- Email: [email protected]
Survival and growth of Bacillus cereus during Gouda cheese manufacturing
Food Control 12 (2001) 31±36
www.elsevier.com/locate/foodcont
Survival and growth of Bacillus cereus during Gouda cheese manufacturing Grace Rukure 1, Bernie H. Bester * Department of Food Science, University of Pretoria, Pretoria 0002, South Africa Received 18 November 1999; received in revised form 10 April 2000; accepted 13 April 2000
Abstract Survival and growth of Bacillus cereus was investigated during manufacturing of Gouda type cheese. The cheese was prepared in the pilot plant from pasteurised milk arti®cially contaminated with spores to give a ®nal concentration of approximately 102 B. cereus spores per millilitre of cheese milk. B. cereus was enumerated by surface plating on B. cereus selective media and lactic acid bacteria were enumerated on lactic agar and MRS agar (de Man-Rogosa-Sharpe). Samples were taken for microbiological analysis of the milk before renneting, curd at cutting, at half whey removal, at ®nal whey removal, at hooping of the curd, the cheese after pressing, after brining, after 1 week, after 2 weeks, after 4 weeks and after 6 weeks. The spores germinated into vegetative cells, which grew and reached a maximum of approximately 104 CFU per gram of cheese at hooping (about 4 h after renneting). After pressing (approximately 16 h after renneting ) the viable cells were reduced to less than 102 CFU per gram. After brining (about 40 h after renneting) B. cereus was not detected in the cheese curd. At this stage the conditions of the cheese, particularly lower moisture content and aw , lower Eh, high salt content, depleted lactose content combined with high acidity may have inhibited the growth of B. cereus. B. cereus did not aect the growth of lactic acid bacteria during cheese manufacturing. Lactic acid bacteria grew from 107 to 109 CFU per gram of curd during cheese manufacturing and stayed fairly constant at about 109 for 6 weeks. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Gouda cheese; B. cereus; Lactic acid bacteria
1. Introduction Gouda is a semi-hard cheese variety made from cowsÕ milk and originated from Holland (Kosikowski, 1978). Milk used for cheese-making should be of a good bacteriological quality to avoid undesirable fermentation and enzymic reactions and without any substances that inhibit or interfere with the growth of the starter bacteria (Chapman & Sharpe, 1981). Milk is an excellent medium for microbial growth. Milk as produced in the alveoli of a healthy cow is free from micro-organisms, but the farm environment makes it impossible to exclude bacteria during milking (Andersson, Ronner, & Granum, 1995). The bacterial load of raw milk ranges from 103 CFU per millilitre when hygiene is good to 107 CFU per millilitre when hygiene is poor. The bacteria consist *
Corresponding author. Tel.: +27-12-420-3209; fax: +27-12-4202839. E-mail address: [email protected] (B.H. Bester). 1 Present Address: Blair Research Institute, P.O. Box CY 573, Causeway, Harare, Zimbabwe. Tel.: 263-4-792747; fax: 263-4-792480.
of psychrotrophs, small numbers of lactic acid bacteria, spore forming Gram-positive rods, coryneform bacteria, micrococci and coliforms. Psychrotrophs will continue to multiply during transport and storage of milk. Pasteurisation (72°C for 15 s) of the milk prior to Gouda cheese manufacture will reduce the microbial load, but the thermoduric micro-organisms with psychrotrophic properties survive (Andersson et al., 1995) and of importance to this research is the species Bacillus cereus. Occurrence of B. cereus in milk has been reported since 1916, and this bacterium is a common contaminant of raw milk produced on some dairy farms (Ahmed, Moustafa, & March, 1983). It can also be found in large numbers in dairy products (Ahmed et al., 1983). In a survey by Wong, Chen and Chen (1988) on dairy products, 52% of ice creams, 35% of soft ice creams, 29% of milk powders, 17% of fermented milks, and 2% of pasteurised milks and fruit ¯avoured milks were found to be contaminated with B. cereus. The concern about its presence in dairy products is that it may cause certain defects in the products or produce toxins and thus cause food poisoning.
0956-7135/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 9 5 6 - 7 1 3 5 ( 0 0 ) 0 0 0 1 6 - 5
32
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
At counts above 2 105 /ml B. cereus causes o ¯avours such as unclean, fruity, bitter, putrid, rancid and yeasty (Meer, Baker, Bodyfelt, & Griths, 1991). When growth continues, the product shows the defect sweet curdling in homogenised low-pasteurised milk and bitty cream in low-temperature pasteurised milk. It has been estimated that more than 25% of the shelf-life problems encountered with pasteurised milk are due to the proliferation of Bacillus species (Griths, 1992). Although low numbers of B. cereus in foods pose no direct health hazard, if these foods are mishandled, growth of this organism might result in a direct health hazard. Large numbers of B. cereus produce enterotoxin in the food before it is consumed and cause illness (Macrae, Robinson, & Salder, 1993). According to the South African regulations relating to Herbs and Spices (R 1468) of the Foodstus, Cosmetics and Disinfectants Act 54 of 1972, any foodstu containing B. cereus in 20 g of sample is deemed harmful or injurious to human health. The objectives of the project were to determine the survival and growth of B. cereus during manufacture of Gouda type cheese and to determine whether B. cereus would aect the growth of lactic acid bacteria during Gouda cheese manufacturing. Such information will indicate whether Gouda cheese manufactured from milk severely contaminated with B. cereus would pose a health hazard to consumers of such cheese or of food products in which such cheese is used. 2. Materials and methods 2.1. Manufacture of Gouda cheese Gouda type cheese was manufactured in the pilot plant of the Department of Food Science, University of Pretoria, using the method described by Kosikowski (1978). Lysozyme was not used during cheese manufacturing. The milk containing spores for the experimental cheese and milk without spores for the control cheese was pasteurised in a batch pasteuriser at 63°C for 30 min and cooled down to 30°C. The milk was then inoculated with 0.75% of a starter culture CH±N22 (a mixture of Lactoccus lactis and L. lactis sub sp. cremoris). The stock cultures were supplied in freeze-dried form by CHR HansenÕs Laboratories (Denmark). The milk was renneted, using Rennilase type T rennet (Novo Industries, Denmark). After 30±45 min at 30°C, the coagulated milk was cut and after 10 min resting of the cut curd, part of the whey (10% of the volume of the milk) was removed and replaced by the same amount of water. The temperature of the curd±whey mixture was then gradually increased from 30°C to 36°C (cooking temperature) over a period of 30±45 min. On reaching the cooking temperature, whey was drained to 50% of the
