Early Fermentation Dynamics and Aerobic Stability of Maize Silage Improved by Dual-Strain Lactic Acid Bacteria Inoculation

Abstract

This study aimed to provide deeper insights into fermentation dynamics, aerobic stability, and bacterial community composition during the short-term ensiling of maize forage with lactic acid bacteria-based inoculants. A 50:50 combination of Lentilactobacillus buchneri DSM2250 and Lactococcus lactis DSM11037 (LBL target application: 150,000 CFU per 1 g forage) was tested alongside an untreated control (C) over fermentation periods of 2, 4, 8, 16, and 32 days. A total of 50 3 L mini-silos were filled with 2 kg of fresh maize each and stored at 20 °C. The pH, dry matter, nutrient profiles, volatile fatty acids, lactic acid, alcohols, ammonia-N, microbiological counts (yeast and mold), and aerobic stability of all samples were analyzed after seven days of air exposure. LBL silage showed higher average dry matter content (DMc) and crude protein (CP) levels by 1.5%, p < 0.001, and 10.8%, p < 0.001, respectively, as well as reduced average dry matter (DM) losses by half (p < 0.001) compared to pure silage. The beneficial effects of inoculation became more pronounced with prolonged storage, particularly by day 32 of fermentation. LBL silage showed increased production of lactic and acetic acids by an average of 55.5% and 5.0%, respectively, (p < 0.01) and significantly reduced butyric acid formation by approximately 14 times. Ethanol and ammonia-N concentrations were also reduced by 55.4% and 25.6%, respectively (p < 0.001), while the pH value remained 0.17 units lower (p < 0.001) compared to the control. The combination of the two strains improved silage aerobic stability by 2.4 days (p < 0.001) and extended shelf life by reducing yeast counts (8.02 vs. 7.35 log₁₀CFU g⁻¹ FM, p < 0.001), while maintaining the pH value close to its initial level. Therefore, compared to the untreated control, the inoculated silage exhibited higher nutritional value, reduced fermentation losses, and suppressed undesirable microbial activity. The positive effects of inoculation became increasingly evident over time, particularly by day 32, highlighting the synergistic benefits of using mixed-strain lactic acid bacteria. These findings support the use of LBL inoculants as an effective strategy to enhance short-term silage quality and stability.

