Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jun 1.
Published in final edited form as: Nat Chem Biol. 2022 Sep 26;18(12):1417–1424. doi: 10.1038/s41589-022-01137-w

Structural basis of AcrIF24 as an anti-CRISPR protein and transcriptional suppressor

Indranil Arun Mukherjee 1, Clinton Gabel 1, Nicholas Noinaj 1, Joseph Bondy-Denomy 2,3,4, Leifu Chang 1,5,6
PMCID: PMC9691602  NIHMSID: NIHMS1834880  PMID: 36163386

Abstract

Anti-CRISPR (Acr) proteins are encoded by phages to inactivate CRISPR-Cas systems of bacteria and archaea and are used to enhance the CRISPR toolbox for genome editing. Here we report the structure and mechanism of AcrIF24, an Acr protein that inhibits the type I-F CRISPR–Cas system from Pseudomonas aeruginosa. AcrIF24 is a homodimer that associates with two copies of the surveillance complex (Csy), and prevents the hybridization between crRNA and target DNA. Furthermore, AcrIF24 functions as an anti-CRISPR-associated (Aca) protein to repress the transcription of the acrIF23-acrIF24 operon. Alone or in complex with Csy, AcrIF24 is capable of binding to the acrIF23-acrIF24 promoter DNA with nanomolar affinity. The structure of a Csy–AcrIF24–promoter DNA complex at 2.7 Å reveals the mechanism for transcriptional suppression. Our results reveal that AcrIF24 functions as an Acr-Aca fusion protein and extend our understanding of the diverse mechanisms utilized by Acr proteins.

Graphical Abstract

graphic file with name nihms-1834880-f0001.jpg

Editor summary:

EMukherjee et al. report that AcrIF24 is an Acr-Aca fusion protein, which inhibits the Csy complex and suppresses transcription from the acrIF23–acrIF24 promoter, and present cryo-EM structures to reveal the mechanism for both roles of AcrIF24.

INTRODUCTION

The Clustered Regularly Interspaced Palindromic Repeats (CRISPR) and CRISPR-associated protein (Cas) systems are widespread adaptive immune systems in prokaryotes used to combat invading mobile genetic elements (MGEs) including viruses and plasmids. These systems utilize RNA-guided effector endonucleases to specifically recognize and cleave foreign DNAs and have been developed as tools for genome editing applications. The class 1 CRISPR-Cas systems (types I, III, and IV) employ multi-protein effector complexes to cleave foreign nucleic acids, whereas the class 2 systems (types II, V, and VI) utilize a single multi-domain Cas effector1,2.

However, MGEs evolved anti-CRISPR (Acr) proteins to inhibit the CRISPR-Cas systems. Since the discovery of the first Acr proteins (AcrIF1–5) in 20133, a total of 24 Acr protein families have been reported to inactivate type I-F CRISPR–Cas systems in Pseudomonas aeruginosa3–6. The type I-F systems contain a surveillance complex (the Csy complex) that is composed of four Cas proteins (Cas5f, Cas6f, Cas7f, and Cas8f in a stoichiometry of 1:1:6:1) and a 60-nt crRNA that integrates all these protein subunits7–11. Target DNA recognition and R-loop formation induce conformational changes in Csy, enabling Cas8f to recruit the Cas2/3 nuclease for degradation of invading MGEs10. Many AcrIF proteins inhibit the type I-F system by blocking target DNA recognition by Csy, such as AcrIF17,8, AcrIF27–9, AcrIF612, AcrIF713, AcrIF812, AcrIF911,12,14, AcrIF108, and AcrIF1413,15. AcrIF3 adopts a similar structure to the C-terminal helical bundle of Cas8f (Cas8fHB), competitively binding to the Cas2/3 nuclease and preventing its recruitment to Csy10,16. AcrIF11 inactivates Csy by specifically ADP-ribosylating a key residue (N250 of Cas8f), preventing the recognition of the protospacer adjacent motif (PAM) of target DNA17. AcrIF4 binds to the Cas8fHB and precludes the conformational changes required for activation of Csy13; it is also shown to inhibit DNA binding in vivo18.

AcrIF24 was recently discovered from a P. aeruginosa prophage and is the largest AcrIF protein (228 amino acids) to date6. Another acr gene (acrIF23) is found upstream of acrIF24, making an acrIF23-acrIF24 operon that is activated by a single promoter. Usually, acr genes are found upstream of a conserved anti-CRISPR-associated (aca) gene, which encodes an Aca protein that functions as a repressor of the acr-aca operon19,20. Without Aca activity, the strong transcription driven by an acr-associated promoter is lethal to a phage19. However, the acrIF23-acrIF24 operon is not associated with an aca gene (Extended Data Fig. 1a). Interestingly, the C-terminus of AcrIF24 contains an HTH domain that is conserved in Aca proteins, indicative of a role also as an Aca protein6.

In this study, we elucidated the mechanisms for the dual function of AcrIF24 combining cryo-EM and biochemical analysis. Our results reveal that AcrIF24 is a dimer, which binds directly to Csy and prevents the hybridization between target DNA and crRNA. Furthermore, AcrIF24 binds tightly to the acrIF23-acrIF24 promoter DNA and is capable of acting as a transcriptional repressor.

RESULTS

AcrIF24 directly binds to Csy, whereas AcrIF23 does not

We first tested the function of AcrIF24 and AcrIF23 using an in vitro DNA cleavage assay17. AcrIF24 can fully inhibit substrate DNA cleavage by Cas2/3 at a 1:1 molar ratio between AcrIF24 and Csy (Extended Data Fig. 1b,c). In comparison, AcrIF23 shows full inhibition at a much lower molar ratio (Extended Data Fig. 1b,c). Electrophoresis mobility shift assay (EMSA) results suggest that AcrIF24 blocks both the formation of the Csy–DNA complex and recruitment of Cas2/3 nuclease, whereas AcrIF23 does not affect either DNA or Cas2/3 recruitment (Extended Data Fig. 1d–f). Size-exclusion chromatography also showed that AcrIF24 directly binds to Csy (Extended Data Fig. 2a).

Overall structure of Csy–AcrIF24

To understand the inhibition mechanism of AcrIF24, we determined the structure of the Csy–AcrIF24 complex using single-particle cryo-EM (Fig. 1, Extended Data Fig. 3, and Supplementary Table 1). We observed two populations of the Csy–AcrIF24 complex: AcrIF24 dimer bound to one copy of Csy (Csy–AcrIF24; ~ 91% of total particles) and two copies of Csy (Csy–AcrIF24 dimer; ~ 9% of total particles), determined at 3.2 and 3.5 Å resolutions, respectively (Extended Data Fig. 3). The cryo-EM maps were of sufficient quality for building the atomic model of AcrIF24 de novo and rebuilding the components in Csy (PDB: 7JZW)13. Csy does not show major conformational changes upon binding to AcrIF24 (PDB: 6B45)8. Briefly, Csy is composed of a Cas6f ‘head’, a Cas8f–Cas5f ‘tail’, and a ‘backbone’ of six copies of Cas7f, integrated by crRNA (Fig. 1a,b). An AcrIF24 dimer binds to the middle region of the Cas7f ‘backbone’, with direct contacts to multiple Cas7 subunits. The AcrIF24 dimer is capable of dimerizing Csy in a ‘back-to-back manner’, placing the Cas6f ‘heads’ at both ends of the assembly (Fig. 1c,d).

Figure 1. Overall structures of Csy–AcrIF24 complexes.

Figure 1.

Open in a new tab

(a) Cryo-EM map of Csy–AcrIF24 in surface representation with each subunit color-coded. Two subunits in the AcrIF24 dimer are shown in cyan and yellow, respectively. (b) Atomic model of Csy–AcrIF24 in cartoon representation. (c) Cryo-EM map of Csy–AcrIF24 dimer in surface representation. (d) Atomic model of Csy–AcrIF24 dimer in cartoon representation.

Structure of AcrIF24

The AcrIF24 monomer contains three domains including an N-terminal domain (NTD, amino acids 1–71), a middle domain (MID, 72–138), and a C-terminal HTH domain (HTH, 139–228) (Fig. 2a and Extended Data Fig. 4a,b). The NTD is composed of an anti-parallel β-sheet capped by a helix in a β1β2β3β4α1 topology (Fig. 2a). A DALI search reveals that the NTD is structurally similar to an auxiliary protein of methane monooxygenase hydroxylase, MMOD21 (PDB: 6D7K; Z-score 6.5) (Extended Data Fig. 4c). The MID domain contains a beta-sheet (β6β5β7β8) followed by α2, with a long loop between β6 and β7 (amino acids 91–109). However, no structural homology of the MID domain was found in a DALI search. HTH is composed of five short helices (α3–α7), and contacts NTD through hydrophobic interactions (Extended Data Fig. 4d). A DALI search reveals that this domain shares a similar structure with anti-CRISPR-associated protein 1 (Aca1) (Z-score 8.0, PDB 7VJN) (Extended Data Fig. 4e)22. In vitro DNA cleavage and EMSA assays with domain deletion mutants suggest that the MID and HTH domain are necessary, and that the NTD or HTH are not sufficient for inhibition of DNA cleavage and association between AcrIF24 and Csy (Fig. 2b,c and Extended Data Figs. 2b and 5a,b).

