Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Jun 26;126(3):244–251. doi: 10.1111/mmi.70092

A Single‐Nucleotide Substitution Generates a de Novo Promoter That Activates a Latent Metabolic Bypass in Escherichia coli

Qiaoqiao Guo 1,2,3, Qi Zou 1, Jilong Qin 4, Makrina Totsika 4, Yaoqin Hong 4,5,6,✉, John E Cronan 1,7,✉
PMCID: PMC13555641  PMID: 42360678

ABSTRACT

The Escherichia coli fabH gene encodes the 3‐ketoacyl‐ACP synthase III that initiates fatty acid synthesis. Deletion of the fabH gene results in severely limited fatty acid synthesis and tiny cells that are unusually sensitive to antibiotics. Genetic investigations identified the yiiD gene, now called madA, that encodes a malonyl‐ACP decarboxylase that suppresses the ∆fabH phenotype, but only when madA is multicopy. We selected vancomycin‐resistant suppressor derivatives of a ∆fabH strain. Although such chromosomal mutations were generally rare and weak, suppressor strain s1 (a single C‐T transition) restored both wild‐type growth, high‐level vancomycin resistance, and wild‐type fatty acid synthesis by creation of a new promoter within the coding sequence of a gene upstream of madA. This provides a caveat to the extensive effort to develop FabH inhibitors as antibacterial drugs.

A chromosomal mutation allows bypass of deletion of the Escherichia coli fabH gene. The bypass results from creation of a new promoter that drives increased madA transcription, resulting in increased synthesis of acetyl‐acyl carrier protein by decarboxylation of malonyl‐acyl carrier protein. The acetyl‐acyl carrier protein acts as the primer for long chain fatty acid synthesis.

graphic file with name MMI-126-244-g002.webp

1. Introduction

Inhibition of a core metabolic pathway is widely assumed to impose a high barrier to evolutionary escape. This assumption underpins many antibacterial discovery strategies. However, this view often overlooks the possibility that latent compensatory activities may already exist within the cell but remain inaccessible due to regulatory constraints. Fatty acid biosynthesis provides a well‐studied example of this problem. In E. coli , FabH initiates fatty acid synthesis by condensing acetyl‐CoA with malonyl‐ACP to form 3‐ketobutyryl‐ACP (Tsay et al. 1992), which is converted to butyryl‐ACP which primes synthesis of the acyl chains of membrane phospholipids and lipid A (White et al. 2005; Cronan Jr. and Rock 2008).

The role of FabH in initiation argued that fabH was an essential gene in E. coli as was first reported (Lai and Cronan 2003). However, others found that ΔfabH strains remained viable, albeit slow growing and of tiny cell size (Baba et al. 2006; Yao et al. 2012). Clone‐bank complementation of a ∆fabH strain identified the yiiD gene as a multicopy (p)ppGpp‐regulated suppressor of fabH null mutations (Sanyal et al. 2019), for which the inability to construct a ∆yiiD ∆fabH strain indicated synthetic lethality (Sanyal et al. 2019). The conflicting results in fabH essentiality were therefore due to different strain backgrounds. The original E. coli ∆fabH deletion was constructed in a strain deficient in synthesis of the stress “alarmone” nucleotides (p)ppGpp (guanosine tetraphosphate and pentaphosphate) whereas the viable ∆fabH strains were proficient in (p)ppGpp synthesis.

Later the yiiD encoded protein (renamed MadA) was shown to decarboxylate malonyl‐ACP to acetyl‐ACP that primed fatty acid synthesis (Whaley et al. 2021). Although the biochemical bypass was known, an unresolved important question was the requirement for multicopy expression of MadA for ∆fabH bypass. In wild‐type strains, loss of FabH sensitizes E. coli to several otherwise ineffective antibiotics (Yao et al. 2012; Hong et al. 2022). Wild‐type E. coli cells are resistant to vancomycin because the intact outer membrane blocks access of the antibiotic to its peptidoglycan target, whereas E. coli fabH null mutants are susceptible due to a defective outer membrane (Yao et al. 2012; Hong et al. 2022). FabH has long been an antimicrobial target (documented by 130 PubMed entries). Hence, ΔfabH suppressors that restored growth and vancomycin resistance would provide a straightforward escape route from FabH inhibition.

A rare spontaneous vancomycin‐resistant ΔfabH‐suppressor, strain s1, containing a single cytosine‐to‐thymine (C → T) transition in the yihY gene upstream of madA created a highly active promoter for madA expression. Elevated MadA expression restored fatty acid synthesis, improved membrane function, and restored vancomycin resistance to the ΔfabH mutant. Thus, a single cis‐regulatory rewiring unlocked a latent metabolic bypass.

2. Results

2.1. A Suppressor Mutation That Restores Vancomycin Resistance to E. coli ΔfabH Strains

The s1 suppressor was isolated by plating a culture of the ∆fabH strain on LB agar plates containing 100 μg/mL vancomycin. The s1 suppressor strain restored wild‐type MG1655 growth rates in LB medium, reestablished essentially full vancomycin resistance and remedied the damaged cell membrane of ∆fabH strains (Figure 1A,B, Figure S2). Moreover, suppressor s1 restored the wild‐type level of fatty acid synthesis to the ∆fabH strain (Figure 1C) shown by [1‐14C] acetate incorporation into membrane phospholipid acyl chains (Figure 1C). Whole‐genome sequencing of strain s1 showed two single‐nucleotide polymorphisms (SNPs), a thymine‐to‐cytosine (T → C) transition in the yihT gene and a C → T transition within the yihY gene, a silent mutation that retained the YihY residue 270 valine (Figure 1D). The vancomycin selection gave other suppressors, but these strains grew as poorly as the ∆fabH strain and were not further analyzed.

