Refine
Document Type
- Doctoral Thesis (10)
Has Fulltext
- yes (10)
Is part of the Bibliography
- no (10)
Keywords
- Biomarker (1)
- Colorectal Cancer (1)
- MEK inhibition (1)
- Organoids (1)
Institute
Despite all advancements in cancer research and clinical practice, cancer remains a life- threatening disease with an increasing incidence. According to a 2018 WHO forecast, cancer incidence will double to approximately 37 million new cancer cases by 2040. Today, clinical management of cancer is based on a "one-fits-all" strategy. Most cancers are still treated by surgical therapy followed by adjuvant or neoadjuvant chemotherapy based on rather strict guidelines (S3 guidelines in Europe) which are based on studies of large cohorts of patients with the same tumor entity. While this approach has led to substantial increases in progression-free survival and overall patient survival, most patients do not benefit from the administered treatment regimen. One reason for this is intra-tumor heterogeneity, which results from clonal evolution between cancer cells and their environment. This means that cancer patients may respond differently to a particular drug due to the different mutation patterns of their tumor cells. Therefore, patients should be screened in advance for reliable cancer biomarkers that definitively predict whether they will respond to a particular therapy. This would increase the probability of a successful treatment.
Colorectal cancer (CRC) is the third most diagnosed cancer and the second leading cause of cancer deaths worldwide. The main cause of death in CRC is a metastatic disease, which is presented in 20 % of patients and eventually develops in more than 30 % of early-stage patients. Despite the significant increase (to more than 30 months) in median survival with the development of cytotoxic agents and the introduction of targeted therapy, the progression-free survival in the first-line setting has remained largely unchanged over the past decade.
The heterogeneity in CRC is characterized by alterations in multiple signaling pathways that affect cellular functions such as cell proliferation or apoptosis. Commonly affected signaling pathways include the mitogen-activated protein kinase (MAPK)- and the transforming growth factor-β/bone morphogenetic protein (TGF-β/BMP)-pathway. Alterations in the TGF-β/BMP pathway, due to mutations in the SMAD4 gene (mothers against decapentaplegic homolog 4), are associated with different drug response and promote resistance to chemotherapy. In addition, they are associated with a higher recurrence rate.
SMAD4 is one of the most common cancer driver genes, and mutations occur in up to 15 % of CRC cases. Therefore, there is an urgent need for therapeutic agents that can specifically target SMAD4-mutated tumors.
The aim of the present study was the identification of the clinical relevance of the SMAD4 gene and the investigation of its suitability as a potential biomarker in CRC.
For this purpose, I investigated sibling patient-derived organoids (PDOs) derived from different regions of a chemo-naïve CRC tumor. PDOs are 3D cell cultures that reliably recapitulate the architecture of the tissue of origin, as well as preserve the genomic background and intra-tumor heterogeneity. The sibling PDOs (R1R361H and R4wt) shared the most common CRC mutations, such as KRASG12D (kirsten rat sarcoma), PIK3CAH1047R (phosphatidylinositol-4,5-bisphosphate 3-kinase, catalytic subunit alpha), and TP53C242F (tumor protein 53), but differed in a SMAD4R361H mutation and showed a different drug response. The single nucleotide variant R361H of the SMAD4 gene is among the most common pathogenic alterations in various cancers, including CRC.
The sibling PDOs showed significant differences in response to the MEK-inhibitors cobimetinib, trametinib, and selumetinib. MEK-inhibitors are antineoplastic agents that inhibit the function of MEK1 and MEK2, preventing phosphorylation of transcription factors, which leads to inhibition of tumor cell proliferation. MEK-inhibitors are approved for the treatment of malignant melanoma. Currently, they are in phase-III clinical trials for the treatment of patients with metastatic CRC.
To investigate whether SMAD4R361H is responsible for sensitivity to MEK-inhibitors, Iestablished three syngeneic PDOs harboring a SMAD4R361H mutation using the CRISPR/Cas9 genome editing system. All CRISPR-PDOs were significantly more sensitive to the MEK-inhibitors, compared to R4wt. I have shown that the SMAD4R361H mutation is responsible for sensitivity to MEK inhibition in CRC models and may be a predictive biomarker.
To test this hypothesis, I examined 62 CRC PDO models and treated them with the MEK-inhibitors cobimetinib, trametinib, and selumetinib. All models that had a pathogenic mutation or deletion in the SMAD4 gene (15 %) were sensitive to cobimetinib, 10 % of models were sensitive to trametinib, and 8 % were sensitive to selumetinib.
I performed transcriptome (RNA sequencing) and proteome analyses using the DigiWest® method to investigate the mechanism underlying MEK-inhibitor sensitivity.
DigiWest® is a Luminex® bead-based analysis that allows the simultaneous analysis of over 100 (phospho-)proteins. The transcriptome and proteome data support the observation that MEK inhibition primarily affects SMAD4R361H PDOs. Furthermore, I have shown that activation of the BMP signaling pathway in organoids with wild-type SMAD4 appears to be responsible for resistance to MEK-inhibitors. Thus, a genetic alteration in the BMP signaling pathway, beyond SMAD4, could lead to sensitivity to MEK-inhibitors.
I identified four genes involved in the TGF-β/BMP signaling pathway that are frequently mutated in CRC and grouped them into the so-called SFAB-signature (SMAD4, FBXW7 (F-box/WD repeat-containing protein 7), ARID1A (AT-rich interactive domain-containing protein 1A), or BMPR2 (Bone morphogenetic protein receptor type II). Clinical data show that approximately 36 % of CRC patients have at least one pathogenic mutation in these genes.
I tested all 62 CRC PDO models and found a significant positive prediction for sensitivity to cobimetinib (95 %) and selumetinib (70 %) for the SFAB-signature. Trametinib and the newly approved MEK-inhibitor binimetinib showed a similar trend. Therefore, the SFAB-signature has high predictive power for response to MEK-inhibitors and could be used as a predictive biomarker panel.
