Possible interaction of b-catenin with BIG1 was first recognized in HepG2 cells more than five years ago, but only later confirmed in HeLa cells. Among proteins immunoprecipitated from HeLa cells, 1% of endogenous b-catenin coprecipitated with antibodies against BIG1 or BIG2. Immunoprecipitation (IP) of endogenous b-catenin collected also about 1 % of BIG1 and 0.5 % of BIG2. To assess potentially direct protein-protein interactions, human b-catenin, BIG1, and BIG2 were synthesized singly in vitro (wheat germ extract) and incubated together before IP. BIG1 was found with IP of b-catenin, and IP of BIG1 yielded also b-catenin, but there was no evidence for direct interaction of in vitro-synthesized BIG2 and b-catenin. To identify regions of the BIG1 molecule that interacted with b-catenin in yeast two-hybrid assays, S. cerevisiae AH109 was co-transformed with constructs encoding BIG1 amino acids 1-1849 (F, full-length), 1-885 (N), or 886-1849 (C) fused to the N-terminal activation domain of Gal4p and full-length b-catenin fused to the Gal4p DNA-binding domain. BIG1-F and BIG1-N, but not BIG1-C, interacted with b-catenin, consistent with the conclusion that b-catenin associated with structural elements in the first 885 amino acids of BIG1. Iodixanol density-gradient fractionation was used to assess intracellular distribution of several endogenous proteins. BIG1 and BIG2 were most abundant in fractions 6 to 11, partially overlapping the lighter of two b-catenin peaks. Endogenous b-catenin was seen microscopically in irregular concentrations along the cell surface, seeming an almost continuous layer in some regions, while perinuclear clusters were larger, corresponding perhaps to clusters of transport vesicles containing b-catenin with BIG1 and/or BIG2. Large and small collections of BIG1 and BIG2 were scattered throughout the cytoplasm, suggesting association of BIG1 and/or BIG2 with b-catenin in some or all sites. To explore BIG1, BIG2, and b-catenin associations further, we compared effects of BIG1 or/and BIG2 depletion on distribution of b-catenin in HeLa cells microscopically. In control cells, b-catenin was seen in punctate collections along cell surfaces and scattered throughout the cytoplasm. The extent of b-catenin overlap with both cis-Golgi GM130 and trans-Golgi syntaxin-6 in perinuclear regions seemed greater after BIG1 or/and BIG2 depletion. Depletion of BIG1 or BIG2 or over-expression of a dominant-negative GEF-inactive Sec7 domain mutant interfered with vesicular transport of specific proteins from TGN or recycling endosomes to the plasma membrane. To determine whether accumulation of b-catenin in HeLa cell Golgi after BIG1 or BIG2 depletion was related to decreased Arf activation, we over-expressed BIG1(E793K) and BIG2(E738K), mutants in which lysine replaced the Sec7 domain glutamate required for GEF activity. Expression of GEF-inactive mutants, but not wild-type (WT) BIG1 or BIG2, failed to restore b-catenin trafficking, and reverse its accumulation in perinuclear clusters. Mutant BIG1(E793K) colocalized with GM130, suggesting b-catenin arrest in cis-Golgi cisternae. Absence of b-catenin in perinuclear regions of cells overexpressing WT BIG1 or BIG2 is consistent with disruption of b-catenin transport via blocking Arf activation by depletion of BIG1 or BIG2, but does not prove it. Both BIG1 and BIG2 contain A kinase-anchoring protein (AKAP) sequences with different specificities for the four kinase regulatory (R) subunits, identified in yeast two-hybrid experiments. We found co-IP of PKA catalytic subunit (Ca) from HeLa cells, with BIG1 or BIG2 antibodies. IP of Ca yielded also BIG1 and BIG2, as well as b-catenin. Phosphorylation of S552 or S675 in the C-terminal armadillo-repeat region of b-catenin by PKA had been associated with increased b-catenin transcriptional activity. Depletion of BIG1 or BIG2 decreased similarly levels of phospho-S675 b-catenin, perhaps due to interference with assembly of b-catenin and PKA-Ca. Production and nuclear accumulation of ABC, newly translated, with no phosphorylation of S37 and T41, are important consequences of Wnt signaling that induced translocation of ABC to plasma membranes where it formed complexes with APC and LRP6, which accumulated in nuclei. ABC levels, like those of phospho-S675 b-catenin, were lower in cells after BIG1- or BIG2-depletion, although total amounts of b-catenin were unchanged. Reversal of the effect of BIG1 depletion on b-catenin phosphorylation in HeLa cells by overexpression of BIG1 confirmed its specificity, implicating BIG1 in regulation of phospho-S675 β-catenin. Overexpression of the GEF-inactive BIG1(E793K) mutant, however, failed to reverse the effects of BIG1 depletion on phospho-S675 b-catenin, consistent with additional need for Arf activation. BIG1-N and F were similarly effective. Overexpression of BIG2 WT, AKAP-A, or B, but not AKAP-C mutants reversed the effects of BIG2 depletion, consistent with critical action of the BIG1 AKAP sequence (identical to BIG2 AKAP-C) in this regulation. Association of PKA-catalyzed phosphorylation of β-catenin S552 and S675 with its greater transcriptional activity has been reported. Immunofluorescence microscopy was used to assess effects of BIG1 and/or BIG2 depletion on intracellular localization of ABC. In control cells, ABC was completely nuclear, and declined notably after depletion of BIG1 and/or BIG2. Total nuclear b-catenin was not altered, and only a very small fraction of total b-catenin was acted upon by BIG1 and BIG2, the pool that was appropriately unphosphorylated and active. In nuclei, b-catenin interacts with transcription factors of the TCF/LEF-1 family, to accomplish transcriptional activation of target genes, such as such as MYC and CCND1, which encodes cyclin D. To explore mechanisms of BIG1 and/or BIG2 roles in b-catenin phosphorylation by PKA, we estimated levels of MYC and CCND1 mRNA using RT-PCR. As amounts of both MYC and CCND1 mRNA in HeLa cells were significantly lower after depletion of BIG1 or BIG2, TCF/LEF-1 transcriptional activities were tested in control or BIG1- and/or BIG2- depleted cells, transfected with TCF/LEF-1 luciferase reporter TOP-FLASH or a negative control vector FOP-FLASH. Treatment of cells with BIG1- and/or BIG2-specific siRNA decreased TCF/LEF-1 transcriptional activity significantly. Lower levels of nuclear ABC were observed after depletion of BIG1 or/and BIG2, and TCF/LEF-1 transcriptional activity was 50% that of control cells. Significance BIG1 and BIG2, guanine nucleotide-exchange factors (GEFs) activate Arfs by accelerating replacement of bound GDP with GTP and contain one or more AKAP(A kinase-anchoring protein) sequence that scaffold multimolecular assemblies to limit cAMP signaling. β-catenin was among HeLa cell proteins co-immunoprecipitated with BIG1 or BIG2, and direct interactions were demonstrated. After BIG1 or BIG2 depletion or overexpression of GEF-inactive mutants, perinuclear accumulation of b-catenin reflected interference with Arf activation. Protein kinase A (PKA) association with b-catenin and levels of PKA-phosphorylated S675 b-catenin were also diminished in depleted cells. Failure of BIG2-AKAP-C mutant to replace endogenous BIG2 function, whereas overexpression of BIG2-AKAP-A, or B mutants was effective confirmed a specific BIG protein AKAP-C requirement for cAMP/PKA regulation of b-catenin signaling.
Showing the most recent 10 out of 12 publications
