Defective lytic transglycosylase disrupts cell morphogenesis by hindering cell wall de-O-acetylation in Neisseria meningitidis

[Editors’ note: the authors resubmitted a revised version of the paper for consideration. What follows is the authors’ response to the first round of review.]

Reviewer #1:

[…]

Major concerns:

1) The morphological defects in Figure 1 need to be quantified, followed by a statistical comparison to wild type and the ltgA complement derivatives. How many cells had these observations? How many ghost cells were observed?

We thank the reviewer for this keen observation. We made the quantification of the morphological abnormalities observed by confocal microscopy. Please see Figure 3. We were unable to do the same for the scanning electron microscopy because in this panel we took high-resolution images of the individual cells or cell clusters to get a closer look at the morphological defects. The ghost cells could be observed in all the clusters we scanned. An approximation of the ratio of ghost to intact cells is 1:10.

2) Figure 7 needs statistics for the cytokines.

This has been updated. The statistical analysis was done by Kruskal-Wallis non-parametric comparison against the complemented strain with a p-value < 0.01. Figure 7 is now the new Figure 6 in the revised manuscript.

3) IL-6 is proinflammatory cytokine, in most cases. The LtgA helix 30 mutant has reduced production of IL-6 (Figure 7). In this context, the statement in the Discussion (paragraph seven) that this mutant version of LtgA has pro-inflammatory potential does not make sense.

We thank the reviewer for this astute observation. This sentence has been deleted in the Discussion. The paper has been refocused and the discussion has been revised.

The mouse phenotypes are somewhat unclear – reduced inflammation but longer persistence of the helix 30 mutant. There is a lot of conjecture in the discussion – but little evidence to substantiate the claims made.

We apologize if this was unclear, however, infection with the helix-30 mutant did not result in longer persistence in the blood stream but it was cleared at a faster rate. See Figure 6. We have refocused the paper and modified the Discussion.

Reviewer #2:

The manuscript entitled, "Crippling the bacterial cell wall molecular machinery". by Williams et al. describes the function of a truncation mutant in LtgA that removes alpha helix 30. While the finding may eventually turn out interesting, but at this point the inconsistencies and poor quality of the writing makes it impossible to fully understand the work.

1)We thank the reviewer for the comments, but we disagree with this assessment. Based on structural knowledge we demonstrate how an LT (LtgA) can be manipulated to affect the integrity of the cell wall composition, create a defect in cell division and separation, and can be manipulated to impair the fitness of the human pathogen Neisseria meningitidis during infection. Our study shows that targeting LTs disrupts the function of its binding partner Ape1 whose stability and activity depends on LtgA. We demonstrate in this verified case, that besides degrading the PG, peptidoglycan degrading enzyme complexes can modulate their enzyme function and stability. These findings highlight the importance of these PG complexes in Neisseria biology and virulence, demonstrating their potential as future targets for drug development.

2) We are sorry that it was unclear, and we have taken significant steps to revise the manuscript both in focus and discussion. The revised paper is more focused and easier to read. It is now primarily focused on LtgA and Ape1 which has shortened the manuscript and improved the overall clarity.

For example, the authors describe a number of proteins that bind the Δ30 protein, but not the wildtype protein, based on Figure 5. However, they go on to validate the binding of these proteins to the wildtype LtgA in Figure 6 not the mutant.

We thank the reviewer for pointing this out. We apologize if this was unclear however the purpose of this experiment was misunderstood because it was not communicated effectively in our original draft of our paper. Therefore, we will explain in more detail below:

In our manuscript we used the Label-free mass spectrometry developed by Mathias Mann (PMID: 25363814). We utilized two quantification strategies: (1) spectral counting and (2) spectrometric signal intensity to measure the protein expression.

All samples were analyzed by mass spectrometry separately using the same protocol. The proteins from each sample were identified using two strategies: 1) the protein expression from each sample is estimated using the number of MS/MS spectra identifying peptide of the protein and 2) the intensity of the corresponding MS spectrum features of the protein. Our output is the list of detected proteins and the absolute or relative abundance of the proteins across all samples for each experimental run, which was done in triplicates. I would like to emphasize that the mass spectrometry can only identify enriched peptides in one sample versus another. It does not tell you whether proteins are absolutely interacting directly or whether all peptides belonging to every single protein was identified, however it can be utilized to define the molecular environment of the protein of interest. For example, in both the wild-type and the delta helix strain PBP1a, LtgE, LtgD were identified but they were more enriched in the delta helix strain. In contrast, Ape1 peptides were not identified by mass spec in either strain which could be because of the abundance of Ape1 in the bacteria or some other technical issue.

