The endodermis is a conserved root cell layer in all higher plants featuring rings of hydrophobic cell wall material, called the Casparian strips (CS). The CS is thought to be crucial for controlling soil-to-vascular translocation of nutrients by limiting diffusion of solutes via the cell wall space (apoplastic pathway), while allowing selective cellular uptake to occur. Since their discovery more than 100 years ago, the CS have received a lot of attention, and such is their importance in current models of root function, that every biology student has to learn about them in basic botany. This textbook knowledge, however, is mainly based on correlative evidence and controversies about the exact role of the CS abound, mainly because specific mutants of endodermal function have been lacking. N. Geldner has obtained first specific mutants in the endodermal barrier and will characterize and engineer more mutants for this project. J. Takano will characterize these CS mutants using latest technologies for quantification, transport and cellular localization of nutrients. These technologies will provide unmatched temporal and spatial resolution. This resolution is necessary for the very small root systems of the Arabidopsis model used. The aim is to discriminate nutrient concentrations inside and outside the endodermal barrier, as well as to gain ideas about the kinetics, and not only steady-state, of nutrient uptake. This combination of competences and technologies will allow, for the first time, an unambiguous demonstration of the importance of this evolutionarily ancient root diffusion barrier for controlled nutrient uptake. Moreover, it will tell us which classes of minerals depend especially on an intact endodermal barrier and the uptake pathways they employ. The proposed work could provide a new framework for root nutrient uptake and profoundly change our current models of root function.

Gamma-aminobutyric acid (GABA) is the major inhibitory neurotransmitter in the vertebrate brain and dysfunction of GABAergic neurons can have profound pathological implications. I propose here to study different aspects of GABAergic neuron development and function in the model organism Caenorhabditis elegans. In C. elegans, 26 neurons express conserved GABAergic terminal differentiation markers. 25 of these are motoneurons and only one neuron, named RIS, is an interneuron which most closely resembles the dominant type of GABA neuron in vertebrates. Preliminary evidence indicates that lim-6 - a LIM homeobox transcription factor - is partly involved in RIS development, and loss of lim-6 suggests that RIS might be involved in the control of metabolism and life span. First, I will delineate cis-regulatory elements necessary for the expression of known RIS terminal differentiation markers using mutational analysis of gfp reporter genes. Second, I propose to identify the factor(s) that control RIS differentiation in association with LIM-6. I will undertake a genetic screen to identify trans-acting factors necessary for the expression of known RIS terminal differentiation markers in a lim-6 mutant background. Third, I will use transgenic animals in which RIS is genetically eliminated to determine RIS function. The proposed project will give a detailed picture of GABA neuron development and function using the C. elegans model system.

Morphologically diverse networks of microtubules (MTs) underlie the regulation of key cellular processes such as chromosome segregation, division, morphogenesis, and motility of eukaryotic cells. In large part, the necessary flexibility in network architecture is made possible by the spatial and temporal regulation of sites from which MT growth is initiated. Although centrosomes are prominent sources of MTs, for many networks spatially dispersed non-centrosomal MT nucleation pathways are essential. Understanding how order arises out of an ensemble of non-centrosomal MTs requires the simultaneous investigation of nucleation and sorting mechanisms. Here, we will focus on the emergence of bipolarity in non-centrosomal spindles involved in chromosome segregation in mitotic and meiotic cells. Our aim is to understand at a mechanistic level how systems evolve to having two poles and how the distribution of MT lengths sets the size of the forming spindle. To this end we identify five conserved MT organizing activities that will be studied in an integrated manner. We propose to change system parameters and quantify the induced response using three complementary approaches. The Goshima lab will study bipolarity in the mitotic spindle of the moss Physcomitrella patens, a system that lacks centrosomal nucleation altogether. The Janson lab will reconstitute MT networks from selected purified components whilst controlling MT nucleation spatially. The Kruse group will develop physical descriptions of MT networks explicitly accounting for MT end dynamics. The integration of the three lines of research will be accomplished by quantitative comparison of protein distributions in spindles and by assessment of system responses to perturbations. The combined studies will allow us to characterize fundamental pathways generating bipolar MT arrangements. Beyond spindles, we expect that these mechanisms will significantly deepen our understanding of the vast variety of non-centrosomal MT network architectures observed throughout biology.

