Salmonella Typhimurium alters galactitol metabolism under ciprofloxacin treatment to balance resistance and virulence

ABSTRACT

Ciprofloxacin-resistant Salmonella Typhimurium (S. Typhimurium) causes a significant health burden worldwide. A wealth of studies has been published on the contributions of different mechanisms to ciprofloxacin resistance in Salmonella spp. But we still lack a deep understanding of the physiological responses and genetic changes that underlie ciprofloxacin exposure. This study aims to know how phenotypic and genotypic characteristics are impacted by ciprofloxacin exposure, from ciprofloxacin-susceptible to ciprofloxacin-resistant strains in vitro. Here, we investigated the multistep evolution of resistance in replicate populations of S. Typhimurium during 24 days of continuously increasing ciprofloxacin exposure and assessed how ciprofloxacin impacts physiology and genetics. Numerous studies have demonstrated that RamA is a global transcriptional regulator that prominently perturbs the transcriptional landscape of S. Typhimurium, resulting in a ciprofloxacin-resistant phenotype appearing first; the quinolone resistance-determining region mutation site can only be detected later. Comparing the microbial physiological changes and RNA sequencing (RNA-Seq) results of ancestral and selectable mutant strains, the selectable mutant strains had some fitness costs, such as decreased virulence, an increase of biofilm-forming ability, a change of “collateral” sensitivity to other drugs, and inability to utilize galactitol. Importantly, in the ciprofloxacin induced, RamA directly binds and activates the gatR gene responsible for the utilization of galactitol, but RamA deletion strains could not activate gatR. The elevated levels of RamA, which inhibit the galactitol metabolic pathway through the activation of gatR, can lead to a reduction in the growth rate, adhesion, and colonization resistance of S. Typhimurium. This finding is supported by studies conducted in M9 medium as well as in vivo infection models.

IMPORTANCE

Treatment of antibiotic resistance can significantly benefit from a deeper understanding of the interactions between drugs and genetics. The physiological responses and genetic mechanisms in antibiotic-exposed bacteria are not well understood. Traditional resistance studies, often retrospective, fail to capture the entire resistance development process and typically exhibit unpredictable dynamics. To explore how clinical isolates of S. Typhimurium respond to ciprofloxacin, we analyzed their adaptive responses. We found that S. Typhimurium RamA-mediated regulation disrupts microbial metabolism under ciprofloxacin exposure, affecting genes in the galactitol metabolic pathways. This disruption facilitates adaptive responses to drug therapy and enhances the efficiency of intracellular survival. A more comprehensive and integrated understanding of these physiological and genetic changes is crucial for improving treatment outcomes.

INTRODUCTION

Salmonellae is an important foodborne and zoonotic pathogen that has emerged as a global cause of severe gastroenteritis, bacteremia, and systemic infection, infecting a wide range of host species, particularly elderly and immunocompromised patients with high percentages (1, 2). In addition, categorized by the World Health Organization (WHO) as a high-priority pathogen for which new treatments and promote research are urgently needed (3), fluoroquinolone (FQs)-resistant Salmonella enterica serovar Typhimurium (S. Typhimurium) causes a significant health burden worldwide (4). FQs are the drug of choice for the treatment of salmonellosis, particularly ciprofloxacin, and are one of the most widely used antibiotics worldwide because of its safety, efficacy, relatively low cost, and promotion of growth in large-scale animal production. It is usually used as a first-line treatment option for life-threatening Salmonella infections in clinical practice (5, 6). With the evolving worldwide crisis of antibiotic resistance to S. Typhimurium, new knowledge on antibiotic resistance mechanisms and the underlying mechanisms is of critical importance (7).

FQ resistance is a major global challenge facing modern medicine. In the current study, the mechanisms of FQ resistance including chromosomal mutations in the target enzymes, DNA gyrase I (GyrA and GyrB) and topoisomerase IV (ParC and ParE), modification of the drug target, and downregulation of outer membrane porin (OmpR) expression reduced membrane permeability, coupled with increases in active multidrug efflux pump (TolC-AcrAB), as well as acquisition of transmissible plasmid-mediated quinolone resistance (PMQR) genes. A brief introduction to the known ciprofloxacin resistance mechanisms in Enterobacteriaceae is outlined in reference (8). It has been suggested that exposure to ciprofloxacin and mutation of chromosomally located primary target genes gyrA, gyrB, parC, and parE represents the first step in the development of quinolone resistance (9). For many Gram-negative bacteria such as Escherichia coli, S. Typhimurium (9), Pasteurella multocida (10), Riemerella anatipestifer (11, 12), and Klebsiella pneumoniae (13), the primary target mutation site in the quinolone resistance-determining region (QRDR) is serine 83 of GyrA. The epidemiological survey showed that highly resistant bacteria typically carry a combination of mutations within gyrA (GyrA 83 or 87) and parC (ParC 80) (14, 15). The chromosomal multidrug efflux pump acrAB-TolC, the main efflux pump described in Enterobacteriaceae, is capable of actively removing FQs and is known to play an important role in the development of high-level FQ resistance (16). Increased levels of acrAB have been proposed as a biomarker of multidrug resistance and suggested to be the first step to facilitate high-level FQ resistance development following the acquisition of target site mutations (17, 18). Several regulators (RamA, SoxS, and MarA) have been reported to influence acrAB operon expression in Salmonella (19, 20). When RamA was overexpressed in S. enterica serovar Typhimurium or E. coli, these strains exhibited decreased susceptibility to multiple drugs (21). It has been confirmed that RamA can bind to the upstream promoter region of acrAB and tolC and increase the expression level of efflux (22). These local and global regulators that control drug resistance mechanisms are cytosolic and/or membranous sensors that respond accordingly to ensure bacterial biological processes, including metabolism and growth.

In addition, as mentioned above, transcriptomic studies have shown that pathways related to drug transport, stress response, DNA repair, and phage induction are upregulated and promote intracellular survival during bacteria exposure to ciprofloxacin (23–25). However, there is still a lack of deep understanding regarding the physiological responses and genetic changes that occur during ciprofloxacin exposure. Specifically, more studies are needed on the dynamics of how ciprofloxacin-susceptible bacterial strains develop resistance to this antibiotic. Previous resistance studies have been conducted retrospectively, which lag in time, do not monitor the entire process of resistance development, and often have unpredictable dynamics. However, little is known about how ciprofloxacin exposure affects the phenotypic and genotypic characteristics of ciprofloxacin-resistant strains in vitro.

Although the regulatory mechanisms of ciprofloxacin resistance are well defined, our understanding of the evolution and adaptation of S. Typhimurium evolves and adapts during continuous stimulation induction exposure to ciprofloxacin which is limited. In particular, there is a lack of evidence regarding clinically resistant or therapeutic strains in continuous studies. In this work, aiming to understand how phenotypic and genotypic characteristics are impacted by ciprofloxacin successive induction, we studied one ciprofloxacin-susceptible S. Typhimurium strain (ancestral strain) under sustained perturbation with ciprofloxacin. Using experimental evolution, ciprofloxacin-resistant S. Typhimurium strains (selectable mutant strains, MIC ≥ 4 µg/mL ciprofloxacin). Focusing on physiological and genetic variations, we found that before the emergence of the QRDR mutation, RamA is a global transcriptional regulator mediating ciprofloxacin resistance that prominently perturbs the transcriptional landscape of S. Typhimurium and that selectable mutant strains undergo distinct fitness costs and transcriptional changes within a long time frame, impacting bacterial microbial galactitol metabolism.

