Evaluation of the Membrane Damage Mechanism of Chlorogenic Acid against Yersinia enterocolitica and Enterobacter sakazakii and Its Application in the Preservation of Raw Pork and Skim Milk

Plant-derived antimicrobial agents have adequate antimicrobial effects on food-borne pathogens, which can be used as food preservatives. The purpose of this study was to evaluate the antibacterial mechanism of chlorogenic acid (CA) against Yersinia enterocolitica and Enterobacter sakazakii. The minimum inhibitory concentration (MIC) of CA was determined by employing the broth microdilution method. Then, the cell function and morphological changes of Y. enterocolitica and E. sakazakii treated with CA were characterized. Finally, the growth inhibition models of Y. enterocolitica in raw pork and E. sakazakii in skim milk were constructed through the response surface methodology. The results demonstrated that CA has a satisfactory inhibitory effect against Y. enterocolitica and E. sakazakii with a MIC of 2.5 mg/mL. In addition, CA inhibited the growth of Y. enterocolitica and E. sakazakii via cell membrane damage, such as depolarization of the cell membrane, reduction in intracellular adenosine triphosphate (ATP) and pH levels, and destruction of cell morphology. Moreover, CA reduced two log cycles of Y. enterocolitica in raw pork and E. sakazakii in skim milk at a certain temperature. According to the corresponding findings, CA has the potential to be developed as an effective preservative to control Y. enterocolitica and E. sakazakii-associated foodborne diseases.


Introduction
Due to the development of society, as well as the continuous improvement of living standards, food safety has garnered increased attention; meanwhile, foodborne diseases serve as one of the main factors affecting food safety [1]. Foodborne diseases are mainly caused by foodborne pathogens, which seriously threaten human health and affect the development of the social economy. The Centers for Disease Control and Prevention estimates that 48 million illnesses have been caused by foodborne diseases, incurring annual economic costs to the U.S. estimated at USD 152 billion to USD 1.4 trillion [2]. Yersinia enterocolitica, a kind of gram-negative bacterium, is widely distributed in foods such as vegetables, meat, dairy, and aquatic products [3]. Y. enterocolitica can proliferate at 4 • C, making it dangerous if contaminated food is stored under refrigerated conditions. Pork is a common host of Y. enterocolitica, and most human infections with Y. enterocolitica are caused by eating raw or undercooked contaminated pork [4]. Y. enterocolitica is mainly transmitted through food or water sources, causing fever, enterocolitis, and sepsis [5]. Therefore, it is of great significance to prevent the contamination of foodborne pathogens during the process of food processing, packaging, storage, and transportation.

Effect of CA on Growth Curve of Y. enterocolitica and E. sakazakii
In order to further analyze the antibacterial activity of CA against Y. enterocolitica and E. sakazakii, growth curves of Y. enterocolitica and E. sakazakii treated with different concentrations of CA were then established. The growth trend of Y. enterocolitica treated with CA lower than 1/8 MIC was basically the same as that of the control group, indicating that CA had no obvious influences on the growth of Y. enterocolitica ( Figure 2a). However, when the concentration of CA reached MIC and 2 MIC, the cell density of Y. enterocolitica did not change within 24 h. The corresponding results confirmed that higher concentrations of CA led to a longer lag phase and lower growth rate of Y. enterocolitica. E. sakazakii in the negative control group began to enter the log phase at 3 h and reached the stable phase at 10 h (Figure 2b). The growth trend of E. sakazakii treated with CA at a lower MIC was basically found to be the same as that of the control group; however, when E. sakazakii treated with CA at MIC and 2 MIC, bacterial growth was slow, and the lag phase was prolonged. The results demonstrated that CA had a good inhibitory effect on the growth of Y. enterocolitica and E. sakazakii; specifically, it performed better on Y. enterocolitica.

Effect of CA on Membrane Potential of Y. enterocolitica and E. sakazakii
The potential difference between two sides of the cell membrane is referred to as the membrane potential, which is an electrical phenomenon accompanied by cell life activities [18]. Maintaining normal membrane potential is vital for ATP synthesis, as well as the normal physiological function of cells. DiBAC4(3) is a lipophilic anionic fluorescent dye, and it is sensitive to cell membrane potential [19], which is commonly used as an indicator

Effect of CA on Growth Curve of Y. enterocolitica and E. sakazakii
In order to further analyze the antibacterial activity of CA against Y. enterocolitica and E. sakazakii, growth curves of Y. enterocolitica and E. sakazakii treated with different concentrations of CA were then established. The growth trend of Y. enterocolitica treated with CA lower than 1/8 MIC was basically the same as that of the control group, indicating that CA had no obvious influences on the growth of Y. enterocolitica ( Figure 2a). However, when the concentration of CA reached MIC and 2 MIC, the cell density of Y. enterocolitica did not change within 24 h. The corresponding results confirmed that higher concentrations of CA led to a longer lag phase and lower growth rate of Y. enterocolitica.
E. sakazakii in the negative control group began to enter the log phase at 3 h and reached the stable phase at 10 h (Figure 2b). The growth trend of E. sakazakii treated with CA at a lower MIC was basically found to be the same as that of the control group; however, when E. sakazakii treated with CA at MIC and 2 MIC, bacterial growth was slow, and the lag phase was prolonged. The results demonstrated that CA had a good inhibitory effect on the growth of Y. enterocolitica and E. sakazakii; specifically, it performed better on Y. enterocolitica.

