Microbial Stability and Fate of Salmonella Enteritidis in Halva, a Low-Moisture Confection

Transcription

1 181 Journal of Food Protection, Vol. 61, No.2, 1998, Pages Copyright, International Association of Milk, Food and Environmental Sanitarians Microbial Stability and Fate of Salmonella Enteritidis in Halva, a Low-Moisture Confection P. KOTZEKIDOU* Department of Food Science and Technology, Faculty of Agriculture, Aristotle University of Thessaloniki, P.B. 250, Thessaloniki, Greece MS 97-26: Received 17 February 1997/Accepted 23 April1997 ABSTRACT A traditional low-moisture confectionery, halva, was studied with respect to microbial stability over prolonged storage. It was kept under refrigeration or at room temperature in air-sealed or vacuum packaging in moisture-proof material. Microbial stability of commercial samples was evaluated with regard to the following groups of microorganisms: aerobic plate count, Enterobacteriaceae, enterococci, sulfite-reducing clostridia, aerobic mesophilic and thermophilic sporeformers, staphylococci, Staphylococcus aureus, Salmonella spp., lipolytic microorganisms, yeasts and molds. In all samples tested the above microorganisms were in acceptable levels, while sulfite-reducing clostridia, Salmonella spp., and molds were not detected. The potential for survival of Salmonella Enteritidis in the product was evaluated by artificial contamination. Inoculum surviving after the immediate significant decrease was still recovered after 8 months of storage. The reduction of salmonellae during storage cannot be predicted on the basis of the a w alone. A low-moisture confection called halva (known also as halvah, chalva, chalwa, halawa) is highly popular in the Eastern Mediterranean countries and the Middle East. It consists of tabena, sugar, citric acid, and Saponaria officinalis (soapwort; Family Caryophyllaceae) root extract (3). Production of halva on an industrial basis has been described in detail by Herda (6), who also reported that halva is composed of 50% tahena, 25 to 35% sugar, 12 to 25% glucose, and 1% additives, such as flour and whipping agents. In some varieties of hal va cocoa powder or nuts such as pistachios and walnuts are mixed in with the halva for a richer, more flavorful taste. The base for production of halva is tabena (or tehineh), an oily paste-like product obtained by the milling of dehulled roasted sesame seeds. Tahena is composed of 24.7% protein, 58.9% fat, 2.3% fiber, 3.0% ash, and <1.0% moisture (15). The oil of tabena has a composition of 42.4% oleic, 39.7% linoleic, 9.8% palmitic, and 6.4% stearic acid. Tabena is rich in sulfur-containing amino acids and tryptophan. Information on the chemical composition of halva is scarce. According to the Standards Institution ofisrael (16) halva has the following composition: :2:24% fat, :2:8.5% protein, :::;55% sucrose, :::;2.0% fiber, :::;3.0% H 2 0, :::;1.7% ash, :::;0.1% saponins, and :::;12 mg of Cu, :::;1.0 mg of As, and :::;2.0 mg of Pb per kg. Halva is a traditional low-moisture confection which is usually packaged in moisture-proof materials in order to protect it from the frequently changing relative humidity conditions of the external environment. Changing temperature and humidity conditions to which the product is exposed during production, storage, distribution, and use * Author for correspondence. Fax: <kotzekid@egno.auth.gr> should be recognized as an important microbial stability factor. Temperature changes can cause surface condensation problems. These localized increases in a w can lead to growth of microorganisms on the surface of the product (17). On the basis of the above considerations there is a need to consider the microbiological quality of halva, which is lacking in the literature. The manufacture of a variety of halva requires the use of cocoa powder, an ingredient that carries a risk of contamination with Salmonella spp. (21). Thus, the extent to which salmonellae are protected by reduced a w and product handling is of special concern to manufacturers of the product. The main objective of this investigation was to assess the microbiological quality of halva produced in a representative Greek processing plant. The microbial stability of the product was evaluated by following the population numbers