A Retro-Aldol Reaction Prompted the Evolvability of a Phosphotransferase System for the Utilization of a Rare Sugar

ABSTRACT The evolution of the bacterial phosphotransferase system (PTS) linked to glycolysis is dependent on the availability of naturally occurring sugars. Although bacteria exhibit sugar specificities based on carbon catabolite repression, the acquisition and evolvability of the cellular sugar preference under conditions that are suboptimal for growth (e.g., environments rich in a rare sugar) are poorly understood. Here, we generated Escherichia coli mutants via a retro-aldol reaction to obtain progeny that can utilize the rare sugar d-tagatose. We detected a minimal set of adaptive mutations in the d-fructose-specific PTS to render E. coli capable of d-tagatose utilization. These E. coli mutant strains lost the tight regulation of both the d-fructose and N-acetyl-galactosamine PTS following deletions in the binding site of the catabolite repressor/activator protein (Cra) upstream from the fruBKA operon and in the agaR gene, encoding the N-acetylgalactosamine (GalNAc) repressor, respectively. Acquired d-tagatose catabolic pathways then underwent fine-tuned adaptation via an additional mutation in 1-phosphofructose kinase to adjust metabolic fluxes. We determined the evolutionary trajectory at the molecular level, providing insights into the mechanism by which enteric bacteria evolved a substrate preference for the rare sugar d-tagatose. Furthermore, the engineered E. coli mutant strain could serve as an in vivo high-throughput screening platform for engineering non-phosphosugar isomerases to produce rare sugars. IMPORTANCE Microorganisms generate energy through glycolysis, which might have preceded a rapid burst of evolution, including the evolution of cellular respiration in the primordial biosphere. However, little is known about the evolvability of cellular sugar preferences. Here, we generated Escherichia coli mutants via a retro-aldol reaction to obtain progeny that can utilize the rare sugar d-tagatose. Consequently, we identified mutational hot spots and determined the evolutionary trajectory at the molecular level. This provided insights into the mechanism by which enteric bacteria evolved substrate preferences for various sugars, accounting for the widespread occurrence of these taxa. Furthermore, the adaptive laboratory evolution-induced cellular chassis could serve as an in vivo high-throughput screening platform for engineering tailor-made non-phosphorylated sugar isomerases to produce low-calorigenic rare sugars showing antidiabetic, antihyperglycemic, and antitumor activities.

