Reduction of counts of Listeria monocytogenes in cheese by means of high hydrostatic pressure

ARTICLE IN PRESS FOOD MICROBIOLOGY Food Microbiology 24 (2007) 59–66 www.elsevier.com/locate/fm

Reduction of counts of Listeria monocytogenes in cheese by means of high hydrostatic pressure Toma´s Lo´pez-Pedemonte, Artur Roig-Sague´s, Silvia De Lamo, Manuela Herna´ndez-Herrero, Buenaventura Guamis Departament de Cie`ncia Animal i dels Aliments, Centre Especial de Recerca, Planta de Tecnologia dels Aliments, CeRTA, XiT, Facultat de Veterina`ria, Universitat Auto`noma de Barcelona, Room VO-238, Edifici V, Campus UAB, 08193 Bellaterra, Barcelona, Spain Received 3 December 2005; received in revised form 17 March 2006; accepted 17 March 2006 Available online 17 April 2006

Abstract Inactivation of Listeria monocytogenes (strains NCTC 11994 and Scott A) was evaluated in model cheeses submitted to 10 min HHP treatments of 300, 400 or 500 MPa at 5 or 20 1C. Counts were measured immediately after high hydrostatic pressure (HHP) treatment (day 1) and after 2, 15 and 30 days of storage at 8 1C. Both strains behaved significantly different after 400 and 500 MPa, being NCTC 11994 more sensitive. Scarce differences were found among final values at both HHP treatment temperatures. Initial reductions (log cfu/ g) for 400 MPa at 20 1C were 2.970.2 for strain NCTC 11994 and 1.570.2 for Scott A. They reached after 30-day storage 5.370.2 and 4.670.4 log cfu/g for NCTC 11994 and Scott A, respectively. For 500 MPa treatments, day-1 reductions of both strains were around 5-log cfu/g, and counts fell below quantification limit after 30 days. Injured cells (around 0.8-log cfu/g) were mostly observed in 400 MPa treated samples on days 1 and 2. Starter cells suffered higher inactivation and injury. For 20 1C treatments, its final counts (log cfu/g) at 300, 400 and 500 MPa were: 8.570.2, 5.470.3 and 2.570.1, respectively. These figures evidence the HHP potential to improve safety of cheese products. r 2006 Elsevier Ltd. All rights reserved. Keywords: Listeria monocytogenes; Lactococcus; Cheese; High pressure

1. Introduction Listeria monocytogenes is an intracellular pathogen which can cause invasive disease in humans and animals. Approximately 99% of human listeriosis infections appear to be food-borne, though the disease process is complex with multiple routes of infection. Listeriosis occurs relatively infrequently but may cause fatality in 20% of the cases, one of the highest figures for a food disease. Neonates, elderly people and immunocompromised patients are particularly at risk. L. monocytogenes is annually responsible for 28% of food-related deaths in the United States. Its incidence per 100,000 persons is: 0.5 in France (1997) and Sweden (1996–2000), 0.4 in Norway Corresponding author. Tel.: +34 935811460; fax: +34 935811494.

E-mail address: [email protected] (A. Roig-Sague´s). 0740-0020/$ - see front matter r 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.fm.2006.03.008

(1996–2000) and Finland (2000–2002), 0.2–0.25 in England and Wales (1991–2001) and 0.07 in Japan (1996–2002; Lunde´n et al., 2004; McLauchlin et al., 2004; Okutani et al., 2004). This zoonotic food-borne pathogen is especially troublesome for the food industry because of it ubiquitous distribution in nature, its ability to grow at low temperatures, in the presence of high salt concentrations and at relatively acid pH. Outbreaks of listeriosis have been often related to the consumption of milks and dairy products which have long shelf lives at refrigerating temperatures such as: ripened soft cheeses (including blue cheeses), Mexican-style soft cheese, chocolate milk, and butter (Carminati et al., 2004; Lunde´n et al., 2004; Waak et al., 2002). L. monocytogenes is in fact, a major cause of product recalls worldwide (Gray et al., 2004). Its presence in milk-based products is a result of either raw-milk or

