Transient upregulation of EGR1 signaling enhances kidney repair by activating SOX9+ renal tubular cells.

Background: Acute kidney injury (AKI) is associated with damage to the nephrons and tubular epithelial cells (TECs), which can lead to chronic kidney disease and end-stage renal disease. Identifying new biomarkers before kidney dysfunction will offer crucial insight into preventive and therapeutic options for the treatment of AKI. Early growth response 1 (EGR1) has been found to be a pioneer transcription factor that can sequentially turn on/off key downstream genes to regulate whole-body regeneration processes in the leopard worm. Whether EGR1 modulates renal regeneration processes in AKI remains to be elucidated.
Methods: AKI models of ischemia-reperfusion injury (IRI) and folic acid (FA) were developed to investigate the roles of EGR1 in kidney injury and regeneration. To further determine the function of EGR1, Egr1-/- mice were applied. Furthermore, RNA sequencing of renal TECs, Chromatin Immunoprecipitation (ChIP) assay, and Dual-luciferase reporter assay were carried out to investigate whether EGR1 affects the expression of SOX9.
Results: EGR1 is highly expressed in the kidney after AKI both in humans and mice through analysis of the Gene Expression Omnibus (GEO) database. Furthermore, we verified that EGR1 rapidly up-regulates in the very early stage of IRI and nephrotoxic models of AKI, and validation studies confirmed the essential roles of EGR1 in renal tubular cell regeneration. Further experiments affirmed that genetic inhibition of Egr1 aggravates the severity of AKI in mouse models. Furthermore, our results revealed that EGR1 could increase SOX9 expression in renal TECs by directly binding to the promoter of the Sox9 gene, thus promoting SOX9+ cell proliferation by activating the Wnt/β-catenin pathway. Conclusions: Together, our results demonstrated that rapid and transient induction of EGR1 plays a renoprotective role in AKI, which highlights the prospects of using EGR1 as a potential therapeutic target for the treatment of AKI. © The author(s).

Introduction

Acute kidney injury (AKI), a common and severe disease, is associated with a high risk of developing chronic kidney disease (CKD) and end-stage renal disease (ESRD), which is a global health burden with high morbidity and mortality 1, 2. AKI affects all segments of the nephron, tubules, glomerulus, interstitium, and vasculature, but proximal tubular cells are the most commonly injured 3. In response to AKI, the molecular characteristics of tubular epithelial cells (TECs) during this process can drive kidney regeneration or interstitial inflammation and fibrosis 3-5. Identifying new biomarkers before kidney dysfunction might help to detect AKI earlier and will be critical in the development of new therapeutics for AKI treatment 6.

TEC injury is the main pathogenic consequence of AKI 7, and resident TECs have a remarkable ability to proliferate and repair after injury 8, 9. During the repair process, resident TECs extend to repopulate the tubule through a process of dedifferentiation, proliferation, and redifferentiation, restoring the functional integrity of the nephrons 10-12. The regenerative and repair abilities of renal TECs after injury are the key to recovery of renal function in patients with AKI. Sex-determining region Y box 9 (SOX9) is a transcription factor that controls cell fate decisions during homeostasis of a wide range of adult tissues and embryonic development 13. Recently, it has been shown SOX9+ cells contribute to renal repair by accelerating the dedifferentiation and proliferation of TECs in the injured kidney 11, 14, 15. However, the upstream and downstream signaling of SOX9 and the mechanisms underlying cellular proliferation and regeneration are still incompletely understood.

Early growth response 1 (EGR1) is an “immediate early” transcription factor that can be rapidly and transiently induced by various cellular stimuli, such as hypoxia, growth factors, and other agents 16, and encodes a protein with a 'zinc finger' motif. EGR1 regulates the expression of many downstream long-term response genes involved in cell differentiation/proliferation, and the inflammatory response 17, 18. EGR was found to bind closed chromatin and rearrange nucleosomes to allow transcription, thus directly activating the expression of genes needed for whole body regeneration in the three-band leopard worm after amputation 19. In normal adult kidneys, the expression of EGR1 was barely detectable 20. By integrated analysis of different high-throughput data and experimental validation, our preliminary research indicated that EGR1 was markedly induced in TECs after injury 20. Previous studies also showed that EGR1 is induced in CKD and functions as a biomarker and pathogenic mediator of kidney fibrosis 21, 22. However, it has not been previously reported whether EGR1 modulates the processes of renal regeneration after AKI.

