LWT - Food Science and Technology 159 (2022) 113211 Contents lists available at ScienceDirect LWT journal homepage: www.elsevier.com/locate/lwt Pathogenic mono-species biofilm formation on stainless steel surfaces: Quantitative, qualitative, and compositional study B.R.H. Cervantes-Huamán , C. Ripolles-Avila *, T. Mazaheri , J.J. Rodríguez-Jerez Area of Human Nutrition and Food Science, Departament de Ciència Animal i dels Aliments, Facultat de Veterinària, Universitat Autònoma de Barcelona, Bellaterra (Cerdanyola del Vallès), CP 08193, Barcelona, Spain A R T I C L E I N F O A B S T R A C T Keywords: The aims of this study were, first, to determine Listeria monocytogenes, Salmonella enterica, ser. Typhimurium, and Surface hygiene Staphylococcus aureus biofilm counts, structure, and their composition in macromolecules by direct epifluor­ Foodborne pathogens escence microscopy (DEM); and second, to evaluate the distribution of the components of the produced biofilms Biofilms through 3D representations obtained by confocal laser scanning microscopy (CLSM). Results showed that all DEM CLSM assessed strains have a high capacity to generate biofilms, with counts greater than 7 log CFU cm⁻2 in all cases. The highest and lowest survival percentages were obtained by S. aureus with 94.13 ± 1.77% and S. Typhimurium with 60.31 ± 4.45%, respectively. Biofilm matrix composition in macromolecules was as follows: (a) L. monocytogenes: proteins 83.35 ± 5.81%, polysaccharides 10.98 ± 5.77%, and extracellular DNA 5.67 ± 2.5%; (b) S. Typhimurium: 83.37 ± 3.35%, 0.68 ± 0.77%, and 15.95 ± 3.39%, respectively; and (c) S. aureus: 82.55 ± 2.39%, 8.48 ± 2.07%, and 8.97 ± 2.27%, respectively. The qualitative analysis of the 3D representations of the biofilms formed by the pathogens modelized showed no homogeneity and/or ordered distribution of their components within the biofilm architecture. These findings could lead to the development of addressed cleaning and disinfection alternatives. 1. Introduction DNA (eDNA) (Jahid & Ha, 2012), and in smaller quantities products from bacterial lysis (Branda et al., 2005). Most microorganisms are The formation of biofilms by pathogenic bacteria on food contact considered capable of forming biofilms under suitable environmental surfaces is one of the main problems faced by the food industry given conditions (Lasa et al., 2005). This capacity represents an adaptive and that this not only puts consumers’ health at risk, but it is also an resistance strategy for microorganisms, allowing them to increase the important cause of reduced economic productivity (OMS, 2015). Ac­ availability of nutrients for their growth, facilitate the use of water or cording to the National Institutes of Health (NIH) and the Centers for even enable the transfer of genetic material. This, in fact, is what is most Disease Control and Prevention (CDC), biofilms are involved in more worrying for the food industry, as biofilms can confer to bacterial cells than 65% of foodborne illnesses. It is currently well documented that resistance against antimicrobial agents (González-Rivas et al., 2018). food pathogens such as Listeria monocytogenes, Salmonella spp., and Consequently, routine cleaning and disinfection operations are often Staphylococcus aureus can produce biofilms on work surfaces and facil­ ineffective against the bacteria that form biofilms (Møretrø et al., 2013). ities, making them leading causes of outbreaks due to It has been reported that 99% of all bacterial cells exist as biofilms and cross-contamination to the final product (Dantas et al., 2018). These that only 1% is in a planktonic state (Ramadan, 2006), providing sufficient three microorganisms present high mortality and incidence rates in reason to continue investigating this predominant form at the microbial Europe and E.E.U.U., respectively. Biofilms are complex microbial level. In this regard, only the presence of a hydrated environment and a communities, irreversibly attached to a surface and embedded in a minimum amount of nutrients are required to originate biofilm formation matrix of self-produced extracellular components (ECM), which exhibit (Terry et al., 2003), factors that are continuously present in the food industry. an altered phenotype in relation to the rate of growth and gene tran­ The biofilm development process is multifactorial and consists of 5 stages: (i) scription of planktonic cells (González-Rivas et al., 2018). The ECM reversible adhesion to the surface; (ii) irreversible adhesion through the produced consists mainly of polysaccharides, proteins, and extracellular production of ECM and quorum sensing; (iii) formation of microcolonies; (iv) * Corresponding author. E-mail address: carolina.ripolles@uab.cat (C. Ripolles-Avila). https://doi.org/10.1016/j.lwt.2022.113211 Received 6 October 2021; Received in revised form 26 December 2021; Accepted 5 February 2022 Available online 8 February 2022 0023-6438/© 2022 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 maturation; and (v) dispersion (Jahid & Ha, 2012; Srey et al., 2013). Once for S. Typhimurium and S. aureus, until an approximate concentration of biofilms are formed on a surface, the cells present in the structure can be 10⁶ CFU ml⁻1 was obtained using a densitometer (DENSIMAT, bio­ between 500 and 1500 times more resistant to biocides than planktonic cells Mérieux, Marcy l’Etoile, France). (Hyde et al., 1998), due to the protective matrix (Carpentier & Cerf, 2011). Furthermore, biofilm cells are more resistant to high temperatures, low pH 2.4. Mono-species biofilm formation on stainless steel surfaces (Castro-Rosas & Escartín, 2005), desiccation, UV rays, and salinity, making their removal more difficult (Speranza et al., 2016). This resistance can This process was carried out following the in vitro biofilm formation allow the persistence of biofilm cells in the food industry, constituting a model established by Ripolles-Avila et al. (2018). For this, 30 μl of the continuous source of microbial contamination (Poimenidou et al., 2016). bacterial inoculum, described in section 2.3, were inoculated in the Thus, it is important to have effective cleaning and disinfection procedures, center of each stainless-steel coupon. The coupons were put into sterile which implies the suitable selection of the products