Double-stranded RNA adenosine deaminases ADAR1 and ADAR2 have overlapping specificities

Adenosine deaminases that act on RNA (ADARs) deaminate adenosines to produce inosines within RNAs that are largely double-stranded (ds). Like most dsRNA binding proteins, the enzymes will bind to any dsRNA without apparent sequence specificity. However, once bound, ADARs deaminate certain adenosines more efficiently than others. Most of what is known about the intrinsic deamination specificity of ADARs derives from analyses of Xenopus ADAR1. In addition to ADAR1, mammalian cells have a second ADAR, named ADAR2; the deamination specificity of this enzyme has not been rigorously studied. Here we directly compare the specificity of human ADAR1 and ADAR2. We find that, like ADAR1, ADAR2 has a 5' neighbor preference (A approximately U > C = G), but, unlike ADAR1, also has a 3' neighbor preference (U = G > C = A). Simultaneous analysis of both neighbor preferences reveals that ADAR2 prefers certain trinucleotide sequences (UAU, AAG, UAG, AAU). In addition to characterizing ADAR2 preferences, we analyzed the fraction of adenosines deaminated in a given RNA at complete reaction, or the enzyme's selectivity. We find that ADAR1 and ADAR2 deaminate a given RNA with the same selectivity, and this appears to be dictated by features of the RNA substrate. Finally, we observed that Xenopus and human ADAR1 deaminate the same adenosines on all RNAs tested, emphasizing the similarity of ADAR1 in these two species. Our data add substantially to the understanding of ADAR2 specificity, and aid in efforts to predict which ADAR deaminates a given editing site adenosine in vivo.     

Deamination of mammalian glutamate receptor RNA by Xenopus dsRNA adenosine deaminase: similarities to in vivo RNA editing.

Double-stranded RNA (dsRNA) adenosine deaminase (dsRAD) converts adenosines to inosines within dsRNA. A great deal of evidence suggests that dsRAD or a related enzyme edits mammalian glutamate receptor mRNA in vivo. Here we map the deamination sites that occur in a truncated glutamate receptor-B (gluR-B) mRNA after incubation with pure Xenopus dsRAD. We find remarkable similarities, as well as distinct differences, between the observed deamination sites and the sites reported to be edited within RNAs isolated from mammalian brain. For example, although deamination at the biologically relevant Q/R editing site occurs, it occurs much less frequently than editing at this site in vivo. We hypothesize that the similarities between the deamination and editing patterns exist because the deamination specificity that is intrinsic to dsRAD is involved in selecting editing sites in vivo. We propose that the observed differences are due to the absence of accessory factors that play indirect roles in vivo, such as binding to and occluding certain sites from dsRAD, or promoting the RNA structure required for correct and efficient editing. The work reported here also suggests that dsRAD is capable of much more selectivity than previously thought; a minimal number of deamination sites (average < or = 5) were found in each gluR-B RNA. We speculate that the observed selectivity is due to the various structural elements (mismatches, bulges, loops) that periodically interrupt the base paired region required for editing.     

RNAi is antagonized by A→I hyper-editing

RNA interference (RNAi) and adenosine to inosine conversion are both mechanisms that respond to double-stranded RNA (dsRNA) and have been suggested to have antiviral roles. RNAi involves processing of dsRNA to short interfering RNAs (siRNAs), which subsequently mediate degradation of the cognate mRNAs. Deamination of adenosines changes the coding capacity of the RNA, as inosine is decoded as guanosine, and alters the structure because A–U base pairs are replaced by I•U wobble pairs. Here we show that RNAi is inhibited if the triggering dsRNA is first deaminated by ADAR2. Moreover, we show that production of siRNAs is progressively inhibited with increasing deamination and that this is sufficient to explain the inhibition of RNAi upon hyper-editing of dsRNAs.     

