Ribonuclease P: Structure and Catalysis
Department of Biochemistry
The Ohio State University
Columbus, Ohio 43210
Department of Molecular, Cellular and Developmental Biology
New Haven, Connecticut 06520
RNase P, ribosomes, and (parts of) spliceosomes are the three RNA enzymes found in nature to date (Guerrier-Takada et al. 1983; Valadkhan and Manley 2001; Steitz and Moore 2003). Other “ribozymes” cleave themselves during their modes of action inside cells and so are not, by definition, enzymes that catalyze multiple turnover while remaining unchanged. Nevertheless, the whole panoply of catalytic RNAs contributes to our understanding of RNA molecules with functions other than those of information content.
All known RNase P enzymes, with the possible exception of those in some plant chloroplasts and trypanosome mitochondria, are ribonucleoproteins (RNPs) made up of an essential RNA subunit and a varying number of protein subunits depending on the source: one in Bacteria, at least four in Archaea, and at least nine in Eukarya (Jarrous 2002; Evans et al. 2006; Walker and Engelke 2006). Despite this remarkable difference in the subunit composition of RNase P in the three domains of life, its primary role in all cells is to catalyze a Mg2+-dependent phosphodiester hydrolysis reaction during 5′ maturation of different precursor tRNAs (ptRNAs). Whereas the fidelity and efficiency of RNase P catalysis would be expected to be dictated by macromolecular recognition, akin to all cellular processes, RNase P does exemplify a rare situation wherein there are striking differences in (1) the sequences around the site of cleavage in its numerous substrates, and (2) the natural variants of the enzyme as reflected in their different sequences/structures and subunit makeup. Despite this plurality, molecular rearrangements in the enzyme–substrate complex somehow permit a common catalytic scheme for a phosphodiester hydrolysis reaction. We summarize new data on enzyme structure and substrate cleavage-site selection that begin to explain some facets of the plasticity that underlies the uniformity in RNase P catalysis.
Structure of the Bacterial RNase P RNA
Although it has been more than two decades since the discovery of the RNA catalyst in bacterial RNase P (Guerrier-Takada et al. 1983), detailed structural insights were not available until recently. The bacterial RNase P RNA (RPR) is typically about 400 nucleotides long. Phylogenetic covariation analysis of a comprehensive set of bacterial RPR sequences established a well-defined secondary structure and identified tertiary interactions (James et al. 1988; Brown 1999; Evans et al. 2006). This analysis also classified bacterial RPR into two major classes with distinct structural modules: type A, represented by Escherichia coli, is found in most bacteria, and type B, represented by Bacillus subtilis, is found in the low-GC content gram-positive bacteria. The universally conserved nucleotides of RPRs are found in five conserved regions (termed CR I through V) that are distal to each other in the primary sequence (Fig. 1). Biochemical studies dissected both types A and B RPRs into two domains: a specificity (S) domain which houses conserved nucleotides that recognize the T stem-loop of the ptRNA substrate; and a catalytic (C) domain which can (1) recognize the leader, acceptor stem, and the ACCA sequence at the 3′ end (by Watson-Crick base-pairing), and (2) cleave the leader sequence of a ptRNA in the presence of the bacterial RNase P protein cofactor (Guerrier-Takada and Altman 1992; Pan 1995; Pan et al. 1995; Loria and Pan 1996; Christian et al. 2002; Harris and Christian 2003).
The first experimental insights on RPR architecture were derived from crystal structures of the S-domain from both types A and B bacterial RNase P. More recently, the crystal structures of the full-length Thermotoga maritima and Bacillus stearothermophilus RPRs have been solved (Fig. 2) (Kazantsev et al. 2005; Torres-Larios et al. 2005). In addition to revealing an overall similar fold, these two structures of bacterial RPR were also fortuitously complementary in resolving disordered or low-resolution regions in each other and have collectively furnished valuable perspectives on both substrate binding and catalysis.
In both type A and type B RPRs, various coaxially stacked helices, held together by tertiary contacts, form the edifice of the three-dimensional fold which can be viewed as two tiers (each of one-helix thickness; Fig. 2). Despite notable differences in the structures of the type A and type B RPRs, the larger layer 1 that encompasses both the substrate-binding regions and the putative catalytic center is similar in both cases. Characteristics of the structure such as the short/long-range interactions that mediate intra- and inter-domain contacts, irregular structural motifs, and unusual backbone contacts are described elsewhere in detail (Kazantsev et al 2005; Torres-Larios et al. 2005), but some key aspects are noted here.
In the T. maritima RPR (type A), layer 2 serves as the underpinning for the larger layer 1 that includes paired regions P1 to P12 and P15 to P17 as well as various joining regions (J11/12, J12/11, and J5/15). Layer 2, which serves an organizing function, is made up of three paired regions (P13, P14 and P18) that contribute to the overall stability through tertiary interactions between the P8/P9 helical stack in the S-domain and the tetraloops in L14 and L18 (Fig. 2a); P8/P9 thus acts as a brace which brings together distal helices in the S- (i.e., P13/P14 stack) and C-domains (i.e., P18). These intramolecular anchors, anticipated from phylogenetic sequence analysis/cross-linking/mutagenesis/stability studies of bacterial RNase P (Chen et al. 1998; Massire et al. 1998; Pomeranz Krummel and Altman 1999b), are reminiscent of near-identical contacts in the hammerhead and group I introns wherein certain loop adenine bases contact the minor groove of two consecutive base pairs in RNA helices.
In the B. stearothermophilus RPR (type B), the absence of P14 and P18 rules out the organizing interactions seen in layer 2 of type A RPR. However, A-minor and ribose zipper interactions between L15.1 (in layer 2) and L5.1 (in layer 1) contribute to a different set of contacts between the two layers (Fig. 2b) (Kazantsev et al. 2005); although P5.1 and P15.1 are absent in type A RPR, these elements in type B RPRs are rich in conserved nucleotides.