Figure 2. Structure of AcrIF24.

Figure 2.

Open in a new tab

(a) Atomic structure of a monomeric subunit of AcrIF24 in cartoon representation with each domain color-coded. (b) Quantification of in vitro DNA cleavage assay in the presence of wild-type and domain deletion mutants of AcrIF24. (c) Quantification of EMSA results showing binding between Csy and target DNA in the presence of wild-type and domain deletion mutants of AcrIF24. (d) AcrIF24 dimer in cartoon representation. Boxed regions are the dimeric interfaces detailed in e and h. (e) MID–MID interaction showing the interacting residues. (f) Quantification of in vitro DNA cleavage assay using wild-type AcrIF24 and AcrIF24 with mutations of residues involved in the dimerization interfaces. (g) Quantification of EMSA assay showing the effects of AcrIF24 variants in inhibiting the formation of the Csy–DNA complex. (h) HTH–HTH interaction with residues in the interface highlighted. The in vitro DNA cleavage assay and EMSA experiments were repeated three times with similar results and representative raw data is shown in Extended Data Fig.5. Error bars represent SD; n = 3. Two-sided t-test was performed (*P < 0.05).

The cryo-EM structures show that AcrIF24 exists as a dimer in the Csy–AcrIF24 complex. To study the stoichiometry of only AcrIF24 in solution, we used inline size-exclusion chromatography small-angle X-ray scattering (SEC-SAXS). AcrIF24 (theoretical mass 25 kDa) eluted as a single peak from SEC with the SAXS analysis showing that AcrIF24 is a dimer with an experimentally estimated molecular weight of 56 kDa (Extended Data Fig. 6a,b). Further, Rg and Dmax values are consistent with a dimer for AcrIF24 (Extended Data Fig. 6c). A cryo-EM reconstruction of AcrIF24 also supports that AcrIF24 in solution is a dimer (Extended Data Fig. 3e). We also performed SEC-SAXS analysis for AcrIF23 and found that AcrIF23 exists in a monomeric state in solution (Extended Data Fig. 6d–f).

The dimerization interface of AcrIF24 lies at the MID–MID and the HTH–HTH interfaces of the two monomers with a buried surface area of 1132 Å2 (Fig. 2d). The MID–MID interface (buried surface 485 Å2) is mediated primarily by hydrophobic interactions between helix α2 from two protomers. Specifically, side chains of L86, L125, and Y128 form a hydrophobic patch at the N-terminal end of the helices (Fig. 2e and Extended Data Fig. 4f). Substitution of any of these residues to alanine or glutamic acid significantly alleviates AcrIF24 mediated inhibition of substrate cleavage in the in vitro cleavage assay (Fig. 2f and Extended Data Figs. 2c and 5c). V133 at the C-terminal end of α2 likely augments the interaction, but substitution to glutamic acid showed no detectable effect in the in vitro cleavage assay (Fig. 2f). EMSA results suggest that wild-type AcrIF24 prevents the formation of the Csy–DNA complex, but mutant AcrIF24 (e.g. L86E, L125E, and Y128A) did so less efficiently (Fig. 2g and Extended Data Fig. 5d). Taken together, the hydrophobic patch at the N-terminal end of α2 is critical for the function of AcrIF24.

The HTH–HTH dimerization interface (buried surface 624 Å2) is primarily stabilized by interactions between helix α7 of both AcrIF24s (Fig. 2h and Extended Data Fig. 4g). Y217, located in the middle of the helix, is critical for the Acr function of AcrIF24. Alanine substitution of Y217 almost completely abolished the AcrIF24 mediated inhibition in in vitro cleavage assays (Fig. 2f and Extended Data Figs. 2c and 5c). A quadruple mutant containing alanine substitutions of Y217 and bulky residues flanking Y217 (H154A/H216A/Y217A/R221A) showed no detectable inhibition mediated by AcrIF24 (Fig. 2f and Extended Data Figs. 2c and 5c). The SEC-SAXS and MALS data suggest that this quadruple mutant fails to form a dimer (Extended Data Figs. 6g,h). In the presence of this quadruple mutant, DNA binding to Csy is restored in the EMSA assay (Fig. 2g and Extended Data Fig. 5d). Together, these results illustrate that AcrIF24 functions as a dimer and prevents target DNA binding to Csy.

Interactions between AcrIF24 and Csy

Both the NTD and MID domains directly contact the Cas7 ‘backbone’ of Csy, and the interactions would block the hybridization between the target DNA and crRNA (Fig. 3a,b). The NTD of AcrIF24 primarily engages with the thumb domain of Cas7.6f and a loop of Cas7.5f (residues 236–276) through main chain interactions (Fig. 3c and Extended Data Fig. 4h). The β2-β3 loop (16-RGLHYGGLP-24) forms three main chain hydrogen bonds with the thumb domain of Cas7.6f (Fig. 3c). Replacement of the loop with a GSGS linker significantly reduced the AcrIF24 inhibition in the in vitro cleavage assay (Fig. 3d and Extended Data Figs. 2c and 5e). In addition, β4 forms a pair of hydrogen bonds with a loop (residues 236–276) of the palm domain of Cas7.5f (Fig. 3c).

Figure 3. Interaction between AcrIF24 and Csy.

Figure 3.

Open in a new tab

(a) Overview of binding between AcrIF24 and Csy. AcrIF24 binds to the Cas7 ‘backbone’ of Csy. (b) Structure of Csy–DNA (PDB ID: 6NE0) shown in the same view of Csy as in a. (c) Detailed interaction between AcrIF24-NTD and Csy. The β2-β3 loop interacting with Csy is colored in red. Specific interactions between the 16–24 region of AcrIF24 and Cas7.6f thumb domain are detailed in the box on the right. (d) Quantification of in vitro DNA cleavage assay in the presence of wild-type AcrIF24 and AcrIF24 with mutations of residues involved in Csy interaction shown in c and e. (e) Detailed interaction between AcrIF24-MID and Csy. The long β6-β7 loop region interacting with Csy is colored in red. Interactions between R118 of AcrIF24 and Cas7.2f thumb domain are detailed in the box on the right. (f) Quantification of EMSA assay showing the effects of wild-type and mutant AcrIF24 variants in inhibiting the formation of Csy–DNA. The in vitro DNA cleavage assay and EMSA experiments were repeated three times with similar results and representative raw data is shown in Extended Data Fig. 5. Error bars represent SD; n = 3. Two-sided t-test was performed (*P < 0.05)

The MID domain of AcrIF24 interacts with the palm domains of Cas7.2f and Cas7.3f, and the thumb domain of Cas7.4f mainly through the long β6-β7 loop (Fig. 3e and Extended Data Fig. 4i). Specifically, the N-terminus of the loop (92-SRLPEA-97) interacts with Cas7.3f and Cas7.4f, with the C=O group of L94 forming a hydrogen bond with the side chain of D82 and E96 establishing a salt bridge with K318, respectively. Replacement of the 92–97 loop with a GSGS linker reduced AcrIF24 function in the in vitro cleavage assay (Fig. 3d and Extended Data Figs. 2c and 5e). The C-terminus of the loop (104–107) engages Cas7.2f, with a salt bridge formed between D105 of the β5-β6 loop and R333 of Cas7.2f. Alanine substitution of the 104–107 region caused impaired inhibition capacity of AcrIF24 (Fig. 3d and Extended Data Figs. 2c and 5e). In addition to the long β6-β7 loop, the β7-β8 loop also contacts Cas7.2f, with R118 of the loop forming salt bridges with E357 of Cas7.2f; alanine substitution also caused impaired inhibition capabilities of AcrIF24 (Fig. 3d,e and Extended Data Figs. 2c and 5e). EMSA data using wild-type and mutant variants of AcrIF24 suggest that AcrIF24 inhibits Csy by preventing the formation of the Csy–DNA complex (Fig. 3f and Extended Data Fig. 5f).

