Effect of E-Beam Treatment on the Safety and Shelf Life of Mayonnaise Potato Salad

Abstract

The radioresistance of Listeria monocytogenes, Salmonella enterica serovar Enteritidis, and S. enterica serovar Typhimurium has been studied in a complex matrix like mayonnaise potato salad. D ₁₀-values of 0.56, 0.32–0.35, and 0.41–0.42 kGy were calculated for each organism, respectively. Keeping in mind these values, the microbiological criteria, the characteristics of the microorganisms, and a shelf life of the products of 20 days stored at 4°C, an irradiation treatment of 1 kGy was calculated to reach the food safety objectives. A duplication of the shelf life is also achieved with a dose of 1 kGy. Mayonnaise potato salad radiated with doses of up to 2 kGy showed negligible off-sensory characteristics.

Introduction

M ayonnaise is a seasoning sauce widely used as an ingredient for mixing with other raw or cooked foods. It is obtained by emulsifying edible oil in an aqueous phase with an acid ingredient (usually lemon juice or vinegar) that is stabilized with egg. The European Model of the Food Codex stipulates that the total fat content of mayonnaise (oil plus egg yolk) should be 78.5% and that of technically pure egg yolk is of no <6%. The pH of mayonnaise ranges typically between 3.2 and 4.2 (Michels and Koning, 2000; ICMSF, 2005). The salt content is usually between 3% and 11% of the aqueous phase. The level of acetic acid in the aqueous phase varies widely between 0.8% and 3.0%. The water activity (a _w) is ∼0.93 (Hwang and Tamplin, 2005). The relatively low value of this parameter restricts the growth of spoilage microorganisms to yeast, a few bacteria that include lactic acid and moulds, although the spoilage by the latter organisms is very rare since they are very sensitive to acetic acid (Smittle and Flowers, 1982). Among the yeasts, the most common are Zygosaccharomyces bailii and Pichia membranaefaciens, which may grow when there is a concentration of acetic acid as low as 3% (Thomas and Davenport, 1985). In addition, several species (L. fructivorans, L. plantarum, L. bucheri, and L. brevis) of the genus Lactobacillus have been reported (Smittle and Flowers, 1982; Muriana and Kanach, 1995) to possibly spoil the product.

Mayonnaise is an essential component of many ready-to-eat (RTE) foods, among which is the mayonnaise potato salad (MPS); other ingredients (minced tuna, carrots, peas, olives, etc.) may be used depending on recipe. The water content of MPS is higher than that of mayonnaise, and, of course, the above-mentioned mayonnaise parameters (pH, a _w, etc.) will be modified depending on the added ingredients. In particular, both the a _w and the aqueous phase of the final product will increase significantly, leading to higher rate of the microbial growth (Hwang and Tamplin, 2005). The pH together with refrigeration is the most useful manner to control the microbial growth. Pathogenic organisms cannot grow in commercial mayonnaise when either the pH is lower than 4.1 or the acidity is at least 0.7% in terms of acetic acid (Glass and Doyle, 1991). However, if any pathogenic microorganism is present in the sauce, when it is added as a condiment, the other ingredients will be contaminated and vice versa. This is the case of the MPS, where the pathogens, if present, may multiply if the storage conditions of the food are favorable. The possible contamination of Listeria monocytogenes in deli salads is one of the main causes of salad recalls in the United States by the Food Safety and Inspection Service (FSIS) of USDA (FDA, 2005; FSIS-USDA, 2005).

Despite the fact that the MPS may be prepared under optimal hygienic conditions, that is, using liquid egg pasteurized and refrigeration of the final product, the MPS is not totally free of contamination by pathogens. They can occasionally reach the product from many common sources, which include the ingredients, the process, and packaging equipment as well as the environment. Due to the refrigeration during storage, L. monocytogenes could be considered as the most dangerous organism because of its ubiquity, persistence, and psychrotrophic character. Nevertheless, the physical–chemical properties of the MPS are not completely favorable for the growth of this bacterium, although, if present, it will grow very slow since the pH is usually lower than 5.0 (Glass and Doyle, 1989, 1991; Farber and Peterkin, 1999). Several studies (Hwang, 2005; Hwang and Tamplin, 2005; Hwang and Marmer, 2007) showed that storage temperature was more significant than mayonnaise pH in influencing the behavior of L. monocytogenes in deli salads. A recent study found that L. monocytogenes was present in 4.7% of seafood salads and in 2.4% of deli salads (i.e., potato, poultry, tuna, pasta, egg, cheese, and coleslaw) in the United States. Data from other countries showed that L. monocytogenes was positive in 9.9% deli salads (FDA/USDA/CDC, 2001). In the Belgian retail market, L. monocytogenes was isolated from 20.8% to 27.3% of deli salads with different components, and an overall incidence of 21.3% in mayonnaise-based salads (Uyttendaele et al., 1999). The behavior of pathogen organisms (including L. monocytogenes) in salads is affected by the presence of food components associated with a more favorable growth environment (such as high nutrient content and pH buffering capacity) and mayonnaise composition (Erickson and Jenkins, 1991; Erickson et al., 1993; Hwang and Tamplin, 2005). For salads prepared with nonreal mayonnaise (e.g., reduced calories and/or fat mayonnaise) that may have a higher pH, a shorter storage time should be considered (Hwang and Marmer, 2007).

The pathogenic power of Salmonella spp. is well known. These bacteria, including S. enterica serovar Enteritidis and S. enterica serovar Typhimurium, are very common ubiquitous organisms in egg and egg products. In 2008 (EFSA, 2010), as in previous years, these serovars were the most frequently reported (79.9% of all known serovars in human cases). Eggs and egg products as well as products containing raw eggs continue to be the most important food vehicles (EFSA, 2010). Mayonnaise and dressings with final pH values above 4.1 offer a potential risk of foodborne outbreaks by Salmonella, or other pathogens (E. coli O157:H7 and L. monocytogenes), either because strains of these bacteria can be particularly acid-tolerant or because the acidulant type and its final concentration may not be adequate to kill such pathogens (ICMSF, 2005). Salmonella present a very low infective dose, lower than 100 cells g⁻¹ food (D'Aoust et al., 1975; Blaser and Newman, 1982; Gorris, 2005), although a value of even 28 cells has been also reported (Vought and Tatini, 1998).

