QUICK SEARCH:   [advanced] Author: Keyword(s):
Home     Help     Feedback     Subscriptions     Archive     Search     Table of Contents

First published online December 8, 2005
doi: 10.1242/10.1242/dev.02177

This Article Summary Full Text Full Text (PDF) Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Related articles in Development Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Moussian, B. Articles by Uv, A. E. Search for Related Content PubMed PubMed Citation Articles by Moussian, B. Articles by Uv, A. E. Social Bookmarking                         What's this?

Drosophila Knickkopf and Retroactive are needed for epithelial tube growth and cuticle differentiation through their specific requirement for chitin filament organization

Bernard Moussian^1,*,, Erika Tång^2,*, Anna Tonning², Sigrun Helms¹, Heinz Schwarz¹, Christiane Nüsslein-Volhard¹ and Anne E. Uv^2,

¹ Department of Genetics, Max-Planck-Institute for Developmental Biology, Tübingen, Germany
² Department of Medical Biochemistry, Gothenburg University, Gothenburg, Sweden.

View larger version (106K): [in a new window]   Fig. 1. knk and rtv are required for uniform tracheal tube expansion. The developing tracheal lumen of wild-type, knk and rtv mutant embryos was visualized with the lumen-specific antibody 2A12. (A-C) Stage 16 knk7A69 (B) and rtv11 (C) mutant embryos display irregular tracheal tube shapes, but no defects in branch patterning, compared with wild-type embryos (A). The box in A depicts the dorsal trunk (DT) region in D-L. (D-F) At stage 14, the 2A12 antigen begins to accumulate in the wild-type DT lumen (D), but is reduced in the lumen of knk7A69 (E) and rtv11 (F) mutants. (G-I) During stage 15, the wild-type DT lumen (G) expands uniformly, whereas the DT lumen of knk7A69 (H) and rtv11 (I) mutants remain constricted at branch fusions (arrows), and the tube between fusion junctions becomes excessively overgrown. (J-L) During stage 16, the knk7A69 (K) and rtv11 (L) mutant DTs in addition become extensively elongated compared with the wild-type trunk (J). (M-O) The lumen of the narrower multicellular ganglionic branches (GB) is discontinuous in knk7A69 (N) and rtv11 (O) mutants at the border of the ventral nerve cord (arrowheads), compared with wild type GB (M). (P) Illustration of tracheal cell shape changes during DT expansion at stage 15. The DT lumen is encircled by three to five cuboidal cells (pale grey) attached to each other by intercellular junctions, apart from the branch fusion lumens, which are surrounded by two toroidal cells (dark grey). Lumen expansion from stage 14 to 15 does not involve cell division and relies on coordinated growth of the apical cell surfaces. (Q) The GB branches are made by rows of single cells folding over themselves, and the arrowhead points to where lumen discontinuities are observed in knk7A69 and rtv11 mutants. Different shades of grey are used to distinguish neighbouring cells in the row. Scale bars: 25 µm. DT, dorsal trunk.

View larger version (90K): [in a new window]   Fig. 2. knk and rtv functions in the chitin-mediated pathway of tube-size control. (A-F) Mutations in knk and rtv do not enhance the tracheal phenotypes of chitin-deficient embryos. Wild-type embryos laid by Nikkomycin-fed parents (1 mg/ml) (D) develop full kkv tracheal phenotype with local dilations and cysts, compared with wild type (A). This phenotype is similar to that of knk7A69 mutants (B) and to knk7A69 mutant embryos laid by Nikkomycin-fed parents (E). Also chitin-deficient rtv11 mutant embryos laid by Nikkomycin-fed parents (F) develop tracheal phenotypes indistinguishable from that of chitin-deficient embryos, but the untreated rtv11 mutant embryos (C) display less severe tube dilations. (G,H) When wild-type flies are fed with a low Nikkomycin dose (0.5 mg/ml) (G), their embryonic offspring display tracheal phenotypes reminiscent of the rtv11 mutant phenotypes (H). Scale bars: 15 µm in A-F; 20 µm in G,H.

