Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A ) Normalized counts of Hnf4a in liver 8 h and 24 h after CLP (RNA-Seq data). n = 3/group. ( B , C ) Western blot analysis of HNF4α protein levels in liver 8 h and 24 h after CLP or sham. HNF4α protein expression (54 kDa) was normalized to actin levels (42 kDa) and expressed relative to sham. n = 5–6/group. ( D , E ) Immunofluorescent images of liver 6 h, 24 h and 48 h after CLP or sham, or from tamoxifen-injected Hnf4a flfl and Hnf4a Liver-i-KO mice, stained with HNF4α antibody (red) and DAPI (blue). Scale bar = 20 µm. Percentage HNF4α positive nuclei was quantified. n = 5–11/group. ( F – J ) HNF4α ChIP-Seq in liver 8 h after sham or CLP. n = 4/group. ( F ) Experimental setup. ( G ) PCA plot. ( H ) Signal intensity as fold over local region per position in a 400-bp region centered on the center of all called peaks. ( I ) Heatmap representing the region scores by deepTools 2.5 kbp centered on the center of all differential peaks (Padj < 0.05). P -values were calculated from DESeq2 (Wald test). Scores were derived from bam file coverage with deepTools. Higher scores indicate higher coverage. ( J ) HOMER transcription factor motif enrichment 200 bp centered on the center of all differential peaks. P -values derived from Fisher’s exact test (Hypergeometric test). Bars: mean ± SD, except for ( E ) mean ± SEM. Each dot represents a single biological replicate. P -values were analyzed with unpaired t-test ( A , E ) or one-way ANOVA ( C , E ). ns: non-significant. .

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: RNA Sequencing, Western Blot, Expressing, Injection, Staining, ChIP-sequencing, Derivative Assay

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A ) Relative abundance of Hnf4a transcripts (Hnf4a-201, Hnf4a-202, and Hnf4a-203) in CLP relative to sham, expressed as log2(TPM) percentages, derived from in-house paired-end RNA-Seq data 8 h post-CLP. n = 3/group (biological replicates). ( B , C ) OD450-OD570 represents the strength of HNF4α binding to dsDNA oligos immobilized on a plate using nuclear lysates from tamoxifen-injected Hnf4a flfl ( n = 4) and Hnf4a Liver-i-KO ( n = 3) mice ( B ) or sham ( n = 8) and CLP ( n = 8) mice 8 h post-CLP ( C ). Bars: mean ± SEM ( A ), central lines: mean ± SD ( B , C ). Each dot represents a single biological replicate. P -values were analyzed with two-way ANOVA ( A ) or unpaired t-test ( B , C ). ns: nonsignificant.

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: Derivative Assay, RNA Sequencing, Binding Assay, Injection

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A ) Experimental setup. Liver was isolated 8 h after sham or CLP for ATAC-Seq, and H3K27ac and H3K4me3 ChIP-Seq analyses. n = 4/group (biological replicates). ( B ) ATAC-Seq PCA plot. ( C ) Signal intensity as fold over local region per position in a 400-bp region centered on the center of all called ATAC-Seq peaks. ( D ) Heatmaps representing the region scores by deepTools centered on the center of all differential ATAC-Seq, and H3K4me3 and H3K27ac ChIP-Seq peaks (Padj < 0.05). P -values were calculated from DESeq2 (Wald test). Higher score indicates higher coverage. ( E ) Sham and CLP ATAC signals 0.5 kbp centered on the center of HNF4α ChIP-Seq peaks with decreased intensity (dn-blue) and increased intensity (up-green) in CLP. ( F ) Boxplots representing H3K27ac and H3K4me3 signals from sham and CLP or Hnf4a flfl and Hnf4a Liver-i-KO 1.5 kbp or 1.0 kbp, respectively, centered on the center of HNF4α ChIP-Seq peaks with decreased intensity (‘sham-specific’) ( n = 2858) and increased intensity (‘CLP-specific’) ( n = 669) in CLP. The boxplots display the distribution from Q1 (the first quartile, 25%) to Q3 (the third quartile, 75%), where the box itself spans the interquartile range (IQR). The line inside the box indicates the median. The whiskers extend from Q1 − 1.5 * IQR to Q3 + 1.5 * IQR. Outliers are defined as values that fall outside this range, calculated using the formulas: min(max(values), Q_3 + 1.5 * IQR) for the upper bound and max(min(values), Q_1 – 1.5 * IQR) for the lower bound. ( G ) Hnf4a flfl and Hnf4a Liver-i-KO ATAC signals 0.5 kbp centered on the center of sham-specific (SO-blue) and CLP-specific (CO-green) ATAC-Seq peaks. ( E – G ) P -values were analyzed with Wilcoxon test. ns: non-significant.

