Effects of bacterial inoculants on microbial community, mycotoxin contamination, and aerobic stability of corn silage infected in field by toxigenic fungi during aerobic exposure

DOI: https://doi.org/10.21203/rs.3.rs-2168901/v1

Abstract

This study was aimed to evaluate the effects of inoculants on the microbial community and mycotoxins contamination during aerobic exposure of corn silage. Whole-crop corn infected with or without mycotoxigenic fungi were ensiled with Lactobacillus buchneri (LB, 1.0×106 cfu g− 1 fresh weight (FW)), Lactobacillus plantarum (LP, 1.0×106 cfu g− 1 FW), or LBLP at 1.0× 106 cfu g− 1 FW each. The higher concentration of AcA (P < 0.05) in LB and LBLP silages than C and LP of NFI silages. Fungal infection resulted in a larger increase of zearalenone (ZEN, P = 0.01), fumonisin B1 (FUB1, P = 0.02), and fumonisin B2 (FUB2, P = 0.02). The RA of Issatchenkia in NFI was higher (P < 0.001) than FI silages, whereas the RA of Kazachstania (P < 0.001), Zygosaccharomyces (P = 0.047), and Candida (P = 0.025) in NFI were lower than these of FI silages. The aerobic stability was improved by the application of LB and LBLP as compared with C of NFI silages. The LB and LBLP had the potential to improve aerobic stability and alleviate mycotoxins contamination of non-fungal infected corn silages, but did not mitigate the negative effect of fungal infection in corn silages.

Introduction

Whole-crop corn silage is currently the predominant conserved forage in dairy systems worldwide, however, it is prone to aerobic spoilage during the feed-out phase [1]. Molds, together with yeasts, are believed to responsible for the aerobic spoilage of silages. Although most of yeast and molds were inhibited during the ensiling, the dominant fermented organic acids showed more fungistatic rather than fungicidal effect [2]. Thus, some acid-tolerant fungi can revive and aggressively proliferate when silage was exposed to air, resulting in the inferior aerobic stability and mycotoxin contamination. Mycotoxin contamination of food and feed is an ongoing global concern because it is currently regarded as a production constraint and a major animal health risk [3]. Fungal infection and mycotoxins contamination and mycotoxins production are impossible to entirely avoid during growing and storing of crops for cattle feed [4]. The oxygen-rich microenvironment was usually observed on the surface and edge of silo in which caked and clumpy areas develop visible green-gray mold indicative of mycotoxin production, including aflatoxins and several Fusarium mycotoxins [5, 6].

Inoculating L. buchneri is a common practice to improve bio-preservation and aerobic stability of corn silage because it delays development of spoilage yeasts and molds [7]. Apart from improvement in fermentation profiles and aerobic stability, some LAB had potential benefits to reduce mycotoxins contamination in silages [8]. Ma, et al. [9] reported that inoculation of L. plantarum or L. buchneri decreased linearly the aflatoxin B1 (AFB1) concentration within 3 d of ensiling, however, it is not clear whether their antifungal and anti-mycotoxigenic property could last to the feed-out phase of silages. L. buchneri have been used widely to enhance the aerobic stability of silages because of its heterolactic nature [10]. Most of the studies aimed at assessing the effects of L. buchneri on aerobic stability of corn silages under normal condition, a few studies have been conducted on the suboptimal conditions (i.e., corn infected by mold in the field) [7].

We hypothesized that infected toxigenic fungi could revive and proliferate during aerobic exposure, while inoculant LAB could mitigate potential negative effect of fungal infection on aerobic stability and hygienic quality of corn silage. Therefore, the study was aimed to evaluate the effects of L. buchneri and L. plantarum on bacterial community and mycotoxins concentration during aerobic exposure of corn silage, which was artificially infected by toxigenic fungi in the field.

Materials And Methods

Crop and Ensiling

Two toxigenic fungi Aspergillus flavus and Fusarium graminearum were cultured on potato dextrose agar (PDA, Shanghai Bio-way Technology Co., Ltd) medium at 30°C under aerobic conditions to fully sporulate. Spores were harvested with sterile glass slides and diluted with 0.1% Tween-80 (V/V) fortified sterile distilled water, followed by filtering with sterile 4-layer gauze to obtain a spore suspension. Then spore suspension was adjusted to a final concentration of approximately 106 cfu·mL− 1. Corn was grown in the experimental field (total 10 plots) of Nanjing Agricultural University (32.04°N, 118.88°E, Nanjing, China). At the silking stage of corn, 5 plots were randomly selected, and the prepared fungal spore mixture (5 mL per plant) was sprayed on ears, husks, silk, and leaves with a vacuum hand sprayer for artificial infection. After two weeks, the same plots of corn were artificially infected with the same procedure again.

Fungal infection (FI) and non-fungal infection (NFI) corn at the 1/2 milk line stage were harvested and chopped to a theoretical length of 2–3 cm. The 5 plots was harvested separately to obtain 5 replicates, and the fresh corn of each plots was divided into 4 piles. Both FI and NFI corn was treated either without (C, untreated), or with L. plantarum (LP, applied at 1.0 × 106 cfu g− 1 fresh weight (FW)), L. buchneri (applied at LB, 1 × 106 cfu g− 1 FW), and combination of L. plantarum and L. buchneri (LBLP, L. plantarum and L. buchneri applied at 0.5 × 106 cfu g− 1 FW each). Then about 3.5 kg of treated corn (DM, 336 ± 6.4 g kg− 1 FW) was filled into a plastic silo (5 L), which was sealed and stored for 90 d at an ambient temperature (24 ± 2°C).