original volume of milk (half-whey). The curd and whey mixture was held for 1±1 12 h at the cooking temperature. The whey was then drained, the curd put into moulds and pressed for 3±4 h. The green cheese was brine-salted for 24 h in saturated brine solution, allowed to dry-o for 2 days, vacuum packed in plastic bags and cured for 6 weeks at 12±15°C. 2.2. Preparation of B. cereus spore suspensions Spores of B. cereus (ATCC 10702) were produced on KG agar slants (Refai, 1979). After incubation for 3±7 days at 37°C, the growth was scraped o the agar slants and the spores collected by centrifugation, washed by resuspending the sedimented spores in sterile distilled water and centrifuging again. This was repeated ®ve times and the suspended spores (about 107 spores/ml) stored at 4±7°C. 2.3. Survival and growth of B. cereus during Gouda cheese manufacturing In order to determine the survival and growth of B. cereus during Gouda cheese manufacturing, two batches of cheese were prepared. Batch 1. Experimental cheese with B. cereus spores added: A batch of 100 l of fresh whole milk was inoculated with 10 ml of spore suspension (approximately 107 spores of B. cereus per millilitre), to give a spore concentration of approximately 102 (between 102 and 103 ) CFU per millilitre in the milk before pasteurisation. The milk was thoroughly mixed and pasteurised at 63°C for 30 min. After pasteurisation the milk was used to manufacture Gouda cheese. Batch 2. Control cheese without B. cereus spores added: A second batch of 100 l of fresh whole milk, to which no B. cereus spores were added, was pasteurised at 63°C for 30 min. After pasteurization the milk was used to manufacture Gouda type cheese. 2.4. Sampling Samples were taken at dierent stages during the Gouda cheese manufacturing and analysed for the parameters as shown in Table 1. 2.5. Microbiological analysis 2.5.1. Preparation of dilutions Tenfold serial dilutions of the samples were made by aseptically transferring 25 g of sample into 225 ml of sterile saline peptone water to give a 10ÿ1 dilution (IDF, 1992). Samples were then homogenised for 20 s using a Stomacher 400 laboratory blender (Seward Laboratory UAC, Britain). Further 10-fold dilutions of up to 10ÿ8 were made by transferring 1 ml of successive serial
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
33
Table 1 Sampling plan during Gouda cheese manufacturing showing process steps and stages at which samples were drawn for microbiological analysesa
a
Process step
Time (h)
B. cereus
LA
MRS
pH
At renneting Curd at cutting At 12 whey removal At ®nal whey removal At hooping (moulding) After pressing After brining After 1 week After 2 weeks After 4 weeks After 6 weeks
0 1.05 2.25 3.15 3.86 15.96 40 168 338 672 1008
+ + + + + + + + + + +
ns + + + + + + + + + +
ns + + + + + + + + + +
+ + + + + + + + + + +
+ sample taken; ns sample not taken; LA Lactic agar; MRS de Man-Rogosa-Sharpe agar.
dilutions into universal bottles containing 9 ml of sterile saline peptone water. 2.5.2. Plate counts by the spread-plate technique Plates for the spread-plate technique were prepared by pouring 12±15 ml of the relevant sterile culture media into sterile petri dishes. The plates were dried at 35°C in an incubator. The dried plates were then inoculated with 0.1 ml of the homogenate or dilutions thereof and appropriately labelled. Each inoculum was spread over the surface of the dried plates using a sterile bent glass rod. After incubation the colonies were counted using a scienti®c colony counter (Stuart Scienti®c, UK) and the results were expressed as CFU per gram or CFU per millilitre depending on whether the sample was a solid (cheese) or a liquid (milk). 2.5.3. B. cereus counts B. cereus selective agar (Oxoid CM 617) was used for the enumeration of B. cereus. The media were prepared according to the manufacturerÕs instructions and poured into sterile petri dishes for the spread-plate technique. The inoculated plates were incubated at 30°C for 24 h, the number of colonies were counted and expressed as CFU per gram (or CFU per millilitre) of sample. The plates were also examined for typical B. cereus colonies which were rough in texture, turquoise to peacock blue in colour, ¯attened, irregular and surrounded by a greyish zone of egg yolk precipitate. Typical B. cereus colonies were con®rmed by Gram staining which showed Gram-positive rods containing spores which did not stain (Harrigan & McCance, 1976). The endospores were ellipsoidal at the centre of the cell and about 1.2 lm in size. 2.5.4. Lactic acid bacteria Lactic agar (Merck) and MRS agar (Oxoid CM361) were used for the enumeration of lactic acid bacteria. The media were prepared according to the manufac-
turersÕ instructions and poured into petri dishes for the spread-plate technique. The inoculated plates were incubated at 37°C for 48 h in anaerobic jars containing Anaerocult A (Merck, Darmstadt, Germany). Using a Harrison disc, a representative number of colonies were isolated from plates selected for computing a plate count and these colonies then examined by the Gram stain and by testing for catalase production with hydrogen peroxide (Harrigan & McCance, 1976). Catalase negative colonies that contained Gram-positive cocci or long or short rods, were recorded as lactic acid bacteria. 2.5.5. Statistical analysis Statistical analyses were done using SPSS version 8.0 for Windows (SPSS, Chicago, USA). Bacterial counts were converted to log10 CFU per gram. Means and standard deviations were calculated. The least signi®cant dierence between the means was determined.
