Short prokaryotic Argonautes provide defence against incoming mobile genetic elements through NAD+ depletion

Argonaute (Ago) proteins are found in all three domains of life. The so-called long Agos are composed of four major domains (N, PAZ, MID and PIWI) and contribute to RNA silencing in eukaryotes (eAgos) or defence against invading mobile genetic elements in prokaryotes (pAgos). The majority (~60%) of pAgos identified bioinformatically are shorter (comprising only MID and PIWI domains) and are typically associated with Sir2, Mrr or TIR domain-containing proteins. The cellular function and mechanism of short pAgos remain enigmatic. Here we show that Geobacter sulfurreducens short pAgo and the NAD+-bound Sir2 protein form a stable heterodimeric complex. The GsSir2/Ago complex presumably recognizes invading plasmid or phage DNA and activates the Sir2 subunit, which triggers endogenous NAD+ depletion and cell death, and prevents the propagation of invading DNA. We reconstituted NAD+ depletion activity in vitro and showed that activated GsSir2/Ago complex functions as a NADase that hydrolyses NAD+ to ADPR. Thus, short Sir2-associated pAgos provide defence against phages and plasmids, underscoring the diversity of mechanisms of prokaryotic Agos. Short pAgo proteins associate with the Sir2 effector from Geobacter sulfurreducens to induce abortive infection via NAD+ depletion and provide defence against invading DNA.

Article https://doi.org/10.1038/s41564-022-01239-0 catalytically active or inactive PIWI domain. Some long pAgos with the catalytically active PIWI, as exemplified by CbAgo and TtAgo, use DNA guides to target and cleave DNA providing defence against invading phages or plasmids, or contributing to chromosome segregation after replication, respectively [12][13][14] . Meanwhile, a long RsAgo from Rhodobacter spaeroides with an inactive PIWI domain guided by small RNA is thought to mobilize an unknown cellular nuclease(s) for degradation of invading plasmids and mobile genetic elements 9,15 . Interestingly, long KmAgo from Kurthia massiliensis can use both DNA and RNA guides to target DNA and RNA in vitro, albeit with different efficiencies 16,17 . In pAgos are quite widespread and are present in 9% of sequenced bacterial and 32% of archaeal genomes 6 . So far, more than ~1,000 pAgos have been identified bioinformatically, revealing a striking diversity. Moreover, pAgos are often associated with additional putative nucleases, helicases and DNA binding proteins that are not linked to eAgos 8 . pAgos are divided into full-length or long pAgos (~40%) sharing conserved N, PAZ, MID and PIWI domain architecture with eAgos, and short pAgos (~60%) composed only of MID and PIWI domains ( Fig. 1a) 8,9 . Long pAgos are relatively well-characterized both structurally and functionally 10,11 and, similar to eAgos, contain either a   Article https://doi.org/10.1038/s41564-022-01239-0 contrast to long pAgos, all short pAgos possess a catalytically inactive PIWI domain and are typically associated with proteins containing a domain initially thought to be analogous to PAZ (APAZ) 18 . Subsequently, however, it was proposed that APAZ may actually be homologous to the N-terminal domains (N and L1) of Ago 11,19 . APAZ-containing proteins are often fused to Sir2 (Silent informator regulator 2), Mrr nucleases or TIR (Toll-interleukin-1 receptor) domains 8,18 . About half of all identified short pAgos are associated with or fused into a single-chain protein with Sir2-APAZ proteins 8 . The Sir2 domain-containing proteins are widely distributed in all domains of life and perform protein deacetylation or ADP-ribosylation functions using NAD + as a co-factor 20,21 . Particularly, bacterial Sir2 proteins are involved in many cellular processes including transcription, translation, carbon and nitrogen metabolism, virulence and resistance to stress 21 . Even though short pAgos, half of which are associated with Sir2 proteins, make up the majority of all pAgos, their function in the cell and in vitro remains to be established.

Results
In this study, we aimed to explore whether short pAgos can act as prokaryotic defence systems against viruses or plasmids. To this end, we selected two short pAgos, GsSir2/Ago from Geobacter sulfurreducens and CcSir2/Ago from Caballeronia cordobensis, each encoded in a putative operon together with a Sir2 domain protein, and PgSir2-Ago from Paraburkholderia graminis, representing a fusion of Sir2 and pAgo (Fig. 1a). The coding regions of the Sir2 and pAgo proteins in GsSir2/Ago and CcSir2/Ago systems overlap by 11    Ago genes, or a single gene, encoding PgSir2-Ago into pBAD expression vectors under a P BAD promoter (Supplementary Table 1) and challenged them with phages or plasmids.

Sir2/Ago systems provide defence against phages
To test whether the GsSir2/Ago system provides defence against phages, we challenged E. coli host carrying the GsSir2/Ago system with a set of six E. coli phages spanning four morphological families including Podoviridae (T7), Siphoviridae (lambda-vir, SECphi27, SECphi18), Myoviridae (T4) and Microviridae (SECphi17, a single-stranded (ss)DNA phage) (Supplementary Table 1). We measured the efficiency of plating (EOP) of these phages with and without l-arabinose induction of the GsSir2/Ago system. The system showed protection against two out of six phages: lambda-vir (~100-fold) and SECphi27 (~1,000-fold) (Fig. 1c).
To probe the role of individual Sir2 and Ago proteins in antiviral defence, we engineered E. coli cells carrying mutant variants of either the Sir2 or the Ago protein of the GsSir2/Ago system and performed small-drop plaque assays using the lambda-vir and SECphi27 phages. In Sir2 variants, the highly conserved D230 residue, presumably involved in NAD + binding, was replaced by alanine (Extended Data Fig. 1c) 18 . Phage challenge assay revealed that the D230A mutation completely abolished defence against both phages (Fig. 1d). The Ago protein is catalytically inactive due to the active site mutations in the PIWI domain; therefore, to obtain a binding-deficient Ago variant, we fused a bulky 29 amino acid His 6 -StrepII-His 6 -tag (HSH-tag) at the C terminus that is important for nucleic acid binding in other Agos 3,6,8,22 . We found that Ago C-terminal modification abolished protection from both phages (Fig. 1d). Thus, both the Sir2 and the Ago proteins are required for protection against phages by the GsSir2/Ago system. Additionally, we probed the ability of homologous CcSir2/Ago and PgSir2-Ago systems to restrict lambda-vir and SECphi27 phages ( Fig. 1e). After l-arabinose induction, the PgSir2-Ago system showed ~500-fold protection against lambda phage and ~400-fold protection against SECphi27, while the CcSir2/Ago system showed ~1,000-fold protection against lambda and no protection against SECphi27 phage. Thus, homologous Sir2/Ago systems from three phylogenetically distant bacteria showed protection against phage infection, albeit with different efficiency and specificity.
Next, E. coli MG1655 cells either lacking or containing the GsSir2/ Ago system were infected in liquid culture with lambda phage at a multiplicity of infection (MOI) of 0.05, 0.5 and 5 ( Fig. 1f). At high MOI where, on average, a single bacterial cell is infected by a single phage, the culture collapses, while at low MOI the culture survives. This phenotype implies that GsSir2/Ago-mediated defence triggers cell death at approximately the same time when the phage-induced lysis occurs.

