- Email: [email protected]
Evolution of Listeria monocytogenes populations during the ripening of naturally contaminated raw ewe’s milk cheese
Food Control 18 (2007) 1258–1262 www.elsevier.com/locate/foodcont
Evolution of Listeria monocytogenes populations during the ripening of naturally contaminated raw ewe’s milk cheese Noe´mia Gameiro a, Suzana Ferreira-Dias b, Mass Ferreira a, Luisa Brito b
a,*
a Laborato´rio de Microbiologia, DBEB, Instituto Superior de Agronomia, Tapada da Ajuda 1349-017 Lisboa, Portugal Centro de Estudos Agro-Alimentares, DAIAT, Instituto Superior de Agronomia, Tapada da Ajuda 1349-017 Lisboa, Portugal
Received 25 April 2006; received in revised form 1 August 2006; accepted 7 August 2006
Abstract The aim of this work was to study, in loco, the evolution of Listeria monocytogenes populations, during ripening (7, 42, 60 and 120 days) of naturally contaminated raw ewe’s milk cheese. Two batches of cheese consisting of 20 or 16 cheeses were obtained from two farmstead cheesemakers, respectively. A significant increase in numbers of L. monocytogenes was observed for both batches, from 7 to 42 days of ripening. These results suggest that this type of cheese has potential to support the survival of L. monocytogenes, while stressing the importance of cheese contamination in the dairies by resident strains. 2006 Elsevier Ltd. All rights reserved. Keywords: Listeria monocytogenes; Cheese ripening; Environmental contamination
1. Introduction Listeria monocytogenes is a foodborne bacterium recognised as pathogenic for both humans and animals. Because of its versatility, Listeria is able to persist in the food industry environment, for several years (Unnerstad et al., 1996), probably in a biofilm state. Post-processing contamination of food with L. monocytogenes is a critical problem of public health. Several outbreaks of listeriosis were linked with the consumption of minimally processed and ready to eat (RTE) foods (Aureli et al., 2000; Brett, Short, & McLauchlin, 1998). These reports highlighted the importance of cross-contamination of processed foods from environmental sources. Cheese is one RTE type of food that has been associated with foodborne listeriosis (Donnelly, 2001). Listeria is widely disseminated in the rural environment and consequently cheese may be contaminated at any stage from farm to table. Presently in Portugal, there is no sur*
Corresponding author. Tel.: +351 21 365 3240/351 21 365 3435; fax: + 351 21 365 3238. E-mail address: [email protected] (L. Brito). 0956-7135/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodcont.2006.08.002
veillance for L. monocytogenes infections and consequently, there is no reported human listeriosis associated with food consumption. The type of cheese analysed in this study is a traditional Portuguese cheese made from raw ewe’s milk coagulated, at 28–30 C, with added salt (NaCl) and plant rennet (cardoon flower), without deliberate addition of any starter culture. The curd is manually worked, molded and pressed at room temperature (20 C) and finally is bandaged with a strap of clean cotton cloth. Attached to the side of each cheese is a casein label (Passport casein marks, DSM Food Specialties, Delft, Netherland) containing a number that identifies the daily production. The cheeses are taken to the first maturation room (6–10 C) where ripening is carried out on wooden shelves for 15–21 days. At the end of this first stage of ripening, the cheeses are moved to the second maturation room (10–14 C) where they are kept also on wooden shelves until the end (30–120 days). From the first day of production, cheeses are turned upside down daily. Usually after one week a viscous smear spontaneously appears on the surface of the cheese. Cheeses are then washed with water, generally with brushes, and bandaged again with
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
new clean cotton straps. This procedure is performed, on average, four and two times during the first and the second stage, respectively (Rodrigues et al., 2000). As protection against transmission of infectious diseases, a minimum age of at least 60 days, at not less than 35 F (1.7 C), is required in the USA since 1949, for cheese made from raw milk (Anonymous, 1950). For this type of cheese, the Portuguese legal directive stands that it cannot be commercialized before 30 days of ripening, although it is generally put on the market with a maturation period of approximately 42 days (semi-soft cheese with 61–69% moisture, on a fat free base, and 45–60% fat, reported to dry matter). Sometimes the ripening period is extended to get a trade mark of aged cheese (at least 120 days of maturation). In this case the resulting semi-hard to hard cheese has on the same basis a moisture content of 49–56% and more than 60% fat (Anonymous, 2002). Previous studies of the pH variation of this type of cheese during ripening showed that the pH of raw milk (6.69 ± .16) in the coagulation vat is statistically equal to the pH of fresh cheese (6.62 ± .23). Then the pH significantly decreases within the seven days thereafter (to 4.90 ± .51) and the following weeks are characterized by a stabilization of the pH (4.78 ± .46 after 15 days, and 4.84 ± .69 after 30 days (Macedo, Malcata, & Oliveira, 1993). Costa, Barr, Margalho, Nabais, & Pereira (1996) also reported that in this type of cheese, the pH value stabilizes after 18 days of cheese maturation. In this study we aimed to investigate the in loco evolution of the numbers of L. monocytogenes in naturally contaminated raw ewe’s milk cheese, during 120 days of ripening. 