The Role of Cdep in the Embryonic Morphogenesis of Drosophila Melanogaster

The Role of Cdep in the Embryonic Morphogenesis of Drosophila melanogaster

Dissertation

zur Erlangung des akademischen Grades Doctor rerum naturalium (Dr. rer.nat.)

vorgelegt

der Fakultät Mathematik und Naturwissenschaften der Technischen Universität Dresden

von

Anne Morbach Geboren am 05. August 1984 in Nordhausen, Deutschland

Eingereicht am 30. Oktober 2015

Die Dissertation wurde in der Zeit von April 2012 bis Oktober 2015 im Max-Planck-Institut für molekulare Zellbiologie und Genetik angefertigt Gutachter:

Prof. Dr. Elisabeth Knust

Prof. Dr. Christian Dahmann Now, here, you see, it takes all the running you can do, to keep in the same place.

Lewis Carroll, Through the Looking-Glass

I Contents

List of Figures V

List of Tables VII

Abstract IX

List of Abbreviations XI

1 Introduction 1 1.1 Epithelial cell polarity ...... 1 1.1.1 Cellularization and formation of the primary epithelium ...... 1 1.1.1.1 Establishment of epithelial polarity and adhesion ...... 2 1.1.2 The epithelial polarity network ...... 3 1.1.3 Cell-cell adhesion ...... 5 1.1.3.1 Adherens junctions ...... 5 1.1.3.2 Septate junctions ...... 6 1.2 Epithelial movements in Drosophila embryonic morphogenesis ...... 6 1.2.1 Epithelial tube formation during Drosophila embryogenesis . . . . . 7 1.2.2 Coordinated migration of epithelial sheets during Dros. embryogenesis 7 1.2.2.1 FERM domain proteins in epithelial migration ...... 9 1.2.2.2 Cdep ...... 10 1.3 Mutagenesis with the CRISPR/Cas9 system ...... 12

2 Aim of My PhD Thesis Work 15

3 Preliminary Work 17

4 Results 19 4.1 A screen for novel regulators in Drosophila embryonic morphogenesis . . . . 19 4.1.1 Deficiencies on the left arm of chromosome 3 cause defects in SG lumen morphology ...... 19 4.1.2 A locus in the overlap of two deficiencies on the right arm of chromosome 3 causes defects in Drosophila embryonic dorsal closure 20 4.2 Two uncharacterized genes regulate SG lumen diameter in Drosophila embryos 22 4.2.1 Mutations in two uncharacterized genes on chromosome 3 cause intermittent tube closure in the Drosophila SG ...... 22 4.2.2 CG17362/ CG9040/ CG17364 play a role in the maintenance of SG lumen width after lumen expansion ...... 24 4.2.3 CG17362 is exclusively expressed in the embryonic SG ...... 25 4.2.4 CG17362 and CG9040 do not contain known protein domains and are only conserved in Drosophilidae ...... 28 4.3 Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 30 4.3.1 Insertions in the ORF of Cdep cause defects in DC and HI . . . . . 30 4.3.2 Embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC} CdepMI00496 show defects in the LE during DC ...... 34 4.3.3 Insertions in the ORF of Cdep cause defects in tracheal and Malpighian tubule morphogenesis ...... 34 4.3.4 The CRISPR/Cas9 system was used to generate loss-of-function mutants of Cdep ...... 35 4.3.5 LOF mutants of Cdep show a variety of phenotypes ...... 37 II Contents

4.3.6 LOF mutants of Cdep cause denticle belt defects and ventral holes in Drosophila larval cuticles ...... 38 4.3.7 Phenotypes in Cdep−/− mutant embryos are likely not caused by maternal defects ...... 44 4.3.8 LOF mutants of Cdep cause segments to fuse ...... 44 4.3.9 Defects in Cdep−/− mutants are not due to actin mislocalization . . 51 4.3.10 Cdep genetically interacts with yurt ...... 51 4.3.11 Cdep genetically interacts with cora ...... 55

5 Discussion 57 5.1 The role of CG17362 and CG9040 in SG lumen stability ...... 57 5.1.1 Impairments in the expression of CG17362 and CG9040 could be the cause for the intermittent tube closure in the embryonic SGs . . . . 57 5.1.2 CG17362 and CG9040 could be necessary for SG lumen dilation and stability of lumen diameter ...... 57 5.2 The role of Cdep in Drosophila embryonic morphogenesis ...... 59 5.2.1 Cdep is a regulator of the JNK pathway ...... 60 5.2.1.1 Mutations in TGF-β pathway components cause LE bunching and gaps in the tracheal dorsal trunk ...... 60 5.2.1.2 The JNK pathway, like Cdep, is instrumental in GBR, DC and HI ...... 61 5.2.1.3 Mouse Farp2 acts upstream of the JNK pathway ...... 61 5.2.2 Does Cdep act as a GEF? ...... 62 5.2.3 Cdep might regulate epithelial migration in parallel to Yrt and Cora 63 5.3 Conclusion and Outlook ...... 64

6 Materials and Methods 67 6.1 Cell strains, plasmids and DNA constructs ...... 67 6.2 Culture media ...... 68 6.2.1 Lysogeny Broth (Bertani, 1951) ...... 68 6.2.2 Super Optimal Catabolite repression (SOC) medium (Hanahan, 1983) 68 6.3 Molecular biology methods ...... 69 6.3.1 Amplifying DNA by standard PCR technology ...... 69 6.3.2 Molecular cloning ...... 70 6.3.2.1 Cloning with restriction endonucleases (Old and Primrose, 1980) ...... 70 6.3.2.2 Gibson Assembly (Gibson et al., 2009) ...... 70 6.3.3 Detecting DNA via agarose gel electrophoresis ...... 71 6.3.4 Purifying DNA from E.coli ...... 71 6.3.5 Extracting DNA from Drosophila adults ...... 71 6.3.5.1 Extracting mRNA from Drosophila embryos and reverse transcription into cDNA ...... 72 6.3.6 Making mRNA probes for in situ hybridization ...... 72 6.3.7 Measuring DNA concentrations using the NanoDrop ...... 72 6.3.8 DNA sequencing ...... 72 6.3.9 Transformation ...... 73 6.3.10 Mutagenesis of Cdep with the CRISPR-Cas9 system ...... 73 6.3.10.1 CRISPR/Cas9 tools for mutagenesis in D.melanogaster . . 73 Contents III

6.3.11 Strategies for CRISPR/Cas9 mutagensis ...... 74 6.3.11.1 vectors and oligomers used for mutagenesis of Cdep via CRISPR-Cas9 system ...... 75 pCFD3-dU6:3-CdepEx2gRNA ...... 75 CdepSTOP-ssODN ...... 76 pCFD4-CdepExc_2sgRNAs ...... 77 pDsRed-5’HA-3’HA ...... 78 6.3.11.2 Screening for successful mutagenesis events in flies modified with the CRISPR/Cas9 system ...... 79 Insertion of an in-frame stop codon into the Cdep ORF . . . 79 Replacing the entire Cdep locus with dsRed ...... 80 6.3.12 Extracting protein from Drosophila embryos ...... 83 6.3.13 Detecting and separating proteins by mass via SDS-PAGE ...... 83 6.3.14 Detection of specific polypeptides by Western blot analysis (Towbin et al., 1979) ...... 84 6.3.14.1 Transferring polypeptides to nitrocellulose membrane . . . 84 6.3.14.2 Detecting specific polypeptides via immonublotting . . . . 85 6.4 Online tools for prediction of protein domains, interactions, sequence alignments, etc...... 86 6.5 Cell culture growth and harvest ...... 86 6.5.1 Growing E.coli cells for vector amplification ...... 86 6.5.2 Cell harvest ...... 86 6.6 Work with Drosophila melanogaster ...... 87 6.6.1 Fly stocks used: deficiencies, alleles, duplications and balancers . . . 88 6.6.2 Recombining alleles located on the same chromosome ...... 93 6.6.3 Collecting Drosophila melanogaster embryos ...... 93 6.6.4 Histochemistry ...... 94 6.6.4.1 Embryo dechorionation ...... 94 6.6.4.2 Embryo fixation for light microscopy of whole specimens . 94 Heat fixation of Drosophila embryos ...... 94 Formaldehyde fixation of Drosophila embryos ...... 94 6.6.4.3 Embryo fixation for light microscopy of semi-thin sections (Tepass and Hartenstein, 1994) ...... 95 6.6.4.4 Antibody staining of embryos for Immunofluorescence . . . 95 6.6.4.5 Phalloidin and DAPI staining of embryos ...... 96 6.6.4.6 Embryo mounting for analysis via fluorescence microscopy 96 6.6.4.7 Embryo mounting for live imaging ...... 97 6.6.4.8 Embedding of embryos for semi-thin sectioning ...... 97 6.6.4.9 Cuticle preparation from hatched and unhatched Drosophila larvae ...... 97 6.6.4.10 Preparation and staining of ovarian follicles from Drosophila females ...... 98 6.6.4.11 mRNA in situ hybridization of whole-mount Drosophila embryos ...... 99

Bibliography XV

Acknowledgements XXIX

Erklärung entsprechend §5.5 der Promotionsordnung XXXI

V List of Figures

1.1 Schematic representation of Drosophila cellularization ...... 2 1.2 The cell polarity network in insects ...... 4 1.3 Cell adhesions in insects ...... 6 1.4 Schematic overview over morphogenesis of tubular organs in the Drosophila embryo...... 8 1.5 Schematic overview over germ band retraction, dorsal closure and head involution ...... 9 1.6 Schematic overview over the type II CRISPR/Cas9 system ...... 12 1.7 Schematic overview over the modiﬁed CRISPR/Cas9 system for use in eukaryotes ...... 13 1.8 Example for indels in Drosophila white gene caused by CRISPR/Cas9-induced mutagenesis ...... 13 1.9 Schematic overview over strategy to replace an endogenous gene with a marker gene via CRISPR/Cas9 ...... 14

4.1 Overview over deficiencies used to find salivary gland phenotypes...... 20 4.2 Examples of salivary gland phenotypes caused by deficiencies ...... 21 4.3 Salivary gland phenotypes found in embryos homozygous for CG17364MB01315 and CG17362MB04765...... 21 4.4 Dorsal closure phenotype in embryos transheterozygous for Df(3R)ED5066 and Df(3R)Exel6143...... 23 4.5 Localization of Df(3R)ED5066 and Df(3R)Exel6143 in the Drosophila genome 23 4.6 Overlap of Df(3L)ED4284 with Df(3L)ED4287 and Df(3L)ED4515 with Df(3L)ED4502...... 24 4.7 Localization of the genomic insertions CG17364MB01315 and CG17362MB04765. 25 4.8 In embryos transheterozygous for CG17364MB01315 and Df(3L)ED4515, the SG lumen collapses during stage 17 ...... 27 4.9 In situ staining of CG17362 mRNA in WT embryos...... 27 4.10 Overlap between Df(3R)ED5066 and Df(3R)Exel6143 ...... 31 4.11 Overview over the Cdep locus with insertion sites of PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 ...... 31 4.12 Dorsal closure phenotypes in embryos homozygous or transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496...... 32 4.13 Overview over the Cdep locus ...... 33 4.14 Defects in leading edge cells in embryos transheterozygous for PBac{5HPw+} CdepB122 and Mi{MIC}CdepMI00496...... 33 4.15 Defects in the trachea and Malpighian tubules in embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496...... 36 4.16 Overview over Cdep alelles created with the CRISPR/Cas9 system . . . . . 37 4.17 Overview over morphological defects in Cdep alelles ...... 39 4.18 Plotted ratio of Cdep−/− embryos showing morphological defects ...... 39 4.19 Plotted ratio of Cdep−/− larvae hatching from eggs ...... 40 4.20 Morphological defects observed in cuticles of Cdep−/− embryos and larvae 42 4.21 Plotted ratio of embryonic and larval Cdep−/− cuticles showing morphological defects ...... 43 4.22 Ovarioles from Cdep−/− females...... 45 4.23 Plotted ratio of cuticle phenotypes from animals heterozygous for Cdep . . 45 4.24 Still frames from live imaging of Cdep−/− embryos ...... 48 4.25 Still frames of details of tracheal and epidermal defects seen in live imaging of Cdep−/− embryos ...... 49 VI List of Figures

4.26 Semi-thin sections of Cdep−/− embryos stained with methylene blue and tolouidine blue...... 50 4.27 Sas and actin staining in the lateral epidermis of Cdep−/− embryos. . . . . 52 4.28 Actin staining in muscles of stage 17 Cdep−/− embryos...... 52 4.29 Agarose gel showing products of BglII digest of PCR product ampliﬁed from exon 2 of CdepE16X , yrt75a and CdepG17X , yrt75a...... 53 4.30 Cuticle preparations of yrt75a embryos and larvae...... 54 4.31 Plot of ratio of Cdep−/−, yrt75a cuticles showing morphological defects . . . 55 4.32 Cuticle preparations of cora5 embryos...... 56 4.33 Plotted ratio of cora5; CdepG17X and cora5; Cdep∆ embryonic cuticles showing morphological defects ...... 56

5.1 Possible roles of CG17362 and CG9040 in SG aECM ...... 59 5.2 Possible mode of action for Cdep in epithelial morphogenesis in the Drosophila embryo ...... 65

6.1 Schematic representation of Gibson Assembly method ...... 71 6.2 pCFD plasmids for delivery of sgRNAs and pHD-DsRed-attP for homologous recombination in Drosophila melanogaster...... 74 6.3 Schematic representation of strategy to insert in-frame stop codon into Cdep ORF...... 76 6.4 Cloning strategy for pCFD4-CdepExc_2sgRNAs ...... 77 6.5 Schematic representation of strategy to replace Cdep ORF with DsRed . . . 78 6.6 Schematic representation of pDsRed-5’HA-3’HA...... 79 6.7 Workflow for crossing and screening flies for succesful insertion of stop codon in Cdep ORF ...... 79 6.8 Example agarose gels with BglII digestion products...... 80 6.9 Workflow for crossing and screening flies for succesful replacement of Cdep ORF with dsRed ...... 81 VII List of Tables

4.1 Genes expressed during Drosophila embryogenesis and aﬀected in Df(3L)ED4284 and Df(3L)ED4515 ...... 26 4.2 Putative interactors of CG17362 and CG9040 proteins, predicted by STRING database ...... 29

6.1 E.coli strains used ...... 67 6.2 Plasmids and vectors used in this thesis ...... 67 6.3 Vectors containing Cdep or parts thereof used for gene modiﬁcation . . . . . 68 6.4 Assay conditions for typical PCR reaction ...... 69 6.5 Typical PCR program ...... 70 6.6 primers used in CRISPR mutagenesis of Cdep ...... 82 6.7 Fly stocks used in this thesis ...... 88

IX Abstract

Many organs and structures formed during the embryonic morphogenesis of animals derive from epithelia. Epithelia are made up of apicobasally polarized cells which adhere to and communicate with each other, allowing for epithelial integrity and plasticity. During embryonic morphogenesis, epithelia change their shape and migrate in a coordinated manner. How these epithelial processes are regulated is still not fully understood. In a forward genetic screen using the embryo of the fruit fly Drosophila melanogaster, candidate genes influencing the morphogenesis of epithelial structures were identified. Three genes, CG17364, CG17362 and CG9040 were identified as possible regulators of lumen stability in the salivary glands, tubular organs deriving from the embryonic epithelium. Furthermore, the gene Cdep was found to play a crucial role in epithelial sheet migration during dorsal closure of the embryo. Embryos carrying genomic insertions that could affect the expression of CG17364, CG17362 and CG9040 show a luminal penotype of the embryonic salivary glands characterized by alternating bloated and seemingly closed sections. Therefore, one of these genes or a combination of them likely plays a role in stabilizing the salivary gland lumen. However, neither CG17364 nor CG17362 or CG9040 contain any known protein domains, hence their molecular roles remain unknown. Cdep (Chondrocyte-derived ezrin-like protein) is a member of the FERM-FA subclass of proteins. Proteins of the FERM family have been shown to interact with the plasma membrane and membrane-bound proteins as well as cytoskeleton components. Accordingly, they have been implicated in stabilizing the cell cortex, and some of them are involved in signal transduction mechanisms. In addition to a FERM domain, Cdep also contains a RhoGEF domain, although is still not clear whether it actually exerts GEF activity. Genomic insertions in the Cdep locus cause defects in embryonic dorsal closure and atypical migratory behaviour in epithelial tubes. In order to study the molecular role of Cdep, the CRISPR/Cas9 system was employed to establish loss-of-function mutants of Cdep. The mutants show aberrations in germ band retraction, dorsal closure and head involution. Moreover, I found that two mutants carrying a premature STOP codon in the Cdep ORF, CdepE16X and CdepG17X , rescue the defects observed in embryonic cuticles mutant for two other FERM-FA members yurt (yrt) and coracle (cora). A deletion of the full Cdep ORF did not rescue those defects. I hypothesize that CdepE16X and CdepG17X encode Cdep variants with increased activity, which compensates for the loss of yrt or cora function, respectively. In conclusion, this leads to a model in which Cdep acts in parallel to Yrt and Cora during Drosophila embryonic morphogenesis. Many of the defects described in this study are reminiscent of phenotypes found in embryos mutant for components and downstream effectors of the Jun-N-terminal Kinase (JNK) pathway. Hence, my work supports an earlier hypothesis according to which a mouse homologue of Cdep, Farp2, acts as an upstream activator of the JNK pathway during epithelial cell migration in vitro (Miyamoto et al., 2003). The data provided here shows that Cdep plays a role in the morphogenesis of a great number of epithelia-derived organs and structures in vivo. My study therefore elucidates a missing link between cell migration cues and JNK pathway activation.

XI List of Abbreviations

A abdominal segment aa amino acid Ab antibody aECM apical extracellular matrix Amp ampicillin AJ adherens junction aPKC atypical Protein kinase C Arm Armadillo (Drosophila β-catenin) AS amnioserosa Baz Bazooka (Drosophila Par3) bp base pair(s) BSA bovine serum albumine Bsk Basket cad cadherin cAMP cyclic adenosine monophosphate cat catenin Cdc Cell division cycle Cdep Chondrocyte-derived ezrin-like protein Chrb Charibde CIP calf intestine phosphatase Cora Coracle Crb Crumbs CREB cAMP response element binding protein CrebA cAMP response element binding protein A C-terminus carboxyterminus D. Drosophila db denticle belts DC dorsal closure dCBP Drosophila CREB-binding protein dH2O distilled water Dlg Discs Large DNA desoxyribonucleic acid Dpp Decapentaplegic ds double-stranded dt dorsal trunk DTT Dithiothreitol, ((2S,3S)-1,4-Bis-sulfanylbutane-2,3-diol) EDTA 2-((2-(bis(carboxymethyl)amino)ethyl)(carboxymethyl)amino)acetic acid E.coli Escherichia coli FA FERM-adjacent Farp FERM, RhoGEF, and pleckstrin homology domain protein FC furrow canal FERM 4.1, Ezrin, Radixin, Moesin FGF fibroblast growth factor fk filzkörper g gram GBE germ band extension GBR germ band retraction h hour Hid Head involution defective hs head skeleton XII List of Abbreviations i.e. Latin id est, "that is" JNK Jun-N-terminal kinase kDa kilodalton l liters LB lysogeny broth lb labium LE leading edge Lgl Lethal giant larvae Mad Mothers against Dpp MAPK Mitogen-activated protein kinase md mandible µg microgram µl microliter µm micrometer µM micromolar mg milligram ml milliliter mm millimeter mM milli molar MCS multiple cloning site MES 2-(N-morpholino)ethanesulfonic acid Mod Modulo Moe Moesin µ micro MW molecular weight mx maxilla Myo Myosin n nano NHS normal horse serum nt nucleotide N-terminus aminoterminus Nrx Neurexin OD optical density o/n overnight ORF open reading frame Patj Pals1-associated Tight Junction Protein Par partition defective PBS phosphate-buffered saline PCR polymerase chain reaction ph pharynx Ps Pasilla ps posterior spiracles Put Punt rpm revolutions per minute Rpr Reaper r/t room temperature SAR subapical region Sas Stranded at second Scyl Scylla Scrib Sribble SDS-PAGE sodium dodecyl sulfate-polyacrylamide gel electrophoresis XIII

Sdt Stardust Sema Semaphorin SG salivary gland Shg Shotgun (Drosophila E-cadherin) SJ septate junction SOC super optimal catabolite-repression ss single-stranded st stomodeum SynCAM synaptogenic adhesion molecule TAE Tris-acetate-EDTA TGF Transforming Growth Factor Tkv Thickveins Tr tracheal metamere Tris 2-Amino-2-hydroxymethyl-propane-1,3-diol UV ultraviolet light V volt Wg Wingless WT wild type Yrt Yurt ZA zonula adherens Zip Zipper (Drosophila non-muscle myosin heavy chain)

1 1 Introduction

In order to study morphogenetic processes in animal development, Drosophila melanogaster has been chosen as a model system for this thesis. D.melanogaster is well-established in the lab and its genetics are relatively simple. Many genetic tools are available, it is easy to keep and produces a lot of oﬀspring. Its embryogenesis is accomplished within 24 hours, allowing for the collection of results in a short amount of time.

Epithelia constitute a great model system to study the regulation of morphogenetic processes during embryogenesis, since they can be easily observed and develop in a highly stereotypical fashion. Furthermore, aberrations in embryonic morphogenesis are often clearly reﬂected in phenotypes aﬀecting epithelia and the organs deriving from them.

1.1 Epithelial cell polarity

Like in all metazoans that produce oﬀspring by sexual reproduction, the development of

Drosophila melanogaster starts with a single cell, the zygote. This cell must undergo a large number of complex changes to bring forth a fully-developed embryo: Cells must divide and organize themselves into tissues, which have to move in a coordinated fashion to form organs. This process requires the cells to keep in close contact with each other while remaining able to rearrange. These requirements are met in epithelia, whose foremost characteristic is their apicobasal polarity. (Gumbiner, 1992; Knust, 1996; Knust and Leptin,

1996; Schöck and Perrimon, 2002). Correct polarity is a prerequisite for proper epithelial cell adhesion, tissue function and organogenesis (Knust, 1994; Müller and Bossinger, 2003;

Pocha and Knust, 2013; Roignot et al., 2013).

1.1.1 Cellularization and formation of the primary epithelium

The ﬁrst step during Drosophila embryonic morphogenesis is to establish the blastoderm epithelium, the ﬁrst epithelium from which all other organs and tissues will derive.

The development of Drosophila melanogaster begins with a number of divisions of the zygotic nucleus, resulting in a syncytium, a single cell containing multiple nuclei. About

5,000 of those nuclei move to the cell periphery, where the blastoderm epithelium is formed, a process called cellularization (Figure 1.1 a, c-e) (Rabinowitz, 1941; Turner and Mahowald,

1976). During cellularization, membrane from the embryo surface invaginates, forming 2 Introduction furrow canals (FC) around all nuclei in the periphery. The FCs advance until they have passed the nuclei, whereupon they expand to fully envelop the nuclei. Hence, a single layer of cells surrounding the embryo, the blastoderm epithelium, is formed (Royou et al., 2004).

During cellularization, key molecules in stereotypical positions of the syncytium membrane act as cues for the establishment of cell polarity and cell-cell junctions (Figure 1.1 b and

Section 1.1.1.1) (Mavrakis et al., 2009). In the mature epithelium, adherens junctions (AJs) form a barrier between apical and basolateral polarity complexes. Furthermore, AJs link the cytoskeleton of neighboring cells, allowing the epithelium to act as a physical entity.

This property is crucial for coordinated epithelial movement.

Figure 1.1: Schematic representation of Drosophila cellularization, adapted from Lecuit (2004). a: Preblastoderm Drosophila embryo. Only a few nuclei (orange) are represented at the embryo periphery. The nuclei in the embryo center are yolk nuclei. b: Dorsal view of syncytial membrane with future apical (magenta) and basolateral (grey) membrane regions. c: Detail of the beginning invagination of surface membrane between two nuclei (grey triangle). d: Establishment of the furrow canal (FC); the invaginating membrane takes the basolateral membrane markers along. e: After the FC has passed the nucleus, it widens to meet FCs from neighboring cells and complete compartmentalization of the primary epithelium. Adherens junctions (AJ), a prerequisite for apicobasal polarity, are being established.

1.1.1.1 Establishment of epithelial polarity and adhesion

For an epithelium to carry out its barrier function and to act as the basic unit for morphogenetic processes, it has to be properly polarized. Therefore, the network of cell polarity determinants must be created during cellularization and maintained throughout the life of the animal. Molecules involved in apicobasal cell polarity and cell-cell adhesion inﬂuence and regulate each other, and some are crucial for the accomplishment of morphogenetic processes (see below). The cues for apicobasal polarity of the primary epithelium are already present in the plasma membrane of the syncytium (Figure 1.1 b).

The membrane above each nucleus possesses apical characteristics whereas in between nuclei, the future basolateral membrane contains basolateral markers and the cell adhesion Epithelial cell polarity 3

protein DE-cadherin (DE-cad, encoded by shotgun, shg) (Mavrakis et al., 2009; Bhat et al., 1999). During invagination of the FC, these markers act as cues to build up the

full apicobasal polarity network (Mavrakis et al., 2009). Bazooka (Baz, Drosophila Par3)

accumulates at the apical cell pole and recruits DE-cad into preliminary sport AJs (Knust,

2002; Harris and Peifer, 2004). Baz also enables the apical localization of Par6 and atypical

Protein Kinase C (aPKC) (Cheeks et al., 2004; Harris and Peifer, 2005). Together, Baz,

aPKC and Par6 form the apical Par complex (Knust, 2002; Tepass, 2012; Chen and Zhang,

2013). Baz recruits Stardust (Sdt, Drosophila Pals1) to the apical domain (Krahn et al.,

2010). Baz is then phosphorylated by aPKC, allowing Sdt to dissociate from the complex

and bind the apical determinat Crumbs (Crb) instead (Krahn et al., 2010). Par6 and

aPKC can also interact with and bind to the Crb complex, which results in Baz being

located slightly more basally at the AJs (Tepass and Knust, 1993; Bachmann et al., 2001;

Roh et al., 2002; Wang et al., 2004; Kempkens et al., 2006). Eventually, the Crb complex

contains Crb, Sdt and the Sdt interactor Pals1-associated tight junction protein (Patj)

(Figure 1.2) (Tepass and Knust, 1993). This complex is localized in the subapical region

(SAR) directly underneath the apical cell membrane. The Par complex consisting of Baz,

Par6, aPKC and the GTPase Cdc42 can also be found in the in the SAR. Immediately

basally to the SAR, the AJs are localized (Figure 1.2) (Knust, 2002).

Crb enables the assembly of belt-like AJs during Drosophila gastrulation (Tepass and

Hartenstein, 1994; Tepass, 1996; Grawe et al., 1996; Bilder et al., 2003). All AJs of one

cell form the zonula adherens (ZA), a belt surrounding the cell basally to the SAR. Crb

regulates ZA assembly by interaction with the spectrin cytoskeleton via direct binding to the

FERM-domain protein Moesin (Moe) (Médina et al., 2002; Wei et al., 2015). Furthermore,

the Membrane-associated guanylate kinase (MAGUK) Lethal giant larvae (Lgl) at the

basolateral membrane domain functionally interacts with Crb, suppressing its activity

basolaterally to ensure proper AJ localization (Tanentzapf and Tepass, 2003).

1.1.2 The epithelial polarity network

Several protein complexes ensure the maintenance of apicobasal epithelial cell polarity

(Figure 1.2). Mutual antagonism between apical and basolateral protein networks ensures

that apical determinants do not overﬂow into basal regions and vice versa (Bilder et al.,

2003). The foremost apical polarity complexes are: 1) the Crb complex containing the 4 Introduction transmembrane protein Crb, its direct interactor Sdt and Patj, which binds to Sdt (Tepass and Knust, 1993; Bachmann et al., 2001; Roh et al., 2002; Bulgakova et al., 2008), and

2) the Par complex with Baz, Par6 and aPKC (Knust, 2002; Wodarz et al., 2000; Huynh et al., 2001; Petronczki and Knoblich, 2001; Benton and St Johnston, 2003a; Tepass, 2012;

Chen and Zhang, 2013). At the basolateral membrane, the Scribble (Scrib) complex is found, containing the PDZ (PSD-95, Discs large and ZO-1) proteins Scrib and Discs large

(Dlg), as well as Lgl (Bilder et al., 2000; Bilder and Perrimon, 2000). A protein complex associated with the basolateral septate junctions (SJs) contains two FERM proteins, Yurt

(Yrt) and Coracle (Cora) and the transmembrane protein Neurexin-IV (NrxIV) (Laprise et al., 2009).

Figure 1.2: Schematic representation of cell polarity markers in Drosophila, adapted from Knust (2002). In the subapical region (SAR), just below the apical cell pole, the Cumbs complex consisting of the transmembrane protein Crumbs (Crb), its binding partner Stardust (Sdt) and the Sdt interactor Patj can be found. This complex deﬁnes the apical membrane region. Another apical protein complex is the Par-complex, with the Drosophila Par3 homologue Bazooka (Baz), atypical Protein Kinase C (aPKC) and its regulatory subunit Par6 as well as the GTPase Cdc42. The adherens junctions (AJ) mark the boundary between apical and basolateral regions in the cell. At the septate junctions (SJ) the basolateral Lgl-complex is locaized. It hosts the proteins Lethal giant larvae (Lgl), Discs Large (Dlg) and Scribble (Scrib). The transmembrane protein Neurexin IV (Nrx-IV) is also localized at the SJ. Its cytoplasmic domain is bound by the FERM domain protein Coracle (Cora) which in turn interacts with another FERM domain protein, Yurt (Yrt).

At the SAR, the serin-threonine kinase aPKC phosphorylates Lgl to release its membrane association, thereby excluding it from the apical cell pole (Wodarz et al., 2000; Suzuki et al., 2001; Betschinger et al., 2003; Hutterer et al., 2004). aPKC is also necessary for

AJ maintenance. It regulates interactions between microtubules and AJs, as well as endocytotic DE-cad internalization; both processes are crucial for AJ stability (Harris and Tepass, 2008; Harris and Peifer, 2007; Georgiou et al., 2008). Par1, a basolateral kinase, phosphorylates Baz to exclude it from the basolateral domain and promote its aﬃliation with the apical Crb complex (Shulman et al., 2000; Benton and St Johnston,

2003b). Crb represses Scrib activity in the apical region of the cell whereas Scrib inﬂuences

Crb endocytosis, a process required for maintenance of Crb levels at the apical membrane Epithelial cell polarity 5

(Bilder et al., 2003; de Vreede et al., 2014; Lu and Bilder, 2005). Some members of the

FERM protein superfamily have emerged to be involved in cell polarity. The SJ components

Yrt and Cora negatively regulate Crb during organ formation in the Drosophila embryo,

maintaining correct apicobasal polarity (Laprise et al., 2006, 2009). In late-stage embryos,

Yrt is recruited to the apical cell pole where it directly interacts with Crb via the FERM

domain binding motif (FDBM) in the intracellular domain of Crb. Yrt limits Crb activity,

thereby restricting apical membrane growth in diﬀerent epithelia-derived tissues such as

the adult photoreceptor cells and the dorsal trunk of the larval trachea (Laprise et al., 2006,

2009, 2010; Gamblin et al., 2014). Furthermore, Yrt and aPKC reciprocally antagonize

each other (Gamblin et al., 2014). Yrt is phosphorylated by aPKC, neutralizing Yrt activity

at the apical membrane, while Yrt inhibits aPKC-mediated signalling to reduce apical

membrane growth (Gamblin et al., 2014). Yrt and Cora are also involved in morphogenetic

processes during Drosophila embryogenesis, albeit it has not been established yet whether

this function is dependent on their role in apicobasal cell polarity (Hoover and Bryant,

2002; Laprise et al., 2006, 2009; Gamblin et al., 2014).

