Chestnut Shell Polyphenols Inhibit the Growth of Three Food-Spoilage Bacteria by Regulating Key Enzymes of Metabolism

The microbial contamination of food poses a threat to human health. Chestnut shells, which are byproducts of chestnut processing, contain polyphenols that exert various physiological effects, and thus have the potential to be used in food preservation. This study investigates the bacteriostatic effect and mechanism(s) of the action of chestnut shell polyphenols (CSPs) on three food-spoilage bacteria, namely Bacillus subtilis, Pseudomonas fragi, and Escherichia coli. To this end, the effect of CSPs on the ultrastructure of each bacterium was determined using scanning electron microscopy and transmission electron microscopy. Moreover, gene expression was analyzed using RT-qPCR. Subsequent molecular docking analysis was employed to elucidate the mechanism of action employed by CSPs via the inhibition of key enzymes. Ultrastructure analysis showed that CSPs damaged the bacterial cell wall and increased permeability. At 0.313 mg/mL, CSPs significantly increased the activity of alkaline phosphatase and lactate dehydrogenase, as well as protein leakage (p < 0.05), whereas the activity of the tricarboxylic acid (TCA) cycle enzymes, isocitrate dehydrogenase and α-ketoglutarate dehydrogenase, were inhibited (p < 0.05). The expression levels of the TCA-related genes gltA, icd, sucA, atpA, citA, odhA, IS178_RS16090, and IS178_RS16290 are also significantly downregulated by CSP treatment (p < 0.05). Moreover, CSPs inhibit respiration and energy metabolism, including ATPase activity and adenosine triphosphate (ATP) synthesis (p < 0.05). Molecular docking determined that proanthocyanidins B1 and C1, the main components of CSPs, are responsible for the antibacterial activity. Therefore, as natural antibacterial substances, CSPs have considerable potential for development and application as natural food preservatives.


Introduction
Microbial activity is a key factor in food spoilage, which not only destroys the original nutritional value of food, but also produces toxins that pose a threat to human health. Currently, food-processing plants commonly use chemical synthetic preservatives and antibiotics, with low prices and remarkable anti-corrosion effects, to prevent food spoilage [1,2]; however, these pose potential safety concerns. The abuse of synthetic chemical preservatives may lead to health risks, such as respiratory diseases and infant and child growth retardation [3], while the excessive use of antibiotics may lead to drug resistance and excessive antibiotic residue [4,5]. Therefore, the search for, and development of, natural bacteriostatic agents to replace synthetic preservatives and antibiotics has become a research hotspot.
Natural bacteriostatic agents originate from a wide range of sources, including animals, microorganisms, plant extracts, and their derivatives. Protamine, an antimicrobial peptide extracted from the sperm cells of vertebrates, such as salmon, exhibits antibacterial

CSP Antibacterial Activity Assay
The AGAR diffusion method was modified [17] to determine the antibacterial activity of the CSPs based on the diameter of the antibacterial zone. The prepared bacterial suspension was uniformly coated with NA or LB media, as indicated in Section 2.1. A sterile filter paper with a diameter of 6 mm was soaked in 10-40 mg/mL CSP solution or sterile water (negative control) for 30 s. The filter paper was then placed on the surface of the medium, and culturing was carried out at 30 • C and 37 • C for 24 h. The minimum inhibitory concentration (MIC) and minimum bactericidal concentration (MBC) of the CSPs were determined using the double gradient dilution method. The bacterial suspension (200 µL) and CSPs (10 mg/mL) were added to the liquid medium (2 mL total volume), and the culture was shaken at 180 rpm for 24 h. Sterile water was used as a negative control, and sodium diacetate (SDA), potassium sorbate, and sodium nitrite were positive controls. The final concentrations of CSPs were 1× MIC and 2× MIC, using the same amount of bacterial suspension added to the liquid medium. The OD 600 of the culture medium at 0, 2, 4, 6, 8, 10, 12, and 24 h was measured using a microplate reader (ELX-800, BIO-TEK, Winooski, VT, USA), and growth curves were plotted according to the absorbance of 600 nm [18].