Aerobic Stability Test

After 90 d of ensiling, all silages of each silo were emptied and mixed thoroughly, followed by filling into a newly large plastic silo (capacity 15L) without compaction and uncovered, which was stored at ambient temperature (24 ± 2 ℃) for 6 d. The silage was sampled on d 0, 2, 4, and 6 d of aerobic exposure. The room and silage temperatures were measured half-hourly by a data logger. Aerobic stability was defined as the number of hours the silage remained stable before its temperature increased by 2°C above the ambient temperature [11].

Sample Preparation and Analyses

The silages were split into four subsamples. One subsample (20 g) was extracted with 60 g of distilled water at 4°C for 24 h, followed by filtering through four layers of medical gauze. The pH of silage filtrate was immediately determined using a pH meter (HANNA pH 211, Hanna Instruments Italia Srl, Padova, Italy). Subsequently, one aliquot of silage extract was centrifuged at 10,000 × g for 10 min at 4°C, and the supernatant was reserved for LA, AcA, and ethanol (EOL) analyses, which were quantified following the protocol of [12].

One subsample of silage was oven-dried at 65°C to constant weight for DM content measure. Dried samples were ground to pass through a 1-mm screen for water-soluble carbohydrate (WSC) and mycotoxins analyses. The WSC content was analyzed by colorimetry after reaction with anthrone reagent. The concentrations of aflatoxins (AFs), zearalenone (ZEN), deoxynivalenol (DON), and fumonisins were quantified using ultra-performance Liquid Chromatography-Tandem Mass Spectrometry system with a Sciex QTRAP® 5500 triple quadruple tandem mass spectrometer (UPLC-MS/MS) (Sciex, Foster City, CA, USA) [13].

Bacterial community analysis

One subsample (10 g) was shaken with 90 mL sodium chloride sterile saline (0.85%) at 150 rpm. One aliquot of solution was 10-fold serially dilution to count the microorganisms, and another aliquot of solution was filtered through four layers of medical gauze for microbial DNA extraction. The LAB was determined on deMan, Rogosa, and Sharp agar after 48 h of anaerobic incubation at 37°C. The number of yeasts and molds was counted on PDA medium after 48–72 h of aerobic incubation at 28°C The microbial data were obtained and transformed to log10 for presentation and statistical analysis.

Three replicates of silages sampled from d 0 and 6 of aerobic exposure were randomly chosen for bacterial and fungal community analyses. Microbial DNA from various solution was extracted with Fast DNA SPIN Kit for Soil (MP Biomedicals, Solon, OH, USA). The Universal primers 338F (ACTCCTACGGGAGGCAGCAG) and 806R (GGACTACHVGGGTWTCTAAT) were used to amplify bacterial 16S rRNA V3-V4 regions of the gene, while fungal ITS was amplified with primers ITS1F (5′-CTTGGTCATTTAGAGGAAGTAA-3′) and ITS2aR (5′-GCTGCGTTCTTCATCGATGC-3′). Then purified PCR amplicons were paired-end sequenced using the Illumina MiSeq PE300 platform (Illumina Inc., San Diego, CA, USA). Paired-end reads were merged and checked by FLASH (Version 1.2.11), and the primers were trimmed using QIIME (Version 1.7.0). Operation taxonomic units (OTUs) were done using open reference clustered with a 97% similarity cutoff using UPARSE (Version 7.1 http://drive5.com/uparse/). Then the chimeric sequences were identified and removed using UCHIME. Alpha-diversity estimates (numbers of observed OUT, Chao1, and Shannon) and beta-diversity evaluation, based on principal coordinate analysis (PCoA), were performed using the Phyloseq and Vegan packages on R. Bacterial and fungal communities at the phylum and genus levels were analyzed based on Silva database with a confidence threshold of 70%. The high-throughput sequencing data were analyzed on the free online platform of Majorbio I-Sanger Cloud Platform (www.i-sanger.com). All DNA sequences have been deposited in the NCBI Short Read Archive database under BioProject PRJNA823478.

Statistical analyses

Data were analyzed using the GLM procedure of SAS (Version 9.3, SAS Institute, Cary, NC, USA). The fermentation profiles, microbial community, and mycotoxins concentrations were analyzed as split plot in a randomized complete block design with fixed effects of fungal infection (F), inoculation (I), aerobic duration (D), and their interaction. The treatments and aerobic duration were considered as main- and sub- plots, respectively. Data of aerobic stability were analyzed as a completely randomized design with fungal infection and inoculation as main factor. The Tukey’s test was employed to compare the differences among treatments, and significance was declared at P < 0.05.

Result

Fermentation profiles of corn silages during aerobic exposure

The dynamics of pH, LA and AcA concentrations during the aerobic exposure are shown in Fig. 1. There was a F × I × D interaction for pH (P = 0.002). The pH of FI silages increased faster and were higher than those of NFI silages over 6 d of aerobic exposure (4.41 vs. 4.11, on average, respectively, P = 0.008). For NFI silages, the pH in LBLP began to lower (P < 0.05) than other silages from d 4. For FI silages, the inoculants decreased pH value compared with C on d 4 (P < 0.05), however, their effects were disappeared on d 6 of aerobic exposure. There were F × I × D interactions for LA (P < 0.001) and AcA (P = 0.008) concentrations. The LA concentration of FI silages decreased faster than that of NFI silages during aerobic exposure. For FI silages, the LP and LBLP silages showed lower (P < 0.05) LA concentration than C and LB on d 4, however, their effects were disappeared on d 6 aerobic exposure. For NFI silages, the higher (P < 0.05) LA concentration in inoculants treated silage than C was only observed on d 6. FI silages had lower AcA concentrations (P < 0.05) than NFI silages (12.5 and 24.4 g kg− 1 DM, respectively) over 6 d of aerobic exposure. For FI silages, the higher (P < 0.05) concentration of AcA in LB and LBLP silages than C and LP lasted 2 and 4 d, respectively.

