ATfaRel2 is a compact, well-structured antitoxin

We determined the crystal structure of ATfaRel2 (residues 1–73) to 1.2-Å resolution (Fig. 1a,b and Supplementary Table 1). The antitoxin protein is monomeric in the crystal (Supplementary Fig. 2a), in good agreement with molecular weight estimates by size-exclusion chromatography (SEC) (Supplementary Fig. 2b and Supplementary Table 2). At the core of its compact, well-folded structure is an antiparallel β-sheet (β-strands β1–β3), with β2 and β3 connected by the central α-helix α1. The C-terminal extension provides an additional β-strand to the β-sheet, β4, which folds parallel to β2, as well as a second α-helix, α2, which forms a part of the protein’s hydrophobic core (Fig. 1b). Despite the lack of sequence similarity, ATfaRel2 is structurally similar to pZFDCapRel, with the two proteins superimposing with a root-mean-square deviation (r.m.s.d.) of 2.8 Å, and they both contain conserved tyrosine residues at the C terminus (Fig. 1c).

Fig. 1: Structure of the ATfaRel2 antitoxin.

a, Cartoon representation of Coprobacillus sp. D7 ATfaRel2 structure. The pZFD core is colored light blue, the C-terminal extension elements are colored salmon and the 51YXXY54 motif is highlighted in dark blue. b, Topology representation of ATfaRel2, colored as in a. c, Superposition of the experimental ATfaRel2 structure (colored as in a) onto the AlphaFold2-predicted structure of the CapRelSJ46 antitoxin domain in the neutralizing state. pZFDCapRel is colored light green and the N-terminal extension that is part of Anchor1 is colored ocher, with the conformational switch region that contains the YXXY neutralization motif colored dark green and labeled in the figure. d, Topology representation of the CapRelSJ46 antitoxin in the neutralizing state, highlighting the circular permutation of ATfaRel2. e, Toxicity neutralization assays probing the 51YXXY54 motif of ATfaRel2 and its scaffolding structure from β1. Serial dilutions of E. coli strains expressing WT FaRel2 alone or coexpressed with ATfaRel2 (WT or V43A, I45A, A50M, Y51A and Y54A variants) were plated on solid LB medium and scored after 16 h at 37 °C. f,g, Binding of WT (f) and Y54A-substituted (g) ATfaRel2 to Y128F-substituted catalytically compromised FaRel2, monitored by ITC.

Interestingly, the predicted topologies of pZFDCapRel in the neutralizing state and ATfaRel2 are similar except for the order of structural elements at the termini. The C-terminal β/α extension of ATfaRel2 is structurally similar to the N-terminal β/α region of pZFDCapRel that connects the antitoxin with toxSYNTHCapRel via Anchor1 (Fig. 1c and Supplementary Fig. 2c–e). This is suggestive of circular permutation (compare Fig. 1b to Fig. 1d). However, as their similarities extend only to the topology level, these two small β/α elements may have evolved independently.

Changes in topology in the antitoxin elements likely reflect the in cis versus in trans neutralization of different toxSASs. Presumably, ATfaRel2 fully dissociates from FaRel2 when the toxin becomes active. Conversely, in the fused CapRels, dissociation of pZFDCapRel is not possible and activation is believed to be mediated by a conformational change10 (Supplementary Fig. 2d,e). Despite these topological variations, both domains retain strong structural similarities, including the YXXY sequence motif linked to toxSYNTH inhibition in CapRelSJ46. This suggests that ATfaRel2 and pZFDCapRel neutralize toxSYNTH domains by a common mechanism.

The YXXY motif is crucial for toxSYNTH inhibition

In CapRelSJ46, the conserved YXXY motif is located in the switch region of the pZFD and is predicted to lock the enzyme in a catalytically inactive state with both Y residues blocking the donor pyrophosphate ATP-binding site of toxSYNTH (Supplementary Fig. 2c–e). Substitutions directly or indirectly targeting this motif render CapRelSJ46 constitutively active10. In the crystal structure of the catalytically active state of CapRelSJ46, the YXXY motif assumes an extended conformation10. Conversely, in the AlphaFold2-generated neutralized state model, the YXXY motif is predicted to fold into a 310-helix that locks into the donor site (Supplementary Fig. 2d,e).