Effect of CA on Growth Curve of Y. enterocolitica and E. sakazakii
In order to further analyze the antibacterial activity of CA against Y. enterocolitica and E. sakazakii, growth curves of Y. enterocolitica and E. sakazakii treated with different concentrations of CA were then established. The growth trend of Y. enterocolitica treated with CA lower than 1/8 MIC was basically the same as that of the control group, indicating that CA had no obvious influences on the growth of Y. enterocolitica ( Figure 2a). However, when the concentration of CA reached MIC and 2 MIC, the cell density of Y. enterocolitica did not change within 24 h. The corresponding results confirmed that higher concentrations of CA led to a longer lag phase and lower growth rate of Y. enterocolitica. E. sakazakii in the negative control group began to enter the log phase at 3 h and reached the stable phase at 10 h (Figure 2b). The growth trend of E. sakazakii treated with CA at a lower MIC was basically found to be the same as that of the control group; however, when E. sakazakii treated with CA at MIC and 2 MIC, bacterial growth was slow, and the lag phase was prolonged. The results demonstrated that CA had a good inhibitory effect on the growth of Y. enterocolitica and E. sakazakii; specifically, it performed better on Y. enterocolitica.

Effect of CA on Membrane Potential of Y. enterocolitica and E. sakazakii
The potential difference between two sides of the cell membrane is referred to as the membrane potential, which is an electrical phenomenon accompanied by cell life activities [18]. Maintaining normal membrane potential is vital for ATP synthesis, as well as the normal physiological function of cells. DiBAC4(3) is a lipophilic anionic fluorescent dye, and it is sensitive to cell membrane potential [19], which is commonly used as an indicator The potential difference between two sides of the cell membrane is referred to as the membrane potential, which is an electrical phenomenon accompanied by cell life activities [18]. Maintaining normal membrane potential is vital for ATP synthesis, as well as the normal physiological function of cells. DiBAC4(3) is a lipophilic anionic fluorescent dye, and it is sensitive to cell membrane potential [19], which is commonly used as an indicator dye for cell membrane potential. The cell is stimulated by the external environment, and the cell membrane is in different potential states. When fluorescence intensity decreases, it indicates that the cell is hyperpolarized; on the contrary, increased fluorescence intensity points to cell depolarization [20].
The results showed that CA significantly increased the membrane potential of Y. enterocolitica and E. sakazakii (Figure 3a), signifying that the cell membranes of Y. enterocolitica and E. sakazakii were depolarized. The change in cell membrane potential affects the power of protons in and out of the cell and inhibits the synthesis of ATP, eventually leading to bacterial death. Similar studies have shown that ginger essential oil depolarizes the cell membranes of S. aureus and E. coli [21]. Depolarization and hyperpolarization occur mainly due to pH changes or increased ion movement, especially the diffusion of K + or K + with several other ions. They diffuse inward and outward to balance the membrane potential. In addition, maintaining ion homeostasis is an integral part of cell growth, which is essential for many energy-related metabolism mechanisms [22]. Therefore, the perturbation of membrane potential caused by CA may affect cell metabolism and eventually lead to bacterial death. dye for cell membrane potential. The cell is stimulated by the external environment, and the cell membrane is in different potential states. When fluorescence intensity decreases, it indicates that the cell is hyperpolarized; on the contrary, increased fluorescence intensity points to cell depolarization [20]. The results showed that CA significantly increased the membrane potential of Y. enterocolitica and E. sakazakii (Figure 3a), signifying that the cell membranes of Y. enterocolitica and E. sakazakii were depolarized. The change in cell membrane potential affects the power of protons in and out of the cell and inhibits the synthesis of ATP, eventually leading to bacterial death. Similar studies have shown that ginger essential oil depolarizes the cell membranes of S. aureus and E. coli [21]. Depolarization and hyperpolarization occur mainly due to pH changes or increased ion movement, especially the diffusion of K + or K + with several other ions. They diffuse inward and outward to balance the membrane potential. In addition, maintaining ion homeostasis is an integral part of cell growth, which is essential for many energy-related metabolism mechanisms [22]. Therefore, the perturbation of membrane potential caused by CA may affect cell metabolism and eventually lead to bacterial death.

