Journal of Dairy Science
Volume 91, Issue 7 , Pages 2566-2574, July 2008

Challenge Testing the Lactoperoxidase System Against a Range of Bacteria Using Different Activation Agents

Department of Food Biosciences, University of Reading, Whiteknights, Reading RG6 6AP, United Kingdom

Received 30 April 2007; accepted 4 March 2008.

Article Outline

Abstract 

Lactoperoxidase (LP) exerts antimicrobial effects in combination with H2O2 and either thiocyanate (SCN−) or a halide (e.g., I−). Garlic extract in the presence of ethanol has also been used to activate the LP system. This study aimed to determine the effects of 3 LP activation systems (LP+SCN−+H2O2; LP+I−+H2O2; LP + garlic extract + ethanol) on the growth and activity of 3 test organisms (Staphylococcus aureus, Pseudomonas aeru-ginosa, and Bacillus cereus). Sterilized milk was used as the reaction medium, and the growth pattern of the organisms and a range of keeping quality (KQ) indicators (pH, titratable acidity, ethanol stability, clot on boiling) were monitored during storage at the respective optimum growth temperature for each organism. The LP+I−+ H2O2 system reduced bacterial counts below the detection limit shortly after treatment for all 3 organisms, and no bacteria could be detected for the duration of the experiment (35 to 55h). The KQ data confirmed that the milk remained unspoiled at the end of the experiments. The LP + garlic extract + ethanol system, on the other hand, had no effect on the growth or KQ with P. aeruginosa, but showed a small retardation of growth of the other 2 organisms, accompanied by small increases (5 to 10h) in KQ. The effects of the LP+SCN−+H2O2 system were intermediate between those of the other 2 systems and differed between organisms. With P. aeruginosa, the system exerted total inhibition within 10h of incubation, but the bacteria regained viability after a further 5h, following a logarithmic growth curve. This was reflected in the KQ indicators, which implied an extension of 15h. With the other 2 bacterial species, LP+SCN−+H2O2 exerted an obvious inhibitory effect, giving a lag phase in the growth curve of 5 to 10h and KQ extension of 10 to 15h. When used in combination, I− and SCN− displayed negative synergy.

Key words: challenge testing, lactoperoxidase, antimicrobial effect, keeping quality

 

Back to Article Outline

Introduction 

Lactoperoxidase (LP) exerts antimicrobial effects when combined with a source of H2O2 and either thiocy-anate (SCN−) or a halide. The enzyme can also catalyze the oxidation of sulfur compounds in garlic, at the expense of oxygen released from ethanol, to generate antimicrobial products (Jandal, 1998). The enzyme together with a source of H2O2/oxygen and an electron donor form the LP system (LP-s). The system has wide applications in the dairy industry, particularly in raw milk preservation in situations where prompt refrigeration is difficult, especially in developing countries. Activation of the LP-s can be used to extend the keeping quality (KQ) before heat treatment.

Different activation strategies and substrate levels have been used in earlier studies to activate the LP+SCN−+H2O2 system. Siva et al. (1991) reported ex-tended KQ of 9.9, 12.2, and 13.6h of raw milk treated with equal concentrations of 10:10, 20:20, and 30:30mg/kg of SCN−:H2O2, respectively, at a storage temperature of 30°C. With an excessive molar concentration of H2O2, Ewais et al. (1985) reported a KQ of 48h for buffalo and cow milk preserved at 20 to 22°C, and treated with 10:25 and 10:30mg/kg of SCN−:H2O2. However, they reported that a greater concentration of H2O2 was necessary for preservation of milk at temperatures as high as 35°C. The use of 70 to 80mg/kg of H2O2 at that temperature kept the acidity of milk unchanged for 24h and greater levels of H2O2 (>80mg/kg) further extended the KQ of buffalo and cow milk. Barraquio et al. (1994) reported a 7-h extended shelf life in LP-s activated raw milk treated with 14:30mg/kg of SCN−:H2O2 at 27 to 32°C ambient temperature. On using excess molar concentration of SCN−, 15:10mg/kg of SCN−:H2O2, Chakraborty et al. (1986) recorded a 3-h extension in shelf life of buffalo milk at 37°C, and Björck et al. (1979) reported the total lag phase of 7 to 8h (30°C), 11 to 12h (25°C), 15 to 16h (20°C), and 24 to 26h (15°C) using 10:8.5mg/kg of SCN−:H2O2. Cailliez-Grimal et al. (2002) observed increased antimicrobial activity of the LP-s associated with greater bacterial inhibition when SCN− was replaced with I− or IO3−. Similarly, Revol-Junelles et al. (2001) reported that activation with I− or IO3− gave greater, permanent inhibition of gram-positive bacteria, particularly Staphylococcus aureus at 25 and 15°C, than SCN−. However, no antibacterial action was observed against gram-negative strains. Jandal (1998) activated the LP-s with garlic extract + ethanol (GE+E) in raw buffalo milk stored at ambient temperature (25°C), which resulted in an increased storage life of up to 15h based on titratable acidity, 9h based on alcohol stability, and 15h based on clot on boiling. Although extensive work has been done on the preservation of raw cow milk using LP+SCN−+H2O2, the LP+I−+H2O2 system is less well understood, and no systematic study has been carried out LP+GE+E systems. The aim of this work was to investigate the influence of the different substrates of the LP-s on antimicrobial activity by challenge testing against bacteria of importance to milk spoilage, and to assess the relative performance of the system on milk preservation. The test organisms used were Staph. aureus, Pseudomonas aeruginosa, and Bacillus cereus. Ultra-high temperature sterilized and raw whole cow milk were used as reaction media. It should be noted that LP activity is completely destroyed by the UHT treatment, and hence enzyme had to be added to UHT milk in these experiments. Growth pattern and keeping quality indicators were used to measure the antimicrobial activities of the LP+SCN−+H2O2, LP+I−+H2O2, and LP+GE+E systems.

