The Bacillus subtilis cell envelope stress-inducible ytpAB operon modulates membrane properties and contributes to bacitracin resistance

ABSTRACT

Antibiotics that inhibit peptidoglycan synthesis trigger the activation of both specific and general protective responses. σ^M responds to diverse antibiotics that inhibit cell wall synthesis. Here, we demonstrate that cell wall-inhibiting drugs, such as bacitracin and cefuroxime, induce the σ^M-dependent ytpAB operon. YtpA is a predicted hydrolase previously proposed to generate the putative lysophospholipid antibiotic bacilysocin (lysophosphatidylglycerol), and YtpB is the branchpoint enzyme for the synthesis of membrane-localized C₃₅ terpenoids. Using targeted lipidomics, we reveal that YtpA is not required for the production of lysophosphatidylglycerol. Nevertheless, ytpA was critical for growth in a mutant strain defective for homeoviscous adaptation due to a lack of genes for the synthesis of branched chain fatty acids and the Des phospholipid desaturase. Consistently, overexpression of ytpA increased membrane fluidity as monitored by fluorescence anisotropy. The ytpA gene contributes to bacitracin resistance in mutants additionally lacking the bceAB or bcrC genes, which directly mediate bacitracin resistance. These epistatic interactions support a model in which σ^M-dependent induction of the ytpAB operon helps cells tolerate bacitracin stress, either by facilitating the flipping of the undecaprenyl phosphate carrier lipid or by impacting the assembly or function of membrane-associated complexes involved in cell wall homeostasis.

IMPORTANCE

Peptidoglycan synthesis inhibitors include some of our most important antibiotics. In Bacillus subtilis, peptidoglycan synthesis inhibitors induce the σ^M regulon, which is critical for intrinsic antibiotic resistance. The σ^M-dependent ytpAB operon encodes a predicted hydrolase (YtpA) and the enzyme that initiates the synthesis of C₃₅ terpenoids (YtpB). Our results suggest that YtpA is critical in cells defective in homeoviscous adaptation. Furthermore, we find that YtpA functions cooperatively with the BceAB and BcrC proteins in conferring intrinsic resistance to bacitracin, a peptide antibiotic that binds tightly to the undecaprenyl-pyrophosphate lipid carrier that sustains peptidoglycan synthesis.

INTRODUCTION

The bacterial cell envelope, minimally consisting of a plasma membrane and a peptidoglycan cell wall, is the major barrier between the cell interior and the extracellular environment (1). Peptidoglycan is a rigid and highly cross-linked structure that confers cell shape and resists turgor pressure to prevent lysis. It is also the target of some of the most clinically relevant antibiotics (2). Bacteria that are exposed to changing environmental conditions, including antibiotics, respond through dedicated cell envelope stress response (CESR) pathways and modulate gene expression to protect the integrity of the cell envelope (3, 4).

Following exposure to a cell envelope stress, an extracellular signal must be communicated across the cell envelope to mediate a transcriptional response. Bacterial CESRs are commonly controlled by two-component system (TCS) regulatory networks or by alternative σ factors [often members of the extracytoplasmic function (ECF) σ family] that are in turn regulated by stress-responsive anti-σ factors (4–6). For example, the Bacillus subtilis BceRS TCS responds selectively to the peptide antibiotic bacitracin that binds to undecaprenyl-pyrophosphate (UPP) and inhibits the recycling of lipid II (7, 8). Activation of the BceRS system upregulates BceAB, an ABC transporter and a primary determinant of intrinsic bacitracin resistance (8). BceAB acts through a target protection mechanism to release bacitracin from inhibited UPP:bacitracin complexes (9). The BceAB complex also collaborates with the BceS sensor kinase to respond to bacitracin stress in a flux-sensing mechanism (10, 11).

In addition to TCSs, ECF σ factors play a prominent role in the regulation of CESRs (4, 6, 12). Upon sensing a cell envelope stress, the membrane-embedded anti-σ factor is inactivated (often by proteolytic cleavage) to release the active σ factor (13–16). One such ECF σ, σ^M, is activated by cell wall-targeting antimicrobials (17–20), although the precise σ^M-activating stimulus is unclear. The activation of σ^M induces the expression of nearly 100 genes, many of which are involved in peptidoglycan synthesis, cell shape determination, and cell division (6, 18). Consistently, deletion of sigM sensitizes the cell to β-lactams (21), moenomycin (20), bacitracin (17), and other antibiotics that target peptidoglycan synthesis (22).

The large σ^M regulon includes several operons with poorly characterized roles in responding to envelope stress (18). The ytpAB operon is one such example. Previously, the ytpA gene product was classified as a Class A₂-phospholipase that cleaves the sn2 acyl chain from phosphatidylglycerol (PG) to produce a discrete lysophospholipid [1-(12-methyltetradecanoyl)-3-phosphoglyceroglycerol]. This lysophospholipid has been named bacilysocin and was suggested to function as an antibiotic to inhibit growth of neighboring microorganisms (23). However, purified bacilysocin has weak antibiotic activity with an MIC of 5 µg/mL for Saccharomyces cerevisiae and 25 µg/mL for Staphylococcus aureus (23), and there is no evidence that it is released into the media at levels sufficient to demonstrate antibiotic activity.

The second gene in the operon, ytpB, encodes an enzyme required for sesquarterpenoid synthesis (tetraprenyl-β-curcumene synthase) (24, 25). Sesquarterpenoids are 35 carbon (C₃₅) cyclic compounds derived from heptaprenyl pyrophosphate (HPP) and have a multi-ring structure similar to C₃₀ hopanoids, which are derived from squalene (26). The major sesquarterpenoid that was produced by B. subtilis is designated as baciterpenol A (24), which can be further modified by autoxidation and dehydration (under non-physiological isolation conditions) to generate baciterpenol B and sporulenes (27). Deletion of ytpB leads to a modest increase in cell sensitivity toward bacitracin (28). Our previous findings revealed that this effect results from increased accumulation of the YtpB substrate, HPP (28). The linear C₃₅ HPP isoprenoid is a structural analog of the longer C₅₅ isoprenoid, UPP, and both contain a membrane-proximal pyrophosphate moiety, which is the ligand for bacitracin (29).