In this study, our objective was to investigate the functional role of EGR1 in AKI and confirm the transcriptional activation role in the regulation of SOX9. To our knowledge, we demonstrate for the first time that the quick induction of EGR1 in renal tubular cells after AKI is an early response that alleviates injury and thus promotes regeneration and that EGR1 thus plays a renoprotective role in ischemic and nephrotoxic AKI.

Methods

Gene Expression Omnibus (GEO) database extraction

The GEO database (https://www.ncbi.nlm.nih.gov/geo/) is a public functional genomics data repository that deposits array- and sequence-based data. To investigate the expression of EGR1 and assess the clinical relevance of EGR1 expression to human and mouse AKI, we used the keyword “AKI, human” or 'AKI, mouse' to search the database and download related gene expression profiles.

Animals and AKI models

Wild-type C57BL/6 mice (18-22 g) were purchased from the Animal Center of Chinese PLA General Hospital. Sox9fl/fl (stock No. 013106), Sox9CreERT2 (stock No. 018829) and mTmG (also named ROSA mTmG, stock No. 007676) mice were obtained from Jackson Labs. Slc34a1CreERT2 mice were constructed by Biocytogen Corporation according to methods previously described 23-27. Egr1 mice were bought from GemPharmatech Corporation (Beijing, China). Slc34a1CreERT2/+ mice were crossed with Sox9fl/fl mice to obtain Slc34a1CreERT2/+:Sox9fl/+ mice, which were then further crossed with Sox9fl/fl mice to obtain Slc34a1CreERT2/+:Sox9fl/fl mice. These mice were housed in a specific pathogen-free facility under a 12-h light/12-h dark cycle with free access to food and water. To induce ischemic AKI, bilateral renal pedicles in the mice were clipped for 30 min using microaneurysm clamps following an established protocol 28. For hydrodynamic-based plasmid delivery 24, 12 h prior to IRI, mice were injected with 20 μg of plasmid DNA in a volume of normal saline (ml) equivalent to 8% of body weight (g) via the tail vein over 5-6 seconds. We constructed a tubule-specific (Pax8 promoter) Egr1 overexpression plasmid (GeneCopoeia) to obtain Egr1 overexpressing (Egr1) mice and constructed the Egr1 (pPax8-Egr1-CFP) plasmid by adding the cyan fluorescence protein (CFP) coding sequence to the Egr1 plasmid to confirm the efficacy of the hydrodynamic plasmid delivery approach. Mice were sacrificed, blood and kidney tissue samples were collected at the indicated time points after AKI. To induce toxic AKI, mice were subjected to a single intraperitoneal injection of FA at a dose of 250 mg/kg as described elsewhere 29. Mice with CreERT2 were subjected to peritoneal injection of 120 mg/kg tamoxifen once every two days 3 times to activate the catalytic activity of the inducible Cre enzyme. The animal protocol and all animal procedures were reviewed and approved by the Institutional Animal Care and Use Committee of the Chinese PLA General Hospital.

Quantitative real-time PCR (qRT-PCR)

Total RNA was isolated with TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. The mRNA levels were determined by qRT-PCR analysis on an Applied Biosystems 7500 system (Applied Biosystems, Foster City, CA). The list of primers is presented in

Western blot analysis

Kidney tissues or TCMK1 cells were lysed on ice with Radio ImmunoPrecipitation Assay (RIPA) lysis buffer containing 100 μg/mL PhenylMethaneSulfonyl Fluoride (PMSF). The supernatants were collected after centrifugation at 12,000×rpm at 4 °C for 30 min. Approximately 30 μg protein from each sample was added to validate protein expression. Primary and second antibodies are listed in

Histology analysis

Samples were isolated at indicated time point. For Hematoxylin-Eosin (H&E) and periodic acid-schiff (PAS) staining, tissues were fixed with 4% formaldehyde, dehydrated and embedded in paraffin. Tissue sections (5 μm) were stained with PAS. Histological examinations of tubular injury were evaluated by acute tubular necrosis (ATN) scores in a blinded manner regarding the grading of tubular necrosis, cast formation, tubular dilation, and loss of the brush border as described previously 20. Fifteen non-overlapping fields (400×) were randomly selected and scored as follows: 0: none; 1: 1 to 10%; 2: 11 to 25%; 3: 26 to 45%; 4: 46 to 75%; and 5: >76%.

