Nuclear accumulation of full-length and truncated adenomatous polyposis coli protein in tumor cells depends on proliferation

Abstract

The adenomatous polyposis coli (APC) tumor suppressor is a nucleocytoplasmic protein. The nuclear accumulation of APC was recently found to vary depending on cell density, suggesting that putative APC function(s) in the nucleus is controlled by the establishment of cell contacts. We report here that the density-dependent redistribution of APC between nucleus and cytoplasm prevails in 6/6 thyroid and colorectal carcinoma cell lines. Moreover, mutated APC lacking known nuclear localization sequences had the similar distribution pattern as the fulllength protein. APC invariably accumulated in the nuclei of Ki-67 expressing cells, but was largely cytoplasmic when cell cycle exit was induced by serum starvation or at high cell density. APC colocalized with b-catenin in the nucleus only in one cell line (SW480). Also, APC maintained a predominantly nuclear position in early confluent states when cytoplasmic b-catenin was recruited to newly formed adherens-like junctions. The results indicate that nuclear targeting of APC is driven by cell cycle entry rather than altered cell-cell contact. The ability of C-terminally truncated APC to accumulate in the nucleus suggests that nuclear import signals other than NLS1 APC and NLS2 APC are functionally important. Residual function(s) of N-terminal APC fragments in tumor cells carrying APC mutations might be beneficial to tumor growth and survival.

Redefining the subcellular location and transport of APC: new insights using a panel of antibodies

“…Moreover, our own previous findings (Brocardo et al, 2001) could not be reproduced (data not shown). Two other studies used M-APC antibody (Fagman et al, 2003;Davies et al, 2004), and using this antibody we also observed a redistribution of APC from nucleus to cytoplasm with increasing density of SW480 cells and MDCK cells (Fig 4A; supplementary Fig S5 online). However, when subconfluent or confluent SW480, MDCK or HCT116 cells were fractionated and the nuclear-cytoplasmic distribution of APC was measured by immunoblotting, a density-dependent redistribution of APC was not detected (Fig 4B), although nuclear accumulation was easily detectable after LMB treatment (Fig 3B).…”

“…The N-terminal antibodies Ab7 and Ali 12-28 produced a weaker staining pattern that was mostly cytoplasmic in all cell lines, except for MDCK in which Ali 12-28 displayed a nuclear-cytoplasmic pattern. The antibody most commonly used for IF detection of APC, named M-APC, generated predominantly nuclear staining as recently described (Fagman et al, 2003;RosinArbesfeld et al, 2003;Davies et al, 2004). In the cytoplasm, all antibodies, except for Ab4, detected accumulation of APC at microtubule-associated membrane protrusions (see the MDCK cells in Fig 1C).…”

“…Several reports have described a correlation between cell density and the localization of APC on the basis of cell staining (Brocardo et al, 2001;Zhang et al, 2001;Fagman et al, 2003;Davies et al, 2004); thus, in highly packed confluent cells, APC redistributes Unambiguous detection of endogenous APC M. Brocardo et al from the nucleus to the cytoplasm. The report by Zhang et al (2001) used cell staining with the antibodies Ab4, Ab1, C20 and N-15, none of which seems to specifically detect APC in the nucleus (Fig 1; supplementary Figs S3, S4 online).…”

“…For example, although APC has been detected predominantly in the nucleus of certain cell types by cell staining (Neufeld & White, 1997;Rosin-Arbesfeld et al, 2003), biochemical fractionation methods showed a predominantly cytoplasmic distribution of APC (Smith et al, 1993;Galea et al, 2001). In addition, a shift of endogenous APC from the nucleus to the cytoplasm in response to increased cell density (Brocardo et al, 2001;Zhang et al, 2001;Fagman et al, 2003;Davies et al, 2004) or proliferation (Fagman et al, 2003;Olmeda et al, 2003) has been observed using antibodies, such as M-APC antisera by cell staining, but has not yet been confirmed by biochemical methods. Our aim was to compare the localization of APC using a panel of the most commonly used APC antibodies, using biochemical fractionation as well as fluorescence microscopy.…”

Interaction between Tumor Suppressor Adenomatous Polyposis Coli and Topoisomerase IIα: Implication for the G2/M Transition

“…Although endogenous fulllength APC exists in both cytoplasm and nucleus (Neufeld and White, 1997;Anderson et al, 2002), the classic NLSs that are thought to facilitate nuclear localization of APC (Zhang et al, 2000) are not present in the M2-APC fragment. The M2-APC region is retained in most truncated forms of APC associated with colorectal cancer and these truncated APC proteins are capable of nucleocytoplasmic shuttling (Fagman et al, 2003). However, the nuclear import ability of these truncated APC proteins has been attributed to the armadillo repeat region (amino acids 334 -625), not the 15-amino acid repeat region (Galea et al, 2001).…”