The current clinically used biomarkers for CRC are based on the mutation status of driver genes KRAS and BRAF, which are present in up to 50 % and 10 % of CRC, respectively. Investigation of molecular alterations in CRC revealed that mutations in the KRAS gene, which is downstream of EGFR (epidermal growth factor receptor) in the MAPK-pathway, interfere with an anti-EGFR-antibody therapy (e.g., cetuximab). Therefore, cetuximab is only relevant for RAS wild-type tumors. However, approximately 40 % of patients with RAS wild-type status do not respond to this treatment.
About 53 % of CRC PDO models carry a pathogenic RAS mutation, about 10 % harbor a pathogenic BRAF mutation. Both, the RAS and RAF status alone as well as the combination of RAS and RAF status with SFAB-signature did not provide a better prediction of sensitivity to MEK inhibition.
Juvenile Neuronal Ceroid Lipofuscinosis (JNCL) is a rare inherited childhood neurodegenerative disease that is caused by a mutation in the gene CLN3. The function of the protein produced by the gene has remained elusive, and therefore the disease mechanism of JNCL is as of yet unknown. The disease is fatal, and no cure is currently available. We believe that simvastatin shows promise as a possible treatment. Simvastatin is well tolerated in children, and as currently no other viable, less invasive treatment for JNCL exists, at least pilot-scale clinical trials for this new off-label use of simvastatin are warranted.
The protein CLN3 has been indicated to have several different subcellular localizations and functions, but conclusive evidence about its role in cellular metabolism is lacking. It is also unclear why the mutation causes the distinct phenotype of the JNCL disease. In order to bring lucidity to the issue, we set out to identify metabolic pathways related to the phenotype of JNCL by using Multi-Epitope Ligand Cartography (MELC) and the related field of toponomics. Toponomic methods are required to process the massive amount of data generated by the MELC runs in order to extract information from them.
Our disease model of choice was the CLN3Δex7/8 knock-in mouse. To separate cause from effect, we compared embryonal wild type and mutant mouse brains to their adult counterparts. The first analyses revealed progressively abnormal Combinatorial Molecular Patterns (CMPs, an unit of toponomic data) related to cholera toxin/ganglioside 1 (Ctx/GM1), which is a membrane microdomain marker.
Cholesterol is an essential part of microdomains, so we utilized filipin staining to see if there were actual changes in cholesterol concentration and localization between healthy and diseased animals. After the disturbance in cholesterol metabolism was verified, we investigated the metabolic pathway that synthesizes cholesterol, the mevalonate pathway. Simvastatin is a drug that specifically down-regulates the mevalonate pathway. Fish oil affects lipid homeostasis and has some effects similar to those of simvastatin, and both of these drugs have previously been studied for their effects on neurodegenerative diseases. After treatment of mice with these drugs, highperformance liquid chromatography (HPLC) measurements on the brain homogenate showed a decrease in levels of farnesyl pyrophosphate (FPP) and geranyl-geranyl pyrophosphate (GGPP), products of the mevalonate pathway, confirming the effect of these drugs on the brains of the animals. Analyses of motor function of the mice further supported the notion that simvastatin had a positive effect on the condition of the diseased animals.
CMP analyses from the simvastatin treated mice showed a rescue of the Ctx/GM1 CMPs, suggesting at least a partial restoration of membrane microdomain homeostasis. Filipin staining revealed reversion of the apparent cholesterol depletion in the adult mutant mouse hippocampus by simvastatin. Interestingly, an additional effect of the treatment was found: simvastatin also affected glutamate receptor homeostasis, especially as regarding to N-methyl-D-aspartate (NMDA) and alphaamino-3-hydroxyl-5-methyl-4-isoxazole-propionate (AMPA) receptors. This finding suggested that excitotoxicity could be a part of the disease process, and pointed towards glutamate receptors as possible therapy targets. This is in line with previous studies that have shown that attenuation of AMPA receptors and L voltage-dependent channels improve the phenotype of a JNCL mouse and cell model, respectively.
Simvastatin mediates many of its effects via downregulation of the mevalonate pathway products, such as isoprenoids and cholesterol. However, simvastatin also has multiple pleiotropic effects that include suppression of excitotoxicity and granting neuroprotection. It is apparent that simvastatin treatment has a positive effect on JNCL mice, but if its effects are mediated via cholesterol (and membrane microdomains), isoprenoids (and isoprenylated proteins) or via a fully cholesterol independent mechanism remains to be solved.
In this study we have shown that with the MELC method and toponomics it is possible to approach rare diseases with confounded disease mechanisms with a hypothesis-free approach, to identify possible drug targets, and to monitor the effects of the drugs on treated individuals. This should open up a new avenue in the research of the many diseases that so far have avoided all attempts at discerning their nature.
