Compensatory mutations demonstrate that P8 and P6 are RNA secondary structure elements important for processing of a group I intron

Deciphering RNA splicing logic with interpretable machine learning

Machine learning methods, particularly neural networks trained on large datasets, are transforming how scientists approach scientific discovery and experimental design. However, current state-of-the-art neural networks are limited by their uninterpretability: Despite their excellent accuracy, they cannot describe how they arrived at their predictions. Here, using an "interpretable-by-design" approach, we present a neural network model that provides insights into RNA splicing, a fundamental process in the transfer of genomic information into functional biochemical products. Although we designed our model to emphasize interpretability, its predictive accuracy is on par with state-of-the-art models. To demonstrate the model's interpretability, we introduce a visualization that, for any given exon, allows us to trace and quantify the entire decision process from input sequence to output splicing prediction. Importantly, the model revealed uncharacterized components of the splicing logic, which we experimentally validated. This study highlights how interpretable machine learning can advance scientific discovery.

Pairwise Engineering of Tandemly Aligned Self-Splicing Group I Introns for Analysis and Control of Their Alternative Splicing

Alternative splicing is an important mechanism in the process of eukaryotic nuclear mRNA precursors producing multiple protein products from a single gene. Although group I self-splicing introns usually perform regular splicing, limited examples of alternative splicing have also been reported. The exon-skipping type of splicing has been observed in genes containing two group I introns. To characterize splicing patterns (exon-skipping/exon-inclusion) of tandemly aligned group I introns, we constructed a reporter gene containing two Tetrahymena introns flanking a short exon. To control splicing patterns, we engineered the two introns in a pairwise manner to design pairs of introns that selectively perform either exon-skipping or exon-inclusion splicing. Through pairwise engineering and biochemical characterization, the structural elements important for the induction of exon-skipping splicing were elucidated.

Flexible Assembly of Engineered Tetrahymena Ribozymes Forming Polygonal RNA Nanostructures with Catalytic Ability

Ribozymes with modular architecture constitute an attractive class of structural platforms for design and construction of nucleic acid nanostructures with biological functions. Through modular engineering of the Tetrahymena ribozyme, we have designed unit RNAs (L-RNAs), assembly of which formed ribozyme-based closed trimers and closed tetramers. Their catalytic activity was dependent on oligomer formation. In this study, the structural variety of L-RNA oligomers was extended by tuning their structural elements, yielding closed pentamers and closed hexamers. Their assembly properties were analyzed by electrophoretic mobility shift assay (EMSA) and atomic force microscopy (AFM).

Heterodimerization of Group I Ribozymes Enabling Exon Recombination through Pairs of Cooperative trans-Splicing Reactions

Group I (GI) self-splicing ribozymes are attractive tools for biotechnology and synthetic biology. Several trans-splicing and related reactions based on GI ribozymes have been developed for the purpose of recombining their target mRNA sequences. By combining trans-splicing systems with rational modular engineering of GI ribozymes it was possible to achieve more complex editing of target RNA sequences. In this study we have developed a cooperative trans-splicing system through rational modular engineering with use of dimeric GI ribozymes derived from the Tetrahymena group I intron ribozyme. The resulting pairs of ribozymes exhibited catalytic activity depending on their selective dimerization. Rational modular redesign as performed in this study would facilitate the development of sophisticated regulation of double or multiple trans-splicing reactions in a cooperative manner.

Modulation of group I ribozyme activity by cationic porphyrins

The effects of cationic porphyrins on the catalytic activities of four group I ribozymes were investigated. A cationic porphyrin possessing four pyridinium moieties (pPyP) inhibited two group IC3 ribozymes (Syn Rz and Azo Rz) and a group IC1 ribozyme (Tet Rz). In the case of a group IA2 ribozyme (Td Rz), however, pPyP served not only as an inhibitor but also as an activator, and the effects of pPyP were dependent on its concentration. To analyze the structural and electronic factors determining the effects of pPyP on group I ribozymes, three cationic porphyrins (pPyNCP, pPyF4P, and TMPyP) were also examined. As interactions between small organic molecules and nucleic acids are attractive and important issues in biochemistry and biotechnology, this study contributes to the development of porphyrin-based molecules that can modulate functions of structured RNA molecules.

