Refine
Document Type
- Doctoral Thesis (8)
Language
- English (8) (remove)
Has Fulltext
- yes (8)
Is part of the Bibliography
- no (8)
Keywords
- Ageing (1)
- Autophagy (1)
- Katabolismus (1)
- Podospora anserina (1)
- Protein-protein interaction (1)
- Proteine (1)
- Transcriptome analysis (1)
Institute
Die akute myeloische Leukämie (AML) ist eine aggressive Erkrankung des Knochenmarks, welche die Hämatopoese beeinträchtigt und zu Knochenmarksversagen führt. Trotz des Fortschritts in der AML-Therapie bleibt die Prognose für die meisten Patienten schlecht, sodass neue Therapieansätze für die Behandlung dringend benötigt werden. Autophagie, ein kataboler Abbauprozess von zellulären Komponenten, ist nachweislich an der Entstehung von AML beteiligt. Als zentraler Regulator von Zellüberleben, Homöostase und Stoffwechsel, dient die Autophagie als Nährstoffquelle durch die Wiederverwertung von Makromolekülen während begrenzter Energieversorgung. AML-Zellen benötigen ein konstantes Nährstoff- und Energieniveau, um ihre Vermehrung aufrechtzuerhalten. Dies wird durch eine Umstellung von Stoffwechselwegen, insbesondere des mitochondrialen Stoffwechsels einschließlich der oxidativen Phosphorylierung (OXPHOS) und des Tricarbonsäurezyklus (TCA), erreicht.
Mehrere Studien haben die Hemmung der Autophagie für die Behandlung von Krebs als vielversprechenden Ansatz vorgestellt. Doch eine Monotherapie mit Autophagie-Inhibitoren erzielte nur eine geringfügige Wirksamkeit. Eine mögliche Erklärung hierfür ist die Entstehung von Kompensationsmechanismen, die zum Ausgleich der Autophagie-Hemmung in Krebszellen entstehen. Bis heute sind diese Kompensationsmechanismen kaum untersucht. Ziel dieser Arbeit ist es, ein geeignetes Autophagie-Gen zu identifizieren, mit dem sich die Rolle der Autophagie-Hemmung für das Überleben von AML-Zellen untersuchen lässt. Zusätzlich sollen die kompensatorischen Mechanismen, die durch die Autophagie-Hemmung in AML-Zellen entstehen können, untersucht werden, um neue metabolische Angriffspunkte zu identifizieren, die für Kombinationstherapien genutzt werden können.
Zu Beginn der Arbeit wurde ein gezielter CRISPR/Cas9 Screen in zwei humanen AML-Zelllinien durchgeführt, um Autophagie-Gene zu identifizieren, deren Verlust eine Proliferationsstörung in AML-Zellen verursacht, welche überwunden werden kann. Validierungsexperimente zeigten, dass der Verlust von ATG3 das Zellwachstum signifikant verminderte. Außerdem zeigte die Messung des Autophagie-Fluxes, dass der Verlust von ATG3 die Autophagie stark beeinträchtigte. Dies wurde durch eine Western-Blot-Analyse, die eine beeinträchtigte LC3-Lipidierung zeigte, und durch eine Immunfluoreszenzanalyse der Autophagosomen-Bildung mittels konfokaler Mikroskopie, die eine geringere Anzahl von Autophagosomen in ATG3-defizienten Zellen ergab, bestätigt. Deshalb wurde der Knockdown von ATG3 in AML Zellen verwendet, um die Mechanismen, die zum Ausgleichen der Autophagie-Hemmung entstehen, zu untersuchen. Zuerst wurde die Zellproliferation in fünf verschiedenen AML Zelllinien über sieben Tage betrachtet. In allen Zellenlinien führte der Verlust von ATG3 mittels small hairpin RNA zu verminderter Zellproliferation. Diese Ergebnisse zeigen die wichtige Rolle von ATG3 in der Autophagie und dass Autophagie-Hemmung durch ATG3-Verlust das Wachstum von AML-Zellen beeinträchtigt.
Da der Verlust von ATG3 die Proliferation von AML-Zellen beeinträchtigte, wurde eine Zellzyklusanalyse durchgeführt. Eine reduzierte S-Phase bestätigte die verminderte Proliferation in ATG3-depletierten AML-Zellen, doch der Zellzyklus war grundsätzlich nicht gestoppt. Darüber hinaus ergab die Analyse der Apoptose, dass diese unter dem Verlust von ATG3 erhöht war, aber etwa 50% der Zellen blieben vital. Diese Beobachtungen deuten darauf hin, dass AML-Zellen trotz des Verlusts der ATG3-abhängigen Autophagie weiter proliferieren können.
