Chronic kidney disease in a murine model of non-alcoholic steatohepatitis (NASH)

Clinical studies suggest that non-alcoholic steatohepatitis (NASH) is an independent risk factor for chronic kidney disease (CKD), but causality and mechanisms linking these two major diseases are lacking. To assess whether NASH can induce CKD, we have characterized kidney function, histological features, transcriptomic and lipidomic profiles in a well-validated murine NASH model. Mice with NASH progressively developed significant podocyte foot process effacement, proteinuria, glomerulosclerosis, tubular epithelial cell injury, lipid accumulation, and interstitial fibrosis. The progression of kidney fibrosis paralleled the severity of the histologic NASH-activity score. Significantly, we confirmed the causal link between NASH and CKD by orthotopic liver transplantation, which attenuated proteinuria, kidney dysfunction, and fibrosis compared with control sham operated mice. Transcriptomic analysis of mouse kidney cortices revealed differentially expressed genes that were highly enriched in mitochondrial dysfunction, lipid metabolic process, and insulin signaling pathways in NASH-induced CKD. Lipidomic analysis of kidney cortices further revealed that phospholipids and sphingolipids were the most significantly changed lipid species. Notably, we found similar kidney histological changes in human NASH and CKD. Thus, our results confirm a causative role of NASH in the development of CKD, reveal potential pathophysiologic mechanisms of NASH-induced kidney injury, and established a valuable model to study the pathogenesis of NASH-associated CKD. This is an important feature of fatty liver disease that has been largely overlooked but has clinical and prognostic importance. Clinical studies suggest that non-alcoholic steatohepatitis (NASH) is an independent risk factor for chronic kidney disease (CKD), but causality and mechanisms linking these two major diseases are lacking. To assess whether NASH can induce CKD, we have characterized kidney function, histological features, transcriptomic and lipidomic profiles in a well-validated murine NASH model. Mice with NASH progressively developed significant podocyte foot process effacement, proteinuria, glomerulosclerosis, tubular epithelial cell injury, lipid accumulation, and interstitial fibrosis. The progression of kidney fibrosis paralleled the severity of the histologic NASH-activity score. Significantly, we confirmed the causal link between NASH and CKD by orthotopic liver transplantation, which attenuated proteinuria, kidney dysfunction, and fibrosis compared with control sham operated mice. Transcriptomic analysis of mouse kidney cortices revealed differentially expressed genes that were highly enriched in mitochondrial dysfunction, lipid metabolic process, and insulin signaling pathways in NASH-induced CKD. Lipidomic analysis of kidney cortices further revealed that phospholipids and sphingolipids were the most significantly changed lipid species. Notably, we found similar kidney histological changes in human NASH and CKD. Thus, our results confirm a causative role of NASH in the development of CKD, reveal potential pathophysiologic mechanisms of NASH-induced kidney injury, and established a valuable model to study the pathogenesis of NASH-associated CKD. This is an important feature of fatty liver disease that has been largely overlooked but has clinical and prognostic importance. Translational StatementNonalcoholic steatohepatitis (NASH) is the second most common indication for a liver transplant, and it is growing rapidly worldwide, paralleling the obesity epidemic. Although many studies have demonstrated that NASH is an independent risk factor for chronic kidney disease (CKD), models and mechanisms linking these 2 diseases are still lacking. Here, we have characterized renal dysfunction, glomerulosclerosis, tubular injury, interstitial fibrosis, lipid accumulation, and mitochondrial damage in a well-validated murine NASH model. Furthermore, we confirmed the regression of kidney damage and CKD progression in NASH mice following the orthotopic liver transplantation. Transcriptomic and lipidomic analyses revealed the alteration of multiple signaling pathways and