This proposal is to examine the hypothesis that an early host defense response is the ubiquitination of essential viral proteins by host E3 ligases and the subsequent degradation of the ubiquitinated viral proteins by 26S proteasomes. Dr Chua’s lab has obtained preliminary results showing that the Turnip mosaic virus (TuMV) HC-Pro is unstable in transgenic plants expressing a 35S-HC-Pro transgene. This viral protein interacts in vitro with an Arabidopsis RING10. I will confirm these in vitro interaction results and define the interaction domain on each partner by analysis of deletion mutants. I will express RING10 in E coli and demonstrate that recombinant RING10 can self-ubiquitinate as well as ubiquitinate HC-Pro in vitro. I will construct RING10 promoter-GUS fusion gene explore whether its expression can be induced by biotic stress, particularly TuMV infection.To access the role of RING10 in anti-virus defense in vivo I will use transgenic technology to manipulate RING10 levels/E3 ligase activity. First, I will make the 35S-HC-Pro transgene in mutant background. Second, I will generate transgenic plants by RNAi suppression technology. Third, I will use an inducible system to over-express the WT RING10 to change in vivo E3 ligases activity. Arabidopsis plants will be infected with TuMV and time course experiments will be performed to examine changes in levels of HC-Pro during the infection and to obtain evidence that its instability is due to ubiquitination by RING10. Finally, based on the results, I will explore novel strategies to engineer TuMV tolerance/resistance in transgenic plants by enhanced degradation of HC-Pro or by over-expression of binding proteins.

A new paradigm of gene regulation has emerged recently with the discovery of small regulatory RNAs and associated proteins that belong to the Argonaute family (Ago and Piwi proteins). microRNAs, a class of ~ 22 nucleotide RNAs are bound to Ago proteins and function by base paring with microRNA-recognition elements (MREs), found in their numerous mRNA targets, and direct either target mRNA endonucleolytic cleavage or translational repression. Piwi proteins in the testis associate with a novel class of 26 to 30 nucleotide small RNAs termed Piwi-associated RNAs (piRNAs). The functions of Piwi proteins and piRNAs are completely unknown. Whether piRNAs exist in other tissues where Piwi proteins are expressed is also unknown. Because of the similarity of domains of Piwi proteins with Ago proteins, I hypothesize that, as is the case of Ago/miRNA ribonucleoprotein, binding of a piRNA/Piwi ribonucleoprotein to a target RNA leads to silencing of its expression and plays an important role in meiosis in male germ cells. To unravel the functions of Piwi proteins and piRNAs, I propose to identify and validate human and mouse piRNA targets. I will use bioinformatics to predict piRNA-recognition elements (PREs). I will also utilize a biochemical approach based on crosslinks between target RNAs and piRNA/Piwi ribonucleoprotein followed by anti-Piwi immunoprecipitations to isolate RNAs that associate physically with piRNA/Piwi ribonucleoproteins. Moreover, I will investigate whether piRNAs are expressed in non-germ cells by performing immunoprecipitations with anti-Piwi antibody and identifying Piwi-associated RNAs. Finally I will validate the piRNA targets using mice as model system.

The aim of this application is to reveal novel principles of adult dopaminergic (DA) neurogenesis in the Parkinsonian vertebrate brain. I will study brain regeneration events in a salamander model of Parkinson's disease, which was recently developed. Salamanders are unique among vertebrates because they can regenerate large body structure as adults, but it was unknown how they respond to brain injury. In our lab we recently demonstrated that after the elimination of midbrain DA neurons by intraventricular injection of 6-hydroxydopamine (6-OHDA), regeneration in salamanders involves DA neurogenesis. Regeneration leads to complete histological restoration and to the recovery of motor behavior within one month. Our unique Parkinson's disease model shows that adult vertebrate midbrain DA neurogenesis is possible. I will take advantage of the fast and robust regeneration events in the salamander brain to understand the molecular and cellular programs of stem cell activation and differentiation, which result in newborn DA neurons. In this application I propose to further develop the salamander model of Parkinson's disease by (1) descriptive analyses of gene expression patterns and (2) by developing ways for functional manipulation of DA regeneration in vivo. The advantage of the salamander Parkinson's disease model is that it allows studying midbrain regeneration and can provide additional information compared to the available mammalian Parkinson's disease models. I believe that our results will lead to better understanding of adult DA neurogenesis and reveal how it could be efficiently activated in mammals.

