RISING STARS: Liver sinusoidal endothelial transcription factors in metabolic homeostasis and disease

in Journal of Molecular Endocrinology
Authors:
Dorka Nagy Department of Metabolism, Digestion and Reproduction, Section of Genetics and Genomics, Imperial College London, London, UK
National Heart and Lung Institute, Imperial College London, London, UK

Search for other papers by Dorka Nagy in
Current site
Google Scholar
PubMed Close
,
Hannah Maude Department of Metabolism, Digestion and Reproduction, Section of Genetics and Genomics, Imperial College London, London, UK

Search for other papers by Hannah Maude in
Current site
Google Scholar
PubMed Close
,
Graeme M Birdsey National Heart and Lung Institute, Imperial College London, London, UK

Search for other papers by Graeme M Birdsey in
Current site
Google Scholar
PubMed Close
,
Anna M Randi National Heart and Lung Institute, Imperial College London, London, UK

Search for other papers by Anna M Randi in
Current site
Google Scholar
PubMed Close
, and
Inês Cebola Department of Metabolism, Digestion and Reproduction, Section of Genetics and Genomics, Imperial College London, London, UK

Search for other papers by Inês Cebola in
Current site
Google Scholar
PubMed Close
https://orcid.org/0000-0003-3528-4982

Correspondence should be addressed to I Cebola: i.dos-santos-cebola@imperial.ac.uk

This paper is part of a collection of articles highlighting the breadth and depth of research being undertaken across the field of basic endocrinology by early- and mid-career researchers. The collection is published across the Journal of Endocrinology and the Journal of Molecular Endocrinology.

DOI:
https://doi.org/10.1530/JME-23-0026
Volume/Issue:
Volume 71: Issue 2
Article Type:
Review Article
Article ID:
e230026
Online Publication Date:
18 Jul 2023
Copyright:
© the author(s) 2023
Open access

Sign up for journal news

Liver sinusoidal endothelial cells (LSECs) are highly specialised endothelial cells that form the liver microvasculature. LSECs maintain liver homeostasis, scavenging bloodborne molecules, regulating immune response, and actively promoting hepatic stellate cell quiescence. These diverse functions are underpinned by a suite of unique phenotypical attributes distinct from other blood vessels. In recent years, studies have begun to reveal the specific contributions of LSECs to liver metabolic homeostasis and how LSEC dysfunction associates with disease aetiology. This has been particularly evident in the context of non-alcoholic fatty liver disease (NAFLD), the hepatic manifestation of metabolic syndrome, which is associated with the loss of key LSEC phenotypical characteristics and molecular identity. Comparative transcriptome studies of LSECs and other endothelial cells, together with rodent knockout models, have revealed that loss of LSEC identity through disruption of core transcription factor activity leads to impaired metabolic homeostasis and to hallmarks of liver disease. This review explores the current knowledge of LSEC transcription factors, covering their roles in LSEC development and maintenance of key phenotypic features, which, when disturbed, lead to loss of liver metabolic homeostasis and promote features of chronic liver diseases, such as non-alcoholic liver disease.

Abstract

Liver sinusoidal endothelial cells (LSECs) are highly specialised endothelial cells that form the liver microvasculature. LSECs maintain liver homeostasis, scavenging bloodborne molecules, regulating immune response, and actively promoting hepatic stellate cell quiescence. These diverse functions are underpinned by a suite of unique phenotypical attributes distinct from other blood vessels. In recent years, studies have begun to reveal the specific contributions of LSECs to liver metabolic homeostasis and how LSEC dysfunction associates with disease aetiology. This has been particularly evident in the context of non-alcoholic fatty liver disease (NAFLD), the hepatic manifestation of metabolic syndrome, which is associated with the loss of key LSEC phenotypical characteristics and molecular identity. Comparative transcriptome studies of LSECs and other endothelial cells, together with rodent knockout models, have revealed that loss of LSEC identity through disruption of core transcription factor activity leads to impaired metabolic homeostasis and to hallmarks of liver disease. This review explores the current knowledge of LSEC transcription factors, covering their roles in LSEC development and maintenance of key phenotypic features, which, when disturbed, lead to loss of liver metabolic homeostasis and promote features of chronic liver diseases, such as non-alcoholic liver disease.

Keywords: liver; endothelial cell; transcription factors; NAFLD; gene transcription

Invited author’s profile

Inês Cebola is a Wellcome Trust Sir Henry Dale Fellow in the Department of Metabolism, Digestion, and Reproduction, at Imperial College London, where she leads the Regulatory Genomics of Metabolic Disease Lab. She made major contributions to the diabetes genetics field during her postdoc with Prof Jorge Ferrer, having a long-standing interest in the role of noncoding DNA in disease. Her lab is composed of computational and wet lab biologists who seek to unravel novel genetic mechanisms of liver disease. Inês also holds an honorary clinical scientist appointment with London North West University Healthcare NHS Trust, leading a cross-disciplinary initiative that brings together academic researchers and clinical geneticists to accelerate genetic diagnosis for patients with rare diseases. Inês was the recipient of the 2019 SfE BES Early Career Prize and of prestigious funding schemes, including the Academy of Medical Sciences Springboard Fellowship (2020) and the Gilead Liver Research Scholars Award (2021).

Introduction

Liver sinusoidal endothelial cells (LSECs) represent a unique and highly specialised endothelial cell (EC) population that forms the liver microvasculature, known as the hepatic sinusoids. LSECs are the most abundant non-parenchymal cell type in the liver, representing 15–20% of all liver cells, but only ~3% of its volume (Blouin et al. 1977, Poisson et al. 2017). In the past two decades, LSECs have emerged as central players in the onset and progression of liver disease, having important implications for metabolic and endocrine health. LSEC dedifferentiation is associated with metabolic conditions such as insulin resistance and non-alcoholic fatty liver disease (NAFLD), a progressive disease that affects ~30% of the world’s population (Younossi et al. 2023) (Box 1).

Box 1 NAFLD and its in vivo models

NAFLD is an umbrella term used to describe a disease spectrum characterised by excessive fat accumulation in the liver (>5% liver mass), called steatosis, in the absence of other secondary causes (e.g. excessive alcohol consumption). NAFLD ranges from simple hepatic steatosis or non-alcoholic fatty liver (NAFL), which is generally benign with a low risk of adverse outcomes, to non-alcoholic steatohepatitis (NASH), hepatic fibrosis, and cirrhosis (Singh et al. 2015). NASH is characterised by steatosis, lobular inflammation, and hepatocellular ballooning and is accompanied by metabolic and endocrine dysfunction (Benedict & Zhang 2017). The prognosis worsens with the development of fibrosis, namely the deposition of scar tissue, which can progress through different stages of severity culminating in permanent scarring or cirrhosis. End-stage disease states include cirrhosis, liver failure, and hepatocellular carcinoma, which are leading causes of liver transplantation (Bataller & Brenner 2005, Younossi et al. 2023).

The liver is a central organ in the metabolism of glucose, free fatty acids, and hormones, and thus plays a major role in metabolic homeostasis. As a result, NAFLD is strongly associated with other metabolic disorders including extrahepatic manifestations of the metabolic syndrome such as insulin resistance and diabetes mellitus; up to 75% of type 2 diabetes patients have NAFLD (Friedman et al. 2018). NAFLD is also tightly associated with cardiovascular disease, which is the most common cause of death in patients with NAFLD (Ekstedt et al. 2006). In light of the strong association between NAFLD, obesity, and other cardiometabolic traits, it was recently proposed to rename NAFLD as metabolic (dysfunction)-associated fatty liver disease (Eslam et al. 2020). It must be noted that NAFLD is highly heterogeneous not only in severity but also in disease trajectory and comorbidity profile. Moreover, the degree of interaction between metabolic risk factors and NAFL/NASH is influenced by patient ancestry, suggesting the role of genetic risk factors in mediating these interactions (Younossi et al. 2018).

The complex natural history of NAFLD and the extensive timeframe required in humans to progress from simple steatosis to NASH has limited the ability of rodent models to fully recapitulate human NASH and cirrhosis. This is illustrated by the nearly 4000 unique animal models of NAFLD reported in the literature (Im et al. 2021). Many of these models involve dietary interventions, which lead to obesity, steatosis, and insulin resistance, but most models fail to show liver injury similar to that observed in NASH patients. This suggests fundamental differences in human and rodent liver biology, which has been recently supported by studies in a human liver chimeric mouse model of NAFLD (Bissig-Choisat et al. 2021) and by the limited number of common genes that are modulated in both mouse and human chronic liver disease (Holland et al. 2022). To induce and accelerate liver damage, models often include a combination of high-fat diet and low doses of toxic agents such as carbon tetrachloride (CCl4) (Tsuchida et al. 2018). However, as these toxins are predominantly metabolised by pericentral hepatocytes (those proximal to the central vein, see Fig. 1), they cause pericentral liver damage, leading to the development of fibrosis in the pericentral area with portal areas being secondary. In contrast, in humans, fibrosis is more frequently distributed in periportal and lobular areas (Delire et al. 2015) (see Box 2 for details on liver zonation).

Box 2 Liver zonation

The liver is structurally organised into lobules: hexagonal units in which oxygen and nutrient-rich blood enter via the portal triad, flow through the sinusoids, and exit via the central vein. As blood passes through the sinusoid, oxygen and nutrients are exported into the LSECs and liver parenchyma, while metabolism products and other cell exports enter the blood (Fig. 1). This structure creates a gradient of oxygen, nutrients, and signalling molecules including Wnt morphogens, a phenomenon called zonation (for detailed reviews see (Ben-Moshe & Itzkovitz 2019, Paris & Henderson 2022)). Importantly, zonation manifests as gradual changes at the gene expression level of LSECs and hepatocytes across the periportal–pericentral axis. Although this gradient is continuous, it is standard practice to delineate specific zones, most often referred to as zone 1 (periportal), zone 2 (midzonal), and zone 3 (pericentral) (Fig. 1). Each zone shows metabolic specialities, for example, hepatocytes in the oxygen-rich periportal zone carry out high-energy processes such as gluconeogenesis, lipid oxidation, and cholesterol synthesis, while the hypoxic pericentral zone carries out glycolysis, lipogenesis, glutamine synthesis, and xenobiotic metabolism (Ben-Moshe & Itzkovitz 2019). It was recently estimated that 67% of LSEC-expressed genes exhibit significant zonation (Inverso et al. 2021), including genes involved in peptide hormone and xenobiotic metabolism, response to gut-derived toxins, canonical Wnt signalling, as well as the binding and uptake of ligands by scavenger receptors (Paris & Henderson 2022). Notably, pericentral LSECs secrete Wnt morphogens, which play an important role in zonation (Preziosi et al. 2018). Zonation may be disrupted in liver disease (reviewed by (Steinman et al. 2021)) and evidence suggests that the zonal distribution of steatosis is relevant for clinical prognosis. For example, patients with NAFLD typically present with pericentral fibrosis which progresses outwards. Pericentral fibrosis independently associates with progressive steatosis and steatohepatitis, while patients with pan-acinar steatosis (i.e. involving all zones equally) are more likely to present with hepatocyte ballooning and advanced fibrosis (Chalasani et al. 2008).

Figure 1
Figure 1

Position, phenotype, and marker fingerprint of LSECs in homeostasis and disease. (A) The liver’s structural unit is the liver lobule where the portal triad and the central vein are connected by sinusoids. (B) Liver sinusoids are the scene for the uptake of nutrients derived from the intestines via the portal vein, where there is also an influx of pathogens and signalling molecules from the gut bacteria (top right). Oxygen-enriched blood enters from the portal artery and mixes with the portal vein blood. The mixed blood flows through the sinusoid where it is filtered by the LSECs and macromolecules and nutrients are transported towards the hepatocytes, while waste products from the liver parenchyma are transported into the sinusoidal blood. This mechanism creates a gradient of oxygen, nutrients, and signalling molecules along the periportal–pericentral axis (Box 2). (C) Spatial molecular imaging of the human liver showing zonated expression of the indicated genes (left). Data were acquired using CosMxTM Spatial Molecular Imager and includes staining for a panel of morphological features including hepatocytes, which were stained with CK8 and CK18 fluorescent antibodies, and quantifications for a panel of 1000 genes. Integrative data analysis enabled the detection of six ‘neighbourhood clusters’ which correspond to the zones labelled (right). Data were retrieved from http://nanostring.com/CosMx-dataset. (D) LSECs, through their unique phenotypic and functional properties, play a key role in several aspects of liver homeostasis (see section ‘LSEC phenotypical properties and function in physiological conditions and disease’). Damaged LSECs display several changes in their phenotype and function (bottom right) such as loss of fenestrae, formation of basal lamina, decreased endocytic activity, deposition of extracellular matrix (ECM), and activation of Kupffer cells. These changes affect macromolecule transport towards the parenchyma, microbe and virus uptake, and can lead to sustained inflammation and fibrosis.

Citation: Journal of Molecular Endocrinology 71, 2; 10.1530/JME-23-0026

Even though LSECs were first isolated five decades ago (Wisse 1970) and have since been morphologically and functionally characterised (Poisson et al. 2017), the profiling of their transcriptomic and epigenomic features has lagged behind due to difficulties in isolating enough cells and their rapid dedifferentiation in culture (Géraud et al. 2010). The predominance of hepatocytes within the liver tissue (~80% of its mass) has made it difficult to assess and analyse LSEC-specific aspects of gene expression, chromatin accessibility, and transcription factor (TF) occupancy using bulk liver tissue. With advances in cell isolation techniques and single-cell sequencing, the quality and quantity of LSEC-specific data have increased in recent years, along with the identification of key developmental drivers and markers of LSEC identity.

Important progress has also been made in identifying the alterations in the LSEC transcriptome that promote LSEC dysfunction and accompany disease progression, as revealed by animal models and, more recently, by single-cell gene expression profiling of both mouse and human liver tissue. This now substantial body of work pinpoints specific LSEC TFs as conductors of the transcriptional changes associated with the onset and progression of metabolic liver disease. In this review, we discuss the contribution of LSEC TF activity to maintain metabolic homeostasis and how specific LSEC TFs have been implicated in liver disease. Given the intricate relationship between liver metabolism and endocrine health, the relevance of LSEC TFs is expected to extend to this field as well.

LSEC phenotypical properties and function in physiological conditions and disease

The phenotypical features of LSECs have been extensively reviewed in the context of liver physiology (Poisson et al. 2017) and are not the primary focus of the present review. Nevertheless, we provide here a brief overview of their major distinctive features and corresponding functions to contextualise phenotypes observed upon ablation of LSEC TFs.

