Inactivation Efficacies and Mechanisms of Gas Plasma and Plasma-Activated Water against Aspergillus flavus Spores and Biofilms: a Comparative Study

The production of mycotoxin-free food remains a challenge in both human and animal food chains. A. flavus, a mycotoxin-producing contaminant of economically important crops, was selected as the model microorganism to investigate the efficiency and mechanisms of ACP technology against fungal contaminants of food. Our study directly compares the antifungal properties of gas plasma (GP) and plasma-activated water (PAW) against fungi as well as reporting the effects of ACP treatment on biofilms produced by A. flavus.

ABSTRACT

Atmospheric cold plasma (ACP) treatment is an emerging food technology for product safety and quality retention, shelf-life extension, and sustainable processing. The activated chemical species of ACP can act rapidly against microorganisms without leaving chemical residues on food surfaces. The main objectives of this study were to investigate the efficiency and mechanisms of inactivation of fungal spores and biofilms by ACP and to understand the effects of the gas-mediated and liquid-mediated modes of application against important fungal contaminants. Aspergillus flavus was selected as the model microorganism. A. flavus spores were exposed to either gas plasma (GP) or plasma-activated water (PAW), whereas gas plasma alone was used to treat A. flavus biofilms. This study demonstrated that both GP and PAW treatments independently resulted in significant decreases of A. flavus metabolic activity and spore counts, with maximal reductions of 2.2 and 0.6 log₁₀ units for GP and PAW, respectively. The characterization of the reactive oxygen and nitrogen species in PAW and spore suspensions indicated that the concentration of secondary reactive species was an important factor influencing the antimicrobial activity of the treatment. The biofilm study showed that GP had detrimental effects on biofilm structure; however, the initial inoculum concentration prior to biofilm formation can be a crucial factor influencing the fungicidal effects of ACP.

IMPORTANCE The production of mycotoxin-free food remains a challenge in both human and animal food chains. A. flavus, a mycotoxin-producing contaminant of economically important crops, was selected as the model microorganism to investigate the efficiency and mechanisms of ACP technology against fungal contaminants of food. Our study directly compares the antifungal properties of gas plasma (GP) and plasma-activated water (PAW) against fungi as well as reporting the effects of ACP treatment on biofilms produced by A. flavus.

INTRODUCTION

Aspergillus flavus is a saprophytic soil fungus that infects and contaminates preharvest and postharvest seed crops. A. flavus has a worldwide distribution resulting from the production of airborne conidia, which easily disperse by air movement and insect vectors (1, 2). According to Klich (3), few fungi have an economic impact as broad as A. flavus, which is a pathogen of plants, animals, and insects, causing storage rot in numerous crops and producing the highly regulated mycotoxin aflatoxin B1 (AFB1). A. flavus can contaminate a wide range of important agricultural crops, such as maize, peanuts, cotton, and wheat, either in the field or in storage, creating major human health concerns and substantial economic losses (4). A. flavus lacks host specificity, and few plant-pathogenic fungi have a similarly broad host range (5). During seed colonization, A. flavus forms biofilm-like structures caused by hypha differentiation at the endosperm-germ interface (6). Contamination with some A. flavus strains may lead to the production of dangerous mycotoxins such as AFB1, one of the most carcinogenic natural compounds, posing tremendous health impacts worldwide (7). As these mycotoxins are both toxigenic and carcinogenic, controlling aflatoxin contamination is an important food safety issue. Several field control techniques that block aflatoxin production are being utilized or explored, such as modification of cultural practices, development of resistant crops through molecular and proteomic techniques, competitive exclusion using nonaflatoxigenic strains to prevent aflatoxin contamination, and development of field treatments. During storage, aflatoxin production is controllable through maintaining available moisture below the levels that will support the growth of A. flavus (3). However, the production of mycotoxin-free food remains a challenge in the food industry due to the low toxic dose, difficulties in contamination prevention, and the inefficiencies and inconsistency of traditional mycotoxin removal strategies such as sorting, washing, dehulling, density segregation, grain milling, and thermal treatment. The limitations of conventional methods used for mycotoxin removal from food commodities drive ongoing research on alternative mycotoxin decontamination approaches, including irradiation (8), ozone treatment (9), and essential oils (10). The focus of this study is atmospheric cold plasma (ACP) for A. flavus inactivation to mitigate the production of aflatoxin. Recent studies have shown that the use of cold plasma can be an effective method for A. flavus inactivation. Simoncicova et al. (11) investigated the effect of diffuse coplanar surface barrier discharge (DCSBD) plasma treatment of A. flavus mycelia on a laboratory medium, malt extract agar, where profound effects of plasma on hyphal cell structure, subcellular organelles, and viability were observed. Only 5 s of plasma exposure led to an approximately 55% reduction in biomass production compared to untreated samples, whereas treatment for 30 s caused a complete or an almost complete loss of cell viability. The inactivation efficacy of plasma against A. flavus has also been successfully tested on various food matrices, including grains and legumes (12, 13), hazelnuts (14, 15), brown-rice cereal bars (16), and groundnuts (17). In addition to A. flavus inactivation, ACP treatment may also be used for the degradation or removal of mycotoxins produced by this fungus, which has been shown in several studies (comprehensively reviewed in reference 18). Limitations across the available studies include the limited consideration of target matrices, the fungal phenotype, the impact of cold plasma as an intervention technology on grain quality factor retention, the treatment mode, and the unsustainability of the application of nonrenewable gases for agricultural or food commodity processing.

Most of the studies on the antimicrobial activity of ACP have been conducted by exposing test microorganisms or artificially inoculated food matrices/models directly to plasma to obtain the maximum microbial inactivation efficiency. Some negative effects, like changes in food color (19), reductions in sensory quality parameters (20), and reductions in seed germination (21, 22), were reported. To overcome these problems, plasma-activated water (PAW), containing long-lived reactive chemical species, could be an alternative approach to address microbiological safety while retaining or promoting other functional properties. Some recent studies have investigated PAW for bacterial inactivation and the control of bacterial growth (23).