For immunofluorescence staining, samples were fixed in 4% paraformaldehyde (PFA) at 4 °C for 18 h, incubated for 2 h in 30% sucrose in phosphate buffered saline (PBS) and embedded in optimal cutting temperature compound (OCT) and then sectioned into 5μm thick slices. The primary and secondary antibodies used are listed in

Multiplex immunofluorescence (mIF) staining

MIF staining was conducted by Opal 7-Color Manual IHC Kits (NEL811001KT, Akoya Biosciences, Marlborough, Massachusetts, USA). The slides from formalin-fixed paraffin-embedded kidney tissues were deparaffinized, rehydrated, and subjected to epitope retrieval by boiling in citrate buffer for 20 min at 97 °C. Endogenous peroxidase was then blocked by incubation in 3% H2O2 for 15 min, and tissue sections were covered with blocking buffer for 10 min at room temperature. Only one antigen was detected in each round, including primary antibody incubation, secondary antibody incubation, and tyramine signal amplification (TSA) visualization, followed by labeling the next antibody after epitope retrieval and protein blocking as before. The antibodies used in this experiment are listed in The slides were scanned using the PerkinElmer Vectra (Vectra 3.0.5; PerkinElmer, Massachusetts, USA).

Cell culture and treatment

Primary renal tubular epithelial cells from mice were isolated as previously described 30. Mouse kidney epithelial cell line, TCMK1, was purchased from Beijing Likeli Biotechnology Co. (Beijing, China) and cultured with DMEM/F12 medium (1:1) (Gibco) supplemented with 10% v/v fetal bovine serum (Gibco).

For small interfering RNA (siRNA) or plasmid transfection experiments, we cultured mouse primary renal TECs and kidney epithelial cell line TCMK1 cells to approximately 50% confluence and transfected them using EndoFectin™ Max (GeneCopoeia, China) 12 h before subjecting them to hypoxia/reoxygenation (H/R) injury to mimic IRI. Briefly, cells were incubated in glucose-free medium in hypoxia condition (1.0% O2) for 6 h. The cells were then incubated in normal conditions with complete medium for 18 h. In this study, we used a Cytomegalovirus (CMV) promoter plasmid (based on the pReceiver-M02 plasmid, GeneCopoeia) to construct an Egr1 overexpression plasmid (Egr1) for use in vitro and used the Pax8 promoter plasmid (Egr1) in vivo. A list of siRNA oligosequences used (GenePharma, China) is provided in

For the in vitro scratch wound assay, 12 h after control siRNA or plasmid transfection, TCMK1 cells were subjected to H/R injury. The confluent cell monolayer was scratched using a sterile 200-μl pipette tip, and washed with PBS. Cell images were captured at 0 h and 18 h after scratching with a wide field microscope. Cell migration was calculated as the ratio of the open area after 18 h to the open area at 0 h.

For cell counting kit-8 (CCK-8) assays, a total of 2000 cells per well were seeded in 96-well plates (Corning, NY) and then transfected with siRNAs and plasmids and treated with 25 µmol/L ICG-001 (Wnt/β-catenin inhibitor, SF6827, Beyotime Biotechnology) as appropriate, after which cell (TCMK1 cell) proliferation was measured using the CCK-8 assay (cell counting kit-8, Dojindo, Japan) according to the manufacturer's instructions. Proliferation rates were determined by measuring the absorbance at 450 nm with a microplate reader (Bio-Rad, USA).

For RNA-Seq and analysis, mouse primary renal TECs were placed in 6 cm plates (Corning, NY) and divided into indicated groups. Cells were then collected with 1 mL of TRIzol (Invitrogen, Carlsbad), RNA was extracted and RNA-Seq was performed by Annoroad Corporation (Beijing, China). Gene set enrichment analysis (GSEA) was performed using GSEA 4.0.3 software (http://www.gsea-msigdb.org/gsea/index.jsp). All raw RNA-Seq data were uploaded to the GEO database GSE174812 (SubSeries are GSE174808 and GSE174811).