Surface maps reveal the modular arrangement of the S- and C-domains in bacterial RPR (Fig. 3). The S-domain structure is identical in isolation and as part of the entire molecule. Like protein enzymes, there is a well-defined substrate-binding crevice in the S-domain in both type A and type B RPRs (Krasilnikov et al. 2003, 2004). Two semi-continuous helical stacks intersect near the hinge at the bottom of the crevice. Both the single-stranded regions J11/12 and J12/11 adopt penta-nucleotide, interweaving T-loop motifs and together with P9 to P11 line the cleft. Nucleotides demonstrated to be involved in T stem-loop interactions are positioned in the walls of the crevice with their bases exposed to the solvent.
The C-domain is made up of four main helical elements (the P1/P4/P5, P2/P3, and P15/16 stacks in layer 1 and P18 in layer 2) that are connected by linker regions containing several conserved nucleotides. Notably, the C-domain adopts a concave structure with a cleft/groove that runs parallel to the P1/P4/P5 stack (Torres-Larios et al. 2005); in addition, various conserved nucleotides in nonhelical regions (e.g., J5/15, J2/4), all of which were previously identified as critical for substrate binding or catalytic function, are found halfway through this helical stack.
Prior to the availability of crystal structures, computer modeling was employed to build atomic scale three-dimensional models of both type A and type B bacterial RPRs using spatial relationship data gathered from cross-linking experiments and comparative sequence analysis (e.g., a tetraloop–tetraloop receptor interaction). The modeling exploited a hierarchical approach beginning with the bundling of helices (determined locally by base-pairing) into linear stacks and subsequently organizing them through higher-order contacts into a complex tertiary fold (Chen et al. 1998; Massire et al. 1998). These models reveal how a ptRNA substrate could be docked via interactions with the S-domain while permitting conserved nucleotides in the C-domain (of both type A and type B RPRs) to converge near the cleavage site (Massire et al. 1998). Notably, central features in these models have now been validated by high-resolution structures.
To correlate form and function, models of the bacterial RPR bound to a tRNA have now been constructed independently for both the type A and type B RPRs using the respective crystal structures (Kazantsev et al. 2005; Torres-Larios et al. 2005). These models are consistent with a wealth of biochemical data on physical contacts in the ES complex (Christian et al. 2002). When the TΨC loop is docked in the opening of the S-domain, the acceptor stem is positioned in a groove present in the C-domain. In this docking scheme (Fig. 3), the cleavage site is proximal to various conserved nucleotides in P4, J5/15, and J2/4 in both type A and type B RPRs and provides a reasonable picture of the catalytic center (without the metal ions essential for catalysis). Moreover, in this model, the L15 loop (with some minor adjustments) can be made to base-pair with the CCA at the 3′ terminus. The robustness of this model is reflected in the excellent agreement between the expected distance from the TΨC loop to the cleavage site in the tRNA (41 Å) and the distance between the contact sites for these regions in the S- and C-domains of the bacterial RPR, respectively (46 Å; Torres-Larios et al. 2005). In both type A and type B RPRs, the precise orientation of the S- and C- domains is the basis of their exquisite cooperation in substrate binding and catalysis (Fig. 3). This is also underscored by the observation that the only common tertiary contact in both type A and type B RPRs is that between L8 and P4, which provides a vital bridge between the P8/P9 stack in the S-domain and P1/P4/P5 stack in the C-domain (Figs. 1 and 2) (Kazantsev et al. 2005; Torres-Larios et al. 2005).
Mondragon and coworkers have suggested grouping of the five bacterial RPR conserved regions (CR I to V) into (1) the S-domain module comprising CR II and III, and (2) the C-domain module comprising CR I, IV, and V (Torres-Larios et al. 2006). Despite the overwhelmingly helical content in the tertiary fold, it is interesting that both conserved modules are nonhelical: CR II and III form two interleaving T-loop motifs whereas CR I, IV, and V are part of loops and turns. Specific tertiary contacts in these conserved modules generate novel structural motifs, whose precise roles in RNase P catalysis remain to be determined.
The bacterial RPR structures highlight important parallels to protein catalysts: (1) the presence of pre-organized substrate-binding sites, and (2) primary and secondary structure versatility that can generate a conserved three-dimensional fold for performing a particular biological function. The latter feature is strikingly illustrated by the use of different structural scaffolds in type A and type B RPRs while maintaining a similar overall RNA structure (Fig. 2) (Kazantsev et al. 2005; Torres-Larios et al. 2005).
Structure of the Bacterial RNase P Protein
The tertiary structures of the bacterial RNase P protein (RPP) from B. subtilis, Staphylococcus aureus, and T. maritima have been determined by either X-ray crystallography or NMR spectroscopy (Stams et al. 1998; Spitzfaden et al. 2000; Kazantsev et al. 2003). The three structures are identical with an overall αβββαβα topology that includes the presence of an uncommon βαβ left-handed crossover connection from β3 to α2 to β4 (Fig. 4a). Bacterial RPP adopts an αβ fold comprising a central, four-stranded β-sheet sandwiched between two helices on one face and a third helix on the other face. This tertiary structure is reminiscent of ribosomal protein S5, domain IV of the translational elongation factor EF-G, and DNA gyrase (Stams et al. 1998). Like other RNA-binding proteins (such as U1A), bacterial RPP possesses a β-sheet whose large surface might be well suited for RNA binding. The bottom part of the solvent-exposed face of the β-sheet packs against the amino-terminal helix to generate a cleft lined with conserved aromatic residues that could engage in stacking interactions with nucleic acid (Fig. 4b). This premise has now been experimentally validated by cross-linking experiments demonstrating interactions between the central cleft and the ptRNA distal leader sequence (Niranjanakumari et al. 1998). Alignment of bacterial RPP sequences also revealed that α2 is rich in conserved amino acid residues and includes an RNR motif (AxxRNRxKRxxR, where x is any amino acid residue and the two R residues are conserved in all 112 bacterial RPP sequences examined; Jovanovic et al. 2002). Footprinting studies place this Arg-rich domain in α2 proximal to conserved nucleotides (in P4, J18/2, and J2/4) of the C-domain of bacterial RPR (Tsai et al. 2003).