An in vitro-selected RNA receptor for the GAAC loop: modular receptor for non-GNRA-type tetraloop

Although artificial RNA motifs that can functionally replace the GNRA/receptor interaction, a class of RNA-RNA interacting motifs, were isolated from RNA libraries and used to generate designer RNA structures, receptors for non-GNRA tetraloops have not been found in nature or selected from RNA libraries. In this study, we report successful isolation of a receptor motif interacting with GAAC, a non-GNRA tetraloop, from randomized sequences embedded in a catalytic RNA. Biochemical characterization of the GAAC/receptor interacting motif within three structural contexts showed its binding affinity, selectivity and structural autonomy. The motif has binding affinity comparable with that of a GNRA/receptor, selectivity orthogonal to GNRA/receptors and structural autonomy even in a large RNA context. These features would be advantageous for usage of the motif as a building block for designer RNAs. The isolated motif can also be used as a query sequence to search for unidentified naturally occurring GANC receptor motifs.

A two-piece derivative of a group I intron RNA as a platform for designing self-assembling RNA templates to promote Peptide ligation

Multicomponent RNA-peptide complexes are attractive from the viewpoint of artificial design of functional biomacromolecular systems. We have developed self-folding and self-assembling RNAs that serve as templates to assist chemical ligation between two reactive peptides with RNA-binding capabilities. The design principle of previous templates, however, can be applied only to limited classes of RNA-binding peptides. In this study, we employed a two-piece derivative of a group I intron RNA from the Tetrahymena large subunit ribosomal RNA (LSU rRNA) as a platform for new template RNAs. In this group I intron-based self-assembling platform, modules for the recognition of substrate peptides can be installed independently from modules holding the platform structure. The new self-assembling platform allows us to expand the repertoire of substrate peptides in template RNA design.

Site-specific isotope labeling of long RNA for structural and mechanistic studies

A site-specific isotope labeling technique of long RNA molecules was established. This technique is comprised of two simple enzymatic reactions, namely a guanosine transfer reaction of group I self-splicing introns and a ligation with T4 DNA ligase. The trans-acting group I self-splicing intron with its external cofactor, ‘isotopically labeled guanosine 5′-monophosphate’ (5′-GMP), steadily gave a 5′-residue-labeled RNA fragment. This key reaction, in combination with a ligation of 5′-remainder non-labeled sequence, allowed us to prepare a site-specifically labeled RNA molecule in a high yield, and its production was confirmed with ¹⁵N NMR spectroscopy. Such a site-specifically labeled RNA molecule can be used to detect a molecular interaction and to probe chemical features of catalytically/structurally important residues with NMR spectroscopy and possibly Raman spectroscopy and mass spectrometry.

Correlated mutations: advances and limitations. A study on fusion proteins and on the Cohesin-Dockerin families

Correlated mutations have been repeatedly exploited for intramolecular contact map prediction. Over the last decade these efforts yielded several methods for measuring correlated mutations. Nevertheless, the application of correlated mutations for the prediction of intermolecular interactions has not yet been explored. This gap is due to several obstacles, such as 3D complexes availability, paralog discrimination, and the availability of sequence pairs that are required for inter- but not intramolecular analyses. Here we selected for analysis fusion protein families that bypass some of these obstacles. We find that several correlated mutation measurements yield reasonable accuracy for intramolecular contact map prediction on the fusion dataset. However, the accuracy level drops sharply in intermolecular contacts prediction. This drop in accuracy does not occur always. In the Cohesin-Dockerin family, reasonable accuracy is achieved in the prediction of both intra- and intermolecular contacts. The Cohesin-Dockerin family is well suited for correlated mutation analysis. Because, however, this family constitutes a special case (it has radical mutations, has domain repeats, within each species each Dockerin domain interacts with each Cohesin domain, see below), the successful prediction in this family does not point to a general potential in using correlated mutations for predicting intermolecular contacts. Overall, the results of our study indicate that current methodologies of correlated mutations analysis are not suitable for large-scale intermolecular contact prediction, and thus cannot assist in docking. With current measurements, sequence availability, sequence annotations, and underdeveloped sequence pairing methods, correlated mutations can yield reasonable accuracy only for a handful of families.