FIGURE 1.

FIGURE 1

Open in a new tab

Identification of a suppressor isolate (s1) of MG1655 ∆fabH mutant. (A) Growth measurement of the E. coli ∆fabH strain and its s1 suppressor in LB medium with and without vancomycin. (B) The suppressor mutation of isolate s1 restored vancomycin resistance to the ∆fabH strain (the numbers above the figure panel indicate the dilution factor from OD600 1.0 overnight cultures). (C) The s1 suppressor mutation reversed the defective de novo fatty acid synthesis in the ∆fabH strain. The numbers above the lanes showed the values for phospholipid acyl chain unsaturation and the radioactive label incorporation relative to the value (100) for the MG1655 strain. (D) Single‐nucleotide polymorphisms (SNPs) identified in suppressor isolate s1. The yihX, yihY, dtd, madA operon is reported to be transcribed from a promoter located immediately upstream of the yihX coding sequence (Mendoza‐Vargas et al. 2009). Note that the lines between the lanes of all TLC plates of this report were made prior to sample loading by scribing the coating of the plate with a rake‐like tool. This prevents bleeding of spots between lanes and simplifies quantitation.

2.2. Suppression of the ΔfabH Null Mutation Is due to a Point Mutation Within the yihY Gene That Created a Promoter

Transcriptomics analysis identified genes that were significantly differentially expressed between the suppressor s1 and parental ∆fabH strains (Figure S1). Among these, madA and dtd were upregulated > 20‐fold in the suppressor s1 (Figure 2A) relative to the parental ∆fabH strain (Figure S1, Figure S2). The E. coli Regulon Database reports that yihY is the second gene in a yihX, yihY, dtd and madA operon transcribed from the yihX promoter (Figure 1D). The yihX and dtd genes, respectively, encode a phosphatase and an aminoacyl‐tRNA editing enzyme. The yihY gene function is unknown, although it encodes a nonessential inner membrane protein (Daley et al. 2005; Baba et al. 2006). Although now corrected (Ezraty et al. 2005), yihY was reported to encode a ribonuclease. Despite their diverse functions, translation of the yihX‐madA operon transcript appears tightly coupled because all four open reading frames overlap.

FIGURE 2.

FIGURE 2

Open in a new tab

A single C → T transition mutation in yihY gives a de novo promoter that drives high‐level expression of the dtd‐madA region. (A) Transcriptomic analysis identified strong upregulation of dtd and madA in the suppressor isolate s1 compared to the parental ∆fabH strain. The yihX–yihY–dtd–madA operon structure is shown schematically. (B) A lacZ transcriptional fusion to the native yihX promoter showed that the yihY C → T mutation did not significantly alter transcription from the promoter upstream of yihX in the MG1655, ∆fabH, and s1 strains. LacZ transcriptional fusions containing the mutant yihY sequence confer a ~30‐fold increase in transcriptional activity of the downstream dtd–madA region in the three strains (C) strain MG1655, (D) strain ΔfabH, and (E) strain s1 whereas the wild‐type sequence had only basal activity. The fusions were constructed using the low copy vector, pHSG575 (Takeshita et al. 1987). AU denotes arbitrary units of β‐galactosidase activity.

Transcriptional fusions to lacZ (with glucose catabolite repression of the chromosomal lac operon) confirmed that the s1 mutation had no effect on the yihX promoter (Figure 2B), consistent with the transcriptomic data (Figure 2A). The altered yihY sequence conferred an ~30‐fold increase in transcription of the downstream dtd and madA genes, whereas the wild‐type yihY sequence had only basal activity (Figure 2C–E). Thus, the suppressor mutation created a de novo promoter that drove increased expression of dtd and madA (Figure 2E).

The second SNP of strain s1, a T → C transition allele in yihT, was tested by plasmid expression of both the wild type and mutant alleles in the ΔfabH strain. Both plasmids failed to restore vancomycin resistance (Figure S3), indicating that the yihT mutation had no role in ∆fabH suppression.

2.3. Mapping the Novel Promoter Within the yihY Gene

The transcription initiation site was determined by 5′‐RACE in parallel with analysis of the yihY gene regions of the MG1655 and ∆fabH strains harvested at mid‐exponential phase. The transcription initiation site of suppressor s1 strain mapped to a site immediately downstream of the promoter created by the C → T transition mutation (Figure 3A). Sequence analysis indicated that the C → T transition generated a near‐canonical σ 70 promoter by creating a properly positioned −10 element upstream of the transcription start site, 17 bp downstream of a high quality −35 element (Figure 3B) (LaFleur et al. 2022).

FIGURE 3.

FIGURE 3

Open in a new tab

De novo promoter formation increases dtd–madA transcription in the s1 suppressor strain. (A) Mapping of the transcriptional start site by 5′‐RACE. Representative sequencing chromatogram from analysis of the s1 strain is shown with the mapped transcription start site immediately downstream of the RACE adaptor sequence. The sharp transition from adaptor to genomic sequence defines a discrete transcription initiation site. (B) Promoter architecture underlying suppressor activation. Sequence alignment of MG1655 and the s1 suppressor reveals a C → T transition that generates a functional −10 element (highlighted in red) in the s1 strain. The position of the mapped transcription start site (+1) is indicated. Predicted σ 70 promoter elements (−35 and −10 regions) are shown, demonstrating that the mutation creates promoter architecture consistent with σ 70 recognition and appropriate spacing between elements, supporting formation of a functional de novo promoter. The numbers above the sequences give the nucleotide positions relative to E. coli dtd gene ATG initiation codon.