Um die Mechanismen zur Kompensation der Autophagie-Hemmung zu untersuchen, wurden die Auswirkungen des ATG3-Verlusts auf die mitochondriale Homöostase untersucht. Die Mitophagie sowie das mitochondriale Membranpotenzial und die Masse unterschieden sich zwischen Kontroll- und ATG3-depletierten AML-Zellen nicht, was darauf hindeutet, dass die mitochondriale Homöostase durch den Verlust von ATG3 nicht beeinträchtigt ist. Als nächstes wurde die mitochondriale Funktion durch Messung des ATP-Spiegels und der OXPHOS untersucht. Die ATP-Level und die OXPHOS waren nach dem Verlust von ATG3 in AML-Zellen erhöht, was auf eine gesteigerte mitochondriale Aktivität bei Autophagie-Defizienz hinweist.
Tectonin β-propeller containing protein 2 (TECPR2) was first identified in a mass- spectrometric approach as an interactor of GABARAP, an ATG8-family protein playing a role in autophagy. The mammalian ATG8 protein family consists of seven members, namely MAP1LC3A (LC3A), MAP1LC3B (LC3B), MAP1LC3C (LC3C), GABARAP, GABARAPL1 and GABARAPL2. All share an ubiquitin-like core and possess two additional N-terminal α-helices, which are important for the distinct functions of the proteins. First determined in various organelles the ATG8 proteins are shown to be involved in autophagy, supporting the formation and cargo recruitment of autophagosomes, the vesicles transporting cargo for autophagic degradation.
Autophagy is the process of recycling cytoplasmic contents by degradation of misfolded proteins or damaged organelles in order to supply nutrients. Also clearance of pathogens can be achieved via autophagy. Importantly, LC3B is incorporated into the autophagosomal membrane and is therefore used as the main marker for autophagosomes. Previous studies exhibited that depletion of TECPR2 leads to a loss of LC3B-positive structures in cells, which suggests TECPR2 to positively regulate autophagic processes.
A frame shift deletion in the gene encoding for TECPR2 causes the generation of a premature stop codon and subsequent an unstable version of the protein, which is then degraded. Mutation in the TECPR2 gene triggers a neurodegenerative disorder termed hereditary spastic paraparesis (HSP). HSPs are a diverse group of neurodegenerative diseases that are characterized by spasticity in prevalent lower extremities and were mediated by a loss of axonal integrity of the corticospinal motor neurons. In the context of HSP more than 50 gene loci were identified by now. While TECPR2 is a human ATG8 binding protein and positive regulator of autophagy causing a form of HSP, the exact function of TECPR2 is unknown.
This study primarily focused on the determination of TECPR2’s binding mode to ATG8 proteins in vitro and in cells. The association of TECPR2 to all ATG8-family proteins was confirmed in in vitro pulldown experiments. Following fragment-based binding and peptide array experiments, the LC3-interacting region (LIR) of TECPR2 could be verified with mutants of TECPR2 lacking the LIR motif. Nuclear magnetic resonance (NMR) and isothermal titration calorimetry (ITC) were conducted to gain deeper insights into the binding preference to the different ATG8-family members. Moreover, the crystal structure of TECPR2-LIR was solved. In cells colocalization studies with overexpressed ATG8 proteins unraveled a preferential binding to the LC3-subfamily.
Further, mass spectrometric analysis revealed novel association partners of TECPR2: SEC24D, HOPS and BLOC-1, all of those participating in different endomembrane trafficking pathways. Interaction and colocalization of TECPR2 with these components was validated with several immunoprecipitation experiments and the N-terminal part of the protein comprising the WD40-domain could be defined as the binding site for all three of the association partners. In further approaches, the requirement of the LIR-motif and the necessity of the availability of LC3 protein for the particular interactions were determined. Interestingly, in the absence of LC3C the binding of TECPR2 to SEC24D was completely disrupted whereas a loss of LC3B only resulted in a decreased association. Notably, the binding proteins were not subjected to autophagosomal degradation, indicating that TECPR2 may operate as a multifunctional scaffold protein. While depletion of TECPR2 destabilized HOPS and BLOC-1, the autophagy defect observed in TECRP2-deficient cells could not be attributed to functional impairment of these two complexes.
Moreover, loss of TECPR2 led to a decline in protein levels of SEC24D and of its heterodimer partner SEC23A. Thus, TECPR2 is required to regulate the protein levels of SEC23A and SEC24D and subsequently the formation of the heterodimers. Together, SEC24D and SEC23A form the inner coat of COPII vesicles. These vesicles are responsible for the anterograde transport of cargo from the ER toward the Golgi compartment. COPII-coated vesicles are secreted form ER at distinct sites, termed ER exit sites (ERES). The small GTPase SAR1A maintains the vesicle budding, coating and secretion at the ERES. Together with SEC13, SEC31 forms the outer coat of the COPII vesicles and therefore serves as a general ERES marker.
