AKR1B1 drives hyperglycemia-induced metabolic reprogramming in MASLD-associated hepatocellular carcinoma

Abstract

Background & aims: The mechanism behind the progressive pathological alteration in metabolic dysfunction-associated steatotic liver disease/steatohepatitis (MASLD/MASH)-associated hepatocellular carcinoma (HCC) is poorly understood. In the present study, we investigated the role of the polyol pathway enzyme AKR1B1 in metabolic switching associated with MASLD/MASH and in the progression of HCC.

Methods: AKR1B1 expression was estimated in the tissue and plasma of patients with MASLD/MASH, HCC, and HCC with diabetes mellitus. The role of AKR1B1 in metabolic switching in vitro was assessed through media conditioning, lentiviral transfection, and pharmacological probes. A proteomic and metabolomic approach was applied for the in-depth investigation of metabolic pathways. Preclinically, mice were subjected to a high-fructose diet and diethylnitrosamine to investigate the role of AKR1B1 in the hyperglycemia-mediated metabolic switching characteristic of MASLD-HCC.

Results: A significant increase in the expression of AKR1B1 was observed in tissue and plasma samples from patients with MASLD/MASH, HCC, and HCC with diabetes mellitus compared to normal samples. Mechanistically, in vitro assays revealed that AKR1B1 modulates the Warburg effect, mitochondrial dynamics, the tricarboxylic acid cycle, and lipogenesis to promote hyperglycemia-mediated MASLD and cancer progression. A pathological increase in the expression of AKR1B1 was observed in experimental MASLD-HCC, and expression was positively correlated with high blood glucose levels. High-fructose diet + diethylnitrosamine-treated animals also exhibited statistically significant elevation of metabolic markers and carcinogenesis markers. AKR1B1 inhibition with epalrestat or NARI-29 inhibited cellular metabolism in in vitro and in vivo models.

Conclusions: Pathological AKR1B1 modulates hepatic metabolism to promote MASLD-associated hepatocarcinogenesis. Aldose reductase inhibition modulates the glycolytic pathway to prevent precancerous hepatocyte formation.

Impact and implications: This research work highlights AKR1B1 as a druggable target in metabolic dysfunction-associated steatotic liver disease (MASLD) and hepatocellular carcinoma (HCC), which could provide the basis for the development of new chemotherapeutic agents. Moreover, our results indicate the potential of plasma AKR1B1 levels as a prognostic marker and diagnostic test for MASLD and associated HCC. Additionally, a major observation in this study was that AKR1B1 is associated with the promotion of the Warburg effect in HCC.

Keywords: HCC; MASLD/MASH; Metabolism; NARI-29; Warburg effect; diethyl nitrosamine; epalrestat; high fructose diet.

Graphical abstract

Fig. 7

Aldose reductase inhibition reverses high fructose-aggravated DEN-induced MASLD in mice. (A) Pictorial representation of in vivo study design. (B) Line plot depicting area under the curve (plasma conc. vs. time) of NARI-29 (25 and 50 mg/kg). Data are represented as mean ± SEM (n = 4). (C) Representative liver images of various experimental groups. Bar graph illustrating (D) percentage change in body weight (each bar represents mean ± SEM and each point represents individual animal data). (E) Serum glucose levels (mg%) (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗∗∗p <0.001 and ^ˆˆˆp <0.001 (one-way ANOVA followed by Tukey test). (F) Liver index: liver weight/body weight (%) (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗∗∗p <0.001 ^ˆp <0.05, and ^ˆˆˆp <0.001) (one-way ANOVA followed by Tukey test), (G) serum AST levels (IU) (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗∗∗p <0.001, ^ˆp <0.05, and ^ˆˆˆp <0.01 (one-way ANOVA followed by Tukey test). (H) Serum ALT levels (IU) (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗∗∗p <0.001, and ^ˆˆˆp <0.001; (one-way ANOVA followed by Tukey test) and (I) Liver triglyceride levels (mg/g tissue) of various experimental groups (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗∗∗p <0.001, ^ˆp <0.05, and ^ˆˆˆp <0.001 (one-way ANOVA followed by Tukey test). DC, disease control; DEN, diethylnitrosamine; EPS, Epalrestat; ORO, Oil red O staining; VC, vehicle control.