Effect of CA on Intracellular ATP of Y. enterocolitica and E. sakazakii
During energy conversion and transmission, ATP serves a key substance and directly provides energy for cell life activities (respiration, proliferation, differentiation, and apoptosis) [23]. The ATP levels in intact cells are in a stable state. However, the destruction of cell homeostasis and integrity may cause changes in intracellular ATP concentrations under stress [24]. The change in ATP content is related to the energy metabolism of various organs. Therefore, influencing the synthesis of ATP is a way for natural preservatives to exert their antibacterial effect.
The effect of CA on the intracellular ATP content of Y. enterocolitica and E. sakazakii is shown in Figure 3b. In the control group, the intracellular ATP contents of Y. enterocolitica and E. sakazakii were found to be 3500 and 3600, respectively, while the intracellular ATP contents of Y. enterocolitica and E. sakazakii treated with CA were noted to decrease significantly. ATP is mainly produced in the mitochondria, which indicates that CA may promote the accumulation of reactive oxygen species (ROS) and lead to the damage of pathogenic mitochondria, resulting in the decline in pathogenic activity. Guo et al. found that luteolin sharply decreased the intracellular ATP concentration of Trueperella pyogenes [25]. The decrease in intracellular ATP concentration may be caused by the rise in the ATP hydrolysis rate by the proton pump, resulting in the rapid consumption of ATP; alternatively, the change in cell membrane permeability may result in ATP leakage.  During energy conversion and transmission, ATP serves a key substance and directly provides energy for cell life activities (respiration, proliferation, differentiation, and apoptosis) [23]. The ATP levels in intact cells are in a stable state. However, the destruction of cell homeostasis and integrity may cause changes in intracellular ATP concentrations under stress [24]. The change in ATP content is related to the energy metabolism of various organs. Therefore, influencing the synthesis of ATP is a way for natural preservatives to exert their antibacterial effect.
The effect of CA on the intracellular ATP content of Y. enterocolitica and E. sakazakii is shown in Figure 3b. In the control group, the intracellular ATP contents of Y. enterocolitica and E. sakazakii were found to be 3500 and 3600, respectively, while the intracellular ATP contents of Y. enterocolitica and E. sakazakii treated with CA were noted to decrease significantly. ATP is mainly produced in the mitochondria, which indicates that CA may promote the accumulation of reactive oxygen species (ROS) and lead to the damage of pathogenic mitochondria, resulting in the decline in pathogenic activity. Guo et al. found that luteolin sharply decreased the intracellular ATP concentration of Trueperella pyogenes [25]. The decrease in intracellular ATP concentration may be caused by the rise in the ATP hydrolysis rate by the proton pump, resulting in the rapid consumption of ATP; alternatively, the change in cell membrane permeability may result in ATP leakage.

Effect of CA on Intracellular pH of Y. enterocolitica and E. sakazakii
As intracellular ATP depletion and decreased membrane potentials after CA treatment suggest the effects of CA on the cellular membrane, pHin was comprehensively investigated in this study. Once the CFDA-SE fluorescent probe enters the cell, it is hydrolyzed by esterase into carboxyfluorescein succinimidyl ester (CFSE), which emits fluorescence and stays there [26]. Therefore, based on the fluorescence response to pH, the pHin changes were monitored in living cells.
The pHin of Y. enterocolitica and E. sakazakii treated with CA was observed to decrease significantly (p < 0.01). Specifically, the pHin of Y. enterocolitica and E. sakazakii without CA treatment was found to be 6.33 ± 0.13 and 6.57 ± 0.15, respectively. After being treated with CA at MIC, the pHin of Y. enterocolitica and E. sakazakii decreased to 4.70 ± 0.10 and 3.44 ± 0.06, respectively. After treatment with CA at 2 MIC, the pHin of Y. enterocolitica and E. sakazakii decreased to 4.22 ± 0.05 and 3.12 ± 0.09 (Figure 4a,b). pHin has been reported to play an essential role in the transcription and synthesis of bacterial DNA, enzyme activity, and protein synthesis [27]. In addition, pHin also controls the cell membrane. The change in pHin means that the permeability of the cell membrane has changed. CA can reduce the level of pHin of Y. enterocolitis and E. sakazakii, demonstrating that CA can change the internal environment of the normal growth of Y. enterocolitis and E. sakazakii, leading to bacterial death. Consistently, a study by Li et al. reported that a tannin-rich fraction from pomegranate rind significantly decreased the pHin of L. monocytogenes [28]. As intracellular ATP depletion and decreased membrane potentials after CA treatment suggest the effects of CA on the cellular membrane, pHin was comprehensively investigated in this study. Once the CFDA-SE fluorescent probe enters the cell, it is hydrolyzed by esterase into carboxyfluorescein succinimidyl ester (CFSE), which emits fluorescence and stays there [26]. Therefore, based on the fluorescence response to pH, the pHin changes were monitored in living cells.
The pHin of Y. enterocolitica and E. sakazakii treated with CA was observed to decrease significantly (p < 0.01). Specifically, the pHin of Y. enterocolitica and E. sakazakii without CA treatment was found to be 6.33 ± 0.13 and 6.57 ± 0.15, respectively. After being treated with CA at MIC, the pHin of Y. enterocolitica and E. sakazakii decreased to 4.70 ± 0.10 and 3.44 ± 0.06, respectively. After treatment with CA at 2 MIC, the pHin of Y. enterocolitica and E. sakazakii decreased to 4.22 ± 0.05 and 3.12 ± 0.09 (Figure 4a,b). pHin has been reported to play an essential role in the transcription and synthesis of bacterial DNA, enzyme activity, and protein synthesis [27]. In addition, pHin also controls the cell membrane. The change in pHin means that the permeability of the cell membrane has changed. CA can reduce the level of pHin of Y. enterocolitis and E. sakazakii, demonstrating that CA can change the internal environment of the normal growth of Y. enterocolitis and E. sakazakii, leading to bacterial death. Consistently, a study by Li et al. reported that a tannin-rich fraction from pomegranate rind significantly decreased the pHin of L. monocytogenes [28].

