Organisms respond to changes in their environment affecting their physiological or

ecological optimum by reactions called stress responses. These stress responses may

enable the organism to survive by counteracting the consequences of the environ-

mental change, the stressor, and usually consist of plastic alterations of traits related

to physiology, behaviour, or morphology. In the ecological model species Daphnia,

the waterflea, stressors like predators or parasites are known to have an important

role in adaptive evolution and have been therefore studied in great detail. However,

although various aspects of stress responses in Daphnia have been analysed, molecu-

lar mechanisms underlying these traits are not well understood so far. For studying

unknown molecular mechanisms, untargeted ‘omics’ approaches are especially suit-

able, as they may identify undescribed key players and processes.

Recently, ‘omics’ approaches became available for Daphnia. Daphnia is a cosmo-

politan distributed fresh water crustacean and has been in research focus for a long

time because of its central role in the limnic food web. Furthermore, the responses of

this organism to a variety of stressors have been intensively studied e.g. to hypoxic

conditions, temperature changes, ecotoxicological relevant substances, parasites or

predation. Of these environmental factors, especially predation and interactions with

parasites have gained much attention, as both are known to have great influence on

the structure of Daphnia populations.

In the work presented in this thesis, I characterised the stress responses of Daphnia

using proteomic approaches. Proteomics is particularly well suited to analyse bio-

logical systems, as proteins are the main effector of nearly all biological processes.

However, performing Daphnia proteomics is a challenging task due to high proteolytic activity in the samples, which most probably originate from proteases located

in the gut of Daphnia, and are not inhibited by proteomics standard sample pre-

paration protocols. Therefore, before performing successful proteomic approaches,

I had to optimise the sample preparation step to inhibit proteolytic activity in Daph-

nia samples. After succeeding with this task, I was able to analyse stress responses of

Daphnia to well-studied stressors like predation and parasites. Furthermore, I stud-

ied their response to microgravity exposure, a stressor not well analysed in Daphnia

so far.

My work on proteins involved in predator-induced phenotypic plasticity is de-

scribed in chapter 2 and 3. Daphnia is a textbook example for this phenomenon and is

known to show a multitude of inducible defences. For my analysis, I used the system

of Daphnia magna and its predator Triops cancriformis. D. magna is known to change its

morphology and to increase the stability of its carapace when exposed to the pred-

ator, which has been shown to serve as an efficient protection against T. cancriformis

predation. In chapter 2, I used a proteomic approach to study predator-induced traits

in late-stage D. magna embryos. D. magna neonates are known to be defended against

Triops immediately after the release from the brood pouch, if mothers were exposed

to the predator. Therefore, the formation of the defensive traits most probably oc-

curs during embryonic development. Furthermore, embryos should have reduced

protease abundances, as they do not feed inside the brood pouch until release. To

study proteins differing in abundance between D. magna exposed to the predator

and a control group, I applied a proteomic 2D-DIGE approach, which is a gel based

method and therefore enables visual monitoring of protein sample quality. I found

differences in traits directly associated with known defences like cuticle proteins and

chitin-modifying enzymes most probably involved in carapace stability. In addition,

enzymes of the energy metabolism and the yolk protein vitellogenin indicat

Eye movements are important to aid vision, and they serve two main functions: to stabilize a moving visual target on the retina and to stabilize gaze during own body movements. Six types of eye movements have been evolved fulfilling this function: saccades, smooth pursuit, vestibulo-ocular reflex, optokinetic response, convergence and gaze holding. In all vertebrates the eyes are moved by six pairs of extraocular muscles that enable horizontal, vertical and rotatory eye movements. The motoneurons of these muscles are located in the oculomotor (nIII), trochlear (nIV) and abducens (nVI) nucleus in the brainstem. Motoneurons of the lateral rectus muscle (LR) in nVI and of the medial rectus muscle (MR) in nIII provide horizontal eye movements, those of inferior oblique (IO) and superior rectus muscle (SR) in nIII upward eye movements. Motoneurons of the superior oblique (SO) and the inferior rectus muscle (IR) in nIII convey downward eye movements. Recently, it was shown that each extraocular muscle is controlled by two motoneuronal groups:

1. Motoneurons of singly innervated muscle fibers (SIF) that lie within the boundaries of motonuclei providing a fast muscle contraction (twitch) and 2. motoneurons of multiply innervated muscle fibers (MIF) in the periphery of motonuclei providing a tonic muscle contraction (non-twitch). Tract-tracing studies indicate that both motoneuronal groups receive premotor inputs from different brainstem areas. A current hypothesis suggests that pathways controlling twitch motoneurons serve to generate eye movements, whereas the non-twitch system is involved in gaze holding. Lesions of inputs to the twitch motoneuron system may lead to supranuclear gaze palsies, whereas impairment of the non-twitch motoneuron system may result in gaze holding deficits, like nystagmus, or strabismus. Up to date only limited data are available about the histochemical characteristics including transmitters to the SIF- (twitch) and MIF (non-twitch) motoneurons.

The present study was undertaken to investigate the histochemical profile of inputs to motoneuronal groups of individual eye muscles mediating horizontal and vertical eye movements including the inputs to MIF- and SIF motoneurons. The MIF motoneurons of the IR and MR are located in the periphery dorsolateral to nIII, close to the Edinger-Westphal nucleus (EW), which is known to contain preganglionic cholinergic neurons. Other scientists have found that the EW is composed of urocortin-positive neurons involved in food intake or stress. In order to delineate these different cell populations within the supraoculomotor area dorsal to nIII, a comparative study in different mammals was conducted to locate the cholinergic preganglionic neurons and urocortin-positive neurons. Only then, it became obvious that the cytoarchitecturally defined EW labels different cell populations in different species. In rat, ferret and human the cytoarchitecturally defined EW is composed of urocortin-positive neurons. Only in monkey the EW contains cholinergic preganglionic neurons, which lie close to the MIF-motoneurons of MR and IR in the C-group.

In monkey, I performed a systematic study on the histochemical profile and transmitter inputs to the different motoneuron subgroups, including MIF- and SIF motoneurons. Brainstem sections containing prelabelled motoneurons were immunostained for the calcium-binding protein calretinin (CR), gamma-aminobutyric acid (GABA) or glutamate decarboxylase (GAD), glycine transporter 2, glycine receptor 1, and the vesicular glutamate transporters (vGlut) 1 and 2.

The study on the histochemical profile of the motoneuron inputs revealed three main results: 1.The inhibitory control of SIF motoneurons for horizontal and vertical eye movements differs. Unlike previous studies in the primate a considerable GABAergic input was found to all SIF motoneuronal groups, but a glycinergic input was confined to motoneurons of the MR mediating horizontal eye movements

Desmosomen sind spezialisierte Haftstrukturen, die die Stabilisierung des Zellverbundes gegenüber Zug- und Scherkräften gewährleisten. Dazu binden desmosomale Cadherine extrazellulär an Haftmoleküle benachbarter Zellen und sind intrazellulär unter anderem über Desmoplakin (DP) und Plakoglobin (PG) an Keratinfilamenten verankert. Insbesondere für das desmosomale Cadherin Desmoglein 3 (Dsg3), das sowohl innerhalb als auch außerhalb der Desmosomen vorkommt, wurde eine wichtige Bedeutung als Adhäsionsprotein in Keratinozyten nachgewiesen. Trotz ihrer Funktion, Widerstand gegen hohe mechanische Belastungen zu vermitteln, sind Desmosomen dynamische Strukturen, die einem stetigen Umbau unterliegen. Die Notwendigkeit einer genauen Regulierung des desmosomalen Auf- und Abbaus wird durch das Vorkommen zahlreicher vererbbarer und autoimmuner Erkrankungen unterstrichen. In der vorliegenden Arbeit wurden Mechanismen, die der geordneten Assemblierung der Desmosomen und der Disassemblierung nach Störung der desmosomalen Zell-Zell-Haftung unterliegen, untersucht.