Fig. 1

AKR1B1 associates with MASLD/MASH, HCC, and DM+HCC progression. (A) Pictorial representation of study outline. (B) AKR1B1 gene expression in normal and HCC tumor samples using TNM plot. Survival analysis of patients with non-alcohol-related HCC in (C) all stages and (D) stage 3. (E). Study design for clinical sample analysis. (F) Immunohistochemistry and (G) violin plots for AKR1B1 expression in MASH/MASLD, HCC, and DM+HCC (each dot represents individual data, and joints represent median values). Level of significance: ∗∗p <0.01, ∗∗∗p <0.0001, and^#p <0.05 (one-way ANOVA followed by Tukey test). Plasma expression of AKR1B1 in (H) MASLD and MASLD+HCC compared with normal control and (I) HCC and DM+HCC compared with normal control. Level of significance: ^∗p <0.05, ∗∗p <0.01; (one-way ANOVA followed by Tukey test). (J, K) expression of AKR1B1 in different stages of HCC. Data is represented as mean±SEM (n ≥6). Each dot represents individual data and the line inside the box represents median values. Level of significance: ∗p <0.05, ∗∗p <0.01; (one-way ANOVA followed by Tukey test). DM, diabetes mellitus; HCC, hepatocellular carcinoma; NAFLD, non-alcoholic fatty liver disease; NASH, non-alcoholic steatohepatitis; NAT, normal adjacent tissue.

Fig. 2

Hyperglycemia or lentiviral transfection-mediated overexpression of AKR1B1 and metabolic reprogramming in human HCC cell lines. (A) Pictorial representation of study outline. (B) Immunoblotting represents the differential expression of AKR1B1 in hepatic cancer cell lines. (C) Graphical representation of lactate secretion in culture media after 24 h of maintenance in LG (2 g/L glucose) and HG (4.5 g/L glucose) (each dot represents individual data and the line inside the box represent median values. Level of significance: ∗∗p <0.01, ∗∗∗p <0.001; (two-way ANOVA followed by Tukey test). (D,G) Immunoblotting represents the differential expression of AKR1B1 and metabolic proteins like HK-II, cMYC, MCT-4, LDHA, and KHK in the HepG2 cell line. (E,H) Densitometric analysis of AKR1B1, HK-II, cMYC, MCT-4, LDHA, and KHK immunoblots (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (F) Oil Red O staining of HepG2 cell line insulted with BSA-palmitate under LG and HG. (G) Immunofluorescence and immunoblotting assay depicting overexpression of AKR1B1 in PLC/PRF-5 cells using lentiviral particle. (J,K) Immunoblotting represents the differential expression of AKR1B1 and metabolic proteins like HK-II, cMYC, LDHA, and KHK. (J,L) Densitometric analysis of AKR1B1, HK-II, cMYC, LDHA, and KHK (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (M) Graphical representation of lactate secretion in culture media of PLC/PRF-5 and AKR+ cell lines after 24 h of maintenance in LG (2 g/L glucose) and HG (4.5 g/L glucose) (each bar represents mean ± SEM). Level of significance: ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (N) Line plot displaying percentage viability of PLC/PRF-5-AKR1B1+ cells upon treatment with EPS and NARI-29 at varying concentrations. BSA-PA, Bovine serum alcohol-palmitic acid; EPS, Epalrestat; HCC, hepatocellular carcinoma; HG, high glucose; LG, low glucose; MOI, multiliplicity of infection; NAFL, non-alcoholic fatty liver.

Fig. 5

AKR1B1 inhibition promotes mitochondrial fusion. (A) Representative immunofluorescence images for HK-II and Mitotracker staining in HepG2 cell line (HG: 4.5 g/L) upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM). (B) Graphical representation of HK-II correlation coefficient (each bar represents Mean ± SEM). Level of significance: ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (C) Representative immunoblots and bar graphs showing differential expression of MFN-1/DRP-1 ratio in HepG2 cell line (HG: 4.5 g/L) (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (D) Representative images for mitochondrial network in single cell generated from mitochondria Analyser ImageJ plugin. (E) Bar graph showing the mitochondrial aspect ratio upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM) (each bar represents mean ± SEM). Level of significance: ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (F) Representative images for mitosox staining assay. (G) Representative images for JC-1 staining and a dot plot showing JC-1 aggregated (red) and monomers (green) were performed using flow cytometry. (H) Bar graph showing mean fluorescence intensity of mitosox upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM) (each bar represents mean ± SEM). Level of significance: ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (I) Graphical representation of JC-1 red fluorescence upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM). (each bar represents mean ± SEM). Level of significance: ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test) (I) Representative immunoblots and bar graphs showing the ratio of Bax to BCL-2 in the HepG2 cell line (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). EPS: Epalrestat, LD: Low dose, and HD: High dose.