Here, we have explored the role of YtpA, a putative phospholipase, on membrane properties and bacitracin sensitivity. Overexpression of ytpA increased membrane fluidity, but in contrast with a prior report (23), YtpA was not required for lysophosphatidylglycerol (LPG) production. Genetic studies reveal that ytpA is critical for the fitness of cells defective in homeoviscous adaptation. Moreover, YtpA contributes to bacitracin resistance in parallel with the BceAB and BcrC resistance systems. We propose that YtpA may support peptidoglycan synthesis by modulating membrane properties to enhance the function of the synthetic machinery and perhaps to facilitate the transmembrane flipping of the UP carrier lipids (30).

RESULTS

Overexpression of YtpA increases membrane fluidity

In previous studies, YtpA was identified as a lysophospholipase responsible for synthesis of bacilysocin [1-(12-methyltetradecanoyl)-3-phosphoglycerol (1-15-LPG)], a lysophospholipid derived from PG with a 15-carbon anteiso branched chain fatty acid (BCFA) (23). In eukaryotes, lysophospholipids participate in membrane remodeling via the Lands cycle in which phospholipids can be deacylated and then reacylated with chemically distinct fatty acids (31). However, the presence of this type of membrane remodeling pathway in bacteria is yet to be established (32). If lysophospholipids persist within the membrane bilayer, they may alter local membrane curvature, permeability, and fluidity (33–35).

To determine if YtpA impacts membrane fluidity, we used the membrane intercalating dye 1,6-diphenyl-1,3,5-hexatriene (DPH) to perform fluorescence anisotropy (FA) (36). The rotational freedom of DPH in the membrane serves as an indicator of membrane fluidity. There was no significant difference in FA between the B. subtilis 168 (trpC2) wild-type (WT) strain and an isogenic ytpA null strain (ΔytpA) (Fig. 1). However, since the ytpAB operon is known to be stress-inducible (18), this may simply mean that ytpA is poorly expressed under these growth conditions. Therefore, we tested the effect of YtpA overexpression using an isopropyl β-D-1-thiogalactopyranoside (IPTG)-inducible promoter. Indeed, expression of ytpA led to a decrease in FA compared to the WT and ΔytpA strains (Fig. 1). This suggests an increase in rotational freedom, which is indicative of an increase in membrane fluidity (36). In a previous study, DPH measurements of FA in vesicles made from B. subtilis membrane lipids revealed a near linear decrease in FA (an increase in fluidity) over the temperature range from 10°C to 45°C (37). The change observed here between WT cells and those with induction of ytpA is comparable to vesicles incubated at temperatures differing by 15°C–20°C (37). Furthermore, a similar magnitude of change was seen in B. subtilis cells with and without induction of the σ^W-dependent membrane stress response, which significantly protects cells against detergents and other agents that increase fluidity (38). Thus, the effect seen here is likely to be physiologically significant.

Fig 1.

Induction of ytpA increases membrane fluidity. Overexpression of ytpA using the spac(hy) promoter (HB27450) with 1 mM IPTG yields statistically significant differences in anisotropy compared to the WT and ytpA knockout (ΔytpA, HB27232) strains. N = 3 biological replicates. A one-way analysis of variance with a Tukey test for multiple comparisons was performed. Columns labeled with different letters are statistically distinct from each other, with a P-value cutoff of <0.05.

Induction of ytpA rescues growth of cells defective in homeoviscous adaptation

To test if the membrane fluidizing effect noted upon induction of ytpA is physiologically relevant, we took advantage of a reporter strain with an artificially rigid membrane (39). This strain, designated Δbkd, is defective in homeoviscous adaptation to conditions of low fluidity due to deletions of the bkd operon and the des gene. These mutations prevent the synthesis of BCFAs and the desaturation of acyl chains by the Des desaturase, respectively (39). Approximately 90% of the B. subtilis membrane is composed of BCFAs that help confer an optimal fluidity necessary for the maintenance of the electron transport chain (40). As a result, the Δbkd strain requires supplementation with precursors to BCFAs for normal growth. For example, supplementing with 2-methylbutyric acid (MB) restores the ability to synthesize anteiso BCFAs and rescues growth in minimal medium (39).

The Δbkd strain has a minor growth defect compared to WT when grown on lysogeny broth (LB) medium at 27°C, 37°C, and 45°C. However, upon deletion of ytpA (Δbkd ΔytpA), fitness was dramatically reduced. The colony size of the Δbkd ΔytpA strain was very small compared to both the WT and Δbkd strains under all the temperatures tested (Fig. 2A). As previously reported, the Δbkd strain is inviable at 22°C on minimal medium lacking BCFA precursors (Fig. 2B). Remarkably, Δbkd with a copy of ytpA expressed from the spac(hy) promoter is able to grow at 22°C, and if MB is additionally present, this strain grows as well as WT (Fig. 2B). The rescue of growth by YtpA is observed both with and without addition of the inducer IPTG, consistent with the known leaky expression of the spac(hy) promoter (41). Using real-time PCR, we estimate that the leaky expression from this promoter leads to a twofold increase in gene expression compared to its native expression in the WT cells. These results suggest that ytpA expression is critical to compensate for the growth-limiting defects that define the Δbkd strain.

Fig 2.