1. Introduction

The quality of silage, apart from factors such as plant species and maturity stage, as well as the contents of dry matter (DM), water-soluble carbohydrates, and crude protein, is influenced by the composition of the microbial community, its dynamic succession, and the resulting fermentation metabolites [1]. Most lactic acid bacteria (LAB) strains are naturally present in spontaneously fermented silage; therefore, they have been isolated, cultivated, and used as single- or mixed-strain inoculants to ensure controlled silage fermentation [2,3,4]. LAB are extensively used as silage additives due to their ability to enhance the production of lactic and acetic acids, promote a rapid decline in pH, prevent the decomposition of organic matter, inhibit the activity of harmful microorganisms, and delay silage spoilage [5]. However, different LAB genera exhibit varying effects on fermentation patterns during the ensiling process [6,7]. A study by Ranjit and Kung [8] suggested that inoculating forage with heterolactic bacteria, such as Lentilactobacillus buchneri, can improve aerobic stability; however, it may also increase DM losses during fermentation. This was confirmed by Kleinschmit and Kung [9], who reported a 1 to 1.8% increase in DM losses when Lentilactobacillus buchneri was applied to corn, grass, and small grain forages. To simultaneously reduce DM losses and enhance fermentation, modern inoculants often include a combination of homolactic or facultative heterolactic bacteria with obligate heterolactic bacteria. Studies have shown that combining homolactic bacteria with Lentilactobacillus buchneri can improve aerobic stability without increasing DM losses [10]. The study by Broberg et al. [11] demonstrated that antifungal compounds produced by LAB strains can also be detected in inoculated silage. The antimicrobial activity of LAB is commonly explained by the synthesis of small organic acids such as lactic, acetic, and formic acids, which may exert their biological effect either directly or by acidification of the growth medium. Other substances produced by LAB have been reported to have antimicrobial effects, e.g., 2,3-butadione, reuterin (3-hydroxypropionaldehyde), acetaldehyde, hydrogen peroxide, hydroxyl radical, and peptides or proteins such as the bacteriocin [12]. Certain strains of Lactococcus lactis can secrete compounds that inhibit spoilage yeasts and molds. Lactococcus lactis is widely recognized for its production of bacteriocins, with the most notable being nisin [13]. Thus, it can be highlighted that heterofermentative LAB, such as Lentilactobacillus buchneri, produce lactic acid along with higher levels of acetic acid and 1.2 propanediol, enhancing the aerobic stability of silage and delaying its deterioration during storage [6,14,15,16]. Lactococcus lactis is a mesophilic species belonging to the family Streptococcaceae. The mechanism of action of Lactococcus lactis primarily centers on its ability to ferment sugars, especially lactose, into lactic acid. This process lowers pH, creating an environment that inhibits many pathogenic and spoilage organisms. However, it is not only passive fermentation. L. lactis produces bacteriocins like nisin, a potent antimicrobial peptide that disrupts the cell membranes of Gram-positive bacteria by forming pores, leading to ion leakage and cell death. This makes Lactococcus lactis a powerful natural preservative and a critical player in food and feed biotechnology [17]. Early fermentation dynamics refers to the initial changes in chemical and microbial activity during the early stages of silage fermentation. Researchers have studied how the addition of LAB strains influenced pH reduction, acid production (especially lactic and acetic acids), and the suppression of undesirable microbes during the initial days of ensiling [18]. The use of a bacterial additive containing two different strains of lactic acid bacteria—Lentilactobacillus buchneri and Lactococcus lactis—may offer complementary benefits. Lactococcus lactis contributes to rapid acidification and exhibits antibacterial effects, while Lentilactobacillus buchneri enhances aerobic stability by producing acetic acid and inhibiting aerobic deterioration. Therefore, gaining a deeper understanding of silage fermentation characteristics across different storage periods could offer valuable insights into fermentation dynamics. During the initial days of fermentation, the rate of acidification plays a crucial role in suppressing undesirable aerobic enterobacteria, yeasts, and certain fungi. This rapid acidification also promotes the production of lactic and acetic acids, thereby minimizing crude protein degradation into ammonia [19]. It is well documented that the Lactobacillus species play a substantial role in silage production due to their rapid growth and ability to lower pH through the production of lactic and acetic acids. Strains of the Lactobacillus genus are commonly used as silage additives to enhance the fermentation process and improve silage quality [6,20,21]. One of the heterofermentative LAB species—Lentilactobacillus buchneri—has been developed as a silage inoculant to prevent aerobic degradation. Under anaerobic conditions, Lentilactobacillus buchneri helps inhibit aerobic deterioration by converting lactic acid into 1,2-propanediol and acetic acid [22,23]. Recently, a novel Lactococcus lactis (DSM11037) LAB strain was introduced, which excels in oxygen scavenging and is also relatively fast in reducing pH [24]. The hypothesis assumes that the combination of fast-acidifying and oxygen-scavenging properties (from Lactococcus lactis) and aerobic stability-enhancing traits (from Lentilactobacillus buchneri) will produce superior results compared to uninoculated silage.

The central innovation lies in the use of a dual-strain LAB inoculant combining Lentilactobacillus buchneri DSM22501 and Lactococcus lactis DSM11037—two strains with distinct and complementary metabolic functions. This combination strategy addresses a key challenge in silage management, that is, ensuring rapid acidification while improving long-term aerobic stability. Additionally, our study focuses on short-term fermentation dynamics (2–32 days), a period that is often neglected but critical in modern silage practices where silos may be opened early. This dual-strain, time-resolved approach offers new insights into fermentation control and microbial succession and contributes original findings to the field.

While most research focuses on long-term silage fermentation, our study aimed to investigate the effects of a dual-strain LAB (Lentilactobacillus buchneri DSM2250 and Lactococcus lactis DSM11037) inoculant on fermentation characteristics, dry matter losses, aerobic stability, and bacterial community dynamics during short-term maize forage ensiling over 2, 4, 8, 16, and 32 days. Understanding these early stages is critical for optimizing fermentation and preventing spoilage, offering a comprehensive view of how bacterial populations evolve during early ensiling.