2.4. MadA Is Responsible for Bypass of fabH Deletion and Vancomycin Resistance in Strain s1

The above data demonstrated formation of a new promoter in strain s1 that greatly increased expression of the dtd‐madA region (Figures 2 and 3). Expression plasmids encoding either dtd or madA were constructed and transformed into the ∆fabH strain. As expected, overexpression of MadA but not Dtd in the ∆fabH strain resulted in recovery of vancomycin resistance (Figure S4A). Expression of the Pseudomonas stutzeri MadB (Figure S5) in the ∆fabH strain also resulted in vancomycin resistance (Figure S4B) and restored wild‐type levels of de novo fatty acid synthesis (Figure S4C). MadB is essentially the C‐terminal malonyl‐ACP decarboxylase domain of MadA (Whaley et al. 2021). The P. stutzeri MadB is 83% identical to the P. putida MadB (McNaught et al. 2023). The MadA N‐terminal domain affects cyclic‐AMP metabolism (Ciemniecki et al. 2026) and thus the MadB data indicate that the MadA N‐domain has no role in ∆fabH suppression. High levels of MadA expression were toxic to growth and inhibit fatty acid synthesis (Figures S6 and S7), arguing that the s1 suppressor promoter had to be sufficiently strong to bypass loss of FabH but not so strong to become toxic.

The increased MadA expression of strain s1 restored two other indications of outer membrane function to the ∆fabH strain: resistance to sodium dodecyl sulfate and swimming motility (Figure S8).

2.5. The Function of Malonyl‐ACP Decarboxylation Is to Prime Fatty Acid Synthesis

To provide a further in vivo test that MadA expression provides an alternate initiation primer and vancomycin resistance, we provided an alternate ACP primer. A V. harveyi AasS acyl‐ACP synthetase plasmid (Jiang et al. 2010) was transformed into the ∆fabH strain with octanoate (C8:0) supplementation to give octanoyl‐ACP, a precursor of both unsaturated and saturated acyl chains (Cronan 2024). The AasS‐octanoate bypass restored vancomycin resistance but only when both AasS expression and octanoate supplementation were present (Figure 4).

FIGURE 4.

FIGURE 4

Open in a new tab

Synthesis of an octanoyl‐ACP primer by Vibrio harveyi acyl‐ACP synthase (AasS) restores wild type vancomycin resistance and growth to the ∆fabH strain. (A) Dependence on AasS induction of [1‐14C]octanoate incorporation into phospholipid acyl chains. (B) Restoration of wild type vancomycin resistance requires both AasS induction and octanoate supplementation (the numbers above each figure panel indicate the dilution factor from the overnight cultures). Note that the trace growth of the ∆fabH/p‐aasS strain can be attributed to the small cytosolic octanoate pool present in E. coli (Hermes and Cronan 2009).

3. Discussion

FabH was considered essential in E. coli because it uniquely initiated fatty acid synthesis (Tsay et al. 1992; Cronan Jr. and Rock 2008). However, MadA showed FabH was not unique (Sanyal et al. 2019; Whaley et al. 2021). MadA decarboxylates malonyl‐ACP to acetyl‐ACP to bypass loss of FabH (Whaley et al. 2021). Wild type MadA provides only limited fatty acid flux. E. coli K‐12 studies showed that ΔfabH cells are markedly smaller, indicating that fatty acid availability constrains membrane biogenesis and thereby cell size (Yao et al. 2012). Our finding that fatty acid synthesis is reduced to ~30% of wild‐type levels fits well (Figure 1C) with ΔfabH cells being ~30% of the wild‐type size (Yao et al. 2012; Whaley et al. 2021). This supports the lipid‐centric framework of Levin and co‐workers (Vadia et al. 2017) in which that fatty acid flux scales closely with envelope expansion capacity. However, our prior work in extra‐intestinal pathogenic E. coli (ExPEC) strains presents an important exception (Hong et al. 2022). Despite severe fatty acid limitation, both CFT073 and EC958 ΔfabH cells retain near wild‐type size, albeit with slower division, suggesting that these ExPEC strains trade growth rate for maintenance of cell size (Hong et al. 2022).

FabH has long been pursued as an antimicrobial target (Verma et al. 2025) which reflects the common assumption in antibacterial discovery that disruption of a core biosynthetic step will be difficult for bacteria to bypass. The s1 suppressor demonstrates upregulation of madA to restore wild‐type growth to the ΔfabH strain showing that bypass of loss of a major biosynthetic step can arise through a minimal cis‐regulatory change. The s1 mutation in yihY converted a sequence into a σ 70–10 element. The newly introduced T resulted in a TACTCT −10 element that contains both bases (the TA and 3′T) that are key in interactions with σ 70 (Feklistov and Darst 2011; Sevostyanova et al. 2007). Moreover, free‐energy modeling predicted reduced local duplex stability at this site (Figure S9).

Excessive madA expression inhibited phospholipid acyl chain synthesis in wild‐type E. coli (Figures S7 and S8). To approximate the level of MadA expression causing toxicity, we expressed MadA from two different promoters in strain MG1655. The weaker arabinose promoter gave a ~3‐fold decrease in phospholipid acyl chain synthesis but only a modest decrease in growth (Figure S7), whereas the more powerful T5 promoter almost completely blocked both growth and acyl chain synthesis (Figure S8). Elevated MadA activity would decrease the malonyl‐ACP pool needed for acyl chain elongation. Expression of an enzyme (BioC) that methylates the free carboxyl of malonyl‐ACP is similarly toxic (Lin and Cronan 2012). Avoidance of toxicity may explain why madA transcription requires alarmone activation.

