Evaluation of the Adhesive Potential of Bacteria Isolated from Meat-Related Sources

Microbial adhesion constitutes the transition of microorganisms from a planktonic mode to a static one. It promotes the formation of biofilm which is responsible for spoilage, foodborne diseases, and corrosion in the food processing industry. In this study, the adhesive potential of fourteen meat-borne bacterial isolates belonging to seven different genera was investigated. All strains were found able to colonize polystyrene surfaces with different levels of firmness. Significant variations were determined in assays of bacterial hydrophobicity and motility. Among the 14 strains, Pseudomonas fragi, Aeromonas salmonicida II, Serratia liquefaciens, Citrobacter braakii, Pseudomonas putida, and Aeromonas veronii had a strong hydrophobic force, while the isolates of Lactobacillus genus showed the most hydrophilic property. In terms of motility, Citrobacter braakii and Escherichia coli exhibited exceptional swarming and swimming abilities, whilst conservatively weak performances were observed in the Lactobacillus strains. Furthermore, the majority of the isolates were predominantly electron donors and weak electron acceptors. Overall, a high level of correlation was observed between biofilm-forming ability with cell surface hydrophobicity and Lewis acid–base properties, whereas the contribution of motility in bacterial adhesion could not be confirmed. Research on the adhesive performance of foodborne bacteria is potentially conducive to developing novel control strategies, such as food processing equipment with specific surfaces, not facilitating attachment.


Introduction
Biofilms are biologically active communities characterized by cells embedded in a matrix of extracellular polymeric substances, in association with a solid surface. The existence of biofilm is ubiquitous in nature because bacteria tend to favor a community-based lifestyle compared to planktonic conditions. Biofilms are culprits of serious engineering problems as well as cross-contamination in the food industry [1]. Microbial adhesion is the first step of biofilm formation, which mainly consists of two stages: the "docking" and the "locking" phases [2,3]. Primary adhesion constitutes the reversible attachment of a planktonic bacterial cell on a conditioned surface, whose occurrence depends on the attractive and repulsive forces, including electrostatic and hydrophobic interactions between the microorganism and the contact surface. It also requires motility property generated by flagella, for movement and direction [4,5]. At the point of the secondary stage, molecularly mediated binding is employed through interactions of extracellular polymers with surface materials and/or cell receptors. This process consolidates the adhesion to an irreversible level that debuts the biofilm maturation.
LABs are mostly known for their useful technological properties in the production of fermented foods, but they could also be involved in the spoilage of anaerobically packed products [6]. Enterobacteriaceae is a capitalized group that commonly contributes to worldwide outbreaks resulting from the consumption of contaminated food, represented by Escherichia coli. Further, it contains species such as Aeromonas and Serratia as frequently mentioned spoilage organisms in various food products [7,8]. Pseudomonas, belonging to the family Pseudomonadaceae, consists of some spoilage strains that also arouse primary attention in the food industry [9]. It is known that the bacterial effects are performed through cross-contamination caused by the presence of biofilms. Among the bacteria mentioned above, Pseudomonas spp. and Escherichia coli have been the protagonists in numerous reports regard to their adhesive ability on various contact surfaces [10,11]. However, few studies have been conducted on adhesion with the other bacteria; in addition, the relationship between biofilm-forming capacity and cell surface characteristics has not yet been evaluated. For the reasons mentioned above, fourteen isolates from meat-related sources, with Pseudomonas, Aeromonas, Serratia spp., and Escherichia coli were investigated in the present study.
The aim of this study was thus (i) to determine the adhesive performance of fourteen isolates from meat-borne sources and characterize individually their cell properties, including surface hydrophobicity, swimming and swarming motility, and electrondonor/electron-acceptor properties; (ii) to inquire into the general relationship between adhesion ability and cell surface characteristics. The study aims to enrich the theoretical framework of microbial biofilm formation and provide potential information for the further development of bacterial control strategies.

Material and Methods
TSA, TSB, and MRS were purchased from Land Bridge Technology Co., Ltd. (Beijing, China). Crystal violet was purchased from Tianjin Chemical Reagent Research Institute (Tianjin, China). Other reagents and chemicals were obtained from Hope Bio-Technology Co., Ltd. (Qingdao, China) and were of analytical grade.