Bacterial Cell Wall and Membrane Damage Assay
Alkaline phosphatase (AKP), lactate dehydrogenase (LDH) activity, and total protein (bicinchoninic acid [BCA]) were measured in the treated bacterial supernatants according to the respective kit instructions (AKFA018C, AKCO003C, AKPR017, Beijing Box Shengong Technology Co., Ltd., Beijing, China). More specifically, 1 mL of CSP solution containing the equivalent of 1× MIC or 2× MIC was added to 10 4 CFU/mL of bacterial suspension for 8 h. The suspension was then centrifuged at 8000× g for 10 min at 4 • C (H2050-R, Xiangtan Xiangyi Instrument Co., Ltd., Xiangtan, China), and the supernatant was tested. Sterile water was used as a negative control, and SDA as a positive control.

Scanning Electron Microscopy (SEM)
The 2× MIC CSPs solution (or sterile water as the negative control and SDA as the positive control) and 10 7 CFU/mL bacterial suspension were combined in equal volumes, and the mixture was cultured for 8 h and then centrifuged (8000× g for 15 min at 4 • C). The precipitate was fixed in 2.5% glutaraldehyde for 12 h, re-suspended twice with phosphatebuffered saline (PBS), and eluted using a 30, 70, 90, and 100% ethanol gradient, centrifuging for 10 min each time. The combined eluates were precipitated, anhydrous ethanol (0.5 mL) was added, and the samples were vacuum freeze-dried. Thallus morphology was observed using a scanning electron microscope (Inspect F50, FEI, Hillsboro, OR, USA) at a magnification of 25 K.

Transmission Electron Microscopy (TEM)
Sample pre-treatment was performed as described in Section 2.4. Morphological changes in the three food-spoilage bacteria after CSP treatment were observed using a transmission electron microscope (Tecnai G2 F20, FEI) (enlarged 25 K).

Activity Measurements of Key Tricarboxylic Acid (TCA) Cycle Enzymes
Bacterial suspensions containing 5-10 million bacterial cells were mixed with different concentrations of CSPs for 8 h, and then centrifuged for 10 min to collect the precipitates. The activities of α-ketoglutarate dehydrogenase (α-KGDH) and isocitrate dehydrogenase (ICDHm) were determined using kits, according to the manufacturer's instructions (AKAC010M and AKAC009M, respectively, Beijing Box Shenggong Technology Co., Ltd., Beijing, China).

RT-qPCR Assay
Total DNA was extracted from B. subtilis, P. fragi, and E. coli using a bacterial genome DNA rapid extraction kit (B518225, Shanghai Bioengineering Co., Ltd., Shanghai, China) and quantified with an ultraviolet spectrophotometer (NanoDrop 2000, Thermo Scientific, Waltham, MA, USA). The mRNA levels of the genes of interest (Table 1) were quantified using qPCR analysis (ExicyclerTM 96, BIONEER, Daejeon, Republic of Korea) under the following conditions: 95 • C for 5 min, 95 • C for 10 s, 60 • C for 15 s, and 72 • C for 15 s, followed by 40 cycles of 72 • C for 90 s, 40 • C for 1 min, melting point 60-94 • C, every 1 • C melting point for 1 s, and 25 • C for 1-2 min. The mRNA expression levels were quantified using the 2 −∆∆CT method.

Molecular Docking
AutoDock Vina [19] was used for the molecular docking of isocitrate dehydrogenase (IDH1) and α-ketoglutarate dehydrogenase (OGDH) with compounds PB1 and PC1, respectively. The 2D structures were downloaded from PubChem (https://PubChem.ncbi. nlm.nih.gov, accessed on 23 December 2022), and the 3D structures of the proteins were downloaded from the Research Collaboratory for Structural Bioinformatics Protein Data Bank (RCSB PDB). The PDB IDs of IDH1 and OGDH were 4UMX and 7WGR, respectively. During docking, the protein structure was converted to a PDB partial charge (Q) and atom type (T) (PDBQT) file containing all polar residues with hydrogen. The compounds were also converted to PDBQT files with all keys rotatable. The Lamarque genetic algorithm was used to simulate flexible docking. The grid box for IDH1 was set to center_x = 11.335, center_y = 27.039, center_z = 80.837, size_x = 40, size_y = 40, and size_z = 40. The grid box for OGDH was set at center_x = 99.908, center_y = 121.999, center_z = 110.413, size_x = 40, size_y = 40, and size_z = 40. The model of the complex was analyzed us- ing Discovery Studio [20]. Interactions between protein-ligand complexes were mapped using PyMol.

