An extended transcription factor regulatory network controls hepatocyte identity

Breakdown in liver functions in advanced stages of liver injuries and in cancer results from hepatocyte loss of identity stemming from compromised expression of the hepatic core transcriptional regulatory circuitry transcription factors.

“breakdown in liver functions in advanced stages of liver injuries and in cancer results from hepatocyte loss of identity stemming from compromised expression of the hepatic CoRC TFs (Dubois et al, 2020a; Berasain et al, 2022).”

• cites Dubois et al, 2020a sourcePartly supported

  The Abstract and Discussion link sustained liver dysfunction after injury to loss of hepatic molecular identity and reduced LIVER-ID transcription-factor activity. The Results examine acute injury in mice and dysfunctional livers from patients with sepsis; they do not establish the same explanation for cancer or advanced stages of liver injury. (LLM, conf=0.87)

  Experimental — accuracy not yet measured
• cites Berasain et al, 2022 sourcePartly supported

  The Key points and Conclusions link loss of liver function during chronic injury to impaired core liver transcription factors and loss of hepatocyte identity. The ‘Loss of hepatic function in liver disease’ section says persistent impairment creates conditions favourable to hepatocellular carcinoma, but does not establish that this mechanism causes liver-function breakdown in cancer itself. The indexed passages do not provide individual page numbers. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Hep-IDCONNECT transcription factor-encoding genes are linked to abnormal liver phenotypes and metabolic functions, albeit less significantly than Hep-ID transcription factors.

“mining genes associated with phenotypes from the Mammalian Phenotype Ontology resource (Smith et al, 2005; Chen et al, 2007) revealed that, albeit less significant when compared to Hep‐ID TFs, Hep‐IDCONNECT TF‐encoding genes are linked to abnormal liver phenotypes and metabolic functions (Fig 2G).”

• cites Smith et al, 2005 sourceSupported
• cites Chen et al, 2007 sourceSupported

Cell-specific transcription factors directly control the expression of non-transcription-factor genes that exert activities/functions characterizing individual cell types (identity effector genes).

“cell‐specific TFs directly control the expression of non‐TF genes exerting activities/functions characterizing individual cell‐types, alternatively known as identity effector genes (Arendt et al, 2016; Almeida et al, 2021).”

• cites Arendt et al, 2016 sourceNot checked — no full text of the source
• cites Almeida et al, 2021 sourceSupported

The moderate transcriptional changes most probably relate to the intrinsic role of identity transcription factor networks, which are built up to allow robustness of individual cell-type transcriptional programs.

“This most probably relates to the intrinsic role of identity TF networks, which are built‐up to allow robustness of individual cell‐type transcriptional programs (Kyrmizi et al, 2006; Almeida et al, 2021).”

• cites Kyrmizi et al, 2006 sourcePartly supported

  In “The complexity of the hepatic cross-regulatory network correlates with the stability of expression of the individual factors” (page unavailable), the paper proposes that a complex network can reduce responses to perturbations and help maintain the hepatic phenotype. That supports the proposed robustness mechanism, but the paper studies hepatocytes, not identity transcription-factor networks across individual cell types generally, and does not establish that mechanism as the explanation for the citing paper’s moderate transcriptional changes. (LLM, conf=0.82)

  Experimental — accuracy not yet measured
• cites Almeida et al, 2021 sourceSupported

Several Hep-IDCONNECT transcription factors are central to establishment of rhythmic hepatocyte gene expression, including both circadian (BHLHE40, CLOCK, DBP, NFIL3, NR1D1/2) and ultradian (XBP1) rhythms.

“Several Hep‐IDCONNECT TFs are central to establishment of rhythmic hepatocyte gene expression (Appendix Fig S11) including both circadian (BHLHE40, CLOCK, DBP, NFIL3, NR1D1/2; Mukherji et al, 2019) and ultradian (XBP1; Meng et al, 2020; Pan et al, 2020) rhythms.”