1.1.3 Cell-cell adhesion

Cell-cell adhesion within epithelia is a prerequisite for proper morphogenesis, as they allow

the epithelial cells to move in a coordinated fashion and to communicate with each other.

Insect epithelia have two kinds of intercellular junctions: The AJs, which form the ZA

around the apical portion of the cell, and the SJs, which lie more basally (Figure 1.3).

While the AJs have more mechanical functions, the SJs mainly act as a molecular barrier.

1.1.3.1 Adherens junctions

The core molecules of AJs are DE-cad and α-catenin (α-cat) as well as β-cat/ Armadillo

(Arm) (Takeichi, 1988; Oda et al., 1994, 1993; Peifer, 1993, 1995). DE-cad is a homophilic

transmembrane adhesion protein whose extracellular domain links neighboring cells (Kemler,

1993; Tepass et al., 1996). The cytoplasmic domain of DE-cad binds to β-cat which in

turn binds α-cat (Kemler, 1993; Nagafuchi et al., 1994). α-cat links the AJ to the actin

cytoskeleton (Yamada et al., 2005; Desai et al., 2013; Chen et al., 2015).

AJs link the plasma membrane with the cytoskeleton, allowing cytoskeletal movements

to result in cell shape changes (Munjal and Lecuit, 2014). Furthermore, AJs distribute 6 Introduction physical forces among the entire tissue by mechanically linking the cytoskeleton of all cells.

Consequentially, AJs are instrumental in tissue movement and tissue remodeling processes

(Borghi et al., 2012; Munjal and Lecuit, 2014).

1.1.3.2 Septate junctions

SJs are situated at the basolateral membrane in insect epithelia (Lane and Skaer, 1980).

Their core components are claudins, transmembrane proteins spanning the membrane four times and engage in homo-and heterophilic interactions both within the same cell and with claudins from neighboring cells (Furuse et al., 1999; Behr et al., 2003; Wu et al.,

2004; Nelson et al., 2010). The SJs mainly serve as a barrier for paracellular transport of molecules (Baumgartner et al., 1996).

Figure 1.3: Schematic representation of cell adhesions in Drosophila, adapted from Knust (2002). Below the subapical region (SAR) the adherens junctions (AJ) mark the boundary between apical and basolateral regions in the cell. Cell-cell adhesion in AJs is achieved by the transmembrane protein DE-cadherin. DE-cadherins from neighboring cells interact with each other. The AJs are linked to the cytoskeleton via β-cat/ Arm and α-cat. In this way, AJs link the cytoskeleton beyond cell borders, enabling a tissue to act as a physical entity. At the septate junctions (SJ), cell-cell adhesion is carried out via claudins. They interact with each other both within the same cell and between neighboring cells. The SJs mainly act as chemical borders to regulate paracellular molecular transport.

1.2 Epithelial movements in Drosophila embryonic

morphogenesis

In the previous sections it has been established how epithelial polarity is set up and maintained, and how epithelial cells are enabled to move in a coordinated fashion. The epithelial movements in the Drosophila embryo studied in this thesis will be introduced here. One example is the formation of epithelial tubes, i.e. during the development of the larval salivary glands (SGs), tracheae and malpighian tubules. Coordinated migration of epithelial sheets on the other hand an be observed during dorsal closure (DC) and head involution (HI). Epithelial movements in Drosophila embryonic morphogenesis 7

1.2.1 Epithelial tube formation during Drosophila embryogenesis

Three examples for epthelial tubes - the SGs, tracheae and Malpighian tubules - are of importance in this thesis. An overview over their respective morphogeneses is given in

Figure 1.4 and the morphogenesis of the SGs will be described as an example. While the

SGs and Malpighian tubules are simple, branchless tubes, the tracheae form a complex tubular network. However, the morphogenesis of all three organs starts with placodes, a group of specialized cells. While the tracheal and SG placodes derive from the embryonic epidermis, the Malpighian tubules are derivatives of the hindgut. In all cases, tubulogenesis happens in three steps: 1) Invagination of the placode cells, 2) elongation and migration of the tube, and 3) establishment of a stable lumen. In the tracheae, tubes from neighboring segments have to fuse before the lumen is stabilized.

The larval SGs of Drosophila melanogaster develop via invagination of two epithelial placodes. Their cells will form the secretory tubes of the SGs. After invagination is

ﬁnished, the tubes migrate to their ﬁnal position in the body cavity. The last step in the morphogenesis of the larval SG in Drosophila melanogaster is the expansion of the lumen.

The expansion and maintenance of the lumen diameter likely requires the deposition of an aECM around the luminal circumference (Abrams, 2006). For this, apical protein secretion is necessary. However, hardly anything is known about the establishment of the aECM in the Drosophila SG.

1.2.2 Coordinated migration of epithelial sheets during Drosophila embryogenesis

The coordinated movement of epithelial sheets in the Drosophila embryo can be observed during germ band retraction (GBR), dorsal closure (DC) and head involution (HI; Figure

1.5). All three require tight intercellular contacts mediated by AJs, intact cell polarity and remodeling of cytoskeletal components (Gorﬁnkiel and Arias, 2007; Houssin et al., 2015;

Bertet et al., 2004; Campbell et al., 2009; Simone and DiNardo, 2010). The processes during DC will be described as an example.

After gastrulation and organogenesis are ﬁnished, the embryo has a dorsal opening covered by the amnioserosa (AS), a transient tissue made of ﬂat cells. The dorsal opening is subsequently closed via the dorsalwards movement of the lateral epidermis during DC. 8 Introduction

a: Schematic overview over b: Schematic overview over c: Schematic overview over tracheal development in the salivary gland development in malpighian tubule development in Drosophila embryo, modified the Drosophila embryo, modified the Drosophila embryo, modified from http://www.sdbonline.org/ from http://www.sdbonline. from http://www.sdbonline. sites/fly/atlas/1819.htm. T1 org/sites/fly/atlas/3031.htm org/sites/fly/atlas/3031.htm. marks the first thoracic semgment, During stage 11, the SG placodes During stage 11, the malpighian A8 marks the last abdominal invaginate. In stage 12, the tube tubules (mt, yellow) bud from segment. During stage 11, the elongates and migrates. At stage the hindgut primordium (hg, tracheal placodes invaginate. In 15, the SG secretory tubes have blue). In consecutive stages the stage 12, each unit, called tracheal reached their final place in the four tubes elongate and migrate metamere, starts branching. embryo. to their stereotypical position. During consectutive stages, the Abbreviations: (mg) midgut, (fg) branches elongate and migrate foregut. and branches from neighboring metameres fuse.

Figure 1.4: Schematic overview over morphogenesis of tubular organs in the Drosophila embryo. All embryos are anterior left, dorsal up. The developmental stage is given in the top left corner.

It starts with the assembly of a supracellular actomyosin cable in the dorsal-most row of

cells, the leading edge (LE) cells. The cable runs around the immediate border of the

dorsal hole and is thought to aid dorsal closure by a ratchet-like contraction mechanism

carried out by actin and myosin (Kiehart et al., 2000; Solon et al., 2009). The LE cells also

extend actin-based ﬁlopodia contacting the AS (Jacinto et al., 2000). These ﬁlopodia aid

in the ﬁrst contact made by LE cells from opposite sides of the dorsal hole. The AS cells

exhibit pulsed actomyosin contractions which are instrumental for DC (Blanchard et al.,

2010). Final closure is accomplished by pulling the cell sheets together via the ﬁlopodia

and establishing new AJs at the novel contact points. (Jacinto et al., 2000; Millard and

Martin, 2008; Choi et al., 2011). In embryos mutant for the FERM domain proteins Yrt

and Cora, a dorsal hole remains, although it is not clear which molecular mechanisms cause

these defects. It is also unknown whether the role of Yrt and Cora in cell polarity plays a Epithelial movements in Drosophila embryonic morphogenesis 9 role in these defects (Hoover and Bryant, 2002; Laprise et al., 2006, 2009; Gamblin et al.,

2014).

Figure 1.5: Schematic overview over germ band retraction, dorsal closure and head involution in chronological order from top to bottom. The embryonic stages are marked in the upper left corner of every embryo. Anterior to the left; stage 13 and 15 are in dorsal view, all others in lateral view. Blue arrows mark direction of tissue movement. The germ band (green) is fully extended in stage 11 and retracts during stage 12. Retraction is ﬁnished in stage 13 and leaves a dorsal hole covered by the amnioserosa (grey). The hole is closed in stages 13 through 16 by dorsal migration of the lateral epidermis. Simultaneously, the dorsal ridge (magenta) moves over the head segments (yellow) which are becoming internalized via head involution. After all morphogenetic tissue movements are ﬁnished, the internalized head segments secrete the head skeleton (dashed line, yellow).

1.2.2.1 FERM domain proteins in epithelial migration

Throughout this introduction, FERM domain proteins have been mentioned as key players in epithelial cell polarity and epithelial migration. In the forward genetic screen carried out for this thesis, another member of the FERM group of proteins, Cdep (Chondrocyte-derived ezrin-like protein), emerged as a regulator of morphogenetic processes in the Drosophila embryo. It will be introduced in the following Section along with the aformentioned Yrt and Cora.

The FERM (Band 4.1, Ezrin, Radixin, Moesin) superfamily of proteins mainly contains scaﬀolding proteins that can interact with membrane-associated proteins, membrane lipids and cytoskeleton components; they therefore aid in stabilizing the cell cortex. The FERM domain can be bound by protein interactors (Tyler et al., 1979; Leto and Marchesi, 1984;

Chishti et al., 1998; Fiévet et al., 2007). The FERM proteins Yrt and Cora are components 10 Introduction of the SJ and regulate apicobasal epithelial polarity (Ward et al., 1998; Lamb et al.,

1998; Hoover and Bryant, 2002; Laprise et al., 2006, 2009). They also play a role in DC and HI (Fehon et al., 1994; Hoover and Bryant, 2002). Both belong to the FERM-FA subclass. Proteins from this subclass have an additional domain immediately C-terminal to their FERM domain, termed FERM-adjacent (FA). The FA domain seems to be a phosphorylation target for negative regulation of protein activity (Manno et al., 2005;

Baines, 2006).

Embryos homozygous for a yrt loss-of-function mutant fail to accomplish GBR, DC and

HI (Hoover and Bryant, 2002). In those embryos, the AS degenerates, which could be one of the reasons for the observed defects, since the AS plays a mechanical role in GBR and

DC (Hoover and Bryant, 2002; Lynch et al., 2013, 2014; Blanchard et al., 2010).

Cora function is essential for the formation and integrity of SJs and the completion of

DC and HI (Ward et al., 1998; Lamb et al., 1998; Genova and Fehon, 2003; Paul et al.,

2003; Laprise et al., 2010). In cuticle preparations from embryos mutant for either yrt or cora, the head skeleton is often misformed or missing and anterior holes can be observed

(Hoover and Bryant, 2002; Fehon et al., 1994; Ward et al., 1998). Embryos mutant for cora also show defects in the salivary gland, reﬂected by necrotic remnants of this organ in cuticle preparations (Lamb et al., 1998). Morphological defects in embryos caused by loss of cora function can be rescued by expression of only its FERM domain (Ward et al., 2001).

However, while the mechanism of Yrt and Cora function in cell polarity has been relatively well-studied (Section 1.1.2), not much is known about how they exert their function in tissue morphogenesis.

1.2.2.2 Cdep

Drosophila Cdep, a FERM-FA protein, is the homologue of human CDEP and mouse

Farp1. CDEP has another homologue in humans, called FARP2 (also known as FRG, FIR and PLEKHC3), which has a homologue in mouse also called Farp2. In addition to the

FERM and FA domains, Cdep and its homologues contain a Rho-GEF (Guanine-nucleotide

Exchange Factor) domain and two pleckstrin-homology (PH) domains. However, studies on human CDEP and FARP2 and mouse Farp1 and Farp2 report contradictory results as towards which small GTPase CDEP or Farp1/2 has GEF activity (Koyano et al., 2001;

Cheadle and Biederer, 2012; Zhuang et al., 2009; Miyamoto et al., 2003; Fukuhara et al., Epithelial movements in Drosophila embryonic morphogenesis 11

2004; Toyofuku et al., 2005; Kubo et al., 2002). Furthermore, He et al. (2013) have found

that some highly conserved amino acid residues in other RhoGEFs are mutated in Farp1

and 2. Farp2 also assumes a tertiary structure that buries the RhoGEF domain deep

inside the protein and autoinhibits it (He et al., 2013). This would necessitate substantial

conformational changes in the protein to activate its GEF activity. Taken together, these

results suggest that Farp2 does not have direct GEF activity (He et al., 2013).

Human CDEP is highly expressed in hypertrophic chondrocytes during vertebrate bone

formation (Koyano et al., 1997, 2001). Chondrocytes produce and maintain cartilage, and

hypertrophic chondrocytes are the most mature stage of cartilage-forming cells. CDEP

has also been found to be upregulated in the human fetal brain and spleen (Koyano et al.,

1997).

Both mouse Farp homologues play a role in axon guidance in mouse embryonic development.

Axons ﬁnd their way with the help of chemical cues from surrounding tissues which can

act either as attractant or repellent (for review see Tessier-Lavigne and Goodman, 1996).

Farp1 functions as an eﬀector of Semaphorin6A (Sema6A), a repulsive axon guidance cue,

and the Sema6A receptor PlexinA4 (reviewed in Haklai-Topper et al., 2010; Zhuang et al.,

2009). Later in development, Farp1 is localized at postsynaptic membranes in mature

neurons (Cheadle and Biederer, 2012). There, it interacts with SynCAM1 (synaptogenic

adhesion molecule 1), promoting synapse formation and F-actin assembly in dendritic

spines (small, actin-rich membraneous outgrowths of neuronal dendrites) (Cheadle and

Biederer, 2012). Farp2 on the other hand binds to the Sema3A receptor PlexinA1 and

therefore acts as a Sema3A eﬀector, resulting in the repulsion of axons from Sema3A cues

(Toyofuku et al., 2005).

Axon guidance in Drosophila is also mediated by Semaphorin cues and Plexin receptors.

Drosophila Sema-1a and Sema-1b are structurally similar to mouse Sema6A, whereas

Drosophila Sema-2a and Sema-2b are more similar to mouse Sema3A (Yazdani and Terman,

2006). Sema3A guidance cues have been implicated in the migration and diﬀerentiation of

epithelia in the mouse and human lung and in the mouse kidney, as well as in epidermal

migration in the nematode Caenorhabditis elegans (Ito et al., 2000; Becker et al., 2011;

Tufro et al., 2008; Roy et al., 2000). It is therefore possible that Drosophila Cdep mediates

epithelial migration by relaying external guidance cues. 12 Introduction

1.3 Mutagenesis with the CRISPR/Cas9 system

One of the objectives of this thesis was to establish loss-of-function mutants of Drosophila

Cdep. To this end, the CRISPR/Cas9-system was utilized for mutagenesis. Since it is a relatively novel method for genetic engineering, its properties and advantages will be introduced here.

The CRISPR/Cas9 system (clustered regularly interspaced short palindromic repeats/

CRISPR-associated) is a bacterial self-defense mechanism against viruses (Barrangou et al.,

2007). It is based on three components: the CRISPR RNA (crRNA), the trans-activting

CRISPR RNA (tracrRNA) and the endonuclease Cas9 (Figure 1.6). Part of the crRNA anneals to the tracrRNA. The tracrRNA forms a speciﬁc structure that can be bound by

Cas9. Together, this complex mediates the silencing of invading nucleic acids. However, potential CRISPR target sites need to contain a 5’-NGG-3’ nucleotide sequence, called a protospacer-adjacent motif (PAM). Cas9 induces a double-strand break (DSB) three bp upstream of the PAM, destroying the invading DNA (reviewed in Hwang et al., 2013;

Bassett et al., 2013; Jinek et al., 2012).

CRISPR/Cas9 has been simpliﬁed to be a versatile tool in genome engineering by fusing the crRNA and tracrRNA into the short guide RNA (sgRNA, also sometimes called gRNA or chimeric RNA, chiRNA; Figure 1.7) (Jinek et al., 2012). Furthermore, the gene encoding

Cas9 has been codon-optimized for use in eukaryotes and a nuclear localization signal has been attached (Cong et al., 2013).

Figure 1.6: Schematic overview over the type II CRISPR/Cas9 system from Streptococcus pyogenes, modiﬁed from Hwang et al. (2013). The target site (green) must be followed by a protospacer adjacent motif (PAM, red). The CRISPR RNA (crRNA) binds to the target site by homologous base pairing. The trans-activting CRISPR RNA (tracrRNA) binds to the crRNA and forms a stereotypical secondary structure that is recognized by the endonuclease Cas9 (light blue). Cas9 induces a double strand break in the DNA target site three bp upstream of the PAM (black triangles). Mutagenesis with the CRISPR/Cas9 system 13

Figure 1.7: Schematic overview over the modiﬁed CRISPR/Cas9 system for use in eukaryotes, modiﬁed from Hwang et al. (2013). Represenatation of the CRISPR/Cas9 components as in Figure 1.6. The crRNA and tracrRNA have been fused into one small guide RNA (sgRNA) still folding into a secondary structure that is recognized by Cas9.

The high potential of this system to serve as a tool for site-speciﬁc mutagenesis is based on

two diﬀerent mechanisms that eukaryotic cells use to repair DNA DSBs: Non-homologous

end-joining (NHEJ) and homologous recombination (HR). In NHEJ, loose DNA ends are

processed before ligation, leading to the deletion or insertion (indels) of a few nucleotides

(Figure 1.8) (for review see van Gent et al., 2001). This mechanism is already a potent

tool for mutagenesis, since indels often induce frameshifts and therefore render the aﬀected

gene non-functional (Bassett et al., 2013). During HR, the homologous strand from the

sister chromatid invades the damaged strand and serves as a template for synthesis of

the missing sequence. The experimenter can make use of this mechanism by introducing

a DNA donor of his choosing that will act as a template for homology repair. In this

way, mutagenesis can be tightly controlled and a wide range of DNA sequences can be

introduced into the genome. Furthermore, with the use of two sgRNAs simultaneously,

longer stretches of DNA can be deleted, oﬀering a simple way to accurately delete whole

genes. Using a donor plasmid carrying a marker gene for homologous recombination in this

context allows for quick and convenient mutant screening (Figure 1.9) (Gratz et al., 2014).

Figure 1.8: Example for indels in Drosophila white gene caused by CRISPR/Cas9-induced mutagenesis, modified from Bassett et al. (2013). Orange: sgRNA target site, pink: PAM. The first sequence is WT. Lowercase letters: inserted nucleotides, dashes: deleted nucleotides. Sequences are from different flies mutagenized with the same sgRNA. 14 Introduction

Figure 1.9: Schematic overview over strategy to replace an endogenous gene with a marker gene via CRISPR/Cas9, modiﬁed from Gratz et al. (2014). The gene that is to be deleted is cut out of the DNA by directing Cas9 to both ends of the coding region with two sgRNAs (black triangles). A donor plasmid for homologous repair is provided. It carries homology arms ﬂanking the deleted region and a marker gene which can be used for convenient mutant screening. 15 2 Aim of My PhD Thesis Work

The aim of this thesis was to elucidate how coordinated morphogenetic processes in animals

are regulated. The Drosophila embryonic salivary gland and the process of embryonic dorsal

closure were used as model systems. In both systems, novel regulators of the embryonic

morphogenesis of Drosophila melanogaster were identiﬁed and characterized.

The gene coding for the Drosophila homologue of Chondrocyte-derived ezrin-like protein

(Cdep) was studied in order to shed light on the molecular functions of this FERM domain

protein in embryonic development. Diﬀerent genetic techniques were used to characterize

the phenotypes caused by loss of Cdep function and to elucidate its role in the genetic

network regulating Drosophila melanogaster embryonic development.

The main questions addressed in this project were:

1) Which genes are involved in regulating the morphogenesis of the

Drosophila salivary glands and the process of dorsal closure?

In order to answer this question, candidate genes were mapped using Drosophila

embryos transheterozygous for partly overlapping deﬁciencies. The embryos were

screened for developmental defects in the salivary glands and during dorsal closure using

immunoﬂuorescence.

2) Which defects does loss of Cdep gene function cause during Drosophila melanogaster embryonic morphogenesis?

To address this question, loss-of-function mutants of Cdep were established

using the CRISPR/Cas9 system. The phenotypes caused by these mutations in

Drosophila melanogaster embryos and ﬁrst-instar larvae were assessed using embryo

immunoﬂuorescence, survival assays and cuticle preparations.

3) Is Cdep part of the genetic network of FERM domain scaﬀolding proteins?

For this, Cdep mutants were combined with mutations in two other FERM domain

proteins, yurt and coracle. The phenotypes in double-mutant embryos were studied to

draw conclusions about genetic interactions between Cdep and yurt or coracle.

17 3 Preliminary Work

David Flores-Benitez, a postdoc from the Knust lab, carried out a genomic screen in

Drosophila melanogaster to ﬁnd novel regulators of embryonic morphogenesis. He chose his candidates based on a possible interaction with the cell polarity determinant Crb. The putative Crb interactors were found in database searches using online resources.

The screen was done using Drosophila melanogaster embryos homozygous for deﬁciencies, large chromosomal deletions uncovering several dozens of genes. The embryos were screened via immunoﬂourescence, looking for defects in the embryonic morphogenesis of either the salivary glands or in the process of dorsal closure. These two morphogenetic processes were chosen as model systems for several reasons: 1) The salivary glands develop from polarized epithelial cells that maintain their polarity through the whole developmental process. Correct cell polarity is also required for proper secretion. 2) During salivary gland morphogenesis, no mitosis or apoptosis takes place, which makes it a relatively simple system to study. 3) In dorsal closure, correct epithelial morphology depending on cell polarity and cell adhesion is crucial. 4) dorsal closure is an ideal system to study coordinated morphogenetic movements involving a reletively large tissue, since most parts of the lateral epidermis of the Drosophila embryo have to move in a tightly regulated, coordinated manner.

Embryos were immunostained for Crb, Stranded-at-second (Sas) and CrebA. Sas, like

Crb, is an apical marker, whereas CrebA is speciﬁcally expressed in nuclei of secretory salivary gland cells. Defects in the morphology of the salivary glands between stages 11 and 16 were recorded, as well as defects in the overall morphology, speciﬁcally in the lateral epidermis and the amnioserosa during stages 13 through 16.

In agreement with D. Flores-Benitez I chose a number of those deﬁciency lines showing interesting phenotypes and continued the project by mapping possible candidate genes responsible for the observed defects.

19 4 Results

4.1 A screen for novel regulators in Drosophila embryonic

morphogenesis

In animal embryonic morphogenesis, many basic processes are shared across phyla. In this thesis, morphogenetic processes involving epithelia-derived organs and tissues are being studied since their regulation is still not fully understood.

In order to ﬁnd novel candidate genes involved in the formation of epithelia-derived organs and epithelial migration, the embryo of Drosophila melanogaster was employed as a model system. Chromosomal deﬁciencies were used to closely map genomic regions causing phenotypes in the embryonic salivary glands (SG), an epithelial tubular organ, or during the process of dorsal closure (DC), an example for coordinated epithelial migration.

4.1.1 Deﬁciencies on the left arm of chromosome 3 cause defects in SG lumen morphology

A number of deficiencies that showed different defects in the SG were taken over from a screen carried out by David Flores-Benitez (Figures 4.1 and 4.2). Embryos homozygous for the deficiency of interest were stained with α-Crb, α-Sas and α-CrebA antibodies to visualize the apical cell poles and the nuclei of the SG secretory cells, respectively. In stage

15 WT embryos, the proximal part of the SG rises up dorsally, then bends posteriorly at approximately half its length (Figure 4.2 top). The lumen is of even width and has a relatively small diameter. In stage 17 WT embryos, the lumen is expanded when compared to earlier stages (Figure 4.2 top right). Some of the deﬁciencies studied caused the distal tip of the SG to point ventrally instead of posterior when homozygous (Figure 4.2 middle and bottom left). In embryos homozygous for other deﬁciencies, the overall shape of the glands was irregular (Figure 4.2 middle right) or the lumen was not evenly expanded and showed bloated portions that alternated with apparently closed parts of the lumen (Figure

4.2 bottom right). Since lumen stabilty usually requires the secretion of an aECM in tubular organs, and apical secretion requires correct apicobasal cell polarity, this phenotype was chosen to study epithelial integrity during Drosophila embryogenesis. Two deficiency lines, Df(3L)ED4287 and Df(3L)ED4502, showed these luminal defects characterized by 20 Results alternating seemingly closed and bloated parts in the SG when homozygous in Drosophila embryos (Figure 4.3). These deficiencies lie in close proximity to each other on the left arm of chromosome 3, but do not overlap (Figure 4.1, green bars). One or several genes deleted by each of these deficiencies is likely responsible for the described phenotype, which will be called "intermittent tube closure" from now on.

Figure 4.1: Overview over deficiencies used to find phenotypes that were judged to be suitable to study Drosophila embryonic morphogenesis in the salivary glands (SGs). Coloured bars mark deleted region of chromosome in each deficiency line. Green bars mark the deficiencies that caused intermittent tube closure phenotypes in the SGs.

4.1.2 A locus in the overlap of two deﬁciencies on the right arm of chromosome 3 causes defects in Drosophila embryonic dorsal closure

In the initial screen for novel regulators of Drosophila embryonic morphogenesis carried out by David Flores-Benitez, two deﬁciencies on the right arm of chromosome 3 showed very similar defects in Drosophila embryonic DC in homozygous embryos (Figure 4.4 b and c). These deﬁciencies are called Df(3R)ED5066 and Df(3R)Exel6143 and overlap by

1.678 bp (Figure 4.5). The first question was whether the DC phenotype observed in embryos homozygous for either deficiency is due to a locus in the overlapping region. To address this, virgins carrying one of the deficiencies were crossed with males carrying the other deficiency, respectively, to yield transheterozygous embryos. Their morphology was compared to that of WT embryos as well as to embryos homozygous for Df(3R)ED5066 A screen for novel regulators in Drosophila embryonic morphogenesis 21

Figure 4.2: Examples of salivary gland (SG) phenotypes found in embryos homozygous for different deficiencies, z-stacks of fluorescent micrographs. One SG of the Drosophila embryos is shown, dorsal is up, anterior to the left. Apical cell poles are marked with Crb (green) and Sas (magenta), the nuclei of secretory SG cells are marked with CrebA (magenta). The brightness of all mutant micrographs has been enhanced to the same extent to improve visibility. Left column: all embryos are embryonic stage 15; right column: all embryos are embryonic stage 17. The SG lumen is clearly visible outlined in green. In stage 15 WT embryos, the SG rises up from the connecting duct (white triangle), then bends posteriorly. The lumen is of even width and small diameter. In stage 17 WT embryos, the lumen is expanded. In some mutants, the tip of the SG bent too far and points ventrally (middle and bottom left). In other mutants, the overall shape of the glands is irregular (middle right) or the lumen is not evenly expanded (bottom right).

Figure 4.3: Salivary gland phenotypes found in embryos homozygous for CG17364MB01315 and CG17362MB04765. Anterior part of stage 17 embryos, dorsal up, anterior to the left. Embryos were stained with α-Crb (green), α-Sas and α-CrebA (both magenta). Top left box: WT salivary gland with evenly expanded lumen. Top right: intermittent tube closure in the SG of embryos homozygous for Df(3L)ED4515. Middle and bottom left: in embryos homozygous for CG17364MB01315 or CG17362MB04765, very similar intermittent tube closure phenotypes can be observed as in embryos homozygous for Df(3L)ED4515. Middle right: Embryos transheterozygous for CG17364MB01315 and Df(3L)ED4515 also show intermittent tube closure in the SG. 22 Results and Df(3R)Exel6143, respectively. All embryos were stained with α-Crb to outline all epithelial cells. In stage 17 WT embryos, DC is finished and the epidermis completely covers the embryo (Figure 4.4 a). All head organs have been internalized during HI. In every segment, the chordotonal organs are visible as a group of five ellipsoids that are elongated along the dorsoventral axis of the embryo (Figure 4.4, white triangles). Stage 17 embryos homozygous for Df(3R)ED5066 also show the chordotonal organs, although they are misformed and not present in every segment. More notably, DC has not proceeded beyond an initial stage. HI has not been completed, either (Figure 4.4 b, arrow). The same is true for stage 17 embryos homozygous for Df(3R)Exel6143, although the overall morphology seems to be less severely affected (Figure 4.4 c). Embryos transheterozygous for Df(3R)ED5066 and Df(3R)Exel6143 showed a very similar phenotype as was observed for homozygous embryos for either deficiency (Figure 4.4 d). It did not matter which deficiency was carried by which parent. In the example shown, GBR has not finished, and

DC and HI were not accomplished, either. No amnioserosa is present, which results in parts of the hindgut emerging from the dorsal hole. Some of the segments splay out in an anterior-posterior direction at the leading edge, whereas others are bunched to an extend that excludes them from the leading edge. From this it can be concluded that one or several of the genes whose ORFs are aﬀected by both Df(3R)ED5066 and Df(3R)Exel6143 causes a GBR, DC and HI defect in Drosophila melanogaster embryos.

The SG and DC phenotypes mentioned in this section were studied further by mapping the genes causing the observed defects and attempting to characterize the molecular roles of the proteins they encode.

4.2 Two uncharacterized genes regulate SG lumen diameter in

Drosophila embryos

4.2.1 Mutations in two uncharacterized genes on chromosome 3 cause intermittent tube closure in the Drosophila SG

In order to map the genes responsible for the intermittent tube closure phenotype, smaller deﬁciencies were aquired from the Bloomington Drosophila Stock Center (BDSC).

