Staphylococcus aureus counters organic acid anion-mediated inhibition of peptidoglycan cross-linking through robust alanine racemase activity

Reviewer #1 (Public review):

Summary:

The manuscript entitled "Staphylococcus aureus counters organic acid anion-mediated inhibition of peptidoglycan cross-linking through robust alanine racemase activity" by Panda, S et al. reports an extensive biochemical analysis of the result from a Tn screen that identified alr1 as being required for acetic acid tolerance. In the end, they demonstrate that reduced D-Ala pools in the ∆alr1 mutant lead to a drastic reduction in D-Ala-D-Ala dipeptide. They show that this is due to the ability of organic acid anions to limit the D-Ala-D-Ala ligase enzyme Ddl. They demonstrate that:

(1) Acetate exposure in the ∆alr1 results in reduced D-Ala-D-Ala dipeptide, but not the monomers.

(2) Acetate can bind to purified Ddl in vitro.

(3) This binding results in reduced enzyme activity.

(4) Other organic acid anions such as lactate, proprionate, and itaconitate can also inhibit Ddl.

The experiments are clearly described and logically laid out.

Comments on revised version:

Given that multiple reviewers noted that determining intracellular acetate levels would strengthen the impact of this manuscript, I still think the comment listed below should be dealt with. Radioactivity is not necessary for this. There are enzymatic kits that will allow for the accurate determination of acetate from a lysate of a known number of cells. This can be used to determine intracellular acetate levels.

(1) It is kind of tricky, but it is possible to measure intracellular acetate. That might be of interest to know where in the Ddl inhibition curve the cells actually are.

Reviewer #2 (Public review):

Summary:

In this manuscript, using Staphylococcus aureus as a model organism, Panda et al. aim to understand how organic acids inhibit bacterial growth. Through careful characterization and interdisciplinary collaboration, the authors present valuable evidence that acetic acid specifically inhibit the activity of Ddl enzyme that converts 2 D-alanine amino acids into D-ala-D-ala dipeptide, which is then used to generate the stem pentapeptide of peptidoglycan (PG) precursors in the cytoplasm. Thus, high concentration of acetic acid weakens the cell wall by limiting PG-crosslinking (which requires D-ala portion). However, S. aureus maintains a high intracellular D-ala concentration to circumvent acetate-mediated growth inhibition.

Strengths:

The authors utilized a well-established transposon mutant library to screen for mutants that struggle to grow in the presence of acetic acid. This screen allowed authors to identify that a strain lacking intact alr1, which encodes for alanine racemase (converts L-ala to D-ala), is unable to grow well in the presence of acetic acid. This phenotype is rescued by the addition of external D-ala. Next, the authors rule out the contribution of other pathways that could lead to the production of D-ala in the cell. Finally, by analyzing D-ala and D-ala-D-ala concentrations, as well as muropeptide intermediates accumulation in different mutants, the authors pinpoint Ddl as the specific target of acetic acid. In fact, synthetic overexpression of ddl alone overcomes the toxic effects of acetic acid. Using genetics, biochemistry, and structural biology, the authors show that Ddl activity is specifically inhibited by acetic acid and likely by other biologically relevant organic acids. Interestingly, this mechanism is different from what has been reported for other organisms such as Escherichia coli (where methionine synthesis is affected). It remains to be seen if this mechanism is conserved in other organisms that are more closely related to S. aureus, such as Clostridioides difficile and Enterococcus faecalis.

Weaknesses:

None noted. With new data the authors have satisfactorily addressed all the concerns of the previous version.

Author response:

The following is the authors’ response to the original reviews.

Public reviews:

Reviewer #1:

(1) Which allele is alr1, the one upstream of mazEF or the one in the lysine biosynthetic operon?

Alr1 is encoded by SAUSA300_2027 and is the gene upstream to mazEF. We have now incorporated this information in the manuscript (Line# 127).

(2) Figure 3B. Where does the C3N2 species come from in the WT and why is it absent in the mutants? It is about 25% of the total dipeptide pool.

In Figure 3B, C3N2 species results from the combination of C3N1 (from Alr1) and C0N1 (from Dat). The reason this species is completely absent in either of the two mutants is because it requires one D-Ala from both Alr1 and Dat proteins to generate C3N2 D-Ala-D-Ala.

