basement membrane. Shown are means and SEM. n=4 for each condition. (E) Quantification of basement membrane thickness in micrometers. Shown are means and SEM. n=4 for each condition. (D, E) ***P<0.001 by unpaired two-tailed t test. (F) Immunofluorescence analysis of kidney sections of 15-month-old control and TrkC-KO mice stained with antibodies specific for Nephrin (green) and ZO-1 (red). Scale bar, 10 µm. Details on the right, 20 µm in total. n=3.

TrkC Is Necessary for Maintaining Filtration Slits

TrkC-KO mice developed albuminuria at the age of 12–15 months (Figure 3B). We then analyzed the ultrastructure of TrkC-KO and control mouse glomeruli in old age (15 months) by TEM to test whether TrkC was necessary for maintaining podocyte slit diaphragms. Although the slit diaphragms of control mice were regular, TrkC-KO mice showed foot process effacement with fewer slit diaphragms per micrometer of glomerular basement membrane (Figure 3, C and D). Additionally, TrkC-KO mice exhibited a thicker basement membrane compared with control mice (Figure 3, C and E). To analyze the expression pattern of typical slit diaphragm proteins, we performed immunofluorescence analysis of kidney sections of 15-month-old TrkC-KO and control mice employing antibodies specific for either nephrin or ZO-1. In TrkC-KO mice, ZO-1 expression was patchier compared with controls (Figure 3F). Nephrin did not exclusively localize to the slit diaphragm but showed a more diffuse expression pattern in TrkC-KO mice compared with controls (Figure 3F). This is consistent with the ultrastructural finding of foot process effacement of TrkC-KO mice.

TrkC Overexpression in the Nephron Results in Proteinuria and Aspects of FSGS

To address whether the level of TrkC expression is crucial for nephron development and maintenance, we generated knock-in mice with the potential of conditional Cre-dependent expression of TrkC. Transgenic TrkC mice were generated employing the pSERCA plasmid that enables the insertion of a stop-cassette flanked by LoxP sites into the ROSA26 gene locus (Supplemental Figure 2, A–D).³² Expression of TrkC by the pSERCA plasmid was first validated in cell culture employing HEK293T cells by coexpression of the pOG231 plasmid which expresses Cre recombinase driven by a cytomegalovirus (CMV) promoter (Supplemental Figure 2E). DNA from ES cells and F1 founder mice lines was characterized by Southern blot analysis (Supplemental Figure 2, F and G).

TrkC was then over-expressed by crossing transgenic TrkC mice with mice that express Cre recombinase driven by the Six2 promoter to generate mice with nephron-specific overexpression of TrkC (referred to as TrkC-overexpressing [OE] mice) and respective littermate controls (Figure 4A).³¹ We first bred mice that carried the TrkC transgene heterozygously (TrkC-OE_het). Successful overexpression of TrkC was tested by analyzing whole-kidney lysates of TrkC-OE_het mice or controls by immunoblot with an antibody specific for TrkC (Figure 4B). TrkC-OE_het mice were born at the expected Mendelian ratio. They developed albuminuria and showed mesangial expansion at an age of 12–15 months alike to TrkC-KO mice (Figure 4, C and D). Ultrastructurally, TrkC-OE_het mice showed foot process effacement at an age of 15 months (Figure 4, E and F). We then generated homozygous nephron-specific TrkC-overexpressing mice (TrkC-OE_hom), employing the Six2 promoter to test whether the phenotype started earlier or was stronger in the homozygous condition. TrkC-OE_hom mice exhibited a phenotype analogous to heterozygous TrkC-OE_het mice with an earlier onset. They developed albuminuria at an age of 9 months concomitant with foot process effacement and aspects of FSGS during aging (Supplemental Figure 3, A–D). Thus, the TrkC gene dose appears to be important for functional glomeruli during aging.

Figure 4.

