Biochemical characterization of the Drosophila wingless signaling pathway based on RNA interference.

Regulation of Armadillo (Arm) protein levels through ubiquitin-mediated degradation plays a central role in the Wingless (Wg) signaling. Although zeste-white3 (Zw3)-mediated Arm phosphorylation has been implicated in its degradation, we have recently shown that casein kinase Ialpha (CKIalpha) also phosphorylates Arm and induces its degradation. However, it remains unclear how CKIalpha and Zw3, as well as other components of the Arm degradation complex, regulate Arm phosphorylation in response to Wg. In particular, whether Wg signaling suppresses CKIalpha- or Zw3-mediated Arm phosphorylation in vivo is unknown. To clarify these issues, we performed a series of RNA interference (RNAi)-based analyses in Drosophila S2R+ cells by using antibodies that specifically recognize Arm phosphorylated at different serine residues. These analyses revealed that Arm phosphorylation at serine-56 and at threonine-52, serine-48, and serine-44, is mediated by CKIalpha and Zw3, respectively, and that Zw3-directed Arm phosphorylation requires CKIalpha-mediated priming phosphorylation. Daxin stimulates Zw3- but not CKIalpha-mediated Arm phosphorylation. Wg suppresses Zw3- but not CKIalpha-mediated Arm phosphorylation, indicating that a vital regulatory step in Wg signaling is Zw3-mediated Arm phosphorylation. In addition, further RNAi-based analyses of the other aspects of the Wg pathway clarified that Wg-induced Dishevelled phosphorylation is due to CKIalpha and that presenilin and protein kinase A play little part in the regulation of Arm protein levels in Drosophila tissue culture cells.

HES1 cooperates with pRb to activate RUNX2-dependent transcription.

UNLABELLED: The retinoblastoma protein, pRb, can activate the transcription factor RUNX2, an essential regulator of osteogenic differentiation, but the mechanism of this activation is unknown. Here we studied the interaction of pRb and RUNX2 with HES1, previously reported to augment RUNX2 activity. PRb can act to promote RUNX2/HES1 association with concomitant promoter occupancy and transcriptional activation in bone cells. INTRODUCTION: RUNX2 (also known as OSF2/CBFA1) is a transcription factor required for osteoblast differentiation and bone formation. We have reported that RUNX2 can associate with the retinoblastoma protein pRb, a common tumor suppressor in bone, and the resultant complex can bind and activate transcription from bone-specific promoters. This activity of the pRb/RUNX2 complex may thus link differentiation control with tumor suppressor activity. However, the mechanism through which pRb can activate RUNX2 is unknown. HES1 is a reported co-activator of RUNX2 that shares a binding site on RUNX2 with pRb. Thus, we have tested the cooperativity among these factors in activating transcription from bone specific promoters. MATERIALS AND METHODS: Coimmunoprecipitation, chromatin immunoprecipitation, and EMSA experiments were used to study the interaction of RUNX2, HES1, and pRb in cell lysates and on DNA. Transcriptional reporter assays were used to analyze the activity of RUNX2 in the presence and absence of HES1 and pRb. RESULTS: We showed that pRb can associate with HES1, a previously described RUNX2 interactor that can itself augment RUNX2-dependent transcription. The association of HES1 with RUNX2 is augmented by pRb. Furthermore, both pRb and HES1 increase the amount of RUNX2 bound to promoter sites in vivo, pRb and HES1 synergistically activate a RUNX2-dependent reporter gene, and depletion of HES1 reduces RUNX2/pRb activity. CONCLUSIONS: These data indicate that pRb acts as a RUNX2 co-activator at least in part by recruiting HES1 into the pRb/RUNX2 complex and further elucidate a novel role for pRb as a transcriptional co-activator in osteogenesis.

GRB10 binds to LRP6, the Wnt co-receptor and inhibits canonical Wnt signaling pathway.