Bacterial Strains and Inoculation
Fourteen food-borne bacteria were tested in this study (Table 1), namely Aeromonas veronii, Aeromonas salmonicida I, Aeromonas salmonicida II, Serratia liquefaciens, Citrobacter braakii, Rahnella aquatilis, Escherichia coli, Pseudomonas putida I, Pseudomonas putida II, Pseudomonas fluorescecs, Pseudomonas fragi, Lactobacillus plantarum, Lactobacillus curvatus, and Lactobacillus sake. They were collected from food manipulated in Key Laboratory of Meat Products Processing, Nanjing Agricultural University, China. The strains stored at −80 • C were each revived twice on TSA plates or MRS plates for the Lactobacillus isolates. A single colony of all the strains was selected and cultured in TSB or MRS broth, and then incubated at 30 • C for 18 to 24 h, allowing the concentration of bacteria to reach 10 8 CFU mL −1 . The activated bacterial culture was then used to prepare a bacterial suspension. An aliquot (200 µL) of each culture was transferred into fresh TSB or MRS, respectively, and incubated at 30 • C for 20 to 24 h. Cells were harvested by centrifugation (Avanti J-E, Beckman Coulter, Brea, CA, USA) at 12,000× g for 5 min and then washed three times with 0.1% BSP. The pellet was resuspended using 0.1% BSP to a final concentration adjusted to 10 9 CFU mL −1 .

Biofilm Assay
The biofilm assay was determined as previously described with slight modifications [12]. An aliquot (180 µL) of prepared bacterial suspension (10 9 CFU mL −1 ) was dispensed into each well of a 96-well polystyrene microplate (Corning, New York, NY, USA). Negative controls were obtained by preparing both broths devoid of bacterial inoculum. Wells were incubated at 20 or 30 • C for 24 or 72 h, then rinsed three times with 0.1% BSP and air-dried for 45 min. Individual wells were stained with 200 µL of 0.25% (m/v) crystal violet for 30 min, then washed three times in 0.1% BSP to remove excessive staining solution. The crystal violet bound to biofilms was solubilized with 0.2 mL of 95% ethanol for 30 min. By measuring the absorbance at 570 nm via a SpectraMax M2 Microplate Reader (Molecular Device, San Jose, CA, USA), the biofilm-forming ability was determined. Five biological replicates were performed in this assay.

Motility Assay
Swimming and swarming motility were evaluated using soft-agar plate assays [13]. The media prepared for the swimming assay consisted of 10 g/L tryptone, 5 g/L NaCl, 2.5 g/L glucose, and 0.3% agar, while that of the swarming assay contained 25 g/L Luria-Bertani, 0.5 g/L glucose, and 0.5% agar. Three microliter aliquots of each bacterial suspension were spotted onto the surface of soft-agar plates, then placed at room temperature for 20 min in order to obtain a better absorption of the inoculum. After incubation at 37 • C for 8 h (swimming) or 20 h (swarming), the motility was recorded by measuring the diameter (mm) of the strain diffusion circle. Six biological replicates were performed in this assay.

Cell Surface Hydrophobicity and Electron-Donor/Acceptor Properties
Microbial adhesions to solvents were measured referring to the previous methods with some modifications [14]. Two milliliters of bacterial suspension was put into contact with an equal volume of the following solvents: xylene (apolar solvent), ethyl acetate (monopolar basic solvent and electron donor) and chloroform (monopolar acidic solvent and electron acceptor), respectively. Once mixed on a vortex for 2 min, the 2-phase system was stored at room temperature for 15 min allowing the hydrocarbon phase to rise completely. The percentage of microbial adhesion to the solvent was calculated as [(A 0 − A 1 )/A 0 ] × 100, where A 0 is the initial OD 600 of the cell suspension and A 1 represents the final OD 600 of the aqueous phase. Each parameter was performed in six biological replicates.

Statistical Analysis
Statistical significance was determined by SPSS 18.0 software (SPSS Inc., Chicago, IL, USA) using a Duncan (D)'s ANOVA at the 0.05 level of significance, and the coefficient ( Table 2)