used for this purpose Petri dishes and subsequently placed in a humidity chamber maintained (Ripolles-Avila, Hascoët, et al., 2019). Among them, enzymatic products at saturated relative humidity of >90% and incubated at 30 ◦ C to pro­ stand out, having been shown to partially or totally degrade biofilm matrices mote biofilm growth under moist conditions (Fuster-Valls et al., 2008). (Coughlan et al., 2016; Mazaheri et al., 2020; Ripolles-Avila, Hascoët, et al., The biofilms were formed in static conditions for a week, with washings 2019). These detergents combined with other disinfecting treatments and nutrient renewal established at 2, 3, 6, and 7 days of incubation. To improve biofilm removal and elimination (Meireles et al., 2016), which this effect, 6 ml of sterile distilled water was used to wash each coupon, explains why understanding the composition in macromolecules of micro­ while 30 μl of TSYEBgluc1%+NaCl2% was used for nutrient renewal for bial biofilms enables enzymatic products to be designed according to the type L. monocytogenes, and TSB was used for S. Typhimurium and S. aureus. of matrix produced by the dominant pathogenic microbiota in each food industry. 2.5. Evaluation of the biofilm formation capacity by DEM The main objectives of the present study were first to determine the biofilm formation capacity and composition in macromolecules of After the incubation period, the surfaces were washed with 6 ml of L. monocytogenes CECT 5672, S. Typhimurium CECT 4594, and S. aureus sterile distilled water to remove all the cells that did not adhere to the CECT 239 as representative pathogenic models, using direct epifluor­ surface and were therefore not part of the generated biofilm. Subse­ escence microscopy (DEM); and second, to evaluate the distribution and quently, the coupons were stained with 5 μl of Live/Dead BacLight architecture of the matrix components of the formed biofilms through bacterial viability kit (Molecular Probes, Oregon, USA) and incubated at 3D representations obtained by confocal laser scanning microscopy room temperature in dark conditions for 15 min. The kit is composed of (CLSM). two fluorescent nucleic acid-binding stains, SYTO9 and propidium io­ dide. The first fluorochrome labels all cells, both damaged and intact 2. Materials and methods membranes, while the second only penetrates cells with damaged membranes, causing a reduction of the first dye. To this effect, viable 2.1. Test surfaces cells with an intact membrane appear in fluorescent green, while dead or damaged cells appear in fluorescent red. AISI 316 grade 2B stainless steel coupons (2 cm in diameter and 1 After incubation, the coupons were evaluated by DEM, using an mm thick) were used in this study. Prior to use, they were subjected to Olympus BX51/BX52 direct epifluorescence microscope (Olympus, cleaning and disinfection procedures. The procedure started with Tokyo, Japan) equipped with a 100 W mercury lamp (USH-103OL, cleaning the surfaces by immersing them in an aqueous solution with a Olympus) and a double pass filter cube (U-M51004 F/R–V2, Olympus), non-bactericidal detergent for 1 h (ADIS Hygiene, Madrid, Spain). The and coupled to a digital camera (DP73, Olympus). The surfaces were surfaces were then washed three times with distilled water for 10 s, observed with a 20X objective. To determine the number of viable and subsequently disinfected with 70% isopropanol (Panreac Química, non-viable cells in each sample, 6 random pictures of 6 separate fields, Castellar del Vallès, Spain), and dried in a laminar flow cabinet (PV-30/ equating to 0.16 mm2, were taken, from which an estimate of the 70, Telstar, Terrassa, Spain), all according to European standard UNE- number of cells per square centimeter was obtained. The image analysis EN 13697:2015 regarding non-porous materials (AENOR, 2015). Last, was carried out using Auto 3.2 software (Soft Imaging System GMBH, the surfaces were autoclaved for 15 min at 121 ◦ C to ensure their sterility Münster, Germany). The means of the cell area and the biofilm occupied before starting the tests. area in each image were thereby obtained. The total number of cells was calculated by dividing the average surface occupied by the biofilm by 2.2. Bacterial strains the area of the cells (Ripolles-Avila, Ríos-Castillo, et al., 2019). The re­ sults were expressed as the logarithm of the number of colony-forming L. monocytogenes CECT 5672, S. Typhimurium CECT 4594, and units (CFU) per square centimeter. S. aureus CECT 239 were obtained as freeze-dried cultures from the Spanish Type Culture Collection (CECT, University of Valencia, Valen­ 2.6. Determination of the composition in macromolecules of biofilms cia, Spain). These freeze-dried cultures were rehydrated in Tryptic Soy Broth (TSB; Oxoid, Madrid, Spain) for 48 h at 30 ◦ C, and then cultured After the incubation period, and in the same manner as described in on Tryptic Soy Agar (TSA; Oxoid, Madrid, Spain) for 48 h at 30 ◦ C. section 2.5, the surfaces were washed with 6 ml of sterile distilled water. Isolated colonies were used to prepare stock cultures on TSA slants, Each coupon was subsequently stained with a staining solution con­ which were incubated for 24 h at 37 ◦ C and stored for up to 1 month at taining three fluorochromes: Concanavalin A-Alexa Fluor 594 (ConA 4 ◦ C. 594; ThermoFisher Scientific, Barcelona, Spain), which binds to α-mannopyranosyl and α-glucopyranosyl residues (Stiefel et al., 2016), 2.3. Inoculum preparation showing up in fluorescent red; 4′ , 6-diamino-2-phenylindole (DAPI, ThermoFisher Scientific, Barcelona, Spain), which has an affinity for The bacterial inoculum was prepared by culturing the stock bacterial DNA, showing up in fluorescent blue; Fluorescein-5-isothiocyanate strains overnight in TSA at 37 ◦ C. Several colonies obtained on the TSA (FITC, Sigma-Aldrich, Madrid, Spain), which has an affinity for pro­ were then inoculated to TSYEBgluc1%+NaCl2% [consisting of TSB supple­ teins through amino groups, and specifically with the isothiocyanate mented with 0.3% w/v yeast extract (BD, Madrid, Spain), 1% w/v group that reacts with terminal amino groups and primary amines (Chen glucose (BioLife, Barcelona, Spain), and 2% w/v sodium chloride et al., 2007), showing up in fluorescent green. To make the staining (Panreac, Castellar