lipid species at both intermediate and late stages. Finally, we found similar kidney histologic and ultrastructural changes in human NASH and CKD. These findings confirm a causative role of NASH in the development of CKD and establish a valuable model to study the pathogenesis of NASH-associated CKD, an important feature of fatty liver disease that has been largely overlooked, yet has clinical and prognostic importance. Nonalcoholic steatohepatitis (NASH) is the second most common indication for a liver transplant, and it is growing rapidly worldwide, paralleling the obesity epidemic. Although many studies have demonstrated that NASH is an independent risk factor for chronic kidney disease (CKD), models and mechanisms linking these 2 diseases are still lacking. Here, we have characterized renal dysfunction, glomerulosclerosis, tubular injury, interstitial fibrosis, lipid accumulation, and mitochondrial damage in a well-validated murine NASH model. Furthermore, we confirmed the regression of kidney damage and CKD progression in NASH mice following the orthotopic liver transplantation. Transcriptomic and lipidomic analyses revealed the alteration of multiple signaling pathways and lipid species at both intermediate and late stages. Finally, we found similar kidney histologic and ultrastructural changes in human NASH and CKD. These findings confirm a causative role of NASH in the development of CKD and establish a valuable model to study the pathogenesis of NASH-associated CKD, an important feature of fatty liver disease that has been largely overlooked, yet has clinical and prognostic importance. Nonalcoholic fatty liver disease (NAFLD) is now the most common chronic liver disease worldwide, associated with the global epidemics of obesity, diabetes, and metabolic syndrome.1Loomba R. Friedman S.L. Shulman G.I. Mechanisms and disease consequences of nonalcoholic fatty liver disease.Cell. 2021; 184: 2537-2564Abstract Full Text Full Text PDF PubMed Scopus (734) Google Scholar, 2Turkish A.R. Nonalcoholic fatty liver disease: emerging mechanisms and consequences.Curr Opin Clin Nutr Metab Care. 2008; 11: 128-133Crossref PubMed Scopus (13) Google Scholar, 3Younossi Z.M. Non-alcoholic fatty liver disease - a global public health perspective.J Hepatol. 2019; 70: 531-544Abstract Full Text Full Text PDF PubMed Scopus (1318) Google Scholar Nonalcoholic steatohepatitis (NASH) is the progressive form of NAFLD, which can lead to cirrhosis, hepatocellular carcinoma, and liver-related mortality, but has limited treatment options. A growing body of clinical evidence established NASH as an independent risk factor for chronic kidney disease (CKD) even after adjustments for traditional risk factors, such as age, sex, body mass index, hypertension, diabetes, smoking, and hyperlipidemia.4Targher G. Bertolini L. Rodella S. et al.Relationship between kidney function and liver histology in subjects with nonalcoholic steatohepatitis.Clin J Am Soc Nephrol. 2010; 5: 2166-2171Crossref PubMed Scopus (189) Google Scholar, 5Sinn D.H. Kang D. Jang H.R. et al.Development of chronic kidney disease in patients with non-alcoholic fatty liver disease: a cohort study.J Hepatol. 2017; 67: 1274-1280Abstract Full Text Full Text PDF PubMed Scopus (116) Google Scholar, 6Park H. Dawwas G.K. Liu X. et al.Nonalcoholic fatty liver disease increases risk of incident advanced chronic kidney disease: a propensity-matched cohort study.J Intern Med. 2019; 286: 711-722Crossref PubMed Scopus (55) Google Scholar Moreover, the extent of proteinuria and reduction of glomerular filtration rate in individuals with NAFLD/NASH parallel the severity of liver fibrosis.4Targher G. Bertolini L. Rodella S. et al.Relationship between kidney function and liver histology in subjects with nonalcoholic steatohepatitis.Clin J Am Soc Nephrol. 2010; 5: 2166-2171Crossref PubMed Scopus (189) Google Scholar,7Mantovani A. Petracca G. Beatrice G. et al.Non-alcoholic fatty liver disease and risk of incident chronic kidney disease: an updated meta-analysis.Gut. 2022; 71: 156-162Crossref PubMed Scopus (132) Google Scholar Although overlapping mechanisms may underly NASH and CKD pathogenesis, such as insulin resistance, activation of the renin-angiotensin system, oxidative stress, and systemic inflammation,1Loomba R. Friedman S.L. Shulman G.I. Mechanisms and disease consequences of nonalcoholic fatty liver disease.Cell. 2021; 184: 2537-2564Abstract Full Text Full Text PDF PubMed Scopus (734) Google Scholar,8Byrne C.D. Targher G. NAFLD as a driver of chronic kidney disease.J Hepatol. 