AU1165797A - Bacterial decontamination method - Google Patents

Description

BACTERIAL DECONTAMINATION METHOD

This invention relates to a method for reducing the levels of bacteria, in particular food-borne human pathogens and spoilage organisms, which is suitable for use in food processing or elsewhere where hygiene requirements mean that bacterial levels should be controlled. The method may be effective against both gram negative and gram positive bacteria. Kits for carrying out the method are also claimed.

PCT patent application WO 93/00822 (MAFF) discloses a method of destroying bacteria by use of osmotic shock treatment in combination with cold shock and/or exposure to the enzyme lysozyme. Lysozyme is well known to be effective against certain gram positive bacteria but the combined treatment extends its usefulness into the class of gram negative bacteria. However, although effective in vitro the combined treatment is not sufficiently effective against Salmonella when the method is used under practical food processing conditions.

In US Patent 5,069,922 there is disclosed a process for treating poultry carcasses to control Salmonella growth, which comprises treating eviscerated and defeathered poultry with a solution containing an alkali metal orthophosphate, e.g. trisodium orthophosphate (TSP) , or an alkali orthophosphate combined with a minor amount of a basic reagent, e.g. sodium carbonate. TSP has been accepted as safe by the US Food and Drug Association and is an ingredient of many food products, but the TSP process as disclosed in the above referenced US patent, and as used to date, has certain disadvantages. Firstly it requires a high TSP concentration (around 0.4M) and thus a high pH (approximately 12.5 at 0. M) , which may reduce its acceptability for poultry and other food treatments. Secondly, it is ineffective against gram positive spoilage bacteria and has questionable effectiveness against certain gram negative bacteria such as Campylobacter.

We have surprisingly found a method of treating products such as foodstuffs, which may be contaminated with human pathogens and/or spoilage bacteria, which is highly effective in the destruction of gram negative bacteria and which will also reduce the levels of gram positive bacteria. The process is particularly useful in the destruction of pathogenic bacteria such as Salmonella and Campylobacter which hitherto have been comparatively resistant to decontamination processes which may be applied to foodstuffs.

According to the present invention there is provided a method for reducing the levels of gram negative and gram positive bacteria in a sample comprising treating the sample with a 0.0005 to 0.2M solution of a trialkali metal orthophosphate, said treatment being combined with one or more of the following further treatments:

a) subjecting the sample to osmotic shock, b) exposing the sample to an enzyme which breaks down peptidoglycan after said treatment with trialkali metal orthophosphate; and c) exposing the sample to a bacteriocin after said treatment with trialkali metal orthophosphate.

A preferred combination of further treatments comprises step (a) with either step (b) or step (c) .

We have found that a combination treatment as disclosed herein is effective in the destruction of gram positive bacteria, against which trialkali metal orthophosphate on its own is ineffective, and for gram negative bacteria such as Salmonella and Campylobacter. Surprisingly the effect of the combination of the trialkali metal orthophosphate process with one or more of the other processes is synergistic in that not only is the range of bacteria destroyed greater than any of the processes separately but also it works with lower concentrations of alkali metal orthophosphate than utilised previously, indeed ones which would be otherwise non-lethal when used without the secondary stage of treatment. Thus concentrations of from 0.0005 to 0.2M, suitably from 0.001 to 0.2M and preferably from 0.005 to 0.01M of trialkali metal orthophosphate are sufficient. This ability to use lower concentrations of trialkali metal orthophosphate reduces the pH of the treatment solution and therefore makes the treatment more attractive to use in food decontamination applications. The precise minimum concentration of TSP which will be effective in any particular case may vary depending upon the nature of the sample. For example, as illustrated hereinafter, the presence of serum in significant quantities may affect the concentration level of TSP, within the above mentioned range, required. However this can be determined using routine methods in any particular case, if it is felt that a minimum amount of TSP is necessary.

The present invention has a further advantage over the previously disclosed osmotic shock plus lysozyme treatment since it does not depend upon the presence of nutrients to promote killing and treatment processes are not markedly dependent on temperature being in the range 4 to 50°C. In addition, using the present invention, organisms washed off chicken skin surfaces were found to be highly susceptible to killing treatments and the resultant treatment solutions were relatively free of bacteria. This could be of significance in reducing cross contamination between carcasses in poultry processing.

Suitably the osmotic shock comprises a hypo-osmotic shock, in which water is induced to enter a cell and particularly a bacterial cell. Hypo-osmotic shock is preferably preceded by administration of a hyper-osmotic shock (where water is induced to leave the cell) . The osmotic shock may i>e administered using the process described in WO 93/008822, the disclosure of which is incorporated herein by reference. In particular, osmotic shock may be induced in a sample by exposing the sample to a first solution having a water activity (a_w) of 0.997 or less (which induces hyper-osmotic shock) and subsequently exposing the sample to a solution of a_k higher than that of said first solution (which then results in hypo-osmotic shock) .

Advantageously, the first solution has a water activity in the range 0.992 to 0.96, more preferably 0.974 to 0.96. It is suitably applied for a time in the range of 5 seconds to 30 minutes, or more preferably between 30 seconds and 20 minutes, or most preferably in the range of 1 to 10 minutes. The said first solution suitably contains NaCl at a concentration sufficient to provide a water activity of 0.997 or less, for example at a concentration of about 0.8M. The a_w of the second solution is suitably of the order of 0.999. The sample is exposed to this solution for a sufficient period of time to produce the desired hypo-osmotic shock. The time may in fact be very short, of the order of a few seconds. However, long exposure times do not adversely affect the reaction. Typically an exposure time of from 5 seconds to 2 hours may be convenient, more particularly from 10 minutes to 1 hour, for example about 30 minutes.

The second solution may be prepared and applied separately from the first solution, or it may be created by introducing an appropriate diluent such as water, to the first solution.

In a preferred embodiment, the said first solution contains 0.001 to 0.2 M of a tri-alkali metal orthophosphate so that the treatment with this reagent is combined with the osmotic shock treatment.

In addition, the second solution may contain reagents used in steps (b) and (c) above, so as to combine these further treatments with the osmotic shock treatment.

