Characterization of the subcellular localization of the TGF-β receptors in the wing imaginal disc

The TGF-β pathway has been extensively studied in vivo in the wing imaginal disc of D. melanogaster. In particular, many investigations focus on the TGF-β ligand Dpp (BMP2/4 orthologue) that acts as a morphogen, regulating patterning and growth of wing disc cells.
It was reported that Dpp forms an extracellular gradient consisting of an apical and a basolateral fraction (Entchev, Schwabedissen, & González-Gaitán, 2000; Gibson, Lehman, & Schubiger, 2002; Teleman & Cohen, 2000). However the function of these distinct fractions and the subcellular localization of the molecules responsible for ligand perception have not yet been characterized in this developmental context.
Therefore, in this thesis I investigated the localization of BMP receptors and co-receptors expressed in the wing disc. I found that the D. melanogaster TGF-β superfamily type-I receptors have different subcellular localizations: while the BMP receptors Tkv and Sax distribute along the apical and the basolateral side of the wing disc epithelium, the TGF-β/Activin receptor Babo is only basolaterally localized. This subcellular bias in localization is also found in the TGF-β type-II receptors, Punt and Wit. Punt localizes only to the basolateral compartment, whereas Wit is also present at the apical compartment and shows a clear enrichment in this domain. The glypicans Dally and Dlp localize to both apical and basolateral compartments, as previously reported (Ayers, Gallet, Staccini-Lavenant, & Thérond, 2010; Gallet, Staccini-Lavenant, & Thérond, 2008).
In order to understand the functional importance of the polarized subcellular localization of TGF-β receptors, I tried to mislocalize the TGF-β receptors using different approaches. In a first approach I have been using membrane-bound nanobodies. Despite the fact that these nanobody-tools were designed to trap secreted GFP-tagged proteins, I will show that they are also potential tools to mislocalize polarized membrane proteins. In another approach, I tried to mislocalize the TGF-β receptors through the mutation of “targeting domains”. Finally, to address whether signalling polarization has important consequences on physiological development, I attempted to modify the subcellular localization of the endogenous type-I receptor Tkv. To this end, I used a landing site in the tkv locus, generated by replacing the last two exons of tkv with an attP landing site and resulting in a tkv null allele. This tool allows manipulation of the receptor at the endogenous locus, important to understand the developmental impact of its proper subcellular localization. I achieved to generate different tagged versions of Tkv, with either an extracellular or a C-terminal mCherry tag. Moreover, I deleted a conserved targeting domain in the Tkv protein and obtained apical enrichment of Tkv localization.
In the following, I will discuss these different methods and their implications for studying TGF-β signalling polarization in the wing disc.