2020; 72: 785-801Abstract Full Text Full Text PDF PubMed Scopus (238) Google Scholar whether NASH is involved in CKD development and potential mechanisms linking NASH to CKD are not known. Previous studies have described kidney injury in NASH animal models induced by a high-fat diet, methionine, and choline-deficient diet,9Kim S.H. Lim Y. Park J.B. et al.Erratum: comparative study of fatty liver induced by methionine and choline-deficiency in C57BL/6N mice originating from three different sources.Lab Anim Res. 2017; 33: 318Crossref PubMed Scopus (2) Google Scholar streptozotocin/high-fat diet,10Saito H. Tanaka T. Sugahara M. et al.Inhibition of prolyl hydroxylase domain (PHD) by JTZ-951 reduces obesity-related diseases in the liver, white adipose tissue, and kidney in mice with a high-fat diet.Lab Invest. 2019; 99: 1217-1232Abstract Full Text Full Text PDF PubMed Scopus (32) Google Scholar,11Xie G. Wang X. Liu P. et al.Distinctly altered gut microbiota in the progression of liver disease.Oncotarget. 2016; 7: 19355-19366Crossref PubMed Scopus (162) Google Scholar or genetic modification.12Permyakova A. Gammal A. Hinden L. et al.A novel indoline derivative ameliorates diabesity-induced chronic kidney disease by reducing metabolic abnormalities.Front Endocrinol (Lausanne). 2020; 11: 91Crossref PubMed Scopus (4) Google Scholar However, these models do not replicate the critical metabolic and histologic features of human NASH.13Ibrahim S.H. Hirsova P. Malhi H. et al.Animal models of nonalcoholic steatohepatitis: eat, delete, and inflame.Dig Dis Sci. 2016; 61: 1325-1336Crossref PubMed Scopus (167) Google Scholar In contrast, we previously developed a new mouse model that closely resembles human NASH in histologic features, metabolic abnormalities, and gene expression pathways by administering a Western diet (WD; high-fat, high-fructose, and high-cholesterol diet), combined with a weekly administration of very low-dose carbon tetrachloride (CCl4).14Tsuchida T. Lee Y.A. Fujiwara N. et al.A simple diet- and chemical-induced murine NASH model with rapid progression of steatohepatitis, fibrosis and liver cancer.J Hepatol. 2018; 69: 385-395Abstract Full Text Full Text PDF PubMed Scopus (320) Google Scholar This "FAT-NASH" (Fibrosis and Tumors) model closely resembles human NASH, including an altered microbiome that parallels those changes seen in humans.15Carter J.K. Bhattacharya D. Borgerding J.N. et al.Modeling dysbiosis of human NASH in mice: loss of gut microbiome diversity and overgrowth of Erysipelotrichales.PLoS One. 2021; 16e0244763Crossref Scopus (29) Google Scholar This model makes it ideal to investigate whether NASH provokes the development of CKD. In the current study, we have characterized the histologic features and transcriptomic and lipidomic analyses of kidneys in this NASH model to assess the relationship and potential mechanisms linking NASH to CKD. In addition, we have confirmed a causal relationship between NASH and CKD by demonstrating that orthotopic liver transplantation (OLT) leads to attenuation of kidney disease. C57BL/6J mice were purchased from Jackson Laboratory. Mice were fed with a WD diet and treated with CCl4 (0.32 μg/g, i.p., once a week), as previously described.14Tsuchida T. Lee Y.A. Fujiwara N. et al.A simple diet- and chemical-induced murine NASH model with rapid progression of steatohepatitis, fibrosis and liver cancer.J Hepatol. 2018; 69: 385-395Abstract Full Text Full Text PDF PubMed Scopus (320) Google Scholar Experimental groups consisted of control mice, treated with chow diet (CD) and vehicle corn oil for 12 or 24 weeks, and NASH mice, treated with WD and CCl4 for 12 or 24 weeks (n = 10 mice per group, comprising 5 males and 5 females). An additional control group of mice on CD with CCl4 treatment was included to examine the potential nephrotoxic effects of CCl4 alone at 12 or 24 weeks after induction (n = 10 mice, 5 males and 5 females; Supplementary Data). All procedures were performed according to protocols approved by the Animal Care and Use Committee of the Icahn School of Medicine at Mount Sinai (IACUC-2015-0112). Male WD/CCl4 mice at 12 weeks, weighing 20 to 25 g, were used for OLT without hepatic artery reconstruction under inhalation anesthesia, according to protocols established by Yokata et al.16Yokota S. Ueki S. Ono Y. et al.Orthotopic mouse liver transplantation to study liver biology and allograft tolerance.Nat Protoc. 