Transcriptomic Analysis Revealed Antimicrobial Mechanisms of Lactobacillus rhamnosus SCB0119 against Escherichia coli and Staphylococcus aureus

Lactic acid bacteria were reported as a promising alternative to antibiotics against pathogens. Among them, Lactobacillus rhamnosus could be used as probiotics and inhibit several pathogens, but its antibacterial mechanisms are still less known. Here, L. rhamnosus SCB0119 isolated from fermented pickles could inhibit bacterial growth or even cause cell death in Escherichia coli ATCC25922 and Staphylococcus aureus ATCC6538, which was mainly attributed to the cell-free culture supernatant (CFS). Moreover, CFS induced the accumulation of reactive oxygen species and destroyed the structure of the cell wall and membrane, including the deformation in cell shape and cell wall, the impairment of the integrity of the cell wall and inner membrane, and the increases in outer membrane permeability, the membrane potential, and pH gradient in E. coli and S. aureus. Furthermore, the transcriptomic analysis demonstrated that CFS altered the transcripts of several genes involved in fatty acid degradation, ion transport, and the biosynthesis of amino acids in E. coli, and fatty acid degradation, protein synthesis, DNA replication, and ATP hydrolysis in S. aureus, which are important for bacterial survival and growth. In conclusion, L. rhamnosus SCB0119 and its CFS could be used as a biocontrol agent against E. coli and S. aureus.


Introduction
Foodborne pathogens, including Escherichia coli and Staphylococcus aureus, cause foodborne diseases and bring serious harm to human health [1][2][3]. Approximately 2.8 billion people worldwide suffer from diarrhea due to foodborne illness, and about 3000 people die each year from illness by ingesting food contaminated with foodborne pathogens in the United States [4]. Existing widely in food, water, food-producing animals, and so on [5], E. coli can cause various human diseases ranging from gastroenteritis, such as diarrhea, to extra-intestinal infections, including those of the urinary tract, bloodstream, and central nervous system [6,7]. For example, Shiga toxin-producing E. coli caused 2,801,000 acute illnesses annually, 3890 cases of hemolytic uremic syndrome, 270 cases of permanent endstage renal disease, and 230 deaths, according to databases from 21 countries published between 1 January 1990 and 30 April 2012 [8]. As one of the main foodborne pathogens contaminating ready-to-eat products, meat, egg, dairy products, cream-filled pastries, and cakes [9] and existing in food production animals [10], S. aureus could produce exotoxins and cause food poisoning with the symptoms including nausea, vomiting, and abdominal cramps with or without diarrhea [11]. Additionally, S. aureus is a leading cause of bacteremia, which harbors high rates in the first year of life, a low incidence through  To further investigate the effects of L. rhamnosus SCB0119 on bacterial growth and its potential mechanisms, CFS was chosen to conduct the following studies. The minimum inhibitory concentrations (MICs) of CFS against E. coli ATCC25922 and S. aureus ATCC6538 were 3.91 and 7.81 mg/mL, respectively. Obviously, 1 MIC of CFS completely inhibited the growth of E. coli ATCC25922 ( Figure 1B) and S. aureus ATCC6538 ( Figure 1C). Moreover, the reduction in bacterial growth of E. coli ATCC25922 and S. aureus ATCC6538 started at the early stationary phase, and the time needed to reach a plateau was delayed by 6 h for E. coli ATCC25922 ( Figure 1B) and 4 h for S. aureus ATCC6538 ( Figure 1C), when 1/2 MIC of CFS was added into Luria Bertani (LB) medium.

Effects of CFS on the Production of Intracellular ATP and Reactive Oxygen Species in E. coli and S. aureus
CFS dose-dependently inhibited the production of intracellular ATP in E. coli ATCC25922 and S. aureus ATCC6538. In comparison to 2.49 and 83.22 nmol/mg protein on average for E. coli ATCC25922 (Figure 2A) and S. aureus ATCC6538 ( Figure 2B), respectively, in the control group, the ATP production decreased by 33.68% and 85.27% in 1/2 MIC group, 57.01% and 96.95% in 1 MIC group, and 93.31% and 99.49% in 2 MIC group, respectively, suggesting CFS inhibited E. coli ATCC25922 and S. aureus ATCC6538 growth or even caused death because the decreases in intracellular ATP levels mean that cells undergo apoptosis, necrosis, or are in a toxic state [22]. The mean diameters of inhibition zones after E. coli ATCC25922 and S. aureus ATCC6538 were incubated with L. rhamnosus SCB0119 culture solution and CFS for 12 h, respectively. (B,C) Effects of CFS on the growth of E. coli ATCC25922 and S. aureus ATCC6538, respectively. Error bars: SD of the mean from three repeated assays.

Effects of CFS on the Production of Intracellular ATP and Reactive Oxygen Species in E. coli and S. aureus
CFS dose-dependently inhibited the production of intracellular ATP in E. coli ATCC25922 and S. aureus ATCC6538. In comparison to 2.49 and 83.22 nmol/mg protein on average for E. coli ATCC25922 (Figure 2A) and S. aureus ATCC6538 ( Figure 2B), respectively, in the control group, the ATP production decreased by 33.68% and 85.27% in 1/2 MIC group, 57.01% and 96.95% in 1 MIC group, and 93.31% and 99.49% in 2 MIC group, respectively, suggesting CFS inhibited E. coli ATCC25922 and S. aureus ATCC6538 growth or even caused death because the decreases in intracellular ATP levels mean that cells undergo apoptosis, necrosis, or are in a toxic state [22].
Moreover, CFS induced oxidative stress in E. coli ATCC25922 and S. aureus ATCC6538 cells. As shown in Figure 2C, the fluorescence intensities using DCFH-DA staining in E. coli ATCC25922 and S. aureus ATCC6538 cells increased by 16.10% and 23.09% in 1/2 MIC group, 26.12% and 41.03% in 1 MIC group, and 43.90% and 64.36% in 2 MIC group, respectively, compared with those in the control group, suggesting CFS increased the reactive oxygen species (ROS) levels in a dose-dependent manner. Figure 2. Effects of CFS from L. rhamnosus SCB0119 on the production of ATP (A,B) and ROS (C) in E. coli ATCC25922 (A,C) and S. aureus ATCC6538 (B,C). ROS production was detected using DCFH-DA staining. Different lowercase letters indicate significant differences among the treatments (p < 0.05). Error bars: SD of the mean from three repeated assays.

