RNA-RNA interactions in the binding site of protein L24 on 23S ribosomal RNA of Escherichia coli: 1. Evidence for their occurrence between widely separated sequence regions.

Protein L24, from the Escherichia coli ribosome, protects a large region of 23S RNA against ribonuclease digestion. This protected RNA consists of a series of non-contiguous subfragments encompassing about 480 nucleotides at the 5' -end of 23S RNA (1). The present work demonstrates that this RNA moiety remains intact, after removal of protein L24, indicating that the subfragments are maintained together by RNA-RNA interactions. Using a urea washing procedure, the weakly bound RNA subfragments were selectively removed leaving more strongly interacting subfragments that were identified, by gel electrophoresis and oligonucleotide fingerprinting, and shown to derive from widely separated sequence regions.     

Evolutionary relationships amongst archaebacteria. A comparative study of 23 S ribosomal RNAs of a sulphur-dependent extreme thermophile, an extreme halophile and a thermophilic methanogen

The 23 S RNA genes representative of each of the main archaebacterial subkingdoms, Desulfurococcus mobilis an extreme thermophile, Halococcus morrhuae an extreme halophile and Methanobacterium thermoautotrophicum a thermophilic methanogen, were cloned and sequenced. The inferred RNA sequences were aligned with all the available 23 S-like RNAs of other archaebacteria, eubacteria/chloroplasts and the cytoplasm of eukaryotes. Universal secondary structural models containing six major structural domains were refined, and extended, using the sequence comparison approach. Much of the present structure was confirmed but six new helices were added, including one that also exists in the eukaryotic 5.8 S RNA, and extensions were made to several existing helices. The data throw doubt on whether the 5' and 3' ends of the 23 S RNA interact, since no stable helix can form in either the extreme thermophile or the methanogen RNA. A few secondary structural features, specific to the archaebacterial RNAs were identified; two of these were supported by a comparison of the archaebacterial RNA sequences, and experimentally, using chemical and ribonuclease probes. Seven tertiary structural interactions, common to all 23 S-like RNAs, were predicted within unpaired regions of the secondary structural model on the basis of co-variation of nucleotide pairs; two lie in the region of the 23 S RNA corresponding to 5.8 S RNA but they are not conserved in the latter. The flanking sequences of each of the RNAs could base-pair to form long RNA processing stems. They were not conserved in sequence but each exhibited a secondary structural feature that is common to all the archaebacterial stems for both 16 S and 23 S RNAs and constitutes a processing site. Kingdom-specific nucleotides have been identified that are associated with antibiotic binding sites at functional centres in 23 S-like RNAs: in the peptidyl transferase centre (erythromycin-domain V) the archaebacterial RNAs classify with the eukaryotic RNAs; at the elongation factor-dependent GTPase centre (thiostrepton-domain II) they fall with the eubacteria, and at the putative amino acyl tRNA site (alpha-sarcin-domain VI) they resemble eukaryotes. Two of the proposed tertiary interactions offer a structural explanation for how functional coupling of domains II and V occurs at the peptidyl transferase centre. Phylogenetic trees were constructed for the archaebacterial kingdom, and for the other two kingdoms, on the basis of the aligned 23 S-like RNA sequences.(ABSTRACT TRUNCATED AT 400 WORDS)     

Identification of cis-Acting Nucleotides and a Structural Feature in West Nile Virus 3′-Terminus RNA That Facilitate Viral Minus Strand RNA Synthesis

The 3′-terminal nucleotides (nt) of West Nile virus (WNV) genomic RNA form a penultimate 16-nt small stem-loop (SSL) and an 80-nt terminal stem-loop (SL). These RNA structures are conserved in divergent flavivirus genomes. A previous in vitro study using truncated WNV 3′ RNA structures predicted a putative tertiary interaction between the 5′ side of the 3′-terminal SL and the loop of the SSL. Although substitution or deletion of the 3′ G (nt 87) within the SSL loop, which forms the only G-C pair in the predicted tertiary interaction, in a WNV infectious clone was lethal, a finding consistent with the involvement in a functionally relevant pseudoknot interaction, extensive mutagenesis of nucleotides in the terminal SL did not identify a cis-acting pairing partner for this SSL 3′ G. However, both the sequence and the structural context of two adjacent base pairs flanked by symmetrical internal loops in the 3′-terminal SL were shown to be required for efficient viral RNA replication. Nuclear magnetic resonance analysis confirmed the predicted SSL and SL structures but not the tertiary interaction. The SSL was previously reported to contain one of three eEF1A binding sites, and G87 in the SSL loop was shown to be involved in eEF1A binding. The nucleotides at the bottom part of the 3′-terminal SL switch between 3′ RNA-RNA and 3′-5′ RNA-RNA interactions. The data suggest that interaction of the 3′ SL RNA with eEF1A at three sites and a unique metastable structural feature may participate in regulating structural changes in the 3′-terminal SL.     