• cites Mukherji et al, 2019 sourcePartly supported

  The Introduction sections on the mammalian circadian-clock system and hepatic functions describe CLOCK, DBP, E4BP4, and REV-ERBα/β in circadian regulation of gene expression, including in the liver. Our copy does not identify BHLHE40 or explicitly connect the name NFIL3 to E4BP4, so it does not substantiate the entire cited list. The XBP1 ultradian claim has separate citations in the printed sentence and is not attributed to Mukherji et al. (LLM, conf=0.84)

  Experimental — accuracy not yet measured
• cites Meng et al, 2020 sourceSupported
• cites Pan et al, 2020 sourceSupported

Besides Hep-ID transcription factors and an additional set of cell-specific transcription factors with well-characterized hepatic functions (Nr1i3/Car, Rorc/Rorg, Hnf1a, and Gata4), the majority of transcription factors from cluster G belonged to cell-type enriched or ubiquitous transcription factors.

“This further highlighted that, besides Hep‐ID TFs and an additional set of cell‐specific TFs with well characterized hepatic functions (Nr1i3/Car, Rorc/Rorg, Hnf1a and Gata4; Tachmatzidi et al, 2021), the majority of TFs from cluster G belonged to CTE or UBQ TFs (Fig 1J and Dataset EV1).”

• cites Tachmatzidi et al, 2021 sourcePartly supported

  The GATA Family section describes GATA4’s role in liver-bud expansion, and the HNF1 section states that HNF1α regulates genes critical for liver function. The cited paper does not establish hepatic functions for Nr1i3/Car or Rorc/Rorg, so it supports only part of the cited list. The cluster-G classification is presented as the citing authors’ own result, not as a finding attributed to this source. (LLM, conf=0.87)

  Experimental — accuracy not yet measured

Many additional transcription factors beyond the hepatic identity core transcriptional regulatory circuitry transcription factors have been ascribed roles in the control of hepatocyte activities.

“many additional TFs beyond these hepatic identity CoRC TFs have been ascribed with roles in the control of hepatocyte activities (Bideyan et al, 2021).”

• cites Bideyan et al, 2021 sourceSupported

Mining transcriptomic data from livers of mice with hepatocyte-specific deletion of individual Hep-IDCONNECT transcription factor-encoding genes indicated a preferential dysregulation of downstream hepatocyte identity effector genes.

“Mining transcriptomic data obtained from the livers of mice with hepatocyte‐specific deletion of individual Hep‐IDCONNECT TF‐encoding genes (hep−/− mice) indicated that this is functionally underlain by a preferential dysregulation of downstream hepatocyte identity effector genes, that is, non‐TF genes with liver‐specific broad H3K4me3 domains defined in Dubois et al (2020a) (Fig 2H).”

• cites Dubois et al, 2020a sourceSupported

A majority of Hep-ID transcription factor genes (9 out of 13), including Mlxipl, were characterized by XBP-1-dependent ultradian expression patterns.

“mining data from Meng et al (2020) revealed that a majority of Hep‐ID TF genes (9 out of 13), including for example Mlxipl, were characterized by XBP‐1‐dependent ultradian expression patterns (Fig 3C and D, and Table EV2).”

• cites Meng et al, 2020 sourcePartly supported

  The Results sections on the hepatic 12-hour transcriptome and XBP1s cistrome report XBP1-dependent oscillating genes in mouse liver, supporting the general basis for the citing authors’ data-mining analysis. Our copy does not identify the Hep-ID transcription factor set, report that 9 of 13 have these patterns, or name Mlxipl as an example; those specifics cannot be checked against it. (LLM, conf=0.83)

  Experimental — accuracy not yet measured

Hep-IDCONNECT transcription factors showed reduced expression levels in microdissected hepatocytes from alcohol-related human liver cirrhosis.

“Similar observations were made using transcriptomic data obtained from microdissected hepatocytes from alcohol‐related human liver cirrhosis (Bou Saleh et al, 2021; Fig 2E).”

• cites Bou Saleh et al, 2021 sourcePartly supported

  The Results section on microdissected tissue (Fig. 3A–D; page unavailable) reports transcriptomic changes in cirrhotic hepatocytes compared with controls, so the cited dataset is relevant. It identifies a significant loss of hepatic-identity gene expression in alcoholic hepatitis, but does not establish reduced expression of Hep-IDCONNECT transcription factors specifically in hepatocytes from alcohol-related cirrhosis. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Hep-IDCONNECT TFs play a role in specifying proper hepatic gene expression in both space and time.