Csy–AcrIF24 binds DNA independent of the spacer sequence

Our EMSA data show that in the presence of wild-type AcrIF24, although the formation of the Csy–DNA complex is prevented, a new band at a higher molecular weight is observed (Extended Data Fig. 1d), indicative of the formation of a Csy–AcrIF24–DNA complex. Consistent with this finding, we were able to purify a Csy–AcrIF24–dsDNA complex by incubating the Csy–AcrIF24 complex with a dsDNA substrate followed by size-exclusion chromatography (Extended Data Fig. 2a). The bound dsDNA was only visible in the dimer form and was determined at a lower resolution (8–10 Å), due to either low abundance of this species, flexibility, or both in this region (Extended Data Fig. 7 and Extended Data Fig. 8a). Nevertheless, the major and minor grooves of the DNA are visible, allowing us to model the bound DNA as a poly G sequence.

The structure shows that the two ends of the dsDNA are bound to both the PAM recognition sites of Csy complexes, and the middle region of the dsDNA (~20 bp) is bound to the AcrIF24 dimer with the connections between the middle region and two ends barely visible (Extended Data Fig. 8a). The middle region of the dsDNA is bound to the HTH domains of the AcrIF24 dimer, which form a highly positively charged surface (Extended Data Fig. 8b). The two ends of the dsDNA are loaded into the PAM recognition sites, similar to the previously studied Csy–dsDNA complex (PDB ID: 6NE0)23. Based on our modeling, the total length of dsDNA spanning the two PAM recognition sites would be around 60–62 bp.

Since no crRNA–target DNA hybridization is observed in the Csy–AcrIF24–DNA complex, we then checked whether Csy–AcrIF24 is capable of binding non-specific DNA. Our EMSA results suggest that Csy–AcrIF24 binds to both specific and non-specific DNA (Extended Data Fig. 8c). Mutations in the PAM recognition site in Csy (K247E/N250D in Cas8f15) show no detectable binding to either specific or non-specific DNA, suggesting that DNA binding to Csy–AcrIF24 relies on this recognition site (Extended Data Fig. 8d). To check whether the specific and non-specific DNA share the same binding site on the Csy–AcrIF24 complex, we performed competition EMSA experiments. Our results show that non-specific DNA can replace specific DNA in the Csy–AcrIF24–DNA complex and vice versa (Extended Data Fig. 8e,f), suggesting they share the same binding site. Taken together, Csy–AcrIF24 is capable of recruiting DNA independent of the PAM and spacer sequences, possibly through interaction with the charged surfaces and the PAM recognition sites.

Both AcrIF24 and Csy–AcrIF24 bind tightly to promoter DNA

We next explored the function of AcrIF24 as an Aca protein. We constructed a plasmid with the acrIF23-acrIF24 promoter upstream of a promoterless lacZ reporter gene, a strategy previously used to test the function of Aca119. Expression of AcrIF24 resulted in a 64% reduction in β-galactosidase activity (Fig. 4a), indicating that AcrIF24 is a transcriptional suppressor. Based on sequence information upstream of the acrIF23-acrIF24 operon, we synthesized a 66-bp dsDNA that covers the predicted –10 and –35 regions of the promoter (predicted by BPROM24) and an inverted repeat (IR) (Fig. 4b). EMSA results suggest that AcrIF24 specifically binds to this promoter DNA (KD: ~ 34 nM) but not to a mutant with interrupted IR sequences (Fig. 4c,d). Csy–AcrIF24 shows a slightly higher affinity to promoter DNA (KD: ~ 20 nM) than AcrIF24 alone (Fig. 4d,e), possibly due to added interactions contributed by the PAM recognition sites on Csy.

Figure 4. AcrIF24 and Csy–AcrIF24 tightly bind to Acr promoter DNA.

Figure 4.

Open in a new tab

(a) β-galactosidase activity assay showing transcriptional repression by AcrIF24. The activity of β-galactosidase in the presence of an acrIF23-acrIF24 promoter is calibrated to 100%. The activities of β-galactosidase are shown as mean ± SD; n = 3. Two-sided t-test was performed (*P < 0.05). (b) Sequence of the 66-bp promoter DNA and a mutant version of it where inverted repeat sequences are interrupted. (c) EMSA results showing that AcrIF24 binds to the promoter DNA but not to the mutant version. (d) Promoter DNA binding curves of AcrIF24 and Csy–AcrIF24. The fraction of DNA bound is determined from EMSA results in c and e. Calculated KD values of AcrIF24 and Csy–AcrIF24 to promoter DNA are ~ 34 nM and ~ 20 nM, respectively. Data are presented as mean values ± SD; n = 3. (e) EMSA results showing that Csy–AcrIF24 binds to the promoter DNA. (f) EMSA results showing the effect of AcrIF24 mutations on promoter DNA binding.

Mutations of key residues in the dimerization interface (e.g. H154A/H216A/Y217A/R221A) abolished promoter DNA binding, indicating dimerization of AcrIF24 is required for binding to promoter DNA (Fig. 4f). Interestingly, the HTH domain of AcrIF24 alone is not capable of binding promoter DNA (Fig. 4f), nor the HTH domain alone shows transcriptional repression in the β-galactosidase activity assay (Fig. 4a). It is likely that other domains such as NTD are important for maintaining the native structure and dimeric form of AcrIF24 to fulfill its Aca function.

Structure of Csy–AcrIF24–promoter DNA

To reveal the structural detail of the interaction between AcrIF24 and the promoter DNA, we determined the structure of the Csy–AcrIF24–promoter DNA complex by cryo-EM. Both the Csy–AcrIF24 monomer and dimer states are bound to promoter DNA and both states were determined to around 2.7 Å, which allowed for unambiguous assignment of the bases in the promoter DNA (Fig. 5a,b and Extended Data Fig. 9, and Supplementary Table 1). A 19 bp promoter DNA spanning the predicted –10 and –35 regions was modeled confidently to the EM density map (Figs. 5c,d and Extended Data Fig. 10a). The IR in this region is involved in direct association with both AcrIF24 protomers. Specifically, a G-C base pair at position 16 is recognized by R194 of AcrIF24.1 (similarly, a G-C base pair at position 4 is recognized by R194 of AcrIF24.2) (Fig. 5e). G-C and A-T base pairs at positions 13 and 14 are recognized by K197 and T193 of AcrIF24.1 (similarly, A-T and G-C base pairs at positions 6 and 7 are recognized by K197 and T193 of AcrIF24.2) (Fig. 5e). Binding of the promoter DNA to Csy–AcrIF24 results in bending in the DNA by 43.7° from position 2 to 18 (calculated by Curve+ program25), similar to ~ 42° bending angle observed from the Aca1–promoter DNA structure22 (Extended Data Fig. 10b). Extra density potentially contributed by DNA flanking the –10 and –35 regions is observed at both PAM recognition sites of Csy; however, the densities are weak, indicative of weak and non-specific interactions (Fig. 5a, b). These results highlight how Csy–AcrIF24 can tightly bind to the acrIF23-acrIF24 promoter.

Figure 5. Cryo-EM structure of Csy–AcrIF24–promoter DNA complex.

Figure 5.

Open in a new tab

(a,b) Cryo-EM map of Csy–AcrIF24–promoter DNA in monomer (a) and dimer (b) forms in surface representation with each subunit color-coded as in Fig. 1a. Two strands of DNA are color-coded in blue and magenta, respectively. Unmodeled density at the two ends of promoter DNA is colored brown. (c) Sequence of the promoter DNA that interacts with AcrIF24 dimer. Inverted repeats are indicated by red color. The base pairs involved in specific interaction with AcrIF24 are indicated by arrows. (d) Structure of AcrIF24–promoter DNA extracted from Csy–AcrIF24–promoter DNA dimer. (e) Specific interaction of promoter DNA with AcrIF24. Three residues (T193, R196, and K197) from each AcrIF24 protomer are involved in the recognition of promoter DNA. Interactions (hydrogen bonds) are indicated by red dashed lines.

DISCUSSION

In this study, we elucidated the structural basis for the dual function of AcrIF24, a fused Acr-Aca protein in type I-F CRISPR-Cas systems or more broadly in class 1 CRISPR-Cas systems, and reported a high-resolution structure of the Csy–AcrIF24–promoter complex. As an Acr, AcrIF24 directly binds to Csy and blocks the hybridization between crRNA and target DNA, thereby inactivating the type I-F CRISPR-Cas system (Fig. 6). Target recognition by Csy involves rapid association and dissociation with dsDNA, searching for a GG PAM sequence26. Once the PAM sequence is identified, DNA is partially melted, which allows the hybridization between melted DNA and the crRNA. When the crRNA-DNA hybridization is successful and a stable R-loop structure is formed, the Csy-dsDNA complex is capable of recruiting Cas2/3 nuclease for target cleavage. If not, Csy’s search for PAM and spacer sequences continues. AcrIF24 binds to the Cas7 ‘backbone’ and prevents the hybridization between target DNA and crRNA, therefore preventing the formation of the R-loop structure. This mechanism is reminiscent of AcrIF17,8, AcrIF911,12, and AcrIF1413,15.