Regulation European Commission No. 2073/2005 on microbiological criteria for food establishes a concentration of L. monocytogenes in food below 100 CFU/g, whereas for Salmonella spp. the requirement is of an absence in 25 g. According to the criterion, there is no doubt that Salmonella spp. is the target organism to be controlled when the food safety objective (FSO) is attempted to achieve. However, in 2007, this EC Regulation was substituted by another No. 1441/2007 (CEC, 2007), which textually designates a limit of 100 CFU/g during the shelf life of “RTE foods unable to support the growth of L. monocytogenes [automatically those with pH ≤4.4 or a _w ≤0.92, products with pH ≤5.0 and a _w ≤0.94] other than those intended for infants and for special medical purposes” but if the RTE is “able to support the growth of L. monocytogenes” the limit is “absence in 25 g” before the food has left the immediate control of the food business operator who has produced it. Additionally, the regulation notifies that the latter criterion is applied to products if the manufacturer is not able to demonstrate, to the satisfaction of the competent authority, that the product will not exceed the limit of 100 CFU g⁻¹ throughout the shelf life. For MPS, it is not possible to demonstrate the above statement. So, L. monocytogenes becomes other organism that needs to be controlled since the final product has a pH of 4.43 and an a _w of 0.99 (see the Materials and Methods section); therefore, it may support the growth of this organism.

Other organisms have also been isolated from both mayonnaise and several added mayonnaise foods, for example, Escherichia coli O157:H7. This bacterium was the causal agent of an outbreak in Oregon (Anonymous, 1993). It is very resistant to acidic conditions and it may remain viable for relatively long periods of time in such foods (Glass et al., 1993), even 36 days in added mayonnaise blue cheeses with a pH of 4.4 (Weagant et al., 1994). However, this organism is more sensitive than Salmonella spp. (Clavero et al., 1994) to the method assayed here for sanitizing MPS. Satphylococcus aureus may also be present, but neither it grows in refrigerated conditions (ICMSF, 1996a) nor does it form enterotoxins at a pH lower than 5.0 (Gomez-Lucía et al., 1987).

According to the former statements L. monocytogenes is excluded only when the food does not support its growth, that is, those foods with a pH ≤4.4 or a _w ≤0.92 or those with pH ≤5.0 and a _w ≤0.94 as well as a product with a shelf life lower than 5 days (ICMFS, 1996a; CEC, 2007). Since the MPS considered in the present work has a pH of 4.93 and a _w of 0.99, it becomes necessary to consider L. monocytogenes as a target organism and, therefore, to take into account the criterion to of absence in 25 g, which is also applied to Salmonella spp.

The irradiation is a very effective way to eliminate pathogens present in foods, including L. monocytogenes (Sommers et al., 2003; Zhu et al., 2005) and Salmonella spp. (Thayer et al., 1990; ICMSF, 1996b). Some researchers (Ahn et al., 2000; Jo and Ahn, 2000) have reported that the irradiation of food should be limited since it can produce changes in the aroma, color, and flavor, which can significantly affect consumer acceptance. Therefore, from a public health point of view, this work was envisaged to carefully adjust irradiation doses to RTE MPS to achieve an adequate level of microbial safety with minimum changes in sensory attributes to avoid rejection of the irradiated product by consumer.

From the technological and commercial points of view, there has been an attempt to extend the shelf life of the RTE MPS maintained at 4°C, which may be of great interest since this RTE product may be exposed a longer time in the refrigerated display cases.

Materials and Methods

Estimation of food safety and performance objectives and performance criteria

The MPS is an RTE product in which many pathogens may grow. It has to be stored under refrigeration (ideally <5°C) during the whole shelf life. Pathogenic organisms may be present, including Salmonella spp., but the low temperature restricts the growth to psychrotrophic organisms. As mentioned above, the EC criterion for L. monocytogenes is the same that for Salmonella spp. Therefore, the FSO of the MPS for both organisms will be 4 CFU/100 g (log₁₀ = −1.39). The fundamentals of risk assessment have been extensively explained in a previous publication (Cabeza et al., 2004).

The establishment of performance objectives (maximum frequency and/or concentration of a hazard in a food at a specific phase in the food chain before the time of consumption that provides or contributes to an FSO or adequate level of protection, as applicable) (Gorris, 2002) including the reasoning used to assess the potential contamination of an RTE meat product during slicing and packaging has also been widely described in a previous article (Cabeza et al., 2007). The conclusion reached in this article is that the contamination of cooked ham with L. monocytogenes during its transformation into an RTE product can reach, in the worst of cases, 10 cells/g (log₁₀ = 1.0). This is the initial level of hazard (H ₀ = 1), which may also be assumed for MPS. The same figure may be considered for Salmonella spp., since this bacterium is also a ubiquitous organism.

As Salmonella spp. do not grow in refrigerated (<5°C) MPS (ICMSF, 1996a; D'Aoust, 2005; Jay et al., 2005), no increase of the hazard will occur during the shelf life of the final product; that is, the performance objectives value is zero. Nevertheless, L. monocytogenes is a psychrotrophic bacterium. Thus, it is necessary to know the potential increase in numbers (Δ) during the storage period. The growth of L. monocytogenes in salads has not been a concern of researchers, probably due the low safety risk for the consumer of this type of product (John and Jenkins, 1991). Actually, the pH value together with the refrigeration is close to the limit reported for growth (Glass and Doyle, 1989, 1991; Farber and Peterkin, 1999). Nevertheless, if the shelf life of the salad is extended, the number of Listeria could be higher than expected, thereby increasing the product risk when consumed. In the report of the FDA (2003), many values of several authors on the increased numbers of L. monocytogenes in various products stored at 4°C–5°C are tabulated, from which an average increase of 0.105 log₁₀ units/day (mean of 40 data) may be calculated. This is a confidence value since the pH of the products mentioned by the FDA (2003) has values of over 5.0, whereas the pH of the MPS is 4.9. Assuming that the salad has a shelf life of 20 days at 5°C, there would be an average increase in numbers of (Δ₂₀) of 2.1 log₁₀ units during the first 20 days of storage. From these data (FSO and Δ₂₀), a performance objectives (PeO = FSO − Δ₂₀) of (−1.39 − 2.1) = −3.49 log CFU g⁻¹, that is, 3.23 × 10⁻⁴ CFU g⁻¹ may be calculated.

