The effect of liver-specific ketohexokinase deletion on the intestinal-liver-kidney axis in high-fructose-induced metabolic syndrome mice

Abstract

The liver is a crucial site for fructose uptake and metabolism, a function intricately linked to fructose-associated pathologies. This study examines the role of hepatic ketohexokinase (KHK) in metabolic syndrome induced solely by high-fructose intake. Liver-specific Khk-deficient mice are generated and fed with a 20% fructose solution for 3 months, after which the features of metabolic syndrome are examined. Compared with fructose-fed floxed controls, fructose-fed liver-specific Khk-deficient mice present alleviated liver injury and hepatic steatosis, along with lower triglyceride levels in the plasma and liver, plasma aspartate transaminase and alanine transaminase levels, and mRNA levels of genes related to triglyceride and fatty acid synthesis. Liver-specific Khk deficiency also leads to lower uric acid levels in the plasma and urine, as well as xanthine oxidase activity and Glut9 mRNA levels in the liver and kidneys of fructose-fed mice. Although intestinal villus length and epithelial barrier integrity remain unaffected, the deletion of liver Khk significantly reduces fructose-stimulated KHK, Glut2, Glut5, and aldolase B expression in the intestine and kidneys, suggesting inhibited fructose absorption and metabolism in these tissues. In the adipose tissue, fructose-induced increases in adipocyte size and tumor necrosis factor-α and interleukin-6 mRNA levels are blocked by liver-specific Khk deficiency, indicating improved remodeling of adipose tissue and reduced inflammation in adipocytes. Overall, liver-specific Khk deletion is sufficient to protect against metabolic syndrome induced by excessive fructose intake. Our findings underscore the critical role of liver KHK-mediated fructose metabolism in driving the physiological and pathological consequences associated with fructose consumption along the intestinal-liver-kidney axis.

Introduction

Although dietary fructose is derived primarily from vegetables and fruits, the consumption of fructose in the form of sugar and high-fructose (HF) corn syrup, two major commercially added sweeteners, has increased dramatically over the past few decades. This rise coincides with a marked global increase in metabolic diseases, including metabolic dysfunction-associated steatotic liver disease [1]. In animal models, HF consumption has been shown to induce all the features of metabolic syndrome, such as weight gain, hepatic steatosis, fatty liver, glucose intolerance, hyperinsulinemia, and insulin resistance [ 2– 4] . Similarly, robust evidence from human studies has revealed a strong association between the consumption of HF-containing beverages and the development of key metabolic syndrome components, including obesity, fatty liver, and insulin resistance [ 5, 6] .

Ketohexokinase (KHK, also known as fructokinase), an initial enzyme for fructose metabolism, catalyzes the phosphorylation of fructose to generate uric acid (UA) and fructose 1-phosphate. The latter is further converted by aldolase B (AldoB) into ihydroxyacetone phosphate and glyceraldehyde, precursors for triglyceride synthesis [7]. KHK exists in two isoforms: KHK-A, a slow-acting fructose metabolizer ubiquitously expressed across multiple tissues, and KHK-C, a rapid metabolizer primarily expressed in the liver, small intestine, and kidney [8]. Growing evidence from animal studies has implicated KHK-mediated fructose metabolism in the pathogenesis of fructose-related metabolic syndrome. Notably, in fructose-fed mice, systemic deficiency of KHK-A/C [ 9– 13] or the administration of KHK-specific inhibitors ( e. g., PF-06835919 [ 14– 16] and compound 14 [17]) has been shown to significantly mitigate various metabolic syndrome phenotypes associated with fructose overconsumption. However, highlighting the divergent roles of KHK-A and KHK-C is critical: systemic deletion of Khk-a exacerbated, whereas global knockout of Khk-c improved, fructose-induced metabolic syndrome in murine models [ 10, 18] . Collectively, these findings consistently support the targeting of KHK as a promising therapeutic strategy for fructose-induced metabolic syndrome.

In the liver, fructose is primarily transported via glucose transporter (GLUT) 2 from the portal circulation into hepatocytes [19], where KHK and AldoB sequentially metabolize fructose to generate UA and triglycerides [7]. Notably, liver-specific Khk knockdown via N-acetylgalactosamine (GalNAc)-conjugated Khk siRNA significantly improved glucose tolerance and attenuated hepatic steatosis in mice fed with a high-fat diet with [20] or without [16] fructose drinking, providing direct evidence for the critical role of hepatic KHK in the development of fructose-related metabolic syndrome. This conclusion was further corroborated by studies from Andres-Hernando et al., which demonstrated that liver-specific KHK-A/C deficiency completely abolished metabolic syndrome in mice administered with a 10% fructose/glucose mixture [13]. However, Andres-Hernando’s studies did not exclude the confounding effect of glucose on fructose-induced metabolic syndrome. Notably, high-glucose intake alone contributes to metabolic syndrome in mice; this effect may be attributed to increased glucose-to-fructose conversion mediated by aldose reductase in the intestinal tract [21]. Indeed, high-glucose-induced metabolic syndrome was significantly attenuated in global KHK-A/C-deficient mice [21]. Thus, the precise impact of liver-specific Khk deletion on metabolic syndrome exclusively induced by HF intake, along with the underlying mechanisms, remains incompletely understood.

In this study, we generated mice with liver-specific deletion of both KHK isoforms (KHK-A and KHK-C) to investigate the role of hepatic KHK in metabolic syndrome induced solely by HF loading and explored the potential mechanisms via transcriptome sequencing.

Materials and Methods

Ethics approval

All animal studies were performed at the Experimental Animal Science and Technology Center of Jiangxi University of Chinese Medicine, following approval by the Institutional Animal Care and Use Committee of the Jiangxi University of Chinese Medicine and in compliance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals.

