Hepatic macrophages express Chi3l1 and upregulate its expression post HFHC diet.

(A) Immunofluorescent staining of TIM4 (white), F4/80 (red), Chi3l1 (green), and nuclear DAPI (blue) in liver sections of mice fed with either NCD or HFHC for 16 weeks, illustrating Chi3l1 expression in hepatic macrophages. Scale bar=20 µm and 10 µm (Inset). Chi3l1+ F4/80+ cells/F4/80+ cells were statistically analyzed. n=4 mice/group. (B) Representative immunofluorescence images of liver sections from WT and Chil1-/- mice stained for Chi3l1(green), F4/80 (macrophages), and TIM4 (Kupffer cells). DAPI (blue) marks nuclei. Scale bar=20 µm and 10 µm (Insets). (C, D) Western blot analysis of Chi3l1 in either isolated Kupffer cells (KCs, F) or whole liver tissue (Liver, G) from mice fed either NCD or HFHC diet. n=2-3 mice/group. (E) mRNA expression levels of Chil1 in liver tissues of patients with metabolic dysfunction-associated fatty liver (MAFL) or with metabolic dysfunction-associated steatohepatitis (MASH) (GEO Datasets: GSE167523, GSE207310, GSE130970). No-MAFLD or Healthy individuals serve as controls. (F) The correlation between mRNA expression levels of Chil1 and MASLD activity score or fibrosis stage was analyzed (GEO Datasets: GSE130970). Representative images were shown in A, B. Mann-Whitney test was performed in E. Pearson’s correlation was performed in F. P value and r value are as indicated.

Deficiency of Chi3l1 in Kupffer cells promotes insulin resistance and hepatic lipid accumulation.

Chil1fl/fl and Clec4fΔChil1 mice were fed either a normal chow diet (NCD) or a high-fat, high-cholesterol (HFHC) diet for 16 weeks. (A, B) Body weight was recorded during HFHC diet feeding (A) and expressed as a percentage of initial body mass (B). (C) H&E (Upper panel) and oil red o staining (Lower panel) was performed to examine liver histology and hepatic lipid accumulation in both genotypes after 16 weeks of NCD or HFHC diet. Scale bar = 20 µm. (D) Liver index (liver weight/body weight × 100%), ALT levels, and serum and liver Cholesterol or Triglyceride levels were measured in both genotypes after 16 weeks on NCD or HFHC diets. n=4-12 mice/group. (E, F) Intraperitoneal glucose tolerance test (IGTT) and insulin tolerance test (ITT) were performed after 16 weeks of NCD or HFHC feeding in both genotypes (n = 4–12 mice per group). Representative images were shown in (C). One-way ANOVA was performed in (A, B, D-F). P-value is as indicated.

Deficiency of Chi3l1 in Kupffer cells promotes liver steatosis and fibrosis in MASH.

Male wildtype C57B/6J mice were fed with NCD or MCD diet for 6 weeks (A-B). Chil1fl/fl and Clec4fΔChil1 mice were fed with a MCD diet for 6 weeks (C-E). (A, B) qRT-PCR (A) and wetern blot (B) analysis of Chi3l1 expression in whole liver tissues under NCD and MCD diets. n=3 mice/group. (C) Body weight of mice with conditional deletion of Chil1 in KCs (Clec4fΔChil1) and their control mice (Chil1fl/fl) was recorded during MCD diet. (D) Histological analyses were performed in liver tissue of Clec4fΔChil1 and Chil1fl/flfed the MCD diet for 6 weeks. Scale bar=20μm. (E) Liver index (liver weight/body weight × 100%), ALT levels and serum and liver Cholesterol or Triglyceride levels were measured in both genotypes fed the MCD diet for 6 weeks. n=4-6 mice/group. Representative images are shown in D. Two-tailed, unpaired student t-test was performed in A, C, D, E. P value is as indicated.

ScRNA-seq reveals upregulated glucose metabolism-related transcripts in KCs, correlating with cell death signatures.