Region of intermolecular complementarity in Escherichia coli 16S rRNA,mRNA, and tRNA molecules

The results of computer analysis of complementarity regions in the sequences of E. coli 16S rRNA, mRNAs and tRNAs are reported in this article. The potential regions of intermolecular RNA-RNA hybridization, or clinger fragments, in 16S rRNA, which are complementary to the sites frequently occurring in mRNAs and tRNAs, were found. Major clinger fragments on 16S rRNA are universal for genes that belong to different functional groups. Our results show there are adaptations of the structural organization of the 16S rRNA molecule to messenger and transport RNA sequences. RNA interaction with clinger fragments may contribute to upregulation of translation process through increasing the local concentration of mRNAs and tRNAs in the vicinity of the ribosome and their proper positioning, as well as decrease the efficiency of translation through non-specific mRNA-16SrRNA interactions.     

Probing the phosphates of the Escherichia coli ribosomal 16S RNA in its naked form, in the 30S subunit, and in the 70S ribosome

Ethylnitrosourea is an alkylating reagent which preferentially modifies phosphates in nucleic acids. It was used to map phosphates in naked Escherichia coli 16S rRNA engaged in tertiary interactions through hydrogen bonds or ion coordination. Of the phosphates, 7% are found involved in such interactions, and 57% of them are located in loops or interhelical regions, where they are involved in maintaining local intrinsic structures or long-distance tertiary interactions. The other phosphates (43%) are found in helical regions. These phosphates often occur at the proximity of bulged nucleotides or in irregular helices containing noncanonical base pairs (and bulges) and are assumed to bind cations in order to neutralize negative charges and to stabilize unusual phosphate backbone folding. In the 30S subunit, ENU allowed mapping of phosphates in contact with proteins. The RNA is not uniformly engaged in RNA/protein interactions. Regions 1-51, 250-310, 567-612, 650-670, and 1307-1382 are particularly buried whereas the 3'-terminal domain and the 5'-proximal region (nucleotides 53-218) are exposed. The conformation of 16S rRNA is not drastically affected by protein binding, but conformational adjustments are detected in several defined regions. They are found in the 5' domain (region 147-172), in the central domain (region 827-872), in the 3' major domain (nucleotides 955-956, 994, 1054, 1181, 1257, and 1262-1263), and in the 3'-terminal domain (around 1400). The 50S subunit shields clusters of phosphates located at the subunit interface. The most extensive protections are observed in the 3'-terminal domain (1490-1542), in the central region of the molecule (770-930), and in the upper 3' major domain.(ABSTRACT TRUNCATED AT 250 WORDS)     

Improved statistical methods reveal direct interactions between 16S and 23S rRNA

Recent biochemical studies have indicated a number of regions in both the 16S and 23S rRNA that are exposed on the ribosomal subunit surface. In order to predict potential interactions between these regions we applied novel phylogenetically-based statistical methods to detect correlated nucleotide changes occurring between the rRNA molecules. With these methods we discovered a number of highly significant correlated changes between different sets of nucleotides in the two ribosomal subunits. The predictions with the highest correlation values belong to regions of the rRNA subunits that are in close proximity according to recent crystal structures of the entire ribosome. We also applied a new statistical method of detecting base triple interactions within these same rRNA subunit regions. This base triple statistic predicted a number of new base triples not detected by pair-wise interaction statistics within the rRNA molecules. Our results suggest that these statistical methods may enhance the ability to detect novel structural elements both within and between RNA molecules.     

A three-dimensional model of domain III of the Escherichia coli small ribosomal subunit

A three-dimensional model of domain III (nucleotides 920 to 1395) of the 30S ribosomal subunit of E. coli is proposed. The data used as a guide in folding the secondary structure of the RNA into a tertiary structure are four long range RNA-RNA interactions proposed by us on the basis of experiments performed in this laboratory plus two sets of data from other laboratories: protein-RNA cross-linking sites for proteins S1, S3, S7, S10 and S12, and the interprotein distances determined by neutron scattering. The model is consistent with nearly all of the published experimental findings on the structure of domain III.     

Evidence for RNA-RNA cross-link formation in Escherichia coli ribosomes.