CAP (c-Cbl assoziiertes Protein) ist ein Adapterprotein, welches zusammen mit ArgBP2 und Vinexin die SoHo-Protein-Familie bildet. Es besitzt in seinem N-terminalen Bereich eine SoHo-Domäne und im C-Terminus drei SH3-Domänen, über welche es mit einer Vielzahl von Proteinen interagieren kann. CAP spielt bei verschiedenen Signaltransduktionsvorgängen und der Reorganisation des Aktinzytoskelettes eine Rolle. So wurde ihm eine Funktion im PI3-Kinase-unabhängigen Insulinsignalweg zugeschrieben. In Zell-Zell-Kontakten ist CAP zusammen mit Nectin und Afadin Bestandteil des NAPSystems, welches parallel zu Cadherin-Catenin-Adhäsionen existiert. In Fokalkontakten bindet CAP an FAK, Paxillin und Vinculin, welche wichtig für die Regulation der Zell-Matrix-Adhäsionen sind. In der vorliegenden Arbeit sollte die Funktion von CAP bei der Assoziation mit dem Aktinzytoskelett näher untersucht werden. Es wurde gezeigt, daß die Kolokalisation von CAP mit Vinculin und Aktin dynamisch und vom Ausbreitungsgrad der Zelle und dem Expressionsniveau des Proteins abhängig ist. Ohne funktionelle SH3-Domänen lokalisiert CAP nicht mehr in Fokalkontakten, kann aber bei Überexpression noch eine Ausbildung von Streßfasern induzieren. Die Funktionalität der zweiten und der dritten SH3-Domäne von CAP ist für die Zelladhäsion von Epithelzellen von Bedeutung, da Konstrukte, die in konservierten Aminosäuren der Domänen mutiert worden sind, eine verringerte Zellausbreitung aufweisen. CAP wurde zudem als ein Interaktionspartner und Substrat der Tyrosinkinasen c-Abl und c-Src charakterisiert. Die Assoziation mit den Kinasen wird dabei vor allem über den C-Terminus von CAP vermittelt, und CAP kann direkt mit der Src-Kinase interagieren. CAP wird von c-Abl überwiegend an Tyrosin 360, und von c-Src bevorzugt an Tyrosin 326 phosphoryliert. Diese Tyrosine liegen innerhalb bekannter Konsensus-Motive der Kinasen. Phosphorylierungsdefiziente CAP-Konstrukte sind noch dazu in der Lage, mit Vinculin und Aktin zu kolokalisieren. Tyrosin 326 hat jedoch einen Einfluß auf die Zellausbreitung auf Fibronektin, da ein in dieser Aminosäure mutiertes Konstrukt einen Defekt hierbei aufweist. Den genauen Mechanismus, über welchen CAP seinen Einfluß auf die Zell-Matrix-Adhäsion ausübt, ist noch ungeklärt. Durch die hier erstmals aufgezeigte Phosphorylierung des Adapterproteins erweitern sich dessen Interaktionsmöglichkeiten durch mögliche Bindungen an SH2-Domänen-enthaltende Proteine. CAP könnte als Bindeglied zwischen c-Abl und c-Src und weiteren zytoskelettalen Komponenten dazu beitragen, die Kinasen an Fokalkontakten zu verankern und ihnen neue Substrate zuzuführen. Die Interaktion mit CAP könnte dabei die Aktivität der Kinasen erhöhen. Weiterführende Studien werden die in diesem Signalweg nachgeschalteten Komponenten untersuchen und auch eine putative Rolle der Phosphorylierung von CAP in den bereits bekannten zellulären Zusammenhängen wie dem Insulinsignalweg näher beleuchten.
Alzheimer’s disease (AD) is a common, age associated neurodegenerative disease that manifests as progressive dementia and is characterized by accumulation of the amyloid beta (Aβ) peptide which is a processing product of a transmembrane protein termed Alzheimer Amyloid Precursor Protein (APP). The Aβ peptide is generated by a sequential proteolytic processing of APP by two distinct proteases that are termed β- and γ-secretase. The β-secretase, also called BACE-1 or memapsin 2, belongs to the family of aspartyl proteases. BACE-1 evidently cleaves APP in an acidic endosomal compartment after endocytosis of APP, thereby facilitating Aβ peptide generation.
Sorting of transmembrane proteins is generally controlled by sorting signals in the cytoplasmic domains of the cargo proteins. The short cytoplasmic tail of BACE-1 with 23 amino acids contains a sorting signal of the acidic cluster, di-leucine (ACDL) type. The two Leu residues in this determinant are important for the clathrin mediated endocytosis of BACE-1, whereas the acidic residues together with the Leu are required for the endosomal sorting and recycling of BACE-1 back to the plasma membrane. The ACDL motif binds to the members of the GGA (Golgi-localized γ ear-containg ARF- binding proteins) family (GGA1-GGA3) that are involved in the sorting of BACE-1.
One of the major aims of this study was to address the role of flotillins in the intracellular sorting of BACE-1. This study shows that flotillin-1 directly binds to the di-leucine motif in the cytoplasmic tail of BACE-1, whereas flotillin-2 only shows an association mediated by flotillin-1. Flotillin-1 competes with GGA2 for the binding to BACE-1 tail, and thus influences the endosomal sorting of BACE-1. Importantly, depletion of flotillins results in an altered localization of the wildtype BACE-1, whereas the plasma membrane resident Leu to Ala (LLAA) mutant is not affected. Flotillin knockdown results in an accumulation of BACE-1, implicating reduced degradation and enhanced stability of this protease. Thus, flotillins appear to be important for the cellular targeting of BACE-1 and also influence the amyloidogenic processing of APP, as demonstrated by an increase in the amyloidogenic C-99 processing fragments.
When flotillin depleted cells were subjected to apoptotic stresses including Aβ25-35 synthetic peptide (inducer of the extrinsic apoptosis pathway) or several chemotherapeutic agents (staurosporine, brefeldin A, doxorubicin, carboplatin and paclitaxel: intrinsic apoptosis pathway) and cytotoxicity was determined, various apoptotic markers were activated in flotillin depleted cells. Caspase-3 and GGA3 are well accepted apoptosis markers and an enhanced caspase-3 cleavage was detected upon STS induced apoptosis in SH-SY5Y, HeLa, and HaCaT cell lines and increased GGA3 cleavage was observed in MCF7 cell line.
One of the major reasons for the apoptotic sensitivity in the absence of flotillins was a PI3K/Akt signaling defect. Neuroblastoma cells depleted of flotillins showed diminished levels of total Akt, phospho-Akt and phospho-ERK upon STS induced apoptosis. Since PI3K/Akt was the primary survival pathway affected upon STS induced apoptosis, ectopic expression of Akt in neuroblastoma cell line reduced caspase-3 cleavage and retarded apoptosis.
The direct downstream target of Akt is FOXO3a, whose localization was investigated in flotillin depleted cells. A major proportion of FOXO3a was localized in the nucleus of flotillin knockdown cells, implicating that FOXOs are active in these cells and subsequently trigger the transcription of death genes. Strikingly, an essential anti-apoptotic molecule and a major cancer target, Mcl-1, was inherently downregulated in flotillin knockdown cells. Mcl-1 is a chief member of the Bcl-2 family as it plays a pivotal role in cell survival and it is a critical protein in cancer therapeutics as suppression of Mcl-1 protein can curtail the survival and growth of tumorous cells.