Dual-luciferase reporter assay

This experiment can be used to verify the binding of transcription factors (reporter gene) to the promoter of downstream target gene. Briefly, the 2000 bp sequence upstream of the transcription initiation site of Sox9 containing potential EGR1-binding sites (-AGTGGGGGTGG-) and its mutant sites (-AGTGGGttTGG-) were constructed into an expression vector (pGL3, GenePharma, China) containing luciferase to construct the pGL3-Sox9 or pGL3-mutSox9 reporter plasmid. The reporter plasmid can regulate the transcriptional expression of Luciferase. The reporter plasmid was then transfected into the 293T cells by using EndoFectin Max (GeneCopoeia, China) according to the manufacturer's instructions, cells were lysed after different treatments (transfected with Egr1 or Control vector). The substrate luciferin was added, and luciferase catalyzed luciferin to emit fluorescence (the strongest wavelength was around 560nm). The fluorescence value can be used to determine the effect of different treatment groups on the transcriptional regulatory element. To avoid errors due to efficiency differences in transfection, Renilla luciferase's reporter plasmid (pRL-CMV) was used as an internal reference (the strongest wavelength was around 465 nm). Luciferase reporter gene expression was measured with a Dual-Luciferase Reporter Assay Kit (Promega, E1910) and a Centro XS LB960 detector (Berthold). Relative firefly luciferase (RFL) activity was obtained by normalizing firefly luciferase activity against Renilla luciferase activity.

Chromatin Immunoprecipitation (ChIP)

ChIP assay can identify the combination between specific genes and the target protein sequence to reflect the interaction between proteins and DNA. Firstly, the specificity antibody targeting the target protein is used to pull down the DNA fragment which binding to the target protein. Secondly, the protein and the interacted DNA fragment were separate to obtain the DNA fragment, and then qRT-PCR was performed to determine the binding site of the target protein in the genome. In our study, ChIP experiments were performed with a ChIP kit (Cell Signaling Technology, ChIP9003) according to the kit instructions. An antibody against EGR1 for ChIP experiments was purchased from CST (#4154). The qRT-PCR (SYBR Green) was used to measure the amounts of DNA fragments and enrichment efficiency. The primers of the mouse Sox9 promoter region are 5'-CAGACTCCAGGCGCAGAAG-3' (forward primer) and 5'-GACTTCGCTGGCGTTTACAG-3' (reverse primer).

Electrophoretic mobility shift assay (EMSA)

EMSA can be used to study the binding of target proteins to specific DNA sequences. The TCMK1 cells were incubated in glucose-free medium in hypoxia condition (1.0% O2) for 6 h, and then incubated in normal conditions with complete medium for 1 h. Nuclear extracts were obtained using a Nuclear and cytoplasmic protein extraction Kit (Beyotime, China) according to the manufacturer's instructions. Single-stranded oligonucleotides were obtained from BGI (Beijing Genomics Institute, China). Double-stranded oligonucleotides were obtained by annealing equal amounts (0.1 mg) of the complementary single-stranded oligonucleotides by heating to 95 °C for 5 min and then gradually cooling to room temperature. Double-stranded oligonucleotides encoding the EGR1 potential binding sequence were 5'-CATCGAAAAGTGGGGGTGGGGGGTTGT-3' and 3'-ACAACCCCCCACCCCCACTTTTCGATG-5'(unlabeled probe), the muted potential binding sequence were 5'-CATCGAAAAGTGGGTTTGGGGGGTTGT-3' and 3'-ACAACCCCCCACCCAAACTTTTCGATG-5' (unlabeled mut-probe). The unlabeled probe was end-labeled with EMSA Probe Biotin Labeling Kit (Beyotime, China). Nuclear extracts were added to 20 µL of binding reactions and incubated for 20 min at room temperature. The EMSA reactions were performed according to the manufacturer's protocol (EMSA/Gel-Shift kit, Beyotime, China). The same unlabeled probe was used as a competitor in the assay. A prominent single super shifted band was observed when nuclear extracts were incubated with an anti-EGR1 antibody.

Statistical analysis

All data were expressed as Mean ± SEM. Statistical analysis of the data was performed using Graphpad Prism 7.0 (GraphPad, CA, USA). Unpaired Student's t test was used for comparisons between two groups. We used one-way ANOVA corrected with Bonferroni coefficient to compare multiple groups. A p value < 0.05 was considered to be significant.

Results

Rapid and transient induction of EGR1 both in ischemic and toxic AKI

To investigate EGR1 expression and assess the clinical relevance of EGR1 expression to AKI, keywords “AKI, human” were used to search the Gene Expression Omnibus (GEO) database and downloaded the gene expression profiles GSE30718 31 and GSE145085 32. A significant increase of Egr1 expression in the AKI human kidney (Figure ) and human kidney organoid () were observed. Data (GSE164647) from a recently published single cell sequencing study of human kidney organoid 33 revealed that Egr1 expression was 1.9 times higher in the injured group than that in the control group. Mice mRNA sequencing series GSE52004 34 and GSE98622 35 showed that Egr1 also increased significantly in mice kidney after IRI (Figure ). These results suggest that EGR1 induction is a common feature of AKI both in human and mouse.

