Skip to main content
NCBI home page
Nature Communications logoLink to Nature Communications
. 2026 Sep 4;17:10523. doi: 10.1038/s41467-026-77369-4

Structural and functional insights into QueC-family protein in QatABCD anti-phage system

Zirui Gao 1,#, Feixue Li 1,#, Hao Wang 2,#, Xi Liu 1,#, Ningning Li 1,#, Weijia Xiong 1, Weidong Ma 1, Demei Sun 1, Dingfei Yan 3, Qingqing Ma 3, Lei Xu 1, Yi Zhang 1,✉
PMCID: PMC13639016  PMID: 42834068

Abstract

QatABCD is a widespread prokaryotic anti-phage defense system comprising four protein components, out of which QatC, a QueC-family protein is the signature component. QueC family proteins are nucleoside biosynthesis enzymes involved in the biosynthesis of queuosine, a 7-deazaguanine derivative. Recently, QueC-family proteins were shown to catalyze a deazaguanylation protein-nucleobase conjugation reaction in type IV CBASS antiphage defense. However, the mechanism of QatABCD, even the function of QatC in this system, remains unclear. Here, we demonstrate that QatBCD forms a complex in which QatD is highly flexible. Crystal structures of the QatBC complex in apo and ATP-bound form support a shared role for QueC-family proteins in targeting protein substrates for N-terminal modification as in type IV CBASS. We show that the QatB N-terminal loop, its binding with QatC and QatC catalytic site are essential for QatABCD defense in vivo, suggesting a modification might occur analogous to CBASS. These findings provide structural and functional insights into QueC-family protein in QatABCD system, suggesting the conserved mechanisms and critical roles of QueC-family in prokaryotic immunity.

Subject terms: X-ray crystallography, Bacteriophages, X-ray crystallography

Here the authors determine the structure and mechanism of the QatABCD anti-phage defense system, showing how the QueCfamily protein QatC recognizes QatB to enable bacterial antiviral immunity and highlighting a conserved role for QueC-family proteins in protein modification.

Introduction

7-Deazapurine and its derivatives, as nucleobase analogs, are important in critical cellular functions widely distributed in nature1, such as RNA and DNA modifications and secondary metabolites2–5. Among the 7-deazapurine derivatives, queuosine (Q) is one of the most structurally complex and evolutionarily conserved tRNA modifications2,6. Incorporated at the anticodon loop of tRNAs, Q plays crucial roles in optimizing translational efficiency and fidelity1,2,6–9.

While eukaryotes salvage its precursor Q from diet or microbiota10, prokaryotes can synthesize Q de novo through a multi-enzyme pathway11–15. QueC (7-cyano-7-deazaguanine synthase, renamed from ybaX) is essential in the early stages of the Q biosynthetic pathway, driving the ATP-dependent two-step transformation of CDG (7-caboxy-7-deazaguanine) to intermediate nucleobase preQ₀11,14,16,17. As diverse QueC homologs increasingly identified in antiphage defense systems18–20, the canonical tRNA-modifying enzyme QueC is now recognized as a critical player in the virus-host arms race21. Recent studies on type IV cyclic oligonucleotide-based antiphage signaling systems (CBASS) revealed that the QueC homolog Cap9 was repurposed for nucleobase-protein conjugation in an immunity context. In these systems, Cap9 mediates a distinct modification called NDG (N-terminal 7-amido-7-deazaguanine) on the cyclic dinucleotide synthase CdnD—a process essential for activating antiphage immunity22. Beyond the role in CBASS system, QueC domains are also found in various anti-phage defense operons, such as the QueC homolog QatC within the qatABCD system19.

The QatABCD system exhibits broad-spectrum phage defense capabilities. The system from Escherichia coli NCTC9009 has been identified to provide protection against phages P1, T3, and λ19. Additionally, the system derived from E. coli strain 46-1 demonstrates low-level protection against T4(C), a mutant lacking gene 56 which encodes a cytosine methyltransferase23. Moreover, the qatABCD homologous system in Pseudomonas aeruginosa demonstrates antiphage activity against Casadabanvirus and Pbunavirus genera24. The qatABCD system in prokaryotes consists of four components: qatA is an ATPase, qatB has no recognizable known domain or motifs, qatC has a QueC domain and qatD is a Tat D family DNase19,25. However, the mechanism of QatABCD system remains unclear.

Here, we determine the crystal structure of QatBC complex, and combine this with in vivo analysis and structure-guided biochemical analyses, allowing us to determine the structural and functional basis of the QueC-family protein QatC within the QatABCD anti-phage system. We show that QatC forms a stable complex with QatB, in which the N-terminal loop of QatB inserts into the catalytic center of QatC. Crystal structures of the QatBC complex in apo and ATP-bound states reveal that QatC retains a conserved QueC catalytic core with distinct N- and C-terminal extensions that mediate specific interaction with QatB. Structure-guided mutational analyses further demonstrate that the catalytic residues of QatC, the zinc-binding motif, and the QatB N-terminal tail are indispensable for anti-phage activity. Together, our findings demonstrate that QatC likely acts as a functional analog of Cap9, mediating protein modification within the QatABCD system, and highlight the evolutionary diversification of QueC-like enzymes as a general strategy in prokaryotic immune defense.

