Inglês (pdf)
Artigo em XML
Referências do artigo
Tradução automática
Enviar este artigo por email
Citado por SciELO 
Similares em
SciELO 
Permalink
1. Introduction
Listeria is a Gram-positive, rod-shaped, non-sporulating, facultative intracellular exquisitely adaptable environmental bacterium belonging to the family Listeriaceae (Klump & Loessner, 2013; Korsak & Szuplewska, 2016; Komora et al., 2020). Currently, the genus Listeria contains 21 different species (NCBI, 2021). In food industry, one of the most important pathogens in this genus is Listeria monocytogenes that can cause listeriosis in humans and animals which is a health concern throughout the world (Korsak et al., 2012; Leylak & Buzrul, 2020). However, very rarely, L. ivanovii, L. innocua, and L. seeligeri have also been associated with disease in humans (Jadhav et al., 2012; Korsak & Szuplewska, 2016). Listeriosis is a sporadic disease, which is often associated with consumption of contaminated foods. Pregnant woman, neonates, adults with underlying disease (cancer, AIDS, diabetes, chronic hepatic disorder, transplant recipients), the elderly (≥ 65 years), and other immunocompromised individuals are at the highest risk (Rahimi et al., 2010).
L. monocytogenes is a serious threat to the food industry since it can survive the most common stress levels present in the food processing environment such as high salinity, acidity, osmotic pressure, refrigeration temperatures and low water activity (Jadhav et al., 2012). L. monocytogenes has been detected in many types of environments including rotting vegetables, untreated or treated sewage water, sludge, agricultural produce and in certain types of food items (Safraz et al., 2017). Many raw and processed foods such as unpasteurized milk, soft cheeses, dairy products, red meats, poultry, sea foods, vegetables, fruits, sausages, smoked fish, salads, deli meat, and refrigerated ready-to-eat (RTE) foods have been identified for their risk of listeriosis (Valimaa et al., 2015). Of these foods, milk and other dairy products especially raw and soft cheeses have been implicated in about half of all the listeriosis outbreaks (Sarker & Ahmed, 2015; Komora et al., 2020).
Biocontrol assays have received renewed attention in recent years as a possible antibiotic alternative to eliminate or control foodborne pathogens infections. Many studies have investigated the use of various phages to control of foodborne pathogens, including L. monocytogenes, Salmonella spp., Staphylococcus aureus, Campylobacter jejuni, Bacillus cereus and Escherichia coli (Goodridge & Bisha, 2011). Biological agents such as bacteriophage (also called phage) or bacteriocin are also used to reduce the viable counts of bacteria on foods (Criscuolo et al., 2017; Komora et al., 2020). Phages are often used in high concentrations to inactivate foodborne pathogens, such as E. coli O157:H7, Salmonella, L. monocytogenes and Camplobacter spp. in different foods (Akhtar et al., 2014). The use of phages in foods is a promising tool for food safety at pre-harvest and post-harvest stages of food production, and storage. There are some advantages of phage application in foods, such as safe, effective, and specific in reducing pathogens in several types of foods, without affecting either normal flora or the organoleptic properties of the food. Bacterial reductions from 0.9 to 6.8 log10 Colony Forming Units (CFU) or even total elimination was reported in many phage applications studies on food (Galarce et al., 2016). Listeria specific phages have been investigated in detail as biocontrol agents on foods so far (Leverentz et al., 2003; Soni et al., 2010; Rossi et al., 2011; Oliveira et al., 2014; Lee et al., 2017; Komora et al., 2020). To date, more than 500 listerial phages isolated from several food products such as meat, dairy, vegetables, fish etc. have been characterized. Most of them have been used in biocontrol assays (Soni & Nannapaneni, 2010; Klump & Loessner, 2013). Some phage preparations have been approved as safe status by the Food and Drug Administration (FDA) and recognized by the United States Department of Agriculture (USDA) as an antimicrobial processing aid to combat L. monocytogenes in foods. ListShield (LMP-102) and Listex P100 are two listerial phage preparations commercialized and marketed so far. ListShield LMP102 was the first phage-based products developed by Intralytic Inc., USA for use as an antimicrobial agent against L. monocytogenes contamination of RTE foods. ListShield LMP102 containing a cocktail of six phages is also used for surfaces in food production facilities. In 2011, Listex P100 containing a single phage was another anti-Listeria phage preparation developed by Micreos Food Safety, The Netherlands, formerly EBI Food Safety (Oliveira et al., 2014; Teng-Hern et al., 2014). Of these two phages, Listex P100 is approved for food preservation on RTE meats and deli products at levels not to exceed 109 plaque forming units (PFU) g-1 (Soni & Nannapaneni, 2010). This phage preparation has also been used in several studies showing its efficacy in removing L. monocytogenes contaminations (Soni & Nannapaneni, 2010; Soni et al., 2010; Rossi et al., 2011; Soni et al., 2012; Silva et al., 2014; Oliveira et al., 2014; Gutierrez et al., 2017; Komora et al. 2020). Many listerial phage preparations have also been tested in the dairy products to reduce the occurrence of L. monocytogenes strains (Carlton et al., 2005;Guenther & Loessner, 2011;Silva et al., 2014;Perera et al., 2015;Lee et al., 2017).
