Screen to identify the novel pancreatic gene Synaptotagmin 13 (Syt13)

Contents

1 Abstract

2 Introduction.
2.1 The early embryonic development
2.2 Development of endodermal derived organs
2.3 Development of the pancreas
2.4 Regulatory networks of pancreas development
2.5 The model of endocrine formation
2.6 Establishment of epithelial asymmetry
2.7 The family of Synaptotagmins (Syt)
2.8 Aim of this thesis

3 Results
3.1 Generation of the Foxa2 Venus mouse line
3.1.1 Design and generation of the Foxa2 Venus (FVF) targeting vector
3.1.2 Analysis of the Foxa2 Venus mouse line in the pancreas
3.2 Genome-wide expression profile of the pancreas in the secondary transition
3.2.1 Bioinformatic analysis of pathways in the secondary transition
3.2.2 Bioinformatic analysis of genes in the secondary transition
3.3 Identification and characterization of pancreatic genes
3.3.1 Temporal and spatial progression of pancreatic genes
3.3.2 Temporal and spatial progression of unknown pancreatic genes
3.4 Analysis of the novel pancreas gene Synaptotagmin 13 (Syt13)
3.4.1 Bioinformatic analysis of Syt13
3.4.1.1 The family of Synaptotagmins
3.4.1.2 Interaction partner of SYT13
3.4.1.3 Target gene prediction of SYT13
3.4.2 Functional analyses of Syt13
3.4.2.1 The gene Syt13
3.4.2.2 The amino acid (aa) sequence of Syt13
3.4.3 Generation of the genetically modified mouse line Syt13
3.4.3.1 Design, generation and verification of the Syt13 GT targeting vector
3.4.4. Syt13 reporter gene expression in the early embryo
3.4.4.1 Characterization of Syt13 reporter gene expression in the crown
3.4.5 Syt13 mutants present defects in the adult pancreas
3.4.6 Syt13 expression in pancreas organogenese
3.4.7 Syt13 associated SNP reveal T2D susceptibility
3.4.8 Delamination of endocrine precursors in Syt13 mutants is impaired
3.4.9 Syt13 mutants show polarity defects

4. Discussion
4.1 FVF marks the multipotent progenitors in the pancreas
4.2 The FVF mouse line is a valuable tool for genome wide expression profiles
4.3 Molecular pathways guiding pancreas organogenesis
4.4 The pancreas gene selection for known and unfamiliar genes
4.5 Generation of different mouse lines
4.6 Generation of the Syt13 GT/GT mouse line
4.7 Syt13 expression in a distinct subset of tissue
4.8 Pancreatic multipotent progenitors and endocrine cells marked by Syt13
4.9 Syt13 initiates morphogenesis in the pancreatic epithelium
4.10 The subcellular localization of Syt13 suggests a role in polarity membrane complexes along with BB positioning
4.11 Potential mechanism of Syt13 in endocrine lineage formation
4.12 Hypothetical molecular function of Syt13

5 Material and Methods
5.1 Material
5.1.1 Equipment
5.1.2 Consumables
5.1.3 Kits
5.1.4 Chemicals
5.1.5 Buffer and solutions
5.1.6 Enzymes
5.1.7 Sera and Antibodies
5.1.8 Oligonucleotides
5.1.9 Cell lines
5.1.10 Culture media
5.1.11 Molecular weigth markers
5.1.12 Mouse lines
5.2 Methods
5.2.1 Bioinformatics methods
5.2.1.1 Affymetrix®Gene 1.0 ST Array
5.2.1.2 Affymetrix®Gene 1.0 ST Array card
5.2.1.3 Affymetrix®Gene 1.0 ST Array card quality control
5.2.1.4 Affymetrix®Gene 1.0 ST Array card analysis
5.2.1.5 Pancreas gene selection using the digital database Genepaint.org
5.2.2 Cell culture
5.2.2.1 Embryonic stem cell culture and spheres culture
5.2.2.2 Culture of primary murine embryonic fibroblasts
5.2.2.3 Treatment of MEF with mytomycin
5.2.2.4 Freezing -Thawing of MEFs
5.2.2.5 Freezing -Thawing of ES cells..
5.2.2.6 Passaging of ES cells..
5.2.2.7 Electroporation of ES cells....
5.2.2.8 Picking of ES cell clones...
5.3 Molecular biology
5.3.1 DNA extraction
5.3.2 RNA preparation
5.3.3 DNA/RNA concentration
5.3.4 Reverse transcription
5.3.5 Gelelectrophorese
5.3.6 DNA sequencing
5.4 Protein biochemistry
5.4.1 Protein extraction from tissue
5.4.2 Bradford assay for determining protein concentration
5.4.3 Western blot
5.4.4 Western blot immunostaining
5.4.5 Immunohistochemistry
5.5 Embryology
5.5.1 Genotyping of mice and embryos
5.5.2 PCR Programs for genotyping
5.5.3 Isolation of embryos and organs
5.5.4 Tissue clearing with BABB
5.5.5 X-gal (5-bromo-4-chloro-3-indolyl β-D-galactoside) staining
5.6 Histology
5.6.1 Paraffin sections
5.6.2 Counterstaining with Nuclear Fast Red (NFR)
5.6.3 Cryosections

6 Supplement
6.1 Abbreviations
6.2 Figures and tables
6.3 Literature....
6.4 Curriculum Vitae

6.5 Congresses and Publications

6.6 Additional Figures

6.7 Alternative Discussion

Danksagung

Ich möchte mich auf dieser Seite bei allen bedanken die mich auf diesem Weg begleitet haben. Insbesondere will ich hier einige beim Namen nennen.

Insbesondere will ich mich bei Herrn Professor Dr. Heiko Lickert bedanken, für die Möglichkeit meine Promotion in seiner Arbeitsgruppe anzufertigen. Die Begeisterung für die Wissenschaft und dein Wissen hat mich stark beeindruckt. Ausserdem will ich mich natürlich noch bei Professor Dr. Martin Hrabe de Angelis bedanken, für die Unterstützung und fachliches Wissen.

Bei Herrn Professor Dr. Grill bedanke ich mich herzlich für den Vorsitz in meiner Prüfungskommission.

Bei dem Team der Arbeitsgruppe Lickert will ich mich für die langjährige Unterstützung bedanken. Besonders Dr. Ingo Burtscher für das offene Ohr bei allen Fragen, die Einführung in konfokale Mikroskopie, Immunhistochemische Färbungen und ES Zellkultur. Bei Dr. Aurelia Raducanu für die Unterstützung in dem nicht immer leichten Screening Projekt und die anschliessende Mausarbeit. Zusätzlich noch bei Dr. Mostafa Bahkti für die Unterstützung und Übernahme im Syt13 Projekt. Ausserdem will ich Aimee Bastidas-Ponce viel Erfolg bei der Weiterführung wünschen - es ist ein vielversprechendes Projekt.

Insbesondere gilt mein Dank an Dr. Bomi Jung, Heide Oller und Dr. Alexander Korostylev für die Unterstützung und hilfreichen Tipps. Auch geht ein herzlicher Dank an Dr. Silvia Engert für die Unterstützung in manchmal schwierigen Situationen. Donna Thompson - du bist eine Perle und dir gebührt extra Dank für deinen besonderen Einsatz in der Arbeitsgruppe. Anne, Wenke, Bianca und Kerstin haben wundervoll geholfen bei alltäglichen Problemen und Arbeiten wie PCRs, Maushausbesuche, Bestellungen und andere kleine Dinge im Laboralltag. Ausserdem noch zu erwähnen sind Dr. Nikola Müller, Steffen Sass, Dr. Martin Irmler und Dr. Harald Staiger für die gute Zusammenarbeit und Austausch in den verschiedenen Projekten.

Ausserdem will ich mich noch bei dem IDR Team bedanken für die Unterstützung und Informationen die das Projekt vorangetrieben haben. Hier geht mein besonderer Dank an Daniela Padula, Martin Preusse, Esra Karaköse, Anika Böttcher, Noah Mozurri, Lisann Heyner und Felizitas Schmid und natürlich nicht zu vergessen Erik Bader.

Aber der größte Dank geht an meine Familie für die Unterstützung in der Zeit - Danke das ihr für mich da wart und immer ein offenes Ohr hattet. Danke auch an Alex, Alex, Manu, Dani, Alex, Elke und nicht zu vergessen Rainer!