Figure 6. Model of the dual function of AcrIF24.

Figure 6.

Open in a new tab

A cartoon model showing the Acr and Aca function of AcrIF24. AcrIF24 exists as a dimer (1), which binds to and dimerizes Csy and prevents hybridization between target DNA and crRNA (2). AcrIF24, alone or in complex with Csy, binds to the promoter DNA (3) and suppresses the transcription of the acrIF23–acrIF24 operon (4).

As an Aca, AcrIF24 is capable of binding to the promoter DNA, thereby suppressing the transcription from the acrIF23-acrIF24 promoter (Fig. 6). The Aca role of AcrIF24 is consistent with a recent study27. Here, we present a high-resolution structure of Csy–AcrIF24–promoter complex to reveal the structural basis for the Aca role of AcrIF24. In the acrIF23-acrIF24 promoter, only one IR sequence between the −35 and −10 regions is identifiable. This IR sequence is in an equivalent position to the IR2 of the acrIF1-aca1 promoter, in which another IR1 is found upstream of the −35 region. The Aca role of Acr proteins was also reported in two studies investigating type II Acrs, including AcrIIA128, and AcrIIA13–1529, The N-terminal domain of AcrIIA1 is a critical transcriptional repressor of the strong acr promoter, whereas and the C-terminal domain inhibits DNA interference by Cas928. The N-termini of AcrIIA13–15 contain HTH motif and are capable of binding the IR regions of the acr promoter29. The fact that dual-function inhibitors could span both class 1 and class 2 CRISPR-Cas systems possibly suggests that this could be a widely successful strategy for Acrs.

Although there are three different domains in AcrIF24 responsible for separate Acr and Aca roles, AcrIF24 functions as a whole with the dimerization interface being essential for both roles. For example, although both NTD and MID directly bind to the Cas7f backbone of Csy to block the hybridization between crRNA and target DNA, HTH is required for optimal inhibition of Csy (Fig. 2b). On the other hand, although only the HTH domain is involved in DNA binding as revealed by the structures reported here, the HTH domain alone is not capable of competent binding to promoter DNA (Fig. 4f). One possible explanation is that all domains are required for maintaining the native structure and dimeric form of AcrIF24 for the dual function of Acr and Aca. In summary, our work reveals the structural basis for the dual function of AcrIF24, a fused Acr-Aca protein that inhibits the type I-F CRISPR–Cas system.

MATERIALS AND METHODS

Plasmid constructs and protein purification

The DNA sequences encoding AcrIF23 and AcrIF24 were ordered as gBlocks from Integrated DNA Technologies, Inc. (IDT). The sequnces were cloned individually into pET His6 Sumo TEV LIC cloning vectors (1S) (Addgene # 29659) using the Gibson Assembly® Master Mix (NEB, Cat. # E2611S). After sequencing verification, these plasmids were transformed into E. coli BL21(DE3) cells for expression in the terrific broth (TB) medium. Protein expression was induced when the optical density at 600 nm (OD600) reached 0.6 by adding 0.5 mM isopropyl β-D-1-thiogalactopyranoside (IPTG) followed by incubation at 16 °C overnight. After harvesting, cell pellets were resuspended in lysis buffer (50 mM HEPES (pH 7.5), 500 mM NaCl, 5% glycerol, 5 mM β-mercaptoethanol) supplemented with 0.2 mM phenylmethylsulfonyl fluoride (PMSF). Cells were disrupted by sonication. After centrifugation, the supernatant was incubated with Ni-NTA beads for 30 minutes and loaded to a gravity flow column followed by an extensive wash using lysis buffer. The target AcrIF24 (or AcrIF23) protein was eluted in a buffer containing 50 mM HEPES (pH 7.5), 500 mM NaCl, 5% glycerol, 1 mM tris(2-carboxyethyl) phosphine hydrochloride (TCEP), and 350 mM imidazole. After digestion by TEV protease overnight to remove the 6XHis and SUMO-tags, AcrIF24 (or AcrIF23) protein was purified with ion-exchange chromatography using HiTrap Q HP. AcrIF24 (or AcrIF23) protein was then concentrated and further purified over a Superdex 200 or a Superdex 75 column (Cytiva) in a buffer containing 20 mM HEPES (pH 7.5), 150 mM NaCl, 5% glycerol, and 1 mM TCEP.

For purification of the PA14 Csy complex, we used a similar protocol as described for AcrIF24. For purification of Cas2/3, the pMBPCas2/3 plasmid (Addgene ID#89238) was transformed into BL21(DE3) cells. Protein expression was induced as described above for AcrIF24. After harvesting, cell pellets were resuspended in lysis buffer (50 mM HEPES (pH 7.5), 500 mM NaCl, 5% glycerol, 1 mM TCEP) supplemented with 0.2 mM PMSF, and disrupted by sonication. After centrifugation, the supernatant was passed through a StrepTrap HP column for binding the tagged Cas2/3 followed by washing with lysis buffer. The protein was eluted with lysis buffer supplemented with 2.5 mM desthiobiotin. The collected elution fractions were concentrated and subjected to a Superdex 200 column (Cytiva) for further purification.

Mutagenesis

All AcrIF24 and Csy mutations were generated by PCR-based QuickChange site-directed mutagenesis. All mutations were confirmed by DNA sequencing. The primers used for making these mutations are listed in Supplementary Table 2. The mutant proteins were expressed and purified using the same protocol as used for the wild-type proteins.

Substrate DNA preparation

To prepare dsDNA substrate for in vitro cleavage assays and EMSAs, a 54-nt non-target DNA strand labeled with 5′-fluorescein (FAM) (synthesized by Genewiz) and a 54-nt target strand without a label (synthesized by IDT) were mixed at a molar ratio of 1:1.25, denatured at 95 °C for 5 minutes, and then annealed at room temperature. The DNA sequences are listed in Supplementary Table 2.

Complex assembly

To assemble the Csy–AcrIF24 complex, purified Csy was incubated with AcrIF24 at a molar ratio of 1:10 for 1 hour on ice. The mixer was then subjected to a Superdex 200 column (Cytiva) equilibrated in buffer containing 20 mM HEPES (pH 7.5), 100 mM KCl, 5% glycerol and 1 mM TCEP. Elution fractions contatining the assembled complex were examined by SDS-PAGE analysis. To assemble the Csy–AcrIF24–dsDNA complex, Csy was incubated with a 10-fold molar excess of AcrIF24 and 1.5 molar excess of target dsDNA for 1 hour on ice. The mixer was then subjected to a Superdex 200 column (Cytiva) equilibrated in buffer containing 20 mM HEPES (pH 7.5), 100 mM KCl, 5% glycerol and 1 mM TCEP, 5 mM MgCl2. Elution fractions were examined by SDS-PAGE and 15% polyacrylamide gels containing 7 M urea.

In vitro DNA cleavage assay

Reactions were typically performed with final concentrations of Csy at 1.2 μM, Cas2/3 at 0.15 μM, and substrate DNA at 0.04 μM. The concentrations of AcrIF24 and its variants are specified in the figures or legends.

We first incubated AcrIF24 with Csy complex at 4 °C in the buffer containing 20 mM HEPES (pH 7.5), 100 mM KCl, 5% glycerol, and 1 mM TCEP for 30 minutes. Substrate DNA was added and incubated for another 15 minutes. Then, Cas2/3 was introduced along with 5 mM MgCl2, 75 μM NiSO4, 5 mM CaCl2, and 1 mM ATP. After 30 minutes at 37 °C, the reaction was quenched with 1% SDS and 8 mM EDTA. The products were separated by electrophoresis over 15% TBE polyacrylamide gels containing 7 M urea and visualized by fluorescence imaging using a GE Amersham Imager 600 system.

For quantification, the intensities of uncleaved substrate DNA were quantified using ImageJ software30. DNA input was calibrated to 100% and uncleaved substrate DNA in each lane was measured. Then the percentage of cleaved DNA was calculated and illustrated as bar graph using GraphPad Prism.

Cas2/3 recruitment assay

The Cas2/3 recruitment assay was performed by incubating 1.2 μM Csy complex with 1.2 μM AcrIF23 or AcrIF24 for 30 minutes at 4°C in a buffer containing 20 mM HEPES (pH 7.5), 150 mM KCl, 1 mM MgCl2 and 5% glycerol. Substrate DNA (0.04 μM) was then introduced into the reaction mixture and further incubated for 15 minutes at 37 °C. Cas2/3 was then added to the reaction mixture and was further incubated for 30 minutes at 37 °C. Products of the reaction were separated using 10% TBE polyacrylamide gels and visualized by fluorescence imaging.