Neuroblastoma cells were rescued from undergoing permanent damage due to STS induced apoptosis by overexpression of anti-apoptotic Bcl-2. Phorbol esters are well known PKC activators, and pre-treatment of neuroblastoma cells with phorbol esters along with staurosporine reduced caspase-3 cleavage.
These results demonstrate that absence of flotillins can sensitize cellular systems to apoptosis induction. The two main characteristics of cancer cells include resistance to apoptosis and unresponsiveness to chemotherapeutic agents. It is a well established fact that impaired apoptosis is central to tumour development. This study implicates that the downregulation of flotillin function can trigger cellular susceptibility and enhances apoptosis in response to conventional chemotherapeutic agents. Therefore, flotillins can serve as vital regulators in providing a more rational approach in molecular-targeted therapies for receding cancer growth and survival.
Reggie-1 (flotillin-2) and reggie-2 (flotillin-1) are membrane microdomain proteins which are associated with the membrane by means of acylation. They influence different cellular signaling processes, such as neuronal, T-cell and insulin signaling. Upon stimulation of the EGF receptor, reggie-1 becomes phosphorylated and undergoes tyrosine 163 dependent translocation from the plasma membrane to endosomal compartments. In addition, reggie-1 was shown to influence actindependent processes. Reggie-2 has been demonstrated to affect caveolin- and clathrin-independent endocytosis. Both proteins form homo- and hetero-oligomers, but the function of these oligomers has remained elusive. Moreover, it has not been clarified if functions of reggie-1 are also influenced by reggie-2 and vice versa. The first aim of the study was to further investigate the interplay and the heterooligomerization of reggie proteins and their functional effects. Both reggie proteins were individually depleted by means of siRNA. In different siRNA systems and various cell lines, reggie-1 depleted cells showed reduced protein amounts of reggie-1 and reggie-2, but reggie-2 knock down cells still expressed reggie-1 protein. The decrease of reggie-2 in reggie-1 depleted cells was only detected at protein but not at mRNA level. Furthermore, reggie-2 expression could be rescued by expression of siRNA resistant wild type reggie-1-EGFP constructs, but not by the soluble myristoylation mutant G2A. This mutant was also not able to associate with endogenous reggie-1 or reggie-2, which demonstrates that membrane association of reggie-1 is necessary for hetero-oligomerization. In addition, fluorescence microscopy studies and membrane fractionations showed that correct localization of overexpressed reggie-2 was dependent on co-overexpressed reggie-1. Thus, hetero-oligomerization is crucial for membrane association of reggie-2 and for its protein stability or protein expression. Moreover, the binding of reggie-2 to reggie-1 required tyrosine 163 of reggie-1 which was previously shown to be important for endosomal translocation of reggie-1. Since reggie-2 was implicated to function in clathrin- and caveolin-independent endocytosis pathways, the effect of reggie-2 depletion on reggie-1 endocytosis was investigated. Indeed, reggie-1 was dependent on reggie-2 for endosomal localization and EGF-induced endocytosis. By FRET-FLIM analysis it could be shown that reggie heterooligomers are dynamic in size or conformation upon EGF stimulation. Thus, it can be concluded that reggie proteins are interdependent in different aspects, such as protein stability or expression, membrane association and subcellular localization. In addition, these results demonstrate that the hetero-oligomers are dynamic and reggie proteins influence each other in terms of function. A further aim was the characterization of reggie-1 and reggie-2 function in actindependent processes, where so far only reggie-1 was known to play a role. Depletion of either of the proteins reduced cell migration, cell spreading and the number of focal adhesions in steady state cells. Thus, also reggie-2 affects actin-dependent processes. Further investigation of the focal adhesions during cell spreading revealed that depletion of reggie-1 displayed different effects as compared to reggie-2 knock down. Reggie-1 depleted cells had elongated cell-matrix-adhesions and showed reduced activation of FAK and ERK2. On the other hand, depletion of reggie-2 resulted in a restricted localization of focal adhesion at the periphery of the cell and decreased ERK2 phosphorylation, but it did not affect FAK autophosphorylation. Hence, reggie proteins influence the regulation of cell-matrix-adhesions differently. A link between reggie proteins and focal adhesions is the actin cross-linking protein -actinin. The interaction of -actinin with reggie-1 could be verified by means of co-immunoprecipitations and FRET-FLIM analysis. Reggie-1 binds -actinin especially in membrane ruffles and in other locations where actin remodeling takes place. Moreover, -actinin showed a different localization pattern during cell spreading in reggie-1 depleted cells, as compared to the control cells. These results provide further insights into the function of both reggie proteins. Their interplay and hetero-oligomerization was shown to be crucial for their role in endocytosis. In addition, both reggie proteins influence actin-dependent processes and differentially affect focal adhesion regulation.
Cell-cell adhesion is an essential process during the development of multicellular organisms. It is based on various cellular junctions and ensures a tight contact between neighboring cells, enabling interactive exchanges necessary for morphological and functional differentiation and maintaining the homeostasis of healthy tissue organization. Two important types of cell-cell adhesions are the adherens junction (AJ) and the desmosome which link the actin cytoskeleton and intermediate filaments to cadherin-based adhesion sites. The core of these structures is composed of single-span transmembrane proteins of the cadherin superfamily which include, among other members, the classical cadherins, e.g. E-cadherin, as well as the desmosomal cadherins, e.g. desmoglein-3. The cytoplasmic domains of the desmosomal and classical cadherins enable interactions with proteins of the catenin family. Classical cadherins preferentially associate with β-catenin and p120-catenin, whereas desmosomal cadherins bind to γ-catenin and plakophilins. Intriguingly, γ-catenin, also known as plakoglobin, is so far the only protein known to be present both in the AJ and the desmosome.