Effect of CA on Cell Membrane Damage of Y. enterocolitica and E. sakazakii
The cell membrane is one of the main components of the cellular structure, which is involved in maintaining cell integrity, material transportation, receptor function, and information transmission [29]. The LIVE/DEAD BacLight™ Bacterial Viability Kit was used to evaluate the effect of CA on the cell membrane integrity of Y. enterocolitica and E. sakazakii. SYTO 9 can enter all cells, which binds to DNA and RNA, emitting green fluorescence. PI can only enter cells with damaged cell membranes, emitting red fluorescence. Moreover, sublethal injured cells with damage and esterase activity emit yellow fluorescence [30]. In this experiment, the fluorescence color of Y. enterocolitica and E. sakazakii after double staining with SYTO 9/PI was observed by the merged images of CLSM, to show the effect of CA on the cell membrane integrity of Y. enterocolitica and E. sakazakii.
The results showed that the cells of Y. enterocolitica and E. sakazakii without CA treatment emitted green fluorescence (Figure 5a,d), indicating that the cell membrane was intact and undamaged. After treatment with CA at MIC, most Y. enterocolitica cells exhibited red fluorescence; however, some cells gave off yellow fluorescence (Figure 5b), indicating that part of the cell membrane was damaged. However, all E. sakazakii cells displayed red  The cell membrane is one of the main components of the cellular structure, which is involved in maintaining cell integrity, material transportation, receptor function, and information transmission [29]. The LIVE/DEAD BacLight™ Bacterial Viability Kit was used to evaluate the effect of CA on the cell membrane integrity of Y. enterocolitica and E. sakazakii. SYTO 9 can enter all cells, which binds to DNA and RNA, emitting green fluorescence. PI can only enter cells with damaged cell membranes, emitting red fluorescence. Moreover, sublethal injured cells with damage and esterase activity emit yellow fluorescence [30]. In this experiment, the fluorescence color of Y. enterocolitica and E. sakazakii after double staining with SYTO 9/PI was observed by the merged images of CLSM, to show the effect of CA on the cell membrane integrity of Y. enterocolitica and E. sakazakii.
The results showed that the cells of Y. enterocolitica and E. sakazakii without CA treatment emitted green fluorescence (Figure 5a,d), indicating that the cell membrane was intact and undamaged. After treatment with CA at MIC, most Y. enterocolitica cells exhibited red fluorescence; however, some cells gave off yellow fluorescence (Figure 5b put forward that the cell membrane of P. aeruginosa became damaged following CA treatment, and the number of membrane damaged cells increased with treatment time and CA concentration [31].

Effect of CA on Cell Morphology of Y. enterocolitica and E. sakazakii
Electron microscopy is a powerful tool, which is used to better understand the influence of external factors on cell morphology. In this study, FEGSEM was used to observe the changes in cell morphology of Y. enterocolitica and E. sakazakii treated with CA. The degree of damage of Y. enterocolitica and E. sakazakii cells was shown to increase with the rise in CA concentration. The cells of Y. enterocolitica and E. sakazakii without CA treatment were complete, and the cells were rod-shaped, full, and smooth with obvious edges (Figure 6a,d), indicating that 2% ethanol had no effect on their cell morphology. When Y. enterocolitica and E. sakazakii were treated with CA at MIC, the bacterial surface was uneven and shrunk, and some cells appeared to adhere to each other (Figure 6b,e). After Y. enterocolitica and E. sakazakii were treated with CA at 2 MIC, the bacteria were seriously damaged, and the inherent corynebacterium morphology of Y. enterocolitica and E. sakazakii was lost, with a large number of contents being leaked (Figure 6c,f). These observations showed that CA induced morphological changes in Y. enterocolitica and E. sakazakii in a concentration-dependent manner.
In this study, the morphological changes of the two bacteria may be caused by the effect of CA on the integrity and permeability of the cell membrane, which led to the separation of the cell membrane from the cell wall, as well as the leakage of intracellular material. These findings were consistent with the results of the membrane potential measurement, indicating that the cell membrane may serve as the main target for the antibac-