2. Materials and Methods

2.1. Crop Material, Microbial Inoculants, and Silage Preparation

The early maturing whole-plant maize hybrid, KWS SANDIAS (FAO number 180), developed by KWS SAAT SE & Co. KGaA, Lochow, Germany, was used as the crop material. The maize was harvested in early October (148 days after sowing) at the late dough stage of grain maturity (343.9 g DM kg⁻¹) using a Claas Jaguar 840 forage harvester from Harsewinkel, Germany, without a kernel processor and without internal inoculant application. The harvester was calibrated to achieve a theoretical chop length of approximately 10 mm for maize. After harvesting, the maize forage was promptly transported to the laboratory for silage preparation, with experiments beginning within two hours. A total of 5 samples of chopped whole crop maize were collected during forage harvesting to minimize the effect of sampling location on silage variables due to different forage composition. The composition of the forage prior to additive application and ensiling is given in

Table 1. Chemical composition and microbial characteristics of whole-plant maize before additive application (n = 5).

Item	Average	SD
DM, g kg−1	343.9	05.60
Crude protein, g kg−1 DM	67.5	06.75
Ether extract g kg−1 DM	24.2	01.2
Crude fibre, gkg−1 DM	205.9	12.35
NFE, g kg−1 DM	603.7	10.04
Crude ash, g kg−1 DM	98.63	01.71
WSC, g kg−1 DM	87.7	03.22
ADF, g kg−1 DM	250.3	13.53
NDF, g kg−1 DM	397.2	23.68
pH	6.04	0.018
Nitrate, mg kg−1 DM	447.76	15.525
Buffer capacity, mequiv100 g−1 DM	19.46	2.056
Yeasts, log10CFU g−1 FM	5.75	0.076
LAB, log10CFU g−1 FM	4.97	0.087
Moulds, log10CFU g−1 FM	5.52	0.049

DM–dry matter, NFE–nitrogen-free extracts, WSC–water-soluble carbohydrates, NDF–neutral detergent fiber, ADF–acid detergent fiber, LAB–lactic acid bacteria, CFU–colony forming units, SE–standard error.

.

Approximately 120 kg of forage was randomly collected, thoroughly mixed, and pre-ensiling samples were taken to establish the baseline values for nutritive content, DM, pH, and yeast and mold counts. Two forage piles, each weighing approximately 50 kg, were prepared and subjected to the following treatments:

• Control (C)—water application only;
• LBL—a 50:50 combination of Lentilactobacillus buchneri (DSM22501) and Lactococcus lactis (DSM11037), SiloSolve^® FC (Novonesis, Lyngby, Denmark) applied at rate 150,000 CFU g⁻¹ forage.

The microbial inoculant LBL was diluted with distilled water to achieve a target concentration (

Table 2. Actual counts of LAB (lactic acid bacteria) in water and suspension used for silage inoculation (CFU mL⁻¹).

Treatment	Expected Counts	Actual Counts	± %
C (water)	0	<1.0 × 10
LBL *	1.5 × 108	1.6 × 108	+6.7%

* LBL 50:50 combination of Lentilactobacillus buchneri DSM22501 and Lactococcus lactis DSM11037.

).

The test product (LBL) was diluted with unchlorinated water immediately before application to obtain a suspension with a concentration of 1.5 × 10⁸ CFU mL⁻¹. Samples of water and inoculant suspension were collected and immediately analyzed for LAB counting using the ISO method 15,214 (Table 2) [25]. The chopped maize forage was divided into two equal portions. One portion of the forage remained untreated, while the other portion was treated with an inoculant. Briefly, 1 mL of suspension and 3 mL of chlorine-free water (to improve spraying and mixing uniformity) was added to 1 kg of forage to obtain the final dosage of 1.5 × 10⁵ CFU g⁻¹ forage. The inoculant suspension and the water were sprayed on the forage using a spray bottle, with periodic mixing of the forage. The negative control (C) forage was sprayed with sterile unchlorinated water at 4 mL kg⁻¹ forage.

The forage was packed into laboratory-scale silos (3 L glass jars, at a density of approximately 200 kg DM/m³. The treated and weighed forage (using 2 kg of fresh forage per mini-silo) was placed into 3 L glass jars (20 cm in height and 14 cm in diameter) manually at a density of approximately 667 kg/m³ (229 kg DM/m³). During the filling process, periodic tamping was carefully performed to minimize air headspace within the jars. Each mini-silo was weighed empty and again after filling to determine the actual amount of forage ensiled. This procedure ensured the uniform preparation of the mini-silos, creating a controlled environment for studying the ensiling process. After packing, the jars were sealed with caps to prevent aerobic exposure until the targeted fermentation durations (2, 4, 8, 16, or 32 days). To release the gases formed during the fermentation process, a micro hole was made in the cap, which was sealed after 7 days.