of certain groups of microorganisms over a period of time exceeding the shelf life of the product. In addition, the potential for survival and growth of Salmonella Enteritidis was evaluated by inoculating the freshly produced halva and following the Salmonella population during storage. MATERIALS AND METHODS Manufacturing of halva. Halva is manufactured in specialized factories in Greece. The methods used vary from one factory to another; however, the basic principles followed for the preparation of halva are the same. Briefly, the dehulled sesame seeds are cleaned, sieved, washed in hot water (approximately 50 C), dried by hot air blowing, and roasted ( to C). The roasted sesame seeds are then ground to a viscous oily paste (tahena) which is the major constituent of halva (approximately 50%, wt/wt). Tahena is mixed with sugar (approximately 35%, wt/wt), glucose (approximately %, wt/wt), citric acid, Saponaria officinalis root extract, and vanillin flavor. The mixture is heated to 1 C, poured into 50-liter round vessels and mechanically handled with a spatula

2 182 KOTZEKIDOU J. Food Prot., Vol. 61, No.2 until the texture is chewy and cohesive. Cocoa powder or nuts are mixed into some varieties. It is then put into plastic containers ranging in size from 500 g to 3 kg. The product was packed first in cellophane and then in a packaging film consisting of polyvinylidene chloride, cellophane, polyvinylidene chloride, aluminum foil-metallized polypropylene with m film thickness; the oxygen transmission rate was approximately ml/m 2 /24 h/atm and the vapor permeability < 1 g HzO/m 2 /24 h. For the purpose of this study commercially produced samples of hal va were packed in air or vacuum packaging, then stored refrigerated (6 C) or at room temperature (18 to C) until analysis. Inoculated microorganism. Two Salmonella Enteritidis strains were obtained from the Institute of Food Hygiene, Thessaloniki, Greece. Stock cultures of the microorganism were maintained in brain heart infusion agar (Difco Laboratories, Detroit, MI) at 4 C. A pool inoculum of two strains was prepared from 24-h cultures grown in Standard I broth (Merck AG, Darmstadt, Germany) at 37 C and diluted to obtain a level of inoculation approximately 1 X 7 CFU/g of hal va. Inoculated samples. A slice of halva was inoculated with 0.1 ml of S. Enteritidis inoculum in multiple locations. The inoculum was evenly distributed over the slice and sandwiched between another slice. The inoculated product was packaged or vacuum packed as the commercial samples and stored refrigerated (6 C) or at room temperature (18 to C) until analysis. Analytical methods. Proximate analysis as well as ph determinations were made using established AOAC methodology (2). Water activity determinations were made using a Novasina aw-measuring system (Novasina, Switzerland). Microbiological analysis. To determine the microbiological quality of halva, 2 duplicate -g samples were taken aseptically from each packaging condition, transferred to sterile plastic pouches and homogenized for 60 s with 1 ml of sterile diluent which contained (per liter) peptone (1 g) and Tween (1 ml) using a Stomacher Lab-Blender 400 (Seward Medical, London, UK). Appropriate dilutions of the samples were prepared in sterile peptone water (0.1 %) and plated in duplicate onto growth media to estimate microbial counts. The aerobic plate count was determined by cultivating suitable dilutions of the sample on plate count agar (Merck) and incubating the plates at a temperature of C for 3 days. Total Enterobacteriaceae were determined by plate count on violet red bile dextrose agar after 24 h of incubation at 37 C. Enterococci were determined by using membrane-filter enterococcus agar according to Slanetz and Bartley (Merck) with incubation at 37 C for 48 h. The counts of sulfite-reducing clostridia were determined by cultivation of diluted samples on SPS agar incubated anaerobically in BBL (BBL Microbiology Systems, Cockeysville, MD) Gas Pak jars with HiCOz cartridges at 37 C for 48 h. The aerobic sporeformers count was determined as described by APHA () on tryptone glucose extract agar with incubation at and 37 C for 5 days and at 55 C for 48 h. Staphylococci were determined by surface plating on Baird-Parker agar (Merck) and