Similarly, from H₀, FSO, and Δ₂₀ values (in logarithmic terms) for Salmonella spp. (1, −1.39, and 0, respectively) and L. monocytogenes (1, −1.39, 2.1, respectively), the process objective (PrO: the effect in frequency and/or concentration of a hazard in a food that must be achieved by the application of one or more control measures to provide or contribute to an FSO or adequate level of protection, as applicable) (Gorris, 2005) may be calculated for both bacteria by the equation (H₀ − PrO + Δ₂₀ = FSO) of the ICMSF (2002). They are the following: 2.39 log units reductions of the load of Salmonella spp. (in this case Salmonella Enteritidis and Salmonella Typhimurium) would be required to reach the FSO in relation with these bacteria; however, a 4.49 log units reduction would be necessary in the case of L. monocytogenes to achieve the FSO for this organism.

Mayonnaise and salad preparation

The mayonnaise composition was (w/w) as follows: soy oil (63.25), water (17.06), pasteurized liquid egg (13.30), 10 degree red wine vinegar (4.20), sugar (1.13), salt (1.03), and potassium sorbate (0.03). Ingredients were mixed with a homogenizer and kept at 4°C until use, which was a period of no more than 2 h.

The MPS composition (w/w) was as follows: precooked vegetables ([potato/carrots/peas] [75/19/6]) (56.00), mayonnaise (25.00), processed tuna in olive oil (7.50), sliced olives (6.50), and minced hard-boiled egg (5.0). Three kilograms of salad was prepared by mixing the ingredients in a domestic blender at room temperature. The final pH and a_w were 4.9 and 0.99, respectively.

Portions of the salad (about 80 g) were placed into clean plastic cylinder containers (5 cm diameter × 6.5 height) and screw cap closed. Samples were stored at 4°C and transferred to the irradiation plant into insulated polystyrene boxes, located 60 km from the laboratory.

Organisms

One strain of Salmonella enterica serovar Enteritidis (CECT 4300) and Salmonella enterica serovar Typhimurium (CECT 443) and another of L. monocytogenes Scott A (CIP 103575) were used. The strains were maintained by freezing (−40°C) in trypticase soy broth (Difco, BD, Sparks, MD) with 10% glycerol added as cryogenic agent. Fresh cultures were prepared for each experiment by removing a piece of frozen culture from vials and inoculating it into 9 mL of trypticase soy broth and then incubating it at 32°C for 24 h. The culture was then centrifuged (at 4°C) and the pellet suspended in a sterile test tube with 10 mL sterile saline, which yielded a bacterial load that was close to 10⁸ cells/mL. A large number of cells were used in experiments to precisely calculate the death kinetic parameters.

Samples contamination and irradiation treatment

For microbiological purposes, several cylinder containers were opened, protected by a Bunsen flame, and contaminated by adding 1 mL of the above bacterial suspension into the material, and then gently mixed with a sterile glass rod.

Contaminated MPS containers (for microbial analysis) and uncontaminated MPS containers (for sensorial analysis) were refrigerated at 4°C and were taken for irradiation in a plant (IONISOS Iberica sterilization SA, Tarancón, Cuenca, Spain) under an electron beam radiation source, which operates at 10 MeV. The radiation doses employed were between 0.5 and 3 kGy and the dose absorbed by samples was checked by determining the absorbance of cellulose triacetate dosimeters (ASTM, 2000) simultaneously irradiated with samples. Experiments were made in triplicate and performed at room temperature (18°C–20°C). The product temperature increase during treatment was <2°C. For nonmicrobiological purposes, doses of 1 and 2 kGy were applied to an appropriate number of containers with the salad. After irradiation treatment, samples were handled as is the norm in microbiological experiments and stored at 5°C until use.

Microbial analyses

To count the salmonella survivors, about 10 g of the material was weighed and homogenized with 90 mL of a sterile saline solution in a Stomacher bag. Counts were determined on the surface of plates with Hektoen and Salmonella Shigella agars. Since the ingredients were not completely sterile, selective media were used to avoid the count of background microbiota. In the case of Salmonella Enteritidis and Salmonella Typhimurium, the two selective media mentioned above, were chosen to detect a potential undercounting of injured cells. For L. monocytogenes, the Palcam medium was employed. Inocula were placed onto the agar surface by use of a spiral plate system (model Eddy Jet; IUL Instrument, Barcelona, Spain). Petri dishes were incubated at 36°C for 24 h. Colonies were enumerated with an automatic counter (model Countermat Flash; IUL Instrument).

Survival curves were obtained by plotting the logarithm of the number of survivors against the dose assayed. Decimal reduction dose (D-values) were calculated from the linear regression equation of survival curves. The correlation coefficients (R ²) of curves and the 95% confidence limits of D-values were calculated by Excel (Microsoft, Redmond, WA

Total viable counts were done by the same method used for salmonella, but the media trypticase soy agar was used. In this case, plates were incubated at 32°C.

Shelf life determination

The shelf life of irradiated salad was determined by periodically counting the bacterial number and analyzing sensory features (odor and visual appearance) in samples stored at 4°C. Nonirradiated samples in aerobically packaging were used as controls. From a bacterial point of view, the end of the shelf life was considered when the total viable counts exceeded the 10⁷ CFU/g level.

Sensory analysis

To determine the possible sensory differences among the nontreated (0 kGy) and irradiated samples (1 and 2 kGy) of MPS, a triangular, a rank order, and a descriptive test were performed. Samples were evaluated by a panel of 20 tasters (10 females and 10 males) selected among the members of the department (Departamento de Nutrición, Bromatología y Tecnología de los Alimentos). The panel members had been previously trained in the sensory assessment of salads. The evaluation was performed between meals, after breakfast, and before the midday meal. To reduce fatigue, panel members performed three sessions per day with a minimum break of 1 h between sessions.