Results

QatB interacts with both QatC and QatD

Previous study showed that all the four components are required for the anti-phage activity against λ phage when the QatABCD system from E. coli NCTC9009 (Fig. 1a) is expressed in E. coli BL21 cells19. To elucidate how QatABCD defends against phage infection, first we tested whether the four components interact with each other. To this end, we co-expressed the four components together, with His-tag on one out of the four proteins each time, respectively. Out of the four expression combinations, we found that QatB/C/D could form a complex, for which the best combination is His-tagged QatC but QatB and QatD without tags (Fig. 1b and Supplementary Fig. 1a–c). The SEC-MALS (Size Exclusion Chromatography-Multi-Angle Light Scattering) analysis also indicated that QatB/C/D interacts with each other with a 1:1:1 ratio in the complex (Fig. 1c). Moreover, QatA does not interact with any of the other components (Supplementary Fig. 1d–f). Next, we sought to analyze the interaction relationship of the three proteins in the QatBCD complex. The native gel results showed that complex can be formed either between QatB and QatC, or between QatB and QatD, but not between QatC and QatD (Fig. 1d). Notably, in the SDS-PAGE gel, QatB migrates slower than QatD, while QatB has a smaller molecular weight. This might result from different SDS binding, conformation or other unknown reasons. The two bands of QatBCD might represent alternative conformational states. Furthermore, we confirmed this relationship through gel filtration or SEC-MALS assay (Supplementary Fig. 2a–e). This suggests that QatB might be a central component in binding to QatC and QatD. Taken together, protein complex can be formed between QatBC, QatBD and QatBCD in the system.

Fig. 1. QatB, QatC and QatD form a stable complex.

Fig. 1

Open in a new tab

a Schematic of the qatABCD system from E.coli NCTC9009. b The gel filtration profiles of QatB (blue), QatC (green), QatD (yellow) and the mixture (black) of them are shown. SDS-PAGE gel of each sample tube is displayed. kDa: kilodalton. Source data are provided as a Source Data file. c Static light scattering (SLS) studies of QatBCD complex. The calculated molecular weight of the main peak is shown above the peak. Source data are provided as a Source Data file. d Native-PAGE showed the binding among QatB, QatC, and QatD. The results of the SDS-PAGE for the same samples are displayed on the left. kDa: kilodalton. Source data are provided as a Source Data file.

Crystal structure of QatBC complex

To reveal the structural basis of QatBCD complex, we subjected the purified QatBCD complex for crystallization. However, despite extensive trials, there is only electron density for the QatBC complex in the crystal, but not QatD, suggesting that the QatD component might be flexible in this complex. Crystallization of the QatBD complex was also unsuccessful. Therefore, we finally solved the crystal structure of QatBC complex at 2.09 Å resolution for structural analysis (Supplementary Fig. 3a, Supplementary Information Table 1). In the asymmetric unit, there are two 1:1 QatBC complex connected through the N-terminal β strand of QatC (Supplementary Fig. 4). However, PISA (Protein Interfaces, Surfaces and Assemblies) analysis26 indicated that QatBC is stable as a 1:1 complex, which is also supported by the SEC-MALS analysis of QatBC complex (Supplementary Fig. 2d). In the structure, QatB and QatC form a compact assembly with a large buried area of 4849.4 Å2 (Fig. 2a). In the complex, the QatC protein can be divided into three parts based on its QueC domain in the middle, that is, the C-terminal extension domain, the N-terminal extension domain and the QueC-like domain (Fig. 2b). Typically, the residues 148-370 of QatC constitute a core domain which is highly structurally similar to Q biosynthetic enzyme QueC from Bacillus subtilis (BsQueC) and Cap9 from Rhizobiales in type IV CBASS (Fig. 2c). Sequence alignment of QatC with BsQueC and Cap9 also reveals active sites S156/S161 are highly conserved across these enzymes20 (Supplementary Fig. 5a). The QueC-like domain includes a Rossman-like fold, consisting of five parallel β-strands with eight α-helices surrounded at both sides (Supplementary Fig. 6a). There is a conserved pyrophosphate-binding SXGXDS motif (S156-G158-D160-S161) located in the loop linking β5/α4 and at the top of α4 (Supplementary Fig. 5a). A highly conserved CxxCxxC motif (residues C352, C355, C358) and another cysteine C332 together coordinates a zinc atom (Fig. 2d). Mutation of the four cysteine residues also severely decreased anti-phage activity of the system (Fig. 2e), suggesting the possible important role of these residues in stabilizing the structure. In addition, the N-terminal extension of QatC comprises four parallel β-strands β1–β4 flanked by three α-helices α1–α3. The C-terminal extension of QatC is mainly composed of three α-helices α12–α14, which form interactions with QatB from one side (Fig. 2b and Supplementary Fig. 6a).

Fig. 2. X-ray crystal structure of QatBC complex.

Fig. 2

Open in a new tab

a The structure of QatBC complex is shown in the cartoon and surface model. b Overall structure of QatC. The QueC-like domain is colored green. c Structural superimposition among QatC, BsQueC (PDB: 3BL5) and Cap9 (PDB: 9NTO). Three proteins are displayed as green, pink, and gray, respectively. d Detail diagram of zinc ion binding site of QatC. e Validation of the zinc-coordinating residues of QatC by plaque assay. 4CA represents mutation of the four Cys residues shown in (d) to Ala. Source data are provided as a Source Data file. f A close view of the binding of the QatC loop between α6 and α8.

QatB is a globular all α-helical structure (α1–α10) with a long N-terminal extension loop (Fig. 2a and Supplementary Fig. 5b). No similar entries with high Z scores returned from Dali search using QatB as a query27. All three domains of QatC are involved in interaction with QatB. The most notable feature of the interaction is that the QatB N-terminal loop is inserted into the putative active site of QatC (Fig. 2a), suggesting that QatB might work as a substrate of QatC, like what the CD-NTase CdnD does in Cap922. While both QatB and CdnD inserts its N-terminal loop into the QueC-like domain of QatC and Cap9, respectively, no structural homology exists between QatB and CdnD (Supplementary Fig. 7). In addition, the QatC residues N263–H280 form a long loop between QatC α7 and α8, which plugs into the surface cavity on QatB formed by its α6 and α8 (Fig. 2f). More detailed interactions between QatB and QatC will be discussed below. Together, the structure of QatBC complex displays how they interact with each other and suggests the possible role of QatB as a substrate of QatC.