“pointing to a role of Hep‐IDCONNECT TFs in specifying proper hepatic gene expression in both space and time.”

• cites González‐Blas et al, 2022 sourcePartly supported

  The section headed “A core enhancer-driven gene regulatory network for hepatocytes is modulated by zonated repressor TFs” (p. 4) identifies hepatocyte TFs and TFs associated with spatial zonation, supporting a role in spatially patterned hepatic gene regulation. Our copy does not identify a group called “Hep-IDCONNECT TFs” or establish a role for those TFs in the timing of hepatic gene expression. (LLM, conf=0.84)

  Experimental — accuracy not yet measured

Inflammatory cytokines such as tumor necrosis factor or IL1B are common drivers of hepatocyte dedifferentiation and dysfunction.

“inflammatory cytokines such as tumor necrosis factor or IL1B are well‐established common drivers of hepatocyte dedifferentiation and dysfunction (Del Campo et al, 2018; Hyun et al, 2020).”

• cites Del Campo et al, 2018 sourcePartly supported

  The section “Release of cytokines by immune cells” says cytokines including TNF-α and IL-1 can cause hepatocyte apoptosis, steatosis, and inflammation, supporting a role in hepatocyte injury and dysfunction. It does not say they drive hepatocyte dedifferentiation; the phenotypic transdifferentiation described in “Activation of HSCs and hepatic fibrosis” concerns hepatic stellate cells instead. Page numbers are unavailable in our copy. (LLM, conf=0.86)

  Experimental — accuracy not yet measured
• cites Hyun et al, 2020 sourcePartly supported

  The paper reports that TNF-α and IL-1β make hepatocytes fetal-like by suppressing the ESRP2-mediated adult splicing program (April 2020 issue, Figure 5 discussion). It also reports impaired ammonia detoxification in cytokine-reprogrammed hepatocytes (Figure 6 discussion). These findings support dedifferentiation and dysfunction, but the paper does not establish that the cytokines are well-established, common drivers of those outcomes across settings. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

T3 treatment does not prevent the loss of Hep-ID TF gene expression that occurs spontaneously when primary mouse hepatocytes are cultured in vitro.

“We could not however observe any protective role for T3 treatment with regards to the loss of Hep‐ID TF gene expression spontaneously occurring when primary mouse hepatocytes (MPH) are cultured in vitro (Ploton et al, 2018; Seirup et al, 2022; Appendix Fig S14).”

• cites Ploton et al, 2018 sourceCould not verify

  The Mice strains and experimentation and Glucose production sections describe primary mouse hepatocyte isolation and culture, but do not report a spontaneous loss of hepatocyte-identity transcription-factor expression. Searches for T3, thyroid hormone, and culture-related loss of transcription-factor expression found no passage assessing whether T3 prevents that loss. The citing sentence presents the T3 observation as the citing authors’ own result, but our copy also does not establish the background loss attributed to Ploton et al. (LLM, conf=0.72)

  Experimental — accuracy not yet measured
• cites Seirup et al, 2022 sourcePartly supported

  The Results (section 2.2) report that mature hepatocyte marker genes decline during culture, and the Methods (section 4.2) identify the hepatocytes as mouse-derived. The paper does not report T3 treatment or establish that T3 fails to prevent loss of Hep-ID transcription-factor gene expression; that treatment result is presented as the citing authors’ own observation. (LLM, conf=0.86)

  Experimental — accuracy not yet measured

TBX3 and TCF7L2 have been identified as regulators of zonated hepatocyte transcriptional programs.

“Other Hep‐IDCONNECT TFs, TBX3 and TCF7L2, have been identified as regulators of zonated hepatocyte transcriptional programs (preprint: González‐Blas et al, 2022)”

• cites González‐Blas et al, 2022 sourceSupported

THRB activation increased the ratio of HNF4A adult promoter P1- to embryonic promoter P2-derived isoforms.

“In addition, THRB activation also increased the ratio of HNF4A adult promoter P1‐ to embryonic promoter P2‐derived isoforms (Fig 4H), a phenomenon known to promote hepatocyte differentiation and function (Dubois et al, 2020b).”