Three independent tests were performed to evaluate appearance, odor, and flavor. Before sensory analysis, samples were removed from the refrigerator and maintained in a temperature-controlled room at 10°C for no more than 1 h. The evaluation of the appearance of the MPS was carried out in the same containers used during the experimental process. These remained closed during the examination period by panel member; thus, they had not direct contact with the content. In the odor and flavor analysis, members of the panel received 50 g of the MPS at about 10°C on a Petri plate. The plastic cylinders were opened shortly before the sensory evaluation took place.

The evaluations were performed in individual booths built according to the International Standards Organization DP 6658 (ISO, 1981a) criteria. The tasters received unsalted crackers and room temperature water to clean the palate between samples. White fluorescent light was used during appearance analysis. Odor and flavor of samples were evaluated under red light conditions just after opening the bags. The sensory analyses were carried out after treatment (zero to first day) and during the storage at 4°C. The evaluation of flavor samples was only performed after the microbiological analysis indicated that the samples were suitable for consumption.

The triangle test (ISO, 1981b) was performed by the forced-choice option, in which the tasters must choose the sample that, in their opinion, is different. All the possible combinations of untreated and irradiated samples were tested. To complement the triangle test, judges were asked to indicate their reasons for selecting one particular sample of the three used in the analysis.

At the rank order test, the judges were instructed to rank samples in order of preference, according to the proximity of the sensory characteristic (appearance or odor or flavor) of the sample analyzed to the optimal sensory quality of the MPS. For this, a 3-point scale (in which 1 corresponded to the lowest preference and 3 to the highest preference) was used. No repetitions were allowed. Results of the rank order test were used to obtain the sum of ranks, which corresponds to the sum of scores of MPS preference for a specific sensory characteristic (calculating the sum of the products of values given to each sample on a 3-point scale [from 1 to 3] by the number of times that each sample was allocated to a specific score). The significance level of data obtained in these tests was determined by Friedman's rank addition according to the model proposed by Joanes (1985) and the tables for multiple comparison procedures for analysis of ranked data (Christensen et al., 2006).

Panelists were also asked to give information about the MPS (appearance, odor, flavor, mayonnaise emulsion stability, solid component texture, and any off-sensory aspect) following a profile descriptive analysis.

For this stage of the evaluation, the panelists were made familiar with the necessary terms to describe the sensory characteristics of the MPS (such as general appearance: color and brightness; odor: richness and intensity, off-odor absence; taste: acid and rancid intensity, richness of taste notes, off-taste absence, juiciness, and after-taste intensity) as well as the expected off-sensory features resulting from the E-beam treatment (off-odors and taste such as hot culture medium, burnt beef broth, sulfuric, metallic, scalded feather, burnt feather, pungent pepper, cooked cabbage, spoiled milk, or spoiled vacuum meat). They were also asked to qualify the intensity of these sensations with the following terms: weak, medium, or strong.

Members of the panel were only given one sample of the MPS (50 g portion at 10°C) to evaluate all the sensory characteristics jointly. The sensory analysis was performed after treatment (0 days) and 7, 14, 22, 30, and 47 days of storage. The MPS were first judged individually by panelists and then collectively by all the panelists.

Statistical analysis

The coefficient of determination (R ²) of curves, the test F, and the 95% confidence limits of D-values were calculated by Excel (Microsoft).

Results and Discussion

Food safety aspects

As expected, the response of organisms to the irradiation treatment fits first-order inactivation kinetics. No survivors of Salmonella Enteritidis, Salmonella Typhimurium, and L. monocytogenes were detected when treatments of 3 kGy were applied. Table 1 shows the survivor equations, the irradiation decimal reduction (D ₁₀-value), and the correlation coefficients (R ²). The thermal processes applied to the MPS ingredients are focused both to attain the sanitation of liquid egg (pasteurization) and to soften the vegetables tissues of potato, carrots, and peas (cooking). The final product may contain some spoilage organisms that either survive the treatments (mainly spore-forming bacteria) or reach the product during postprocess operations. In fact, this was observed in the shelf-life experiments, when mesophiles counts of close to 10⁴ CFU/g were observed (Table 2) in the first day. Accordingly, selective media were necessary for counting the Salmonella and Listeria survivors to the irradiation treatments. Since the survivor equations for Salmonella Enteritidis and Salmonella Typhimurium in both media were very close (compare in Table 1 the D ₁₀-values in the Salmonella Shigella agar vs. those determined in the Hektoen medium) and the correlation coefficients very similar, it was assumed that selective media did not affect the growth of these bacteria. The highest D ₁₀-values for Salmonella Enteritidis and Salmonella Typhimurium will be those used to estimate the process criterion. The selective Palcam medium was the only one used for counting surviving Listeria against the irradiation treatment.

Table 1.

Destruction Equations, Irradiation Reduction Values (D ₁₀), Correlation Coefficients (R ²), 95% Confidence Limits, and Process Criteria for Salmonella Enteritidis, Salmonella Typhimurium, and Listeria monocytogenes in Mayonnaise Potato Salad

Organism ^a	Equation	D₁₀ -Value (kGy)	95% CL	R²	Process criterion (kGy)
Salmonella Enteritidis (SS)	log CFU/g = 8.121 − 3.1692D	0.32	0.30–0.34	0.98	NU
Salmonella Enteritidis (HK)	log CFU/g = 7.921 − 2.4390D	0.35	0.33–0.38	0.98	0.84
Salmonella Typhimurium (SS)	log CFU/g = 7.921 − 2.4390D	0.41	0.39–0.43	0.99	NU
Salmonella Typhimurium (HK)	log CFU/g = 8.091 − 2.3988D	0.42	0.40–0.45	0.99	1.00
Listeria monocytogenes (PL)	log CFU/g = 6.051 − 1.7794D	0.56	0.52–0.59	0.98	2.51 (0.68)^b

SS, HK, and PL: bacteria counts were made in Salmonella Shigella (SS), Hektoen (HK), and Palcam (PL) agar, respectively.

