Cell Cycle and Senescence Regulation by Podocyte Histone Deacetylase 1 and 2

Abstract

Significance Statement

The loss of integrity of the glomerular filtration barrier results in proteinuria that is often attributed to podocyte loss. Yet how damaged podocytes are lost remains unknown. Germline loss of murine podocyte-associated Hdac1 and Hdac2 (Hdac1/2) results in proteinuria and collapsing glomerulopathy due to sustained double-stranded DNA damage. Hdac1/2 deletion induces loss of podocyte quiescence, cell cycle entry, arrest in G1, and podocyte senescence, observed both in vivo and in vitro. Through the senescence secretory associated phenotype, podocytes secrete proteins that contribute to their detachment. These results solidify the role of HDACs in cell cycle regulation and senescence, providing important clues in our understanding of how podocytes are lost following injury.

Background

Intact expression of podocyte histone deacetylases (HDAC) during development is essential for maintaining a normal glomerular filtration barrier because of its role in modulating DNA damage and preventing premature senescence.

Methods

Germline podocyte-specific Hdac1 and 2 (Hdac1/2) double-knockout mice were generated to examine the importance of these enzymes during development.

Results

Podocyte-specific loss of Hdac1/2 in mice resulted in severe proteinuria, kidney failure, and collapsing glomerulopathy. Hdac1/2-deprived podocytes exhibited classic characteristics of senescence, such as senescence-associated β-galactosidase activity and lipofuscin aggregates. In addition, DNA damage, likely caused by epigenetic alterations such as open chromatin conformation, not only resulted in podocyte cell-cycle entry as shown in vivo by Ki67 expression and by FUCCI-2aR mice, but also in p21-mediated cell-cycle arrest. Through the senescence secretory associated phenotype, the damaged podocytes secreted proinflammatory cytokines, growth factors, and matrix metalloproteinases, resulting in subsequent podocyte detachment and loss, evidenced by senescent podocytes in urine.

Conclusions

Hdac1/2 plays an essential role during development. Loss of these genes in double knockout mice leads to sustained DNA damage and podocyte senescence and loss.

Introduction

CKD affects approximately 15% of the United States population.¹ Approximately 80% of CKD cases manifest glomerular damage, resulting in a pathological loss of large molecular weight proteins, such as albumin.^2–4 Loss of podocytes, which are terminally differentiated epithelial cells that lines the outermost surface of the glomerulus, has been implicated in the pathogenesis of proteinuria, but unanswered questions remain on how damaged podocytes are lost.

Major pathways that have been postulated in podocyte loss include anoikis,^5,6 autophagy,^7,8 pyroptosis,⁹ mitotic catastrophes,^10–13 and apoptosis.^14–19 Yet, apoptosis as the major cause of podocyte death is still unclear, as nuclear fragmentation and apoptotic bodies have rarely been observed in vivo.^14,18,19 Recently, cellular senescence, a biological process that limits the proliferation of damaged mitotic cells, has also been suggested as another mechanism that may lead to podocyte loss.^20–24

Growing evidence suggests that postmitotic cells are also capable of entering a state of senescence by entering the cell-cycle after DNA damage or in an age-related manner.^25–28 Senescent cells can negatively affect surrounding cells through the senescence-associated secretory phenotype (SASP), which play an important role in degenerative disease.^25,29 It is becoming increasingly clear that chromatin reorganization plays a key role in senescence, although the precise mechanism remains unclear. One possibility is that certain chromatin states are more susceptible to DNA breaks and prevent successful DNA damage repair,³⁰ hence, leading to senescence. Histone deacetylases (HDACs) are enzymes that modify chromatin organization by removing acetyl groups from lysine residues on histone tails. Specifically, HDAC1/2 has been identified as an important participant in the DNA-damage response.³¹ Our group previously reported that pharmacological inhibition of HDAC1/2 and doxycycline-induced podocyte-specific Hdac1/2 knockout mice in adulthood improved proteinuria and glomerulosclerosis after podocyte injury.³² To further determine whether germline podocyte-specific loss of Hdac1/2 would mirror the adult phenotype, Nphs2-Cre Hdac1/2 knockout mice were generated. Unexpectedly, these mutant mice developed proteinuria and histological features suggestive of collapsing glomerulopathy (CG). These findings prompted us to further examine the role of Hdac1/2 in podocyte loss.

Ultrastructural examination of Hdac1/2 knockout podocytes demonstrated foot process effacement, and lipofuscin droplet accumulation within the podocyte cell body, a feature suggestive of cellular senescence.³³ These findings were further corroborated by senescence-associated β-galactosidase (SA-β-gal) activity, a well-known hallmark of senescence.^34–36 We further determined that sustained DNA damage, likely caused by epigenetic alterations such as open chromatin and deficient DNA damage repair, leads to senescence. The absence of Hdac1/2 not only resulted in podocyte re-entry into the cell-cycle as studied with the FUCCI technology³⁷ (i.e., Fluorescence-Ubiquitination-based Cell-Cycle Indicator) and Ki67 expression but also produced p21-mediated cell-cycle arrest in G1, resulting in senescent podocytes being observed in urine. Through SASP, senescent podocytes were found to secrete proinflammatory cytokines, growth factors, and matrix metalloproteases (MMPs) that contribute to their detachment and loss. Collectively, our results indicate that an intact Hdac1/2 in podocytes is necessary during development, as loss of these genes results in podocyte senescence and podocytopenia.

Methods

Antibodies

Mouse anti-Wilms tumor 1 (anti-WT-1; Novus Biologicals, catalog: NB11-60011); rabbit anti-WT-1 (Abcam, ab89901); mouse anti-HDAC1 (Cell Signaling Technology, 5356S); rabbit anti-HDAC2 (Abcam, ab16032); guinea pig anti-nephrin (Progen, GP-N2); rabbit anti-p21 (Abcam, ab188224); mouse anti-phospho-ataxia telangiectasia mutated (ATM; Millipore Sigma, 05–740); mouse anti-p53 (Cell Signaling Technology, 2524S); mouse anti-Cyclin D3 (Cell Signaling Technology, 2936S); mouse anti-S319-phosphorylated histone H2AX (γH2AX; Millipore Sigma, 05-636); rat anti-mCherry (Invitrogen, M11217); goat anti-mVenus (MyBioSource, MBS448126); mouse anti-p16 (Abcam, ab189034); mouse anti-Ki67 (Cell Signaling Technology, 8D5); rabbit anti-GAPDH (Cell Signaling Technology, 2118S); Alexa Fluor 488 goat anti-mouse IgG antibody (Invitrogen, A11029); Alexa Fluor 488 goat anti-rabbit IgG antibody (Invitrogen, A11034); Alexa Fluor 488 donkey anti-goat IgG antibody (Invitrogen, A11055); Alexa Fluor 594 goat anti-rabbit IgG antibody (Invitrogen, A21207); Alexa Fluor 594 goat anti-mouse (Invitrogen, A11032); Alexa Fluor 594 goat anti-guinea pig (Invitrogen, A11076); Alexa Fluor 594 goat anti-rat (Invitrogen, A11007); Alexa Fluor 647 goat anti-rabbit (Invitrogen, A21245); and Alexa Fluor 647 goat anti-rabbit (Invitrogen, A21235) were purchased commercially.