2. Materials and methods 2.1. Dairy farmhouses background Two farmhouses (A and B) with their own flocks and dairies were selected for this research. Farmhouse A had a flock with 60 sheep and the flock from farmhouse B had 240 sheep. Milking was done mechanically in farmhouse A, and manually in farmhouse B. At each dairy one batch of cheese consisted of cheeses produced daily with the milk obtained from the evening milking of the day before, mixed with the milk from the morning milking of the day of cheese production. The two dairies were selected from a group previously surveyed by our team, which showed recognised problems regarding this pathogenic bacterium. There were no incidents of listeriosis in the flocks, to our knowledge, during the surveyed periods. The bulk milk that was in dairy A the source of the analysed batch of cheeses was negative for the presence of L. monocytogenes. In dairy B, the bulk milk from which the cheeses were made was positive for the presence of L. monocytogenes, though the level of the pathogen in the milk was not ascertained. The existence of L. monocytogenes in cheeses from both dairies led to the interdiction of
1259
their trade and to their quarantine for the purpose of this study. The maturation of the reported batches of cheese proceeded with cleaning and sanitation procedures followed as usual in these dairies. 2.2. Sampling Two batches of cheese, one from each farmhouse (A and B), were analysed in this study. The batch consisted in farmhouse A of 20 cheeses weighing from 500 g to 750 g, and in farmhouse B the batch was formed by 16 cheeses weighting about 1000 g each. In each batch four sampling times within the ripening period, were considered (7, 42, 60 and 120 days). Five or four cheeses, for dairy A and B, respectively, were considered at each time and each cheese was taken as a sample. The cheeses were transported to the laboratory under refrigerated conditions (4–8 C) and analysed in less than 24 h. An aliquot of 25 g of cheese was removed by cutting radially and vertically three nearly equidistant wedges, which included approximately equal amounts of material from the inner and outer parts of the cheese. The remaining amount of cheese was destroyed after removing the cheese aliquots for analysis. 2.3. Enumeration and detection of L. monocytogenes At each ripening time enumeration of L. monocytogenes was performed basically according to EN ISO/11290-2 (Anonymous, 2004) as follows: from each cheese the 25 g aliquot was blended for 2 min, in a Stomacher 400 (IUL, Barcelona), with 225 ml of LEB (Listeria Enrichment Broth, Oxoid). Portions of 0.2 ml of the homogenised or of the appropriately diluted samples were surface plated on each of five replicate plates of ALOA (Ottaviani & Agosti, 1997) (AES Laboratoire, Bruz, France) and PALCAM agar (Merck, Darmstadt, Germany), respectively. Plates were incubated aerobically at 37 C for 24–48 h. Colonies presumed to be Listeria spp. were further confirmed as described in Leite et al. (2006). The counts of L. monocytogenes on each out of five plates of two different culture media, respectively, were summed (five replicate plates · 0.2 ml of a 1:10 dilution). In order to continue with the enrichment steps the sample in LEB was homogenised with the suggested supplements (SR141, Oxoid, Hampshire, UK) and detection of L. monocytogenes was performed, according to the vertical standard IDF143A:1995 (Anonymous, 1995). Two Listeria reference strains obtained from CECT (Collecio´n Espan˜ola de Cultivos Tipo) were used in this study as positive and negative controls, respectively: L. monocytogenes CECT4032 (=NCTC11994) (serovar 4b) and Listeria innocua CECT910T (=NCTC11288) (serovar 6a). 2.4. Data analysis ANOVA of the bacterial counts were carried out using LSD post-hoc multiple comparison tests using the program
1260
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
at 42 (A1, B1) and 120 (A2, B2) days of cheese maturation, are plotted in Fig. 1. Significant differences were found (P 6 0.01) between cheeses with the same ripening time for both batches. Similar results were obtained for the other two ripening times and also for ALOA counting data.
StatisticaTM, version 5, from Statsoft, Tulsa, OK 74104, USA. 3. Results 3.1. Enumeration and detection of L. monocytogenes Table 1 summarizes the statistics on counting data throughout the ripening time, at dairies A and B. The counts of L. monocytogenes obtained from cheeses of producer B were always higher than the counts obtained from the cheeses produced at dairy A, for both ALOA and PALCAM media. Results of the enrichment were negative for all samples in which L. monocytogenes was not detected by surface plating on ALOA and PALCAM media. For each batch of cheese and for each ripening time, a wide range of variation on the counting data was observed, which was accompanied by large standard deviation values (Table 1). A one-way ANOVA was carried out on the mean values of CFU presented in Table 1. Eight groups were considered per dairy (four ripening times · 2 media). No significant differences were observed for both dairies between L. monocytogenes counts obtained on ALOA and on PALCAM media. However, a significant increase (P 6 0.05) in the mean values of CFU from 7 to 42 days of cheese maturation, for both dairies A and B, was observed. In addition, when ripening was extended to 60 days, at dairy B, a significant decrease in L. monocytogenes counts was detected. Another ANOVA of L. monocytogenes counts on ALOA and PALCAM was carried out, in order to compare data within cheeses of the same batch and ripening time. Each cheese was taken as a group and all the five replicate plates in each group were considered. The results obtained, on PALCAM medium, for dairy A and dairy B,
