Effective inactivation of food pathogens Listeria monocytogenes and Salmonella enterica by combined treatment of hypericin-based photosensitization and high power pulsed light

Authors


Zivile Luksiene, Vilnius university, Institute Applied Sciences, Sauletekio 10, 10223 Vilnius, Lithuania. E-mail: zivile.luksiene@tmi.vu.lt

Abstract

Aims:  The aim of this study was to evaluate the inactivation efficiency of Listeria monocytogenes ATCL3C 7644 and Salmonella enterica serovar Typhimurium strain DS88 by combined treatment of hypericin (Hyp)-based photosensitization and high power pulsed light (HPPL).

Methods and Results:  Cells were incubated with Hyp (1 × 10−5 or 1 × 10−7 mol l−1) in PBS and illuminated with a light λ = 585 nm. For the combined treatment, bacteria were, after photosensitization, exposed to 350 pulses of HPPL (UV light dose = 0·023 J cm−2). Fluorescence measurements were performed to evaluate optimal time for cell–Hyp interaction.

Results indicate that Hyp tends to bind both Listeria and Salmonella. After photosensitization treatment, Listeria population was reduced 7 log, whereas Salmonella was inactivated just 1 log. Electron photomicrograps of Salmonella and Listeria confirmed that photosensitization induced total collapse of the Listeria cell wall, but not that of Salmonella. After combined photosensitization–HPPL treatment, the population of Listeria was diminished by 7 log and Salmonella by 6·7 log.

Conclusions: Listeria can be effectively inactivated by Hyp-based photosensitization (7 log), whereas Salmonella is more resistant to photosensitization and can be inactivated just by 1 log in vitro. Combined treatment of photosensitization and pulsed light inactivates effectively (6·7–7 log) both the Gram-positive and the more resistant to photosensitization Gram-negative bacteria.

Significance and Impact of the Study:  A new approach to combat Gram-positive and Gram-negative bacteria is proposed, combining photosensitization with high power pulsed light.

Introduction

Food-borne diseases are still a significant problem in the world. It is estimated that more than 1·5 billion cases of food-borne illness and more than 3 million deaths from these diseases occur every year (WHO 2007). The main food pathogens Listeria monocytogenes and Salmonella enterica cause serious food-borne infections, meningitis, bacteraemia, septicaemia, miscarriage and stillbirths (Kendall 2003). It indicates that existing conventional chemical antibacterial treatments used to combat food pathogens are not effective enough. Moreover, the usage of chemical sanitizers is suspected to be environmentally unsound as it is associated with occupational and operational hazards and is potentially harmful for humans (Bintsis et al. 2000). Consequently, the development of nonthermal and nonchemical sterilization methods offers prospects. Well-known emerging antibacterial technologies such as ionizing radiation, high power pulsed light, ultraviolet light, atmospheric pressure plasma treatment and high frequency ultrasound have the ability to inactivate micro-organisms at ambient or near-ambient temperatures, thereby avoiding the deleterious effects that heat has on flavour, colour and nutrient value of foods (Aguilar-Rosas et al. 2007; Odriozola-Serrano et al. 2009; Oms-Oliu et al. 2010). The novel approach to control pathogens in nonthermal and nonchemical ways is based on photosensitization. Effective inactivation of main food pathogens by photosensitization in vitro and on the food-related packaging surface has been demonstrated in several studies (Buchovec et al. 2009, 2010; Le Marc et al. 2009; Luksiene et al. 2009). Moreover, photosensitization exhibited pronounced efficiency in inhibiting bacterial biofilms, spores, yeasts and fungi in vitro (Luksiene et al. 2010a,b; Luksiene and Paskeviciute 2011). Interesting and promising data were obtained when chlorophyllin-based photosensitization was used to control harmful and pathogenic bacteria, yeasts and fungi on the surface of strawberries (Luksiene and Paskeviciute 2011). Comparative analysis indicated that photosensitization was a more effective antimicrobial treatment to control food packaging contamination than the traditional surface treatment with hypochlorite (Luksiene et al. 2010a).

