Skip to main content
NCBI home page
RNA logoLink to RNA
. 2022 Nov;28(11):1509–1518. doi: 10.1261/rna.079327.122

Structures of RNA ligase RtcB in complexes with divalent cations and GTP

Agata Jacewicz 1, Swathi Dantuluri 1, Stewart Shuman 1
PMCID: PMC9745838  PMID: 36130078

Abstract

Pyrococcus horikoshii (Pho) RtcB exemplifies a family of binuclear transition metal- and GTP-dependent RNA ligases that join 3′-phosphate and 5′-OH ends via RtcB-(histidinyl-N)–GMP and RNA3'pp5'G intermediates. We find that guanylylation of PhoRtcB is optimal with manganese and less effective with cobalt and nickel. Zinc and copper are inactive and potently inhibit manganese-dependent guanylylation. We report crystal structures of PhoRtcB in complexes with GTP and permissive (Mn, Co, Ni) or inhibitory (Zn, Cu) metals. Zinc and copper occupy the M1 and M2 sites adjacent to the GTP phosphates, as do manganese, cobalt, and nickel. The identity/positions of enzymic ligands for M1 (His234, His329, Cys98) and M2 (Cys98, Asp95, His203) are the same for permissive and inhibitory metals. The differences pertain to: (i) the coordination geometries and phosphate contacts of the metals; and (ii) the orientation of the His404 nucleophile with respect to the GTP α-phosphate and pyrophosphate leaving group. M2 metal coordination geometry correlates with metal cofactor activity, whereby inhibitory Zn2 and Cu2 assume a tetrahedral configuration and contact only the GTP γ-phosphate, whereas Mn2, Co2, and Ni2 coordination complexes are pentahedral and contact the β- and γ-phosphates. The His404-Nε–Pα–O(α-β) angle is closer to apical in Mn (179°), Co (171°), and Ni (169°) structures than in Zn (160°) and Cu (155°) structures. The octahedral Mn1 geometry in our RtcB•GTP•Mn2+ structure, in which Mn1 contacts α-, β-, and γ-phosphates, transitions to a tetrahedral configuration after formation of RtcB•(His404)–GMP•Mn2+ and departure of pyrophosphate.

Keywords: RNA ligase, RNA repair, covalent catalysis, GTP

INTRODUCTION

RtcB enzymes comprise a family of bacterial, archaeal, and eukaryal RNA ligases that participate in tRNA splicing, RNA repair, nonspliceosomal mRNA splicing, and RNA recombination (Englert et al. 2011; Popow et al. 2011; Tanaka and Shuman 2011; Tanaka et al. 2011a; Jurkin et al. 2014; Kosmaczewski et al. 2014; Lu et al. 2014; Moldovan et al. 2019). Unlike classic RNA and DNA ligases that catalyze ATP- and magnesium-dependent joining of 3′-OH and 5′-phosphate ends via ligase-(lysyl-Nζ)-AMP and A5′pp5′RNA/DNA intermediates, RtcB ligases seal broken RNAs with 5′-OH and either 2′,3′-cyclic phosphate (RNA > p) or 3′-phosphate (RNAp) ends in a reaction dependent on GTP and a transition metal cofactor, typically manganese. The distinctive chemistry of end-joining by RtcB was revealed via studies of Escherichia coli RtcB (EcoRtcB), which executes a multistep pathway entailing: (i) reaction of the enzyme with GTP to form a covalent RtcB-(His337-N)-GMP intermediate; (ii) hydrolysis of RNA > p to RNAp; (iii) transfer of guanylate from His337 to the polynucleotide 3′-phosphate to form a polynucleotide-3′pp5′G intermediate; and (iv) attack of a 5′-OH on the –NppG end to form a 3′–5′ phosphodiester splice junction and liberate GMP (Tanaka et al. 2011b; Chakravarty and Shuman 2012; Chakravarty et al. 2012).

The overall EcoRtcB ligation pathway, and the histidine guanylylation and phosphodiester synthesis steps in particular, require manganese as the divalent cation cofactor. Neither magnesium, calcium, cobalt, nickel, copper, nor zinc can replace manganese in the histidine guanylylation reaction (Tanaka et al. 2011b). Metal mixing experiments, in which EcoRtcB guanylylation reactions containing 2 mM manganese were supplemented with 2 mM of another divalent cation, revealed that cobalt, nickel, copper, and zinc virtually abolished EcoRtcB guanylylation in the presence of manganese. In contrast, magnesium and calcium had no effect on EcoRtcB guanylylation in combination with manganese. The metal requirement for the isolated step of phosphodiester synthesis was satisfied by manganese and to a feeble extent by nickel, but not by magnesium, calcium, cobalt, or zinc (Das et al. 2013). Mixing equivalent concentrations of manganese and other metals showed that: (i) cobalt and zinc effaced phosphodiester synthesis in the presence of manganese; (ii) the extent of sealing in the presence of nickel plus manganese was reduced to the level with nickel alone; and (iii) magnesium and calcium had no effect on manganese-dependent phosphodiester synthesis. These results engendered a hypothesis that EcoRtcB binds its metal cofactor via “soft” enzymic ligands (histidine nitrogens and a cysteine sulfur), such that “hard” metals like calcium and magnesium do not bind the active site, whereas soft metals such as zinc, copper, nickel, and cobalt do bind the active site (and out-compete manganese), but when so engaged are unable to sustain catalysis (Tanaka and Shuman 2011; Tanaka et al. 2011b).

Structural insights regarding the RtcB mechanism have emerged from crystal structures of Pyrococcus horikoshii RtcB (PhoRtcB), captured in several functional states: RtcB apoenzyme; RtcB complexes with Mn2+ or Mn2+•GTP(αS); a covalent RtcB-(His404-Nε)–GMP•Mn2+ intermediate; and RtcB in complex with a 5′-OH polynucleotide substrate (Okada et al. 2006; Englert et al. 2012; Desai et al. 2013; Banerjee et al. 2021). A key finding was that PhoRtcB binds two Mn2+ ions, separated by 3.5 Å, one of which (Mn1) coordinates the α phosphate of GTP (and the GMP phosphate in the covalent RtcB–pG complex) while the other (Mn2) contacts the GTP γ phosphate (Desai et al. 2013). Mn1 is coordinated by Cys98, His234, and His329 (conserved as Cys78, His185, and His281 in EcoRtcB). Mn2 is coordinated by Asp95, Cys98, and His203 (conserved as Asp75, Cys78, and His168 in EcoRtcB). The active site cysteine bridges the two manganese ions and its mutation to alanine abolishes all steps of the RtcB nucleic acid ligation pathway (Englert et al. 2012; Maughan and Shuman 2016). Cys98 was oxidized to cysteine sulfonic acid in the structure of PhoRtcB complexed with 5′-OH polynucleotide substrate, thereby precluding binding of the manganese cofactors, and raising the prospect that RtcB activity might be sensitive to modulation during oxidative stress (Banerjee et al. 2021). Indeed, human RtcB was shown to be susceptible to copper-dependent oxidative inactivation and could be shielded from inactivation by the flavoprotein oxidoreductase PYROXD1 (Asanović et al. 2021).

