Refine
Has Fulltext
- yes (47)
Is part of the Bibliography
- yes (47)
Year of publication
Document Type
- Journal article (25)
- Doctoral Thesis (22)
Language
- English (47) (remove)
Keywords
- Drosophila (5)
- Maus (4)
- Raf <Biochemie> (4)
- signal transduction (4)
- Embryonalentwicklung (3)
- Signaltransduktion (3)
- Stammzelle (3)
- phosphorylation (3)
- Apoptosis (2)
- Biochemie (2)
Institute
- Institut für Medizinische Strahlenkunde und Zellforschung (47) (remove)
Sonstige beteiligte Institutionen
Cellular proliferation, differentiation and survival in response to extracellular signals are controlled by the signal transduction pathway of Ras, Raf and MAP kinase. The Raf proteins are serine/threonine kinases with essential function in growth/differentiation/survival - related signal transduction events. In mammals, three functional (A-, B-, and C-Raf) genes were described. Biochemical studies suggest overlapping and differential utilization of Raf isozymes. However, the frequent co-expression of Raf isozymes and their multiple activators and effectors impedes the full understanding of their specific roles. The elucidation of these roles is important due to the involvement of the Ras/Raf/MEK/MAP kinase cascade in human disorders especially in tumor development and progression. B-Raf was shown to posses the strongest kinase activity among Raf kinases and display antiapoptotic properties. Mice deficient in B-Raf show overall growth retardation and die between E10.5 and E12.5 of vascular defects caused by excessive death of differentiated endothelial cells. To elucidate the redundancy of Raf isozymes during embryonic development and to rescue B-Raf-/- (KO) phenotype, B-Raf alleles were disrupted by introducing A-Raf cDNA under the control of endogenous B-Raf promoter. The resulting BRaf A-Raf/A-Raf (KIN) phenotype depends on genetic background. The living embryos displaying normal development but size reduction were found with low incidence at E12.5d-16.5d. All of them displayed the rescue of vascular system. One adult p20 mouse without any visible defects in development and behavior was obtained. On the other hand, the processes of neurogenesis and neural precursors migration in survived embryos were disturbed which led in some cases to underdevelopment of different brain compartments. TUNEL and cell proliferation (PCNA staining) assays revealed more apoptotic (E13.5d) and less proliferating(E12.5d cells within ventricular and sub-ventricular zones of brain ventricles and in striatum of KIN embryos. In addition, more apoptotic cells were detected in many other tissues of E13.5d and in lung of E16.5d KIN embryos but not in adult KIN mouse. p20 KIN mouse demonstrated reduced fraction of neural precursor cells in sub-granular zone of hippocampus and mature neurons in olfactory bulb. The other processes of neurogenesis were not disturbed in adult KIN animal. Fibroblasts obtained from KIN embryos demonstrated less proliferative ability and were more susceptible to apoptotic stimuli compared to WT. This was accompanied by the reduction of active ERK and Akt required for survival, and with decrease of inactive phosphorylated BAD. The kinetic of both ERK and Akt phosphorylation upon serum stimulation was delayed. All these data indicate that moderate A-Raf kinase activity can prevent the endothelial apoptosis but is not enough to completely rescue the other developmental consequences.
The experimental work of this thesis addresses the questions of whether established cell lines injected into murine blastocysts find their way back home and seed preferentially at the site of their origin. Furthermore, can they change their fate and differentiate to unrelated cell types when exposed to the embryonic environment. This survey was based on the fact that different cell lines have different potentials in developing embryos, dependent on their cellular identity. The cell lines used in this survey were AGM region-deriving DAS 104-4, DAS 104-8 cells, yolk sac-deriving YSE cells and bone marrow-deriving FDCP mix cells. These cells were injected into mouse blastocysts. Donor cells were traced in developing embryos via specific markers. Analysis of the embryos revealed that DAS cells are promiscuous in their seeding pattern, since they were found in all analysed tissues with similar frequencies. YSE cells showed preferences in seeding yolk sac and liver. YSE donor cells in chimaeric tissues were not able to change their immuno-phenotype, indicating that they did not change their destiny. Analysis of adult mice did not reveal any of YSE-derived cells donor contribution. In contrast, FDCP mix cells mostly engrafted haematopoietic tissues, although the embryos analysed by in situ hybridization had donor signals frequently in cartilage primordia, heads, and livers. Analysis of whether FDCPmix-derived cells found in foetal livers were of haematopoietic or hepatocytes nature showed that progeny of injected FDCP mix cells do not differentiate into cells that express a hepatocyte-specific marker. Further analysis showed that FDCPmix-derived donor cells found in brain express neural or haematopoietic markers. In order to reveal if they transdifferentiate to neurons or fuse with neurons/glial cells, nuclear diameters of donor and recipient cells were determined. Comparison of the nuclear diameters of recipient and donor cells revealed no differences. Therefore this suggests that progeny of FDCP mix in brain are not fusion products. Analysis of adult mice tissues revealed that presence of FDCP mix-derived cells was the highest in brains. These results confirmed the assumption that the developmental potential of the analysed cells cannot be easily modified, even when exposed to early embryonic environment. Therefore one can conclude that the analysed cell types had different homing patterns depending on their origins.