Sir2/Ago systems interfere with plasmid transformation
Further, to test whether heterologous expression of the GsSir2/Ago, CcSir2/Ago and PgSir2-Ago systems in E. coli cells (strains BL21-AI and DH10B) provides a barrier for plasmid transformation, four plasmids (pCDF, pCOLA, pACYC184 and pRSF) with different ori regions and copy numbers were used in a plasmid interference assay (Supplementary Table 1, Fig. 2 and Extended Data Fig. 2). We found that the GsSir2/ Ago system prevented only pCDF plasmid transformation, reducing its efficiency nearly ~100-fold (Fig. 2b). Next, we tested pCDF transformation efficiency in E. coli cells expressing the GsSir2/Ago mutants (Fig. 2c). Both Sir2 D230A mutation and pAgo HSH-tag modification that impaired phage restriction also abolished plasmid interference making cells permissive to pCDF plasmid transformation (Fig. 2c). Similar to GsSir2/Ago, the CcSir2/Ago system provided resistance only for pCDF plasmid transformation (Extended Data Fig. 2b), while cells carrying the single-chain PgSir2-Ago system were permissive to transformation of all four plasmids (Extended Data Fig. 2b). Interestingly, although the GsSir2/Ago system in E. coli interfered with the pCDF plasmid transformation, the pCOLA plasmid that differs mainly in the ori region and antibiotic resistance gene was permissive. To test whether the ori region determines differences in the transformation efficiency between pCDF and pCOLA plasmids, we swapped the ori sequences of pCDF and pCOLA (Fig. 2d). The pCDF plasmid with ColA ori instead of CloDF13 became permissive in E. coli cells expressing the GsSir2/Ago system, whereas transformation of pCOLA plasmid bearing CloDF13 ori instead of ColA ori was prevented. These results indicate that the CloDF13 ori is a key element that controls plasmid transformation efficiency in E. coli cells expressing the GsSir2/Ago system. The plasmid interference by the GsSir2/Ago system could be due to either plasmid entry exclusion, replication inhibition or plasmid degradation. To eliminate the possible role of plasmid entry barriers on the pCDF plasmid transformation efficiency, we engineered heterologous E. coli cells carrying two plasmids: pBAD plasmid providing carbenicillin (Cb) resistance and expressing GsSir2/Ago (or its mutants) under the control of P BAD inducible promoter and the pCDF plasmid providing streptomycin (Str) resistance, and tested cell viability in the presence or absence of the inducer. In this case, the pCDF plasmid is already in the cell and provides streptomycin resistance; however, antibiotic resistance should be lost if the plasmid is restricted after GsSir2/Ago expression. In the absence of induction, cell viability of E. coli cells carrying wild-type (wt) GsSir2/Ago (or its mutants) and an empty pBAD vector (Extended Data Fig. 2e) was identical. In the presence of the inducer, the viability of cells expressing the wt GsSir2/Ago system, but not its mutants, significantly decreased ( Fig. 2e), indicating that GsSir2/Ago interferes with the pCDF plasmid already present in the cell. Notably, a decrease in cell viability is observed in E. coli BL21-AI cells (tetracycline-resistant) without the cell selection for Str and Cb resistance. It cannot be ruled out that upon recognition of pCDF, the GsSir2/Ago system becomes activated and triggers cell death. A similar cell death phenotype triggered by the GsSir2/Ago was observed during phage infection in liquid cultures (Fig. 1f). Taken together, these data show that the GsSir2/Ago system acts as a defence system against phages and plasmids via cell death or suicidal mechanism.

Short pAgo and Sir2 form a stable heterodimeric complex
To characterize the Sir2/Ago systems biochemically, we aimed to express individual Sir2 and Ago proteins in E. coli. The GsSir2 and CcSir2 proteins (but not PgSir2-Ago) were expressed and purified by chromatography ( Fig. 3a and Extended Data Fig. 3). The N-terminal His 6 -tagged GsSir2 and CcSir2 proteins co-expressed with Ago proteins co-purified on the Ni 2+ -affinity column (Extended Data Fig. 3), indicating that Sir2 and pAgo proteins form a stable complex. We failed to express Sir2 and Ago proteins individually, suggesting that they form an obligatory Sir2/Ago complex. Functionally compromised GsSir2(D230A)/Ago and GsSir2/Ago-HSH variants also formed a complex, indicating that while the introduced mutations abolished the activity in vivo, they did not affect the protein complex structure ( Fig. 2c and Extended Data Fig. 3c). Further analysis of the oligomeric state of GsSir2/Ago and CcSir2/Ago complexes in solution using multi-angle light scattering coupled with size exclusion chromatography (SEC-MALS), mass photometry and small-angle X-ray scattering (SAXS) showed that heterodimeric complexes are formed in a wide range of protein concentrations (from 20 nM to 6.5 µM) ( Fig. 3b-d and Extended Data Fig. 4). According to the SAXS data, the heterodimeric GsSir2/Ago complex acquires a notably asymmetric shape (Fig. 3d, Extended Data Fig. 4 and Supplementary Table 3) that is consistent with a structural model of the heterodimer (Supplementary Note). In summary, the results show that pAgos and associated Sir2 proteins encoded by a single operon form a stable heterodimeric complex.

