Artigo Acesso aberto Revisado por pares

Role for intraflagellar transport in building a functional transition zone

2018; Springer Nature; Volume: 19; Issue: 12 Linguagem: Inglês

10.15252/embr.201845862

ISSN

1469-3178

Autores

Victor L. Jensen, Nils J. Lambacher, Chunmei Li, Swetha Mohan, Corey Williams, Peter N. Inglis, Bradley K. Yoder, Oliver E. Blacque, Michel R. Leroux,

Tópico(s)

Urological Disorders and Treatments

Resumo

Article14 November 2018free access Transparent process Role for intraflagellar transport in building a functional transition zone Victor L Jensen Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Nils J Lambacher Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Chunmei Li Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Swetha Mohan Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Corey L Williams Department of Cell, Developmental, and Integrative Biology, University of Alabama at Birmingham Medical School, Birmingham, AL, USA Search for more papers by this author Peter N Inglis Department of Biology, Kwantlen Polytechnic University, Surrey, BC, Canada Search for more papers by this author Bradley K Yoder Department of Cell, Developmental, and Integrative Biology, University of Alabama at Birmingham Medical School, Birmingham, AL, USA Search for more papers by this author Oliver E Blacque orcid.org/0000-0003-1598-2695 School of Biomolecular and Biomedical Science, University College Dublin, Belfield, Dublin 4, Ireland Search for more papers by this author Michel R Leroux Corresponding Author [email protected] orcid.org/0000-0003-0788-9298 Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Victor L Jensen Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Nils J Lambacher Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Chunmei Li Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Swetha Mohan Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Corey L Williams Department of Cell, Developmental, and Integrative Biology, University of Alabama at Birmingham Medical School, Birmingham, AL, USA Search for more papers by this author Peter N Inglis Department of Biology, Kwantlen Polytechnic University, Surrey, BC, Canada Search for more papers by this author Bradley K Yoder Department of Cell, Developmental, and Integrative Biology, University of Alabama at Birmingham Medical School, Birmingham, AL, USA Search for more papers by this author Oliver E Blacque orcid.org/0000-0003-1598-2695 School of Biomolecular and Biomedical Science, University College Dublin, Belfield, Dublin 4, Ireland Search for more papers by this author Michel R Leroux Corresponding Author [email protected] orcid.org/0000-0003-0788-9298 Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada Search for more papers by this author Author Information Victor L Jensen1, Nils J Lambacher1, Chunmei Li1, Swetha Mohan1, Corey L Williams2, Peter N Inglis3, Bradley K Yoder2, Oliver E Blacque4 and Michel R Leroux *,1 1Department of Molecular Biology and Biochemistry, and Centre for Cell Biology, Development and Disease, Simon Fraser University, Burnaby, BC, Canada 2Department of Cell, Developmental, and Integrative Biology, University of Alabama at Birmingham Medical School, Birmingham, AL, USA 3Department of Biology, Kwantlen Polytechnic University, Surrey, BC, Canada 4School of Biomolecular and Biomedical Science, University College Dublin, Belfield, Dublin 4, Ireland *Corresponding author. Tel: +1 778 782 6683; E-mail: [email protected] EMBO Rep (2018)19:e45862https://doi.org/10.15252/embr.201845862 PDFDownload PDF of article text and main figures. Peer ReviewDownload a summary of the editorial decision process including editorial decision letters, reviewer comments and author responses to feedback. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info Abstract Genetic disorders caused by cilia dysfunction, termed ciliopathies, frequently involve the intraflagellar transport (IFT) system. Mutations in IFT subunits—including IFT-dynein motor DYNC2H1—impair ciliary structures and Hedgehog signalling, typically leading to "skeletal" ciliopathies such as Jeune asphyxiating thoracic dystrophy. Intriguingly, IFT gene mutations also cause eye, kidney and brain ciliopathies often linked to defects in the transition zone (TZ), a ciliary gate implicated in Hedgehog signalling. Here, we identify a C. elegans temperature-sensitive (ts) IFT-dynein mutant (che-3; human DYNC2H1) and use it to show a role for retrograde IFT in anterograde transport and ciliary maintenance. Unexpectedly, correct TZ assembly and gating function for periciliary proteins also require IFT-dynein. Using the reversibility of the novel ts-IFT-dynein, we show that restoring IFT in adults (post-developmentally) reverses defects in ciliary structure, TZ protein localisation and ciliary gating. Notably, this ability to reverse TZ defects declines as animals age. Together, our findings reveal a previously unknown role for IFT in TZ assembly in metazoans, providing new insights into the pathomechanism and potential phenotypic overlap between IFT- and TZ-associated ciliopathies. Synopsis A temperature-sensitive mutation in C. elegans CHE-3, the orthologue of mammalian DYNC2H1, reveals a role for the intraflagellar transport (IFT) retrograde dynein motor in anterograde IFT transport, cilium structure maintenance, and assembly of a functional transition zone ("ciliary gate"). Reversible temperature-sensitive (ts) IFT-dynein (che-3) mutant uncovered in C. elegans. Loss of CHE-3 function abrogates retrograde IFT and eventually halts anterograde IFT. CHE-3 is required for cilium structure maintenance. CHE-3 facilitates assembly and function of the transition zone "ciliary gate". Introduction Cilia are evolutionarily conserved eukaryotic organelles that perform motility and/or sensory functions in protists, plants and multicellular animals (metazoans) 1234. The microtubule-based axonemes of motile and non-motile (primary) cilia are templated from a basal body and harbour within their proximal-most region a transition zone (TZ) characterised by Y-shaped structures that connect the microtubules to the overlying ciliary membrane 56789. The TZ consists of numerous proteins, most of which are membrane-associated, that form a diffusion barrier or "ciliary gate" 10111213141516. The function of this gate is important for the correct localisation of signalling proteins within the organelle, including components of the Hedgehog signalling pathway in vertebrates 17181920. Mobilisation of ciliary proteins across this barrier and within the compartment depends on an intraflagellar transport (IFT) trafficking machinery 212223. Discovered by the Rosenbaum Lab in 1993 24, IFT has a well-established role in building functional cilia 212225. Five IFT modules participate in moving ciliary cargo, such as tubulin building blocks and receptors, into or out of cilia 22262728. Anterograde trafficking from the basal body to the ciliary tip is carried out by kinesin-2 family motors, namely heterotrimeric kinesin-II and homodimeric KIF17/OSM-3 2930. Retrograde trafficking from the ciliary tip to the base depends on a multi-protein IFT-dynein assembly that includes cytoplasmic dynein-2 heavy chain (DYNC2H1), as well as light, intermediate and light intermediate chains 313233343536. The kinesin and dynein molecular motors mobilise two cargo-binding subcomplexes, termed IFT-A (containing 7 or more proteins) and IFT-B (at least 15 proteins), which play important roles in retrograde and anterograde trafficking, respectively. Disruption of IFT-dynein or IFT-A subunits in mammals, Caenorhabditis elegans and Chlamydomonas reinhardtii leads to short cilia with accumulations of IFT proteins at expanded ("bulbous") ciliary tips 83233373839. In contrast, mutations in IFT-B subunits typically result in near-complete loss of the ciliary axoneme 84041. Finally, an accessory complex containing at least 8 Bardet–Biedl syndrome (BBS) proteins, termed BBSome, associates with the IFT machinery and helps to transport specific ciliary cargo 424344. Notably, several components of the Hedgehog signalling pathway are dependent on IFT-associated proteins to regulate their dynamic ciliary localisation 