RESULTS

Evolution of ciprofloxacin-resistant Salmonella Typhimurium

Using experimental evolution, the primary aim of our study was to acquire ciprofloxacin-resistant S. Typhimurium strains and to dynamically detect the development of drug resistance. We serially passaged sensitive S. Typhimurium parent strains (MIC ≤ 0.0625 mg/L) in twofold gradients of antibiotic concentration daily for 24 days, dynamically selecting for high levels of resistance. At the end of each growth cycle (16 hours), the culture with the highest ciprofloxacin concentration, permitting bacterial growth [optical density at 600 nm (OD₆₀₀) > 0.4], was propagated into a fresh antibiotic gradient (1:200 dilution). In each cycle, the MICs of the S. Typhimurium were measured. Ciprofloxacin resistance increased up to 64-fold (MIC ≥ 4.0 mg/L) after the 24-day cycle. The resulting categories were designated susceptible (S), intermediate (I), and resistant (R) (Fig. 1A). We then selected strains at different time points for a detailed study of gene expression and carriage of specific mutations. We found that the MIC values of the strains fluctuated less during the first 13 days (S) of induction. The “I” MIC values fluctuate from day 13 to 15; especially on day 18 when the MIC (≥4.0 mg/L) is close to the threshold of resistance criteria, no drug-resistant phenotypes or mutations were detected during this process (Fig. 1A; for details, see File Table S3 in the supplemental material). However, the “R” MIC values steadily increased from day 19 to day 24, with the emergence of drug-resistant phenotypes and mutations (GyrA S83F or D87N and ParC S80R). We also compared the expression levels of RamA (a transcriptional activator, which influences the expression of genes including those for antimicrobial resistance and the acrAB-tolC efflux pump. Studying its behavior under various conditions can elucidate regulatory mechanisms involved in these processes) and acrA (efflux pump) at different time points by real-time fluorescence quantitative PCR (RT-qPCR). The results showed that ciprofloxacin induction was accompanied by the emergence of selectable mutant strains with varying degrees of upregulation of RamA and acrA compared with the ancestral strain: pre-induction (day 1), with ancestral strain as control (Fig. 1B), and no difference between ancestral and selectable mutant strains post-induction (day 9) (Fig. 1C). However, we found differences in the expression levels of RamA (>1.5-fold) and acrA (1.5-fold) between the ancestral and selectable mutant strains on post-induction day 15 (Fig. 1D) and significant differences on post-induction days 18 (Fig. 1E) and 24 (Fig. 1F), with RamA (approximately 3.8-fold) being more significantly upregulated. Therefore, in S. Typhimurium, increased ciprofloxacin resistance is caused primarily by overexpression of ramA and the AcrAB-TolC efflux pumps; a number of previous studies have supported this conclusion. Interestingly, the resistance phenotype first appeared, and then, mutations in the target gene sites in the QRDR were detected (Fig. 1).

Fig 1.

Monitoring of the development of ciprofloxacin resistance. (A) The ordinate indicates the MIC of ciprofloxacin to the strain, and the abscissa indicates the date of continuous stimulation. S (Sensitive) indicated that there was no drug resistance and no point mutation within 13 days; I (Intermediate) indicated that there was drug resistance but no point mutation during the period of 14–18 days; R (resistance) showed that there was drug resistance and point mutations of gryA S83F/D87N and parC S80R after 19 days. (B-F) During continuous stimulation with ciprofloxacin for 24 consecutive days, the expression levels of RamA and acrA in ancestral and selectable mutant strains were monitored at 1, 9, 15, 18, and 24 days. Statistical analysis was performed using Student’s t-test. Data are presented as mean ± SEM. Mean values from at least three independent cultures are shown with standard deviations (ns: P > 0.05, *P < 0.05, **P < 0.01, and ***P < 0.001).

Ciprofloxacin induces physiological changes in S. Typhimurium

Beceiro et al. believed that antibiotics brought great pressure on the growth of microorganisms, which forced them to damage toxicity and bacterial adaptability in the process of continuous adaptation to these pressures (26). To investigate the potential fitness implications of resistance evolution, we tested microbial physiology such as growth, biofilm formation, virulence, collateral drug resistance, and metabolism of ancestral and selectable mutant strains. First, the growth curve of the strain was measured in LB medium, and it was found that a significant retardation of growth was observed in the early stages of strain induction by ciprofloxacin (the growth rate exhibited a delay of at least 5 hours compared with the control); however, this growth lag disappeared as the resistant phenotype gradually stabilized; especially when comparing ancestral and selectable mutant strains (24th day, MIC ≥ 4 µg/mL ciprofloxacin), there were no significant growth differences (Fig. 2A). It appears that the acquisition of resistance did not affect the growth of the strain in the same environment. However, further comparisons revealed that according to classical crystal violet staining, mutant strains were significantly more capable of forming biofilms than ancestral strains (Fig. 2B) and biofilm formation is often positively correlated with drug resistance. The selectable mutant strain showed a consistent and significant decline in invasion (Fig. 2C) and adhesion (Fig. 2D) abilities in the HeLa and Caco cell lines. In response to ciprofloxacin stimulation, the strain acquired a resistant phenotype while reducing its virulence, possibly by downregulating key virulence genes.

Fig 2.

Microbial physiological detection. (A) Determination of growth curve. Growth curves of ancestral strains (black) and selectable mutant strains (blue) in LB medium. The OD₆₀₀ of the strains were detected every hour and continuously for 15 hours. (B) Detection of biofilm formation ability. Detection of biofilm of S. Typhimurium by crystal violet staining. The top of the picture shows the crystal violet staining results of the biofilm. (C, D) Bacterial invasion and adhesion. HeLa and Caco cells were used to detect the invasion and adhesion of ancestral and selectable mutant strains. The changes of invasion and adhesion were judged by antibiotic treatment and plate counting. (E) Collateral drug sensitivity. The collateral resistance of ancestral and selectable mutant strains to 17 antibiotics was detected by measuring MIC values in 96-well plates. (F) Detection of metabolic substrate spectrum of bacteria. Biochemical vials were used to detect the changes of metabolic capacity of ancestral and selectable mutant strains. Observe the metabolism at 24 hours according to the instructions (purple indicates no metabolism, yellow indicates metabolism). The above experiments were repeated at least three times. Statistical analysis was performed using Student’s t-test. Data are presented as mean ± SEM. Mean values from at least three independent cultures are shown, with standard deviations (***P＜0.001).