YtpA is physiologically important in cells with defects in membrane fluidity (A). A clean, unmarked deletion of ytpA (ΔytpA) in the Δbkd strain (HB27482) reduces the growth of the cells at the permissive temperatures of 27°C, 37°C, and 45°C on LB medium. The colony size of the Δbkd ΔytpA strain is significantly smaller than that of the Δbkd strain (HB27373). (B) Overexpression of ytpA in Δbkd cells (HB27384) restores viability on minimal media when grown at non-permissive low temperature (22°C). A representative image is shown (N = 3). Untreated column represents cells plated on minimal media without any supplementation. IPTG column represents cells plated on minimal media supplemented with 1 mM IPTG, and MB column represents cells plated on minimal media supplemented with 100 µM MB. (C) Induction of ytpA from the IPTG-inducible spac(hy) promoter partially restores fluidity in a Δbkd strain. The data presented are the average of three biological replicates where errors bars represent the standard deviation. A one-way analysis of variance with a Tukey test for multiple comparisons was performed. Columns labeled with different letters are statistically distinct from each other with a P-value cutoff of <0.05.

We next used FA to test if induction of YtpA increases membrane fluidity of the Δbkd strain (Fig. 2C). As reported previously (39), the Δbkd strain shows an increase in FA compared to WT. Induction of ytpA in the Δbkd strain led to a significant decrease in FA, although not a complete restoration back to the levels of WT cells (Fig. 2C). This is consistent with the partial rescue of growth by induction of ytpA in the Δbkd strain at 22°C without MB (Fig. 2B). We conclude that expression of YtpA increases membrane fluidity and restores growth of a strain defective in biochemical pathways that normally serve to increase membrane fluidity.

YtpA is not the major phospholipase in vivo

YtpA was proposed to be a phospholipase A₂ responsible for the release of 1-15-LPG into the medium (23). Because lysophospholipids may impact membrane biology, we performed a targeted lipidomic analysis to determine if deletion of ytpA altered the phospholipid and lysophospholipid content of the cell. In both the WT and ytpA mutant strains (ΔytpA), the major PG and phosphatidylethanolamine (PE) species were the same, with the most abundant species having a total of 30, 31, or 32 carbons in the acyl chains, with a C₁₅ fatty acid in the 2-position. The minor 28 PG/PE and the 29 PG/PE peaks have a C₁₃ or C₁₄ acyl chain in the 2-position, respectively (Fig. S1A and B).

Next, we analyzed the LPG and lysophosphatidylethanolamine (LPE) composition of cells (Fig. 3; Fig. S1C and D and S2). For quantitation, the signals arising from the 1- and 2-acyl-lysophospholipids with the same carbon number were combined. Note that bacilysocin, the previously described C₁₅ 1-acyl-lysophospholipid (23), most likely results from the action of an unidentified A₁ phospholipase (producing a 2-acyl-lysophospholipid as the product) followed by fatty acyl chain migration (Fig. S1), which reaches ~90% at the 1-position at equilibrium (42–44). Similar acyl chain migration was seen in recent studies monitoring lysophospholipid production in S. aureus (45).

Fig 3.

Lysophospholipid content of cells and media in WT and ΔytpA strains (HB27232). Strains were grown in LB to late-log phase, the cells or media were extracted with methanol, and the LPG/LPE molecular species were determined by liquid chromatography tandem mass spectrometry (LC-MS/MS). The LPG/LPE abundances were determined relative to a [d5]17-LPG internal standard. WT (red); ΔytpA (blue). (A) Cellular LPG, (B) media LPG, (C) cellular LPE, and (D) media LPE. Student t-test was done to compare the values of WT and ytpA samples for each molecular species separately. * indicates a P value of <0.05.

In growing cells and in the medium, the dominant lysophospholipid was 15-LPG (Fig. 3A and B), with minor amounts of 13- and 14-LPG in cells (Fig. 3A). 15-LPE was also a major species in both cells (Fig. 3C) and media (Fig. 3D) in late-log phase. There was a modest, but statistically significant, reduction in lysophospholipids in the ΔytpA strain in growing cells (Fig. 3A and C). There was little if any effect noted in stationary phase cells (Fig. S2A and C). One notable change in the stationary phase cultures was that the amount of lysophospholipids in the medium was significantly elevated (10-fold) compared to samples taken in late-log phase (Fig. 2B and D). The presence of lysophospholipids in the cellular fraction of the ΔytpA strain suggests that YtpA is not responsible for the bulk of lysophospholipid synthesis in B. subtilis. Consistently, induction of ytpA with IPTG did not result in an increase in lysophospholipids compared to either the uninduced condition or WT (Fig. S3).

YtpA is a member of a large superfamily of serine-dependent hydrolases (alpha-beta hydrolases) with a wide variety of substrates. Bioinformatic searches indicate that YtpA is likely a cytosolic-facing, membrane-associated protein. Sequence homology searches consistently yield sequence and domain similarities between YtpA and other phospholipases, including PldB, a poorly characterized phospholipase and the namesake of the COG2267 superfamily. Because traditional homology searches are limited to sequence similarity, we additionally used the YtpA AlphaFold2-generated structure to search protein structure databases using FoldSeek (46–48). Among the many proteins with similar predicted structures, biochemical information is available for only a handful. For example, YtpA has 30% identity to a secreted monoacylglycerol hydrolase from Mycobacterium tuberculosis (UNIPROT 007427) that hydrolyzes glycerol monoesters of long-chain fatty acids (49, 50). The lack of a clear functional role for YtpA highlights a recurring problem for this large family of alpha-beta hydrolases, enzymes that often still have enigmatic functions (51).

Cell envelope active antibiotics induce expression of ytpAB in a σ^M-dependent manner

Next, we evaluated the regulation of ytpA in conditions that lead to cell envelope stress. The ytpAB operon is regulated by σ^M, an alternative ECF sigma factor that is activated in response to various peptidoglycan synthesis inhibitors and other cell wall stressors (6). Early genome-wide transcriptome studies have revealed that ytpAB is most strongly induced by inhibitors of the membrane-associated steps of peptidoglycan biosynthesis and in particular by those compounds that interfere with the lipid II cycle such as bacitracin and vancomycin (18, 52). This response pattern is consistent with a recent comprehensive profiling study using both RNA-sequencing and tiling array methodologies in which bacitracin and vancomycin were the strongest inducers, followed by tunicamycin, moenomycin, and lysozyme (53).