Ribozymes: recent advances in the development of RNA tools

The discovery 20 years ago that some RNA molecules, called ribozymes, are able to catalyze chemical reactions was a breakthrough in biology. Over the last two decades numerous natural RNA motifs endowed with catalytic activity have been described. They all fit within a few well-defined types that respond to a specific RNA structure. The prototype catalytic domain of each one has been engineered to generate trans-acting ribozymes that catalyze the site-specific cleavage of other RNA molecules. On the 20th anniversary of ribozyme discovery we briefly summarize the main features of the different natural catalytic RNAs. We also describe progress towards developing strategies to ensure an efficient ribozyme-based technology, dedicating special attention to the ones aimed to achieve a new generation of therapeutic agents.

Putative intermediary stages for the molecular evolution from a ribozyme to a catalytic RNP

A hypothetical evolutionary pathway from a ribozyme to a catalytic RNA-protein complex (RNP) is proposed and examined. In this hypothesis for an early phase of molecular evolution, one RNA-RNA interaction in the starting ribozyme is replaced with an RNA-protein interaction via two intermediary stages. At each stage, the original RNA-RNA interaction and a newly introduced RNA-protein interaction are designed to coexist. The catalytic RNPs corresponding to the intermediary stages were constructed by employing the Tetrahymena ribozyme together with molecular modeling. Analyses of the RNPs indicate that the protein can fully replace the original role of the RNA-RNA interaction in the starting ribozyme and that the association of a protein with a ribozyme might be beneficial for improving the ribozymatic activity.

Selections for constituting new RNA-protein interactions in catalytic RNP

In vitro and in vivo selection techniques are developed to constitute new RNA-peptide interactions. The selection strategy is designed by employing a catalytic RNP consisting of a derivative of the Tetrahymena ribozyme and an artificial RNA-binding protein. An arginine-rich RNA-binding motif and its target RNA motif in the RNP are substituted with randomized sequences and used for the selection experiments. Previously unknown binding motifs are obtained and the newly established interactions have been indispensable for assembling a catalytically active RNP. The method employed in this study is useful for making customized self-splicing intron RNAs whose activity is regulated by protein cofactors.

Two conserved structural components, A-rich bulge and P4 XJ6/7 base-triples, in activating the group I ribozymes

BACKGROUND

The A-rich bulge of the group I intron ribozyme, a highly conserved structural element in its P5 peripheral region, plays a significant role in activating the ribozyme. The bulge has been known to interact with the P4 stem forming P4 XJ6/7 base-triples in the conserved core. The base-triples by themselves have also been identified as a distinctive element responsible for enhancing the activity of the ribozyme.

RESULTS

A weakly active variant of the Tetrahymena ribozyme lacking the P5 extension was dramatically activated by the addition of an A-rich bulge at the peripheral region, or by replacement of the original P4 XJ6/7 base-triples in the core structure with more stabilized isosteric ones. Biochemical analyses showed that the two methods of activation affect the ribozyme differently.

CONCLUSIONS

The long-range interaction between the A-rich bulge and P4 or additionally stabilized P4 XJ6/7 base-triples can contribute dramatically to activation of the Tetrahymena ribozyme. Both improve the kcat value, which represents the rate of the limiting step of the ribozyme reaction when its binding site is saturated with GTP. However, the bulge or the modified base-triples gave a moderate reduction or considerable increase, respectively, to the Km(GTP) value.

Mispaired P3 region in the hierarchical folding pathway of the Tetrahymena ribozyme

BACKGROUND

The Tetrahymena group I ribozyme folds into a complex three-dimensional structure for performing catalytic reactions. The catalysis depends on its catalytic core consisting of two helical domains, P4-P6 and P3-P7, connected by single stranded regions. In the folding process, most of this ribozyme folds in a hierarchical manner in which a kinetically stable intermediate determines the overall folding rate.

RESULTS

Although the nature of this intermediate has not yet been elucidated, a mispaired P3 stem (alt-P3) appears a likely candidate. To examine the effects of the alt-P3 structure on the kinetic and thermodynamic properties of the active structure of the ribozyme or its P3-P7 domain formation, we prepared and analysed variant ribozymes in which relative stabilities of the original P3 and alt-P3 structure were altered systematically.

CONCLUSION

The results indicate that the alt-P3 structure is not the major rate-limiting factor in the folding process.

