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

Staphylococcus aureus, weak acids, acetate, D-alanyl-D-alanine ligase, Alanine 21 racemase 22 23 Running Title: Organic acid anions inhibit Ddl 24 25 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 2

26 Weak organic acids are commonly found in host niches colonized by bacteria, and they can 27 inhibit bacterial growth as the environment becomes acidic. This inhibition is often attributed to 28 the toxicity resulting from the accumulation of high concentrations of organic anions in the 29 cytosol, which disrupts cellular homeostasis. However, the precise cellular targets that organic 30 anions poison and the mechanisms used to counter organic anion intoxication in bacteria have 31 not been elucidated. Here, we utilize acetic acid , a weak organic acid abundantly found in the 32 gut to investigate its impact on the growth of Staphylococcus aureus . We demonstrate that 33 acetate anions bind to and inhibit D-alanyl-D-alanine ligase (Ddl) activity in S. aureus . Ddl 34 inhibition reduces intracellular D-alanyl-D-alanine ( D-Ala-D-Ala) levels, compromising 35 staphylococcal peptidoglycan cro ss-linking and cell wall integrity. To overcome the effects of 36 acetate-mediated Ddl inhibition, S. aureus maintains a high intracellular D-Ala pool through 37 alanine racemase (Alr1) activity and additionally limits the flux of D-Ala to D-glutamate by 38 controlling D-alanine aminotransferase (Dat) activity. Surprisingly, the modus operandi of 39 acetate intoxication in S. aureus is common to multiple biologically relevant weak organic acids 40 indicating that Ddl is a conserved target of small organic anions. These findings suggest that S. 41 aureus may have evolved to maintain high intracellular D-Ala concentrations, partly to counter 42 organic anion intoxication. 43 44 45 46 47 48 49 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 3 Significance 50 Under mildly acidic conditions, weak organic acids like acetic acid accumulate to high 51 concentrations within the cytosol as organic anions. However, the physiological consequence of 52 organic anion accumulation is poorly defined. Here we investigate how the acetate anion 53 impacts S. aureus. We show that acetate anions directly bind Ddl and inhibit its activity. The 54 resulting decrease in intracellular D-Ala-D-Ala pools impacts peptidoglycan integrity. Since 55 acetate is a weak inhibitor of Ddl, mechanisms that maintain a high intracellular D-Ala pools are 56 sufficient to counter the effect of acetate-mediated Ddl inhibition in S. aureus. 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 4 Organic acids produced by host and bacterial metabolism are critical determinants of 72 infection outcomes (1, 2). During infection, the host macrophages produce millimolar amounts 73 of itaconate, a dicarboxylic acid known to inhibit bacterial growth (3). Conversely, many 74 bacterial pathogens and the gut microflora excrete short-chain organic fatty acids, which exhibit 75 immunomodulatory functions and can skew the host response during infection (4, 5). Upon 76 entry into the bacterial cell, organic acids can become toxic to bacteria when they disassociate 77 in the cytosol as protons and organic anions. The influx of protons can result in cytoplasmic 78 acidification and prove lethal for some pathogens if not adequately controlled (6). Similarly, 79 organic anions have been shown to accumulate to toxic levels in the bacterial cytoplasm (7) . 80 However, the precise consequences of organic anion toxicity and the mechanisms pathogens 81 employ to withstand the effects of anion perturbations within cells are not clearly understood. 82 Here we focus on the response of Staphylococcus aureus to acetic acid, which is th e 83 primary end-product of glucose catabolism under aerobic conditions. S. aureus also likely 84 encounters high concentrations (up to 100 mM) of acetic acid and other short-chain fatty acids 85 produced by human gut microbiota during intestinal colonization (8). On average, 20% of adults 86 carry S. aureus in their intestines (9), and the burden there often surpasses that found in nasal 87 passages by more than three orders of magnitude establishing the gut as a primary site for S. 88 aureus colonization (10). We have previously shown that excreted acetic acid can promote 89 cytoplasmic acidification in cultures of S. aureus, especially when the external environment 90 becomes sufficiently acidic (pH< 5) (11). Cytoplasmic acidification promotes protein oxidation 91 and triggers a staphylococcal ClpP-dependent damage response that eliminates unfit cells from 92 the population (12). In contrast, in mildly acidic environments (pH 5.5-6.5), although S. aureus 93 actively buffers its intracellular environment against acidification, the transmembrane pH 94 gradient (ΔpH) of S. aureus will drive the accumulation of millimolar quantities of acetate anions 95 into the cytoplasm. Previous studies in Escherichia coli have shown that acetate intoxication 96 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 5 causes an osmotic imbalance that can transiently be accommodated by the efflux of 97 physiological anions like glutamate (7). In addition, acetate anions have also been reported to 98 impact enzymes in the methionine biosynthetic pathway, resulting in a toxic accumulation of 99 homocysteine and a reduction in intracellular methionine leading to growth inhibition of E. coli 100 (13, 14). However, it remains unclear if these effects are common to other bacteria. 101 Here, we demonstrate that the primary target of acetate intoxication in S. aureus is Ddl. 102 This crucial enzyme produces the D-Ala-D-Ala dipeptide, that is incorporated into peptidoglycan 103 precursors and is necessary for cell wall cross-linking. We also demonstrate that carbon flux 104 through alanine racemase and a tight control of Dat activity increases the cytosolic D-Ala pools 105 to counter acetate-mediated inhibition of Ddl. Importantly, these phenotypic effects are not 106 unique to acetate but are conserved across multiple biologically important organic acid anions. 107 Therefore, we propose that S. aureus may have evolved to maintain a high intracellular D-ala 108 pool partly to offset the inhibition of Ddl by organic anions typically encountered during human 109 colonization. 110

111 Alanine racemase counters acetate intoxication 112 To identify genetic determinants that counter the effects of acetate intoxication, we 113 screened the Nebraska Transposon Mutant Library (NTML) for mutants sensitive to 20 mM 114 acetic acid in Tryptic Soy Broth (TSB) media, pH 6.0. Under these conditions, S. aur eus 115 maintains its intracellular pH approx. 1.5 units above the external pH (15) and is estimated to 116 accumulate over 600 mM acetate in the cytosol (16). The NTML strains were grown under static 117 conditions at 37 °C, and the extent of growth was determined at 24 h by measuring the optical 118 density at 600 nm (OD 600). As a control, we performed an identical screen without acetic acid 119 supplementation. We normalized the growth of each mutant in both screens ( ± acetic acid) to 120 their isogenic wild-type (WT) strain. A comparison of growth indices (OD 600 Tn-mut/WT) for each 121 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 6 mutant in the presence and absence of 20 mM acetic acid revealed that most mutants clustered 122 close to an index of 1 in the plot (Figure 1A), which suggested that most mutants tolerated 123 acetate intoxication reasonably well. A few mutants that grew poorly following acetate 124 intoxication due to inherent growth defects were observed close to the plot diagonal, whereas 125 those mutants that did not have intrinsic growth deficiencies were located further away from the 126 diagonal. Among the latter class of mutants, we observed that the alr1 mutant had the most 127 substantial reduction in growth when subjected to acetate stress (Figure 1A, B). To confirm that 128 the acetate- dependent growth defect of the alr1 mutant was not due to polar effects, we 129 complemented the mutant by inserting a functional copy of alr1 under the control of its native 130 promoter into the S. aureus pathogenicity island (SaPI) attachment site. Genetic 131 complementation completely restored the alr1 mutant phenotype to WT levels (Figure 1B). 132 These results suggest that acetate intoxication impairs the growth of S. aureus in the absence 133 of a functional alanine racemase. Further supporting this conclusion, we could reduce acetate 134 toxicity in the alr1 mutant by culturing this strain in glucose-free TSB media, which alleviates 135 carbon catabolite repression and activates TCA cycle-dependent acetate catabolism ( Figure 136 1C) (17) . Conversely, the inactivation of citrate synthase ( citZ), the first enzyme of the TCA-137 cycle, re-imposed acetate toxicity in the alr1 mutant when cultured in glucose-free TSB media 138 (Figure 1C). 139 Acetate intoxication alters the intracellular D-Ala-D-Ala pools 140 Alr1 catalyzes the conversion of L-Ala to D-Ala during staphylococcal growth (Figure 1-figure 141 supplement 1A). The D-Ala is further converted to D-Ala-D-Ala dipeptide by the ATP- dependent 142 Ddl (Figure 1-figure supplement 1A) and incorporated into peptidoglycan (PG) muropeptide, 143 thus playing a crucial role in PG biosynthesis, cross-linking, and integrity (18, 19). Therefore, 144 we hypothesized that under acetate stress, low concentrations of D-Ala in the alr1 mutant might 145 concomitantly reduce D-Ala-D-Ala concentrations in the cell resulting in a growth defect. To test 146 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 7 this hypothesis, we determined the intracellular pool of D-Ala-D-Ala using liquid 147 chromatography-tandem mass spectrometry (LC-MS/MS). In regular growth media (TSB), we 148 observed that the inactivation of alr1 decreased the D-Ala-D-Ala pool by approximately 80% 149 compared to the WT strain (Figure 1D). However, following acetate intoxication the level of D-150 Ala-D-Ala was depleted by more than 99% (Figure 1D). The external supplementation of D-Ala 151 (5 mM) in the media fully restored the growth of the alr1 mutant to WT levels under acetic acid 152 stress (Figure 1E), which suggests that increased intracellular D-Ala pools can overcome the 153 detrimental impact of acetate intoxication. 154 The depletion of D-Ala-D-Ala following acetate intoxication is surprising since S. aureus is 155 predicted to have two additional pathways that can synthesize D-Ala and channel it to the 156 production of this dipeptide. For instance, S. aureus harbors a second predicted alanine 157 racemase (Alr2) that could compensate for the lack of Alr1 activity (Figure 1-figure supplement 158 1A). Alternatively, Dat, which catalyzes the formation of D-Ala from pyruvate and D-glutamate 159 (D-Glu), may functionally complement the alr1 mutant under acetate stress (Figure 1-figure 160 supplement 1A). However, the lack of functional complementation from these alternate 161 pathways of D-Ala biosynthesis following acetate intoxication suggests that not all metabolic 162 routes to D-Ala are operational or that regulatory bottlenecks limit pathway activity. To test 163 these possibilities, we constructed a series of mutants in which all three predicted routes of D-164 Ala biosynthesis (alr1 , alr2 and dat) were disrupted either individually or in various 165 combinations and performed growth assays (Figure 1-figure supplement 1B). Surprisingly, we 166 observed that the inactivation of alr1 and dat simultaneously ( alr1dat mutant) was synthetic 167 lethal in S. aureus, suggesting that alr1 and dat were the sole contributors of D-Ala in S. aureus. 168 Indeed, the supplementation of D-Ala fully restored the growth of the alr1dat mutant (Figure 1-169 figure supplement 1C). 