Catalytic site of QatC

Since QueC enzymes catalyze a two-step ATP-dependent reaction in which CDG (7-carboxy-7-deazaguanine) reacts with ATP as the first step, we moved on to incubate QatBC complex with both CDG and ATP, and solved a 2.66 Å X-ray crystal structure (Supplementary Fig. 3b, c, Supplementary Information Table 1). However, there is no electron density for CDG but only ATP, and there is no variation in the density of the N-terminus of QatB either. This suggests that CDG might not have reacted with ATP in QatC, therefore, we turned to analyze the binding pocket of ATP in QatC. Binding of ATP does not show significant impact on the conformation of the QatBC complex compared to the apo complex, with an RMSD of 0.308 Å among 656 Cα atoms (Fig. 3a). Investigation into the active site of QatC reveals conserved ATP binding and catalytic residues among QatC homologs as well as QueC and Cap9 (Fig. 3b and Supplementary Fig. 5a). Specifically, QatC F235 displays a marked conformational change compared with its apo state, to form π-π stacking with the adenine base of the ATP molecule. Moreover, the pyrophosphate-binding SXGXDS motif (S156-G158-D160-S161) together coordinates the β- and γ-phosphates of ATP in the structure (Fig. 3b). In addition, the ribose group of ATP is also coordinated by a hydrogen bond from carbonyl oxygen atoms of P255 (Fig. 3b). Finally, the α-phosphate group is also stabilized by a hydrogen bond from the N-terminal G2 of QatB.

Fig. 3. ATP-bound state structure of QatBC complex.

Fig. 3

Open in a new tab

a Structural comparison between QatBC in its apo (gray) and ATP-bound form (green and cyan). b Structural alignment of the binding pocket of ATP between QatBC in its apo (gray) and ATP-bound form (green). Hydrogen bonds are shown as yellow dashed lines. c Structural comparison of the binding pocket of NDG among QatC (green), QueC (pink) and Cap9 (gray). The RMSD values are 3.396 Å (QatC-Cap9), 3.954 Å (QatC-QueC) and 2.149 Å (Cap9-QueC). d Validation of the catalytic activity of QatC by plaque assay. SSAA represents QatC mutation S156A/S161A. Source data are provided as a Source Data file.

Next, we investigated the potential QatC substrate binding site, since there is no density for the CDG molecule. In the reported Cap9-CdnD structure, the N-terminus of CdnD is modified by the NDG nucleobase, while there is no density for the CDG molecule either22. Notably, the NDG modification is stabilized by the surrounding residues Y115 and D117, which are also conserved in BsQueC (Y129 and D131), suggesting the importance of these two residues in coordination of CDG and stabilization of the NDG moiety of the product. However in QatC structure, they are replaced by T276 and T278, respectively (Fig. 3c). Moreover, these two Thr residues are also conserved among QatC homologs (Supplementary Fig. 5a), suggesting their conserved function among QatC proteins, as well as the possibility of another molecule distinct from CDG as the substrate of QatC proteins. Mutations of QatC S156 and S161 significantly reduced anti-phage activity, indicating that the catalytic activity of QatC is indispensable for the function of the system (Fig. 3d). To investigate whether QatC is a functional QueC enzyme, we incubated EcQatC with ATP in the presence or absence of CDG and then analyzed the product through High Performance Liquid Chromatography (HPLC). Despite extensive trials, the results showed that EcQatC does not exhibit obvious ATP pyrophosphatase activity either in the presence or absence of CDG (Supplementary Fig. 8). Moreover, no obvious ATP pyrophosphatase activity was observed either when QatC was replaced by QatBC or QatBCD complex (Supplementary Fig. 8). Together these results indicate that QatC catalytic activity is essential for the immune function of the system but its substrate apart from QatB still awaits further studies.

QatB N-terminal tail is essential for anti-phage activity

As mentioned above, the N-terminus of QatB is inserted into the active site of QatC. Like CdnD in the Cap9 active site, the electron density also starts from the second Gly residue of QatB, suggesting that the M1 residue was removed by post-translational proteolysis in bacteria, especially occurring when the second residue carries a small sidechain such as glycine28. Specifically, the N-terminal loop of QatB is stabilized by multiple hydrogen-bond and hydrophobic interactions from the active site cavity of QatC (Fig. 4a). The similar position of the N-terminal loop of QatB and that of CdnD in QatC and Cap9 (Fig. 4b), respectively, suggests that the G2 residue of QatB might also undergo modification by QatC. Notably, the N-terminal four residues “MGTS” of QatB are highly conserved among QatB homologs (Supplementary Fig. 5b), suggesting the importance of the N-terminus of QatB. Consistently, deletions of residues 14–20, mutation of G2 to aspartate or adding an N-terminal His-tag before QatB M1 all severely decreased the anti-phage activity of the Qat system (Fig. 4c). Although intact-protein mass spectrometry did not reveal detectable NDG modification on QatB (Supplementary Fig. 9a, b), LC–MS/MS analysis of chymotrypsin-digested QatB, obtained from the purified QatBCD complex co-expressed, identified an NDG-modified N-terminal peptide (GTSKAY) from QatB (Supplementary Fig. 9c). MS/MS fragmentation further confirmed the identity of the peptide and the modification site, indicating that QatB can undergo N-terminal NDG modification (Supplementary Fig. 9d). The discrepancy between the conclusions drawn from the intact-protein analysis and the peptide-level analysis may arise from differences in the analytical approaches and detection sensitivities of the two methods. This might also reflect that the efficiency of the modification is low under present experimental conditions. Taken together, these results collectively define the essential role of QatB N-terminus in anti-phage activity.

Fig. 4. The N-terminus of QatB is essential for phage resistance.

Fig. 4

Open in a new tab

a Detailed binding between the QatB N-terminus and QatC. Residues involved in binding are shown as sticks. Hydrogen bonds and hydrophobic interactions are shown as yellow and blue dashed lines. b Structural comparison between QatB (cyan)-QatC (green) complex and CdnD (yellow)-Cap9 (gray) complex. c Validation of the function of the N-terminus of QatB by plaque assay. Source data are provided as a Source Data file.

