Bariatric Surgery Associates with Nonalcoholic Steatohepatitis/Hepatocellular Carcinoma Amelioration via SPP1 Suppression

Figures

• 

  Figure 1. Identification of DEGs. DEGs were screened out from NASH_BS (A) and NASH_Diet (C) subgroups. The top 20 total significant upregulated and downregulated DEGs were mapped with gene profiles from NASH_BS (B) and NASH_Diet (D) subgroups.

• 

  Figure 2. Function enrichment. Hallmark enrichment analyses were conducted in the NASH_BS and NASH_Diet subgroups. (A) 20 significant hallmark terms were shown with enrichment scores (ES) in NASH_BS subgroup (FDR adjusted p < 0.05). (B) 9 top-ranked (but not significant) hallmark terms were shown in the NASH_Diet subgroup (FDR adjusted p > 0.05). KEGG and GO enrichment analyses were conducted in NASH_BS and NASH_Diet subgroups. Top significant KEGG and GO terms from upregulated and downregulated gene clusters were respectively displayed (C–F) (p < 0.05).

• 

  Figure 3. Cell type enrichment. Cell type enrichment levels of in total 7 parenchyma and 16 non-parenchyma cell types were displayed as NASH_BS (A) and NASH_Diet (B) subgroups. Immune scores were displayed as NASH_BS (C) and NASH_Diet (D) subgroups (*: p < 0.05). Pearson’s correlation analyses were carried out mutually among 23 cell populations in NASH_BS (E) and NASH_Diet (F) subgroups (*: p < 0.1, **: p < 0.01, ***: p < 0.001, ****: p < 0.0001).

• 

  Figure 4. Identification of cancer-associated key genes in NASH_BS livers. Enriched cancer modules were assessed and displayed as (A) NASH_BS and (B) NASH_Diet subgroups. (C) High macrophage-associated gene modules in the NASH_BS subgroup were selected according to WGCNA. Coef [p (scientific notation: e.g., 2e-05 = 2 × 10−5)] values were indicated in each block. (D) DEGs were screened out from GSE164760 (NASH vs. NASH-derived HCC). (E) Overlapping genes were generated from downregulated genes in the NASH_BS subgroup, upregulated genes in the NASH-HCC group, and macrophage associated WGCNA module. (F) The PPI network was generated with 54 overlapping genes from downregulated genes in the NASH_BS subgroup and the macrophage-associated WGCNA module. Hub gene clusters were colored light blue, purple, and green. And cancer associated genes were boarded in red. (G) Fold changes (FC) of gene (MNDA, ALOX5AP, PECAM1, SPP1, CD86, FGF21 and CSTA) expression were meta-analyzed in 5 datasets. FC values with statistical significance (p < 0.05) were highlighted. (H) Gene (MNDA, ALOX5AP, PECAM1, SPP1, CD86, FGF21 and CSTA) expression in healthy and tumor tissues were displayed based on TCGA database (*: p…

• 

  Figure 5. Annotation of SPP1 in LIHC/HCC. (A) SPP1 expression was compared in LIHC (TCGA) dataset. (B) The correlation of SPP1 expression and tumor stages was analyzed in LIHC (TCGA) dataset. (C) Correlation of SPP1 expression and survival time (OS, DSS, PFS, and DFS) was analyzed in LIHC (TCGA) dataset. p value is displayed in scientific notation (e.g., 1.6e-3 = 1.6 × 10−3). (D) Correlation of OS, SPP1 expression, and macrophage infiltration levels was analyzed in the LIHC (TCGA) dataset.

• 

  Figure 6. SPP1 expression in mouse models and patients. (A) Post-BS and -Diet on obese mouse models were performed. Relative gene expressions of Spp1 and 9 NASH/HCC biomarkers (Fas, Cd36, Ppara, Pparg, Fgf21, Cmyc, Ctnnb1, Ki67, and Tert) were displayed in obese (B), Post-BS, and -Diet mouse livers (C). (D) Correlation analyses between Spp1 and 9 NASH/HCC biomarkers (Fas, Cd36, Ppara, Pparg, Fgf21, Cmyc, Ctnnb1, Ki67, and Tert) were respectively conducted following the Pearson’s method. (E) SPP1 expressions in situ (normal tissue vs. tumor tissue) were generated from the HPA database.