Results and Discussion
Bacterial biofilm formation is the source of microorganism cross-contamination in the food industry. This study involved 14 meat-borne bacteria for the biofilm-forming assay. Since the optimal temperature of each genus varies, an average of 30 • C was proposed to be used for coordination, while 20 • C was applied in order to simulate a lower environmental temperature that occurs during processing in some food facilities. Both short-term and long-term formations of biofilm were investigated by adjusting the incubation time to 24 h and 72 h. Results showed that all of the bacteria were able to attach to the polystyrene surface with a variation which revealed the differences among genera and within species (Figure 1). A. salmonicida displayed the best adhesive performance, regardless of time and temperature, with a consistency between the two isolates (A. salmonicida I and II), while a moderate amount of biofilm was exhibited by A. veronii, indicating a speciesspecific distinction. A similar finding was observed when studying the adhesive strength of 14 Aeromonas trains [15]. (Dias et al., 2018). Except P. putida I, all strains of Pseudomonas (including P. putida II) formed considerable amount of biofilms at 20 • C for 24 h, whereas at 72 h, a higher temperature was more favorable for attachment of these bacterial cells. The different behaviors presented by P. putida I and P. putida II were in contrast to the case of A. salmonicida, this phenomenon demonstrated that isolates of one species are not necessarily supposed to perform in the same way. As for the other strains tested, a relatively strong biofilm-forming capacity was obtained in both C. braakii and S. liquefaciens at 30 • C for 24 h or 72 h, respectively, while the Lactobacillus genus was so "modest" that it formed the least amount of biofilm whether it was on the first day or the third day. This might result from the alternation of substratum. Lactobacillus is recognized as inhabitants in the human gastrointestinal tract where surface proteins can take advantage of mediating adhesion with the help of mucus [16]. However, the polystyrene surface used for the present in vitro experiment is of significant difference to the epithelial one, so that related proteins could be limited to play their original utility. The performance of biofilm formation was temperature-dependent. At 24 h, nine bacteria among the fourteen displayed a high attachment under 20 • C compared to 30 • C, especially R. aauqtilis, E. coli, and the Pseudomonas species, showing a significant advantage. This phenomenon was probably due to the diversity of optimal temperature of the strains engaged. A large cell density in culture broths results possibly in providing a comfortable environment for bacterial communication, thus enhancing the attachment of bacteria [17]. The dynamic process of attachment evolved as time went on and 48 h later, and a great variation was observed at 72 h. More bacteria showed a stronger biofilm-forming ability at 30 • C, especially S. liquefaciens, Pseudomonas isolates putida II, fluorescecs, and fragi, which formed markedly improved amounts of biofilm and even exceeded that on the first day. Meanwhile, A. veronii maintained steady and a slightly fluctuated gap of biofilm amount between the two temperatures tested. Additionally, less biofilm of few isolates represented by A. salmonicida and C. braakii was observed at 72 h than 24 h, probably due to the biofilm dispersal at the end of incubation periods. The results above indicated that incubation time had an influence on the biofilm-forming ability. One complete development cycle of biofilm formation includes dispersion as the last step, which promotes the detachment of cells from the mature biofilm [18]. Moreover, this phenomenon results from the starvation of cells in regard to limited perfusion of nutrients and restricted removal of waste [19]. The period during which bacteria showed an impaired adhesion ability in the present study might refer to this development phase. Under 20 • C, the biofilm formed by the majority of the isolates at 24 h (e.g., A. salmonicida and Pseudomonas) grew over time to mature and disperse, leading to a decrease in the amount of biofilm at 72 h. While under 30 • C, the development of biofilm might be delayed so that at 24 h, the isolates seemed only to colonize, and the biofilm was not formed in a large quantity until 72 h. This partly proved that the lower temperature was preferable for most of the involved bacteria. Nevertheless, attention should be given to the existence of particularity in the biofilm-forming pace of individual strains.
Appl. Sci. 2021, 11, x FOR PEER REVIEW 5 of 9 indicated that incubation time had an influence on the biofilm-forming ability. One complete development cycle of biofilm formation includes dispersion as the last step, which promotes the detachment of cells from the mature biofilm [18]. Moreover, this phenomenon results from the starvation of cells in regard to limited perfusion of nutrients and restricted removal of waste [19]. The period during which bacteria showed an impaired adhesion ability in the present study might refer to this development phase. Under 20 °C, the biofilm formed by the majority of the isolates at 24 h (e.g., A. salmonicida and Pseudomonas) grew over time to mature and disperse, leading to a decrease in the amount of biofilm at 72 h. While under 30 °C, the development of biofilm might be delayed so that at 24 h, the isolates seemed only to colonize, and the biofilm was not formed in a large quantity until 72 h. This partly proved that the lower temperature was preferable for most of the involved bacteria. Nevertheless, attention should be given to the existence of particularity in the biofilm-forming pace of individual strains. Cell adhesion requires contact with the bacterial cell envelope and interacting surface. Motile properties are involved in the meeting of these two objects at the docking stage during the multistep process of adhesion. Two types