Structure of 5 S ribosomal RNA from Escherichia coli: identification of kethoxal-reactive sites in the A and B conformations.

The topographies of the A and B conformers of free 5 S RNA have been examined using kethoxal as a probe of single-stranded, accessible guanine residues. Each of the kethoxal-reactive guanines has been identified using diagonal electrophoresis, and the relative rate of modification at each site has been studied.Free 5 S RNA in the A form has several reactive guanines in addition to G₁₃ and G₄₁, which are the only two available for reaction in the intact 50 S ribosomal subunit (Noller &; Herr, 1974). The relative reactivities of these sites are G₄₁ ? G₁₃ > G₆₉ > G₂₄ > G₈₆ > G₁₀₇ > G₁₆, G₂₃, G₄₄. Modification at G₂₃ and G₄₄ reaches maximum values of only about 0.05 mol per mol 5 S RNA, suggesting that these residues are unreactive in the major conformer of the A form population. These results are compatible with a secondary structure model based on phylogenetic sequence conservation (Fox &; Woese, 1975), but imply that 12 of the 18 unpaired guanines in this model are involved in further molecular interactions.The modification pattern of the B conformer demands a different base-pairing arrangement and shows that the B form contains less structure than the A form. The relative reactivities in the B form are G₁₃ > G₁₀₂ > G₁₆ > G₂₄, G₄₄ > G₆₁, G₁₀₀ > G₂₃, G₅₁, G₁₀₇ > G₅₄, G₅₆. Several sites show plateaux at submolar modification levels, indicating the existence of some conformational heterogeneity in preparations of the B form of 5 S RNA. Heat-denatured 5 S RNA appears to contain a mixture of conformers including the A and B form.These results place limitations on certain structural and functional models for 5 S RNA. For example, G₄₄, which has often been implicated in base-pairing with tRNA, is accessible in the B form but not in the A form. Yet the B form does not bind the 5 S RNA-specific ribosomal proteins, nor is there evidence for its existence in the ribosome.     

Accessibility of the 5 S RNA in yeast ribosomes

The accessibility of yeast 5 S RNA to modification by diethyl pyrocarbonate was compared in the free 5 S RNA molecule, 60 S subunits and whole ribosomes. All the reactive sites in the free RNA were eliminated or suppressed in ribosomes but two sites. A₅₁ and A₆₄, remained accessible and a slight reactivity was observed at four new sites (G₃₀, G₄₉, G₅₂ and A₇₂). Nucleotide sequences that have been implicated in initiator transfer RNA binding or subunit interactions are not accessible.     

Protection of specific sites in 23 S and 5 S RNA from chemical modification by association of 30 S and 50 S ribosomes.

The effect of 30S subunit attachment on the accessibility of specific sites in 5 S and 23 S RNA in 50 S ribosomal subunits was studied by means of the guanine-specific reagent kethoxal. Oligonucleotides surrounding the sites of kethoxal substitution were resolved and quantitated by diagonal electrophoresis. In contrast to the extensive protection of sites in 16 S RNA in 70 S ribosomes (Chapman &; Noller, 1977), only two strongly (approx. 90%) protected sites were detected in 23 S RNA. The nucleotide sequences at these sites are

in which the indicated kethoxal-reactive guanines (with K above them) are strongly protected by association of 30 S and 50 S subunits. The latter sequence has the potential to base-pair with nucleotides 816 to 821 of the 16 S RNA, a site which has been shown to be protected from kethoxal by 50 S subunits and essential for subunit association. Six additional sites in 23 S RNA are partially (30 to 50%) protected by 30 S subunits. One of these sequences,

is complementary to nucleotides 787 to 792 of 16 S RNA. a site which is also 50 S-protected and essential for association. Of the two kethoxal-reactive 5 S RNA sites in 50 S subunits, G₁₃ is partially protected in 70 S ribosomes. while G₄₁ remains unaffected by subunit association.The relatively small number of kethoxal-reactive sites in 23 S RNA that is strongly protected in 70 S ribosomes suggests that subunit association may involve contacts between single-stranded sites in 16 S RNA and 50 S subunit proteins or non-Watson-Crick interactions with 23 S RNA. in addition to the two suggested base-paired contacts.     

