Genetic Subtyping, Biofilm-Forming Ability and Biocide Susceptibility of Listeria monocytogenes Strains Isolated from a Ready-to-Eat Food Industry

Listeria monocytogenes is a foodborne pathogen of special concern for ready-to-eat food producers. The control of its presence is a critical step in which food-grade sanitizers play an essential role. L. monocytogenes is believed to persist in food processing environments in biofilms, exhibiting less susceptibility to sanitizers than planktonic cells. This study aimed to test the susceptibility of L. monocytogenes in planktonic culture and biofilm to three commercial food-grade sanitizers and to benzalkonium chloride; together with the genetic subtyping of the isolates. L. monocytogenes isolates were collected from raw materials, final products and food-contact surfaces during a 6-year period from a ready-to-eat meat-producing food industry and genetically characterized. Serogrouping and pulsed-field gel electrophoresis (PFGE) revealed genetic variability and differentiated L. monocytogenes isolates in three clusters. The biofilm-forming ability assay revealed that the isolates were weak biofilm producers. L. monocytogenes strains were susceptible both in the planktonic and biofilm form to oxidizing and ethanol-based compounds and to benzalkonium chloride, but not to quaternary ammonium compound. A positive association of biofilm-forming ability and LD90 values for quaternary ammonium compound and benzalkonium chloride was found. This study highlights the need for preventive measures improvement and for a conscious selection and use of sanitizers in food-related environments to control Listeria monocytogenes.


Introduction
Listeria monocytogenes is an ubiquitous small Gram-positive bacterium widespread in the natural environment [1]. It is also an opportunistic pathogen responsible for human listeriosis, a severe disease with high hospitalization and case fatality rates [2,3]. Its psychrotrophic nature and the ability to survive and multiply under extreme physicochemical conditions [4] may explain the difficulty of controlling its presence in refrigerated environments [5].
This pathogen is often associated to ready-to-eat (RTE) food products, with contamination occurring during food processing production [6,7]. Incoming raw materials, food handlers, and even processed ingredients and products are frequent sources of L. monocytogenes contamination [8]. After entering a food producing facility, L. monocytogenes can become a long-term resident, being able to persist for months or years within the premises, including food contact equipments [9]. Once established, 1 L. monocytogenes isolates share the same code with the sample from which they were recovered. Figure 1 presents the resulting dendrogram of 17 L. monocytogenes strains considering ApaI and AscI restriction patterns and serogroups. Pulsotypes were considered to be clones when they had at least 90% of similarity.