(3) Figure 3D could perhaps be omitted. I understand that the authors attained statistical significance in the fitness defect, but biologically this difference is very minor. One would have to look at the isotopomer distribution in the Dat overexpressing strain to make sure that increased flux actually occurred since there are other means of affecting activity (e.g. allosteric modulators).

Thank you for the suggestion. We agree with the reviewer that the fitness defect observed after increased dat expression is relatively minor and have moved this figure to the supplementary section as Figure 3-figure supplement 1.

Although we attempted to amplify the fitness defect of dat expression by cloning dat on to a multicopy vector, we couldn't maintain its stable expression in S. aureus. This instability may be due to the depletion of D-Ala when dat is overexpressed. As a result, we switched to expressing dat from a single additional copy integrated into the SaPI locus, which was sufficient to cause the expected fitness defect, albeit a minor one.

(4) In Figure 4A, why is the complete subunit UDP-NAM-AEKAA increasing in each strain upon acetate challenge if there was such a stark reduction in D-Ala-D-Ala, particularly in the ∆alr1 mutant? For that matter, why are the levels of UDP-NAM-AEKAA in the ∆alr1 mutant identical to that of WT with/out acetate?

Thank you for raising this important point. We have addressed this in line# 299-302 and 451-455 of the revised manuscript. In short, we believe that the inhibition of Ddl by acetate significantly increases the intracellular pool of the tripeptide UDP-NAM-AEK, which then outcompetes the substrate (pentapeptide; UDP-NAM-AEKAA) of MraY. As a result, the intracellular concentration of the pentapeptide increases since it is no longer efficiently consumed by MraY. This explanation is also supported by a kinetic study conducted in Ref (1), where the competition between UDP-NAM-AEKAA and UDP-NAM-AEK as substrates for MraY is demonstrated.

(5) Figure 4B. Is there no significant difference between ddl and murF transcripts between WT and ∆alr1 under acetate stress? This comparison was not labeled if the tests were done.

Thank you for suggesting this comparison. The ddl and murF transcripts between WT and alr1 under acetate stress were significantly different. We have added this comparison to Figure 4B.

(6) Although tricky, it is possible to measure intracellular acetate. It might be of interest to know where in the Ddl inhibition curve the cells actually are.

Thank you for the suggestion. We agree this would have been an excellent addition to the manuscript. However, accurately measuring intracellular acetate would require the use of radiolabeled acetate (2), and we currently lack the expertise to do this experiment. However, since our study clearly shows that acetate-mediated growth impairment is due to Ddl inhibition, and the IC50 of acetate for Ddl is around 400 mM, we predict that the intracellular concentration must be close to or above this IC50 to observe the growth phenotypes we report.

Reviewer #2:

Although the authors have conclusively shown that Ddl is the target of acetic acid, it appears that the acetic acid concentration used in the experiments may not truly reflect the concentration range S. aureus would experience in its environment. Moreover, Ddl is only significantly inhibited at a very high acetate concentration (>400 mM). Thus, additional experiments showing growth phenotypes at lower organic acid concentrations may be beneficial.

Thank you for the suggestion. In response to the reviewer, we have measured growth at various acetate concentrations and demonstrate a concentration-dependent effect (Figure 1C).

We use 20 mM acetic acid in our study. In the gut, where S. aureus colonizes, acetate levels can reach up to 100 mM, so we believe our concentrations are physiologically relevant. When S. aureus encounters 20 mM acetate, the intracellular concentration can rise to 600 mM if the transmembrane pH gradient is 1.5 units, which is well above the ~400 mM IC50 we report for Ddl.

Another aspect not adequately discussed is the presence of D-ala in the gut environment, which may be protective against acetate toxicity based on the model provided.

Thank you for pointing this out. We agree that D-Ala from the gut microbiota could protect against acetate toxicity, and we’ve included this in the discussion. However, our study clearly indicates that S. aureus itself maintains high intracellular D-Ala levels through Alr1 activity which is sufficient to counter acetate anion intoxication.

Recommendation for the authors:

Reviewer #2:

Major Comments:

(1) In Line 85, authors indicate S. aureus may encounter a high concentration of ~100 mM acetic acid (extracellular?). Could the authors cite more (and recent) references indicating S. aureus encounters >100 mM acetic acid in the environment?