Changes in Cell Morphology and Structure of E. coli and S. aureus Caused by CFS
Cell morphology and structure of E. coli ATCC25922 and S. aureus ATCC6538 were affected by CFS. Firstly, scanning electron microscopy (SEM) images showed that fewer cells and more shrunk cells were observed in the CFS group than that in the control group ( Figure 3). Moreover, S. aureus ATCC6538 cells became dispersed rather than clustered Figure 2. Effects of CFS from L. rhamnosus SCB0119 on the production of ATP (A,B) and ROS (C) in E. coli ATCC25922 (A,C) and S. aureus ATCC6538 (B,C). ROS production was detected using DCFH-DA staining. Different lowercase letters indicate significant differences among the treatments (p < 0.05). Error bars: SD of the mean from three repeated assays.
Moreover, CFS induced oxidative stress in E. coli ATCC25922 and S. aureus ATCC6538 cells. As shown in Figure 2C, the fluorescence intensities using DCFH-DA staining in E. coli ATCC25922 and S. aureus ATCC6538 cells increased by 16.10% and 23.09% in 1/2 MIC group, 26.12% and 41.03% in 1 MIC group, and 43.90% and 64.36% in 2 MIC group, respectively, compared with those in the control group, suggesting CFS increased the reactive oxygen species (ROS) levels in a dose-dependent manner.

Changes in Cell Morphology and Structure of E. coli and S. aureus Caused by CFS
Cell morphology and structure of E. coli ATCC25922 and S. aureus ATCC6538 were affected by CFS. Firstly, scanning electron microscopy (SEM) images showed that fewer cells and more shrunk cells were observed in the CFS group than that in the control group ( Figure 3). Moreover, S. aureus ATCC6538 cells became dispersed rather than clustered after being treated with CFS ( Figure 3). Additionally, transmission electron microscopy (TEM) images displayed that E. coli ATCC25922 and S. aureus ATCC6538 cells treated with CFS were deformed, ruptured, and less distinctly outlined compared with untreated cells (Figure 3). after being treated with CFS ( Figure 3). Additionally, transmission electron microscopy (TEM) images displayed that E. coli ATCC25922 and S. aureus ATCC6538 cells treated with CFS were deformed, ruptured, and less distinctly outlined compared with untreated cells (Figure 3). Secondly, CFS impaired cell wall integrity based on alkaline phosphatase (AKP) activity. As shown in Figure 4A, the AKP contents in S. aureus ATCC6538 cells increased by 73.81% in the 1/2 MIC group, 144.76% in the 1 MIC group, and 197.62% in the 2 MIC group, and those in E. coli ATCC25922 cells increased 24.69% in 2 MIC group, compared with those in the control group, though CFS at 1/2 and 1 MIC did not change the AKP contents in E. coli ATCC25922 cells.
Moreover, CFS increased outer membrane permeability and impaired inner membrane integrity. The fluorescence intensities using 1-N-phenylnaphthylamine (NPN) staining, which has become a common fluorescent detection signal to evaluate the extent of bacterial outer membrane damage, increased by 36.92-66.18% in E. coli ATCC25922 and S. aureus ATCC6538 cells treated with CFS at different concentrations, compared with those in untreated cells ( Figure 4B). Additionally, the fluorescence intensities using propidium iodide (PI) staining in E. coli ATCC25922 and S. aureus ATCC6538 cells increased by 51.39% and 69.13% in the 1/2 MIC group, 88.44% and 123.12% in the 1 MIC group, and 141.69% and 194.72% in the 2 MIC group ( Figure 4C), respectively, compared with those in the control group, suggesting CFS impaired inner membrane integrity in a dose-dependent manner.
Accompanied by the destruction of cell membrane integrity, CFS increased the membrane potential (ΔΨ) and pH gradient (ΔpH). As shown in Figure 4D, the fluorescence intensities using DiBAC4(3) staining in E. coli ATCC25922 and S. aureus ATCC6538 cells increased by 14.55% and 55.03% in the 1/2 MIC group, 25.83% and 80.51% in the 1 MIC group, and 46.30% and 110.69% in the 2 MIC group, respectively, compared with those in the control group, suggesting CFS caused depolarization of E. coli ATCC25922 and S. aureus ATCC6538 cells. Similarly, the fluorescence intensities using BCECF-AM staining in E. coli ATCC25922 cells increased by 11.73% in the 1/2 MIC group, 22.73% in the 1 MIC group, and 45.31% in the 2 MIC group, and those in S. aureus ATCC6538 cells increased Secondly, CFS impaired cell wall integrity based on alkaline phosphatase (AKP) activity. As shown in Figure 4A, the AKP contents in S. aureus ATCC6538 cells increased by 73.81% in the 1/2 MIC group, 144.76% in the 1 MIC group, and 197.62% in the 2 MIC group, and those in E. coli ATCC25922 cells increased 24.69% in 2 MIC group, compared with those in the control group, though CFS at 1/2 and 1 MIC did not change the AKP contents in E. coli ATCC25922 cells.  , outer membrane permeability (B), inner membrane integrity (C), membrane potential (D) and pH gradients (E) in E. coli ATCC25922 and S. aureus ATCC6538. Changes in the outer membrane permeability, inner membrane integrity, membrane potential, and pH gradients were assayed using NPN, PI, DiBAC4(3), and BCECF-AM staining, respectively. Different lowercase letters indicate significant differences among the treatments (p < 0.05). Error bars: SD of the mean from three repeated assays.