3. Results 3.1. Changes in B. cereus, lactic acid bacteria numbers and pH during cheese manufacture. Table 2 shows the changes in B. cereus, lactic acid bacteria numbers and pH in cheese made from milk inoculated with B. cereus spores (experimental cheese) and cheese without B. cereus spores (control cheese). B. cereus spores survived and germinated into vegetative cells which grew from approximately 102 CFU per gram and reached a maximum of 104 CFU per gram at hooping (about 4 h after renneting). With the inoculum level of approximately 102 CFU per millilitre of milk, a maximun level of approximately 104 CFU per gram of curd (log10 4.4) was reached at the stage of hooping of the curd. Thus, B. cereus increased by 2.2 log cycles, whereafter they rapidly decreased to less than 102 CFU
34
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
Table 2 Changes in numbers of B. cereus and lactic acid bacteria during manufacturing of Gouda cheese Time (h)
Experimental cheese (log10 CFU per gram)
Control cheese (log10 CFU per gram)
(MRS)A
(LA)A
BCA
pH
(MRS)A
(LA)A
BCA
pH
0
nsB
ns
ns
< 2:0
7.3a (0.1) 7.6b (0.5) 8.1c (0.1) 8.2b (0.1) 8.8d (0.1) 9.0dc (0.1) 9.1e (0.1) 9.0de (0.0) 9.0de (0.0) 9.0de (0.0)
7.3a (0.1) 7.6a (0.1) 8.1b (0.1) 8.1b (0.0) 8.8 (0.1) 9.0c (0.1) 9.0c (0.1) 9.0c (0.1) 9.0c (0.0) 9.0c (0.0)
6.60 (0.00) 6.43 (0.06) 6.40 (0.00) 6.30 (0.00) 6.15 (0.07) 6.11 (0.01) 5.19 (0.01) 5.17 (0.01) 5.16 (0.01) 5.15 (0.01) 5.15 (0.01)
ns
1.07
2.2a (0.0) 2.4a (0.1) 3.5b (0.5) 3.5b (0.0) 4.4c (0.1) < 2:0
7.4a (0.0) 7.0b (0.1) 7.8b (0.1) 8.1c (0.1) 8.7d (0.1) 9.0c (0.0) 9.1c (0.0) 9.1e (0.0) 9.0e (0.0) 8.9e (0.0)
7.4a (0.1) 7.4a (0.1) 7.9b (0.1) 8.1c (0.1) 8.7d (0.1) 9.0c (0.0) 9.1c (0.0) 9.1e (0.0) 9.0e (0.0) 9.0e (0.0)
< 2:0
6.6 (0.00) 6.40 (0.00) 6.40 (0.00) 6.15 (0.07) 6.11 (0.01) 5.19 (0.01) 5.17 (0.01) 5.16 (0.01) 5.16 (0.01) 5.16 (0.01) 5.16 (0.01)
2.25 3.15 4.26 15.96 40.00 168 336 672 1008
2
< 2:0 < 2:0 < 2:0 < 2:0 < 2:0
< 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0
* Means with dierent superscripts in the same column are signi®cantly dierent from each other (P < 0.05). Numbers in parentheses are standard deviations. A Lactic acid bacteria on MRS and Lactic agar (LA), B. cereus (BC) on Oxoid CM 617 agar. B ns, not sampled.
Table 3 Means of pH, salt content, moisture content and fat content of the experimental and the control cheese at 6 weeks and compositional results of commercial Gouda cheese obtained from Lolkema and Blaauw (1974) Analysis
Experimental cheese
Control cheese
Commercial Gouda cheese
pH Fat (%) Moisture (%) Salt (%) Salt in moisture (%) Fat in dry matter (%)
5.16 31.2 40.2 2.0 4.74 52.22
5.15 30.0 39.8 1.88 4.51 49.83
5.2 29.2 41 1.5±2.2 ) 49.5
per gram after pressing (approximately 16 h after renneting). A general decrease in pH was observed from pH 6.6 to pH 5.15 for the experimental cheese and from pH 6.6 to pH 5.16 for the control cheese. Maximum numbers of B. cereus were reached when the pH was 6.15. B. cereus did not aect the growth of lactic acid bacteria. Lactic acid bacteria increased by 1.7 log cycles in both the experimental and control cheese until 1 week after renneting of the milk and then remained fairly constant throughout the ripening process of 6 weeks. B. cereus was not found in the control cheese and there was no signi®cant dierence P > 0:05 in the numbers of lactic acid bacteria between the experimental and the control cheese.