181945. The vast majority of known TZ and IFT proteins (over 60 in total) are linked to human disorders termed ciliopathies 4647. Disruption of an IFT-A, IFT-B or IFT-dynein subunit typically causes so-called "skeletal" ciliopathies, such as Jeune asphyxiating thoracic dystrophy (JATD), cranioectodermal dysplasia (CED) and Ellis-van Creveld syndrome (EVC) 47484950. In more rare cases, mutations in the same IFT proteins can be associated with ciliopathies that principally affect the eye and kidney—including retinitis pigmentosa (RP), cone-rod dystrophy (CRD) and nephronophthisis (NPHP) 47515253. The reason for such different phenotypic presentations is unclear. However, eye- and/or kidney-associated ciliopathies, sometimes coupled with liver and brain anomalies as well as polydactyly, are the hallmarks of TZ-associated ciliopathies; these include NPHP, Meckel syndrome, Joubert syndrome and Senior-Løken syndrome 464754. Ciliopathies are therefore complex—exhibiting multigenic and multiallelic traits, and a spectrum of partially overlapping clinical phenotypes. A central question relevant to ciliary biology and human disease is how specific mutations in particular components of different ciliary structures/processes correlate with discrete clinical ailments. Moreover, whether ciliary defects can be reversed in ciliopathy patients remains virtually unknown. In this study, we identify the first metazoan temperature-sensitive (ts) cilia mutant, which we show affects the function of CHE-3, the C. elegans orthologue of the retrograde IFT-dynein-2 motor DYNC2H1 33. We use this novel ts-che-3 mutant to reveal roles for retrograde IFT in maintaining the ciliary axoneme, anterograde IFT, as well as correct assembly and function of the TZ ciliary gate. Notably, the ts-che-3 mutant enables us to show post-developmental regrowth of cilia and restoration of TZ gate function. Our findings suggest a functional interplay between IFT and the TZ, whereby disruption of IFT can manifest in defects in TZ assembly and function. This suggests a potential mechanism for how IFT mutations could cause typical IFT-associated skeletal ciliopathies, TZ-associated renal/retinal/brain ciliopathies or, potentially, a combination of both. Furthermore, the ability to rebuild the ciliary axoneme and TZ post-developmentally in our model system raises the possibility of therapeutic intervention for some post-natal aspects of IFT- and TZ-associated ciliopathies. Results Genetic screen uncovers a temperature-sensitive dynein-2 heavy chain mutant To provide new insights into retrograde IFT, we conducted an unbiased genetic screen to obtain mutants affecting this process (Fig 1A). A C. elegans paraquat-resistant mutant library, known to be enriched for mutants with defective cilia 55, was subjected to two sequential screens. The first screen identified animals deficient in their uptake of a fluorescent dye into their head and tail sensory neurons (Fig 1A), a process that depends on intact chemosensory cilia which are normally exposed to the environment 85657. Such dye-filling (dyf) mutants frequently have cilia structure defects caused by mutations in core IFT or BBS proteins 8374258596061. The second screen involved introducing GFP-tagged IFT markers (IFT-A subunit CHE-11/IFT140, IFT-B subunit OSM-5/IFT88 and IFT-dynein-associated XBX-1/DYNC2LI1) in the dyf mutants and visualising them by microscopy. This helped reveal IFT protein accumulations (Fig 1A) typical of retrograde IFT mutants 333760. Figure 1. Screen for retrograde IFT defects identifies a temperature-sensitive allele of the dynein heavy chain CHE-3, the orthologue of mammalian DYNC2H1 Genetic screen for retrograde IFT defects uncovers a temperature-sensitive (ts) mutation in C. elegans CHE-3 (dynein-2 heavy chain). Paraquat-resistant strains were first screened for dye-filling defects (Dyf phenotype) typically indicative of cilia structure anomalies (examples of normal and defective DiI dye-filling are shown). Three different IFT reporters were then introduced into the dyf mutants (CHE-11::GFP shown as an example) to reveal