Generation of Mice

For selective deletion of Hdac1/2 in glomerular podocytes, Hdac1^fl/fl and Hdac2^fl/fl mice were obtained as a gift from Dr. Eric Olson, UT Southwestern at Dallas, and crossed with Nphs2-Cre (Pod-Cre) and Rosa-Diphtheria toxin receptor, Rosa-Dtr^fl/fl, mice on a BALB/c background. Podocyte-specific Hdac1/2 double KO (DKO) mice were obtained by breeding Hdac1^fl/fl Hdac2^+/fl Pod-Cre Rosa-DTR^fl/fl mice. DKO mice and littermate Hdac1^fl/fl Hdac2^+/fl Pod-Cre Rosa-DTR^fl/fl controls were used for the experiments. R26Fucci2aR mice on a BALB/c background were obtained from Pr. Ian James Jackson (University of Edinburgh, UK) and mated with Hdac1^fl/fl Hdac2^+/fl Pod-Cre Rosa-DTR^fl/fl mice to generate R26Fucci2aR Hdac1^fl/fl Hdac2^+/fl Pod-Cre Rosa-DTR^fl/fl mice. Tail genotyping was performed by PCR using previously described protocols.^32,37–39

Biochemical Measurements Plasma Creatinine, Urine Albumin, and Urine Creatinine

Urine samples were collected from the DKO and littermate controls. Urine albumin levels were measured in duplicate using an albumin ELISA quantitation kit according to the manufacturer's protocol (Bethyl Laboratories Inc), and the absorbance read at 450 nm (Synergy LX Multi-Mode Reader, BioTek) as previously described.^40–43 Urine and plasma creatinine were measured in duplicate for each sample with an ELISA quantitation kit (Bioassay Systems) at an absorbance of 490 nm.

Cell Culture

Isolation of primary podocytes from P14 DKO and littermate control pups were performed, as previously described.^39,40 Briefly, glomerular cells were seeded on type-1 collagen (Corning, 354231)-coated cell-culture dishes and cultured in RPMI 1640 medium (Gibco, 11875093), supplemented with 10% FBS (Gibco, 10082147), 100 U/ml penicillin, 100 µg/ml streptomycin (Gibco, 15140163), 100 mM HEPES (Gibco, 15630080), 1 mM sodium pyruvate (Gibco, 11360070) and 1 mM sodium bicarbonate in a humified 5% carbon dioxide (CO₂) incubator. Two days after the cells were seeded, the medium, including 0.1 µg/ml diphtheria toxin (Sigma-Aldrich, D0564), was changed every other day. Subculture was performed by detaching podocytes with 0.05% trypsin-EDTA (Gibco, 25300054) and cultured on collagen type1-coated dishes.

Histology and Immunofluorescence Staining

Mice were anesthetized by intraperitoneal injection of ketamine and xylazine followed by perfusion fixation with 4% paraformaldehyde through the left ventricle, either with or without glutaraldehyde, for histology and immunofluorescence or transmission electron microscopy (TEM), respectively. For histology, sections were sent to Yale Pathology Core Tissue Services for hematoxylin and eosin (H&E), periodic acid-Schiff (PAS), and Masson's trichrome and Jones' Methenamine Silver staining. To evaluate collapsed glomeruli, glomerulosclerosis, and interstitial fibrosis, kidney sections were assessed as previously described.^40,41,43 Briefly, 15 glomeruli from each specimen were assessed on H&E-stained sections, and the degree of glomerular collapse was quantified as the percentage of affected glomeruli per total number of glomeruli. The severity of glomerulosclerosis in each glomerulus was semiquantitatively scored on PAS-stained sections as follows: 0, no sclerosis; 1, sclerosis of <10% of glomeruli; 2, sclerosis of 10%–25% of the glomeruli; 3, sclerosis of 25%–50% of the glomeruli; and 4, sclerosis of >50% of glomeruli. To evaluate interstitial fibrosis, 15 fields of each section were examined on Masson's trichrome and semiquantitative analysis of each field was assessed as follows: 0, no fibrosis; 1, fibrosis of <10% of areas; 2, fibrosis of 10%–25% areas; 3, fibrosis of 25%–50% of areas; and 4, fibrosis of >50% of areas. For quantitative ultrastructural analysis of the glomerulus by TEM, the number of podocyte foot processes present in each micrograph was divided by the total length of the glomerular basement membrane (GBM) to calculate the mean density of podocyte foot processes. Thickness of GBM was quantified by NIH ImageJ software (version 1.52a). For immunofluorescence, kidney cryosections (4 µm) were subjected to antigen retrieval at 95°C for 10 minutes in 10 mM sodium citrate buffer (pH 6.0), followed by blocking with 3% BSA in PBS for 1 hour at room temperature (RT). Immunostaining of cryosections was performed as follows: slides were incubated with the appropriate primary antibodies overnight at 4°C, followed by incubation with Alexa Fluor 488, 594, and/or 691-conjugated secondary antibodies at RT for 1 hour, and three times washing with PBS. Then, coverslips were mounted with mounting medium antifade with or without 4′,6-diamidino-2-phenylindole (DAPI) if triple immunofluorescence staining. For in vitro experiments, cultured cells were fixed with 4% paraformaldehyde for 10 minutes at room temperature, washed with PBS, then incubated for 10 minutes in 0.1% Triton X-100 in PBS, followed by blocking with 5% BSA for 30 minutes. Cells were then incubated with primary antibodies at 4°C overnight, followed by incubation with the appropriate secondary antibodies for 1 hour at RT, washed, and mounted with mounting medium antifade with DAPI. Images were taken using an Andor CSU-WDi spinning disk confocal microscope equipped with a Nikon Ti-E CFI Plan Apochromat Lambda 60× oil immersion objective for immunofluorescence analysis, and images were processed using ImageJ software or Adobe Photoshop S 2018. For quantification of podocyte density, 20 glomeruli were evaluated for each section by counting the WT-1 positive nuclei in each glomerulus in relation to the volume of the disc of glomerular tuft as previously reported.⁴⁴

Immunoblotting

Primary podocytes were lysed in lysis buffer containing 50 mM Tris-hydrochloride (pH 7.6), 500 mM sodium chloride, 0.1% SDS, 0.5% deoxycholate, 1% Triton X-100, 0.5 mM magnesium chloride, phosphatase inhibitor cocktail, and protease inhibitor cocktail (Roche, 11697498001). Protein concentrations were quantified using the Pierce BCA Protein Assay Kit (Thermo Fisher, 23225). Equal amounts of podocyte lysates were separated by SDS-PAGE and transferred to polyvinylidene difluoride membranes (EMD Millipore). The membrane was blocked with 5% nonfat milk (American Bio) or 5% BSA (Sigma-Aldrich) in Tris-buffered saline and Tween 20, the appropriate peroxidase-labeled anti-IgG secondary antibody (EMD Millipore) was added, and signals were detected using enhanced chemiluminescence reagents (Bio-Rad) and exposed with Odyssey (LI-COR Biosciences). For quantification, densitometry was performed using ImageJ software.

