1. Introduction
Various prevention strategies against contaminations of food produces are relevant among the European Union. Consequently, a vast number of now well-established practices and concepts such as good agricultural practice, good manufacturing practice and hazard analysis (HACCP) were designed to minimize the risk of contaminations with pathogenic bacteria and food spoilers. However, the whole value chain of fresh and fresh-cut produce harbors numerous sources of contaminations such as raw material, processing tools, surfaces with food contact, or insufficiently trained employees. Once contaminated it is very likely that a produce carries the contamination along the whole production chain. Therefore, the in-company process hygiene is under special surveillance and new, innovative techniques recently gain attention [
1]. Nowadays the situation is aggravated by the fact that inconsistent food safety standards were developed into economic, political, and social problems. The reason for that confusingly amount of sanitation standards is especially due to the presence of new microbial loads such as altered focuses on accurately described bacteria or bacteria that conquered new habitats, which might alter locally. However, the main body of sterilization and disinfection methods based on wet chemical processes (peracetic acid, hydrogen peroxide, chlorine dioxide, chlorine, active oxygen, ammonium compounds, isopropanol, phosphoric acid, formic acid or nitric acid), which are highly effective but also increase the development of new resistances [
1]. Certainly, cleaning and disinfection cycles that clean the production environment in terms of the removal of organic (proteins, lipids, carbohydrates) and inorganic (salts) contaminations were integrated into every process chain. Nevertheless, it will be a future challenge to develop new maybe unified safety standards, which base on alternative techniques and may help to overcome those economic, political, and social barriers. The complex geometries of food processing plants made from various materials (stainless steel, glass, polymers, rubber and polytetrafluoroethylene (PTFE)) make effective cleaning difficult. For instance, empirically designed routine processes can lead to incomplete microbial decontamination, so that pathogenic microorganisms such as
Listeria monocytogenes can survive on surfaces and contaminate the further process chain [
2,
3,
4]. Under these circumstances, the risk of a microbial adhesion and biofilm formation is particularly high. Cross-contaminations, bacterial adhesion, and biofilm formation was avoided by an optimized choice of materials, which can be further achieved by efficient and sustainable agents with antimicrobial effects [
5,
6]. Manufacturers are therefore faced with a wide range of resources-consuming and cost-intensive challenges (environment, energy and water consumption, residues as well as storage and disposal etc.), which raised the demand for innovative process concepts that supplement or replace existing systems. When the food production is resumed after the heat decontamination step, the cooling time appears to be especially crucial. Certainly, active cooling is possible, but it is also energy consuming.
At boundaries between two materials with a different thermal expansion, e.g., at observation windows, heat decontamination induces tensions that may lead to cracks in the glass window. For instance, windows for optical sensors like in-line process photometers for in-line phase switch control may be such a weak point. Particularly, calcifications will become critical, if heat decontamination is combined with water vapor [
7]. When heat have been used for decontamination, the chamber where the decontamination takes place needs to be pre-heated. Due to the low heat transfer coefficient for the heat transfer between gaseous and solid phases and the high heat capacity of steel in contrast to ambient air, stainless steel part needs a long time to heat up or cool down. Therefore, colder plant parts lower the surrounding temperature, which may result in an insufficient decontamination rate, a pre-heating appears to be a very crucial step. Additionally, overpressure has the drawback to increase the mechanical load on machine parts in contact, which increases the frequencies of maintenance for those plants [
7]. Therefore, all machine components have to be pressure-proofed. If heat is used these drawbacks lack when PPA (plasma-processed air) is applied. The use of PTW (plasma-treated water) enables the user to apply the shearing forces of the water for cleaning; which is not possible in decontamination steps using gaseous compounds. Additionally, when using heat, special attention must also be paid to the work safety of the user, i.e. to prevent burns. This is not necessary with the plasma technology presented here where the work safety requirements are easy to implement. Despite many advantages, plasma technology is an individual solution for special requirements and can currently be seen as a niche product.
