Group I intron self-splicing with adenosine: evidence for a single nucleoside-binding site.

For self-splicing of Tetrahymena ribosomal RNA precursor, guanosine binding is required for 5' splice-site cleavage and exon ligation. Whether these two reactions use the same or different guanosine-binding sites has been debated. A double mutation in a previously identified guanosine-binding site within the intron resulted in preference for adenosine (or adenosine triphosphate) as the substrate for cleavage at the 5' splice site. However, splicing was blocked in the exon ligation step. Blockage was reversed by a change from guanine to adenine at the 3' splice site. These results indicate that a single determinant specifies nucleoside binding for both steps of splicing. Furthermore, it suggests that RNA could form an active site specific for adenosine triphosphate.

Imidazole rescue of a cytosine mutation in a self-cleaving ribozyme.

Ribozymes use a number of the same catalytic strategies as protein enzymes. However, general base catalysis by a ribozyme has not been demonstrated. In the hepatitis delta virus antigenomic ribozyme, imidazole buffer rescued activity of a mutant with a cytosine-76 (C76) to uracil substitution. In addition, a C76 to adenine substitution reduced the apparent pKa (where Ka is the acid constant) of the self-cleavage reaction by an amount consistent with differences in the pKa values of these two side chains. These results suggest that, in the wild-type ribozyme, C76 acts as a general base. This finding has implications for potential catalytic functions of conserved cytosines and adenines in other ribozymes and in ribonuclear proteins with enzymatic activity.

RNA as an RNA polymerase: net elongation of an RNA primer catalyzed by the Tetrahymena ribozyme.

A catalytic RNA (ribozyme) derived from an intervening sequence (IVS) RNA of Tetrahymena thermophila will catalyze an RNA polymerization reaction in which pentacytidylic acid (C5) is extended by the successive addition of mononucleotides derived from a guanylyl-(3',5')-nucleotide (GpN). Cytidines or uridines are added to C5 to generate chain lengths of 10 to 11 nucleotides, with longer products being generated at greatly reduced efficiency. The reaction is analogous to that catalyzed by a replicase with C5 acting as the primer, GpNs as the nucleoside triphosphates, and a sequence in the ribozyme providing a template. The demonstration that an RNA enzyme can catalyze net elongation of an RNA primer supports theories of prebiotic RNA self-replication.

Cis- and trans-acting ribozymes from a human pathogen, hepatitis delta virus.

Hepatitis delta virus (HDV) contains two self-cleaving RNA sequences (ribozymes) that may naturally function as such in human cells. A pseudo-knot-containing structural motif, which is distinct from the well-characterized secondary structures of self-cleaving RNAs common to the plant pathogenic RNAs, is shared by the cis-acting HDV ribozymes. Definition of the sequences and secondary structures of the HDV ribozymes has facilitated the design of novel catalytic molecules, such as small RNA circles, capable of site-specific cleavage of RNA in trans.

HDV ribozymes.

The self-cleaving RNA sequences, or ribozymes, in the genomic and antigenomic strands of hepatitis delta virus (HDV) RNA fold into structures that are similar to each other but distinct from those of small ribozymes associated with the RNA replicons that infect plants. HDV ribozymes have provided a tractable system for studying the mechanism of catalytic RNA, and results of biochemical and structural studies on the HDV ribozymes, from a number of labs, have enhanced our understanding and expanded our thinking about the potential for catalytic roles of RNA side chains. The results of these studies are consistent with models suggesting that both an active-site cytosine and a divalent metal ion have catalytic roles in facilitating the cleavage reaction in the HDV ribozymes. Despite recent advances, details about the catalytic mechanism of the HDV ribozyme continue to be debated, and these ribozymes should serve as a good system for further study.

DNA strand breakage by wheat germ type 1 topoisomerase.

Properties of strand breakage in duplex and single-stranded DNA by the wheat germ type 1 DNA topoisomerase were investigated. Strand breakage in duplex DNA is dependent upon the use of denaturing conditions to inactivate the enzyme and terminate the reaction, whereas breakage of single-stranded DNA occurs under the normal reaction conditions and is not dependent upon denaturation. Breakage generates a free 5' hydroxyl group and enzyme bound to the 3' side of the break, presumably via the 3' phosphate group. The location of sites of breakage with both duplex and single-stranded DNA is not random. In all these respects the wheat germ enzyme closely resembles the rat liver type 1 topoisomerase. A comparison of the locations of the sites of breakage in duplex DNA generated by the topoisomerases from wheat germ and rat liver indicates a number of common sites, although the patterns of breakage are not identical.