It is possible that resistance to one antibiotic may result in “collateral” sensitivity to other antibiotics and the existence of “collateral” sensitivity also makes it possible to design new drug delivery schemes (27). To investigate the effects of ciprofloxacin on S. Typhimurium, we tested the sensitivity of the 17 antibiotics. Some, but not all, drugs are clinically relevant for treating Salmonella infections. The MIC of ciprofloxacin was also tested as a control. Our results indicate that collateral effects are heterogeneous, including sensitivity or resistance (Fig. 2E). Noteworthy increases were as follows: AMP (8-fold), TET (4-fold), CHL (16-fold), FLO (4-fold), STR (2-fold), and APR (2-fold), and decreases were as follows: SMD (4-fold), SPT (16-fold), and ROX (8-fold). Symptoms include cross-resistance to at least one drug within the same class and an eightfold increase in resistance to enrofloxacin (ENR). Several previous studies have indicated and these results suggest that under selective pressure, Salmonella can respond to increased AcrAB-TolC efflux pumps or other regulatory genes (primarily RamA), allowing for the evolution of multi-drug resistance (MDR) (20). Surprisingly, ciprofloxacin induction led to a decline in drug resistance among the three drugs, particularly SPT. These results provide valuable clues for studying the mechanism of ciprofloxacin resistance in S. Typhimurium.

The metabolic ability of this strain is closely associated with its virulence and viability. It is necessary to explore changes in the metabolic capacity of strains to explain their biological phenotype. The Salmonella Biochemical Identification Kit (HuanKai Biology, HK1002, CHINA) (for details, see File Fig. S1 in the supplemental material) and the ability of the selected isolates to ferment nine different carbohydrates (glucose, fructose, lactose, sucrose, maltose, mannitol, galactitol, ribitol, and xylose) were examined. By comparing the colorimetric cards provided by the manufacturers, we found differential metabolites between ancestral and selectable mutant strains. In contrast, the results of the biochemical identification plate showed that the selectable mutant strain had differences in the metabolism of galactitol and urea (Fig. S1) and further verification showed that the selectable mutant strain could not normally use galactitol (Fig. 2F). We speculate that some mysterious mechanisms in drug-resistant strains directly or indirectly lead to downregulation of the gene encoding galactitol metabolism.

Ciprofloxacin induces S. Typhimurium-specific transcriptional responses

Physiological changes were observed across the bacterial populations following ciprofloxacin exposure after the appearance of a resistant phenotype, as demonstrated above. We aimed to determine whether these biological characteristics distinguish bacteria based on their different transcriptional properties. To gain insight into the transcriptional profile of the ancestral to selectable mutant strains during ciprofloxacin exposure, we performed RNA sequencing analysis 1 day vs. 24 days (Fig. 3A). Generally, gene expression analysis revealed that 169 S. Typhimurium genes were differentially regulated (Fig. 3B; Table S4), with 32 genes upregulated and 137 genes downregulated (FDR < 0.01, P < 0.01) in all selectable mutant strains compared with ancestral strains. Differentially regulated genes were assigned to functional categories using the Clusters of Orthologous Groups of proteins (COG) database, and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis was performed based on the KEGG database. Interestingly, the majority of genes (53 genes) were assigned to the category ‘‘cellular processes and signaling’’ and specifically to the subcategories ‘‘intracellular trafficking, secretion, and vesicular transport’’ (Fig. 3C; Table S5). The KEGG pathway enrichment results showed that the DEGs were enriched in 30 KEGG pathways; the most significantly enriched pathways were the “Bacterial secretion system,” among which 22 KEGG pathways were related to metabolism (Table S6). This suggests that the virulence and metabolic capacity of S. Typhimurium may have changed significantly during the development of ciprofloxacin resistance (Fig. 3B and D). Table S7 shows the top 30 upregulated and downregulated genes in S. Typhimurium. In particular, significant changes were observed in the expression of MDR-related genes ramA (upregulated 6.005-fold), tolC (upregulated 2.350-fold), acrA (upregulated 2.246-fold), and acrB (upregulated 2.419-fold), whereas ompF (downregulated 2.967-fold) and siiABCDEF (downregulated from 2.825- to 7.389-fold) encode membrane proteins (23); only STY1854 (biofilm development protein YmgB/AriR, upregulated 2.309-fold) was detected in the genes associated with the biofilm. Furthermore, the transcription levels of the type III secretion system and invasion-related genes were dramatically downregulated (from 1.939- to 6.321-fold) (Fig. 3B and D) (28). Interestingly, there were also several metabolism- and biosynthesis-associated genes that were commonly changed, especially genes for galactitol utilization gatABC (downregulated from 2.118- to 2.616-fold), STY3437 (downregulated 2.218-fold), and gatR (upregulated 2.316-fold) (29), which is consistent with the biochemical test results shown in Fig. 3B and D. Interestingly, the global regulator RamA and the negative regulator gatR of galactitol metabolism were significantly upregulated (Fig. 3D). Overall, this suggests that ciprofloxacin activates high expression of RamA locus and regulates not only multidrug efflux pumps and accessory genes but also genes involved in galactitol metabolic pathways.

Fig 3.

Comparative RNA-seq analysis of ancestral and selectable mutant strains. (A) Laboratory evolution experiment scheme. Three separate ancestral strains were used in laboratory evolution experiments. The ancestral strains were stimulated with ciprofloxacin at a concentration lower than the lethal concentration for 24 consecutive days to obtain drug-resistant strains. This experiment was repeated twice. (B) Volcano map shows up-regulated and down-regulated genes in S.Typhimurium. The ordinate: log-fold-change in expression; the Abscissa: negative logarithm10 transformed adjusted P values. Genes up-regulated in mutant strain (fold-change ≥ 0.5 and adjusted P < 0.05) are shown in blue and downregulated shown in red (0 ≤ fold-change ≤ 0.5 and adjusted P ≤ 0.05). (C) Differentially up- and down-regulated mutant genes were classified into the COG functional categories and subcategories. Some genes were assigned to more than one COG functional category. (D) Genes with significant differences and known functions were selected and displayed as heat maps. Use the per million mapped reads (RPKM) values of selected genes in all samples to draw a heat map. The log-fold-change and adjusted P values for all the genes are visualized as bar plot and one-column heatmap, respectively.

RT-qPCR was used to corroborate the selected values from significantly different pathways. Five important differentially expressed gene enrichment pathways in the transcriptome sequencing analysis were selected for RT-qPCR analysis (Bio-Rad CFX software). The results showed that the gene difference times showed a high degree of concordance between data from the two methodologies (for details, see Fig. S2 in the supplemental material).

