Roderick P. Tas

Microtubule Minus-End Binding Protein CAMSAP2 Controls Axon Specification and Dendrite Development

In neurons, most microtubules are not associated with a central microtubule-organizing center (MTOC), and therefore, both the minus and plus-ends of these non-centrosomal microtubules are found throughout the cell. Microtubule plus-ends are well established as dynamic regulatory sites in numerous processes, but the role of microtubule minus-ends has remained poorly understood. Using live-cell imaging, high-resolution microscopy, and laser-based microsurgery techniques, we show that the CAMSAP/Nezha/Patronin family protein CAMSAP2 specifically localizes to non-centrosomal microtubule minus-ends and is required for proper microtubule organization in neurons. CAMSAP2 stabilizes non-centrosomal microtubules and is required for neuronal polarity, axon specification, and dendritic branch formation in vitro and in vivo. Furthermore, we found that non-centrosomal microtubules in dendrites are largely generated by γ-Tubulin-dependent nucleation. We propose a two-step model in which γ-Tubulin initiates the formation of non-centrosomal microtubules and CAMSAP2 stabilizes the free microtubule minus-ends in order to control neuronal polarity and development.

Differentiation between Oppositely Oriented Microtubules Controls Polarized Neuronal Transport

Summary Microtubules are essential for polarized transport in neurons, but how their organization guides motor proteins to axons or dendrites is unclear. Because different motors recognize distinct microtubule properties, we used optical nanoscopy to examine the relationship between microtubule orientations, stability, and modifications. Nanometric tracking of motors to super-resolve microtubules and determine their polarity revealed that in dendrites, stable and acetylated microtubules are mostly oriented minus-end out, while dynamic and tyrosinated microtubules are oriented oppositely. In addition, microtubules with similar orientations and modifications form bundles that bias transport. Importantly, because the plus-end-directed Kinesin-1 selectively interacts with acetylated microtubules, this organization guides this motor out of dendrites and into axons. In contrast, Kinesin-3 prefers tyrosinated microtubules and can enter both axons and dendrites. This separation of distinct microtubule subsets into oppositely oriented bundles constitutes a key architectural principle of the neuronal microtubule cytoskeleton that enables polarized sorting by different motor proteins.

Myosin-V Induces Cargo Immobilization and Clustering at the Axon Initial Segment

The selective transport of different cargoes into axons and dendrites underlies the polarized organization of the neuron. Although it has become clear that the combined activity of different motors determines the destination and selectivity of transport, little is known about the mechanistic details of motor cooperation. For example, the exact role of myosin-V in opposing microtubule-based axon entries has remained unclear. Here we use two orthogonal chemically-induced heterodimerization systems to independently recruit different motors to cargoes. We find that recruiting myosin-V to kinesin-propelled cargoes at approximately equal numbers is sufficient to stall motility. Kinesin-driven cargoes entering the axon were arrested in the axon initial segment (AIS) upon myosin-V recruitment and accumulated in distinct actin-rich hotspots. Importantly, unlike proposed previously, myosin-V did not return these cargoes to the cell body, suggesting that additional mechanism are required to establish cargo retrieval from the AIS.

MICAL3 flavoprotein monooxygenase forms a complex with centralspindlin and regulates cytokinesis

During cytokinesis, the antiparallel array of microtubules forming the central spindle organizes the midbody, a structure that anchors the ingressed cleavage furrow and guides the assembly of abscission machinery. Here, we identified a role for the flavoprotein monooxygenase MICAL3, an actin disassembly factor, in organizing midbody-associated protein complexes. By combining cell biological assays with cross-linking mass spectrometry, we show that MICAL3 is recruited to the central spindle and the midbody through a direct interaction with the centralspindlin component MKLP1. Knock-out of MICAL3 leads to an increased frequency of cytokinetic failure and a delayed abscission. In a mechanism independent of its enzymatic activity, MICAL3 targets the adaptor protein ELKS and Rab8A-positive vesicles to the midbody, and the depletion of ELKS and Rab8A also leads to cytokinesis defects. We propose that MICAL3 acts as a midbody-associated scaffold for vesicle targeting, which promotes maturation of the intercellular bridge and abscission.

Three-Step Model for Polarized Sorting of KIF17 into Dendrites

Kinesin and dynein motors drive bidirectional cargo transport along microtubules and have a critical role in polarized cargo trafficking in neurons [1, 2]. The kinesin-2 family protein KIF17 is a dendrite-specific motor protein and has been shown to interact with several dendritic cargoes [3-7]. However, the mechanism underlying the dendritic targeting of KIF17 remains poorly understood [8-11]. Using live-cell imaging combined with inducible trafficking assays to directly probe KIF17 motor activity in living neurons, we found that the polarized sorting of KIF17 to dendrites is regulated in multiple steps. First, cargo binding of KIF17 relieves autoinhibition and initiates microtubule-based cargo transport. Second, KIF17 does not autonomously target dendrites, but enters the axon where the actin cytoskeleton at the axon initial segment (AIS) prevents KIF17 vesicles from moving further into the axon. Third, dynein-based motor activity is able to redirect KIF17-coupled cargoes into dendrites. We propose a three-step model for polarized targeting of KIF17, in which the collective function of multiple motor teams is required for proper dendritic sorting.