del Vallès, Spain)] for L. monocytogenes, and to TSB solution, 1 mg ml⁻1 of each fluorochrome was combined with 0.1 M 2 B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 sodium bicarbonate (NaHCO, Panreac, Castellar del Vallès, Spain). that S. aureus is a highly pathogenic biofilm producer, thus supporting Moreover, 10 μl of ConA, 10 μl of DAPI, and 5 μl of FITC were added to the results obtained in the present study where S. aureus was determined each coupon together with 75 μl of 0.1 M NaHCO₃. Once the 100 μl was to be the largest biofilm producer with a total count of 7.76 ± 0.03 log deposited on the coupons, the samples were incubated at room tem­ CFU cm⁻2. It is important to highlight that the type of surface is a perature in dark conditions for 1 h so that the dyes could penetrate the fundamental factor that directly influences biofilm formation (White­ structure. This combination has proven to be effective in studying the head et al., 2009). Although high counts were obtained for this pathogen composition of biofilm macromolecules (Ripolles-Avila et al., 2020). when using stainless steel surfaces, other authors such as Di Ciccio et al. After the incubation period, the staining solution was removed from (2015) and Poimenidou et al. (2016) indicated that S. aureus possesses a the coupons by gravity and washed with 3 ml of sterile distilled water to greater capacity to form biofilms on polystyrene than on stainless steel. eliminate the remaining stain from the surface. The coupons were then However, it is known that the capacity to form biofilms among strains evaluated by DEM. As in section 2.5, 6 different fields were taken for belonging to the same species is varied (Borges et al., 2018; Ripolle­ each sample, but with the difference that 3 photographs corresponding s-Avila, Hascoët, et al., 2019). In this context, Jang et al. (2018) to 3 different filters were taken for each field, enabling the portions of demonstrated that modifying stainless steel as a food contact material the macromolecules studied in each field photographed to be distin­ through electrochemical engravings effectively inhibits the adhesion of guished. The software described in section 2.5 was again used to analyze S. aureus. the images, and the results were expressed as a percentage of each of the In descending order, L. monocytogenes was next, the average count of marked elements. which presented statistically significant differences (P < 0.05) compared with the rest. Ripolles-Avila, Cervantes-Huaman, et al. (2019) evaluated 2.7. Qualitative evaluation of the distribution of biofilm components the capacity to form biofilms of 17 L. monocytogenes strains using DEM, resulting in the CECT 5672 strain presenting the highest count, with a This evaluation was carried out following the same methodology similar value to the one obtained in the present study. In third and last described in sections 2.5 and 2.6, with the difference in this case that the place was S. Typhimurium, whose average count also presented signif­ samples were analyzed by CLSM using a Spectal Multiphoton Leica TCS icant differences (P < 0.05) compared with the rest. Overall, the results SP5 microscope (Leica Microsistemas, Barcelona, Spain). The samples, of the present study reveal that the Gram-positive bacteria used have a each consisting of 6 photographs, were analyzed using a 20X objective, greater capacity for biofilm formation than Gram-negative bacteria. and according to the emission of each dye. The specialized software However, these results cannot be extrapolated to a fact dependent on the IMARIS 7.1 (Bitplane, Zurich, Switzerland) was used to analyze the type of cell wall that the bacteria have, since biofilm formation depends images, from which the 3D models were obtained. on different factors, among which the generation of a protective matrix stands out, for which different metabolic pathways are used (Hassan-­ 2.8. Statistical methods Muhammad et al., 2020). To discuss that, in the study conducted by Borges et al. (2018), the biofilm formation capacity of 243 strains of The tests corresponding to the evaluation by DEM and CLSM were Salmonella spp. at four different incubation temperatures (37 ◦ C, 28 ◦ C, carried out in duplicate in three different experiments (n = 6). The 12 ◦ C and 3 ◦ C) were evaluated. A total of 92.2% of the strains analyzed bacterial counts obtained by DEM were converted to decimal logarith­ were able to produce biofilms at least one of the tested temperatures. In mic values, thus coinciding with the assumption of a normal distribu­ the tests, 71.6% of the strains produced biofilms at 37 ◦ C, 63% at 28 ◦ C, tion. IBM SPSS Statistics version 23 was used to analyze the data 52.3% at 12 ◦ C, and 39.5% at 3 ◦ C. It is important to emphasize the statistically. The One Way Anova test, with a posterior contrast using capacity of Salmonella spp. to produce biofilms at 3 ◦ C (Silva et al., Tukey’s test, was used for the means comparison to observe if there were 2008). Piras et al. (2015) reported a higher biofilm production at 22 ◦ C significant differences between the counts obtained. A P < 0.05 was than at 35 ◦ C. These findings indicate that the factors involved in the considered as statistically significant. production of biofilms have different responses depending on the incu­ bation temperature (Cabarkapa et al., 2015). Based on these studies and 3. Results and discussion under the test conditions established in the present study, S. Typhimu­ rium was expected to present higher counts, together with a better ca­ 3.1. Evaluation of the mono-species biofilm formation capacity on pacity to form biofilms, when incubated at 30 ◦ C. stainless steel surfaces As Table 1 shows, the survival percentage of the generated biofilms was also measured. The proportion of viable cells in the biofilms formed Table 1 shows that the three strains evaluated showed a high ca­ by the assessed Gram-positive bacteria (i.e. L. monocytogenes and pacity to form biofilms, presenting a total count of adhered cells greater S. aureus) ranged from 90.48% to 94.13%, with the highest percentage than 7 log CFU cm− 2 in all cases. To this effect, S. aureus turned out to be corresponding to S. aureus biofilms, although with no significant dif­ the strain with the greatest capacity to form biofilms