To investigate whether QatBC interaction is essential for anti-phage activity, we analyzed their interface in detail. Notably, a network of hydrophobic and electrostatic interactions is observed at their interface (Fig. 5a–e), in which hydrophobic residues of QatC (A404, P422, F282, L269, P268, and F415) engage in van der Waals interactions with hydrophobic residues of QatB (V218, I214, P147, A6, I208, M200, L138, and A165). Additionally, positively charged residues of QatC (R454 and R412) form salt bridges with negatively charged residues of QatB (D212, D166, and E162). To confirm the roles of these residues in QatBC interaction, we introduced point mutations on these critical interfacial residues and analyzed them using pull-down assay (Fig. 5f). The results showed that all mutants except QatC P268A/L269A display much weaker QatB-QatC binding. The mutation of QatC R412A/K414A/F415A resulted in the complete disappearance of its band, indicating the complete elimination of the binding of QatB and QatC, suggesting that these residues are crucial for the binding of QatB and QatC. Interestingly, this mutation still reduced the anti-phage effects of the system (Fig. 5g), suggesting that some differences might exist in vivo compared to the in vitro binding. Together, the results suggested that these interfacial interactions are indispensable for the anti-phage activity of QatABCD system.

Fig. 5. Interface interactions of QatBC are essential for activity of qatABCD system.

Fig. 5

Open in a new tab

a–e Detailed interactions of QatBC complex are shown. Blue dashed lines represent hydrophobic and red dashed lines represent electrostatic interactions. f Pull-down assay, showing the binding ability between QatB, QatC and their mutants, in which QatB or its mutant was C-terminally His-tagged. kDa: kilodalton. Source data are provided as a Source Data file. g Validation of the functions of the interface residues of QatBC by plaque assay. Source data are provided as a Source Data file.

Insights into QatD in the QatBCD complex

Since QatD cannot be resolved through structural studies of the QatABCD complex, we used AlphaFold3 to predict a QatBCD complex (Supplementary Fig. 10a). In the predicted structure (interface predicted template modeling (ipTM) score of 0.61 and a predicted template modeling (pTM) score of 0.68), interestingly, QatC and the main body of QatB are in the same position as they are in the crystal structure, except that the QatB N-terminal flexible loop mainly interacts with QatD but not QatC, despite low confidence for this loop region (Supplementary Fig. 10a, b). Interestingly, in the structural alignment of the top five predicted structures of QatBCD, QatC and the main body of QatB overlaps well with their solved structure, but the position of QatD varies greatly (Supplementary Fig. 10c). This is consistent with our finding based on the crystal structure that QatD binding is highly flexible in the QatBCD complex. However, in each predicted structure, the QatB N-terminal flexible loop is involved in QatD binding (Supplementary Fig. 10b). To investigate the function of this loop in QatD binding, we generated a QatB Δ2–46 truncated mutant and native gel assay showed that this truncated mutant lost QatD binding (Supplementary Fig. 10d). Interestingly, this QatB truncated mutant still retains QatC binding (Supplementary Fig. 10d), suggesting that the other interfaces other than those involving QatB N-terminal loop play a key role in QatBC binding. The function of QatD in this system and its binding by QatB N-terminus still awaits further study. We also analyzed the interaction interface between QatB and QatD, excluding the N-terminus of QatB. We have investigated these potential interactions by mutating the interface residues (Supplementary Fig. 10e) and testing their effect on QatB-QatD binding. The native gel results showed that mutation of the other interfaces only has a minor role, if any, in disrupting QatBD binding (Supplementary Fig. 10f), suggesting that QatB-QatD interaction relies almost exclusively on the insertion of the N-terminal flexible loop of QatB into QatD. To test whether the nuclease activity of QatD is essential for the anti-phage activity of QatABCD system, we analyzed the residues responsible for the nuclease activity of qatD and mutated it. Nuclease assays showed that QatD exhibits single-stranded DNA (ssDNA) cleavage activity, which was not inhibited by QatBC (Supplementary Fig. 10g). E95A/H133A mutation of QatD reduced the nuclease activity, and also displayed weakened phage defense (Supplementary Fig. 10h), suggesting that this activity is needed for defense. Together, these results suggest that QatD is flexibly bound in the QatBCD by the N-terminus of QatB.

Discussion

In this study, we mainly characterize the QatBC complex in the QatABCD system through structural biology, biochemistry and in vivo studies. As the signature component of this system, QatC is a QueC-domain protein, whose catalytic activity is essential for the anti-phage activity of the system. Together with the recently published study into type IV CBASS22, this work illustrates the role of QueC-family proteins as key catalytic components in prokaryotic defense systems (Supplementary Fig. 11). Moreover, these studies also extend the substrate scope of QueC-family proteins beyond nucleic acids29.

Crystal structures of QatBC complex reveal a series of essential elements in the anti-phage activity of QatABCD, including the catalytic active site of QatC, the Zinc-coordinating residues of QatC, the N-terminal flexible loop of QatB and the interface between QatB and QatC, which have been proved by phage plaque assays30. Interestingly, we attempted to crystallize the QatBCD complex but obtained the QatBC structure. Predicted structures of QatBCD support the notion that QatD is flexibly bound in the complex and suggest that QatB N-terminus is responsible for this binding, which was further confirmed by native gel assays. Previous study indicated that mutation of the putative active site of QatD decreases the anti-phage activity of QatABCD22,23. However, the exact role of QatD in the system needs further study. During the preparation of our manuscript, the structure of QatBC has also been reported by two other studies31,32. Interestingly, one of the studies reported that QatA and QatD in the Pseudomonas aeruginosa qatABCD system are not required for the anti-phage activity31. However, these two genes in the E. coli 46 NCTC9009 system have been confirmed to be required for anti-phage19. This suggests that for systems from different species distinct factors are needed for anti-phage activity. Despite differences in experimental systems and species analyzed, both studies converge on the central role of the QatBC complex and highlight QatC as a critical catalytic component of this defense pathway. These independent structural analyses collectively reinforce the idea that QueC-family enzymes have been broadly repurposed in prokaryotic immune systems.