Structure of the Bacterial RNase P Holoenzyme
Using protein-tethered Fe-EDTA–based hydroxyl radical footprinting, a list of possible physical contacts between the two subunits of the E. coli RNase P holoenzyme was obtained (Tsai et al. 2003). Fierke and coworkers established through photochemical cross-linking experiments that bacterial RPP uses its central cleft to interact with the nucleotides –4 to –7 in the ptRNA leader sequence (the cleavage site is between –1 and +1; Niranjanakumari et al. 1998). Taken together with earlier observations that the catalytic RNA moiety makes specific tertiary contacts with the T stem-loop and base-pairs to the CCA sequence at the 3′ end of the ptRNA (Kirsebom and Svard 1994; Pan et al. 1995), these footprinting and cross-linking data helped build three-dimensional models of the bacterial RNase P holoenzyme in the absence and presence of its ptRNA substrate (Fig. 5) (Tsai et al. 2003). Although atomic details are lacking, the model indicates how the holoenzyme could indeed simultaneously make multiple contacts with the ptRNA and how functional cooperation between the RPR and RPP (during ptRNA binding) might impose entropic constraints to facilitate substrate recognition/metal-ion binding and catalysis.
Structure of Archaeal and Eukaryal RPRs
Although no high-resolution structures are available for archaeal and eukaryal RPRs, identification of at least 50 sequences from each domain has permitted phlylogenetic sequence analysis that in turn has allowed refinement of secondary structure models (Harris et al. 2001; Li and Altman 2004a; Marquez et al. 2005). Despite the marked reduction in size of the archaeal and eukaryal RPRs (at least 10–20% smaller than typical bacterial RPRs), at least 13 nucleotides are universally conserved in identity and possibly in spatial location. In terms of differences, the archaeal and eukaryal RPRs are clearly missing some sequence/structure elements present in bacterial RPR that are either important for tertiary contacts or for direct interactions with the substrate (Fig. 1). For instance, the tertiary contacts in the smaller layer 2 of bacterial RPR (between P8 and L14/L18) are not possible in either archaeal or eukaryal RPR because they lack these elements and compensatory structural motifs are not evident. The L15 loop that base-pairs with the CCA sequence at the 3′ end of ptRNAs is also absent in all eukaryal and some archaeal RPRs.
Structure of Archaeal and Eukaryal RPPs
Four RPPs have been proven to associate with archaeal RNase P by virtue of co-purification (Hall and Brown 2002). Since a native RNase P holoenzyme has not been isolated, the caveat remains that there could be additional RPPs. Nevertheless, the four identified proteins are homologs of Rpp21, Rpp29, Rpp30, and Pop5, proteins associated with human and yeast RNase P. The tertiary structures of all four archaeal RPPs have been solved by X-ray crystallography and/or NMR spectroscopy, helped in some measure by the fact that they are both thermostable and smaller in size relative to their eukaryal homologs. The availability of structures of all four archaeal RPPs will aid structure/function relationship studies and elucidation of how these proteins aid archaeal RPR catalysis.
High-resolution structures of Pop5 from both Pyrococcus horikoshii (Pho) and P. furiosus (Pfu) have been determined using X-ray crystallography; Pho Pop5 was solved as a complex with Rpp30, its binary interaction partner (Kawano et al. 2006; Wilson et al. 2006). Pho and Pfu Pop5 display a compact overall αβ sandwich structure that is similar to the RNA-binding RRM fold (Fig. 4c). Moreover, there is striking similarity between Pop5 and bacterial RPP despite different secondary structure connectivities (βααββαβα in Pfu/Pho Pop5 versus αβββαβα in bacterial RPP). Pop5 exhibits a central, four-stranded antiparallel β-sheet surrounded by four α-helices. Like bacterial RPP, one face of the β-sheet packs against the amino-terminal helices to form the hydrophobic core; the other, rich in aromatic residues and partly solvent exposed, packs against the carboxy-terminal helix but does not generate an obvious cleft as found in bacterial RPP. It remains to be seen whether the structural homology between Pop5 and bacterial RPP translates to functional equivalence as postulated (Wilson et al. 2006).
The crystal structure of Pho Rpp21 confirmed the presence of a zinc ribbon with a single zinc ion coordinated by the sulfur atom in four highly conserved Cys residues (Fig. 4d) (Kakuta et al. 2005). The overall L-shaped structure results from two interacting α-helices (amino-terminal domain) along one arm and three antiparallel β-strands (carboxy-terminal domain) on the other; a linker connects the two arms. Two of the Cys residues needed for zinc binding are in the linker, and the remaining pair is proximal and present in the carboxy-terminal domain. The linker, through its participation in zinc binding, is important in bridging these two domains that make little intramolecular contacts; the planar nature of the zinc-binding fold also imposes structural constraints on the relative orientation of the amino- and carboxy-terminal domains. Not surprisingly, mutations of the conserved Cys residues result in loss of structure and activity (Kakuta et al. 2005).
The tertiary structure of Rpp29 from Archaeoglobus fulgidus (Afu), Methano-thermobacter thermoautotrophicus (Mth), and Pho has been solved by either NMR spectroscopy or X-ray crystallography (both for Afu Rpp29; Boomershine et al. 2003; Sidote and Hoffman 2003; Numata et al. 2004; Sidote et al. 2004). Whereas all four structures reveal the presence of a twisted, six-stranded antiparallel β-sheet built around a conserved hydrophobic core, the crystal structures additionally reveal the presence of amino- and carboxy-terminal helices, which were not seen in the NMR studies, likely due to these regions being disordered or flexible (Fig. 4e). The Rpp29 fold resembles Sm/Sm-like proteins that play a role in pre-mRNA splicing and snRNP biogenesis (Numata et al. 2004; Sidote et al. 2004).