Cleavage of dsRNAs hyper-edited by ADARs occurs at preferred editing sites

Long double-stranded RNAs (dsRNAs) may undergo covalent modification (hyper-editing) by adenosine deaminases that act on RNA (ADARs), whereby up to 50–60% of adenosine residues are converted to inosine. Previously, we have described a ribonuclease activity in various cell extracts that specifically targets dsRNAs hyper-edited by ADARs. Such a ribonuclease may play an important role in viral defense, or may alternatively be involved in down-regulation of other RNA duplexes. Cleavage of hyper-edited dsRNA occurs within sequences containing multiple IU pairs but not in duplexes that contain either isosteric GU pairs or Watson–Crick base pairs. Here, we describe experiments aimed at further characterizing cleavage of hyper-edited dsRNA. Using various inosine-containing dsRNAs we show that cleavage occurs preferentially at a site containing both IU and UI pairs, and that inclusion of even a single GU pair inhibits cleavage. We also show that cleavage occurs on both strands within a single dsRNA molecule and requires a 2′-OH group. Strikingly, we show that ADAR1, ADAR2 or dADAR all preferentially generate the preferred cleavage site when hyper-editing a long dsRNA.     

Crystal structure of a model branchpoint-U2 snRNA duplex containing bulged adenosines

Bulged nucleotides play a variety of important roles in RNA structure and function, frequently forming tertiary interactions and sometimes even participating in RNA catalysis. In pre-mRNA splicing, the U2 snRNA base pairs with the intron branchpoint sequence (BPS) to form a short RNA duplex that contains a bulged adenosine that ultimately serves as the nucleophile that attacks the 5' splice site. We have determined a 2.18-A resolution crystal structure of a self-complementary RNA designed to mimic the highly conserved yeast (Saccharomyces cerevisiae) branchpoint sequence (5'-UACUAACGUAGUA with the BPS italicized and the branchsite adenosine underlined) base paired with its complementary sequence from U2 snRNA. The structure shows a nearly ideal A-form helix from which two unpaired adenosines flip out. Although the adenosine adjacent to the branchsite adenosine is the one bulged out in the structure described here, either of these adenosines can serve as the nucleophile in mammalian but not in yeast pre-mRNA splicing. In addition, the packing of the bulged RNA helices within the crystal reveals a novel RNA tertiary interaction in which three RNA helices interact through bulged adenosines in the absence of any divalent metal ions.     

Functional importance of conserved nucleotides at the histone RNA 3' processing site.

Histone pre-mRNA 3' processing is controlled by a hairpin element preceding the processing site that interacts with a hairpin-binding protein (HBP) and a downstream spacer element that serves as anchoring site for the U7 snRNP. In addition, the nucleotides following the hairpin and surrounding the processing site (ACCCA'CA) are conserved among vertebrate histone genes. Single to triple nucleotide mutations of this sequence were tested for their ability to be processed in nuclear extract from animal cells. Changing the first four nucleotides had no qualitative and little if any quantitative effects on histone RNA 3' processing in mouse K21 cell extract, where processing of this gene is virtually independent of the HBP. A gel mobility shift assay revealing HBP interactions and a processing assay in HeLa cell extract (where the contribution of HBP to efficient processing is more important) showed that only one of these mutations, predicted to extend the hairpin by one base pair, affected the interaction with HBP. Mutations in the next three nucleotides affected both the cleavage efficiency and the choice of processing sites. Analysis of these novel sites indicated a preference for the nucleotide 5' of the cleavage site in the order A > C > U > G. Moreover, a guanosine in the 3' position inhibited cleavage. The preference for an A is shared with the cleavage/polyadenylation reaction, but the preference order for the other nucleotides is different [Chen F, MacDonald CC, Wilusz J, 1995, Nucleic Acids Res 23:2614-2620].     

The mechanism of adenosine to inosine conversion by the double-stranded RNA unwinding/modifying activity: a high-performance liquid chromatography-mass spectrometry analysis.

We have used directly combined high-performance liquid chromatography-mass spectrometry (LC/MS) to examine the mechanism of the reaction catalyzed by the double-stranded RNA unwinding/modifying activity [Bass & Weintraub (1988) Cell 55, 1089-1098]. A double-stranded RNA substrate in which all adenosines were uniformly labeled with 13C was synthesized. An LC/MS analysis of the nucleoside products from the modified, labeled substrate confirmed that adenosine is modified to inosine during the unwinding/modifying reaction. Most importantly, we found that no carbons are exchanged during the reaction. By including H2(18)O in the reaction, we showed that water serves efficiently as the oxygen donor in vitro. These results are consistent with a hydrolytic deamination mechanism and rule out a base replacement mechanism. Although the double-stranded RNA unwinding/modifying activity appears to utilize a catalytic mechanism similar to that of adenosine deaminase, coformycin, a transition-state analogue, will not inhibit the unwinding/modifying activity.     

