03 Fakultät Chemie

Permanent URI for this collectionhttps://elib.uni-stuttgart.de/handle/11682/4

Browse

Now showing 1 - 20 of 64

Open Access
The active subunits of the 20S Proteasome in Saccharomyces cerevisiae : mutational analysis of their specificities and a C-terminal extention
(2008) Estiveira, Rui José Cabrita; Heinemeyer, Wolfgang (PD Dr.)
The proteasome is a large multi-subunit complex ubiquitous in eukaryotes and archaebacteria. It contains proteolytic subunits that function simultaneously to digest protein substrates into oligopeptides. In eukaryotic cells, it is involved in the removal of abnormal, misfolded or incorrectly assembled proteins, but additionally it has regulatory functions. For example it is responsible for the degradation of cyclins in cell-cycle control and for the destruction of transcription factors or metabolic enzymes in metabolic adaptation. Finally, the proteasome is also involved in MHC (major histocompatibility complex) class I mediated cellular immune response. These cellular functions are linked to an ubiquitin- and ATPrequiring protein degradation pathway involving the 26S proteasome whose proteolytic core is formed by the 20S proteasome. The 20S proteasome has a cylindrical shape and is composed of four rings, each formed by seven α- or seven β-subunits and stacked in the order αββα. In eukaryotic cells, the 20S proteasome is composed of two copies of 14 different subunits, 7 distinct α-type and 7 distinct β-type subunits. Only three of the β-type subunits are proteolytically active and have N-terminal threonine residues acting as nucleophiles. They differ in their major specificities: β5/Pre2, β2/Pup1 and β1/Pre3 are classified as having "chymotrypsin-like", "trypsin-like" and "peptidylglutamyl peptide hydrolysing" (PGPH or caspase-like) activities, respectively. This classification is based on the preferred amino acid residues found at the site of hydrolysis in peptide or protein substrates. These three active β-type subunits have a fixed location in the proteasome, with the two β5/Pre2 copies separated from the clustered β2/Pup1 and β1/Pre3 subunits. In yeast a hierarchy of individual subunit activities for proteasomal function was established: β5/Pre2>> β2/Pup1 > β1/Pre3. Part of this work aimed at clarifying whether this hierarchy is solely dependent on the specificities or whether topological conditions lead to the dominance of the β5/Pre2 activity over the others, which could involve inter-subunit communication mediated by interjacent inactive β-subunits. Stepwise site-directed mutagenesis of key residues forming the substrate binding pockets was used to swap specificities between the yeast β5/Pre2 and β1/Pre3 active sites. Consequences of these mutations were then analysed in regard to maturation of the modified subunits, their specificities towards peptide substrates diagnostic for chymotrypsin-like and PGPH activity and changes in their rank in the hierarchy of functional importance. By mutating the key residue methionine 45 into an arginine, the β5/Pre2 was able to mature and showed some PGPH activity. Combinations with mutant strains, having the other active subunits inactivated, revealed that Pre2 lost its functional dominance. When other key residues were additionally replaced by those present in β1/Pre3 (A20T, V31T, I35T), instead of an further increase in PGPH activity, the β5/Pre2 subunit showed an overall decrease in activity. An unexpected exception was the pre2-M45R-I35T mutant with a strong increase in chymotrypsin-like activity. Maturation of the Pre2 subunit occurred normally like in wild-type in all combinations of mutations tested. When the residue arginine 45 was mutated into a methionine in Pre3, this subunit lost any detectable peptidase activity. Attempts to stabilise the methionine 45 by introducing strategic point mutations at residue 52 were unsuccessful. Additional alterations in the substrate binding site of β1/Pre3 (T20A, T31V, T35I) completely abolished the autolytic maturation and thus any gain of activity. Strains lacking both the Pre3 propeptide and Nα-acetyltransferase were used to confirm that the mutations result in activity loss, even when the autolytic removal of the propeptide was not required. In a second project, the role of the long C-terminal extension of the yeast β2/Pup1 subunit was examined. This 37 amino acid structure embraces the β-ring neighbor subunit β3/Pup3 and reaches the next subunit β4/Pre1. It also contacts β7/Pre6 of the opposite β-ring. Mutations of residues that could loosen contact to the surface of β3/Pup3 (Y204A, R208G and T211A) where without effect. Complete deletion of this extension or truncation of 25 residues was lethal and deletion of the last 20 amino acids caused a strong cell growth defect. When replacing the last 20 amino acids of this C-terminal extension by a FLAG tag, the growth phenotype was lost. The lethal mutations were over-expressed in wild type strains, but the mutated β2 subunits did not incorporate into proteasomes. This indicates that removal of the distal half from the β2/Pup1 C-terminal extension impedes the integration of this subunit during early assembly stages of the 20S proteasome.
Open Access
Analysis of SNAREs, Arf1p and regulators in intracellular transport
(2007) Schindler, Christina; Wolf, Dieter H. (Prof. Dr.)
Vesicular transport is an essential process allowing communication between different organelles in an eukaryotic cell. The small GTPase Arf1p regulates the generation of coated vesicles from donor organelles at many different levels of the secretory pathway. Arf1p cooperates with GTPase-activating proteins (GAPs) to hydrolyze the bound GTP, which subsequently induces the shedding of the proteinacious vesicle coat. Only then the vesicle is capable to undergo SNARE-mediated (soluble NSF attachment protein receptor) fusion with a target organelle to deliver its content. Previously, our lab showed that ArfGAPs can catalytically induce an altered conformation in vesicle SNARE proteins (v-SNAREs). The SNARE in the altered conformation is able to interact with Arf1p. Thus, the uptake of v-SNAREs in budding vesicles is facilitated. The current study extends the previous results to target membrane SNAREs (t-SNAREs) and shows that the ArfGAP-induced conformation enhances the formation of SNARE complexes. SNARE complex formation is an essential step during membrane fusion. Thus, the altered SNARE conformation is not only important during vesicle generation, but also for the consumption of transport vesicles. This let us speculate that ArfGAP proteins might act in a chaperone-like function on SNARE proteins. We were also able to show that SNARE proteins in the ArfGAP-induced SNARE conformation can interact with Sec17p and Sec18p, the yeast homologs of alpha-SNAP (soluble NSF attachment protein) and AAA-ATPase NSF (N-ethylmaleimide-sensitive factor). Both factors play a key role in membrane fusion by resolving cis-SNARE complexes. Data from this study indicate that the ATPase Sec18p has a second function during vesicle fusion as it can displace Arf1p from SNAREs. This process requires neither ATP-hydrolysis nor Sec18p´s co-factor Sec17p. Residual coat on the vesicle could still be required during an initial contact between the two fusing membranes. However, it would likely obstruct the final fusion event. Thus, Sec18p might be able to remove residual coat from a vesicle and allow final fusion to take place. After one round of vesicle fusion, v-SNAREs are recycled back to the donor compartment. Data presented in this study show that Snc1p and Snc2p, which are v-SNAREs involved in yeast exo- and endocytosis, interact physically and genetically with the ArfGAP Gcs1p, and that this interaction is essential for recycling of SNAREs via the trans-Golgi-Network (TGN) and endosomes. Furthermore, we were able to show that the COPI vesicle coat, which has been implicated in retrograde trafficking from the Golgi apparatus to the endoplasmic reticulum (ER) as well as in retrograde traffic within the Golgi apparatus, plays also an important role in the recycling of the post-Golgi SNAREs Snc1p and Snc2p. Besides Arf1p´s role in the formation of retrograde directed vesicles, Arf1p also participates in anterograde trafficking from the TGN, which is thought to be the main sorting station for anterograde cargo in an eukaryotic cell. In an attempt to identify new interactors of the small GTPase Arf1p, we found a novel fungi-specific protein family: the ChAPs (for Chs5p and Arf1p-interacting proteins). The ChAP family of proteins consists out of four members: Bch1p, Bch2p, Chs6p and Bud7p. These factors at least partially localize to the TGN, dependent on their interaction with Chs5p. Chs5p has been previously reported to be essential for the delivery of chitin-synthase III (Chs3p) to the yeast bud neck-region; a process Arf1p is also implicated in. We are able to show that Arf1p, Chs5p and individual members of the ChAP family interact genetically and physically. Arf1p, Chs5p and the ChAP proteins form high molecular complexes that contain the potential cargo Chs3p. Based on our findings, we suggest that the ChAP proteins are required for the transport of certain cargo in specialized transport vesicles. The ChAPs might function as cargo receptors, coat adaptors or even as novel coat. Altogether, this work has highlighted new interactors for the small GTPase Arf1p and new modes of function for its regulatory GAP proteins.
Open Access
Apoptosis in the yeast saccharomyces cerevisiae : a novel cell death process regulated by the ubiquitin-proteasome system
(2001) Ligr, Martin; Wolf, Dieter H. (Prof. Dr.)