Sir2/Ago prefers an RNA guide to bind a DNA target
Long pAgos use ssDNA and/or ssRNA guides to recognize their complementary DNA and/or RNA targets 10 . To establish the guide preference of Gs and Cc Sir2/Ago, we analysed binding of single-or double-stranded (ds)DNA or RNA by electrophoretic mobility shift assay (EMSA) (Extended Data Fig. 6). Both Sir2/Ago heterodimers showed a strong preference for ssDNA and ssRNA binding. RNA/DNA heteroduplex was bound with an intermediate affinity, while dsRNA or dsDNA showed only weak binding (Extended Data Fig. 6a,b,e and Table 1). Interestingly, neither the 5'-terminal phosphate nor the 3'-OH end or Mg 2+ ions were required for ssDNA binding as both Gs and Cc complexes bound the circular ssDNA and linear oligonucleotides with similar affinity (Extended Data Fig. 6a,b). As expected, Gs and Cc Sir2/Ago containing the inactivated PIWI domains showed no cleavage activity for any nucleic acid (NA) substrate tested (Extended Data Fig. 6f). In summary, EMSA experiments suggest that in vitro ssRNA or ssDNA are preferable GsSir2/Ago guides.
Next, we analysed ssDNA or ssRNA target binding by the binary GsSir2/Ago complexes pre-loaded with either ssRNA or ssDNA guides. In a separate set of experiments, reaction mixtures also contained heparin, a competitor of nucleic acid binding ( Fig. 4a and Table 1). The binary GsSir2/Ago-ssRNA complex showed ~10-fold better binding to the ssDNA than to ssRNA, and heparin addition had only a little effect (~4-fold decrease) on binding affinity in this case. The GsSir2/Ago-ssDNA binary complex bound to the matching ssDNA target with affinity similar to the GsSir2/Ago-ssRNA binary complex; however, heparin addition abolished binding (Extended Data Fig. 6a,d). Furthermore, the binary Article https://doi.org/10.1038/s41564-022-01239-0 GsSir2/Ago-ssRNA complex bound to the complementary ssDNA target ~200-fold better than apo-GsSir2/Ago-bound pre-annealed gRNA/tDNA heteroduplex, indicating that GsSir2/Ago requires binding of the RNA guide first to interact with the DNA target ( Fig. 4a and Table 1). The binding affinity of the functionally compromised in vivo GsSir2(D230A)/Ago mutant was similar to the wt, while the binding affinity of the GsSir2/ Ago-HSH mutant was slightly (~7-fold) weaker (Fig. 4a, Extended Data Fig. 6c and Table 1). Charge reversal mutations of positively charged residues that are involved in interactions with the guide and target NAs according to the GsSir2/Ago model abolished both the target NA binding and the pCDF plasmid interference (Extended Data Figs. 2f,g, 3b and 6c). Taken together, our data suggest that GsSir2/Ago uses ssRNA as a guide for the recognition of an ssDNA target. To identify NAs bound by GsSir2/Ago in vivo, we purified the GsSir2/Ago-NA complex from E. coli transformed with the pBAD_GsSir2/ Ago expression vector and the pCDF target plasmid, extracted NAs and subjected them to sequencing. Subsequent analysis revealed that GsSir2/Ago is associated with small (predominantly 21 nt) RNAs with or without the 5'-phosphate (Fig. 4b,c). Different from other Argonaute proteins that show base selectivity for the first nucleotide at the 5'-end of the guide [13][14][15]23 , GsSir2/Ago-associated small RNAs show preference for the 5'-AU dinucleotide (Fig. 4d). This preference is more pronounced for small RNAs containing 5'-phosphate, implying that GsSir2/Ago uses as a guide small RNAs containing the 5'-AU dinucleotide. Most co-purified small RNAs (~95%) matched the E. coli genome, whereas the smaller fraction (~5%) originated from the pBAD_GsSir2/Ago and pCDF plasmids (Supplementary Data 1). Interestingly, small plasmid-borne RNAs that matched CloDF13 and ColE1 ori regions of corresponding plasmids were noticeably enriched (Fig. 4e). Taken together, RNA-seq data suggest that GsSir2/Ago could use small 5'-AU-RNAs originating from the invader transcripts (for example, pCDF ori region) as guides to target the invaders' DNA.  Article https://doi.org/10.1038/s41564-022-01239-0