Relationship between the self-splicing activity and the solidity of the master domain of the Tetrahymena group I ribozyme

The highly conserved P3-P7 domain of the Group I intron ribozymes is known to contain essential elements, such as the binding site for the cofactor guanosine, required for conducting the splicing reaction. We investigated the domain of the Tetrahymena intron ribozyme and its variants in order to clarify the relationship between its stability and function. We found that the destabilization of the P3-P7 domain facilitates the active structure formation at high magnesium ion concentrations where the formation is retarded for the wild type. The destabilized domain also increases K(GTP)(m) although this can be compensated by increasing the concentration of Mg(2+), indicating that the stable domain is required for establishing a tight guanosine binding site. The results suggest that the stability of the domain affects the rate-limiting step in the RNA folding pathway and also regulates the efficiency of the splicing reaction.

Structure-Function Analysis of RNAs Generated by In Vivo and In Vitro Selection

Today, the concept of Darwinian evolution plays a significant role in studying structure-function relationships concerning known molecules and in helping to design previously unknown molecules with desired functionalities. Results from

Design and development of a catalytic ribonucleoprotein

Ribonucleoproteins (RNPs) consisting of derivatives of a ribozyme and an RNA-binding protein were designed and constructed based upon high-resolution structures of the corresponding prototype molecules, the Tetrahymena group I self-splicing intron RNA and two proteins (bacteriophage lambdaN and HIV Rev proteins) containing RNA-binding motifs. The splicing reaction proceeds efficiently only when the designed RNA associates with the designed protein either in vivo or in vitro. In vivo mutagenic protein selection was effective for improving the capability of the protein. Kinetic analyses indicate that the protein promotes RNA folding to establish an active conformation. The fact that the conversion of a ribozyme to an RNP can be accomplished by simple molecular design supports the RNA world hypothesis and suggests that a natural active RNP might have evolved readily from a ribozyme.

Conserved base-pairings between C266-A268 and U307-G309 in the P7 of the Tetrahymena ribozyme is nonessential for the in vitro self-splicing reaction

P7 is highly conserved in Group I self-splicing intron ribozymes. This region is known to coordinate metal ions and bind a cofactor guanosine required for the self-splicing. To further investigate the fundamental role of the corresponding region in the Tetrahymena ribozyme, we attempted to identify minimal requirements for the base-paired region excluding the guanosine binding site. We discovered that a variety of sequences are eligible and its derivatives possessing extra nucleotide(s) can still conduct the first step of splicing, indicating that no particular base-pairing is essential in this region for conducting the reaction in vitro. The results provide two hypotheses for the fundamental role of this region: (i) if the region contains element(s) that are strictly required in the catalysis, they are not necessarily tightly fixed in the ribozyme and (ii) if not, its fundamental role may simply be to coordinate neighboring regions that are directly involved in the catalysis.

Self-splicing of the Tetrahymena group I ribozyme without conserved base-triples

BACKGROUND

Group I introns share a conserved core region consisting of two domains, P8-P3-P7 and P4-P6, joined by four base-triples. We showed previously that the T4 td intron can perform phosphoester transfer reactions at two splice sites in the absence of both P4-P6 and the conserved base-triples, whereas it is barely able to perform the intact splicing reaction due to the difficulty of conducting the sequential reactions.

RESULTS

Based on previous findings, we constructed a bimolecular ribozyme lacking a large portion of P4-P6 and the base-triples from the Tetrahymena intron, on the assumption that the long-range interactions of the peripheral regions in the two RNAs can compensate for the deteriorated core. The bimolecular ribozyme performed the intact splicing reaction.

CONCLUSION

The present analysis indicates that the base-triples are nonessential, but that L4 and the distal part of P4 in P4-P6 are important for conducting the splicing reaction. The reconstituted self-splicing ribozyme provides an amenable system for analysing the role(s) of elements in the core region in the self-splicing reaction mechanism.