Influence of the three nucleotides upstream of the initiation codon on expression of the Escherichia coli lacZ gene in Saccharomyces cerevisiae.

By introducing synthetic oligonucleotides into a lacZ-yeast expression vector a set of 47 plasmids (out of 64 possible) was generated, differing only in the three bases immediately upstream of the AUG initiation codon of the Escherichia coli lacZ gene. Expression of the beta-galactosidase fusion protein encoded by the different plasmids was determined in Saccharomyces cerevisiae by immunogel electrophoresis. Among the clones tested we found a factor 3 difference in expression. A slight nucleotide preference was found in positions -3(A > G > C = U) and -2 (G > C = U > A). The choice of the nucleotide at position -1 immediately 5' of the AUG did not effect translation efficiency. Increasing homology to the yeast consensus sequence (AAAAAAAUGUCU) was not concomitant with an increased translation efficiency. Our results indicate that the choice of nucleotides immediately preceding the initiation codon in yeast does not dramatically influence translation efficiency, as in prokaryotes or higher eukaryotes.     

Reovirus Replicase-Directed Synthesis of Double-Stranded Ribonucleic Acid

After the incubation of reovirus replicase reaction mixtures (containing labeled ribonucleoside triphosphates), partially double-stranded ribonucleic acid (pdsRNA) products were isolated by cellulose column chromatography followed by precipitation with 2 m NaCl. The pulse-labeled reaction product contained a significantly large amount of pdsRNA that became complete dsRNA as reaction time increased, indicating that pdsRNA was an intermediate of the replicase reaction. The newly synthesized RNA strand (³H-labeled) of the pdsRNA was resistant to ribonuclease digestion, suggesting that single-stranded RNA regions were part of a preexistent unlabeled RNA template. These observations, together with the electrophoretic behavior of the pdsRNA in polyacrylamide gel, are consistent with the hypothesis that dsRNA is synthesized by the elongation of a complementary RNA strand upon a preexistent template of single-stranded RNA (i.e., messenger RNA). The direction of the RNA strand elongation was determined by carrying out the replicase reaction in the presence of ³H-cytidine triphosphate (or ³H-uridine triphosphate) and adenine triphosphate-α-³²P followed by a chase with excess unlabeled cytidine triphosphate (or uridine triphosphate). The dsRNA product was digested with T1 ribonuclease and the resulting 3′-terminal fragments were isolated by chromatography on a dihydroxyboryl derivative of cellulose. Examination of the ratio of ³H to ³²P in these fragments indicated that RNA synthesis proceeded from the 5′ to 3′ terminus.     

The RISC subunit Tudor-SN binds to hyper-edited double-stranded RNA and promotes its cleavage

Long perfect double-stranded RNA (dsRNA) molecules play a role in various cellular pathways. dsRNA may undergo extensive covalent modification (hyper-editing) by adenosine deaminases that act on RNA (ADARs), resulting in conversion of up to 50% of adenosine residues to inosine (I). Alternatively, dsRNA may trigger RNA interference (RNAi), resulting in silencing of the cognate mRNA. These two pathways have previously been shown to be antagonistic. We show a novel interaction between components of the ADAR and RNAi pathways. Tudor staphylococcal nuclease (Tudor-SN) is a subunit of the RNA-induced silencing complex, which is central to the mechanism of RNAi. Here we show that Tudor-SN specifically interacts with and promotes cleavage of model hyper-edited dsRNA substrates containing multiple I.U and U.I pairs. This interaction suggests a novel unsuspected interplay between the two pathways that is more complex than mutual antagonism.     

Effects of hypoxanthine substitution on bleomycin-mediated DNA strand degradation in DNA-RNA hybrids.