Consistent with a defect in COPII coat assembly, the number of ERES diminished in the absence of TECPR2. These phenotypes could be rescued by the wildtype TECPR2 protein but not by the LIR-mutant. Intriguingly, these results were mimicked by depletion of LC3C, which localized to ERES. By monitoring the release of various cargos from ER in dependency of TECPR2 or LC3C, a role of both proteins in ER export was determined. These facts indicated that TECPR2 cooperates with LC3C to facilitate COPII assembly, ERES maintenance and ER export. Notably, fibroblast derived from a HSP patient carrying mutated TECPR2 showed diminished SEC24D protein levels and delayed ER export.
Concurrent with emerging evidence for a role of ERES in autophagosome formation, depletion of TECPR2 or LC3C or overexpression of a constitutive inactive SAR1 mutant reduced puncta formation of the early autophagosomal protein WIPI2.
In summary, this study uncovered a role for TECPR2 in ER export at ERES through interaction and stabilization of SEC24D, a COPII coat protein. This process also depended on ATG8-family protein LC3C, which is localized at ERES. Both proteins are required for correct COPII-mediated secretion. Moreover, the presence of TECPR2 and LC3C on ER allows development of omegasomes, membranous structures budding ER to form autophagosomes, by stabilization of WIPI2 and therefore contribute to autophagosome formation.
Autophagy, meaning “self-eating”, is an important cellular waste disposal mechanism. Thereby, damaged proteins, lipids and organelles are enclosed by autophagosomes and subsequently transported to the lysosomes for degradation into basic, cellular building blocks. Under basal conditions autophagy prevents the accumulation of defective and harmful material and generally promotes cell survival. However, several studies reported that hyperactivated autophagy, e.g. during developmental processes in lower eukaryotes, or during chemotherapeutic treatment of cancer cells, can also trigger cell death.
In recent years, autophagic cell death (ACD) has been considered as an alternative cell death pathway for tumor therapy, especially for solid tumors with high apoptosis resistance such as glioblastoma. Glioblastoma (GBM) is a very aggressive, malignant primary brain tumor with a median survival of ~ 15 months despite surgery and chemoradiotherapy. Accordingly, there is a great interest in improving GBM therapy through alternative cell death mechanisms. Interestingly, it has been shown that various substances, e.g. AT 101, cannabinoids and the combination of imipramine and ticlopidine (IM+TIC), induce ACD in GBM cells.
The aim of this project was to identify the underlying mechanisms of stress- and drug-induced ACD and its therapeutic potential for glioblastoma treatment. For detailed investigation of ACD, a CRISPR/Cas9-based approach was used to generate ATG5 and ATG7 knockouts as genetic models of autophagy deficiency. In a previous study of our lab it was demonstrated that administration of AT 101 triggers ACD in glioblastoma cells, which was associated with early mitochondrial fragmentation but no signs of apoptosis. Since mitochondrial fragmentation often precedes mitophagy, the first part of this thesis explored the potential role of mitophagy in AT 101-induced cell death.
ATG5-depleted cells confirmed that AT 101 induces ACD. In addition, treatment with AT 101 resulted in a pronounced mitochondrial depolarization, which was at least partly caused by the opening of the mitochondrial permeability pore. Global proteome analysis of AT 101-treated GBM cells revealed a robust decrease in mitochondrial protein clusters as well as a strong increase in the enzyme heme oxygenase-1 (HMOX1). Subsequent experiments for detailed investigation of mitophagy following AT 101 treatment (western blot, flow cytometric MTG and mt-mKeima, qRT-PCR of mitochondrial vs nuclear DNA) consistently indicated strong mitophagy induction by AT 101, which could be reduced by genetic or pharmacological inhibition of autophagy. Furthermore, siRNA-mediated knockdown experiments revealed that the selective mitophagy receptors BNIP3 and BNIP3L and the HMOX1 enzyme play an essential role in AT 101-induced mitophagy and subsequent cell death. Taken together, these data demonstrate that AT 101-induced mitochondrial dysfunction and HMOX1 induction synergize to promote excessive mitophagy with a lethal outcome in glioma cells.
