Drosophila syncytial nuclear divisions limit transcription unit size of early zygotic genes. As mitosis inhibits not only transcription, but also pre-mRNA splicing, we reasoned that constraints on splicing were likely to exist in the early embryo, being splicing avoidance a possible explanation why most early zygotic genes are intronless. We isolated two mutant alleles for a subunit of the NTC/Prp19 complexes, which specifically impaired pre-mRNA splicing of early zygotic but not maternally encoded transcripts. We hypothesized that the requirements for pre-mRNA splicing efficiency were likely to vary during development. Ectopic maternal expression of an early zygotic pre-mRNA was sufficient to suppress its splicing defects in the mutant background. Furthermore, a small early zygotic transcript with multiple introns was poorly spliced in wild-type embryos. Our findings demonstrate for the first time the existence of a developmental pre-requisite for highly efficient splicing during Drosophila early embryonic development and suggest in highly proliferative tissues a need for coordination between cell cycle and gene architecture to ensure correct gene expression and avoid abnormally processed transcripts.DOI: http://dx.doi.org/10.7554/eLife.02181.001

BackgroundPostembryonic development, including metamorphosis, of many animals is under control of hormones. In Drosophila and other insects these developmental transitions are regulated by the coordinate action of two principal hormones, the steroid ecdysone and the sesquiterpenoid juvenile hormone (JH). While the mode of ecdysone action is relatively well understood, the molecular mode of JH action remains elusive.Methodology/Principal FindingsTo gain more insights into the molecular mechanism of JH action, we have tested the biological activity of 86 structurally diverse JH agonists in Drosophila melanogaster. The results were evaluated using 3D QSAR analyses involving CoMFA and CoMSIA procedures. Using this approach we have generated both computer-aided and species-specific pharmacophore fingerprints of JH and its agonists, which revealed that the most active compounds must possess an electronegative atom (oxygen or nitrogen) at both ends of the molecule. When either of these electronegative atoms are replaced by carbon or the distance between them is shorter than 11.5 Å or longer than 13.5 Å, their biological activity is dramatically decreased. The presence of an electron-deficient moiety in the middle of the JH agonist is also essential for high activity.Conclusions/SignificanceThe information from 3D QSAR provides guidelines and mechanistic scope for identification of steric and electrostatic properties as well as donor and acceptor hydrogen-bonding that are important features of the ligand-binding cavity of a JH target protein. In order to refine the pharmacophore analysis and evaluate the outcomes of the CoMFA and CoMSIA study we used pseudoreceptor modeling software PrGen to generate a putative binding site surrogate that is composed of eight amino acid residues corresponding to the defined molecular interactions.

Members of the highly conserved LAMMER family of protein kinases have been described in all eukaryotes. LAMMER kinases possess markedly similar peptide motifs in their kinase catalytic subdomains that are responsible for phosphotransfer and substrate interaction, suggesting that family members serve similar functions in widely diverged species. This hypothesis is supported by their phosphorylation of SR and SR-related proteins in diverged species. Here we describe a 3-dimensional homology model of the catalytic domain of DOA, a representative LAMMER kinase, encoded by the Drosophila locus Darkener of apricot (Doa). Homology modeling of DOA based on a Sky1p template revealed a highly conserved structural framework within conserved core regions. These adopt typical kinase folding like that of other protein kinases. However, in contrast to Sky1p, some structural features, such as those in helix αC suggest that the DOA kinase is not a constitutively active enzyme but requires activation. This may occur by phosphorylation within an activation loop that forms a broad turn and in which interactions between the side chains occur across the loop. The fold of the activation loop is stabilized through interactions with residues in the C-terminal tail, which is not part of the conserved kinase core and is variable among protein kinases. Immediately following the activation loop in the segment between the β9 sheet and helix αF is a P + 1 loop. The electrostatic surface potential of the DOA substrate-binding groove is largely negative, as it is in other known SR protein kinases, suggesting that DOA substrates must be basic. All differences between D. melanogaster and other Drosophila species are single amino acid changes situated in regions outside of any α-helices or β-sheets, and after modeling these had absolutely no visible effect on protein structure. The absence of evolved amino acid changes among 12 Drosophila species that would cause at least predictable changes in DOA structure indicate that evolution has already selected evolved mutations for having minimal effect on kinase structure.