Michael M. Kessels lab - Research Group Cell Biology of the Cytoskeleton
(Research Institute of the FSU Jena / Institute for Biochemistry I)
The actin cytoskeleton is composed of a plethora of single molecules, which are organized in filaments and superstructures thereof. The actin cytoskeleton is continuously being remodeled and this dynamic remodeling is essential for eukaryotic cells. It allows cells to respond to intracellular and extracellular stimuli and is indispensable for cell division, establishment and changes in cellular morphology and cell adhesion-processes. Such processes for example play an important role during embryonic development, wound healing as well as the formation of neuronal networks.
The required morphological and functional organization of cellular membrane structures and compartments as well as of membrane transport processes depends on a coordinated work of a variety of molecules of especially the membrane-associated, cortical actin cytoskeleton. Such a coordinated work seems to be of special importance in cells that depend on efficient and tightly regulated membrane transport processes and defined cell morphologies, such as neuronal cells. In neurons, this interplay may be of high efficiency to ensure rapid compensatory endocytosis after synaptic transmitter release and to thereby ensure an efficient and rapid synaptic signal transmission. Furthermore, postsynaptic modulation of receptor content within the membrane by insertion and endocytic uptake is thought to be an important factor in synaptic plasticity. The specialized cytomatrix structures on both sides of the synaptic cleft orchestrate these synaptic functions.
Our main research focus therefore is to identify and characterize molecular players at the functional interface between endocytosis and the cortical actin cytoskeleton. We have discovered novel proteins that are able to mediate such interconnections, e.g. the actin-binding protein Abp1. Analysis of such proteins represents a route to a better understanding of the working mechanisms connecting cellular membranes and their functions with the actin cytoskeleton and special cytomatrix structures. In neurons, these cell biological insights significantly advance our understanding of how the speed and efficiency of synaptic transmission is accomplished and how the complex and dynamic environment of the synapse is established, maintained and remodeled during synaptic plasticity processes and during the formation and reorganization of neuronal networks.
As cytoskeletal forces are a major source for defining, keeping and changing the morphology and subcellular organization of cells and these processes are crucial for the formation of the complex tissues and organs of multicellular organisms, it is furthermore crucial to understand actin dynamics in detail. We therefore also study the molecular mechanisms and the cellular functions of components of the actin cytoskeleton. We hereby focus on the machines that catalyze the critical step of actin filament formation, i.e. the assembly of an actin nucleus. These components are called actin nucleators or actin nucleation promoting factors (NPFs). Prominent are the arp2/3 complex, formins and novel WH2-domain-containing nucleators, such as Cobl (Qualmann & Kessels 2009, Trends Cell Biol.). Studying the organization and the dynamics of the actin cytoskeleton yields deep insights into how cells are shaped and plastically reorganized.
Research Projects
Mechanisms of Arp2/3 complex-mediated actin filament formation and of its spatial and temporal control
The Arp2/3 complex is the most well characterized machine for the formation of new actin filaments (nucleation) and thus of crucial importance for the life of eukaryotic cells. Our studies show that certain SH3 domain-containing proteins control the activity of the Arp2/3 complex by determining the activity of arp2/3 complex activators, such as the Neural Wiskott-Aldrich-Syndrom Protein (N-WASP). The F-actin binding protein Abp1 (Kessels et al. 2000, Mol. Biol. Cell) interacts with the Arp2/3 complex activator N-WASP, releases its autoinhibition and thereby steers its functions (Pinyol et al. 2007 PLoS ONE). The project examines the functions of the Abp1/N-WASP interaction and the and molecular mechanisms Abp1 uses to control neuronal development and morphology.
Michael M. Kessels Lab - AG Zellbiologie des Cytoskeletts
Arp2/3 complex deficiency in developing neurons (induced via RNA interference and highlighted via GFP expression and fluorescence mircroscopy) leads to defects in proper development of axons.
A powerful nucleator of actin filaments is the Arp2/3 complex. Purified Arp2/3 complex alone, however, only poorly nucleates actin filaments and thus requires activating factors for efficient activity. Since the Arp2/3 complex is involved in diverse actin cytoskeletal functions, specificity for individual processes and morphological features can only be brought about by distinct regulatory pathways and by further accessory components of this core actin nucleation machinery.