In this study, we showed that the two homologous, membrane raft-associated proteins flotillin-1 and flotillin-2 associate with core proteins of the AJ and the desmosome in vitro and in vivo. In confluent human, non-malignant epithelial MCF10A cells and human skin cryosections, flotillin-2 colocalized with E-cadherin, desmoglein-3 and γ-catenin at cell-cell contact sites, whereas flotillin-1 showed barely any overlap with these proteins. In addition, we detected a colocalization of both flotillins with the actin-binding protein α-actinin in membrane ruffles in subconfluent and at cell-cell contact sites in confluent MCF10A cells as well as in human skin cryosections. The interaction with α-actinin was later shown to be flotillin-1 dependent by performing indirect GST pulldown experiments with purified α-actinin-1-GST in MCF10A cell lysates.
Since flotillin-2 strongly colocalized with cell-cell junctions, this suggested that flotillins might be found in complex with cell adhesion proteins. Thus, we performed coimmunoprecipitation experiments in murine skin lysates and various cell lines of epithelial origin, such as human breast cancer MCF7 cells, human keratinocyte HaCaT cells and primary mouse keratinocytes. These experiments demonstrated that flotillins, especially flotillin-2, coprecipitated with E-cadherin, desmosomal cadherins and γ-catenin in relation to the respective cell type and the maturation status of these cell-cell adhesion structures. However, since γ-catenin is so far the only protein known to be present in the AJ and the desmosome, we further assumed that the complex formation of flotillins with cell adhesion structures is mediated by γ-catenin. For this, we performed indirect GST pulldown experiments in MCF10A cell lysates with bacterially expressed, purified flotillin-1-GST, flotillin-2-GST and γ-catenin-GST and were able to verify the complex formation of adhesion proteins and flotillins in vitro. To further test if the interaction of γ-catenin and flotillins is a direct one, we used purified flotillin-1-GST or flotillin-2-GST and γ-catenin-MBP fusion proteins. Both flotillins directly interacted with γ-catenin in this in vitro assay. In addition, mapping of the interaction domains in γ-catenin by using GST fusion proteins carrying different parts of γ-catenin suggested that flotillins bind to a discontinuous γ-catenin binding domain which consists of a Major determinant around ARM domains 6-12, most likely with a major contribution of the ARM domain 7, and possibly including the NT part of γ-catenin.
To study the effect of flotillin depletion on cell-cell adhesion, we generated stable MCF10A cell lines in which flotillins were knocked down by means of lentiviral shRNAs. Staining of E-cadherin and γ-catenin in these cells showed that the localization at the cell-cell borders was significantly altered after flotillin-2 depletion, which pointed to a role for flotillin-2 in the formation of cell-cell adhesion structures in epithelial cells. Furthermore, isolation of detergent resistant membranes (DRMs) from these cells demonstrated that upon depletion of flotillin-2, a significant amount of E-cadherin and γ-catenin shifted into raft fractions. On the contrary, no change was detected in flotillin-1 knockdown cells. These observations point to a functional role of flotillin-2 in the regulation of raft association of cell-cell adhesion proteins. To gain more insight into the in vivo relevance of our findings, we next studied the function of flotillins in the skin of Flot2-/- knockout mice. Analysis of lysates prepared from the skin of one year old female animals revealed an increased expression of E-cadherin, desmoglein-1 and γ-catenin but not β-catenin, implicating that specific adhesion proteins are upregulated in flotillin-2 knockout skin.
Since flotillins are tightly associated with membrane microdomains we next studied the interaction of flotillin-2 with membrane cholesterol. Using the photoreactive cholesterol analog azocholestanol, we were able to show that flotillin-2 and cholesterol directly interacted. In addition, previous studies speculated that flotillin-2 interacts with cholesterol via two putative cholesterol recognition/interaction amino acid consensus (CRAC) motifs. Analysis of the flotillin-2 sequence revealed that flotillin-2 actually contains four putative CRAC motifs. However, using various flotillin-2 CRAC mutant GFP fusion proteins, we were able to show that none of the putative CRAC motifs is functional, which suggested that flotillin-2 interacts with membrane cholesterol, e.g., via posttranslational modifications, such as myristoylation and palmitoylation which were previously shown to be essential for membrane association of flotillin proteins.
The long sought molecular function of membrane raft-associated flotillin proteins is slowly becoming resolved, partially owing to the increasing knowledge about their interaction partners. Being ubiquitously expressed and evolutionarily highly conserved, flotillins carry out important cellular functions, one of which is the regulation of signal transduction pathways. This study shows that the signaling adaptor protein fibroblast growth factor receptor substrate 2 (FRS2) directly interacts both in vivo and in vitro with flotillin-1 (flot-1). FRS2 is an important docking protein of many receptor tyrosine kinases. It regulates downstream signaling by forming molecular complexes with other adaptor proteins and tyrosine phosphatases, and seems to be a critical mediator of sustained extracellular signal regulated kinase (ERK) activity. Flot-1 has also been implicated in the regulation of ERK activity upon EGF and FGF stimuli. Furthermore, flot-1 forms signalosomes with EGFR and the downstream components of the MAP kinase pathway. The newly discovered interaction between FRS2 and flot-1 was shown to be mediated by the phosphotyrosine binding (PTB) domain and, to a lesser extent, the C-terminus (CT) of FRS2 and by the C-terminus of flot-1. Flot-1 coprecipitated together with FRS2 from murine tissues and cell lysates, demonstrating that this interaction also takes place in vivo. Interestingly, flot-2, which shows a high homology to flot-1 and forms stable oligomeric complexes with it, does not appear to directly interact with FRS2. Novel insights into the functional role of the interaction between flot-1 and FRS2 were provided by the results showing that depletion of flot-1 affects the cellular localization of FRS2. In hepatocytes stably depleted of flot-1, FRS2 appeared to be more soluble. Furthermore, upon pervanadate stimulation of the cells, a small fraction of FRS2 was recruited into detergent resistant membranes, but the recruitment did not take place in the absence of flot-1. Triggered by the same stimulus, a fraction of FRS2 was translocated to the nucleus independently of flot-1. Overexpression of FRS2 has previously been shown to result in increased ERK activation. However, in cells depleted of flot-1, FRS2 was not able to compensate for the compromised ERK activation after EGF or FGF stimulation. This might imply that FRS2 and flot-1 are functionally interconnected and that FRS2 resides upstream of flot-1. Taken together, the results presented here indicate that this complex may be involved in the control of signaling downstream of receptor tyrosine kinases and is important for ensuring a proper signaling response. In the absence