Those smaller deﬁciencies deleted some of the genes uncovered by Df(3L)ED4287 and Two uncharacterized genes regulate SG lumen diameter in Drosophila embryos 23

Figure 4.4: Dorsal closure phenotype in embryos homozygous and transheterozygous for Df(3R)ED5066 and Df(3R)Exel6143, z-stacks of fluorescent micrographs. a: In WT stage 17 embryos, dorsal closure and head involution have finished, the epidermis covers the whole embryo. In each segment the chordotonal organs are visible (triangles), which were used as a mark for stage 17 in mutant embryos. b: In Df(3R)ED5066 embryos, dorsal closure has not finished by stage 17 and the amnioserosa is absent. Some of the segments splay out in an anterior-posterior direction at the leading edge (bar). The head organs have not been internalized (arrow). c: Df(3R)Exel6143 embryos have a more regular overall morphology, albeit dorsal closure has not finished. d: In stage 17 embryos transheterozygous for Df(3R)ED5066 and Df(3R)Exel6143, the phenotype is similar as seen in Df(3R)ED5066 or Df(3R)Exel6143 homozygous embryos, respectively. Dorsal closure has not finished and the dorsal hole is not covered by the amnioserosa, so parts of the gut are visible (asterisk). Some of the segments splay out in an anterior-posterior direction at the leading edge (bar), whereas others are bunched to an extend that excludes them from the leading edge (diamond). The head organs have not been internalized (arrow).

Figure 4.5: Localization of Df(3R)ED5066 and Df(3R)Exel6143 in the Drosophila genome. These two deﬁciencies overlap by 1,678 bp.

Df(3L)ED4502. The smaller deficiency Df(3L)ED4284 affects 35 genes also affected in

Df(3L)ED4287 (Figure 4.6 a), whereas Df(3L)ED4515 deletes nine genes uncovered by

Df(3L)ED4502 (Figure 4.6 b). An overview over genes expressed during embryogenesis and uncovered by Df(3L)ED4284 and Df(3L)ED4515 is given in Table 4.1.

For the genes listed in Table 4.1, mutant lines were aquired from BDSC, if available.

Usually, those mutant lines carried a P-element or Minos insertion within the gene’s open reading frame (ORF). For the same genes, a database search was carried out on flybase.org to ﬁnd available expression data. The expression data for CG17362 and 24 Results

Figure 4.6: Overlap of Df(3L)ED4284 with Df(3L)ED4287 (a) and Df(3L)ED4515 with Df(3L)ED4502 (b) and genes aﬀected by both deﬁciencies.

CG9040 showed that both genes are highly expressed toward the end of embryogenesis.

Expression data sorted by organ showed that both genes are also highly and exclusively expressed in the SG of third instar larvae. There was no detailed expression data available for embryonic organs and tissues. Since the time of CG17362 and CG9040 expression peaks coincides with SG lumen expansion, these two genes were chosen as candidates that could inﬂuence SG lumen morphology. However, only two mutant lines carrying Minos insertions that might aﬀect the expression of CG17362 and CG9040 were available, namely

CG17364MB01315 and CG17362MB04765 (Figure 4.7). Both CG17362 and CG9040 are located on the plus-strand of chromosome 3, within the open reading frame of CG17364 which is located on the minus-strand. Therefore, an insertion in CG17364 that lies upstream of CG17362 and CG9040 could inﬂuence their expression. When embryos homozygous for either CG17364MB01315 or CG17362MB04765 were stained for Crb, Sas and CrebA, the SGs showed a very similar intermittent tube closure phenotype that was also observed in embryos homozygous for Df(3L)ED4515. The same was true for embryos transheterozygous for

CG17364MB01315 and Df(3L)ED4515. The ﬂy line carrying CG17362MB04765 only became available late during the project and could therefore not be crossed with Df(3L)ED4515.

Taken together, these results indicate that CG17362 or CG9040 or CG17364 or possibly a combination of these genes play a role in SG lumen formation or stability.

4.2.2 CG17362/ CG9040/ CG17364 play a role in the maintenance of SG lumen width after lumen expansion

The question that arose when studying the intermittent tube closure phenotype was whether the SG lumen in CG17362MB04765 and CG17364MB01315 embryos did not expand properly Two uncharacterized genes regulate SG lumen diameter in Drosophila embryos 25

Figure 4.7: Localization of the genomic insertions CG17364MB01315 and CG17362MB04765 on the left arm of the third chromosome of Drosophila melanogaster. CG17362 and CG9040 are located on the plus-strand of chromosome 3, within the open reading frame of CG17364, which is located on the minus-strand. CG17364MB01315 could aﬀect the expression of CG17362 and/ or CG9040 since it is located upstream of their coding sequences. The Minos insertion CG17362MB04765 is also located upstream of the coding regions of CG17362 and CG9040, albeit closer to the transcription start site of CG17362 than CG17364MB01315. during stage 17 or whether it initially expanded, but subsequently collapsed in some places.

It is notable that during SG development, the apical cell poles of the secretory cells do not come in contact with each other, there is always a small lumen. In order to answer the question at hand, embryos transheterozygous for CG17364MB01315 and Df(3L)ED4515 were collected. The lumen width in all stages of SG development after invagination was assessed.

In WT stage 13 embryos, the secretory tubes of the SGs have turned and started migrating towards the posterior pole of the embryo (Figure 4.8 top row). The lumen is relatively narrow. By stage 14, the SG tubes have reached their ﬁnal position in the embryo. In stage 15, the lumen expands slightly. Full expansion of the SG lumen is reached in stage

17 WT embryos. Embryos transheterozygous for CG17364MB01315 and Df(3L)ED4515 did not show any diﬀerences to WT embryos in respect to the morphology of the SG lumen up until stage 17. Only in stage 17, the intermittent tube closure phenotype arose (Figure

4.8 bottom row). Therefore, the lumen collapses during or after its ﬁnal expansion. This hints at the possibility that CG17362/ CG9040/ CG17364 play a role either in SG lumen expansion itself or in the stabilization of the lumen diameter after expansion.

4.2.3 CG17362 is exclusively expressed in the embryonic SG

Next, it was asked whether CG17362, CG9040 and/ or CG17364 might also regulate lumen diameter in other tubular organs. To quickly assess this question, probes for mRNA in situ hybridization were made. Fixed embryos of a diﬀerent range of developmental stages after the beginning of organogenesis were stained. Even though two diﬀerent probes were made for the mRNA of each of the three genes, only the probe for CG17362 resulted in a signal. 26 Results

Table 4.1: Genes expressed during Drosophila embryogenesis and aﬀected in Df(3L)ED4284 and Df(3L)ED4515 gene name/ expression∗ gene ontology∗ mutant?/ CG no. tested? Df(3L)ED4284 Patj 0-6 hpf† , also SG; ubiquitous binds aPKC; AJ organization, cell y/y in adults polarity α-Spectrin 6-24 hpf, ubiqiutous in adults actin-binding; links y/y transmembrane proteins to actin-cytoskeleton Rap1 0-12 hpf, ubiquitous in adults GTPase; cell adhesion, cell y/y migration, neuroblast division and polarity CG12025 early pupal stages not known n/n CG13924 early larval stages not known n/n CG12024 0-12 hpf, larvae, pupae, adults Ubiquitin system component n/n CNMa 12-24 hpf, larvae, pupae, neuropeptide hormone, G-coupled n/n adults receptor signalling pathway GV1 pupal period DNA binding y/y CG42676 all times not known n/n JTBR 0-12 hpf, larvae, pupae, adults integral membrane component y/y dre4 0-12 hpf, adult ovary Component of FACT complex; y/y activates transcription elongation Df(3L)ED4515 CG32137 18-24 hpf, also embryonic SG, neuron projection morphogenesis y/y ubiquitous in adults CG17362 larval SG not known y/y CG9040 larval SG not known y/y Meics 0-6 hpf, adult female transcription factor n/n ssp2 0-12 hpf, adult female mitotic spindle elongation y/y Nxf 3 not known mRNA export from nucleus n/n CG13738 12-24 hpf not known n/n CG17364 18-24 hpf, larvae, pupae, GTP binding; microtubule-based y/y adults process CG17361 0-12 + 18-24 hpf, adult zinc and DNA or RNA binding n/n female CG17359 0-6 hpf, adult female zinc and DNA or RNA binding n/n Nprl3 0-6 hpf, adult female TORC1 signalling n/n upSET embryo, pupae, adult female zinc ion binding y/y ptip 0-12 hpf, adult female transcription factor binding; y/y histone methylation endos 0-12 hpf, pupae, adult female sulfonylurea receptor binding; n/n mitotic spindle organization CG6650 0-6 hpf, also SG; larvae, phosphotransferase; carbohydrate y/y pupae, adults metabolism ∗ gene expression and gene ontology data was obtained from flybase.org † all speciﬁcations of hpf (hours post fertilization) refer to gene expression peaks in Drosophila embryos.

CG17362 was detected strongly and exclusively in the secretory tubes of the SGs (Figure

4.9). In late stages, after the tracheae had ﬁlled with air, the signal was also detectable in Two uncharacterized genes regulate SG lumen diameter in Drosophila embryos 27

Figure 4.8: In embryos transheterozygous for CG17364MB01315 and Df(3L)ED4515, the salivary gland lumen collapses during stage 17. Anterior part of embryos, dorsal up, anterior to the left. Embryos were stained with α-Crb (green), α-Sas and α-CrebA (both magenta). In all stages after complete invagination of the salivary gland, the secretory salivary gland lumen has a normal, even width (stages 13 through 15). Only after lumen expansion in stage 17, the intermittent tube closure phenotype becomes apparent (white triangles). The inset shows a 2.5 ×blowup of the boxed area, the CrebA signal has been darkened to enhance visibility of the lumen.

Figure 4.9: In situ staining of CG17362 mRNA in WT embryos, Brightfield micrographs. dorsal up, anterior to the left. In stage 14, the SGs have reached their final position in the embryo. The pharynx (ph) and hindgut (hg) are discernible. CG17362 mRNA is detected in the cytoplasm of secretory SG cells (inset). The same is true for stage 17 embryos. (dt) dorsal trunk, (fk) filzkörper, (st) stomodeum.

the tracheal system. This was also the case for the non-binding negative control probe.

I therefore assume that the probe enters the trachea during the hybridization step, but cannot be washed out again subsequently.

I conclude that CG17362 is exclusively expressed in the secretory cells of the embryonic

SG. 28 Results

4.2.4 CG17362 and CG9040 do not contain known protein domains and are only conserved in Drosophilidae

In order to elucidate possible molecular functions of CG17362 and CG9040, a database search using the protein sequences was carried out. Protein sequences were obtained from www.uniprot.org on 24/08/2015. The UniProt identiﬁers used were Q9VUB0 for CG17362 and Q8MQZ5 for CG9040. The gene CG17362 is predicted to code for a protein of 153 aa length whereas CG9040 is predicted to code for a 175 aa long protein. The SMART tool does not predict any conserved domains for either CG17362 or CG9040. Submitting the protein sequences to HMMTOP turned out a prediction of no transmembrane domains or signal peptides. When searching for putative interactors for both proteins on STRING, some proteins expressed in the SG of late third-instar larvae come up. Most of these proteins are predicted to play a role in glue secretion to attach the pupa to the substrate (Table 4.2).

All of those putative interactions are predicted based on coexpression data. STRING does not predict any interactors for CG17364. When carrying out a protein BLAST alignment for all three predicted proteins, all hits with a sequence coverage above 80 % and an identity of 30 % or more are uncharacterized proteins from Drosophila melanogaster and other

Drosophilidae. Based on this data it was judged that elucidating the role of CG17362,

CG9040 and CG17364 would not enhance our knowledge about morphogenetic processes in animals in general. Therefore, the project was discontinued. Two uncharacterized genes regulate SG lumen diameter in Drosophila embryos 29

Table 4.2: Putative interactors of CG17362 and CG9040 proteins, predicted by STRING database gene name/ expression∗ genomic gene ontology∗ CG no. locus CG17362 new glue 1 3rd instar wandering larvae 3C10 pupal adhesion SG new glue 2 3rd instar wandering larvae 3C10 pupal adhesion SG Pre-intermoult 3rd instar wandering larvae 3C10 unknown gene 1 SG Mucin 68Ca 3rd instar wandering larvae 68C15 ECM structural SG constituent CG14265 3rd instar wandering larvae 3C10 unknown SG CG17134 3rd instar wandering larvae 68C15 proteolysis SG CG13460 3rd instar wandering larvae 71B2 unknown SG CG14852 3rd instar wandering larvae 88C6 unknown SG and imaginal discs CG8087 3rd instar wandering larvae 88C6 unknown SG and imaginal discs CG9040 new glue 3 3rd instar wandering larvae 3C10 pupal adhesion SG Mucin 68Ca 3rd instar wandering larvae 68C15 ECM structural SG constituent Ecdysone-induced 3rd instar wandering larvae 71E5 pupal adhesion, gene 71Ee SG, imaginal discs and CNS heamolymph coagulation PHD ﬁnger 3rd instar wandering larvae 19B3 - 19C1 histone binding, zinc protein 7 ortholog imaginal discs ion binding Salivary gland 3rd instar wandering larvae 3C10 pupal adhesion secretion 4 SG, digestive system, fat body, imaginal discs, CNS CG14265 3rd instar wandering larvae 3C10 unknown SG CG12310 3rd instar wandering larvae 71B2 unknown SG

∗ gene expression and gene ontology data was obtained from flybase.org † all speciﬁcations of hpf (hours post fertilization) refer to gene expression peaks in Drosophila embryos. 30 Results

4.3 Loss of Cdep causes diﬀerent defects in Drosophila

embryonic morphogenesis

4.3.1 Insertions in the ORF of Cdep cause defects in DC and HI

In order to ﬁnd out which gene located in the overlap of Df(3R)ED5066 and Df(3R)Exel6143

(Figure 4.10) causes the defects observed during DC and HI, insertion mutants were ordered for the three protein-coding genes likely aﬀected by both deletions. Those genes are called

Cdep, Ubc6 (also known as UbcD6 ) and CG14661. Two non-protein-coding genes also lie in the overlap, CR45188 and snRNA:U1:82Eb. Their role was not assessed further. At the time that these experiments were carried out, no mutant line was available for CG14661. Two insertion lines were available for Cdep, PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496

(Figure 4.11). For Ubc6, there were also two insertion lines, P{EPgy2}Ubc6EY 04634 and

Mi{MIC}Ubc6MI02360. While embryos homozygous for either insertion in Ubc6 showed no apparent defects (data not shown), embryos homozygous for either PBac{5HPw+}

CdepB122 or Mi{MIC}CdepMI00496 showed defects that were very similar to those oberved in embryos homozygous or transheterozygous for Df(3R)ED5066 and Df(3R)Exel6143. In

PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 embryos, DC seemed to be arrested, resulting in a dorsal hole in stage 17 embryos (Figure 4.12 d and e). In the same embryos, head organs were still on the outside of the embryo, leading to the conclusion that HI does not take place in those mutants (Figure 4.12 d and data not shown). Homozygous Mi{MIC}

CdepMI00496 animals did not survive until adulthood. There were some homozygous adult escapers for PBac{5HPw+}CdepB122, albeit they were too weak to keep as a homozygous stock. PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 animals were crossed with each other (Figure 4.12 f) and with animals carrying either deﬁciency (data not shown) to yield transheterozygous embryos. This was done to show that the observed defects were not due to a background mutation elsewhere in the genome. Indeed, both the DC and the HI defects recurred in embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}

CdepMI00496 (Figure 4.12 f).

From this data, it can be concluded that Cdep plays a role in morphogenetic processes during Drosophila embryogenesis. Loss of Cdep causes diﬀerent defects in Drosophila embryonic morphogenesis 31

Figure 4.10: Overlap between Df(3R)ED5066 and Df(3R)Exel6143 with genes likely aﬀected by both deletions.

Figure 4.11: Overview over the Cdep locus with insertion sites of PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496. Only two of the Cdep isoforms are shown, which are almost identical (see text). PBac{5HPw+}CdepB122 carries a P-element insertion before the last two exons whereas Mi{MIC}CdepMI00496 carries a Minos insertion right before the last exon. 32 Results

Figure 4.12: Dorsal closure phenotypes in embryos homozygous or transheterozygous for PBac {5HPw+}CdepB122 and Mi{MIC}CdepMI00496, z-stacks of fluorescent micrographs. Drosophila embryos stained with α-Crb, anterior to the left, dorsal up. a: In stage 14 WT embryos, DC is in progress. The dorsal hole is transiently covered by the amnioserosa (AS). b: In stage 17 embryos, when DC is finished, the epidermis covers the whole embryo. This includes the head organs that have been internalized during head involution, which happens in parallel to DC. c: Embryos transheterozygous for Df(3R)ED5066 and Df(3R)Exel6143 show defects in DC characterized by a dorsal hole. Since the dorsal hole is not covered by the AS, parts of the gut are visible (asterisk). Head involution is also defective, resulting in external head organs (arrow). d: Embryos homozygous for PBac{5HPw+}CdepB122 show the same defects seen in Df(3R)ED5066/ Df(3R)Exel6143 transheterozygous embryos. e: Embryo homozygous for Mi{MIC}CdepMI00496 also show defects in DC and head involution (defect in head involution not shown here due to frame of micrograph) f: Embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 show the same defects that can be observed in Df(3R)ED5066/ Df(3R)Exel6143 transheterozygous embryos. Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 33

Figure 4.13: Overview over the Cdep locus. Information about exons and introns as well as the predicted isoforms was retrieved from www.flybase.org in May 2013. There are three predicted splice variants of Cdep. All of them contain exons 2 through 8 (yellow), which code for the FERM and FERM-adjacent domain (FA) of Cdep. In Cdep-RF, exon 8 has 3 additional base pairs at its 5’ end. Only splice variants Cdep-RE and RF contain exons 15 through 17 (green), which code for the RhoGEF and PH domains. Splice variant Cdep-RG has two large exons that do not code for any known conserved protein domains (magenta). Cdep-RC only contains the exons coding for the FERM and FA domains.

Figure 4.14: Defects in leading edge (LE) cells in embryos transheterozygous for PBac {5HPw+}CdepB122 and Mi{MIC}CdepMI00496, z-stacks of fluorescent micrographs. Embryos stained with α-Crb and α-Sas. anterior to the left, dorsal up. a: During stage 13, LE cells of WT embryos elongate in a dorsal-ventral direction and extend filopodia that contact the amnioserosa (white triangles). b: During stage 14, the LE becomes straight in WT embryos. LE cells mostly have the same width and are parallel to each other. c, d: In stage 16 and 17 PBac{5HPw+}CdepB122/ Mi{MIC}CdepMI00496 transheterozygous embryos, dorsal closure has not finished. LE cells are not elongated and of very uneven width, some splaying out along the anterior-posterior axis (white asterisks). There are hardly any filopodia visible (white triangles).

Cdep lies on the right arm of the third chromosome of Drosophila melanogaster. The locus is about 35 kb long and has been predicted to encode four protein isoforms: Cdep-PE,

Cdep-PF, Cdep-PG and Cdep-PC (Figure 4.13). The mRNAs of Cdep-RE and RF only diﬀer by 3 bp at the 5’-end of exon 8. All predicted isoforms share the same N-terminus containing a FERM domain and a FA domain. Only predicted isoforms PE and PF also contain a RhoGEF domain and two PH domains. The mRNA encoding Cdep-PG contains two long exons that do not encode any known conserved protein domains. Since only 34 Results

Cdep-PE and PF are predicted to contain all protein domains, they will be the main focus of this thesis.

4.3.2 Embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC} CdepMI00496 show defects in the LE during DC

In order to ﬁnd out what might cause the defects in DC, the LE cells were observed more closely. Embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 were stained with α-Sas and α-Crb. Sas is localized in the apical membrane and the

ﬁlopodia extended by the LE cells. In stage 13 WT embryos, the LE is not yet perfectly straight. However, the LE cells already start elongating along the dorsoventral axis and extend ﬁlopodia that make contact with the AS (Figure 4.14 a, white triangles). The

LE becomes straight during stage 14, the LE cells are elongated and evenly distributed

(Figure 4.14 c). In embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}

CdepMI00496, the dorsal hole is never closed (Figure 4.12). When taking a closer look at the LE in those embryos, it becomes obvious that the LE cells are not properly elongated

(Figure 4.14 b and d). They are of diﬀerent width and some of them splay out in an anterior-posterior direction (Figure 4.14 d, asterisks). There is also a much smaller number of ﬁlopodia and they do not extend as far as seen in WT (Figure 4.14 b and d, triangles).

Taken together, these results show that the insertions PBac{5HPw+}CdepB122 and

Mi{MIC}CdepMI00496 cause defects in DC and failures in LE cell elongation and extension of ﬁlopodia.

4.3.3 Insertions in the ORF of Cdep cause defects in tracheal and Malpighian tubule morphogenesis

The defects in DC and head involution caused by the insertions PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 could be due to the role of Cdep in a general process that is important for embryonic morphogenesis of epithelial structures. If this is the case, other epithelia-derived organs could also be aﬀected by these mutations. To ﬁnd out whether this was the case, stage 17 embryos were stained with α-Crb and the morphology of the trachea, SGs and Malpighian tubules was assessed (Figure 4.15).

In WT stage 17 embryos, the trachea are fully developed. All tracheal metameres have formed branches that fused with each other in a stereotypical manner. Most notably, the Loss of Cdep causes diﬀerent defects in Drosophila embryonic morphogenesis 35

dorsal trunk forms a continuous tube that is connected to the exterior at the anterior and

posterior spiracles (Figure 4.15 a; spiracles not shown). The SGs have reached their ﬁnal

position in the embryo, rising from their contact point in the larval mouth ﬁrst dorsally,

then bending posteriorly (Figure 4.15 b). The anterior Malpighian tubules have evaginated

from the hindgut, loop forward into the thoracic region and then bend to point posteriorly

(Figure 4.15 b, b’). Stage 17 embryos transheterozygous for PBac{5HPw+}CdepB122 and

Mi{MIC}CdepMI00496 showed defects in the morphogenesis of the trachea and Malpighian

tubules (Figure 4.15 c and d) The branches of tracheal metameres were very short and had

often not fused. Therefore, the dorsal trunk was not continuous (Figure 4.15 c’ and d’,

triangles). The anterior Malpighian tubules did not reach their ﬁnal position, although

they formed a loop (Figure 4.15 c” and d”). However, the secretory tubes of the SGs

showed a near-WT morphology (Figure 4.15 e and e’).

Thus, in addition to impairing DC and HI, PBac{5HPw+}CdepB122 and Mi{MIC}

CdepMI00496 also aﬀect the morphogenesis of the trachea and Malpighian tubules in the

Drosophila embryo.

4.3.4 The CRISPR/Cas9 system was used to generate loss-of-function mutants of Cdep

The insertions PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 are located in the intron

before the second-to-last and last exon of Cdep, respectively (Figure 4.5). It is therefore

possible that the insertions aﬀect the splicing of the Cdep mRNA. However, they might

also disturb distal regulatory elements for another gene. Taking these possibilities into

account, it became necessary to create loss-of-function (LOF) mutants of Cdep in order to

identify its true function. For this, the CRISPR/Cas9 system was used for two diﬀerent

mutagenetic approaches: A full deletion of the Cdep ORF was created by replacing the

Cdep locus with a DsRed gene under the control of an eye-speciﬁc promoter. Since the

success of this approach could not be estimated beforehand, an easier way of mutagenizing

Cdep was carried out in parallel. In this attempt, an in-frame stop codon was inserted

shortly after the start codon of Cdep. In the latter approach, only 13 out of 120 embryos

injected with the CRISPR components survived to adulthood. The progeny of two of them

screened positive for the novel BglII restriciton site that was inserted alongside with the

stop codon (see Materials and Methods). Surprisingly, sequencing of the positive progeny 36 Results

Figure 4.15: Defects in the trachea and Malpighian tubules in embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496, z-stacks of ﬂuorescent micrographs. Embryos stained with α-Crb, anterior is to the left, dorsal up. All embryos are stage 17. a: In WT stage 17 embryos, the trachea are fully developed. Most notably, the dorsal trunk (DT) forms a continuous tube that is connected to the exterior at the posterior spiracles (not visible here, optical slices not included in z-stack). b: The secretory part of the salivary glands (SG) lies in the ventral anterior part of the embryo. It has a bent shape and its tip points posteriorly. The anterior Malpighian tubules, one of which is visible here (box and b’) have evaginated from the hindgut and loop forward into the thoracic region. The tip points posteriorly. c and d: Stage 17 embryos transheterozygous for PBac{5HPw+}CdepB122 and Mi{MIC}CdepMI00496 show defects in the morphogenesis of the trachea and Malpighian tubules (the boxed regions in c and d are enlarged in c’,c”, d’ and d”). The tracheal metameres have not joined and therefore the dorsal trunk is not continuous (c’ and d’, triangles). The anterior Malpighian tubules do not reach their full length, although they form a loop (c” and d”). The secretory tubes of the salivary glands show a near-WT morphology.

revealed two diﬀerent Cdep alleles. In one of them, codon 16, originally coding for Glutamic acid, is replaced by a stop codon. This allele was called CdepE16X . In the other allele,

Glycine in position 17 is replaced by a stop; it was called CdepG17X . It is notable that

CdepG17X originated from an incomplete insertion of the donor DNA that was used to create the in-frame stop codon. The attempt to replace the complete Cdep locus with DsRed under control of an eye-speciﬁc promoter was also successful. Out of about 200 injected embryos, 87 survived to adulthood. The progeny of three of them screened positive for

DsRed expression in the adult eyes after crossing out the nos::Cas9 transgene on the first Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 37 chromosome (see Materials and Methods). The correct replacement of Cdep with DsRed was confirmed by sequencing. The deletion begins 37 bp downstream of the Cdep start codon and includes all coding exons. It expands to 8 bp upstream of the 3’UTR of Cdep.

The complete genotype of the established deletion allele is Cdep∆,attP,loxP,3xP 3::DsRed,loxP .

It will be called Cdep∆ from now on. It is of interest to mention that all Cdep LOF mutants mentioned here can be kept as homozygous stocks. Figure 4.16 gives an overview over the

Cdep alleles described in this section.

Figure 4.16: Overview over Cdep alelles created with the CRISPR/Cas9 system. a: Scheme of how Cdep∆ was created. Black triangles mark the cut sites for Cas9. A plasmid encoding DsRed under control of the 3xP3 promoter was used for homology repair, which yielded the sequence shown on the bottom. b: Sequences immediately following the transcription start sites of CdepE16X and CdepG17X . The part of the sequences that is still translated into protein is set in bold typeface. While in CdepE16X , the BglII restriction site used for screening (see Materials and Methods) is downstream of the stop codon, it is upstream of the stop codon in CdepG17X . In CdepG17X , only part of the insert used to introduce the stop codon was inserted due to unknown reasons. The part that was inserted lacks the stop codon, but includes the BglII restriction site. By coincidence, a stop codon is formed of the last Thymine of the BglII restriction site and the endogenous Guanine and Adenine nucleotides that follow it.

4.3.5 LOF mutants of Cdep show a variety of phenotypes

In order to investigate the morphological defects caused by Cdep LOF mutants, embryos homozygous for CdepE16X , CdepG17X and Cdep∆ were ﬁxed and stained with DAPI. In 38 Results general, all statements made are true for CdepE16X , CdepG17X and Cdep∆, unless stated otherwise. The three genotypes will be collectively referred to as Cdep−/−.

When looking at the DAPI-stained Cdep−/− embryos, it became immediately apparent that a lot of embryos had obvious morphological defects. The defects ranged from irregularities in the body segments over missing body parts (usually the head), to an entirely diverged morphology that caused embryos to dissolve into several pieces (Figure

4.17). At a low magnification, a high number of such pieces could be observed, probably originating from embryos that have fallen apart during the fixation process. Even though the observed defects were quite severe, a relatively large number of embryos also showed a WT-like morphology. In order to find the percentage of WT-like embryos, all embryos and pieces were counted; each piece was counted as one aberrant embryo since it could not be retraced which pieces belonged together. In case of CdepE16X and Cdep∆, more than 70 % of the units counted were WT-like embryos, whereas for CdepG17X , this was the case for about 50 % of the counted units (Figure 4.18). However, the embryos appearing to have a WT-like morphology in this approach could have milder morphological defects which could not be observed under these experimental conditions. The ratio of embryos affected by loss of Cdep function was also reflected in the ratio of larvae that hatch. In

WT, about 86% of larvae had hatched from eggs incubated for 36 h at 25 ◦C. Under the same conditions, about 60% of CdepE16X larvae hatched, 49 % hatched from Cdep∆ eggs and only 33 % from CdepG17X eggs (Figure 4.19). Altogether, these results show that LOF mutants of Cdep cause morphologigal defects of varying severity in Drosophila embryos.

4.3.6 LOF mutants of Cdep cause denticle belt defects and ventral holes in Drosophila larval cuticles

In order to draw conclusions about the possible function of Cdep, the morphological defects caused by its loss of function needed to be characterized in more detail. For this, milder phenotypes had to be analyzed, as grossly affected embryos showed too many different features. It was assumed that embryos with less severe phenotypes would reach a late stage of embryogenesis, when the larval cuticle is secreted. Since the cuticle is secreted by the epidermis, it reflects morphological features of the epidermis. Therefore, preparing larval cuticles from Cdep−/− embryos and larvae was considered a straightforward way to get an overview over less severe Cdep−/− phenotypes. Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 39

Figure 4.17: Overview over morphological defects in Cdep alelles, fluorescent micrographs. Embryos stained with DAPI. Stage 10 and 13 WT embryos are in dorsal view, anterior to the left. All other WT embryos are dorsal up, anterior to the left. Cdep−/− embryos are depicted with their presumed anterior pole to the left, dorsoventral orientation is arbitrary. a: Different stages of WT embryos stained with DAPI. The body segments are discernible. The dorsal hole is visible as a dark grey area. b-e: Cdep−/− embryos show a variety of phenotypes. The embryo in b seems to be at an early stage, as no body segments are visible. However, the morphology is abnormal. c: This embryo has apparently suffered an arrest in germ band retraction and it shows multiple other defects. d: Germ band retraction seems to be arrested. The tissue normally forming the head has mostly dissolved. e: This embryo has no discernible morphological features and appears to be breaking into several pieces. f: Overview over DAPI-stained WT embryos at a low magnification, stitched fluorescent micrograph. One piece probably originating from a dissolved embryo is visible (triangle). g: Overview over DAPI-stained Cdep−/− embryos at a low magnification, stitched fluorescent micrograph. A relatively high number of pieces can be observed. They were likely caused by embryos dissolving during fixation due to severe morphological defects as seen in e.

Figure 4.18: Plotted ratio of Cdep−/− embryos showing morphological defects. Light grey represents the percentage of embryos showing a noticable morphological defect in DAPI stainings (see also Figure 4.17). Dark grey represents the percentage of embryos not showing any observable defects. The total number of embryos counted per genotype (n) is given behind each bar. 40 Results

Figure 4.19: Ploted ratio of Cdep−/− larvae hatching from eggs. Light grey represents the percentage of eggs not yielding larva after an incubation period of 36 h at 25 ◦C. Dark grey represents the percentage of empty egg shells found on the substrate. The total number of laid eggs (n) is given behind each bar.