170 The inactivation of alr2, either alone or in combination with other D-alanine-generating 171 enzymes, did not affect growth (Figure 1-figure supplement 1B). This observation suggests that 172 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 8 alr2 is unlikely to be a functional alanine racemase under the growth conditions tested. 173 Collectively, these results indicate that Dat activity accounts for D-Ala production in the absence 174 of alr1, but its contribution is insufficient to counter acetate intoxication. 175 Insufficient translation of dat impacts the alr1 mutant following acetate intoxication 176 Since Dat activity contributes to D-Ala production in the alr1 mutant, we questioned why Dat 177 is insufficient to sustain D-Ala-D-Ala pools under conditions of acetate intoxication. One possible 178 explanation may relate to the maintenance of osmotic balance by S. aureus . It has been 179 proposed that the intracellular accumulation of acetate anions may bring about an efflux of L/D-180 Glu from cells to adjust for osmolarity, thus exhausting one of the key substrates for Dat activity 181 and limiting D-Ala production (7). However, this hypothesis is improbable since the expression 182 of dat from a multicopy vector rescued the alr1 mutant from the effects of acetate intoxication 183 (Figure 2-figure supplement 1A), suggesting that the intracellular D-Glu pools are sufficient to 184 support D-Ala production through Dat activity. Alternatively, we hypothesized that the alr1 185 mutant's heightened sensitivity to acetate toxicity could be due to a decrease in dat 186 transcription which would effectively reduce intracellular D-Ala. However, we found no 187 detrimental effect of acetate intoxication on dat transcription in the alr1 mutant (Figure 2-figure 188 supplement 1B). Together, these observations raise the possibility that the depletion of D-Ala-D-189 Ala in the alr1 mutant following acetate intoxication may arise from a post-transcriptional 190 regulatory bottleneck that limits dat from meeting the demand for intracellular D-Ala. 191 In S. aureus, dat is part of a bicistronic operon (Figure 2A). The first gene, pepV, encodes 192 an extracellular dipeptidase (20, 21). Transcriptional start site (TSS) mapping of the pepV-dat 193 operon by the adaptor and radioactivity-free (ARF-TSS) method revealed a 30-nucleotide 194 untranslated region (5'-UTR) extending upstream from the pepV initiation codon. The 5'-UTR 195 includes a Shine-Dalgarno motif (ribosome binding site, SD1) upstream of the pepV start codon 196 (Figure 2A). In addition, a second SD motif (SD2) associated with dat was identified within the 197 pepV coding region (Figu re 2A) and did not overlap with the pepV termination codon. The 198 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 9 location of SD2 within pepV suggests that the insufficient production of D-Ala by dat following 199 acetate intoxication could be attributed to suboptimal translation of dat. This could occur as 200 ribosomes (70S) that are moving from SD1 may interfere with the translation of dat from SD2. 201 To test this hypothesis, we engineered a nonsense mutation in pepV (alr1pepVQ12STOP mutant) 202 that would prevent the ribosomes originating from SD1 from moving forward (Figure 2A). 203 However, the alr1pepVQ12STOP mutant grew poorly compared to the alr1 mutant following 204 acetate intoxication (Figure 2B). This suggested that the translation of dat is coupled to that of 205 pepV presumably through stable mRNA secondary structures that form within pepV. These 206 structures may not be effectively resolved in the alr1pepVQ12STOP mutant due to the absence of 207 ribosome traffic on pepV mRNA. 208 As an alternative approach to determine if SD2 positioning within pepV impeded dat 209 translation, we deleted pepV along with SD1 in the alr1 mutant (alr1pepVΔSD1-467, Figure 2A). In 210 the resulting strain, dat translation was under the sole control of its native SD2. Remarkably, 211 the alr1pepVΔSD1-467 mutant did not display a heightened sensitivity to acetate stress and grew 212 identical to the WT strain following acetate intoxication (Figure 2C). Similarly, an alr1 mutant in 213 which dat was linked to SD1 ( alr1pepV mutant, Figure 2A) also phenocopied the WT strain 214 following acetate intoxication (Figure 2D). Notably, the observed growth differences in 215 alr1pepVΔSD1-467, alr1pepV and alr1pepVQ12STOP mutants following acetate intoxication did not 216

from any changes in dat transcription (Figure 2E). These findings collectively suggest 217 that the native promoter elements, as well as the SD sites of pepV and dat can independently 218 support the robust expression and translation of dat to levels required for countering acetate 219 intoxication. However, the genetic arrangement of the dat translation initiation region (TIR) 220 within pepV offered tight control of dat translation and prevented cells from producing sufficient 221 enzyme following acetate intoxication. 222 Why is the Dat tightly controlled? 223 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 10 The need to tightly control Dat activity suggests that flux between D-Ala and D-Glu pools 224 must be carefully balanced during staphylococcal growth. To gain insight into this process, we 225 profiled the mass isotopologue distribution (MID) of D-Ala-D-Ala in the WT, alr1, and dat 226 mutants under isotopic steady-state conditions using 13C3 15N1-L-Ala as the tracer during growth 227 experiments in chemically defined medium (CDM). The flux of 13C3 15N1-L-Ala through Alr1 228 should result in 13C3 15N1-D-Ala production (Figure 3A, D-Ala retains labeled nitrogen). On the 229 other hand, staphylococcal alanine dehydrogenases (Ald1 and Ald2) catalyze the conversion of 230 13C3 15N1-L-Ala to 13C3-pyruvate and finally 13C3-D-Ala through Dat activity (Figure 3A). Thus, the 231 labeled nitrogen in 13C3 15N1-L-Ala is lost as 15N1-NH4 when fluxed through the Ald/Dat pathway 232 (Figure 3A). Since the intracellular pools of D-Ala are converted to D-Ala-D-Ala, the MID of the 233 latter metabolite should mirror the isotopologue ratios of D-Ala produced from either Alr1 or Dat 234 activities. 235 LC-MS/MS analysis revealed that ~ 80% of the intracellular D-Ala-D-Ala pool had 236 incorporated the labeled L-Ala supplemented in media (fractional contribution, 0. 80). As 237 expected, the majority (~ 55%) of the D-Ala-D-Ala in the WT was composed of the C 6N2 238 isotopologue (in which both units of D-Ala contain labeled carbon and nitrogen), which 239 suggested that alr1 was the major contributor of D-Ala in S. aureus (Figure 3B). Surprisingly, 240 the sole contribution of dat activity (C6N0, C3N0, C 0N0) to D-Ala-D-Ala was less than 1% in the 241 WT strain and D-Ala-D-Ala isotopologues with at least one D-Ala originating from dat activity 242 (C6N1, C3N1, C3N2) although readily observed , were still in the minority. However, the D-Ala-D-243 Ala originating from Dat activity expanded substantially upon alr1 mutation (Figure 3B). 244 Inactivation of dat itself displayed few differences in the MID of D-Ala-D-Ala, compared to the 245 WT strain (Figure 3B). These results suggest that flux through Dat is most likely driven toward s 246 D-Glu in the WT strain rather than D-Ala. Only upon inactivation of alr1 does the Dat activity 247 reverse towards the production of D-Ala. 248 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 11 To confirm these predictions, we measured the levels of 15N1-D-Glu in the WT, alr1, and dat 249 mutants following growth with the 13C3 15N1-L-Ala tracer. Consistent with Dat activity funneling D-250 Ala to D-Glu in the WT, approximately 78% of the D-Glu pool in the WT strain was 15N labeled. 251 Furthermore, we observed that inactivation of dat resulted in the complete depletion of 252 intracellular levels of 15N1-D-Glu (Figure 3C). Inactivation of alr1 also had a similar outcome with 253 loss of 15N1-D-Glu pools due to the lack of 13C3 15N1-D-Ala in this mutant (Figure 3C). Together, 254 these results strongly suggest that in the WT strain, Dat activity diverts D-Ala towards D-Glu 255 production. 256 Given the critical need to produce D-Ala-D-Ala during acetate intoxication, any diversion of 257 its precursor pool (D-Ala) to D-Glu through Dat activity is bound to decrease cell fitness and thus 258 may justify its tight translational control. To test this hypothesis, we determined the mean 259 competitive fitness (w) of cells that overexpressed dat compared to those that had native levels 260 of expression. Accordingly, we performed coculture competition assays of the WT strain with 261 an isogenic mutant strain that either harbored an empty vector (pAQ59) integrated into the 262 SaPI chromosomal site or a vector containing dat under control of its native promoter (pAS8), 263 following acetate intoxicati on. Consistent with increased Dat activity in the WT strain being 264 detrimental to the cell, the mean competitive fitness of the dat overexpressing strain was 265 significantly lower (w 4h= 0.91) in the exponential growth phase than its isogenic WT strain that 266 harbored the empty vector ( w4h= 1.26) (Figure 3D). Collectively, these results suggest that Dat 267 catalyzes the production of D-Glu in the WT strain, and its tight regulation prevents excessive 268 flux of D-Ala to D-Glu which is necessary to maintain cell fitness following acetate intoxication. 269 Acetate intoxication impacts PG biosynthesis 270 Since acetate intoxication ultimately affects D-Ala-D-Ala pools (Figure 1D), we predicted 271 potential alterations to PG biosynthesis and cell wall integrity. To test this hypothesis, we 272 quantified various cytosolic PG intermediates in the WT strain by LC -MS/MS analysis. Acetate 273 intoxication caused a significant increase in the intracellular pools of multiple PG biosynthetic 274 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 12 intermediates, including Uridine diphosphate N-acetylglucosamine (UDP-NAG), UDP-N-275 acetylmuramic acid (UDP-NAM), UDP-NAM-L -Ala, UDP-NAM-L -Ala-D-Glu-L-Lys and UDP-276 NAM-L-Ala-D-Glu-L-Lys-D-Ala-D-Ala (UDP-NAM-AEK AA) in the WT strain when compared to 277 the unchallenged control (Figure 4A). However, the growth of the WT strain was slightly 278 inhibited by acetic acid (Figure 1B), which suggests that the observed accumulation of PG 279 intermediates may have been due to an imbalance between the rates of PG biosynthesis and 280 growth. Notably, the alr1 mutant showed higher levels of UDP-NAM-AEK compared to the WT 281 and the dat mutant following acetate intoxication (Figure 4A), indicating a metabolic block in the 282 production of UDP-NA M-AEKAA due to insufficient D-Ala-D-Ala. The effect of this metabolic 283 block is also evident from the increased transcription of ddl and murF (Figure 4B) which encode 284 enzymes that incorporate D-Ala-D-Ala into PG precursors, suggesting a greater need to 285 maintain peptidoglycan cross-linking following acetate intoxication. 286 Unsurprisingly, the dysregulation of D-Ala-D-Ala homeostasis following acetate intoxication 287 was also reflected in the extent of cell wall cross-linking in the WT, alr1 and dat mutants. 288 Muropeptide analysis revealed that acetate intoxication in the WT strain increased levels of 289 monomeric muropeptides (Figure 4C, Figure 4-figure supplement 1). Conversely, the 290 percentage of di- and trimeric muropeptides decreased relative to the WT control, as did the 291 percent cross-linking (Figure 4C). These observations suggest that acetate intoxication 292 constrains the D-Ala-D-Ala pool in the WT strain and alters PG cross-linking despite Alr1 293 activity. The extent of PG cross-linking in the dat mutant was similar to WT in the presence or 294 absence of acetate, consistent with our finding that the Dat activity plays a limited role in 295 maintaining the D-Ala-D-Ala pool in the WT strain (Figure 4C). In contrast, PG cross-linking in 296 the alr1 mutant was lower than the WT strain by ~10% (Figure 4C). Acetate intoxication further 297 decreased the cross-linking approximately 20% relative to WT as well as the ratio of dimeric to 298 monomeric muropeptides in the alr1 mutant, which inevitably reduced the growth of this strain 299 (Figure 4C). 