of flot-1, increased Tyr phosphorylation of FRS2 was observed. It is known that Tyr and Thr phosphorylation of FRS2 are reciprocally regulated. Since ERK is a known executor of the FRS2 Thr phosphorylation, and ERK activity was shown to be severely diminished upon flot-1 depletion, the increased Tyr phosphorylation of FRS2 was in agreement with this and might be a direct consequence of a decreased ERK activity upon flot-1 depletion. FRS2 owes its name to the major and the first described function of this protein as a substrate for FGFR. PTB domain of FRS2 was published to constitutively bind the juxtamembrane domain of FGFR. In this study, the PTB domain was mapped to be involved in the constitutive interaction with flot-1 and the competition was shown to exist between flot-1 and FGFR1 for binding to FRS2. Another novel interaction partner of FRS2 was discovered in the present study. Cbl-associated protein (CAP) is an adaptor protein with three SH3 domains and it plays a role during insulin signaling by recruiting the signaling complex to lipid rafts. CAP was previously shown to interact with flot-1 via the SoHo domain, and this interaction was found to be crucial for the lipid raft recruitment of other signaling components. Both the PTB domain and CT of FRS2 were found to mediate the interaction with CAP, whereas in CAP, the SoHo domain, together with the third SH3 domain, seems to bind to FRS2. SH3 domains mediate the assembly of specific protein complexes by binding to proline rich sequences, several of which are present in FRS2. Due to overlapping interaction domains, FRS2 and flot-1 competed for the binding to CAP. However, the interaction with neither CAP nor flot-1 was necessary for the observed nuclear translocation of FRS2. Since CAP is expressed as several tissue- and developmental stage-specific isoforms, a further aim of this study was to analyze the expression of its isoforms in mouse embryonic fibroblasts (MEFs). Many new isoforms were discovered here which have not been described in the literature so far. They all contain the SoHo domain and three SH3 domains, but differ among themselves by the presence and length of a proline-rich region that preceeds the SoHo domain and by a novel 20-amino acid (AA) stretch between the second and the third SH3 domain. The length of the proline-rich region turned out to be an important factor determining the strength of the interaction with FRS2. The interaction was found to be weakened by the increasing length of this region. The new isoforms possessing the 20-AA stretch are specifically expressed in murine muscular tissues, with the highest level in the heart. During adipogenesis, we observed a shift in the abundance of the isoforms, in that only the isoforms without the insertion were shown to be upregulated on mRNA level. However, during myogenesis, preferentially expressed isoforms were those with the insertion. The collected data implicate that isoforms with the 20-AA insertion might be more ubiquitous in nondifferentiated/embryonic cells and that the observed "isoform-switch" might be dependent on the cell fate and differentiation state.
Multicellular organisms require that cells adhere to each other. This cell-cell adhesion is indispensable for the formation and the integrity of epithelial structures, tissues and organs. Mammals have developed four different cell-cell adhesion structures, the adhering junctions, which ensure the tight contact between cells but are also important platforms for communication and exchange in tissues. Two of these adhering junctions are cadherin based, the belt-like adherens junctions and the spot-like desmosomes. Both structures have in common that they are composed of single membrane spanning proteins, the cadherins, which accomplish adhesion in a calcium-dependent manner. The intracellular parts of classical as well as desmosomal cadherins bind to different adaptor proteins of the armadillo-protein family and others which build a protein plaque underneath the membrane and link the cadherins to the actin or intermediate filament cytoskeleton.
Desmosomes are of special importance for tissues that have to withstand mechanical stress. Although they are essential to stabilize tissues they have to be highly flexible and dynamic structures, as processes like wound healing or tissue remodeling require that adhesive interactions can be modulated. The molecular dynamics within desmosomes are not jet understood in detail, but it is assumed that two different membrane associated pools of desmosomal cadherins exist in cells. Cadherins that are incorporated in mature desmosomes are part of the junctional pool, whereas cadherins that are not associated with firm desmosomes and the intermediate filament cytoskeleton belong to the non-junctional pool. Lateral movements between the two pools results in a dynamic equilibrium and allows for example the exchange of old cadherins. Little is known about the breakdown of desmosomal cadherins. Several studies found that desmosome assembly or endocytosis are cholesterol dependent processes and claimed that membrane microdomains play a role in the regulation of desmosome dynamics. Moreover, membrane rafts may be involved in the pathomechanism of the desmosome associated disease pemphigus, were autoantibodies bind to the cadherin desmoglein-3 and trigger its internalization which results in a loss of adhesion in skin cells.
Membrane rafts are cholesterol dependent nanoscale structures of cellular membranes that are able to regulate the distribution of proteins within the plasma membrane and thus form platforms for cell signaling and membrane trafficking. Flotillins are proteins that are associated with membrane rafts and are reported to be involved in processes like endocytosis, endosomal sorting and a multitude of different signaling events. We could recently show that the membrane raft associated proteins flotillin-1 and flotillin-2 bind directly to the armadillo protein y-catenin which can be part of both, the adherens junction and the desmosome. The aim of this study was to eluciadate a possible role of flotillins in the regulation of desmosomes.
HaCaT keratinocytes were chosen as the main cell system for this study and at first the association of desmosomal components with flotillins was analyzed in detail. It was found that flotillins are clearly associated with desmosomal proteins. They colocalize with desmoglein-3 at cell borders and precipitate the other desmogleins. Further binding assays revealed that both flotillins bind to all desmogleins and the long isoforms of the second class of desmosomal cadherins, the desmocollins. The interaction is a direct one and was mapped to the ICS sequence within the cadherins. This close association rendered the question whether flotillins are functionally implicated in desmosome regulation. To address this issue, stable flotillin knockdown HaCaT cells were analyzed in detail. The molecular morphology of desmoglein-3, desmoglein-1 and two plaque proteins was clearly altered in the absence of flotillins. The membrane staining of all tested desmosomal proteins was derailed and disordered. Furthermoore, the loss of flotillins had an impact on the adhesive capacity of HaCaT keratinocytes. The cell-cell adhesion was weakened in the absence of flotillins, which was monitored by an increased fragmentation of knockdown cells in a cell dissociation assay.