Suitable enzymes which break down peptidoglycan for use in the method of the invention are those which are not harmful to humans, and a particular example is lysozyme. Lysozyme is suitably applied to the sample in the form of an aqueous solution, for example having a concentration of at least lμgml^'1 and preferably at least δμgml¹. Such a lysozyme solution is conveniently provided in the form of a solution of freeze dried egg white which is at a concentration of at least 0.lmgml ^:.

We have found that further treatment with lysozyme is particularly effective. The lysozyme may be applied in a rinse water, following treatment with trialkali metal orthophosphate, at a concentration of at least lμg ml^'1 and preferably at least 5μg ml^'1. Alternatively, lysozyme treatment is combined with osmotic shock treatment by adding lysozyme to the second solution of higher a„.

Suitably bacteriocins are also those which are considered to be fit for human consumption such as nisin and pediocin, and preferably nisin is used in the method of the invention. Suitable nisin treatment solutions contain nisin at a concentration of at least O.lμM, preferably lμM or more. In a preferred embodiment, the sample is rinsed with water after treatment with the trialkali metal orthophosphate and prior to treatment with the solution containing nisin.

A particular advantage of the use of nisin in this combined treatment is that nisin is also particularly effective in killing gram positive organisms, for example, Staphylococcus aureus giving further improved kills over the other combined treatments.

Thus further treatment with nisin is particularly effective where there is contamination by gram positive bacteria.

In the use of the process of the invention, particularly in the treatment of poultry, control of TSP concentrations is desirable in order to prevent the pH of the enzyme or bacteriocin treatment, in particular lysozyme or nisin treatments respectively, being adversely affected. It may be desirable to acidify the lysozyme or nisin solution to optimise the treatment.

Therefore, suitably, treatment solutions of bacteriocin or enzyme solutions are acidified, for example to a pH of approximately 5.0. This may be effected by inclusion of an acid to the solution, preferably an organic acid such as lactic acid. Lactic acid is suitably present at a concentration of at least 0.25mM.

A suitable trialkali metal orthophosphate for use in the method of the invention is trisodium phosphate.

The sample is suitably a portion of a foodstuff, although it may also comprise other consumable products such as pharmaceuticals, cosmetics and toiletries. However, it may also comprise surfaces of machinery, instruments or utensils, such as pipework, or working surfaces, such as food preparation surfaces or 'clean-room' surfaces where the presence of bacteria would be problematic.

Treatment methods used will depend upon the nature of the sample and the nature of the solution being applied. However, treatments may suitably be effected, for example by immersing the sample in a solution, or by washing and particularly spray washing the sample with a treatment solution, or a combination of these. Where treatment requires that the sample is exposed to the treatment solution for an extended period of time, for instance, where a solution with a water activity of less than 0.997 is being applied in order to induce a hyperosmotic shock, immersion may be the most convenient option, but suitably spray wash solutions may be prepared, for example by using electrostatic spraying techniques or by including film- forming chemicals such as surfactants into the solution prior to application.

Thus in a preferred embodiment, a sample is treated with a first solution comprising a trialkali metal orthophosphate at a concentration in the range of from 0.0005 to 0.2M, suitably from 0.001 to 0.2M, said solution having a water activity (a_w) of 0.997 or less, and subsequently treating the sample with a second solution of an enzyme which breaks down peptidoglycan or a bacteriocin, said second solution having an a„ higher than that of said first solution.

Suitably the sample is sequentially immersed in said first and second solutions but alternatives such as immersion in the first solution followed by spray washing in the second solution may be used.

In a preferred process a sample of foodstuff is immersed in a solution of TSP at a concentration in the range 0.0005 to 0.2M, suitably from 0.001 to 0.2M and preferably 0.005 to 0.01M, together with a solution of sodium chloride, preferably at a concentration of about 0.8M, at a temperature in the range 4 to 50°, preferably 37°C and then spray washing with lysozyme solution, which is suitably at a concentration of >10μg ml¹. The lysozyme solution may be provided in the form of freeze dried egg white at a concentration of about O.lmg ml¹. This preferred process is highly effective in the killing of gram negative bacteria on vegetable material such as lettuce.

In an alternative preferred process a sample of foodstuff is immersed in a solution of trisodium orthophosphate at a concentration in the range 0.001 to 0.2M, preferably 0.005 to 0.01M, at a temperature in the range 4 to 50°C, preferably 37°C and then spray washed with nisin at a concentration of greater than lμM.

Although the treatments of the present invention are fairly independent of temperature, it is preferred that the treatment with trialkali metal orthophosphate and the further treatment is carried out at a temperature in the range 4 to 50°C, preferably about 37°C.

Kits for carrying out the above-described method form a further aspect of the invention. These kits may comprise a trialkali metal orthophosphate and one or more of the following components:

a) a reagent which can be used to induce osmotic shock in a cell; b) an enzyme which can break down peptidoglycan; and c) a bacteriocin.

The reagents may be supplied per se with instructions for forming suitable treatment solutions, or they may be supplied as ready made aqueous solutions. The components of the kit will be separated in various containers, for example in multipack containers.

Products treated in accordance with the above-described method and particularly foodstuffs so treated form a further aspect of the invention. The method of the present invention is now illustrated further with reference to the following non-limiting examples.

In the work conducted, the following investigations were carried out:

(i) the effect of combining TSP with osmotic shock and/or lysozyme treatment.

(ii) the effectiveness of TSP and combined osmotic shock/lysozyme treatments on the killing of organisms attached to food surfaces (chicken skin and lettuce)

(iii) the effect of using nisin to supplement or replace lysozyme in killing procedures.

The organisms used were: Escherichia coli , Listeria monocytogenes , Pseudomonas fluorescens, Salmonella enteri tidis, Campylobacter jejuni and Staphylococcus aureus .

Preparation of Bacteria

Campylobacter jejuni (NCTC 11626) , Listeria monocytogenes (NCTC 7973), Pseudomonas fluorescens (NCTC 10038) Salmonella enteri tidis (NCTC 6676) and Staphylococcus aureus (NCTC 8532) were obtained as freeze-dried cultures from the National Collection of Type Cultures, Colindale, UK. Escherichia coli (NCIMB 9485) was obtained from the Division of Life Sciences Collection, King's College.