1 Abstract

Screen to identify the novel pancreatic gene Synaptotagmin 13 (Syt13)

The multipotent pancreatic progenitors segregate in the secondary transition (E12.5-15.5) into the pancreatic lineages endocrine, ductal and acinar in mouse development. Different signaling pathways as WNT, FGF, Notch, BMP, SHH and RA define anterior-posterior axis and the lineage decision of the organ. The precise mechanism of the formation of the endocrine lineage will allow generation of in vitro β-cells for cell-replacement therapy and identify potential disease genes. Thus, we utilized the Foxa2-Venus fusion (FVF) knock-in reporter mouse to separate the pancreatic endodermal epithelium (FVF +) from the surrounding non-endodermal tissue (FVF -) in the secondary transition. Subsequent global gene expression analyses identified 886 temporal significantly regulated genes in both tissue compartments. By using the public database Genepaint, spatial expression of the genes in the embryo and in the pancreatic region was analyzed in classifying the in situ mRNA expression pattern at E14.5 into tip/acinar, epithelium and mesenchyme. In a next step, GO term analyses classified the majority of the genes according to their predicted protein function. The selection of pancreatic genes, yet functionally not described are under the criteria of temporal regulated expression in the secondary transition, lineage prognosis by spatial pancreatic mRNA hybridization pattern and predicted protein function. In a next step, we generated Knock in/knock out mice lines to analyze the function and mechanism of these pancreatic candidate genes. Taken together, we successfully accomplished a global gene expression profile during secondary transition of the pancreas, inspected WNT, FGF, Notch, BMP, SHH pathway components and pancreatic lineage determinants. The pancreatic gene candidate, Synaptotagmin 13 (Syt13) is expressed in the secondary transition, the trunk mRNA hybridization pattern similar to endocrine TF factor Ngn3 and the gene functionally not described. Syt13 is a member of the large family of Synaptotagmins that are known to regulate vesicle docking and fusion for neurotransmitter and secretory granule release. Expression analysis using the Syt13 lacZ reporter gene revealed endocrine-specific expression in pancreas organogenesis and in the adult Islets of Langerhans. In addition, we observe in Syt13 deficient pancreas delamination defects of endocrine progenitors during secondary transition. Also, we quantified a decline of endocrine differentiated cells characterized through Insulin. In collaboration with Harry Staiger and Hans-Ullrich Häring from the University in Tübingen we identified a single nucleotide polymorphisms (SNP; rs11038374) in SYT13 that correlates with insulin secretion defects in a cohort of 2100 individuals with increased risk for diabetes. Thus, Syt13 is expressed and regulates the endocrine lineage in mouse and may present a promising target for diabetes therapy.

Screening und Identifizierung des neuen Pankreas Gen Synaptotagmin 13 (Syt13)

In der Pankreasentwicklung differenzieren aus einem Pool von Vorläuferzellen in der sekundären Transition (E12.5-15.5) die endokrine, exokrine und duktale Zelllinien. In dem Prozess selbst sind Gewebeinteraktionen und komplexe Signalwege involviert. Wir haben ein Transkriptionsprofil erstellt um die zeitliche und räumliche Expression regulatorischer und funktioneller Gene zu charakterisieren. Dabei haben wir spezifisch als Zeitpunkt die sekundäre Transition gewählt in der sich die verschiedenen pankreatischen Ziellinien abspalten.

Mit der Forkhead Transkriptionsfaktor Foxa2 Venus (FVF) Mauslinie konnten wir pankreatisches Epithelium (FVF +) von umgebenden Mesenchym (FVF -) trennen. In dem anschließenden globalen Transkriptionsprofil identifizierten wir 886 signifikant regulierte Gene in beiden Kompartimenten, was auf essentielle Gewebeinteraktionen in der Pankreasentwicklung hinweist. Die Expression aus unserem Profil konnten wir dann mit der öffentlich zugänglichen Datenbank Genepaint für bekannte und unbekannte Gene bestätigen. Mit diesem Ansatz haben wir zudem die unterschiedlichen pankreatischen Linien nach bestimmten Mustern klassifiziert und zwar in Exokrin, Endokrin und Mesenchym. Für bekannte und unbekannte pankreatische Gene konnten wir somit die zeitlich und räumlich begrenzte Expression bei der Abspaltung in unterschiedlichen pankreatische Linien und die entsprechende Linie aufzeigen. In einem nächsten Schritt generierten wir verschiedene transgene Mauslinien um Funktion und Mechanismus der unbekannten Gene näher zu untersuchen. Zusammengefasst zeigt unser Transkriptionsprofil, das bekannte und unbekannte Gene in Gewebeinteraktionen und Stoffwechselwegen notwendig sind für die Entwicklung des Pankreas.

Eines der unbekannten identifizierten Gene ist Synaptotagmin 13 (Syt13), mit räumlich und zeitlich begrenzter Expression in der sekundären Transition. Erste LacZ Reportergen Expressionsdaten weisen auf Syt13 in endokrinen Zellen und adulten Langerhans’schen Inseln hin. Wir zeigen, das Syt13 im Pankreas notwendig ist für die Abspaltung in die endokrine Linie und für die Bildung der endokrinen Zellmasse. In Kollaboration mit Harry Staiger (Universität Tübingen) konnten wir zudem Syt13 direkt auf Diabetes beziehen, Genome-wide association studies (GWAS) Studien zeigen auf single nucleotide polymorphism (SNP; rs11038374) in SYT13 die Defekte in der Insulinsekretion aufweisen. Unsere Daten schlagen vor, das Syt13 in der endokrinen Zellentstehung eine Rolle spielt. Die weiteren Aussichten für Syt13 sind dabei therapeutische Ansätze um endokrine Linien zu expandieren oder um diese zu induzieren.

2 Introduction

2.1 The early embryonic development

The development of the embryo starts with the formation of the zygote out of the fertilized egg. Thus, the fertilized zygote is capable of producing all the different lineages responsible for the embryo proper. Subsequent cell divisions onwards lead from the zygote to the 8-cell stage, termed as morula (Gardner 2001; Piotrowska and Zernicka-Goetz 2001; Piotrowska-Nitsche and Zernicka-Goetz 2005). These 5 cell cycles are commonly referred to as cleavage decisions, which occur in the absence of cellular growth or in the increase of total cell mass - leading to the blastocyst. At embryonic stage (E) 3.5, a cavity will form in the early blastocyst and will separate the embryo into two cell populations: the outer lining extraembryonic trophectoderm (TE) and the inner cell mass (ICM), oriented at one side of the cavity. This lineage segregation process is determined as the first cell fate decision. Further differentiation occurs within the ICM, specifying in the second cell fate decision the primitive Endoderm (PrE) and the Epiblast (Epi) (Figure 2.1 - late blastocyst). The PrE does not contribute to the embryo itself, presumably later as extra-embryonic (ExE) tissue to the yolk sac, whereas the ICM gives rise to the embryo proper (Cockburn and Rossant 2010; Johnson et al. 1986). Prior implantation into the maternal uterus at E4.5, the embryo elongates to an egg-shaped structure. The PrE covers the Epi in the cavity, so that the embryonic lineage will be enclosed by the PrE on one side and by the TE on the other side. In the next step, the PrE expands and lines the luminal surface of the TE and the distal visceral endoderm (DVE) (Figure 2.1 - implanted embryo). Thus, the DVE mediates activity of wingless type MMTV integration family (WNT), fibroblast growth factor (FGF), Bone morphogenetic protein (BMP) and Nodal for the differentiation and patterning of the Epi (Lewis and Tam 2006; Sumi et al. 2013; Sumi et al. 2008; de Sousa Lopes et al., 2004; Beddington and Robertson 1999). In a next step, the Anterior-Posterior (AP) patterning of the body axis is initiated in a dynamic process by variations in the endodermal structure, regionalized gene expression domains and morphogenetic signaling (Tam and Loebel, 2007). At E6.5, the cells of the DVE migrate into the prospective anterior region of the embryo and become the anterior visceral endoderm (AVE) of the late cylinder (Figure 2.1 - late cylinder). In the AVE activity of Wnt and Nodal inhibitors restrict the action of this signaling pathways to the posterior side of the embryo. Thus, formation of the primitive streak (PS) is initiated posterior of the AVE. With the formation of the PS at E6.5, the gastrulation and establishment of the three principal germ layers takes place (Beddington and Robertson 1999; Burtscher and Lickert 2009; Lawson et al. 1986).

Abbildung in dieser Leseprobe nicht enthalten

modified from Arias et al., 2013

Figure 2.1: Early stages of mouse embryo development

In the blastocyst cells get orientated into Epi, distal visceral endoderm (DVE) and primitive endoderm (PrE). The Extraembryonic tissue (ExE) does not contribute to the embryo itself. After implantation Epi cells are immigrating into the primitive streak, egress and displace the AVE. Prior to gastrulation, proximal-distal (PD) and AP are established through morphogenetic signals.

Abbreviations: ExE = Extraembryonic; VE = Visceral endoderm; PrE = Primitive endoderm; Epi = Epiblast; AP = Anterior-Posterior; DVE = Distal visceral endoderm; AVE = Anterior visceral endoderm; PS = Primitive streak; PrE = primitive endoderm.

2.2 Development of endodermal derived organs

The gastrulation is initiated through formation of the PS on the opposite side of the AVE. In the steps of gastrulation, Epi cells migrate into the PS and segregate into distinct cell populations. The clonal analysis of Epi fate highlighted that temporal and spatial ingression of the Epi in the PS defines the three principal germ layers: endoderm, mesoderm and ectoderm.

With this in mind, Epi cells remaining in the blastocyst are still competent of the different embryonic lineages (Morani et al., 2013). As gastrulation is initiated, the Epi cells migrate through the posterior region in the PS and transform into the Exe mesoderm (Kinder et al., 1999). The center and most likely anterior part of the PS contributes to the mesodermal derived cells as cardiac, cranial, paraxial and axial mesoderm (Kinder et al., 2001). Contrary, the most anterior end of the PS forms the definitive endoderm (DE) (Wells and Melton 1999; Engert et al. 2013). In further steps migrates the DE to the anterior region of the embryo via different routes and displaces the AVE. At the end of gastrulation, the DE covers as a cell-layer sheet the embryo with approximately 500 cells. (Tam and Loebel 2007; Lawson and Pedersen 1987). The DE will form the primitive gut tube through folding of this epithelial sheet. The folding appears at the anterior and posterior ends. Thus, anterior endoderm forms the anterior intestinal portal (AIP) defining the foregut and equally, the posterior endoderm folds to establish the caudal intestine portal (CIP) defining the hindgut region. At the end of folding of the DE, turning of the whole embryo completes internalization as a tube. The derivatives of the gut regions are regionalized in foregut, midgut and the hindgut. The foregut gives rise to the largest number of structures; it forms the pharynx and its derivative as the respiratory part, esophagus, stomach, duodenum and the liver with the pancreas. The midgut is the supplier of the gastrointestinal part of the embryo, with the exception of the upper duodenum. Contrary, the hindgut contributes to the urogenital tract (The atlas of mouse development).