Hypericum perforatum L. (St John’s Wort, Hypericaceae) is one of the most consumed medicinal plants in the world. It was shown to possess antidepressant (Cervo et al. 2002), antioxidant (Conforti et al. 2002), anti-inflammatory (Kumar et al. 2001), antifungal (Fenner et al. 2005), antiviral (Jacobson et al. 2001), antitumor, anti-HIV (Darbinian-Sarkissian et al. 2006) and antibacterial activities (Mazandarani et al. 2007). In addition, the main naphtodianthrone hypericin (Hyp) exhibits photosensitizing properties (Schey et al. 2000). However, just a few studies have been published on the evaluation of antimicrobial efficiency of Hyp-based photosensitization against food pathogens. Traditionally, different susceptibility of Gram-positive and Gram-negative bacteria to photosensitization was confirmed in these studies (Jankowski et al. 2005; Engelhardt et al. 2010). Thus, to increase the inactivation of Gram-negative bacteria, it seems promising to combine Hyp-based photosensitization with high power pulsed light, which in 2000 was approved by the USA Food and Drug Administration as save technology for food surface decontamination (FDA 2000).

Thus, this study is focused on the possibility to inactivate Gram-positive and Gram-negative food-borne pathogens (L. monocytogenes and Salm. enterica) by combining Hyp-based photosensitization with high power pulsed light in vitro.

Materials and methods

Chemicals

Hypericin was synthesized and donated by Dr P. Adomėnas from the Institute of Applied Sciences of Vilnius University. A stock solution of 1 × 10−2 mol l−1 Hyp was prepared in dimethyl sulphoxide (5 mg ml−1) and stored at −20°C in the dark. Two concentrations of Hyp (1 × 10−7 and 1 × 10−5 mol l−1) were used in the experiments. Appropriate final concentrations of Hyp were prepared by further dilution of a stock solution in 0·1 mol l−1 phosphate-buffered saline (PBS, pH 7·2).

Bacterial cultures and growth conditions

Two food pathogens were used for experiments: L. monocytogenes ATCL3C 7644 (3rd passage of ATCC7644-test organism) was kindly provided by the National Veterinary Laboratory (Vilnius, Lithuania), and Salm. enterica serovar Typhimurium strain DS88 [SL5676 SmR (pLM32)], resistant to tetracycline, was kindly provided by Prof. D.H. Bamford (University of Helsinki, Finland). All bacteria were maintained at 37°C for 24 h onto Luria-Bertani Agar (LBA; Liofilchem, Roseto degli Abruzzi, Italy). Listeria and Salmonella cultures were grown overnight (∼16 h) at 37°C in 20 ml of tryptone soya medium supplemented with 0·6% yeast extract (TSYE) (Liofilchem) and in 20 ml of Luria-Bertani medium (LB; Liofilchem), respectively, with agitation of 120 rev min−1 (Environmental Shaker – Incubator ES – 20; Biosan, Riga, Latvia).

Afterwards, Listeria and Salmonella bacterial cultures were diluted 20 times by the fresh TSYE and LB medium (OD540 = 0·164), respectively, and grown at 37°C in a shaker (120 rev min−1) to the mid-log phase (∼1·16 × 109 colony-forming units (CFU ml−1), OD540 = 0·9 for Listeria; ∼5 × 108 CFU ml−1, OD540 = 1·3 for Salmonella). Bacterial optical density was determined in a 10·01-mm cuvette at λ = 540 nm (Heλios Gamma & Delta spectrophotometers; ThermoSpectronic, Waltham, MA, USA). Cells were then harvested by centrifugation (10 min, 3420 g) (Mikro 200; Hettich Zentrifugen, Tuttlingen, Germany) and resuspended in a 1 ml of 0·1 mol l−1 PBS (pH 7·2) to 5·8 × 109 CFU ml−1 final concentration of Listeria and to 2·5 × 109 CFU ml−1 final concentration of Salmonella cells. These stock suspensions were PBS diluted to ∼1 × 10−7 CFU ml−1 and used for the further experiments.