Ischemia and toxins are two common causes of AKI in the clinical settings, and we have successfully constructed models of ischemic AKI induced by IRI () and toxic AKI induced by folic acid (FA) () respectively. Results of qRT-PCR (Figure ), immunohistochemical staining (Figure ), immunofluorescence staining () and Western blotting (Figure ) revealed that EGR1 was hardly detected in normal kidney, but was immediately upregulation after IRI. The highest Egr1 mRNA expression is 30 min to 60 min after IRI, and the highest EGR1 protein expression is 1 h to 6 h after IRI. The expression of EGR1 was earlier than kidney injury molecule-1(KIM1, a marker of kidney injury) in both mRNA and protein level. Similar results were observed in FA-induced AKI (), suggesting that induction of EGR1 is a common feature that characterizes not only in ischemic AKI but also in toxic AKI. To clarify the tubular location of EGR1, we performed co-staining of the kidney tubules for EGR1 with various markers, lotus teragonolobus lectin (LTL, a proximal tubule marker), peanut agglutinin (PNA, a henle/distal tubule loop marker), dolichos biflorus agglutinin (DBA, a collecting duct marker), aquaporin 2 (AQP2, a collecting duct marker) 36, 37, and Endomucin (EMCN, an endothelial cell marker 38). Immunostaining results indicated that EGR1 could be detected in most renal tubule segments and in some of interstitial endothelial cell of the kidneys with AKI (Figure ).

EGR1 decreases tubular injury and drives renal tubule repair and regeneration

To further investigate the function of EGR1 in renal tubule repair after AKI, we constructed Egr1 overexpressing mice by injecting a pPax8-Egr1-CFP plasmid into mice to facilitate kidney tubule-specific Egr1 overexpression (Egr1). The pPax8-Egr1-CFP plasmid contains promoter of paired box 8 (Pax8), Egr1, and a cyan fluorescence protein (CFP), which can be used to confirm the efficacy of plasmid delivery. Figure shows the experimental protocols. By using intravital two-photon microscopy to trace CFP expression in live mTmG mouse, the boost expression of CFP in renal tubules was detected (Figure ), which indicated that the gene transfection system could effectively increase Egr1 expression.

To determine the protective function of EGR1 in AKI, mice were subjected to IRI for 12 h after plasmid injection and sacrificed 3 d after surgery (Figure ). Renal function analysis of serum creatinine (SCr) revealed that EGR1 could significantly ameliorate IRI (Figure ). PAS staining also showed less morphological injury to the kidneys in mice with EGR1 overexpression (Figure ). Immunostaining of KIM-1 revealed that EGR1 could improve IRI with notable decreasing expression of KIM-1 (Figure ). Proliferating cell nuclear antigen (PCNA) staining confirmed that EGR1 could increase the proliferation of TECs (Figure ). PAX2 have been characterized as a marker of dedifferentiated proximal tubule cells 39, 40. Significant upregulation of PAX2 was observed in the cortical area after IRI in Egr1 mice (Figure ). Multiplex immunofluorescence (mIF) staining showed that EGR1 can double staining with PAX2 and PCNA in the same cells, while few co-stained with KIM1 3 d after IRI (). Similar results were observed in analyzing the role of EGR1 in the FA-induced AKI model ().

EGR1 deficiency exacerbates kidney injury and inhibits tubule repair

To further confirm the protective role of EGR1 in renal tubule repair after AKI, we raised Egr1 knockout (Egr1) mice, both Egr1 and WT mice were subjected to IRI and sacrificed 72 h after surgery. The efficacy of Egr1 knockout was confirmed by immunohistochemistry. Egr1 was barely detected in Egr1 mice (). The Egr1 mice were phenotypically normal and showed no appreciable defects in renal morphology or function. However, Egr1 deficiency significantly aggravated renal IRI with SCr level significantly increased (Figure ). AKI histopathological damage was evaluated by PAS (periodic acid-Schiff) and KIM-1 staining. PAS staining revealed the cast formation, a marker of renal impairment (Figure ). KIM-1 staining also confirmed the aggravation of kidney injury in Egr1 mice (Figure ). In addition, similar renal damage was elucidated in cell proliferation as labeled by PCNA (Figure ). The number of PCNA-positive cells was significantly decreased in Egr1 mice compared to WT mice 3 d after IRI. These results confirm the role of EGR1 in kidney regeneration after IR-induced AKI.