within the group of ferences with L. monocytogenes survival rate (P = 0.91). In contrast, S. microorganisms evaluated. An average count was significantly different Typhimurium survival rate presented significant differences (P < 0.05) (P < 0.05) from the others. Ibarra-Trujillo et al. (2012) also demonstrate with both Gram-positive bacteria. In this case, it is highly interesting to point out that the total cell counts of the biofilms produced by the three Table 1 bacteria were around 7 logs although S. Typhimurium survival rate was Count of cells that made up the biofilms of the studied strains, expressed in log much lower (i.e. 60.31%), implying therefore that approximately 40% of CFU cm⁻2 and evaluated by DEM. The survival value expressed as a percentage is cells conforming S. Typhimurium biofilms were dead. It has been also shown. Each value corresponds to an average of two repetitions carried out described for other Gram-negative bacteria, such as Klebsiella pneumo­ in three different experiments (n = 6). The standard error of the mean has also niae, that for conforming strong biofilms, eDNA as well as other matrix been calculated. components appear because of increased cell death, what makes these Bacterial strains Total count Survival rate (%) structures stronger (Desai et al., 2019). Moreover, the fact that Log CFU cm− 2 non-viable cell counts were obtained, which in some cases were very L. monocytogenes CECT 5672 7.56 ± 0.04b 90.48 ± 1.62a high, could be because with long incubation times the bacteria that form S. Typhimurium CECT 4594 7.26 ± 0.03c 60.31 ± 4.45b the biofilms can overcome their own exponential growth curve, reach­ S. aureus CECT 239 7.76 ± 0.03a 94.13 ± 1.77a ing cell death. This not only helps to structure the system, but it also ᵃ-ᶜ Values within the column that lack a common superscript differ significantly provides the cells that remain viable in the biofilm with a new source of (P < 0.05). energy (Ripolles-Avila et al., 2018). This fact has been observed in the 3 B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 study of biofilms of other microorganisms such as Bacillus subtilis, in visualization of these channels (Fig. 1-c), concurring with the observa­ whose non-viable cells complex three-dimensional structures are tions of Ibarra-Trujillo et al. (2012). generated. It also constitutes a stress response at the community level to improve the resistance of biofilms to unfavorable environmental con­ ditions (Asally et al., 2012). 3.2. Determination of the composition in macromolecules of biofilms Fig. 1 depicts the shape of the cells that make up the various biofilms and their distribution on the surface, which is sometimes geometric and Fig. 2 compiles the results of the macromolecules composition other times more distributed. In this regard, it has been suggested that evaluation. Proteins were the major component of the biofilm matrix for the organization of the cells from which biofilms are created, as detected all of them (Fig. 2-A). The protein content of biofilms matrices formed by by DEM, can be used to predict their creation (Ripolles-Avila et al., L. monocytogenes, S. Typhimurium, and S. aureus (83.35 ± 5.81%, 83.37 2018). With this in mind, it can be proven that all of the bacterial strains ± 3.35% and 82.55 ± 2.39%, respectively) did not present significant utilized in the study developed biofilms after a one-week incubation differences (P = 0.98) among them. Colagiorgi et al. (2016) described period, albeit with varying degrees of organization. This finding is that L. monocytogenes ECM is mainly composed of proteins. Likewise, consistent with that of Centorame et al. (2017), who found that biofilms surface treatments with protease have been shown to trigger an alter­ have a more complicated organization and a larger density of connected ation in the development of biofilms produced by L. monocytogenes or to cells after a week of incubation. Furthermore, conditioning in a induce the dispersion of the cells that form them, indicating the key role humidity-saturated chamber led to the formation of mature biofilms, as that proteins play in the development and maintenance of the structure also stated by Mai et al. (2006), since this is a primary determinant in the (Nguyen & Burrows, 2014). Moormeier et al. (2014) determined that adhesion of microorganisms and a way to help them to distribute on the adding proteases during the cell multiplication stage cancels the for­ surface. mation of S. aureus biofilms, revealing that the formation of micro­ A characteristic that shows that a biofilm has reached its maturity colonies and the subsequent maturation of the structure involves a stage is the presence of water channels (Bryers, 2000). These formations component of a protein nature. Furthermore, Gonzáles-Machado et al. promote the constant circulation of nutrients and the elimination of (2018) determine that the major component of ECM formed by Salmo­ residues within the biofilm structure (Donlan & Costerton, 2002). In nella enterica Agona is proteins, a comparable result to the one obtained Fig. 1-a and 1-b, corresponding to L. monocytogenes and S. Typhimu­ in the present study. All this supports the results obtained in the present rium, respectively, the presence of interstitial spaces that can be study, where none of the pathogens showed significant differences (P > assumed to be water channels was revealed. Contrarily, S. aureus 0.05) in the proteins content, thus proving their role importance on showed dense colonization on the surface, preventing the clear biofilms integrity. Fig. 2-B shows that L. monocytogenes and S. aureus were the strains Fig. 1. DEM images of biofilms of (a) L. monocytogenes CECT 5672, (b) S. Typhimurium CECT 4594, and (c) S. aureus CECT 239, stained with Live/Dead Baclight. Magnification 20X. 4 B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 Fig. 2. Content of (A) proteins, (B) polysaccharides, and (C) eDNA, present in the matrix of biofilms formed by L. monocytogenes CECT 5672, S. Typhimurium CECT 4594, and S. aureus CECT 239. Data expressed in percentages. The error bars represent the standard error of the mean (n = 6). ᵃ⁻ᵇDifferent letters indicate that the values differ significantly (P < 0.05). that presented the highest percentage of polysaccharides in the