A deteriorated triple-helical scaffold accelerates formation of the Tetrahymena ribozyme active structure

The Tetrahymena group I ribozyme requires a hierarchical folding process to form its correct three-dimensional structure. Ribozyme activity depends on the catalytic core consisting of two domains, P4-P6 and P3-P7, connected by a triple-helical scaffold. The folding proceeds in the following order: (i) fast folding of the P4-P6 domain, (ii) slow folding of the P3-P7 domain, and (iii) structure rearrangement to form the active ribozyme structure. The third step is believed to directly determine the conformation of the active catalytic domain, but as yet the precise mechanisms remain to be elucidated. To investigate the folding kinetics of this step, we analyzed mutant ribozymes having base substitution(s) in the triple-helical scaffold and found that disruption of the scaffold at A105G results in modest slowing of the P3-P7 folding (1.9-fold) and acceleration of step (iii) by 5.9-fold. These results suggest that disruption or destabilization of the scaffold is a normal component in the formation process of the active structure of the wild type ribozyme.

A small structural element, Pc-J5/5a, plays dual roles in a group IC1 intron RNA

The P4-P6 domain of group IC1 intron ribozymes such as that of the Tetrahymena autonomously folds into a hairpin-shaped structure in which the J5/5a region serves as a hinge. Phylogenetic comparisons of these IC1 introns suggested that the J5/5a region (termed Pc-J5/5a motif) in a subclass of IC1 introns such as the one from Pneumocystis carinii functions not only as a hinge but also as a receptor for a GAAA-tetraloop. We investigated the role of this motif by transplanting the structural unit, Pc-J5/5a motif, of Pneumocystis carinii into the P4-P6 domain of the Tetrahymena intron. The results showed that the Pc-J5/5a motif binds to a GAAA loop with high affinity and also facilitates the bending of the Tetrahymena P4-P6 domain more positively than the original J5/5a region.

Characterization of P8 and J8/7 elements in the conserved core of the tetrahymena group I intron ribozyme

The universally conserved core region in the group I intron ribozymes is responsible for its catalytic activity. The structural elements in this region have been known to organize the active site of this class of ribozymes. However, it has been unclear whether all elements are requisite or some elements are dispensable for conducting the catalysis. To investigate the necessity of these elements in the catalysis, we prepared and examined a series of mutants having a nick or deletion in these elements. In this report, we show that two elements, P8 and 5' portion of J8/7, are nonessential for activity.

Antibiotic inhibition of RNA catalysis: neomycin B binds to the catalytic core of the td group I intron displacing essential metal ions

The aminoglycoside antibiotic neomycin B induces misreading of the genetic code during translation and inhibits several ribozymes. The self-splicing group I intron derived from the T4 phage thymidylate synthase (td) gene is one of these. Here we report how neomycin B binds to the intron RNA inhibiting splicing in vitro. Footprinting experiments identified two major regions of protection by neomycin B: one in the internal loop between the stems P4 and P5 and the other in the catalytic core close to the G-binding site. Mutational analyses defined the latter as the inhibitory site. Splicing inhibition is strongly dependent on pH and Mg2+ concentration, suggesting electrostatic interactions and competition with divalent metal ions. Fe2+-induced hydroxyl radical (Fe-OH.) cleavage of the RNA backbone was used to monitor neomycin-mediated changes in the proximity of the metal ions. Neomycin B protected several positions in the catalytic core from Fe-OH. cleavage, suggesting that metal ions are displaced in the presence of the antibiotic. Mutation of the bulged nucleotide in the P7 stem, a position which is strongly protected by neomycin B from Fe-OH. cleavage and which has been proposed to be involved in binding an essential metal ion, renders splicing resistant to neomycin. These results allowed the docking of neomycin to the core of the group I intron in the 3D model.

P5abc of the Tetrahymena ribozyme consists of three functionally independent elements

P5abc domain of Tetrahymena LSU intron functions as an activator that is not essential for but enhances the activity of the ribozyme either when present in cis or when added in trans. This domain contains three regions (A-rich bulge, L5b, and L5c) that have been demonstrated to interact with the rest of the intron. Although these regions are presumably important for efficient activation, the role of each element is not understood in the mechanism of activation. We employed circularly permuted introns and examined the roles of each element. The results show that each of the three elements can activate the intron independently. We also found that a correlation between the activation by P5abc and the physical affinity of P5abc to the intron exists.