The GsSir2/Ago complex binds NAD + and causes its depletion
Computational analysis of Sir2 domains showed that they possess a conserved NAD + -binding pocket (Extended Data Figs. 1c and 5). To determine whether Sir2 domains can indeed bind endogenous NAD + , purified GsSir2/Ago and CcSir2/Ago complexes were heat-treated, protein aggregates removed by centrifugation and the supernatant analysed by mass spectrometry-high-performance liquid chromatography (MS-HPLC) ( Fig. 5 and Extended Data Fig. 7). The quantitative analysis showed that both the GsSir2/Ago and CcSir2/Ago complexes co-purified with bound endogenous NAD + in approximately 1:1 (Sir2:NAD + ) molar ratio (Fig. 5a, and Extended Data Figs. 3 and 7). However, in the case of the functionally inactive GsSir2(D230A)/Ago mutant, only 0.6% of all complexes were NAD + -bound, indicating that the mutation severely compromised NAD + binding by the Sir2 domain (Fig. 5a). NAD + binding by the GsSir2 subunit and the similarity of GsSir2 to the N-terminal NADase of the ThsA protein from the anti-phage Thoeris system 24 prompted us to investigate the level of endogenous NAD + in the presence of the induced GsSir2/Ago system and its target pCDF plasmid. In these experiments, the corresponding E. coli cells were lysed, proteins were removed and the amount of NAD + in the supernatant was examined by MS-HPLC. When the wt GsSir2/Ago expression was induced in the presence of pCDF plasmid, NAD + was depleted (~120-fold decrease), whereas in the case of the functionally inactive GsSir2(D230A)/Ago and GsSir2/Ago-HSH mutants, the level of endogenous NAD + was similar to that of the empty pBAD vector (Fig. 5b and Extended Data Fig. 7). It should be noted that a significant ~30-fold decrease in the NAD + level was observed when the wt GsSir2/Ago expression was induced even in the absence of pCDF, suggesting that the heterologously expressed system may be toxic to the cells resulting in their slower growth ( Fig.  5b and Extended Data Fig. 7). To test the hypothesis that the activated Article https://doi.org/10.1038/s41564-022-01239-0 GsSir2/Ago system, similar to the Thoeris anti-phage system, depletes the endogenous NAD + through hydrolysis or cyclization, we attempted to identify possible products (ADPR, cADPR, AMP, cAMP, ADP, cADP, nicotinamide and adenine) using MS-HPLC, albeit without success. It is possible that NAD + conversion products were not detected due to either being processed in the cell to other reaction intermediates or to ion suppression during MS analysis of the cell lysates.
Next, we investigated whether the NAD + depletion activity detected in cells could be reconstituted in vitro. To this aim we mixed the binary GsSir2/Ago-gRNA complex with ssDNA and monitored NAD + concentration using a commercial kit (Fig. 5c). We found that NAD + concentration decreased when wt GsSir2/Ago-gRNA complex was added to the complementary ssDNA target; however, no changes were observed in the case of non-complementary ssDNA. Mutations in the Sir2 domain (D230A) or the APAZ/Ago part (GsSir2/Ago-HSH, GsSir2 APAZ /Ago and GsSir2/Ago PIWI mutants) compromised NAD + depletion (Fig. 5c). MS analysis revealed that the GsSir2/Ago-gRNA complex, in the presence of the complementary ssDNA, hydrolyses NAD + to ADPR (Fig. 5d,e), similar to the Thoeris anti-phage system 24 . Taken together, these results demonstrate that GsSir2/Ago functions as a NADase that becomes activated upon target DNA binding.

Discussion
The association of Sir2-like domains with short pAgo proteins has been identified bioinformatically in the pioneering Makarova et al. paper 18 . It has been speculated that Sir2-domain proteins can act as nucleases, but the structure and function of Sir2 proteins have so far not been elucidated. Here we show that the APAZ-containing Sir2 and short pAgo proteins form a heterodimeric complex (Fig. 3), similar to a short pAgo and a Mrr nuclease domain-containing protein 25 . Furthermore, our structure modelling results show that the APAZ region of Sir2 proteins shares similarity with the N, L1 and L2 domains of canonical Agos, substantiating previous sequence-based predictions 11,19 At the same time, Sir2 proteins entirely lack the PAZ domain ( Fig. 1a and Extended Data Fig. 5a). Thus, apparently both split and single-chain Sir2/Ago systems evolved from long pAgos by the loss of the PAZ domain and acquisition of the Sir2 domain.
Next, we provide the experimental evidence that the Sir2/Ago complex functions as a defence system against invading phages and plasmids ( Figs. 1 and 2). Intriguingly, plasmid interference assay using four plasmids with different replicons (Extended Data Fig. 2) revealed that the GsSir2/Ago system prevents transformation only of the pCDF plasmid that contains CloDF13 ori (Fig. 2), suggesting that GsSir2/ Ago may recognize specific replicon elements or structures. Indeed, ori swap between pCDF and permissive pCOLA plasmid made the latter sensitive to GsSir2/Ago interference. We show here that GsSir2/ Ago co-purifies from E. coli cells together with small (predominantly 21 nt long) RNAs that preferentially contain the 5'-AU dinucleotide (Fig. 4b-d). Interestingly, a fraction of small RNAs that originates from pCDF CloDF13 and pBAD ColE1 ori regions is enriched (Fig. 4e). ColE1-like origins, including CloDF13, use two small RNAs (RNAI and RNAII) for priming of the replication that involves the R-loop intermediate [26][27][28] . It is tempting to speculate that GsSir2/Ago is able to bind nucleic acids of different lengths (Fig. 4b,c) and preferentially binds ori-associated small RNAs that can be subjected to further processing by cellular RNases to produce ~21 nt gRNAs similarly to long RsAgo that shares an inactivated PIWI domain with GsSir2/Ago 15 .
We further show that in vitro, the reconstituted wt GsSir2/ Ago-gRNA complex becomes activated after binding the complementary DNA target and triggers NAD + hydrolysis generating ADPR (Fig. 5c-e). It is likely that in E. coli cells, the APAZ/Ago part of the GsSir2/ Ago complex guided by the ori-associated RNA guides could bind to the complementary plasmid or phage DNA target, activating the Sir2 effector domain that depletes endogenous NAD + leading to cell death, thereby restricting plasmid and phage propagation (Fig. 5f, Supplementary Text and ref. 29 ). A similar anti-phage defence mechanism based on NAD + exhaustion has been shown for the Thoeris and the Pycsar systems, CBASS (cyclic oligonucleotide-based signalling system) and DSR (defence-associated sirtuins) [29][30][31] . In the Thoeris system of Bacillus cereus MSX-D12, the Sir2 domain is similar to that of the GsSir2/Ago system and performs the hydrolysis of NAD + to ADPR and nicotinamide 31 , similar to the GsSir2/Ago system. Further structural and biochemical studies are underway to establish the structure of the heterodimeric Sir2/Ago complex and the mechanism of Sir2 domain activation that triggers NAD + hydrolysis.