300 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 13 Muropeptide analysis also revealed the accumulation of a disaccharide tripeptide (NAG-301 NAM-AEK (M3); m/z. Da, 826.4080) in the peptidoglycan (PG) extracted from the alr1 mutant 302 (Figure 4-figure supplement 1B). This finding suggests that the significantly elevated levels of 303 UDP-NAM-AEK in the alr1 mutant could efficiently outcompete the substrate specificity of 304 phospho-N-acetylmuramyl pentapeptide translocase (MraY) for UDP- NAM-AEKAA, ultimately 305 becoming integrated into the PG structure itself. Interestingly the incorporation of UDP-NAG-306 NAM-AEK into the alr1 mutant's PG only marginally increased following acetate treatment 307 (Figure 4-figure supplement 1B, see inset). The increase of UDP- NAG-NAM-AEK is most likely 308 an underestimate since cells with higher levels of incorporation are more likely to lyse due to a 309 reduction in PG cross-linking. Overall, these observations support a model wherein the 310 immediate consequences of acetate intoxication are defects in PG crosslinking and 311 biosynthesis. 312 Acetate intoxication inhibits Ddl activity 313 While the above observations point to the consequences of acetate intoxication of S. 314 aureus, its molecular target was not initially identified. Since acetate intoxication dramatically 315 reduces D-Ala-D-Ala levels in the alr1 mutant (Figure 1D), we reasoned that acetate might 316 inhibit either Dat or Ddl activity. To distinguish between these two targets, we measured the 317 levels of D-Ala in the alr1 mutant following acetate intoxication. Surprisingly, we observed that 318 the D-Ala pools in the alr1 mutant did not significantly change in response to acetate 319 intoxication compared to the untreated control (Figure 5A). This suggested that Dat activity was 320 preserved in the alr1 mutant to the same extent as its untreated control and was not affected by 321 acetate. 322 Conversely, these findings also indicate that the acetate-dependent decrease of the D-Ala-323 D-Ala pool in the alr1 mutant was most likely due to the inhibition of Ddl. To test this hypothesis, 324 we cloned S. aureus ddl under the control of a cadmium inducible promoter and induced its 325 expression in the alr1 mutant following acetate intoxication (Figure 5B). Indeed, the growth of 326 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 14 the alr1 mutant was restored to WT levels when ddl was overexpressed, strongly suggesting 327 that Ddl was the target of acetate anion (Figure 5B). 328 To confirm that acetate inhibits Ddl through direct interactions, we undertook two separate 329 approaches. As the first approach, 6xHis-tagged S. aureus Ddl was purified, and in-vitro 330 enzyme kinetic assays were performed to determine the possible inhibitory mechanism of Ddl 331 by acetate. Considering the high concentration of acetate estimated to accumulate in the 332 cytoplasm, a concentration of 300 mM sodium acetate was used in the initial reactions to test 333 inhibition (Figure 5C). Interestingly, variation of acetate concentration showed that Ddl was 334 inhibited in vitro , and these conditions suggest an IC 50 of 400.3 ± 8 mM (Figure 5D). This 335 indicates significant inhibition of Ddl by acetate when the cellular concentration is near the 336 hypothesized 600 mM, further confirming that Ddl is a direct target of inhibition by acetate 337 anion. Furthermore, based on kinetic experiments performed under varying concentrations of 338 either ATP or D-Ala, the kcat values are shown to be distinctly different for each acetate 339 concentration, which strongly suggests a mixed inhibition mechanism for acetate (Figure 5E 340 and F, Table S1). 341 Differential Scanning Fluorometry (DSF) was used as another approach to assess the 342 direct binding of acetate to Ddl (Table S2). The Ddl protein without any ligand bound shows a 343 melting temperature (Tm) of 45 °C. After adding 300 mM sodium acetate, Ddl exhibited a 3.7 344 °C Tm shift indicating a slight thermal stabilization upon binding acetate. This is higher than the 345 shift in the Tm exhibited by a Ddl/ATP complex. The addition of ADP to Ddl results in a 346 decrease of 3.2 °C, indicating a decrease in thermal stability compared to Ddl alone. 347 Intriguingly, when adding acetate to Ddl complexes with ATP or ADP, the Tm increased to 48.9 348 °C and 49.9 °C, respectively (Table S2). This represents a Tm increase of 2.3 °C when acetate 349 is added to a Ddl/ATP complex but a Tm increase of 8.1 °C when acetate is added to a 350 Ddl/ADP complex. The addition of D-Ala to the reaction mixture increases the Tm of Ddl by 4.2 351 °C and adding acetate to the Ddl/ D-Ala mixture shows only a 0.3 °C Tm shift (Table S2). The 352 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 15 widely varying changes in Tm for the tested complexes, particularly when comparing the Tm 353 values for ligand-free, ADP-bound, and the ADP/Acetate complex, further support a mixed 354 inhibition mechanism as these data suggest acetate may bind to multiple sites on Ddl or the 355 location of these binding sites may change depending on the ligand-bound state of the enzyme 356 due to Ddl conformational changes as observed in Ddl orthologs (22). 357 Ddl/Acetate complex structure shows binding of acetate at both substrate binding sites 358 To gain further insight into the mechanism of acetate inhibition, the X-ray crystal structure of 359 a Ddl/acetate complex was obtained using co-crystals of Ddl and acetate. The crystal diffracted 360 to 1.9 Å and data were consistent with a P 2 2 1 21 space group possessing one molecule of Ddl 361 in the asymmetric unit (Table S3). The crystal structure of the Ddl/acetate complex (PDB:8FFF) 362 shows difference density corresponding to acetate at two different sites of the protein. One 363 acetate is positioned within the adenine binding subsite of the ATP binding site and the other 364 acetate ion is positioned in the second D-Ala binding site (Figure 5G). The acetate ion in the 365 ATP binding site interacts with the side chain of Lys177 and the backbone nitrogen of Val216. 366 Also, the methyl group of acetate forms van der Waals interactions with the side chain of 367 Leu145 (Figure 5H). The acetate ion that binds to the D-Ala binding site forms a bidentate polar 368 interaction with the side chain of Arg291 and a hydrogen-bonded interaction with the backbone 369 nitrogen of Gly312 (Figure 5I). These two residues are conserved in Ddl homologs and 370 previous structural data clearly illustrate the crucial role these residues play in D-Ala binding 371 (22). 372 The acetate-bound structure shows conformational differences compared to the previously 373 published ligand-free and ADP-bound structures (22). The 𝛚 loop, which is associated with 374 substrate binding, is disordered in both the S. aureus Ddl ligand-free (PDB:2I87) and the Ddl-375 ADP complex structures (PDB:2I8C) as well as other available Ddl crystal structures that lack 376 bound substrates or ligands (PDB:3K3P, 5DMX and 6U1C) (23-25) . Interestingly, this loop is 377 well ordered in the acetate-bound structure described here (Figure 5J), which gives the first 378 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 16 view of the S. aureus Ddl 𝛚 loop and the interactions it may form with substrates or inhibitors. 379 The structural stabilization of the 𝛚 loop is consistent with the DSF results exhibiting an 380 increase in the melting temperature upon binding acetate. The 𝛚 loop is shifted towards the 381 ATP binding site and repositions the conserved Tyr246 side chain within the ATP binding site, 382 which likely hinders the binding of ATP (Figure 5J). This positioning is comparable with the 383 Mycobacterium tuberculosis Ddl (PDB:3LWB) ligand-free structure, which also takes a closed 384 conformation showing the 𝛚 loop positioned within the ATP binding site and obstructing ATP 385 binding (26). Taken together, the kinetic, DSF, and structural data suggest that while acetate 386 can directly bind within both substrate binding pockets of Ddl, it also stimulates conformational 387 changes in the dynamic 𝛚 loop to afford more allosteric-like effects on enzyme activity. Each of 388 these observations support a mixed inhibition modality. 389 Multiple organic acids inhibit the alr1 mutant in a D-Ala-dependent manner 390 Finally, we determined whether the growth inhibition of the alr1 mutant is unique to acetate 391 anion or is a more general phenomenon mirrored by addition of other small organic acids. 392 Accordingly, we initially performed molecular docking studies of three biologically relevant 393 organic anions: lactate, propionate and itaconate, in both the ATP and D-Ala binding pockets of 394 Ddl (Figure 6A-D). The acetate anion-bound structure of Ddl was used as a reference for 395 analysis. The docking results suggest reasonable poses for lactate, propionate and itaconate 396 within the ATP binding site forming polar interactions with Ddl residues conserved for binding 397 ATP. Upon docking, the carboxylate moieties of both lactate and propionate form ionic 398 interactions with the Lys177 side chain similar to those observed in the Ddl/acetate crystal 399 structure (Figure 6A and B). Also, the side chain of Glu213 in Ddl forms a hydrogen bonded 400 interaction with the hydroxyl of lactate (Figure 6A) and van der Waals interactions between 401 propionate and nearby side chains of Phe175 and Phe295 (Figure 6B) were indicated. The two 402 carboxylate groups of itaconate form hydrogen bonded interactions with backbone amide 403 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 17 nitrogen atoms of Ala218 and Tyr246 as well as a van der Waals interaction with the nearby 404 side chain of Phe175 (Figure 6C). 405 The molecular docking results for lactate, propionate, and itaconate in the D-Ala binding site 406 of Ddl also show similar types of interactions but with variable poses and slight orientation 407 differences compared to that observed for acetate in the crystal structure (Figure 6D). The D-408 Ala binding site, consisting of primar ily charged and polar atoms, allows for a range of binding 409 modes for these small anions, where the ligand size is a stronger factor in determining the 410 binding location. Acetate and propionate, being smaller and less sterically hindered, bind 411 preferentially near the Arg 291 side chain that coordinates the acid moiety of D-ala during the 412 enzymatic reaction (Figure 6D). Meanwhile, itaconate and L-lactate bind in the more spacious 413 region between Lys251 and Ser317 (Figure 6D). The Glide scores from the docking results, 414 which provide a rough estimate of the G of binding for each ligand suggest modest affinity to 415 the identified binding sites (Table S4). 416 To determine if these organic acids could impact Ddl function, the WT and the alr1 mutant 417 were challenged with lactic, propionic and itaconic acids (Figure 6E-G). All three organic acids 418 inhibited the growth of the alr1 mutant. The addition of D-Ala to the culture media rescued the 419 growth of the alr1 mutant to WT levels (Figure 6E-G) consistent with Ddl being the target of 420 lactate, propionate and itaconate. Moreover, overexpression of ddl in the alr1 mutant also 421 restored growth of the alr1 mutant following the organic acid challenge (Figure 6-figure 422 supplement 1A-C). These findings collectively suggest that various organic acid anions can 423 inhibit Ddl activity in S. aureus. 424