In parentheses is the actual process criterion according to European Commission regulation (see text).

95% CL, 95% confidence limits; D, absorbed radiation (kGy); NU, not used for the process criterion determination (see text).

Table 2.

Changes (log CFU/g) in the Aerobic Mesophiles Counts of E-Beam-Treated Mayonnaise Potato Salad During Storage at 4°C

	Days of storage
Doses (kGy)	1	22	30	47
0	3.76	6.60	7.83	>8.00
1	<3.00	<3.00	<3.00	<3.00
2	<3.00	<3.00	<3.00	<3.00
3	<3.00	<3.00	<3.00	<3.00

The D ₁₀-values for this strain of Salmonella Enteritidis were similar to those determined previously (Ordóñez et al., 2007) in broiler raw meat (D ₁₀-value = 0.37 kGy) but slightly lower (D ₁₀-value = 0.41–0.43 kGy) than those calculated in dry fermented sausage (Cabeza et al., 2009). However, these radioresistant parameters may be considered to be normal since they are in the range reported by several authors in a variety of products (Thayer et al., 1990; Serrano et al., 1997; Jakabi et al., 2003). Publications (Tarkowski et al., 1984; Thayer et al., 1990; Grant and Patterson, 1992) have repeatedly confirmed that the resistance of Salmonella Typhimurium against irradiation is significantly higher than that of Salmonella Enteritidis. The same assumption was found in the present work and previously in dry fermented sausages for the same strains here used (Cabeza et al., 2009) although in the latter foods the radioresistance was higher (0.42 for Salmonella Enteritidis and 0.54 kGy for Salmonella Typhimurium), probably as a consequence of the lower a _w of sausages, <0.90 versus 0.99 in the MPS (Cabeza et al., 2009). L monocytogenes presented higher resistance than both strains of Salmonellae against E-beam treatment although that of Salmonella Typhimurium was nearer. This observation is in general agreement with data reported by several authors (Thayer et al., 1990; Grant and Patterson, 1992; Jakabi et al., 2003; Sommers et al., 2003; Mendonca et al., 2004).

The process criteria (PrC, the effect in frequency, and/or concentration of a hazard in a food that must be achieved by the application of one or more control measures to provide or contribute to an FSO or adequate level of protection, as applicable) (Gorris, 2005) for MPS are showed in Table 1. They may be obtained by multiplying the specific PrO (2.39 D₁₀ for Salmonella Enteritidis and Salmonella Typhimurium and 4.49 D₁₀ for L. monocytogenes) by the appropriate D ₁₀-value. Thus, the corresponding PrC values (Table 1) are the following: for Salmonella Enteritidis, it is 0.35 (the most unfavorable) × 2.39 = 0.84 kGy, for Salmonella Typhimurium it is 0.42 × 2.39 = 1.00 kGy and, for L. monocytogenes, 0.56 × 4.49 = 2.51 kGy, respectively. As reported below (Tables 3 and 4), doses of 2 kGy yield a product in which the consumer may detect undesirable odor and flavor. Therefore, the calculated PrC to be applied for guaranteeing the absence of L. monocytogenes for a 20-day term under refrigeration may be an excessive treatment from a sensory point of view. Nevertheless, according to the EC microbiological criteria specifications (CEC, 2007), it is possible, in this case, to demonstrate that the product will not exceed the 100 CFU/g limit throughout its shelf life since it has received an appropriate listericide treatment and, consequently, the 100 CFU/g criterion, must be applied. In this case, the H ₀ and Δ₂₀ are the same values (1 and 2.1 in logarithmic terms, respectively), but the FSO will be 1.0 instead of −1.39. From equation H ₀ − PrO + Δ₂₀ = FSO, the PrO of 1.21 D is determined. As a result, the PrC will be 0.68 kGy (Table 1, in parentheses). Then, it is necessary to apply the PrC determined for Salmonella Typhimurium, that is, 1.0 kGy, which is very low since it is far from the threshold of 2 kGy. As the irradiation treatment provokes an important increase of the shelf life (see below), the PrC for a shelf life of 1 month has also been assayed. In this case, the Δ₃₀ is log 4.20 CFU/g, which results in a PrO of 3.20 and a PrC of 1.80 kGy, still lower than the 2 kGy threshold.

Table 3.

Sensory Characteristics of Untreated (0 kGy) and Irradiated (1, 2 kGy) Mayonnaise Potato Salad That Showed Significant Differences (p < 0.05) in the Triangular Test After Several Periods of Refrigerated Storage at 4°C^a

Doses (kGy)	1	2	Storage days
0	NS	NS	0
	O	O, F	7
	O	O, F	14
	NS	O	22
	O (SP)	O (SP)	30
	O (SP)	O (SP)	47
1		NS	0
		F	7
		NS	14
		NS	22
		F	30
		F	47

No significant differences (NS) were detected in odor, flavor, and appearance. Significant differences (p < 0.05) for odor (O) and/or flavor (F). Samples nonirradiated were spoiled and flavor test was not performed (SP).

Table 4.

Sensory Evaluation ^a of Appearance, Odor, and Flavor by Rank Order Test of Untreated and E-Beam-Irradiated Mayonnaise Potato Salad Just After Treatment (0 Days) and After Several Periods of Refrigerated Storage (4°C)

	Days of storage at 4°C
Dose (kGy)	0	7	14	22	30	47
0	39_A	41_A	42_A	41_A	42_A	21_B	Appearance
1	40_A	37_A	40_A	40_A	37_A	51_A
2	41_A	42_A	38_A	39_A	41_A	48_A
0	51_A	52_A	50_A	29_B	30_B	28_B	Odor
1	36_B	36_B	41_AB	51_A	40_AB	40_AB
2	33_B	32_B	29_B	40_AB	50_A	52_A
0	51_A	51_A	60_A	NP	NP	NP	Flavor
1	48_A	49_A	33_B	31_A	28_A	29_A
2	21_B	20_B	27_B	29_A	32_A	31_A

Different letters in the same column for a similar sensorial attribute indicate significant differences (p < 0.05).