Back to Article Outline

Materials and Methods 

Materials 

Sodium percarbonate (Na2CO3·1·5H2O2) was used as a source of H2O2 for LP system activation; SCN− was in the form of NaSCN and I− in the form of KI (all supplied by Sigma-Aldrich, Gillingham, UK). Bovine LP was kindly donated by DMV International (New York, NY).

Homogenized whole UHT milk was supplied by Arla Foods (Leeds, UK). This milk was used because it is sterile, thereby minimizing or avoiding contamination with microorganisms other than the test organisms.

Test Organisms and Growth Conditions 

Three strains, Staph. aureus (NCDO 10651), P. aeru-ginosa (NCTC 10299), and B. cereus (NCDO 0577) obtained from the departmental stock culture maintained at −80°C in cryoprotective media, were used in this study. The cultures were selected on the basis that they have the potential to cause spoilage in raw and pasteurized products. Cultures were regenerated in nutrient broth and grown overnight on nutrient agar slants at 30°C. Single colonies were isolated from stock slants by streaking onto agar plates at 30°C for 24h, and the purity of the test organisms was confirmed by microscopic examination. Working inocula for antibacterial assays were prepared by subculturing single colonies from agar plate streaks into nutrient broth and incubating at the optimum growth temperatures of the organisms: 25°C (P. aeruginosa), 37°C (Staph. aureus), and 30°C (B. cereus) for 18 to 20h. The number of viable cells was around 107 to 108 cfu/mL of broth.

Preparation of the LP+GE+E System Substrates 

Garlic was skinned and the central part ground in a sterile mortar. The resulting concentrated extract was first filtered through muslin cloth and then through Whatman filter paper no. 4. The concentrated garlic extract filtrate was diluted (1% vol/vol) in distilled water. Each 1mL of 1% garlic extract combined with 1mL of 1% ethanol was considered to be equivalent to 10mg/kg as claimed previously by Jandal (1998).

Activation of the Different LP-s 

The concentration of LP used was 30mg/kg, based on its natural concentration in milk (De Wit and Van Hooydonk, 1996). The H2O2 concentration of 80mg/kg was adopted in this study because the findings of Ewais et al. (1985) demonstrated effective performance of the LP+SCN−+H2O2 system with H2O2 concentrations of 70 to 80mg/kg, ensuring an extended shelf life of 24h. The recommended SCN− level is between 3.5 and 12mg/kg (Rossi and De Oliveira, 1993–1994) or ≤10mg/kg (Ewais et al., 1985). Therefore, the concentration of 7:80mg/kg of SCN−:H2O2 was chosen to activate the LP system in these experiments. For comparison, the same substrate level was adopted for LP+I−+H2O2 system; that is, 7:80mg/kg of I−:H2O2. Substrates for the LP+GE+E system were similarly established based on earlier work (Jandal, 1998), and preliminary trials in raw milk (data not shown) gave the best KQ for 110:30mg/kg garlic extract:ethanol combination level.

The composition of the reaction mixtures for the different LP-s were as follows: LP+SCN−+H2O2 (30:7:80mg/kg), LP+I−+H2O2 (30:7:80mg/kg), and LP+GE+E (30:110:30mg/kg).

Lactoperoxidase, thiocyanate, iodide, and sodium percarbonate were dissolved separately in 1mL of HPLC water and subsequently added to whole UHT milk (inoculated with test strains) individually through syringe filters.

Determination of Antimicrobial Activity 

To determine the antimicrobial activity of the different LP-s, UHT milk (2L) was inoculated with pure cultures of the test bacteria to give an initial concentration of approximately 102 cfu/mL. The milk was thoroughly agitated for 5min to ensure uniform suspension of the bacteria and then divided into 4 subsamples. The LP-s for 3 subsamples were activated using different substrates as described above and the untreated sub-sample was used as a control. After thorough mixing, all samples were aliquoted (25-mL samples) into 30-mL sterile containers and incubated at each culture's respective optimum growth temperatures: 25°C (P. aer-uginosa), 37°C (Staph. aureus), and 30°C (B. cereus). The growth pattern was followed and the KQ determined. Samples were taken at 0h (just after inoculation) and at 5-h intervals and the total viable count (TVC) of surviving cells was determined by the surface plate agar method as described previously (Marks et al., 2001). The TVC was determined using 0.1mL of the required dilution (100 to 105 serial dilutions) on predried and appropriately marked duplicate Petri plates (plate count agar CM183, Oxoid Ltd., Basing-stoke, UK) with sterile spreaders. The plates were incubated in an inverted position for 48h and the colonies counted promptly after the incubation period. The TVC measurements were aimed at monitoring the growth pattern of the bacteria and relate to the antibacterial effects of the different LP systems.