Statistical Analysis
SPSS software (version 20; IBM, Chicago, IL, USA) was used to analyse data. Duncan's multiple comparison test was used for difference analysis. A p-value < 0.05 was considered statistically significant. Origin 2018 software (version 9.0; OriginLab, Bethesda, MD, USA) was used to plot the data. All experiments were repeated at least three times, and the data were presented as the mean ± standard deviation (SD), n = 3. The AGAR diffusion method can measure the bacteriostatic effects of applied samples. The antibacterial zone diameters of CSPs (40 mg/mL) against B. subtilis, P. fragi, and E. coli were 13.00 ± 0.42 mm, 10.30 ± 0.07 mm, and 6.00 mm, respectively ( Figure 2). Sterile water did not produce a bacteriostatic effect on the three tested bacteria. CSPs had the best inhibitory effect on B. subtilis, followed by P. fragi, while E. coli did not produce a bacteriostatic zone. Similarly, Nunes et al. [21] found that E. coli treated with the polyphenol extract of Portuguese red wine at different concentration gradients did not generate an inhibitory zone. This may be related to bacterial structure. That is, E. coli and P. fragi were not inhibited or weakly inhibited because they are gram-negative with double-layer cell membranes, preventing most small molecules from crossing the outer membrane and aggregating in the cell, making the drug less active [22]. In contrast, B. subtilis is a grampositive bacterium with a simple cell wall structure composed mainly of peptidoglycan and teichoic acid, resulting in greater inhibition than gram-negative bacteria [23]. This was further demonstrated by Snoussi et al. [24] who assessed the antibacterial activity of D. flabellifolia hydroalcoholic extract (3 mg/disc), revealing an antibacterial zone diameter of 8.00 ± 0.00 mm for the gram-negative bacterium Enterobacter cloacae and 36.33 ± 0.58 mm for the gram-positive bacterium Staphylococcus epidermidis. Meanwhile, a microshoot extract of Nasturtium officinale exhibits narrow-spectrum antibacterial activity against grampositive bacteria and no inhibitory activity against gram-negative bacteria [25]. Similarly, although phenolic compounds extracted from geranium showed significant antibacterial activity against gram-positive and gram-negative bacteria, the effect was more considerable against gram-positive bacteria [26]. Therefore, CSPs may be used as natural preservatives to protect against food spoilage primarily caused by gram-positive bacteria, such as findings and may be due to the culture time and CSP treatment method. That is, the CS content absorbed by the filter paper of the bacteriostatic zone was uncertain. The growt curve adopted quantitative CSPs mixed with bacterial suspension for 8 h. Therefore, com pared with the results of the bacteriostatic zone, the growth curve had a more obviou inhibitory effect on E. coli. These results showed that CSPs could inhibit the growth an propagation of the three tested bacteria; higher CSP concentrations corresponded wit more obvious inhibitory effects.  The diameter of the filter paper used in the experiment was 6 mm. Different lowercase letters within the same species indicate significant differences (p < 0.05) based on variance analysis. CSPs, chestnut shell polyphenols; sterile water, negative control. The 1× MIC and 2× MIC of CSPs for the three tested bacteria were 0.313 mg/mL and 0.625 mg/mL, respectively. The 1× MIC of SDA against B. subtilis, P. fragi, and E. coli was 0.625, 0.313, and 1.25 mg/mL, respectively, and the 2× MIC was 1.25, 0.625, and 2.5 mg/mL, respectively. CSPs, chestnut shell polyphenols; SDA, sodium diacetate; MIC, minimum inhibitory concentration.

MIC and MBC Analysis
The growth of E. coli under the filter paper of CSP treatment groups was inhibited, indicating that CSPs had an inhibitory effect on E. coli. The MIC of CSPs against the three tested bacteria was 0.313 mg/mL ( Table 2). Under the same conditions, the MIC range of the positive control (SDA) was 0.313-5 mg/mL, depending on the type of bacterium. The MBC results showed that CSPs, with an MBC of 0.625 mg/mL, were more effective against the three tested bacteria than were the positive controls (SDA, potassium sorbate, and sodium nitrite), with an MBC range of 5-10 mg/mL. Among the positive controls, SDA showed the best inhibitory effect ( Table 3). The inhibitory effects (MBC) of CSPs against B. subtilis, P. fragi, and E. coli were 16-, 16-, and 8-times that of SDA, respectively, suggesting that CSPs have natural preservative potential. Meanwhile, Zhao et al. [27] assessed the antibacterial effect of tea saponin derived from oil tea shell on E. coli and reported its MIC as 1 mg/mL and MBC as 4 mg/mL. Compared with the effect of CSPs on E. coli, the MIC and MBC of tea saponin were inferior. This may be related to the structure, composition, and purity of the polyphenols. SDA, potassium sorbate, and sodium nitrite are commonly used as food chemical preservatives, among which, sodium nitrite is not only anticorrosive but also impacts coloring and is thus more commonly used in meat products. Chestnut shells contain a large amount of brown pigment and is, thus, also used as a food colorant [28]. If applied to biscuits, chocolate cakes, and meat products, extracts of chestnut shells can serve as both a preservative and colorant, demonstrating their potential in food applications.