The logic of their plan in this first paragraph is not clear. They start by saying that probably BL21 is missing the PTS genes for the transport of tagatose that are found in other Tag+ bacteria. Then they switch to talk about aldolases! They need to explain that in order to select for mutations leading to evolution of existing PTS genes to transport an alternative substrate they decided to overexpress in the bacteria a downstream gene required for tag metabolism. They should thus explain the expected degradation pathway and the choice of enzymes. Fig. 1 shows the gat (galacticol) PTS operon but its role and the presence of endogenous tagatose-1,6-P aldolase is not mentioned in the text although they do test the purified enzymes in vitro. L.94,95 This sentence is only clear to people who know that the KbaY and GatY aldolases are heterodimeric and only fully active with the subunits KbaZ and GatZ. At least they should add "required for full activity" to their sentence before reference (24). Why did not they include both subunits in their overproduction purification test? BL21(DE3) is used to express proteins for overproduction and purification from a plasmid (e.g. pET28a in their case) under control of the T7 promoter and usually produces large amounts of the protein in the cell. They are using it for their ALE with a high amount of IPTG presumably completely changing cellular physiology. Did they look at the protein profile of their strains? Did they think to try with a more physiological level of the cloned aldolase? The mutation in the lacUV5 promoter was crucial to reduce expression of the BL_GatY and hence might have reduced the time required to obtain YF1, if they had used a lower expression system. To this reader it does not seem to be the best strain to start with. Maybe they wanted the Gal minus to isolate arabinose isomerase mutants? Have others isolated AraA mutants with improved catalysis of galactose? How do their araA mutations compare to these? L. 101, 409 they use 0.05% fructose as "minimum (should be minimal) helper substrate" which will induce the fru genes including fruA the fructose PTS transporter which might be expected to transport another ketose sugar like tagatose or evolve to a new or wider specificity. They actually say that FruBA might function as a D-tagatose transporter (L.112-113) but it seems out of place here as they next talk of curing the BY_GatY plasmid expressing the aldolase, without explaining why. As the cured strain (YF1) did grow slowly on 0.05% fru+0.45% tagatose they conclude it had acquired mutations allowing utilization of tagatose, which could be in the expressed fructose PTS. Importantly it only took one additional transfer (if I've understood Fig.  2A correctly) to get YF2 which grew as well as YF1 with the BY_GatY plasmid. Interestingly they did not isolate mutations in the Fructose PTS genes themselves. This could either be because it is equally efficient at transporting fructose and tagatose or the high expression of the operon from the loss of Cra/FruR repression compensated for a lower affinity for tagatose. Was the presence of the 0.05% fructose with 0.45% tagatose important for their selection? Did they compare with glucose or another sugar? It seems that the way that their selection was set up with low fructose in the media they were expecting/directing that the pathway would be via the fructose operon.
L.124, 132 On reading it I wanted to know here what is the predicted effect of the lacUV5 promoter mutation expressing the T7 polymerase and hence its possible effect on BY_GatY expression from the T7 promoter? L. 180 maybe gives the answer, the mutation changes the lacUV -10 promoter to the original lac promoter TATAAT to TATGTT. However this does not agree with the GA to AT change listed as the lacUV5 mutation in Table S1). In any case they should say it is a mutation reducing the T7 expression here and refer to further discussion later, if necessary.
Compared to the limited details of the selection the analysis of the mutations obtained is rather long and also not always easy to follow (L.138 -222). This analysis would be more appropriate if they had several different mutations to compare rather than the three mutations expected from such a selection.
Other points: L.30 Peripheral PTS ? explain L. 91, L. 428 Class II aldolase Is this important? What is difference from class I ? L. 99 Not "Subsequently" better to say "In order to achieve this ...." L.106 Replace "in over 500h" with "after 500h" L.108 Serial transfer culture? Presumably cultureS. Explain of from what to what L.123 Cra is also known as FruR and it is its role as the fructose-specific repressor which is important here. The authors should state that Cra is FruR and the delta67 is removing the Cra/FruR binding site and derepressing the fruBAK operon.
L.142 onwards please refer to the figures where these results are shown. The text is not easy to follow. None of their growth curves give the original BL21(DE3) for comparison.
L.155 a phenotype e.g. growth on fructose is not REversed by growth on another sugar (which is a different phenotype).
L.158 loss of a specific regulator (AgaR) removes a sugar-operon specific repression and not general catabolite repression (generally due to cAMP, CAP and EIIA of the PTS) L.160 Their nomenclature of the strains carrying reconstructed mutations in BL21 is not orthodox but I appreciate their attempt to produce simple names for multimutated strains. The use of capital letters for the genes mutated and for the number of genes mutated is confusing and it might be easier if they use lower case letters for the in the single, double, triple (e.g. triple mutation tCFA). In theory the s, d and t are redundant with the number of letters in the name (e.g CFA is a triple mutation).
L.163 the growth curves for TCFA and TCFA(PTC) Fig. S2F are very different. Any comment? Surprisingly some of the combinations of mutations do affect growth on glucose.
L.181 ......reduced to that of the original.... The sentence is badly constructed. The mutation in the T7 RNAP promoter changed the -10 of the promoter from (... TATAAT) of the high expression lacUV5 promoter to .. (TATGTT) corresponding to the original lac promoter and should thus have reduced expression of the T7 polymerase and of its downstream targets. (Was this verified?) L. 186 Fig.S4A does not seem to show lag times. Please verify that the labels on the abscissa are correct. What does "except for D-Fru in a plasmid dependent manner" refer to ? Do they mean "did not exhibit a lag time on glucose or fructose, except for Y1 in the absence of the GatY plasmid"??
L.213 "located in GalN PTS and galactitol PTS" Do they mean that kbaY is required for the metabolism of sugars transported by two different PTS. The kbaY gene is located with genes for the galactitol PTS (Fig.1A).
L. 220 It is only novel in E. coli but a similar pathway exists already in Salmonella and Klebsiella! L.230 Fig. 3D What is the bar diagram to the right? L.260 fits well L.271 we need some preliminary information. Please say that arabinose isomerase can catalyse the formation of tagatose from galactose (ref). They take advantage of the fact that BL21(DE3) is missing genes of the Leloir pathway to show cloned arabinose isomerase can allow growth on galactose via the tagatose utilization pathway.
L.282 what do they mean "grow solely by enzymatic activity"? L. 291 YFI (BL_GAT/EC_Ara ML) Their strain contain two pET plasmids with different antibiotic resistances but are they the same origin of replication? In which case the copy numbers of the two high copy plasmids are going to vary. 50 microg/ml of ampicillin will soon be used up in any case. Fig. 4D and 4F why is the lag for use of galactose so much longer in 4F than 4D? Fig. 4D should have the AraA plasmid indicated like the araA mutation in 4C.
L. 414 "without antibiotics, and an inducer" reads as if it is without antibiotics but with an inducer. I suppose they mean "without antibiotics and IPTG" i.e. without the inducer L.555 "mutation(s) were introduced" it is unlikely they get a single mutation L.570 when cells HAD grown L.571 I do not understand why "growth was not faster than the growth rate of the original culture "?
L.734 what is the "clinker strategy"?