ARTICLE IN PRESS 60

T. Lo´pez-Pedemonte et al. / Food Microbiology 24 (2007) 59–66

post-processing contamination (Borucki et al., 2004; Carminati et al., 2004). In most cases, contamination due to environmental hygiene during cheese-making is weak (below 3 cfu/mL) being the most probable concentration 0.1 cfu/mL. However, direct contamination from dairy cattle can be higher than 3-log cfu/mL (Sanaa et al., 2004). In the last years the capability of inactivating microorganisms by high hydrostatic pressure (HHP) has been object of intensive research (O’Reilly et al., 2000; Smelt, 1998). HHP destroys microbial cells by inducing changes in the morphology, wall and cell membrane and by modifying biochemical reactions and genetic mechanisms (Abee and Wouters, 1999; Patterson et al., 1995; Smelt, 1998). Some of the factors affecting HHP inactivation of microorganisms are: temperature, time and amount of pressure applied, presence of antimicrobial substances and the food matrix involved (Patterson et al., 1995; Smelt, 1998; Wuytack et al., 2002). In particular, for L. monocytogenes, exponential-growth cells are significantly less resistant to pressure than stationary-phase cells (Mackey et al., 1995). Moreover significant variation in pressure sensitivity between different strains of L. monocytogenes (more than 3-log cfu/mL) has been reported by several authors (Alpas et al., 2000; Cheftel, 1995; Patterson et al., 1995; Simpson and Gilmour, 1997). The organism has shown more resistance to pressure when treated in UHT milk than in poultry meat or phosphate buffer saline (Patterson et al., 1995; Simpson and Gilmour, 1997), stressing the need to study HHP inactivation into real food matrixes. HHP is known to produce a substantial amount of sublethally injured cells (Wuytack et al., 2002, 2003). The degree of injury can be measured by comparing the growth on nutritive media to the one on selective media or nutritive media supplemented with NaCl, sodium dodecil sulfate, or acid pH (McClements et al., 2001; Patterson et al., 1995). However, for food samples which background microbiota interfere with the counts onto non-selective media, this is not a reliable tool to assess HHP injury. The thin agar layer (TAL) method was proposed by Kang and Fung (1999) for recovery of heat-injured L. monocytogenes and by Chang et al. (2003) for recovery of freeze or acid injured cells. This technique proved to produce significant higher counts of L. monocytogenes than selective media alone. It seemed a good alternative to estimate the amount of sublethally injured cells inside a complex food matrix. In general, the higher the pressure and the temperature applied the higher inactivation we get, especially for treatment temperatures superior to 50 1C. However, not all substrates withstand such severe conditions. In particular, traditional soft curd cheeses cannot be submitted to temperatures above 40 1C. The main objectives of this work were firstly, to study the inactivation of L. monocytogenes in model cheeses after applying mild HHP treatments and the influence of HHP on the starter cells added for cheesemaking. Secondly, to study the evolution of their counts and the presence of sublethal damage over a 30-day storage period at 8 1C.

2. Materials and methods 2.1. Organisms The strains used were L. monocytogenes NCTC 11994 and L. monocytogenes Scott A, provided by Coleccio´n Espan˜ola de Cultivos Tipo (University of Valencia, Valencia, Spain) and Instituto Nacional de Investigacio´n y Tecnologı´ a Agraria y Alimentaria (Spain), respectively. Both have proved baroresistance and belong to serobar 4b, responsible for most human listeriosis. In particular, strain NCTC 11994 has been associated with meningitis intoxication after cheese consumption (Cheftel, 1995; Gray et al., 2004; Lunde´n et al., 2004; Patterson et al., 1995). 2.2. Preparation of inocula of L. monocytogenes Both strains were kept at –20 1C on cryobeads (Nalgenes System 100TM, Microkit Iberica S.L., Madrid, Spain) and revived by placing one bead in Brain Heart Infusion Broth (Oxoid Limited, Basingstoke, UK) supplemented with 6 g/L of yeast extract powder (Oxoid, BHIYE) and incubated at 37 1C for 24 h. Each broth culture was checked by plating onto brain heart infusion agar (Oxoid) containing also 6 g/L of yeast extract powder (BHAYE). One mL of the broth was poured in 9 mL of BHIYE and left 18 to 24 h at 37 1C in order to obtain cells in stationary phase of growth. The suspension was centrifuged 15 min at 3000 rpm. The harvested cells were washed twice and finally resuspended in 9 mL of phosphate buffered saline pH 7.3, 10 mM (Oxoid, PBS). 2.3. Pasteurization of milk Raw milk was collected from a local farm and transported under refrigerated conditions in less than 2 h to the laboratory. It was then pasteurized (30 min at 65 1C), quickly refrigerated and then stored below 4 1C. 2.4. Starter culture preparation A mixture of commercial lyophilized strains of Lactococcus lactis subsp. lactis and Lactococcus lactis subsp. cremoris (Ezal MAO 11, Rhodia Iberia S.A., Madrid, Spain), known as a non-bacteriocin producer, was used as a starter culture for the washed-curd model cheese manufacture. The culture was revived inoculating 0.015 g of mixture in 1000 mL of commercial sterilized skimmed milk and incubated at 30 1C for 24 h. A volume of 50 mL was used to prepare a subculture in 200 mL of sterilized skimmed milk, which was also incubated at 30 1C for 24 h. The final concentration was approximately 9-log cfu/g. 2.5. Manufacture of model cheeses Model washed-curd cheeses were manufactured under controlled microbiological conditions following a modification

