Kupffer cell death is a characteristic feature of MASLD progression.

(A-D) Male Wild-type C57BL/6J mice were fed a high-fat high-cholesterol diet (HFHC) for 0, 4, or 16 weeks. (A) KCs death was assessed by immunostaining of liver sections for TIM4(KCs marker, red), TUNEL (green), and DAPI (nuclei, blue). Scale bar: 50 µm (main panels) and 20 µm (Inset). (B) KCs death was quantified. n=4 mice/group. (C) Flow cytometry analysis of KCs (CD45+ F4/80hi CD11blow TIM4+) and IMs (CD45+ F4/80lowCD11bhi TIM4-) among isolated NPCs. (D) KCs counts were quantified. n=4-5 mice/group. (E-F) Male wild-type C57BL/6J mice were fed either: (E) Normal chow diet (NCD) or high-fat diet (HFD) for 20 weeks, or (F) NCD or methionine-choline-deficient diet (MCD) for 6 weeks. KCs death was assessed by immunostaining of liver sections for TIM4(green), TUNEL (red), and DAPI (nuclei, blue). Scale bar: 50 µm (main panels) and 20 µm (Inset). KCs death was quantified. n=4 mice/group. Representative images are shown in A, C, E, F. One-way ANOVA (B, D). Unpaired Student’s t-test (E, F). P value as indicated.

Kupffer cells exhibit early and zone-specific susceptibility to death during MASLD progression.

(A-E) Male Wild-type C57BL/6J mice were fed a HFHC diet for 0, 4, or 16 weeks. (A-C) Hepatic cell death was assessed by co-staining TUNEL with: (A) HNF4α (hepatocytes), (B) Desmin (hepatic stellate cells, HSCs), (C) Iba1 (hepatic macrophages), and DAPI (nuclei, blue). Scale bars: 50 µm (main panels). Hepatic cell death was quantified (n = 4 mice/group). (D) Zonal distribution of KCs death was evaluated by co-staining TIM4 (KCs), TUNEL, Glutamine Synthetase (GS, central vein marker) and DAPI (nuclei, blue). Scale bars: 50 µm. Zonal distribution of KCs death was quantified (n = 6-7 mice/group). FOV: field of view. PV: portal vein. CV: central vein. Representative images are shown in A-D. One-way ANOVA (A-D). P value as indicated.

Kupffer cells exhibit metabolic reprogramming with increased glycolysis during early MASLD.

(A) Experimental design for metabolomic analysis of KCs isolated from male wild-type mice fed a HFHC diet for 0, 4 or 8 weeks. n=3 mice/group. (B) Principal component analysis (PCA) of enriched metabolites in KCs across different dietary durations. (C-D) Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of metabolic pathways upregulated in KCs at 4 weeks (C) or 8 weeks (D). The glucose metabolism pathway is highlighted by red rectangles. (E) Heatmap depicting significantly altered metabolites involved in glucose metabolism pathways in KCs across different dietary durations. (F) Heatmap depicting significantly altered metabolites involved in cell death in KCs across different dietary durations.

Excessive glucose metabolic activity contributes to Kupffer cell death.

(A-B) Isolated KCs were treated for 24 h with: 5.5 mM glucose + isopropanol (control), 5.5 mM glucose + 800 µM palmitic acid (PA), 10 mM glucose + 800 µM PA. Cell viability was assessed by Cleaved caspase-3 (Cl-Casp3) staining (Cl-Casp3+ cells = dead). Scale bars: 20 µm (main panels), 5 µm (insets). Cl-Casp3 was detected by Western blot. (C-D) Isolated Kupffer cells were treated for 24 h with: Blank (no treatment), DMSO (vehicle control), 20 µM PS48 (PDK1 activator). Scale bars: 20 µm (main panels), 5 µm (insets). Cell death were analyzed as above. (E) Experimental design. Male WT mice were fed a high-fat, high-cholesterol (HFHC) diet for 5 weeks. From the 3rd week onward, mice received intraperitoneal injections of either vehicle or 2-DG (50 mg/kg) every other day (n = 3-4 mice/group). (F-G) Effects of glycolysis inhibition on KCs death after 5 weeks of HFHC feeding. (F) Representative images of liver sections co-stained with TUNEL and the KCs marker TIM4. (G) Quantification of TUNEL+ KCs. Data are presented as mean ± SEM. Statistical analysis was performed using one-way ANOVA (A, C) and an unpaired Student’s t-test (G); P values are indicated.