candidate retrograde IFT mutants showing strong IFT protein accumulation at cilia tips. Whole-genome sequencing revealed the likely causative mutation in one of the strains, namely in the retrograde IFT-dynein motor CHE-3 (DYNC2H1 human orthologue). A variable Dyf phenotype at 20°C suggested the possibility of a ts mutation. Scale bar, 20 μm. The Dyf phenotype of the ts-che-3 mutant was confirmed, with animals grown at the permissive temperature (15°C) showing the ability to uptake dye, while those raised at the restrictive temperature (25°C) are Dyf. Scale bar, 20 μm. Using the homodimeric kinesin OSM-3 as a marker for IFT, proper localisation is seen in the ts-che-3 mutant at the permissive temperature, with the strongest signal at the basal body (bb). At the restrictive temperature, most of the OSM-3 signal is found within the axoneme of the short, bulbous cilia. Scale bar is 4 μm. The ts-che-3 mutant shows normal osmotic avoidance at permissive temperature but not at the restrictive temperature, indicating defective chemosensation. The osm-9 mutant is included as a positive control. n = 50 animals; error bars are standard error. Download figure Download PowerPoint Several candidate retrograde IFT mutants were uncovered, one of which was subjected to whole-genome sequencing and found to contain a likely deleterious mutation in the gene encoding the dynein-2 heavy chain, CHE-3. A genetic complementation assay with an existing che-3 null allele (e1124) confirmed that che-3 is indeed the gene responsible for the Dyf phenotype. The mutation, G1997E, alters a glycine residue conserved across all protist, plant and metazoan DYNC2H1 orthologues (Fig EV1A). It is located immediately adjacent to an essential aspartic acid residue that contacts Mg2+-ATP in the second AAA ATPase domain of the motor protein (Fig EV1B) 62636465. The replacement of glycine with glutamate likely interferes with the binding or affinity of ATP in the AAA2 domain, which is required for dynein motor activity 62636465. Click here to expand this figure. Figure EV1. The temperature-sensitive mutation in C. elegans CHE-3 (orthologue of human DYNC2H1) is in a glycine residue next to the ATP binding site and immediately adjacent to an evolutionarily conserved aspartic acid (Asp; D) likely required for ATP binding/hydrolysis in the AAA2 domain Multiple sequence alignment showing a portion of the AAA2 domain containing the glycine (G1997) residue mutated to glutamic acid (Glu; E) in CHE-3 (G2056 residue in human DYNC2H1). Close-up of Gly and Asp amino acids in relation to ATP-Mg2+ within the AAA2 domain of human DYNC2H1 (crystal structure PDB accession number 4RH7). Download figure Download PowerPoint Although a che-3 null mutant has already been characterised 3366, we discovered that the G1997E mutation in CHE-3 causes a temperature-sensitive (ts) Dyf phenotype (Fig 1). We therefore sought to study this mutant further, as it represents the first metazoan temperature-sensitive cilia mutant to be uncovered. The ts-che-3 mutant exhibits temperature-sensitive cilia defects At the permissive temperature of 15°C, the ts-che-3 mutant features no dye-filling defective phenotype and no overt ciliary defects, as judged by the correct localisation of IFT markers (Fig 1B and C). In contrast, 100% of ts-che-3 animals are dye-filling defective at the restrictive temperature of 25°C (Fig 1B). This dye-filling phenotype can be explained by severe cilia structure anomalies, since the ts-che-3 mutant has short cilia with large bulbous accumulations of IFT reporters (Fig 1C), similar to the previously reported phenotype for the che-3 null allele 33. Consistent with these structural ciliary defects, the ts-che-3 animals display a temperature-dependent sensory/behavioural phenotype. At permissive temperature, ts-che-3 mutants avoid a high-osmolarity barrier, similar to wild-type animals (Fig 1D). However, at restrictive temperature, the ts-che-3 animals largely fail to sense and avoid the barrier. This phenotype is comparable to that of a mutant with defects in the osmosensing ciliary TRPV ion channel (osm-9) (Fig 