Although the available PhoRtcB structures highlight the soft quality of many of the enzymic manganese ligands, they do not provide insights into why metals other than manganese might not be permissive for catalysis. A plausible hypothesis is that the coordination geometries of permissive and nonpermissive metals differ at one or both metal-binding sites. Confounding this simple view is the fact that the coordination geometries of the catalytically permissive M1 and M2 manganese ions differ from one PhoRtcB structure to the next in the set of five manganese-containing PhoRtcB structures reported by two independent groups of investigators (Englert et al. 2012; Desai et al. 2013). The M1 manganese complex is variously modeled as tetrahedral, pentahedral, or octahedral according to whether the metal has four, five, or six coordinated atoms. The M2 metal complex is variously modeled as tetrahedral or pentahedral. Moreover, the geometry of the (Mn2+)2•GTPαS complex cannot be unambiguously equated to a Michaelis complex for the PhoRtcB guanylylation step in light of the authors’ statement that GTPαS is “an unreactive GTP analog” (Desai et al. 2013). New insights into RtcB metal specificity emerged from a recent study of human RtcB, wherein it was noted that: (i) cobalt was more effective than manganese as the cofactor for ligation; (ii) magnesium, zinc, and calcium were feeble cofactors; and (iii) nickel and copper were inert (Kroupova et al. 2021). A crystal structure of human RtcB in complex with GMP and two cobalt ions (taken as mimetic of a product complex subsequent to dissociation of the ligated RNA) highlighted that M1 and M2 cobalt ions adopt tetrahedral and octahedral geometries, respectively (Kroupova et al. 2021). This configuration differs from the manganese coordination complexes observed in the various PhoRtcB structures.

Our aim in the present study was to establish the metal specificity of PhoRtcB guanylylation and then interrogate the structures of PhoRtcB in complexes with GTP and a spectrum of permissive and nonpermissive metals. With respect to the former issue, we report that: (i) PhoRtcB is active in the presence of manganese (optimal), cobalt, or nickel; (ii) calcium and magnesium have feeble cofactor activity and do not inhibit manganese-dependent guanylylation; (iii) zinc and copper do not support guanylylation and strongly inhibit the manganese-dependent activity.

Here we report structures of PhoRtcB•GTP•(Me2+)2 complexes obtained by infusing GTP and either manganese, cobalt, nickel, copper, or zinc into crystals of PhoRtcB apoenzyme at room temperature. In the nonpermissive GTP•(Cu2+)2 and GTP•(Zn2+)2 complexes, the M2 metal is coordinated tetrahedrally while the M1 coordination complexes differ in their geometry (octahedral for Zn1 versus tetrahedral for Cu1). In the permissive GTP•(Mn2+)2, GTP•(Co2+)2 and GTP•(Ni2+)2 complexes, M2 is pentahedrally coordinated while M1 coordination is either octahedral (Mn) or pentahedral (Co, Ni). The His404-Nε–Pα–O(α-β) angle is closer to apical in the permissive Mn, Co, and Ni GTP complexes than in the inhibitory Zn and Cu GTP-bound structures. Thus, metal specificity appears to correlate with M2 coordination geometry and closeness to in-line orientation of the histidine nucleophile and the pyrophosphate leaving group.

We also report that crystals grown after preincubation of PhoRtcB with GTP and manganese at 75°C yielded two distinct RtcB–GMP adducts: (i) an “on-pathway” state in which GMP is covalently attached to His404-Nε and manganese occupies the M1 site with tetrahedral geometry, signaling a remodeling of the precatalytic octahedral GTP•Mn1 geometry subsequent to guanylate transfer; and (ii) an “off-pathway” state in which GMP is linked covalently to His234-Nε and the metal sites are unoccupied.

RESULTS

Divalent cation specificity of PhoRtcB in formation of the covalent enzyme-GMP intermediate

The RtcB ligation pathway is initiated by the reaction of RtcB with GTP and a divalent cation cofactor to form a covalent RtcB-(histidinyl-N)–GMP adduct (Tanaka et al. 2011b). To query the metal specificity of PhoRtcB, we incubated 5 µM enzyme with 10 µM [α32P]GTP for 20 min at 75°C, either in the absence of added divalent cation or in the presence of 2 mM manganese, cobalt, nickel, magnesium, calcium, zinc, or copper. Transfer of [32P]GMP from GTP to the 54 kDa PhoRtcB polypeptide was detected by SDS-PAGE analysis of the reaction mixtures and was strictly dependent on inclusion of a divalent cation (Fig. 1A). Guanylylation was optimal in the presence of manganese. Cobalt and nickel supported guanylylation at 30% and 21% of the level seen with manganese. Magnesium and calcium were weakly active (<5% of the activity with manganese). Zinc and copper were inert (Fig. 1A). We proceeded to conduct metal mixing experiments in which PhoRtcB guanylylation reactions containing 1 mM manganese were supplemented with 1 mM of another divalent cation. Zinc and copper abolished PhoRtcB guanylylation in the presence of manganese (Fig. 1B). In contrast, nickel, magnesium, and calcium had no deleterious effect on PhoRtcB guanylylation in combination with manganese, while cobalt reduced the extent of guanylylation by one-third compared to the manganese control (Fig. 1B). These findings engender the following inferences: (i) the soft metals manganese, cobalt, and nickel can bind the histidine/cysteine-rich enzymic metal ligands and are permissive for catalysis of PhoRtcB guanylylation; (ii) the hard metals magnesium and calcium bind weakly to the M1 and M2 sites and are thereby feeble activators of guanylylation and unable to inhibit manganese-dependent guanylylation under competitive conditions; and (iii) inhibitory soft metals copper and zinc out-compete manganese for one or both metal-binding sites on the enzyme, wherein engaged they are unable to support guanylylation reaction chemistry. To further elucidate the inhibition by copper and zinc, we titrated the nonpermissive metals into reactions containing 2 mM manganese and observed potent concentration-dependent inhibition of PhoRtcB guanylylation (Fig. 1C,D). The yield of RtcB–[32P]GMP was reduced by half by 50 µM zinc (Fig. 1C) and by two thirds by 50 µM copper (Fig. 1D), indicating that at least one of the two essential metal-binding sites has a ≥40-fold greater affinity for Zn2+/Cu2+ than for Mn2+.

