Stability of nucleolar versus non-nucleolar forms of human p14(ARF).

Fusion proteins containing the amino-terminal domain of human p14(ARF) linked to green fluorescent protein are able to bind MDM2 and stabilize p53 without localization in the nucleolus. However, these fusion proteins are inherently unstable, with half-lives considerably shorter than either authentic ARF or chimaeras containing the entire coding domain, both of which are predominantly nucleolar. We present evidence that the unstable fusion proteins are significantly stabilized if redirected to the nucleolus by addition of a basic motif based on the nuclear localization signal of SV40 T-antigen. Moreover, the stability of these proteins can be enhanced by modulating the functions of MDM2 and p53. These data are consistent with a model in which ARF interacts with MDM2 in the nucleoplasm but is consequently subject to proteasomal degradation. Nucleolar localization may serve to store or stabilize ARF.

Lysophosphatidic acid attenuates the cytotoxic effects and degree of peroxisome proliferator-activated receptor gamma activation induced by 15-deoxyDelta12,14-prostaglandin J2 in neuroblastoma cells.

PPARgamma (peroxisome proliferator-activated receptor gamma) is a ligand-activated transcription factor that responds to 15dPGJ2 (15-deoxy-Delta12,14-prostglandin J2). 15dPGJ2, in vitro, halts neuroblastoma cell growth, but reported mechanisms vary. Here we evaluated the modulatory effects of endogenous serum lipid mitogens upon the extent of 15dPGJ2-induced growth inhibition and on the precise cellular responses of neuroblastoma cells to PPARgamma activation. We show that 15dPGJ2 specifically inhibited cell growth in both complete and delipidated media. 15dPGJ2-induced growth inhibition was accompanied by decreased cell viability, although the effect was far more marked in delipidated medium than in complete medium. Incubation with 15dPGJ2 in complete medium resulted in cytoplasmic changes characteristic of type II programmed cell death (autophagy), while prior serum lipid removal resulted in cell death via an apoptotic mechanism. These distinct, serum lipid-dependent cellular responses to 15dPGJ2 were accompanied by increases in the expression of a reporter gene construct containing a PPAR response element of 2.3-fold in complete medium, but of 4.8-fold in delipidated medium. Restoration of the serum lysolipid LPA (lysophosphatidic acid) to cells in delipidated medium reduced 15dPGJ2-mediated PPARgamma activation, growth inhibition and cell death; following addition of S1P (sphingosine 1-phosphate), decreases were apparent but more marginal. Further, while the effects of LPA in delipidated medium were mediated through a G(i)/phosphoinositide 3-kinase/MAPK (mitogen-activated protein kinase) pathway, those of S1P did not involve the MAPK component. These data suggest that the serum lysolipid LPA modulates the degree of PPARgamma activation and the precise cellular response to 15dPGJ2 via activation of a G(i)/phosphoinositide 3-kinase/MAPK pathway.