In the ventral foregut endoderm development, organ domains are established through transcription factors and signals from the mesenchyme. The hepatic endoderm is induced by FGF signaling from the mesenchyme and is marked by Hematopoietically expressed Homeobox (Hhex) expression, establishing the liver (Zorn and Wells 2010; Zorn and Wells 2009) The first obvious indication of pancreas morphogenesis takes place approximately at embryonic stage E8.0. It is noteworthy that all pancreatic epithelial cells derive from a common pool of pancreatic progenitors that evaginate from the dorsal and ventral section of the foregut/midgut junction. A signaling cascade mainly contributed by mesodermal-derived FGF, BMP, WNT, Activin A, retinoic acid (RA) and repression of sonic hedgehog (SHH) leads to the outgrowth of the pancreatic protrusions (Raducanu and Lickert 2012; Pan and Wright 2011). Upon a gut roation between E11 and E12, the ventral pancreas switches dorsally and the proximity to the dorsal pancreatic bud leads to the fusion of the two lobes. This period involves tubulogenesis in the ongoing organogenesis which includes micro lumen fusion and the rearrangement of the epithelial sheet into a stratified epithelium (Kesavan et al. 2009). Along with plexus remodeling and proliferation, the pancreatic epithelium (PE) transforms into a branched single luminal ductal system. Simultaneously, in the so called secondary transition, the PE differentiates into the pancreatic lineages Subsequently, the formerly multipotent pancreatic progenitors (MPP) will be specified by lineage commitment through a temporal and spatial restricted transcription factor program. Lineage tracing experiments and Gain/Loss-of function analyses in genetically modified mice further gained insights into the morphogenetic and transcriptionally mechanism which facilitate maturation of the pancreatic anlagen (Mehta and Gittes 2005; Li et al. 2004; Gittes et al. 1996).

The earliest transcription factors that characterize the pancreatic buds are Pancreatic and duodenal factor 1 (Pdx1), pancreas transcription factor 1 alpha (Ptf1 α) and Homeobox Protein HB9 (Hlxb9). In addition, combined ectopic expression of Pdx1 and Ptf1 α converts posterior endoderm into endocrine and exocrine pancreatic tissue, highlighting the importance of both factors for pancreas initiation (Afelik et al. 2006; Sherwood et al. 2009) Although, wingled helix/forkhead box 1/2 (Foxa1/2), NK homeobox x 6-1 (Nkx6-1) and Nkx2-2 are co-expressed in the pancreatic initiation stage, but their expression is broadly prevalent in the foregut endoderm. These multipotent cells give later on rise to all the different lineages of the adult pancreas, consisting of exocrine, ductal and endocrine cell. A landmark article by Golosow and Grobstein demonstrated that cytodifferentiation of the PE is promoted by the surrounding non-endodermal tissue. Experiments either stripping of the mesenchyme of the endodermal-derived, as well as pancreatic explants separated by a porous barrier of the non-endodermal part in trans filter experiments, could define the necessity of tissue interactions. In both projections the PE failed to proliferate. Thus, highlighting the tissue interaction in pancreas organogenesis (Golosow et al., 1962; Willmann et al., 2016).

2.3 Development of the pancreas

The adult pancreas is a compound gland which is important for the nutrient metabolism and composes of an exocrine compartment with acinar cells and ductal cells and an endocrine compartment composed of and pancreatic polypeptide (PP) cells organized in structures called the Islets of Langerhans. The acinar cells secrete digestive enzymes which are transported through the ductal system into the duodenum where they catalyze the digestion of nutrients. The endocrine cells secrete hormones such as glucagon, insulin, ghrelin, somatostatin and PP into the blood stream to maintain a fine balanced hormone homeostasis. Understanding the developmental process that leads to the formation of the functional organ is essential to generate potential therapeutic strategies for diabetes (Zorn and Wells 2010; Zorn and Wells 2009).

Pancreas formation appears at around E8.0 likely in the last steps of gastrulation with the separation of the primitive gut tube into foregut and hindgut. In this stage, inductive mesodermal signals including FGFs, BMPs, Activin and RA induce an area in the lateral endoderm and a dorsal domain in the midline endoderm to become pancreatic tissue as shown by expression of the transcription factor (TF) Pdx1, a key regulator of pancreas development (Stanger et al. 2007; Mehta and Gittes 2005; Gittes et al. 1996; Gittes et al., 2009). Continuous signals of Wnt and FGF-signaling from the surrounding tissues activate expansion of the two buds into the surrounding mesenchyme. The stratified epithelium becomes polarized and microlumens are generated throughout the bud which later fuse together to form a luminal epithelial plexus (Figure 2.2). The stratified epithelial bud contains MPP and few differentiated endocrine cells which are mainly glucagon-positive (Herrera et al., 2000). This first differentiation event takes place between E9.0 and E11.5 is called the primary transition. At around E11.5, gut tube rotation brings the two pancreatic buds in close proximity to one another and this event leads to their fusion into a single pancreatic primordium, which shapes the future organ. Massive plexus remodeling leads to the organization of a tubular structure from which branches continue to form through an epithelial branching morphogenesis process. The epithelial cells are still multipotent and capable of differentiation into the three pancreatic lineages: duct, acinar and endocrine. From E12.5 until E15.5, the three pancreatic lineage decisions are subsequently being made in the process of the secondary transition. The segregation of the epithelium initiates patterning into a distal tip region and a proximal trunk. Thereby, the cells representing the tip region will be assigned for the exocrine lineage, which are characterized through expression of Carboxypeptidase 1 (Cpa1), which expresses cells and co-expression of Ptf1 α. The pattern of the trunk region is commonly represented by ductal/endocrine factors illustrated through Sry related Homeobox 9 (Sox9) and Neurogenin 3 (Ngn3). Thus, the ductal cells are bi-potent for either the ductal or endocrine lineage and restricted in the MPP PE. In the secondary transition at E12.5 - 15.5 in the ductal compartment the progenitor pool of the endocrine progenitors marked by Ngn3 + - cells delaminate out of the ductal cord through a process which is likely regarded as epithelial-mesenchymal transition (EMT) (Gouzi et al. 2011). Shortly after delamination, the endocrine precursors migrate into the PE and aggregate into the precursor Islets of Langerhans. From E18.5 onwards and in the first days after birth, the Islet of Langerhans morphogenesis takes place, which are indicated through the α-, δ-, ε-,PP- and β-cells that form clusters, invade the PE and acquire the typical architecture of the adult Islet of Langerhans (Zorn & Wells 2009; Gittes 2009).

In pancreas organogenesis, tubulogenesis is tightly spatio- and temporal regulated, as the lineages including branching morphogenesis develop in the proliferating pancreas (Willmann et al. 2016). The expansion of the pancreas alters compared to branching morphogenesis of different tubular organs as lung, kidney and ureteric bud (Karihaloo et al. 2005). The classical branching morphogenesis suggests that peripheral growth accompanied with epithelial expansion, contrary to the pancreatic branching, implicates plexus formation, proliferation and remodeling and further plexus formation. Postnatal (P), the PE is transformed into a branched single luminal ductal system that is necessary to transporting digestive enzymes. Currently, tubulogenesis remains widely unknown as the polarity establishment in the different lineages of the pancreas especially affecting the evolving β-cells. This might be an interesting point of research, especially regarding the polarity within the adult Islets of Langerhans, respective β-cells (Kone et al. 2014; Granot et al. 2009; Lo et al. 2012; Martin-Belmonte and Perez- Moreno 2011).

Abbildung in dieser Leseprobe nicht enthalten

Figure 2.2: Tubulogenesis of the embryonic pancreas

The pancreas buds out of the anterior endoderm and at E11, micro lumen within the PE appear. Through a process called tubulogenesis at the beginning of the secondary transition, patterning into the different lineages determines specific regions. Epithelial remodeling (plexus) leads to the adult pancreas including defined compartments as the exocrine lineage (adult pancreas - red ball) and the endocrine Islets of Langerhans (adult pancreas - blue balls with β-cells; outer layer in yellow α-, δ-, ε- and pp-cells)

Abbreviations: E = embryonic stage; PP = pancreatic polypeptide; PE = pancreatic epithelium

2.4 Regulatory networks of pancreas development

In pancreas organogenesis, a transcription factor hierarchy orchestrates the segregation into the different cell types. The entry point of pancreas initiation is well characterized by the TF Pdx1, as in the MPP after pancreas initiation Pdx1 serves as a marker for the PE (Gao et al. 2008).