The second part of this thesis focused on the identification of new substances that cause ACD and the investigation of the underlying cell death pathways. Using a cell death screen of the ENZO Screen-Well™ autophagy library in MZ-54 wild-type vs ATG5 and ATG7-depleted cells, loperamide, pimozide, and STF-62247 were identified as ACD-inducing agents. The increase of the autophagic flux and the induction of ACD by these substances was confirmed by using different ATG5 and ATG7 knockout cell lines and the already established positive control IM+TIC.
In contrast to AT 101, IM+TIC, STF-62247, loperamide and pimozide produced neither mitochondrial dysfunction nor mitophagy. Interestingly, it has been described that imipramine, loperamide and pimozide inhibit the lysosomal enzyme acid sphingomyelinase, which is associated with impaired lipid transport. Global proteome analysis and cholesterol staining confirmed that all four substances, but especially loperamide and pimozide, inhibit cellular lipid transport, leading to massive lipid accumulation in the lysosomes. In the further course of the experiments, the connection between defective lipid transport and autophagy was investigated in more detail. On the one hand, the defective lipid transport contributed to the induction of autophagy, on the other hand the massive accumulation of lipids led to lysosomal membrane damage, inhibition of lysosomal degradation at later time points and finally to a lysosomal cell death. Remarkably, it has been shown that hyperactivated autophagy by IM+TIC, loperamide and pimozide massively promotes lysosomal membrane damage. This result highlights the difficulties of a clear distinction between autophagic and lysosomal cell death.
In summary, two new signaling pathways that induce autophagic cell death in GBM cells and may be relevant for glioblastoma therapy were investigated in this study.
RNA modifications are present in all three kingdoms of life and detected in all classes of cellular RNAs. RNA modifications are diverse, with more than 100 types of chemical modifications identified to date. These chemical modifications expand the topological repertoire of RNAs and are expected to fine-tune their functions. Ribosomal RNA (rRNA) contains two types of covalent modifications, either methylation on the sugar (Nm) or bases (mN), or base isomerization (conversion of uridine into pseudouridines, "). Pseudouridylations and ribose methylations are catalyzed by site-specific H/ACA and C/D box snoRNPs, respectively. The RNA component (snoRNA) of both types of snoRNPs is responsible for the site selection by base pairing with the rRNA substrate, whereas the protein component catalyzes the modification reaction: Nop1 in C/D box and Cbf5 in H/ACA box snoRNPs. Contrastingly, base methylations are performed by snoRNA independent, ‘protein-only’, methyltransferases (MTases). rRNA modifications occur at highly conserved positions, all clustering around functional ribosomal sites. Mutations in factors involved in rRNA modification have been linked to severe human diseases (e.g. X-linked Dyskeratosis congenita). Emerging evidences indicate that heterogeneity in RNA modification prevails, i.e. not all positions are modified at all time, and the concept of ‘specialized ribosomes’ has been coined. rRNA modification heterogeneity has been correlated with disease etiology (cancer), and shown to play a role in cell differentiation(hematopoiesis). Remarkably, alteration in rRNA modification patterns profoundly affects the preference of ribosomes for cap- versus IRESdependent translation initiation, with major consequences on cell physiology.
This dissertation aimed to shed light on changes of the epigenetic landscape in heart and skeletal muscle tissue of the turquoise Killifish N. furzeri, a novel, short-lived animal model for aging research. The following results could be obtained:
1. A global trend towards closed chromatin conformation could be observed; histone markers for H3K27me3, H3K9me3 and H4K20me3 accumulated in skeletal muscle tissue from old N. furzeri. Markers for open chromatin conformation such as H3K4me3, H3K9ac and H4K16ac decreased in old skeletal muscle tissue. In old hearts from N. furzeri an accumulation of H3K27me3 could be detected while H3K9ac was found to increase with age as well. mRNA expression levels of methylating enzymes were higher in skeletal muscle tissue from old N. furzeri when compared to expression levels in skeletal muscle tissue from young N. furzeri.
2. The shift of epigenetic pattern was accompanied by a change of gene expression. Via mRNA sequencing in collaboration with the MPI, Bad Nauheim it could be shown that genes associated with cell cycle and DNA repair were lower expressed in skeletal muscle tissue from old N. furzeri than in tissue from young N. furzeri. Genes, associated with inflammatory signaling and glycolysis, displayed increased mRNA levels in skeletal muscle tissue from old N. furzeri. These results could be confirmed by Western blot and qRT-PCR analyses.
3. Markers for DNA damage and senescence increased in skeletal muscle tissue from old N. furzeri.
4. Cells derived from young and old N. furzeri skeletal muscle could be isolated and cultured for many passages. These cells were a mix of different cell types with properties and features of the native tissue. They could be used for treatment with drugs and/small compounds modulating the epigenetic landscape via specific interference with methylating enzymes.