2016; 11: 1163-1174Crossref PubMed Scopus (24) Google Scholar In brief, a liver graft was procured from 18-week-old wild-type C57BL/6J donor male mice after dissecting vessels, ligaments, and connective tissues around the liver. After clamping the portal vein, intrahepatic inferior vena cava, and suprahepatic inferior vena cava, the recipient liver of NASH mice was removed, and the liver graft was placed orthotopically. Vessels and bile duct were anastomosed through a combined cuff and suture technique. Recipient mice were given CD and corn oil for another 12 weeks after OLT and euthanized at 24 weeks. For control mice, sham operation was performed in parallel, and mice were switched to normal CD and corn oil for another 12 weeks. Three NASH mice completed the procedure with successful OLT and survived. Liver and kidney tissues were fixed in 10% neutral-buffered formalin, embedded in paraffin, and sectioned to 3-μm thickness. Liver sections were stained with hematoxylin and eosin for assessment of liver histology. Periodic acid–Schiff staining was used to examine kidney histology. Kidney and liver fibrosis were evaluated by Sirius red staining. Kidney sections were deparaffinized and rehydrated, and then processed with primary antibodies against α-smooth muscle actin (Abcam; ab124964) following an immunohistochemistry standard protocol. Lipid accumulation was determined by oil red O staining. Optimal cutting temperature compound–embedded frozen kidney sections (8 μm thick) were used for oil red O staining, according to the manufacturer's instructions. Frozen kidney sections (3 μm thick) were used for immunofluorescence staining for collagen I (Servicebio; GB11022-3). Mouse kidney tissues were fixed in glutaraldehyde. Sections were mounted on a copper grid, and the ultrastructures were imaged under a Hitachi H-7650 microscope. The thickness of the glomerular basement membrane and widths of the podocyte foot processes were determined as previously described.17Fujimoto M. Maezawa Y. Yokote K. et al.Mice lacking Smad3 are protected against streptozotocin-induced diabetic glomerulopathy.Biochem Biophys Res Commun. 2003; 305: 1002-1007Crossref PubMed Scopus (172) Google Scholar,18Chen A. Feng Y. Lai H. et al.Soluble RARRES1 induces podocyte apoptosis to promote glomerular disease progression.J Clin Invest. 2020; 130: 5523-5535Crossref PubMed Scopus (35) Google Scholar NASH scores were evaluated from the steatosis grade, inflammation score, and fibrosis stage, according to the NASH Clinical Research Network scoring system.19Kleiner D.E. Brunt E.M. Van Natta M. et al.Design and validation of a histological scoring system for nonalcoholic fatty liver disease.Hepatology. 2005; 41: 1313-1321Crossref PubMed Scopus (8151) Google Scholar The cross-sectional area of the glomerular tuft and the proportion of the mesangial matrix relative to the glomerular area were measured by using ImageJ software on an equal number of pictures per mouse (20 images) under constant magnification. The severity of tubular vacuolization was assessed by the following grading schema: 0, no vacuolization; 1, <10%; 2, 10% to 25%; 3, 26% to 50%; and 4, >50%. The amount of fibrosis was determined by quantifying the percentage of positive staining areas with Sirius red staining. The proportion of lipid area was quantified by the positive areas with oil red O staining. Semiquantitative histologic analysis was performed by a trained pathologist blinded to the groups (XL), and images were acquired by FZ and YY. The human kidney biopsy sample was collected from the Department of Pathology, Columbia University, under Columbia University Medical Center Institutional Review Board number AAAT7999 for archival use of kidney biopsy slides for research purposes. The