TrkC overexpression in the nephron results in albuminuria and mesangial expansion. (A) Schematic of the ROSA26 locus before (upper part) and after (lower part) Cre event. Blue box, ROSA26 locus flanking regions; gray box, promoter region; black arrows, LoxP sites; red box, stop cassette; blue arrow, hTrkC cDNA; yellow box, IRES site; green arrow, eGFP. (B) Immunoblot analysis of mouse kidney lysates with antibodies specific for TrkC and α-Actinin 4 of three control and three heterozygous nephron-specific TrkC-overexpressing mice (TrkC-OE_het). (C) SDS–PAGE analysis of 5 µl of urine of representative 15-month-old control and TrkC-OE_het mice. As controls, 1 and 5 µg of BSA were loaded. (D) PAS staining of kidney sections of 15-month-old control and TrkC-OE_het mice. Scale bars, 50 µm each. (E) TEM analysis of 15-month-old representative control and TrkC-OE_het mice. Scale bars, upper panel, 5 µm; middle panel, 2 µm; lower panel, 1 µm. (F) Quantification of slit diaphragms per micrometer of basement membrane. Shown are means and SEM. n=3 for each condition, *P<0.05 by unpaired two-tailed t test.

TrkC Activation Results in Over-Representation of Phospho-Proteins in Insulin-, Erbb-, and TLR-Related Signaling Pathways

To identify signaling mechanisms after TrkC activation, we performed a phospho-proteomic analysis of cellular lysates of a cultured mouse podocyte line that inducibly expressed TrkC after doxycycline (+dox) stimulation and that was incubated either with the TrkC ligand Nt-3 for 10 or 60 minutes (10’ or 60’) or with respective solvent controls (please see the Methods sections for details). In total, 14,019 unique phospho-peptides were identified, corresponding to 4545 unique phosphoproteins. Of these phospho-proteins, 75% were identified in all experimental conditions and thus not influenced by TrkC activation. The overlapping phospho-protein identification signatures of the respective podocyte lysates are visualized in a Venn diagram (Figure 5A).

Figure 5.

TrkC activation in podocytes regulates insulin-associated signaling pathways. (A) Venn diagram of 4545 unique phospho-proteins identified in lysates of cultured mouse podocytes expressing TrkC after doxycycline treatment. Four conditions were analyzed: control (without doxycycline treatment); dox (with doxycycline-induced TrkC overexpression); Nt3–10’ (with dox-induced TrkC overexpression and treatment with Nt-3 for 10 minutes); Nt3–60’ (with dox-induced TrkC overexpression and treatment with Nt-3 for 60 minutes). Percentages and total number of phospho-proteins identified in each group are listed. n=3 for each condition. (B) Analysis of over-represented pathways of phospho-proteins identified in the Nt3–60’–treated group in distinction to the control and dox groups. Shown are P values and fold enrichment (logarithmic scale). (C) Quantitative PCR of glomerular lysates of 5-month-old control and nephron-specific TrkC knockout mice (TrkC-KO). n=4–8, *P<0.05 by unpaired two-tailed t test. Neurotrophic tyrosine kinase, receptor, type 3 (Ntrk3, gene encoding TrkC). (D) Quantitative PCR of glomerular lysate (Glom), lysate of prepared PTs, and a prepared CCD/TAL fraction of wild-type mice shown as means and SEM. The following genes were analyzed: soluble glucose transporter 2 (Sglt2), glucose transporter 2 (Glut2), solute carrier family 12 member 1 (Nkcc2), aquaporin 2 (Aqp2), epithelial sodium channel (ENaC), Nphs1 (gene encoding Nephrin), Igf1 receptor (Igf1r), Igf2 receptor (Igf2r), insulin receptor (Insr), Igf binding protein 1 (Igfbp1), Igf binding protein 2 (Igfbp2), neurotrophic tyrosine kinase, receptor, type 3 (Ntrk3; gene encoding TrkC). Sglt2 and Glut2: controls for the PT; Nkcc2: control for the TAL; Aqp2 and ENaC: controls for the CCD; Nphs1: control for the glomerular fraction (podocytes). n=4.

We then performed an enrichment analysis to functionally characterize the phospho-proteins identified in the Nt-3–treated group (60’) compared with both controls (control, +dox). Thereby, we identified several over-represented pathways among these known TrkC-dependent signaling cascades such as neurotrophin and MAPK signaling (Figure 5B). Furthermore, Erbb, insulin, and TLR signaling pathways were over-represented (Figure 5B). Over-represented biologic processes and protein domains are listed in Supplemental Figure 4, A and B. Unique phospho-proteins identified in lysates of podocytes treated with Nt-3 for either 10 or 60 minutes and phospho-proteins identified in both lysates of podocytes treated with Nt-3 for 10 or 60 minutes are listed in Supplemental Tables 2–4.