incubating the plates at 37 C for 48 h. Staphylococcus aureus was determined as described by APHA (). The counts of yeasts and molds were determined on malt extract-yeast extract 40% (wt/wt) glucose agar () at C for 5 to 7 days. The counts of lipolytic microorganisms were determined by cultivation of diluted samples on Tributyrin agar (Merck) with incubation at 2SOC for 3 days. The examination of commercial samples for Salmonella spp. was carried out according to the AOAC procedure (2). For preenrichment, a 25-g sample was blended with 225 ml of lactose broth and incubated at 3SOC for 24 h. For enrichment, replicate I-ml portions were transferred from lactose broth to ml of selenite cystine broth and incubated at 3SOC for 24 h. For isolation of Salmonella spp., 3-mm loops of selenite cystine cultures were streaked on xylose lysine desoxycholate agar and bismuth sulfite agar, followed by incubation at 3SOC for 24 h. Salmonella counts were determined in inoculated halva samples according to the following procedure. A 25-g sample was blended with 225 ml of lactose broth. The homogenate was enriched at 3SOC for 18 to 24 h for revival of injured cells and then inoculated by surface plating on salmonella-shigella agar (Merck). The plates were incubated at 37 C for 24 h. RESULTS AND DISCUSSION Microbial stability of commercial product. The results from chemical and microbiological analyses of halva are presented in Tables 1 and 2, respectively. The high percentage of solids, which results in an a w of 0.176, acts as antimicrobial barrier to provide stability. However, product handling is another source of microbial stability problems. Even though the product is heated at 1 C, slicing and packaging provide ample opportunity for surface recontamination. Localized increases in a w may lead to growth of microorganisms on the surface of the product. Migration of moisture can take place within a product in a moisture-proof package, especially if there is a temperature differential (5). The cycling of high and low temperatures such as may occur during storage can result in microbial stability problems due to creation of surface conditions with high a w Most companies overpack the product in order to counter moisture gain problems and carefully select packaging materials. A substantial amount of the product is packaged in nonhermetic containers or sold in bulk. So the product is subject to moisture gain dependent upon storage relative humidity, the permeability of the package material to water vapor, etc. Therefore, the surface of the product, where oxygen could be more readily available, is the region which should be more concerned with potential outgrowth of aerobic microorganisms. In addition, it is a region where stability has to be considered independently from the bulk of the product. Vegetative cells and spores of aerobic microorganisms as well as lipolytic microorganisms predominated in the product. The higher populations of spores in comparison to vegetative cells can be explained by the fact that spores exhibit maximum heat resistance around an a w of 0.2 to 0.4, while vegetative cells are most heat resistant in the intermediate moisture food water activity range (8). Heating of bacterial spores in a dry environment causes the removal of TABLE 1. Physicochemical analysis of commercially produced halva Parameter Value ph 6.0 aw Moisture content (%) 2.8 Protein (%) 14.9 Fat (%) 31.8 Ash (%) 1.7 Total sugars (%) 49.5

3 J. Food Prot., Vol. 61, No. 2 MICROBIAL STABILITY AND SALMONELLA SURVIVAL IN HALVA 183 TABLE 2. Microbial analyses of freshly produced Microorganisms halva Aerobic plate count 1.1 X 2 Enterobacteriaceae 1.2 X 1 Enterococci. 2.3 X 1 Sulfite-reducing clostridia < at C 3.5 X 2 at 3rC 2.1 X 2 at 55 C 1.2 X 2 Staphylococci 2.5 X 2 Staphylococcus aureus NDa yeasts... 