The main objective of this study was to compare the fungicidal activities and inactivation mechanisms of two modes of ACP treatment, i.e., gas plasma (GP) and PAW, by investigating spore survival rates, viability, and cell wall integrity. The chemical characterization of PAW and aqueous spore suspensions posttreatment was carried out to investigate how changes in chemistry, namely, changes in pH levels and concentrations of secondary reactive species, including hydrogen peroxide and nitrates, contribute to the overall antifungal effects. The effect of gas plasma on the A. flavus biofilm phenotype was also investigated.

RESULTS

Effects of gas plasma and PAW treatment on aqueous spore suspensions of A. flavus.

The fungicidal properties of two different types of ACP treatment, i.e., GP and PAW, against spore suspensions of A. flavus were compared in terms of spore culturability and metabolic activity (Fig. 1). For gas plasma-treated samples, apart from the 5 min of indirect treatment, which insignificantly reduced spore counts, all gas plasma treatments tested (direct [5 and 20 min] and indirect [20 min]) resulted in significant decreases (P < 0.05) of both spore counts and metabolic activity. A maximal reduction of 2.2 log₁₀ units and 96.8% viability reduction were achieved after 20 min of direct plasma exposure (Fig. 1a). Contact with PAW resulted in significant decreases (P < 0.05) of spore counts for all treatment parameters, i.e., using water prepared by treatment directly for 5 min (PAW-5) or 20 min (PAW-20), with a subsequent time of contact with fungal spores of 2 h or 24 h. The reductions achieved using PAW contact were lower than with gas plasma treatment: 0.2- and 0.6-log₁₀ reductions after 2 and 24 h of contact time with PAW-20, respectively. However, metabolic activity was decreased by 42.2% and 55.2% for PAW-20 with contact times of 2 and 24 h, respectively. PAW-5 had no significant effect on metabolic activity (Fig. 1b).

FIG 1.

ACP treatment efficacy against A. flavus spores. The effects of GP (a) and PAW (b) on spore culturability (log₁₀ CFU per milliliter) and metabolic activity (percent) are shown. Error bars show the standard deviations. Different letters indicate a significant difference at the 0.05 level between various treatment parameters.

Both types of ACP treatment, GP and PAW, resulted in the formation of malondialdehyde (MDA) (Fig. 2). Gas plasma treatment for 20 min and treatment with PAW-20 for both contact times, i.e., 2 and 24 h, significantly increased the concentration of MDA generated (P < 0.05). Recorded MDA concentrations were maximally increased by 235.7 and 228.2 nM after 20 min of direct gas plasma treatment and PAW-20 with a 24-h contact time, respectively (Fig. 2a and b).

FIG 2.

Effect of ACP treatment of A. flavus spores on MDA concentrations (nanomolar) using GP (a) and PAW (b). Different letters indicate a significant difference at the 0.05 level between various treatment parameters.

Concentrations of extracellular DNA and protein of A. flavus spores depended on the dose and the duration of GP and PAW exposure (Fig. 3). The application of either GP directly or PAW-20 for 2 or 24 h increased the detected extracellular DNA and protein concentrations significantly. Direct gas plasma treatment for 20 min caused maximal increases in DNA and protein concentrations, by 2.9 μg/ml and 34.2 mg/ml, respectively, whereas treatment of samples with PAW-20 with a 2-h contact time increased the maximal concentrations of DNA and protein by 1.4 μg/ml and 22.9 mg/ml, respectively (Fig. 3a and b).

FIG 3.

Effect of ACP treatment of A. flavus spores on concentrations of extracellular DNA (micrograms per milliliter) and protein (milligrams per milliliter) using GP (a) and PAW (b). Different letters indicate a significant difference at the 0.05 level between various treatment parameters.

Changes in the chemistry of water and fungal suspensions subjected to plasma.

Cold plasma treatment of liquids generates longer-lived reactive oxygen and nitrogen metastable species in liquids, which is accompanied by changes in the pH values. Gas plasma treatment resulted in dramatic pH decreases (Table 1) in both water and fungal suspensions, with the lowest value of pH ∼2.4 being recorded after 20 min of direct exposure. There were no significant differences between water and fungal suspensions subjected to the same ACP treatments.

TABLE 1.

Effect of ACP treatment on pH values of water and fungal suspensions^a

Treatment parameter	Mean pH ± SD
	Deionized water + gas plasma	Fungal suspension + gas plasma	Fungal suspension + PAW
			0 h	2 h	24 h
Control	5.42 (A; 1) ± 0.06	5.41 (A; 1) ± 0.10	5.42 (A; 1) ± 0.02	5.43 (A; 1) ± 0.02	5.43 (A; 1) ± 0.02
Indirect
5 min	2.87 (B; 1) ± 0.02	2.84 (B; 1) ± 0.06	2.83 (B; 1) ± 0.02	2.86 (B; 1) ± 0.05	2.86 (B; 1) ± 0.03
20 min	2.80 (C; 1) ± 0.01	2.79 (B; 1) ± 0.03	2.81 (B; 1) ± 0.03	2.79 (C; 1) ± 0.02	2.80 (C; 1) ± 0.02
Direct
5 min	2.62 (D; 1) ± 0.08	2.64 (C; 1) ± 0.05	2.66 (C; 1) ± 0.02	2.62 (D; 1) ± 0.02	2.65 (D; 1) ± 0.02
20 min	2.37 (E; 1) ± 0.05	2.38 (D; 1) ± 0.06	2.42 (D; 1) ± 0.02	2.43 (E; 1) ± 0.02	2.42 (E; 1) ± 0.02

^aDifferent parenthetical letters within the column indicate a significant difference between the untreated control and ACP-treated samples (P < 0.05); different parenthetical numbers within the row indicate a significant difference between the corresponding samples subjected to the same ACP treatment parameters (P < 0.05).

