Primary Cilia, Ciliogenesis and the Actin Cytoskeleton: A Little Less Resorption, A Little More Actin Please

Abstract

Primary cilia are microtubule-based organelles that extend from the apical surface of most mammalian cells, forming when the basal body (derived from the mother centriole) docks at the apical cell membrane. They act as universal cellular "antennae" in vertebrates that receive and integrate mechanical and chemical signals from the extracellular environment, serving diverse roles in chemo-, mechano- and photo-sensation that control developmental signaling, cell polarity and cell proliferation. Mutations in ciliary genes cause a major group of inherited developmental disorders called ciliopathies. There are very few preventative treatments or new therapeutic interventions that modify disease progression or the long-term outlook of patients with these conditions. Recent work has identified at least four distinct but interrelated cellular processes that regulate cilia formation and maintenance, comprising the cell cycle, cellular proteostasis, signaling pathways and structural influences of the actin cytoskeleton. The actin cytoskeleton is composed of microfilaments that are formed from filamentous (F) polymers of globular G-actin subunits. Actin filaments are organized into bundles and networks, and are attached to the cell membrane, by diverse cross-linking proteins. During cell migration, actin filament bundles form either radially at the leading edge or as axial stress fibers. Early studies demonstrated that loss-of-function mutations in ciliopathy genes increased stress fiber formation and impaired ciliogenesis whereas pharmacological inhibition of actin polymerization promoted ciliogenesis. These studies suggest that polymerization of the actin cytoskeleton, F-actin branching and the formation of stress fibers all inhibit primary cilium formation, whereas depolymerization or depletion of actin enhance ciliogenesis. Here, we review the mechanistic basis for these effects on ciliogenesis, which comprise several cellular processes acting in concert at different timescales. Actin polymerization is both a physical barrier to both cilia-targeted vesicle transport and to the membrane remodeling required for ciliogenesis. In contrast, actin may cause cilia loss by localizing disassembly factors at the ciliary base, and F-actin branching may itself activate the YAP/TAZ pathway to promote cilia disassembly. The fundamental role of actin polymerization in the control of ciliogenesis may present potential new targets for disease-modifying therapeutic approaches in treating ciliopathies.

Keywords: ROCK inhibitors; actin cytoskeleton; cilia; ciliogenesis; ciliopathies; cytoskeletal drugs; drug screen; polycystic kidney disease.

FIGURE 1

The primary cilium, actin dynamics and the actin cytoskeleton. Main panel: The primary cilium and the actin cytoskeleton. The primary cilium forms when the mother centriole docks at the apical membrane to form the basal body. Microtubules (gray cylinders) then nucleate at the basal body to initiate formation of the axoneme with a 9+0 microtubule arrangement. The sub-distal appendages mediate microtubule anchoring, and the pericentriolar material may act as a microtubule organizing center. Y-shaped links (dark brown) connect the microtubule doublets to the ciliary membrane at the ciliary transition zone, which acts as a gate to control the entry and exit of proteins and lipids. Intraflagellar transport (IFT) protein complexes (orange and yellow) transport cargo (indicated by arrows) along the axoneme by using microtubule based motor proteins. The ciliary pocket is an endocytic membrane domain around the base of the cilium implicated in actin dynamics, the transport of membrane associated proteins to cilia and in ciliary disassembly. The ciliary tip is a source of extracellular vesicles and is involved in early ciliary disassembly through decapitation. The actin cytoskeleton modulates ciliogenesis through effects on both vesicle trafficking (lower left; tan symbols and dashed arrow) and actin cytoskeleton remodeling mediated by acto-myosin contractions (lower right; green symbols and arrows). Actin-binding proteins are indicated by dark red symbols. Inset left: Microfilament assembly and treadmilling. Actin is present in cells as both a monomer (globular, G-actin) and a polymer (filamentous, F-actin). Actin binds and can hydrolyze ATP to ADP. There is preferential additional of ATP-bound monomers to the (+) end of the polymer and ATP is hydrolyzed upon filament assembly. The preferential dissociation of ADP-bound actin results in a “treadmilling” effect, whereby the F-actin filaments exhibit net growth at their (+) ends and net dissociation at their (−) ends. Both polymerization and depolymerization require actin binding proteins. Profilin can sequester G-actin from the pool of actin monomers but can also catalyze the exchange of ADP to ATP, converting monomers to the more polymerizable ATP-bound form. Actin polymerization requires nucleating factors, such as Spire or formins, or the actin-related protein complex (ARP2/3). Actin depolymerization remains incompletely understood but members of the yeast actin depolymerization factor (ADF)/cofilin family can enhance dissociation of monomers from the (−) end. Inset right: Actin cytoskeleton organization. Many actin binding proteins can alter the arrangement and structure of F-actin. ARP2/3 can elicit F-actin branching. Members of the yeast actin depolymerization factor (ADF)/cofilin family can both enhance dissociation of G-actin monomers from the (−) end and can sever filaments to produce additional (−) ends. Filamins and various other actin binding proteins can crosslink actin to form complex networks, can anchor F-actin to membranes and can bundle actin into stress fibers.

FIGURE 2

Imaging the actin cytoskeleton and actin-binding proteins within cells. (A) Zeiss Airyscan microscopy images of COS-7 cells probed with an actin-binding artificial non-antibody binding protein (“Affimer”) biotinylated on the C-terminal cysteine and visualized with fluorescent streptavidin (green). Cells were co-stained with fluorescent phalloidin (left panel; red) or antibody against non-muscle myosin 2B/MYH10 (right panel; red). Scale bars 10 or 5 μm, as indicated. (B) Confocal microscopy (upper panel) and high content imaging (lower panel) of ciliated serum-starved mouse inner medullary collecting duct (mIMCD3) cells expressing LifeAct-GFP (upper panel; green) or probed with AlexaFluor 488-phalloidin conjugate (lower panel; green). Primary cilia are marked by ARL13B (red). Frame indicates magnified inset showing detail of actin stress fibers in the upper panel. Scale bars = 20 μm. (C) Confocal microscopy image of human female dermal fibroblasts heterozygous for the FLNA frameshift mutation c.1587delG, p.(K529Nfs*40) (Adams et al., 2012) probed with antibodies for the actin-binding proteins filamin A (green) and filamin B (red), and counterstained with phalloidin-AlexaFluor633 conjugate (blue). The FLNA gene is carried on the X chromosome and random X-inactivation in different cells results in either haploinsufficiency (yellow-colored cell; indicated by arrow) or complete loss of filamin A protein (magenta-colored cell; arrowhead). Scale bar = 10 μm.

FIGURE 3

Ciliary membrane remodeling of the actin-binding protein PCARE in the primary cilium and photoreceptor connecting cilium. The proposed role of PCARE in the cilium (left panel) and in the specialized cilium of the photoreceptor outer segment and connecting cilium (right panel) are indicated. PCARE (red) has been shown to interact with actin-binding proteins (dark red symbols) associated with de novo F-actin network assembly including the ARP2/3 complex, gelsolin, profilin and WASF3 (orange). PCARE in mice localizes to the basal body, the microtubules of the connecting cilium and extends into the newer outer segment discs (Corral-Serrano et al., 2020). Overexpression of PCARE in hTERT RPE-1 cells resulted in WASF-3 translocating from F-actin to the cilium and induced expansion of the ciliary membrane to form a bulbous tip (left panel). PCARE is a retinal specialist protein that facilitates actin remodeling via recruitment of WASF3 to induce membrane remodeling and expansion that produces new outer segment discs for photo-sensation in the photoreceptor cell (Corral-Serrano et al., 2020).