3.2. Chemical analyses of the cheese at 6 weeks Table 3 shows the pH, salt content, moisture content and fat content of the experimental and the control cheese at 6 weeks. For comparison the average composition of commercial Gouda cheese is also given. 4. Discussion In cheese milk inoculated with approximately 102 B. cereus spores per millilitre of milk in the present study, B. cereus multiplied during cheese-making to a maximum of approximately 104 CFU per gram at hooping,
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
i.e., after about 4 h. Ungerminated spores could have remained but according to Davies and Wilkinson (1973) the spores must germinate, outgrow and multiply to be of any potential danger. After pressing (approximately 16 h after renneting) B. cereus numbers were reduced to less than 100 CFU per gram of cheese curd (the lowest number detectable with the method used). This is less than the number of 105 per millilitre or per gram which has been reported to cause illness (Overcast & Atmaram, 1974). The initial growth of B. cereus was expected because after pasteurisation the milk was free from other competitor micro-organisms and conditions were still quite favourable for their growth. B. cereus grew nearly at the same rate as the lactic acid bacteria during the early stages of lactic acid fermentation until the LAB reached numbers of 108 per gram and the pH dropped to below 6.15. Thereafter the numbers of B. cereus decreased rapidly. Work by Wong and Chen (1988) on germination and growth of B. cereus in non-fat milk with added lactic acid bacteria also showed that B. cereus was not aected by lactic acid bacteria at the beginning of the fermentation process, but was aected strongly with continued fermentation. The reduction of B. cereus at pH lower than 6.15 is also in accordance with the results found by Driessen (1992) during the manufacture of yoghurt. Killing or inactivation of vegetative B. cereus cells at pH below 6.15 may not be attributed to reduction in pH and lactic acid production only but to a number of factors which act synergistically such as substrate competition, changes in oxidation±reduction potential and production of antimicobial agents (Frank, Marth, & Olson, 1978; Wong et al., 1988). According to Wong et al. (1988) antimicrobial agents such as hydrogen peroxide, formate, acetate or lactate enhanced the inhibitory activity of lactic acid bacteria. However, in their work on the survival and growth of B. cereus in mageu, a sour maize beverage, Byaruhanga, Bester and Watson (1999) came to the conclusion that the inhibition of the Bacillus was attributable to the eect of undissociated lactic acid. Lactic acid bacteria ¯ourished throughout the manufacturing and ripening process reaching numbers as high as 109 CFU per gram of cheese. According to Kosikowski (1978) bacterial numbers of 108 to 109 CFU per gram are enough to enhance cheese ripening. Both experimental and control cheese thus contained enough lactic acid bacterial cells to accomplish ripening of the cheese. Analysis of variance showed no signi®cant dierence P > 0:05 in the isolation rate of lactic acid bacteria on the two media, Lactic agar and MRS agar. The overall composition of both the experimental and control cheese compared well with that of a good Gouda cheese (Lolkema & Blaauw, 1974).
35
5. Conclusion B. cereus spores germinated into vegetative cells, grew and multiplied during Gouda cheese manufacturing. Considering the level of spores inoculated and the detection limits of the plating method used, no viable cells of B. cereus were found in the cheese at the end of the manufacturing process. Therefore, it can be concluded that B. cereus spores were able to survive, germinate and grow into vegetative cells during the early stages of cheese manufacturing but the vegetative cells were not able to survive during the ®nal stages of the manufacturing process and ripening of the cheese.
References Ahmed, A. H., Moustafa, K., & Marth, E. H. (1983). Incidence of Bacillus cereus in milk and some milk products. Journal of Food Protection, 46, 126±128. Andersson, A., Ronner, U., & Granum, P. (1995). What problems does the food industry have with sporeforming pathogens, Bacillus cereus and Clostridium perfrigens?. International Journal of Food Microbiology, 28, 145±155. Byaruhanga, Y. B., Bester, B. H., & Watson, T. G. (1999). Growth and survival of Bacillus cereus in mageu, a sour maize beverage. World Journal of Microbiology and Biotechnology, 15, 329±333. Chapman, H. R., & Sharpe, M. E. (1981). Microbiology of cheese. In: R. K. Robinson, Dairy microbiology (pp. 157±243), vol. 2, London: Applied Science Publishers. Davies, F. L., & Wilkinson, G., 1973. Bacillus cereus in milk and dairy products. In: B. C. Hobbs, & J. H. B. Christian, The microbial safety of food. Proceedings of the Eighth International Symposium on Food Microbiology (pp. 57±67). London: Academic Press. Driessen, F. M. (1992). Importance of Bacillus cereus in fermented milks and processed non-fermented dairy foods. Bulletin of the International Dairy Federation, 287, 11±15. Frank, J. F., Marth, E. H., & Olson, N. F. (1978). Behavior of enteropathogenic Escherichia coli during manufacture and ripening of brick cheese. Journal of Food Protection, 41, 111±115. Griths, M. W. (1992). Bacillus cereus in milk. Bulletin of International Dairy Federation, 287, 18. Harrigan, W. F., & McCance, C. (1976). Laboratory methods in food and dairy microbiology. London: Academic Press. IDF (1992). Preparation of samples and dilutions for microbiological examination. International Dairy Federation Standard, 122B, 1±4. Kosikowski, F. (1978). Cheese and fermented milk foods (2nd ed.) (pp. 16±36, 281±282). New York: F. V. Kosikowski and Associates. Lolkema, H., & Blaauw, J. (1974). Kaasbereiding. Apeldoorn, Netherlands: Landelijk Stichting Beroepsopleiding Levensmiddelindustrie. Macrae, R., Robinson, R. K., & Sadler, M. J. (1993). Encyclopaedia of food science, food technology and nutrition. New York: Academic Press. Meer, R. R., Baker, J., Bodyfelt, F. W., & Griths, M. W. (1991). Psychrotrophic Bacillus spp. in ¯uid milk products: A review. Journal of Food Protection, 54, 968±979.
36
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
Overcast, W. W., & Atmaram, K. (1974). The role of Bacillus cereus in sweet curdling of ¯uid milk. Journal of Milk Food Technology, 37, 233±236. Refai M. K. (1979). Microbiological analysis. Manual of food quality control, vol. 4, Rome: Food and Agricultural Organization of the United Nations.