The jars were stored at room temperature (20 °C) in the dark. Each mini-silo was considered an experimental unit. The experiment comprised 10 treatments (two treatments × five storage durations) and 50 silos (five replicates per treatment combination).

2.2. Silages Sampling, Chemical and Microbiological Analyses

At 2, 4, 8, 16, or 32 days post-ensiling, the 3 L glass jars (mini-silos) were opened, and each mini-silo was weighed since the onset of fermentation. The initial weight (prior to fermentation) and the final weight (recorded at the end of each fermentation period), along with the DM content of the material, were used to calculate the DM losses during the fermentation period. One 500 g representative sample was collected from each mini-silo (five replications per treatment) for further chemical and microbial analysis. These analyses included the determination of the following properties: DM, pH, CP, neutral detergent fiber (NDF), acid detergent fiber (ADF), water-soluble carbohydrates (WSCs), volatile fatty acids (VFAs), including key fermentation products such as lactate, acetate, butyrate, and propionate, ammonia-N, yeast and mold counts, and total LAB counts. All chemical and microbial analyses were performed according to the methods described by Jatkauskas et al. [26].

2.3. Calculations for Corrected Dry Matter (DMc) Concentration and Dry Matter Loss

To account for the loss of volatile fermentation products during the drying process of the maize silage samples, the following equation proposed by Weissbach and Strubelt [27] was applied:

DMc = DMn + 0.95 FA + 0.08 LA + 0.77 PD + 1.00 OA (g kg⁻¹ FM)

where FA represent fatty acids (C₂ to C₆), LA represent lactic acid, PD represent 1,2-propanediol, and OA represent other alcohols (C₂ to C₄), including 2,3-butanediol. Each component (FA, LA, PD, OA) was quantified in grams per kilogram of fresh matter (FM) and added to the uncorrected dry matter (DMn) using the specified coefficients.

The DM loss was calculated based on the initial DM content at the time of ensiling and the corrected DM content of the silage (DMc). Considering the DM content of both the forage and the silage, the DM loss was determined using the following formula:

DM loss = (DM at ensiling − DMc silage)/DM at ensiling

2.4. Aerobic Stability Evaluation of the Silages

The aerobic stability of silages was evaluated by monitoring several parameters over a 7-day aerobic exposure period for each fermentation duration. Temperature fluctuations in the aerobically exposed silages were continuously recorded during this period, following the methodology outlined by Jatkauskas et al. [26]. Aerobic stability was defined as the number of hours the silage remained stable before rising more than 3 °C above the ambient temperature [28].

At the end of the aerobic exposure, yeast and mold counts were measured to evaluate microbial activity, potential spoilage, and overall aerobic stability. Additionally, pH levels were assessed throughout the 7-day exposure period.

2.5. Statistical Analyses

The data were analyzed using the GLIMMIX procedure of the SAS Statistical Software (version 9.4) [29], employing a Gaussian distribution for the response variables. The model included the main effects of inoculation (control and treated), the ensiling period (2, 4, 8, 16, and 32 days), and their interactions. Significant differences between means were identified using Tukey’s Studentized Range Test, with p ≤ 0.05 considered statistically significant. Data related to the crop composition were analyzed using the non-parametric NPAR1WAY procedure with the ANOVA and WILCOXON options and applying the Kruskal–Wallis Test.