7 X 1 Lipolytic microorganisms 8 X 2 Salmonella spp. ND a ND, not detected. Mean CFU/g ± SD (n = 6) moisture from within the spore, thereby preventing plasticization of the matrix and subsequent inactivation (14). There were no detectable levels of sulfite-reducing clostridia, Staphylococcus aureus, Salmonella spp., and molds. Effective handling of the product can prevent such pathogenic contaminants from becoming established and subsequent cross-contamination. Through environmental (postprocess) contamination controls the presence of significant levels of the above microorganisms is avoidable. The high population of staphylococci can be explained by the pronounced increase in heat resistance of cocci in dry fat. But staphylococci which survive heating by reason of fat protection are incapable of subsequent growth because they remain in the fat. The mechanism of fat protection in foods appears to rest upon the absence of localized moisture (7). The aerobic plate count of halva stored under refrigeration or at room temperature in air-sealed or vacuum packaging did not increase significantly during storage (Fig. 1). In aw-reduced environments, the lag phase lengthened when the temperature was lowered, and this variation depends on the aw-controlling solute. Sucrose, which is mainly used to control aw in hal va, is not a permeable solute, so cells need time to synthesize an intracellular compatible solute such as glutamic acid or proline to balance the external osmolality. This osmoregulation process requires considerable energy, and since at low temperature chemical reactions are slowed, the combined effect of low temperature and low aw controlled by impermeable solutes lengthens the lag phase dramatically (11). An awincrease, especially at the surface of the product, in a range permitting bacterial growth may cause an increase of aerobic plate counts. In vacuum packaging the low aw allowed marked bacterial survival that was much greater at refrigeration temperatures. The mesophilic aerobic sporeformers remained constant after 60 days of storage in air-sealed packaging stored under refrigeration, whereas they declined in all other storage conditions (Fig. 2). Lipolytic microorganisms did not increase significantly during storage and remained in low numbers (Fig. 3). In fatty foods like halva the presence of lipolytic microorganisms is very important since lipase is extremely resistant against heat denaturation. Even though sterilization processes before storage kill the bacteria, the heat-resistant lipase previously produced by them remains active. Since the enzyme is also stable and active during storage at low temperatures as well as at low water activities, lipase can cause changes in quality in fatty foods during prolonged chilled or dehydrated storage (1). After 8 months of storage vegetative cells and spores of mesophilic bacteria and lipolytic microorganisms predominated in the product (Table 3). Staphylococci and enterococci survived only in refrigerated samples. Although microorganisms were unable to Mesophilic aerobic sporeformers Aerobic plate count ~ 2.;:! u III :a '" u 1.5 -' '" 0 -' FIGURE 1. Aerobic plate counts of halva stored under refrigeration (6 C) or at room temperature (18 to C) in air-sealed or vacuum packaging. Air-sealed packaging, stored at 6 C; 0 vacuum packaging, stored at 6 C;.. air-sealed packaging, stored at 18 to C; D. vacuum packaging, stored at 18 to C FIGURE 2. Counts (log CFU/g of halva) of mesophilic aerobic sporeformers in air-sealed or vacuum-packed halva stored under refrigeration (6 C) or at room temperature (18 to C). Air-sealed packaging, stored at 6 C; 0 vacuum packaging, stored at 6 C;.. air-sealed packaging, stored at 18 to C; D. vacuum packaging, stored at 18 to C. 60

4 184 KOTZEKIDOU J. Food Prot., Vol. 61, No Lipolytic microorganisms TABLE 4. Range of microbial populations of 5 commercial samples of halva from each of four different processing plants Mean CFU/g ± SD (n = ) in processing plant no. Microorganisms ~ 2.a u III ~ 1, FIGURE 3. Counts of lipolytic microorganisms in air-sealed or vacuum-packed halva stored under refrigeration (6 C) or at room temperature (18 to C). Air-sealed packaging, stored at 6 C; D vacuum packaging, stored at 6 C;... air-sealed packaging, stored at 18 to C; 1::" vacuum packaging, stored at 18 to C. grow, the a w and/or fat contents permitted marked microbial survival. The distribution of microbial counts obtained from halva samples from four factories is summarized in Table 4. TABLE 3. Microbial analyses of halva after 8 months of storage in air-sealed or vacuum packaging under refrigeration (6 C) or at room temperature (18 to C) 60 Mean CFU/g ± SD (n = 6) Air-sealed Vacuum Air-sealed Vacuum