Generation of liver-specific Khk knockout mice

Mice harboring LoxP sequences flanking exons 3, 4, and 5 of the Khk gene ( Khk ^fl/fl, strain ID: C57BL/6J-Khk ^em1cyagen, CKOCMP-16548-Khk-B6J-VA) were initially obtained from Cyagen (Suzhou, China). Liver-specific Khk knockout (KO) mice were generated by crossing Khk ^fl/fl mice with liver-specific Cre recombinase-expressing mice (Albumin-Cre strain C001006; Cyagen, Suzhou, China) [13]. Briefly, male Khk ^fl/fl mice were bred with female Albumin-Cre transgenic mice to produce offspring with Albumin-Cre-driven Khk deletion (designated Khk ^fl/fl AlbCre ⁺, liver-specific Khk knockout). Khk ^fl/fl mice without Cre recombinase expression (defined as Khk ^fl/fl AlbCre ^–) served as control animals. All the mice were maintained on a C57BL/6J genetic background, and male mice aged 2–3 months were used for the experiments. The experimental mice were housed in a temperature- and humidity-controlled specific pathogen-free (SPF) facility under a 12:12-h light-dark cycle, with ad libitum access to tap water and standard chow.

Animal treatment

Mice aged 8–12 weeks were maintained on a standard diet and randomly assigned to two groups ( n = 5 or 8 per group): the regular drinking water control group and the HF treatment group. Fructose was administered as a 20% (w/v) solution in the drinking water ad libitum for 12 weeks, while the water of the control mice received normal drinking water. At week 12, the mice were individually housed in metabolic cages within a temperature- and humidity-controlled environment under a 12:12-h light-dark cycle for 24-h urine collection, along with measurements of body weight, 24-h food intake, and 24-h water intake. At the experimental endpoint, all the mice were anesthetized via isoflurane inhalation (R510-22-10; RWD Life Science, Shenzhen, China), and tissues, including blood, jejunum, liver, kidneys, and adipose tissues, were carefully collected.

At week 12, an insulin tolerance test (ITT) was conducted on 8-h fasted mice. Following intraperitoneal insulin injection (0.75 U/kg), blood samples were collected via tail vein sampling at predetermined time points (0, 15, 30, 60, 90, and 120 min). For the glucose tolerance test (GTT), the mice were similarly fasted for 8 h, and blood samples were collected via the tail vein at the same time points (0, 15, 30, 60, 90, and 120 min) after intraperitoneal glucose injection (1 g/kg). Blood glucose concentrations were measured via a glucometer (Sinocare, Changsha, China). To assess fructose uptake capacity, portal vein blood was collected [22], and portal vein fructose levels were measured at 0, 5, 10, 15, 20, 25, and 30 min following oral gavage of 1 g/kg fructose. The area under the curve (AUC) was statistically analyzed. To evaluate Na ⁺-H ⁺ exchanger 3 (NHE ₃) activity, the mice were administered with 1.5 mL of normal saline intraperitoneally at week 12 and housed in metabolic cages. Urine was then collected over the subsequent 6 hours, and the urinary volume and sodium levels were determined.

Biochemical analysis

Blood samples were collected from the inferior vena cava of the mice under isoflurane anesthesia, and the plasma was isolated via centrifugation at 1000 g for 10 min at 4°C. Urine and plasma parameters (UA: C012-2-1; fructose: A085-1-1; aspartate transaminase [AST]: C010-2-1; alanine transaminase [ALT]: C009-2-1; insulin: H203-1-2; sodium [Na ⁺]: C002-1-1; triglyceride: A110-1-1) were biochemically assayed using kits purchased from Nanjing Jiancheng Bioengineering Institute (Nanjing, China) according to the manufacturer’s instructions. Hepatic triglyceride level was determined using Triglyceride assay kit (A110-1-1; Nanjing Jiancheng Bioengineering Institute), and xanthine oxidase (XOD) activity in freeze-clamped liver tissue was determined via biochemical assay using XOD activity detection kit (BC1095; Solarbio, Beijing, China) following the manufacturer’s protocol.

Histopathology

Tissues, including the jejunum, liver, kidney, perigonadal white adipose tissue (WAT), and brown adipose tissue (BAT), were fixed overnight at 4°C in 3% (w/v) paraformaldehyde/0.1 M phosphate buffer and then processed for histological analysis. Formalin-fixed paraffin-embedded (FFPE) sections of the jejunum, liver, kidneys, WAT, and BAT were subjected to hematoxylin (G1005-1; Servicebio, Wuhan, China) and eosin (G1005-2; Servicebio) (H&E) staining. For lipid deposition assessment, formalin-fixed optimal cutting temperature (OCT) compound-embedded liver samples were stained with Oil Red O (O8010; Solarbio). Images were acquired under consistent imaging conditions via a DMI4000B fluorescence microscope (Leica, Wetzlar, Germany). The size and number of adipocytes in WAT and BAT were analyzed via ImageJ software.

Transcriptome sequencing and data processing

Liver samples from all groups ( n  = 3 per group) were immediately harvested for total RNA extraction, and RNA quality was assessed to construct the transcriptomic sequencing library. Transcriptome sequencing and analysis were performed by Novogene Biotechnology (Beijing, China) via the Illumina NovaSeq platform ( https://magic-plus.novogene.com/). Sequence reads from all the samples were mapped to the GRCm39 reference genome via the HISAT2 tool (version 2.2.1). Differential expression analysis between groups was conducted with DESeq2 software (v3.19). In this study, |log ₂FC| > 1 and adjusted P-value ( P _adj) < 0.05 were set as the criteria for defining differentially expressed genes (DEGs). A Protein-Protein Interaction (PPI) network was constructed via the STRING database ( https://cn.string-db.org/) combined with Cytoscape software (version 3.10.3) to identify key hub genes. All DEGs were mapped to Gene Ontology (GO) terms in the GO database to identify the top 10 enriched GO terms, with a significance cutoff of P  < 0.05. Pathway enrichment analysis was performed via the Kyoto Encyclopedia of Genes and Genomes (KEGG) database to identify the top 20 enriched signaling pathways among the DEGs, with a P _adj < 0.05 threshold.