3. Results

3.1. Characterization of Maize Silage Stored in Mini-Silos and Opened at 2, 4, 8, 16, and 32 Days

3.1.1. Fermentation Profile of the Silages upon Opening

The results obtained indicate that both inoculation and storage duration had a significant effect on the accumulation of fermentation products. Silage treated with LBL showed a mean lactic acid concentration that was 55.5% higher (p < 0.001) compared to the untreated control. This difference was even more pronounced at specific time points, with lactic acid concentrations in the LBL-treated silage being 56.4% higher (p < 0.001) on day 16 and 52.1% higher (p < 0.001) on day 32 of storage (Figure 1). The mean acetic acid concentration was 0.5% higher (p < 0.001) in the LBL-treated silage compared to the control. However, the LBL treatment did not significantly affect acetic acid formation during individual storage periods, except on day 2, when the acetic acid concentration in the LBL-treated silage was 41.5% higher (p < 0.001) than in the control (Figure 2). In the control silage, the butyric acid concentration increased 15.6-fold (p < 0.001) from day 2 to day 32, whereas in the LBL-treated silage, the butyric acid concentration decreased 18-fold (p < 0.001) during the same storage period (Figure 3). There was a slight increase in propionic acid in the LBL-treated silage at 16 and 32 days, but overall, differences between treatments and storage days were minor, with values remaining relatively stable across storage periods in both treatments (Figure 4). Silage treated with LBL exhibited mean ethanol and ammonia-N concentrations that were 35.6% and 20.4% lower, respectively (p < 0.001), than those of the untreated control. These differences became more pronounced over time, with the ethanol and ammonia-N concentrations in the LBL-treated silage being 38.4% and 4.5% lower, respectively (p < 0.001), on day 8, and 42.8% and 29.4% lower, respectively (p < 0.001), on day 32 of storage (Figure 5 and Figure 6). Silage treated with LBL showed a mean pH value that was 0.17 units lower (p < 0.001) than the untreated control (Figure 7). The pH of the LBL-treated silage was lower (by 0.16 units, p < 0.001) than that of the control from day 2 of fermentation. This difference increased to 0.20 units by day 4, then gradually decreased over time. On days 8, 16, and 32 of storage, the pH differences were 0.19, 0.16, and 0.12 units, respectively (p < 0.001).

3.1.2. Nutritional Composition of Silages upon Opening

The addition of the inoculant had no effect on the DM concentration of the silages up to day 8 of fermentation (

Table 3. Nutrient characteristics of maize silages without or with the inoculation and stored for 2, 4, 8, 16, and 32 days before opening.

Items	C	LBL	Mean	SE
DM, %
2 days	34.0 aA	34.0 aA	34.0 A	0.116
4 days	33.8 aA	33.8 aA	33.8 A	0.116
8 days	32.5 cB	33.1 bB	32.8 B	0.116
16 days	31.9 dC	32.9 bcB	32.4 C	0.116
32 days	31.6 dC	32.8 bcB	32.2 C	0.116
Mean	32.8 b	33.3 a
DMc, %
2 days	34.3 aA	34.3 aA	34.3 A	0.076
4 days	34.3 aA	34.3 aA	34.3 A	0.076
8 days	33.4 cdB	33.8 abcA	33.6 B	0.076
16 days	32.9 deB	33.8 bcA	33.3 BC	0.076
32 days	32.7 eB	33.7 cdA	33.4 C	0.076
Mean	33.5 b	34.0 a
CP, % DM
2 days	6.43 aA	6.64 aA	6.53 A	0.110
4 days	6.13 aA	6.64 aA	6.38 AB	0.110
8 days	5.42 bB	6.65 aA	6.03 B	0.110
16 days	5.05 bcB	5.44 bB	5.25 C	0.110
32 days	4.75 cB	5.41 bB	5.08 C	0.110
Mean	5.56 b	6.16 a
NDF, % DM
2 days	41.32 aA	41.46 aA	41.40	0.481
4 days	41.92 aA	41.51 aA	41.72	0.481
8 days	42.32 aA	42.18 aA	42.25	0.481
16 days	42.17 aA	41.66 aA	41.91	0.481
32 days	41.85 aA	41.30 aA	41.58	0.481
Mean	41.91 a	41.62 a
ADF,
2 days	24.53 aA	24.41 aA	24.471 A	0.523
4 days	24.87 aA	23.9 aA	24.38 A	0.523
8 days	24.18 aA	23.52 aA	23.85 A	0.523
16 days	24.12 aA	22.58 aA	23.35 A	0.523
32 days	23.93 aA	22.6 aA	23.26 A	0.523
Mean	24.33 a	23.40 b
DM losses, %
2 days	0.58 eC	0.61 eB	0.60 C	0.193
4 days	0.94 deC	0.66 eB	0.80 C	0.193
8 days	3.97 bB	2.21 cdA	3.09 B	0.193
16 days	5.46 aA	2.44 cA	3.95 A	0.193
32 days	6.10 aA	2.66 bcA	4.38 A	0.193
Mean	3.41 a	1.72 b

C–control; LBL–50:50 combination of Lentilactobacillus buchneri DSM22501 and Lactococcus lactis DSM11037; DM–dry matter; DMc–dry matter corrected for volatiles; CP–crude protein; NDF–neutral detergent fiber; ADF–acid detergent fiber; 2–32 day–storage period of the silages at opening; SE–standard error; a–e–within a row, different letters indicate a significant difference (p <0.001) between treatment groups; A–C–within a column, different letters indicate significant differences (p < 0.001) between days (2–32).