Oligonucleotides used in this work
All synthetic DNA oligonucleotides used for cloning and site-specific mutagenesis were purchased from Metabion and are listed in Supplementary Table 2.
The E. coli codon optimized genes (IDT codon optimization tool) encoding the Sir2 (WP_053571900.1) and Ago (WP_053571899.1) of the CcSir2/Ago system (from C. cordobensis, NCBI taxon_id 1353886) were synthesized and cloned into pBAD/HisA expression vector by Twist Bioscience. The CcSir2 protein contains at its N terminus a His 6 -tag that can be cleaved by TEV protease. For purification of the CcSir2/Ago complex, a TwinStrep-tag (35 amino acids: MGGSAWSHPQFEKGGGSGGGSGG-SAWSHPQFEKGS) was additionally fused to the N terminus of the CcSir2 protein already containing a His 6 -tag.
To swap ori regions between pCOLA and pCDF plasmids, the DNA fragments containing ColA and CloDF13 ori were amplified by PCR  Table 2). The resulting DNA fragments were digested by NheI and XbaI (ThermoFisher, FD0684), ligated into pCOLA and pCDF vectors pre-cleaved with NheI and XbaI, and dephosphorylated using FastAP.
To swap streptomycin resistance for kanamycin in the pCDF plasmid, plasmids pCDF and pCOLA were cleaved with NheI and Eco81I (ThermoFisher, FD0374), and isolated using a runView electrophoresis system (Cleaver Scientific). The purified fragments were then ligated into pCDF to yield a pCDF_Kn plasmid.
All gene sequences were confirmed by sequencing; links to DNA and protein sequences are presented in Supplementary Table 1.

Phage restriction assay
E. coli MG1655 (ATCC 47076) cells carrying pBAD plasmids expressing wt or mutated Sir2/Ago systems were used for phage infection assays as described below. Whole-plasmid sequencing was applied to all transformed E. coli clones to verify the integrity of the system and lack of mutations, as previously described 33 .
E. coli phages (T4, T7, lambda-vir) were kindly provided by U. Qimron. Phages SECphi17, SECphi18 and SECphi27 were isolated by the Sorek lab as previously described 34 . Small-drop plaque assay was performed as previously described 35 . An overnight culture of E. coli bacteria was diluted 1:100 in MMB medium (LB + 0.1 mM MnCl 2 + 5 mM MgCl 2 + 5 mM CaCl 2 ) supplied with 0.1% l-arabinose for expression induction. Bacterial cultures were incubated at 37 °C until early log phase (optical density (OD) 600 = 0.3), and 500 µl of bacteria were mixed with 25 ml of MMB agar (LB + 0.1 mM MnCl 2 + 5 mM MgCl 2 + 5 mM CaCl 2 + 0.5% agar + 0.1% l-arabinose) and poured into a square Petri dish. Serial dilutions of phage lysate in MMB were dropped on top of the cell lawn. After the drops dried up, plates were incubated at room temperature for 24 h. EOP was determined via comparison of phage lysate titre on control bacteria and bacteria containing the Argonaute system with and without induction with l-arabinose.
Liquid culture phage infection experiments were performed as previously described 33 . After overnight incubation, the liquid suspensions of pAgo-lacking and pAgo-containing E. coli cells were diluted 1:100 in MMB medium supplied with 0.2% l-arabinose and dispensed (180 µl volume) into a 96-well plate. Plates were incubated at 37 °C until the early log phase (OD 600 = 0.3), then 20 µl of phage lysate was added to each well at multiplicities of infection of 5, 0.5 or 0.05, and each experiment was performed in three replicates. Optical density measurements at a wavelength of 600 nm were taken every 15 min using a TECAN Infinite 200 plate reader.

Plasmid interference assay
E. coli (BL21-AI and DH10B strain) cells were pre-transformed with pBAD/HisA plasmid encoding a GsSir2/Ago (either wild-type or mutant variants), CcSir2/Ago or PgSir2-Ago under the control of araBAD promoter. After 2 h of induction with either 0.01% (w/v) (CcSir2/Ago) or 0.1% (GsSir2/Ago and PgSir2-Ago) l-arabinose at 37 °C and 200 r.p.m., cells were heat-shock transformed with pCDF (pCDF_Kn in the case of PgSir2-Ago in DH10B), pCOLA, pACYC184 or pRSF plasmids. After recovery, cells were either serially diluted and aliquots spotted on selection medium, or undiluted suspensions were spread on selection medium and colony-forming units counted manually. In parallel, viability and overexpression of pAgo-containing E. coli cells were monitored using serial dilutions and western blot, respectively.
In separate experiments, E. coli BL21-AI strain cells were heat-shock transformed with both GsSir2/Ago operon-encoding pBAD/HisA construct and pCDF plasmid. Protein expression in selected double transformants was then induced by the addition of l-Ara (final concentration of 0.1% (w/v)) into liquid LB culture. After a 2 h induction at 37 °C and 200 r.p.m., OD 600 equalized, cultures were serially diluted and aliquots were spotted on a selection medium containing different antibiotics.

Expression and purification of GsSir2/Ago complexes
For GsSir2/Ago protein expression, E. coli DH10B strain was transformed with a corresponding plasmid (Supplementary Table 1). Cells were grown at 37 °C in LB medium in the presence of 50 µg ml −1 ampicillin until OD 600 = 0.7 was reached. Then, the temperature was decreased to 16 °C and proteins were expressed for 16 h by adding 0.2% w/v l-arabinose. Next, collected cells were disrupted by sonication in buffer A (20 mM Tris-HCl (pH 8.0 at 25 °C), 500 mM NaCl, 2 mM phenylmethylsulfonyl fluoride, 5 mM 2-mercaptoethanol), and cell debris was removed by centrifugation. GsSir2/Ago complexes were purified to >90% homogeneity by chromatography through HisTrap HP chelating, HiTrap Heparin HP and HiLoad Superdex 200 columns (GE Healthcare). Purified proteins were stored at −20 °C in a buffer containing 20 mM Tris-HCl (pH 8.0 at 25 °C), 200 mM KCl, 1 mM dithiothreitol (DTT) and 50% v/v glycerol. The identity of the purified proteins was confirmed by mass spectrometry. Protein concentrations were determined from OD 280 measurements using the theoretical extinction coefficients calculated with the ProtParam tool available at http://web.expasy. org/protparam/. GsSir2/Ago complex concentrations are expressed in terms of heterodimer. The GsSir2/Ago MID surface mutant could not be purified due to its poor expression.