4. Discussion The evaluation of L. monocytogenes concentration in contaminated cheeses, during ripening, was performed on two batches of cheese from two different producers by using ALOA and PALCAM media. The selectivity of ALOA is obtained by the presence of some agents also present in PALCAM omitting however, acriflavine, and adding nalidix acid and cycloheximide. The advantage of ALOA over PALCAM is the possibility to differentiate between L. monocytogenes and other Listeria spp. Nevertheless, attention has to be paid to some L. ivanovii strains since, like L. monocytogenes, they can also produce blue colonies with an opaque halo surrounding the colonies. This drawback can be overcome by the observation of the extent of haemolysis reaction using horse blood agar. The larger haemolysis zone produced by L. ivanovii, in this medium, allows its differentiation from L. monocytogenes. Concerning the suitability of both media to evaluate the level of cheese contamination with L. monocytogenes, the present results show that no significant differences were obtained between counts on both media. Regarding the evolution of L. monocytogenes populations during ripening of naturally contaminated raw ewe’s milk cheese, at 42 days of maturation (when soft cheese is marketed) cheeses had levels of contamination that were not in accordance with the zero mandatory level of the cheese regulation at the time of these assays (Anonymous, 1996). Moreover, in both batches of cheese, there was an
Table 1 Descriptive statistics on Listeria monocytogenes counting, throughout cheese ripening in two batches of cheese, from dairies A and B, respectively Batch A
Ripening time (days) 7 42 60 120
B
7 42 60 120
Culture medium
Mean (·10 CFU/g)
Minimum (·10 CFU/g)
Maximum (·10 CFU/g)
Standard deviation (·10 CFU/g)
P A P A P A P A
0.4 0 25.0 37.8 16.2 28.6 8.4 5.8
0 0 0 1 0 1 2 2
2 0 74 103 81 130 21 13
0.9 0 29.7 41.1 36.2 56.7 7.6 4.6
P A P A P A P A
113.8 499.0 408.8 720.3 81.8 204.3 248.0 434.8
56 215 60 158 0 0 13 54
150 807 982 2002 158 329 475 835
44.3 307.9 401.4 861.2 67.9 151.8 216.5 380.5
(P)—PALCAM medium; (A)—ALOA medium.
800 700 600 500 400 300 200 100 0
c a
a
a 2
150
a
100 50
a
a
a
0
3 Samples
4
1
5
2
3 Samples
A1
4
5
A2
a
5000
1250
b
750 500
a
a
4000 CFU/ g
1000 CFU/ g
b
200
1
250
1261
250
b CFU/ g
CFU/ g
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
a
3000 2000
b
1000
a
0
b
0 1
2
3 Samples
4
B1
1
2
3 Samples
4
B2
Fig. 1. Mean values of L. monocytogenes counts (CFU/g), on PALCAM medium, of cheeses from dairy A (A) and dairy B (B). A1 and B1, cheese samples with 42 days of ripening; A2 and B2, cheese samples with 120 days of ripening. Equal letters mean no significant differences between samples. Different letters mean that sample values are statistically different at P 6 0.01.
increase in L. monocytogenes counts from 7 to 42 days (Table 1), particularly evident for the batch from dairy A made from Listeria free milk. This fact probably reflects cheese cross-contamination in the dairies, possibly due to a higher number of manipulations, at this stage of ripening. Nevertheless in the batch from dairy B, there was a decline in the numbers of L. monocytogenes CFU from 42 to 60 days of cheese maturation. However, in the batch from dairy A, at the end of the 120 days of ripening, although the cheeses were not in accordance with the zero mandatory level, they were in accordance with the new EC regulation in use since the first of January of 2006 (<100 CFU/g at the point of consumption) (Anonymous, 2005). For other types of cheese, like Swiss and Parmesan, several studies have shown that the aging process, up to more than 60 days of ripening, might inhibit the growth of pathogens (Ahmed & Marth, 1990; Mahmoud, Johnson, & Marth, 1992). Significant differences in counts within the same groups of cheese (sample repetitions from the same batch, and with the same time of maturation), were detected (Fig. 1). This may explain the high values of standard deviation and the wide range of variation observed (Table 1), and most likely reflects the heterogeneity created by cheese washing conditions. Actually cheeses from the same batch may be washed together with other cheeses from different batches, as a bulk in the same sink with stand water. The washing of cheeses, usually with brushes, is a very critical point in the production of this type of cheese. Although
the advice for cheese washing under running water, producers have difficulty to understand the reason for spending so much water, since the government encourages the public to spare water for environmental reasons. Listeria spp. have been frequently found in cheese brushing machines in other studies on dairies (Canillac & Mourey, 1993). In the dairies investigated here the cheese brushes are chemically sanitized, but the sanitizers are not always properly used, and consequently resistant/persistent strains became residents in the dairies (Lourenc¸o & Brito, unpublished data). Actually, permanent contaminations within these farmhouses’ facilities were found in a previous work (Leite et al., 2006). The batch of analysed cheeses in dairy A harboured the same strain of L. monocytogenes from serovar 4b (pulsetype 6). This genotype was systematically collected, during one year, from different batches. A similar situation occurred in dairy B, but the strain harboured was from serovar 1/2c (pulsetype 2). This genotype was recovered in this dairy from other batches of cheese throughout seven months. The results obtained then suggested that the most frequent cause of L. monocytogenes contamination in these dairy farmhouses was the environmental contamination of the milk during milking, and that the contaminant strains persisted throughout the cheese ripening period (Leite et al., 2006). Different adaptative responses, to salt and pH stress, of persistent L. monocytogenes strains, isolated from this type of cheese with 7 and 42 ripening days, respectively, were reported (Ribeiro, Manha, & Brito, 2006). These results suggested the adaptation of