images provided to investigators were deidentified. No patient consent was required, as per Institutional Review Board protocol. Urine albumin was measured using a commercial assay enzyme-linked immunosorbent assay kit (E99-134; Bethyl Laboratory). Urine creatinine levels were quantified using a quantiChrom Creatinine Assay Kit (DICT-500; Bioassay Systems). Urine albumin excretion was expressed as the ratio of albumin/creatinine. Blood urea nitrogen (BUN) was measured from mouse sera using a commercial kit (BioAssay Systems), according to the manufacturer's protocol. Total RNA was isolated from kidney cortices by using the RNeasy mini kit (Qiagen; 74104), according to the manufacturer's protocol. mRNA sequencing was performed at the CLC Genomics and Epigenomics Core Facility at Weil Cornell Medical College. The RNA-sequencing data were analyzed by following the procedure described below. Briefly, after sequence quality filtering at a cutoff of a minimum quality score Q20 in at least 90% bases for the paired-end 150-bp FASTQ sequencing data, the good-quality reads aligned to reads were processed and aligned to the University of California, Santa Cruz, Mus musculus reference genome and transcriptome (build mm10) using the Burrows-Wheeler Aligner.20Li H. Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform.Bioinformatics. 2009; 25: 1754-1760Crossref PubMed Scopus (31394) Google Scholar The reads that are uniquely aligned to the exon and splicing junction sites for each transcript were combined to calculate an expression level for a corresponding transcript and further normalized on the basis of reads per kilobase per million reads21Mortazavi A. Williams B.A. McCue K. et al.Mapping and quantifying mammalian transcriptomes by RNA-Seq.Nat Methods. 2008; 5: 621-628Crossref PubMed Scopus (10545) Google Scholar to compare transcription levels among samples. The transcripts with a low raw read count of >100 in all the samples were excluded for downstream analysis. Gene expression value was transformed to the log 2 base scale. Principle component analysis was first performed to assess the sample correlations using the expression data of all the genes. The differentially expressed genes in NASH mice compared with control mice were identified by the R package limma test.22Ritchie M.E. Phipson B. Wu D. et al.limma powers differential expression analyses for RNA-sequencing and microarray studies.Nucleic Acids Res. 2015; 43: e47Crossref PubMed Scopus (20346) Google Scholar A specific gene was considered differentially expressed if the P value was ≤ 0.05. The Gene Ontology and pathway analysis for the differentially expressed genes were then performed with a fold change cutoff of ≥1.5 using INGENUITY IPA (www.ingenuity.com/products/ipa) and the online tool Enrichr.23Chen E.Y. Tan C.M. Kou Y. et al.Enrichr: interactive and collaborative HTML5 gene list enrichment analysis tool.BMC Bioinformatics. 2013; 14: 128Crossref PubMed Scopus (3944) Google Scholar The read coverage of gene functional elements was also visualized by the Integrative Genome Viewer tool (www.broadinstitute.org/igv/) from the genome alignment file. Heat map analysis was performed for the top 50 differentially expressed genes after the median center was transformed using Multi-Experiment Viewer software.24Saeed A.I. Sharov V. White J. et al.TM4: a free, open-source system for microarray data management and analysis.Biotechniques. 2003; 34: 374-378Crossref PubMed Scopus (4097) Google Scholar Kidney cortex (10 ± 0.2 mg) tissues were homogenized in 40 μl of cold double-distilled water at 20 Hz. A total of 20 μl of the resulting supernatants was mixed with 225 ul methanol by vortexing. Then, we added 5 ul internal standard and 750 ul methyl tert-butyl ether. Then, the mixture was vibrated for 30 minutes, and 188 μl double-distilled water was added and vortexed to form a 2-phase system. After equilibration for 10 minutes at 4 °C, the mixture was centrifuged at 10,000g for 10 minutes at 4 °C. A total of 700 μl of supernatant was dried under nitrogen. All samples were reconstituted in acetonitrile/isopropyl alcohol/water (65:30:5, v/v/v), and samples were mixed with 20 μl as quality control, and 2 μl was injected into the liquid chromatography–mass