Microbial Quality of Industrial Liquid Egg White: Assumptions on Spoiling Issues in Egg-Based Chilled Desserts Clarisse Techer, Amina Daoud, Marie-No¨elle Madec, Michel Gautier, Sophie Jan, and Florence Baron
As a 1st step, this study aimed at investigating the microbial quality of liquid egg white in a French egg processing company. Thirty raw and 33 pasteurized liquid egg white samples were analyzed. Pasteurization was globally found efficient on mesophilic contaminants (1.7 ± 1.6 and 0.8 ± 0.9 log CFU/mL in raw and pasteurized samples, respectively), including for the control of Salmonella. However, Gram-positive enterococci were still detected in the pasteurized samples. As a 2nd step, a representative bacterial collection was built for exploring the spoilage issue in egg-based chilled desserts. Custard cream was chosen as growth medium since this food is widely used for the production of French chilled desserts. All of the 166 isolates of the bacterial collection were shown to be able to grow and to induce spoilage of the custard cream at refrigeration temperature (10 °C). Several spoilage types were highlighted in the custard cream, on the basis of changes regarding pH, consistency, production of holes or gas. As a 3rd step, bacterial enzymatic activities were explored on custard cream-based agar media. The bacterial collection was reduced to 43 isolates, based on further selection regarding the genera and the spoilage types previously highlighted. Albeit to different degrees, all these isolates were able to produce proteases. A large part of these isolates also expressed lipolytic and amylolytic activities. This study emphasizes the need to control egg white contamination and especially with Gram-positive heat-resistant Enterococi, in order to guarantee the shelf life of egg-based chilled desserts. Keywords: bacterial contamination, chilled desserts, enzymatic activities, liquid egg white, spoilage
This study provides new insights on the bacterial contamination of industrial liquid egg white and represents the starting point for the control of spoilage issues in the sector of chilled egg-based desserts potentially hosting egg white bacterial contaminants.
Practical Application:
Introduction Over the past few decades, the production of egg products has been constantly increasing. In 2012, France was among the 1st European producer, with approximately 290000 tons in liquid equivalent (Itavi 2013). Egg products, defined as eggs removed from their shells, are sold in the form of whole egg, egg yolk, or egg white products, either in liquid, dried, or frozen forms. They are used for the production of a wide range of foods such as pastry, sea-food products, meat products, or dairy products, including raw or poorly cooked finished products that are particularly sensitive from a microbiological point of view. While the content of shell eggs is generally sterile under safe conditions of breeding, the average level of contamination of the eggshell surface is comprised between 3.8 and 6.3 log CFU/egg for the aerobic mesophilic flora, depending on the studies (Baron and Jan 2010 for review). Eggshells are therefore a likely source of crosscontamination at the step of industrial egg breaking (Baron and Jan 2010 for review). In order to ensure the microbial stabilization of the egg products, heat treatments are almost systematically carried out. In the French egg processing industry, pasteurization is carried out for 2 to 6 min at 65 °C to 68 °C for whole egg and egg MS 20141105 Submitted 6/27/2014, Accepted 12/4/2014. Authors are with Equipe Microbiologie de l’Oeuf et des Ovoproduits (Micov), UMR1253 Science et Technologie du Lait et de l’Œuf, Agrocampus Ouest, INRA, F-35042, Rennes, France. Direct inquiries to author Baron (E-mail: [email protected]).
R C 2015 Institute of Food Technologists
doi: 10.1111/1750-3841.12764 Further reproduction without permission is prohibited
yolk and at lower temperatures (55 to 57 °C) for egg white, due to the higher heat-sensitivity of the constitutive proteins of this later. After this pasteurization step, the liquid egg products are rapidly cooled and they are conditioned, stored, and delivered at refrigeration temperature. Their shelf life ranges from few weeks to 3 mo at 4 °C. There are numerous studies relating to egg contamination with Salmonella Enteritidis, due to the strong involvement of this microorganism in food-poisoning outbreaks in relation to egg consumption (EFSA 2014). By contrast, the contamination with heat-resistant bacteria has been poorly investigated in this sector, whereas they can shorten the shelf life of the egg products themselves or of food products containing egg products as ingredients, with damaging economic consequences for the egg processing industry. On the few studies reporting the contamination of pasteurized liquid egg products by heat resistant bacteria, Enterococcus sp. (Hidalgo and others 2008; Miller and others 2010) and bacteria of the Bacillus cereus group (Baron and others 2007; Techer and others 2014) were already highlighted in pasteurized liquid whole egg products. However, little is known on the microbial quality of industrial egg white. This issue is of great importance when foams of egg white are used for the production of highly perishable products, including chilled egg-based desserts such as the French floating island. This chilled dessert comprises a sugared egg white foam deposited onto custard cream. Contrary to egg white, which is recognized as not readily supporting bacterial growth (Baron and Jan 2010 for review), the custard cream is rich Vol. 80, Nr. 2, 2015 r Journal of Food Science M389
M: Food Microbiology & Safety
Abstract:
Bacteria from liquid egg white . . . in various substrates such as proteins, lipids, and carbohydrates. Since it is sterilized at 132 °C for a few minutes, this ingredient is unlikely to host bacteria, provided that postcontaminations are avoided. The spoilage events recorded along storage of the dessert may then rather involve a bacterial population provided by the pasteurized egg white foam. While weakly allowed to grow in this medium, this bacterial population may find in the custard cream all the nutrients for growth. In order to test this assumption, the present study aimed at characterizing the microbial quality of industrial liquid egg white and at testing, at low temperature, the spoilage potential of egg white bacterial contaminants inoculated in custard cream. This later was used as such or in an agar test allowing the quantification of specific bacterial enzymatic activities, which are recognized as being fairly induced in food spoilage (Pirttijarvi and others 1996; Cosentino and others 1997).