The structure of the free ATfaRel2 reveals that the 51YXXY54 motif indeed folds into a 310-helix structure that was predicted to be key for toxSAS neutralization, scaffolded by β-strands β1 and β3 (Fig. 1a and Supplementary Fig. 1b). This suggests that, even in the absence of the toxin, ATfaRel2 is primed for efficient FaRel2 neutralization. Guided by the structural similarities between ATfaRel2 and pZFDCapRel, we probed the role of the individual residues of the 51YXXY54 motif in the neutralization of FaRel2 toxicity in vivo (Fig. 1e). While Y54A substitution fully ablates the neutralizing activity of ATfaRel2, I45A and Y51A result in modest but clearly detectable defects in FaRel2 neutralization. Our isothermal titration calorimetry (ITC) assays lend further support to the in vivo neutralization results. To overcome the challenges of producing an otherwise highly toxic FaRel2, we followed a well-established substitution strategy for RSH enzymes10,12,13 and we used a catalytically impaired Y128F-substituted variant, FaRel2-Y128F. The substitution interferes with the accommodation of the acceptor nucleotide in the active site but is located far from the predicted pZFDCapRel–toxSYNTH interface. We characterized complex formation between FaRel2-Y128F and wild-type (WT) ATfaRel2 or variants carrying substitutions in the 51YXXY54 motif (Fig. 1f,g and Supplementary Table 3). V43A and A50M substitutions had a negligible effect on complex stability. These mutant antitoxins bound FaRel2-Y128F with KD values of 78.8 nM and 38.0 nM, respectively, compared with a KD value of 35.2 nM in the case of WT ATfaRel2. The destabilizing effect was more pronounced in the case of I45A and Y51A variants. While these substitutions resulted in 25-fold and 43-fold reductions in affinity, respectively, the remaining affinity was still sufficient for partial neutralization in vivo (Fig. 1e). Lastly, the Y54A variant, which was unable to neutralize FaRel2 in vivo, had a 250-fold lower affinity to the toxin than WT ATfaRel2.

FaRel2 binds ATP in a partially folded state

Next, we determined the structure of the catalytically impaired FaRel2-Y128F bound to the non-hydrolyzable ATP analog adenosine-5′-[(α,β)-methyleno]triphosphate (APCPP) at 2.6-Å resolution (Fig. 2a). The toxSYNTH domain of FaRel2 shares the overall topology of other nucleotide pyrophophotransferases, retaining a presumably ancestral fold composed of a central five-stranded β-sheet framed by two α-helices, α3 and α6 (refs. 13,14) (Fig. 2b). The core toxSYNTH domain is very similar to that of the (pp)pApp alarmone synthetase Tas1 (Protein Data Bank (PDB) 6OX6) and the tRNA-pyrophosphokinase CapRelSJ46 (PDB 7ZTB), superimposing with r.m.s.d. values of 1.0 Å and 0.8 Å, respectively (Supplementary Fig. 3a,b).

Fig. 2: Structure of FaRel2 in complex with APCPP.

a, The structure of Coprobacillus sp. D7 FaRel2 in complex with APCPP bound to the pyrophosphate donor site. The adenine base is coordinated by conserved residues R64, R95 and Q147, as well as residue G93, while E145 provides the specificity for adenosine nucleotides. R64 provides additional van der Waals contacts with the ribose moiety. The 5′ triphosphate moiety is stabilized by R64, K66, S70, K74 and R77, while the Mg2+ ion is held in place by the catalytic D90. b, Topology diagram of FaRel2, highlighting the N-terminal region involved in tRNA recognition (yellow), the toxSYNTH domain (dark blue, with the catalytic region outlined by a dashed line) and the disordered C-terminal region (dark magenta). The acceptor site is highlighted by the G-loop (β3–β4 loop) in red and the donor site involving β1, β2 and α3 is shaded in pink. c, Superposition of the FaRel2:APCPP complex on the structure of S. aureus RelP in complex with APCPP (PDB 6EWZ; turquoise). The ordered N terminus of FaRel2 that is lacking in RelP is highlighted in yellow. d, Superposition of the complexes of FaRel2:APCPP (dark blue), RelP:APCPP (turquoise) and Thermus thermophilus RelNTD:AMP (dark orange; PDB 6S2U) illustrates the conserved mode of pyrophosphage donor coordination in RSH pyrophosphotransferases.