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: • Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER. • Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file. • Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file. For complete guidelines on revision requirements, please see the journal Submission and Review Process requirements at https://journals.asm.org/journal/Spectrum/submission-review-process. Submissions of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. " 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 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 Publication Fees, including supplemental material costs, please visit our website.

Response to Reviewers (Spectrum03660-22)
 Editor "Data Availability Statement" is missing in the "Materials and Methods" section. ▶We included it in the revised text (L616). Please clearly label your introduction "Introduction." ▶We labeled "Introduction" (L51). Please callout supplementary table s4 in the manuscript. ▶We revised the text for " Table S4" (L216 and L490) and "Table S3" (L437).

 Reviewer #1 (Comments for the Author):
This is very nice work. The authors successfully obtained E. coli mutants to utilize the rare sugar, and then took a systematic approach to determine the evolutionary trajectory of E. coli mutants to utilize the rare sugar. The mechanism by which E. coli evolved rare sugar utilization can initiate the utilization of new sugars. In my opinion, it adds valuable data to our understanding of sugar utilization in E. coli. ► We thank you so much for your nice appreciation for our paper.

 Reviewer #2 (Comments for the Author):
The authors have produced an E. coli B (BL21(DE3)) strain capable of growing on the rare ketose sugar tagatose. They start with an E. coli B strain BL21(DE3) in which they express aldolases capable of cleaving tagatose-1,6-phosphate as a crucial step in tagatose metabolism. They isolate a 1st strain which grows well on tagatose in the presence of the plasmid and from this, another strain no longer requiring the plasmid. Sequencing both strains identified mutations causing the overproduction of the fruBAK operon and a point mutation in fruK allowing further metabolism of tagatose to tagatose 1,6-P, the substrate of the aldolase. In the second strain, without the plasmid, an agaR mutation derepressed the expression of kbaYZ genes encoding an authentic E. coli tagatose1,6P aldolase. They have analysed three fruK mutations which appeared intermediately but only one of which fixed, and found lower activity towards the authentic substrate and some activity on alternative substrates. Subsequently they used their tagatose-utilising strain to select mutations in arabinose isomerase which allow better activity on galactose converting it to tagatose. They propose their system as a platform for engineering sugar isomerases. ► We greatly appreciate your critical specific comments, which help us to improve our manuscript clearer to the potential readers. Please, see our point-by-point responses to your specific comments below.  In the present study, however, we introduced a foreign gatY gene from B. licheniformis to E. coli, resulting in the occurrence of adaptive evolution to yield D-Tag+ E. coli mutant strains, and investigated how to gain the D-Tag preference through endogenous genes in the core metabolic pathways. Such an approach enabled us to reveal a new evolutionary path to gain promiscuous sugar preference via adaptive evolution primarily occurring in the fructosespecific fruBKA operon, which was not observed in both studies. This result reveals that the Fru PTS may evolve to use two different ketoses (i.e., D-Tag as well as D-Fru), implying the resilience of enteric bacteria in preferable sugar-limited environments. Furthermore, we found that tagatose usability was enhanced through additional mutations in endogenous regulators and non-coding regions for fine-tuned expression levels, even in the absence of a foreign aldolase gene. This intriguing result demonstrates that the evolved strain by fruBKA-based adaptation utilizes D-Tag. To clarify this issue, we rephrased the relevant parts in the revised text (L88-90)