Evidence is presented in three separate cases for the formation of RNA-RNA cross-links in intact E. coli ribosomes and ribosomal subunits. The first case is a cross-link between the 18S and 13S regions of the 23S RNA, induced by ultraviolet irradiation. The second is a cross-link at the subunit interface, generated by the bifunctional reagent bis-(2-chloroethyl)-amine. The third example is a cross-link between sections O'-D and P-A of the 16S RNA, induced as in the first case by ultraviolet irradiation. The RNA-RNA cross-links can be identified as such, despite the complications introduced by concomitant RNA-protein cross-linking reactions. The experiments represent a first attempt to introduce RNA-RNA cross-linking into studies of the topographical organization of the RNA within the ribosome.     

Protein-independent folding pathway of the 16S rRNA 5' domain

Evolution of the ribosome from an RNA catalyst suggests that the intrinsic folding pathway of the rRNA dictates the hierarchy of ribosome assembly. To address this possibility, we probed the tertiary folding pathway of the 5' domain of the Escherichia coli 16S rRNA at 20 ms intervals using X-ray-dependent hydroxyl radical footprinting. Comparison with crystallographic structures and footprinting reactions on native 30S ribosomes showed that the RNA formed all of the predicted tertiary interactions in the absence of proteins. In 20 mM MgCl2, many tertiary interactions appeared within 20 ms. By contrast, interactions between H6, H15 and H17 near the spur of the 30S ribosome evolved over several minutes, likely due to mispairing of a central helix junction. The kinetic folding pathway of the RNA corresponded to the expected order of protein binding, suggesting that the RNA folding pathway forms the basis for early steps of ribosome assembly.     

Multiple crosslinks of proteins S7, S9, S13 to domains 3 and 4 of 16S RNA in the 30S particle.

Functionally active 70S ribosomes containing 4-thiouracil in place of uracil (substitution level 2%) were prepared by an in vivo substitution method. RNA-protein crosslinks were introduced by 366 nm photoactivation of 4-thiouracil in the purified 30S subunits. Seven single stranded M13 probes containing rDNA inserts complementary to domains 3 and 4 of 16S RNA were constructed. These inserts approximately 100 nucleotides long starting at nucleotide 868 and ending at the 3' OH terminus were used to select contiguous RNA sections. The proteins covalently linked to each selected RNA section were identified by 2D gel electrophoresis. Proteins S7, S9, S13 were shown to be efficiently crosslinked to multiple sites belonging to both domains.     

Escherichia coli 5S RNA A and B conformers. Characterisation by enzymatic and chemical methods

The structures of the two stable conformers of Escherichia coli 5 S RNA, the and B form, were compared. Information about the structures were obtained using the methods of limited enzymatic hydrolysis and chemical modification of accessible nucleotides. Base-specific modifications were performed for adenosines and cytidines using diethylpyrocarbonate and dimethylsulfate in combination with a strand-scission reaction at the modified site. Base-specific (RNase T1) as well as conformation-specific (nuclease S1, cobra venom nuclease) enzymes were employed for the limited enzymatic hydrolysis. Clear differences in the accessibility of the two 5 S RNA conformers to the enzymes and the chemical reagents were established and the regions with altered reactivities were localized in the 5 S RNA structure. The results are consistent with the disruption of the secondary structural interactions in helix II and partly in helices III and IV during the transition from the A to the B form. (The numbering of the helices is according to the generally accepted Fox and Woese model.) In addition some regions presumably involved in the tertiary structure are distorted. There is evidence, however, for the new formation of structural regions between two distant sites in the 5 S RNA B form. The results enable us to refine the existing 5 S RNA A-form model and provide insight into the structural dynamics that lead to the formation of the 5 S RNA B form.     

Action of Ribonuclease T1 on 30S Ribosomes of Escherichia coli and Its Role in Sequence Studies on 16S Ribonucleic Acid

Two large ribonucleic acid (RNA) fragments have been obtained from T1-RNase-treated 30S ribosomes of Escherichia coli. One fragment, about 475 nucleotides long, contains all the unique oligonucleotides found by Fellner and associates in sections of 16S RNA designated P, E, E', and K, and one-half the large oligonucleotides of section A. The other large fragment is about 300 nucleotides long and contains the oligonucleotides found in sections C, C', C'. The isolation of these large fragments seems to confirm the arrangement of sections within 16S RNA. There are also recovered from nuclease-treated ribosomes three small fragments, one (120 nucleotides long) from the 5' end, one (26 nucleotides long) from the 3' OH end of the chain, and another section (66 nucleotides long) from the middle of the 16S RNA chain. Small molecular weight material is also generated by nuclease treatment, and about half this material is derived from a region close to the 3' OH end of the 16S RNA chain. This indicates that the most accessible part of the rRNA of E. coli 30S ribosomes is a region 100 to 150 nucleotides long near the 3' end of the chain. A general scheme is proposed to explain the generation of the various-sized RNA products from the rRNA of the 30S ribosome.