Lipofuscin Staining

The Sudan-Black B (SBB) histochemical stain was used to detect lipofuscin as previously described.^33,45 Briefly, a solution of 0.7% w/v of SBB (Sigma-Aldrich, 199664) was prepared in 70% ethanol, stirred overnight, and filtered through filter paper. Primary podocytes and snap-frozen mouse kidney cryosections were fixed in 1% (wt/vol) formaldehyde/PBS for 5 minutes and washed with PBS. Kidney sections and primary podocytes were dehydrated for 5 minutes with 50% and 70% ethanol, then incubated with 20 µl of the freshly prepared SBB for 8 minutes, washed with 50% ethanol and water, before being counterstained with nuclear fast red and analyzed by light microscopy. Human kidney tissue sections were deparaffinized with xylene, 100% ethanol, 95% ethanol, and 70% ethanol previous SSB.

SA-β-Galactosidase Activity Staining

For detection of SA-β-gal activity, primary podocytes or snap-frozen cryosections in optimal cutting temperature compound of mouse kidneys were processed with an SA-β-gal staining kit (Cell Signaling Technology, 9860) in accordance with the manufacturer's protocol. Counterstaining of kidney sections was performed with nuclear fast red. The percentage of podocytes with SA-β-gal activity was quantified as the blue-positive cells per total number of cells in a microscopic field visualized through a 20× objective lens.

Glomerular RNA Isolation and RNAseq Analysis

Glomeruli were isolated in control and DKO mice at 25 days old. Samples were prepared in duplicate, and each sample contained the glomeruli of two mice. Glomeruli were isolated from kidney cortical tissue with 45% Percoll solution, as previously described.^46,47 Glomerular RNA was extracted using the RNeasy kit (Qiagen) and sent to the Yale Center for Genome Analysis. The raw TRAPseq fastq files were processed using fastp tool (version 0.20). With a default setting, sequencing reads with low-quality bases were trimmed or filtered. Alignment was performed for cleaned reads using STAR (version 2.7.9)⁴⁸ and mouse reference genome (gencode version GRCm38.p6 with vM25 gene annotation). Expression quantification was performed for alignment results using featureCounts (version 2.0.0).⁴⁹ Genes with low expression levels (< six read counts in at least 20% samples) that likely represent noise, were excluded before downstream analysis. The filtered read counts matrix was then normalized by the transcripts per million methods. Detection of differentially expressed genes (DEGs) was performed using R package DESeq2 (version 1.30.1)⁵⁰ and The Benjamin-Hochberg procedure was used for multiple test correction, and FDR ≤0.05 as a threshold for detection of DEGs.

Quantitative PCR Analysis

Total RNA was extracted from the primary podocytes from control and DKO mice by using RNeasy kit (Qiagen). The RNA was measured by spectrophotometry (Nanodrop Technologies), and 2 µg of total RNA was used for reverse transcription by a High-Capacity cDNA Reverse Transcription Kit (Applied Biosystems, 4368814) according to the manufacture's instruction. The quantitative PCR (qPCR) amplifications were performed using Power SYBR Green PCR Master Mix (Applied Biosystems) with a 7300 Real-Time PCR machine.

Comet Assay in Neutral Conditions

For in vitro detection and quantification of DNA double-strand breaks (DSBs), primary podocytes from DKO and control mice were trypsinized, counted using a hemacytometer, and performed as previously described.⁵¹

Urinary Podocytes

Aliquots of freshly collected urine from DKO and control mice were sedimented on type I-collagen-coated coverslips (18 mm diameter) contained in 24-well plates and air-dried for 30 minutes. For immunofluorescence, the samples were fixed with 4% paraformaldehyde for 10 minutes at RT, washed with PBS, then permeabilized with 0.1% Triton X-100 and blocked with 5% BSA as aforementioned. Samples were then incubated with rabbit anti-WT-1 antibody at 4°C overnight, followed by incubation with the appropriate secondary antibody for 1 hour at RT, mounted with antifade mounting medium with DAPI, and analyzed by confocal microscopy. For lipofuscin detection, samples were fixed in 1% formaldehyde and incubated in 70% ethanol before staining with SSB as previously described.

Cell Count with Crystal Violet

Primary DKO and control podocytes were trypsinized, counted using a hemacytometer, and seeded with equal amounts of cells (6 × 10³ per well) on 96-well plates coated with collagen for 3 days. After the removal of nonviable cells by washing them with PBS three times, the viable cells were fixed with 95% ethanol. Podocytes were stained with 0.1% crystal violet for 15 minutes at RT, washed with water, and then lysed with 1% SDS. The absorbance was measured at 595 nm. For MMP2/9 inhibition experiments, the MMP2/9 inhibitor (Abcam, ab145190) was added at 5 µM in the serum-free medium after seeding and the viability was counted after 3 days as aforementioned.

Zymography

The enzymatic activity of the in vitro secreted MMP2/9 was assessed using gelatin zymography.^52,53 Briefly, primary DKO and control podocytes were washed with PBS, placed in serum-free media for 16 hours, and media were collected and centrifugated at 400g for 5 minutes at 4°C to remove cells and debris. For zymography, 25 µl media (containing approximately 25 µg/µl protein as quantified by BCA Protein Assay Kit) was mixed with 4X Laemmli sample buffer (Bio-Rad, 1610747) and separated by 7.5% polyacrylamide SDS gel containing 1 mg/ml gelatin. After electrophoresis, the gel was washed twice with washing buffer (2.5% Triton X-100, 50 mM Tris HCl, 5 mM CaCl₂, 1 µM ZnCl₂ in water, pH 7.5) for 30 minutes at RT, equilibrated in developing buffer (1% Triton X-100, 50 mM Tris HCl, 5 mM CaCl₂, 1 µM ZnCl₂ in water, pH 7.5) for 10 minutes at 37°C, followed by incubation in developing buffer at 37°C for 18 hours. The gel was stained with 0.2% Coomassie in water for 1 hour at RT and incubated in destaining solution (50 ml methanol, 80 ml acetic acid, and 870 distilled water) until clear bands were visualized.