FIGURE 1.

FIGURE 1.

Open in a new tab

Metal specificity of PhoRtcB. (A) Reaction mixtures (10 µL) containing 50 mM Tris-HCl (pH 8.0), 10 µM [α32P]GTP, 5 µM RtcB, and 2 mM of the indicated divalent cation (as the chloride salt) were incubated at 75°C for 20 min. Divalent cation was omitted from an RtcB-containing control reaction (lane –). (B) Reaction mixtures (10 µL) containing 50 mM Tris-HCl (pH 8.0), 10 µM [α32P]GTP, 5 µM RtcB, 1 mM MnCl2, and 1 mM of the indicated divalent cation were incubated at 75°C for 20 min. (C,D) Reaction mixtures (10 µL) containing 50 mM Tris-HCl (pH 8.0), 10 µM [α32P]GTP, 5 µM RtcB, 2 mM MnCl2, and increasing concentrations of ZnCl2 or CuCl2 (as specified above the lanes) were incubated at 75°C for 20 min. The reaction products were analyzed by SDS-PAGE. Autoradiographs of the gels are shown. The positions and sizes of prestained marker proteins are indicated on the left. The relative extents of guanylylation in these assays are specified below the lanes, normalized to those of the manganese-only reactions (defined as 100%).

Structures of PhoRtcB in complexes with GTP and zinc or copper

PhoRtcB crystals were grown at room temperature by sitting drop vapor diffusion after mixture with an equal volume of precipitant solution containing 2.1 M ammonium sulfate and 0.2 M lithium sulfate. As reported previously by Desai et al. (2013), who used an ammonium sulfate-based crystallization procedure, the crystals were in space group P212121 and contained two RtcB protomers in the asymmetric unit. Several sulfate anions were present near the active sites of both protomers, at locations that are presumed to mimic phosphates of the nucleic acid substrates. The crystals of the protein per se had no nucleotide or divalent cation in the active site of either protomer. To obtain structures of PhoRtcB in complexes with nonpermissive metals and GTP, the apoenzyme crystals were soaked for 60 min in precipitant solution containing 10 mM GTP and 2 mM zinc chloride or 10 mM copper chloride. The structure of the PhoRtcB•GTP•(Zn2+)2 complex at 2.22 Å resolution was refined to Rwork/Rfree of 17.0/19.7 (Supplemental Table S1). As reported previously (Englert et al. 2012; Desai et al. 2013), the guanosine nucleoside of GTP receives base-specific contacts from Lys480 to the guanine O6 atom and Glu206 to the guanine N1 and N2 atoms. Anomalous difference peaks (contoured at 15σ in Fig. 2A) corresponding to two zinc atoms are seen adjacent to the GTP phosphates (Fig. 2A). Zn1 is engaged with octahedral geometry to His234-Nε (2.3 Å), His329-Nε (2.3 Å), Cys98-Cγ (2.3 Å), an α-phosphate oxygen (2.2 Å), a β-phosphate oxygen (2.3 Å), and a water (2.2 Å) that bridges to the γ-phosphate (Fig. 2A). Zn2 adopts a tetrahedral coordination complex filled by Asp95-Oδ (2.0 Å), Cys98-Cγ (2.3 Å), His203-Nε (2.2 Å), and a γ-phosphate oxygen (1.8 Å) (Fig. 2A). Zn2 is positioned 2.7 Å from the β-phosphate oxygen that is engaged by Zn1. Asn202-Nδ donates a hydrogen bond to the α-β bridging oxygen; Asn330-Nδ makes a hydrogen bond to a γ-phosphate oxygen. In the PhoRtcB•GTP•Zn2+ complex, the His404-Nε atom is situated 3.6 Å from the α-phosphorus atom (magenta dashed line in Fig. 2A) and subtends an N–P–O angle of 160° with respect to the GTP α-β bridging oxygen.

FIGURE 2.

FIGURE 2.

Open in a new tab

Structures of PhoRtcB in complexes with GTP and zinc or copper. Stereo views of the GTP•(Zn2+)2 (A) and GTP•(Cu2+)2 complexes (B) are shown. GTP is rendered as a cartoon model with gray carbons and yellow phosphorus atoms. Amino acids contacting GTP and metals are labeled and depicted as stick models with beige carbons. Atomic contacts are indicated by black dashed lines. The distance between the His404-Nε nucleophile and the GTP α-phosphorus is denoted by a magenta dashed line. Two Zn atoms (A) and Cu atoms (B) are rendered as orange spheres and green spheres, respectively. Anomalous difference peaks overlying the Zn and Cu atoms (blue mesh) are contoured at 15σ and 7σ, respectively.

The structure of the PhoRtcB•GTP•(Cu2+)2 complex at 2.42 Å resolution was refined to Rwork/Rfree of 16.5/20.4 (Supplemental Table S1). Anomalous difference peaks overlying the two copper atoms are contoured at 7σ in Figure 2B. Cu1 is engaged with tetrahedral geometry to His234-Nε (2.1 Å), His329-Nε (2.1 Å), Cys98-Cγ (2.1 Å), and an α-phosphate oxygen (2.8 Å). Cu2 is tetrahedrally coordinated to Asp95-Oδ (2.9 Å), Cys98-Cγ (2.0 Å), His203-Nε (2.0 Å), and the β-γ bridging oxygen (2.8 Å) (Fig. 2B). Here, the His404-Nε atom is situated 3.9 Å from the α-phosphorus atom at an N–P–O angle of 155° with respect to the GTP α-β bridging oxygen.

Whereas the RtcB active site amino acids and the GMP moiety of GTP are superimposable in the zinc-bound and copper-bound structures, the pyrophosphate leaving groups adopt slightly different conformations (the β and γ phosphorus atoms are shifted by 0.5 and 0.7 Å, respectively), the M1 and M2 metals are shifted by 0.5 and 0.7 Å, respectively, and the metal coordination complexes differ in their geometry (octahedral for Zn1 versus tetrahedral for Cu1), phosphate contacts (Zn2 to γ-phosphate oxygen versus Cu2 contact to β-γ bridging oxygen), and metal-phosphate interatomic distances.