On the other hand, there is little information about the phage application on L. monocytogenes in milk. If there is a flaw during the pasteurization of milk, there may be a risk of L. monocytogenes existence. Moreover, there may also be a possibility of contamination of milk by L. monocytogenes after the heat treatment.
Therefore, the objectives of this study were (i) to determine the effectiveness of Listex P100 in reducing L. monocytogenes in pasteurized milk and broth as a function of storage temperature and storage duration (ii) to describe the reduction of L. monocytogenes counts in broth and milk by use of a suitable mathematical model.
2. Materials and methods
Pasteurized milk samples
Pasteurized (72 °C for 15 s) cow’s milk used in this study was purchased commercially on the day of the assay in 1000 mL tetra packs or glass bottle from local supermarket in Ankara, Turkey. Shelf life of the milk was between 10 - 14 days if stored in refrigerator without opening the package. Milk samples close to the expiry date were not selected. Prior to the experimental studies, covers of pasteurized milk samples were opened in a sterile cabin at room temperature.
Bacteria, phage and growth condition
The bacteria used in this study were as follows: L. ivanovii ATCC19119, L. monocytogenes ATCC7644, L. monocytogenes PL2, L. monocytogenes PL3, L. monocytogenes PL9, and L. monocytogenes PL10. L. monocytogenes PL2, PL3, PL9, and PL10 strains (serotype 1/2a) were isolated from RTE foods in Ankara/Turkey and molecular characterized previously. L. ivanovii ATCC19119 was used as a helper strain to determine the titer of the Listex P100 (phage P100 or P100 phage) as described by Silva et al. (2014). All strains used in this study were obtained from the culture collection of Food Microbiology Laboratory, Department of Food Engineering, Ankara University, Ankara, Turkey. Strains were grown individually on Tryptic Soy Broth supplemented with 0.6% of yeast extract (TSBYE) (Sigma, Germany) at 35 °C for 20 - 24 h. The Listeria spp. strains were maintained at -20 °C with 30% (v/v) glycerol (Merck, Germany).
Phage P100 characterized by its broad spectrum toward L. monocytogenes strains, was used in this study. The phage concentration was approximately 1011 PFU mL-1 in physiological saline buffer (PBS) (0.85% NaCl). Phage P100 was stored at refrigerated conditions (4 °C) during the study. Phage P100 stock concentration was also approximately 1011 PFU mL-1 by plaque formation assay. Stock solution of phage P100 was diluted in PBS for preparing the desired concentrations for phage application.
Preparation of bacteria for inoculum
L. monocytogenes strains were grown individually on Tryptic Soy Agar supplemented with 0.6% of yeast extract (TSAYE) (Sigma, Germany) at 35 °C for 20 - 24 h. An individual colony of each strain was transferred into a flask with 10 mL of TSBYE at 35 °C for 20-24 h. Overnight broth cultures of L. monocytogenes strains were centrifuged at 6000×g for 5 min at 4 °C and resuspended in PBS. Suspensions were diluted with PBS to approximately 106 CFU mL-1. (Oliveira et al., 2014). The concentration of L. monocytogenes strains was confirmed by MacFarland Densitometers (Den-1B).