Influence of CSPs on the Growth Curve of Food-Spoilage Bacteria
Microorganisms typically exhibit an "S"-shaped growth curve, generally divided into a hysteretic period, logarithmic growth period, stable period, and decline period [29]. The growth curves of B. subtilis, P. fragi, and E. coli, following treatment with saline (control), showed clear logarithmic and stable growth periods, consistent with normal bacterial growth (Figure 2A-C). However, the growth curves of the three bacteria were altered after CSP treatment. The growth curves of P. fragi and E. coli showed obvious retardation, whereas that of B. subtilis did not. Compared with the control group, the logarithmic growth stages of the three bacteria lagged following treatment with 1× MIC CSP. At 2× MIC CSP, the growth curve of the three bacteria was slowed, and the growth inhibition effect was more obvious. Similarly, Li et al. [30] combined chitosan with gallic acid (CTS-GA) and assessed its bactericidal effect on E. coli by assessing changes in its growth curve. The growth of E. coli was significantly inhibited when cultured under 1× MIC (p < 0.01). Moreover, in the current study, the inhibitory effect of 2× MIC CSP was significantly higher than that of 1× MIC SDA for the three bacteria. At 2× MIC CSP, relatively no reproductive growth was observed. Meanwhile, CSPs significantly inhibited E. coli compared with the control group ( Figure 2C). This result contradicted the E. coli bacteriostatic zone findings and may be due to the culture time and CSP treatment method. That is, the CSP content absorbed by the filter paper of the bacteriostatic zone was uncertain. The growth curve adopted quantitative CSPs mixed with bacterial suspension for 8 h. Therefore, compared with the results of the bacteriostatic zone, the growth curve had a more obvious inhibitory effect on E. coli. These results showed that CSPs could inhibit the growth and propagation of the three tested bacteria; higher CSP concentrations corresponded with more obvious inhibitory effects.