• cites Dubois et al, 2020b sourcePartly supported

  Section 2.1 reports a switch from P2- to P1-derived HNF4A isoforms during hepatocyte differentiation, supporting the association invoked by the citation. It does not establish that increasing the P1:P2 ratio itself promotes hepatocyte differentiation and function. The THRB-activation result is attributed to the citing paper’s Fig. 4H, not to Dubois et al. (LLM, conf=0.82)

  Experimental — accuracy not yet measured

Overexpression or activation of several Hep-IDCONNECT transcription factors, GATA6, KLF9, TBX3, NR1H3, and THRB, has been shown to suppress hepatocarcinogenesis.

“overexpression or activation of several Hep‐IDCONNECT TFs, GATA‐binding protein 6 (GATA6), krüppel‐like factor 9 (KLF9), T‐box transcription factor 3 (TBX3) and the nuclear receptors liver X receptor alpha (NR1H3 also known as LXR alpha) and thyroid hormone receptor beta (THRB), has been shown to suppress hepatocarcinogenesis (Fig 4C; Sun et al, 2014; Tan et al, 2019; Kowalik et al, 2020; Lin et a”

• cites Sun et al, 2014 sourceNot checked — no full text of the source
• cites Tan et al, 2019 sourceNot checked — no full text of the source
• cites Kowalik et al, 2020 sourcePartly supported

  The Discussion (page unavailable) supports an antitumor role for the THRB/KLF9 axis: T3 impaired HCC-cell growth, and KLF9 silencing partially reversed that effect. The cited paper does not establish suppression of hepatocarcinogenesis by GATA6, TBX3, or NR1H3, which are also included in the citing sentence. (LLM, conf=0.87)

  Experimental — accuracy not yet measured

In a rat model of hepatocarcinogenesis, T3 was able to counteract the cancer-driven loss of expression of Hep-ID transcription factors.

“interrogating the transcriptomic data obtained in a rat model of hepatocarcinogenesis (Kowalik et al, 2020) revealed that T3 was able to counteract the cancer‐driven loss of expression of Hep‐ID TFs (Fig 4J).”

• cites Kowalik et al, 2020 sourceSupported

Hepatocyte partial loss of identity is a hallmark of diseased livers.

“we and others have reported that hepatocyte partial loss of identity is a hallmark of diseased livers [recently reviewed in (Berasain et al, 2022)].”

• cites Berasain et al, 2022 sourceSupported

Hepatic TF cistromes can be faithfully inferred from mouse liver ChIP-seq data.

“hepatic TF cistromes can be faithfully inferred from mouse liver ChIP‐seq data (Kyrmizi et al, 2006; Schmidt et al, 2010; Dubois‐Chevalier et al, 2017a; Sommars et al, 2019; Dubois et al, 2020a).”

• cites Kyrmizi et al, 2006 sourcePartly supported

  The section on recruitment of hepatic transcription factors reports using ChIP in mouse liver to map factor occupancy at selected regulatory regions. The methods section says those regions were assessed by real-time PCR, not sequencing. The paper therefore supports mapping some hepatic TF binding in mouse liver, but not the claim that whole cistromes can be faithfully inferred from mouse liver ChIP-seq data. (LLM, conf=0.88)

  Experimental — accuracy not yet measured
• cites Schmidt et al, 2010 sourceSupported
• cites Dubois‐Chevalier et al, 2017a sourcePartly supported

  The Results section on NR1H4-bound CRMs describes using mouse-liver ChIP-seq data to define the NR1H4 cistrome and compare it with 47 other transcriptional regulators. The Methods section on ChIP-seq processing describes replicate and false-positive controls. These passages support deriving hepatic regulator cistromes from mouse-liver ChIP-seq, but do not establish that the resulting cistromes can be inferred *faithfully* in general. (LLM, conf=0.81)

  Experimental — accuracy not yet measured

Many Hep-IDCONNECT TFs, including NR1H3 and THRB, do not display typically used features to define hepatocyte identity TFs such as hepatocyte-specific expression and association with super-enhancers.

“Many Hep‐IDCONNECT TFs, including NR1H3 and THRB, do not display typically used features to define hepatocyte identity TFs such as hepatocyte‐specific expression and association with super‐enhancers (Hnisz et al, 2013; Dubois et al, 2020a; Table EV3).”