Reflecting their various specialised functions, LSECs lack basal lamina and display fenestrae (large diaphragm-free pores organised into sieve plates) to aid macromolecule transport towards the perisinusoidal space and subsequently towards the hepatocytes, which are responsible for the metabolic functions of the liver (Fig. 1) (Poisson et al. 2017). The main function of LSEC fenestrae is to aid macromolecule transport towards the space of Disse (perisinusoidal space). Some metabolites such as small chylomicron remnants, drug molecules, and other small molecules including water can freely pass into the space of Disse, resulting in the formation of a para-vascular volume of plasma in this space (Braet & Wisse 2002). Larger molecules are transported through the fenestrae by ‘sieving’, or by transcytosis.

Another important feature of LSECs is the high concentration of scavenger receptors on the cell surface, which enable efficient metabolite clearance and tackling of the viral and bacterial influx from the gut (Sørensen et al. 2015). The better-characterised receptors responsible for macromolecule uptake by LSECs are the mannose receptor (MRC1/CD206), stabilin-1, stabilin-2 (STAB1,2), and the Fc Gamma Receptor IIb (FCGR2B/CD32B). FCGR2B enables the elimination of circulating small soluble immune complexes as the only Fc receptor expressed in LSECs (Ganesan et al. 2012, Sørensen et al. 2012). STAB1/2 mediate the uptake of oxidised and acetylated LDL, glycation end products and waste products, while MRC1 is important in glycoprotein homeostasis and immunity (Elvevold et al. 2008, Schledzewski et al. 2011). Reflecting the high level of expression of these receptors in LSECs compared to other liver cells, the genes encoding these receptors are often used as markers to identify LSECs in transcriptomic studies (Fig. 2).

Figure 2
Figure 2

Marker genes of LSECs identified across eight independent studies employing different experimental methods and/or analysis. The summary of rodent and human datasets presents genes identified by at least four independent studies. The genes are ranked by the number of identifying studies. scRNA-seq studies were considered if they provided marker gene sets for clusters defined as LSECs. All data used for the identification of LSEC clusters are of healthy human/mouse liver. Ramachandran et al. provided a pre-filtered LSEC marker list (n = 80) (Ramachandran et al. 2019). For the remaining scRNA-seq studies, we obtained the top 100 genes ranked by fold-change of LSECs vs other liver cell types (Guilliams et al. 2022). In studies that defined two (MacParland et al. 2018) or three (Aizarani et al. 2019) LSEC clusters, the final list combined the top 100 genes from each of those clusters. Haan et al. identified 27 LSEC-enriched genes by comparing LSECs with heart and brain ECs (de Haan et al. 2020). Géraud et al. provided a 46-gene marker list based on gene expression comparison between EC groups (Géraud et al. 2010). De Smedt et al. developed a computational workflow to identify TFs central in differentiation and specification (CenTFinder). The application of CenTFinder to a series of LSEC gene expression datasets resulted in a list of 80 putative marker genes (De Smedt et al. 2021). On the right, we present scaled mean protein expression for LSECs across four zones (portal node (PN), periportal (PP), pericentral (PC), and central vein (CV)) (n = 4 samples per zone) (Inverso et al. 2021). Genes/proteins not reported in the datasets described are shown with a small empty dot. The full list of marker genes is available in Supplementary Table 1 (see section on supplementary materials given at the end of this article).

Citation: Journal of Molecular Endocrinology 71, 2; 10.1530/JME-23-0026

LSECs display dynamic immune functions and present antigens to CD8+ naïve T cells, contributing to tolerogenic response, while they can also promote T cell activation and local inflammatory response when antigen concentration increases (Limmer et al. 2000, Burgdorf et al. 2007). LSECs are involved in the regulation of angiogenesis and vascular tone in homeostatic conditions (Poisson et al. 2017). Their function in equalising the blood pressure between the portal structures (hepatic artery and portal vein) and the central vein is indispensable, as they produce endothelial vasodilator agents. LSECs are the main source of the vasodilator nitrogen oxide in the liver (Shah et al. 1997), and they also produce paracrine factors (carbon monoxide, thromboxane A2, and prostacyclin) responsible for suppressing the vasoconstrictive function of hepatic stellate cells (i.e. keeping them in a quiescent state) (Fernandez 2015). LSECs also exhibit anti-thrombotic and anti-fibrotic phenotypes (Lafoz et al. 2020).

LSECs help maintain the periportal–pericentral zonation of metabolic pathways in hepatocytes (Box 2) through the conserved, zonated secretion of Wnt2 and Wnt9b morphogens by pericentral LSECs and central venous ECs (Hu et al. 2022). While a detailed examination is outside the scope of the current review, one notable example showed that knockout of the Wnt cargo receptor Evi (Wls) in liver ECs using Stab2-Cre mice resulted in alterations to lipid metabolism, cholesterol production and hepatocyte zonation (Leibing et al. 2018).

Altogether, the features described earlier paint a picture in which LSECs are essential gatekeepers of metabolic homeostasis; moreover, there is a growing body of evidence linking endothelial dysregulation to liver disease onset. During NAFLD progression, LSECs undergo morphological changes that indicate the loss of their specialised phenotype and dedifferentiation. This phenomenon is called sinusoidal capillarisation and involves the loss of fenestrae (defenestration) and the formation of basal lamina (Hammoutene & Rautou 2019). Defenestration results in chylomicron retention in the sinusoidal blood, which leads to hypertriglyceridaemia and hyperlipoproteinaemia (Poisson et al. 2017). Despite several hypotheses, the molecular mechanisms behind capillarisation leading to NAFLD have not been conclusively determined. The timing of sinusoidal capillarisation during NAFLD progression is also still subject to debate. Miyao and colleagues reported that capillarisation was an early event during disease progression, appearing already with simple steatosis in two different mouse models (the choline-deficient, l-amino acid-defined and high-fat diet models) and preceding the activation of Kupffer cells and hepatic stellate cells (Miyao et al. 2015), both of which are critical events in fibrosis (Hernandez-Gea & Friedman 2011). In contrast, a more recent study by Kus et al. has challenged this view, reporting that while LSECs showed an inflammatory response in early NAFLD pathogenesis, there was no sinusoidal capillarisation, but instead increased fenestrae diameter in high fat diet-induced liver steatosis with no immune activation or fibrosis (Kus et al. 2019). These opposing results likely reflect differences in the experimental models employed and encourage further investigation in rodent models, but also in human tissue. Besides defenestration and capillarisation, LSECs display regenerative angiocrine signalling after acute injury, undergoing a fibrogenic switch if the injury is sustained over time (Ding et al. 2014). Furthermore, during the early phase of liver regeneration after hepatectomy, the downregulation of Angiopoietin-2 in LSECs leads to hepatocyte proliferation by releasing the angiocrine proliferative brake. In later stages of regeneration, Angiopoietin-2 expression recovers, enabling angiogenesis in the newly formed tissue (Hu et al. 2014).

LSEC marker genes

The specialised LSEC phenotype is the result of a unique transcriptional and presumably epigenomic signature. It must be noted that LSECs share some important features with other ECs, like the expression of pan-endothelial TFs such as ERG (Dufton et al. 2017). Lymphatic ECs in particular share salient transcriptional features with LSECs, including the expression of lymphatic marker genes LYVE1 and FLT4 (VEGFR3) (Strauss et al. 2017, Aizarani et al. 2019, Inverso et al. 2021) and the expression of Maf in both EC types (Gómez-Salinero et al. 2022b). These and other common features between LSECs and other liver ECs have made the isolation of pure LSEC populations difficult in the past. Therefore, characterising the healthy LSEC transcriptional profile can help identify marker genes for use in LSEC isolation. The definition of marker genes differs between studies but broadly refers to genes enriched in the cell type of interest, often including genes involved in specialised cell functions. LSEC marker genes have been defined by comparing LSECs to other EC types, or to other liver cell types, as is the case in single-cell analysis of liver tissue. These distinct methods identified different marker genes (Fig. 2). For instance, GATA4 was not identified as a marker gene of LSECs in single-cell studies of liver tissue (MacParland et al. 2018, Aizarani et al. 2019, Ramachandran et al. 2019, Andrews et al. 2022, Guilliams et al. 2022), likely because hepatocytes and hepatic stellate cells also express GATA4 (Fig. 3). However, GATA4 is typically not expressed in other ECs and has therefore been identified as an LSEC marker gene in several EC-based studies (Géraud et al. 2017, de Haan et al. 2020, Winkler et al. 2021). Measuring the expression of marker genes can also help to characterise the loss of LSEC identity in disease models or patient biopsies and in developing better in vitro systems to study disease. Despite the established roles of LSECs in liver physiology and disease, what truly constitutes an LSEC transcriptional profile, and which are the marker genes of this cell population is still the focus of active research and debate. In this section, we provide an overview of the progress made in the identification of LSEC marker genes, with a focus on TFs (Box 3).

Box 3 Transcription factors in a nutshell

Transcription factors (TFs) are proteins that bind DNA in a sequence-specific manner and regulate gene transcription. Through binding to specific target regions, which are commonly located in cis-regulatory elements such as promoters, enhancers, or silencers, TFs can activate or repress gene expression in response to regulatory cues. Such regulatory cues can be diverse and include both intrinsic (e.g. developmental processes) and extrinsic signals (e.g. signalling cascades activated by nutrients), making TFs centrepieces in the coordination of complex and dynamic gene regulatory networks. They can either recruit other TFs and/or transcriptional complexes or disable those functions by occupying target regions (Lambert et al. 2018).

Different TF families are characterised by a preferred DNA recognition sequence or motif. The current approaches to identify the consensus motif of specific TFs involve either the analysis of high-throughput binding assays or chromatin immunoprecipitation followed by sequencing (ChIP-seq). The results of these experiments have been compiled in several databases that host vast collections of known TF motifs for vertebrates (e.g. JASPAR, Hocomoco, SwissRegulon). However, the presence of a binding motif does not necessarily translate into TF occupancy, which also depends on conditions including (i) the TF post-translational modifications (Filtz et al. 2014), (ii) the concentration of the TF in the nucleus, (iii) the binding of necessary TFs and co-factors to nearby sequences, and (iv) the specificity of the TF, in other words, its ability to distinguish between a high vs a low affinity binding site (Zabet & Adryan 2015). Most TFs also require chromatin to be accessible at their binding site, free from nucleosomes and thus exposing the motif for recognition. However, a subset of TFs termed pioneer TFs can induce chromatin remodelling: they modify chromatin architecture by binding condensed, nucleosome-bound chromatin and displacing nucleosomes, enabling other TFs and transcriptional machinery to bind to the DNA (Zaret 2020). Pioneer TFs are particularly important in development and lineage identity, with the first pioneer TFs (GATA4 and FOXA1) being identified in the context of hepatic specification (Bossard & Zaret 1998, Gualdi et al. 1996). Regions of accessible chromatin may also have varying levels of activity, reflected by different degrees of deposition of histone marks and DNA methylation. These chromatin features may in turn affect TF binding frequency. In addition to these mechanisms, the activity of TFs is heavily influenced by their post-translational modifications and by protein–protein interactions with binding partners (Jolma et al. 2015). This multitude of mechanisms influencing how TFs act also highlights that the expression of a TF in a cell type and/or location within a tissue does not necessarily correlate with its level of activity. Altogether these factors contribute to the binding of one TF to the DNA, but it must be noted that TFs usually work as part of an intricate system of coordinated action by multiple TFs.

Figure 3
Figure 3

Summary of key transcription factors of LSEC identity presented in this review. (A, B) Single-cell gene expression levels from the Human Liver Cell Atlas (https://www.livercellatlas.org) (Guilliams et al. 2022). Gene expression data were downloaded and annotated with the provided cell annotation matrix, including cell type and UMAP coordinates. UMAPs and violin plots show normalised expression levels, obtained using SCTransform from the Seurat package (v4.3.0) in R. (C) The gene expression levels across the indicated liver cell types are presented as scaled pseudo-count counts per million (CPM) values from scRNA-seq data, as described earlier (Guilliams et al. 2022). Macrovascular ECs represent the mean expression of central vein and portal vein endothelial cells, as seen in panel (A). The LSEC-specific knockouts of TFs that lead to liver fibrosis in mouse models without additional treatments (e.g. KO of another TF, CCl4 treatment) are listed at ‘LSEC KO leads to fibrosis’. The position weight matrices (PWMs) corresponding to the preferred binding motifs of each TF are shown on the right. Motif logos were retrieved from the Jaspar database (Castro-Mondragon et al. 2021). The ZEB2 PWM is not available from the Jaspar database; therefore, the highly homologous ZEB1 (ZEB2 paralog) motif is listed; however, the PWMs of both motifs are highly concordant (HOMER database, not shown). 1: (Winkler et al. 2021); 2: (Dufton et al. 2017).

Citation: Journal of Molecular Endocrinology 71, 2; 10.1530/JME-23-0026

There have been several attempts to describe the transcriptional features of LSECs, particularly to pinpoint the set of LSEC lineage-determining TFs (Fig. 2). A major challenge in the study of LSECs is their rapid in vitro dedifferentiation. This dedifferentiation was demonstrated at the transcriptional level in a comparative analysis of freshly isolated rat LSECs vs LSECs cultured for 42 h, which revealed 465 genes downregulated by culture, including the TFs Gata4, Tfec (Tcfec), and Maf (Géraud et al. 2010). In a recent study, de Haan et al. described an LSEC fingerprint composed of 27 LSEC-enriched genes, defined as genes enriched in LSECs vs heart and brain ECs, plus three previously identified important genes (LYVE1, STAB1, and CLEC4M (LSIGN)). This LSEC fingerprint included seven genes encoding TFs: GATA4, TFEC, MAF, ZEB2, MEIS2, HOXB5, and CUX2 (de Haan et al. 2020).

As mentioned earlier, it is challenging to isolate pure LSEC populations due to the overlapping expression of markers across different types of ECs. de Haan et al. used a pan-endothelial marker, Tie2, to isolate mouse liver ECs, reporting a purity of >99% for the microvascular marker CD36 and low expression of lymphatic EC markers (de Haan et al. 2020). However, CD36 is less abundant in central venous LSECs than in periportal LSECs (Su et al. 2021) and therefore may not be an ideal total-LSEC reporter for cell sorting. Similarly, other markers used for cell isolation may be zonated, showing varying expression along the periportal–pericentral axis (Ben-Moshe & Itzkovitz 2019) (Box 2), which raises an important consideration around the methods used to isolate LSECs for molecular characterisation.

Recent years have seen the application of sequencing at single-cell resolution, including single-cell RNA-seq (scRNA-seq), which can be used to profile a complex pool of cells and capture cell-to-cell heterogeneity. scRNA-seq can be applied to liver tissue or to pre-sorted ECs, with similar cells clustered by computational algorithms based on shared transcriptomic features. Thus far, most scRNA-seq studies of LSECs have relied on annotating LSECs based on the enriched expression of established LSEC marker genes compared with other cells. These studies have identified additional sets of genes that distinguish LSECs from other liver cell types (Ben-Moshe & Itzkovitz 2019, Saviano et al. 2020) as well as genes with heterogeneous expression across LSEC sub-populations and that are zonated (Inverso et al. 2021, Su et al. 2021).