biofilm At a qualitative level, the structures were similar in organization to matrix (10.98 ± 5.77% and 8.48 ± 2.07%, respectively), with no sig­ those previously found in a cell viability study (Fig. 3). On this regard, nificant differences (P = 0.54). Contrarily, S. Typhimurium presented a S. aureus showed highest cell count, highest survivability, and highest low percentage of polysaccharides of 0.68 ± 0.77% in its matrix. The structural density, although the composition of the biofilms matrix’s composition of the exopolysaccharides as ECM is different in each bac­ macromolecules was similar to the other Gram-positive bacteria. On the terium, for example, it is cellulose in S. Typhimurium, poly-N- contrary, polysaccharide and eDNA content was significantly different acetylglucosamine in S. aureus (Lasa et al., 2005), and teichoic acids (P < 0.05) than S. Typhimurium, what can also be the reason of higher in L. monocytogenes (Brauge et al., 2016). survivability. Different studies focused on enzymatic technology to Last, Fig. 2-C shows that S. Typhimurium presented the highest eliminate biofilms have determined that its action largely depends on percentage of eDNA in the biofilm matrix (15.95 ± 3.39%), and that this the biofilm-forming species, the interactions between the species that percentage was significantly different (P < 0.05) from those obtained for form the biofilm, and the environmental conditions. For example, the rest of the studied strains. L. monocytogenes and S. aureus did not Lequette et al. (2010) found that polysaccharides have a significant ef­ show significant differences (P = 0.23) in terms of the content of eDNA fect in terms of the matrix destabilization of biofilms formed with present in the biofilm matrix. Although a lower proportion of eDNA was Pseudomonas fluorescens, while serine proteases have a greater release obtained for both these pathogens, this component has been demon­ effect on the biofilms formed by Bacillus spp. (Lequette et al., 2010). strated to be an important determinant in biofilms. In this regard, Rice Lefebvre et al. (2016) found that alpha-amylase is more effective to et al. (2007) determined that eDNA is an important structural compo­ detach P. aeruginosa biofilm cells than S. aureus biofilm cells. This fact nent of the matrix of S. aureus biofilms. Nguyen and Burrows (2014) and indicates the importance of studying the type of resident microbiota in Zetzmann et al. (2015) found eDNA to be a cohesive and structural each food industry and the matrix composition they produce. This matrix component for L. monocytogenes biofilms. Last, eDNA not only knowledge will be useful to design and define cleaning strategies and plays a structural role, but it also serves as a source of energy and nu­ effective disinfection and personalize industrial treatments. trients (Harmsen et al., 2010). In the present study, lipids were not included in the determination of components that conform biofilm matrix as, normally, it is composed of 3.3. Evaluation of the distribution of the components that form the highly hydrated hydrophilic molecules and just, in some cases, ECM biofilms have hydrophobic properties. It seems that the most important type of lipids in terms of biofilm formation are rhamnolipids, which can be Several studies have used CLSM as a tool for the morphological and found in the matrix of Pseudomonas aeruginosa. Interestingly, they structural analysis of biofilms (Bridier et al., 2010; Ramírez-Granillo display surface activity and have been proposed to act in initial micro­ et al., 2015; Reis-Teixeira et al., 2017; Walker & Horswill, 2012), and for colony formation, facilitating surface-associated bacterial migration and determining the composition in macromolecules of the biofilm matrix the formation of mushroom-shaped structures, preventing colonization (Fish et al., 2017). The combination of CLSM with an image analysis of channels, and playing a part in biofilm dispersion (Flemming & system represents an important and accepted research tool for analyzing Wingender, 2010). However, modelized biofilms of the present study and understanding highly complex systems such as biofilms (Neu et al., are not distinguished for producing such substances. Preliminary studies 2010). There is currently no fluorescence labeling method that visual­ showed that fat content on L. monocytogenes biofilms was negligible izes the biofilm matrix in general, due to the complex and highly vari­ (Ripolles-Avila et al., 2018). able composition of the matrix produced by different bacteria and under different environmental conditions (Schlafer & Meyer, 2017). For this 5 B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 Fig. 3. DEM images of the macromolecular composition of the matrix of biofilms formed by (a) L. monocytogenes CECT 5672, (b) S. Typhimurium CECT 4594, and (c) S. aureus CECT 239, stained with Concanavalin A (red), FITC (green), and DAPI (blue). Magnification 20X. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) reason, the majority macromolecules of the biofilm matrix were indi­ Regarding the composition in macromolecules of the matrix, all the vidually stained. The distribution of the components within the biofilms species obtained a higher percentage of proteins. However, the can be observed with respect to cell viability (Fig. 4) and the majority composition of polysaccharides and eDNA differed between the strains macromolecules (Fig. 5). These figures show similar results to those studied. Last, the 3D representations of the produced biofilms did not observed in Figs. 1 and 2, respectively. However, there is no evidence of show a homogeneous and/or ordered distribution of their components a homogeneous and/or ordered distribution of the elements studied (viable and non-viable cells, proteins, polysaccharides, and eDNA). This within the biofilm for each case. Bridier et al. (2010) use CLSM to quantitative determination of the majority macromolecules that form determine the biovolume generated in biofilms formed by various the biofilm matrix and the analysis of their qualitative characteristics pathogens, determining that S. aureus has a high biovolume value, represents an advance for the development of new products and the L. monocytogenes an intermediate value, and S. enterica a lower value. design