ARTICLE IN PRESS T. Lo´pez-Pedemonte et al. / Food Microbiology 24 (2007) 59–66

of the procedure developed by Ur-Rehman et al. (1998), as described by Lo´pez-Pedemonte et al. (2003). Pasteurized milk was brought to 31 1C in a water bath and 2% (v/v) starter culture (prepared as indicated above) was added together with 0.01% (v/v) of a 35% (w/v) calcium chloride solution (Arroyo, Santander, Spain) to improve coagulation. A 0.02% (v/v) liquid rennet extract of calf origin (520 mg/l active chymosin, Arroyo) was used as coagulating agent. Milk was poured in previously autoclaved longnecked centrifuge bottles of 225 mL containing L. monocytogenes inocula (except blanks). Coagulation took place at 31 1C. After 45 min, curds were gently cut with sterile stainless steel tools and heated for 15 min at 37 1C. About 40% of whey was discarded and replaced by sterile tap water. Bottles were centrifuged at 7000 g for 40 min at room temperature. Then they were kept into a water bath at 37 1C until pH reached 5.5. The whey was discarded and 20% sterile brine (200 g of NaCl per litre of tap water) was added directly inside the bottle. After 15 min, the brine was removed and the cheeses were taken out from the bottles and dried with sterile paper. All these conditions had been previously adjusted to obtain model cheeses of approximately 23 g, 55% of dry matter, 1.5% salt-in-moisture content and a final pH around 5, measured 24 h after the manufacture. Inocula level of L. monocytogenes was around 7.5-log cfu/g of cheese. They were vacuum packed in plastic bags (bb4.l, Cryovac Packaging, Sant Boi de Llobregat, Spain) and stored at 8 1C for 30 days. An independent series of experiments with cheeses not inoculated with L. monocytogenes but identically made was carried out to study HHP inactivation of starter bacteria and its evolution during storage at 8 1C. 2.6. pH measurement A potentiometric measurement of the pH was performed with a penetration pHmeter (model 2001; Crison Instruments, Barcelona, Spain). 2.7. HHP treatments Cheese samples were HHP treated 24 h after manufacture using discontinuous HHP equipment (Model S-FL850-9-W, Stansted Fluid Power). The pressure chamber has 37 mm bore internal diameter
245 mm long and has inside the product loading canister. The vessel body and pressure-transmitting fluid inside (water) were held at treatment temperature by circulating water through an integral heat transfer jacket fitted to the outside of the high pressure barrel assembly. The temperature was monitored with a thermocouple positioned in the transmission fluid in the upper part of the chamber. On each run two vacuumpacked cheese samples (approx. 100 cm3) were placed inside the product canister. They were stabilized at the initial treatment temperatures of 2071 or 571 1C and then immediately submitted to a 10 min HHP treatment of 300, 400 or 500 MPa (as measured on the digital and analogue

61

pressure indicators). Time of pressure come up was 75, 90 and 110 s for the three pressures assayed; depressurization time was 30–35 s. Maximum temperatures of the transmitting fluid (adiabatic heating) for the three pressures assayed were as follows: 10.870.3, 14.870.2 and 16.570.5 1C, respectively for 5 1C HHP treatments. For 20 1C HHP treatments they were 26.570.4, 28.570.4 and 31.570.6 1C, respectively.

2.8. Microbiological analysis Microbiological cheese analysis was performed 24 h after manufacture, immediately after HHP treatment (day 1). Cheeses were also analysed after 2, 15 and 30 days of storage at 8 1C, by homogenizing for one and half minute 10 g of sample in 90 mL of BHIYE in an electromechanical blender (BagMixers, Interscience, France). Decimal dilutions in Peptone Water (10 g/L peptone and 5 g/L NaCl, Oxoid) were surface spread onto each media. Counts of L. monocytogenes were assessed onto Listeria selective medium (Oxford formulation) with Listeria selective supplement (Oxoid, LSM media). Counts of L. monocytogenes were also surface spread according to the Thin Agar Layer Method (Kang and Fung, 1999) on LSM covered with two layers of 7 mL of BHAYE (TAL media), to recover both injured and not injured cells. The dishes were incubated at 37 1C at least 48 h. Every analysis day, the remaining first dilutions in BHIYE of every sample, were incubated 18 h at 32 1C in order to determine whether complete inactivation of L. monocytogenes was achieved or not. A loopful of this culture was streaked onto a plate of LSM medium and incubated at 37 1C in order to confirm presence or absence of L. monocytogenes cells. Blank samples (not inoculated with L. monocytogenes) were included to assess the efficacy of pasteurization and manufacturing processes. Inoculated model cheeses not submitted to any HHP treatment were named Controls. Reductions of L. monocytogenes for every analysis day were calculated comparing counts of Control samples (No) with those of HHP treated samples (N). Injured L. monocytogenes cells were estimated subtracting LSM media counts to TAL media counts when both counts differed (Po0.05). In cheeses not inoculated with L. monocytogenes, counts of Lactococcus spp. were followed by spreading their decimal dilutions in peptone water (10 g/L peptone and 5 g/L NaCl) onto M-17 medium (Oxoid) supplemented with lactose bacteriological (Oxoid, 5 g/L). These samples were also spread onto tryptone soya agar (Oxoid, TSA) to determine the counts of total viable cells and to detect those not able to recover on M-17. In addition, spreads on TSA supplemented with 30 g/L of NaCl (TSA+ 3%NaCl) were made to estimate the number of injured cells by comparing counts with those on TSA alone (Patterson et al., 1995; Wuytack et al., 2002). Those samples not inoculated with L. monocytogenes and not pressurized were

ARTICLE IN PRESS T. Lo´pez-Pedemonte et al. / Food Microbiology 24 (2007) 59–66

called Starter Controls. The dishes were incubated at 30 1C, for at least 48 h. 2.9. Statistical analysis Each HHP experiment was performed three separate times with duplicate analysis in each replicate. ANOVA as implemented in SPSS 12.0 for Windows (SPSS Inc, Chicago, USA) was used to test effects of pressure, strain, and storage day on the final logarithm of colony count. Tukey, Duncan and Student–Newman–Keuls post-hoc tests were used as paired comparisons between sample means. Level of significance was set to 0.05. 3. Results and discussion 3.1. Reductions of counts of L. monocytogenes The initial reductions of L. monocytogenes inside cheese can be seen in Fig. 1. Reductions of counts of L. monocytogenes significantly increased with the pressure applied (at both 5 and 20 1C initial HHP treatment temperatures). These initial values are similar to those Szczawinski et al. (1997) obtain with a mixture of three L. monocytogenes strains (isolated from milk) inoculated to slices of Gouda, Edanski and Podlaski ripened hard cheeses. Gallot-Lavalle´ (1998) pressurized goat cheese manufactured from raw milk finding successful reductions of more than 5-log cfu/g of L. monocytogenes F13 (a mutant of strain L028 isolated from a human listeriosis case) even from 350 MPa. On the contrary, Carminati et al. (2004) studied the HHP inactivation of a mixture of seven strains of L. monocytogenes into removed Gorgonzola cheese rinds reporting that pressure treatments of 600 and