WT embryonic and larval cuticle preparations clearly show the head skeleton, which has a fishhook-shaped appearance from a lateral view and the shape of a long, narrow "V" from a ventral or dorsal view (Figure 4.20 a, b and f). Eight denticle belts are discernible on the ventral side of the cuticle. At the dorsal-posterior pole, the filzkörper with the posterior spiracles are clearly evident. Two different features became immediately apparent when observing cuticles from Cdep−/− embryos: In more severely affected specimens, anterior holes were visible in the cuticle, leading to the assumption that these embryos had no head. The holes could be rather large, including big parts of the ventral side of the cuticle as well. In less severe cases, denticle belts were fused and defects in the head skeleton could be noticed. Some of the larvae that had hatched also showed fused denticle belts or belts that were only present in about one half of the ventral cuticle. In these cases it was especially noticable that the belts were not simply cut off, but formed tips towards the ventral midline. When several neighboring denticle belts were only half-formed on the same side of the cuticle, their medial-directed ends seemed to converge towards one point on the midline (Figure 4.20 h). When this was the case, the larva had a shortened, kinked appearance when compared to WT.

When counting the number of embryonic and larval cuticles aﬀected by the aforementioned defects, it turned out that the maximum percentage of cuticles with aﬀected morphologies was 14%. This number was found for cuticles from CdepE16X embryos and larvae (Figure

4.21). In Cdep∆ cuticles, 12% showed defects, whereas only 7% of CdepG17X cuticles showed an abnormality, the same number as was found in WT cuticles. However, in WT cuticles the most abundant abnormality was that the embryos were still surrounded by a vitelline Loss of Cdep causes diﬀerent defects in Drosophila embryonic morphogenesis 41

membrane, meaning that the larvae had not hatched, albeit the cuticle showed a WT-like

phenotype. In Cdep−/− cutcicle preparations, the most abundant phenotypes were those

described above and depicted in 4.20. Furthermore, as shown above, only about 50 % of

CdepG17X embryos showed a WT-like morphology in DAPI stainings. One can therefore

assume that only a small portion of CdepG17X embryos even develop far enough to secret

a cuticle.

In summary, Cdep−/− embryonic cuticles show anterior holes and defects in the head

skeleton. Some Cdep−/− larval cuticles display fused denticle belts, which leads to a

shortened, kinked body. This conﬁrms the earlier hypothesis that Cdep plays a role during

morphogenetic processes involving epithelia in Drosophila embryogenesis. 42 Results

Figure 4.20: Morphological defects observed in cuticles of Cdep−/− embryos and larvae, darkfield micrographs. Brightness and contrast has been enhanced to improve visibility. Embryonic cuticles (a-e) are surrounded by the vitelline membrane and are anterior to the left. Larval cuticles (g-h) are anterior up. a: Cuticles of WT embryos, ventral view. In the upper image, the forked head skeleton (hs) is visible at the anterior pole. In the lower image, the eight denticle belts (db, numbered) are visible. b: Cuticles of WT embryos, lateral view. In the upper image, the hs is visible at the anterior pole, the filzkörper with the posterior spiracles (ps) can be seen at the posterior pole. In the lower image, the eight db are visible on the ventral side of the embryo. c: Cuticle of Cdep−/− embryo in ventral view. The distances between the db are irregular, two of them contact each other (triangle). c’: 2×blowup of the anterior pole of the cuticle in c showing defects in the hs. d: Cuticle of Cdep−/− embryo in lateral view. At the anterior pole, a hole is apparent (asterisk). e: Cuticle of Cdep−/− embryo in ventral view. The whole anterior-ventral part of the cuticle is missing (asterisk). f: Cuticle of a hatched WT L1 larva in lateral view, ventral to the right. The hs, ps and the eight db are clearly discernible. g: Cuticle of a hatched Cdep−/− larva in ventral view. The hs and ps have a WT-like morphology. Db 2 and 3 as well as 4 and 5 have fused with each other (triangles). h: Cuticle of a hatched Cdep−/− larva in ventral view. The WT-like hs seems to be visible in a lateral view whereas the rest of the cuticle appears to be in a ventral orientation. Db 2 through 4 as well as 6 are only present in the left half of the cuticle (triangles). Their medial ends seem to converge towards one point on the midline. Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 43

Figure 4.21: Plotted ratio of embryonic and larval Cdep−/− cuticles showing morphological defects. The phenotype called "unhatched WT-like" refers to cuticles that show no obvious morphological defects but are still surrounded by the vitelline membrane. "Cdep-like" refers to embryonic cuticles with denticle belt defects, mouth hook defects and/ or anterior holes that can extend ventrally (Figure 4.20). 44 Results

4.3.7 Phenotypes in Cdep−/− mutant embryos are likely not caused by maternal defects

Given the high number of embryos dissolving into pieces, the question arose whether this was due to a maternal defect. It was reasoned that a defect in egg formation or deposition of the vitelline membrane could cause the observed phenotypes. In order to answer this question, two approaches were chosen: Ovaries were dissected from WT and

Cdep−/− females and stained with DAPI and Phalloidin to assess possible morphological defects in the follicles. Furthermore, Cdep−/− virgins were crossed with WT males and vice-versa, and cuticle preparations were carried out. The ovarioles and follicles of Cdep−/− females had a WT-like appearance (Figure 4.22). When comparing the ratios of diﬀerent phenotypes seen in the cuticles of embryos from WT females and Cdep−/− males to those of homozygous WT embryos, no marked diﬀerences occured (Figure 4.23). The same is true for the cuticles of embryos from Cdep−/− females and WT males, with one exception:

The ratio of embryos with cuticular defects was higher when the embryos received one

CdepG17X allele from the mother and a Cdep +/+ allele from the father, when comparing to the ratio of defects observed for the cuticles of CdepG17X homozygous embryos. This is surprising as it hints at the possibility that CdepG17X is a dominant allele. However, the experiment would have to be repeated several more times to allow for a deﬁnite conclusion.

These results do not completely rule out the possibility that the phenotypes observed in

Cdep−/− embryos derive from maternal defects. However, no aberrations could be found in Cdep−/− follicles. The chorion and vitelline membrane of Cdep−/− eggs did not show any apparent abnormalities and seemed as firm and stable as WT chorion and vitelline membrane (not shown). Furthermore, in almost all cases the offspring of a Cdep−/− mother and WT father did not show a different distribution of phenotypes from embryos with reversed parental genotypes or homozygous WT embryos. It is therefore unlikely that a maternal defect causes the full range of phenotypes observed in Cdep−/− embryos, although there could be a small maternal contribution.

4.3.8 LOF mutants of Cdep cause segments to fuse

The fused denticle belts in Cdep−/− embryonic and larval cuticles and the kinked appearance of some larvae raised the question whether body segments were initially not being established Loss of Cdep causes diﬀerent defects in Drosophila embryonic morphogenesis 45

Figure 4.22: Ovarioles from Cdep−/− females, z-stacks of ﬂuorescent micrographs. Ovarioles were stained with Phalloidin (green) and DAPI (biue). Cdep−/− ovaries have a WT-like appearance. The egg chamber has a regular shape and is surrounded by the monolayered follicular epithelium. Actin levels are not diﬀerent from those in WT.

Figure 4.23: Plotted ratio of cuticle phenotypes from animals heterozygous for Cdep. When the Cdep allele is mentioned first in superscript, i.e. "CdepE16X/+", it means that the Cdep allele came from the mother. If the Cdep allele is mentioned second, as in "Cdep+/E16X ", the Cdep allele came from the father. Black represents ratio of larval cuticles that had a WT-like appearance whereas blue represents the ratio of embryonic cuticles showing phenotypes typical for Cdep−/− mutants. The uppermost plot shows the ratios for WT embryos and larvae. The upper plot for each group of three shows the ratios for embryos and larvae homozygous for each Cdep LOF allele. The second plot in each group shows the ratios for embryos and larvae from a Cdep−/− mother and a WT father, whereas in the lowest plot for each group the parents’ genotypes are reversed. All heterozygous combinations of a Cdep allele and WT show WT-like numbers, except for CdepG17X/+, which shows a higher number of affected cuticles than was found for cuticles from embryos and larvae homozygous for CdepG17X . during embryonic development or whether they were generated at first, but then disappeared during later stages of development. In order to find out which was the case, embryos carrying the allele DE-cad::GFP (Huang et al., 2009) combined with the Cdep−/− alleles 46 Results were used for live imaging. In DE-cad::GFP embryos not carrying a Cdep−/− allele,

GBR progressed normally (Figure 4.24 a; Supplementary movie WT_whole). The AS covered the dorsal hole, which was subsequently closed via DC. The embryo showed a WT-like morphology with clearly discernible body segments. When observing the morphogenesis of a DE-cad::GFP; Cdep−/− embryo, several defects became apparent

(Figure 4.24 b; Supplementary movie Cdep_whole). The embryo showed an irregular morphology from the start of the experiment, when it was presumably at stage 11, judged by the beginning of GBR. While GBR progressed, part of the amnioserosa appeared to be attached to a posterior area of the epidermis instead of the epidermal edge. Furthermore, the establishment of body segments could not be observed during the entire movie. However, some segmental characteristics were present, such as tracheal metameres. When looking at the movie in more detail, some intriguing features can be noticed: Two tracheal metameres fused during GBR, giving rise to an X-shaped structure of dorsal and ventral branches

(Figure 4.25 b; Supplementary movie Cdep_trachea). Furthermore, a hole appeared in the ventral epidermis, but was healed in the course of GBR (Figure 4.25 c; Supplementary movie Cdep_ventral-hole). However, these morphological defects were not observed in all DE-cad::GFP; Cdep−/− used for live imaging. As mentioned above (Section 4.3.5), a number of Cdep−/− embryos showed a WT-like morphology and therefore it is not surprising that the defects described here could only be found in some DE-cad::GFP;

Cdep−/− embryos. From this, it is reasonable to assume that body segments are initially established in aﬀected embryos. Some segments or parts thereof collpase during embryonic development, causing neighboring structures to fuse.

Especially the appearance and subsequent healing of an epidermal hole, but also the fusion of two tracheal metameres raised the question whether parts of embryonic segments died and were replaced by neighboring tissues moving in to close the gap. This would also explain the fusion of denticle belts described above. To answer this question, semi-thin sections of Cdep−/− embryos were prepared and stained with methylene blue and tolouidine blue to visualize dead tissue. In semi-thin sections of WT embryos, the yolk appears as a mass of dark spots, all other tissues are light grey (Figure 4.26 a). Morphological features are clearly distinguishable. In semi-thin sections of some Cdep−/− embryos a number of dark clusters bigger than single cells were visible within body structures (Figure 4.26 b and c). While the dark structures did not appear in regions large enough to represent whole Loss of Cdep causes diﬀerent defects in Drosophila embryonic morphogenesis 47 segments, they often occured in body regions close to each other (Figure 4.26 b, stage

12). In a stage 7 CdepG17X embryo showing a severely aﬀected morphology, only a small number of dark spots was visible (Figure 4.26 c). Taken together, these results suggest that apoptosis plays a role in the emergence of the observed phenotypes. It is possible that cell death mainly occurs in earlier stages, causing enough damage to bring forth the progressive morphological defects observed in later stages. 48 Results

Figure 4.24: Still frames from live imaging of Cdep−/− embryos. Time series of fluorescent micrograph z-stacks. Areas where no epidermal cells outlined with DE-cad::GFP are visible where caused by an unforeseen upwards shift of the microscopic stage during the imaging process. a: DE-cad::GFP embryo. Germ band retraction begun about 100 minutes after the experiment was started and proceeded until about 250 minutes into the experiment. Subsequently, dorsal closure was carried out. The regular morphology of the embryo can be appreciated. The amnioserosa is clearly visible. The edge of the epidermis abutting the amnioserosa is smooth and regular. Segments start becoming apparent at 200 minutes and are fully established 300 minutes into the experiment. b: DE-cad::GFP; Cdep−/− embryo. Germ band retraction had just begun when the experiment was started and finished at about 200 minutes. Dorsal closure started, but was not quite completed at 370 minutes, when the experiment was finished. After 50 minutes, some irregularities in the morphology become apparent. The epidermal edge is uneven. Part of the amnioserosa seems to reach over the posterior epidermis and is attached there instead of the epidermal edge (triangle). This attachment is maintained until germ band retraction is finished (The insets show 2 ×blowups of the boxed areas). Furthermore, the beginning establishment of the body segments cannot be observed and the overall morphology appears disordered. As a consequence, some defects can be seen in the trachea at 370 minutes: a tracheal metamere in the anterior part of the embryo extends two dorsal and two ventral branches, when it should only have one of each (arrow). Furthermore, the tracheal dorsal trunk shows gaps (asterisk). Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 49

Figure 4.25: Still frames of details of tracheal and epidermal defects seen in live imaging of Cdep−/− embryos. Time series of ﬂuorescent micrograph z-stacks. The boxed areas in a - c are blown up in a’ - c’. The contrast in all images was enhanced for improved visibility. a’: DE-cad::GFP embryo, showing two tracheal metameres (arrows). Two tracheal pits appear 30 minutes into the experiment. During germ band retraction, they move along with the epidermis while they invaginate, maintaining their distance to each other. b’: DE-cad::GFP; Cdep−/− embryo, two tracheal metameres (arrows). Two tracheal pits have been established at the beginning of the experiment. During germ band retraction, they move along with the epidermis while they invaginate, but also reduce the distance between them until, after 140 minutes, they have fused. At the end of the experiment, the structure originating from these two metameres extends two dorsal and two ventral tracheal branches that form an "X" (triangles). c’: DE-cad::GFP; Cdep−/− embryo, part of the ventral epidermis. 60 minutes into the experiment, a ventral hole appears in the epidermis (arrow). It travels along with germ band retraction, becoming smaller. Eventually, after 290 minutes, it is not visible anymore. 50 Results

Figure 4.26: Semi-thin sections (2 µm) of Cdep−/− embryos stained with methylene blue and tolouidine blue. Brightfield micrographs. Anterior to the left. a: Sagittal (st 10 and 14) or medial (st 17) section through WT embryos. In st 10, the fully extended germ band is visible. The yolk appears as a mass of round dark spots. In stage 14, the pharynx (ph) is apparent at the anterior end, whereas at the posterior end the hindgut (hg) is visible. In stage 17, the stomodeum (st) can be distinguished at the anterior end. Parts of the yolk-filled midgut (mg) can be seen. Also the posterior parts of the tracheal dorsal trunks (dt) are visible as symmetrical structures. The epidermal segments are clearly evident. b: Two sagittal sections through a stage 11 and two paramedial sections through a stage 12 CdepE16X embryo. In both embryos, there are several relatively large dark spots visible in the anterior part of the embryo, in the stage 12 embryo dark spots can also be observed in the posterior part around the retracting germband (arrows). In the stage 11 embryo, the epidermis is damaged in the ventral posterior part. The rounded edges of the epidermis around the breaking point implicate that this is not due to mechanical damage of the specimen (asterisk). c: Two paramedial sections through a stage 17 CdepG17X embryo. The embryo shows multiple morphological defects. There seem to be no epidermal segments. However, the presence of the head skeleton, discernible by small, spiky structures in the stomodeum, marks stage 17. Part of the posterior-lateral body, possibly the epidermis, seems to be missing. The rounded edges of the present body parts implicate that this is not due to mechanical damage of the specimen (asterisk). Some dark structures appear at the anterior and posterior pole (arrows). Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 51

4.3.9 Defects in Cdep−/− mutants are not due to actin mislocalization

The cytoskeleton component actin plays a major role in embryonic morphogenesis, since the cell rearrangements that take place during this process rely on the reorganization of the actin cytoskeleton. It was therefore asked whether the defects observed in Cdep−/− mutants occured due to defects in actin localization. Embryos homozygous for Cdep−/− were stained with α-Sas, to show the overall morphology, and Phalloidin to visualize actin.

When looking at the lateral epidermis of stage 14 WT embryos, the Sas signal outlines the epidermal cells. Actin is localized at the cell cortex in WT epidermal cells (Figure

4.27 a and a’). In Cdep−/− embryos, Sas localization seemed to be less deﬁned, the cell outlines could not be distinguished as clearly as in WT (Figure 4.27 b and c). However, actin localization was very similar to WT (Figure 4.27 b’ and c’).

Actin is also a major component of muscles. When staining stage 17 embryos with

Phalloidin, the larval muscles become clearly visible. In WT embryos, the stereotypical muscle pattern is visible (Figure 4.28 a). While in Cdep−/− embryos the muscles contained similar amounts of actin like in WT embryos, the pattern of the muscles was sometimes disrupted (Figure 4.28 b and c). In some embryos only parts of the muscle pattern were aﬀected, while in others the pattern was completely aberrant. From these results it can be concluded that actin is not mislocalized in epidermal cells of Cdep−/− embryos. The aberrant muscle pattern likely reﬂects earlier morphological defects.

4.3.10 Cdep genetically interacts with yurt

As established in Section 4.3.6, the Cdep−/− phenotype is characterized, among other things, by a malformation of the head skeleton or an anterior hole in the embryonic cuticle.

This phenotype was reminiscent of the yrt75a phenotype, where the head is also often impaired. It was asked whether Cdep functionally interacts with yrt. Therefore, double mutants of yrt75a and Cdep−/− were established. The alleles had to be recombined, since both genes are located on the right arm of chromosome 3. Double mutants were examined for their cuticular phenotypes.

After recombination of each Cdep allele with yrt75a, the presence of the Cdep alleles carrying a premature stop codon was conﬁrmed via PCR ampliﬁcation of the region around the stop codon and subsequent digestion of the PCR product with BglII, the novel 52 Results

Figure 4.27: Sas and actin staining in the lateral epidermis of Cdep−/− embryos, z-stacks of ﬂuorescent micrographs. The brightness of all stainings was enhanced to the same extend to improve visibility. Dorsal up, anterior to the left. In all Sas stainings, the insets are 2 ×blowups of the boxed area. a and a’: Lateral epidermis of WT embryo. Sas (a) outlines all epidermal cells. Actin (a’) is localized at the cell cortex. The dark areas mark the nuclei. b - c’: Lateral epidermis of CdepG17X and Cdep∆ embryos. The Sas staining (b, c) is more diﬀuse than in WT in some areas (insets). Actin localization is WT-like (b’, c’).

Figure 4.28: Actin staining in muscles of stage 17 Cdep∆ and CdepG17X embryos, z-stacks of fluorescent micrographs. Dorsal up, anterior left in a and b, ventral view with anterior to the left in c. a: Phalloidin-stained muscles of WT embryo. The stereotypical pattern of the larval muscles is visible. b and c: Phalloidin-stained muscles of Cdep−/− embryos. The embryo in b seems to have a relatively normal muscle pattern in its posterior body half. However, in the anterior half a number of muscle fibers are missing, one fiber diverts from the visible muscles. In the embryo depicted in c, the normal muscle pattern is severely disrupted. restriction site inserted together with the stop codon (Figure 4.29). The presence of Cdep∆ was confirmed by the red fluorescence of the adult eyes. In order to confirm the absence of Yrt protein from all Cdep−/−, yrt75a double mutant lines, a Western Blot was carried out, using protein extracted from stage 12 or older embryos homozygous for the desired alleles. The membrane was probed with α-Yrt Ab. However, several attempts to detect Yrt Loss of Cdep causes different defects in Drosophila embryonic morphogenesis 53

even in extracts from WT embryos proved unsuccessful. Therefore, a diﬀerent approach

was chosen. All yrt75a and Cdep−/−, yrt75a lines are homozygous lethal. It had to be

excluded that in the double mutants this was due to a background mutation. Therefore,

the Cdep−/−, yrt75a lines were crossed with yrt75a, reasoning that if yrt75a was present in

the double mutant, all adult oﬀspring must keep the balancer chromosome. This was the

case, therefore it is very likely that the phenotypes observed in Cdep−/−, yrt75a double

mutants are indeed caused by the combination of the desired mutant alleles.

This is an example for a crossing scheme to test Cdep−/−, yrt75a double mutants:

75a − 75a 75a w−;;yrt × w−;;Cdep , yrt ⇒ w−;; yrt (lethal) TTG TTG Cdep−,yrt75a ♂ In case of a background mutation, the genotype of the oﬀspring would be transheterozygous

for Cdep − and yrt75a, allowing for the segregation of the balancer:

75a − − 75a w−;;yrt × w−;;Cdep , other ⇒ w−;; yrt (viable) TTG TTG Cdep−, other− ♂ Cuticles from yrt75a embryos frequently

show a hole at the dorsal-posterior pole

(Figure 4.30 b and d). This hole often

pushes the ﬁlzkörper with the posterior

spiracles away from each other. If the hole is Figure 4.29: Agarose gel showing products of BglII digest of PCR product ampliﬁed from exon too small to clearly distinguish, its presence 2 of CdepE16X , yrt75a and CdepG17X , yrt75a. Genomic DNA from embryos homozygous for each can still be concluded from the aberrant allele was used as a template. The amplicon was generated with the primer pair AM241+AM242 appearance of the ﬁlzkörper (Figure 4.30 b). and has a length of 725 bp (triangle). After a BglII Another characteristic yrt75a phenotype is digest, the amplicon is cut into two fragments of 512 and 213 bp length, respectively (arrows). The E16X 75a a defect in the head skeleton or an anterior absence of a 725 bp band for Cdep , yrt and CdepG17X , yrt75a shows that no WT allele of hole (Figure 4.30 b-d). If the latter is the Cdep is present in either strain.

case, the head skeleton is missing entirely.

Altogether, 51 % of analyzed yrt75a cuticles showed one of these phenotypes or a combination

thereof (Figure 4.31). However, some yrt75a mutant larvae hatch, although they do not

reach adulthood. In these larvae a dorsal kink can often be observed (Figure 4.30 e). This

was the case for 13 % of the examined yrt75a animals. When recombining the yrt75a with

Cdep−/−, a surprising ﬁnding was made. In cuticles from embryos and larvae homozygous 54 Results for CdepE16X , yrt75a and CdepG17X , yrt75a no more yrt-like phenotypes were observed

(Figure 4.31). In fact, no more phenotypes seen in CdepE16X and CdepG17X mutants were observed, either. This is especially noticeable when comparing CdepE16X to CdepE16X , yrt75a mutants (14 % of Cdep-like phenotypes in CdepE16X vs. 1 % in CdepE16X , yrt75a).

Conversely, the number of embryonic and larval cuticles showing yrt-like defects in Cdep∆, yrt75a homozygotes was higher than that in yrt75a embryos and larvae. However, it has to be noted that in Cdep∆, yrt75a double mutants the ratio of hatchlings is higher than that in yrt75a single mutants (67 % in Cdep∆, yrt75a vs. 53 % in yrt75a). To investigate whether this is a signiﬁcant diﬀerence, the experiment will have to be repeated.

Taken together, these results show that there is a genetic interaction between Cdep and yrt. Surprisingly though, the results were very diﬀerent for double mutants of yrt75a and the Cdep alleles carrying a premature stop codon (CdepE16X and CdepG17X ) and yrt75a and the deletion allele Cdep∆.

Figure 4.30: Cuticle preparations of yrt75a embryos and larvae. Darkﬁeld micrographs. Cuticle in a is in lateral view. Cuticles in b, d and e are in dorsal view, c is in ventral view. All cuticles are with their anterior to the left. a: Cuticle of WT embryo. The head skeleton and posterior spiracles are visible. b - d: Cuticles of yrt75a embryos. Typical phenotypes observed are dorsal-posterior holes (triangles) and anterior holes (asterisks). e: Cuticle of a yrt75a larva. A dorsal-posterior kink is visible (triangle). Loss of Cdep causes diﬀerent defects in Drosophila embryonic morphogenesis 55

Figure 4.31: Plot of ratio of Cdep−/−, yrt75a cuticles showing morphological defects. The term "yrt-like" refers to embryonic cuticles showing a dorsal hole and/ or defects in the head skeleton or an anterior hole. Since Cdep−/− embryos also show defects in the head skeleton and anterior holes, these phenotypes were also counted as "yrt-like" in double mutants. The total number of embryos counted per genotype (n) is given behind each bar.

4.3.11 Cdep genetically interacts with cora

Another member of the FERM-FA subclass of scaﬀolding proteins is Cora. Loss of cora

function results in dorsal holes in the embryonic cuticle, and Lamb et al. (1998) reported

one allele to cause anterior holes. Therefore, ﬂies carrying the cora allele cora5 (cora lies on

the second chromosome of D.melanogaster) were crossed with CdepG17X and Cdep∆. Due

to time restrictions, one initial experiment could be carried out, assessing the cuticular

phenotypes of cora5; CdepG17X and cora5; Cdep∆ double mutants.

Embryos homozygous for cora5 show a dorsal hole in the cuticle (Ward et al., 1998; Lamb et al., 1998). Small holes are restricted to the medial cuticle area, whereas large holes can

extend to the posterior pole, resulting in a gap between the ﬁlzkörper. Furthermore, cora5

causes defects in the SGs. This is reﬂected by necrotic SG remnants in cuticle preparations

(Figure 4.32 Ward et al., 1998). This phenotype was partly rescued when combining cora5

with CdepG17X (100 % of cora-like phenotypes in cora5 vs. 19 % in cora5; CdepG17X ).

However, as was the case for yrt75a, Cdep∆ does not rescue the cora5 phenotype (100 % of

cora-like phenotypes in cora5 vs. 96 % in cora5; Cdep∆). All ratios of the diﬀerent cuticular 56 Results phenotypes observed in cora5 single mutants and cora5; CdepG17X and cora5; Cdep∆ double mutants are given in Figure 4.33.

Figure 4.32: Cuticle preparations of cora5 embryos. Darkﬁeld micrographs. Cuticle in a is in lateral view. Cuticles in b and c are in dorsolateral view. All cuticles are with their anterior to the left. a: Cuticle of WT embryo. The head skeleton and posterior spiracles are visible. b and c: Cuticles of cora5 embryos. Typical phenotypes observed are dorsal-medial holes that can extend to the posterior pole (asterisks). Usually, necrotic remnants of the salivary glands can also be seen (triangles).

Figure 4.33: Plotted ratio of cora5; CdepG17X and cora5; Cdep∆ embryonic cuticles showing morphological defects. The term "cora-like" refers to embryonic cuticles showing a dorsal hole and necrotic remnants of the salivary glands. The total number of embryos counted per genotype (n) is given behind each bar. 57 5 Discussion

5.1 The role of CG17362 and CG9040 in SG lumen stability

5.1.1 Impairments in the expression of CG17362 and CG9040 could be the cause for the intermittent tube closure in the embryonic SGs

In this thesis, I have shown that two diﬀerent Minos insertions in the ORF of the

uncharacterized Drosophila melanogaster gene CG17364 cause an intermittent tube closure

phenotype in the embryonic SGs.

Two other genes, CG17362 and CG9040, lie upstream of the Minos insertions on the

DNA strand complementary to CG17364, within an intron of CG17364. Since no antibodies

are available for any of the protein products of the aforementioned genes, it could not

be assessed which of the three genes was aﬀected by the Minos insertions. The sequence

covering both Minos insertions as well as CG17362 and CG9040 was used for a prediciton of

potential promoter regions with McPromoter006. The only part of the sequence that has a

probability to be a promoter is within the ORF of CG17362. A homeodomain transcription

factor binding site identiﬁed by ChIP analysis is disrupted by Mi{ET1}CG17362MB04765

(modENCODE Consortium et al., 2010). Otherwise, no known or predicted sequences

inﬂuencing expression of CG17362 and CG9040 are impaired by the Minos insertions used

in this thesis. However, expression data obtained from FlyBase showed that CG17362 and

CG9040 both have extremely high expression levels in late-stage Drosophila embryos as well

as in the SGs of third-instar larvae. Furthermore, RNA in situ hybridization carried out

for this thesis showed that CG17362 is strongly and exclusively expressed in the embryonic

SGs in WT. Onthe other hand, CG17364 shows moderate to high expression levels in late

embryos, pupae and adult heads, but not in the larval SGs, according to data from FlyBase.

This hints at the possibility that the observed SG phenotype stems from the impairment

of CG17362 and probably CG9040 expression.

5.1.2 CG17362 and CG9040 could be necessary for SG lumen dilation and stability of lumen diameter

In embryos homozygous for either CG17364MB01315 or CG17362MB04765, the SGs initially

develop normally during embryogenesis. Their invagination, migration and growth is very 58 Discussion similar to WT embryos. Only in stage 17 embryos does the SG lumen show a phenotype that is characterized by bloated areas alternating with seemingly collapsed lumen. Both

CG17362 and CG9040 are expressed late in embryogenesis, when the SG lumen normally widens. In the Drosophila tracheae, a secretory burst at the end of embryogenesis dilates the lumen of the dorsal trunk (Tsarouhas et al., 2007). During this process, proteins important for the assembly of the chitinous aECM of the tracheal dorsal trunk lumen are also secreted (Devine et al., 2005; Luschnig et al., 2006). The tracheal aECM plays a crucial role in the length and diameter of the dorsal trunk. A SG phenotype strongly reminiscent of what I report here has been shown in Drosophila embryos homozygous for either of two prolyl hydroxylases PH4α SG1 and PH4α SG2 (Abrams, 2006). Both are subunits of ER enzymes and hydroxylate prolin residues in secreted proteins (Kivirikko and Pihlajaniemi, 1998). Another Drosophila mutant showing an intermittent tube closure phenotype in the embryonic SGs is pasilla4 (Seshaiah et al., 2001). Pasilla (Ps) is a nuclear splicing regulator. In both studies it was marked that the secretory content of SGs from embryos mutant for the studied gene was lower than in SGs from WT embryos (Abrams,

2006; Seshaiah et al., 2001). The ps mutant allele ps4 caused a reduction in the number and size of apical secretory granules in the embryonic SG (Seshaiah et al., 2001). While Ps probably has an indirect role in SG lumen maintenance, PH4α SG1 and PH4α SG2 likely directly modify secreted proteins and could therefore contribute to the aECM of the SG lumen in Drosophila. The aECM of the SGs does not contain chitin. However, Abrams

(2006) reported ﬁbrillar structures in the lumen of late-stage SGs, although the molecular nature of these ﬁbrils is still unknown. It is likely that the dilation of the SG secretory tubes also requires a secretory burst, since apical secretory vesicles become visible in the

SG secretory cells in late embryos (Abrams, 2006). It is therefore possible that CG17362 and CG9040 encode proteins that play a role in this secretory burst, i.e. by assisting in the splicing of the mRNA of aECM components or by being part of the secretory machinery.

However, from this data it cannot be concluded what the molecular role of the proteins encoded by CG17362 and CG9040 could be. In order to elucidate this further, the number of secretory vesicles and the content of late-stage SG lumina would have to be determined.

An overview over possible roles of CG17362 and CG9040 in the building or maintenance of the SG aECM is given in Figure 5.1. The role of Cdep in Drosophila embryonic morphogenesis 59

Figure 5.1: Possible roles of CG17362 and CG9040 in the apical extracellular matrix (aECM) of the Drosophila larval salivary glands (SGs). a: CG17362 and CG9040 could encode splicing factors that process mRNAs of aECM components. b: CG17362 and CG9040 might aid in the packing of aECM components into vesicles c: CG17362 and CG9040 might assist in the transport of secretory vesicles. d: CG17362 and CG9040 could aid the apical fusion of secretory vesicles.

