al levels, it is likely that the high concentrations of the proteins may force interactions that would not occur 
at physiological levels of expression. To diminish that concern, we examined the interaction of exogenously expressed FLAG-Glis3 or its PY461 mutant in HEK293T cells with endogenous ITCH protein using immunoprecipitation and an anti-Flag M2 antibody. Indeed, as shown in Fig 2F, endogenous ITCH interacted robustly with Glis3, while no interaction was detected with the PY461 mutant, reinforcing the requirement for that motif. Although several commercial Glis3 antibodies are available none of these were able to recognize endogenous Glis3 protein order TL32711 thereby preventing us from examining the interaction with endogenous Glis3. Interaction between Glis3 and HECT E3 ubiquitin ligases is mediated by WW-domains Members of the Nedd4/Rsp5p family contain in addition to the centrally located WWdomains, an N-terminal C2 domain and a C-terminal HECT domain. To determine whether these domains influence the interaction PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19741364 between Itch and Glis3, IP was performed with Glis3 and Itch lacking its C2 or HECT domains. The results showed that deletion of either the C2 or the HECT domain had little effect on the interaction between Itch and Glis3 and therefore are not required for the interaction. In addition, the four WW domains from Itch were sufficient for interaction with Glis3 although the interaction was considerably weaker than observed with full-length Itch and might be due to differences in the WW domain conformation and therefore affinity for Glis3 between full-length Itch and the Itch WW domain only. The WW-domains of Smurf2 and NEDD4 were also sufficient for interaction with Glis3 in a PY461 motif dependent manner. These data indicate that the association of Glis3 with these HECT ubiquitin ligases is mediated through a direct interaction of the WW domains with the PY461-motif of Glis3. In order to determine whether the interaction required secondary proteins associated with either Itch or Glis3, an in vitro pulldown was performed 9 / 22 Regulation of Glis3 Activity by the HECT E3 Ubiquitin Ligases Fig 3. Itch directly interacts with Glis3 through its WW-domains. A. HEK293T cells were transfected with FLAG-Glis3 and Myc empty vector, Itch, Itch-C2, Itch–HECT, or Itch-WWI containing only the four WW-domains. Co-immunoprecipitation was performed as described in Fig 2A2C. B. FLAG-Glis3 or the PY461 mutant and Myc Empty or Myc-Itch were transcribed and translated in vitro and an in vitro pulldown assay was performed using anti-Myc antibody as described in Methods and Materials. G. Schematic showing the interaction between Itch with the PY461 motif of Glis3. doi:10.1371/journal.pone.0131303.g003 using in vitro transcribed and translated proteins. As seen in Fig 3B, TnT-Myc-Itch immunoprecipitated with TnT-FLAG Glis3 and PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19743978 a weaker interaction was observed between TnT-Myc Itch and TnT-FLAG PY461 mutant suggesting that the interaction is direct and does not require intermediate proteins. Collectively, these results demonstrate that the HECT E3 ubiquitin ligases interact directly with the PY461 motif of Glis3 through their WW-domains and that mutation of two residues of the core PPxY motif greatly, but not completely, disrupt their interaction. A schematic summary of the interaction between Glis3 and Itch is shown in Fig 3C. HECT E3 ubiquitin ligases promote Glis3 polyubiquitination To determine whether the E3 ubiquitin ligases were capable of promoting