Changes in WSC and EOL concentrations of corn silage over 6 d of aerobic exposure are shown in Fig. 2. The concentration of WSC decreased (P < 0.001) over 6 d of aerobic exposure regardless of fungal infection. FI silages showed lower (P < 0.001) WSC concentration than NFI silages (22.2 vs. 33.4 g kg− 1 DM). The WSC concentrations in C and LP were lower (P < 0.01) than those of LB and LBLP for NFI silages after 6 d of aerobic exposure. There was a F × I × D interaction for concentration of EOL (P < 0.001). For NFI silages, the EOL concentration gradually (P > 0.05) decreased and there was no significant difference (P > 0.05) among NFI silages over 6 d of aerobic exposure. For FI silages, there was a slight increase in EOL concentration with 2 d followed by a rapid drop, and LP and LBLP silages had lower EOL concentration than C and LB on d 4 of aerobic exposure (P < 0.05).

The changes in the number of LAB, as well as yeasts and molds over 6 d of aerobic exposure are shown in Fig. 3. A F ×D interaction (P < 0.001) was detected for LAB counts. There was a slight increase in LAB counts regardless of fungal infection over 2 d of aerobic exposure, followed by a gradual decrease for FI silages, while LAB counts in NFI silages rapidly decreased after 4 d of aerobic exposure. Furthermore, an I × D interaction (P = 0.01) was also observed for LAB counts. LBLP of NFI silages had the lowest LAB counts after 6 d of aerobic exposure, while there was no significant difference in LAB counts among FI silages over during aerobic exposure. An F × D interaction was observed for yeast and mold (P < 0.001) counts. The count of yeasts and molds in FI silages increased faster and remained higher levels than that of NFI silages over 6 d of aerobic exposure (4.52 vs. 3.66 log10 cfu g− 1 FW, on average, respectively, P = 0.008). A two-way interaction of I × D was also observed for counts of yeasts and molds (P = 0.01). The inoculants decreased (P < 0.05) the counts of yeasts and molds compared with C of FI silages after 6 d of aerobic exposure, however, there was no significant difference (P > 0.05) in yeasts and molds counts among NFI silages during the entire aerobic exposure.

Bacterial and fungal communities of corn silages during aerobic exposure

A total of 2,294,390 and 2,948,299 reads were obtained by high-throughput amplicon sequencing of bacterial 16S rRNA and fungal ITS genes, respectively. The Good’s coverage for each sample was greater than 99% and rarefaction curves approached the plateau phase for all samples, revealing that most of the bacteria and fungi were detected and the sequencing depth was adequate for reliable analysis of microbial community. The Chao1and Shannon indices of bacterial and fungal communities decreased after 6 d of aerobic exposure, and inoculants reduced the Chao1and Shannon indices in NFI silages but did not affect those of FI silages (Fig. S1 and S2). The PCoA plot based on Unweighted-UniFrac distance of the bacterial (A, R = 0.5640, P = 0.001) and fungal (B, R = 0.5376, P = 0.001) communities are shown in Fig. S3, there was no same centroid among treatments, and aerobic duration had greater effects on the separations of bacterial and fungal communities than fungal infection and inoculants.

Changes in the genus-level bacterial community during aerobic exposure are shown in Fig. 4A. Acetobacter (57.1%), Lactobacillus (27.4%), Acinetobacter (2.95%), Klebsiella (1.93%), and Enterobacter (1.31%) were the dominant bacterial genus in corn silages. The relative abundance (RA) of Acetobacter, Lactobacillus, Klebsiella, and Enterobacter were neither affected (P > 0.05) by fungal infection nor inoculation. During 6 d of aerobic exposure, the RA of Acetobacter in silage increased (P < 0.001), while the RA of Lactobacillus (P < 0.001), Klebsiella (P = 0.003), and Enterobacter (P < 0.001) significantly decreased. There was an F × I interaction (P = 0.007) for the RA of Acinetobacter because it was the greatest (P < 0.05) in C among NFI silages, but it was the lowest in C among FI silages.

The changes of fungal genera with RA above 1% in the corn silages during 6 d of aerobic exposure are shown in Fig. 4B. The fungal genera including Issatchenkia (52.3%), Kazachstania (18.1%), Zygosaccharomyces (9.78%), Candida (7.14%), Aspergillus (2.23%), and Pichia (1.96%) were the dominant fungal genus in corn silages during aerobic exposure. The RA of Issatchenkia was higher (P < 0.001) in NFI than FI silages, whereas the RA of Kazachstania (P < 0.001), Zygosaccharomyces (P = 0.047), Candida (P = 0.025), and Pichia (P = 0.013) in NFI were lower than FI silages. there were F×D interaction for the RA of Issatchenkia (P = 0.013), Candida (P < 0.001), and Pichia (P < 0.001) because they significantly (P < 0.01) decreased for FI silages, but there were no marked changes for NFI silages during aerobic exposure. The F×I interaction were also found for RA of Kazachstania (P < 0.001) and Zygosaccharomyces (P < 0.001). The RA of Kazachstania and Zygosaccharomyces in FI silages significantly (P < 0.01) increased during aerobic exposure, but there were no significant changes except C for NFI silages.