Materials and Methods
M: Food Microbiology & Safety
Microbiological analyses of the egg white samples Sixty-three samples of liquid egg white were provided by a single egg processing industry from Western France for a 1-y period. The egg white products came from clean and uncraked eggs. They contained no ingredients, no stabilizers, no whipping or processing aids, and no preservatives. For the batches 1 to 19, the samples were analyzed both before (raw egg white products) and after (pasteurized egg white products) pasteurization at 57 °C for 6 min. The batches 20 to 30 were exclusively analyzed before pasteurization, while the batches 31 to 44 were exclusively analyzed after pasteurization. The concentration of total mesophilic bacteria was assessed after serial decimal dilutions in Tryptone salt (TS; 0.1% Tryptone, AES laboratoire, Combourg, France; 0.85% NaCl, Labogros, Buchs, France), plating onto Brain Heart Infusion-Yeast Extract agar (BHI-YE) (3.7% BHI (Merck, Darmstadt, Germany); 0.1% Yeast extract (Merck); 1.5% agar (Merck) and incubation under aerobic and anaerobic (Anaerocult A, Merck) conditions for 48 h at 30 °C. Each sample was analyzed in triplicate under both aeration conditions. After incubation, the colony forming units (CFU) were counted and the results were expressed as the mean of 3 replicates in log CFU/mL ± standard deviation (SD) for each sample. Identification of the bacteria and construction of the bacterial collection For each egg white sample, 5 to 10 colonies were randomly collected in the BHI-YE agar counting plates. In order to avoid redundancy, 3 to 5 purified colonies coming from the same egg white sample were selected, based on macroscopic and microscopic criteria. This selection resulted in the creation of a bacterial collection comprising 166 isolates. All these isolates were (i) identified by 16S rDNA sequencing and (ii) tested in liquid custard cream for assessing their growth potential and the type of spoilage they induced. A 43-isolate subset was further selected inside the collection in order to guarantee the best representativeness of each genus and of each spoilage type. These 43 isolates were then submitted to enzymatic assays on custard cream-based agar media. 16S rDNA sequencing Three hundred microliters of a 25% (m/v) sterile suspension of Chelex beads (Grosseron, Saint-Herblain, France) prepared in sterile Milli-Q water (Sigma Aldrich, Saint-Quentin Fallavier, France) was added to 5 mL of each bacterial frozen culture previously thawed and incubated overnight in BHI-YE at 30 °C. M390 Journal of Food Science r Vol. 80, Nr. 2, 2015
The mixture was vortexed and centrifuged at 5700 × g for 7 min at 4 °C. The cell pellet was resuspended in 200 µL sterile Milli-Q water (Sigma Aldrich) and lyzed by heating at 100 °C for 10 min. After centrifugation at 5700 × g for 7 min at 4 °C, 150 µL supernatant was collected and recentrifuged under the same conditions. The supernatant was then used for 16S rDNA gene amplification. The final PCR mixture contained 50 ng DNA template, 0.18 mM dNTPs (Eurogentec, Seraing, Belgium), 0.7 µM of each 16S primers (5’- GCCAGCAGCCGCGGTAA -3’ as forward degenerate primer and 5’- GACGGGCGGTGTGTAC -3’ as reverse primer), 53.6 U/mL AmpliTaq polymerase and 0.9X AmpliTaq buffer (Biolabs, Evry, France). Thermal cycling was carried out in a PCR thermocycler (iCycler optical module 584BR, BioRad, Marnes-la-Coquette, France) with the following run: a starting cycle of 4 min at 95 °C, followed by 30 cycles of 30 s at 95 °C, 30 s at 50 °C and 30 s at 72 °C, and a final extension of 7 min at 72 °C. The PCR products were verified after migration on 1% agarose gel (80V, 30 min) in TBE 1X (Eurobio, Courtaboeuf, France) together with the DNA Smart Ladder SF molecular weight marker (Eurogentec). Gels were stained with ethidium bromide (Sigma Aldrich) and digitized using a gel imager (Bioblock, Illkirch, France). The 16S rDNA amplified fragment was identified according to its apparent length (about 800 bp). Purification and sequencing were performed by GATC Biotech AG (Konstanz, Germany) using the following primer: 5’- GCCAGCAGCCGCGGTAA3’. DNA sequences were compared with available sequences deposited in GenBank (http://www.ncbi.nlm.nih.gov/Genbank) using nucleotide-nucleotide Basic Local Alignment Search Tool (BLAST) for nucleotides (blastn) alignments accessed from the website (http://www.ncbi.nlm.nih.gov/BLAST/).