1D) and mutants (e.g., IFT) with ciliogenic defects 42606768. Thus, our genetic screen for IFT retrograde mutants has uncovered a novel dynein-2 temperature-sensitive mutation that exhibits largely normal ciliary structure and function at the permissive temperature, while resembling the null mutant at the restrictive temperature. Dynein-2 is required for correct assembly of transition zone proteins There is growing evidence for physical and functional connections between the IFT system and the TZ ciliary gate (see 3). Yet, in C. elegans, the function of IFT is largely unimpaired when the TZ is completely disrupted 12. Whether IFT is required for the correct assembly and function of the TZ has not been specifically investigated in metazoans. As part of our molecular analyses of the ts-che-3 mutant, we discovered that at the restrictive temperature, a fluorescently tagged TZ protein (MKS-6; orthologue of human CC2D2A/MKS6/JBTS9) displays, in addition to seemingly normal TZ localisation, ectopic distribution to more distal regions of the short and bulbous cilia (Fig 2A). We imaged additional TZ proteins in the ts-che-3 mutant at restrictive temperature (25°C), namely NPHP-4 (NPHP4) and CEP-290 (CEP290). We found similar ectopic localisation with NPHP-4 (Fig 2A). Despite CEP-290 appearing similar at the restrictive and permissive temperatures, there is a small, yet significant amount of ectopic TZ protein in the ciliary middle segment. We confirmed that the TZ defects are not exclusive to the ts-che-3 mutant, since similar ectopic localisation of several TZ proteins (NPHP-1, NPHP-4, MKS-1, MKSR-1 and MKS-6) is observed in the che-3 null mutant (Fig EV2A). Furthermore, to rule out any gain-of-function or dominant-negative effects, we looked at animals heterozygous for the ts-che-3 allele grown at the restrictive temperature. We found that they appear wild-type in terms of MKS-6 protein localisation (Fig EV2B). Figure 2. Loss of IFT-dynein function at restrictive temperature results in TZ protein localisation defects that are reversed at permissive temperature Three TZ markers (NPHP-4, CEP-290 and MKS-6) are shown at both the permissive (15°C) and restrictive (25°C) temperatures in the ts-che-3 mutant. At the permissive temperature, all three markers localise to the TZ, while at the restrictive temperature exhibit ectopic localisation (leakage) distally into the ciliary axoneme. While NPHP-4 and MKS-6 show significant accumulations, CEP-290 is grossly normal. At the permissive temperature, both IFT co-markers (CHE-13 and XBX-1) display strong localisation to the basal body (bb) and along the axoneme. At the restrictive temperature, the IFT markers show accumulation in the axoneme. Accumulations emphasised using asterisks. To determine whether ciliary compartmentalisation could be restored, young ts-che-3 mutant adults were transferred to the permissive temperature after being raised at the restrictive temperature. The TZ proteins reassembled at the TZ, with loss of the ectopic accumulations, and IFT protein localisation was also restored—strongest at the basal body, similar to animals grown at the permissive temperature. After 24 h at the restrictive temperature, TZ proteins appear to begin to accumulate distally within the axoneme in ts-che-3 mutants. This is seen, in small amounts, for NPHP-4 and CEP-290 but with stronger ectopic localisation for MKS-6. The IFT proteins CHE-13 and XBX-1 also show accumulations in the axoneme after 24 h at the restrictive temperature. Data information: Fluorescence quantification is shown for each marker at the indicated temperature in the heat maps on the right. Each point in the plot represents one pixel along the centre of the basal body (BB), transition zone (TZ) and middle segment (MS) regions. Dotted areas (three pixels) in the MS were used to quantify ectopic localisation (TZ markers) or accumulation (IFT markers) for statistical analyses. n = 10 cilia (4–7 animals), Kruskal–Wallis test, *P < 0.05, **P < 0.01. tz, transition zone; bb, basal body; scale bars are 4 μm. Download figure Download PowerPoint Click here to