Inhibition of core metabolic pathways may be especially vulnerable to evolutionary bypass when latent compensatory functions already exist that need only transcriptional activation. Indeed, de novo promoter formation is highly accessible in E. coli . Experimental evolution has shown that ~60% of forty ~100 bp random sequences are only a single base change away from promoter activity whereas ~10% are already active (Yona et al. 2018). Second, the E. coli K‐12 genome retains ~200 unannotated putative coding sequences (Goodall et al. 2018). These findings suggest that inhibition of core metabolic pathways is more vulnerable to evolutionary bypass than generally assumed.

4. Materials and Methods

4.1. Materials

Sigma‐Aldrich and New England Biolabs were the sources of most reagents. Sodium[1‐14C] acetate (57.0 mCi/mmol) and sodium[1‐14C] octanoate (55.0 mCi/mmol) were supplied by Moravek and American Radiolabeled Chemicals, respectively.

4.2. Bacterial Strains, Plasmids and Growth Media

Bacterial strains and plasmids are listed in Table S1 and primers in Table S2. Cultures were incubated at 37°C or 30°C in LB medium (tryptone, 10 g/L; yeast extract, 5 g/L; NaCl, 10 g/L) or RB (LB lacking yeast extract). Antibiotics were added at (in μg/L) sodium ampicillin, 100; chloramphenicol, 30; vancomycin, 50. Growth was determined by OD600. with a Bioscreen C.

4.3. Genome Sequencing

Genomic DNA using the Qiagen DNeasy blood and tissue kit. Whole genome sequencing was performed by BGI (Hong Kong) using DNBseq platforms generating > 800 bp read data. Sequencing data were aligned to the E. coli MG1655 reference genome.

4.4. Construction of Expression Plasmids

The plasmids were constructed as described in the legends of the relevant Supporting Information: figures.

4.5. Analysis of 14C Labeled Phospholipid Acyl Chains

De novo synthesis was assayed as described previously (Zou et al. 2022). Cultures (5 mL) from dilution of overnight cultures into fresh medium at OD600 0.05 and adding 1 mCi/L sodium [1‐14C]acetate was followed by incubation for 3 h at 37°C. The same protocol except that 0.1 mM cold octanoic acid and 0.1 mCi/L sodium[1‐14C]octanoate was used. The [14C]‐labeled fatty acid methyl esters were synthesized and analyzed as given (Zou et al. 2022).

4.6. LacZ Transcriptional Fusion Constructions and β‐Galactosidase Assay

The effects of the s1 suppressor mutation yihX promoter output were tested. A 500 bp upstream region including the first 36 bp of yihX (−500 to +36 relative to the yihX start codon) was amplified from MG1655 DNA using primers P‐yihX F and 5 ‐yihX R. Similarly, the lacZ coding sequence was obtained using primers lacZ F and lacZ R. These fragments were inserted into pHSG575 to give the yihX–lacZ fusion plasmid. Similar manipulations gave the dtd‐lacZ fusion. β‐Galactosidase was assayed in triplicate (Bi et al. 2014; Zou et al. 2022). LB containing glucose was used to repress expression of chromosomal lacZ.

4.7. Mapping of the yihY Transcription Initiation Site

The transcription initiation site was determined using the Sangon Biotech 5′‐RACE Kit Briefly, E. coli MG1655, ∆fabH, and its suppressor isolate s1 were cultured in LB medium at 37°C to OD600 of 0.6. The cells were harvested and total RNA extracted using the Tiangen RNA extraction kit. The concentration and purity of the RNA were by A260/A280 RNA integrity was confirmed by gel demonstration of discrete 23S and 16S ribosomal RNA bands. A specific reverse transcription primer RT1 for dtd‐madA was utilized to synthesize cDNA from the mRNA template by the Reverse Transcriptase Mix (RNase H‐), followed by addition of a homopolymeric (dC) tail of 10–15 residues to the 3′ end of the cDNA by terminal transferase (TdT) to allow pairing with the 5′ adaptor primer (GCTGTCAACGATACGCTACGTAACGGCATGACAGTGGGGGGGGGGGGG) containing an oligonucleotide anchor sequence. The first‐round PCR amplification was performed using this adaptor primer and a specific outer primer R1of the dtd‐madA region with the single‐stranded cDNA as the template. A second (nested) PCR was performed using the 5′ RACE outer primer (GCTGTCAACGATACGCTACGTAAC), which contained a partial adapter sequence and R2 another specific inner primer of the dtd‐madA cluster to amplify the 5′‐terminal cDNA fragment of the dtd gene. The transcription initiation site wase determined by sequencing of the final PCR products by Sangon Biotech.

Author Contributions

Qiaoqiao Guo: investigation, methodology, validation, writing – review and editing, writing – original draft. Jilong Qin: investigation, formal analysis. Yaoqin Hong: conceptualization, investigation, methodology, validation, writing – original draft, writing – review and editing. Qi Zou: investigation, methodology, validation, writing – original draft, writing – review and editing. John E. Cronan: conceptualization, investigation, funding acquisition, writing – original draft, methodology, validation, writing – review and editing, project administration, supervision. Makrina Totsika: investigation, project administration, funding acquisition.

Funding

This work was supported by the National Institute of Allergy and Infectious Diseases, AI15650. Queensland University of Technology, 2017HIG0119. James Cook University.

Ethics Statement

The authors have nothing to report.

Conflicts of Interest

Makrina Totsika is an employee of the GSK group of companies. This research was conducted independently of any commercial interests. The remaining authors declare no conflicts of interest.