The main focus of the proposed project is the role of stromal fibroblasts-DICER1/microRNAs in cancer growth and metastasis .This will be accomplished by using elegant in vivo experiments with the generation of knockout mice as a model.
Increasing evidence suggests that cancer development is facilitated by continuing interaction between tumor cells and activated stromal cells indicating that fibroblasts are critical for cancer progression. However the importance of stromal fibroblast-specific DICER1 and microRNAs in the crosstalk between tumor and stroma has not been specifically addressed. We will address this significant topic of the role of DICER1, a key protein in the biogenesis pathway of microRNAs, in the complex interactions between tumor and stroma by using innovative in vivo experiments through conditional deletion of DICER1 in different lineages of tumor associated fibroblasts and mesenchymal cells. These mice will be back crossed with MMTV-pyMT mice, a reliable mousse model that develops metastatic mammary cancer. Fibroblasts-specific Dicer1 knockout mice (FSP-1-Cre+/Dicer1flox/flox and α-SMA-Cre+/Dicer1flox/flox) will be studied for phenotype and their effect in cancer progression.
These mice will also allows us for the first time the study of the possible existence of microRNAs and/or miRNA machinery proteins working as paracrine signals to the tumor. The characterization of important roles of microRNAs and microRNA machinery proteins in the crosstalk between stromal and tumor cells that could possibly be influencing tumor’s proliferation and/or invasion capacity, will gives us a new scenario to account for in cancer therapeutics.

Biological systems are composed of different components interacting in a complex manner. Evolution has shaped their organization and molecular reality. Proteins exhibit a remarkable evolvability, but a detailed understanding of why a protein performs a certain task better than other remains elusive. I plan to develop evoMAGE, a platform to evolve components to carry out different biological functions.
MAGE, a technique to massively modify the genome will be coupled with different selection systems to create a controllable implementation of evolution. The proposed platform will be presented in a device which integrates controlled diversity generation, expression, and selection in multiple cycles with a high throughput and effectiveness. Different components will be evolved and co-evolved to interact with the
different layers of the HIV machinery -from the genome to the end products- permitting to study the evolutionary process induced.
With evoMAGE multiple paths of evolution can be simulated and monitored as well as their interdependences. Besides creating a tool to generate new protein-protein and protein-DNA interactions, observing how proteins change along with their host systems will greatly contribute our understanding of living beings architecture.

Bacterial restriction-modification systems (RMs) provide a tractable, experimental model for the evolutionary dynamics of multi-level conflicts, self-nonself recognition in immunity, and host-parasite coevolution. Focusing on Type II RMs, which are minimally composed of two genes – a restriction endonuclease gene, R, and a methyltransferase, M – this proposal studies the dynamics of these sophisticated mobile genetic elements, both in single-cells and in evolving populations. The evolutionary function of RMs in bacterial populations is a matter of debate: while RMs can provide protection against bacteriophage infection, evolution experiments revealed that alternative pathways of resistance are typically employed by bacteria. RMs also behave as selfish elements, which kill host cells when RM systems are lost, resulting in RM persistence in populations. A multi-level conflict, between the best interests of the gene, the cell, and the population, thus exists in this simple genetic system, and will be studied quantitatively. Being at the interface between bacterial host and bacteriophage parasite, RMs (which themselves can be viewed as parasitic) are also a key ecological player in a host-parasite coevolution.

