N-terminal polyubiquitination and degradation of the Arf tumor suppressor

The Arf tumor suppressor protein (p19Arf in the mouse, and p14ARF in humans) antagonizes the functions of the p53-negative regulator Mdm2 to trigger p53-dependent cell-cycle arrest or apoptosis (Sherr 2001). Arf also has additional activities that are p53-independent, including an ability to inhibit ribosomal RNA (rRNA) processing (Sugimoto et al. 2003) and effects on gene expression (Eymin et al. 2001; Fatyol and Szalay 2001; Martelli et al. 2001; Datta et al. 2002; Kuo et al. 2003; Rocha et al. 2003). The p19Arf protein is unusual in several respects. It is encoded by the Ink4a locus (now designated Ink4a/Arf; Quelle et al. 1995), which also specifies a totally unrelated polypeptide, p16Ink4a, a potent inhibitor of cyclin D-dependent kinases (Serrano et al. 1993). Two transcripts initiated at separate promoters incorporate sequences from different first exons (1α and 1β) that are spliced to the products of a common downstream exon translated in alternative reading frames (from which Arf gets its name). Mouse p19Arf is a basic protein containing ∼22% arginine but only a single lysine residue that is not conserved in human p14ARF. Given its basic nature, p19Arf must be “buffered” by binding to other cellular proteins and/or nucleic acids. Most Arf protein localizes within the nucleolus where it is bound in complexes of high molecular mass (2–5 MDa) that include many other proteins (Weber et al. 1999; Bertwistle et al. 2004). Among the latter is nucleophosmin (NPM or B23), an abundant cellular protein which binds to p19Arf with high stoichiometry (Itahana et al. 2003; Bertwistle et al. 2004). Expression of Arf transcripts is difficult to detect in developing mouse embryos (Zindy et al. 1997), and mice lacking the gene develop normally but are highly prone to tumor formation (Kamijo et al. 1997). The levels of p19Arf expressed in normal tissues in vivo are also quite low. Arf is induced by elevated and sustained oncogenic signals, such as those resulting from Myc overexpression (Zindy et al. 1998) or Ras mutation (Palmero et al. 1998) but not directly by other genotoxic stresses that lead to p53 stabilization. Studies using reporter mice in which a gene encoding green fluorescent protein (GFP) was substituted for Arf exon 1β have implied that Arf activation in vivo can efficiently eliminate incipient tumor cells (Zindy et al. 2003). The p19Arf protein progressively accumulates when primary mouse embryo fibroblasts (MEFs) are passaged in culture, correlating with their decreasing proliferative capacity and eventual senescence. In contrast, Arf-null MEFs continue to proliferate much like established cell lines (Kamijo et al. 1997). Although one might imagine that p19Arf accumulation in senescing primary cells need not be accompanied by appreciable degradation, the mechanisms that regulate Arf turnover have not been elucidated. We show here that p19Arf is subject to more dynamic controls than previously thought, being degraded by the proteasome following N-terminal polyubiquitination.