Gas plasma treatment of both water and fungal suspensions resulted in significant increases in hydrogen peroxide (H₂O₂) concentrations in all tested samples (Table 2). The exposure of fungal spores to either water or PAW caused a significant increase in H₂O₂. The lowest concentrations of H₂O₂ were noted for fungal suspensions treated with gas plasma and fungal suspensions after a 24-h contact time with PAW. For samples treated with PAW, concentrations of H₂O₂ decreased with prolonged contact times (P < 0.05), reaching the lowest recorded values after 24 h of contact with PAW (values for untreated controls remained at the same levels).

TABLE 2.

Effect of ACP treatment on hydrogen peroxide concentrations in water and fungal suspensions^a

Treatment parameter	Mean hydrogen peroxide concn (μM) ± SD
	Deionized water + gas plasma	Fungal suspension + gas plasma	Fungal suspension + PAW
			0 h	2 h	24 h
Control	0.59 (A; 1) ± 0.69	10.97 (A; 2) ± 1.69	12.07 (A; 2) ± 0.52	11.98 (A; 2) ± 1.68	12.17 (A; 2) ± 0.95
Indirect
5 min	114.04 (B; 1) ± 2.51	49.53 (A, B; 2) ± 6.47	99.76 (B; 3) ± 2.75	69.39 (B; 4) ± 1.46	33.37 (B; 5) ± 0.89
20 min	104.56 (B; 1) ± 7.30	90.25 (B; 2) ± 13.18	167.81 (C; 3) ± 3.62	136.43 (C; 4) ± 1.08	94.48 (C; 2) ± 3.43
Direct
5 min	403.84 (C; 1) ± 47.50	280.18 (C; 2) ± 29.51	418.74 (D; 1) ± 9.30	384.76 (D; 1) ± 2.39	249.11 (D; 2) ± 15.36
20 min	2,014.04 (D; 1) ± 85.93	1,774.3 (D; 2) ± 106.07	2,110.69 (E; 3) ± 12.24	2,079.67 (E; 1, 3) ± 2.20	2,005.13 (E; 3) ± 4.02

^aDifferent parenthetical letters within the column indicate a significant difference between the untreated control and ACP-treated samples (P < 0.05); different parenthetical numbers within the row indicate a significant difference between the corresponding samples subjected to the same ACP treatment parameters (P < 0.05).

Similarly, ACP treatment increased nitrate (NO₃⁻) concentrations significantly in the tested samples (Table 3). The levels of NO₃⁻ decreased during contact time for samples treated with PAW, as observed for H₂O₂.

TABLE 3.

Effect of ACP treatment on nitrate concentrations in water and fungal suspensions^a

Treatment parameter	Mean nitrate concn (mM) ± SD
	Deionized water + gas plasma	Fungal suspension + gas plasma	Fungal suspension + PAW
			0 h	2 h	24 h
Control	0.00 (A; 1) ± 0.00	0.00 (A; 1) ± 0.00	0.00 (A; 1) ± 0.00	0.00 (A; 1) ± 0.00	0.00 (A; 1) ± 0.00
Indirect
5 min	0.15 (A, B; 1) ± 0.01	0.04 (A; 2) ± 0.01	0.13 (A; 1, 3) ± 0.03	0.09 (A; 3, 4) ± 0.05	0.07 (A, B; 2, 4) ± 0.05
20 min	0.25 (B; 1) ± 0.02	0.15 (B; 2) ± 0.05	0.27 (A; 1) ± 0.04	0.25 (A; 1) ± 0.04	0.15 (B; 2) ± 0.06
Direct
5 min	1.14 (C; 1, 3) ± 0.09	1.10 (C; 1) ± 0.09	1.38 (B; 2) ± 0.09	1.22 (B; 1) ± 0.07	1.15 (C; 1, 3) ± 0.05
20 min	4.46 (D; 1) ± 0.29	4.06 (D; 2) ± 0.05	5.23 (C; 1) ± 0.66	5.05 (C; 1) ± 0.50	4.54 (D; 1) ± 0.16

^aDifferent parenthetical letters within the column indicate a significant difference between the untreated control and ACP-treated samples (P < 0.05); different parenthetical numbers within the row indicate a significant difference between the corresponding samples subjected to the same ACP treatment parameters (P < 0.05).

Effect of low pH on A. flavus spores.

The effect of low pH on A. flavus spores was evaluated using acidified water (AW) of pH ∼2.4 and compared to that using PAW-20 (Fig. 4). Treatment with AW decreased cell viability by 21.4% and 26.2% after contact times of 2 and 24 h, respectively, whereas reductions of spore counts for both contact times were of 0.1 log₁₀ units. Overall, PAW-20 was more effective against A. flavus spores than AW at the same pH.

FIG 4.

Effects of acidified water and PAW-20 on A. flavus spores. Effects on spore culturability (log₁₀ CFU per milliliter) and metabolic activity (percent) are shown. Error bars show the standard deviations. Different letters indicate a significant difference at the 0.05 level between various treatment parameters.

Gas plasma treatment of A. flavus biofilms.

Biomass production of A. flavus biofilms was estimated using the crystal violet (CV) assay. The initial biofilm biomass, i.e., biomass after 24 h of incubation, was compared to those of untreated and ACP-treated biofilms after an additional 24 h of posttreatment storage at 15°C. Similar trends were observed for both inoculum concentrations used to form the fungal biofilms (6 and 7 log₁₀ CFU/ml); it was noted that compared to the initial biofilm biomass, the untreated controls significantly increased their biomass during 24 h of posttreatment storage, by 45.8% and 79.8% for the 6- and 7-log₁₀ CFU/ml inocula, respectively. In contrast, the biomasses of ACP-treated biofilms decreased by 20.3% and 13.9% for 6- and 7-log₁₀ CFU/ml initial inocula, respectively. Compared to their respective untreated controls, the biomasses of ACP-treated biofilms decreased significantly, by 45.4% and 52.1%, using 6- and 7-log₁₀ CFU/ml inoculum concentrations, respectively (Fig. 5).

