Our studies of nonsense-containing transcripts together with results from a survey in 1998 that we undertook indicate that transcripts from the majority of mammalian genes are subject to NMD when they prematurely terminate translation more than ~50-55 nt upstream of the final exon-exon junction. Therefore, we now understand why disease-associated nonsense codons generally reduce mRNA abundance but normal termination codons, which usually reside within the final exon, generally do not. One of the most significant outcomes of our work has been the finding that nuclear pre-mRNA splicing influences cytoplasmic mRNA translation by influencing mRNP structure. Data indicate that a complex of proteins is deposited on mRNA immediately upstream of exon-exon junctions as a consequence of pre-mRNA splicing. This complex consists of Upf3 or Upf3X, a mostly nuclear shuttling protein involved in NMD. The Upf3- or Upf3X-bound complex then recruits Upf2, a mostly cytoplasmic but perinuclear protein also ivolved in NMD. Upf2 ultimately recruits Upf1, which is largely cytoplasmic. If translation terminates more than ~50-55 nt upstream of an exon-exon junction marked by the Upf proteins, then EJC-associated Upf1 elicits NMD. However, if translation terminates less than ~50-55 nt upstream of or downstream of the 3’-most exon-exon junction, then translating ribosomes are thought to remove all exon junction complexes and the mRNA is immune to NMD. Consistent with this model, we have found that the substrate for NMD is mRNA bound by the cap binding protein (CBP) heterodimer CBP80 and CBP20.  CBP80/20-bound mRNA undergoes what we have called a “pioneer” round of translation, and it is during this round (which may involve one or more ribosomes) that nonsense codon recognition can lead to NMD. Once CBP80/20 has been replaced by eukaryotic initiation factor (eIF) 4E, the mRNA is immune to NMD. Consistent with NMD targeting CBP80/20-bound but not eIF4E-bound mRNA, the Upf-containing exon junction complexes are detected only on CBP80/20-bound mRNA. Furthermore, recent studies indicate that CBP80 promotes NMD by interacting directly with Upf1 and promoting the interaction of Upf1 with Upf2. Thus, our studies have led to the discovery of a new template for protein synthesis and new insights into rearrangements of mRNP structure during the lifetime of an mRNA.

Approximately one-third of genetic diseases are due to frameshift and nonsense mutations that result in the premature termination of translation. Studies in progress will significantly advance our understanding of the factors involved in splicing, translation termination and mRNA decay that are required for NMD. Our results will be useful when designing therapies that aim to abrogate NMD in order to abrogate the severity of nonsense-generated diseases.

We have found that Upf1 also interacts with the double-stranded RNA binding protein Staufen (Stau)1. To understand the significance of this finding, microarray analyses were used to identify mRNAs that bind Stau1. For those mRNAs tested, Stau1 interacts with the 3’ untranslated region, and down-regulating the cellular level of Stau1 or Upf1 up-regulates mRNA half-life. These and other results indicate that Stau1 mediates the decay of specific cellular mRNAs when translation terminates normally. This Stau1-mediated mRNA decay (SMD) contributes significantly to the network of posttranscriptional regulatory pathways in mammalian cells. SMD differs from NMD because it occurs independently of splicing and after downregulating Upf NMD factors other than Upf1. Furthermore, SMD, unlike NMD, targets not only CBP80/20-bound mRNA but also eIF4E-bound mRNA. However, SMD is similar to NMD because it requires translation and the recruitment of Upf1 sufficiently downstream of a termination codon.

NMD is a splicing-dependent and translation-dependent pathway that targets not only disease-associated but also naturally occurring transcripts (for recent reviews, see Maquat, 2004, Nat. Rev. Mol. Cell Biol 5:89-99; Maquat, 2004, Curr. Genomics 5:174-190), many of which are mistakes made during alternative splicing (Pan et al., 2006, Genes & Dev. 2006 20:153-8).  Currently, we are interested in further characterizing the pioneer round of translation, during which nonsense codon recognition leads to NMD (Ishigaki, Li and Maquat, 2001 Cell 106:607-617; Lejeune et al., 2002, EMBO J. 21:3536-3545).  We have made important progress in identifying components of the pioneer translation initiation complex, which consists of CBP80/20 at the mRNA cap, PABP2 and PABP1 at the mRNA poly(A) tail, and the exon junction complex of proteins that includes the NMD factors Upf3 or Upf3X, Upf2, and, finally, Upf1 (Chiu, Lejeune, Ranganathan and Maquat, 2004, Genes & Dev. 18:645-754; Lejeune, Ranganathan and Maquat, 2004, Nat. Struct. Mol. Biol. 11:992-1000; Hosoda, Lejeune and Maquat, 2006, Mol. Cell. Biol. 26:3085-3097).  We have found that CBP80 promotes NMD by promoting the interaction between Upf1 and Upf2 (Hosoda, Kim, Lejeune and Maquat, 2005, Nat. Struct. Mol. Biol. 12:893-901).  We are particularly interested in understanding additional changes in mRNP structure that occur during the pioneer round of translation and its remodeling to the steady-state initiation complex, eIF4E-bound mRNA.  We are also interested in understanding the mechanistic difference between nucleus-associated and cytoplasmic NMD, the degradative enzymology of NMD (Lejeune, Li and Maquat, 2003, Mol. Cell 12:675-687), and factor function in NMD (Chiu, Serin, Ohara and Maquat, 2003, RNA 9:77-87; Brumbaugh et al., 2004, Mol. Cell 14:585-598).

Opportunities are also available to study a new mRNA decay pathway that we recently uncovered (Kim, Furic, DesGrosseillers and Maquat, 2005, Cell 120:195-208). This pathway, which we call Staufen1-mediated mRNA decay (SMD), has opened up a whole new field of research for us. We have found that Staufen1, which is a double-stranded RNA binding protein, recruits the NMD factor Upf1 to certain mRNA 3’ untranslated regions so as to elicit SMD in a translation-dependent fashion. Using microarray analyses, we have identified a number of mRNAs that are naturally down-regulated by SMD. Future studies aim to elucidate how mammalian cells utilize SMD to regulate gene expression. Included in these studies are identifying mRNA sequences that bind Staufen1, characterizing the complex Staufen1-containing mRNA binding complex, defining the mRNP rearrangements that occur during SMD, and characterizing the physiological significance of SMD. Unlike NMD, SMD targets not only CBP80/20-bound mRNA but also its rearranged product, eIF4E-bound mRNA. This makes sense for a conditionally regulated pathway.

In summary, significant projects are available in molecular biology, biochemistry and cell biology. Successful candidates will join a well-equipped group of interactive lab members with diverse backgrounds and broad expertise in newly remodeled labs. The University of Rochester is unique for its sizeable community of RNA researchers.