expand this figure. Figure EV2. The null and ts-che-3 mutant, but not the heterozygous ts mutant show ectopic TZ protein localisation, which recovers after FRAP in the ts mutant In the null allele che-3(e1124), five transition zone (TZ) proteins show distal ectopic accumulation (marked by asterisks) in the bulbous axonemes of the mutant cilia, something not observed in wild-type animals. TZs are indicated with an arrowhead. Dotted lines and arrows indicate direction and location of either cilia or dendrites (den). Scale bar is 4 μm. When grown at the restrictive temperature the cilia of animals heterozygous for the ts-che-3 allele resemble those found in wild-type, with no apparent ectopic TZ (MKS-6) or IFT (XBX-1) protein accumulations seen in the homozygous mutant. Scale bar is 4 μm. After photobleaching the distal end of a truncated axoneme in a ts-che-3 mutant grown at the restrictive temperature, MKS-6 shows fluorescence recovery within seconds. White boxes indicate pixels analysed for the graph, which represents a ratio of fluorescence intensity between the area photobleached (indicated by the red box) to an area not photobleached. Scale bar is 0.5 μm. Download figure Download PowerPoint Interestingly, unlike previous reports which determined that TZ proteins are immobile at the TZ 69, FRAP (fluorescence recovery after photobleaching) analysis shows that ectopic MKS-6 exhibits fast, diffusion-like mobility (Fig EV2C). This suggests that the protein is not correctly assembled/tethered in this distal ciliary region, compared to its normal localisation at the TZ. Importantly, the temperature-sensitive nature of our che-3 mutant allowed us to test for the reversibility of this phenotype. After shifting the ts-che-3 mutant from restrictive temperature (25°C) to permissive temperature (15°C) for 24 h, the accumulation of TZ proteins within the distal ciliary region was lost, and the TZ proteins became largely confined to the correct TZ region (Fig 2B). Next, we wondered whether IFT is required to maintain the proper localisation of TZ proteins, and if turning "off", IFT would result in ectopic TZ protein localisation. To answer this, we shifted the ts-che-3 mutant raised at the permissive (15°C) to restrictive (25°C) temperature for 24 h and assessed NPHP-4, CEP-290 and MKS-6 localisation. At the permissive temperature, these proteins localise correctly to the TZ (Fig 2A). Upon temperature upshift (young adults) for 24 h, a small but significant amount of TZ protein localises ectopically distally within the ciliary axoneme (Fig 2C), with MKS-6 being most visibly affected. Our findings reveal that TZ proteins show ectopic localisation when the animals are raised at restrictive temperature (Fig 2A); this ectopic localisation can be removed by restoring IFT function (Fig 2B). Furthermore, TZ proteins show delocalisation distally when IFT is turned "off" by a temperature shift to 25°C (Fig 2C). Altogether, this provides evidence that IFT is required for retrieving ectopically distributed TZ proteins and maintaining their correct localisation/assembly in the TZ region. Dynein-2 is required for correct function of the transition zone "ciliary gate" Based on our observation that TZ proteins show ectopic localisation in the ts-che-3 mutant, we queried whether the function of the ciliary gate is also compromised. Two different membrane-associated proteins that normally localise at the base of cilia (more specifically the periciliary membrane compartment or PCMC), namely RPI-2 (human RP2 orthologue) and TRAM-1 (TRAM1), were used to test for the integrity of the TZ or "ciliary gate" as previously done 10. At the restrictive temperature, the ts-che-3 mutant shows abnormal ciliary accumulation of both markers (Fig 3A), a phenotype observed in most TZ mutants 101216. The ts-che-3 strain permitted us to test the reversibility of this phenotype. Upon shifting from restrictive to permissive temperature for 24 h, ts-che-3 animals re-established their ability to exclude RPI-2 and TRAM-1 from the ciliary compartment (Fig 3A). Hence, loss of CHE-3 function causes a partial