5. DNA methylation and hydroxy-methylation were found to go in different directions in skeletal muscle and heart tissue from N. furzeri: while increasing in skeletal muscle tissue, a both DNA modifications declined in heart tissue with age.
6. In the heart of N. furzeri microRNA expression changes with age were assed with sequencing in collaboration with the FLI, Jena. It could be demonstrated that miRNA expression is age-dependent. Particular focus was on miR-29 and its target genes: miR-29 was highly upregulated in heart and skeletal muscle tissue, while target genes such as collagens and dnmts were reduced with age in the heart of N. furzeri.
7. Cardiac function remained stable with age and no accumulation of collagens could be found when comparing hearts of young and old N. furzeri despite the increase of markers for oxidative stress.
8. Cell culture experiments with human cardiac fibroblasts revealed that miR-29 is upregulated with increasing age of the donor. In addition to that, it could be shown that miR-29 is positively regulated by oxidative stress.
9. A zebrafish mutant with modified expression of miR-29 that was created in collaboration with the SNS, Pisa, presented a severe hypoxic phenotype and an altered mRNA expression profile compared to wild type control zebrafish. Cardiac dysfunction and hypertrophy were observed as well as an increase in DNA methylation and collagens.
Taken together, it could be shown that the aging process in skeletal muscle and heart tissue from N. furzeri leads to a series of changes on epigenetic levels. It remains to be elucidated whether these changes are result or cause for further changes of mRNA expression, protein levels and pathophysiology, yet the N. furzeri represents a promising research model for further aging studies.
In the first part of this work, the development of a novel two-dimensional native gel electrophoretic system (2-D BN/hrCNE) is described. This new system simplifies proteomics and biochemical analysis of mega protein complexes that are dissociated into the constituent complexes during 2-D electrophoresis, thereby reducing the complexity of the system considerably. This technique is exceptionally well suited for the in-gel detection of fluorescence-labeled proteins and the identification of individual enzymes and protein complexes by specific in-gel assays on native gels.
In the second part, a new technique for the native immunoblotting of blue native gels (NIBN) was developed. This new technique allows for the identification of conformation-specific antibodies and the discrimination of antibodies recognizing linear epitopes of denatured proteins. Identification of conformation-specific antibodies is becoming increasingly important not only for the electron microscopic identification of native proteins but also for structural investigations in general. For this purpose, a commonly used protocol for Western blotting of blue native gels was modified in such a way that the native state of proteins and protein complexes was retained throughout the complete protocol. Instead of using the denaturing methanol in Western blotting protocols, mild detergents such as Tween 20, digitonin and Brij 35 were used for the obligatory removal of protein bound Coomassie-dye.
The detection of respiratory complex I by activity staining on the blot membrane demonstrated that all three non-ionic detergents preserved the native state of complex I. The native state of the enzyme on the blot membrane was also monitored and confirmed with the help of a set of conformation-specific antibodies. NIBN can be used as a simple alternative method to the demanding native ELISA to screen for conformation-specific antibodies for structural studies. Unlike the time consuming native ELISA, NIBN does not require introduction of appropriate affinity tags and purification of the target protein by chromatography. Thus, the NIBN technique is especially useful for microscale projects and for proteins not easily accessible to genetic manipulation.
The third part aimed at identification of the immediate protein interaction partners of Cox26, a hydrophobic protein that has been identified by our group as a novel component of yeast respiratory supercomplex. Multi-dimensional electrophoretic techniques were applied to identify non-covalent and covalent protein-protein interactions of Cox26. Three-dimensional electrophoresis (BNE/BNE/SDS-PAGE) gave both qualitative and quantitative information on covalent and non-covalent interactions of Cox26 and subunits of cytochrome c oxidase (complex IV), and showed that most of the Cox26 protein was non-covalently bound to the complex IV moiety of the respirasomes. Four-dimensional electrophoresis (BNE/BNE/SDS/SDS-PAGE) applying reducing and non-reducing conditions revealed that a minor fraction of Cox26 used a single cysteine residue in the center of a predicted transmembrane helix to form a disulfide bond with the Cox2 subunit of complex IV. A structural role of Cox26 protein in the assembly/stability of respiratory strings or patches has been suggested.
The last part of this work focused on the isolation and characterization of native and morphologically intact nucleoids from bovine heart mitochondria, since only a few studies on nucleoid organization and composition have been carried out on mammalian tissues. The nucleoids appeared as distinct bands (apparent mass around 30-36 MDa) in blue native-PAGE on large pore gels. The moderate variation in particle size seems to reflect variations in the binding of loosely nucleoid-associated components like respiratory chain complexes. The estimated 30-36 MDa mass of nucleoids on native gels suggested that each nucleoid contains one mtDNA molecule provided that nucleoids contains equal amounts of DNA, protein and RNA (Miyakawa et al., 1987).