1262
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
L. monocytogenes to these two different cheese matrices environments which accounts for their persistence. The present study finished at the end of the milking season. The dairies were then closed, all the contaminated batches of cheese destroyed, and the facilities and equipments submitted to an intensive program of sanitation and disinfection. In spite of these measures, it was very difficult to completely eliminate the contaminant. Ultimately farmhouse B was able to eradicate the bacteria, after the implementation of a new sheep-housing and a parlour with mechanical milking. The problem was finally solved in farmhouse A after the producer decided to sell his flock, and started to produce cheese from reliable milk purchased from other local flocks. The results obtained in this work support previous investigations in these dairies, and suggest that this type of cheese has potential to support the survival of L. monocytogenes. Before this work was performed there was a conviction among these producers that the prolonged ripening of the cheeses (from soft to hard cheese) could solve L. monocytogenes contamination. The results presented in this study suggested that the extending of the ripening time, depending on the level of contamination, will hardly eradicate the pathogen. Therefore maximum care must be taken to avoid contamination with this organism, in every stage of cheese manufacture. The regular use of good manufacturing practices during both milking and cheesemaking, together with udder heath management, are important contributions to effectively control dairy plant contamination with L. monocytogenes. Acknowledgement This work was funded by the project PO AGRO 8.1-1 Concurso, Projecto 292. Noe´mia Gameiro was recipient of a grant from PRODEP III. References Ahmed, E. Y., & Marth, E. H. (1990). Fate of Listeria monocytogenes during the manufacture and ripening of Parmesan cheese. Journal of Dairy Science, 73, 3351–3356. Anonymous (1950). US Food and Drug Administration, Part 19-Cheeses; Processed Cheeses; Cheese Foods; Cheese Spreads, and Related Foods: Definition and Standards of Identity. [Docket No. FDC-46] Final Rule. Federal Register. August 24, 1950, 5656–5690. Anonymous (1995). IDF International Dairy Federation, 1995. Milk and milk products. Detection of Listeria monocytogenes. Provisional IDF
Standart No. 143A:1995. Brussels, Belgium: International Dairy Federation. Anonymous (1996). Imprensa Nacional. Dia´rio da Repu´blica. I Se´rie-B. No. 45 de 22-02-1996. Portaria No. 56/96, 340–341. Imprensa Nacional, Casa da Moeda, Lisbon, Portugal. Anonymous (2002). Imprensa Nacional. Dia´rio da Repu´blica. II Se´rie. No. 97 de 26-04-2002. Despacho No. 8487/2002 (2ase´rie), 7675–7676. Imprensa Nacional, Casa da Moeda, Lisbon, Portugal. Anonymous (2004). EN ISO 11290-2:1998/FDAM 1:2004. Microbiology of food and animal feeding stuffs—Horizontal method for the detection and enumeration of Listeria monocytogenes—Part 2: Enumeration method. Anonymous (2005). Regulamento (CE) N 2073/2005 da Comissa˜o, de 15 de Novembro de 2005, Jornal Oficial da Comunidade Europeia, L 338/ 1–26. Aureli, P., Fiorucci, G. C., Caroli, D., Marchiaro, G., Novara, O., Leonello, L., et al. (2000). An outbreak of febrile gastroenteritis associated with corn contaminated by Listeria monocytogenes. The New England Journal of Medicine, 342, 1236–1241. Brett, M. S. Y., Short, P., & McLauchlin, P. (1998). A small outbreak of listeriosis associated with smoked mussels. International Journal of Food Microbiology, 43, 223–229. Canillac, N., & Mourey, A. (1993). Sources of contamination by Listeria during the making of semi-soft surface-ripened cheese. Sciences des aliments, 13, 533–544. Costa, P., Barr, C., Margalho, C., Nabais, R., & Pereira, C. (1996). Dinaˆmica fı´sico-quı´mica da maturac¸a˜o de queijo Serra da Estrela. Terra Fe´rtil, 2, 11–19. Donnelly, C. (2001). Factors associated with hygienic control and quality of cheeses prepared from raw milk: a review. Bulletin International Dairy Federation, 369, 16–27. Leite, P., Rodrigues, R., Ferreira, MASS, Ribeiro, G., Jacquet, C., Martin, P., & Brito, L. (2006). Comparative characterization of Listeria monocytogenes isolated from Portuguese farmhouse ewe’s cheese and from humans. International Journal of Food Microbiology, 106, 111–121. Macedo, A. C., Malcata, F. X., & Oliveira, J. C. (1993). The technology, chemistry, and microbiology of Serra cheese: a review. Journal of Dairy Science, 76, 1725–1739. Mahmoud, M. B., Johnson, M. E., & Marth, E. H. (1992). Survival of Listeria monocytogenes during the manufacture and ripening of Swiss cheese. Journal of Dairy Science, 75, 380–386. Ottaviani, F., & Agosti, M. (1997). Esperienza su un agar selettivo e differenziale per Listeria monocytogenes. Industrie Alimentari, 36, 1–3. Ribeiro, M. H., Manha, S., & Brito, L. (2006). The effects of salt and pH stress on the growth rates of persistent strains of Listeria monocytogenes collected from specific ecological niches. Food Research International, 39, 816–822. Rodrigues, R. C., de Almeida, J. C., Pereira, C. D., Gomes, D. S., Madanelo, J. P., Oliveira, M. J., & Fonseca, M. L. (2000). Queijo Serra da Estrela, processos tradicionais e inovac¸o˜es tecnolo´gicas. Coimbra, Direcc¸a˜o Regional de Agricultura da Beira Litoral. Unnerstad, H., Bannerman, E., Bille, J., Danielsson-Tham, M.-L., Waak, E., & Tham, W. (1996). Prolonged contamination of a dairy with Listeria monocytogenes. Netherlands Milk and Dairy Journal, 50, 493–499.