We constructed a luciferase transcriptional reporter to monitor the expression of the ytpAB operon in response to cell envelope stresses. Consistent with expectation, the ytpAB reporter fusion was strongly induced by high levels of bacitracin (31.25 µg/mL), and this induction was lost if either the σ^M promoter site or the sigM gene was deleted (Fig. 4). The reporter fusion was also induced by cefuroxime (0.16 µg/mL), a drug that inhibits the activity of enzymes involved in peptidoglycan synthesis. Using real-time PCR, we observed a fourfold induction of ytpA after 15 min of treatment with 31.25 µg/mL of bacitracin and a twofold increase when cells were grown to mid-log phase in the presence of the same bacitracin concentration.

Fig 4.

Induction of the ytpAB operon is σ^M-dependent. Induction of P_ytpAB-lux (HB27246) by bacitracin (BAC) or cefuroxime (CEF) is lost in the absence of sigM (HB27287; purple) or the σ^M-specific consensus sequence of the ytpAB promoter (HB27247; yellow). N = 3 biological replicates; error bars represent standard deviation.

Deletion of bacitracin-resistant genes (bcrC and bceAB) significantly sensitizes the cell to bacitracin and is known to alter the expression of bacitracin-responsive genes (7). While a WT cell had a bacitracin MIC of 125 µg/mL, deletion of the intrinsic bacitracin resistance determinants bceAB and bcrC significantly reduced the MIC. In a bceAB or bcrC deletion background, ytpAB expression was induced by concentrations of bacitracin as low as 1.25 µg/mL (Fig. S4). The observation that ytpA expression is induced in response to cell wall acting drugs is suggestive of a role in intrinsic drug resistance.

The ytpAB operon confers bacitracin resistance

Next, we sought to determine if the ytpAB operon contributes to bacitracin resistance. The individual deletions of ytpA (ΔytpA) or ytpB (ΔytpB) did not have a significant effect on the growth of the cells with 62.5 µg/mL bacitracin (0.5× MIC). However, the ytpAB double mutant (ΔytpAB) had a notable growth lag (Fig. 5). An effect of YtpB on bacitracin resistance was noted previously in studies in Mueller-Hinton medium and was ascribed to the accumulation of the YtpB substrate HPP, a close chemical analog of UPP (28). HPP likely complexes with bacitracin and may reduce the efficiency of BceAB-dependent detoxification by competition for the active site of the BceAB resistance protein (9). HPP might also serve as a competitive substrate for the BcrC-dependent phosphatase (28). Although ΔytpB did not affect bacitracin resistance under the conditions we tested (LB medium), ytpA and ytpB together clearly contribute to bacitracin resistance.

Fig 5.

Effects of ΔytpA and ΔytpB (unmarked, in-frame deletions) on bacitracin sensitivity. Individual deletions of ytpA (HB27232) or ytpB (HB27253) do not have a significant effect on bacitracin (BAC) sensitivity compared to the WT cells. The ytpAB operon deletion (HB27407) has an increased lag in the presence of 62.5 µg/mL bacitracin compared to either single mutant. N = 6; standard deviation in the growth of each strain has been shown by shading.

Loss of YtpA increases bacitracin sensitivity in strains lacking BcrC or BceAB

We reasoned that if YtpA were contributing to bacitracin resistance by increasing membrane fluidity, one mechanism might be through facilitation of UP (or UPP) flipping across the membrane. To explore this, we monitored the bacitracin sensitivity of strains defective in recycling of the undecaprenyl carrier lipid due to lack of a UPP phosphatase (BcrC or UppP). The BcrC and UppP phosphatases are individually dispensable, but the double mutant is not viable (54, 55). In the presence of low levels of bacitracin (5 µg/mL), neither the ΔbcrC nor the ΔytpA mutant displayed much of a growth lag. However, the ΔytpA ΔbcrC double mutant was greatly inhibited with a >4-h growth lag (Fig. 6A). In contrast, there is very little additivity between ΔytpA and ΔuppP, even with high bacitracin levels (62.5 µg/mL) (Fig. 6B).

Fig 6.

Loss of YtpA increases bacitracin (BAC) sensitivity in genetically sensitized strains. (A) Deletion of ytpA (HB27360) increases the sensitivity of the bacitracin-sensitive bcrC mutant (HB27277) as measured with 5 µg/mL bacitracin. (B) Neither the loss of ytpA (HB27232) or uppP (HB27446) alone nor the combination (HB27447) has a major impact on bacitracin sensitivity as measured with 62.5 µg/mL bacitracin. (C) Deletion of ytpA (HB27273) increases the sensitivity of the bacitracin-sensitive bceAB mutant (HB27271) as measured with 5 µg/mL bacitracin. N = 5 biological replicates; standard deviation in the growth of each strain has been shown by shading.

Next, we explored the role of YtpA in strains lacking the BceAB resistance pathway. The BceAB proteins function in the dissociation of UPP:bacitracin complexes in a target protection mechanism of resistance (9). Consistent with prior work (9), ΔbceAB was highly sensitive to bacitracin with decreased growth observed at 5 µg/mL. At this concentration, the ΔytpA strain was unaffected, whereas the ΔytpA ΔbceAB double mutant was unable to grow (Fig. 6C). The additivity of YtpA with both BcrC and BceAB, the two major players of bacitracin resistance network, suggests a role for YtpA in bacitracin resistance. By increasing membrane fluidity, YtpA may reduce UPP levels on the outer leaflet of the membrane, perhaps by allowing it to flip inside.