Determination of KQ 

Keeping quality of whole UHT milk inoculated with pure cultures of the test bacteria, activated with the appropriate systems and incubated at their respective optimum temperatures, was measured using pH, titratable acidity, clot on boiling, and alcohol stability tests as described below. All KQ measurements were carried out in duplicate.

For titratable acidity, 1mL of phenolphthalein solution (0.5% wt/wt) was added to 10mL of milk. The mixture was titrated against 0.11 M NaOH (VWR International Ltd., Poole, UK) to an end point marked by a pink color persistent for at least 5s. The milk was judged to be spoiled when TA had increased by 0.02% (Chakraborty et al., 1986).

The clot on boiling test was performed by boiling 2mL of milk sample in a test tube for 5min, after which it was examined for curdling by tilting the tube gently. The sample was judged to be spoiled when clotting was observed (Harding, 1995).

For the alcohol stability test, equal volumes of milk and 70% ethanol (prepared from 99% ethanol, VWR International Ltd.) were mixed in a test tube. The tube was inverted several times and then examined for coagulation. Coagulation of milk indicated spoilage (Marks et al., 2001).

The pH was determined with a pH meter (ABB Kent EIL 7045/46, Kent, UK) and spoilage was judged to have occurred by a decrease in pH of 0.4 units (Barrett et al., 1999).

There was generally very good agreement between the 4 methods of KQ assessment, and hence to simplify presentation, KQ is summarized as the range of values of time for spoilage to occur over the 4 methods in duplicate experiments.

Back to Article Outline

Results 

Antimicrobial Effects of the LP-s 

This series of experiments was designed to measure the antibacterial effect of the 3 LP-s. Two parameters, bacterial viability of P. aeruginosa, Staph. aureus, and B. cereus, and KQ of the respective samples, were employed as measures of the antibacterial effects of the different LP systems. The experiments were carried out in duplicate on separate dates.

Figure 1 shows the differences in susceptibility of P. aeruginosa to the different LP systems. The LP+I− +H2O2 system reduced the bacterial counts to below the detection limit (log cfu/mL = 1) just after treatment. No counts were detected up to the end of the experiment, which suggests a rapid and complete inactivation of the strain. The effect of the LP+SCN−+H2O2 was time dependent. The system exerted complete bacterial inhibition within a 10-h incubation period, but the bacteria regained viability 5h later. There was little effect on bacterial viability and growth with the LP+GE+E system. The KQ (Table 1) of the LP+GE+E treated sample was comparable to that of the control, implying no increase in shelf life. The LP+SCN−+H2O2 system extended the KQ by 15h, whereas the LP+I−+H2O2 sample had a KQ more than 10h longer than the former system, and the milk was unspoiled at the end of the experiment.

  • View full-size image.
  • Figure 1. 

    The antimicrobial effects of the different lactoperoxidase (LP) systems against Pseudomonas aeruginosa in whole UHT milk. Error bars show the range of data from 2 separate experiments (detection limit = 1 log cfu/mL). Systems: LGE = LP + garlic extract + ethanol; SCN−/H2O2 = LP + thiocyanate + hydrogen peroxide; I−/H2O2 = LP + iodide + hydrogen peroxide.

Table 1. The keeping quality of whole UHT milk inoculated with Staphylococcus aureus, Bacillus cereus, or Pseudomonas aeruginosa, and preserved by activation of lactoperoxidase (LP) systems using different substrates1
Preservation treatment1Keeping quality (h)
P. aeruginosaStaph. aureusB. cereus
Control25−3025−3015−25
LP + GE + E25−3030−3525−30
LP + SCN− + H2O240−4535−4030−35
LP + I− + H2O2>55>45>35

1LP + GE + E = LP + garlic extract + ethanol; LP + SCN− + H2O2 = LP + thiocyanate + hydrogen peroxide; LP + I− + H2O2 = LP + iodide + hydrogen peroxide.

The LP+I−+H2O2 system showed similar lethal effects on Staph. aureus (Figure 2) as on P. aeruginosa. However, the reduction of counts to below the detection limit was achieved after 5h of incubation. The antimicrobial activity was powerful enough to maintain the counts below detection limit for up to 14 d (data not shown), which suggests complete inactivation of the bacteria. The LP+GE+E system resulted in a small retardation in the bacterial growth. The LP+SCN−+H2O2 system produced definite growth retardation and extended the lag phase by 5h before the bacteria recovered and resumed logarithmic growth. The increase in lag resulted in an apparent increase in the bacterial generation time and so enhanced KQ by about 10h (Table 1). The LP+I− +H2O2 activated sample was of acceptable quality up to the end of the experiment as demonstrated by the KQ results (Table 1), whereas the LP+GE+E sample had an extended KQ of 5h.

  • View full-size image.
  • Figure 2. 