Correlation relationship between microbial community and fermentation profiles

For NFI silages, redundancy analysis (RDA) indicated that the microbial community of silage was greatly affected by pH (Fig. 5A), followed by LA, EOL, WSC, and AcA. Lactobacillus was positively correlated with WSC, LA, and AcA, while negatively correlated with pH in NFI silages. Positive correlations were found between Issatchenkia and WSC, LA, and AcA, while negative correlation between Issatchenkia and pH was observed. For FI silages (Fig. 5B), the WSC, EOL, LA, and AcA concentrations were positively correlated with Lactobacillus, Issatchenkia, and Candida, while negatively correlated with Kazachstania, Zygosaccharomyces, and Acetobacter. The pH was positively correlated with Kazachstania, Zygosaccharomyces, and Acetobacter for FI silages.

Mycotoxin concentrations of corn silages

The mycotoxins concentration of silages showed high standard error, however, all data were normal distributed. The change of mycotoxins in corn silages during aerobic exposure are shown in Fig. 6. There were F×I interactions b for concentrations of AFB1 (P = 0.03) and AFB2 ( P = 0.02), because the increase of AFB1 in C was larger (P < 0.05) than other treatments for NFI silages but there was no significant difference in AFB1 concentration among FI silages. The change of AFB2 was similar among all silages for NFI silage, while the decline of AFB2 was only (P < 0.05) found in LB of FI silages. The FI silages showed larger increase of ZEN (P = 0.01), FUB1 (P = 0.02), and FUB2 (P = 0.02) than NFI silages. The increase of ZEN in LP was the smallest (P < 0.05) among all FI silages. The concentrations of FUB1 (P = 0.15) and FUB2 (P = 0.89) were not affected by inoculation. Neither inoculation nor fungal infection affected the concentration of DON.

Aerobic stability of corn silages

There was an F×I interaction (P < 0.001) for the aerobic stability of corn silages (Fig. 7), because LB (82 h) and LBLP (85 h) showed longer aerobic stability than C (62 h) for NFI silages, whereas there was no significant difference (P > 0.05) in aerobic stability among FI silages. The fungal infection enhanced the aerobic deterioration compared with NFI silages (72.7vs. 17.7, on average for NFI and FI, respectively, P < 0.001).

Discussion

Effect of inoculant and fungal infection on the fermentation profiles and mycotoxins of corn silages during aerobic exposure

Once the silage is opened and fed, air freely accesses the silo face and aerobic microorganisms that survived the ensiling process, e.g. bacilli, yeast and acid-tolerant bacteria, can rapidly proliferate, metabolizing residual sugars and organic acids to CO2, H2O, and heat [14, 15]. Consequently, silage temperature increases and the silage mass becomes aerobically unstable. The consumption of acids by aerobic microorganisms is accompanied with the increase of silages pH, thus, the variation of silage pH is used as the criteria of aerobic deterioration [16]. In the study, the pH of NFI silages showed a slight increase within 2 d followed by a marked increase, while that in FI silages rapidly increased during the initial 2 d of aerobic exposure, this might be attributed to the rapid revival of aerobic microorganisms including infected toxigenic fungi [16]. This hypothesis was confirmed by the changes of LA: the LA in FI sharply decreased once aerobic exposure, while the rapid drop of LA was following a transient stable (2 d) in NFI silages. Kung, et al. [17] also reported that the increase of silage pH was accompanied by a decline of LA in corn silages during aerobic exposure. The faster and larger changes of pH and LA in FI than NFI silages were related to its lower AcA concentration. The AcA has been proven to be responsible for enhancing aerobic stability because it acts as an inhibitor of spoilage organisms. Danner, et al. [18] stated that exponentially increased silage stability with AcA concentration were attributed that the lipophilic and undissociated form (around pH 4) of AcA, which could penetrate the bacterial plasma membrane and disorder metabolism in cell. Queiroz, et al. [19] also found that fungal infection worsens the fermentation of corn silage, resulted the lower AcA production during ensiling, which might be partly contributed the faster and larger changes of pH and LA in FI than NFI silages in the present study.

Although other modes of action may exist, the production of AcA has been the most accepted explanation of how organisms from the Lb. buchneri group of bacteria increases the aerobic stability of silages [7]. In the study, the higher AcA concentrations in LB and LBLP than other treatments of NFI silages lasted to d 2 and 4, respectively, while no significant difference in AcA concentrations was observed among FI silages. This indicated that the fungal infection might disturb the microbial community and fermentation of silages, diminishing the effect of LB on the AcA production. The decrease of WSC decreased with prolonged aerobic exposure was attributed to the extensive proliferation of aerobic microorganisms, which can oxidize WSC to CO2 and H2O [20]. FI silages always had lower WSC concentrations than NFI silages. It can be conjectured that infected fungi consumed more WSC during the initial transient aerobic stage of ensiling, resulting in the lower residual WSC being in FI silages than NFI silages. For NFI silages, the concentration of WSC in LB and LBLP was consistently higher than that of C and LP over the 6 d of aerobic exposure. This was attributed to the more AcA produced by L. buchneri, which inhibited the proliferation of undesirable microorganisms, preserving more residual WSC [21]. Kung, et al. [22] reported that higher dissociation constants (pKa) of AcA (4.75) than LA (3.86) contributed to the higher antimicrobial activity of AcA in surroundings where the pH values are low (around pH 4), since a greater proportion of the acetate is undissociated status, which could pass through the bacterial or fungal cell membrane and release protons to acidify the cytoplasm, thereby inhibiting or killing microorganisms [23, 24]. In the study, there was a transient increase in the ethanol concentration of FI silages, which was possibly due to the metabolism of yeast within the initial 2 d of aerobic exposure [25]. FI silages had higher number of yeasts and molds than NFI silages. Irrespective of fungal infection, the lower numbers of yeast and molds in LB and LBLP than C and LP on d 2 and 4 were attributed to the fungistatic property of AcA produced by the inoculants of L. buchneri [2]. With the progress of aerobic exposure, the gradual decline of ethanol concentration might be due to its volatilization or metabolized by Acetobacter [26].

