Nat Cell Biol
October 1, 2009;
The planar cell polarity effector Fuz is essential for targeted membrane trafficking, ciliogenesis and mouse embryonic development.
The planar cell polarity (PCP) signalling pathway is essential for embryonic development because it governs diverse cellular behaviours, and ''core PCP'' proteins, such as Dishevelled
and Frizzled, have been extensively characterized. By contrast, the ''PCP effector'' proteins, such as Intu
, remain largely unstudied. These proteins are essential for PCP signalling, but they have never been investigated in mammals and their cell biological activities remain entirely unknown. We report here that Fuz
mutant mice show neural tube
defects, skeletal dysmorphologies and Hedgehog
signalling defects stemming from disrupted ciliogenesis. Using bioinformatics and imaging of an in vivo mucociliary epithelium
, we established a central role for Fuz
in membrane trafficking, showing that Fuz
is essential for trafficking of cargo to basal bodies and to the apical tips of cilia
is also essential for exocytosis in secretory cells. Finally, we identified a Rab-related small GTPase as a Fuz
interaction partner that is also essential for ciliogenesis and secretion. These results are significant because they provide new insights into the mechanisms by which developmental regulatory systems such as PCP signalling interface with fundamental cellular systems such as the vesicle trafficking machinery.
Nat Cell Biol
[+] show captions
Figure 2. RSG1 controls ciliogenesis and secretion. (a) SEM of intact, control Xenopus ciliated epidermis reveals multi-ciliated cells and surrounding mucous secreting cells (b) RSG1 morphants display defects in ciliogenesis and absence of mucous granules and exocytic pits. (c) Higher-magnification view of RSG1 morphant ciliated epidermis displaying diminished cilia numbers and lengths and a decrease in exocytic pits in neighboring secretory cells. (d) Fuz morphants also display diminished cilia numbers and lengths and a decrease in exocytic pits in secretory cells. (e, f) Epidermal targeted over-expression of RSG1T65N, but not wild type RSG1, results in defects in ciliogenesis as well as decreases in mucous granules and exocytic pits in secretory cells. (g, h) GFP-RSG1 (low-level expression) localizes to the basal body region of multi-ciliated cells, whereas GFP-RSG1T65N (low-level expression) in multi-ciliated cells is diffuse and not tightly associated with basal bodies. Observations of fluorescence levels following expression of GFP-RSG1 and GFP-RSG1T65N mRNA were comparable, suggesting similar rates of translation for the two proteins.
Figure 3. Homology modeling, network predictions, and experimental validation suggest a trafficking function for Fuz. (a, b) Rendered protein models (Open-Source PyMOL 0.99rc6 software). (a) Sedl N-terminal domain (pdb:1H3Q). (b) 1H3Q based homology threaded model of the C- terminal -aa(s) 287–419 - of Xenopus Fuz protein. (c) Illustration of experimentally-derived protein-protein interactions (see Supplemental Methods) linking Dvl2 with longin-domain proteins AP1s, AP2s, and SedlP as well as CLAMP. (d) Mosaic imaging of live embryo expressing CLAMP-GFP, which localizes to ciliary axonemes and apical tips. A nucRFP marks the nuclei, serving as a lineage tracer for co-injected Fuz MO. The confocal slice reveals a loss of apical localized CLAMP-GFP at ciliary tips in FUZMO cells (inset shows merge of (e) and a more basal slice to display nucRFP signal). Scale bar in (e) = 10uM.
Figure 4. Fuzzy and RSG1 control trafficking to basal bodies as well as to the tips of cilia. (a, b) Fuz knockdown disrupts localization of CLAMP-GFP to the ciliary rootlet. Confocal stacks of formaldehyde fixed Xenopus epidermis expressing centrin-RFP and CLAMP-GFP mRNAs. (a) Multi-ciliated cell in x-y view from an uninjected control embryo exhibits elongated CLAMP-GFP signal extending from relatively evenly spaced basal bodies (centrin-RFP). (a′) Higher magnification view of the x-y section from [a]. (a″) X-Z projection of the stack shown in [a′] displays apical co-localization of centrin-RFP and CLAMP-GFP. (b) Ciliated cell in an x-y view of the apical surface in a Fuz morphant reveals defects in the spacing of centrin-RFP signal and defects in elongated CLAMP-GFP signal. Additionally CLAMP-GFP signal is not faithfully co-localized with centrin-RFP signal in many cases. (b′) Higher magnification view of the x-y section from [b]. (b″) X-Z-projection of the stack shown in [b′] reveals apical alignment of the centrin-RFP signal however in many cases the CLAMP-GFP signal is below the apical surface in large punctae. (c–e) Mosaic imaging of live agarose embedded embryo. CLAMP-GFP highlights a variety of epidermal structures. RFP-Histone 2B (nuc-RFP) serves as a lineage tracer for morpholino-injected cells. (c, d) 3D projections (x-z). RSG1 MO (+ nucRFP cells) display defects in normal CLAMP-GFP localization along the apical surface. (e) Confocal slice (x-y) exhibiting a loss of apical localized CLAMP-GFP to ciliary tips in RSG1 MO cells (inset shows merge of (e) and more basal slice to display nucRFP signal). Scale bars in (c, d) are 3uM; scale bar in (e) is 10μm.
Figure 5. Knockdown of Fuz disrupts exocytosis in mucus-secreting cells. (a) Wild type multi-ciliated cell (left) flanked by secretory cells in Xenopus mucociliary epidermis. Control cells have an average of over 90 open exocytic pits per cell. (b) Fuz morphants display defects in ciliogenesis in multi-ciliated cell (left) and failure of exocytosis in mucus-secreting cells (note the absence of exocytic pits indicated by white arrowheads in [a]). Green arrowheads indicate apical membrane blebs (see also panel F, below). In a representative experiment, Fuz morphant cells had fewer than 5 exocytc pits per cell on average (difference from control is significant by the Mann-Whitney U-test; p<0.0001). Scale bars in A, B = 5μm (c) Mosaic epidermal tissue, with morphant cells outlined in red. Scale bar = 10μm (d) Confocal section (x-y) of mosaic embryo in which membrane-RFP (memRFP, red) mRNA was co-injected with Fuz morpholino and processed for Xeel (interlectin-2) antibody (green). Cell expressing a high level memRFP (right) lacks apical Xeel antibody staining comparable to the neighboring cell (right). (d′) Confocal projection (x-z) of [d] illustrates the loss of apical Xeel staining (green) in the Fuz morphant cell correlated with apical memRFP expression (right). Scale bars in D, d′ = 3μm (e) TEM of section of control Xenopus epidermis shows empty and mucus granule-containing vesicles docked at the apical surface, whereas Fuz morphants (f) display a defect of vesicle fusion with the plasma membrane release, as illustrated by large membrane protrusions (green arrows in [b, f]). Additionally, frequent homotypic vesicle fusion events were observed in secretory cells of Fuz morphants (red arrow). Scale bar in E = 500nm; Scale bar in F= 100nm.