Contrary, the adult pancreas consists of different cell types represented in the different lineages: acinar and ductal cells and the endocrine compartment comprising of the Islets of Langerhans which include α-, δ-, ε- and PP-cells and the β-cells. As the pancreas evolves, the PE is characterized by Pdx1 expression, thereby reflecting subpopulations as Pdx1low and Pdx1high. The Pdx1low population is specifically required for the exocrine lineage, whereas the lineage determining factor for the endocrine compartment is represented by the subpopulation Pdx1high (Holland et al. 2002). A transcription factor hierarchy is guiding the pancreatic outgrowth, consisting of Pdx1low expressing PE within acinar cells represented by Ptf1αHigh expression, as suppression of Ptf1 α induces acinar-to-endocrine formation (Hesselson et al. 2011). In addition, members of the GATA family such as GATA binding protein 4 (Gata4) and basic helix-loop-helix 1 (Mist1) regulate acinar cell proliferation (Pan and Wright 2011; Jia et al. 2008). Thus, Mist1 is scattered on the foregut wall at stage E10.5 and after E13.5 and will be restricted to the acinar progenitors and mature exocrine cells. Gata4 expression is restricted to the acinar fate in the E, in the adult pancreas Gata4 expression localizes specifically in the α - and β-cells of the Islets of Langerhans (Xuan et al. 2012). The factor nuclear receptor subfamily 5 group a2 (Nr5a2) controls the early MPP the progenitor cells in the PE and after separation into tip and trunk pattern the acinar differentiation. It acts directly or through regulatory interactions with the acinar determining factors as Gata4, Ptf1 α and the recombining binding protein suppressor of hairless (Rbpjl) (Hale et al. 2014).

In addition, the SRY related gene 9 (Sox9) determines the ductal lineage independently of Foxa2 and Hnf6 (Dubois et al. 2011). In the adult pancreas, the ductal lineage is responsible for the transportation of digestive enzymes from the surrounding exocrine tissue into the duodenum. Intercalating blood vessels in the Islets of Langerhans transport the hormones into the blood stream for maintaining the precise blood glucose level in the adult pancreas. Through the use of a gain-of-function, in mice constantly misexpressing Aristaless related homeobox (Arx), β-cells were converted into pancreatic PP and α-cells (Courtney et al. 2013). Besides decreasing β-cell mass, mice developed hyperglycemia and died (Figure 2.3 A). There are several more regulatory genes and combinations, thus pointing to a fine- tuned spatio- and temporal network of TF in the secondary transition in pancreas organogenesis.

Abbildung in dieser Leseprobe nicht enthalten

modified from Pan and Wright et al., 2011

Figure 2.3: Lineage hierarchy of the pancreatic lineages with cell specific genes for the adult pancreas (A) and the embryonic pancreas from multipotent state at E8.0 to the mature β cell (B).

(A) The adult pancreas consist of acinar cells (Ptf1α high, GATA4, Mist1 and Nr5a2), ductal lineage (Hnf6, Hnf1β, Sox9 and Foxa2) and the Islets of Langerhans (α-, δ-, ε- and PP-cells and β-cell).

(B) Multipotent progenitors get restricted to the β cell by different lineage commitments.

The transcription factor cascade starts with multipotent pancreatic cells (MPC) (Mnx1, Pdx1, Ptf1α low, Sox9, Nkx6-1, Hnf1β, Hnf6, Foxa2, GATA4), the second multipotent state is illustrated in a classical tip pattern (Pdx1low, Ptf1α low, Sox9low, Nkx6-1, Hnf1βhigh, Hnf6 and Foxa2). Bi-potent progenitors are although characterizing the trunk pattern (Pdx1low, Sox9high, Nkx6-1, Hnf1βhigh, Hnf6, Foxa2). In the next step, endocrine progenitors segregate in the PE (Ngn3high, Myt1, Isl1, NeuroD, Snail2). Endocrine progenitors will led to endocrine precursors (Mnx1, Myt1, Isl1, NeuroD, Rfx6, Pax6, Nkx6-1, Nkx2-2, Snail2) and to immature β cells (Pdx1high, Mnx1, Nkx6-1, NeuroD, Nkx2-2, MafB, MafA, Pax4, Snail2low) to β cells (Pdx1high, Mnx1, Nkx6-1, NeuroD, Nkx2-2, MafB, Pax4, Foxa1 and Foxa2)

Abbreviations: Ptf1α = pancreas specific transcription factor 1 α; Gata4 = Gata binding protein 4; Mist1 = basic helix-loop-helix protein; Nr5a2 = nuclear receptor subfamily 5 group a; Hnf6 = Hnf homeobox 6; Hnf1ß = Hnf homeobox 1ß; Sox9 = Sry related homeobox 9; Foxa2 = wingled helix/forkhead box 2; Foxa1 = wingled helix/forkhead box 1; PP = pancreatic polypeptide; MPC = multipotent pancreatic cells; Mnx1 = Moto neuron and pancreas homeobox 1; Pdx1 = pancreatic and duodenal homeobox 1; Nkx6-1 = Nk homeobox 6-1; Nkx2-2 = Nk homeobox 2-2; Ngn3 = Neurogenin 3; Myt1 = Myelin transcription factor 1; Isl1 = Isl lim homeobox 1; NeuroD = Neuronal Differentiation 1; Snail2 = Snail family Zinc Finger 2; Rfx6 = regulatory factor x 6; MafA = V-maf musculoaponeurotic fibrosarcoma oncogene homolog A; MafB = V-maf musculoaponeurotic fibrosarcoma oncogene homolog B; Pax4 = paired homeobox x 4.

In the development of the pancreas, in conjunction with Ptf1 α and Foxa2, Pdx1 facilitates the expansion of the PE (Stanger et al. 2007). Thereby in later stages, it acts as an glucose-responsive regulator of insulin gene expression, likely for glucagon-like peptide-1 (GLP-1) (Hay et al. 2005). Ptf1 α and Foxa2 act upstream of Pdx1 with the combined expression in the MPP of the PE as early as the pancreatic formation is initiated at E8.0. In the later steps of pancreas organogenesis and finally in the adult pancreas, the total pancreatic cell mass depends strikingly on these marker onset (Stanger et al., 2007). The first lineage segregation appears as early as E9.5 which does not affect the adult pancreas. After the first lineage segregation, a second lineage segregation arises with a typical expression pattern that is described as a tip and trunk. The typical tip pattern represents the MPP at E14.5. The marker onset illustrating the typical bi-potent trunk pattern for the ductal and endocrine lineage in the PE at E14.5 is illustrated by the TF Pdx1low, Ptf1α low, Sox9low, HNF1βlow, NK homeobox, family 6- 1(Nkx6-1), Gata4 and Nr5a2 (Zhou et al., 2007). The compartmentalization of the ductal and endocrine lineage leads to a subset of Neurogenin3high (Ngn3high), Myelin transcription factor 1 (Myt1), ISL LIM homeobox 1 (Isl1), neuronal differentiation 1 (NeuroD) and Snail Family Zinc Finger 2 (Snail2) expressing cells (Figure 2.3 B). Thus, the endocrine progenitors still maintain the MPP state of the endocrine lineage, meaning that all the different subcells in the mature Islets of Langerhans derive out of this progenitor pool. In a next step in pancreas organogenesis, the immature β-cells will develop in the precursor Islets of Langerhans to the final mature β-cells. Factors that determine this specific lineage commitment include Pdx1high, motor neuron and pancreas homeobox 1 (Mnx1), Nkx6-1, NeuroD, Nkx-.2, V-Maf Avian Musculoaponeurotic Fibrosarcoma Oncogene Homolog B (MafB), MafA, Paired Box 4 (Pax4) and Snail2low (Figure 2.3 B). As Snail factors determine the process of EMT, in the mature Islets of Langerhans Snail2 is no longer expressed (Schaffer et al. 2013; Seymour et al. 2012; Herrera 2000).

2.5 The model of endocrine formation

In line with pancreas development, the endocrine lineage at approximately E13.5 will be restricted. Specific factors that drive the endocrine lineage are described as trunk pattern in the PE and is reflected by an epithelial cord pattern, which is illustrated by Sox9 expression. Within this cord-like structure, the expression of the Ngn3 -transient population represents the endocrine progenitors (Gradwohl et al. 2000; Gu et al. 2002; Schwitzgebel et al. 2000; Jensen 2004; Zhou et al. 2007). Mainly, lineage tracing experiments implicated that Ngn3 as a basic helix-loop-helix transcription factor is necessary for the establishment of all endocrine cells as glucagon, Insulin, somatostatin and PP which first assemble into the Islets of Langerhans at E18.5 (Gradwohl et al. 2000; Herrera 2000). The latest attempts showed that Sox9 deficient mice illustrate a severe reduction in endocrine progenitors marked by Ngn3, which implicates that Sox9 acts upstream of the endocrine lineage (Seymour et al. 2012). Furthermore, Lynn and Seymour proposed a cell-autonomous role for Sox9 in Ngn3 induction, which suggests an negative feedback-loop for co-related expression of Sox9 and Ngn3 (Shih et al. 2012; Seymour et al. 2012). These results further determine the importance of Ngn3 as a key TF for the endocrine progenitor. In addition, upstream TF as Sox9 and Pdx1, specifically in the duct, regulate expression of endocrine precursor Ngn3. The signal cascade for the regulation of Ngn3 is still controversial, as there might be extrinsic and intrinsic signals that affect the proliferation of the endocrine lineage. Interestingly, signals from mesenchyme might not play a role in endocrine formation - as close proximity of epithelium to the mesenchyme accelerates the exocrine fate, whereas missing contact of epithelium to the mesenchyme leads to the endocrine lineage (Li et al. 2004).

In the 1970s, Pictet and Rutter already postulated that these endocrine progenitors delaminate out of the epithelial sheet and cluster to precursor Islets of Langerhans. This delamination process accompanies with alteration in contact to other cells. In the neuron crest system, delamination is well studied and described as transcriptionally controlled by the Snail family of TF (Sanitarias et al. 2002; Pictet et al. 1972).