SEC-MALS and mass photometry
Size-exclusion chromatography of GsSir2/Ago complexes was carried out at room temperature using a Superdex 200 10/300 GL column (GE Healthcare) pre-equilibrated with a buffer (20 mM Tris-HCl (pH 8.0 at 25 °C), 500 mM NaCl). A calibration curve was generated by measuring the elution volumes of a series of standard proteins of known molecular mass (Bio-Rad). The molecular masses of pAgos complexes were calculated by interpolating their elution volume onto the calibration curve. SEC-MALS of GsSir2/Ago and CcSir2/ Ago complexes was performed at room temperature using a Superdex 200 10/300 GL column (GE Healthcare) pre-equilibrated with a buffer (20 mM Tris-HCl (pH 8.0 at 25 °C), 500 mM NaCl, 0.03% NaN 3 , 1 mM DTT), at 0.4 ml min −1 flow rate. Sample concentrations were 6 µM and 6.5 µM for GsSir2/Ago and CcSir2/Ago, respectively. Light scattering signals were monitored on a miniDawn TREOS II detector, and concentrations of protein samples were measured using an Optilab T-rEX refractive index detector (Wyatt Technologies). Data were analysed in Astra software (Wyatt Technologies) using a specific refractive index increment (dn/dc) value of 0.185 ml g −1 .
Mass photometry of the GsSir2/Ago complex was performed using a Refeyn OneMP system (Refeyn). The protein complex was diluted to 20 nM in a buffer containing 20 mM Tris-HCl (pH 8.0) and 500 mM NaCl before measurement.

SAXS analysis
The synchrotron SAXS data were collected at beamline P12 operated by EMBL Hamburg at the PETRA III storage ring (DESY, Hamburg, Germany) 36 . The GsSir2/Ago sample in the storage buffer was transferred into the sample buffer (20 mM Tris-HCl (pH 7.5), 200 mM NaCl, 5 mM MgCl 2 , 2 mM β-mercaptoethanol) using gel-filtration NAP column (GE Healthcare) and concentrated by ultrafiltration to 1.2, 1.3, 1.6 and 5.5 mg ml −1 concentrations. The data were collected at 0.124 nm wavelength and the distance to the detector (Pilatus 2 M, Dectris) was set to 3 m. Samples in the sample changer were kept at 10 °C and capillary temperature was set to 20 °C. Twenty frames exposed for 0.045 s were averaged for each concentration. The s-range of collected Article https://doi.org/10.1038/s41564-022-01239-0 data was from 0.0133796 to 3.7925 nm −1 , where s = 4πsinθ/λ is the momentum transfer (2θ is the scattering angle, λ is the wavelength of the X-rays). The data were analysed using programmes of the ATSAS 2.8.4 (r10552) suite 37 . Data were normalized to an absolute scale with water as standard. As the data collected for the sample with a concentration of 1.22 mg ml −1 were noisy at higher s (Extended Data Fig. 4), and higher-concentration data showed more aggregation at low s, we used a merged dataset produced with PRIMUS 38 . Scattering data were parameterized and indirectly Fourier transformed with GNOM5 (ref. 39 ). Structural parameters of this dataset are summarized in Supplementary Table 3. The dimensionless Kratky plot in Extended Data Fig. 4 was calculated as described previously 40 . The ab initio models were calculated by GASBOR 41 software. Molecular mass estimations of the apo-GsSir2/Ago complex in solution, assessed by ATSAS tools (DATVC, DATMW) and server SAXSMoW (http://saxs.ifsc.usp.br/) 42 , are presented in Supplementary Table 3.
Binding experiments for the binary GsSir2/Ago complex were conducted by first pre-mixing 5'-phosphorylated ssRNA or ssDNA guide with the equimolar GsSir2/Ago complex in the same buffer as above. The binary GsSir2/Ago:NA guide complex was then diluted to two times the final reaction concentration (with respect to the guide) in the same buffer and mixed with a complementary 5'-32 P-target oligonucleotide in the presence or absence of 67 ng µl −1 heparin sodium salt (Sigma-Aldrich, H3149). The final reaction contained 10 pM target NA and 0, 0.02, 0.05, 0.1, 0.2, 0.5 and 1 nM of GsSir2/Ago:NA guide complex. A control (C g *) contained 0.1 nM GsSir2/Ago:NA guide complex with the guide labelled with [γ-32 P]ATP, and 10 pM unlabelled target NA. Three independent replicates were performed.
The binding reaction mixtures were analysed by EMSA in a PAA gel (8% 29:1 acrylamide/bis-acrylamide in TAE). The electrophoresis TAE buffer was supplemented with 5 mM magnesium acetate. Radiolabelled substrates were detected and quantified using a phosphor imager. The results were analysed with OptiQuant and OriginPro software. The K d was calculated from the following formula: where S NB is the unbound substrate (nM), S 0 is the initial substrate concentration (nM), E 0 is the initial protein complex concentration (nM), and A1 is the non-binding fraction of substrate (%).