To provide inocula for experiments, the cryoprotectant glycerol (final concentration 15% v/v) was added to early stationary phase cultures. One ml aliquots of cultures were then dispensed into sterile cryotubes and stored at -70°C.

Except for C. jejuni, bacteria were grown aerobically at 37°C or 30°C (Pseudojnonas only) on Brain Heart Infusion broth (Oxoid) or Nutrient agar plates (Oxoid) . Broth cultures were grown in 250ml flasks containing 25 ml medium, on a shaking incubator. C. jejuni was grown statically in 70 ml tissue culture bottles containing 25ml Brain Heart Infusion broth (Oxoid) enriched with 10% Horse serum (Oxoid) and 0.25% Yeast extract (Oxoid), or on Blood agar plates (7% v/v Horse blood (Oxoid) in Oxoid Blood agar base No 2) . Cultures were incubated at 37°C in an atmosphere of 10%(v/v) CO, and 10% (v/v) 0₂, obtained using a gas jar and gas generating kit (Oxoid BR 60) .

Example 1: Killing of suspended cells. The effective cell kills achieved on gram negative and gram positive organisms subjected to treatments with TSP and lysozyme and/or osmotic shock, at various temperatures and in the presence or absence of serum was tested as follows.

The methodology adopted was based on the disclosure of WO 93/00822 incorporated herein by reference which indicated that (i) early stationary phase cells were the most resistant to osmotic shock/lysozyme treatments; (ii) 0.8 M NaCl allowed a maximal or near-maximal killing effect in this system for all the gram negative organisms tested; (iii) exposure to hyper and hypo-osmotic shocks for 10 and 30 minutes respectively was adequate to allow access of lysozyme to gram negative cells and longer treatment times would be commercially unacceptable;

(iv) 20μgml ^α lysozyme gave close to optimal cell kills in osmotically shocked gram negative cells.

Overnight broth cultures were diluted 1/25 in fresh medium and similarly grown to the early stationary phase (approximately 4h incubation) ; growth was monitored by following culture optical density using an EEL colorimeter at 550 nm.

Stationary phase cultures were diluted 1/100 into appropriate test media, which included: TSP, TSP plus sodium chloride (NaCl) , TSP plus heat inactivated newborn calf serum (up to 50% v/v,-Gibco) , and TSP plus NaCl and serum; concentrations of TSP and NaCl were varied and are given in the results tables.

After incubation in test media for 10 min at 4 or 37°C, cell suspensions were further diluted (1/100) in distilled water or distilled water containing 20μgml ' lysozyme, (Sigma L6876) .

After incubation for a further 30 min, cells were diluted as appropriate in deionised water and plated on nutrient agar.

Plates were incubated for up to 48h at 37°C or 30°C

( Pseudomonas only) and cell kills estimated.

The results are shown in Tables 1 to 8. In the following Tables, annotations used are as follows; ns, no survivors detected; cell kill > 99.7% *pH of initial treatment solutions, i.e. TSP, NaCl, TSP + NaCl, or water.

Table 1. The combined effect of TSP. osmotic shock and lysozvme on the survival of Gram -ve bacteria at 37°C

Organism NaCl (M) TSP(M) pH* % ce l survival

-lysozyme +lysozyme

E. coli none none 7.01 100 87 none 0.002 10.46 100 2.3

0.4 none 6.57 73 11

0.8 none 6.39 33 0.59

1.2 none 6.26 13 ns

0.8 0.002 9.56 ns ns

P . none none 7.12 100 100 fluorescens 0.8 none 6.36 100 12 none 0.001 7.59 100 100 none 0.002 9.07 99 66

0.8 0.001 6.81 ns ns

S. enteri tidis none none 7.01 100 99

0.8 none 6.28 86 41 none 0.001 8.70 91 71 none 0.002 10.69 63 28 none 0.005 10.87 0.2 0.1

0.8 0.001 8.42 ns ns

C. jejuni none none 7.34 100 100

0.8 none 6.62 100 95 none 0.001 8.897 100 76 none 0.002 9.89 100 32 none 0.005 10.67 ns ns

0.8 0.001 7.38 2 2

0.8 0.002 9.42 ns ns

Table 2. The combined effect of TSP. osmotic shock and

Ivsozvme on the survival of Gram -ve bacteria at 4°C

Organism NaCl (M) TSP(M) pH" % cell survival -lysozyme - lysozyme

E. coli none none 6.80 100 100

0.8 none 6.36 31 ns none 0.002 9.33 51 10 none 0.005 11.30 ns ns

0.8 0.002 9.54 ns ns

fluorescens none none 7.04 100 100

0.8 none 6.36 100 3 none 0.001 7.54 65 85 none 0.002 9.26 99 56

0.8 0.001 7.01 ns ns

S. enteri tidis none none 7.15 100 100

0.8 none 6.32 29 17 none 0.001 7.40 80 77 none 0.002 8.42 52 34 none 0.005 11.26 1 ns

0.8 0.001 6.77 9 ns

0.8 0.002 8.80 10 ns

0.8 0.005 10.29 1 ns

C. jejuni none none 7.42 100 100

0.8 none 6.64 100 90 none 0.001 7.84 100 67 none 0.002 8.68 100 89 none 0.005 10.05 4 3

0.8 0.001 7.66 95 33

0.8 0.002 9.00 3 2

0.8 0.005 9.68 ns ns

Table 3. The combined effect of TSP. osmotic shock and Ivsozvme on the survival of Gram +ve bacteria at 37°C

Organism NaCl (M) TSP(M pH fe Cell Survival lysozyme +lysozyme

L. mono¬ none none 6.89 100 1 cytogenes 0.8 none 6.32 100 5 none 0.005 10.81 88 3 none 0.010 11.54 88 8 none 0.050 12.22 4 ns none 0.100 12.50 ns ns

0.8 0.005 10.35 27 ns

0.8 0.010 10.80 9 ns

0.8 0.050 11.87 ns ns

St. aureus none none 6.99 100 100

0.8 none 6.32 64 77 none 0.005 9.92 100 53 none 0.01 11.51 77 78 none 0.05 12.23 ns ns

