The formation of foregut endoderm cells from definitive e...

The formation of foregut endoderm cells from definitive endoderm is a crucial step in early embryonic development that leads to the formation of several vital organs, including the lungs, liver, pancreas, and part of the digestive tract (reviewed in Lewis and Tam 2006). This process is guided by localized signaling from adjacent mesodermal tissues, including the notochord and cardiogenic mesoderm, which influence the differentiation of specific regions of the foregut into distinct organs (reviewed in Lewis and Tam 2006).

The definitive endoderm is formed during gastrulation, a process early in embryonic development where the three primary germ layers (ectoderm, mesoderm, and endoderm) are established (reviewed in Lewis and Tam 2006). As the definitive endoderm develops early after fertilization, before the pregnancy is usually known, but after the ethically approved window of in vitro growth of human embryos, our knowledge of growth factors that affect human definitive endoderm and markers of definitive endoderm is derived from in vitro experimentation with human embryonic stem cells (hESCs) or induced pluripotent cells (iPCs) in comparison with in vivo and in vitro studies of mouse embryonic development (reviewed in Ghimire et al. 2021). More recently, studies on hESC/iPC-derived pancreatic organoids grown in vitro or orthotopically transplanted in mouse are providing new insights into pancreas development and disease (Hohwieler et al. 2017).

The definitive endoderm arises from the epiblast and is a precursor to the entire gut tube and associated organs (reviewed in Lewis and Tam 2006). Once the definitive endoderm is formed, it undergoes patterning along the anterior-posterior, ventral-dorsal, and medial-lateral axes of the embryo (reviewed in Lewis and Tam 2006, Kraus and Grapin-Botton 2012). This patterning is influenced by gradients of signaling molecules, including NODAL, WNT, FGF, BMP, and retinoic acid, which establish regional identities within the endoderm (reviewed in Lewis and Tam 2006).

The anterior region of the definitive endoderm, where the foregut forms, is characterized by low levels of WNT and BMP signaling, and inhibition of these pathways is essential for foregut specification (reviewed in Zorn and Wells 2010). NODAL, a member of the TGF-? superfamily, is crucial for endoderm formation and foregut patterning (reviewed in Lewis and Tam 2006). In in vitro protocols, Activin A is usually used as a substitute for NODAL as it is more easily obtained in the active form (D?Amour et al. 2005). Higher levels of NODAL signaling specify more anterior fates and the foregut is more sensitive to the reduction of NODAL than the midgut and hindgut (reviewed in Lewis and Tam 2006). Fibroblast growth factors (FGFs) are involved in anterior-posterior regionalization of the definitive endoderm and enable foregut specification (reviewed in McCracken and Wells 2012).

Not many specific definitive endoderm markers are available, as many are also detectable in other embryonic layers and some are not ubiquitously present in definitive endoderm cells, due to rapid regionalization of the definitive endoderm (reviewed in Lewis and Tam 2006).

Definitive endoderm cells are characterized by expression of the following markers:

CER1 (Cerberus, also known as CERL) is a secreted protein that was reported as one of the earliest definitive endoderm markers during mouse embryonic development (Iwashita et al. 2013, reviewed in Lewis and Tam 2006), also produced by human definitive endoderm (Iwashita et al. 2013). CER1 acts a regulator of NODAL signaling and plays an important role in endoderm formation (reviewed in Schier and Shen 2000). Expression of mRNA of LEFTY1, an antagonist of NODAL (reviewed in Schier and Shen 2000), has been reported as a marker of definitive endoderm (Genga et al. 2019; Weng et al. 2020). Related LEFTY2 mRNA has been reported as a marker of definitive endoderm in a single study (Genga et al. 2019) and has therefore not been annotated.

CXCR4 is a chemokine receptor that plays a role in cell migration and the organization of the developing endoderm (reviewed in Katsumoto and Kume 2013). CXCR4 protein is often used as a marker for definitive endoderm, particularly in in vitro differentiation protocols (d?Amour et al. 2005).

FOXA2 (also known as HNF3B) is a transcription factor critical for the formation of the definitive endoderm (reviewed in Lewis and Tam 2006) and for the differentiation of several endodermal derivatives, including the liver, pancreas, and lungs (reviewed in Friedman and Kaestner 2006). FOXA2 knockdown in definitive endoderm impairs formation of foregut endoderm (Genga et al. 2019). FOXA2 is considered as pan-endodermal marker, since it is expressed in both definitive endoderm and extraembryonic endoderm (Seguin et al. 2008).

GATA4 and GATA6 are transcription factors expressed in definitive endoderm that contribute to the regulation of genes involved in endodermal differentiation (reviewed in Lentjes et al. 2016). GATA4 and GATA6 are considered as pan-endodermal markers, since they are expressed in both definitive endoderm and extraembryonic endoderm (Seguin et al. 2008). In mice, Gata6 is required for the maintenance of acinar cell differentiation and its ablation leads to acinar-to-ductal metaplasia (ADM) and accelerates tumor development in the presence of an activating Kras G12V mutation (Martinelli et al. 2016). In mice, nicotine was shown to decrease the activity of the Gata6 gene promoter, leading to loss of Gata6 expression, dedifferentiation of acinar cells, and development of Kras-induced pancreatic cancer (Hermann et al. 2014). A fraction of human pancreatic ductal adenocarcinomas (PDACs) shows overexpression of GATA6 (Badea et al. 2008), and amplification of the GATA6 gene (Martinelli et al. 2017), which is associated with better prognosis (Martinelli et al. 2017). Although ectopic GATA6 overexpression in human PDAC cell lines increases proliferation, it inhibits epithelial-to-mesenchymal transition (EMT) and invasiveness (Martinelli et al. 2017). On the other hand, a significant fraction of PDACs show loss of GATA6 by immunohistochemistry (Martinelli et al. 2017). GATA6 has therefore been proposed as a tumor suppressor in PDAC (Martinelli et al. 2017).