Final scoring = (N ₁ × 1) + (N ₂ × 2) + (N ₃ × 3), where N ₁, N ₂, and N ₃ are the number of panelists that ranked the sample in position 1 (least preference), 2, or 3 (most preference) in the order ranked test.

NP, samples nonirradiated were spoiled and flavor test was not performed. In this case, only the flavor of the irradiated samples was evaluated [final scoring = (N ₁ × 1) + (N ₂ × 2), where N ₁ and N ₂ are the number of panelists that ranked the sample in position 1 (least preference) or 2 (most preference)].

Another dangerous pathogenic bacterium that may occasionally reach the salad while it is being prepared is E. coli O157:H7. Its radioresistance is very low. D-values of about 0.2 kGy have been reported by several authors (Clavero et al., 1994; Buchanan et al., 1998; Chirinos et al., 2002; Cabeza et al., 2009). This D-value would yield a PrC (0.2 × 2.39) of 0.48 kGy for this bacterium, which is even lower than that of Salmonella Enteritidis. Therefore, this organism is not of great concern to this respect.

Besides controlling pathogens, irradiation also has a bactericidal effect on spoilage bacteria, which leads to an extended shelf life of the treated product. Table 2 shows the changes in the number of aerobic mesophiles during storage. The counts show that the E-beam treatment at a dose of 1 kGy may not only be effective in guaranteeing the microbiological safety of the MPS, but it also causes a significant reduction of total spoilage microbiota. This was not appreciated on the first day since the microbiological counts were designed to detect more than 10³ CFU/g. However, the next bacterial analyses (days 22, 30, and 47) demonstrated the very strong reduction mentioned above. Even in day 47, the level of organisms in the sample irradiated with 1 kGy was lower than the detection limit, whereas the nonirradiated samples were close to being spoiled by day 22. It may be concluded that the E-beam treatment at doses as low as 1 kGy are enough to provoke a noticeable increase (over a month) of the MPS shelf life.

Sensory properties

Table 3 shows the results of the sensory triangle of untreated and samples treated with 1 and 2 kGy for selected (appearance, odor, and flavor) sensory features. No differences were detected in the appearance and flavor between untreated samples and those subjected to 1 kGy E-beam treatment. However, significant differences (P < 0.05) in the first 2 weeks after treatment were found for the odor, which were due, according to the panelist description, to effluvia expelled to “burnt beef broth” or “scalded feather” but they were not detected afterward (day 22), similar to that previously reported in other irradiated product (Jo and Ahn, 2000; Cabeza et al., 2007). Differences were detected again after 1 month of storage, but they were due to other phenomenon, namely, the growth of the natural microbiota in nontreated samples (Table 2), probably lactic acid bacteria according to the reports of other authors (Smittle and Flowers, 1982; Muriana and Kanach, 1995). As was to be expected, the higher the doses applied, the greater the sensory modifications became, as it is reflected when samples were treated with 2 kGy, in which differences affected also to the flavor (Table 3). Since the dose to obtain the process criterion (PrC) was set at 1.0 kGy (Table 1), it may say that only the odor is slightly affected.

The rank order tests were performed to rank samples untreated and irradiated (at 1 and 2 kGy) in order of preference, according to the proximity to the optimal sensory quality of the MPS (Table 4). The appearance of control samples were not different from those treated by E-beam treatment until the 47th day of storage. Then, it was observed a partial liquefaction in control samples and the bag slightly swollen, which has been attributed to the growth of spoilage microbiota in untreated MPS.

No differences in the odor were found between samples irradiated with 1 and 2 kGy. The pattern was the same, including along the whole period of storage. However, this feature was the most sensitive to the irradiation since differences were observed when irradiated samples were compared versus control (nonirradiated) ones. In fact, the latter reached a higher scoring still 2 weeks of storage. The irradiated MPS expelled odors that denoted slight irradiation notes (slight “sulfured,” “burnt beef broth,” “scalded feather,” or “hot culture medium”) but they were not designated as unpleasant nor were they rejected. Nevertheless, from 22 days of storage the ranking tests showed that the irradiated samples had a more similar odor to that which is desired for this type of food, reaching a higher punctuation than control samples. This paradoxical behavior can be explained by the microbial spoilage of untreated samples by the natural microbiota (Table 2). In the descriptive analysis carried out after this storage time, the members of the panel stated that the untreated MPS expelled acid odors, whereas in the treated samples the slight off-odor notes mentioned above disappeared.

The results of microbiological analysis indicated that the untreated MPSs were not suitable for consumption after 20 days of refrigerated storage. For this reason, in the flavor analysis there were two stages of research: an initial one (up to day 14), in which control samples (0 kGy) were evaluated as well as those irradiated, and a second stage (test performed after day 14), in which only the irradiated MPSs were analyzed (notice that in this kind of sensory evaluation the final scoring depend of the number of samples tested). No differences between untreated MPSs and those with a dose of 1 kGy were detected at 0 and after 7 days of storage, although a slight difference was noticed after 2 weeks. When 2 kGy were applied, differences were already detected just after treatment (day 0). In the descriptive analysis, the members of the panel stated that the MPS with 2 kGy presented off-flavors with slight burnt, sulfuric, and light astringent notes, whereas the flavor modifications produced by 1 kGy were practically negligible. No doubt they were due to the generation of off-substances (probably, most of them volatiles) by the irradiation, affecting to the odor and flavor when 2 kGy was applied. The flavor peculiarities of irradiated samples disappeared after 22 days of storage, which leads to postulate that substances generated by the E-beam treatment (at least at low doses) are transitory and they dissipate as the time goes on.

Conclusion

From the results, it may be concluded that the FSO for Salmonella enterica serovar Enteritidis, Salmonella enterica serovar Typhimurium, and L. monocytogenes in MPS may be achieved with an E-beam treatment of ∼1.0 kGy. The E-beam treatment of MPS at a dose of 1 kGy approximately doubles its life. If the storage is extended to 30 days, a 1.8 kGy treatment must be applied. The MPS irradiated with doses of 1 kGy, or slightly higher, only causes negligible sensory changes.