The role of the basic N-terminal region of protein L18 in 5S RNA-23S RNA complex formation.

Of the three proteins, L5, L18 and L25, which bind to 5S RNA, the former two effect the interaction of 5S RNA with 23S RNA. We have used trypsin as a probe to investigate the roles of the proteins in this RNA-RNA assembly, with the following results: (1) In complexes with 5S RNA, the highly basic N-terminal region of L18 is accessible to trypsin. This accessibility is unaffected by L25. However, its presence is essential for stimulating L5 binding. (2) In 5S RNA-protein-23S RNA complexes proteins L5 and L18 are both strongly resistant to proteolysis. (3) No 5S RNA-23S RNA complex formation occurs in the presence of L5 and the C-terminal L18 fragment. Two possible models for the mechanism of RNA-RNA assembly are proposed.     

Effect of 50 S subunit proteins L5, L18 and L25 on the fluorescence of 5 S RNA-bound ethidium bromide

Among the three Escherichia coli 50 S subunit proteins L5, L18 and L25, which have an affinity for 5 S RNA, only protein L18 exerts a strong effect on the fluorescence of 5 S RNA-ethidium bromide complexes, without changing the quantum yield of the fluorescence. Proteins L5 and L25, although they have little effect on the fluorescence, have a strong stabilizing influence on the 5 S RNA-L18 complex. The results are discussed in terms of the secondary and tertiary structures of 5 S RNA in relation to ribosomal protein binding.     

Letters to the editor: Accessibility of 5 S RNA in 50 S ribosomal subunits

Only two sites in 5 S RNA react with Kethoxal in 50 S ribosomal subunits. These two sites, G₁₃ and G₄₁, have previously been found to be accessible in free 5 S RNA. Nucleotide sequences which have been suggested as possible binding sites for the T-ψ-C-G loop of tRNA are not accessible.     

Alteration of 5S RNA conformation by ribosomal proteins L18 and L25.

The effects of ribosomal proteins L18, L25 and L5 on the conformation of 5S RNA have been studied by circular dichroism and temperature dependent ultraviolet absorbance. The circular dichroism spectrum of native 5S RNA is characterized in the near ultraviolet by a large positive band at 267 nm and a small negative band at 298 nm. The greatest perturbation in the spectrum was produced by protein L18 which induced a 20% increase in the 267 nm band and no change in the 298 nm band. By contrast, protein L25 caused a small decrease in both bands. No effect was observed with protein L5. Simultaneous binding of proteins L18 and L25 resulted in CD changes equivalent to the sum of their independent effects. The UV absorbance thermal denaturation profile of the 5S RNA L18 complex lacked the pre-melting behavior characteristic of 5S RNA. Protein L25 had no effect on the 5S RNA melting profile. We concluded that protein L18 increases the secondary, and possible the tertiary structure of 5S RNA, and exerts a minor stabilizing effect on its conformation while protein L25 causes a small decrease in 5S RNA secondary structure. The implications of these findings for ribosome assembly and function are discussed.     

Incorporation of 5S RNA into 16S-23S RNA complex

5S RNA as such is not incorporated into 16S-23S RNA complex formed under reconstitution condition. However, the addition of 50S ribosomal proteins, L5, L18 and L25/L15 results in its incorporation in stoichiometric amount. None of the proteins added individually is capable of incorporating 5S RNA into the complex. Of the different combinations in pairs that are possible out of the four proteins, the pairs L5, L18 and L15, L18 stimulate the incorporation to some extent. Of the four possible triplets, L5, L18, L25 or L5, L15, L18 is the most efficient for maximum incorporation of 5S RNA. The presence of all the four proteins is no more effective than the combinations of the three.     

Identification of Escherichia coli and Bacillus stearothermophilus ribosomal protein binding sites on Escherichia coli 5S RNA.