Several studies prove the efficiency of non-thermal atmospheric pressure plasma applications for a surface decontamination. The resulting reactive species (including oxygen and hydroxyl radicals, nitrogen oxides and other oxidizing species) proved to be highly effective against many yeasts, viruses and bacteria (including spores). The results have been found on various surfaces such as glass, polypropylene (PP), paper or stainless steel [
8,
9]. The exact inactivation mechanism of the different plasma sources has already been the subject of several reviews [
10,
11]. The removal of biomolecules such as proteins and peptides, which are also required for purification, is also possible by plasma application [
12].
For the indirect plasma process based on PLexc microwave plasma, antimicrobial effects have already been demonstrated on both laboratory and industrial scale on abiotic surfaces (glass, plastic) and biological surfaces (fruit, vegetables, meat) [
13,
14,
15,
16,
17,
18]. It applies to both PPA and PTW.
Currently, synergistic effects of both methods have not yet been deeply investigated and are the object of this study. Our findings strongly suggest a serial application of the two methods. If the production process allows such an application, a PTW rinsing/washing step should be followed by a PPA- drying step, subsequently. Many processes in the food industry fulfill these requirements, which give the PTW/PPA processes a wide range for their industrial application.
4. Discussion
The production of food and other produce (e.g., pharmaceuticals or cosmetics) requires a high hygienic standard, which ensures a high quality and a safe usage of these products. A factor that gains importance when the produce must fulfill high microbiologic demands. Especially in the food industry, poor or inadequate hygienic surroundings lead to health infections and foodborne diseases as well as to high production losses. The high nutrient content and the relatively high pH value (pH 5.5 to 7) of some fruits and vegetables may increase the growth of pathogens [
37]. Usually, bacterial and fungal contaminations of foods or their processing environments plus interrupted freezer chains are responsible for foodborne diseases, production and quality losses. Typical human pathogens found in food are
L. monocytogenes,
Salmonella sp.,
Clostridium sp.,
E. coli,
S. aureus, and
Aspergillus sp. Many molds (e.g.,
Fusarium sp.), oomycetes,
Xanthomonas sp.,
Erwinia sp. and
Pseudomonas sp. may parasitically live and found in any phyto-pathogen. Therefore, the inactivation of human and phyto-pathogens is the foundation for a safe consumption of any product and, thus, radiates (effects downstream) in many social and economic areas. Our chosen bacteria (
P. carotovorum,
L. monocytogenes and
E. coli) and the endospores of
B. atrophaeus cover a broad spectrum of possible food contaminations. The picked microorganisms may be responsible for human or plant diseases and for the formation of biofilms.
In the field of conventional cleaning methods, a distinction has been made between COP (cleaning out of place) and CIP (cleaning in place). The COP process includes high-pressure cleaning in combination with foam cleaning supported by a manual cleaning of individual plant components. It is a step which is irregularly necessary. The foam used during cleaning is rinsed off with water after a defined contact time. The CIP process is often an automated cleaning process in a closed circuit, also known as recirculation cleaning. Water, steam, alkalis, acids or enzymatic solutions have been used as cleaning agents. The third possibility in the food industry is “washing in place”, which is the cleaning of different components in dipping baths or dishwashers. Furthermore, UV radiation has often been used to ensure disinfection after cleaning. Other decontamination techniques including ozone that is a potent antimicrobial agent comprise the drawback of material damage when the concentration of the agent is too high. Regular cleaning and disinfection cycles of the production plants and rooms for contaminants removal means high costs, less productivity, maybe highly polluted wastewater and material damage for the manufacturer. Hence, the food industry faces a wide range of challenges. Therefore, there is a great demand for innovative process concepts, which complements or replaces the existing systems with resource-saving, versatile and efficient technologies. Behind that background, the use of direct and indirect non-thermal plasma-based decontamination methods appears to be promising. The advantages of a decontamination agent that acts quickly, does not leave toxic residues on the food surface or in the exhaust gasses and an adjustable, product-specific process temperature compensate the disadvantage of high energy costs by the production of the PTW and PPA [