Rpp30, the largest of the four archaeal RPPs (~25 kD), displays an α/β barrel structure similar to the TIM (triose phosphate isomerase) barrel family members with particularly strong resemblance to E. coli adenosine deaminase (Fig. 4f) (Takagi et al. 2004). TIM family members adopt an open barrel structure typically made up of eight α-helices and eight β-strands in repeating αβ units. Pho Rpp30 contains ten α-helices and seven β-strands and forms an oblate ellipsoid; the carboxy-terminal helix serves as a lid to the barrel. It is unclear how thematic variations from the typical TIM barrel are related to the function of Rpp30.
A recent study demonstrated that addition of the putative Pho L7Ae ribosomal protein to in vitro reconstituted Pho RNase P, comprising Pho RPR and the four RPPs, elevated the optimal temperature from 55°C to 70°C (Fukuhara et al. 2006). Concrete evidence that L7Ae is a bona fide subunit and part of a functional archaeal RNase P holoenzyme complex is lacking; however, if proven, it will make L7Ae unique among RNA-binding proteins in playing multiple roles since it is also part of RNPs which catalyze pseudouridylation and ribose 2′-O-methylation (Fukuhara et al. 2006).
The Nature of the Catalytic Event
The putative nature of the catalytic event during E. coli RNase P catalysis was outlined several years ago (Haydock and Allen 1985; Guerrier-Takada et al. 1986). The SN2 reaction mechanism is still thought to be appropriate for catalysis by RNase P, but the exact structural details have not been worked out fully. By analyzing solvent nucleophile isotope effects, Cassano et al. (2004) have inferred that a Mg2+-hydroxide complex is important in attacking the scissile phosphodiester linkage in the catalytic reaction.
The role of the –1 nucleotide and Mg2+ ions is critical in terms of a model of catalysis (Brannvall et al. 2002; Harris and Christian 2003; Persson et al. 2003; Zahler et al. 2005). It is apparent that at least one Mg2+ ion must be brought in by the substrate to the active site, a fact obvious from kinetic studies to date (Fig. 6). In fact, two or more divalent metal ions must be needed in this reaction (Warnecke et al. 1996). Their functions are supported by recent studies (Christian et al. 2006) on the utility of P4 of E. coli RPR in positioning a metal ion and a more general view of other metal-requiring enzymes (Yang et al. 2006).
A change in the identity of the base or the 2′OH at nucleotide –1, as well as alterations in the 3′-RCCA of the substrate in E. coli, affects the cleavage site and efficiency. Indeed, it was shown that A248 in M1 RNA interacts with the –1 nucleotide in ptRNAs, especially when the –1 nucleotide is a U (Zahler et al. 2003). Perturbations in the “standard” consensus recognition structures will lead to cleavage one nucleotide upstream of the canonical cleavage site. In addition, Kikovska et al. (2006) have shown that for ptRNAs which have a G at position +1, the exocyclic amine group contributes to both correct cleavage and rate of cleavage. This cannot be said for other nucleotides at this position. Kirsebom and Svard (1994) have also provided ample evidence that the P15 loop from M1 RNA is important for catalysis in E. coli. The function of this loop in organisms lacking the 3′-CCA in their ptRNAs, however, may not be the same.
An optimal fit between substrate and enzyme, including metal ions, is necessary for catalysis and cleavage at the correct site. In a parallel reaction that results in cleavage at the –1 position, not all steps of the optimal fit are achieved, and mis-cleavage occurs. The possible variations in the two processes allow the enzyme to make a fit with several precursor RNAs that have slightly different structural features and sequences at the cleavage site and can account for some of the apparent flexibility of the catalytic structure.
Evolution of Catalysis in the RNP Complex
The functional interplay between the RPR and RPPs is a common theme in RNase P from all organisms. Multiple roles for the single bacterial RPP have been proposed based on studies using E. coli and B. subtilis RNase P with different substrates: (1) stabilizing the RPR tertiary structure, (2) promoting direct interactions with the 5′ leader sequence to selectively increase the affinity for the ptRNA (substrate) over mature tRNA (product), (3) enhancing the rate of chemical cleavage by the cognate RPR, and (4) decreasing the requirement for Mg2+ due to an increased affinity for metal ions (Crary et al. 1998; Kurz et al. 1998; Niranjanakumari et al. 1998; Buck et al. 2005; Sun et al. 2006). A recent study demonstrated that E. coli RPP confers uniformity in both binding and catalysis by the RPR regardless of whether ptRNA substrates contain consensus recognition elements; the protein compensates for differences in ptRNA structure by altering its energetic contributions to leader binding (Sun et al. 2006). Although some of the above-mentioned roles appear to be realized in in vitro studies either with certain substrates or only with type A or type B RNase P holoenyzmes, hetero-reconstitution experiments both in vitro and in vivo reveal that a type A RPP can substitute for its type B counterpart, indicating the presence of conserved functional cores in these protein subunits that confer versatility to the RNA catalyst (Wegscheid et al. 2006).
Studies with archaeal RNase P are not as far along as with its bacterial counterpart. By assaying numerous archaeal RPRs in vitro, Pannucci et al. (1999) concluded that some archaeal RPRs are active at high ionic strength (300 mm Mg2+ and 4 m NH4+). Kimura and coworkers reported that recombinant Pho RPR, Pop5, Rpp21, Rpp29, and Rpp30 could be used to reconstitute (under single-turnover conditions) Pho RNase P in vitro, a first for any archaeal or eukaryal RNase P holoenzyme (Kouzuma et al. 2003). Subsequently, reconstitution of functional Mth and Pfu RNase P holoenzymes has been accomplished (Boomershine et al. 2003; Tsai et al. 2006). The reconstitution assay has permitted an evaluation both of the effect of single amino acid substitutions in archaeal RPPs and whether all RPPs are required for functional activity (Takagi et al. 2004; Kakuta et al. 2005; Tsai et al. 2006). Although no single Pfu RPP is able to activate Pfu RPR individually, two pairs of RPPs (one consisting of Rpp30 and Pop5, and the other made of Rpp21 and Rpp29) will result in partially active RNase P complexes that have a decreased Mg2+ requirement and possibly increased substrate affinity (Tsai et al. 2006). Even though the current estimate of only four archaeal RPPs might be revised to a higher number once a native enzyme is fully characterized, it is clear that the archaeal RPR is the catalytic entity and that protein cofactors play indispensable supporting roles.
