Association of Tumor Suppressor Protein Pdcd4 With Ribosomes Is Mediated by Protein-Protein and Protein-RNA Interactions

Abstract

The Pdcd4 (programmed cell death gene 4) gene has been implicated as a novel tumor suppressor gene in the development of several types of human cancer. The Pdcd4 protein is believed to act as a translation suppressor of mRNAs containing structured 5′ UTRs. Pdcd4 contains 2 copies of so-called MA3 domains that mediate tight interactions with the translation initiation factor eIF4A, resulting in the inhibition of the eIF4A helicase activity. The N-terminal part of Pdcd4, which has been less well characterized, binds RNA in vitro, but as yet, it has not been clear whether RNA binding by Pdcd4 plays a role in vivo. Here, the authors have identified 2 highly conserved clusters of basic amino acid residues that are essential for the RNA binding activity of Pdcd4. They also show that a substantial fraction of Pdcd4 is present, together with small ribosomal subunits, in translation preinitiation complexes. Using mutants that disrupt RNA binding or the Pdcd4-eIF4A interaction, they demonstrate that the ribosomal association of Pdcd4 is dependent on its RNA binding activity as well as on its ability to interact with eIF4A. Their work provides the first direct evidence for an essential role of the Pdcd4 RNA binding activity in vivo and suggests that RNA binding is required for recruiting Pdcd4 to the translation machinery.

Pdcd4 directly binds the coding region of c-myb mRNA and suppresses its translation

“…Initially, in vitro binding experiments using nonspecific RNA suggested that the N-terminal part of Pdcd4 is involved in RNA binding (Bo¨hm et al, 2003). More recently, we have shown that the Pdcd4 RNAbinding domain is required for the association of Pdcd4 with ribosomal complexes in vivo, providing the first evidence for a role of RNA binding in targeting Pdcd4 to the translation machinery (Wedeken et al, 2010). The identification of a 'Pdcd4-response region' in c-myb mRNA has allowed, for the first time, to study RNA binding of Pdcd4 in the context of a physiological RNA target site.…”

“…Previous work has shown that the amino terminal domain of Pdcd4 binds to RNA and that its RNA-binding activity is mediated by two clusters of basic amino acids. Mutation of both clusters (which are referred to as RBM1 and 2) results in a complete loss of the RNAbinding activity (Bo¨hm et al, 2003;Wedeken et al, 2010). As shown in Figure 5, mutant Pdcd4 carrying mutations of the basic amino acid clusters inhibited c-Myb translation less strongly than wild-type Pdcd4, Nco, E, X, N, S and B refer to restriction sites for NcoI, EcoRI, XmaI, NaeI, SalI and BglII, respectively.…”

“…pCMVluc was obtained by inserting the CMV promoter into pGL3-basic (Promega, Mannheim, Germany). Expression vectors for hemagglutinin-tagged chicken Pdcd4 and wild-type human Pdcd4 have been described (Schlichter et al, 2001a;Waters et al, 2007) pCDNA4-hPdcd4-mutRBM1 þ 2 encodes a mutant Pdcd4 protein with amino-acid substitutions of several basic amino acids in the N-terminal part of Pdcd4 (Wedeken et al, 2010). Bacterial expression vectors for GST-Pdcd4 and GST-Pdcd4 mutRBM1 þ 2 have been described (Wedeken et al, 2010).…”

“…Density gradient centrifugation was performed as described (Wedeken et al, 2010). Ribosomal RNAs were visualized by agarose gel electrophoresis.…”

“…The RNA was extracted with TriZOL, ethanolprecipitated and dissolved in diethylpyrocarbonate-treated water. GST-Pdcd4-wt and GST-Pdcd4-mutRBM1 þ 2 have been described (Wedeken et al, 2010). The proteins were purified by binding to GST sepharose in buffer containing 1 Â phosphate-buffered saline, 1% NP-40, 2 mM DTT, 1 mM phenylmethanesulfonylfluoride and a protease inhibitor cocktail containing pepstatin A, leupeptin and aprotinin.…”