This proposal applies a combination of single-cell microscopy, microfluidics, optical traps, and modeling to elucidate the selective pressures responsible for the evolution and maintenance of RM systems in bacteria. Inducible, fluorescent-tagged constructs of R and M genes will be introduced into E.coli, and their expression dynamics will be measured and modeled. Simulations of populations of cells containing RM systems will be performed, using fitness cost functions obtained in single-cell measurements, and then used to predict and interpret the results of laboratory evolution. Modeling will be used to bridge individual-level measurements and population-level dynamics observed in the lab, with the goal of elucidating the selective pressures that are encountered by cells in conflict with resident RM systems. To assess more broadly these pressures, experiments to identify host-beneficial functions of RM systems will be undertaken. Bacteria with and without specific RM systems will be evolved under various environmental stresses.

The combination of approaches proposed here will be used in an integrative manner to bring these sophisticated genetic systems to the forefront of evolutionary dynamics research, and to provide a deeper understanding of the larger biological questions, outlined above, that RM systems are uniquely positioned to answer.

Most proteins that are to be integrated into membranes or secreted from the cell are threaded through a conserved, membrane-embedded conduit – the Sec61 complex in the eukaryotic endoplasmic reticulum (ER) or the SecY complex in the prokaryotic plasma membrane. This so-called "translocon" forms a gated channel across the membrane ("vertical gating"), but can also open laterally to release hydrophobic segments of membrane proteins into the lipid phase ("lateral gating"). Targeting of proteins to the translocon is determined by divergent cleavable signal sequences or noncleavable trans-membrane segments. Priming of the translocation channel for cargo binding requires its prior interaction with additional ligands, such as the ribosome during co-translational translocation of a nascent polypeptide chain. The structural details of cargo recognition by the translocon are elusive and the mechanisms of vertical and lateral gating are only poorly understood. I propose to address these problems by combining biochemistry, X-ray crystallography and cryo-electron microscopy. I will establish systems to reconstitute and purify stable intermediates of co-translational protein translocation, which are suited for functional and structural analyses. The main objective of this study is to determine the crystal structure of a translocon bound to a ribosome-nascent chain complex. This will reveal how the translocon can specifically recognize a wide range of different signal sequences. The project will further provide mechanistic insight into ligand- and cargo-dependent gating of the channel, as well as into general principles of molecular recognition and membrane protein dynamics.

The birth of new genes offers an enormous evolutionary potential by providing substrate for natural selection and drift to work on and promises the emergence of novel functions. Recent studies uncovered an astonishing array of mechanisms by which new genes are formed and provided insights into several fascinating cases of adaptive new gene functions. Nevertheless, the repertoires of new functional genes in mammals remain poorly characterized. In my project, I want to examine processes involved in new gene birth by conducting comprehensive comparative analyses across all major mammalian lineages. Specifically, I want to study the processes by which new genes arise, the frequency and timing of these events in different mammals and the contribution of new genes to phenotypic evolution. I will mainly focus on so far little studied processes of gene origination such as de novo origination, genomic parasite domestication and gene fusion. To this end, I will use publically available genomic resources as well as the unique collection of tissue and lineage-specific transcriptome data generated in the host lab, which will allow to evaluate the special role of particular tissues in the process of new gene origination and to functionally characterize new genes. Bioinformatics procedures will be used to conduct comparative evolutionary studies of new gene emergence in mammals on a so far unprecedented scale and in great detail. I will attempt initial experimental characterization of several promising candidate genes. The proposed project has the potential to uncover general rules of new gene birth across mammals and highlight cases of particular evolutionary interest and importance.

Cancer is a dynamic ecological and evolutionary community of cells where mutations in genes give rise to new clones of cells that either survive or become extinct. We propose that cancers can evolve by cooperation, whereby two or more mutations occur in separate cell clones, which can then cooperate to produce an emergent malignant phenotype. This contrasts with the conventional theory, in which cancer-causing mutations accumulate in one clone that has every property needed to be malignant. However, often more than one clone persist in cancers. If more than one clone persist, why should only one be active? And if more than one are active, how do they interact? Cooperation is possible, but does it occur? We will test this new idea of how cancers evolve using an integrated interdisciplinary approach of genetic analysis, biological assays and mathematical modeling. We will genotype microscopic parts of human tumours and use innovative mouse models to determine how often tumours arise from two or more clones. At the same time, we will use a mathematical multicellular simulation model of cancer to explore the biological consequences of clonal cooperation and to guide interpretation of the data obtained from the laboratory studies. The outcome will be a new genetic paradigm for cancer that could revolutionise cancer research and therapy.