Several scRNA-seq studies of the liver have highlighted TFs with enriched expression in LSECs. A seminal study of the healthy and cirrhotic human liver using scRNA-seq identified a cluster of LSECs with high expression of TFs including GATA3/4, STAT2/3, MEIS1, TBX2, NR5A2, NR2F1, ELK1, ETS1, HIF1A, SREBF1, RARB, IRF1, ATF3, MLXIP, XBP1, FOXO1, MEF2C, CEBPB, JUND, BACH1, and JUN (Ramachandran et al. 2019). Many of these are recognised core LSEC TFs, yet the lack of standardised scRNA-seq methods, including on how LSEC clusters are identified, and the high transcriptional heterogeneity of LSECs means that it is still difficult to reach a consensus list of marker genes (Fig. 2).

In addition to markers of mature LSECs, some interesting results have come from the application of single-cell methods to different stages of liver development. For example, the Rafii team observed that LSECs constitute ‘the most transcriptional diverse’ liver EC type when comparing developmental stages (Gómez-Salinero et al. 2022b ). In this study, the authors employed scRNA-seq on CD45-CD31+ cells to characterise the mouse liver endothelium throughout embryonic (E12-E18) and postnatal (P2, P8, P15, and P30) development, describing marker genes for a total of ten LSEC clusters. The genes Cd34, Pgk1, and Mif were reported as markers of undifferentiated LSECs; whilst Aqp1, Mrc1, Fcgr2b, Clec4g, and Kit were markers of adult differentiated LSECs.

Another important angle to the identification of LSEC markers is the possibility of improving the understanding of liver pathophysiological processes. Studies using the CCl4 liver injury model have revealed that the transcriptional profile of LSECs is markedly changed during liver injury with over 7000 differentially expressed genes (Manicardi et al. 2021). The transcriptional changes included increased expression of genes encoding secretory proteins (e.g. Cxcl10, Inhbb, Tpbpb, and Il4ra) and decreased expression of membrane and transport-related genes (Manicardi et al. 2021), possibly revealing LSEC dedifferentiation under stress. Single-cell transcriptomics has proved to be a powerful tool to investigate liver disease states as well, revealing for instance that cirrhotic livers show a substantial depletion of LSECs along with the appearance of cells with distinct transcriptional signatures, as revealed by studies in both human (Ramachandran et al. 2019) and mouse liver (Xiong et al. 2019, Su et al. 2021). Su et al. observed that periportal LSECs may be the most vulnerable to injury, implicating in this process the TFs Klf2 and Klf4, and components of the activator protein 1 (AP-1) complex, with reduced expression in cirrhotic mouse LSECs (Su et al. 2021).

The transcription factors that orchestrate LSEC transcriptional identity

The specialised LSEC phenotype we described earlier is likely the result of the orchestrated action of multiple TFs (Box 3) that are responsible for initiating and maintaining the LSEC-specific gene expression programme. The sinusoidal endothelium develops from the mesenchyme of the septum transversum (mesoderm) and differentiates under the transcriptional control of GATA4 and c-MAF (Asahina et al. 2011, Géraud et al. 2017, Gómez-Salinero et al. 2022b ). Interestingly, these two TFs remain central to the maintenance of LSEC function and identity in adulthood. A less investigated but seemingly important TF is MEIS2, which has been identified by a few studies in recent years as a main regulator of LSEC fate (de Haan et al. 2020, Liang et al. 2022). Overexpression of this TF triad was shown to be partially successful in promoting an LSEC-like gene programme in human umbilical vein endothelial cells (HUVECs), inducing the expression of a number of LSEC markers, albeit with an incomplete recapitulation of LSEC functionality (de Haan et al. 2020). Whole transcriptome analyses were not carried out to investigate the extent to which the co-expression of GATA4, c-MAF, and MEIS2 induced an LSEC transcriptional profile (de Haan et al. 2020). Other TFs are therefore likely required for the complete establishment of the LSEC transcriptional programme (Danoy et al. 2020). Recent studies have implicated SPI1 and ZEB2 in the regulation of genes related to LSEC immune and angiogenic functions, respectively (De Smedt et al. 2021, de Haan et al. 2022). Besides LSEC-enriched TFs, pan-endothelial TFs also play key roles in LSEC identity and function. ERG, as the better-established example of a pan-endothelial TF, is essential for LSEC homeostasis, protecting from liver fibrosis (Dufton et al. 2017). In this section, we describe in more detail the TFs that have been observed to drive the LSEC transcriptional machinery (Fig. 3) and discuss how they have been implicated in liver metabolic disease.

c-MAF

MAF factors, including c-MAF, are subunits of the AP-1 transcriptional complex and have a strong preference for nucleosome-depleted regions (Grossman et al. 2018). c-MAF, encoded by the MAF gene in humans, has been identified by several studies as a TF of central importance in LSEC development and function (Géraud et al. 2010, de Haan et al. 2020, Gómez-Salinero et al. 2022b ). Recently, Gómez-Salinero et al. identified c-Maf as the top-enriched TF in liver ECs using publicly available scRNA-seq datasets representing multiple mouse tissues (Gómez-Salinero et al. 2022b ). Maf was shown to increase expression continuously throughout mouse liver embryonic and postnatal development by scRNA-seq analysis (Gómez-Salinero et al. 2022b ). This contrasted with Gata4 (described in detail in the next section), which showed uniform expression across liver EC types and developmental stages. Thus, despite the important roles of these two TFs, their specific contributions to the regulation of LSEC homeostasis seem to be distinct. scRNA-seq analysis of adult human liver ECs further revealed that MAF expression is higher in the liver sinusoid compared to either portal or central vein ECs (Fig. 3) (Gómez-Salinero et al. 2022b, Guilliams et al. 2022).

c-Maf has a key role in regulating LSEC maturation and identity. Gómez-Salinero et al. also observed increased co-expression of Maf with the LSEC markers Mrc1 and Fcgr2b: inducible endothelial-specific ablation of Maf (VE-cadherin(Cdh5)-CreERT2/c-Mafflox-flox) at E12–14, postnatally and in adult mice led to a reduction in the expression of Mrc1 and Fcgr2b (Gómez-Salinero et al. 2022b). RNA velocity analysis revealed that postnatal c-Maf deletion led to an immature EC phenotype, with decreased expression of LSEC markers and overexpression of arterial genes, such as Cd34, Ly6a, Aplnr, and Cd9, which are associated with the retention of liver haematopoiesis. In adult mice, deletion of c-Maf also led to decreased expression of LSEC markers, but this time it associated with increased expression of genes characteristic of portal vein ECs (Gómez-Salinero et al. 2022b).

c-Maf also seems to be required for phenotypical specification along the periportal–pericentral axis, as its deletion during embryonic development resulted in an aberrant zonation phenotype and in the appearance of EC clusters detected by scRNA-seq that were not present in wildtype mice (Gómez-Salinero et al. 2022b). In adult mice, c-Maf removal resulted in a mild liver zonation phenotype with the expansion of hepatocytes expressing glutamine synthase, an enzyme known to be exclusively expressed in the pericentral zone (Gebhardt et al. 2007), even though other zonation markers such as Cdh1 (E-cadherin) or Cyp2E1 were not affected (Gómez-Salinero et al. 2022b). Consistent with the role of c-Maf in regulating and maintaining LSEC maturation, its expression in liver ECs was diminished upon induction of fibrosis with CCl4 for a month (Gómez-Salinero et al. 2022b). Liver EC-specific c-Maf deletion alone did not induce fibrosis in mice, but its combination with CCl4 resulted in an exacerbation of the fibrotic phenotype (Gómez-Salinero et al. 2022b). These observations in murine models suggest that c-MAF loss may be a contributing factor to human NASH. Furthermore, the implications of c-MAF in liver disease are likely not restricted to the endothelial compartment, as c-MAF is also highly expressed in Kupffer cells and scar-associated macrophages (Fig. 3) (Ramachandran et al. 2019, Guilliams et al. 2022). Moreover, in addition to LSECs, lymphatic ECs also express c-MAF, rendering further investigations of the role of LSEC c-MAF in liver metabolic homeostasis and disease with complementary systems, for example, using other liver EC-specific Cre recombinases.

TFs may contribute to gene regulatory programmes via different mechanisms and not all TFs can initiate chromatin remodelling as pioneer TFs (Box 3). Still, TFs such as c-MAF can modulate the accessibility of regulatory elements and in this way promote gene programmes. Two studies have reported that in vitro overexpression of c-MAF in HUVECs induced a pro-sinusoidal transcriptional programme and the expression of sinusoidal identity genes (de Haan et al. 2020, Gómez-Salinero et al. 2022b). Microvascular and LSEC markers CD36, DPP4 (CD26), and STAB1/2 were induced at the mRNA level, while CD14 and MRC1 induction was also observed at the protein level. RNA-seq analysis of ‘induced LSECs’, identified as CD26+CD36+, recapitulated broad transcriptional features of human primary LSECs. Still, the expression of a subset of sinusoidal markers, including FCGR2B, was not induced by c-MAF overexpression (Gómez-Salinero et al. 2022b).

Until the recent studies investigating the transcriptional properties of LSECs, c-MAF was primarily known as a regulator of lymphocyte differentiation and function (Imbratta et al. 2020). c-MAF has been described to promote chromatin accessibility at its binding sites in subsets of lymphoid cells, which is in line with its prominent developmental role (Parker et al. 2020). While the function of c-MAF in directing LSEC chromatin accessibility has not been investigated, it is plausible to assume it has similar functions in promoting transcriptional programmes in the liver endothelium.

The identification of TFs able to induce transcriptional remodelling, such as c-MAF, may be of particular importance to develop in vitro models and potential sources of LSECs and other liver cell types for regenerative medicine applications. Encouraging results were already observed in co-cultures of ‘induced LSECs’ generated by overexpression of c-MAF with human primary hepatocyte aggregates, which showed formation of cytoplasmic fenestration gaps in ‘induced LSECs’ and sustained induction of hepatocyte functionality (CYP1A2 expression and albumin secretion) for at least 28 days (Gómez-Salinero et al. 2022b).

GATA4

GATA4 is a zinc-finger TF that is crucial in liver development (Watt et al. 2007) and in the maintenance of a differentiated LSEC state. Like other members of the GATA TF family, GATA4 is capable of both activating and repressing gene expression, acting together with co-regulators to assemble a transcriptional complex and recruit chromatin remodelling machinery (GATA transcriptional complexes are described in detail in (Tremblay et al. 2018)). During embryonic development, GATA4 primes liver-specific regulatory elements through its pioneer TF activity (Bossard & Zaret 1998). In the adult liver, GATA4 is expressed in LSECs and hepatocytes, albeit its level is lower in hepatocytes.

Comparing freshly isolated rat LSECs to lung microvascular ECs and to LSECs cultured for 42 h, Géraud et al. identified Gata4 as a TF that is preferentially expressed in LSECs and rapidly lost upon culturing (Géraud et al. 2010). The same group later demonstrated that Gata4 expression in LSEC is essential for fetal development. Using Stab2-Cre to delete Gata4 only in LSEC, the authors observed that all mouse embryos with LSEC-specific Gata4 knockout showed lethality at E15.5-E17.5 (Géraud et al. 2017). This severe phenotype was associated with very early sinusoidal capillarisation, detected at E10.5, and loss of key LSEC identity genes, including Lyve1 and Stab2, as well as upregulation of endothelial genes that are mainly involved in cell junction formation, such as Pecam1 (Cd31) (Géraud et al. 2017). LSEC-specific Gata4 deletion also led to reduced numbers of stem and progenitor cells in the liver at E11.25, suggesting that Gata4 is necessary for hematopoietic stem cell (HSC) migration into the fetal liver (Géraud et al. 2017). Overexpression of GATA4 in HUVECs suppressed expression of the junctional molecule Cadherin-5 (VE-cadherin), suggesting that GATA4 decreases junctional stability in fetal LSECs and thus the permissiveness of the liver parenchyma for HSCs (Géraud et al. 2017). Whether other angiocrine factors or cytokines also play a role in establishing the liver HSC niche remains unknown.

The studies described above provide evidence that Gata4 is essential for mouse embryonic development; furthermore, other studies inform about the LSEC Gata4 role in liver metabolic homeostasis. Deletion of Gata4 in LSECs at E17.5 using a Clec4g-driven inducible Cre recombinase (Clec4g-iCretg/0 ) resulted in impaired hepatocyte function in 3-month-old mice, including elevated aspartate and alanine aminotransferase levels (Winkler et al. 2021), measurements that often correlate with NAFLD in humans (Sookoian et al. 2016). Analysis of livers from mice with LSEC-specific Gata4 ablation (Gata4LSEC-KO ) showed disruption of hepatic zonation and of major metabolic functions, including the metabolism of fatty acids, bile acids, and xenobiotics, as well as oxidative phosphorylation. These pathophysiological alterations were accompanied by marked transcriptional remodelling, showing activation of angiogenic and MYC-dependent gene programmes (Winkler et al. 2021). Notably, the ablation of Gata4 in LSEC led to the downregulation of c-Maf, a result that is consistent with the hypothesis that these two core LSEC TFs are part of the same TF network (Wilkinson et al. 2017).

In agreement with the studies of c-Maf deletion models, the early capillarisation phenotype observed in Gata4LSEC-KO mice led to the hypothesis that Gata4 could be important in the prevention of liver fibrosis (Winkler et al. 2021). LSEC-restricted deletion of Gata4 at E10.5 was associated with increased ECM deposition, as shown by upregulation of ECM-associated genes at the mRNA and protein levels (Géraud et al. 2017); while its deletion at a later developmental stage also led to sinusoidal capillarisation and caused perisinusoidal fibrosis in adult mice, with activation and expansion of hepatic stellate cells, and increased presence of infiltrating inflammatory cells (Winkler et al. 2021). Perisinusoidal fibrosis is characteristic of NASH (Takahashi & Fukusato 2014); and indeed, the LSEC transcriptional profiles of Gata4LSEC-KO mice were similar to those observed in a diet-induced liver fibrosis model (Winkler et al. 2021). Further highlighting the active role of GATA4 in liver metabolic disease, analysis of livers from a diet-induced model of NASH showed a strong downregulation of Gata4 (Winkler et al. 2021), which was also observed in scRNA-seq from three human cirrhotic livers (CD45- non-parenchymal cells were profiled from one NAFLD and two alcoholic liver disease donors) (Ramachandran et al. 2019).