Wildtype C57BL/6J mice were fed either a normal chow diet (NCD) or HFHC for 16 weeks. NPCs were isolated and subjected to BD Rhapsody scRNA sequencing. (A) Uniform manifold approximation and projection (UMAP) plots illustrate the clustering of NPCs in the livers of mice fed NCD and HFHC. Cell clusters are color-coded, with monocytes/macrophages clusters outlined. (B) UMAP plots depict the clustering of Monocytes/Macrophages in the livers of mice fed NCD and HFHC. Cell clusters are color-coded. (C) Dot plot displays the scaled gene expression levels of lineage-specific marker genes in different cell clusters. (D) Quantification of each cell cluster is presented. (E) KEGG analysis reveals the top 12 enriched pathways for up-regulated genes when comparing HFHC versus NCD in KCs, monocytes, and MoMFs, respectively. (F) Gene set variation analysis (GSVA) shows pathway activity for cell death, glucose metabolism, and cell proliferation in KCs, monocytes, and MoMFs of WT mice fed NCD or HFHC for 16 weeks, respectively. (G) The correlation between cell death and glucose metabolism pathways, based on GSVA score, is depicted.

Chi3l1 deficiency promote KCs death during MASLD.

(A) GSVA analysis showed the enrichment of cell death-related pathways in KCs from WT mice fed with either NCD or HFHC or Chil1-/- mice fed with HFHC. (B) Dot plot showing the scaled gene expression levels of Apoptosis-related genes and repressor genes in KCs from either WT or Chil1-/-fed with HFHC. (C) Strategy used to gate KCs (CD45+ F4/80hi CD11blow TIM4hi) and MoMFs (CD45+ F4/80low CD11bhi Ly6G- TIM4-) in the liver by flow cytometry. (D) Number of KCs and MoMFs /liver or gram(g) liver were statistically analyzed. n= 3-4 mice per group. (E) Immunofluorescent staining to detect TIM4(green), TUNEL (red), and nuclear DAPI (blue) in liver sections. Scale bar=50µm and 20µm (Insets). TUNEL+ TIM4+ cells/TIM4+ cells were statistically analyzed. n=4 mice/group. Representative images are shown in C, E. One-way ANOVA was performed in D. Two-tailed, unpaired student t-test was performed in E. P value is as indicated.

Molecular interaction between Chi3l1 and glucose.

(A) A comparison of chemical structures between glucose and chitin. (B) Prediction of Chi3l1-glucose interaction using STITCH database (http://stitch.embl.de). (C) Strategy for pulling down glucose-binding proteins in murine serum. (D) Biotin-conjugated glucose was incubated with murine serum from mice fed with HFHC for 16 weeks. Proteins bound to glucose were precipitated by streptavidin beads. Biotin or biotin-conjugated glucose plus glucose were used as negative controls. Western blot was performed to examine Chi3l1 in the precipitate. (E) Microscale thermophoresis assay to detect the interaction between recombinant mouse Chi3l1 (rChi3l1) and glucose. Kd=4.95±0.66mM. (F) Western blot to detect Chi3l1 expression in murine serum before and after HFHC feeding. n=3 mice/group.

Chi3l1 limits glucose uptake and protects hepatic macrophages from cell death.

(A) Following 12 h of glucose starvation, isolated KCs or BMDM were divided into two groups: one treated with no 2-NBDG and the other with 2-NBDG. Within each group, KCs or BMDM were further treated without or with recombinant murine Chi3l1 (rChi3l1) for 6 h. Glycogen aggregate formation labeled by 2-NBDG (Green) in KCs or BMDM was examined after counterstaining with nuclear DAPI (Blue). Scale bar=2μm. Area of 2-NBDG in KCs were quantified. (B) Following 12 h of glucose starvation, BMDM were treated with either no glucose or high glucose (25mM). Concurrently, BMDM were treated without or with rChi3l1 for 24 h under each condition. glycogen aggregate formation in BMDM was detected using immunofluorescence staining for Stbd1 (red) and nuclear DAPI (blue). Scale bar = 10 μm. (C and D) BMDM cells were treated without or with rChi3l1 for 24 h and subjected to Seahorse metabolic analysis to measure the extracellular acidification rate (ECAR). (E and F) KCs were treated without (blank) or with either Isopropyl alcohol (Iso) or 800uM palmitic acid (PA) or 100ng rChi3l1 with 800 uM PA for 24 h. Western blot was performed to detect cleaved caspase 3 (Cl-Casp3) in E. Calcein/PI staining was quantified to detect cell viability in F. Scale bar=50μm. (G) Measurement of 2-NBDG (a fluorescent glucose analog) uptake by KCs in vivo. WT and Chil1-/-mice, either untreated or supplemented with rChi3l1, were injected intraperitoneally with 12 mg/kg 2-NBDG. After 45mins, KCs were isolated and glucose uptake assessed by spectrophotometry. (H) Representative immunofluorescence images of liver sections stained for TIM4 (red) and 2-NBDG uptake (green) to visualize glucose uptake by KCs in situ. Scale bar = 10 µm (Insets). Quantification is shown as the percentage of TIM4⁺ cells that are also 2-NBDG⁺. Representative images were shown in A, B, H. One-way ANOVA was performed in A, F, G, H. Two-tailed, unpaired student t-test was performed in D. P value is as indicated.

