Impact of Erwinia gerundensis as a Biocontrol Agent on the Sanitary and Technological Quality of Barley Malt

Abstract

This study investigates the use of a barley-associated Erwinia gerundensis strain as a biocontrol agent for the protection of malt against fungal development and mycotoxin production during the malting of barley. First, the antifungal activity of E. gerundensis was studied via simultaneous co-cultures in Yeast Malt broth and on barley kernels artificially infected with Fusarium tricinctum. Then, the effect of E. gerundensis on the fungal load and mycotoxin content of malt obtained from naturally contaminated barley was studied by applying different doses of the bacterial strain at the steeping step of the malting process. Mycotoxin concentrations were determined using the technique of liquid chromatography coupled to the mass spectrometry detector in tandem (LC-MS/MS). Also, the effect of E. gerundensis on the technological quality of the final malt was evaluated using European Brewery Convention standard methods. The bacterial strain reduced the production of enniatin (ENN) by 97.7–99.8% and 38–68% in the liquid medium and on barley kernels, respectively, along with 0–82% Fusarium tricinctum growth reduction on barley kernels. Application of E. gerundensis at the beginning of the steeping step reduced the fungal load of the final malt by 67–72% and its ENN and deoxynivalenol (DON) contents by 32–50% and 15–33%, respectively, depending on the treatment dose. Barley malts treated with E. gerundensis showed no degradation in their technological quality, thus making E. gerundensis a potentially interesting biocontrol agent for the malting and brewing industries. Further investigation should be focused on improving the bioprotective capability of the proposed biocontrol agent.

Keywords:

• Pathogenic fungi
• mycotoxins
• Erwinia gerundensis
• biocontrol agent
• malting

Introduction

Beer is one of the oldest fermented beverages obtained by yeast-mediated alcoholic fermentation and the most consumed alcoholic beverage worldwide. In the EU, about 29.7 billion liters of beer were consumed in 2020/2021, out of the approximately 34.1 billion liters produced during that season.^[1] Beer is mainly made from water, hops, yeast, and malt, the latter being a germinated cereal grain (usually barley). As such, barley quality significantly affects the characteristics of the final product. Unfortunately, barley crops can be contaminated by several pathogenic fungal species, including Fusarium species, which represent a generic problem for cereals, as they cause severe damage and considerable crop losses that account for up to a loss of 10% of global food production.^[2] The main Fusarium species that contaminate barley include Fusarium avenaceum, Fusarium culmorum, Fusarium graminearum, Fusarium langsethiae, Fusarium poae, Fusarium sporotrichioides, and Fusarium tricinctum.^[3] The genus Fusarium is responsible for Fusarium Head Blight (FHB),^[4] one of the most destructive plant diseases, and it represents serious agricultural consequences in all small cereal grain crops and particularly in barley.^[5,6] Fusarium strains can also cause several carcinogenic and mutagenic diseases in humans and animals due to their natural production of harmful secondary metabolites known as mycotoxins, with the most important ones being deoxynivalenol (DON), enniatins (ENN), fumonisins, nivalenol (NIV), T2 & HT2 toxins, and zearalenone (ZEA).^[7,8] The thermostability of these toxins renders them capable of contaminating foodstuffs at different stages of the food production chain, such as in the field, during harvest, during storage, and during processing,^[9] with their global prevalence being estimated to be as high as 60–80%.^[10,11] In the field, the presence of Fusarium species and of their mycotoxins in cereal crops such as barley is mainly due to favorable climatic conditions such as high humidity (>90%) and temperature (23–29 °C) that lead to fungal development.^[12]

The malt used for beer production is obtained by a process called “malting,” during which the barley grain is encouraged to germinate under controlled conditions. Malting is a complex biological process that includes many biochemical and biological steps in which the microorganisms (bacteria, fungi, and yeasts) that are naturally present on the grains play a major role. The effects related to the grain’s microbiota can be positive or negative, depending on the nature of the microorganisms and on other factors.^[13,14] Grain moisture, nutrient abundance, and process duration are all among the factors that can induce fungal growth and mycotoxin production during malting.^[3] The excessive development of fungi, especially Fusarium, negatively affects the quality of barley and malt. The malting conditions (temperatures ranging from 16–22 °C and high humidity) are obviously favorable to microbial development, with an amplifying effect of contamination.^[15,16] The first step of the malting process is steeping, consisting of a succession of “under-water” and “air-rest” stages, during which the moisture of the barley grain increases from about 12–14% to 42–46%, and the embryo starts to develop. Despite the important washing effect that occurs during this step, steeping remains a critical point for microbial development on the grain, as spore forms are activated, and bacteria and yeasts significantly grow. The “air-rest” stages also lead to significant fungal development.^[17] The second step of malting is germination, a stage of great metabolic changes by the action of enzymes in the grain, where microbial development is also favored given the conditions (temperatures around 18-20 °C and moisture content of ∼45%). The last step of the malting process is kilning, where the main goal is to kill the embryo and stabilize the germinated grain (malt) by gradual drying (from 50 °C up to 85-110 °C) to a moisture content of 4-5%. Although the malted grain’s microflora is reduced at this step, its microbial load remains higher than that of the ungerminated grain. Mycotoxins persist at this stage as they are highly thermostable and are susceptible to finding their way into the beer during the brewing process,^[18,19] thus causing significant technological issues. For example, deoxynivalenol (DON) has been shown to inhibit both top and bottom fermentation yeasts,^[20] thus affecting their alcohol production capability. Other studies have also shown that technological phenomena such as premature yeast flocculation (PYF) and gushing are associated with Fusarium strains and their metabolites.^[21]

Considering the economic, health, and technological consequences related to pathogenic fungi and their mycotoxins, it is crucial to limit their presence and development on cereals throughout the food production chain. For this purpose, several strategies employing physical, chemical, and biological methods are implemented as preventive or curative approaches. Nowadays, consumers want better quality products with a more natural preservation, limiting the use of chemical means.^[22] Chemical or thermal means do not eliminate all risks, and they often modify the properties of raw materials (barley). Furthermore, in the case of malting barley (used for beer production) where it is vital to preserve a viable embryo for malting, heat treatment is not a good alternative to reduce commodity contamination. Therefore, biological methods have gained interest over the past two decades, especially with the use of microorganisms as biocontrol agents to reduce fungal contaminations. In this context, studies have shown that the use of bacteria,^[23–31] yeasts,^[13,32–35] and fungi^[36] as biocontrol agents can reduce the development of pathogenic fungi and their mycotoxin production, thus improving the quality of malt, and eventually of beer.