Evolution of Listeria monocytogenes populations during the ripening of naturally contaminated raw ewe’s milk cheese Noe´mia Gameiro a, Suzana Ferreira-Dias b, Mass Ferreira a, Luisa Brito b
a,*
a Laborato´rio de Microbiologia, DBEB, Instituto Superior de Agronomia, Tapada da Ajuda 1349-017 Lisboa, Portugal Centro de Estudos Agro-Alimentares, DAIAT, Instituto Superior de Agronomia, Tapada da Ajuda 1349-017 Lisboa, Portugal
Received 25 April 2006; received in revised form 1 August 2006; accepted 7 August 2006
Abstract The aim of this work was to study, in loco, the evolution of Listeria monocytogenes populations, during ripening (7, 42, 60 and 120 days) of naturally contaminated raw ewe’s milk cheese. Two batches of cheese consisting of 20 or 16 cheeses were obtained from two farmstead cheesemakers, respectively. A significant increase in numbers of L. monocytogenes was observed for both batches, from 7 to 42 days of ripening. These results suggest that this type of cheese has potential to support the survival of L. monocytogenes, while stressing the importance of cheese contamination in the dairies by resident strains. 2006 Elsevier Ltd. All rights reserved. Keywords: Listeria monocytogenes; Cheese ripening; Environmental contamination
1. Introduction Listeria monocytogenes is a foodborne bacterium recognised as pathogenic for both humans and animals. Because of its versatility, Listeria is able to persist in the food industry environment, for several years (Unnerstad et al., 1996), probably in a biofilm state. Post-processing contamination of food with L. monocytogenes is a critical problem of public health. Several outbreaks of listeriosis were linked with the consumption of minimally processed and ready to eat (RTE) foods (Aureli et al., 2000; Brett, Short, & McLauchlin, 1998). These reports highlighted the importance of cross-contamination of processed foods from environmental sources. Cheese is one RTE type of food that has been associated with foodborne listeriosis (Donnelly, 2001). Listeria is widely disseminated in the rural environment and consequently cheese may be contaminated at any stage from farm to table. Presently in Portugal, there is no sur*
Corresponding author. Tel.: +351 21 365 3240/351 21 365 3435; fax: + 351 21 365 3238. E-mail address: [email protected] (L. Brito). 0956-7135/$ - see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.foodcont.2006.08.002
veillance for L. monocytogenes infections and consequently, there is no reported human listeriosis associated with food consumption. The type of cheese analysed in this study is a traditional Portuguese cheese made from raw ewe’s milk coagulated, at 28–30 C, with added salt (NaCl) and plant rennet (cardoon flower), without deliberate addition of any starter culture. The curd is manually worked, molded and pressed at room temperature (20 C) and finally is bandaged with a strap of clean cotton cloth. Attached to the side of each cheese is a casein label (Passport casein marks, DSM Food Specialties, Delft, Netherland) containing a number that identifies the daily production. The cheeses are taken to the first maturation room (6–10 C) where ripening is carried out on wooden shelves for 15–21 days. At the end of this first stage of ripening, the cheeses are moved to the second maturation room (10–14 C) where they are kept also on wooden shelves until the end (30–120 days). From the first day of production, cheeses are turned upside down daily. Usually after one week a viscous smear spontaneously appears on the surface of the cheese. Cheeses are then washed with water, generally with brushes, and bandaged again with
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
new clean cotton straps. This procedure is performed, on average, four and two times during the first and the second stage, respectively (Rodrigues et al., 2000). As protection against transmission of infectious diseases, a minimum age of at least 60 days, at not less than 35 F (1.7 C), is required in the USA since 1949, for cheese made from raw milk (Anonymous, 1950). For this type of cheese, the Portuguese legal directive stands that it cannot be commercialized before 30 days of ripening, although it is generally put on the market with a maturation period of approximately 42 days (semi-soft cheese with 61–69% moisture, on a fat free base, and 45–60% fat, reported to dry matter). Sometimes the ripening period is extended to get a trade mark of aged cheese (at least 120 days of maturation). In this case the resulting semi-hard to hard cheese has on the same basis a moisture content of 49–56% and more than 60% fat (Anonymous, 2002). Previous studies of the pH variation of this type of cheese during ripening showed that the pH of raw milk (6.69 ± .16) in the coagulation vat is statistically equal to the pH of fresh cheese (6.62 ± .23). Then the pH significantly decreases within the seven days thereafter (to 4.90 ± .51) and the following weeks are characterized by a stabilization of the pH (4.78 ± .46 after 15 days, and 4.84 ± .69 after 30 days (Macedo, Malcata, & Oliveira, 1993). Costa, Barr, Margalho, Nabais, & Pereira (1996) also reported that in this type of cheese, the pH value stabilizes after 18 days of cheese maturation. In this study we aimed to investigate the in loco evolution of the numbers of L. monocytogenes in naturally contaminated raw ewe’s milk cheese, during 120 days of ripening. 2. Materials and methods 2.1. Dairy farmhouses background Two farmhouses (A and B) with their own flocks and dairies were selected for this research. Farmhouse A had a flock with 60 sheep and the flock from farmhouse B had 240 sheep. Milking was done mechanically in farmhouse A, and manually in farmhouse B. At each dairy one batch of cheese consisted of cheeses produced daily with the milk obtained from the evening milking of the day before, mixed with the milk from the morning milking of the day of cheese production. The two dairies were selected from a group previously surveyed by our team, which showed recognised problems regarding this pathogenic bacterium. There were no incidents of listeriosis in the flocks, to our knowledge, during the surveyed periods. The bulk milk that was in dairy A the source of the analysed batch of cheeses was negative for the presence of L. monocytogenes. In dairy B, the bulk milk from which the cheeses were made was positive for the presence of L. monocytogenes, though the level of the pathogen in the milk was not ascertained. The existence of L. monocytogenes in cheeses from both dairies led to the interdiction of
1259