Meiosis is a specialized cell division process that produces haploid gametes through two rounds of chromosome segregation. The first meiotic chromosome segregation is different from that in mitosis, because homologous chromosome pairs, not sister chromatid pairs, segregate from each other. To allow this unique segregation, it is important that homologous chromosome biorientation is established until metaphase I, by which homologous kinetochore pairs face the opposite spindle poles whereas sister kinetochores face the same poles. Previous studies suggest that chromosome condensation and specific action of microtubules to the chromosome surface may contribute to formation of the bioriented structure of the homologous chromosomes. However, it has not been examined how spatiotemporal structural change and dynamic behavior of the chromosomes lead to establishment of homologous chromosome biorientation. Thus, I will visualize chromosome dynamics with high-resolution microscopy in live mammalian meiosis. To this end, I will use the mouse oocyte system, because induction of meiosis, expression of engineered proteins and perturbation of protein function are available in an in vitro culture. I will use advanced techniques of automated confocal microscopy and also selective plane illumination microscopy (SPIM) to establish quantitative high-resolution 4D imaging assays for chromosome biorientation. I will then use these assays to probe the molecular mechanisms for homologous chromosome biorientation, by functional perturbations of the cytoskeleton and candidate proteins such as condensins and Polo kinase by RNAi, specific drugs, or female germline specific conditional knockouts.

Estrogen plays an essential role in reproduction and has been implicated in endometrial and breast cancers. Estrogen-regulated gene expression is exerted by the two estrogen receptors ERalpha and ERbeta. Although significant progress has been made over the last 20 years for the understanding of estrogen receptor action, the current picture of estrogen signaling is fragmentary. A recent genome-wide study of ERalpha binding sites and estrogen-regulated gene expression in my host lab suggests a highly complex mechanism of ERalpha action, implicating additional transcription factors and epigenetic changes through several chromatin factors. I aim to explore the function of the identified chromatin factors in ERalpha action to provide mechanistic insights in the role of epigenetic changes in estrogen signaling. I will focus on two questions: 1. How does the histone variant H2A.Z, whose expression is stimulated by estrogen, contribute to ERalpha action? 2. How does GATA3, whose binding sites are enriched at ERalpha chromatin targets, affect ERalpha recruitment? To answer these questions I will study the effects of H2A.Z depletion and overexpression on gene expression, analyse the binding of H2A.Z to specific targets and analyse the modification states of H2A.Z upon estrogen stimulation. To dissect the role of GATA3, I will analyse the effects of GATA3 depletion on gene expression and ERalpha binding and the effects of GATA3 binding on the local nucleosome structure. While my PhD work focused on RNAi screens in cell biology, this detailed mechanistic analysis of a regulatory transcriptional network presents a unique opportunity to grow in new research directions.

The refinement of approximate protein structures is a crucial step in protein structure prediction with applications in various other areas of molecular biology. Specifically, knowledge of high resolution protein structures is an important pre-requisite in molecular medicine and rational drug design. The generic protein folding potentials however fail when used for structure refinements and hence atomic refinement of low-resolution structures is a significant challenge to the community. The goal of the proposed research is the development of a novel algorithm for the atomic refinement of structural models of proteins. At the core of the algorithm is the automated design of a unique energy function, based on rigorous mathematical principles, that is continuous and is locally guaranteed to bring the approximate protein structures closer to the native conformation. A mathematical programming algorithm is proposed for learning this energy function by solving inequalities of the nature, U(model)-U(native)>0 (where U is the potential energy), under local and global constraints, using optimization protocols from applied mathematics. The potential obtained from present algorithm has many advantages over traditional statistical potentials because of its continuous nature, training from positive and negative examples and the local and global constraints used during learning. In practical terms, the new energy function will make it possible to perform straightforward and local energy minimization to consistently improve the quality of the model, something that has so far eluded investigators in the field of protein structural biology.

The progressive development of multi-antibiotic resistance and the partially endemic appearance in hospitals and health care facilities, make Acinetobacter baumannii a major emerging threat to public health. The characterization of the iron-sequestering strategy via the non-ribosomal peptide siderophore acinetobactin is of particular interest. The siderophore acinetobactin has been identified as a pathogenic factor with a demonstrable hemolytic ability. The non-ribosomal peptide synthetase that prepares acinetobactin is atypical as it exists as a dissociated system of functional independent proteins units. BasD is a 112 kDalton protein of the acinetobactin synthetase. Based on homology searches, it consists of an oxazoline cyclase domain, a putative dehydrogenase domain, a region of approx. 250 amino acid residues with no known homology and a C-terminal carrier domain of unknown function. This organization of the acinetobactin synthetase represents a model system to study protein-protein interactions and molecular recognition even of large multifunctional proteins without disrupting any of the domains. The goal of the proposed research is the biochemical characterization of (1) the novel oxazoline cyclization domain in BasD, (2) the characterization of the specific interactions between the subsequent domains in BasD and (3) the required interactions with other proteins in the biosynthetic pathway of the siderophore acinetobactin.