To analyze whether the expression of corresponding genes of proteins associated with differentially regulated phosphorylation were also regulated after TrkC activation, we analyzed genes involved in insulin signaling and Erbb receptor signaling and metabolic genes by quantitative PCR in cultured podocytes. Wild-typic podocytes or podocytes inducibly expressing TrkC after doxycycline stimulation were incubated with Nt-3. The expression of Erbb4 and Igf1 and several genes associated with metabolic processes was differentially regulated after activation of TrkC (Supplemental Figure 5).

TrkC Knockout in Glomerular Epithelial Cells Regulates Metabolic and Igf-Associated Signaling Pathways

Next, we tested whether genes regulated in vitro in cultured podocytes were also regulated in TrkC-KO mice. For this, lysates of isolated TrkC-KO mouse glomeruli and controls at the age of 5 months before the onset of albuminuria were used. Among the regulated genes was the insulin growth factor binding protein 2 (Igfbp2) which functions in Igf-associated signal transduction (Figure 5C).⁴⁹ Furthermore, several genes involved in other metabolic processes such as enoyl-CoA hydratase and 3-hydroxyacyl CoA dehydrogenase (Ehhadh), dimethylarginine dimethylaminohydrolase 1 (Ddah1), aminoadipate-semialdehyde synthase (Aass), and solute carrier family 27 fatty acid transporter member 2 (Slc27a2) were increasingly expressed in TrkC-KO mouse glomeruli (Figure 5C). Genes that were not regulated significantly are shown in Supplemental Figure 6.

Next, we analyzed the expression profile of TrkC and Igf-signaling–associated genes in various segments of the nephron in wild-type mice. TrkC was expressed at a higher level by glomerular cells than cells of the PT or a combined cell fraction of TAL and CCD (Figure 5D). The PT expressed only minimal amounts of TrkC. Igf1R, Igf2R, and Igfbp2 expression was enriched in the glomerular fraction compared with TAL and CCD (Figure 5D). The InsR was expressed in all fractions with the highest expression in the glomerulus. Igfbp1 was also expressed in all tested segments with the highest expression in the glomerulus and PT (Figure 5D).

TrkC Transactivates the Igf1R

In podocyte culture, activation of TrkC by its ligand results in phosphorylation of insulin-signaling–related proteins (Figure 5, A and B). Furthermore, Igf-related pathways are regulated in vivo in TrkC-KO mice (Figure 5C). To test whether there is crosstalk between TrkC and insulin-related signaling, we generated mouse podocytes allowing conditional over-expression of TrkC after doxycycline treatment. Wild-type mouse podocytes (wt-podocytes) or podocytes that inducibly over-express TrkC (TrkC^OE-podocytes) were then treated with insulin or Nt-3. We subsequently analyzed whether the InsR was phosphorylated on a tyrosine residue (Y1146) described to facilitate the activation of the InsR and downstream signaling.⁵⁰ Although insulin treatment resulted in robust phosphorylation of the InsR on Y1146 either in wt-podocytes or in TrkC^OE-podocytes, incubation of podocytes with Nt-3 did not result in phosphorylation of the InsR on Y1146. InsR and TrkC activation are both known to result in phosphorylation of Akt and Erk (Figure 6, A and B).^14,51–53 Thus, this is not suitable to distinguish between InsR and TrkC activation. Because we found that members of the Igf receptor signaling cascade were regulated in TrkC-KO mice, we tested whether Igf1R was phosphorylated on activating tyrosine residues 1135 and 1136 after incubation with either insulin or Nt-3. Incubation of wt-podocytes or TrkC^OE-podocytes with either insulin or Nt-3 resulted in phosphorylation of the Igf1R on the activating residues Y1135/Y1136 (Figure 6, A and B). To exclude that Nt-3 might directly activate the Igf1R, we tested whether stimulation of HEK293T wild-type cells that do not endogenously express TrkC (Supplemental Figure 2E) or HEK293T cells transfected with V180-TrkC led to phosphorylation of Igf1R on Y1135/Y1136. Although stimulation of HEK293T cells that overexpress TrkC with Nt-3 resulted in phosphorylation of the Igf1R on Y1135/Y1136, this was not the case in wild-type HEK293T cells (Supplemental Figure 7). Next, we tested whether overexpression of TrkC in mice in the nephron resulted in increased activation of the Igf1R. Whole-kidney lysates of control and homozygous TrkC-OE mice were analyzed by immunoblot for phosphorylation of the activating tyrosine residues Igf1R Y1135/Y1136. TrkC-OE mice exhibited significantly more Igf1R phosphorylation in kidney cells than controls (Figure 6, C and D). In contrast, phosphorylation of Igf1R on Y1135/Y1136 was reduced in TrkC-KO mice whole-kidney lysates compared with controls (Figure 6, E and F).