In addition to phenotypical and transcriptomic analyses, the direct interrogation of chromatin activity and TF binding has the potential to provide additional insights into the modes of action of TFs. To unravel the mechanisms of GATA4 activity, Winkler et al. employed ATAC-seq to identify accessible chromatin regions in LSECs from wildtype and Gata4 LSEC-KO mice (Winkler et al. 2021). This analysis revealed that GATA4 represses a continuous EC gene programme via a reduction in chromatin accessibility. Integration of these results with whole liver Gata4 ChIP-seq implicated chromatin occupancy by Gata4 at sites with reduced accessibility. For instance, Pdgfb, which is bound by Gata4 in the liver, was upregulated and its promoter became more accessible in LSECs from Gata4 LSEC-KO mice (Winkler et al. 2021). This association between Gata4 and Pdgfb is quite interesting since platelet-derived growth factor (PDGF) signalling is associated with stellate cell activation and fibrosis (Bataller & Brenner 2005). Pdgfb is normally expressed in continuous ECs but not in LSECs (Winkler et al. 2021).

MEIS2

MEIS2 belongs to the TF superclass TALE (three amino acid loop extension), which is a highly conserved family of homeobox proteins (Bürglin 1997). Several members of the TALE family have been implicated in vertebrate embryogenesis gene programmes, driving cell fate specification during segmentation and in later developmental stages (Moens & Selleri 2006), contributing to the development of multiple tissues and organs, including the brain, limb, lens, retina, and heart (Machon et al. 2015). Mutant mice lacking Meis2 (Hprt1-Cre) displayed embryonic lethality between E13.5–E14.5, showing haemorrhaging and a small liver size compared to healthy controls (Machon et al. 2015), suggesting a role of Meis2 in embryonic megakaryopoiesis and haematopoiesis. This is consistent with studies showing that Meis2 overexpression in mouse embryonic stem cells promoted haematopoiesis (Cai et al. 2012).

Recent screens for lineage-determining TFs using comparative transcriptomics have suggested MEIS2 to have an important role in the establishment and maintenance of an LSEC-differentiated phenotype, with primary LSECs rapidly losing MEIS2 expression in vitro (de Haan et al. 2020, Liang et al. 2022). scRNA-seq analysis indicated that Meis2 expression and regulatory activity increase along the developmental trajectory, possibly contributing to sinusoidal build-up during embryonic and postnatal development (Danoy et al. 2020). In the adult liver, MEIS2 is most expressed in ECs compared to other cell types, although we note that its expression is also relatively high in cholangiocytes and hepatic stellate cells (Fig. 3). Transcriptomic analysis comparing human pluripotent stem cells (hiPSCs) vs hiPSC-derived LSECs using nanoCAGE revealed that MEIS2 was upregulated in hiPSC-derived LSECs, along with ERG, MAF and other known LSEC TFs (Danoy et al. 2020). In this study, the DNA-binding motif of MEIS2 was identified within the top 10 most important motifs in driving the hiPSC-derived LSEC transcriptional profile. Moreover, a regulatory network analysis implicated MEIS2 in the direct regulation of the LSEC marker gene LYVE1 and of the vascular endothelial growth factor (VEGF) signalling pathway (Danoy et al. 2020). Overexpression studies in HUVECs have also shown that MEIS2 induces the expression of classic LSEC genes, including F8, IL1A, CLEC4M, and STAB1 (de Haan et al. 2020). Studies in other developmental contexts, namely palatal bone development, suggested that modulation of Meis2 activity associates with changes in target gene accessibility at both promoter and distal regulatory regions (Machon et al. 2015). It remains to be investigated if Meis2 regulates gene expression through the same mechanisms during LSEC specification.

ERG

The ETS-related gene (ERG) is a lineage-specific master regulator of endothelial gene expression (reviewed in (Shah et al. 2016)). ERG is required during embryonic vascular development, angiogenesis, and for maintenance of vascular homeostasis. ERG expression appears first in the mesoderm during early development (E8.5 in mice) and is maintained throughout adulthood at consistent levels in arterial, venous and microvascular endothelium (Vlaeminck-Guillem et al. 2000). ERG is indispensable for vascular development: constitutive homozygous deletion of endothelial Erg causes embryonic lethality in mice (E10.5–12.5) due to disruption of cardiovascular development (Vijayaraj et al. 2012, Birdsey et al. 2015). The crucial role of ERG as a lineage-determining TF is demonstrated by studies where ectopic expression of ERG contributed to cell fate reprogramming. For instance, lentiviral overexpression of ERG in embryonic or adult murine somatic fibroblasts along with hematopoietic lineage-determining TFs (GATA2, LMO2, RUNX1c, and TAL1) reprogrammed fibroblasts to hematopoietic progenitors (Batta et al. 2014). Expression of ETS factors ERG, FLI1, and ETV2 in combination with TGFβ pathway inhibition was able to reprogramme human amniotic cells into vascular endothelium (Ginsberg et al. 2012). Moreover, ectopic expression of the pioneer factor ETV2 (Gong et al. 2022) converted primary human adult skin fibroblasts into functional ECs through the activation of ERG (Morita et al. 2015). Besides pointing to the importance of ERG as a developmental TF, these studies highlight the potential of combining ERG with other endothelial TFs for the generation and/or improvement of in vitro EC models through the induction of endothelial-specific cell fate. This is exemplified by the recent finding that ERG and FLI1 cooperate to activate a vascular gene expression programme in adult human mesenchymal stromal cells (Gómez-Salinero et al. 2022a).

ChIP-seq analysis of HUVECs revealed ERG binding to multiple sites of accessible open chromatin, including promoters and enhancers (Box 4) (Kalna et al. 2019). Globally, ERG binding in HUVECs was greatest at active enhancers; ~67% of ERG-bound sites were located at either distal intergenic or intragenic regions, supporting a role for ERG-mediated transactivation of gene expression through EC enhancers (Kalna et al. 2019). In line with its role as a lineage-determining TF, ERG is bound at the vast majority (93%) of HUVEC super-enhancers (Kalna et al. 2019), a class of enhancer clusters that are associated with genes that define cell lineage identity and regulate tissue-specific functions (Hnisz et al. 2013). Notably, siRNA-mediated inhibition of ERG in HUVECs led to changes in H3K27ac enrichment at enhancers and to the redistribution of a subset of core super-enhancers associated with essential endothelial genes (Kalna et al. 2019).

In agreement with many other TFs, ERG can both drive and repress gene expression. Studies have shown that ERG controls transcription through either cooperation or competition with other TFs. For example, ERG cooperates with the TF KLF2 to drive the expression of the anticoagulant cell surface protein thrombomodulin. Interestingly, this mechanism is organotypic, since it was observed in LSECs, but not aortic ECs. Molecular studies showed that ERG is required for KLF2 to access chromatin, by recruiting p300 and mediating H3K27ac in low shear stress conditions, as found in LSECs, but is dispensable in high shear stress conditions, as found in the aorta (Peghaire et al. 2019). An example of competition is the relationship between ERG and SMAD3: here, ERG prevents SMAD3 binding to chromatin at genes driven by the TGFβ-TGFBR1 (ALK5) pathway, thus promoting homeostasis and preventing endothelial-to-mesenchymal transition (EndoMT) (Dufton et al. 2017). EndoMT is characterised by the loss of endothelial lineage markers, morphology, and function and is associated with multiple chronic diseases. EndoMT also contributes to LSEC capillarisation and ECM production in liver fibrosis (Ruan et al. 2021). In the liver, Dufton et al. showed that EC-specific deletion of Erg in mice (constitutive heterozygous deletion with Tie2-Cre/Ergfl/+ and inducible homozygous deletion with Pdgfb-iCreER-eGFP/Ergfl/fl) led to spontaneous EndoMT and liver fibrogenesis (Dufton et al. 2017). This was associated with disrupted portal tracts and increased periportal collagen deposition – features consistent with a fibrotic phenotype (Dufton et al. 2017). The association between loss of ERG expression and liver fibrogenesis was further supported by the observation that ERG expression is significantly downregulated in human liver fibrosis and cirrhosis (Dufton et al. 2017).

Box 4 Enhancers

Enhancers are non-coding cis-regulatory elements that activate target gene expression through the recruitment of TFs, cofactors (coactivators and corepressors), and basal transcriptional machinery. Enhancers are mostly intergenic or intronic regions of open chromatin that can affect target genes located distally, even a million base pairs away, in an orientation-independent manner (Panigrahi & O’Malley 2021). Since enhancers are free from nucleosomes, they are accessible to enzymatic action. Frequently used methods to map enhancers are based on profiling chromatin accessibility by ATAC-seq or DNase-seq in conjunction with ChIP-seq for active histone marks (e.g. H3K27ac). The most accepted theory of enhancer mechanism of action is chromatin looping, during which a distal enhancer comes into physical contact with its target gene’s promoter. The activation of gene expression occurs through the recruitment and binding of activating TF(s) by the enhancer region, which boosts RNA polymerase II-dependent transcription at the promoter of the target gene. Genetic variants at enhancers can affect the binding of TFs, leading to altered target gene expression. Interestingly, common disease variants have been shown to locate in non-coding regions, with enhancer elements being particularly enriched for them (Maurano et al. 2012). Although enhancers have been investigated for decades, several challenges still hamper the characterisation of enhancer role in disease, recently reviewed by Zaugg and colleagues (Zaugg et al. 2022).

SPI1

SPI1 (often referred to as PU.1) is another member of the ETS family that has been highlighted recently as of potential relevance in LSEC biology. However, it must be noted that most studies highlighting SPI1 in LSECs were based on TF motif enrichment analyses and may thus reflect the high degree of similarity between the motifs recognised by SPI1 and other ETS factors (Fig. 3), particularly ERG which we discussed in the previous section. A computational workflow, CenTFinder, was employed to identify and rank TFs driving lineage specification (De Smedt et al. 2021), identifying SPI1 as a regulator of immune response transcriptional programmes in LSECs. Similarly, the activity of the SPI1 motif was reported to be induced during the differentiation of hiPSC-derived LSECs (Danoy et al. 2020), and when overexpressed in stem cells along with ETV2, SPI1 induced an LSEC-like phenotype, including the expression of the LSEC markers FCGRB2 (CD32B), LYVE1, MRC1, CRHBP, FCN3, and OIT3 (De Smedt et al. 2021). However, marker genes that are not related to immune functions, such as STAB1/2, CLEC4G, and CLEC4M, were not upregulated (De Smedt et al. 2021). These results may reflect an inherent affinity of SPI1 to regulate immune genes, given its broadly reported role as a regulator of immune cell differentiation (Li et al. 2020). The evidence linking SPI1 to liver metabolic disease is limited, but SPI1 has been implicated in the regulation of transcriptional changes observed in NASH (Steensels et al. 2020). The SPI1 binding motif has also been reported to be enriched in NAFLD genetic risk variants (Namjou et al. 2019). It remains to be investigated if these associations between SPI1 activity and NAFLD pertain to LSEC and/or other liver cell types, especially since its expression in LSECs is largely undetectable in adult human and mouse scRNA-seq (Fig. 3, https://www.livercellatlas.org/, https://tabula-muris.ds.czbiohub.org/). Future studies should also investigate if SPI1 drives LSEC identity during embryonic development, a question that will likely require investigations using tailored in vivo models, as those already created for other LSEC TFs. In a developmental context, it is possible that it acts as a pioneer TF during LSEC differentiation, similar to its function in hematopoietic development where it opens stem cell heterochromatin (Pham et al. 2013).

ZEB2

ZEB2 is one of the two members of the vertebrate family of zinc-finger E-box binding homeodomain (or ZEB) TFs. ZEB2 was originally named SIP1, as it was first discovered in a 2-hybrid screen for SMAD-interacting proteins (SIPs). ZEB2 is a repressor of SMAD signalling (Postigo et al. 2003) predominantly known for its role in regulating nervous system development and in epithelial-to-mesenchymal transition in cancer (Fardi et al. 2019). ZEB2 was recently shown to be enriched in liver microvascular ECs (de Haan et al. 2020). However, experimental follow-up for the role of ZEB2 in LSECs showed uniform expression of Zeb2 in mouse LSECs and non-sinusoidal vessels of the liver (de Haan et al. 2022), and scRNA-seq also indicates that Zeb2 is highly expressed in immune cells in adult human liver (Fig. 3). Conditional deletion of Zeb2 in mouse ECs (VE-cadherin(Cdh5)-CreERT2/Zeb2flox-flox, Zeb2 ECKO) revealed that loss of Zeb2 affected Pdgf-signalling and angiogenic genes, but Lyve1 was the only LSEC marker showing decreased expression. Still, the Zeb2 ECKO mice presented a denser and irregularly shaped liver microvasculature. Upon CCl4-induced fibrosis, immunostaining in Zeb2 ECKO livers showed a gain of continuous EC markers (Cd34, Pecam1), loss of the LSEC marker Fcgr2b, and formation of basal lamina (de Haan et al. 2022). Considering its important role in other developmental programmes, these studies uncover ZEB2 as an interesting candidate for future investigations in LSECs and liver disease.

Future directions

There is an unmet need to identify the drivers of LSEC transcriptional programmes and cellular identity to create better in vitro model systems and provide therapeutic targets and biomarkers for liver diseases such as NAFLD. As highlighted by the studies discussed in this review, liver tissue and freshly isolated LSECs are still the gold standards for LSEC expression analysis, with many studies carried out in rodent LSECs. However, this creates limitations in terms of species differences and availability of this difficult-to-obtain cell type and has important ethical implications. Moreover, the development of reliable LSEC in vitro models may improve liver organoid modelling systems and platforms for regenerative therapy. The collected evidence on developmental LSEC TFs and marker genes brings us closer to the possibility of creating a TF overexpression model in more readily available, less differentiated ECs, such as HUVECs. Overexpression of c-MAF, GATA4, and MEIS2 in HUVECs induced partial LSEC phenotype in an additive way, surpassing the level of induction achieved by either of the TFs alone (de Haan et al. 2020). The combination of TF overexpression with the addition of microenvironmental molecules may also be necessary to induce and maintain the LSEC signature. For instance, bone morphogenic protein 9 (BMP9), a circulating endothelial quiescence factor, has been suggested as a necessary maintenance factor for LSEC fenestration (David et al. 2008). Bmp9 knockout in mice reduced fenestration frequency, as opposed to BMP9 treatment in cultured LSEC, which lead to prolonged fenestrated phenotype (Desroches-Castan et al. 2019). Bmp9 treatment also induced c-Maf expression and has been suggested to promote LSEC identity (Desroches-Castan et al. 2019, Gómez-Salinero et al. 2022b). Recently, Gage et al. differentiated LSEC-like cells from hiPSCs by sequentially manipulating signalling molecule levels and pathways, however, some important marker levels (e.g. F8) still remained low in hiPSC-derived LSECs, demonstrating incomplete conversion (Gage et al. 2020). Heterotypic interactions between LSECs and other liver resident cell types are also important for the maintenance of LSEC identity in vitro, as it has been exemplified in LSEC-hepatocyte co-culture experiments (Gómez-Salinero et al. 2022b) and in more complex systems such as organoids and liver-on-a-chip (Rezvani et al. 2023). Future work should address the completion of the LSEC model based on the overexpression of TFs described in this review and looking in more detail at the cell-to-cell signalling pathways that contribute to specification of the liver endothelium during embryonic and postnatal development.