Transcriptomic Analysis of E. coli and S. aureus Cells in Response to CFS
In E. coli ATCC25922, the RNA sequencing results showed that the raw sequence outputs were 32823244, 35152464, 35876796, and 33148174 reads with 88.70%, 90.15%, 88.54%, and 88.21% total matches for the two CFS-free experiments and two CFS experiments, respectively (Table S2). Compared with those in CFS-free experiments, 1/2 MIC of CFS differentially upregulated the expression of 327 genes and downregulated the expression of 70 genes (Table S3). For S. aureus ATCC6538, the raw sequence outputs were . Effects of CFS from L. rhamnosus SCB0119 on AKP contents (A), outer membrane permeability (B), inner membrane integrity (C), membrane potential (D) and pH gradients (E) in E. coli ATCC25922 and S. aureus ATCC6538. Changes in the outer membrane permeability, inner membrane integrity, membrane potential, and pH gradients were assayed using NPN, PI, DiBAC4(3), and BCECF-AM staining, respectively. Different lowercase letters indicate significant differences among the treatments (p < 0.05). Error bars: SD of the mean from three repeated assays.
Moreover, CFS increased outer membrane permeability and impaired inner membrane integrity. The fluorescence intensities using 1-N-phenylnaphthylamine (NPN) staining, which has become a common fluorescent detection signal to evaluate the extent of bacterial outer membrane damage, increased by 36.92-66.18% in E. coli ATCC25922 and S. aureus ATCC6538 cells treated with CFS at different concentrations, compared with those in untreated cells ( Figure 4B). Additionally, the fluorescence intensities using propidium iodide (PI) staining in E. coli ATCC25922 and S. aureus ATCC6538 cells increased by 51.39% and 69.13% in the 1/2 MIC group, 88.44% and 123.12% in the 1 MIC group, and 141.69% and 194.72% in the 2 MIC group ( Figure 4C), respectively, compared with those in the control group, suggesting CFS impaired inner membrane integrity in a dose-dependent manner.
Accompanied by the destruction of cell membrane integrity, CFS increased the membrane potential (∆Ψ) and pH gradient (∆pH). As shown in Figure 4D, the fluorescence intensities using DiBAC4(3) staining in E. coli ATCC25922 and S. aureus ATCC6538 cells increased by 14.55% and 55.03% in the 1/2 MIC group, 25.83% and 80.51% in the 1 MIC group, and 46.30% and 110.69% in the 2 MIC group, respectively, compared with those in the control group, suggesting CFS caused depolarization of E. coli ATCC25922 and S. aureus ATCC6538 cells. Similarly, the fluorescence intensities using BCECF-AM staining in E. coli ATCC25922 cells increased by 11.73% in the 1/2 MIC group, 22.73% in the 1 MIC group, and 45.31% in the 2 MIC group, and those in S. aureus ATCC6538 cells increased by 6.12% and 12.92 in the 1 MIC and 2 MIC groups, respectively, but decreased by 10.57% in the 1/2 MIC group ( Figure 4E), compared with those in the control group.

Transcriptomic Analysis of E. coli and S. aureus Cells in Response to CFS
In E. coli ATCC25922, the RNA sequencing results showed that the raw sequence outputs were 32823244, 35152464, 35876796, and 33148174 reads with 88.70%, 90.15%, 88.54%, and 88.21% total matches for the two CFS-free experiments and two CFS experiments, respectively (Table S2). Compared with those in CFS-free experiments, 1/2 MIC of CFS differentially upregulated the expression of 327 genes and downregulated the expression of 70 genes (Table S3). For S. aureus ATCC6538, the raw sequence outputs were 25524088, 21597232, 26838638, and 24129642 reads with 91.41%, 86.25%, 90.26%, and 80.69% total matches for the two CFS-free experiments and two CFS experiments, respectively (Table S2). When treated with 1/2 MIC of CFS, 814 genes were significantly up-regulated, and 233 genes were differentially down-regulated (Table S4). Moreover, all five upregulated and five downregulated genes selected for evaluation in E. coli ATCC25922 and S. aureus ATCC6538 were differentially expressed in response to CFS ( Figure 5), confirming the reliability of the transcriptome data.
in the 1/2 MIC group ( Figure 4E), compared with those in the control group. , outer membrane permeability (B), inner membrane integrity (C), membrane potential (D) and pH gradients (E) in E. coli ATCC25922 and S. aureus ATCC6538. Changes in the outer membrane permeability, inner membrane integrity, membrane potential, and pH gradients were assayed using NPN, PI, DiBAC4 (3), and BCECF-AM staining, respectively. Different lowercase letters indicate significant differences among the treatments (p < 0.05). Error bars: SD of the mean from three repeated assays.