of motility were investigated in this study: swimming and swarming. The former is a characteristic related to individual movement while the latter shows the movement of a bacterial population. Great variability among genera was observed (Figure 2). Significant distinction existed of a certain single strain (P. putida) as well. E. coli and C. braakii showed a superior motility capacity. In the swimming media, they developed a relatively large strain diffusion circle whose diameter was over 75 mm, three times as large as that of the second echelon. P. putida Ⅱ, R. aquatilis, and S. liquefaciens exhibited moderate swimming capacity with a diameter of around 20 mm. Less disparity was found in the results of the swarming assay. The five strains mentioned above also possessed a strong swarming ability but were much more reserved compared to their swimming performances. This finding suggested that the fluidity of the medium could possibly influence the diffusion of the microorganisms, a flowing state might favor the cross-contamination in food processing as subsequent biofilm development for instance, whereas no remarkable differences between the two Cell adhesion requires contact with the bacterial cell envelope and interacting surface. Motile properties are involved in the meeting of these two objects at the docking stage during the multistep process of adhesion. Two types of motility were investigated in this study: swimming and swarming. The former is a characteristic related to individual movement while the latter shows the movement of a bacterial population. Great variability among genera was observed (Figure 2). Significant distinction existed of a certain single strain (P. putida) as well. E. coli and C. braakii showed a superior motility capacity. In the swimming media, they developed a relatively large strain diffusion circle whose diameter was over 75 mm, three times as large as that of the second echelon. P. putida II, R. aquatilis, and S. liquefaciens exhibited moderate swimming capacity with a diameter of around 20 mm. Less disparity was found in the results of the swarming assay. The five strains mentioned above also possessed a strong swarming ability but were much more reserved compared to their swimming performances. This finding suggested that the fluidity of the medium could possibly influence the diffusion of the microorganisms, a flowing state might favor the cross-contamination in food processing as subsequent biofilm development for instance, whereas no remarkable differences between the two motility-related properties appeared in P. fluorescens, A. veronii, and P. putida I. Both swimming and swarming motility are flagella-dependent. It was reported that flagellar motility played an essential role in assisting microorganisms to reach proximate to the surface at the early beginning of adhesion [20]. The wide range of performances among strains in the present study might mainly result from the diversity of flagellar characteristics, but also other infectors such as biosurfactant production and QS signaling [21,22]. However, all bacteria with remarkable motile capacity were not strong in forming biofilms, taking E. coli and C. braakii as examples. Furthermore, no correlation was determined in this study between motility and biofilm formation from a general view (Table 2). This finding suggested that the role of flagella-based motility might not be of critical importance to the attachment of microorganisms and it might alter the adhesion process of different bacteria. In spite of the transition from planktonic mode to biofilm growth mode promoted by motility, it might keep cells away from the surface through active movements in the meantime. Moreover, many other factors (e.g., surface interactions) exist contributing to the adhesion process. Nevertheless, the function of flagella cannot be ignored. It was found by video microscopy that the flagellum was required for rapid attachment of Vibrio cholerae to a surface, and flagellar mutant was defective in initial attachment to abiotic surfaces [23]. (Watnick et al., 2001). In addition, if focusing alone on the tested Lactobacillus spp., all three isolates had the weakest motility capacity, which is reasonable because lactobacilli rarely carry flagella and move with them. Their consistently poor abilities displayed in both adhesion and motility assays provide a hint to the intrinsic connection between these two features.
Appl. Sci. 2021, 11, x FOR PEER REVIEW 6 of 9 motility-related properties appeared in P. fluorescens, A. veronii, and P. putida Ⅰ. Both swimming and swarming motility are flagella-dependent. It was reported that flagellar motility played an essential role in assisting microorganisms to reach proximate to the surface at the early beginning of adhesion [20]. The wide range of performances among strains in the present study might mainly result from the diversity of flagellar characteristics, but also other infectors such as biosurfactant production and QS signaling [21,22]. . However, all bacteria with remarkable motile capacity were not strong in forming biofilms, taking E. coli and C. braakii as examples. Furthermore, no correlation was determined in this study between motility and biofilm formation from a general view (Table 2). This finding suggested that the role of flagella-based motility might not be of critical importance to the attachment of microorganisms and it might alter the adhesion process of different bacteria. In spite of the transition from planktonic mode to biofilm growth mode promoted by motility, it might keep cells away from the surface through active movements in the meantime. Moreover, many other factors (e.g., surface interactions) exist contributing to the adhesion process. Nevertheless, the function of flagella cannot be ignored. It was found by video microscopy that the flagellum was required for rapid attachment of Vibrio cholerae to a surface, and flagellar mutant was defective in initial attachment to abiotic surfaces [23]. (Watnick et al., 2001). In addition, if