EGR1 promotes the migration and proliferation of TECs

Next, we investigate the role of EGR1 in the regulation of tubular cell proliferation and migration after H/R injury in vitro. We used ethynyl deoxyuridine (EdU) methods for labelling dividing cells. Our results revealed an increased number of EdU-positive cells after hypoxia for 6 h and reoxygenation for 18 h. However, EdU was significantly decreased after silencing Egr1 expression (siEgr1), but increased significantly when Egr1 was overexpressed (Figure ), which suggested that EGR1 promoted the proliferation of renal tubular epithelial cells. Furthermore, the scratch wound assay showed that EGR1 promoted TCMK1 cell migration after H/R injury (Figure ).

The key molecular networks of EGR1 in kidney repair

To clarify the molecular mechanism by which EGR1 regulates renal TEC proliferation after AKI, we analyzed isolated mouse primary renal tubular epithelial cells by RNA sequencing (RNA-Seq). A total of 3090 differentially expressed genes (DEGs) met the criteria (|log2-fold change (FC) ≥ 0.378 and adjusted false discovery rate (FDR) < 0.05) upon comparison of the Sham group and the Egr1 group. A total of 594 DEGs met the criteria (| log2-FC ≥ 0.263 and FDR < 0.05) upon comparison of the Sham group and the H/R group (Figure ). By searching the Search Tool for the Retrieval of Interacting Genes/Proteins (STRING) database (http://string-db.org/), 130 of the 224 overlapping DEGs met the criterion of a combined score > 0.4. Hub genes were further extracted using the cyto-Hubba 41 plugin in Cytoscape software 42 as previously described 20. The top 30 genes were identified (Figure ) and 22 genes were associated with the regulation of cell population proliferation (Figure ). Among these hub genes, Sox9 is associated with the positive regulation of epithelial cell proliferation. It has been reported that Sox9 was required for kidney regeneration 10. In brief, our results collectively indicate that EGR1 may drive Sox9 pathway activation in kidney repair after AKI.

EGR1 promotes Sox9 expression in vivo and in vitro

In the GEO dataset of GSE98622 35, Sox9 mRNA was initially upregulated in renal tissue 4 h after IRI and kept expression for 6 months, whereas EGR1 expression restored to normal at 24 h after IRI (Figure ). The similar results were found in our own IRI-AKI model by qRT-PCR analysis (Figure ). Western blot analysis results revealed that SOX9 was markedly upregulated 24h after IRI (Figure ), with the highest expression from 24-48 h. The delayed expression of Sox9 indicates that Egr1 could be upstream element of the Sox9 gene. To verify this hypothesis, we delivered a tubule-specific Egr1 plasmid (Egr1, promoted by Pax8) into kidneys of Sox9+/- mice (Figure ) 12 h before IRI. As shown in Figure , the GFP+ cells, which represent SOX9 renal tubular cells, were upregulated significantly by Egr1 overexpression. Furthermore, immunohistochemical assays revealed that SOX9 expression was up-regulated after IRI and Egr1 overexpression increased SOX9 expression, but Egr1 knockout decreased SOX9 expression (Figure ). The majority of SOX9+ cells (81.31%) were also Ki67+ (Figure ), indicating that SOX9 renal cells are the predominant proliferative cell population when Egr1 overexpression. Similar results were observed in the FA-induced AKI model ().

EGR1 mediates SOX9 expression by binding to the promoter of Sox9

The relationship between EGR1 and SOX9 expression was further verified in the mouse kidney epithelial cell line, TCMK1. The qRT-PCR analysis revealed that Egr1 expression immediately peaked at 0.5 h after H/R, while Sox9 expression gradually increased at 0.5 h and reached a plateau at 4 h after H/R (Figure ), which is consistent with Western blot analysis (Figure ). The immunofluorescence assay showed the increased expression of SOX9 in Egr1 overexpressing TCMK1 cells after H/R, but decreased expression in Egr1 knockdown cells (Figure ). Quantitative RT-PCR analysis showed the same results (). Immunofluorescence staining revealed that EGR1 was co-stained with SOX9 after Egr1 plasmid transfected in TCMK1 cells which were subjected to H/R (Figure S9C).

