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

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 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 Bradley, P. L. Articles by Andrew, D. J. Search for Related Content PubMed PubMed Citation Articles by Bradley, P. L. Articles by Andrew, D. J. Social Bookmarking                         What's this?

ribbon encodes a novel BTB/POZ protein required for directed cell migration in Drosophila melanogaster

Department of Cell Biology and Anatomy, The Johns Hopkins University School of Medicine, 725 N. Wolfe Street, Baltimore, Maryland 21205-2196, USA

View larger version (26K): [in a new window]   Fig. 1. Multiple signaling pathways are required for tracheal branching. The top half of the figure depicts tracheal placodes (blue) at embryonic stage 10 and the sources of activation (magenta) of four signaling pathways known to function during tracheogenesis (EGFR, FGF, DPP, WG). The lower portion of the figure depicts tracheal metameres at stage 13/14. By this stage, wild-type trachea have formed the five primary branches: dorsal branch (DB), dorsal trunk anterior and posterior (DTa, DTp), visceral branch (VB), lateral trunk (LT), and ganglionic branch (GB). TC, transverse connective. Defects in the components of the pathways are schematized (representative genotypes of the mutants are noted below each metamere). White circles represent absence of appropriately migrating tracheal cells in mutant metameres; blue circles represent the observed positions of the tracheal cells. EGFR signaling is initiated by a localized source of RHO within the placode (magenta; Bier et al., 1990). In the absence of EGFR signaling, many cells fail to invaginate and remain clustered on the surface of the embryo, and cells are missing from every branch (see also Fig. 3G-H'). The BNL/FGF ligand is expressed in patches outside the trachea (magenta) and signals to the BTL/FGFR, which is expressed in the tracheal placode. In the absence of FGF signaling, all branches fail to migrate. The DPP ligand has localized sources dorsal and ventral to the placode (magenta) and signals through the receptors TKV (expressed in the placode) and PUT (expressed ubiquitously; Affolter et al., 1994; Ruberte et al., 1995). kni expression is DPP-dependent in the DB, LT, and GB. In the absence of DPP signaling, kni expression is lost in the DB, LT and GB, and these branches do not form. WG is expressed adjacent to the tracheal placodes (magenta) and signals to the tracheal placode to direct the transcriptional activities of ARM and dTCF. One downstream target of WG signaling is sal. In the absence of WG signaling, sal expression is lost, and DT cells fail to migrate away from the transverse connective (TC). It should be noted that other WNT molecules may be required to activate signaling through ARM/dTCF since arm and dTCF mutant phenotypes are stronger than the wg phenotype alone (Llimargas, 2000). In embryos lacking rib function, no DT is formed and the LT and GB are stunted in their migration. Stages are according to Campos-Ortega and Hartenstein (Campos-Ortega and Hartenstein, 1985).

View larger version (78K): [in a new window]   Fig. 2. rib mutants have defects in tracheal development. Embryos in the left column are wild-type; embryos in the right column are rib1 homozygotes. A-H, K, and L are lateral views; I and J are dorsal views. 2A12 or anti-CRB was used to visualize the tracheal lumen, and anti-TRH to visualize nuclear TRH in all tracheal cells. The tracheal network is abnormal in rib1 mutants (B). The most obvious difference is a complete loss of the main tracheal tube, the DT (arrow in B). The specification of tracheal placodes and early events of tracheal invagination appear normal in rib mutants (D). (E,F) By late stage 12, many of the tracheal branches in rib mutants have not migrated and lumen size is expanded (F), as compared with wild type (E). Insets in E,F,G,H are of metamere 4 (black arrow). (G,H) At stage 14, the DT (white arrow) is clearly absent in rib mutants (H) and the LT (black arrowhead) and GB (arrow) are stunted. The VB (white arrowhead) is visible out of the plane of focus (H). In late stage rib embryos (J), VBs reach the gut and perform terminal branching (arrows), as in wild type (I). (K,L) In embryos carrying UAS-rib and the tracheal driver btl-Gal4, tracheal phenotypes are rescued. (K) The lumina are less dilated, the two ventral branches are less stunted (black arrow and arrowhead), and DTs are migrating (white arrow). (L) Rescue of DT formation is obvious by late stages (arrow). On average, seven of the nine DT fragments form in these rescued embryos. Wild-type genotypes are rib1/CFL, which were also stained with anti-ßgal (brown staining in C,E,G), or Oregon R (A,I). No differences were observed in the phenotypes of embryos carrying one versus two wild-type alleles of rib.