It remains unknown what exact reaction is catalyzed by QatC. The crystal structure of QatBC suggests that QatB meets the conditions to be a substrate of QatC: its N-terminus inserted into the catalytic center of QatC and in the same context as CdnD in the Cap9 catalytic center. However, despite extensive trials, we could not reconstitute the activity of QatC using CDG, ATP and QatB in vitro. It suggests that QatC either needs a phage trigger to be activated or utilizes a substrate distinct from CDG. Future studies into this will help characterize the mechanism of modification of QatB N-terminus and its role in anti-phage activity. While the complete mechanism of qatABCD-mediated defense remains unsolved, our results illustrate that qatABCD utilizes a mechanism in which QueC-family also mediates protein-ligand conjugation.

Methods

Bacterial strains

The E. coli BL21 (DE3) strains were grown in Lysogeny broth (LB) medium at 37 °C both with aeration at 225 r.p.m. When indicated, ampicillin was used to maintain the pBAD24 plasmid. Gene expression was induced by the addition of 0.2% L-arabinose.

The E. coli BL21 (DE3) strain was used for recombinant protein overexpression and grown in Lysogeny broth (LB) medium. The cells were grown at 37 °C until OD600 nm reached 0.8 and then induced at 18 °C for 12 h.

Protein expression and purification

The qatABCD gene was bought from AddGene. The full-length qatA/qatB/qatC/qatD gene was amplified by PCR and cloned into a modified pET28a vector (kanamycin resistance) in which the expressed protein contains an N-terminal His6-SUMO tag. Specifically, in order to ensure that the N-terminus of QatB does not carry excess amino acid residues after cleavage by Ulp1, we directly linked the codon corresponding to the second glycine residue of QatB to the codon corresponding to the cleavage site of Ulp1. For the Ni-column pull-down assay, qatB was cloned into the pET22b vector (ampicillin resistance) and carried a C-terminal His6 tag, while qatC was cloned into the pRSFDuet vector (kanamycin resistance) and carried an N-terminal His6 tag. The mutants were generated by two-step PCR and were subcloned, overexpressed and purified in the same way as for the WT protein.

To collectively express QatABCD and allow one of the proteins to carry a His6 tag, we cloned qatA and qatB onto the pRSFDuet vector (kanamycin resistance), and qatC and qatD onto the pETDuet vector (ampicillin resistance). We engineered one of the proteins, QatA, QatC or QatD, to carry an N-terminal His6 tag, while the other three proteins did not carry any tag. Specifically, to ensure that QatB carries a C-terminal His6 tag while the other three proteins do not, we cloned qatB onto the pET22b vector (ampicillin resistance), qatA onto the pACYCDuet vector (chloramphenicol resistance), and qatC and qatD onto the pRSFDuet vector (kanamycin resistance).

All of the proteins were expressed in E. coli strain BL21 (DE3) and induced by 0.2 mM IPTG when the cell density reached an OD600 of 0.8. After growth at 18 °C for 12 h, the cells were collected, resuspended in lysis buffer (50 mM Tris-HCl pH 8.0, 300 mM NaCl, 10 mM imidazole and 1 mM PMSF) and lysed by sonication. The cell lysate was centrifuged at 20,000 × g for 50 min at 4 °C to remove cell debris. The supernatant was applied onto a self-packaged Ni-affinity column (2 ml Ni-NTA, Genscript) and contaminant proteins were removed with wash buffer (50 mM Tris pH 8.0, 300 mM NaCl, 30 mM imidazole). The fusion protein was elution with elution buffer (50 mM Tris pH 8.0, 300 mM NaCl, 300 mM imidazole). The protein of QatA/QatB/QatC/QatD with His6–SUMO tag was digested with Ulp1 on the Ni-NTA column at 18 °C for 2 h after removing contaminant proteins with wash buffer. The protein was then eluted with wash buffer. The eluant of protein was concentrated and further purified using a Superdex-200 increase 10/300 GL (GE Healthcare) column equilibrated with a buffer containing 10 mM Tris-HCl pH 8.0, 200 mM NaCl and 5 mM DTT. The purified proteins were analyzed by SDS–PAGE. The protein bound to the column was eluted with a gradient concentration of 10–1000 mM NaCl, and the protein purity was then verified using SDS–PAGE. The fractions containing the target protein were pooled and concentrated.

Gel filtration assay

The QatA, QatB, QatC, QatD purified as described above and the mixture of the two or three proteins involved were subjected to gel filtration analysis (Superdex-75 increase 10/300 GL, GE Healthcare). The mixture of two or three proteins was incubated at a molar ratio of 1:1 or 1:1:1 overnight on ice before the gel filtration analysis in buffer containing 10 mM Tris-HCl pH 8.0, 200 mM NaCl, and 5 mM DTT. The assays were performed with a flow rate of 0.5 mL/min and an injection volume of 1 mL for each run. Samples from relevant fractions were subjected to SDS-PAGE and visualized by Coomassie blue staining.

Crystallization

After size-exclusion chromatography purification, the fractions containing the target protein were pooled and concentrated to 15 mg/mL. Both crystal forms of protein complexes were obtained using the sitting-drop vapor diffusion method at 18 °C by mixing 0.8 μL of protein solution with 0.8 μL of reservoir solution. For the apo form, the reservoir solution contained 2 M ammonium sulfate. For the ATP-bound form, the reservoir solution was composed of 0.1 M ammonium sulfate, 0.3 M sodium formate, 0.1 M Tris pH7.8, 3% w/v γ-PGA (Na+ form, LM), 5% w/v PEG 4000. The protein was pre-incubated with 1 mM ATP, 1 mM CDG and 2 mM MgCl₂ on ice for 1 h prior to crystallization. Before harvesting, the crystals were cryoprotected in the reservoir solution supplemented with 20% (v/v) glycerol and then flash-frozen in liquid nitrogen.