A tissue-specific protein purification approach in Caenorhabditis elegans identifies novel interaction partners of DLG-1/Discs large

BackgroundAffinity purification followed by mass spectrometry (AP/MS) is a widely used approach to identify protein interactions and complexes. In multicellular organisms, the accurate identification of protein complexes by AP/MS is complicated by the potential heterogeneity of complexes in different tissues. Here, we present an in vivo biotinylation-based approach for the tissue-specific purification of protein complexes from Caenorhabditis elegans. Tissue-specific biotinylation is achieved by the expression in select tissues of the bacterial biotin ligase BirA, which biotinylates proteins tagged with the Avi peptide.ResultsWe generated N- and C-terminal tags combining GFP with the Avi peptide sequence, as well as four BirA driver lines expressing BirA ubiquitously and specifically in the seam and hyp7 epidermal cells, intestine, or neurons. We validated the ability of our approach to identify bona fide protein interactions by identifying the known LGL-1 interaction partners PAR-6 and PKC-3. Purification of the Discs large protein DLG-1 identified several candidate interaction partners, including the AAA-type ATPase ATAD-3 and the uncharacterized protein MAPH-1.1. We have identified the domains that mediate the DLG-1/ATAD-3 interaction, and show that this interaction contributes to C. elegans development. MAPH-1.1 co-purified specifically with DLG-1 purified from neurons, and shared limited homology with the microtubule-associated protein MAP1A, a known neuronal interaction partner of mammalian DLG4/PSD95. A CRISPR/Cas9-engineered GFP::MAPH-1.1 fusion was broadly expressed and co-localized with microtubules.ConclusionsThe method we present here is able to purify protein complexes from specific tissues. We uncovered a series of DLG-1 interactors, and conclude that ATAD-3 is a biologically relevant interaction partner of DLG-1. Finally, we conclude that MAPH-1.1 is a microtubule-associated protein of the MAP1 family and a candidate neuron-specific interaction partner of DLG-1.

Exploring cytoskeletal diversity in neurons

Cytoskeletal architecture underlies the diversity and function of neuronal compartments Often in biology, form follows function. For example, the ability of neurons to receive, process, and transmit information depends on their polarized organization into axons and dendrites. The cytoskeleton and associated motor proteins shape cells and establish spatial organization. Microtubules (MTs) and actin are core components of the cytoskeleton and are assembled through head-to-tail polymerization of α- and β-tubulin heterodimers and actin monomers, respectively, resulting in asymmetric, polarized polymers with two different ends, called plus and minus ends. The spatially regulated polymerization of MTs and actin can drive morphological transitions, such as local protrusion of the plasma membrane, to drive cell migration or the development of specialized extensions, such as axons or dendrites and their branches. In addition, the structural asymmetry of MTs and actin enables cytoskeletal motor proteins (myosin, kinesin, and dynein) to walk toward a specific end of the fibers. Given the extreme dimensions and functional compartmentalization of neurons, such active transport is critical to sort and distribute cellular cargoes. Recent studies have used advanced microscopy to reveal how the cytoskeleton takes many different forms to facilitate local functions in neurons (see the figure).

Nogo Receptor crystal structures with a native disulfide pattern suggest a novel mode of self-interaction

The Nogo Receptor (NgR) is a glycophosphatidylinositol-anchored cell-surface protein and is a receptor for three myelin-associated inhibitors of regeneration: myelin-associated glycoprotein, Nogo66 and oligodendrocyte myelin glycoprotein. In combination with different co-receptors, NgR mediates signalling that reduces neuronal plasticity. The available structures of the NgR ligand-binding leucine-rich repeat (LRR) domain have an artificial disulfide pattern owing to truncated C-terminal construct boundaries. NgR has previously been shown to self-associate via its LRR domain, but the structural basis of this interaction remains elusive. Here, crystal structures of the NgR LRR with a longer C-terminal segment and a native disulfide pattern are presented. An additional C-terminal loop proximal to the C-terminal LRR cap is stabilized by two newly formed disulfide bonds, but is otherwise mostly unstructured in the absence of any stabilizing interactions. NgR crystallized in six unique crystal forms, three of which share a crystal-packing interface. NgR crystal-packing interfaces from all eight unique crystal forms are compared in order to explore how NgR could self-interact on the neuronal plasma membrane.