A non-directed, hydroxylamine-generated suppressor mutation in the P3 pairing region of the bacteriophage T4 td intron partially restores self-splicing capability

Hydroxylamine (HA) mutagenesis of an HA-induced splicing-defective bacteriophage T4 td intron mutant with a mutation in the intron P3 RNA pairing region was used to generate pseudorevertants. Because HA can only cause GC to AT transitions, the original mutant (H104A) could not undergo true reversion, yet the compensatory mutation on the opposite side of the P3 helix, which was complementary to the original H104A mutation, could occur. A pseudorevertant was isolated that contained both the original H104A mutation and the compensatory mutation HS9. By phenotypic and molecular genetic criteria, this double mutant (H104A-HS9) was shown to be able to undergo significant RNA splicing, thus confirming the existence and functional importance of the long-range P3 pairing region in this phage intron. The second-site suppressor mutation (HS9) was isolated by phage cross and also exhibited some self-splicing ability. A correlation exists between the strength of P3 helix Watson-Crick base pairing and the apparent level of splicing when wild-type, H104A, HS9, and H104A-HS9 are compared. This suggests that the primary role of the P3 RNA pairing region in the T4 td intron is structural in contributing to the critical RNA secondary structure.

A modified group I intron can function as both a ribozyme and a 5' exon in a trans-exon ligation reaction

Here, we show that a single RNA molecule derived from a group-I intron can provide the catalytic activity, the substrate recognition domain and the attacking nucleophile in a reaction that mimics the exon ligation step of splicing. To accomplish this reaction, we have linked a 5' exon sequence to the 3' end of an attenuated form of the self-splicing Tetrahymena rRNA intron. The ribozyme (I-E1) attacks an oligoribonucleotide analog of the 3' splice site (I'-E2) to generate a product containing ligated exons (I-E1-E2) and a small intron fragment (I'). Two modified introns were constructed and tested for activity. A construct designed to interact with the 3' splice site through intermolecular P9.0 and P10 helices was found to be inactive due to failure to form a stable ribozyme-substrate complex. A second modified intron and substrate combination was engineered, in which the complex was further stabilized by an intermolecular P9.2 helix. In this case, stable complexes and reaction products were identified. The reaction efficiency was low compared to splicing of the unmodified intron-containing precursor, and will be optimized in future experiments. Following optimization, we believe that this system may be exploited to examine the functional consequences of a wide variety of 3' splice-site modifications, and may provide the basis for development of highly selective trans-acting ribozymes.

Selection of novel forms of a functional domain within the Tetrahymena ribozyme

P5abc is an RNA structure within the self-splicing Tetrahymena group I intron that provides an activation function to the remainder of the ribozyme, either when present in cis or when added in trans. This 69-nucleotide activator domain was replaced with randomized sequence of 20 or 40 nt in length, and individuals among these pools with sequences that could functionally replace P5abc were selected. The basis of selection was a reaction in which two separate halves of the ribozyme became joined; selection was completed by reverse transcription and the polymerase chain reaction, using primers with sequence from either side of the ligation junction. Selectant sequences fell into three families that appear unrelated to P5abc; for example they lack the A-rich bulge thought to be a important feature of P5abc. Thus, rather than defining some consensus sequence for activator domains, this result reveals a certain tolerance in the ribozyme in its ability to derive activation function from diverse sequence types. In the context of splicing precursor RNA, the new sequences supported self-splicing, but failed to activate a related reaction, hydrolysis of the 3' splice site, implying that this region of the intron can differentially control two related reactions.

A region of group I introns that contains universally conserved residues but is not essential for self-splicing

The catalytic core of the self-splicing group I intron RNAs is composed of six paired regions together with their connecting sequences; these are thought to form two elongated domains, with paired regions P5, P4, and P6 aligned along one axis and P8, P3, and P7 along the other. Most of the very highly conserved residues of the group I introns lie in or near P7, but two occur in L4, the internal loop connecting P4 and P5. It is generally believed that such bases are conserved because they are essential for splicing. Mutants were created in a member of each of the two major subclasses of group I introns, in which P5, L4, and the distal portion of P4 were deleted. Splicing activity was still detected in these mutants, albeit substantially weakened; splicing was accurate and occurred by the normal group I mechanism, with addition of a guanosine molecule to the intron. Thus the deleted region, containing two universally conserved bases, is not essential but facilitates splicing. Another reaction characteristic of group I introns, hydrolysis of the 3' splice site, was less severely affected by the deletions. The results are discussed in terms of the prevailing three-dimensional model for the core structure of the group I introns.