Membrane protein structure determination is in most cases performed in phospholipid or detergent micelles, which is far away from physiological conditions. Therefore the use of phospholipid bilayers in so called nanodiscs with a stable planar lipid bilayer is more suited for studying membrane protein structure in a native environment. This membrane mimic will be used for the structure determination of the VDAC-1 ion channel using NMR spectroscopy and for studying the complex formation with pro- and anti-apoptotic Bcl-2 proteins. Formation of these complexes relates to apoptosis but the detailed role has remained unclear. Furthermore, by using cell-free protein expression and subsequent reconstitution in phospholipid nanodiscs, the structure and ligand binding of G-protein coupled receptors will be investigated. For sample validation, a fluorescence-based thermal stability assay will be optimized for membrane proteins and combined with light scattering and electron microscopy. Another system of interest will be the T-cell receptor (TCR) / CD3 system. There is no structural information on the topology of TCR/CD3 available so far. By using membrane-anchored T-cell receptor proteins it should be possible to measure interactions between the individual components. This wide range of topics ranging from membrane biology to immunology and apoptosis will give me the opportunity to complement my NMR knowledge by other biophysical techniques like electron microscopy but more importantly give me the opportunity to learn new techniques like cell-free protein expression, membrane biochemistry, Sortase-mediated protein ligation and cell-based assays for functional characterization.

Cyclic nucleotide-gated channels (CNG channels) are part of the cyclic nucleotide-regulated family of channels. These are nonselective cation channels activated by the binding of cyclic nucleotides (cAMP and cGMP) to a cyclic-nucleotide binding domain (CNBD). Native CNG channels are heterotetramers of CNGA and CNGB subunits. Heterotetrameric rod or cone CNG channels has been proposed to be composed from the unique stochiometry of 3:1 CNGA:CNGB. The olfactory channel has been proposed to be 2:1:1 CNGA2:CNGA4:CNGB1b. This heteromeric channel stoichiometry was surprising since it represents a breakdown in symmetry within the channel tetramer. The distal, post-CNBD region of CNGA subunits was shown to contain a leucine zipper domain (CLZ), which trimerized in solution. However, the mechanism of this exclusive stoichiometry is largely unknown. Utilizing an improved single molecule subunit counting method, I will directly measure the stoichiometry of intact CNG channels and determine the three dimensional structure of the CLZ domain. Structural based mutations will be further used in functional and biochemical assays, to unveil the molecular mechanism of heteromeric subunit assembly. This study will provide with a deeper molecular insight of the unique asymmetric stoichiometry of CNG channels and will pave the road towards global structural perception of cyclic nucleotide gated channels.

The formation of mature neural circuits requires the selective pruning of inappropriate axonal branches. Glial cells have a central role in the axon pruning events by engulfing axonal debris, but the molecular mechanisms underlying axon pruning are largely unknown. The engulfment is shown to be mediated through the proteins in glial cells, which were previously implicated in the engulfment of apoptotic cells. Apoptotic cells express phosphatidylserine (PS) on their surface as an “eat me” signal to macrophages, and macrophages engulf the cells by recognizing the PS. Our preliminary experiments indicate the degenerating axons also express PS on their surface. By blocking PS with MFG-E8/D89E, that has a very high affinity for PS, we propose to examine whether PS exposure on degenerating axons triggers the engulfment by glial cells. By generating MFG-E8/D89E transgenic mice and examining several regions of axon pruning events, we will confirm that glial cells engulf degenerating axons by recognizing PS in vivo. Finally, our goal is to identify the molecular mechanisms of PS exposure by degenerating axons and the mechanisms of PS recognition by glial cells. We recently identified a gene family that promotes PS exposure. We will examine whether one of them mediates PS exposure in degenerating axons. To identify a novel receptor for PS on glial cells, we have established a high-throughput in vitro screening assay system to quantify the engulfment. Our research will provide not only important clues for the molecular mechanisms underlying developmental axon pruning but also new pathophysiological insights into neurodegenerative diseases.

