Characterization of a recombinant thermotolerant argonaute protein as an endonuclease by broad guide utilization

Abstract

Background

Prokaryotic argonaute proteins (pAgos) play an important role in host defense in vivo. Most importantly, the thermophilic pAgos with endonuclease activity hold great potential for programmable genetic manipulation. Therefore, exploring argonaute proteins with unique enzyme properties is desired for understanding their diverse catalytic mechanisms and promoting their applications in biotechnology.

Results

The argonaute protein from archaeon Methanocaldococcus fervens (MfAgo) was cloned and overexpressed in Escherichia coli BL21 (DE3). The recombinant protein showed the expected molecular weight of ~ 85.8 kDa by SDS-PAGE. The activity assays demonstrate that MfAgo has cleavage activities toward single-stranded DNA (ssDNA) targets specifically at the site complementary to the position between nucleotides 10 and 11 of the guide strand. Interestingly, MfAgo utilizes small 5′-phosphorylated ssDNA (5′-P ssDNA), 5′-hydroxylated ssDNA (5′-OH ssDNA), and 5′-phosphorylated ssRNA (5′-P ssRNA) as the guides for catalysis. The optimal temperatures are highly dependent on the type of guide and have a range of 80–90 °C. The addition of 0.5 mM Mn²⁺, Mg²⁺ or Co²⁺ to the reaction system significantly enhanced the enzyme activity. Meanwhile, MfAgo is quite active at NaCl concentrations less than 500 mM. Furthermore, structural modeling analyses suggested that its unique wide guide-dependent activity might be related to differing multiple interactions between guides and the MID domain of MfAgo.

Conclusions

MfAgo shows efficient endonuclease activity for ssDNA cleavage. In contrast to most known pAgos, which recognize only one type of guide, MfAgo uses diverse guides, including 5′-P ssDNA, 5′-OH ssDNA, and 5′-P ssRNA, to specifically cleave targets. Characterization of MfAgo expands the understanding of catalysis in the Ago family and provides clues for future genetic manipulation.

Background

Argonaute proteins (Agos) are widely present in all domains of life and bind small DNA or RNA guides to specifically recognize—and sometimes cleave—complementary nucleic acid targets (Swarts et al. 2014a). Eukaryotic argonaute proteins (eAgos) are the key components in RNA interference (RNAi) pathways that participate in the regulation of post-transcriptional gene expression (Bartel 2009; Ketting 2011; Meister 2013; Siomi et al. 2011). Recently, pAgos, including bacterial and archaeal Agos, were supposed to participate in host defense processes by interfering with exogenous nucleic acid infection (Olovnikov et al. 2013; Swarts et al. 2014a, 2015a, b). Genome analysis of prokaryotes reveals that pAgos are more diverse than eAgos since they are distributed in ~ 30% of bacterial and ~ 10% of archaeal sequenced genomes (Koonin 2017; Makarova et al. 2009; Swarts et al. 2014b). The functions of pAgos are relatively poorly understood and need to be further investigated.

Previous structural and biochemical studies of pAgos from thermophilic prokaryotes showed that they exhibit endonuclease activity directed by nucleic acid guides, which is similar to CRISPR–Cas9 activity. The characterized pAgos showed a preference for high temperature, generally ranging from 65 to 95 °C, and a divalent cation requirement. As an endonuclease of Watson–Crick guide-target pairing, the pAgos have unique guide preferences. To date, most pAgos perform specific cleavage via one type of guide from all four natural types of small nucleic acids: 5′-P ssDNA, 5′-OH ssDNA, 5′-OH ssRNA, and 5′-P ssRNA. For example, Agos from Pyrococcus furiosus (PfAgo) (Rivas et al. 2005; Song et al. 2004; Swarts et al. 2015a), Aquifex aeolicus (AaAgo) (Rashid et al. 2007; Yuan et al. 2005, 2006), Methanocaldococcus jannaschii (MjAgo) (Willkomm et al. 2017; Zander et al. 2014, 2017), and Thermus thermophilus (TtAgo) (Sheng et al. 2014; Swarts et al. 2014a, 2015b, 2017; Wang et al. 2008a, b, 2009) cleave complementary nucleic acid targets using 5′-P ssDNA guides. Meanwhile, Agos from Marinitoga piezophila (MpAgo) (Doxzen and Doudna 2017; Kaya et al. 2016) and Thermotoga profunda (TpAgo) (Kaya et al. 2016) use the 5′-OH ssRNA guide to cleave nucleic acid targets, whereas the Rhodobacter sphaeroides Ago (RsAgo) (Miyoshi et al. 2016; Olovnikov et al. 2013) utilizes a 5′-P ssRNA guide to recognize cognate nucleic acid targets (Table 1). Notably, the structural analysis indicates that the distinct guide preferences are strongly associated with the guide 5′-end-binding pocket of the MID domain, as reasoned by the available space and hydrophobic or charged environment (Kaya et al. 2016; Wang et al. 2008b).