Differential regulation of KCs and MoMFs fate by Chi3l1-glucose interaction.

KCs maintain a high-glucose activation state, while MoMFs exhibit a relatively low-glucose metabolic program. Chi3l1-glucose binding inhibits glucose uptake in KCs, thereby delaying KCs death and alleviating MASLD progression and metabolic dysfunction. In contrast, although Chi3l1-glucose binding similarly inhibits glucose uptake in MoMFs, their low basal glucose metabolism renders them resistant to this metabolic perturbation, resulting in minimal impact on MASLD pathogenesis.

MASLD progression in the HFHC diet-induced mouse model.

(A) Representative liver sections from wild-type C57BL/6J mice fed either a normal chow diet (NCD) or a high-fat, high-cholesterol (HFHC) diet for 16 weeks. H&E and Sirius Red staining were used to assess lipid deposition, inflammation and fibrosis. Scale bar: 20 µm. Quantification of Sirius Red–positive area is shown. (B) Western blot analysis of α-SMA expression in whole liver lysates from NCD-and HFHC-fed mice (n = 3 mice/group) to evaluate activation of hepatic stellate cells.

Generation and validation of Chil1-/- mice.

(A) The construction, genotyping strategy and genotyping results of Chil1-/- mice. P: positive control; WT: Wild-type; Neg: Blank control(ddH2O). (B) qRT-PCR analysis of mRNA expression levels of Chil1 in liver tissues of WT and Chil1-/- mice fed with HFHC for 0, 8 and 16 weeks. n=3-4 mice/group. Two-tailed, unpaired student t-test was performed in B. P value is as indicated.

The construction and genotype of Clec4fΔChil1 mice

(A) The construction, genotyping strategy and genotyping results of Clec4fΔChil1 mice. P: positive control; WT: Wild-type; Neg: Blank control (ddH2O). (B) qRT-PCR analysis of mRNA expression levels of Chil1 in KCs (CD45+ F4/80hi CD11blow TIM4hi) or MoMFs (CD45+ F4/80low CD11bhi Ly6G- TIM4-) FACS sorted from Chil1fl/fl and Clec4fΔChil1 mice at 0 and 4 weeks post HFHC diet. n=3 mice/group. (C) Western blot to detect Chi3l1 expression in isolated KCs of Chil1fl/fl and Clec4fΔChil1 mice. n=2 mice/group. (D) The expression specificity of Clec4f was examined in various tissues in Clec4f cre-Rosa tdtomato mice, which is generated by crossing Clec4f-cre with Rosa-tdtomato mice.

Deficiency of Chi3l1 in Kupffer cells promotes insulin resistance and hepatic lipid accumulation.

Clec4f cre and Clec4fΔChil1 mice were fed with a HFHC diet for 16 weeks. (A, B) Body weight was recorded during HFHC diet feeding (A) and expressed as a percentage of initial body mass (B). (C, D) H&E (C) and oil red o staining (D) was performed to examine liver histology and hepatic lipid accumulation in in both genotypes after 16 weeks of HFHC diet. Scale bar = 20 µm. (E) Liver index (liver weight/body weight × 100%), ALT levels and serum and liver Cholesterol or Triglyceride levels were measured in both genotypes after 16 weeks of HFHC diet. n=3-6 mice/group. (F&G) Intraperitoneal glucose tolerance test (IGTT) and insulin tolerance test (ITT) were performed after 16 weeks of HFHC feeding in both genotypes.n=3-6 mice/group. Representative images were shown in C, D. Two-tailed, unpaired student t-test was performed in A,B,E-G. P value is as indicated.

The construction and genotype of Lyz2ΔChil1 mice.