    The antimicrobial effects of the different lactoperoxidase (LP) systems against Staphylococcus aureus in whole UHT milk. Error bars show the range of data from 2 separate experiments (detection limit = 1 log cfu/mL). Systems: LGE = LP + garlic extract + ethanol; LP/SCN−/H2O2 = LP + thiocyanate + hydrogen peroxide; LP/I−/H2O2 = LP + iodide + hydrogen peroxide.

The growth and survival of B. cereus cells (Figure 3) gave generally similar results to Staph. aureus and P. aeruginosa. The LP+I−+H2O2 system resulted in similar bactericidal action against B. cereus as with other strains. The counts were below detection limits within 10h of incubation and the level was maintained up to the end of the experiment. The LP+SCN−+H2O2 system prolonged the lag phase by 10h, but this was then followed by the normal bacterial growth rate. There was no pause in growth due to LP+GE+E treatment; however, there was some growth retardation. This is supported by the KQ (Table 1), which was greater for the LP+GE+E treated sample than the control sample but lower than the LP+SCN−+H2O2 and LP+I−+H2O2 treated samples.

  • View full-size image.
  • Figure 3. 

    The antimicrobial effects of the different lactoperoxidase (LP) systems against Bacillus cereus in whole UHT milk. Error bars show the range of data from 2 separate experiments (detection limit = 1 log cfu/mL). Systems: LGE = LP + garlic extract + ethanol; LP/SCN−/H2O2 = LP + thiocyanate + hydrogen peroxide; LP/I−/H2O2 = LP + iodide + hydrogen peroxide.

To eliminate the possibility that I− alone gave rise to the bactericidal effect and H2O2 alone to the bacteriostatic effect on the population, the antimicrobial effects of individual substrates of the LP systems against Staph. aureus at the same concentrations as used in the LP systems were investigated. Figure 4 shows that I− alone had no effect on bacterial growth and there was no effect on the KQ of the milk. Although H2O2 exerted some antibacterial effect (Figure 5), the LP+SCN−+H2O2 system was more effective than H2O2 alone, and resulted in better KQ (35 to 40h for LP+SCN−+H2O2; 25 to 30h for H2O2 alone). These results indicate that LP-catalyzed oxidation of SCN− and I−, respectively, were responsible for the bacteriostatic and bactericidal effects exerted against the test strain population.

  • View full-size image.
  • Figure 4. 

    The antimicrobial effects of iodide (KI; 7mg/kg) alone, and in the presence of H2O2 (80mg/kg) and lactoperoxidase (LP/I−/H2O2) against Staphylococcus aureus (detection limit = 1 log cfu/mL).

  • View full-size image.
  • Figure 5. 

    The antimicrobial effects of H2O2 (80mg/kg) alone and in the presence of lactoperoxidase and SCN− (LP/SCN−/H2O2; 7mg/kg) against Staphylococcus aureus in whole UHT milk.

The LP+I−+H2O2 system with or without SCN− at a greater microbial load (Figure 6) showed variation in susceptibility of Staph. aureus to the LP system activated with either I− or SCN− or combinations of both in different ratios. The LP+I−+H2O2 system displayed similar bactericidal activity against Staph. aureus at a greater microbial load (5 log cycles) as was the case with a lower microbial load (2 log; Figure 2). However, complete inactivation was achieved after a 15-h incubation compared with a ≤5-h incubation time (Figure 2) with a lower initial cell density. The bacteriocidal effect of the LP+I−+H2O2 system was reduced in the presence of SCN− and its effectiveness depended on the relative amount of SCN− to I−. The bacteriocidal effect and growth inhibition was greater with the LP+I−+H2O2 system with a 1:2 SCN−:I− ratio than with a 1:1 SCN−:I− ratio. The KQ also decreased with increasing concentration of SCN− in the LP+I−+H2O2 system (>80h with I− alone; 70h with 1:2 SCN−:I− ratio; 45h with 1:1 SCN− :I− ratio; 25h with SCN− alone). A similar trend was observed when a mixed inoculum of Staph. aureus, B. cereus, and P. aeruginosa and more combination levels of SCN−:I− were used (Figure 7). The least growth inhibition was observed for the LP+SCN−+H2O2 system, whereas the greatest growth inhibition was recorded for the LP+I−+H2O2 system and the LP+I−+SCN−+H2O2 system with a 1:7 SCN−:I− ratio, with intermediate ratios following the expected trend. The KQ results followed a similar trend with 20 to 30h for SCN− alone to 50 to 55h for I− alone and intermediate KQ values for intermediate ratios. The LP+I−+H2O2 system with or without SCN−, however, failed to show such an effective bacteriocidal effect (as had been demonstrated earlier with individual cultures) against the mixed cultures inoculated at 5 log cycle cell density. The bacterial survival rate was greater under these experimental conditions, which led to a shorter KQ implying a reduction in the system's toxicity. The quantitative effect of microbial load was further investigated by carrying out another set of experiments using a lower microbial load (2 log cycles) of the same strains (Figure 8). At this cell density, the LP+−+H2O2 system with or without SCN− replicated the bactericidal effect and enhanced the KQ (25h with SCN− alone; >85h at all levels of I−), which suggests that the lethality of the LP systems varies among strains and decreases with increasing microbial load.