The RAF family of protein kinases consists of three members, A-RAF, B-RAF and C-RAF. Unlike the other isotypes, B-RAF has been found to have an important function for normal development of the central nervous system (CNS), because newly generated embryonic neurons lacking B-RAF cannot respond to survival factors and undergo cell death in vitro. A second cell lineage affected by the absence of B-RAF are endothelial cells and their death leads to internal bleedings and lethality of B-RAF-/- mice between embryonic day 10.5 (E10.5) and E12.5 precluding an opportunity to further analyze neural B-RAF function at a later stage. In contrast to B-RAF-/- mice, B-RAFKIN/KIN mice, which are B-RAF deficient but express a chimeric protein consisting of the unique N terminus of B-RAF and all the domains of A-RAF in the B-RAF gene locus, survive after midgestation because their endothelial cells are protected from apoptosis. More importantly, overall prevention of abnormal neural apoptosis in the forebrain allows us to study proliferation- or differentiation-oriented function of B-RAF other than its survival effects in CNS development. The detailed investigation of B-RAFKIN/KIN animals was concentrated on cortical development. There were apparent cortical defects in B-RAFKIN/KIN forebrain: Loss of B-RAF led to severe reduction of Brn-2 expressing pyramidal projection neurons accompanied by a disruption of dendrite formation in the upper layers. In further analysis, BrdU labelling experiments showed that from E14.5 to E16.5 cell proliferation in the ventricular zone of the mutant mice was reduced and that the late-born cortical neurons failed to migrate properly. While the proliferation defect of cortical progenitors was associated with reduced ERK activation, the mechanism causing impaired neuronal migration remains to be determined. Our hypothesis is that the subcellular localization of phospho-ERK may be altered in migrating cortical neurons in B-RAFKIN/KIN mice. To confirm in vivo function of B-RAF and further study unknown roles in embryonic neurogenesis as well as other morphogenesis, conditional B-RAF knockouts would be the ideal models, which can efficiently avoid embryonic lethality, prevent unwanted pleiotropic side effects and exclude accumulative compensatory developmental changes from the earliest developmental stage on, through the deletion of genetic material/gene function in selected cells at a specific time. The use of site-specific recombinases such as Cre and the successful development of the reversible tetracycline-based switch have provided powerful venues for creating conditional loss-of-function mouse models. Generation of tetracycline-regulated B-RAF and floxed B-RAF mouse embryonic stem (ES) cell lines was performed. Up to now, high-grade chimeric mice were obtained after blastocyst injection of the modified ES cell clones. The germline transmission from these chimeric mice is currently under investigation. When either of conditional mouse lines is ready, detailed examination in their CNS development would be done to reveal how B-RAF plays a real role for normal development of the nervous system.