Electrophoretic mobility shift assays (EMSA)

To test the effects of AcrIF24 (or its variants) on the recognition of substrate DNA by Csy (Extended Data Fig. 5b,d,f), 1.2 μM Csy complex was incubated with AcrIF24 (or its variants) in a 1:1 molar ratio at 4 °C for 30 minutes. Then, 40 nM DNA substrate was added and incubated at 4 °C for another 30 minutes. Products were separated using 6% TBE polyacrylamide gels and visualized by fluorescence imaging. All binding reactions were conducted in buffer A containing 20 mM HEPES (pH 7.5), 150 mM KCl, 1 mM MgCl2, and 5% glycerol.

To test DNA binding to Csy–AcrIF24 (Extended Data Fig. 8c,d), 20 nM 5’-FAM-labeled DNA was incubated with increasing concentration of Csy–AcrIF24 (1, 20, 50, 100, 150, 200, 250, 300, 600, 1200, 1500 nM) in buffer A at 4 °C for 30 minutes. Products were separated using 6% TBE polyacrylamide gels and visualized by fluorescence imaging.

In competition EMSA experiments (Extended Data Fig. 8e,f), Csy and AcrIF24 were incubated in a 1:1 molar ratio at 4 °C for 30 minutes, followed by the addition of 0.25 μM FAM-labeled specific dsDNA (dsDNAsp), and the reaction mixture was incubated at 37°C for 30 minutes in buffer A. Then the nonspecific dsDNA (dsDNAns with no FAM label) was added at increasing concentrations (0.125, 0.25, 0.5, 1 μM) and incubated at 4°C for another 30 minutes. Products were separated using 6% TBE polyacrylamide gels and visualized by fluorescence imaging.

To calculate the affinities between AcrIF24 (or Csy–AcrIF24) and the promoter DNA (Fig. 4c,e), 20 nM 5’-FAM-labeled wild type or mutant promoter DNA was incubated with increasing concentrations (1, 5, 10, 15, 20, 40, 60, 80, 1000, 1250, 1500 nM) of AcrIF24 (or Csy–AcrIF24) in the buffer A for 30 minutes at 4°C. Products were separated using 10% (for AcrIF24) and 6% (for Csy–AcrIF24) TBE polyacrylamide gels and visualized by fluorescence imaging.

AcrIF24 mutant DNA binding assay (Fig. 4f), 20 nM 5’-FAM-labeled promoter DNA was incubated with an increasing concentration of AcrIF24 mutants (1.5, 3, 6 μM) in buffer A for 30 minutes at 4°C. Products were separated using 10% TBE polyacrylamide gels and visualized by fluorescence imaging.

The quantification of the EMSA data was performed by measuring the intensities of free DNA bands using ImageJ30 with DNA input calibrated at 100%. The percentage of bound DNA was calculated and illustrated as bar graphs. For calculation of binding affinity, the binding curves were fitted using the ‘specific binding with Hill slope’ function in GraphPad Prism.

β-galactosidase assay

A promoterless β-galactosidase reporter shuttle vector pQF50 and a vector with constitutive promoter pCM-Str were gifts from Dr. Alan R. Davidson Lab18. The whole plasmids were sequenced for confirmation and construction purposes. AcrIF24 was PCR amplified from its gBlock using primers listed in Supplemental Table 2 and cloned into pCM-Str using Gibson isothermal assembly. To insert the promoter DNA to pQF50, a pair of primers with 30 nt overhangs matching the promoter DNA are designed. pQF50 plasmid was PCR amplified using the primers. PCR products were cleaned up with a PCR clean-up kit (Qiagen) and then DpnI digested to remove the original pQF50 with no promoter insertion. Blunt ends of the PCR product were annealed using the Blunt/TA Ligase Master Mix (New England Biolabs Inc.). The insertion of promoter and acrIF24 gene was sequence verified.

The modified pQF50 and pCM-Str plasmids designed for the β-galactosidase assay were transformed into competent DH5α E. coli cells. A single colony was selected and cultured overnight. The cells were then diluted by 1:100 and grown for ~ 3 hours with shaking till the OD600 reached ~ 0.6. 20 μL of culture was mixed with 80 μL of permeabilization solution (0.8 mg mL−1 CTAB, 0.4 mg mL−1 sodium deoxycholate, 100 mM Na2HPO4, 20 mM KCl, 2 mM MgSO4, 5.4 μL mL−1 β-mercaptoethanol) and incubated at 30 °C for 30 minutes. 600 μL of substrate solution (60 mM Na2HPO4, 40 mM NaH2PO4, 1 mg mL−1 o-nitrophenyl-β-galactosidase) was added and the reaction was allowed to proceed at 30 °C for 30 minutes. The reaction was stopped with the addition of 700 μL of 1 M Na2CO3, absorbance at 420 nm (A420) was measured, and Miller units were calculated. The β-galactosidase activity was then quantified using a method derived from31.

Size-exclusion chromatography small-angle X-ray scattering (SEC-SAXS)

Purified samples were subjected to SEC-SAXS with or without multi-angle light scattering (MALS) using either a Superdex 200 Increase GL or Superose 6 column (Cytiva). The scattering data were collected at beamline 18-ID of the Biophysics Collaborative Access Team (BioCAT) of the Advanced Photon Source, Argonne National Laboratory. Briefly, for SEC-SAXS the sample(s) were analyzed using an ATKA Pure (Cytiva) FPLC, while for SEC-MALS-SAXS, the sample was analyzed using a 1260 Infinity II HPLC (Agilent Technologies) in line with a MALS+DLS detector (DAWN Helios II, Wyatt Technologies) and an RI detector (Optilab T-rEX, Wyatt Technologies). The SAXS data were collected as the sample was passed through a co-flow cell using a Pilatus3 X 1M detector (Dectris). The SAXS data were analyzed, and final plots were made using BioXTAS RAW32 and ATSAS33. The data were reduced and the data range for scattering curves was selected. Upon averaging the data, the q-range and molecular weight information were obtained by Guinier analysis. The pair-distance distribution curves were calculated using GNOM. Theoretical scattering for the X-ray and cryo-EM structures were calculated and compared with experimental scattering curves using Crysol, with structural superpositions performed with SUPCOMB34. Analysis of the MALS data was performed using the ASTRA 7 software (Wyatt Technologies).

Electron microscopy

Aliquots of 3 μL Csy–AcrIF24 or Csy–AcrIF24–DNA complexes at ~2 mg mL−1 were applied to glow-discharged UltrAuFoil holey gold grids (R1.2/1.3, 300 mesh). The grids were prepared using a ThermoFisher Scientific Mark IV Vitrobot and transferred into a Titan Krios microscope operated at 300 kV. Cryo-EM data were collected at a nominal magnification of 81,000x, resulting in a calibrated physical pixel size of 1.05 Å/pixel. The images were recorded on a K3 electron direct detector in super-resolution mode at the end of a GIF-Quantum energy filter operated with a slit width of 20 eV. Imaged were collected using Leginon35 with a defocus range of 1.2–2.5 μm. A dose rate of 20 electrons per pixel per second and an exposure time of 3.12 seconds were used, generating 40 movie frames with a total dose of ~ 54 electrons per Å2. Statistics for cryo-EM data are listed in Supplemental Table 1.

The Csy–AcrIF24–promoter DNA complex was studied similarly, except for that the data were collected on a Titan Krios G4 microscope and the images have a calibrated physical pixel size of 1.054 Å/pixel.

Image processing

As for the Csy–AcrIF24 dataset, the movie frames were imported to cryoSPARC36. Movie frames were aligned patch motion correction with a binning factor of 2. Contrast transfer function (CTF) parameters were estimated using Patch CTF37. A few thousand particles were auto-picked without a template to generate 2D averages for subsequent template-based auto-picking. The auto-picked and extracted particles were screened by 2D classification and selection of to exclude false and bad particles that fall into 2D averages with poor features. A subset of the particles (100,000 particles) were used to generate an initial model in cryoSPARC.

The screened particles were imported to RELION-338 for 3D classification, using an initial model generated by cryoSPARC. All three resultant 3D classes contain AcrIF24; however, class 2 lacks the Cas8–Cas5 heterodimer, and class 3 shows poor features (Extended Data Fig. 3f). Particles in class 1 (111,955) were further refined in RELION, resulting in a map at an overall resolution of 3.2 Å based on the FSC = 0.143 criterion. Focused refinements around the inhibitors were further performed to improve the local resolutions. A combined map was calculated by the vop maximum program in Chimera39 and used for model building. A workflow for data processing is summarized in Extended Data Fig. 3.

A similar workflow was used for the Csy–AcrIF24–DNA and Csy–AcrIF24–promoter DNA datasets, except for that all processing was done in cryoSPARC (Extended Data Figs. 7 and 9).