Supporting information

Figure S1: Gene ontology (GO) enrichment analysis was performed with genes that exhibit a differential expression log2 ratio ≥ 3.0 using ShinyGO (Ge et al. 2019).

Figure S2: Differentially expressed genes with ultra‐high absolute confidence between three replicates of ∆fabH and suppressor s1. RNA was extracted and purified using the Qiagen RNeasy kits for RNA purification, in accordance with manufacturer's protocol. Transcriptomic sequencing was performed by BGI (Hong Kong) using DNBse platforms. Sequencing data were mapped to the E. coli MG1655 reference genome. (A) The differentially expressed genes between ∆fabH and suppressor s1 (hits with absolute confidence at −1000 or 1000 are highlighted by yellow transparent strips at the top of the volcano plot). Principal component analysis showed clear separation of conditions along PC1 (98.6% variance), indicating the primary source of transcriptional variation is driven by the experimental condition. Replicates clustered tightly within each group with only minor variation observed along PC2 (0.52%), suggesting limited technical or biological noise. (B) Sankey plot illustrating the identity of genes highlighted in the volcano plot and categorized into relevant pathways (genes in bold were upregulated and vice versa). The madA gene is highlighted with an orange oval. Many of the transcriptional changes can be attributed to restoration of a functional outer membrane in the s1 suppressor strain. The Sankey plot was visualized with Omicshare tools (Mu et al. 2024).

Figure S3: Plasmid expression of the wild‐type and mutant yihT failed to reverse vancomycin sensitivity of the ∆fabH strain. The C‐T transition mutant allele of yihT is denoted as yihT*. Overnight cultures were adjusted to OD600 and serially diluted 10‐fold, with 5 μL spotted onto LB agar containing 50 μM vancomycin and grown overnight at 37°C.

Figure S4: Overexpression of MadA or expression of MadB restored growth in the presence of vancomycin and wild type fatty acid synthesis. (A) Vancomycin resistance of MG1655 ∆fabH was restored upon overexpression of MadA or the P. stutzeri MadB homologue. (B) Expression of the P. stutzeri MadB restored vancomycin resistance to the ∆fabH strain. In panels A and B, the numbers above each figure panel indicate the dilution factor from the overnight cultures. (C) [1‐14C]acetate incorporation to measure de novo fatty acid biosynthesis in the ∆fabH strain with either overexpression of MadA or expression of MadB. In panel C the numbers above the lanes give the phospholipid unsaturation and the radioactive label incorporation values relative to the value (100) for the wild type MG1655 strain. The dtd and madA expression plasmids for E. coli dtd and madA genes, were constructed from MG1655 strain genomic DNA using primer sets dtd NdeI F and dtd HindIII R, and madA NdeI F and madA HindIII R, respectively, inserted into the NdeI and HindIII sites of vector pBAD24M. The dtd‐madA fragment was amplified using primer set dtd NdeI F and madA HindIII R and cloned as above. The yihT coding region was amplified from the genomic DNAs from MG1655 or the s1 suppressor strains by primer set yihT F1 and yihT R1 and then ligated with linearized vector pHSG575 by Gibson assembly.

Figure S5: The Pseudomonas stutzeri ATCC 17588 = LMG 11199 madB gene and MadB protein (UniProt A0A0H3YTL2_STUST). Note that P. stuteri is now called Stutzerimonas stutzeri in some data bases.

Figure S6: Effects of paraBAD promoter driven MadA overproduction on fatty acid synthesis and growth of strain MG1655 (denoted as WT). (A) [1‐14C]acetate incorporation into phospholipid acyl chains as referenced in Materials and Methods. The cultures were started at an OD600 of 0.05 for 1 h followed by addition of 1 mCi/L [1‐14C]acetate with or without supplementation with 0.2% arabinose for another 3 h. (B) Growth of MG1655 and MG1655 carrying the p‐madA plasmid in the presence or absence of 0.2% arabinose. The medium was RB, a relatively poor medium such that arabinose consumption stimulated growth. Note that growth of the p‐madA culture overexpressing MadA that contained arabinose ceased growth after about 4 h. The expression plasmid the madA genes, was amplified from E. coli MG1655 strain genomic DNA using primer sets madA NdeI F and madA HindIII R, respectively, and then inserted into the NdeI and HindIII sites of vector pBAD24M.

Figure S7: Effects of pT5 promoter driven MadA overproduction on growth of strain MG1655 (denoted as WT) and fatty acyl chain synthesis. (A) Growth on LB plate (essentially the same result was given on RB plates). (B) Growth in RB medium, IPTG 1 mM was added as given. (C) [14C]acetate incorporation into membrane phospholipid acyl chains. The cultures were started at 37°C in RB medium at an OD600 of 0.05 and incubated to log phase (OD600 of about 0.3–0.4) followed by addition of 1 mCi/L of [1‐14C]acetate with or without supplementation with 1 mM IPTG for another 3 h. The radiolabeled fatty acyl chains were analyzed as described in Materials and Methods. For madA expression from the phage T5 promoter, the madA coding region was applified from MG1655 DNA using primer set Ec madA F2 and Ec madA R2 and then ligated to vector pQE2 vector.

Figure S8: Other properties of the s1 and ∆fabH/p‐aasS strains. The top and middle panels show relief of the sodium dodecyl sulfate (SDS) sensitivity of ∆fabH strains by the s1 mutation and the octanoate‐dependent production of acyl‐ACP synthase, respectively. The bottom panel shows the s1 mutation restores swimming motility to the ∆fabH strain assayed in 0.3% LB agar after spotting 10 μL of each culture and incubation for ~40 h. The strains were: top left MG1655, top right ∆fabH and lower right, s1 (Bp7 was an early name for strain s1).