A self-splicing group I intron in the nuclear pre-rRNA of the green alga, Ankistrodesmus stipitatus

The nuclear small subunit ribosomal RNA gene of the unicellular green alga Ankistrodesmus stipitatus contains a group I intron, the first of its kind to be found in the nucleus of a member of the plant kingdom. The intron RNA closely resembles the group I intron found in the large subunit rRNA precursor of Tetrahymena thermophila, differing by only eight nucleotides of 48 in the catalytic core and having the same peripheral secondary structure elements. The Ankistrodesmus RNA self-splices in vitro, yielding the typical group I intron splicing intermediates and products. Unlike the Tetrahymena intron, however, splicing is accelerated by high concentrations of monovalent cations and is rate-limited by the exon ligation step. This system provides an opportunity to understand how limited changes in intron sequence and structure alter the properties of an RNA catalytic center.

Effects of mutations of the bulged nucleotide in the conserved P7 pairing element of the phage T4 td intron on ribozyme function

The P7 element of group I introns contains a semiconserved "bulged" nucleotide, a C in group IA introns (nt 870 in the td intron) and an A in group IB introns [Cech, T.R. (1988) Gene 73, 259-271]. Variants U870, G870, and A870, isolated by a combination of in vitro and in vivo genetic strategies, indicate that C and A at position 870 are consistent with splicing whereas U and G are not. Although mutants G870 and U870 could be activated in vitro by increasing the Mg2+ concentration, their Km for GTP at pH 7 was 20-100-fold elevated, and they were unable to undergo site-specific hydrolysis. The dependence of the mutants on high guanosine concentrations could be substantially overcome by an increase in pH, suggesting that a tautomeric change, which makes U and G mimic C and A, is responsible for restoring function. In contrast to the striking Km effect, Vmax for the mutants differed by less than a factor of 2 from the wild type. Furthermore, streptomycin, an aminoglycoside antibiotic that competes with guanosine for its binding site, inhibited splicing of the U870 and G870 constructs at least as well as of the C870 and A870 variants, indicating that the guanosine-binding site of the mutants is proficient at interacting with a guanidino group. While our experiments argue against a hydrogen-bonding interaction between the C6-O of the cofactor and C4-NH2 of the bulged nucleotide, they are consistent with other models in which the C4-NH2 and/or N3 groups of the bulged C are involved in establishing an active ribozyme.

Three-dimensional folding of Tetrahymena thermophila rRNA IVS sequence: a proposal

We studied the Tetrahymena thermophila rRNA IVS sequence with the aim of obtaining a model of the structure characterized by the bases proximity of the self-reactions sites. The considered sequence kept up those fragments essential for its catalytic activity as demonstrated by deletion mutants. The first step was the theoretical analysis with a computer method previously proposed, to find optimal free energy secondary structures with the required features, under the suitable constrains. Then we tried folding the obtained secondary structures, in low resolution tertiary models, which kept up the proximity of the catalytic sites also in the space. The proposed tertiary folding seems to provide for a better explanation to the transesterification mechanisms and moreover it is in good agreement with the experimental data (activity of mutants, enzymatic cleavages, phylogenetically conserved regions).

Mutational analysis of conserved nucleotides in a self-splicing group I intron

We have constructed all single base substitutions in almost all of the highly conserved residues of the Tetrahymena self-splicing intron. Mutation of highly conserved residues almost invariably leads to loss of enzymatic activity. In many cases, activity could be regained by making additional mutations that restored predicted base-pairings; these second site suppressors in general confirm the secondary structure derived from phylogenetic data. At several positions, our suppression data can be most readily explained by assuming non-Watson-Crick base-pairings. In addition to the requirements imposed by the secondary structure, the sequence of the intron is constrained by "negative interactions", the exclusion of particular nucleotide sequences that would form undesirable secondary structures. A comparison of genetic and phylogenetic data suggests sites that may be involved in tertiary structural interactions.