Electron microscopic analysis of native nucleoids, which was performed by Dr. Karen Davies from the Max-Planck-Institute of Biophysics, Department of Structural Biology, Frankfurt, showed homogenous pool of particles with dimensions in 85x100 nm (in negative stain) and 100x150 nm (in cryo-tomography). Some of the nucleoids showed dumbbell-shape indicating dimerization of nucleoids. Recent EM and high-resolution light microscopy analysis of mammalian nucleoids have reported that nucleoids have a size of 70 nm in average. We also observed the same size of 70 nm in cryo-tomogramms when we applied harsher treatment of the native nucleoid particles with dimensions 100x150 nm. This observation is in agreement with published nucleoid sizes from both EM and high-resolution light microscopy, if we assume that native nucleoids have been dissociated under harsher treatment.
The protein composition of bovine heart mt-nucleoids was analyzed by a number of complementary approaches to identify low and highly abundant, easily dissociating and tightly bound proteins, and to rank the 90 most abundant mt-nucleoid proteins. Native and denaturing gel electrophoresis techniques were coupled to LC-MS/MS to achieve a comprehensive protein component analysis. Qualitative MS analysis of highly purified nucleoids identified more than 400 proteins, including well known nucleoid proteins such as mitochondrial transcription factor and mtDNA-binding protein (TFAM), mitochondrial single-stranded DNA-binding protein (mtSSB), mitochondrial DNA polymerase subunit gamma-2 (POLG2) and mitochondrial helicase C26H10ORF2 protein (Twinkle). These proteins were ranked according to Mascot scores, and sorted according to presumed functional properties. A large group of proteins involved in protein synthesis comprised an almost complete set of subunits of mitochondrial ribosomes suggesting that the nucleoids contained significant amounts of mitochondrial ribosomes. Identification of sixty six proteins from the oxidative phosphorylation (OXPHOS) system comprising around 100 proteins in total suggested that OXPHOS proteins are also associated with mt-nucleoids.
Interestingly, TFAM, described as a main mtDNA packaging factor in human and other mammalian cells, was not confirmed here as a major nucleoid component from bovine heart mitochondria. Fluorescence staining of protein spots on 2-D IEF/SDS gels clearly identified TFAM, but according to the stain intensity, this protein did not rank in the list of the 90 most abundant nucleoid proteins. Western blot analysis of sucrose gradient fractions revealed an enrichment of putative TFAM isoform in nucleoid fractions. Unexpectedly, the uncharacterized mitochondrial protein Es1 was identified as the most abundant nucleoid protein in bovine heart nucleoids instead. This implicates that nucleoid organization may differ between species and tissues. A functional characterization of Es1 is required to clarify its role in mammalian nucleoids.
Ubiquitin and the ubiquitin-like protein ATG8 are covalently attached to their respective targets via a coordinated cascade involving E1 activating, E2 conjugating and E3 ligating enzymes. Whereas ubiquitin is conferred to proteins as mono- and/or polymer(s) to alter their stability, localization and/or activity, the ubiquitin-like modifier (UBL) ATG8 is conjugated to the phospholipid phosphatidylethanolamine (PE). The best understood function of ATG8 is during autophagy where ATG8-PE conjugates are incorporated into both layers of incipient autophagosomes and serve as multipurpose docking sites for autophagosomal cargo receptors as well as regulatory factors (termed adaptors) that drive formation and maturation of autophagosomes. Mammalian cells harbor six ATG8 family members that can be subclassified into the LC3- and GABARAP-family and that can all be lipidated. However, it is currently unclear to what extent these proteins are functionally redundant or fulfil unique roles.
Cullin-RING ligase complexes (CRLs) are modular E3 ubiquitin ligases that comprise a RING-finger protein that associates with the ubiquitin-charged E2 enzyme, a substrate recruiting module as well as a cullin scaffold as a linker between RING protein and substrate adaptor. Whereas SCF (SKP1-CUL1-F-box protein) complexes, the most studied CRLs, harbor cullin-1 (CUL1) as scaffold and F-box proteins as substrate binding modules, CUL3-containing CRL complexes employ cullin-3 (CUL3), RING-box protein 1 (RBX1) and BTB proteins as substrate adaptors. Here, the BTB domain serves as binding interface for CUL3 and is usually complemented by an additional protein-protein interaction domain such as MATH or Kelch that mediates binding to the substrate of the E3 ligase complex.
