Crystal structure of the RluD pseudouridine synthase catalytic module, an enzyme that modifies 23S rRNA and is essential for normal cell growth of Escherichia coli

Pseudouridine (5-beta-D-ribofuranosyluracil, Psi) is the most commonly found modified base in RNA. Conversion of uridine to Psi is performed enzymatically in both prokaryotes and eukaryotes by pseudouridine synthases (EC 4.2.1.70). The Escherichia coli Psi-synthase RluD modifies uridine to Psi at positions 1911, 1915 and 1917 within 23S rRNA. RluD also possesses a second function related to proper assembly of the 50S ribosomal subunit that is independent of Psi-synthesis. Here, we report the crystal structure of the catalytic module of RluD (residues 68-326; DeltaRluD) refined at 1.8A to a final R-factor of 21.8% (R(free)=24.3%). DeltaRluD is a monomeric enzyme having an overall mixed alpha/beta fold. The DeltaRluD molecule consists of two subdomains, a catalytic subdomain and C-terminal subdomain with the RNA-binding cleft formed by loops extending from the catalytic sub-domain. The catalytic sub-domain of DeltaRluD has a similar fold as in TruA, TruB and RsuA, with the location of the RNA-binding cleft, active-site and conserved, catalytic Asp residue superposing in all four structures. Superposition of the crystal structure of TruB bound to a T-stem loop with RluD reveals that similar RNA-protein interactions for the flipped-out uridine base would exist in both structures, implying that base-flipping is necessary for catalysis. This observation also implies that the specificity determinants for site-specific RNA-binding and recognition likely reside in parts of RluD beyond the active site.     

An arginine-aspartate network in the active site of bacterial TruB is critical for catalyzing pseudouridine formation

Pseudouridine synthases introduce the most common RNA modification and likely use the same catalytic mechanism. Besides a catalytic aspartate residue, the contributions of other residues for catalysis of pseudouridine formation are poorly understood. Here, we have tested the role of a conserved basic residue in the active site for catalysis using the bacterial pseudouridine synthase TruB targeting U55 in tRNAs. Substitution of arginine 181 with lysine results in a 2500-fold reduction of TruB’s catalytic rate without affecting tRNA binding. Furthermore, we analyzed the function of a second-shell aspartate residue (D90) that is conserved in all TruB enzymes and interacts with C56 of tRNA. Site-directed mutagenesis, biochemical and kinetic studies reveal that this residue is not critical for substrate binding but influences catalysis significantly as replacement of D90 with glutamate or asparagine reduces the catalytic rate 30- and 50-fold, respectively. In agreement with molecular dynamics simulations of TruB wild type and TruB D90N, we propose an electrostatic network composed of the catalytic aspartate (D48), R181 and D90 that is important for catalysis by fine-tuning the D48-R181 interaction. Conserved, negatively charged residues similar to D90 are found in a number of pseudouridine synthases, suggesting that this might be a general mechanism.     

Domain organization and crystal structure of the catalytic domain of E.coli RluF, a pseudouridine synthase that acts on 23S rRNA

Pseudouridine synthases catalyze the isomerization of uridine to pseudouridine (Psi) in rRNA and tRNA. The pseudouridine synthase RluF from Escherichia coli (E.C. 4.2.1.70) modifies U2604 in 23S rRNA, and belongs to a large family of pseudouridine synthases present in all kingdoms of life. Here we report the domain architecture and crystal structure of the catalytic domain of E.coli RluF at 2.6A resolution. Limited proteolysis, mass spectrometry and N-terminal sequencing indicate that RluF has a distinct domain architecture, with the catalytic domain flanked at the N and C termini by additional domains connected to it by flexible linkers. The structure of the catalytic domain of RluF is similar to those of RsuA and TruB. RluF is a member of the RsuA sequence family of Psi-synthases, along with RluB and RluE. Structural comparison of RluF with its closest structural homologues, RsuA and TruB, suggests possible functional roles for the N-terminal and C-terminal domains of RluF.     