Spoilage experiments Inoculum. For each selected isolate, a purified colony was transferred into 10 mL BHI-YE and the mixture was incubated at 30 °C for 18 h under aerobic conditions. Aliquots of each culture were frozen at –20 °C after glycerol addition at a final concentration of 25%. Before each experiment, frozen aliquots were thawed and twice propagated in BHI-YE at 30 °C for 18 h under aerobic conditions. Culture media. For liquid-state experiments, the culture medium consisted in commercial UHT custard creams provided by a local supermarket. According to the manufacturer, the creams contain 80% milk, 12% sucrose, 7% whole egg, around 1% starch and unknown quantities of vanilla aroma and stabilizer. For solid-state experiments, the custard cream-based agar medium was prepared with 10% sterile commercial custard cream added to a previously sterilized (121 °C for 21 min) medium consisting in 1.5% agar in distilled water (pH 6.9). Twelve milliliters of this mixture were poured into each Petri dish. For the detection of the lipolytic activity, 2% of rhodamine B (Sigma Aldrich) was added before pouring. This later becomes fluorescent when complexed with the free fatty acids released by the bacterial lipases, according to the studies of Kouker and Jaeger (1987), Feng and others (2005), and Yan and others (2007). Growth in liquid custard cream. A collection of 166 isolates was tested for growth experiments in sterile liquid custard cream (see the section Identification of the bacteria and construction of the bacterial collection). Each inoculum was previously diluted in tryptone salt in order to reach an initial concentration of 3 to 5 log CFU/mL in sterile commercial custard cream (10 mL in Falcon tubes). The contaminated custard cream samples were incubated at 10 °C for 21 d under aerobic conditions. Before and after this
incubation period, enumeration was carried using a plate-counting micro-method (Baron and others 2006). Cells counts were determined after 24 h growth at 30 °C on BHI-YE agar. Experiments were repeated for visual confirmation of bacterial growth. Enumerations were carried out only once, and the results were expressed in log CFU/mL for each tested isolate. Spoilage characteristics of the liquid custard cream. Spoilage criteria were based on visual observations (thickening, coagulation, production of holes), on a hearing test (sound detection at the opening of the tube) and on pH measurements (pH meter 315i, WTW, Weilheim, Germany). Results were expressed as ࢞pH, corresponding to the difference between the pH of the custard cream at time zero of inoculation and after 21 d of incubation at 10 °C. Enzymatic activities in custard cream agar-based media. A collection of 43 isolates was selected (see the section Identification of the bacteria and construction of the bacterial collection). Two microliters of each inoculum was spotted onto the agar plates. Incubation was carried out for 48 h at 30 °C under aerobic conditions. After this incubation period and according to the protocol of Loperena and others (2012), proteolytic activities were evaluated by the measure of the clearing zone surrounding the colonies (diameter of the clearing zone – diameter of the colony)/2, expressed in mm. Results were expressed as the mean of 2 replicates. According to the protocol of Hankin and others (1975), amylolytic activities were evaluated after addition of an iodine solution on the agar plates (lugol, Sigma Aldrich) and measure of the clearing zone surrounding the colonies (diameter of the clearing zone – diameter of the colony)/2. Results were expressed as the mean of 2 replicates. According to Feng and others (2005), lipolytic activities were detected under Rhodamine B fluorescent illumination (Olympus BX51TF, Olympus, Hamburg, Germany). Wavelengths used were 553 and 627 nm for excitation and emission, respectively. Experiments were performed in duplicate. Controls. The sterility of the custard cream was systematically checked before each experiment, as follows: 1 mL of custard cream was poured into a Petri dish containing BHI-YE agar. The absence of growth was checked after overnight aerobic incubation at 30 °C. For the characterization of the type of spoilage of the custard cream, a noninoculated sample was systematically incubated under the same conditions as those for the assays. For the enzymatic tests, both negative and positive controls were carried out. Positive controls were implemented independently by deposit of 2 µL of a purified protease from Bacillus licheniformis (ref. P5459, Sigma Aldrich), a purified lipase from Pseudomonas fluorescens (ref. 95608, Sigma Aldrich), and an amylase from Bacillus sp. (ref. A7720, Sigma Aldrich) for proteolytic, lipolytic, and amylolytic activities, respectively. The absence of interference between the 3 enzymatic tests was also checked by deposit of each purified enzymes onto each performed medium.
