Ruprecht-Karls-Universität Heidelberg

Projects within the SFB 1036 (2nd funding period)

The SFB 1036 funding has been renewed for 2016 to 2020 (press release)

TP 1: Renate Voit & Ingrid Grummt - Transcriptional and epigenetic control by Sirtuins in response to stress

Project leaders' home page

Downregulation of transcription under metabolic, osmotic or oxidative stress is brought about by decreased transcription initiation or by adjusting the number of genes that are involved in transcription. We propose to investigate both mechanisms, which are intimately linked. The team of Renate Voit will study the role of the nuclear NAD+-dependent deacetylase SIRT7 in stress-dependent transcriptional regulation, the team of Ingrid Grummt will focus on the role of long noncoding RNA (lncRNA) in stress-mediated changes of chromatin structure, which connect epigenetic regulation to maintenance of genome integrity. We will use genome-wide approaches to identify novel targets and pathways that are linked to SIRT7 activity. We will compare the interaction of SIRT7 with specific proteins and RNAs, and monitor acetylation of SIRT7-associated proteins in unstressed, stressed and in SIRT7-depleted cells. This approach will be complemented by ChIP-seq and iCLIP-seq to analyze the association of SIRT7 with specific gene loci and with RNAs. To decipher the impact of SIRT7-dependent deacetylation on the function of SIRT7-associated proteins, we will analyze acetylation of fibrillarin, the enzyme that mediates site-specific 2'-O-methylation of rRNA, and correlate SIRT7-dependent deacetylation of FBL to methylation of rRNA and ribosome translational fidelity. A main research focus of both groups will be the elucidation of the mechanisms that form and suppress Rloops, i.e. structures that cause hyper-recombination, genome rearrangements, and genome instability. We will investigate whether natural antisense transcripts mediate epigenetic changes and affect transcription of overlapping genes by R-loop formation, and whether lncRNA-mediated changes in the structure of DNA and chromatin are recognized by protein complexes that modulate gene expression. We postulate that SIRT7 regulates the activity of DDX21, a DNA-RNA helicase which may resolve R-loops and safeguard genome integrity. Consequently, we will investigate (i) whether DDX21 resolves R-loops, (ii) whether the activity of DDX21 is regulated by SIRT7-dependent deacetylation, (iii) whether the acetylation state of DDX21 is altered under stress conditions, and (iv) whether the antisense RNA TARID activates TCF21 expression by forming an R-loop at the transcription start site. We will measure R-loop accumulation by DRIP-seq upon overexpression and knockdown of SIRT7 or DDX21, and monitor DNA damage by γ-H2AX accumulation and COMET assays. These approaches are straightforward and we believe they will uncover novel regulatory mechanisms that impact gene expression to preserve genome stability.

TP 3: Thomas Hofmann - Molecular mechanisms of DNA damage-induced cell fate decisions

Project leader's home page

In response to DNA damage, cells need to decide between different cell fate options including DNA repair and cell death. The right decision-making is of fundamental importance to prevent genomic instability and cancer development, but the underlying mechanisms are only partially understood. DNA damage-activated protein kinases, such as ATM and HIPK2, play a central role in DNA damage-induced cell fate control. In the first funding period we focussed on the mechanism of HIPK2 activation upon DNA damage and on the regulation of the deacetylase Sirtuin 1 (SIRT1) through site-specific phosphorylation by HIPK2 and ATM.

In the second funding period our research plan is centred on projects. In project 1 we will investigate the role of a novel adaptor protein in the DNA damage response. We identified this protein during the first funding period as binding partner of both HIPK2 and p53. Our unpublished results indicate that this protein is regulated in its subcellular distribution between cytoplasmic stress granules (SG) and the cell nucleus by HIPK2-dependent phosphorylation. The nuclear-targeted isoform of this protein interacts with p53, a master cell fate regulator of the DNA damage response. Our goal is to elucidate the nuclear and cytoplasmic functions of this factor with focus on its interaction with p53 and its function in SG biology in response to DNA damage.

Our second project aims at deciphering the molecular functions of HIPK2 in the DNA damage response in more detail. We will use unbiased mass spectrometry approaches to dissect its function and regulation in response to genotoxic stress. Our results are expected to deepen the understanding on the molecular mechanisms which guide and specify DNA damage-induced cell fate decisions.

TP 5: Andreas Kulozik & Matthias Hentze- Stress dependent responses of mRNA 3‘-end processing

Project leaders´ home page

Cells respond to a variety of different stress insults by widespread changes of gene expression. In this context, transcriptional and posttranscriptional stress-responsive mechanisms influence gene expression in a quantitative and qualitative manner. Regulating the 3' end processing of nascent mRNAs has recently been recognized as a mechanism that fine-tunes the abundance of specific gene products upon intrinsic and extrinsic stimuli. This is exemplified by the prothrombin (F2) gene whose expression is controlled via regulated 3' end processing at a polyadenylation (poly(A)) site containing a cis-acting RNA sequence element recruiting a stress sensitive ribonucleoprotein particle (mRNP). In addition, approx. 50% of mammalian genes harbor more than one functional polyadenylation site and can undergo alternative polyadenylation (APA). Such a change in poly(A) site usage may alter the length of 3'UTRs or change the coding regions thus qualitatively influencing gene expression.

