Identification of the locations of the methyl groups in 18 S ribosomal RNA from Xenopus laevis and man

The 18 S ribosomal RNA from a variety of vertebrate species contains some 40 to 47 methyl groups. The majority of these are 2'-O-ribose substituents; the remaining few are on bases. Several lines of evidence have permitted the identification of the precise locations of the methyl groups in the primary structure of 18 S ribosomal RNA of Xenopus laevis and man. Digestion of RNA with T1 ribonuclease, followed by analysis of the methylated oligonucleotides yielded data on sequences immediately surrounding the methyl groups. Preparative hybridization of X. laevis 18 S ribosomal RNA restriction fragments of ribosomal DNA, followed by fingerprinting analysis on RNA recovered from the hybrids, allowed each methylated oligonucleotide to be mapped to a specific region within 18 S ribosomal RNA. The data on RNA oligonucleotides were correlated with Xenopus ribosomal DNA sequence data in the regions defined by the mapping experiments to identify the precise locations of most of the methyl groups in the X. laevis 18 S RNA sequence. The remaining uncertainties in Xenopus were solved with the aid of data from ribonuclease A fingerprints and, in a few instances, relevant oligonucleotide or sequence data from other laboratories. The locations of most of the methyl groups in human 18 S ribosomal RNA were deduced from the high degree of correspondence between methylated oligonucleotides from human and X. laevis 18 S RNA, together with knowledge of the human 18 S ribosomal DNA sequence. The remaining methylation sites in human 18 S RNA were located with assistance from relevant published comparative data. In the aligned sequences, human and other mammalian 18 S RNA are methylated at all the same positions as in X. laevis, and there are seven additional 2'-O-methylation sites in mammalian 18 S RNA. Further features of the methyl group distribution are briefly reviewed.     

Ribonucleic acids of human milk

The milk feeding is the most essential process laying the foundation of human health at the postnatal development. However little is known about nucleic acids secreted into mother's milk during lactation. In order to investigate the composition and abundance of human milk NA we adapted the conventional isolation method to achieve high yield of total nucleic acids from milk samples. Concentration of total NA in milk samples of different donors varies from 20 to 68 mkg/ml at early stages of lactation. The average concentration tends to fall down to the end of lactation. The chain length of the major forms of NA varies from mononucleotides up to approximately 100 bases. Compositions of milk oligonucleotides are similar in samples of different donors. Major milk oligonucleotides are formed of RNA. Human milk contains the set of long-chain oligonucleotides with a developed secondary structure. Sequences of some oligo-RNAs correspond to the 3'-part of 5.8 S human ribosomal RNA and to the 3'-parts of tRNAVal and tRNATyr Primary structures of some others oligo-RNAs were related to fragments of human 18S and 28S rRNAs.     

Sequence determination of the 3'' terminal T1 oligonucleotide of 18S ribosomal RNA.

We have reexamined the primary structure of the 3' terminal oligonucleotide of 18S RNA from chicken fibroblasts and have shown, contrary to previously published results that this extremity G-A-U-C-A-U-U-AOH is identical to that of the rabbit, drosophila and bombyx. Furthermore the electrophoretic mobility and composition of the 3' terminal oligonucleotides of 18S RNA from rat and human cells are similar to that of other RNAs and show that the identity of structure for this region of 18S RNA extends to include all tested species between yeast and man. This finding reveals a marked degree of evolutionary constraint on the structure of this region.     

Ribonucleic Acids of Human Milk

The milk feeding is the most essential process laying the foundation of human health at the postnatal development. However little is known about nucleic acids secreted into mother's milk during lactation. In order to investigate the composition and abundance of human milk NA we adapted the conventional isolation method to achieve high yield of total nucleic acids from milk samples. Concentration of total NA in milk samples of different donors varies from 20 to 68 mkg/ml at early stages of lactation. The average concentration tends to fall down to the end of lactation. The chain length of the major forms of NA varies from mononucleotides up to approximately 100 bases. Compositions of milk oligonucleotides are similar in samples of different donors. Major milk oligonucleotides are formed of RNA. Human milk contains the set of long‐chain oligonucleotides with a developed secondary structure. Sequences of some oligo‐RNAs correspond to the 3′‐part of 5.8 S human ribosomal RNA and to the 3′‐parts of tRNAVal and tRNATyr Primary structures of some others oligo‐RNAs were related to fragments of human 18S and 28S rRNAs.     