Pulsed-Field Gel Electrophoresis Typing
The different food and environment samples presented six PFGE types. Three clusters were identified (indicated as A, B and C in Figure 1), while FP5, RM5 and FP1 pulsotypes had a distinct PFGE profile.
The first cluster (Figure 1, cluster A) includes 9 strains, corresponding to 52.9% of all the analyzed isolates. These strains with identical restriction patterns and exhibiting the same serogroup (serogroup IIc) were collected from raw materials, intermediate products, finished products, and food-contact surfaces in a time frame of 14 months (from February 2013 to April 2014). When comparing cluster A food-contact surfaces and finished products strains' profiles, results suggest the possibility of a common source. It is noteworthy that L. monocytogenes strain RM4 collected in 2014 has 91.6% similarity with strains collected in 2013. This is suggestive of a potential persistent contamination within the food industry, although more studies should be considered in order to establish source attribution. The different food and environment samples presented six PFGE types. Three clusters were identified (indicated as A, B and C in Figure 1), while FP5, RM5 and FP1 pulsotypes had a distinct PFGE profile.
The first cluster ( Figure 1, cluster A) includes 9 strains, corresponding to 52.9% of all the analyzed isolates. These strains with identical restriction patterns and exhibiting the same serogroup (serogroup IIc) were collected from raw materials, intermediate products, finished products, and food-contact surfaces in a time frame of 14 months (from February 2013 to April 2014). When comparing cluster A food-contact surfaces and finished products strains' profiles, results suggest the possibility of a common source. It is noteworthy that L. monocytogenes strain RM4 collected in 2014 has 91.6% similarity with strains collected in 2013. This is suggestive of a potential persistent contamination within the food industry, although more studies should be considered in order to establish source attribution. Kurpas et al. [34] linked L. monocytogenes presence in food processing environments, such as abattoirs, RTE meat-processing industries and retail establishments to crosscontamination.
Cluster B includes three strains (FP7, FP8 and FP9) collected from different finished products between May 2014 and April 2015, all belonging to serogroup IIa. L. monocytogenes strain FP7 shares an indistinguishable profile with strain FP9, collected 1 year later. Apart from being suggestive of persistence over time, which might be due to L. monocytogenes survival and growth in niches within the food environment, these strains belong to serogroup IIa, which is the one most commonly associated to food-related environments [1,30].
Cluster C includes 2 strains (RM2 and RM3) collected between 2013 and 2014, from different raw materials. As seen before in cluster A, L. monocytogenes pulsotypes identified in raw materials exhibit high similarity with pulsotypes from equipment and finished products. These pulsotypes may persist due to the repeated re-introduction of strains from the external environment into food processing facilities over time [35]. Suppliers should be addressed to understand the origin of some strains, although results underline cross-contamination as a possible way of disseminating L. monocytogenes in the assessed food industry. A strict selection and control of suppliers seems to be a preventive measure of upmost importance [36]. Three distinct pulsotypes can also be seen in the resulting dendrogram. FP5 and RM5 strains were collected 1 year apart from each other and presented distinct pulsotypes (64.9% of similarity), belonging to serogroups IIb and IVb, respectively. FP1 isolate exhibits a different PFGE profile from other serogroup IIa strains (71.4% of similarity), which might be due to the fact that serogroup IIa includes atypical strains [27,37]. Cluster B includes three strains (FP7, FP8 and FP9) collected from different finished products between May 2014 and April 2015, all belonging to serogroup IIa. L. monocytogenes strain FP7 shares an indistinguishable profile with strain FP9, collected 1 year later. Apart from being suggestive of persistence over time, which might be due to L. monocytogenes survival and growth in niches within the food environment, these strains belong to serogroup IIa, which is the one most commonly associated to food-related environments [1,30].
Cluster C includes 2 strains (RM2 and RM3) collected between 2013 and 2014, from different raw materials. As seen before in cluster A, L. monocytogenes pulsotypes identified in raw materials exhibit high similarity with pulsotypes from equipment and finished products. These pulsotypes may persist due to the repeated re-introduction of strains from the external environment into food processing facilities over time [35]. Suppliers should be addressed to understand the origin of some strains, although results underline cross-contamination as a possible way of disseminating L. monocytogenes in the assessed food industry. A strict selection and control of suppliers seems to be a preventive measure of upmost importance [36]. Three distinct pulsotypes can also be seen in the resulting dendrogram. FP5 and RM5 strains were collected 1 year apart from each other and presented distinct pulsotypes (64.9% of similarity), belonging to serogroups IIb and IVb, respectively. FP1 isolate exhibits a different PFGE profile from other serogroup IIa strains (71.4% of similarity), which might be due to the fact that serogroup IIa includes atypical strains [27,37].
The presence of serogroups IIa, IIb and IVb isolates suggests a potential public health hazard associated with these RTE meat-based products consumption, since these are the serogroups more commonly associated to human infection [38,39].

Biofilm Formation Assay
After serogrouping and PFGE typing, 10 L. monocytogenes strains were selected for the biofilm formation assay in order to have representatives with different profiles (serogroups and pulsotypes). L. monocytogenes CECT 4031, CECT 911, CECT 935, and CECT 937 were also included in order to investigate differences between strains of different serogroups. The assessed strains in biofilms revealed cvOD values ranging from 0.068 ± 0.001 to 0.1240 ± 0.006 and viable cell counts of 6.0 ± 0.4 log cfu/mL to 7.6 ± 0.4 log cfu/mL after 5 days of growth in polystyrene microtiter wells ( Figure 2).