In our previous study, we highlighted the ability of a barley-associated Erwinia gerundensis strain to reduce fungal growth and mycotoxin production.^[37] The study showed that Erwinia gerundensis reduced fungal growth by up to 69% and the production of enniatins by 74–100% over a 72h culture in the liquid medium, thus supporting the use of this bacterial strain as a biocontrol agent against pathogenic fungi and their mycotoxins in cereal-based products. Our present study aims to evaluate the impact of Erwinia gerundensis on mycotoxin production by pathogenic fungal species during the malting process and on the technological quality of malt. For this purpose, the antifungal property of the bacterial strain was first confirmed in the liquid medium, then evaluated during co-cultures on barley kernels, and finally during micro-malting assays.

Experimental

Chemicals, reagents, culture media, and barley kernels

The chemicals and solvents used in this study were of analytical grade and were purchased from Sigma Aldrich (France) and Carlo Erba (Val-de-Reuil, France). The composition and suppliers of the culture media used in this study are detailed in Table 1. Once prepared, the culture media were sterilized by autoclaving at 121 °C and 1 bar for 20 min.

Table 1. Culture media composition and suppliers.

Culture medium	Composition	Supplier
Oxytetracycline-Glucose-Yeast Extract Agar (OGYE Agar)	Yeast extract: 5 g/L, Glucose: 20 g/L, Oxytetracycline: 0.1 g/L	Biokar Diagnostics (France)
Potato Dextrose Agar (PDA)	Potato Starch (from infusion): 4 g/L, Dextrose: 20 g/L, Agar: 15 g/L	Difco Labs. (France)
Spezieller Nährstoffärmer Agar (SNA)	K2HPO4: 1 g/L, KNO3: 1 g/L, MgSO4.7H2O: 0.5 g/L, KCl: 0.5 g/L, Glucose: 0.2 g/L, Saccharose: 0.2 g/L, Agar: 20 g/L	–
Yeast Malt Broth (YM)	Glucose: 10 g/L, Malt extract: 3 g/L, Peptic digest of animal tissue: 5 g/L, Yeast extract: 3 g/L	Sigma Aldrich (India)
Yeast Malt Agar (YMA)	Agar: 20 g/L, Glucose: 10 g/L, Malt extract: 3 g/L, Peptic digest of animal tissue: 5 g/L, Yeast extract: 3 g/L	Sigma Aldrich (India)

The barley kernels used in this study belonged to two-row spring barley cultivars named “RGT PLANET” and “EXPLORER”, both harvested in France. They were provided by the French Institute of Beverages, Brewing and Malting (IFBM).

Bacterial strain and culture conditions

The bacterial strain Erwinia gerundensis was used in this study. It was previously isolated from contaminated barley kernels harvested in 2018 in the eastern region of France and identified using the 16S rRNA gene sequencing method. The strain was provided by the Joint Technological Unit UMT OPTIMALT (Nancy, France). It was initially grown in Yeast Malt broth (YM: 21 g/L) at 30 °C for 48 h. Then, the liquid culture was centrifuged (3000 x g for 5 min), and the pellet was recovered in fresh YM broth with 10% glycerol and stored at −80 °C as stock cultures or used to prepare bacterial suspension.

The bacterial suspension was prepared by inoculating 1 mL of the bacterial stock into a 15 mL Falcon tube containing 10 mL of YM broth. After 24 h of incubation at 28 °C, the entire volume was transferred to a 250-mL Erlenmeyer flask containing 100 mL of YM broth, and the flask was incubated at 28 °C in a shaking incubator (210 rpm) for another 24 h. Finally, the bacterial biomass concentration was determined by plate count technique (28 °C, 48h in YM Agar), and the suspension was ultimately used for co-culture assays and micro-malting assays.

Fusarium strains and culture conditions

Two Fusarium strains were used in this study: F. tricinctum 2399 and F. tricinctum 2502. The fungal strains were previously isolated from contaminated barley kernels harvested in 2018 and 2020 from various regions of France, using the official method established by the French Ministry of Agriculture and Food.^[38] They were identified and collected by the French Institute of Beverages, Brewing and Malting (IFBM), who provided them for this study. The Fusarium strains were pre-cultured in Potato Dextrose Agar (PDA: 39 g/L) at 22 °C for 7 days. Then, agar plugs containing the vegetative form (mycelium) of each strain were preserved in microtubes containing sterile mineral oil and maintained at 5 °C as stock cultures or used to prepare Fusarium spore suspensions.

Fusarium strain sporulation was induced in Spezieller Nährstoffärmer Agar (SNA) medium. Firstly, at least 3 plugs of each Fusarium strain from a seven-day-old solid (PDA) pre-culture were inoculated in Petri dishes (diameter 90 mm) containing SNA medium in such a way that the mycelium side was in direct contact with the solid medium and incubated at 22 °C for 7–10 days in a climatic chamber with alternating light and dark. At the end of the incubation time, 3 mL of sterile water was added to each Petri dish to collect the fungal spores. Finally, the spores were counted on a Malassez cell counting chamber and ultimately used for co-culture assays in YM broth and on barley kernels.

Co-culture between Erwinia gerundensis and Fusarium tricinctum in liquid medium

The bacterial and fungal strains were simultaneously co-cultured in 250-mL Erlenmeyer flasks containing 100 mL of YM liquid medium. The bacterial strain was inoculated at a concentration of 4 x 10⁷ CFU/mL into the flasks at the beginning of the co-cultures. Then, F. tricinctum 2399 or F. tricinctum 2502 was inoculated at a concentration of 10⁴ spores/mL. For control conditions, each F. tricinctum strain was inoculated alone (separately) at the same concentrations as those in co-cultures. The flasks were incubated at 28 °C in a rotary shaker operating at 210 rpm for 7 days, with samples taken every 24h from day 3 to day 7. The samples were filtered using a syringe filter (pore size 0.2 µm) and stored in chromatography vials at −25 °C until required for enniatin analysis. All experiments were conducted in triplicate.

Enniatin analysis by LC-MS/MS method

Enniatins (ENN) produced during the cultures throughout the study were determined by the method of liquid chromatography-mass spectrometry (LC-MS/MS), using a Famos-Switchos-Ultimate capillary LC system coupled to a triple-quadrupole mass spectrometer (ThermoFisher Scientific, U.S.A.), as explained in our previous study.^[37] The same method was also used for the quantification of other mycotoxins such as deoxynivalenol (DON), deoxynivalenol-3-glucoside (DON-3G), and T2 and HT2 toxins produced during micro-malting assays. The limits of quantification (LOQ) for ENN, DON, DON3G, T2, and HT2 toxins are 10 µg/kg, 60 µg/kg, 15 µg/kg, 3 µg/kg, and 6 µg/kg, respectively.