Sequence homologies in mammalian 5.8S ribosomal RNA.

Oligonucleotide products of complete pancreatic or T1 RNase digestion or partial T1 RNase digestion of HeLa cell (human) and MPC-11 cell (mouse) 5.8S rRNA are identical with those obtained from Novikoff hepatoma (rat) 5.8S rRNA except for minor differences at the termini. pCp is the only major 5' terminus of both human and mouse RNAs; both pGp and pCp 5' termini were found in rat 5.8S RNA. Furthermore, HeLa cells contain C-U-U at the 3' end rather than the C-U terminus of mouse and rat. The results indicate that the nucleotide sequence has been highly conserved during the evolution of mammals and suggest that, as reported for 5S rRNA, this sequence is essentially constant throughout the Mammalia.     

The nuclear 5S RNAs from chicken, rat and man. U5 RNAs are encoded by multiple genes

Preparations of chicken, rat and human nuclear 5S RNA contain two sets of molecules. The set with the lowest electrophoretic mobility (5Sa) contains RNAs identical or closely related to ribosomal 5S RNA from the corresponding animal species. In HeLa cells and rat brain, we only detected an RNA identical to the ribosomal 5S RNA. In hen brain and liver, we found other species differing by a limited number of substitutions. The results suggest that mutated 5S genes may be expressed differently according to the cell type. The set with the highest mobility corresponds to U5 RNA. In both rat brain and HeLa cells, U5 RNA was found to be composed of 4 and 5 different molecules respectively (U5A, U5B1-4) differing by a small number of substitutions or insertions. In hen brain, no U5B was detected but U5A' differing from U5A by the absence of the 3'-terminal adenosine. All the U5 RNAs contain the same set of modified nucleotides. They also have the same secondary structure which consists of two hairpins joined together by a 17 nucleotide long single-stranded region. The 3' half of the molecule has a compact conformation. Together, the results suggest that U5 RNAs are transcribed from a multigene family and that mutated genes may be expressed as far as secondary structure is conserved. The conformation of U5 RNA is likely to be related to its function and it is of interest to mention that several similarities of structure are found between U5 and U1A RNA.     

Demonstration of the "5-8 S" ribosomal sequence in HeLa cell ribosomal precursor RNA

HeLa cell “5.8 S” ribosomal RNA was digested with T₁ ribonuclease and the digestion products were characterized. In particular several hexa-, or larger, oligonucleotides were well fractionated by T₁ ribonuclease plus alkaline phosphatase fingerprints. The sequences of these large products were determined. The same large products were identified in fingerprints of “native” 28 S RNA, that is, 28 S RNA to which 5.8 S RNA is attached. The products were demonstrably absent in fingerprints of heat-denatured 28 S RNA, which lacks the 5.8 S fragment. The oligonucleotides were present in fingerprints of 32 S RNA, whether previously heated or not. One of the largest 5.8 S oligonucleotides contains an alkali-stable (2′-O-methylated) dinucleotide, Gm-C. This product was identified in fingerprints of methyl-labelled 45 S RNA. These findings prove that the 5.8 S ribosomal sequence is present within HeLa cell ribosomal precursor RNA. In addition to the methylated nucleotide, two pseudouridylate residues were discovered in HeLa cell 5.8 S RNA.     

Role of the 5.8S rRNA in ribosome translocation.

Studies on the inhibition of protein synthesis by specific anti 5.8S rRNA oligonucleotides have suggested that this RNA plays an important role in eukaryotic ribosome function. Mutations in the 5. 8S rRNA can inhibit cell growth and compromise protein synthesis in vitro . Polyribosomes from cells expressing these mutant 5.8S rRNAs are elevated in size and ribosome-associated tRNA. Cell free extracts from these cells also are more sensitive to antibiotics which act on the 60S ribosomal subunit by inhibiting elongation. The extracts are especially sensitive to cycloheximide and diphtheria toxin which act specifically to inhibit translocation. Studies of ribosomal proteins show no reproducible changes in the core proteins, but reveal reduced levels of elongation factors 1 and 2 only in ribosomes which contain large amounts of mutant 5.8S rRNA. Polyribosomes from cells which are severely inhibited, but contain little mutant 5.8S rRNA, do not show the same reductions in the elongation factors, an observation which underlines the specific nature of the change. Taken together the results demonstrate a defined and critical function for the 5.8S rRNA, suggesting that this RNA plays a role in ribosome translocation.     