Effect of CA on Cell Morphology of Y. enterocolitica and E. sakazakii
Electron microscopy is a powerful tool, which is used to better understand the influence of external factors on cell morphology. In this study, FEGSEM was used to observe the changes in cell morphology of Y. enterocolitica and E. sakazakii treated with CA. The degree of damage of Y. enterocolitica and E. sakazakii cells was shown to increase with the rise in CA concentration. The cells of Y. enterocolitica and E. sakazakii without CA treatment were complete, and the cells were rod-shaped, full, and smooth with obvious edges (Figure 6a,d), indicating that 2% ethanol had no effect on their cell morphology. When Y. enterocolitica and E. sakazakii were treated with CA at MIC, the bacterial surface was uneven and shrunk, and some cells appeared to adhere to each other (Figure 6b,e). After Y. enterocolitica and E. sakazakii were treated with CA at 2 MIC, the bacteria were seriously damaged, and the inherent corynebacterium morphology of Y. enterocolitica and E. sakazakii was lost, with a large number of contents being leaked (Figure 6c,f). These observations showed that CA induced morphological changes in Y. enterocolitica and E. sakazakii in a concentration-dependent manner.
In this study, the morphological changes of the two bacteria may be caused by the effect of CA on the integrity and permeability of the cell membrane, which led to the separation of the cell membrane from the cell wall, as well as the leakage of intracellular material. These findings were consistent with the results of the membrane potential measurement, indicating that the cell membrane may serve as the main target for the antibacterial effects of CA. Severe membrane damages due to cell permeability result in a bacterial morphological transition from rod-shaped cells to amorphous cells, which is consistent with FEGSEM observations of Shigella dysentery and Streptococcus pneumoniae treated with CA. Lou et al. demonstrated that CA significantly increased the permeability of the outer membrane and plasma membrane of Shigella dysentery and Streptococcus pneumoniae, resulting in the loss of barrier function, as well as the slight leakage of nucleotides [32]. Citrus essential oil can also change the permeability and integrity of the plasma membrane of E. coli and L. rhamnosus, thus achieving antibacterial effects [33]. terial effects of CA. Severe membrane damages due to cell permeability result in a bacterial morphological transition from rod-shaped cells to amorphous cells, which is consistent with FEGSEM observations of Shigella dysentery and Streptococcus pneumoniae treated with CA. Lou et al. demonstrated that CA significantly increased the permeability of the outer membrane and plasma membrane of Shigella dysentery and Streptococcus pneumoniae, resulting in the loss of barrier function, as well as the slight leakage of nucleotides [32]. Citrus essential oil can also change the permeability and integrity of the plasma membrane of E. coli and L. rhamnosus, thus achieving antibacterial effects [33].

Inhibitory Effect of CA on Y. enterocolitica in Raw Pork and E. sakazakii in Skim Milk
Gram-negative bacteria Y. enterocolitica and E. sakazakii are common pathogens, which are found in raw pork and skim milk, respectively. The effect of CA on the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk was evaluated according to the count of viable bacteria. Positive values represented reduction while negative values represented promotion (Tables 1 and 2). According to the standard deviation in parentheses next to the Y value, the difference between the parallel samples was found to be very small. The number of bacteria of Y. enterocolitica and E. sakazakii in the untreated group increased by 2-3 log cycles, which was positively correlated with CA concentration and temperature. In addition, when Y. enterocolitica and E. sakazakii were treated with CA at MIC, the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk was almost completely inhibited. Moreover, when Y. enterocolitica and E. sakazakii were treated with CA at 2 MIC, the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk was inhibited by 1.7-2 log cycles. Therefore, CA was shown to effectively inhibit the growth

Inhibitory Effect of CA on Y. enterocolitica in Raw Pork and E. sakazakii in Skim Milk
Gram-negative bacteria Y. enterocolitica and E. sakazakii are common pathogens, which are found in raw pork and skim milk, respectively. The effect of CA on the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk was evaluated according to the count of viable bacteria. Positive values represented reduction while negative values represented promotion (Tables 1 and 2). According to the standard deviation in parentheses next to the Y value, the difference between the parallel samples was found to be very small. The number of bacteria of Y. enterocolitica and E. sakazakii in the untreated group increased by 2-3 log cycles, which was positively correlated with CA concentration and temperature. In addition, when Y. enterocolitica and E. sakazakii were treated with CA at MIC, the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk was almost completely inhibited. Moreover, when Y. enterocolitica and E. sakazakii were treated with CA at 2 MIC, the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk was inhibited by 1.7-2 log cycles. Therefore, CA was shown to effectively inhibit the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk. Using Design Expert 8.06 software to build the following quadratic model, the regression equations of Y. enterocolitica and E. sakazakii were: Y = −1.6771 − 0.0701 X 1 − 0.0210 X 2 + 1.2490 X 3 + 0.0001 X 1 X 2 + 0.0043 X 1 X 3 − 0.0012 X 2 X 3 + 0.0014 X 12 + 0.0001 X 22 − 0.0790 X 32 and Y = −1.4329 − 0.0809 X 1 − 0.0329 X 2 + 1.2847 X 3 + 0.0001 X 1 X 2 + 0.0035 X 1 X 3 − 0.0012 X 2 X 3 + 0.0017 X 12 + 0.0002 X 22 − 0.0766 X 32 , where Y represents the logarithm of Y. enterocolitica and E. sakazakii cell inhibition; and X 1 , X 2 , and X 3 refer to time, temperature, and CA concentration, respectively. In order to test the validity of the regression equation and evaluate the associations of Y. enterocolitica and E. sakazakii inhibition with the various factors, the regression equation was analyzed by variance analysis (Tables 3 and 4). Accordingly, an overall association was presented between them, with the correlation coefficient R-Squared being 0.9986 and 0.9981, while the correction coefficient Adj R-Squared was 0.9925 and 0.9892, respectively. The regression model was noted to be very significant (p < 0.0001), and no conspicuous Lack of Fit (p > 0.05) was presented, indicating that the model fitting degree was good and the test error was small. Hence, the model can be applied to predict the antibacterial effect of CA on the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk. The variance test of the regression equation showed that the interaction between temperature and CA concentration had a significant impact on the antibacterial effect of CA on Y.
enterocolitica and E. sakazakii (X 2 X 3 , p < 0.0001). In addition, temperature had no significant effect on the bacterial number of Y. enterocolitica within 12-24 h (X 2 , p > 0.05). Furthermore, the interaction between treatment time and CA concentration (X 1 X 3 , p > 0.05) had no significant influence on Y. enterocolitica and E. sakazakii growth inhibition. The 3D response surface plot and contour plot (Figures 7 and 8) intuitively displayed the impact of the three factors on the response value and trend of change. The CA of MIC was shown to basically inhibit the increase in Y. enterocolitica in raw pork and E. sakazakii in skim milk, which further verified the inhibitory effect of CA on Y. enterocolitica and E. sakazakii in food systems. Similarly, Shan et al. reported that oregano, clove, pomegranate peel, and grape seed extracts can be used as aerobic packaging to inhibit foodborne pathogens (L. monocytogenes, Salmonella enteritidis, and S. aureus) in pork at 20 • C [34].
The 3D response surface plot and contour plot (Figures 7 and 8) intuitively displayed the impact of the three factors on the response value and trend of change. The CA of MIC was shown to basically inhibit the increase in Y. enterocolitica in raw pork and E. sakazakii in skim milk, which further verified the inhibitory effect of CA on Y. enterocolitica and E. sakazakii in food systems. Similarly, Shan et al. reported that oregano, clove, pomegranate peel, and grape seed extracts can be used as aerobic packaging to inhibit foodborne pathogens (L. monocytogenes, Salmonella enteritidis, and S. aureus) in pork at 20 °C [34]. The 3D response surface plot and contour plot (Figures 7 and 8) intuitively displayed the impact of the three factors on the response value and trend of change. The CA of MIC was shown to basically inhibit the increase in Y. enterocolitica in raw pork and E. sakazakii in skim milk, which further verified the inhibitory effect of CA on Y. enterocolitica and E. sakazakii in food systems. Similarly, Shan et al. reported that oregano, clove, pomegranate peel, and grape seed extracts can be used as aerobic packaging to inhibit foodborne pathogens (L. monocytogenes, Salmonella enteritidis, and S. aureus) in pork at 20 °C [34].