It is clear from previous studies that the spatiotemporal expression pattern of LSEC TFs is diverse. In terms of cellular specificity, c-MAF, GATA4, MEIS2, and ZEB2 are LSEC-enriched TFs. We refer here to LSEC-enriched TFs in opposition to LSEC-specific because even though the expression of these TFs is higher in LSECs in comparison to other ECs and/or other liver cells, their expression is not exclusively detected in LSECs. As we noted previously, c-MAF is expressed in LSECs, Kupffer cells and macrophages, while GATA4 is present in LSECs, hepatocytes and hepatic stellate cells. The temporal expression patterns can also differ between TFs, as illustrated when comparing Gata4 and Maf during embryonic development: Gata4 expression is constant throughout development, whereas Maf expression increases (Gómez-Salinero et al. 2022b). These features have implications for the potential roles of these and other TFs in liver pathophysiology and suggest that dysfunction of at least some of these LSEC core TFs may lead to pleiotropic effects across multiple cell types.

This review highlights a variety of methods used to study LSECs, particularly at the transcriptional level. The integration of different datasets has been important for addressing the limitations of any one method – this was shown recently in the differences of single-cell technologies (single-nuclei- vs scRNA-seq) for capturing zonated LSECs (Andrews et al. 2022). In addition to differential gene expression, other mechanisms may drive zonated transcriptional networks. For instance, differences in translation rates or protein stability may create a disconnection between mRNA and protein levels for certain TFs along the periportal–pericentral axis (Inverso et al. 2021). Post-translational modifications are often required to regulate TF activity and their ability to partner with other TFs (Filtz et al. 2014), meaning that the detection of a TF in a cell type and/or location within a tissue does not necessarily correlate with TF activity. Similarly, differential DNA methylation of TF binding sites may contribute to the zonated activity of LSEC TFs, as has been observed in human hepatocytes (Brosch et al. 2018). Thus, future single-cell multi-omics analyses may reveal additional layers of the LSEC regulome. This and other new advances, including spatial transcriptomics (Hu et al. 2022) (Fig. 1A), and proteomics (Inverso et al. 2021) are sure to add new dimensions to our understanding of LSEC biology. Moreover, access to LSEC-specific data from patients with acute and chronic liver disease is currently very limited. While single-cell data can effectively resolve specific cell types, there is not yet published data comparing NAFLD or its spectrum of phenotypes to a healthy liver. For example, two published studies of the LSEC transcriptome capture cirrhotic liver from alcoholic liver disease patients (bulk RNA-seq (Manicardi et al. 2021)) and a combination of alcoholic liver disease and NAFLD patients (scRNA-seq (Ramachandran et al. 2019)). We anticipate that the generation of new single-cell datasets from liver disease patients vs healthy donors, with gold-standard data sharing practices such as sharing full differential gene expression results, will greatly enhance our understanding of LSEC damage and its consequences in liver disease.

Cis-regulatory networks involve the coordinated activity of multiple TFs that bind to promoters and enhancers to activate broad gene programmes. The liver has been extensively studied in the contexts of health and disease, particularly NAFLD, to show that there is impaired hepatic TF activity in liver metabolic disease (reviewed in (Cebola 2020)). However, as mentioned earlier, such studies focused on bulk liver tissue analyses and did not have the granularity needed to capture TF activity in specific liver cell populations, such as LSECs. Thus, future studies employing a combination of better in vitro models and single-cell omics are expected to contribute to the better characterisation of the cistromes and transcriptional networks under the control of specific LSEC TFs. It is equally important that we seek to understand how LSEC TFs work in a coordinated fashion to promote the different LSEC phenotypical attributes, especially those implicated in disease. A fundamental and still unanswered question is ‘How does the pan-endothelial TF ERG act with LSEC-enriched TFs to drive the LSEC transcriptional profile?’. Are these TFs part of a single cis-regulatory network, whereby for instance ERG controls the expression of GATA4? Or are these TFs part of separate transcriptional networks and thus in charge of regulating distinct sets of genes in LSECs? We predict that the application of single-cell ATAC-seq (Cusanovich et al. 2015) and single-cell CUT&Tag (Bartosovic et al. 2021), among other methods, will accelerate such findings and provide answers to some of these questions. Finally, more efficient and standardised data processing protocols along with advanced data integration methods, including the application of deep learning algorithms (Ma & Xu 2022), will further enable new discoveries into the functions of LSEC TFs and their individual contributions to liver homeostasis and disease.

It is well established that genetic variants can contribute to a higher risk of developing NAFLD and other liver metabolic diseases. Some variants may affect TF activity by altering their protein sequence, which may lead for example to changes in DNA-binding affinity. A recent genome-wide association study for NAFLD in a histology-characterised cohort, identified a missense variant in the PYGO1 gene, which encodes a TF from the canonical Wnt signalling pathway (Anstee et al. 2020). Wnt signalling is essential for appropriate sinusoidal differentiation (Birdsey et al. 2015) and its aberrant activation has been associated with fibrosis (Inverso et al. 2021). It would therefore be interesting to characterise Wnt signalling and TF activity when PYGO1 carries the recently identified NAFLD risk allele.

Most common disease-associated genetic variants however do not affect protein coding sequences and are instead located within cell type-specific enhancers (Maurano et al. 2012). These common variants may act by disrupting TF binding events, which could lead to changes in how genes are regulated or how they respond to metabolic cues. Given the central role of LSECs in regulating metabolic homeostasis, it is possible that LSECs are also key players in liver disease genetic susceptibility. The assignment of such non-coding variants to their target genes and thus to affected pathways remains challenging, but it is increasingly obvious that it is one of the next steps in identifying disease-effector genes for common pathologies. Thus, the combined analysis of LSEC transcriptomic and epigenomics datasets should be considered in the future investigation of genetic factors of liver disease.

Conclusions

Healthy, functioning LSECs play a key role in protecting the liver from inflammation and fibrosis, and even contribute to regeneration of the liver in early stages of fibrosis or in hepatectomy. In this review, we collected the available evidence for the roles of six TFs, which have been implicated in initiating and driving LSEC development and in the maintenance of liver metabolic homeostasis. Given the many challenges of working with primary ECs, we propose that these LSEC TFs should be harnessed to develop better LSEC in vitro modelling systems, along with other currently known LSEC marker genes (Fig. 2) and secreted factors. There are still significant gaps in our understanding of LSEC TF activity, which we envision will be tackled in future studies, deploying state-of-the-art approaches such as the epigenomic profiling of complex tissue samples with single-cell resolution.

Supplementary materials

This is linked to the online version of the paper at https://doi.org/10.1530/JME-23-0026.

Declaration of interest

I.C.’s partner holds stock options in Ochre Bio. The other authors declare no conflict of interest.

Funding

DN is recipient of a British Heart Foundation 4-year PhD studentship (FS/4yPhD/F/20/34128). IC is recipient of a Sir Henry Dale Fellowship jointly funded by the Wellcome Trust and the Royal Society (224662/Z/21/Z) and a Research Grant by the Royal Society (RGS\R1\221086). GMB and AMR are supported by the British Heart Foundation (PG/20/16/35047; RG/17/4/32662). The authors are also supported by the National Institute for Health Research (NIHR) Imperial Biomedical Research Centre.

References

  • Aizarani N, Saviano A, Mailly L, Durand S, Herman JS, Pessaux P, Baumert TF & & Grün D 2019 A human liver cell atlas reveals heterogeneity and epithelial progenitors. Nature 572 199–204. (https://doi.org/10.1038/s41586-019-1373-2)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Andrews TS, Atif J, Liu JC, Perciani CT, Ma XZ, Thoeni C, Slyper M, Eraslan G, Segerstolpe A, Manuel J, et al.2022 Single-cell, single-nucleus, and spatial RNA sequencing of the human liver identifies cholangiocyte and mesenchymal heterogeneity. Hepatology Communications 6 821–840. (https://doi.org/10.1002/hep4.1854)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Anstee QM, Darlay R, Cockell S, Meroni M, Govaere O, Tiniakos D, Burt AD, Bedossa P, Palmer J, Liu YL, et al.2020 Genome-wide association study of non-alcoholic fatty liver and steatohepatitis in a histologically characterised cohort. Journal of Hepatology 73 505–515. (https://doi.org/10.1016/j.jhep.2020.04.003)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Asahina K, Zhou B, Pu WT & & Tsukamoto H 2011 Septum transversum-derived mesothelium gives rise to hepatic stellate cells and perivascular mesenchymal cells in developing mouse liver. Hepatology 53 983–995. (https://doi.org/10.1002/hep.24119)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bartosovic M, Kabbe M & & Castelo-Branco G 2021 Single-cell CUT&Tag profiles histone modifications and transcription factors in complex tissues. Nature Biotechnology 39 825–835. (https://doi.org/10.1038/s41587-021-00869-9)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bataller R & & Brenner DA 2005 Liver fibrosis. Journal of Clinical Investigation 115 209–218. (https://doi.org/10.1172/JCI24282)

  • Batta K, Florkowska M, Kouskoff V & & Lacaud G 2014 Direct reprogramming of murine fibroblasts to hematopoietic progenitor cells. Cell Reports 9 1871–1884. (https://doi.org/10.1016/j.celrep.2014.11.002)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Benedict M & & Zhang X 2017 Non-alcoholic fatty liver disease: an expanded review. World Journal of Hepatology 9 715–732. (https://doi.org/10.4254/wjh.v9.i16.715)

  • Ben-Moshe S & & Itzkovitz S 2019 Spatial heterogeneity in the mammalian liver. Nature Reviews. Gastroenterology and Hepatology 16 395–410. (https://doi.org/10.1038/s41575-019-0134-x)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Birdsey GM, Shah AV, Dufton N, Reynolds LE, Osuna Almagro L, Yang Y, Aspalter IM, Khan ST, Mason JC, Dejana E, et al.2015 The endothelial transcription factor ERG promotes vascular stability and growth through Wnt/β-catenin signaling. Developmental Cell 32 82–96. (https://doi.org/10.1016/j.devcel.2014.11.016)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bissig-Choisat B, Alves-Bezerra M, Zorman B, Ochsner SA, Barzi M, Legras X, Yang D, Borowiak M, Dean AM, York RB, et al.2021 A human liver chimeric mouse model for non-alcoholic fatty liver disease. JHEP Reports 3 100281. (https://doi.org/10.1016/j.jhepr.2021.100281)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Blouin A, Bolender RP & & Weibel ER 1977 Distribution of organelles and membranes between hepatocytes and nonhepatocytes in the rat liver parenchyma. A stereological study. Journal of Cell Biology 72 441–455. (https://doi.org/10.1083/jcb.72.2.441)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bossard P & & Zaret KS 1998 GATA transcription factors as potentiators of gut endoderm differentiation. Development 125 4909–4917. (https://doi.org/10.1242/dev.125.24.4909)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Braet F & & Wisse E 2002 Structural and functional aspects of liver sinusoidal endothelial cell fenestrae: a review. Comparative Hepatology 1 1. (https://doi.org/10.1186/1476-5926-1-1)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Brosch M, Kattler K, Herrmann A, Von Schönfels W, Nordström K, Seehofer D, Damm G, Becker T, Zeissig S, Nehring S, et al.2018 Epigenomic map of human liver reveals principles of zonated morphogenic and metabolic control. Nature Communications 9 4150. (https://doi.org/10.1038/s41467-018-06611-5)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Burgdorf S, Kautz A, Böhnert V, Knolle PA & & Kurts C 2007 Distinct pathways of antigen uptake and intracellular routing in CD4 and CD8 T cell activation. Science 316 612–616. (https://doi.org/10.1126/science.1137971)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bürglin TR 1997 Analysis of TALE superclass homeobox genes (MEIS, PBC, KNOX, Iroquois, TGIF) reveals a novel domain conserved between plants and animals. Nucleic Acids Research 25 4173–4180. (https://doi.org/10.1093/nar/25.21.4173)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Cai M, Langer EM, Gill JG, Satpathy AT, Albring JC, Kc W, Murphy TL & & Murphy KM 2012 Dual actions of Meis1 inhibit erythroid progenitor development and sustain general hematopoietic cell proliferation. Blood 120 335–346. (https://doi.org/10.1182/blood-2012-01-403139)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Castro-Mondragon JA, Riudavets-Puig R, Rauluseviciute I, Berhanu Lemma RB, Turchi L, Blanc-Mathieu R, Lucas J, Boddie P, Khan A, Manosalva Pérez N, et al.2022 JASPAR 2022: the 9th release of the open-access database of transcription factor binding profiles. Nucleic Acids Research 50 D165–D173. (https://doi.org/10.1093/nar/gkab1113)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Cebola I 2020 Liver gene regulatory networks: contributing factors to nonalcoholic fatty liver disease. Wiley Interdisciplinary Reviews. Systems Biology and Medicine 12 e1480. (https://doi.org/10.1002/wsbm.1480)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Chalasani N, Wilson L, Kleiner DE, Cummings OW, Brunt EM, Unalp A & NASH Clinical Research Network 2008 Relationship of steatosis grade and zonal location to histological features of steatohepatitis in adult patients with non-alcoholic fatty liver disease. Journal of Hepatology 48 829–834. (https://doi.org/10.1016/j.jhep.2008.01.016)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Cusanovich DA, Daza R, Adey A, Pliner HA, Christiansen L, Gunderson KL, Steemers FJ, Trapnell C & & Shendure J 2015 Multiplex single-cell profiling of chromatin accessibility by combinatorial cellular indexing. Science 348 910–914. (https://doi.org/10.1126/science.aab1601)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Danoy M, Poulain S, Koui Y, Tauran Y, Scheidecker B, Kido T, Miyajima A, Sakai Y, Plessy C & & Leclerc E 2020 Transcriptome profiling of hiPSC-derived LSECs with nanoCAGE. Molecular Omics 16 138–146. (https://doi.org/10.1039/c9mo00135b)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • David L, Mallet C, Keramidas M, Lamandé N, Gasc JM, Dupuis-Girod S, Plauchu H, Feige JJ & & Bailly S 2008 Bone morphogenetic Protein-9 is a circulating vascular quiescence factor. Circulation Research 102 914–922. (https://doi.org/10.1161/CIRCRESAHA.107.165530)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • de Haan W, Øie C, Benkheil M, Dheedene W, Vinckier S, Coppiello G, Aranguren XL, Beerens M, Jaekers J, Topal B, et al.2020 Unraveling the transcriptional determinants of liver sinusoidal endothelial cell specialization. American Journal of Physiology: Gastrointestinal and Liver Physiology 318 G803–G815. (https://doi.org/10.1152/ajpgi.00215.2019)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • de Haan W, Dheedene W, Apelt K, Décombas-Deschamps S, Vinckier S, Verhulst S, Conidi A, Deffieux T, Staring MW, Vandervoort P, et al.2022 Endothelial Zeb2 preserves the hepatic angioarchitecture and protects against liver fibrosis. Cardiovascular Research 118 1262–1275. (https://doi.org/10.1093/cvr/cvab148)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • De Smedt J, Van Os EA, Talon I, Ghosh S, Toprakhisar B, Furtado Madeiro Da Costa R, Zaunz S, Vazquez MA, Boon R, Baatsen P, et al.2021 PU.1 drives specification of pluripotent stem cell-derived endothelial cells to LSEC-like cells. Cell Death and Disease 12 84. (https://doi.org/10.1038/s41419-020-03356-2)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Delire B, Stärkel P & & Leclercq I 2015 Animal models for fibrotic liver diseases: what we have, what we need, and what is under development. Journal of Clinical and Translational Hepatology 3 53–66. (https://doi.org/10.14218/JCTH.2014.00035)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Desroches-Castan A, Tillet E, Ricard N, Ouarné M, Mallet C, Belmudes L, Couté Y, Boillot O, Scoazec JY, Bailly S, et al.2019 Bone morphogenetic Protein 9 is a paracrine factor controlling liver sinusoidal endothelial cell fenestration and protecting against hepatic fibrosis. Hepatology 70 1392–1408. (https://doi.org/10.1002/hep.30655)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Ding BS, Cao Z, Lis R, Nolan DJ, Guo P, Simons M, Penfold ME, Shido K, Rabbany SY & & Rafii S 2014 Divergent angiocrine signals from vascular niche balance liver regeneration and fibrosis. Nature 505 97–102. (https://doi.org/10.1038/nature12681)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Dufton NP, Peghaire CR, Osuna-Almagro L, Raimondi C, Kalna V, Chauhan A, Webb G, Yang Y, Birdsey GM, Lalor P, et al.2017 Dynamic regulation of canonical TGFβ signalling by endothelial transcription factor ERG protects from liver fibrogenesis. Nature Communications 8 895. (https://doi.org/10.1038/s41467-017-01169-0)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Ekstedt M, Franzén LE, Mathiesen UL, Thorelius L, Holmqvist M, Bodemar G & & Kechagias S 2006 Long-term follow-up of patients with NAFLD and elevated liver enzymes. Hepatology 44 865–873. (https://doi.org/10.1002/hep.21327)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Elvevold K, Simon-Santamaria J, Hasvold H, Mccourt P, Smedsrød B & & Sørensen KK 2008 Liver sinusoidal endothelial cells depend on mannose receptor-mediated recruitment of lysosomal enzymes for normal degradation capacity. Hepatology 48 2007–2015. (https://doi.org/10.1002/hep.22527)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Eslam M, Sanyal AJ & & George J 2020 MAFLD: a consensus-driven proposed nomenclature for metabolic associated fatty liver disease. Gastroenterology 158 1999-2014.e1. (https://doi.org/10.1053/j.gastro.2019.11.312)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Fardi M, Alivand M, Baradaran B, Farshdousti Hagh M & & Solali S 2019 The crucial role of ZEB2: from development to epithelial-to-mesenchymal transition and cancer complexity. Journal of Cellular Physiology 234 14783–14799. (https://doi.org/10.1002/jcp.28277)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Fernandez M 2015 Molecular pathophysiology of portal hypertension. Hepatology 61 1406–1415. (https://doi.org/10.1002/hep.27343)

