Studies on HeLa cell nuclear DNA-like RNA by RNA-DNA hybridization.

Abstract

A class of rapidly labeled nuclear RNA distinct from ribosomal precursor RNA has recently been characterized in various types of animal cells.1-' This RNA has a base composition resembling DNA (substituting uracil for thynline) and is therefore clearly distinguishable from nucleolar ribosomal RNA precursor molecules. On the basis of sedimentation behavior after various treatments3' B and contour lengths in electron micrographs,2 it has been suggested that this RNA represents molecules of various lengths, including some polyribonucleotide chains as large as 107 molecular weight. We have used the abbreviation HS-nRNA (heterogeneously sedimenting nuclear RNA) to refer to this class of RNA obtained from HeLa cells.5' 6 The cytoplasmic mRNA of HeLa cells comprises molecules which sediment much more slowly than the HS-rnRNA,7' 8 although the base composition of both is DNA-like. It is therefore possible that cytoplasmic mRNA is derived from HS-nRNA by a specific cleavage mechanism(s). Such a mechanism, by which long polynucleotides are converted to specific shorter molecules, has been described in HeLa cells: 28S and 16S ribosomal RNA derive from a 45S ribosomal precursor molecule in the nucleolus of the cell.6' 9, 10 Work on nucleated duck erythroblasts (cells which primarily synthesize hemoglobin) suggests that the majority of HS-nRNA molecules in that cell never leave the nucleus but are apparently constantly synthesized and degraded.2' On the other hand, consideration of the kinetics of pulse-chase experiments have led Brown and Gurdon to suggest that in developing frog embryos at least some of the shorter DNA-like RNA may arise from cleavage of longer molecules. In Hela, cells, the rate of incorporation of H3-uridine into HS-nRNA was compared to what would be expected if the HS-nRNA served as a precursor to cytoplasmic mRNA.5 Those kinetic experiments gave no clear indication of whether any of the HS-nRNA might be converted to cytoplasmic mRNA. To explore further the relationship of these two classes of RNA molecules, we turned to the technique of RNA-DNA hybridization.1 It was anticipated that experiments utilizing competition hybridization would be especially useful. For example, if HS-nRNA or RNA derived from it never entered the cytoplasm, hybrid formation by radioactive HS-nRNA should not be affected by cytoplasmic mRNA. It was found, however, that while unlabeled HeLa cell cytoplasmic RNA would interfere with hybridization of labeled RNA molecules, no true competition for DNA sites by unlabeled RNA molecules could be demonstrated. Therefore, the extent to which nucleotide sequences in these two classes of DINTAlike RNA might be related could niot be determined from such experiments. Materials and Methods.-Labeling procedures and preparation of nucleic acids: Hela cells were grown, labeled, and fractionated into cytoplasmic and nuclear fractions as previously described.5 7t 10 The total nuclear RNA was extracted by a hot phenol-sodium dodecyl sulfate (phenol-SDS) procedure and subjected to zonal sedimentation.A 12 The RNA sedimenting faster than the 45S ribosomal precursor peak (approximately 50-IOO(S) was collected by ethanol pre-