5.2 The role of Cdep in Drosophila embryonic morphogenesis

In this thesis, three diﬀerent kinds of Cdep alleles were assessed: a) PBac{5HPw+}CdepB122

and Mi{MIC}CdepMI00496 are intronic insertion allles. b) CdepE16X and CdepG17X carry

a premature stop codon. c) Cdep∆ is a full deletion of the Cdep ORF.

I have shown that insertions in the ORF of Cdep cause defects in DC and impair fusion

of the tracheal dorsal trunk in Drosophila embryos. During DC, LE cells do not elongate

properly along the dorsoventral axis and instead often splay out in an anterior-posterior

direction. This can lead to the anterior-posterior expansion of the LE of whole segments, in

turn excluding neighboring segments from the LE. Many branches of tracheal metameres do

not migrate properly, causing the metameres to remain isolated from each other all through

embryonic morphogenesis. The P-element inserted in PBac{5HPw+}CdepB122 is 94 bp

upstream of exon 15, whereas in Mi{MIC}CdepMI00496, a Minos element, is inserted 2 bp

downstream of exon 17. Intronic mutations have been reported to interfere with splicing in

some human diseases including Retinitis Pigmentosa (Davidson et al., 2013; Khan et al.,

2015), and in defects in mice and C.elegans (Kameya et al., 2002; Wells et al., 2015). PBac

{5HPw+}CdepB122 could aﬀect splicing, causing exon 15 to be skipped. Since exon 15

encodes the PH sites necessary for membrane association of Cdep, the PBac{5HPw+}

CdepB122 allele could cause mislocalization of Cdep. Mi{MIC}CdepMI00496 possibly impairs

splicing of exon 17. This exon does not encode any known domains. However, its absence

may cause problems in protein folding.

A premature stop codon or deletion of the Cdep ORF cause multiple morphological defects

in Drosophila embryos. These defects are characterized by the fusion of epithelia-derived

segmental body structures such as tracheal metameres and denticle belts. Furthermore, in

some embryos the AS degenerates long before DC is accomplished. Also, anterior defects 60 Discussion reﬂected in the dissolution of the head and anterior holes in the embryonic cuticle have been observed. While ﬂies carrying either CdepG17X , CdepE16X or Cdep∆ are homozygous viable and fertile, the rate of larvae hatching from eggs is reduced when compared to

WT. While it is very probable that no Cdep protein is left in Cdep∆ mutants, alternative start codons could be used to translate a 5’ truncated version of Cdep in CdepE16X and

CdepG17X . Indeed, Cdep mutants carrying a premature stop codon rescue the cuticular defects seen in yrt75a and cora5 embryos, while a full deletion of the Cdep locus does not, or only in a very limited fashion. It is therefore possible that CdepE16X and CdepG17X mutants produce an altered version of Cdep that could be structurally diﬀerent from the

WT version (Section 5.2.3).

5.2.1 Cdep is a regulator of the JNK pathway

5.2.1.1 Mutations in TGF-β pathway components cause LE bunching and gaps in

the tracheal dorsal trunk

In Drosophila DC, expression of the TGF-β ligand Decapentaplegic (Dpp) is activated by the Jun-N-terminal Kinase (JNK) pathway (Riesgo-Escovar and Hafen, 1997). The

Drosophila JNK is encoded by basket (bsk) (Sluss et al., 1996). Ricos et al. (1999) have shown that mutations in the Dpp receptor genes thickveins (tkv) and punt (put) cause anterior-posterior stretching of LE cells, extending to whole segments, and consequentially the bunching of neighboring cells or segments. The same is true for mutants of the

TGF-β transcription factor Schnurri (Shn) which acts downstream of Dpp signalling (Ricos et al., 1999). The CREB (cAMP response element-binding protein) binding protein dCBP is a coactivator of the Dpp-activated transcription factor Mothers against Dpp (Mad).

Mutations in dCBP and its interactor Modulo (Mod) cause fusion of denticle belts and head defects (Bantignies et al., 2002).

Eivers et al. (2009) report occasional denticle belt fusions in embryos homozygous for the dpp null allele dppH64. Migration of dorsal and ventral branches of tracheal metameres require Dpp signalling (Wappner et al., 1997; Vincent et al., 1997; Llimargas and Casanova,

1997). A mutation in the Dpp receptor tkv causes occasional gaps in the dorsal trunk and general failure in development of the tracheal tree (Llimargas and Casanova, 1997).

The phenotypes described in these reports closely resemble those observed in the Cdep−/− The role of Cdep in Drosophila embryonic morphogenesis 61

mutants characterized in this thesis. Hence, a model in which Cdep is involved in the JNK

or TGF-β pathway is conceivable.

5.2.1.2 The JNK pathway, like Cdep, is instrumental in GBR, DC and HI

Activation of the JNK pathway and its subsequent activation of dpp expression during

Drosophila DC culminates in the the expression of zipper (zip, Drosophila non-muscle

myosin heavy chain) and the assembly of a supracellular actomyosin cable in LE cells (Young et al., 1993; Ricos et al., 1999; Arquier et al., 2001). JNK activity is also instrumental in

the assembly of the actin-based ﬁlopodia in LE cells (Jacinto et al., 2000). Moreover, the

Dpp signal is crucial for the pulsed AS cell contractions that partly drive DC, and in the

adhesion between AS and LE cells (Fernández et al., 2007; Blanchard et al., 2010).

JNK also activates expression of the transcription factor u-shaped (ush) through Dpp

(Fernández et al., 2007). Ush is required for AS survival during GBR (Frank and Rushlow,

1996). Since the AS is mechanically aiding to accomplish GBR, the JNK pathway therefore

also plays a role in this epithelial movement.

The pro-apoptotic genes hid, rpr, grim, scylla (scyl) and charybde (chrb) are instrumental

in Drosophila HI (Abbott and Lengyel, 1991; Grether et al., 1995; Nassif et al., 1998; Chen et al., 1996; Scuderi et al., 2006). JNK is a well-known activator of apoptosis, it activates the

expression of hid, rpr and grim (Kanda and Miura, 2004; Adachi-Yamada and O’Connor,

2002; Igaki, 2009). Both scyl and chrb are transcriptionally regulated by Dpp (Scuderi et al., 2006). However, although it is highly likely that the JNK pathway directly regulates

HI, this has not yet been studied.

Thus, the JNK pathway likely mediates the morphogenetic processes in those tissues

whose morphogenesis is aﬀected by defects in Cdep. An model in which Cdep inﬂuences

the JNK pathway or vice versa is therefore possible.

5.2.1.3 Mouse Farp2 acts upstream of the JNK pathway

JNK has been shown to play a role during cell migration in many diﬀerent systems, i.e.

during neurogenesis, wound healing and cancer cell invasion in mouse models, where it

can both activate and repress cell migration, depending on the cellular and developmental

context (Mizuno et al., 2005; Westerlund et al., 2011; Zhang et al., 2014; Chen et al., 2014;

Wang et al., 2015; Cho et al., 2014). Activation of the JNK pathway via the small GTPases 62 Discussion

Rac1 and Cdc42 is capable of suppressing cell migration in human kidney epithelial cells as a reaction to a chemorepellent (Yamauchi et al., 2002). JNK activation was suppressed in those cells when they were transfected with a mouse Farp2 mutant lacking the RhoGEF domain, placing Farp2 upstream of the JNK pathway (Miyamoto et al., 2003). In axon guidance, mouse Farp1 and Farp2 also relay chemorepellent signals to cell migratory behaviour (Zhuang et al., 2009; Toyofuku et al., 2005). Sema3A, the chemorepellent signal relayed by Farp2, also acts as a guidance cue for epithelial migration in the mouse and human lung and in the mouse kidney, and in epidermal migration in the nematode

Caenorhabditis elegans (Ito et al., 2000; Becker et al., 2011; Tufro et al., 2008; Roy et al.,

2000).

A direct inﬂuence of a Cdep homologue on the JNK pathway has therefore been shown.

Furthermore, a chemorepellent whose signal is transferred to downstream eﬀectors by the same Cdep homologue plays a role in multiple epithelial migration processes. Hence, I suggest a model in which Cdep relays extracellular guidance cues to the JNK pathway in epithelial migration. However, since Cdep itself is likely membrane-associated, but not a transmembrane protein, the extracellular cues must be sensed by other receptors. Indeed,

Toyofuku et al. (2005) propose that the Cdep mouse homologue Farp2 initially binds to the transmembrane receptor plexin-A1 via the FERM domain of Farp2. Plexin-A1 and neuropilin-1 comprise the Sema3A receptor complex. Upon Sema3A binding to neuropilin-1,

Farp2 is released from plexin-A1 and transduces the chemorepellent signal (Toyofuku et al.,

2005).

5.2.2 Does Cdep act as a GEF?

Human CDEP has been claimed to be a RhoA-GEF (Koyano et al., 2001) whereas its mouse homologue Farp1 has been found to bind Rac1 (Cheadle and Biederer, 2012; Zhuang et al., 2009). Mouse Farp2 has been found to exert its GEF activity exclusively on Cdc42 in two studies (Miyamoto et al., 2003; Fukuhara et al., 2004) or exclusively on Rac1 in two other studies (Toyofuku et al., 2005; Kubo et al., 2002). In none of these studies a well-characterized GEF has been used as a control to estimate the strength of the GEF activity of the examined Cdep homologue. Furthermore, He et al. (2013) have found that two highly conserved aa residues in other RhoGEFs are mutated in human FARP1 and mouse Farp2. When aligning the sequences of either of these two proteins to mouse The role of Cdep in Drosophila embryonic morphogenesis 63

Farp1, human FARP2 and Drosophila Cdep, it becomes clear that these key aa residues are

mutated in them as well. Furthermore, Farp2 assumes a tertiary structure that autoinhibits

the RhoGEF domain (He et al., 2013). Very substantial conformational changes, mediated

by phosphorylation on multiple residues, and interactions with several binding partners

are necessary to relieve this autoinhibition (He et al., 2013). However, a Farp2 mutant

lacking the RhoGEF domain fails to activate the JNK pathway (Miyamoto et al., 2003).

Furthermore, Toyofuku et al. (2005) showed that Rac-GTP levels rose in cells expressing

Farp2 when compared to cells not expressing Farp2. It is thus possible that Drosophila

Cdep acts as a GEF in some, but not all, developmental contexts. Farp2 has also been

hypothesized to act as an indirect GEF, possibly by strengthening the interaction of other

GEFs with their substrate (He et al., 2013). Other FERM proteins have been implicated in

the regulation of small GTPases: Moesin antagonizes the activity of Rho, whereas Merlin,

the Drosophila homologue of the neuroﬁbromatosis type II tumor suppressor in humans

and mice, downregulates active Rac (Speck et al., 2003; Shaw et al., 2001). However, none

of them has protein domains that imply a direct interaction with GTPases. Cdep therefore

promises to provide a direct link between scaﬀolding function and regulation of cellular

signalling cascades, a function that FERM domain proteins have only been shown to carry

out indirectly to this date.

5.2.3 Cdep might regulate epithelial migration in parallel to Yrt and Cora

I have shown that Cdep mutants carrying a premature stop codon shortly downstream of

the transcription start site of Cdep rescue the morphological defects observed in yrt75a or cora5 loss-of-function mutant embryos. In CdepE16X and CdepG17X mutants, translation

could start at an alternative ATG. This could lead to diﬀerences in activity levels or

modiﬁed interaction behaviors of the resulting truncated Cdep protein. As discussed

above, the RhoGEF domain of the mouse homologue Farp2 requires large conformational

changes to be activated (He et al., 2013).In a truncated version of Cdep, the RhoGEF

domain could be more accessible and therefore have a higher activity than in WT Cdep.

The truncation of the protein might also aﬀect the FA domain which has been implied

in phosphorylation-dependent negative regulation of FERM-FA proteins (Manno et al.,

2005; Baines, 2006). If the FA domain is absent or misfolded in CdepE16X and CdepG17X

mutants, the remaining protein might be a constitutively active version of Cdep which 64 Discussion compensates for the missing activities of Yrt or Cora during epithelial morphogenesis in the

Drosophila embryo. While the roles of Yrt and Cora in epithelial cell polarity are relatively well-studied, it is not known how they exert their molecular functions during epithelial movements (Hoover and Bryant, 2002; Laprise et al., 2006, 2009; Gamblin et al., 2014).

FERM domain proteins usually act as scaﬀolding proteins binding membrane lipids and membrane-associated proteins to the cytoskeleton (for review see Tepass, 2009). According to a BLAST search carried out with the actin binding site of the human Cora homologue

Ezrin, neither Yrt nor Cora have an actin binding site (Turunen et al., 1994). However,

Cora has been shown to mediate the association of a glutamate receptor to the actin cytoskeleton in Drosophila neuromuscular junctions (Chen et al., 2005). Additionally, postsynaptic actin organization in neuromuscular junctions is regulated by Baz in an aPKC-dependent manner; and aPKC activity is mediated by Yrt (Ramachandran et al.,

2009; Gamblin et al., 2014). Thus, Cora and Yrt could have an indirect eﬀect on the interaction of actin with membrane-associated proteins. Since actin rearrangement is an important driver of epithelial migration, this is a possible pathway for the inﬂuence of

Cora and Yrt on embryonic morphogenesis.

If CdepE16X and CdepG17X indeed produce constitutively active proteins that comepensate for the loss of Yrt and Cora, Cdep would in fact act in parallel to them rather than in the same pathway. Cdep could inﬂuence cytoskeleton organization via the JNK pathway, although no apparent abnormalities in the actin cytoskeleton have been observed in Cdep−/− embryos. It is possible that the defects in the cytoskeleton are too subtle to observe under the given circumstances. Alternatively, the changes might not aﬀect the overall cytoskeletal organization, but rather the stability of its interaction with the plasma membrane. To elucidate this, further experiments have to be carried out. An overview over a possible mode of action for Cdep in epithelial morphogenesis in the Drosophila embryo is given in

Figure 5.2.

5.3 Conclusion and Outlook

Due to the role of the JNK pathway in a myriad of different dovelopmental processes, it is a challenge to find bona fide interactors. My findings support the hypothesis established in Miyamoto et al. (2003) that Cdep is an activator of the JNK pathway; they also fit in Conclusion and Outlook 65

Figure 5.2: Possible mode of action for Cdep in epithelial morphogenesis in the Drosophila embryo. A transmembrane receptor picks up a signal from an extracellular guidance cue. Cdep is associated, direclty or indirectly, with the cytoplasmic part of the receptor and transmits the signal to the JNK pathway, activating it. This might be dependent on the RhoGEF domain of Cdep. JNK pathway activity results in the activation of diﬀerent downstream eﬀectors, depending on the tissue and developmental time point. In tubular organs, i.e. the trachea and malpighian tubules, JNK regulates speed and direction of migration. During head involution (HI) JNK promotes apoptosis. In epithelial migration events that require cell contractions mediated by actin and myosin, the JNK signal activates the TGF-β pathway by promoting decapentaplegic dpp expression. This is necessary for the completion of germ band retraction (GBR), dorsal closure (DC) and possibly HI. The FERM domain proteins Yurt and Cora likely act in a redundant pathway to drive The same morphogenetic movements, possibly by regulating the interaction of other proteins with actin.

well with the roles found for the Cdep homologues CDEP and FARP2 in human and Farp1 and Farp2 in mouse. I and other studies before have shown that Cdep inﬂuences multiple developmental processes where epithelial movement plays an instrumental role. Hence, it is likely that Cdep is a canonical activator of the JNK pathway in epithelia. The model proposed in this thesis elucidates one of the missing links between external cues and the onset of the JNK signalling cascade.

In order to quickly confirm whether Cdep activates the JNK pathway, Cdep−/− embryos will be stained with antibodies specifically binding the phosphorylated, and therefore active, form of Bsk, the Drosophila JNK. This will also allow to estimate whether mildly affected embryos have higher p-Bsk levels than severely affected individuals.

Furthermore, to ﬁnd out where Cdep is localized in the embryo and in epithelial cells,

Cdep will be endogenously tagged with an mRFP transgene utilizing the CRISPR/Cas9 66 Discussion system. The embryos will be stained with an α-mRFP antibody. Former approaches using two diﬀerent Cdep antibodies have been unsuccessful. The antibody signal was undetectable with confocal laser scanning microscopy and the the antibodies produced many unspeciﬁfc bands in Western Blots.

It will be interesting to see whether Cdep inﬂuences the protein levels of Yrt or Cora.

Therefore, Cdep−/− embryos will be stained with antibodies against both these proteins.

Conversely, the Cdep-mRFP transgene will be combined with yrt75a and cora5, respectively.

From this experiment it will become more clear if Cdep acts in parallel to the Yrt-Cora group of proteins or in the same pathway, and how the three proteins inﬂuence each other.

Finally, the question remains whether Cdep exerts GEF activity. To assess this possibility, cell lines could be transfected with Cdep, CdepE16X , CdepG17X and Cdep∆ and the amounts of active Rho, Rac and Cdc42 can be measured. This will also allow a conclusion about whether CdepE16X and CdepG17X produce proteins with higher GEF activity than WT

Cdep. However, this experiment will still not clarify if Cdep directly interacts with small

GTPases or via other GEFs. To demonstrate this, an elaborate study will be necessary. 67

6 Materials and Methods

6.1 Cell strains, plasmids and DNA constructs

Table 6.1: E.coli strains used Escherichia coli resistance genotype reference strain DH5α –F− Φ80lacZ∆M15 ∆(lacZYA-argF)U169 Bethesda Res. − + deoR recA1 endA1 hsdR17 (rk , mk ) phoA Laboratories supE44 thi-1 gyrA96 relA1 λ−

E.coli DH5α cells were used for transformation, ampliﬁcation and -80 ◦C long-term storage of vectors in form of glycerol stocks.

Table 6.2: Plasmids and vectors used in this thesis plasmid vector resistance description reference backbone pValium22 pCFD3- AmpR Drosophila U6:3 promoter followed (Port et al., dU6:3 by 2×BbsI restriction site, T3 2014) gRNA promoter, T7 promoter, LacZ-a pValium22 pCFD-4_ AmpR Drosophila U6:3 and U6:1 promoter, (Port et al., U6:1_U6:3 T3 promoter, T7 promoter, LacZ-a 2014) tandemgRNAs pJ204 pHD- AmpR LoxP-attP-3xP3-DsRed-LoxP (Gratz et al., DsRed-attP 2014)

pCFD-3 and pCFD-4 are vectors for cloning of sgRNA templates. After injection of one of these vectors into D.melanogaster embryos, the sgRNAs will be produced by transcription from the ubiquitously active U6:3 and U6:1 pomoters, respectively (Port et al., 2014). pHD-DsRed-attP was used for homology repair after CRSIPR-Cas9-induced deletion of the whole Cdep locus. 68 Materials and Methods

Table 6.3: Vectors containing Cdep or parts thereof used for gene modiﬁcation vector description reference

pOT2-SD10794 contains cDNA of Cdep splice variant RE Dros. Genomics Res. (GenBank: BT024186.1) Center, Bloomington, IN, USA BAC attB-P[acman]- bp no. 669466-772460 on chromosome BACPAC Res. CH321-58D21 arm 3R of Drosophila melanogaster Center, Children’s cloned into the BAC attB-P[acman] Hospital Oakland, CA. USA pCFD3-dU6:3- gRNA template for recognition of this thesis CdepEx2gRNA CRISPR target site closest to START codon in Cdep exon 2 pCFD-4_U6:1_U6:3 gRNA templates for recognition of this thesis CdepExcgRNAs CRISPR target sites in exon 2 and exon 17 of Cdep pHD-Cdep5’HA-DsRed- ∼1kb DNA stretches homologous to Cdep this thesis Cdep3’HA 5’UTR (5’HA) or Cdep 3’UTR (3’HA), ﬂanking 3xP3-DsRed of pHD-DsRed-attP

6.2 Culture media

6.2.1 Lysogeny Broth (Bertani, 1951)

For growth of E.coli BL21 cells expressing a gene from a plasmid, Lysogeny Broth

(LB-medium) was used. LB medium was supplied ready-made and autoclaved by in-house facilities. Ampicillin was added to the medium by the experimenter after autoclaving to yield a ﬁnal concentration of 100 µg/ ml. To make LB-agar, 1.5 % (w/v) bacto-agar was added. LB-agar-plates containing a variety of antibiotics were supplied by in-house facilities.

Lysogeny Broth 1 % (w/v) Bacto-Tryptone 0.5 % (w/v) Yeast extract 0.5 % (w/v) NaCl pH7.0

6.2.2 Super Optimal Catabolite repression (SOC) medium (Hanahan, 1983)

For ﬁrst transformations of newly cloned plasmids into E.coli DH5α, the Super Optimal

Catabolite repression (SOC) medium was used. The autoclaved medium was supplied by Molecular biology methods 69 in-house facilities. SOC is a rich medium that increases the eﬃciency of transformation and promotes the recovery of the transformed cells.

SOC medium 2 % (w/v) Bacto-Tryptone 0.5 % (w/v) Yeast extract 8.6 mM NaCl 2.5 mM KCl

20 mM MgSO4 20 mM Glucose pH7.0

6.3 Molecular biology methods

6.3.1 Amplifying DNA by standard PCR technology

Phusion DNA polymerase was used for PCR ampliﬁcation (New England Biolabs (NEB),

Ipswich, MA, USA), together with the supplied buffer and DMSO. The protocol was taken from https://www.neb.com/protocols/1/01/01/pcr-protocol-m0530. Single dNTP solutions (dATP, dTTP, dCTP, dGTP) were purchased from Promega (MI, USA). They were mixed to yield a final concentration of 10 mM (2.5 mM of each dNTP). Primers were usually purchased in a lyophilized form from MWG Eurofins Operon (Huntsville, AL,

USA), resolubilized in ddH2O and diluted to a working concentration of 10 µM. After some initial optimization, the PCR setup and conditions shown in Tables 6.4 and 6.5 were used. PCR reactions were set up in an Eppendorf Mastercycler ep Gradient S (Eppendorf,

GER).

Table 6.4: Assay conditions for typical PCR reaction amount reagent 1.0 µl template (approx. 100 ng/µl) 1.0 µl of each forward and reverse primer (stock solution 10 µM) 0.4 µl dNTPs 4.0 µl 10×GC- or HF-buﬀer 0.6 µl Dimethylsulfoxide 0.1 µl Phusion DNA Polymerase

11.9 µl dH2O Σ 20 µl 70 Materials and Methods

Table 6.5: Typical PCR program temperature duration step 98◦C 30 sec initial denaturation 98◦C 10 sec DNA denaturation 40 repeats * 30 sec primer annealing 72◦C ** extension 72◦C 10 min ﬁnal extension 4◦C ∞ hold ∗ suitable annealing teperatures were calculated with the NEB Tm Calculator at http://tmcalculator. neb.com. ∗∗ 15-20 seconds of extension time were allowed per kb.

6.3.2 Molecular cloning

6.3.2.1 Cloning with restriction endonucleases (Old and Primrose, 1980)

For most constructs, conventional cloning assays utilizing restriction endonucleases were used. Restriction enzymes were purchased from NEB and used according to the manufacturer’s specifications with the supplied buffers. Restriction reactions were usually done for one to four hours. Calf Intestine Phosphatase (CIP; NEB) was only used for 5’-dephosphorylation of the vector when a lot of religation events occurred in a first unsuccessful cloning attempt.

In such cases, CIP was added to the restriction digest assay, which was then incubated for at least one hour. For ligation of inserts and vector, the T4 ligase from NEB was used according to the manufacturer’s speciﬁcations.

6.3.2.2 Gibson Assembly (Gibson et al., 2009)

With this cloning technique, multiple dsDNA fragments can be assembled in one reaction step. Primers for each fragment are designed in such a way that they partly overlap with the adjacent fragment. The reaction mix contains all fragments that are to be assembled, a suitable buﬀer and the following enzymes:

• a 5’ exonuclease to create 3’ overhangs for each fragment

• a DNA polymerase to ﬁll gaps after fragment assembly

• a DNA ligase to repair nicks (see also Figure6.1)

The Gibson Assembly Master Mix containing all enzymes in reaction buﬀer was obtained from NEB and used according to the manufacturer’s speciﬁcations. Molecular biology methods 71

Figure 6.1: Gibson Assembly, schematic representation, modiﬁed from (Gibson et al., 2009). The vector is linearized either via restriction digest or PCR. The inserts are ampliﬁed with primers that create a 15 bp overlap with the adjacent fragment. The linerized vector and all inserts are then assembled in one reaction step.

6.3.3 Detecting DNA via agarose gel electrophoresis

To ensure correct DNA fragment lengths during cloning or for genomic analysis, DNA

samples were separated on a 1.5 % (w/v) agarose gel.

For the gel UltraPure Agarose powder (Invitrogen, Carlsbad, USA) was solubilized in

1×TBE buﬀer Maniatis (1). After a short cooling period, ethidiumbromide was added

to yield a ﬁnal concentration of 0.05 (v/v) The gel was run in 1×TBE at 80 V for approximately 35 minutes. As a standard, 10 µl of the GeneRuler 1kb DNA Ladder (Thermo Scientiﬁc, Waltham, USA) was run in parallel to the DNA samples. DNA bands

were detected under UV light.

50× TBE 40 mM Tris-acetate 1 mM EDTA pH 7.5 - 8.0

6.3.4 Purifying DNA from E.coli

To purify plasmids from E.coli cell cultures, the QIAprep Spin Miniprep Kit or the QIAGEN

Plasmid Midi Kit was used. Puriﬁcation was carried out as given in the QIAGEN R Plasmid

Puriﬁcation Handbook from April 2012. During a midiprep, lysate was cleared using a

QIAﬁlter Cartridge (all materials mentioned were supplied by Qiagen, Hilden, GER).

6.3.5 Extracting DNA from Drosophila adults

DNA from adult ﬂies was extracted as described in Winkler et al. (2005). Usually, ﬁve

ﬂies were used and the DNA pellet was resuspended in 50 µl of dH2O. Theis DNA was diluted 1:10 before using it as a template for PCR. 72 Materials and Methods

6.3.5.1 Extracting mRNA from Drosophila embryos and reverse transcription into

cDNA

Embryos were collected and incubated to reach the desired developmental stage. They dechorionated and suspended in dH2O. They were lysed with QIAzol lysis reagent (QIAGEN) following two freeze-thaw cycles to facilitate lysis. For mRNA extraction, the RNeasy MiniKit (QIAGEN) was used according to the manufacturer’s speciﬁcations.

For reverse transcription, SuperScript III reverse transcriptase (Invitrogen) and the reaction was assembled following the manufacturer’s instructions. 1.5 µg of mRNA were used as template. If mRNA concentrations were not high enough, the maximum possible amount was used. The resulting cDNA was diluted 1:10 or 1:100 depending on the initial mRNA concentration, and 1 µl was used as a PCR template.

6.3.6 Making mRNA probes for in situ hybridization

Target sites for mRNA probes of ca. 500 bp length were chosen from the coding sequence of the respective gene. Primers were designed introducing a promoter region into the

PCR product recognized by either SP6 or T7 RNA polymerase. The PCR products were used as templates for transcription into RNA probes with digoxygenin (DIG) labeled

Uracil. For this, the DIG RNA Labeling Kit (SP6/T7) (Roche) was used according to the manufacturer’s speciﬁcations. The quality and correct length of the RNA probe was conﬁrmed via agarose gel electrophoresis.

6.3.7 Measuring DNA concentrations using the NanoDrop

DNA concentration was measured with the NanoDrop 1000 spectrophotometer (Thermo

Scientiﬁc) using 1 µl of sample diluted in water or Buﬀer EB (Qiagen).

6.3.8 DNA sequencing

Approximately 50 ng per kb of each vector was given to the MPI-CBG Sequencing Facility.

Primers for common vectors were supplied by the facility. The sequenced data was analyzed with the Geneious software version 6.1.8 (Biomatters Ltd., Auckland, NZ). The sequences acquired by the facility were aligned with the desired DNA sequence of the appropriate DNA construct (sequencing alignments for relevant constructs are given in the Supplemental Molecular biology methods 73

Material on the USB ﬂash drive attached to this thesis). Only if a construct proved to

have the correct DNA sequence it was used for further experiments.

6.3.9 Transformation

Ligation assays were transformed into chemocompetent E.coli DH5α cells. A 50 µl cell

aliquot and 2 µl of the ligation assay were used. The cells were incubated on ice for 30

minutes. Transformation was triggered by a 42 ◦C heat-shock for 60 to 90 seconds and cells

were left on ice for another two minutes before adding 500 µl SOC-medium prewarmed

to room temperature. The assay was shaken at 37◦C for at least 60 minutes. Cells were

plated on LB-Agar including a selective antibiotic dependent on the expression vector.

Vectors were transformed in the same way, using 50 µl of E.coli DH5α cells and 0.5 µl of

vector.

All plates were incubated at 37◦C overnight (16 to 18 hours).

6.3.10 Mutagenesis of Cdep with the CRISPR-Cas9 system

6.3.10.1 CRISPR/Cas9 tools for mutagenesis in D.melanogaster

A ﬂy line carrying a Cas9 transgene under the control of the germline-speciﬁc nanos

promoter (nos::cas9 ) has been established (Port et al., 2014). This transgene has been

inserted on the X-chromosome and is marked with a miniwhite gene and 3xP3-mRFP

(see below). Several plasmids placing expression of the sgRNAs under the control of the

ubiquitously active U6:3 or U6:1 promoters are available (Figure 6.2a and 6.2b) (Port et al., 2014). Furthermore, a donor plasmid for homolgy repair called pHD-DsRed-attP has

been synthesized and made available by Gratz et al. (2014). It contains a 3xP3::DsRed

marker which drives expression of a red ﬂuorescent protein in the eyes of adult ﬂies and

the ocelli of larvae and adults. 3xP3::DsRed is ﬂanked by multiple cloning sites to allow

cloning of homology arms (Figure 6.2c and 6.2d). The same group also established an

online tool, the CRISPR Optimal Target Finder, to identify potential CRISPR target

sites in the Drosophila genome and predict oﬀ-target sites. This tool can be found at

http://tools.flycrispr.molbio.wisc.edu/targetFinder/ (Gratz et al., 2014). 74 Materials and Methods

a: pCFD3-dU6:3gRNA from Port et al. b: pCFD4-dU6:1_U6:3tandemgRNA from (2014), image generated with Geneious. Port et al. (2014), image generated with Geneious.

c: pHD-DsRed-attP from Gratz et al. (2014), image d: Left: w− and 3xP3::DsRed flies under generated with Geneious. visible light. Right: w− and 3xP3::DsRed flies under UV light with RFP filter.

Figure 6.2: pCFD plasmids for delivery of sgRNAs and pHD-DsRed-attP for homologous recombination in Drosophila melanogaster.

6.3.11 Strategies for CRISPR/Cas9 mutagensis

In order to establish loss-of-function mutants of Cdep, two diﬀerent approaches were chosen with the CRISPR/Cas9 system: A full deletion of the Cdep ORF was created by replacing the Cdep locus with 3xP3::DsRed, following a protocol decribed in Gratz et al. (2014).