Effect of inoculant and fungal infection on the bacterial and fungal communities of corn silages

During aerobic exposure, application of LAB reduced the Shannon and Chao 1 indexes of bacterial and fungal communities in NFI silages, but not FI silages. This might be due to the proliferation of aerobic bacteria and yeasts, which occupied the predominant roles of the microbial community [20].

In the study, Acetobacter became the dominant bacteria after 6 d of aerobic exposure, because it can oxidize ethanol to AcA initially, followed by oxidation of LA and AcA to CO2 and H2O [27]. Acetobacter is non-fermenting aerobic bacteria and can be found in various environments, it is able to initiate aerobic deterioration of corn silage with or without the presence of yeasts [28]. Guan, et al. [29] also observed that the significant increase in the RA of Acetobacter in Napier grass after 2 d of aerobic exposure. Lactobacillus is the main bacteria involved in LA fermentation during ensiling and usually dominates the well-fermented silages [30], however, it is rapidly replaced by aerobic bacteria (e.g. Acetobacter spp.) once silage is exposure to air [31]. In the present study, the RA of Lactobacillus reduced below 1% in silage except C and LB of NFI silages after 6 d of aerobic exposure. The genus Klebsiella and Enterobacter, belonging to enterobacteria, are capable to produce ammonia, which can cause animal health issues [32]. In the study, the decline of RA of Klebsiella and Enterobacter might be attributed to the anaerobic property and intolerance to acid conditions, as reported by McGarvey, et al. [33], who found Enterobacteriaceae could thrive in anaerobic and weak acidic conditions (pH > 5.4). In the study, the silage pH was not increased up 5.4 until d 6 of aerobic exposure, which resulted in the substitution of Klebsiella and Enterobacter by Acetobacter. The higher RA of Acinetobacter in NFI than FI silages might be related to the higher AcA concentration, which could be used by Acinetobacter as substrate [34].

Issatchenkia, an acid-tolerant yeast, is the dominant fungal genus in NFI silages regardless of aerobic exposure or inoculation. However, there are marked variations in fungal composition for FI silages during 6 d of aerobic exposure. This discrepancy between NFI and FI silages was attributed to the fungal infection, which resulted in more yeast and molds present and revived in FI silages during aerobic exposure. For FI silages, the marked decline in the RA of Issatchenkia was accompanied by the increase of RA of Kazachstania and Zygosaccharomyces during 6 d of aerobic exposure. Wang, et al. [35] indicated that Kazachstania had a strong tolerance to LA and was crucially involved in initiating the aerobic deterioration of corn silage with a relatively low pH and AcA content. Hao, et al. [36] reported that Zygosaccharomyces bailii was the sole yeast species isolated from spoilage total mixed ration (TMR) silages and confirmed that the Z. bailii could initiate aerobic deterioration of TMR silages. In the study, the RA of Candida and Pichia decreased below 1% after 6 d of aerobic exposure, this is contrast to the report by Duniere, et al. [37], who found that Candida and Pichia were the main spoilage genera after aerobic exposure. Pahlow, et al. [38] indicated that Issatchenkia, Candida, and Pichia are lactate-assimilating yeasts that the initiators of aerobic degradation of silage. We speculated that Kazachstania and Zygosaccharomyces underwent more vigorous growth and outcompeted other yeast and molds.

The correlation between microbial community and fermentation profiles was analyzed by RDA. In the present study, bacterial genera such as Lactobacillus were found to be positively correlated with LA and AcA, while Acetobacter was negatively correlated with LA and AcA regardless of fungal infection. When silage is exposed to air, LA and AA were metabolized by aerobic bacteria, increasing silage pH, which further inhibited the proliferation of Lactobacillus and boosted the dominance of Acetobacter [39]. Fungal genera Kazachstania and Zygosaccharomyces were negatively correlated with LA and were positively correlated with pH. This indicated that Kazachstania and Zygosaccharomyces were considered as dominant fungi during aerobic exposure of silages, which could assimilate lactate and increase pH, initiating the spoilage of silages.

Effect of inoculant and fungal infection on mycotoxins of corn silages during aerobic exposure

In the study, the concentrations of most mycotoxins increased over 6 d of aerobic exposure, indicating that toxigenic fungi revived during the aerobic exposure [40]. The larger increases of AFs, ZEN, and DON concentrations in FI than NFI silages confirmed that artificially infected A. flavus and F. graminearum might revive and produce mycotoxins during the aerobic exposure stage. Ferrero, et al. [41] indicated that the aerobic environment allowed proliferation of A. flavus and enhanced the production of AFB1. Vandicke, et al. [42] also reported that some Fusarium species were even able to survive at low oxygen levels (< 0.5%), such as silage conditions, and the inactive fungal spores in the silage may be reactivated and produce Fusarium toxins due to exposure to oxygen during feed-out.