Our studies have revealed that knock-down of the F-actin-binding protein Abp1, which is important for endocytosis and synaptic organization, results in changes in axon development similar to one of the two most prominent phenotypes of Arp2/3 complex inhibition - a selective increase of axon length (Pinyol et al. 2007, J. Neurosci.). In vitro and in vivo experiments demonstrate that Abp1 interacts directly with N-WASP, an activator of the Arp2/3 complex, releases the autoinhibition of N-WASP in cooperation with the small GTPase Cdc42 and thereby promotes N-WASP-triggered Arp2/3 complex-mediated actin polymerization. In line with these mechanistic studies and the colocalization of Abp1, N-WASP and Arp2/3 at sites of actin polymerization in neurons, we could show that Abp1 plays an essential and cooperative role with Cdc42 in N-WASP-induced rearrangements of the neuronal cytoskeleton. We furthermore showed that introduction of N-WASP mutants lacking the ability to bind Abp1 or Cdc42, Arp2/3 complex inhibition, Abp1 knock down, N-WASP knock down and Arp3 knock down all cause identical defects in axon length control (Pinyol et al. 2007, PLoS ONE; Dharmalingam et al. 2009, J. Neurosci.). Our data thus strongly suggest that these proteins and their complex formation are important for cytoskeletal processes underlying neuronal network formation.
It is obvious that powerful Arp2/3 complex activation by Abp1 and further components binding to Arp2/3 complex activators needs to be carefully coordinated with the different cellular functions involving Arp2/3 complex-mediated actin nucleation in both time and space. Further work of our lab showed that distinct molecules acting on the arp2/3 complex activator N-WASP ensure specificity in cellular function. Whereas Abp1/N-WASP/Arp2/3 complex-mediated actin nucleation is crucial for proper axon length control, interactions of N-WASP with another SH3 domain-containing protein, syndapin I, are crucial for N-WASP/arp2/3 complex-mediated suppression of axon branching (Dharmalingam et al. 2009, J. Neurosci.).
How the powerful Arp2/3 complex actin nucleation machine is controlled in time and space during different developmental and physiological processes is an area of intense current and future work in the lab.
The novel actin nucleator Cordon-Bleu (Cobl) and its role in neuronal morphogenesis and network formation
In the course of this project we have been able to identify a novel actin nucleator that also seems to interact with both Abp1 and syndapins (Ahuja et al. 2007, Cell). Cobl was a thus far uncharacterized protein. We therefore started broad and extensive studies to understand its actin nucleation mechansims, its functions and its importance in different cellular functions. We for example study the mechansims Cobl uses by mutational and structural examinations, analyze the protein interactions of cobl with a variety of methods in vitro as well as in living cells and furthermore investigate its cell biological functions in neuromorphogenesis and in the proper formation of neuronal structures.
Cobl is a 180 kD protein with little sequence homology to other known proteins. It is marked by three C-terminal Wiskott-Aldrich-Syndrome-Protein-Homology 2 (WH2) domains that interaction with monomeric actin and is predominantly expressed in the brain (Ahuja et al. 2007, Cell). An excess of Cobl gives rise to elaborate F-actin-rich ruffles in non-neuronal cells and to excessive arborization in neuronal cells, a phenotype, which we demonstrated to depend on its capability to trigger actin nucleation. Cobl-mediated actin nucleation - in contrast to that mediated by the arp2/3 complex - gives rise to unbranched actin filaments and is as powerful as actin nucleation induced by activated arp2/3 complex. Loss-of-function-analyses via RNAi showed that Cobl is a crucial factor in the development of the dendritic arbor and thus seems to have a function distinct from that of the arp2/3 complex during neuromorphogenesis (Ahuja et al. 2007, Cell).
Comparative analyses of WH2 domain-mediated functions of the actin cytoskeleton
Cobl uses multiple Wiskott-Aldrich-Syndrom-Protein-Homology 2 (WH2) domains for interaction with monomeric actin. The project therefore studies the properties and the functions of the different WH2 domains of Cobl. First results of such studies suggest that the individual properties and the organization of WH2 domains within the Coble molecule are of high importance for the actin nucleation activity of Cobl. Our working hypothesis is that Cobl uses the third WH2 domain, which is located at the end of a long linker (L2), to attach an actin monomer to the back side of a linear actin dimer preassembled by the action of Wh2 domains number 1 and 2. This would result in a trimeric actin nucleus, which already corresponds to the full diameter of an actin filament and therefore offers a complete plus end for further - now spontaneous - attachment of actin monomers (polymerization) and therefore leads to rapid filament formation and elongation (Ahuja et al. 2007, Cell).
Schematic representation of the WH2 domain order in Cobl. Cobl has three WH2 domains. Each binds one actin monomer. Cobl can thus associate with three actin monomers - biophysical calculations suggest that three is the minimal number of actin monomers for overcoming the kinetic barrier of actin nucleation. The ability of Cobl to nucleate filaments experimentally proves that indeed an assembly of three monomers is sufficient for the generation of an actin nucleus. The molecular mechanism cobl uses, however, is not restricted on assembling three actin monomers but Cobl also seems to actively orchestrate their exact position in three dimensions.