Nucleic acid extraction and analysis
To obtain GsSir2/Ago-bound nucleic acids, E. coli DH10B was transformed with pBAD/HisA_TwinStrep_TEV_GsSir2/Ago and pCDF_Kn plasmids (Supplementary Table 1). Cells were grown at 37 °C in LB medium in the presence of 50 µg ml −1 ampicillin and 25 µg ml −1 kanamycin until OD 600 = 0.7 was reached. Then, expression was induced by adding 0.1% w/v l-arabinose, and cells were collected after 2 h. Cells were disrupted using B-PER bacterial protein extraction reagent (ThermoFisher, 78248) containing 6 mg ml −1 lysozyme. The GsSir2/ Ago-NA complex was purified as described above, except that all buffer solutions contained 100 mM NaCl.
To extract nucleic acids co-purified with the GsSir2/Ago complex, 800 µl of Roti-phenol/chloroform/isoamyl alcohol (Carl-Roth, A156) was added to 800 µl of purified protein-NA fractions in 5PRIME phase lock gel tubes (Quantabio, 733-2477). The upper aqueous phase was isolated and 0.1 volume of 1 M sodium acetate, 3 volumes of 100% ethanol and 10 µl glycogen (ThermoFisher, R0561) were added. This mixture was vortexed briefly and incubated at −20 °C for 20 h. Samples were centrifuged for 20 min and the supernatant was removed from the pellet. The pellet was washed with cold (−20 °C) 70% ethanol. The pellets containing the co-purified nucleic acids were dried for 20 min at room temperature, and pellets were resuspended in 30 µl water (free of nucleases).
In a control sample, total RNA from induced cells was extracted using SPLIT RNA extraction kit (Lexogen, 008). Then ribosomal RNA was removed using RiboCop for Gram-negative bacteria (Lexogen, 126).

RNA sequencing and analysis
Half of extracted RNA was treated with T4 PNK (ThermoFisher, EK0031) according to the protocol of the manufacturer. Then T4 PNK treated and untreated RNA samples were converted to DNA libraries using the Small RNA-seq Library Prep kit (Lexogen, 052). Concentration and quality of libraries were measured with a Qubit fluorometer (ThermoFisher) and a 2100 Bioanalyzer (Agilent).
Both libraries were sequenced using Illumina MiniSeq sequencing with single-end reads and 75 bp read length. Single-end reads were processed by trimming adapters with AdapterRemoval v2.3.0 (ref. 43 ). Then the processed reads were aligned to the E. coli strain K12 substrain DH10B genome (GenBank: CP000948.1) and the additional pBAD/ HisA_TwinStrep_TEV_GsSir2/Ago, pCDF_Kn plasmids (Supplementary Table 1) using BWA-MEM v0.7.17 (ref. 44 ). To avoid filtering out shorter reads during the alignment process, aligned reads with mapping quality values greater than or equal to 15 were chosen. FastQC Article https://doi.org/10.1038/s41564-022-01239-0 v0.11.8 (ref. 45 ) was used for read quality control and SAMtools v1.7 (ref. 46 ) for indexing, sorting and analysing alignment files. A custom script (fragmentation-bias.jl) in combination with Weblogo v3.7.4 (ref. 47 ) was used to produce nucleotide frequency plots. The custom script had to be implemented to ensure that only aligned reads would be used for nucleotide frequency analysis. Gene enrichment analysis was performed with bedtools v2.26.0 (ref. 48 ) and FPKM_count.py v4.0.0 of the RSeqQC package 49 . IGV v2.5.2 (ref. 50 ) was mainly used to inspect and visualize read coverage along the genomes. A control DNA library of total RNA was prepared using CORALL Total RNA-seq Library Prep kit (Lexogen, 095). Concentration and quality of the library were measured with a Qubit Fluorometer (ThermoFisher) and a 2100 Bioanalyzer (Agilent) according to the protocol of the manufacturer.
The control DNA library was sequenced using Illumina NextSeq sequencing with paired-end reads and 75 bp read length. Read processing, alignment and alignment analysis were similar to those for Illumina MiniSeq sequencing.

Preparation of E. coli cells for NAD + quantification
Overnight cultures of single colonies of E. coli DH10B strain harbouring a pBAD-His construct with either wt GsSir2/Ago or mutant system (GsSir2/Ago-HSH or GsSir2(D230A)/Ago), or empty vector (negative control) were diluted and grown in LB broth (BD) supplemented with respective antibiotics (50 µg ml −1 ampicillin and 25 µg ml −1 streptomycin) at 37 °C until they reached OD 600 = 0.4-0.5. Cell cultures were either induced to express the protein or not (control samples). l-Ara (0.1% final concentration) was added to induce protein expression. Induced and non-induced cultures were collected 2 h later. The cultures were normalized to OD 600 of approximately 0.7 and the pellet from 1 ml of culture suspension was stored at −80 °C until further analysis. All cell pellets were lysed by adding B-PER solution (ThermoFisher) supplemented with 6 mg ml −1 lysozyme (62971, Fluka) for 20 min at room temperature while gently rocking (Multi Bio 3D Mini-Shaker, Biosan). Cell debris was removed by centrifugation and metabolites were isolated by phenol:chloroform:isoamyl alcohol (25:24:1, v/v/v) extraction. Metabolites were stored at −20 °C until MS-HPLC analysis. Additionally, the endogenous NAD + concentration was estimated using NAD/NADH quantitation kit (Sigma Aldrich, MAK037) from four independent measurements.

In vitro NADase assay
Reaction mixtures with a volume of 25 µl were prepared with the following final concentrations: 0.5 µM GsSir2/Ago or mutant complex, 50 µM NAD + , 1× Tango buffer (33 mM Tris-acetate (pH 7.9 at 37 °C), 10 mM magnesium acetate, 66 mM potassium acetate, 0.1 mg ml −1 BSA; ThermoFisher, BY5), 1 mM DTT, 0.5 µM 5'P-RNA guide (TF-A) and/or 0.5 µM ssDNA (MZ-949 or MZ-589) (Supplementary Table 2). Reactions with RNA guide were pre-incubated for 15 min at 37 °C, then ssDNA was added and the mixture incubated for 1 h at 37 °C. A volume of 3 µl of each sample was used as input for the NAD/NADH quantitation kit (Sigma Aldrich, MAK037) according to the instructions provided by the manufacturer. All experiments were performed in triplicate. These samples were also used for mass spectrometry.