700 MPa were necessary to obtain more than 2.4-log cfu/g reductions (strains 23, 28, 29, 30, 34, 51 and 54 were isolated from Gorgonzola cheeses produced in different dairy plants). The diverse strains, phase of growth and food matrixes used may explain the differences among the reductions obtained. The evolution of counts of L. monocytogenes can be followed in Figs. 2 and 3. Counts of Control samples for both strains of L. monocytogenes did not significantly differ during storage. In contrast, counts of all treated samples diminished with storage time. As expected, and in accordance with previous reports (Alpas and Bozoglu, 2003; Cheftel 1995; Simpson and Gilmour 1997; Tay et al. 2003) significant differences were found in the behaviour of both strains. From the moment of pressurization (Fig. 1), strain Scott A proved to be significantly more pressureresistant than strain NCTC 11994. This might have caused the higher counts of strain Scott A over strain NCTC 9 8 7 6 Log cfu/g

62

5 4 3 2 1 0 012

15 Days

30

0 12

15 Days

30

(A) 0 -1

9 a

a

a

8

a

6 Log cfu/g

Log N/No (log cfu/g)

7 -2 -3 b b,c c

d

-4

5 4 3

-5 -6

2 L. monocytogenes Scott A, HHP at 5°C L. monocytogenes Scott A, HHP at 20°C L. monocytogenes NCTC 11994, HHP at 5°C L. monocytogenes NCTC 11994, HHP at 20°C e e,f

1 f

f

0

-7 300

400

500

Pressure (MPa) Figure 1. Reductions of L. monocytogenes NCTC 11994 and Scott A in model cheeses immediately after pressurization at 5 and 20 1C (day 1). All values are expressed as mean 7 confidence interval of three replications in selective media. No stands for counts of control samples and N for those of treated samples. For each pressure, bar means with different letters (a, b, c, d, e, f) differ (Po0.05).

(B)

Figure 2. Evolution of counts of Listeria monocytogenes Scott A (mean 7 confidence interval, three replications) spread onto Listeria selective media (LSM, full lines and filled symbols) and onto thin agar layer media (TAL, dotted lines and empty symbols) during 30-day storage at 8 1C: (A) HHP at 5 1C and (B) HHP at 20 1C. Reference: Controls counts (~/B), 300 MPa counts (’/&), 400 MPa counts (n/m), 500 MPa counts (/J). When interval bars are not visible they fall within plot symbols.

ARTICLE IN PRESS T. Lo´pez-Pedemonte et al. / Food Microbiology 24 (2007) 59–66

9 8 7

Log cfu/g

6 5 4 3 2 1 0 012

15 Days

30

012

15 Days

30

(A) 9 8 7

Log cfu/g

6 5 4 3 2 1 0 (B)

Figure 3. Evolution of counts of Listeria monocytogenes NCTC 11994 (mean 7 confidence interval, three replications) spread onto Listeria selective media (LSM, full lines and filled symbols) and onto thin agar layer media (TAL, dotted lines and empty symbols) during 30-day storage at 8 1C: (A) HHP at 5 1C and (B) HHP at 20 1C. Reference: Controls counts (~/B), 300 MPa counts (’/&), 400 MPa counts (n/m), 500 MPa counts (/J). When interval bars are not visible they fall within plot symbols.

11994 which consistently persisted during the 30-day storage period (Figs. 2 and 3). If we compare the counts of L. monocytogenes through the 30 day storage at 8 1C (i.e. counts in days 2, 15 and 30 relative to day 1 counts), for 300 MPa treated samples significant decreases were observed mostly on day 30 (except for day 15 for NCTC 11994 HHP treated at 5 1C). For 400 MPa treated samples significant decreases in counts were observed on day 15, except for strain Scott A pressurized at 20 1C (probably due to the variability found in the counts on this day). At this pressure a considerable amount of cell inactivation was achieved and the differences in the behaviour of both strains became more evident. Finally, when 500 MPa were applied, minimum detectable counts were already achieved immediately after pressurization (except for Scott A pressurized at 5 1C) and were sustained during the 30 days of storage,