• cites Hnisz et al, 2013 sourceSupported
• cites Dubois et al, 2020a sourcePartly supported

  The Results section on the LIVER-ID TF network describes high, liver-specific expression as one way to define identity TFs, while the super-enhancer Results section links super-enhancers to liver-identity genes. Neither section establishes the claim’s central point: that many Hep-IDCONNECT TFs, specifically NR1H3 and THRB, lack those features. Searches for Hep-IDCONNECT and both TF names found no passage addressing that comparison; page numbers were unavailable in our copy. (LLM, conf=0.76)

  Experimental — accuracy not yet measured

Hep-IDCONNECT transcription factors control identity effector genes.

“Besides controlling identity effector genes, Hep‐IDCONNECT TFs also engage in reciprocal transcriptional regulation with TFs of the CoRC.”

Hep-IDCONNECT transcription factors engage in reciprocal transcriptional regulation with transcription factors of the core transcriptional regulatory circuitry.

“Besides controlling identity effector genes, Hep‐IDCONNECT TFs also engage in reciprocal transcriptional regulation with TFs of the CoRC.”

The transcriptional regulatory network controlling hepatocyte identity has a more complex organization than a core transcriptional regulatory circuitry limited to a small set of interconnected cell-specific transcription factors.

“By mining global hepatic TF regulons, we reveal a more complex organization of the transcriptional regulatory network controlling hepatocyte identity.”

Tight functional interconnections controlling hepatocyte identity extend to non-cell-specific transcription factors beyond the core transcriptional regulatory circuitry (Hep-IDCONNECT TFs).

“We show that tight functional interconnections controlling hepatocyte identity extend to non‐cell‐specific TFs beyond the CoRC, which we call hepatocyte identity (Hep‐ID)CONNECT TFs.”

In homeostatic basal conditions, Hep-IDCONNECT transcription factors are involved in fine tuning core transcriptional regulatory circuitry transcription factor expression, including their rhythmic expression patterns.

“In homeostatic basal conditions, this translates into Hep‐IDCONNECT TFs being involved in fine tuning CoRC TF expression including their rhythmic expression patterns.”

Activation of NR1H3 or THRB in hepatocarcinoma or in hepatocytes subjected to inflammation-induced loss of identity resets core transcriptional regulatory circuitry transcription factor expression.

“This is observed upon activation of NR1H3 or THRB in hepatocarcinoma or in hepatocytes subjected to inflammation‐induced loss of identity.”

TF-encoding genes captured by cluster G were not limited to those displaying the strongest and most liver-specific expression.

“Despite being enriched, TF‐encoding genes captured by cluster G were not limited to those displaying the strongest and most liver‐specific expression (Fig 1G and H).”

H3K4me3 labeling at Hep-IDCONNECT transcription factor genes displayed intermediate tissue-specificity when compared to Hep-ID transcription factor genes and remaining transcription factor-encoding genes from cluster G.

“Moreover, H3K4me3 labelling at Hep‐IDCONNECT TF genes displayed intermediate tissue‐specificity when compared to Hep‐ID TF genes, on the one hand, and remaining TF‐encoding genes from cluster G, on the other hand (Appendix Fig S7).”

Hep-IDCONNECT transcription factors showed reduced expression levels in the livers of mouse liver injury models triggering hepatocyte dedifferentiation.

“we found that Hep‐IDCONNECT TFs showed reduced expression levels in the livers of these mouse models (Fig 2D).”

Expression of Hep-IDCONNECT transcription factors was decreased in transcriptomic data from mouse primary hepatocytes of adult mice with hepatocyte-specific deletion of the Hep-ID transcription factor Hnf4a.

“In line, expression of Hep‐IDCONNECT TFs was decreased in transcriptomic data obtained from MPH of adult mice with hepatocyte‐specific deletion of the Hep‐ID TF Hnf4a (Fig 2C).”

Similar to Hep-ID transcription factors and different from CTCF, binding of most Hep-IDCONNECT transcription factors was stronger at hepatocyte identity effector gene promoters.