E. coli strains cannot grow on tagatose but some Klebsiella and
The paper is not always easy to follow, lots of things need to be explained to make it easily comprehensible to the average reader. In particular for the selection procedure. ► As advised, we rephrased and explained the relevant parts throughout the paper to make them easily comprehensible to the average reader (See the specific responses to your comments below).
Retro-aldol reaction needs to be explained: it is a chemical term, reverse of a synthetic aldol condensation. Biochemists and molecular biologists are more familiar with the enzyme aldolase which performs the reaction and they should explain this. In fact, they start with genes expressing known aldolases so the retro-aldol term just adds an unnecessary complication. ► As advised, we explained the retro-aldol reaction to retro-aldol cleavage reactions derived from sugar aldolase (L110-111).
L. 112, 339 etc "Rare sugar-rich conditions", "Rare sugar-utilizing strains" etc are ambiguous. The authors want to say (rare sugar)-rich conditions and (rare sugar)-utilizing strains but in fact it reads as rare "sugar-utilizing" strains i.e., as if strains utilizing sugars are rare! To be clear the authors need to write "strains utilizing rare sugars" etc ► As advised, we revised these terms throughout the text (L77, L79, L125, L356, and L381). Did they try their protocol in a Gal + strain or is the delta gal implicated? ► We tried to use a Gal -E. coli BL21(DE3) strain lacking galTKEbecause we sought to obtain a platform cell for screening tailor-made L-arabinose isomerase (AI) for the production of Dtagatose. For this reason, we required the strain with Gal negative phenotype (optionally including the ompTand lon -) B strain with T7 promoter, which facilitates us to express recombinant AI mutant library plasmid.  Fig. 1G). Therefore, to accelerate the adaptive evolution, we tried to heterologously express the single subunit of BL_GatY, which has better activity than EC_GatY and EC_KbaY (Fig. 1D), known as tagatose aldolase. Furtheremore, the purpose of the aldolase activity assay in the present study was not to identify whether there was an aldolase activity as a complete complex, but to compare and select an appropriate aldolase with the highest aldol cleavage activity as a single subunit. Therefore, we tested EC_KbaY and EC_GatY subunits separately only for their monomeric aldolase activity.
4 BL21(DE3) is used to express proteins for overproduction and purification from a plasmid (e.g. pET28a in their case) under control of the T7 promoter and usually produces large amounts of the protein in the cell. They are using it for their ALE with a high amount of IPTG presumably completely changing cellular physiology. Did they look at the protein profile of their strains? Did they think to try with a more physiological level of the cloned aldolase? The mutation in the lacUV5 promoter was crucial to reduce expression of the BL_GatY and hence might have reduced the time required to obtain YF1, if they had used a lower expression system. To this reader it does not seem to be the best strain to start with. Maybe they wanted the Gal minus to isolate arabinose isomerase mutants? ► Yes, we also found that when we used E. coli K-12 BW25113 WT strain harboring pBAD-BL_GatY plasmid with a relatively lower expressed promoter than a strong T7lac promoter in  ► As you pointed out, we corrected it (L113) and rephrased the relevant part to explain the purpose of plasmid curing (L128-L129). For your information, we clearly stated that we also resequenced YF2 strain the following BL_GatY plasmid curing (L133-L144). As a result, we found that only agaR deletion was the main genetic mutation in YF2 strain, which is also mainly described in the results of this study (Fig. 2B and Table. S1).
Was the presence of the 0.05% fructose with 0.45% tagatose important for their selection? Did they compare with glucose or another sugar? It seems that the way that their selection was set up with low fructose in the media they were expecting/directing that the pathway would be via the fructose operon. ► Yes, when we designed the experiment for ALE, we also tested another type of combination (0.05% glucose (D-Glc) and 0.  Table S1). In any case they should say it is a mutation reducing the T7 expression here and refer to further discussion later, if necessary. ► We corrected mistyped lacUV5 SNVs in (Table S1) Compared to the limited details of the selection the analysis of the mutations obtained is rather long and also not always easy to follow (L.138 -222). This analysis would be more appropriate if they had several different mutations to compare rather than the three mutations expected from such a selection. ► As shown in Fig 2, there were six mutations found in the ALE-induced strains, and only four of them, CraBS Δ67, FruK_A39S, AgaR PTC , and T7RNAP pro , were directly related to the sugar uptake system. We found these mutations are the essential factors for the D-tag availability (L.

