IFT (intraflagellar transport) assembles and maintains sensory cilia on the dendritic endings of chemosensory neurons within the nematode Caenorhabditis elegans. During IFT, macromolecular protein complexes called IFT particles (which carry ciliary precursors) are moved from the base of the sensory cilium to its distal tip by anterograde IFT motors (kinesin-II and Osm-3 kinesin) and back to the base by retrograde IFT-dynein [Rosenbaum and Witman (2002) Nat. Rev. Mol. Cell Biol. 3, 813–825; Scholey (2003) Annu. Rev. Cell Dev. Biol. 19, 423–443; and Snell, Pan and Wang (2004) Cell 117, 693–697]. In the present study, we describe the protein machinery of IFT in C. elegans, which we have analysed using time-lapse fluorescence microscopy of green fluorescent protein-fusion proteins in concert with ciliary mutants.
- intraflagellar transport
- microtubule-based motility
Signal reception by chemosensory neurons, which control chemotaxis in Caenorhabditis elegans, depends on sensory cilia, specialized non-motile axonemes that detect environmental chemicals via chemoreceptors that are concentrated in the ciliary membranes [1–7]. To understand how IFT (intraflagellar transport) [1–3] contributes to the assembly, maintenance and function of these sensory cilia, we identified previously two candidate anterograde IFT motors and localized them to sensory cilia ; we developed an in vivo motility assay for visualizing the movement of these IFT motors and the corresponding IFT-particle subunits at rates of the order of 1 μm/s within sensory cilia of living animals ; and we used this assay, in concert with a chemotaxis-deficient ciliary mutant, to investigate the role of IFT-dynein in retrograde IFT . The results we obtained are consistent with the model displayed in Figure 1.
The prototypical anterograde IFT motor, heterotrimeric kinesin-II (Figure 1), was first purified from sea-urchin embryos, where it is required for ciliogenesis [11–14]. This protein contains two heterodimerized N-terminal motor subunits, which move towards the plus ends of microtubules at 0.4 μm/s and have a non-motor subunit, the KAP, linked to their cargo-binding C-terminal ends [12,13]. In C. elegans, mutations in a closely related kinesin subunit, OSM-3, was found to cause defects in sensory ciliogenesis ([15,16]; Figure 2), which initially led to the hypothesis that the KAP homologue in worm probably associates with OSM-3 to drive ciliogenesis . Careful fractionation experiments, however, showed that sensory cilia on C. elegans neurons contain two distinct kinesin-II complexes, heterotrimeric kinesin-II (which is very similar to sea-urchin kinesin-II) and homodimeric Osm-3-kinesin . Although both these motor proteins move along sensory cilia and are, hence, strong candidates for being anterograde IFT motors, the precise functional relationships between them is still not clear (Figure 1). The retrograde IFT motor, IFT-dynein, was also first identified in sea-urchin embryos as a form of cytoplasmic dynein that is up-regulated during ciliogenesis . In C. elegans, the IFT-dynein heavy chain is CHE-3, mutations in which give rise to truncated sensory cilia that are filled with IFT particles (; Figure 2) and lack retrograde IFT . The subunit composition of IFT-dynein is unknown; however, a probable accessory subunit, the light intermediate chain, DLIC, has been identified .
We wish to improve our understanding of the roles of these IFT motors in sensory ciliogenesis in C. elegans (Figure 1). To this end, our current work is aimed at using the IFT motility assay in concert with ciliary mutants to answer the following: (i) what is the relationship between the two anterograde IFT motors and how do they relate to the transport of IFT-particle complexes A and B? (ii) do IFT motors/particles carry precursors of axonemes (e.g. tubulin subunits) or membranes (e.g. chemoreceptors)? (iii) do IFT motors contribute directly to sensory signalling? and (iv) do IFT motors also transport IFT particles in axons and dendrites? To do this, it is necessary to identify the full inventory of molecules involved in IFT. Accordingly, a number of candidate IFT motor and IFT-particle subunits can be identified among the deduced protein sequences encoded by the worm genome (Table 1). In addition, defective sensory cilia are represented in at least four classes of mutant C. elegans, namely chemotaxis-defective (che), osmotic avoidance-defective (osm), dauer larva formation-defective (daf) and dye-filling-defective (dyf) mutants (Figure 2). For some of these mutants, the observed defects in sensory ciliary structure and chemosensory behaviour are probably due to defective IFT. For example, the deformed sensory cilia and the behavioural phenotypes associated with mutations in the anterograde IFT-motor subunit OSM-3 and the retrograde IFT-motor subunits, CHE-3 and D2LIC, are supposed to result from defects in IFT [2,4,10,16,18,19]. Similarly, mutations in several IFT-particle subunits, e.g. OSM-1, -5 and -6, CHE-2, -11 and -13 and DAF-10, probably result in defects in intraflagellar transport and delivery of ciliary precursors [20–23].
Thus genome sequencing and ciliary mutant isolation have provided a rich repository of candidate components of the IFT machinery required for building C. elegans sensory cilia. Our identification of Che-3-dynein as a retrograde IFT motor suggests that in vivo IFT motility assays of these proteins by time-lapse microscopy in appropriate mutant backgrounds provides a powerful method for dissecting the mechanisms and functions of IFT in sensory ciliogenesis. Our own efforts in this direction have been frustrated by our inability, so far, to sort out the functional relationship between the two anterograde IFT motors. For example, it is puzzling that the motility of the IFT-particle subunit, green fluorescent protein–OSM-6, persists in cilia of mutants lacking OSM-3-kinesin function, even though these mutants display defects in sensory ciliogenesis and chemotactic behaviour. It is probable that heterotrimeric kinesin-II drives the transport of this IFT particle; however, so far, single mutants in subunits of this putative IFT motor display no defects in ciliary structure or chemosensory behaviour. We are currently working towards resolving this puzzle.
This work was supported by NIH grant no. GM50718.
Lipids, Rafts and Traffic: A Focus Topic at BioScience2004, held at SECC Glasgow, U.K., 18–22 July 2004. Edited by G. Banting (Bristol, U.K.), N. Bulleid (Manchester, U.K.), C. Connolly (Dundee, U.K.), S. High (Manchester, U.K.) and K. Okkenhaug (Babraham Institute, Cambridge, U.K.)
Abbreviations: IFT, intraflagellar transport
- © 2004 The Biochemical Society