of possible strategies for their elimination. These findings are consistent with those of the current study, which used DEM to assess the number of cells that form biofilms (Table 1). Figs. 4a Declaration of interest and 5a show that L. monocytogenes produces a thin biofilm, similar to the findings of Reis-Teixeira et al. (2017) and Bridier et al. (2010), who did The authors declare no conflict of interest. not observe that L. monocytogenes has multiple layers and so are shown with a thin profile. This qualitative analysis of the distribution of biofilm Funding sources components and architecture provides valid criteria to be considered in the design of strategies for their elimination. This study was supported by Research Project grants RTI2018- 098267-R-C32 from the Ministerio de Ciencia, Innovación y 4. Conclusions Universidades. L. monocytogenes CECT 5672, S. Typhimurium CECT 4594, and CRediT authorship contribution statement S. aureus CECT 239 showed great biofilm forming capacity on stainless steel surfaces under the established experimental conditions, based on B.R.H. Cervantes-Huamán: Conceptualization, Methodology, both the number of cells that form the biofilms and the structure Formal analysis, Investigation, Writing – original draft. C. Ripolles- generated. Among them, S. aureus was shown to be the pathogen with Avila: Conceptualization, Methodology, Validation, Writing – review & the highest cell count, highest survival, and highest structural density. editing, Supervision. T. Mazaheri: Methodology, Investigation. J.J. 6 B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 Fig. 4. 3D images obtained by CLSM of the biofilms formed by (a) L. monocytogenes CECT 5672, (b) S. Typhimurium CECT 4594, and (c) S. aureus CECT 239, stained with Live/Dead Baclight. Viable cells are marked in green and non-viable cells in red. The second column represents only living cells and the second column only deaths. Magnification 20X. (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.) Fig. 5. 3D images obtained by CLSM of the biofilms formed by (a) L. monocytogenes CECT 5672, (b) S. Typhimurium CECT 4594, and (c) S. aureus CECT 239, stained with Concanavalin A (red) which marks polysaccharides, FITC (green) which marks proteins, and DAPI (blue) which marks eDNA. The second column represents only proteins, the third only polysaccharides, and the fourth only eDNA. Magnification 20X. Rodríguez-Jerez: Conceptualization, Validation, Writing – review & Acknowledgements editing, Supervision, Project administration, Funding acquisition. The authors thank Judit, Helena, and Nuria from the Microscopy 7 B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 Service of the UAB for their for her technical support in CLSM. The González-Rivas, F., Ripolles-Avila, C., Fontecha-Umaña, F., Ríos-Castillo, A. G., & Rodríguez-Jerez, J. J. (2018). Biofilms in the spotlight: Detection, quantification, authors alone are responsible for the content and writing of the paper. and removal methods. Comprehensive Reviews in Food Science and Food Safety, 17(5), The authors acknowledge Ms. Sarah Davies for the English grammar 1261–1276. https://doi.org/10.1111/1541-4337.12378 review. Harmsen, M., Lappann, M., Knøchel, S., & Molin, S. (2010). Role of extracellular DNA during biofilm formation by Listeria monocytogenes. Applied and Environmental Microbiology, 76(7), 2271–2279. https://doi.org/10.1128/AEM.02361-09 References Hyde, J., Darouiche, R., & Costerton, J. (1998). Strategies for prophylaxis against prosthetic valve endocarditis: A review article. Journal of Heart Valve Disease, 7(3), AENOR. (2015). ISO 13697. Antisépticos y desinfectantes químicos. Ensayo cuantitativo de 316–326. Retrieved from https://www.ncbi.nlm.nih.gov/pubmed/9651846. superficie no porosa para la evaluación de la actividad bactericida y/o fungicida de los Ibarra-Trujillo, C., Villar-Vidal, M., Gaitán-Cepeda, L. A., Pozos-Guillen, A., Mendoza-de desinfectantes químicos utilizados en productos alimenticios, en la industria, en el hogar y Elias, R., & Sánchez-Vargas, L. O. (2012). Ensayo de formación y cuantificación de en colectividades. biopelículas mixtas de Candida albicans y Staphylococcus aureus. Revista Asally, M., Kittisopikul, M., Rue, P., Du, Y., Hu, Z., Cagatay, T., & Süel, G. M. (2012). Iberoamericana De Micologia, 29(4), 214–222. https://doi.org/10.1016/j. Localized cell death focuses mechanical forces during 3D patterning in a biofilm. riam.2012.02.003 Proceedings of the National Academy of Sciences, 109(46), 18891–18896. https://doi. Jahid, I. K., & Ha, S. (2012). A review of microbial biofilms of produce: Future challenge org/10.1073/pnas.1212429109 to food safety. Food Science Biotechnology, 21, 299–316. https://doi.org/10.1007/ Borges, K. A., Furian, T. Q., Souza, S. N., Menezes, R., Tondo, E. C., Salle, C. T. P., s10068-012-0041-1 Moraes, H. L. S., & Nascimento, V. P. (2018). Biofilm formation capacity of Jang, Y., Choi, W. T., Johnson, C. T., García, A. J., Singh, P. M., Breedveld, V., & Salmonella serotypes at different temperature conditions. Pesquisa Veterinária Champion, J. A. (2018). Inhibition of bacterial adhesion on nanotextured stainless Brasileira, 38(1), 71–76. https://doi.org/10.1590/1678-5150-pvb-4928 steel 316L by electrochemical etching. ACS Biomaterials Science & Engineering, 4(1), Branda, S. S., Vik, S., Friedman, L., & Kolter, R. (2005). Biofilms: The matrix revisited. 90–97. https://doi.org/10.1021/acsbiomaterials.7b00544 Trends in Microbiology, 13, 20–26. https://doi.org/10.1016/j.tim.2004.11.006 Lasa, I., Del Pozo, J. L., Penadés, J. R., & Leiva, J. (2005). Bacterial biofilms and Brauge, T., Sadovskaya, I., Faille, C., Benezech, T., Maes, E., Guerardel, Y., & Midelet- infection. Anales del Sistema Sanitario de Navarra, 28(2), 163–175. Retrieved from Bourdin, G. (2016). Teichoic acid is the major polysaccharide present in the Listeria http://scielo.isciii.es/scielo.php?script=sci_arttext&pid=S1137-662720050003 monocytogenes biofilm matrix. FEMS Microbiology Letters, 363(2), Article fnv229. 