CSPs Damage the Structure of the Bacterial Cell Wall and Membrane
AKP is a protease located between the bacterial cell membrane and cell wall. When the cell wall is damaged, a relatively large amount of AKP escapes, resulting in apoptosis. Therefore, the degree of cell wall integrity can be estimated by measuring the activity of AKP in the supernatant before and after treatment [31,32]. Figure 3A shows that, compared with the control group, AKP activity in the CSP-treated experimental group significantly increased in a dose-dependent manner (p < 0.05). The activity of AKP in the supernatant of B. subtilis treated with 1× and 2× MIC CSP was 2-and 2.66-times higher than that of the control group, respectively. The AKP activity of P. fragi increased from 0.58 to 1.26 U/mL after treatment with different CSP concentrations, while the AKP activity of E. coli treated with CSPs or SDA exhibited no significant change at 1× MIC (p > 0.05). The results showed that treatment with different concentrations of CSPs damaged the three tested bacteria to different degrees; in particular, the 2× MIC damaged the cell wall of all three bacteria, causing content leakage. Due to the increased concentration of AKP in the supernatants, the bacteriostatic effect of CSPs is likely related to cell wall destruction, which is consistent with the findings of Lin et al. [33].
LDH activity and protein leakage of the three bacteria were significantly increased after CSP treatment (p < 0.05; Figure 3B,C). At 1× MIC of CSPs, the protein released by B. subtilis significantly increased by 34% compared with the control group (p < 0.05). However, there was no significant change in the protein released by E. coli compared with the control group (p > 0.05), which may be related to differences in bacterial structure, confirming the conclusion made in Section 3.1.1. At 2× MIC CSP, the leaked protein content of B. subtilis, P. fragi, and E. coli was 3.3-, 3.6-, and 3.3-times that of the control group, respectively. The LDH activity of all three bacteria treated with different concentrations of CSPs was significantly higher than that of the control group (p < 0.05; Figure 3C). At 1× MIC, there was no significant difference in the LDH activity between B. subtilis and P. fragi treated with CSPs or SDA (p > 0.05). The LDH activity of E. coli treated with 1× MIC of SDA was 45% higher than that of E. coli treated with CSPs. At 2× MIC, the LDH activity of B. subtilis treated with CSPs or SDA did not differ significantly (p > 0.05), indicating that CSPs at 2× MIC had the best bacteriostatic effect on B. subtilis. This result is consistent with the results of the BCA protein leakage assay.
The cell membrane is another important structure that determines the integrity of bacterial cells [34]. Hence, we assessed the extent of cell membrane damage based on LDH activity and protein leakage [35]. Polyphenols interact with proteins, altering the structure of the plasma membrane and leading to the rapid loss of proteins, which can affect energy metabolism [36,37]. Our results indicated that CSPs caused irreparable damage to the bacterial cell membrane, resulting in the release of AKP, protein, and LDH, while interfering with the normal growth of the tested microorganisms. Zhao et al. [38] reported that treatment of Staphylococcus aureus with a 3× MIC of sugarcane (Saccharum officinarum L.) bagasse extract reduced the bacterial protein content and the inhibited growth and metabolism of the bacterial cells, findings which were consistent with our results. treated with CSPs or SDA exhibited no significant change at 1× MIC (p > 0.0 showed that treatment with different concentrations of CSPs damaged th bacteria to different degrees; in particular, the 2× MIC damaged the cell w bacteria, causing content leakage. Due to the increased concentration of AK natants, the bacteriostatic effect of CSPs is likely related to cell wall destru consistent with the findings of Lin et al. [33]. LDH activity and protein leakage of the three bacteria were significa after CSP treatment (p < 0.05; Figure 3B,C). At 1× MIC of CSPs, the protein subtilis significantly increased by 34% compared with the control gro

CSPs Damage Bacterial Morphology
To further investigate the mechanism of bacterial inhibition by CSPs, SEM and TEM were used to observe changes in cell morphology and ultrastructure. SEM images of control B. subtilis, P. fragi, and E. coli revealed regular rod-shaped morphologies, with smooth surfaces and clear and complete cell walls ( Figure 4A-C). Compared with the control groups, the cell surface of the three tested bacteria treated with 2× MIC CSP showed varying degrees of damage ( Figure 4D-F), with obvious cracks, dents, and adhesions. Similarly, at 2× MIC SDA, the cracks, adhesions, fractures, and fragmentation of cells were observed ( Figure 4G-I).
To investigate the effect of CSPs on the internal structure of the bacteria, TEM observations were carried out ( Figure 5), which verified the SEM conclusions. In the control group, there was no damage or voids on the cell surface of the three tested bacteria; the cell wall and cell membrane were intact, and the cytoplasm was compact and uniform without any leakage of cell contents ( Figure 5A-C). After treatment with 2× MIC CSP ( Figure 5D-F) and 2× MIC SDA ( Figure 5G-I), the three types of bacterial cells appeared severely damaged: the cell wall and membrane were damaged, certain cell membrane structures were discontinuous, cell content leakage was severe, there were more particles on the cytoplasmic edge, and the cell contents appeared condensed. Compared with the control bacterial groups, the 2× MIC CSP treatment groups showed an obvious cavity formation phenomenon and many black clumps in the cells. Similarly, Alshuniaber et al. [39] found that Staphylococcus aureus and E. coli treated with extracted Spirulina polyphenols exhibited deformation, cell damage, and cell wall dissolution. By affecting the composition of the cell membrane, polyphenolic compounds alter the composition of fatty acids in bacterial cell membranes, resulting in significant changes in long-chain unsaturated fatty acids, thereby reducing membrane viscosity, inhibiting the synthesis of ergozite, reducing the fluidity of cell membranes, and increasing cell membrane permeability [40], which is consistent with the experimental results of this study.  To investigate the effect of CSPs on the internal structure of the bacteria, TEM observations were carried out ( Figure 5), which verified the SEM conclusions. In the control group, there was no damage or voids on the cell surface of the three tested bacteria; the cell wall and cell membrane were intact, and the cytoplasm was compact and uniform without any leakage of cell contents ( Figure 5A-C). After treatment with 2× MIC CSP (Figure 5D-F) and 2× MIC SDA ( Figure 5G-I), the three types of bacterial cells appeared severely damaged: the cell wall and membrane were damaged, certain cell membrane structures were discontinuous, cell content leakage was severe, there were more particles on the cytoplasmic edge, and the cell contents appeared condensed. Compared with the control bacterial groups, the 2× MIC CSP treatment groups showed an obvious cavity formation phenomenon and many black clumps in the cells. Similarly, Alshuniaber et al. [39] found that Staphylococcus aureus and E. coli treated with extracted Spirulina polyphenols exhibited deformation, cell damage, and cell wall dissolution. By affecting the composition of the cell membrane, polyphenolic compounds alter the composition of fatty acids in bacterial cell membranes, resulting in significant changes in long-chain unsaturated Therefore, we concluded that the inhibition mechanism of CSPs on the three bacteria included damage to the cell wall and membrane structures, resulting in the aggregation and leakage of AKP, LDH, and other proteins. This damage affected the growth and metabolism of the bacteria and ultimately led to their death due to being unable to maintain their original cell morphology.