The catalytic core of RSH enzymes is typically decorated by regulatory elements that exert allosteric control on the enzymatic domains12,15. The FaRel2 toxSYNTH domain has a well-resolved N-terminal α-helical extension comprising helices α1 and α2, with residues K28 and R29 from the α2–α3 loop being crucial for tRNA binding7 (Supplementary Fig. 1c). While the N terminus is unresolved (Fig. 2c) in the non-toxic (p)ppGpp alarmone synthetases RelP15,16 and RelQ17, the N terminus folds back toward the toxSYNTH core in FaRel2 and intercalates α1 and α2 between α3 and α5, providing an anchor point for tRNAs near the active site G-loop (Fig. 2a,c). The C-terminal region of SYNTH and toxSYNTH domains of pyrophosphotransferases have varied architectures composed predominantly of small α-helical domains6,10,13,14,18,19; the C-terminal bundle of four α-helices of SASs RelQ and RelP, which synthesizes (p)ppGpp, acts as an oligomerization interface15,17, toxSYNTH of the monomeric Tas1 is followed by a small α-helical domain6 and the SYNTH domain of long RHSs is followed by a flexible core domain with a high α-helical propensity18. In the case of FaRel2 complexed with APCPP, the C terminus is disordered and not visible in the electron density (Fig. 2d).

Despite these differences, the FaRel2-bound APCPP superimposes remarkably well with the APCPP bound in the donor site of other (p)ppGpp-synthesizing RSHs SAS RelP and RelQ, as well as the long RSH Rel13,15,17 (Fig. 2d). As in the other alarmone synthetases, the adenosine group stacks with the conserved R64 and R95 from β1 and β2, with β5 E145 hydrogen bonding the adenosine NH2 group (Supplementary Fig. 1c). The R64 residue has a crucial role in providing van der Waals contacts to accommodate the ribose while directly coordinating the 5′ α and β phosphates together with K66. The strongly basic α4 that follows β1 further stabilizes the triphosphates through S70, K74 and R77 (Fig. 2a). At the pyrophosphate acceptor side of the active site, the conformation of the G-loop and the orientation of the base-coordinating F128 (Y128 in the WT FaRel2) deviate from what was observed in precatalytic and postcatalytic complexes of (p)ppGpp synthetases15,17 (Supplementary Fig. 3c,d). It is tempting to speculate that, while the ground state of FaRel2 is primed to bind ATP, efficient tRNA binding would likely involve further conformational rearrangements, such as folding of the C terminus and alignment of the active site.

ATfaRel2 binds FaRel2 by β-sheet extension

To uncover the mechanism of toxin neutralization, we determined the structure of the ATfaRel2:FaRel2 complex. The structure reveals a ATfaRel22:FaRel22 heterotetrametric arrangement (Fig. 3a–c). The 2:2 stoichiometry was confirmed in solution by SEC (62 kDa versus the theoretical 68 kDa; Fig. 3d) and was consistent with the ITC measurements (Fig. 1f).

Fig. 3: Structure of the ATfaRel2:FaRel2 complex.