In spite of the progress made in deciphering regulatory networks of cancer cells on the molecular level, the interaction of tumour cells with their stroma has not been adequately analyzed. Earlier, we have addressed the hypothesis that the murine embryonic microenvironment can induce the differentiation of human tumour cells. To examine such interactions, human leukaemic AML cells were injected into pre-implantation murine blastocysts at embryonic day 3.5 of gestation. Analysis of developing mice revealed the presence of human AML cells in chimaeric embryos and adults and the appearance of haematopoietic differentiation markers on progeny of injected human AML cells. This finding strengthens the notion that the embryonic microenvironment is capable of regulating the proliferation and differentiation of leukaemic AML cells. Based on these results, I embarked to analyse the consequences of stromal environment-induced changes in human AML cells upon in vitro coculture with selected haematopoietic stromal cell lines in terms of changes in differentiation and proliferation properties of AML cells. For this purpose, established human AML cell lines were cocultured on a variety of mitotically inactivated stromal cell lines derived from different murine embryonic/foetal haematopoietic sites such as yolk sac, aorta-gonad-mesonephros (AGM) region and foetal liver. To score for coculture-induced changes, I compared the morphology, histo-chemical properties, immunophenotype, proliferation rate, and gene expression profile in cocultured and non-cocultured AML cells. Results show that, upon coculture of Kasumi-1 cells- a cell line established from a FAB class M2 patient - with AGM-derived DAS 104-4, but not with other stromal cell lines, Kasumi-1 AML cells exibit decreased proliferation and colony formation capabilities and acquire differentiated morphologies. Along this line, coculturing of Kasumi-1 cells resulted in the up-regulation of the myelo-monocytic lineage cell surface markers CD11b and CD14. Coculture also resulted in increase in lysosomal marker CD68, a hallmark of myeloid differentiation. Interestingly, apart from cell lines, coculture on DAS 104-4 stroma was also efficient in inducing myeloid differentiation of patient derived primary M2-AML cells. Moreover, cocultivation of KG-1 cell line on DAS 104-4 showed activation of -globin transcription and up-regulation of Glycophorin A on its surface, which indicate DAS 104-4 coculture-induced erythroid differentiation of KG-1 cells. Analysis of the proliferation rate of Kasumi-1 cells using the CFSE retention assay revealed that upon cocultivation on DAS 104-4, but not on NIH 3T3 cells, there is a decrease both in the proliferation rate and in the frequency of colony forming cells in clonogenic methyl cellulose cultures. Cell cycle analysis revealed the coculture-induced accumulation of G1-G0 stage cells. Gene-expression analysis by quantitative RT-PCR revealed a substantial decrease in the amount of AML1 and AML1-ETO fusion transcripts in parallel with an increase in p16, p21, C/EBP and PU.1 transcription levels. Interestingly, AML1-ETO transcription down-regulation of AML cells needs direct contact with DAS 104-4 cells. Knocking down AML1-ETO expression by siRNA strategy led to reduction in proliferation and depletion of colony forming cells in Kasumi1 cell population. siRNA-mediated AML1-ETO knock-down Kasumi-1 cells showed increased susceptibility to stroma-induced myeloid differentiation. However, on its own, AML1-ETO down-regulation was not sufficient to induce myeloid differentiation. This indicates that AML1-ETO down-regulation may have an active role on the coculture-induced effect but in addition to AML1-ETO down-regulation, further stimuli are required for the coculture-induced myeloid differentiation in the AML cells. In summary, in the present study I established and characterised a coculture-based in vitro system, which is capable of reducing the proliferation while inducing differentiation of human AML cells. The concept emerging from the studies indicates that the stroma environment can affect leukaemic cell proliferation and differentiation in contact-dependent and CD44 activation-independent manner. Furthermore, this study emphasizes the role of AML1-ETO in AML and indicates that AML1-ETO down-regulation is involved in the stroma-induced differentiation of Kasumi-1 cells. The result described here encourages further investigation into the mechanistic details of molecular and cellular interactions between the leukaemic cells and their stroma, which in turn may lead to the identification of new paradigms for a knowledge-based control and reprogramming of leukaemic cells.