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: Isolation, ChIP-sequencing

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A – C ) Liver was isolated 8 h after sham or CLP for ATAC-Seq, and H3K4me3 and H3K27ac ChIP-Seq analyses. n = 4/group (biological replicates). ( A ) H3K4me3 and H3K27ac ChIP-Seq PCA plot. ( B ) HOMER transcription factor motif enrichment 200 bp centered on the center of all differential peaks. P -values derived from Fisher’s exact test (Hypergeometric test). Differential peaks were identified by direct region overlap and DESeq2 (using Wald test), with the FDR set at 5%. ( C ) Heatmaps representing H3K27ac and H3K4me3 signals from sham and CLP or Hnf4a flfl and Hnf4a Liver-i-KO 1.5 kbp or 1.0 kbp, respectively, centered on the center of differential HNF4α ChIP-Seq peaks with decreased intensity (‘sham-specific’) and increased intensity (‘CLP-specific’) in CLP. Differential peaks were identified by DESeq2 (using Wald test), with the FDR set at 5%.

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: Isolation, ChIP-sequencing, Derivative Assay

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A ) Overlap between RNA-Seq, HNF4α ChIP-Seq and ATAC-Seq-associated pathway analysis for genes with Padj < 0.05 & LFC < 0 in liver 8 h after sham or CLP. P -values were calculated from DESeq2 (Wald test). RNA-Seq: n = 3/group; ChIP-Seq & ATAC-Seq: n = 4/group (biological replicates). ( B ) Western blot analysis of hepatic HNF4α protein levels (54 kDa) relative to actin (42 kDa) in Hnf4a Liver-i-KO and Hnf4a flfl mice 3 days after tamoxifen or vehicle treatment. n = 12 (4 female, 8 male). ( C ) Hnf4a Liver-i-KO and Hnf4a flfl mice were subjected to CLP 3 days after the last tamoxifen injection and mortality was monitored over 10 days. Data were analyzed with a Log-rank test. n = 24 (11 female, 13 male). ( D ) Hnf4a Liver-i-KO and Hnf4a flfl mice were i.p . injected with tamoxifen for 5 consecutive days. Three days later, sham or CLP was performed, and liver was isolated for RNA-Seq analysis 8 h later. n = 4/group (9 female, 7 male) (biological replicates). ( E ) Metascape pathway analysis of genes downregulated or upregulated (Padj < 0.05) in Hnf4a Liver-i-KO sham relative to Hnf4a flfl , and downregulated (Padj < 0.05, LFC < −0.8) ( n = 716) or upregulated (Padj < 0.05, LFC > 0.8) ( n = 245), respectively, in CLP relative to sham. ( F ) Scatterplot representing LFC of genes with Padj < 0.05 in Hnf4a Liver-i-KO sham relative to Hnf4a flfl sham, and in Hnf4a flfl CLP relative to Hnf4a flfl sham. Key genes of interest are indicated. ( G ) Metascape pathway analysis of genes further downregulated (Padj < 0.05, LFC < −0.8) ( n = 288) or upregulated (Padj < 0.05, LFC > 0.8) ( n = 133) in Hnf4a Liver-i-KO CLP, relative to Hnf4a flfl CLP. ( H ) Normalized counts of representative genes from (G), with the expression level of Hnf4a flfl sham set as 100%. n = 4/group. Bars: mean ± SD. Each dot represents a single biological replicate. ( E , G ): P -values derived from Fisher’s exact test (Hypergeometric test). ( F , H ): P -values were calculated from DESeq2 (Wald test). .