Figure S9: Predicted changes in promoter free energy (ΔGtotal) associated with the s1 suppressor mutation. Predictions were generated using a model‐based promoter calculator that integrates sequence‐dependent contributions to transcription initiation, including σ 70–DNA interactions and local DNA energetics (LaFleur et al. 2022). The s1 suppressor exhibits a local decrease in ΔGtotal relative to ΔfabH, consistent with increased promoter favorability at the mutated site. Transcriptional prediction and ∆G energy/position‐weight modellings were performed as previously described (LaFleur et al. 2022).

Table S1: Strains and Plasmids used in this work.

Table S2: Sequences of the PCR primers used.

MMI-126-244-s001.docx (4.7MB, docx)

Acknowledgements

This work was supported by National Institutes of Health grant AI15650 from the National Institute of Allergy and Infectious Diseases to John E. Cronan, and Early Career Researcher Award Scheme from the Faculty of Health of Queensland University of Technology to Yaoqin Hong and Grant Submission Support Scheme from the James Cook University awarded to Yaoqin Hong.

Contributor Information

Yaoqin Hong, Email: yaoqin.hong@jcu.edu.au.

John E. Cronan, Email: jecronan@illinois.edu.

Data Availability Statement

The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Zhang et al. 2025) in National Genomics Data Center (Members and Partners 2026), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: CRA043388) that are publicly accessible at https://ngdc.cncb.ac.cn/gsa. Additional data that support the findings of this study are available on request from the corresponding author.