  • Filtz TM, Vogel WK & & Leid M 2014 Regulation of transcription factor activity by interconnected post-translational modifications. Trends in Pharmacological Sciences 35 76–85. (https://doi.org/10.1016/j.tips.2013.11.005)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Friedman SL, Neuschwander-Tetri BA, Rinella M & & Sanyal AJ 2018 Mechanisms of NAFLD development and therapeutic strategies. Nature Medicine 24 908–922. (https://doi.org/10.1038/s41591-018-0104-9)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Gage BK, Liu JC, Innes BT, MacParland SA, Mcgilvray ID, Bader GD & & Keller GM 2020 Generation of functional liver sinusoidal endothelial cells from human pluripotent stem-cell-derived venous angioblasts. Cell Stem Cell 27 254–269.e9. (https://doi.org/10.1016/j.stem.2020.06.007)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Ganesan LP, Kim J, Wu Y, Mohanty S, Phillips GS, Birmingham DJ, Robinson JM & & Anderson CL 2012 FcγRIIb on liver sinusoidal endothelium clears small immune complexes. Journal of Immunology 189 4981–4988. (https://doi.org/10.4049/jimmunol.1202017)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Gebhardt R, Baldysiak-Figiel A, Krügel V, Ueberham E & & Gaunitz F 2007 Hepatocellular expression of glutamine synthetase: an indicator of morphogen actions as master regulators of zonation in adult liver. Progress in Histochemistry and Cytochemistry 41 201–266. (https://doi.org/10.1016/j.proghi.2006.12.001)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Géraud C, Schledzewski K, Demory A, Klein D, Kaus M, Peyre F, Sticht C, Evdokimov K, Lu S, Schmieder A, et al.2010 Liver sinusoidal endothelium: a microenvironment-dependent differentiation program in rat including the novel junctional protein liver endothelial differentiation-associated protein-1. Hepatology 52 313–326. (https://doi.org/10.1002/hep.23618)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Géraud C, Koch PS, Zierow J, Klapproth K, Busch K, Olsavszky V, Leibing T, Demory A, Ulbrich F, Diett M, et al.2017 GATA4-dependent organ-specific endothelial differentiation controls liver development and embryonic hematopoiesis. Journal of Clinical Investigation 127 1099–1114. (https://doi.org/10.1172/JCI90086)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Ginsberg M, James D, Ding BS, Nolan D, Geng F, Butler JM, Schachterle W, Pulijaal VR, Mathew S, Chasen ST, et al.2012 Efficient direct reprogramming of mature amniotic cells into endothelial cells by ETS factors and TGFβ suppression. Cell 151 559–575. (https://doi.org/10.1016/j.cell.2012.09.032)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Gómez-Salinero JM, Itkin T, Houghton S, Badwe C, Lin Y, Kalna V, Dufton N, Peghaire CR, Yokoyama M, Wingo M, et al.2022a Cooperative ETS transcription factors enforce adult endothelial cell fate and cardiovascular homeostasis. Nature Cardiovascular Research 1 882–899. (https://doi.org/10.1038/s44161-022-00128-3)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Gómez-Salinero JM, Izzo F, Lin Y, Houghton S, Itkin T, Geng F, Bram Y, Adelson RP, Lu TM, Inghirami G, et al.2022b Specification of fetal liver endothelial progenitors to functional zonated adult sinusoids requires c-Maf induction. Cell Stem Cell 29 593-609.e7. (https://doi.org/10.1016/j.stem.2022.03.002)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Gong W, Das S, Sierra-Pagan JE, Skie E, Dsouza N, Larson TA, Garry MG, Luzete-Monteiro E, Zaret KS & & Garry DJ 2022 ETV2 functions as a pioneer factor to regulate and reprogram the endothelial lineage. Nature Cell Biology 24 672–684. (https://doi.org/10.1038/s41556-022-00901-3)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Grossman SR, Engreitz J, Ray JP, Nguyen TH, Hacohen N & & Lander ES 2018 Positional specificity of different transcription factor classes within enhancers. Proceedings of the National Academy of Sciences of the United States of America 115 E7222–E7230. (https://doi.org/10.1073/pnas.1804663115)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Gualdi R, Bossard P, Zheng M, Hamada Y, Coleman JR & & Zaret KS 1996 Hepatic specification of the gut endoderm in vitro: cell signaling and transcriptional control. Genes and Development 10 1670–1682. (https://doi.org/10.1101/gad.10.13.1670)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Guilliams M, Bonnardel J, Haest B, Vanderborght B, Wagner C, Remmerie A, Bujko A, Martens L, Thoné T, Browaeys R, et al.2022 Spatial proteogenomics reveals distinct and evolutionarily conserved hepatic macrophage niches. Cell 185 379–396.e38. (https://doi.org/10.1016/j.cell.2021.12.018)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Hammoutene A & & Rautou PE 2019 Role of liver sinusoidal endothelial cells in non-alcoholic fatty liver disease. Journal of Hepatology 70 1278–1291. (https://doi.org/10.1016/j.jhep.2019.02.012)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Hernandez-Gea V & & Friedman SL 2011 Pathogenesis of liver fibrosis. Annual Review of Pathology: Mechanisms of Disease 6 425–456. (https://doi.org/10.1146/annurev-pathol-011110-130246)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Hnisz D, Abraham BJ, Lee TI, Lau A, Saint-André V, Sigova AA, Hoke HA & & Young RA 2013 Super-enhancers in the control of cell identity and disease. Cell 155 934–947. (https://doi.org/10.1016/j.cell.2013.09.053)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Holland CH, Ramirez Flores RO, Myllys M, Hassan R, Edlund K, Hofmann U, Marchan R, Cadenas C, Reinders J, Hoehme S, et al.2022 Transcriptomic cross-species analysis of chronic liver disease reveals consistent regulation between humans and mice. Hepatology Communications 6 161–177. (https://doi.org/10.1002/hep4.1797)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Hu J, Srivastava K, Wieland M, Runge A, Mogler C, Besemfelder E, Terhardt D, Vogel MJ, Cao L, Korn C, et al.2014 Endothelial cell-derived angiopoietin-2 controls liver regeneration as a spatiotemporal rheostat. Science 343 416–419. (https://doi.org/10.1126/science.1244880)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Hu S, Liu S, Bian Y, Poddar M, Singh S, Cao C, Mcgaughey J, Bell A, Blazer LL, Adams JJ, et al.2022 Single-cell spatial transcriptomics reveals a dynamic control of metabolic zonation and liver regeneration by endothelial cell Wnt2 and Wnt9b. Cell Reports. Medicine 3 100754. (https://doi.org/10.1016/j.xcrm.2022.100754)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Im YR, Hunter H, De Gracia Hahn D, Duret A, Cheah Q, Dong J, Fairey M, Hjalmarsson C, Li A, Lim HK, et al.2021 A systematic review of animal models of NAFLD finds high-fat, high-fructose diets most closely resemble human NAFLD. Hepatology 74 1884–1901. (https://doi.org/10.1002/hep.31897)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Imbratta C, Hussein H, Andris F & & Verdeil G 2020 c-MAF, a Swiss Army Knife for Tolerance in lymphocytes. Frontiers in Immunology 11 206. (https://doi.org/10.3389/fimmu.2020.00206)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Inverso D, Shi J, Lee KH, Jakab M, Ben-Moshe S, Kulkarni SR, Schneider M, Wang G, Komeili M, Vélez PA, et al.2021 A spatial vascular transcriptomic, proteomic, and phosphoproteomic atlas unveils an angiocrine Tie-Wnt signaling axis in the liver. Developmental Cell 56 1677–1693.e10. (https://doi.org/10.1016/j.devcel.2021.05.001)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Jolma A, Yin Y, Nitta KR, Dave K, Popov A, Taipale M, Enge M, Kivioja T, Morgunova E & & Taipale J 2015 DNA-dependent formation of transcription factor pairs alters their binding specificity. Nature 527 384–388. (https://doi.org/10.1038/nature15518)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Kalna V, Yang Y, Peghaire CR, Frudd K, Hannah R, Shah AV, Osuna Almagro L, Boyle JJ, Göttgens B, Ferrer J, et al.2019 The transcription factor ERG regulates super-enhancers associated with an endothelial-specific gene expression program. Circulation Research 124 1337–1349. (https://doi.org/10.1161/CIRCRESAHA.118.313788)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Kus E, Kaczara P, Czyzynska-Cichon I, Szafranska K, Zapotoczny B, Kij A, Sowinska A, Kotlinowski J, Mateuszuk L, Czarnowska E, et al.2019 LSEC fenestrae are preserved despite pro-inflammatory phenotype of liver sinusoidal endothelial cells in mice on high fat diet. Frontiers in Physiology 10 6. (https://doi.org/10.3389/fphys.2019.00006)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Lafoz E, Ruart M, Anton A, Oncins A & & Hernández-Gea V 2020 The endothelium as a driver of liver fibrosis and regeneration. Cells 9 929. (https://doi.org/10.3390/cells9040929)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Lambert SA, Jolma A, Campitelli LF, Das PK, Yin Y, Albu M, Chen X, Taipale J, Hughes TR & & Weirauch MT 2018 The human transcription factors. Cell 172 650–665. (https://doi.org/10.1016/j.cell.2018.01.029)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Leibing T, Géraud C, Augustin I, Boutros M, Augustin HG, Okun JG, Langhans CD, Zierow J, Wohlfeil SA, Olsavszky V, et al.2018 Angiocrine Wnt signaling controls liver growth and metabolic maturation in mice. Hepatology 68 707–722. (https://doi.org/10.1002/hep.29613)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Li G, Hao W & & Hu W 2020 Transcription factor PU.1 and immune cell differentiation (Review) (Review). International Journal of Molecular Medicine 46 1943–1950. (https://doi.org/10.3892/ijmm.2020.4763)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Liang Y, Kaneko K, Xin B, Lee J, Sun X, Zhang K & & Feng GS 2022 Temporal analyses of postnatal liver development and maturation by single-cell transcriptomics. Developmental Cell 57 398–414.e5. (https://doi.org/10.1016/j.devcel.2022.01.004)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Limmer A, Ohl J, Kurts C, Ljunggren HG, Reiss Y, Groettrup M, Momburg F, Arnold B & & Knolle PA 2000 Efficient presentation of exogenous antigen by liver endothelial cells to CD8+ T cells results in antigen-specific T-cell tolerance. Nature Medicine 6 1348–1354. (https://doi.org/10.1038/82161)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Ma Q & & Xu D 2022 Deep learning shapes single-cell data analysis. Nature Reviews. Molecular Cell Biology 23 303–304. (https://doi.org/10.1038/s41580-022-00466-x)

  • Machon O, Masek J, Machonova O, Krauss S & & Kozmik Z 2015 Meis2 is essential for cranial and cardiac neural crest development. BMC Developmental Biology 15 40. (https://doi.org/10.1186/s12861-015-0093-6)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • MacParland SA, Liu JC, Ma X-Z, Innes BT, Bartczak AM, Gage BK, Manuel J, Khuu N, Echeverri J, Linares I, Gupta R, et al.2018 Single cell RNA sequencing of human liver reveals distinct intrahepatic macrophage populations. Nature Communications 9 4383. (https://doi.org/10.1038/s41467-018-06318-7)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Manicardi N, Fernández-Iglesias A, Abad-Jordà L, Royo F, Azkargorta M, Ortega-Ribera M, Sanfeliu-Redondo D, Martínez-Alcocer A, Elortza F, Hessheimer AJ, et al.2021 Transcriptomic profiling of the liver sinusoidal endothelium during cirrhosis reveals stage-specific secretory signature. Cancers 13 2688. (https://doi.org/10.3390/cancers13112688)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Maurano MT, Humbert R, Rynes E, Thurman RE, Haugen E, Wang H, Reynolds AP, Sandstrom R, Qu H, Brody J, et al.2012 Systematic localization of common disease-associated variation in regulatory DNA. Science 337 1190–1195. (https://doi.org/10.1126/science.1222794)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Miyao M, Kotani H, Ishida T, Kawai C, Manabe S, Abiru H & & Tamaki K 2015 Pivotal role of liver sinusoidal endothelial cells in NAFLD/NASH progression. Laboratory Investigation 95 1130–1144. (https://doi.org/10.1038/labinvest.2015.95)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Moens CB & & Selleri L 2006 Hox cofactors in vertebrate development. Developmental Biology 291 193–206. (https://doi.org/10.1016/j.ydbio.2005.10.032)