Biofilm Formation Assay
After serogrouping and PFGE typing, 10 L. monocytogenes strains were selected for the biofilm formation assay in order to have representatives with different profiles (serogroups and pulsotypes). L. monocytogenes CECT 4031, CECT 911, CECT 935, and CECT 937 were also included in order to investigate differences between strains of different serogroups.
The assessed strains in biofilms revealed cvOD values ranging from 0.068 ± 0.001 to 0.1240 ± 0.006 and viable cell counts of 6.0 ± 0.4 log cfu/mL to 7.6 ± 0.4 log cfu/mL after 5 days of growth in polystyrene microtiter wells ( Figure 2). According to Stepanović et al. [40] classification, all the strains (n = 10) revealed a weak biofilmforming ability. Similar results were obtained by Meloni et al. [41] when studying L. monocytogenes isolates from fermented sausage processing plants: 65% of all isolates were weak biofilm producers. However, in our work, the assessed strains exhibited significantly different degrees of biofilmforming ability based on cvOD values (p = 0.0066), and VCC results did not reflect the same biofilmforming ability classes as those obtained using cvOD values. Considering VCC values, all the strains, except L. monocytogenes FP1 and RM3 isolates, revealed lower values than L. monocytogenes CECT 935, which exhibited 7.4 ± 0.2 log cfu/mL. On the other hand, when considering cvOD values, L. monocytogenes FP1, RM1, and FP6 isolates exhibited higher cvOD values than reference L. monocytogenes CECT 935 (0.1078 ± 0.005). Considering cvOD and VCC values, L. monocytogenes CECT 4031 revealed the lowest values for both parameters at 30 °C. The obtained difference between these two parameters is due to the nature of each method of determination. While cvOD measures the turbidity of a suspension and quantifies total biomass (viable and non-viable cells, but also extracellular matrix components), VCC only considers live cells [42]. Taking into account the selected methods to analyze biofilm formation-VCC (log cfu/mL) and cvOD, Pearson's correlation analysis was performed. According to Pearson's correlation coefficient (ρ = 0.7749, p = 0.009), there is a positive and strong correlation between both parameters, which indicates that both methods present a good relationship, being reliable to quantifying L. monocytogenes biofilm formation, complementing each other.
When relating the biofilm-forming ability using cvOD values with the assessed L. monocytogenes serogroups, no significant differences were found (p = 0.526) and the same happened for VCC values (p = 0.929) ( Table 3). Table 3. Biofilm-forming ability of L. monocytogenes strains according to the respective serogroups. According to Stepanović et al. [40] classification, all the strains (n = 10) revealed a weak biofilm-forming ability. Similar results were obtained by Meloni et al. [41] when studying L. monocytogenes isolates from fermented sausage processing plants: 65% of all isolates were weak biofilm producers. However, in our work, the assessed strains exhibited significantly different degrees of biofilm-forming ability based on cvOD values (p = 0.0066), and VCC results did not reflect the same biofilm-forming ability classes as those obtained using cvOD values. Considering VCC values, all the strains, except L. monocytogenes FP1 and RM3 isolates, revealed lower values than L. monocytogenes CECT 935, which exhibited 7.4 ± 0.2 log cfu/mL. On the other hand, when considering cvOD values, L. monocytogenes FP1, RM1, and FP6 isolates exhibited higher cvOD values than reference L. monocytogenes CECT 935 (0.1078 ± 0.005). Considering cvOD and VCC values, L. monocytogenes CECT 4031 revealed the lowest values for both parameters at 30 • C. The obtained difference between these two parameters is due to the nature of each method of determination. While cvOD measures the turbidity of a suspension and quantifies total biomass (viable and non-viable cells, but also extracellular matrix components), VCC only considers live cells [42]. Taking into account the selected methods to analyze biofilm formation-VCC (log cfu/mL) and cvOD, Pearson's correlation analysis was performed. According to Pearson's correlation coefficient (ρ = 0.7749, p = 0.009), there is a positive and strong correlation between both parameters, which indicates that both methods present a good relationship, being reliable to quantifying L. monocytogenes biofilm formation, complementing each other.
When relating the biofilm-forming ability using cvOD values with the assessed L. monocytogenes serogroups, no significant differences were found (p = 0.526) and the same happened for VCC values (p = 0.929) ( Table 3). Similar results were obtained by Di Bonaventura et al. [43] when studying the association of phylogeny and biofilm production. Nevertheless, this study's results counteract the ones obtained by Meloni et al. [41], in which serotypes 1/2a, 1/2b and 4b isolates presented a higher adherence when compared to serotype 1/2c isolates.
Other authors have shown that L. monocytogenes strains from different sources and serogroups are able to produce biofilms on a variety of surfaces, depending on the strain, surface and culture conditions [13,44]. Previous works reported that L. monocytogenes strains varied significantly in their ability to produce biofilm, but no trends could be observed when isolates' serotype and source were compared [3,40]. It is important to highlight that since there is a link between virulence and L. monocytogenes serotype, a continuous discussion relating biofilm formation and serotypes goes on, in order to determine whether biofilm formation is related to disease incidence [1,14].