Acquisition of ciprofloxacin resistance by Salmonella Typhimurium affects the galactitol metabolic

As we could not cover all of the induced pathways in this study, we focused on the galactitol metabolism of S. Typhimurium to illustrate how the acquisition of ciprofloxacin resistance by S. Typhimurium affects galactitol metabolism. We observed that the selectable mutant strains exposed to ciprofloxacin differed from the ancestral strain in galactitol biochemical assays and transcriptomic data. This suggests that ciprofloxacin may affect the ability of S. Typhimurium to utilize galactitol. Therefore, we tested this hypothesis by assessing the growth status of the strain in medium containing galactitol as the sole carbon source. For example, the phenotype of ancestral and selectable mutant strains was verified using M9 agar plates with 1% galactitol as an additional carbon source and supplemented with ciprofloxacin at different concentrations (2–6 μg/mL); CFU was significantly affected by ciprofloxacin concentration (Fig. 4A). In growth studies, the ancestral strain grew normally, and selectable mutant strains were defective (P < 0.001) when using 0.2% galactitol as a carbon source, exhibiting virtually no growth (Fig. 4B). The results showed that with the stimulation of the mutant strain by increasing the concentration of antibiotics, the utilization capacity of galactitol decreased gradually, which suggested that there might be a quantitative correlation between them. In mice fed adequate amounts of galactitol, colonization loads for S. Typhimurium in feces and cecum were compared (Fig. 4C). In mouse infection assays, ancestral and mutant bacterial strains, including STΔramA and STΔgatR deletion mutants, were assessed following treatments with PBS and galactitol. Our findings revealed no significant differences in fecal bacterial loads among the four bacterial strains in the PBS-treated group (P > 0.05) (Fig. 4E). However, cecal colonization showed a marked reduction in bacterial loads for mutant strains compared with ancestral strains (P < 0.01). Notably, the STΔramA mutants exhibited lower bacterial loads than the STΔgatR mutants (P < 0.05) (Fig. 4E). Under galactitol treatment, significant disparities were observed across the strains. Mutant strains displayed reduced bacterial counts in both feces (P < 0.05) and cecum (P < 0.001) compared with ancestral strains (Fig. 4D). Specifically, the fecal (P < 0.05) and cecal (P < 0.01) bacterial loads in STΔgatR mutants approached those of ancestral strains, surpassing the STΔramA mutants (Fig. 4D). Meanwhile, the growth of the ancestral strains, the STΔramA mutants, and the STΔgatR mutants in LB medium and M9 medium (0.2% galactitol) was also evaluated (for details, see Fig. S3 in the supplemental material). The results indicated that mutant bacterial strains unable to metabolize galactitol exhibited significantly reduced colonization in mice. The absence of ramA, a global regulatory factor, further decreased colonization efficiency due to its integral role in strain activity. Conversely, removal of gatR eliminated an inhibitory effect on galactitol metabolism. This led to enhanced utilization of exogenous galactitol, increasing both the number and colonization capabilities of the strain, restoring it to levels comparable to the ancestral strain. According to recent studies, E. coli enhances colonization resistance against S. Typhimurium by competitively metabolizing galactitol (30). In conclusion, this demonstrates that galactitol utilization contributes to the S. Typhimurium shift in gut colonization and blocks the metabolism of galactitol as a mechanism for ciprofloxacin treatment of S. Typhimurium.

Fig 4.

Effect of non-utilization of galactitol on S. Typhimurium (A and B) Effect of ciprofloxacin concentration on the utilization of galactitol by selectable mutant strain. The growth status of the selectable mutant strain in M9 medium containing 1% galactitol and different concentrations of ciprofloxacin were indicated (STR01, STR02, and STR03 represent the strains in the medium containing 2 µg/mL, 4 µg/mL, and 6 µg/mL ciprofloxacin, respectively). (C, D, and E) Mouse experimental scheme. BALB/c mice inoculated with galactitol or sterile PBS as control (n = 7; n = 7) for 10 days, and then, the experimental groups were orally infected with ancestral and selectable mutant bacterial strains, including STΔramA and STΔgatR deletion mutants. Cecal contents were collected 1 day after infection, and mice were sacrificed 2 days after infection. The colonization of ancestral and selectable mutant strains in mice was determined by monitoring the bacterial load of feces and cecal contents. (F) Detection of galactitol operon gene expression in selectable mutant strains stimulated by different concentrations of ciprofloxacin by RT-qPCR; the expression levels of RamA and gatR increase with rising concentrations of CIP. The above experiments were repeated at least three times. Statistical analysis was performed using Student’s t-test. Data are presented as mean ± SEM. Mean values from at least three independent cultures are shown, with standard deviations.

RT-qPCR was performed on genes in the galactitol operon of selectable mutant strains stimulated with different concentrations of ciprofloxacin. To verify the quantitative relationship in the above results, we detected the expression levels of RamA and galactitol metabolism gene clusters, including STY3435-STY3445 (Fig. 4F). Through data analysis, we found that, compared with the ancestral strain, nine genes (gatY-gatD) in the galactitol metabolism pathway were significantly downregulated in selectable mutant strains and the negative regulatory gene gatR in this pathway was significantly upregulated (Fig. 4F). In addition, the transcription levels of gatR and RamA increased synchronously with the increase in drug resistance of the selectable mutant strains and the transcription levels of gatR and RamA increased synchronously (Fig. 4F). Our data demonstrate that ciprofloxacin induces differential changes in key gene expression between mutant strains and their parental counterparts. This variation in gene expression correlates with the dosage of ciprofloxacin, illustrating a dose-dependent response in these genetic alterations. Therefore, we speculate that during the development of drug resistance, RamA is highly expressed as a global regulatory factor and then regulates the high expression of gatR, which negatively regulates genes related to the metabolism of galactitol (29), resulting in a decrease in the ability of selectable mutant strains to metabolize galactitol.

Activation of the gatR gene by the RamA regulator is mediated by ciprofloxacin

As we all know, active RamA plays an important role in conferring ciprofloxacin resistance in S. Typhimurium. To investigate whether RamA regulates galactitol operon genes directly or indirectly (Fig. 5A), we used an electrophoretic mobility shift assay (EMSA) to verify whether it can specifically bind to the promoter of the galactitol operon (PgatY, PgatZ, and PgatR) (Fig. 5A). SDS-PAGE (Fig. 5B) and Western blot (WB) (Fig. 5C) results of purified RamA-GST are shown. We employed qPCR to compare the ramA deletion strains and mutant strains. The results, which are now presented in Fig. 5D, clearly demonstrate that deletion of the ramA gene in these strains results in the abolition of gatR expression activation, which is typically induced by ciprofloxacin. Fragments and DNA-protein complexes were separated by gel electrophoresis at 8°C. At increasing concentrations of the RamA protein, slow-migrating bands were observed for the gatR promoter (Fig. 5E). These results indicated that RamA binds specifically to the promoter region of gatR in vitro. RamA did not bind to the gatZ (Fig. 5F) and gatY (Fig. 5G) promoters under the same experimental conditions. This observation substantiates the hypothesis that ramA is crucial for the upregulation of gatR in response to ciprofloxacin exposure. It provides new insights into the specific genetic pathways that are influenced by this antibiotic.

Fig 5.

Binding of RamA to the gatY, gatZ, and gatR promoter regions. (A) Galactitol operon of S. Typhimurium. The galactitol operon consists of 10 genes (gatY-gatR). Putative promoters are indicated by arrows, and putative terminator regions are shown as hairpins. Probe sequences of the three promoters, gatY, gatZ, and gatR are shown below. The gatR is bound by and regulated by RamA, marking the −10 and −35 motifs in the promoter region. (B, C) RamA protein purification. SDS-PAGE and WB results of the purified RamA are shown. (D) qPCR to compare the ramA deletion strains and selectable mutant strains. The expression levels of RamA and gatR increase with rising concentrations of CIP. In the absence of RamA, CIP cannot activate RamA, resulting in no change in the expression level of gatR. (E–G) Results of EMSA using the gatY, gatZ, and gatR promoter regions and RamA fusion protein. The DNA fragments are indicated at the bottom of the gels. Each well contained 10-nM DNA fragments. Lanes 1 to 5 show the EMSA results using 0, 5, 10, 20, and 40 pmol of RamA protein, respectively.

