Nuclear integration of positive Dpp signals, antagonistic Wg inputs and mesodermal competence factors during Drosophila visceral mesoderm induction

Abstract

Tissue induction during embryonic development relies to a significant degree on the integration of combinatorial regulatory inputs at the enhancer level of target genes. During mesodermal tissue induction in Drosophila, various combinations of inductive signals and mesoderm-intrinsic transcription factors cooperate to induce the progenitors of different types of muscle and heart precursors at precisely defined positions within the mesoderm layer. Dpp signals are required in cooperation with the mesoderm-specific NK homeodomain transcription factor Tinman (Tin) to induce all dorsal mesodermal tissue derivatives, which include dorsal somatic muscles, the dorsal vessel and visceral muscles of the midgut. Wingless (Wg) signals modulate the responses to Dpp/Tin along anteroposterior positions by cooperating with Dpp/Tin during dorsal vessel and somatic muscle induction while antagonizing Dpp/Tin during visceral mesoderm induction. As a result, dorsal muscle and cardiac progenitors form in a pattern that is reciprocal to that of visceral muscle precursors along the anteroposterior axis. Our present study addresses how positive Dpp signals and antagonistic Wg inputs are integrated at the enhancer level of bagpipe (bap), a NK homeobox gene that serves as an early regulator of visceral mesoderm development. We show that an evolutionarily conserved bap enhancer element requires combinatorial binding sites for Tin and Dpp-activated Smad proteins for its activity. Adjacent binding sites for the FoxG transcription factors encoded by the Sloppy paired genes (slp1 and slp2), which are direct targets of the Wg signaling cascade, serve to block the synergistic activity of Tin and activated Smads during bap induction. In addition, we show that binding sites for yet unknown repressors are essential to prevent the induction of the bap enhancer by Dpp in the dorsal ectoderm. Our data illustrate how the same signal combinations can have opposite effects on different targets in the same cells during tissue induction.