425 Intracellular anion accumulation has long been hypothesized to drive weak organic acid 426 toxicity in bacteria (16, 27, 28). However, few studies have investigated the mechanism by 427 which weak acid anions inhibit bacterial growth. Acetic acid is particularly interesting among 428 weak acids, given that it is a common byproduct of glucose catabolism in bacteria and is 429 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 18 excreted in high concentrations (29). S. aureus does not catabolize acetate as a carbon source 430 unless glucose is first exhausted from the environment (30). This results in the intracellular 431 accumulation of acetate in S. aureus as a function of the bacterial transmembrane pH gradient, 432 especially when acetic acid concentrations are high in the immediate vicinity of cells. Here we 433 determine that at high intracellular concentrations, acetate anions directly bind Ddl and inhibit 434 D-Ala-D-Ala production to adversely impact peptidoglycan cross-linking (Figure 7). However, S. 435 aureus exhibits a remarkable tolerance to acetate intoxication due to the robust production of D-436 Ala by Alr1, which ultimately increases D-Ala-D-Ala pools (Figure 7). 437 Multiple lines of evidence demonstrate Ddl to be the target of acetate anions. First, LC-438 MS/MS analysis revealed that acetate intoxication decreased D-Ala-D-Ala pools but not D-Ala in 439 S. aureus, pointing to Ddl as the target of acetate. Second, DSF and in-vitro enzyme kinetic 440 studies showed that acetate could bind and inhibit purified rDdl through a mixed inhibition 441 mechanism. Third, structural analysis of the Ddl-inhibitor complex confirmed that acetate binds 442 to both the ATP-binding and D-Ala binding sites within Ddl and further induced conformational 443 changes to the dynamic 𝛚 loop, which weakens the binding of ATP to the Ddl active site. 444 Finally, overexpression of ddl alone was sufficient to overcome acetate-mediated inhibition of 445 the alr1 mutant and restore growth to WT levels. 446 Inhibitors that bind an enzyme's catalytic substrate binding sites are usually competed out 447 by high concentrations of substrates. However, acetate inhibits Ddl through a mixed inhibition 448 mechanism despite binding to the substrate binding pockets of Ddl. We suspect this is due to 449 additional conformational changes observed in the dynamic 𝛚 loop that affords more allosteric-450 like effects on enzyme activity. However, we cannot rule out that acetate might bind to 451 additional sites in the Ddl-ATP complex, Ddl-ADP complex, or a Ddl-ADP-phospho-D-Ala 452 complex with varying affinities. The differences in the temperature shifts observed in DSF with 453 various substrate complexes support th is possibility. The crystal structures of Ddl/acetate 454 complexes with different substrates could provide a more precise conclusion about the 455 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 19 inhibitory modality of Ddl by acetate . In line with acetate's inhibitory effect on Ddl, we observed 456 that acetate intoxication in the alr1 mutant led to a disproportionate increase in the cytosolic 457 pool of PG tripeptide intermediate (UDP- NAM-AEK) compared to the pentapeptide form (UDP-458 NAM-AEKAA). Previous reports have suggested that MraY might facilitate the integration of 459 UDP-NAM-tripeptide into S. aureus PG, especially when its concentration within cells exceeds 460 that of UDP-NAM -pentapeptide (31, 32). Our findings strongly support this hypothesis, as the 461 analysis of the alr1 mutant's cell wall muropeptides revealed a clear elevation in the level of the 462 disaccharide-tripeptide NAG-NAM-AEK. The inhibition of Ddl by acetate would further reduce 463 the presence of terminal D-Ala-D-Ala moieties within alr1 muropeptides which likely leaves 464 these cells incapable of withstanding the outward-directed cell turgor pressure, ultimately 465 leading to cell death (32). 466 Despite acetate inhibiting Ddl through a mixed inhibition mechanism, it should be noted that 467 a functional Alr1 or even the supplementation of D-Ala in culture media can provide significant 468 tolerance against acetate intoxication in S. aureus. These observations suggest that Ddl is only 469 weakly inhibited by acetate, which is also evident from the relatively high IC 50 of approximately 470 400 mM observed in our kinetic experiments with S. aureus Ddl. The weak inhibition of Ddl 471 would suggest that inflating the cytosolic D-Ala pools could promote sufficient generation of D-472 Ala-D-Ala to counter acetate intoxication. Indeed, it has been estimated that S. aureus 473 maintains a high concentration of roughly 30 mM intracellular D-Ala (33), which we now 474 demonstrate to be critical in countering acetate intoxication. 475 The existence of pepV and dat within the same operon suggests that these genes may 476 have evolved related functions. In Lactococcus lactis the PepV dipeptidase activity was shown 477 to be important for supplying cells with L-Ala which was eventually incorporated into PG (34). In 478 this context, pepV and dat may have a similar role in modulating the intracellular alanine pool. 479 A surprising finding of our study was that dat expression is relatively stable and tightly 480 controlled in S. aur eus due to its SD motif being located within the coding region of pepV. 481 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 20 Furthermore, such a genetic arrangement has been linked to translational coupling (35), 482 wherein active translation from the first gene promotes the translation of the following gene in 483 the operon, which in the case of dat was not sufficient to overcome acetate toxicity in the alr1 484 mutant. Two central mechanisms of translational coupling have been proposed. The first 485 involves secondary and tertiary mRNA structures that either occlude or encompass the SD 486 motif of downstream genes and shield it from ribosomes, thus preventing its translation (36). 487 These mRNA structures can be relieved when a ribosome initiates translation from the first 488 gene of the operon and exposes the downstream intragenic SD sequences to new 30S 489 ribosomal subunits (34). In the second mechanism, continued translation of the first gene of the 490 operon is necessary to increase the abundance of ribosomes in the TIR of the second gene 491 resulting in its enhanced translation (35). Irrespective of the mechanism of translational 492 coupling, our results suggest that genetic arrangements that promote translational coupling 493 might also limit the overall production of dat and thus prevent it from functionally 494 complementing the alr1 mutant following acetate intoxication. Since our data suggest that Dat 495 primarily promotes flux from D-Ala to D-Glu when Alr1 is active, the tight control of dat through 496 translational coupling could prevent the depletion of the intracellular reserves of D-Ala 497 necessary to overcome Ddl inhibition during acetate intoxicat ion. Thus. the elevated D-Ala pool 498 maintained within the cell could represent a strategic adaptation by S. aureus to combat Ddl 499 inhibition caused by organic acids typically present in the niches colonized by this bacterium. 500 In conclusion, our findings demonstrate that Ddl is the primary target of acetate anion 501 intoxication in S. aureus . However, other biologically relevant organic anions like lactate, 502 propionate and itaconate could also inhibit the alr1 mutant similar to acetate. Furthermore, the 503 growth inhibition of the alr1 mutant by these organic acids could be rescued following D-Ala 504 supplementation, which suggests that Ddl is a bona fide and conserved target of various 505 organic acid anions. Indeed, it is tempting to speculate that the robust Alr1 activity leading to 506 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 21 the accumulation of millimolar levels of D-Ala may have evolved in part to offset the inhibition of 507 Ddl from the toxic effects of organic anions. 508 Acknowledgments 509 This work was funded by NIH/NIAID R01AI125588 and 2P01A1083211 Metabolomics Core to 510 VCT, 2P 01AI083211 Project 4 to TK, respectively. This work was also supported in part by 511 NIH/NIAID R21AI151924 to DRR. X -ray diffraction data were collected at the Life Sciences 512 Collaborative Access Team beamline 21 -ID-F at the Advanced Photon Source, Argonne 513 National Laboratory, which is a U.S. Department of Energy (DOE) Office of Science User 514 Facility operated for the DOE Office of Science by Argonne National Laboratory under Contract 515 No. DE -AC02-06CH11357. Use of the LS -CAT Sector 21 was supported by the Michig an 516 Economic Development Corporation and the Michigan Technology Tri -Corridor (Grant 517 085P1000817). The University of Nebraska Medical Center Mass Spectrometry and 518 Proteomics Core Facility is administrated through the Office of the Vice Chancellor for 519 Research and supported by state funds from the Nebraska Research Initiative (NRI). Research 520 in the Cava lab is supported by the Swedish Research Council, the Laboratory for Molecular 521 Infection Medicine Sweden (MIMS), Umeå University, the Knut and Alice Wallenber g 522 Foundation (KAW) and the Kempe Foundation. The funders had no role in the study design, 523 data collection, interpretation, and decision to submit this work for publication. The authors 524 have no conflict of interest to declare. 525 Data Availability 526 The atomic coordinates and structure factors have been deposited in the Protein Data Bank, 527 accessible at www.pdb.org, with the PDB ID code 8FFF. 528

529 Bacterial strains and growth conditions 530 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 22 The S. aureus WT and mutant strains described in this study were cultured in TSB 531 containing 14 mM glucose. S. aureus JE2 mutants were mainly obtained from the Nebraska 532 Transposon Mutant Library (37). These mutants were re-transduced into the WT strain using 533 Ф11- bacteriophage to eliminate any off-target effects . To generate double or triple mutants, 534 the Erm R antibiotic cassette in the transposon mutants w as exchanged with Kan R or Tet R 535 cassettes by allelic exchange before introducing an additional mutation. The allelic exchange 536 was performed as described previously (38) . In -frame gene deletion mutants were created 537 using a temperature-sensitive vector, pJB38, as described previously (38) . S. aureus mutants 538 were complemented by inserting the WT allele of mutated genes under the control of their 539 native promoter into the SaPI1 chromosomal site using the pJC1111 suicide vector (39) . For 540 experiments involving the over-expression of ddl in S. aureus , ddl was cloned into a CdCl 2 541 inducible multicopy vector, pJB68 (38). The concentration of CdCl 2 was titrated to achieve full 542 growth complementation. All bacterial isolates, plasmids, and primers used in this study are 543 listed in Table S5, S6, and S7, respectively. 544 Nebraska Transposon Mutant Library (NTML) screen 545 The NTML mutants were grown in 96-well plates in the presence and absence of 20 mM 546 acetic acid (pH~6.1) in TSB for 24 hours at 37 ºC. The growth of bacteria was determined by 547 measuring the optical density at 600 nm (OD 600) after 24 hours using a TECAN Infinite 200 548 spectrophotometer. To account for well-to -well variances that accompany 96-well cultures, the 549 WT strain was independently grown in all the wells of a 96-well plate, both in the presence and 550 absence of acetic acid. Area under the curve (AUC) values for each mutant under a particular 551 condition were obtained by normalizing the values to WT AUC. The graph was generated by 552 plotting the normalized AUC of a mutant under acetate stress versus the control (g rowth without 553 acetate). 