The more effective of inoculants on reducing AFB1 contamination in NFI than FI silages might be attributed to the revival of toxigenic fungi and higher mycotoxins in FI silage, which attenuated the effects of inoculants on reducing AFB1 contamination of FI silages. Ma, et al. [9] reported that L. plantarum, L. buchneri, and Pediococcus acidilactici could bind to AFB1 in an in vitro medium. Dogi, et al. [43] suggested that inoculation with Lactobacillus rhamnosus strongly inhibited the fungal growth (F. graminearum, Aspergillus parasiticus, etc.) and mycotoxin production (AFs, ZEN, DON, etc.). However, it is unexplainable that the AFB2 in LB of FI silage was decreased during aerobic exposure in the study.

Effect of inoculants and fungal infection on aerobic stability of corn silages

The fungal infection shortened aerobic stability of silages regardless of inoculation. This was attributed to the artificial fungal infection, which resulted in more yeast and molds survived in FI than NFI silages. Keshri, et al. [44] reported that the lower aerobic stability of corn silages was due to the high number of lactate-assimilating yeasts. In the study, the aerobic stability of NFI silages was improved by the application of LB and LBLP as compared with C of NFI silages. Kung, et al. [17] also found that inoculant containing L. buchneri could extend the aerobic stability of silages challenged with air stress, which was attributed to the antifungal property of AcA produced by L. buchneri. Romero, et al. [45] reported that applying a combination inoculant of L. buchneri and Pediococcus pentosaceus increased the AcA concentration and decreased yeasts and molds in corn silages, which in turn indirectly extended the aerobic stability.

Conclusion

The fungal infection disturbs the microbial community and silage fermentation, weakening the effect of LB on AcA production. Kazachstania and Zygosaccharomyces were dominant fungi and contributed to the aerobic spoilage of FI silages. There were larger increases of AFs, ZEN, and DON concentrations in FI than NFI, and the inoculants showed more effect on reducing AFB1 contamination in NFI than FI silages. The application of LB and LBLP inoculants improved aerobic stability and delayed the onset of aerobic deterioration, which in turn indirectly reduced the risk of mycotoxins production after silo opening, but did not mitigate the negative effect of fungal infection in corn silages.

Declarations

Acknowledgments

This work was partially supported by National Natural Science Foundation of China (31872421) and National Key Research and Development Program of China (2017YFE0104300).

Author contributions

Wenbo Wang, Wenkang Wang, and Pengfei Ma: performed the experiment, analysis, and writing. Junfeng Li, Jie Zhao, and Antonio Gallo: performed the editing and revision. Xianjun Yuan and Tao Shao: designed the experiment. All authors have read and agreed to the published version of the manuscript.

Funding

National Natural Science Foundation of China (31872421) and National Key Research and Development Program of China (2017YFE0104300).

Availability of data and materials

The datasets used or analyzed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

Not applicable.

Consent for publication

All authors listed have read the complete manuscript and have approved submission of the paper.