133-143). The result part you pointed out deals with how the loss of tight regulation occurred in a sugar-specific PTS to gain sugar usability. First, we verified each of the mutation effects in the growth profile on D-Tag by introduction of single, double, and triple mutations (L159-182). In turn, we tried to identify minimal genetic elements for rare sugar utilization. We analyzed the correlations between genotypic and phenotypic changes by 14 mutant strains with all types of mutant combinations (L183-L210). And we finally figured out how the novel flux was induced and controlled by the minimal mutations (L. 211-239). If we have a chance, we would like to analyze other mutations that are not essential but for enhancing flux stabilization, which we did not refer to in this study.
Other L. 99 Not "Subsequently" better to say "In order to achieve this ...." ► As you recommended, we replaced it (L. 111).

L.123 Cra is also known as FruR and it is its role as the fructose-specific repressor which is important here. The authors should state that Cra is FruR and the delta67 is removing the Cra/FruR binding site and derepressing the fruBAK operon.
► As you advised, we additionally described it (L137, 138).

L.142 onwards please refer to the figures where these results are shown. The text is not easy to follow.
None of their growth curves give the original BL21(DE3) for comparison. ► As advised, we rearranged figure 2 and included the descriptions of corresponding results cited with figures (L156~L158). In this text, we introduced mutations to WT with a stable growth profile (Fig. 1B) and tried to compare the growth rate and lag time between the created mutant strains for the three sugars media not with WT.
L.155 a phenotype e.g., growth on fructose is not reversed by growth on another sugar (which is a different phenotype). ► As advised, we deleted it (L173).

L.158 loss of a specific regulator (AgaR) removes a sugar-operon specific repression and not general catabolite repression (generally due to cAMP, CAP and EIIA of the PTS) ► As advised, we corrected it (L177)
L.160 Their nomenclature of the strains carrying reconstructed mutations in BL21 is not orthodox but I appreciate their attempt to produce simple names for multimutated strains. The use of capital letters for the genes mutated and for the number of genes mutated is confusing and it might be easier if they use lowercase letters for the in the single, double, triple (e.g. triple mutation tCFA). In theory the s, d and t are redundant with the number of letters in the name (e.g. CFA is a triple mutation). ► As you advised, we replaced upper case with lower case to indicate the number of mutated genes throughout the text (L160~L217).
L.163 the growth curves for TCFA and TCFA(PTC) Fig. S2F are very different. Any comment?
8 Surprisingly some of the combinations of mutations do affect growth on glucose. ► Yes, as you mentioned, there are differences in growth curves of tCFA and tCFA PTC stains in fructose media (S2F). We did not try further analysis to compare the effects of premature termination codon (PTC) and whole gene deletion in the repressor-coding gene. To our speculation based on these data, there are AT-rich regions in upstream and downstream of the agaR coding region. We believe that PTC mutation has more advantageous in genomic stability than deletion of agaR. It is currently unknown what effect the wild-type of AgaR ( L. 186 Fig. S4A does not seem to show lag times. Please verify that the labels on the abscissa are correct. ► To clarify this, we re-depicted Fig. S4A. What does "except for D-Fru in a plasmid-dependent manner" refer to? Do they mean "did not exhibit a lag time on glucose or fructose, except for Y1 in the absence of the GatY plasmid"?? ►Two optimized strains (tCFA and tCFA PTC ) with no lag time on D-Tag and D-Glc ( Fig. 2I and  Fig. S4A) showed lags when cells were grown on D-Fru with and without the pET_BL_GatY plasmid ( Fig. S2F and Fig. S4B). As you can see in (Fig. S4B), other recombinant strains harboring pET_BL_GatY plasmid showed lag on D-Fru due to a physiological burden by heterologous expression. To clarify this, we replaced the relevant sentence (L204).

