RNA EXPORT

Background: The presence of a nucleus is a defining characteristic of eukaryotic cells. Subdivision of the cell into nuclear and cytoplasmic compartments creates the requirement for some mechanism to move macromolecules between the nucleus and the cytoplasm. Proteins required within the nucleus are synthesized in the cytoplasm and must be imported into the nucleus. And RNA molecules required in the cytoplasm (mRNA, tRNA, rRNA) are synthesized in the nucleus, assembled into nucleoprotein complexes, and exported. All transport between these compartments takes place through nuclear pore complexes (NPCs), very large (66 megadaltons in yeast, >100 megadaltons in metazoans) multiprotein complexes which contain multiple copies of more than 30 polypeptides .

To identify nuclear transport proteins, we developed an in situ hybridization assay to detect poly(A)+ mRNA and used this assay to screen temperature-sensitive mutant strains of Saccharomyces cerevisiae to identify those that showed nuclear accumulation of poly(A)+ RNA following a shift to the non-permissive temperature of 37°C. By screening more than 1200 ts strains, we identified about 40 strains showing nuclear accumulation. Further genetic testing eliminated those where multiple mutations were required to produce the observed phenotype. This left approximately 24 mutants in 10 complementation groups. The defective genes were identified by cloning the wild-type form by complementation of the ts growth defect.

Genetic identification of proteins important for mRNA export: Five of the genes identified encode nuclear pore complex proteins, called nucleoporins. These are Nup120p, Nup133p, Nup159p, Nup145p, and Nup85p. These are five of the seven yeast nucleoporins (Nup82p and Nup84p are the other two) that play important roles in mRNA export but are not involved in protein transport. Interestingly, Nup85p, Nup84p, Nup120p, Nup133p, along with Seh1p, are parts of an NPC subcomplex. None are essential but deletion of any one causes ts growth defects (deletion of SEH1 is cold-sensitive). Nup159p is part of a different NPC subcomplex. We determined that Nup159p is linked to the NPC through Nup82p and Nsp1p. All three of these are essential nucleoporins,. All three are also heptad repeat proteins (coiled coil), and interact through their heptad repeats. Truncation of the heptad repeat region of Nup159p or Nup82p results in mRNA export defects, temperature-sensitive growth, and loss of Nup159p from NPCs.

Another protein identified by this screen is Dbp5p, a DEAD-box protein. Since DEAD-box proteins are involved in all other steps of RNA metabolism, from synthesis to turnover, it was to be expected that one or more DEAD-box proteins would play a role in mRNA export. All DEAD-box proteins tested have ATPase activity; for most, this is dependent on or stimulated by RNA. Some have RNA unwinding activity as well, while others show RNA unwinding activity only in the presence of other proteins or a cell extract. Dbp5p falls into this latter class. Dbp5p is a shuttling protein that associates with the NPC through the N-terminal portion of Nup159p. Thus, when Nup159p is lost from the poor, the functional defect in mRNA export most likely reflects the absence of Dbp5p from the NPC.

The role in mRNA export of the DEAD-box protein, Dbp5p: Understanding the precise function of Dbp5p is one primary goal of this laboratory. We are attempting to identify proteins with which Dbp5p interacts. We are also testing the hypothesis that Dbp5p uses its ATPase activity to disassemble the mRNA/protein complex exported from the nucleus. This would allow hnRNP proteins to return to the nucleus for another round of export, and allow the mRNA to associate with other proteins and be translated. The Dead-box protein to which Dbp5p is most closely related is a translation initiation factor, eIF4A. This protein requires eIF4B, an RNA binding protein, to show strong RNA unwinding activity. Since eIF4A and Dbp5p have a very high degree of identity, we suspect that Dbp5p will also require a co-factor in order to play its role and have full activity. We are performing genetic screens to search for an interacting protein. We are also conducting studies to determine if any of the known soluble mRNA transport factors (e.g. Mex67p, Gle1p, Mtr2p) function as co-factors for Dbp5p.

Many recent studies indicate that the many events of gene expression, from initiation of transcription through RNA processing, RNA export, translation, and RNA degradation, are coupled. Many processing factors travel with RNA polymerase and this may allow them easy access to processing signals in the elongating mRNA. Similarly, mRNA export factors are recruited during transcription but how this couples RNA synthesis and processing to mRNA export is uncertain. Our studies indicate that Dbp5p may associate with the mRNA early during transcription.