Model building and refinement

De novo model building of the AcrIF24 structures was performed manually in COOT40. Secondary structure predictions by PSIPRED41 were used to assist manual building. To build models of Csy, each subunit of the structure of the Csy–AcrIF4 complex (PDB: 6NE0) was fitted into the maps as a rigid-body in UCSF Chimera39 and manually adjusted in COOT. Refinement of the structure models against corresponding maps was performed using phenix.real_space_refine tool in Phenix42. The buried interface area was calculated by the PDBePISA web server. The AcrIF23 model used for fitting the SEC-SAXS bead model was generated by AlphaFold243. The poly G model was built by the ‘other modeling tools’ in COOT. Specifically, a double-stranded poly G structure was built in an ideal B-form. The structure was fitted onto the cryo-EM map as a rigid body using the ‘fit in map’ program in UCSF Chimera. The model was then fitted in COOT with all molecule self-restrains to maintain hydrogen bonds and stacking interactions between bases.

Visualization

Figures were generated using PyMOL and UCSF Chimera39.

Extended Data

Extended Data Fig. 1. AcrIF23 and AcrIF24 inhibit in vitro DNA cleavage by Csy and Cas2/3.

Extended Data Fig. 1

Open in a new tab

(a) Cartoon diagram of the acrIF23-acrIF24 operon found in a P. aeruginosa prophage. No aca gene downstream of the acrIF23–24 locus was found. In comparison, the aca9 gene was found downstream of acrIF23 in a D. carbinolicus prophage. Adapted from6. (b) AcrIF23 and AcrIF24 inhibits substrate DNA cleavage by the type I-F CRISPR-Cas system in the in vitro DNA cleavage assay. (c) Quantitative result of the experiment in (b). Error bars represent SD; n = 3. Two-sided t test was performed (*P < 0.05). (d) EMSA results reveal that AcrIF24 inhibits substrate DNA binding to Csy whereas AcrIF23 does not. Note: Csy–AcrIF24 is capable of binding DNA independent of PAM and spacer sequences, resulting in a high molecular weight shift. * indicates a faint band where two copies of Csy bound to DNA, similar to previous report14 (doi.org/10.1093/nar/gkab092). (e) Quantitative result of the experiment in (d). Error bars represent SD; n = 3. Two-sided t test was performed (*P < 0.05). (f) EMSA results reveal that Cas2/3 recruitment by Csy is abolished by AcrIF24 but not by AcrIF23.

Extended Data Fig. 2. Biochemical and biophysical analysis of AcrIF23 and AcrIF24.

Extended Data Fig. 2

Open in a new tab

(a) Size exclusion chromatography (SEC) profiles of Csy, Csy–AcrIF24, Csy(Cas8f K247E/N250D)–AcrIF24 and Csy–AcrIF24–DNA complexes. Fractions of Csy–AcrIF24–DNA as indicated were analyzed by SDS-PAGE (to visualize proteins) and 10% Native PAGE gel (to visualize DNA), results of which are shown below the SEC profiles. SDS-PAGE of fractions of Csy(Cas8f K247E/N250D)–AcrIF24 is also shown below the SEC profiles. (b) SEC profiles (top) and SDS-PAGE (bottom) of AcrIF24 and its domain deletion mutants. (c) SEC profiles (left) and SDS-PAGE (right) of AcrIF24 mutants.

Extended Data Fig. 3. Cryo-EM data processing workflow for Csy–AcrIF24.

Extended Data Fig. 3

Open in a new tab

(a) A representative raw cryo-EM micrograph of Csy–AcrIF24. (b) Representative 2D class averages. (c–e) Three major 3D classes were identified from 2D classification and heterogenous refinement: Csy–AcrIF24 (c), Csy–AcrIF24 dimer (d), and AcrIF24 only (e). (f) Major 3D classes from the 3D classification of Csy–AcrIF24. Class 2 shows missing density for Cas5f, and Cas8f subunits. (g) 3D refinement for particles from the 3D classification of Csy–AcrIF24 as indicated. (h) The plot of the global half map FSC of Csy–AcrIF24 indicates an average resolution of 3.2 Å. (i) Two focused refinements with a soft mask either in the head or tail of the Csy complex to improve local resolutions. (j) The final Csy–AcrIF24 map generated by combining maps from focused refinement. (k) 3D refinement for particles from the 3D classification of Csy–AcrIF24 dimer as indicated. (l) The plot of the global half map FSC of Csy–AcrIF24 dimer indicates an average resolution of ~ 3.5 Å. (m) Two focused refinements with a soft mask to each Csy complex bound to AcrIF24 to improve local resolutions. (n) The final Csy–AcrIF24 dimer map generated by combining maps from focused refinement.

Extended Data Fig. 4. Structure of Csy–AcrIF24.

Extended Data Fig. 4

Open in a new tab

(a) Structure of AcrIF24 protomer color-coded as in Fig. 2a. EM density map is shown in orange mesh, and the atomic model of AcrIF24 is shown in sticks. (b) The protein sequence and secondary structure of AcrIF24. Residues involved in the dimeric interaction and tested by mutagenesis are marked with triangles. (c) Structural superimposition between AcrIF24 NTD domain and MMOD (PDB: 6D7K). (d) Interaction between HTH and NTD, with hydrophobic residues in the interface highlighted. (e)Structural superimposition between AcrIF24 HTH domain and Aca1 (PDB: 7CJK). (f) MID-MID dimerization interface of AcrIF24 as in Fig. 2e, with corresponding EM density shown in mesh. (g) HTH-HTH dimerization interface of AcrIF24 as in Fig. 2h, with corresponding EM density shown in mesh. (h) Interaction between AcrIF24-NTD and Csy as in Fig. 3c, with corresponding EM density shown in mesh. (i) Interaction between AcrIF24-MID and Csy as in Fig. 3e, with corresponding EM density shown in mesh.

Extended Data Fig. 5. Effect of AcrIF24 mutants on DNA cleavage and binding by Csy.

Extended Data Fig. 5

Open in a new tab

(a,b) In vitro DNA cleavage assays (a) and EMSAs (b) with Csy and target DNA in the presence of wild-type or domain deletion mutants of AcrIF24. (c,d) In vitro DNA cleavage assays (c) and EMSAs (d) with Csy and target DNA in the presence of wild-type AcrIF24 or AcrIF24 with mutations of residues in the dimerization interfaces. (e,f) In vitro DNA cleavage assays (e) and EMSAs (f) in the presence of wild-type AcrIF24 or AcrIF24 with mutations of residues involved in Csy interaction.

Extended Data Fig. 6. SAXS analysis of AcrIF23 and AcrIF24.

Extended Data Fig. 6

Open in a new tab

(a) SAXS assay of purified wild type AcrIF24 in solution. The experimental SAXS scattering curve for AcrIF24 is shown as gray points, while calculated curves for the monomeric (red) and dimer (blue) forms of AcrIF24 are shown as solid lines. The fit of each calculated curve to the experimental data is indicated by χ2 values of 194 and 1.4, respectively. (b) A superposition of the AcrIF24 dimer with an ab initio bead model from the SAXS analysis (DAMMIF/N). (c) The ‘estimated molecular weight’ of AcrIF24 as the average of Vc, Vp, Bayes, and Shape/Size. (d) The experimental SAXS scattering curve for AcrIF23 is shown as gray points, while calculated curve for the AcrIF23 monomer model (AlphaFold) (green) is shown as a solid line. The fit of the calculated curve to the experimental data is indicated by the χ2 value of 1.2. (e) A superposition of the AcrIF23 monomeric model (AlphaFold) with the ab initio bead model from the SAXS analysis (DAMMIF/N). (f) The ‘estimated molecular weight’ of AcrIF23 as the average of Vc, Vp, Bayes, and Shape/Size. AcrIF23 has a theoretical molecular weight of 17.5 kDa. (g) SEC-SAXS scattering curve (gray; Dmax of 75.0 Å; Rg of 25.3 Å; MWavg of 36 kDa) for the AcrIF24(H154A/H216A/Y217A/R221A) mutant aligned with the calculated scattering curve for the AcrIF24 monomer (red, χ2 of 1.3). The residual plot from the fit of the experimental data (gray) and fit of the model (red) is shown in the bottom panel. (h) SEC-MALS analysis of the AcrIF24(H154A/H216A/Y217A/R221A) mutant showing a calculated molecular weight consistent with a monomer.

Extended Data Fig. 7. Cryo-EM data processing workflow for Csy–AcrIF24–DNA.