38]. The promising and easy integration of antimicrobial potent species, which have been directly provided by the plasma itself or from plasma-processed compounds upstream into a non-thermal treatment mode, makes non-thermal plasmas particularly attractive for the decontamination in food processing [
39]. Plasma processes are usually easy to use, inexpensive and suitable for the decontamination of products where high heat is not desired [
40]. The experimental set-up described in this publication for the PTW- and PPA-based inactivation of the presented microorganisms can exclude stresses such as temperature, pressure or radiation. However, PPA and PTW embrace reactive nitrogen species such as NO and NO
2 and HNO
2 and HNO
3, respectively, which may have a harmful impact on the produce’s surface. Nevertheless, only about 3% of used air has been processed and most of the water leaved untreated. If compared to their expected natural resistance and virulence, the presented antimicrobial inactivation (separate and combined) on PET-stripes will remain surprising. The different formation and composition of the cell wall and membrane of Gram-negative and Gram-positive bacteria can lead to a higher resistance of Gram-positive bacteria compared to Gram-negative bacteria to stress factors (e.g., plasma). Fungi (mold and yeasts), conidiospores and endospores can be even more resistant to such stresses [
10]. However, the results present a different bacterial behavior, the Gram-positive
L. monocytogenes has been inactivated in most cases with an additive or synergistic effect, no inhibitory case has been seen. On the other hand, the two Gram-negative bacteria
E. coli and
P. carotovorum showed a weaker inactivation where inhibitory cases have been also detected. A more general aspect for both types of PTW/ PPA treatment, which has been obtained for all bacteria, was the fact that an increased pre- and post-treatment time let to an increase in the decontamination rate. Another unexpected point was the inactivation kinetics of endospores of
B. atrophaeus. Commonly, endospores of bacteria are known as the most resistant microorganisms. However, in our experiments, the combination of all PTW-treatments with short and middle PPA treatments (5 s + 3 min up to 15 s+ 5 min) was synergistic. Only the five s + 1 min treatment shows also additive effects. Extension of the PPA treatment to 50 s combined with one, three and five min lead to more additive and inhibitory effects. A previous work showed promising inactivation results for different microorganisms on polymeric surfaces and plant tissue by PPA and PTW in single use [
13,
35,
41,
42].
There are many possible mechanisms induced by PTW/PPA treatment, which may underlie the observed results. However, the effective inactivation mechanisms, and any involved signal pathways induced by direct and indirect plasma treatment are still unidentified. A first aspect of inactivation could be an acidification [
43] as known for lactic acid. Causing a depression of the pH value inside a bacterial cell by ionization of the undissociated acidic molecules or a disorder of the substrate transport by an alteration of the cell membrane´s permeability may be addressed as a mechanism that cause cell inactivation. Additionally, pH values lower than 4.0 influence the growth of most foodborne bacteria negatively. Due to the chemical behavior of PTW and PPA, regardless whether applied separately or combined, an acidification on the specimen’s surface and therefore, in the direct microbial environment can be expected [
14,
35,
44,
45]. This expected acidification may lead to a similar inactivation mode as it has been observed for lactic acid.
Another reason for the inactivation of microorganisms by PTW and PPA can be the humidity in the PPA- and PTW-process, which is provided by the ambient/compressed air or the water itself. Maeda et al., (2003) have reported about the influence of moisture in air on bacterial inactivation [
46]. They observed a nearby dependency of microbiological inactivation on moisture in air for
E. coli. The investigated microorganism showed a maximum inactivation rate at 43% relative humidity as it occurs in laboratory/room air.
The results for E. coli K12 in the present work showed an increase of inactivation by increased treatment times for a separated and combined PTW and PPA treatment. However, an improvement of the detection limit would lead to more clarity about synergistic efficacy.
