A Signaling Pathway to Translational Control

Translation initiation is generally the rate limiting step Over twenty years ago, an obligatory role for protein in the overall process and is influenced by elements translation in G1 cell cycle progression was brought to within the 59 untranslated region (UTR) of the mRNA. light (reviewed by Norbury and Nurse, 1992). Transit The rapamycin-sensitive translation of an IGF-II mRNA through the cell cycle is thought to require a general appears to be regulated by such a mechanism. Tranincrease in the rate of translation following growth facscripts of the hormone IGF-II are alternatively spliced tor-stimulation; in fact, even a partial inhibition of protein to produce multiple mRNAs that differ only in their 59 synthesis causes fibroblasts to accumulate in the G1 UTRs. The translation of two such splice variants (4.8 phase (Norbury and Nurse, 1992). It is now well known and 6.0 kb) are discordant in their sensitivity to rapathat growth factor stimulation causes a 2-3 fold mean mycin treatment. While the 4.8 kb message is translated increase in overall protein synthesis, implicating transequally in rapamycin treated and untreated cells, transmembrane signaling events in translational control (relation of the 6.0 kb transcript is both rapamycin-sensitive viewed by Sonenberg, 1996). This average increase apand dependent on the presence of growth factors (Nielpears modest; however, while translation of some sen et al., 1995). Thus, the 1.2 kb leader sequence in transcripts remains unaffected by mitogen stimulation, the 59 UTR of the 6.0 kb variant confers translational others increase greatly (discussed below). Although our regulation to this gene. understanding of the events that regulate gene expresWhat structural elements in 59 UTRs may be involved sion through transcription has made steady progress, in their translational regulation? One element implicated the signal transduction pathways involved in translain the regulated translation of transcripts encoding ribotional control have remained relatively uncharacterized. somal proteins and the eukaryotic elongation factors 1A Catalyzed through use of cell-permeable inhibitors of and 2 (eEF1A and eEF2) is the polypyrimidine tract. This cell cycle progression, one signaling pathway that leads 59 terminal oligopyrimidine tract (59 TOP) typically conto translational control has recently become better unsists of a stretch of 4-14 pyrimidines following the Nderstood. This pathway is distinct from now well known methylguanosine cap structure of the mRNA. Translaras/MAP kinase and Jak/Stat pathways that, among tional regulation of the ribosomal protein S16 transcript other outcomes, are involved in membrane-to-nuclear is abolished by replacement of five pyrimidines with signaling events. purines in the 59 UTR (Levy et al., 1991). A short stretch Rapamycin, initially characterized as an inhibitor of of pyrimidines is also found within 139 bp of the cap G1 cell cycle progression, has recently been used to site of the 6.0 kb IGF-II transcript. Therefore, 59 TOP illuminate a growth factor-regulated signaling pathway tracts are candidate translational cis-regulatory elethat leads to the enhanced translation of a specific subments, TLREs (Levy et al., 1991), that may be modulated set of mRNAs. Translation of most transcripts is unafby the rapamycin-sensitive signaling pathway. fected by the presence of rapamycin (Jefferies et al., However, there are other structural elements in the 59 1994; Terada et al., 1994; Nielsen et al., 1995). In fact, UTR of mRNAs that are known to influence translation treatment with rapamycin for 2-3 hours has only a slight initiation. It has been speculated that regulation of traninhibitory effect (15%) on overall protein synthesis (Jefscripts with long 59 UTRs is due to the formation of feries et al., 1994; Terada et al., 1994; Mendez et al., secondary structures that must be overcome to provide 1996). However, translation of mRNAs derived from a efficient translation initiation (Sonenberg, 1996). In supfew genes is inhibited significantly. The gene transcripts port of this hypothesis, sequences that form stable secthus far identified as being rapamycin-sensitive include ondary structures (DG#264 kcal/mol) efficiently inhibit those encoding ribosomal proteins (S3, S6, S14, and translation when inserted into the 59 UTR (Sonenberg, S24), translation elongation factors (eEF1A and eEF2), 1996, and references therein). Examples of translaand a secreted peptide growth factor called insulin-like tionally-regulated mRNAs with long and complex 59 growth factor II (IGF-II). As indicated by an increased UTRs include those that encode human FGF-5, c-myc association with multiple ribosomes (polyribosomes), and ornithine decarboxylase (Sonenberg, 1996). It is these rapamycin-sensitive transcripts are translated at possible then that formation of stable secondary struca 2-10 fold increased rate following mitogen stimulation tures within the 59 UTR of some transcripts, like that of (Jefferies et al., 1994; Terada et al., 1994; Nielsen et IGF-II, may confer translational regulation that is depenal., 1995). This shift to polyribosomes occurs with a dent on rapamycin-sensitive signaling pathway. Such a concomitant decrease in the level of these mRNAs in hypothesis is supported by recent findings concerning