Data collection, structure determination and refinement

All diffraction data were collected at the SSRF beamlines BL02U1 using a wavelength of 0.979 Å and a DECTRIS EIGER2 X 9M detector. Data integration and scaling were performed using Aquarium33. The initial model was solved by molecular replacement with a structure predicted by AlphaFold234 as a template and refined manually using COOT35. All structures were further refined in PHENIX36 with stereochemistry restraints and TLS refinement. The final models were obtained after several rounds of refinement. All structural illustrations were generated using PyMOL (https://pymol.org/). Detailed statistics of data collection and structure refinement are summarized in Supplementary Information Table 1.

DNA digestion assays

The QatD/QatB/QatC (1 μM) and ssDNA (100 nM) were incubated at 37 °C for 30 min in 50 mM Tris pH 8.2, 50 mM KCl and 5% glycerol. 5 mM MgCl2 were added in the marked samples concurrently. The mixture was separated using 14% Urea-PAGE and observed under 475 nm wavelength.

SEC-MALS (Size Exclusion Chromatography-Multi-Angle Light Scattering)

Static light scattering experiments of QatA/QatA+ATP/QatBC/QatBD/QatC+QatD/QatBCD were performed in 10 mM Tris-HCl pH 8.0, 200 mM NaCl, 5 mM DTT with a GE Healthcare Superdex-200 increase 10/300 GL size-exclusion column connected to the Wyatt DAWN HELEOS Laser photometer and Wyatt Optilab T-rEX differential refractometer37. Wyatt ASTRA 7.3.2 software was used for the data analysis.

Native-PAGE assay

For the assessment of native-PAGE for the binding between proteins, each protein was pre-incubated at 20 μM at 18 °C for 20 min. Products of the reaction were analyzed using 8% native polyacrylamide gels and visualized by Coomassie blue staining.

QatC activity assay

For testing the activity of QatC and its mutants, depyrophosphate reactions were performed in a 100 μL reaction volume containing 50 mM Tris pH 8.2, 50 mM KCl, 2 mM MgCl2, 2 mM CDG and 2 mM ATP if needed. After incubation at 37 °C for 60 min, the reactions were terminated by adding acetonitrile with a final concentration of 20%. The products were filtered with a 0.22 μm filter and were subsequently used for HPLC experiments. The HPLC analysis was used an 100 Å AQ C18 column (4.6 × 250 mm; 5 μm; Bonnasil-BS). Chromatographic separation was performed at a flow rate of 1 mL/min under isocratic elution with 97% 20 mM potassium phosphate buffer (pH 6.0) and 3% acetonitrile. Detection was carried out at 254 nm.

Ni-column pull-down assay

20 μM each protein were first incubated together for 30 min at 37 °C, then the mixtures were incubated with Ni-NTA beads for 30 min at 4 °C. The Ni-NTA beads were first washed with buffer containing 50 mM Tris-HCl pH 8.0, 30 mM imidazole, and 300 mM NaCl three times and then, eluted with buffer containing 50 mM Tris-HCl pH 8.0, 300 mM imidazole, and 300 mM NaCl once. Samples of input and pull-down were separated using SDS–PAGE.

Phage plaque assay

For the plaque assay, the qatABCD operon cloned into the pBAD24 vector (ampicillin resistance) with any indicated mutations or deletions and transformed into Escherichia coli strain BL21(DE3). Cultures were incubated while shaking at 37 °C over night in LB medium. Bacterial culture was mixed with 0.8% LB-agar to an OD600 of 0.06 supplemented 0.2% L-arabinose, poured onto the surface of a 2.5% LB-agar plate, and allowed to solidify at room temperature for 1 h. 5 mL of top agar was used for 850 × 15 mm round plates and 10 mL was used for 100 × 100 mm square plates. Ten-fold serial dilutions of the phages were prepared, and 1.2 µL of each dilution was spotted onto the bacterial layer. After the spots had dried, the plates were inverted and incubated at 37 °C for 3 to 6 h before imaging.

Mass spectrometric data collection and spectral deconvolution of intact proteins

QatBCD complex was analyzed by intact protein LC/MS using a Waters SYNAPT G2-Si Q-TOF system equipped with an ACQUITY UPLC BEH C4 1.7 µM column. The mobile phase were 0.1% formic acid water (A) and 0.1% formic acid acetonitrile (B). Peak integration and spectral deconvolution (MaxEnt1, iterated to convergence at 0.5 Da resolution) was performed using MassLynx software. All experiments were independently repeated three times with consistent results.

Mass spectrometric data collection and analysis of QatB N-terminal peptide

To determine the N-terminal modification of QatB, 5 μM QatB, 5 μM QatC, 1 mM ATP and 1 mM CDG were reacted in a buffer containing 25 mM HEPES pH 7.5, 100 mM KCl, 5 mM MgCl2 at 37 °C for 1 h. The complex was subjected to SDS-PAGE and the band corresponding to QatB was cut from the gel, resuspended in PBS and digested with chymotrypsin at 40:1 protein-to-protease ratio for 16 h at 37 °C. Peptides were injected onto a UHPLC 3000 system coupled to a Thermo Scientific Orbitrap Exploris 480 mass spectrometer using a C-18 analytical column (300 Å, 5 μm, Thermo Fisher Scientific, USA). The peptides were eluted with a gradient elution program at a flow rate of 0.300 μL/min. Mobile phase A consisted of 0.1% formic acid, and mobile phase B consisted of 100% acetonitrile and 0.1% formic acid. The mass spectrometer was operated in the data-dependent acquisition (DDA) mode using Xcalibur 4.5.445.18 software. MS1 spectra were acquired at a mass range of 300–1800 m/z with a resolution of 60,000. The spray voltage was set at 2100 V, and the automatic gain control (AGC) target was set to 3e6. For MS2 scans, the top 40 most intense precursor ions were fragmented in the HCD collision cell at a normalized collision energy of 32% using a 0.4 Da isolation window. The dynamic exclusion duration was set to 15 s, and the AGC target was set to 1e5 while the maximum injection time was set to 100 ms29. All experiments were independently repeated three times with consistent results.