In order to find out the mechanism by which flotillins influence the membrane morphology and the adhesiveness in keratinocytes, the association of desmosomal proteins with membrane microdomains was examined, at first. A predominant part of desmoglein-3 is associated with membrane rafts in HaCaT keratinocytes, whereas only a minor part of desmoglein-1 is found there. However, the raft-association of none of the examined proteins was altered in the absence of flotillins. Furthermore, flotillin depletion did not change the distribution of desmogleins with the two different cadherin pools. Less desmoglein-3 is found in the junctional pool of the flotillin depleted cells compared to the control cells, but this is due to an overall diminished desmoglein-3 protein level in these cells.
Flotillins are involved in endocytic processes but their exact role there is under debate. The endocytic uptake of desmosomal cadherins requires intact membrane rafts, but the precise mechanism is still unknown. A possible involvement of flotillins on the endocytosis of desmoglein-3 was addressed next. It is known that the internalization of desmoglein-2 is dependent on the GTPase dynamin, arguing for an involvement of dynamin in the endocytosis of desmoglein-3 as well. When dynamin and thus desmoglein-3 endocytosis was inhibited using chemical compounds, the mislocalization of desmoglein-3 that was observed in flotillin knockdown cells was restored. This suggest that inhibition of desmoglein-3 endocytosis enhances the amount and/or availability of desmoglein-3 at the plasma membrane, which then normalizes the morphological alterations caused by a knockdown of flotillins. Furthermore the morphological alterations in the flotillin knockdown HaCaT cells were found to be similar to the localization of desmoglein-3 that was observed upon treatment of keratinocytes with PV IgG These structures have been described before as linear arrays and are assumed to be sites of endocytic uptake. This strengthens the idea that enhanced desmoglein-3 internalization takes place in the absence of flotillins, which then results in a weakened adhesion.
Altogether this study revealed flotillins as novel players in desmosome mediated cell-cell adhesion processes. By binding to desmosomal cadherins and desmosomal plaque proteins, flotillins stabilize desmosomes at the plasma membrane and are required for a proper cell-cell adhesion.
Batten disease refers to neuronal ceroid lipofuscinoses (NCLs), which are inherited lysosomal storage diseases with diverse ages of onset and cause progressive neurodegeneration. The most common NCL is Juvenile NCL (JNCL), which begins in early childhood and is characterized by lysosomal accumulation of subunit c of the mitochondrial ATP synthase (subunit c). JNCL is caused by mutations in the gene CLN3. This gene encodes the CLN3 protein, a transmembrane protein of unknown structure. Localization of CLN3 is ambiguous, and its exact cellular function is not known. Thereby, it is unclear what mechanisms lead to neurodegeneration in JNCL. Models of JNCL present disturbed membrane bound organelles and cytoskeleton as well as impaired autophagy and lysosomal function. The JNCL gene defect that most patients harbor is deletion of the exons 7 and 8 of CLN3. In the Cln3Δex7/8/Δex7/8 mouse model of JNCL, this deletion has been introduced to the mouse Cln3 gene.
The actin cytoskeleton consists of filaments formed through polymerization of actin and provides a framework which defines cellular morphology and also facilitates cell motility, cytokinesis, and cell surface remodeling. Rho GTPases are signaling proteins which regulate the assembly and dynamics of the actin cytoskeleton and play an important role in neuronal morphology. Rho GTPases need to be membrane-anchored in order to become active and initiate a signaling cascade. Their membrane anchorage is achieved through their geranylgeranyl tails, which they acquire through prenylation. Protein prenylation refers to the attachment of a geranylgeranyl or farnesyl group to the C-terminus of a protein. The enzyme geranylgeranyl transferase (GGTase) catalyzes geranylgeranylation, whereas geranylgeranyl pyrophosphate (GGPP) is the donor of the geranylgeranyl group. Cells produce GGPP as well as cholesterol and other lipids through the mevalonate pathway (MVA pathway).
The aim of this study was to analyze how the JNCL gene defect affects cellular morphology, especially the actin cytoskeleton and Rho GTPases, and the MVA pathway which is connected with Rho GTPase activation. These important cellular components play crucial roles in neurons and are implicated in other neurodegenerative diseases, but have received little attention in JNCL. The immortalized CbCln3Δex7/8/Δex7/8 cerebellar precursor cell line from Cln3Δex7/8/Δex7/8 mice was used for the experiments and provides a genetically accurate, neuronal cell model of JNCL. CbCln3Δex7/8/Δex7/8 cells present subunit c accumulation only when aged at confluency, but sub-confluent cells display other phenotypes. The experiments of this study were performed both with confluency-aged and sub-confluent cells. Filamentous actin was visualized, and protein levels as well as membrane localization of several small Rho GTPases was analyzed biochemically. Also the protein levels of GGTase and the key enzymes of the mevalonate pathway were determined.
Staining pattern of filamentous actin was disturbed in confluency-aged CbCln3Δex7/8/Δex7/8 cells. Additionally it was found out that these cells did not grow to wild-type size and exhibited an elongated peroxisomal morphology. Rho GTPases had reduced total levels and showed a tendency of decreased membrane localization. Levels of GGTase and the MVA pathway enzymes were altered. Results of sub-confluent CbCln3Δex7/8/Δex7/8 cells were similar with the exception of HMG-CoA reductase, which is the rate-limiting enzyme of the MVA pathway: while its level in confluency-aged CbCln3Δex7/8/Δex7/8 cells was increased, at sub-confluency it showed a reduced level. Also, in contrast with the confluency-aged cells, Rho GTPases presented a tendency of increased membrane localization.