During the first phase of this funding period we have tested the global role of the cis-acting acting RNA element and the associated mRNP that we have previously identified in the prothrombin 3'UTR. By performing RNAseq and pre-mRNAseq analyses in stressed and unstressed cells we have first tested the hypothesis that the motif identified in the F2 mRNA, the upstream sequence element (USE), may play a more general role in controlling 3' end processing. By using iCLIP, we also tested whether one of the proteins that we had previously shown to bind to the USE in a stress-dependent fashion binds to its cognate sequence in USE-containing mRNAs more generally. The results obtained revealed that the USE and its binding proteins is not likely to play a general role in controlling 3' processing. However, in the course of this work we developed a novel and generally applicable analysis tool of iCLIP data sets. Interestingly, this work revealed that anisomycin induced ribotoxic stress that activates the MAP38 kinase (p38), oxidative and endoplasmic reticulum (ER) stress cause alternative polyadenylation (APA). Stress-induced APA thereby affects the 3'UTR length (3'UTR-APA) or the coding region (CR-APA) of hundreds of human genes whose function in the cellular stress reponse or during the recovery period will now have to analyzed (see below).

We now aim at unraveling the molecular mechanism(s) and the physiological consequences of MAP38K, ER- and oxidative stress-induced alternative polyadenylation. During the course of this funding period we aim at (1) dissecting the molecular mechanism underlying alternative polyadenylation and (2) identifying the physiological role of APA-regulated mRNAs. Specifically, we aim at determining the cis-acting RNA sequence elements and trans-acting proteins that influence poly(A) site choice in stress in a system-wide manner. We further want to identify the cellular signaling pathways that underlie stress-induced alternative polyadenylation. Based on the results obtained during the first funding period, these analyses will be focused on common stress related signaling pathways such as p38, JNK and ERK. Also based on results of the first funding period which implicated oxidative stress as one of the stressors that induce alternative polyadenylation and based on previous work of the Muckenthaler group (TP16), we will use a mouse model with an increased level of oxidative stress caused by iron-overload (FPN C326S ki) and a mouse model with a reduced level of oxidative stress under conditions of iron depletion (FPN +/- ko), which are made available within this SFB by the Muckenthaler group to study the role of stress induced alternative polyadenylation on an organismal level. Finally, we intend to specify the role of APA-regulated protein-encoding mRNAs during the stress recovery phase to unravel the physiological impact of stress-regulated alternative polyadenylation. The results of this project will provide detailed mechanistic and physiological insights into stress-induced alternative polyadenylation.

TP 7: Georg Stoecklin - Coordination of protein synthesis and cell cycle control under stress conditons

Project leader's home page

Cell cycle arrest and inhibition of protein synthesis are both hallmarks of stress responses. Thereby, cells avoid segregating their genomes with damaged DNA, reduce energy consumption under adverse conditions, reallocate their resources to the repair of cellular structures, and prevent the accumulation of misfolded proteins. Translation suppression goes along with the assembly of cytoplasmic stress granules (SGs), which harbor stalled mRNAs together with RNA-associated proteins and signaling molecules. Within this SFB, we conducted a siRNA screen to identify novel regulators of translation and SG assembly, and identified cell cycle-related kinases as candidates. Our goal is to determine the mechanism by which cell cycle-related kinases control translation, and examine more generally the role of proliferation in the response to and recovery from stress conditions.

As a second project within the SFB, we want to determine the exact role of TIAR in regulating the cell cycle. TIAR is an RNA-binding protein that controls splicing in the nucleus, translation suppression in the cytoplasm, and SG formation through its aggregation-prone C-terminal domain. We found that TIAR is essential for genome stability by inhibiting cell cycle progression under conditions of DNA damage and replication stress. Our goal is to determine the molecular mechanism by which TIAR controls the cell cycle. In collaboration with Sylvia Erhardt, we further assess whether this function of TIAR is evolutionary conserved in Drosophila melanogaster. Our research will deepen our understanding of how essential cellular programs such as cell cycle control and protein synthesis are coordinated during the onset and resolution of stress responses.

TP 8: Bernd Bukau & Axel Mogk - Mechanisms and physiological functions of organized protein aggregation and disaggregation

Project leaders' home page

Perturbation of protein homeostasis caused by exposure to stress conditions, aging or pathophysiological states results in the accumulation of misfolded proteins, which is linked to cellular toxicity. Protein quality control systems use diverse strategies to fight misfolded proteins, including chaperone-mediated refolding, proteolytic removal and active sequestration into deposits of aggregated proteins. The long-term goal of our research is to elucidate the mechanisms and physiological implications of organized protein aggregation and how the active sequestration of misfolded proteins is intertwined with subsequent protein disaggregation and refolding. Concerning the aggregation of misfolded proteins, in the first funding period, we discovered a novel nuclear deposit (INQ) for misfolded proteins in Saccharomyces cerevisiae. We identified compartment-specific aggregases promoting sequestration of misfolded proteins in the cytosol (the small Hsp member, Hsp42) and nucleus (Btn2), thereby establishing a novel concept for organized protein aggregation in yeast that involves nucleo-cytoplasmic shuttling of misfolded proteins for aggregate deposition. Btn2 is an unstable protein and its expression is strongly induced upon stress such as heat shock, limiting Btn2 accumulation and INQ formation to the immediate phases of folding stress. Btn2-dependent nuclear INQ also forms upon DNA replication stress and includes regulators of DNA damage checkpoint, suggesting DNA surveillance integrates into protein quality control via INQ and Btn2. Concerning the molecular features of protein aggregates and subsequent disaggregation, biochemical analysis of Hsp42 revealed that it sequesters early unfolding intermediates, thereby preserving a native-like core structure of bound substrates. This high degree of structure preservation facilitates substrate recovery by disaggregating Hsp70-Hsp100 chaperones and defines a powerful cellular strategy for protein repair. In the next funding period we will continue unraveling the mechanisms of organized protein aggregation by cellular aggregases at the molecular level and define the physiological roles of compartment-specific sequestration sites. Our three major research aims are:
(1) To dissect the mechanism of Btn2 driven protein aggregation in the nucleus. We will determine how this uncharacterized aggregase sequesters misfolded proteins and how and why Btn2 levels are regulated so tightly.
(2) To analyze the interplay between aggregase and disaggregases in the nucleus by determining the impact of Btn2 on substrate structure and investigating how Btn2-bound substrates are transferred to disaggregating machineries.
(3) To determine the physiological roles of compartment-specific sequestrations in stress adaptation and cellular aging. We will also analyze the potential role of the nuclear INQ deposit in coordinating stress signaling pathways, including the DNA damage checkpoint and the heat shock response.