Numerous group I introns with variable distributions in the ribosomal DNA of a lichen fungus.

The length of the small subunit ribosomal DNA (SSU rDNA) differs significantly among individuals from natural populations of the ascomycetous lichen complex Cladonia chlorophaea. The sequence of the 3' region of the SSU rDNA from two individuals, chosen to represent the shortest and longest sequences, revealed multiple insertions within a region that otherwise aligned with a 520-nucleotide sequence of the SSU rDNA in Saccharomyces cerevisiae. The high degree of variability in SSU rDNA size can be accounted for by different numbers of insertions; one individual had two group I introns and the second had five introns, two of which were clearly related to introns at identical positions in the other individual. Yet, introns in different positions, whether within an individual or between individuals, were not similar in sequence. The distribution of introns at three of the positions is consistent with either intron loss or acquisition, and clearly indicates the dynamic variability in this region of the nuclear genome. All seven insertions, which ranged in size from 210 to 228 nucleotides, had the conserved sequence and secondary structural elements of group I introns. The variation in distribution and sequence of group I introns within a short highly conserved region of rDNA presents a unique opportunity for examining the molecular evolution and mobility of group I introns within a systematics framework.

A toggle duplex in hepatitis delta virus self-cleaving RNA that stabilizes an inactive and a salt-dependent pro-active ribozyme conformation.

The antigenomic RNA of hepatitis delta virus (HDV) can form a short duplex, P2a, in which a four-nucleotide sequence within the self-cleaving domain pairs with a sequence just outside the previously defined 3'-boundary of the ribozyme. Both sequences that would participate in forming P2a were previously determined to be non-essential for self-cleavage activity. Ribozymes able to form P2a were less active than those lacking the 3' P2a sequence when preincubated under the standard low-Na+ conditions. Chemical probing of the RNA correlated base-pairing in P2a with this inhibition. Furthermore, mutagenesis and 3' truncation experiments mapped the inhibitory sequence to P2a. However, raising the NaCl concentration in the preincubation prior to adding Mg2+ reversed the inhibitory effect. Moreover, with NaCl preincubation, the P2a-containing ribozyme was more active than an otherwise identical ribozyme lacking the 3' P2a sequence. Non-denaturing gels provided evidence for alternative conformations of the P2a-containing precursor with only the faster-migrating species correlating with the active form. A difference in the temperature-dependence for the rate of cleavage of the P2a-containing ribozyme with and without NaCl, together with a difference in the melting behavior of the RNA in NaCl with and without P2a, suggested that P2a favors the native structure in NaCl. Many derivatives of the HDV ribozymes form inactive conformers; however, this study reveals details of a specific structure that stabilizes both inactive and active conformations of the HDV ribozyme.

Breakage of single-stranded DNA by eukaryotic type 1 topoisomerase occurs only at regions with the potential for base-pairing.

Eukaryotic type 1 DNA topoisomerases break single-stranded DNA at specific sites. A preferred site for rat liver topoisomerase breakage in single-stranded phi X174 DNA was located within a region of the DNA with the potential for duplex formation. To investigate the relationship between sites of breakage in duplex and single-stranded DNA, a restriction fragment containing sequences from the transcriptional regulatory and enhancer region of the simian virus 40 genome was used as a substrate for topoisomerase. While different patterns of breakage in the native and denatured forms of the DNA were observed, some sites of breakage were common to both forms. The break sites in the denatured DNA were a subset of the break sites in the duplex DNA and were located in regions which had the potential for intrastrand base-pairing due to distal complementary sequences. A series of single-stranded fragments were generated with the distal complementary sequences deleted and these fragments were used as substrates for topoisomerase breakage. The lack of detectable breakage at a site when the complementary sequence was deleted, suggests that topoisomerase acts at duplex regions in the single-stranded DNA and that it is not active on regions of single-stranded DNA that are not base-paired.

A pH-sensitive RNA tertiary interaction affects self-cleavage activity of the HDV ribozymes in the absence of added divalent metal ion.