Wong, H., & Chen, Y. (1988). Eects of lactic acid bacteria and organic acids on growth and germination of Bacillus cereus. Applied and Environmental Microbiology, 54, 2179±2184. Wong, H. C., Chen, L. Y., & Chen, C. L. (1988). Growth germination and toxigenic activity of Bacillus cereus in milk products. Journal of Food Protection, 51, 707±710.
www.elsevier.com/locate/foodcont
Survival and growth of Bacillus cereus during Gouda cheese manufacturing Grace Rukure 1, Bernie H. Bester * Department of Food Science, University of Pretoria, Pretoria 0002, South Africa Received 18 November 1999; received in revised form 10 April 2000; accepted 13 April 2000
Abstract Survival and growth of Bacillus cereus was investigated during manufacturing of Gouda type cheese. The cheese was prepared in the pilot plant from pasteurised milk arti®cially contaminated with spores to give a ®nal concentration of approximately 102 B. cereus spores per millilitre of cheese milk. B. cereus was enumerated by surface plating on B. cereus selective media and lactic acid bacteria were enumerated on lactic agar and MRS agar (de Man-Rogosa-Sharpe). Samples were taken for microbiological analysis of the milk before renneting, curd at cutting, at half whey removal, at ®nal whey removal, at hooping of the curd, the cheese after pressing, after brining, after 1 week, after 2 weeks, after 4 weeks and after 6 weeks. The spores germinated into vegetative cells, which grew and reached a maximum of approximately 104 CFU per gram of cheese at hooping (about 4 h after renneting). After pressing (approximately 16 h after renneting ) the viable cells were reduced to less than 102 CFU per gram. After brining (about 40 h after renneting) B. cereus was not detected in the cheese curd. At this stage the conditions of the cheese, particularly lower moisture content and aw , lower Eh, high salt content, depleted lactose content combined with high acidity may have inhibited the growth of B. cereus. B. cereus did not aect the growth of lactic acid bacteria during cheese manufacturing. Lactic acid bacteria grew from 107 to 109 CFU per gram of curd during cheese manufacturing and stayed fairly constant at about 109 for 6 weeks. Ó 2000 Elsevier Science Ltd. All rights reserved. Keywords: Gouda cheese; B. cereus; Lactic acid bacteria
1. Introduction Gouda is a semi-hard cheese variety made from cowsÕ milk and originated from Holland (Kosikowski, 1978). Milk used for cheese-making should be of a good bacteriological quality to avoid undesirable fermentation and enzymic reactions and without any substances that inhibit or interfere with the growth of the starter bacteria (Chapman & Sharpe, 1981). Milk is an excellent medium for microbial growth. Milk as produced in the alveoli of a healthy cow is free from micro-organisms, but the farm environment makes it impossible to exclude bacteria during milking (Andersson, Ronner, & Granum, 1995). The bacterial load of raw milk ranges from 103 CFU per millilitre when hygiene is good to 107 CFU per millilitre when hygiene is poor. The bacteria consist *
Corresponding author. Tel.: +27-12-420-3209; fax: +27-12-4202839. E-mail address: [email protected] (B.H. Bester). 1 Present Address: Blair Research Institute, P.O. Box CY 573, Causeway, Harare, Zimbabwe. Tel.: 263-4-792747; fax: 263-4-792480.
of psychrotrophs, small numbers of lactic acid bacteria, spore forming Gram-positive rods, coryneform bacteria, micrococci and coliforms. Psychrotrophs will continue to multiply during transport and storage of milk. Pasteurisation (72°C for 15 s) of the milk prior to Gouda cheese manufacture will reduce the microbial load, but the thermoduric micro-organisms with psychrotrophic properties survive (Andersson et al., 1995) and of importance to this research is the species Bacillus cereus. Occurrence of B. cereus in milk has been reported since 1916, and this bacterium is a common contaminant of raw milk produced on some dairy farms (Ahmed, Moustafa, & March, 1983). It can also be found in large numbers in dairy products (Ahmed et al., 1983). In a survey by Wong, Chen and Chen (1988) on dairy products, 52% of ice creams, 35% of soft ice creams, 29% of milk powders, 17% of fermented milks, and 2% of pasteurised milks and fruit ¯avoured milks were found to be contaminated with B. cereus. The concern about its presence in dairy products is that it may cause certain defects in the products or produce toxins and thus cause food poisoning.
0956-7135/00/$ - see front matter Ó 2000 Elsevier Science Ltd. All rights reserved. PII: S 0 9 5 6 - 7 1 3 5 ( 0 0 ) 0 0 0 1 6 - 5
32
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
At counts above 2 105 /ml B. cereus causes o ¯avours such as unclean, fruity, bitter, putrid, rancid and yeasty (Meer, Baker, Bodyfelt, & Griths, 1991). When growth continues, the product shows the defect sweet curdling in homogenised low-pasteurised milk and bitty cream in low-temperature pasteurised milk. It has been estimated that more than 25% of the shelf-life problems encountered with pasteurised milk are due to the proliferation of Bacillus species (Griths, 1992). Although low numbers of B. cereus in foods pose no direct health hazard, if these foods are mishandled, growth of this organism might result in a direct health hazard. Large numbers of B. cereus produce enterotoxin in the food before it is consumed and cause illness (Macrae, Robinson, & Salder, 1993). According to the South African regulations relating to Herbs and Spices (R 1468) of the Foodstus, Cosmetics and Disinfectants Act 54 of 1972, any foodstu containing B. cereus in 20 g of sample is deemed harmful or injurious to human health. The objectives of the project were to determine the survival and growth of B. cereus during manufacture of Gouda type cheese and to determine whether B. cereus would aect the growth of lactic acid bacteria during Gouda cheese manufacturing. Such information will indicate whether Gouda cheese manufactured from milk severely contaminated with B. cereus would pose a health hazard to consumers of such cheese or of food products in which such cheese is used. 2. Materials and methods 2.1. Manufacture of Gouda cheese Gouda type cheese was manufactured in the pilot plant of the Department of Food Science, University of Pretoria, using the method described by Kosikowski (1978). Lysozyme was not used during cheese manufacturing. The milk containing spores for the experimental cheese and milk without spores for the control cheese was pasteurised in a batch pasteuriser at 63°C for 30 min and cooled down to 30°C. The milk was then inoculated with 0.75% of a starter culture CH±N22 (a mixture of Lactoccus lactis and L. lactis sub sp. cremoris). The stock cultures were supplied in freeze-dried form by CHR HansenÕs Laboratories (Denmark). The milk was renneted, using Rennilase type T rennet (Novo Industries, Denmark). After 30±45 min at 30°C, the coagulated milk was cut and after 10 min resting of the cut curd, part of the whey (10% of the volume of the milk) was removed and replaced by the same amount of water. The temperature of the curd±whey mixture was then gradually increased from 30°C to 36°C (cooking temperature) over a period of 30±45 min. On reaching the cooking temperature, whey was drained to 50% of the