Reagents
Chlorogenic acid (HPLC ≥ 98%) was purchased from Biotech Bioengineering (Shanghai) Co., Ltd. (Shanghai, China) and dissolved in Luria-Bertani (LB) medium and phosphate-buffered saline (PBS) containing 2% (v/v) ethanol in order to prepare different concentrations of CA. All other reagents were of analytical grade.

Bacterial Strains and Culture Conditions
Yersinia enterocolitica (BNCC 108930) and Enterobacter sakazakii (BNCC 186080) were purchased from BeNa Culture Collection (BNCC, Beijing, China) and cultured in LB medium. Y. enterocolitica and E. sakazakii frozen at −80 °C were activated on a LB plate, after which a single colony was inoculated into LB medium and cultured overnight. In order to obtain a fresh bacterial suspension, the overnight cultured bacterial suspension was transferred to a 100 mL LB medium and cultivated to the logarithmic phase (approximately 1 × 10 8 CFU/mL).

Determination of MIC
The MIC was measured via microdilution in 96-well microtiter plates [35]. The preparation of the bacterial suspension was described in Section 2.2, and the optical density (OD) of the bacterial suspension at 600 nm was adjusted to 0.5. The OD600 of 0.5 corresponded to approximately 1 × 10 8 colony forming units (CFU)/mL. Then, 190 μL of bacterial suspension and 10 μL of CA were added into the 96-well plate (Nunc, Copenhagen, Denmark). Afterward, the mixture was cultured at 37 °C for 8 h. Finally, the OD600 was monitored with a multimode reader (Synergy H1, BioTek, Winooski, VT, USA). The final concentrations of CA were 10, 5, 2.5, 1.25, 0.625, 0.3125, and 0.15625 mg/mL. The LB-containing 2% ethanol group and 2.5 mg/mL ampicillin group were labeled as the negative control group and positive control group, respectively. The MIC was the lowest CA concentration without visible bacterial growth.

Bacterial Growth Curve
The antibacterial activity was reflected by the determination of the bacterial growth curve according to Kang et al. with a few modifications [36]. Briefly, 190 μL of bacterial suspension with an OD600 of 0.5 and 10 μL of CA were added into the 96-well plate (Nunc, Copenhagen, Denmark). Then, the mixture was cultured at 37 °C for 24 h. Finally, the

Reagents
Chlorogenic acid (HPLC ≥ 98%) was purchased from Biotech Bioengineering (Shanghai) Co., Ltd. (Shanghai, China) and dissolved in Luria-Bertani (LB) medium and phosphatebuffered saline (PBS) containing 2% (v/v) ethanol in order to prepare different concentrations of CA. All other reagents were of analytical grade.

Bacterial Strains and Culture Conditions
Yersinia enterocolitica (BNCC 108930) and Enterobacter sakazakii (BNCC 186080) were purchased from BeNa Culture Collection (BNCC, Beijing, China) and cultured in LB medium. Y. enterocolitica and E. sakazakii frozen at −80 • C were activated on a LB plate, after which a single colony was inoculated into LB medium and cultured overnight. In order to obtain a fresh bacterial suspension, the overnight cultured bacterial suspension was transferred to a 100 mL LB medium and cultivated to the logarithmic phase (approximately 1 × 10 8 CFU/mL).