FIG 5.

Effect of gas plasma treatment on biomass production (optical density at 590 nm [OD₅₉₀]) of A. flavus biofilms. Different letters indicate a significant difference at the 0.05 level between the initial biofilm biomass (after 24 h of incubation) and those of untreated control and ACP-treated samples.

To investigate the effect of gas plasma treatment on A. flavus biofilms, two initial inoculum concentrations were used, namely, 6 and 7 log₁₀ CFU/ml. This resulted in concentrations of 6.2 and 9.6 log₁₀ CFU/ml for cells within the biofilms after 24 h of incubation plus a subsequent 24 h of storage at 15°C. Regardless of the initial inoculum concentration, a 20-min direct gas plasma treatment significantly decreased both cell counts and viability, with a 2.3-log₁₀ reduction of cell counts and an 88.3% reduction of cell viability for the 6-log₁₀ CFU/ml inoculum and a 1.5-log₁₀ reduction of cell counts and a 63.2% reduction of cell viability for the 7-log₁₀ CFU/ml inoculum (Fig. 6).

FIG 6.

Effect of gas plasma treatment on A. flavus biofilms. The effects on spore culturability (log₁₀ CFU per milliliter) and metabolic activity (percent) are shown. Error bars show the standard deviations. Different letters indicate a significant difference at the 0.05 level between untreated control and ACP-treated samples.

Although ACP treatment resulted in significant increases of MDA concentrations for both A. flavus biofilms used in the study, the changes were more pronounced for the lower inoculum concentration: an increase of 1,436.4 nM for the 6-log₁₀ CFU/ml inoculum compared to an increase of 534.6 nM for the 7-log₁₀ CFU/ml inoculum (Fig. 7).

FIG 7.

Effect of gas plasma treatment of A. flavus biofilms on MDA concentrations (nanomolar). Different letters indicate a significant difference at the 0.05 level between untreated control and ACP-treated samples.

A similar phenomenon was observed for extracellular DNA and protein measurements: ACP treatment significantly increased the concentrations of both DNA, by 11.6 and 3.6 μg/ml for the 6- and 7-log₁₀ CFU/ml inocula, respectively, and protein, by 146.6 and 44.6 mg/ml for the 6- and 7-log₁₀ CFU/ml inocula, respectively (Fig. 8). The ACP treatment applied was the same for the two biofilm densities investigated.

FIG 8.

Effect of gas plasma treatment of A. flavus biofilms on concentrations of extracellular DNA (micrograms per milliliter) and protein (milligrams per milliliter). Different letters indicate a significant difference at the 0.05 level between untreated control and ACP-treated samples.

DISCUSSION

Atmospheric cold plasma technology has received considerable attention in recent years for potential food decontamination applications, primarily using gas plasma, whether direct, indirect, or in package. However, poor process product compatibility may negatively affect food quality or functional properties. Therefore, plasma-activated water has been considered an alternative method for food disinfection (24, 25). Recently, numerous studies reported the useful antimicrobial efficacy of PAW (26–28); however, to date, reports have focused mainly on PAW efficacy or interactions with bacteria, and the reports focusing on the utilization of PAW against fungi are limited. However, spore-forming fungi are important in food spoilage and may also produce toxic metabolites that impact the safety and sustainability of the cereals sector and associated food supply chains globally. To our knowledge, there have been no previous reports directly comparing gas plasma and PAW treatment efficacies against fungi; the main objective of this study was to evaluate the mechanisms of inactivation of fungal spores of A. flavus by investigating effects on spore survival rates, viability, and cell wall integrity. Our study also reports the effects of ACP treatment on biofilms produced by A. flavus.

Aspergillus spp. are typically classified as “storage” contaminants of cereal grains, and their growth is associated with inefficient drying. ACP technology is flexible in application, and treatments could be applied using different modes of delivery at seed, in the field, at harvest, in transport, prior to storage (to decrease the number of spores present on grains), and during storage. With respect to storage, if the conditions were not sufficient to inhibit the germination of the remaining spores, this could result in the formation of biofilm-like structures; at this stage, ACP could be used again as a reprocess or sparge mode to inhibit biofilm growth to avoid the potential for mycotoxin production. In this study, the posttreatment storage time in a contained environment can be analogous to grain storage in silos or large grain tanks. When combined with an extended posttreatment storage time, the effects of plasma treatment would persist longer than during the exposure time only, due to a longer time of contact with longer-lived reactive species generated during ACP treatment and present in a contained environment. The choice of storage conditions was motivated by the results obtained in our previous study (21), in which extending the posttreatment storage time up to 24 h resulted in higher microbial inactivation levels than with 0 or 2 h of posttreatment storage at 15°C.