TP 9: Matthias Mayer - Molecular mechanism of the eukaryotic heat shock transcription factor Hsf1

project leader's home page

The heat shock response (HSR) is the major response of cells to disturbances of cellular homeostasis. It is not only induced by different types of physical and chemical stress, including temperature increase, mechanical strain, oxidative stress, heavy metals, and pH changes, but also during development and pathophysiological processes such as cancer and neurodegeneration, and is critical for aging and longevity. In all eukaryotic cells the diverse cues that induce the HSR converge on the heat shock transcription factor 1 (Hsf1), which influences the expression of more than 2000 genes, some or which encode chaperones like Hsp70 and Hsp90 that negatively regulate Hsf1 activity. The molecular mechanism how Hsf1 integrates the different signals, is still unclear. Some basic principles of Hsf1 activation and attenuation are similar in all eukaryotes, and yeast has been used to gain important insights into the regulation of Hsf1 in higher eukaryotes. Metazoan Hsf1 was proposed to directly sense temperature and its activation involves transition from the monomeric to the trimeric state, which is prerequisite for DNA binding, and a large number of posttranslational modifications with uncertain effects on Hsf1 activity. Upon prolonged stress exposure the HSR attenuates by an unknown mechanism involving Hsp70 and it is also unclear how recovery and return to the pre-shock state is achieved.

The long-term goal of our investigations is to elucidate the molecular mechanism of Hsf1 activation, attenuation and recovery. In the past funding period we aimed at (1) elucidation of the effects of temperature on the conformation of Hsf1; (2) how chaperones influence temperature-induced conformational changes in Hsf1; and (3) identification of new interaction partners of Hsf1, that may regulate temperature-induced activation and attenuation of the HSR. We used hydrogen-1H/2H-exchange mass spectrometry (HX-MS) to analyze the effects of temperature on the conformational dynamics of human Hsf1. We found that Hsf1 directly measures temperature through local unfolding of a small region within its regulatory domain, and that the unfolding temperature unexpectedly depends on Hsf1 concentration, indicating that temperature sensing does not occur in the monomeric state. Our data yielded a novel model for the temperature-induced trimerization and DNA binding of Hsf1. Surprisingly, Hsp90, generally supposed to repress the HSR by keeping Hsf1 in the monomeric state, reduced transition temperature and steepness of the activation curve in vitro, thus promoting Hsf1 trimerization and DNA binding at lower temperatures and widening the temperature window of Hsf1 activation. We also created stable tagged Hsf1-transfected cell lines, which allowed us to identify wellestablished and previously unsuspected proteins in complex with Hsf1 during the attenuation phase of the HSR. Finally, to gain insights into the role of Hsf1 in aging and longevity we established a method for analyzing the stress response and key cytosolic parameters that influence the stress response, as pH, H2O2-levels, and glutathione redox buffer, in replicatively aging yeast cells. The central finding of this last study is that the HSR is deregulated in aging yeast cells and that the redox state of the glutathione buffer does not influence longevity in yeast, which is in contrast to the oxidative stress theory of aging.

TP 10: Michael Knop & Anton Khmelinskii - Protein homeostasis during conditions of chronic stress in yeast

Project leaders' home page

Organisms are continuously challenged by various stresses that can be external (toxic agents, nutrient limitations, ionizing radiation) or internal (aneuploidy, changes in gene copy number or mutations). Cellular responses that are necessary for growth under these conditions aim at counteracting the effects of the stresses by removal and degradation of damaged components and the by eliciting an adaptive response.

In the first funding period we developed and employed a systemic approach to explore changes in protein turnover under conditions of chronic stress. This involved the use of a proteome-wide yeast resource to screen for stress-induced changes in protein abundance or stability and to characterize underlying responses with a focus on selective protein degradation by the ubiquitin-proteasome system (UPS). This work resulted in a large data set and initial analysis of this data set led to the identification of a novel protein quality control pathway, mediated by the Asi E3 ubiquitin ligase, that removes mislocalized trans membrane proteins from the inner nuclear membrane.

In the second funding period we will continue and expand our work on proteome turnover under normal and chronic stress conditions along three lines of research. In the first Aim we will seek a global understanding of the UPS and its roles in proteome turnover to identify general principles of proteome organization and dynamics. We will explore the data set of changes in proteome turnover in UPS mutants to (i) define substrates and functions for different UPS enzymes by integrating additional data such as protein-protein interactions, genetic interaction and ubiquitination sites, (ii) search for potential sequence motifs specific to substrates of individual enzymes, (iii) construct a network of the UPS based on phenotypic similarities between UPS components and (iv) examine changes in network organization in chronic stress conditions to identify potential condition-specific regulators.

In combination, the work proposed will allow for a systemic but detailed exploration of the regulatory pathways underlying cellular protein homeostasis and degradation upon exposure to a set of chronic stresses imposed by environmental challenges.