Determination of host range of phage
To determine phage lytic spectrum against four different L. monocytogenes strains used in this study, double agar overlay plaque assay was used first to screen the strains as previously described by Akhtar et al. (2017) with a few minor modifications. Briefly, overnight cultures were used to prepare an exponential phase culture. Phage suspensions (100 µL phage solution with a titer of 108 PFU mL-1 and 10 mM CaCl2) were prepared in a sterile test tube, and then 150 µL of the exponential phase L. monocytogenes strains with a concentration of approximately 106 CFU mL-1 was added, separately. After 30 min of incubation at 35 °C, 5 mL of soft TSAYE (7.5% agar) was poured on the phage-bacterium mixture. The resulting mixture was gently vortexed and spotted on the pre-solidified TSAYE plate containing 10 mM CaCl2. After solidification of agar for 30 min at room temperature, the plates were incubated at 35 °C for 20 - 24 h to determine plaque formation. The lytic activity was checked for the formation of clear areas. Control plates were untreated with phage.
Preparation of pasteurized milk samples
Pasteurized milk samples were screened for the presence or absence of Listeria spp. Pasteurized milks used in this study were diluted in PBS, followed by spread plating on Palcam agar (Merk, Germany) to test for the presence of typical Listeria spp. colonies as described by Oliveira et al. (2014). L. monocytogenes ATCC7644 positive control and one uninoculated media negative control were used for each set of concurrently analyzed samples. After incubated at 35 °C for 20 - 24 h, samples found to be negative were used in biocontrol studies.
Phage treatment in fresh broth medium and pasteurized milk
Phage application assays were carried out at 4 and 30 °C in parallel batches in 30 mL fresh TSBYE and pasteurized milk added of 10 mM CaCl2. All the batches were incubated at several incubation times (on days 1, 2, 3, 4 and 5) in this study. Three experimental sets were formed in this study: (i) fresh TSBYE and pasteurized milk untreated with phage and bacteria (negative control), (ii) fresh TSBYE and pasteurized milk contaminated with only bacteria (positive control), and (iii) fresh TSBYE and pasteurized milk inoculated with bacteria and after phage (biocontrol experiments).
All the batches of 30 mL were contaminated with 100 µL of bacterial suspensions with a concentration of approximately 106 CFU mL-1. Then, one aliquot of each L. monocytogenes infected batch was inoculated with 100 µL of the phage suspension at approximately 108 PFU mL-1. Finally, 10 mM CaCl2 was added. All samples were mixed thoroughly to ensure homogenous distribution of the pathogen and phage, as reported by McLean et al. (2013) with some minor modifications. All samples were then incubated at 4 and at 30 °C for a total of 5 days.
Determination of bacterial counts and phage titer
Bacterial viable counts (CFU mL-1) and phage concentrations (PFU mL-1) from samples were performed immediately, after addiction of bacteria and phage, and after 1, 2, 3, 4 and 5 days of incubation of storage at 4 and 30 °C. To determine viable bacterial counts, aliquots were removed from each sample at defined intervals and serially diluted in PBS. Then, 100 µL of dilutions were spread plated on TSAYE and incubated at 35 °C for 24 - 48 h until typical Listeria colonies could be enumerated (Guenther et al., 2009). The data were plotted as log10 CFU mL-1.
To determine the phage titration from broth medium and pasteurized milk samples of phage treatments, the samples were centrifuged at 6000×g for 10 min. The supernatants of TSBYE medium samples were then filtered through a 0.45 µm por size filter (Sartorius, Germany), but supernatants of pasteurized milk samples were treated with chloroform (Merck, Germany) at 1% (v/v) after centrifugation to destroy bacterial cells and other potential contaminant bacteria (Arachchi et al., 2013). In this assay, the bacteria free samples of TSBYE medium and pasteurized milk were first serially diluted in PBS later. Phage titers were determined using double agar overlay plaque assay using a TSAYE as described previously (Soni et al., 2010). The plates were incubated at 35 °C for 24 - 48 h until plagues could be enumerated. At the end of the incubation period, the visible plaques were counted, and the data were expressed as log10 PFU mL-1.