Transcriptomic Analysis of E. coli and S. aureus Cells in Response to CFS
In E. coli ATCC25922, the RNA sequencing results showed that the raw sequence outputs were 32823244, 35152464, 35876796, and 33148174 reads with 88.70%, 90.15%, 88.54%, and 88.21% total matches for the two CFS-free experiments and two CFS experiments, respectively (Table S2). Compared with those in CFS-free experiments, 1/2 MIC of CFS differentially upregulated the expression of 327 genes and downregulated the expression of 70 genes (Table S3). For S. aureus ATCC6538, the raw sequence outputs were 25524088, 21597232, 26838638, and 24129642 reads with 91.41%, 86.25%, 90.26%, and 80.69% total matches for the two CFS-free experiments and two CFS experiments, respectively (Table S2). When treated with 1/2 MIC of CFS, 814 genes were significantly upregulated, and 233 genes were differentially down-regulated (Table S4). Moreover, all five upregulated and five downregulated genes selected for evaluation in E. coli ATCC25922 and S. aureus ATCC6538 were differentially expressed in response to CFS ( Figure 5), confirming the reliability of the transcriptome data. Moreover, gene ontology (GO) enrichment analysis classified the functions of the differentially expressed genes (DEGs) into biological process, cellular component, and molecular function, which included 57, 1, and 23 for E. coli ATCC25922, and 44, 15, and 4 for S. aureus ATCC6538, respectively. In E. coli ATCC25922, ion transmembrane transport and ion transport, outer membrane-bounded periplasmic space, and anion transmembrane transporter activity and ion transmembrane transporter activity were dominant for biological process, cellular component, and molecular function, respectively ( Figure 6A). Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis demonstrated that DEGs were mapped into nitrogen metabolism, amino acid metabolism, synthesis and degradation of ketone bodies, sulfur metabolism, ABC transporters, carbohydrate metabolism, metabolism of cofactors and vitamins, benzoate degradation, microbial metabolism in diverse environments, drug metabolism-cytochrome P450, and fatty acid metabolism ( Figure 6B), which were crucial for lipid metabolism, amino acid metabolism, cellular structure and processes, and membrane transport. For S. aureus ATCC6538, GO enrichment analysis showed that gene expression and peptide biosynthetic process, ribonucleoprotein complex and ribosome, and structural molecule activity and structural constituent of ribosome were dominant for biological process, cellular component, and molecular function, respectively ( Figure 7A). KEGG enrichment analysis demonstrated that DEGs were mapped into the ribosome, aminoacyl-tRNA biosynthesis, peptidoglycan biosynthesis, protein export, mismatch repair, DNA replication, fatty acid metabolism, bacterial secretion system, metabolism of cofactors and vitamins, the PPAR signaling pathway, RNA polymerase, carbohydrate metabolism, and oxidative phosphorylation ( Figure 7B), which are crucial for lipid metabolism, energy metabolism, translation, and replication and repair. These data implied that the targeted genes of CFS in E. coli might differ from those in S. aureus.
was used as an internal standard. The red line at value 1 represented the expression of chosen genes in the control groups. Error bars: SD of the mean from three repeated assays.
Moreover, gene ontology (GO) enrichment analysis classified the functions of the differentially expressed genes (DEGs) into biological process, cellular component, and molecular function, which included 57, 1, and 23 for E. coli ATCC25922, and 44, 15, and 4 for S. aureus ATCC6538, respectively. In E. coli ATCC25922, ion transmembrane transport and ion transport, outer membrane-bounded periplasmic space, and anion transmembrane transporter activity and ion transmembrane transporter activity were dominant for biological process, cellular component, and molecular function, respectively ( Figure 6A). Kyoto encyclopedia of genes and genomes (KEGG) enrichment analysis demonstrated that DEGs were mapped into nitrogen metabolism, amino acid metabolism, synthesis and degradation of ketone bodies, sulfur metabolism, ABC transporters, carbohydrate metabolism, metabolism of cofactors and vitamins, benzoate degradation, microbial metabolism in diverse environments, drug metabolism-cytochrome P450, and fatty acid metabolism ( Figure 6B), which were crucial for lipid metabolism, amino acid metabolism, cellular structure and processes, and membrane transport. For S. aureus ATCC6538, GO enrichment analysis showed that gene expression and peptide biosynthetic process, ribonucleoprotein complex and ribosome, and structural molecule activity and structural constituent of ribosome were dominant for biological process, cellular component, and molecular function, respectively ( Figure 7A). KEGG enrichment analysis demonstrated that DEGs were mapped into the ribosome, aminoacyl-tRNA biosynthesis, peptidoglycan biosynthesis, protein export, mismatch repair, DNA replication, fatty acid metabolism, bacterial secretion system, metabolism of cofactors and vitamins, the PPAR signaling pathway, RNA polymerase, carbohydrate metabolism, and oxidative phosphorylation ( Figure 7B), which are crucial for lipid metabolism, energy metabolism, translation, and replication and repair. These data implied that the targeted genes of CFS in E. coli might differ from those in S. aureus.  To illuminate the antimicrobial mechanisms of CFS against E. coli and S. aureus, we further analyzed the DEGs involved in the biosynthesis of amino acids, ABC transporters, and fatty acid degradation in E. coli ATCC25922, and ribosome, aminoacyl-tRNA biosynthesis, DNA replication, oxidative phosphorylation, and fatty acid degradation in S. aureus ATCC6538. In E. coli ATCC25922, CFS increased the expressions of ilvN, ilvI, leuA, leuB, and lysC involved in the biosynthesis of amino acids by 2.08-8.75-fold and decreased the expressions of ABC transporter nikE, nikD, nikC, nikB, nikA, and fepD by 52.89-81.45% and the expression of adhP involved in fatty acid degradation by 27.94% (Table 1). Moreover, the expression of osmY encoding one periplasmic protein was down-regulated by 18.14% (Table S3). In S. aureus ATCC6538, CFS upregulated the expression of genes encoding 30S small subunits, 50S large subunits, and aminoacyl-tRNA by 3.66-104.47 folds and the expressions of atpD, atpA, atpE, and atpB related to oxidative phosphorylation by 5.48-16.79 folds, but decreased the expressions of adhP and EKM74_RS04795 related to DNA replication by 38.05% and 53.22%, respectively (Table 2). Moreover, the expressions of frr and rbfA related to ribosome cofactor were up-regulated by 3.19-and 12.50-fold (Table S4) To illuminate the antimicrobial mechanisms of CFS against E. coli and S. aureus, we further analyzed the DEGs involved in the biosynthesis of amino acids, ABC transporters, and fatty acid degradation in E. coli ATCC25922, and ribosome, aminoacyl-tRNA biosynthesis, DNA replication, oxidative phosphorylation, and fatty acid degradation in S. aureus ATCC6538. In E. coli ATCC25922, CFS increased the expressions of ilvN, ilvI, leuA, leuB, and lysC involved in the biosynthesis of amino acids by 2.08-8.75-fold and decreased the expressions of ABC transporter nikE, nikD, nikC, nikB, nikA, and fepD by 52.89-81.45% and the expression of adhP involved in fatty acid degradation by 27.94% (Table 1). Moreover, the expression of osmY encoding one periplasmic protein was down-regulated by 18.14% (Table S3). In S. aureus ATCC6538, CFS upregulated the expression of genes encoding 30S small subunits, 50S large subunits, and aminoacyl-tRNA by 3.66-104.47 folds and the expressions of atpD, atpA, atpE, and atpB related to oxidative phosphorylation by 5.48-16.79 folds, but decreased the expressions of adhP and EKM74_RS04795 related to DNA replication by 38.05% and 53.22%, respectively (Table 2). Moreover, the expressions of frr and rbfA related to ribosome cofactor were up-regulated by 3.19-and 12.50-fold (Table S4), respectively.