packaging packaging packaging packaging Microorganisms WC) (6 C) D (l8- C) D (l8- C) D Aerobic plate count 1.2 X 2 9 X 1 6 X 1 3 X 1 Enterobacteriaceae < < < < Enterococci < 1.2 X ' < < Sulfite-reducing clostridia < < < < at C 1.8 X X X X 2 at 37 C lx 2 7 X X 2 8 X 1 at 55 C < < < < Staphylococci 7 X 1 9 X 1 < < Staphylococcus aureus ND" ND ND ND Yeasts < < < < Lipolytic microorganisms 2 X X X X 2 Salmonella spp. ND ND ND ND a ND,not detected. Aerobic plate count 6 X X X 2 4 X 1 Enterobacteriaceae 3 X 1 2 X 1 2 X 1 < Enterococci < 7.6 X X 1 < Sulfite-reducing clostridia < < < < at C 9 X 1 5X 3 8 X 2 < at 37 C 8 X 1 4X 3 6 X 2 < at 55 C 4 X 1 5X 2 5 X 2 < Staphylococci 7 X X X 2 < Staphylococcus aureus ND" 8 X 1 7 X 1 ND Yeasts 6 X X X 2 4 X 1 Lipolytic microorganisms 5 X 1 2x 2 4X 2 6 X 1 Salmonella spp. ND ND ND ND a ND,not detected. Enterobacteriaceae counts were not greater than CFU/g and there was not a correlation between the counts of Enterobacteriaceae and the presence of Salmonella cells in the product. In the samples from two factories S. aureus was detected and the average counts were not greater than CFU/g. It is of particular interest if S. aureus is present as a surface contaminant. In this case oxygen may be not limited and furthermore it may be assumed that if surface growth occurred, ph and a w may be changed (13). In the main constituent of halva, tahena, Ayaz et al. (3) reported that the average S. aureus count in ten processing plants was 56 CFU/g and Salmonella spp. were isolated from samples from two plants. The high aerobic sporeformers counts observed in some samples would be capable of directly causing spoilage if there is surface condensation, but the limiting a w for the growth of sporeformers is 0. (18). Molds were not detected in any samples tested. The yeast counts in one factory were 4.9 X 3 CFU/g. High yeast populations in halva are important because the limiting a w for osmotolerant yeasts in nutrient-rich confectioneries is 0.60 (18). Special care should be given to the control of osmotolerant yeasts in processing equipment. Product residues in equipment are naturally selective for osmotolerant yeasts. Air, dust, and moisture provide means of microbial transmission in the plant environment. Uncontrolled moisture may provide opportunities for microbes to proliferate and establish themselves permanently in the plant (). Fate of inoculated Salmonella Enteritidis. The behavior of artificially inoculated S. Enteritidis during storage of halva is shown in Table 5. The batches of halva used for these experiments contained no Salmonella spp. in 25-g samples tested by the preenrichment-enrichment procedure.

5 J. Food Prot., Vol. 61, No.2 MICROBIAL STABILITY AND SALMONELLA SURVIVAL IN HALVA 185 TABLE 5. Survival of Salmonella Enteritidis in air-sealed or vacuum packed halva stored under refrigeration (6 C) or at room temperature (18 to C), initial viable inoculum, 3.87 log CFU/g Salmonella Enteritidis Mean CFU/g (SD), n = 6 Air-sealed Vacuum Air-sealed Vacuum Storage packaging packaging packaging packaging time (6 C) (6 C) (I8- C) (I8- C) 6 days 2.82 (0.22) 2.91 (0.13) 3.58 (0.32) 3.63 (0.38) 9 days 2.54 (0.) 2.65 (0.24) 3.35 (5) 3.53 (0.46) 15 days 2.53 (0.33) 2.68 (0.26) 2.41 (0.74) 2.81 (7) 25 days 2.83 (0.45) 2.94 (0.75) 2.14 (3) 2.94 (0.37) 40 days 3.14 (4) 3.59 (0.36) 3.06 (0.34) 3.50 (0.67) 2 months 2.24 (0.85) 2.98 (0.48) 2.29 (0.38) 2.83 (2) 7 months 2.35 (0.67) 2.94 (6) 2.12 (0.15) 2.72 (0.44) 8 months 2. (0.49) 2.76 (0.49) 2.15 (4) 2. (0.31) The wet inoculum of 7 log CFU/g added to halva samples decreased by about 3 log/g immediately after inoculation and the S. Enteritidis population recovered was 3.87 log CFU/g due to an osmotic shock effect. Several investigators have reported a decrease in viable counts occurring a short time after salmonellae cells were added to dry materials as a wet inoculum (9, 12). As shown in Table 5, although S. Enteritidis cells do not multiply in an environment with an a w of 0.176, they may survive for at least 8 months. Salmonella cells held at low a w levels may maintain viability for relatively long periods of time (19). Interactions between a w and extrinsic environmental factors, such as storage in air or under vacuum, and temperature, appear to play