Fig. 3

AKR1B1 inhibition reverses polyol flux and GLUT-1 expression. (A) Line plot showing percentage cell viability of NARI-29 and EPS in the different hepatic cancer cell lines. Representative immunoblots and bar graphs show differential expression of AKT, pAKT, cMYC, GLUT-1, and KHK on (B) PLC/PRF-5-AKR+ cell line and (F) HepG2 cell line (HG: 4.5 g/L) (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (C) Representative images of fluorescence diacetate and propidium iodide dual staining assay depicting live and dead cells in HepG2 spheroid culture upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM) (each bar represents mean ± SEM). Level of significance: ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). Representative immunofluorescence images for KHK on (D) HepG2 cell line (HG: 4.5 g/L) and (E) PLC/PRF-5-AKR+ cell line. EPS: Epalrestat, LD: Low dose, and HD: High dose.

Fig. 4

Reversal of Warburg effect through AKR1B1 inhibition. (A) Representative immunoblots and bar graphs show differential expression of glycolytic markers HK-II, PFK1, PKM2, LDHA, and MCT-4 on HepG2 cell line (HG: 4.5 g/L) and PLC/PRF-5-AKR+ cell line coupled with metabolomics data for glucose-6 phosphate, fructose 1, 6 bisphosphate, phosphoenolpyruvate, lactate and extracellular lactate in HepG2 cell line upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM) for 24 h (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). (B) Representative images of ORO staining about oil globule deposition (captured using 200X magnification) and bar graph showing ORO relative absorbance in HepG2 cell line after 24 h of insult with BSA-palmitate under hyperglycemia (4.5 g/L glucose) with or without AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM) (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001; (two-way ANOVA followed by Tukey test). (C) Representative immunofluorescence images for LDHA in HepG2 and PLC/PRF-5-AKR+ cell line (HG: 4.5 g/L). BSA-PA, Bovine serum alcohol-palmitic acid; EPS, Epalrestat; HD, high dose; HG, high glucose;LD: Low dose; ORO, Oil red O staining.

Fig. 6

AKR1B1 inhibition impairs the TCA cycle and depletes ATP to induce apoptosis. (A) Graphical representation of ATP content measured using ELISA (each bar represents mean ± SEM). Level of significance: ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test) (B) Heat map for change in metabolic proteins analyzed by proteomics in HepG2 cell line. (C) Pictorial representation for expression of TCA cycle enzymes upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM) (C: Control, H: High glucose). (D) Representative images of acridine orange and ethidium bromide dual staining. (E) Dot plot showing annexin V/PI dual staining to confirm apoptosis performed using flow cytometry. (F) Representative immunoblots and (G) bar graph showing differential expression of PARP and CL-PARP upon AKR1B1 inhibition with NARI-29 (LD: 10 μM and HD: 20 μM) and EPS (50 μM) (each bar represents mean ± SEM). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001; (one-way ANOVA followed by Tukey test). LD: Low dose, HD: High dose, EPS: Epalrestat, H: High glucose, and C: Control.

Fig. 8

Aldose reductase inhibition reverses the metabolic modulation in mouse liver. Representative immunofluorescence images for (A) AKR1B1 and CD44, (D) GLUT-1, (E) KHK and E-cadherin, and (G) Ki-67 in mouse liver tissue sections of various experimental groups. (B) Bar graph showing the expression of AKR1B1 in mouse plasma (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗∗∗p <0.001, ^ˆp <0.05, and ^ˆˆp <0.01 (one-way ANOVA followed by Tukey test). (C) Representative (i) immunoblots and bar graph showing differential expression of (ii) cMYC, (iii) AKR1B1, and (iv) KHK in live tissue lysate (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗p <0.05, ∗∗p <0.01, ∗∗∗p <0.001, ^ˆp <0.05, ^ˆˆp <0.01, and ^ˆˆˆp <0.001 (one-way ANOVA followed by Tukey test). (F) Representative (i) immunoblots and bar graph showing differential expression of (ii) PFK1, (iii) PKM2, (iv) LDHA, and (v) MCT-4 in live tissue lysate (each bar represents mean ± SEM and each point represents individual animal data). Level of significance: ∗∗p <0.01, ∗∗∗p <0.001, ^ˆˆp <0.01, and ^ˆˆˆp <0.001 (one-way ANOVA followed by Tukey test). DC, disease control; EPS, Epalrestat; VC, vehicle control.