In the process itself, Snail family Zinc finger TF are involved and members of the Rho subfamily of GTPases. Recently published by Rukstalis, Snail2/Slug is co-expressed with endocrine progenitor Ngn3 and still maintained in a subset of differentiated endocrine cells (Rukstalis et al. 2006; Rukstalis and Habener 2007). These results link to epithelial and to EMT of endocrine progenitors as they leave the ductal cord. In development, cancer cells and metastases the mechanism of EMT is well studied. In the mechanism of EMT the epithelial cells remodel their polarity as they change from epithelial state to mesenchyme. The characteristic of the epithelium is described as apical-basal (AB) polarity including a polarized actin cytoskeleton. As cells change into a mesenchymal state, polarity changes from AB to front-rear and cell junction remodel as the cell moves out of the epithelial sheet. Main TF of EMT are Snail2, Zinc Finger E-Box Binding Homeobox (Zeb) and Twist family BHLH transcription factor (Twist). Another hallmark of EMT is described as the switch of E-cadherin (Ecad) to N-cadherin. The loosening of the epithelial character includes the down regulation of E-cad. Graphin-Botton had already published that Ngn3 overexpression leads to endocrine differentiation. Further work by Gouzi showed that Ecad is transcriptionally down regulated in endocrine precursors with Snail2 under the control of Ngn3. These results suggest that endocrine progenitors undergo delamination with at least partial EMT. Which mechanism regulates endocrine formation remains elusive and is in focus in the field of diabetes research, as it will help trigger the endocrine commitment in vivo (Johansson et al. 2007; Gouzi et al. 2011).

2.6 Establishment of epithelial asymmetry

Epithelial sheets are the boundary between the extracellular matrix (ECM) and the epithelial tissue itself. The junctional complexes in the membrane surface participate in the establishment and maintenance of the epithelial asymmetry that separates cellular compartments (Rodriguez-Fraticelli et al. 2012). Thus, organization of polarity complexes, the cytoskeleton, vesicle-trafficking and adhesive junctions are coordinated within the cytoplasm and between neighboring cells. (Figure 2.4) (Apodaca et al. 2012).

The adhesion and barrier functions are carried out through Tigth junctions (TJ) and adherens junctions (AJ). In the initiation of polarization mainly junctional adhesion molecules (JAMs), nectin and cadherins define the basolateral surface. A Nectin-Afadin complex mediates via homophilic adhesion receptor domains calcium dependent intercellular adhesion. Thus, they form the first junctional connection to neighboring cells (Figure 2.4). The cytoskeleton illustrates dynamic patterning of actin by the cadherin/ß-catenin complexes, which generates a pool of α-catenin that drives along with p120 (Figure 2.4). Actin-Related Protein 2/3 Homolog (Arp2/3) nucleates actin filaments and establishes an extensive array of branched actin filaments. The interaction to neuronal Wiskott-Aldrich Syndrome Protein (N-WASP), a target of cell division control protein 42 (Cdc42) and Ras-Related C3 Botulinum Toxin Substrate 1 (Rac1), regulates the junctional architecture and cortical tension as demonstrated in vertebrate model systems (Otani et al. 2006). Next, Afadin associates with the cytoplasmic domain of nectins and JAM are recruited to the AJ through an interaction with α-catenin where it directly interacts with actin filaments (Mandai et al. 1997; Tachibana et al. 2000) (Figure 2.4). Thus, Afadin is a major organizer of the apical junctional complex (AJC), and is essential for the development of AB polarity in vertebrate embryogenesis (Zhadanov et al. 1999; Ikeda et al. 1999). In further steps, the partioning defective protein 3 (PAR3), a member of the PAR polarity complex, incorporates with the Nectin-Afadin adhesion complex and initiates at the JAM that the primordial cell contacts.

In the intermediate phase of cellular polarization, PAR3 dissociates and localizes to RAC1, which is followed by subapical association to the Par3-Par6- atypical protein kinase C (αPKC) polarity complex. Besides AJ, TJ are also structurally and functionally linked to the actin cytoskeleton (Figure 2.4). At TJ, TJ proteins (ZO-1, ZO-2, ZO-3) interact with Par3 and stabilize the AB complex through association to actin (Figure 2.4). Cells depleted of ZO-1 show defects in the barrier to larger molecules, and changes in the junctional complexes associated actin, indicating that ZO-1 forms a stabilizing link between the barrier and the junctional actomyosin through PAR3 binding (Li et al. 2005; Wittchen et al. 1999; D’Atri and Citi 2001). In line with these results, depletion of zonula occurrence 2 (ZO-2) does not lead to either actin reorganization or altered plasma membrane permeability, whereas a depletion of either ZO-1 and ZO-2 leads to a drastic expansion of the actomyosin associated with AJ (Nomme et al. 2011; Li et al. 2005). Thus, junctional complexes maintain the AB polarity within the cells and in the epithelial sheet.

The protein network establishing AB polarity includes Crumbs (CRB), partioning defective (PAR, respective Par3-Par6-αPKC) and the Scribble (SCRIB)-lethal (2) giant larvae homologue (DLG) complex. In the initial stage of polarization, it is reported that Par3 forms the primordial cell contacts through interaction with JAM (Ebnet 2003). Par3 dissociates from the junctional primordial complex and forms a ternary complex of Par3 and Par6 linked through αPKC (Lin et al., 2000). Dissociation of Par3 is initiated through the αPKC phosphorylation step, which leads to relocalization of Par3 at TJ and the CRB complex. Simultaneously, the CDC42-Par6- αPKC complex localizes apically, determining the AB polarity in the cell. Thus, the CRB complex defines the apical plasma membrane domain, the PAR complex establishes AB surfaces and the SCRIB-DLG basolateral domains, restricted to epithelial cells (Figure 2.4). As described above, the polarity proteins are ultimately linked to intracellular protein complexes, which make the AJ and TJ (Martín-Belmonte et al. 2008).

Abbildung in dieser Leseprobe nicht enthalten

modified from Belmonte et al.,2012

Figure 2.4: The epithelial polarity program players

The TJ and AJ associate to the PAR complex member PAR3 for establishment of lateral and AB polarity in the initial polarization. Subsequently, establishment and persistence of the polarity within the cell is determined by structural components at the apical and apical-lateral (AL) plasma membrane domains. TJ are represented by TJ protein ZO 1-3 which binds to F-Actin. The AJ are illustrated by Afadin, respective ß-catenin. The intracellular localization of the polarity proteins complexes in a mature AB polarized epithelial cells is achieved through 3 different complexes. The crumbs (CRB) complex (in red), the PAR complex (in blue) and the Scribble (SCRIB)-lethal (2) giant larvae homologue (DLG) complex cooperate to form AJC.

Abbreviations: PAR = partitioning defective protein; PAR3 = partitioning defective protein 3; ZO = zonula occludens; AB = apical-basal; CRB = crumbs; SCRIB = Scribble; DLG = giant larvae homologue; AJC = adherens junction complex; TJ = tight junctions; AJ = adherens junctions; AL = apical-lateral; AJC = adherens junctions complexes.

2.7 The family of Synaptotagmins (Syt)

Synaptotagmins are commonly known as synaptic vesicle membrane proteins. Representative family member Synaptotagmin 1 (Syt1) undergoes calcium-dependent interaction for neurotransmitter release and vesicle exocytosis (Yoshihara & Littleton 2002). Thus, it may characterize the large family of Syts, with at least 19 member in mammals (Adolfsen and Littleton, 2001) . In general, Syts are composed of a Transmembrane domain (TMD) including a preceding N-terminal sequence and 2C2 domains, which are defined as Ca² + binding sites (Ullrich and Südhof 1995; Südhof et al., 2002).

The family can be separated into Ca² + binder and Ca² + non-binder, suggesting that different Syt isoforms have distinct functions. Syt1 thereby triggers the Ca² + release of the soluble trans-soluble N- ethylmaleimide-sensitive-factor attachment receptor (tSNARE) complex. SNARE complexes are membrane trafficking vesicles that establish polarity within the cells. In a developmental context, apical/basolateral SNARE sorting coordinates epithelial morphogenesis (Rodriguez-Fraticelli et al., 2015). Best studied are SNARE-Synaptotagmin complexes in neurons. Syt1 acts as a membrane binding machine likely due to the fact that the C2 domains interact with the plasma membrane. Upon Ca² + binding, the plasma membrane is restrained by the C2 domains and in a further step leads the tension to membrane-fusion and exocytosis of neurotransmitters in neurons and neuroendocrine cells (Figure 2.5).

Similar molecular characteristics of members in the Syt family are shown by Sytnaptotagmin 7 (Syt7). Thus, Syt7 presents a higher affinity to Ca² + and deletion leads to the partial redcution in Ca² + triggered release. Remarkably, deletion of Syt7 along with Syt1 nearly abolishes exocytosis (Wen et al. 2010). Also, Syt7 is expressed in β-cells, α-cells and insulin-responsive tissues as fat. With this in mind, the protein is involved in Ca² + triggered exocytosis of insulin and glucagon and in glucose uptake (Li et al., 2007)

The non-calcium binder in the family of Synaptotagmins are currently in focus due to its diverse function. Until now the isoforms Syt4, 8 and 13 are unlikely able to bind calcium and for Syt4 published data shows broad expression in non-neuronal tissues (Fukuda and Mikoshiba 2001; Tsuboi and Rutter 2003). The gene Syt13 may be regarded as atypical Syt, as the TMD is missing the preceding N-terminal sequence and the 2C2 domains lacks of the Ca² + binding pockets (Fukuda and Mikoshiba 2001; Poser et al., 2001). Moreover, previous results suggest expression in non-neuronal tissue. Therefore, we are interested in expression, function and mechanism of Syt13 in transgenic mouse models.