Co-culture between Erwinia gerundensis and Fusarium strains on barley kernels

Co-culture experiments were performed in 500 mL bottles containing 100 g of barley kernels hydrated with 100 mL distilled water. The barley grains were left in contact with water for 1 h to allow the grains to absorb the water before being sterilized at 121 °C for 20 min. After cooling overnight at ambient temperature, the grains were artificially contaminated by Fusarium tricinctum 2399 or Fusarium tricinctum 2502 at a concentration of 10⁷ spores/kg of barley for each fungal strain, then inoculated with Erwinia gerundensis at a concentration of 4 x 10¹⁰ CFU/kg of barley. For control conditions, each microorganism was inoculated alone (separately) at the same concentrations as those in co-cultures. A negative control was prepared with sterilized barley kernels without microorganism inoculation. The bottles were incubated at 28 °C in alternating light/darkness for 4 days, with manual agitation three times per day. At the end of the incubation time, bacterial and Fusarium biomass as well as enniatin concentrations were determined. All experiments were conducted in triplicate.

Determination of bacterial and fungal biomass and enniatin concentration during co-cultures on barley kernels

For the determination of bacterial and Fusarium biomass concentrations, 25 g of post-culture barley kernels were diluted (1/5) in a diluent prepared as follows: 1.6 g/L tryptone, 8.5 g/L sodium chloride, and 0.033 g/L Tween80. The mixture was left in contact for 25 min, with manual shaking every 5 min. Then, the supernatant was collected and used to determine biomass concentrations. The bacterial biomass was determined by the plate count method using YM agar medium (28 °C, 48 h), and the fungal biomass was determined by the same method using Oxytetracycline-Glucose-Yeast Extract Agar (28 °C, 5 days). At the same time, 5 g of post-culture barley kernels were dried at 105 °C for 24 h for the determination of dry weight percentage. Biomass concentrations of the microorganisms were finally expressed as CFU/kg of dry barley kernel.

For enniatin quantification, 70 g of post-culture barley kernels were dried at 40 °C for 24 h, then ground into very fine particles before being analyzed by the LC-MS/MS method described above.

Malting parameters and application of Erwinia gerundensis

Malting was performed in a micro-malting plant equipped with software for controlling malting parameters owned by IFBM (Vandœuvre-lès-Nancy, France). Micro-malting assays were carried out in individual boxes containing 1.2 kg of naturally contaminated barley kernels belonging to two-row spring barley cultivars obtained from IFBM (Vandœuvre-lès-Nancy, France). The kernels were calibrated using a sieve, and only those with a size of > 2.5 mm were used, as per malting requirements.

The three main stages of the malting process include steeping, germination, and kilning. To determine the optimal inoculation rate, three different inoculation rates of E. gerundensis were tested: 4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg of barley. Then, the optimal inoculation rate (of the three inoculation rates) was tested again in triplicate. In both cases, the bacterial strain was inoculated at the beginning of the steeping step. Steeping was performed at 18 °C for 28h with alternating under-water/air-rest intervals at 8/17/3 h in steeping tanks equipped with pumps to ensure proper mixing. At the end of steeping, the barley kernels were transferred to the germination chamber, where they were left to germinate at 16 °C for 4 days. Finally, the germinated kernels were kilned (dried) with a hot air stream at different temperatures using the following kilning regime: 50 °C for 8 h, 64 °C for 7 h, 80 °C for 3 h, and 85 °C for 3 h. At the end of kilning, the rootlets were removed from the kilned malts.

Samples were taken throughout the malting process, especially before steeping, at the end of the first under-water (immersion) step, at the end of the steeping step, at the end of the germination step, and at the end of the kilning step. Then, these samples were used for the monitoring of E. gerundensis throughout the malting process via metagenomic analysis, for the determination of fungal biomass concentration, mycotoxin concentrations (DON, DON-3G, ENN, T2, and HT2 toxins), and malt quality parameters.

Monitoring of Erwinia gerundensis inoculum throughout the malting process

The monitoring of E. gerundensis (EG) throughout the malting process was performed through the characterization of the composition and diversity of the bacterial communities using a high throughput sequencing-based protocol that targets PCR-generated amplicon. Prior to the analysis, the barley samples were finely ground, and about 5 g of each sample was used for each condition: untreated barley (sample M1), barley treated with 4 x 10¹⁰ CFU of EG/kg (sample M2), barley treated with 4 x 10⁷ CFU of EG/kg (sample M3), and barley treated with 4 x 10⁴ CFU of EG/kg barley (sample M4). A fresh liquid suspension of E. gerundensis (obtained as explained above) was also used for sequence comparison purposes (monitoring) in the samples. Microbial DNA extractions were performed on each sample to extract DNA from both epiphyte and endophyte microorganisms. Total genomic DNA was extracted from the sample powder using the NucleoSpin Soil Kit (Macherey-Nagel, Germany), following the manufacturer’s instructions. Bacterial communities were characterized from the variable regions V3-V4 of the 16S rRNA gene using the universal primers 341 F (5-CCTAYGGGRBGCASCAG-3) and 806 R (5-GGACTACNNGGGTATCTAAT-3).^[39] DNA amplification was performed by PCR using the Type-it Microsatellite PCR kit (Qiagen, Hilden, Germany) in a total volume of 15 µl containing 1X master mix, 0.13 µM of each primer, and 2 µl of ten-fold dilution genomic DNA under the following conditions: 95 °C for 5 min, 35 cycles of 30 s at 95 °C, 55 °C for 1 min and 72 °C for 30 s, followed by 30 min at 60 °C. Next-generation sequencing library constructions were performed following the manufacturer’s instructions (Illumina, U.S.A.). Negative controls from the extraction step and PCR reaction were sequenced with the samples to evaluate and exclude contaminant reads from the sample data set. The library was sequenced with Illumina’s paired-end 2 × 250-bp technology and v2 chemistry. Then, the sequences were analyzed by FROGS pipeline.^[40] Like the other pipelines, FROGS also relies on the principles of sorting, cleaning, and simplifying the data from sequencing. All analyses were performed by ADNID laboratory (Montferrier-sur-Lez, France), a subsidiary of the French Institute of Beverages, Brewing and Malting (IFBM).

Determination of fungal load and mycotoxin content of malt

Fungal biomass concentration, as well as mycotoxin (DON, DON-3G, ENN, T2, and HT2 toxins) concentrations of malt, were determined following the protocol used for the same analyses during co-culture assays on barley kernels, as explained above.