Determination of MIC
The MIC was measured via microdilution in 96-well microtiter plates [35]. The preparation of the bacterial suspension was described in Section 2.2, and the optical density (OD) of the bacterial suspension at 600 nm was adjusted to 0.5. The OD 600 of 0.5 corresponded to approximately 1 × 10 8 colony forming units (CFU)/mL. Then, 190 µL of bacterial suspension and 10 µL of CA were added into the 96-well plate (Nunc, Copenhagen, Denmark). Afterward, the mixture was cultured at 37 • C for 8 h. Finally, the OD 600 was monitored with a multimode reader (Synergy H1, BioTek, Winooski, VT, USA). The final concentrations of CA were 10, 5, 2.5, 1.25, 0.625, 0.3125, and 0.15625 mg/mL. The LB-containing 2% ethanol group and 2.5 mg/mL ampicillin group were labeled as the negative control group and positive control group, respectively. The MIC was the lowest CA concentration without visible bacterial growth.

Bacterial Growth Curve
The antibacterial activity was reflected by the determination of the bacterial growth curve according to Kang et al. with a few modifications [36]. Briefly, 190 µL of bacterial suspension with an OD 600 of 0.5 and 10 µL of CA were added into the 96-well plate (Nunc, Copenhagen, Denmark). Then, the mixture was cultured at 37 • C for 24 h. Finally, the OD 600 was monitored with a multimode reader (Synergy H1, BioTek, Winooski, VT, USA) every 1 h. The final concentrations of CA were 10, 5, 2.5, 1.25, 0.625, 0.3125, 0.15625, and 0.0781 mg/mL. The LB-containing 2% ethanol group and 2.5 mg/mL ampicillin group were labeled as the negative control group and positive control group, respectively.

Determination of Membrane Potential
The experiment was carried out according to Wu et al. with some modifications [37]. The preparation of the bacterial suspension was described in Section 2.2. The supernatant of the bacterial suspension was removed by centrifugation (4000× g, 10 min, 4 • C), and the bacteria were washed twice with PBS and suspended in PBS. CA was then added to the bacterial suspension with an OD 600 of 0.5 to achieve final concentrations of 0, MIC, and 2 MIC. Next, the sample was cultured at 37 • C for 2 h, and 200 µL of bacterial suspension was added to the black, opaque 96-well microtiter plates (Nunc, Copenhagen, Denmark), while 1 mM of fluorescent probe bis-(1,3-dibutylbarbituric acid) trimethine oxonol (DiBAC4(3); Molecular Probes, Sigma, Louis, MO, USA) was added to the sample. After incubating at 37 • C for 15 min in the dark, the fluorescence intensity was measured by a multimode reader (Synergy H1, BioTek, Winooski, VT, USA). The excitation wavelength was 492 nm, and the emission wavelength was 515 nm.

Measurement of Intracellular ATP Concentrations
The intracellular ATP concentration was determined through the method put forward by Han et al. [38], and preparation of the bacterial suspension was described in Section 2.2. The supernatant of the bacterial suspension was removed by centrifugation (4000× g, 10 min, 4 • C), and the bacteria were washed twice with PBS and suspended in PBS. CA was then added to the bacterial suspension with an OD 600 of 0.5 to achieve final concentrations of 0, MIC, and 2 MIC. Then, the sample was cultured at 37 • C for 2 h, and the bacteria were washed 3 times with PBS and collected via centrifugation (4000× g, 10 min, 4 • C). Afterward, 100 µL of cell lysate buffer was added to the sample in order to lyse the cell, after which the supernatant was collected via centrifugation (4000× g, 10 min, 4 • C) and stored on ice to prevent ATP degradation. The content of intracellular ATP was then determined using an ATP assay kit according to the manual's instructions (Beyotime Bioengineering Institute, Shanghai, China). Specifically, 100 µL of detection working solution was added to the black 96-well microplate (Nunc, Copenhagen, Denmark), after 10 min, and the background ATP in the microplate was eliminated. Then, 20 µL of supernatant was added to the 96-well microplate. Finally, the chemiluminescence intensity of the sample was determined using a multimode reader (Synergy H1, BioTek, Winooski, VT, USA).

