The nucleolytic ribozymes use general acid–base catalysis to contribute significantly to their rate enhancement. The VS (Varkud satellite) ribozyme uses a guanine and an adenine nucleobase as general base and acid respectively in the cleavage reaction. The hairpin ribozyme is probably closely similar, while the remaining nucleolytic ribozymes provide some interesting contrasts.
- catalytic mechanism
- general acid–base catalysis
- hairpin ribozyme
- RNA catalysis
- Varkud satellite ribozyme (VS ribozyme)
The nucleolytic ribozymes
The nucleolytic ribozymes are a class of five different ribozymes that cleave or ligate RNA at a specific internal site . They carry out cleavage by transesterification reactions in which the 2′-oxygen attacks the 3′-phosphorus, with departure of the 5′-oxygen to leave a cyclic 2′,3′-phosphate. In the ligation reaction the 5′-oxygen attacks the phosphorus (as the cyclic phosphate) with breakage of the 2′-O–P bond. The reactions are accelerated ≥105-fold when catalysed by the ribozymes.
At present, five distinct nucleolytic ribozymes are known. These are the hammerhead, hairpin, HDV (hepatitis delta virus), VS (Varkud satellite) and GlmS ribozymes (Figure 1). Their structures are all different, although it can be noted that the hammerhead, hairpin and VS ribozymes are based upon helical junctions, whereas the HDV and GlmS ribozymes are based upon complex pseudoknots. Perhaps this represents two alternative ways to create a stable structure in small autonomously folding RNA species. Crystal structures have been determined for all these ribozymes except for the VS [2–9], for which a structure has been determined from SAXS (small-angle X-ray scattering) data .
The VS ribozyme
The sequence and secondary structure of the VS ribozyme  indicates that it comprises seven helical segments that are connected by three different three-way junctions (Figure 1). A trans-acting core of the ribozyme can be released, formed by five helices (II–VI) connected by two of the junctions to form a nominal H shape. Cleavage and ligation reactions occur within the internal loop of stem–loop I, which is connected to the end of helix II. Collins and co-workers showed that the terminal loop of helix I interacts with that of helix V .
We have used SAXS in solution to determine the shape of the complete ribozyme comprising helices I–VII , as well as its components. Scattering curves were analysed to reconstruct a family of closely related structures. These enabled us to propose a model for the complete structure (Figure 2). This was in good agreement with an earlier model based on studies of the individual three-way junctions [13,14].
The critical catalytic components of the VS ribozyme have been identified. An internal loop within helix VI (termed the A730 loop) was found to be sensitive to ethylation interference , and sequence variants in this loop led to large effects on cleavage  and ligation  activity without measureable changes to the structure. Within the A730 loop, substitution of A756 resulted in loss of activity by three orders of magnitude . NAIM (nucleotide analogue interference mapping) experiments revealed that this position was the most sensitive nucleotide to substitution by a range of analogues  and UV cross-linking data placed A756 physically close to the cleavage site in the substrate . It was shown that some activity could be retained when A756 was replaced by an imidazole nucleoside , suggesting a role in general acid–base catalysis akin to the protein enzyme RNaseA. A second key nucleobase was found in the internal loop of the substrate helix I. Changes to G638 resulted in reduced activity by four orders of magnitude, while leaving the binding affinity of the substrate to the ribozyme unaltered . These studies collectively point to an involvement of the nucleobases of A756 and G638 in the catalytic mechanism of the VS ribozyme.
The catalytic mechanism of the VS ribozyme
The nucleobases of A756 and G638 might lower the activation barrier by stabilization of the transition state, but present evidence points towards general acid–base catalysis by these nucleobases as providing significant rate enhancement. The pH-dependence of the reaction provides insight into this. General acid–base catalysis requires that the participating nucleobases be in the correct state of ionization, i.e. that the acid is protonated and the base unprotonated at the outset of the reaction. The observed rate of reaction (kobs) will be the rate of cleavage catalysed by the ribozyme in the correct state of protonation (kcat) multiplied by the product of the fractions of protonated acid and unprotonated base (fA and fB respectively), i.e. fA and fB can be readily calculated if the pKa values of the nucleobases are known. Assuming values of around 5 and 9 for adenine and guanine respectively, this generates a bell-shaped curve of fA·fB against pH. The experimentally determined pH-dependence of the cleavage reaction in the presence of a high concentration of Mg2+ ions is indeed bell-shaped with pH (Figure 3), fitting a double-ionization model with apparent pKa values of 5.2 and 8.4 . The pH profile changed as expected when G638 was replaced by nucleobases of different pKa.