Extended Data Fig. 7

Open in a new tab

(a) A representative raw cryo-EM micrograph of the Csy–AcrIF24–dsDNA complex. (b) Representative 2D class averages. (c,d) Two 3D classes separated by heterogenous refinement: Csy–AcrIF24 in monomer (c) and dimer (d) states. (e) Three major classes from the heterogenous refinement of Csy–AcrIF24 monomer. (f) Refined map of Csy–AcrIF24 in monomer state. (g) Homogenous refinement for particles from the dimer state. (h) The plot of the global half map FSC of Csy–AcrIF24–dsDNA indicates an average resolution of 3.62 Å. (i) Refined map of Csy–AcrIF24–dsDNA after applying Gaussian filter to show the weak density of DNA.

Extended Data Fig. 8. Csy–AcrIF24 dimer recruits DNA independent of PAM and spacer sequences.

Extended Data Fig. 8

Open in a new tab

(a) Cryo-EM structure of Csy–AcrIF24 dimer bound to dsDNA. Csy–AcrIF24 subunits are color-coded as Fig. 1a. Two strands of DNA are color-coded in blue and magenta, respectively. (b) Electrostatic potential maps of Csy–AcrIF24 dimer (left) and AcrIF24 dimer (right), showing that DNA is bound to the positively changed surface of the assembly. (c) EMSA assay showing that Csy–AcrIF24 binds to both specific DNA (dsDNAsp) and non-specific DNA (dsDNAns) without a PAM or a spacer sequence. Sequences of dsDNAsp and dsDNAns are shown at the top, with the spacer highlighted and the GG PAM sequence colored in red. (d) EMSA assay showing that Csy (Cas8f K247E/N250D)–AcrIF24 is not capable of binding to both dsDNAsp or dsDNAns. (e,f) Competition EMSA assay to test whether the binding sites of dsDNAsp and dsDNAns overlap. In e, Csy or Csy–AcrIF24 was incubated with dsDNAsp (5′-FAM in the TS), followed by adding increasing concentrations of unlabeled dsDNAns. In f, Csy or Csy–AcrIF24 was incubated with dsDNAns (5′-FAM in the TS), followed by adding increasing concentrations of unlabeled dsDNAsp.

Extended Data Fig. 9. Cryo-EM data processing workflow for Csy–AcrIF24–promoter DNA.

Extended Data Fig. 9

Open in a new tab

(a) A representative raw cryo-EM micrograph of the Csy–AcrIF24–promoter DNA complex. (b) Representative 2D class averages. (c,d) Two 3D classes separated by heterogenous refinement: Csy–AcrIF24-promoter DNA in monomer (c) and dimer (d) states. (e) Two major classes from the heterogenous refinement of Csy–AcrIF24-promoter DNA monomer. (f) Refined map of Csy–AcrIF24-promoter DNA in monomer state. (g) The plot of the global half map FSC of Csy–AcrIF24–promoter DNA in monomer state indicates an average resolution of ~ 2.7 Å. (h) Two major classes from the heterogenous refinement of Csy–AcrIF24-promoter DNA dimer. (i) Refined map of Csy–AcrIF24-promoter DNA in dimer state. (j) The plot of the global half map FSC of Csy–AcrIF24–promoter DNA in dimer state indicates an average resolution of ~ 2.7 Å.

Extended Data Fig. 10. Structure of AcrIF24–promoter DNA.

Extended Data Fig. 10

Open in a new tab

(a) Modeling of the 19-bp promoter DNA bound to AcrIF24. Shown are individual base pairs of the 19-bp promoter DNA, with corresponding EM densities in mesh. (b) Structural comparison between AcrIF24–promoter DNA and Aca1–promoter DNA.

Supplementary Material

1834880_Sup_Info
1834880_SD_ED_Fig_1
1834880_SD_Fig_4
1834880_SD_ED_Fig_5
1834880_SD_ED_Fig_2
1834880_RS
1834880_SD_ED_Fig_8

ACKNOWLEDGMENTS

We thank Thomas Klose and Jingchuan Sun for help with cryo-EM, Steven Wilson for computation, Robert V. Stahelin for providing access to the GE Amersham Imager 600 system, Alan R. Davidson for sharing plasmids, Jesse Hopkins for his assistance with the SAXS and MALS experiments and analyses, and other beamline staff of the BioCAT beamline at the Advanced Photon Source (APS), Argonne National Laboratory for their help with data collection. This work was supported by the NIH grant R01GM138675 and a Core Pilot grant from Indiana Clinical and Translational Sciences Institute (CTSI) to L.C. C.G. is supported by a grant from the NIH (T32GM132024). The BioCAT resources are supported by grant P30 GM138395 from the NIGMS of the NIH; the use of the Pilatus 3 1M detector was provided by grant 1S10OD018090 from NIGMS. This work made use of the Purdue Cryo-EM Facility, and resources of the Advanced Photon Source, a U.S. Department of Energy (DOE) Office of Science User Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract No. DE-AC02-06CH11357.

Footnotes

COMPETING INTERESTS

J.B.-D. is a scientific advisory board member of SNIPR Biome and Excision Biotherapeutics and a scientific advisory board member and co-founder of Acrigen Biosciences. The Bondy-Denomy lab receives research support from Felix Biotechnology. The remaining authors declare no competing interests.

Data availability

Cryo-EM reconstructions of Csy–AcrIF24, Csy–AcrIF24 dimer, Csy–AcrIF24–DNA dimer, Csy–AcrIF24–Promoter DNA, and Csy–AcrIF24–Promoter DNA dimer have been deposited in the Electron Microscopy Data Bank under the accession numbers EMD-25660, EMD-25661, EMD-25662, EMD-25789, and EMD-25788, respectively. Coordinates for atomic models of Csy–AcrIF24, Csy–AcrIF24 dimer, Csy–AcrIF24–DNA dimer, Csy–AcrIF24–Promoter DNA, and Csy–AcrIF24–Promoter DNA dimer have been deposited in the Protein Data Bank under the accession numbers 7T3J, 7T3K, 7T3L, 7TAX, and 7TAW, respectively. Structures of Csy alone and Csy–AcrIF4 are available in the Protein Data Bank under the accession numbers 6B45 and 7JZW, respectively. Source data are provided with this paper.