Results and Discussion Microbial quality of liquid egg white Regardless of the type of sample analyzed, the bacterial counts were of the same order of magnitude under aerobic and anaerobic conditions of enumeration (Table 1). It can therefore be concluded that egg white contaminants were aero-anaerobic bacteria. This observation allowed interpreting the results regardless of the aeration conditions of enumeration. Considering the raw egg products, the bacterial concentrations were fairly low with, on average, less than 2 log CFU/mL. These
results are in accordance with the intrinsic properties of egg white. This fluid is known as unlikely supporting microbial growth due to the cumulative effect of alkaline pH, iron deficiency, and the expression of antimicrobial molecules, such as lysozyme, ovotransferrin, proteinase inhibitors (cystatin, ovomucoid, and ovoinhibitor), and vitamin-binding proteins (riboflavin-, avidin-, and thiaminbinding proteins) (Baron and Jan 2010 for review). However, 2 raw samples were highly contaminated, exhibiting more than 5 (Nr 29) and more than 8 log (Nr 30 in Table 1) CFU/mL. These high levels of contamination may be ascribed to the use of endogenously contaminated eggs or to the release of bacteria from a biofilm present in the processing equipment. Another explanation should be that a poor separation of egg white from egg yolk would have resulted in the driven of egg yolk into these egg white samples, resulting in the impairment of the egg white antimicrobial activities. The presence of as low as 0.5% egg yolk in egg white is recognized as sufficient to allow bacteria to grow in egg white at permissive temperatures (Baron 1998). Despite the globally low bacterial concentrations, the contamination concerned 93% of the analyzed raw samples (Table 1). Events of cross-contaminations with eggshells represent the most likely explanation for this result. Indeed, the contact of the egg content with eggshells is unavoidable at the step of egg breaking. Contrary to the egg content, which is generally sterile under safe conditions of breeding (Baron and Jan 2010 for review), the eggshells are systematically contaminated (Musgrove and others 2004, 2005; De Reu and others 2005, 2006). This systematic cross-contamination confirms the need to stabilize industrial egg products, including liquid egg white. The European regulation describes specific requirements for the manufacture of egg products. Considering the microbiological criteria, the regulation Nr 2073/2005 defines (i) a food safety criterion for Salmonella in “egg products, excluding products where the manufacturing process or the composition of the product will eliminate the Salmonella risk” for egg products placed on the market during their shelf life (absence in 25 g) and (ii) a process hygiene criterion for Enterobacteriaceae at the end of the process of egg product manufacturing (less than 2 log CFU/g). In the present study, pasteurization was found efficient since the percentage of positive samples was reduced between 64% and 50% after pasteurization under aerobic and anaerobic conditions, respectively (Table 1). The efficiency of pasteurization was globally confirmed regarding the average bacterial concentrations which were reduced to less than 1 log CFU/mL in the positive samples (Table 1). The range of concentrations was narrower than in the raw samples, that is, from less than 1 CFU/mL to around 2 log CFU/mL (Table 1). Comparison with literature is difficult since, to our knowledge, the microbial quality of industrial egg white has so far been poorly investigated in Europe. Correa and others (2008) have reported a bacterial concentration of 2 log CFU/mL in Spanish pasteurized liquid egg white products, corroborating the average low concentrations highlighted in the present study. The literature is more abundant regarding the contamination of pasteurized whole egg products. Whole egg products are more likely to be contaminated due to the fact that, unlike egg white, they represent optimal media for bacterial growth (Baron and Jan 2010 for review). However, the pasteurization of whole egg products is possible under stronger conditions than those used for egg white, due to the higher heat-resistance of egg yolk proteins. The resulting level of contamination reported in the literature is nearly similar as that found in pasteurized liquid egg Vol. 80, Nr. 2, 2015 r Journal of Food Science M391
M: Food Microbiology & Safety
Bacteria from liquid egg white . . .
