Listeria monocytogenes ATCC19112 used in this study was purchased from the China Microbial Preservation Center (Beijing, China). The strain of PL 28 was previously isolated from chilled beef in our laboratory (Fang et al., 2022). These strains stored at −80°C were revitalized on tryptic soy broth (TSB, Bio-Tech, Qingdao, China) at 30°C with consecutive transfers after 24 h shaking incubation (180 rpm).

Briefly, overnight activated cultures of the two strains were diluted at a ratio of 1:1,000, with 5.0 log cfu/ml as the final cell concentration for each species. The diluted LM and PL were inoculated separately or mixed in tube or polystyrene plates with TSB, and the inoculated samples were cultivated statically for 24 and 48 h at 30 and 15°C. For enumeration of planktonic cells, bacterial suspension as mono- or dual species from each tube was serially diluted with sterile peptone water (0.85% NaCl and 0.1% peptone), and then aliquots of 0.1 ml were spread on Pseudomonas CFC Selective Agar (Bio-Tech, Qingdao, China) for PL and PALCAM Agar (Bio-Tech, Qingdao, China) for LM. For enumeration of the biofilm cells, mono- and dual-species biofilm cells were cultivated on the surface of stainless steel coupons (SS, AISI304, 1 cm × 1 cm × 0.2 cm) in six-well polystyrene plates (one coupon in each well) (Wang et al., 2020). The coupons were soaked in 70% ethanol for 5 min and autoclaved at 121°C for 15 min prior to the experiment. Wells containing TSB alone were considered as control. After incubation at 30 and 15°C, SS coupons were taken out and washed three times with phosphate buffered saline (PBS, pH 7.0) to release loosely attached cells. Subsequently, the biofilm cells on the SS surfaces were obtained by vigorously vortexing in 0.1% peptone water for 5 min. The homogenized suspension was then diluted and enumerated on the above two selective agar plates, and the total number of colonies was counted after incubation at 30°C for 24 h.

The activated LM and PL as mono- or dual-species biofilms were quantified using crystal violet (CV) assay. Briefly, the above diluted cultures were inoculated in 96-well polystyrene microtiter plates as mono- or co-culture, and uninoculated TSB was used as negative control. Following incubation, the biofilm in each well of microtiter plates was carefully washed thrice with sterile PBS to remove unattached cells. Biofilm cells on the bottom and side of each well were stained with 0.2% (w/v) CV for 15 min. After a second washing step, biofilm cell-associated CV in each well was resolubilized with 95.0% ethanol (v/v) for 5 min. The optical absorbance of the samples was measured at 590 nm using a microplate reader (Infinite 200, Tecan, Switzerland).

Extracellular polysaccharides in biofilm were extracted according to Wang et al. (2020). The above diluted LM and PL containing about 5 log cfu/ml of the two strains as mono- or dual species were transferred into six-well polystyrene microtiter plate and incubated at 30 and 15°C for 24 or 48 h. After the cultures in each well were carefully aspirated, the biofilm cells were washed with sterile PBS, redissolved with PBS, and sonicated at 50 kHz for 5 min. Bacterial pellets were then removed by centrifugation (8,000 × g for 30 min) and filtering by a 0.22 μm membrane, and the extracting supernatant was added with 5% phenol and sulfuric acid and bathed at 100°C for 10 min. The absorbance value was determined at 490 nm to quantitatively measure exopolysaccharides using glucose as the standard.

The biofilms formed by PL and LM as mono- or dual species were further observed by optical microscopy (OM), confocal laser scanning microscopy (CLSM), and scanning electron microscopy (SEM). After incubation for 24 h at 30°C, the glass slips (8 mm × 8 mm) in six-well polystyrene microtiter plates were taken out and rinsed with PBS three times to remove the planktonic bacteria. The biofilm cells on the surface were then stained with 0.2% (w/v) CV and directly observed using an OM (Olympus CX21, Tokyo, Japan). Meanwhile, the biofilm structures of mono- or dual species were examined using CLSM with objective lens (Carl Zeiss LSM710, Jena, Germany) after stained with 0.3% SYTO-9 and 0.3% PI (Sigma-Aldrich, Shanghai, China) in the dark at 30°C for 15 min. The maximum excitation/emission used for these stains was approximately 505/525 for SYTO-9 (green channel) and 493/635 for PI (red channel). Representative CLSM images from each coupon were acquired by scanning z-stacks at a scanning step size of 1 μm and were processed in IMARIS 7.6 software (Bitplane AG, Zurich, Switzerland). Three representative areas were selected by scanning the total surface area with a 20× objective. The Zeiss confocal software was used to analyze the biofilm images, allowing for collection of z-stacks. Quantitative structural parameters of the biofilms, including bacterial biovolume and thickness, were calculated using PHLIP (Almeida et al., 2003).