TP 11 Sabine Strahl - The role of protein O-mannosylation in the ER quality control and the unfolded protein response

Project leader's home page

In all eukaryotes, proteins that enter the endoplasmic reticulum (ER) are potentially N- or O-glycosylated and fold with the help of molecular chaperones and other folding helpers. Proteins that fail to assume a final native state are retained in the ER by a conserved process referred to as ER quality control and eventually degraded by ER associated degradation pathway(s) (ERAD). When the ER load of unfolded proteins becomes too high, the unfolded protein response (UPR) adapts the secretory pathway to this ER protein load and increases the efficiency of protein folding. N-glycosylation serves an important role in protein folding and the monitoring of the folding state of proteins as well as the targeting of unfolded proteins for degradation by ERAD. In fungi and animals, also O-mannosylation is an essential protein modification that is initiated in the ER by a conserved family of protein O-mannosyltransferases (PMTs). In recent years evidence is accumulating that this type of glycosylation is also crucial for ER quality control and the UPR. However, the molecular mechanism and components of the so-called unfolded protein O-mannosylation (UPOM) system are still poorly understood.

In the previous funding period we set the stage to further unravel the role of protein O-mannosylation for ER homeostasis. We developed a mass spectrometry-based approach that allowed us to describe the first yeast O-mannose glycoproteome and established O-mannose directed monoclonal antibodies, a valuable tool for the analysis of O-mannosylated glycoproteins. Furthermore, we identified candidate components of the UPOM pathway, and its connection to other stress response modules that help cells mounting an effective ER quality control and UPR. The fact that we found major mediator and effector proteins of ER quality control and UPR on the side of PMT substrates potentially adds a hitherto unrecognized layer of regulation in ER homeostasis. Based on this work, we continue and extend our ongoing study to further unravel the UPOM system and to clarify how O-Man glycans impact on the performance of ER chaperones and folding helpers.

TP 12: Marius Lemberg - The role of rhomboid proteases in membrane protein quality control

Project leader's home page

As the main entry point of the secretory pathway, the Endoplasmic Reticulum (ER) harbors complex quality control machinery that targets damaged and misfolded proteins into ER-associated degradation (ERAD). Similar checkpoints exist for the inner nuclear membrane, mitochondria, the Golgi compartment, and the late secretory pathway. A fundamental challenge is to understand how damaged membrane proteins are recognized and extracted from the lipid bilayer such that the rest of the organelle remains unaffected. While most of our understanding of ERAD comes from the p97-mediated dislocation of full-length substrates, we previously showed that the ubiquitin-dependent rhomboid protease RHBDL4 cleaves certain membrane proteins with unstable transmembrane helices thereby facilitating their degradation. However, the physiological function of RHBDL4 and how it is related to classical ERAD remains unclear. In order to allow physiological conclusions, we have established proteomic techniques to identify endogenous substrates. The continuous central objective of this project will be to decipher the molecular function of RHBDL4 and to define the conserved role of rhomboid proteases in the control of membrane protein homeostasis and stress response. We expect that our comprehensive analysis of rhomboid proteases will identify native substrates and regulatory principles in membrane protein homeostasis and stress response.

TP 13: Rüdiger Hell & Markus Wirtz - The importance of N-terminal acetylation dependent protein degradation for stress responses

Project leaders' home page

In the first period of the SFB we showed that N-terminal acetylation of proteins by the two major Nacetyltransferases (NatA and NatB) affects approximately 70% of proteins in plants. While originally thought to be constitutive, we uncovered the dynamic regulation of the plant protein acetylome. It was shown that the drought stress-induced hormone abscisic acid supresses N-terminal acetylation by NatA. Indeed, genetically engineered down-regulation of NatA in plants was sufficient to induce an effective drought stress response. Down-regulation of NatA was demonstrated to cause increased global protein turnover by activation of the ubiquitination dependent proteasome system. Among the affected proteins were key response factors like the plant immune receptor SNC1, indicating that altered acetylation can affect specific stress response pathways. In line with these observations we found that several stress responses such as UV and redoxstress are modulated by N-terminal acetylation via NatA or NatB.

The goal of the submitted project is to elucidate the mechanism of stress-induced protein quality control by NatA and NatB, which contributes significantly to cellular surveillance. At present it is unknown how stresses after their perception are translated into alterations of the acetylome, i.e. how Nat activity and specificity are modulated. Once the acetylome has been modified, it is unclear how the altered acetylation status of target proteins is signalled to stress response pathways. We will therefore focus on two major aims: 1) how are distinct stresses translated into a defined alteration of the protein acetylome and 2) how is acetylation of NatA and NatB substrates affecting the response of plants towards stresses.

In order to address aim 1 we will quantify stress-induced changes of the protein acetylation status with high spatial and temporal resolution and test the impact of these stresses on abundance and ribosome association of NatA and NatB. This will allow unravelling stress-induced adaptation of the acetylome in a sequence context specific manner, since substrate specificities of both Nats have been determined in the previous project. We will furthermore test the postulated role of the NatA complex as a hub for hormone signalling by quantification of transcriptional and post-translational regulation of NatA in response to diverse stress-related plant hormones. The impact of acetylation on protein stability and function (aim 2) will be analysed by crossing established Nat mutants with mutants deficient in selected components of the ubiquitin dependent proteasome machinery (Ac/N end rule pathway) and pharmacological inhibition of translation and degradation.

Stable isotope labelling of cell cultures will be applied to dissect if NatA depletion destabilizes selectively NatA substrates or induces protein degradation by global activation of ubiquitination mediated proteolysis. Alternatively, the fate of non-acetylated proteins in response to stresses will be followed by specific biotin labelling of free protein N-termini. The innovative design of this approach allows mass-spectrometry based quantification of non-acetylated proteins in response to stresses by multiplex peptide stable isotope dimethyl labelling.

The expected results will demonstrate the role of global and specific proteome imprinting by the plant Nterminal acetylation machinery as a hormone controlled safe guarding system that regulates manifold stressresponse pathways by determining protein-half life time of key stress response factors, like immune receptors or transcription factors.