554 Competitive fitness assay 555 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 23 The cultures of WT ( S. aureus JE2) and isogenic mutant strain with either pAQ59 (empty 556 vector) or pAS8 (containing dat gene under control of its native promoter) inserted at the SaPI1 557 chromosomal site were used to assess competitive fitness. Following the growth of these 558 cultures for 24 h, 10 7 colony forming units (cfu) per milliliter of each strain were used to 559 measure the competitive fitness in presence of 20 mM acetate. The bacterial cfu were 560 enumerated on TSA plates with or without 0.1 mM cadmium chloride immediately after initiation 561 of competition and at 4 h between tested strains allowing the bacteria to undergo approximately 562 seven replications to reach 10 9 cfu/ ml . The competitive fitness was calculated using the 563 Malthusian parameter for competitors using the following formula: w = ln (M f/Mi)/ln (W f/Wi), 564 where f and i represent cfu counts at final (4 h) and initial (time 0) of competition assay, 565 respectively (11). M and W refer to mutant and WT, respectively. 566 Sample collection for mass-spectrometry analysis 567 Overnight cultures of WT, alr1 and dat mutants were inoculated to an OD 600 of 0.06 units 568 into 250 ml flasks containing 25ml of TSB 14 mM glucose. Acetic acid (20 mM) was added to 569 the flasks whenever necessary. The flasks were incubated in a shaker incubator at 37 ºC and 570 250 rpm. A total of 10 OD 600 units of cells were collected following 3 hours of incubation by 571 centrifuging the cultures at 10,000 rpm at 4 ºC. The pellet was then washed once with ice-cold 572 saline (0.85% NaCl) and centrifuged again at 10,000 rpm at 4 ºC. The bacterial cells were then 573 resuspended in ice cold quenching solution consisting of 60% ethanol, 2 µM Br-ATP and 2 µM 574 ribitol. The cytosolic metabolites were obtained by bead beating the cells, followed by 575 centrifugation. The supernatant was collected and stored at -80 ºC until further use. For stable 576 isotope experiments, overnight cultures were inoculated into a chemically defined medium 577 (CDM, (40)) containing 13C3 15N1-L-Ala (100 mg/ L) in place of L-Ala and the samples were 578 collected in the exponential phase following 4 hours of incubation at 37 ºC. 579 Chromatography for mass-spectrometry analysis 580 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 24 The chromatographic separation of PG intermediates was performed by liquid 581 chromatography using XBridge Amide (150 × 2.1 mm ID; 1. 7 µm particle size, Waters, USA) 582 analytical column; whereas D-Ala-D-Ala was analysed using XBridge Amide (10 0 × 2.1 mm ID; 583 1.7 µm particle size, Waters, USA). A guard XBridge Amide column (20 × 2.1 mm ID; 1.7µm 584 particle size, Waters, USA) was connected in front of the analytical column. Mobile phase A 585 was composed of 10 mM ammonium acetate, 10 mM ammonium hydroxide containing 5 % 586 acetonitrile in LC-MS grade water; mobile phase B was 100% LC-MS grade acetonitrile. The 587 column was maintained at 35 °C and the autosampler temperature was maintained at 5 °C. The 588 gradient was started with the A/B solvent ratio at 15/85 for over 1 minute, followed by a gradual 589 increase of A. A was reduced to 15% after separation and elution of all the interested 590 compounds and equilibrated for 6.0 minutes before the next run. The needle was washed with 1 591 mL of strong wash solvent containing 100% acetonitrile followed by 1 mL of weak wash solvent 592 comprised of 10% aqueous methanol after each injection. The sample injection volume was 5µl. 593 Chiral separation of D- and L-isomers of alanine and glutamate was achieved on Astec 594 CHIROBIOTIC® T column (150 x 2.1 mm, 5 µm particle size) from Supelco. Mobile phase A 595 was 20 mM ammonium acetate and mobile phase B was 100% ethanol. The mobile phase 596 composition was 40:60 v/v of A:B in isocratic elution mode pumped at 100 L/min flow rate. The 597 injection volume was 5 L and the column was maintained at room temperature. Multiple 598 reaction monitoring (MRM) for D- and L- isomers of alanine are listed in Table S8. All other MS 599 parameters are discussed in the LC-MS/MS analysis section. The L-enantiomer of alanine and 600 glutamate elutes faster than their D-counterparts. The total run time was 15 minutes. 601 Targeted LC-MS/MS analysis 602 Triple-quadrupole-ion trap hybrid mass spectrometer viz., QTRAP 6500+ (Sciex, USA) 603 connected with Waters UPLC was used for targeted analysis. The QTRAP 6500+ was operated 604 in polarity switching mode for targeted quantitation of amino acids through the Multiple Reaction 605 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 25 Monitoring (MRM) process. LC-MS MRM data for each metabolite w as acquired in centroid 606 mode as a default setting. MRM details for each analyte are listed in Table S8. The optimized 607 electrospray ionization (ESI) parameters were as follows: electrospray ion voltage of -4200 V 608 and 5500 V in negative and positive mode, respectively, source temperature of 500 °C, curtain 609 gas of 40, and gas 1 and 2 of 40 and 40 psi, respectively. Compound-specific parameters were 610 optimized for each compound using manual tuning. These parameters include a declustering 611 potential (DP) of 65 V and -60 V in positive and negative mode, respectively, entrance potential 612 (EP) of 10 V and -10 V in positive and negative mode, respectively, and collision cell exi t 613 potential (CXP) maintained at 10 V and -10 V in positive and negative mode respectively. Other 614 compound-specific parameters, such as Q1, Q3, and collision energies, are listed in Table S8. 615 MRM conditions for PG intermediates were adopted from Vemula et al (41). 616 High Resolution Mass Spectrometry 617 HRMS Orbitrap (Exploris 480) operated in polarity switching mode was used for the untargeted 618 analysis of isotopologues of D-Ala-D-Ala and D-Glu in data-dependent MS/MS acquisition mode 619 (DDA). Electrospray ionization (ESI) parameters were optimized are as follows: electrospray 620 ion voltage of -2700V and 3500V in negative and positive mode respectively , Ion transfer tube 621 temperature was maintained at 350°C, m/z scan range was 140-180 Da for non-chiral LC-622

using Amide column whereas, it was 80-160 Da for chiral column method. Sheath gas, 623 auxiliary gas and sweep gas were optimized according to the UHPLC flow rate. Orbitrap 624 resolution for precursor ion as well as for fragment ion scan was maintained at 240000 and 625 60000 respectively. Normalized collision energies at 30, 50 and 150% were used for the 626 fragmentation. Data was acquired in profile mode. Xcaliber software from Thermo was used for 627 instrument control and data acquisition. This software was equipped with Qual-, Quant- and 628 FreeStyle browsers which were used for profiling metabolites and their isotopologues in all 629 samples. Selected precursor ion for each isotopologue is listed in Table S9. Identification and 630 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 26 detection of all metabolites was aided by the Compound Discoverer (CD) software procured 631 from Thermo USA. The KEGG and HMDB databases plugged-in with CD software were used 632 for metabolite identifications and annotations . Mass accuracy for all the ions was maintained at 633 or below 5 ppm. To correct for natural abundance, we utilized FluxFix, an open-source online 634 software (42), and independently verified these calculations using the ChemCalc software (43). 635 Fractional contribution of D-Ala-D-Ala from imported 13C3 15N1-L-Ala 636 An estimate of the fractional contribution (FC) of labeled carbon from 13C3 15N1-L-Ala tracer 637 incorporated in to the intracellular D-Ala-D-Ala pool was calculated using equation 1, as 638 previously described (44). 639 𝐹𝐶 = ∑ 𝑖.𝑚 𝑖 𝑛 𝑖=0 𝑛.∑ 𝑚 𝑖 𝑛 𝑖=0 eq. 1 640 where, n is the number of carbon atoms in D-Ala-D-Ala, i represents the various carbon 641 isotopologues of D-Ala-D-Ala and m the abundance of the D-Ala-D-Ala isotopologues. 642 Transcription site identification of the dat operon 643 The adaptor- and radiation-free transcription start site (ARF-TSS) identification method was 644 employed to identify the 5 ՚-UTR region of the dat operon (45). In brief, 1 ug of RNA isolated 645 from JE2 WT was subjected to reverse transcription by using 5 ՚-phosphorylated primer 646 pepV_TSS_R1 and the first strand cDNA synthesis kit (Invitrogen, Superscript III First-Strand 647 Synthesis System). RNA was degraded by using 1M NaOH at 65 ºC for 30 min and then 648 neutralized with 1M HCl. The resultant cDNA was ligated by using T4 RNA Ligase I (Thermo 649 Scientific) to generate a circular cDNA. Two inverse primers: pepV_TSS_R2 and 650 pepV_TSS_F3 were used to amplify the circular cDNA. The amplified product was cloned into a 651 TOPO Cloning vector and then sequenced using M13F(-20) and M13R primers. All the primers 652 used in this procedure are mentioned in Table S7. 653 Quantitative real-time PCR 654 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 27 Quantitative real-time PCR was performed to estimate the transcript levels of dat, ddl and 655 murI in the presence and absence of acetate. The samples were collected during the 656 exponential growth phase and RNA was isolated using a Qiagen RNA isolation kit following the 657 manufacturer's protocol. A total of 500 ng of RNA was used to synthesize cDNA using the 658 QuantiTech reverse transcription kit (Qiagen). The cDNA samples were then diluted 1:10 and 659 used as a template to perform RT-qPCR. The RT-qPCR was carried out using SYBR green 660 master mix (Roche Applied Science) in a QuantiFast light cycler (Applied Biosystems). The 661 relative transcript levels were estimated by using the comparative threshold cycle method 662 (ΔΔCT) and sigA was used as the internal control for normalization. Primers used to perform 663 RT-qPCR are listed in Table S7. 664 Muropeptide analysis 665 The WT and isogenic mutants were inoculated to an OD 600 of 0.06 into 1-liter flasks 666 containing 100 mL of TSB 14 mM glucose. Acetic acid (20 mM) was added to the media when 667 appropriate. A total of 95 OD 600 units of cells were collected following 6 hours of growth at 37 668 ºC, 250 rpm. The pelleted cells were then resuspended in 50 % SDS and boiled for 3 hours. 669 Once boiled, cell wall material was pelleted by ultracentrifugation and washed with water. 670 Clean sacculi was digested with muramidase (100 µg/ml) and soluble muropeptides reduced 671 using 0.5 M sodium borate pH 9.5 and 10 mg/mL sodium borohydride. The pH of the samples 672 was then adjusted to 3.5 with phosphoric acid. UPLC analyses was performed on a Waters-673 UPLC system equipped with an ACQUITY UPLC BEH C18 Column, 130 Å, 1.7 µm, 2.1 mm × 674 150 mm (Waters Corporation, USA) and identified at Abs. 204 nm. Muropeptides were 675 separated using a linear gradient from buffer A (0.1 % formic acid in water) to buffer B (0.1 % 676 formic acid in acetonitrile). Identification of individual peaks was assigned by comparison of the 677 retention times and profiles to validated chromatograms (46-48) . The identity of peak belonging 678 to disaccharide tripeptide , NAG-NAM-AEK (M3) was assigned by mass spectrometry using 679 UPLC system coupled to a Xevo G2/XS Q-TOF mass spectrometer (Waters Corp.). Data 680 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 28 acquisition and processing were performed using UNIFI software package (Waters Corp.). The 681 relative amount of each muropeptide was calculated relative to the total area of the 682 chromatogram. Representative chromatograms for each sample type are depicted in (Figure 4-683 figure supplement 1). The abundance of PG (total PG) was assessed by normalizing the total 684 area of the chromatogram to the OD 600. The degree of cross-linking refers to the number of 685 peptide bridges and was calculated as % of dimers + % of trimers x 2 + % of tetramers x 3 (49). 