The long-term health of all metazoan cells is inextricably linked to protein quality control. An imbalance in protein homeostasis, whether caused by the expression of mutant proteins or environmental and physiological stress, can result in severe molecular damage to the cell, dysregulation of tissues leading to pathology, and increased risk of protein conformational diseases including cancer, immunodeficiencies, metabolic disease, and neurodegeneration. Adaptation to stress and survival, when challenged by disease, requires an ability to sense damaged proteins and to coordinate induction of protective stress response pathways, chaperones and clearance networks.

This proposal brings together the complementary strengths of the Hartl lab at the Max Planck Institute of Biochemistry (Martinsried, Germany), the Nagata lab at Kyoto Sangyo University (Kyoto, Japan) and the Morimoto lab at Northwestern University (Evanston, USA) to decipher the mechanisms underlying the maintenance of protein homeostasis (proteostasis) at the biochemical, cellular, and organismal levels. We will employ quantitative proteomics to map the proteins that are sequestered by disease protein aggregates, identify different aggregation states and the toxic species that mediate aberrant protein interactions. Bioinformatic analysis of the physico-chemical properties of the candidate proteins will provide insight on function, localization and conservation and reveal features of proteins at-risk and the ability of proteostasis modulators to maintain functionality. Together, this represents a comprehensive analysis of the primary causes for proteostasis collapse in stress and aging and strategies to prevent or reduce damage and decline.

Key to the success of this proposal is the innovation described above, and that this can only be accomplished in a concerted effort by which we form new teams comprised of lab members from each group. We posit that only together will it be possible to address the ideas presented here. We will train talented investigators with a much broader scientific perspective and capabilities employing tools and approaches normally limited to cell biological, proteomics, or organismal approaches. Such efforts will be invaluable for the next generation of biomedical researchers as we work together to tackle such large-scale biological problems as proposed in this grant.

The development of nanotechnology to treat cancer promises to impact medicine and human health in the future. The most complex nanoparticles can carry out multiple functions, such as targeting, signaling and treating malignant cells. However, designing such multi-functional nanoparticles is challenging and current nanoparticles are still subject to low efficacy and limited tumor penetration.

Instead, cooperation could allow simple or specialized nanoparticles to work together towards the efficient and modular treatment of cancer. To investigate this, we take inspiration from swarm intelligence that studies systems in nature and robotics where simple agents with limited capabilities are able to cooperate towards a common goal. In these systems, the intelligence of the group emerges from local interactions among the agents and through the environment (stigmergy) without any centralized planning or global communication.

We identify three mechanisms of cooperation which will be studied and compared at the nano-scale, namely, implicit, direct and stigmergic cooperation. These mechanisms will be simulated and optimized in-silico and validated using in-vivo experiments aimed at treating brain tumors in mice models. Results will compare the benefits and costs across the different mechanisms of cooperation.

The final aim of this work is to issue a methodology and guidelines for the design and optimization of cooperative nanosystems.

A key challenge in biology is to understand the significance of genome duplication for adaptation and speciation. We propose to address this issue by analyzing gene network function following genome duplication in polyploid species of the plant genus Cardamine, which is related to the model species Arabidopsis thaliana. To this end, we will integrate bioinformatics and plant ecology to infer polyploid gene networks from transcriptome datasets of homeologous gene pairs. To functionally test network inferences we will generate synthetic ‘partial polyploid’ lines by transforming genes from one parent species into the other parent.

Network merging occurs in many diverse contexts. For example, in the successful merger and acquisition of companies, a strong leader reorganizes the social network in a top-down approach by removing redundant sections and generating new and enhanced functions. In contrast to this, little is known about how polyploid species acquire adaptive traits via the merging of parental gene networks. We hypothesize that adaptations for a new niche may arise via novel network reconfigurations. We further expect that sub-networks for stress responses that are unique to each parent may be safeguarded in the polyploid species, thus allowing them to exploit fluctuating habitats. To test each hypothesis we will infer and test nodes in the polyploid gene network that may have adaptive significance.