Determination of malt quality parameters

The technological quality of the malt obtained from the malting process was evaluated based on the following parameters: extract fine grind (%DM, EBC method 4.5.1), moisture content (%, EBC method 4.2), total protein (%DM, EBC method 4.3 Dumas), soluble protein (%DM, EBC method 4.9 Dumas), Kolbach Index (soluble to total protein ratio), diastatic power (°WK, EBC method 4.12), wort viscosity (mPa.s, EBC method 4.8), friability (%, EBC method 4.15), β-glucan (mg/L, EBC method 8.13.2), free amino nitrogen (mg/L, EBC method 4.10), Tepral filtration, and wort pH. Some of these analyses were performed on the malt grain, while others were performed on the malt wort. All analyses were performed following the official methods recommended by the European Brewery Convention.^[41]

Statistical analysis

One-way ANOVAs, followed by Tukey’s multiple comparison tests, were used to evaluate the differences between control and co-culture conditions in a liquid medium and on barley kernels, as well as for the comparison of micro-malting results. Differences were statistically significant when the p-value < 0.05. Graphical values and values in the tables are presented as mean ± standard deviation unless stated otherwise. The statistical analysis was performed using Minitab Statistical Software, version 21.1.1.0 (Minitab Inc., U.S.A.).

Results and discussion

In vitro antifungal activity of Erwinia gerundensis in YM liquid medium

The impact of the bacterial strain E. gerundensis on enniatin production was evaluated via a simultaneous co-culture between the bacterium and enniatin-producing Fusarium strains, F. tricinctum 2399 and F. tricinctum 2502. During the co-culture experiments, total enniatin (ENN) concentration was determined using LC-MS/MS method every day from 72 h to 168 h. In the control culture, F. tricinctum 2399 exhibited its highest ENN production at 72 h (40504 µg/kg ± 6742), and ENN concentrations detected from 96 h to 168 h (between 21321 µg/kg ± 2973 and 24761 µg/kg ± 1094) were not significantly different from each other (p = 0.472) (Figure 1, Panel A). In co-culture (F. tricinctum 2399 + E. gerundensis 6E4), ENN concentrations significantly and radically decreased, ranging from 438 µg/kg (± 44) to 555 µg/kg (± 77). In other words, the bacterial strain reduced ENN concentrations by 98.6%, 97.9%, 98.0%, 98.2%, and 97.7% at 72, 96, 120, 144, and 168 h, respectively. E. gerundensis also exhibited similar ENN-reducing capacity against the fungus F. tricinctum 2502. ENN production of the fungus ranged from 22703 µg/kg (± 1769) to 29408 µg/kg (± 1697), with the highest quantifications occurring at 96 h and 168 h (Figure 1, Panel B). When co-cultured with E. gerundensis, ENN production of the fungus was reduced by over 99.7% throughout the incubation time. The anti-mycotoxin activity of the bacterium E. gerundensis could be linked to several mechanisms such as competition for resources and space, inhibition of ENN biosynthesis pathways, and transformation of ENN into non-toxic metabolites. These results confirmed that E. gerundensis is not only effective against enniatin biosynthesis at 72 h, as evidenced in our previous study^[37] but also that its efficiency could last for at least 7 days, a period of time that is similar to that of the malting process, where the bacterial strain could eventually be applied as a biocontrol agent.

Figure 1. Enniatin (ENN) concentration in control (F. tricinctum alone) and co-culture (F. tricinctum + E. gerundensis) experiments in liquid medium. Panel A: Co-culture between F. tricinctum 2399 and E. gerundensis 6E4; Panel B: Co-culture between F. tricinctum 2502 and Erwinia gerundensis 6E4. Histograms (mean ± standard deviation) with different letters (a, b, c) are significantly different (p < 0.05). *Total ENN refers to the sum of enniatins A, A1, B, and B1.

Effect of Erwinia gerundensis on Fusarium tricinctum and enniatin production on barley kernels

The impact of E. gerundensis on fungal growth and enniatin (ENN) production was evaluated during simultaneous co-cultures between the bacterium and two strains of F. tricinctum (2399 and 2502) on barley kernels for 4 days at 28 °C, with the aim of evaluating the implementation of the bacterial strain and validating its efficiency against Fusarium mycotoxinogenesis on the barley matrix prior to malting. The results showed that the bacterial strain not only successfully colonized the barley matrix, reaching a final biomass of approximately 10¹³ CFU/kg of barley in 4 days, but also exhibited the same growth pattern regardless of the culture conditions (control culture or co-culture) (Table 2). The growth of the fungus F. tricinctum 2502 was not affected by the co-culture, as the fungus yielded a biomass concentration of approximately 6 x 10⁹ CFU/kg of barley at the end of the incubation time in both the control and co-culture conditions. On the contrary, the growth of F. tricinctum 2399 was affected by the co-culture. In control conditions, this fungus reached a biomass concentration of 1.8 x 10¹⁰ CFU/kg at the end of the incubation time, whereas its growth was reduced by approximately 82% in the presence of the bacterial strain in co-culture, suggesting that the strain F. tricinctum 2399 showed a higher sensitivity towards E. gerundensis in terms of growth.

Table 2. Biomass and enniatin concentrations during co-culture on barley kernels. For total ENN, values (mean ± standard deviation) with different letters (a, b, c, d) are significantly different (p < 0.05).

	E. gerundensis			F. tricinctum 2399		F. tricinctum 2502
	Control	vs. TRI 2399	vs. TRI 2502	Control	vs. EG	Control	vs. EG
Initial biomass (CFU/kg)	4.0 x 10^10	4.0 x 10^10	1.0 x 10^10	1.0 x 10^7	1.0 x 10^7	1.0 x 10^7	1.0 x 10^7
Final biomass (CFU/kg)	1.3 x 10^13	1.4 x 10^13	1.2 x 10^13	1.8 x 10^10	3.2 x 10^9	5.5 x 10^9	6.0 x 10^9
Total ENN (µg/kg)	NA	NA	NA	625 ± 89^c	199 ± 55^d	3507 ± 74^a	2167 ± 704^b

TRI: Fusarium tricinctum; EG: Erwinia gerundensis; ENN: enniatin; NA: Not Applicable.

As far as fungal mycotoxinogenesis is concerned, both fungi produced ENN at relatively high concentrations (Table 2). In control conditions, the concentration of ENN produced by F. tricinctum 2399 was 625 µg/kg (±89) whereas only about one-third of this concentration was detected in co-culture, meaning that the toxinogenic potential of the fungus was significantly reduced by about 68% by E. gerundensis (p < 0.05). Similarly, ENN production of F. tricinctum 2502 was significantly reduced by 38% by the bacterial strain, thus decreasing from 3507 µg/kg (±74) in control culture conditions to 2167 µg/kg (±704) in co-culture conditions (p < 0.05). Analyzing the results, it is important to emphasize that the fungal strain F. tricinctum 2399 proved once again to be more sensitive towards E. gerundensis in terms of ENN biosynthesis. The ENN reduction percentages achieved for both fungi on barley kernels (38–68%) were lower than those obtained in the liquid medium (over 98%), suggesting that there may have been a matrix effect. This phenomenon could be explained by the physical composition of the media. Solid-state fermentations are much more complex than fermentations in liquid medium due to the matrix effect generated by the solid medium and the different water activity (a_w) of liquid and solid media. As such, the solid medium may limit mass transfer and probably the activity of the bacterial strain. Hence, the bacterial strain may have benefited from a better fermentation environment in the liquid medium to exhibit its antifungal activity. In addition, the behavior of mycotoxins is known to be strongly influenced by the matrix, as mycotoxins can bind with several matrix components, such as starch or proteins, through the formation of covalent adducts.^[42,43]