Apoptosis is co-regulated by the conserved family of Bcl-2-related proteins, which includes both its agonists (Bax) and antagonists (Bcl-XL). A mutant strain of the yeast Saccharomyces cerevisiae has been shown to express all morphological signs of apoptosis. Overexpression of Bax is lethal in S. cerevisiae, whereas simultaneous overexpression of Bcl-XL rescues the cells. We report that overexpression of mammalian Bax in a S. cerevisiae wild type strain triggers morphological changes similar to those of apoptotic metazoan cells: the loss of asymmetric distribution of plasma membrane phosphatidylserine, plasma membrane blebbing, chromatin condensation and margination, and DNA fragmentation. Simultaneous overexpression of Bcl-XL prevents these changes. We demonstrate that Bax triggers phenotypic alterations in yeast strongly resembling those it causes in metazoan apoptotic cells. Oxygen radicals are important components of metazoan apoptosis. Oxygen radicals accumulate in yeast cells overexpressing Bax, whereas radical depletion or hypoxia prevents apoptosis. This suggests that the generation of oxygen radicals is a key event in the ancestral apoptotic pathway and offer an explanation for the mechanism of Bax-induced apoptosis in the absence of any established apoptotic gene in yeast. I have identified the yeast gene STM1 in an overexpression screen for new proteasomal substrates. Stm1 is a bona fide substrate of the proteasome. It is localized in the perinuclear region and is required for growth in the presence of mutagens. Overexpression in cells with impaired proteasomal degradation leads to cell death accompanied with cytological markers of apoptosis. Cells lacking Stm1 display deficiency in the apoptosis-like cell death process induced by treatment with low concentrations of H2O2. I suggest that Stm1 is involved in the control of the apoptosis-like cell death in yeast.
Open Access
Biochemical characterisation of tRNA-Asp methyltransferase Dnmt2 and its physiological significance
(2014) Shanmugam, Raghuvaran; Jeltsch, Albert (Prof. Dr.)
Methylation of tRNA plays important roles in the stabilisation of tRNAs and accurate protein synthesis in cells. In eukaryotes various tRNA methyltransferases exist, among them DNMT2 which methylates tRNAAsp at position C38 in the anticodon loop. It is also called tRNA-aspartate methyltransferase 1 (Trdmt1) and the enzyme is highly conserved among eukaryotes. In this work, I investigated the mechanism of DNMT2 interaction with tRNAAsp, characterised the function of the only prokaryotic Dnmt2 homolog found in G. sulfurreducens and studied the physiological importance of the C38 methylation of tRNAAsp in mammalian cells. The molecular details of the interaction of DNMT2 and tRNAAsp are unknown due to lack of the co-crystal structure. Here, I characterised the important residues in DNMT2 required for the tRNA binding and catalysis. By site-directed mutagenesis of 20 conserved lysine and arginine residues in DNMT2, I show that 8 of them have a strong effect on the catalytic activity of the enzyme. They map to one side of the enzyme where the catalytic pocket of DNMT2 is located. The binding of most of the mutant enzymes to tRNA was unaffected suggesting a role of these residues in transition state stabilisation. Manual docking of tRNAAsp into the surface cleft decorated by the 8 residues suggested that DNMT2 interacts mainly with the anticodon stem/loop of tRNAAsp. In my second project, I characterised the function of Dnmt2 homolog found in G. sulfurreducens (GsDnmt2). Here, I show that GsDnmt2 methylates tRNAGlu more efficiently than tRNAAsp. I also report the molecular basis for the swapped substrate specificity of GsDnmt2 and show that the variable loops of G.sulfurreducens tRNAAsp and tRNAGlu of eukaryotes contain a -GG- dinucleotide which is not preferred by Dnmt2. Exchange of the variable loop of mouse tRNAAsp to that tRNAGlu led to dramatic decrease in the activity of human DNMT2. This identifies the variable loop of tRNA as a specificity determinant in the recognition by Dnmt2. In my final project, I investigated the physiological importance of the tRNAAsp C38 methylation in aminoacylation and cellular protein synthesis. Here, I report that C38 methylation enhances the rate of aspartylation on tRNAAsp by 4-5 folds. Concomitant with this, a decrease in the charging levels of tRNAAsp was observed in Dnmt2 knockout MEF cells, which also showed a reduced efficiency in the synthesis of proteins containing poly-Asp sequences. A gene ontology searches for proteins with poly-Asp sequences showed that a significant number of these proteins are associated with transcriptional regulation and gene expression functions. With this I propose that the mild phenotype observed with the Dnmt2 KO cells under stress condition could be correlated to a disregulation of protein synthesis.
Open Access
Biochemical investigation of the substrate specificity of protein methyltransferases and the identification of novel substrates
(2016) Kusevic, Denis; Jeltsch, Albert (Prof. Dr.)
Posttranslationale Proteinmodiﬁkationen (PTMs) sind wichtig, um verschiedene Proteinfunktionen, wie z. B. Lokalisation, Aktivität, Stabilität und Protein-Protein Interaktionen zu regulieren. In Proteinen können viele Aminosäuren methyliert werden, darunter auch Lysin, Arginin und Glutamin. Methylierungen sind auf vielen verschieden Protein zu ﬁnden, jedoch sind Histonproteine die bedeutendsten. Die Histonmethylierung beeinﬂusst die Chromatinstrukur und spielt eine große Rolle in der Regulation der Transkription. Die Enzyme, die für den Transfer von Methylgruppen auf die Proteine zuständig sind, werden Protein Methyltransferasen (PMTs) genannt. Sie sind sehr speziﬁsch und methylieren immer nur eine Art von Aminosäuren. Dabei zeigt die schnell steigende Anzahl an Berichten über die Methylierung von Proteinen, dass die Methylierung als posttranslationale Modiﬁkation in den letzten Jahren immer mehr an Bedeutung gewinnt. In dieser Doktorarbeit wurde die Substratspeziﬁtät dreier unterschiedlicher Protein Methyltransferasen untersucht, und zwar von HEMK2, einer Glutamin Methyltransferase, sowie von NSD2 und Clr4, zwei Protein Lysin Methyltransferasen (PKMTs). Die Glutamin Methyltransferase HEMK2 methyliert Q185 des Terminationsfaktors eRF1 (eukaryotic translation release factor 1), der für die Termination der Peptidsynthese und für die Hydrolyse der Polypeptidkette von der tRNA am Ribosom verantwortlich ist. Zur Bestimmung der Substratspeziﬁtät von HEMK2 wurde die Aminosäuresequenz von eRF1 als Vorlage verwendet und die erhaltenen Daten zeigen, dass das Substrat für eine Methylierung ein G-Q-X3-K Sequenzmotiv besitzen muss. Eine Suche nach dieser Sequenz in einer Proteindatenbank ergab, dass mehrere humane Proteine dieses Sequenzmotiv besitzen. Von diesen identiﬁzierten Substratkandidaten wurden 125 von HEMK2 auf Peptidebene methyliert. Außerdem konnte gezeigt werden, dass von diesen 125 Kandidaten 16 auf Proteinebene methyliert werden. Zuletzt wurde eine Methylierung der „Chromodomain helicase DNA binding protein 5“ (CHD5) und „Nuclear protein in Testis“ (NUT) Proteine mit Hilfe eines glutaminspeziﬁschen Antikörpers in menschlichen HEK293 Zellen nachgewiesen. NSD2 ist ein Mitglied der „nuclear receptor SET domain-containing“ Enzymfamilie und dimethyliert Lysin K36 von Histon H3 und Lysin K44 von Histon H4. Es wurde gezeigt, dass eine abnormale Expression von NSD2 zu verschiedenen Arten von Krebs und dem Wolf-Hirschhorn Syndrom führen kann. Die Analyse der Substratspeziﬁtät von NSD2 zeigte, dass dieses Enzym die Aminosäuren G33 bis P38 von H3 erkennt. Dabei werden hydrophobe Aminosäuren an den Positionen -1 und +2 (das Ziellysin wird hierbei als Position 0 deﬁniert) bevorzugt. Mit Hilfe des Speziﬁtätsproﬁls von NSD2 wurden mehrere humane Proteine identiﬁziert, die dieses Sequenzmotiv enthalten. Von diesen identiﬁzierten Substratkandidaten wurden 45 durch NSD2 auf Peptidebene methyliert. Des Weiteren wurde gezeigt, dass 3 Kandidaten (ATRX, FANCM und SET8) auf Proteinebene methyliert wurden und zusätzlich konnte die Methylierung von ATRX und FANCM durch NSD2 in HEK293 Zellen nachgewiesen werden. Da die Methylierungen einen erheblichen Einﬂuss auf die Eigenschaften und Funktionen von Proteinen besitzen, müssen weitere Experimente an den neuen Substraten von HEMK2 (CHD5 und NUT) und NSD2 (ATRX und FANCM) durchgeführt werden, um die Auswirkungen auf die biologischen Funktionen der Methylierung herauszuﬁnden. Abgesehen von den menschlichen Enzymen, wurden ähnliche Untersuchungen auch an der Histon Lysin Methyltransferase Clr4, einem SUV39H1-Homolog aus S. pombe, durchgeführt. Clr4 trimethyliert Lysin K9 des Histonproteins H3. Zur Bestimmung des Speziﬁtätsproﬁls von Clr4 wurde die Aminosäuresequenz von H3 (1 - 18) verwendet. Die Ergebnisse zeigten, dass Clr4 speziﬁsch die Aminosäuren der Positionen -1 bis +3 der Zielsequenz erkennt. Zusätzlich wurden 6 neue Peptidsubstrate aus S. pombe identiﬁzieren, die durch Clr4 methyliert wurden. Um die Detektion von Proteinmethylierungen weiter zu verbessern, wurde eine neue radioaktivitätsfreie, Mikrotiter-Untersuchungsmethode entwickelt, die natürlich vorkommende Lese-Domänen anstelle von methylspeziﬁschen Antikörpern zur Erkennung von Methylierungen auf Histonpeptiden verwendet. Es wurde gezeigt, dass diese Methode erfolgreich die Methyltransferaseaktivität bestimmen und für die Suche nach PKMT Inhibitoren verwendet werden kann.