The roles of the essential Asp-48 and highly conserved His-43 elucidated by the pH dependence of the pseudouridine synthase TruB

All known pseudouridine synthases have a conserved aspartic acid residue that is essential for catalysis, Asp-48 in Escherichia coli TruB. To probe the role of this residue, inactive D48C TruB was oxidized to generate the sulfinic acid cognate of aspartic acid. The oxidation restored significant but reduced catalytic activity, consistent with the proposed roles of Asp-48 as a nucleophile and general base. The family of pseudouridine synthases including TruB also has a nearly invariant histidine residue, His-43 in the E. coli enzyme. To examine the role of this conserved residue, site-directed mutagenesis was used to generate H43Q, H43N, H43A, H43G, and H43F TruB. Except for phenylalanine, the substitutions seriously impaired the enzyme, but all of the altered TruB retained significant activity. To examine the roles of Asp-48 and His-43 more fully, the pH dependences of wild-type, oxidized D48C, and H43A TruB were determined. The wild-type enzyme displays a typical bell-shaped profile. With oxidized D48C TruB, logk(cat) varies linearly with pH, suggesting the participation of specific rather than general base catalysis. Substitution of His-43 perturbs the pH profile, but it remains bell-shaped. The ascending limb of the pH profile is assigned to Asp-48, and the descending limb is tentatively ascribed to an active site tyrosine residue, the bound substrate uridine, or the bound product pseudouridine.     

Role of cysteine residues in pseudouridine synthases of different families.

The pseudouridine synthases catalyze the isomerization of uridine to pseudouridine in RNA molecules. An attractive mechanism was proposed based on that of thymidylate synthase, in which the thiol(ate) group of a cysteine side chain serves as the nucleophile in a Michael addition to C6 of the isomerized uridine. Such a role for cysteine in the pseudouridine synthase TruA (also named Psi synthase I) has been discredited by site-directed mutagenesis, but sequence alignments have led to the conclusion that there are four distinct "families" of pseudouridine synthases that share no statistically significant global sequence similarity. It was, therefore, necessary to probe the role of cysteine residues in pseudouridine synthases of the families that do not include TruA. We examined the enzymes RluA and TruB, which are members of different families than TruA and each other. Substitution of cysteine for amino acids with nonnucleophilic side chains did not significantly alter the catalytic activity of either pseudouridine synthase. We conclude, therefore, that neither TruB nor RluA require thiol(ate) groups to effect catalysis, excluding their participation in a Michael addition to C6 of uridine, although not eliminating that mechanism (with an alternate nucleophile) from future consideration.     

Dissecting the roles of a strictly conserved tyrosine in substrate recognition and catalysis by pseudouridine 55 synthase

Sequence alignment of the TruA, TruB, RsuA, and RluA families of pseudouridine synthases (PsiS) identifies a strictly conserved aspartic acid, which has been shown to be the critical nucleophile for the PsiS-catalyzed formation of pseudouridine (Psi). However, superposition of the representative structures from these four families of enzymes identifies two additional amino acids, a lysine or an arginine (K/R) and a tyrosine (Y), from a K/RxY motif that are structurally conserved in the active site. We have created a series of Thermotoga maritima and Escherichia coli pseudouridine 55 synthase (Psi55S) mutants in which the conserved Y is mutated to other amino acids. A new crystal structure of the T. maritima Psi55S Y67F mutant in complex with a 5FU-RNA at 2.4 A resolution revealed formation of 5-fluoro-6-hydroxypseudouridine (5FhPsi), the same product previously seen in wild-type Psi55S-5FU-RNA complex structures. HPLC analysis confirmed efficient formation of 5FhPsi by both Psi55S Y67F and Y67L mutants but to a much lesser extent by the Y67A mutant when 5FU-RNA substrate was used. However, both HPLC analysis and a tritium release assay indicated that these mutants had no detectable enzymatic activity when the natural RNA substrate was used. The combined structural and mutational studies lead us to propose that the side chain of the conserved tyrosine in these four families of PsiS plays a dual role within the active site, maintaining the structural integrity of the active site through its hydrophobic phenyl ring and acting as a general base through its OH group for the proton abstraction required in the last step of PsiS-catalyzed formation of Psi.     