In addition, the biofilms formed by PL and LM as mono- or dual species were further observed by SEM. After the removal of culture media and washing, the glass slips incubated for 24 h at 30°C were collected and fixed with ice-cold 2.5% glutaraldehyde overnight at 4°C. After washing, biofilm cells on glass slip were post-fixed with 1% osmium tetroxide for 1 h. The samples were washed in PBS (pH 7.0) and dehydrated with a graded series of ethanol solutions from 30 to 100% for about 15 min at each step. The dehydrated samples were coated with gold-palladium and observed using SEM (Hitachi SU8010, Japan).

ε-Polylysine hydrochloride (>90% pure, w/w, Silver-Elephant Bio-engineering Co., Ltd., Zhejiang, China) was prepared as a stock solution (20 mg/ml) in distilled sterile water and diluted to work solution (5 mg/ml). First, minimal inhibitory concentration (MIC) and maximum sublethal concentrations (MSC) values of ε-PLH against PL and LM in TSB were obtained using the microwell dilution assay (Wang et al., 2020). MIC was defined as the lowest concentration of the antimicrobial that inhibited >90% of the growth of the tested microorganism (Oo et al., 2010). MSC was considered as the last tested concentration that allowed bacterial growth compared with control (Di Pasqua et al., 2006). Then, the diluted cultures of LM and PL were inoculated separately or mixed in TSB, which were supplemented with ε-PLH at the different concentrations of 0, 2, 4, 8, 16, 32, 64, and 128 μg/ml at the initial time. After incubation at 30°C for 24 h, the populations of biofilm cells as mono- or dual species were counted as mentioned in the “Counting of Planktonic and Biofilm Cell Populations” section, and biofilm biomass and extracellular polysaccharides were determined using the above CV assay and phenol-sulfuric acid method as described in the “Crystal Violet Biofilm and Extracellular Polysaccharides Assay” section, respectively.

The inhibitory activity of ε-PLH in combination with CEO on the formation of dual biofilms of LM and PL was further analyzed. CEO was obtained from Borui Spice Oil Co., Ltd. (>95% pure, v/v, Jiangxi, China). The diluted LM and PL cultures were mixed in equal volumes for dual-species inoculation in fresh TSB, with 5.0 log cfu/ml as the final cell concentration for each species. The two preservatives at the MSCs using alone or combination, including ε-PLH at 8 (P8) or 16 μg/ml (P16), CEO at 25 (C25) or 50 μg/ml (C50), and four combined groups (P8 + C25, P8 + C50, P16 + C25, P16 + C50), were supplemented to wells of the microtiter plate. After incubation at 30°C for 24 h, biofilm biomass and extracellular polysaccharides of the multispecies biofilms were determined. Then, CLSM and SEM were utilized to observe the effect of ε-PLH combined with CEO on the spatial structure of the dual biofilms.

In addition, ε-PLH and CEO were further tested against 24 h performed dual biofilms formed by LM and PL in microtiter plates for their ability to reduce the CV biomass (biofilm removal, %) and metabolic activity (biofilm inactivation, %). Briefly, the 24 h preformed biofilms in each well of 24-microtiter plates were carefully removed and washed with PBS (pH 7.0) to remove unattached cells. ε-PLH and CEO either alone or combination at high than MICs were supplemented to the preformed biofilm in the microtiter plates, including ε-PLH at 64 (P64) or 128 μg/ml (P128), CEO at 50 (C50) and 100 μg/ml (C100), four combined groups (P64 + C50, P64 + C100, P128 + C50, P128 + C100), and sterile peptone water as control. After treatment at 30°C for 30 min, the biomass of inhibitors exposed and non-exposed biofilms was quantified by CV staining. The percentage of biomass removal (% BR) was determined by Equation 1, where is the OD value for preservative non-exposed biofilms (control, C) and T is the OD value for inhibitor exposed biofilms.