The results of this study reveal that the accurate JNCL gene defect alters cellular morphology and the activity of the MVA pathway in neuronal cells. Small cell size and disrupted architecture of the actin cytoskeleton are confirmed as neuronal JNCL phenotypes, and the peroxisome is introduced as a novel cellular component affected in JNCL. Through defects in endocytosis, autophagy, lysosomal and mitochondrial function, and cytoskeleton, the JNCL gene defect may prevent cells from growing to wild-type size. The JNCL gene defect may attenuate the MVA pathway via mitochondrial dysfunction and/or upregulation of degradative processes. Attenuation of the MVA pathway may contribute to impaired membrane rafts, which are an established phenotype of JNCL cells. As indicated by reduced GGTase level and supported by downregulation of lipid production through the MVA pathway, the JNCL gene defect might also decrease prenylation of proteins.
Reggie-1 und Reggie-2 stellen eine über diverse Spezies konservierte Proteinfamilie dar und werden in den meisten Geweben exprimiert. Die physiologische Funktion dieser Proteine ist bisher nicht geklärt. Die Reggie-Proteine wurden zunächst im Zusammenhang mit der Regeneration von Axonen retinaler Ganglienzellen des Goldfisches beschrieben. In diesen Zellen wird die Expression beider Proteine nach Läsion des Nervs hochreguliert. Unabhängig davon wurden Reggies aufgrund ihres Sedimentationsverhaltens in Dichtegradienten als Flotilline beschrieben. Dabei ist Reggie-1 identisch mit Flotillin-2 und Reggie-2 mit Flotillin-1. Beide Reggie-Proteine sind Raft-assoziiert. Bei Rafts handelt es sich um Membran-Mikrodomänen, deren Rolle in verschiedenen zellulären Prozessen wie Signaltransduktion, Endozytose und dem Transport von Proteinen beschrieben wurde. Strukturell zeichnen sich Rafts durch einen im Vergleich zur restlichen Membran erhöhten Anteil an Cholesterol und Sphingolipiden aus, der unter anderem in einer herabgesetzten lateralen Beweglichkeit der beteiligten Moleküle resultiert, und so Sortierungs- und Signalvorgänge innerhalb der Zelle begünstigt. Aus der Beobachtung, dass Reggie-2 mit dem Struktur-Vorläuferprotein des Rotavirus interagiert, ergab sich die Fragestellung zu dieser Arbeit. Das Ausschleusen von Rotaviren aus infizierten Zellen ist von Rafts abhängig. Die strukturellen Eigenschaften des Rotavirus-Proteins ähneln Gag, dem Struktur-Vorläuferprotein des Humanen Immundefizienz-Virus und verwandter Retroviren. Das HIV Gag-Protein ist außerhalb des viralen Kontext nach Expression in der Lage, an zellulären Membranen zu assemblieren und Virus-ähnliche Partikel zu bilden. Die molekularen Abläufe dieses Vorgangs, der als budding bezeichnet wird, sind noch nicht im Detail verstanden. Sie werden jedoch sowohl für Gag allein, als auch für die Virusproduktion in infizierten Zellen mit Rafts in Verbindung gebracht. Ein Hinweis darauf ergibt sich aus der Tatsache, dass HIV-infizierte Zellen nach Cholesterol-Depletion kein produktives Virus mehr abschnüren. Im Rahmen dieser Arbeit konnte gezeigt werden, dass Reggie-2 mit HIV-1 Gag interagiert und sich Veränderungen der Reggie-2 Expressions sowohl auf die zelluläre Lokalisation von Gag als auch auf die Produktion Virus-ähnlicher Partikel auswirkt. Die Überexpression von Reggie-2 in HeLa-Zellen führt zu einer Umverteilung von Gag mit erhöhter Lokalisation an der Plasmamembran, an der es dann stark mit Reggie-2 kolokalisiert. Die Verminderung der Reggie-2 Expression mittels siRNA resultiert in einer erhöhten Freisetzung Virus-ähnlicher Partikel aus Gag-transfizierten HeLa-Zellen und in einer eher löslichen Lokalisation des viralen Proteins im Zytoplasma. Der Ort des buddings von HIV ist Zelltyp-abhängig und involviert die zelluläre ESCRT-Maschinerie, die physiologisch für das Sortieren von Proteinen in multivesicular bodies (MVBs) zuständig ist. Gag rekrutiert das ESCRT-System, indem es über ein kurzes Aminosäuremotif an das ESCRT-I Protein TSG101 bindet und dabei die Eigenschaft des zellulären Proteins HRS nachahmt. Im Rahmen dieser Arbeit konnte gezeigt werden, dass Reggie-2 in HeLa-Zellen mit Komponenten der ESCRT-Maschinerie kolokalisiert. Dabei ist denkbar, dass eine Verbindung zwischen ESCRT und Reggie-Proteinen besteht, da Rafts als Signal- und Sortier-Plattformen die physiologischen Prozesse begünstigen könnten, die bei der Funktion von MVBs eine Rolle spielen. Im Gegensatz zu Reggie-2 findet keine Interaktion von Gag mit Reggie-1 statt. Beide Reggie-Proteine sind jedoch biochemisch in produktiven Viren infizierter primärer T-Zellen nachweisbar. Des Weiteren zeigen Immunfluoreszenz-Studien infizierter Lymphozyten, dass beide Reggies in diesen Zellen in einem großen Ausmaß mit Gag kolokalisieren. Der Grund dafür ist wahrscheinlich die Fähigkeit der Reggies zur Homo- und Heterooligomerisierung. Die direkte und spezifische Interaktion von Gag mit dem Raft-assozierten Protein Reggie-2 konnte mit verschiedenen Methoden gezeigt werden. Dabei können die Ergebnisse als Grundlage für ein besseres Verständnis der HIV-Pathogenese dienen, und zusätzlich gibt die mögliche Verbindung von Reggies zum ESCRT-Komplex neue Hinweise auf die zelluläre Funktion dieser Raft-Proteine.