00002. https://doi.org/10.1093/femsle/fnv229 Lefebvre, E., Vighetto, C., Di Martino, P., Larreta-Garde, V., & Seyer, D. (2016). Bridier, A., Dubois-Brissonnet, F., Boubetra, A., Thomas, V., & Briandet, R. (2010). The Synergistic antibiofilm efficacy of various commercial antiseptics, enzymes and biofilm architecture of sixty opportunistic pathogens deciphered using a high EDTA: A study of Pseudomonas aeruginosa and Staphylococcus aureus biofilms. throughput CLSM method. Journal of Microbiological Methods, 82(1), 64–70. https:// International Journal of Antimicrobial Agents, 48(2), 181–188. https://doi.org/ doi.org/10.1016/j.mimet.2010.04.006 10.1016/j.ijantimicag.2016.05.008 Bryers, J. D. (2000). Biofilm formation and persistence. In J. D. Bryers (Ed.), Biofilms II: Lequette, Y., Boels, G., Clarisse, M., & Faille, C. (2010). Using enzymes to remove Process analysis and applications (pp. 45–88). New York: Wiley-Liss Inc. biofilms of bacterial isolates sampled in the food-industry. Biofouling, 26(4), Cabarkapa, I., Skrinjar, M., Levic, J., Kokic, B., Blagojev, N., Milanov, D., & Suvajdzic, L. 421–431. https://doi.org/10.1080/08927011003699535 (2015). Biofilm forming ability of Salmonella enteritidis in vitro. Acta Veterinaria Brno, Mai, T. L., Sofyan, N. I., Fergus, J. W., Gale, W. F., & Conner, D. E. (2006). Attachment of 65(3), 371–389. https://doi.org/10.1515/acve-2015-0031 Listeria monocytogenes to an austenitic stainless steel after welding and accelerated Carpentier, B., & Cerf, O. (2011). Review - persistence of Listeria monocytogenes in food corrosion treatments. Journal of Food Protection, 69(7), 1527–1532. https://doi.org/ industry equipment and premises. International Journal of Food Microbiology, 145, 10.4315/0362-028X-69.7.1527 1–8. https://doi.org/10.1016/j.ijfoodmicro.2011.01.005 Mazaheri, T., Ripolles-Avila, C., Hascoët, A. S., & Rodríguez-Jerez, J. J. (2020). Effect of Castro-Rosas, J., & Escartín, E. F. (2005). Increased tolerance of Vibrio cholerae O1 to an enzymatic treatment on the removal of mature Listeria monocytogenes biofilms: A temperature, pH, or drying associated with colonization of shrimp carapaces. quantitative and qualitative study. Food Control, 114, Article 107266. https://doi. International Journal of Food Microbiology, 102, 195–201. https://doi.org/10.1016/j. org/10.1016/j.foodcont.2020.107266 ijfoodmicro.2004.12.015 Meireles, A., Borges, A., Giaouris, E., & Simões, M. (2016). The current knowledge on the Centorame, P., D’Angelo, A. R., Di Simone, F., Salini, R., Cornacchia, A., Marrone, R., & application of antibiofilm enzymes in the food industry. Food Research International, Pomilio, F. (2017). Listeria monocytogenes biofilm production on food packaging 86, 140–146. https://doi.org/10.1016/j.foodres.2016.06.006 materials submitted to physical treatment. Italian Journal of Food Safety, 6(3), Moormeier, D. E., Bose, J. L., Horswill, A. R., & Bayles, K. W. (2014). Temporal and 106–109. https://doi.org/10.4081/ijfs.2017.6654 stochastic control of Staphylococcus aureus biofilm development. mBio, 5(5), Chen, M. Y., Lee, D. J., Tay, J. H., & Show, K. Y. (2007). Staining of extracellular e01341–14. https://doi.org/10.1128/mBio.01341-14 polymeric substances and cells in bioaggregates. Applied Microbiology and Møretrø, T., Langsrud, S., & Heir, E. (2013). Bacteria on meat abattoir process surfaces Biotechnology, 75(2), 467–474. https://doi.org/10.1007/s00253-006-0816-5 after sanitation: Characterisation of survival properties of Listeria monocytogenes and Colagiorgi, A., Di Ciccio, P., Zanardi, E., Ghidini, S., & Ianieri, A. (2016). A look inside the commensal bacterial flora. Advances in Microbiology, 3, 255–264. https://doi. the Listeria monocytogenes biofilms extracellular matrix. Microorganisms, 4(22), 1–12. org/10.4236/aim.2013.33037 https://doi.org/10.3390/microorganisms4030022 Muhammad, M. H., Idris, A. L., Fan, X., Guo, Y., Yu, Y., Jin, X., & Huang, T. (2020). Coughlan, L. M., Cotter, P. D., Hill, C., & Alvarez-Ordóñez, A. (2016). New weapons to Beyond risk: Bacterial biofilms and their regulating approaches. Frontiers in fight old enemies: Novel strategies for the (bio) control of bacterial biofilms in the Microbiology, 11(May), 1–20. https://doi.org/10.3389/fmicb.2020.00928 food industry. Frontiers in Microbiology, 7, 1–21. https://doi.org/10.3389/ Neu, T. R., Manz, B., Volke, F., Dynes, J. J., Hitchcock, A. P., & Lawrence, J. R. (2010). fmicb.2016.01641 Advanced imaging techniques for assessment of structure, composition and function Dantas, S. T. A., Rossi, B. F., Bonsaglia, E. C. R., Castilho, I. G., Hernandes, R. T., in biofilm systems. FEMS Microbiology Ecology, 72(1), 1–21. https://doi.org/ Fernandes, A., & Rall, V. L. M. (2018). Cross-contamination and biofilm formation by 10.1111/j.1574-6941.2010.00837.x Salmonella enterica serovar Enteritidis on various cutting boards. Foodborne Pathogens Nguyen, U. T., & Burrows, L. L. (2014). DNase I and proteinase K impair Listeria and Disease, 15, 81–85. https://doi.org/10.1089/fpd.2017.2341 monocytogenes biofilm formation and induce dispersal of pre-existing biofilms. Desai, S., Sanghrajka, K., & Gajjar, D. (2019). High adhesion and increased cell death International Journal of Food Microbiology, 187, 26–32. https://doi.org/10.1016/j. contribute to strong biofilm formation in Klebsiella pneumoniae. Pathogens, 8(4), ijfoodmicro.2014.06.025 1–16. https://doi.org/10.3390/pathogens8040277 OMS. (2015). Informe de la OMS señala que los niños menores de 5 años representan casi un Di Ciccio, P., Vergara, A., Festino, A. R., Paludi, D., Zanardi, E., Ghidini, S., & Ianieri, A. tercio de las muertes por enfermedades de transmisión alimentaria. (2015). biofilm formation by Staphylococcus aureus on food contact surfaces: Piras, F., Fois, F., Consolati, S. G., Mazza, R., & Mazzette, R. (2015). Influence of Relationship with temperature and cell surface hydrophobicity. Food Control, 50, temperature, source, and serotype on biofilm formation of Salmonella enterica 930–936. https://doi.org/10.1016/j.foodcont.2014.10.048 isolates from pig slaughterhouses. International Association for Food Protection, 78 Donlan, R., & Costerton, J. (2002). Biofilms: Survival mechanisms of clinically relevant (10), 1875–1878. https://doi.org/10.4315/0362-028X.JFP-15-085 microorganisms. Clinical Microbiology Reviews, 15(2), 167–193. https://doi.org/ Poimenidou, S. V., Chrysadakou, M., Tzakoniati, A., Bikouli, V. C., Nychas, J. J., & 10.1128/CMR.15.2.167-193.2002 Skandamis, P. N. (2016). Variability of Listeria monocytogenes strains in biofilm