Absorption and fluorescence measurements

Absorption spectrum of Hyp was recorded in a 10·01-mm cuvette using a Heλios Gamma spectrophotometer, while the fluorescence spectrum of Hyp was recorded with Perkin-Elmer model LS-55 fluorescence spectrophotometer (Beaconsfield, UK). The cell suspensions for the measurements of Hyp fluorescence were prepared as follows. Cells (∼1 × 10−7 CFU ml−1 in 0·1 mol l−1 PBS (pH 7·2)) with 1 × 10−7 or 1 × 10−5 mol l−1 Hyp concentration were incubated in the dark at 37°C for the indicated time. To evaluate the amount of cell-attached Hyp, 3 ml aliquots of bacterial suspension after incubation were centrifuged and resuspended in 0·1 mol l−1 PBS (pH 7·2). Supernatant and cells in PBS were used immediately for fluorescence measurements. Each sample contained 20 μl Triton X-100. Scan range parameters were as follows:

1. Excitation wavelength = 556 nm

2. Emission = 570–750 nm

3. Excitation slit = 9 nm

4. Emission slit = 5 nm

5. Scan speed (nm min−1) = 200.

The fluorescence data were analysed with Origin 7·5 software (OriginLab Corp., Northhampton, MA, USA).

Hyp-based photosensitization treatment

Aliquots (20 ml) of bacterial suspension (∼1 × 107 CFU ml−1 in 0·1 mol l−1 PBS (pH 7·2)) with appropriate concentration of Hyp (1 × 10−5 or 1 × 10−7 mol l−1) were incubated in the dark at 37°C. For the following experiments, the cells were incubated in the shaker (130 rev min−1) for various periods (5–60 min). Afterwards, 150 μl aliquots of bacterial suspension were withdrawn, placed into sterile flat bottom wells and exposed to light for various times (0–60 min). A LED-based light source for photosensitization experiments (λ = 585 nm; intensity = 3·84 mW cm−2) was constructed in our Institute. Light dose delivered to the sample was calculated as light intensity multiplied on irradiation time.

Combined treatment of Hyp-based photosensitization and high power pulsed light (HPPL)

The device for HPPL treatment constructed in our laboratory consisted of a chamber, a reflector with a flash lamp and a power supply unit. The illumination spectrum of the xenon lamp was broad (200–1000 nm) and had maximal emission at 260 nm (UV is the main antibacterial light). Pulse duration was t = 112 ms and pulse frequency 5 Hz. The power of each pulse ranged from 0·07 to 0·9 MW. For combined treatment, 100 μl of photosensitization-treated bacterial cells was surface inoculated on the LBA plates and exposed to 350 pulses of high power pulsed light (UV light dose 0·023 J cm−2). UV light energy density was 0·324 mJ cm−2, frequency = 5Hz and duration = 70 s.

Bacterial cell survival assay

The antibacterial effect of photosensitization and combined treatment on Listeria and Salmonella cells was evaluated by the spread plate method. Thus, 100 μl of appropriate dilutions of bacterial test culture after treatment in vitro was surface inoculated on the LBA plates. Afterwards, the bacteria were kept in the thermostat for 24 h at 37°C. The surviving cell populations were enumerated and expressed by log10 (CFU ml−1).

Scanning electron microscopy (SEM)

The effect of Hyp-based photosensitization on the morphology of Listeria and Salmonella was examined by SEM. Bacterial suspensions (1 × 107 CFU ml−1) in 0·1 mol l−1 PBS were incubated in the dark for the appropriate time at 37°C with 1 × 10−5 or 1 × 10−7 mol l−1 Hyp and afterwards illuminated with light of 585 nm wavelength. In the next step, the samples consisting of 20 μl of bacterial suspension were withdrawn, transferred to aluminium stubs, air-dried and sputter coated with 15-nm gold layer using an Q150T ES (Quorum Technologies, Ashford, Kent, UK). The scanning was performed with an Apollo 300 scanning electron microscope at an accelerating voltage of 20 kV.

Statistics

The experiments were carried out in triplicate (from different inocula) for each set of exposure. A standard error was estimated for every experimental point and marked in a figure as an error bar. Sometimes the bars were too small to be visible. The data were analysed with Origin 7·5 software (OriginLab). The significance of both irradiation time and the type of treatment on bacterial survival was assessed by the analysis of variance (anova) model with the Bonferroni post hoc test. A value of P < 0·05 was considered as significant.

Results

Fluorescence measurements of cell-associated hypericin

Data presented in Fig. 1 show chemical formula of hypericin, its absorption and fluorescence spectra. It is clear that the main absorption band of Hyp is about 590 nm, and the main fluorescence peaks are at 602 and 651 nm.