Statistical Analysis

All data are represented as mean±SEM. The number of replicates for each experiment is shown in the figure legends. Statistical analysis was performed by GraphPad Prism 9 software using two-tailed t test or one-way ANOVA. Statistical significance was determined with P<0.05.

Study Approval

The Committee on the Use and Care of Animals Institutional Review Board at Yale University approved all animal experiments. All work was carried out in accordance with the principles and procedures outlined in the NIH Guide for the Care and Use of Laboratory Animals (2011).

Results

Loss of Podocyte Hdac1/2 Results in Severe Proteinuria and Kidney Failure

To examine the importance of Hdac1/2 in podocytes, we generated podocyte-specific Hdac1/2 DKO mice by mating Hdac1^fl/fl and Hdac2^fl/fl with Nphs2 (Pod)-Cre Rosa-DTR ^fl/fl mice. DKO mice were obtained by breeding Hdac1^fl/fl Hdac2^+/fl Pod-Cre Rosa-DTR^fl/fl and identified by tail genotyping for the appropriate genes (Supplemental Figure 1a). To ensure a pure population of primary podocytes, we leveraged the loxP sites flanked diphtheria toxin-receptor (DTR) of our Rosa-DTR^fl/fl mice, where podocytes are spared cell death after diphtheria toxin administration while eliminating all non-Cre expressing kidney cells.⁵⁴ Lack of immunoreactivity was observed for podocyte Hdac1/2, when co-labeled with WT-1, a podocyte-specific marker by immunofluorescence and by western blot (Supplemental Figure 1b–1d). However, Hdac1/2 remains expressed in adulthood when examining podocytes in glomerular sections from control mice (Supplemental Figure 1e).

DKO mice had no observable phenotype at birth, as demonstrated by WT-1 immunostaining and podocyte density (Supplemental Figure 2a–c). However, 3 weeks postnatally, compared with littermate controls, the mutant mice demonstrated mesangial expansion, proteinaceous casts, and mild proteinuria (Supplemental Figure 3a–c), which further progressed by 4–5 weeks with failure to gain weight (Figure 1, A and B) and 100% death before 6 weeks of age (Figure 1C). Hdac1^+/fl Hdac2^fl/fl control mice were also generated and displayed no overt phenotype after 12 months postnatal (Figure 1A, Supplemental Figure 4), indicating that loss of both Hdac1 and Hdac2 are required in podocytes to manifest a kidney phenotype.

Figure 1.

Loss of podocyte-associated Hdac1/2 results in severe proteinuria, kidney failure, and collapsing glomerulopathy. (A) Podocyte-specific Hdac1/2 DKO mice fail to gain weight by 5 weeks of age (red) relative to controls Hdac1^fl/fl Hdac2^+/fl littermate (blue) and Hdac1^+/fl Hdac2^fl/fl (green). n=6 mice. (B) Representative image of control Hdac1^fl/fl Hdac2^+/fl littermate (Ctrl) and DKO mice at 5 weeks of age. (C) Survival curve of DKO mice (red) demonstrates 100% death by 6 weeks of age. n=6 mice. (D) Quantification of urinary albumin normalized to creatinine in control and DKO mice at 5 weeks of age. n=6 mice, *P<0.01. (E) Plasma creatinine in control and DKO mice at 5 weeks of age. n=6 mice. *P<0.01. (F) Representative light microscopic images (H&E, PAS, and Trichrome) of a single glomerulus from control and DKO mice at 5 weeks of age. Scale bar, 10 µm. (G) Representative light microscopic images (H&E, PAS, and Trichrome) of kidney interstitium from control and DKO mice at 5 weeks of age. Scale bar, 50 µm. (H) Quantification of the percentage of collapsed glomeruli per total number of glomeruli in (F). n=15 glomeruli from three different mice, *P<0.01. (I) Quantification of glomerulosclerosis in (E). n=15 glomeruli from three different mice, *P<0.01. (J) Quantification of interstitial fibrosis in (F). n=15 glomeruli from three different mice, *P<0.01. Alb, albumin; Cr, urinary creatinine.

Measured albumin/creatinine ratio by ELISA, demonstrated robust albuminuria (Figure 1D) and elevated plasma creatinine (Figure 1E), whereas histological examination of DKO kidneys by H&E, PAS, and Trichrome staining revealed that approximately 50% of glomeruli had undergone global or segmental collapse and glomerulosclerosis (Figure 1F; quantification Figure 1, H and I). Moreover, we found severe interstitial fibrosis, tubular dilatation, and proteinaceous casts (Figure 1G; quantification Figure 1J), suggestive of end-stage kidney disease.

Loss of Podocyte Hdac1/2 Results in Collapsing Glomerulopathy, Foot Process Effacement, and Lipofuscin Accumulation

To confirm whether the observed phenotype corresponds to CG, Jones' silver staining was performed, confirming glomerular tuft collapse with marked proliferation of parietal epithelium (Figure 2, A and B), which characterizes CG.⁵⁵ TEM demonstrated dramatic foot process effacement and modest increases in GBM thickness in DKO mice (Figure 2C, quantification Figure 2D). Intriguingly, podocyte loss of Hdac1/2 resulted in hypertrophy, quantified as a 1.5-fold increase in cell-body area (Figure 2E), and a striking presence of cytoplasmic hypodense regions, suggestive of lipofuscin accumulation.

Figure 2.

Loss of podocyte Hdac1/2 results in collapsing glomerulopathy, severe foot process effacement, and lipofuscin accumulation. (A) Representative light microscope image of Jones' silver staining of a single glomerulus from control and DKO mice at 5 weeks of age. The arrow indicated marked proliferating parietal epithelium. Scale bar, 10 µm. (B) Representative light images of Jones' silver staining of kidney interstitium from control and DKO mice at 5 weeks of age. Scale bar, 50 µm. (C) Transmission electron micrographs illustrating foot process effacement by 5 weeks of age. Scale bar, 1.0 µm. (D) Quantification of the number of foot processes per micrometer of GBM and the GBM thickness in the control and DKO glomeruli. n=12 glomerular capillaries, from three different mice, *P<0.01. (E) High magnification micrographs from control and DKO podocytes showing accumulation of lipofuscin as indicated by the arrowheads. DKO podocytes exhibited a 1.5-fold increase in cellular body area as quantified by ImageJ (Control podocyte: 26.7 µm² cell body, nucleus 15.7 µm²; DKO podocyte: 42.4 µm² cell body, nucleus 14.4 µm²). Scale bar, 1 µm.