Gas plasma was generally more efficient against aqueous suspensions of A. flavus than PAW, resulting in a maximal reduction of 2.2 log counts and a 96.8% reduction in metabolic activity after 20 min of direct treatment, compared to a 0.6-log reduction of counts and a 55.2% metabolic activity reduction after 24 h of contact with PAW. Increases in MDA, DNA, and protein concentrations were more pronounced when GP treatment was applied. As shown in Tables 1 to 3, there were significant changes in the chemistry of both water and fungal suspensions subjected to plasma. The antimicrobial properties of noncomplex plasma-treated liquids such as water are reported to be caused mainly by a pH reduction (increased acidity) and the generation of long-lived secondary chemical species such as hydrogen peroxide, nitrite, nitrate, and other short-lived species (28), whereas the higher inactivation efficiency of GP in addition to the liquid-mediated changes in chemical composition could be explained by additional physical processes generated during plasma discharges, such as a high electric field, overpressure shock waves, and intense UV radiation, which are known to inactivate microorganisms by DNA or RNA damage (29, 30). Although many studies reported high efficacies of PAW treatment against bacterial vegetative cells, we report that the inactivation levels obtained for fungal cells were lower than those for the treatment of bacteria. The fungicidal effect of cold plasma mechanisms is weaker than the bactericidal effect under comparable conditions, which is caused by differences in the structures and compositions of prokaryotic and eukaryotic microbial cells. The cell structure of fungi is more sophisticated than the bacterial one; therefore, different plasma conditions must be adopted to achieve a comparable inactivation effect (31). Kamgang-Youbi et al. (26) evaluated the microbial disinfection efficacy of PAW against a range of bacteria (Staphylococcus epidermidis, Leuconostoc mesenteroides, and Hafnia alvei) and a yeast (Saccharomyces cerevisiae) model and found that inactivation was more effective for bacteria than for the yeast. Xu et al. (32) investigated the effects of soaking on the postharvest preservation of button mushrooms (Agaricus bisporus) in PAW over 7 days of storage at 20°C. It was reported that PAW reduced the microbial counts by 1.5 log₁₀ and 0.5 log₁₀ units for bacteria and fungi during storage, respectively. Although GP is reported to be a more efficient microbial decontamination tool against fungi, it should be considered that the application of PAW might be more suitable, alone or in combination with other intervention technologies, for the disinfection of foods due to minimized negative effects on food quality parameters compared to GP treatment. The effect of food surface topography might also contribute to the potentially higher antimicrobial efficacy of PAW: as various microorganisms can be present within various food irregularities, such as cracks, grooves, or gaps, transient exposure to GP might be insufficient, and PAW could penetrate these areas easier, as liquids are amenable to different application modalities, including misting, coating, and spraying.

Gas plasma was also evaluated against A. flavus biofilms. Direct treatment with GP for 20 min was selected for biofilm treatment due to the highest levels of inactivation and decreases of cell viability among all the parameters tested against A. flavus spores in aqueous suspensions (Fig. 1). Although filamentous fungi possess a natural tendency to grow adhered to surfaces, the influence of this type of growth on fungal physiology has not yet been thoroughly studied (33). Biofilms produced by Aspergillus spp., similarly to bacterial biofilms, show defined developmental phases: following initial spore seeding, there are a lag phase (spore adhesion), spore germination, filamentation, and the formation of a monolayer, followed by increased structural complexity, exopolysaccharide (EPS) production, and maturation. Spore contact and attachment to the surface are required to trigger germination and hypha formation, leading to biofilm development (34). From an agriculture and food (agri-food) sector perspective, the interest in fungal biofilms is due mainly to the fact that some of the most devastating and universal crop diseases are caused by plant-pathogenic fungi. Furthermore, resistance to conventionally used biocides is a defining characteristic of fungal biofilms (35). The target microorganism used in this study, A. flavus, was shown to form biofilm-like structures during seed colonization (6); however, the morphological and molecular changes that occur in the fungus during seed colonization are not yet well understood. Dolezal et al. (36) demonstrated that A. flavus could infect all tissues of the immature maize kernel by 96 h after infection. It was found that at the endosperm-germ interface, hyphae appeared to differentiate and form a biofilm-like structure that surrounded the germ, whereas mycelia were observed in and around the point of inoculation in the endosperm and were found growing down to the germ. It is probable that similar structures are formed by A. flavus during seed colonization of other economically important crops, such as wheat and barley. In this study, we compare the efficacies of ACP treatment against in vitro A. flavus biofilm models formed using two different initial inoculum concentrations, namely, 6 and 7 log₁₀ spores/ml. In general, the initial inoculum concentration used for biofilm formation was a crucial factor in plasma treatment efficacy: the antimicrobial effects of ACP were more pronounced when a lower concentration (6 log₁₀ spores/ml) was used. Via both optical microscopic (Fig. 9) and scanning electron microscopy (SEM) (Fig. 10) observations, it was demonstrated that increasing the inoculum concentration from 6 to 7 log₁₀ CFU/ml resulted in high spore numbers remaining within the structure of biofilms; i.e., not all spores from the inoculum underwent germination, which likely reduced their sensitivity to plasma treatment. Similarly, Mowat et al. (37) observed that the structural morphology and integrity of biofilms formed by Aspergillus fumigatus were dependent on the concentration of the inoculum (conidial seeding density). It should also be noted that the inoculum concentrations used in this experiment were much higher than the A. flavus counts present on naturally contaminated foods; contamination levels of A. flavus reported for corn (38), soybean (39), and various tree nut samples (40) did not exceed 4.0 log₁₀ CFU/g. However, in our preliminary work, it was observed that biofilms initiated using inoculum concentrations of 5 log₁₀ spores/ml and lower did not form stable fungal biofilms for the in vitro settings used here; they were easily disrupted by simple handling and were not reproducible.

FIG 9.

Structure of A. flavus biofilms before and after ACP treatment. Optical microscope images (magnification, ×400) at initial inoculum concentrations of 6 log₁₀ CFU/ml for control (a) and ACP-treated (b) samples and 7 log₁₀ CFU/ml for control (c) and ACP-treated samples (d) are shown.

FIG 10.

Structure of A. flavus biofilms before and after ACP treatment. Images from SEM analysis (magnification, ×2,000) at initial inoculum concentrations of 6 log₁₀ CFU/ml for control (a) and ACP-treated (b) samples and of 7 log₁₀ CFU/ml for control (c) and ACP-treated (d) samples are shown. White arrows indicate spores with visibly deformed structures.