0.8 0.005 9.66 33 17

0.8 0.01 9.88 ns ns

0.8 0.05 11.65 ns ns

Table 4. The combined effect of TSP. osmotic shock and Ivsozvme on the survival of Gram +ve bacteria at 4°C

Organism NaCl (M) TSP(M) pH' % cell survival lysozyme -(-lysozyme

L. mono¬ none none 7.06 100 20 cytogenes 0.8 none 6.26 100 20 none 0.005 11.38 100 33 none 0.010 11.82 100 3 none 0.050 12.34 100 ns none 0.100 12.46 9 ns

0.8 0.005 10.62 77 ns

0.8 0.010 11.24 53 ns

0.8 0.050 11.93 15 ns

St. aureus none none 7.28 100 100

0.8 none 6.33 100 100 none 0.01 11.51 100 100 none 0.05 12.28 85 65 none 0.10 12.44 83 40

0.8 0.01 11.29 100 91

0.8 0.05 11.85 55 94

0.8 0.10 12.23 13 20 Table 5. The combined effect of TSP. osmotic shock and lysozvme on the survival of Gram -ve bacteria at 37°C in the presence of 50% v/v serum

Organism NaCl (M) TSP(M) pH" % cell survival

-lysozyme +lysozyme

E. coli none none 7.54 100 100

0.8 none 7.00 100 15 none 0.002 8.01 40 3 none 0.005 8.35 56 5 none 0.01 9.27 84 ns

0.8 0.002 7.88 100 ns

0.8 0.005 8.56 1 ns

0.8 0.01 9.20 ns ns

P. none none 7.20 100 97 fluorescens 0.8 none 6.94 46 5 none 0.001 7.45 99 88 none 0.002 7.72 81 72 none 0.005 8.59 74 70

0.8 0.001 7.15 48 2

0.8 0.002 7.43 34 3

0.8 0.005 8.45 ns ns

S. enteri tidis none none 7.22 100 100

0.8 none 6.00 32 4 none 0.002 7.60 100 100 none 0.005 8.35 100 71 none 0.01 9.32 90 82

0.8 0.002 7.35 35 1

0.8 0.005 8.10 2 1

0.8 0.01 8.91 ns ns

C. jejuni none none 7.13 100 75

0.8 none 6.68 75 53 none 0.005 8.59 95 88 none 0.01 9.56 21 23

0.8 0.005 7.88 15 7

0.8 0.01 9.14 ns ns

Table 6. The combined effect of TSP. osmotic shock and lysozvme on the survival of Gram -ve bacteria at 4°C in the presence of 50% v/v serum

Organism NaCl (M) TSP(M) pH* % cell survival -lysozyme -t-lysozyme

E. coli none none 7.42 100 100

0.8 none 7.18 100 20 none 0.002 7.87 100 100 none 0.005 8.92 100 100 none 0.01 9.66 100 94

0.8 0.002 7.18 53 10

0.8 0.005 8.37 25 6

0.8 0.01 9.34 1 ns

P. fluorescens none none 7.18 100 100

0.8 none 6.93 66 14 none 0.001 7.43 100 100 none 0.002 7.74 100 91 none 0.005 8.59 89 88 none 0.01 9.60 76 89

0.8 0.001 7.15 39 7

0.8 0.002 7.36 28 7

0.8 0.005 8.56 ns ns

0.8 0.01 9.08 ns ns

S . enteri tidis none none 7.18 100 82

0.8 none 6.99 100 90 none 0.002 7.50 100 100 none 0.005 8.24 97 90 none 0.01 9.36 100 98

0.8 0.002 7.24 85 61

0.8 0.005 8.08 89 68

0.8 0.01 9.09 59 10

C. jejuni none none 7.21 100 66

0.8 none 6.53 78 68 none 0.005 8.34 71 66 none 0.01 9.48 47 42

0.8 0.005 7.38 37 37

0.8 0.01 9.42 ns ns Table 7. The combined effect of TSP. osmotic shock and lysozvme on the survival of Gram +ve bacteria at 37°C in the presence of 50% serum

Organism NaCl (M) TSP(M pH* % Cell Survival

- lysozyme +lysozyme

L . mono¬ none none 7.12 100 ns cytogenes 0.8 none 6.99 100 2 none 0.010 9.48 98 3 none 0.050 11.44 52 ns none 0.100 12.01 28 ns

0.8 0.01 9.42 90 3

0.8 0.05 11.21 22 ns

0.8 0.10 11.60 ns ns

St . aureus none none 7.26 100 99

0.8 none 7.07 84 85 none 0.010 9.57 92 80 none 0.050 11.50 89 82 none 0.100 11.89 5 7 none 0.150 12.11 ns ns

0.8 0.010 9.14 88 89

0.8 0.050 10.98 92 ns

0.8 0.100 11.48 1 2

0.8 0.150 11.72 ns ns

Table 8. The combined effect of TSP. osmotic shock and Ivsozvme on the survival of Gram +ve bacteria at 4°C in the presence of 50% serum

Organism NaCl (M) TSP(M) pH" % cell survival lysozyme -i-lysozyme

L. mono ¬ none none 7.15 100 37 cytogenes 0.8 none 6.90 100 23 none 0.010 9.71 100 54 none 0.050 11.38 74 5 none 0.100 11.71 12 2

0.8 0.010 9.36 74 27

0.8 0.050 10.94 2 ns

0.8 0.10 11.71 2 ns

St . aureus none none 7.44 100 100

0.8 none 7.18 100 100 none 0.01 9.67 100 100 none 0.05 11.53 100 100 none 0.10 11.82 100 100 none 0.15 12.10 100 89

0.8 0.01 9.39 100 100

0.8 0.05 11.07 100 100

0.8 0.10 11.51 100 100

0.8 0.15 11.69 95 82 Tables 1 and 3 show results at 37°C on gram negative and gram positive organisms respectively and Tables 2 and 4 show results at 4°C. The TSP concentrations tested ranged from 0.OM to 0.005M for the gram negative organisms and 0.OM to 0.10M for gram positive.