DISCUSSION

Since the 1980s, ciprofloxacin has been extensively employed in clinical medicine, and it has currently emerged as the predominant antibacterial agent worldwide (31, 32). Despite the extensive research conducted on S. Typhimurium multi-drug resistance, which is regarded as a highly significant bacterial model organism, the precise conditions and mechanisms underlying its involvement in ciprofloxacin resistance remain elusive. Through conducting an evolutionary experiment on S. Typhimurium induced by ciprofloxacin, we have demonstrated the physiological and genetic changes that occur within the same strain upon acquiring a ciprofloxacin-resistant phenotype. Our findings not only confirm previously established phenomena, such as the significant role of RamA in resisting ciprofloxacin invasion and the main mutation sites being GyrA S83F or D87N and ParC S80R, but also reveal novel and intriguing phenomena, such as the alteration of strain metabolism (6, 15, 22). Significantly, these data provide a clearer understanding of the combined influence of plastic physiological adaptations and microevolutionary genetic changes on treatment outcomes. Consequently, we have determined that treating S. Typhimurium with ciprofloxacin is a double-edged sword, offering both benefits and risks.

Of note, through an analysis of the available data, it is not difficult to summarize the studies relating to ciprofloxacin and Salmonella, which involve (i) differential gene expression by RamA—RamA was the main factor controlling the susceptibility of S. Typhimurium to ciprofloxacin by activating MdtK as well as increasing the expression level of AcrAB and decreased expression of porin protein OmpF (20, 24, 33); (ii) inactivation of RamA caused changes in the response of Salmonella to at least 100 compounds, including but not limited to susceptibility to antimicrobials and toxic compounds, better in utilization of carbon and phosphorus sources (24); (iii) overexpression of RamA increases mutation rate and influences drug resistance phenotype (20); and (iv)ciprofloxacin-treated S. Typhimurium induces specific morphological (elongated morphology) and transcriptional signatures associated with bacterial DNA repair, SOS response, and intracellular survival (25). In our evolutionary experiment, we confirmed that both RamA and AcrB activities and QRDR (GyrA S83F or D87N and ParC S80R) mutations contributed to ciprofloxacin resistance. In this process, we found an interesting phenomenon: the target gene site mutation is slightly different from the time node of drug resistance, which often appears in the drug resistance phenotype first and then detects the target gene mutation in QRDR. Therefore, we speculate that the emergence of drug-resistant phenotypes is often a coordinated effect of multiple resistance mechanisms and that mutations in drug targets are the final step in confirming drug resistance. It has been suggested that the use of QRDR mutations to determine ciprofloxacin resistance in previous epidemiological surveillance of drug-resistant strains is time lagged, incomplete, and unreasonable.

However, after acquiring the resistant phenotype of ciprofloxacin, the selectable mutant strain underwent a number of physiological changes compared with the ancestral strain. First, compared with the ancestral strain, the mutant strain grew slowly and formed small colonies during the initial induced phase of ciprofloxacin (data not shown), which disappeared as the resistant phenotype (ciprofloxacin MIC ≥ 4.0 mg/L) of the mutant strain. Survival and growth at high antibiotic concentrations are the result of multiple adjustments in many metabolic pathways, and this needs to be studied further (7). However, bacteria become more resistant to antibiotics due to biofilm formation, which protects them against exogenous stressors. The selectable mutant strain can form biofilms, which can significantly increase antibiotic resistance, except ciprofloxacin, and exhibited qualitatively different profiles of collateral sensitivity to drugs (resistance to AMP, TET, CHL, FLO, STR, and APR). The three antibiotics showed a significant decline (SMD, SPT, and ROX). A study by Maltas et al. found that collateral effects are pervasive but difficult to predict because independent populations selected by the same drug can exhibit qualitatively different profiles of collateral sensitivity as well as markedly different fitness costs (27).

Additionally, the selectable mutant strain showed a significant decrease in invasion and adhesion of HeLa and Caco cells. The main reason for this is the downregulation of the expression of a large number of genes related to virulence, particularly the type III secretion system. Furthermore, the biochemical indices of each group were tested and compared. We showed that there are substantial differences in galactitol metabolic processes between ancestral and selectable mutant strains. The findings of these phenotypes once again break the traditional view that antibiotics act in a simple linear chain of events, where the antibiotic enters the cell and binds and inhibits an essential target, thus stopping growth or killing the bacteria (7, 34, 35). Bacteria immediately respond to drugs as part of their tolerance physiology and can potentially evolve genetic changes as a result of their response (36). Ultimately, this process may affect additional properties and pathways in the bacterium, as it involves a fitness cost. Subsequently, COG and KEGG annotations indicated that they were involved in a wide range of metabolic pathways, supporting this conclusion.

As described by others, the degradation of galactitol by Salmonella remains under-investigated, with the exception of its ability to contribute to in vivo growth in animals, mainly evidenced by (i) competing for galactitol, E. coli enhances colonization resistance against S. Typhimurium (30), (ii) transposon insertion in this region (gat gene cluster of 9.6 kb length encompassing 10 genes) results in a slightly attenuated growth phenotype in different hosts, including mice, chickens, pigs, and calves (29, 37), and (iii) S. Typhimurium LTL strain is defective in growth on galactitol sole carbon source M9 minimal medium, while its RamA-deficient strain grows well (24). It is further shown that galactitol is important for Salmonella colonization, invasion, and selection of host diversity and that the absence of RamA affects the metabolism of galactitol by Salmonella. Our work shows the metabolism of the ciprofloxacin-induced selectable mutant strain of galactitol compared with the ancestral strain, mainly in terms of the inability to grow on a medium that is the sole carbon source of galactitol, and significant differences in the in vivo bacterial loads of mice enriched with galactitol. Importantly, we also showed that the response to ciprofloxacin is specific and dosage dependent and the upregulation of RamA and gatR transcription machinery may influence bacterial survival and persistence; meanwhile, deleting the ramA gene in these strains inhibits the typical activation of gatR expression triggered by ciprofloxacin, underscoring ramA’s essential role in the upregulation of gatR in response to the antibiotic, and in the ciprofloxacin induced, RamA directly binds and activates the gatR gene responsible for the utilization of galactitol.

We proposed a model for this ciprofloxacin-induced regulatory pathway, which is illustrated in Fig. 6; briefly, in the scheme depicting the galactitol degradation pathway according to Nolle et al., these data demonstrate that the gene cluster STY3435-STY3445, which encodes a phosphotransferase system and catabolic enzymes (Fig. 6, left), is responsible for the utilization of galactitol by salmonellae; in the environment where ciprofloxacin exists, it enters the bacteria through the porin, overexpression of RamA, which influences drug resistance phenotype, and then directly activates gatR gene expression by binding the gatR promoter, after which gatR (autoregulated repressor) was bound to these two main gat gene cluster promoters (gatY and gatZ) as well as to its own promoter, demonstrating that ciprofloxacin activates RamA, which can directly activate gatR gene transcription and control galactitol degradation (Fig. 6, right).