63

reaching reductions of almost 6.0-log cfu/g. No significant differences were observed for strain and HHP temperature treatment, probably because counts often fell below the quantifying limit of the technique. However, total inactivation was not achieved since L. monocytogenes colonies always grew after incubating the homogenized initial dilution 18 h at 32 1C. Regarding the global effect of HHP treatment temperature, despite some significant variations found between samples pressurized at 5 or 20 1C in the evolution of L. monocytogenes counts, the figures reached at the end of the storage period do not justify the choice of one temperature after the other. In Control samples pH decreased from 5.570.2 at the end of manufacture to 4.9870.01 in 24 h and in 30 days reached 5.1070.01. Both strains of L. monocytogenes had the ability of maintaining its counts despite the ripening process that took place. Its acid tolerance and osmotolerance are characteristics that explain its presence in foods with low pH and high levels of NaCl such as cheese. Acid tolerance itself may have been induced by growing at relatively low pH such as 5.5 (at the end of the manufacture) and it could have been enhanced by osmoadaptation and vice versa (Faleiro et al., 2003; Kroll and Patchett 1992). However, when the pressurized samples were stored again at 8 1C, significant decreases in its counts were observed. Presumably, HHP capability of causing membrane damage and perturbation of enzymes in bacteria cells (Simpson and Gilmour, 1997; Tholozan et al., 2000), might have induced a certain loss of tolerance to acid pH or to the presence of organic acids. This effect can be seen in the evolution of counts of 400 MPa treated samples (Figs. 2 and 3). 3.2. Evaluation of injured L. monocytogenes cells The thin agar layer method (Kang and Fung, 1999) was used to evaluate the amount of injured L. monocytogenes cells present inside the cheeses after HHP treatments. Counts on TAL media can also be seen in Figs. 2 and 3. Comparing them with counts on selective media, significant differences (and consequently significant injury) were found only for samples HHP treated at 20 1C. For those samples containing strain NCTC 11994 pressurized at 300 MPa, the number of injured cells reached significance on day 30 of storage (1.170.3-log cfu/g). On the contrary, in cheeses pressurized at 400 MPa, injured cells were observed only on days 1 and 2 (0.870.2 and 0.770.1log cfu/g, respectively). For strain Scott A, a similar injury pattern was observed in samples pressurised at 400 MPa. In samples where significant differences among TAL and selective media were shown immediately after HHP and on day 2, pressurization itself is likely to be the main agent of cell injury. In the cases where the number of sublethally injured cells increased towards the end of storage time, this was probably due to the effects of the ripening of the matrix on HHP sensitized cells. In previous studies carried out with strains Scott A and NCTC 11994 inside PBS,

ARTICLE IN PRESS T. Lo´pez-Pedemonte et al. / Food Microbiology 24 (2007) 59–66

64

10 min HHP treatments of 400 MPa showed injured cell values of 3–4-log cfu/g (McClements et al., 2001; Patterson et al., 1995). They were obtained by subtracting counts on nutritive media supplemented with 30 g/L of NaCl from counts on nutritive media alone. This amount of injured cells was never achieved inside our cheeses using the TAL media. Even when cheese is a food matrix that could provide nutrients and protect bacteria from HHP more than PBS, it must be taken into account that the amount of cell damage measured using the TAL media may be influenced by the migration of components from the selective to the nutritive media, giving values somehow lower than nutritive media alone. Nevertheless, the TAL method seemed, a priori, a reasonable alternative to study HHP cell injury in cheese samples, since it is not possible to prevent starter bacteria from growing on TSA together with L. monocytogenes cells. Further studies employing more accurate and specific methods of quantification of injured cells inside complex food matrixes shall be performed. 3.3. Reductions on counts of Lactococcus spp. In Starter Control cheeses counts observed during the whole storage period were similar for the three spread media (M-17, TSA and TSA+3%NaCl), as expected. Considering that cheeses were made from pasteurized milk, it can be assumed that TSA counts were representative of

Lactococcus spp. cells. Inactivation of starter cells increased significantly with the pressure applied (see Table 1) and with storage time in 400 and 500 MPa treated samples. In this case, the temperature of HHP treatments had a significant effect. These results are concordant with those obtained by Messens et al. (1999) in slices of Gouda cheese and by Wick et al. (2004) in Cheddar cheese. Injury caused by HHP to starter cells was evaluated by subtracting counts on TSA supplemented with 30 g/L of NaCl from those of TSA. As can be deduced from Table 1, immediately after applying HHP at both treatment temperatures, the amount of sublethally injured cells significantly increased with the pressure applied (around 1.6 7 0.4 and 2.6 7 0.5-log cfu/g for 400 and 500 MPa, respectively). The evolution of counts of Lactoccocus spp. can also be seen on Table 1. For those samples HHP treated at 20 1C, 300 MPa produced reductions in starter counts of approximately 1.1-log cfu/g. 300 MPa treatments proved to produce similar reductions and an acceptable ripening process in real scale ewe’s milk cheese using the same starter culture (Juan et al., 2004). However, the reductions caused in L. monocytogenes cells seem too low to improve cheese safety. The fact that on day 15 a significant increase was observed in counts of TSA+3%NaCl equalling the counts of the other media may imply that some cells were able to recover. For cheeses pressurized at 400 MPa, counts on the three media did not significantly vary with storage time,

Table 1 Mean counts (n ¼ 6) of Lactococcus spp. cells 7 standard error (log cfu/g), during the 30-day storage of model cheeses at 8 1C Samples