Table 1 Guide and target preferences of characterized pAgo proteins

Recently, the application of Agos has gradually attracted researchers’ interest. Based on PfAgo-specific dsDNA cleavage via a pair of 5′-P ssDNA guides, the artificial restriction enzymes’ (AREs) platform has been developed to enable to cleave target DNA sequences at virtually any arbitrary site (Enghiad and Zhao 2017). Further, mining and biochemical characterizations of pAgos are greatly demanded to understand their basic biological roles and to explore their application potentials for genome manipulation (Ryazansky et al. 2018).

Based on a phylogenetic tree analysis, we found a putative endonuclease pAgo from the hyperthermophilic archaeon Methanocaldococcus fervens (MfAgo). To better understand its catalytic function, MfAgo was cloned and overexpressed in E. coli BL21 (DE3) for biochemical characterization. Moreover, computational simulation analysis was carried out to reveal its nucleic acid guide preferences. Our study demonstrates that MfAgo is a unique endonuclease with DNA cleavage activity directed by a wide range of guides that provides a new enzyme resource for future gene manipulation.

Methods

Bacterial strain, plasmid, and medium

The host strain E. coli BL21(DE3) was purchased from Novagen (Madison, WI). The recombinant plasmid pET-28a(+)-MfAgo, containing the synthesized codon-optimized MfAgo gene, was constructed (Genscript, China). Luria–Bertani (LB) medium (tryptone 10 g/L, yeast extract 5 g/L, NaCl 10 g/L) was used for MfAgo expression.

Phylogenetic tree and sequence alignment

A similarity search for the PfAgo amino acid sequence was performed by BLAST in the NCBI database, and Ago sequences with high sequence identity were selected and analyzed by MEGA 7.0 software (Kumar et al. 2016) to construct a phylogenetic tree. The sequence alignments of the Ago family were carried out using ClustalW (Thompson et al. 1994). For clarity, only residues forming the active sites are displayed.

Cloning and expression of MfAgo in E. coli BL21 (DE3)

The expression vector pET-28a(+)-MfAgo was transformed into the E. coli BL21 (DE3) strain to express the recombinant MfAgo. Then, the positive clones were propagated overnight in a shaker incubator at 37 °C and 220 rpm in 5-mL LB medium containing 100 μg/mL kanamycin. After overnight incubation, the seed culture (1%) was inoculated into 1-L LB medium containing 100 μg/mL kanamycin at 37 °C and incubated at 220 rpm until an OD₆₀₀ value of 0.6–0.8 was reached. Cultures were cold-shocked by incubation in an ice bath for 15 min. Protein expression was then induced by the addition of isopropyl β-d-thiogalactopyranoside (IPTG) to a final concentration of 1 mM, followed by incubation for 16 h at 20 °C and 220 rpm in a shaker incubator. Cells were harvested for centrifugation for 20 min at 6000 rpm, and the cell pellet was collected for later purification.