To validate the targeting efficacy of EGR1 on Sox9, we constructed a double luciferase reporter system. The promoter region of Sox9 contains multiple potential binding sites for the transcription factor EGR1, which was predicted through Jaspar database (Jaspar.genereg.net) (). Chromatin immunoprecipitation (ChIP) assay showed that the ChIP enrichment efficiency for EGR1 in the Sox9 promoter region reached 0.35%, about 7 times higher than that of the negative control IgG group (Figure ). Among these potential binding sites, only 2 were in sense strand. According to the ranking order, we selected the potential binding site in sense chain with the highest score for the double luciferase assay. The reporter plasmid pGL3-Sox9 was constructed by cloning the 2000 bp sequence upstream of the transcription initiation site of Sox9 into the plasmid pGL3 (Figure ). Our luciferase reporter gene assay elucidated that EGR1 significantly increased luciferase activity by targeting Sox9 promoter-binding region (Figure ). The EMSA assay showed a specific binding of labeled probe (this potential binding site) to EGR1 (Figure ). Taken together, EGR1 regulates Sox9 transcription expression by binding the promoter of the Sox9 gene after AKI.

EGR1 requires SOX9 to drive renal tubule repair and regeneration

To further investigate whether EGR1 drives renal tubule repair and regeneration through SOX9, we utilized Slc34a1CreERT2/+: Sox9fl/fl mice in which the Sox9 gene was specifically knocked out in renal TECs (). The high efficiency of the Cre recombinant enzyme was shown in . Slc34a1CreERT2/+: Sox9+/+ mice (Sox9 WT) and Slc34a1CreERT2/+: Sox9fl/fl (Sox9 cKO) mice were injected with tamoxifen and then injected with Control vector or Egr1 plasmid 12 h before IRI and sacrificed 3 d after IRI (Figure ). Compared to Sox9 WT mice, Sox9 cKO mice exhibited severe morphological and functional damage to renal tubules, particularly in the cortex and corticomedullary border region. Egr1OV plasmid injection could not rescue this injury (Figure ). Meanwhile, immunofluorescence staining showed an increased KIM-1-positive area in Sox9 cKO mice at 3 d after IRI, and EGR1 overexpression did not reduce the area of injury induced when Sox9 was knockout (Figure ). Furthermore, Sox9 cKO mice showed fewer PCNA positive cells than Sox9 WT mice, and EGR1 overexpression did not improve the reduced proliferative capacity caused by Sox9 knockout (Figure ).

SOX9 promotes the proliferation of TECs via the Wnt/β-catenin pathway

To further investigate the mechanism by which SOX9 promotes proliferation after AKI, Sox9 expression in primary TECs was knockdown by siRNA and RNA-Seq analysis was applied. Gene set enrichment analysis (GSEA) of SOX9-responsive genes showed significant enrichment of the WNT signaling pathway in primary TECs (Figure ). The expression of genes associated with the Wnt/β-catenin pathway was significantly altered, and most of these genes decreased significantly after Sox9 knockdown (Figure ). A previous study indicated that SOX9 can promote nuclear translocation of β-catenin to activate the Wnt/β-catenin pathway 43. Increased β-catenin expression and nuclear translocalization were observed when Sox9 was overexpressed, while lower β-catenin expression was observed when Sox9 was knocked down (Figure ). The Wnt/β-catenin pathway inhibitor ICG001 could reduce the expression of genes activated by the Wnt/β-catenin pathway (Cyclin D1, cMyc) and the proliferation of TCMK1 cells, and Sox9 overexpression can't rescue the expression of Cyclin D1 and cMyc (Figure ). In brief, SOX9 is required for Wnt/β-catenin pathway to drive renal tubule repair and regeneration after AKI (Figure ).

Discussion

In the present study, our results demonstrated that EGR1 is rapidly and transiently in AKI and describe the role of EGR1 in mediating renal epithelial cell regeneration and repair after AKI for first time (Figure ). Increased expression of EGR1 in renal TECs decreases tubular injury and drives the repair and regeneration of renal tubules both in ischemic and toxic AKI. EGR1 deficiency exacerbates kidney injury and inhibits tubule repair after AKI. The transcription factor EGR1 increases SOX9 expression in renal TECs by directly binding to the promoter of the Sox9 gene, thus promoting the regeneration of SOX9 renal tubular cells regeneration by activating the Wnt/β-catenin pathway. We identify the EGR1-SOX9-Wnt/β-catenin axis as a potential target for the prevention and/or attenuation of ischemic and nephrotoxic AKI.