Structure of PhoRtcB in complex with GTP and manganese

The structure of the PhoRtcB•GTP•(Mn2+)2 complex at 2.47 Å resolution was refined to Rwork/Rfree of 18.3/21.8 (Supplemental Table S1). Anomalous difference peaks overlying the two manganese atoms are contoured at 3σ in Figure 3A. Mn1 is coordinated with octahedral geometry to His234-Nε (2.4 Å), His329-Nε (2.3 Å), Cys98-Cγ (2.3 Å), an α-phosphate oxygen (2.1 Å), a β-phosphate oxygen (2.2 Å), and a γ-phosphate oxygen (2.3 Å). Mn2 is pentahedrally coordinated to Asp95-Oδ (2.2 Å), Cys98-Cγ (2.2 Å), His203-Nε (2.2 Å), a β-phosphate oxygen (2.2 Å), and a γ-phosphate oxygen (2.1 Å) (Fig. 3A). Asn202-Nδ makes a bifurcated hydrogen bond to the α-β bridging oxygen and to the β-phosphate oxygen not coordinated by Mn1 and Mn2. The His404-Nε atom is poised 3.6 Å from the α-phosphorus atom at an N–P–O angle of 179° with respect to the GTP α-β bridging oxygen.

FIGURE 3.

FIGURE 3.

Open in a new tab

Structures of PhoRtcB in complexes with GTP and manganese, cobalt, or nickel. Stereo views of the GTP•(Mn2+)2 (A), GTP•(Co2+)2 (B), and GTP•(Ni2+)2 complexes (C) are shown. Atomic contacts to GTP and metals are indicated by black dashed lines. The distance between the His404-Nε nucleophile and the GTP α-phosphorus is denoted by a magenta dashed line. Mn atoms (A), Co atoms (B), and Ni atoms (C) are rendered as magenta, cyan, and light green spheres, respectively. Anomalous difference peaks overlying the Mn and Co atoms (green mesh) and Ni atoms (blue mesh) are contoured at 3σ, 14σ, and 10σ, respectively.

Structures of PhoRtcB in complexes with GTP and cobalt or nickel

The structure of the PhoRtcB•GTP•(Co2+)2 complex at 2.43 Å resolution was refined to Rwork/Rfree of 17.7/20.3 (Supplemental Table S1). Anomalous difference peaks overlying the two cobalt atoms are contoured at 14σ in Figure 3B. Co1 is coordinated pentahedrally to His234-Nε (2.4 Å), His329-Nε (2.3 Å), Cys98-Cγ (2.6 Å), an α-phosphate oxygen (2.0 Å), and a β-phosphate oxygen (2.5 Å) (Fig. 3B). [Co1 is 2.9 Å from the γ-phosphate oxygen that occupies a sixth ligand site for Mn1 in the manganese-bound RtcB structure (Fig. 3A).] Co2 is pentahedrally coordinated to Asp95-Oδ (2.2 Å), Cys98-Cγ (2.4 Å), His203-Nε (2.3 Å), a β-phosphate oxygen (2.5 Å), and a γ-phosphate oxygen (2.3 Å) (Fig. 3A). The His404-Nε atom is situated 3.4 Å from the α-phosphorus atom at a N–P–O angle of 171° to the GTP α-β bridging oxygen.

The active site of the 2.60 Å structure of a PhoRtcB•GTP•(Ni2+)2 complex (Rwork/Rfree of 17.0/20.9) is shown in Figure 3C with anomalous difference peaks overlying the two nickel atoms contoured at 10σ. In this case, Ni1 is coordinated octahedrally to His234-Nε (2.3 Å), His329-Nε (2.2 Å), Cys98-Cγ (2.6 Å), an α-phosphate oxygen (2.1 Å), a β-phosphate oxygen (2.2 Å), and a water (2.3 Å) that bridges to the γ-phosphate (Fig. 3C). (The Ni1 coordination complex is nearly identical to the octahedral Zn1 complex shown in Fig. 2A). Ni2 is pentahedrally coordinated to Asp95-Oδ (2.5 Å), Cys98-Cγ (2.2 Å), His203-Nε (2.2 Å), a β-phosphate oxygen (2.4 Å), and a γ-phosphate oxygen (2.2 Å) (Fig. 3C). The His404-Nε atom is poised 3.5 Å from the α-phosphorus atom at a N–P–O angle of 169° to the α-β bridging oxygen.

Structures of PhoRtcB as covalent His–GMP adducts

The fact that structures of PhoRtcB obtained after soaking apoenzyme crystals at room temperature with GTP and permissive metal ions (Mn2+, Co2+, Ni2+) did not result in conversion of PhoRtcB•GTP•(Me2+)2 to the covalent PhoRtcB–GMP intermediate accords with the thermophilic nature of the enzyme, whereby PhoRtcB guanylylation is optimal at 70°C–80°C (Englert et al. 2012; Desai et al. 2013). Here we conducted crystallization trials of PhoRtcB that had been preincubated for 60 min at 75°C with 2 mM GTP and 2 mM MnCl2. A crystal grown by mixture with precipitant solution containing 2.0 M ammonium sulfate and 0.2 M potassium formate diffracted to 2.21 Å, was in space group P212121, and was isomorphous with the GTP-soaked crystals described above. The A protomer was guanylylated on His404-Nε, whereas the active site of the B protomer was unoccupied. The RtcB-(His404)–GMP complex included a single manganese ion at the M1 site, tetrahedrally coordinated to His234-Nε, His329-Nε, Cys98-Cγ, and an α-phosphate oxygen, and is virtually identical to the structures reported previously for PhoRtcB-(His404)–GMP in which the M1 coordination complex was also tetrahedral (Englert et al. 2012; Desai et al. 2013). Comparison to our PhoRtcB•GTP•(Mn2+)2 structure (a putative mimetic of a Michaelis complex in the RtcB guanylylation reaction in which Mn1 is octahedrally engaged) suggests that the M1 metal coordination complex is remodeled subsequent to catalysis and ejection of the pyrophosphate leaving group (likely in association with the M2 manganese).

Another crystal grown from the same preincubated RtcB preparation using a precipitant solution containing 2.2 M ammonium sulfate was in space group P43 and contained two protomers in the asymmetric unit. The structure was refined to 2.35 Å resolution (Rwork/Rfree = 19.1/21.7; Supplemental Table S2). Both protomers were guanylylated, albeit on His234-Nε, as affirmed by the omit map in Figure 4 showing continuous density between His234 and the GMP phosphate. In this structure, His329-Nδ donates a hydrogen bond to a GMP phosphate oxygen. No metal is present in the active site, consistent with the fact that His234 (an M1 ligand) is otherwise deployed in this structure. It is most likely that the His234–GMP adduct is off-pathway, insofar as: (i) mutating PhoRtcB His234 to alanine did not diminish RtcB guanylylation (Englert et al. 2012); and (ii) the equivalent His185 to alanine mutation in EcoRtcB did not affect HORNAp ligation (Maughan and Shuman 2016). We speculate below in the Discussion section on how the His234–GMP adduct might arise.

FIGURE 4.

FIGURE 4.