View larger version (73K): [in a new window]   Fig. 3. rib does not function upstream of FGF, DPP or EGFR. All views are lateral. Stage 14 wild-type (A) and rib1 (B) embryos hybridized with an antisense btl RNA probe. rib mutants have similar expression, except that btl RNA expression is prolonged in TC cells (arrowhead; this difference is more obvious at earlier stages; not shown). Stage 11 wild-type (C) and rib1 (D) embryos hybridized with an antisense bnl RNA probe show similar expression patterns. Wild-type embryos were co-hybridized with a btl probe (pink staining in C). Stage 12 wild-type (E) and rib1 (F) embryos stained with anti-KNI have identical expression patterns; KNI is expressed in the DB, LT, and GB, and is lost from the TC and DT cells (arrow). Stage 14 wild-type (G,G') or rhoP38 (H,H') embryos stained with anti-TRH. rho embryos exhibit loss of several DBs (black arrows in H) and have fewer cells in the DT (arrowhead in H), LT (arrowhead in H'), and GB (arrow in H'), when compared with wild type. (Markings in G,G' are identical to H,H', except the view of the wild-type embryo in G is slightly more ventral than in H, placing the DBs out of the plane of focus.) A group of tracheal cells do not invaginate in rho mutants and are found in the same plane of focus as the epidermis (white arrowhead in H'). Early stage 12 wild-type (I) and rhoP38 mutant (J) embryos hybridized with an antisense sal probe show expression in the dorsal cells of tracheal pits (arrow). Embryos were co-hybridized with a salivary gland-specific probe to distinguish mutant from heterozygous embryos (arrowhead in I,J). Wild-type embryo genotypes are as follows: rib1/CFL hybridized with lacZ (A) or stained with anti-ßgal (E), Oregon R (C), and rhoP38/TUL (G,G',I).

View larger version (87K): [in a new window]   Fig. 4. rib functions downstream of, or in parallel to, WG signaling and is independent of sal. All views are lateral. Early stage 12 wild-type (A) and rib1 (B) embryos hybridized with an antisense sal RNA probe show sal accumulation in dorsal tracheal cells (arrow). Wild-type (C) and rib1 mutant (D) embryos carrying the sal-TSE-lacZ reporter construct stained with anti-ß-gal, which detects expression in the dorsal tracheal cells (arrow), and with anti-CRB to visualize the trachea. Slightly lower levels of sal-ß-gal are detected in rib mutants. Anti-CRB staining of a rib1 embryo carrying both UAS-sal and btl-Gal4 transgenes (F) reveals that DT migration (arrow) is not rescued with increased sal expression compared with rib1 alone (E). The DT (arrow) in rib1 mutants fails to migrate (G), whereas the DB (arrowhead), LT, and GB fail to form in tkvA12 mutants (H). Embryos doubly mutant for rib1 and tkvA12 (I,J) form only the TC (arrow in I) and VB (arrow in J).

View larger version (87K): [in a new window]   Fig. 5. rib mutants fail to complete dorsal closure and have ventral cuticle patterning defects. Dark-field images of the lateral cuticle of wild type (A) and rib mutant larvae (B,C) photographed at the same magnification. Of the mutant larvae with scorable cuticles, small anterior dorsal holes and puckering of the dorsal epidermis occur in 36% of rib mutants (B) and large dorsal holes occur in 64% of rib mutants (C; arrows indicate extent of dorsal opening). Lateral views of dpp RNA expression in the leading edge cells of the lateral epidermis (large arrow) and the midgut (arrowhead) of rib/CFL (D) and rib mutant (E) embryos are shown. In rib mutants, dpp staining at the leading edge is more disorganized and in some regions extends into the lateral epidermis (small arrows in E). This apparent increase may be due to an increase in the number of cells expressing dpp or the morphology of these cells and the leading edge at late stages. (F-J) Representative phase contrast images of ventral cuticles of first instar larvae of an allelic series of rib mutations. Anterior (A) is oriented up, and larvae were photographed at the same magnification. There is a prominent narrowing of the lateral extent of denticle belts relative to the ventral surface of the larva, increasing with allele severity. Loss of denticle diversity also increases with the allelic series. The most severe phenotype (rib1/rib1) is shown in J and is equivalent to rib1/Df(2R)P34 (not shown); in such embryos, only a few similarly shaped denticles form.

View larger version (77K): [in a new window]   Fig. 6. rib mutants have defects in salivary gland formation. (A-J) Salivary gland secretory cells (arrowheads in A,B) are visualized with an antibody to dCREB-A. (K-N) Salivary duct cells are visualized with an antibody to DRI. A-F, I, and J are lateral views; G,H,K-N are ventral views. In the formation of the wild-type salivary gland, cells are internalized (A), migrating first dorsally and then redirecting to migrate posteriorly (arrow in C,E) until the dorsal tip reaches the level of the third thoracic segment (T3), and the glands lie along the body wall (G). Initially, rib1 secretory cells invaginate similarly to wild-type (compare B and A); however, once cells reach the dorsal position at which wild-type cells would normally turn to migrate posteriorly, rib1 secretory cells are stalled in their migration (arrow in D,F). In late stage rib mutants, the salivary glands become reoriented, which is likely a secondary effect, and the lumina of the gland become greatly distorted (H). In embryos carrying UAS-rib and the secretory cell-specific driver fkh-Gal4, the posterior migration of secretory cells is restored (I), and secretory cells reach their normal position in the embryo (J). Wild-type salivary ducts are composed of two individual ducts (arrows in K) and a single common duct (arrowhead in K). Salivary ducts in rib mutants fail to complete normal development. Images representing the range of duct defects are shown (L,M). In some embryos, one or two rudimentary tubes are formed, most likely corresponding to the individual ducts (arrows in L), but these never elaborate. In other embryos, no tubes form and the anterior portion of the secretory gland is found in a hole in the DRI-stained duct primordia (arrows in M). In embryos carrying UAS-rib and the secretory cell-specific driver fkh-Gal4, formation of both the common (arrowhead in N) and individual ducts (arrows in N) is significantly restored. Wild-type embryos are rib1/CFL co-stained with anti-ßgal (brown staining in A,C,E) or Oregon R (G,K).