The results (Table 1, 37°C, Table 2, 4°C) confirm that gram negative cells are resistant to lysozyme in the absence of osmotic shock. Osmotic shock or TSP (up to 0.002M) alone, also give relatively low kills. The combination of osmotic shock and lysozyme treatment gave high kills, as predicted by WO 93/00822, for E. coli and P. fluorescens.

At 37°C, in all cases, no surviving cells were detected when a low concentration of TSP (0.001 or 0.002M) was used in combination with osmotic shock, even in the absence of a subsequent lysozyme treatment. TSP treatment in the absence of osmotic shock also enhanced killing of cells subsequently exposed to lysozyme. The pH of cell suspensions treated with TSP (up to 0.002M) were <10. Thus, these results show marked improvement in cell kills when TSP treatment was combined with either lysozyme or osmotic shock treatment at 37°C.

Results at 4°C (Table 2) were essentially similar to those at the higher temperature. However, Campylobacter proved more resistant than the other bacteria tested; cell kills of

Campylobacter were not increased by lysozyme treatment, and, in combination with osmotic shock, 0.005M TSP was required to reduce the number of Campylobacter to an undetectable level. Nevertheless, even for Campylobacter, TSP killing was clearly promoted by osmotic shock.

Osmotic shock treatment alone had little or no effect on the viability of L. monocytogenes or St.aureus at 37 and 4°C (Tables 3 and 4, respectively). However, lysozyme (20μg ml^'1) reduced the viable count of L. monocytogenes by 84-99%; killing appeared temperature dependent, being greater at 37 than 4°C.

For Listeria, combining TSP(>0.05M) treatment with a subsequent lysozyme treatment gave significantly higher kills than for either treatment alone. Osmotic shock also enhanced killing by TSP (cf. killing by 0.005-0.05M TSP in presence and absence of NaCl-treatment; Tables 3 and 4), particularly at 37°C. Thus, in combination with osmotic shock/lysozyme treatments, TSP gave maximal killing (no survivors detected; kills >99.7%) with very low concentrations of TSP (0.005M).

St.aureus was more susceptible to TSP when this treatment was combined with osmotic shock. However, in contrast to results obtained with L. monocytogenes, exposure to lysozyme did not enhance killing following either TSP and/or osmotic shock treatment. However, St. aureus was markedly sensitive to low concentrations of nisin (see Example 3) .

The effect of the presence of serum on the TSP/osmotic shock/lysozyme combined killing treatment was determined by incubating cells in 50% v/v serum plus TSP and NaCl and subsequently diluting the cell suspension in lysozyme solution. At 37°C (Table 5) , 100% kills of the gram negative bacteria, C. jejuni , E. coli, P. fluorescens and S. enteritidis were obtained. However, the TSP concentration required (5 to 10 mM) was higher than for cells suspended without serum. The presence of serum lowered the pH of the TSP (5-lOmM) /NaCl treatment solution to between 8 and 9. At 4°C (Table 6) , a similar result was obtained, except with S. enteritidis, for which a kill of 90% was obtained in the combined treatment (TSP/osmotic shock/lysozyme) with lOmM TSP. In the presence of serum, 100% kills of the gram positive bacterium, L. monocytogenes were achieved at 4 and 37°C (Tables

7 and 8) using 50mM TSP in the combined TSP/osmotic shock/lysozyme procedure. Without lysozyme treatment, kills were 78% and 98% respectively. St. aureus was again more resistant than L. monocytogenes. At 37°C, a TSP concentration of 150mM was required to give a 100% kill and kills were not significantly enhanced by osmotic shock or lysozyme treatment. At 4°C, St. aureus was resistant (<20% kill) to TSP (>150m) and combined treatments using TSP (>150m) . Example 2; Killing of cells adhered to food surfaces

The effective kills achieved for gram negative organisms (E. coli and S. enteri tidis) attached to chicken skin for various concentrations of TSP in the presence or absence of osmotic shock and/or a subsequent lysozyme treatment was tested as follows.

Overnight cultures of test bacteria were diluted 1/25 in fresh medium and grown to the early stationary phase (approximately 4h incubation) as described above. The experimental food surfaces used were: lg chicken skin samples (removed from wing or thighs and purchased from a local supermarket) ; and circular sections (diameter 3 cm) of lettuce leaves. Food samples were immersed in early stationary phase cultures for 30 minutes at ambient temperature and subsequently dried for 3 minutes in a stream of cold air, or alternatively, the surfaces of the samples were inoculated with 20μl of the test culture and left to air dry for 30 min.

Inoculated test samples were immersed in 10 ml TSP solution, 10 ml TSP plus 0.8M NaCl, 0.8M NaCl or distilled water, as appropriate. After incubation on a rotary shaker for 10 mins at ambient temperature, or 4 or 37°C, the samples were shaken to remove excess fluid and then immersed in 10ml distilled water or distilled water containing lOOμgml^'1 lysozyme (Sigma L6876) . Additionally, in some experiments, nisin (3.33 x 10⁵M) or lactic acid (2.5 x 10^'3 or 2.5 x lθ"M) was added to the final treatment solution, or an additional water rinse was included prior to the final treatment. After incubation for 30 minutes, samples were stomached individually for 2 minutes with 50 ml distilled water in a stomacher (Colworth 80) . Serial dilutions of stomached samples, and where indicated, treatment fluids, were plated on plate count agar (Oxoid) or blood agar for C. jejuni only. Plates were incubated as described above and the percentage cell kill calculated and recorded in Tables 9 and 10. Table 9. (A) The combined effect of TSP. osmotic shock and lysozvme on the survival of E. coli attached to chicken skin at room temperature. (B) PH values for processing solutions following initial treatments with or without Q.20M TSP.