The mechanism of selective mRNA export and selective protein import following stress: Following stress in yeast (heat shock or 10% ethanol shock), poly(A)+ mRNA accumulates in the nucleus. At the same time, various stress response/heat shock genes are strongly induced at the transcriptional level. We adapted our in situ assay to detect heat shock mRNA (SSA4 mRNA encoding hsp70) and determined that heat shock mRNA was exported efficiently when bulk poly(A)+ mRNA accumulated in nuclei following heat shock. We showed that this specific export requires Nup42p (Rip1p), a non-essential nucleoporin that becomes essential when other transport proteins have been mutated. However, export of heat shock mRNA does not require the hnRNP protein, Npl3p. Otherwise, the requirements for export appear similar to those for mRNA under normal growth conditions. We are attempting to define the mechanisms responsible for this selective mRNA export. We have also observed selective nuclear protein import after stress: proteins with canonical NLSs are prevented from entering the nucleus. In contrast, a fraction of Npl3p leaves the nucleus after stress and can gradually return to the nucleus. This requires protein but not RNA synthesis, but the protein requirement is not Hsp70. We are attempting to define the basis for regulated protein import.

Coupling of 3’ mRNA processing with mRNA export: A screen for additional mutants required for heat shock mRNA export identified 3’ processing factors. This suggests that formation of the 3’ end of an mRNA may be linked to mRNA export. Evidence over the years suggests that mRNAs that receive their 3’ poly(A) tails by an abnormal mechanism are exported very inefficiently. We hypothesize that the process of 3’ cleavage and polyadenylation leaves some 3’ processing factors associated with the mRNA after cleavage and polyadenylation. Their presence is hypothesized to indicate that the mRNA has been generated by proper 3’ processing, marking the mRNA for efficient mRNA export. We now know that some of these 3’ factors shuttle between the nucleus and cytoplasm, a behavior consistent with a role in transport, and unlikely to be seen for nuclear mRNA processing factors that stay are required solely within the nucleus. We are attempting to determine how 3’ processing and mRNA export are coupled.

MICRO RNAs AND BREAST CANCER

Background: Breast cancer is a genetic disease which results from multiple mutations. Most of the mutations causing breast cancer are not inherited but occur during a personÕs lifetime. Mutations which lead towards breast cancer occur in genes encoding many classes of proteins which together guard our genome and maintain control over normal cell growth and behavior. There are many alternative combinations of mutations which can result in breast cancer. The status of the estrogen receptor in different tumors provides important information about treatment (should tamoxifen be used?) and is correlated with different prognoses. Thousands to tens of thousands of genes are expressed in most types of cells. and these patterns changes dramatically when cells become malignant. In fact, recent studies indicate that the pattern of expression of fewer than 100 genes might be a powerful tool for sorting breast cancers into categories with different optimal treatment strategies and prognoses. To understand how breast cancer develops, we need to catalog the changes in gene expression that occur during breast tumor development and progression, and determine how these changes modify cellular behavior.

Our cells contain many classes of RNA. Only one class, mRNA, encodes proteins. The machinery which reads mRNAs and directs protein synthesis contains two types of non-coding RNAs (e.g. rRNAs and tRNAs). U-snRNAs and snoRNAs are non-coding and involved in RNA processing. Recently, it was reported that our cells and those of most multicellular organisms contain a novel class of non-coding RNAs. These are micro RNAs, ~22 nucleotides long, and they act to regulate expression of protein-coding genes by forming imperfect double-stranded hybrids with target mRNAs of specific sequence. These microRNA/mRNA hybrids are loaded onto polyribosomes but translation is blocked by an unknown mechanism.

The objective of our study is the identification of micro RNAs expressed in normal human breast cells. We also wish to determine whether any of these are absent or expressed at an abnormal level in breast cancer cell lines as compared with normal breast cells. Such a finding would indicate that miRNAs have potential to contribute to development and progression of breast cancer, and their patterns of expression could also be diagnostic of breast tumor category. Subsequent studies will be directed at identifying the mRNAs targeted by specific microRNAs and study of how micro RNA regulation of these mRNAs affects cellular growth properties and oncogenesis.