Abbildung in dieser Leseprobe nicht enthalten

modified from Martens et al.,2007

Figure 2.5: Model of Synaptotagmin 1 promoting membrane fusion

The classical Syt is represented by Syt1. Syt1 is transported to the Ca² + channel by presumably a member of the kinesin motor family. In a docked stage, the synaptic vesicles are primed for the release of the neurotransmitter. This is indicated through the tSNARE complex. Upon Ca² + influx, the C domains frame the plasma membrane, resulting in a buldge pointing towards the vesicle. The close proximity of the membranes leads to fusion and finally to exocytosis of the vesicle content into the channel (17nm pore).

Abbreviations: Syt = Synaptotagmin; tSNARE = trans-soluble N-ethylmaleimide-sensitive-factor attachment receptor

2.8 Aim of this thesis

Foxa2 is an important player in the organizer tissue with protein expression in the embryo already at the early stage E6.5. During gastrulation, the endoderm germ layer forms and covers the egg-shaped embryo past emerging out of the PS. Endodermal-derived organs are the glandular structures of the pharynx, the entire gastrointestinal tract, respiratory tract and associated organs such as the liver and the pancreas. Taking the advantage of the recently in our laboratory generated Foxa2 Venus mouse line, we focused on pancreas organogenesis. By accomplishing an expression profile of the endodermal and non-endodermal tissue compartment during the secondary transition, we wanted to enlargen the knowledge of pancreatic regulatory mechanism and single gene expression in pancreas organogenesis. Thus, we separated both tissue compartments and identified conserved pathways and genes in pancreas development.

In addition, novel pancreatic related genes were selected and analyzed for predicted molecular function and mechanism. In using the digital database Genepaint.org, pancreatic candidate genes had been sorted by in silico in situ hybridization pattern into the different pancreatic lineages. Our interest lies in the identified gene Syt13, as the in situ reflects a trunk pattern, respective endocrine lineage segregation in the secondary phase at E12.5-15.5 in pancreas organogenesis. In generating a knock-in/knock-out mouse line (Syt13 GT /GT) we analyzed the function and molecular mechanism of Syt13. Restricted expression pattern during pancreas organogenesis and subcellular localization lead us to the hypothetical model of Syt13 in endocrine lineage segregation in the secondary transition phase in pancreas organogenesis.

3 Results

3.1 Generation of the Foxa2 Venus mouse line

In the gastrulation of the mouse embryo at E6.5 the three principal germ layers evolve: the ectoderm, the mesoderm and the endoderm (Ichikawa et al., 2013). Subsequently, the family of Fox TF is required for the formation of the DE, as well as the organs which develop out of the epithelial squamous cell sheet lining the egg-shaped embryo. Thus, Foxa2 plays a master role in the formation of endoderm and Foxa2 -/- mice already show lethality at E10-11 with severe defects in node, notochord, neural tube and closure of the gut tube (Burtscher et al., 2009; Uetzmann et al., 2008; Weinstein et al., 1994). Also, the family members Foxa1 and Foxa2 are key TF in metabolism, mutant mice reflect an absent of the liver and Cre mediated deletions leads to hypoglucagonemia, respective hypoglycemia in the adult Islets of Langerhans (Lee at al., 2005). Further analysis of Foxa2 in the pancreas will determine lineage allocation and metabolic impact of the gene in endocrine lineage segregation, clustering of precursor Islet of Langerhans and in the mature Islets of Langerhans. Hence, by using a knock-in strategy to express a Venus reporter protein under the control of Foxa2 expression, we generated in our laboratory the Foxa2 Venus (FVF) mouse line to investigate these processes.

3.1.1 Design and generation of the Foxa2 Venus(VF) targeting vector

For the FVF targeting strategy, we generated a targeting construct and replaced the Stop-Codon in the Exon 3 in the open reading frame (ORF) of the Venus reporter gene, followed by the original Foxa2 3’- untranslated region (3’-UTR) (Figure 3.1, yellow arrow). Downstream of the Venus reporter gene the inserted polyadenylation site (pA) represents a translational stop codon (Figure 3.1, green box). The stop codon insertion followed the phosphor-glycerate kinase (PGK) driven neomycin (neo) resistance gene flanked by the Flippase (Flp) recognition target (FRT) (Figure 3.1, white arrow and red box).

Abbildung in dieser Leseprobe nicht enthalten

modified from Burtscher et al., 2013

Figure 3.1: Targeting strategy of the FVF allele

The WT allele consists of 3 Exons. In the stop codon of Exon 3 the targeting vector with the Venus-tag was fused. The targeting allele can be distinguished from the WT allele by using the Primer EP 397 and 398 for genotyping. Mice carrying the FVF-Neo allele were crossed with Flp-e mice for excision of the PGK driven neo- resistant cassette. Targeting and verification was accomplished by Dr. Ingo Burtscher.

Abbreviations: TSR = transcriptional start site/region; En2 = engrailed homeobox; SA = splice acceptor; pA = SV40 large T gen; neo = neomycin, PGK = phospho-glycerate-kinase; UTR = untranslated region; EP = Primer; FRT = Flipase (Flp) recognition target; Flp-e = enhanced Flipase; FVF = Foxa2-Venus fusion; WT = wild type; loxP = Sitespecific recombinase Cre; bp = base pair.

IDG3.2 embryonic stem (ES) cells were electroporated after linearization of the targeting construct with AscI and neoresistent clones selected upon 300μg/ml G418. The chimeras were generated by diploid aggregation of the recombinant ES cells with CD1 morula (Nagy et al., 2000) and chimerism was scored by coat colour. Upon removal of the neo cassette by using Flp-e (enhanced Flipase) mediated excision (Dymecki et al., 1996) the FVF mice had been mated for eigth generations into the C57Bl/6 background and also on a mixed background including CD1/D57Bl/6/129Sc. The targeting was accomplished by Dr. Ingo Burtscher and Wenke Barkey.

For genotyping of the FVF allele using polymerase chain reaction (PCR) the following primer (EP) were designed to distinguish between the WT allele with 207 base pairs (bp) (EP397, EP398), the FVF Neo allele at 317bp (EP64, EP398) and the FVF delta Neo allele with 506bp (EP38, EP398) (Figure 3.1). For further utilizing the FVF mouse line, heterozygous intercrosses were generated.

3.1.2 Analysis of the Foxa2 Venus mouse line in the pancreas

In the initial stage of pancreas organogenesis at E8.0, the pancreatic primordia buds out into the surrounding mesenchyme. Thereupon, the pancreatic cell population is assigned as MPP until in the secondary transition the MPP segregate into the different lineages (Koop et al., 2011). With this in mind, lineage segregation of the PE at E14.5 may be illustrated by a specified marker lineage analysis. Thus, we were interested in the MPP characteristics of the FVF positive (FVF +) PE and lineage specification within the PE in the secondary transition at E14.5 (Burtscher et al., 2009).

A Tamoxifen inducible promoter of Cpa1 determines lineage segregation of the formerly MPP at E14.5 into the exocrine compartment illustrated by a typical expression pattern in the PE referred to as tip (Zhou et al., 2007). Interestingly, we observed expression of FVF in the PE along with co-expression of Cpa1 in these tip epithelial boundaries (Figure 3.2, A-D). Also, FVF + cells are represented to a larger extend in the PE, all Cpa1+ cells co-localize to FVF+ cells at E14.5 in the tip compartment of the PE (Figure 3.2, C - D; C *). This co-localization not only illustrates cytoplasmic distribution of Cpa1 compared to the nuclear localization of the endogenous FVF protein. In addition, rosette-like structures represent the branching morphogenesis in pancreas organogenesis at E14.5 (Figure 3.2, D‘) (Villasenor et al. 2010). Remarkably, Cpa1+ cells co-express FVFhigh to a lesser extend compared to FVFlow expression within the cells (Figure 3.2, D‘, red arrow CPA1+ and FVF+, yellow arrow CPA1- and FVF+). In summary, our previous results do not only confirm the tip pattern of the formerly MPP in the PE, we characterized the lineage segregation into the exocrine lineage at E14.5, identified 2 subpopulations of endogenous FVF, stated as FVFhigh, respective FVFlow and directly categorized a FVFlowCpa1+ cell population likely as protrusions at the ceiling of the PE.

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.2: FVF and Cpa1 co-localize in the PE

(A-D’) The 4’,6-Diamidin-2-phenylindol (DAPI) counterstain and immunhistochemistry (IHC) against FVF and Cpa1 on a coronal PE section at E14.5 (A-D’). All FVF+ expressing cells in the PE express Cpa1 at the onset of pancreas lineage segregation in the tip region (C, *). Cpa1+ expressing cells are mainly in the tip region of the PE (D). The FVF+ area is broader since it is also expressed within the PE, illustrating the subpopulations FVFhigh and FVFlow (D’, FVFhigh yellow arrow and FVFlow (red arrow) .

Scale is set for 25μm

Abbreviations: DAPI = 4’,6-Diamidin-2-phenylindol; FVF = Foxa2-Venus-Fusion; CPA1 = Carboxypeptidase 1; HIGH. MAGN. = Higher Magnification; PE = pancreatic epithelium; IHC = Immunhistochemistry.

Similar to Cpa1, the paired homeodomain Pax6 is present as early as E9.0 in the dorsal and ventral pancreatic primordia, suggesting a crucial role in pancreas organogenesis. Contrary to Cpa1, Pax6 is required for endocrine cell development and mutations in the gene of PAX6 are associated to diabetic diseases in respect to α-cells (Gosmain et al., 2010; St-Onge et al., 1997; Zorn and Wells, 2011). Hence, we performed further analysis of Pax6 in the secondary transition at E14.5.