Enhanced glycolytic flux in Chil1-/- macrophages.

(A) Schematic diagram depicting the fate of glucose-derived ribose carbons in WT mouse primary hepatocytes. (B) Principal component analysis (PCA) of metabolites in WT and Chil1-/- BMDMs cultured with [U-13C]glucose. (C) Heatmap depicting significantly altered Glycolysis and Pentose phosphate (PPP) metabolites in WT and Chil1-/- BMDMs. (D) Glucose metabolic flux analysis in WT and Chil1-/- BMDMs cultured with [U-13C]glucose showing mass isotopologue distributions of: Glycolytic intermediates (Glc, F6P, FBP, 3PGA, 2PGA, PEP, PA, LA, G6P). PPP intermediates (Ru5P, R5P, S7P, DHAP). Data represent n= 6 biological replicates/group. (E-F) Extracellular acidification rate (ECAR) analysis of WT or Chil1-/- BMDMs cells. BMDM were sequentially treated with Glucose, oligomycin and 2-DG as indicated during seahorse. Unpaired student t-test (D,F). P*<0.05, P**<0.01, P***<0.001.

Enhanced glycolysis accelerated Kupffer cell death during MASLD.

(A) Cleaved caspase-3 (Cl-Casp3) staining to detect WT and Chil1-/- Kupffer cell death. Cells were under treatment without (Blank) or with either Isopropyl alcohol (Iso) or Palmitic Acid (PA) for 24 h. Scale bar: 20μm. (B) Cl-Casp3+ cells were quantified. (C) LDH release measurement in culture medium of KCs isolated from male WT and Chil1-/-mice was measured after treatment for 24 h with: Blank (no treatment), ISO (vehicle control), 800 µM PA. (D) Flow cytometry analysis of KCs (CD45+ F4/80hi CD11blow TIM4+) and MoMFs (CD45+ F4/80low CD11bhi TIM4-) among NPCs in Clec4f-cre and Clec4fΔChil1 mice fed HFHC diet for 0 or 16 weeks. (E) KCs counts were quantified. n= 4 mice/group. (F) Kupffer cell death was assessed by immunostaining of TIM4 (KCs marker, green), TUNEL (red), and DAPI (nuclei, blue) in liver sections from Clec4f-cre and Clec4fΔChil1 mice fed HFHC diet for 0 or 16 weeks. Scale bar: 50μm (main panels) and 20μm (Inset). (G) KCs death was quantified. n=4 mice/group. Representative images shown (A, D, F). Unpaired student t-test (B,C,E,G). P value as indicated.

Excessive glycolysis enhancement promotes Kupffer cell death in MASLD.

(Left) Under physiological conditions, Kupffer cells (KCs) maintain basal glucose metabolism supporting cellular homeostasis and survival. (Right) During MASLD progression, KCs undergo excessive glycolysis enhancement, which accelerates KCs death.

The generation of HFHC-induced MASLD mouse model.

Male wildtype C57B/6J mice were fed with HFHC diet for 0, 4 or 16 weeks. (A) H&E (top), Oil red O (middle) and Sirius red staining (bottom) were performed to detect MASLD progression in liver sections at 0, 4 or 16 weeks after HFHC diet, respectively. Scale bar: 20μm. Liver fibrosis was quantified. MASLD Activity Score is diagnosed. (B) Body weight was recorded during HFHC diet feeding. (C) Serum ALT, AST, cholesterol, triglyceride or liver cholesterol, triglyceride is measured at 16 weeks after HFHC diet. n=4 mice/group. Unpaired Student’s t-test (B, C); one-way ANOVA (A). P value as indicated.ns: not significant.

Examination of KCs death and MoMFs recruitment in HFHC mice.

Male wildtype C57B/6J mice were fed with HFHC diet for 0, 4 or 16 weeks. (A) KCs death was examined by immunofluorescence staining of Clec4f and cleavaged caspase 3 (Cl-Casp3) and DAPI (Nuclei) in livers of mice. KCs death was quantified. n=4 mice/group. (B) Flow cytometry analysis of MoMFs (CD45⁺ Ly6G⁻ CD11b⁺ F4/80lowTIM4low/-) among NPCs of WT mice fed HFHC deit for 0, 4, or 16 weeks. (C) MoMFs counts were quantified. n=4-5 mice/group. (D) Proliferation-associated TUNEL+ KCs was examined by co-staining of Ki67, TIM4, TUNEL and DAPI (Nuclei). Ki67+KCs among TUNEL+ KCs was quantified. Representative images are shown in A, D. One-way ANOVA (A, C, D). P value as indicated. Scale bar: 20μm (main panels), 5μm(inset).