original volume of milk (half-whey). The curd and whey mixture was held for 1±1 12 h at the cooking temperature. The whey was then drained, the curd put into moulds and pressed for 3±4 h. The green cheese was brine-salted for 24 h in saturated brine solution, allowed to dry-o for 2 days, vacuum packed in plastic bags and cured for 6 weeks at 12±15°C. 2.2. Preparation of B. cereus spore suspensions Spores of B. cereus (ATCC 10702) were produced on KG agar slants (Refai, 1979). After incubation for 3±7 days at 37°C, the growth was scraped o the agar slants and the spores collected by centrifugation, washed by resuspending the sedimented spores in sterile distilled water and centrifuging again. This was repeated ®ve times and the suspended spores (about 107 spores/ml) stored at 4±7°C. 2.3. Survival and growth of B. cereus during Gouda cheese manufacturing In order to determine the survival and growth of B. cereus during Gouda cheese manufacturing, two batches of cheese were prepared. Batch 1. Experimental cheese with B. cereus spores added: A batch of 100 l of fresh whole milk was inoculated with 10 ml of spore suspension (approximately 107 spores of B. cereus per millilitre), to give a spore concentration of approximately 102 (between 102 and 103 ) CFU per millilitre in the milk before pasteurisation. The milk was thoroughly mixed and pasteurised at 63°C for 30 min. After pasteurisation the milk was used to manufacture Gouda cheese. Batch 2. Control cheese without B. cereus spores added: A second batch of 100 l of fresh whole milk, to which no B. cereus spores were added, was pasteurised at 63°C for 30 min. After pasteurization the milk was used to manufacture Gouda type cheese. 2.4. Sampling Samples were taken at dierent stages during the Gouda cheese manufacturing and analysed for the parameters as shown in Table 1. 2.5. Microbiological analysis 2.5.1. Preparation of dilutions Tenfold serial dilutions of the samples were made by aseptically transferring 25 g of sample into 225 ml of sterile saline peptone water to give a 10ÿ1 dilution (IDF, 1992). Samples were then homogenised for 20 s using a Stomacher 400 laboratory blender (Seward Laboratory UAC, Britain). Further 10-fold dilutions of up to 10ÿ8 were made by transferring 1 ml of successive serial
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
33
Table 1 Sampling plan during Gouda cheese manufacturing showing process steps and stages at which samples were drawn for microbiological analysesa
a
Process step
Time (h)
B. cereus
LA
MRS
pH
At renneting Curd at cutting At 12 whey removal At ®nal whey removal At hooping (moulding) After pressing After brining After 1 week After 2 weeks After 4 weeks After 6 weeks
0 1.05 2.25 3.15 3.86 15.96 40 168 338 672 1008
+ + + + + + + + + + +
ns + + + + + + + + + +
ns + + + + + + + + + +
+ + + + + + + + + + +
+ sample taken; ns sample not taken; LA Lactic agar; MRS de Man-Rogosa-Sharpe agar.
dilutions into universal bottles containing 9 ml of sterile saline peptone water. 2.5.2. Plate counts by the spread-plate technique Plates for the spread-plate technique were prepared by pouring 12±15 ml of the relevant sterile culture media into sterile petri dishes. The plates were dried at 35°C in an incubator. The dried plates were then inoculated with 0.1 ml of the homogenate or dilutions thereof and appropriately labelled. Each inoculum was spread over the surface of the dried plates using a sterile bent glass rod. After incubation the colonies were counted using a scienti®c colony counter (Stuart Scienti®c, UK) and the results were expressed as CFU per gram or CFU per millilitre depending on whether the sample was a solid (cheese) or a liquid (milk). 2.5.3. B. cereus counts B. cereus selective agar (Oxoid CM 617) was used for the enumeration of B. cereus. The media were prepared according to the manufacturerÕs instructions and poured into sterile petri dishes for the spread-plate technique. The inoculated plates were incubated at 30°C for 24 h, the number of colonies were counted and expressed as CFU per gram (or CFU per millilitre) of sample. The plates were also examined for typical B. cereus colonies which were rough in texture, turquoise to peacock blue in colour, ¯attened, irregular and surrounded by a greyish zone of egg yolk precipitate. Typical B. cereus colonies were con®rmed by Gram staining which showed Gram-positive rods containing spores which did not stain (Harrigan & McCance, 1976). The endospores were ellipsoidal at the centre of the cell and about 1.2 lm in size. 2.5.4. Lactic acid bacteria Lactic agar (Merck) and MRS agar (Oxoid CM361) were used for the enumeration of lactic acid bacteria. The media were prepared according to the manufac-
turersÕ instructions and poured into petri dishes for the spread-plate technique. The inoculated plates were incubated at 37°C for 48 h in anaerobic jars containing Anaerocult A (Merck, Darmstadt, Germany). Using a Harrison disc, a representative number of colonies were isolated from plates selected for computing a plate count and these colonies then examined by the Gram stain and by testing for catalase production with hydrogen peroxide (Harrigan & McCance, 1976). Catalase negative colonies that contained Gram-positive cocci or long or short rods, were recorded as lactic acid bacteria. 2.5.5. Statistical analysis Statistical analyses were done using SPSS version 8.0 for Windows (SPSS, Chicago, USA). Bacterial counts were converted to log10 CFU per gram. Means and standard deviations were calculated. The least signi®cant dierence between the means was determined.