NaCl (M) TSP (M) TSP,TSP/NaCl % Cell Si rviva pH - lysozyme + lysoz none none 6.64 100 98

0.8 none 6.39 86 20 none 0.01 9.82 100 84 none 0.05 11.69 93 53 none 0.10 11.99 0.1 1.8 none 0.2 12.34 0.8 0.4

0.8 0.01 9.24 45 45

0.8 0.05 11.04 53 39

0.8 0.10 11.92 0.2 0.1

0.8 0.2 12.27 0.7 0.4

B

NaCl (M ) TSP (M) PH of initial pH of final treatment treatment solution * solution **

- lysozyme + lysozyme none none 6.59 6.62 6.48

0.8 none 6.56 6.50 6.52 none 0.20 12.41 11.67 11.68

0.8 0.20 12.32 11.56 11.56

* initial treatment solutions were: TSP, TSP + NaCl, or water ** final treatment solutions were: lysozyme solution or water

Table 10. The combined effect of TSP. osmotic shock and lysozyme on the survival of E. colj attached to chicken skin incubated at 4°C

5 NaCl (M) TSP (1 pH % Cell Survival -lysozyme +lysozyme none none 6.73 100 91

0.8 none 6.36 100 41 υ0 none 0.10 12.22 1 1 none 0.20 12.46 0.2 0.1

0.8 0.10 11.95 4 3 5 0.8 0.20 12.30 6 1

* pH of initial treatment solutions, i.e TSP, NaCl, TSP + NaCl, or water

0

The TSP concentrations used was relatively high, being based on those required to kill suspended organisms in the presence of high organic load, such as might exist at the skin surface. 5

The results show synergy between TSP and osmotic shock treatments. However, TSP and TSP/osmotic shock treatments did not, in this case, appear to significantly enhance sensitivity to lysozyme, possibly due to carry over of TSP to the lysozyme 0 solution, which became alkaline.

The experiments were repeated at ambient temperature using lower TSP concentrations (0.002 to 0.01M). In these experiments the skin surface was directly inoculated with test 5 organism rather than being immersed in culture. This was to minimise the access of organisms to the underside of skin samples, where they might be more easily occluded by fatty material. The results are shown in Tables 11 and 12. Table 11. The combined effect of low TSP concentration, osmotic shock and Ivsozvme on the survival of E. coli attached to chicken skin at room temperature

NaCl (M) TSP (M) pH* % Cell Survival -lysozyme +lysozyme none none 6.93 100 44 0.8 none 6.45 32 14 none 0.002 8.83 42 14 none 0.005 10.34 50 7 none 0.01 11.08 44 9

0.8 .002 9.16 18 3 0.8 ,005 9.90 25 5 0.8 0.01 10.58 25 7

* pH of initial treatment solutions, i.e TSP, NaCl, TSP + NaCl, or water

Table 12. The combined effect of TSP. osmotic shock and lysozvme on the survival of S. enteri tidis dried onto the surface of chicken skin

NaCl (Mlι TSP (M) pH* % Cell Survival -lysozyme +lysozyme none none 6.82 100 87

0.8 none 6.29 100 30 none 0.001 7.00 100 57 none 0.002 8.56 87 85 none 0.005 10.31 11 9 none 0.01 10.90 27 7

0.8 0.001 6.73 97 19

0.8 0.002 7.74 17 9

0.8 0.005 9.908 23 8

0.8 0.01 10.81 5 6

* pH of initial treatment solutions, i.e TSP, NaCl, TSP + NaCl, or water Results (Table 11) show that at these low TSP concentrations, there was synergy between TSP/lysozyme, TSP/osmotic shock and TSP/osmotic shock/lysozyme treatments. The maximum cell kill, following treatment with 0.002M TSP (pH approximately 9) was 97%, comparable to that for cells inoculated onto the skin surface and treated for 10 min with 0.4M TSP (pH 12.5) alone. The pH of lysozyme treatment solutions was in all cases <7.

At the TSP concentrations used (up to 0.01M), the maximum kill of S. enteri tidis cells attached to chicken skin was 95% (Table 12) and there was some indication of synergy between TSP/lysozyme and TSP/osmotic shock treatments. To test the possible effect of TSP carry over on lysozyme activity in the final washing solutions, the solutions were acidified using 2.5 or 0.25mM lactic acid. At these concentrations, the pH of lysozyme solutions, and controls without lysozyme, decreased to approximately 5.0 (2.5mM lactic acid) and 6.5 (0.25mM lactic acid). However, cell kills were not increased, except following treatment with 0.01M TSP/NaCl. In this case, cell kill rose from 94 to 99%, in samples subsequently treated with lysozyme in 2.5mM lactic acid. In part, this enhanced killing may be due to an increased pH shock.

Lettuce was used as a contrasting surface to that of chicken. Kills of attached E. coli clearly showed a marked synergy between TSP and subsequent lysozyme treatments, especially at lower TSP concentrations (less than 0.01M) . At higher concentrations the effect of lysozyme was less marked. The results on lettuce leaves also showed a synergy between TSP and osmotic shock treatments, especially at 0.01M TSP (the highest concentration tested) . Maximal kills of E. coli on lettuce were >99% for both TSP/lysozyme and TSP/osmotic shock combined treatments. Thus, in contrast to the results observed using chicken skin, data for lettuce were much closer to those obtained for suspended cells. This implies that the process of attachment to a surface does not in itself provide protection against the treatments used, but that the nature of the surface is critical to survival. A single experiment was also conducted using S. enteri tidis attached to lettuce. Results were qualitatively similar to those for E. coli , though cell kills were generally less. Example 3: Killing effect of nisin

Experiments similar to those described above for lysozyme were also carried out using nisin or nisin plus lysozyme. Nisin was obtained as a freeze-dried powder (2.5% nisin with NaCl and denatured milk solids; Sigma) . It was dissolved in water and stored frozen (-70°C) as a stock solution (3mM) . After thawing, it was sterilised by membrane filtration and used at concentrations in the range 0.114μM to 34.2μM. The results are shown in Tables 13 to 15.

Table 13. The effect of nisin and TSP treatment on the survival of C. eiuni . E. coli and P. fluorescens

Organism Temperature TSP % cell kill

(°C) Concentration (with log

(mM) reduction)

C. jejuni 37 0 98.0 (1.6) 0.5 >99.9 (>5.8)

4 0 61.0 (0.7) 0.5 58.2 (0.4) 1.0 54.5 (0.3) 2.0 65.8 (0.5) 5.0 95.1 (1.3)

E. coli 37 0 72.0 (0.8) 0.5 98.1 (1.7) 2.0 98.6 (1.9) 5.0 >99.9 (7.4)

E. coli 45 0 42.1 (0.5) 0.5 94.1 (>6.5)

P. 37 0 41.5 (0.5) fluorescens 37 0.5 >99.9 (6.8) 1.0 >99.9 (5.0) ambient 0.5 68.8 (0.5) 1.0 99.5 (2.3) 2.0 99.9 (6.4)

St . aureus 37 0 >99.9 (4.8)

1 >99.9 (>6.5)

The results shown are for cells incubated in TSP at various temperatures and diluted (1:10) in nisin solution (1.14μM) at ambient temperature. Table 14. The effect of nisin and TSP treatment on the survival of C. ieiuni . E. coli and S. enteri tidis attached to chicken s-kin.