HHP at 5 1C M17

TSA

Starter Controls Day 1 Day 2 Day 15 Day 30

9.570.1 9.670.1 9.370.1 8.770.2

A a

Treated at 300 MPa Day 1 Day 2 Day 15 Day 30

7.670.1 8.170.2 7.570.3 7.170.2

A a

Treated at 400 MPa Day 1 Day 2 Day 15 Day 30

7.470.3 7.670.1 5.570.1 4.770.1

A a

Treated at 500 MPa Day 1 Day 2 Day 15 Day 30

6.370.1 6.070.1 6.370.1 3.170.1

A a

a,b,c

HHP at 20 1C

A a A a B a

A a A a A a

A a B a C a

A a A a B a

TSA+3%NaCl

9.570.1 9.470.2 9.370.1 8.870.3

A a

8.170.2 8.470.2 8.370.1 7.270.2

A a

7.870.1 7.370.1 6.670.2 5.170.1

A a

7.670.1 7.370.1 6.770.3 3.170.2

A b

A a A a B a

A a A a B a

B a C b D a

A b B a C a

9.270.1 9.670.1 9.170.1 8.170.2

A a

7.670.1 8.070.1 7.370.3 6.770.1

A a

6.270.4 6.470.4 4.770.3 3.470.1

A b

4.470.2 4.670.5 3.870.3 2.670.1

A c

A a A a B a

A a A,B a B a

A b B c C b

A c A b B b

M17

TSA

9.770.1 9.570.1 9.870.1 9.670.1

A a

8.770.3 8.870.2 8.570.2 8.570.2

A a

5.270.2 5.270.3 5.570.3 5.470.3

A a

A a A a A a

A a A a A a

A a A a A a

5.170.5 A a 5.570.2 A a 3.8 7 0.3 B a 2.570.1 C a

TSA+3%NaCl

9.770.1 9.570.1 9.670.1 9.670.1

A a

8.870.2 9.070.2 8.570.2 8.570.2

A a

5.170.1 6.070.6 5.970.3 5.870.2

A a

5.770.2 6.170.3 4.270.2 2.670.1

A a

A a A a A a

A a A a A a

B a B a B a

A b B a C a

9.870.1 9.670.1 9.470.3 9.470.1

A a

7.470.2 7.770.1 8.370.3 8.470.3

A b

3.670.3 3.370.1 3.770.3 3.770.2

A b

3.270.2 3.170.2 2.870.2 2.470.1

A b

A a A a A a

A b B a B a

A c A b A b

A c A,B b B a

Within each HHP treatment temperature (5 and 20 1C) and for the same storage day, column means of counts onto M17, TSA and TSA + 3% NaCl with different lower case superscripts differ (Po0.05). A,B,C,D Within each kind of sample (Starter Controls, treated at 300, 400 and 500 MPa), and for the same media, row means counts of each storage day with different upper case superscripts differ (Po0.05).

ARTICLE IN PRESS T. Lo´pez-Pedemonte et al. / Food Microbiology 24 (2007) 59–66

neither those on TSA+3%NaCl, despite being significantly smaller than those on TSA and M-17 agar. It seemed that after 400 MPa pressure injured cells were not able to recover, probably due to a more intense sublethal damage. For samples treated at 500 MPa, even when initial reductions on cells of Lactococcus were similar to those found at 400 MPa, a significant and sharp decrease of counts on TSA and M-17 agar was seen as storage time increased. Although counts on TSA+3%NaCl did not diminish so fast, it seems reasonable to conclude that death instead of recovery of injured cells prevailed. After 30 days of storage no significant differences among counts were found on the three media. The number of starter cells fell below 3-log cfu/g of cheese, but remained always above that of L. monocytogenes cells. Although the pattern showed by samples pressurized at 5 1C differed, after 30 days of storage at 8 1C the final counts of starter cells in the three media were quite similar to those of cheeses HHP treated at 20 1C. O’Reilly et al. (2002) applied pressures in the range of 100 to 400 MPa to Cheddar cheese containing single strains of Lactococcus lactis (303, 223, 227 and AM2) for 20 min at 25 1C; they stated that these pressures do not enhance starter autolysis, affect chymosin activity or significantly increase levels of proteolysis in cheese. On the other hand, Wick et al. (2004) studied the effect of HHP on Cheddar cheese made with a multiple Lactococcus spp. starter. In part of their work they applied to one month ripened samples pressures of 400 to 800 MPa at 25 1C for 5 min and followed evolution of counts for 160 days. The reductions observed immediately after pressurising cheeses were similar to ours. However, 400 MPa treated samples managed to reach control cell counts after 120 days of storage at 10 1C, while counts of 500 and 800 MPa treated cheeses did not. The starter counts we obtained after 500 MPa with model cheese samples (24 h after manufacture) might question the starter capability of producing an adequate ripening process. Malone et al. (2003) studied HHP effects on proteolytic and glycolytic enzymes involved in cheese manufacturing. According to them, many enzymes (including chymosine) reduce their activity at 500 MPa, while other enzymes activity remains unchanged or even increased. In short, in terms of pressure, 400 MPa seems the most appropriate choice in order to achieve significant reductions of pathogen counts while preserving starter cells from excessive depletion. The scarce differences in the behaviour of L. monocytogenes for both HHP temperatures assayed would allow us to select the temperature in the 5–20 1C range that best suits each cheese variety. However, at 400 MPa a difference of approximately 2-log cfu/g between cheese counts of Lactococcus cells pressurized at 5 and 20 1C was obtained. This fact leads us to recommend the use of refrigerating HHP treatment temperatures if the starter variety behaves like the one chosen in our work after HHP treatment.