Whereas FVF + expressing cells represent the PE, Pax6 + cells are restricted to certain territories combined in clusters in the PE (Figure 3.3, A-C). More precise, the Pax6 + regions are within the PE, compared to the tip pattern of FVF low Cpa1 + expressing compartments. Also, FVFhigh co-localizes to Pax6, these regions suggest to represent the trunk pattern and thereby endocrine cell fate (Figure 3.3, D) (Zhou et al., 2007; Shih et al., 2013). We could observe the direct correlation of FVFlowPax6- (Figure 3.3, D’ red arrow) and FVFhighPax6+ (Figure 3.3, D’ yellow arrow) and thereby define the different subpopulations in the PE at E14.5 in more detail.

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.3: FVF and Pax6 co-localize in the PE

(A-D’) IHC on PE and mesenchyme (DAPI counterstain) at E14.5 on a coronal pancreatic section against FVF and Pax6 at E14.5(A-D’). Pax6 + expressing cells are restricted to the trunk region of the PE (C, *). Contrary, the FVF+ ,compartment represents the PE. Consistently, the PE illustrates the subpopulations FVFhigh and FVFlow (D’, FVFhigh yellow arrow and FVFlow red arrow).

Scale is set for 25μm

Abbreviations: DAPI = 4’,6-Diamidin-2-phenylindol; FVF = Foxa2-Venus-Fusion; Pax6 = paired homeodomain x 6; HIGH. MAGN. = Higher Magnification; PE = pancreatic epithelium; IHC = Immunhistochemistry; E = embryonic stage.

Likewise the endocrine progenitor Pax6, the endocrine precursor Nkx6-1 becomes restricted to the particular endocrine trunk compartment in the secondary transition at E14.5 (Kopp et al., 2011; Pan et al., 2013; Shih et al., 2013). We confirmed the FVF+ expressing cell population, representing the PE and a regionalized Nkx6-1 + cell subpopulation illustrated by a typical trunk pattern (Figure 3.4; A-D; C *). The observation of the FVFhighNkx6-1+, respective FVFlowNkx6-1- verified previous results of endocrine versus exocrine lineage segregation (Figure 3.4, B-D, D red arrow, see also Figure 3.3). Remarkably, different subpopulation characterized by FVFhighNkx6-1- and FVF-Nkx6-1+, (Figure 3.4, D’, FVFhighNkx6- 1- yellow arrow and FVF-Nkx6-1+ black arrow) suggesting a competence windows for the generation of the different endocrine cell types along with different protein expression levels of FVF and Nkx6-1.

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.4: FVF and Nkx6-1 co-localize in the PE

(A-D’) In the PE at E14.5 on coronal sections IHC illustrates FVF and Nkx6-1 (DAPI counterstain). The PE is marked with a FVF + expressing cell compartment, Nkx6-1 + expressing cells are restricted to the trunk compartment of the PE (C *). Interestingly, the PE illustrates the subpopulations FVFhighNkx6-1- and FVFhighNkx6-1, (D’, FVFhighNkx6-1- yellow arrow, FVFhighNkx6-1+ red arrow and FVF-Nkx6-1- yellow arrow).

Scale is set for 25μm

Abbreviations: DAPI = 4’,6-Diamidin-2-phenylindol; FVF = Foxa2-Venus-Fusion; NKX6-1 = NK homeobox 6-1; HIGH. MAGN. = Higher Magnification; PE = pancreatic epithelium; IHC = Immunhistochemistry; E = embryonic stage.

In the next step, we determined the regionalized compartmentalization of the ductal and endocrine lineage commitment in the bi-potent trunk pattern at E14.5. The gene Hnf1 β consequently forms the duct in the secondary transition and on the mRNA level in the regulatory transcriptional network that regulates Ngn3 expression (Solar et al., 2009; Oliver-Krasinski et al., 2009). Thus, in the FVF+ PE, HNF1 β + cells are expressed in the trunk pattern (Figure 3.5, C *) - interestingly, our observation suggesting rather a FVFlowHNF1β+ subpopulation as the expected FVFhighHNF1β+ expressing cells (Figure 3.5, A-D). The co-localization of FVFlowHNF1ß+ characterizes the duct, endocrine lineage segregation of the formerly bi-potent precursors is marked through FVFhighHNF1β- protein levels (Figure 3.5, D’, FVFlowHNF1ß+ yellow arrow and FVFhighHNF1- red arrow).

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.5: FVF and Hnf1β co-localize in the PE

(A-D’) In coronal sections at E14.5, the PE is represented by FVF IHC, respective a regionalized trunk compartment by HNF1β IHC (DAPI counterstain). The HNF1β+ cell population co-localizes to the FVF+ cell population (A-D; C *). It is to note, that the FVFlow cells are characterized through HNF1β+ regionalized expressing cells. On that account, FVFhigh cells are HNF1β- (D; D’, FVFlow HNF1β+ yellow arrow and FVFhigh HNF1β- red arrow).

Scale is set for 25μm

Abbreviations: DAPI = 4’,6-Diamidin-2-phenylindol; FVF = Foxa2-Venus-Fusion; HNF1β = HNF1 Homeobox β; HIGH. MAGN. = Higher Magnification; PE = pancreatic epithelium; IHC = Immunhistochemistry; E = embryonic stage.

We further focused on the ductal compartment in respect to Sox9. The Sox9 gene expression is excluded from the MPP cells at the beginning of the secondary transition and localized in the trunk pattern (Seymour et al., 2007). Notably, we confirmed previous results of the FVF + PE, FVF low Sox9 + and FVF high Sox9 - and FVF low Sox9 - cell populations in the PE at E14.5 (Figure 3.6, A-D, C *). The co-localization of Sox9 and FVFlow is restricted to the cord-like structure, defining the common progenitor pool of the formerly bi-potent trunk precursors (Figure 3.6, D’ red arrow). Thus, the endocrine committed cells are represented by FVF high Sox9 - expression (Figure 3.6, D’ yellow arrow; Figure 3.1-3.5).

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.6: Figure 3.5: FVF and Sox9 co-localize in the PE

(A-D’) Repetitive coronal sections at E14.5 of the PE and IHC of FVF and Sox9 identified the PE with a FVF + cell population and the trunk compartment as a Sox9 + cell population (DAPI counterstain) (A-D, C *). The Sox9+ cells co-localizes to the FVFlow cell population (A-D, C *, D’). Contrary, FVFhigh does not contribute to Sox9+ PE regions (D; D’, FVFlowSox9+ red arrow and FVFhighSox9- yellow arrow).

Scale is set for 25μm

Abbreviations: DAPI = 4’,6-Diamidin-2-phenylindol; FVF = Foxa2-Venus-Fusion; Sox9 = SRY related gene 9; HIGH. MAGN. = Higher Magnification; PE = pancreatic epithelium; IHC = Immunhistochemistry; E = embryonic stage.

Previous results suggest that the FVF high cell subpopulation segregates into the endocrine lineage, also we were interested at which E the MPP of the PE will become restricted, mainly described at FVF expression levels. Our first observation indicated lineage segregation of FVF between E18.5 and P1, the formerly FVF+Ecad+ PE at P1 is specified as FVF- Ecad + PE (Figure 3.7, A-B and D). As opposed to this, the Pdx1 + cell population illustrates FVF+ Ecad - and more detailed FVF high Ecad - Pdx1 low expression, respective a FVF low Ecad - Pdx1 high cell subpopulations (Figure 3.7, B-D; D’, FVFhighEcad-Pdx1low red arrow and FVFlowEcad-Pdx1high yellow arrow). At E18.5 the FVF high expressing cell population lining the outer edge of the precursor Islets of Langerhans, suggesting that this progenitor cell pool will give rise to α- cells (Figure 3.7, E18.5, B; D’, FVFhighEcad-Pdx1low red arrow). Contrary, Pdx1high marked cells are expressed in the inner perimeter indicating β-cell commitment (Figure 3.7, E18.5, B; D’, FVFlowEcad- Pdx1high yellow arrow) (Willmann et al., 2016; Wang et al., 2010; Wang et al., 2005). In later E at P1, precursor Islets of Langerhans mature as indicated in regionalized FVF high expression, respective Pdx1 high and the PE will become compartmentalized into the different lineages (Figure 3.7, P1, D’, FVFhighEcad-Pdx1low red arrow and FVFlowEcad-Pdx1high yellow arrow). Also in the adult Islets of Langerhans FVF co-localized to Pdx1 with a slight number of FVF high cell population which may represent the α-cells (Figure 3.7, P36, D’, FVFhighEcad-Pdx1low red arrow and FVFlowEcad-Pdx1high yellow arrow).

Abbildung in dieser Leseprobe nicht enthalten

Figure 3.7: FVF expression after secondary transition correlates to the endocrine lineage commitment.

(A-D’) Consecutive pancreatic coronal sections and IHC after secondary transition (E15.5, E18.5, P1 and P36) identified lineage restriction of the formerly MPP PE (DAPI for counterstain). The FVF + expression will become restricted to regionalized compartments in the PE (E15.5 - P36, A-B). Thus, the PE is characterized through Ecad; contrary the endocrine progenitors, precursors and mature endocrine cells are illustrated by Pdx1 (E15.5 - P36, C; E15.5 - P36, D). The different subpopulations of FVF anticorrelate to Pdx1, by means FVF high cells are Pdx1 -, whereas the FVF low cell population represents the Pdx1 + cell pool (E15.5 - P36, D’, FVFhigh red arrow and Pdx1+ yellow arrow).