spectrometry system. The lipidomic profiling was performed using the Ultra-Performance Liquid Chromatography Mass Spectrometry system (ultra-performance liquid chromatography, Agilent 1290; mass spectrometry, Applied Biosystems SCIEX 6500 + QTRAP) at Core Facility of Basic Medical Sciences, Shanghai Jiaotong University School of Medicine. The samples were separated with reverse-phase chromatography (Kinetex 2.6 μm; C18; 2.1 × 100 mm; Phenomenex). The experiment was performed using multi-reaction monitoring mode, in which each type of lipid was corrected with corresponding lipid isotope standards to ensure the accuracy of the experiment. LipidSig R package was used to analyze the data.25Lin W.J. Shen P.C. Liu H.C. et al.LipidSig: a web-based tool for lipidomic data analysis.Nucleic Acids Res. 2021; 49: W336-W345Crossref PubMed Scopus (28) Google Scholar R package, version was 2.11.1, was also used. All samples were normalized using the probabilistic quotient normalization method. Differential analysis was performed using normalized and log-transformed data with the limma package. Lipid with adjusted P ≤ 0.05 was considered statistically significant. Data are expressed as means ± SEM. Analysis of variance, followed by the Bonferroni correction, was used to analyze means between groups. GraphPad Prism 9 software was used for statistical analyses. P < 0.05 was considered statistically significant. To assess whether NASH can induce CKD, we first assessed their renal function at 12 weeks (intermediate NASH stage) and 24 weeks (late stage) post-NASH induction. At 12 weeks, NASH mice began to show albuminuria development compared with control mice, without significant accumulation of BUN (Figure 1a and b). By 24 weeks, overt albuminuria and BUN accumulation were observed in NASH mice (Figure 1a and b) that were not influenced by sex (Supplementary Figure S1A and B). The renal function decline was not due to a direct kidney injury by CCl4 administration, as no change in renal function parameters was observed in an independent cohort of control mice after either 12 or 24 weeks post-CCl4 administration fed with a normal diet that did not develop NASH14Tsuchida T. Lee Y.A. Fujiwara N. et al.A simple diet- and chemical-induced murine NASH model with rapid progression of steatohepatitis, fibrosis and liver cancer.J Hepatol. 2018; 69: 385-395Abstract Full Text Full Text PDF PubMed Scopus (320) Google Scholar (Supplementary Figure S2A and B). Histologic analysis of periodic acid–Schiff–stained kidney sections showed mild to moderate glomerulomegaly (red arrowheads in Figure 2a, top panel) in NASH mice at the intermediate stage, which was marked by increased glomerular cross-sectional area (Figure 2b) and the percentage of mesangial area (Figure 2b). At the late stage, NASH mice developed diffuse moderate to severe mesangial expansion with increased mesangial cell number and accumulation of mesangial matrix and focal segmental glomerular sclerosis, some of which was perihilar focal segmental glomerular sclerosis (Figure 2a, top panel). The glomerular cross-sectional area of the late-stage NASH mice was 78% larger than that of control mice (Figure 2b), and the percentage of mesangial area to the glomerular cross-sectional area of these NASH mice was 2.4-fold higher than that of control mice and 1.25-fold higher than that of the intermediate-stage NASH mice (P < 0.001) (Figure 2b). None of the control mice fed with a normal diet developed glomerulosclerosis or tubulointerstitial fibrosis (Figure 2a, top panel).Figure 2Nonalcoholic steatohepatitis (NASH) promotes glomerulosclerosis and tubulointerstitial fibrosis. (a) Representative images of periodic acid–Schiff (PAS)–stained glomeruli (top panel) and transmission electron microscopy (TEM) (bottom panel) are shown. Bar in PAS stain = 20 μm, bar in TEM = 500 nm. (b) Graph of glomerular cross-sectional area, percentage of mesangial area to glomerular area, the podocyte foot process (FP) widths, and width of glomerular basement membrane (GBM) in indicated mice. n = 10 in each group. Values represent mean ± SEM. ∗P < 0.05, ∗∗P < 0.01, ∗∗∗P < 0.001. NASH mice developed (a) significant glomerulomegaly and glomerulosclerosis (top panel, red arrowhead), (b) deposition