Fig 6.

Metabolism and regulation mechanism of galactitol. (Left) Scheme depicting the galactitol degradation pathway according to Nolle et al. (29). (Right) Mechanism of ciprofloxacin on galactose metabolism. Ciprofloxacin enters cells through porin, resulting in an increase in the expression of RamA. RamA inhibits the expression of gatY and gatZ by binding with gatR, thus making S. Typhimurium unable to utilize galactitol.

The relationship between galactitol and resistance to ciprofloxacin likely involves complex interactions within bacterial metabolic and regulatory networks, for instance, metabolic pathways and efflux pumps, oxidative stress response, and genetic and regulatory networks. Understanding these interactions at a molecular level could provide new insights into the mechanisms of antibiotic resistance and inform the development of more effective therapeutic strategies. Our study reveals that ciprofloxacin induces distinct physiological and genetic changes in bacteria compared with untreated counterparts, suggesting complex implications for patients undergoing fluoroquinolone therapy. While we have documented some of the metabolic capacities utilized by pathogens during proliferation under antibiotic stress, the full spectrum remains uncharted, particularly the role of RamA in this context. We noted transcriptional variations between ciprofloxacin-treated and untreated S. Typhimurium subpopulations. However, our analysis does not encompass the complete range or depth of bacterial responses to ciprofloxacin nor does it compare bacteria within host cells exposed to the antibiotic. Advancing our understanding of how ciprofloxacin treatment influences Salmonella infections would benefit from further investigation into the genes involved in resistance to ciprofloxacin.

In conclusion, from a drug perspective, ciprofloxacin affects the colonization and invasion of Salmonella by inhibiting its virulence-related genes and altering its physiological and metabolic patterns in the gut. From a bacterial perspective, owing to lower intracellular galactitol levels than in the intestine, galactitol accumulation in the cell causes osmotic stress and produces reactive oxygen species (38, 39), and the strain is better able to evade the therapeutic effects of the drug by turning off the metabolic pathway of galactitol and improving the efficiency of intracellular survival. Molecular interactions, bacterial genetics, physiology, and evolutionary biology aspects of antibiotic interactions are old but recently expanding fields with numerous publications (7). This study elucidates the complex relationship between bacterial metabolism and resistance, highlighting that a comprehensive understanding of antibiotic interactions necessitates insight into how bacteria adjust physiologically to antibiotic concentrations, which subsequently affects their adaptive responses. Our investigation into antibiotic interactions not only advances fundamental biological knowledge but also has potential clinical implications.

MATERIALS AND METHODS

Bacterial strains, plasmids, and growth conditions

The strains and plasmids used in the present study are listed in Table S1. The various constructs were confirmed using PCR and DNA sequencing (QINGKE, Xi’an, China). The ancestral strain was wild type (WT), and a wild-type S. Typhimurium strain isolated from poultry was analyzed in this study. In a previous study, this isolate was susceptible to all quinolone antimicrobials and had no PMQR gene and was particularly sensitive to ciprofloxacin (≤0.0625 mg/L) (Fig. 1A). The minimal inhibitory concentration assay was performed using the standard broth microdilution method, according to the procedure recommended by the Clinical and Laboratory Standards Institute (CLSI) (40), using the internal quality control strain E. coli ATCC 25922. Aliquots of a culture inoculated with a single colony and grown overnight at 37°C were stored in 60% (vol/vol) glycerol at −80°C and used for all ancestral controls, and to initiate evolution experiments S. Tm and E. coli, cultures were grown in Luria-Bertani (LB) broth (10 g/L tryptone, 5 g/L yeast extract, and 5 g/L NaCl) or in M9 medium supplemented with 2 mM MgSO₄, 0.1 mM CaCl₂, and 54.9 mM (1%, wt/vol) galactitol. If necessary, the medium was supplemented with one of the following antibiotics: ampicillin (100 µg/mL), kanamycin (50 µg/mL), ciprofloxacin (2–8 μg/mL), or chloramphenicol (20 µg/mL). For the solid media, 1.5% (wt/vol) agar was added. For all growth experiments, single colonies of the ancestral and selectable mutant strains were inoculated into LB or M9 medium, cultured at 37°C overnight, and inoculated 1:100 in the appropriate liquid growth medium for specific experimental applications. Growth curves were obtained from bacterial cultures incubated at 37°C with or without agitation in 250-mL bulb flasks containing 50 mL of medium. OD₆₀₀ was measured at appropriate time intervals.

All mutant strains were generated utilizing the λ Red recombination system, as detailed by Datsenko and Wanner (2000). Antibiotic resistance genes along with FRT sites were amplified by PCR using specific knockout primers, with plasmid pKD4 (kanamycin resistance cassette, 3267bps) serving as the template. The resulting PCR products were electroporated into the recipient S. Tm strain, which harbored plasmid pKD46. The genotypes of the resultant strains were confirmed by PCR using specific check-up primers, as listed in Table S2.

Evolution experiment

The evolution experiment was performed according to the method described previously by Kim et al. (41). Briefly, evolution experiments were conducted in 96-well microtiter plates (Thermo 266120) with a final volume of 200 µL per well for MIC and 10-mL shaker flasks for bacterial proliferation. Each day, frozen cultures from the previous day of evolution were thawed and diluted 1:200 each day into a 96-well plate, twofold gradient of the appropriate antibiotics, beginning with approximately 1/8 the MIC of the ancestral strain. Each plate included one blank well and one well for the ancestral control. Ciprofloxacin gradient plates were prepared for up to 3 days in advance and stored at 4°C until use. The plates were incubated at 37°C, 50% humidity, and 600 rpm for 16–20 hours. After each round of growth, the OD₆₀₀ of the culture was measured using a UV spectrophotometer (Beckman Coulter, DU730, USA). For each population, the wells at the highest drug concentration and with OD₆₀₀ > 0.4 after background subtraction were expanded in shaker flasks and proliferation-selectable mutant strains. These cultures were stored in 60% (vol/vol) glycerol at −80℃ for subsequent experiments and sequencing. Populations evolved for 24 days (Fig. 3A).

Detection of target gene mutations in QRDR

Identification of mutations in the quinolone resistance-determining regions (QRDR) of gyrA, gyrB, parC, and parE sequences using the genomes of ancestral and selectable mutant strains as templates was performed using PCR amplification; the primers used are listed in Table S2. All PCR amplicons were sequenced using QINGKE (Xian, China). The wild-type S. Typhimurium LT2 strain genome sequence in the NCBI genome database was used for comparison to identify mutations.