Also, an in-frame stop codon was inserted shortly after the START codon of Cdep as described in Port et al. (2014). The constructs, strategies and workﬂows used in each approach are given below. The sgRNAs were encoded on plasmids under the control of Molecular biology methods 75 the ubiquitously active U6:3 or U6:1 promoters. The in-frame stop codon was introduced using a single-stranded Oligodeoxynucleotide (ssODN) carrying an 11 long insert between

60 nt homology arms. The entire Cdep locus was replaced by 3xP3::DsRed supplied on pHD-DsRed-attP (Gratz et al., 2014). Homology arms of roughly 1 kb length were cloned into pHD-DsRed-attP. The CRISPR components - the plasmid encoding the sgRNA and the ssODN or the construct for homology repair - were injected into the posterior pole of preblastoderm nos::cas9 embryos by Sven Ssykor. He also collected the surviving larvae. Usually, 500 ng/ µl of each construct was injected into the embryos (1 µg/ µl

ﬁnal concentration with both components in the same solution). If this solution proved too viscous for injection, it was diluted to 400 ng/ µl per component (800 µg/ µl ﬁnal concentration).

6.3.11.1 vectors and oligomers used for mutagenesis of Cdep via CRISPR-Cas9

system pCFD3-dU6:3-CdepEx2gRNA The vector backbone pCFD3-dU6:3gRNA was obtained from Addgene (#49410; Port et al. (2014)). pCFD3-dU6:3-CdepEx2gRNA encodes sgRNA for the Cas9 target site GtgtctaccaccttctccgcCGG (UPPERCASE: G for eﬃcient transcription from U6 promoter, UPPERCASE italics: PAM ) found on the minus strand in exon 2 of Cdep via the CRISPR Optimal Target Finder). The DNA oligomers CAM005 and CAM002 (Table 6.6) were ordered as 5’-phosphorylated ssDNAs from MWG Euroﬁns

Operon. Both oligomers were resuspended in dH2O to yield a concentration of 100 µM. An annealing reaction was assembled in a 1,5 ml Eppendorf tube as given on http://flycrispr.molbio.wisc.edu/protocols/crRNA:

1 µl of each oligomer

1 µl 10x T4 ligase buﬀer (NEB)

7 µl H2O

About 500 ml of water was heated to boiling point and allowed to cool for several minutes.

The tube containing the annealing reaction was then submerged in the hot water and the water was allowed to cool down to room temperature. The resulting dsDNA with sticky overhangs was subsequently ligated into the pCFD3 dU6:3gRNA backbone that had been digested with BbsI and puriﬁed with the QIAGEN PCR cleanup kit. NEB T4 ligase was 76 Materials and Methods used with a standard ligation protocol. E.coli DH5α was transformed with 2 µl of the resulting vector pCFD3-dU6:3-CdepEx2gRNA and plated on LB-Agar with Amp. On the next day, colonies were picked and grown in 5 ml LB-Amp. A miniprep was carried out to extract the plasmid, which was then given for sequencing with the M13 reverse primer supplied by the MPI-CBG Sequencing Facility.

CdepSTOP-ssODN In order to insert an in-frame stop codon shortly after the Start codon of Cdep and thereby create a loss-of-function mutant, a single-stranded DNA oligomer was ordered from Integrated DNA Technology (Coralville, IA, USA). It contained two

60 nt homology arms ﬂanking the cut made by Cas9 guided by the sgRNA encoded in pCFD3-dU6:3-CdepEx2gRNA. The stop codon TAA was encoded by the 11 nt insert, as well as a BglII restriction site for convenient screening. The full sequence of the ssODN is

5’-AGATTCTCGTCTACAAAATGTCCCTGGCCGACATGGGCACAGCCTCTCGAT

CCGCCGGCGAGTAAagatctGAGAAGGTGGTAGACACTATGATCTAGCCACGG

GCGGAGCTGGAAGTGGGGGTCATCCAG-3’ (UPPERCASE italics: nucleotides to keep stop codon in frame with Cdep ORF; UPPERCASE bold: stop codon; lowercase:

BglII restriction site). An overview over the strategy is given in Figure 6.3.

Figure 6.3: Strategy to insert in-frame stop codon into Cdep ORF. The introns (colored boxes) and exons of Cdep isoforms RE and RF are shown. Underneath is a blow-up of the region around the Start codon (green), the PAM (magenta) and the sgRNA target site (orange). The insert was inserted after the ﬁrst two base pairs of the sgRNA target site. Successful insertion resulted in an in-frame stop codon and a novel BglII restriction site. Screening for insertion events was carried out by amplifying the region around the insert with the primer pair AM241 (fwd) and AM242 (rev), resulting in a 725 bp amplicon. Molecular biology methods 77

pCFD4-CdepExc_2sgRNAs This vector encodes two sgRNAs for the deletion of the

whole Cdep ORF, one guiding Cas9 to the target site tacacagccatgctgtgtTGG (UPPERCASE italics: PAM ; found in exon 2 of Cdep) and the other to guide Cas9 to the target site

gtgtctaccaccttctccgcCGG (UPPERCASE italics: PAM ; found in exon 17 of Cdep). The

DNA oligomers CAM009 and CAM010 (Table 6.6) were used as primers for PCR, using 10

ng of pCFD4-U6:1_U6:3tandemgRNAs as template (Addgene #49411; Port et al. (2014)).

A schematic of the resulting amplicon is depicted in Figure 6.4. This DNA was cloned into

pCFD4-U6:1_U6:3tandemgRNAs via Gibson Assembly after linearization of the vector

with BbsI endonuclease. Treating the vector with BbsI resulted in the removal of the ﬁrst

sgRNA core sequence and the U6:3 promoter, both of which were replaced by the insert

(Figures 6.4a and 6.4b).

a: Part of pCFD4-U6:1_U6:3tandemgRNAs involved in encoding sgRNA cores and U6 promoters. The sequence between the two BbsI restrcition sites will be eliminated during the restriction digest.

b: PCR product resulting from amplifying pCFD4-U6:1_U6:3tandemgRNAs with CAM009 and CAM010.

c: Gibson Assembly of one insert encoding sgRNAs targeting Cdep exon2 and Cdep exon 17 into pCFD4-U6:1_U6:3tandemgRNAs, yielding the vector pCFD4-CdepExc_2sgRNAs.

Figure 6.4: Cloning strategy for pCFD4-CdepExc_2sgRNAs 78 Materials and Methods pDsRed-5’HA-3’HA For the replacement of the Cdep ORF with dsRed, the vector pHD-DsRed-attP was obtained from Addgene (#51019; Gratz et al. (2014)). An overview over the strategy used for the mutagenesis approach is shown in Figure 6.5. The 5’-homology arm (5’HA), a DNA sequence homologous to 1,021 bp upstream of the sgRNA target site in exon 2 of Cdep was cloned into the vector. This was done using the EcoRI and

NotI restriction site of the multiple cloning site provided on the vector. The primer pair

AM257+AM266 was used with 10 ng of the Cdep BAC attB-P[acman]-CH321-58D21 as template.

The 3’-homology arm (3’HA) covered 989 bp in the 3’UTR of Cdep, starting 3 bp downstream of the endogenous stop codon. It was inserted into the vector via Gibson assembly since all other attempts using restriction endonucleases failed. The 3’HA was ampliﬁed from attB-P[acman]-CH321-58D21 using the Gibson Assembly primer pair

AM273+AM274. The vector already containing the 5’HA was linearized using the restriction endonuclease XhoI.

An overview over the relevant part of the vector is shown in Figure6.6 and primers mentioned are listed in Table 6.6.

Figure 6.5: Schematic representation of strategy to replace Cdep ORF with DsRed. Black triangles mark the cut sites for Cas9. pDsRed-5’HA-3’HA carries 1 kb homology arms allowing it to act as a template for homology repair after the Cdep ORF has been deleted. In this way, the DsRed gene will replace Cdep, allowing for convenient mutant screening. Molecular biology methods 79

Figure 6.6: Schematic representation of pDsRed-5’HA-3’HA. Image generated with Geneious. The 5’-homology arm (HA) binds to a region upstream of the sgRNA target site in exon 2 of Cdep. The 3’HA binds to the 3’UTR of Cdep, 3 bp downstream of the endogenous stop codon. Not shown is the Amicillin ressistance gene and a pBR322 origin of replication.

6.3.11.2 Screening for successful mutagenesis events in ﬂies modiﬁed with the

CRISPR/Cas9 system

Insertion of an in-frame stop codon into the Cdep ORF The screen for successful

insertion events of CdepSTOP-ssODN into the Cdep locus was carried out by PCR and

subsequent restriction digest with BglII. For this, ﬁve ﬂies presumably carrying the insert

were sacriﬁced and their genomic DNA was extracted. It was diluted 1:10 and used as

a template with the primer pair AM241+AM242 (Figure 6.3 and Table 6.6). The PCR

product was cleaned with the QIAquick PCR puriﬁcation kit (Qiagen) and digested with

BglII (NEB) for about two hours. The entire volume of DNA was loaded onto an agrose gel.

As a negative control, DNA from WT ﬂies was also ampliﬁed and digested. Furthermore,

when it became apparent that there were ﬂies carrying the insertion and that they were

homozygous viable, the PCR and BglII digest were repeated to show that there was no

WT allele of Cdep present (Figure 6.8). The PCR products were given for sequencing.

Figure 6.7 shows the crosses that were carried out to establish ﬂy lines homozygous for the

Cdep STOP alleles.

Figure 6.7: Workﬂow for crossing and screening ﬂies for succesful insertion of stop codon in Cdep ORF. From the F1 and F2, single males were crossed with two or three virgins carrying a third-chromosome balancer. 80 Materials and Methods

Figure 6.8: Example agarose gels with BglII digestion products of amplicons generated with primers AM241+AM242. a: Agarose gel showing BglII digestion products of DNA amplified from WT flies or flies carrying Cdep STOP over a balancer chromosome. Triangle: undigested PCR product. Arrows: digestion products. A part of the image between the DNA marker and the bands was cut out for better visibility. b: Agarose gel showing BglII digestion products of DNA amplified from WT flies or flies homozygous for Cdep STOP . There is no undigested PCR product left in the right lane, showing that there is no WT allele of Cdep present in these flies.

Replacing the entire Cdep locus with dsRed The dsRed marker is clearly visible as bright red fluorescence in the adult eyes and ocelli. Nonetheless, a problem had to be overcome when screening offspring for the presence of 3xP3::DsRed: the nos::cas9 transgene on the first chromosome is marked with 3xP3::mRFP which gives the same red fluorescent eye phenotype as 3xP3::DsRed. nos::cas9 is also marked with the miniwhite gene which produces an orange eye color. Therefore, only male F1 offspring was crossed further. From the F2, males with white eyes (absence of miniwhite, therefore absence of nos::cas9 ) that were red fluorescent (presence of 3xP3::DsRed) were taken and crossed again. The flies carrying this allele proved to be homozygous viable, therefore a homozygous stock could be established. Five homozygous flies were sacrificed and their DNA was extracted and used as template for PCR with the primer pair AM280+AM281 (Table 6.6). This primer pair flanks the coding region of Cdep and produces a 4,276 bp long product only after the replacement of the Cdep locus with 3xP3::DsRed. If the Cdep ORF was still present, the amplicon would be about 30 kb long. However, amplification will most likely break off before finishing this long amplicon, therefore in WT, no PCR product is expected. The

PCR product was given for sequencing. Molecular biology methods 81

Figure 6.9: Workflow for crossing and screening flies for succesful replacement of Cdep ORF with dsRed. From the F2, males with white eyes that showed red fluorescence under UV light were crossed with two or three virgins carrying a third-chromosome balancer. 82 Materials and Methods

Table 6.6: primers used in CRISPR mutagenesis of Cdep Gibson Assembly name sequence overlaps with

AM273 (fwd) gcataaggcgcgcctaggGCAACTAAAAGAA pDsRed-5’HA 3’-end after CTCAGTGTTTC XhoI digest AM274 (rev) ctcgattgacggaagagccCAATAAAAGATTT pDsRed-5’HA 5’-end after AAGAACTTTGTAAAATC XhoI digest

sgRNA target sites name sequence restr. site/ overlaps with

CAM002 AAACGCGGAGAAGGTGGTAGACAC BbsI CAM005 GTCGGTGTCTACCACCTTCTCCGC BbsI CAM009 tatataggaaagatatccgggtgaacttcgGTGT U6:1 promoter, sgRNA CTACCACCTTCTCCGCgttttagagct core agaaatagcaag CAM010 attttaacttgctatttctagctctaaaacAACAC sgRNA core, U6:3 AGCATGGCTGTGTACcgacgttaaat promoter tgaaaataggtc

classic cloning name sequence restriction site AM257 (fwd) ctactGAATTCCCAGATAGAGAAACT EcoRI GAACTTCC AM266 (rev) gatcaGCGGCCGCGGATCGAGAGGC NotI TGTGC

Mutant screening name sequence AM241 (fwd) GAGGAAATCCAGTGCCACACATGCGTG AM242 (rev) GTACTTGGAACATGGTGATCGAGTCGTC AM280 (fwd) GTATCTAGCAAATTGACAGCCAG AM281 (rev) GTATTTCGCATTTATCGTGAGAC

lowercase italics: overlap with neighboring fragment for Gibson Assembly UPPERCASE: sequence annealing to template blue: overlap with U6 promoter region lowercase bold: guanine for eﬃcient transcription from U6 promoter UPPERCASE BOLD: sequence encoding sgRNA underlined: overlap with sgRNA core lowercase: overhang for binding of restriction enzyme UPPERCASE ITALICS: restriction site Molecular biology methods 83

6.3.12 Extracting protein from Drosophila embryos

For protein extraction, embryos of the desired genotype and developmental stage were

collected from apple juice agar plates and suspended in PBS. They wewre ecntrifuged and

the PBS was discarded. Subsequently, 1 µl of embryo lysis bufer for protein extraction was

added and the embryos were ground with a pistill and BioVortexer (Biospec, Bartlesville,

USA). SDS-PAGE loading dye was added to yield a concentration of 1×. Samples were

then used for SDS-PAGE (Section 6.3.13).

Embryo lysis buﬀer for protein extraction 50mM Tris pH8.0 150mM NaCl 0.5 % Triton X-100

1mM MgCl2 1× freshly added cOmplete inhibitor tablet (Roche) 1mM freshly added PMSF

6.3.13 Detecting and separating proteins by mass via SDS-PAGE

The sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) (Shapiro et al., 1967) is a method to separate proteins by their molecular weight on a gel matrix.

Principally the separation is based on the charge of the proteins, but a more or less identical

charge per mass unit of each protein is ensured by adding SDS to the protein sample

and incubating it for ten minutes at 65 ◦C. SDS is a detergent that denatures proteins

and charges them negatively, so they will move towards the anode in an electric ﬁeld.

The incubation at a high temperature supports the protein denaturation and makes the

polypeptide chain more accesible to the SDS. 4×SDS-PAGE loading dye (Laemmli dye

(Laemmli, 1970)) consists of the components given below.

SDS-PA minigels (6 ×8 cm) were cast several hours before use. The upper 2 cm

also containing the loading wells had a lower acrylamide concentration and allowed the

polypeptides to pass through faster, collecting or "stacking" them at the interface with the

separation gel. 84 Materials and Methods

SDS-PA gels stacking gel separating gel Tris HCl pH8.8 – 375 mM Tris HCl pH6.8 125 mM – SDS 1.25 0.625 Acrylamide 3.75h % 10h % APS 1.67 0.833 Temed 1.67 h 0.833 h dH2O ﬁll up to desiredh volume h

After preparation the samples were loaded on a SDS-PA gel containing a concentration of acrylamide suitable for the size of the protein that was to be detected. The cathode and anode chambers were ﬁlled with SDS-PAGE running buﬀer. Polypeptides were stacked at

80 V for 20 minutes and separated for another 90 minutes at 120 V. As a standard, 5 µl of the SeeBlue Plus2 Pre-Stained Standard marker (Invitrogen) was used. Polypeptides were then transferred onto a nitrocellulose membrane via Western blotting.

SDS-PAGE loading dye SDS-PAGE running buﬀer (Laemmli dye) 25 mM Tris 20 % (w/v) SDS 192mM glycine 1 M Tris-HCl pH 6.8 0.1% SDS 20 mM 2-mercaptoethanol 40 % (w/v) glycerol 0.2 % (w/v) bromphenolblue

6.3.14 Detection of speciﬁc polypeptides by Western blot analysis (Towbin et al., 1979)

6.3.14.1 Transferring polypeptides to nitrocellulose membrane

After the polypeptides had been separated by SDS-PAGE, they were transferred to a

0.45 µm Hybond ECL Nitrocellulose Membrane (GE Healthcare, Little Chalfont, UK) by

Western blotting. For this, two sponges, two gel-sized sheets of Whatman paper (GE

Healthcare) and one equally sized sheet of nitrocellulose mebrane were soaked in Western blot transfer buffer. One sponge and one sheet of Whatman paper were then placed on the cathode side of the blot cassette (Biorad, Hercules, CA, USA). The gel and subsequently the nitrocellulose membrane were added and covered first with another sheet of Whatman paper and the remaining sponge. The blot cassette was inserted into the Mini Trans-Blot gel system (Biorad). And covered with transfer buffer. Blotting occured at 100 V for two hours. Molecular biology methods 85

Western blot transfer buﬀer 25 mM Tris 192 mM glycine 20 % methanol

6.3.14.2 Detecting speciﬁc polypeptides via immonublotting

After transferring polypeptides from a SDS-acrylamide gel to a nitrocellulose membrane, the membrane was blocked for one hour in 5 % BSA solubilized in TBST (1×TBS containing

0.2 % Tween 20). The primary antibody was diluted to a suitable concentration in TBST containing 1 % BSA and added to the membrane. The membrane was incubated with the primary antibody overnight at 4 ◦C. Subsequently, the membrane was washed three times for ten minutes in TBST. A peroxidase-coupled secondary antibody, directed against the species the primary antibody was raised in, was diluted to a ﬁnal concentration of 1:5,000 in TBST and added to the membrane. The membrane was incubated with the secondary antibody for one hour at room temperature. Afterwards, the membrane was washed two times for ten minutes in TBST and two times for ten minutes in TBS. The two components of Amersham ECL chemiluminescent detection solution (GE Healthcare) were mixed 1:1.

After removing excess liquid from the membrane, it was placed on a ﬂat, even surface and covered in detection solution for ﬁve minutes. The detection solution was then drained and the membrane was placed in a light-proof cassette. In a dark room, three sheets of

Amersham Hyperﬁlm ECL (GE Healthcare) were simultaneously exposed for ten minutes to the signal emitted by the membrane. The ﬁlms were developed using a Kodak X-Omat

2000A (Eastman Kodak Company, Rochester, NY, USA).

1×TBS 100 mM Tris base 100 mM boric acid 2 mM EDTA 86 Materials and Methods

6.4 Online tools for prediction of protein domains, interactions,

sequence alignments, etc.

For predicting transmembrane domains or signal peptides, the HMMTOP protein topology prediction tool (Tusnády and Simon, 1998, http://www.enzim.hu/hmmtop/index.php) was used. Protein interactions were looked up on the STRING database (Snel et al.,

2000, http://string-db.org). Similar sequences of proteins or DNA were searched with

BLAST on (Altschul et al., 1997, 2005, http://blast.ncbi.nlm.nih.gov/Blast.cgi) using the default algorithm. The SMART tool was used to predict protein known protein domains (Schultz et al., 1998, http://smart.embl-heidelberg.de/). Potential promoter regions were predicted with the Markov Chain Promoter Finder McPromoter006 (Ohler,

2006, http://tools.igsp.duke.edu/generegulation/McPromoter/).

6.5 Cell culture growth and harvest

6.5.1 Growing E.coli cells for vector ampliﬁcation

In order to highly amplify a vector from a single E.coli clone, small-volume cultures were prepared from single colonies and used for inoculation of the growth medium. For such a culture, 5 ml of LB medium with the appropriate antibiotic for selective growth were prepared. One colony from the respective transformant plate was picked to inoculate the medium. It was left shaking at 37 ◦C for approximately 4 to 6 hours. A culture ﬂask containing 50 ml of LB with the appropriate antibiotic was inoculated with 50 - 100 µl of the culture and left shaking overnight at 37 ◦C.

If only small amounts of vector were needed, 5 ml of LB medium conatining the appropriate antibiotic were inoculated with one single E.coli colony from an agar plate.

The culture was shaken overnight at 37 ◦C.

6.5.2 Cell harvest

The E.coli cells containing a certain vector were harvested by decanting the medium into a

50 ml Falcon tube and centrifuging it at for 30 minutes at 3,000 g. The resulting cell pellet was resuspended in buﬀer P1 (Qiagen). Work with Drosophila melanogaster 87

6.6 Work with Drosophila melanogaster

All techniques regarding the establishment and maintenance of fly stocks were carried out as described in (Ashburner, 1989). Techniques included stock keeping, crossing of genetically distinct fly lines, genetic recombination, etc. Fly stocks were usually kept in plastic tubes at room temperature. Stocks that needed to be amplified quickly were kept at 25 ◦C, as were embryo collections and crosses. Stocks not frequently used were kept at

18 ◦C. All fly stocks that are mentioned in this thesis are listed in Table 6.6.1. Fly embryos were staged according to Campos-Ortega and Hartenstein (2013). 88 Materials and Methods et al. , et al. , et al. et al. et al. et al. reference (Ryder Ryder (Ryder Ryder Ryder Parks 2007) 2007) (2007) (2007) (2007) (2004) Continued on next page D.melanogaster D.melanogaster D.melanogaster D.melanogaster D.melanogaster D.melanogaster in description uncovering 95 genes uncovering 34 genes, completely overlapping with Df(3L)ED4287 uncovering 83 genes uncovering 17 genes, 9deleted of in which Df(3L)ED4502 are also uncovering 30 genes includingcomplete the Cdep locus uncovering 12 genes including aof small the part Cdep locus 3’ of its coding region location 82E3-E7 Chromosomal deletion in genomic duplications and balancers ;; Df(3L)ED4287/ TTG 62B5-E5 Chromosomal deletion ;; Df(3L)ED4284/ TTG 62B4-B12 Chromosomal;; deletion Df(3L)ED4502/ in TTG 70A3-C10;; Df(3L)ED4515/ TTG Chromosomal deletion in 70C6-C15 Chromosomal;; deletion Df(3R)ED5066/ in TTG 82C5-E5;; Chromosomal Df(3R)Exel6143/ deletion in 1118 1118 1118 1118 1118 1118 genotype TTG Fly stocks used: deficiencies, alleles, duplications and balancers chromosome rd name Deficiencies 3 Df(3L)ED4287 w Df(3L)ED4284 w Df(3L)ED4502 w Df(3L)ED4515 w Df(3R)ED5066 w Df(3R)Exel6143 w Table 6.7: 6.6.1 Fly stocks used: deficiencies, alleles, Work with Drosophila melanogaster 89 (2014) et al. , et al. et al. reference this thesis Port Huang this thesis this thesis this thesis (Lamb this thesis 1998) (2009) Continued on next page X alleles in 16 E ∆ nanos combined with combined with combined with Cdep Cdep shg , DE-cadherin shg shg shg and and , no detectable 5 5 cora cora coracle r the control of the allele in 3rd chromosome allele in 3rd chromosome k-in allele of X allele in 3rd chromosome 16 E G 17 X ∆ alleles in one fly line description promoter for germ-line specific expression labeled with GFP in all cells GFP knock-in allele of GFP knock-in allele of GFP knock-in allele of combination of one fly line protein expression Cas9 unde Cdep Cdep Cdep 57B15-16, 82E2 56C4; 82E2 combination of 56C4; 82E2 location 57B15-16 GFP knoc 57B15-16, 82E2 57B15-16, 82E2 56C4 Null allele of genomic / / ∗ ∆ 5 5 / 5 cora cora Cdep ; ; ; cora ) GF P GF P GF P GF P + X 16 ∆ E previous page X X Cdep Cdep 16 ; TI{TI}shg ; TI{TI}shg ; TI{TI}shg ; TI{TI}shg ; P{neoFRT}43D ; P{neoFRT}43D E G 17 ∗ ∗ ∗ ∗ w P(nos-cas9, w w w w ; P{neoFRT}43D 1118 1118 ∗ 1 1 1 1 1 CTG; y w w CTG; CTG; Dr/ TTG y y w genotype M(3xP3-RFP.attP)ZH-2A w Cdep Cdep 16 X Continued from E ∆ X Cdep Cdep 16 ; ; E G 17 X ∆ 5 5 5 chromosome chromosome st nd DE-cad::GFP; cora cora Table 6.7 – name Alleles 1 nos::Cas92 y DE-cad::GFP y DE-cad::GFP; DE-cad::GFP; cora Cdep Cdep Cdep 90 Materials and Methods et al. et al. et al. et al. reference Laprise Metaxakis Metaxakis Venken this thesis this thesis this thesis (2006) (2005) (2005) (2011) Continued on next page et al. t insertion in the intron including the first ATG,and exons part 11 of and exon 10 9 description between exons 6 and 7 of CG17364 coding region Minos element insertion shortlyof upstream exon 17 of Cdep exons 15 and 16 of Cdep Bellen nucleotide triplet 45 bptranscription upstream start of site Cdep replacedcodon by TAA stop nucleotide triplet 48 bptranscription upstream start of site Cdep replacedcodon by TAA stop and replaced by dsRedthree under copies control of of the P3 promoter (2004) 82E2 location 70C970C9 Minos elemen 82E3 Minos element insertion 5’ of CG17364 82E2 P-element insertion in intron between 82E2 82E2 Complete coding region of Cdep deleted genomic / TTG 87E11 Deletion of 4.9 kb of the yurt locus } a + 75 / TTG / TTG , loxP yrt X 16 E / TTG 01315 04765 3:: dsRed / TTG MB MB 00496 previous page Cdep CdepG 17 X 122 ;; ;; Mi{ET1} ;; Mi{ET1} ;; Mi{MIC} ;; PBac{5HPw ;; ;; attP, loxP, MI B ∆ , 3 xP 1118 1118 1118 1118 1118 1118 1118 w w CG17364 w CG17362 w Cdep w Cdep w w Cdep P{neoFRT}82B genotype 01315 04765 Continued from MB MB X 00496 122 16 X E MI B G 17 ∆ chromosome 75 a rd Cdep Table 6.7 – name 3 CG17364 CG17362 Cdep Cdep Cdep Cdep yrt Work with Drosophila melanogaster 91 reference this thesis this thesis this thesis Chen and Ganetzky (2012) this thesis this thesis Continued on next page a a , alleles 75 75 a yrt yrt 75 yrt and and X and 16 X ∆ E G 17 D.melanogaster Cdep Cdep Cdep in one fly line in one fly line . Duplicatied region X 16 G 17 X E Cdep one fly line Cdep Cdep description recombination of Duplication of 188 genesof on chromosome the 3 right of arm alleles in including inserted into first chromosome.covered Genes by the duplicationchromosome are 3 deleted by on Df(3R)2-2. recombination of alleles in one fly line recombination of in one fly line with with 81F4 - 83A Combination of duplication Dp(3;1)2-2 82E2, 87E11 81F4 - 83A Combination of duplication Dp(3;1)2-2 location 81F4 - 83A 82E2, 87E11 82E2, 87E11 genomic X X 16 , G 17 E ∆ / TTG / TTG / TTG a a a Cdep Cdep 75 75 75 Cdep ;; ;; Df(3R)2-2 ;; , , yrt yrt yrt X X 1118 1118 1118 16 E G 17 1 previous page Cdep Cdep ;; ;; attP, loxP, ;; 1118 1118 1118 Dp(3;1)2-2, w w P{neoFRT}82B Dp(3;1)2-2, w w P{neoFRT}82B w P{neoFRT}82B genotype 3xP3::dsRed, loxP, /TM3, Sb a a 75 75 a Continued from yrt yrt 75 , , X X X yrt 16 , 16 E G 17 ∆ G 17 X E chromosome rd Dp(3;1)2-2;; Table 6.7 – name Cdep Duplications 3 Dp(3;1)2-2 Dp(3;1)2-2, w Cdep Dp(3;1)2-2;; Cdep Cdep Cdep 92 Materials and Methods reference Alan Michelson Alan Michelson Paul Schedl, Liz Gavis and Tubby Stubble promoter promoter. Balancer chromosome and twist twist description CyO balancer expressing EGFPthe driven by TM3 balancer expressing EGFPthe driven by is marked with loss-of-function alleles of loss-of-function alleles of Serrate Stubble chr. 2 location chr. 3 TM6B balancer, marked with chr. 3 genomic 1 / 1 Sb , Tb Gla − 1 1 / TM3, / TM6B, Sb Mio =GAL4-twi.G}2.2, =GAL4-twi.G}2.3, previous page Mio mC ; In(2LR)Gla, wg ;; Dr + + mC 1 ; Dr 1118 1118 ∗ P{UAS-2xEGFP}AH2.2 CyO, P{w P{w P{UAS-2xEGFP}AH2.3, Ser genotype w w Continued from chromosome chromosome nd rd Table 6.7 – name Balancer chromosomes 2 CTG 3 TTG TM6B w Work with Drosophila melanogaster 93

6.6.2 Recombining alleles located on the same chromosome

When wanting to assess phenotypes in embryos doubly mutant for two diﬀerent alleles

located on the same chromosome, these alleles have to be recombined. Recombination is an

event naturally occurring in the early meiotic prophase of Drosophila oocytes. The term

"recombination" describes the exchange of alleles between non-sister chromatids (Merriam,

1967). In order to achieve this, the alleles that are to be recombined need to be brought

into the same female ﬂies by crossing the lines carrying each allele. Subsequently, female

virgin oﬀspring from this cross is crossed with a male balancer line (see crossing scheme

below). Balanced oﬀspring from this second cross can now be screened for the presence of

both desired alleles, i.e. by phenotypic markers or PCR. − − − gene a − gene b w ;; TTG × w ;; TTG

' ↓ ♂ − w−;; gene a × w−;; Dr gene b− TTG ' ↓ ♂ − − − gene a , gene b w ;; TTG

6.6.3 Collecting Drosophila melanogaster embryos

For the collection of Drosophila embryos, adult virgins and males were kept in a plastic

cage on apple juice agar plates (Cold Spring Harbor Protocols, 2011a) with some yeast

paste. To stimulate rapid egg laying, the plates were changed every 2 - 3 hours for one day.

This way, females would lay eggs soon after fertilization so that the developmental stage of

the embryos could be estimated as accurately as possible from the time after egg lay. Adult

ﬂies were allowed to lay eggs for a certain period of time according to the developmental

stage of interest. Embryos were harvested by covering the plates in dH2O and swirling a soft paintbrush to loosen the embryos. This supension was poured through a stainless

steel mesh to retain only embryos.

Apple juice agar yeast paste

600 ml dH2O 5 ml dH2O 18 g agar 5 ml 10 % acetic acid (Merck Millipore, 200 ml apple juice Billerica, MA, USA 20 g sucrose 7.5 g baker’s yeast (Sigma-Aldrich, St. Louis, 20 ml 20 % Nipagin M MO, USA) 94 Materials and Methods

6.6.4 Histochemistry

6.6.4.1 Embryo dechorionation

After recovering Drosophila eggs from apple juice agar plates, embryos were covered in a 1:2 dilution of Sodium hypochlorite (stock solution: 6 - 14 %, Merck Millipore) for 3 minutes. The sodium hypochlorite removes the chorion without damaging the vitelline membrane.

