Small molecule ligand-RNA interactions have the potential to influence gene expression at a variety of steps and in a number of ways. Here, we demonstrate that such interactions are sufficiently stable to inhibit translation of eukaryotic mRNAs in vitro and in vivo. Inhibition is only observed when the 59 UTR of the mRNA is targeted, and the response is proportional to the number of binding sites within this region. We find that small molecule ligand-RNA interactions can be sufficiently stable to prevent 80S ribosome assembly on an mRNA template. The ability to specifically ablate expression of a defined mRNA with a small molecule ligand demonstrates proof of principle for pharmacological targeting aimed at controlling translation of specific mRNAs.

The National Cancer Institute (NCI) Human Tumor Cell Line Anti-Cancer Drug Screen has evaluated the cytotoxicity profiles of a large number of synthetic compounds, natural products, and plant extracts on 60 different cell lines. The data for each compound/extract can be assessed for similarity of cytotoxicity pattern, relative to a given test compound, using an algorithm called COMPARE. In applying a chemical biology approach to better understand the mechanism of eukaryotic protein synthesis, we used these resources to search for novel inhibitors of translation. The cytotoxicity profiles of 31 known protein synthesis inhibitors were used to identify compounds from the NCI database with similar activity profiles. Using this approach, two natural products, phyllanthoside and nagilactone C, were identified and characterized as novel protein synthesis inhibitors. Both compounds are specific for the eukaryotic translation apparatus, function in vivo and in vitro, and interfere with translation elongation. Our results demonstrate the feasibility of utilizing cytotoxicity profiles to identify new inhibitors of translation.

No abstract

Matrix metalloproteinases (MMPs) are enzymes thought to be involved in tumor invasion. We hypothesized that MMP-2 and MMP-11 overexpression was associated with the aggressiveness of ovarian carcinoma. This study was performed on samples from 100 patients with stage III ovarian carcinomas treated surgically between 1990 and 2000. Immunohistochemical staining was performed on ovarian tumors and peritoneal implants using monoclonal antibodies. Overexpression was defined as more than 10% of cells expressing the marker. Multivariate analyses showed that only MMP-2 overexpression by cancer cells in peritoneal implants was associated with a significant risk of death by disease (hazard ratio, 2.65; 95% confidence interval, 1.41-4.97; P =.003). MMP-11 overexpression was not predictive of survival. These results suggest that MMP-2 overexpression by cancer cells in peritoneal implants and not in the primary ovarian cancer is predictive of ovarian cancer prognosis and more likely reflects the presence of particularly aggressive clones of cancer cells.

The use of small molecule inhibitors in the study of cellular processes is a powerful approach to understanding gene function. During the course of a high throughput screen for novel inhibitors of eukaryotic translation, we identified a number of nucleic acid binding ligands that showed activity in our assay. When tested on a panel of mRNA transcripts displaying different modes of translation initiation, these ligands showed a range of biological activities -with some inhibiting both cap-dependent and internal initiation and others preferentially blocking internal initiation. We used this information to identify a novel threading intercalator that inhibits Hepatitis C virus internal initiation.