686 Protein purification 687 The coding region of ddl was cloned into pET28a vector to generate a C-terminal 6 His tag 688 fusion protein before being transferred into E. coli BL21(DE3). The cells were grown in Luria 689 Broth Media (Research Product Internationals) containing 50 µg/mL Kanamycin (Gold 690 Biotechnology) at 37 °C. When OD 600 reached 0.6, 1 mM IPTG (Gold Biotechnology) was 691 added to induce the protein expression. The cells were harvested by centrifugation (3724 g) 692 after inducing them at 16 °C for 20 h. The harvested cells were resuspended in lysis buffer 693 comprising 25 mM Tris pH 7.5, 150 mM NaCl, and 5 mM 2-Mercaptoethanol. The cells were 694 lysed by adding Lysozyme ( MP-Biomedicals) and DNase I (Roche Applied Sciences) and 695 incubating them on ice for 30 minutes. Then cells were subjected to sonication (Sonicator 696 3000, Misonix) to further lyse the cells. The crude cell lysate was refined by centrifuging at 697 18514 g for 40 min (Fixed angle rotor, 5810-R Centrifuge, Eppendorf). The clarified lysate was 698 applied to a 5 mL HisTrap™TALON™ crude cobalt column (Cytiva) aft er equilibrating the 699 column with lysis buffer. The column was washed using the same buffer and the protein was 700 eluted isocratically using 150 mM imidazole-containing buffer. The purified protein was dialyzed 701 in 20 mM Tris pH 8.0 buffer and 0.5 mM Tris (2-carboxyethyl) phosphine to use in 702 crystallization experiments and biochemical assays. 703 Crystallization of Ddl and data collection 704 The crystals of Ddl in complex with acetate were obtained by co-crystallization experiments 705 using the hanging drop vapor diffusion method. The 10 mg/mL of protein was incubated with 30 706 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 29 mM potassium acetate, 5 mM magnesium chloride hexahydrate, and 1 mM ADP for 20 min 707 before the crystallization experiments. The co-crystals were achieved in crystallization drop 708 against a well solution consisting of 0.2 M sodium thiocyanate and 20 % polyethylene glycol 709 monomethyl ether 2000. The crystals were flash cooled in liquid nitrogen immediately after 710 adding 40% polyethylene glycol 3350 to the crystallization drop for cryoprotection. The data 711 were collected at the Advance Photon Source Argonne National Laboratory (APS-ANL, IL), LS-712 CAT ID-F beamline. 713 Ddl enzyme kinetic assays 714 The Invitrogen™ EnzChek™ Phosphate Assay Kit was used to detect the release of 715 inorganic phosphate by continuously monitoring the absorbance at 360 nm. The reaction 716 components were added as specified by the kit with 200 nm Ddl (containing 1mM MgCl 2), 100 717 mM Potassium chloride, and ATP. The reaction mixture was incubated for 10 min and D-Ala 718 substrate was added to initiate the reaction. The inhibition of Ddl by acetate was determined 719 using various concentrations of sodium acetate, D-Ala, and ATP to determine kinetic 720 parameters. 721 Data processing and refinement 722 The data was processed by CCP4 software (50) and S. aureus D-alanyl D-alanine ligase 723 apoprotein (PDB:2I87) was used for the molecular replacement followed by a rigid body 724 refinement using PHENIX (51). Manual model refinement was performed using Coot (52). The 725 XYZ coordinate, B-factor, occupancy, and real space refinements were executed using 726 PHENIX between manual model refinements. The acetate was modeled using eLBOW and 727 positioned at the corresponding difference density. The structure was refined using PHENIX 728 and validated using Molprobity (53). 729 Molecular Docking Experiments 730 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 30 The docking experiments of small organic acids were performed with the acetate-bound Ddl 731 structure (PDB:8FFF) with acetate removed. The protein structure was first prepared with the 732 protein preparation wizard. The lactate, propionate and itaconate ligands were prepared by 733 LigPrep. The docking experiments were performed using Schrödinger Glide (New York, NY). 734 Differential Scanning Fluorometry 735 The reaction mixture was prepared using 22 µM Ddl, 5 mM magnesium chloride, 100 mM 736 potassium chloride, 1 mM ADP, 300 mM potassium acetate, and 20 mM Tris pH 7.5 buffer as 737 required. The SyPro orange dye was added to a final concentration of 1 X Protein Thermal 738 Shift™ Dye (Thermofisher) in the reaction mixture. The reactions were performed in triplicate. 739 The samples were centrifuged in MicroAmp™ Optical 96 -Well Reaction Plate (Applied 740 Biosystems) at 2325 g for 10 minutes. The protein denaturation was monitored by obtaining the 741 fluorescence signal by increasing the temperature from 22 °C - 95 °C at 0.5 °C/minute rate 742 using QuantStudio 3 real-time PCR (ThermoFisher). The melting temperature (Tm) was 743 determined by calculating the derivative of the fluorescent signal and identifying the centroid of 744 the observed melting peak. 745 746 747 748 749 750 751 752 753 754 755 756 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 31 757 758 759 760 761 762 763 764 765 766 767 768 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 32 Figures 769 770 771 772 Figure 1. Alanine racemase activity counters acetate intoxication (A) The Nebraska 773 Transposon Mutant library was screen against 20 mM acetic acid, pH 6.0 to identify mutants 774 with altered growth phenotypes. The WT strain and transposon mutants were grown for 24 h in 775 TSB ± 20 mM acetic acid. The bacterial growth at 24 h was measured spectrophotometrically 776 (OD600) and normalized to WT growth. The X and Y-axis on the plot represent normalized 777 growth values for each mutant in the presence or absence of acetate. (B) The growth of the WT, 778 alr1 mutant, and alr1 complemented strain in TSB supplemented with 20 mM acetic acid. (C) 779 Aerobic growth of WT, alr1, citZ, citZalr1 mutants in TSB media lacking glucose but 780 supplemented with 20 mM acetic acid. (D) LC-MS/MS analysis was performed to quantify the 781 intracellular D-Ala-D-Ala pool in strains cultured for 3 h (exponential phase) in TSB ± 20 mM 782 acetic acid. (E) The growth of strains was monitored following D-Ala supplementation (5 mM) in 783 TSB + 20 mM acetate, (n=3, mean ± SD). Ac, acetate. 784 785 786 787 788 789 790 791 792 793 794 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 33 795 796 Figure 1-figure supplement 1. Alr1 and Dat are the primary routes of D-Ala production in 797 S. aureus. (A) Schematic of the predicted D-Ala-D-Ala generating pathways in S. aureus . 798 Growth curves of various S. aureus strains grown in the (B) absence or (C) presence of 5 mM 799 D-Ala (n=3, mean ± SD). 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 34 822 823 Figure 2. Translational coupling of dat to pepV limits the alr1 mutant from countering 824 acetate intoxication (A) Schematic representation of various engineered mutations in the 825 pepV-dat locus.SD, Shine-Dalgarno motif; TSS, transcriptional start site. (B)-(D) Growth of 826 engineered mutants was monitored spectrophotometrically (OD 600) in TSB supplemented with 827 20 mM acetate (n=3, mean ± SD). (E) RT-qPCR to determine dat transcription in various 828 mutants relative to the WT strain. 829 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 35 830 831 Figure 2-figure supplement 1 . Overexpression of dat rescues the growth defect of the 832 alr1 mutant (A) dat was cloned in a multicopy vector (pSP4) controlled by its native promoter. 833 The S. aureus strains containing pSP4 and pLI50 (empty vector) were grown in TSB 834 supplemented with 20 mM acetate (B) RT-qPCR analysis of dat expression in the alr1 mutant in 835 the presence or absence of 20 mM acetic acid (n=3, mean ± SD). 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 36 865 866 Figure 3. Reaction orientation and fluxes through Alr1 and Dat (A) Schematic 867 representation of various isotopologues of D-Ala-D-Ala and D-Glu generated from 13C3 15N1 868 labeled L-Ala. Metabolites in blue mainly arise from Alr1, red, through the Ald1/2-Dat pathway 869 and yellow are unlabeled intermediates within cells. The mass isotopologue distribution of (B) D-870 Ala-D-Ala and (C) D-Glu were determined by LC-MS/MS following the growth of S. aureus in 871 chemically defined media supplemented with 13C3 15N1 L-Ala (n=3, mean ± SD). Isotopologues of 872 D-Ala-D-Ala shown in grey color are minor species and are noted in Table S9. (D) The mean 873 competitive fitness ( w) was determined by co-culturing the WT strain with an isogenic mutant 874 that contained either the empty pAQ59 vector integrated into the SaPI chromosomal site (WT EV) 875 or the pAS8 vector containing dat under the control of its native promoter (WT dat) (n=18, the 876 dotted lines indicate the median and quartiles). 877 878 879 880 881 882 883 884 885 886 887 888 889 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 37 890 Figure 4. Acetate intoxication impacts soluble PG precursor pools and cell wall cross-891 linking. (A) The intracellular pool of PG intermediates in exponential phase cultures of S. 892 aureus was estimated using LC-MS/MS analysis. cps, counts per second (B) ddl and murF 893 transcription in the exponential growth phase was determined by RT-qPCR analysis (n=3, mean 894 ± SD). (C) Cell wall muropeptide analysis of the WT, alr1 and dat mutants was determined 895 following growth in TSB ± 20 mM acetate for 3 h. Cell wall cross-linking was estimated as 896 previously described (49). Ac, acetate. 897 898 899 900 901 902 903 904 905 906 907 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 38 908 909 Figure 4-figure supplement 1 . Muropeptide analysis. Representative c hromatograms of 910 muropeptide extracts from (A-C) WT, alr1 and dat mutants, and following (D-F) acetate 911 intoxication. A unique peak corresponding to NAG-NAM-AEK (M3, m/z, Da: 826.4080) was 912 identified in the alr1 mutant. The peak area of M3 was normalized to the total area of peaks 913 observed in the chromatogram and expressed as percent (see inset figure in B, n=3, mean ± 914 SD). M, monomer; D, dimer; T, trimer, Tt, tetramer. 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 39 940 Figure 5. Acetate anion inhibits Ddl activity. (A) The intracellular D-Ala was determined by 941 LC-MS/MS analysis. (B) The ddl gene was overexpressed in S. aureus using a cadmium 942 inducible expression system (pSP36). CdCl 2, 0.312 µM. (C) Inhibition of recombinant His-943 tagged Ddl activity in the presence of 300 mM sodium acetate (D) IC50 curve of the inhibition of 944 rDdl by acetate. Michaelis-Menten kinetics of rDdl in varying concentrations of (E) D-Ala, and (F) 945 ATP in the presence of acetate to assess the inhibition mechanism . (G) Structure of the acetate 946 bound Ddl (PDB:8FFF) . (H) Acetate bound to the ATP binding site of Ddl (I) Acetate bound to 947 the second D-Ala binding site of Ddl. The calculated Fo-Fc omit maps are contoured to 3σ and 948 the mesh is shown in blue. (J) Superimposed structure of acetate bound Ddl (slate blue) with 949 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 40 StaDdl apo structure (PDB:2I87, beige) and StaDdl-ADP complex structure (PDB:2I8C, grey) 950 showing a shift of 𝛚 loop (red) to ATP binding site. The D-Ala-D-Ala was modeled at the D-Ala 951 binding site using Thermos thermophius HB8 Ddl structure (PDB:2ZDQ). The bound ADP (grey) 952 of PDB:2I87 and modeled D-Ala-D-Ala (light blue) indicates the positioning of Ac at ATP and 953 second D-Ala binding sites respectively. Ac, acetate; V, velocity. 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 41 1001 1002 Figure 6. Biologically relevant weak acids inhibit growth of the alr1 mutant . Molecular 1003 docking of (A) lactate (B) propionate and (C) itaconate to the ATP binding site of Ddl. (D) The 1004 relative positions and poise of different organic anions in relation to acetate in the D-Ala binding 1005 site of Ddl was determined using Schrödinger Glide. The growth (OD 600) of the WT and alr1 1006 mutant in TSB containing (E) lactic acid (40 mM) (F) propionic acid (20 mM) and (G) itaconic 1007 acid (20 mM) in the presence or absence of 5 mM D-Ala. 