Western blot analysis

The detailed experimental protocols have been previously described in our laboratory’s publication [23]. Briefly, tissue samples from the jejunum, liver, and kidneys were lysed and sonicated in homogenization buffer containing a protease inhibitor cocktail (Roche, Berlin, Germany). Protein concentrations were determined via the Pierce BCA Protein Assay Kit (NCI3225CH; Thermo Fisher Scientific, Rockford, USA) following the manufacturer’s instructions. The samples were separated via 10% SDS-PAGE and transferred to polyvinylidene fluoride membranes (Immobilon ^®-P Transfer Membrane; Millipore, Bedford, USA). The membranes were blocked with 5% (w/v) bovine serum albumin (BSA) in Tris-buffered saline containing 0.1% Tween-20 (TBST) for 1 h at room temperature and then incubated with primary antibodies against KHK-A (1:1000 dilution, 21708; SAB Signalway, College Park, USA); KHK-C (1:1000 dilution, 21709; SAB Signalway); AQP2 (1:2000 dilution, A7310; Sigma-Aldrich, St Louis, USA); and HSP90 (1:10,000 dilution, TA500494; OriGene, Rockville, USA) diluted in antibody dilution buffer (1.5 g BSA, 0.1 g NaN ₃, 50 mL TBST) overnight at 4°C. Following TBST washes, the membranes were incubated with goat anti-mouse antibody (IRDye 800CW; LI-COR, Lincoln, USA) goat anti-rabbit antibody (IRDye 680LT; LI-COR) as secondary antibodies for 1 h at room temperature. The immunoblot signals were detected via an Odyssey system (OSA-0323; LI-COR) and quantified via Image-Pro Plus v6.0 software. The values were normalized to the mean intensity of the control groups, and the relative protein expression was normalized to that of the loading control HSP90.

RT-qPCR

The experimental protocol for RT-qPCR followed the methods outlined in our laboratory’s publication [23]. Briefly, total RNA was isolated from snap-frozen jejunum, liver, adipose, and kidney samples via TRIzol reagent 15596018; Invitrogen, Carlsbad, USA), followed by reverse transcription into cDNA via HiScript Q RT SuperMix (R122; Vazyme, Nanjing, China) according to the manufacturer’s instructions. The total RNA concentration was measured with a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific). Quantitative real-time PCR was performed via the LightCycler ^® 96 System (Roche, Ricardo Rojas, Argentina) and Hieff ^® qPCR SYBR Green Master Mix (11201ES; Yeasen, Shanghai, China) following the manufacturer’s protocol. The primer sequences were obtained from the PrimerBank database [24] and are provided in Supplementary Table S1. The relative mRNA expression of each gene was calculated as a relative value normalized to that of glyceraldehyde-3-phosphate dehydrogenase ( Gapdh).

Statistical analysis

The data are presented as the mean ± SEM. Statistical analysis was performed via IBM SPSS Statistics 19 software. Multiple comparisons were conducted via one-way analysis of variance (ANOVA) with a Bonferroni post hoc test. Statistical significance was defined as P  < 0.05.

Results

Liver-specific Khk deletion ameliorates hepatic steatosis and liver injury in fructose-fed mice

To elucidate the role of hepatic KHK in metabolic syndrome induced solely by HF, we generated liver-specific Khk-a/c knockout ( Khk ^fl/fl AlbCre ⁺) mice. In these mice, LoxP sequences flanked exons 3, 4, and 5 of the Khk gene, enabling Albumin-Cre-mediated deletion of both the KHK-A and KHK-C isoforms in the liver ( Figure 1A). Khk ^fl/fl AlbCre ^– (control) and Khk ^fl/fl AlbCre ⁺ (knockout) mice were administered with 20% fructose solution for 12 weeks, after which metabolic syndrome features were analyzed. Following 20% fructose consumption, hepatic Khk-a and Khk-c mRNA and KHK-A and KHK-C protein expression were significantly upregulated in Khk ^fl/fl AlbCre ^– mice. In contrast, expression levels were dramatically reduced to nearly undetectable levels in Khk ^fl/fl AlbCre ⁺ mice, regardless of fructose exposure ( Figure 1B,C). HF-induced liver injury (assessed by H&E staining, Figure 1D) and hepatic steatosis (assessed by Oil Red O staining, Figure 1E) were markedly attenuated in the Khk ^fl/fl AlbCre ⁺ mice compared with the control mice. This was accompanied by significantly decreased triglyceride levels in the plasma and liver ( Figure 1F), as well as reduced plasma ALT and AST levels ( Figure 1G). These data indicate that liver-specific Khk deletion protects against HF-induced hepatic steatosis and liver injury.

Figure 1 .

Effects of liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺ ) on liver histopathological features, plasma ALT and AST activities, and hepatic lipid accumulation

Mice were exposed to control water or 20% fructose (HF) in the drinking water for 12 weeks. (A) The gene targeting strategy was used to generate mice with liver-specific Khk knockout. (B) RT-qPCR analysis of Khk-a and Khk-c mRNA expression in the liver; expression was normalized to that of Gapdh. (C) Representative western blots and densitometry results for the KHK-A and KHK-C proteins in the liver. (D,E) Representative images of livers stained with H&E (D) and Oil Red O (E). (F) Plasma triglyceride levels and hepatic triglyceride levels. (G) Plasma ALT and AST levels. *P < 0.05, **P < 0.01, and ***P < 0.001 vs water; #P < 0.05, ##P < 0.01, and ###P < 0.001 vs Khkfl/flAlbCre–.

Liver-specific Khk deletion reduces body weight gain, fructose-derived caloric intake, and UA production in fructose-fed mice

When fed with 20% fructose solutions, Khk ^fl/fl AlbCre ⁺ mice displayed significantly less body weight gain than Khk ^fl/fl AlbCre ^– mice did ( Figure 2A), despite equivalent food consumption ( Figure 2B). Compared with Khk ^fl/fl AlbCre ^– mice, Khk ^fl/fl AlbCre ⁺ mice presented reduced fructose fluid intake ( Figure 2C) and urinary output ( Figure 2D). Although 24-h food-derived caloric content was similar, both 24-h total caloric intake and fructose-derived caloric content were significantly lower in Khk ^fl/fl AlbCre ⁺ mice than in Khk ^fl/fl AlbCre ^– mice ( Figure 2E), which correlated with the pronounced reduction in body weight gain.