Scale is set for 50μm

Abbreviations: DAPI = 4’,6-Diamidin-2-phenylindol; IHC = Immunhistochemistry; FVF = Foxa2-Venus-Fusion; Pdx1 = pancreatic and duodenal homeobox 1; E = embryonic stage; PE = pancreatic epithelium, P = postnatal, Ecad = E-cadherin, MPP = multipotent pancreatic progenitors.

Taken together, this analysis reveals that FVF co-localizes to the MPP in the secondary transition at E14.5. Interestingly, PE patterning and FVF levels indicate the different lineages in pancreas organogenesis. Thus, the FVFlow cell compartment is characterized through a Cpa1 + Hnf1 β + Sox9 + Ecad + marker onset likely describing the exocrine and ductal progenitor pool. On the other hand, endocrine lineage segregation in the PE at E14.5 is illustrated by Nkx6-1 + Pax6 + Ecad - expressing cells. Although the subpopulations of FVF follows different expression cascades, our results emphasize the FVF + cell population within the PE and thus utilizing the FVF knock-in reporter mouse for a global gene expression profile of the PE and surrounding tissue in the secondary transition.

3.2 Genome-wide expression profile of the pancreas in the secondary transition

The TF factor Foxa2 is the key regulator of early development, previous work in Heiko Lickerts laboratory (Tamplin et al., 2008) identified in an in situ hybridization screen novel nodal genes selected by their specific mRNA expression in the organizer region in the node at E7.5. These first results indicated for two of the selected genes, Pitchfork (Pifo) and Flattop (fltp), a role in left-right asymmetry, cilium disassembly and basal body (BB) docking (Tamplin et al., 2008; Gegg et al., 2014; Kinzel et al., 2010). Thus, in the global gene expression profile of FVF + PE and FVF — non-endodermal derived tissue we aimed to characterize regulatory and functional genes in the secondary transition of the pancreas by utilizing the FVF knock-in reporter mouse. (Burtscher and Lickert, 2013).

In general, different global gene expression analysis of the pancreas focuses on the endocrine lineage, although exocrine lineages determine pancreatic endocrine fate in the first secondary transition phase. This is illustrated by knockdowns of exocrine determined TF as Ptf1 α, Cpa1, Nr5a2 which will led to hypoplasia and/ or failure in pancreas organogenesis (Kawaguchi et al., 2002; Krapp et al., 1996; Zhou et al., 2007). Hence, we analyzed tissue interactions in the secondary transition and identified known and yet functionally not described pancreatic genes, metabolic pathways and dissected classical EMT factors in the PE and non-endodermal derived tissue compartment.

3.2.1 Bioinformatic analysis of pathways in the secondary transition

For generating a global transcriptional expression profile during secondary transition, we dissected the pancreata of endodermal (FVF +) and non-endodermal (FVF -) derived tissue compartment at consecutive stages from E12.5 until E15.5 of FVF mice, respective WT. Analysis of lineage allocations is mandatory during secondary transition, in earlier E established endocrine progenitors will be depleted (Heller et al. 2004) and from E16.5 onwards segregation into precursor Islets of Langerhans indicates maturation (Gradwohl et al. 2000; Herrera et al., 2000). Thus, the total RNA of three biological replicates for each stage of two the distinct cell populations, the pancreatic epithelium (FVF +) and the FVF - population, was used to generate the gene regulatory network (GRN) of the pancreas in the secondary transition. The application of the Affymetrix® array card (GeneChip® Gene 1.0 ST Array Card), subsequent processing of the probe set (GeneChip® Scanner 3000 7G Whole-Genome Association System) lead to the probe set outcome. The Affymetrix® Expression console normalized the robust multichip-analyses (RMA) on gene-level (ratio >200 compared endodermal and non- endodermal). Comprehensive R based microarray analyses web frontend (CARMAweb) inducted genewise testing by Limma t-test and Benjamini-Hochberg multiple testing correction with a false discovery rate (FDR<5%) offered statistically significant differential expression of 2921 probe sets. The statistical Principal component (PC) analysis thereby linearized the different E for both tissue compartments. The first PC (PC1) shows the spatial distant difference between the endodermal, respective non-endodermal tissue compartment with a variance of 68.7% variance Whereas PC2 describes the different E with a total variance of 13.3%. (See Figure 3.8 B) (Willmann et al., 2016).

Abbildung in dieser Leseprobe nicht enthalten

modified from Willmann et al., 2016

Figure 3.8: Experimental setup for the GRN.

(A) The GenePaint in silico in situ demonstrates mRNA expression of Foxa2 at E14.5, the black box indicates the pancreatic region in the whole mount embryo. The FVF + cell population represents the PE, contrary the FVF - cell population is marked through DAPI IHC at E14.5. After pancreas dissection and single cell suspension, fluorescent-activated cell sorting (FACs) sorting separated the endodermal (FVF + ) of the non-endodermal (FVF - ) cell compartment. Global transcriptional profiling was accomplished by Affymetrix® GeneChip® Gene 1.0 ST Array Card and statistical analyses through the Affymetrix® Expression console.

(B) The Affymetrix® Expression console was used for statistical analysis, thereby using PC analysis separated the different E. In taking triplicates of the RNA samples, each sample in the different embryonic stage was referenced against each other. The results are blotted with variances between 68.7% (PC1) and 13.3% (PC2).

Abbreviations: RNA = ribo nucleotide acid; E = embryonic stage; FACs = fluorescent activated cell sorting; PC = principal component; DAPI = 4’,6-Diamidin-2-phenylindol; PE = pancreatic epithelium; IHC = immunhistochemistry; FVF+ = Foxa2-Venus-Fusion positive; FVF- = Foxa2-Venus-Fusion negative.

First of all, we determined the quantity of dysregulated genes at each E for both tissue compartments (Figure 3.9, A). In total, 7346 genes at E14.5 are differentially expressed, indicating increasing morphogenetic alterations within the PE (Pan and Wright, 2011a; Pictet et al., 1972). Due to lower sample sizes and less concordant ribonucleotide acid (RNA) in the FVF - at E13.5, the number of significantly regulated genes was reduced by multiple testing corrections (Willmann et al., 2016).

Abbildung in dieser Leseprobe nicht enthalten

modified from Willmann et al., 2016

Figure 3.9 : The GRN in the pathway analysis.

(A) The Histogramm describes the quantity of dysregulated genes of each developmental stage (p-value < 0.01). At E13.5 sample size represent duplicates (*, n=2); E12.5 - E15.5 are represented in triplictes (n=3, exceptian E13.5 n=2).

(B) The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analsis was performed with indicated hierarchical pathway enrichments in the different E of the secondary transition. Significant differences in FVF + to FVF - compartment are color-coded (red color-code illustrates for the terms high dysregulated number of genes in comparison of both tissue populations, respectively (log10(p)<1 with p<0.1)).

Abbreviations: n = number; E = embryonic stage; KEEG = Kyoto Encyclopedia of Genes and Genomes; E = embryonic stage; log = logarithm; FVF+ = Foxa2-Venus-Fusion positive; FVF- = Foxa2-Venus-Fusion negative; ECM = extracellular matrix; tRNA = transfer ribo nucleid acid; No/n = number.

The endodermal population shows discrepancy compared to the non-endodermal population in metabolic pathways mainly due to diverse assignments in both tissue compartments. Functional pathway analysis was performed by using the database resource Kyoto encyclopedia of genes and genome (KEGG). We wanted to achieve a more comprehensive understanding of high-level functions and pathways which differ in both populations (Figure 3.9 B). The key pathway as maturity onset diabetes of the young (MODY) and type 2 diabetes (T2D) are upregulated, indicating the ongoing organogenesis of the pancreas. Interestingly, ECM-receptor interaction (ECM) and cell adhesion molecules (CAM) are highly deregulated. This result is in line with already published data by Kesavan for Laminin-1 inhibition, which leads to failure in branching morphogenesis and differentiation. It may reflect the importance of ECM and mesenchyme for proper proliferation of the pancreas during secondary transition (Kesavan et al., 2014). Mitogen-activated protein kinases (MAPK kinases) indicate a role in lineage formation by MAPK/ERK pathway and transcriptional activation of lineage specific factors. Recently published by Morris might suggest for the pathway regulation of actin cytoskeleton and AJ/TJ remodeling of the epithelial polarity sheet as the endocrine lineage establishes (Morris et al., 2015). Which mechanism as classical EMT or delamination occurs, is still controversial. More interestingly, axon guidance within the epithelial population is dysregulated, indicating the correlation of the gene activity program of endocrine cells in neural tissue and in the pancreas. Our results confirm published data by Arensbergen, in which insulin-producing β-cells adopt a neural gene activity program by repressing the Polycomb complex (Van Arensbergen et al., 2010). Thus, highlighting again the close relationship of neuronal and pancreatic derived progenitors. One more point to mention might be the pathways for amino-acid related biosynthesis and the metabolic pathway which are regulated from E14.5 onwards. This may implicate ongoing differentiation of the different lineages and expansion and branching of the pancreas. As the KEGG pathway analysis only offers pathways and does not reflect single gene expression in a spatial-temporal manner, in the next step we wanted to elucidate specific gene expression (Willmann et al., 2016).

3.2.2 Bioinformatic analysis of genes in the secondary transition

The KEEG pathway analysis illustrates dysregulated pathways in correlation between the FVF + - and the FVF --cell population, respective endodermal versus non-endodermal tissue compartment. In the next step, the GRN was utilized for deciphering single gene expression of a selected subset of genes. Therefore, the selected genes were arranged in clusters due to their expression profile in terms of the FVF + - and the FVF - cell compartment, spatio-temporal separated in the E of the secondary transition (Willmann et al., 2016).

[...]