Figure 1.
LTM muscle fibers 21, 22 and 23 frequently overshoot their attachment sites in Wnt5, drl and dnt mutant embryos.
Stage 16 embryo body wall muscle preparations stained with anti-Muscle Myosin are shown for the wild type control (w1118) (A), Wnt5400 (B), drlRed2 (C), Drl-2E124 (D), dnt42.3 (E) and Df(2L)Exel6043 (F). Two hemisegments are displayed for each genotype with one set of muscles 21–23 labelled. In Wnt5, drl and dnt mutants, LTMs frequently bypass their normal attachment at the epidermis at muscle 12 and instead extend ventrally beyond muscle 13 and attach at a novel epidermal site located close to muscle fiber 7. Df(2L)Exel6043 mutant embryos, that lack both DNT and DRL, display this phenotype in all hemisegments of the homozygous animals. The penetrance of these phenotypes is shown in Table 1. The muscle bypass phenotype is schematically shown in panel (G). The * indicates the location of the novel, ectopic epidermal attachment in panels (B), (C), (E), (F) and (G). Anterior is up and ventral is left.
Table 1.
LTM muscle bypass phenotype in Wnt5, drl and dnt mutants and restoration of attachment when Wnt5 is present in tendon cells or muscle.
Figure 2.
Generation of a dnt mutant by imprecise excision of an adjacent P-element.
The P{EP}dntEP(2)2158 insert, situated 350 bp upstream of the dnt ATG initiator codon, was mobilized by providing a source of transposase and imprecise excisions were selected for by loss of the w+ marker in the P-element insert and molecularly characterized by sequencing cloned genomic PCR products (Materials and Methods). The locations of the insertion, the extent of the deletion in the dnt42.3 allele and gene landmarks, e.g., exons and the location of the WIF encoding segments, are shown in (A). The dnt42.3 allele displays dramatically decreased expression of dnt mRNA. Stage 11 wild type (B) and contemporaneously processed dnt42.3 mutant (C) embryos hybridized with a dnt antisense probe are shown. Anterior is to the left and dorsal is up in panels (B) and (C).
Figure 3.
Muscle attachment defects persist from the embryonic to larval stages in Wnt5 and drl mutants.
Third instar larval body walls of w1118 (A), Wnt5400 (B) and drlRed2 (C) mutant larvae are stained with anti-FAS2 (mAb 1D4). Wnt5400 larvae and drlRed2 larvae frequently bypass their normal attachment sites and extend ventrally where they form new stable attachments. The original and ectopic tendons cells are indicated by + and *, respectively. FAS2 protein is evident at both sites. The penetrance of the bypass phenotypes is indicated in Table 1. Anterior is up and ventral is left.
Figure 4.
Wnt5 protein and mRNA expression domains in epidermis, muscle and tendon cells during embryonic development.
WNT5 is predominantly expressed in subsets of neurons in the CNS from stage 12 onwards throughout embryonic development (data not shown; [28]). However, there is also strong expression from this stage onwards in the epidermis and the musculature. At stage 12, Wnt5 protein (A, B) and Wnt5 mRNA (E, F) expression is observed in the epidermis, most prominently in two clusters (arrows), and throughout the somatic mesoderm that will give rise to the body wall musculature. Later in embryonic development at early stage 16 WNT5 protein and Wnt5 mRNA are present in the attachment sites (arrows in panels C and G) and at low levels in most muscle fibers including the LTMs 21, 22 and 23 (C, G). At the end of embryonic development at late stage 17, Wnt5 protein (D) and Wnt5 mRNA (H) are almost undetectable in the somatic mesoderm. In all panels anterior is up and ventral is left.
Figure 5.
The WNT5 and SR epidermal expression domains partially overlap.
Wild type embryos of stage 13 were double labeled with anti-SR and anti-WNT5 antibodies and visualized by confocal microscopy. SR protein (red) is present in a number of tendon precursor cells (A). The WNT5 epidermal expression domains (green) in these same embryos are shown in (B). The overlay of these panels is shown in (C). The SR expression domains partially overlap with the larger WNT5 domains. Anterior is up and ventral is left.
Figure 6.
The new attachment sites of the bypassed muscle fibers in Wnt5 and drl mutants frequently do not express SR, while the bypassed attachment sites do.
Double labeled stage 16 embryos are shown of w1118 (A), Wnt5400 (B) and drlRed2 (C) with anti-Muscle Myosin in green and anti-SR in red (Material and Methods). Asterisks mark the novel attachment sites of the overshooting LTM muscles; white arrowheads mark the locations of the original attachment sites. In Wnt5 mutants the novel target sites do not express SR in 65% of the segments containing overshooting muscles, while the bypassed attachment sites usually express SR. The SR positive, original tendon cell is also present in drlRed2 mutants, but is partly masked by the overshooting muscle fiber in panel (C), but clearly visible in panel (F)). These results were confirmed in embryos that express Tau-MYC under the control of a stripe promoter in both Wnt5 and drl mutants (data not shown). The following genotypes are shown, the control UAS-Tau-MYC; sr-GAL4 embryos (D), Wnt5400; UAS-Tau-MYC/sr-GAL4 (E) and drlRed2; UAS-Tau-MYC/sr-Gal4 (F). Anti-Muscle Myosin is shown in green and anti-MYC in red. No MYC protein is observed in the ectopic attachment sites. The photographs in Panels (A–C) were taken on a compound microscope and those in Panels (D–F) on a confocal microscope. Anterior is up and ventral is left.
Figure 7.
The new attachment sites of the bypassed muscle fibers in Wnt5 and drl mutants express βPS integrin.
Wild type (A), Wnt5400 (B) and drlred2 (C) embryos were labelled with anti-βPS Integrin. Muscles 21–23 do exhibit an accumulation of βPS Integrin protein at the tip of the overshooting fibers (white asterix).