a, Structure of the heterotetrametic ATfaRel22:FaRel22 complex. Antitoxin units are colored light blue and pink, while the toxin units are colored dark blue and gold. The C-terminal region of FaRel2, disordered in the FaRel2:APCPP structure, in the TA complex folds into a small α-helical subdomain. b, Stabilization of the ATfaRel2:FaRel2 heterodimer by the primary complex interface. The ATP-binding catalytic sites have key structural roles in the interface. c, The secondary interface ATfaRel2:FaRel2 is formed at the two-fold symmetry axis of the complex. The interface comprises (1) cross-coordination of R47 from each antitoxin unit; (2) formation of an alternative hydrophobic interface between ATfaRel2 and FaRel2 (that is, different from the YXXY motif); and (3) interactions between the C-terminal α-helical regions of the two FaRel2 units. d, Analytical SEC of ATfaRel22:FaRel2-Y128F2 (dark-blue trace), ATfaRel2-R47A2:FaRel2-Y128F2 (lilac), FaRel2-Y128F (yellow) and ATfaRel2 (red). e, Probing of the secondary FaRel2:ATfaRel2 interface through toxicity neutralization assays. Serial dilutions of E. coli strains expressing FaRel2 alone or coexpressed with ATfaRel2 (WT or R47A, Y57A and Y59A variants) were plated on solid LB medium and scored after 16 h at 37 °C. f,g, Binding of R47A-substituted (f) and F59A-substituted (g) ATfaRel2 to Y128F-substituted FaRel2, monitored by ITC.

The C-terminal region of FaRel2, disordered in the FaRel2:APCPP complex, folds into an α-helical subdomain upon complex formation and provides a dimerization interface mediated only by toxin–toxin interactions. By contrast, in the bound state, the conformation of ATfaRel2 matches that of the free antitoxin, with both structures superimposing with an r.m.s.d. of 0.4 Å (Supplementary Fig. 3e). The primary interface between ATfaRel2 and FaRel2 has an area of 1,190.0 Å2, with ATfaRel2 sterically blocking access of the ATP substrate to the pyrophosphate donor binding site of the toxSYNTH domain (Fig. 3b). Through this large interface, ATfaRel2 contacts several key functional regions of FaRel2: (1) the long N-terminal α-helix α3 (a structural element that is often involved in allosteric crosstalk in many RSH enzymes and interacts with α1 from ATfaRel2); (2) the basic α-helix α4 (involved in the stabilization of the ATP triphosphate group, which is coordinated by ATfaRel2 through β3 and the β3–α1 loop); (3) the central β-sheet (through β1, β2 and β5, which harbor the catalytic center and adenine coordinating residues); and (4) the C-terminal cap of the α-helix α6 (which is disordered in the FaRel2:APCPP complex and folds into an α-helical dimerization region when bound to ATfaRel2). The only notable exception was the predicted tRNA recognition site that remains solvent accessible (Supplementary Fig. 3f).

The center of the neutralization interface is formed by the antiparallel β-strand interaction between FaRel2 β1 and ATfaRel2 β3 that connects the β-sheets of both proteins, extending the core of the complex. Residues I42–T46 of β3 form multiple van der Waals contacts with FaRel2 β1, further stabilizing the complex. Residues V43 and I45 serve as a scaffold, orienting the YXXY 310-helix that anchors ATfaRel2 (Fig. 3b). As predicted for pZFDCapRel (ref. 10), this hydrophobic tether projects Y51 and Y54 into the ATP-binding site through a π-stacking arrangement with R64 and R95, which precludes adenine coordination to the pyrophosphate donor site of FaRel2. These results suggest that, while this mechanism of neutralization is likely the same at the structural level to that proposed for CapRelSJ46, the energetics of neutralization are certainly different, with the dynamic association of pZFDCapRel regulating toxSYNTHCapRel in cis, contrasting with the stable in trans neutralization in the ATfaRel2:FaRel2 complex. These differences could have important implications for triggering these systems.

Dimerization enhances toxin neutralization by ATfaRel2

Compared with pZFDCapRel, ATfaRel2 is considerably more tolerant to substitutions in the toxin-binding interface (compare Fig. 1e with Extended Data Fig. 3j in ref. 10). The structure of the ATfaRel22:FaRel22 complex reveals that ATfaRel2 engages the neighboring FaRel2 in the heterotetramer through a secondary interface that is half the size of the primary one (~550.0 Å2 versus 1,190.0 Å2), thus effectively crosslinking the complex (Supplementary Fig. 4a–c). We hypothesized that the stable oligomeric nature of ATfaRel22:FaRel22 compensates for the lack of TA colocalization enforced in the monomeric CapRelSJ46 through fusion of the toxin and antitoxin domains into one polypeptide. On this basis, substitutions disrupting the ATfaRel22:FaRel22 oligomerization interface (and not affecting the primary neutralization interface at the active site) would compromise the TA recognition.