“In line, similar to Hep‐ID TFs and different from CTCF, binding of most Hep‐IDCONNECT TFs was stronger at hepatocyte identity effector gene promoters (Fig 2I).”

The breadth of H3K4me3 labeling at the promoters of Hep-IDCONNECT transcription factor genes was intermediate between that observed for Hep-ID transcription factor genes and other transcription factor-encoding genes.

“the breadth of H3K4me3 labeling at the promoters of Hep‐IDCONNECT TF genes was also comprised in between that observed for the Hep‐ID TF and other TF‐encoding genes (Fig 2F).”

A subset of cell-type enriched/ubiquitous transcription factors from cluster G, called Hep-IDCONNECT transcription factors, is characterized by privileged hepatocyte expression (mouse primary hepatocytes ranked among the top 10 expressing cells and expression in mouse primary hepatocytes above average expression in other cells).

“we found that a subset of the CTE/UBQ TFs from cluster G, hereafter called Hep‐IDCONNECT TFs, was nevertheless characterized by privileged hepatocyte expression (i.e., MPH ranked among the top 10 expressing cells and expression in MPH above average expression in other cells; Figs 2A and EV3).”

Hep-IDCONNECT transcription factors show greater recruitment to Hep-ID transcription factor gene promoters in mouse liver compared to control activity-matched transcription factor-encoding gene promoters.

“Mining Hep‐IDCONNECT TFs binding in mouse liver indicated greater recruitment to Hep‐ID TF gene promoters when compared to control activity‐matched TF‐encoding gene promoters, reminiscent of data obtained for Hep‐ID TFs (Fig 3A and Appendix Fig S10A–C).”

Hierarchical clustering of TF-encoding gene promoters based on TF-binding pattern similarity identified seven main clusters (A-G).

“Hierarchical clustering was next performed and identified seven main clusters of TF‐encoding gene promoters (Fig 1C and Appendix Fig S2A).”

Single-nuclei RNA-seq data from adult mouse and human livers confirmed privileged expression of Hep-IDCONNECT transcription factors in hepatocytes.

“mining single‐nuclei RNA‐seq data from adult mouse and human livers confirmed privileged expression in hepatocytes (Appendix Fig S5A and B).”

Hep-ID transcription factor binding was most pronounced at promoters from cluster G, both individually and collectively (Hep-ID TF co-recruitment).

“Moreover, Hep‐ID TF binding was also most pronounced at promoters from cluster G, both when considered individually (Appendix Figs S2D and E, and S3G) or collectively, i.e., Hep‐ID TF co‐recruitment (Fig 1F).”

Cluster G was enriched for Hep-ID transcription factor-encoding genes (odds ratio = 5.2, P = 0.006).

“In line, cluster G was enriched for Hep‐ID TF‐encoding genes (odds ratio = 5.2, P = 0.006).”

Hep-IDCONNECT transcription factor binding to Hep-ID transcription factor gene promoters only translated into moderate transcriptional changes in expression of Hep-ID transcription factor-encoding genes.

“However, this only translated into moderate transcriptional changes in expression of Hep‐ID TF‐encoding genes.”

Promoters from cluster G were overall the most strongly active as revealed by chromatin accessibility (DHS), hepatic histone acetylation (H3K27ac), or mRNA expression levels of associated TF-encoding genes.

“Promoters from cluster G were also overall the most strongly active as revealed by chromatin accessibility (DHS), hepatic histone acetylation (H3K27ac) or mRNA expression levels of associated TF‐encoding genes (Fig 1E and Appendix Fig S4A–C).”

Mouse liver cistromes of eight Hep-ID transcription factors show extensive co-recruitment at Hep-ID transcription factor-encoding gene promoters compared to control non-Hep-ID transcription factor-encoding gene promoters.

“monitoring the mouse liver cistromes of eight Hep‐ID TFs (CEBPA, FOXA2, HNF4A, NR1H4, NR5A2, ONECUT1, PPARA, PROX1; Dataset EV2) pointed to their extensive co‐recruitment at the Hep‐ID TF‐encoding gene promoters when compared to a control set of non‐Hep‐ID TF‐encoding gene promoters (Figs 1A and EV1A and B).”

Hep-IDCONNECT TFs are not required to sustain Hep-ID TF gene expression and cell identity in basal condition.

