Transcriptional profiling of C. elegans DAF-19 uncovers a ciliary base-associated protein and a CDK/CCRK/LF2p-related kinase required for intraflagellar transport

Abstract

Cilia are ubiquitous cell surface projections that mediate various sensory- and motility-based processes and are implicated in a growing number of multi-organ genetic disorders termed ciliopathies. To identify new components required for cilium biogenesis and function, we sought to further define and validate the transcriptional targets of DAF-19, the ciliogenic C. elegans RFX transcription factor. Transcriptional profiling of daf-19 mutants (which do not form cilia) and wild-type animals was performed using embryos staged to when the cell types developing cilia in the worm, the ciliated sensory neurons (CSNs), still differentiate. Comparisons between the two populations revealed 881 differentially regulated genes with greater than a 1.5-fold increase or decrease in expression. A subset of these was confirmed by quantitative RT-PCR. Transgenic worms expressing transcriptional GFP fusions revealed CSN-specific expression patterns for 11 of 14 candidate genes. We show that two uncharacterized candidate genes, termed dyf-17 and dyf-18 because their corresponding mutants display dye-filling (Dyf) defects, are important for ciliogenesis. DYF-17 localizes at the base of cilia and is specifically required for building the distal segment of sensory cilia. DYF-18 is an evolutionarily conserved CDK7/CCRK/LF2p-related serine/threonine kinase that is necessary for the proper function of intraflagellar transport, a process critical for cilium biogenesis. Together, our microarray study identifies targets of the evolutionarily conserved RFX transcription factor, DAF-19, providing a rich dataset from which to uncover-in addition to DYF-17 and DYF-18-cellular components important for cilium formation and function.

Figure 1

Gene expression analyses using transcriptional GFP reporter fusions of select ciliary candidate genes revealed by microarrays and quantitative real-time PCR. Transgenic worms at the 3-fold embryonic stage (A) and the L1 stage (B–K) are shown: (A–H, K) head region, (I–J) whole worms. (A) C31C9.1 is expressed in amphid and labial neurons. (B) F28A12.3 is expressed exclusively in labial neurons. (C–D) F35D2.4 and T26A8.2 are expressed in amphid and labial neurons. (E) ZK418.3 is broadly expressed in head neurons, including both amphid and labial neurons. (F–H) F55A4.3, Y53G8AM.4, and F10E9.1 are expressed in amphid sensory neurons. (I) C42C11.7 has a combined amphid neuronal and hypodermal expression pattern (J–K) M01A8.2 and F58E6.11 are candidate genes lacking neuronal localization patterns. (L) A schematic diagram depicting the head (left) and tail (right) ciliated sensory neurons (CSNs).

Figure 2

Characterization of the gene dyf-17. (A) Schematic diagram of the dyf-17 (Y39B6A.11) gene structure, which resides on linkage group V (LG V) in the C. elegans genome. Exons (boxes) and the X-box promoter motif (X) relative to the translational start site are indicated. The Mos1 transposon insertion allele ox175 in the third exon is indicated (open arrowhead). Sequencing of the Y39B6A.11 gene in an ox175 mutant background revealed that a Mos1 transposon had inserted in the third exon of dyf-17 (YAC Y39B6A-nt 170512 / 170513). (B, E) Transcriptional reporter analyses of dyf-17 promoter::GFP in daf-19 (+) and daf-19 (−) backgrounds. (C, F) Translational reporter analyses of dyf-17 (promoter + gene) fused to GFP showing subcellular localization in amphid (C, head outlined) and phasmid (F, tail outlined) CSNs. Cutout boxes contain 2× magnified regions highlighting the ciliary base localization patterns in both amphid and phasmid neurons. (D, G) Transcriptional reporter fusion of bbs-7::GFP showing the length of cilia in both a wild type (full-length) and dyf-17 (truncated) phasmid cilium. The ciliary base is demarcated (asterisk). (H–J) Dye filling assays of wild type (H), dyf-17 mutant (Dyf −) animals (I) and rescued (Dyf +) animals (J). Worm heads are outlined (dashed lines) and amphid neuronal cell bodies are demarcated (solid arrow heads). (K) Osmotic avoidance (Osm) assay of wild type (positive control), che- 13 (negative control), dyf-17 (mutant) and dyf-17 rescued worms. (*) Student’s t-tests: n = 100 for all genotypes; p = 1.04 E -13 for wild type versus dyf-17; p = 3.1 E -13 for dyf-17 mutant versus dyf-17 rescue. (L) Cilia length measurements using bbs-7::GFP (cf. D, G). (*) Student’s t-tests: n = 50 for both genotypes; p = 6.8 E -66 for wild type versus dyf-17.

Figure 3

DYF-18 is an evolutionarily conserved CDK/CCRK/LF2p-related Ser/Thr kinase required for proper cilium-dependent fluorescent dye filling and chemosensation. (A, B) Fluorescence images of the head region of wild type and dyf-18 (ok200) mutant worms, respectively, taken at the same exposure time after a dye filling assay. dyf-18 mutant animals show weak dye-filling defects in amphid neurons (not shown for phasmid neurons). (C) Quantification of the fluorescent dye filling defects, showing a statistically significant decrease in the dye uptake in a dyf-18 mutant strain as compared to wild type. (D) dyf-18 mutants display an osmotic avoidance defective (Osm) phenotype that is rescued by introduction of a DYF-18::GFP translational fusion construct. Wild type serves as a positive control, osm-5 mutants serve as a negative control. (C, D) n denotes the number of animals tested. (*) Student’s t-tests were performed for statistical significance (p values are shown). (E, F) Expression pattern and localization of the DYF-18 protein in head and tail neurons, respectively. A DYF-18::GFP translational fusion construct is expressed in amphid, phasmid and labial neurons in wild-type animals. The encoded protein localizes diffusely within the cell bodies, dendrites, and ciliary compartments. Cell bodies, dendrites and ciliary regions are indicated.

Figure 4

DYF-18 regulates the localization of intraflagellar transport (IFT) machinery components. Both amphid and phasmid cilia are shown. (A, B) Localization of IFT subcomplex B proteins. CHE-2 shows wild-type localization whereas OSM-5 is largely excluded from cilia and accumulates at the base of cilia. (C, D) Localization of IFT motors. KAP-1 shows essentially wild-type localization with occasional accumulations at the tip of the middle segment (marked by *) whereas the OSM-3 motor is largely unable to enter the distal segment and shows large accumulations (marked by *) between the middle and the distal segment. (E) Shows the essentially normal localization of the IFT subcomplex A protein CHE-11 to the middle and distal ciliary segments. In all panels the dotted line divides the middle and the distal segments. BB, degenerate basal bodies; *, marks protein accumulation. Presence or absence of IFT in the mutant has been denoted.