Discussion
Lactic acid bacteria were reported as a promising alternative to antibiotics against several pathogens [18]. Here, L. rhamnosus SCB0119 could inhibit the growth of E. coli ATCC25922 and S. aureus ATCC6538. Moreover, its CFS, the main component for antibacterial activity, inhibited bacterial growth or even caused death, induced oxidative stress, and impaired cell shape and structure, as discussed below.
CFS from L. rhamnosus SCB0119 reduced bacterial growth of E. coli ATCC25922 and S. aureus ATCC6538. According to the diameter of the inhibition zone, CFS from L. rhamnosus SCB0119 exhibited greater antibacterial activity against E. coli ATCC25922 and S. aureus ATCC6538 than that from L. rhamnosus XN2 (7.2 ± 0.5 mm for E. coli CICC10302 and 15.1 ± 1.2 mm for S. aureus CICC10384) and LMG 23,522 (8.9 ± 0.2 mm for E. coli O157:H7 and 11.3 ± 0.4 mm for S. aureus) [21,23]. Additionally, the decreases in intracellular ATP levels suggested that CFS from L. rhamnosus SCB0119 might cause the death of E. coli and S. aureus cells [22]. The cell death of pathogens caused by L. rhamnosus was also reported in S. typhimurium treated with L. rhamnosus SQ511 [20].
Moreover, ROS production plays an important role in commencing cessation of the cell during the initiation of cell death [24]. Here, CFS from L. rhamnosus SCB0119 accumulated the ROS levels in E. coli ATCC25922 and S. aureus ATCC6538, which was similar to the increases of ROS production in S. typhimurium caused by CFS from L. rhamnosus SQ511 [20]. The high level of ROS production causes oxidative stress, which is closely associated with adverse effects on cellular components and causes cell death [25,26].
Furthermore, CFS from L. rhamnosus SCB0119 destroyed the cell wall, outer membrane, and plasma membrane. First, the cell wall is the first line to protect cells, including the cell membrane and organelles [27,28]. When the cell wall is damaged, AKP existing between the cell wall and the membrane of bacteria leaks into the extracellular milieu, causing an increase in extracellular AKP activity [29]. In this study, CFS increased AKP contents in E. coli ATCC25922 and S. aureus ATCC6538, suggesting the cell wall of E. coli and S. aureus was impaired by CFS from L. rhamnosus SCB0119. The impairment of the cell wall might affect the cell membrane. Generally, effective antibacterial compounds must penetrate or destroy the plasma membrane of the pathogen [30]. The increases in fluorescence intensity of PI staining, ∆Ψ, and ∆pH indicated that CFS from L. rhamnosus SCB0119 destroyed the integrity of the cell membrane and improved the cell membrane permeability in E. coli ATCC25922 and S. aureus ATCC6538, which resulted in the dissipation of membrane ∆Ψ and ∆pH and depolarization of the cytoplasmic membrane [31,32]. Changes in PI staining, ∆Ψ, and ∆pH were consistent with that in E. coli cells treated with a bacteriocin produced by L. rhamnosus 1.0320 [32]. Additionally, changes in PI staining coincided with that in S. aureus cells conditioned with CFS from L. rhamnosus XN2 [21], and the depolarization of cell membrane was in agreement with that in S. enteritidis caused by CFS from L. rhamnosus SQ511 [20]. Besides, disruption of membrane integrity leads to intracellular ATP leakage, which is also responsible for the decreased intracellular ATP concentration [33]. The elimination of the transmembrane electrochemical gradient might also lead to the loss of the ability of bacterial cells to synthesize ATP [34]. Moreover, the outer membrane is important for barrier function and nutrient flow [35]. The increases in the amount of NPN staining demonstrated that CFS from L. rhamnosus SCB0119 increased the permeability of the outer membrane in E. coli ATCC25922 and S. aureus ATCC6538, which concurred with the findings found in E. coli cells treated with a bacteriocin produced by L. rhamnosus 1.0320 [32]. These data indicated that CFS from L. rhamnosus SCB0119 destroyed the integrity of the cell wall and membrane, increased the permeability of the cell membrane, released the cell contents, and induced cell death in E. coli and S. aureus.