TP 14: Tobias Dick - Molecular mechanisms of protein thiol oxidation in peroxide sensing and stress signalling

Project leader's home page

Over the last ten years it has become clear that there are many physiological and stress situations in which hydrogen peroxide (H2O2) acts as a messenger in signal transduction and triggers highly specific adaptive responses. H2O2 is sensed through the oxidation of thiol groups which act as molecular switches of protein function. These reversible posttranslational modifications can either be inhibitory or activating. For example, thiol oxidation transiently inactivates certain protein tyrosine phosphatases during growth factor signalling. The role of H2O2 as a signaling molecule presents us with a number of difficult questions which are in need of clarification: How exactly does H2O2 lead to protein oxidation? How can the oxidation process be specific? How can thiol oxidation happen on time scales relevant to cell signaling, considering kinetic barriers and kinetic competition with peroxidases?

In the course of the first funding period we discovered and described the first example of a peroxiredoxinbased redox relay in mammalian cells: Peroxiredoxin-2 (Prx-2) serves as the primary H2O
2 receptor and facilitates oxidation of the transcription factor STAT3 under signaling and stress conditions. Thus, it is now clear that peroxiredoxin relays do exist and operate in mammalian cells and therefore may generally explain the high sensitivity and target specificity of protein thiol oxidation. However, the next crucial question is: how common are peroxiredoxin redox relays in mammalian cells? Are they rather exceptional or do they apply to most cases of H2O2 sensing and signalling? For example, are protein tyrosine phosphatases, classical examples of protein redox regulation, also oxidized by peroxiredoxins? Answering these questions about the overall significance of peroxiredoxin-based redox relays is the first specific aim for the next funding period.

Within the first funding period we also discovered a molecular mechanism that explains how a protein can be directly oxidized by H2O2, in a highly specific manner, yet without the need for a mediator: the glycolytic enzyme GAPDH was found to harbour a highly conserved peroxidase-like catalytic mechanism that facilitates 'self-oxidation'. However, there is significant evidence that enzymes with 'built in' self-oxidation mechanisms are exceptional and not representative of the majority of redox-regulated proteins. Importantly, our results directly refute the wide-spread assumption that oxidation sensitivity of thiols can be explained or predicted by their nucleophile properties, in particular pKa. Recognizing that S-sulfenylation is a diagnostic feature of direct thiol oxidation, and thus a molecular marker that allows to detect direct thiol oxidation events, it will be important to clarify the extent of intracellular protein S-sulfenylation under different signaling and stress conditions. Answering questions about the prevalence of S-sulfenylation on redox-regulated proteins is the second specific aim for the next funding period.

TP 15 Frauke Melchior - SUMOylation in Oxidative Stress Response

Project leader's home page

To sense - and defend against - oxidative stress, cells depend on signal transduction cascades that involve redox-sensitive proteins. We previously identified SUMO (Small Ubiquitin related modifier) enzymes as downstream effectors of reactive oxygen species (ROS). Hydrogen peroxide transiently inactivates SUMO E1 and E2 enzymes by inducing a disulfide bond between their catalytic cysteines (Bossis and Melchior, Mol Cell 2006). How important their oxidation is in light of many other redox-regulated proteins has however been unclear. During the first funding period we could demonstrate that redox-regulation of SUMO enzymes in response to oxidative stress is an essential process. Our findings indicate that that one essential function of this pathway is the ability of cells to activate the ATM-dependent DNA damage response pathway in response to oxidative stress (Stankovic-Valentin et al., EMBO 2016). During the second funding period, we plan to dissect the molecular mechanisms that link SUMO enzyme oxidation to ATM-dependent processes and following the idea that SUMO E1 and E2 enzymes are components of specialized complexes with roles in cytoplasmic and/or nuclear signal transduction.

TP 16: Martina Muckenthaler - Surveillance of systemic iron levels in response to stress by the hepcidin/ferroportin regulatory system

Project leader's home page

Iron is an essential nutrient and a potential toxin. It is a critical component of heme groups, iron-sulfur cluster- containing proteins and of enzymes involved in mitochondrial respiration and DNA synthesis and thus plays an important role in cellular metabolism, survival and proliferation. However, free excess iron is toxic as it reacts with oxygen to generate reactive oxygen species (ROS) which trigger cell damage. The organism protects itself from toxic iron excess by activating the hepatic expression of the circulating peptide hormone hepcidin, the master regulator of systemic iron homeostasis. Hepcidin binds its receptor ferroportin, the only known iron exporter to diminish iron efflux into the blood stream. Our recent work uncovered the importance of the hepcidin-ferroportin interaction in preventing fatal iron overload (Altamura et al., 2014) and within SFB1036 we identified novel pathways that control hepcidin and ferroportin expression.

The acute phase response is an important stress condition, in which the organism protects itself from iron excess in the blood stream. Inflammatory cytokines activate the expression of the iron hormone hepcidin, to sequester iron in reticuloendothelial macrophages. Thus, hepcidin is a crucial effector of inflammatory hypoferremia. However, our recent findings within SFB1036 challenge the exclusive role of hepcidin in controlling the hypoferremic response and suggest that rapid hepcidin-independent ferroportin downregulation in the major sites of iron recycling may represent a first-line response to restrict iron availability during the acute phase response. We showed that acute inflammatory stress, mediated by Toll-like receptors 2 and 6 (TLR2 and TLR6) induces ferroportin-controlled hypoferremia in mice injected with TLR ligands. In future studies we will analyze the cellular pathways that control ferroportin levels in response to inflammatory stress and investigate whether different stress conditions (e.g. iron or redox stress) impact on the control of ferroportin levels. Cell-autonomous effects in macrophages due to ferroportin regulation will be differentiated from systemic, hepcidin-controlled ferroportin regulation by analyzing macrophages isolated from a mouse model with a disrupted hepcidin/ferroportin circuitry. Finally we will ask the question how iron sequestration in macrophages in response to decreased ferroportin expression affects the macrophage phenotype (e.g. macrophage polarization). Overall, this work will provide insight into the regulation of the most important surveillance mechanisms of iron levels in response to stress.