The hypothesis that cell division failure that result in genetically unstable tetraploid (4N) cells promotes tumorigenesis was proposed almost 100 years ago, and it has been supported by extensive in vitro and clinical correlations. The host laboratory recently provided direct experimental evidence in support of this hypothesis: tetraploidy, at least in combination with p53 loss, could promote the formation of aneuploid tumors. However, the mechanisms underlying transformation of 4N cells are poorly understood. To define the mechanisms, I propose experiments with the following aims. Aim 1: although numerous links have been demonstrated between p53 loss and tetraploidy, the role of p53 loss in transformation of 4N cells has not been determined. Therefore I will investigate whether p53 loss promotes tumorigenesis by 4N cells through aberrant DNA damage checkpoint response and/or suppression of apoptosis. This will be performed using mouse 4N embryonic stem (ES) cells that are genetically stable and wild type for p53. The p53 in this system will be depleted by lentivirus-mediated RNA interference. Aim 2: since mitotic 4N cells contain 4 centrosomes, they can form an abnormal multipolar spindle that results in catastrophic multipolar mitoses, leading to genome instability and tumorigenesis by 4N cells. However, this idea has never been tested yet, because the extra centrosomes in such cells cluster into two pairs. To test this idea, I will knock down a motor protein required for centrosome clustering (HSET) and force 4N ES cells to undergo multipolar mitoses. This experiment will determine whether multipolar mitoses promote DNA damage, genome instability, and tumorigenesis.

Understanding the mechanisms behind speciation is a central challenge of biology. The process of speciation often begins with postzygotic reproductive isolation: A cross between two populations can still produce a zygote, but the hybrids suffer from drastic impairment in their fitness and are either sterile or die. Only a few genes causing hybrid incompatibility have been identified, none in plants. Even for the few known in animals, it is unclear how important they were for establishing reproductive barriers. Through hundreds of crosses among wild arabidopsis strains, the receiving laboratory has identified several hybrid incompatibilities, most of which show very similar phenotypes, characterized by a pathogen response in the absence of pathogens. Analysis of the first incompatibility revealed that one of the causal genes is a disease resistance gene belonging to the R gene family. In my project, I will address the following questions: (1) What is the other, non-R gene partner in hybrid incompatibility, and what is the mechanism of interaction? (2) Are R genes causal for all of the incompatibilities in arabidopsis? If yes, is there a common theme to the type of R genes involved? (3) Do the incompatibility genes evolve faster than other genes in the genome? The proposed work uses plants as a study object, like my PhD work, but dealing with a very different area of biology, evolutionary genetics of population level processes rather than developmental biology and pattern formation. Moreover, postdoctoral training in the Weigel laboratory will introduce me to new concepts and approaches used in studies of natural genetic and genomic variation.

In an allosteric enzyme, substrate binding affinity, and thereby enzyme activity, is regulated by binding at spatially distant sites. Although many examples of allosterically regulated enzymes are known, the molecular mechanisms of the propagation of the signal through the structure are - up to now - poorly understood. An important feature of allosteric motion, is the coupling between small scale conformational changes and global collective motions of the protein. However, the accuracy and resolution of current methods does not suffice to study this coupling in detail. To identify intermediate structures we propose to use the successful structure prediction method ROSETTA. This will allow us to study the allosteric motion in molecular detail and to test our understanding by predicting allosteric motions from single protein structures. Crucial for this approach will be an enhanced conformational sampling. Thus, we combine ROSETTA with CONCOORD, a very fast sampling method, which has already been shown to sample a major part of the large collective motions present in allosteric enzymes. To test our predictions we will consider specific examples, where both conformational states are known from Xray crystallography. We believe that the allosteric signal is propagated via coupling of the changes near the effector site with those global elastic modes that are inherent to every protein structure. To test this hypothesis, we plan to design allostery into the dimeric protein triose-phosphate isomerase (TIM) whose active sites are known to be non-allosteric. Our design will be experimentally confirmed by measuring the increase in binding cooperativity of the enzyme.

