Structural and functional insights into a novel two-component endolysin encoded by a single gene in Enterococcus faecalis phage.

Using bacteriophage-derived endolysins as an alternative strategy for fighting drug-resistant bacteria has recently been garnering renewed interest. However, their application is still hindered by their narrow spectra of activity. In our previous work, we demonstrated that the endolysin LysIME-EF1 possesses efficient bactericidal activity against multiple strains of Enterococcus faecalis (E. faecalis). Herein, we observed an 8 kDa fragment and hypothesized that it contributes to LysIME-EF1 lytic activity. To examine our hypothesis, we determined the structure of LysIME-EF1 at 1.75 Å resolution. LysIME-EF1 exhibits a unique architecture in which one full-length LysIME-EF1 forms a tetramer with three additional C-terminal cell-wall binding domains (CBDs) that correspond to the abovementioned 8 kDa fragment. Furthermore, we identified an internal ribosomal binding site (RBS) and alternative start codon within LysIME-EF1 gene, which are demonstrated to be responsible for the translation of the truncated CBD. To elucidate the molecular mechanism for the lytic activity of LysIME-EF1, we combined mutagenesis, lytic activity assays and in vivo animal infection experiments. The results confirmed that the additional LysIME-EF1 CBDs are important for LysIME-EF1 architecture and its lytic activity. To our knowledge, this is the first determined structure of multimeric endolysin encoded by a single gene in E. faecalis phages. As such, it may provide valuable insights into designing potent endolysins against the opportunistic pathogen E. faecalis.

Introduction

Enterococcus faecalis (E. faecalis) is a Gram-positive bacterium that commonly inhabits oral cavity, lower intestinal tract, and vaginal tract of healthy humans or other mammals [1]. However, E. faecalis is also a major opportunistic pathogen that causes a large number of community-acquired and hospital infections, where high levels of antibiotic resistance in E. faecalis increase its pathogenicity [2]. E. faecalis is particularly life-threatening for immunocompromised patients, and is known to cause endocarditis, bacteremia, urinary tract infections, meningitis etc. In addition, Enterococcus has been found related to root canal-treated teeth and diabetic foot infections [3-6].

Endolysins are hydrolytic enzymes that specifically recognize the host cell-wall and cleave the peptidoglycan to digest the bacterial cell-wall for release of phage progeny during the lytic cycle [7, 8], which is why they hold promise for the treatment of multidrug-resistant bacterial infections as an alternative to antibiotics [9-11]. Generally, most endolysins are composed of a single polypeptide chain divided into an N-terminal catalytic domain (CD) and a C-terminal cell-wall binding domain (CBD), which are interconnected by a short flexible linker [12-14]. Some endolysins contain single CD and some are composed of multiple CDs linked to CBD [PLoS pathogens. 2014 ">15-17]. Catalytic domains are responsible for cleavage of peptidoglycan bonds within the cell-wall and can be classified into four groups: N-acetylglucosaminidases, N-acetylmuramoyl-L-alanine amidases, N-acetylmuramidases (lysozymes), and endopeptidases [18-20]. In contrast, CBDs display much greater variety and distinguish discrete epitopes present within the host cell-wall, typically carbohydrates or teichoic acids, thus giving rise to species-specific or strain-specific activity of a particular endolysin [21, 22]. The modular structure of lysins makes it possible to design bioengineered endolysins that have desired properties, such as higher activity, or broader killing spectrum. Chimeric endolysins such as ClyH and ClyJ are engineered to fight against methicillin-resistant Staphylococcus aureus (MRSA) and multidrug-resistant Streptococcus pneumoniae (MRSP) [23, 24].

Although most known endolysins are composed of a single polypeptide chain containing one CD and one CBD [25-27], some exceptions are known. For example, in some cases two or three CDs are linked to a single CBD made of two separated functional regions [15, 28, 29]. Moreover, PlyC, a unique multimeric endolysin from streptocococcal C1 phage, consists of two components encoded by two separated genes located in one single operon [30]. The structure of PlyC comprises an enzymatic component (PlyCA) associated with eight copies of the cell-wall binding component (PlyCB) [31]. Interestingly, recent studies reported the existence of multimeric endolysins encoded by one single gene, such as Lys170 from enterococcal phage F170/08 [32], CTP1L that targets Clostridium tyrobutyricum, and CD27L that targets Clostridium difficile [33, 34].

LysIME-EF1, an endolysin isolated from E. faecalis phage possesses a highly efficient lytic activity to multidrug resistance E. faecalis [35]. In this study, we found that LysIME-EF1 possesses efficient bactericidal activity and can lyse 29 E. faecalis clinical strains, including the lethal vancomycin-resistant Enterococcus faecalis 002 (VREF). Furthermore, the discovery that a truncated CBD fragment is co-purified with the full-length LysIME-EF1 prompted us to solve the crystal structures of LysIME-EF1 holoenzyme and the CBD, which were determined both at 1.75 Å resolution. These structures and the accompanying biochemical experiments and in vivo animal infection experiments provided insights into the molecular mechanism of LysIME-EF1 lytic activity.

Results

Expression of the LysIME-EF1 gene appears to result in two stable polypeptides

In our previous work, we isolated an endolysin derived from E. faecalis bacteriophage [35]. PSI-BLAST search predicted that LysIME-EF1 consists of an N-terminal catalytic CHAP (cysteine, histidine-dependent amidohydrolases/peptidases) domain and a C-terminal CBD (cell-wall binding domain) (. Remarkably, LysIME-EF1 displays efficient bactericidal activity against 29 clinical strains of E. faecalis (). Subsequently, we found that expression of full-length N-terminally linked to a 6xHis-tag LysIME-EF1 in E. coli resulted in a strongly co-purified polypeptide with the approximate molecular mass of 8 kDa ( The 8 kDa band was overlooked in our previous works because of relatively longer running time for SDS-PAGE [35]. To identify this protein, we conducted N-terminal sequencing and found that the first four amino acid residues coincided with the sequence of M168F169I170Y171, which corresponds to the middle of LysIME-EF1. These results suggest that the 8 kDa fragment is the CBD of LysIME-EF1 rather than a product of protein degradation.

Expression of LysIME-EF1 gene appears to results in production of two stable polypeptides.

(A) Domain organization of LysIME-EF1. The full-length LysIME-EF1 contains two domains, i.e. the N-terminal CHAP domain (residues 1–143, colored green) and the C-terminal CBD (residues 168–237, colored red). The linker (residues 144–167) interconnecting the two domains is shown as a dot-dashed blue line. (B) The SDS-PAGE analysis of the expression of wild type (WT) LysIME-EF1. (C) Survival test of E. faecalis 002 on LB agar dishes after the cells were lysed by WT LysIME-EF1 (0.1 μM) as well as CHAP domain protein alone (0.1 μM) or CBD alone (0.1 μM) or mixture of CHAP domain protein plus CBD (10 μM) at 37°C for 2 hours. (D) Survival test of E. faecalis 002 on LB agar dishes after the cells were lysed by WT LysIME-EF1 (0.1 μM) that mixed with CBD at different molar ratio at 37°C for 2 hours.

To further examine the hypothesis that the extra CBD fragment is incorporated into protein complex, size-exclusion chromatography was performed to analyze the solution state of LysIME-EF1. LysIME-EF1 CBD (8 kDa) and full-length LysIME-EF1 (30 kDa) were co-eluted at 15.5 ml, with the estimated molecular mass of approximately 60 kDa (). To obtain the exact molecular mass, analytical ultracentrifugation (AUC) was performed, indicating that LysIME-EF1 exhibited a molecular mass of 55.6 kDa, higher than the theoretical molecular mass of LysIME-EF1 alone (). Taken together, these results suggested that the truncated CBD fragments may form a complex with full-length LysIME-EF1.

To assess the effect of additional CBDs on LysIME-EF1 lytic activity, we tested the bactericidal activity of the wild type (WT) LysIME-EF1, LysIME-EF1 CHAP, LysIME-EF1 CBD and mixed WT LysIME-EF1 and LysIME-EF1 CBD at different molar ratios. All the constructs were expressed in E. coli and purified as soluble recombinant proteins ( The results revealed that WT LysIME-EF1 has efficient bactericidal activity against E. faecalis. In contrast, CHAP alone, CBD alone or a mixture of these two polypeptides exhibited very low bactericidal activity, even at high concentrations (10 μM) ( Interestingly, addition of the CBD mixed with full-length LysIME-EF1 at different molar ratios showed that lytic efficiency is inversely proportional to the molar ratio of CBD fragment to full-length LysIME-EF1 (.

Overall structure of WT LysIME-EF1 and the CBD fragment

To better understand the role of the CBD in WT LysIME-EF1 lytic activity, we aimed to determine the structure of LysIME-EF1. Thus, full-length LysIME-EF1 was cloned into the expression vector pCold I, expressed in E. coli and subsequently purified using Ni affinity chromatography and size-exclusion chromatography. Following extensive crystallization trials, we obtained crystals that diffracted well in the space group I422, with cell parameters a = b = 59.88 Å, c = 87.01 Å, α = β = γ = 90.00° ( However, we failed to solve the structure by molecular replacement using the CHAP domain of LysGH15 (PDB entry: 4OLK) as search model. Fortunately, we obtained diffractive crystals of selenium-labeled protein in different crystallization conditions, which were then optimized and diffracted well in the space group P1211 with cell parameters a = 47.03 Å, b = 56.92 Å, c = 91.52 Å, α = γ = 90.00°, β = 96.29° (). Finally, the structure was solved successfully using the data sets belonging to the space group P1211 by selenium single-wavelength anomalous dispersion (Se-SAD) method. Interestingly, the structure of the abovementioned datasets in the space group I422 was solved later by molecular replacement and determined as LysIME-EF1 CBD homotetramer ().

The crystal structure of WT LysIME-EF1 refined to 1.75 Å resolution revealed that WT LysIME-EF1 possesses a CHAP domain (residues 1–143) and a CBD (residues 168–237), which are connected by a linker (residues 144–167) (). The electron densities for the CHAP and the CBD domains are unambiguously defined, whereas the electron density for the linker is missing due to its high flexibility. In addition to the full-length LysIME-EF1, three additional CBD fragments were successfully built into the final model (). The four CBDs together form a tetrameric ring 45 Å in length and 45 Å in width. One side of LysIME-EF1 tetrameric CBDs, which we speculate to be responsible for recognition of peptidoglycan, is arranged in a planar surface, while the other side is more convex in shape (right panel in ).

Overall structure of wild type LysIME-EF1.

(A) Ribbon diagram of an asymmetric unit of LysIME-EF1. The full-length LysIME-EF1 was colored in green for CHAP domain and in red for CBD, and the other three CBDs were colored different shades of grey. (B) Surface representations of LysIME-EF1 shown the same views and colors as in (A). The four CBDs form a tetrameric ring with 45 Å in length and 45 Å in width. The N-terminal CHAP domain is positioned entirely above the CBD tetrameric ring.

The N-terminal CHAP domain of LysIME-EF1 folds into a typical globular domain containing two α-helices, two 310-helices and six β-strands (). Dali search [36] against the PDB database indicates that the most closely related structures are those of the CHAP domains of staphylococcal phage-derived endolysins LysGH15 (PDB: 4OLK) and LysK (PDB: 4CSH) () [15, 29]. Structural alignment of the CHAP domains of LysIME-EF1 and LysGH15 shows that the two α-helices and the six β-strands of both proteins are aligned well with each other with an average root mean square deviation (RMSD) of 0.717 Å and 0.742 Å, respectively (). Analogously to LysGH15, LysIME-EF1 CHAP domain also contains a calcium-binding site and a classical amidohydrolase catalytic triad (C29, H90 and N110) (). Electron density of the calcium ion is clearly identified in the crystal structure, and the calcium ion is coordinated by the conserved residues D20, D22, W24, G26, D31 and one water molecule (). EDTA completely inactivates the lytic activity of LysIME-EF1, which can only be restored by addition of calcium ions (. Similar to LysGH15 [15], mutation of the catalytic residues to alanine reduced the bactericidal activity of LysIME-EF1 (.

Structural and functional analysis of the LysIME-EF1 CHAP domain.

(A) Overall structure of the CHAP domain (residues 1–143). The N- termini and C-termini are labeled with the respective letters. The Ca2+ ion bound to the LysIME-EF1 CHAP domain is shown as sphere in grey. (B) A detailed view of the catalytic triad and the calcium-binding site in the CHAP domain. (C) Sequence alignment of the residues involved in calcium binding, which are conserved when aligned with LysGH15 and LysK. The blue triangles indicate the residues involved in Ca2+ ion binding. (D) The effects of different metal ions (1 μM) on the lytic activity of the EDTA-inactivated WT LysIME-EF1 protein (5 μM) (excess EDTA was removed by dialysis). The values represent the mean ± SD (n  =  3). (E) Survival test of E. faecalis 002 on LB agar dishes after the cells were lysed by the catalytic triad residue (C29A, H90A and N110A) mutants of LysIME-EF1 (0.1 μM) at 37°C for 1 hours.

Ultimately, the dataset diffracted in the space group I422 was proved to be LysIME-EF1 CBD and the structure was solved by molecular replacement method at 1.75 Å resolution using the CBD from WT LysIME-EF1 as a search model. There was one molecule in an asymmetric unit. The CBD comprises a four-stranded β-sheet capped on each side by one α-helix (). Crystallographic symmetry analysis suggested that LysIME-EF1 CBD forms a stable homotetramer (). The oligomeric interface contains 1318.9 Å2 of the buried surface area, accounting for 28.25% of the monomer surface area calculated by PISA [37]. To investigate the oligomerization state of LysIME-EF1 CBD, size-exclusion chromatography was performed. The LysIME-EF1 CBD was cloned into pET 21a(+) with an N-terminal His tag and the purified LysIME-EF1 CBD was subjected to size-exclusion chromatography and the elution profile is 17.7 ml in a Superdex 200 column (GE Healthcare) with an estimated ~32 kDa (, supporting that LysIME-EF1 CBD forms a stable tetramer in solution.

Detailed analysis shows that the tetramerization is mainly mediated by extensive hydrogen bonds and hydrophobic interactions (. The hydrogen bonds are mainly formed between the residues located at the α1 helix of one protomer and the residues located at the β4 of the adjacent protomer. More specifically, the residue Y218 forms hydrogen bonds with R208 and Y209 of the adjacent protomer, the carboxyl of R208 and N212, K189 and A231, E201 and N222 also form hydrogen bonds between two adjacent protomers (). In addition, a sizeable hydrophobic interface is formed by the neighboring protomers, where the residues M227, M228, A230, A231, L232 of α2 helix from one protomer pack against hydrophobic patches 183WFVIGG188 and 191IYLP194 of the other protomer. Notably, the M227 nearly inserts into the hydrophobic cavity formed by residues F184, I186, I191 and L193 (). In order to evaluate the importance of these residues for tetramerization, we created LysIME-EF1 CBD variants carrying mutants of the corresponding residues. The double mutant that carries an Arg instead of Phe184 and Ala instead of Tyr218 was tested by size-exclusion chromatography. Differing from that of LysIME-EF1 CBD, the elution profile of the LysIME-EF1 CBD F184R&Y218A is 18.7 ml (), confirming that the residues F184 and Y218 are essential for maintaining the oligomeric tetramerization of LysIME-EF1.

Additional CBD fragments are required for the lytic activity of LysIME-EF1

To further investigate if the three truncated CBD fragments are necessary for the functional LysIME-EF1 holoenzyme, we compared the lytic activities of WT LysIME-EF1 and LysIME-EF1 obligate-monomer (LysIME-EF1 F184R/Y218A mutant). Antibacterial activity was assessed as the decrease in viable bacterial counts (Fig 4D). The results showed that the antibacterial activity was nearly lost when the interactions between the CBDs were disrupted, suggesting that tetramerization of LysIME-EF1 is required for its highly efficient lysis of E. faecalis cells.

Fig 4

Analysis of the oligomerization interface of LysIME-EF1 CBD.

(A) Ribbon representation of LysIME-EF1 CBD tetramer. The four monomers (Chains A to D) are shown in red, blue, yellow and palegreen, respectively. (B) Ribbon representation showing the detailed interaction between the adjacent CBDs. The key residues are shown in sticks. (C) Combined surface representation and a ribbon-stick model showing the molecular interface of the hydrophobic interactions, one LysIME-EF1 CBD is shown in the surface model and the other is shown in the ribbon-sticks model. The hydrophobic residues of LysIME-EF1 in the model are shown in white. (D) Survival test of E. faecalis 002 on LB agar dishes after the cells were lysed by WT LysIME-EF1 (0.1 μM) as well as LysIME-EF1 F184R/Y218A mutant (0.1 μM) at 37°C for 2 hours.

Identification of an internal translation site within the LysIME-EF1 gene

The high-resolution structures of WT LysIEM-EF1 and LysIME-EF1 CBD raised the question about how the extra CBD was produced. To this end, we performed a thorough analysis of the amino acid sequence of LysIME-EF1. Notably, the presence of a methionine (M168) at the N-terminus of the three CBDs in our protein structure prompted us to consider the hypothesis that the site of M168 acts as an alternative in-frame start codon site for translation (502ATG504) (). Coincidentally, inspection of LysIME-EF1 nucleotide sequence revealed a putative ribosome binding site (RBS) upstream of the alternative start codon ().

Identification of an internal ribosomal binding site and alternative start codon within LysIME-EF1 gene.

(A) Sequence logo created with Weblogo [51] showing the linker region of LysIME-EF1 gene with nucleotides numbered according to its nucleotide sequence and based on a sequence alignment of LysIME-EF1-related endolysins. The internal ribosomal binding site sequence is underlined with red line, and the start codon is underlined with a blue line. (B) SDS-PAGE analysis showing products of the expression of WT LysIME-EF1 and mLysIME-EF1. (C) Survival test of E. faecalis 002 on LB agar dishes after the cells were lysed by WT LysIME-EF1 (0.1 μM) as well as mLysIME-EF1 (0.1 μM) at 37°C for 2 hours. (D) Survival test of E. faecalis 002 on LB agar dishes after the cells were lysed by mLysIME-EF1 (0.1 μM) that mixed with CBD at different molar ratio at 37°C for 2 hours.

To establish the relationship between these sequences and the production of truncated CBDs, a methionine-to-leucine substitution (M168L) and mutation of the putative ribosome binding site AAGGAGA to AACCAGA (mutated nucleotides are underlined) were carried out (hereafter the double mutant protein is designated mLysIME-EF1). Unlike the WT LysIME-EF1, mLysIME-EF1 is translated into a single product with the molecular mass of ~30 kDa, corresponding to the full-length LysIME-EF1, whereas the ~ 8 kDa product was absent (). These results clearly confirmed that both M168 and putative RBS are associated with the production of the ~ 8 kDa CBD fragments. Intriguingly, a similar phenomenon, in which one full-length protein associates with three CBDs produced by secondary translation, has been reported for endolysin Lys170 derived from another E. faecalis bacteriophage [32].

In order to further examine the effect of the CBD fragments on lytic activity of full-length LysIME-EF1 against E. faecalis, we compared bactericidal potentials of the mLysIME-EF1 with that of the WT LysIME-EF1. As shown in mLysIME-EF1 has significantly reduced bactericidal activity in comparison to the WT LysIME-EF1 protein. This suggests that the co-assembly with three CBDs is a prerequisite for WT LysIME-EF1 to form a functional unit to fight against E. faecalis. However, different from Lys170 and CTP1L [32, 34], mixing mLysIME-EF1 with the truncated LysIME-EF1 CBD (residues 168–237) at different molar ratios also could not significantly improve the lytic activity (

The results of the current study and previous reports made us wonder whether the mechanisms of secondary translation and oligomerization are common to other endolysins homologous to LysIME-EF1 and what extent [32]. We performed a protein BLAST using the LysIME-EF1 CBD domain as a query sequence. As a result, 36 unique protein sequences with an E value < 0.01 were identified (. All these proteins are described as bacteriophage-derived endolysins that target Enterococcus species. Notably, methionine as the first amino acid residue of their CBD domains is a feature common to all of them. In addition, nucleotide sequence alignment revealed the same ribosome binding site (AAGGAGA) sequence present in all 36 proteins just upstream of the alternative start codon site. These results are consistent with the previous bioinformatics analysis by Proenca et al [32] and imply that the use of a secondary translation initiation site may be widespread among endolysins targeting Enterococcus species.

Identification of the key residues involved in binding to the host cell

Next, for further understanding of the structural basis of how LysIME-EF1 binds to the Enterococcus species strains, we aimed to identify the key residues involved in binding to the host cell. A positively charged patch that may contain residues involved in cell-wall binding is located on the planar surface of the LysIME-EF1 tetramer (). In order to verify the involvement of this positively charged patch in cell-wall binding, we substituted the residues K173, K176, and R190 with glutamic acid (E) (). Then, we generated recombinant N-terminally GFP-tagged LysIME-EF1 CBD variants and assessed their cell-wall binding capability using fluorescence-activated cell sorting (FACS) (. When compared to the WT LysIME-EF1, mutations K173E and K176E had a slight impact on LysIME-EF1 cell-wall binding ability, whereas the mutation R190E nearly abolished the cell-wall binding ability of LysIME-EF1. In addition, we also compared bactericidal activities of the abovementioned LysIME-EF1 variants against E. faecalis 002. As expected, the R190E mutant exhibited almost no lytic activity (), which may be due to substantial loss of host cell binding ability. These results suggest that R190 is the key residue mediating binding of LysIME-EF1 to host cells, which is in accordance with its high level of conservation in LysIME-EF1 and homologous endolysins ().

Key residues of LysIME-EF1 CBD involved in binding to E. faecalis cell-wall.

(A) Cartoon depiction of LysIME-EF1 monomer shows the predicted residues (K173, K176 and R190) involved in cell-wall binding in sticks. (B) Flow cytometry was used to assess the ability of LysIME-EF1 and its variants (WT LysIME-EF1 and K173E, K176E and R190E mutants) to bind E. faecalis 002. (C) Survival test of E. faecalis 002 on LB agar dishes after the cells were lysed by WT LysIME-EF1 (0.1 μM) as well as K173E, K176E and R190E mutants (0.1 μM) at 37°C for 2 hours.

Evaluation of protective capabilities in mice

Finally, to investigate whether the additional CBDs and the residues involved in cell-wall binding are necessary for protection against E. faecalis challenge in mice in vivo, the purified WT LysIME-EF1 and its derivatives (mLysIME-EF1, LysIME-EF1 C29A/H90A/N110A, K173E, K176E, R190E, or F184R/Y218A mutants) were administered to mice inoculated intraperitoneally with E. faecalis 002. When determining the minimal lethal dose (MLD) of E. faecalis 002, we found that injecting 2×109 CFU (colony-forming unit)/mouse of E. faecalis 002 caused nearly a 100% mortality rate within 4 days. The dose of 2×MLD (i.e. 4×109 CFU/mouse) was then used as the challenge dose for testing the protection capabilities of WT LysIME-EF1 and its derivatives. Equal amounts of WT LysIME-EF1 (20 μg) or LysIME-EF1 derivatives (20 μg) were administered to respective experimental groups (n = 24) 1 h post challenge inoculation. The results revealed that the mice administered WT LysIME-EF1 provided a 95.83% (23 out of 24) protection rate. However, the catalytic triad triple mutant (C29A/H190A/N110A) provided only a 12.5% protection rate and mLysIME-EF1 provided a 25% protection rate, whereas the LysIME-EF1 F184R/Y218A rescued only 16.67% of mice. The protection rates of K173E, K176E, R190E LysIME-EF1 were 54.17%, 45.83%, 16.67%, respectively (). Taken together, the animal experiments further confirmed that the additional LysIME-EF1 CBD and the residues involved in cell-wall binding on CBD are important for the lytic activity of LysIME-EF1.

LysIME-EF1 rescues mice from E. faecalis infection.

Mice were injected intraperitoneally with 4×109  CFU/mouse of E. faecalis 002. One hour later they were treated with WT LysIME-EF1 and LysIME-EF1 variants proteins (20 μg in 200 μl 1xPBS buffer) by the same route and followed for 4 days (n  =  24). As a blank control, mice were treated with an equal volume of 1xPBS buffer. Mice survival rate was evaluated using Kaplan–Meier analysis. For clarity, overlapping curve lines are shown as dotted lines.

Discussion

Use of bacteriophage-derived endolysins is nowadays considered a potential strategy for treatment of bacterial infections [38-40], particularly caused by multidrug-resistant pathogenic microorganisms [41]. Despite its potent bactericidal effect and potential for therapeutic application, limited structural and functional information of endolysins has been available. The high-resolution X-ray structures of full-length LysIME-EF1 and its CBD fragments reported in this study contribute to our understanding of the multimeric endolysins and provide a template for elucidating the biological function and underlying mechanisms of LysIME-EF1 and its highly efficient lytic activity.

The structure of WT LysIME-EF1, which is distinct among known structures of endolysins, consists of one full-length protein associated with three additional N-terminally truncated CBD peptides. The structure of the CBD in the full-length LysIME-EF1 protein does not show any differences from that of the truncated CBD peptides. However, disruption of the oligomerization interface between CBDs results in significant decrease in lytic activity against E. faecalis, suggesting that LysIME-EF1 is a functional endolysin that requires the assembly of additional three CBDs to achieve an oligomeric conformation for optimal activity. A similar result was reported in other multimeric endolysins, such as the previously identified endolysin Lys170 from the enterococcal phage F170/08 or CTP1L, which lyses Clostridium tyrobutyricum [32, 34]. However, the molecular mechanism of the multimeric endolysins recognition of their substrates remains to be illuminated.

The presence of secondary translation products has been described for endolysins of several phages [42-44], which is consistent with the generally accepted notion that bacteriophages make very economical use of their genetic material. For instance, in-frame secondary translation site has previously been reported for endolysin of the staphylococcal phage 2638A, which is lytic for Staphylococcus aureus [42]. Concerning oligomerization, only three endolysins have been reported to form an active multimer: CTP1L and CD27L from Clostridium tyrobutyricum phages, and Lys170 from E. faecalis phage [32-34]. The results of our study and the abovementioned reports suggest that the use of an internal translation site may be common among endolysin genes. Structure-based analysis of the LysIME-EF1 CBD domain and homology search identified 36 endolysins from E. faecalis phages, which display amino acid sequence identity ranging from 37% to 98% with LysIME-EF1. The common feature of these endolysins is the presence of an internal ribosomal binding site upstream of a putative alternative start codon.

Based on the results of the structural and biochemical analysis, it was demonstrated that the residue R190 is essential for LysIME-EF1 binding to host cell. The animal experiments performed in this study further corroborated the results of structural and biochemical analysis.

In conclusion, our work sheds light on the molecular mechanism by which WT LysIME-EF1 lyses host cells, and thus lays the basis for designing endolysins with high activity and broader antimicrobial spectrum that can be useful for pharmaceutical applications.

Materials and methods

Ethics statement

Animal use protocols are approved by the Animal Ethics and Welfare Committee of Beijing Institute of Microbiology and Epidemiology (#2018–583) and implemented in accordance with the national standards of the People's Republic of China (GB / T 35892–2018: Guidelines for the review of Experimental mice-animal welfare ethics, GB / T 35823–2018: General requirements for laboratory animal-animal experiments).

Protein expression and purification

The coding sequence of the LysIME-EF1 (NC_041959.1) was PCR-amplified using primers (forward: 5'-gcggatccatggttaaattaaacgatgtac-3', reverse: 5'- gctctagactatactttaacttgtggtaaagc-3'), after which the amplified fragment was cloned into the expression vector pCold I using restriction endonuclease Xba I and BamH I (NEB). The construct was introduced into E. coli BL21 (DE3) cells for expression of recombinant proteins. The recombinant strains were cultured in LB medium (10 g/L NaCl, 10 g/L tryptone, and 5 g/L yeast extract) at 37°C until the OD600 reached 0.8. After 16 h of induction with 0.5 mM isopropylthio-beta-D-galactoside (IPTG) at 16°C, the cells were harvested by centrifugation at 2,000×g for 30 min at 4°C. The pellet was resuspended in the lysis buffer containing 50 mM Tris-HCl (pH 8.0) and 150 mM NaCl, and lysed by ultrasonication. The lysate was then centrifuged at 23,000×g for 30 min at 4°C, and the supernatant was loaded onto a Ni-NTA column (Qiagen). The target protein was eluted in buffer containing 50 mM Tris-HCl (pH 8.0), 150 mM NaCl and 250 mM imidazole. After sample analysis with SDS-PAGE, the fractions containing pure protein were pooled and concentrated to 0.5 ml, and then loaded onto a Superdex 200 increase column (GE Healthcare) equilibrated with a buffer containing 20 mM Tris-HCl (pH 8.0) and 150 mM NaCl. Selenomethionine derivate of LysIME-EF1 was purified using the same procedure, but with 2 mM DTT added to the buffer. All mutants of LysIME-EF1 were purified in the same procedure as the WT LysIME-EF1.

The CHAP domain (residues 1–143) and the CBD (residues 168–237) of LysIME-EF1 were sub-cloned into pET21a with an N-terminal 6xHis-tag, the recombinant protein was expressed and purified as the WT LysIME-EF1.

Site-directed mutagenesis

The Quick-Change Site-Directed Mutagenesis Kit (Stratagene) was used to generate LysIME-EF1 mutants. The primers used to generate the mutants are listed in . All mutants were constructed and verified by sequencing of the entire gene. Expression and purification of the mutants were performed using the same method as the production of the WT LysIME-EF1.

Crystallization

Initial crystallization screens were conducted with sitting-drop vapor diffusion method using commercial crystallization kits. The protein concentration used for crystallization was 15 mg/ml. Hampton Research kits were used in the sitting drop vapor diffusion method to identify preliminary crystallization conditions at 16°C. Crystallization drops contained 0.5 μl of the protein solution mixed with 0.5 μl of reservoir solution. Crystals appeared in two conditions. Cubic-shaped crystals appeared in 0.1 M HEPES (pH 6.5–7.5), 50 mM magnesium acetate and 20% (w/v) PEG 4,000. Diamond-shaped crystals appeared in 20% PEG8,000 and 0.2 M sodium potassium tartrate tetrahydrate 0.1 M Tris pH 8.5. Crystals were harvested with 20% (v/v) ethylene glycol as cryoprotectant before flash freezing in liquid nitrogen. To solve phase problem, selenomethionine was introduced with the standard method. The protein concentration of the selenomethionine derivate of LysIME-EF1 used for crystallization was 30 mg/ml. Crystals of LysIME-EF1SeMet were grown in 0.1 M MES (pH 5.6–6.2), 0.2 M Ca(OAc)2, and 10–14% (v/v) 2-propanol. All crystals were flash-frozen in liquid nitrogen, with the addition of 20%-25% (v/v) glycerol as cryo-protectant.

Data collection and structure determination

X-ray diffraction data were collected at the beamline BL-17U1 of the Shanghai Synchrotron Radiation Facility (SSRF). All data were processed using HKL-2000 [45, 46]. The original structure of LysIME-EF1 was determined by using the single-wavelength anomalous dispersion (SAD) phasing. Phases were calculated using AutoSol implemented in PHENIX [47]. AutoBuild in PHENIX was used to automatically build the atom model. The structure of the CBD of LysIME-EF1 was determined by using molecular replacement with Phaser [48]. After several rounds of positional and B-factor refinement using Phenix, as well as refinement with TLS parameters alternated with manual model revision using Coot [49], the quality of final models were checked using the PROCHECK program. Details of the data collection and refinement statistics are given in All of the figures showing protein structures were prepared with PyMOL [36]. All coordinates and structure factor files are available at the PDB database as 6IST (WT LysIME-EF1) and 6L00 (LysIME-EF1 CBD).

Size-exclusion chromatography

The purified WT LysIME-EF1, mLysIME-EF1, LysIME-EF1 CHAP (residues 1–143), LysIME-EF1 CBD (residues 168–237), LysIME-EF1 F184R/Y218A and LysIME-EF1 CBD F184R/Y218A were subjected to size-exclusion chromatography (Superdex 200 10/300 GL column, GE Healthcare, 20 mM Tris–HCl pH 8.0, 150 mM NaCl), respectively. The assay was performed at a flow rate of 0.5 ml/min at 16°C. Standard proteins containing Thyroglobulin (669 kDa), Aldolase (158 kDa), Conalbumin (75 kDa), Ovalbumin (44 kDa), Aprotinin (6.5 kDa) were performed in the same buffer at the same condition.

Analytic ultracentrifugation (AUC)

Sedimentation velocity experiments were used to assess the molecular sizes of the WT of LysIME-EF1 and the CBD of LysIME-EF1 at 20°C on a Beckman XL-A analytical ultracentrifuge equipped with absorbance optics and an An60 Ti rotor (Beckman Coulter, Inc. Fullerton, CA). Samples were diluted to an optical density at 280 nm (OD280) of 1 in a 1.2 cm path length. The rotor speed was set to 35,000×g for all samples. The sedimentation coefficient was obtained using the c(s) method with the Sedfit Software [50].

Turbidity reduction assay

E. faecalis 002 was grown to mid-exponential growth phase until OD600 reached 0.8, after which the bacteria were centrifugated at the speed of 3,000×g for 10 mins. The cells were washed three times with 1xPBS buffer and then resuspended in 3 ml of fresh LB medium. Eighteen microliters of the purified endolysin protein (1 mg/ml) were added to the bacterial sample, and the mixture was cultured on a shaking platform at 37°C. Changes in value of OD600 were measured every 20 min.

EDTA (1 mM) was added to the purified wild-type LysIME-EF1 and the excess EDTA was removed by dialysis using 1xPBS buffer at 4°C for 16 h. Afterwards, equal amounts of different metal ions (1 μM) were added to the EDTA-inactivated LysIME-EF1 and incubated for 2 h at 4°C. Peptidoglycan lytic activity was measured as the decrease in the OD600.

Bactericidal activity assay

For the antimicrobial activity assay, equimolar concentrations (0.1 μM) of LysIME-EF1 and LysIME-EF1 mutants were added into E. faecalis 002 bacterial suspensions (OD600 = 0.8), respectively. After 2 hours of incubation at 37°C, 220 rpm. The cell suspension was washed twice using 1xPBS buffer by centrifugation (17,000×g, 10 min), and then 4 μl of tenfold serially diluted cell suspension were spot plated onto LB agar dishes for survival assay. Residual viable cell numbers on the plate were measured after incubation at 37°C for 12 h. For the blank control, the same volume of 1xPBS buffer was added instead of endolysin LysIME-EF1. The antibacterial activity was expressed as the decrease in viable bacterial counts. All experiments were performed in triplicate.

To test the effect of the CBDs on the wild-type LysIME-EF1 and the mLysIME-EF1, wild-type LysIME-EF1 or mLysIME-EF1 were incubated with LysIME-EF1 CBD at different molar ratios (1:1, 1:3, 1:6, 1:9) at 4°C for 2 h. The method of measuring bactericidal activity is described as above.

Flow cytometry analysis

E. faecalis 002 was cultured in LB medium until the OD600 reached 0.8, after which the cells were centrifuged at the speed of 3,000×g. The obtained pellet was washed and resuspended in 1xPBS buffer. The purified GFP-LysIME-EF1 CBD was mixed with the resuspended E. faecalis 002 and incubated at 4°C for 1 hour. After two washes with 1xPBS buffer, the percentage of GFP-positive cells was determined using the Cell Quest Software (Becton Dickinson).

Protection by LysIME-EF1 in mouse infected with E. faecalis 002

The infection model was based on methods described previously [35], using 6–8 weeks-old female BALB/c mice (weight range, 18–20 g) obtained from the Laboratory Animal Center of the Academy of Military Medical Sciences, Beijing, China. E. faecalis 002 was grown to OD600 = 1.6, and the bacteria were centrifuged at 10,000×g, 10 min, 4°C. The bacterial pellet was resuspended in 1xPBS buffer. Bacterial inoculation titers were calculated by serial dilution and plating onto BHI agar plates for each experiment. Mice were injected with different inocula of the E. faecalis 002, i.e. 2×107, 2×108, 2×109 and 2×1010 CFU (colony forming unit) in 200 μl of 1xPBS buffer to determine the minimal lethal dose (MLD) that would cause 100% mortality over a 4-day period. Once the MLD was determined to be 2×109 CFU, 2xMLD (i.e. 4×109 CFU) was used as the infective inoculum (the challenge dose) in the mouse model. To determine the protection effect of the LysIME-EF1 and its variants in mice challenged with 2xMLD of the E. faecalis 002, a single dose (20 μg in 200 μl of 1xPBS buffer) of LysIME-EF1 proteins (WT LysIME-EF1, mLysIME-EF1, LysIME-EF1 C29A/H90A/N110A, K173E, K176E, R190E, and F184R/Y218A mutants) were administered intraperitoneally on the other side of abdomen 1 h post bacterial inoculation. Each experimental group contained 24 mice. In the blank control group, the infected mice were treated with equal volume (200 μl) of 1xPBS buffer. The survival rate for each experimental group was monitored every 8 h during the first 48 h and then every 12 h for up to 4 days post-infection.

Statistical analysis

Statistical significance was determined using the unpaired two-tailed Student’s t-test at a level of significance of P<0.05. Survival data were analyzed using a Kaplan–Meier survival analysis with a log-rank method.

LysIME-EF1 possesses efficient bactericidal spectrum against 29 clinical strains of E. faecalis.

A decrease in OD600 was used to evaluate the lytic activity of LysIME-EF1. WT LysIME-EF1 was used to lyse against E. faecalis. All assays were performed in triplicate, and the data are expressed as means ± SD (n  =  3).

(TIF)

Click here for additional data file.

The oligomerization of wild type LysIME-EF1 in solution.

(A) Size-exclusion chromatography (GE Healthcare, Superdex 200 increase, buffer: 20 mM Tris-HCl, pH 8.0, 150 mM NaCl) of WT LysIME-EF1. The CBD fragment of LysIME-EF1 was co-eluted with the full-length LysIME-EF1 at 15.5 mL, standard proteins containing Thyroglobulin (669 kDa), Aldolase (158 kDa), Conalbumin (75 kDa), Ovalbumin (44 kDa), Aprotinin (6.5 kDa) were performed in the same buffer. (B) Analytical ultracentrifugation (AUC) of the exact molecular mass of WT LysIME-EF1, which yielded a sedimentation coefficient of 3.264 S with a molecular mass of 55.6 kDa. The c(s) represents continuous (sedimentation coefficient distribution) analysis model.

(TIF)

Click here for additional data file.

The size-exclusion chromatography and purification of LysIME-EF-1 CHAP and CBD.

(A) Size-exclusion chromatography (GE Healthcare, Superdex 200 increase, buffer: 20 mM Tris-HCl, pH 8.0, 150 mM NaCl) of LysIME-EF1 CHAP, standard proteins containing Thyroglobulin (669 kDa), Aldolase (158 kDa), Conalbumin (75 kDa), Ovalbumin (44 kDa), Aprotinin (6.5 kDa) were performed in the same buffer. (B) The SDS-PAGE analysis of LysIME-EF1 CHAP and CBD.

(TIF)

Click here for additional data file.

LysIME-EF1 possesses a classical CHAP.

(A) The structure of the CHAP domain of LysIME-EF1. (B) The superposition of the CHAP domains of LysIME-EF1, LysK and LysGH15. (C) Structure-based sequence alignment of LysIME-EF1, LysGH15 and LysK CHAP domain. Strictly conserved residues are boxed in white on a red background and highly conserved residues are boxed in red on a white background. At the top of the sequences, the secondary structure elements of LysIME-EF1 CHAP are shown, and every 10 residues are indicated with a dot (.) shown above the sequences. Sequence alignment was generated by ClustalW (https://www.genome.jp/tools-bin/clustalw) and the figure was generated by ESpript 3 (http://espript.ibcp.fr/ESPript/ESPript/). The blue triangles indicate the residues involved in Ca2+ ion binding and the red triangles indicate the catalytic triad residues on CHAP domain.

(TIF)

Click here for additional data file.

Crystal structure of LysIME-EF1 CBD.

(A) Overview of LysIME-EF1 CBD monomer with labeled secondary elements. The LysIME-EF1 CBD monomer was shown in red color. (B) Topological diagram of LysIME-EF1 CBD with β-strands drawn in black arrows and the α-helices shown as cylinders.

(TIF)

Click here for additional data file.

LysIME-EF1 CBD is a homotetramer in solution.

(A) Size-exclusion chromatography (GE Healthcare, Superdex 200 increase, buffer: 20 mM Tris-HCl, pH 8.0, 150 mM NaCl) of LysIME-EF1 CBD and the mutant LysIME-EF1 CBD F184R/Y218A. The elution peaks for the two proteins were 17.7 ml and 18.7 ml, respectively. (B) The SDS-PAGE analysis of LysIME-EF1 CBD F184R&Y218A protein.

(TIF)

Click here for additional data file.

The N-terminal residue identification of the LysIME-EF1 CBD.

A blue colored monomer extracted from the structure of Fig 4A. The first four residues of LysIME-EF1 CBD including M168F169I170Y171 are shown in sticks and the 2Fo-Fc map was contoured at 1.0 σ level.

(TIF)

Click here for additional data file.

Sequence alignment of LysIME-EF1 CBD with CBD of other endolysins derived from E. faecalis phages.

Multiple sequence alignment of LysIME-EF1 CBD and CBD of other endolysins of E. faecalis phages produced by ClusterW (https://www.genome.jp/tools-bin/clustalw) and ESpript 3 (http://espript.ibcp.fr/ESPript/ESPript/). Every 10 residues are indicated with a dot (.) shown above the sequences. Strictly conserved residues are indicated in white on a red background. The blue triangles indicate the positive residues involved in putative cell-wall binding and the red triangles indicate the residues responsible for the oligomerization of the CBD.

(TIF)

Click here for additional data file.

The electrostatic surface of LysIME-EF1.

The positively charged patch is located on the planar surface of the LysIME-EF1 CBD tetramer. The putative cell binding residues K173, K1746 and R176 are shown in sticks. Positively and negatively charged regions are colored in blue and red, respectively.

(TIF)

Click here for additional data file.

Primers used to generate the LysIME-EF1 mutants in this study.

(TIF)

Click here for additional data file.

24 Oct 2019

Dear Prof. Ouyang,

Thank you very much for submitting your manuscript "Structural and functional insights into a novel two-component endolysin encoded by a single gene in Enterococcus faecalis phage" (PPATHOGENS-D-19-01755) for review by PLOS Pathogens. Your manuscript was fully evaluated at the editorial level and by independent peer reviewers. The reviewers appreciated the attention to an important problem, but raised some substantial concerns about the manuscript as it currently stands. These issues must be addressed before we would be willing to consider a revised version of your study. We cannot, of course, promise publication at that time.

We therefore ask you to modify the manuscript according to the review recommendations before we can consider your manuscript for acceptance. Your revisions should address the specific points made by each reviewer.

In addition, when you are ready to resubmit, please be prepared to provide the following:

(1) A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

(2) Two versions of the manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org.

Additionally, to enhance the reproducibility of your results, PLOS recommends that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see http://journals.plos.org/plospathogens/s/submission-guidelines#loc-materials-and-methods

We hope to receive your revised manuscript within 60 days. If you anticipate any delay in its return, we ask that you let us know the expected resubmission date by replying to this email. Revised manuscripts received beyond 60 days may require evaluation and peer review similar to that applied to newly submitted manuscripts.

There is additional documentation related to this decision letter. To access the file(s), please click the link below. You may also login to the system and click the 'View Attachments' link in the Action column.

[LINK]

We are sorry that we cannot be more positive about your manuscript at this stage, but if you have any concerns or questions, please do not hesitate to contact us.

Editor added:  It is essential to address all questions raised by the third reviewer for revised version.

Sincerely,

Gongyi Zhang

Associate Editor

PLOS Pathogens

Michael Otto

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Grant McFadden

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-2556-3526

***********************

Reviewer's Responses to Questions

Part I - Summary

Please use this section to discuss strengths/weaknesses of study, novelty/significance, general execution and scholarship.

Reviewer #1: Zhou and coworkers report the first structure of a known E. faecalis-targeted endolysin, LysIME-EF1 and find it posing multimeric composition, i.e., a full-length endolysin with three extra CBDs. A lysin (Lys170, the endolysin from enterococcal phage F170/08) with similar composition has been reported by reference 28 (Proenca D, Velours C, Leandro C, Garcia M, Pimentel M, et al. A two-component, multimeric endolysin encoded by a single gene. Mol Microbiol 2015，95: 739-753). The difference is that the current finding is based on the crystallographic study of LysIME-EF1, which is important for further understanding the action model of similar lysins against E. faecalis.

Reviewer #2: Lysins are hydrolytic enzymes produced by bacteriophages in order to cleave the host’s cell wall and are considered as a potential strategy for treatment of bacterial infections caused by multidrug-resistant pathogenic microorganisms. This study focuses on LysIME-EF1, an endolysin that specifically lyses Enterococcus faecalis, a major pathogen with high levels of antibiotic resistance.

In this study, the authors observed that expression of LysIME-EF1 results in two polypeptides, a full length product and a C-terminal CBD domain. Both polypeptides are necessary for the assembly of the functional lysin. They further determined the crystal structure of LysIME-EF1 at a high resolution. The structure revealed an asymmetric assembly containing one full length LysIME-EF1 and three CBD domains. Based on the structure, the authors identified potential key residues, designed mutants and assessed their cell binding capability. Overall, the study constitutes a novel insight into this therapeutically important enzyme and open the possibilities for the future structure-based design of lysins with higher activity and broader antimicrobial spectrum.

Reviewer #3: See attached review

**********

Part II – Major Issues: Key Experiments Required for Acceptance

Please use this section to detail the key new experiments or modifications of existing experiments that should be absolutely required to validate study conclusions.

Generally, there should be no more than 3 such required experiments or major modifications for a "Major Revision" recommendation. If more than 3 experiments are necessary to validate the study conclusions, then you are encouraged to recommend "Reject".

Reviewer #1: 1. Fig 1B and Fig 5B. The images show that the band intensity of the CBD (8 kD) is much lighter compared with that of the full length 30 kD LysIME-EF1. How to explain this result if it is correct that the purified LysIME-EF1 consists of a full-length of 30 kD LysIME-EF1 and three extra 8 kD of CBDs? In theory, the concentration of the CBD should be 3 times of that of the full length 30 kD LysIME-EF1 during SDS-PAGE electrophoresis. For instance, both bands of Lys170 in the SDS-PAGE showed similar density in the previous report (Ref. 20).

2. Fig 5C: Additional experiments should be done by mixing M168L/RBS LysIME-EF1 mutant with different ratio of the purified CBD to see if the lytic activity could be restored in order to confirm the multimeric composition.

3. Fig 1C, 3E and 6C: From the lysis curves of the mutants, it is hard to judge the activity difference between them. It is suggested to use the more sensitive culture methods, such as plate counting, to confirm the activity of different mutants, especially K173E, K176E, C29A, H90A, N110A, CHAP, CBD, and CHAP+CBD.

4. Fig 7 and all related text: The in-vivo infection model might just prove that the protection rates are related to the activity of the lysin used, which did not add more information to the structure of LysIME-EF1 and could be deleted.

Reviewer #2: (No Response)

Reviewer #3: See attached review

**********

Part III – Minor Issues: Editorial and Data Presentation Modifications

Please use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity.

Reviewer #1: 1. Page 2, lines 27-2: Most natural endolysins has limited host range, but there are several lysins with extended killing spectra, for instance, chimeric lysin ClyR, natural lysin PlySs2 and PlyV12. What does low persistence mean? Please rephase this sentence.

2. Page 4, lines 59: Enterococcus faecalis should be written as E. faecalis after appearing the first time. Correct this throughout the text.

3. Line 10, page 187, Supporting figure is missing.

4. Page 12, line 214：Interaction of lysin with bacterial cells is not infection. Post-infection could be changed to post-addition. The same mistake also goes to page 13, line 239.

5. Page 12, line 216: What does “ability” mean?

6. Page 15, line 277: 10 hours post infection? In the Figure 7 legend and the material method, it is mentioned that lysins were administrated 1 h post infection.

Reviewer #2: Nonetheless, I have the following comments about the manuscript:

1. The author reported that one full length LysIME-EF1 and three CBD domains co-assemble to a functional lysin. Is this the only possible assembly? Does the CHAP domain block the additional CHAP domain to assemble on a CBD tetramer?

2. The authors reported that the CHAP alone, CBD alone or a mixture of the two constructs, exhibited no bactericidal activity. This indicates that 1) CHAP domain and CBD tetramer can not co-assemble in vitro. 2) assembly with the CBD tetramer likely has an allosteric effect on the CHAP domain. The structural comparison of CHAP domain alone and that in complex with CBD will be helpful.

3. The presence of secondary translation products has been observed previously in other proteins, such as MthK (Jiang et al, Nature, 2002) and Tvok (Parfenova et al, J Biol Chem, 2007).

Reviewer #3: See attached review

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Hongping Wei

Reviewer #2: No

Reviewer #3: No

Submitted filename: This is an interesting manuscript about a multimeric bacteriophage endolysin.docx

Click here for additional data file.

26 Dec 2019

Submitted filename: Responses to PPATHOGENS-D-19-01755.docx

Click here for additional data file.

29 Jan 2020

Dear Prof. Ouyang,

Thank you very much for submitting your manuscript "Structural and functional insights into a novel two-component endolysin encoded by a single gene in Enterococcus faecalis phage" for consideration at PLOS Pathogens. As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. The reviewers appreciated the attention to an important topic. Based on the reviews, we are likely to accept this manuscript for publication, providing that you modify the manuscript according to the review recommendations.

Please prepare and submit your revised manuscript within 30 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to all review comments, and a description of the changes you have made in the manuscript.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Thank you again for your submission to our journal. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Gongyi Zhang

Associate Editor

PLOS Pathogens

Michael Otto

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

***********************

Reviewer Comments (if any, and for reference):

Reviewer's Responses to Questions

Part I - Summary

Please use this section to discuss strengths/weaknesses of study, novelty/significance, general execution and scholarship.

Reviewer #1: (No Response)

Reviewer #2: The authors have done a good job at explaining some key points and carrying out additional informative experiments. I believe the paper is ready for publication.

Reviewer #3: I was reviewer 3 in the original manuscript and this revised manuscript is significantly improved. In particular, the authors have added new bacteriolytic assays for most experiments, added additional purification gels, shown standards on gel filtration, and included additional details for some of the methodology. Moreover, they have adequately addressed most (not all) reviewer comments to a reasonable level of satisfaction. Nonetheless, there still remain a couple of errors and questions that were not fully addressed, requiring some additional revisions.

**********

Part II – Major Issues: Key Experiments Required for Acceptance

Please use this section to detail the key new experiments or modifications of existing experiments that should be absolutely required to validate study conclusions.

Generally, there should be no more than 3 such required experiments or major modifications for a "Major Revision" recommendation. If more than 3 experiments are necessary to validate the study conclusions, then you are encouraged to recommend "Reject".

Reviewer #1: (No Response)

Reviewer #2: (No Response)

Reviewer #3: Line numbers in the comments below refer to the marked up version of the revised manuscript.

- Lines 108/109 and 122/123. The authors state that their enzyme “displays a broader host range than other homologous endolysins, with bactericidal activity against 29 clinical strains of E. faecalis”. What do they mean by this comment? A broader host range than what? The fact that they tested 29 strains does not make this broad activity or broad spectrum since all 29 strains were the exact same species. All endolysins show activity toward at least one species and many show activity toward multiple species. Thus, to make a statement about a broader host range than other endolysins when they only test one species is nonsensical.

- Line 209. The authors reference figure S3C, which does not exist. I believe they mean to reference figure S4C.

- Line 225. The authors reference figures S4A and S4B, but I believe they mean to reference figures S5A and S5B.

- Lines 313-325. Proenca et al. 2015 did this same bioinformatics analysis on Lys170 and found the same results (i.e. methionine as first amino acid residue of the CBD and upstream Shine-Dalgarno sequence suggesting an alternative start codon site that generates oligomer). At the least, the authors need to acknowledge this previous work. Perhaps a statement saying that their analysis is consistent with the findings of Proenca et. al or something along these lines.

- Lines 332-336. The authors discuss residues 173, 176, and 190 and label these residues in figure 6A. This is fine; however, it would be very insightful to the reader if they also pointed out the location of these residues in Figure S9, which shows the electrostatic map of the planar surface so the reader can visualize exactly where these critical residues are located on the CBD surface.

- Lines 338-340. The authors should insert a reference to Figure 6B at the end of this sentence describing the FACS results.

- Line 346. The authors reference Figure 6B here, but I believe they mean to reference Figure S8 instead.

- Figure 7. This figure and all of the associated in vivo work is still fraught with issues. The authors have already shown that the mutants are deficient in antimicrobial activity, so to show the same thing in vivo is not very impactful. As another reviewer pointed out, the in vivo data provides no additional structural or mechanistic insight into LysIME-EF1, and I agree. I think the data should be removed. If the authors choose to keep the in vivo data, then many other issues will need to be addressed:

- In vivo, Issue 1. The authors did a pilot experiment to determine the minimum lethal dose (MLD) in lines 554-557. However, they never report what dose was experimentally determined to be the MLD. On line 353, it is stated that the mice were treated with 2 x 10^10 cfus, so one presumes this was the MLD, yet the figure legend (line 665) states the mice were treated with 4 x 10^9 cfus. Which is it? Also, the 4 x 10^9 dose was not one of the doses tested for MLD, so there is no rationale for how they derived this dose.

- In vivo, Issue 2. It is stated that mice were divided into 10 groups of 8 mice (lines 559-560). Figure 7 only shows 8 groups (WT, control, and 6 mutants). What is the purpose of the 2 remaining groups?

- In vivo, Issue 3. In both the original manuscript and the revised manuscript the authors consistently state that they used 8 mice for each group. However, in their rebuttal letter, they state they did the experiments in triplicate, indicating N=24 per group, not N=8 per group. This information is not stated anywhere in the manuscript. Were these new experiments since the original submission? If so, things need to be revised to state N=24 per group.

- In vivo, Issue 4. Statistics. Error bars are presented, which is odd in an in vivo experiment where normally the total N value is reported. As stated above, the authors never signify that experiments were done in triplicate, but this must be the case as they report 95.83% protection for WT (line 357), which cannot be derived from a group of 8 mice, but can be derived from a group of 24 mice (i.e. 23 out of 24 protected). They should report the data this way (i.e. 23/24). More importantly, the authors present no statistical data to compare results between groups, which is a significant flaw and must be addressed.

- In vivo, Issue 5. What is VREF (lines 549 and 665). This term is never defined.

- In vivo, Issue 6. The mice are monitored for 4 days according to the methods (lines 564-566). However, the authors state that the experiments were only 2 days in the results (line 356) and figure legend (line 667). Which is it?

- In vivo, Issue 7. Lines 354-356 are awkward. I think some sentences were combined or there is an extra period in line 355.

- In vivo, Issue 8. The results only provide life or death metrics. Was there any attempt to measure health status…i.e. even if a mouse is alive at day 2 or day 4 does not mean the mouse is healthy or otherwise will not die the next day, which is why these types of experiments should either last longer than a few days or should include some type of health monitoring matrix.

- In vivo, Issue 9. Graph. In the original manuscript, the authors presented the data in Kaplan Meier survival curves, which was appropriate. However, because several lines overlapped, they chose to present the data in a bar graph in the revised manuscript. This is not advisable because it loses all the temporal associated data displayed in a Kaplan Meier graph. I would go back to the KM graph even if multiple KM graphs are needed due to the overlap.

- Final conclusions. In lines 422-423 the authors state that residue R190 is essential for binding to the host cell wall and hence the biological activity of the endolysin. They then state in lines 424-26 that alteration of this residue could provide an opportunity to design endolysins with high specificity, efficacy, or broader spectrum. If R190 is essential for binding, it is not clear how alteration of this residue in particular can potentiate all of these actions. While I agree that the authors have described essential residues and mechanisms that increase our general understanding of endolysins and this general understanding may one day inform future engineering efforts, I think the comments specifically aimed at alteration of residue R190 without supporting data is a bit overstated.

- Turbidity reduction assays. In lines 521-528 the authors describe the general methods for the turbidity reduction assay. In the revised manuscript, all of the turbidity reduction assay experiments have been removed and replaced by the bacteriolytic assays. The one exception is the experiment with EDTA and metal ions presented in Figure 3D. Therefore, rather than general methods, these methods should specifically include experimental details for the EDTA/metal ion data. For example, specify how much EDTA was used, describe the dialysis steps, and describe the supplemental addition of metal ions.

- Entire manuscript. The entire manuscript should be better proofread by a native English speaker. This was a comment in the first review and the authors state in their cover letter that the language has been improved in the revised manuscript. However, this is not the case. Many language errors remain and some of the newly edited additions are even worse grammatically than the original version.

**********

Part III – Minor Issues: Editorial and Data Presentation Modifications

Please use this section for editorial suggestions as well as relatively minor modifications of existing data that would enhance clarity.

Reviewer #1: The revised manuscript is improved with additional experiments and has addressed most of my previous concerns. However, the authors chose to keep the results on animal model, which is all right to show some additional data but there are some things to be clarified.

1. Bacterial dose. In the materials and methods section, the authors evaluated the virulence of E. faecalis 02 with different bacterial dosage, i.e., 2×107, 2×108, 2×109 and 2×1010 CFU/mouse, to seek the minimal lethal dose and the authors said the minimal lethal dose of bacteria was used in the animal infection experiments (page 25, line 449). While, in the results section, bacterial dosage is ≥2 × 1010 cfu/mouse (page 15, line 272). And in the corresponding Figure 7 legend, the injected bacterial dosage is 4×109 cfu/mouse (page 30, line 519). These data are not consistent. What is the minimal lethal dose of the strain tested?

2. In the materials and methods section, survival rates of the animals were recorded every 8 h for the first 48 h and then every 12 h for up to 4 days post-infection. The survival rates shown in Figure 7 are the survival rates at 2 days post-infection? Why did the authors choose 4 days of observation and present the survival rates on day 2?

3. In both methods section and figure legend, a number of n=8 animal/group was given. What do the error bars in Fig 7 stand for? Did the authors repeat the experiments (n=8 animal/group) for several times? If that is the case, the repeat times should be given accordingly. Also it looks odd for percentages of 96% mortality rate (Line 275), 95.83% (Line 276) and 16.66% (line 279) 54.17% (line 281), and 45.83% (line 282). How were these percentages calculated if the denominator is 8?

Minor points：

-Methods for Fig 1C and 1D, as well as Fig 5D etc were missing.

--Page 6, line 91, Fig S1 (p30, line 523), broader bactericidal spectrum. Please indicate which lysin LysIME-EF1 was compared to?

--page 15, lines 274-275, the sentence is confusing. Please change to “… variants (20 ug) 1 h post-infection. Mice that were not….”.

--page 20, site-directed mutagenesis. All primers used should be given in the supplementary.

--Line 416, turbidity reduction assay. It’s better to use g instead of rpm? Please check other places.

--page 24, line 424, OD600 should be shown as OD600.

--Line 428, PBS is not negative control, it is blank control.

--page 29, line 514 and line 516, delete the second LysIME-EF1.

--Page 30, line 519, what VREF stands for? A full name should be given when it was shown first time.

--Page 30, line 523, delete r between broader and bactericidal.

Reviewer #2: (No Response)

Reviewer #3: (No Response)

**********

PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

Reviewer #3: No

Figure Files:

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, . PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at .

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, PLOS recommends that you deposit laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see

3 Feb 2020

Submitted filename: Response to PPATHOGENS-D-19-01755R2-0203.docx

Click here for additional data file.

10 Feb 2020

Dear Prof. Ouyang,

We are pleased to inform you that your manuscript 'Structural and functional insights into a novel two-component endolysin encoded by a single gene in Enterococcus faecalis phage' has been provisionally accepted for publication in PLOS Pathogens.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch within two working days with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Pathogens.

Best regards,

Gongyi Zhang

Associate Editor

PLOS Pathogens

Michael Otto

Section Editor

PLOS Pathogens

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

***********************************************************

Reviewer Comments (if any, and for reference):

10 Mar 2020

Dear Prof. Ouyang,

We are delighted to inform you that your manuscript, "Structural and functional insights into a novel two-component endolysin encoded by a single gene in Enterococcus faecalis phage," has been formally accepted for publication in PLOS Pathogens.

We have now passed your article onto the PLOS Production Department who will complete the rest of the pre-publication process. All authors will receive a confirmation email upon publication.

The corresponding author will soon be receiving a typeset proof for review, to ensure errors have not been introduced during production. Please review the PDF proof of your manuscript carefully, as this is the last chance to correct any scientific or type-setting errors. Please note that major changes, or those which affect the scientific understanding of the work, will likely cause delays to the publication date of your manuscript. Note: Proofs for Front Matter articles (Pearls, Reviews, Opinions, etc...) are generated on a different schedule and may not be made available as quickly.

Soon after your final files are uploaded, the early version of your manuscript, if you opted to have an early version of your article, will be published online. The date of the early version will be your article's publication date. The final article will be published to the same URL, and all versions of the paper will be accessible to readers.

Thank you again for supporting open-access publishing; we are looking forward to publishing your work in PLOS Pathogens.

Best regards,

Kasturi Haldar

Editor-in-Chief

PLOS Pathogens

​orcid.org/0000-0001-5065-158X

Michael Malim

Editor-in-Chief

PLOS Pathogens

orcid.org/0000-0002-7699-2064

Table 1

X-ray data collection and refinement statistics.

Dataset	Se-Met WT LysIME-EF1	LysIME-EF1 CBD
Data collection
Wavelength (Å)	0.9792	0.9792
Space group	P1211	I422
Cell dimensionsa, b, c (Å)α, β, γ (°)	47.03, 56.92, 91.5290.00, 96.29, 90.00	59.88, 59.88, 87.0190.00, 90.00, 90.00
Resolution range(Å)	46.75–1.75(1.813–1.75)	42.34–1.75(1.813–1.75)
unique reflections	47621(4713)	8236(552)
Rmerge	0.103	0.089
CC1/2	0.995	0.985
I/σ(I)	16.5(2.1)	24.3(2.3)
Completeness (%)	97.83(97.42)	98.81 (92.48)
Refinement
Resolution (Å)	1.75	1.75
Completeness (%)	97.83(97.42)	99.0(92.8)
Rwork	0.1793(0.3173)	0.2082 (0.2986)
Rfree	0.2163(0.3674)	0.2334 (0.3582)
No. atomsion	33971	5820
Ramachandran plot (%)Favored regionAllowed regionOutliers region	99.520.480.00	100.000.000.00

Values in parentheses are for highest-resolution shell.