To test this hypothesis, we subjected the secondary interface to single-residue substitutions and assessed the complex stability in vivo through toxicity neutralization assays (Fig. 3e). The three targeted residues R47A, Y57A and F59A of ATfaRel2 are all distant from the main contact interface that blocks access of ATP to the active site (Fig. 3c). R47 is located on the two-fold symmetry axis of the complex and the side chains of R47 from each ATfaRel2 interlock through π–π interactions. Y57 and F59 are part of a small hydrophobic core that defines the secondary oligomerization interface. The R47A substitution resulted in a modest defect, whereas Y57A and F59A compromised the neutralization severely (Fig. 3d).

The direct interrogation of these interactions by ITC was in good agreement with the in vivo data. The R47A substitution efficiently perturbs the secondary interface and decouples the highly cooperative tetramer formation observed in the WT protein (Fig. 3f and Supplementary Table 3). The first high-affinity binding event (KD = 35 nM with a stoichiometry of 0.4) followed a lower-affinity recognition event (KD = 750 nM with a stoichiometry of 0.8). This likely represents the initial neutralization of FaRel2 (with a 2:1 TA ratio) followed by the formation of a less stable 2:2 tetramer. The impact of F59A on the affinity was even stronger, with a 60-fold decrease in affinity and a confirmed 1:1 binding molar ratio, indicating an interaction mediated by only the primary interface (Fig. 3g and Supplementary Table 3). These results were consistent with SEC experiments that revealed a decrease in size of the TA complex from the estimated 70.8 kDa of the WT (consistent with a 2:2 A:T stoichiometry) to 51.3 kDa suggestive of a ATfaRel2-R47A:FaRel2-Y128F2 complex (1:2 A:T stoichiometry) (Fig. 3d and Supplementary Table 2). The observation of a stable ATfaRel2-R47A:FaRel2-Y128F2 complex in SEC matched the high-affinity interaction observed by ITC with ATfaRel2-R47A (Supplementary Table 2). It is, thus, likely that the C-terminal region of FaRel2 that folds upon TA complex formation and provides a large FaRel2:FaRel2 interface in the complex is still capable of partially stabilizing the oligomer against the effect of mild substitutions such as R47A but not against F59A, which had a major effect on complex formation.

FaRel2 C terminus stabilizes the heterotetrameric complex

Upon formation of the ATfaRel22:FaRel22 complex, the C-terminal regions of the two individual FaRel2 toxin polypeptides fold into a dimerization region with four α-helices that contributes 670 Å2 to the TA interface (Supplementary Fig. 4d). This folding upon binding interaction likely has an important role in the overall stability of the heterotetramer. Guided by the structure of FaRel2 bound to APCPP, we constructed a truncated version of FaRel2 (FaRel2-Δ166–206) lacking the C-terminal disordered part. FaRel2-Δ166–206 interacted with ATfaRel2 with a KD of 3.3 μM, as measured by ITC (Supplementary Fig. 4e). This ~90-fold drop in affinity of FaRel2-Δ166–206 for ATfaRel2 underscores the strong contribution of oligomerization to the overall energetics of complex formation. In the case of the WT TA complex, the binding was both entropically and enthalpically driven. The entropic penalty from the folding of the FaRel2 C terminus was likely compensated for by the configurational entropy associated with the large hydrophobic surface buried upon binding (Supplementary Table 3). Together with the strong enthalpic component that accompanied the oligomerization through the C-terminal α-helical region, this resulted in very stable heterotetramerization. Because the FaRel2-Δ166–206 truncation removes the enthalpic contribution of the C terminus, binding to ATfaRel2 was, as expected, predominantly entropically driven, resulting in a less stable complex (Supplementary Table 3). Collectively, these results suggest that, while the main TA interface drives toxin neutralization, the oligomerization further stabilizes the interaction between ATfaRel2 and the toxSYNTH domain of FaRel2. Additional contacts of the main interface YXXY motif with the folded C-terminal α-helical FaRel2:FaRel2 interface link the oligomerization with toxin neutralization, hinting at a potential allosteric path for activation and toxin release.