  • Morita R, Suzuki M, Kasahara H, Shimizu N, Shichita T, Sekiya T, Kimura A, Sasaki K, Yasukawa H & & Yoshimura A 2015 ETS transcription factor ETV2 directly converts human fibroblasts into functional endothelial cells. Proceedings of the National Academy of Sciences of the United States of America 112 160–165. (https://doi.org/10.1073/pnas.1413234112)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Namjou B, Lingren T, Huang Y, Parameswaran S, Cobb BL, Stanaway IB, Connolly JJ, Mentch FD, Benoit B, Niu X, et al.2019 GWAS and enrichment analyses of non-alcoholic fatty liver disease identify new trait-associated genes and pathways across eMERGE Network. BMC Medicine 17 135. (https://doi.org/10.1186/s12916-019-1364-z)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Panigrahi A & & O’Malley BW 2021 Mechanisms of enhancer action: the known and the unknown. Genome Biology 22 108. (https://doi.org/10.1186/s13059-021-02322-1)

  • Paris J & & Henderson NC 2022 Liver zonation, revisited. Hepatology 76 1219–1230. (https://doi.org/10.1002/hep.32408)

  • Parker ME, Barrera A, Wheaton JD, Zuberbuehler MK, Allan DSJ, Carlyle JR, Reddy TE & & Ciofani M 2020 c-Maf regulates the plasticity of group 3 innate lymphoid cells by restraining the type 1 program. Journal of Experimental Medicine 217. (https://doi.org/10.1084/jem.20191030)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Peghaire C, Dufton NP, Lang M, Salles-Crawley II, Ahnström J, Kalna V, Raimondi C, Pericleous C, Inuabasi L, Kiseleva R, et al.2019 The transcription factor ERG regulates a low shear stress-induced anti-thrombotic pathway in the microvasculature. Nature Communications 10 5014. (https://doi.org/10.1038/s41467-019-12897-w)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Pham TH, Minderjahn J, Schmidl C, Hoffmeister H, Schmidhofer S, Chen W, Längst G, Benner C & & Rehli M 2013 Mechanisms of in vivo binding site selection of the hematopoietic master transcription factor PU.1. Nucleic Acids Research 41 6391–6402. (https://doi.org/10.1093/nar/gkt355)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Poisson J, Lemoinne S, Boulanger C, Durand F, Moreau R, Valla D & & Rautou PE 2017 Liver sinusoidal endothelial cells: physiology and role in liver diseases. Journal of Hepatology 66 212–227. (https://doi.org/10.1016/j.jhep.2016.07.009)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Postigo AA, Depp JL, Taylor JJ & & Kroll KL 2003 Regulation of Smad signaling through a differential recruitment of coactivators and corepressors by ZEB proteins. EMBO Journal 22 2453–2462. (https://doi.org/10.1093/emboj/cdg226)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Preziosi M, Okabe H, Poddar M, Singh S & & Monga SP 2018 Endothelial Wnts regulate β-catenin signaling in murine liver zonation and regeneration: a sequel to the Wnt–Wnt situation. Hepatology Communications 2 845–860. (https://doi.org/10.1002/hep4.1196)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Ramachandran P, Dobie R, Wilson-Kanamori JR, Dora EF, Henderson BEP, Luu NT, Portman JR, Matchett KP, Brice M, Marwick JA, et al.2019 Resolving the fibrotic niche of human liver cirrhosis at single-cell level. Nature 575 512–518. (https://doi.org/10.1038/s41586-019-1631-3)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Rezvani M, Vallier L & & Guillot A 2023 Modeling non-alcoholic fatty liver disease in the dish using human-specific platforms – strategies and limitations. Cellular and Molecular Gastroenterology and Hepatology 15 1135–1145. (https://doi.org/10.1016/j.jcmgh.2023.01.014)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Ruan B, Duan JL, Xu H, Tao KS, Han H, Dou GR & & Wang L 2021 Capillarized liver sinusoidal endothelial cells undergo partial endothelial-mesenchymal transition to actively deposit sinusoidal ECM in liver fibrosis. Frontiers in Cell and Developmental Biology 9 671081. (https://doi.org/10.3389/fcell.2021.671081)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Saviano A, Henderson NC & & Baumert TF 2020 Single-cell genomics and spatial transcriptomics: discovery of novel cell states and cellular interactions in liver physiology and disease biology. Journal of Hepatology 73 1219–1230. (https://doi.org/10.1016/j.jhep.2020.06.004)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Schledzewski K, Géraud C, Arnold B, Wang S, Gröne HJ, Kempf T, Wollert KC, Straub BK, Schirmacher P, Demory A, et al.2011 Deficiency of liver sinusoidal scavenger receptors stabilin-1 and -2 in mice causes glomerulofibrotic nephropathy via impaired hepatic clearance of noxious blood factors. Journal of Clinical Investigation 121 703–714. (https://doi.org/10.1172/JCI44740)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Shah V, Haddad FG, Garcia-Cardena G, Frangos JA, Mennone A, Groszmann RJ & & Sessa WC 1997 Liver sinusoidal endothelial cells are responsible for nitric oxide modulation of resistance in the hepatic sinusoids. Journal of Clinical Investigation 100 2923–2930. (https://doi.org/10.1172/JCI119842)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Shah AV, Birdsey GM & & Randi AM 2016 Regulation of endothelial homeostasis, vascular development and angiogenesis by the transcription factor ERG. Vascular Pharmacology 86 3–13. (https://doi.org/10.1016/j.vph.2016.05.003)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Singh S, Allen AM, Wang Z, Prokop LJ, Murad MH & & Loomba R 2015 Fibrosis progression in nonalcoholic fatty liver vs nonalcoholic steatohepatitis: a systematic review and meta-analysis of paired-biopsy studies. Clinical Gastroenterology and Hepatology 13 643–54.e1 .e1-9. (https://doi.org/10.1016/j.cgh.2014.04.014)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Sookoian S, Castaño GO, Scian R, Fernández Gianotti T, Dopazo H, Rohr C, Gaj G, San Martino J, Sevic I, Flichman D, et al.2016 Serum aminotransferases in nonalcoholic fatty liver disease are a signature of liver metabolic perturbations at the amino acid and Krebs cycle level. American Journal of Clinical Nutrition 103 422–434. (https://doi.org/10.3945/ajcn.115.118695)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Sørensen KK, Mccourt P, Berg T, Crossley C, Couteur DL, Wake K & & Smedsrød B 2012 The scavenger endothelial cell: a new player in homeostasis and immunity. American Journal of Physiology. Regulatory, Integrative and Comparative Physiology 303 R1217–R1230. (https://doi.org/10.1152/ajpregu.00686.2011)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Sørensen KK, Simon-Santamaria J, Mccuskey RS & & Smedsrød B 2015 Liver sinusoidal endothelial cells. Comprehensive Physiology 5 1751–1774. (https://doi.org/10.1002/cphy.c140078)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Steensels S, Qiao J & & Ersoy BA 2020 Transcriptional regulation in non-alcoholic fatty liver disease. Metabolites 10. (https://doi.org/10.3390/metabo10070283)

  • Steinman JB, Salomao MA & & Pajvani UB 2021 Zonation in NASH – A key paradigm for understanding pathophysiology and clinical outcomes. Liver International 41 2534–2546. (https://doi.org/10.1111/liv.15025)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Strauss O, Phillips A, Ruggiero K, Bartlett A & & Dunbar PR 2017 Immunofluorescence identifies distinct subsets of endothelial cells in the human liver. Scientific Reports 7 44356. (https://doi.org/10.1038/srep44356)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Su T, Yang Y, Lai S, Jeong J, Jung Y, Mcconnell M, Utsumi T & & Iwakiri Y 2021 Single-cell transcriptomics reveals zone-specific alterations of liver sinusoidal endothelial cells in cirrhosis. Cellular and Molecular Gastroenterology and Hepatology 11 1139–1161. (https://doi.org/10.1016/j.jcmgh.2020.12.007)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Takahashi Y & & Fukusato T 2014 Histopathology of nonalcoholic fatty liver disease/nonalcoholic steatohepatitis. World Journal of Gastroenterology 20 15539–15548. (https://doi.org/10.3748/wjg.v20.i42.15539)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Tremblay M, Sanchez-Ferras O & & Bouchard M 2018 GATA transcription factors in development and disease. Development 145. (https://doi.org/10.1242/dev.164384)

  • Tsuchida T, Lee YA, Fujiwara N, Ybanez M, Allen B, Martins S, Fiel MI, Goossens N, Chou HI, Hoshida Y, et al.2018 A simple diet- and chemical-induced murine NASH model with rapid progression of steatohepatitis, fibrosis and liver cancer. Journal of Hepatology 69 385–395. (https://doi.org/10.1016/j.jhep.2018.03.011)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Vijayaraj P, Le Bras A, Mitchell N, Kondo M, Juliao S, Wasserman M, Beeler D, Spokes K, Aird WC, Baldwin HS, et al.2012 Erg is a crucial regulator of endocardial-mesenchymal transformation during cardiac valve morphogenesis. Development 139 3973–3985. (https://doi.org/10.1242/dev.081596)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Vlaeminck-Guillem V, Carrere S, Dewitte F, Stehelin D, Desbiens X & & Duterque-Coquillaud M 2000 The Ets family member Erg gene is expressed in mesodermal tissues and neural crests at fundamental steps during mouse embryogenesis. Mechanisms of Development 91 331–335. (https://doi.org/10.1016/s0925-4773(9900272-5)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Watt AJ, Zhao R, Li J & & Duncan SA 2007 Development of the mammalian liver and ventral pancreas is dependent on GATA4. BMC Developmental Biology 7 37. (https://doi.org/10.1186/1471-213X-7-37)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Wilkinson AC, Nakauchi H & & Göttgens B 2017 Mammalian transcription factor networks: recent advances in interrogating biological complexity. Cell Systems 5 319–331. (https://doi.org/10.1016/j.cels.2017.07.004)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Winkler M, Staniczek T, Kürschner SW, Schmid CD, Schönhaber H, Cordero J, Kessler L, Mathes A, Sticht C, Nesslin M, et al.2021 Endothelial GATA4 controls liver fibrosis and regeneration by preventing a pathogenic switch in angiocrine signaling. Journal of Hepatology 74 380–393. (https://doi.org/10.1016/j.jhep.2020.08.033)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Wisse E 1970 An electron microscopic study of the fenestrated endothelial lining of rat liver sinusoids. Journal of Ultrastructure Research 31 125–150. (https://doi.org/10.1016/s0022-5320(7090150-4)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Xiong X, Kuang H, Ansari S, Liu T, Gong J, Wang S, Zhao XY, Ji Y, Li C, Guo L, et al.2019 Landscape of intercellular crosstalk in healthy and NASH liver revealed by single-cell secretome gene analysis. Molecular Cell 75 644–660.e5. (https://doi.org/10.1016/j.molcel.2019.07.028)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Younossi Z, Anstee QM, Marietti M, Hardy T, Henry L, Eslam M, George J & & Bugianesi E 2018 Global burden of NAFLD and NASH: trends, predictions, risk factors and prevention. Nature Reviews. Gastroenterology and Hepatology 15 11–20. (https://doi.org/10.1038/nrgastro.2017.109)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Younossi ZM, Golabi P, Paik JM, Henry A, Van Dongen C & & Henry L 2023 The global epidemiology of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH): a systematic review. Hepatology 77 1335–1347. (https://doi.org/10.1097/HEP.0000000000000004)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Zabet NR & & Adryan B 2015 Estimating binding properties of transcription factors from genome-wide binding profiles. Nucleic Acids Research 43 84–94. (https://doi.org/10.1093/nar/gku1269)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Zaret KS 2020 Pioneer transcription factors initiating gene network changes. Annual Review of Genetics 54 367–385. (https://doi.org/10.1146/annurev-genet-030220-015007)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Zaugg JB, Sahlén P, Andersson R, Alberich-Jorda M, De Laat W, Deplancke B, Ferrer J, Mandrup S, Natoli G, Plewczynski D, et al.2022 Current challenges in understanding the role of enhancers in disease. Nature Structural and Molecular Biology 29 1148–1158. (https://doi.org/10.1038/s41594-022-00896-3)

    • PubMed
    • Search Google Scholar
    • Export Citation

Journal of Molecular Endocrinology is committed to supporting researchers in demonstrating the impact of their articles published in the journal.

The two types of article metrics we measure are (i) more traditional full-text views and pdf downloads, and (ii) Altmetric data, which shows the wider impact of articles in a range of non-traditional sources, such as social media.

More information is on the Reasons to publish page.

Sept 2018 onwards Past Year Past 30 Days
Full Text Views 2450 2450 124
PDF Downloads 1136 1136 108

 

  • Collapse
  • Expand
Online ISSN:
1479-6813
Print ISSN:
0952-5041
Copyright:
© the author(s) 2023
Article ID:
e230026
Article Type:
Review Article
Received Date:
04 May 2023
Accepted Date:
12 Jun 2023
Print Publication Date:
01 Aug 2023
Online Publication Date:
18 Jul 2023
Keywords:
liver; endothelial cell; transcription factors; NAFLD; gene transcription
  • Figure 1
    View in gallery
    Figure 1

    Position, phenotype, and marker fingerprint of LSECs in homeostasis and disease. (A) The liver’s structural unit is the liver lobule where the portal triad and the central vein are connected by sinusoids. (B) Liver sinusoids are the scene for the uptake of nutrients derived from the intestines via the portal vein, where there is also an influx of pathogens and signalling molecules from the gut bacteria (top right). Oxygen-enriched blood enters from the portal artery and mixes with the portal vein blood. The mixed blood flows through the sinusoid where it is filtered by the LSECs and macromolecules and nutrients are transported towards the hepatocytes, while waste products from the liver parenchyma are transported into the sinusoidal blood. This mechanism creates a gradient of oxygen, nutrients, and signalling molecules along the periportal–pericentral axis (Box 2). (C) Spatial molecular imaging of the human liver showing zonated expression of the indicated genes (left). Data were acquired using CosMxTM Spatial Molecular Imager and includes staining for a panel of morphological features including hepatocytes, which were stained with CK8 and CK18 fluorescent antibodies, and quantifications for a panel of 1000 genes. Integrative data analysis enabled the detection of six ‘neighbourhood clusters’ which correspond to the zones labelled (right). Data were retrieved from http://nanostring.com/CosMx-dataset. (D) LSECs, through their unique phenotypic and functional properties, play a key role in several aspects of liver homeostasis (see section ‘LSEC phenotypical properties and function in physiological conditions and disease’). Damaged LSECs display several changes in their phenotype and function (bottom right) such as loss of fenestrae, formation of basal lamina, decreased endocytic activity, deposition of extracellular matrix (ECM), and activation of Kupffer cells. These changes affect macromolecule transport towards the parenchyma, microbe and virus uptake, and can lead to sustained inflammation and fibrosis.