References

  1. Baba, T. , Ara T., Hasegawa M., et al. 2006. “Construction of Escherichia coli K‐12 In‐Frame, Single‐Gene Knockout Mutants: The Keio Collection.” Molecular Systems Biology 2: 0008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bi, H. , Zhu L., Wang H., and Cronan J. E.. 2014. “Inefficient Translation Renders the Enterococcus faecalis fabK Enoyl‐Acyl Carrier Protein Reductase Phenotypically Cryptic.” Journal of Bacteriology 196: 170–179. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ciemniecki, J. A. , Will J. L., and Escalante‐Semerena J. C.. 2026. “In Salmonella typhimurium , YiiD Modulates cAMP Levels in Lag Phase During Growth on Succinate.” Molecular Microbiology. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Cronan, J. E. 2024. “Unsaturated Fatty Acid Synthesis in Bacteria: Mechanisms and Regulation of Canonical and Remarkably Noncanonical Pathways.” Biochimie 218: 137–151. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cronan, J. E., Jr. , and Rock C. O.. 2008. “Biosynthesis of Membrane Lipids.” EcoSal Plus 3: 10‐1128. [DOI] [PubMed] [Google Scholar]
  6. Daley, D. O. , Rapp M., Granseth E., Melen K., Drew D., and von Heijne G.. 2005. “Global Topology Analysis of the Escherichia coli Inner Membrane Proteome.” Science 308: 1321–1323. [DOI] [PubMed] [Google Scholar]
  7. Ezraty, B. , Dahlgren B., and Deutscher M. P.. 2005. “The RNase Z Homologue Encoded by Escherichia coli elaC Gene Is RNase BN.” Journal of Biological Chemistry 280: 16542–16545. [DOI] [PubMed] [Google Scholar]
  8. Feklistov, A. , and Darst S. A.. 2011. “Structural Basis for Promoter −10 Element Recognition by the Bacterial RNA Polymerase σ Subunit.” Cell 147: 1257–1269. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Ge, S. X. , Jung D., and Yao R.. 2019. “ShinyGO: a Graphical Gene‐Set Enrichment Tool for Animals and Plants.” Bioinformatics 36: 2628–2629. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Goodall, E. C. A. , Robinson A., Johnston I. G., et al. 2018. “The Essential Genome of Escherichia coli K‐12.” MBio 9: e02096‐17. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Hermes, F. A. , and Cronan J. E.. 2009. “Scavenging of Cytosolic Octanoic Acid by Mutant LplA Lipoate Ligases Allows Growth of Escherichia coli Strains Lacking the LipB Octanoyltransferase of Lipoic Acid Synthesis.” Journal of Bacteriology 191: 6796–6803. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Hong, Y. , Qin J., Verderosa A. D., et al. 2022. “Loss of Beta‐Ketoacyl Acyl Carrier Protein Synthase III Activity Restores Multidrug‐Resistant Escherichia coli Sensitivity to Previously Ineffective Antibiotics.” mSphere 7: e0011722. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Jiang, Y. , Morgan‐Kiss R. M., Campbell J. W., Chan C. H., and Cronan J. E.. 2010. “Expression of Vibrio harveyi Acyl‐ACP Synthetase Allows Efficient Entry of Exogenous Fatty Acids Into the Escherichia coli Fatty Acid and Lipid A Synthetic Pathways.” Biochemistry 49: 718–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. LaFleur, T. L. , Hossain A., and Salis H. M.. 2022. “Automated Model‐Predictive Design of Synthetic Promoters to Control Transcriptional Profiles in Bacteria.” Nature Communications 13: 5159. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Lai, C. Y. , and Cronan J. E.. 2003. “Beta‐Ketoacyl‐Acyl Carrier Protein Synthase III (FabH) is Essential for Bacterial Fatty Acid Synthesis.” Journal of Biological Chemistry 278: 51494–51503. [DOI] [PubMed] [Google Scholar]
  16. Lin, S. , and Cronan J. E.. 2012. “The BioC O‐Methyltransferase Catalyzes Methyl Esterification of Malonyl‐Acyl Carrier Protein, an Essential Step in Biotin Synthesis.” Journal of Biological Chemistry 287: 37010–37020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. McNaught, K. J. , Kuatsjah E., Zahn M., et al. 2023. “Initiation of Fatty Acid Biosynthesis in Pseudomonas putida KT2440.” Metabolic Engineering 76: 193–203. [DOI] [PubMed] [Google Scholar]
  18. Members, C. N. , and Partners . 2026. “Database Resources of the National Genomics Data Center, China National Center for Bioinformation in 2026.” Nucleic Acids Research 54: D28–D47. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Mendoza‐Vargas, A. , Olvera L., Olvera M., et al. 2009. “Genome‐Wide Identification of Transcription Start Sites, Promoters and Transcription Factor Binding Sites in E. coli .” PLoS One 4: e7526. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Mu, H. , Chen J., Huang W., et al., 2024. “OmicShare Tools: A zero‐Code Interactive Online Platform for Biological Data Analysis and Visualization.” Imeta 3: e228. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Sanyal, R. , Singh V., and Harinarayanan R.. 2019. “A Novel Gene Contributing to the Initiation of Fatty Acid Biosynthesis in Escherichia coli .” Journal of Bacteriology 201: 10‐1128. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Sevostyanova, A. , Feklistov A., Barinova N., et al. 2007. “Specific Recognition of the −10 Promoter Element by the Free RNA Polymerase Sigma Subunit.” Journal of Biological Chemistry 282: 22033–22039. [DOI] [PubMed] [Google Scholar]
  23. Takeshita, S. , Sato M., Toba M., Masahashi W., and Hashimoto‐Gotoh T.. 1987. “High‐Copy‐Number and Low‐Copy‐Number Plasmid Vectors for lacZ Alpha‐Complementation and Chloramphenicol‐ or Kanamycin‐Resistance Selection.” Gene 61: 63–74. [DOI] [PubMed] [Google Scholar]
  24. Tsay, J. T. , Oh W., Larson T. J., Jackowski S., and Rock C. O.. 1992. “Isolation and Characterization of the Beta‐Ketoacyl‐Acyl Carrier Protein Synthase III Gene (fabH) From Escherichia coli K‐12.” Journal of Biological Chemistry 267: 6807–6814. [PubMed] [Google Scholar]
  25. Vadia, S. , Tse J. L., Lucena R., et al. 2017. “Fatty Acid Availability Sets Cell Envelope Capacity and Dictates Microbial Cell Size.” Current Biology 27: 1757–1767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Verma, S. , Gupta P., Lal S., et al. 2025. “Recent Developments of the Beta‐Ketoacyl‐Acyl Carrier Protein Synthase III (FabH) Inhibitors: Emerging Indication as Antibacterial Agents.” European Journal of Medicinal Chemistry 298: 117995. [DOI] [PubMed] [Google Scholar]
  27. Whaley, S. G. , Radka C. D., Subramanian C., Frank M. W., and Rock C. O.. 2021. “Malonyl‐Acyl Carrier Protein Decarboxylase Activity Promotes Fatty Acid and Cell Envelope Biosynthesis in Proteobacteria.” Journal of Biological Chemistry 297: 101434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. White, S. W. , Zheng J., and Zhang Y. M.. 2005. “The Structural Biology of Type II Fatty Acid Biosynthesis.” Annual Review of Biochemistry 74: 791–831. [DOI] [PubMed] [Google Scholar]
  29. Yao, Z. , Davis R. M., Kishony R., Kahne D., and Ruiz N.. 2012. “Regulation of Cell Size in Response to Nutrient Availability by Fatty Acid Biosynthesis in Escherichia coli .” Proceedings of the National Academy of Sciences of the United States of America 109: E2561–E2568. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Yona, A. H. , Alm E. J., and Gore J.. 2018. “Random Sequences Rapidly Evolve Into de Novo Promoters.” Nature Communications 9: 1530. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Zhang, S. , Chen X., Jin E., et al. 2025. “The GSA Family in 2025: A Broadened Sharing Platform for Multi‐Omics and Multimodal Data.” Genomics, Proteomics & Bioinformatics 23: qzaf072. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Zou, Q. , Dong H., Zhu L., and Cronan J. E.. 2022. “The Enterococcus faecalis FabT Transcription Factor Regulates Fatty Acid Biosynthesis in Response to Exogeneous Fatty Acids.” Frontiers in Microbiology 13: 877582. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1: Gene ontology (GO) enrichment analysis was performed with genes that exhibit a differential expression log2 ratio ≥ 3.0 using ShinyGO (Ge et al. 2019).

Figure S2: Differentially expressed genes with ultra‐high absolute confidence between three replicates of ∆fabH and suppressor s1. RNA was extracted and purified using the Qiagen RNeasy kits for RNA purification, in accordance with manufacturer's protocol. Transcriptomic sequencing was performed by BGI (Hong Kong) using DNBse platforms. Sequencing data were mapped to the E. coli MG1655 reference genome. (A) The differentially expressed genes between ∆fabH and suppressor s1 (hits with absolute confidence at −1000 or 1000 are highlighted by yellow transparent strips at the top of the volcano plot). Principal component analysis showed clear separation of conditions along PC1 (98.6% variance), indicating the primary source of transcriptional variation is driven by the experimental condition. Replicates clustered tightly within each group with only minor variation observed along PC2 (0.52%), suggesting limited technical or biological noise. (B) Sankey plot illustrating the identity of genes highlighted in the volcano plot and categorized into relevant pathways (genes in bold were upregulated and vice versa). The madA gene is highlighted with an orange oval. Many of the transcriptional changes can be attributed to restoration of a functional outer membrane in the s1 suppressor strain. The Sankey plot was visualized with Omicshare tools (Mu et al. 2024).