tRNA-pyrophosphorylating toxins specifically bind tRNA

Given the low concentrations toxSAS toxins are typically found in the cell (below the detection levels of current techniques)20,21, tRNA-phosphorylating activity would likely depend on a strong and specific association with tRNAs. We used ITC to examine the tRNA binding capacity of a representative toxSAS and housekeeping SASs: tRNA-phosphorylating Coprobacillus sp. D7 FaRel2 and Mycobacterium phage Phrann PhRel, (pp)pApp-synthesizing C. marina FaRel and (p)ppGpp-synthesizing Staphylococcus aureus SAS RelQ. FaRel2 and PhRel bound deacylated initiator tRNAifMet with similar affinities (KD values of 483 nM and 825 nM, respectively; Fig. 4a,b and Supplementary Table 3). (pp)pApp synthetase FaRel had a 37-fold lower affinity to tRNAifMet (KD value of 17.8 µM). No tRNAifMet binding was observed for (p)ppGpp-producing Enterococcus faecalis SAS RelQ (Fig. 4c,d and Supplementary Table 3). Lastly, E. faecalis RelQ was shown to interact with a short single-stranded model mRNA(MF) coding for MF dipeptide22. Our ITC experiments demonstrated that S. aureus RelQ similarly bound mRNA(MF) with submicromolar affinity (KD value of 922 nM) (Fig. 4e and Supplementary Table 3).

Fig. 4: Energetic and structural basis of substrate recognition by toxSASs and non-toxSASs.

a–d, Binding of FaRel2-Y128F (a), PhRel2-Y143F (b), FaRel-Y175F (c) and WT RelQSa (d) to deacylated initiator tRNAifMet, monitored by ITC. e, Binding of mRNA(MF) to WT RelQSa, monitored by ITC. f,g, Binding of APCPP to FaRel2-Y128F (f) and ATfaRel22:FaRel2-Y128F2 complex (g), monitored by ITC. h, Binding of tRNAifMet to the ATfaRel22:FaRel2-Y128F2 complex, monitored by ITC. i, Structure of ATfaRel22:FaRel2-Y128F2 bound to APCPP (green). The unbiased mFo-DFc electron density map corresponding to the bound APCPP is shown in gray. j, Details of the coordination of APCPP (green) in the acceptor site when bound to ATfaRel22:FaRel2-Y128F2. The adenosine base is coordinated by the G-loop F128 and the 5′ β and γ phosphates extend to the basic patch of α4. There, they bind in a reverse orientation compared with APCPP in the FaRel2:APCPP complex, underscoring that the nucleotide is bound in a state incompatible with phosphate transfer. For comparison, the APCPP in the orientation observed in the complex with FaRel2 is shown in light yellow. Active site residues of FaRel2 are labeled in black and the residues from the YXXY motif of ATfaRel2 are labeled in light blue. k, FaRel2-Y128F:APCPP complex with APCPP placed in the donor site in a catalytically compatible orientation, presented in the same pose as i. In the absence of ATfaRel2, α7 and α8 are not visible in the electron density.

ATfaRel2 interferes with APCPP binding but not tRNA recognition

While association with ATfaRel2 decreased the affinity (KD) to APCPP ~20-fold, from 2.1 to 41.3 μM (Fig. 4f,g and Supplementary Table 3), the affinity to deacylated tRNAifMet was virtually the same for the free toxin and inactive TA complex (KD value of 483 nM versus 460 nM) (Fig. 4h and Supplementary Table 3). In the cases of both free monomeric FaRel2 and the heterotetrametric ATfaRel2:FaRel2 complex, the tRNA binding had 1:1 stoichiometry with respect to FaRel2.