  • View full-size image.
  • Figure 6. 

    Susceptibility of Staphylococcus aureus to the hypoio-dite-generating lactoperoxidase (LP) system with different combination levels of SCN−:I− (detection limit = 1 log cfu/mL).

  • View full-size image.
  • Figure 7. 

    The antimicrobial effects of lactoperoxidase system activated using different combination levels of SCN−:I− in whole UHT milk inoculated with a high initial level (105) of mixed culture (Staphylococcus aureus, Bacillus cereus, and Pseudomonas aeruginosa).

  • View full-size image.
  • Figure 8. 

    The antimicrobial effects of lactoperoxidase system activated using different combination levels of SCN−:I− in whole UHT milk inoculated with a low initial level (2 log10) of mixed bacterial strains (Staphylococcus aureus, Bacillus cereus, and Pseudomonas aeruginosa) (detection limit = 1 log cfu/mL).

A final experiment investigated the preservation of raw milk incubated at 30°C with LP+I−+H2O2 and LP+SCN−+H2O2 systems (Figure 9). Although there was immediate microbial growth for the control sample and a 5-h extension in lag phase for the LP+SCN−+H2O2 sample, the LP+I−+H2O2 sample demonstrated a partial lethal effect and a more extended lag phase for the first 10h before resuming the normal growth. The KQ results showed better preservative effects of the LP+I− +H2O2 system (35 to 45h) compared with the LP+SCN− +H2O2 system (30 to 40h) or the control raw milk (15 to 25h).

  • View full-size image.
  • Figure 9. 

    The antimicrobial effects of lactoperoxidase + thiocya-nate + hydrogen peroxide (LP/SCN−/H2O2) and lactoperoxidase + iodide + hydrogen peroxide (LP/I−/H2O2) in raw milk incubated at 30°C.

Back to Article Outline

Discussion 

Antimicrobial Effects of the LP Systems 

The 3 combinations of LP-s substrates produced wide variations in antimicrobial effects of the LP-s against Staph. aureus, B. cereus, and P. aeruginosa. The wide variations in the antimicrobial activity among the LP-s may be correlated with the stability of the oxidizing agent(s) produced or their ability to penetrate microbial membranes (Thomas, 1985). Barraquio et al. (1994) related the degree of inhibition to the strength of the interaction of the antimicrobial products and free SH groups of vital enzymes in the bacterial cell wall. It has been established that the oxidizing agents formed from halides and SCN− differ in oxidation potential, stability, and solubility (Thomas, 1985). The higher charge density of I− compared with SCN− leads to the formation of more stable compounds between I− and SH groups of bacterial enzymes, and hence the greater bacteriocidal effect of the oxidation products of I− compared with those of SCN− (De Wit and Van Hooydonk, 1996). They also described the bacteriocidal characteristics of the iodide system against a mixture of Micrococcus, Staphylococcus, Listeria, and B. cereus as reminiscent of the irreversible oxidation of SH groups by I−-induced LP effects. Although the antimicrobial activity of LP+SCN− +H2O2 is more dependent on temperature and pH, LP+− +H2O2 promotes killing independently of these factors (Bosch et al., 2000). It is probable that the performance of the LP+GE+E system is also influenced by similar factors. The LP+GE+E system was much the least effective among the 3 LP systems and also imparted a strong garlic smell to milk, which adds to the limitations of this system.

The lethal effect of LP+I−+H2O2 system is consistent with previous findings that reported 0% survival rates of both Escherichia coli (Klebanoff, 1967) and Candida albicans (Majerus and Courtois, 1992) in the activated LP+I−+H2O2 system in acetate buffer (pH 5) and phosphate buffer (pH 7.4), respectively. The variation in the incubation time necessary for the complete inactivation to occur on exposure to LP+I−+H2O2 reflects the possible variation in sensitivity among strains. This may depend on the cell-wall structure and inhibitory compounds generated by the organism in question (Korhonen, 2004), or on the susceptibility of the primary target enzyme to the LP system (Shin et al., 2001). Whereas complete inactivation was rapid in the case of P. aerugi-nosa, it was slower against B. cereus and Staph. aureus, which suggests that the system is more effective against gram-negative bacteria than gram-positive bacteria. This is in contrast to the findings of Revol-Junelles et al. (2001), who concluded that activation with I− or IO3− was effective against Staph. aureus, but ineffective against 2 gram-negative strains (Pseudomonas fluo-rescens and E. coli). An initial reduction in microbial counts observed for LP+SCN−+H2O2 suggests that the system acts bacteriocidally with an overall bacterio-static effect on the population. The antibacterial effects of LP+SCN−+H2O2 coupled with a reduction in colony count at the early stage has been reported previously in both buffalo and cow milk (Björck et al., 1979; Ewais et al., 1985; Chakraborty et al., 1986). Kamau et al. (1990) described the decrease in colony count of both Listeria monocytogenes and Staph. aureus in milk during the initial 0 to 2h as the killing of a small fraction of the bacteria, and attributed it to the killing effect of short-lived products of the LP-s, which are more bactericidal than OSCN−. However, it should also be emphasized that the antimicrobial effect of LP system on any particular organism depends on the reaction conditions (Pruitt and Reiter, 1985).