Loss of YtpA has only a small effect in strains lacking the UptA UP flippase

Following the transglycosylation reaction, the UPP lipid carrier is dephosphorylated on the outer leaflet of the membrane. Then, the transmembrane flipping of the UP product is facilitated by DedA family membrane proteins (22, 56). In B. subtilis, the σ^M-regulated uptA (formerly yngC gene) encodes one such protein (22). A null mutant of uptA (ΔuptA) has no overt growth defect but displays an increased sensitivity to MX-2401 (22), an antibiotic that binds selectively to UP that is exposed on the outer leaflet of the membrane (57).

We speculated that YtpA-dependent membrane changes might also help to support the flipping of UP. However, deletion of ytpA did not increase the sensitivity of the uptA strain (ΔytpA ΔuptA) to the sub-MIC level of 0.6 µg/mL MX-2401 (Fig. 7A). Consistent with the notion that UptA mediates flipping of UP but not UPP, the uptA deletion (ΔuptA) had little effect on bacitracin sensitivity (Fig. 7B), as shown previously (22). Moreover, the ytpA and uptA mutations did not exhibit an additive effect on bacitracin sensitivity (Fig. 7B). The absence of additivity with UptA on sensitivity to MX-2401 suggests that YtpA has no significant role in modulating UP levels on the outside of the membrane.

Fig 7.

Epistasis of ytpA and uptA as measured by the growth of the ytpA (HB27232); uptA (HB27393) mutants alone and in combination (HB27362) on treatment with (A) 0.6 µg/mL MX-2401 and (B) 62.5 µg/mL bacitracin. N = 5 biological replicates; standard deviation in the growth of each strain has been shown by shading.

DISCUSSION

Antibiotics that interfere with peptidoglycan synthesis activate a large regulon of genes associated with σ^M-dependent promoters that collectively function to sustain cell wall synthesis even if one or more steps are inhibited (6, 18). The σ^M stress response is triggered when the membrane-localized anti-σ^M complex (YhdK/YhdL) is inactivated by a still unknown mechanism (13). Induction is amplified by a very strong positive autoregulation that leads to high level but transient expression from an autoregulatory promoter for the sigM-yhdL-yhdK operon (58). Prolonged and un-regulated induction of the σ^M regulon is lethal due, in part, to toxicity from high-level production of numerous integral membrane proteins (59). The transient induction of the σ^M stress response can counteract the action of many cell wall-acting antibiotics, and sigM mutants display heightened sensitivity to moenomycin, bacitracin, β-lactams, and other peptidoglycan synthesis inhibitors (20, 21).

To define the roles of σ^M-activated operons in protection against cell envelope stress, we and others have sought to identify σ^M target genes and their functions. This task is complex due to the large number of σ^M-activated operons (18) and the overlapping regulation with other ECF σ-dependent regulons (19–21, 60). In addition, many stress-induced operons, including those for essential genes, are expressed independent of σ^M and then further upregulated in times of stress.

The role of σ^M in protecting against peptidoglycan synthesis inhibitors can be attributed, at least in part, to the upregulation of genes for peptidoglycan synthesis. The σ^M regulon includes genes for both cytosolic steps of peptidoglycan biosynthesis (Ddl, MurB, and MurF) and membrane-associated steps, including the alternate lipid II flippase (Amj), components of the Rod complex for peptidoglycan assembly (RodA and MreBCD), and a Class A penicillin-binding protein (PBP1) (18, 61–63). In the specific case of moenomycin, σ^M regulation of the RodA transglycosylase is sufficient for resistance (61, 63). Other σ^M-activated functions include enzymes for UPP synthesis (IspD and IspF) (64), the BcrC UPP-phosphatase (18, 60, 65), and the UptA UPP flippase (22), which can all function to help sustain sufficient levels of the undecaprenyl phosphate (UP) lipid carrier (7, 64). In addition, σ^M increases synthesis of stress-induced isozymes for synthesis of lipoteichoic acid [LtaSa (66)] and wall teichoic acid [TagT (67)]. Finally, σ^M activates genes that control secondary stress responses. The latter include genes encoding the SasA (YwaC) small alarmone synthase responsible for generation of (p)ppGpp, pGpp, ppApp, and AppppA (68), the DisA synthase for cyclic-di-AMP, and Spx, a transcription factor that controls a large regulon that contributes to protection from oxidative stress (69).

Although the roles of many σ^M-regulated operons have now been defined, there are still others with poorly understood functions. Here, we demonstrate that the ytpAB operon contributes to a high level of intrinsic bacitracin resistance in B. subtilis. Bacitracin is a cyclic dodecapeptide metalloantibiotic produced by some Bacillus licheniformis and B. subtilis species (70). Bacitracin binds together with a Zn²⁺ ion to sequester the pyrophosphate moiety and first prenyl group of the UPP carrier lipid released on the outer leaflet of the membrane following the transglycosylation reaction (29). To sustain the lipid II cycle, the UPP carrier must be dephosphorylated to the UP, which is required by the MraY enzyme for the synthesis of lipid I.

Consistent with its close genetic relationship to known producer species, B. subtilis expresses a robust intrinsic resistance to bacitracin (7, 64). The first line of defense is the BceAB ABC transporter, which dissociates the bacitracin:UPP complex in a target protection mechanism (9). The BceAB transporter is specifically induced by bacitracin through the action of the BceRS two-component system (10). The BceS sensor kinase forms a complex with the BceAB proteins to allow for a flux-sensing regulation mechanism (10, 11). The second line of defense is BcrC, a σ^M-dependent UPP phosphatase (7). By dephosphorylating UPP to UP, BcrC converts the bacitracin target to a form no longer recognized by this antibiotic (65). However, UP is the target for amphomycin antibiotics, including the semi-synthetic derivative MX-2401 (57). Here, we identify the ytpAB operon as an additional contributor to intrinsic resistance to bacitracin.