TP 17 Michael Brunner - Light- and heat-shock related stress response in Neurospora

Project leader's home page

The major aim of this proposal is the analysis of mechanisms that ensure genome-wide quantitative and temporal coordination of transcriptional responses to stress signals. As a convenient model system we use predominantly the filamentous fungus Neurospora crassa, which rapidly forms asexual spores in response to exposure to light. We will analyze the light response of Neurospora as a paradigm of a stress-signal induced gene expression program. Considering that most light-induced mRNAs are transcribed in only a few copies per gene we are particularly interested in mechanisms assessing the number of transcripts per gene to reproducibly ensure precision of the transcriptional stress response. We have shown in preliminary work that this is in part achieved by a transcriptional memory based on a transcription-dependent physical inactivation of promoters. This negative memory appears to be implemented by chromatin after a number of mRNAs copies characteristic for the promoter have been transcribed (transcriptional burst) and, renders the gene refractory towards further stimulation. The major aim of this project is to uncover the underlying mechanisms. We will particularly focus on transcription induced long-lasting alterations in the chromatin state of refractory promoters and genes. In addition to light-inducible promoters we will also analyze burst size and refractoriness of heat-inducible promoters in Neurospora. Furthermore, to exploit the potential of yeast genetics, we will setup a heat-shock based screening system in yeast to identify genes and analyze components that determine transcription burst size and transcription-induced refractoriness of promoters.

TP 18: Sylvia Erhardt (new project leader) - Cellular stress response at centromeres

Project leader's home page

Chromosome segregation during mitosis is a vital process of a proliferating cell that needs to be carefully controlled by surveillance mechanisms to avert segregation defects, aneuploidy and resulting developmental defects, and diseases such as cancer. Therefore, the process of mitosis needs to react to cellular stress, especially genotoxic stress to ensure that replication errors or DNA damage defects are not inherited into the next generation. This is controlled by checkpoints that stop the progression of the cell cycle to allow the repair of any lesion before cells are allowed to enter mitosis. Chromosome segregation is mediated by the centromere, a defined chromatin region on every chromosome. Centromeres do not rely on a specific DNA sequence for their spatial determination but rather on epigenetic mechanisms. The epigenetic determinants of centromeres include the centromere-specific histone variant CENP-A and, more recently discovered, noncoding RNAs. Even though the regulation and centromeric loading of CENP-A and the function of noncoding RNA are essential for genome stability, we know little about the regulation of centromeres during genotoxic stress. The overarching aim of this project is to determine the role of centromeres during genotoxic stress. We have strong evidence that there are regulatory mechanisms in place to specifically regulate CENP-A during DNA damage repair (DDR) and -based on unpublished results - we also propose a regulatory role for non-coding RNA.

We have recently shown that CENP-A deposition is carefully regulated in the presence of DNA damage. The histone fold protein and subunit of chromatin remodelling complexes CHRAC14 binds to CENP-A specifically during genotoxic stress and either prevents loading or evicts CENP-A from sites of DNA damage. Flies that lack CHRAC14 have a reduced tolerance to DNA damage and a defective G2-M checkpoint. In the first aim of this proposal, we will explore the role of CHRAC14 in DDR and CENP-A loading from a mechanistic point of view. Recently identified common interaction partners of CHRAC14 and CENP-A will be in the focus of this proposal.

The second aim of this proposal is designed to characterize the role of centromeric non-coding RNAs in DDR. We will identify protein complexes associated with the centromere-associated RNAs that are formed during genotoxic stress situations where DNA damage checkpoints are activated and cells cannot undergo mitosis. We will furthermore characterize potential RNA processing mechanisms that are specific to DDR. In addition, we will systematically identify and characterize RNAs that associate with centromeres during stress induction only. We will elucidate their function in centromere biology and chromosome segregation. These experiments are designed to identify and characterize novel surveillance mechanisms of non-coding RNAs under genotoxic stress.

TP 19: Claudio Joazeiro (new project leader) - Biology and mechanisms of the translational stress signalling protein, Rqc2

Project leader's home page

Ribosomes can halt translation elongation and become stalled under various circumstances, such as upon translation of mRNA templates lacking stop codons, containing sequential rare codons, or encoding homopolymeric Lys tracts. If left unchecked, ribosomal stalling can be deleterious - whether by reducing the pool of available ribosomes, or by generating aberrantly-synthesized polypeptides as potentially toxic byproducts. Accordingly, cells have evolved systems that recognize stalled ribosomes, degrade the associated aberrant mRNAs and nascent polypeptides, recycle ribosomal subunits, and elicit a translational stress response. The importance of these systems is evidenced by the fact that they are present from bacteria to mammals, and by the harmful effects caused by their inactivating mutations in various organisms.

A key player in translational surveillance is the Ribosome-associated Quality Control (RQC) complex, which targets large ribosomal subunits stalled with nascent chains. In eukaryotes, the RQC is minimally composed of Ltn1 (Listerin in mammals), Rqc1, and Rqc2/Tae2 (NEMF in mammals) subunits. Rqc2 is the central subject of this proposal. According to the prevailing model, Rqc2 has three main functions in RQC: (1) Rqc2 directly recognizes the aberrant structure of a stalled 60S ribosomal subunit and prevents reassociation of 40S subunits; (2) Rqc2 stabilizes binding of Ltn1, the E3 ubiquitin ligase which, together with Rqc1, dictates nascent chain proteasomal degradation; and (3) Rqc2 can modify 60S-stalled nascent chains with a heterogeneous COOH-terminal extension containing Ala- and Thr residues, known as the 'CAT tail'. This 'CATylation' activity of Rqc2 was shown to correlate with activation of translational stress signalling via Heat Shock Factor 1 (Hsf1). More recently, we have found that CATylation promotes formation of insoluble nascent chain aggregates. Stalled nascent chain CATylation and aggregation were observed either in cells deficient for Ltn1 or when utilizing Lysine-less reporters that are expected to be targeted less efficiently by Ltn1. We hypothesize that Rqc2 primarily modifies RQC substrates that are resistant to ubiquitylation.