Our working hypothesis is based on the topology of WH2 domains in Cobl. WH2 domains number 1 and 2 are so close to each other that for steric reasons only a linear actin dimer can result from both domains binding an actin monomer. In contrast, the linker L2 is long enough to reach around an actin dimer. Also, actin binding of WH2 domain number 3 is weak (Ahuja et al. 2007, Cell). This may suggest that actin monomer addition by this Wh2 domain is the last step of Cobl-mediated actin nucleation and is supported by the multiple actin-actin-interactions, which would become possible if an actin dimer was preassembled by WH2 domains 1 and 2. If this scenario is true, two predictions can be made, Cobl-mediated actin nucleation should be dependent on all three WH2 domains and the long linker L2. Mutational analyses indeed show that this is the case (Ahuja et al. 2007, Cell).
Further work is required to fully reveal the mechanism of Cobl-mediated actin nucleation and the different steps during this complex process.
Functional analyses of Abp1
Genetic, cell biological and biochemical examinations in the model organism Drosophila melanogaster, which are conducted in collaboration with the Leibniz-Institute for Neurobiology, Magdeburg, show that the correct development of sensory organs, such as eyes and bristles, as well as the proper generation of neuromuscular synapses require actin filament formation mediated by the F-actin binding protein Abp1 (Kessels et al. 2000 Mol. Biol. Cell) at the cell cortex. The project therefore examines the abilities of Abp1 to bind to actin filament, to membranes and to activate the Arp2/3 complex in vitro as well as in the complete organism (also compare Pinyol et al. 2007, PLoS ONE).
The role of cytoskeletal and cytomatrix components in the structural and functional organization of neurons
Nerve cell communication requires specialy subcellular compartments and structures, which need to be organized on the molecular level (Gundelfinger, Kessels & Qualmann 2003, Nature Rev. Cell Biol.). We were able to demonstrate that the actin binding protein Abp1, a functional link between the actin cytoskeleton and membrane transport processes (Kessels et al. 2001 J. Cell Biol.), interacts with scaffold proteins specifc for the pre- and the postsynapse (Fenster et al. 2003, J. Biol Chem; Qualmann et al. 2004, J. Neurosci.) and that it is required for proper formation and maturation of synaptic nerve cell contacts (Haeckel et al. 2008, J. Neurosci.). The project examines the molecular basis of Abp1 functions in the formation and the plasticity of synaptic cell-cell contacts and in the neuromorphogenesis processes underlying the formation of neuronal networks.
.
Abp1 connects the actin cytoskeleton with membran transport processes and synaptic organization. Protein interaction studies and functional studies suggest a role of the actin binding protein Abp1 at the interface of several cellular processes. While the two N-terminal modules of Abp1 associate with actin filaments, Abp1 also associates with components of membrane trafficking processes via its C-terminal SH3 domain. Based on complex formation with components of the pre- and postsynaptic cytomatrix Abp1 furthermore seems to play a role in the formation, organization, function and plasticity of synaptic nerve cell connections.
On the presynaptic side Abp1 directly and tightly associates with the so-called Q domain of piccolo, a huge protein specifically located at the active zone of vesicle release. This interaction is dependent on the SH3 domain of Abp1. Further studies in cells show that Abp1 can connect piccolo and newly formed F-actin structures - a function that may help to organize and structurally modulate the active zone of the presynapse (Fenster et al. 2003, J. Biol. Chem.).
On the postsynaptic side Abp1 is a component of postsynaptic densities (PSDs), dense scaffold structures that contain a variety of membrane receptors and downstream signalling components and are located in small nerve cell protrusions called postsynaptic spines. Abp1 is recruited to the postsynapse by interactions with two scaffold proteins of the PSD, ProSAP1/Shank2 and ProSAP2/Shank3 (Qualmann et al 2004 J. Neurosci.) and further studies show that both Abp1-mediated connections to actin filaments in spines and Abp1-mediated Arp2/3 complex-dependent actin nucleation play crucial roles for the formation and the lateral size extension of the heads of postsynaptic spines (Haeckel et al. 2008, J. Neurosci.). As the postsynaptic functions of Abp1 furthermore respond to incoming signals from neurotransmitter receptors, it seems that Abp1 is one of the components mediating the cytoskeletal reorganization processes occurring during synaptic plasticity (Qualmann et al. 2004, J. Neurosci.; Haeckel et al. 2008, J. Neurosci.). As such alterations allow for changes in the reception and transmission of a reoccurring, signal of similar kind, synaptic plasticity, i.e. modulations of the postsynaptic organization, are thought to be the basis for learning and memory formation.