Figure 6.

TrkC activation results in phosphorylation of the Igf1R on activating tyrosine residues. (A) Immunoblot analysis of wild-type K8 podocyte lysates or lysates of podocytes treated with doxycycline (Dox) to overexpress TrkC. Podocytes were additionally incubated either with insulin or with the TrkC ligand neurotrophin-3 (Nt-3) for 10 minutes. Immunoblots were performed with antibodies specific for Erk1/2, pErk1/2, Akt, pAkt, the InsR, pInsR Y1146, Trk, pTrk Y516, insulin growth factor 1 receptor (Igf1R), and pIgf1R Y1135/Y1136. α-Actinin 4 was employed as a loading control. One representative experiment is shown. n=3. (B) Densitometric quantification of the immunoblot bands in (A) is shown as a ratio of pIgf1R/Igf1R, pInsR/InsR, pAkt/Akt, and pErk/Erk. Shown are means and SEM. *P<0.05, **P<0.01, ***P<0.001 by two-way ANOVA with Bonferroni post hoc analysis. n=3. (C) Whole-kidney lysates of 15-month-old control and homozygous TrkC-OE mice were analyzed by immunoblot for phosphorylation of the Igf1R on tyrosine residues Y1135/1136. TrkC was blotted to show the overexpression; TrkC_long: longer exposure time; loading control: β-Tubulin. (D) Densitometric quantification of the immunoblot bands in (C) shown as ratio of pIgf1R/Igf1R. n=6, *P<0.05 by unpaired t test. (E) Whole-kidney lysates of control and TrkC-KO mice were analyzed by immunoblot for phosphorylation of the Igf1R on tyrosine residues Y1135/1136. TrkC was blotted to show the knockout; loading control: α-Actinin-4. pIgf1R_{(Y1135/Y1136)_short} or pIgf1R_{(Y1135/Y1136)_long} indicates the exposure time. (F) Densitometric quantification of the immunoblot bands in (E) shown as ratio of pIgf1R/Igf1R. n=6, *P<0.05 by unpaired t test.

TRKC and IGF1R Expression Are Regulated in Human DN

Next, we wondered whether TRKC- or IGF-signaling–related genes were differentially expressed in human glomerular disease. Gene expression analysis of glomerular lysates prepared by the European Renal cDNA Bank of patients suffering from MCD, FSGS, membranous glomerulonephropathy (MGN), and DN showed that TrkC expression was significantly elevated in glomerular tissue of patients with DN (Figure 7). This was not the case for the glomerular pathologies MCD, FSGS, and MGN. The TrkC ligand NT-3 was not differentially expressed. In glomerular tissue of patients with DN, the INSR, IGF1R, and IGF1 were significantly less expressed than in control tissue (glomerular tissue of living kidney donors).

Figure 7.

TRKC and IGF1R expression is regulated in human DN. Quantitative PCR analysis of glomerular lysate of patients with MCD, FSGS, MGN, or DN. Shown are the Entrez Gene ID, the gene symbol, and the log fold change of the analyzed genes compared with controls (glomerular lysate of living kidney donors). Neurotrophic tyrosine kinase receptor 3 (NTRK3): gene encoding TrkC, neurotrophin-3 (NT-3), INSR, IGF 1 receptor (IGF1R), IGF 2 receptor (IGF2R), IGF binding protein 1 (IGFBP1), IGF binding protein 2 (IGFBP2), IGF1. Red color indicates upregulation of gene expression; violet color indicates downregulation of gene expression shown as a heat map.

Discussion

In this study, we show that TrkC is not essential for nephron development, but is necessary for maintaining intact glomeruli during aging. Mice with loss of TrkC in the nephron exhibit enlarged glomeruli with mesangial proliferation, basement membrane thickening, foot process effacement with albuminuria, podocyte loss, and aspects of FSGS during aging. The regulation of TrkC expression in the nephron appears crucial for healthy glomeruli as heterozygous and homozygous mice that over-express TrkC in the nephron exhibit a similar phenotype. Phospho-proteome analysis of podocyte lysates shows that activated TrkC induces signaling via insulin-, Erbb-, and TLR-related pathways. In TrkC-KO mouse glomeruli, expression of genes associated with Igf signaling is regulated. Nt-3–dependent TrkC activation in cultured podocytes and overexpression of TrkC in the nephron in mice lead to phosphorylation of the Igf1R on activating tyrosine residues, whereas Igf1R phosphorylation is diminished in TrkC-KO mice kidney lysates. Furthermore, TrkC expression is elevated in glomerular tissue of patients with DN. Thus, TrkC is necessary to maintain healthy glomeruli and transactivates Igf1R in podocytes in vitro and in mouse kidneys in vivo.