L.211 Fig.S5C Why should deletion of fruK (specific for fructose-1P) reduce growth on glucose in YF2?
► Generally, D-Glc and D-Fru catabolic pathways are highly linked. It is uncertain, but there has been reported that FruK can cause a conformation change of transcriptional regulator Cra through physical contact interaction (Singh et al. BioRxiv 2017). Therefore, it is considered that the absence of FruK may have delayed glucose availability by increasing the amount of the normal conformation of Cra, which inhibits the expression of glucose operon. This remains to further validate the effect of the bacterial fructose-1p kinase (FruK) on glucose availability at the molecular level.
L.213 "located in GalN PTS and galactitol PTS" Do they mean that kbaY is required for the metabolism of sugars transported by two different PTS. The kbaY gene is located with genes for the galactitol PTS (Fig.1A). ► We corrected wrong information (L230); As shown in Fig. 1A and Fig. S5B, the kbaY gene is located in the GalN PTS only, and not in the galactitol PTS.
L. 220 It is only novel in E. coli but a similar pathway exists already in Salmonella and Klebsiella! ► As we described above, Salmonella and Klebsiella have independent tag operons (tagKTH) along with the fru genes (STM2204, STM2205). We changed the term "novel" to "new route" (L129).
L.230 Fig. 3D What is the bar diagram to the right? Fig.3F What is the diagram to the right? ► As you pointed out, we added the missing information about the bar diagram in Fig. 3D and 3F (L806-L811).

L.260 fits well ► We revised it (L277).
L.271 we need some preliminary information. Please say that arabinose isomerase can catalyse the formation of tagatose from galactose (ref). They take advantage of the fact that BL21(DE3) is missing genes of the Leloir pathway to show cloned arabinose isomerase can allow growth on galactose via the tagatose utilization pathway.

► As you advised, we included additional information on L-arabinose isomerase (AI) that can catalyze the formation of D-Tag from D-Gal (Lee et al. AEM 2004) (L290-291).
L.282 what do they mean "grow solely by enzymatic activity"? ► This means that the platform host can grow on D-Gal only through AI activity for D-Gal to D-Tag. We also included an additional explanation to clarify this (L299).
L. 291 YFI (BL_GAT/EC_Ara ML) Their strain contain two pET plasmids with different antibiotic resistances but are they the same origin of replication? In which case the copy numbers of the two high copy plasmids are going to vary. 50 microg/ml of ampicillin will soon be used up in any case. ► Yes, we also concerned about the compatibility of two pET plasmids since the two plasmids are of the same ori. In our case, we tested the pET_Duet system, but in the confirmation of expression through SDS-PAGE in a time-dependent manner, we concluded that the two plasmids system was much more effective for expression than the Duet plasmid system (see below). Furthermore, we used the two types of antibiotics, 50 µg/ml of ampicillin and 25 µg/ml of kanamycin, and culture transfers were conducted to fresh media with the presence of essential elements for growth; it was believed to prevent segregation of both plasmids. L. 414 "without antibiotics, and an inducer" reads as if it is without antibiotics but with an inducer. I suppose they mean "without antibiotics and IPTG" i.e. without the inducer ► As you recommended, we revised it (L430).
L.555 "mutation(s) were introduced" it is unlikely they get a single mutation ► We corrected it (L570).

L.570 when cells HAD grown ► We corrected it (L585).
L.571 I do not understand why "growth was not faster than the growth rate of the original culture"? ► As you pointed out, we rephrased it (L.586).
L.734 what is the "clinker strategy"? ► clinker is a pipeline for generating a gene cluster map. We deleted the unnecessary word "strategy," and we have mentioned and citied the paper ( Thank you for submitting your manuscript to Microbiology Spectrum. When submitting the revised version of your paper, please provide (1) point-by-point responses to the issues raised by the reviewers as file type "Response to Reviewers," not in your cover letter, and (2) a PDF file that indicates the changes from the original submission (by highlighting or underlining the changes) as file type "Marked Up Manuscript -For Review Only". Please use this link to submit your revised manuscript -we strongly recommend that you submit your paper within the next 60 days or reach out to me. Detailed instructions on submitting your revised paper are below.