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 42 1032 1033 Figure 6-figure supplement 1. Overexpression of ddl rescues the growth defect of the 1034 alr1 mutant The growth (OD 600) of the WT and alr1 mutants overexpressing Ddl (pSP36; 1035 cadmium inducible expression of ddl) in TSB supplemented with (A) lactic acid (40 mM), (B) 1036 propionic acid (20 mM) and (C) Itaconic acid (20 mM) in the presence or absence of 5 mM D-Ala 1037 (n=3, mean ± SD). 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 43 1072 1073 Figure 7. Model depicting the role of Alr1 in countering organic acid anion-mediated 1074 inhibition of Ddl. During its growth, S. aureus (WT) maintains a substantial intracellular pool of 1075 D-Ala through the activity of Alr1. Any excess D-Ala is subsequently converted into D-Glu by the 1076 action of the Dat enzyme. The high concentration of D-Ala is crucial for the optimal functioning 1077 of Ddl and serves to prevent the inhibition of Ddl by acetate (Ac -) and other organic acid anions. 1078 This process generates sufficient D-Ala-D-Ala, which is rapidly incorporated into the PG 1079 tripeptide precursor UDP-NAM-AEKAA to form UDP-NAM-AEKAA, which ultimately contribut es 1080 to a robust cross-linked PG (murein) sacculus. In the alr1 mutant, the Dat reaction orientation is 1081 switched to preserve intracellular D-Ala. Nevertheless, this change is inadequate to maintain 1082 sufficient D-Ala pool to shield Ddl from inhibition by Ac -, due to tight control of dat translation. 1083 This results in an excess of UDP-NAM-AEK, which competes effectively with UDP-NAM-AEKAA 1084 for PG incorporation. The absence of a terminal D-Ala-D-Ala in the PG hinders crosslinking and 1085 leads to impaired growth following acetate intoxication. 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 44 Table S1: Effect of acetate on Ddl activity 1100 1101 Substrate Condition Km (mM) Vmax (μM min - 1) kcat (min-1) D-Ala 0 mM Acetate 7.0 ± 0.6 16.0 ± 0.4 80.1 ± 2.1 100 mM Acetate 6.8 ± 0.6 10.9 ± 0.3 54.4 ± 1.6 300 mM Acetate 8.4 ± 0.6 6.3 ± 0.1 31.5 ± 0.7 ATP 0 mM Acetate 0.6 ± 0.1 12.3 ± 1.1 61.7 ± 5.7 200 mM Acetate 0.8 ± 0.2 6.9 ± 0.8 34.4 ± 3.8 300 mM Acetate 0.8 ± 0.3 4.0 ± 0.7 19.9 ± 3.6 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 45 Table S2: Impact of acetate on Ddl stability assessed by DSF Sample Tm D (ºC) ΔTm D (ºC)* Ddl 45.0 ± 0.0 - Ddl + Sodium acetate (Ac) 48.7 ± 0.0 3.7 Ddl + ATP 46.4 ± 0.0 1.4 Ddl + ATP + Ac 48.9 ± 0.1 3.9 Ddl + ADP 41.9 ± 0.1 -3.2 Ddl + ADP + Ac 49.7 ± 0.1 4.0 Ddl + D-Ala 49.2 ± 0.1 4.2 Ddl + D-Ala + Ac 49.6 ± 0.0 4.5 Ddl + D-Ala + ATP + Ac 47.1 ± 0.1 2.0 Ddl + D-Ala + ADP + Ac 49.8 ± 0.0 4.8 1136 *The ∆Tm D values are calculated as the difference in melting temperature 1137 of the Ddl apo protein to Ddl with added substrates or acetate inhibitor. 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 46 1167 Table S3: Refinement statistics of Ddl/Acetate structure Resolution Range 48.21 – 1.92 (1.989 – 1.92) Space group P 2 21 21 Unit cell 55.123 65.817 99.423 90 90 90 Total reflections 173526 (12202) Unique reflections 20366 (2806) Multiplicity 8.5 (9.3) Completeness (%) 94.80 (50.46) Mean I/sigma(I) 10.99 (1.72) Wilson B-factor 37.29 R-merge 0.097 (0.33) R-meas 0.104 (0.35) R-pim 0.036 (0.12) CC1/2 0.998 (0.95) CC* 0.999 (0.99) Reflections used in refinement 26870 (1416) Reflections used for R-free 1601 (92) R-work 0.21 (0.42) R-free 0.27 (0.47) CC (work) 0.223 (0.07) CC (free) 0.217 (-0.16) Number of non-hydrogen atoms 2963 macromolecules 2777 ligands 10 solvent 176 Protein residues 355 RMS(bonds) 0.009 RMS(angles) 1.10 Ramachandran favored (%) 93.70 Ramachandran allowed (%) 6.02 Ramachandran outliers (%) 0.29 Rotamer outliers (%) 0.00 Clashscore 6.88 Average B-factor 43.60 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 47 macromolecules 43.60 ligands 45.09 solvent 43.64 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 48 1214 1215 Table S4: Glide scores from molecular docking studies of organic anions Organic Anion ATP binding site D-Ala binding site L-lactate -4.851 -4.431 Propionate -2.317 -1.883 Itaconate -3.572 -3.575 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 49 1256 1257 Table S5: Strains used in this study Strains Description Source E. coli Electro-Ten-Blue General plasmid maintenance strain Stratagene S. aureus RN4220 Restriction-deficient strain is routinely used as a transformation intermediate (54) S. aureus RN4220: pRN7023 Restriction deficient strain carrying pRN7023 plasmid containing integrase gene routinely used as a transformation intermediate (39) E. coli DH5α General plasmid maintenance strain Thermo Fisher E. coli BL21(DE3) Protein over-expression and purification strain Novagen S. aureus JE2 S. aureus USA300 LAC cured of all 3 native plasmids (37) JE2 alr1 bursa aurealis transposon mutant, ErmR NTML JE2 alr1::alr1 WT copy of alr1 complemented at the SaPI1 site of bursa aurealis transposon mutant, ErmR This study JE2 citZ bursa aurealis transposon mutant, ErmR NTML JE2 citZalr1 bursa aurealis transposon mutant, ErmR, KanR This study JE2 dat bursa aurealis transposon mutant, ErmR NTML JE2 Δalr2 Inframe isogenic deletion mutant of JE2 This study JE2 alr1Δalr2 bursa aurealis transposon mutant, Erm R, transduced into inframe isogenic deletion mutant JE2 Δalr2 This study JE2 alr1dat bursa aurealis transposon mutant, ErmR, TetR This study JE2 alr1: dat pLI50 dat plasmid (pSP4) transduced into bursa aurealis transposon mutant, ErmR JE2 pepVΔSD1-467 Isogenic deletion mutant of SD1 and pepV This study JE2 alr1pepVΔSD1-467 bursa aurealis transposon mutant, Erm R transduced into isogenic deletion mutant of SD1 and pepV This study JE2 ΔpepV Inframe isogenic deletion mutant of JE2 This study JE2 alr1ΔpepV bursa aurealis transposon mutant, Erm R transduced into inframe isogenic deletion mutant JE2 ΔpepV This study JE2 pepVQ12STOP Glutamine to STOP codon substitution at 12 th amino acid position in PepV This study JE2 alr1pepVQ12STOP Bursa aurealis transposon mutant, Erm R transduced into glutamine to STOP codon substitution at 12 th amino acid position in PepV This study JE2 WT: pJB68 pJB68 transduced into WT JE2 This study JE2 WT: pSP36 pSP36 transduced into WT JE2 This study JE2 alr1: pJB68 pJB68 transduced into JE2 alr1 This study JE2 alr1: pSP36 pSP36 transduced into WT alr1 This study JE2 WT::pAQ59 pAQ59 empty vector inserted at the SaPI1 site of WT JE2 This study JE2 WT::pAS8 dat (under its native promoter) inserted at the SaPI1 site of WT JE2 This study 1258 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 50 1259 1260 Table S6: Plasmids used in this study Plasmids Description Source pLI50 E. coli- S. aureus shuttle vector (55) pJB38 E. coli- S. aureus allelic exchange vector (38) pJC1111 E. coli- S. aureus SaPI1 site integration vector (39) pJB68 E. coli- S. aureus cadmium inducible shuttle vector (38) pET28a Expression vector for purification of protein in E. coli BL21 (DE3) Novagen pAS3 pJC1111 based vector f or integration of WT copy of alr1 at the SaPI1 site This study pAS2 pJB38 based vector for alr2 chromosomal deletion This study pSP4 pLI50 dat (under control of its native promoter) This study pSP19 pJB38 based vector for pepV SD1 chromosomal deletion This study pSP20 pJB38 based vector for pepV SD1-467 chromosomal deletion This study pSP16 pJB38 based vector for pepV chromosomal deletion This study pSP15 pJB38 based vector for substitution of chromosomal pepV with pepVQ12STOP This study pSP36 pJB68 based vector for overexpression of ddl This study pSP32 pET28a based vector for purification of full length Ddl (C -terminal his tag) This study pAQ59 E. coli - S. aureus SaPI1 site integration vector with pSC101 ori region (56) pAS8 pAQ59 based vector for integration of dat gene under its native promoter at the SaPI1 site This study 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 51 1277 1278 1279 Table S7: Primers used in this study 1280 1281 Gene/Modification Primer name Primer sequence (5' – 3') alr1 alr1_F TGCTGACGAACCAGGAGATA alr1_R TGTAGTTGGGTCAGTAGCTG alr1 complementation alr1_comp_F CGGCCGCTGCATGCCTGCAGACATGAGCAACGTAAA ATTG alr1_comp_R AGCTCGGTACCCGGGGATCCAATGACCTTTAATTACT CTAATGATAAC citZ 1641_F CAGCGGAGACTAAAATAAGTTC 1641_R CCCAATCTCAGATAACATCGTC dat dat_F ACTATAGGTGGCGGTACTTA dat_R ACCATCGGATATCTTCAACG alr2 deletion alr2_UP_F CGAGGCCCTTTCGTCTTCAATACTTAGAAGGTAATGG CTC alr2_UP2_R TCATAGCACTTGCTGTCAATGTATTACAC alr2_DN2_F ATTGACAGCAAGTGCTATGAATCATGATTC alr2_DN_R TTGCATGCCTGCAGGTCGACGCTTCTTCATTTCTATTA ACAAG dat complementation dat promoter_F CCTTTCGTCTTCAAGAATTCGATGTGAGTAGGACAGA AATG dat promoter_R TTTTTTCCATTCGAAATCGACTTCCTTTTTTC dat_pLI50_F TCGATTTCGAATGGAAAAAATTTTTTTAAATGGTG dat_pLI50_R TTGCATGCCTGCAGGTCGACCGAAAGTTGATAAATTT AAGTAATTTAATC .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 52 TSS identification of dat operon pepV_TSS_R1 P-CCATCTCTATGTGCAATTTC pepV_TSS_R2 GCGTCTTCTGATGCTTTTGC pepV_TSS_F3 GTCCTCGTAAGGCATTAGAC M13F (-20) GTAAAACGACGGCCAG M13R CAGGAAACAGCTATGAC pepVΔSD1-467 RBS1pepV_UP_F CCTTTCGTCTTCAAGAATTCAGCGACGCAATTAGGAA C RBS1pepV_UP_R TTATTCCTCCTTTTTCTATAAGTTAAATTCTATTTTACAT GAAAAG 1282 RBS1pepV_DN_F TATAGAAAAAGGAGGAATAATATATGGAAAAAATTTTT TTAAATG RBS1pepV_DN_R TATAGAAAAAGGAGGAATAATATATGGAAAAAATTTTT TTAAATG pepV deletion pepV_UP2_F CCTTTCGTCTTCAAGAATTCAACAATTAAAGAAGTAAA AACAAATC pepV_UP2_R TTTTTTCCATTCGAAATCGACTTCCTTTTTTC pepV_DN2_F TCGATTTCGAATGGAAAAAATTTTTTTAAATGGTG pepV_DN2_R TTGCATGCCTGCAGGTCGACTTTCAACTGAAAATGAG AAAC pepVQ12STOP pepV_STOP_UP _F CCTTTCGTCTTCAAGAATTCCAAATCCGAAAGAATATG C pepV_STOP_UP _R TAATGATTTAATCTTCGTATTGTTGAACTTTTTC pepV_STOP_D N_F ATACGAAGATTAAATCATTAATGACTTAAAAGGATTATT AG pepV_STOP_D N_R TTGCATGCCTGCAGGTCGACAAAGACCTGCGTTTTCA TTATC ddl overexpression plasmid ddl_pJB68_F TTTATAAGGAGGAAAAACATATGACAAAAGAAAATATT TGTATCG .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 53 ddl_pJB68_R GAATAGGCGCGCCTGAATTCATCCATGATTGAATTTG CTTTAATG Ddl purification plasmid ddl_C_Histag_F CTTTAAGAAGGAGATATACCATGACAAAAGAAAATATT TGTATCG ddl_C_Histag_R CAGTGGTGGTGGTGGTGGTGGTCAATTTTGTATTTAT TTTTCTGTTTATC ddl RT-qPCR ddl_RT_F GGGCTTTTTGAAGTTTTGGA ddl_RT_R TGGTAACCCTCGATGTTCAA murF RT-qPCR murF_RT_F TCACAATTGATTCACGAGCA murF_RT_R CCCAGCACCATCTTGTAATG dat RT-qPCR dat RT_F GATGGTTACGTTGCGACATT dat RT_R CACCTCGATGTTGAATTGCT sigA RT-qPCR JE2_RT_sigA_F AACTGAATCCAAGTGATCTTAGTG JE2_RT_sigA_R TCATCACCTTGTTCAATACGTTTG dat insertion at the SaPI1 site (pAS8) dat_UP2_F GAGCCGCTGCATGCCTGCAGGATGTGAGTAGGACAG AAATG dat_UP_R TTTTTTCCATTCGAAATCGACTTCCTTTTTTC dat_DN_F TCGATTTCGAATGGAAAAAATTTTTTTAAATGGTG dat_DN_R AGCTCGGTACCCGGGGATCCCGAAAGTTGATAAATTT AAGTAATTTAATC 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 54 1294 1295 1296 1297 Table S8: Table of Multiple Reaction Monitoring (MRM) transitions Metabolite Polarity MRM (Q1/Q3) CE (V) DP (V) RT Column L-Ala (+) 90.1 / 44.0 17 65 6.4 C D-Ala (+) 90.1 / 44.0 17 65 9.5 C D-Ala-D-Ala (+) 161.0 / 44.2 30.5 65 3.4 XB UDP-NAG (-) 606.0 / 79.0 -149 -80 12.7 XB UDP-NAM (-) 678.1 / 79.0 -120 -80 13.2 XB UDP-NAM-A (-) 749.1 / 403.0 -42 -90 13.4 XB UDP-NAM-AE (-) 878.2 / 403.0 -48 -105 14.3 XB UDP-NAM-AEK (-) 1006.2 / 403.0 -50 -130 14.9 XB UDP-NAM-AEKAA (-) 1148.5 / 403.0 -55 -140 14.6 XB NAM (-) 292.0 / 89.0 -16 -30 6.2 XB NAG (+) 204.0 / 138.1 18.9 30 5.6 XB Br-ATP (IS) (-) 588.0 / 159.0 -38.9 -60 6.0 XB Ribitol (IS) (-) 151.1 / 89.0 -14.8 -60 5.3 XB 1298 CE: Collision energy ;DP: Declustering potential; RT: Retention Time; C: CHIROBIOTIC® T column; XB: 1299 XBridge Amide column 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 55 1325 1326 1327 1328 1329 1330 Table S9: HRMS base peak identification of isotopologues Metabolite Isotopologues(13C15N) Base peak (m/z) D-Ala-D-Ala (Positive mode) 1 C6H12N2O3 C0N0 161.0921 2 [13]C1C5H12N2O3 C1N0 162.0954 3 [13]C2C4H12N2O3 C2N0 163.0988 4 [13]C3C3H12N2O3 C3N0 164.1021 5 [13]C3C3H12[15]N1N1O3 C3N1 165.09917 6 [13]C3C3H12[15]N2O3 C3N2 166.0962 7 [13]C4C2H12N2O3 C4N0 165.10549 8 [13]C5C1H12N2O3 C5N0 166.10884 9 [13]C6H12N2O3 C6N0 167.1122 10 [13]C6H12[15]N1N1O3 C6N1 168.1092 11 [13]C6H12[15]N2O3 C6N2 169.1063 12 [13]C1C5H12[15]N1N1O3 C1N1 163.09246 13 [13]C1C5H12[15]N2O3 C1N2 164.0895 14 [13]C2C4H12[15]N1N1O3 C2N1 164.0958 15 [13]C2C4H12[15]N2O3 C2N2 165.09285 16 [13]C4C2H12[15]N1N1O3 C4N1 166.1025 17 [13]C4C2H12[15]N2O3 C4N2 167.09956 18 [13]C5C1H12[15]N1N1O3 C5N1 167.10588 19 [13]C5C1H12[15]N2O3 C5N2 168.1029 20 C6H12[15]N1N1O3 C0N1 162.0891 21 C6H12[15]N2O3 C0N2 163.0861 D-Glu (Negative mode) 1 C5H9NO4 C0N0 146.04588 2 C5H9[15]NO4 C0N1 147.04292 3 [13]C1C4H9NO4 C1N0 147.04924 4 [13]C2C3H9NO4 C2N0 148.05259 5 [13]C2C3H9[15]NO4 C2N1 149.04963 6 [13]C1C4H9[15]NO4 C1N1 148.04627 7 [13]C3C2H9[15]NO4 C3N1 150.05298 8 [13]C4C1H9[15]NO4 C4N1 151.05634 9 [13]C5H9[15]NO4 C5N1 152.05969 10 [13]C3C2H9NO4 C3N0 149.05595 11 [13]C4C1H9NO4 C4N0 150.0593 12 [13]C5H9NO4 C5N0 151.06266 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 56 1331 RT: D-Ala-D-Ala, 3.4 mins on 10 cm XBridge amide column; D-Glu, 6.4 mins on CHIROBIOTIC® T column 1332 1333 1334 1335 1336

