Inserting of a ~150 aa long protein (your ligand-dependent degron) somewhere in a protein without paying attention to the structural properties will most likely destroy your POI; it might not fold correctly anymore or the inserted domain will destroy the POI's function by disrupting the natural structure. If you want insert something in your POI, check first if there is a structure solved for you POI; insertion sites with higher chance for viability of your construct are hinge regions on the surface of your protein but still, you cannot insert everything you want. Also check if there is also a structure available for your domain which you want to insert inside your POI. If there is no structure available you can predict the POI structure using ITASSER, PHYRE or HHpred. The challenge behind inserting a domain in your POI is the same like creating split-proteins (e.g. split-GFP). Check related papers to understand how they design the split-sites. Your potential viable insertion site would also be a potential viable split-site.

/edit/ Ah, I just see you that you has the tag "membrane fusion proteins" marked; so, I guess you want anchor a protein on the cytosolic site of the TMD, thus the degron has to be C-terminal from the TMD (I guess you use a Type I TMD). This means you want to fuse the ligand-dependent degron between TMD and the fusion partner? If yes, make sure that the degron is accesssible by flanking it with flexible linker amino acids, e.g. (GGGGS)xn.

/edit2/ Here they fused your sytem to a multipass transmembrane protein TRPV5: http://www.ncbi.nlm.nih.gov/pubmed/18842822