Biofilm assays

The action on biofilm formation was quantitatively measured by crystal violet staining, as previously described (42). Briefly, ancestral and selectable mutant strains were inoculated into LB medium and cultured at 37°C overnight. The broth was diluted with LB medium to 1.0 × 10⁷ CFU/mL, and 100 µL was added to the sterilized 96-well micro culture plate and incubated at 37°C for 24 hours. Each group contained five duplicate wells, and a blank culture solution group was used as a negative control. The culture solution was carefully absorbed, gently rinsed twice with distilled water, and placed in a ventilated and cool place to be inverted and air dried for fixation. Then, 120 µL of 1% crystal violet solution was added for 15–20 minutes. The samples were then rinsed with running water until no obvious color remained in the blank staining well and placed in a ventilated area to dry. After air drying, 150 µL of 33% glacial acetic acid solution was added, and the staining solution was fully dissolved for 10 min. The OD₆₀₀ was measured using a Multimodal Plate Reader (Bio-Rad).

Incidental sensitivity experiment

The ancestral and selectable mutant strains were cultured in LB medium to OD₆₀₀ = 0.6, and experiments to estimate IC₅₀ were performed in replicates in 96-well plates. According to CLSI, the twofold serial broth microdilution method was used (40). A total volume of 200 µL of the microbial suspension (10⁶ CFU/mL) was added to each well, and each antimicrobial agent was added in successive rows to attain the test concentrations. We measured the OD₆₀₀ at 16 hours using a Multimodal Plate Reader (Bio-Rad). All experiments were performed at least three times for verification. The OD₆₀₀ measurements for each drug concentration were normalized to OD₆₀₀ in the absence of the drug. The panel of antimicrobial compounds tested included amoxicillin (AMO; 0.125–32 μg/mL), ceftiofur sodium (CFS; 0.125–8 μg/mL), cefotaxime (CTX; 0.0625–8 μg/mL), tetracycline (TET; 0.25–64 μg/mL), doxycycline (DOX; 0.25–32 μg/mL), chloramphenicol (CHL; 0.25–64 μg/mL), florfenicol (FLO; 0.25–64 μg/mL), ciprofloxacin (CIP; 0.0625–16 μg/mL), enrofloxacin (ENR; 0.125–16 μg/mL), sulfamethoxydiazine (SMD; 2–128 μg/mL), sulfamethoxazole (SMZ; 2–128 μg/mL), polymyxin E (CT; 0.125–8 μg/mL), polymyxin B (PB; 0.125–8 μg/mL), spectinomycin (SPT; 2–128 μg/mL), roxithromycin (ROX; 2–128 μg/mL), streptomycin (SM; 2–128 μg/mL), and apramycin (APR; 0.25–32 μg/mL) (27). The MIC was defined as the lowest concentration of the antimicrobial agent that can inhibit the visible growth of bacteria (CLSI, 2016). The ancestral strains, E. coli ATCC25922 and Staphylococcus aureus ATCC29213, were used as controls.

Cell adhesion and invasion assays

Referring to Fattinger et al.’s assay method (43), HeLa and Caco cells were maintained in Dulbecco’s modified Eagle’s medium (Gibco) supplemented with 10% heat-inactivated fetal calf serum (Ausbian) and antibiotics. However, antibiotics were omitted in preparation for infection. The cell lines were maintained at 37°C in a CO₂-buffered incubator. For adhesion assays, the cells were also infected by adding a given S. Typhimurium strain for the indicated time. After the incubation period, to remove non-adherent bacteria, infected cells were washed three times with PBS. The cells were then lysed with 0.1% sodium deoxycholate (Sigma Aldrich). Serial dilutions of the lysates were plated on LB agar with appropriate antibiotics. The plates were incubated at 37°C for 18–20 hours, and single colonies were counted to determine the number of cell-associated (adherent) bacteria. For invasion assays, the cells were infected by adding a given S. Typhimurium strain for the indicated time. For enumeration of S. Typhimurium invasion efficiency by plating, infected cells were washed with fresh culture medium, incubated with culture medium containing 200 µg/mL gentamicin (AppliChem) for 40 min to kill extracellular bacteria, washed with PBS, and lysed in 0.1% sodium deoxycholate (Sigma Aldrich), and serial dilutions of the lysates were plated on LB agar with appropriate antibiotics. The culture medium was incubated at 37°C for 18–20 hours, and single colonies were counted.

Biochemical tests

The ancestral and selectable mutant strains were inoculated into solid medium and cultured for 18–24 hours. Monoclonal colonies in the solid culture medium were picked for staining and microscopic observation, and monoclonal colonies meeting the morphological characteristics of the bacteria were inoculated into LB and cultivated to a turbidity of 0.5 McDonnell. The bacterial solution (80 µL) was pipetted into a biochemical reaction tube and incubated at 37°C for 18–24 hours. The biochemical reaction tube (HuanKai Biology, HK1002 EasyID Salmonella Biochemical Identification Kit, China), including malonate, urea, lactose, salicin, mannitol, sorbitol, galactitol, indole, and potassium chloride, was used to verify the result against the color chart supplied by the manufacturer (Fig. 2A).

The sugar fermentation profile (nine different carbohydrates: glucose, fructose, lactose, sucrose, maltose, trehalose, mannitol, galactitol, ribitol, and xylose) was determined by incubating the cultures with different sugars at 37°C for 24 hours in bromocresol purple broth. Bromocresol purple was used to indicate a pH change, resulting in a color shift from purple to yellow. All biochemical experiments were repeated at least thrice.

RNA extraction, preparation of RNA-Seq sample, and sequencing

The methods mentioned in the literature were used to culture bacteria and extract RNA (28). Total RNA was extracted from three replicates of each strain [fig ancestral (n = 3) and selectable mutant strain (n = 3), MIC ≥ 4 µg/mL ciprofloxacin]. In addition, the samples were subjected to a second treatment with DNase to remove any residual genomic DNA. Finally, RNA integrity was assessed using a Bioanalyzer (Agilent), and RNA purity and concentration were detected using a spectrophotometer (DHS, Tianjing, China). Total RNA samples were stored at −80°C until further use.

RNA-seq libraries were prepared according to the manufacturer’s instructions (Biomarker Biotechnology Co., Beijing, China). The removal of 16S and 23S rRNAs from total RNA was performed using a Ribo-Zero rRNA Removal Kit (Bacteria). The mRNA was used to prepare individually barcoded (indexed) RNA-Seq libraries with a NEBNext Ultra II RNA Library Prep Kit for TRAN (BioLabs, Beijing, China). Using the Sequencing by Synthesis method, six RNA-Seq libraries prepared from S. Typhimurium strains were sequenced in a HiSeq2500 instrument (Illumina Inc., USA), and reads were base called and quality filtered using the Consensus Assessment of Sequence and Variance (CASAVA version 1.8). The genome sequence and functional annotation information of S. Typhimurium CT18 were obtained from the NCBI database (GenBank accession number AL513382.1). Bioinformatics analysis was completed using the Biomarker Cloud Platform (Biomarker Biotechnology Co., Beijing, China). Each read was mapped to the reference sequence, which could be considered a mapped read. After the reads were mapped with Bowtie 2, the total raw read counts for each gene were generated. These Mapped Data were used for further statistical analyses to determine the differentially expressed genes, as described below.