The overall effects of plasma treatments on fungal biofilms were similar to those on aqueous spore suspensions (obtained by both GP and PAW) and included significant decreases in the remaining spore counts and viability as well as the release of intracellular material such as MDA (a product of membrane lipid peroxidation), DNA, and proteins, suggesting a partial or total disruption of cell walls leading to compromised cell wall integrity and, subsequently, cell leakage; a similar phenomenon after ACP treatment was previously observed in bacteria (41). These results are in line with the SEM observations of untreated and ACP-treated A. flavus biofilms: considerable morphological alterations and structural damage to fungal structures, particularly spores, were observed within plasma-treated biofilms (Fig. 10). Similarly, Lee et al. (42) reported via SEM observations that exposure of spores of the insect-pathogenic fungus Cordyceps bassiana to ACP resulted in a dramatic change in cell morphology. After plasma treatment, spores were ruptured, shrunken, and flattened, indicating that the intracellular space had been at least partially emptied of its contents. Additionally, the plasma-treated spores showed a rough, wrinkled surface. Avramidis et al. (43) evaluated the effect of plasma treatment on mycelia of other agriculturally relevant fungal species, Ascochyta pinodella and Fusarium culmorum. Using light microscopy, they observed that the complete mycelium was affected by plasma exposure: the cell wall and cell membrane structures were damaged, resulting in leakage of the cytoplasm. After 180 s of ACP treatment, several cracks occurred on the cell wall of the hypha, and after 360 s of treatment, deformation progressed: cracks were broadened, and the whole hypha was flattened. Suhem et al. (16) reported damaged structures of A. flavus treated with a cold atmospheric plasma jet: microscopic observations showed broken conidiophores and vesicles, which resulted in cell leakage and a loss of viability. SEM observations by Dasan et al. (44) demonstrated that spores of Aspergillus parasiticus lost their integrity after plasma treatment and that the cell contents dispersed into clusters. Based on these observations, it may be hypothesized that plasma treatment results in fungal cell wall disruption, making it permeable and thus allowing the leakage of intracellular components, such as nucleic acids and proteins, consequently leading to metabolic activity decreases resulting in cell death (18, 42). The proposed mechanisms of fungal inactivation by cold plasma are summarized in Fig. 11.

FIG 11.

Mechanisms of fungal inactivation by cold plasma.

Other considerable factors that contributed to the fungicidal effects of both GP treatment of aqueous fungal suspensions as well as treatment with PAW were changes in chemistry caused by exposure to plasma, namely, pH reduction and the generation of H₂O₂ and NO₃⁻, which had a combined effect on microbial inactivation. Changes in the chemistry of water and fungal suspensions subjected to plasma were evaluated to better understand the mechanisms of inactivation of the gas-mediated versus liquid-mediated mode of application of ACP. This is also an important aspect of the use of ACP technology successfully in food processing. Naïtali et al. (45) investigated the contributions of nitrites, nitrates, and H₂O₂ to the lethal effect of PAW on a Gram-negative bacterium, Hafnia alvei, by evaluating the disinfection potential of acidified (by HCl) solutions prepared using these compounds alone or in a mixture at the concentrations found in PAW. It was reported that after 30 min of contact with PAW, a log₁₀ reduction of >5.9 was achieved, whereas after the same time of contact with acidified water, only a 0.4-log₁₀ reduction was found. An acidified mixture of nitrites, nitrates, and H₂O₂ at the same concentrations as the ones in PAW resulted in a 5.7-log₁₀ reduction. This study also underlined the need for an acidic pH control to confirm the effect of PAW: buffered PAW reduced bacterial counts by only 0.2 log₁₀ units. Similarly, Schnabel et al. (46) compared the inactivation efficacies of various plasma-activated liquids to those of solutions acidified with HNO₃: the results showed that only a very strongly acidified HNO₃ solution led to inactivation rates comparable to those of PAW. Also, in our study, PAW treatment was more effective against A. flavus spores than acidified water at the same pH (Fig. 4). While the lower pH strongly supports the inactivation process, an antimicrobial effect of PAW based exclusively on increased acidification can be excluded. It can therefore be concluded that microbial inactivation using PAW is based on synergistic effects between acidification and the generation of long-lived secondary chemical species.

Conclusion.

Both GP and PAW treatments can be effective tools against A. flavus spores: both types of ACP treatment resulted in significant decreases of spore counts and viability, with GP being generally more effective. It was concluded that fungal inactivation using PAW is based on synergistic effects between acidification and the generation of long-lived secondary chemical species, whereas changes in the chemical composition of treated samples combined with additional physical processes generated during plasma discharge are factors for the higher inactivation efficiency of GP. Of relevance to agri-food sectors, our study presents the effects of ACP treatment on spores and biofilms produced by A. flavus, showing that the initial spore concentration leading to biofilm formation was a crucial factor in plasma treatment efficacy. Overall, pronounced effects of plasma on both external surface elements and internal structures of fungal cells were observed, providing new insights into the mechanisms resulting in the inactivation of fungi by cold plasma.

MATERIALS AND METHODS

Fungal strain and culture conditions.

The fungal strain used was an A. flavus strain isolated from commercially available wheat grains (origin, United Kingdom) and identified by matrix-assisted laser desorption ionization–time of flight mass spectrometry (MALDI-TOF MS). Spore suspensions of A. flavus were prepared as described previously (21). Briefly, potato dextrose agar (PDA; Biokar Diagnostics, France) was inoculated with A. flavus and incubated at 30°C until visible sporulation was obtained (5 to 7 days). The spores were harvested by flooding the agar surface with 10 ml of sterile phosphate-buffered saline (PBS) containing Tween 20 (1%) and scraping the spores from mycelia with a sterile spreader. The suspension was washed twice in sterile PBS, adjusted to 8 to 9 log₁₀ CFU/ml, and stored at 4°C for up to 3 weeks, with germination ability assessed prior to each experiment.

Biofilm formation.

A. flavus biofilms were formed on 6-well flat-bottom polystyrene plates (Sarstedt, Ireland) according to the protocol described previously by Ramirez Granillo et al. (47), with modifications. Working spore suspensions of A. flavus were prepared in RPMI medium (Merck, Ireland) supplemented with glucose to 2% at two initial concentrations, ∼6.0 and 7.0 log₁₀ spores/ml. Biofilms were formed by adding 2 ml of the working spore suspension to each well. The plates were incubated at 37°C for 24 h. After incubation, the medium containing unattached mycelium was discarded, and the wells were washed with 2 ml PBS to remove the remaining unattached cells.