Overall, our aim is to elucidate the basis of substrate targeting by Rqc2 and its orthologs. Knowledge derived from the proposed research will be critical for a more complete understanding of RQC biology and mechanisms, and, in more general terms, of strategies utilized for protein quality control and stress signalling. In the longer term, our studies are also expected to contribute towards elucidating the relationship between RQC defects and neurodegeneration.

TP 20: Carmen Nussbaum-Krammer (new project leader) - Combating protein misfolding at the cellular and organismal level

Project leader's home page

In protein misfolding disorders (PMDs), such as Alzheimer's and Parkinson's disease (PD), specific proteins form alternative pathogenic conformations that can grow from small oligomeric assemblies to large amyloids. PMDs exhibit a complex pathology involving non-cell autonomous proteotoxic effects and progressive spreading of protein misfolding. Our previous work established a metazoan prion model in C. elegans using NM, the prion domain of the yeast prion protein Sup35. We showed that intercellular spreading of NM is mediated by autophagic uptake from the cytosol into membrane-bound vesicles mediating its exchange between cells and tissues. By investigating the prion-like characteristics of disease-associated proteins in C. elegans, we recently found that α-synuclein is spreading in a similar fashion as the prion domain. It has been reported that intercellular transmission of misfolded α-synuclein contributes to the propagation of disease pathology in PD, but the exact pathways involved are insufficiently understood. A better understanding of these pathways might reveal new therapeutic strategies to combat PD. Within this SFB, we aim to identify pathways that regulate prion-like propagation of α-synuclein within and between cells. By performing organism-wide and tissue-specific RNAi screens in C. elegans, we aim to identify novel genes and pathways that are involved in the uptake, sorting and transport of α-synuclein into transmissible vesicles and their transfer between cells. We will also examine the influence of mutations and posttranslational modifications as well as the role of the Hsp70 disaggregation machinery in α-synuclein propagation and spreading.

TP 21: Gislene Pereira (new project leader) - Signaling integration in the surveillance mechanisms of the spindle position checkpoint

Project leader's home page

The spindle apparatus is a complex proteinaceous assembly formed by microtubules and associated proteins that fulfil essential functions during mitosis. These include the physical segregation of chromosomes as well as mitotic spindle orientation, which in asymmetrically dividing cells is critical for cell fate determination. In contrast to other types of damage control, in which a defective protein or organelle is eliminated (e.g. by degradation), cells cannot just eliminate a defective spindle; otherwise, mitosis would not occur. To avoid error in chromosome segregation upon spindle damage, cells have developed cellular surveillance mechanisms (also named checkpoints) that constantly monitor spindle function. Mitotic checkpoints detect spindle errors and transmit a hold-signal to the cell cycle machinery in order to stop cell cycle progression until defects are corrected. In budding yeast, the spindle position checkpoint (SPOC) is a surveillance mechanism that halts mitotic exit and cytokinesis (cell division) when the mitotic spindle fails to orient along the mother to daughter cell polarity axis. In the absence of the SPOC, cells with a misaligned spindle undergo cell division even though chromosome segregation into the daughter cell is disrupted. In cells with damaged spindles SPOC deficiency causes chromosome mis-segregation (genome instability), which ultimately leads to cell death.

The SPOC inhibits the activation of the mitotic exit network, which is a centrosome-based signaling cascade that drives cells out of mitosis. The most downstream SPOC component is the GTPase activating protein (GAP) complex composed of Bub2 and Bfa1, which directly inhibits activation of the mitotic exit network. How SPOC senses spindle defects on a molecular level and efficiently blocks cell cycle progression are still important unsolved questions. Previous work indicated that the mother cell specific kinase, Kin4, regulates Bub2-Bfa1 localisation to the yeast centrosome in response to spindle misalignment. However, our preliminary data indicated that depending on the genetic background, Kin4 becomes dispensable for SPOC in cells with misaligned spindles; yet cells still require Bub2-Bfa1 for cell cycle arrest. Therefore, we postulate that multiple signaling pathways contribute to the spindle defect sensing and converge on Bub2-Bfa1 regulation in a Kin4-independent manner.

The overall aim of this project is to provide a comprehensive evaluation of SPOC functioning upon spindle perturbations caused by microtubule depolymerising drugs or by defective spindle components. For this, we will dissect the molecular mechanisms that control Bub2-Bfa1 activity and localisation independently of Kin4.

TP 22: Irmgard Sinning (new project leader) - Molecular mechanisms of ribosome associated enzymes in stress responses

Project leader's home page

Ribosomes synthesize proteins directed by the mRNA, and already during translation, the emerging polypeptide chain is subject to enzymatic modification, to targeting factors for localization, and to chaperones for de novo protein folding. These "early" factors are key to cellular homeostasis and ¬allow adaptation and tolerance to stress conditions. N-terminal acetylation (NTA) is the most common modification of cytosolic proteins and it has been shown that the acetylome of the cell is not static, but regulated upon cellular stress. The aim of this proposal is to dissect the molecular mechanisms of the major N-acetyl transferases (NatA, NatB and NatC) and their role protein quality control under stress conditions.