Footnotes

Acknowledgments

The present work was supported by the Ministerio de Educación y Ciencia as projects AGL2007-65235-CO2-02 and CARNISENUSA (CSD2007-00016), included in the CONSOLIDER-INGENIO 2010 issue. The authors belong to the research group UCM-BSCH 920276 of the Universidad Complutense. The authors are grateful to IONISOS Iberica for the technical assistance provided during the E-beam irradiation.

Disclosure Statement

No competing financial interests exist.

References

Ahn

, Jo

, Du

et al. Quality characteristics of pork patties irradiated and stored in different packaging and storage conditions. Meat Sci, 2000; 56:203–209.

Anonymous. Cross contamination/different strains in Oregon E. coli case. Food Chem News, 1993; 35:32–33.

[ASTM] American Society for Testing and Materials. E1650-97e1 standard practice for use of cellulose acetate dosimetry systems. Standards on disc, 12.02. American Society for Testing and Materials: West Conshohocken, PA, 2000.

Blaser

, Newman

. A review of human salmonellosis: I Infective dose. Rev Infect Dis, 1982; 4:1096–1106.

Buchanan

, Edelson

, Snipes

et al. Inactivation of Escherichia coli O157:H7 in apple juice by irradiation. Appl Environ Microbiol, 1998; 64:4533–4535.

Cabeza

, Ordóñez

, Cambero

et al. Effect of thermoultrasonication on Salmonella enterica serovar Enteritidis in distilled water and intact shell eggs. J Food Prot, 2004; 67:1886–1891.

Cabeza

, Cambero

, de la Hoz

et al. Optimization of E-beam treatment to eliminate Listeria monocytogenes from ready-to-eat (RTE) cooked ham. Innov Food Sci Emerg Technol, 2007; 8:299–305.

Cabeza

, de la Hoz

, Velasco

et al. Safety and quality of ready-to-eat dry fermented sausages subjected to E-beam radiation. Meat Sci, 2009; 83:320–327.

Clavero

MRS

, Monk

, Beuchat

et al. Inactivation of Escherichia coli O157:H7, salmonellae and Campylobacter jejuni in raw ground beef gamma irradiation. Appl Environ Microbiol, 1994; 60:2069–2075.

10.

[CEC] Commission of the European Communities. Commission Regulation (EC) No. 1441/2007 of 5 December 2007 amending Regulation (EC) No. 2073/2005 on microbiological criteria for foodstuffs. Official Journal of the European UnionL 322/122007.

11.

Christensen

, Ogden

, Dunn

et al. Multiple comparison procedures for analysis of ranked data. J Food Sci, 2006; 71:132–143.

12.

Chirinos

RRO

, Vizeu

, Destro

et al. Inactivation of Escherichia coli O157:H7 in hamburgers by gamma irradiation. Braz J Microbiol, 2002; 33:53–56.

13.

D'Aoust

. Salmonella. Guide to Foodborne Pathogens. Labbé

, García

. New York: Wiley-Interscience, 2005; 163–191.

14.

D'Aoust

, Aris

, Thisdele

et al. Salmonella eastborne outbreak associated with chocolate. Can Inst Food Sci Technol J, 1975; 8:181–184.

15.

[EFSA] European Food Safety Authority. The Community Summary Report. Trends and sources of zoonoses and zoonotic agents and food-borne outbreaks in the European Union in 2008. EFSA J, 2010; 8:1496.

16.

Erickson

, Jenkins

. Comparative Salmonella spp. and Listeria monocytogenes inactivation rates in four commercial mayonnaise products. J Food Prot, 1991; 54:913–916.

17.

Erickson

, Mckenna

, Woodruff

, Bloom

. Fate of Salmonella spp., Listeria monocytogenes, and spoilage microorganisms in home-style salads prepared with commercial real mayonnaise or reduced calorie mayonnaise dressings. J Food Prot, 1993; 56:1015–1021.

18.

Farber

, Peterkin

. Listeria monocytogenes a food-borne pathogen. Microbiol Rev, 1999; 55:476–511.

19.

[FDA] Food, Drug Administration, and U.S. Food Safety and Inspection Service. Quantitative assessment of the relative risk to public health from foodborne Listeria monocytogenes among selected categories of ready-to-eat foods. Available in Docket No. 1999N-1168, 23–28. 2003.

20.

[FDA] Food and Drug Administration. 2005. Recalls, market withdrawals and safety alerts archive. www.fda.gov/oc/po/firmrecalls/archive.html. 2010 November . (Online.)

21.

[FDA/USDA/CDC] Food, Drug Administration/United States Department of Agriculture/Centers for Disease Control and Prevention. 2001. Draft assessment of the relative risk to public health from foodborne Listeria monocytogenes among selected categories of ready-to-eat foods. www.fsis.usda.gov/OPPDE/rdad/FRPubs/99N-1168FDA-052201.htm. 2010 November . (Online.)

22.

[FSIS-USDA] Food Safety and Inspection Service, United States Department of Agriculture. Recalls. 2005. www.fsis.usda.gov/Fsis_Recalls/index.asp. 2010 November . (Online.)

23.

Glass

, Doyle

. Fate of Listeria monocytogenes in processed meat during refrigerated storage. Appl Environ Microbiol, 1989; 55:1565–1569.

24.

Glass

, Doyle

. Fate of Salmonella and Listeria monocytogenes in commercial, reduced-calorie mayonnaise. J Food Prot, 1991; 54:691–695.

25.

Glass

, Loeffeelholz

, Harried

et al. Survival of Escherichia coli O157:H7 in mayonnaise, mayonnaise dressing. Food Research. Institute Annual Meeting, Madison, WI, May 13, 1993.

26.

Gomez-Lucía

, Goyache

, Blanco

et al. Growth of Staphylococcus aureus and enterotoxin production in homemade mayonnaise prepared at different pH values. J Food Prot, 1987; 50:872–875.

27.

Gorris

LGM

. Food safety objective: an integral part of food chain management. Food Control, 2005; 16:801–809.

28.

Gorris

LGM

. Impact of risk analysis on food safety. Inaugural address European Chair in Food Safety Microbiology. Wageningen University and Research: Wageningen, The Netherlands, 2002. http://edepot.wur.nl/33030. 2010 November . (Online.)