Open Access
Characterization of novel proteins involved in catabolite degradation of fructose-1,6-bisphosphatase in saccharomyces cerevisiae
(2006) Pfirrmann, Thorsten; Wolf, Dieter (Prof. Dr.)
Glycolysis and gluconeogenesis are reciprocally controlled central metabolic pathways in cells. Catabolite degradation of fructose-1,6-bisphosphatase (FBPase) is a key regulatory step, when Saccharomyces cerevisiae cells switch from anabolic gluconeogenesis to catabolic glycolysis. Addition of glucose to cells growing on a non-fermentable carbon source causes FBPase phosphorylation resulting in a decrease of enzymatic activity. This is followed by a proteolytic breakdown of the enzyme via the ubiquitin-proteasome system with a half-life of 20-30 min. In a genome wide screen nine so called gid mutants (glucose induced degradation deficient) defective in proteasome-dependent catabolite degradation of FBPase were identified. Analysis of Gid2 revealed that this protein is a part of a soluble, cytosolic protein complex with a molecular mass of at least 600kDa (Regelmann et al., 2003). The work of this thesis focuses on the analysis of the novel Gid proteins and their possible role in a higher molecular mass protein complex. To be able to detect Gid proteins immunologically, functional chromosomally HA eptitope tagged versions of Gid5, Gid6, Gid7, Gid8 and Gid9 proteins were generated. Using step glycerol gradient centrifugation it could be shown that Gid5/Vid28, Gid7, Gid8 and Gid9 are also components of a higher molecular mass complex of about 600kDa. Gid1/Vid30, Gid/Ubc8 and Gid4/Vid24 exhibit a sedimentation profile of lower molecular mass slightly overlapping with 600kDa aminopeptidase I. Gid6/Ubp14 is only present in its monomeric form. Use of Gid7 as a bait protein in a co-immunoprecipitation experiment, led to the identification of Gid1/Vid30, Gid2, Gid4/Vid24, Gid5/Vid28, Gid7, Gid8 and Gid9 as interacting components. The protein Gid6/Ubp14 is not part of this protein complex. The direct interaction of Gid4 with Gid5 could be shown via the two hybrid method. Expression profiles on ethanol or glucose of Gid1, Gid2, Gid5, Gid6, Gid7, Gid8 and Gid9 were similar. Gid4/Vid24 was not expressed on ethanol but appears when cells are treated with glucose. As found for Gid3/Ubc8 (Schüle et al., 2000), Gid4/Vid24 seems to disappear during incubation on glucose in a time-dependent fashion. Fructose-1,6-bisphosphatase was found to interact with Gid1 and Gid7 protein. As shown for GID2 (Regelmann et al., 2003) deletion of GID1 and GID7 leads to a block in fructose-1,6-bisphosphatase polyubiquitination. This shows that Gid proteins are directly involved in the ubiquitination process which preceeds proteasome degradation. Two discovered short RING domains in Gid2 and Gid9 (ShRING domains) as well as the discovery of 5 WD40 domains within Gid7 suggest a role of the Gid complex as a novel E3 ubiquitin ligase. The targeted mutation of conserved cysteine residues within the shRING domain of Gid2 could support this theory. Biochemical and molecular methods were used to identify the localization of Gid1, Gid6, Gid7 and Gid8. Interestingly all four Gid proteins were found to be localized in the nucleus. The direct interaction of FBPase with these Gid proteins raised the question of whether FBPase itself had a function in the nucleus of the cell or not. To investigate this question GFP-fusions with FBPase were constructed and localisation studies were performed. An increasing signal of FBPase within the nucleus after onset of catabolite degradation gave proof of the existence of this enzyme in the nucleus as well. Several mutants known to have a defect in nuclear import were tested for the catabolite degradation of FBPase. The protein kinaseA pathway was shown to be the signal transduction pathway triggering FBPase degradation. This led to the discovery of novel putative phosphorylation sites within FBPase by bioinformatics.
Open Access
Characterization of the yeast proteins Neo1p and Sjl2p, two highly conserved regulators of phospholipid composition within endosomal membranes
(2004) Wicky John, Sidonie; Singer-Krüger, Birgit (Priv.-Doz. Dr.)
Endocytosis is a key membrane trafficking pathway by which all eukaryotic cells internalize extracellular material as well as portions of the plasma membrane. In the budding yeast Saccharomyces cerevisiae, the major fraction of the material internalized at the plasma membrane is transported through the early and late endosomal compartments to the vacuole. The endosomal compartments are highly dynamic and are connected to the plasma membrane, the late Golgi complex and the vacuole by several routes, which all involve vesicular transport. Thus, the formation of vesicles is essential to maintain the dynamic exchanges between these different compartments. In the present study, I analyzed two yeast proteins, Neo1p and Sjl2p, which are suggested to function as regulators of the phospholipid composition of endosomal organelles and thereby seem to participate in vesicle formation (Neo1p) and in vesicle uncoating (Sjl2p), two processes essential during vesicular membrane transport. Neo1p is an essential member of the Drs2 family of P-type ATPases with proposed function as aminophospholipid (APL) translocases. Genetic and biochemical data in the laboratory of B. Singer-Krüger indicated that Neo1p might function within the endosomal system. Consistent with these findings, I could confirm by indirect immunofluorescence that the major fraction of Neo1p localizes to the endosomal compartments, while a smaller part associates with late Golgi membranes. In agreement with this localization, two temperature-sensitive neo1 mutants were shown to be defective in endocytosis (B. Singer-Krüger's laboratory) and in vacuolar protein sorting (my studies). While these defects were already observed at permissive temperature, at nonpermissive temperature the neo1-ts mutants exhibited additional impairments in membrane transport between the ER and the early Golgi compartment. However, these defects were most likely a consequence of the accumulation of mutant Neo1 proteins in the ER under these conditions. In support of previous results in the laboratory of B. Singer-Krüger, I identified further links between Neo1p and the endosomal proteins Ysl2p and Arl1p. I could show that Neo1p interacts in vivo with Ysl2p and that the subcellular localization and the stability of Ysl2p are affected in the neo1-69 mutant. Furthermore, the subcellular distribution of Arl1p was also found to be impaired in neo1-69 cells at permissive temperature. Based on these findings and the work of B. Singer-Krüger, Neo1p was proposed to act together with Ysl2p and Arl1p in membrane trafficking within the endosomal/late Golgi system. In the second part of this work, the subcellular localization of Sjl2p, a polyphosphoinositide- and 5'-inositide phosphatase of the synaptojanin family, was examined. Based on genetic analyses, Sjl2p has been suggested to participate in early steps of endocytic transport. Here, I determined by indirect immunofluorescence that the Sjl2p-positive punctate structures were distinct from those containing typical early and late endosomal marker proteins and were not sensitive to mutations affecting the structure of either the endosomal compartments or the Golgi complex. Sjl2p did not show a typical plasma membrane staining pattern either. However, Sjl2p colocalized with cortical actin patch components found within clumps that accumulated in cells lacking the two actin-regulating kinases Prk1p and Ark1p. Within these aberrant structures, Sjl2p also colocalized with FM4-64 endocytosed for a short time, suggesting that Sjl2p localized to primary endocytic vesicles that interact with the cortical actin cytoskeleton. This interaction may at least to some extent be mediated by Bsp1p, a binding partner of Sjl2p isolated in B. Singer-Krüger's laboratory, which in my studies was found to be part of the cortical actin cytoskeleton. In cells deficient for the PtdIns(4)P kinase (pik1-ts), the staining pattern of Bsp1p was changed, suggesting that the subcellular distribution of Bsp1p is dependent on the levels of phosphoinositides. Consistent with that, Bsp1p was found to flotate with liposomes containing acidic phospholipids including phosphoinositides. Thus, Bsp1p may act as an adapter that directly connects Sjl2p-containing vesicles to the cortical actin cytoskeleton during early stages of endocytosis via interactions with Sjl2p (B. Singer-Krüger) and with a subset of phospholipids within the vesicle membrane (B. Singer-Krüger and my results). In summary, the results of my PhD thesis brought new insights into the localization and function of Neo1p and Sjl2p within the endosomal system. These data are relevant for future studies to elucidate the precise mechanism of Neo1p and Sjl2p during endocytosis.