View larger version (55K): [in a new window]   Fig. 3. knk and rtv are required for tracheal luminal chitin filament assembly. (A-E) Early stage 15 embryos double labelled with luminal antibody 2A12 (red; top panel) and FITC-conjugated chitin-binding probe (CBP; green, bottom panel), where analogous DT segments from each genotype are shown. All embryos were fixed and labelled in parallel and the confocal images taken in the same session with identical settings to display comparative levels of luminal chitin. (In the middle merged panel, colour levels are adjusted to enable visualization of both 2A12 and CBP labelling.) (A) In wild-type embryos, CBP labels a luminal chitinous fibre with `threads' running parallel to tube length. The merged image shows that the CBP-labelled fibre is confined to only a part of the lumen, leaving narrow gaps to the surrounding epithelium. (B) The trachea of wild-type embryos laid by Nikkomycin-fed parentsdisplay kkv phenotype and barely detectable luminal CBP levels. (C,D) In both knk7A69 (C) and rtv11 (D) mutant trachea, CBP labelling reveals a broad chitinous matrix, which fills the entire lumen. (E) The intensity of this CBP labelling is weaker than in the wild type, but stronger than in embryos with reduced CS-1 activity upon treatment with a low Nikkomycin dose (0.5 mg/ml) to reproduce the kkv and rtv mutant phenotypes. Scale bars: 5 µm in A-E.

View larger version (190K): [in a new window]   Fig. 4. Chitin is present in knk and rtv mutant procuticles, but lamellogenesis is defective. (A,B) Comparison of cuticle preparations of wild-type (A) and knk7A69/knk5C77 (B) larvae shows that the knk cuticle is dilated and the knk head skeleton is strongly deformed and melanized (insets in A,B). (C,D) TEM analysis of cross-sections of wild-type (C) and knk5C77 mutant (D) larval cuticles reveals a defected epicuticle and procuticle in knk mutant larvae, but a normal envelope. The knk mutant procuticle is not clearly separable from the upper epicuticle (epi/pro), and inclusions of electron-dense material occupy the relicts of the procuticle. The knk procuticle is also devoid of chitin lamellar texture (arrow in inset in C, compare with inset in D). (E-H) Detection of chitin by gold-conjugated WGA (seen as black spots in the procuticle) reveals the presence of chitin in the wild-type procuticle (E), knk5C77 (F) and rtv11 (H) mutant procuticles, and not in kkv14D79 mutant cuticle (G). Insets in G and H illustrate fully contrasted magnifications of rtv and kkv cuticles to allow comparison with wild-type and knk cuticles in the insets in C and D. Scale bars: 0.5 µm; 0.25 µm in insets. env, envelope; epi, epicuticle; pro, procuticle.

View larger version (65K): [in a new window]   Fig. 5. Knk is an apical GPI-anchored protein expressed in the developing trachea and epidermis. (A) The wild-type Knk protein (top) is predicted to contain an N-terminal signal peptide (green), two tandem DM13 domains (blue), one DOMON (red) domain and a C-terminal GPI-anchor (black). Seven knk alleles, which all cause the same phenotypes, harbour premature stop codons to produce truncated Knk proteins of various lengths (illustrated below the wild Knk protein). The closest relatives to Knk in plant and worms are represented by Arabidopsis thaliana (Accession number BAB08762), with four transmembrane domains (TM, black) and a cytochrome BS61 domain (yellow), and Caenorhabditis elegans (Accession number AAG49388). (B) Western blot analysis of stage 17 embryonic extracts show that Knk protein is present in both the membrane (pel) and the soluble fraction (sup), compared with the transmembrane Syntaxin1A protein, which only precipitates with the membrane fraction. (C) Incubation of the membrane fraction in (B; pel) with Phospholipase C releases some Knk into the membrane-free supernatant (sup), indicating that Knk is a GPI-anchored protein. By contrast, Tout-velu (Ttv), which has a C-terminal type 1b transmembrane domain, is resistant to Phospholipase C treatment. (D-F) In-situ hybridization with knk anti-sense RNA probes detects knk transcripts in the developing trachea from stage 13 (arrows in D-F). From stage 15 the Knk transcript is also detected in the pharynx, hindgut and epidermis (D). (G-L) Co-labelling with anti-Knk and anti-Fas3 reveals apical Knk localization in the trachea and epidermis. In stage 15 wild type tracheal epithelium anti-Knk (G and J; red) highlights the apical cell surface and parts of the apico-lateral surface as seen from the slight overlap with Fas3 (J; green). Knk-labelling is absent in knk mutant trachea (H,K). Stage 16 wild-type epidermis (I) also displays apical Knk labelling (green) compared with the lateral Fas3 (red), which is absent in the knk mutant epidermis (L). (M-O) Stage 15 embryos labelled with the 2A12 antibody shows that the tracheal phenotype of knk mutants (M) is rescued by ectopic expression of UAS-knk (N) and UAS-knkTM (O) driven with Btl-GAL4. Scale bars: 50 µm in D; 25 µm in E,F,M-O; 7 µm in G-L.