Crystal structure of the catalytic domain of RluD, the only rRNA pseudouridine synthase required for normal growth of Escherichia coli

Escherichia coli pseudouridine synthase RluD makes pseudouridines 1911, 1915, and 1917 in the loop of helix 69 in 23S RNA. These are the most highly conserved ribosomal pseudouridines known. Of 11 pseudouridine synthases in E. coli, only cells lacking RluD have severe growth defects and abnormal ribosomes. We have determined the 2.0 A structure of the catalytic domain of RluD (residues 77-326), the first structure of an RluA family member. The catalytic domain folds into a mainly antiparallel beta-sheet flanked by several loops and helices. A positively charged cleft that presumably binds RNA leads to the conserved Asp 139. The RluD N-terminal S4 domain, connected by a flexible linker, is disordered in our structure. RluD is very similar in both catalytic domain structure and active site arrangement to the pseudouridine synthases RsuA, TruB, and TruA. We identify five sequence motifs, two of which are novel, in the RluA, RsuA, TruB, and TruA families, uniting them as one superfamily. These results strongly suggest that four of the five families of pseudouridine synthases arose by divergent evolution. The RluD structure also provides insight into its multisite specificity.     

Differential roles of archaeal box H/ACA proteins in guide RNA-dependent and independent pseudouridine formation.

RNA-guided pseudouridine (Psi) synthesis in Archaea and Eukarya requires a four-protein one-RNA containing box H/ACA ribonucleoprotein (RNP) complex. The proteins in the archaeal RNP are aCbf5, aNop10, aGar1 and L7Ae. Pyrococcus aCbf5-aNop10 is suggested to be the minimal catalytic core in this synthesis and the activity is enhanced by L7Ae and aGar1. The protein aCbf5 is homologous to eukaryal Cbf5 (dyskerin, NAP57) as well as to bacterial TruB and eukaryal Pus4; the last two produce YPsi55 in tRNAs in a guide RNA-independent manner. Here, using recombinant Methanocaldococcus jannaschii proteins, we report that aCbf5 and aGar1 together can function as a tRNA Psi55 synthase in a guide RNA-independent manner. This activity is enhanced by aNop10, but not by L7Ae. The aCbf5 alone can also produce Psi55 in tRNAs that contain the canonical 3'-CCA sequence and this activity is stimulated by aGar1. These results suggest that the roles of accessory proteins are different in guide RNA-dependent and independent Psi synthesis by aCbf5. The presence of conserved C (or U) and A at tRNA positions 56 and 58, respectively, which are required for TruB/Pus4 activity, is not essential for aCbf5-mediated Psi55 formation. Conserved A58 in tRNA normally forms a tertiary reverse Hoogstein base pair with an equally conserved U54. This base pair is recognized by TruB. Apparently aCbf5 does not require this base pair to recognize U55 for conversion to Psi55.     

The handling of the mechanistic probe 5-fluorouridine by the pseudouridine synthase TruA and its consistency with the handling of the same probe by the pseudouridine synthases TruB and RluA

RNA containing 5-fluorouridine (F(5)U) had previously been used to examine the mechanism of the pseudouridine synthase TruA, formerly known as pseudouridine synthase I [Gu et al. (1999) Proc. Natl. Acad. Sci. U.S.A. 96, 14270-14275]. From that work, it was reasonably concluded that the pseudouridine synthases proceed via a mechanism involving a Michael addition by an active site aspartic acid residue to the pyrimidine ring of uridine or F(5)U. Those conclusions rested on the assumption that the hydrate of F(5)U was obtained after digestion of the product RNA and that hydration resulted from hydrolysis of the ester intermediate between the aspartic acid residue and F(5)U. As reported here, (18)O labeling definitively demonstrates that ester hydrolysis does not give rise to the observed hydrated product and that digestion generates not the expected mononucleoside product but rather a dinucleotide between a hydrated isomer of F(5)U and the following nucleoside in RNA. The discovery that digestion products are dinucleotides accounts for the previously puzzling differences in the isolated products obtained following the action of the pseudouridine synthases TruB and RluA on F(5)U in RNA.     

Crystal structure of the highly divergent pseudouridine synthase TruD reveals a circular permutation of a conserved fold

The pseudouridine (Psi) synthases Pus7p and TruD define a family of RNA-modifying enzymes with no sequence similarity to previously characterized Psi synthases. The 2.2 A resolution structure of Escherichia coli TruD reveals a U-shaped molecule with a catalytic domain that superimposes closely on that of other Psi synthases. A domain that appears to be unique to TruD/Pus7p family enzymes hinges over the catalytic domain, possibly serving to clasp the substrate RNAs. The active site comprises residues that are conserved in other Psi synthases, although at least one comes from a structurally distinct part of the protein. Remarkably, the connectivity of the structural elements of the TruD catalytic domain is a circular permutation of that of its paralogs. Because the sequence of the permuted segment, a beta-strand that bisects the catalytic domain, is conserved among orthologs from bacteria, archaea and eukarya, the permutation likely happened early in evolution.     