The metabolic activity of preservative exposed and non-exposed biofilms was evaluated using XTT [2,3-bis (2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carbox-anilide] assay kit (Jiangsu Kaiji Biotechnology Co., Ltd., Nanjing, China). After incubation at 37°C for 4 h in the dark, the absorbance in each well was measured at 450 nm. The percentage of biofilm inactivation (% BI) was determined according to Equation 2, where C is the absorbance for control and T is the absorbance for inhibitor exposed biofilms.

Additionally, the effects of ε-PLH combined with CEO on the spatial structure of the 24 h performed dual biofilms were further examined by CLSM.

According to Bodor et al. (2008), autoinducer-2 (AI-2) activities of PL and LM as mono- or dual species were assayed using Vibrio harveyi BB170. Overnight culture of V. harveyi BB170 was diluted (1:5,000) in a fresh AB medium, and the diluted cells were dispensed into 96-well microplates, fresh AB medium as blank control. After incubation for 24 h at 30°C, the supernatant of mono and co-culture treated with ε-PLH and CEO at the MSCs was centrifuged (10,000 × g, 3 min), and then filtered through a 0.22 μm membrane filter. Each cell-free filter was added to each well of microplates containing the diluting BB170 strain. After incubated at 30°C with shaking, the bioluminescence of V. harveyi BB170 was monitored every 30 min using a luminometer (PerkinElmer Victor X, PerkinElmer Inc., Waltham, MA, United States), continuous for 6 h.

All the experiments were performed in three biological replicates, and the mean and standard deviation of experimental values were calculated. The figures were processed using Prism 8.0 software (GraphPad Software Inc., La Jolla, CA, United States), and statistical significance was assessed by one-way ANOVA method using SPSS 17.0 software package (SPSS Inc., Chicago, IL, United States). Significant differences among experimental groups of mono- and dual cultures treated with or without ε-PLH and CEO are expressed as p values of <0.05* and <0.01^**.

The planktonic and biofilm cell population of LM and PL as mono- and dual-species cultures incubated at 30°C or 15°C are described in Table 1. The planktonic cell number of single LM reached 9.0 log cfu/ml during the stationary phase, which was 0.2 log cfu/ml higher than in the co-culture (p > 0.05). The planktonic population of PL in both mono- and dual-species cultures reached about 9.0 log cfu/ml in TSB. In addition, LM and PL strains were able to form mono-specie biofilm on the surface of SS, but their biofilm cell count of each strain was significantly lower than the corresponding planktonic cell (p < 0.05). In dual-species biofilms, PL attained approximately 0.6 log cfu/cm² higher cell density compared to LM, thus constituting approximately 85% of the total population both at 30 and 15°C. Moreover, there was significant difference of LM or PL biofilm cell count between mono- and dual-species biofilms (p < 0.05). Biofilm cell densities of LM in dual-species biofilms reached 6.47–6.58 log cfu/cm² at 30 and 15°C, in contrast to about 7.0 log cfu/cm² in mono-species biofilm (p < 0.05). The dual-biofilm cell populations of PL were 7.26–7.40 log cfu/cm² during mature phase, which were slightly lower than mono-species-attained numbers of biofilm cells (p < 0.05).

Moreover, PL showed significantly higher quantities of biofilm biomass (Figure 1A) and exopolysaccharides (Figure 1B) than LM in mono species (p < 0.01) using microtiter plates, indicating the strong biofilm-forming ability of PL. In spite of no difference of viable cell number, biofilm biomass and exopolysaccharide both in mono- or dual-species LM and PL were greatly higher at 15°C than those at 30°C, indicating that the low temperature could stimulate the secretion of biofilm extracellular matrix of the two strains. Unexpectedly, the levels of biofilm formation and polysaccharide production in the co-culture were significantly lower than that of mono-species PL both at two culture temperatures (p < 0.05).