Bacteria from liquid egg white . . . Table 1–Plate-counting of bacteria in raw and pasteurized (57 °C for 6 min) liquid egg white samples in Brain Heart Infusion-Yeast extract (BHI-YE) agar under aerobic and anaerobic conditions at 30 °C for 48 h. Results are expressed as the mean of 3 technical replications in log CFU/mL ± standard deviation. Before pasteurization Batch number
M: Food Microbiology & Safety
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Mean Number (percentages) of contaminated samples
After pasteurization
Aerobic flora
Anaerobic flora
Aerobic flora
Anaerobic flora
± 0.03 ± 0.1 ± 0.3 ± 0.1 ± 0.1 ± 0.2 ± 0.1 ± 0.0 ± 0.2 ± 0.1 ± 0.1 ± 0.1 ± 0.3 ± 0.1 ± 0.1 ± 0.1 ± 0.0 ± 0.1 ± 0.2 ± 2.8 ± 0.1 ± 0.3
As a 1st step, this study aimed at investigating the microbial quality of liquid egg white in a French egg processing company. Thirty raw and 33 pasteurized liquid egg white samples were analyzed. Pasteurization was globally found efficient on mesophilic contaminants (1.7 ± 1.6 and 0.8 ± 0.9 log CFU/mL in raw and pasteurized samples, respectively), including for the control of Salmonella. However, Gram-positive enterococci were still detected in the pasteurized samples. As a 2nd step, a representative bacterial collection was built for exploring the spoilage issue in egg-based chilled desserts. Custard cream was chosen as growth medium since this food is widely used for the production of French chilled desserts. All of the 166 isolates of the bacterial collection were shown to be able to grow and to induce spoilage of the custard cream at refrigeration temperature (10 °C). Several spoilage types were highlighted in the custard cream, on the basis of changes regarding pH, consistency, production of holes or gas. As a 3rd step, bacterial enzymatic activities were explored on custard cream-based agar media. The bacterial collection was reduced to 43 isolates, based on further selection regarding the genera and the spoilage types previously highlighted. Albeit to different degrees, all these isolates were able to produce proteases. A large part of these isolates also expressed lipolytic and amylolytic activities. This study emphasizes the need to control egg white contamination and especially with Gram-positive heat-resistant Enterococi, in order to guarantee the shelf life of egg-based chilled desserts. Keywords: bacterial contamination, chilled desserts, enzymatic activities, liquid egg white, spoilage
This study provides new insights on the bacterial contamination of industrial liquid egg white and represents the starting point for the control of spoilage issues in the sector of chilled egg-based desserts potentially hosting egg white bacterial contaminants.
Practical Application:
Introduction Over the past few decades, the production of egg products has been constantly increasing. In 2012, France was among the 1st European producer, with approximately 290000 tons in liquid equivalent (Itavi 2013). Egg products, defined as eggs removed from their shells, are sold in the form of whole egg, egg yolk, or egg white products, either in liquid, dried, or frozen forms. They are used for the production of a wide range of foods such as pastry, sea-food products, meat products, or dairy products, including raw or poorly cooked finished products that are particularly sensitive from a microbiological point of view. While the content of shell eggs is generally sterile under safe conditions of breeding, the average level of contamination of the eggshell surface is comprised between 3.8 and 6.3 log CFU/egg for the aerobic mesophilic flora, depending on the studies (Baron and Jan 2010 for review). Eggshells are therefore a likely source of crosscontamination at the step of industrial egg breaking (Baron and Jan 2010 for review). In order to ensure the microbial stabilization of the egg products, heat treatments are almost systematically carried out. In the French egg processing industry, pasteurization is carried out for 2 to 6 min at 65 °C to 68 °C for whole egg and egg MS 20141105 Submitted 6/27/2014, Accepted 12/4/2014. Authors are with Equipe Microbiologie de l’Oeuf et des Ovoproduits (Micov), UMR1253 Science et Technologie du Lait et de l’Œuf, Agrocampus Ouest, INRA, F-35042, Rennes, France. Direct inquiries to author Baron (E-mail: [email protected]).
R C 2015 Institute of Food Technologists
doi: 10.1111/1750-3841.12764 Further reproduction without permission is prohibited
yolk and at lower temperatures (55 to 57 °C) for egg white, due to the higher heat-sensitivity of the constitutive proteins of this later. After this pasteurization step, the liquid egg products are rapidly cooled and they are conditioned, stored, and delivered at refrigeration temperature. Their shelf life ranges from few weeks to 3 mo at 4 °C. There are numerous studies relating to egg contamination with Salmonella Enteritidis, due to the strong involvement of this microorganism in food-poisoning outbreaks in relation to egg consumption (EFSA 2014). By contrast, the contamination with heat-resistant bacteria has been poorly investigated in this sector, whereas they can shorten the shelf life of the egg products themselves or of food products containing egg products as ingredients, with damaging economic consequences for the egg processing industry. On the few studies reporting the contamination of pasteurized liquid egg products by heat resistant bacteria, Enterococcus sp. (Hidalgo and others 2008; Miller and others 2010) and bacteria of the Bacillus cereus group (Baron and others 2007; Techer and others 2014) were already highlighted in pasteurized liquid whole egg products. However, little is known on the microbial quality of industrial egg white. This issue is of great importance when foams of egg white are used for the production of highly perishable products, including chilled egg-based desserts such as the French floating island. This chilled dessert comprises a sugared egg white foam deposited onto custard cream. Contrary to egg white, which is recognized as not readily supporting bacterial growth (Baron and Jan 2010 for review), the custard cream is rich Vol. 80, Nr. 2, 2015 r Journal of Food Science M389
M: Food Microbiology & Safety
Abstract:
Bacteria from liquid egg white . . . in various substrates such as proteins, lipids, and carbohydrates. Since it is sterilized at 132 °C for a few minutes, this ingredient is unlikely to host bacteria, provided that postcontaminations are avoided. The spoilage events recorded along storage of the dessert may then rather involve a bacterial population provided by the pasteurized egg white foam. While weakly allowed to grow in this medium, this bacterial population may find in the custard cream all the nutrients for growth. In order to test this assumption, the present study aimed at characterizing the microbial quality of industrial liquid egg white and at testing, at low temperature, the spoilage potential of egg white bacterial contaminants inoculated in custard cream. This later was used as such or in an agar test allowing the quantification of specific bacterial enzymatic activities, which are recognized as being fairly induced in food spoilage (Pirttijarvi and others 1996; Cosentino and others 1997).