Uniparental zygotes with two genomes from the same sex can be established from fertilised oocytes after pronuclear exchange. They contain two maternal (gynogenetic; GG) or paternal (androgenetic; AG) pronuclei and are not competent to develop into viable offspring but they can form blastocysts from which embryonic stem cells (ES cells) can be derived. The developmental potential of uniparental ES cells is not fully investigated. The restricted developmental potential of uniparental cells is cell-intrinsic and probably reflects the different roles maternal and paternal genomes play during development. Following blastocyst injection, both GG and AG ES cells show biased and parent-of-origin-specific chimaera formation. While the in vitro and in vivo neural differentiation potential of GG ES cells is well characterised the neural developmental potential of AG ES cells is less clear. In an earlier study the group of K. John McLaughlin reported that AG and GG ES cell-derived hematopoietic stem cells conveyed long-term, multi-lineage hematopoietic engraftment with no associated pathologies (Eckardt et al., 2007). The aim of this study was to investigate the potential of AG uniparental murine ES cells to differentiate in vitro and in vivo into neural progenitor / stem cells and further into neurons, astro- and oligodendroglia in comparison to GG and biparental (normal fertilised; N) ES cells. Uniparental and biparental ES cells were obtained from K. John McLaughlin’s group and a cell culture system was established to expand uniparental (AG, GG) and biparental N ES cells on murine embryonic fibroblasts (MEF). A multistep-protocol was used to differentiate ES cells towards pan-neural progenitor cells and neuronal and glial cell types (Brüstle et al., 1997). The ability of terminal neural differentiation in vitro was analysed by fluorescence microscopy using neuronal and glial lineage markers. In parallel, eGFP+ AG or N ES cells were injected into blastocysts prior to their transfer into foster mothers. At E12.5 and E14.5, embryos were isolated, forebrains were dissected and by means of fluorescence activated cell sorting (FACS) eGFP+ donor cells were isolated from chimeric brains. Both eGFP+ donor and corresponding eGFP- blastocyst-derived brain cells were expanded and analyses of differentiation potential and self-renewal capacity were performed. Also, cryosections of E12.5 chimeric brains were analysed for donor contribution to the neuronal lineage by immunofluorescence microscopy. Here it is described that following in vitro differentiation, AG pan-neural progenitor cells have similar abilities to differentiate into neuronal and glial lineages as GG and N pan-neural progenitor cells. In cryosections of E12.5 chimeric brains no differences in brain engraftment and formation of immature neuronal cells between uniparental AG and N donor cells were detected. AG and N ES cell-derived cells isolated from chimeric foetal brains by FACS exhibited similar neurosphere initiating cell frequencies and neural multi-lineage differentiation potential. Therefore, the data of this study suggest that the previously described differences in the in vivo engraftment pattern of uniparental inner cell mass (ICM) cells in foetal brains (Keverne et al., 1996) are not primarily due to limitations in the proliferation or differentiation properties of uniparental neural progenitor cells. The results presented here indicate that AG ES cell-derived neural progenitor / stem cells did not differ from N neural progenitor / stem cells in their self-renewal and their neural multi-lineage differentiation potential. Also AG ES cell-derived cells contributed to developing brains at early foetal developmental stages showing a widespread and balanced distribution in chimeric brains. AG brain cells form neurospheres with self-renewal and neural differentiation capacity similar to N ES cell-derived brain cells. Thus, the data of this study together indicate that the neural developmental potential in vivo and in vitro of AG and N ES cells does not differ.
In neoplastic diseases the tumor stroma and especially tumor-associated macrophages (TAMs) play an important role in tumor growth and progression. TAMs exhibit an intensive cross-talk with tumor cells resulting in the promotion of angiogenesis and the inhibition of local protective immune responses in certain tumor entities. Therefore, TAMs are a potential target for tumor therapy. Here it was shown that intravenously applied intracellular bacteria like Salmonella and Shigella primarily target TAMs. To exploit this feature a growth attenuated Shigella strain with the capacity to induce apoptosis in macrophages was designed. Shigella are invasive bacteria that penetrate the colonic tissue and initiate an acute inflammation. In macrophages, Shigella rapidly induces caspase-1 processing and apoptosis via the virulence factor IpaB. By genomic deletion of the aroA-locus a metabolically attenuated strain defective in intracellular growth but with retained capacity of infection, cell-to-cell spread, caspase-1 processing and apoptosis induction in macrophages was designed. It was shown that this strain primarily targets TAMs in 4T1 cell induced and transgenic MMTV-HER2/new breast cancer models. Shigella were almost exclusively found intracellularly, whereas growth attenuated Salmonella were also found extracellularly at late time points. The metabollically attenuated Shigella strain with retained virulence, but not avirulent Shigella strains, was able to activate caspase-1 and induce apoptosis in TAMs at all time points (4 h, 6 h and 7 d p.i.) in both breast cancer models. This unrestricted apoptosis induction translated into a substantial, long-lasting and highly significant reduction of TAMs number (up to 70 %) in both models. In contrast, Salmonella could only induce apoptosis in TAMs at early time points (6 h p.i.) and failed to reduce TAMs in both models. In the 4T1 model, the effect on tumor size was monitored and treatment of the mice with the attenuated Shigella strain resulted in a complete block of tumor growth. Finally, Shigella primarily infected the macrophage fraction, activated caspase-1 and induced apoptosis in cells derived from a human ovarian carcinoma ex vivo. Taken together, this data suggests that growth attenuated intracellular bacteria capable of inducing apoptosis in TAMs are a promising therapeutic option for certain cancer diseases where TAMs have a proven role for tumor growth or progression.