Renal tubules have a remarkable capacity to proliferate and repair after injury 8, 9, and increasing evidences indicate that intrinsic viable epithelial cell migration, proliferation, and dedifferentiation are primarily responsible for tubular regeneration after AKI 44, 45. Applying the IRI or FA injury to Egr1 and Egr1 mice and the H/R injury to renal cells, we found that EGR1 can improve the regeneration and repair of renal tubules after AKI in mouse model and promote proliferation, dedifferentiation and migration of TECs. Ishibe 46 reported that epithelial dedifferentiation can activate the morphological and transcriptional events involved in cell spreading, migration, and proliferation. The mechanism by which EGR1 promotes dedifferentiation and migration of TECs remains to be elucidated, but could also be related to regeneration and repair.

By performing the RNA-Seq, we found that Sox9 was one of the 30 Egr1-related hub genes and reflects the positive regulation of epithelial cell proliferation. It has been shown that upregulation of Sox9 is an early cellular response to AKI 10, 11, and Sox9 has been shown to contribute to renal repair by accelerating the dedifferentiation and proliferation of TECs in the injured kidney 14. SOX9+ renal tubular cells play crucial roles in subsequent repair processes after the initial injury phase. Our results showed that Egr1 and Sox9 were expressed in sequence by in vivo and in vitro experiments. A previous report indicated that the canonical Wnt pathway may be actively involved in the kidney-regeneration process 47. Furthermore, SOX9 could promote nuclear translocation of β-catenin to activate the Wnt/β-catenin pathway 43. Therefore, the GSEA of SOX9-responsive genes showed a significant enrichment of the WNT signaling pathway in our RNA-Seq. Our further experiments suggested that SOX9 requires the Wnt/β-catenin pathway to drive renal tubule repair and regeneration after AKI.

Interestingly, our study confirmed that EGR1 expression is an early response by which the kidney alleviates injury and promotes regeneration after AKI, and EGR1 was also reported up-regulated in patients with CKD 21, while EGR1 deficiency attenuated the normal responses of tubular cells to inflammatory factors and alleviated renal fibrosis in tubule interstitial nephritis. Silencing EGR1 could alleviate renal injury in diabetic kidney disease (DKD) 22. Beyond CKD, persistent expression of EGR1 aggravates nephrotic progression, which has also been reported in many other chronic kidney diseases, such as unilateral ureteral obstruction 48 and proteinuric kidney diseases 49. The dual roles of EGR1 may be contradictory. However, many genes, such as SOX9 13, 50-52, epidermal growth factor receptor (EGFR) 53-55, MMP7 30, 56, 57, KIM1 58, 59, and Wnt pathway genes 60-62, also have similar seemingly contradictory effects; that is, they play a role in promoting repair in AKI but also promote fibrosis progression in CKD. This may be because the injury that stimulates AKI is relatively short and mostly transient, while in CKD, sustained injury stimulation and continuous proliferation lead to the occurrence of fibrosis. The above conundrum also relates to the issue of adaptive repair and maladaptive repair after AKI. It would be interesting to identify the key factors that stop persistent activation of the above genes and determine how much EGR1 involves in those processes.

The present study has some limitations and drawbacks. First, because we mainly focused on observing changes in proliferation after AKI, the main experimental observation time was 3 d after AKI. Therefore, we did not monitor the long-term prognosis 14 d or 28 d after AKI. Second, our study did not explicitly examine the mechanism by which EGR1 is activated in AKI. Some reports 16, 17, 63 have shown that EGR1 expression can be induced by growth factors, hypoxia, and pro-inflammatory cytokines, but it remains unclear why EGR1 expression increased so quickly and to such a great extent after AKI. These issues will be illustrated in our future research.

In summary, our findings reveal that EGR1 is induced rapidly and transiently in the renal tubular epithelium after IRI and FA-induced AKI. We demonstrated that EGR1 has a renoprotective effect through increasing SOX9 expression in renal tubular epithelial cells by binding directly to the promoter of the Sox9 gene, further promoting SOX9 positive tubular regeneration by activating the Wnt/β-catenin pathway. This study provides novel insights into the role and mechanism of EGR1 in protecting renal tubules after AKI.

Supplementary figures and tables.

Click here for additional data file.