Taken together, the results show that E. gerundensis is efficient at reducing ENN biosynthesis both in an in vitro liquid fermentation system and on the barley matrix. To the best of our knowledge and after thorough coverage of the available literature about the biological reduction of ENN, this is the first report of ENN reduction by E. gerundensis on a barley matrix. However, there are a few references about bacterium-mediated biological reduction of Fusarium mycotoxins in cereal matrices. For instance, Roig et al.^[44] used the bacteria Bb. adolescents, Lb. rhamnosus, and S. thermophilus to reduce ENN contents in wheat flour by 15–50%, depending on the type of ENN and on the bacterium. The effect of two strains of P. fluorescens on deoxynivalenol (DON) accumulation in wheat and barley kernels was studied by Khan and Doohan,^[45] who reported that both strains reduced DON levels in both cereals by 74–78% in field conditions and that one of the strains (MKB 158) reduced the DON content of wheat and barley by 82% and 98% respectively under glasshouse conditions.

We did not directly investigate how Erwinia gerundensis exhibits its mycotoxin-reducing ability in this study. However, the decrease in the final biomass concentration and in ENN production of the fungus F. tricinctum 2399 when co-cultured with the bacterium hints to the fact that the ability of E. gerundensis to reduce mycotoxin production in the grains might arise from the fact that it prevents or inhibits fungal growth and consequently mycotoxin production. E. gerundensis managed to reduce ENN production of the fungus F. tricinctum 2502 on barley kernels by about 38% without inhibiting its growth, suggesting that the bacterial strain might have exerted its action by directly inhibiting mycotoxin production or by degrading the mycotoxin. The latter mechanism has been evidenced by many authors for several Fusarium mycotoxins. In particular, Roig et al.^[44] evaluated the degradation of ENN by nine bacteria of the gastrointestinal tract (Bb. longum, Bb. bifidum, Bb. breve, Bb. adolescents, Lb. rhamnosus, Lb. casei-casei, S. thermophilus, Lb. ruminis, and Lb. casei). All nine bacteria significantly degraded ENN, yielding three degradation products that were characterized as ENN B lacking a structural component represented by the hydroxyisovaleric acid, ENN B lacking two molecules of water and a methyl group, and ENN B1 lacking the valine unit. Likewise, the degradation of ENN B and ENN B1 by six Bacillus subtilis strains was studied by Meca et al.,^[46] and all of the tested strains were able to degrade both types of ENN, with degradation products resulting from the loss of hydroxyisovaleric acid unit in the structure of ENN B and ENN B1. Suchfort^[25] also reported on the ability of a strain of Bacillus licheniformis to degrade ENN, resulting in the hydrolysis of an ester or an amide bond in the cyclic ring of ENN, thus forming four degradation products. In another study, Meca et al.^[47] evaluated the interaction between the Fusarium mycotoxin beauvericin (BEA), a minor mycotoxin structurally similar to ENN, by thirteen bacterial strains characteristic of the gastrointestinal tract, and concluded that the mycotoxins BEA and ENN could be degraded by bacteria of different species. Awad et al.^[48] described the degradation properties of a Eubacterium strain (BBSH 797) that completely transformed deoxynivalenol (DON) to a less toxic chemical de-epoxy DON (DOM-1). The transformation of DON into its lesser toxic forms, such as 3-keto-4-deoxynivalenol (3-keto-DON) and 3-epi-DON has also been documented for the bacteria Agrobacterium–Rhizobium,^[49] Nocardioides WSN05-2,^[50] and Devosia.^[51,52] Other studies have also reported on the degradation of zearalenone (ZEA) by bacteria belonging to the genera Bacillus,^[53,54] Pseudomonas,^[55] and Rhodococcus,^[56] with ZEA degradation percentages ranging from 50 to 85%, depending on the bacterium and the matrix.

The ability of Erwinia gerundensis to reduce ENN production in both liquid and solid-state fermentation systems being particularly striking, the strain was deemed appropriate to be employed in the biocontrol assays during malting.

Developing the application protocol of Erwinia gerundensis in malting

This part of the study was devoted to the development of the application protocol of Erwinia gerundensis in malting in order to evaluate the bioprotective capability of the bacterial strain. For this purpose, a preliminary micro-malting assay was carried out (without replications) with three different bacterial inoculations (4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg of barley) applied at the beginning of the steeping step of malting. The goal was to determine the lowest bacterial inoculation rate that could be used to achieve the desired effects with respect to the assessment parameters of the bioprotection capability of the bacterium, such as the implantation of E. gerundensis on barley kernels throughout the malting process, the fungal load and mycotoxin content of malt, and the technological quality of malt. Therefore, all results regarding three different bacterial inoculation rates (4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg of barley) tested at the same time were obtained from a micro-malting experiment that was performed only once (no replications). The subsequent results are based on the optimal bacterial inoculation rate (as determined based on the first micro-malting experiment), and were obtained from micro-malting experiments performed in triplicate.

Implantation of Erwinia gerundensis on barley kernels during malting

For the micro-malting assay with three different inoculation rates of E. gerundensis, samples were collected before and during malting (at different stages of the malting process) to monitor the relative abundance of E. gerundensis inoculum throughout the malting process. Sequence data of each sample analyzed using FROGS pipeline^[40] yielded about 23136 normalized sequence reads per sample. The relative abundance of E. gerundensis (for each of the three inoculation rates tested) with respect to the total bacterial community of each sample is shown in Figure 2.

Figure 2. Relative abundance of E. gerundensis (EG) throughout the malting process. Expressed in number of reads of EG sequences per a total of 23136 reads. M1: no EG treatment; M2: barley treated with 4 x 10¹⁰ CFU EG/kg; M3: Barley treated with 4 x 10⁷ CFU EG/kg; M4: barley treated with 4 x 10⁴ CFU/kg; Seed: unmalted barley.

The sequence data analysis revealed a relative abundance ranging from 0 to 15% for E. gerundensis across all samples. An analysis of variance (ANOVA) followed by a Tukey test on the number of E. gerundensis sequences in each sample showed that the number of sequences in M1 and M4 samples differed significantly at p < 0.05; neither M1 sample nor M4 sample was significantly different from the other two samples (M2 and M3). Hence, considering the number of reads (Figure 2), E. gerundensis is significantly less abundant in M1 compared to M4. This is a logical observation as there is no additional E. gerundensis inoculation in sample M1. Another important finding is that the presence of E. gerundensis logically seems to be more pronounced in the inoculated samples (M2, M3, and M4) compared to the control sample M1. More interestingly, it was found that E. gerundensis was present throughout each step of the malting process, regardless of the treatment dose, thus suggesting that the bacterial strain had successfully implanted itself and developed on the barley grains during malting. This is a crucial part of the bioprotective activity expected of E. gerundensis, as it would need to settle down, adapt, and grow in its new environment before being able to exert its bioprotection capability.