CSPs Reduce the Content of ATP and the Activity of ATPases
ATP is largely the product of oxidative phosphorylation, which can directly provide energy to living organisms. ATP content is typically stable, however, it decreases rapidly when the cell membrane is damaged or cells die; therefore, the ATP content can be used to reflect the number of living cells [41]. The ATP content of the bacteria was negatively correlated with the CSP concentration ( Figure 6A). In B. subtilis, the intracellular ATP content decreased by 8.1% and 70.3% after treatment with 1× and 2× MIC CSP, respectively, compared with the control group. In turn, ATP content decreased by 29.7% and 75.7% after treatment with 1× and 2× MIC of SDA, respectively, compared with the control group. The ATP content in P. fragi and E. coli was similar to that in B. subtilis. Compared with the control groups, the ATP content in the CSP treatment groups decreased significantly (p < 0.05), and the effect on the three bacteria of the 2× MIC high-dose vs. the 1× MIC low-dose was significantly stronger (p < 0.05), showing a concentration-dependent trend. These results showed that CSPs inhibited ATP synthesis, and the bacteria likely died due to insufficient energy supplies. Similarly, the bacteriostatic mechanism of eugenic acid against Cronobacter sakazakii is related to a decrease in intracellular ATP content [42]. Therefore, we concluded that the inhibition mechanism of CSPs on the three bacte included damage to the cell wall and membrane structures, resulting in the aggregat and leakage of AKP, LDH, and other proteins. This damage affected the growth a metabolism of the bacteria and ultimately led to their death due to being unable maintain their original cell morphology.

CSPs Reduce the Content of ATP and the Activity of ATPases
ATP is largely the product of oxidative phosphorylation, which can directly prov energy to living organisms. ATP content is typically stable, however, it decreases rapi when the cell membrane is damaged or cells die; therefore, the ATP content can be us to reflect the number of living cells [41]. The ATP content of the bacteria was negativ correlated with the CSP concentration ( Figure 6A). In B. subtilis, the intracellular A Na + K + -ATPase and Ca ++ Mg ++ -ATPase are proteases in biofilms that maintain the balance of ions inside and outside the cells and catalyze the synthesis and hydrolysis of ATP for energy. Both cell membrane damage and ion leakage affect the energy conversion system associated with the cell membrane [43]. The effects of different concentrations of CSPs on the ATPase activity of B. subtilis, P. fragi, and E. coli showed a trend consistent with that of the ATP content. With increased CSP concentration, the activities of Na + K + -ATPase and Ca ++ Mg ++ -ATPase decreased in the bacteria. Interestingly, in B. subtilis, the Na + K + -ATPase activity (0.20 U/10 4 cell) following treatment with 2× MIC CSP was not significantly different (p > 0.05) from that measured after treatment with 2× MIC SDA. Na + K + -ATPase and Ca ++ Mg ++ -ATPase are proteases in biofilms that maint balance of ions inside and outside the cells and catalyze the synthesis and hydro ATP for energy. Both cell membrane damage and ion leakage affect the energy con system associated with the cell membrane [43]. The effects of different concentrat CSPs on the ATPase activity of B. subtilis, P. fragi, and E. coli showed a trend con with that of the ATP content. With increased CSP concentration, the activities of ATPase and Ca ++ Mg ++ -ATPase decreased in the bacteria. Interestingly, in B. subt Na + K + -ATPase activity (0.20 U/10 4 cell) following treatment with 2× MIC CSP w significantly different (p > 0.05) from that measured after treatment with 2× MIC S The Ca ++ Mg ++ -ATPase activity of the three bacteria was significantly decreased after treatment with different concentrations of CSPs and SDA, compared with that of the untreated group (p < 0.05). Treatment with 2× MIC CSPs significantly decreased enzyme activity by 44.2%, 31.5%, and 52.9%, respectively, compared with that following treatment with 1× MIC of SDA (p < 0.05). These results indicate that CSPs inhibited the activities of Na + K + -ATPase and Ca ++ Mg ++ -ATPase by changing the permeability of the cell membrane, resulting in the failure of the ATP energy supply and the death of the bacterial cells, which may be another mechanism of CSP inhibition. Polyphenolic compounds, such as mulberry pigment, silymarin, baicalin, and silybin, can completely inhibit E. coli ATP syn-thase, whereas hesperidin, kaempferol, apigenin, and rutin have relatively weak inhibitory abilities (approximately 40-60%) [44], indicating that the structure of the phenol hydroxyl group is a key factor in bacteriostasis.