Low-density lipoprotein receptor-related protein 6 (LRP6) is a component of cell-surface receptors for Wnt proteins and Wnt is known to promote recruitment of Axin by LRP6 thereby inhibiting beta-catenin's degradation. We show here that growth factor receptor-bound protein10 (GRB10), a multi-modular adaptor protein that is known to associate with several transmembrane tyrosine kinase receptors, binds to the intracellular portion of LRP6 and negatively regulates Wnt signaling. GRB10 overexpression suppressed Wnt3a-, and LRP6-induced but not beta-catenin-induced TCF-dependent-reporter activities in HEK293T cells, suggesting that GRB10 functions upstream of beta-catenin. Actually, GRB10 overexpression attenuated the Wnt3a-induced accumulation of beta-catenin. In addition, RNAi-mediated down-regulation of endogenous GRB10 stimulated Wnt3a-induced reporter activities, indicating that GRB10 is indeed a novel negative regulator of the Wnt signaling pathway. The finding that GRB10 interferes with the binding of Axin to LRP6 indicated a possible molecular mechanism by which the overexpression of GRB10 suppresses Wnt signaling.

Axin and the Axin/Arrow-binding protein DCAP mediate glucose-glycogen metabolism.

Axin was found as a negative regulator of the canonical Wnt pathway. Human LRP5 was originally found as a candidate gene of insulin dependent diabetes mellitus (IDDM), but its Drosophila homolog, Arrow, works as a co-receptor of the canonical Wnt signal. In our previous paper, we found a new Drosophila Axin (Daxin)-binding SH3 protein, DCAP, a homolog of mammalian CAV family protein. Among the subtypes, DCAPL3 shows significant homology with CAP, an essential component of glucose transport in insulin signal. Further binding assay revealed that DCAP binds to not only Axin but also Arrow, and Axin binds to not only GSK3beta but also Arrow. However, overexpression and RNAi experiments of DCAP do not affect the canonical Wnt pathway. As DCAP is expressed predominantly in insulin-target organs, and as RNAi of DCAP disrupts the pattern of endogenous glycogen accumulation in late stage embryos, we suggest that DCAP is also involved in glucose transport. Moreover, early stage embryos lacking maternal Axin show significant delay of initial glycogen decomposition, and RNAi of Axin in S2 cells revealed quite increase of endogenous glycogen level as well as GSK3beta. These results suggest that Axin and DCAP mediate glucose-glycogen metabolism in embryo. In addition, the interaction among Axin, Arrow, and DCAP implies a possible cross-talk between Wnt signal and insulin signal.

Casein kinase I phosphorylates the Armadillo protein and induces its degradation in Drosophila.

Casein kinase I (CKI) was recently reported as a positive regulator of Wnt signaling in vertebrates and Caenorhabditis elegans. To elucidate the function of Drosophila CKI in the wingless (Wg) pathway, we have disrupted its function by double-stranded RNA-mediated interference (RNAi). While previous findings were mainly based on CKI overexpression, this is the first convincing loss-of-function analysis of CKI. Surprisingly, CKIalpha- or CKIepsilon-RNAi markedly elevated the Armadillo (Arm) protein levels in Drosophila Schneider S2R+ cells, without affecting its mRNA levels. Pulse-chase analysis showed that CKI-RNAi stabilizes Arm protein. Moreover, Drosophila embryos injected with CKIalpha double-stranded RNA showed a naked cuticle phenotype, which is associated with activation of Wg signaling. These results indicate that CKI functions as a negative regulator of Wg/Arm signaling. Overexpression of CKIalpha induced hyper-phosphorylation of both Arm and Dishevelled in S2R+ cells and, conversely, CKIalpha-RNAi reduced the amount of hyper-modified forms. His-tagged Arm was phosphorylated by CKIalpha in vitro on a set of serine and threonine residues that are also phosphorylated by Zeste-white 3. Thus, we propose that CKI phosphorylates Arm and stimulates its degradation.