An ultraviolet-inducible adenosine-adenosine cross-link reflects the catalytic structure of the Tetrahymena ribozyme

When a shortened enzymatic version of the Tetrahymena self-splicing intervening sequence (IVS) RNA is placed under catalytic conditions and irradiated at 254 nm, a covalent cross-link forms with high efficiency. The position of the cross-link was mapped by using three independent methods: RNase H digestion, primer extension with reverse transcriptase, and partial hydrolysis of end-labeled RNA. The cross-link is chemically unusual in that it joins two adenosines, A57 and A95. Formation of this cross-link depends upon the identity and concentration of divalent cations present and upon heat-cool renaturation of the IVS in a manner that parallels conditions required for optimal catalytic activity. Furthermore, cross-linking requires the presence of sequences within the core structure, which is conserved among group I intervening sequences and necessary for catalytic activity. Together these correlations suggest that a common folded structure permits cross-linking and catalytic activity. The core can form this structure independent of the presence of P1 and elements at the 3' end of the IVS. The cross-linked RNA loses catalytic activity under destabilizing conditions, presumably due to disruption of the folded structure by the cross-link. One of the nucleotides participating in this cross-link is highly conserved (86%) within the secondary structure of group I intervening sequences. We conclude that A57 and A95 are precisely aligned in a catalytically active conformation of the RNA. A model is presented for the tertiary arrangement in the vicinity of the cross-link.

Deletion-tolerance and trans-splicing of the bacteriophage T4 td intron. Analysis of the P6-L6a region

Non-directed mutagenesis and phylogenetic comparison suggest that certain elements of the bacteriophage T4 td group Ia intron are dispensable to self-splicing. The L6-P6a-L6a region was identified as a potential non-essential element, and was removed by sequential deletions extending from the L6a loop toward the P6 pairing. Assays for splicing indicate that as long as the P6 pairing is maintained, the 1016 nucleotide td intron can be reduced to less than 250 nucleotides while maintaining function in vivo and in vitro. The P6 pairing appears to be essential for splicing while P6a is not. In addition, a spontaneous pseudorevertant of a splicing-defective deletion was isolated and shown to result from a single nucleotide change in the predicted L6a loop. This genetic suppressor mimics the ability of Mg2+ to reverse the phenotype of the deletion, suggesting that function is restored by structural stabilization of P6. The tolerance of this region to deletion prompted us to split the ribozyme core in L6a, to generate precursors that might function in trans. Indeed, the two half-molecules do associate to form a bimolecular complex that yields accurately ligated exons both in vitro and in vivo. The biological implications of these results, as well as the usefulness of trans-splicing for generating unprocessed precursors in vitro are discussed.

Sequence specificity of the P6 pairing for splicing of the group I td intron of phage T4

Seventeen non-directed td- (thymidylate synthase-deficient) splicing-defective mutations isolated in phage T4 were localized within the catalytic core of the ribozyme. All of the mutations occur in conserved structural elements that form part of the td intron core secondary structure. Remarkably, seven of the seventeen independently isolated mutations clustered in the dinucleotide 5' element (P6[5']) of the putative two-base-pair P6 stem. An analysis of this region was undertaken by site-directed mutagenesis of the plasmid-borne td gene, leading to the following findings: First, the short P6 pairing in the td secondary structure model was verified with appropriate P6[5'] and P6[3'] compensatory mutations. Second, all P6[5'] and P6[3'] mutants are defective in the first step of splicing, guanosine-dependent 5' splice site cleavage, whereas their activity at the 3' splice site is variable. Third, residual in vitro splicing activity of the mutants altered on only one side of the P6 pairing is correlated with the ability to form an alternative two-base-pair P6 stem. Fourth, the degree to which the compensatory mutants have their splicing activity restored is highly condition-dependent. Restoration of phenotype of the compensatory P6[5']:[3'] constructs is weak under stringent in vitro conditions as well as in vivo. This sequence specificity is consistent with phylogenetic conservation of the P6 pairing elements in group I introns, and suggests either structural constraints on the P6 stem or a dual function of one or both pairing elements.

Catalytic activity is retained in the Tetrahymena group I intron despite removal of the large extension of element P5

We have made sizeable internal deletions within the self-splicing group I intron of Tetrahymena thermophila. Deletions were made in a piecewise manner in order to remove secondary structural elements thought to be extraneous to the catalytic center of the molecule. The resulting deletion mutants retain self-splicing activity, albeit under modified reaction conditions that enhance duplex stability. Considering those portions of the molecule that can be deleted without a loss of catalytic activity, one is left with a catalytic center of approximately 130 nucleotides that is solely responsible for the molecule's activity.