CSPs Inhibit the Activity of Key TCA Cycle Enzymes in Bacteria
The TCA cycle provides biological energy for cell survival by oxidizing nutrients [45].
Two key enzymes of the TCA cycle are α-KGDH and ICDHm. α-KGDH catalyzes the oxidation and decarboxylation of α-ketoglutaric acid to produce succinyl CoA and NADH in the TCA cycle; thus, the enzyme plays an important role in carbon, amino acid, and energy metabolism. In the absence of α-KGDH activity, α-ketoglutarate accumulates, posing a lethal threat to obligate aerobic cells [46]. Meanwhile, ICDHm catalyzes the oxidative decarboxylation of isocitrate to α-ketoglutaric acid and reduces NAD to NADH in the TCA cycle; thus, the enzyme plays an important role in metabolism, synthesis, and antioxidant stress. α-KGDH and ICDHm activity was inhibited in a concentrationdependent manner in the three CSP-treated bacteria ( Figure 6B). Compared with the control group, α-KGDH activity decreased significantly in the three bacteria (B. subtilis; P. fragi, and E. coli.) treated with CSPs or SDA (p < 0.05). Treatment with CSPs at 2× MIC decreased enzyme activity by 22.5%, 50.8%, and 31% compared with SDA treatment at 1× MIC (p < 0.05). The activity of ICDHm was similar to that of α-KGDH. In P. fragi, there was no significant difference in the ICDHm activity in samples treated with 2× MIC CSPs or 1× MIC SDA (p > 0.05). Oregano essential oil can inhibit the respiratory metabolism of S. aureus by affecting the synthesis of metabolic products and the activity of key enzymes of the TCA cycle [47], suggesting that this may be another mechanism of bacterial inhibition by CSPs.

CSPs Regulate the Expression of Key TCA Cycle-Related Genes
In the TCA pathway, detrimental changes in the expression of genes encoding key enzymes may lead to an increase in reactive oxygen species, at which time the signaling pathway associated with apoptosis is activated, ultimately leading to bacterial cell apoptosis [48]. To further investigate the inhibitory mechanism of CSPs on the three tested bacterial species, the expression levels of 12 TCA cycle-related genes were determined using RT-qPCR (Table 3). In B. subtilis, the expression of citA (citrate synthase), icd (ICDHm), odhA (α-KGDH), and atpA (ATP synthase) was significantly downregulated in the CSP-treated groups compared with the control (p < 0.05). The expression of icd was downregulated by 40% after treatment with 1× MIC CSP and 70% (maximum extent) at 2× MIC, compared with the control group (p < 0.05).
In P. fragi, the expression of gltA (citrate synthase), IS178_RS16090 (ICDHm), IS178_RS16290 (α-KGDH), and atpA (ATP synthase) was significantly downregulated in the CSP-treated groups compared with the control group (p < 0.05). The expression of IS178_RS16290 was downregulated by 49% after treatment with 1× MIC CSP, and the expression of gltA was downregulated by 71% after treatment with 2× MIC CSP, compared with the control.
In E. coli, the expression of gltA (citrate synthase), icd (ICDHm), sucA (α-KGDH), and atpA (ATP synthase) was significantly decreased in the CSP-treated groups by 17-65% compared with the control (p < 0.05). The downregulation of the four genes after treatment with 2× MIC CSP was superior to that following treatment with 1× MIC SDA, and the antibacterial effect was stronger. Among the four genes, CSPs showed the strongest inhibitory effect on sucA, which encodes α-KGDH.
The gene expression results indicate that CSP treatment mainly affected α-KGDH and ICDHm in the TCA cycle, resulting in abnormal gene expression and the inhibition of respiratory metabolism. This suggests that CSPs can destroy the bacterial cell structure at the molecular level by affecting gene expression and key enzyme activity, resulting in energy metabolism obstruction and cell death.