Model
Weibull model proposed by Mafart et al. (2002) was used to describe the reduction of L. monocytogenes by the phage Listex P100 in this study:
()1
where S(t) is the survival ratio [S(t) = N0/N(t) with N0 and N(t) are the initial (t = 0) and momentary (at time t) numbers of L. monocytogenes, respectively], δ is the time parameter (in day) which represents time necessary to obtain the first decimal reduction i.e., it is the time required reduce N0 to N0/10, n is the unitless shape parameter which determines the shape of the survival curve. If n < 1 then the shape is concave, if n = 1 then a straight line is observed and if n > 1 the shape is convex.
Note that different versions of the Weibull model were proposed by different researchers (see Peleg, 1999 and van Boekel, 2002 for example); however, they all produce the same fit since their shape parameters are common (Buzrul, 2007). The above equation was used because time parameter (δ) could be used to understand the phage resistance of L. monocytogenes - see below.
The non-linear regression was performed by using SigmaPlot (Version 12, Chicago, IL, USA). The goodness-of-fit of the Weibull model was evaluated by using adjusted determination coefficient (R2 adj) and root mean square error (RMSE) value. Although R2 is generally used to understand how good the model fits the data, it is not recommended (Granato et al., 2014; Leylak et al., 2020). R2 adj may be preferable since it also considers the number of parameters in the model (Granato et al., 2014). Nevertheless, RMSE is the most informative indices of all (Ratkowsky, 2004; Öksüz & Buzrul, 2020). Smaller RMSE values indicate better the fit.
Statistical analysis
All experiments were repeated three times and all treatments were replicated at least three times in each experiment in this study. The bacterial and phage data were transformed to log10 units. All statistical analyses were carried out using SPSS (Version 16, Chicago, IL, USA). The analysis of one-way variance (ANOVA) followed by Tukey’s test was applied to determine the differences between broth model system and pasteurized milk treated with phage and control. Significant differences between incubation times were also analyzed. A significant difference was defined at p ≤ 0.05.
3. Results and discussion
The ability of phage P100 against four different L. monocytogenes strains in pasteurized milk and fresh TSBYE samples stored at 4 and 30 °C was tested on days 1, 2, 3, 4 and 5 in this study. The sensitivities of L. monocytogenes strains to P100 phage were confirmed by soft agar overlay assay. P100 phage produced large plagues of approximately between 1.7 - 2.0 mm in diameter against four different L. monocytogenes strains - results not shown. Although L. monocytogenes PL2 and PL9 displayed 1.7 mm and 1.9 mm plaque size, L. monocytogenes PL3 and PL10 had both 2.0 mm plaque size. No Listeria spp. were found in the samples which was confirmed by the microbiological evaluation of pasteurized milk before inoculation with phage P100.
Fig. 1,Fig. 2,Fig. 3andFig. 4 show the impact of Listex P100 on L. monocytogenes strains PL2, PL3, PL9 and PL10, respectively. The data of the remaining phage titer in both samples after phage treatment at both temperature levels during storage were not determined in this study. More reductions were observed in broth medium than in pasteurized milk regardless of the temperature level where the strain PL2 was the exception at 4 °C. Almost identical reductions were obtained for L. monocytogenes PL2 at 4 °C in broth medium and pasteurized milk (Fig. 2band2d). Moreover, reductions were significantly (p ≤ 0.05) higher at 30 °C than at 4 °C for all the strains showing the effectiveness of phage at higher temperature levels. The reductions of L. monocytogenes strains by the phage treatment were between 2.7 to 3.4 log10 units at 30 °C and 1.4 to 2.1 log10 units at 4 °C after 4 days of incubation in broth medium whereas 1.9 to 2.9 log10 units and 1.0 to 1.6 log10 units were observed after 4 days of incubation in milk at 30 and 4 °C, respectively. It should be noted that lower reduction effects of phage P100 were found in this study in comparison to previous works (Soni & Nannapani, 2010; Oliveira et al., 2014; Migueis et al., 2017). L. monocytogenes counts in both media could not be eliminated by the phage P100 at both temperature levels and this may be explained by several ways. Firstly, it was possible that some compounds such as proteins, carbohydrates, calcium, and phosphorus in pasteurized milk (Tolba et al., 2014) may inhibit the attachment of phages to cells. This could also be the reason of observing lower reductions in milk than in broth. Secondly, applied phage dosage, which is a critical factor, may be insufficient. High dose treatment of phage (> 108 PFU mL-1) can be used to eliminate more L. monocytogenes viable cells in pasteurized milk. Another reason to have less phage effectiveness may be the development of phage resistant mutants since phage cocktails were not used in this study. Isolation of natural origin listerial phages may be showed high specify against L. monocytogenes isolated in Turkey.