To further investigate the effect of AtfaRel2 binding on the interaction of FaRel2 with ATP, we determined the structure of the ATfaRel2:FaRel2 complex bound to APCPP (Fig. 4i). The high protein concentration intrinsic of the crystal lattice combined with high nucleotide concentration used for soaking facilitated the binding of APCPP to a partially blocked active site. As predicted from the structure of ATfaRel2:FaRel2, the coordination sites for the adenine, ribose and α phosphate groups at the pyrophosphate donor site are blocked by ATfaRel2. Thus, APCPP is bound in the pyrophosphate acceptor site in a conformation incompatible with pyrophosphate transfer (Fig. 4j). The adenine base is coordinated by Y128 and R95, resembling the expected coordination of the terminal adenine of the CCA tRNA moiety. The β and γ phosphates further anchor the nucleotide; however, they are observed in a reversed orientation compared with the FaRel2:APCPP complex (Fig. 4j–k).

It is instructive to compare the local charge distributions in the active sites of tRNA-modifying FaRel2 to those of (pp)pGpp and (p)ppApp alarmone synthetases Rel13, RelP15 and Tas1 (ref. 6). All alarmone synthetases have a large positive patch that accommodates diphosphate and triphosphate nucleotide substrates, located on the acceptor site close to the conserved Y residue that interacts with the acceptor base (Supplementary Fig. 4f–h). This positive patch is considerably smaller in FaRel2 (Supplementary Fig. 4i), which explains the misorientation of the β and γ phosphates of APCPP bound in the acceptor site of FaRel2 and the lack of alarmone synthetase activity of the tRNA-targeting toxSAS.

Collectively, our results demonstrate that FaRel2 is neutralized by the ATfaRel2 antitoxin by compromising the accommodation of ATP in the toxSYNTH active site without affecting the interaction with uncharged tRNAs. This suggests that the ATfaRel22:FaRel22:tRNA2 ternary complex could be preformed in the cell, with the toxSAS neutralized in the complex until activation is triggered.

toxSAS neutralization is defined by catalytic activity

Prompted by the conceptual differences between Tis1-mediated neutralization of Tas1 and ATfaRel2-mediated neutralization of FaRel2, we next used AlphaFold223 to explore the general principles underlying the mechanisms of toxin neutralization across known toxSAS functional diversity (Fig. 5 and Supplementary Fig. 5a–l). On the toxSAS toxin side, AlphaFold2 predicts a strong conservation of a core toxSYNTH fold decorated with a variety of insertions at the N and C termini (Fig. 5a–e). On the antitoxin side, the pZFD fold is found as either a standalone neutralizing domain or part of multidomain antitoxins combined with either Tis1 or Tis1-like (Fig. 5f–h) or PanA domains11,24 (Supplementary Fig. 5a–j).

Fig. 5: Neutralization of (pp)pApp-producing FaRel toxSAS.

a–f, AlphaFold2-generated structural models of Coprobacillus sp. D7 FaRel2 (a), M. tuberculosis CapRel (unfused) (b), M. tuberculosis PhRel (c), bacteriophage Lily PhRel2 (d) and C. marina FaRel (e) and the AT2faRel:FaRel complex (f). The Tis1-like NTD of AT2faRel is colored dark green and the pZFD/ATfaRel2-like CTD is colored light cyan. AlphaFold2 predicts that FaRel is neutralized by the Tis1-like domain through the pyrophosphate acceptor site. g,h, Structural superposition of the Tis1-like NTD of AT2faRel (dark green) on Tis1 (PDB 6OX6), colored brown (g) and the CTD (light cyan) on pZFDCapRel (PDB 7ZTB), colored light teal (h). i, Probing of the secondary FaRel2:ATfaRel2 interface through toxicity neutralization assays. Serial dilutions of E. coli strains expressing either AT2faRelNTD or AT2faRelCTD alone or together with FaRel were plated on solid LB medium and scored after 16 h at 37 °C.

Structural predictions of the different neutralized complexes uncovered a general trend. Translation-targeting tRNA-pyrophosphorylating toxSASs such as fused and split CapRel, FaRel2, PhRel and PhRel2 are inhibited through the pyrophosphate donor site (Supplementary Fig. 5a–j). Conversely, metabolism-targeting (pp)pApp-producin