Activity towards L. monocytogenes Planktonic Suspension
The effect of food-grade commercial sanitizers, including an oxidizing compound (OxC), a quaternary ammonium compound (QaC) and an ethanol-based compound (EthC) on the selected five L. monocytogenes strains was assessed. Tested concentrations were selected based on the manufacturer's recommendation for use in food contact surfaces. The manufacturer's recommended concentrations for OxC and EthC were found to be equally effective in inactivating the five tested strains in planktonic suspension, although this was not observed for QaC.
L. monocytogenes planktonic cells were inactivated by 50 ppm or more (100 and 150 ppm) of OxC. Norwood and Gilmour [45] reported that a 30 sec exposure to 10 ppm free chlorine was enough to completely eliminate planktonic L. monocytogenes culture.
L. monocytogenes strains were exposed to increasing concentrations of EthC (50%, 70%, and 100%) that seemed to be effective in inactivating planktonic cells. Similar results were obtained by Aarnisalo et al. [46].
L. monocytogenes planktonic forms enumeration after QaC treatment was not possible to perform within the tested concentration range. Some authors have reported resistance to QaCs in L. monocytogenes strains [47][48][49] and active efflux pumps are considered the main mechanism for L. monocytogenes tolerance to QaCs [50]. Because it was not possible to determine L. monocytogenes susceptibility to QaC, benzalkonium chloride (BaC) was used to evaluate L. monocytogenes planktonic cells susceptibility. Figure 3 presents the effects of BaC treatment on the five selected L. monocytogenes strains planktonic suspensions. As shown, all strains in the planktonic form presented different susceptibilities to BaC, being affected by different concentrations. Reference strains L. monocytogenes CECT 4031 and CECT 935 were the most susceptible, presenting more than 4-log cfu/mL reduction when exposed to 0.8 ppm of BaC. L. monocytogenes RM1, RM3 and RM5 strains were less susceptible, presenting 4-log cfu/mL reduction only for concentrations higher than 12.5 ppm for RM1 and RM5 and 20 ppm for RM3. To have an 8-log cfu/mL reduction, L. monocytogenes CECT 4031 and CECT 935 planktonic cells were exposed to 2 ppm of BaC. The same was observed when RM1 and RM5 and RM3 were subjected to 25 ppm and 150 ppm, respectively. In line with our results, Nocker et al. [51] reported that the exposure of L. monocytogenes strains to BaC concentrations higher than 30 ppm for 30 min was able to reduce bacterial colonies as measured by plate counts.