65

Concerning raw milk cheeses production, several factors need to be taken into account. Their relative high moisture content and increase of pH during ripening enhances survival and growth of L. monocytogenes; in fact, concerns about sufficiency of curing as a single control pathogen step have been heightened (Altekruse et al., 1998). Soft curd cheeses made from raw milk present L. monocytogenes levels from 0 to 2-log cfu/g in 75% of the cases, but may reach up to 6-log cfu/g (Brisabois et al., 1997; Lunde´n et al., 2004). The infective dose is hard to define but acute listeriosis is not likely to occur (or was rarely related) with foods containing less than 100 cfu/g (Brisabois et al., 1997; Chen et al., 2003). In this study it was shown that the use of moderate HHP treatments at mild temperatures coupled with a reasonable ripening time at refrigerating temperatures, can substantially improve cheese safety regarding L. monocytogenes. As stated in Chen and Hoover (2003), foods containing low levels of L. monocytogenes pose very little health risk and assuring the elimination of the higher concentrations can reduce the number of predicted listeriosis cases by more than 99%. Acknowledgements The authors acknowledge the financial support received from the research project CAL-00-005-C2-1 (Instituto Nacional de Investigacio´n y Tecnologı´ a Agraria y Alimentaria) and the grant given to Toma´s Lo´pez-Pedemonte by the Age`ncia de Gestio´ d’Ajuts Universitaris i de Recerca de la Generalitat de Catalunya. We also thank the Coleccio´n Espan˜ola de Cultivos Tipo and Instituto Nacional de Investigacio´n y Tecnologı´ a Agraria y Alimentaria for providing the strains. References Abee, T., Wouters, J.A., 1999. Microbial stress response in minimal processing. Int. J. Food Microbiol. 50, 65–91. Alpas, H., Bozoglu, F., 2003. Efficiency of high pressure treatment for destruction of Listeria monocytogenes in fruit juices. FEMS Immunol. Med. Microbiol. 35, 269–273. Alpas, H., Kalchayanand, N., Bozoglu, F., Ray, B., 2000. Interactions of high hydrostatic pressure, pressurisation temperature and pH on death injury of pressure resistant and pressure-sensitive strains of foodborne pathogens. Int. J. Food Microbiol. 60, 33–42. Altekruse, S.F., Timbo, B.B., Mowbray, J.C., Bean, N.H., Potter, M.E., 1998. Cheese-associated outbreaks of human illness in the United States, 1973 to 1992: sanitary manufacturing practices protect consumers. J. Food Prot. 61, 1405–1407. Borucki, M., Reynolds, J., Gay, C.G., McElwain, K.L., Kim, S.H., Knowles, D.P., Hu, J., 2004. Dairy farm reservoir of Listeria monocytogenes sporadic and epidemic strains. J. Food Prot. 67 (11), 2496–2499. Brisabois, A., Lafarge, V., Brouillaud, A., De Buyser, M.L., Collette, C., Garin-Bastuji, B., Thorel, M.-F., 1997. Les germes pathogens dans le lait et les produits laitiers: situation en France et en Europe. Rev. Sci. Tech. Off. Int. Epiz. 16 (1), 452–471. Carminati, D., Gatti, M., Bonvini, B., Neviani, E., Mucchetti, G., 2004. High-pressure processing of Gorgonzola cheese: influence on Listeria monocytogenes inactivation and on sensory characteristics. J. Food Prot. 67 (8), 1671–1675.

ARTICLE IN PRESS 66

T. Lo´pez-Pedemonte et al. / Food Microbiology 24 (2007) 59–66

Chang, V.P., Mills, E.W., Cutter, C.N., 2003. Comparison of recovery methods for freeze-injured Listeria monocytogenes, Salmonella typhimurim and Campylobacter coli in cell suspensions and associated with pork surfaces. J. Food Prot. 66, 798–803. Cheftel, J.C., 1995. Review: high-pressure, microbial inactivation and food preservation. Food Sci. Technol. Int. 1, 75–90. Chen, H., Hoover, D.G., 2003. Modeling the combined effect of high hydrostatic pressure and mild heat on the inactivation kinetics of Listeria monocytogenes Scott A in whole milk. Innovative Food Sci. Emerging. Technol. 4, 25–34. Chen, Y., Ross, W.H., Scott, V.N., Gombas, D.E., 2003. Listeria monocytogenes: low levels equal low risk. J. Food Prot. 66, 570–577. Faleiro, M.L., Andrew, P.W., Power, D., 2003. Stress response of Listeria monocytogenes isolated from cheese and other foods. Int. J. Food Microbiol. 84, 207–216. Gallot-Lavalle´, T., 1998. Efficacite´ du traitement par les hautes pressions sur la destruction de Listeria monocytogenes dans les fromages de che`vre au lait cru. Sci. Aliments 18, 647–655. Gray, M.J., Zadoks, R.N., Fortes, E.D., Dogan, B., Cai, S., Chen, Y., Scott, V.N., Gombas, D.E., Boor, K.J., Wiedmann, M., 2004. Listeria monocytogenes isolates from foods and humans form distinct but overlapping populations. Appl. Environ. Microbiol. 70 (10), 5833–5841. Juan, B., Ferragut, V., Guamis, B., Buffa, M., Trujillo, A., 2004. Proteolysis of a high-pressure treated ewe’s milk cheese. Milchwissenschaft 59, 616–619. Kang, D.-H., Fung, D.Y.C., 1999. Thin agar layer method for recovery of heat-injured Listeria monocytogenes. J. Food Prot. 62, 1346–1349. Kroll, R.G., Patchett, R.A., 1992. Induced acid tolerance in Listeria monocytogenes. Lett. Appl. Microbiol. 14, 224–227. Lo´pez-Pedemonte, T., Roig-Sague´s, A., Capellas, M., Trujillo, A., Guamis, B., 2003. Inactivation of spores of Bacillus cereus in cheese by high hydrostatic pressure with the addition of nisin or lysozyme. J. Dairy Sci. 86, 3075–3081. Lunde´n, J., Tolvanen, R., Korkeala, H., 2004. Human listeriosis outbreaks linked to dairy products in Europe. J. Dairy Sci. 87 (E. suppl.), E6–E11. Mackey, B.M., Forestie`re, K., Isaacs, N., 1995. Factors affecting the resistance of Listeria monocytogenes to high hydrostatic pressure. Food Biotechnol. 9 (1&2), 1–11. Malone, A.S., Wick, C., Shellhammer, T.H., Courtney, P.D., 2003. High pressure effects on proteolytic and glycolytic enzymes involved in cheese manufacturing. J. Dairy Sci. 86, 1139–1146. McClements, J.M.J., Patterson, M.F., Linton, M., 2001. The effect of growth stage and growth temperature on high hydrostatic pressure inactivation of some psychrotrophic bacteria in milk. J. Food Prot. 64 (4), 514–522. McLauchlin, J., Mitchell, R.T., Smerdon, W.J., Jewell, K., 2004. Listeria monocytogenes and listeriosis: a review of the hazard characterisation for use in microbiological risk assessment of foods. Int. J. Food Microbiol. 92, 15–33. Messens, W., Estepar-Garcia, J., Dewettinck, C., Huyghebaert, A., 1999. Proteolysis of high-pressure-treated Gouda cheese. Int. Dairy J. 9, 775–782.