The complete inhibition of P. aeruginosa by LP+SCN− +H2O2 and its recovery appears to suggest a complete inhibition of respiration without loss of viability as reported in earlier studies (Thomas and Aune, 1978), although there could be other explanations including adaptation of a small number of survivors, or the presence of resistant mutants that survive and grow. It also con-firms the findings of Bosch et al. (2000) who demonstrated complete inhibition of P. aeruginosa in the presence or absence of iodide. Incorporation of SCN− into SH groups of bacteria effectively removes the reacted SH groups from a functional role such as respiration and glycolysis and may result in cessation of growth (Mickelson, 1979). The complete inhibition observed for P. aeruginosa but not for B. cereus and Staph. aureus similarly suggests it being more effective against gram-negative than gram-positive bacteria. The LP+GE+E system was barely effective against P. aeruginosa, but had a slight effect against gram-positive bacteria, B. cereus and Staph. aureus.

The failure of I− alone to replicate the lethal effects of LP+I−+H2O2 or exert any antimicrobial effect suggests that the resultant bactericidal effect was not just due to I−, but rather the complete LP+I−+H2O2 system. It also appears to confirm the reliance of the antibacterial effect of I− on intermediate oxidation products (HOI, OI−). Identical findings were reported by (Klebanoff, 1968), who demonstrated lack of effect due to I− alone at 0.01μM against Lactobacillus acidophilus in acetate buffer (pH 5), which reinforces the view that the antibacterial effect of the LP+I−+H2O2 system is due to the conversion of iodide by oxidation to a strong antibacterial agent.

The greater antimicrobial activity observed for LP+SCN−+H2O2 compared with H2O2 alone implies that the amount of H2O2 used for activation was still adequate to function as a component of the LP system. Excessive amounts of H2O2 inactivate the enzyme, and H2O2 on its own is an effective antimicrobial agent when applied at a much greater concentration (300ppm; FAO/WHO, 1991), although these levels of addition are not generally permissible. Greater H2O2 concentrations or the presence of catalysts such as LP overcomes the kinetic barrier to H2O2 toxicity (Thomas, 1985).

Single or Mixed Cultures in the LP-H2O2 plus SCN− or I− or Both 

The effectiveness of the LP+I−+H2O2 system at a greater initial microbial load, as in the case of poor quality milk, and its performance in the presence of SCN− (which occurs naturally in milk at 1 to 15mg/L) was tested using greater counts of Staph. aureus and different combination levels of SCN−/I−. The decrease in the bactericidal effect of LP+I−+H2O2 with increasing SCN− concentration suggests an effective competition of SCN− with I− for LP-catalyzed oxidation (Thomas, 1985). Majerus and Courtois (1992) observed a more lethal effect of hypoiodite produced by the LP system in saliva against Candida albicans ATCC 10231 in a group of nonsmokers with salivary SCN− concentrations of 1.24±0.28mM compared with a group of smokers with concentrations of 2.63±0.25mM. Klebanoff (1967, 1968) demonstrated the decrease in the bactericidal effect of myeloperoxidase-iodide-H2O2 against E. coli in acetate buffer (pH 5.0) with increasing SCN− concentration. It appears that, rather than contributing to the antibacterial action, incorporation of SCN− interferes and lowers the bactericidal effects of the iodide system. The effect is probably due to the preferential interaction of SCN− with compound I (reaction product of LP with H2O2), resulting in limited oxidation of I− and hence limited production of antimicrobial compounds. This is supported by the hierarchy of reduction potential (Furtmüller et al., 2002), which demonstrates the relatively greater reactivity of SCN− with compound I with a rate constant of 2.0×108 mol−1 s−1 compared with I− with compound I with a rate constant of (1.2±0.4)×108 mol−1 s−1. These results emphasize the existence of competition for peroxidase and the importance of I−:SCN− ratio for the performance of the LP+I−+H2O2 system (Majerus and Courtois, 1992).

The investigation of the quantitative effect of microbial load on the performance of the LP+I−+H2O2 system with or without SCN− demonstrated the variation in the lethality of the system with changes in cell density of both single and mixed strains. The system appeared to be more lethal at low cell density as reflected by either a shorter incubation time for complete inactivation to occur or reduced survival rate of the test strains, which is probably a reflection of the fact that there are more cells to kill at greater cell densities. Wolfson and Sumner (1994) demonstrated the dependency of the bacte-riocidal effect on the initial inoculum of Salmonella typhimurium in trypticase soy broth. The LP+SCN−+H2O2 system was bacteriocidal to an inoculum level of 102 cfu/mL at 37°C but bacteriostatic to the population at inoculum levels of 106 to 107 cfu/mL. The decrease in cell density seemed to increase the amount of the antimicrobial agent to which each cell is exposed and raised the plateau levels of bactericidal action (Thomas and Aune, 1978). The difference in inactivation of Staph. aureus and mixed cultures at almost the same cell density also suggests that the lethality of the LP systems varies among strains.