Open in a new tab

Structure of PhoRtcB as covalent His234–GMP adduct. Stereo view of the His234–GMP adduct overlaid with a simulated annealing omit density map (green mesh) contoured at 2.5 σ. Amino acids in the vicinity are depicted as stick models with beige carbons.

DISCUSSION

The present study affords new insights into the fastidious metal specificity of Pyrococcus RtcB, which is: (i) preferentially activated for RtcB guanylylation by manganese and less effectively by cobalt and nickel; (ii) indifferent to magnesium and calcium, which are feeble activators and do not inhibit manganese-driven activity; and (iii) antagonized by zinc and copper, which fail to support activity themselves and potently inhibit RtcB guanylylation in the presence of manganese. It was unclear whether the “bad” metals could occupy one or both manganese sites documented in previous PhoRtcB structures. By solving a new series of structures of PhoRtcB in complexes with GTP and either permissive (Mn, Co, Ni) or nonpermissive/inhibitory (Zn, Cu) metals, we draw the following conclusions and inferences. First, that zinc and copper occupy the M1 and M2 sites adjacent to the GTP phosphates, in more or less the same positions as the permissive metals manganese, cobalt and nickel (Fig. 5). Second, that the identity and positions of the enzymic ligands for M1 (His234, His329, Cys98) and M2 (Cys98, Asp95, His203) are also the same for the permissive and bad metals (Fig. 5). The pertinent differences between the structures concern: (i) the coordination geometries of the M1 and M2 metals; (ii) the contacts and conformations of the GTP β-phosphate and γ-phosphate (Fig. 5); and (iii) the orientation of the His404 nucleophile with respect to the α-phosphate and the pyrophosphate leaving group. It remains an open question whether the potent inhibition of manganese-dependent RtcB guanylylation by zinc and copper entails occupancy of one or both of the manganese sites by the bad metal.

FIGURE 5.

FIGURE 5.

Open in a new tab

Superposition of PhoRtcB•GTP•(Me2+)2 structures. The five GTP•(Me2+)2 complexes are superimposed. The amino acids M1 and M2 sites are labeled. The carbon, phosphorus, and metal atoms are colored as depicted above the stereo image.

The M2 metal coordination geometry correlates with metal cofactor activity in guanylylation versus metal inhibition, insofar as inhibitory Zn2 and Cu2 assume a tetrahedral configuration and contact only the GTP γ-phosphate, whereas the pentahedral Mn2, Co2, and Ni2 coordination complexes entail contacts to the β- and γ-phosphates. In the “nonproductive” complex of PhoRtcB with GTPαS and manganese, the Mn2 ion is also tetrahedrally coordinated and contacts only the γ-phosphate (Desai et al. 2013). In contrast, the M1 geometry does not strictly correlate with activity: to wit, the octahedral Zn1 and Mn1 coordination complexes are nearly identical. The Co1 and Ni1 complexes are pentahedral in the structures reported here, which might explain why cobalt and nickel are less effective cofactors for guanylylation than manganese. The Cu1 complex with GTP is distinctively tetrahedral in this study and this might contribute to copper's inhibition of PhoRtcB guanylylation.

The formation of the RtcB–GMP pathway intermediate is presumed to entail an associative in-line attack of the histidine-Nε nucleophile on the GTP α-phosphorus via a pentacoordinate phosphorane transition state. This obliges the His-Nε to closely approach the α-phosphate and assume an apical orientation with respect to the GTP α-β bridging oxygen. The degree to which this state is approximated in the RtcB•GTP substrate complex will influence the progress of the guanylylation reaction. The trend we see here is that the His404-Nε–Pα–O(α-β) angle is closer to apical in the permissive Mn (179°), Co (171°), and Ni (169°) structures than in the inhibitory Zn (160°) and Cu (155°) structures. The angle is better correlated with metal cofactor activity than is the His404-Nε to Pα distance, which is similar in the structures solved: Mn (3.6 Å), Co (3.4 Å), Ni (3.5 Å), Zn (3.6 Å), Cu (3.9 Å).

The present study highlights the emerging theme that RtcB enzymes from divergent taxa display differences in their ability to use metal cofactors other than manganese. Whereas EcoRtcB is stringent in its reliance on manganese (Tanaka and Shuman 2011; Tanaka et al. 2011b), PhoRtcB is accepting of cobalt and nickel (this study), while human RtcB prefers cobalt over manganese and can also use magnesium and zinc, albeit less effectively than manganese (Kroupova et al. 2021). EcoRtcB and PhoRtcB share the property that zinc and copper are potently inhibitory when mixed with manganese. Copper is an ineffective cofactor per se for human RtcB (Kroupova et al. 2021); to our knowledge, the effects of metal mixing on human RtcB activity have not been reported. The reasons for these differences in metal utilization and inhibition among RtcB enzymes are unclear given that the amino acids that serve as M1 and M2 ligands are conserved.

In agreement with prior studies (Englert et al. 2012; Desai et al. 2013), we find that the M1 manganese adopts a tetrahedral configuration in the context of the covalent PhoRtcB-(His404)–GMP intermediate, wherein the Mn1 ligands are His234, His329, Cys98-Cγ, and a GMP α-phosphate oxygen. This contrasts with the octahedral Mn1 geometry in our PhoRtcB•GTP•Mn2+ structure, in which the Mn1 complex is filled by His234, His329, Cys98-Cγ, an α-phosphate oxygen, a β-phosphate oxygen, and a γ-phosphate oxygen. The transition from octahedral to tetrahedral Mn1 configuration pre- versus post-catalysis of RtcB guanylylation is sensible insofar as the departure of the pyrophosphate leaving group severs the Mn1 contacts to the β-phosphate and γ-phosphate oxygens. An equivalent tetrahedral Co1 coordination geometry is seen in the structure of human RtcB with GMP bound noncovalently (Kroupova et al. 2021). It is conceivable that the M1 and M2 geometries undergo additional transitions at downstream steps along the RtcB ligation reaction pathway that have yet to be captured structurally.