Pre-steady-state kinetic analysis of the three Escherichia coli pseudouridine synthases TruB, TruA, and RluA reveals uniformly slow catalysis

Pseudouridine synthases catalyze formation of the most abundant modification of functional RNAs by site-specifically isomerizing uridines to pseudouridines. While the structure and substrate specificity of these enzymes have been studied in detail, the kinetic and the catalytic mechanism of pseudouridine synthases remain unknown. Here, the first pre-steady-state kinetic analysis of three Escherichia coli pseudouridine synthases is presented. A novel stopped-flow absorbance assay revealed that substrate tRNA binding by TruB takes place in two steps with an overall rate of 6 sec(-1). In order to observe catalysis of pseudouridine formation directly, the traditional tritium release assay was adapted for the quench-flow technique, allowing, for the first time, observation of a single round of pseudouridine formation. Thereby, the single-round rate constant of pseudouridylation (k(Ψ)) by TruB was determined to be 0.5 sec(-1). This rate constant is similar to the k(cat) obtained under multiple-turnover conditions in steady-state experiments, indicating that catalysis is the rate-limiting step for TruB. In order to investigate if pseudouridine synthases are characterized by slow catalysis in general, the rapid kinetic quench-flow analysis was also performed with two other E. coli enzymes, RluA and TruA, which displayed rate constants of pseudouridine formation of 0.7 and 0.35 sec(-1), respectively. Hence, uniformly slow catalysis might be a general feature of pseudouridine synthases that share a conserved catalytic domain and supposedly use the same catalytic mechanism.     

Formation of the conserved pseudouridine at position 55 in archaeal tRNA

Pseudouridine (Ψ) located at position 55 in tRNA is a nearly universally conserved RNA modification found in all three domains of life. This modification is catalyzed by TruB in bacteria and by Pus4 in eukaryotes, but so far the Ψ55 synthase has not been identified in archaea. In this work, we report the ability of two distinct pseudouridine synthases from the hyperthermophilic archaeon Pyrococcus furiosus to specifically modify U55 in tRNA in vitro. These enzymes are _pfuCbf5, a protein known to play a role in RNA-guided modification of rRNA, and _pfuPsuX, a previously uncharacterized enzyme that is not a member of the TruB/Pus4/Cbf5 family of pseudouridine synthases. _pfuPsuX is hereafter renamed _pfuPus10. Both enzymes specifically modify tRNA U55 in vitro but exhibit differences in substrate recognition. In addition, we find that in a heterologous in vivo system, _pfuPus10 efficiently complements an Escherichia coli strain deficient in the bacterial Ψ55 synthase TruB. These results indicate that it is probable that _pfuCbf5 or _pfuPus10 (or both) is responsible for the introduction of pseudouridine at U55 in tRNAs in archaea. While we cannot unequivocally assign the function from our results, both possibilities represent unexpected functions of these proteins as discussed herein.     

Modification of human U4 RNA requires U6 RNA and multiple pseudouridine synthases.

Small nuclear RNAs (snRNA), cofactors in the splicing of pre-mRNA, are highly modified. In this report the modification of human U4 RNA was studied using cell extracts and in vitro synthesized, and therefore unmodified, U4 RNA. The formation of pseudouridine (Psi) at positions 4, 72 and 79 in U4 RNA was dependent on an RNA-containing cofactor, since the activities in the extracts were micrococcal nuclease (MN) sensitive. Extracts were fractionated on glycerol gradients and there was a broad peak of reconstitution activity centered at 14 S. Reconstitution was not due to additional enzymatic activity, since the peak fraction was MN sensitive. Oligodeoxynucleotide-mediated RNase H digestion of U6 RNA in the extracts inhibited formation of Psi in U4 RNA. From glycerol gradient analysis we determined that exogenously added U4 RNA that is associated with U6 RNA (sedimentation velocity 16 S) was significantly higher in Psi content than U4 RNA not associated with U6 RNA (8 S). Competitive inhibitors of Psi synthases, 5-fluorouridine-containing (5-FU) wild-type and mutant U4 RNAs, were used to investigate formation of Psi in U4 RNA. Deletions and point mutations in these 5-FU-containing U4 RNAs affected their ability to inhibit Psi synthase in vitro. With the aid of these potent inhibitors it was determined that at least two separate activities modify the uridines at these positions.     