Comparative studies of the primary structures of ribosomal RNAs of several eukaryotic cell lines by the fingerprinting method.

Comparisons of the primary structures of 18S and 28S ribosomal RNAs of man, rat, mouse and chicken were made by two-dimensional fractionation including electrophoresis at pH 3.5 and homochromatography. All large T1 oligonucleotides were recovered from the different fingerprints and their radioactivity was measured. They were then hydrolysed with pancreatic RNase and the pancreatic products were digested with alkali to determine their base composition and detect modified residues. Finally, residues bearing a modification on the ribose were analysed by hydrolyses with snake venom and spleen phosphodiesterases. For the 18A RNAs 23, 27, 26, 24 oligonucleotides, whose lengths range from 22 to 10 residues, were analyzed respectively for man, rat, mouse and chicken. Among these, 14 are identical in the four species, two at least are common to man, rat, mouse but differ by the presence of A-Cps in chicken spot 4' instead of A-Up in spot 4 and A2-Gp in chicken spot 14 instead of A2-Gp in spot 13. For the 28S RNAs of man, rat, mouse and chicken, 20, 19, 21 and 22 oligonucleotides ranging in length from 27 to 12 residues were analyzed. 11 of them are common to the four species; 4 of them are found in man, rat, mouse and one of these (spot 1) has a corresponding spot in chicken from which it differs only by the existence of A3-Up instead of A2-Up. Another mammalian oligonucleotide (spot 6) differs from its homologous chicken spot (spot 6') bytwo point mutations. The same modified residues as found by Khan and Maden in man, chicken, and xenopus, have been found in rat and mouse. Moreover when these modified residues are common to several species they are found within an identical nucleotide sequence, as can be seen in the case of spots 1, 3, 9, 11 of 18S RNAs and 4, 7, 13 for 28S RNAs. The number of differences observed between the ribosomal RNAs of the four species were compared to the number of differences observed in the same species for several proteins, globins alpha and beta, insulin, cytochrome C and lysozyme.     

Isolation and characterization of a novel ribonucleoprotein particle: large structures contain a single species of small RNA

Rat liver coated vesicle preparations were frequently found to contain small ovoid bodies, which resembled coated vesicles in morphology. We have purified these bodies to homogeneity using sucrose density gradients and preparative agarose gel electrophoresis. When negatively stained and viewed by electron microscopy, the purified structures display a very distinct and complex morphology, resembling the multiple arches which form cathedral vaults. They measure 35 X 65 nm and are therefore considerably larger than ribosomes. When subjected to SDS PAGE, these structures, which we refer to as vaults, appear to contain several minor and five major species: Mr 210,000, 192,000, 104,000, 54,000, and 37,000. One of these (Mr 104,000) greatly predominates, accounting for greater than 70% of the total Coomassie Brilliant Blue-staining protein. Another major species of Mr 37,000 has been identified as a species of small RNA of unusual base composition (adenosine 12.0%, guanosine 29.7%, uridine 30.9%, and 27.4% cytidine), which migrates as a single species in urea PAGE between the 5S and 5.8S ribosomal standards, containing approximately 140 bases. Although the RNA constitutes only 4.6% of the entire structure, the large size of the particle requires that each one contains approximately 9 molecules of this RNA. Antibodies prepared against the entire particle are largely specific for the major (Mr 104,000) polypeptide species. Although they do not directly react with the RNA constituent on Western blots, these antibodies immunoprecipitate a 32P-labeled RNA of identical size from metabolically-labeled rat hepatoma cells. Vaults are observed in partially purified fractions from human fibroblasts, murine 3T3 cells, glial cells, and rabbit alveolar macrophages. It therefore appears that these novel ribonucleoprotein structures are broadly distributed among different cell types. The function of vaults is at present unknown.     

