Growth Inhibition by the Tumor Suppressor p33ING1 in Immortalized and Primary Cells: Involvement of Two Silencing Domains and Effect of Ras

Abstract

ING1 was identified as an inhibitor of growth and has been described as a tumor suppressor. Furthermore, the expression of ING1 is induced in senescent cells and antisense ING1 extends the proliferative life span of primary human fibroblasts. Cooperation of p33ING1 with p53 has been suggested to be an important function of ING1 in cell cycle control. Intriguingly, it has been shown that p33ING1 is associated with histone acetylation as well as with histone deacetylation function. Here we show that p33ING1 is a potent transcriptional silencer in various cell types. However, the silencing function is independent of the presence of p53. By use of deletion mutants two potent autonomous and transferable silencing domains were identified, but no evidence of an activation domain was found. The amino (N)-terminal silencing domain is sensitive to the histone deacetylase inhibitor trichostatin A (TSA) whereas the carboxy-terminal silencing function is resistant to TSA, suggesting that p33ING1 confers gene silencing through both HDAC-dependent and -independent mechanisms. Interestingly, the presence of oncogenic Ras, which is able to induce premature senescence, increases the p33ING1-mediated silencing function. Moreover, ING1-mediated silencing was reduced by coexpressing dominant-negative Ras or by treatment with the mitogen-activated protein kinase inhibitor PD98059 but not by treatment with SB203580, an inhibitor of the p38 pathway. In addition, we show that both silencing domains of ING1 are involved in cell cycle control, as measured by inhibition of colony formation of immortalized cells and by thymidine incorporation of primary human diploid fibroblasts (HDF). Interestingly, p33ING1 expression induces features of cellular senescence in HDFs.

A functional link between the tumour suppressors ARF and p33ING1

“…We have also observed that deletions including either the N-terminus or the Cterminus of p33ING1 are unable to trigger cell-cycle arrest to the same extent as the full-length protein, this effect being more obvious for the C-terminal construct. This is in agreement with previous observations in human diploid fibroblasts (Goeman et al, 2005). Interestingly, both deletion constructs showed an impaired capacity to increase p53 levels, and accordingly p21CIP1 and Mdm2 (Supplementary Figure 2).…”

“…ING proteins have been linked functionally to the p53 pathway at different levels, including the control of p53 protein stability, as transcriptional cofactors, or through post-translational modifications of p53 (Nagashima et al, 2001;Nourani et al, 2001;Gozani et al, 2003). ING proteins also participate in chromatin remodelling through their association with histone deacetylases and histone acetyltransferases (Nourani et al, 2001;Feng et al, 2002;Kuzmichev et al, 2002;Vieyra et al, 2002;Xin et al, 2004;Goeman et al, 2005). The most extensively studied member of the ING family is p33ING1, one of the products of the ING1 locus (also referred to as p33ING1b, or ING1b elsewhere; Feng et al, 2002).…”

“…Next, we sought to define the p33ING1 domains involved in this interaction. We have recently defined two functional motifs, in the Nand C-terminus of p33ING1, which are important for its activity (Goeman et al, 2005). After co-transfection of p14ARF with specific deletion constructs for p33ING1 in 293T cells, we were able to observe co-precipitation between p14ARF and a mutant protein containing residues 1-171 of p33ING1 (p33ING1-N), whereas no interaction was observed with a C-terminal construct containing residues 172-279 (p33ING1-C), indicating that the N-terminal domain of p33ING1 is sufficient for binding to p14ARF ( Figure 1d).…”

“…More importantly, there is ample evidence, both in mammals and yeast, showing that ING proteins can form complexes with histone deacetylases and histone acetyltransferases. Based on this, it has been suggested that ING proteins could play a role in transcriptional control, through chromatin remodelling (Nourani et al, 2001;Skowyra et al, 2001;Feng et al, 2002;Kuzmichev et al, 2002;Vieyra et al, 2002;Xin et al, 2004;Goeman et al, 2005). An increasing number of evidences indicate that ARF can have a direct role in transcriptional control (Kamijo et al, 1998;Rocha et al, 2003Rocha et al, , 2005Calabro et al, 2004;D'Amico et al, 2004;Datta et al, 2004;Kalinichenko et al, 2004;Qi et al, 2004) and, therefore, it is feasible that ARF might cooperate with p33ING1 in transcriptional regulation, through chromatin remodelling processes.…”

“…Our data showing a clear effect of ARF on p33ING1-dependent transactivation, in the absence of effects on p53 protein levels, are consistent with this hypothesis. In this regard, it is important to note that the N-terminal domain of p33ING1, which is sufficient for binding to p14ARF, is implicated in histone deacetylase-mediated transcriptional repression (Goeman et al, 2005). Further understanding of the functional connection between ARF and p33ING1 should provide important information about the mechanisms underlying the role of these two proteins in tumour suppression.…”