Fish, K., Osborn, A. M., & Boxall, J. B. (2017). Biofilm structures (EPS and bacterial formation on stainless steel and polystyrene materials and resistance to peracetic communities) in drinking water distribution systems are conditioned by hydraulics acid and quaternary ammonium compounds. International Journal of Food and influence discolouration. The Science of the Total Environment, 593–594, Microbiology, 237, 164–171. https://doi.org/10.1016/j.ijfoodmicro.2016.08.029 571–580. https://doi.org/10.1016/j.scitotenv.2017.03.176 Ramadan, H. H. (2006). Chronic rhinosinusitis and bacterial biofilms. Current Opinion in Flemming, H., & Wingender, J. (2010). The biofilm matrix. Nature Reviews Microbiology, Otolaryngology & Head and Neck Surgery, 14(3), 183–186. https://doi.org/10.1097/ 8(9), 623–633. https://doi.org/10.1038/nrmicro2415 01.moo.0000193177.62074.fd Fuster-Valls, N., Hernández-Herrero, M., Marín-de-Mateo, M., & Rodríguez-Jerez, J. J. Ramírez-Granillo, A., Canales, M. G., Espíndola, M. E., Martínez-Rivera, M. A., de (2008). Effect of different environmental conditions on the bacteria survival on Lucio, V. M., & Tovar, A. V. (2015). Antibiosis interaction of Staphylococccus aureus stainless steel surfaces. Food Control, 19(3), 308–314. https://doi.org/10.1016/j. on Aspergillus fumigatus assessed in vitro by mixed biofilm formation. BMC foodcont.2007.04.013 Microbiology, 15, 33. https://doi.org/10.1186/s12866-015-0363-2 González-Machado, C., Capita, R., Riesco-Peláez, F., & Alonso-Calleja, C. (2018). Reis-Teixeira, F. B. dos, Alves, V. F., & Martinis, E. C. P. de (2017). Growth, viability and Visualization and quantification of the cellular and extracellular components of architecture of biofilms of Listeria monocytogenes formed on abiotic surfaces. Salmonella Agona biofilms at different stages of development. PLoS One, 13(7). Brazilian Journal of Microbiology, 48, 587–591. https://doi.org/10.1016/j. https://doi.org/10.1371/journal.pone.0200011 bjm.2017.01.004 8 B.R.H. Cervantes-Huamán et al. LWT 159 (2022) 113211 Rice, K. C., Mann, E. E., Endres, J. L., Weiss, E. C., Cassat, J. E., Smeltzer, M. S., & Silva, D. L., Celidonio, F. A., & Oliveira, K. M. P. (2008). Verificação da temperatura de Bayles, K. W. (2007). The cidA murein hydrolase regulator contributes to DNA refrigeradores domésticos para minimizar a deterioração e possíveis doenças release and biofilm development in Staphylococcus aureus. Proceedings of the National veiculadas por alimentos. Revista Higiene Alimentar, 22(164), 42–45. Retrieved from Academy of Sciences of the United States of America, 104(19), 8113–8118. https://doi. https://pesquisa.bvsalud.org/bvs-vet/resource/pt/vti-45357. org/10.1073/pnas.0610226104 Speranza, B., Monacis, N., Sinigaglia, M., & Corbo, M. (2016). Approaches to removal Ripolles-Avila, C., Cervantes-Huaman, B. H., Hascoët, A. S., Yuste, J., & Rodríguez- and killing of Salmonella spp. biofilms. Journal of Food Processing and Preservation, 41 Jerez, J. J. (2019). Quantification of mature Listeria monocytogenes biofilm cells (1), Article e12758. https://doi.org/10.1111/jfpp.12758 formed by an in vitro model: A comparison of different methods. International Journal Srey, S., Jahid, I., & Ha, S. D. (2013). biofilm formation in food industries: A food safety of Food Microbiology, 289, 209–214. https://doi.org/10.1016/j. concern. Food Control, 31(2), 572–585. https://doi.org/10.1016/j. ijfoodmicro.2018.10.020 foodcont.2012.12.001 Ripolles-Avila, C., Hascoët, A. S., Guerrero-Navarro, A. E., & Rodríguez-Jerez, J. J. Stiefel, P., Mauerhofer, S., Schneider, J., Maniura-Weber, K., Rosenberg, U., & Ren, Q. (2018a). Establishment of incubation conditions to optimize the in vitro formation of (2016a). Enzymes enhance biofilm removal efficiency of cleaners. Antimicrobial mature Listeria monocytogenes biofilms on food-contact surfaces. Food Control, 92, Agents and Chemotherapy, 60(6), 3647–3652. https://doi.org/10.1128/AAC.00400- 240–248. https://doi.org/10.1016/j.foodcont.2018.04.054 16 Ripolles-Avila, C., Hascoët, A. S., Martínez-Suárez, J. V., Capita, R., & Rodríguez- Terry, A., Curtis, E., Pantle, R., & Penny, S. A. (2003). Boundaries for biofilm formation: Jerez, J. J. (2019). Evaluation of the microbiological contamination of food Humidity and temperature. Applied and Environmental Microbiology, 69(8), processing environments through implementing surface sensors in an Iberian pork 5006–5010. https://doi.org/10.1128/AEM.69.8.5006–5010.2003 processing plant: An approach towards the control of. Listeria monocytogenes. Food Walker, J. N., & Horswill, A. R. (2012). A coverslip-based technique for evaluating Control, 99, 40–47. https://doi.org/10.1016/j.foodcont.2018.12.013 Staphylococcus aureus biofilm formation on human plasma. Frontiers in Cellular and Ripolles-Avila, C., Ramos-Rubio, M., Hascoët, A. S., Castillo, M., & Rodríguez-Jerez, J. J. Infection Microbiology, 2, 39. https://doi.org/10.3389/fcimb.2012.00039 (2020). New approach for the removal of mature biofilms formed by wild strains of Whitehead, K. A., Benson, P., Smith, L. A., & Verran, J. (2009). The use of Listeria monocytogenes isolated from food contact surfaces in an Iberian pig physicochemical methods to detect organic food soils on stainless steel surfaces. processing plant. International Journal of Food Microbiology, 323, Article 108595. Biofouling. The Journal of Bioadhesion and Biofilm Research, 25(8), 749–756. https:// https://doi.org/10.1016/j.ijfoodmicro.2020.108595 doi.org/10.1080/08927010903161299 Ripolles-Avila, C., Ríos-Castillo, A. G., Fontecha-Umaña, F., & Rodríguez-Jerez, J. J. Zetzmann, M., Okshevsky, M., Endres, J., Sedlag, A., Caccia, N., Auchter, M., (2019). Removal of Salmonella enterica serovar Typhimurium and Cronobacter Waidmann, M. S., Desvaux, M., Meyer, R. L., & Riedel, C. U. (2015). DNase-sensitive sakazakii biofilms from food contact surfaces through enzymatic catalysis. Journal of and -resistant modes of biofilm formation by Listeria monocytogenes. Frontiers in Food Safety, 40, Article e12755. https://doi.org/10.1111/jfs.12755 Microbiology, 6(DEC), 1–11. https://doi.org/10.3389/fmicb.2015.01428 Schlafer, S., & Meyer, R. L. (2017). Confocal microscopy imaging of the biofilm matrix. Journal of Microbiological Methods, 138, 50–59. https://doi.org/10.1016/j. mimet.2016.03.002 9