We have reported on the differences in site-specific cleavage between DNA and DNA-RNA hybrids by various prototypic DNA cleavers (accompanying paper). In the case of bleomycin (BLM), degradation at 5'-GC-3'sites was suppressed relative to the same sequence in double-stranded DNA, while 5'-GT-3' damage remained constant. We now present results of our further investigation on the chemical and conformational factors that contribute to BLM-mediated DNA strand cleavage of DNA-RNA hybrids. Substitution of guanine by hypoxanthine on the RNA strand of hybrids resulted in a significant enhancement of 5'-GC-3' site damage on the DNA strand relative to double-stranded DNA, thus reversing the suppression noted at these sites. Additionally, 5'-AT-3' sites, which are damaged significantly more in the hybrid than in DNA, exhibit decreased product formation when hypoxanthine is present on the RNA strand of hybrids. However, when hypoxanthine is substituted for guanine on the DNA strand (a GC cleavage site becomes IC), 5'-IT-3' and 5'-IC-3' site cleavage is almost completely suppressed, whereas AT site cleavage is dramatically enhanced. The priority in metallobleomycin site-specific cleavage of hybrids changes with hypoxanthine substitution: the cleavage priority is AT > GT > GC in native hybrid; GC > GT > AT in hybrids substituted with hypoxanthine in the RNA strand; AT >> GT approximately GC in hybrids substituted with hypoxanthine in the DNA strand. The results of kinetic isotope effect studies on BLM cleavage are presented and, in most cases, the values are larger for the hypoxanthine-substituted hybrid. The results suggest that the 2-amino groups of guanine residues on both strands of the nucleic acid play an important role in modulation of the binding and cleavage specificity of BLM.     

The Helicase Activity of Ribonuclease R Is Essential for Efficient Nuclease Activity

RNase R, which belongs to the RNB family of enzymes, is a 3′ to 5′ hydrolytic exoribonuclease able to digest highly structured RNA. It was previously reported that RNase R possesses an intrinsic helicase activity that is independent of its ribonuclease activity. However, the properties of this helicase activity and its relationship to the ribonuclease activity were not clear. Here, we show that helicase activity is dependent on ATP and have identified ATP-binding Walker A and Walker B motifs that are present in Escherichia coli RNase R and in 88% of mesophilic bacterial genera analyzed, but absent from thermophilic bacteria. We also show by mutational analysis that both of these motifs are required for helicase activity. Interestingly, the Walker A motif is located in the C-terminal region of RNase R, whereas the Walker B motif is in its N-terminal region implying that the two parts of the protein must come together to generate a functional ATP-binding site. Direct measurement of ATP binding confirmed that ATP binds only when double-stranded RNA is present. Detailed analysis of the helicase activity revealed that ATP hydrolysis is not required because both adenosine 5′-O-(thiotriphosphate) and adenosine 5′-(β,γ-imino)triphosphate can stimulate helicase activity, as can other nucleoside triphosphates. Although the nuclease activity of RNase R is not needed for its helicase activity, the helicase activity is important for effective nuclease activity against a dsRNA substrate, particularly at lower temperatures and with more stable duplexes. Moreover, competition experiments and mutational analysis revealed that the helicase activity utilizes the same catalytic channel as the nuclease activity. These findings indicate that the helicase activity plays an essential role in the catalytic efficiency of RNase R.     

RNA mimetics: oligoribonucleotide N3'-->P5' phosphoramidates.

The synthesis and properties of novel RNA mimetics, oligoribonucleotide N3'-->P5' phosphoramidates, are described. These oligonucleotides contain 3'-aminoribonucleosides connected via N3'-->P5' phosphoramidate linkages, replacing the native RNA O3'-->P5' phosphodiester counterparts. The key monomers 2'-t-butyldimethylsilyl-3'-(monomethoxytrityl)-amino-5'-phospho ramidi tes were synthesized and used to prepare the oligonucleotide phosphoramidates using a solid phase methodology based on the phosphoramidite transfer reaction. Oligoribophosphoramidates are very resistant to enzymatic hydrolysis by snake venom phosphodiesterase. These compounds form stable duplexes with complementary natural phosphodiester DNA and RNA strands, as well as with 2'-deoxy N3'-->P5' phosphoramidates. The increase in melting temperature, Delta T m, was 5-14 degrees C relative to the 2'-deoxy phosphoramidates for decanucleotides. Also, the thermal stability of the ribophosphoramidatehomoduplex was noticeably higher (Delta T m +9.5 degrees C) than that for the isosequential 2'-deoxy phosphoramidate complex. Furthermore, the oligopyrimidine ribo N3'-->P5' phosphoramidate formed an extremely stable triplex with an oligopurine/oligopyrimidine DNA duplex with Delta T m +14.3 degrees C relative to the 2'-deoxy N3'-->P5' phosphoramidate counterpart. The properties of the oligoribonucleotide N3'-->P5' phosphoramidates indicate that these compounds can be used as hydrolytically stable structural and functional RNA mimetics.     

trans-spliceosomal U6 RNAs of Crithidia fasciculata and Leptomonas seymouri: deviation from the conserved ACAGAG sequence and potential base pairing with spliced leader RNA.