6.6.4.2 Embryo ﬁxation for light microscopy of whole specimens

Dechorionated embryos were ﬁxed using either a heat ﬁxation protocol or a formaldehyde

ﬁxation protocol.

Heat ﬁxation of Drosophila embryos Dechorionated embryos were immersed in 2 ml of Triton-X-Salt solution (TSS) that had been heated to boiling point in a glass vial.

Immediately afterwards, the vial was ﬁlled up with 15 ml of ice-cold TSS and incubated on ice for 5 minutes. After decanting the TSS, 2 ml of heptane and 2 ml of methanol were added (both Merck Millipore). The vial was shaken vigorously for 30 seconds and the upper phase, consisting of heptane, was removed with a Pasteur Pipette. The embryos were then transferred to a 1.5 ml reaction tube and washed with methanol three times.

Embryos covered in methanol were stored at -20 ◦C until further use.

10×TSS 3 ml Triton-X-100 (Merck Millipore) 40 g NaCl (Sigma-Aldrich)

to 1 l dH2O

Formaldehyde fixation of Drosophila embryos Dechorionated embryos were immersed in a solution of 4 ml of PBS or embryo fixation buffer in a glass vial. 4 ml of heptane and formaldehyde to a final concentration of 4 % (v/v) (Merck Millipore) were added. The vial was shaken gently for 20 minutes at r/t. Subsequently, the lower formaldehyde phase was removed and 3 - 4 ml of methanol were added. The vial was shaken vigorously for 30 seconds and the upper phase, consisting of heptane, was removed with a plastic pasteur pipette. The embryos were then transferred to a 1.5 ml reaction tube and washed with methanol three times. Embryos were stored in methanol at -20 ◦C until further use. Work with Drosophila melanogaster 95

When ﬁxing embryos for Phalloidin staining, methanol was replaced with 90 % ethanol in all steps.

Embryo ﬁxation buﬀer 100 mM Hepes (Merck Millipore)

2 mM MgSO4 (Sigma-Aldrich) 1 mM EGTA (Sigma-Aldrich)

6.6.4.3 Embryo ﬁxation for light microscopy of semi-thin sections (Tepass and

Hartenstein, 1994)

For the production of semi-thin sections, dechorionated embryos were pre-ﬁxed by submerging them in 250 µl 100 µM phosphate buﬀer pH 7.2 mixed with 250 µl of 50 % glutaraldehyde and adding 2 ml of heptane. This assay was shaken gently for 20 minutes at r/t. The

Heptane phase was then removed and the embryos were retrieved from the vial using a nylon mesh that had been shaped into a small spoon-like shape and fixed to a 200 µl pipet tip using heat. With a soft brush, the embryos were transferred to a strip of clear double-sided tape fixed to a glass slide. After letting the embryos dry for about one minute, they were covered with phosphate buffer. Subsequently, the embryos were devitellinized by hand using a pair of fine forceps. Embryos were then stored at 4 ◦C in phosphate buffer containing 1 % glutaraldehyde for a short amount of time before they were fixed again with osmium and glutaraldehyde. For this, the same amounts of osmium solution and glutaraldehyde solution were freshly mixed and the embryos were fixed vor 30 minutes on ice. After three washing steps using phosphate buffer, the samples were fixed again using only osmium solution.

Osmium solution Glutaraldehyde solution 500 µl 4 % osmium 100 µl 50 % glutaraldehyde 500 µl 100 µM phosphate 1.15 ml 100 µM phosphate buﬀer pH 7.2 buﬀer pH 7.2

6.6.4.4 Antibody staining of embryos for Immunoﬂuorescence

Embryos stored in methanol or 90 were rehydrated by washing them three times in PBST for 5 minutes. Embryos were then blocked for one hour in 1×PBST + 5 % normal horse serum. After blocking, embryos were incubated with 50 - 100 µl of primary antibodies diluted in 1×PBST at 4 ◦C overnight. On the next day, the primary antibody solution 96 Materials and Methods was removed and the embryos were washed three times with 1×PBST. Subsequently, the embryos were incubated with the secondary antibody solution for one hour (secondary antibodies were coupled to a ﬂuorophore, usually Alexa-488, Alexa-568 or Alexa-647).

Finally, the embryos were washed again six times with 1×PBST before mounting.

6.6.4.5 Phalloidin and DAPI staining of embryos

When staining actin, embryos were ﬁxed and washed using 90 % ethanol for all steps. Actin stainings were carried out using Phalloidin-Alexa555 (stock solution: 6.6 µM, Thermo

Scientiﬁc) diluted 1:100 in PBST. Embryos were incubated with Phalloidin-Alexa555 for one hour at room temperature.

For DAPI stainings, embryos could be ﬁxed with any protocol. Staining was carried out with DAPI diluted 1:5,000 in PBST (stock solution: 4mg/ml; Roche) for ten minutes at room temperature. Embryos were washed twice for ﬁve minutes with PBST and then mounted in glycerol propyl gallate on microscopy slides as explained below.

6.6.4.6 Embryo mounting for analysis via ﬂuorescence microscopy

Embryos stained with ﬂuorescent antibodies, Phalloidin or DAPI were mounted in glycerol propyl gallate. For this, embryos suspended in 1×PBST were drawn into a 200 µl pipet tip of which the end had been cut diagonally. A drop of propyl gallate was placed on a 76x26x1 mm microscopy glass slide (Marienfeld, Lauda-Königshofen, GER). The embryos were then pushed out of the pipet tip very carefully, trying to get as many embryos as possible into a very small drop of PBST at the end of the pipet tip. This drop was dipped into the propyl gallate and the embryos gently spread over the glass slide. An appropriately-sized cover slip with a thickness of 0.17 +/- 0.005 mm (Menzel-Gläser, Braunschweig, GER) for confocal

1 laser scanning microscopy or a glass coverslip with a thickness of 1 2 (Corning, Corning, NY, USA) for all other kinds of light microscopy was lowered onto the slide and its edges were sealed with clear nail polish (Yves Rocher, La Gacilly, FRA). Fluorescent specimens were imaged with a Zeiss LSM 780 NLO confocal laser scanning microscope using a Zeiss

C-Apochromat 40x 1.2 W objective (Carl Zeiss AG, Oberkochen, GER). Specimens stained with Alexa-488 were illuminated with an Argon multiline 458/ 477/ 488/ 514 nm laser, specimens stained with Alexa-647 were illuminated with a 633 nm Helium-Neon laser and Work with Drosophila melanogaster 97 specimens stained with Alexa-555 or Alexa-568 were illuminated with a diode-pumped solid state 561 nm laser.

Glycerol propyl gallate 50 mg/ ml propyl gallate 75 % glycerol

25 % dH2O

6.6.4.7 Embryo mounting for live imaging

To observe morphogenetic processes in vivo, embryos carrying the allele DE-cad::GFP were used (Huang et al., 2009). This allele is a GFP knock-in allele of shotgun, which results in

DE-cad being marked with GFP in all cells. This allele was combined with alleles of Cdep to observe how morphogenetic defects arise.

Flies homozygous for both DE-cad::GFP and the desired Cdep allele were allowed to lay eggs on an applejuice agar plate for one hour. The plates were then incubated for the appropriate amount of time the embryos needed to reach the desired developmental stage.

A square of about 4mm × 4 mm apple juice agar was cut from an apple juice agar plate.

Embryos were dechorionated by rolling them on double-sided tape (3M, Maplewood, MN,

USA). The dechorionated embryos were put onto the apple juice agar square upside-down.

A Nunc R glass bottom petri dish (ThermoScientiﬁc) whose glass slide was covered with dried heptane glue was gently touched to the embryos on the apple juice agar square.

The embryos attached to the glass slide were then covered with dH20 and imaged using a Zeiss N-Achroplan 20x 0.5 W immersion objective on a Zeiss LSM 780 NLO confocal laser scanning microscope (Carl Zeiss AG).

6.6.4.8 Embedding of embryos for semi-thin sectioning

Embryos ﬁxed with osmium were embedded in Durcopan. They were cut into sections with a thickness of 2 µm and mounted on a glass slide. The sections were then stained with a toluidinblue/ methyleneblue/ borate solution for 10 minutes at 65 ◦C. Sections were imaged using brightﬁeld microscopy on a Zeiss Axio Imager 2 (Carl Zeiss AG).

6.6.4.9 Cuticle preparation from hatched and unhatched Drosophila larvae

In order to quickly assess the overall morphology of Drosophila embryos and L1 larvae, cuticle preparations were carried out. Adult ﬂies were allowed to mate and lay eggs for a 98 Materials and Methods short amount of time in the morning, i.e. for one or two hours. The apple juice agar plates with the eggs were then incubated at 29 ◦C. If necessary, heterozygote embryos carrying a balancer chromosome, discernible by the ﬂuorescent proteins expressed by marker genes on the balancer chromosomes, were picked from the plates and discarded after about 8 hours of incubation. The same was done after an additional incubation time of 14 - 16 hours.

Altogether, the eggs were allowed to develop for 36 hours to make sure that all larvae could hatch. All eggs and larvae were collected from the plate and a dechorionation step was carried out. Finally, all embryos and larvae were mounted in Hoyer’s medium (Cold

Spring Harbor Protocols, 2011b) on 76×26×1mm glass slides. The mounted specimens

1 ◦ were covered with glass coverslips that had a thickness of 1 2 , and incubated at 65 C overnight. On the next day, the slides were sealed with clear nail polish and subsequently imaged using darkﬁeld microscopy on a Zeiss Axio Imager 2.

Hoyer’s medium 15 g gum arabic

25 ml dH2O 100 g chloral hydrate 10 g glycerol

Dissolve gum arabic in water heated to 60 ◦C in a fume hood, stir overnight. Add chloral hydrate, let dissolve. Add glycerol. Centrifuge for 30 minutes at 10,000 g, then

ﬁlter through glass wool. Before use, dilute with 10 % - 15 % H2O and centrifuge for ten minutes at maximum speed in a table-top centrifuge.

6.6.4.10 Preparation and staining of ovarian follicles from Drosophila females

To assess the morphology of egg chambers of Drosophila melanogaster, ﬁve to ten virgin females were put into a food vial with some yeast powder. The females were incubated for 36 - 48 hours and subsequently, the ovaries were dissected. For this, the females were anesthesized and their abdomen sliced open with one tip of a pair of sharp forceps. The ovaries were pushed out of this opening, cleaned of all other organs and the ovarioles were separated with a thin dissection needle. The ovaries were ﬁxed with 4 % PFA in ice-cold

PBS pH7.2 for 20 minutes. After ﬁxation, several washing steps were carried out: twice for

ﬁve minutes with PBS pH7.2, three times for 15 minutes with PBST pH7.2 (PBS pH7.2 with 0.5 % Triton X-100) and three times for 20 minutes in blocking solution (PBST pH7.2 with 5 % BSA). Work with Drosophila melanogaster 99

Ovaries were stained with Phalloidin-Alexa488 diluted 1:100 in PBST pH7.2 for one hour at room temperature. They were washed three more times for 20 minutes in PBST. During the second washing step, they were incubated with DAPI diluted 1:2,000 (stock solution:

4 mg/ ml). Egg chambers were mounted in prolongGold (invitrogen) on a 76x26x1 mm microscopy glass slide (Marienfeld) and covered with a 22 mm × 22 mm cover slip with a thickness of 0.17 +/- 0.005 mm (Menzel-Gläser). During the mounting process, large egg chambers with very mature eggs were discarded and ovarioles were seperated as much as possible. The specimens were imaged with a Zeiss LSM 780 NLO confocal laser scanning microscope using a Zeiss C-Apochromat 40x 1.2 W objective (Carl Zeiss AG).

PBS pH7.2 for ovaries 68.4 mM NaCl

8.2 mM KH2PO4 28 mM Na2HPO4 pH7.2

6.6.4.11 mRNA in situ hybridization of whole-mount Drosophila embryos

Embryos were collected and incubated to reach the desired developmental stage. They were dechorionated and fixed with 4 % PFA, using embryo fixation buffer (see below).

Hybridization of the DIG-labeled RNA probes to the transcripts in the ﬁxed embryo and subsequent probe detection was carried out as described in Tautz (2000). After washing the embryos ﬁrst in methanol and then in PBST (1×PBS with 0.02 % Tween 20) after

ﬁxation, a postﬁxation step was carried out with 4 % PFA in PBST. Hybridization was carried out at 65 ◦C overnight. Both the binding (antisense) and non-binding (sense) probe were used for hybridization in separate experiments. The non-binding probe served as a negative control.

XV Bibliography

Abbott MK and Lengyel JA. Embryonic head involution and rotation of male terminalia require the Drosophila locus head involution defective. Genetics, 129(3):783–789, 1991.

Abrams EW. Fork head and Sage maintain a uniform and patent salivary gland lumen through regulation of two downstream target genes, PH4 SG1 and PH4 SG2. Development, 133(18):3517–3527, 2006.

Adachi-Yamada T and O’Connor MB. Morphogenetic apoptosis: a mechanism for correcting discontinuities in morphogen gradients. Dev Biol, 251(1):74–90, 2002.

Altschul SF, Madden TL, Schäﬀer AA, Zhang J, Zhang Z, Miller W and Lipman DJ. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic acids research, 25(17):3389–3402, 1997.

Altschul SF, Wootton JC, Gertz EM, Agarwala R, Morgulis A, Schäﬀer AA and Yu YK. Protein database searches using compositionally adjusted substitution matrices. The FEBS journal, 272(20):5101–5109, 2005.

Arquier N, Perrin L, Manfruelli P and Sémériva M. The Drosophila tumor suppressor gene lethal(2)giant larvae is required for the emission of the Decapentaplegic signal. Development, 128(12):2209–2220, 2001.

Ashburner M. Drosophila: A laboratory manual. Cold Spring Harbor Laboratory Pr, 1989.

Bachmann A, Schneider M, Theilenberg E, Grawe F and Knust E. Drosophila Stardust is a partner of Crumbs in the control of epithelial cell polarity. Nature, 414(6864):638–643, 2001.

Baines AJ. A FERM-adjacent (FA) region deﬁnes a subset of the 4.1 superfamily and is a potential regulator of FERM domain function. BMC genomics, 7(1):85, 2006.

Bantignies F, Goodman RH and Smolik SM. The interaction between the coactivator dCBP and Modulo, a chromatin-associated factor, aﬀects segmentation and melanotic tumor formation in Drosophila. Proc Natl Acad Sci U S A, 99(5):2895–2900, 2002.

Barrangou R, Fremaux C, Deveau H, Richards M, Boyaval P, Moineau S, Romero DA and Horvath P. CRISPR provides acquired resistance against viruses in prokaryotes. Science, 315(5819):1709–1712, 2007.

Bassett AR, Tibbit C, Ponting CP and Liu JL. Highly eﬃcient targeted mutagenesis of Drosophila with the CRISPR/Cas9 system. Cell reports, 4(1):220–228, 2013.

Baumgartner S, Littleton JT, Broadie K, Bhat MA, Harbecke R, Lengyel JA, Chiquet-Ehrismann R, Prokop A and Bellen HJ. A Drosophila neurexin is required for septate junction and blood-nerve barrier formation and function. Cell, 87(6):1059–1068, 1996.

Becker PM, Tran TS, Delannoy MJ, He C, Shannon JM and McGrath-Morrow S. Semaphorin 3A contributes to distal pulmonary epithelial cell diﬀerentiation and lung morphogenesis. PLoS ONE, 6(11):e27449, 2011.

Behr M, Riedel D and Schuh R. The claudin-like megatrachea is essential in septate junctions for the epithelial barrier function in Drosophila. Dev Cell, 5(4):611–620, 2003. XVI Bibliography

Bellen HJ, Levis RW, Liao G, He Y, Carlson JW, Tsang G, Evans-Holm M, Hiesinger PR, Schulze KL, Rubin GM, Hoskins RA and Spradling AC. The BDGP gene disruption project: single transposon insertions associated with 40% of Drosophila genes. Genetics, 167(2):761–781, 2004.

Benton R and St Johnston D. A conserved oligomerization domain in drosophila Bazooka/PAR-3 is important for apical localization and epithelial polarity. Current Biology, 13(15):1330–1334, 2003a.

Benton R and St Johnston D. Drosophila PAR-1 and 14-3-3 inhibit Bazooka/PAR-3 to establish complementary cortical domains in polarized cells. Cell, 115(6):691–704, 2003b.

Bertani G. Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. Journal of Bacteriology, 1951.

Bertet C, Sulak L and Lecuit T. Myosin-dependent junction remodelling controls planar cell intercalation and axis elongation. Nature, 429(6992):667–671, 2004.

Betschinger J, Mechtler K and Knoblich JA. The Par complex directs asymmetric cell division by phosphorylating the cytoskeletal protein Lgl. Nature, 422(6929):326–330, 2003.

Bhat MA, Izaddoost S, Lu Y, Cho KO, Choi KW and Bellen HJ. Discs Lost, a novel multi-PDZ domain protein, establishes and maintains epithelial polarity. Cell, 96(6):833–845, 1999.

Bilder D, Li M and Perrimon N. Cooperative regulation of cell polarity and growth by Drosophila tumor suppressors. Science, 289(5476):113–116, 2000.

Bilder D and Perrimon N. Localization of apical epithelial determinants by the basolateral PDZ protein Scribble. Nature, 403(6770):676–680, 2000.

Bilder D, Schober M and Perrimon N. Integrated activity of PDZ protein complexes regulates epithelial polarity. Nat Cell Biol, 5(1):53–58, 2003.

Blanchard GB, Murugesu S, Adams RJ, Martinez Arias A and Gorﬁnkiel N. Cytoskeletal dynamics and supracellular organisation of cell shape ﬂuctuations during dorsal closure. Development, 137(16):2743–2752, 2010.

Borghi N, Sorokina M, Shcherbakova OG, Weis WI, Pruitt BL, Nelson WJ and Dunn AR. E-cadherin is under constitutive actomyosin-generated tension that is increased at cell-cell contacts upon externally applied stretch. Proc Natl Acad Sci U S A, 109(31):12568–12573, 2012.

Bulgakova NA, Kempkens O and Knust E. Multiple domains of Stardust diﬀerentially mediate localisation of the Crumbs-Stardust complex during photoreceptor development in Drosophila. J Cell Sci, 121(Pt 12):2018–2026, 2008.

Campbell K, Knust E and Skaer H. Crumbs stabilises epithelial polarity during tissue remodelling. Journal of Cell Science, 122(15):2604–2612, 2009.

Campos-Ortega JA and Hartenstein V. The Embryonic Development of Drosophila melanogaster. Springer Science & Business Media, Berlin, Heidelberg, 2013. Bibliography XVII

Cheadle L and Biederer T. The novel synaptogenic protein Farp1 links postsynaptic cytoskeletal dynamics and transsynaptic organization. The Journal of Cell Biology, 199(6):985–1001, 2012.

Cheeks RJ, Canman JC, Gabriel WN, Meyer N, Strome S and Goldstein B. C. elegans PAR proteins function by mobilizing and stabilizing asymmetrically localized protein complexes. Current Biology, 14(10):851–862, 2004.

Chen CS, Hong S, Indra I, Sergeeva AP, Troyanovsky RB, Shapiro L, Honig B and Troyanovsky SM. α-Catenin-mediated cadherin clustering couples cadherin and actin dynamics. The Journal of Cell Biology, 105(1):13, 2015.

Chen J and Zhang M. The Par3/Par6/aPKC complex and epithelial cell polarity. Experimental Cell Research, 319(10):1357–1364, 2013.

Chen JC, Lin BB, Hu HW, Lin C, Jin WY, Zhang FB, Zhu YA, Lu CJ, Wei XJ and Chen RJ. NGF accelerates cutaneous wound healing by promoting the migration of dermal ﬁbroblasts via the PI3K/Akt-Rac1-JNK and ERK pathways. BioMed research international, 2014(7):547187–13, 2014.

Chen K, Merino C, Sigrist SJ and Featherstone DE. The 4.1 protein coracle mediates subunit-selective anchoring of Drosophila glutamate receptors to the postsynaptic actin cytoskeleton. The Journal of neuroscience : the oﬃcial journal of the Society for Neuroscience, 25(28):6667–6675, 2005.

Chen P, Nordstrom W, Gish B and Abrams JM. grim, a novel cell death gene in Drosophila. Genes & development, 10(14):1773–1782, 1996.

Chen X and Ganetzky B. A neuropeptide signaling pathway regulates synaptic growth in Drosophila. The Journal of Cell Biology, 196(4):529–543, 2012.

Chishti AH, Kim AC, Marfatia SM, Lutchman M, Hanspal M, Jindal H, Liu SC, Low PS, Rouleau GA, Mohandas N, Chasis JA, Conboy JG, Gascard P, Takakuwa Y, Huang SC, Benz EJ, Bretscher A, Fehon RG, Gusella JF, Ramesh V, Solomon F, Marchesi VT, Tsukita S and Hoover KB. The FERM domain: a unique module involved in the linkage of cytoplasmic proteins to the membrane. Trends in biochemical sciences, 23(8):281–282, 1998.

Cho HJ, Hwang YS, Mood K, Ji YJ, Lim J, Morrison DK and Daar IO. EphrinB1 interacts with CNK1 and promotes cell migration through c-Jun N-terminal kinase (JNK) activation. Journal of Biological Chemistry, 289(26):18556–18568, 2014.

Choi W, Jung KC, Nelson KS, Bhat MA, Beitel GJ, Peifer M and Fanning AS. The single Drosophila ZO-1 protein Polychaetoid regulates embryonic morphogenesis in coordination with Canoe/afadin and Enabled. Molecular biology of the cell, 22(12):2010–2030, 2011.

Cold Spring Harbor Protocols. Drosophila apple juice-agar plates. Cold Spring Harbor protocols, 2011(9):pdb.rec065672–pdb.rec065672, 2011a.

Cold Spring Harbor Protocols. Hoyer’s Medium. Cold Spring Harbor protocols, 2011(4):pdb.rec12429–pdb.rec12429, 2011b.

Cong L, Ran FA, Cox D, Lin S, Barretto R, Habib N, Hsu PD, Wu X, Jiang W, Marraﬃni LA and Zhang F. Multiplex genome engineering using CRISPR/Cas systems. Science, 339(6121):819–823, 2013. XVIII Bibliography

Davidson AE, Schwarz N, Zelinger L, Stern-Schneider G, Shoemark A, Spitzbarth B, Gross M, Laxer U, Sosna J, Sergouniotis PI, Waseem NH, Wilson R, Kahn RA, Plagnol V, Wolfrum U, Banin E, Hardcastle AJ, Cheetham ME, Sharon D and Webster AR. Mutations in ARL2BP, encoding ADP-ribosylation-factor-like 2 binding protein, cause autosomal-recessive retinitis pigmentosa. American journal of human genetics, 93(2):321–329, 2013. de Vreede G, Schoenfeld JD, Windler SL, Morrison H, Lu H and Bilder D. The Scribble module regulates retromer-dependent endocytic traﬃcking during epithelial polarization. Development, 141(14):2796–2802, 2014.

Desai R, Sarpal R, Ishiyama N, Pellikka M, Ikura M and Tepass U. Monomeric α-catenin links cadherin to the actin cytoskeleton. Nat Cell Biol, 15(3):261–273, 2013.

Devine WP, Lubarsky B, Shaw K, Luschnig S, Messina L and Krasnow MA. Requirement for chitin biosynthesis in epithelial tube morphogenesis. Proc Natl Acad Sci U S A, 102(47):17014–17019, 2005.

Eivers E, Fuentealba LC, Sander V, Clemens JC, Hartnett L and De Robertis EM. Mad is required for wingless signaling in wing development and segment patterning in Drosophila. PLoS ONE, 4(8):e6543, 2009.

Fehon RG, Dawson IA and Artavanis-Tsakonas S. A Drosophila homologue of membrane-skeleton protein 4.1 is associated with septate junctions and is encoded by the coracle gene. Development, 120(3):545–557, 1994.

Fernández BG, Arias AM and Jacinto A. Dpp signalling orchestrates dorsal closure by regulating cell shape changes both in the amnioserosa and in the epidermis. Mechanisms of development, 124(11-12):884–897, 2007.

Fiévet B, Louvard D and Arpin M. ERM proteins in epithelial cell organization and functions. Biochimica et biophysica acta, 1773(5):653–660, 2007.

Frank LH and Rushlow C. A group of genes required for maintenance of the amnioserosa tissue in Drosophila. Development, 122(5):1343–1352, 1996.

Fukuhara T, Shimizu K, Kawakatsu T, Fukuyama T, Minami Y, Honda T, Hoshino T, Yamada T, Ogita H, Okada M and Takai Y. Activation of Cdc42 by trans interactions of the cell adhesion molecules nectins through c-Src and Cdc42-GEF FRG. J Cell Biol, 166(3):393–405, 2004.

Furuse M, Sasaki H and Tsukita S. Manner of interaction of heterogeneous claudin species within and between tight junction strands. J Cell Biol, 147(4):891–903, 1999.

Gamblin CL, Hardy ÉJL, Chartier FJM, Bisson N and Laprise P. A bidirectional antagonism between aPKC and Yurt regulates epithelial cell polarity. The Journal of Cell Biology, 2014.

Genova JL and Fehon RG. Neuroglian, Gliotactin, and the Na+/K+ ATPase are essential for septate junction function in Drosophila. J Cell Biol, 161(5):979–989, 2003.

Georgiou M, Marinari E, Burden J and Baum B. Cdc42, Par6, and aPKC regulate Arp2/3-mediated endocytosis to control local adherens junction stability. Current Biology, 18(21):1631–1638, 2008. Bibliography XIX

Gibson DG, Young L, Chuang RY, Venter JC, Hutchison CA and Smith HO. Enzymatic assembly of DNA molecules up to several hundred kilobases. Nature methods, 6(5):343–345, 2009.

Gorﬁnkiel N and Arias AM. Requirements for adherens junction components in the interaction between epithelial tissues during dorsal closure in Drosophila. J Cell Sci, 120(Pt 18):3289–3298, 2007.

Gratz SJ, Ukken FP, Rubinstein CD, Thiede G, Donohue LK, Cummings AM and O’Connor-Giles KM. Highly Speciﬁc and Eﬃcient CRISPR/Cas9-Catalyzed Homology-Directed Repair in Drosophila. Genetics, 2014.

Grawe F, Wodarz A, Lee B, Knust E and Skaer H. The Drosophila genes crumbs and stardust are involved in the biogenesis of adherens junctions. Development, 122(3):951–959, 1996.

Grether ME, Abrams JM, Agapite J, White K and Steller H. The head involution defective gene of Drosophila melanogaster functions in programmed cell death. Genes & development, 9(14):1694–1708, 1995.

Gumbiner BM. Epithelial morphogenesis. Cell, 69(3):385–387, 1992.

Haklai-Topper L, Mlechkovich G, Savariego D, Gokhman I and Yaron A. Cis interaction between Semaphorin6A and Plexin-A4 modulates the repulsive response to Sema6A. The EMBO Journal, 29(15):2635–2645, 2010.

Hanahan D. Studies on transformation of Escherichia coli with plasmids. Journal of molecular biology, 166(4):557–580, 1983.

Harris KP and Tepass U. Cdc42 and Par proteins stabilize dynamic adherens junctions in the Drosophila neuroectoderm through regulation of apical endocytosis. The Journal of Cell Biology, 183(6):1129–1143, 2008.

Harris TJC and Peifer M. Adherens junction-dependent and -independent steps in the establishment of epithelial cell polarity in Drosophila. J Cell Biol, 167(1):135–147, 2004.

Harris TJC and Peifer M. The positioning and segregation of apical cues during epithelial polarity establishment in Drosophila. J Cell Biol, 170(5):813–823, 2005.

Harris TJC and Peifer M. aPKC controls microtubule organization to balance adherens junction symmetry and planar polarity during development. Dev Cell, 12(5):727–738, 2007.

He X, Kuo YC, Rosche TJ and Zhang X. Structural basis for autoinhibition of the guanine nucleotide exchange factor FARP2. Structure (London, England : 1993), 21(3):355–364, 2013.

Hoover KB and Bryant PJ. Drosophila Yurt is a new protein-4.1-like protein required for epithelial morphogenesis. Development genes and evolution, 212(5):230–238, 2002.

Houssin E, Tepass U and Laprise P. Girdin-mediated interactions between cadherin and the actin cytoskeleton are required for epithelial morphogenesis in Drosophila. Development, 142(10):1777–1784, 2015.

Huang J, Zhou W, Dong W, Watson AM and Hong Y. From the Cover: Directed, eﬃcient, and versatile modiﬁcations of the Drosophila genome by genomic engineering. Proc Natl Acad Sci U S A, 106(20):8284–8289, 2009. XX Bibliography

Hutterer A, Betschinger J, Petronczki M and Knoblich JA. Sequential Roles of Cdc42, Par-6, aPKC, and Lgl in the Establishment of Epithelial Polarity during Drosophila Embryogenesis. Developmental Cell, 6(6):845–854, 2004.

Huynh JR, Petronczki M, Knoblich JA and St Johnston D. Bazooka and PAR-6 are required with PAR-1 for the maintenance of oocyte fate in Drosophila. Current Biology, 11(11):901–906, 2001.

Hwang WY, Fu Y, Reyon D, Maeder ML, Tsai SQ, Sander JD, Peterson RT, Yeh JRJ and Joung JK. Eﬃcient genome editing in zebraﬁsh using a CRISPR-Cas system. Nat Biotechnol, 31(3):227–229, 2013.

Igaki T. Correcting developmental errors by apoptosis: lessons from Drosophila JNK signaling. Apoptosis : an international journal on programmed cell death, 14(8):1021–1028, 2009.

Ito T, Kagoshima M, Sasaki Y, Li C, Udaka N, Kitsukawa T, Fujisawa H, Taniguchi M, Yagi T, Kitamura H and Goshima Y. Repulsive axon guidance molecule Sema3A inhibits branching morphogenesis of fetal mouse lung. Mechanisms of development, 97(1-2):35–45, 2000.

Jacinto A, Wood W, Balayo T, Turmaine M, Martinez-Arias A and Martin P. Dynamic actin-based epithelial adhesion and cell matching during Drosophila dorsal closure. Current Biology, 10(22):1420–1426, 2000.

Jinek M, Chylinski K, Fonfara I, Hauer M, Doudna JA and Charpentier E.A programmable dual-RNA-guided DNA endonuclease in adaptive bacterial immunity. Science, 337(6096):816–821, 2012.

Kameya S, Hawes NL, Chang B, Heckenlively JR, Naggert JK and Nishina PM. Mfrp, a gene encoding a frizzled related protein, is mutated in the mouse retinal degeneration 6. Human molecular genetics, 11(16):1879–1886, 2002.

Kanda H and Miura M. Regulatory roles of JNK in programmed cell death. Journal of biochemistry, 136(1):1–6, 2004.