REFERENCES

  • 1.Makarova KS et al. An updated evolutionary classification of CRISPR–Cas systems. Nature Reviews Microbiology 13, 722 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Koonin EV, Makarova KS & Zhang F Diversity, classification and evolution of CRISPR-Cas systems. Curr Opin Microbiol 37, 67–78, doi: 10.1016/j.mib.2017.05.008 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Bondy-Denomy J, Pawluk A, Maxwell KL & Davidson AR Bacteriophage genes that inactivate the CRISPR/Cas bacterial immune system. Nature 493, 429–432, doi: 10.1038/nature11723 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Pawluk A et al. Inactivation of CRISPR-Cas systems by anti-CRISPR proteins in diverse bacterial species. Nat Microbiol 1, doi:Artn 16085 10.1038/Nmicrobiol.2016.85 (2016). [DOI] [PubMed] [Google Scholar]
  • 5.Marino ND et al. Discovery of widespread type I and type V CRISPR-Cas inhibitors. Science 362, 240–242, doi: 10.1126/science.aau5174 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Pinilla-Redondo R et al. Discovery of multiple anti-CRISPRs highlights anti-defense gene clustering in mobile genetic elements. Nat Commun 11, 5652, doi: 10.1038/s41467-020-19415-3 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Chowdhury S et al. Structure Reveals Mechanisms of Viral Suppressors that Intercept a CRISPR RNA-Guided Surveillance Complex. Cell 169, 47–57 e11, doi: 10.1016/j.cell.2017.03.012 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Guo TW et al. Cryo-EM Structures Reveal Mechanism and Inhibition of DNA Targeting by a CRISPR-Cas Surveillance Complex. Cell 171, 414–426 e412, doi: 10.1016/j.cell.2017.09.006 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Peng R et al. Alternate binding modes of anti-CRISPR viral suppressors AcrF1/2 to Csy surveillance complex revealed by cryo-EM structures. Cell Res 27, 853–864, doi: 10.1038/cr.2017.79 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Rollins MF et al. Structure Reveals a Mechanism of CRISPR-RNA-Guided Nuclease Recruitment and Anti-CRISPR Viral Mimicry. Molecular Cell 74, 132-+, doi: 10.1016/j.molcel.2019.02.001 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Hirschi M et al. AcrIF9 tethers non-sequence specific dsDNA to the CRISPR RNA-guided surveillance complex. Nat Commun 11, doi:ARTN 2730 10.1038/s41467-020-16512-1 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Zhang KM et al. Inhibition mechanisms of AcrF9, AcrF8, and AcrF6 against type I-F CRISPR-Cas complex revealed by cryo-EM. P Natl Acad Sci USA 117, 7176–7182, doi: 10.1073/pnas.1922638117 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Gabel C, Li Z, Zhang H & Chang L Structural basis for inhibition of the type I-F CRISPR-Cas surveillance complex by AcrIF4, AcrIF7 and AcrIF14. Nucleic Acids Res 49, 584–594, doi: 10.1093/nar/gkaa1199 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Lu WT, Trost CN, Muller-Esparza H, Randau L & Davidson AR Anti-CRISPR AcrIF9 functions by inducing the CRISPR-Cas complex to bind DNA non-specifically. Nucleic Acids Res 49, 3381–3393, doi: 10.1093/nar/gkab092 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Liu X et al. Insights into the dual functions of AcrIF14 during the inhibition of type I-F CRISPR-Cas surveillance complex. Nucleic Acids Research 49, 10178–10191, doi: 10.1093/nar/gkab738 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Wang XF et al. Structural basis of Cas3 inhibition by the bacteriophage protein AcrF3. Nat Struct Mol Biol 23, 868–870, doi: 10.1038/nsmb.3269 (2016). [DOI] [PubMed] [Google Scholar]
  • 17.Niu Y et al. A Type I-F Anti-CRISPR Protein Inhibits the CRISPR-Cas Surveillance Complex by ADP-Ribosylation. Mol Cell, doi: 10.1016/j.molcel.2020.09.015 (2020). [DOI] [PubMed] [Google Scholar]
  • 18.Bondy-Denomy J et al. Multiple mechanisms for CRISPR-Cas inhibition by anti-CRISPR proteins. Nature 526, 136–139, doi: 10.1038/nature15254 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Stanley SY et al. Anti-CRISPR-Associated Proteins Are Crucial Repressors of Anti-CRISPR Transcription. Cell 178, 1452–1464 e1413, doi: 10.1016/j.cell.2019.07.046 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Birkholz N, Fagerlund RD, Smith LM, Jackson SA & Fineran PC The autoregulator Aca2 mediates anti-CRISPR repression. Nucleic Acids Res 47, 9658–9665, doi: 10.1093/nar/gkz721 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Kim H et al. MMOD-induced structural changes of hydroxylase in soluble methane monooxygenase. Sci Adv 5, eaax0059, doi: 10.1126/sciadv.aax0059 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Liu Y et al. Structural basis for anti-CRISPR repression mediated by bacterial operon proteins Aca1 and Aca2. J Biol Chem 297, 101357, doi: 10.1016/j.jbc.2021.101357 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Rollins MF et al. Structure Reveals a Mechanism of CRISPR-RNA-Guided Nuclease Recruitment and Anti-CRISPR Viral Mimicry. Mol Cell 74, 132–142 e135, doi: 10.1016/j.molcel.2019.02.001 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Li RW Metagenomics and its applications in agriculture, biomedicine, and environmental studies. (Nova Science Publisher’s, 2011). [Google Scholar]
  • 25.Lavery R, Moakher M, Maddocks JH, Petkeviciute D & Zakrzewska K Conformational analysis of nucleic acids revisited: Curves+. Nucleic Acids Res 37, 5917–5929, doi: 10.1093/nar/gkp608 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Rollins MF, Schuman JT, Paulus K, Bukhari HS & Wiedenheft B Mechanism of foreign DNA recognition by a CRISPR RNA-guided surveillance complex from Pseudomonas aeruginosa. Nucleic Acids Res 43, 2216–2222, doi: 10.1093/nar/gkv094 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Yang LG et al. Insights into the inhibition of type I-F CRISPR-Cas system by a multifunctional anti-CRISPR protein AcrIF24. Nat Commun 13, doi:ARTN 1931 10.1038/s41467-022-29581-1 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Osuna BA et al. Critical Anti-CRISPR Locus Repression by a Bi-functional Cas9 Inhibitor. Cell Host Microbe 28, 23–30 e25, doi: 10.1016/j.chom.2020.04.002 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Watters KE et al. Potent CRISPR-Cas9 inhibitors from Staphylococcus genomes. Proc Natl Acad Sci U S A 117, 6531–6539, doi: 10.1073/pnas.1917668117 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Schneider CA, Rasband WS & Eliceiri KW NIH Image to ImageJ: 25 years of image analysis. Nature Methods 9, 671–675, doi: 10.1038/nmeth.2089 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Zhang X & Bremer H Control of the Escherichia coli rrnB P1 promoter strength by ppGpp. J Biol Chem 270, 11181–11189, doi: 10.1074/jbc.270.19.11181 (1995). [DOI] [PubMed] [Google Scholar]
  • 32.Hopkins JB, Gillilan RE & Skou S BioXTAS RAW: improvements to a free open-source program for small-angle X-ray scattering data reduction and analysis. J Appl Crystallogr 50, 1545–1553, doi: 10.1107/S1600576717011438 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Manalastas-Cantos K et al. ATSAS 3.0: expanded functionality and new tools for small-angle scattering data analysis. J Appl Crystallogr 54, 343–355, doi: 10.1107/S1600576720013412 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Kozin MB & Svergun DI Automated matching of high- and low-resolution structural models. Journal of Applied Crystallography 34, 33–41, doi:Doi 10.1107/S0021889800014126 (2001). [DOI] [Google Scholar]
  • 35.Suloway C et al. Automated molecular microscopy: the new Leginon system. J Struct Biol 151, 41–60, doi: 10.1016/j.jsb.2005.03.010 (2005). [DOI] [PubMed] [Google Scholar]
  • 36.Punjani A, Rubinstein JL, Fleet DJ & Brubaker MA cryoSPARC: algorithms for rapid unsupervised cryo-EM structure determination. Nat Methods 14, 290–296, doi: 10.1038/nmeth.4169 (2017). [DOI] [PubMed] [Google Scholar]
  • 37.Zhang K Gctf: Real-time CTF determination and correction. J Struct Biol 193, 1–12, doi: 10.1016/j.jsb.2015.11.003 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Zivanov J et al. New tools for automated high-resolution cryo-EM structure determination in RELION-3. Elife 7, doi: 10.7554/eLife.42166 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Pettersen EF et al. UCSF Chimera--a visualization system for exploratory research and analysis. J Comput Chem 25, 1605–1612, doi: 10.1002/jcc.20084 (2004). [DOI] [PubMed] [Google Scholar]
  • 40.Emsley P, Lohkamp B, Scott WG & Cowtan K Features and development of Coot. Acta Crystallogr D Biol Crystallogr 66, 486–501, doi: 10.1107/S0907444910007493 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Jones DT Protein secondary structure prediction based on position-specific scoring matrices. J Mol Biol 292, 195–202, doi: 10.1006/jmbi.1999.3091 (1999). [DOI] [PubMed] [Google Scholar]
  • 42.Afonine PV et al. Real-space refinement in PHENIX for cryo-EM and crystallography. Acta Crystallogr D Struct Biol 74, 531–544, doi: 10.1107/S2059798318006551 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Jumper J et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589, doi: 10.1038/s41586-021-03819-2 (2021). [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

1834880_Sup_Info
NIHMS1834880-supplement-1834880_Sup_Info.pdf (92KB, pdf)
1834880_SD_ED_Fig_1
NIHMS1834880-supplement-1834880_SD_ED_Fig_1.xlsx (1.3MB, xlsx)
1834880_SD_Fig_4
NIHMS1834880-supplement-1834880_SD_Fig_4.xlsx (3.8MB, xlsx)
1834880_SD_ED_Fig_5
NIHMS1834880-supplement-1834880_SD_ED_Fig_5.xlsx (2.1MB, xlsx)
1834880_SD_ED_Fig_2
NIHMS1834880-supplement-1834880_SD_ED_Fig_2.xlsx (4.4MB, xlsx)
1834880_RS
NIHMS1834880-supplement-1834880_RS.pdf (1.2MB, pdf)
1834880_SD_ED_Fig_8
NIHMS1834880-supplement-1834880_SD_ED_Fig_8.xlsx (2.7MB, xlsx)

Data Availability Statement

Cryo-EM reconstructions of Csy–AcrIF24, Csy–AcrIF24 dimer, Csy–AcrIF24–DNA dimer, Csy–AcrIF24–Promoter DNA, and Csy–AcrIF24–Promoter DNA dimer have been deposited in the Electron Microscopy Data Bank under the accession numbers EMD-25660, EMD-25661, EMD-25662, EMD-25789, and EMD-25788, respectively. Coordinates for atomic models of Csy–AcrIF24, Csy–AcrIF24 dimer, Csy–AcrIF24–DNA dimer, Csy–AcrIF24–Promoter DNA, and Csy–AcrIF24–Promoter DNA dimer have been deposited in the Protein Data Bank under the accession numbers 7T3J, 7T3K, 7T3L, 7TAX, and 7TAW, respectively. Structures of Csy alone and Csy–AcrIF4 are available in the Protein Data Bank under the accession numbers 6B45 and 7JZW, respectively. Source data are provided with this paper.

RESOURCES