29.

Grant

, Patterson

. Sensitivity of foodborne pathogens to irradiation in the components of a chilled ready meal. Food Microbiol, 1992; 9:95–103.

30.

Hwang

. The effect of mayonnaise pH and storage temperature on the behavior of Listeria monocytogenes in ham salad and potato salad. J Food Prot, 2005; 68:1628–1634.

31.

Hwang

, Marmer

. Growth of Listeria monocytogenes in egg salad and pasta salad formulated with mayonnaise of various pH and stored at refrigerated and abuse temperatures. Food Microbiol, 2007; 24:211–218.

32.

Hwang

, Tamplin

. The influence of mayonnaise pH and storage temperature on the growth of Listeria monocytogenes in seafood salad. Int J Food Microbiol, 2005; 102:277–285.

33.

[ICMSF] International Commission on Microbiological Specifications for Foods. Listeria monocytogenes. Microorganisms in Foods. 5 Microbiological Specifications of Food Pathogens. London: Blackie Academic & Professional, 1996a. 141–182.

34.

[ICMSF] International Commission on Microbiological Specifications for Foods. Salmonellae. In: Microorganisms in Foods. 5 Microbiological Specifications of Food Pathogens. London: Blackie Academic & Professional, 1996b. 217–264.

35.

[ICMSF] International Commission on Microbiological Specification for Foods. Listeria monocytogenes in cooked sausage (frankfurters). In: Microorganisms in Foods. 7 Microbiological Testing in Food Safety Management. New York: Kluivert Academic/Plenum Publishers, 2002; 285–312.

36.

[ICMSF] International Commission on Microbiological Specifications for Foods. Oil and fat-based foods (II. Mayonnaise and dressings), In: Microorganisms in Foods. 6 2 ^nd Microbial Ecology of Food Commodities. New York: Kluwer Academic/Plenum Publishers, 2005; 481–493.

37.

[ISO] International Standards Organization. Analyze sensorielle. Guide pour l'implantation d'un local destiné aux analysses sensoriellesISO/DP 66.58.International Organization for Standardization: Geneva, 1981a. (In French.)

38.

[ISO] International Standards Organization. Methodologie essai triangulaireISO TC 34/SC 12.International Organization for Standardization: Geneva, 1981b. (In French.)

39.

Jakabi

, Gelli

, Torre

JCMD

et al.

Inactivation by ionising radiation of Salmonella Enteritidis, Salmonella infantis and Vibrio parahaemolyticus in oysters (Crassostrea brasiliana)

J Food Prot, 2003; 66:1025–1029.

40.

Jay

, Loessner

, Golden

. Modern Food Microbiology, 7 ^th. New York: Springer, 2005; 619–636.

41.

, Ahn

. Production of volatile compounds from irradiated oil emulsions containing amino acids or proteins. J Food Sci, 2000; 65:612–616.

42.

Joanes

. On a rank sum test due to Kramer. J Food Sci, 1985; 50:1442–1444.

43.

John

, Jenkins

. Comparative Salmonella Spp. and Listeria monocytogenes inactivation rates in four commercial mayonnaise products. J Food Prot, 1991; 54:913–916.

44.

Mendonca

, Romero

, Lihono

et al. Radiation resistance and virulence of Listeria monocytogenes Scott A following starvation in physiological saline. J Food Prot, 2004; 67:470–474.

45.

Michels

MJM

, Koning

. Mayonnaise, dressings, mustard, mayonnaise-based salads, and acid sauces. 7 Microbiological Safety and Quality of Food, 1 Lind

, Baird-Parker

, Gould

. Gaithersburg: Aspen Publish, Inc., 2000; 807–835.

46.

Muriana

, Kanach

. Use of NisaplinTM to inhibit spoilage bacteria in buttermilk ranch dressing. J Food Prot, 1995; 58:1109–1113.

47.

Ordóñez

, Cabeza

, de la Hoz

et al. Uso de radiaciones beta para conseguir el objetivo de seguridad alimentaria (FSO) en carne de pollo con respecto a Salmonella spp. Rev Com Científ AESA, 2007; 6:67–87. (In Spanish.)

48.

Serrano

, Murano

, Shenoy

et al. D-Values of Salmonella Enteritidis isolates and quality attributes of shell eggs and liquid whole eggs treated with irradiation. Poultry Sci, 1997; 76:202–205.

49.

Smittle

, Flowers

. Acid tolerant microorganisms involved in the spoilage of salad dressing. J Food Prot, 1982; 45:977–983.

50.

Sommers

, Fan

, Niemira

et al. Radiation (gamma) resistance and postirradiation growth of Listeria monocytogenes suspended in beef bologna containing sodium diacetate and potassium lactate. J Food Prot, 2003; 66:2051–2056.

51.

Tarkowski

, Stoffer

SCC

, Beumer

et al. Low dose gamma irradiation of raw meat. I. Bacteriological and sensory effects in artificially contaminated samples. Int J Food Microbiol, 1984; 1:13–23.

52.

Thayer

, Boyd

, Muller

et al. Radiation resistance of Salmonella. J Ind Microbiol, 1990; 5:383–390.

53.

Thomas

, Davenport

. Zygosaccharomyces baily. A profile of characteristics and spoilage activities. Food Microbiol, 1985; 2:157–169.

54.

Uyttendaele

, De Troy

, Debevere

. Incidence of Listeria monocytogenes in different types of meat products on the Belgian retail market. Int J Food Microbiol, 1999; 53:75–80.

55.

Vought

, Tatini

. Salmonella enteritidis contamination of ice cream associated with a 1994 multistate outbreak. J Food Prot, 1998; 61:5–10.

56.

Weagant

, Bryant

, Bark

. Survival of Escherichia coli O157:H7 in mayonnaise and mayonnaise-based sauces at room and refrigerated temperatures. J Food Prot, 1994; 57:629–631.

57.

Zhu

, Du

, Cordray

, Ahn

. Control of 1 Listeria monocytogenes contamination in ready-to-eat meat products. Compr Rev Food Sci F, 2005; 4:34–42.