Summary E. coli [³²P]-labelled 5S RNA was complexed with E. coli and B. stearothermophilus 50S ribosomal proteins. Limited T₁ RNase digestion of each complex yielded three major fragments which were analysed for their sequences and rebinding of proteins. The primary binding sites for the E. coli binding proteins were determined to be sequences 18 to 57 for E-L5, 58 to 100 for E-L18 and 101 to 116 for E-L25. Rebinding experiments of purified E. coli proteins to the 5S RNA fragments led to the conclusion that E-L5 and E-L25 have secondary binding sites in the section 58 to 100, the primary binding site for E-L18. Since B. stearothermophilus proteins B-L5 and BL22 were found to interact with sequences 18 to 57 and 58 to 100 it was established that the thermophile proteins recognize and interact with RNA sequences similar to those of E. coli. Comparison of the E. coli 5S RNA sequence with those of other prokaryotic 5S RNAs reveals that the ribosomal proteins interact with the most conserved sections of the RNA.Paper number 12 on structure and function of 5S RNA.Preceding paper: Wrede, P. and Erdmann, V.A. Proc. Natl. Acad. Sci. USA 74, 2706–2709 (1977)     

Sequence and conformation of 5 S RNA from Chlorella cytoplasmic ribosomes: comparison with other 5 S RNA molecules

The sequence of Chlorella cytoplasmic 5 S RNA has been determined by fingerprinting techniques. Partial digests were fractionated by a two-dimensional acrylamide gel electrophoretic technique, which indicates whether specific fragments are paired in the molecule. In this way, the four main base-paired regions of the molecule were located. The sequence of Chlorella cytoplasmic 5 S RNA is related to, but different from, that of other eukaryotic 5 S RNAs: it shows approximately 60% homology with vertebrate 5 S RNA and 40% homology with yeast 5 S RNA. In some respects the conformation of the molecule in solution is quite different from that of other sequenced 5 S RNAs: in particular, the highly accessible region found around position 40 in all other 5 S RNAs (prokaryotic and eukaryotic) does not exist in this molecule.     

Unique arrangement of coding sequences for 5 S, 5.8 S, 18 S and 25 S ribosomal RNA in Saccharomyces cerevisiae as determined by R-loop and hybridization analysis.

The arrangement of the coding sequences for the 5 S, 5.8 S, 18 S and 25 S ribosomal RNA from Saccharomyces cerevisiae was analyzed in λ-yeast hybrids containing repeating units of the ribosomal DNA. After mapping of restriction sites, the positions of the coding sequences were determined by hybridization of purified rRNAs to restriction fragments, by R-loop analysis in the electron microscope, and by electrophoresis of S₁ nuclease-treated rRNA/rDNA hybrids in alkaline agarose gels. The R-loop method was improved with respect to the length calibration of RNA/DNA duplexes and to the spreading conditions resulting in fully extended 18 S and 25 S rRNA R-loops. The qualitative results are: (1) the 5 S rRNA genes, unlike those in higher eukaryotes, alternate with the genes of the precursor for the 5.8 S, 18 S and 25 S rRNA; (2) the coding sequence for 5.8 S rRNA maps, as in higher eukaryotes, between the 18 S and 25 S rRNA coding sequences. The quantitative results are: (1) the tandemly repeating rDNA units have a constant length of 9060 ± 100 nucleotide pairs with one SstI, two HindIII and, dependent on the strain, six or seven EcoRI sites; (2) the 18 S and 25 S rRNA coding regions consist of 1710 ± 80 and 3360 ± 80 nucleotide pairs, respectively; (3) an 18 S rRNA coding region is separated by a 780 ± 70 nucleotide pairs transcribed spacer from a 25 S rRNA coding region. This is then followed by a 3210 ± 100 nucleotide pairs mainly non-transcribed spacer which contains a 5 S rRNA gene.     