Additionally, CFS from L. rhamnosus SCB0119 affected the transcriptional expression of genes involved in fatty acid degradation, biosynthesis of amino acids, ABC transporters, ribosome, aminoacyl-tRNA biosynthesis, DNA replication, and/or oxidative phosphorylation in E. coli ATCC25922 and S. aureus ATCC6538. Firstly, adhP encoding an alcohol dehydrogenase contributes to the reduction of fatty aldehydes to fatty alcohols [36], and its deletion resulted in the accumulation of fatty aldehydes, which caused damage to the cell or the inhibition of enzymes activities by metabolic disorders or potential toxicity [37]. Here, CFS inhibited the expression of adhP in E. coli ATCC25922 and S. aureus ATCC6538, which might be an explanation for the inhibition of bacterial growth caused by CFS from L. rhamnosus SCB0119. Secondly, nickel is an essential catalytic cofactor of enzymes allowing organisms to inhabit diverse environmental niches, and its pumping through the cytoplasmic membrane is regulated by NikABCDE [38]. Iron intake positively correlated with the synthesis of some ferritin proteins, such as catalases, whose reduction will cause the intracellular accumulation of H 2 O 2 , induce the oxidative stress response inside the thallus, and bring adverse effects on cellular components such as DNA, proteins, and membrane lipids [26,39,40]. The deletion of E. coli fepD caused a reduction in iron intake [41]. Therefore, the downregulation of nikA, nikB, nikC, nikD, nikE, and fepD might result in the damage of nickel and iron homeostasis and then induce the inhibition of bacterial growth in E. coli caused by CFS from L. rhamnosus SCB0119. Thirdly, the accumulation of amino acids could be attributed to related to protein denaturation and blockage of protein synthesis [42]. The overexpression of lysC, ilvI, ilvN, leuA, and leuB might increase the production of lysine, leucine, isoleucine, and valine [43], suggesting the presence of CFS from L. rhamnosus SCB0119 might promote the biosynthesis of lysine, leucine, isoleucine, and valine and then cause the damage to cell survival in E. coli. In addition, the decrease in the expression of osmY might be another explanation for the antibacterial activities of CFS from L. rhamnosus SCB0119 against E. coli because OsmY is a molecular chaperone through a genetic selection that forces cells to optimize unstable protein folding in vivo [44] and positively regulates the integrity of the outer membrane [45]. In S. aureus ATCC6538, the inhibition of bacterial growth caused by CFS from L. rhamnosus SCB0119 might be due to the changes in protein synthesis by mediating the expression of the ribosome cofactor genes (frr and rbfA), genes encoding 30S ribosomal proteins (rpsT, rpsD, rpsI, rpsK, rpsM, rpsE, rpsH, rpsJ, rpsQ, rpsC, and rpsS), genes encoding 50S ribosomal proteins (rpmG, rpmA, rplU, rpmI, rpmJ, rplO, rplR, rplF, rplE, rplN, and rplP), and genes in the pathway of the aminoacyl-tRNA biosynthesis (leuS, gatB, lysS, gltX, pheT, and ileS), because they were involved in the initiation, elongation, and termination of the assembly of the ribosome [46] and important for protein synthesis [47,48]. Moreover, gene EKM74-RS04795 encodes a single-stranded DNA-binding protein, which is crucial in DNA replication, repair, and recombination [49], was down-regulated by CFS in S. aureus ATCC6538, suggesting CFS from L. rhamnosus SCB0119 might cause cell damage or death in S. aureus by interfering with DNA repair pathways and leading to genomic instability. Besides, the expression of four genes (atpA, atpB, atpD, and atpE) encoding ATP synthases hydrolyzing ATP [50] were up-regulated by CFS from L. rhamnosus SCB0119 in S. aureus ATCC6538, which might result in the decreases of ATP content and alive cells.
Lastly, previous studies demonstrated that the consumption of L. rhamnosus strains gave positive results in preventing diarrhea and bacterial vaginosis, restoring normal gastrointestinal microflora, and so on [17]. In addition, when directly mixed with beef, CFS from Lactobacillus with stable antimicrobial activity could effectively reduce the microbial load of inoculated pathogens and enhance the quality and safety of beef products [51], and the addition of L. rhamnosus into milk could inhibit the growth of P. fluorescens and P. putida from raw milk and improve the raw milk quality [52]. Therefore, effectiveness in reducing the growth of E. coli ATCC25922 and S. aureus ATCC6538 suggested that L. rhamnosus SCB0119 and its CFS might be directly consumed by humans and animals or mixed with various foods to prevent the infection or contamination of E. coli and S. aureus, although the findings need further validation of a clinical, practical, and/or industrial scale. In the future, the safety of L. rhamnosus SCB0119 and its CFS, active components from CFS, and the evaluation of their application potential in preventing the infection and contamination of E. coli and S. aureus should be further investigated.