Meanwhile, OM, CLSM, and SEM were employed to visualize the adhesion and structure of mono- and dual-species biofilms at micro level (Figure 1C). Massive adhesive cells were distributed in the mono-species biofilm of LM and PL after 24 h by OM observation. Compared to LM, PL formed the denser and thicker maturing biofilm with multiple layers followed by CLSM and SEM analysis, while the dual-species biofilms developed sparer and higher heterogeneity than PL. In addition, LM and PL as mono-culture, and their co-culture formed the biofilms with thickness of 20.3, 100.5, and 90.4 μm, respectively. Furthermore, PL as mono- or dual species produced the large EPS wrapping cell aggregation forming spatial structure observed by SEM, especially mono-species PL. Generally, the results of these microscopies were consistent with those as determined by CV staining and exopolysaccharides.

The MICs of ε-PLH against LM and PL were 16 and 100 μg/ml in TSB, respectively (Supplementary Table 1). ε-PLH at the concentration lower than 8 and 50 μg/ml had no significant effect on the planktonic cell densities of LM and PL during the stationary phase, respectively, and these latter concentrations of 8 and 50 μg/ml were thus as MSCs of ε-PLH against the two strains. Effects of ε-PLH treatment on the biofilm formation of the two strains, cultivated under either mono- or dual-species conditions on SS coupons, were evaluated (Figure 2A). The treatment of ε-PLH resulted in a significant reduction of mono- or dual-species biofilm biomass of LM and PL compared with untreated control (p < 0.05). The biofilm cell counts of dual-species LM were apparently decreased after 32 μg/ml ε-PLH treatment compared with mono-species LM treated with 4 μg/ml ε-PLH. In contrast, the viable cell population of PL both in mono- and dual-culture biofilms treated with 64 μg/ml ε-PLH was significantly decreased (p < 0.05) and did not differ between two biofilm status (p > 0.05). More specifically, the ε-PLH treatment at 64 μg/ml achieved 4.32 and 0.85 log decrease of dual-species biofilms of LM and PL compared with 5.87 and 0.96 log reduction of individual mono-culture, respectively.

The effects of ε-PLH at the different concentrations on the biofilm biomass and exopolysaccharide production of LM and PL as mono- or dual-species biofilms are illustrated in Figures 2B,C. When treated with ε-PLH ranging from 2 to 128 μg/ml, biofilm biomass of LM and PL as mono-culture was significantly reduced at 2 and 16 μg/ml of ε-PLH (p < 0.05), respectively, whereas the dual-species biofilms were apparently dropped at 32 μg/ml of ε-PLH. When treated with ε-PLH at 32 μg/ml, biofilm biomass for LM and PL as mono-species reduced by 83.6 and 36.9%, respectively, which were higher than that for the dual-species biofilms (31.0%). LM failed to form the biofilm in supplement with 64 μg/ml ε-PLH. Likewise, the exopolysaccharide production of LM and PL as mono-culture was decreased in the presence of ε-PLH at 2 and 16 μg/ml, respectively, compared with LM and PL as dual-species biofilms at 32 μg/ml ε-PLH (p < 0.05). When treated with ε-PLH at 32 μg/ml, exopolysaccharide of LM and PL as mono- and dual-species biofilms reduced by 79.22, 27.66, and 25.84%, respectively. These findings indicated that ε-PLH effectively inhibited the biofilm development of LM as mono species in a concentration-dependent manner; however, multispecies biofilms of the two strains exhibited their community-wide resistance, resulting in higher tolerance to ε-PLH than their counterparts in mono-species biofilm.

Due to the high tolerance of multispecies biofilms to ε-PLH, the impacts of ε-PLH combined with CEO at the MSCs against biofilm formation as dual-species LM and PL were further assayed (Figure 3A). Compared with no effect of alone ε-PLH and CEO at two MSCs, the inhibitory effect of four combinations of ε-PLH and CEO significantly enhanced against the dual biofilms (p < 0.05). After incubation for 24 h, the combined treatments of P8 + C25, P8 + C50, P16 + C25, and P16 + C50 exhibited 19.4, 43.2, 37.17, and 55.27% (p < 0.05) decrease in CV biomass, respectively. The inhibitory rate of the three combined groups except P8 + C25 was apparently higher compared with either of the compounds used alone (p < 0.05). Similarly, the four combined treatments showed 24.13, 46.87, 45.03, and 54.03% reduction in exopolysaccharide levels of dual-species biofilms, respectively, in contrast to treatment of P16 (7.4%) or C50 (25.87%) used alone (Figure 3B). It was revealed that ε-PLH in combination with CEO constituents at the MSCs induced a good synergistic effect to inhibit multispecies biofilm formation of LM and PL.