View larger version (12K): [in a new window]   Fig. 7. Genomic region surrounding rib. (A) rib maps to a small interval of the 56C region (gray) by complementation analysis with overlapping deficiencies in the region (Df(2R)P34, Df(2R)F7, Df(2R)GC8, and Df(2R)GC10). coracle (cora) and wbl also map within the rib region, whereas enabled (enb) maps to a distinct interval (Gertler et al., 1995). (B) Two Celera Genomics DNA contigs span the rib region. AE003796 begins distal to the region, ending at nt 268,419, which overlaps with the first 60 nt of AE003797; AE003797 continues proximally, ending outside of the region. Breakpoints for Df(2R)GC10 and Df(2R)P34 are shown as hatched boxes. 1, 2, 3, 4, 5 and 6 are predicted genes in the region. Genes are depicted as arrows, which indicate the direction and approximate size of the transcription unit. gene 6 maps completely outside of deficiency Df(2R)P34. EP(2)2445 is the viable P-element insertion line used to generate the lethal line EP(2)24451, which deletes DNA from enb to cora and does not affect rib function, leaving only three candidates. Since wbl complements rib (Table 1), rib is gene 4 or gene 5. Proximal (P) and distal (D) is relative to the centromere.

View larger version (97K): [in a new window]   Fig. 8. rib mRNA is expressed throughout embryonic development. Whole-mount in situ hybridization to wild-type embryos with an antisense rib RNA probe. Images with primed letters are ventral views of the same embryo; all others are lateral views. Embryonic stages are indicated in the lower left corner. rib expression is first detected in the termini (A,B), but is absent from pole cells (arrowhead in A). Segmental stripes appear in the epidermis (arrowheads in B,C,C',E). Expression of rib RNA in the salivary gland primordia is evident by stage 10 (arrow in D), and is expressed throughout invagination (arrows in E,E',F'). The Malpighian tubules also express rib (arrowhead in G). By later stages, rib RNA is detected in most cells of the epidermis, including cells of the lateral epidermis. Of note is the lack of elevated expression in the later central nervous system and midgut, two tissues whose formation is abnormal in rib mutants (Jack and Myette, 1997), perhaps indicating that these defects are indirect. Alternatively, earlier expression or a lower level of rib is required in these tissues.

View larger version (50K): [in a new window]   Fig. 9. (A) Gene structure of rib. The salient features of the longest cDNA (3959 nt; BDGP LD16058) are depicted, and the position of the single intron is noted. A consensus polyadenylation signal (AATAAA) is present near the end of the 3' UTR. By northern analysis, we detected a single transcript of 4.3 kb (data not shown), which correlates well with cDNA length and suggests the cDNA is nearly full length. (B) Conceptual translation of the 1963 nt ORF yields a 661-residue protein. Translation of the corresponding GadFly gene CG7230 reveals an identical protein sequence. An N-terminal BTB/POZ domain is denoted in bold. There are four consensus NLSs (underlined). A predicted coiled-coil region in the C terminus is double underlined. RIB contains seven MAPK consensus phosphorylation sites (PX1-2S/TP, where X is any amino acid), which are boxed. The mutations in rib1 and rib2 alleles are double-boxed: rib1 encodes a stop codon after residue 282, and rib2 has an arginine to histidine substitution at residue 58 (R58H). (C) Sequence alignment of the RIB BTB/POZ domain with BTB/POZ domains of Drosophila Longitudinals lacking (LOLA), Drosophila Bric a brac (BAB), mouse Zinc finger protein 161 (mZfp161), and human Promyelocytic leukemia zinc finger (hPLZF). The eleven N-terminal residues (under the bar) are not included in the BTB/POZ domain as defined by the InterPro program; however, this short stretch is highly conserved in these three and other Drosophila BTB/POZ domain proteins. Percentage identity and similarity of the BTB/POZ domains with respect to RIB are noted. Residues that were examined in a structure/function analysis of hPLZF (Melnick et al., 2000) are indicated by an open circle, and residues that, when mutated, disrupt BTB/POZ domain function are also indicated by a filled circle. Note that the rib2 allele has a change in one of these essential residues (*). Black shading, white letters denotes identical residues; dark gray shading, white letters denotes conserved residues; grey shading denotes similar residues.

CiteULike

Complore

Connotea

Del.icio.us

Digg

Reddit

Technorati

Twitter