The ability to perceive our environment through interactions with light represent billions of years of cumulative engineering resulting in the complex process of human vision. This required the careful optimization of subtle interactions between chromophores that absorb visible light and their surrounding rhodopsin proteins. Detailed knowledge of these interactions is important for not only understanding vision, but also for the development of novel proteins possessing new light activated functions. Often, the direction and efficiency of photo-induced reactions in proteins are controlled by rapid dynamics occurring as fast as a picosecond, which requires the use of ultrafast techniques to directly explore. This proposal describes a multi-disciplinary approach involving direct feedback between time-resolved measurements and rational protein design to develop new proteins that mimic the photodynamic features of the rhodopsin pigment protein. This allows for a more comprehensive investigation without the inherent limitations to the direct study of rhodopsin. This "bottom-up" approach to reengineering proteins serves as a powerful platform for designing specific protein interactions of interest. Key to the success of this approach is the construction of a state-of-the-art ultrafast laser laboratory capable of performing different time-resolved techniques for exploring transient electronic and vibrational properties. Support from Career Development Award will be highly instrumental in enticing talented international researchers to aid in its construction, which will allow for the samples to dictate which techniques are the most appropriate for their study and not the opposite.

Embryonic stem (ES) cells are defined by pluripotency and their capacity to self-renew. Epigenetic modifications play a crucial part in controlling ES cell identity, and an expanding body of evidence suggests that particularly histone methylation is involved in this process. Histone methyltransferase (HMT) enzymes, that catalyse the addition of methyl groups to histones, and histone demethylases (HDM) that are required for the reverse reaction, carry specific domains that are responsible for catalytic activity. Based on these domains, I have identified 43 potential HMT and 32 HDM enzymes in the mouse genome. The goal of this project is to determine which of the HMT and HDM enzyme in the genome is crucial for ES cell maintenance and differentiation. To this end, lentiviral-mediated expression of siRNA will be used to knock down HMTs and HDMs specifically transcribed in undifferentiated ES cells. If the targeted enzymes are crucial for ES cell maintenance, a knock-down would result in the promotion of differentiation. Moreover, lentiviral transduction to overexpress HMT and HDM genes, which are activated in the course of differentiation, will be employed to identify enzymes that can induce differentiation of ES cells. Furthermore, target genes regulated by the identified enzymes will be determined by genome-wide mapping of their binding sites using ChIP-chip analysis. This project will contribute to our understanding of which role chromatin regulation plays in the maintenance of ES cell identity.

New neurons are continually recruited throughout the adulthood and participate in functionally remodeling the neural network in the brain. One of the most important questions is how the new-born neurons integrate into the pre-exiting circuitry in the adult brain. Recent accumulating evidence supports a role for tonic and phasic GABA induced depolarize and NR1 activation of neural progenitors and new-born neurons in the dentate gyrus in promoting the differentiation and the synaptic integration in the adult brain. However, little is known regarding the role of intrinsic activity of progenitor cells in their incorporation in the adult brain. In this proposal, I plan to utilize retrovirus specific manipulate of new born neurons in the adult mouse brain in vivo, a technique developed by Professor Gage’s lab and sheds light on the study of neural stem cells. I propose to suppress the excitability in the new born neuron within active and mature hippocampal neural network by over-expression inward rectifier K+ channel using retroviral vectors. I plan to examine whether the supression of activity in new born neurons have any effect on their axon and dendrit development, on their circuit formation and competation with surrounded mature neurons. This project contributes to the mechanism of the integration of new born neurons into the preexisting circuitry and strong current interest of scientists in the neural stem cell field. This project will also allow me to investigate a new field, gain insights into new biological concepts and learn new technologies. This fellowship experience will undoubtedly help me to head my own research laboratory in future.

CLC chloride transport proteins are conserved from prokaryotes to eukaryotes and are involved in many physiological functions, such as muscle excitability, hippocampal development, or endocystosis. Defects in CLCs lead to many diseases. The CLC family encompasses both Cl- channels and Cl-/H+ antiporters. Crystal structures of the antiporters have been solved, but there are as yet no structures of the channels. Although the antiporters have been crystallized under a variety of different conditions, none of the structures are significantly different. This is surprising because the process of catalyzing stoichiometric exchange of Cl- for H+ must involve a series of conformational changes. Thus, it is likely that additional conformations exist but that only one of the conformations crystallizes. Additional methods of examining conformational changes are essential. I aim to study conformational changes of the CLC-ec1 antiporter using hydrogen/deuterium exchange (HDX) detected by mass spectrometry. HDX does not require crystallization of the protein and therefore allows mapping of all conformational states. Likewise, I propose to use HDX to study conformational changes associated with gating in the CLC-0 channel. This will yield the first structural data on a full-length CLC channel. Comparison of CLC-0 and CLC-ec1 will reveal to what extent the structure of a eukaryotic channel resembles that of prokaryotic transporter and will allow me to evaluate how conformational changes that lead to Cl-/H+ exchange in transporters are related to gating in channels. These results will help us understand how similar structures have evolved to carry out different physiological functions.