Besides ubiquitylation, guanine nucleotide binding is another common way to regulate protein activity and signaling in cells. Here, small Rho GTPases cycle between active and inactive states by binding of the guanine nucleotides GTP or GDP with the help of regulatory proteins. Whereas GTPase-activating proteins (GAP) render RAC1 inactive by facilitating GTP hydrolysis, guanine exchange factors (GEF) such as T-lymphoma invasion and metastasis-inducing protein 1 (TIAM1) activate RAC1 by stimulating the exchange of GDP to GTP. Local control of RAC1 activity is essential to allow a specific cellular response to stimuli such as growth factors or migratory impulses.
This study reports an unexpected link between the GABARAP subfamily of mammalian ATG8 proteins, the ubiquitin proteasome system and RAC1 through the ubiquitylation of the RAC1 GEF TIAM1. The Kelch repeat and BTB domain-containing proteins 6 (KBTBD6) and 7 (KBTBD7) were established as heterodimeric substrate adaptors for CUL3. Interestingly, a thorough proteomic analysis revealed a number of putative substrates but, out of 11 substrate candidates tested, only the RAC1 GEF TIAM1 appeared to be influenced by depletion of CUL3KBTBD6/KBTBD7. Binding studies showed that KBTBD7 binds TIAM1 via the Kelch repeats and that this binding was markedly enhanced when CUL3 activation was abolished upon treatment with the neddylation inhibitor MLN4924. Also, total TIAM1 abundance was increased upon CUL3KBTBD6/KBTBD7 depletion and accumulation of TIAM1 upon proteasome inhibition suggested that TIAM1 is degraded via the proteasome. In vivo ubiquitylation assays and denaturing immunoprecipitations as well as mass spectrometrical analysis confirmed that CUL3KBTBD6/KBTBD7 ubiquitylates TIAM1 at two distinct lysines (K1404 and K1420) close to its C-terminus.
Previously, KBTBD6 and KBTBD7 were found as interactors of several members of the human ATG8 family of proteins in a proteomic study analyzing the human autophagy network. This association was confirmed in the present work. Furthermore, peptide array technology and mutational analysis revealed that KBTBD6 and KBTBD7 employ a classical ATG8-family interacting motif (AIM; also referred to as LC3-interacting region or LIR) as binding interface. The AIMs of KBTBD6 (W-V-R-V) and KBTBD7 (W-V-Q-V) fulfil the consensus AIM sequence motif (F/W/Y1-X2-X3-I/L/V4) and are preceded by several acidic residues and serines. A series of structural and cell biological experiments revealed a binding preference for the GABARAP subfamily of human ATG8 proteins and most importantly, a requirement of the GABARAP-KBTBD6 and -KBTBD7 interaction for TIAM1 ubiquitylation. The finding that TIAM1 binding to KBTBD6 and KBTBD7 AIM mutants was diminished raised the possibility that GABARAP binding mediates the recruitment of CUL3KBTBD6/KBTBD7 to membranes where TIAM1 is localized. Interestingly, colocalization of KBTBD6, GABARAPL1 and TIAM1 in punctuate structures could be observed. Since only a very small fraction of GABARAPL1 colocalized with LC3B, and colocalization between KBTBD6 and LC3B was not observed, these vesicular structures are most likely distinct from autophagosomes. Furthermore, TIAM1 ubiquitylation was reduced when GABARAP, but not LC3B, was depleted or when lipidation of GABARAP was prevented.
Stabilization of TIAM1 upon KBTBD6 and/or KBTBD7 depletion led to elevated TIAM1-dependent RAC1 activity, altered actin morphology with increased cortical actin and loss of vinculin foci. Re-introduction of wild-type KBTBD6 or KBTBD7 but not AIM mutants reverted all these phenotypes. Moreover, depletion of KBTBD6 or KBTBD7 in human breast cancer cells massively increased their invasiveness, whereas TIAM1 knockdown had the opposite outcome. All physiological effects of KBTBD6 and KBTBD7 depletion were inhibited by additional depletion of TIAM1 or RAC1 confirming that the phenotypes observed are indeed mediated by the CUL3KBTBD6/KBTBD7-TIAM1-RAC1 signaling pathway. Intriguingly, KBTBD6 and KBTBD7 were not subject to autophagosomal degradation, thereby establishing a new function for GABARAP proteins beyond autophagosomal degradation in providing a signaling platform for recruitment of the E3 ligase CUL3KBTBD6/KBTBD7 in close proximity to its substrate TIAM1, enabling localized ubiquitylation.
