Dorsoventral patterning of the Drosophila embryo is regulated by graded distribution of bone morphogenetic proteins (BMPs) composed of two ligands, decapentaplegic (Dpp) a BMP2/4 ortholog and screw (Scw) a BMP5/6/7/8 family member. scw(E1) encodes an unusual allele that was isolated as a dominant enhancer of partial loss-of-function mutations in dpp. However, the molecular mechanisms that underlie this genetic interaction remain to be addressed. Here we show that scw(E1) contains a mutation at the furin cleavage site within the prodomain that is crucial for ligand production. Furthermore, our data show that Scw(E1) preferentially forms heterodimers with Dpp rather than homotypic dimers, providing a possible explanation for the dominant negative phenotype of scw(E1) alleles. The unprocessed prodomain of Scw(E1) remains in a complex with the Dpp:Scw heterodimer, and thus could interfere with interaction of the ligand with the extracellular matrix, or the kinetics of processing/secretion of the ligand in vivo. These data reveal novel mechanisms by which post-translational regulation of Scw can modulate Dpp signaling activity.

Genes on the long arm of the Drosophila melanogaster 4th chromosome are difficult to study because the chromosome lacks mitotic and meiotic recombination. Without recombination numerous standard methods of genetic analysis are impossible. Here we report new resources for the 4th. For mitotic recombination we generated a chromosome with an FRT very near the centromere in 101F and a derivative that carries FRT101F with a distal ubiquitously expressed GAL80 transgene. This pair of chromosomes enables both unmarked and MARCM clones. For meiotic recombination we demonstrate that a Bloom syndrome helicase and recombination defective double mutant genotype can create recombinant 4th chromosomes via female meiosis. All strains will be available to the community via the Bloomington Drosophila Stock Center. Additional resources for studies of the 4th are in preparation and will also be made available. The goal of the 4th Chromosome Resource Project is to accelerate the genetic analysis of protein coding genes on the 4th, including the 44 genes with no demonstrated function. Studies of these previously inaccessible but largely conserved genes will close longstanding gaps in our knowledge of metazoan development and physiology.

The bone morphogenetic protein (BMP) signaling network, comprising evolutionary conserved BMP2/4/Decapentaplegic (Dpp) and Chordin/Short gastrulation (Sog), is widely utilized for dorsalventral (DV) patterning during animal development. A similar network is required for posterior crossvein (PCV) formation in the Drosophila pupal wing. Although both transcriptional and post-transcriptional regulation of co-factors in the network gives rise to tissue-specific and species-specific properties, their mechanisms are incompletely understood. In Drosophila, BMP5/6/7/8-type ligands, Screw (Scw) and Glass bottom boat (Gbb), form heterodimers with Dpp for DV patterning and PCV development, respectively. Sequence analysis indicates that the Scw ligand contains two N-glycosylation motifs: one being highly conserved between BMP2/4-and BMP5/6/7/8-type ligands, and the other being Scw ligand specific. Our data reveal that N-glycosylation of the Scw ligand boosts BMP signaling both in cell culture and in the embryo. In contrast, N-glycosylation modifications of Gbb or Scw ligands reduce the consistency of PCV development. These results suggest that tolerance for structural changes of BMP5/ 6/7/8-type ligands is dependent on developmental constraints. Furthermore, gain and loss of N-glycosylation motifs in conserved signaling molecules under evolutionary constraints appear to constitute flexible modules to adapt to developmental processes.