Identification of the nucleotide sequences involved in the interaction between Escherichia coli 5 RNA and specific 50 S subunit proteins

Pancreatic RNase partial digests of ³²P-labelled 5 S RNA-protein complexes have been fractionated by electrophoresis on polyacrylamide gels. Specific fragments of the 5 S RNA molecule have been recovered from electrophoresis bands containing polynucleotide-protein complexes. These digestion-resistant complexes are only found if RNase treatment is carried out in the presence of at least one of the two 50 S subunit proteins L18 and L25, which are able to bind to 5 S RNA individually and specifically. The sequences of the isolated fragments have been determined. From the results, it can be concluded that sequence 69 to 120 and, possibly, sequence 1 to 11, are involved in the 5 S RNA-protein interactions which are responsible for the insertion of 5 S RNA in the 50 S subunit structure. Sequence 12 to 68, on the other hand, has no strong interactions with proteins L18 and L25. Each protein certainly binds to several nucleotide residues, which are not contiguous in the primary structure. In particular, good experimental evidence has been obtained in favour of the binding of protein L25 to two distant regions of the 5 S RNA molecule, which must have a bihelical secondary structure. The importance of the 5 S RNA conformation for its proper insertion in the 50 S subunit is thus confirmed.     

A ribonuclease-resistant region of 5S RNA and its relation to the RNA binding sites of proteins L18 and L25.

An RNA fragment, constituting three subfragments of nucleotide sequences 1-11, 69-87 and 89-120, is the most ribonuclease-resistant part of the native 5S RNA of Escherichia coli, at 0 degrees C. A smaller fragment of nucleotide sequence 69-87 and 90-110 is ribonuclease-resistant at 25 degrees. Degradation of the L25-5S RNA complex with ribonuclease A or T2 yielded RNA fragments similar to those of the free 5S RNA at 0 degrees C and 25 degrees C; moreover L25 remained strongly bound to both RNA fragments and also produced some opening of the RNA structure in at least two positions. Protein L18 initially protected most of the 5S RNA against ribonuclease digestion, at 0 degrees C, but was then gradually released prior to the formation of the larger RNA fragment. It cannot be concluded, therefore, as it was earlier (Gray et al., 1973), that this RNA fragment contains the primary binding site of L18.     

Nucleotide sequences of accessible regions of 23S RNA in 50S ribosomal subunits.

Nucleotide sequences around kethoxal-reactive guanine residues of 23S RNA in 50S ribosomal subunits have been determined. By use of the diagonal paper electrophoresis method )Noller, H.F. (1974), Biochemistry 13, 4694-4703), 41 ribonuclease T1 oligonucleotides, originating from about 25 sites, were identified and sequenced. These sites are single stranded and accessible in free 50S subunits, and are thus potential sites for interaction with functional ligands during protein synthesis. Examination of these sequences for potential intermolecular base-pairing reveals the following: (1) There are 19 possible complementary combinations between exposed sequences in 16S and 23S RNA containing more than 4 base pairs: 15 containing 5 base pairs and 4 containing 6 base pairs. Nine of these complementary combinations contain 16S RNA sequences which we have previously shown to be protected from kethoxall by 50S subunits (Chapman, N.M., and Noller, H.F. (1977), J. Mol. Biol. 109, 131-149). (2) One of the exposed sites in 23S RNA has a sequence which is complementary to the invariant GT psi CR sequence in tRNA.     

Physical studies of 5S RNA variants at position 66.

Two variants of the 5S RNA of E. coli have been examined by imino proton NMR spectroscopy, one of them a deletion of A66 (Christiansen, J., Douthwaite, S.R., Christensen, A. and Garrett, R.A. (1985) EMBO J. 4, 1019-1024) and the other a replacement of A66 with a C (Goringer, H.U. and Wagner, R. (1986) Biol. Chem. Hoppe-Seyler 367, 769-780). Both are of interest because the role the bulged A in helix II of 5S RNA is supposed to play in interactions with ribosomal protein L18. The data show that the structural perturbations that result from these mutations are minimal, and assign the resonances of some of the imino protons around position 66. Some mutations at or near position 66 greatly reduce the L18-dependent increase in the circular dichroism of 5S RNA at 267 nm first observed by Bear and coworkers (Bear, D.G., Schleich, T., Noller, H.F. and Garrett, R.A. (1977) Nucl. Acids Res. 4, 2511-2526).     