Loss of Podocyte Hdac1/2 Results in Podocyte Senescence

To confirm lipofuscin accumulation in DKO podocytes, we further stained kidney sections with the lipofuscin-specific histological agent, SBB. Micrographs revealed robust presence of lipofuscin staining in the perimeter of DKO glomeruli, which was absent in control glomeruli (Figure 3A).

Figure 3.

Loss of podocyte Hdac1/2 results in cellular senescence. Representative images of (A) SBB staining and (B) SA-β-gal activity in glomeruli from control and DKO mice at 5 weeks of age. Lipofuscin is stained as positive SBB granules within senescent cells in (A) and blue coloration represents cellular senescence in SA-β-gal staining in (B). Counterstain: nuclear fast red. Scale bar, 10 µm. (C) Representative images of SBB staining in primary control and DKO podocytes. Lipofuscin stained as positive SBB granules in DKO cells. Scale bar, 2 µm. (D) Representative images of SA-β-gal staining in primary control and DKO podocytes. Scale bar, 5 µm. (E) Quantification of the percentage of podocytes positive for SA-β-gal per total number of podocytes in (D). n=9, *P<0.01. (F) Representative images of SBB staining in human glomeruli from healthy and biopsy-proven FSGS. The arrow indicates lipofuscin accumulation in a collapsed glomerulus. Scale bar, 20 µm. (G) Quantification of lipofuscin accumulation in (F). n=10 glomeruli, *P<0.01. (H) Immunofluorescence images of p21 and WT-1 in control and DKO podocytes in vivo and in vitro. Scale bar, 5 µm. (I) Immunoblot images of p21 and GAPDH as a loading control in control and DKO mice primary podocytes. (J) Quantification of p21 immunoblots in (G). n=3, *P<0.01. FSGS, focal segmental glomerulosclerosis.

Considering these results, we sought to investigate whether DKO podocytes present other classic features of cellular senescence. Expectedly, DKO mice glomeruli demonstrated SA-β-gal activity, the hallmark senescence biomarker (Figure 3B). To further validate these findings, DKO primary podocytes showed a remarkable increase in cytoplasmic lipofuscin granules on SBB staining (Figure 3C), and increased SA-β-gal activity was observed in approximately 80% DKO podocytes (Figure 3D; quantification Figure 3E). Furthermore, SBB staining of healthy and diseased human kidney biopsies revealed lipofuscin accumulation on damaged glomeruli from patients with FSGS (Figure 3F, quantification Figure 3G), suggesting evidence of senescent cells in glomerular diseases.

To determine whether other key senescence regulators, p21 and p16,⁵⁶ play an essential role in our DKO podocytes, immunofluorescence was performed. A marked increase in nuclear p21 was observed in DKO podocytes in vivo and in vitro (Figure 3H), which was further confirmed by western blot (Figure 3I, quantification Figure 3J). However, no overt differences were observed for p16 (Supplemental Figure 5a–b). One explanation for these findings is that p53/p21 pathway is activated in response to DNA damage,^26,57 whereas the p16 is more related to age-associated senescence and telomerase-shortening.⁵⁸ Together, our findings suggest that deletion of Hdac1/2 results in podocyte senescence, mediated by p21 in response to DNA damage.

DNA Damage and DNA Damage Response Play a Role in Podocyte Senescence

Given the increased p21 podocyte expression and the reported connections between senescence and severe DNA damage, especially those caused by double-strand breaks (DSBs) in terminally differentiated cells, we studied whether the deletion of podocyte Hdac1/2 resulted in DSBs. We assessed the presence of phosphorylated histone H2AX (γH2AX), a sensitive molecular marker of DSBs.⁵⁹ Through immunofluorescence, we observed an increased podocyte nuclear and cytosolic γH2AX staining in DKO glomeruli when compared with control (Figure 4A, quantification Supplemental Figure 6a).

Figure 4.

Loss of podocyte Hdac1/Hdac2 results in DNA damage and activates DNA damage. Representative immunofluorescence images of (A) γH2AX, (B) phospho-ATM, (C) p53, and (D) cyclin D and WT-1 in control and DKO mice glomeruli at 5 weeks of age. Scale bar, 10 µm. Representative immunofluorescence images of (E) γH2AX, (F) phospho-ATM, (G) p53, and (H) cyclin D and WT-1 in primary control and DKO podocytes. Scale bar, 5 µm. (I) Immunoblot images of p53, cyclin D, and GAPDH as a loading control in control and DKO mice primary podocytes. (J) Quantification of immunoblots in (I). n=3, *P<0.01.

Furthermore, we sought to determine the pathway that activates podocyte senescence after DNA damage. In response to DSBs, the DNA damage repair is initiated by ATM, which activates downstream effector protein p53.^60,61 Podocyte expression and nuclear localization of phosphorylated-ATM and p53 were confirmed by immunofluorescence in DKO mice glomeruli (Figure 4, B and C, quantification Supplemental Figure 6b and c). Conversely, in littermate control mice, DNA damage response (DDR) was not observed. It has been shown that in postmitotic cells, DDR is involved in the activation of cell-cycle to induce DSBs repair and survival, or its failure results in cellular death.²⁸ To investigate whether podocytes activate the cell-cycle on DNA damage, we assessed the expression of cyclin D, which is synthesized during the G1 phase of the cell cycle.⁶² Immunofluorescence in DKO kidneys revealed increased cyclin D expression in podocyte nuclei (Figure 4D, quantification Supplemental Figure 6d), suggesting cell-cycle activation. In agreement with in vivo results, primary podocytes lacking Hdac1/2 also revealed γH2AX expression (Figure 4E). Furthermore, to validate DNA damage in primary DKO podocytes, we used the comet assay under neutral conditions.⁵¹ DKO podocytes exhibited a distinct comet pattern, which was absent in controls (Supplemental Figure 6e, quantification Supplemental Figure 6f), thus confirming podocyte DNA damage.

Using immunostaining, we also observed a dramatic increase of phosphorylated-ATM, nuclear p53 and cyclin D (Figure 4, F–H, quantification Supplemental Figure 6a–d) in primary DKO podocytes in comparison with controls, along with increased protein expression levels for p53 and cyclin D by immunoblotting (Figure 4I, quantification Figure 4J), indicating that activated DDR may play a role in podocyte senescence.