Molecular Docking Simulation of Major CSPs Components and Key TCA Cycle Enzymes
Simulated models of molecular docking for PB1 and PC1 with IDH1 and OGDH were used to study the main components and potential mechanism of the inhibition of CSPs on cell membrane dehydrogenase activity. The docking scores of PB1 with isocitrate dehydrogenase and α-ketoglutarate dehydrogenase were −11.7211 kcal/mol and −11.2414 kcal/mol, respectively. The docking scores of PC1 with isocitrate dehydrogenase and α-ketoglutarate dehydrogenase were −14.0050 kcal/mol and −12.8227 kcal/mol, respectively. The more negative the docking score, the better the binding affinity between the ligand and receptor [20]. The results showed that PB1 and PC1 could bind tightly to the bacterial cell membrane enzymes IDH1 and OGDH.
OGDH forms hydrogen bonds, as well as metal-acceptor, hydrophobic, and VDW interactions with PB1 and PC1 ( Figure 7C,D). PB1 forms hydrogen bonds with OGDH chain A residues Thr450, Ser375, and Leu377, and the residue Gln746 in OGDH chain B. PB1 and PC1 form metal receptor interactions with OGDH chain A residue Mg 2+ ion. PB1 forms hydrophobic interactions with OGDH chain A residues Phe278, His513, Leu377, and Ala412, and the residues Phe750 and Leu720 in OGDH chain B. PB1 and PC1 form VDW interactions with surrounding residues in OGDH. PC1 forms hydrogen bond interactions with OGDH chain A residues Arg456, Ser375, His311, and Asp411, and the residues Leu720 and Gln746 in OGDH chain B. PC1 forms hydrophobic interactions with OGDH chain A residues His790, Gly447, Asp451, Leu377, Ala412, Pro452, and Ala455 and OGDH chain B residue Leu720. The binding energies of OGDH to PB1 and PC1 were determined primarily by these interactions.
PB1, PC1, and other proanthocyanidins are polyphenolic flavonoids that inhibit the growth and metabolism of microorganisms. Proanthocyanidins can reduce the stability of bacterial cell membranes by inhibiting β-lactamase activity, thereby inhibiting microbial growth [49]. PB1 and PC1, the main components of CSPs, played important roles in destroying bacterial cell membranes and inhibiting bacterial growth in this study.

Conclusions
The extracted CSPs exhibit antibacterial activity against B. subtilis, P. fragi, and E. coli, among which, B. subtilis is the most susceptible, and the inhibitory effect on gram-positive bacteria is stronger than that on gram-negative bacteria. Our results suggest the following bacteriostatic mechanism of CSPs: the bacterial cell wall and plasma membrane are damaged, cellular structural integrity is diminished, the permeability barrier of the membrane is reduced, and the leakage of cell contents increases. These processes lead to the inhibition of TCA cycle-related gene expression, resulting in the reduced activity of key enzymes, hindered respiration and energy metabolism, and the inhibition of ATP synthesis, thus, inhibiting cell proliferation and leading to bacterial cell death. PB1 and PC1 are the main components of the CSPs that exert antibacterial effects. At the same dose, the bacteriostatic effect of the CSPs was superior to the positive control SDA. Hence, collectively, this study demonstrates the potential of CSP extract as a low-price natural food preservative that can alleviate the harmful effects of synthetic chemical preservatives, reduce the bacterial contamination of food, and provide a theoretical basis for the high-value-added use of CSPs. However, while this study demonstrates that CSPs have an inhibitory effect on bacteria, it remains necessary to determine whether they exert an inhibitory effect on molds and yeasts. Moreover, it is necessary to evaluate the dose safety for adding CSP extracts to food products and to conduct toxicology analyses.

Data Availability Statement:
The data used to support the findings of this study can be made available by the corresponding author upon request.