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

References

  1. Benjamim da Silva E, Costa DM, Santos EM, Moyer K, Hellings E and Kung L, Jr., The effects of Lactobacillus hilgardii 4785 and Lactobacillus buchneri 40788 on the microbiome, fermentation, and aerobic stability of corn silage ensiled for various times. J Dairy Sci. 2021;104:10678–10698.
  2. Kung L, Jr., Smith ML, Benjamim da Silva E, Windle MC, da Silva TC and Polukis SA, An evaluation of the effectiveness of a chemical additive based on sodium benzoate, potassium sorbate, and sodium nitrite on the fermentation and aerobic stability of corn silage. J Dairy Sci. 2018;101:5949–5960.
  3. García-Díaz M, Gil-Serna J, Vázquez C, Botia MN and Patiño B, A Comprehensive Study on the Occurrence of Mycotoxins and Their Producing Fungi during the Maize Production Cycle in Spain. Microorganisms. 2020;8:141–155.
  4. Smith LE, Stoltzfus RJ and Prendergast A, Food chain mycotoxin exposure, gut health, and impaired growth: a conceptual framework. Adv Nutr. 2012;3:526–531.
  5. Cheli F, Campagnoli A and Dell’Orto V, Fungal populations and mycotoxins in silages: From occurrence to analysis. Anim Feed Sci Tech. 2013;183:1–16.
  6. Gallo A, Fancello F, Ghilardelli F, Zara S, Froldi F and Spanghero M, Effects of several lactic acid bacteria inoculants on fermentation and mycotoxins in corn silage. Anim Feed Sci Tech. 2021;277.
  7. Muck RE, Nadeau EMG, McAllister TA, Contreras-Govea FE, Santos MC and Kung L, Jr., Silage review: Recent advances and future uses of silage additives. J Dairy Sci. 2018;101:3980–4000.
  8. Gallo A, Fancello F, Ghilardelli F, Zara S and Spanghero M, Effects of several commercial or pure lactic acid bacteria inoculants on fermentation and mycotoxin levels in high-moisture corn silage. Anim Feed Sci Tech. 2022;286.
  9. Ma ZX, Amaro FX, Romero JJ, Pereira OG, Jeong KC and Adesogan AT, The capacity of silage inoculant bacteria to bind aflatoxin B1 in vitro and in artificially contaminated corn silage. J Dairy Sci. 2017;100:7198–7210.
  10. Ferrero F, Tabacco E, Piano S, Casale M and Borreani G, Temperature during conservation in laboratory silos affects fermentation profile and aerobic stability of corn silage treated with Lactobacillus buchneri, Lactobacillus hilgardii, and their combination. J Dairy Sci. 2021;104:1696–1713.
  11. Ranjit NK and Kung L, Jr., The effect of Lactobacillus buchneri, Lactobacillus plantarum, or a chemical preservative on the fermentation and aerobic stability of corn silage. J Dairy Sci. 2000;83:526–535.
  12. Yuan XJ, Yang X, Wang WB, Li JF, Dong ZH, Zhao J, et al., The effects of natamycin and hexanoic acid on the bacterial community, mycotoxins concentrations, fermentation profiles, and aerobic stability of high moisture whole-crop corn silage. Anim Feed Sci Tech. 2022;286.
  13. Wang W, Yang X, Li J, Dong Z, Zhao J, Shao T, et al., Effects of hexanoic acid on microbial communities, fermentation, and hygienic quality of corn silages infested with toxigenic fungi. J Sci Food Agric. 2022;102:3522–3534.
  14. Drouin P, Tremblay J, Renaud J and Apper E, Microbiota succession during aerobic stability of maize silage inoculated with Lentilactobacillus buchneri NCIMB 40788 and Lentilactobacillus hilgardii CNCM-I-4785. Microbiologyopen 10:e1153 (2021).
  15. Shan G, Buescher W, Maack C, Lipski A, Acir IH, Trimborn M, et al., Dual sensor measurement shows that temperature outperforms pH as an early sign of aerobic deterioration in maize silage. Sci Rep 11:8686 (2021).
  16. Wilkinson JM and Davies DR, The aerobic stability of silage: key findings and recent developments. Grass Forage Sci. 2013;68:1–19.
  17. Kung L, Jr., Savage RM, da Silva EB, Polukis SA, Smith ML, Johnson ACB, et al., The effects of air stress during storage and low packing density on the fermentation and aerobic stability of corn silage inoculated with Lactobacillus buchneri 40788. J Dairy Sci. 2021;104:4206–4222.
  18. Danner H, Holzer M, Mayrhuber E and Braun R, Acetic acid increases stability of silage under aerobic conditions. Appl Environ Microbiol. 2003;69:562–567.
  19. Queiroz OCM, Kim SC and Adesogan AT, Effect of treatment with a mixture of bacteria and fibrolytic enzymes on the quality and safety of corn silage infested with different levels of rust. J Dairy Sci. 2012;95:5285–5291.
  20. Yuan XJ, Wen AY, Wang J, Desta ST, Dong ZH and Shao T, Effects of four short-chain fatty acids or salts on fermentation characteristics and aerobic stability of alfalfa (Medicago sativa L.) silage. J Sci Food Agric. 2018;98:328–335.
  21. Romero JJ, Joo Y, Park J, Tiezzi F, Gutierrez-Rodriguez E and Castillo MS, Bacterial and fungal communities, fermentation, and aerobic stability of conventional hybrids and brown midrib hybrids ensiled at low moisture with or without a homo- and heterofermentative inoculant. J Dairy Sci. 2018;101:3057–3076.
  22. Kung L, Jr., Shaver RD, Grant RJ and Schmidt RJ, Silage review: Interpretation of chemical, microbial, and organoleptic components of silages. J Dairy Sci. 2018;101:4020–4033.
  23. Gheller LS, Ghizzi LG, Takiya CS, Grigoletto NTS, Silva TBP, Marques JA, et al., Different organic acid preparations on fermentation and microbiological profile, chemical composition, and aerobic stability of whole-plant corn silage. Anim Feed Sci Tech. 2021;281.
  24. Kung L, Stokes MR and Lin CJ, Silage Additives, in Silage Science and Technology. 2003;305–360.
  25. Koc F, Unal EO, Okuyucu B, Esen S and Isik R, Effect of Different Kefir Source on Fermentation, Aerobic Stability, and Microbial Community of Alfalfa Silage. Animals. 2021;11.
  26. Carvalho BF, Ávila CLS, Miguel MGCP, Pinto JC, Santos MC and Schwan RF, Aerobic stability of sugar-cane silage inoculated with tropical strains of lactic acid bacteria. Grass Forage Sci. 2015;70:308–323.
  27. da Silva EB, Savage RM, Biddle AS, Polukis SA, Smith ML and Kung L, Effects of a chemical additive on the fermentation, microbial communities, and aerobic stability of corn silage with or without air stress during storage. J Anim Sci. 2020;98:1–11.
  28. Spoelstra SF, Courtin MG and Vanbeers JAC, Acetic-Acid Bacteria Can Initiate Aerobic Deterioration of Whole Crop Maize Silage. J Agr Sci. 1988;111:127–132.
  29. Guan H, Shuai Y, Ran Q, Yan Y, Wang X, Li D, et al., The microbiome and metabolome of Napier grass silages prepared with screened lactic acid bacteria during ensiling and aerobic exposure. Anim Feed Sci Tech 269 (2020).
  30. McAllister TA, Duniere L, Drouin P, Xu S, Wang Y, Munns K, et al., Silage review: Using molecular approaches to define the microbial ecology of silage. J Dairy Sci. 2018;101:4060–4074.
  31. Guan H, Ran Q, Li H and Zhang X, Succession of Microbial Communities of Corn Silage Inoculated with Heterofermentative Lactic Acid Bacteria from Ensiling to Aerobic Exposure. Fermentation. 2021;7:258.
  32. Driehuis F, Wilkinson JM, Jiang Y, Ogunade I and Adesogan AT, Silage review: Animal and human health risks from silage. J Dairy Sci. 2018;101:4093–4110.
  33. McGarvey JA, Franco RB, Palumbo JD, Hnasko R, Stanker L and Mitloehner FM, Bacterial population dynamics during the ensiling of Medicago sativa (alfalfa) and subsequent exposure to air. J Appl Microbiol. 2013;l114:1661–1670.
  34. Poduch E and Kotra LP, Acinetobacter Infections, in xPharm: The Comprehensive Pharmacology Reference, ed. by Enna SJ and Bylund DB. 2007;1–9.
  35. Wang C, Sun L, Xu H, Na N, Yin G, Liu S, et al., Microbial Communities, Metabolites, Fermentation Quality and Aerobic Stability of Whole-Plant Corn Silage Collected from Family Farms in Desert Steppe of North China. Processes. 2021;9:784.
  36. Hao W, Wang HL, Ning TT, Yang FY and Xu CC, Aerobic Stability and Effects of Yeasts during Deterioration of Non-fermented and Fermented Total Mixed Ration with Different Moisture Levels. Asian-Australas J Anim Sci. 2015;28:816–826.
  37. Duniere L, Xu S, Long J, Elekwachi C, Wang Y, Turkington K, et al., Bacterial and fungal core microbiomes associated with small grain silages during ensiling and aerobic spoilage. BMC Microbiol 17:50 (2017).
  38. Pahlow G, Muck RE, Driehuis F, Elferink SJWHO and Spoelstra SF, Microbiology of Ensiling, in Silage Science and Technology. 2015;31–93.
  39. Liu B, Huan H, Gu H, Xu N, Shen Q and Ding C, Dynamics of a microbial community during ensiling and upon aerobic exposure in lactic acid bacteria inoculation-treated and untreated barley silages. Bioresour Technol. 2019;273:212–219.
  40. Bernardes TF, De Oliveira IL, Casagrande DR, Ferrero F, Tabacco E and Borreani G, Feed-out rate used as a tool to manage the aerobic deterioration of corn silages in tropical and temperate climates. J Dairy Sci 104:10828–10840 (2021).
  41. Ferrero F, Prencipe S, Spadaro D, Gullino ML, Cavallarin L, Piano S, et al., Increase in aflatoxins due to Aspergillus section Flavi multiplication during the aerobic deterioration of corn silage treated with different bacteria inocula. J Dairy Sci. 2019;102:1176–1193.
  42. Vandicke J, De Visschere K, Ameye M, Croubels S, De Saeger S, Audenaert K, et al., Multi-Mycotoxin Contamination of Maize Silages in Flanders, Belgium: Monitoring Mycotoxin Levels from Seed to Feed. Toxins. 2021;13:202–223.
  43. Dogi CA, Fochesato A, Armando R, Pribull B, de Souza MM, da Silva Coelho I, et al., Selection of lactic acid bacteria to promote an efficient silage fermentation capable of inhibiting the activity of Aspergillus parasiticus and Fusarium gramineraum and mycotoxin production. J Appl Microbiol. 2013;114:1650–1660.
  44. Keshri J, Chen Y, Pinto R, Kroupitski Y, Weinberg ZG and Sela Saldinger S, Microbiome dynamics during ensiling of corn with and without Lactobacillus plantarum inoculant. Appl Microbiol Biotechnol. 2018;102:4025–4037.
  45. Romero JJ, Park J, Joo Y, Zhao Y, Killerby M, Reyes DC, et al., A combination of Lactobacillus buchneri and Pediococcus pentosaceus extended the aerobic stability of conventional and brown midrib mutants-corn hybrids ensiled at low dry matter concentrations by causing a major shift in their bacterial and fungal community. J Anim Sci. 2021;99:1–16.