For two decades, discussions of the properties of eukaryotic RNase P have always been prefaced by the statement that the RNA component of the enzyme has not been shown to be catalytic under conditions tested to date, or has not been catalytic at all. Recent experiments in the laboratory of L. Kirsebom have indicated that this statement may not be true. H1 RNA, the RNA subunit of the human RNase P, appears to be catalytically active, albeit with an efficiency 106-fold lower than that for E. coli RPR (E. Kikovska et al., in prep.). The same is true for the RPR from Giardia lamblia, a lower eukaryote. These recent findings suggest that many RPRs that had not been shown to be catalytic in fact do contain catalytic activity, but at very low levels compared to that found with the E. coli RPR. The reactions that tested these activities were carried out under conditions in which there is little or no breakdown of the substrate during prolonged incubation of the reaction mixture.
Taken together, these studies indicate that the RNA component in archaeal and eukaryal RNase P have coevolved to display a greater dependence on their cognate RPPs to participate effectively in the enzymatic mechanism (in contrast to the bacterial RPR). This factor must be an important general feature of the advent of the world of proteins on the world of RNA. Although it is unclear why different numbers of RPPs were recruited to the RPRs in the three domains of life, it appears that independent protein-facilitated solutions to enhance RNA catalysis in distinct lineages have converged on similar functional themes (Andrews et al. 2001; Kouzuma et al. 2003; Hsieh et al. 2004; Tsai et al. 2006).
Since the simple “one catalytic RNA and one protein cofactor” composition of bacterial RNase P has been replaced by an RNP complex of “one RNA and four or more protein subunits” in archaeal and eukaryal RNase P, a commonly held premise was that the catalytic role of the bacterial RPR had been replaced by some of the archaeal or eukaryal RPPs. The recent findings that some archaeal and eukaryal RPRs are active, albeit only weakly, in the absence of their protein cofactors are inconsistent with this notion. Rather than the RPRs being relegated to a secondary role by their protein cofactors, which have the potential to be more efficient and versatile enzymes, it is likely that the catalytic capability rests with the RPR through evolution of RNase P. This then raises the question why the extant cell, which relies almost solely on proteins for catalytic and structural roles, uses RNA-mediated catalysis in RNase P and ribosomes and has even retained these evolutionary vestiges from a hypothetical RNA World. Is it possible that in another billion years, no RNA will be left in these enzymes with the apparent catalytic activity they now have? Or have these RNA enzymes (as part of RNP complexes) evolved to possess properties that render them uniquely well suited for catalyzing a single chemical reaction on multiple substrates with high efficiency and extraordinary fidelity?
New Substrates of RNase P
The T stem and loop of conventional ptRNAs interacts by H-bonding of certain riboses and nucleotides with the S-domain (P9 and P11) in bacterial RPR (Christian et al. 2002; Harris and Christian 2003). However, it has also been shown that this substrate-binding segment, which is missing in certain deletion mutations of E. coli RPR, does bind other substrates like p4.5S RNA, but it does so much more strongly when the protein cofactor, C5 protein, is present. Fierke and colleagues have shown that the protein is important in binding the precursor segment of ptRNAs (Crary et al. 1998; Kurz et al. 1998; Loria et al. 1998; Niranjanakumari et al. 1998). There are still some missing pieces of this puzzle. First, how do non-ptRNA substrates like p4.5S RNA bind to the enzymatic RNA (and protein), and second, how do precursors with relatively short precursor segments bind to the enzymatic RNA (and protein), and what are the kinetic characteristics that determine each case?
The use of temperature-sensitive mutants in RNase P function in E. coli was important in the early days of the study of this enzyme. The use of such a mutant in the gene for E. coli RPP, combined with microarray data, yielded new and unexpected substrates for RNase P in E. coli. Several operons were found that had one or a few cleavage sites in the spaces between certain ORFs. One of these is the lac operon (Li and Altman 2003, 2004b). In the latter case, this result explained polarity between the lacZ and the lacA gene products. It had been previously mentioned that hairpin-like substrates, which should appear very frequently in many RNAs in any cell type, would be hidden by folding of large mRNAs. The sites available in operons are clear exceptions in which these sites are accessible. Similarly, some riboswitches (vitamin B12 metabolism, adenine and guanine biosynthesis) are also cleaved by RNase P, but the net biological effect of these cleavages is not completely known.
Another RNA sequence, OLE, which is involved in isoprenoid biosynthesis, is also cleaved by RNase P (J. Ko and S. Altman, unpubl.), but it is not certain whether OLE is part of an operon, an actual riboswitch, or a noncoding RNA (E. Puerta-Fernandez, J.E. Barrick, A. Roth, and R.R. Breaker, unpubl.). This sequence is found in extremophiles (e.g., B. halodurans) and may have some function in cell wall or membrane protein synthesis.
Stolc and colleagues have performed in S. cerevisiae a microarray experiment similar to the one cited above in E. coli (Samanta et al. 2006). In that case, the amount of RNase P was decreased using a regulatable RRP. Microarray data were gathered in which it was shown that certain regions are transcribed more so than in control experiments in which RNase P levels are wild type. Stolc has advertised some of these regions as possible RNase P substrates. One of these novel RNAs can be cleaved by yeast RNase P, as expected (L. Yang and S. Altman, unpubl.).