Steps involved in the progression of non-small cell lung cancer (NSCLC) to metastasis are poorly understood. Expression of oncogenic C-RAF in lung epithelial cells has yielded a model for non-small cell lung cancer (NSCLC). The induced adenomas are characterised by high genomic stability, a lack of tumor progression and pronounced cell-cell contacts raising the question whether disruption of E-cadherin complexes would promote progression to metastasis. Two genetic approaches were used to evaluate the role of adherens junctions in a C-RAF driven mouse model for NSCLC: conditional ablation of the Cdh1 gene and expression of dominant negative (dn) E-cadherin. Disruption of E-cadherin function caused massive formation of intratumoral vessels that was reversible in the early phase of induction. Vascularized tumors grew more rapidly, developed invasive fronts and gave rise to micrometastasis. ß-catenin was identified as a critical effector of E-cadherin disruption leading to up-regulation of angiogenic inducers (VEGF-A and VEGF-C) in mouse and human lung tumor cell lines. In vivo, lung tumor cells with disrupted E-cadherin expressed ß-catenin target genes of endodermal and other lineages suggesting that reprogramming may be involved in metastatic progression.
The proteins of the RAF family (A-RAF, B-RAF, and C-RAF) are serine/threonine-kinases that play important roles in development, mature cell regulation and cancer. Although it is widely held that their localization on membranes is an important aspect of their function, there are few data addressing this aspect of their mode of action. Here, we report that each member of the RAF family exhibits a specific distribution at the level of cellular membranes, and that C-RAF is the only isoform that directly targets mitochondria. We find that the RAF kinases exhibit intrinsic differences in terms of mitochondrial affinity, and that C-RAF is the only isoform that binds this organelle efficiently. This affinity is conferred by the C-RAF amino-terminal domain, and does not depend on the presence of RAS GTPases on the surface of mitochondria. Furthermore, we analyze the consequences of C-RAF activation on the cellular and molecular level. C-RAF activation on mitochondria dramatically changes their morphology and their subcellular distribution. On the molecular level, we examine the role of C-RAF in the regulation of the pro-apoptotic Bcl-2 family member BAD. This protein exhibits the original mode of regulation by phosphorylation. Although several reports addressed the regulation of BAD by C-RAF, the exact mode of action as well as the consequences of C-RAF activation on BAD are still not completely understood. We show that the inducible activation of C-RAF promotes the rapid phosphorylation of BAD on Serine-112 (Ser-75 in the human protein), through a cascade involving the kinases MEK and RSK. Our findings reveal a new aspect of the regulation of BAD protein and its control by the RAF pathway: we find that C-RAF activation promotes BAD poly-ubiquitylation in a phosphorylation-dependent fashion, and increases the turn-over of this protein through proteasomal degradation.
In neurons the Ras signaling pathway is activated by a large number of various stimuli, including trophic factors, neurotransmitters and modulatory peptides. Guanine nucleotide exchange factors (GEFs) mediate the activation of Ras GTPases, by catalyzing the exchange of GDP for GTP, and facilitate signaling networks crosstalk. In this work, very-KIND (VKIND), a new brain specific RasGEF was structurally and functionally characterized. VKIND belongs to the KIND protein family along with the non-receptor tyrosine phosphatases type 13 and Spir actin nucleation factors. The kinase non-catalytic C-lobe domain (KIND) is similar to the C-terminal protein kinase catalytic fold (C-lobe) of the p21-activated kinase (PAK). The open reading frame (ORF) of the VKIND gene of 5229 base pairs was cloned. The VKIND ORF translates into a protein of 1742 amino acids residues with a size of 191 kD. The VKIND protein structure is highly conserved among species and at present the protein is found only in Vertebratae and Echinodermatae. The arrangement of two KIND domains at its amino-terminal region, KIND1 and KIND2, is depicted in its name. The KIND module functions as a molecular interaction structure that is deprived of any enzymatic activity. While the precise occupation of the KIND1 domain remains elusive, the KIND2 domain binds to the microtubules-associated protein 2 (MAP2). The protein central portion features two clusters of high conservation of yet unknown function as well as a coiled-coil motif with a putative multiple protein-protein interaction activity. At the carboxy-terminal region VKIND features a guanine nucleotide exchange factor for Ras-like small GTPases (RasGEF) with