YtpA is annotated as a phospholipase responsible for removal of a fatty acyl chain from PG to generate a lysophospholid species (bacilysocin) reported to have weak antibiotic activity (23). We have confirmed that B. subtilis does produce lysophospholipids, including LPG, detectable both in the supernatant and membrane fractions. However, YtpA is not required for LPG production (Fig. 3) nor are lysophospholipid levels enhanced in a strain in which ytpA is induced (Fig. S3). The enzyme, presumably an A₁ phospholipase, that is responsible for forming these species remains to be determined, and the products that result from YtpA activity are also still unclear.

The ytpAB operon additionally encodes YtpB, an enzyme that converts HPP into the monocyclic tetraprenyl-β-curcumene in the committed step in the synthesis of the C₃₅ terpenoid-designated baciterpenol A (24, 25). Baciterpenol A presumably modulates membrane properties and may thereby contribute to stress resistance, but its physiological role remains poorly characterized. A ytpB mutant strain was previously reported to be sensitized to bacitracin, an effect attributed to an increased accumulation of the HPP substrate rather than the absence of product (28). The co-regulation of YtpA and YtpB is consistent with a model in which they both function to modulate membrane properties in response to stress.

We here present evidence that ytpA affects membrane fluidity and interacts genetically with proteins that function in homeoviscous adaptation. Specifically, we observed a striking growth defect in Δbkd ΔytpA cells lacking the ability to make BCFAs and desaturated phospholipids and also missing YtpA (Fig. 2). This indicates that YtpA modulates membrane properties. Whether these effects are due to the modest impact that YtpA has on lysophospholipid synthesis (Fig. 3) or due to some other mechanism is still unknown.

YtpA also contributes to intrinsic bacitracin resistance as revealed in cells defective in other intrinsic resistance mechanisms (Fig. 6). The known mechanisms of bacitracin resistance all affect the availability of the bacitracin target UPP (BcrC) or the stability of its complex with bacitracin (BceAB) (Fig. 8). In B. subtilis, there are two UPP phosphatases, UppP and BcrC, and at least one is required for viability (54, 55). Circumstantial evidence suggests that BcrC may be the major UPP phosphatase active on the outer leaflet of the membrane. Specifically, loss of BcrC leads to a significant increase in sensitivity to bacitracin, whereas no effect is seen with a strain lacking uppP, the second UPP phosphatase. The strong additive effect on bacitracin sensitivity between bcrC (which increases UPP levels in the outer leaflet) and ytpA (Fig. 6) suggests that increased membrane fluidity may facilitate UPP flipping, possibly through a spontaneous reaction or one involving an unidentified protein partner (Fig. 8). In addition, YtpA may alter membrane properties that affect the function of membrane-anchored enzymes involved in cell wall synthesis.

Fig 8.

Model for lipid II recycling. Lipid II consists of peptidoglycan precursors bound to a UPP moiety and is synthesized in the inner leaflet of the membrane. Flippases MurJ/Amj flip it to the outer leaflet where PBPs incorporate the precursors into the peptidoglycan meshwork. Subsequently, UPP is dephosphorylated to UP and then UptA (or other unidentified proteins) flips UP back to the inner membrane. MraY initiates the incorporation of peptidoglycan precursors to reinitiate the lipid II cycle. Bacitracin binds to UPP on the outer leaflet of the membrane, inhibiting its recycling and limiting cell wall synthesis. In response to bacitracin, BceAB is upregulated to remove bacitracin from UPP, and BcrC is upregulated to dephosphorylate UPP into UP, thereby eliminating the bacitracin target. In addition, YtpA, which increases membrane fluidity by an unknown mechanism, contributes to bacitracin resistance. We speculate that YtpA may aid in flipping of UPP from the outer to the inner leaflet of the membrane.

MATERIALS AND METHODS

Growth conditions, bacterial strains, and plasmids

All strains were cultured in LB medium at 37°C and aerated on an orbital shaker at 300 rpm. Before each experiment, glycerol stocks were streaked onto fresh LB agar plates and grown overnight at 37°C. Antibiotics were used as required at the following concentrations: 100 µg/mL ampicillin, 10 µg/mL chloramphenicol, macrolide-lincosamide-streptogramin (MLS) (1 µg/mL erythromycin and 2.5 µg/mL lincomycin), and 10 µg/mL kanamycin. Bacterial strains used in this study are listed in Table 1. Bacitracin was used as the biologically active Zn salt (Zn bacitracin; Sigma #B5150), unless otherwise indicated. Deletion strains were created utilizing the BKK/BKE genomic library available at the Bacillus Genome Stock Center (BGSC) (71). Taking advantage of the natural competence of B. subtilis, all gene deletions with either a kanamycin or an erythromycin cassette were moved into the WT B. subtilis 168 strain. For transformation, B. subtilis was grown in modified competence media to stationary phase ~0.8–1.0 OD_600nm, incubated with desired DNA for 1–2 h, and plated on appropriate antibiotics. Null mutations were created by removing the antibiotic resistance cassette to create clean, in-frame deletion mutants using the pDR244 plasmid, as described (71). All gene deletions were confirmed via colony PCR using designated check primers (Table S1). Genes were overexpressed ectopically at the amyE locus using the pPL82 plasmid (72). Ectopic overexpression from the Pspac(hy) promoter was induced with 1 mM IPTG. Long-flanking homology PCR was used to construct operon deletions using primers as shown in Table S1. The Δbkd strain contains a deletion of the entire bkd operon (ptb, bcd, ipdV, bkdAA, bkdAB, and bkdB), in addition to a des deletion (39).

TABLE 1.