Open Access
Charakterisierung der Autophagozytosegene AUT9 und AUT1 in der Hefe Saccharomyces cerevisiae
(2002) Reiche, Steffen; Thumm, Michael (Priv. Doz. Dr.)
Um den Mechanismus der Autophagozytose näher zu untersuchen, werden in dieser Arbeit das Aut9p und das Aut1p in der Hefe Saccharomyces cerevuisiae lokalisiert und näher charakterisiert. Es wird gezeigt, dass AUT9 allelisch mit CVT7 ist. Die Deletion des AUT9-Gens hat keinen Einfluss auf die Vakuolenmorphologie in Saccharomyces cerevuisiae. Das Wachstum auf nicht fermentierbaren Kohlenstoffquellen ist ebenfalls ungestört. Das Aut9p konnte mit Hilfe einer N-terminalen Fusion mit dem grün fluoreszierenden Protein (GFP) in der Hefezelle an punktförmigen Strukturen im Zytosol lokalisiert werden. Die vesikelartigen Strukturen wurden von Noda, Kim et al. (2000) als neues Zellkompartiment identifiziert, und als Präautophagosom bezeichnet. Das Aut1p wurde als lösliches, zytosolisches Protein lokalisiert. Es konnte gezeigt werden, dass die Expression des Aut1p stark hungerinduziert ist.
Open Access
Components and mechanisms of cytoplasmic protein quality control and elimination of regulatory enzymes
(2011) Eisele, Frederik; Wolf, Dieter H. (Prof. Dr.)
Relatively little is known about cytoplasmic protein quality control in eukaryotic cells. After proteins have been translated on ribosomes, they have to achieve their native conformation, get to their place of action and be assembled into protein complexes when indicated. Errors in the protein sequence caused by DNA mutations, mistakes during transcription or translation, as well as folding disorders caused by chemical or physical stress can impair the proper functionality of the cell and evoke diseases. Therefore, it is the task of the cellular protein quality control system to assist proteins while folding into their native conformation, to unfold misfolded proteins and to refold them. Finally, irreversibly misfolded proteins have to be transferred for degradation to the proteolytic systems of the cell, the 26S proteasome or the vacuole (lysosome). The components that are involved in the control of protein folding and in the transfer of misfolded cytoplasmic proteins to the proteolytic systems have been poorly investigated. In this work, novel components of the cytoplasmic quality control system have been discovered by studying mutated variants of carboxypeptidase Y (CPY*), a vacuolar enzyme, which due to deletion of its signal sequence cannot be imported into the endoplasmic reticulum (ER) for further transfer into the vacuole and therefore is permanently located to the cytoplasm of the budding yeast Saccharomyces cerevisiae. Studies investigating ∆ssCPY* (signal sequence deleted CPY*), ∆ssCG* (∆ssCPY* carrying a C-terminal GFP tag) and the corresponding wild-type enzyme ∆ssCPY showed that for proteasomal degradation of these substrates the cytoplasmic chaperone Hsp70 (Heat shock protein) Ssa1, the Hsp40 co-chaperone Ydj1 and the ubiquitin-conjugating enzymes (E2) Ubc4 and Ubc5 are necessary. It could be shown that Ssa1 and Ydj1 are involved in the resolubilization of precipitated ∆ssCG*, in keeping ∆ssCG* in solution and in the transport of ubiquitylated ∆ssCG* to the 26S proteasome. The following study searched for further factors of the cytoplasmic quality control, especially a ubiquitin ligase (E3), which is capable of targeting misfolded cytoplasmic proteins for proteasomal degradation. Yeast mutants were isolated in a genetic screen, which are able to stabilize the fusion protein ∆ssCL*myc (∆ssCPY* C-terminally fused to myc-tagged 3-isopropylmalate dehydrogenase (LEU2myc)) and are therefore able to grow on media lacking leucine. This led to the discovery of the E3 Ubr1. Subsequent investigations revealed that the proteasomal degradation of ∆ssCL*myc is strongly dependent on Ubr1 and that the misfolded substrate physically interacts with this E3. Furthermore, it could be shown that for degradation of ∆ssCL*myc and ∆ssCG* the Hsp110s Sse1 and Sse2 are necessary, probably functioning as nucleotide exchange factors for Ssa1. Besides the degradation of finally misfolded cytoplasmic proteins, the eukaryotic cell utilizes its proteolytic systems to eliminate regulatory enzymes upon changes in the cellular environment. After switching cells from non-fermentable to fermentable media, a key regulatory enzyme in the gluconeogenesis pathway, fructose-1,6-bisphosphatase (FBPase), is ubiquitylated by the Gid-E3 complex and then degraded by the ubiquitin proteasome system (UPS) to allow switching from gluconeogenesis to glycolysis. In a further study we found that for degradation of ubiquitylated FBPase procession by the AAA-ATPase Cdc48 and its co-factors Ufd1 and Npl4 is necessary. This is the first time that for degradation of a native substrate by the UPS a dependency on the Cdc48-Ufd1-Npl4 complex could be shown. In addition, it could be shown that the ubiquitin receptor proteins Dsk2 and Rad23 are also necessary for the proteasomal degradation of FBPase. Before a ubiquitylated substrate of the 26S proteasome is degraded, its ubiquitin chains are cleaved off. The ubiquitin-specific protease Ubp14 cleaves these free chains to single ubiquitin molecules. Cells deleted in UBP14 accumulate ubiquitin chains, which leads to impairment of the UPS dependent protein degradation. In a further study we demonstrated that inhibition of proteasomal degradation by deletion of UBP14 does not occur in the degradation process of all substrates tested. While e.g. UPS dependent degradation of the gluconeogenic enzyme FBPase is impaired in ∆ubp14 strains, degradation of ∆ssCG* is only slightly reduced and degradation of a misfolded substrate of the ER, CPY*HA is not at all affected. This finding suggests that there are several substrate specific pathways to proteasomal degradation, which can be defined by a varying dependency on Ubp14.
Open Access
Cytosolic protein quality control of the orphan protein Fas2, a novel physiological substrate of the E3 ligase Ubr1
(2013) Scazzari, Mario; Wolf, Dieter H. (Prof. Dr.)
Cellular protein quality control (PQC) monitors the proper folding of polypeptides, assembly of protein subunits into protein complexes as well as the delivery of terminally misfolded proteins to degradation. The components of PQC known best at the moment are molecular chaperones and the ubiquitin proteasome system. In contrast to the well-described protein quality control system of the endoplasmic reticulum (ERAD), less is known about how misfolded proteins in the cytosol are recognized and degraded. The cytosolic fatty acid synthase complex (FAS) of Saccharomyces cerevisiae, which is composed of six Fas1- and six Fas2-subunits, is rather stable to proteolysis in vivo. In the absence of the Fas1 subunit (FAS1 deletion strain) the remaining Fas2 subunit becomes an orphan protein which is proteolytically unstable and is targeted to the 26S proteasome for degradation (Egner et al, 1993). In my work, I used the orphan Fas2 protein as object of investigation in order to identify new cellular components that are involved in the recognition and degradation of a natural unassembled protein subunit in S. cerevisiae. In addition, it was elucidated how these newly identified factors act in the quality control process of a naturally occurring orphan protein. Due to previous reports (Heck et al, 2010; Prasad et al, 2010) showing that some cytosolic misfolded proteins are imported into the nucleus for proteasomal degradation the cellular localization of orphan Fas2 was determined. Using laser-scanning microscopy it could be shown that C-terminally EGFP-tagged (enhanced green fluorescent protein) orphan Fas2 is localized to the cytosol, thus representing a potential substrate for the cytosolic quality control system (CytoQC). Furthermore, glycerol step density gradient centrifugation experiments revealed that the majority of the orphan Fas2 proteins are organized in high molecular assembly intermediates, which consist mostly of Fas2 homohexamers. By using the thermosensitive ssa1-45 mutant carrying in addition the gene deletions of SSA2, SSA3 and SSA4, it could be shown that the proteasomal degradation of the orphan protein is dependent on the Hsp70 chaperone Ssa1. It is likely that Ssa1 is required for keeping orphan Fas2 soluble. All members of the Hsp90-, Hsp100-, and Hsp110-chaperone family as well as the small heat shock proteins Hsp26 and Hsp42 were shown to have no effect on the degradation of orphan Fas2. Selected members of the Hsp40 chaperone family, including Apj1, Xdj1 and even Ydj1 also did not show a significant influence on the Ssa1-dependent elimination of the substrate. To prove whether other components of the UPS than the proteasome are required for degradation of orphan Fas2 different E2- and E3 gene deletion mutants were analyzed. It was found that the elimination of orphan Fas2 is strongly delayed in a strain carrying a UBC2 UBC4 double deletion. As single deletions of UBC2 and UBC4 have no significant effect on the turnover of the substrate, it can be assumed that these E2 enzymes have complementing functions in the degradation process of orphan Fas2. In a search for the responsible E3 ubiquitin ligase(s) required for orphan Fas2 degradation the E3 RING ligase Ubr1 was identified. Deletion of UBR1 leads to a strongly delayed degradation of orphan Fas2. The expression of an Ubr1 RING mutant (C1220S) or of an Ubr1 type-1 N-end rule mutant (D176) from a high-copy plasmid in the UBR1 deletion strain cannot complement the strongly delayed degradation of orphan Fas2. In contrast, the stabilization of the orphan protein in the UBR1 deletion strain is reversed, when the same strain harbours a high-copy plasmid expressing wild type Ubr1 or an Ubr1 type-2 N-end rule mutant (P406S) or even a cytosolically-located version of the nuclear E3 RING ligase San1 due to deletion of the nuclear localization sequence. Interaction studies revealed that the E3 RING ligase Ubr1 is physically associated with orphan Fas2. In addition, it was found that Ubr1 mutants harbouring either a defect RING domain or a defect in one of the N-end rule substrate-binding sites (type-1 or type-2) were still able to physically interact with orphan Fas2. Further studies showed that the physical association of orphan Fas2 and Ubr1 remains stable in the conditional ssa1-45 mutant carrying in addition deletions of SSA2 to SSA4. This indicates that already E3-bound orphan Fas2 may not require a functional peptide-binding domain of Ssa1 to maintain the physical contact to Ubr1. Finally, the AAA-ATPase Cdc48 was identified to be necessary for the elimination process of orphan Fas2. Cdc48 may function in the dissociation process of orphan Fas2 assembly intermediates, which mainly consist of Fas2-homohexamers.