A common conformational feature in several prokaryotic and eukaryotic 5 S RNAs: a highly exposed, single-stranded loop around position 40

Psendomonas fluorescens, yeast and HeLa cells ³²P-labelled 5 S RNAs were submitted to partial hydrolysis with T₁, T₂ or pancreatic ribonucleases; the fragments were separated by two-dimensional acrylamide gel electrophoresis. First splits (obtained when only one cleavage takes place in the molecule) were found to occur essentially around position 40 in the sequence, as already demonstrated for Escherichia coli 5 S RNA. The existence in prokaryotic and eukaryotic 5 S RNAs of this very accessible region is thus proved. Eukaryotic 5 S RNAs also display a very accessible region around position 90 of the sequence.     

CONTROL OF 5S RNA SYNTHESIS DURING EARLY DEVELOPMENT OF ANUCLEOLATE AND PARTIAL NUCLEOLATE MUTANTS OF XENOPUS LAEVIS

Ribosomes of all eukaryotes contain a single molecule of 5S, 18S, and 28S RNA. In the frog Xenopus laevis the genes which code for 18S and 28S RNA are located in the nucleolar organizer, but these genes are not linked to the 5S RNA genes. Therefore the synthesis of the three ribosomal RNAs provides a model system for studying interchromosomal aspects of gene regulation. In order to determine if the synthesis of the three ribosomal RNAs are interdependent, the relative rate of 5S RNA synthesis was measured in anucleolate mutants (o/o), which do not synthesize any 18S or 28S RNA, and in partial nucleolate mutants (p^l-1/o), which synthesize 18S and 28S RNA at 25% of the normal rate. Since the o/o and p^l-1/o mutants have a complete and partial deletion of 18S and 28S RNA genes respectively, but the normal number of 5S RNA genes, they provide a unique system in which to study the dependence of 5S RNA synthesis on the synthesis of 18S and 28S RNA. Total RNA was extracted from embryos labeled during different stages of development and analyzed by polyacrylamide gel electrophoresis. Quite unexpectedly it was found that 5S RNA synthesis in o/o and p^l-1/o mutants proceeds at the same rate as it does in normal embryos. Furthermore, 5S RNA synthesis is initiated normally at gastrulation in o/o mutants in the complete absence of 18S and 28S RNA synthesis.     

Structural analyses of E. coli 5S RNA fragments, their associates and complexes with proteins L18 and L25.

The structure of Escherichia coli 5S RNA fragments 1-41 and 42-120 has been studied by the read-off gel sequencing technique using S1 nuclease and cobra venom RNase as probes. Comparison of the digestion patterns with those of reassociated and intact 5S RNA suggests that the structure of both fragments is very similar to that of the corresponding regions in the intact molecule. Six different fragments obtained by partial digestion with T1 RNase and S1 nuclease have been used for reconstitution of 5S RNA, its certain structural regions and complexes with ribosomal proteins L18 and L25 recognizes the double-helix consisting of nucleotides 79-97 (i.e. prokaryotic stem), whereas a loop-region around position 40 (possible positions 39-47) is involved in the interaction with protein L18.     

Precursors of 5 S ribosomal RNA in Bacillus subtilis

Bacillus subtilis 168 accumulates subnormal quantities of mature 5 S ribo-somal RNA in the presence of inhibitors of protein synthesis, such as chloramphenicol, or during pulse-labeling experiments. However, two RNA species, evidently precursors of m5 rRNA and therefore designated as p5_A and p5_B, do accumulate under these conditions. These RNA species are substantially longer than B. subtilis m5 rRNA: p5_A is about 179 nucleotides in length and p5_B is composed of approximately 152 nucleotides. The sum of p5_A, p5_B and m5 rRNA accumulating in the absence of protein synthesis, less excess chain length associated with p5_A and p5_B, equals the expected quantities of m5 rRNA in growing cells. p5_A and p5P_B both contain all t₁ RNase-generated oligonucleotides characteristic of m5 rRNA plus additional sequences. At least the 5′ termini of p5_A and p5_B differ from that of m5. If chloramphenicol is removed from a culture in which p5_A and p5_B have accumulated and further RNA synthesis is inhibited, then a quantitative reciprocal loss of p5_A and p5_B occurs as m5 rRNA accumulates. No evidence suggests any p5_A to p5_B transition under these conditions.