Figure 1 Impact of P100 phage application on L. monocytogenes PL2 counts in broth (a) and in pasteurized milk (b). Gray squares and red circles are the experimental data points at 4 and 30 °C, respectively. Solid green lines indicate the fit of the Weibull model [Eq. (1)].
Figure 2 Impact of P100 phage application on L. monocytogenes PL3 counts in broth (a) and in pasteurized milk (b). Gray squares and red circles are the experimental data points at 4 and 30 °C, respectively. Solid green lines indicate the fit of the Weibull model [Eq. (1)].
Figure 3 Impact of P100 phage application on L. monocytogenes PL9 counts in broth (a) and in pasteurized milk (b). Gray squares and red circles are the experimental data points at 4 and 30 °C, respectively. Solid green lines indicate the fit of the Weibull model [Eq. (1)].
Figure 4 Impact of P100 phage application on L. monocytogenes PL10 counts in broth (a) and in pasteurized milk (b). Gray squares and red circles are the experimental data points at 4 and 30 °C, respectively. Solid green lines indicate the fit of the Weibull model [Eq. (1)].
Table 1 Parameters (δ and n) of the Weibull model ± their standard errors and goodness-of-fit indices (R2 adj and RMSE) of each fit
Although the ability of phage P100 against four different L. monocytogenes strains in fresh broth medium and pasteurized milk samples stored at 4 and 30 °C was tested up to 5 days (results not shown), reductions up to 4 days were displayed. This is because slight increases were observed at day 5 (0.2 - 0.6 log10 units at 30 °C and 0.1 - 0.2 log10 units at 4 °C in broth, and 0.2 - 0.4 log10 units at 30 °C and 0.1 - 0.3 log10 units at 4 °C). Hence, for modeling purposes data up to 4 days were used.
The fit of the Weibull model [Eq. (1)] are also shown in Fig. 1,Fig. 2,Fig. 3andFig. 4. It could be said that model produced reasonable fit to the data which can be also judged by the high R2 adj (≥ 0.936) and low RMSE (≤ 0.212) values given in Table 1. Furthermore, survival curves obtained in broth medium were all concave (Fig. 1a,Fig. 2a,Fig. 3aandFig. 4a) indicating that the shape parameters (n) were less than 1 (Table 1). On the other hand, both concave (n < 1) and convex (n > 1) survival curves were observed in pasteurized milk. In fact, some of the concave survival curves obtained in milk were close to linear. For example, strains PL2 (Fig. 1b), and PL9 at 4 and 30 °C (Fig. 3b) were almost linear with n ≥ 0.9 (Note that survival curves are linear if n = 1).
Time parameter (δ) was used to differentiate the phage resistances of the strains. Low δ values revealed sensitive strains whereas resistant strains had higher δ values (Table 1). It was also possible to evaluate the higher reductions in broth than in milk by comparing the δ values given in Table 1. Much lower δ values were estimated by the model fits in broth medium. Moreover, at low temperature higher δ values were observed which indicates reduction at 4 °C was more difficult than 30 °C. Overall, it could be said that PL2 was the most phage-resistant L. monocytogenes strain while PL10 was the most sensitive one in broth at both temperature levels (Table 1). Maximum reduction in viable counts for the strain PL10 in broth were 3.4 and 2.1 log10 units after 4 days at 30 and 4 °C, respectively. On the other hand, in milk although PL2 was still the most phage-resistant L. monocytogenes strain at 30 °C, PL9 was the most-resistant one at 4 °C. Only 1 log10 reduction was observed for PL9 in milk at 4 °C.