Open Access
Design of synthetic epigenetic circuits featuring memory effects and reversible switching based on DNA methylation
(2017) Maier, Johannes A. H.; Möhrle, Raphael; Jeltsch, Albert
Epigenetic systems store information in DNA methylation patterns in a durable but reversible manner, but have not been regularly used in synthetic biology. Here, we designed synthetic epigenetic memory systems using DNA methylation sensitive engineered zinc finger proteins to repress a memory operon comprising the CcrM methyltransferase and a reporter. Triggering by heat, nutrients, ultraviolet irradiation or DNA damaging compounds induces CcrM expression and DNA methylation. In the induced on-state, methylation in the operator of the memory operon prevents zinc finger protein binding leading to positive feedback and permanent activation. Using an mf-Lon protease degradable CcrM variant enables reversible switching. Epigenetic memory systems have numerous potential applications in synthetic biology, including life biosensors, death switches or induction systems for industrial protein production. The large variety of bacterial DNA methyltransferases potentially allows for massive multiplexing of signal storage and logical operations depending on more than one input signal.
Open Access
Development and application of experimental tools for studying the distribution and dynamics of chromatin modifications
(2015) Kungulovski, Goran; Jeltsch, Albert (Prof. Dr.)
All cells in a multicellular organism carry the same genetic information, and yet throughout their lifetimes they follow unique transcriptional programs, which lead to phenotypical and functional differences. These differential gene expression programs are enacted by highly coordinated epigenetic mechanisms, which include modifications of chromatin, such as DNA methylation and histone post-translational modifications. Their involvement in chromatin-associated processes, association with different genomic elements and the means of their establishment and maintenance are crucial scientific issues. The primary focus of this study was to shed light on the genome-wide distribution of chromatin modifications and the effects of their local establishment. First, we focused our efforts into developing and applying novel affinity reagents for local and genome-wide characterization of histone modifications. We made use of native and engineered recombinant proteins that have an intrinsic ability to interact specifically with modified histones. In a rigorous side-by-side comparison with high quality histone modification antibodies, we successfully applied these novel affinity reagents in approaches such as western blot and chromatin precipitation coupled with quantitative PCR or massively parallel sequencing, following established quality control criteria. By this, we validated the feasibility of this strategy. We also discussed in detail the advantages of using recombinant proteins in lieu of antibodies, such as their cheap production with high yield, ease of protein engineering and consistent quality and reproducibility. Secondly, we wanted to clarify some of the principal mechanisms by which epigenetic modifications operate inside the cell. To this aim, we established and applied an approach based on zinc finger targeted promoter methylation of VEGF-A (vascular endothelial growth factor) in order to study the in vivo effects and dynamics of histone modifications and DNA methylation. Adenoviral constructs made of the targeting zinc finger were fused with catalytic domains from epigenetic enzymes such as DNA and histone methyltransferases and were used to infect cells. By these means we were able to successfully follow in detail the establishment, effects and kinetics of the installed chromatin modifications. Our data indicate that local chromatin editing of a target locus can change its intial chromatin state and consequently modulate its transcriptional output, albeit for a short period of time, before it returns to its native configuration. In conclusion, in this body of work we were able to successfully develop and apply novel affinity reagents for studying the distribution of histone modifications. Along the same lines, we also successfully developed and applied a strategy for targeted chromatin editing, which allowed us to study the dynamics of establishment, downstream effects and maintenance of chromatin modifications.
Open Access
Development of zinc finger methyltransferase fusion proteins for targeted DNA methylation and gene silencing in human cells
(2014) Nunna, Suneetha; Jeltsch, Albert (Prof. Dr.)
Epigenetic modifications such as DNA methylation and histone modifications play important roles in the regulation of gene expression. DNA methylation occurs at C5 position of cytosine residues mainly in CpG dinucleotides. In the human genome, about 70% of the CpGs are methylated. Most of the gene promoters are accompanied by CpG islands (regions rich in CpG dinucleotides) and methylation in these regions is inversely correlated with gene expression. Aberrant DNA methylation at the promoter region leads to variety of diseases including cancer. In the present study, we used catalytic domains of the Dnmt3a DNA methyltransferase and the GLP H3K9me3 lysine methyltransferase to silence two important oncogenes by targeted methylation. Zinc-finger proteins with predefined specificity were used as the targeting device. The first target gene was the vascular endothelial growth factor A (VEGF-A), which plays an important role in the vasculogenesis and angiogenesis. A zinc finger (VAZF) binding to the VEGF-A promoter was fused to either the catalytic domain of Dnmt3a (VAZF-Dnmt3aC) or to a fusion of Dnmt3a with its stimulator Dnmt3L (VAZF-Dnmt3a3Lsc). After transient transfection in ovarian cancer cells and magnetic activated cell sorting (MACS), we observed 25% methylation of the target region in the cells transfected with VAZF-Dnmt3aC and 49% in VAZF-Dnmt3a3Lsc transfected cells. VEGF-A expression was measured by quantitative -RTPCR and we observed a 36% reduction of VEGF-A expression in the cells that were transfected with VAZF-Dnmt3aC, and 56% in VAZF-Dnmt3a3Lsc transfected cells. However, transfection yields after MACS were only around 60-80% in these studies such that untransfected cells were still present. The second target gene of my study was the epithelial cell adhesion molecule (EpCAM), which is a transmembrane glycoprotein and is required for homophilic cell-cell adhesion. EpCAM is overexpressed in numerous cancers and its expression is inversely correlated with the promoter methylation status. In the present study, I used a zinc finger binding to the EpCAM promoter region, fused to the catalytic domain of the Dnmt3a DNA methyltransferase (EpZF-Dnmt3aC) for targeted methylation of the EpCAM promoter. After magnetic activated cell sorting, transfection yields of 60-80% were reached. With this approach 29% DNA methylation was achieved at the EpCAM promoter region in SKOV3 ovarian cancer cells. In stable cell lines expressing EpZF-Dnmt3aC, the methylation reached up to 48% which in turn led to 80% reduction of EpCAM protein expression. I also observed a reduction of cell proliferation in these stable cell lines which is a promising result suggesting that EpCAM repression might be an approach in cancer treatment. To improve the efficiency of gene delivery into the ovarian cancer cells, recombinant adenoviral vectors encoding the zinc-finger fused DNA methyltransferases and histone H3K9me3 methyltransferase catalytic domain were generated. Using adenovirus mediated gene delivery we achieved 53% methylation and 80% gene suppression at the VEGF-A promoter in SKOV3 ovarian cancer cells. Similarly, the zinc finger fused to GLP resulted in appearance of H3K9me3 at the promoter and 64% gene repression. At the EpCAM promoter, an EpCAM zinc finger fused to DNA methyltransferases led to 32% of methylation and 90% gene repression in the same ovarian cancer cells. Another important objective of the study was to measure the stability of DNA methylation and gene silencing of VEGF-A. After infection of SKOV3 ovarian cancer cells with adenovirus expressing VAZF-Dnmt3aC or VAZF-GLP, the methylation and gene expression was measured at different time points. In the cells that were infected with VAZF-Dnmt3aC constructs, establishment of DNA methylation was initiated 24 h after infection. The methylation reached to a maximum after five days, but surprisingly afterwards DNA methylation gradually declined to its basal level till day 15. VEGF-A supression correlated with the methylation levels, five days after infection 75% suppression of VEGF-A was observed which gradually declined to 8% after 15 days. In SKOV3 cells that were infected with recombinant adenoviral vectors encoding VAZF-GLP, H3K9 histone methylation peaked at day 5, but like the DNA methylation it gradually was lost afterwards. This result indicates that both silencing marks could not be introduced in a stable manner. As it was shown that DNA methylation and histone methylation acts synergistically, we measured DNA methylation after targeted introduction of histone H3K9 methylation and vice versa. However, in both the experiments we could not observe synergistic effects at the VEGF-A promoter. We started experiments to measure whether the decreased VEGF-A and EpCAM expression has an anti-tumor effect in a mouse model, which will explore the therapeutic potential of targeted methylation in cancer treatment.