The merge of species living in wet and dry habitats occurred recurrently in Cardamine, and the resultant polyploid species occupy new ecological niches with fluctuating water levels. We can exploit transgenic methodologies developed in C. hirsuta to test network inferences and gene functions at critical nodes in polyploid gene networks. Moreover, two polyploids arose during the last century, providing a unique opportunity to study recent network merging in its native environment.

We will develop new algorithms in order to infer gene networks in merged polyploid genomes. We will design next generation sequencing and custom microarray approaches to distinguish homeologous gene expression, and we will develop novel network comparison algorithms to infer critical genes in the polyploid networks. Thus, we will advance the state of the art in network analysis and reveal principles underpinning successful network mergers in the formation of polyploid species.

Collective migration of endothelial cells is responsible for the formation and maintenance of the vasculature, for healing following injury, stroke or cardiac infarct as well as for cancer formation. While necessary for migration, the role and molecular regulators of membrane trafficking during the migration process are not yet understood. Here I propose to investigate the molecular mechanisms and physiological role of membrane trafficking during collective endothelial cell migration following three main experimental strategies. First, during planar cell migration in vitro, by validating a focused siRNA screen targeting known trafficking pathways. Second, during endothelial tube formation, where endothelial cells growing on beads form vascular sprouts in a 3D environment. Third, in zebrafish, during regenerative angiogenesis following caudal fin amputation. This combination of in vitro and in vivo methodologies will allow us to rapidly evaluate the physiological relevance of identified key regulators. Detailed biochemical and cell biological analyses involving transport assays, high-resolution imaging of localized signaling activities by confocal and TIRF microscopy, and real-time analysis of vascular sprouting, will provide the basis for the development of mechanistic models of how membrane trafficking orchestrates migration and vascular morphogenesis. Together, the proposed research will provide mechanistic insight of broad interest, as collective cell migration is a widespread phenomenon in animal physiology, while also identifying potential targets for anti-angiogenic therapies.

Polycomb group (PcG) proteins are transcriptional repressors that have been implicated in various developmental and disease processes in a broad spectrum of animals, from nematode worms to humans. PcG proteins mediate chromatin remodelling and are required for the stable silencing of homeotic (Hox) genes, which control patterning along the anteroposterior body axis. Despite intensive efforts in a variety of systems, the exact mechanisms of PcG function have remained elusive.

Recent studies suggest that PcG proteins interfere with the transcriptional state of RNA Polymerase II (Pol II) at paused Hox genes in the Drosophila embryo. We propose to experimentally validate this interaction and show that it eliminates plasticity in the expression of Hox genes, and therefore represents a mechanism that ensures the robustness of embryonic patterning and maintenance of organ identity.

Our model will be tested using a combination of whole-genome assays and functional genetics methods. We will analyse the transcriptional state of Pol II in contexts where PcG function is impaired in the absence of a phenotype. Our aim is to reveal the existence of a 'poised' state in Drosophila: although silent, paused genes contain fully activated Pol II and are ready for rapid activation by inducing signals. We will test the hypothesis that PcG enzymatic complexes mediate a conversion from a 'poised' state into a 'stalled' state at Hox gene promoters and that this interaction contributes to homeotic silencing. Finally, we aim to demonstrate that a poised transcriptional state introduces plasticity in Hox expression and reduces gene pattern accuracy in a fluctuating environment.