U6 RNA genes from the trypanosomatids Crithidia fasciculata and Leptomonas seymouri have been isolated and sequenced. As in Trypanosoma brucei, the U6 RNA genes in both C. fasciculata and L. seymouri are arranged in close linkage with upstream tRNA genes. The U6 RNA sequences from C. fasciculata and L. seymouri deviate in five and three positions, respectively, from the published T. brucei sequence. Interestingly, both C. fasciculata U6 RNA genes carry a C-->T change at the second position of the ACAGAG hexanucleotide sequence, which is important for splicing function and has been considered phylogenetically invariable. A compensatory base change of the C. fasciculata spliced leader RNA at the highly conserved 5' splice site position +5, G-->A, suggests that an interaction between the 5' splice site region and U6 RNA recently proposed for the yeast cis-splicing system may also occur in trans splicing.     

Contribution of structural elements to Thermus thermophilus ribonuclease P RNA function.

We have performed a deletion and mutational analysis of the catalytic ribonuclease (RNase) P RNA subunit from the extreme thermophilic eubacterium Thermus thermophilus HB8. Catalytic activity was reduced 600-fold when the terminal helix, connecting the 5' and 3' ends of the molecule, was destroyed by deleting 15 nucleotides from the 3' end. In comparison, the removal of a large portion (94 nucleotides, about one quarter of the RNA) of the upper loop region impaired function only to a relatively moderate extent (400-fold reduction in activity). The terminal helix appears to be crucial for the proper folding of RNase P RNA, possibly by orientating the adjacent universally conserved pseudoknot structure. The region containing the lower half of the pseudoknot structure was shown to be a key element for enzyme function, as was the region of nucleotides 328-335. Deleting a conserved hairpin (nucleotides 304-327) adjacent to this region and replacing the hairpin by a tetranucleotide sequence or a single cytidine reduced catalytic activity only 6-fold, whereas a simultaneous mutation of the five highly conserved nucleotides in the region of nucleotides 328-335 reduced catalytic activity by > 10(5)-fold. The two strictly conserved adenines 244 and 245 (nucleotides 248/249 in Escherichia coli RNase P RNA) were not as essential for enzyme function as suggested by previous data. However, additional disruption of two helical segments (nucleotides 235-242) adjacent to nucleotides 244 and 245 reduced activity by > 10(4)-fold, supporting the notion that nucleotides in this region are also part of the active core structure.(ABSTRACT TRUNCATED AT 250 WORDS)     

Extensive C->U transition biases in the genomes of a wide range of mammalian RNA viruses; potential associations with transcriptional mutations,damage- or host-mediated editing of viral RNA

The rapid evolution of RNA viruses has been long considered to result from a combination of high copying error frequencies during RNA replication, short generation times and the consequent extensive fixation of neutral or adaptive changes over short periods. While both the identities and sites of mutations are typically modelled as being random, recent investigations of sequence diversity of SARS coronavirus 2 (SARS-CoV-2) have identified a preponderance of C->U transitions, proposed to be driven by an APOBEC-like RNA editing process. The current study investigated whether this phenomenon could be observed in datasets of other RNA viruses. Using a 5% divergence filter to infer directionality, 18 from 36 datasets of aligned coding region sequences from a diverse range of mammalian RNA viruses (including Picornaviridae, Flaviviridae, Matonaviridae, Caliciviridae and Coronaviridae) showed a >2-fold base composition normalised excess of C->U transitions compared to U->C (range 2.1x–7.5x), with a consistently observed favoured 5’ U upstream context. The presence of genome scale RNA secondary structure (GORS) was the only other genomic or structural parameter significantly associated with C->U/U->C transition asymmetries by multivariable analysis (ANOVA), potentially reflecting RNA structure dependence of sites targeted for C->U mutations. Using the association index metric, C->U changes were specifically over-represented at phylogenetically uninformative sites, potentially paralleling extensive homoplasy of this transition reported in SARS-CoV-2. Although mechanisms remain to be functionally characterised, excess C->U substitutions accounted for 11–14% of standing sequence variability of structured viruses and may therefore represent a potent driver of their sequence diversification and longer-term evolution.