Effect of different doses of Erwinia gerundensis on fungal load and mycotoxin content of malt

The effect of three different inoculation rates of Erwinia gerundensis (4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg of barley) on fungal load and mycotoxin content of malt was evaluated through micro-malting. At the end of malting, the LC-MS/MS method was used to determine the mycotoxin (DON, DON3G, ENN, T2 and HT2 toxins) content of malt, while its fungal biomass concentration was determined by the classical agar-plating method. As shown in Figure 3, the fungal load of the original barley (∼ 9.5 x 10⁵ CFU/kg of barley) increased by about 86%, reaching 1.8 x 10⁶ CFU/kg for the resulting (untreated) malt. This can be explained by the fact that the fungi originally present on barley are met with good environmental conditions (high water activity and temperature around 18 °C) during the malting process, which favored their growth. The addition of E. gerundensis before steeping resulted in a reduction of the fungal load of malt by 72, 67, and 69% for the bacterial inoculation rates 4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg of barley, respectively, when compared to untreated malt. This shows that E. gerundensis has proven to be efficient at controlling the total fungal load of malted barley, regardless of the employed inoculation rate. The results also suggest that there was no bacterial dose effect on the extent of reduction of the fungal biomass concentration in the final malt, as the reduction percentages are quite similar for all three inoculation rates.

Figure 3. Effect of E. gerundensis on the total fungal load of malt. Error bars represent the standard error.

For further evaluation of the bioprotective capability of E. gerundensis during malting, enniatin (ENN), deoxynivalenol (DON), deoxynivalenol-3-glucoside (DON3G), T2 and HT2 toxins contents of malt were quantified. Although E. gerundensis was found to be efficient against the reduction of ENN on barley kernels, it was decided to evaluate its efficiency against DON and DON3G mycotoxins, as they are the most frequently detected Fusarium mycotoxins in malt and beer samples, and they have been reported to cause the majority of mycotoxin-related problems in malt and beer.^[57] Also, T2 and HT2 toxins were included in the evaluation as these are the most toxic compounds in the type-A trichothecenes family.^[3] The concentrations of ENN, DON, DON3G, and T2 + HT2 toxins in unmalted barley were 679 µg/kg, 409 µg/kg, 47 µg/kg, and 85 µg/kg, respectively. The final concentrations of ENN, DON, and DON3G in malt increased by 19-fold, 1.6-fold, and 10-fold, respectively, whereas the concentration of T2 + HT2 toxins decreased by 64%, as compared to those of unmalted barley (Figure 4). ENN emerged as the highest mycotoxin produced in untreated malt. This was an expected result, as the barley kernels used in the experiment (RGT PLANET cultivar) were selected for their high contamination by Fusarium tricinctum, one of the main ENN producers. The addition of E. gerundensis at the beginning of the steeping step resulted in different effects on the mycotoxins, depending on the employed bacterial inoculation rates and on the type of mycotoxin. For instance, the ENN content of the malt obtained with the application of E. gerundensis was reduced by 50%, 32%, and 43% for bacterial inoculation rates of 4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg barley respectively, compared to the untreated malt (Figure 4, Panel A). However, considering the uncertainty of the LC-MS/MS method that was used to quantify the mycotoxins (15%), the reduction percentages of ENN were not different among the inoculation rates, indicating that there was no bacterial dose effect on the extent of ENN reduction. As for DON, its concentration was reduced by 18%, 33%, and 15% in the malt treated with E. gerundensis at 4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg barley, respectively, compared to that of the untreated malt (Figure 4, Panel B). Similarly, DON reduction percentages were quite similar among the three bacterial inoculation rates within the uncertainty of the LC-MS/MS, thus suggesting once again that there was no bacterial dose effect on the extent of DON reduction. An opposite trend was observed with DON3G contents of the treated malt. Malts treated with E. gerundensis were found to be more contaminated in DON3G than their untreated counterparts (Figure 4, Panel C). This effect was more pronounced with the two highest bacterial inoculation rates (4 x 10⁷ CFU/kg and 4 x 10¹⁰ CFU/kg), where an increase of about 50% was observed in DON3G content of malt treated with these two rates. Contamination levels of T2 + HT2 toxins in treated malts also increased by 55% and 190%, respectively, with the addition of E. gerundensis by 4 x 10⁴ CFU/kg and 4 x 10¹⁰ CFU/kg. For the malt sample treated with 4 x 10⁴ CFU/kg of E. gerundensis, T2 + HT2 toxins were reduced by 63% compared to the untreated malt. This phenomenon of increasing mycotoxin contamination levels (for DON3G and T2 + HT2 toxins) with increasing E. gerundensis dose suggests that, when the dose of the biocontrol agent is relatively high, it might become an important stress factor for Fusarium species that produce these toxins, thus boosting their metabolite (including mycotoxin) production as a response to the stress, as mycotoxins are thought to serve as a means of protection for Fusarium species against other microorganisms.^[58]

Figure 4. Effect of E. gerundensis (EG) on mycotoxin content of malt. A: ENN; B: DON; C: DON3G; D: T2 + HT2 toxins. Error bars represent the uncertainty of the LC-MS/MS method (15%).

The results have shown that contamination levels for enniatin (ENN), deoxynivalenol (DON), and deoxynivalenol-3-glucoside (DON3G) have strongly increased during malting, resulting in 1.6 to 19 times higher levels of these mycotoxins in the malted barley (untreated), compared to the unmalted one. This is mainly due to the temperature and moisture conditions that barley is subjected to in malting, which provide a favorable environment for Fusarium growth and, consequently, for mycotoxin production. However, it has been reported that the steeping process removes water-soluble mycotoxins initially present in cereal grains such as DON, although the rest of the malting process (especially germination) leads to their production in the final malt.^[59] Hence, the steeping step would not have much of an effect on mycotoxins with very low water solubility, such as ENN.^[60] This is probably why ENN was the mycotoxin to have shown the highest increase (19-fold) in its contamination levels from barley to malt in our study. Considering that ENNs have received very little attention regarding their cytotoxic effect compared to the other mycotoxins,^[61] this is another example of the risks posed by ENNs that will require more research attention, especially in the malting and brewing industries where mycotoxins may pose serious threats.