Link Not Available
Below you will find instructions from the Microbiology Spectrum editorial office and comments generated during the review.
ASM policy requires that data be available to the public upon online posting of the article, so please verify all links to sequence records, if present, and make sure that each number retrieves the full record of the data. If a new accession number is not linked or a link is broken, provide production staff with the correct URL for the record. If the accession numbers for new data are not publicly accessible before the expected online posting of the article, publication of your article may be delayed; please contact the ASM production staff immediately with the expected release date.
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.
Rereading this paper I still find it difficult to follow and it is a shame that the authors do not have access to a competent English speaker who can understand and clearly present their results to a wide audience. I suppose it will be understandable to an interested specialist. In general, although grammatically correct, the English sentence construction is not good and thus not easy to follow.
They have produced a detailed rebuttal to my numerus questions and suggestions to the first version but have made only minimal changes to the actual text, apparently they do not think other readers will have similar questions. E.g. they explain in the rebuttal the advantages of their system for constructing a tagatose-utilizing strain for selection of AI mutations compared to other published methods but do not explain that to the reader. machinery. Staff Comments:

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: • Point-by-point responses to the issues raised by the reviewers in a file named "Response to Reviewers," NOT IN YOUR COVER LETTER.
• Upload a compare copy of the manuscript (without figures) as a "Marked-Up Manuscript" file.
• Each figure must be uploaded as a separate file, and any multipanel figures must be assembled into one file. For complete guidelines on revision requirements, please see the journal Submission and Review Process requirements at https://journals.asm.org/journal/Spectrum/submission-review-process. Submissions of a paper that does not conform to Microbiology Spectrum guidelines will delay acceptance of your manuscript. " 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 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 Publication Fees, including supplemental material costs, please visit our website.

► We have included sequence identity values in the revised text (L243245)
L. 290 as a candidate to be a D-galactose isomerase, which after mutagenesis might be responsible for the conversion of galactose to tagatose. They might like to explain that any tagatose produced from galactose by AI needs to be phosphorylated at both 1 and 6 positions to produce the substrate for the KbaY aldolase. They presumably assume that the 1 position could be phosphorylated by FruA, if it is acting intracellularly (and not part of the PTS membrane transporter) thus allowing the second phosphorylation by the FruK A39S allele they have isolated. ► Our comprehensive data indicated that intracellular D-gal is converted to D-tag by L-AI expressed in E. coli BL21(DE3), and, in turn, D-tag is phosphorylated to yield either D-tag-1P or D-tag-6P by fructose-specific PTS EIIAB components of membraneanchored FruA (EIIBC) and cytosolic FruB directed towards the intracellular side. It has been reported that PfkA and PfkB (also known as Fru-6P kinase) are involved in phosphorylation of Tag-6-P to generate Tag-1,6-BP (J. Babul J. Biol. Chem. 1978). However, our unpublished data below demonstrated that the essential components for D-tag catabolism are not PfkAB, but rather FruKBA (unpublished data C and D). Likewise, wild-type BL21(DE3) harboring AI and BL_GatY plasmids did not grow on galactose. Only the BL21(DE3) mutant strain with mutations in fructoserelated genes and harboring the AI plasmid grew on D-gal, indicating that PfkA and PfkB are not relevant to supporting growth on D-tag. Accordingly, it is plausible that initiation of phosphorylation and D-tag catabolism are related to FruBKA not only via extracellular uptake of tagatose, but also through intracellularly produced tagatose by AI.
Unfortunately, this is not yet clear because we failed to obtain tag-1P to determine the primary substrate for FruK_A39S among tag-1P and tag-6P. Thus, we have re-depicted Figs. 1C, 3C, and 4A to reflect the fact that the specific position initially phosphorylated is unclear.  Fig. 6A and B, with NMR spectra of Fru1P, here? Is this relevant to the failed synthesis of Tag1P?? Did they synthesis Fru1P? but not Fru6P and Tag6P? ►Yes, we synthesized D-Fru 1-P (L571) and obtained D-Fru 6-p and D-Tag 6-P from Sigma. As advised, we have moved Fig. S6 to the Materials and methods section (L542) and renamed ' Fig. S8' in the Supplementary information (L5075).

L.273 Why do they cite Supp
L.500 Did they really measure their 96 well plates with an Ultra8000 spectrophotometer, which seems to be a standard dual beam spectrometer and not a plate reader? ► We have corrected this (L530).

Fig. 1C
They could say here that BL21(DE3) is also missing genes for galactose catabolism as well as the tagatose uptake machinery. ► As advised, we have corrected this (L808).