Finally, we realize that the mass spectrometry and validation experiments of the LtgA interactome broadens the story and therefore based on suggestions from reviewer 3 we have refocused the manuscripts on LtgA and Ape1. The proteomic experiments have been moved to the supplementary and its limitation and precise utility discussed briefly in the main text. We apologize again for this confusion.

They also claim that the Δ30 strain was cleared faster in mice, but they don't account for the slower growth rate of this crippled strain.

We apologize there was no detail explanation in the text. The Discussion has been modified. It now reads:

“Finally, to understand what role a defective LtgA that interferes with the normal function of the PG machinery plays in the pathogenesis of N. meningitidis we used a mouse infection model and showed ΔltgA^ltgAΔ30 strain of type N. meningitidis was cleared from the blood at a significantly faster rate than the wild-type, ΔltgA^ltgA or ΔltgA strains. […] Additionally, the helix deleted strain is hyperacetyaled and it is known that modification of the PG makes the cell wall more susceptible to complement-mediated lysis (Zarantonelli et al., 2013; Rosain et al., 2017), and PG modification is also associated with a decreased inflammatory response (Taguchi et al.,2018).”

There are many other problems with the work that needs to be ironed out before this could be considered for publication.

We thank the reviewer for your critic. We addressed all the problems that were pointed out. The study was thorough, and comprehensive as well as cross-disciplined and while the presentation was unclear the experiments were executed with the utmost rigor in mind. As mentioned above we have focused the manuscript on LtgA and Ape1 which improved its clarity, focus and message.

Reviewer #3:

[…]

From the results of these experiments the authors conclude that LtgA mediates the function of a hub of PG glycan strand modifying enzymes, orchestrates cell division, cell wall integrity, and contributes to pathogenicity. Although there are several interesting aspects to this study, most of these stated conclusions are not justified by the data. I believe the authors should carefully reconsider their interpretation of the results and consider changing the focus of the manuscript to the Ape1-LtgA connection.

We agree with the reviewer that such a statement would be too preliminary, and we revised our conclusion: Please see below:

“We devised a multidisciplinary approach using structural biology to show that it is possible to target a ‘hot spot’ on an LT in order to affect bacterial growth, cell division, and cell membrane integrity, which resulted in lethal consequences for the bacteria during host infection. […] A small molecule binding to alpha helix 30 could interfere with growth and simultaneously promote bacterial clearance, mimicking the enhanced clearance of the ltgA^ltgAΔ30 mutant in a murine infection model”.

My major concerns are as follows:

1) The observation that deletion of ltgA has no phenotype but that LtgA^Δ30 expression is detrimental to growth indicates that a defective LtgA is worse than not having it at all. One cannot conclude from this observation that LtgA orchestrates cell division, cell wall integrity, or the function of an important enzymatic complex. If this were true, then the loss of ltgA function would result in division and cell wall defects. The more likely explanation of the results is that the catalytically defective LtgA^Δ30 binds glycans and/or components of the PG biogenesis machinery and somehow impairs their function. Thus, the finding that LtgA^Δ30 expression is detrimental provides a clue to LtgA function, but more information about the mechanism behind the dominant-negative phenotype is needed to make strong statements/conclusions about LtgA's role in the cell. At present, it is hard to tell whether or not the detrimental effects of LtgA^Δ30 production has anything to do with its normal function, especially since it is expressed at about 3-5x greater concentration than the native enzyme in the reported experiments.

We thank the reviewer for this observation. We have changed the sentence in the Abstract to read:

“Here we show, that the active site of LtgA can be genetically modified to, affect the integrity of the cell wall, cell division, cell separation, and can impair the fitness of the human pathogen Neisseria meningitidis during infection.”

We apologize if this was unclear however, LtgA^Δ30 is produced relative to the wild type or the complemented strains at roughly 2 times greater amounts. Please see the exact quantification in Figure 2B and C. However, we agree and appreciate your point about the precision of the statement.

2) I think it is important to know whether an active site point mutant of LtgA has a similar phenotype to LtgA^Δ30. This will help determine whether LtgA^Δ30 is special in some way, or if any catalytically defective enzyme will cause similar effects on cell division etc.

This is a very interesting point, however the mutant we created is defective in substrate binding and therefore could affect the ability of the LtgA^Δ30 mutant catalyze efficiently the PG. However, this would be a part of a future more extensive exploration.

3) The observation that LtgA interacts with PBP1a and LtgE in vitro is interesting. However, none of the results indicate that these interactions have a functional consequence in cells or on the enzymes in vitro. Thus, the conclusion that LtgA mediates the function of a hub of PG glycan strand modifying enzymes is not an appropriate one to make without further studies.

The reviewer is correct. As pointed out above in the revised version we have removed this statement from the manuscript. Additionally, primarily based on your suggestion we have further refocused the paper on Ape1 and LtgA.