The importance of the moisture of a plasma gas for the inactivation of
Bacillus subtilis endospores and
Aspergillus niger conidiospores was also investigated by Muranyi et al., (2008) [
47]. These scientists showed reductions up to five lg within seven s at zero percent relative gas humidity and up to four lg at 80% gas humidity for the endospores. Their observation was that high amounts of gas humidity weaken the inactivation of plasma gasses. This could be an explanation for the decreasing inactivation effect, which was seen in our investigations for endospores of
B. atrophaeus. Longer combined treatment times that ensure a longer contact time between the PTW treated spores and the PPA let to more additive and inhibitory effects.
The third and maybe most important explanatory approach may result from the chemical composition of the PTW and the PPA. Behind that background, many antimicrobial efficient reactive oxygen and nitrogen species tending into the focus. Therefore, experimental investigations of the chemical composition of PPA and PTW are crucial to understand the biochemical reactions, which are responsible for the microbial inactivation. The chemical analysis showed that reactive oxygen species (ROS) and reactive nitrogen species (RNS) were detectable in both, PTW and PPA. The detected chemical species in PPA and PTW were OH, NO, NO
2 and NO, NO
2, respectively. The detection of hydrogen peroxide as well as peroxynitrite in PTW will be a part of further investigations. However, it is most probably that nitrite and nitrate are only the final products of complex reaction chains in PPA and PTW. Hydroxyl radicals (OH*) are well known as strong oxidizing agents that can damage microbial cells [
10,
43,
48]. The photo dissociation of ozone to oxygen and singlet oxygen may also be a possibility to generate hydroxyl radicals. Oxygen and singlet oxygen can react with water to hydroxyl radicals [
49]. Another highly antimicrobial effective chemical, hydrogen peroxide, can be formed by the reaction of two OH* [
50]. Due to the plasma set-up and used dry air (below 32% relative humidity), chemical reactions and species, which mainly based on RNS are predominantly expected in our work. Nitrogen monoxide (NO*) is generated by nitrogen and oxygen, both can be found in air. The generation of NO* further reacts to nitrogen dioxide (NO
2*) if oxygen (O
2) is available as reaction component. The long-living radicals NO* and NO
2* are known for their antimicrobial effectivity. In addition, NO* may react with ozone (O
3) to NO
2* and O
2. Another possible reaction of NO* and NO
2* can form dinitrogen trioxide (N
2O
3), which may react with O
3 to nitrogen trioxide radical (NO
3*) via dinitrogen hexoxide. Peroxynitrite, assumed to be highly antimicrobial, could be a product by the reaction of NO* with a superoxide radical [
43,
51]. The summarized reactions are theoretical possible in dry air after a plasma ignition. Reactions that are taking place in moisture surroundings may also be taken into account. High water amounts of treated products itself or wash water residues on abiotic surfaces may provide moisture reaction chambers. NO*, O
2 and water (H
2O) react to nitrite (NO
2−) and hydrogen (H
+). If instead of NO*, NO
2* reacts with O
2 and H
2O, H
+ and nitrate (NO
3− are the reaction products. The reaction of N
2O
3, which is generated in gas and gas/water phase, with water may additionally occur and result in NO
2− and H
+. The two radicals NO
3* and NO
2* form dinitrogen pentoxide in occurrence of H
2O [
43,
51]. Dinitrogen pentoxide can react with water to NO
3− and H
+.
However, in our experimental set-up, OH radicals have not been detected. This might be due to the absence of oxygen radicals because of a restrained energy uptake (from the microwave). Another possibility could be the generation of water clusters such as (H2O)NO or (H2O)OH.
Nevertheless, the solutions applied in the experiments showed a strong acidification, which might be a result of the long-living generated radicals NO* and NO2* that react with water to nitrous acid (HNO2) and nitric acid (HNO3). Usually HNO2 decays to H+ and NO2−, but a pH value beneath 2.75 could lead to a spontaneous forming of OH* and NO* radicals. Numerous of the postulated ions, radicals and molecules are known for their strong antimicrobial effects. In combination with the antimicrobial potential of a pH shift into acidic surroundings, the occurrence of such compounds, which are also influenced by the pH, may result in the observed microbial inactivation. Further investigations are necessary and will provide a better insight into the chemical and biochemical processes underlying the antimicrobial effects observed and assumed in the presented work.