Figure 1.

 Chemical formula, absorption (----) and fluorescence (–––) spectrums of 1 × 10−5 mol l−1 Hyp prepared in dimethyl sulphoxide.

Analysis of fluorescence emission spectra of Hyp incubated with cells indicates that Hyp in bacterial suspension tends to bind both bacteria: 0- to 60-min incubation of bacteria with Hyp and following washing with PBS indicate that part of Hyp was associated both with Listeria and Salmonella (Figs 2, 3). It is interesting to note that the fluorescence intensity of cell-associated Hyp in Listeria cells grows rapidly within the first 5 min of incubation. A further increase of the incubation time (10–60 min) did not affect the cell-associated Hyp fluorescence intensity (Fig. 2). Similarly, data presented in Fig. 3 demonstrate that about 70% of Hyp binds to Salmonella already after a few minutes of incubation and then the amount of cell-associated photosensitizer just slowly increases.

Figure 2.

 Fluorescence of cell-associated hypericin as function of incubation time: inserted fluorescence spectra of Listeria monocytogenes ATCL3C 7644 cells incubated with 1 × 10−7 mol l−1 Hyp for different time interval: (–––) cell fluorescence without Hyp, (–·–·–) 5-min incubation, (–··–··) 10-min incubation, (----) 20-min incubation, (––··––) 40-min incubation, (······) 60-min incubation.

Figure 3.

 Fluorescence of cell-associated hypericin as function of incubation time: inserted fluorescence spectra of Salmonella enterica DS88 cells incubated with 1 × 10−5 mol l−1 Hyp for different time intervals: (–––) cell fluorescence without Hyp, (–·–·–) 5-min incubation, (–··–··) 10-min incubation, (----) 20-min incubation, (––··––) 40-min incubation, (······) 60-min incubation.

Inactivation of Listeria and Salmonella by Hyp-based photosensitization in vitro

Data obtained indicate that used Hyp concentrations (1 × 10−5 or 1 × 10−7 mol l−1) were not toxic to the investigated bacteria because no effect on their viability was found in the dark (data not shown). Just incubation of bacteria with Hyp and following illumination with 585 nm light resulted in the reduction of bacterial count. Data presented in Fig. 4 reveal that Listeria is very susceptible to Hyp-based photosensitization. Already 5-min incubation and 13·8 J cm−2 illumination dose lead to about 5 log reduction of Listeria (P < 0·05). Extending the incubation time to 10 min increases the inactivation of Listeria to more than 6 log (P < 0·05). After 60-min incubation with 1 × 10−7 mol l−1 Hyp and following 9·22 J cm−2 illumination, it can be inactivated by 7 log (P < 0·05).

Figure 4.

 Photoinactivation of Listeria monocytogenes ATCL3C 7644 incubated with 1 × 10−7 mol l−1 Hyp as a function of illumination dose: (•) control, (▲) 5-min incubation, (□) 10-min incubation and (♦) 60-min incubation. Every point is the average of 3–6 experiments, and error bars sometimes are too small to be more visible.

In contrast, Salmonella is very resistant to Hyp-based photosensitization: 60-min incubation with 1 × 10−7 mol l−1 Hyp and following 13·8 J cm−2 illumination had no effect on Salmonella growth (data do not differ from the control), while the incubation with higher Hyp concentration (1 × 10−5 mol l−1) and following illumination diminished cell population just by 1 log (Fig. 5). It is important to note that the increase of incubation time with Hyp from 5 to 60 min had a negligible effect on the inactivation of Salmonella. Data presented in Fig. 6 show the morphological changes induced immediately after Hyp-based photosensitization in Listeria and Salmonella (Figs 4, 5). It is obvious that Listeria cells immediately after Hyp-based photosensitization start surface bleb formation, whereas most of Salmonella were not damaged at all.

Figure 5.

 Photoinactivation of Salmonella enterica DS88 incubated with 1 × 10−5 mol l−1 Hyp as a function of illumination dose: (•) control, (▲) 40-min incubation and (□) 60-min incubation. Every point is the average of 3–6 experiments, and error bars sometimes are too small to be more visible.

Figure 6.