To the best of our knowledge, no studies have specifically examined whether S. aureus encounters high mM concentration of acetate in the gut. Line 85 was surmised from multiple studies: recent findings that S. aureus colonizes the gut (3, 4) and that the gut environment has high acetate levels (~100 mM) (5). In response to the reviewers request, more recent references supporting high acetate concentrations in the gut (6, 7) have been added in Line# 86.

(2) In Line 117, it is mentioned that S. aureus when grown in vitro at 20 mM acetic acid can accumulate ~600 mM acetic acid in the cytoplasm.

a. Does the intracellular concentration go up proportionally if grown in 100 mM acetic acid? Given the IC50 of acetic acid-mediated inhibition of Ddl is ~400 mM, I wonder how physiologically relevant this finding presented here is.

Thank you for the opportunity to explain this further. If S. aureus encounters a concentration of 100 mM acetate and its transmembrane pH gradient (pHin-pHout) is held at 1.5, the intracellular concentration of acetate could theoretically increase up to 3 M based on Ref (8). However, previous studies have shown that bacteria can lower the magnitude of transmembrane pH gradient by decreasing their intracellular pH to limit accumulation of anions within cells (9, 10).

Although our study shows that the IC50 of Ddl inhibition by acetate is relatively high (~400 mM), we believe it’s still relevant because just 20 mM of environmental acetate at a pH of 6.0 can raise the intracellular concentration of acetate to over 600 mM, which is well above the IC50 we report for Ddl. Moreover, since S. aureus may encounter high concentrations of acetate during gut colonization, we believe our findings are physiologically relevant.

b. Could the authors show concentration-dependent growth inhibition in alr::tn by titrating a range of acetic acid concentrations (for example 0, 0.5, 1, 5, 10, 20 mM)? Measuring intracellular acetate concentration may be beneficial as well.

Thank you for this question. We now provide data to support that acetate-mediated inhibition of the alr1 mutant is concentration-dependent (see Figure 1C).

c. It appears that there may be excess D-ala in the gut environment (PMIDs: 30559391; 35816159), which could counter the high acetate based on the model presented here. Could the authors clarify and/or include this information in the manuscript?

This is an excellent point, and we have now included it in the discussion (Line# 470-475). It is indeed possible that D-Ala produced by the gut microbiome may further enhance S. aureus resistance to organic acid anions, in addition to the inherent contribution of Alr1 activity.

(3) The following is not needed; however, it would be interesting if the authors could show that S. aureus cells grown in the presence of acetate are highly sensitive to cycloserine (which targets Alr and Ddl) compared to cells grown in the absence of acetate.

Thank you for the suggestion. We are currently studying D-cycloserine (DCS) resistance in S. aureus. Although we provide the data below for clarification, it is not included in the current manuscript as it is part of a separate study.

As the reviewer speculated, S. aureus is more susceptible to DCS when grown in the presence of acetate (see figure below). Normally, complete growth inhibition occurs at 32 µg/ml of DCS. However, with 20 mM acetic acid present, complete inhibition is achieved at just 8 µg/ml of DCS. Furthermore, the growth inhibition is completely rescued when externally supplemented with 5 mM D-Ala. We believe that DCS works synergistically with acetate to inhibit Ddl activity, and we are conducting additional studies to explore this further.

Minor Comments:

(1) Many commas are missing.

Missing commas are now incorporated.

(2) Line 77: disassociate --> dissociate

Corrected.

(3) Line 103: that --> which

Corrected.

(4) Lines 199-203: authors could have used gfp/luciferase reporter to test their hypotheses.

Thank you for the suggestion. Initially, we created GFP translational fusions for all the mutants mentioned in Line# 199-203. However, the fluorescence intensity was too low to test the hypothesis, as these were single-copy fusions inserted at the SaPI site of the S. aureus genome. Because of this limitation, we took advantage of the essentiality of D-Ala-D-Ala in S. aureus to report on various mutants instead of a fluorescent reporter. In hindsight, a LacZ reporter assay might have been equally effective.

(5) Line 339: It would be beneficial to introduce that Ddl has two independent ATP and D-ala binding sites.

We have now added that information (Line# 338-339).

(6) Is ddl an essential gene? If so, explicitly mention that.

Yes, ddl is an essential gene and we have now incorporated this information in Line 103.

(7) Line 354: shows a difference in density?