Figure S3: Plasmid expression of the wild‐type and mutant yihT failed to reverse vancomycin sensitivity of the ∆fabH strain. The C‐T transition mutant allele of yihT is denoted as yihT*. Overnight cultures were adjusted to OD600 and serially diluted 10‐fold, with 5 μL spotted onto LB agar containing 50 μM vancomycin and grown overnight at 37°C.

Figure S4: Overexpression of MadA or expression of MadB restored growth in the presence of vancomycin and wild type fatty acid synthesis. (A) Vancomycin resistance of MG1655 ∆fabH was restored upon overexpression of MadA or the P. stutzeri MadB homologue. (B) Expression of the P. stutzeri MadB restored vancomycin resistance to the ∆fabH strain. In panels A and B, the numbers above each figure panel indicate the dilution factor from the overnight cultures. (C) [1‐14C]acetate incorporation to measure de novo fatty acid biosynthesis in the ∆fabH strain with either overexpression of MadA or expression of MadB. In panel C the numbers above the lanes give the phospholipid unsaturation and the radioactive label incorporation values relative to the value (100) for the wild type MG1655 strain. The dtd and madA expression plasmids for E. coli dtd and madA genes, were constructed from MG1655 strain genomic DNA using primer sets dtd NdeI F and dtd HindIII R, and madA NdeI F and madA HindIII R, respectively, inserted into the NdeI and HindIII sites of vector pBAD24M. The dtd‐madA fragment was amplified using primer set dtd NdeI F and madA HindIII R and cloned as above. The yihT coding region was amplified from the genomic DNAs from MG1655 or the s1 suppressor strains by primer set yihT F1 and yihT R1 and then ligated with linearized vector pHSG575 by Gibson assembly.

Figure S5: The Pseudomonas stutzeri ATCC 17588 = LMG 11199 madB gene and MadB protein (UniProt A0A0H3YTL2_STUST). Note that P. stuteri is now called Stutzerimonas stutzeri in some data bases.

Figure S6: Effects of paraBAD promoter driven MadA overproduction on fatty acid synthesis and growth of strain MG1655 (denoted as WT). (A) [1‐14C]acetate incorporation into phospholipid acyl chains as referenced in Materials and Methods. The cultures were started at an OD600 of 0.05 for 1 h followed by addition of 1 mCi/L [1‐14C]acetate with or without supplementation with 0.2% arabinose for another 3 h. (B) Growth of MG1655 and MG1655 carrying the p‐madA plasmid in the presence or absence of 0.2% arabinose. The medium was RB, a relatively poor medium such that arabinose consumption stimulated growth. Note that growth of the p‐madA culture overexpressing MadA that contained arabinose ceased growth after about 4 h. The expression plasmid the madA genes, was amplified from E. coli MG1655 strain genomic DNA using primer sets madA NdeI F and madA HindIII R, respectively, and then inserted into the NdeI and HindIII sites of vector pBAD24M.

Figure S7: Effects of pT5 promoter driven MadA overproduction on growth of strain MG1655 (denoted as WT) and fatty acyl chain synthesis. (A) Growth on LB plate (essentially the same result was given on RB plates). (B) Growth in RB medium, IPTG 1 mM was added as given. (C) [14C]acetate incorporation into membrane phospholipid acyl chains. The cultures were started at 37°C in RB medium at an OD600 of 0.05 and incubated to log phase (OD600 of about 0.3–0.4) followed by addition of 1 mCi/L of [1‐14C]acetate with or without supplementation with 1 mM IPTG for another 3 h. The radiolabeled fatty acyl chains were analyzed as described in Materials and Methods. For madA expression from the phage T5 promoter, the madA coding region was applified from MG1655 DNA using primer set Ec madA F2 and Ec madA R2 and then ligated to vector pQE2 vector.

Figure S8: Other properties of the s1 and ∆fabH/p‐aasS strains. The top and middle panels show relief of the sodium dodecyl sulfate (SDS) sensitivity of ∆fabH strains by the s1 mutation and the octanoate‐dependent production of acyl‐ACP synthase, respectively. The bottom panel shows the s1 mutation restores swimming motility to the ∆fabH strain assayed in 0.3% LB agar after spotting 10 μL of each culture and incubation for ~40 h. The strains were: top left MG1655, top right ∆fabH and lower right, s1 (Bp7 was an early name for strain s1).

Figure S9: Predicted changes in promoter free energy (ΔGtotal) associated with the s1 suppressor mutation. Predictions were generated using a model‐based promoter calculator that integrates sequence‐dependent contributions to transcription initiation, including σ 70–DNA interactions and local DNA energetics (LaFleur et al. 2022). The s1 suppressor exhibits a local decrease in ΔGtotal relative to ΔfabH, consistent with increased promoter favorability at the mutated site. Transcriptional prediction and ∆G energy/position‐weight modellings were performed as previously described (LaFleur et al. 2022).

Table S1: Strains and Plasmids used in this work.

Table S2: Sequences of the PCR primers used.

MMI-126-244-s001.docx (4.7MB, docx)

Data Availability Statement

The raw sequence data reported in this paper have been deposited in the Genome Sequence Archive (Zhang et al. 2025) in National Genomics Data Center (Members and Partners 2026), China National Center for Bioinformation/Beijing Institute of Genomics, Chinese Academy of Sciences (GSA: CRA043388) that are publicly accessible at https://ngdc.cncb.ac.cn/gsa. Additional data that support the findings of this study are available on request from the corresponding author.

Articles from Molecular Microbiology are provided here courtesy of Wiley

RESOURCES