Purification of recombinant MfAgo

The harvested cell pellets were resuspended in lysis buffer (20 mM Tris–HCl, 1 M NaCl, pH 8.0) and then disrupted using a high-pressure homogenizer (Gefran, Italy) at 600–800 bar for 3 min. The lysates were incubated at 65 °C for 15 min to eliminate most of the endogenous proteins of E. coli BL21 (DE3). Then, the solution was centrifuged for 30 min at 4 °C and 12,000 rpm, and then, the supernatants were loaded onto an Ni–NTA column. The N-terminal His-tagged MfAgo was eluted with elution buffer (20 mM Tris–HCl, 1 M NaCl, 200 mM imidazole, pH 8.0). Finally, size-exclusion chromatography (Superdex 200, GE) was further carried out with FPLC (Agilent, USA). The eluted recombinant protein was detected and analyzed by 15% SDS-PAGE. The concentration of purified MfAgo was measured by the bicinchoninic acid (BCA) method (Yeasen, China), and the fractions containing the protein were frozen at − 80 °C in storage buffer (20 mM Tris–HCl, 1 M NaCl, 15% (v/v) glycerol, pH 8.0).

Enzymatic characteristics of the recombinant MfAgo in vitro

Endonuclease activity assays

The ssDNA or ssRNA guides and fluorescently labeled ssDNA or ssRNA targets were synthesized commercially (Genscript, China). For activity assays, 2.5 μM purified MfAgo, 2 μM ssDNA or ssRNA guide, and 1 μM fluorescently labeled ssDNA or ssRNA target were mixed in a reaction buffer (15 mM Tris–HCl, 250 mM NaCl, 0.5 mM MnCl₂, pH 8.0) and then incubated for 35 min at 85 °C in a thermocycler (Eppendorf, GER). After high-temperature incubation, the reactions were stopped via the addition of loading buffer (95% formamide, 0.5 mmol/L EDTA, 0.025% bromophenol blue, 0.025% xylene cyanol FF) at a 1:1 ratio (v/v). Then, the samples were resolved on 16% denaturing polyacrylamide gels. The gels were visualized on a Fuji FLA7000 scanner with fluorescence measurements with a λ_ex of 535 nm and a λ_em of 595 nm. All the nucleic acids used in this study are listed in Table 2.

Table 2 Nucleic acids used in the study

$Mf$Ago activity at the varied temperatures

The effects of temperature on MfAgo activity mediated by different guides were determined at various temperatures from 55 to 99 °C. Generally, 2.5 μM MfAgo, 2 μM ssDNA or ssRNA guide, and 0.8 μM ssDNA targets were mixed in the reaction buffer and then incubated for 30 min at 54.8, 58.3, 60.7, 66.0, 71.0, 75.0, 80.0, 85.0, 90.0, 95.1, 96.9, and 98.7 °C, respectively. The samples were resolved on 16% denaturing polyacrylamide gels. Gels were stained using GelRed (Biotium, USA). The nucleic acids were visualized using a G:BOX Chemi imager (Syngene, USA) and quantitatively analyzed by Quantity One (Bio-Rad, USA).

For the stability of varied guides at high temperatures, 2 μM ssDNA or ssRNA guide was first pre-incubated in the reaction buffer at 95 °C for different times: 0, 5, 10, 15, and 30 min, respectively. Subsequently, with the addition of 2.5 μM MfAgo and 0.8 μM ssDNA targets, the cleavage assay was performed at 80 °C for 30 min. The samples were stained and visualized as above.

Effect of divalent cations on $Mf$Ago activity

The effects of divalent cation on MfAgo cleavage activity were determined using different divalent cations. For the assays, 2.5 μM MfAgo, 2 μM ssDNA or ssRNA guide, and 0.8 μM ssDNA targets were mixed in a reaction buffer (15 mM Tris–HCl, 250 mM NaCl, pH 8.0) to which different divalent cations (FeCl₂, CoCl₂, CuCl₂, MgCl₂, MnCl₂, ZnCl₂, and CaCl₂) were added to a final concentration of 0.5 mM, and the samples were then incubated for 30 min at 85 °C. The cleavage activity without the addition of a divalent cation was used as a control.