Finally, we observed here a new off-pathway state of PhoRtcB guanylylation in which GMP is covalently attached to His234 instead of His404. We can envision two scenarios to account for its formation: (i) GTP binds to PhoRtcB with its triphosphate moiety in a different conformation that allows for direct in-line attack of His234-Nε on the GTP α-phosphorus; or (ii) PhoRtcB reacts with GTP to form the His404–GMP intermediate, which is then subject to attack by His234-Nε on the α-phosphorus resulting in guanylate transfer from His404 to His234. The prospect that PhoRtcB can bind GTP in more than one conformation had been suggested earlier by Englert et al. (2012), who highlighted two ambiguous electron density peaks adjacent to His404–GMP that they interpreted either as: (i) partial occupancy by a pyrophosphate product of the guanylylation reaction; or (ii) the β- and γ-phosphates of residual unreacted GTP. As they discuss in their publication supplement, GTP fit the density better than pyrophosphate plus His–GMP. In this GTP conformation, the His404-Nε is not properly aligned with Pα–O(α-β), indicating that the orientation of the GTP triphosphate is not on pathway for His404 guanylylation (Englert et al. 2012). Rather, they speculate that another histidine (e.g., His329) might react with this form of GTP. If it is the case that the alternative GTP-bound state discussed by Englert et al. can set up an alternative direct histidine guanylylation reaction, our structure here indicates that His234 is the culprit. As for the alternative model of sequential GMP transfer from GTP to His404 to His234, the structure of the His404–GMP•Mn2+ intermediate indicates that it is not poised for such a transfer, insofar as His234-Nε is 5 Å from the GMP phosphorus and the geometry is highly unfavorable (His234-Nε–Pα–His404-Nε angle of 108°). However, the reaction might proceed if the M1 metal departs (as is the case in the His234–GMP structure) and thereby allows for repositioning of active site constituents leading to GMP transfer to His234.

MATERIALS AND METHODS

PhoRtcB

A pET28a-based bacterial expression plasmid with a codon-optimized complete PhoRtcB open reading frame inserted between the NdeI and XhoI restriction sites was purchased from GenScript. His6-tagged PhoRtcB was produced in a 2-L culture of plasmid-bearing E. coli BL21(DE3)-CodonPlus cells, initially grown at 37°C in Terrific Broth containing 0.4% (v/v) glycerol, 50 µg/mL kanamycin, and 35 µg/mL chloramphenicol, and induced during an overnight incubation at 17°C with 0.5 mM IPTG. Cells were harvested by centrifugation and the pellets were resuspended in 25 mL lysis buffer (50 mM Tris-HCl, pH 8.0, 1 M NaCl, 10 mM TCEP, 20 mM imidazole, 10% glycerol) and stored at −80°C. Subsequent purification was performed at 4°C unless noted otherwise. The thawed cell suspension in lysis buffer was adjusted to 1 mg/mL lysozyme and supplemented with one EDTA-free protease inhibitor cocktail tablet (Roche; cat. no. 11836170001) and incubated for 30 min. The lysates were sonicated to reduce viscosity, and insoluble material was removed by centrifugation in a Fiberlite F18-12X50 rotor for 45 min at 14,000 rpm. The soluble extract was mixed for 1 h with 5 mL of Ni-NTA agarose (Qiagen) that had been equilibrated in lysis buffer. The resin was collected by centrifugation and resuspended in 50 mL of buffer A (50 mM Tris-HCl, pH 8.0, 300 mM NaCl, 5 mM TCEP) containing 20 mM imidazole, then recollected by centrifugation. The resin was resuspended in 50 mL of buffer A containing 50 mM imidazole. The resin was then poured into a gravity flow column and the bound material was serially step-eluted with 150, 250, and 500 mM imidazole in buffer A. The elution profiles were monitored by SDS-PAGE. The peak RtcB-containing fractions were pooled, concentrated by centrifugal ultrafiltration at room temperature, and then subjected to gel filtration through a 120-mL Superdex 200 column that was equilibrated with 20 mM Tris-HCl, pH 8.0, 200 mM NaCl, 2 mM TCEP, 5% glycerol. Peak fractions were pooled and concentrated at room temperature to 4.2 mg/mL by centrifugal ultrafiltration and stored at −80°C. Protein concentration was determined by using the BioRad dye reagent with bovine serum albumin as the standard.

Assay of PhoRtcB guanylylation

Reaction mixtures (10 µL) containing 50 mM Tris-HCl (pH 8.0), 10 µM [α32P]GTP (100 pmol GTP), 50 pmol (5 µM) PhoRtcB, and divalent cations (chloride salts) as specified in figure legends were incubated at 75°C for 20 min. The reactions were quenched by adding 15 µL of a solution containing 125 mM Tris (pH 6.8), 250 mM DTT, 5% SDS, 15% glycerol, and 0.01% bromphenol blue. The mixtures were analyzed by SDS-PAGE (12.5% acrylamide). 32P-GMP transfer to PhoRtcB was visualized by autoradiography. The gels were scanned with a Fuji BAS-2500 imaging apparatus, and 32P-PhoRtcB was quantified using ImageQuant software.

Crystallization of PhoRtcB•Me•GTP complexes

Crystals of the PhoRtcB protein in complex with metals and GTP were obtained in a series of ligand-soaking experiments with preformed crystals of the apoenzyme. Apoenzyme PhoRtcB crystals were grown by the sitting drop vapor diffusion at 22°C. Aliquots (2 µL) of 0.12 mM protein were mixed with an equal volume of reservoir solution containing 2.05 to 2.1 M ammonium sulfate and 0.2 M lithium sulfate. Crystals grew within 1 to 2 d and were used for soaking experiments after a week. Soaks were prepared by direct transfer of apocrystals to soaking solutions containing 2.1 M ammonium sulfate, 0.2 M lithium sulfate, 2 to 10 mM MnCl2, CoCl2, NiCl2, ZnCl2, or CuCl2, and 10 mM GTP (pH 7.5). Crystals were incubated with ligands for 60 min (Co/GTP, Ni/GTP, Zn/GTP and Cu/GTP soaks) or 16–18 h (Mn/GTP soak). Crystals were cryoprotected in 2.0 M ammonium sulfate, 0.2 M lithium sulfate, 2 to 10 mM MnCl2, CoCl2, NiCl2, ZnCl2, or CuCl2, 5 to 8 mM GTP, 20% (w/v) sucrose and then flash-frozen in liquid nitrogen.

X-ray diffraction and structure determination of PhoRtcB•Me•GTP complexes

X-ray diffraction data from single GTP and metal-soaked PhoRtcB crystals in space group P212121 were collected at the Advanced Photon Source beamlines 24ID-C and 24ID-E. Reduction of all crystallographic data was performed using XDS (Kabsch 2010) and AIMLESS (CCP4 suite) (Evans and Murshudov 2013). The structures were solved by molecular replacement in combination with single-wavelength anomalous dispersion (MR-SAD) as implemented in PHENIX.PHASER (Adams et al. 2010) using pdb 4ISZ or 4ISJ as search models (protein only). All crystals contained two RtcB protomers in the asymmetric unit. Phases were optimized by density modification in PHENIX.RESOLVE. The models were iteratively improved in PHENIX.REFINE without imposing noncrystallographic symmetry (NCS) restraints and included translation libration screw (TLS) B-factor refinement, interspersed with manual adjustments of models in COOT (Emsley and Cowtan 2004). Data collection and refinement statistics are compiled in Supplemental Table S1.