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: RNA Sequencing, ChIP-sequencing, Western Blot, Injection, Isolation, Expressing, Derivative Assay

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: Enrichr pathway analysis (MSIGDB Hallmark 2020) of genes significantly downregulated or upregulated (Padj < 0.05) in Hnf4a Liver-i-KO sham relative to Hnf4a flfl (= HNF4α-dependent genes) and downregulated (Padj < 0.05, LFC < −0.8) or upregulated (Padj < 0.05, LFC > 0.8), respectively, in CLP relative to sham. P -values derived from Fisher’s exact test (Hypergeometric test). Differential genes were identified by DESeq2 (using Wald test). The pathways we are mainly interested in, as evidenced by literature, are highlighted in bold.

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: Derivative Assay

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A – C ) Analysis of livers from tamoxifen-injected Hnf4a Liver-i-KO and Hnf4a flfl mice 8 h after sham or CLP. ( A ) Heatmap of differentially expressed nuclear receptor genes from RNA-Seq (unit scale bar: normalized counts). Rows are centered by unit variance scaling and clustered using correlation distance and average linkage. n = 4/group (9 female, 7 male) (biological replicates). ( B ) RT-qPCR Ppara mRNA expression relative to Gapdh and Hprt . ( C ) Western blot analysis of PPARα protein levels (52 kDa) relative to actin (42 kDa). n = 5/group. ( D ) HNF4α ChIP-qPCR on Ppara promoter in liver 8 h after sham or CLP. n = 4/group. ( E ) PPARα agonist GW7647 or vehicle were injected i.p . in Hnf4a Liver-i-KO and Hnf4a flfl mice 3 days after tamoxifen treatment. Livers were isolated for qPCR 4 h later. n = 5–6/group (2 female, 20 male). ( F ) RT-qPCR mRNA expression of Ppara and several PPARα target genes relative to Rpl and Hprt . ( G ) Hnf4a Liver-i-KO and Hnf4a flfl mice were i.p . injected with tamoxifen on 5 consecutive days. Three days later, sham or CLP was performed, and liver and blood were isolated 8 h and 24 h later. ( H ) Plasma FFA levels 8 h after sham or CLP. n = 31 (12 female, 19 male). ( I ) Volume of lipid droplets (represented by voxel counts), relative to liver tissue volume, was calculated for each z-stack and averaged over all z-stacks per section 8 h after sham or CLP. n = 31 (13 female, 18 male). ( J , K ) Plasma IL6, aspartate aminotransferase (AST) and lactate dehydrogenase (LDH) levels 24 h after sham or CLP. n = 17–19 (7 female, 10–12 male). Bars: mean ± SD. Each dot represents a single biological replicate. P -values were analyzed with two-way ANOVA ( B , C , F , H – K ) or unpaired t-test ( D ). ns: non-significant. .

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: Injection, RNA Sequencing, Quantitative RT-PCR, Expressing, Western Blot, ChIP-qPCR, Isolation, Clinical Proteomics

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A ) Structural formula of N-trans-caffeoyltyramine (NCT). ( B ) Mice were injected with NCT (200 mg/kg) or DMSO i.p . twice daily for 6 days and subjected to CLP on day 7. Mortality was monitored over 12 days, during which NCT was injected. Data were analyzed with a Log-rank test. n = 11–12/group. ( C ) Mice were injected with NCT (200 mg/kg) or vehicle i.p . 3 h and 24 h post-CLP. Mortality was monitored over 10 days. Data were analyzed with a Gehan-Breslow-Wilcoxon test. n = 17–18/group. ( D – N ) Mice were injected with NCT (200 mg/kg) or DMSO i.p . twice daily for 6 days, subjected to sham or CLP on day 7, and liver and blood were isolated 24 h later. n = 3–8/group. ( D ) Experimental setup. ( E ) Body temperature. ( F ) Systemic and hepatic bacterial load (log CFU/mL). ( G ) HNF4α ChIP-qPCR relative to mouse IgG, on Hes6 , Ppara , Apoa2 and Ugt2b1 promoters in liver. ( H ) RT-qPCR Ppara mRNA expression in liver relative to Hprt and Rpl . ( I ) Plasma FFA levels. ( J ) Lipid droplets volume (voxel counts) relative to liver tissue volume was calculated for each z-stack and averaged over all z-stacks per section. ( K ) RT-qPCR Apcs mRNA expression in liver relative to Hprt and Rpl . ( L – N ) Plasma serum amyloid P (SAP), IL6, aspartate aminotransferase (AST) and lactate dehydrogenase (LDH) levels. Bars: mean ± SD, except for ( E , F ) mean ± SEM. Each dot represents a single biological replicate. P -values were analyzed with two-way ANOVA. ns: non-significant. .