 Scanning electron microscopy analysis of Listeria monocytogenes ATCL3C 7644 and Salmonella enterica DS88 cells after Hyp-based photosensitization in comparison with control, not treated ones: (a) untreated Listeria cells; (b) Listeria cells incubated with 1 × 10−7 mol l−1 Hyp for 10 min and illuminated by 4·6 J cm−2 light dose; (c) untreated Salmonella cells; (d) Salmonella incubated with 1 × 10−5 mol l−1 Hyp for 60 min and illuminated by 13·8 J cm−2 light dose.

Inactivation of Listeria and Salmonella by combined treatment of Hyp-based photosensitization and high power pulsed light

Results presented in Fig. 7 indicate that at special experimental conditions, when a lower photosensitization dose was used (bacteria were incubated 5 min with 1 × 10−7 mol l−1 Hyp and after illuminated with light 585 nm, 2·3 J cm−2), the population of Listeria was diminished by 2·2 log. Reduced high power pulsed light dose (350 pulses, total UV dose 0·023 J cm−2) diminished the population of Listeria by 5·3 log. Following the combination of the lowered dose of photosensitization with the lowered dose of HPPL reduced the survival fraction of Listeria from 7 log in control to 0 log in treated samples (P < 0·05). Gram-negative Salmonella were more resistant to photosensitization: even incubation of cells for 60 min with 1 × 10−5 mol l−1 Hyp and illumination with light 585 nm (13·8 J cm−2) reduced the bacterial population by just 1 log. Reduced high power pulsed light dose diminished the bacterial Salmonella population by 5 log. After combined treatment of photosensitization and HPPL, a remarkable reduction of Salmonella counts (6·7 log) was found (P < 0·05). Further increase in pulsed light dose diminishes Salmonella population to undetectable level (data not shown).

Figure 7.

 Inactivation of bacteria by photosensitization and high power pulsed light (HPPL): inline image control; inline image photoinactivation of Listeria monocytogenes ATCL3C 7644 (1 × 10−7 mol l−1 Hyp, 5-min incubation, 2·3 J cm−2 illumination dose) and Salmonella enterica DS88 (1 × 10−5 mol l−1 Hyp, 60-min incubation, 13·8 J cm−2 illumination dose); inline image treatment by HPPL (0·023 J cm−2); inline image combined treatment of photosensitization and HPPL (0·023 J cm−2). Every point is the average of 3–6 experiments.

Discussion

Much work has been performed on the photoinactivation of food pathogens Bacillus cereus, Salmonella Typhimurium, Listeria monocytogenes, their spores and biofilms, using endogenous photosensitizers produced by micro-organisms metabolizing colourless and odourless aminolevulinic acid to endogenous porphyrins (Nitzan et al. 2004; Buchovec et al. 2009, 2010; Le Marc et al. 2009; Luksiene et al. 2009). The other interesting approach to control food pathogens was developed in our previous studies (Luksiene et al. 2010a,b; Luksiene and Paskeviciute 2011), where chlorophyllin (food additive E140) distinguished for its antimutagenic and anticancerogenic properties (Surh 1998) was evaluated as a compound with pronounced photosensitizing activity. In further studies, chlorophyllin-based photosensitization was used to control food pathogens on the surface of food packaging material (Luksiene et al. 2010a), fruits and vegetables (Paskeviciute and Luksiene 2009). Taking into account that photosensitization as nonthermal environmentally friendly antibacterial treatment was patented as an effective tool for surface sterilization (Luksiene and Buchovec 2009), it was interesting to expand the list of promising plant-based photosensitizers and to evaluate photosensitizing properties of hypericin.

Analysis of fluorescence emission spectra of cell-associated Hyp reveals that Hyp tends to bind Gram-positive as well as Gram-negative bacteria. It must be noted that interaction of Hyp with both bacteria was very fast and did not require more than 5-min incubation time to reach saturation (Figs 2–3). However, the photoinactivation of Salmonella cells perhaps is restricted by LPS (lipoprotein polysaccharide) barrier through which singlet oxygen and/or oxyradicals (generated outside the cells) must pass to interact with a membrane or cytoplasmic components (Malik et al. 1992). Nitzan and Pechatnikov (2011) suggested that neutral photosensitizers are not effective against Gram-negative bacteria because of their failure to attach to the bacteria’s inner membrane after penetration through the outer membrane. Therefore, reactive oxygen species generated outside the cell wall cannot affect Gram-negative bacteria because of their distance from the inner membrane.