ACP treatment.

(i) ACP system setup. The ACP system used in this study was a high-voltage (HV) dielectric barrier discharge (DBD) system with a maximum voltage output in the range of 0 to 120 root mean square kV (kV_rms) at 50 Hz, as described in detail previously by Pankaj et al. (48) and as fully characterized by Moiseev et al. (49) and Milosavljević and Cullen (50). The experimental approach and schematic diagram of the system are shown in Fig. 12a and b, respectively. All samples were subjected to ACP treatment at 80 kV at atmospheric pressure using atmospheric air as the working gas. The distance between the two 15-cm-diameter aluminum disk electrodes was 40 mm, which was equal to the height of the polypropylene container (310 by 230 by 20 mm), which served as both a sample holder and a dielectric barrier. Samples were placed inside the container and subjected to either direct or indirect plasma treatment as described previously (51). For direct exposure, samples were placed between the electrodes, i.e., within the plasma discharge, with a 10-mm distance between the sample and the top electrode. For indirect plasma treatment, samples were placed in the corner of the container, the distance between the samples and the center of the electrodes ranged from 120 to 160 mm owing to the sample distribution on the plate. Direct and indirect plasma treatments were conducted simultaneously. Before treatment, each container was sealed with a high-barrier polypropylene bag (catalog number B2630; Cryovac, Ireland).

FIG 12.

Experimental setup (a) and schematic diagram (b) of the DBD ACP reactor. Either petri dishes containing sterile deionized water or A. flavus spore suspensions or a 6-well plate containing A. flavus biofilm was treated inside a polypropylene container. For direct exposure, samples were placed within the plasma discharge, and for indirect exposure, they were placed outside the discharge.

(ii) Gas plasma treatment of A. flavus spore suspensions and biofilms. An A. flavus spore suspension (10 ml) with an average concentration of 6.0 log₁₀ spores/ml was transferred to a petri dish (diameter, 90 mm), placed inside the container without the petri dish lid, and subjected to both direct and indirect treatments for either 5 or 20 min (two separate petri dishes were used to achieve two different modes of treatment). Plates containing A. flavus biofilms were placed inside the container and subjected to direct ACP treatment for 20 min. There was a slight increase in temperature observed as a result of direct treatment. On average, the increases in temperatures measured immediately after treatment for up to 5 and 20 min were not higher than 5°C and 10°C, respectively. No change in temperature was noted for indirect GP exposure.

(iii) Generation of PAW. Sterile deionized water (10 ml) within a petri dish (diameter, 90 mm) was placed without the lid inside the container and subjected to both direct and indirect ACP treatments using the same contained plasma reactor system, for either 5 or 20 min, to generate PAW-5 and PAW-20, respectively.

(iv) Posttreatment storage. After the treatment, all samples were stored unopened for 24 h at 15°C. Control samples were stored under identical conditions. Unless otherwise stated, all experiments were performed three times.

(v) PAW treatment of A. flavus spore suspensions. For the determination of the antifungal effect of PAW, 200 μl of an A. flavus spore suspension (7 log₁₀ spores/ml) was added to 2-ml microtubes containing 1,800 μl of PAW, which resulted in a final concentration of 6 log₁₀ spores/ml. The samples were vortexed and incubated at 15°C for either 2 or 24 h of PAW contact. Due to the low antimicrobial effects of PAW generated by the indirect mode of plasma exposure (data not shown), only PAW prepared via direct treatment (5 and 20 min) was used for the treatment of A. flavus spores (water subjected to both direct and indirect ACP treatment was used for chemical analyses for comparative purposes). PAW treatment was performed at room temperature. After each contact time, samples were immediately subjected to further analysis.

Effect of low pH on spore inactivation levels and viability.

An additional experiment evaluated the effect of low pH on A. flavus spores. The purpose of this part of the study was to compare the antifungal efficacies of PAW and acidified water (AW) at the same pH value. As PAW treatment was most effective using PAW-20 at a pH value of 2.37, AW with approximately the same pH value (∼2.4) was selected to perform a comparative pH control experiment. The AW solution at pH ∼2.4 was adjusted with 1 M HCl. Treatment of A. flavus spore suspensions with acidified water was conducted as described above [see “ACP treatment. (v) PAW treatment of A. flavus spore suspensions”], using AW instead of PAW.

Microbiological analysis.

A. flavus spore suspensions treated with GP were transferred from petri dishes to 25-ml tubes and vortexed. Spore suspensions treated with PAW or AW were vortexed. For analysis of A. flavus biofilms, 2 ml of PBS was added to the wells containing biofilms. Biofilms were scraped thoroughly with a pipette tip, and the PBS-cell suspension was transferred into 2-ml tubes. All samples were analyzed for viability using a standard plate count method and a 2,3-bis-(2-methoxy-4-nitro-5-sulfophenyl)-2H-tetrazolium-5-carboxanilide salt (XTT) assay. For GP- and PAW-treated samples, levels of lipid peroxidation and extracellular DNA and protein contents were also recorded. Optical microscopy and scanning electron microscopy (SEM) were used to study A. flavus biofilms.

(i) Plate count assay. ACP-treated samples and the corresponding controls were serially diluted in maximum-recovery diluent (MRD; Scharlau, France). Aliquots (0.1 ml) of appropriate dilutions as well as undiluted samples (1 ml) were plated onto PDA plates and incubated at 30°C for 24 to 48 h. The limit of detection was 1 CFU/ml.