We further observed the adhesion of dual-species biofilms formed by LM and PL treated with ε-PLH and CEO compounds by CLSM and SEM to assess the changes in biofilm spatial structures (Figure 3C). The two strains formed the dual biofilms with dense matrices, while the biofilm homogeneity gradually increased with the treatment of ε-PLH and CEO, as well as the decrease in viable cells. The spatial structure of dual biofilms was impaired by CEO at two MSCs in a concentration-dependent manner, in contrast to weak effect of ε-PLH at 8 and 16 μg/ml. Compared to the two compounds used alone, the four combined treatments apparently decreased the biovolume of viable cells with green signals and increased the dead cells with red signals in the dual-species biofilms. When treated with the combined groups of P8 + C25, P8 + C50, P16 + C25, and P16 + C50, the multispecies biofilm thickness was decreased to 71, 36, 42, and 23.7 μm, respectively, in contrast to 89.7 μm of thickness for dual-species biofilms control. Observations by SEM showed that many small amorphous aggregates (LM) appeared around the bacilli (PL) in the spare biofilms with less EPS matrix treated by P16 + C25 compared to control of dual-species biofilms (Figure 3D). As the concentration of combined CEO increased to 50 μg/ml (P16 + C50), few monolayers of PL adhered to surface and formed the biofilms without EPS matrix. The observations by CLSM and SEM were in line with the strong inhibition of biofilm biomass and exopolysaccharides secretion when combining ε-PLH with CEO. Thus, the combined treatments of ε-PLH and CEO at the MSCs repressed effectively adhesion and destroyed the multispecies biofilm spatial structures, indicating their synergistic inhibitory activity.

Autoinducer-2 activities of LM and PL were performed in co-culture treated by ε-PLH or CEO, as presented in Figure 4. The supernatants of the LM and PL as mono- or co-culture induced bioluminescence in V. harveyi BB170, especially PL. The AI-2 activity in the co-culture of the two strains was significantly lower than that of single PL, which might be consistent with the biofilm-forming ability. Moreover, AI-2 activity of co-culture was reduced by 3.42–14.92% and 5.54–17.81% after exposure to ε-PLH or CEO used alone, while the inhibitory rates of P8 + C50, P16 + C25, and P16 + C50 were as high as 35.55–46.89%, respectively, indicating the inhibitory efficiency of combined compounds in a concentration-dependent manner (p < 0.05).

The combination of ε-PLH with CEO at higher than MICs was further applied to eradicate the preformed biofilms of LM and PL as dual species after incubation for 24 h. In Figures 5A,B, biofilm removal and inactivation of combined treatments significantly increased compared to the two compounds used alone. Approximately 47.08–60.36% of dual biofilms were removed after treatment with four combined groups of P64 + C50, P64 + C100, P128 + C50, and P128 + C100 compared with 43.82% with P128 and 36.38% with C100, respectively. As expected, the combined treatments of ε-PLH and CEO resulted in the great inactivation of cellular viability in performed biofilms as dual species. The four combined treatments showed 72.46–80.5% decrease in the cell viability of 24 h performed biofilms as dual species, which were higher than P128 (36.99%) and C100 (72.05%) used alone (p < 0.05). Additionally, ε-PLH was effective in the biofilm removal of dual-species biofilms, while CEO exhibited the strong inactivation of biofilm cell, which could be associated with their synergistic eradication of performed biofilms. Furthermore, CLSM images also confirmed that ε-PLH in combination with CEO significantly enhanced eradication of the multispecies biofilms, in contrast to either of the compounds used alone (Figure 5C). Following the combined treatment, viable cells with green disappeared and dead cells in red increased throughout the biofilms, indicating that the dual biofilms were susceptible to combined treatments, and the combined treatment of P128 + C100 showed the highest biofilm-eradicating efficacy. In addition, the spatial structure of LM and PL as dual-species biofilms was effectively destroyed by the combined compounds, as well as the significant decrease in biofilm thickness (p < 0.05). More specifically, the biofilm thickness treated with P128 + C100 was decreased by nearly 50% compared to the dual-biofilms control.