Deletion of Hdac1/2 in Podocytes Causes Loss of Quiescence, Cell-Cycle Entry and G1 Arrest

To confirm that DKO podocytes lose their quiescent phenotype and acquire a proliferative state, we assessed the immunoreactivity for Ki67, a biomarker that detects proliferating cells observed in CG⁵⁵ and cell-cycle stage G1, S, G2, and M, but not in G0.⁶³ DKO podocytes demonstrated Ki67 expression, confirming the proliferative phenotype (Figure 5A, quantification Figure 5C). Next, to study the cell-cycle stage of the DKO podocytes, we crossed the FUCCI2aR with our Hdac1^fl/fl Hdac2^+/fl Pod-Cre Rosa-DTR^fl/fl mice and generated specifically the FUCCI reporters in the DKO (FUCCI2aR-DKO) and the control (FUCCI2aR-Ctrl) podocytes. The FUCCI2aR system incorporates the fluorescent probes mCherry coupled to Cdt1 protein, which accumulates in G1 phase and colors the nuclei in red, and mVenus coupled to germinin, which accumulates in early S through the M phase, staining the nuclei in green. During the G1/S transition, the nucleus appears yellow, whereas in the quiescent state, G0 nuclei do not display any fluorescence.³⁷

Figure 5.

Loss of podocyte Hdac1/Hdac2 in podocytes induces Ki67 expression, cell-cycle entry, and G1 arrest. (A) Representative immunofluorescence of Ki67 and WT-1 in control and DKO glomeruli at 5 weeks of age. (B) Representative immunofluorescence images of mCherry (red), mVenus (green), and WT-1 (blue) in glomeruli from FUCCI2aR-Ctrl and FUCCI2aR-DKO mice at 5 weeks of age. mCherry accumulates during G1 and is lost during the G1/S transition as mVenus accumulates and peaks at G2/M.(C) Quantification of percentage of podocytes positive for Ki67 in (A). n=10 glomeruli, *P<0.01. (D) Distribution of podocyte in the cell cycle phases, in FUCCI2aR-Ctrl (blue) and FUCCI2aR-DKO (red) mice in (A). n=12 glomeruli. (E) Distribution of positive-γH2AX podocytes in the cell cycle in glomeruli from FUCCI2aR-DKO mice at 3 weeks and 5 weeks of age. n=7 glomeruli, *P<0.01. (F) Representative immunofluorescence images of (E), mCherry, mVenus, and γH2AX (blue). (G) Representative immunofluorescence images of mCherry, mVenus and p21 (blue) in a collapsed glomerulus from FUCCI2aR-DKO mice at 5 weeks of age. Scale bar, 10 µm.

For FUCCI2aR-DKO mice at 5 weeks postnatally, immunofluorescence micrographs revealed more than 40% red fluorescent podocytes co-labeled with WT-1, whereas no yellow or green cells were observed (Figure 5B, quantification Figure 5D). By contrast, approximately 90% of podocytes were not in the cell-cycle in FUCCI2aR-Ctrl. However, we observed periodic red fluorescent cells, suggesting that terminally differentiated cells may reversibly and temporarily re-enter into G1 phase to repair DNA damage during normal maintenance. Green-stained nuclear podocytes were sparse suggesting that very few podocytes reach the G2/M phase before eventual podocyte death.

Next, we sought to determine whether DNA damage precedes podocyte cell-cycle entry by leveraging our FUCCI2aR mice. We used a triple immunofluorescence staining for mVenus, mCherry, and γH2AX in DKO mice at 3 and 5 weeks of age. In line with our hypothesis, at 3 weeks several glomerular cells displayed DSBs, but only approximately 10% were FUCCI positive (Figure 5F, quantification Figure 5E); whereas at 5 weeks old, a higher percentage of cells displayed mCherry-positive fluorescence. These results indicate that DSBs likely occur before G1 phase and that the percentage of podocytes in the cell-cycle correlates with disease progression.

We next investigated whether cell-cycle arrest in the DKO podocytes occurred given that G2 staining was absent. Moreover, cellular senescence has been associated with occurring during cell-cycle arrest in G1.⁶⁴ Thus, we stained for p21 in FUCCI2aR-DKO glomeruli and observed colocalization with the red-fluorescent podocytes (Figure 5G), suggesting that podocytes that lack Hdac1/2, not only re-enter the cell-cycle on DSBs but also undergo senescence through cell-cycle arrest in G1.

Senescent Podocytes Secrete SASP Components

Hallmark of senescent cells includes altered metabolism and secretion of SASP elements that reinforce senescence in an autocrine/paracrine fashion and stimulates the recruitment of the immune system to eliminate injured cells.^29,65 To study this phenomenon in our DKO model, we performed glomerular RNA-seq profiling from control and DKO mice at 25 days old. Gene expression data were analyzed with fastp tool and the expression values in biological replicates were averaged (Supplemental Material S1). The genes with P<0.05 and higher than two-fold change were determined as DEGs. Our analysis identified 1193 DEGs, 715 upregulated and 478 downregulated in the DKO when compared with the control. The genes with the highest upregulated and downregulated are shown in the heatmap (Figure 6, A and B, respectively). Gene ontology and KEGG pathways analysis⁶⁶ identified inflammatory response, cytokine production, cell-cycle activation, and regulation of cell migration as the most significant enriched pathways in the upregulated DEGs (Figure 6C), whereas metabolic processes were identified as the significant enriched pathways from the downregulated DEGs (Figure 6D). In agreement with the literature, our findings suggest that the identified DEGs may represent glomerular SASP and the deregulated metabolism of senescent cells.

Figure 6.

RNA-seq analysis indicates that DKO glomeruli secret SASP components. (A and B) Heatmap representing color-coded highest differentially upregulated and downregulated glomerular genes analyzed by Z ratio in DKO compared with control mice at 25 days old. (C and D) Most significant enriched pathways in the upregulated and downregulated genes according to Gene Ontology and KEGG pathways. (E) Heatmap representing color-coded highest DEGs in the RNA-seq analysis that have been previously associated with the SASP. (F) Real-time PCR in primary DKO and control podocytes for the SASP-associated genes in (E) with a fold change higher than three. (G) Real-time PCR in primary DKO and control podocytes for Wt1. ****P<0.00001, ***P<0.001, **P<0.01, *P<0.05. (H) Gelatin zymography of media of primary control (lanes 1 and 2) and DKO podocytes (lanes 3 and 4).

Furthermore, the heterogeneity of SASP in various cell types and different senescence inducers have made it difficult to identify SASP as a single phenotype. Yet, there are proteomic databases, such as “SASP Atlas,” that includes a myriad of proteins elevated in senescent cells in vitro.⁶⁷ To give a greater validity to our findings, we compared our DEGs with the “SASP Atlas” and found that 132 DEGs have been associated with SASP (Supplemental Material 2). In the same investigation, the researchers also defined “core-SASP” elements, common to all senescence inducers, which include GDF15, SERPINs, and MMPs. Accordingly, we found these genes among our DEGs, with expression values four times higher in DKO glomeruli compared with the control. Moreover, to confirm that primary DKO podocytes express increased SASP elements, we performed qPCR for the SASP genes with a fold-change higher than 3.0 according to our RNA-seq data (Figure 6, E and F). In addition, we assessed the expression of Wt1 because in vivo results showed a decreased expression in podocyte markers such as Nphs2. Consistently, we observed an important reduction in DKO podocytes (Figure 6G).