displacement of TZ proteins, which correlates with impaired TZ (ciliary gate) activity. Interestingly, only a small proportion of the TRAM-1 appears to "leak" into the cilium at restrictive temperature, whereas a majority of the RPI-2 accumulates in the cilium when IFT is impaired. This suggests that RPI-2 is targeted to cilia in an IFT-independent manner and might subsequently be removed from cilia by IFT, thus resulting in its final localisation at the ciliary base. Figure 3. Abrogating CHE-3 function causes defects in transition zone gate function and ultrastructure In the ts-che-3 mutant grown at the permissive temperature (15°C), both RPI-2/RP2 and TRAM-1/TRAM1 are localised to the PCMC, just proximal to the ciliary axoneme. At the restrictive temperature, TRAM-1 shows leakage into the axoneme with its strongest localisation at the PCMC. Intriguingly, RPI-2 now displays strong localisation to the axoneme, indicating that in the absence of IFT, RPI-2 may be targeted into the cilium. When ts-che-3 mutant animals grown at the restrictive temperature (25°C) are shifted at young adult to the permissive (15°C) for 24 h, both TRAM-1 and RPI-2 exhibit restoration of their PCMC localisation. Consistent with other TZ markers, both MKS-2 and MKSR-1 show ectopic localisation in the axoneme when grown at the restrictive temperature compared to the permissive temperature. Also consistent with what was observed for other markers, after 24 h at the permissive temperature, animals grown at the restrictive temperature show loss of the ectopic localisation. Fluorescence quantification is shown for each marker at the indicated temperature in the heat maps on the right. Each point in the plot represents one pixel along the centre of the basal body (BB), transition zone (TZ) and middle segment (MS) regions. Dotted areas (three pixels) in the MS were used to quantify ectopic localisation (MKS-2 and MKSR-1) or accumulation (RPI-2 and TRAM-1) for statistical analyses. n = 10 cilia (4–6 animals), Kruskal–Wallis test, *P < 0.05, **P < 0.01; scale bars are 4 μm. Loss of CHE-3 function results in short, bulbous cilia containing electron-dense accumulations and abnormal membrane–microtubule connections. Shown are transmission electron micrograph (TEM) cross sections of an amphid channel cilium in wild-type and che-3(e1124) null mutant animals. Representative images of wild-type cilia show intact middle segment (containing doublet microtubules) and transition zone (with Y-shaped links) (left top and bottom panels) regions. Representative images of che-3(e1124) cilia reveal apparently intact transition zones, with visible Y-link structures, but enlarged ciliary ends filled with electron-dense accumulations, which often appear vesicular. The bulbous structures reveal doublet microtubules associated with the membrane, and occasionally ectopic microtubule-to-membrane connections, which sometimes appear similar to transition zone Y-links in the region just distal to the seemingly "normal" TZ structure. Schematics show longitudinal (left images) and cross section (right images) representations of wild-type and che-3(e1124) cilia. tz, transition zone; pcmc, periciliary membrane compartment; scale bars are as indicated in (nm) for TEM images. Download figure Download PowerPoint Given the improper localisation of TZ proteins in the ts-che-3 mutant, we wondered whether there are ultrastructural defects within the TZ compartment, which could explain our observed ciliary gate phenotypes. Interestingly, both the che-3 null and the ts-che-3 mutant (grown at restrictive temperature) animals appear to have, from Transmission Electron Microscopy (TEM) analysis, all of the ultrastructural features expected of the TZ; namely, the TZ regions feature doublet microtubules connected to a constricted ciliary membrane via Y-links (Figs 3B and EV3) 8. Whether more subtle TZ ultrastructure defects are present cannot be excluded. Based on fluorescent reporters, the largely intact TZ of ts-che-3 animals (Figs 2A and 3A) is consistent with the presence of Y-links at the TZ. However, the ts-

Referência(s)