Activity towards L. monocytogenes 5-day-old Biofilms
The biocide activity testing assay on biofilms was based on the enumeration of viable cells. The three commercial compounds were tested on 5-day-old biofilms according to the manufacturer's The biocide activity testing assay on biofilms was based on the enumeration of viable cells.
The three commercial compounds were tested on 5-day-old biofilms according to the manufacturer's recommended concentrations. As was observed for L. monocytogenes planktonic cells, both OxC and EthC tested concentrations, which were within the manufacturer's recommended concentrations, were able to eliminate biofilms of all the tested isolates in 5 min at 20 • C. In fact, it was reported that 200 ppm of sodium hypochlorite, an OxC, is enough to eliminate at least 20% of L. monocytogenes biofilms [3,45]. In contrast, after QaC's treatment, no susceptibility to this biocide was observed. Figure 4 presents the effect on VCC after treatment with QaC on the selected L. monocytogenes 5-day-old biofilms.
Antibiotics 2020, 9, x FOR PEER REVIEW 8 of 18 recommended concentrations. As was observed for L. monocytogenes planktonic cells, both OxC and EthC tested concentrations, which were within the manufacturer's recommended concentrations, were able to eliminate biofilms of all the tested isolates in 5 min at 20 °C. In fact, it was reported that 200 ppm of sodium hypochlorite, an OxC, is enough to eliminate at least 20% of L. monocytogenes biofilms [3,45]. In contrast, after QaC's treatment, no susceptibility to this biocide was observed. Figure 4 presents the effect on VCC after treatment with QaC on the selected L. monocytogenes 5-dayold biofilms. In general, QaC was not effective in removing L. monocytogenes 5-day-old biofilms. As shown in Figure 4, when exposed to 150 ppm of QaC, L. monocytogenes CECT 4031 presented the highest reduction (from 6.4 log cfu/mL to 4.0 log cfu/mL). The remaining L. monocytogenes strains presented approximately 1-log cfu/mL reduction in VCC values. QaC resistance in L. monocytogenes biofilms has been reported [52,53]. Taking into account that this biocide is commonly used in food-related environments, these results are worrisome, as L. monocytogenes biofilms present a potential risk in food safety [54]. Figure 5 presents the tested concentration range of BaC's in L. monocytogenes 5-day-old biofilms. In general, L. monocytogenes 5-day-old biofilms' VCC were affected by different BaC concentrations, as happened for planktonic suspensions. While L. monocytogenes CECT 4031 was the most susceptible to BaC's treatment and also presented the lowest biofilm-forming ability, L. monocytogenes RM3 strain was the less susceptible, but presented the highest biofilm-forming ability based on VCC values. A 3-log cfu/mL reduction was observed for L. monocytogenes CECT 4031 after 5 min of exposure to 10 ppm of BaC. On the other hand, for a similar reduction on L. monocytogenes RM3 biofilm, 250 ppm of BaC were necessary. Comparing these results to those obtained for planktonic cells, it seems that L. monocytogenes biofilms are less susceptible to BaC's tested concentrations, since a higher BaC's concentration is needed to have an equivalent log cfu/mL reduction. In general, QaC was not effective in removing L. monocytogenes 5-day-old biofilms. As shown in Figure 4, when exposed to 150 ppm of QaC, L. monocytogenes CECT 4031 presented the highest reduction (from 6.4 log cfu/mL to 4.0 log cfu/mL). The remaining L. monocytogenes strains presented approximately 1-log cfu/mL reduction in VCC values. QaC resistance in L. monocytogenes biofilms has been reported [52,53]. Taking into account that this biocide is commonly used in food-related environments, these results are worrisome, as L. monocytogenes biofilms present a potential risk in food safety [54]. Figure 5 presents the tested concentration range of BaC's in L. monocytogenes 5-day-old biofilms. In general, L. monocytogenes 5-day-old biofilms' VCC were affected by different BaC concentrations, as happened for planktonic suspensions. While L. monocytogenes CECT 4031 was the most susceptible to BaC's treatment and also presented the lowest biofilm-forming ability, L. monocytogenes RM3 strain was the less susceptible, but presented the highest biofilm-forming ability based on VCC values. A 3-log cfu/mL reduction was observed for L. monocytogenes CECT 4031 after 5 min of exposure to 10 ppm of BaC. On the other hand, for a similar reduction on L. monocytogenes RM3 biofilm, 250 ppm of BaC were necessary. Comparing these results to those obtained for planktonic cells, it seems that L. monocytogenes biofilms are less susceptible to BaC's tested concentrations, since a higher BaC's concentration is needed to have an equivalent log cfu/mL reduction.
One example is L. monocytogenes RM3 isolate that in biofilm presented a 2-log cfu/mL reduction when exposed to 100 ppm of BaC and a 3-log cfu/mL reduction when exposed to 250 ppm, while the exposure to 150 ppm of BaC in the planktonic form was enough to cause a 8-log cfu/mL reduction. It has been previously discussed that in biofilm form, L. monocytogenes is more resistant to stress and sanitizing agents than planktonic cells [41,55]. Nakamura et al. [54], when assessing the sanitizing effect of BaC in L. monocytogenes planktonic cells and biofilms, reported that biofilm formation and tolerance to BaC might be related. Tolerance to BaC has also been reported by Piercey et al. [23] after testing BaC resistance and susceptibility based on the minimum inhibitory concentration, and by Xu et al. [24] after investigating phenotypic and genotypic tolerance to BaC based on susceptibility One example is L. monocytogenes RM3 isolate that in biofilm presented a 2-log cfu/mL reduction when exposed to 100 ppm of BaC and a 3-log cfu/mL