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: Injection, Isolation, ChIP-qPCR, Quantitative RT-PCR, Expressing, Clinical Proteomics

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: ( A ) Experimental setup porcine sepsis model. By intraperitoneal installation of 3 g/kg of autologous feces, the animals were permitted to develop sepsis until reaching a state of severe hypotension characterized by a mean arterial pressure (MAP) of ≤50 mmHg. Once severe hypotension, defined as a MAP ranging between 45 and 50 mmHg, was achieved, it was maintained for a duration of one hour. Around 18 h after sepsis initiation, liver samples were collected for RNA-Seq analysis ( n = 3 sham, n = 9 sepsis; biological replicates). ( B ) Enrichr analysis of the genes downregulated (Padj < 0.05, LFC < 0) in pig sepsis. Differential genes were identified by DESeq2 (using Wald test). The upstream regulators are ordered by their enrichment P -value (derived from Fisher’s exact test (Hypergeometric test)). ( C ) HOMER transcription factor motif enrichment 1000 bp upstream of TSS from the genes downregulated in pig sepsis. Differential genes were identified by DESeq2 (using Wald test). P -values were derived from Fisher’s exact test (Hypergeometric test)). ( D ) At 2 weeks, Alb-uPA +/+ -SCID mice were injected with human hepatocytes into their spleens, enabling migration to the liver and repopulation of empty niches. After several weeks, mice underwent sham or CLP surgery, and livers were isolated 24 h later. ( E ) Heatmaps displaying the expression levels of HNF4α-dependent genes (genes differentially expressed in Hnf4a Liver-i-KO sham relative to Hnf4a flfl sham, with Padj < 0.05) in mouse liver 24 h post-CLP, pig sepsis liver and humanized mice liver 24 h post-CLP. We selected HNF4α target genes known to be downregulated (Padj < 0.05) in mouse liver 24 h post-CLP, focussing on the ones mentioned in the paper. Differential genes were identified by DESeq2 (using Wald test). Expression levels were quantified by RT-qPCR relative to Hprt and Rpl , and log transformed. For the humanized mice, human-specific and mouse-specific primers were used to investigate the response in human and mouse hepatocytes, respectively.

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: RNA Sequencing, Derivative Assay, Injection, Migration, Isolation, Expressing, Quantitative RT-PCR, Transformation Assay

Journal: EMBO Molecular Medicine

Article Title: A critical role for HNF4α in polymicrobial sepsis-associated metabolic reprogramming and death

doi: 10.1038/s44321-024-00130-1

Figure Lengend Snippet: Reagents and tools table

Article Snippet: The membrane was blocked with a 1⁄2 dilution of Starting Block/PBST0,1% (Thermo Fisher Scientific) and incubated overnight at 4 °C with primary antibodies against HNF4α (1:1000; PP-H1415-0C, R&D Systems), PPARα (1:1000; ab126285, Abcam), STAT3 (1:2000; 4904, Cell Signaling Technology), phospho-STAT3 (1:2000; 9145, Cell Signaling Technology) and β-actin (1:5000; BA3R, Thermo Fisher Scientific), as an internal control.

Techniques: Recombinant, Control, Sequencing, Software, cDNA Synthesis, Acid Assay, Enzyme-linked Immunosorbent Assay, Purification