Im ersten Teil der vorliegenden Studien standen die Vorgänge der Desmosomenbildung in humanen Keratinozyten im Fokus. Adhärenskontakte und deren Zusammenwirken mit Actinfilamenten spielen eine wichtige Rolle in der Ausbildung der Desmosomen. Für die Actin-Bindeproteine Adducin und Cortactin wurde durch siRNA-Interferenzstudien eine essentielle Funktion für die Vermittlung der desmosomalen Zell-Zell-Haftung nachgewiesen. Die siRNA-induzierte Depletion von Adducin verursachte eine Reduktion der zytoskelettal-gebundenen Dsg3-Moleküle, was mit einer reduzierten Membranmobiltät korrelierte. Für Cortactin wurde eine direkte Interaktion mit Dsg3 mittels zweier unabhängiger molekularbiologischer Methoden nachgewiesen. Dies deutet auf eine direkte Rolle des Cortactins in der Regulierung der Desmosomen hin. Die siRNA-induzierte Depletion von E-Cadherin führte zum Verlust der membranständigen Lokalisation von Dsg3 und zu einer verminderten Verankerung der Dsg3-Moleküle innerhalb der zytoskelettalen Proteinfraktion. Es wurde ein Signalkomplex aus extradesmosomalen Dsg3, E-Cadherin und der Tyrosinkinase Src identifiziert, dessen Stabilität durch Src reguliert wurde. Hierbei wurden Dsg3 und E-Cadherin an Tyrosinresten durch Src phosphoryliert, deren Aktivität sowohl für die Inkorporation von Dsg3 in die Desmosomen als auch für die Reifung der Desmosomen zu stabilen Haftkontakten essentiell war.

Im zweiten Teil der vorliegenden Arbeit wurden die Prozesse der desmosomalen Disassemblierung nach Inkubation mit Pemphigus vulgaris-Autoantikörpern (PV-IgG) analysiert.PV ist eine etablierte Modellerkrankung zur Untersuchung der Desmosomen-vermittelten Zelladhäsion in Keratinozyten. Die Bindung der gegen Dsg1 und Dsg3 gerichteten PV-IgGs induziert eine Reduktion der Dsg3-Proteinmengen und eine Aktivierung verschiedener Signalwege, u.a. von RhoA und PKC. Da diese Signalwege ebenfalls Adducin regulieren und PV-IgGs eine Umorganisierung des Actin-Zytoskeletts verursachen, die durch exogene Aktivierung von RhoA verhindert wird, wurde das Zusammenspiel von PV-IgGs, RhoA und Adducin untersucht. Die protektive Wirkung der RhoA-Aktivierung auf die Zell-Zell-Haftung und die Verteilung von Dsg3 nach Applikation der PV-IgGs war sowohl von der Expression als auch von der Phosphorylierung von Adducin an Serin726 abhängig. Interessanterweise verursachten PV-IgGs über den Ca2+-Einstrom und über PKC, unabhangig von RhoA, eine schnelle Phosphorylierung von Adducin an Serin726. Die durch den Ca2+-Einstrom- und PKC-vermittelte Phosphorylierung von Adducin könnte somit einen Rettungsmechanismus der Keratinozyten darstellen, der in Reaktion auf die PV-IgG-Bindung einsetzt und die desmosomale Assemblierung induziert. Ferner wurde die reduzierte Verankerung der Keratinfilamente an Desmosomen, ein weiteres Merkmal der PV-Pathogenese, mit der Aktivität von PKC korreliert. Keratinfilamente, die einer dynamischen Regulierung durch