their trade and to their quarantine for the purpose of this study. The maturation of the reported batches of cheese proceeded with cleaning and sanitation procedures followed as usual in these dairies. 2.2. Sampling Two batches of cheese, one from each farmhouse (A and B), were analysed in this study. The batch consisted in farmhouse A of 20 cheeses weighing from 500 g to 750 g, and in farmhouse B the batch was formed by 16 cheeses weighting about 1000 g each. In each batch four sampling times within the ripening period, were considered (7, 42, 60 and 120 days). Five or four cheeses, for dairy A and B, respectively, were considered at each time and each cheese was taken as a sample. The cheeses were transported to the laboratory under refrigerated conditions (4–8 C) and analysed in less than 24 h. An aliquot of 25 g of cheese was removed by cutting radially and vertically three nearly equidistant wedges, which included approximately equal amounts of material from the inner and outer parts of the cheese. The remaining amount of cheese was destroyed after removing the cheese aliquots for analysis. 2.3. Enumeration and detection of L. monocytogenes At each ripening time enumeration of L. monocytogenes was performed basically according to EN ISO/11290-2 (Anonymous, 2004) as follows: from each cheese the 25 g aliquot was blended for 2 min, in a Stomacher 400 (IUL, Barcelona), with 225 ml of LEB (Listeria Enrichment Broth, Oxoid). Portions of 0.2 ml of the homogenised or of the appropriately diluted samples were surface plated on each of five replicate plates of ALOA (Ottaviani & Agosti, 1997) (AES Laboratoire, Bruz, France) and PALCAM agar (Merck, Darmstadt, Germany), respectively. Plates were incubated aerobically at 37 C for 24–48 h. Colonies presumed to be Listeria spp. were further confirmed as described in Leite et al. (2006). The counts of L. monocytogenes on each out of five plates of two different culture media, respectively, were summed (five replicate plates · 0.2 ml of a 1:10 dilution). In order to continue with the enrichment steps the sample in LEB was homogenised with the suggested supplements (SR141, Oxoid, Hampshire, UK) and detection of L. monocytogenes was performed, according to the vertical standard IDF143A:1995 (Anonymous, 1995). Two Listeria reference strains obtained from CECT (Collecio´n Espan˜ola de Cultivos Tipo) were used in this study as positive and negative controls, respectively: L. monocytogenes CECT4032 (=NCTC11994) (serovar 4b) and Listeria innocua CECT910T (=NCTC11288) (serovar 6a). 2.4. Data analysis ANOVA of the bacterial counts were carried out using LSD post-hoc multiple comparison tests using the program
1260
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
at 42 (A1, B1) and 120 (A2, B2) days of cheese maturation, are plotted in Fig. 1. Significant differences were found (P 6 0.01) between cheeses with the same ripening time for both batches. Similar results were obtained for the other two ripening times and also for ALOA counting data.
StatisticaTM, version 5, from Statsoft, Tulsa, OK 74104, USA. 3. Results 3.1. Enumeration and detection of L. monocytogenes Table 1 summarizes the statistics on counting data throughout the ripening time, at dairies A and B. The counts of L. monocytogenes obtained from cheeses of producer B were always higher than the counts obtained from the cheeses produced at dairy A, for both ALOA and PALCAM media. Results of the enrichment were negative for all samples in which L. monocytogenes was not detected by surface plating on ALOA and PALCAM media. For each batch of cheese and for each ripening time, a wide range of variation on the counting data was observed, which was accompanied by large standard deviation values (Table 1). A one-way ANOVA was carried out on the mean values of CFU presented in Table 1. Eight groups were considered per dairy (four ripening times · 2 media). No significant differences were observed for both dairies between L. monocytogenes counts obtained on ALOA and on PALCAM media. However, a significant increase (P 6 0.05) in the mean values of CFU from 7 to 42 days of cheese maturation, for both dairies A and B, was observed. In addition, when ripening was extended to 60 days, at dairy B, a significant decrease in L. monocytogenes counts was detected. Another ANOVA of L. monocytogenes counts on ALOA and PALCAM was carried out, in order to compare data within cheeses of the same batch and ripening time. Each cheese was taken as a group and all the five replicate plates in each group were considered. The results obtained, on PALCAM medium, for dairy A and dairy B,
4. Discussion The evaluation of L. monocytogenes concentration in contaminated cheeses, during ripening, was performed on two batches of cheese from two different producers by using ALOA and PALCAM media. The selectivity of ALOA is obtained by the presence of some agents also present in PALCAM omitting however, acriflavine, and adding nalidix acid and cycloheximide. The advantage of ALOA over PALCAM is the possibility to differentiate between L. monocytogenes and other Listeria spp. Nevertheless, attention has to be paid to some L. ivanovii strains since, like L. monocytogenes, they can also produce blue colonies with an opaque halo surrounding the colonies. This drawback can be overcome by the observation of the extent of haemolysis reaction using horse blood agar. The larger haemolysis zone produced by L. ivanovii, in this medium, allows its differentiation from L. monocytogenes. Concerning the suitability of both media to evaluate the level of cheese contamination with L. monocytogenes, the present results show that no significant differences were obtained between counts on both media. Regarding the evolution of L. monocytogenes populations during ripening of naturally contaminated raw ewe’s milk cheese, at 42 days of maturation (when soft cheese is marketed) cheeses had levels of contamination that were not in accordance with the zero mandatory level of the cheese regulation at the time of these assays (Anonymous, 1996). Moreover, in both batches of cheese, there was an
Table 1 Descriptive statistics on Listeria monocytogenes counting, throughout cheese ripening in two batches of cheese, from dairies A and B, respectively Batch A
Ripening time (days) 7 42 60 120
B
7 42 60 120
Culture medium
Mean (·10 CFU/g)
Minimum (·10 CFU/g)
Maximum (·10 CFU/g)
Standard deviation (·10 CFU/g)
P A P A P A P A
0.4 0 25.0 37.8 16.2 28.6 8.4 5.8
0 0 0 1 0 1 2 2
2 0 74 103 81 130 21 13
0.9 0 29.7 41.1 36.2 56.7 7.6 4.6
P A P A P A P A
113.8 499.0 408.8 720.3 81.8 204.3 248.0 434.8
56 215 60 158 0 0 13 54
150 807 982 2002 158 329 475 835
44.3 307.9 401.4 861.2 67.9 151.8 216.5 380.5
(P)—PALCAM medium; (A)—ALOA medium.