reduction when exposed to 250 ppm, while the exposure to 150 ppm of BaC in the planktonic form was enough to cause a 8-log cfu/mL reduction. It To assess L. monocytogenes susceptibility to QaC and BaC, LD 90 values were calculated. Figure 6 presents QaC LD 90 values. These values ranged from 298.0 to 532.2 ppm, and were higher than the manufacturer's recommended concentrations to be used in food-related surfaces (maximum recommended concentration: 150 ppm). effect of BaC in L. monocytogenes planktonic cells and biofilms, reported that biofilm formation and tolerance to BaC might be related. Tolerance to BaC has also been reported by Piercey et al. [23] after testing BaC resistance and susceptibility based on the minimum inhibitory concentration, and by Xu et al. [24] after investigating phenotypic and genotypic tolerance to BaC based on susceptibility testing and molecular methods. Although in the last years several studies have focused on biofilm elimination, possible facilitating strategies are still unclear.
To assess L. monocytogenes susceptibility to QaC and BaC, LD90 values were calculated. Figure 6 presents QaC LD90 values. These values ranged from 298.0 to 532.2 ppm, and were higher than the manufacturer's recommended concentrations to be used in food-related surfaces (maximum recommended concentration: 150 ppm). This fact is relevant, because QaC is a commercial biocide that might be used in sublethal concentrations, which might induce L. monocytogenes resistance [46,56]. L. monocytogenes QaC resistance has been previously described, both for planktonic cells and biofilms [10,22,35].
BaC estimated LD90 values for L. monocytogenes tested strains (Figure 7) that ranged from 1.0 to 102.0 ppm in the planktonic form and from 17.8 to 675.2 ppm in biofilms, presenting significant differences (p < 0.0001). L. monocytogenes biofilms exhibited a reduced susceptibility to BaC, compared to the planktonic forms. The biofilm structure may play an important role as the extracellular matrix acts like a barrier, preventing contact with antimicrobial agents [57,58]. In this study, the higher the biofilm-forming ability, the higher were the LD90 values for QaC and BaC. This positive association of biofilm-forming ability and LD90 values was moderate, both for QaC and for BaC (Table 4). This fact is relevant, because QaC is a commercial biocide that might be used in sublethal concentrations, which might induce L. monocytogenes resistance [46,56]. L. monocytogenes QaC resistance has been previously described, both for planktonic cells and biofilms [10,22,35].
BaC estimated LD 90 values for L. monocytogenes tested strains (Figure 7) that ranged from 1.0 to 102.0 ppm in the planktonic form and from 17.8 to 675.2 ppm in biofilms, presenting significant differences (p < 0.0001).
tolerance to BaC might be related. Tolerance to BaC has also been reported by Piercey et al. [23] after testing BaC resistance and susceptibility based on the minimum inhibitory concentration, and by Xu et al. [24] after investigating phenotypic and genotypic tolerance to BaC based on susceptibility testing and molecular methods. Although in the last years several studies have focused on biofilm elimination, possible facilitating strategies are still unclear.
To assess L. monocytogenes susceptibility to QaC and BaC, LD90 values were calculated. Figure 6 presents QaC LD90 values. These values ranged from 298.0 to 532.2 ppm, and were higher than the manufacturer's recommended concentrations to be used in food-related surfaces (maximum recommended concentration: 150 ppm). This fact is relevant, because QaC is a commercial biocide that might be used in sublethal concentrations, which might induce L. monocytogenes resistance [46,56]. L. monocytogenes QaC resistance has been previously described, both for planktonic cells and biofilms [10,22,35].
BaC estimated LD90 values for L. monocytogenes tested strains (Figure 7) that ranged from 1.0 to 102.0 ppm in the planktonic form and from 17.8 to 675.2 ppm in biofilms, presenting significant differences (p < 0.0001). L. monocytogenes biofilms exhibited a reduced susceptibility to BaC, compared to the planktonic forms. The biofilm structure may play an important role as the extracellular matrix acts like a barrier, preventing contact with antimicrobial agents [57,58]. In this study, the higher the biofilm-forming ability, the higher were the LD90 values for QaC and BaC. This positive association of biofilm-forming ability and LD90 values was moderate, both for QaC and for BaC (Table 4). L. monocytogenes biofilms exhibited a reduced susceptibility to BaC, compared to the planktonic forms. The biofilm structure may play an important role as the extracellular matrix acts like a barrier, preventing contact with antimicrobial agents [57,58]. In this study, the higher the biofilm-forming ability, the higher were the LD 90 values for QaC and BaC. This positive association of biofilm-forming ability and LD 90 values was moderate, both for QaC and for BaC (Table 4). These results emphasize the importance of the cautious selection and use of sanitizers in food-producing premises. In fact, the equipment's sanitizing method should be re-assessed and validated in order to control L. monocytogenes contamination, as it might select isolates that are able to survive and adapt to the food processing environment [59], acting as potential contamination sources for RTE food produced in those surfaces. Taken together, biofilm-forming ability and LD 90 values underline the need to select different sanitizers, using rotating schemes, in order to prevent biocide resistance over time. Also, different strategies should be considered, other than the use of chemical biocides, as novel technologies, to control L. monocytogenes in the food production environment [60,61].