3. Results 3.1. Changes in B. cereus, lactic acid bacteria numbers and pH during cheese manufacture. Table 2 shows the changes in B. cereus, lactic acid bacteria numbers and pH in cheese made from milk inoculated with B. cereus spores (experimental cheese) and cheese without B. cereus spores (control cheese). B. cereus spores survived and germinated into vegetative cells which grew from approximately 102 CFU per gram and reached a maximum of 104 CFU per gram at hooping (about 4 h after renneting). With the inoculum level of approximately 102 CFU per millilitre of milk, a maximun level of approximately 104 CFU per gram of curd (log10 4.4) was reached at the stage of hooping of the curd. Thus, B. cereus increased by 2.2 log cycles, whereafter they rapidly decreased to less than 102 CFU
34
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
Table 2 Changes in numbers of B. cereus and lactic acid bacteria during manufacturing of Gouda cheese Time (h)
Experimental cheese (log10 CFU per gram)
Control cheese (log10 CFU per gram)
(MRS)A
(LA)A
BCA
pH
(MRS)A
(LA)A
BCA
pH
0
nsB
ns
ns
< 2:0
7.3a (0.1) 7.6b (0.5) 8.1c (0.1) 8.2b (0.1) 8.8d (0.1) 9.0dc (0.1) 9.1e (0.1) 9.0de (0.0) 9.0de (0.0) 9.0de (0.0)
7.3a (0.1) 7.6a (0.1) 8.1b (0.1) 8.1b (0.0) 8.8 (0.1) 9.0c (0.1) 9.0c (0.1) 9.0c (0.1) 9.0c (0.0) 9.0c (0.0)
6.60 (0.00) 6.43 (0.06) 6.40 (0.00) 6.30 (0.00) 6.15 (0.07) 6.11 (0.01) 5.19 (0.01) 5.17 (0.01) 5.16 (0.01) 5.15 (0.01) 5.15 (0.01)
ns
1.07
2.2a (0.0) 2.4a (0.1) 3.5b (0.5) 3.5b (0.0) 4.4c (0.1) < 2:0
7.4a (0.0) 7.0b (0.1) 7.8b (0.1) 8.1c (0.1) 8.7d (0.1) 9.0c (0.0) 9.1c (0.0) 9.1e (0.0) 9.0e (0.0) 8.9e (0.0)
7.4a (0.1) 7.4a (0.1) 7.9b (0.1) 8.1c (0.1) 8.7d (0.1) 9.0c (0.0) 9.1c (0.0) 9.1e (0.0) 9.0e (0.0) 9.0e (0.0)
< 2:0
6.6 (0.00) 6.40 (0.00) 6.40 (0.00) 6.15 (0.07) 6.11 (0.01) 5.19 (0.01) 5.17 (0.01) 5.16 (0.01) 5.16 (0.01) 5.16 (0.01) 5.16 (0.01)
2.25 3.15 4.26 15.96 40.00 168 336 672 1008
2
< 2:0 < 2:0 < 2:0 < 2:0 < 2:0
< 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0 < 2:0
* Means with dierent superscripts in the same column are signi®cantly dierent from each other (P < 0.05). Numbers in parentheses are standard deviations. A Lactic acid bacteria on MRS and Lactic agar (LA), B. cereus (BC) on Oxoid CM 617 agar. B ns, not sampled.
Table 3 Means of pH, salt content, moisture content and fat content of the experimental and the control cheese at 6 weeks and compositional results of commercial Gouda cheese obtained from Lolkema and Blaauw (1974) Analysis
Experimental cheese
Control cheese
Commercial Gouda cheese
pH Fat (%) Moisture (%) Salt (%) Salt in moisture (%) Fat in dry matter (%)
5.16 31.2 40.2 2.0 4.74 52.22
5.15 30.0 39.8 1.88 4.51 49.83
5.2 29.2 41 1.5±2.2 ) 49.5
per gram after pressing (approximately 16 h after renneting). A general decrease in pH was observed from pH 6.6 to pH 5.15 for the experimental cheese and from pH 6.6 to pH 5.16 for the control cheese. Maximum numbers of B. cereus were reached when the pH was 6.15. B. cereus did not aect the growth of lactic acid bacteria. Lactic acid bacteria increased by 1.7 log cycles in both the experimental and control cheese until 1 week after renneting of the milk and then remained fairly constant throughout the ripening process of 6 weeks. B. cereus was not found in the control cheese and there was no signi®cant dierence P > 0:05 in the numbers of lactic acid bacteria between the experimental and the control cheese.