800 700 600 500 400 300 200 100 0
c a
a
a 2
150
a
100 50
a
a
a
0
3 Samples
4
1
5
2
3 Samples
A1
4
5
A2
a
5000
1250
b
750 500
a
a
4000 CFU/ g
1000 CFU/ g
b
200
1
250
1261
250
b CFU/ g
CFU/ g
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
a
3000 2000
b
1000
a
0
b
0 1
2
3 Samples
4
B1
1
2
3 Samples
4
B2
Fig. 1. Mean values of L. monocytogenes counts (CFU/g), on PALCAM medium, of cheeses from dairy A (A) and dairy B (B). A1 and B1, cheese samples with 42 days of ripening; A2 and B2, cheese samples with 120 days of ripening. Equal letters mean no significant differences between samples. Different letters mean that sample values are statistically different at P 6 0.01.
increase in L. monocytogenes counts from 7 to 42 days (Table 1), particularly evident for the batch from dairy A made from Listeria free milk. This fact probably reflects cheese cross-contamination in the dairies, possibly due to a higher number of manipulations, at this stage of ripening. Nevertheless in the batch from dairy B, there was a decline in the numbers of L. monocytogenes CFU from 42 to 60 days of cheese maturation. However, in the batch from dairy A, at the end of the 120 days of ripening, although the cheeses were not in accordance with the zero mandatory level, they were in accordance with the new EC regulation in use since the first of January of 2006 (<100 CFU/g at the point of consumption) (Anonymous, 2005). For other types of cheese, like Swiss and Parmesan, several studies have shown that the aging process, up to more than 60 days of ripening, might inhibit the growth of pathogens (Ahmed & Marth, 1990; Mahmoud, Johnson, & Marth, 1992). Significant differences in counts within the same groups of cheese (sample repetitions from the same batch, and with the same time of maturation), were detected (Fig. 1). This may explain the high values of standard deviation and the wide range of variation observed (Table 1), and most likely reflects the heterogeneity created by cheese washing conditions. Actually cheeses from the same batch may be washed together with other cheeses from different batches, as a bulk in the same sink with stand water. The washing of cheeses, usually with brushes, is a very critical point in the production of this type of cheese. Although
the advice for cheese washing under running water, producers have difficulty to understand the reason for spending so much water, since the government encourages the public to spare water for environmental reasons. Listeria spp. have been frequently found in cheese brushing machines in other studies on dairies (Canillac & Mourey, 1993). In the dairies investigated here the cheese brushes are chemically sanitized, but the sanitizers are not always properly used, and consequently resistant/persistent strains became residents in the dairies (Lourenc¸o & Brito, unpublished data). Actually, permanent contaminations within these farmhouses’ facilities were found in a previous work (Leite et al., 2006). The batch of analysed cheeses in dairy A harboured the same strain of L. monocytogenes from serovar 4b (pulsetype 6). This genotype was systematically collected, during one year, from different batches. A similar situation occurred in dairy B, but the strain harboured was from serovar 1/2c (pulsetype 2). This genotype was recovered in this dairy from other batches of cheese throughout seven months. The results obtained then suggested that the most frequent cause of L. monocytogenes contamination in these dairy farmhouses was the environmental contamination of the milk during milking, and that the contaminant strains persisted throughout the cheese ripening period (Leite et al., 2006). Different adaptative responses, to salt and pH stress, of persistent L. monocytogenes strains, isolated from this type of cheese with 7 and 42 ripening days, respectively, were reported (Ribeiro, Manha, & Brito, 2006). These results suggested the adaptation of
1262
N. Gameiro et al. / Food Control 18 (2007) 1258–1262
L. monocytogenes to these two different cheese matrices environments which accounts for their persistence. The present study finished at the end of the milking season. The dairies were then closed, all the contaminated batches of cheese destroyed, and the facilities and equipments submitted to an intensive program of sanitation and disinfection. In spite of these measures, it was very difficult to completely eliminate the contaminant. Ultimately farmhouse B was able to eradicate the bacteria, after the implementation of a new sheep-housing and a parlour with mechanical milking. The problem was finally solved in farmhouse A after the producer decided to sell his flock, and started to produce cheese from reliable milk purchased from other local flocks. The results