Pavel Rodriguez

FUNCTIONAL DOMAINS OF RNA BINDING PROTEIN CUGBP2. Pavel Rodriguez¹ and Shrikant Anant^2,3, ¹ Department of Biology, Washington University, St Louis, MO; ²Department of Medicine, Washington University School of Medicine, St Louis, MO; ³Department of Medicine and Cell Biology University of Oklahoma Health Sciences Center, Oklahoma City, OK

CUGBP2 (also called NAPOR, ETR-3, or BRUNOL3) is a ubiquitously expressed RNA binding protein with highest expression observed in muscle and brain. It was originally believed to play a role in regulating alternative splicing. We identified the protein based on its ability to interact with apobec-1, the catalytic component of the apolipoprotein B (apoB) mRNA complex in a yeast two-hybrid screen of liver cDNA library. Furthermore, we have determined that CUGBP2 binds to AU-rich sequences immediately upstream of theedited cytidine in apoB mRNA, and represses C to U editing ofapoB transcripts. In addition, we have determined that CUGBP2 is a nuclear protein that is sometimes found in the cytoplasm where it can regulate stability and translation of Cyclooxygenase-2 mRNA. However, the mechanism of CUGBP2 function is currently unknown. As a first step to understanding CUGBP2 function, we proposed to perform a structure: function study by generating deletion mutants. CUGBP2 has significant structural homology to HuR, a member of the ELAV RNA binding protein family. Both CUGBP2 and HuR encode three RNA recognition motifs (RRMs) with the first two RRMs (RRM1, RRM2) in tandem, and the last one (RRM3) separated from the first two by a linker region. A major difference between the proteins is the presence of an A/Q rich linker region that is present only in CUGBP2. The hypothesis being tested is that RRM1 and RRM2 are involved in mRNA binding and stabilization, while the A/Q rich linker region and possibly RRM3 are engaged in the inhibition of COX-2 mRNA translation. To understand the role of the different domains, we have generated a series of deletion mutants, including deletion of the individual RRMs and the A/Q-rich bridge region, as well as deletion of both RRM1 and RRM2. In addition we generated constructs that encoded just the individual domains. The full-length cDNA was used as template for PCR with specific primers and Pfu DNA polymerase. The cDNAs were cloned into plasmid pPCR-Amp Script, and sequenced to determine that no PCR-mediated mutations were introduced into the sequence. After confirmation of the validity of the sequence, the cDNAs were isolated and cloned into plasmids pCMV-Tag2B (for expression in mammalian cells) and pGEX-4T3 (for expression in bacteria). Validity of the cloned cDNA in these two plasmids was further confirmed by sequencing the forward and reverse strands. Further studies were carried with these deletion mutants. We have recently identified that the full length CUGBP2 localizes to the nucleus and shuttles between the nucleus and cytoplasm. Using immunocytochemistry, we have determined the localization of the individual proteins in two human cell lines, HCT116 colon adenocarcinomas cells and HeLa cervical carcinoma cells. We have also determined the effects of the mutants on stability and translation of COX-2 mRNA. These studies delineate the role of the different domains of CUGBP2 in stability and translation.