3.2. Chemical analyses of the cheese at 6 weeks Table 3 shows the pH, salt content, moisture content and fat content of the experimental and the control cheese at 6 weeks. For comparison the average composition of commercial Gouda cheese is also given. 4. Discussion In cheese milk inoculated with approximately 102 B. cereus spores per millilitre of milk in the present study, B. cereus multiplied during cheese-making to a maximum of approximately 104 CFU per gram at hooping,
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
i.e., after about 4 h. Ungerminated spores could have remained but according to Davies and Wilkinson (1973) the spores must germinate, outgrow and multiply to be of any potential danger. After pressing (approximately 16 h after renneting) B. cereus numbers were reduced to less than 100 CFU per gram of cheese curd (the lowest number detectable with the method used). This is less than the number of 105 per millilitre or per gram which has been reported to cause illness (Overcast & Atmaram, 1974). The initial growth of B. cereus was expected because after pasteurisation the milk was free from other competitor micro-organisms and conditions were still quite favourable for their growth. B. cereus grew nearly at the same rate as the lactic acid bacteria during the early stages of lactic acid fermentation until the LAB reached numbers of 108 per gram and the pH dropped to below 6.15. Thereafter the numbers of B. cereus decreased rapidly. Work by Wong and Chen (1988) on germination and growth of B. cereus in non-fat milk with added lactic acid bacteria also showed that B. cereus was not aected by lactic acid bacteria at the beginning of the fermentation process, but was aected strongly with continued fermentation. The reduction of B. cereus at pH lower than 6.15 is also in accordance with the results found by Driessen (1992) during the manufacture of yoghurt. Killing or inactivation of vegetative B. cereus cells at pH below 6.15 may not be attributed to reduction in pH and lactic acid production only but to a number of factors which act synergistically such as substrate competition, changes in oxidation±reduction potential and production of antimicobial agents (Frank, Marth, & Olson, 1978; Wong et al., 1988). According to Wong et al. (1988) antimicrobial agents such as hydrogen peroxide, formate, acetate or lactate enhanced the inhibitory activity of lactic acid bacteria. However, in their work on the survival and growth of B. cereus in mageu, a sour maize beverage, Byaruhanga, Bester and Watson (1999) came to the conclusion that the inhibition of the Bacillus was attributable to the eect of undissociated lactic acid. Lactic acid bacteria ¯ourished throughout the manufacturing and ripening process reaching numbers as high as 109 CFU per gram of cheese. According to Kosikowski (1978) bacterial numbers of 108 to 109 CFU per gram are enough to enhance cheese ripening. Both experimental and control cheese thus contained enough lactic acid bacterial cells to accomplish ripening of the cheese. Analysis of variance showed no signi®cant dierence P > 0:05 in the isolation rate of lactic acid bacteria on the two media, Lactic agar and MRS agar. The overall composition of both the experimental and control cheese compared well with that of a good Gouda cheese (Lolkema & Blaauw, 1974).
35
5. Conclusion B. cereus spores germinated into vegetative cells, grew and multiplied during Gouda cheese manufacturing. Considering the level of spores inoculated and the detection limits of the plating method used, no viable cells of B. cereus were found in the cheese at the end of the manufacturing process. Therefore, it can be concluded that B. cereus spores were able to survive, germinate and grow into vegetative cells during the early stages of cheese manufacturing but the vegetative cells were not able to survive during the ®nal stages of the manufacturing process and ripening of the cheese.
References Ahmed, A. H., Moustafa, K., & Marth, E. H. (1983). Incidence of Bacillus cereus in milk and some milk products. Journal of Food Protection, 46, 126±128. Andersson, A., Ronner, U., & Granum, P. (1995). What problems does the food industry have with sporeforming pathogens, Bacillus cereus and Clostridium perfrigens?. International Journal of Food Microbiology, 28, 145±155. Byaruhanga, Y. B., Bester, B. H., & Watson, T. G. (1999). Growth and survival of Bacillus cereus in mageu, a sour maize beverage. World Journal of Microbiology and Biotechnology, 15, 329±333. Chapman, H. R., & Sharpe, M. E. (1981). Microbiology of cheese. In: R. K. Robinson, Dairy microbiology (pp. 157±243), vol. 2, London: Applied Science Publishers. Davies, F. L., & Wilkinson, G., 1973. Bacillus cereus in milk and dairy products. In: B. C. Hobbs, & J. H. B. Christian, The microbial safety of food. Proceedings of the Eighth International Symposium on Food Microbiology (pp. 57±67). London: Academic Press. Driessen, F. M. (1992). Importance of Bacillus cereus in fermented milks and processed non-fermented dairy foods. Bulletin of the International Dairy Federation, 287, 11±15. Frank, J. F., Marth, E. H., & Olson, N. F. (1978). Behavior of enteropathogenic Escherichia coli during manufacture and ripening of brick cheese. Journal of Food Protection, 41, 111±115. Griths, M. W. (1992). Bacillus cereus in milk. Bulletin of International Dairy Federation, 287, 18. Harrigan, W. F., & McCance, C. (1976). Laboratory methods in food and dairy microbiology. London: Academic Press. IDF (1992). Preparation of samples and dilutions for microbiological examination. International Dairy Federation Standard, 122B, 1±4. Kosikowski, F. (1978). Cheese and fermented milk foods (2nd ed.) (pp. 16±36, 281±282). New York: F. V. Kosikowski and Associates. Lolkema, H., & Blaauw, J. (1974). Kaasbereiding. Apeldoorn, Netherlands: Landelijk Stichting Beroepsopleiding Levensmiddelindustrie. Macrae, R., Robinson, R. K., & Sadler, M. J. (1993). Encyclopaedia of food science, food technology and nutrition. New York: Academic Press. Meer, R. R., Baker, J., Bodyfelt, F. W., & Griths, M. W. (1991). Psychrotrophic Bacillus spp. in ¯uid milk products: A review. Journal of Food Protection, 54, 968±979.
36
G. Rukure, B.H. Bester / Food Control 12 (2001) 31±36
Overcast, W. W., & Atmaram, K. (1974). The role of Bacillus cereus in sweet curdling of ¯uid milk. Journal of Milk Food Technology, 37, 233±236. Refai M. K. (1979). Microbiological analysis. Manual of food quality control, vol. 4, Rome: Food and Agricultural Organization of the United Nations.
Wong, H., & Chen, Y. (1988). Eects of lactic acid bacteria and organic acids on growth and germination of Bacillus cereus. Applied and Environmental Microbiology, 54, 2179±2184. Wong, H. C., Chen, L. Y., & Chen, C. L. (1988). Growth germination and toxigenic activity of Bacillus cereus in milk products. Journal of Food Protection, 51, 707±710.