Finally, to validate that podocytes secrete SASP elements, we performed zymography by examining the enzymatic activity of MMP-2 and MMP-9, type IV collagenases, given that collagen IV is a major component of the GBM. Furthermore, our RNA-seq data demonstrated increased Mmp14 expression (Figure 6F), a well-known modulator of MMP-2 activity.⁶⁸ Our results revealed higher MMP-2/9 activities in DKO in comparison with control (Figure 6H), suggesting that DKO podocytes express and secrete SASP components.

Podocyte Senescence Results in Podocytopenia

Because we observed that senescent podocytes secret SASP elements such as MMPs that can facilitate their clearance and detachment from the GBM, we probed whether the senescent phenotype results in podocyte loss. Crystal violet staining showed a reduction in cell number in DKO primary podocytes compared with control (Figure 7A) and was improved on MMP-2/9 inhibition, demonstrating the role of SASP in podocyte loss.

Figure 7.

Podocyte senescence results in podocytopenia. (A) Quantification and representative images of podocyte cell number by crystal violet in primary control and DKO podocytes after three days of plating in the presence or absence of MMP2/9 (MMPi). n=9, *P<0.01. (B) Podocyte density, quantified from the number of WT-1–stained nuclei in glomeruli from control and DKO mice at 3 weeks and 5 weeks of age. n=20 glomeruli from three different mice, **P<0.01, *P<0.05. (C) Immunofluorescence images of WT-1 and DAPI (blue) in urinary podocytes from DKO mice. Scale bar, 10 µm. (D) Representative image of lipofuscin staining by SBB in urinary podocytes from DKO mice. Scale bar, 5 µm.

Accordingly, DKO mice presented a reduced podocyte number, quantified by podocyte density (Figure 7B). Compared with controls, DKO glomeruli stained with WT-1 revealed a mild reduction in podocyte number by 3 weeks of age, which was more pronounced at 5 weeks. To evaluate whether apoptosis contributes to podocyte death, we performed TUNEL assay (Supplemental Figure S7). We did not detect positive staining in glomerular cells suggesting that DKO podocytes do not die by apoptosis.

Compared with controls (Supplemental Figure S8), positive WT-1 and DAPI cells were observed in urine samples from DKO mice at 5 weeks by immunofluorescence (Figure 7C). Furthermore, lipofuscin-positive cells were observed in DKO urine samples on SSB staining (Figure 7D), suggestive of senescent cells. Collectively, these findings suggest that podocyte senescence results in podocytopenia, and SASP elements, such as MMPs, represent a mechanism contributing to the podocyte detachment and loss.

Discussion

Cellular senescence has been postulated to play a vital role in health and disease after first being described in 1961 by Hayflick.^25,57,69 Recent studies also suggest that podocytes can also undergo age-related senescence.^21,22,24,70 To build on this framework, we found that in mice, germline loss of podocyte-associated Hdac1/2 results in podocyte senescence, as demonstrated through SA-β-gal, p21 expression, and lipofuscin staining. Although lipofuscin accumulation has been observed in senescence-associated illnesses such as end-stage of heart failure,⁷¹ Alzheimer's, or Parkinson's disease,^26,72 to our knowledge, this is the first report providing evidence for lipofuscin accumulation in damaged podocytes. This novel observation may provide new insights into how damaged podocytes behave before their loss secondary to detachment.

Chromatin reorganization plays a key role in senescence, although the precise mechanism by, which epigenetic alterations are involved in senescence is still unknown.^73,74 Moreover, several studies suggest a progressive transition from an open chromatin configuration in dedifferentiated cellular stages to a more compact state in differentiated/quiescent cells. Indeed, one of the hallmarks of quiescent cells is highly condensed chromatin.^75–77 Thus, deletion of Hdac1/2 may impede podocyte quiescence by maintaining an open chromatin, which in turn results in more susceptibility to DNA breaks and prevents successful DNA damage repair, leading to senescence. Many investigations have associated the inhibition of HDAC1/2 with prolonged DNA damage due to the inability to successfully repair DNA.^31,75–79 The combination of these factors is reflected by (1) the presence of DSBs and DDR, as demonstrated by γH2AX staining and activation of the ATM-p53 pathway, and (2) the dysregulation of the podocyte quiescent phenotype (i.e., loss of WT-1 expression) and re-expression of proliferative markers (i.e., Ki67), which have been described as important features in CG.⁵⁵ These findings together suggest that epigenetic alterations may be at the origin of DNA damage-induced senescence and thus, the phenotype observed.

Our group had previously reported that podocyte-specific Hdac1/2 loss in an adult mice model, or pharmacologically by valproic acid, mitigated proteinuria and glomerulosclerosis.³² Compared with our germline model where HDACs are indispensable in attaining podocyte quiescence as the podocyte matures postnatally until 2 weeks,⁸⁰ Hdac1/2 deletion in adulthood, where highly condensed chromatin already exists, may not have detrimental effects. Instead, the deletion of Hdac1/2 activities may facilitate processes that required a more relaxed chromatin such as DNA damage repair.⁸¹ We do not rule out the possibility that in adulthood, HDACs may deacetylate nonhistone proteins instead of acting directly on chromatin. Certainly, further studies will be needed to identify HDAC1/2 substrates on podocyte health.

Podocytes, as ones of the most metabolically active cells, are highly susceptible to harm.⁸² It has been postulated that on injury, podocytes re-enter the cell-cycle, resulting in podocyte loss and progression toward glomerular disease.⁸³ In this study, we suggest that re-entry into the cell-cycle may be important, where injured podocytes may temporarily and reversibly re-enter because some DNA repair pathways are only active in the G1 phase.²⁸ Accordingly, by using the FUCCI system, we observed that in some glomeruli of our healthy model (FUCCI2aR-Ctrl), 5%–10% of podocytes were in the G1 phase. A similar population of podocytes in G1 (3%–6%) was also reported using the FUCCI technology in control.⁸⁴ Interestingly, research strongly supports the existence of a single switching point in G1, the restriction point, which regulates either the return to G0 or the “entry” into a new cell-cycle in which cells are committed to mitosis.⁸⁵

Our Hdac1/2-deprived podocytes, enter the cell-cycle and seem to progress to late G1 as immunofluorescence reveals intense-red fluorescence by the FUCCI technology and increased cyclin D expression.⁸⁶ Because yellow (G1/S) or green (G2/M) podocytes were not observed within the FUCCI system, we presumed a cell-cycle arrest in G1, driven by p21. Indeed, senescence is frequently defined as a cell-cycle arrest in G1,^64,87,88 and p21 expression has been associated with linking DDR by the ATM-p53 pathway to cell-cycle arrest.^60,61,89 The likely role of p21 is to prevent damaged DNA to be duplicated in the S-phase.⁹⁰ Thus, p21 expression strengthens the evidence that DNA damage is a mediator of podocyte-senescence in our model. Interestingly, p21 expression in podocytes has been associated with CG,⁹¹ but whether its expression relates to DNA-damage-mediated senescence requires further investigation.