Despite impressive advances in elucidation of the tertiary structure of bacterial RPRs, details of the catalytic site are still lacking. Since there is evidence of least two structures of the enzyme, with and without an appropriate substrate (Pomeranz Krummel and Altman 1999a), it will also be instructive to examine the conformational changes in the ES complex. A crystal structure of the RPR (or holoenzyme)–substrate complex has yet to be achieved. A. Mondragon (pers. comm.) has stated that part of the problem, aside from inadequate resolution in the current structures, may be that the active site is “flexible” and adapts slightly differently to different substrates. This explanation fits well with the properties of substrates that have been studied. A model noncleavable substrate, with a 5′ single-stranded leader sequence and a double-stranded region (mimicking the acceptor stem) with an interstrand disulfide to prevent helix denaturation and RNase P cleavage, merits consideration in studies aimed at solving the ES complex (Pomeranz-Krummel et al. 2000).
Some aspects of the substrate recognition problem remain to be worked out. Since non-ptRNA substrates (e.g., p4.5S RNA), which lack the T-stem loop, are cleaved by bacterial RPR, both in the absence and presence of its protein cofactor, it is unclear whether (and how) the S-domain is involved in recognizing these substrates. In addition, because there are many examples of the bacterial RPP broadening the substrate specificity of bacterial RPR, it also remains to be seen how the protein cofactor or its induced structural alterations in the RPR facilitate recognition and cleavage of a wider array of substrates (Liu and Altman 1994; Sun et al. 2006). Of course, comparing the RNA alone and holoenzyme structures will be a necessary first step in this regard.
RPRs play a central role in catalysis regardless of the variations on structure of RPRs and RPPs. Are RNA–protein interactions in archaeal and eukaryal RNase P able to replace some structural aspects of the bacterial RPR? Is there a near-identical placement of universally conserved nucleotides in the active site despite considerable structural plasticity in the different RNase P holoenzymes? High-resolution structures are needed to gain reliable insights into these problems.
S.A. thanks Yale University for support, and V.G. acknowledges the support of the National Science Foundation (grant MCB-0238233) and the National Institutes of Health (grant GM067947). We are grateful to Drs. Mark Foster and Lien Lai (OSU) for suggestions and assistance with preparation of figures. We also thank our colleagues for providing unpublished material, and we regret that, due to space constraints, we could not acknowledge the work of several researchers in this report.
Andrews A.J., Hall T.A., and Brown J.W. 2001. Characterization of RNase P holoenzymes from Methanococcus jannaschii and Methanothermobacter thermo-autotrophicus. Biol. Chem. 382: 1171–1177.
Boomershine W.P., McElroy C.A., Tsai H.Y., Wilson R.C., Gopalan V., and Foster M.P. 2003. Structure of Mth11/Mth Rpp29, an essential protein subunit of archaeal and eukaryotic RNase P. Proc. Natl. Acad. Sci. 100: 15398–15403.
Brannvall M., Petterson B.M.F., and Kirsebom L.A. 2002. The residue immediately upstream of the RNase P cleavage site is a positive determinant. Biochimie 84: 693–703.
Brown J.W. 1999. The ribonuclease P database. Nucleic Acids Res. 27: 314.
Buck A.H., Dalby A.B., Poole A.W., Kazantsev A.V., and Pace N.R. 2005. Protein activation of a ribozyme: The role of bacterial RNase P protein. EMBO J. 24: 3360–3368.
Cassano A.G., Anderson V.E., and Harris M.E. 2004. Analysis of solvent nucleophile isotope effects: Evidence for concerted mechanisms and nucleophilic activation by metal coordination in non-enzymatic and ribozyme-catalyzed phosphodiester hydrolysis. Biochemistry 43: 10547–10559.
Chen J.L., Nolan J.M., Harris M.E., and Pace N.R. 1998. Comparative photocross-linking analysis of the tertiary structures of Escherichia coli and Bacillus subtilis RNase P RNAs. EMBO J. 17: 1515–1525.
Christian E.L., Smith K.M.J., Perera N., and Harris M.E. 2006. The P4 metal binding site in RNase P RNA affects active site metal affinity through substrate positioning. RNA 12: 1463–1467.
Christian E.L., Zahler N., Kaye N.M., and Harris M.E. 2002. Analysis of substrate recognition by the ribonucleoprotein endonuclease RNase P. Methods 28: 307–322.
Crary S.M., Niranjanakumari S., and Fierke C.A. 1998. The protein component of Bacillus subtilis ribonuclease P increases catalytic efficiency by enhancing interactions with the 5′ leader sequence of pre-tRNAAsp. Biochemistry 37: 9409–9416.
Evans D., Marquez S.M., and Pace N.R. 2006. RNase P: Interface of the RNA and protein worlds. Trends Biochem. Sci. 31: 333–341.
Fukuhara H., Kifusa M., Watanabe M., Terada A., Honda T., Numata T., Kakuta Y., and Kimura M. 2006. A fifth protein subunit Ph1496p elevates the optimum temperature for the ribonuclease P activity from Pyrococcus horikoshii OT3. Biochem. Biophys. Res. Commun. 343: 956–964.
Guerrier-Takada C. and Altman S. 1992. Reconstitution of enzymatic activity from fragments of M1 RNA. Proc. Natl. Acad. Sci. 89: 1266–1270.
Guerrier-Takada C., Haydock K., Allen L., and Altman S. 1986. Metal ion requirements and other aspects of the reaction catalyzed by M1 RNA, the RNA subunit of ribonuclease P from Escherichia coli. Biochemistry 25: 1509–1515.
Guerrier-Takada C., Gardiner K., Marsh T., Pace N., and Altman S. 1983. The RNA moiety of ribonuclease P is the catalytic subunit of the enzyme. Cell 35: 849–857.
Hall T.A. and Brown J.W. 2002. Archaeal RNase P has multiple protein subunits homologous to eukaryotic nuclear RNase P proteins. RNA 8: 296–306.