Self-cleavage of the genomic and antigenomic ribozymes from hepatitis delta virus (HDV) requires divalent cation for optimal activity. Recently, the HDV genomic ribozyme has been shown to be active in NaCl in the absence of added divalent metal ion at low pH (apparent pKa 5.7). However, we find that the antigenomic ribozyme is 100 to 1000-fold less active under similar conditions. With deletion of a three-nucleotide sequence (C41-A42-A43) unique to the genomic ribozyme, the rate constant for cleavage decreased substantially, while activity of the antigenomic ribozyme was enhanced by introducing a CAA sequence. From the crystal structure, it has been proposed that C41 in this sequence is protonated. To investigate a possible connection between activity at low pH and protonation of C41, mutations were made that were predicted to either eliminate protonation or alter the nature of the tertiary interaction upon protonation. In the absence of added Mg2+, these mutations reduced activity and eliminated the observed pH-rate dependence. Thermal denaturation studies revealed a pH-sensitive structural feature in the genomic ribozyme, while unfolding of the mutant ribozymes was pH-independent. We propose that, in the absence of added Mg2+, protonation of C41 contributes to enhanced activity of the HDV genomic ribozyme at low pH.

Energetic contribution of non-essential 5' sequence to catalysis in a hepatitis delta virus ribozyme.

Hepatitis delta virus (HDV) ribozymes employ multiple catalytic strategies to achieve overall rate enhancement of RNA cleavage. These strategies include general acid-base catalysis by a cytosine side chain and involvement of divalent metal ions. Here we used a trans-acting form of the antigenomic ribozyme to examine the contribution of the 5' sequence in the substrate to HDV ribozyme catalysis. The cleavage rate constants increased for substrates with 5' sequence alterations that reduced ground-state binding to the ribozyme. Quantitatively, a plot of activation free energy of chemical conversion versus Gibb's free energy of substrate binding revealed a linear relationship with a slope of -1. This relationship is consistent with a model in which components of the substrate immediately 5' to the cleavage site in the HDV ribozyme-substrate complex destabilize ground-state binding. The intrinsic binding energy derived from the ground-state destabilization could contribute up to 2 kcal/mol toward the total 8.5 kcal/mol reduction in activation free energy for RNA cleavage catalyzed by the HDV ribozyme.

Assessment of disparate structural features in three models of the hepatitis delta virus ribozyme.

Three models for the secondary structure of the hepatitis delta virus (HDV) antigenomic self-cleaving RNA element were tested by site-directed mutagenesis. Two models in which bases 5' to the cleavage site are paired with sequence at the 3' end of the element were both inconsistent with the data from the mutagenesis. Specifically, mutations in the 3' sequence which decrease self-cleavage activity could not be compensated by base changes in the 5' sequence as predicted by these models. The evidence was consistent with a third model in which the 3' end pairs with a portion of a loop within the ribozyme sequence to generate a pseudoknot structure. This same pairing was also required to generate higher rates of cleavage in trans with a 15-mer ribozyme, thus ruling out a proposed hammerhead-like 'axehead' model for the HDV ribozyme.

Evidence that genomic and antigenomic RNA self-cleaving elements from hepatitis delta virus have similar secondary structures.

The two sequences that define the self-cleaving elements from the genomic and antigenomic RNA of hepatitis delta virus were folded into secondary structures with similar features. Evidence in support of the two models was obtained from limited ribonuclease digestion of genomic and antigenomic RNA fragments containing the sequence 3' of the cleavage site. Under conditions where the rates of self-cleavage are enhanced by addition of 5 M urea (2-10 mM Mg2+ at 37 degrees C), ribonucleases T1, U2, A and V1 generated digestion patterns consistent with the proposed RNA structures. The evidence for a relatively stable structure in urea when Mg2+ is present suggests that denaturant-enhanced rates of self-cleavage could result from destabilization of competing inactive structures.

A pseudoknot-like structure required for efficient self-cleavage of hepatitis delta virus RNA.

Hepatitis delta virus genomic and antigenomic RNAs contain a self-cleavage site hypothesized to function in processing the viral RNA during replication. Self-cleavage requires only a divalent cation and is mediated at the genomic site by a sequence of less than 85 nucleotides. We propose that the genomic self-cleaving sequence element and a corresponding sequence from the anti-genomic RNA could generate related secondary structures. The region of the antigenomic sequence, predicted from the proposed structure, was synthesized and shown to be sufficient for self-cleavage. Evidence for two stems which form a tertiary interaction was obtained by site-specific mutagenesis of the antigenomic sequence. Efficient self-cleavage in 10 M formamide or 5 M urea, also a property of the genomic sequence, was dependent on base-pairing in both stems. But in the absence of denaturants, the stem distal to the site of cleavage was not required, suggesting that the tertiary interaction stabilizes the structure required for self-cleavage.