Organism TSP concentration % ce 11 kil.

(mM) (wi .th log reduct :ion)

- nisin + nisin

C. j ej uni 5 87.8 (0.9) 99.1 (2.0) 10 89.0 (1.0) 95.6 (1.4)

E. coli 5 82.6 (0.8) 99.9 (3.6) 10 82.9 (0.8) 96.2 (1.4)

5* 18.2 (0.1) 97.0 (1.5) 10* 56.8 (0.4) 96.6 (1.5)

S. enteri tidis 5 69.9 (0.5) 55.1 (0.4) 10 72.2 (0.6) 97.1 (1.6)

St . aureus 0 0 >99.9 (3.1)

5 0 >99.9 (3.9)

10 0 >99.9 (3.5)

* water rinse omitted

The results shown are for cells dried on the surface of the skin, incubated in TSP at 37°C, rinsed quickly (5sec) in water and transferred to nisin solution (34.2μM) or distilled water at 37°C.

Table 15. The effect of TSP and nisin plus lactic acid or nisin plus Ivsozvme treatments on the survival of S. en teri tidis attached to chicken skin.

TSP concentration % cell kill (mM) (with log reduction) + nisin + nisin + lysozyme

5 59 (0.4) 87.5 (0, 9) 10 96 (1.5) 93.8 (1.2)

B

TSP concentration % cell kill (mM) (with log reduction) lactic acid lactic acid + nisin

(pH5! (pH5)

Experiment 1

5 96.3 (1.4) 90.0 (1.0) 10 96.4 (1.5) 93.9 (1.2)

Experiment 2

5 99.2 (2.1) 99, (2. 2) 10 99.4 (2.2) 98, (1.7)

* after addition of chicken skin sample

The results shown are for cells dried on the surface of skin, incubated in TSP at 37°C, and either (A) rinsed quickly (5 sec) in water and transferred to nisin (3μM) or nisin plus lysozyme (lOOμg ml^'1) at 37°C; or (B) transferred to nisin in 2.5mM lactic acid.

Table 13 illustrates the effect of 1.14μM nisin on suspended cells previously treated with various concentrations of TSP. The cells were: C. jejuni (4 and 37°C) E. coli (37°C) and P. fluorescens (room temperature and 37°C) . In experiments conducted at 37°C, nisin caused high kills (up to 7 logs) of cells previously treated with only low TSP concentrations (0.5-1.OmM, depending upon the organism) . Under the conditions used, nisin or TSP treatment did not cause significant cell death when applied singly. In addition, at higher TSP concentrations, where TSP treatment alone did cause some cell death, a marked synergy between TSP and nisin treatment was evident. Under the conditions used, nisin gave high kills of St. aureus in both TSP-treated and untreated cells (Table 13) .

Table 14 illustrates the results of experiments where E. coli cells were dried on the outer surface of chicken skin. It can be seen that nisin markedly enhanced killing at 37°C of cells previously treated with 0.01 and 0.005M TSP; at 0.05M TSP, kills were similar in the presence and absence of nisin (data not shown) . Cell killing by nisin following exposure to low TSP concentrations was further enhanced by the introduction of a quick rinse in water, after TSP treatment and immediately prior to immersion in nisin solution. Cell kills at 0.005M TSP were consistently >99%. The high kills of E. coli obtained using 0.005M TSP in combination with a further nisin treatment were comparable to those seen at very high TSP concentrations (0.4M) without nisin. Similarly high kills (99%) were also observed for C. jejuni (Table 14). S.enteritidis was more resistant, nevertheless, kills of up to 97% were observed (Table 14) . (Under the conditions used, nisin was highly effective in killing the gram positive bacterium St. aureus . Even in the absence of prior treatment with TSP, kills were >99.9% (Table 14) .

Table 15(B) shows the effect on killing of Salmonella when lactic acid was incorporated into the nisin washing solution. The lactic acid was added to ensure an acid pH (approximately 5.0) which was near optimal for nisin activity. Cell kills in the presence of nisin increased to 99%; however, of potential interest, kills in the absence of nisin were similarly high, suggesting that an increased pH shock was a significant cause of cell death.

Table 15(A) shows the effect of lysozyme plus nisin following TSP and combined osmotic shock/TSP treatments on E. coli attached to chicken skin as the test system. In experiments conducted at 37°C, both nisin and lysozyme reduced the viable count of TSP treated chicken skin by >90%. However, kills were not further improved when skin was treated with nisin plus lysozyme. A similar conclusion was reached in experiments using combined osmotic shock/TSP treatment prior to nisin and/or lysoyme treatment at both ambient and 37°C. Kills of up to 95% were obtained; however, comparison of the data with that for lysozyme or nisin alone, showed no increased killing effect by the combination. It is possible that nisin and lysozyme may act antagonistically or that they affect the same population of cells.

In experiments using chicken, optimal combined treatments involving nisin gave high cell kills. In these experiments it was possible that cell kills were underestimated due to the presence of naturally-acquired bacteria. Therefore, some experiments were repeated using chicken portions sterilised by γ-irradiation. In addition, the effectiveness of a combined TSP plus osmotic shock and nisin treatment was determined, i.e. chicken skin was exposed to TSP in the presence of NaCl and subsequently transferred to nisin solution. The results are shown in Table 16.

Table 16. The effect of nisin, osmotic shock and TSP treatment on survival of E. coli . P. fluorescens. S . enteri tidis , L . monocytogenes and St. aureus attached to previously irradiated (sterilised) chicken skin.