(A) The construction, genotyping strategy and genotyping results of Lyz2ΔChil 1 mice. pos: positive control; WT: Wild-type; Neg: Blank control(ddH2O). (B) qRT-PCR analysis of mRNA expression levels of Chil1 in KCs (CD45+ F4/80hi CD11blow TIM4hi) or MoMFs (CD45+ F4/80low CD11bhi Ly6G- TIM4-) FACS sorted from Chil1fl/fl and Lyz2ΔChil1 mice at 0 and 4 weeks post HFHC diet. n= 3 mice/group. (C) Western blotting analysis of protein levels of Chi3l1 in BMDM and primary KCs of Chil1fl/ f l and Lyz2ΔChil1 mice. n=2-3 mice/group.

Deficiency of Chi3l1 in MoMFs barely affect insulin resistance and hepatic lipid accumulation.

Chil1fl/fl and Lyz2ΔChil1 mice were fed either a normal chow diet (NCD) or a high-fat, high-cholesterol (HFHC) diet for 16 weeks. (A, B) Body weight was recorded during HFHC diet feeding (A) and expressed as a percentage of initial body mass (B). (C) H&E (Upper panel) and oil red o staining (Lower panel) was performed to examine liver histology and hepatic lipid accumulation in both genotypes after 16 weeks of NCD or HFHC diet. Scale bar = 20 µm. (D) Liver index (liver weight/body weight × 100%), ALT levels, and serum and liver Cholesterol or Triglyceride levels were measured in both genotypes after 16 weeks on NCD or HFHC diets. n=4-9 mice/group. (E, F) Intraperitoneal glucose tolerance test (IGTT) and insulin tolerance test (ITT) were performed after 16 weeks of NCD or HFHC feeding in both genotypes (n = 4–9 mice per group). Representative images were shown in (C). One-way ANOVA was performed in (A, B, D-F). P-values are as indicated.

Gene expression levels of lineage-specific marker genes in monocytes/macrophages clusters.

(A) Scaled gene expression levels of each lineage-specific marker gene are shown in UMAP plots of monocytes/macrophages clusters. Colors indicate gene expression levels.

Chi3l1 deficiency promote KCs death during MASLD

WT and Chil1-/- mice were fed with a HFHC diet for 0, 8, 16 weeks. (A) Flow cytometry analysis of KCs (CD45+ F4/80hi CD11blow TIM4hi) and MoMFs (CD45+ F4/80low CD11bhi Ly6G- TIM4-) among NPCs between WT and Chil1-/- mice. (B) Immunofluorescent staining to detect TIM4(red), Cleaved caspase3(green), and nuclear DAPI (blue) in liver sections. Scale bar=20μm and 5μm(Insets). Cleaved caspase3+ TIM4+ cells/ TIM4+ cells were statistically analyzed. n=4-6 mice/group. Representative images are shown in A, B. Student t-test was performed in B. P value is as indicated.

Deficiency of Chi3l1 in KCs but not MoMFs promote KCs death during MASLD.

Chil1fl/fl and Clec4fΔChil1 mice were fed with a MCD diet for 6 weeks. Chil1fl/fl and Lyz2ΔChil1 mice were fed with a HFHC diet for 20 weeks. (A) Immunofluorescent staining to detect TIM4(green), TUNEL(red), and nuclear DAPI (blue) in liver sections of Chil1fl/fl and Clec4fΔChil1 mice. Scale bar=50µm and 20µm (Insets). TUNEL+ TIM4+ cells/TIM4+ cells were statistically analyzed. n=4-6 mice/group. (B) Immunofluorescent staining to detect TIM4(green), TUNEL(red), and nuclear DAPI (blue) in liver sections of Chil1fl/fl and Lyz2ΔChil1 mice. Scale bar=50µm and 20µm (Insets). TUNEL+ TIM4+ cells/TIM4+ cells were statistically analyzed. n=4-5 mice/group. (C) Flow cytometry analysis of KCs (CD45+ F4/80hi CD11blow TIM4hi) and MoMFs (CD45+ F4/80low CD11bhi Ly6G- TIM4-) among NPCs between Chil1fl/fl and Lyz2ΔChil1 mice. (D) Number of KCs or MoMFs/gram(g) liver were statistically analyzed. n= 3 mice/group. Representative images are shown in A-C. Student t-test was performed in A-B. One-way ANOVA was performed in D. P value is as indicated.