Given that LtgA^Δ30 expression affects PG acetylation in cells and LtgA promotes the activity of the deacetylase Ape1 in vitro, there is a much more convincing functional connection between these enzymes. I would therefore recommend rewriting the paper to focus on the dominant negative phenotype of LtgA^Δ30 and potentially other catalytically defective LtgA variants and how this uncovered a connection with Ape1 function.

We thank the reviewer for this very important insight, and we have revised the manuscript to focus on these two main players Ape1 and LtgA.

4) There is a disconnect between the co-immunoprecipitation/proteomic results and the in vitro SEC experiments. The co-IP data indicate that only LtgA^Δ30 associates with PBP1a and other proteins, yet the SEC experiments are performed with LtgA (wild type). If the LtgA (wild type) is capable of interacting with these proteins, why did they not show up as hits in the proteomic analysis? Does LtgA^Δ30 have a greater affinity for the interaction partners identified? This discrepancy raises concerns about the validity and reproducibility of the proteomic experiments.

We thank the reviewer for pointing this out. We apologize if this was unclear however, since the purpose of this experiment was largely misunderstood, because it was not communicated effectively in our original draft of our paper. We will explain below:

In our manuscript we used the Label-free mass spectrometry developed by Mathias Mann (PMID: 25363814). We utilized two quantification strategies: (1) spectral counting and (2) spectrometric signal intensity to measure the protein expression. All samples were analyzed by mass spectrometry separately using the same protocol. The proteins from each sample were identified using two strategies: 1) the protein expression from each sample is estimated using the number of MS/MS spectra identifying peptide of the protein and 2) the intensity of the corresponding MS spectrum features of the protein. Our output is the list of detected proteins and the absolute or relative abundance of the proteins across all samples for each experimental run, which was done in triplicates. I would like to emphasize that the mass spectrometry can only identify enriched peptides in one sample versus another. It does not tell you whether proteins are absolutely interacting directly or whether all peptides belonging to every single protein was identified, however it can be utilized to define the molecular environment of the protein of interest. For example, in both the wild-type and the delta helix strain PBP1a, LtgE, LtgD were identified but they were more enriched in the delta helix strain. In contrast, Ape1 peptides were not identified by mass spec in either strain which could be because of the abundance of Ape1 in the bacteria or some other technical issue.

We acknowledge that the Mass spec and validation experiments of the LtgA interactome broadens the story and therefore based on your suggestions we have further refocused the manuscripts on LtgA and Ape1. The proteomic experiments have been moved to the supplementary and its limitation and precise utility discussed briefly in the main text.

[Editors’ note: what follows is the authors’ response to the second round of review.]

Major issues requiring new experimentation:

1). The focus on the LtgA-Ape1 connection in the revised manuscript represents a major step in the right direction. However, with this renewed focus, it becomes important to provide some data that helps explain why Ape1 activity appears to be negatively impacted by expression of the LtgA^Δ30 mutant. At a minimum, Ape1 levels should be assessed by western blot to determine whether or not it is stably expressed in these cells.

We thank the reviewer for this important insight. Ape1 is produced comparatively across each strain. Please see Figure 4—figure supplement 2.

2) Additional support for the model that Ape1 is activated by LtgA in vivo should also be provided by examining whether or not Ape1 depletion results in a morphological defect that resembles that induced by the LtgA^Δ30 mutant.

We thank the reviewers for this suggestion. In our previous study we did note that a knockout of Ape1 resulted in the increased cell size of Neissera meningitidis (Veyrier et al. 2013). Here we examine the Δape1 strain for morphological defects using SEM or labeling the cell wall with FM-64. While there were no pronounced differences in the division and separation, we did find significant cell shape abnormalities and lysed bacteria i.e. irregular cell surfaces, high molecular weight blebs, asymmetrical diplococci, and lysed bacteria Please see Figure 3—figure supplement 2.

3) Testing whether or not expression of a catalytically defective LtgA mutant induces the same phenotype as the LtgA^Δ30 mutant is an important experiment. It will indicate whether or not there is something special about the LtgA^Δ30 mutant or if any catalytic defect will cause the observed effects. This experiment is not a difficult one, and given its importance, is not beyond the scope of the current manuscript or a subject for further study.

We thank the reviewer for the suggestion. A mutant of the catalytic residue did not result in any observable phenotype. Please see Figure 3—figure supplement 2.

4) The title of the manuscript is too vague and does not convey any useful information to a potential reader. Consider something sharper and more on-point such as, "A catalytically defective variant of a lytic transglycosylase disrupts cell morphogenesis by interfering with peptidoglycan deacetylation in Neisseria meningitidis". Or something similar to this effect.

We loved your suggestion. However, it was too long based on the character limit from eLife. So we revised it to: “A defective lytic transglycosylase disrupts cell morphogenesis by hindering peptidoglycan de-O-acetylation in meningococcus.”