Materials and Methods
M: Food Microbiology & Safety
Microbiological analyses of the egg white samples Sixty-three samples of liquid egg white were provided by a single egg processing industry from Western France for a 1-y period. The egg white products came from clean and uncraked eggs. They contained no ingredients, no stabilizers, no whipping or processing aids, and no preservatives. For the batches 1 to 19, the samples were analyzed both before (raw egg white products) and after (pasteurized egg white products) pasteurization at 57 °C for 6 min. The batches 20 to 30 were exclusively analyzed before pasteurization, while the batches 31 to 44 were exclusively analyzed after pasteurization. The concentration of total mesophilic bacteria was assessed after serial decimal dilutions in Tryptone salt (TS; 0.1% Tryptone, AES laboratoire, Combourg, France; 0.85% NaCl, Labogros, Buchs, France), plating onto Brain Heart Infusion-Yeast Extract agar (BHI-YE) (3.7% BHI (Merck, Darmstadt, Germany); 0.1% Yeast extract (Merck); 1.5% agar (Merck) and incubation under aerobic and anaerobic (Anaerocult A, Merck) conditions for 48 h at 30 °C. Each sample was analyzed in triplicate under both aeration conditions. After incubation, the colony forming units (CFU) were counted and the results were expressed as the mean of 3 replicates in log CFU/mL ± standard deviation (SD) for each sample. Identification of the bacteria and construction of the bacterial collection For each egg white sample, 5 to 10 colonies were randomly collected in the BHI-YE agar counting plates. In order to avoid redundancy, 3 to 5 purified colonies coming from the same egg white sample were selected, based on macroscopic and microscopic criteria. This selection resulted in the creation of a bacterial collection comprising 166 isolates. All these isolates were (i) identified by 16S rDNA sequencing and (ii) tested in liquid custard cream for assessing their growth potential and the type of spoilage they induced. A 43-isolate subset was further selected inside the collection in order to guarantee the best representativeness of each genus and of each spoilage type. These 43 isolates were then submitted to enzymatic assays on custard cream-based agar media. 16S rDNA sequencing Three hundred microliters of a 25% (m/v) sterile suspension of Chelex beads (Grosseron, Saint-Herblain, France) prepared in sterile Milli-Q water (Sigma Aldrich, Saint-Quentin Fallavier, France) was added to 5 mL of each bacterial frozen culture previously thawed and incubated overnight in BHI-YE at 30 °C. M390 Journal of Food Science r Vol. 80, Nr. 2, 2015
The mixture was vortexed and centrifuged at 5700 × g for 7 min at 4 °C. The cell pellet was resuspended in 200 µL sterile Milli-Q water (Sigma Aldrich) and lyzed by heating at 100 °C for 10 min. After centrifugation at 5700 × g for 7 min at 4 °C, 150 µL supernatant was collected and recentrifuged under the same conditions. The supernatant was then used for 16S rDNA gene amplification. The final PCR mixture contained 50 ng DNA template, 0.18 mM dNTPs (Eurogentec, Seraing, Belgium), 0.7 µM of each 16S primers (5’- GCCAGCAGCCGCGGTAA -3’ as forward degenerate primer and 5’- GACGGGCGGTGTGTAC -3’ as reverse primer), 53.6 U/mL AmpliTaq polymerase and 0.9X AmpliTaq buffer (Biolabs, Evry, France). Thermal cycling was carried out in a PCR thermocycler (iCycler optical module 584BR, BioRad, Marnes-la-Coquette, France) with the following run: a starting cycle of 4 min at 95 °C, followed by 30 cycles of 30 s at 95 °C, 30 s at 50 °C and 30 s at 72 °C, and a final extension of 7 min at 72 °C. The PCR products were verified after migration on 1% agarose gel (80V, 30 min) in TBE 1X (Eurobio, Courtaboeuf, France) together with the DNA Smart Ladder SF molecular weight marker (Eurogentec). Gels were stained with ethidium bromide (Sigma Aldrich) and digitized using a gel imager (Bioblock, Illkirch, France). The 16S rDNA amplified fragment was identified according to its apparent length (about 800 bp). Purification and sequencing were performed by GATC Biotech AG (Konstanz, Germany) using the following primer: 5’- GCCAGCAGCCGCGGTAA3’. DNA sequences were compared with available sequences deposited in GenBank (http://www.ncbi.nlm.nih.gov/Genbank) using nucleotide-nucleotide Basic Local Alignment Search Tool (BLAST) for nucleotides (blastn) alignments accessed from the website (http://www.ncbi.nlm.nih.gov/BLAST/).