focusing alone on the tested Lactobacillus spp., all three isolates had the weakest motility capacity, which is reasonable because lactobacilli rarely carry flagella and move with them. Their consistently poor abilities displayed in both adhesion and motility assays provide a hint to the intrinsic connection between these two features. Surface characteristics such as hydrophobicity are also estimated to be able to influence the outcome of primary adhesion. Bacterial surfaces with hydrophilic properties were reported to be beneficial for electroactive biofilm formation [24]. These factors are usually species and strain-dependent. The differences in affinity may be due to the composition and content of different lipopolysaccharides, capsular material, or appendages on the cell surface. In our study, xylene, which is an apolar solvent, was used Surface characteristics such as hydrophobicity are also estimated to be able to influence the outcome of primary adhesion. Bacterial surfaces with hydrophilic properties were reported to be beneficial for electroactive biofilm formation [24]. These factors are usually species and strain-dependent. The differences in affinity may be due to the composition and content of different lipopolysaccharides, capsular material, or appendages on the cell surface. In our study, xylene, which is an apolar solvent, was used to determine cell surface hydrophobicity, and the expression of cell affinity to this solvent varied significantly among strains evaluated (Table 1). High affinity to xylene was detected in Pseudomonas and Aeromonas genera (with exception of isolates P. fluorescens and A. salmonicida I), represent-ing an advantageous hydrophobic ability, while the Lactobacillus strains showed a strong hydrophilic property, with an affinity value to xylene under 20%, and even lower than 5% in L. curvatus. Furthermore, S. liquefaciens and C. braakii were another two competitors in hydrophobicity. The existing distinctions in this assay might be a consequence of varied envelope molecules, especially proteins and polysaccharides. Some evidence showed that the attendance of protein was likely to promote hydrophobic behavior whereas the participation of polysaccharides may result in better hydrophilic capacity [25,26]. General correlation was evident between cell surface hydrophobicity and biofilm formation ( Table 2). Bacteria that possessed high hydrophobic surfaces such as Pseudomonas and Aeromonas attached readily to the model surface, while Lactobacillus strains that were more hydrophilic displayed poor biofilm-forming ability. It is in line with the finding of an investigation into surface characteristics of 50 Lactococcus lactis strains that more hydrophobic isolates were found to adhere more to polystyrene [27]. This would suggest a positive correlation between biofilm formation on polystyrene and the existence of attractive van der Waals interactions at the interface of hydrophobic cells and the inert material.
Many studies focusing on the fundamental mechanisms governing microbial adhesion have proved the importance of Lewis acid-base, i.e., electron-donor/electron-acceptor, interactions [28,29]. In the present study, apart from xylene, bacterial affinities to another two solvents, namely ethyl acetate and chloroform, were investigated to determine Lewis acidbase properties (Table 1). Among the fourteen stains, only L. curvatus showed a slightly stronger affinity for ethyl acetate than for chloroform, implying an electron-accepting characteristic. For L. sake, equal affinity was observed to both solvents. L. plantarum was predominantly an electron donor but a weak electron acceptor referring to its better affinity for chloroform. The three strains of Lactobacillus exhibited completely different behaviors in this test. However, their affinity values were maintained at a consistently low level. This is in line with all passive performances in regard to this genus through the previously mentioned assays. All other strains displayed a better affinity for chloroform than for ethyl acetate, indicating their possession of a superior ability to donate electrons and a basic character. Among them, Pseudomonas and Aeromonas spp. had an affinity value for chloroform calculated higher than 70% (surpassing 90% for P. fragi and P. putida II), which made them the strongest electron donors, and were followed by S. liquefaciens and R. aquatilis with a value of around 60%, concerning the same indicator. Different from the outstanding performances in the swarming test (Figure 2), both C. braakii and E. coli showed low potential for electron-related affinities. Taking all the 14 isolates into consideration, the electron donor/acceptor properties were correlated with the biofilm-forming process of engaged microorganisms ( Table 2). On the whole, isolates with a better capacity to donate electrons (basic character) attached relatively easier to the polystyrene surface. It was demonstrated that Lewis acid-base interactions played a major role compared to electrostatic and van der Waals forces in the process of biofilm development [30]. However, these are not in line with a finding when investigating cell surface properties of marine bacteria, which suggested that the contribution of an electron-donating character in bacterial adhesion was at a low level [31]. (Grasland et al., 2003). The disagreement might result from the difference in the source of bacteria.

Conclusions
In conclusion, the 14 foodborne isolates tested in our study were capable of colonizing the polystyrene surface. Moreover, it was demonstrated that the biofilm formation was significantly influenced by hydrophobicity and electron-donor/electron-acceptor interactions. However, bacterial motility could not be considered as playing a constructive role in the adhesion process. Consistent performances were detected within genus or species for these cell surface characteristics. It would be advisable to conduct future research that engages in other bacteria from a food-associated source.