Organism % cell kill (log reduction)

TSP + nisin NaCl + nisin TSP + NaCl + nisin

E. coli 97. (1.59) 85.8 (0.90) 99.4 (2.20)

P. fluorescens 99.1 (2.06) 99.4 (2.24) 99.4 (2.20)

S. enteritidis 98.5 (1.84) 96.6 (1.47) 99.1 (2.05)

L. monocytogenes >99.9 (>5.0) 99.9 (3.38) >99.9 (>5.0)

St. aureus >99.9 (3.7) 99.9(>6.0) >99.9 (4.9)

The results shown are for cells dried on the surface of the skin, incubated in TSP (5mM) , NaCl (0.8M) or TSP plus NaCl at 37°C, and transferred to nisin (33.3μM) solution. Where TSP was used, a brief water rinse (5 sec) was applied immediately prior to transfer to nisin solution.

For gram negative bacteria, kills determined for TSP-nisin treatment (>97%) were generally higher than those determined in experiments using unsterilised chicken (cf. Tables 14 and 16) . High kills (>85%) were also observed for osmotic shock- nisin treatment (Table 16) . However, the combined TSP plus osmotic shock and nisin treatment was most effective (cell kills >99%) . Under the conditions used, this treatment also reduced the viable count of L. moncytogenes and St. aureus (gram positive species) by approximately 5 log cycles (Table 16) .

Claims (33)

1. A method for reducing the levels of gram negative and gram positive bacteria in a sample comprising treating the sample with a 0.0005 to 0.2M solution of a trialkali metal orthophosphate, said treatment being combined with one or more of the following further treatments:

a) subjecting the sample to osmotic shock, b) exposing the sample to an enzyme which breaks down peptidoglycan after said treatment with trialkali metal orthophosphate; and ii) exposing the sample to a bacteriocin after said treatment with trialkali metal orthophosphate.

2. A method according to claim 1 wherein the further treatment comprises step (a) and step (b) or step (c) .

3. A method according to claim 1 or claim 2 which includes subjecting said sample to osmotic shock and said osmotic shock comprises a hypo-osmotic shock.

4. A method according to claim 3 wherein said hypo-osmotic shock is preceded by administration of a hyperosmotic shock.

5. A method according to Claim 4 wherein said osmotic shock treatment includes the steps of exposing the sample to a first solution having a water activity (a_w) of 0.997 or less and subsequently exposing the sample to a solution of a_u higher than that of said first solution.

6. A method according to claim 5 wherein the said first solution contains NaCl at a concentration sufficient to provide a water activity of 0.997 or less,

7. A method according to Claim 5 or claim 6 wherein said first solution contains 0.0005 to 0.2 M of a tri-alkali metal orthophosphate.

8. A method according to any one of the preceding claims which includes a step of exposing the sample to an enzyme which breaks down peptidoglycan, and said enzyme comprises lysozyme.

9. A method according to any of Claim 7 wherein the sample is treated with a solution of lysozyme at a concentration of at least lμgml ^!.

10. A method according to Claim 8 wherein the lysozyme is provided in the form of a solution of freeze dried egg white which is at a concentration of at least O.lmgml^'1.

11. A method according to any one of the preceding claims wherein the sample is exposed to a bacteriocin and said bacteriocin is selected from nisin or pedocin.

12. A method according to claim 11 wherein the bacteriocin is nisin.

13. A method according to claim 12 wherein the sample is treated with a solution containing nisin of concentration of at least O.lμM.

14. A method according to Claim 13 wherein prior to treatment with the solution containing nisin, the sample is rinsed with water.

15. A method according to any one of Claims 8 to 14 wherein the bacteriocin or enzyme solutions are acidified.

16. A method according to claim 15 wherein the solutions are acidified to a pH of approximately 5.0.

17. A method according to claim 15 or claim 16 wherein the solutions are acidified by addition of lactic acid.

18. A method according to claim 17 wherein the lactic acid is present at a concentration of at least 0.25mM.

19. A method according to any one of the preceding claims wherein the trialkali metal orthophposphate is trisodium phosphate.

20. A method according to any one of the preceding claims wherein the concentration of the trialkali metal orthophosphate solution is in the range of from 0.001 to 0.02M

21. A method according to claim 20 wherein the concentration of the trialkali metal orthophosphate solution is in the range of from 0.005 to 0.01M.

22. A method according to any one of the preceding claims wherein the sample comprises a consumable product or a surface.

23. A method according to claim 22 wherein the sample comprises a portion of a foodstuff.

24. A method according to any one of the preceding claims wherein a treatment step is effected by immersing the sample into a treatment solution.

25. A method according to any one of the preceding claims wherein a treatment step is effected by spray washing the sample with a treatment solution.

26. A method according to claim 1 wherein a sample is treated with a first solution comprising a trialkali metal orthophosphate at a concentration in the range of from 0.0005 to 0.2M, said solution having a water activity (a_w) of 0.997 or less, and subsequently treating the sample with a second solution of an enzyme which breaks down peptidoglycan or a bacteriocin, said second solution having an a_w higher than that of said first solution.

27. A method according to claim 25 wherein the sample is sequentially immersed in said first and second solutions.

28. A method for reducing the levels of gram negative and gram positive bacteria in a portion of a foodstuff comprising immersing the portion in a solution of trisodium phosphate at a concentration in the range 0.0005 to 0.2M, and sodium chloride at a concentration of about 0.8M, at a temperature in the range 4 to 50°C, and then spray washing with lysozyme solution.

29. A method for reducing the levels of gram negative and gram positive bacteria in a portion of a foodstuff comprising immersing the portion in a solution of trisodium phosphate at a concentration in the range 0.0005 to 0.2M, at a temperature in the range 4 to 50°C, and then spray washing with nisin at a concentration of greater than lμM.

30. A kit for carrying out a method according to any one of the preceding claims, which comprises a trialkali metal orthophosphate and one or more of the following components:

a) a reagent which can be used to induce osmotic shock in a cell; b) an enzyme which can break down peptidoglycan; and c) a bacteriocin.

31. A kit according to claim 30 wherein the components are in the form of aqueous solutions.

32. A product which has been treated by a method according to any one of claims 1 to 29.

33. A product according to claim 32 which comprises a foodstuff.