While these results are consistent with general acid–base catalysis by A756 and G638, pH profiles alone cannot tell us which nucleobase is the acid, and which the base in the cleavage reaction. This was achieved using 5′-PS (5′-phosphorothiolate) substitution at the scissile phosphate. We found that the cleavage activity of VS A756G was impaired 1000-fold cleaving the oxy (5′-PO) substrate, but the activity was completely restored for the 5′-PS-containing substrate . Sulfur is a much better leaving group than oxygen, and therefore no longer requires acid catalysis, and we conclude that A756 is the general acid for the cleavage reaction. By contrast, the rate of cleavage of a G638DAP (i.e. the guanine at position 638 has been replaced by diaminopurine) plus 5′-PS substrate was similar to that observed for a G638DAP plus 5′-PO substrate, and both were lower than the natural sequence. The pH-dependence of the cleavage reaction corresponded to a single ionization, with a pKa value of 5.3, consistent with general base catalysis by the diaminopurine at position 638. Thus all the available data are consistent with a catalytic mechanism for the VS ribozyme cleavage reaction in which G638 acts as general base to deprotonate the 2′-oxygen nucleophile, and A756 is the general acid protonating the 5′-oxyanion leaving group (Figure 4). Of course, by the principle of microscopic reversibility, protonated G638 should act as the general acid protonating the 2′-oxyanion leaving group, and unprotonated A756 as general base deprotonating the 5′-oxygen nucleophile that attacks the cyclic phosphate in the ligation reaction.
The hairpin ribozyme
The hairpin ribozyme is centred on a four-way helical junction (Figure 1). Adjacent arms A and B contain loops that include most of the nucleotides shown to be essential for catalytic activity, and the site of cleavage/ligation is located in arm A. The four-way junction is not essential for activity, but assists the folding of the ribozyme under physiological conditions [23–25], rather like the loops of the hammerhead ribozyme [26,27]. The loops are required to dock together to generate the active form of the ribozyme, shown first by FRET (fluorescence resonance energy transfer)  and later by crystallography . Single-molecule FRET studies of both a simplified ribozyme lacking two helices  and the complete ribozyme  revealed metal ion-dependent dynamics, with repeated docking and undocking of the loops. Single-molecule FRET was also used to follow the cleavage and ligation reactions in the individual hairpin ribozyme .
Two key nucleotides have been identified in the hairpin ribozyme. The first was G8, located in the internal loop of the A helix, on the opposite strand from the scissile phosphate. Nucleotide substitution at that position gave rates of cleavage reduced by two orders of magnitude in both the simplified form [33–35] and the junction form  of the ribozyme, without affecting ion-induced folding. A crystal structure of a transition-state analogue of the ribozyme  revealed G8 hydrogen bonded to the 2′-oxygen and the proS non-bridging oxygen of the scissile phosphate and thus positioned to participate in the catalytic chemistry. The structure also revealed A38 (contributed by loop B) juxtaposed with the scissile phosphate, making hydrogen bonds to the 5′-oxygen and the proR oxygen. NAIM experiments showed that ligation activity was sensitive to functional group changes at this position , whereas its removal led to a 10 000-fold loss in activity .
Similarity between the VS and hairpin ribozymes
There are thought-provoking similarities between the VS and hairpin ribozymes. Both appear to generate their active conformations by the intimate association of two internal loops, and both have critical adenine and guanine nucleobases. Moreover, the relative positioning of these and the scissile phosphate is the same in the two ribozymes.
This suggests that the two ribozymes could share a common mechanism. The evidence for general acid–base catalysis in the hairpin ribozyme is more controversial than for the VS ribozyme, but we believe that all of the available data are consistent with this proposal. It has long been recognized that G8 and A38 of the hairpin ribozyme might participate in general acid–base catalysis [4,29,35,39]. The observed pH-dependence of the reaction is different from that of the VS ribozyme, with both cleavage and ligation activity rising with pH until a plateau level is achieved around neutrality, corresponding to a pKa close to 6. At first sight, this appears to correspond to a single pKa [32,40]; however, it is perfectly understandable in terms of a higher pKa for the guanine base in this ribozyme. Replacing G8 with nucleobases of lower pKa (such as diaminopurine) provides good evidence in support of general acid–base catalysis . Ribozyme substituted with imidazole nucleoside at position 8 also gave bell-shaped pH profiles for both cleavage and ligation .
The evidence for and against this mechanism in the hairpin ribozyme is discussed in detail elsewhere [41a]. But, in summary, I believe that, despite very different architecture, the hairpin and VS ribozymes share a common mechanism in which a significant contribution to the catalytic rate enhancement is provided by general acid–base catalysis involving an adenine and a guanine nucleobase, with the guanine acting as general base in the cleavage reaction.