The optimal Mn²⁺ concentration for MfAgo cleavage activity was also determined using buffer with different final concentrations of Mn²⁺: 5, 10, 25, 50, 100, 250, 500, 1000, 2000, and 2500 μM, respectively. All the samples were stained and analyzed as above.

Effect of NaCl concentration on $Mf$Ago activity

The influence of NaCl concentration on the catalytic activity of recombinant MfAgo was investigated using buffer systems with various NaCl concentrations (50, 100, 250, 500, 750, 1000, 1500, and 2000 mM, respectively). The samples were stained and analyzed as above.

Kinetic performance of $Mf$Ago mediated by different guides

For a cleavage kinetic analysis, the concentrations of MfAgo, ssDNA targets, and different guides were the same as above. The assays were performed at 250 mM NaCl and 2 mM Mn²⁺ at 85 °C for different times: 0, 1, 3, 5, 10, 20, 30, 45, 60, 80, 100, 120, 150, and 180 min, respectively. The samples were resolved on 16% denaturing polyacrylamide gels and quantitatively analyzed by Quantity One (Bio-Rad, USA) to measure the cleavage efficiency of MfAgo mediated by different guides.

Homology modeling and structural analysis

Homology modeling for MfAgo was determined using the SWISS-MODEL online tool (https://www.swissmodel.expasy.org/) (Waterhouse et al. 2018) by using the binary structure of MjAgo with the 5′-P ssDNA guide (PDB ID: 5G5T, identity 68%) (Willkomm et al. 2017) as the template; then, the 5′-P ssDNA, 5′-OH ssDNA, and 5′-P ssRNA guides were docked into the MfAgo model structure using the Schrodinger software (Repasky et al. 2007). An electrostatic heat map and the specific region where MfAgo binds the guide were analyzed by PyMol.

Results and discussion

Phylogenetic tree and sequence analysis of $Mf$Ago

To discover new functional biological resources in pAgos, we constructed a phylogenetic tree by aligning the PfAgo amino acid sequences in the NCBI database (Fig. 1a). MfAgo was chosen as the candidate because it is most closely related to hyperthermophilic MjAgo, with 68% shared identity. Then, the sequence analysis of MfAgo was performed in detail. The MfAgo gene (GenBank Accession No. NC013156) was an intact open reading frame encoding the protein of 706 amino acids. Accordingly, target cleavage by all active Ago proteins is mediated by the conserved tetrad of the DEDX motif (where X can be D, H, N or K) (Kaya et al. 2016; Swarts et al. 2014b, 2015a; Yuan et al. 2005). Multiple sequence alignment was carried out, especially focusing on the regions containing the DEDX motif (Fig. 1b). Most importantly, we found that the DEDD tetrad occurred in MfAgo, suggesting that MfAgo may have endonuclease catalytic activity that needs characterization in vitro.

Fig. 1

Phylogenetic tree and sequence analysis. a Neighbor-joining phylogenetic tree analysis of MfAgo based on amino acid sequences. The numbers at the nodes indicate the bootstrap values for neighbor-joining analysis of 1000 resampled data sets. The scale bar represents the evolutionary distance between species. Numbers in parentheses represent the sequence accession numbers in the NCBI database. b Multiple sequence alignments of MfAgo with other known pAgos. MjAgo: Ago from Methanocaldococcus jannaschii. PfAgo: Ago from Pyrococcus furiosus. AaAgo: Ago from Aquifex aeolicus. TtAgo: Ago from Thermus thermophilus. Only regions containing the DEDX catalytic residues (indicated in red) are shown. The alignment was carried out using ClustalW, shaded manually and using ESPript 3.0 (Robert and Gouet 2014)

Expression and purification of the recombinant $Mf$Ago

The gene-encoding MfAgo was codon-optimized and built into the plasmid pET-28a(+), which has an N-terminal His-tag and was then transformed into E. coli BL21(DE3). Expression in the recombinant E. coli BL21(DE3) was induced by IPTG at 20 °C, and the resulting protein was then purified using an Ni–NTA affinity column. SDS-PAGE analysis showed that the purified protein size was consistent with the estimated molecular weight of recombinant protein (85.8 kDa), which indicated that the recombinant MfAgo was successfully expressed in E. coli BL21(DE3) in a soluble form (Fig. 2a). Then, the size-exclusion chromatography elution showed a single peak, indicating that the purified recombinant MfAgo was a monomer (Fig. 2b). Here, the recombinant MfAgo, purified by Ni–NTA and size exclusion, could be used for subsequent cleavage activity assays in vitro.

Fig. 2

Purification of the MfAgo protein. a SDS-PAGE analysis of Ni–NTA-purified MfAgo. M: protein marker; lane 1: purified MfAgo. b Chromatograms of size-exclusion chromatography for MfAgo. The measured UV absorbance (mAU) at 280 nm is shown against the elution volume (mL)

Enzymatic characterization of the recombinant $Mf$Ago in vitro

Endonuclease activity of $Mf$Ago

To investigate the enzyme properties of the recombinant MfAgo, we investigated nucleic acid cleavage via a variety of nucleic acid guides in vitro. First, we designed 16-nucleotide (nt) ssDNA or ssRNA guides with a 5′ phosphate or 5′ hydroxyl group to cleave complementary FAM-labeled ssDNA or ssRNA targets (Fig. 3a). Strikingly, the results showed that MfAgo cleaved complementary ssDNA targets using 5′-P ssDNA, 5′-P ssRNA, or 5′-OH ssDNA guides (Fig. 3b). Different kinds of guide preferences have not been observed in other characterized Agos from eukaryotes or prokaryotes.

Fig. 3

Endonuclease activity assays of MfAgo in vitro. a Synthetic DNA or RNA guides (red) and targets (blue). Predicted cleavage positions are indicated with a black triangle, black lines indicate the predicted 11- and 34-nt cleavage products. b Cleavage activity assay of MfAgo with FAM-labeled DNA or RNA targets

Meanwhile, the results also indicated that, like other characterized Agos, MfAgo cleaved the complementary target DNA strand at the complementary site between nucleotides 10 and 11 of the guide strand, but there were some differences in cleavage efficiency mediated by three different kinds of guides.

$Mf$Ago catalytic activity affected by temperature

The cleavage activity of active pAgos is strongly temperature dependent (Swarts et al. 2015a; Wang et al. 2009); thus, we tested the influence of temperature on recombinant MfAgo ssDNA cleavage activity mediated by three different kinds of guides (Fig. 4). The optimum temperatures of MfAgo were significantly different when directed by varied guides: the 5′-P ssDNA guide-mediated MfAgo was most active in the range from 85 to 98.7 °C, achieving nearly 100% cleavage for the ssDNA target; the activity of 5′-OH ssDNA guide-mediated MfAgo was enhanced with increasing temperature from 54.8 to 90 °C and then decreased rapidly with higher temperature (Fig. 4a), which might be caused by the weaker interaction between 5′-OH ssDNA guide and MfAgo than interaction between 5′-P ssDNA guide and MfAgo.

Fig. 4

Effect of temperature on MfAgo activity. a Effect of temperature on MfAgo activity mediated by 5′-P ssDNA and 5′-OH ssDNA guides. b Effect of temperature on MfAgo activity mediated by 5′-P ssRNA guide. Error bars represent the SDs of three independent experiments

For the DNA cleavage of 5′-P ssRNA guide-mediated MfAgo, we observed that the activity increased at the temperature range from 54.8 to 85 °C, then decreased rapidly at higher temperature and almost lost at 95 °C (Fig. 4b). Distinct to highly stable ssDNA guides, the 5′-P ssRNA guide could be degraded partially under high temperature (Additional file 1: Fig. S1). Therefore, we assumed that the reduction of MfAgo cleavage activity over 85 °C might be a result of the 5′-P ssRNA degradation (Additional file 1: Fig. S2) together with the weakening interaction between guide and MfAgo under high temperatures.

$Mf$Ago catalytic activity affected by divalent cations

Ago proteins are divalent cation-dependent endonucleases (Nowotny et al. 2005; Song et al. 2004), and the presence of divalent cations is essential for pAgos to specifically bind the 5′ end of the guide strand (Parker 2010; Sheng et al. 2014; Wang et al. 2009). To investigate which divalent cations MfAgo can be utilized for cleavage, different divalent cations at a final concentration of 0.5 mM were added to the reaction system. MfAgo was active with Mn²⁺, Co²⁺, and Mg²⁺, with Mn²⁺ giving higher activity than Co²⁺ and Mg²⁺ (Fig. 5a).

Fig. 5

Effects of different divalent cations and Mn²⁺ concentrations on MfAgo activity. a Effect of different divalent cations on enzyme activity. b Effect of Mn²⁺ concentration on enzyme activity. Error bars represent the SDs of three independent experiments

To further determine the optimal concentration of Mn²⁺ required for MfAgo activity, a series of Mn²⁺ concentrations was used for MfAgo activity analysis (Fig. 5b). Increased concentrations of Mn²⁺ improved MfAgo activity mediated by three different guides. Interestingly, MfAgo guide-mediated cleavage of ssDNA targets has different Mn²⁺ concentration requirements depending on the guide. MfAgo that binds a 5′-P ssDNA guide cleaved almost 100% of ssDNA targets when the concentration of Mn²⁺ was greater than 10 μM. However, MfAgo that binds 5′-OH ssDNA or 5′-P ssRNA guides cannot cleave the target nucleic acid completely, even at Mn²⁺ concentrations as high as 2000 μM.

$Mf$Ago catalytic activity strongly related to NaCl concentration

As NaCl plays an important role in maintaining the catalytic activity and stability of pAgos (Swarts et al. 2014a, 2015a), we investigated the effects of the NaCl concentration on MfAgo activity (Fig. 6). MfAgo activity mediated by the 5′-P ssDNA guide was highest in the NaCl concentration range from 100 to 1000 mM. MfAgo activity mediated by 5′-P ssRNA and 5′-OH ssDNA guides achieves maximum cleavage at NaCl concentrations of 100 mM and 500 mM, respectively. Furthermore, higher NaCl concentrations could inhibit the cleavage activity of MfAgo.

Fig. 6

Effect of NaCl on the enzyme activity of MfAgo. Error bars represent the SDs of three independent experiments

Kinetic performance of $Mf$Ago

To further investigate the catalytic properties of MfAgo mediated by three different guides, we performed a cleavage kinetics assay at 85 °C with 250 mM NaCl and 2 mM Mn²⁺ (Fig. 7). The results indicated that MfAgo exhibited the highest cleavage efficiency for the DNA target when utilizing the 5′-P ssDNA guide. In contrast, MfAgo activity mediated by the 5′-OH ssDNA guide was less efficient, and the cleavage efficiency of MfAgo using the 5′-P ssRNA guide was the lowest. Therefore, MfAgo prefers to use 5′-P ssDNA nucleic acid as a guide strand.

Fig. 7

Kinetic performance of MfAgo mediated by three different kinds of guides. Assays were performed in three independent experiments and plotted using nonlinear regression. SDs are represented by error bars

Homology modeling and structural analysis of key interactions between guides and the MID domain of MfAgo

Homology modeling has been extensively used to generate reliable three-dimensional protein structure models. With the complex structure of MjAgo with its 5′-P ssDNA guide (PDB ID: 5G5T, identity 68%) as a template, we used SWISS-MODEL to construct the structure of MfAgo (data not shown). Similarly, the architecture of MfAgo displays a typical pAgo protein structure, which has four domains and two linker regions, forming a bilobal scaffold composed of a PIWI (P element-induced wimpy testis) domain, a MID (middle) domain, a PAZ (Piwi–Argonaute–Zwille) domain and an N-terminal domain, and connecting linkers L1 and L2 (Parker et al. 2005; Rashid et al. 2007; Schirle and MacRae 2012; Song et al. 2004).

To better understand the specific guide recognition mechanism of MfAgo, the guides 5′-OH ssDNA, 5′-P ssDNA, and 5′-P ssRNA were individually docked into the MfAgo structure. Ago proteins use a pocket in the MID domain for binding the 5′ end of guides (Kaya et al. 2016; Sheng et al. 2014); thus, potential interactions between the MID domain and different nucleic acid guides were examined (Fig. 8a–c). In previous studies, TtAgo contains a binding pocket with multiple charged residues for recognizing and binding the 5′ phosphate group of the guide (Fig. 8d), while the MID-binding pocket of MpAgo contains a hydrophobic pocket devoid of charged residues around the 5′-hydroxy group (Fig. 8e). In contrast, it seems that the MID-binding pocket of MfAgo possesses both hydrophobic residues and hydrophilic residues and that the hydrophobic residues Leu⁴⁵³ and Tyr⁴³⁴ might interact with the 5′-hydroxyl group of guide DNA (Fig. 8a). Furthermore, the lack of a key α-helix that exists in MpAgo at the C-terminus of the PIWI domain would result a less-compressed MID-binding pocket in MfAgo, therefore providing sufficient space to bind the 5′ phosphate group.

Fig. 8

The structural analysis and electrostatic heat map (red = negatively charged, blue = positively charged, white = neutral) of the binding pocket of MfAgo for varied guides. The predicted binding pocket of MfAgo with a 5′-OH ssDNA guide (a), 5′-P ssDNA guide (b), and 5′-P ssRNA guide (c). d The binding pocket of TtAgo with a 5′-P ssDNA guide. e The binding pocket of MpAgo with a 5′-OH ssRNA guide

Inspection of the structure of MfAgo complexed with a 5′-OH ssDNA guide indicated that only residue Asp⁴³³ might form a single hydrogen bond with 5′-OH in ssDNA (Table 3). However, when the 5′-phosphate guide was docked into MfAgo, the residues Arg⁴²⁸ and Asp⁴³³ in the MID domain created multiple hydrogen-bonding interactions with the phosphate group of the guide, and another key residue, Lys⁶⁸⁵, is involved in hydrogen-bonding interactions with the 5′-terminal base of guide DNA/RNA, implying a stronger MfAgo-binding affinity for 5′-P guides than for 5′-OH guides. These results might be the reason that the catalytic activity and optimal temperature of MfAgo are higher with a 5′-P ssDNA guide than with a 5′-OH ssDNA guide.

Table 3 Residues of MfAgo to interact with the 5ʹ end of the guide

Conclusion

MfAgo was cloned and overexpressed in E. coli BL21(DE3). The biochemical characterization revealed that MfAgo catalyzes ssDNA target cleavage directed by either 5′-P ssDNA, 5′-OH ssDNA, or 5′-P ssRNA guides, which is distinguished from the behavior of all other characterized eukaryotic or prokaryotic Agos. MfAgo functions as an endonuclease for DNA target cleavage that requires divalent cations such as Mn²⁺, Mg²⁺, or Co²⁺ for its activity, with Mn²⁺ being the cation resulting in the highest activity. The maximum cleavage activity was achieved at 90 °C for 5′-OH ssDNA and 80–85 °C for 5′-P ssRNA guide-directed MfAgo, whereas 85–98.7 °C was optimal for the 5′-P DNA guide-directed MfAgo. Meanwhile, MfAgo is quite active in reactions with an NaCl concentration less than 500 mM, whereas higher NaCl concentrations inhibit the activity of MfAgo. Furthermore, our structural modeling studies revealed that MfAgo has a distinct pocket in the MID domain for binding the 5′-end group of guides and that the different catalytic activities of MfAgo with diverse guides might drive from varied hydrogen-binding interactions between the 5′-end group of guides and several key residues in this binding pocket. Characterization of MfAgo expands the understanding of the pAgo family and will inspire the development of these new Ago proteins’ potential applications in genetic manipulation.