View larger version (112K): [in a new window]   Fig. 6. Analysis of epithelial polarity and SJ integrity in knk mutants. (A-D) The knk mutant tracheal epithelium displays normal levels and distribution of the two SJ proteins Fas3 (B) and Dlg1 (D), compared with the wild type (A,C). (E,F) TEM analysis of wild-type (E) and knk mutant (F) epidermis reveals a normal `ladder'-like SJ structure (arrows) in the mutants. (G,H) The apical marker Crumbs localizes normally along the apical cell surface in knk mutants (H), compared with the wild type (G). (I,L) Embryos homozygous for loss-of-function alleles of Fas2 and knk display additive reduced luminal 2A12 levels. Early stage 15 wild-type (I), Fas2 (J), knk (K) and Fas2;knk (L) mutant embryos labelled with antibody 2A12. The images were obtained with identical confocal settings, and show that luminal 2A12 levels are reduced in Fas2 mutants, reduced and unevenly distributed in knk mutants and absent in Fas2;knk7A69 double mutants. The cellular 2A12 levels are, however, comparable. (M,N) The luminal Pio protein appears unaffected in knk mutant DT (N), compared with wild type (M), as shown by co-labelling for Pio (green) and 2A12 (red). Scale bars: 5 µm in A-D; 10 nm in E,F; 7 µm in G-N.

View larger version (100K): [in a new window]   Fig. 7. Levels and localization of Knk in tube-expansion mutants. (A) A western blot with embryonic extract from stage 17 wild type and knk, grh and rtv mutants labelled with anti-Knk reveals similar Knk levels in embryos of all genotypes (top row). Anti-Tubulin was used for loading control (bottom row). (B-E) Anti-Knk labelling of stage 15 wild-type (B), kkv (C), Fas2 (D) and sinu (E) mutant DTs shows that Knk localizes to the apical surface in wild type and kkv mutants, but occupies the entire cell surface in Fas2 and sinu mutants. (F-H) Anti-Knk also labels the apical surface in wild-type stage 16 DT (F), but in Fas2 (G) and mega (H) mutants Knk is detected along the lateral and basal cell surface (arrowhead). (I-L) UAS-knkTM-expression, driven with the tracheal Btl-GAL4 line, rescues the knk (L), but not the Fas2 (K) mutant tracheal phenotypes, as analysed by co-labelling with anti-Knk (red) and CBP (green). In Fas2 mutants (J) luminal chitin appears amorphous and expanded compared with wild type (I), and UAS-knkTM-expression in Fas2 mutants (K) does not rescue chitin filament organization nor tube-size defects. (M) A model illustrating the requirement of different tube expansion genes for uniform lumen diameter expansion. Chitin chains (green) in the expanding lumen assemble into a filamentous cable in a process that requires the apical surface proteins Knk and Rtv, and perhaps additional components (left figure). SJ components are needed for the correct localization of Knk and possibly other proteins involved in chitin filament assembly. In knk and rtv mutants, chitin fails to assemble into a filamentous cable (right figure) and loses its function in tracheal tube-size regulation. Loss of SJ components also leads to defects in chitin matrix assembly, but not to the severe fusion branch constrictions seen in knk and rtv mutants. Scale bars: 4 µm in B-E,I-L; 5 µm in F-H.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter