Construction of Leaderless-Bacteriocin-Producing Bacteriophage Targeting E. coli and Neighboring Gram-Positive Pathogens

ABSTRACT Lytic bacteriophages are expected as effective tools to control infectious bacteria in human and pathogenic or spoilage bacteria in foods. Leaderless bacteriocins (LLBs) are simple bacteriocins produced by Gram-positive bacteria. LLBs do not possess an N-terminal leader peptide in the precursor, which means that they are active immediately after translation. In this study, we constructed a novel antimicrobial agent, an LLB-producing phage (LLB-phage), by genetic engineering to introduce the LLB structural gene into the lytic phage genome. To this end, lnqQ (structure gene of an LLB, lacticin Q) and trxA, an essential gene for T7 phage genome replication, were integrated in tandem into T7 phage genome using homologous recombination in Escherichia coli host strain. The recombinant lnqQ-T7 phage was isolated by a screening method using ΔtrxA host strain. lnqQ-T7 phage formed a clear halo in agar plates containing both E. coli and lacticin Q-susceptible Bacillus coagulans, indicating that lnqQ-T7 phage could produce a significant amount of lacticin Q. Lacticin Q production did not exert a significant effect on the lytic cycle of T7 phage. In fact, the production of lacticin Q enhanced T7 phage lytic activity and helped to prevent the emergence of bacterial populations resistant against this phage. These results serve as a proof of principle for LLB-phages. There are different types of LLBs and phages, meaning that in the future, it may be possible to produce any number of LLB-phages which can be designed to efficiently control different types of bacterial contamination in different settings. IMPORTANCE We demonstrated that we could combine LLB and phage to construct promising novel antimicrobial agents, LLB-phage. The first LLB-phage, lnqQ-T7 phage, can control the growth of both the Gram-negative host strain and neighboring Gram-positive bacteria while preventing the emergence of phage resistance in the host strain. There are several different types of LLBs and phages, suggesting that we may be able to design a battery of LLB-phages by selecting novel combinations of LLBs and phages. These constructs could be tailored to control various bacterial contaminations and infectious diseases.

The manuscript by Masuda et al. presents a very interesting approach and concept for development of a novel type of combined antimicrobial with dual specificity, where the gene encoding a gram-positive bacteriocin is expressed from a Gram-negative phage. In the work, a leaderless bacteriocin is combined with T4 phage is shown as a proof of principle, but the concept can in theory be expanded to numerous other combinations. In addition to demonstrating the dual activity, the authors also assess the burst size and lytic activity of the modified phage compared to the parental phage. In general, the manuscript is clearly written and results nicely presented and discussed.
Some comments: -The introduction could be expanded to provide more background information. Please describe more about the target range, biosynthesis and mechanism-of-action of lacticin Q in particular. For example, related to discussion (line 264-266); what is known about the activity of this bacteriocin when expressed in E. coli? -Furthermore, some background about the T4 phage multiplication (including the dependency of trxA) and life cycle would be useful in the introduction.
-Regarding production and release of LnqQ. The bacteriocin was not detected in the phage lysate and the authors discuss possible reasons for this. It would be interesting if the authors discuss possible strategies to increase the production of bacteriocin, eg. increasing the gene copy number, increasing the latency period.
-Along the same lines, the authors highlight the use of leaderless bacteriocins as ideal for the purpose, since they are non-modified and without a leader. However, if I interpret the results correctly, the leaderless bacteriocins are mostly released from the E. coli cells due to lysis and not export from intact cells. If this is ture, I believe other non-modified bacteriocins could be equally well suited for the purpose (as long as they are expressed/cloned without their leader). Please discuss.
- Figure 3 and line 135-137. When the parental T7 phage was exposed to Bacillus coagulans overlay the authors state that "the growth of B. coagulans was shown inside the plaques of T7 phage formed against E. coli". I do appreciate that the plaques are not as clear as in the other panels and that there is grown on the plaques. However, there also seems to be a zones with not growth at the edges of the plaques. Could the authors explain or speculate on this?
- Figure 4A. The setup of the figure is a bit confusing due to the placement of the "Bc (Alone)" exposed to LnqQ only. Consider reorganizing the figure to avoid confusion, e.g. by making a frame around this part of the figure.
-Line 27: the trxA-gene needs some kind of introduction. Please rephrase to shortly explain the function of the trxA gene. -Line 30: replace "plate" with "plates" -Line 43: add "the" before "Gram-negative" -Line 79: consider replacing "approach" with "classification". Remove the word "of" -Line 80. To my knowledge, "listeria-specific bacteriocins" are not widely used a group in the classification of bacteriocins. Would be more suitable to use e.g., "most non-modified bacteriocins" -Line 111: explain the expression dynamics of the T7 promoter -Line 274: consider rephrasing to "... LLB phage could also slow down development of resistance to ...". -Line 594: "10 10-8 PFU/ml". Please correct.
-References: formatting of bacterial names is lacking Reviewer #2 (Comments for the Author): The work describes the engineering of phage T7 to produce lnqQ. This is an interesting approach and the engineering of phages in such as way can be useful for manipulating bacterial communities. In its current form the work seems preliminary. The engineering of phages generally is still in its infancy, so the production of an engineered phage is quite an achievement. Phage T7 has been previously engineered using trxA as a selective marker, so the method per se is not novel, only the gene added (which is interesting). The previous work that developed the approach of trxA should be cited. Clearly phage mutants have been obtained and when used in a clever double-double agar overlay, there can be seen to be an inhibitory effect on Bacillus. The amount of lnqQ produced was not fully quantified and only tested on plates. If it inhibits growth when grown in liquid medium was not presented and would seem to be critical. If there is such tiny amounts, will the approach work if the bacteria are in liquid culture and not forced together as they are on plates ?
The data presented on one-step experiments is not conclusive. The text suggests there was a difference in burst size, but no statistical analysis was presented to support this. Given the small difference in burst size, it is not clear if the difference is simply a result of the variation that occurs when performing one-step experiments. To fully test if there are quantitative differences between the WT and mutant phage, the use of a phage virulence index (Storms 2020) would be more appropriate. This would incorporate all elements of phage replication The explanation of the differences in observed phage production, is not supported by the data presented.
L20 -word missing in the sentence L71 -clarify what is meant by stress ? Is this possibly better phrased as not stable in environmentally relevant conditions? L73-probably needs a reference here L105-Qimron work should be cited that showed this was essential L177 are these reported differences statistically different ? Looking at the overlap of error bars in Figure 5, it is unclear if this is the case. The error associated with these numbers should be reported and tested for significant difference L182-I am unclear on the rationale for this ? The authors have been through in testing this, but it is not clear why it would target a Gram-ve bacteria, when all published tests of this bacteriocin to date have shown it targets Gram +ve bacteria.
L189 -the change in the point when resistance is observed, is surprising. If lanQ does not effect growth of E.coli, why would it alter resistance ? This could be discussed further L214 -as mentioned previously this method has previously been used and is not a novel technique for T7, previous work using this technique should be cited. L216 -the method used for creating phage mutants in phage T7 is based on previously described methods: Qimron et al, Grigonyte et al and others have used for engineering phage T7 in the same manner. The fact this method has been used before should really be cited as the principle behind it is not new .
L230 -the differences between these two strains is intriguing. Also how accurate this increase is, is not clear. The level of production was not quantified between the two strains. Furthermore, the amount of phage produced appears to be the same between both phages based on the one step experiment. Certainly not a 20x difference in the number of phages produced. Why there would be a decrease in protein synthesis of just one protein (lnqQ) and not all phage proteins is not clear. This suggests trxA has very specific role in the production of lnqQ ? as well as its general role binding to the RNA polymerase subunits that allow transcription of T7.
L246 -all of the points raised around the use qPCR are true. But these have been raised previously with many other systems. eg qPCR has been used widely estimate burst size and latent period of cyanophages . Currently it reads as if this is the first time such an approach has been taken for phages, which it isnt , but might be for T7.

Preparing Revision Guidelines
To submit your modified manuscript, log onto the eJP submission site at https://spectrum.msubmit.net/cgi-bin/main.plex. Go to Author Tasks and click the appropriate manuscript title to begin the revision process. The information that you entered when you first submitted the paper will be displayed. Please update the information as necessary. Here are a few examples of required updates that authors must address: Please return the manuscript within 60 days; if you cannot complete the modification within this time period, please contact me. If you do not wish to modify the manuscript and prefer to submit it to another journal, please notify me of your decision immediately so that the manuscript may be formally withdrawn from consideration by Microbiology Spectrum.
If you would like to submit an image for consideration as the Featured Image for an issue, please contact Spectrum staff.
If your manuscript is accepted for publication, you will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail. Arrangements for payment must be made before your article is published. For a complete list of Publicat ion Fees, including supplemental material costs, please visit our website.
Corresponding authors may join or renew ASM membership to obtain discounts on publication fees. Need to upgrade your membership level? Please contact Customer Service at Service@asmusa.org.
Thank you for submitting your paper to Microbiology Spectrum.

Response to reviewers
Answer for Reviewer 1 Thank you very much for reviewing our manuscript and giving us helpful comments.
According to your kind suggestions, we revised our manuscript and revised parts are marked with yellow in the text.
1. Regarding the information about lacticin Q, as you suggested, we added more background information especially about mode of action and biosynthesis (line 96-104). So far, there have been no publication demonstrating directly lacticin Q expression in E. coli without any N-terminal extension, but from the result of previous study by Iwatani et al. (ref. 20), it is assumed that lacticin Q can be immediately active after translation in E. coli as well. In addition, it is not published data but when we tried to introduce expression plasmid with lacticin Q structure gene with T7 promoter into E. coli BL21 carrying T7 RNA polymerase, transformation could not succeed. Therefore, we used DH5α strain for homologous recombination to avoid leaked lacticin Q expression which damage E. coli.
2. As you recommended, we added more information about T7 phage, especially about its genome replication and the role of thioredoxin in it (line 106-110).
3. Regarding the production and release of LnqQ, we discussed shortly at line 255-257, and we also added another sentence at line 296-298. 4. As you suggested it is theoretically possible that not only LLB but other mature peptides from other non-modified bacteriocins will be used as candidates. So, we described this in discussion part at line 299-301. 5. Regarding the clear area between E. coli growth and B. coagulans growth in T7 phage halo (Fig. 3), we are sorry, but we have no concrete explanation based on any experiments, but we assume that this could be because of some compounds produced from E. coli and defused into halo area which inhibited growth of B. coagulans.
Because this is not actually related with our purpose in this study, we did not mention in our manuscript.
6. Regarding Fig. 4, we revised its setup as you suggested.
Response to minor comments Title: we replaced "against" with "targeting".
Line 25: we revised as you suggested.
Line 27: we added short explanation of trxA.
Line 111: we added explanation of T7 phage gene expression including T7 promoter at line 127-132.
Line 274: we revised as you suggested at line 291.
References: we corrected bacterial names.

Answer for Reviewer 2
Thank you very much for reviewing our manuscript and giving us helpful comments. According to your kind suggestions, we revised our manuscript and revised parts are marked with yellow in the text.

1.
Regarding previous reports about trxA by Qimron et al., we are very sorry that we forgot to add references, and as you suggested, we added these citations at proper places in revised manuscript.
2. Regarding the challenge for tiny productivity of lnqQ, as you suggested, this must be improved in future by such as 1) re-evaluating the position of LLB structure gene for suitable expression timing or 2) increasing the number of structure gene to enhance expression level. We discussed this shortly at line 255-257, and we also added another sentence at line 296-298 3. Regarding the one-step growth experiments, we used constructed sigmoidal curves for determining burst sizes but as you suggested, experimental raw data are more suitable to evaluate statistical differences. Therefore, we revised the part of one-step growth experiments at line 192-204. In revised manuscript, the significance of difference of burst size was evaluated by T-test and p value was 0.1329. The infectious times for both phages were also described more clearly.
The use of a phage virulence index by Storms is also good protocol to evaluate phage activity, but unfortunately, we do not have the plate reader machine which allow us to perform continuous bacterial culture experiments, and our protocol is also appropriate enough to evaluate the difference between T7 and lnqQ-T7 phages.

Response to minor comments
Line 20: we are sorry, but we cannot figure out what word is missing in this sentence.
Could you tell us more precisely?
Line 71: we revised as "not stable in environmentally relevant conditions".
Line 73: we combined two sentence and added reference (7)  In addition, it is not published data but when we tried to introduce expression plasmid carrying lacticin Q structure gene with T7 promoter into E. coli BL21 carrying T7 RNA polymerase, transformation could not succeed. Therefore, we used DH5α strain for homologous recombination to avoid leaked lacticin Q expression which damage E. coli. Line 230: as you mentioned, this differences between two host strains are of interest for us as well. However, unfortunately, we have no experimental or published information to explain specific impact of trxA on lacticin Q production. We hypothesized reduced protein synthesis in general as one of the reasons. Therefore, to avoid this trxA dependent problem, we are trying to establish new platforms with CRISPR-Cas systems in which we do not need to manipulate host strains genomes.
Line 246: we are very sorry that we did not realize the previous reports precisely demonstrating qPCR method to quantify phages. Therefore, we deleted the discussion part mentioning qPCR method, and we added reference 27 by Peng et al.  promoter, so they will be expressed by T7 RNA polymerase. Thank you for submitting a revised version of your paper. Your manuscript has now been accepted, and I am forwarding it to the ASM Journals Department for publication. You will be notified when your proofs are ready to be viewed.
The ASM Journals program strives for constant improvement in our submission and publication process. Please tell us how we can improve your experience by taking this quick Author Survey.
As an open-access publication, Spectrum receives no financial support from paid subscriptions and depends on authors' prompt payment of publication fees as soon as their articles are accepted. You will be contacted separately about payment when the proofs are issued; please follow the instructions in that e-mail. Arrangements for payment must be made before your article is published. For a complete list of Publicat ion Fees, including supplemental material costs, please visit our website.