1337 1338 1. Passalacqua KD, Charbonneau ME, & O'Riordan MXD (2016) Bacterial metabolism 1339 shapes the host-pathogen interface. Microbiol Spectr 4(3). 1340 2. Brestoff JR & Artis D (2013) Commensal bacteria at the interface of host metabolism 1341 and the immune system. Nat Immunol 14(7):676-684. 1342 3. Tomlinson KL , et al. (2021) Staphylococcus aureus induces an itaconate-dominated 1343 immunometabolic response that drives biofilm formation. Nat Commun 12(1):1399. 1344 4. Heim CE, et al. (2020) Lactate production by Staphylococcus aureus biofilm inhibits 1345 HDAC11 to reprogramme the host immune response during persistent infection. Nat 1346 Microbiol 5(10):1271-1284. 1347 5. Schlatterer K, Peschel A, & Kretschmer D (2021) Short-chain fatty acid and FFAR2 1348 Activation - A new option for treating infections? Front Cell Infect Microbiol 11:785833. 1349 6. Salmond CV, Kroll RG, & Booth IR (1984) The effect of food preservatives on pH 1350 homeostasis in Escherichia coli. J Gen Microbiol 130(11):2845-2850. 1351 7. Roe AJ, McLaggan D, Davidson I, O'Byrne C, & Booth IR (1998) Perturbation of anion 1352 balance during inhibition of growth of Escherichia col i by weak acids. J Bacteriol 1353 180(4):767-772. 1354 8. Cummings JH, Pomare EW, Branch WJ, Naylor CP, & Macfarlane GT (1987) Short 1355 chain fatty acids in human large intestine, portal, hepatic and venous blood. Gut 1356 28(10):1221-1227. 1357 9. Acton DS, Plat-Sinnige MJ, van Wamel W, de Groot N, & van Belkum A (2009) Intestinal 1358 carriage of Staphylococcus aureus: how does its frequency compare with that of nasal 1359 carriage and what is its clinical impact? Eur J Clin Microbiol Infect Dis 28(2):115-127. 1360 10. Piewngam P , et al. (2023) Probiotic for pathogen-specific Staphylococcus aureus 1361 decolonisation in Thailand: a phase 2, double-blind, randomised, placebo-controlled trial. 1362 Lancet Microbe 4(2):e75-e83. 1363 11. Thomas VC , et al. (2014) A central role for carbon-overflow pathways in the modulation 1364 of bacterial cell death. PLoS Pathog 10(6):e1004205. 1365 12. Alqarzaee AA , et al. (2021) Staphylococcal ClpXP protease targets the cellular 1366 antioxidant system to eliminate fitness-compromised cells in stationary phase. Proc Natl 1367 Acad Sci U S A 118(47). 1368 13. Mordukhova EA & Pan JG (2013) Evolved cobalamin-independent methionine synthase 1369 (MetE) improves the acetate and thermal tolerance of Escherichia coli . Appl Environ 1370 Microbiol 79(24):7905-7915. 1371 14. Roe AJ, O'Byrne C, McLaggan D, & Booth IR (2002) Inhibition of Escherichia coli growth 1372 by acetic acid: a problem with methionine biosynthesis and homocysteine toxicity. 1373 Microbiology (Reading) 148(Pt 7):2215-2222. 1374 15. Gries CM , et al. (2016) Potassium uptake modulates Staphylococcus aureus 1375 metabolism. mSphere 1(3). 1376 16. Carpenter CE & Broadbent JR (2009) External concentration of organic acid anions and 1377 pH: key independent variables for studying how organic acids inhibit growth of bacteria 1378 in mildly acidic foods. J Food Sci 74(1):R12-15. 1379 17. Halsey CR , et al. (2017) Amino acid catabolism in Staphylococcus aureus and the 1380 function of carbon catabolite repression. mBio 8(1). 1381 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 57 18. Moscoso M, Garcia P, Cabral MP, Rumbo C, & Bou G (2018) A D-Alanine auxotrophic 1382 live vaccine is effective against lethal infection caused by Staphylococcus aureus . 1383 Virulence 9(1):604-620. 1384 19. Cava F, Lam H, de Pedro MA, & Waldor MK (2011) Emerging knowledge of regulatory 1385 roles of D-amino acids in bacteria. Cell Mol Life Sci 68(5):817-831. 1386 20. Burlak C , et al. (2007) Global analysis of community-associated methicillin-resistant 1387 Staphylococcus aureus exoproteins reveals molecules produced in vitro and during 1388 infection. Cell Microbiol 9(5):1172-1190. 1389 21. Goldstein JM, Kordula T, Moon JL, Mayo JA, & Travis J (2005) Characterization of an 1390 extracellular dipeptidase from Streptococcus gordonii FSS2. Infect Immun 73(2):1256-1391 1259. 1392 22. Liu S , et al. (2006) Allosteric inhibition of Staphylococcus aureus D-Alanine:D-Alanine 1393 ligase revealed by crystallographic studies. Proc Natl Acad Sci U S A 103(41):15178-1394 15183. 1395 23. Lu Y, Xu H, & Zhao X (2010) Crystal structure of the apo form of D-Alanine:D-Alanine 1396 ligase (Ddl) from Streptococcus mutans. Protein Pept Lett 17(8):1053-1057. 1397 24. Pederick JL, Thompson AP, Bell SG, & Bruning JB (2020) D-Alanine-D-alanine ligase as 1398 a model for the activation of ATP-grasp enzymes by monovalent cations. J Biol Chem 1399 295(23):7894-7904. 1400 25. Huynh KH , et al. (2015) The crystal structure of the D-alanine -D-alanine ligase from 1401 Acinetobacter baumannii suggests a flexible conformational change in the central 1402 domain before nucleotide binding. J Microbiol 53(11):776-782. 1403 26. Bruning JB, Murillo AC, Chacon O, Barletta RG, & Sacchettini JC (2011) Structure of the 1404 Mycobacterium tuberculosis D-Alanine:D-Alanine ligase, a target of the antituberculosis 1405 drug D-Cycloserine. Antimicrob Agents Chemother 55(1):291-301. 1406 27. Russell JB & Diez-Gonzalez F (1998) The effects of fermentation acids on bacterial 1407 growth. Adv Microb Physiol 39:205-234. 1408 28. Russell JB (1992) Another explanation for the toxicity of fermentation acids at low pH: 1409 anion accumulation versus uncoupling. Journal of Applied Bacteriology 73(5):363-370. 1410 29. Wolfe AJ (2005) The acetate switch. Microbiol Mol Biol Rev 69(1):12-50. 1411 30. Somerville GA & Proctor RA (2009) At the crossroads of bacterial metabolism and 1412 virulence factor synthesis in staphylococci. Microbiol Mol Biol Rev 73(2):233-248. 1413 31. Hammes WP & Neuhaus FC (1974) On the specificity of phospho-N-acetylmuramyl-1414 pentapeptide translocase. The peptide subunit of uridine diphosphate -N-actylmuramyl-1415 pentapeptide. J Biol Chem 249(10):3140-3150. 1416 32. Sobral RG, Ludovice AM, de Lencastre H, & Tomasz A (2006) Role of murF in cell wall 1417 biosynthesis: isolation and characterization of a murF conditional mutant of 1418 Staphylococcus aureus. J Bacteriol 188(7):2543-2553. 1419 33. Vemula H, Ayon NJ, & Gutheil WG (2016) Cytoplasmic peptidoglycan intermediate 1420 levels in Staphylococcus aureus. Biochimie 121:72-78. 1421 34. Huang C, Hernandez-Valdes JA, Kuipers OP, & Kok J (2020) Lysis of a Lactococcus 1422 lactis dipeptidase mutant and rescue by mutation in the pleiotropic regulator CodY. Appl 1423 Environ Microbiol 86(8). 1424 35. Huber M , et al. (2019) Translational coupling via termination-reinitiation in archaea and 1425 bacteria. Nat Commun 10(1):4006. 1426 36. Rex G, Surin B, Besse G, Schneppe B, & McCarthy JE (1994) The mechanism of 1427 translational coupling in Escherichia coli. Higher order structure in the atpHA mRNA acts 1428 as a conformational switch regulating the access of de novo initiating ribosomes. J Biol 1429 Chem 269(27):18118-18127. 1430 37. Fey PD, et al. (2013) A genetic resource for rapid and comprehensive phenotype 1431 screening of nonessential Staphylococcus aureus genes. mBio 4(1):e00537-00512. 1432 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint 58 38. Bose JL, Fey PD, & Bayles KW (2013) Genetic tools to enhance the study of gene 1433 function and regulation in Staphylococcus aureus . Appl Environ Microbiol 79(7):2218-1434 2224. 1435 39. Chen J, Yoong P, Ram G, Torres VJ, & Novick RP (2014) Single-copy vectors for 1436 integration at the SaPI1 attachment site for Staphylococcus aureus. Plasmid 76:1-7. 1437 40. Hussain M, Hastings JG, & White PJ (1991) A chemically defined medium for slime 1438 production by coagulase-negative staphylococci. J Med Microbiol 34(3):143-147. 1439 41. Vemula H, Bobba S, Putty S, Barbara JE, & Gutheil WG (2014) Ion-pairing liquid 1440 chromatography-tandem mass spectrometry-based quantification of uridine 1441 diphosphate-linked intermediates in the Staphylococcus aureus cell wall biosynthesis 1442 pathway. Anal Biochem 465:12-19. 1443 42. Trefely S, Ashwell P, & Snyder NW (2016) FluxFix: automatic isotopologue 1444 normalization for metabolic tracer analysis. BMC Bioinformatics 17(1):485. 1445 43. Patiny L & Borel A (2013) ChemCalc: a building block for tomorrow's chemical 1446 infrastructure. J Chem Inf Model 53(5):1223-1228. 1447 44. Fendt SM , et al. (2013) Metformin decreases glucose oxidation and increases the 1448 dependency of prostate cancer cells on reductive glutamine metabolism. Cancer Res 1449 73(14):4429-4438. 1450 45. Wang C, Lee J, Deng Y, Tao F, & Zhang LH (2012) ARF-TSS: an alternative method for 1451 identification of transcription start site in bacteria. Biotechniques 52(4). 1452 46. de Jonge BL, Chang YS, Gage D, & Tomasz A (1992) Peptidoglycan composition of a 1453 highly methicillin-resistant Staphylococcus aureus strain. The role of penicillin binding 1454 protein 2A. J Biol Chem 267(16):11248-11254. 1455 47. De Jonge BL, Gage D, & Xu N (2002) The carboxyl terminus of peptidoglycan stem 1456 peptides is a determinant for methicillin resistance in Staphylococcus aureus. Antimicrob 1457 Agents Chemother 46(10):3151-3155. 1458 48. Kuhner D, Stahl M, Demircioglu DD, & Bertsche U (2014) From cells to muropeptide 1459 structures in 24 h: peptidoglycan mapping by UPLC-MS. Sci Rep 4:7494. 1460 49. Alvarez L, Hernandez SB, de Pedro MA, & Cava F (2016) Ultra-sensitive, high-resolution 1461 liquid chromatography methods for the high-throughput quantitative analysis of bacterial 1462 cell wall chemistry and structure. Methods Mol Biol 1440:11-27. 1463 50. Winn MD , et al. (2011) Overview of the CCP4 suite and current developments. Acta 1464 Crystallogr D Biol Crystallogr 67(Pt 4):235-242. 1465 51. Afonine PV , et al. (2012) Towards automated crystallographic structure refinement with 1466 phenix.refine. Acta Crystallogr D Biol Crystallogr 68(Pt 4):352-367. 1467 52. Emsley P, Lohkamp B, Scott WG, & Cowtan K (2010) Features and development of 1468 Coot. Acta Crystallogr D Biol Crystallogr 66(Pt 4):486-501. 1469 53. Chen VB, et al. (2010) MolProbity: all-atom structure validation for macromolecular 1470 crystallography. Acta Crystallogr D Biol Crystallogr 66(Pt 1):12-21. 1471 54. Kreiswirth BN , et al. (1983) The toxic shock syndrome exotoxin structural gene is not 1472 detectably transmitted by a prophage. Nature 305(5936):709-712. 1473 55. Lee CY, Buranen SL, & Ye ZH (1991) Construction of single-copy integration vectors for 1474 Staphylococcus aureus. Gene 103(1):101-105. 1475 56. Jacquet R , et al. (2019) Dual gene expression analysis identifies factors associated with 1476 Staphylococcus aureus virulence in diabetic mice. Infect Immun 87(5):e00163-00119. 1477 1478 .CC-BY 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted January 17, 2024. ; https://doi.org/10.1101/2024.01.15.575639doi: bioRxiv preprint