Tables

Table 1
Probability of the effects of fungal infection, inoculants, days of aerobic exposure, and their interactions on chemical composition, fermentation products, and microbial populations of corn silage (n = 5)
Item
Main effect1
 
Interaction effect2
F
I
D
 
F×I
F×D
I×D
F×I×D
pH
< 0.001
< 0.001
0.002
 
0.727
< 0.001
0.126
0.002
Lactic acid
< 0.001
< 0.001
< 0.001
 
< 0.001
< 0.001
< 0.001
< 0.001
Acetic acid
< 0.001
< 0.001
< 0.001
 
0.463
< 0.001
0.002
0.008
Ethanol
0.009
< 0.001
< 0.001
 
0.003
< 0.001
< 0.001
< 0.001
WSC
< 0.001
< 0.001
< 0.001
 
0.073
0.460
0.408
0.289
Y&M
< 0.001
< 0.001
< 0.001
 
0.428
< 0.001
0.011
0.710
LAB
0.780
< 0.001
< 0.001
 
0.120
< 0.001
0.010
0.099
1 F, Fungal infection; I, inoculants; D, days of aerobic exposure;
2 two-way and three-way interaction among main factors;
WSC, water-soluble carbohydrates; LAB, lactic acid bacteria; Y&M, yeasts and molds.