a structural RasGEFN motif attached at its N-terminal site. The VKIND RasGEF motif is structurally related to the yeast catalytic domain CDC25. The closest relation of the VKIND RasGEF domain with an average sequence identity of 23% is assigned to the RasGEF domains of exchange factors specific for Rap GTPases with two unique insertions: the first one of 24 amino acids in the N-terminal end of the domain (between helixes αA and αB of the SOS1 RasGEF module) and the second one of 11 amino acids in the C-terminal part (between, helixes αJ and αK of the Sos1 RasGEF module). The RasGEFN domain plays a critical role in sustaining the structural and catalytical integrity of the guanidine exchange factor. VKIND is specifically and highly expressed in the murine nervous system during embryonic development and adulthood. During embryogenesis VKIND expression is present in the murine neural tube, telencephalon, retinal ganglion cells, and rhombencephalon. In the adult murine brain VKIND expression is most prominent in the cerebellum, however exclusively restricted to the granular and Purkinje cell layers. Subcellular distribution studies and time-lapse analysis revealed the gradual accumulation of VKIND into highly motile circular particles which featured estimated maximum velocity of 12 μm/min. By merging the nascent structures progressively grew to estimated 2 μm in size suggesting a role for VKIND in the vesicular transport process. Furthermore, the KIND1/KIND2 region of the VKIND protein was found to be phosphorylated by the p38 mitogen-activated protein kinase (MAPK), recently discovered to induce neurite outgrowth in response to hyperosmotic shock. In the light of VKIND negatively controlling neurite outgrowth, further elucidation of the complex Ras pathways may provide rewarding insights in the neuronal physiology.
Extracellular signals are translated and amplified via cascades of serially switched protein kinases, MAP kinases (MAPKs). One of the MAP pathways, the classical RAS/RAF/MEK/ERK pathway, transduces signals from receptor tyrosine kinases and plays a central role in regulation of cell proliferation. RAF kinases (A-, B- and C-RAF) function atop of this cascade and convert signals emanating from conformational change of RAS GTPases into their kinase activity, which in turn phosphorylates their immediate substrate, MEK. Disregulated kinase activity of RAF can result in tumor formation, as documented for many types of cancer, predominantly melanomas and thyroid carcinomas (B-RAF). A-RAF is the least characterized RAF, possibly due to its low intrinsic kinase activity and comparatively mild phenotype of A-RAF knockout mice. Nevertheless, the unique phenotype of araf -/- mice, showed predominantly neurological abnormalities such as cerebellum disorders, suggesting that A-RAF participates in a specific process not complemented by activities of B- and CRAF. Here we describe the role of A-RAF in membrane trafficking and identify its function in a specific step of endocytosis. This work led to the discovery of a C-terminally truncated version of A-RAF, AR149 that strongly interfered with cell growth and polarization in yeast and with endocytosis and actin polymerization in mammalian cells. As this work was in progress two splicing isoforms of ARAF, termed DA-RAF1 and DA-RAF2 were described that act as natural inhibitors of RAS-ERK signaling during myogenic differentiation (Yokoyama et al., 2007). DA-RAF2 contains the first 153 aa of A-RAF and thus is nearly identical with AR149. AR149 localized specifically to the recycling endosomal compartments as confirmed by colocalization and coimmunoprecipitation with ARF6. Expression of AR149 interferes with recycling of endocytosed transferrin (Tfn) and with actin polymerization. The endocytic compartment, where internalized Tfn is trapped, was identified as ARF6- and RAB11- positive endocytic vesicles. We conclude that the inhibition of Tfn trafficking in the absence of A-RAF or under overexpression of AR149 occurs between tubular- and TGNassociated recycling endosomal compartments. siRNA-mediated depletion of endogenous A-RAF or inhibition of MEK by U0126 mimic the AR149 overexpression phenotype, supporting a role of ARAF regulated ERK signalling at endosomes that is controlled by AR149 and targets ARF6. Our data additionally suggest EFA6 as a partner of A-RAF during activation of ARF6. The novel findings on the A-RAF localization and the interaction with ARF6 have led to a new model of ARAF function were A-RAF via activation of ARF6 controls the recycling of endocytic vesicles.Endocytosis and rapid recycling of synaptic vesicles is critically important for the physiological function of neurons. The finding, that A-RAF regulates endocytic recycling open a new perspective for investigation of the role of A-RAF in the nervous system.