Figure 2 .

General parameters of wild-type ( Khk ^{fl/ fl} AlbCre ^– ) and liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺ ) mice

Mice were exposed to control water or 20% fructose (HF) in the drinking water for 12 weeks. (A) Body weight gain. (B,C) Daily intake of food (B) and fluid (C). (D) Daily urine output. (E) Food-derived caloric intake, fructose-derived caloric intake, and daily total caloric intake. (F) Plasma fructose levels and daily urinary fructose excretion. (G) Fructose levels in the portal vein after oral gavage of 1 g/kg fructose. *P < 0.05, **P < 0.01, and ***P < 0.001 vs water; #P < 0.05 and ##P < 0.01 vs Khkfl/flAlbCre–.

Notably, while plasma fructose concentrations were comparable, fructose-fed Khk ^fl/fl AlbCre ⁺ mice exhibited a marked increase in urinary fructose excretion relative to fructose-fed Khk ^fl/fl AlbCre ^– mice ( Figure 2F). Under basal conditions, portal vein fructose levels after oral fructose gavage were similar between Khk ^fl/fl AlbCre ⁺ and Khk ^fl/fl AlbCre ^– mice ( Figure 2G). However, following 12 weeks of 20% fructose water consumption, portal vein fructose levels (as indicated by the AUC) postoral gavage were significantly lower in Khk ^fl/fl AlbCre ⁺ mice than in Khk ^fl/fl AlbCre ^– mice ( Figure 2G). These findings suggest that hepatic KHK-mediated fructose catabolism may exert a positive feedback regulatory effect on the rate of fructose delivery to the liver under conditions of HF intake.

Liver-specific Khk deletion also significantly reduced plasma UA concentrations and urinary UA excretion ( Figure 3A). We also detected the activity of xanthine oxidase (XOD), the rate-limiting enzyme for UA synthesis, and the mRNA levels of Glut9, a UA transporter. The consumption of 20% fructose solution significantly elevated XOD activity ( Figure 3B) and Glut9 mRNA levels ( Figure 3C) in hepatic and renal tissues in mice, all of which were notably attenuated by liver-specific Khk deficiency. These results suggest that UA signaling is inhibited in liver-specific Khk-deficient mice fed with 20% fructose solution.

Figure 3 .

Uric acid levels in wild-type ( Khk ^{fl/ fl} AlbCre ^– ) and liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺ ) mice

Mice were exposed to control water or 20% fructose (HF) in the drinking water for 12 weeks. (A) Plasma uric acid levels and daily urinary uric acid excretion. (B) Xanthine oxidase (XOD) activity in the liver and kidneys. (C) RT-qPCR analysis of Glut9 mRNA expression in the liver and kidneys, with expression normalized to Gapdh. **P < 0.01 and ***P < 0.001 vs water; #P < 0.05 and ##P < 0.01 vs Khkfl/flAlbCre–.

Liver-specific Khk deletion improves glucose resistance and insulin resistance in fructose-fed mice

HF intake is closely associated with glucose intolerance and insulin resistance [25]. The consumption of 20% fructose solution significantly elevated plasma glucose and insulin concentrations in mice; however, the hepatic-specific deletion of Khk reduced only plasma insulin levels without affecting plasma glucose levels ( Figure 4A,B). In response to 20% fructose administration, liver-specific Khk-deficient mice still exhibited glucose intolerance ( Figure 4C) and insulin resistance ( Figure 4D); however, these phenotypes were milder than those observed in the floxed controls. These results indicate that hepatic-specific Khk deficiency may play a protective role against fructose-induced hyperinsulinemia, glucose intolerance, and insulin resistance.

Figure 4 .

Effects of liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺) on glucose and insulin resistance of mice

Mice were exposed to water control or 20% fructose (HF) in the drinking water for 12 weeks. (A) Fasting blood glucose. (B) Plasma insulin levels. (C) Glucose tolerance test. (D) Insulin tolerance test. *P < 0.05, **P < 0.01, and ***P < 0.001 vs water; #P < 0.05 and ###P < 0.001 vs Khkfl/flAlbCre–.

Liver-specific Khk deletion suppresses the expression of genes related to fatty acid and triglyceride synthesis in the livers of fructose-fed mice

To further elucidate the potential mechanisms underlying the role of hepatic KHK in HF intake responses, transcriptomic sequencing was performed on liver tissues from floxed ( Khk ^fl/fl AlbCre ^–) and hepatic-specific Khk-deficient ( Khk ^fl/fl AlbCre ⁺) mice that were administered with either regular water or 20% fructose solution. Under normal water consumption, hepatic-specific Khk deletion resulted in the upregulation of 93 genes and the downregulation of 108 genes ( Figure 5A and Supplementary Figure S1). GO enrichment analysis revealed that these DEGs were enriched primarily in biological processes, including organic acid and carboxylic acid biosynthetic processes ( Figure 5B). In contrast, KEGG enrichment analysis highlighted a significant association with chemical carcinogenesis pathways ( Figure 5C). Using the STRING database combined with Cytoscape software for PPI network construction, we found that the downregulated Cyp2c37 (cytochrome P450, family 2, subfamily c, polypeptide 37), Cyp2c50, Cyp2c54 and upregulated Agxt (alanine-glyoxylate aminotransferase) and Cyp2b10 genes presented the highest degree values, suggesting that they may serve as key hub genes regulated by KHK ( Figure 5D). These findings indicate that Khk deletion potentially impacts cytochrome P450 isozymes, which modulate arachidonate and linoleic acid metabolic processes in the liver.

Figure 5 .