Time parameter (δ) of the Weibull model was a good indicator to evaluate the phage-resistances of the strains (Table 1); however, it represents only the first decimal reduction (Note that time necessary for 2 log10 reduction is not 2·δ, but 21/n δ. Similarly, time for “m” number of reduction is m1/n δ). It may be a better approach to compare resistances based on the time required for 5 or 6 log10 reduction, but maximum reduction attained in this study was 3.4 log10 (strain PL10 in broth at 30 °C). In some cases, only 1 log10 reduction was observed (strain PL9 in milk at 4 °C). Moreover, the initial inoculum was about 106 CFU mL-1 for all strains. This is, of course, higher than the expected contamination in pasteurized milk hence δ can be a good option to better understand the phage-resistances of the strains in this study. Comparison of the δ values given in Table 1 revealed that huge variabilities between strains were possible. Therefore, calculations should be done according to the most phage-resistant strain. In addition, resistant phage in a medium may be sensitive in another medium and this was also observed in this study. PL2 was the most phage-resistant strain among all in broth at 4 °C; however, it was the second after the strain PL9 in milk at the same temperature (Table 1).
Shape parameters (n) listed in Table 1 and survival curves shown in Fig. 1, Fig. 2, Fig. 3 and Fig. 4 indicated both concave (n < 1) and convex (n > 1) survival curves. This may be because of temperature and medium (Peleg & Cole, 2000). For example, in broth medium only concave survival curves were observed at 30 and 4 °C. On the other hand, in pasteurized milk although concave survival curves were abundant some of them were close to linear (Fig. 1b and 3b). In addition, convex survival curves in milk at 4 °C were also obtained (Fig. 1b and 4b).
Biocontrol assay against contamination by L. monocytogenes by using phages is a promising tool of green strategy. The use of phages is also a viable alternative to chemical antimicrobials against foodborne pathogens. Earlier works on using phage to control the growth of L. monocytogenes were done in fresh-cut fruits and vegetables (Leverentz et al., 2003), honeydew melon (Leverentz et al., 2004), fresh-cut apples (Leverentz et al., 2006), cooked ham (Holck & Berg, 2009), in RTE poultry (Bigot et al., 2011), soft ripened white mold and red-smear cheeses (Guenther & Loessner, 2011), lettuce, apples, cheese, smoked salmon and frozen foods (Perera et al., 2015) and chicken breast (Yang et al., 2017).
Phage P100, which was received GRAS status for application in all foods (Guenther et al., 2009), has been used effectively against L. monocytogenes in soft cheeses (Carlton et al., 2005), RTE foods (Guenther et al., 2009), catfish fillets (Soni et al., 2010), raw salmon fillet tissue (Soni & Nannapani, 2010), fresh sausage (Rossi et al., 2011), queso fresco cheese (Soni et al., 2012), RTE sliced deli meat products, (Prabhu et al., 2016), fresh cut fruits and fruit juices (Oliveira et al., 2014), soft cheeses (Silva et al., 2014), dry-cured ham (Gutierrez et al., 2017) and sashimi (Migueis et al., 2017) so far. Although L. monocytogenes is effectively controlled by pasteurization, its presence in the finished product is possible because of post-pasteurization contamination from sources in the milk processing plant environment or by the operator error. Even very low numbers of L. monocytogenes in processed dairy products can multiply rapidly to reach dangerous levels despite the proper refrigeration (Xanthiakos et al., 2006). To the best of our knowledge, little is known about the efficacy of phage P100 to control of L. monocytogenes strains in pasteurized milk. A notable example is the work of Lee et al. (2017) and Komora et al. (2020). However, to test of the lytic ability of the phage against E. coli strains, some researchers have been carried out in milk and milk fermentation so far (Mclean et al., 2013; Tomat et al., 2013; Tolba et al., 2014).
In all assays performed, application of phage P100 in both samples was able to reduce L. monocytogenes cells at both temperature levels when the initial number of Listeria cells was about 106 CFU mL-1. It was interesting that, on day 5, there was no effect on L. monocytogenes populations with treatment phage P100 in both samples.
4. Conclusions
Listex P100 was used to reduce L. monocytogenes strains in broth and milk in this study. Reductions up to 3.4 log10 in broth and 2.9 log10 in milk were achieved. Survival data was successfully described by the Weibull model and the time parameter of the Weibull model was used to evaluate the phage-resistances of the strains. The results obtained in this study were noteworthy because the use of phages for biocontrol of L. monocytogenes in milk has received little attention. Further studies should be carried out at different inoculum levels of both bacterium and phage, and with the use of combined natural chemicals such as nisin.