Characterization of L. monocytogenes Isolates Collection
A collection of presumptive L. monocytogenes isolates (n = 20) was gathered from raw materials, intermediate, and final products, as well as industrial environment samples (food contact surfaces) of a RTE meat-based food producing industry (Table 1). This industry was located in Évora, Alentejo and produced pork meat delicatessens. L. monocytogenes isolates were collected during a 6-year period (2010-2015) as a result of routine microbiological sampling for industrial hygiene and food safety verification purposes, according to ISO 11290:1996 [62]. From a total of 76 collected samples, five raw materials, three intermediate products, nine finished RTE meat products and three food-contact surfaces were positive for L. monocytogenes. The strains were preserved in brain hearth infusion (BHI) broth (Scharlab, S.B., Barcelona, Spain) with 15% glycerol (Merck KGaA, Darmstadt, Germany) at −80 • C and revivified before use.

L. monocytogenes Confirmation and Serogrouping by PCR
Presumptive L. monocytogenes isolates (n = 20) were confirmed by PCR and serogrouped using a multiplex PCR and an additional PCR based on the amplification of the flaA gene [27]. L. monocytogenes confirmed isolates (n = 17) were selected for further genetic characterization.

Pulsed-Field Electrophoresis Typing
PFGE typing of the selected isolates was performed according to the PulseNet standardized procedure for L. monocytogenes [63]. Briefly, bacterial genomic DNA in 1.5% agarose (SeaKem Gold Agarose, Cambrex, East Rutherford, NJ, USA) plugs was digested in separate reactions with 10U AscI (NZYTech, Lisbon, Portugal) for 2h at 37 • C, and with 50U ApaI (NZYTech) for 2h at 25 • C. Electrophoresis of the resulting DNA fragments was performed in 1% agarose gel (SeaKem Gold), with a lambda PFG ladder standard (New England Biolabs, Massachusetts, USA) in 0.5 X solution of Tris-borate-EDTA buffer (NZYTech) at 14 • C, with 6 V/cm, initial pulsed time of 4.0 s and final pulsed time of 40 s, included angle of 120 • over 19 h using a CHEF-Dr III System (Bio-Rad Laboratories, Hercules, CA, USA). Gels were stained with ethidium bromide (Sigma, St. Louis, MO, USA) and photographed under UV transillumination.

Biofilm-Forming Ability Assay
To assess biofilm formation, six L. monocytogenes strains were selected (RM1, RM3, RM5, FP1, FP5, and FP8) to have representatives from different serogroups and PFGE types. Also, four L. monocytogenes reference strains from the Spanish Type Culture Collection (CECT) were used: CECT 4031 (serogroup IIa), CECT 937 (serogroup IIb), CECT 911 (serogroup IIc), and CECT 935 (serogroup IVb). These strains present the same serogroups as the ones detected in this study isolates (Table 2), allowing for the comparison with existing studies.
The protocol proposed by Romanova et al. [51] was used with some modifications to obtain a 5-day L. monocytogenes mono-cultural biofilm. A single colony of each selected strain was inoculated in buffered peptone water (BPW) (Scharlab, S.B) and incubated for 16-18 h at 30 • C. Optical density at 600 nm (OD 600nm ) was assessed in Ultrospec 2000 (Pharmacia Biotech, Washington, WA, USA) to obtain a concentration of 8 log cfu/mL. For each strain, 4 µL were transferred into three separate wells of polystyrene flat-bottomed microtiter plates (Normax, Marinha Grande, Portugal) filled with 200 µL of BPW. Three wells were used as negative controls, with BPW alone. The plates were lidded and statically incubated at 30 • C for 5 days. After this, the solution was removed from the wells that were rinsed once with sterile distilled water to remove loosely associated bacteria and the attached biofilms were assessed by viable cells enumeration and crystal violet staining.
Considering both evaluation methods, this assay was performed in triplicate, with three replicates for each strain.

Enumeration of Viable Cells in Biofilms
The biofilm was detached from the well surface mechanically into 100 µL of BPW using a mini cell scraper (VWR International, Monroeville, PA, USA). The microtiter plate was sonicated (Ultrasonic bath MXB14, Grant Instruments, Royston, England) for 5 min to detach and collect sessile cells. Then, 100 µL of BPW were pipetted into each well. Serial 10-fold dilutions of the sample in BPW were prepared and 10 µL were dropped onto the surface of a tryptone soy agar (TSA) (Scharlab, S.B) plate. After overnight incubation at 30 • C, colonies were enumerated in a stereoscopic magnifier (Nikon SMZ645, Tokyo, Japan).

Biofilm Assessment by Crystal Violet Staining
The microtiter plate was left air drying for 45 min in the laminar flow hood. Biofilm was stained by adding 220 µL of 0.1% crystal violet (bioMérieux, France) solution to each well for 15 min at room temperature. After stain removal, the wells were washed three times with sterile distilled water and left air drying for 30 min in the laminar flow hood. Then, 220 µL of detaining solution (ethanol: acetone 80:20 v/v) were added to each well 15 min. The microtiter plate was then shaken (Ultrasonic bath MXB14, Grant Instruments, Royston, England) for 5 min and the crystal violet OD (cvOD) was measured in SpectraMax 340PC (Molecular Devices, Silicon Valley, San Jose, CA, USA). Each absorbance value was corrected by subtracting the average absorbance readings of the blank control wells. Adherence capability of the tested strains was based on the cvOD exhibited by bacterial biofilms, according to Stepanović et al. [40] classification. The cut-off cvOD (cvODc) was defined as three standard deviations above the negative control mean cvOD. Strains were classified as no biofilm producers (cvOD ≤ cvODc), weak biofilm producers (cvODc < cvOD ≤ 2 × cvODc), moderate biofilm producers (2 × cvODc < cvOD ≤ 4 × cvODc), and strong biofilm producers (4 × cvODc < cvOD).