Transcriptome sequencing results of the livers of wild-type ( Khk ^{fl/ fl} AlbCre ^– ) and liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺ ) mice

Mice were exposed to control water or 20% fructose (HF) in the drinking water for 12 weeks. (A–D) Volcanic diagram (A), GO enrichment plot (B), KEGG enrichment plot (C), and PPI network (D) of differentially expressed genes between Khkfl/flAlbCre– and Khkfl/flAlbCre+ mice with normal water intake. (E–H) Volcanic diagram (E), GO enrichment plot (F), KEGG enrichment plot (G), and PPI network (H) of DEGs in Khkfl/flAlbCre– mice given normal water or 20% fructose solution. (I–L) Volcanic diagram (I), GO enrichment plot (J), KEGG enrichment plot (K), and PPI network (L) of differentially expressed genes between Khkfl/fl AlbCre– and Khkfl/fl AlbCre+ mice with 20% fructose solution intake. (M–P) Venn diagram (M), GO enrichment plot (N), KEGG enrichment plot (O), and protein-protein interaction (PPI) network (P) of the intersection genes of the DEGs caused by fructose consumption (E) and liver-specific Khk deficiency (I).

Compared with those in floxed mice ( Khk ^fl/fl AlbCre ^–), which consumed normal water, 32 genes were upregulated, and 22 genes were downregulated in fructose-drinking floxed mice ( Figure 5E and Supplementary Figure S2). GO enrichment analysis revealed that these DEGs were enriched primarily in biological processes, including organic acid and carboxylic acid biosynthetic processes, as well as coenzyme and cofactor metabolic processes ( Figure 5F). Additionally, KEGG enrichment analysis revealed a significant association with carbon metabolism pathways ( Figure 5G). PPI network analysis identified key hub genes among the fructose-induced upregulated genes, including AldoB, Acly (ATP citrate lyase), Acaca ( Accα, acetyl-CoA carboxylase alpha), Apoa4 (apolipoprotein A-IV), Gpam (glycerol-3-phosphate acyltransferase, mitochondrial), Cyp4a14 (cytochrome P450, family 4, subfamily a, polypeptide 14), Me1 (malic enzyme 1, NADP ⁺ dependent, cytosolic), Elovl6 (ELOVL fatty acid elongase 6), and Pgd (phosphogluconate dehydrogenase) ( Figure 5H). These genes are closely associated with hepatic fatty acid and triglyceride synthesis. Collectively, these results indicate that fructose intake stimulates fatty acid and triglyceride synthesis pathways in the liver.

Compared with fructose-fed floxed mice ( Khk ^fl/fl AlbCre ^–), fructose-fed liver-specific Khk-deficient mice ( Khk ^fl/fl AlbCre ⁺) presented upregulation of 22 genes and downregulation of 37 genes ( Figure 5I and Supplementary Figure S3). GO enrichment analysis indicated that these DEGs were enriched primarily in biological processes, including organic acid and carboxylic acid biosynthetic processes, as well as fatty acid, coenzyme, and cofactor metabolic processes ( Figure 5J). KEGG enrichment analysis further revealed a significant association with carbon and fatty acid metabolism pathways ( Figure 5K). Among these genes, the downregulated genes, including Fasn (fatty acid synthase), AldoB, Acly, Acaca, Acss2 (acyl-CoA synthetase short-chain family member 2), Gpam, Elovl6, Me1, Hmgcr (3-hydroxy-3-methylglutaryl-coenzyme A reductase), Cyp4a14, Pgd, and Apoa4, mediated by liver-specific Khk deletion were identified as key hub genes ( Figure 5L). Collectively, these findings suggest that hepatic-specific Khk deficiency inhibits fructose-stimulated hepatic fatty acid and triglyceride synthesis.

To further characterize the impact of hepatic-specific Khk deficiency, we analyzed the overlapping genes between fructose consumption-induced differential genes ( Figure 5E) and liver-specific Khk deficiency-induced differential genes ( Figure 5I). The intersection comprised 23 genes ( Figure 5M). GO and KEGG enrichment analyses revealed that these genes were enriched primarily in fatty acid biosynthetic/metabolic processes and carbon metabolism ( Figure 5N,O). PPI network analysis identified AldoB, Acly, Acaca, Gpam, Elovl6, Me1, Cyp4a14, Pgd, and Apoa4 as key hub genes ( Figure 5P). The reduced expression levels of these genes in the liver may underlie the inhibitory effect of hepatic-specific Khk deficiency on fructose-induced hepatic steatosis and liver injury.

We further examined the mRNA expression levels of key genes in the livers of floxed ( Khk ^fl/fl AlbCre ^–) and liver-specific Khk-deficient ( Khk ^fl/fl AlbCre ⁺) mice via RT-qPCR, including Glut2 (the primary transporter mediating fructose uptake from the portal circulation into hepatocytes [19]), genes associated with triglyceride synthesis ( AldoB, Acaca, Acacb, Acly, Gpam, Apoa4, Chrebp-α, and Chrebp-β), and genes involved in fatty acid synthesis ( Fasn, Me1, Elovl6, Cyp4a14, and Pgd). Notably, the fructose-induced upregulation of the expression of these genes in the liver was significantly attenuated in Khk ^fl/fl AlbCre ⁺ mice ( Figure 6). Collectively, these results demonstrate that liver-specific Khk deficiency effectively inhibits hepatic triglyceride and fatty acid synthesis pathways under HF conditions, which contributes, at least in part, to the amelioration of hepatic steatosis and liver injury.

Figure 6 .

Effects of liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺) on the expression of genes related to fatty acid and triglyceride synthesis in the mouse livers

Mice were exposed to control water or 20% fructose (HF) in the drinking water for 12 weeks. RT-qPCR analysis of Glut2, AldoB, Fasn, Acaca, Acacb, Chrebp-α, Chrebp-β, Acly, Gpam, Apoa4, Cyp4a14, Me1, Elovl6, and Pgd mRNA expression in the liver, with expression normalized to Gapdh. *P < 0.05, **P < 0.01, and ***P < 0.001 vs water; ##P < 0.01 and ###P < 0.001 vs Khkfl/flAlbCre–.