Statistics and reproducibility

For Figs. 1b–d, 2e, 3d, 4c, 5f, g; Supplementary Figs. 1, 2, 7, 8, 9d, f–h, each experiment was repeated independently three times with similar results.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Supplementary information

SUPPLEMENTARY INFORMATION (9.1MB, pdf)
41467_2026_77369_MOESM2_ESM.pdf (27.5KB, pdf)

Description of Additional Supplementary Files

Supplementary Data 1 (19.1KB, xlsx)
reporting summary (73.5KB, pdf)
Transparent Peer Review file (1.2MB, pdf)

Source data

Source data 1 (1.3MB, pdf)
Source data 2 (1.4MB, pdf)
Source data 3 (9.1MB, xlsx)

Acknowledgements

We thank the staff at beamlines BL02U1 and BL19U1 of the Shanghai Synchrotron Radiation Facility for their assistance with data collection; the staff at the Tsinghua University Branch of China National Center for Protein Sciences Beijing and S. Fan for providing facility support for X-ray diffraction of the crystal samples.

Author contributions

Y.Z. conceived and supervised the project and designed experiments. Z.G., F.L., N.L., W.X., W.M., D.S. and L.X. purified the proteins and performed in vitro activity analysis and in vivo assays. H.W. and F.L. collected the diffraction data and solved the structures. Y.Z. wrote the original manuscript with the help of all the other authors.

Peer review

Peer review information

Nature Communications thanks Tian-Min Fu, and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. A peer review file is available.

Funding

Y.Z. is supported by National key research and development program of China (2022YFC2104800) and the National Natural Science Foundation of China (32371329).

Data availability

The crystal structures generated in this study have been deposited in the Protein Data Bank (PDB) under accession codes 9WVJ (QatBC) and 9X6C (ATP-bound QatBC) [https://doi.org/10.2210/pdb9X6C/pdb]. The mass spectrometry proteomics data generated in this study have been deposited in the PRIDE38 database under accession codes PXD079562 and PXD079571. All other data generated in this study are provided in the Supplementary Information or Source Data file. Source data are provided with this paper.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

These authors contributed equally: Zirui Gao, Feixue Li, Hao Wang, Xi Liu, Ningning Li.

Supplementary information

The online version contains supplementary material available at https://doi.org/10.1038/s41467-026-77369-4.