Dynamic changes in mRNA expression levels of rate-limiting enzyme genes involved in glucose metabolism.

(A) Purity of isolated KCs was examined by immunofluorescence staining of TIM4 and DAPI (nuclei). Scale bar:20μm (main panels), 5μm(inset). Purity was quantified (n=6 independent experiments). (B) qRT-PCR analysis of mRNA expression levels of key rate-limiting enzymes in glycolysis, pentose phosphate pathway (PPP), glycogenolysis, glycogenesis, TCA cycle, and β-oxidation in WT KCs from mice for indicated dietary durations (n = 4 mice/group).One-way ANOVA (B), P values as indicated.

Excessive glucose metabolic activity contributes to Kupffer cell death.

(A) Calcein-AM was used to assess primary KCs cells viability treated as in main Figure 4. Scale bar: 20μm. Representative images are shown. (B-C) Isolated Kupffer cells were treated for 24 h with: Blank (no treatment), DMSO (vehicle control), 20 µM oligomycin (Oligo, ATP synthase inhibitor). Scale bars: 20 µm (main panels), 5 µm (insets). Cell death were analyzed as above. One-way ANOVA (B). P value as indicated.

Comparison of KCs polarization between WT and Chil1-/- mice during MASLD progression.

(A) qPCR analysis of Key genes involved in macrophage polarization pathway in liver tissues of WT and Chil1-/- mice fed with HFHC at indicated week. n=4-7 mice/group. Representative images are shown in A. One-way ANOVA (A). P value as indicated.

Recombinant Chi3l1(rChi3l1) inhibits glucose utilization in Chil1-/- BMDMs.

(A) Comparative principal component analysis (PCA) of metabolites in Chil1-/-and Chil1-/- supplemented with rChi3l1 BMDMs cultured with uniformly labeled [U-13C]glucose. (B) Heatmap depicting significantly altered Glycolysis and Pentose phosphate (PPP) metabolites in Chil1-/- and Chil1-/- supplemented with rChi3l1 BMDMs . (C) Glucose metabolic flux analysis in Chil1-/- and Chil1-/- supplemented with rChi3l1 BMDMs showing mass isotopologue distributions of: glycolysis pathway intermediates (Glc, F6P, FBP, 3PGA, 2PGA, PEP, PA, LA, G6P). Pentose phosphate pathway intermediates (Ru5P, R5P, S7P, DHAP). Data represent n= 6 biological replicates. (D) Lactate dehydrogenase (LDH) activity in culture medium of Chil1-/-BMDMs treated for 24 h with: 10 mM glucose (high glucose), 10 mM glucose + 100 ng/mL rChi3l1. Unpaired Student’s t-test (D). P value as indicated.

Chi3l1 is majorly expressed in hepatic macrophages and its expression is upreguated during MASLD.

(A) Cellular source of Chi3l1 was assessed by immunohistochemical analysis of consecutive liver sections from mice fed an HFHC diet for 16 weeks. Serial sections were independently stained for: Chi3l1, Lineage markers: HNF4α (hepatocytes), Desmin (hepatic stellate cells; HSCs), or Iba1 (Hepatic macrophages). Cellular localization was determined by aligning morphological profiles across sequential sections. Scale bars: 20 μm (main panels), 5 μm (insets). Hepatocytes are outlined in Chi3l1+ images in green dashed lines. (B) Chi3l1 expression in KCs were assessed by co-staining with F4/80 (red) and Timd4 (white) in livers of mice fed either normal chow diet (NCD, 0 weeks) or HFHC diet (16 weeks). Scale bar:20μm (main panels), 5μm (inset). The intensity of Chi3l1 expression in KCs was quantified. Unpaired Student’s t-test (B). P value as indicated.

Cre insertion promotes KCs self-proliferation.

(A) Comparsion of KCs self-proliferation in WT or Clec4f cre mice under NCD or Clec4f cre mice under HFHC diet feeding by costaining of Ki67 (proliferation marker) and TIM4 (KCs marker). Nuclei are counterstained with DAPI (blue). Scale bar: 20μm (main panels) and 10μm (Inset). (B) KCs proliferation was quantified. Representative images are shown in A. One-way ANOVA (B). P value as indicated.