An important aspect to mention here is the possible VBNC (Viable but nonculturable) status of many bacteria, which is assumed as an active bacterial response (surviving strategy) to unusual stresses. A process that is comparable to the sporulation of
Bacillus or
Clostridium as well as the dormancy in
Micrococcus luteus [
52,
53]. This status means that bacteria do not proliferate on/in nutrient media and therefore are not detectable in common proliferation assays. However, their metabolic activity, like the intracellular ATP level or membrane potential, remains unchanged [
54,
55]. Contrary to this, the synthesis of macromolecules and cell breath is limited [
56]. Furthermore, changes in fatty acid and protein composition of the double lipid layer (cytoplasmic membrane and cell wall) are possible [
57,
58,
59,
60]. Different triggers for VBNC status are described in literature. For instance, nutrient limitation (e.g., starvation in distilled water), salt and oxygen content, unfavorable incubation temperatures (e.g., low temperatures of 5 °C), shaking of growth containers, disinfectants and toxic metal ions (e.g., copper) are identified as VBNC inducers. Additionally, it also depends on the type of bacteria [
61,
62,
63,
64,
65,
66,
67,
68,
69,
70,
71]. The resuscitation of VBNC cells is possible but not imperatively required [
72,
73]. Resuscitation, which means that the bacteria gains back the proliferative ability, is triggered by different factors divided in three categories: (1) reversal of the negative environmental factor, e.g., increasing the incubation temperature to optimum, transferring bacteria to nutrient rich media [
63,
64]; (2) with substances that can repair cell damage, e.g., ammonium chloride or sulphate [
74,
75]; (3) addition of cell-own or cell-typical signaling molecules, e.g., pyruvate, RpF protein [
76,
77,
78,
79]. During VBNC status, bacteria retain their virulence or, with regaining their cultivability, have a pathogenic effect again. In addition, antibiotic resistance has been associated with the VBNC status; especially antibiotics that are directed against the reproductive system of the bacteria have been affected [
54,
72,
80,
81,
82]. Currently, literature describes for more than 60 species of bacteria the ability to change to VBNC status, including
E. coli and
L. monocytogenes [
83,
84]. Many human pathogens are included in this VBNC group, therefore, they should be recognized as a serious health risk to public health and future work should include investigations for protein synthesis (e.g., FISH) as well as Live/Dead assays (fluorescence) in combination with common proliferation assays.
However, the direct application of PTW and PPA seems to be possible, regardless of whether used separated or combined, on abiotic surfaces in contact with fresh produce. Currently, strong impacts to the fresh produce quality are not expected. First investigations with PPA on apple showed no impact on texture and appearance and a very low impact on odor. The use of PTW for lettuce washing also resulted in no negative effects on texture and color [
15,
85].
However, plants use radicals like ROS and RNS as a defense mechanism. This mechanism can have a more local or a wider, systemic appearance [
86,
87]. An important messenger in their defense signal pathways is NO* [
88,
89]. It has been shown by the literature that NO* is a crucial regulator of many physiological processes in plants, including stomatal closure and plant growth as well as their individual development [
90,
91,
92,
93,
94,
95,
96]. Furthermore, NO* can regulate these processes directly by governing gene transcription. Therefore, NO*-regulated genes are included in signal transduction, defense, cell death, transport, basic metabolism and production as well as degradation of ROS. These radicals react as activators or inhibitors of enzymes, ion channels, transcription factors and therefore regulate specific processes during stress situations in plants. In the presented plasma set-up and PTW/PPA treatment, the formation of RNS, especially NO*, occurs. The observed antimicrobial effects indicate the efficiency of generated RNS. The future perspective of using PTW and/or PPA in food industry is promising, due to low-priced consumables (compressed or ambient air, drinking water, current) and high grad of adaptability for different applications (production lines). The agents can be produced on-demand, and no special training for other chemicals is needed.