Spoilage experiments Inoculum. For each selected isolate, a purified colony was transferred into 10 mL BHI-YE and the mixture was incubated at 30 °C for 18 h under aerobic conditions. Aliquots of each culture were frozen at –20 °C after glycerol addition at a final concentration of 25%. Before each experiment, frozen aliquots were thawed and twice propagated in BHI-YE at 30 °C for 18 h under aerobic conditions. Culture media. For liquid-state experiments, the culture medium consisted in commercial UHT custard creams provided by a local supermarket. According to the manufacturer, the creams contain 80% milk, 12% sucrose, 7% whole egg, around 1% starch and unknown quantities of vanilla aroma and stabilizer. For solid-state experiments, the custard cream-based agar medium was prepared with 10% sterile commercial custard cream added to a previously sterilized (121 °C for 21 min) medium consisting in 1.5% agar in distilled water (pH 6.9). Twelve milliliters of this mixture were poured into each Petri dish. For the detection of the lipolytic activity, 2% of rhodamine B (Sigma Aldrich) was added before pouring. This later becomes fluorescent when complexed with the free fatty acids released by the bacterial lipases, according to the studies of Kouker and Jaeger (1987), Feng and others (2005), and Yan and others (2007). Growth in liquid custard cream. A collection of 166 isolates was tested for growth experiments in sterile liquid custard cream (see the section Identification of the bacteria and construction of the bacterial collection). Each inoculum was previously diluted in tryptone salt in order to reach an initial concentration of 3 to 5 log CFU/mL in sterile commercial custard cream (10 mL in Falcon tubes). The contaminated custard cream samples were incubated at 10 °C for 21 d under aerobic conditions. Before and after this
incubation period, enumeration was carried using a plate-counting micro-method (Baron and others 2006). Cells counts were determined after 24 h growth at 30 °C on BHI-YE agar. Experiments were repeated for visual confirmation of bacterial growth. Enumerations were carried out only once, and the results were expressed in log CFU/mL for each tested isolate. Spoilage characteristics of the liquid custard cream. Spoilage criteria were based on visual observations (thickening, coagulation, production of holes), on a hearing test (sound detection at the opening of the tube) and on pH measurements (pH meter 315i, WTW, Weilheim, Germany). Results were expressed as ࢞pH, corresponding to the difference between the pH of the custard cream at time zero of inoculation and after 21 d of incubation at 10 °C. Enzymatic activities in custard cream agar-based media. A collection of 43 isolates was selected (see the section Identification of the bacteria and construction of the bacterial collection). Two microliters of each inoculum was spotted onto the agar plates. Incubation was carried out for 48 h at 30 °C under aerobic conditions. After this incubation period and according to the protocol of Loperena and others (2012), proteolytic activities were evaluated by the measure of the clearing zone surrounding the colonies (diameter of the clearing zone – diameter of the colony)/2, expressed in mm. Results were expressed as the mean of 2 replicates. According to the protocol of Hankin and others (1975), amylolytic activities were evaluated after addition of an iodine solution on the agar plates (lugol, Sigma Aldrich) and measure of the clearing zone surrounding the colonies (diameter of the clearing zone – diameter of the colony)/2. Results were expressed as the mean of 2 replicates. According to Feng and others (2005), lipolytic activities were detected under Rhodamine B fluorescent illumination (Olympus BX51TF, Olympus, Hamburg, Germany). Wavelengths used were 553 and 627 nm for excitation and emission, respectively. Experiments were performed in duplicate. Controls. The sterility of the custard cream was systematically checked before each experiment, as follows: 1 mL of custard cream was poured into a Petri dish containing BHI-YE agar. The absence of growth was checked after overnight aerobic incubation at 30 °C. For the characterization of the type of spoilage of the custard cream, a noninoculated sample was systematically incubated under the same conditions as those for the assays. For the enzymatic tests, both negative and positive controls were carried out. Positive controls were implemented independently by deposit of 2 µL of a purified protease from Bacillus licheniformis (ref. P5459, Sigma Aldrich), a purified lipase from Pseudomonas fluorescens (ref. 95608, Sigma Aldrich), and an amylase from Bacillus sp. (ref. A7720, Sigma Aldrich) for proteolytic, lipolytic, and amylolytic activities, respectively. The absence of interference between the 3 enzymatic tests was also checked by deposit of each purified enzymes onto each performed medium.