Our study has shown the effectiveness of Erwinia gerundensis in reducing the total fungal load and the mycotoxin content of malt, with the lowest dose (4 x 10⁴ CFU/kg) yielding the best results. This is an interesting finding as it is the first report of E. gerundensis being evaluated as a biocontrol agent for barley malt protection against mycotoxin-producing fungi and their mycotoxins during malting. Various biocontrol agents have been studied and proposed for the bioprotection of malt against fungi, and some are being used in the malting industry. For instance, IFBM has patented a biocontrol process where Geotrichum candidum is successfully used to reduce fungal growth and minimize mycotoxin production (especially T2 and HT2 toxins) in barley malting. However, some strains of G. candidum possess lipase activity that may lead to the production of undesirable oxidation products.^[13] The fungus Pythium oligandrum has been used as a starter culture to suppress the growth of DON- and DON3G-producing Fusarium culmorum in the malting of artificially contaminated barley kernels.^[62] The same fungus has recently been used as a biocontrol agent for wheat malt protection during malting, where its treatment before the steeping stage resulted in 88%, 87%, and 85% reduction of Fusarium culmorum contamination, DON, and DON3G production, respectively, in wheat malt.^[36] Moreover, P. oligandrum has been approved as a biocontrol agent for plant protection by the European Union.^[63] Bacteria, and especially lactic acid bacteria (LAB), have also been reported to reduce the growth of mycotoxinogenic fungi and the production of the most studied mycotoxins, such as DON and zearalenone (ZEA) in malting.^[17,23,64] However, some bacterial strains belonging to the genera Lactobacillus, Pediococcus, Pectinatus, and Megasphaera are considered to be principal spoilage agents in the brewing industry.^[65]

As presented above, none of the biocontrol agents proposed in malting to date was directed toward the reduction of ENN or ENN-producing fungi. Our study contributes to filling up the gap by proposing E. gerundensis as a biocontrol agent applicable in the malting industry. However, it is still crucial to evaluate the effect of E. gerundensis on the technological quality of malt.

Effect of different doses of Erwinia gerundensis on the technological quality of malt

The technological quality of the malted barley samples was evaluated based on amylolytic (enzyme activity, diastatic power), cytolytic (friability, β-glucan content, viscosity), and proteolytic (total protein content, soluble protein content, Kolbach index, free amino nitrogen) criteria in order to investigate the effect of E. gerundensis treatment on the malt quality. The quality parameter values of the untreated malt were all found to be within the required range for a malt designated for brewing (Table 3).

Table 3. Technological quality of malted barley with different doses of E. gerundensis (EG) applied before steeping.

Malt quality parameters	Reference range*	Untreated malt	Malt with 4 x 10^4 CFU EG/kg barley	Malt with 4 x 10^7 CFU EG/kg barley	Malt with 4 x 10^10 CFU EG/kg barley
Diastatic Power (°WK)	> 250	290	300	290	250
Extract fine grind (% DM)	> 80	83.7	83.3	83.2	83.7
Friability (% flour)	> 80	90	87	88	90
Total β-glucan (mg/L)	< 200	219	233	210	173
Wort viscosity (mPa.s)	1.5 − 1.7	1.54	1.52	1.54	1.55
Free amino nitrogen (mg/L)	> 140	191	179	179	196
Total protein (% DM)	10 − 11.5	10.4	10.2	10.3	10.2
Soluble protein (% DM)	3.5 − 4.5	4.8	4.8	4.8	4.8
Kolbach index (%)	> 35	46	47	47	47
Filtration time (min per 350 g of wort)	NA	33.8	30.2	31.9	30.4
Moisture content (%)	< 4.5	3.6	3.6	3.6	3.6
Wort pH	5.7 − 6	5.8	5.8	5.8	5.8

*Reference range was provided by the French Institute of Beverages, Brewing and Malting (IFBM), member of the Analysis Committee of the European Brewery Convention; NA: Not Applicable.

Except for total β-glucan and diastatic power values, the addition of E. gerundensis at different doses before the steeping phase did not result in any notable changes in the technological quality parameter values of the treated malt compared to the untreated malt. The value of total β-glucan obtained for the malt treated with E. gerundensis at 4 x 10¹⁰ CFU/kg barley (173 mg/L) was about 21%, 26%, and 18% lower than those obtained in the untreated malt, and in the malt treated with 4 x 10⁴ CFU/kg and 4 x 10⁷ CFU/kg barley respectively. This suggests that the highest bacterial treatment dose was interesting in terms of the β-glucan content of the malt wort, as the β-glucan concentration recommended for a malt designated for brewing purposes is < 200 mg/L, and a good malting performance is associated with low levels of β-glucans.^[66–69] A high β-glucan content is also known to increase wort viscosity, which causes reduced filtration and unwanted beer turbidity.^[68] However, the wort viscosity values observed for all of the malt samples in our study were quite similar, regardless of the β-glucan content. It was observed that it took about the same amount of time (∼ 30 min) to filter 350 g of malt wort obtained from malt samples treated with 4 x 10⁴ CFU E. gerundensis/kg barley or 4 x 10¹⁰ CFU E. gerundensis/kg barley, where the β-glucan contents were different (233 mg/L and 173 mg/L respectively), thus suggesting that the differences in β-glucan content observed in our study were not significant enough to impact the wort viscosity or the filtration.

The three different doses of Erwinia gerundensis treatment (4 x 10⁴ CFU/kg, 4 x 10⁷ CFU/kg, and 4 x 10¹⁰ CFU/kg barley) yielded acceptable values for the technological quality parameters. In the previous experiment, the lowest treatment dose was found to be the optimal dose that yielded the best results in terms of the mycotoxin content of the malt (Figure 4). Therefore, the lowest treatment dose (4 x 10⁴ CFU of E. gerundensis/kg of barley) was chosen as the optimal one and was applied during the subsequent micro-malting experiments where bacterial treatment was required.

The bioprotective capability of Erwinia gerundensis during malting

After determining the optimal treatment dose of E. gerundensis (4 x 10⁴ CFU/kg of barley) as explained above, subsequent micro-malting experiments were performed in triplicate with this treatment dose inoculated at the beginning of the steeping step. The barley kernels used in these experiments belonged to a two-row spring barley cultivar named “EXPLORER” and were naturally contaminated by ENN (491 ± 256 µg/kg) and T2 & HT2 toxins (84 ± 57 µg/kg). The goal was to confirm the effect of E. gerundensis on the mycotoxin content and the technological quality of the final malt.