Open Access
DNA-Array-Technologie für transkriptionelle Untersuchungen des Genoms der Hefe Saccharomyces cerevisiae
(2001) Hauser, Nicole; Wolf, Dieter (Prof. Dr.)
DNA-Array-Technologie ermöglicht die hoch-parallele Untersuchung transkriptioneller Aktivität. Sie wurde am vollständigen Gensatz von Saccharomyces cerevisiae methodisch etabliert und angewendet. Hierfür wurden 6116 offene Leserahmen der Hefe mittels PCR isoliert und auf Membranen aufgebracht. Auf diese wurden revers transkribierte Proben hybridisiert, die die Gesamt-mRNA aus Hefezellen repräsentierten. Die Signalintensität an jeder Belegposition des Rasters dient als Maß für die relative Häufigkeit des entsprechenden Transkripts. Die Methodik wurde bezüglich ihrer Aussagekraft bewertet und kontinuierlich technisch weiter entwickelt. Systematische Variation der Hybridisierungsparameter, speziell für die immobilisierte und freie DNA, schuf die Grundlage für eine Quantifizierung der absoluten Transkripthäufigkeiten. Die gewonnenen Rohdaten wurden mit Hilfe mathematischer Routinen prozessiert und neue Visualisierungs- und Analyseverfahren wurden auf die nach Signifikanz und anderen Kriterien gefilterten Datensätze angewendet, um eine biologische Interpretation zu unterstützen. Aus eigenen Experimenten und zahlreichen Kooperationen wurden beispielhaft die Ergebnisse aus vier experimentellen Ansätzen vorgestellt: (1) Vergleich einer industriell genutzten Weinhefe mit einem Laborstamm, sowohl auf transkriptioneller als auch genomischer Ebene. (2) Untersuchungen einer Disruptionsmutante im Gen des tRNA-Export-Faktors Los1p. (3) Erhöhung der Wachstumstemperatur von 30°C auf 37°C. (4) Adaptation von Hefezellen an hohe externe Salzkonzentration.
Open Access
Enantioselective hydrolysis of racemic naproxen nitrile and naproxen amide to S-naproxen by new bacterial isolates
(1994) Layh, Norman; Stolz, Andreas; Böhme, Joachim; Effenberger, Franz; Knackmuss, Hans-Joachim
Bacteria were enriched from soil samples with succinate as a carbon source and racemic naproxen nitrile [2-(6-methoxy-2-naphthyl)propionitrile] as sole source of nitrogen. Since naproxen nitrile was only poorly soluble in water media amended with different water-immiscible organic phases were used for the enrichments. With pristane (2,6,10,14-tetramethylpentadecane) as the organic phase two bacterial strains were isolated (strain C3II and strain MP50) which were identified as rhodococci. Cells of both strains converted naproxen nitrile via naproxen amide to naproxen. From racemic naproxen nitrile Rhodococcus sp. C3II formed S-naproxen amide and subsequently S-naproxen. Racemic naproxen amide was hydrolysed to S-naproxen. Rhodococcus sp. MP50 converted racemic naproxen nitrile predominantly to R-naproxen amide and racemic naproxen amide to S-naproxen. With both strains racemic naproxen amide was converted to S-naproxen with an enantiomeric excess >99% at a conversion rate up to 80% of the theoretical value. In strain C3II the enzymes which hydrolysed naproxen nitrile and naproxen amide were present only at a low constitutive level. In contrast, in Rhodococcus sp. MP50 these activities were induced when grown in the presence of various nitriles.
Open Access
Die Endoplasmatisches Retikulum assoziierte Degradation (ERAD) eines fehlgefalteten Membran verankerten Proteins mit variierender zytoplasmatischer Domäne und die Entdeckung eines neuen ERAD-Weges
(2012) Besser, Stefanie; Wolf, Dieter H. (Prof. Dr.)
Die Funktionalität neu synthetisierter Proteine ist unter anderem durch ihre korrekte dreidimensionale Faltung gegeben. Der Abbau von nicht funktionellen oder fehlgefalteten Proteinen schützt die Zelle vor Proteotoxizität, die zu schweren Beeinträchtigungen der Zellfunktion sowie Alzheimer-, Parkinson- und Creutzfeld-Jakob-Krankheit führt. Die Genauigkeit des Proteinreifungsprozesses ist so grundlegend wichtig, dass sich mehrere zelluläre Mechanismen entwickelt haben, die die Proteinfaltung überwachen. Das Endoplasmatische Retikulum (ER) spielt eine entscheidende Rolle bei der Proteinqualitätskontrolle (PQC; engl. protein quality control) und der Reifung sekretorischer Proteine. Der Austritt aus dem ER ist durch ein strenges Qualitätssystem geregelt, das fehlgefaltete Polypeptide und unvollendet zusammengebaute Untereinheiten erkennt und zurückhält. Unvollständig gefaltete oder fehlgefaltete Proteine werden dem ER-assoziierten Abbauweg (ERAD; engl. ER associated degradation) zugeführt. Die Retrotranslokation der fehlgefalteten Proteine über die ER-Membran ins Zytosol und der Abbau durch das Ubiquitin-Proteasom-System (UPS) ist für den ERAD-Prozess erforderlich. Die ER-Qualitätskontrolle (ERQC; engl. ER quality control) und die ERAD sind eng miteinander verbunden und wird auch ER-Qualitätskontrolle und -assoziierte Degradation (ERQD; engl. ER-quality control and degradation) genannt. Um zu untersuchen, ob die Faltungsintensität der zytosolischen Domäne für die substratspezifischen Anforderungen einiger ERAD-Komponenten verantwortlich ist, wurde eine Auswahl an fehlgefalteten Fusionsproteinen hergestellt, die die mutierte Carboxypeptidase Y (CPY*) im ER-Lumen als Degradationsmotiv tragen. Die Kollektion der fehlgefalteten Proteine mit unterschiedlicher zytosolischer Topologie besteht aus CTG*, CTL*myc und CTD*. Die durchgeführten Versuche mit Fusionsproteinen mit zytosolisch unterschiedlichen Topologien zeigten, dass Chaperone und andere Faktoren für die Entfaltung der zytosolisch stark gefalteten Domänen vor dem proteasomalen Abbau unerlässlich sind. Auftretende Unterschiede im Abbauverhalten können nur aufgrund der unterschiedlichen Faltungsstärken der zytosolischen Domänen der Substrate zustande kommen. Nichtsdestotrotz unterscheiden Zellen weniger zwischen schwach oder stark gefalteten Domänen. Die für den Abbau rekrutierten Komponenten werden abhängig von den charakteristischen Eigenschaften der zytosolischen Domänen ausgewählt. Frühere Arbeiten zur ERAD ließen erkennen, dass die Deletion einer oder beider bekannter ER-Membran lokalisierter E3-Ligasen, Hrd1/Der3 und Doa10, nicht die vollständige Stabilisierung der untersuchten Substrate bewirkt (Gnann et al., 2004; Huyer et al., 2004). Diese Ergebnisse lassen darauf schließen, dass mindestens eine weitere E3-Ligase involviert sein muss, die den endgültigen Abbau einleitet. Ubr1 ist als E3-Ligase im N-end rule Weg (engl.) bekannt (Bartel et al., 1990; Varshavsky, 1996a) und ist Teil der Abbaumaschinerie fehlgefalteter zytosolischer Substrate (Eisele et al., 2008; Heck et al., 2010). Bei Fehlen der klassischen ER-E3-Ligasen Hrd1/Der3 und Doa10 werden die ERAD-L-Substrate CTG* und CTL*myc ubr1-abhängig vollständig abgebaut. Um herauszufinden, welche weiteren Komponenten in den neuen Ubr1-abhängigen Weg involviert sind, wurden bekannte Komponenten aus dem klassischen ERAD-Weg zusätzlich in einer Doppelmutante der3/hrd1 doa10 deletiert. Das ERAD-L-Substrat CTG* wird weiterhin polyubiquitiniert und über das 26S Proteasom abgebaut. Das Hsp40-Co-Chaperon Hlj1 ist am vollständigen Abbau von CTG* beteiligt. Die AAA-ATPase Cdc48, sowie Der1 und Dfm1 sind in diesem Abbauweg entbehrlich. Keines der bisher untersuchten Ubiquitin-konjugierenden E2-Enzyme, die aus dem klassischen ERAD-Weg und N-end rule Weg bekannt sind, sind in den ubr1-abhängigen Weg involviert. Vermutliche handelt es sich bei diesem Abbauweg um eine Art Reservesystem, das an geschalten wird, wenn der klassische ERAD-Weg mit dem Abbau der zytosolischen Domänen der ERAD-L-Substrate überfordert ist.
Open Access
Endoplasmic reticulum associated protein degradation (ERAD): the function of Dfm1 and other novel components of the pathway
(2011) Stolz, Alexandra; Wolf, Dieter H. (Prof. Dr.)