MIXL1, a target of NODAL signaling, is a transcription factor involved in meso-endodermal patterning and endoderm formation (reviewed in Lewis and Tam 2006). During mouse embryonic development, Mixl1 mRNA is expressed during differentiation of mesendoderm into definitive endoderm and is required for formation of definitive endoderm (Izumi et al. 2007). In humans, MIXL1, together with EOMES, has been reported as a marker of mesendoderm, which may serve as precursors to both definitive endoderm and mesoderm (Teo et al. 2014, Li et al. 2020), but MIXL1 mRNA has also been reported as a marker of definitive endoderm (D?Amour et al. 2005, Teo et al. 2012, Genga et al. 2019).

SOX17 is a key transcription factor involved in the formation and maintenance of definitive endoderm and is one of the earliest markers of the definitive endoderm during mouse embryonic development (reviewed in Lewis and Tam 2006). SOX17 is essential for the proper development of endodermal organs (reviewed in Tan et al. 2020). A long noncoding RNA (lncRNA) T-REX17, transcribed 230 kb upstream of the SOX17 gene, was reported to be induced following SOX17 gene activation and restricted to early definitive endoderm (Landshammer et al. 2023).

CDH1 (E-Cadherin) is a cell adhesion molecule implicated as a marker of definitive endoderm in mouse (Iwashita et al. 2013), but in humans it is a marker of pluripotency and is downregulated in definitive endoderm (D?Amour et al. 2005, Teo et al. 2014, Li et al. 2020), while CDH2 has been reported to be upregulated (D?Amour et al. 2005; Li et al. 2020). In the study by Liang et al. 2020, CDH1 was used as a marker of human definitive endoderm, but it was expressed in less than one third of cells positive for the definitive endoderm marker CXCR4 upon the induction of definitive endoderm formation.

Patterning of definitive endoderm occurs through a complex cross talk between mesoderm and endoderm and involves gradients of FGFs, BMPs, WNTs, all-trans retinoic acid (atRA), and sonic hedgehog (SHH) (reviewed in Zorn and Wells 2010, McCracken and Wells 2012). The foregut part of the definitive endoderm is patterned along the anterior-posterior, ventral-dorsal, and medial-lateral axes (reviewed in Kraus and Grapin-Botton 2012) through a molecularly incompletely characterized process. Ventral foregut endoderm is surrounded by cardiac mesoderm, while dorsal foregut endoderm is in contact with the notochord and then with the dorsal aorta (Seymour and Serup 2019).

The dorsal foregut endoderm contributes to the formation of the dorsal pancreas (reviewed in Jennings et al. 2015), duodenum (Jennings et al. 2013), and esophagus (reviewed in Fausett and Klingensmith 2012). The dorsal foregut endoderm lies adjacent to the notochord and the splanchnic mesenchyme, which later constitutes the dorsal aortae (reviewed in Jennings et al. 2015, Dassaye et al. 2016). The notochord patterns the development of the dorsal foregut endoderm into the dorsal pancreatic bud by excluding sonic hedgehog (SHH), which allows expression of the key transcription factor PDX1 (Hebrok et al. 1998; reviewed in Pan and Wright 2011, Jennings et al. 2015).

FGF2, secreted by the notochord, is crucial for specifying the dorsal foregut endoderm (Hebrok et al. 1998; Ameri et al. 2010). Studies in chicken have demonstrated that FGF2, together with Activin B, inhibits SHH signaling and enables expression of pancreatic genes in the dorsal foregut (Hebrok et al. 1998).

All-trans retinoic acid (atRA), synthesized in mouse lateral plate mesoderm in cells expressing retinaldehyde dehydrogenase (Raldh2), acts as a posteriorizing agent in the developing foregut (Molotkov et al. 2005, reviewed in Duester 2008) and based on studies in mouse and Xenopus is needed for the development of dorsal pancreas (reviewed in Duester 2008). atRA was shown to induce expression of Pdx1 in embryoid bodies derived from mouse embryonic stem cells (mESCs) (Micallef et al. 2005). Human definitive endoderm cells obtained in culture by treatment of hESCs with Activin A and WNT3A express the atRA receptor beta (RARB) (Johannesson et al. 2009).

In culture, definitive endoderm cells derived from hESCs, differentiate into foregut endoderm cells when treated with FGF4 and atRA, where atRA stimulates expression of PDX1 while FGF4 primarily acts to promote cell viability (Johannesson et al. 2009). The authors speculate that their protocol produces dorsal foregut endoderm (Johannesson et al. 2009), which is in line with gene expression profiling showing that hPSC-derived definitive endoderm cells differentiate along dorsal rather than ventral pancreatic program (Jennings et al. 2017).

A detailed immunohistochemical study of the dorsal foregut endoderm at Carnegie stage 10 (CS10) of human embryogenesis was published by Jennings et al. 2013.

Dorsal foregut endoderm cells are characterized by expression of the following markers:

FOXA2 is expressed in the nuclei of all epithelial cells of the developing human endoderm (Jennings et al. 2013).

SOX17 is expressed in the nuclei of dorsal foregut endoderm cells (Jennings et al. 2013).

Human dorsal foregut endoderm cells are negative for SHH, while SHH is expressed in ventral endoderm cells (Jennings et al. 2013).

At CS10, human dorsal foregut endoderm cells do not express PDX1 or GATA4 (Jennings et al. 2013).

Retinoic acid receptor beta (RARB) is upregulated in hESC-derived definitive endoderm and required for its differentiation along the dorsal pancreatic trajectory (Johannesson et al. 2009). RARB is therefore annotated as a required input.