Strains used in this study

Strain number	Genotype	Construction	Reference
168	trpC2	Lab strain	Lab stock
HB27141	trpC2 ytpA::erm	BGSC → 168	Lab stock
HB27232	trpC2 ytpA null	pDR244 → HB27141	This study
HB27450	trpC2 amyE::pSpac(hy)-ytpA	pPL82-ytpA → 168	This study
HB27142	trpC2 des::kan	BGSC → 168	This study
HB27373	trpC2 des::kan bkd::erm (Δbkd)	gDNA bkd::erm (see primers) → HB27142	This study
HB27384	trpC2 des::kan bkd::erm amyE:: pSpac(hy)-ytpA	pPL82-ytpA → HB27373	This study
HB27482	trpC2 ytpA null des::kan bkd::erm	gDNA bkd::erm (see primers) → HB27232	This study
HB27246	trpC2 sacA::PytpAB-luxABCDE-erm	pBS3E-lux-PytpAB → 168	This study
HB27247	trpC2 sacA::PytpAB reduced-luxABCDE-erm	pBS3E-lux-PytpAB reduced → 168	This study
HB27260	trpC2 sigM::erm	BGSC → 168	Lab stock
HB27287	trpC2 sigM::erm sacA::PytpAB-luxABCDE-erm	pBS3K-lux-PytpAB → HB27260	This study
HB27272	trpC2 bcrC::erm	BGSC → 168	Lab stock
HB27277	trpC2 bcrC null	pDR244 → HB27272	This study
HB27291	trpC2 bcrC null sacA::PytpAB-luxABCDE-erm	pBS3K-lux-PytpAB → HB27277	This study
HB27271	trpC2 bceAB::kan	HB0928	Lab stock
HB27289	trpC2 bceAB::kan sacA::PytpAB-luxABCDE	pBS3E-lux-PytpAB → HB27271	This study
HB27249	trpC2 ytpB::erm	BGSC → 168	Lab stock
HB27253	trpC2 ytpB null	pDR244 → HB27249	This study
HB27407	trpC2 ytpAB::erm	See primers	This study
HB27275	trpC2 ytpA null bcrC::erm	gDNA HB27272 → HB27232	This study
HB27360	trpC2 ytpA null bcrC null	pDR244 → HB27275	This study
HB27442	trpC2 uppP::erm	BGSC → 168	Lab stock
HB27446	trpC2 uppP null	pDR244 → HB27442	This study
HB27443	trpC2 ytpA null uppP::erm	gDNA HB27442 → HB27232	This study
HB27447	trpC2 ytpA null uppP null	pDR244 → HB27443	This study
HB27344	trpC2 uptA::erm	BGSC → 168	Lab stock
HB27393	trpC2 uptA null	pDR244 → HB27344	This study
HB27351	trpC2 ytpA null uptA::erm	gDNA HB27344 → HB27232	This study
HB27362	trpC2 ytpA null uptA null	pDR244 → HB27351	This study
HB27273	trpC2 ytpA null bceAB::kan	gDNA HB27271 → HB27232	This study
HB27490	trpC2 ytpA uptA bceAB::kan	gDNA HB27271 → HB27362	This study

Fluorescence anisotropy

FA was performed as described with modification (73). Briefly, 5 mL of cells were grown in LB medium at 37°C with shaking to an OD_600nm ~1.0 with or without 1 mM IPTG induction, where applicable. Cells were harvested and centrifuged at 2,500 × g for 3 min. Cell pellets were washed twice and then resuspended in phosphate buffer (100 mM, pH 7.0) to OD_600nm 0.15. Cells were treated with DPH (Sigma) to a final concentration of 3.2 µM. An unlabeled control was also prepared. Cells were incubated in the dark in a 30°C water bath for 30 min. FA was performed with a PerkinElmer LS55 luminescence spectrometer (λ_ex = 358 nm, slit width = 10 nm; λ_em = 428 nm, slit width = 15 nm). A correction for the fluorescence intensity of unlabeled cells was performed as described (74). The data shown are average and standard deviation of three biological replicates.

Spot dilution assay

Cells were streaked onto LB agar plates and grown overnight at 37°C. From a colony, 5 mL cells were grown in LB till ~0.4 OD_600nm. Ten-fold serial dilutions were prepared, and 5 µL of the cells were plated on LB medium. Plates were allowed to air-dry for 20 min and then incubated at 27°C, 37°C, and 42°C. Images were captured after 2 days for plates incubated at 27°C and 1 day for plates incubated at 37°C and 42°C. For the cold sensitivity assay, cells were streaked onto LB agar plates supplemented with 100 µM MB (Sigma) and grown at 37°C. Five milliliters of cells were grown from an isolated colony in LB medium in the absence of MB at 37°C to ~1.0 OD_600nm. Cells were harvested and centrifuged at 2,500 × g for 5 min, and the pellets were washed with an equal volume of standard lab minimal media (15 mM (NH₄)₂SO₄; 0.8 mM MgSO₄ 7H₂O; 3.4 mM sodium citrate dihydrate; 2 mM KPO₄; 4.2 mM potassium glutamate; 40 mM morpholinepropanesulfonic acid, pH 7.4; 0.25 mM tryptophan; 5 µM FeSO₄; 5 µM MnCl₂; and 2% glucose). Ten-fold serial dilutions were performed in minimal medium, and 10 µL of cells were spotted onto minimal medium plates. Plates were allowed to air-dry for 20 min and then incubated at 22°C. Minimal media agar plates were either unsupplemented, supplemented with 100 µM MB, or supplemented with 1 mM IPTG. Spot dilutions were photographed every 24 h to monitor growth (N = 3). A representative image is shown.

LPG/LPE mass spectrometry

WT and ytpA deletion strains were grown in LB media till late-log phase and overnight for stationary phase cultures. For the strains harboring IPTG-inducible ytpA, cells were grown with or without 1 mM IPTG till late-log phase. LPG and LPE were extracted from 5 mL of cells or 1 mL of supernatant from 0.2µm filtered media. The cells were resuspended in 0.5 mL of water, and 0.5 mL of cold methanol containing 1% acetic acid was added. To 1 mL of filtered media, 1 mL of cold methanol containing 1% acetic acid was added. Samples were incubated on ice for 10 min and centrifuged at 20,000 × g for 20 min. Supernatants were dried in a speed vac concentrator and resuspended in 80% methanol containing 100 ng/mL of [d5]17-LPG.