Data obtained in this study confirmed that photoactivated hypericin remarkably inactivated Gram-positive Listeria but had an insignificant effect on the Gram-negative Salmonella. The population of Listeria cells was diminished by 7 log using 1 × 10−7 mol l−1 Hyp and 9·22 J cm−2 illumination dose, where as the population of Gram-negative Salmonella decreased by just 1 log after incubation with 1 × 10−5 mol l−1 Hyp and illumination with 13·8 J cm−2 light dose. These results are in agreement with others studies. For example, Engelhardt et al. (2010) showed that incubation of Gram-positive Staph. aureus with 1 × 10−7 mol l−1 water-soluble Hyp for 5 min and illumination at 135 J cm−2 leads to the 4–5 log reduction in bacterial count. Staph. sobrinus can be inactivated by 2 log after 15-min incubation with 5 × 10−6–2 × 10−5 mol l−1 Hyp and 128 J cm−2 illumination dose (Lüthi et al. 2009). To the contrary, several Gram-negative bacteria (Shigella flexnery 1a, S. flexnery 2a and Escherichia coli K1 no. 959) were more resistant to Hyp-based photosensitization and were inactivated insignificantly (Jankowski et al. 2005).

The different susceptibilities of Listeria and Salmonella to photosensitization can be attributed to the particular structure of their cell wall (Demidova and Hamblin 2005). The cytoplasmic membrane in Gram-positive bacteria is surrounded by a porous layer composed of peptidoglycan and lipoteichoic acid. This structure allows easy interaction of photosensitizer with cell wall components and promotes their accumulation inside the cell. Therefore, photosensitizers whose molecular weight does not generally exceed 1500–1800 Da can readily diffuse to the inner plasma membrane of these bacteria (Jori et al. 2006). In contrast, the cell wall of Gram-negative bacteria consists of an inner and outer membrane with incorporated lipopolysaccharides directed outside. Moreover, the two membranes are separated by a peptidoglycan layer, which acts as a permeability barrier to many of the molecules from the external environment (Brovko 2010). Electron microphotographs of Listeria and Salmonella obtained after Hyp-based photosensitization confirm the intensive surface bleb formation in Listeria, suggesting breakage of contact between the cell wall and the membrane. No significant structural changes were observed in treated Salmonella cells.

It is known that combination of several antimicrobial tools (concept of hurdle technologies) can enhance the microbial inactivation and allow the use of lower individual treatment intensities, which do not cause adverse or sometimes destructive changes of surface and matrix properties (Wood and Bruhn 2000). Combined treatment is more effective because a number of sublethal stresses on a microbial cell cause the micro-organism to expend energy to overcome the stressful environment, potentially leading to metabolic exhaustion and death (Leistner 2000). High power pulsed light is a nonthermal food preservation technology that proposes decontamination of surfaces by intense and short duration (microseconds) pulses of broad spectrum light (200–1000 nm) and was approved for food surface decontamination by the Food and Drug Administration (FDA 2000). It offers effective inactivation of pathogens by destructive action on their DNA (Miller et al. 1999), is mercury free and has limited energy costs and short exposure time (Elmnasser et al. 2007). Moreover, no bacterial resistance to this treatment was observed. As Hyp-based photosensitization alone failed to inactivate Salmonella cells efficiently, combined treatment with HPPL was performed. When photosensitization and HPPL treatments were combined, a remarkable reduction in viable count of both bacteria was observed: Listeria was inactivated by 7 log, whereas population of resistant to photosensitization Salmonella was reduced by 6·7 log.

Conclusion

In conclusion, Hyp-based photosensitization inactivated significantly Gram-positive Listeria, but not Gram-negative Salmonella. In addition, rather low (1 log) inactivation of Salmonella requires higher hypericin concentration, longer incubation time and higher illumination dose, whereas Listeria being more sensitive to this treatment was killed by 7 log using milder experimental conditions.

Remarkable enhancement of the inactivation of Gram-negative Salmonella can be obtained combining Hyp-based photosensitization and high power pulsed light. Thus, our results confirm that combination of these two treatments might be an effective tool to combat both Gram-positive and Gram-negative bacteria in an effective and environmentally friendly way.

Ancillary