Bioinformatic analysis of ciliary transition zone proteins reveals insights into the evolution of ciliopathy networks

Abstract

Background: Cilia are critical for diverse functions, from motility to signal transduction, and ciliary dysfunction causes inherited diseases termed ciliopathies. Several ciliopathy proteins influence developmental signalling and aberrant signalling explains many ciliopathy phenotypes. Ciliary compartmentalisation is essential for function, and the transition zone (TZ), found at the proximal end of the cilium, has recently emerged as a key player in regulating this process. Ciliary compartmentalisation is linked to two protein complexes, the MKS and NPHP complexes, at the TZ that consist largely of ciliopathy proteins, leading to the hypothesis that ciliopathy proteins affect signalling by regulating ciliary content. However, there is no consensus on complex composition, formation, or the contribution of each component.

Results: Using bioinformatics, we examined the evolutionary patterns of TZ complex proteins across the extant eukaryotic supergroups, in both ciliated and non-ciliated organisms. We show that TZ complex proteins are restricted to the proteomes of ciliated organisms and identify a core conserved group (TMEM67, CC2D2A, B9D1, B9D2, AHI1 and a single TCTN, plus perhaps MKS1) which are present in >50% of all ciliate/flagellate organisms analysed in each supergroup. The smaller NPHP complex apparently evolved later than the larger MKS complex; this result may explain why RPGRIP1L, which forms the linker between the two complexes, is not one of the core conserved proteins. We also uncovered a striking correlation between lack of TZ proteins in non-seed land plants and loss of TZ-specific ciliary Y-links that link microtubule doublets to the membrane, consistent with the interpretation that these proteins are structural components of Y-links, or regulators of their formation.

Conclusions: This bioinformatic analysis represents the first systematic analysis of the cohort of TZ complex proteins across eukaryotic evolution. Given the near-ubiquity of only 6 proteins across ciliated eukaryotes, we propose that the MKS complex represents a dynamic complex built around these 6 proteins and implicated in Y-link formation and ciliary permeability.

Figure 1

Non-seed land plants lack ciliary Y-links. A - Cartoon of cilium structure indicating the basal body (BB), transition zone (TZ) and axoneme. The central pair microtubules (present in motile cilia) and stellate structure (present in Plantae) are shown in grey. B, C - Transmission electron micrographs of axoneme and transition zone architecture in mammals/trypanosomes (B) and non-seed land plants (C). Boxes represent a single enlarged doublet indicated by the asterisk; doublets were rotated to position the ciliary/flagellar membrane towards the top and right of the boxed area. Scale bars = 50 nm. Arrows indicate Y-links visible in mammalian and trypanosome cilia/flagella (B, top 3 panels). Note that these structures are absent in all non-seed plants examined (C, top 6 panels).

Figure 2

Evolutionary distribution of ciliary transition zone genes. Stylised eukaryotic tree showing distribution of TZ complex proteins across 6 eukaryotic supergroups. All non-ciliated organisms have been excluded from the table. Plantae are divided into two sub-groups: green algae and mosses/ferns (higher plants are non-ciliate). Both these sub-groups contain ciliated organisms; note that mosses and ferns lack all known TZ complex components. Black circles denote presence of a putative orthologue in >75% ciliated/flagellate organisms analysed in one supergroup. Dark grey, light grey and white circles represent presence in 50-74%, 25-49% and <25% organisms analysed in one supergroup, respectively. The boxed area shows proteins that are present in >50% organisms in every supergroup; these represent the core TZ components. MKS1 is present in >50% of every supergroup but missing from the Rhizaria, which are represented by only a single sequenced organism; additional genome sequences may indicate that MKS1 is also core. Additional file 2 shows a more detailed version listing individual organisms, and the full bioinformatic analysis is available in Additional file 1.