Measurement of Intracellular pH Level
pHin was determined according to a modified method put forward by Shi et al. [39], with the preparation of the bacterial suspension described in Section 2.2. The supernatant of the bacterial suspension was removed by centrifugation (4000× g, 10 min, 4 • C), and the bacteria were washed twice with HEPES buffer solution. Next, 3 mM of fluorescent probe carboxyfluorescein diacetate succinimidyl ester (cFDA-SE; Molecular Probes, Sigma, Louis, MO, USA) was added to the bacterial suspension. After 20 min, the cells were washed twice with PBS, and 10 mM of glucose solution was added and incubated at 37 • C for 30 min to eliminate non-conjugated cFDA-SE. Finally, the cells were washed twice with PBS and suspended in PBS. CA (0, MIC, and 2 MIC) was added to the cell culture and stained by fluorescence in black, opaque 96-well microtiter plates (Nunc, Copenhagen, Denmark) for 2 h. The fluorescence intensity was measured by a multimode reader (Synergy H1, BioTek, Winooski, VT, USA). The excitation wavelength was 490 nm and 440 nm, and the emission wavelength was 520 nm.
Calibration curves were determined by the cFDA-SE-loaded cells with different pH buffers. The pH buffer was prepared using glycine (50 mM), potassium chloride (50 mM), citric acid (50 mM), and sodium dihydrogen phosphate (50 mM). The pH (3, 4, 5, 6, 7, 8, 9, and 10) was then adjusted with NaOH/HCL. The bacteria were suspended in different pH buffers, and valinomycin was added to equilibrate the pHin and pHout. Finally, the fluorescence intensity was determined, and the pHin of the cells was evaluated according to the ratio of the fluorescence signal of the pH-sensitive wavelength (490 nm) and pHinsensitive wavelength (440 nm).

Confocal Laser Scanning Microscopy Analysis
Membrane permeability and dynamic changes were evaluated using the LIVE/DEAD BacLight™ Bacterial Viability Kit (Molecular Probes, Thermo Fisher, Waltham, MA, USA) according to the procedure put forward by Du et al. with slight modifications [40]. The preparation of the bacterial suspension was described in Section 2.2. The supernatant of the bacterial suspension was removed by centrifugation (4000× g, 10 min, 4 • C), and the bacteria were washed twice with normal saline and suspended in normal saline. CA was added to the bacterial suspension with an OD 600 of 0.5 to achieve final concentrations of 0, MIC, and 2 MIC, respectively. Then, the sample was cultured at 37 • C for 2 h, and the bacterial suspension was washed 3 times with normal saline and suspended in normal saline. Next, 3 µL of the 2× staining solution (SYTO9/PI) was added to 1 mL of the bacterial suspension. After incubating at 37 • C for 15 min in the dark, the bacteria were washed 3 times with normal saline and suspended in 100 µL of normal saline. Finally, 2 µL of bacterial suspension was transferred to a glass slide and observed by CLSM (LSM800, Carl Zeiss, Yarra, Germany).

Field Emission Gun Scanning Electron Microscope Analysis
The effects of CA on the morphology of Y. enterocolitica and E. sakazakii were determined using FEGSEM [41]. The preparation of the bacterial suspension was carried out according to Section 2.2. The supernatant of the bacterial suspension was removed by centrifugation (4000× g, 10 min, 4 • C), and the bacteria were washed twice with PBS and suspended in PBS. CA was added to the bacterial suspension with an OD 600 of 0.5 to achieve final concentrations of 0, MIC, and 2 MIC. The sample was then cultured at 37 • C for 2 h, and the cells were washed 3 times with PBS and collected by centrifugation (4000× g, 10 min, 4 • C). They were then fixed with 2.5% glutaraldehyde solution overnight at 4 • C. Then, the bacteria were collected via centrifugation (4000× g, 10 min, 4 • C) and dehydrated with a graded series of ethanol (30%, 50%, 70%, 90%, and 100%) for 10 min each. Finally, the bacteria were incubated with isoamyl acetate for 30 min and collected by centrifugation (4000× g, 10 min, 4 • C). The bacteria were then dried through freeze-drying, and the dried sample was sprayed with gold on the FEGSEM support. The morphology of the cells was observed using FEGSEM (MLA 650, FEI, Hillsboro, OR, USA).

Modeling the Inhibitory Effect of Chlorogenic Acid on the Growth of Y. enterocolitica in Raw Pork and E. sakazalii in Skim Milk
Frozen pork was purchased from a local supermarket and transported to the laboratory. The raw pork was aseptically cut into small pieces (approximately 1.5 cm × 1.5 cm × 1.5 cm) and exposed to ultraviolet light for 30 min to reduce the presence of other contaminants. A suspension of Y. enterocolitica was added to the 10 g raw pork sample in order to artificially contaminate the raw pork, with the final bacterial concentration being 10 7 CFU/g [42]. Skim milk was purchased from a local supermarket. A suspension of E. sakazakii bacteria was added to the 10 mL skim milk sample to artificially contaminate the skim milk, with the final bacterial concentration being 10 7 CFU/mL [43]. The samples were then divided into three groups, the control group (without antibacterial drugs), sample group treated with CA at MIC, and sample group treated with CA at 2 MIC.

Statistical Analysis
All experimental results were analyzed using the SPSS software (SPSS 8.0 for Windows). All data were expressed as mean ± standard deviation (n = 3). Analysis of variance (ANOVA) was carried out to determine any significant differences (p ≤ 0.01).

Conclusions
Within the food industry, the application of plant-derived preservatives has received intensive attention. This study demonstrated the effective antibacterial activity of CA against Y. enterocolitica and E. sakazakii by increasing the permeability of the cell membrane, destroying the integrity of the cell membrane, and causing the leakage of cell contents. Response surface methodology analysis illustrated that CA can be used to inhibit the growth of Y. enterocolitica in raw pork and E. sakazakii in skim milk. Therefore, CA, as a natural antibacterial preservative, possesses broad application prospects within the food industry.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data that support the findings of this study are available on request from the corresponding author.