important roles in the survival of salmonellae. The survival of inoculated S. Enteritidis under refrigeration as well as at room temperature was enhanced in product stored under vacuum rather than air. A similar observation was reported by Christian and Stewart (4) on the survival of S. newport at 25 C in a w between 0.00 and During the first 9 days of storage S. Enteritidis survived better at room temperature than under refrigeration, but thereafter survival varied little between the two storage temperatures. After 8 months of storage the greatest decline in viable S. Enteritidis counts was observed in air-sealed packed product stored at room temperature, whereas S. Enteritidis survived better in vacuumpacked halva stored under refrigeration. Although osmotic shock causes an immediate significant decrease in viable counts of Salmonella CFU, the other factors such as temperature and oxygen did not greatly affect the survival of S. Enteritidis during storage. Salmonella spp. contaminating the product after processing may survive for at least 8 months. Salmonellosis could result, depending on the extent of contamination and surface condensation. REFERENCES 1. Andersson, R. E. 19. Microbial lipolysis at low temperatures. Appl. Environ. Microbiol. 39: AOAC. 19. Official methods of analysis, 15th ed. Association of Official Analytical Chemists, Washington, D.C. 3. Ayaz, M., W. N. Sawaya, and A. AI-Sogair Microbiological quality of tehineh manufactured in Saudi Arabia. J. Food Prot. 49: Christian, J. H. B., and B. J. Stewart Survival of Staphylococcus aureus and Salmonella newport in dried foods, as influenced by water activity and oxygen, p In Microbiological safety of food. Proceedings of the 8th International Symposium on Food Microbiology. Academic Press, London. 5. Hazeu, W., and H. J. Heuck Changes of humidity inside packages due to environmental conditions, p Monograph Series 23. Society of Chemical Industries, London. 6. Herda, J.-M. 19. Large-scale production ofhalawa in Saudi Arabia. Food Eng. Int. 5 : Hersom, A. c., and E. D. Hulland. 19. Canned foods: thermal processing and microbiology, 7th ed. Churchill Livingstone, Edinburgh. 8. Hsieh, E, K. Acott, and T. P. Labuza Death kinetics of pathogens in a pasta product. J. Food Sci. 41: Juven, B. J.,N. A. Cox, J. S. Bailey, J. E. Thomson, O. W. Charles, and J. V. Shutze Survival of Salmonella in dry food and feed. J. Food Prot. 47: Lenovich, L. M., and P. J. Konkel Confectionery products, p In C. Vanderzant and D. E Splittstoesser (ed.), Compendium of methods for the microbiological examination of foods, 3rd ed. American Public Health Association, Washington, D.C. 11. Li, K.-Y., and J. A. Torres Water activity relationships for selected mesophiles and psychotrophs at refrigeration temperature. J. Food Prot. 56: Mossel, D. A. A., and M. J. Koopman Losses in viable cells of salmonellae upon inoculation into dry animal feeds of various types. Poult. Sci. 44: Notermans, S., and C. J. Heuvelman Combined effect of water activity, ph and sub-optimal temperature on growth and enterotoxin production of Staphylococcus aureus. J. Food Sci. 48: , Sapru, V., and T. P. Labuza Glassy state in bacterial spores predicted by polymer glass-transition theory. J. Food Sci. 58: Sawaya, W. N., M. Ayaz, J. K. Khalil, and A. E Ai-Shalhat Chemical composition and nutritional quality of tahineh (sesame butter). Food Chern. 18: Standards Institution of Israel Sesame halvah. Food Sci. Technol. Abstr. (5): Torres, J. A Microbial stabilization of intermediate moisture food surfaces, p In L. B. Rockland and L. R. Beuchat (ed.), Water activity: theory and applications to food. Marcel Dekker, Inc., New York. 18. Troller, J Food spoilage by microorganisms tolerating low-aw environments. Food Technol. 33(1): Troller, J. A Adaptation and growth of microorganisms in environments with reduced water activity, p In L. B. Rockland and L. R. Beuchat (ed.), Water activity: theory and applications to food. Marcel Dekker, Inc., New York.. Vanderzant, C., and D. E Splittstoesser (ed.) Compendium of methods for the microbiological examination of foods, 3rd ed. American Public Health Association, Washington, D.C. 21. World Health Organization Weekly epidemiological record. 48 (39, 40). World Health Organization, Geneva.