We will characterize NatA, NatB and NatC structurally and functionally. Our goal is to integrate atomic knowledge into a detailed understanding of the function of Nats, their dynamic interactions and adaptation to cellular stress. We will exploit different model organisms including Neurospora crassa, Arabidopsis thaliana and Homo sapiens to understand the role of Nat regulation by specific interaction partners upon cellular stress and their impact on NTA levels. Starting from structural analyses, we want to unravel the molecular basis of stress-induced regulation of NTA activity by NatA in vitro and in vivo starting with A. thaliana and N. crassa. Additional binding partners of NatA, for example Naa50, have been suggested to interact with the NatA complex under specific cellular conditions. We want to understand the interplay between these factors and NatA during cellular stress using an integrated structural biology approach in combination with biochemical and in vivo characterization. Taken together, we aim to dissect the molecular mechanisms underlying the regulation of the three major Nats upon stress by a combination of structural, biochemical and in vivo characterization.

TP 23: Aurelio Teleman (new project leader) - Surveillance of amino acid levels by mTORC1

Project leader's home page

The kinase mTORC1 is a key regulator of cellular growth and metabolism. When active, mTORC1 promotes cell growth by activating anabolic pathways (e.g. lipid biosynthesis, nucleotide biosynthesis, and translation), and by inhibiting catabolic processes (e.g. autophagy). Because of this, mTORC1 activity is carefully regulated by sensing mechanisms that respond to cellular stresses such as amino acid removal, low cellular energy, or hypoxia. Since many of the processes regulated downstream of mTORC1 (such as translation) affect the levels of metabolites that are upstream of mTORC1 (such as amino acids), mTORC1 can be considered a surveillance rheostat that is embedded in negative feedback loops. In the present application we propose to study how mTORC1 surveys cellular amino acid levels via the negative feedback loop whereby depletion of cellular amino acids leads to reduced mTORC1 activity, which dampens cellular translation and activates autophagy, thereby leading to restoration of amino acids and hence reactivation of mTORC1. This homeostatic mechanism couples cellular translation rates to amino acid availability, thereby ensuring protein synthesis quality control in the face of limiting amino acids.

Although amino acids are known to regulate mTORC1, and recent discoveries have elucidated signalling mechanisms involving the Rag GTPases by which the presence of amino acids activates mTORC1, comparatively little is known about how amino acid removal leads to mTORC1 inhibition. To discover novel components in this sensing pathway, we performed a genome-wide RNAi screen in mammalian cells, looking for genes that affect the ability of mTORC1 to become inactive upon amino acid withdrawal. In this manner, we identified 7 genes previously not known to participate in this process. Loss of these genes causes mTORC1 to remain aberrantly active when amino acids are removed. In addition, we have identified 7 novel mTORC1 regulators that regulate mTORC1 activity under standard cell-culture conditions, and potentially affect mTORC1 activity in response to other stresses. The goal of this project is to study two of these genes. We plan (1) to understand the role of these genes in amino acid sensing, (2) to uncover the molecular mechanism by which these genes affect mTORC1 activity, and (3) to uncover the physiological and/or pathophysiological relevance of this regulation.

Z1: Thomas Ruppert - Mass spectrometry and Proteomics

Project leader's home page

Stressful conditions elicit in cells a highly complex response that is characteristic for the kind of environmental impact and for the physiological state of the cell. This response in general involves qualitative and quantitative changes in protein content of a cell as well as in posttranslational modifications (PTMs) of specific proteins. Monitoring such changes on a global level is only possible by high resolution mass spectrometry.

The Z1 project is provided by the core facility for mass spectrometry and proteomics at the ZMBH. Our mission is to provide state-of-the-art mass spectrometry services to the research groups of the Heidelberg campus and we establish and improve methods in close collaboration with them. Over the years we analyzed samples for more than 90 research groups, several thousand analyses each year.

That allowed to improve and extend the repertoire of workflows in the field of protein characterization, which means low complexity, but tiny amounts of sample. Now we have workflows for a variety of posttranslational modifications, for identification of protein interaction partners and their interaction sites after chemical cross-linking.

Our main focus in the first funding period, however, was to adapt our workflows for the analysis of whole proteomes for which sample amount is normally not limited but complexity is extremely high. In the meantime a variety of quantitative proteomics workflows has been established using stable isotope labeling in cell culture - SILAC - or after chemical labeling on the peptide level. In addition, label free quantification is frequently used. Quantitative phosphoproteomics was established with Elmar Schiebels group using IMAC enrichment and very recently we established a glycoproteomics workflow with Sabine Strahls group using 2m long lectin columns for enrichment of O-mannosylated peptides. With increasing complexity of MS data sets it is important that in addition to our data interpretation users can do the data interpretation by their own, but with their knowledge about the biology behind. Therefore, we offer hands on training in data interpretation for the commonly used workflows.

In the next funding period we expect this demand for mass spectrometric analysis to grow even further. To cope with this, we will not only increase our sample throughput but also our quality standards. One major focus will be an increased coverage of identified and quantified proteins by improving all levels of our proteomics workflows. We will, for example, offer extended support for our users during early stages of sample preparation, like protein extraction, and implement self-packed HPLC columns, which offer improved peptide separation power. Another key factor will be the addition of the new LC-QExactive mass spectrometry system that will increase our performance on all levels and render new methods possible, like data independent acquisition mass spectrometry (DIA) with its improved quantitation capabilities. Last but not least, we will extend the offer for training opportunities for Core Facility users to better bridge the gap between mass spectrometry specialists and biological scientists.

Former SFB projects/members:

TP 2: Brian Luke - The regulation of telomere transcription and structure to promote genome stability (home page)

TP 4: Elmar Schiebel - Release of hCdc14B phosphatase from the nucleolus in response to DNA damage (home page)