Preservation of Raw Milk with LP+I−+H2O2 and LP+SCN−+H2O2 Systems 

The LP+I−+H2O2 system showed better preservative effects in raw milk compared with the LP+SCN−+H2O2 system, implying that H2O2 catalyzed oxidation of I− results in the formation of more effective antimicrobial compounds than H2O2 catalyzed oxidation of SCN−. Both LP systems showed weaker antimicrobial activity in raw milk than in inoculated UHT milk. This could be because the resident microflora in the raw milk was more resistant than the 3 bacteria used as inoculum, although this was not tested here. The results are consistent with those obtained for field trials carried out in tropical conditions in Tanzania using 15:15mg/L of I−:H2O2 (unpublished data). They also suggest that oxidation of only a small amount of I− has a huge impact on the KQ of raw milk. The interference with SCN− reduces the amount of I− oxidized in the LP system to yield antimicrobial compounds. However, because of the relatively lower concentration of iodide in dairy milk (220±120μg/L; Lacroix and Wong, 1980, and 9.6 to 382μg/L; Kirk et al., 2005) compared with SCN− (1 to 15mg/L;Reiter and Harnulv, 1984; Wolfson and Sumner, 1993; De Wit and Van Hooydonk, 1996, and 4 to 59mg/L; Korhonen, 1973), it is unlikely that the LP+I−+H2O2 system would exert any antimicrobial effect without reactivation. Iodine in the milk of ruminants exists primarily in the iodide form (Lacroix and Wong, 1980) and varies widely: 140 ng/L (Swanson et al., 1990) and 30 to 3,484μg/kg (Bruhn and Franke, 1985). Iodine is important in human nutrition; however, excess as well as deficiency is of concern (Lacroix and Wong, 1980). Its deficiency may lead to retardation of fetus development, death, or weak neonates with hyper-plastic thyroid goiter, whereas excessive amounts may lead to thyrotoxicosis or hypersensitivity in the general population (Paulíková et al., 2002). It is thus important to investigate the implication of using iodide as an LP-s substrate on the health of the milk-drinking population and on the quality of milk and milk products.

Back to Article Outline

Conclusions 

The study generally confirms the dependency of the antibacterial effects of LP systems on substrate composition, microbial composition, and microbial load. The H2O2-dependent oxidation of I− provides the most effective system compared with SCN− or garlic extract as an electron donor. The negative synergistic effects of I− and SCN− on the performance of LP system demonstrated in this study imply the dependence of the levels of use of I− for the activation of the LP system on the natural SCN− concentration in milk. However, despite the greater levels of SCN− in milk and its interference with LP+I−+H2O2, the latter is a more effective system in milk preservation than LP+SCN−+H2O2. The current observation opens further avenues for the establishment of minimum and optimum levels of use of I− in the LP system, its health implications for milk consumers, and the general acceptability of LP+I−+H2O2 to activate milk.

Back to Article Outline

Supplementary data 

Download file

download text

Interpretive summary.