Quantitative real-time PCR

RT-qPCR was performed (12). Briefly, 1 mg of total RNA isolated as described above was used to synthesize cDNA using a high-capacity cDNA reverse transcription kit (Applied Biosystems) according to the manufacturer’s protocol. cDNA templates were diluted 10-fold; 2 mL of each diluted sample was used as a template for RT-qPCR, which contained 1× SYBR green PCR master mix (Applied Biosystems), 2.5 mM of each primer, and H₂O to a final volume of 25 mL. PCR conditions included 10 min at 95°C, followed by 40 cycles of 95°C for 10 s, 60°C for 1 min, and 72°C for 30 s. Transcription levels of target genes were normalized using the 16S rRNA gene as an internal standard. A 7500 Real-Time PCR Detection System (Applied Biosystems) was used for RT-qPCR measurements under standard conditions. Experiments were performed in triplicate, with three biological replicates. The relative gene expression levels were calculated using the comparative threshold cycle method.

Phenotypic assay for galactitol utilization

A phenotypic assay was performed using chemically defined M9 minimal medium to determine the utilization of the selected carbon sources. M9 minimal medium was supplemented with 10 mg/L thiamine, 500 mg/L histidine, a combination of vitamin solution, and 1% glucose as the carbon source. For visual inspection of growth by the scribing method, a selectable mutant strain was grown on M9 agar plates with 1% galactitol and ciprofloxacin at different concentrations and kept humid at 37°C for 24 hours. For the growth curves, ancestral and selectable mutant strains were grown overnight in LB medium and then incubated in M9 minimal medium supplemented with 1% galactitol. Growth at 37°C was monitored by measuring optical density at 600 nm for 24 hours using a UV spectrophotometer (Beckman Coulter, DU730, USA).

Expression and purification of RamA protein

Using E. coli BL21 (DE3) as a prokaryotic expression system, RamA protein was prepared in a soluble form. Briefly, a single colony carrying the recombinant plasmid pGEX-4T-1-rama was inoculated into 5 mL of liquid LB medium containing 100 µg/mL ampicillin and grown for 12 hours at 37°C. Then, the overnight cultures (~5 mL) were diluted into 1 L of LB medium and grown to an OD₆₀₀ of approximately 0.4 at 37°C. Protein expression was induced by the addition of isopropyl β-D-1-thiogalactopyranoside to a final concentration of 0.5 mM for 5 hours at 37°C. Cells were harvested by centrifugation at 5,000 × g for 10 min at 4°C, washed three times with lysis buffer, and stored at −80°C until use. Frozen cells were resuspended in lysis buffer containing phenylmethylsulfonyl fluoride and lysozyme and lysed by sonication. After centrifugation, the supernatant was extracted using a Pierce GST Spin Purification Kit (Thermo) as described by the manufacturer. Proteins were eluted, recovered, and dialyzed against storage buffer. Finally, protein concentrations were determined using a protein assay kit (Bio-Rad), and purity was assessed by SDS-PAGE.

Electrophoretic mobility shift assay

To detect the binding of RamA to the DNA probes, EMSAs were performed using a Light Shift Chemiluminescent EMSA Kit (Thermo). Fragments representing parts of the putative promoter regions of gatY, gatZ, and gatR were amplified using biotinylated primers as described above (for oligonucleotides, see Fig. S1), and 100 ng of DNA was mixed with increasing amounts of purified RamA-GST in binding buffer [10 mM Tris pH 7.5, 1 mM EDTA, 1 mM dithiothreitol, 90 mM KCl, 10 mM MgCl₂, 10 mM acetyl phosphate, 50 ng/µL poly (dI-dC), 1 µg/mL bovine serum albumin, and 5% glycerol]. The total volume was 20 µL. After incubation for 30 min at room temperature, the samples were loaded with 4 µL of 6× loading dye on a 12% polyacrylamide gel and separated at 120 V for 150 min at 4°C in the same buffer that was pre-cooled to 4°C. The gel was transferred to a nylon membrane and detected according to the manufacturer’ s instructions. Images were visualized and captured under UV irradiation.

Mouse colonization experiments

All animal experiments were approved by the local authorities. Specific pathogen-free mice were obtained from the Lanzhou Veterinary Research Institute (Lanzhou, China). Mice (BALB/c) aged 6–8 weeks were used in all experiments and were randomly assigned to experimental groups. During the experiment, the mice were housed in groups of seven mice per cage, rather than being single housed. The health status of all mice was assessed twice daily. The indicated mouse lines were treated with 2% galactitol or sterile PBS as a control 10 days before further treatment (Fig. 4C). For infection experiments, the S. Typhimurium ancestral strain was grown on LB plates containing streptomycin (50 mg/mL), while the selectable mutant strain was grown on LB plates containing streptomycin (50 mg/mL) and ciprofloxacin (4 µg/mL) at 37°C. One colony was resuspended in 5 mL LB and grown for 12 hours at 37°C on a wheel rotor. A subculture (1:100 dilution) was prepared in fresh LB and incubated for an additional 6 hours. Bacteria were washed in ice-cold sterile PBS, pelleted, and resuspended in fresh PBS. Mice were infected with ancestral and selectable mutant strains by oral gavage with 100 µL of a bacterial suspension (approximately 1 × 10⁸ CFU). Mice were given either non-supplemented drinking water or water supplemented with 2% galactitol during infection with S. Typhimurium to determine the role of galactitol in colonization resistance. All the mice were sacrificed by cervical dislocation. S. Typhimurium loads in feces 1 day and cecum contents 2 days post-infection were determined by plating on LB supplemented with the appropriate antibiotics.

Statistical analysis

Statistical details for each experiment are provided in the figure legends. The data were analyzed using GraphPad Prism 9.0 Windows software (GraphPad Software, La Jolla California USA, www.graphpad.com). The levels of significance were set as follows: ∗P < 0.05, considered statistically significant; ∗∗P < 0.01 and ∗∗∗P < 0.001 were all considered highly significant; only those are indicated in the figures.

ETHICS APPROVAL

All animal experiments were handled following strict procedures (Guide for the Care and Use of Laboratory Animals published by the Institute of Laboratory Animal Resources, Reference No. LVRIAEC-2020-019). This study was approved by the Institutional Animal Care and Use Committee of the State Key Laboratory of Veterinary Etiological Biology, Lanzhou Veterinary Research Institute. Every experiment was conducted to minimize animal suffering and to reduce the number of animals used.

DATA AVAILABILITY

All data included in this study are available upon request by contact with the corresponding authors (Qiwei Chen, chenqiwei@caas.cn, and Xiaowei Gong, gongxiaowei@caas.cn).

SUPPLEMENTAL MATERIAL

The following material is available online at https://doi.org/10.1128/jb.00178-24.

Supplemental material. jb.00178-24-s0001.docx.

Figures S1 to S3; Tables S1 to S3.

Table S4. jb.00178-24-s0002.xls.

Significantly differentially regulated genes.

Table S5. jb.00178-24-s0003.xlsx.

COG analysis.

Table S6. jb.00178-24-s0004.xlsx.

KEGG pathway.

Table S7. jb.00178-24-s0005.xls.

Top 30 upregulated and downregulated genes.

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.