Harris J.K., Haas E.S., Williams D., Frank D.N., and Brown J.W. 2001. New insight into RNase P RNA structure from comparative analysis of the archaeal RNA. RNA 7: 220–232.
Harris M.E. and Christian E.L. 2003. Recent insights into the structure and function of the ribonucleoprotein enzyme ribonuclease P. Curr. Opin. Struct. Biol. 13: 325–333.
Haydock K. and Allen L.C. 1985. Molecular mechanism of catalysis by RNA. Prog. Clin. Biol. Res. 172A: 87–98.
Hsieh J., Andrews A., and Fierke C.A. 2004. Roles of protein subunits in RNA-protein complexes: Lessons from ribonuclease P. Biopolymers 73: 79–89.
James B.D., Olsen G.J., Liu J.S., and Pace N.R. 1988. The secondary structure of ribonuclease P RNA, the catalytic element of a ribonucleoprotein enzyme. Cell 52: 19–26.
Jarrous N. 2002. Human ribonuclease P: Subunits, function, and intra-nuclear localization. RNA 8: 1–7.
Jovanovic M., Sanchez R., Altman S., and Gopalan V. 2002. Elucidation of structure-function relationships in the protein subunit of bacterial RNase P using a genetic complementation approach. Nucleic Acids Res. 30: 5065–5063.
Kakuta Y., Ishimatsu I., Numata T., Kimura K., Yao M., Tanaka I., and Kimura M. 2005. Crystal structure of a ribonuclease P protein Ph1601p from Pyrococcus horikoshii OT3: An archaeal homologue of human nuclear ribonuclease P protein Rpp21. Biochemistry 44: 12086–12093.
Kawano S., Nakashima T., Kakuta Y., Tanaka I., and Kimura M. 2006. Crystal structure of protein Ph1481p in complex with protein Ph1877p of archaeal RNase P from Pyrococcus horikoshii OT3: Implication of dimer formation of the holoenzyme. J. Mol. Biol. 357: 583–591.
Kazantsev A.V., Krivenko A.A., Harrington D.J., Holbrook S.R., Adams P.D., and Pace N.R. 2005. Crystal structure of a bacterial ribonuclease P RNA. Proc. Natl. Acad. Sci. 102: 13392–13397.
Kazantsev A.V., Krivenko A.A., Harrington D.J., Carter R.J., Holbrook S.R., Adams P.D., and Pace N.R. 2003. High-resolution structure of RNase P protein from Thermotoga maritima. Proc. Natl. Acad. Sci. 100: 7497–7502.
Kikovska E., Branvall M., and Kirsebom, L.A. 2006. The exocyclic amine at the RNase P cleavage site contributes to substrate binding and catalysis. J. Mol. Biol. 359: 572–584.
Kirsebom L.A. and Svard S.G. 1994. Base pairing between Escherichia coli RNase P RNA and its substrate. EMBO J. 13: 4870–4876.
Kouzuma Y., Mizoguchi M., Takagi H., Fukuhara H., Tsukamoto M., Numata T., and Kimura M. 2003. Reconstitution of archaeal ribonuclease P from RNA and four protein components. Biochem. Biophys. Res. Commun. 306: 666–673.
Krasilnikov A.S., Xiao Y., Pan T., and Mondragon A. 2004. Basis for structural diversity in homologous RNAs. Science 306: 104–107.
Krasilnikov A.S., Yang X., Pan T., and Mondragon A. 2003. Crystal structure of the specificity domain of ribonuclease P. Nature 421: 760–764.
Kurz J.C. and Fierke C.A. 2002. The affinity of magnesium binding sites in the Bacillus subtilis RNase P x pre-tRNA complex is enhanced by the protein subunit. Biochemistry 41: 9545–9558.
Kurz J.C., Niranjanakumari S., and Fierke C.A. 1998. Protein component of Bacillus subtilis RNase P specifically enhances the affinity for precursor-tRNAAsp. Biochemistry 37: 2393–2400.
Li Y. and Altman S. 2003. A specific endoribonuclease, RNase P, affects gene expresson of polycistronic operon mRNAs. Proc. Natl. Acad. Sci. 100: 13213–13218.
———. 2004a. In search of RNase P RNA from microbial genomes. RNA 10: 1533–1540.
———. 2004b. Polarity effects in the lactose operon of Escherichia coli. J. Mol. Biol. 339: 31–39.
Liu F. and Altman S. 1994. Differential evolution of substrates for an RNA enzyme in the presence and absence of its protein cofactor. Cell 77: 1093–1100.
Loria A. and Pan T. 1996. Domain structure of the ribozyme from eubacterial ribonuclease P. RNA 2: 551–563.
Loria A., Niranjanakumari S., Fierke C.A., and Pan T. 1998. Recognition of a pre-RNA substrate by the Bacillus subtilis RNase P holoenzyme. Biochemistry 37: 15466–15473.
Marquez S.M., Harris J.K., Kelley S.T., Brown J.W., Dawson S.C., Roberts E.C., and Pace N.R. 2005. Structural implications of novel diversity in eucaryal RNase P RNA. RNA 11: 739–751.
Massire C., Jaeger L., and Westhof E. 1998. Derivation of the three-dimensional architecture of bacterial ribonuclease P RNAs from comparative sequence analysis. J. Mol. Biol. 279: 773–793.
Niranjanakumari S., Stams T., Crary S.M., Christianson D.W., and Fierke C.A. 1998. Protein component of the ribozyme ribonuclease P alters substrate recognition by directly contacting precursor tRNA. Proc. Natl. Acad. Sci. 95: 15212–15217.
Numata T., Ishimatsu I., Kakuta Y., Tanaka I., and Kimura M. 2004. Crystal structure of archaeal ribonuclease P protein Ph1771p from Pyrococcus horikoshii OT3: An archaeal homolog of eukaryotic ribonuclease P protein Rpp29. RNA 10: 1423–1432.