“While not being required to sustain Hep‐ID TF gene expression and cell identity in basal condition, our data obtained with NR1H3 and THRB in dedifferentiated hepatocytes revealed that Hep‐IDCONNECT TF activation can be leveraged to reset the CoRC and hepatocyte identity.”

Treatment with GW3965, T3, or GC-1 induced a global increase in Hep-ID transcription factor gene expression in HCC cell-lines.

“we found that treatment with GW3965, T3, or GC‐1 induced a global increase in Hep‐ID TF gene expression in HCC cell‐lines (Fig 4E).”

In precancerous rat liver nodules, Hep-ID TF gene expression is modulated compared to control rat livers, differing from matched control genes.

“Modulation of Hep‐ID TF gene expression in precancerous nodules compared to control rat livers and in liver nodules of rats treated with T3 compared to nodules of non‐treated rats.”

Hep-ID TFs of the CoRC are tightly connected with an additional layer of TFs through reciprocal co-recruitment to their promoters.

“we identified that Hep‐ID TFs of the CoRC are tightly connected with an additional layer of TFs through reciprocal co‐recruitment to their promoters.”

The ability of GW3965, T3, or GC-1 to induce Hep-ID transcription factor gene expression was linked to their ability to interfere with HCC cell-line growth in vitro.

“This was linked to the ability of these ligands to interfere with HCC cell‐line growth in vitro (Appendix Fig S12D).”

Activation of Hep-IDCONNECT TFs NR1H3 and THRB can reset the CoRC and hepatocyte identity in dedifferentiated hepatocytes.

“our data obtained with NR1H3 and THRB in dedifferentiated hepatocytes revealed that Hep‐IDCONNECT TF activation can be leveraged to reset the CoRC and hepatocyte identity.”

Mining transcriptomic data of human hepatocarcinoma from TCGA (n = 367) indicated a positive correlation between expression of Hep-IDCONNECT and Hep-ID transcription factor-encoding genes.

“Mining transcriptomic data of human hepatocarcinoma (HCC) from the cancer genome atlas (TCGA; n = 367) indicated a positive correlation between expression of Hep‐IDCONNECT and Hep‐ID TF‐encoding genes (Fig 4A).”

Similar data were obtained using a non-cancerous human immortalized hepatic cell-line (IHH) used as another model of partially dedifferentiated hepatocytes.

“Similar data were obtained using a non‐cancerous human immortalized hepatic cell‐line (IHH) used as another model of partially dedifferentiated hepatocytes (Schippers et al, 1997; Fig EV4A–C).”

• cites Schippers et al, 1997 sourceNot checked — no full text of the source

Binding of Hep-IDCONNECT transcription factors to Hep-ID transcription factor genes was also observed in hepatic cancer cells.

“binding of Hep‐IDCONNECT TFs to Hep‐ID TF genes was also observed in hepatic cancer cells (Fig 4D).”

Survival analyses indicated better overall survival of patients with HCC expressing high levels of KLF9 and THRB.

“Survival analyses indicated better overall survival of patients with HCC expressing high levels of KLF9 and THRB (Appendix Fig S12C).”

Western blot assays confirmed T3-induced changes in Hep-ID transcription factor levels.

“Western blot assays confirmed T3‐induced changes in Hep‐ID TF levels (Fig 4I).”

Hep-IDCONNECT TFs comprise regulators of rhythmic gene expression including XBP1.

“Hep‐IDCONNECT TFs comprise regulators of rhythmic gene expression including XBP1.”

In the IL1B + T3 high responsiveness group, Hep-ID TF gene expression recovered to levels higher than in the IL1B only and/or IL1B + T3 low responsiveness groups.

“the IL1B + T3 high responsiveness group showed expression recovery of Hep‐ID TF genes whose levels were higher in this group when compared to those in the IL1B only and/or IL1B + T3 low responsiveness groups (Fig 5D).”

• AssumptionRests on an assumption: Stimulation of Dio1 and Hectd3 expression is a valid readout of T3 responsiveness in mouse liver.

Induction of Hep-ID transcription factor gene expression by T3 and GC-1 in HepG2 cells was independently reproduced using RT-qPCR analyses, with induction of DIO1 used as a positive control.