References

  • 1.de Crécy-Lagard, V. et al. Biosynthesis and function of 7-deazaguanine derivatives in bacteria and phages. Microbiol. Mol. Biol. Rev.88, e0019923 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Harada, F. & Nishimura, S. Possible anticodon sequences of tRNA His, tRNA Asm, and tRNA Asp from Escherichia coli B. Universal presence of nucleoside Q in the first postion of the anticondons of these transfer ribonucleic acids. Biochemistry11, 301–308 (1972). [DOI] [PubMed] [Google Scholar]
  • 3.Thiaville, J. J. et al. Novel genomic island modifies DNA with 7-deazaguanine derivatives. Proc. Natl. Acad. Sci. USA113, E1452–E1459 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Hutinet, G. et al. 7-Deazaguanine modifications protect phage DNA from host restriction systems. Nat. Commun.10, 5442 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Tolman, R. L., Robins, R. K. & Townsend, L. B. Pyrrolo[2,3-d]pyrimidine nucleoside antibiotics. Total synthesis and structure of toyocamycin, unamycin B, vengicide, antibiotic E-212, and Sangivamycin (BA-90912). J. Am. Chem. Soc.90, 524–526 (1968). [DOI] [PubMed] [Google Scholar]
  • 6.Kasai, H., Kuchino, Y., Nihei, K. & Nishimura, S. Distribution of the modified nucleoside Q and its derivatives in animal and plant transfer RNA’s. Nucleic Acids Res.2, 1931–1939 (1975). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Meier, F., Suter, B., Grosjean, H., Keith, G. & Kubli, E. Queuosine modification of the wobble base in tRNAHis influences ‘in vivo’ decoding properties. EMBO J.4, 823–827 (1985). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Tuorto, F. et al. Queuosine-modified tRNAs confer nutritional control of protein translation. EMBO J.37, e99777 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Müller, M. et al. Queuine links translational control in eukaryotes to a micronutrient from bacteria. Nucleic Acids Res.47, 3711–3727 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Reyniers, J. P., Pleasants, J. R., Wostmann, B. S., Katze, J. R. & Farkas, W. R. Administration of exogenous queuine is essential for the biosynthesis of the queuosine-containing transfer RNAs in the mouse. J. Biol. Chem.256, 11591–11594 (1981). [PubMed] [Google Scholar]
  • 11.Reader, J. S., Metzgar, D., Schimmel, P. & de Crécy-Lagard, V. Identification of four genes necessary for biosynthesis of the modified nucleoside queuosine. J. Biol. Chem.279, 6280–6285 (2004). [DOI] [PubMed] [Google Scholar]
  • 12.Van Lanen, S. G. et al. From cyclohydrolase to oxidoreductase: discovery of nitrile reductase activity in a common fold. Proc. Natl. Acad. Sci.102, 4264–4269 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Phillips, G. et al. Biosynthesis of 7-deazaguanosine-modified tRNA nucleosides: a new role for GTP cyclohydrolase I. J. Bacteriol.190, 7876–7884 (2008). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.McCarty, R. M., Somogyi, A., Lin, G., Jacobsen, N. E. & Bandarian, V. The deazapurine biosynthetic pathway revealed: in vitro enzymatic synthesis of PreQ(0) from guanosine 5’-triphosphate in four steps. Biochemistry48, 3847–3852 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Gaur, R. & Varshney, U. Genetic analysis identifies a function for the queC (ybaX) gene product at an initial step in the queuosine biosynthetic pathway in Escherichia coli. J. Bacteriol.187, 6893–6901 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Winkler, M., Dokulil, K., Weber, H., Pavkov-Keller, T. & Wilding, B. The nitrile-forming enzyme 7-cyano-7-deazaguanine synthase from geobacillus kaustophilus: a reverse nitrilase? Chembiochem16, 2373–2378 (2015). [DOI] [PubMed] [Google Scholar]
  • 17.Cicmil, N. & Huang, R. H. Crystal structure of QueC from Bacillus subtilis: an enzyme involved in preQ1 biosynthesis. Proteins Struct. Funct. Bioinform.72, 1084–1088 (2008). [DOI] [PubMed] [Google Scholar]
  • 18.Millman, A., Melamed, S., Amitai, G. & Sorek, R. Diversity and classification of cyclic-oligonucleotide-based anti-phage signalling systems. Nat. Microbiol.5, 1608–1615 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Gao, L. et al. Diverse enzymatic activities mediate antiviral immunity in prokaryotes. Science369, 1077–1084 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Hobbs, S. J. & Kranzusch, P. J. Nucleotide immune signaling in CBASS, Pycsar, Thoeris, and CRISPR antiphage defense. Annu. Rev. Microbiol.78, 255–276 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Díaz-Rullo, J. & González-Pastor, J. E. tRNA queuosine modification is involved in biofilm formation and virulence in bacteria. Nucleic Acids Res.51, 9821–9837 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Wassarman, D. R. et al. Deazaguanylation is a nucleobase-protein conjugation required for type IV CBASS immunity. Science389, 1347–1352 (2025). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Wang, S. et al. Landscape of new nuclease-containing antiphage systems in Escherichia coli and the counterdefense roles of bacteriophage T4 genome modifications. J. Virol.97, e0059923 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Costa, A. R. et al. Accumulation of defense systems in phage-resistant strains of Pseudomonas aeruginosa. Sci. Adv.10, eadj0341 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Wexler, M. et al. TatD is a cytoplasmic protein with DNase activity. No requirement for TatD family proteins in sec-independent protein export. J. Biol. Chem.275, 16717–16722 (2000). [DOI] [PubMed] [Google Scholar]
  • 26.Krissinel, E. & Henrick, K. Inference of macromolecular assemblies from crystalline state. J. Mol. Biol.372, 774–797 (2007). [DOI] [PubMed] [Google Scholar]
  • 27.Holm, L., Laiho, A., Toronen, P. & Salgado, M. DALI shines a light on remote homologs: one hundred discoveries. Protein Sci.32, e4519 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Wingfield, P. T. N-terminal methionine processing. Curr. Protoc. Protein Sci.88, 6.14.1–6.14.3 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Li, Q. et al. Epoxyqueuosine reductase QueH in the biosynthetic pathway to tRNA queuosine is a unique metalloenzyme. Biochemistry60, 3152–3161 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Aksnes, H., Hole, K. & Arnesen, T. Molecular, cellular, and physiological significance of N-terminal acetylation. Int. Rev. Cell Mol. Biol.316, 267–305 (2015). [DOI] [PubMed] [Google Scholar]
  • 31.Gao, A., Wassarman, D. R. & Kranzusch, P. J. Structural basis of QueC-family protein function in qatABCD anti-phage defense. Nat. Commun.17, 5420 (2026). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Oh, H. & Bae, E. Structural investigation of QatB and QatC proteins in QatABCD anti-phage defense. Nat. Commun.17, 6378 (2026). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Yu, F. et al. Aquarium: an automatic data-processing and experiment information management system for biological macromolecular crystallography beamlines. J. Appl. Crystallogr.52, 472–477 (2019). [Google Scholar]
  • 34.Jumper, J. et al. Highly accurate protein structure prediction with AlphaFold. Nature596, 583–589 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Emsley, P., Lohkamp, B., Scott, W. G. & Cowtan, K. Features and development of Coot. Acta Crystallogr. D Biol. Crystallogr.66, 486–501 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Adams, P. D. et al. PHENIX: building new software for automated crystallographic structure determination. Acta Crystallogr. D Biol. Crystallogr.58, 1948–1954 (2002). [DOI] [PubMed] [Google Scholar]
  • 37.Some, D., Amartely, H., Tsadok, A. & Lebendiker, M. Characterization of proteins by size-exclusion chromatography coupled to multi-angle light scattering (SEC-MALS). J. Vis. Exp. 148, e59615 (2019). [DOI] [PubMed]
  • 38.Perez-Riverol, Y. et al. The PRIDE database at 20 years: 2025 update. Nucleic Acids Res.53, D543–D553 (2025). [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

SUPPLEMENTARY INFORMATION (9.1MB, pdf)
41467_2026_77369_MOESM2_ESM.pdf (27.5KB, pdf)

Description of Additional Supplementary Files

Supplementary Data 1 (19.1KB, xlsx)
reporting summary (73.5KB, pdf)
Transparent Peer Review file (1.2MB, pdf)
Source data 1 (1.3MB, pdf)
Source data 2 (1.4MB, pdf)
Source data 3 (9.1MB, xlsx)

Data Availability Statement

The crystal structures generated in this study have been deposited in the Protein Data Bank (PDB) under accession codes 9WVJ (QatBC) and 9X6C (ATP-bound QatBC) [https://doi.org/10.2210/pdb9X6C/pdb]. The mass spectrometry proteomics data generated in this study have been deposited in the PRIDE38 database under accession codes PXD079562 and PXD079571. All other data generated in this study are provided in the Supplementary Information or Source Data file. Source data are provided with this paper.

Articles from Nature Communications are provided here courtesy of Nature Publishing Group

RESOURCES