Liver-specific Khk deletion affects renal physiological functions in fructose-fed mice

Hepatic-specific Khk deficiency significantly decreased renal KHK-A and KHK-C protein expression levels in mice receiving 20% fructose solution ( Figure 7A). Moreover, the fructose-induced upregulation of Khk-a, Khk-c, Glut2, Glut5, and AldoB mRNA expression in both the renal cortex and medulla was notably attenuated in Khk ^fl/fl AlbCre ⁺ mice ( Figure 7B). These findings may be linked to the increased urinary fructose excretion ( Figure 2F) and decreased urinary UA excretion ( Figure 3A) observed in fructose-fed Khk ^fl/fl AlbCre ⁺ mice relative to fructose-fed Khk ^fl/fl AlbCre ^– controls. No significant pathological damage was detected in the renal tissues of either fructose-fed Khk ^fl/fl AlbCre ⁺ or Khk ^fl/fl AlbCre ^– mice ( Figure 7C). Notably, the fructose-induced downregulation of renal AQP2 protein ( Figure 7D) and Aqp2 and V ₂R mRNA expression ( Figure 7E) were also significantly reversed by hepatic-specific Khk deletion, which may be correlated with the reduced fluid intake ( Figure 2C) and urine output ( Figure 2D) in fructose-fed Khk ^fl/fl AlbCre ⁺ mice. Notably, compared with fructose-fed Khk ^fl/fl AlbCre ^– mice, fructose-fed Khk ^fl/fl AlbCre ⁺ mice presented significantly lower renal Slc9a3 (encoding NHE ₃ protein) mRNA levels ( Figure 7F), greater 24-h urinary Na ⁺ excretion ( Figure 7G), and greater urinary Na ⁺ excretion following acute intraperitoneal saline administration ( Figure 7H). Collectively, these results suggest that hepatic KHK inhibition may regulate renal physiological functions, including urine concentration capacity and Na ⁺ reabsorption, by modulating fructose absorption and catabolism in renal proximal tubules.

Figure 7 .

Renal impact of liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺) mice

Mice were exposed to water control or 20% fructose (HF) in drinking water for 12 weeks. (A) Representative western blots and densitometry results for the KHK-A and KHK-C proteins in the kidney. (B) RT-qPCR analysis of Khk-a, Khk-c Glut2, Glut5, and AldoB mRNA expression in the kidney; expression was normalized to that of Gapdh. (C) Representative images of kidneys stained with H&E. (D) Representative western blots and densitometry results for the AQP2 protein in the kidney. (E,F) RT-qPCR analysis of Aqp2, V2R (E) and Slc9a3 (F) mRNA expression in the kidney; expression was normalized to that of Gapdh. (G) 24-h urinary Na+ excretion. (H) Changes in urinary Na+ excretion (6 h) in response to i.p. administration of 1.5 mL of normal saline. *P < 0.05, **P < 0.01, and ***P < 0.001 vs water; #P < 0.05 and ##P < 0.01 vs Khkfl/flAlbCre–.

Liver-specific Khk deletion inhibits fructose absorption and metabolism in the intestines of fructose-fed mice

Hepatic-specific Khk-deficient mice presented reduced KHK-A and KHK-C protein expression in the jejunum relative to fructose-fed Khk ^fl/fl AlbCre ^– control mice ( Figure 8A). Additionally, the fructose-induced upregulation of Khk-a, Khk-c, Glut2, Glut5, and AldoB mRNA expression in the jejunum was significantly attenuated in Khk ^fl/fl AlbCre ⁺ mice ( Figure 8B). However, neither the shortened jejunal villus length ( Figure 8C,D) nor the reduced mRNA expression of tight junction genes (including Zo-1, Occludin, and Claudin-1) ( Figure 8E) was affected by hepatic-specific Khk deletion. Collectively, these findings suggest that under HF conditions, hepatic KHK may specifically modulate intestinal fructose metabolism and absorption, thereby regulating fructose delivery to the liver without impacting intestinal epithelial barrier integrity in mice.

Figure 8 .

Effects of liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺) on intestinal morphology and gut permeability

Mice were exposed to water control or 20% fructose (HF) drinking for 12 weeks. (A) Representative western blots and densitometry results for KHK-A and KHK-C in the jejunum. (B) RT-qPCR analysis of Khk-a, Khk-c, Glut2, Glut5, and AldoB mRNA expression in the jejunum; expression was normalized to that of Gapdh. (C,D) Representative H&E-stained images of villi in the jejunum (C), and the villus length was quantified (D). (E) RT-qPCR analysis of the mRNA expression of tight junction genes (Zo-1, Occludin, and Claudin-1) in the jejunum; expression was normalized to that of Gapdh. *P < 0.05, **P < 0.01, and ***P < 0.001 vs water; #P < 0.05 and ##P < 0.01 vs Khkfl/flAlbCre–.

Liver-specific Khk deletion improves adipose tissue changes and inflammation in fructose-fed mice

HE staining revealed that fructose-induced phenotypes in white adipose tissue (WAT) and brown adipose tissue (BAT), including increased adipocyte size and reduced adipocyte number, as previously reported [26], were significantly attenuated by hepatic-specific Khk deficiency ( Figure 9A,B). Consistent with previous reports of reduced expression of metabolic regulatory factors, such as uncoupling protein 1 (UCP1) and peroxisome proliferator-activated receptor gamma coactivator 1-alpha (PGC-1α), in the BAT of fructose-fed mice [26], we detected downregulated Ucp1 and Pgc-1α mRNA expression in the BAT of fructose-fed floxed mice; notably, this downregulation was significantly reversed in hepatic-specific Khk-deficient mice ( Figure 9C). However, under baseline conditions, while hepatic-specific Khk deletion increased Pgc-1α mRNA levels but decreased Ucp1 mRNA levels in WAT, it had no effect on fructose-induced Pgc-1α mRNA upregulation or fructose-mediated Ucp1 mRNA downregulation in mice ( Figure 9C). These findings suggest potential differential regulation of WAT versus BAT Ucp1 and Pgc-1α mRNA expression by fructose or hepatic KHK. Additionally, fructose-stimulated mRNA expression of the inflammatory cytokines tumor necrosis factor-α ( Tnf-α) and interleukin-6 ( Il-6) in the WAT and BAT was significantly abolished in hepatic-specific Khk-deficient mice ( Figure 9D). Overall, hepatic-specific Khk deficiency may influence the thermogenic capacity of adipose tissues and adipocyte inflammation during fructose intake.