Back to Article Outline

References 

  1. Barraquio VL, Resubal LE, Bantoc IBM, Almazan EN. Preservation of raw milk with lactoperoxidase/hydrogen peroxide/thiocyanate system, hydrogen peroxide system and by refrigeration. Asia Life Sci. 1994;3:1–10
  2. Barrett NE, Grandison AS, Lewis MJ. Contribution of the lactoperoxidase system to the keeping quality of pasteurized milk. J. Dairy Res. 1999;66:73–80
  3. Björck L, Claesson O, Schulthess W. The lactoperoxidase/thiocyanate/hydrogen peroxide system as a temporary preservative for raw milk in developing countries. Milchwissenschaft. 1979;34:726–729
  4. Bosch EH, Van Doorne H, De Vries S. The lactoperoxi-dase system: The influence of iodide and the chemical and antimicrobial stability over the period of about 18 months. J. Appl. Microbiol. 2000;89:215–224
  5. Bruhn JC, Franke AA. Iodine in cows’ milk produced in the USA in 1980–1981. J. Food Prot. 1985;48:397–399
  6. Cailliez-Grimal C, Revo-Juvenelles AM, Linder M, Milli-ere JB. Antimicrobial activity of the glucose/glucose-oxidase and the lactoperoxidase system (SCN−) modified by I− or IO3− anion. Milchwissenschaft. 2002;57:656–660
  7. Chakraborty BK, Chaudry SS, Alex KA, Jacob G, Soni GJ. Application of the lactoperoxidase system for preserving buffalo milk produced in Indian villages. Milchwissenschaft. 1986;41:16–19
  8. De Wit JN, Van Hooydonk ACM. Structure, functions and applications of lactoperoxidase in natural antimicrobial systems. Neth. Milk Dairy J. 1996;50:227–244
  9. Ewais SM, Hefnawy AS, El-Salam A. Utilisation of lactoeproxidase system in preservation of raw milk under local condition. Egyptian J. Dairy Sci. 1985;13:1–7
  10. FAO/WHO. Guidelines for the preservation of raw milk by use of the lactoperoxidase system. Report of joint FAO/WHO committee on food additives CAC/GL 13. Rome, Italy: FAO; 1991;
  11. Furtmüller PG, Jantschko W, Regelsberger G, Jakopitsch C, Arnhold J, Obinger C. Reaction of lactoperoxidase compound I with halides and thiocyanate. Biochemistry. 2002;41:11895–11900
  12. Harding F. Hygienic quality. In:  Harding F editors. Milk Quality. London, UK: Blackie Academic and Professional; 1995;p. 40–59
  13. Jandal JM. Lactoperoxidase/garlic extract/ethanol system in the preservation of buffalo milk. Buffalo J. 1998;1:95–101
  14. Kamau DN, Doores S, Pruitt KM. Antibacterial activity of the lactoperoxidase system against Listeria monocytogenes and Staphylococcus aureus in milk. J. Food Prot. 1990;53:1010–1078
  15. Kirk AB, Martinelango PK, Tian K, Dutta A, Smith EE, Dasgupta PK. Percholate and iodide in dairy and breast milk. Environ. Sci. Technol. 2005;39:2011–2017
  16. Klebanoff SJ. Iodination of bacteria: A bactericidal mechanism. J. Exp. Med. 1967;126:1063–1078
  17. Klebanoff SJ. Myeloperoxidase-halide-hydrogen peroxide antibacterial system. J. Bacteriol. 1968;95:2131–2138
  18. Korhonen H. Untersuchungen zur Bacterizidie der Milch Im-munisierung del bovinen Milchdruse. Finnish J. Dairy Sci. 1973;32:1–15
  19. Korhonen H. The Lactoperoxidase system in mastitic milk FAO Lactoperoxidase Expert Meeting, Cape Town, South Africa. Rome, Italy: FAO/WHO; 2004;
  20. Lacroix DE, Wong NP. Determination of iodide in milk using the iodide specific ion electrode and its application to market milk samples. J. Food Prot. 1980;43:672–674
  21. Majerus PMC, Courtois PAP. Susceptibility of Candida albicans to peroxide catalysed oxidation of thiocynate, iodide and bromide. J. Biol. Buccale. 1992;20:241–245
  22. Marks NE, Grandison AS, Lewis MJ. Challenge testing of the lactoperoxidase system in pasteurised milk. J. Appl. Microbiol. 2001;91:735–741
  23. Mickelson MN. Antibacterial action of lactoperoxidase-thiocy-anate-hydrogen peroxide on Streptococcus agalactiae. Appl. Environ. Microbiol. 1979;38:821–826
  24. Paulíková I, Kovác G, Bíreś J, Paulík Ś, Seidel H, Nagy O. Iodine toxicity in ruminants. Vet. Med. (Czech). 2002;47:343–350
  25. Pruitt KM, Reiter B. Biochemistry of peroxidase system: Antimicrobial effects. In:  Pruitt KM,  Tenovuo J editor. The Lactoperoxidase System: Chemistry and Biological Significance. New York, NY: Marcel Dekker Inc.; 1985;p. 143–178
  26. Reiter B, Harnulv BG. Lactoperoxidase antibacterial system: Natural occurrence, biological functions and practical applications. J. Food Prot. 1984;47:724–732
  27. Revol-Junelles A-M, Boussouel N, Ramet J-P, Milliere J-B. Antibacterial activities of lactoperoxidase systems (LPS) modified by I− and IO3− anions. Milchwissenschaft. 2001;56:329–332
  28. Rossi EA, De Oliveira JS. Effect of thiocyanate concentration on the effectiveness of the lactoperoxidase system activated in raw milk. Aliment. Nutr. 1993-1994;5:49–56
  29. Shin K, Hayasawa H, Lonnerdal B. Inhibition of Esche-richia coli respiratory enzymes by the lactoperoxidase-hydrogen peroxide-thiocyanate antimicrobial system. J. Appl. Microbiol. 2001;90:489–493
  30. Siva CV, Upadhyay KG, Sannabhadti SS. Lactoperox-idase/thiocyanate/H2O2 system, its uses and implication in manufacture of dairy products. Indian Dairyman. 1991;43:240–246
  31. Swanson EW, Miller JK, Mueller FJ, Patton CS, Bacon JA, Ramsey N. Iodine in milk and meat of dairy cows fed different amounts of potassium iodide or ethylenediamine dihydroiodide. J. Dairy Sci. 1990;73:398–405
  32. Thomas EL. Products of lactoperoxidase catalyzed oxidation of thiocyanate and halides. In:  Pruitt KM,  Tenovuo J editor. The Lactoperoxidase System: Chemistry and Biological Significance. New York, NY: Marcel Dekker Inc.; 1985;p. 179–202
  33. Thomas EL, Aune TM. Cofactor role of iodide in peroxidase antimicrobial action against Escherichia coli. Antimicrob. Agents Chemother. 1978;13:1000–1005
  34. Wolfson LM, Sumner SS. Antibacterial activity of the lactoperoxidase system: A review. J. Food Prot. 1993;56:887–892
  35. Wolfson LM, Sumner SS. Antibacterial activity of the lactoperoxidase system against Salmonella typhimirium in trypticase soy broth in the presence and absence of a heat treatment. J. Food Prot. 1994;57:365–368

PII: S0022-0302(08)71130-5

doi:10.3168/jds.2007-0322

Journal of Dairy Science
Volume 91, Issue 7 , Pages 2566-2574, July 2008

Article Tools

* Image Usage & Resolution