Local restricted control of RAC1 activity by ubiquitylation has been described for TIAM1-RAC1 signaling previously. Examples are HECT, UBA and WWE domain-containing protein 1 (HUWE1)-mediated TIAM1 ubiquitylation that occurs predominantly at cell-cell-junctions in response to hepatocyte growth factor stimulation in MDCKII cells or inhibition of RAC1 activity by the RAC1 GAP protein BCR (breakpoint cluster region) at the leading edge of astrocytes through binding to the TIAM1-Par (polarity) complex. SCFBTRC mediates ubiquitylation of TIAM1 in response to mitogens or DNA damage, though it has not been explored whether this regulation is spatially restricted. Thus, this study adds a novel layer of complexity to the spatial regulation of RAC1 signaling by implicating membrane-bound human ATG8 proteins in this process.
Also, this study is the first report specifically implicating the GABARAP proteins in cellular signaling events. It will be interesting to explore whether the concept of localized signaling mediated by GABARAPs applies to other substrates of CUL3KBTBD6/KBTBD7 and membranerelated signaling processes in which GABARAP proteins are involved. Controlling RAC1 activity at GABARAP-decorated membranes might also be important for trafficking events or autophagy since it was described that RAC1 has an inhibitory function on autophagy. Therefore, spatial restricted ubiquitylation of TIAM1 resulting in specific deactivation of RAC1 could promote the autophagic process when locally needed. Although the catalytic mTOR inhibitor Torin1 and the lysosomal H+ ATPase inhibitor BafilomycinA1 promoted TIAM1 ubiquitylation by increasing the pool of membrane-conjugated GABARAP, but other signals that stimulate GABARAP-KBTBD6/KBTBD7 association and subsequent TIAM1 ubiquitylation are to be identified. Besides, determining the KBTBD6/KBTBD7 binding site in TIAM1 or uncovering a deubiquitylating enzyme (DUB) that locally counteracts the ubiquitylation of TIAM1 will enable a better comprehension of the complete localized signaling cascade.
Biological ageing is a degenerative and irreversible process, ultimately leading to death of the organism. The process is complex and under the control of genetic, environmental and stochastic traits. Although many theories have been established during the last decades, none of these are able to fully describe the complex mechanisms, which lead to ageing. Generally, biological processes and environmental factors lead to molecular damage and an accumulation of impaired cellular components. In contrast, counteracting surveillance systems are effective, including repair, remodelling and degradation of damaged or impaired components, respectively. Nevertheless, at some point these systems are no longer effective, either because the increasing amount of molecular damages can not longer be removed efficiently or because the repairing and removing mechanisms themselves become affected by impairing effects. The organism finally declines and dies. To investigate and to understand these counteracting mechanisms and the complex interplay of decline and maintenance, holistic and systems biological investigations are required. Hence, the processes which lead to ageing in the fungal model organism Podospora anserina, had been analysed using different advanced bioinformatics methods. In contrast to many other ageing models, P. anserina exhibits a short lifespan, a less biochemical complexity and it provides a good accessibility for genetic manipulations.
To achieve a general overview on the different biochemical processes, which are affected during ageing in P. anserina, an initial comprehensive investigation was applied, which aimed to reveal genes significantly regulated and expressed in an age-dependent manner. This investigation was based on an age-dependent transcriptome analysis. Sophisticated and comprehensive analyses revealed different age-related pathways and indicated that especially autophagy may play a crucial role during ageing. For example, it was found that the expression of autophagy-associated genes increases in the course of ageing.
Subsequently, to investigate and to characterise the autophagy pathway, its associated single components and their interactions, Path2PPI, a new bioinformatics approach, was developed. Path2PPI enables the prediction of protein-protein interaction networks of particular pathways by means of a homology comparison approach and was applied to construct the protein-protein interaction network of autophagy in P. anserina.
The predicted network was extended by experimental data, comprising the transcriptome data as well as newly generated protein-protein interaction data achieved from a yeast two-hybrid analysis. Using different mathematical and statistical methods the topological properties of the constructed network had been compared with those of randomly generated networks to approve its biological significance. In addition, based on this topological and functional analysis, the most important proteins were determined and functional modules were identified, which correspond to the different sub-pathways of autophagy. Due to the integrated transcriptome data the autophagy network could be linked to the ageing process. For example, different proteins had been identified, which genes are continuously up- or down-regulated during ageing and it was shown for the first time that autophagy-associated genes are significantly often co-expressed during ageing.
The presented biological network provides a systems biological view on autophagy and enables further studies, which aim to analyse the relationship of autophagy and ageing. Furthermore, it allows the investigation of potential methods for intervention into the ageing process and to extend the healthy lifespan of P. anserina as well as of other eukaryotic organisms, in particular humans.