Indoxyl sulfate induces intestinal barrier injury through IRF1-DRP1 axis-mediated mitophagy impairment

Rationale: The dysfunctional gut-kidney axis forms a vicious circle, which eventually becomes a catalyst for the progression of chronic kidney disease (CKD) and occurrence of related complications. However, the pathogenic factors of CKD-associated intestinal dysfunction and its mechanism remain elusive. Methods: We first identified the protein-bound uremic toxin indoxyl sulfate (IS) as a possible contributor to intestinal barrier injury. Transepithelial electrical resistance, permeability assay and transmission electron microscopy were carried out to evaluate the damaging effect of IS on intestinal barrier in intestinal epithelial cells, IS-injected mice and CKD mice. In vitro and in vivo experiments were performed to investigate the role of IS in intestinal barrier injury and the underlying mechanism. Finally, CKD mice treated with AST-120 (an oral adsorbent for IS) and gene knockout mice were used to verify the mechanism and to explore possible interventions for IS-induced intestinal barrier injury. Results: Transepithelial electrical resistance and the expressions of tight junction-related genes were significantly suppressed by IS in intestinal epithelial cells. In vitro experiments demonstrated that IS inhibited the expression of dynamin-related protein 1 (DRP1) and mitophagic flux, whereas DRP1 overexpression attenuated IS-induced mitophagic inhibition and intestinal epithelial cell damage. Furthermore, IS suppressed DRP1 by upregulating the expression of interferon regulatory factor 1 (IRF1), and IRF1 could directly bind to the promoter region of DRP1. Additionally, the decreased expression of DRP1 and autophagosome-encapsulated mitochondria were observed in the intestinal tissues of CKD patients. Administration of AST-120 or genetic knockout of IRF1 attenuated IS-induced DRP1 reduction, mitophagic impairment and intestinal barrier injury in mice. Conclusions: These findings suggest that reducing IS accumulation or targeting the IRF1-DRP1 axis may be a promising therapeutic strategy for alleviating CKD-associated intestinal dysfunction.

nephrectomy two weeks later, as previously described [1]. Laparotomy without damaging the kidney served as sham mice. CKD mice were fed a diet containing 5% AST-120 (Kremezin, Kure ha Chemical Industry, Tokyo, Japan) for 8 weeks. WT C57BL/6J and IRF1 -/mice were intraperitoneally injected with IS (100 mg/kg, Sigma-Aldrich, St. Louis, MO, USA) daily for 8 weeks. Each group included at least 8 mice. All mice were randomly divided into various groups and blinded to the investigators. Mice were euthanized by carbon dioxide inhalation, and the serum and intestinal tissues were collected. Macroscopic score of the disease grade was assessed as previously described [2]. All animal procedures were approved by the Committee of Ethics on Animal Experiments of the Army Medical University.

16s rRNA sequencing and bioinformatic analysis
Fresh fecal samples of sham and CKD mice were collected for 16s rRNA sequencing and analysis by Beijing Genomics Institute (Beijing, China) as previously described [3]. Briefly, the quality of DNA extracts was determined using NanoDrop ND-2000 (Thermo Fisher, Waltham, MA, USA), and DNA integrity was examined by agarose gel electrophoresis. The qualified DNA was used to construct a library. The V4 region of 16s rRNA gene was obtained using PCR, and the products were purified for sequencing on the Illumina (HiSeq 2500) plateform (Illumina, Inc, San Diego, CA).
The numbers of tag and operational taxonomic unit (OTU) were generated from the above samples. Based on the OUT abundance, Venn diagram was drawn with VennDiagram of software R (v3.1.1). Principal Component Analysis (PCA) was performed with the relative OTU abundance value using package "ade4" of software R (v3.1.1). A dot represents each sample, and different colors represent different groups. The alpha diversity comparison was analyzed by Wilcoxon Rank-Sum test, and the plotbox was drawn with software R (v3.1.1). Heatmap was generated using the package "gplots" of software R (v3.1.1) and the distance algorithm is "euclidean", the clustering method is "complete".

High pressure liquid chromatography (HPLC)
Blood samples of patients and mice were centrifuged for 10 minutes (min) at 3000 rpm. Serum level of IS was measured using HPLC as previously described [1].

Enzyme-linked immunosorbent assay (ELISA)
Serum levels of TNF-α, IL-1β and IL-6 in mice were determined using the corresponding ELISA kits (Boster Biological Technology, Wuhan, China) according to the manufacturer's protocol.