Okutani, A., Okada, Y., Yamamoto, S., Igimi, S., 2004. Nationwide survey of human Listeria monocytogenes infection in Japan. Epidemiol. Infect. 132, 769–772. O’Reilly, C.E., O’Connor, P.M., Kelly, A.L., Beresford, T.P., Murphy, P.M., 2000. Use of hydrostatic pressure for inactivation of microbial contaminants in cheese. Appl. Environ. Microbiol. 66 (11), 4890–4896. O’Reilly, C.E., O’Connor, P.M., Murphy, P.M., Kelly, A.L., Beresford, T., 2002. Effects of high-pressure treatment on viability and autolysis of starter bacteria and proteolysis in Cheddar cheese. Int. Dairy J. 12, 915–922. Patterson, M.F., Quinn, M., Simpson, R., Gilmour, A., 1995. The sensitivity of vegetative pathogens to high hydrostatic pressure treatments in phosphate-buffered saline and foods. J. Food Prot. 58, 524–529. Sanaa, M., Coroller, L., Cerf, O., 2004. Risk assessment of listeriosis linked to the consumption of two soft cheeses made from raw milk: Camembert and Brie of Meaux. Risk Anal. 24 (2), 389–399. Simpson, R.K., Gilmour, A., 1997. The resistance of Listeria monocytogenes to high hydrostatic pressure in foods. Food Microbiol 14, 567–573. Smelt, J.P.P.M., 1998. Recent advances in the microbiology of high pressure processing. Trends Food Sci. Technol. 9, 152–158. Szczawinski, J., Szczawinska, M., Stanczak, B., Fonberg-Broczek, M., Szczepeck, J., 1997. Effects of high pressure on survival of Listeria monocytogenes in ripened, sliced cheeses at ambient temperature. In: Heremans, K. (Ed.), High Pressure Research in the Biosciences and Biotechnology. University Press, Leuven, Belgium, pp. 295–298. Tay, A., Shellhammer, T.H., Yousef, A.E., Chism, G.W., 2003. Pressure death and tailing behaviour of Listeria monocytogenes strains having different barotolerances. J. Food Prot. 66, 2057–2061. Tholozan, J.L., Ritz, M., Jugiau, F., Federighi, M., Tissier, J.P., 2000. Physiological effects of high hydrostatic pressure treatments on Listeria monocytogenes and Salmonella typhimurium. J. Appl. Microbiol. 88, 202–215. Ur-Rehman, S., Fox, P.F., McSweeney, P.L.H., Fox, P.M., 1998. A protocol for the manufacture of miniature cheeses. Lait 78, 607–620. Waak, E., Tham, W., Danielsson-Tham, M., 2002. Prevalence of Listeria monocytogenes strains isolated from raw whole milk in farm bulk tanks and in dairy plant receiving tanks. Appl. Environ. Microbiol. 68, 3366–3370. Wick, Ch., Nienabert, U., Anggraeni, O., Shellhammer, T.H., Courtney, P.D., 2004. Texture, proteolysis and viable lactic acid bacteria in commercial Cheddar cheeses treated with high pressure. J. Dairy Res. 71, 107–115. Wuytack, E.Y., Diels, A.M.J., Michiels, Ch.W., 2002. Bacterial inactivation by high-pressure homogenisation and high hydrostatic pressure. Int. J. Food Microbiol. 77, 205–212. Wuytack, E.Y., Thi Puong, L.D., Aertsen, A., Reyns, K.M.F., Marquenie, D., de Ketelaere, B., Masschalk, B., Van Opstal, I., Diels, A.M.J., Michiels, C.W., 2003. Comparison of sublethal injury induced in Salmonella enterica serovar typhimurium by heat and by different non thermal treatments. J. Food Prot. 66 (1), 31–37.