Kemler R. From cadherins to catenins: cytoplasmic protein interactions and regulation of cell adhesion. Trends in genetics : TIG, 9(9):317–321, 1993.

Kempkens O, Médina E, Fernandez-Ballester G, Özüyaman S, Le Bivic A, Serrano L and Knust E. Computer modelling in combination with in vitro studies reveals similar binding aﬃnities of Drosophila Crumbs for the PDZ domains of Stardust and DmPar-6. European Journal of Cell Biology, 85(8):753–767, 2006.

Khan SY, Ali S, Naeem MA, Khan SN, Husnain T, Butt NH, Qazi ZA, Akram J, Riazuddin S, Ayyagari R, Hejtmancik JF and Riazuddin SA. Splice-site mutations identiﬁed in PDE6A responsible for retinitis pigmentosa in consanguineous Pakistani families. Molecular vision, 21:871–882, 2015.

Kiehart DP, Galbraith CG, Edwards KA, Rickoll WL and Montague RA. Multiple forces contribute to cell sheet morphogenesis for dorsal closure in Drosophila. J Cell Biol, 149(2):471–490, 2000.

Kivirikko KI and Pihlajaniemi T. Collagen hydroxylases and the protein disulﬁde isomerase subunit of prolyl 4-hydroxylases. Advances in enzymology and related areas of molecular biology, 72:325–398, 1998. Bibliography XXI

Knust E. Control of epithelial cell polarity in Drosophila. Trends in genetics : TIG, 10(8):275–280, 1994.

Knust E. Drosophila morphogenesis: follow-my-leader in epithelia. Current Biology, 6(4):379–381, 1996.

Knust E. Composition and Formation of Intercellular Junctions in Epithelial Cells. Science, 298(5600):1955–1959, 2002.

Knust E and Leptin M. Adherens junctions in the Drosophila embryo: the role of E-cadherin in their establishment and morphogenetic function. BioEssays : news and reviews in molecular, cellular and developmental biology, 18(8):609–612, 1996.

Koyano Y, Kawamoto T, Kikuchi A, Shen M, Kuruta Y, Tsutsumi S, Fujimoto K, Noshiro M, Fujii K and Kato Y. Chondrocyte-derived ezrin-like domain containing protein (CDEP), a rho guanine nucleotide exchange factor, is inducible in chondrocytes by parathyroid hormone and cyclic AMP and has transforming activity in NIH3T3 cells. Osteoarthritis and cartilage / OARS, Osteoarthritis Research Society, 9 Suppl A:S64–8, 2001.

Koyano Y, Kawamoto T, Shen M, Yan W, Noshiro M, Fujii K and Kato Y. Molecular cloning and characterization of CDEP, a novel human protein containing the ezrin-like domain of the band 4.1 superfamily and the Dbl homology domain of Rho guanine nucleotide exchange factors. Biochemical and biophysical research communications, 241(2):369–375, 1997.

Krahn MP, Buckers J, Kastrup L and Wodarz A. Formation of a Bazooka-Stardust complex is essential for plasma membrane polarity in epithelia. The Journal of Cell Biology, 190(5):751–760, 2010.

Kubo T, Yamashita T, Yamaguchi A, Sumimoto H, Hosokawa K and Tohyama M. A novel FERM domain including guanine nucleotide exchange factor is involved in Rac signaling and regulates neurite remodeling. The Journal of neuroscience : the oﬃcial journal of the Society for Neuroscience, 22(19):8504–8513, 2002.

Laemmli UK. Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature, 227(5259):680–685, 1970.

Lamb RS, Ward RE, Schweizer L and Fehon RG. Drosophila coracle, a member of the protein 4.1 superfamily, has essential structural functions in the septate junctions and developmental functions in embryonic and adult epithelial cells. Molecular biology of the cell, 9(12):3505–3519, 1998.

Lane NJ and Skaer Hl. Intercellular Junctions in Insect Tissues. 1980.

Laprise P, Beronja S, Silva-Gagliardi NF, Pellikka M, Jensen AM, McGlade CJ and Tepass U. The FERM protein Yurt is a negative regulatory component of the Crumbs complex that controls epithelial polarity and apical membrane size. Dev Cell, 11(3):363–374, 2006.

Laprise P, Lau KM, Harris KP, Silva-Gagliardi NF, Paul SM, Beronja S, Beitel GJ, McGlade CJ and Tepass U. Yurt, Coracle, Neurexin IV and the Na(+),K(+)-ATPase form a novel group of epithelial polarity proteins. Nature, 459(7250):1141–1145, 2009. XXII Bibliography

Laprise P, Paul SM, Boulanger J, Robbins RM, Beitel GJ and Tepass U. Epithelial polarity proteins regulate Drosophila tracheal tube size in parallel to the luminal matrix pathway. Current biology : CB, 20(1):55–61, 2010.

Lecuit T. Junctions and vesicular traﬃcking during Drosophila cellularization. J Cell Sci, 117(Pt 16):3427–3433, 2004.

Leto TL and Marchesi VT. A structural model of human erythrocyte protein 4.1. The Journal of biological chemistry, 259(7):4603–4608, 1984.

Llimargas M and Casanova J. ventral veinless, a POU domain transcription factor, regulates diﬀerent transduction pathways required for tracheal branching in Drosophila. Development, 124(17):3273–3281, 1997.

Lu H and Bilder D. Endocytic control of epithelial polarity and proliferation in Drosophila. Nat Cell Biol, 7(12):1232–1239, 2005.

Luschnig S, Bätz T, Armbruster K and Krasnow MA. serpentine and vermiform encode matrix proteins with chitin binding and deacetylation domains that limit tracheal tube length in Drosophila. Current Biology, 16(2):186–194, 2006.

Lynch HE, Crews SM, Rosenthal B, Kim E, Gish R, Echiverri K and Hutson MS. Cellular mechanics of germ band retraction in Drosophila. Dev Biol, 384(2):205–213, 2013.

Lynch HE, Veldhuis J, Brodland GW and Hutson MS. Modeling cell elongation during germ band retraction: cell autonomy versus applied anisotropic stress. New journal of physics, 16(2014):055003, 2014.

Maniatis T. Fritsch, EF, and Sambrook, J.(1982) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory, 1.

Manno S, Takakuwa Y and Mohandas N. Modulation of erythrocyte membrane mechanical function by protein 4.1 phosphorylation. The Journal of biological chemistry, 280(9):7581–7587, 2005.

Mavrakis M, Rikhy R and Lippincott-Schwartz J. Plasma membrane polarity and compartmentalization are established before cellularization in the ﬂy embryo. Dev Cell, 16(1):93–104, 2009.

Médina E, Williams J, Klipfell E, Zarnescu D, Thomas G and Le Bivic A. Crumbs interacts with moesin and beta(Heavy)-spectrin in the apical membrane skeleton of Drosophila. J Cell Biol, 158(5):941–951, 2002.

Merriam JR. The initiation of nonhomologous chromosome pairing before exchange in female Drosophila melanogaster. Genetics, 57(2):409–425, 1967.

Metaxakis A, Oehler S, Klinakis A and Savakis C. Minos as a genetic and genomic tool in Drosophila melanogaster. Genetics, 171(2):571–581, 2005.

Millard TH and Martin P. Dynamic analysis of ﬁlopodial interactions during the zippering phase of Drosophila dorsal closure. Development, 135(4):621–626, 2008.

Miyamoto Y, Yamauchi J and Itoh H. Src kinase regulates the activation of a novel FGD-1-related Cdc42 guanine nucleotide exchange factor in the signaling pathway from the endothelin A receptor to JNK. The Journal of biological chemistry, 278(32):29890–29900, 2003. Bibliography XXIII

Mizuno N, Kokubu H, Sato M, Nishimura A, Yamauchi J, Kurose H and Itoh H. G protein-coupled receptor signaling through Gq and JNK negatively regulates neural progenitor cell migration. Proc Natl Acad Sci U S A, 102(35):12365–12370, 2005. modENCODE Consortium, Roy S, Ernst J, Kharchenko PV, Kheradpour P, Negre N, Eaton ML, Landolin JM, Bristow CA, Ma L, Lin MF, Washietl S, Arshinoff BI, Ay F, Meyer PE, Robine N, Washington NL, Di Stefano L, Berezikov E, Brown CD, Candeias R, Carlson JW, Carr A, Jungreis I, Marbach D, Sealfon R, Tolstorukov MY, Will S, Alekseyenko AA, Artieri C, Booth BW, Brooks AN, Dai Q, Davis CA, Duff MO, Feng X, Gorchakov AA, Gu T, Henikoff JG, Kapranov P, Li R, MacAlpine HK, Malone J, Minoda A, Nordman J, Okamura K, Perry M, Powell SK, Riddle NC, Sakai A, Samsonova A, Sandler JE, Schwartz YB, Sher N, Spokony R, Sturgill D, van Baren M, Wan KH, Yang L, Yu C, Feingold E, Good P, Guyer M, Lowdon R, Ahmad K, Andrews J, Berger B, Brenner SE, Brent MR, Cherbas L, Elgin SCR, Gingeras TR, Grossman R, Hoskins RA, Kaufman TC, Kent W, Kuroda MI, Orr-Weaver T, Perrimon N, Pirrotta V, Posakony JW, Ren B, Russell S, Cherbas P, Graveley BR, Lewis S, Micklem G, Oliver B, Park PJ, Celniker SE, Henikoff S, Karpen GH, Lai EC, MacAlpine DM, Stein LD, White KP and Kellis M. Identification of functional elements and regulatory circuits by Drosophila modENCODE. Science, 330(6012):1787–1797, 2010.

Müller HAJ and Bossinger O. Molecular networks controlling epithelial cell polarity in development. Mechanisms of development, 120(11):1231–1256, 2003.

Munjal A and Lecuit T. Actomyosin networks and tissue morphogenesis. Development, 141(9):1789–1793, 2014.

Nagafuchi A, Ishihara S and Tsukita S. The roles of catenins in the cadherin-mediated cell adhesion: functional analysis of E-cadherin-alpha catenin fusion molecules. J Cell Biol, 127(1):235–245, 1994.

Nassif C, Daniel A, Lengyel JA and Hartenstein V. The role of morphogenetic cell death during Drosophila embryonic head development. Dev Biol, 197(2):170–186, 1998.

Nelson KS, Furuse M and Beitel GJ. The Drosophila Claudin Kune-kune is required for septate junction organization and tracheal tube size control. Genetics, 185(3):831–839, 2010.

Oda H, Uemura T, Harada Y, Iwai Y and Takeichi M. A Drosophila homolog of cadherin associated with armadillo and essential for embryonic cell-cell adhesion. Dev Biol, 165(2):716–726, 1994.

Oda H, Uemura T, Shiomi K, Nagafuchi A, Tsukita S and Takeichi M. Identiﬁcation of a Drosophila homologue of alpha-catenin and its association with the armadillo protein. J Cell Biol, 121(5):1133–1140, 1993.

Ohler U. Identiﬁcation of core promoter modules in Drosophila and their application in accurate transcription start site prediction. Nucleic acids research, 34(20):5943–5950, 2006.

Old RW and Primrose SB. Principles of Gene Manipulation. An Introduction to Genetic Engineering. Univ of California Press, 1980. XXIV Bibliography

Parks AL, Cook KR, Belvin M, Dompe NA, Fawcett R, Huppert K, Tan LR, Winter CG, Bogart KP, Deal JE, Deal-Herr ME, Grant D, Marcinko M, Miyazaki WY, Robertson S, Shaw KJ, Tabios M, Vysotskaia V, Zhao L, Andrade RS, Edgar KA, Howie E, Killpack K, Milash B, Norton A, Thao D, Whittaker K, Winner MA, Friedman L, Margolis J, Singer MA, Kopczynski C, Curtis D, Kaufman TC, Plowman GD, Duyk G and Francis-Lang HL. Systematic generation of high-resolution deletion coverage of the Drosophila melanogaster genome. Nature genetics, 36(3):288–292, 2004.

Paul SM, Ternet M, Salvaterra PM and Beitel GJ. The Na+/K+ ATPase is required for septate junction function and epithelial tube-size control in the Drosophila tracheal system. Development, 130(20):4963–4974, 2003.

Peifer M. The product of the Drosophila segment polarity gene armadillo is part of a multi-protein complex resembling the vertebrate adherens junction. J Cell Sci, 105 ( Pt 4):993–1000, 1993.

Peifer M. Cell adhesion and signal transduction: the Armadillo connection. Trends in Cell Biology, 5(6):224–229, 1995.

Petronczki M and Knoblich JA. DmPAR-6 directs epithelial polarity and asymmetric cell division of neuroblasts in Drosophila. Nat Cell Biol, 3(1):43–49, 2001.

Pocha SM and Knust E. Complexities of Crumbs Function and Regulation in Tissue Morphogenesis. Current biology : CB, 23(7):R289–93, 2013.

Port F, Chen HM, Lee T and Bullock SL. Optimized CRISPR/Cas tools for eﬃcient germline and somatic genome engineering in Drosophila. Proc Natl Acad Sci U S A, 111(29):E2967–76, 2014.

Rabinowitz M. Studies on the cytology and early embryology of the egg of Drosophila melanogaster. Journal of Morphology, 69(1):1–49, 1941.

Ramachandran P, Barria R, Ashley J and Budnik V. A critical step for postsynaptic F-actin organization: regulation of Baz/Par-3 localization by aPKC and PTEN. Developmental neurobiology, 69(9):583–602, 2009.

Ricos MG, Harden N, Sem KP, Lim L and Chia W. Dcdc42 acts in TGF-beta signaling during Drosophila morphogenesis: distinct roles for the Drac1/JNK and Dcdc42/TGF-beta cascades in cytoskeletal regulation. J Cell Sci, 112 ( Pt 8):1225–1235, 1999.

Riesgo-Escovar JR and Hafen E. Drosophila Jun kinase regulates expression of decapentaplegic via the ETS-domain protein Aop and the AP-1 transcription factor DJun during dorsal closure. Genes & development, 11(13):1717–1727, 1997.

Roh MH, Makarova O, Liu CJ, Shin K, Lee S, Laurinec S, Goyal M, Wiggins R and Margolis B. The Maguk protein, Pals1, functions as an adapter, linking mammalian homologues of Crumbs and Discs Lost. J Cell Biol, 157(1):161–172, 2002.

Roignot J, Peng X and Mostov K. Polarity in mammalian epithelial morphogenesis. Cold Spring Harbor perspectives in biology, 5(2):a013789–a013789, 2013.

Roy PJ, Zheng H, Warren CE and Culotti JG. mab-20 encodes Semaphorin-2a and is required to prevent ectopic cell contacts during epidermal morphogenesis in Caenorhabditis elegans. Development, 127(4):755–767, 2000. Bibliography XXV

Royou A, Field C, Sisson JC, Sullivan W and Karess R. Reassessing the role and dynamics of nonmuscle myosin II during furrow formation in early Drosophila embryos. Molecular biology of the cell, 15(2):838–850, 2004.

Ryder E, Ashburner M, Bautista-Llacer R, Drummond J, Webster J, Johnson G, Morley T, Chan YS, Blows F, Coulson D, Reuter G, Baisch H, Apelt C, Kauk A, Rudolph T, Kube M, Klimm M, Nickel C, Szidonya J, Maróy P, Pal M, Rasmuson-Lestander A, Ekström K, Stocker H, Hugentobler C, Hafen E, Gubb D, Pﬂugfelder G, Dorner C, Mechler B, Schenkel H, Marhold J, Serras F, Corominas M, Punset A, Roote J and Russell S. The DrosDel deletion collection: a Drosophila genomewide chromosomal deﬁciency resource. Genetics, 177(1):615–629, 2007.

Schöck F and Perrimon N. Molecular mechanisms of epithelial morphogenesis. Annual Review of Cell and Developmental Biology, 18(1):463–493, 2002.

Schultz J, Milpetz F, Bork P and Ponting CP. SMART, a simple modular architecture research tool: identiﬁcation of signaling domains. Proc Natl Acad Sci USA, 95(11):5857–5864, 1998.

Scuderi A, Simin K, Kazuko SG, Metherall JE and Letsou A. scylla and charybde, homologues of the human apoptotic gene RTP801, are required for head involution in Drosophila. Dev Biol, 291(1):110–122, 2006.

Seshaiah P, Miller B, Myat MM and Andrew DJ. pasilla, the Drosophila homologue of the human Nova-1 and Nova-2 proteins, is required for normal secretion in the salivary gland. Dev Biol, 239(2):309–322, 2001.

Shapiro AL, Viñuela E and Maizel JV. Molecular weight estimation of polypeptide chains by electrophoresis in SDS-polyacrylamide gels. Biochemical and biophysical research communications, 28(5):815–820, 1967.

Shaw RJ, Paez JG, Curto M, Yaktine A, Pruitt WM, Saotome I, O’Bryan JP, Gupta V, Ratner N, Der CJ, Jacks T and McClatchey AI. The Nf2 tumor suppressor, merlin, functions in Rac-dependent signaling. Dev Cell, 1(1):63–72, 2001.

Shulman JM, Benton R and St Johnston D. The Drosophila homolog of C. elegans PAR-1 organizes the oocyte cytoskeleton and directs oskar mRNA localization to the posterior pole. Cell, 101(4):377–388, 2000.

Simone RP and DiNardo S. Actomyosin contractility and Discs large contribute to junctional conversion in guiding cell alignment within the Drosophila embryonic epithelium. Development, 137(8):1385–1394, 2010.

Sluss HK, Han Z, Barrett T, Goberdhan DC, WILSON C, Davis RJ and Ip YT. A JNK signal transduction pathway that mediates morphogenesis and an immune response in Drosophila. Genes & development, 10(21):2745–2758, 1996.

Snel B, Lehmann G, Bork P and Huynen MA. STRING: a web-server to retrieve and display the repeatedly occurring neighbourhood of a gene. Nucleic acids research, 28(18):3442–3444, 2000.

Solon J, Kaya-Copur A, Colombelli J and Brunner D. Pulsed forces timed by a ratchet-like mechanism drive directed tissue movement during dorsal closure. Cell, 137(7):1331–1342, 2009. XXVI Bibliography

Speck O, Hughes SC, Noren NK, Kulikauskas RM and Fehon RG. Moesin functions antagonistically to the Rho pathway to maintain epithelial integrity. Nature, 421(6918):83–87, 2003.

Suzuki A, Yamanaka T, Hirose T, Manabe N, Mizuno K, Shimizu M, Akimoto K, Izumi Y, Ohnishi T and Ohno S. Atypical protein kinase C is involved in the evolutionarily conserved par protein complex and plays a critical role in establishing epithelia-speciﬁc junctional structures. J Cell Biol, 152(6):1183–1196, 2001.

Takeichi M. The cadherins: cell-cell adhesion molecules controlling animal morphogenesis. Development, 102(4):639–655, 1988.

Tanentzapf G and Tepass U. Interactions between the crumbs, lethal giant larvae and bazooka pathways in epithelial polarization. Nat Cell Biol, 5(1):46–52, 2003.

Tautz D. Whole Mount In Situ hybridization for the Detection of mRNA in Drosophila Embryos. In Nonradioactive Analysis of Biomolecules, pp. 573–580. Springer Berlin Heidelberg, Berlin, Heidelberg, 2000.

Tepass U. Crumbs, a component of the apical membrane, is required for zonula adherens formation in primary epithelia of Drosophila. Dev Biol, 177(1):217–225, 1996.

Tepass U. FERM proteins in animal morphogenesis. Current opinion in genetics & development, 19(4):357–367, 2009.

Tepass U. The apical polarity protein network in Drosophila epithelial cells: regulation of polarity, junctions, morphogenesis, cell growth, and survival. Annual Review of Cell and Developmental Biology, 28:655–685, 2012.

Tepass U, Gruszynski-DeFeo E, Haag TA, Omatyar L, Török T and Hartenstein V. shotgun encodes Drosophila E-cadherin and is preferentially required during cell rearrangement in the neurectoderm and other morphogenetically active epithelia. Genes & development, 10(6):672–685, 1996.

Tepass U and Hartenstein V. The development of cellular junctions in the Drosophila embryo. Dev Biol, 161(2):563–596, 1994.

Tepass U and Knust E. Crumbs and stardust act in a genetic pathway that controls the organization of epithelia in Drosophila melanogaster. Dev Biol, 159(1):311–326, 1993.

Tessier-Lavigne M and Goodman CS. The molecular biology of axon guidance. Science, 274(5290):1123–1133, 1996.

Towbin H, Staehelin T and Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci U S A, 76(9):4350–4354, 1979.

Toyofuku T, Yoshida J, Sugimoto T, Zhang H, Kumanogoh A, Hori M and Kikutani H. FARP2 triggers signals for Sema3A-mediated axonal repulsion. Nature neuroscience, 8(12):1712–1719, 2005.

Tsarouhas V, Senti KA, Jayaram SA, Tiklová K, Hemphälä J, Adler J and Samakovlis C. Sequential pulses of apical epithelial secretion and endocytosis drive airway maturation in Drosophila. Dev Cell, 13(2):214–225, 2007.

Tufro A, Teichman J, Woda C and Villegas G. Semaphorin3a inhibits ureteric bud branching morphogenesis. Mechanisms of development, 125(5-6):558–568, 2008. Bibliography XXVII

Turner FR and Mahowald AP. Scanning electron microscopy of Drosophila embryogenesis. 1. The structure of the egg envelopes and the formation of the cellular blastoderm. Dev Biol, 50(1):95–108, 1976.

Turunen O, Wahlström T and Vaheri A. Ezrin has a COOH-terminal actin-binding site that is conserved in the ezrin protein family. J Cell Biol, 126(6):1445–1453, 1994.

Tusnády GE and Simon I. Principles governing amino acid composition of integral membrane proteins: application to topology prediction. Journal of molecular biology, 283(2):489–506, 1998.

Tyler JM, Hargreaves WR and Branton D. Puriﬁcation of two spectrin-binding proteins: biochemical and electron microscopic evidence for site-speciﬁc reassociation between spectrin and bands 2.1 and 4.1. Proc Natl Acad Sci U S A, 76(10):5192–5196, 1979. van Gent DC, Hoeijmakers JH and Kanaar R. Chromosomal stability and the DNA double-stranded break connection. Nature Reviews Genetics, 2(3):196–206, 2001.

Venken KJT, Schulze KL, Haelterman NA, Pan H, He Y, Evans-Holm M, Carlson JW, Levis RW, Spradling AC, Hoskins RA and Bellen HJ. MiMIC: a highly versatile transposon insertion resource for engineering Drosophila melanogaster genes. Nature methods, 8(9):737–743, 2011.

Vincent S, Ruberte E, Grieder NC, Chen CK, Haerry T, Schuh R and Aﬀolter M. DPP controls tracheal cell migration along the dorsoventral body axis of the Drosophila embryo. Development, 124(14):2741–2750, 1997.

Wang J, Liu G, Li Q, Wang F, Xie F, Zhai R, Guo Y, Chen T, Zhang N, Ni W, Yuan H and Tai G. Mucin1 promotes the migration and invasion of hepatocellular carcinoma cells via JNK-mediated phosphorylation of Smad2 at the C-terminal and linker regions. Oncotarget, 6(22):19264–19278, 2015.

Wang Q, Hurd TW and Margolis B. Tight junction protein Par6 interacts with an evolutionarily conserved region in the amino terminus of PALS1/stardust. The Journal of biological chemistry, 279(29):30715–30721, 2004.

Wappner P, Gabay L and Shilo BZ. Interactions between the EGF receptor and DPP pathways establish distinct cell fates in the tracheal placodes. Development, 124(22):4707–4716, 1997.

Ward RE, Lamb RS and Fehon RG. A conserved functional domain of Drosophila coracle is required for localization at the septate junction and has membrane-organizing activity. J Cell Biol, 140(6):1463–1473, 1998.

Ward RE, Schweizer L, Lamb RS and Fehon RG. The protein 4.1, ezrin, radixin, moesin (FERM) domain of Drosophila Coracle, a cytoplasmic component of the septate junction, provides functions essential for embryonic development and imaginal cell proliferation. Genetics, 159(1):219–228, 2001.

Wei Z, Li Y, Ye F and Zhang M. Structural basis for the phosphorylation-regulated interaction between the cytoplasmic tail of cell polarity protein Crumbs and the actin binding protein Moesin. Journal of Biological Chemistry, p. jbc.M115.643791, 2015. XXVIII Bibliography

Wells KL, Rowneki M and Killian DJ. A splice acceptor mutation in C. elegans daf-19/Rfx disrupts functional specialization of male-speciﬁc ciliated neurons but does not aﬀect ciliogenesis. Gene, 559(2):196–202, 2015.

Westerlund N, Zdrojewska J, Padzik A, Komulainen E, Björkblom B, Rannikko E, Tararuk T, Garcia-Frigola C, Sandholm J, Nguyen L, Kallunki T, Courtney MJ and Coﬀey ET. Phosphorylation of SCG10/stathmin-2 determines multipolar stage exit and neuronal migration rate. Nature neuroscience, 14(3):305–313, 2011.

Winkler S, Schwabedissen A, Backasch D, Bökel C, Seidel C, Bönisch S, Fürthauer M, Kuhrs A, Cobreros L, Brand M and González-Gaitán M. Target-selected mutant screen by TILLING in Drosophila. Genome research, 15(5):718–723, 2005.

Wodarz A, Ramrath A, Grimm A and Knust E. Drosophila atypical protein kinase C associates with Bazooka and controls polarity of epithelia and neuroblasts. J Cell Biol, 150(6):1361–1374, 2000.

Wu VM, Schulte J, Hirschi A, Tepass U and Beitel GJ. Sinuous is a Drosophila claudin required for septate junction organization and epithelial tube size control. J Cell Biol, 164(2):313–323, 2004.

Yamada S, Pokutta S, Drees F, Weis WI and Nelson WJ. Deconstructing the cadherin-catenin-actin complex. Cell, 123(5):889–901, 2005.

Yamauchi J, Miyamoto Y, Kokubu H, Nishii H, Okamoto M, Sugawara Y, Hirasawa A, Tsujimoto G and Itoh H. Endothelin suppresses cell migration via the JNK signaling pathway in a manner dependent upon Src kinase, Rac1, and Cdc42. FEBS letters, 527(1-3):284–288, 2002.

Yazdani U and Terman JR. The semaphorins. Genome Biology, 7(3):211, 2006.

Young PE, Richman AM, Ketchum AS and Kiehart DP. Morphogenesis in Drosophila requires nonmuscle myosin heavy chain function. Genes & development, 7(1):29–41, 1993.

Zhang M, Sun L, Wang X, Chen S, Kong Y, Liu N, Chen Y, Jia Q, Zhang L and Zhang L. Activin B promotes BMSC-mediated cutaneous wound healing by regulating cell migration via the JNK-ERK signaling pathway. Cell transplantation, 23(9):1061–1073, 2014.

Zhuang B, Su YS and Sockanathan S. FARP1 promotes the dendritic growth of spinal motor neuron subtypes through transmembrane Semaphorin6A and PlexinA4 signaling. Neuron, 61(3):359–372, 2009. Acknowledgements

I would like to sincerely thank Elisabeth Knust for allowing me to work in her lab and providing me with all the help and resources I needed. I am especially grateful for her kind and patient manner, and for being so understanding of the myriad of diﬀerent ways a scientist’s life can go.

I need to thank David Flores-Benitez for teaching me many facts and techniques and for being readily available to answer questions. Also for pushing me far enough to realize that I need to know my own potential, and not let my skills and conﬁdence be questioned and compromised by others.

Thank you to the Dresden International PhD Program and the Max Planck Institute of Molecular Cell Biology and Genetics for creating such a competeitive and challenging atmosphere.

A big Thank You goes to the Sequencing Facility, the Light Microscopy Facility and the Antibody Facility for providing me with their expertise and excellent services. The same goes for the fly keepers Sven, Conny and Ruth and Stefan and Katja in the fly kitchen. Another big Thank You to Franziska and Kostas for assisting me numerous times, especially with my demanding needs during four Long Nights of Science. Kostas, thank you on top of that for filming even though you were really busy! Thank you, Florian, for creating SO many opportunities to quench my thirst for being in the limelight. If you weren’t so driven and resourceful, I would never have been so sure where I belong. Thank you, Marko, Linda, Ali, Katja, Satu, Anna, Srija and Cátia for helpful discussions and just being great colleagues! Shradha, thank you for doing all those awesome things with me and for being such a funny, smart, open-minded person! It was a true pleasure getting to know you.

I especially want to thank my family and friends for being there for me during the years, you are the ones that actually helped me stick with it: To my mom for everything she has ever done, for always believing in me and being on my side. For driving hundreds of kilometers to be at whatever thing or show I was involved in and for providing me with all the things I ever needed to get where wanted to go. Or at least thought I wanted to go. To Frenne for still being there, through all changes and ups and downs. Most of all, for bringing out the best in me when I didn’t know there was anything good there. Mark for being the perfect significant other. For being so infinitely sweet and patient and kind. For being just as happy with me as I am with you! Алёнка, for coming out of the blue and instantly becoming one of the best friends I ever had. And for teaching me something very relevant: Осторожно, какашки! Anneke for sometimes gently nudging, sometimes downright shoving me out of my comfort zone. I’m getting more and more Dutch every day! Also, awesome job inviting your two friends over for a weekend in December 2011. And Irena, for being such a loving and patient friend, always welcoming, always comforting, always there to help. I am SO grateful I met the two of you and I hope we’ll still be friends a long time from now. And Sven, for being a wonderful desk neighbor, for encouraging me to open my eyes and my mind to all the possibilities and sometimes uncomfortable truths out there. I consider myself lucky for knowing you! I also want to cordially thank the Nematology lab at Wageningen UR, especially Jaap, Joost, Rikus, Casper, Koen, Paula, Ocavina, Lizeth and Savio for welcoming me so openly and instantly including me in their circles (That outing honestly was one of the most fun days I ever had)! The same is true for Francien and Niek. Thanks to Bas Zwaan as well, for allowing me access to the fly facilities, and to Jelle and Klaas for spontanteously giving me the opportunity to expand my network. Erklärung entsprechend §5.5 der Promotionsordnung

Hiermit versichere ich, dass ich die vorliegende Arbeit ohne unzulässige Hilfe Dritter und ohne Benutzung anderer als der angegebenen Hilfsmittel angefertigt habe; die aus fremden Quellen direkt oder indirekt übernommenen Gedanken sind als solche kenntlich gemacht. Die Arbeit wurde bisher weder im Inland noch im Ausland in gleicher oder ähnlicher Form einer anderen Prüfungsbehörde vorgelegt. Die Dissertation wurde im Zeitraum vom 01. April 2012 bis 30. Oktober 2015 verfasst und von Prof. Dr. Elisabeth Knust, MPI-CBG Dresden betreut. Meine Person betreﬀend erkläre ich hiermit, dass keine früheren erfolglosen Promotionsverfahren stattgefunden haben. Ich erkenne die Promotionsordnung der Fakultät für Mathematik und Naturwissenschaften, Technische Universität Dresden an.

Datum, Unterschrift