Reverse transcription and quantitative PCR (qPCR)
Total RNA was extracted using Trizol reagent (Invitrogen, Carlsbad, CA, USA) and reversely transcribed with a reverse-transcript kit (Promega, Madison, WI, USA). qPCR was performed with SYBR Green qPCR kit (Takara, Dalian, China), as previously described [7]. Total DNA was extracted from Caco2 cells using a DNA extraction kit (Takara), and mitochondrial DNA (mtDNA) copy number was analyzed using qPCR to detect mitochondrially encoded cytochrome c oxidase II (MT-CO2), taking glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an internal control.
The mouse and human primers for qPCR are listed in Tables S2 and S3.

Western blot analysis
Total protein was extracted using a cell lysis buffer containing 50 mM Tris (pH 7.4), 150 mM NaCl, 1% Triton X-100, 1% sodium deoxycholate, 0.1% SDS (Beyotime) supplemented with each protease and phosphatase inhibitor cocktail tablet (Roche Diagnostics GmbH, Mannheim, Germany) per 10 ml solution . The protein concentration was measured with a BCA kit (Beyotime).Western blot analysis was performed as previously described [7]. Briefly, the proteins were separated with 12% SDS-PAGE, transferred into a PVDF membrane, and incubated with primary antibody at 4 °C overnight. Primary antibodies against IRF1 (sc-514544x) and β-actin

ROS detection
After treatment with IS or Indole, Caco2 cells were incubated with CM-H2DCF-DA at 37 °C for 20 min, and analyzed using the Accuri C6 flow cytometer (BD Biosciences, San Jose, CA, USA).

Mitochondrial membrane potential analysis
After treatment, Caco2 cells were incubated with JC-1 at 37 °C for 20 min, washed with JC-1 buffer and placed in culture medium on ice. Cells were imaged using a laser scanning confocal microscope (Zeiss, LSM780, Germany) or harvested for C6 flow cytometry analysis (BD Biosciences).

Transmission electron microscopy
Cells and intestinal tissues were harvested and imaged using a TEM (JEM-1400PLUS, Japan) as previously described [8].
Then, they were incubated with Cy3-conjugated goat anti-rabbit antibody or FITC-conjugated goat anti-mouse antibody in the dark for 1 hour and DAPI for 3 min.
Cells or tissue sections were observed using a laser scanning confocal microscope (Zesis).

mCherry-GFP-LC3 assay
To evaluate autophagic flux, Caco2 cells were seeded on coverslips (NEST) and transfected with mCherry-GFP-LC3 plasmids (Beyotime) using Lipofectamine 2000 (Invitrogen), according to the manufacturer's protocol. After 24 h, cells were treated with control, IS, Rapa or Baf for another 24 h, and then examined using a laser scanning confocal microscope (Zesis). Images were analyzed with the ImageJ software.

Overexpression and downregulation of target genes
PCR was performed to obtain full-length cDNA of target genes. After digestion, recombination, transformation and culture of the positive monoclones, the bacteria were identified by Beijing Genomics Institute (Beijing, China). The restriction enzyme for DRP1 and IRF1 was BamHI and EcoRI, respectively, and the vector was Then, cells were treated with control or IS for another 24 h and harvested for subsequent analysis.

Construction of reporter plasmids and point mutation
Putative IRF1 binding sites in DRP1 promoter region are listed in Table S1. Truncated DRP1 promoter fragments were amplified by PCR using genomic DNA of Caco2 cell (the primers are listed in Table S4)

Dual-luciferase reporter assay
The recombinant plasmids and pGL3-basic were co-transfected with pRL-TK vector (Promega) into Caco2 cells separately using Lipofectamine 2000 (Invitrogen), according to the manufacturer's protocol. Cells were treated with control or IS for another 24 h. Luciferase activity was detected using a Dual-luciferase reporter assay system (Promega). The firefly activity was normalized against Renilla activity.

Chromatin immunoprecipitation (ChIP) assay
ChIP was performed using a ChIP kit (Millipore) as previously described [7]. Briefly, Caco2 cells were treated with control or IS for 24, fixed with 1% formaldehyde, lyzed in SDS lysis buffer, sonicated to shear DNA, and immunoprecipitated with 2 μg antibody against IRF1 (Santa Cruz Biotechnology), taking IgG as a negative control.
The precipitated DNA was amplified by PCR and qPCR with the primers (-584 ~ -401) that cover the IRF1 binding sites (-508 ~ -497) in pGL3-PGC1α-P3. Primers (-1136 ~ -1020) without IRF1 binding sites served as a negative control, while the total DNA (Input) served as a positive control. The primers for ChIP are listed in Table S3.

Statistical analysis
Data were presented as mean ± SEM unless otherwise stated. Comparisons between two groups were analyzed by two-tailed unpaired Student's t test. Comparisons among multiple groups were tested by one-way analysis of variance (ANOVA).