Tiki casts a spell on Wnt

From injury to patterning—MAPKs and Wnt signaling in Hydra

Hydra has a regenerative capacity that is not limited to individual organs but encompasses the entire body. Various global and integrative genome, transcriptome and proteome approaches have shown that many of the signaling pathways and transcription factors present in vertebrates are already present in Cnidaria, the sister group of Bilateria, and are also activated in regeneration. It is now possible to investigate one of the central questions of regeneration biology, i.e., how does the patterning system become activated by the injury signals that initiate regeneration. This review will present the current data obtained in Hydra and draw parallels with regeneration in Bilateria. Important findings of this global analysis are that the Wnt signaling pathway has a dual function in the regeneration process. In the early phase Wnt is activated generically and in a second phase of pattern formation it is activated in a position specific manner. Thus, Wnt signaling is part of the generic injury response, in which mitogen-activated protein kinases (MAPKs) are initially activated via calcium and reactive oxygen species (ROS). The MAPKs, p38, c-Jun N-terminal kinases (JNKs) and extracellular signal-regulated kinases (ERK) are essential for Wnt activation in Hydra head and foot regenerates. Furthermore, the antagonism between the ERK signaling pathway and stress-induced MAPKs results in a balanced induction of apoptosis and mitosis. However, the early Wnt genes are activated by MAPK signaling rather than apoptosis. Early Wnt gene activity is differentially integrated with a stable, β-Catenin-based gradient along the primary body axis maintaining axial polarity and activating further Wnts in the regenerating head. Because MAPKs and Wnts are highly evolutionarily conserved, we hypothesize that this mechanism is also present in vertebrates but may be activated to different degrees at the level of early Wnt gene integration.

Wg secreted by conventional Golgi transport diffuses and forms Wg gradient whereas Wg tethered to extracellular vesicles do not diffuse

Wingless (Wg)/Wnt family proteins are essential for animal development and adult homeostasis. Drosophila Wg secreted from the dorsal-ventral (DV) midline in wing discs forms a concentration gradient that is shaped by diffusion rate and stability of Wg. To understand how the gradient of extracellular Wg is generated, we compared the secretion route of NRT-Wg, an artificial membrane-tethered form of Wg that is supposedly not secreted but still supports fly development, to that of wild-type Wg. We found that wild-type Wg is secreted by both conventional Golgi transport and via extracellular vesicles (EVs), and NRT-Wg can be also secreted via EVs. Furthermore, wild-type Wg secreted by Golgi transport diffused and formed Wg gradient but Wg-containing EVs did not diffuse at all. In case of Wg stability, Sol narae (Sona), a metalloprotease that cleaves Wg, contributes to generate a steep Wg gradient. Interestingly, Wg was also produced in the presumptive wing blade region, which indicates that NRT-Wg on EVs expressed in the blade allows the blade cells to proliferate and differentiate without Wg diffused from the DV midline. We propose that EV-associated Wg induces Wg signaling in autocrine and juxtaposed manners whereas Wg secreted by Golgi transport forms gradient and acts in the long-range signaling, and different organs differentially utilize these two types of Wg signaling for their own development.

Tiki, at the head of a new superfamily of enzymes

Tiki proteins appear to antagonize Wnt signalling pathway by acting as Wnt proteases, thereby affecting Wnt solubility by its amino-terminal cleavage. Tiki1 protease activity was shown to be metal ion-dependent and was inhibited by chelating agents and thus was tentatively proposed to be a metalloprotease. Nevertheless, Tiki proteins exhibit no detectable sequence similarity to previously described metalloproteases, but instead have been reported as being homologues of TraB proteins (Pfam ID: PF01963), a widely distributed family of unknown function and structure. Here, we show that Tiki proteins are members of a new superfamily of domains contained not just in TraB proteins, but also in erythromycin esterase (Pfam ID: PF05139), DUF399 (domain of unknown function 399; Pfam ID: PF04187) and MARTX toxins that contribute to host invasion and pathogenesis by bacteria. We establish the core fold of this enzymatic domain and its catalytic residues.