“Induction of Hep‐ID TF gene expression by T3 and GC‐1 in HepG2 cells was independently reproduced as judged using RT‐qPCR analyses, induction of DIO1 being used as a positive control in these assays (Fig 4F and G).”

Extensive co-recruitment of Hep-ID transcription factors is not limited to their own highly hepatocyte-specific genes within the core transcriptional regulatory circuitry but extends to an additional set of transcription factors.

“Altogether, these analyses indicated that extensive co‐recruitment of Hep‐ID TFs is not limited to their own highly hepatocyte‐specific genes within the CoRC but extends to an additional set of TFs.”

These data point to the potential importance of Hep-IDCONNECT transcription factors in fully differentiated hepatocytes.

“Altogether, these data point to the potential importance of Hep‐IDCONNECT TFs in fully differentiated hepatocytes.”

Hepatocyte identity is controlled by an extended array of transcription factors beyond the core transcriptional regulatory circuitry.

“Our study establishes that hepatocyte identity is controlled by an extended array of TFs beyond the CoRC.”

Many Hep-IDCONNECT transcription factors have individually been ascribed a role in the control of hepatic metabolic activities.

“many Hep‐IDCONNECT TFs have individually, that is, without an identified unifying rationale, been ascribed, at least to some extent, a role in the control of hepatic metabolic activities (Appendix Fig S8).”

Hep-IDCONNECT transcription factors might be directly dependent upon Hep-ID transcription factors for their enhanced hepatocyte expression.

“These observations suggested that Hep‐IDCONNECT TFs might be directly dependent upon Hep‐ID TFs for their enhanced hepatocyte expression.”

Cluster G is the prominent subset of promoters capturing hepatic core transcriptional regulatory circuitry auto-/cross-binding.

“Therefore, cluster G is the prominent subset of promoters capturing hepatic CoRC auto−/cross‐binding.”

T3-mediated effects towards promoting a mature hepatocyte cell-state are not limited to cancer cells.

“defining that T3‐mediated effects towards promoting a mature hepatocyte cell‐state are not limited to cancer cells.”

• cites Ma et al, 2022Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

TF binding to cell-specific target genes translates into a more critical role in (re)establishment and dynamic regulation rather than maintenance of steady-state gene expression levels.

“These findings are consistent with other recent observations where TF binding to cell‐specific target genes translates into a more critical role in (re)establishment and dynamic regulation rather than maintenance of steady‐state gene expression levels (Lo et al, 2022).”

• cites Lo et al, 2022Not matched

  Couldn’t be matched to a paper in Crossref, so it can’t be checked.

The tight connection between Hep-IDCONNECT TFs and the CoRC provides an additional layer of regulation to finely tune Hep-ID TF expression.

“the tight connection between Hep‐IDCONNECT TFs and the CoRC provides an additional layer of regulation to finely tune Hep‐ID TF expression.”

Our data provide an explanation for the reported beneficial effects exerted by NR1H3 and THRB in HCC.

“With regard to HCC, our data provide explanation for the reported beneficial effects exerted by NR1H3 and THRB.”

Hep-IDCONNECT TFs are involved in identity stabilization loops whose functional importance for cell identity is revealed in pathophysiological situations of dedifferentiation.

“we propose that Hep‐IDCONNECT TFs are involved in identity stabilization loops, whose functional importance with regards to cell identity is revealed in pathophysiological situations of dedifferentiation (Fig 6).”

THRB is an actionable target to reset expression of hepatic identity TF genes in partially dedifferentiated hepatocytes.

“Altogether, these data have identified the Hep‐IDCONNECT TF THRB as an actionable target to reset expression of hepatic identity TF genes in partially dedifferentiated hepatocytes.”

No major difference in immune cell infiltration was present between the groups.

“We also ruled out any major difference in immune cell infiltration as a potential confounder in these analyses (Fig EV5D).”

Our study uncovered an unforeseen role for an extended transcriptional regulatory network beyond the CoRC in the control of hepatocyte identity.

“our study has allowed to uncover an unforeseen role for an extended transcriptional regulatory network beyond the CoRC in the control of hepatocyte identity (Fig 6).”