Probing the conformational changes in 5.8S, 18S and 28S rRNA upon association of derived subunits into complete 80S ribosomes.

The participation of 18S, 5.8S and 28S ribosomal RNA in subunit association was investigated by chemical modification and primer extension. Derived 40S and 60S ribosomal subunits isolated from mouse Ehrlich ascites cells were reassociated into 80S particles. These ribosomes were treated with dimethyl sulphate and 1-cyclohexyl-3-(morpholinoethyl) carbodiimide metho-p-toluene sulfonate to allow specific modification of single strand bases in the rRNAs. The modification pattern in the 80S ribosome was compared to that of the derived ribosomal subunits. Formation of complete 80S ribosomes altered the extent of modification of a limited number of bases in the rRNAs. The majority of these nucleotides were located to phylogenetically conserved regions in the rRNA but the reactivity of some bases in eukaryote specific sequences was also changed. The nucleotides affected by subunit association were clustered in the central and 3'-minor domains of 18S rRNA as well as in domains I, II, IV and V of 5.8/28S rRNA. Most of the bases became less accessible to modification in the 80S ribosome, suggesting that these bases were involved in subunit interaction. Three regions of the rRNAs, the central domain of 18S rRNA, 5.8S rRNA and domain V in 28S rRNA, contained bases that showed increased accessibility for modification after subunit association. The increased reactivity indicates that these regions undergo structural changes upon subunit association.     

Oligonucleotide sequences of pancreatic and T1 ribonuclease digests of 5S ribosomal RNA from mouse cells.

Oligonucleotides produced by complete pancreatic and T1 RNase digestion of 5S ribosomal RNA from a mouse hepatoma, MH 134, have been separated with two-dimensional electrophoresis and their nucleotide sequences determined. Except for the presence of a 5'-terminal diphosphate, these nucleotide sequences were identical with those of KB cells, confirming the identity of the primary structure of 5S RNA between these animals. Both oligonucleotide patterns produced with these enzymes from 5S RNA of the liver were also identical with those of the hepatoma. All these agree with the strong conservation of 5S RNA genes in animal species. However, when 5S ribosomal RNA was extracted from ribosomes which were prepared from microsomal pellet, pancreatic RNase digest contained two trinucleotides (A-G-Cp and G-A-Cp) that were not found in 5S RNA prepared with a one-step procedure. It was concluded that different isolation procedure might indeed cause artifactual fragments on enzymatic digestion due to internal nicks produced during isolation. The significance of 5'-terminal diphosphate in relation to the biosynthesis of 5S ribosomal RNA is also discussed.     

Analysis of large specific T1 oligonucleotides of 17S and 25S ribosomal RNAs from Saccharomyces cerevisiae.

The primary structure of 17S and 25S ribosomal RNAs from Saccharomyces cerevisiae has been analysed by two-dimensional fractionation of T1 oligonucleotides. This method consists of an electrophoresis at pH 3.5 followed by a homochromatography on DEAE-cellulose plates. After the second dimension, the large T1 oligonucleotides were hydrolyzed by pancreatic RNAse, followed by alkaline hydrolysis of the pancreatic products. By fractionating a mixture of tritiated HeLa cell ribosomal RNAs and 32 P yeast cell ribosomal RNAs, two autoradiographs were obtained; one corresponding to the 32P labelled material and the other to the tritiated labelled material. By superposition of the two autoradiographs, the mobility of the various T1 oligonucleotides can be accurately compared and it is shown that yeast 17S rRNA and human 18S rRNA have in common 5 large oligonucleotides and that yeast 25S rRNA and human 28S rRNA have 4 identical oligonucleotides.     

Internal transcribed spacer 1 of the yeast precursor ribosomal RNA. Higher order structure and common structural motifs

The higher order structure of the first internal transcribed spacer between the 18S and the 5.8S rRNA sequences in the Saccharomyces cerevisiae precursor ribosomal RNA has been investigated. Sites of potential base pairing in the RNA region have been determined by using a combination of enzymatic and chemical structure sensitive probes. Data generated have been used to evaluate secondary structure models predicted by minimum free energy calculations. Several alternative suboptimal structures were also evaluated. The derived model contains several stable hairpins. Theoretical secondary structural models for the corresponding RNA region from S. carlsbergensis, S. pombe, N. crassa, X. laevis, and mung bean have also been derived from identical calculations and assumptions. Certain structural motifs appear to be conserved despite extensive divergence in the base sequence. The yeast model should be a useful prototype for investigation of structure and function of precursor ribosomal RNA molecules.     