As shown in Table 4, the mycotoxin content of the untreated malt was 259 µg/kg (± 265) for ENN and 33 µg/kg (± 34) for T2&HT2 toxins, showing that the malting process led to a 47% and 61% reduction of ENN and T2&HT2 toxins respectively in the final malt, as compared to the original barley; however, the reductions were not statistically significant (p > 0.05). The addition of E. gerundensis at a concentration of 4 x 10⁴ CFU/kg of barley at the beginning of the steeping step of the malting process resulted in a reduction of ENN and T2 + HT2 toxins content of the final malt by 38% and 22%, respectively, as compared to the untreated malt. However, these reductions were not found to be statistically significant as they all returned p-values > 0.05. This might be due to the huge variability in mycotoxin concentration values, as shown by the high standard deviations. Some reasons that could justify this outcome are: i) the fact that the samples were highly diluted before analysis by the LC-MS/MS method might have led to large differences in the results; ii) the proposed biocontrol agent (Erwinia gerundensis) may have been unable to effectively sustain its viability and a sufficient antifungal or anti-mycotoxin activity during the malting process in these experiments. It should be noted that the temperature range applied during the steeping step of the malting process (16–18 °C) is not favorable for the growth of E. gerundensis, as it is a mesophilic strain with growth temperatures ranging from 20 to 45 °C and optimal growth temperatures in the range of 30–39 °C.^[70] Nevertheless, these findings are interesting as they highlight the anti-mycotoxin activity of E. gerundensis during malting, although the application process should be improved for more robustness.

Table 4. Effect of E. gerundensis on the mycotoxin content of malt. Values (mean ± standard deviation) with the same letters are not significantly different (p > 0.05).

Conditions	ENN (µg/kg)	T2 + HT2 (µg/kg)
Unmalted barley	491 ± 256^a	84 ± 57^a
Untreated malt	259 ± 265^a	33 ± 34^a
Malt treated with E. gerundensis	160 ± 66^a	26 ± 14^a
P-value	0.236	0.361

The malt samples obtained from the micro-malting experiments were subjected to further analysis in order to investigate the impact of E. gerundensis on the technological quality of the final malt. For most of the quality parameters, the results showed that there was no statistically significant difference (p > 0.05) between the untreated malts and the malts treated with E. gerundensis (Table 5). For malt quality parameters such as wort viscosity (p = 0.047) and wort pH (p < 0.001), where statistically significant differences were observed, the values were still within the required range in the malting and brewing industries.

Table 5. Effect of E. gerundensis on the technological quality parameters of malt. Values (mean ± standard deviation) with different letters are significantly different (p < 0.05).

Malt quality parameters	Reference range*	Untreated malt	Malt treated with E. gerundensis	P-value
Diastatic Power (°WK)	> 250	273 ± 6^a	280 ± 0^a	0.116
Extract fine grind (% DM)	> 80	82 ± 0.3^a	82 ± 0.2^a	0.877
Friability (% flour)	> 80	92 ± 0.6^a	92 ± 0.6^a	1.000
Total β-glucan (mg/L)	< 200	132 ± 14^a	127 ± 3^a	0.585
Wort viscosity (mPa.s)	1.5 − 1.7	1.51 ± 0.0^a	1.49 ± 0.0^b	0.047
Free amino nitrogen (mg/L)	> 140	192 ± 1.7^a	190 ± 0.6^a	0.189
Total protein (% DM)	10 − 11.5	10.4 ± 0.0^a	10.4 ± 0.2^a	0.725
Soluble protein (% DM)	3.5 − 4.5	5.0 ± 0.2^a	5.1 ± 0.0^a	0.374
Kolbach index (%)	> 35	48 ± 2^a	49 ± 1^a	0.275
Filtration time (min per 350 g of wort)	NA	26.3 ± 2.9^a	21.6 ± 4.6^a	0.205
Moisture content (%)	< 4.5	3.8 ± 0.0^a	3.9 ± 0.1^a	0.116
Wort pH	5.7 − 6	5.8 ± 0.0^a	5.6 ± 0.0^b	< 0.001

*Reference range was provided by the French Institute of Beverages, Brewing and Malting (IFBM), member of the Analysis Committee of the European Brewery Convention; NA: Not Applicable.

Each malt quality parameter contributes to providing a good assessment of the level of the physicochemical modification that the barley kernels undergo as they are transformed into malts designated for beer production. The diastatic power (DP), which is one of the main parameters, expresses the activity of enzymes that are produced during the malting process. These enzymes are important for transforming barley starch into fermentable sugars essential in beer production. In our study, the treated and the untreated malt samples presented similar DP values (280 ± 0°WK and 273 ± 6°WK), and these are considered acceptable in malting and brewing industries and showed that the bacterial strain E. gerundensis did not have any negative impact on the malt’s enzymatic activity. Another important parameter is wort viscosity (WV). High viscosity is known to affect the rheological properties of wort and to cause filtration issues during brewing,^[68] whereas low viscosity was reported to be an indicator of a successful malting, reflecting the cytolytic activity in the barley kernels.^[71] The viscosity of the untreated malt (WV = 1.51 mPa.s) and treated malt (WV = 1.49 mPa.s) samples generated in this study were all < 1.53 mPa.s, and WV values < 1.53 mPa.s are classified as very good and are considered very appropriate for the filtration process.^[67] Free amino nitrogen (FAN), Kolbach index (KI), and soluble protein (SP) are parameters that reflect the degree of proteolysis that barley proteins have undergone during the malting process. All results obtained for these parameters met the desired values, thus pointing towards a good proteolytic activity in the malt. Taken together, these results indicate that the addition of E. gerundensis as a biocontrol agent at the beginning of the steeping step did not result in any degradation of the technological quality of the malt. Hence, the barley-associated strain E. gerundensis can be considered as a potential biocontrol agent for barley protection during malting.

Conclusion

The issues related to pathogenic fungi and their mycotoxin production are worsening due to the highly variable climatic conditions that favor fungal development and to the societal desire to reduce the use of phytosanitary products such as chemical fungicides. In this context, the use of technological microorganisms as a biological alternative is gaining interest among the scientific community as it is a potentially interesting and natural solution for protection against these fungi and their mycotoxins. This study investigated the use of a barley-associated Erwinia gerundensis strain as a biocontrol agent for the protection of malt against fungal development and mycotoxin production during the malting process. The initial results of our study could be very useful for the malting and brewing industries, as they demonstrate that the treatment of naturally contaminated barley with Erwinia gerundensis, at the beginning of the steeping step of malting, reduces the fungal load and mycotoxin content of the final malt without causing any degradation in its technological quality. Further investigations should be focused on improving the bioprotective capability of the proposed biocontrol agent.

Acknowledgment

The authors would like to thank Noémie Wiatrak, Elodie Hoch, Annelore Elfassy, Cécile Mangavel, Marcia Leyva-Salas, Sophie Slezack, Séverine Piutti, and Julie Genestier for their valuable support.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research work was supported by the French Ministry of Agriculture and Food, within the framework of the Joint Technological Unit OPTIMALT (Nancy, France), and funded by the Fondation de la Brasserie et de la Malterie Françaises (France), the Région Grand-Est (France), and the University of Lorraine (France) as part of a PhD fellowship.