Pan T. 1995. Higher order folding and domain analysis of the ribozyme from Bacillus subtilis ribonuclease P. Biochemistry 34: 902–909.
Pan T., Loria A., and Zhong K. 1995. Probing of tertiary interactions in RNA: 2′-hydroxyl-base contacts between the RNase P RNA and pre-tRNA. Proc. Natl. Acad. Sci. 92: 12510–12514.
Pannucci J.A., Haas E.S., Hall T.A., Harris J.K., and Brown J.W. 1999. RNase P RNAs from some archaea are catalytically active. Proc. Natl. Acad. Sci. 96: 7803–7808.
Persson T., Cuzic S., and Hartmann R.K. 2003. Catalysis by RNase P RNA: Unique features and unprecedented active site plasticity. J. Biol. Chem. 278: 43394–43401.
Pomeranz Krummel D.A. and Altman S. 1999a. Multiple binding modes of substrate to the catalytic RNA subunit of RNase P from Escherichia coli. RNA 8: 1021–1033.
———. 1999b. Verification of phylogenetic predictions in vivo and the importance of the tetraloop motif in a catalytic RNA. Proc. Natl. Acad. Sci. 96: 11200–11205.
Pomeranz Krummel D.A., Kent O., MacMillan A.M., and Altman S. 2000. Evidence for helical unwinding of an RNA substrate by the RNA enzyme RNase P: Use of an interstrand disulfide crosslink in substrate. J. Mol. Biol. 295: 1113–1118.
Samanta M.P., Tongprasit W., Sethi H., Chin C.S., and Stolc V. 2006. Global identification of noncoding RNAs in Saccharomyces cerevisiae by modulating an essential RNA processing pathway. Proc. Natl. Acad. Sci. 103: 4192–4197.
Sidote D.J. and Hoffman D.W. 2003. NMR structure of an archaeal homologue of ribonuclease P protein Rpp29. Biochemistry 42: 13541–13550.
Sidote D.J., Heideker J., and Hoffman D.W. 2004. Crystal structure of archaeal ribonuclease P protein aRpp29 from Archaeoglobus fulgidus. Biochemistry 43: 14128–14138.
Spitzfaden C., Nicholson N., Jones J.J., Guth S., Lehr R., Prescott C.D., Hegg L.A., and Eggleston D.S. 2000. The structure of ribonuclease P protein from Staphylococcus aureus reveals a unique binding site for single-stranded RNA. J. Mol. Biol. 295: 105–115.
Stams T., Niranjanakumari S., Fierke C.A., and Christianson D.W. 1998. Ribonuclease P protein structure: Evolutionary origins in the translational apparatus. Science 280: 752–755.
Steitz T.A. and Moore P.B. 2003. RNA, the first macromolecular catalyst: The ribosome is a ribozyme. Trends Biochem. Sci. 8: 411–418.
Sun L., Campbell F.E., Zahler N.H., and Harris M.E. 2006. Evidence that substrate specific effects of C5 protein lead to uniformity in binding and catalysis by RNase P. EMBO J. 25: 3998-4007.
Takagi H., Watanabe M., Kakuta Y., Kamachi R., Numata T., Tanaka I., and Kimura M. 2004. Crystal structure of the ribonuclease P protein Ph1877p from hyperthermophilic archaeon Pyrococcus horikoshii OT3. Biochem. Biophys. Res. Commun. 319: 787–794.
Torres-Larios A., Swinger K.K., Pan T., and Mondragon A. 2006. Structure of RNase P, a universal ribozyme. Curr. Opin. Struct. Biol. 16: 327–335.
Torres-Larios A., Swinger K.K., Kraslinikov A.S., Pan T., and Mondragon A. 2005. Crystal structure of the RNA component of bacterial ribonuclease P. Nature 437: 584–587.
Tsai H.Y., Pulukkunat D. K., Woznick W., and Gopalan V. 2006. Functional reconstitution and characterization of Pyrococcus furiosus RNase P. Proc. Natl. Acad. Sci. 103: 16147–16152.
Tsai H.Y., Masquida B., Biswas R., Westhof E., and Gopalan V. 2003. Molecular modeling of the three-dimensional structure of the bacterial RNase P holoenzyme. J. Mol. Biol. 325: 661–675.
Valadkhan S. and Manley J.L. 2001. Splicing-related catalysis by protein-free snRNAs. Nature 413: 701–707.
Walker S.C. and Engleke D.R. 2006. Ribonuclease P: The evolution of an ancient RNA enzyme. Crit. Rev. Biochem. Mol. Biol. 41: 77–102.
Warnecke J.M., Furste J.P., Hardt W.D., Erdmann V.A., and Hartmann R.K. 1996. Ribonuclease P (RNase P) RNA is converted to a Cd (2+)-ribozyme by a single Rp-phosphorothioate modification in the precursor tRNA at the RNase P cleavage site. Proc. Natl. Acad. Sci. 93: 8924–8928.
Wegscheid B., Condon C., and Hartmann R.K. 2006. Type A and type B RNase P RNAs are interchangeable in vivo despite substantial biophysical differences. EMBO Rep. 7: 411–417.
Wilson R.C., Bohlen C.J., Foster M.P., and Bell C.E. 2006. Structure of Pfu Pop5, an archaeal RNase P protein. Proc. Natl. Acad. Sci. 103: 873–878.
Yang W., Lee J. Y., and Nowotny M. 2006. Making and breaking nucleic acids: Two-Mg2+-ion catalysis and substrate specificity. Mol. Cell 22: 5–13.
Zahler N.H., Christian E.L., and Harris M.E. 2003. Recognition of the 5′ leader of pre-tRNA substrates by the active site of ribonuclease P. RNA. 9: 734–745.
Zahler N.H., Sun L., Christian E.L., and Harris M.E. 2005. The pre-tRNA nucleotide base and 2′-hydroxyl at N(–1) contribute to fidelity in tRNA processing by RNase P. J. Mol. Biol. 345: 969–985.