LPG and LPE were analyzed using a Shimadzu Prominence UFLC attached to a QTrap 4500 equipped with a Turbo V ion source (Sciex). Samples were injected onto an Acquity UPLC HSS C18, 2.5 mm, 3.0 × 150 mm of column at 30°C (Waters) using a flow rate of 0.2 mL/min. Solvent A was 5 mM ammonium acetate + 1% formic acid, and Solvent B was 95% methanol + 5 mM ammonium acetate + 1% formic acid. The HPLC program was the following: starting solvent mixture of 35% A/65% B, 0 to 1 min isocratic with 65% B, 1 to 3 min linear gradient to 100% B, 3 to 30 min isocratic with 100% B, 30 to 32 min linear gradient to 65% B, and 32 to 35 min isocratic with 65% B. The QTrap 4500 was operated in the negative mode, and the ion source parameters were ion spray voltage, −4500 V; curtain gas, 30 psi; temperature, 500°C; collision gas, medium; ion source gas 1, 20 psi; ion source gas 2, 35 psi; declustering potential, −80 V; and collision energy, −30 V. The multiple reaction monitoring transitions for LPG and LPE species are listed in Table S2. [d5]17-LPG was used as the internal standard. The system was controlled by the Analyst software (Sciex) and analyzed with MultiQuant 3.0.2 software (Sciex). Peaks corresponding to individual LPG species were quantified relative to the internal standard.

PG mass spectrometry

WT and ytpA deletion strains were grown in LB media till late-log phase. Lipids were extracted from 5 mL of culture by the Bligh and Dyer method. Lipid extracts were resuspended in chloroform/methanol (1:1). PG was analyzed using a Shimadzu Prominence UFLC system attached to a QTrap 4500 equipped with a Turbo V ion source (Sciex). Samples were injected onto an Acquity UPLC BEH HILIC, 1.7 µm, 2.1 × 150 mm of column (Waters) at 45°C with a flow rate of 0.2 mL/min. Solvent A was acetonitrile, and solvent B was 15-mM ammonium formate, pH 3. The HPLC program was the following: starting solvent mixture of 96% A/4% B; 0–2 min, isocratic with 4% B; 2–20 min, linear gradient to 80% B; 20–23 min, isocratic with 80% B; 23–25 min, linear gradient to 4% B; and 25–30 min, isocratic with 4% B. The QTrap 4500 was operated in the Q1 negative mode. The ion source parameters for Q1 were as follows: ion spray voltage, −4,500 V; curtain gas, 25 psi; temperature, 350°C; ion source gas 1, 40 psi; ion source gas 2, 60 psi; and declustering potential, −40 V. The system was controlled and analyzed by the Analyst software (Sciex).

The samples were introduced to the QTrap 4500 by direct injection to perform product scans to verify the fatty acids present in a particular PG molecular species along with the positional distribution of the fatty acids. The ion source parameters for negative mode product scan were as follows: ion spray voltage, −4500 V; curtain gas, 10 psi; collision gas, medium; temperature, 270°C; ion source gas 1, 10 psi; ion source gas 2, 15 psi; declustering potential, −40 V; and collision energy, −50 V.

Luciferase reporter construction and measurement

Luciferase reporters were constructed by inserting designated promoters into the multiple cloning site of pBS3Elux, or pBS3Klux (75), and transformed into B. subtilis using natural competence as described above. For luciferase measurements, strains were grown in LB medium at 37°C to ~0.4 OD_600nm. Two microliters of culture was inoculated into 99 µL of fresh LB medium in a 96-well plate. Where applicable, cultures were treated with 0.005 µg/mL cefuroxime. The concentration of bacitracin used varied depending on the strains and has been mentioned in the figure legend. The plate was incubated at 37°C with orbital shaking in a Synergy H1 Plate Reader (BioTek Instruments, Inc.) and OD_600nm and luminescence were measured every 6 min. Relative light units for promoter activity were determined by luciferase intensity normalized for cell density (OD_600nm). The data shown are the average and standard deviation of three biological replicates.

Growth kinetics assay

From a single colony, cells were grown in 5-mL LB medium at 37°C with shaking to OD_600nm ~0.4–0.5. One microliter of culture was added to 199 µL of fresh LB medium in a 100-well Honeycomb plates (Steri). Where applicable, cells were treated with sub-lethal concentrations of bacitracin as determined by the relative bacitracin sensitivity of each strain. The OD_600nm of each well was measured at 37°C with shaking in a Bioscreen C Pro growth analyzer (Growth Curves USA, NJ, USA) every 30 min for 24 h. The data shown are the average and standard deviation from biological replicates.

Real-time PCR

Gene expression was determined by real-time PCR using primers mentioned in Table S1. Cultures were grown up to an OD_600nm of ~0.4. RNA was purified from 1.5 mL of cells using the RNeasy kit from Qiagen as per the manufacturer’s instructions. The isolated RNA was then given a DNase treatment with a Turbo DNA-free kit (Invitrogen, AM1907). Approximately 15 µg of RNA was incubated with 2 µL of DNase and 2 µL of buffer at 37°C for 15 min, followed by a 5 min incubation with the DNase-inactivating agent. The samples were then centrifuged at 8,000 rpm for 3 min, and the supernatant was collected in a fresh microcentrifuge tube. cDNA was prepared with 2 µg of the treated RNA in 20 µL total volume of reaction mix using a high-capacity cDNA reverse transcription kit from Applied Biosystems (4368814). The cDNA was further diluted 1:10 to obtain a final concentration of 10 ng/µL. Gene expression levels were measured using 10 ng of cDNA, 0.5 µM gene-specific primers, and 1 × SYBR Green Master Mix (Applied Biosystems, A25742). The gyrA gene was used as an internal control.

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jb.00015-24.

Supplemental material. jb.00015-24-s0001.pdf.

Figures S1 to S4; Tables S1 and S2.

ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.