Bacterial Strains and Culture Condition
Lactobacillus rhamnosus SCB0119 (CGMCC NO: 17618) was isolated from fermented pickles by ourselves and cultivated in De Man Rogosa Sharpe (MRS) medium at 37 • C under anaerobic conditions. Escherichia coli ATCC25922 and Staphylococcus aureus ATCC6538 were obtained from the American Type Culture Collection (ATCC) and cultivated at 37 • C in LB medium under constant shaking at 180 rpm.

Preparation of Cell-Free Culture Supernatants and Cells of L. rhamnosus SCB0119
Briefly, L. rhamnosus SCB0119 was inoculated into MRS medium at a ratio of 3:100 (v/v) and statically cultured at 37 • C for 60 h. The supernatants were collected, filtered with a 0.22 µm pore filter (Merck-Millipore, Darmstadt, Germany), and served as CFS. Bacterial cells were washed with phosphate buffer saline (PBS) and then resuspended in an equal volume of PBS.

Determination of the Antibacterial Activity of L. rhamnosus SCB0119
The antimicrobial activity of L. rhamnosus SCB0119 was evaluated using the agar well diffusion method [53] with some modifications. Briefly, the dilutions of E. coli ATCC25922 and S. aureus ATCC6538 cells (10 7 CFU/mL) were added into LB medium with 0.75% agar, respectively. Then, the mixture was poured onto LB agar plates (1.5% agar) with several Oxford cups (6 × 7.8 mm), which were removed when the agar solidified. Aliquots of 100 µL of L. rhamnosus SCB0119 culture, CFS, resuspended bacterial cells, and MRS medium were added to the wells and incubated at 37 • C. After 12 h of incubation, the mean diameters of the inhibition zones were recorded as antibacterial activity of L. rhamnosus SCB0119 against E. coli ATCC25922 and S. aureus ATCC6538. The experiments were conducted in triplicate.

Minimum Inhibitory Concentrations of CFS against E. coli and S. aureus
MICs were determined via the microdilution method in 96-well polystyrene plates (Corning Inc., Corning, NY, USA) [54]. Briefly, CFS was evaporated, dissolved using ultrapure water to 500 mg/mL and stored at −20 • C. Wells containing 100 µL of E. coli ATCC25922/S. aureus ATCC6538 dilutions (10 7 CFU/mL) in LB medium were supplemented with different concentrations (0.49-250 mg/mL) of CFS and incubated at 37 • C for 12 h. Optical densities at 600 nm (OD 600 ) were read using a Multiskan GO Reader (Thermo Scientific, Waltham, MA, USA). The experiments were conducted in triplicate.

Growth Curve of E. coli and S. aureus after Treatment with CFS
The bacterial growth curve was assayed using the method reported by Yan et al. [55]. Briefly, E. coli ATCC25922/S. aureus ATCC6538 dilutions (10 7 CFU/mL) in LB medium were supplemented with 0, 1/2 MIC, and 1 MIC of CFS and incubated in Bioscreen C (Oy Growthcurves Ab Ltd., Turku, Finland), where the temperature was set at 37 • C and the rotating speed was set at 180 rpm. During a 34-h incubation, OD 600 s were read at 2-h intervals. The experiments were conducted in triplicate.

Scanning Electron Microscopy and Transmission Electron Microscopy
The potential effects of CFS on cell morphology and structure of E. coli ATCC25922 and S. aureus ATCC6538 were determined by SEM and TEM [56,57]. Briefly, bacterial dilutions (10 7 CFU/mL) in LB medium were supplemented with 1/2 MIC of CFS and incubated for 10 h at 37 • C by shaking. After centrifugation at 4 • C for 5 min, bacterial cells were collected, washed with PBS (0.1 mM, pH 7.2) three times, fixed overnight with 4% glutaraldehyde, washed with PBS once, and post-fixed in 1% OsO 4 for 30 min. Some post-fixed cells were dehydrated in a graded ethanol series (50%, 70%, 80%, 90%, and 100%), dried by CO 2 for 4 h, coated with gold, and observed under a field emission scanning electron microscope (FEI/Talos L120C, Thermo Fisher Scientific, New York, NY, USA). Other post-fixed cells were dyed with uranium acetate overnight, dehydrated in a graded ethanol series (30%, 50%, 70%, 85%, 95%, and 100%), embedded in Spurr resin, polymerized at 55 • C for 48 h, and observed under a transmission electron microscope (ZEISS, Oberkochen, Germany).

Concentrations of Alkaline Phosphatase and Intracellular ATP
Bacterial dilutions of E. coli ATCC25922/S. aureus ATCC6538 (10 8 CFU/mL) in PBS for the activity of AKP or 5 mM HEPES buffer for the concentration of intracellular ATP were supplemented with 0, 1/2 MIC, 1 MIC, and 2 MIC of CFS and incubated at 37 • C by shaking. After 1 h, the supernatants or bacteria were collected and used to measure the AKP activity using an AKP kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China) and ATP content using an ATP detection kit (Beyotime Biotechnology, Shanghai, China) [58] according to the method described by the manufacturer. The concentration of ATP was determined using external calibration curves of the standard ATP solution at different concentrations. Moreover, the protein concentration in the sample was measured using the total protein quantitative determination kit (Nanjing Jiancheng Bioengineering Institute, Nanjing, China), and the concentration of ATP was converted into nmol/mg protein. All of the experiments were conducted in triplicate.