The use of the term “difference density” is a technical crystallographic term commonly used to connote density observed for ligands in X-ray crystal structures. In this case, it simply refers to the observed density that corresponds to the two acetate ions bound within the Ddl active site.

(8) Line 498: "Thus." Typo, change period to comma.

We have corrected as suggested in Line 496.

(9) Figure 1 legend says "was screen" instead of screened.

This is now corrected.

(10) Figure 1- Figure Supplement 1B: including data for alr2::tn dat::tn may ensure no redundancy (Lines 171-172). It is currently missing.

Thank you for the suggestion. We now include both alr2dat double mutant and the alr1alr2dat triple mutant in Figure 1 - Figure Supplement 1B. In addition we also show that the alr1alr2dat mutant is resuced by the addition of D-Ala in Figure 1 - Figure Supplement 1C. The mutant information is also added to Table S5.

(11) Figure 7: pentaglycine coming off of NAM is misleading. Remove untethered pentaglycine bridges.

We thank you for pointing this out. We have modified the figure in the manuscript as suggested by the reviewer.

(12) Are alr1/ddl cells (with limited 4-3 PG crosslink) less sensitive to vancomycin?

On the contrary, the alr1 mutant is slightly more sensitive to vancomycin compared to the wild-type strain (see Figure below). We believe this happens because the alr1 mutant incorporates less D-Ala-D-Ala into the peptidoglycan, reducing the number of targets for vancomycin. As a result, vancomycin may be able to saturate the available D-Ala-D-Ala targets on the cell wall at a lower concentration in the alr1 mutant than in the wild type strain, leading to increased sensitivity. We haven’t included this data in the manuscript as it is part of a separate study.

(13) Based on the structural studies, could the authors mutate the residues of Ddl involved in acetic acid binding, thereby making it resistant to acetic acid stress?

The residues that the acetate anion interacts with are located within the ATP-binding and D-Ala-binding sites of Ddl. Since these residues are essential for Ddl function, we are unable to mutate them.

(14) Microscopy to show the cell morphologies of wild-type and mutants exposed to acetic acid (and with D-ala supplementation) could be potentially interesting.

Thank you for the suggestion. We did perform microscopy, expecting changes in cell shape or size, but the results were unremarkable and not included in the manuscript.

References:

(1) Hammes WP & Neuhaus FC (1974) On the specificity of phospho-N-acetylmuramyl-pentapeptide translocase. The peptide subunit of uridine diphosphate-N-actylmuramyl-pentapeptide. J Biol Chem 249(10):3140-3150.

(2) Roe AJ, McLaggan D, Davidson I, O'Byrne C, & Booth IR (1998) Perturbation of anion balance during inhibition of growth of Escherichia coli by weak acids. J Bacteriol 180(4):767-772.

(3) Acton DS, Plat-Sinnige MJ, van Wamel W, de Groot N, & van Belkum A (2009) Intestinal carriage of Staphylococcus aureus: how does its frequency compare with that of nasal carriage and what is its clinical impact? Eur J Clin Microbiol Infect Dis 28(2):115-127.

(4) Piewngam P_, et al._ (2023) Probiotic for pathogen-specific Staphylococcus aureus decolonisation in Thailand: a phase 2, double-blind, randomised, placebo-controlled trial. Lancet Microbe 4(2):e75-e83.

(5) Cummings JH, Pomare EW, Branch WJ, Naylor CP, & Macfarlane GT (1987) Short chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut 28(10):1221-1227.

(6) Correa-Oliveira R, Fachi JL, Vieira A, Sato FT, & Vinolo MA (2016) Regulation of immune cell function by short-chain fatty acids. Clin Transl Immunology 5(4):e73.

(7) Hosmer J, McEwan AG, & Kappler U (2024) Bacterial acetate metabolism and its influence on human epithelia. Emerg Top Life Sci 8(1):1-13.

(8) Carpenter CE & Broadbent JR (2009) External concentration of organic acid anions and pH: key independent variables for studying how organic acids inhibit growth of bacteria in mildly acidic foods. J Food Sci 74(1):R12-15.

(9) Russell JB (1992) Another explanation for the toxicity of fermentation acids at low pH: anion accumulation versus uncoupling. Journal of Applied Bacteriology 73(5):363-370.

(10) Russell JB & Diez-Gonzalez F (1998) The effects of fermentation acids on bacterial growth. Adv Microb Physiol 39:205-234.