Mass spectrometry of NAD +
To quantitate NAD + bound to GsSir2/Ago and CcSir2/Ago complexes, high-performance liquid chromatography-mass spectrometry/mass spectrometry (HPLC-MS/MS) analysis was used. First, purified pAgos complexes were diluted to 5 µM in a buffer containing 20 mM Tris-HCl (pH 8.0 at 25 °C) and 200 mM NaCl. Then 20 µl of the solution was incubated at 70 °C for 20 min and centrifuged for 30 min (16,100 g at 4 °C) to remove unfolded proteins. The supernatants and NAD + standards were analysed by electrospray ionization mass spectrometry (ESI-MS) using an integrated HPLC/ESI-MS system (1290 Infinity, Agilent Technologies/Triple Quadrupole 6410, Agilent Technologies) equipped with a Supelco Discovery HS C18 column (7.5 cm × 2.1 mm, 3 µm; Agilent Technologies). HPLC/ESI-MS/MS was performed using two ion transitions to detect NAD + in the samples: 662.1→540.1 and 662.1→426.0. Ion transition 662.1→540.1, being the most abundant, was used for the quantitative analysis. Mobile phase A was 5 mM ammonium acetate in water (pH 7.0) and mobile phase B was 5 mM ammonium acetate in methanol (pH 7.0). The HPLC parameters were as follows: flow 0.25 ml min −1 ; column temperature 30 °C; 0-3 min, 0% B; 3-9 min, 0-40% B; 9-10 min, 40-100% B; 10-13 min, 100% B. The MS was operated using negative electrospray ionization at 2,500 V, the gas temperature was set to 300 °C and the fragmentor voltage was 135 V. Multiple reaction monitoring was used with a collision energy of 15 V to measure ion m/z 540.1 (ion transition 662.1 → 540.1) and also with a collision energy of 20 V to measure ion m/z 426.0 (ion transition 662.1 → 426.0).
To quantitate endogenous NAD + , HPLC-MS analysis was performed by ESI-MS using an integrated HPLC/ESI-MS system (1290 Infinity, Agilent Technologies/Q-TOF 6520, Agilent Technologies) equipped with a Supelco Discovery HS C18 column (7.5 cm × 2.1 mm, 3 µm; Agilent Technologies). The samples were investigated in both negative and positive ionization modes. For negative ionization mode, solvents A (5 mM ammonium acetate in water, pH 7.0) and B (5 mM ammonium acetate in methanol, pH 7.0) were used. For positive ionization mode, solvents C (0.02% formic acid in water) and D (0.02% formic acid in acetonitrile) were used. In both cases, elution was performed with a linear gradient of solvents at a flow rate of 0. NAD + hydrolysis products generated by GsSir2/Ago in vitro were analysed as above. Using negative ionization mode only accumulation of ADPR was detected.

Notes
During the revision of this manuscript, a paper was published that shows plasmid-induced degradation of NAD + in vivo by short pAgo-associated TIR-APAZ systems (named SPARTA) 51 . The paper also shows that expression of Sir2/Ago systems (named SPARSA) in E. coli triggers NAD depletion.

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

Code availability
The Julia script used to identify nucleotide frequency at the beginning of the aligned reads and prepare the input for the Weblogo programme is available in the GitHub repository at https://github.com/ agrybauskas/argonaute-bound-rna-manuscript. Fig. 1 | Bioinformatic analysis. a, PIWI catalytic tetrad DEDX alignment. The 4 catalytic residues (red numbers indicate positions of corresponding GsAgo positions) are shown in 4 motifs of ±3 positions. The motifs are separated by vertical blue lines. Sequence names consist of the following: NCBI sequence ID, abbreviated phylum (for example, 'ProG'gamma-proteobacteria) and organism name. b, Top -a circular phylogenetic tree was generated according to supplementary data provided with Ryazansky et al. 8 Long-A pAgo variants are coloured in green (truncated variants without the PAZ domain, light green), long-B pAgo proteins are light green (truncated variants without PAZ, green), and short pAgo proteins are orange. pAgo proteins containing the catalytic tetrad DEDX in their PIWI domain are indicated in blue on the outer circle; pAgos with inactivated PIWI domain are indicated in light grey on the outer circle. pAgo proteins of the GsSir2/Ago, CcSir2/Ago and PgSir2-Ago systems are indicated by 'Gs', 'Cc' and 'Pg', respectively. Bottom -Circular phylogenetic tree of APAZ domains. The circular phylogenetic tree of the five groups of APAZ domains was generated using APAZ domain alignments from Ryazansky et al. 8 supplementary file 7. c, Top -Combined alignment of Sir2 domains. Alignment consists of 3 parts, separated by horizontal black lines. In the top part, the Sir2 domain sequences of the GsSir2, CcSir2, PgSir2-Ago and homologues are shown. Logos above depict the conservation of Sir2 domains of Ia and Ib groups. The indicated position numbers correspond to the GsSir2 sequence. In the bottom part, homologues (sirtuins) of catalytically active Thermotoga maritima Sir2 (TmSir2) deacetylase are shown. Logos below indicate the conservation of these homologues. The position numbers correspond to the TmSir2 sequence. Sequences of six motifs that include all positions that form the NAD + -binding pocket, as seen in the TmSir2 structure (PDB ID 2H4F) are shown. Sequence names for the top alignment consist of sequence ID, abbreviated phylum and organism name. Sequence names for bottom alignment all start with 'Sir2hom' followed by sequence ID, organism name and short protein name (based on annotation). Stars above the logos indicate residues in the NAD + -binding pocket of canonical sirtuins (for example, TmSir2) that are also conserved. Star colours indicate conservation between the two groups: green -conserved in both canonical sirtuins and GsSir2-like; blue -conserved in both groups, but different; yellow -conserved only in canonical sirtuins; red -conserved only in GsSir2-like proteins. In the middle, alignment of ThsA homologues with Sir2 domains. Bottom -MID domain alignment. Red stars indicate positions of amino acids involved in the binding of the 5'-P end of the guide nucleic acid. The numbering above corresponds to the GsAgo sequence. Additionally, concatenated alignment of just the 6 indicated positions is shown on the right. The three sequences of interest are indicated with red rectangles. Numbers on the left and right of the alignment indicate the first and last positions in the alignment for each sequence.