Structures of guanylylated PhoRtcB

PhoRtcB (0.17 mM) was preincubated with 2 mM MnCl2 and 2 mM GTP for 60 min at 75°C. The mixture was centrifuged for 20 min at 16,800g in the benchtop centrifuge at 22°C. The supernatant was spun through a 0.22 µm Spin-X centrifuge tube filter (Corning Inc.). The clarified solution contained 8.4 mg/mL PhoRtcB. Crystals were grown by sitting-drop vapor-diffusion at 22°C by mixing protein aliquots with equal volumes of reservoir buffers. In one variation, 2 µL of protein sample was mixed with 2 µL of reservoir solution containing 0.2 M potassium formate and 2.0 M ammonium sulfate. Crystals appeared overnight and were grown for 2 wk before they were harvested and used for streak seeding into the same reservoir solution to improve crystal size and shape. Crystals were cryoprotected with 20% sucrose in reservoir solution before flash-freezing in liquid nitrogen. Diffraction data from a single crystal were collected at the Advanced Photon Source beamline 24ID-E. The crystal diffracted to 2.21 Å resolution, belonged to space group P212121, contained two protomers in the ASU, and was isomorphous with the metal/GTP-soaked crystals described above. The structure was solved by MR-SAD implemented PHENIX.PHASER using 4IT0 as search model (protein only). The electron density map revealed a covalent adduct between GMP and His404-Nε and a single manganese ion in the M1 site. The structure of the RtcB-(His404)–GMP•Mn2+ complex was virtually identical to that reported previously (Desai et al. 2013) and is therefore not depicted here or deposited in PDB.

In a second variation of the crystallization experiment, 0.5 µL of the clarified preincubated PhoRtcB solution was mixed with 0.5 µL of precipitant solution containing 2.2 M ammonium sulfate. Crystals were harvested after 6 d, cryoprotected with 22% sucrose in reservoir solution and flash-frozen in liquid nitrogen. Diffraction data from a single crystal were collected at the Advanced Photon Source beamline 24ID-C. The crystal diffracted to 2.35 Å resolution, belonged to space group P43, and contained two protomers in the ASU. The structure was solved by MR using 4DWQ (without heteroatoms) as a search ensemble. Following density modification in PHENIX.RESOLVE, refinement was executed in Phenix (without NCS and including TLS B-factor refinement). Iterative model building into electron density was performed in COOT. The electron density map revealed that GMP was covalently attached to His234-Nδ in both PhoRtcB protomers and that the M1 and M2 sites were unoccupied. Data collection and refinement statistics for the PhoRtcB-(His234)–GMP structure are compiled in Supplemental Table S2.

DATA DEPOSITION

Structural coordinates have been deposited in Protein Data Bank under accession codes 8DC9, 8DCA, 8DCB, 8DCD, 8DCF, and 8DCG.

SUPPLEMENTAL MATERIAL

Supplemental material is available for this article.

Supplementary Material

Supplemental Material
supp_28_11_1509__DC1.html (800B, html)

ACKNOWLEDGMENTS

This research was supported by National Institutes of Health (NIH) grant R35-GM126945 (S.S.). The MSKCC structural biology core laboratory is supported by National Cancer Institute grant P30-CA008748. X-ray diffraction data were collected at synchrotron facilities supported by grants and contracts from the National Institutes of Health (P30-GM124165, HEI-S10OD021527) and the Department of Energy (DE-AC02-06CH11357). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Footnotes

MEET THE FIRST AUTHOR

Agata Jacewicz.

Agata Jacewicz

Open in a new tab

Meet the First Author(s) is a new editorial feature within RNA, in which the first author(s) of research-based papers in each issue have the opportunity to introduce themselves and their work to readers of RNA and the RNA research community. Agata Jacewicz is the first author of this paper, “Structures of RNA ligase RtcB in complexes with divalent cations and GTP.” Agata is a senior research scientist in Stewart Shuman's laboratory in the Molecular Biology program at Sloan Kettering Institute in New York. The main focus of her research is to understand mechanisms and structures of enzymes that perform and regulate essential nucleic acid transactions.

What are the major results described in your paper and how do they impact this branch of the field?

Previous studies had shown that, in the first step in RNA ligation catalyzed by E. coli RtcB, an essential histidine nucleophile performs an in-line attack on the α-phosphorus of GTP, which results in the formation of the covalent enzyme-GMP intermediate and release of pyrophosphate. This reaction requires manganese as the divalent cation cofactor. In our study, using a thermophilic RtcB ligase from Pyrococcus horikoshii (PhoRtcB), we show that metals other than manganese, cobalt, and nickel are activators of PhoRtcB guanylylation, whereas zinc and copper are inhibitory. Our manuscript provides structural insights into PhoRtcB metal specificity. By solving a series of crystal structures of PhoRtcB in complexes with the GTP and either permissive or nonpermissive metals, we observed that both metal sites (M1 and M2) can be occupied by “bad” or “good” metals. However, the bad metals in the M2 site adopt different coordination geometry and make fewer contacts to the GTP phosphates when compared to good metals. Additionally, in “productive” enzyme complexes, the histidine-Nε nucleophile assumes closer-to-optimal orientation with respect to the GTP α-β bridging oxygen as compared to its position observed in “nonproductive” enzyme complexes. In sum, our work provides added granularity that speaks to the mechanisms involved in RNA ligation.

What led you to study RNA or this aspect of RNA science?

Our study highlights the emerging theme that RtcB enzymes from divergent taxa display differences in their ability to use metal cofactors other than manganese. For instance: RtcB from E. coli displays a stringent preference for manganese, whereas the human RtcB homolog preferably uses cobalt, but also performs the reaction when supplemented with manganese, magnesium and zinc. The reasons for these differences in metal utilization and inhibition among RtcB enzymes are unclear, and hence interesting to study.

What are some of the landmark moments that provoked your interest in science or your development as a scientist?

There are many important small-to-medium sized proteins that have been recalcitrant to crystallization in my hands so far. I am fascinated by the possibility of studying those proteins using cryogenic electron microscopy, which to date, has been largely applied to the study of large proteins and/or protein complexes.

If you were able to give one piece of advice to your younger self, what would that be?

Trust your intuition, ask questions, and always strive to evolve as a scientist. Remember, that technical competency is as important as keeping a critical mind.