Childhood abuse is one of the major risk factors for the development of adult

psychopathology though the response to childhood abuse and other types of early

life adversities is not uniform. Genetic predisposition modulates the exposure to

environmental factors in form of gene by environment interaction. This has been

shown for FKBP5, a modulator of the stress hormone axis, with certain alleles in

FKBP5 conferring a higher risk towards PTSD in adulthood in response to childhood

abuse. This thesis investigates the potential molecular mechanism behind this gene

by environment interaction and delineates an allele-specific demethylation

mechanism in response to childhood abuse. In addition, data on genome-wide gene

expression and DNA methylation profiles in peripheral blood in response to

childhood abuse is presented providing evidence for the hypothesis that childhood

trauma leads to a different molecular trajectory towards adult psychopathology

compared to adult traumatization. The data presented here contribute to our

understanding of the molecular mechanisms underlying gene by environment

interactions in psychiatry and the pathophysiology of trauma- and stress-induced

psychiatric disorders.

Super-resolution fluorescence microscopy performed via 3D structured illumination microscopy (3D-SIM) features an 8-fold volumetric resolution improvement over conventional microscopy and is well established on flat, adherent cells. However, blastomeres in mammalian embryos are non-adherent, round and large. Scanning whole mount mammalian embryos with 3D-SIM is prone to failure due to non-adherent embryos moving during scanning and a large distance to the cover glass. The biggest challenge and achievement of this doctorate thesis was the development of a novel method to perform 3D-SIM on mammalian embryos (“3D structured illumination microscopy of mammalian embryos and spermatozoa” published in BMC Developmental Biology). The development and fine-tuning of this method took over two years due to the time-intense generation of embryos and the subsequent two day long embryo staining, embedding and scanning with steps that required novel techniques such as micromanipulation which was not associated with sample preparation prior to this protocol. Problem identification was time-intensive since each of the numerous steps necessary could negatively affect the image quality. This method was fine-tuned during three studies. The first study “Reprogramming of fibroblast nuclei in cloned bovine embryos involves major structural remodeling with both striking similarities and differences to nuclear phenotypes of in vitro fertilized embryos” (published in Nucleus) investigates the profound changes of nuclear architecture during cattle preimplantation development of embryos generated by somatic cell nuclear transfer (SCNT) and in vitro fertilization (IVF). Fibroblast nuclei in embryos generated by SCNT go through similar changes in nuclear architecture as embryos generated by IVF. In both embryo types the occurrence of a large, chromatin-free lacuna in the center of nuclei around major embryonic genome activation (EGA) was noted. Similarly, the chromosome territory-interchromatin compartment (CT-IC) model applied to both types of embryos, featuring a lacuna or not, with an enrichment of RNA polymerase II and H3K4me3, a histone modification for transcriptionally competent chromatin, in less concentrated chromatin and an enrichment of H3K9me3, a transcriptionally restrictive histone modification, in more concentrated chromatin. However, large, highly concentrated H3K4me3 and H3K9me3 clusters were noted in both embryo types at chromatin concentrations that did not fit to the model. The chromatin-free lacunas were highly enriched in newly synthesized mRNA.

The second study “Remodeling of the Nuclear Envelope and Lamina during Bovine Preimplantation Development and Its Functional Implications” (published in PLOS ONE) presents the changes of the nuclear envelope and lamina during bovine preimplantation development. Before major EGA, chromatin-free areas of the nuclear periphery were also free of nuclear pore complexes (NPCs), whereas after major EGA, the entire nuclear periphery was equipped with at least a fine layer of chromatin and associated NPCs. Three types of nuclear invaginations were predominant at different stages. The most common invagination was lamin B and NUP153 positive and was most prominent between the 2-cell and 8-cell stages until the onset of major EGA. Lamin B positive, but NUP153 negative invaginations were most prominent during stages with large nuclear volume and surface reductions. The least common invagination was lamin B negative but NUP153 positive and occurred almost exclusively at the morula stage. RNA-Seq and 3D-SIM data showed large deposits of spliced NUP153 mRNA and cytoplasmic NUP153 protein clusters until shortly after major EGA. NUP153 association with chromatin was initiated at metaphase.

The third study “Stage-dependent remodeling of the nuclear envelope and lamina during rabbit early embryonic development” (published in the Journal of Reproduction and Development) demonstrated that rabbit embryonic nuclei fe