Because mice with a conventional TrkC knockout develop albuminuria directly after birth,⁶ we were astonished to find that nephron-specific TrkC knockout mice age normally and develop albuminuria around 12 months of age. Conventional TrkC knockout mice die in the early postnatal period due to severe cardiac and neuronal abnormalities.⁶ Albuminuria is associated with severe outcomes in critically ill patients and may reflect microvascular effects of systemic inflammation.⁵⁴ Thus, albuminuria might have been a secondary symptom of the critically ill conventional TrkC knockout mice before death.

In adult mouse kidneys, TrkC appears to be more highly expressed in glomeruli than in some tubular segments. Although TrkC may be expressed in other Six2-targeted cells in the nephron besides podocytes, it is interesting that mice develop glomerular proteinuria, which argues that podocytes are most likely the predominant cell population involved in developing this phenotype. By employing the Six2 promoter to generate TrkC-OE mice, we ectopically overexpressed TrkC in the proximal and distal tubules where little endogenous TrkC is present. It is interesting to note that TrkC-OE mice do not show any histologic alterations in the tubular compartment. Thus, an increase in TrkC expression does not appear to influence the function of proximal and distal tubules. Nephron-specific TrkC-KO mice also do not show a tubular phenotype when aging, probably due to very low expression of endogenous TrkC in tubules of adult mice. During development, TrkC is only expressed in the collecting duct of embryonic kidneys.⁵⁵ This may explain why knockout of TrkC in glomerular parietal and visceral epithelial cells and proximal and distal tubules does not result in a developmental phenotype in mice.

Balanced TrkC expression in the nephron appears to be necessary for healthy glomeruli because nephron-specific TrkC knockout and overexpressing mice exhibit a similar expression level–dependent phenotype. We cannot exclude that excessive expression of hTrkC in mice may function in a dominant-negative fashion in TrkC-OE mice, thus mimicking the knockout phenotype. However, our data showing that the Igf1R phosphorylation is being increased in TrkC-OE and decreased in TrkC-KO mice argue that physiologic TrkC expression is tightly regulated and that changes in the quantity independently of this being an increase or decrease in the amount results in stimuli that lead to breakdown of the slit diaphragm and consecutive sclerosis of the glomerulus. It is not uncommon that signaling cascades are dysregulated by either excess or lack of a signaling component. The importance of a balanced expression was previously shown for other receptors and cell adhesion proteins with signaling properties such as the glucocorticoid receptor or the Drosophila ortholog of the slit diaphragm protein Nephrin, Sticks and stones.^56,57 A physiologic expression is also essential to prevent tumorigenesis. The paradigm that proteins function either as oncogenes or tumor suppressors was challenged by findings that, during the development of hepatocellular carcinoma, molecules such as the hepatocyte growth factor receptor (HGFR), NF-κB, or β-catenin can function either as oncogenes or tumor suppressors.⁵⁸ In the Akt pathway, both decreased and increased levels of SR protein-specific kinase (SRPK1) promote cancer.^59,60 Many proteins function in multiprotein complexes. Stoichiometric changes in protein complexes may result in dysfunction of the respective protein complex independent of the cause being excess or lack of one protein component. It remains to be elucidated whether this is the case for TrkC.