REFERENCES

  1. Adams PD, Afonine PV, Bunkóczi G, Chen VB, Davis IW, Echols N, Headd JJ, Hung LW, Kapral GJ, Grosse-Kunstleve RW, et al. 2010. PHENIX: a comprehensive Python-based system for macromolecular structure solution. Acta Crystallogr D Biol Crystallogr 66: 213–221. 10.1107/S0907444909052925 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Asanović I, Strandback E, Kroupova A, Pasajlic D, Meinhart A, Tsung-Pin P, Djokovic N, Anrather D, Schuetz T, Suskiewicz MJ, et al. 2021. The oxidoreductase PYROXD1 uses NAD(P)+ as an antioxidant to sustain tRNA ligase activity in pre-tRNA splicing and unfolded protein response. Mol Cell 81: 2520–2532. 10.1016/j.molcel.2021.04.007 [DOI] [PubMed] [Google Scholar]
  3. Banerjee A, Goldgur Y, Shuman S. 2021. Structure of 3′-PO4/5′-OH RNA ligase RtcB in complex with a 5′-OH oligonucleotide. RNA 27: 584–590. 10.1261/rna.078692.121 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chakravarty AK, Shuman S. 2012. The sequential 2′,3′ cyclic phosphodiesterase and 3′-phosphate/5′-OH ligation steps of the RtcB RNA splicing pathway are GTP-dependent. Nucleic Acids Res 40: 8558–8567. 10.1093/nar/gks558 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Chakravarty AK, Subbotin R, Chait BT, Shuman S. 2012. RNA ligase RtcB splices 3′-phosphate and 5′-OH ends via covalent RtcB-(histidinyl)-GMP and polynucleotide-(3′)pp(5′)G intermediates. Proc Natl Acad Sci 109: 6072–6077. 10.1073/pnas.1201207109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Das U, Chakravarty AK, Remus BS, Shuman S. 2013. Rewriting the rules for end joining via enzymatic splicing of DNA 3′-PO4 and 5′-OH ends. Proc Natl Acad Sci 110: 20437–20442. 10.1073/pnas.1314289110 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Desai KK, Bingman CA, Phillips GN, Raines RT. 2013. Structures of the noncanonical RNA ligase RtcB reveal the mechanism of histidine guanylylation. Biochemistry 52: 2518–2525. 10.1021/bi4002375 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Emsley P, Cowtan K. 2004. Coot: model-building tools for molecular graphics. Acta Crystallogr D Biol Crystallogr 60: 2126–2132. 10.1107/S0907444904019158 [DOI] [PubMed] [Google Scholar]
  9. Englert M, Sheppard K, Aslanian A, Yates JR, Söll D. 2011. Archaeal 3′-phosphate RNA splicing ligase characterization identified the missing component in tRNA maturation. Proc Natl Acad Sci 108: 1290–1295. 10.1073/pnas.1018307108 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Englert M, Xia S, Okada C, Nakamura A, Tanavde V, Yoa M, Eom SH, Konigsberg WH, Söll D, Wang J. 2012. Structural and mechanistic insights into guanylylation of RNA-splicing ligase RtcB joining between 3′-terminal phosphate and 5′-OH. Proc Natl Acad Sci 109: 15235–15240. 10.1073/pnas.1213795109 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Evans PR, Murshudov GN. 2013. How good are my data and what is the resolution? Acta Crystallogr D Biol Crystallogr 69: 120411214. 10.1107/S0907444913000061 [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Jurkin J, Henkel T, Nielsen AF, Minnich M, Popow J, Kaufmann T, Heindl K, Hoffmann T, Busslinger M, Martinez J. 2014. The mammalian tRNA ligase complex mediates splicing of XBP1 mRNA and controls antibody secretion in plasma cells. EMBO J 33: 2922–2936. 10.15252/embj.201490332 [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Kabsch W. 2010. XDS. Acta Crystallogr D Biol Crystallogr 66: 125–132. 10.1107/S0907444909047337 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Kosmaczewski SG, Edwards TJ, Han SM, Eckwahl MJ, Meyer BI, Peach S, Hesselberth JR, Wolin SL, Hammarlund M. 2014. The RtcB RNA ligase is an essential component of the metazoan unfolded protein response. EMBO Rep 15: 1278–1285. 10.15252/embr.201439531 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Kroupova A, Ackle F, Asanovic I, Weitzer S, Boneberg FM, Faini M, Leitner A, Chui A, Aebersold R, Martinez J, et al. 2021. Molecular architecture of the human tRNA ligase complex. eLife 10: e71656. 10.7554/eLife.71656 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Lu Y, Liang FX, Wang X. 2014. A synthetic biology approach identifies the mammalian UPR RNA ligase RtcB. Mol Cell 55: 758–770. 10.1016/j.molcel.2014.06.032 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Maughan WP, Shuman S. 2016. Distinct contributions of enzymic functional groups to the 2′,3′-cyclic phosphodiesterase, 3′-phosphate guanylylation, and 3′-ppG/5′-OH ligation steps of the Escherichia coli RtcB nucleic acid splicing pathway. J Bacteriol 198: 1294–1304. 10.1128/JB.00913-15 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Moldovan JB, Wang Y, Shuman S, Mills RE, Moran JV. 2019. RNA ligation precedes the retrotransposition of U6/LINE-1 chimeric RNA. Proc Natl Acad Sci 116: 20612–20622. 10.1073/pnas.1805404116 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Okada C, Maegawa Y, Yao M, Tanaka I. 2006. Crystal structure of an RtcB homolog protein (PH1502-extein protein) from Pyrococcus horikoshii reveals a novel fold. Proteins 63: 1119–1122. 10.1002/prot.20912 [DOI] [PubMed] [Google Scholar]
  20. Popow J, Englert M, Weitzer S, Schleiffer A, Mierzwa B, Mechtler K, Trowitzsch S, Will CL, Lürhmann R, Söll D, et al. 2011. HSPC117 is the essential subunit of a human tRNA splicing ligase complex. Science 331: 760–764. 10.1126/science.1197847 [DOI] [PubMed] [Google Scholar]
  21. Tanaka N, Shuman S. 2011. RtcB is the RNA ligase component of an Escherichia coli RNA repair operon. J Biol Chem 286: 7727–7731. 10.1074/jbc.C111.219022 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Tanaka N, Meineke B, Shuman S. 2011a. RtcB, a novel RNA ligase, can catalyze tRNA splicing and HAC1 mRNA splicing in vivo. J Biol Chem 286: 30253–30257. 10.1074/jbc.C111.274597 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Tanaka N, Chakravarty AK, Maughan B, Shuman S. 2011b. Novel mechanism of RNA repair by RtcB via sequential 2′,3′-cyclic phosphodiesterase and 3′-phosphate/5′-hydroxyl ligation reactions. J Biol Chem 286: 43134–43143. 10.1074/jbc.M111.302133 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material
supp_28_11_1509__DC1.html (800B, html)
supp_079327.122_Supplemental_Tables.pdf (77.5KB, pdf)

Articles from RNA are provided here courtesy of The RNA Society

RESOURCES