Mechanistic investigations of the pseudouridine synthase RluA using RNA containing 5-fluorouridine

The pseuoduridine synthases (psi synthases) isomerize uridine (U) to pseudouridine (psi) in RNA, and they fall into five families that share very limited sequence similarity but have the same overall fold and active-site architecture, including an essential Asp. The mechanism by which the psi synthases operate remains unknown, and mechanistic work has largely made use of RNA containing 5-fluorouridine (f5U) in place of U. The psi synthase TruA forms a covalent adduct with such RNA, and heat disruption of the adduct generates a hydrated product of f5U, which was reasonably concluded to result from the hydrolysis of an ester linkage between the essential Asp and f5U. In contrast, the psi synthase TruB, which is a member of a different family, does not form an adduct with f5U in RNA but catalyzes the rearrangement and hydration of the f5U, which labeling studies with [18O]water showed does not result from ester hydrolysis. To extend the line of mechanistic investigation to another family of psi synthases and an enzyme that makes an adduct with f5U in RNA, the behavior of RluA toward RNA containing f5U was examined. Stem-loop RNAs are shown to be good substrates for RluA. Heat denaturation of the adduct between RluA and RNA containing f5U produces a hydrated nucleoside product, and labeling studies show that hydration does not occur by ester hydrolysis. These results are interpreted in light of a consistent mechanistic scheme for the handling of f5U by psi synthases.     

Dye label interference with RNA modification reveals 5-fluorouridine as non-covalent inhibitor

The interest in RNA modification enzymes surges due to their involvement in epigenetic phenomena. Here we present a particularly informative approach to investigate the interaction of dye-labeled RNA with modification enzymes. We investigated pseudouridine (Ψ) synthase TruB interacting with an alleged suicide substrate RNA containing 5-fluorouridine (5FU). A longstanding dogma, stipulating formation of a stable covalent complex was challenged by discrepancies between the time scale of complex formation and enzymatic turnover. Instead of classic mutagenesis, we used differentially positioned fluorescent labels to modulate substrate properties in a range of enzymatic conversion between 6% and 99%. Despite this variegation, formation of SDS-stable complexes occurred instantaneously for all 5FU-substrates. Protein binding was investigated by advanced fluorescence spectroscopy allowing unprecedented simultaneous detection of change in fluorescence lifetime, anisotropy decay, as well as emission and excitation maxima. Determination of K_d values showed that introduction of 5FU into the RNA substrate increased protein affinity by 14× at most. Finally, competition experiments demonstrated reversibility of complex formation for 5FU-RNA. Our results lead us to conclude that the hitherto postulated long-term covalent interaction of TruB with 5FU tRNA is based on the interpretation of artifacts. This is likely true for the entire class of pseudouridine synthases.     

Functional importance of motif I of pseudouridine synthases: mutagenesis of aligned lysine and proline residues

On the basis of sequence alignments, the pseudouridine synthases were grouped into four families that share no statistically significant global sequence similarity, though some common sequence motifs were discovered [Koonin, E. V. (1996) Nucleic Acids. Res. 24, 2411-2415; Gustafsson, C., Reid, R., Greene, P. J., and Santi, D. V. (1996) Nucleic Acids Res. 24, 3756-3762]. We have investigated the functional significance of these alignments by substituting the nearly invariant lysine and proline residues in Motif I of RluA and TruB, pseudouridine synthases belonging to different families. Contrary to our expectations, the altered enzymes display only very mild kinetic impairment. Substitution of the aligned lysine and proline residues does, however, reduce structural stability, consistent with a temperature sensitive phenotype that results from substitution of the cognate proline residue in Cbf5p, a yeast homologue of TruB [Zerbarjadian, Y., King, T., Fournier, M. J., Clarke, L., and Carbon, J. (1999) Mol. Cell. Biol. 19, 7461-7472]. Together, our data support a functional role for Motif I, as predicted by sequence alignments, though the effect of substituting the highly conserved residues was milder than we anticipated. By extrapolation, our findings also support the assignment of pseudouridine synthase function to certain physiologically important eukaryotic proteins that contain Motif I, including the human protein dyskerin, alteration of which leads to the disease dyskeratosis congenita.