GH/Igf signaling is implicated to play a role in several kidney pathologies such as proteinuric CKD, DN, AKI, and polycystic kidneys.²⁷ Igf1 mediates many actions of GH.¹⁶ In podocytes, Igfbp3 enhances the proapoptotic effect of TGF β and inhibits the antiapoptotic effect of bone morphogenetic protein-7.⁶¹ In urinary podocytes of patients with FSGS, Igfbp3 and Igfbp1 are increased compared with controls or patients with MCD with the same level of proteinuria.⁶² Thus, this may indicate a specific role for the Igf-signaling axis in the pathology of FSGS. The GH/Igf system is recognized to be involved in DN and several interventions targeting this system were shown to ameliorate experimental DN.^63–65 We show that TrkC can transactivate the Igf1R in podocytes in culture and that TrkC overexpression results in increased Igf1R phosphorylation in the kidney in mice, whereas knockout of TrkC results in diminished activation of Igf1R in vivo. Dysregulated Igf1R activation by TrkC in podocytes could thus contribute to the development of albuminuria and foot process effacement of TrkC-OE and TrkC-KO mice in vivo. Although TrkC-KO and TrkC-OE mice develop aspects of FSGS during aging, it is worth noting that TrkC expression is not regulated in glomeruli of patients with FSGS, whereas TrkC is expressed at higher levels in glomeruli of patients with DN. In contrast, in human diabetic kidneys the InsR, Igf1R, and Igf1 are downregulated. Downregulating the expression of Igf-associated signaling proteins may be the cellular response to increased TrkC expression and potentially increased Igf1R activation. It will be interesting to further analyze whether TrkC plays a role in glomerular diseases that exhibit dysregulated Igf signaling. Further investigation is necessary to understand the biologic effect of the TrkC-Igf1R transactivation and its importance for kidney health.

In addition to Igf-related signaling, we identified TrkC activation to have an effect on TLR- and Erbb-associated signaling in podocytes. Furthermore, TrkC loss or overexpression results in an FSGS-like phenotype in mice. Chronic FSGS is a scarring process that is associated with chronic inflammation partly explained by TLR activation.^66–68 Inhibition of TLR2 and TLR4 attenuates the development of FSGS in rats with 5/6 nephrectomy.⁶⁹ We find that TrkC expression is regulated in DN in human biopsy samples. Both Toll-like and Erbb receptors play important roles in the pathogenesis of DN. TLR4 promotes DN in mice.^70,71 Antagonizing TLR4 protects against Streptozotocin-induced DN in mice.⁷² EGF receptor (Erbb1, EGFR) depletion in podocytes attenuates DN.⁷³ Thus, crosstalk of TrkC with Toll-like and Erbb receptors could be important in the pathogenesis of FSGS or DN. Elucidating the molecular mechanisms that connect TrkC with Erbb and TLR signaling may further our understanding of FSGS or DN and potentially uncover molecular targets to treat these diseases.

Disclosures

H. Pavenstädt reports research funding from Sanofi; and reports serving as a scientific advisor or membership with Deutsche Forschungsgemeinschaft (DFG). All remaining authors have nothing to disclose.

Funding

This work was supported by German Research Foundation (Deutsche Forschungsgemeinschaft) grants GE2158/3-1, GE2158/5-1 (to B. George) and SFB/TRR57, SFB/TRR219, BO3755/3-1, and BO3755/6-1 (to P. Boor), Interdisciplinary Centre for Clinical Research (IZKF) Münster grant Geo2/003/16 (to B. George), German Federal Ministry of Education and Research (Bundesministerium für Bildung und Forschung) grant STOP-FSGS-01GM1901A (to P. Boor), and by a scholarship of the German Kidney Foundation (to B. George).

Supplemental Material

This article contains the following supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2020040424/-/DCSupplemental.

Supplemental Figure 1. TrkC knockout in the nephron results in podocyte loss.

Supplemental Figure 2. Generation of transgenic TrkC mice for tissue-specific overexpression.

Supplemental Figure 3. Homozygous mice with TrkC overexpression in the nephron develop albuminuria earlier than heterozygous mice.

Supplemental Figure 4. Over-represented biologic processes and protein domains after neurotrophin-3 treatment of cultured podocytes.

Supplemental Figure 5. Regulated gene expression after TrkC activation in cultured podocytes.

Supplemental Figure 6. Additional data: Expression of genes regulating metabolic processes in nephron-specific TrkC knockout mice.

Supplemental Figure 7. Nt-3 does not induce phosphorylation of the Igf1R on activating tyrosine residues in HEK293T cells that do not express TrkC.

Supplemental Table 1. List of primers used for rt-PCR analysis.

Supplemental Table 2. List of unique phospho-proteins identified in the group treated with Nt-3 for 10 minutes.

Supplemental Table 3. List of unique phospho-proteins identified in both Nt-3–treated groups.

Supplemental Table 4. List of unique phospho-proteins identified in the group treated with Nt-3 for 60 minutes.