Results and Discussion Microbial quality of liquid egg white Regardless of the type of sample analyzed, the bacterial counts were of the same order of magnitude under aerobic and anaerobic conditions of enumeration (Table 1). It can therefore be concluded that egg white contaminants were aero-anaerobic bacteria. This observation allowed interpreting the results regardless of the aeration conditions of enumeration. Considering the raw egg products, the bacterial concentrations were fairly low with, on average, less than 2 log CFU/mL. These
results are in accordance with the intrinsic properties of egg white. This fluid is known as unlikely supporting microbial growth due to the cumulative effect of alkaline pH, iron deficiency, and the expression of antimicrobial molecules, such as lysozyme, ovotransferrin, proteinase inhibitors (cystatin, ovomucoid, and ovoinhibitor), and vitamin-binding proteins (riboflavin-, avidin-, and thiaminbinding proteins) (Baron and Jan 2010 for review). However, 2 raw samples were highly contaminated, exhibiting more than 5 (Nr 29) and more than 8 log (Nr 30 in Table 1) CFU/mL. These high levels of contamination may be ascribed to the use of endogenously contaminated eggs or to the release of bacteria from a biofilm present in the processing equipment. Another explanation should be that a poor separation of egg white from egg yolk would have resulted in the driven of egg yolk into these egg white samples, resulting in the impairment of the egg white antimicrobial activities. The presence of as low as 0.5% egg yolk in egg white is recognized as sufficient to allow bacteria to grow in egg white at permissive temperatures (Baron 1998). Despite the globally low bacterial concentrations, the contamination concerned 93% of the analyzed raw samples (Table 1). Events of cross-contaminations with eggshells represent the most likely explanation for this result. Indeed, the contact of the egg content with eggshells is unavoidable at the step of egg breaking. Contrary to the egg content, which is generally sterile under safe conditions of breeding (Baron and Jan 2010 for review), the eggshells are systematically contaminated (Musgrove and others 2004, 2005; De Reu and others 2005, 2006). This systematic cross-contamination confirms the need to stabilize industrial egg products, including liquid egg white. The European regulation describes specific requirements for the manufacture of egg products. Considering the microbiological criteria, the regulation Nr 2073/2005 defines (i) a food safety criterion for Salmonella in “egg products, excluding products where the manufacturing process or the composition of the product will eliminate the Salmonella risk” for egg products placed on the market during their shelf life (absence in 25 g) and (ii) a process hygiene criterion for Enterobacteriaceae at the end of the process of egg product manufacturing (less than 2 log CFU/g). In the present study, pasteurization was found efficient since the percentage of positive samples was reduced between 64% and 50% after pasteurization under aerobic and anaerobic conditions, respectively (Table 1). The efficiency of pasteurization was globally confirmed regarding the average bacterial concentrations which were reduced to less than 1 log CFU/mL in the positive samples (Table 1). The range of concentrations was narrower than in the raw samples, that is, from less than 1 CFU/mL to around 2 log CFU/mL (Table 1). Comparison with literature is difficult since, to our knowledge, the microbial quality of industrial egg white has so far been poorly investigated in Europe. Correa and others (2008) have reported a bacterial concentration of 2 log CFU/mL in Spanish pasteurized liquid egg white products, corroborating the average low concentrations highlighted in the present study. The literature is more abundant regarding the contamination of pasteurized whole egg products. Whole egg products are more likely to be contaminated due to the fact that, unlike egg white, they represent optimal media for bacterial growth (Baron and Jan 2010 for review). However, the pasteurization of whole egg products is possible under stronger conditions than those used for egg white, due to the higher heat-resistance of egg yolk proteins. The resulting level of contamination reported in the literature is nearly similar as that found in pasteurized liquid egg Vol. 80, Nr. 2, 2015 r Journal of Food Science M391
M: Food Microbiology & Safety
Bacteria from liquid egg white . . .
Bacteria from liquid egg white . . . Table 1–Plate-counting of bacteria in raw and pasteurized (57 °C for 6 min) liquid egg white samples in Brain Heart Infusion-Yeast extract (BHI-YE) agar under aerobic and anaerobic conditions at 30 °C for 48 h. Results are expressed as the mean of 3 technical replications in log CFU/mL ± standard deviation. Before pasteurization Batch number
M: Food Microbiology & Safety
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 Mean Number (percentages) of contaminated samples
After pasteurization
Aerobic flora
Anaerobic flora
Aerobic flora
Anaerobic flora
± 0.03 ± 0.1 ± 0.3 ± 0.1 ± 0.1 ± 0.2 ± 0.1 ± 0.0 ± 0.2 ± 0.1 ± 0.1 ± 0.1 ± 0.3 ± 0.1 ± 0.1 ± 0.1 ± 0.0 ± 0.1 ± 0.2 ± 2.8 ± 0.1 ± 0.3