obtained in this work support previous investigations in these dairies, and suggest that this type of cheese has potential to support the survival of L. monocytogenes. Before this work was performed there was a conviction among these producers that the prolonged ripening of the cheeses (from soft to hard cheese) could solve L. monocytogenes contamination. The results presented in this study suggested that the extending of the ripening time, depending on the level of contamination, will hardly eradicate the pathogen. Therefore maximum care must be taken to avoid contamination with this organism, in every stage of cheese manufacture. The regular use of good manufacturing practices during both milking and cheesemaking, together with udder heath management, are important contributions to effectively control dairy plant contamination with L. monocytogenes. Acknowledgement This work was funded by the project PO AGRO 8.1-1 Concurso, Projecto 292. Noe´mia Gameiro was recipient of a grant from PRODEP III. References Ahmed, E. Y., & Marth, E. H. (1990). Fate of Listeria monocytogenes during the manufacture and ripening of Parmesan cheese. Journal of Dairy Science, 73, 3351–3356. Anonymous (1950). US Food and Drug Administration, Part 19-Cheeses; Processed Cheeses; Cheese Foods; Cheese Spreads, and Related Foods: Definition and Standards of Identity. [Docket No. FDC-46] Final Rule. Federal Register. August 24, 1950, 5656–5690. Anonymous (1995). IDF International Dairy Federation, 1995. Milk and milk products. Detection of Listeria monocytogenes. Provisional IDF
Standart No. 143A:1995. Brussels, Belgium: International Dairy Federation. Anonymous (1996). Imprensa Nacional. Dia´rio da Repu´blica. I Se´rie-B. No. 45 de 22-02-1996. Portaria No. 56/96, 340–341. Imprensa Nacional, Casa da Moeda, Lisbon, Portugal. Anonymous (2002). Imprensa Nacional. Dia´rio da Repu´blica. II Se´rie. No. 97 de 26-04-2002. Despacho No. 8487/2002 (2ase´rie), 7675–7676. Imprensa Nacional, Casa da Moeda, Lisbon, Portugal. Anonymous (2004). EN ISO 11290-2:1998/FDAM 1:2004. Microbiology of food and animal feeding stuffs—Horizontal method for the detection and enumeration of Listeria monocytogenes—Part 2: Enumeration method. Anonymous (2005). Regulamento (CE) N 2073/2005 da Comissa˜o, de 15 de Novembro de 2005, Jornal Oficial da Comunidade Europeia, L 338/ 1–26. Aureli, P., Fiorucci, G. C., Caroli, D., Marchiaro, G., Novara, O., Leonello, L., et al. (2000). An outbreak of febrile gastroenteritis associated with corn contaminated by Listeria monocytogenes. The New England Journal of Medicine, 342, 1236–1241. Brett, M. S. Y., Short, P., & McLauchlin, P. (1998). A small outbreak of listeriosis associated with smoked mussels. International Journal of Food Microbiology, 43, 223–229. Canillac, N., & Mourey, A. (1993). Sources of contamination by Listeria during the making of semi-soft surface-ripened cheese. Sciences des aliments, 13, 533–544. Costa, P., Barr, C., Margalho, C., Nabais, R., & Pereira, C. (1996). Dinaˆmica fı´sico-quı´mica da maturac¸a˜o de queijo Serra da Estrela. Terra Fe´rtil, 2, 11–19. Donnelly, C. (2001). Factors associated with hygienic control and quality of cheeses prepared from raw milk: a review. Bulletin International Dairy Federation, 369, 16–27. Leite, P., Rodrigues, R., Ferreira, MASS, Ribeiro, G., Jacquet, C., Martin, P., & Brito, L. (2006). Comparative characterization of Listeria monocytogenes isolated from Portuguese farmhouse ewe’s cheese and from humans. International Journal of Food Microbiology, 106, 111–121. Macedo, A. C., Malcata, F. X., & Oliveira, J. C. (1993). The technology, chemistry, and microbiology of Serra cheese: a review. Journal of Dairy Science, 76, 1725–1739. Mahmoud, M. B., Johnson, M. E., & Marth, E. H. (1992). Survival of Listeria monocytogenes during the manufacture and ripening of Swiss cheese. Journal of Dairy Science, 75, 380–386. Ottaviani, F., & Agosti, M. (1997). Esperienza su un agar selettivo e differenziale per Listeria monocytogenes. Industrie Alimentari, 36, 1–3. Ribeiro, M. H., Manha, S., & Brito, L. (2006). The effects of salt and pH stress on the growth rates of persistent strains of Listeria monocytogenes collected from specific ecological niches. Food Research International, 39, 816–822. Rodrigues, R. C., de Almeida, J. C., Pereira, C. D., Gomes, D. S., Madanelo, J. P., Oliveira, M. J., & Fonseca, M. L. (2000). Queijo Serra da Estrela, processos tradicionais e inovac¸o˜es tecnolo´gicas. Coimbra, Direcc¸a˜o Regional de Agricultura da Beira Litoral. Unnerstad, H., Bannerman, E., Bille, J., Danielsson-Tham, M.-L., Waak, E., & Tham, W. (1996). Prolonged contamination of a dairy with Listeria monocytogenes. Netherlands Milk and Dairy Journal, 50, 493–499.