of the mesangial matrix, (b) larger glomerular cross-sectional area, and (b) the percentage of mesangial area to glomerular area than that of control (Ctrl) mice. (a, bottom panels, and b) TEM revealed podocyte FP widening in intermediate-stage NASH and extensive FP effacement in late-stage NASH mice. Green arrows: slit diaphragms of podocytes. (c) Representative images of PAS-stained kidneys (top and middle panels) and Sirius red–stained kidneys (bottom panel) are shown. Bar = 100 μm. (d) Representative images of immunostaining of collagen I (top panel) and α-smooth muscle actin (SMA) (bottom panel) are shown. Bar = 50 μm. (e) The average percentage of Sirius red–stained positive area of kidney, and collagen I– and α-SMA–stained positive area per mouse. n = 10 in each group. Values represent mean ± SEM. ∗∗P < 0.01, ∗∗∗P < 0.001. (a) NASH mice developed significant vacuolation on tubules (asterisks, top panel), (c) interstitial fibrosis and infiltrate of inflammation cells (middle panel, arrowheads), and (c) have large irregular scars on the cortical surface (top panel, arrows) at the late stage of NASH (24 weeks). (e) In addition, NASH mice have larger Sirius red–stained positive area (left panel) and the percentage of collagen I–stained positive area (middle panel) and α-SMA–stained positive area (right panel) than that of control mice. (f) Representative images of hematoxylin and eosin (H&E)–stained (top panel) and Sirius red–stained (bottom panel) liver are shown. NASH mice developed severe steatosis, inflammation, hepatocyte ballooning, and fibrosis. Bar = 50 μm. (g,h) The average percentage of NASH active score and Sirius red–stained positive area of liver per mouse. n = 10 in each group. Values represent mean ± SEM. ∗∗∗P < 0.001. (i) Renal fibrosis area was positively correlated with NASH active score. (j) Renal fibrosis area was positively correlated with liver fibrosis area. NS, not significant. To optimize viewing of this image, please see the online version of this article at www.kidney-international.org.View Large Image Figure ViewerDownload Hi-res image Download (PPT) The evaluation of the ultrastructure of podocytes by transmission electron microscopy revealed foot process widening at the intermediate stage and extensive effacement at the late stage of NASH (Figure 2a, bottom panel). The mean foot process width in NASH mice at 12 weeks was 1.25-fold greater than control, and 1.78-fold greater than control mice in late stage of NASH mice (Figure 2b). However, there was no difference in the thickness of the glomerular basement membrane between groups (Figure 2b). Kidney histology also showed significant vacuolation in tubules (asterisks in Figure 2a, top panel) and mild tubulointerstitial fibrosis (Figure 2c, bottom panel) in NASH mice at the intermediate stage. At the late stage, NASH mice had irregular macroscopic scars on the kidney cortical surface (black arrows in Figure 2c, top panel) that represented a large fibrotic area. Microscopically, NASH mice showed a significant increase in tubulointerstitial fibrosis, tubular atrophy, and diffuse infiltrates of inflammatory cells (arrowheads in Figure 2c, middle panel), most of which were macrophages (Supplementary Figure S3). Sirius red histologic staining and immunostaining for collagen and α-smooth muscle actin confirmed a moderate but significantly increased renal fibrosis at the intermediate stage and a more remarkable increase in renal fibrosis by the late stage in NASH kidneys (Figure 2c–e). These results indicate the development of renal fibrosis at the intermediate stage of NASH, which is significantly worsened by the advanced stage of NASH. Consistent with our original description of the model, CCl4-treated mice on WD rapidly developed hallmarks of NASH (steatosis, inflammation, hepatocyte ballooning, and fibrosis) within 12 weeks, progressing to hepatocellular carcinoma by 24 weeks (Figure 2f–h). Notably, the extent of renal fibrosis was positively correlated with the NASH activity score and the percentage of liver fibrosis area (Figure 2i and j). These results indicated that renal fibrosis was strongly associated with the histologic severity of NASH, consistent with those reported in human NASH and CKD.4Targ