Figure 9 .

Effect of liver-specific Khk knockout ( Khk ^{fl/ fl} AlbCre ⁺ ) on adipose tissue changes and inflammation

Mice were exposed to water control or 20% fructose (HF) in the drinking water for 12 weeks. (A,B) Representative histological results of brown adipose tissue (BAT, A) and white adipose tissue (WAT, B) by hematoxylin and eosin staining. The size and number of adipocytes were quantified. (C) RT-qPCR analysis of Ucp1 and Pgc-1α (peroxisome proliferator-activated receptor, gamma, coactivator 1 alpha) mRNA expression in BAT and WAT, with expression normalized to Gapdh. (D) RT-qPCR analysis of the mRNA expression of the inflammatory cytokines Tnf-α and Il-6 in BAT and WAT, with the expression normalized to that of Gapdh. *P < 0.05, **P < 0.01, ***P < 0.001 vs water; #P < 0.05 and ##P < 0.01 vs Khkfl/flAlbCre–.

Discussion

The liver serves as a primary organ for fructose uptake and metabolism and is closely linked to fructose-induced pathological conditions [19]. In the present study, as an extension of prior work [13], we investigated the impact of liver-specific Khk knockout on metabolic syndrome directly caused by HF intake alone to rule out the potential influence of high-glycemic carbohydrates in driving metabolic syndrome. We demonstrated that liver-specific KHK-A/C deficiency inhibited 20% fructose solution-induced metabolic dysfunction in mice, as evidenced by reduced body weight gain, alleviated liver injury and hepatic steatosis, improved glucose and insulin tolerance, and mitigated adipose tissue alterations and inflammation. These findings underscore the importance of hepatic KHK in fructose-loading-induced metabolic syndrome. By excluding the confounding effects of high glucose, our results more accurately reflect the role of liver KHK in fructose-induced metabolic syndrome. Furthermore, we made progress in the following key aspects.

First, our study not only further corroborates previous investigations demonstrating the protective effects of global or liver-specific KHK inhibition against fatty liver and metabolic syndrome [ 10, 11, 13– 16, 20] but also reveals distinct findings regarding energy intake. Unlike the observations of Andres-Hernando et al. [13], where total caloric intake and fructose-derived caloric intake remained comparable between liver-specific Khk-deficient mice and control mice fed ad libitum with fructose/glucose, we observed that fructose-fed liver-specific Khk-deficient mice presented significantly lower total caloric intake and fructose-derived caloric intake than fructose-fed controls did. This discrepancy may be attributed to the presence or absence of glucose in the ad libitum diet. Notably, while a hypothesis [2] suggests that fructose-induced metabolic syndrome may occur independently of excessive energy intake, this hypothesis has been challenged by a recent original study [13]. Specifically, although gut Khk-a/c knockout alleviated fructose/glucose mixture-induced metabolic syndrome and reduced energy intake, when total calories (including those from the fructose/glucose mixture) were matched between genotypes, gut Khk-a/c knockout mice and controls presented comparable obesity and metabolic syndrome severity [13]. These findings indicate that reduced total and fructose-derived caloric intake contributes, at least in part, to the improvements in fatty liver and metabolic syndrome observed in liver-specific Khk-deficient mice under HF exposure. This mechanism requires validation through experiments with matched energy intake between liver-specific Khk-deficient and floxed control mice.

Second, liver-specific Khk deletion may influence intestinal fructose absorption and metabolism under HF conditions. Fructose is believed to be transported primarily from the intestinal lumen into epithelial cells via GLUT5 (located in the apical membrane) [27] and then diffuses out of enterocytes into the hepatic portal vein through GLUT2 (in the basolateral membrane) [28]. In the liver, GLUT2 further mediates fructose transport from the portal circulation into hepatocytes [19]. Notably, compared with HF-fed floxed controls, liver-specific KHK-A/C-deficient mice fed with an HF diet presented reduced jejunal expression of KHK-A, KHK-C, Glut5, Glut2, and AldoB. This effect was accompanied by decreased hepatic fructose delivery, indicating that the inhibition of jejunal fructose absorption, transport, and catabolism may contribute to the amelioration of metabolic syndrome features in these HF-fed mice. Notably, the reduced jejunal expression of KHK-A and KHK-C may represent the core driver of these changes. Findings from intestinal-specific KHK-A/C-deficient mice may support this hypothesis: (1) intestinal KHK-A/C deletion resulted in greater hepatic fructose delivery following acute fructose/glucose loading, a phenomenon likely linked to significantly elevated jejunal Glut5 and Glut2 expression under normal conditions [ 13, 29] ; (2) while intestinal Khk-a/c knockout mice and controls presented comparable metabolic syndrome features when total caloric intake and fructose/glucose-derived caloric intake were matched, intestinal KHK-A/C deletion still partially mitigated fructose/glucose-induced metabolic syndrome by reducing total caloric intake and fructose-derived caloric intake [13]. However, further investigation is needed to clarify whether the downregulation of jejunal KHK-A and KHK-C induced by liver-specific Khk deficiency contributes to alleviating fructose-induced metabolic syndrome, particularly in intestinal Khk-deficient mice subjected to only HF loading.