3'-terminal processing of ribosomal RNA precursors in mammalian cells.

The 3'-terminal structures of ribosomal 28S RNA and its precursors from rat and mouse were analyzed by means of periodate oxidation followed by reduction with 3H-borohydride. 3'-terminal labeled nucleoside derivatives produced by RNase T2 digestion were determined by thin-layer chromatography and oligonucleotides generated by RNase T1 digestion were analyzed by DEAE-Sephadex chromatography. In the rat, the major 3'-terminal sequences of ribosomal 28S RNA, nucleolar 28S, 32S, 41S, and 45S RNAs were YGUoh, GZ2Uoh, GZ12Uoh, GZ2Uoh, and GZ7Goh, respectively, whereas in the mouse corresponding sequences were YGUoh, GZ1,2, or 3Uoh, Goh, Uoh and GZ 13Uoh, respectively. (Y: pyrimidine nucleoside, Z: any nucleoside other than guanosine) These results suggest that a "transcribed spacer" sequence is present at the 3'-terminus of the 45S pre-ribosomal RNA, which is gradually removed during the steps of processing.     

Probing the structure of mouse Ehrlich ascites cell 5.8S, 18S and 28S ribosomal RNA in situ.

The secondary structure of mouse Ehrlich ascites 18S, 5.8S and 28S ribosomal RNA in situ was investigated by chemical modification using dimethyl sulphate and 1-cyclohexyl-3-(morpholinoethyl) carbodiimide metho-p-toluene sulphonate. These reagents specifically modify unpaired bases in the RNA. The reactive bases were localized by primer extension followed by gel electrophoresis. The three rRNA species were equally accessible for modification i.e. approximately 10% of the nucleotides were reactive. The experimental data support the theoretical secondary structure models proposed for 18S and 5.8/28S rRNA as almost all modified bases were located in putative single-strand regions of the rRNAs or in helical regions that could be expected to undergo dynamic breathing. However, deviations from the suggested models were found in both 18S and 28S rRNA. In 18S rRNA some putative helices in the 5'-domain were extensively modified by the single-strand specific reagents as was one of the suggested helices in domain III of 28S rRNA. Of the four eukaryote specific expansion segments present in mouse Ehrlich ascites cell 28S rRNA, segments I and III were only partly available for modification while segments II and IV showed average to high modification.     

DNA primase of human mitochondria is associated with structural RNA that is essential for enzymatic activity

DNA primase isolated from human mitochondria sediments in glycerol density gradients at 30S and 70S. These unusually high sedimentation coefficients are a result of association of the primase activity with RNA. Treatment of primase with nuclease not only affects its sedimentation behavior, but also inactivates the primase activity. The major RNA species that cofractionates with primase activity is shown by direct sequence analysis to be cytosolic 5.8S ribosomal RNA (rRNA). Specific degradation of endogenous 5.8S rRNA using ribonuclease H and oligonucleotides complementary to 5.8S rRNA results in reduction of primase activity. Other small RNAs may play a structural role in the formation of an active DNA primase complex.     

Nucleotide sequences of the 5.8S rRNAs of a mollusc and a porifer, and considerations regarding the secondary structure of 5.8S rRNA and its interaction with 28S rRNA

We report the primary structures of the 5.8 S ribosomal RNAs isolated from the sponge Hymeniacidon sanguinea and the snail Arion rufus. We had previously proposed (Ursi et al., Nucl. Acids Res. 10, 3517-3530 (1982)) a secondary structure model on the basis of a comparison of twelve 5.8 S RNA sequences then known, and a matching model for the interaction of 5.8 S RNA with 26 S RNA in yeast. Here we show that the secondary structure model can be extended to the 25 sequences presently available, and that the interaction model can be extended to the binding of 5.8 S RNA to the 5'-terminal domain of 28 S (26 S) RNA in three species.