Proteins, featured with a multitude of enzymatic activities as well as structural and other physiological functions are the main operators in the cell. Proteins are synthesized in the cytosol by ribosomes, which use m-RNA as a template to translate DNA based structural information into an amino acid sequence. During translation many errors occur resulting in so-called defective ribosomal products. In addition, stresses as heat, heavy metal ions and oxygen lead to the formation of partially unfolded and misfolded proteins. In human accumulation of these proteins results in severe diseases as are Alzheimer’s disease, Parkinson’s disease, Huntington’s disease and many others. Therefore quality control systems exist, which recognize unfolded or misfolded proteins and support their folding process. If a protein is unable to reach its native conformation or to refold, the quality control system marks it as terminally misfolded and hands it over to the degradation machinery of the cell. In case of proteins of the secretory pathway this process is called endoplasmic reticulum quality control and associated protein degradation (ERQD). ERQD includes the recognition of the misfolded protein species, the trimming of glycan trees to signal misfolding, retrograde transport out of the ER lumen into the cytosol, ubiquitylation of the misfolded protein and degradation by the proteasome. The following thesis was engaged in the identification of new components of ERQD and tried to get insights into some mechanistic functions of the involved proteins. The proteins Dfm1, Mnl2 and Ubr1 were found as new components of the endoplasmic reticulum associated protein degradation (ERAD) machinery. Mnl2 was identified as a putative α-1,2-mannosidase. It was shown to be involved in the degradation of the misfolded glycoprotein CPY*. Most probably Mnl2 trims down the glycan trees of ERAD substrates, which are subsequently recognized by the lectin Yos9. Yos9 accelerates the degradation of terminally misfolded glycoproteins which expose these glycan structures. However, Yos9 does not seem to act only on glycosylated proteins but also seems to affect the degradation kinetics of unglycosylated ERAD substrates. In contrast to misfolded glycoproteins Yos9 delays degradation in case of the unglycosylated ERAD substrate CPY*0000. Most likely Yos9 has a chaperone like function in addition to its lectin function and provides more time for refolding of the misfolded protein. This function is, however, independent of its MRH domain that recognizes glycans. The other new ERAD component, the polytopic ER membrane localized Dfm1 protein, was found to form distinct complexes with the ligases Hrd1/Der3 and Doa10 as well as with the AAA type ATPase Cdc48. Degradation of different ERAD substrates containing a transmembrane domain was tested for Dfm1 involvement. The degradation and ubiquitylation of the ERAD-C substrate Ste6* was shown to depend on Dfm1. In addition, Dfm1 seems to be involved in a new degradation pathway, which acts independently of the ubiquitin ligases Hrd1/Der3 and Doa10. In the absence of these canonical ER ligases the cytosolic ubiquitin ligase Ubr1 seems to be recruited to maintain degradation of at least some ERAD substrates by the proteasome. Extraction of the misfolded protein species no longer depends on Cdc48 in all cases, but the driving force of other machines, most probably chaperones of the Ssa family of Hsp70 chaperones, were found to be sufficient to keep extraction and degradation of the substrates going.
Open Access
ER-associated protein degradation (ERAD): an unexpected function of Yos9 and the discovery of Mnl2, a new component of the pathway
(2011) Martínez Benítez, Elena; Wolf, Dieter H. (Prof. Dr.)
In eukaryotes, membrane and soluble secretory proteins are synthesized at the rough endoplasmic reticulum (ER). A protein that cannot fold properly will be degraded in a process called ER associated degradation (ERAD). Failures in ERAD either by loss of function or by premature degradation of proteins cause a range of severe diseases in humans. In 1989 the gene responsible for the human disease cystic fibrosis (CF), cystic fibrosis transmembrane conductance regulator (CFTR), encoding a chloride channel was found. The mutation ΔF508 present in 80% of the patients provokes the most severe symptoms and shortest life expectancy. The disease is a consequence of the accelerated degradation of the protein CFTRΔF508, which never reaches its site of action. Moreover, wild type CFTR has a 25% success in reaching its final destination. Is of great value to understand the ERAD of CFTR and CFTRΔF508 in order to understand the disease. Many components involved in ERAD of these substrates had been discovered. The first part of this work involves a systematic study of human CFTR degradation in yeast (the turnover of CFTR in yeast cells behaves like CFTRΔF508 in human cells). All components examined were found not to be required for CFTR ERAD or had a very mild effect. The ER protein quality control recognizes misfolded proteins in two ways: via exposure of hydrophobic patches on the surface of a protein and modification of its glycan structure. The majority of the proteins that enter the ER are N-glycosylated. During folding of a protein, several enzymes trim these glycan trees generating a degradation signal, which is recognized by the lectin Yos9. There is little known about proteins that enter the ER but are not glycosylated. The second part of the work refers to findings that shed light onto the differences between glycosylated and non-glycosylated substrates in ERAD. To define the ERAD pathway for non-glycosylated proteins, ERAD deficient mutants were checked in their capacity to deliver a non-glycosylated protein for elimination. It is shown that unglycosylated CPY* (CPY*0000) is degraded by the same pathway as is glycosylated CPY* (ERAD-L). However, the Yos9 protein, known to be the recognition component of glycosylated misfolded proteins in ERAD, is shown in this work to have a tuning role in the ERAD of unglycosylated CPY*0000. Yos9 promotes degradation of glycosylated substrates while it hinders degradation of unglycosylated CPY*0000. Additional ERAD components are still to be discovered. In this work a putative mannosidase was found. This protein was named mannosidase like protein 2 (Mnl2). Mnl2 accelerates CPY* degradation. The effect of the deletion of MNL2 is most notable when its homologue MNL1/HTM1 is absent. Substrate degradation in the deletion strain is affected because the glycan structure on the ERAD substrate is no longer a degradation signal. This section of the work introduces a novel ER quality control component involved in glycan trimming.
Open Access
ER-assoziierte Proteindegradation (ERAD): Untersuchungen von Komponenten zum Abbau glykosylierter und unglykosylierter Substrate
(2009) Fischer, Oliver; Wolf, Dieter H. (Prof. Dr.)
Die Funktion von Proteinen hängt von ihrer korrekten Struktur ab. In eukaryontischen Zellen existieren deshalb Protein Qualitätskontrollsysteme, die missgefaltete Proteine erkennen und abbauen können. Für neusynthetisierte sekretorische Proteine befindet sich eine solche Qualitätskontrolle im endoplasmatischen Retikulum (ER) und wird allgemein als ER-abhängige Protein Qualitätskontrolle (ERQC) bezeichnet. Missgefaltete Proteine werden im ER über einen retrograden Transport zurück ins Zytosol transportiert und dort über das Ubiquitin-Proteasom-System abgebaut. Man nennt diesen ERQC-abhängigen Abbau von Proteinen die ER-assoziierte Degradation (ERAD). Diese Systeme sind evolutionär hoch konserviert. Man findet sie in allen Eukaryontenzellen von der Hefe bis zum Menschen. Ein für proteinbiochemische und molekularbiologische Methoden leicht zugänglicher eukaryoter Modellorganismus ist die Bäckerhefe Saccharomyces cerevisiae. Mehrere Studien mit dem ERAD-Modellsubstrat CPY* zeigten die Bedeutung der Glykane an Proteinen für die ER-assoziierte Degradation von missgefalteten glykosylierten sekretorischen Proteinen auf. In Wildtypzellen wird glycosyliertes CPY* mit einer Halbwertszeit von t1/2 = 20 min abgebaut; die entsprechend unglykosylierte Variante CPY*0000 wird stabilisiert mit einer Halbwertszeit von ca. t1/2 = 90 min. In dieser Arbeit konnte zum ersten Mal gezeigt werden, dass die zwar verlangsamte Degradation von CPY*0000 noch eine von Schlüsselkomponenten des ERAD abhängige Degradation ist. Zunächst wurde in mehreren unabhängigen Pulse-Chase Analysen eine Stabilisierung von CPY*0000 in einem pre1-1 pre4-1 Stamm gefunden. Eine Stabilisierung von CPY*0000 wurde in weiteren Pulse-Chase Analysen mit einem delta-der3-Stamm entdeckt, der die für ERAD-L und -M Weg notwendige E3-Ubiquitin-Ligase Der3 depletiert. Zudem wurde in einer Pulse-Chase Analyse das Abbauverhalten von CPY*0000 in vakuolären Proteinasen defizientem Zellen untersucht. Die beiden ER lumenalen Proteine Yos9 und Htm1 (auch Mnl1 genannt) wurden auf Grund von ersten Studien als Lektine vermutet, die den Glykan-Status an glykosylierten sekretorischen Proteinen detektieren können. In dieser Arbeit wurden in vivo Bindungspartner von Yos9 nachgewiesen. In Immunopräzipitationsexperimenten wurden Bindungen von Yos9 zu seinem Substrat CPY* und zu ERAD Komponenten Cdc48, Der3, Htm1 und Kar2 gefunden. Die Bindungen von Yos9 zu Cdc48 und Der3 sind Hrd3 abhängig, nicht aber die Bindung zum Substrat, wie weitere Immunopräzipitationen zeigten. In weiteren Co-Immunopräzipitationen wurde auch eine Bindung von Htm1 zu der ER-α1,2-Mannosidase (Mns1) ausgeschlossen. Htm1 besitzt eine zu Mns1 homologe Mannosidase-Domäne. In dieser Arbeit wurden deshalb genetische Analysen mit Htm1 und Mns1 durchgeführt. Erstmals konnte ein genetischer Zusammenhang zwischen den beiden Genen gefunden werden: Die Gene MNS1 und HTM1 haben auf die Degradation von CPY* einen additiven Effekt. Mit verschiedenen Deletionstämmen wurden Pulse-Chase Analysen im Hinblick auf die Degradation CPY* durchgeführt. Zellen mit eine Deletion in den Genen MNS1 oder Htm1 zeigten jeweils nur noch eine Degradation von CPY* mit einer Halbwertszeit von ca. t1/2 = 90 min auf. Eine delta-mns1 delta-htm1 Doppeldeletion in den Zellen führte darüber hinaus zu einer fast vollständigen Stabilisierung von CPY*. Der delta-mns1 delta-htm1 Doppeldeletionsstamm wurde mit einem „high-copy“ Plasmid transformiert, das Htm1 kodierte. In Pulse-Chase Analysen wurde nun gezeigt, dass diese Überexpression von Htm1 in den Zellen den delta-mns1 Phänotyp hinsichtlich der CPY* Degradation nicht ändert. Analog wurde dieses Experiment mit überexprimierter Mns1 im delta-htm1 Deletionsstamm durchgeführt und nachgewiesen, dass auch eine Überexpression von Mns1 nicht den delta-htm1 Phänotyp ändert. Die Funktion von Usa1 im ERAD ist in Hinblick auf seine Substratspezifität widersprüchlich. Im Rahmen einer Kooperation mit der Gruppe von T. Sommer, MDC Berlin, wurde in dieser Arbeit der Abbau von CTG* in DF5-Stämmen in Abhängigkeit von USA1 untersucht. Es wurden der DF5-Stamm Wildtyp für USA1 und der DF5-Stamm delta-usa1 mit einer USA1 Deletion mit dem Plasmid, welches CTG* kodiert, transformiert. Mit den Transformanden wurden zwei unabhängige Pulse-Chase-Analysen durchgeführt. In beiden Versuchen zeigte sich eine deutliche verlangsamte Degradation von CTG* in den delta-usa1-Zellen im Vergleich zu den USA1-Wildtypzellen.