How structural novelties arise is a key question in evolutionary developmental biology. Land plants evolved several morphological innovations such as roots, leaves and vascular system. Vascular cambium is a proliferative tissue essential for radial thickening of plant body, which is found in seed plants: angiosperms and gymnosperms. In this tissue, cambial stem cells continue to produce daughter cells that will differentiate into specific types of vascular cells. To address how the cambial stem cells and therefore vascular cambium originate in the ancestor of seed plants, I propose to explore the functional evolution of TDIF peptide signaling, which regulates cambial stem cell maintenance in the angiosperm Arabidopsis thaliana. Because the ligand and receptor genes for TDIF signaling are conserved in the lycophyte (non-seed plant) Selaginella moellendorffii, these genes provide a clue to investigate this issue. Specific aims of this project are: 1. To identify TDIF signaling genes from species in the major taxa of vascular plants including lycophytes, ferns, gymnosperms and angiosperms. 2. To compare the expression pattern, physiological activity and molecular function of the identified genes/proteins to those in Arabidopsis. From the analyses, I expect to clarify how TDIF signaling is incorporated into cambial stem cell maintenance and whether the vascular cambia of seed plants share a common origin as suggested in previous anatomical studies. Experimental methods developed in this project will contribute greatly to investigate evolution of other traits in land plants.

The catacombs of ancient Egypt contain literally millions of well-preserved Sacred Ibis (Threskiornis aethiopica) mummies. These birds were revered by the Egyptians from the Predynastic period (7,500 – 5,100yrBP), by which time their relationship to the god Thoth had been well established. Ibis have been a focus of worship ever since and by the Late Period (~2600yrBP) the numbers being mummified had reached the millions. At one site alone in a single year 10,000 were estimated to have been interned. These mummies are common in many museums around the world and large numbers are still preserved in catacombs from at least five sites in Egypt. Using archaeological and Optically Stimulated Luminescence methods we can precisely age these mummies. The research team aims to use ancient DNA and genomic methods to study both rates and modes of DNA evolution using this remarkable resource. In recent years, a number of ancient DNA studies have made significant contributions to our understanding of the processes that govern DNA change. However, generally these studies have been limited by the lack of large numbers of known-age remains from multiple time points. Specifically, we will recover more than 200 nuclear microsatellite DNA loci from samples of different ages and use these data to test mutational models that have been formulated to explain the evolution of these DNA repeat loci. In order to develop powerful methods of analysis we need to have a clear understanding of the factors affecting the development and evolution of microsatellite DNA alleles in vertebrates. The proposed research will directly contribute to this improved understanding. In addition, we will sequence more than 100 mitochondrial genomes from ancient Ibis. With the aid of new mathematical and evolutionary models we can greatly improve our estimation of molecular rates. These mitochondrial data will also help us to understand the Egyptian traditions relating to animal mummification. In preparation for this work, we have shown that we can recover both mitochondrial and single copy nuclear DNA sequences from Ibis mummies. The large number of known-age mummified Sacred Ibis in Egypt, together with the multidisciplinary nature of the team (ancient genomics, evolutionary biology, mathematics and Egyptology), provides a unique opportunity to study DNA changes over time.

Cellular bistable systems draw increasing attention because of their roles in epigenetics, differentiation and cellular memory. It remains unclear what parameters are tuned by the cells in order to improve cellular memory or stabilize cellular differentiation states. We will explore the strategies evolution employs for adjusting parameters in bistable systems to optimize their function. In Saccharomyces cerevisiae, the galactose regulatory network induces ON and OFF cells with stable, high and low expression states, respectively. The state of a cell population does not only depend on the present galactose concentration but also on the past condition. This bistable memory requires positive feedback loops. After whole genome duplication, homologous GAL1 and GAL3 subfunctionalized and assumed negative and positive regulatory roles, enclosing negative and positive feedback loops, respectively. K. lactis, a relative of S. cerevisiae, maintains also a positive feedback loop through Gal4, the transcriptional activator of the network. We will investigate the evolution of the GAL network by measuring the system parameters. In addition, we will apply synthetic networks to mimic the natural network for the flexibility to tune the parameters and observe the responses. This project will focus on the evolutionary reparametrization of the network and how nonlinear responses and stochastic gene expression contribute to the development of cellular memory. Combining experimental and computational approaches, this project will lead to a better understanding of the stability of cellular differentiation states and the roles and origins of the cellular memory.