Third, liver-specific Khk deletion exerts multifaceted influences on renal physiological functions under HF intake. On the one hand, compared with HF-fed floxed controls, liver-specific Khk-a/ c-knockout mice fed with an HF diet presented reduced renal expression of KHK-A, KHK-C, Glut5, Glut2, and AldoB. This downregulation indicates weakened renal fructose absorption and catabolism, which may contribute to elevated urinary fructose excretion and reduced UA excretion. The attenuated renal fructose metabolism may further enhance the protective effect of liver-specific Khk knockout against fructose-induced fatty liver and metabolic syndrome. However, the mechanism by which hepatic KHK-mediated fructose metabolism defects affect renal fructose absorption and catabolism remains unclear. On the other hand, liver-specific Khk knockout significantly mitigated fructose-induced reductions in renal AQP2 and V ₂R expression, as well as polydipsia and polyuria, in mice [30]. This was evidenced by reduced 24-h water intake and urine volume, along with increased renal AQP2 and V ₂R expression levels in HF-fed hepatic-specific Khk knockout mice compared with those in HF-fed floxed controls. These findings suggest that liver-specific Khk deficiency may improve fructose-induced water imbalance by increasing the urine concentration capacity through activation of the renal V ₂R/AQP2 signaling pathway. Finally, liver-specific Khk deficiency abolished fructose-stimulated renal NHE ₃ expression and activity [ 31– 34] , as well as fructose-induced reductions in urinary Na ⁺ excretion [35]. This observation corroborates the inhibitory effect of global Khk deletion on fructose-stimulated renal NHE ₃ in mice [36]. Collectively, these results consistently indicate that KHK blockade inhibits fructose-induced renal NHE ₃ activation, which may be attributed to the prevention of fructose-induced UA generation [35]. Notably, fructose-stimulated NHE ₃ activation in proximal tubules has been shown to contribute to HF-induced salt-sensitive hypertension in rats, as determined by radiotelemetry [ 31– 33] . Although Hayasaki et al. [36] reported that a combination of 10% fructose solution and a 4% NaCl diet induced salt-sensitive hypertension in mice (monitored via the tail-cuff method), which was attenuated in global Khk knockout mice, we did not observe blood pressure elevation (via radiotelemetry) in either liver-specific Khk-deficient mice or floxed mice fed with a 20% fructose solution plus an 8% NaCl diet (data not shown). This discrepancy may be related to differences in blood pressure monitoring methods: radiotelemetry, rather than tail-cuff plethysmography, is more suitable for accurately assessing blood pressure variability when changes in murine blood pressure are relatively subtle [37].

UA, a key metabolite of fructose, plays a critical role in high-fructose HF-induced metabolic syndrome [ 38– 41] . Hepatic-specific Khk knockout significantly reduced plasma UA levels, urinary UA excretion, XOD activity and Glut9 mRNA levels in the liver and kidneys of mice that consumed 20% fructose solution. This was accompanied by a marked increase in urinary fructose excretion. The reduction in UA levels may contribute to the amelioration of fatty liver and metabolic syndrome observed in liver-specific Khk-deficient mice. Previous studies have shown that both allopurinol (an XOD inhibitor that blocks UA production) [ 38, 40] and benzbromarone (a uricosuric agent that promotes UA excretion) [38] significantly improved HF-induced hepatic steatosis and metabolic syndrome while also reducing intracellular UA levels in the livers of rats. Notably, the administration of allopurinol to block UA production has been shown to significantly downregulate fructose-stimulated KHK expression in both the liver [40] and kidneys [35]. Conversely, UA stimulates ChREBP-mediated Khk mRNA transcription, thereby increasing the sensitivity of hepatocytes to fructose metabolism [40]. These findings suggest that the reduced UA generation in liver-specific Khk-deficient mice may explain the observed lower KHK levels in the jejunum and kidneys. In summary, the reduction in UA may contribute to the improvement in HF-induced fatty liver and metabolic syndrome in liver-specific Khk-deficient mice; this hypothesis requires further validation through exogenous UA supplementation experiments in these mice.

In conclusion, our study underscores the critical role of liver KHK-mediated fructose metabolism in driving fructose-induced physiological and pathological effects along the intestinal-liver-kidney axis. Specifically, we demonstrated that liver-specific Khk deletion confers protection against fructose-induced fatty liver and metabolic syndrome through multiple mechanisms: inhibition of hepatic fatty acid and triglyceride synthesis, reduction in UA production, and attenuation of intestinal and renal fructose absorption and metabolism. Notably, while the intestine, liver, and kidney are primary sites of fructose metabolism, tissues, including skeletal muscle, adipose tissue, pancreas, brain, testes, red blood cells, and ocular lens, can also metabolize substantial amounts of fructose, particularly under conditions of KHK deficiency in primary metabolic organs or HF intake [42]. Thus, the potential roles of KHK in extraintestinal-liver-kidney axis tissues warrant further clarification. Moreover, KHK-A and KHK-C have opposite effects on fructose-induced fatty liver and metabolic syndrome [ 10, 18, 43] . Our findings highlight the need for targeted investigations into the distinct functions of liver-specific KHK-A and KHK-C in fructose-induced metabolic disorders. This is especially critical when models with liver-specific deletion of Khk-a or Khk- c are used to dissect their contributions.

Supplementary Data

Supplementary data are available at Acta Biochimica et Biophysica Sinica online.

COMPETING INTERESTS

The authors declare that they have no conflict of interest.

Funding Statement

This work was supported by the grants from Jiangxi “Double Thousand Plan” (No. jxsq2020101074), the Jiangxi Provincial Natural Science Foundation (No. 20232BAB206018), the National Natural Science Foundation of China (Nos. 82160051 and 32100908), the Science and Technology Research Project in Education Department of Jiangxi Province (No. GJJ2200904), the Ph.D. Start-up Research Fund in Jiangxi University of Chinese Medicine (No. 2020BSZR009), the Science and Technology Research Project in Health Commission of Jiangxi Province (No. 202311143), the Discipline of Chinese and Western Integrative Medicine in Jiangxi University of Chinese Medicine (Top Discipline of Jiangxi Province, No. zxyylxk20220103), and the Scientific and Technological Innovation Team Grant of the Jiangxi University of Chinese Medicine (No. CXTD22014).