Assay for Outer Membrane Permeability and Inner Membrane Integrity
For the outer membrane permeability, E. coli ATCC25922/S. aureus ATCC6538 dilutions (10 8 CFU/mL) in 5 mM HEPES buffer containing 5 mM glucose (pH 7.0) were incubated with 10 µM NPN for 30 min at 25 • C and then supplemented with 0, 1/2 MIC, 1 MIC, and 2 MIC of CFS and incubated at 37 • C by shaking [59]. After 30 min, the fluorescence intensities were recorded using a Varioskan Flash (spectraMax i3x, Molecular Devices, San Francisco, CA, USA) at the excitation/emission wavelengths of 350/420 nm. The experiments were conducted in triplicate.
For the inner membrane integrity, E. coli ATCC25922/S. aureus ATCC6538 dilutions (10 8 CFU/mL) in PBS were incubated with 10 µM PI for 30 min at 25 • C and then supplemented with 0, 1/2 MIC, 1 MIC, and 2 MIC of CFS and incubated at 37 • C by shaking [60]. After 1 h, the fluorescence intensities were recorded using the Varioskan Flash at the excitation/emission wavelengths of 535/615 nm. The experiments were conducted in triplicate.

Transcriptome Sequencing and Analysis
Total RNA was extracted from E. coli ATCC25922 and S. aureus ATCC6538 cells grown in LB medium with 1/2 MIC of CFS for 10 h at 37 • C using TRIzol ® Reagent (Invitrogen, Carlsbad, CA, USA) according to the manufacturer's instructions and genomic DNA was removed using DNase I (TaKaRa, Kyoto, Japan). Then RNA quality was determined using 2100 Bioanalyser (Agilent, Palo Alto, CA, USA) and quantified using the ND-2000 (NanoDrop Technologies, Thermo Fisher Scientific, New York, NY, USA). RNAseq strand-specific libraries were created using 5 g of total RNA and the TruSeq RNA sample preparation Kit (Illumina, San Diego, CA, USA). rRNA was removed by RiboZero rRNA removal kit (Epicenter, Stockholm, Sweden) and fragmented using fragmentation buffer. The Illumina-indexed adaptors were ligated after cDNA synthesis, end repair, A-base addition, and ligation according to Illumina's protocol. cDNA target fragments of 200-300 bp were selected on 2% Low Range Ultra Agarose followed by PCR amplified using Phusion DNA polymerase (NEB) for 15 PCR cycles. Paired-end libraries were sequenced by Illumina NovaSeq 6000 sequencing (150 bp × 2, Shanghai BIOZERON Co., Ltd, Shanghai, China) after being quantified by TBS380. The sequencing data were deposited in NCBI Sequence Read Archive (http://www.ncbi.nlm.nih.gov/Traces/sra) under the accession numbers SRP395161 for E. coli ATCC25922 and SRP394988 for S. aureus ATCC6538.
Clean data were mapped to the data from the E. coli genome (NCBI accession: ASM301845v1) and the S. aureus genome (NCBI accession: ASM394542v1), respectively. The expression level of each gene was evaluated by calculating the fragments per kilobase per million fragments mapped (FPKM), and edgeR (https://bioconductor.org/packages/release/bioc/ html/edgeR.html) was used for differential expression analysis. To compare the differential gene expression among the libraries, the false discovery rate (FDR) ≤0.05 and the absolute value of the log 2 ratio ≥1 were used as the threshold to screen the DEGs. Functional classification and pathway analysis of DEGs were performed using GO functional enrichment (Goatools, https://github.com/tanghaibao/Goatools) and KEGG pathway analysis (KOBAS, http://kobas.cbi.pku.edu.cn/kobas3), respectively. The experiments were conducted in twice.

qRT-PCR
Total RNA was extracted from E. coli ATCC25922 and S. aureus ATCC6538 cells grown in LB medium with 1/2 MIC of CFS for 10 h at 37 • C using TRIzol ® Reagent (Invitrogen, Carlsbad, CA, USA), detected by agarose gel electrophoresis, quantified using the ND-2000 (NanoDrop Technologies, Thermo Fisher Scientific, New York, NY, USA), and transcribed reversely into cDNA with PrimeScript RT reagent kit (TaKaRa, Kyoto, Japan). Tenfold dilutions of each cDNA were used as templates to analyze the expression of tested genes via qRT-PCR with paired primers (Table S1) using SYBR ® Premix Ex Taq™ (TaKaRa, Kyoto, Japan). The reaction cycling consisted of initial denaturation at 95 • C for 30 s, followed by 40 cycles of denaturation at 95 • C for 30 s, annealing and extension at 60 • C for 34 s, and ended up with a dissolution curve. The analysis of the dissolution curve evaluated the specificity of the primers. The 16 s rRNA was used as an internal standard. The relative transcription level of each gene was defined as the ratio of its transcript in cells treated with CFS over that without CFS using the 2 −∆∆Ct method, where Ct means the threshold cycle, ∆Ct is equal to the difference in threshold cycles for target and reference (Ct x − Ct r ), and ∆∆Ct equals (∆Ct in CFS group − ∆Ct in control group ) [63]. The experiments were conducted in triplicate.

Data Analysis
Results from replicates were expressed as mean ± standard deviation (SD), and the data analysis was subjected to a one-factor analysis of variance, followed by Tukey's HSD test. p < 0.05 was considered a significant difference in all experiments.

Conclusions
Lactobacillus rhamnosus SCB0119 could inhibit the growth and induce cell death of E. coli and S. aureus, and its antibacterial activity was mainly attributed to the CFS. Its CFS could induce oxidative stress involving ROS and cause damage to the integrity and permeability of the cell wall and membrane, which resulted in the release of the cell contents. Moreover, the transcriptomic analysis showed that CFS aimed at the expression of genes involved in fatty acid degradation, ion transport, and the biosynthesis of amino acids in E. coli and affected the expression of genes involved in fatty acid degradation, protein synthesis, DNA replication, and ATP hydrolysis in S. aureus. These findings demonstrated L. rhamnosus SCB0119 and its CFS have the potential for use as a biocontrol agent against pathogens, including E. coli and S. aureus.