General acid–base catalysis is unlikely to be the only source of catalytic rate enhancement. Orientation of the attacking nucleophile may increase the reaction rate to some degree, although such alignment is unlikely to contribute a large rate enhancement. The transition state may be stabilized by bonding with nucleobases (as observed in the crystal ), and this is not inconsistent with general acid–base catalysis. But if, for example, A756 of the VS ribozyme were acting solely by stabilizing the phosphorane, then the 1000-fold effect of substitution by guanine should not be reversed by 5′-PS substitution at the scissile phosphate . The pH-dependence of the cleavage reaction by the G638DAP VS ribozyme indicates that proton transfer contributes >100-fold to its catalytic power , so that general acid–base catalysis is likely to be a very significant, if not the major, source of the ribozyme's observed rate enhancement.
The other nucleolytic ribozymes
To what extent can the remaining known nucleolytic ribozymes fit with the pattern of the VS and hairpin ribozymes? It seems that all five probably exploit general acid–base catalysis, and all use nucleobases for at least some of this. In two cases, guanine bases are again used. In the hammerhead ribozyme, the nucleobase of G12 is well-placed to act as a general base in the cleavage reaction , in agreement with experiments of Han and Burke . Similarly, in the GlmS ribozyme, a guanine nucleobase (G40 in Thermoanaerobacter tengcongenis or G33 in Bacillus anthracis) is positioned to act as general base in cleavage [9,44,45]. Thus guanine nucleobases appear to be used in general acid–base catalysis in four of the five nucleolytic ribozymes. There is no evidence for adenine playing a specific role in the remaining ribozymes. However, there are data suggesting that cytosine (also of low pKa) acts as a general acid in the cleavage reaction of the HDV ribozyme [46,47]. The remaining catalytic players of the HDV and GlmS ribozymes are very interesting, since neither is a nucleobase. The HDV ribozyme appears to employ a hydrated metal ion as the general base used to deprotonate the 2′-hydroxy group [46,48]. But, most interestingly, the GlmS ribozyme uses the amine of its bound glucosamine 6-phosphate (it is, of course, a riboswitch too) as the general acid to protonate the 5′-oxyanion leaving group [9,44,49]. The ability to exploit bound small-molecule cofactors in catalysis is characteristic of protein enzymes, and GlmS shows that ribozymes can do this too. In a sense, the use of exogenous guanosine by the group I ribozyme in the first step of splicing is similar [50,51], although, strictly speaking, that is not used catalytically since it is not released at the end of the reaction. However, the ability to recruit and use small molecules to enhance catalytic power provides a precedent, and a suggestion of how peptide and ultimately protein-based enzymes might have emerged from an RNA world .
Are there other nucleolytic ribozymes still to be discovered?
The GlmS ribozyme was the only truly novel ribozyme to be discovered  within the last two decades, and even that was over 5 years ago. New examples of known ribozymes have been turning up reasonably regularly. These include hammerhead ribozymes in plants  and mammals , and bioinformatics searching has indicated that they are widespread in the genomes of bacteria, chromalveolata, plantae and metazoa kingdoms . HDV ribozymes have been found encoded in the human genome , and GlmS ribozymes are widespread in the Gram-positive bacteria. But are there genuinely novel nucleolytic ribozymes still to be discovered?
Although it is hard to answer this question with any certainty, there may be no more to be found. And in any case, the number is probably rather small, if not zero. It may be simply that there are rather few ways to solve the problem of how to make a functional nucleolytic ribozyme. Although the structures of proteins (with their 20 different amino acids) may be almost infinitely tuneable, that may not be true for RNA. Clearly RNA can bind ligands with impressive selectivity, but catalysis may be trickier to achieve. It is interesting that the hairpin and VS ribozyme seem to have converged on apparently very similar ways to perform their chemistry, yet the fact that only one VS ribozyme has been discovered in a single isolate of Neurospora (found in the eponymous village in India, i.e. Varkud) suggests that it really is extremely rare. It is to be hoped that there are further nucleolytic ribozymes, but it just may turn out that we have already found them all.
Ribozyme studies in this laboratory are funded by Cancer Research UK.
I especially thank Dr Tim Wilson and Dr Joe Piccirilli for many valuable discussions and major contributions to the work described here.
NACON VIII: 8th International Meeting on Recognition Studies in Nucleic Acids: An Independent Meeting held at The Edge, University of Sheffield, Sheffield, U.K., 12–16 September 2010. Organized by Mike Blackburn, Mark Dickman, Jane Grasby, David Hornby, Chris Hunter, John Rafferty, Jim Thomas, David Williams and Nick Williams (Sheffield, U.K.).
Abbreviations: FRET, fluorescence resonance energy transfer; HDV, hepatatis delta virus; NAIM, nucleotide analogue interference mapping; 5′-PS, 5′-phosphorothiolate; SAXS, small-angle X-ray scattering; VS, Varkud satellite
- © The Authors Journal compilation © 2011 Biochemical Society