Furthermore, podocyte senescence results in podocytopenia and urinary podocytes were identified by positive WT-1 staining in our DKO mice, along with the presence of urinary SBB-stained cells that likely indicate the senescent podocytes, which have detached from the GBM. Accordingly, we demonstrated by RNA-seq profiling, qPCR, and gelatin zymography that glomeruli and podocytes from DKO mice feature an altered metabolism and develop a SASP phenotype consisting of extracellular matrix-degrading proteins, growth factors, and proinflammatory molecules that not only affect their environment and damage the neighboring glomerular cells but also attract the immune system to eliminate them.⁵⁵ Activation of the immune system has been reported as a major pathway in CG, where glomerular and tubular damage is secondary to released cytokines.⁵⁵ Furthermore, the expression of inflammatory genes that lead to senescence can be activated by sustained and cytosolic DNA damage through the cGAS-STING pathway.⁹² Further investigations on these pathways in podocyte senescence will be valuable.

Moreover, recent studies about age-related senescence in podocytes have also stated an impaired metabolism and activation of inflammatory pathways.^21–24,70 Indeed, the aging process has been associated with altered epigenetic mechanisms, dedifferentiation processes, and accumulation of unrepaired DSBs,^93,94 suggesting that podocyte senescence may have similar mechanisms during disease and aging but develop different SASP elements that lead to divergent phenotypes.

In summary, the phenotype observed on podocyte-ablation of Hdac1/2 results in severe proteinuria, and histologic features of CG because of dysregulation of the podocyte quiescent phenotype through DNA damage and cell-cycle entry that mediates podocytes senescence-induced podocytopenia through the SASP (Figure 8). Nevertheless, important questions remain including (1) the mechanism of how podocytes repair DNA damage to return the cell to its quiescent state, (2) what level of podocyte injury or damage results in “a point of no return” marking irreparability, (3) whether epigenetic factors and DNA damage, caused by viral proteins or medication, are at the origin of CG, and (4) the mechanism of how the immune system and other SASP elements result in podocyte loss. Our study provides evidence that terminally differentiated cells, such as podocytes, re-enter cell-cycle after injury and undergo senescence, opening further studies to define mechanisms that will translate into future therapeutic targets to maintain the integrity of the glomerular filtration barrier.

Figure 8.

Cartoon schematic showing that loss of Hdac1/2 in podocytes results in DNA damage and senescence. Loss of Hdac1/2 in podocytes results in sustained DNA double-stranded-breaks (γH2AX), likely due to epigenetic alterations, i.e., open chromatin, and activation of the DNA-damage response, mediated by ATM phosphorylation. The DNA damage response causes podocyte cell-cycle entry, arrest in G1 mediated by the p53-p21 pathway, and podocyte senescence, characterized by lipofuscin aggregates and SA-β-gal activity. Senescent podocytes secrete matrix metalloproteinases that may contribute to the detachment of podocytes.

Supplementary Material

jasn-34-433-s001.pdf ^{(5.1MB, pdf)}

jasn-34-433-s002.xlsx ^{(6MB, xlsx)}

jasn-34-433-s003.docx ^{(56.4KB, docx)}

Disclosures

C.E. Pedigo is currently working at Goldfinch Bio, a late-stage pharmaceutical company. All work was performed during his tenure at Yale and has no relevance to his current position. C.E. Pedigo also reports Employment with Angion Biomedica Corp and Omega Therapeutics; and Ownership Interest: Angion Biomedica Corp. S. Ishibe reports Consultancy: Inxmed and Walden Pharmaceuticals; and Honoraria: InxMed and Walden Pharmaceuticals. S. Perincheri reports Research Funding: Paige AI. K. Inoue reports Speakers Bureau: Mochida Pharmaceutical Co., Ltd. All remaining authors have nothing to disclose.

Funding

This work was supported by George O'Brien Kidney Center at Yale School of Medicine grant P30DK078310 and National Institutes of Health grants DK083294 (to S. Ishibe) and DK093629 (to S. Ishibe).

Author Contributions

P.X. Medina Rangel and S. Ishibe conceptualized the study; J. Gu was responsible for data curation; H. Zhao was responsible for the formal analysis; S. Ishibe was responsible for the funding acquisition; P. Bunda, E. Cross, E. Gutiérrez-Calabrés, K. Inoue, G. Lerner, C. Liu, P.X. Medina Rangel, S. Nagata, C.E. Pedigo, S. Perincheri, A. Priyadarshini, X. Tian, and Y. Wang were responsible for the investigation; S. Ishibe and P.X. Medina Rangel reviewed and edited the manuscript; S. Ishibe and X. Tian were responsible for the supervision; and P.X. Medina Rangel wrote the original draft.

Data Sharing Statement

Original data reported in this study of type Spreadsheets/Data Sets have been deposited to Gene Expression Omnibus (GEO) accession no. GSE216088.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/JSN/A909, http://links.lww.com/JSN/A910, http://links.lww.com/JSN/A911.

This article contains the following supplemental material:

Supplemental Figure 1. Podocyte-specific deletion and expression of Hdac1/2 in germline and control mouse, respectively.

Supplemental Figure 2. DKO mice have no observable phenotype at birth.

Supplemental Figure 3. DKO mice manifested a phenotype at 3 weeks old that rapidly decline in kidney function.

Supplemental Figure 4. Hdac1^+/fl Hdac2^fl/fl Pod-Cre Rosa-DTR^fl/fl mice have no observable phenotype after 12 months of age.

Supplemental Figure 5. No overt differences are observed for p16 immunostaining in vivo and in vitro.

Supplemental Figure 6. Loss of Hdac1/2 results in DNA damage and activates DNA damage response.

Supplemental Figure 7. No apoptotic cells are observed in DKO glomeruli as evaluated by the TUNEL assay.

Supplemental Figure 8. Urinary podocytes were not observed in control mice.

Supplemental Material 1. DEGs in control and DKO mice according to RNA-seq profiling.

Supplemental Material 2. List of DEGs from our RNA-seq profiling found in the SASP Atlas.