(ii) Crystal violet assay. A crystal violet (CV) assay was used for the quantification of fungal biofilm biomass. The assay was performed according to the procedure described previously by Peeters et al. (52), with minor modifications. Briefly, fungal biofilms were fixed with 99% methanol (1 ml) for 15 min, after which supernatants were removed and the plates were air dried. Subsequently, 2 ml of a 0.2% CV solution (Merck, Portugal) was added to the wells, and the plates were incubated for 20 min. After this time, the plates were rinsed with tap water and air dried. The cell-bound crystal violet was dissolved by adding 2 ml of 33% acetic acid (Sigma-Aldrich, Ireland), and the absorbance was measured at 590 nm using a microplate reader (Synergy HT; BioTek Instruments Inc.). All steps were carried out at room temperature. Each absorbance value was corrected by subtracting the mean absorbance of a blank (uninoculated medium). Experiments were performed in triplicate and repeated twice.

(iii) XTT assay. An XTT assay (VWR, Ireland) was used to estimate the viability of aqueous suspensions of fungal spores and biofilms based on their metabolic activity and was carried out according to the procedure described previously by Peeters et al. (52), with minor modifications. Before each assay, fresh solutions of XTT were prepared by dissolving 4 mg in 10 ml of prewarmed PBS. The solution was supplemented with 5.5 mg menadione (Sigma-Aldrich Co., Ireland) in 10 ml acetone (Sigma-Aldrich Co., Ireland). To measure the viability of aqueous suspensions of fungal spores (treated with either gas plasma or PAW), 100 μl of each sample was transferred into 96-well plates, and the XTT-menadione (100 μl) complex was added. The wells containing A. flavus biofilms were filled with sterile PBS (500 μl), and XTT-menadione (500 μl) was then added. Plates were incubated for 4 h at 37°C in the dark. After incubation, the supernatant (100 μl) from each well was transferred into the wells of a new 96-well plate, and the absorbance at 486 nm was measured with a microplate reader (Synergy HT; BioTek Instruments Inc.). Each absorbance value was corrected by subtracting the mean absorbance of a blank (uninoculated medium). Experiments were performed in triplicate and replicated twice. The percentage of surviving spores was calculated as [(A_ACP − A_C)/A₀] × 100%, where A_ACP, A_C, and A₀ are the absorbances of ACP-treated, negative-control, and untreated control samples, respectively.

(iv) Lipid peroxidation-MDA assay. To determine malondialdehyde (MDA) concentrations, a thiobarbituric acid (TBA) test was conducted according to the procedure described previously by Heath and Packer (53), with minor modifications. For protein precipitation, 0.5 ml of the sample was mixed with 0.5 ml of 30% trichloroacetic acid (TCA; Sigma-Aldrich) and centrifuged at 10,000 rpm for 5 min. After centrifugation, 0.5 ml of the supernatant was mixed with 0.5 ml of 30% TCA containing 0.5% (wt/vol) TBA (Sigma-Aldrich, Ireland) and 25 μl 4% (wt/vol) butylated hydroxytoluene (BHT; Sigma-Aldrich, Ireland) in ethanol. The mixture was heated at 98°C for 30 min and cooled on ice for 5 min. The samples were centrifuged at 4,000 rpm for 1 min. One hundred microliters of the supernatant was transferred into wells of a 96-well plate, and the absorbance was measured at 532 nm. The concentration of MDA was calculated using an MDA standard curve.

(v) Extracellular DNA and protein measurements. Extracellular DNA and protein levels were determined spectrophotometrically, and the values were estimated according to the protocol described previously by Wen et al. (30). The DNA concentration was calculated using the following formula (54): dsDNA (double-stranded DNA) concentration (micrograms per milliliter) = 50 μg/ml × A₂₆₀ × dilution factor.

The protein concentration was calculated using the following formula (30): protein concentration (milligrams per liter) = 1.45 × A₂₈₀ − 0.74 × A₂₆₀.

(vi) Optical microscopy. An optical microscopy technique was used to visualize structural changes in ACP-treated A. flavus biofilms compared to the untreated control. Biofilms grown on 6-well plates were observed using a reverse light microscope (Optika, Italy) under a 40× objective (total magnification, ×400). Images were acquired with a digital camera and analyzed using TCapture computer software (Tucsen Photonics Co. Ltd.).

(vii) Scanning electron microscopy. For SEM analysis, ACP-treated and untreated A. flavus biofilms were prepared according to the procedure described previously by Ziuzina et al. (55) and examined using an ultrahigh-resolution scanning electron microscope (Regulus SU-8230; Hitachi) at 1.0 kV.

Chemical analysis.

Chemical analysis was performed for PAW-5 and PAW-20 generated using both direct and indirect modes of treatment, fungal suspensions subjected to gas plasma, and fungal suspensions treated with PAW. The pH values as well as concentrations of hydrogen peroxide (H₂O₂) and nitrate (NO₃⁻) in untreated and gas plasma/PAW-treated samples were determined.

The pH was determined by recording potentiometric measurements of the suspensions at 25°C using an Orion pH meter (model 420A). H₂O₂ concentration measurement was performed according to the protocol described previously by Boehm et al. (56). The concentration of NO₃⁻ was determined photometrically with 2,6-dimethyl phenol (DMP) using the Spectroquant nitrate assay kit (Merck, Germany). Samples were incubated with reagents for 20 min. After that, 25 μl of 33% TCA was added, and samples were centrifuged at 10,000 × g for 2 min to precipitate proteins. Subsequently, 100 μl of the supernatant was transferred to wells of a 96-well plate, and the absorbance at 340 nm was measured.

A standard curve of known H₂O₂ (Perhydrol for analysis, Emsure ISO) and NO₃⁻ (sodium nitrate; Sigma-Aldrich, Ireland) concentrations was included on each plate and used to convert absorbances into concentration values.

Statistics.

Analysis of variance (ANOVA) was performed to determine significant differences among means at a 95.0% confidence level (α = 0.05). Fisher’s least significant difference (LSD) test was used. Analysis was performed using the Statgraphics Centurion XVI.I package (Statistical Graphics, WA, USA). Test statistics were regarded as significant when the P value was ≤0.05.