Biocides Activity Testing Assay
Based on serogrouping and biofilm formation parameters, L. monocytogenes strains RM1, RM3 and RM5 and L. monocytogenes reference strains CECT 4031 and CECT 935 were selected to be further assessed.
Biocides Commonly used biocides in food contact surfaces and equipment sanitization in food-producing establishments were selected for further testing. Commercial sanitizers (HigiaBlue, Portugal) containing oxidizing compounds (OxC), ethanol-based compounds (EthC) and quaternary ammonium compounds (QaC) were tested. Benzalkonium chloride (BaC; Merck KGaA) was also evaluated. Table 5 exhibits the tested concentrations for each biocide (diluted in hard water) and respective neutralizers. L. monocytogenes strains were incubated in BHI agar (Scharlab, S.B.) at 37 • C for 18 h. Then, 10 mL of the bacterial suspension were prepared to have an OD 600nm of 0.15-0.5, corresponding to a concentration of approximately 1.5-5 × 10 8 cfu/mL. To each tube containing 1 mL of interfering substance, 1 mL of the bacterial suspension was added, and the mixture was vortexed. After 2 min, 8 mL of one of the desired biocide test concentration were added, incubating for 5 min at 20 • C. Then, 1 mL was collected and mixed with 1 mL of hard water and 8 mL of the appropriate neutralizer. After neutralization (5 min at 20 • C), 1 mL was incorporated in TSA (Scharlab, S.B.) in duplicate. After overnight incubation at 37 • C, colonies were enumerated.

Activity towards L. monocytogenes 5-day-old Biofilms
For the biocide activity testing on L. monocytogenes 5-day-old biofilms, to each well containing the biofilm, 20 µL of interfering substance and 20 µL of tryptone salt solution (Scharlab, S.B.) were added. After 2 min, 160 µL of one of the biocide test concentrations was added and incubated for 5 min at 20 • C. After medium removal, the wells were washed with 40 µL of hard water and 160 µL of the appropriate neutralizer. After neutralization (5 min at 20 • C), the medium was removed and the wells were washed with sterile distilled water, which was also removed. The biofilm quantification was performed according to the procedure described in Section 3.4.1. for biofilm detachment, dilution and colony enumeration.
LD 90 was then calculated for both planktonic and biofilm assays in order to determine the biocide concentration that reduced 90% of VCC.

Data Analyses
All quantitative data are presented as mean values with standard deviation (SD) from three independent experiments. Using BioNumerics software package version 6.10 (Applied Maths, Sint-Martens-Latem, Belgium), a dendrogram was constructed based on PFGE patterns of the 17 L. monocytogenes strains, with an optimization setting of 1.5% and a band-position tolerance of 1.5% for AscI and ApaI restriction. Cluster analysis was performed using the unweighted pair group method (UPGMA) with arithmetic averages and band-based Dice correlation coefficient.
To assess L. monocytogenes biofilm formation parameters, Pearson's correlation analyses were used to evaluate the interdependency of cvOD and VCC. To relate biofilm formation parameters and L. monocytogenes serogroups, one-way analysis of variance (ANOVA) followed by Tukey's test were performed.
To evaluate the susceptibility of selected L. monocytogenes strains to biocides, LD 90 values were obtained by adjusting experimental data of mortality obtained in biocide testing assays to a polynomial equation or to a linear regression adjusted to a scatter plot of mortality versus biocide concentration in MS Excel 2016 software (Microsoft Corporation, Redmond, USA). Two-way ANOVA was used to compare BaC LD 90 values in planktonic and biofilm forms. To compare L. monocytogenes biofilms QaC LD 90 values and also BaC LD 90 values, one-way ANOVA followed by Tukey's test were performed. Pearson's correlation coefficient was also used to relate biofilm formation parameters and QaC and BaC LD 90 . When p < 0.05, a statistically significant difference was considered.

Conclusions
Overall, this study provided an assessment of L. monocytogenes isolates from a RTE meat-based food industry, using phenotypic and genetic characterization. The use of molecular and subtyping techniques is an important tool to understand the routes and sources of contamination. Our results reveal that L. monocytogenes contamination of finished products seems to be related to food-contact surfaces, but also to raw materials. Moreover, some of the obtained pulsotypes revealed high homology (>90%) but were not temporally matched, being collected with months of interval. These results might point out to a common source of contamination and are consistent with the hypothesis that there are stable clonal groups of L. monocytogenes, which persist over time, in foods and food-related environments.
All of the studied L. monocytogenes strains demonstrated biofilm-forming ability at 30 • C, revealing to be weak biofilm producers. Strains in biofilms were not susceptible to one of the used commercial sanitizers in the industrial premises, QaC, within the recommended concentration range. Similar results were obtained when testing a pure substance biocide, benzalkonium chloride (BaC) in L. monocytogenes biofilms. In contrast, L. monocytogenes planktonic forms were susceptible to BaC tested concentrations. A positive association was found between biofilm formation parameters and LD 90 values for QaC and BaC.
Taken together, our results suggest that preventive measures need improvement in the assessed food industry. It also reinforces the necessity of an appropriate selection and application of biocides in food premises, to prevent L. monocytogenes biofilm formation and biocide resistance development over time.