Drosophila transition fibers are essential for IFT-dependent ciliary elongation but not basal body docking and ciliary budding

Cilia are highly conserved organelles critical for animal development and perception. Dysfunction of cilia has been linked to a wide spectrum of human genetic diseases, termed ciliopathies.^1,2 Transition fibers (TFs) are striking ciliary base structures essential for cilia assembly. Vertebrates' TFs that originate from centriole distal appendages (DAs) mediate basal body docking to ciliary vesicles to initiate ciliogenesis and regulate the entry of ciliary proteins for axoneme assembly via intraflagellar transport (IFT) machinery.³ Although no distal appendages can be observed on Drosophila centrioles,^4,5 three key TF proteins, FBF1, CEP164, and CEP89, have obvious homologs in Drosophila. We aimed to compare their functions with their mammalian counterparts in Drosophila ciliogenesis. Here, we show that all three proteins are localized like TF proteins at the ciliary base in both sensory neurons and spermatocytes, the only two types of ciliated cells in flies. Fbf1 and Cep89 are essential for the formation of IFT-dependent neuronal cilia, but Cep164 is dispensable for ciliogenesis in flies. Strikingly, none are required for basal body docking and transition zone (TZ) assembly in IFT-dependent neuronal cilia or IFT-independent spermatocyte cilia. Furthermore, we demonstrate that Unc is essential to recruit all three TF proteins and establish a hierarchical order, with Cep89 acting on Fbf1. Collectively, our results not only demonstrate that TF proteins are required for IFT-dependent ciliogenesis in Drosophila, in agreement with an evolutionarily conserved function of these proteins in regulating ciliary protein entry, but also that the basal body docking function of TFs has diverged during evolution.

The cilium-like region of the Drosophila spermatocyte: an emerging flagellum?

Primary cilia and flagella are distinct structures with different functions in eukaryotic cells. Despite the fact that they share similar basic organization and architecture, a direct developmental continuity among them has not been reported until now. The primary cilium is a dynamic structure that typically assembles and disassembles during mitotic cell cycles, whereas the sperm axoneme is nucleated by the centriole inherited by the differentiating spermatid at the end of meiosis. Fruit flies display a remarkable exception to this general rule. Drosophila spermatocytes have an unusual axoneme-based structure reminiscent of primary cilia (the cilium-like region, or CLR). This structure persists through the meiotic divisions when it is internalized with the centriole to organize the centrosome and is finally inherited by young spermatids. Examination of elongating spermatids by transmission electron microscopy (EM) and cold regrowth experiments suggests that the motile axoneme derives directly from the elongation and remodelling of the immotile CLR. Both the CLR and elongating spermatid flagella have incomplete C-tubules that form longitudinal sheets associated with the B-tubule wall, unlike axonemes of other organisms in which C-tubules stop growing at the transition between the basal body and the axonemal doublets. Moreover, both the CLR and spermatid flagella lack a structured transition zone, a characteristic feature of ciliated cells. Uncoordinated (unc) mutants that lack C-remnants have short centrioles, suggesting that the C-sheets play a role in the elongation of the centriole after it docks to the cell membrane. The structural similarities between CLR and sperm axoneme suggest that the CLR can be considered as the basal region of the future axoneme and could represent the start point for its elongation.