Open Access
The function of the beta6/Pre7 propeptide for 20S proteasome biogenesis in baker’s yeast
(2004) Iyappan, Saravanakumar; Heinemeyer, Wolfgang (PD Dr. )
In eukaryotes, the regulated proteolysis of intracellular proteins occurs through a specialized enzymatic machinery, the ubiquitin-proteasome system. Here proteins are specifically recognized and marked for degradation by addition of poly-ubiquitin chains before being degraded by the 26S proteasome. The 20S proteasome is the catalytic core of the 26S proteasome, composed of two copies of each 7 different alpha- and beta-type subunits forming a barrel-shaped complex. Assembly and maturation of the eukaryotic 20S proteasome is a multi-step process, in which the free alpha- and beta-type subunits are first assembled into the complete but still inactive complex. Final activation occurs after conversion of three of the beta-type subunits into their mature form by autocatalytic removal of N-terminal propeptides leading to the exposure of catalytic threonines at their N-termini. The contribution of the propeptides of these active beta-type subunits to particle assembly in baker's yeast ranges from absolute necessity to dispensability. In addition, of the four inactive beta-type subunits three are also synthesized with a propeptide which stays either unprocessed (beta-3/Pup3) or is only partially removed (beta-7/Pre4 and beta-6/Pre7) by the action of the activated neighbour beta-type subunits. The resulting short N-terminal extensions of 8-9 residues are found in the crystal structure of the yeast 20S proteasome in an extended conformation each reaching the restrictions separating the central proteolytic chamber from the antechambers. In this work multiple questions are addressed regarding the function of propeptides of the inactive subunits that have been preserved during proteasome evolution. The work focusses especially on the structure and function of the beta-6/Pre7 propeptide and its contribution to proteasome biogenesis, because it is essential for yeast cell viability in contrast to the beta-3/Pup3 propeptide that is dispensable for proteasome function, like already reported for the beta7/Pre4 propeptide. The region in the beta-6/Pre7 propeptide that is cleaved off by the neighboring active beta-type subunit beta-2/Pup1 is not essential, but the remaining 9 residues found in the mature proteasome are indispensable for its function. Stepwise truncation analysis in the Pre7 propeptide remnant region revealed that the six most C-terminal amino acids are sufficient for cell survival, but cell growth is considerably retarded if this piece is shortened from 9 to 6 residues. Surprisingly, the need for this propeptide remnant is restricted to the presence of the proteasome assembly factor Ump1. Ump1 was shown to enhance the dimerization of two half-proteasome precursor complexes and in addition to promote proteasome maturation. Fractionation of precursor and mature proteasome species by gel-filtration analysis as well as pulse-chase analysis elucidated that in a UMP1 wild-type strain background the Pre7 mutant bearing a propeptide remnant with 6 residues has severe defects in particle assembly and maturation. In contrast to this, in ump deletion cells such perturbation in assembly and maturation was found even in the presence of wild-type Pre7 and these defects were not enhanced by the truncations in the Pre7 propeptide region. Similarly, the proteasomal peptidase activities were reduced gradually along with stepwise truncations in the Pre7 propeptide region in the presence of Ump1, but the activity profiles were not altered in the ump1deletion strain background. As seen in non-denaturing gel electrophoresis, truncations in the Pre7 propeptide remnant region by up to 3 residues interfered with the dimerization of two half-proteasomes to the pre-holoproteasome and thus led to accumulation of these assembly products, but again only in the presence of Ump1. The obstruction stage in the proteasome assembly in those strains that are viable only in the absence of Ump1 was analysed by expressing the Ump1 protein from an inducible promoter. As expected, in cells carrying a Pre7 propeptide remnant shorter than 6 residues, the appearance of Ump1 completely inhibited their proliferation after few cell divisions. Analysis of these cells revealed that the proteasome assembly process is blocked at the state of half-proteasome assembly intermediates containing Ump1. Furthermore, interaction of Ump1 was restricted to the half-proteasome, whereas in strains carrying wild-type Pre7 Ump1 was found in a putative earlier assembly intermediate. Thus, the binding of Ump1 to early assembly stages might trigger the dimerization process more efficiently than its interaction with completely assembled half-proteasomes. In vitro interaction and GST-pull down analysis additionally indicated that formation of half-proteasomes is coupled with incorporation of Ump1 and that its association depends on the late incorporating subunits Pre2, Pre4 and Pre7, most likely mediated by their propeptides. Replacement analysis of the Pre7 propeptide remnant region showed that its function specifically depends on its amino acid sequence. Sequence comparison of the primary structure of beta-6 orthologues from different species enlightened that a proline at position -6 and a tyrosine at -5 have been absolutely conserved during evolution. In addition to this, a stretch of hydrophobic amino acids that forms a hydrophobic pocket and binds the propeptide in the mature proteasome was also found to be conserved in all known beta 6 subunits. The tyrosine at position -5 turned out to be crucial for the function of the Pre7 propeptide remnant, since its exchange with alanine was lethal in UMP1 wild-type cells. The aromatic side chain of this tyrosine is required to fix the propeptide remnant on the Pre7 subunit surface by interacting with the hydrophobic pocket, whereas the hydroxyl group is dispensable. Two phenylalanines in the hydrophobic pocket were also found to be essential, suggesting that these residues are either crucial to fix the propeptide at any stage of the assembly process or required for the structural integrity of the Pre7 subunit. Remarkably, the lethality caused by the mutations in the hydrophobic pocket was again coupled to the presence of the Ump1 protein. Probably, the exposure of this hydrophobic surface due to the absence of the Pre7 propeptide or due to loss of its ability to bind it might lead to unfavorable interactions of the Ump1 protein with this area and so might preclude further assembly and maturation steps. Although the propeptide of the Pre7 subunit seems to be crucial to avoid such unfavorable interactions of Ump1 with the hydrophobic surface, a primary role might lie in maintaining a structural integrity, which is a prerequisite for the proceeding of the assembly and maturation process. This is implicated by the fact that replacing the tyrosine in the propeptide region in combination with replacements of any of the hydrophobic amino acids in the hydrophobic pocket causes severe growth defects even in ump1 deletion cells and that exchanging all conserved residues in the pocket by alanines is lethal. Taken together this work establishes a critical role of Pre7 propeptide in the yeast 20S proteasome assembly and in its maturation. The conserved residues found in the propeptide and in the hydrophobic pocket of the Pre7 subunit are concertedly necessary for proper folding into a tertiary structure that allows efficient incorporation of the subunit into the proteasome. Moreover, the propeptide of Pre7 is specifically required for the function of the proteasome assembly and maturation factor Ump1 during proteasome biogenesis. In addition, it might be involved in interactions with other proteasomal proteins or in the transmission of structural alterations in the proteasomal assembly intermediates.