Group I permuted intron-exon (PIE) sequences self-splice to produce circular exons.

Circularly permuted group I intron precursor RNAs, containing end-to-end fused exons which interrupt half-intron sequences, were generated and tested for self-splicing activity. An autocatalytic RNA can form when the primary order of essential intron sequence elements, splice sites, and exons are permuted in this manner. Covalent attachment of guanosine to the 5' half-intron product, and accurate exon ligation indicated that the mechanism and specificity of splicing were not altered. However, because the exons were fused and the order of the splice sites reversed, splicing released the fused-exon as a circle. With this arrangement of splice sites, circular exon production was a prediction of the group I splicing mechanism. Circular RNAs have properties that would make them attractive for certain studies of RNA structure and function. Reversal of splice site sequences in a context that allows splicing, such as those generated by circularly permuted group I introns, could be used to generate short defined sequences of circular RNA in vitro and perhaps in vivo.

DNA breakage and closure by rat liver type 1 topoisomerase: separation of the half-reactions by using a single-stranded DNA substrate.

Circular single strands of bacteriophage phi X174 DNA are broken by rat liver DNA nicking-closing enzyme (type 1 topoisomerase) in low salt (50 mM KCl) at 37 degrees C, generating linear strands containing covalently bound enzyme [Been, M. D. & Champoux, J. J. (1980) Nucleic Acids Res. 8, 6129-6142]. The linear strands can be recircularized in the presence of 10 mM MgCl2 at 24 degrees C and 37 degrees C or 250 mM KCl at 24 degrees C. Recircularization is blocked when the hydroxyl group at the 5' terminus is phosphorylated. The linears generated by the nicking-closing enzyme can also be joined to other DNA fragments containing 5' hydroxyls, but not 5' phosphates. The linkage formed in both the intrastrand and interstrand reactions is stable to alkali. Reclosure of broken single strands is presumed to be analogous to the closure step that occurs durng nicking and closing cycles on duplex DNA.

Optimal self-cleavage activity of the hepatitis delta virus RNA is dependent on a homopurine base pair in the ribozyme core.

A non-Watson-Crick G.G interaction within the core region of the hepatitis delta virus (HDV) antigenomic ribozyme is required for optimal rates of self-cleavage activity. Base substitutions for either one or both G's revealed that full activity was obtained only when both G's were replaced with A's. At those positions, substitutions that generate potential Watson-Crick, G.U, heteropurine, or homopyrimidine combinations resulted in dramatically lower cleavage activity. A homopurine symmetric base pair, of the same type identified in the high-affinity binding site of the HIV RRE, is most consistent with this data. Additional features shared between the antigenomic ribozyme and the Rev binding site in the vicinity of the homopurine pairs suggest some structural similarity for this region of the two RNAs and a possible motif associated with this homopurine interaction. Evidence for a homopurine pair at the equivalent position in a modified form of the HDV genomic ribozyme was also found. With the postulated symmetric pairing scheme, large distortions in the nucleotide conformation, the sugar-phosphate backbone, or both would be necessary to accommodate this interaction at the end of a helix; we hypothesize that this distortion is critical to the structure of the active site of the ribozyme and it is stabilized by the homopurine base pair.

Generation of nuclease resistant circular RNA decoys for HIV-Tat and HIV-Rev by autocatalytic splicing.

Circular exon sequences can be generated by splicing permuted intron-exon (PIE) sequences. The Anabaena pre-tRNA group I self-splicing PIE sequence was modified to generate circular forms of the HIV-TAR and the high affinity region of the HIV-RRE (RBE). RNA products containing TAR and the RBE were purified from splicing reactions and demonstrated to be circular. The circular form of these sequences was shown to be resistant to nuclease degradation in cellular extracts. Gel shift assays demonstrate that the circular form of the RBE is specifically bound by a Rev derived peptide. These data suggest that PIE-circularization of RNA may be an effective way to express small stable RNAs designed for therapeutics (eg-decoys).