  • Figure 2
    View in gallery
    Figure 2

    Marker genes of LSECs identified across eight independent studies employing different experimental methods and/or analysis. The summary of rodent and human datasets presents genes identified by at least four independent studies. The genes are ranked by the number of identifying studies. scRNA-seq studies were considered if they provided marker gene sets for clusters defined as LSECs. All data used for the identification of LSEC clusters are of healthy human/mouse liver. Ramachandran et al. provided a pre-filtered LSEC marker list (n = 80) (Ramachandran et al. 2019). For the remaining scRNA-seq studies, we obtained the top 100 genes ranked by fold-change of LSECs vs other liver cell types (Guilliams et al. 2022). In studies that defined two (MacParland et al. 2018) or three (Aizarani et al. 2019) LSEC clusters, the final list combined the top 100 genes from each of those clusters. Haan et al. identified 27 LSEC-enriched genes by comparing LSECs with heart and brain ECs (de Haan et al. 2020). Géraud et al. provided a 46-gene marker list based on gene expression comparison between EC groups (Géraud et al. 2010). De Smedt et al. developed a computational workflow to identify TFs central in differentiation and specification (CenTFinder). The application of CenTFinder to a series of LSEC gene expression datasets resulted in a list of 80 putative marker genes (De Smedt et al. 2021). On the right, we present scaled mean protein expression for LSECs across four zones (portal node (PN), periportal (PP), pericentral (PC), and central vein (CV)) (n = 4 samples per zone) (Inverso et al. 2021). Genes/proteins not reported in the datasets described are shown with a small empty dot. The full list of marker genes is available in Supplementary Table 1 (see section on supplementary materials given at the end of this article).

  • Figure 3
    View in gallery
    Figure 3

    Summary of key transcription factors of LSEC identity presented in this review. (A, B) Single-cell gene expression levels from the Human Liver Cell Atlas (https://www.livercellatlas.org) (Guilliams et al. 2022). Gene expression data were downloaded and annotated with the provided cell annotation matrix, including cell type and UMAP coordinates. UMAPs and violin plots show normalised expression levels, obtained using SCTransform from the Seurat package (v4.3.0) in R. (C) The gene expression levels across the indicated liver cell types are presented as scaled pseudo-count counts per million (CPM) values from scRNA-seq data, as described earlier (Guilliams et al. 2022). Macrovascular ECs represent the mean expression of central vein and portal vein endothelial cells, as seen in panel (A). The LSEC-specific knockouts of TFs that lead to liver fibrosis in mouse models without additional treatments (e.g. KO of another TF, CCl4 treatment) are listed at ‘LSEC KO leads to fibrosis’. The position weight matrices (PWMs) corresponding to the preferred binding motifs of each TF are shown on the right. Motif logos were retrieved from the Jaspar database (Castro-Mondragon et al. 2021). The ZEB2 PWM is not available from the Jaspar database; therefore, the highly homologous ZEB1 (ZEB2 paralog) motif is listed; however, the PWMs of both motifs are highly concordant (HOMER database, not shown). 1: (Winkler et al. 2021); 2: (Dufton et al. 2017).

Figure 1

Position, phenotype, and marker fingerprint of LSECs in homeostasis and disease. (A) The liver’s structural unit is the liver lobule where the portal triad and the central vein are connected by sinusoids. (B) Liver sinusoids are the scene for the uptake of nutrients derived from the intestines via the portal vein, where there is also an influx of pathogens and signalling molecules from the gut bacteria (top right). Oxygen-enriched blood enters from the portal artery and mixes with the portal vein blood. The mixed blood flows through the sinusoid where it is filtered by the LSECs and macromolecules and nutrients are transported towards the hepatocytes, while waste products from the liver parenchyma are transported into the sinusoidal blood. This mechanism creates a gradient of oxygen, nutrients, and signalling molecules along the periportal–pericentral axis (Box 2). (C) Spatial molecular imaging of the human liver showing zonated expression of the indicated genes (left). Data were acquired using CosMxTM Spatial Molecular Imager and includes staining for a panel of morphological features including hepatocytes, which were stained with CK8 and CK18 fluorescent antibodies, and quantifications for a panel of 1000 genes. Integrative data analysis enabled the detection of six ‘neighbourhood clusters’ which correspond to the zones labelled (right). Data were retrieved from http://nanostring.com/CosMx-dataset. (D) LSECs, through their unique phenotypic and functional properties, play a key role in several aspects of liver homeostasis (see section ‘LSEC phenotypical properties and function in physiological conditions and disease’). Damaged LSECs display several changes in their phenotype and function (bottom right) such as loss of fenestrae, formation of basal lamina, decreased endocytic activity, deposition of extracellular matrix (ECM), and activation of Kupffer cells. These changes affect macromolecule transport towards the parenchyma, microbe and virus uptake, and can lead to sustained inflammation and fibrosis.
Figure 2

Marker genes of LSECs identified across eight independent studies employing different experimental methods and/or analysis. The summary of rodent and human datasets presents genes identified by at least four independent studies. The genes are ranked by the number of identifying studies. scRNA-seq studies were considered if they provided marker gene sets for clusters defined as LSECs. All data used for the identification of LSEC clusters are of healthy human/mouse liver. Ramachandran et al. provided a pre-filtered LSEC marker list (n = 80) (Ramachandran et al. 2019). For the remaining scRNA-seq studies, we obtained the top 100 genes ranked by fold-change of LSECs vs other liver cell types (Guilliams et al. 2022). In studies that defined two (MacParland et al. 2018) or three (Aizarani et al. 2019) LSEC clusters, the final list combined the top 100 genes from each of those clusters. Haan et al. identified 27 LSEC-enriched genes by comparing LSECs with heart and brain ECs (de Haan et al. 2020). Géraud et al. provided a 46-gene marker list based on gene expression comparison between EC groups (Géraud et al. 2010). De Smedt et al. developed a computational workflow to identify TFs central in differentiation and specification (CenTFinder). The application of CenTFinder to a series of LSEC gene expression datasets resulted in a list of 80 putative marker genes (De Smedt et al. 2021). On the right, we present scaled mean protein expression for LSECs across four zones (portal node (PN), periportal (PP), pericentral (PC), and central vein (CV)) (n = 4 samples per zone) (Inverso et al. 2021). Genes/proteins not reported in the datasets described are shown with a small empty dot. The full list of marker genes is available in Supplementary Table 1 (see section on supplementary materials given at the end of this article).
Figure 3

Summary of key transcription factors of LSEC identity presented in this review. (A, B) Single-cell gene expression levels from the Human Liver Cell Atlas (https://www.livercellatlas.org) (Guilliams et al. 2022). Gene expression data were downloaded and annotated with the provided cell annotation matrix, including cell type and UMAP coordinates. UMAPs and violin plots show normalised expression levels, obtained using SCTransform from the Seurat package (v4.3.0) in R. (C) The gene expression levels across the indicated liver cell types are presented as scaled pseudo-count counts per million (CPM) values from scRNA-seq data, as described earlier (Guilliams et al. 2022). Macrovascular ECs represent the mean expression of central vein and portal vein endothelial cells, as seen in panel (A). The LSEC-specific knockouts of TFs that lead to liver fibrosis in mouse models without additional treatments (e.g. KO of another TF, CCl4 treatment) are listed at ‘LSEC KO leads to fibrosis’. The position weight matrices (PWMs) corresponding to the preferred binding motifs of each TF are shown on the right. Motif logos were retrieved from the Jaspar database (Castro-Mondragon et al. 2021). The ZEB2 PWM is not available from the Jaspar database; therefore, the highly homologous ZEB1 (ZEB2 paralog) motif is listed; however, the PWMs of both motifs are highly concordant (HOMER database, not shown). 1: (Winkler et al. 2021); 2: (Dufton et al. 2017).

  • Aizarani N, Saviano A, Mailly L, Durand S, Herman JS, Pessaux P, Baumert TF & & Grün D 2019 A human liver cell atlas reveals heterogeneity and epithelial progenitors. Nature 572 199–204. (https://doi.org/10.1038/s41586-019-1373-2)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Andrews TS, Atif J, Liu JC, Perciani CT, Ma XZ, Thoeni C, Slyper M, Eraslan G, Segerstolpe A, Manuel J, et al.2022 Single-cell, single-nucleus, and spatial RNA sequencing of the human liver identifies cholangiocyte and mesenchymal heterogeneity. Hepatology Communications 6 821–840. (https://doi.org/10.1002/hep4.1854)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Anstee QM, Darlay R, Cockell S, Meroni M, Govaere O, Tiniakos D, Burt AD, Bedossa P, Palmer J, Liu YL, et al.2020 Genome-wide association study of non-alcoholic fatty liver and steatohepatitis in a histologically characterised cohort. Journal of Hepatology 73 505–515. (https://doi.org/10.1016/j.jhep.2020.04.003)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Asahina K, Zhou B, Pu WT & & Tsukamoto H 2011 Septum transversum-derived mesothelium gives rise to hepatic stellate cells and perivascular mesenchymal cells in developing mouse liver. Hepatology 53 983–995. (https://doi.org/10.1002/hep.24119)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bartosovic M, Kabbe M & & Castelo-Branco G 2021 Single-cell CUT&Tag profiles histone modifications and transcription factors in complex tissues. Nature Biotechnology 39 825–835. (https://doi.org/10.1038/s41587-021-00869-9)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bataller R & & Brenner DA 2005 Liver fibrosis. Journal of Clinical Investigation 115 209–218. (https://doi.org/10.1172/JCI24282)

  • Batta K, Florkowska M, Kouskoff V & & Lacaud G 2014 Direct reprogramming of murine fibroblasts to hematopoietic progenitor cells. Cell Reports 9 1871–1884. (https://doi.org/10.1016/j.celrep.2014.11.002)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Benedict M & & Zhang X 2017 Non-alcoholic fatty liver disease: an expanded review. World Journal of Hepatology 9 715–732. (https://doi.org/10.4254/wjh.v9.i16.715)

  • Ben-Moshe S & & Itzkovitz S 2019 Spatial heterogeneity in the mammalian liver. Nature Reviews. Gastroenterology and Hepatology 16 395–410. (https://doi.org/10.1038/s41575-019-0134-x)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Birdsey GM, Shah AV, Dufton N, Reynolds LE, Osuna Almagro L, Yang Y, Aspalter IM, Khan ST, Mason JC, Dejana E, et al.2015 The endothelial transcription factor ERG promotes vascular stability and growth through Wnt/β-catenin signaling. Developmental Cell 32 82–96. (https://doi.org/10.1016/j.devcel.2014.11.016)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bissig-Choisat B, Alves-Bezerra M, Zorman B, Ochsner SA, Barzi M, Legras X, Yang D, Borowiak M, Dean AM, York RB, et al.2021 A human liver chimeric mouse model for non-alcoholic fatty liver disease. JHEP Reports 3 100281. (https://doi.org/10.1016/j.jhepr.2021.100281)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Blouin A, Bolender RP & & Weibel ER 1977 Distribution of organelles and membranes between hepatocytes and nonhepatocytes in the rat liver parenchyma. A stereological study. Journal of Cell Biology 72 441–455. (https://doi.org/10.1083/jcb.72.2.441)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bossard P & & Zaret KS 1998 GATA transcription factors as potentiators of gut endoderm differentiation. Development 125 4909–4917. (https://doi.org/10.1242/dev.125.24.4909)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Braet F & & Wisse E 2002 Structural and functional aspects of liver sinusoidal endothelial cell fenestrae: a review. Comparative Hepatology 1 1. (https://doi.org/10.1186/1476-5926-1-1)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Brosch M, Kattler K, Herrmann A, Von Schönfels W, Nordström K, Seehofer D, Damm G, Becker T, Zeissig S, Nehring S, et al.2018 Epigenomic map of human liver reveals principles of zonated morphogenic and metabolic control. Nature Communications 9 4150. (https://doi.org/10.1038/s41467-018-06611-5)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Burgdorf S, Kautz A, Böhnert V, Knolle PA & & Kurts C 2007 Distinct pathways of antigen uptake and intracellular routing in CD4 and CD8 T cell activation. Science 316 612–616. (https://doi.org/10.1126/science.1137971)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Bürglin TR 1997 Analysis of TALE superclass homeobox genes (MEIS, PBC, KNOX, Iroquois, TGIF) reveals a novel domain conserved between plants and animals. Nucleic Acids Research 25 4173–4180. (https://doi.org/10.1093/nar/25.21.4173)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Cai M, Langer EM, Gill JG, Satpathy AT, Albring JC, Kc W, Murphy TL & & Murphy KM 2012 Dual actions of Meis1 inhibit erythroid progenitor development and sustain general hematopoietic cell proliferation. Blood 120 335–346. (https://doi.org/10.1182/blood-2012-01-403139)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Castro-Mondragon JA, Riudavets-Puig R, Rauluseviciute I, Berhanu Lemma RB, Turchi L, Blanc-Mathieu R, Lucas J, Boddie P, Khan A, Manosalva Pérez N, et al.2022 JASPAR 2022: the 9th release of the open-access database of transcription factor binding profiles. Nucleic Acids Research 50 D165–D173. (https://doi.org/10.1093/nar/gkab1113)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Cebola I 2020 Liver gene regulatory networks: contributing factors to nonalcoholic fatty liver disease. Wiley Interdisciplinary Reviews. Systems Biology and Medicine 12 e1480. (https://doi.org/10.1002/wsbm.1480)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Chalasani N, Wilson L, Kleiner DE, Cummings OW, Brunt EM, Unalp A & NASH Clinical Research Network 2008 Relationship of steatosis grade and zonal location to histological features of steatohepatitis in adult patients with non-alcoholic fatty liver disease. Journal of Hepatology 48 829–834. (https://doi.org/10.1016/j.jhep.2008.01.016)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Cusanovich DA, Daza R, Adey A, Pliner HA, Christiansen L, Gunderson KL, Steemers FJ, Trapnell C & & Shendure J 2015 Multiplex single-cell profiling of chromatin accessibility by combinatorial cellular indexing. Science 348 910–914. (https://doi.org/10.1126/science.aab1601)

    • PubMed
    • Search Google Scholar
    • Export Citation

  • Danoy M, Poulain S, Koui Y,