). However, from day 8 to day 32, the LBL-treated silage maintained a DM content that was 1.8–3.8% higher (p < 0.001) than that of the control. From day 2 to day 32, the DM content decreased by 7.0% (p < 0.001) in the control (C) silage and by 3.5% (p < 0.001) in the LBL-treated silage. A similar trend was observed in the DMc content, with the LBL-treated silage maintaining a mean DMc that was 1.5% higher (p < 0.001) than that of the control silage. During the same period, the DMc content decreased by 4.7% (p < 0.001) in the control silage and by 1.7% (p < 0.001) in the LBL-treated silage. Crude protein content declined by 26.1% (p < 0.001) in the control silage and by 18% (p < 0.001) in the LBL-treated silage over the 32 days of fermentation. The LBL-treated silage maintained a mean crude protein that was 10.8% higher (p < 0.001) than that of the control silage. There were no significant differences in NDF content between the treatments and the storage periods. The ADF content remained relatively stable over the 32-day storage period. The LBL-treated silage had a mean ADF content that was 3.8% lower (p < 0.001) than that of the control silage. The efficiency of the fermentation process was improved by reducing the mean DM losses by half (p < 0.001) in the LBL treatment compared to the control. Although DM losses increased over time in both treatment groups, the silage treated with LBL showed significantly lower DM losses—by 80% on day 8 (p < 0.001), by 2.2 times on day 16 (p < 0.001), and by 2.3 times on day 32 (p < 0.001) of storage.

3.1.3. Microbial Profile of the Silages upon Opening

The silage treated with LBL showed mean yeast and mold counts that were 41.0% and 31.6% lower (p < 0.001), respectively, than the untreated control (

Table 4. Microbial characteristics (FM basis) of maize silages without or with the inoculation and stored for 2, 4, 8, 16, and 32 days before opening (data presented as LS means in log₁₀CFU g⁻¹ FM.)

Items	C	LBL	Mean	SE
Yeast, log10CFU g−1 FM
2 days	5.08 aA	4.32 cA	4.70 A	0.051
4 days	4.72 bB	3.63 deB	4.17 B	0.051
8 days	4.63 bcC	2.47 fBC	3.55 C	0.051
16 days	3.93 dD	1.34 gD	2.63 D	0.051
32 days	3.32 eE	1.06 gD	2.19 E	0.051
Mean	4.34 a	2.56 b
Mold, log10CFU g−1 FM
2 days	3.49 aA	2.52 cdA	3.09 A	0.063
4 days	3.12 bB	2.41 dA	2.77 B	0.063
8 days	2.86 bcB	1.97 eB	2.42 C	0.063
16 days	2.71 cdB	1.78 efB	2.25 CD	0.063
32 days	2.67 cdB	1.45 fC	2.06 D	0.063
Mean	2.97 a	2.03 b

C–control; LBL–50:50 combination of Lentilactobacillus buchneri DSM22501 and Lactococcus lactis DSM11037; 2–32 day–storage period of the silages at opening; CFU–colony forming units; SE–standard error; a–g–within a row, different letters indicate a significant difference (p < 0.001) between treatment groups; A–E–within a column, different letters indicate significant differences (p < 0.001) between days (2–32).

). At 4 days of storage, the yeast and mold counts in the LBL-treated silage were 23.1% and 23.0% lower, respectively, compared to the untreated control silage (p < 0.001). These differences became more pronounced over time. At day 8, the yeast and mold counts in the LBL-treated silage were 46.6% and 31.1% lower, respectively (p < 0.001), compared to the control silage. At day 16, the difference reached 65.9% for yeast and 34.3% for mold (p < 0.001). By day 32, the yeast and mold counts in the LBL-treated silage were 68.1% and 45.7% lower, respectively, than in the control (p < 0.001).

3.1.4. Characteristics of Maize Silage Aerobic Stability

The LBL-treated silage exhibited significantly longer aerobic stability time (AST) than the control silage throughout the 32-day fermentation period. Notably, at days 2 and 4 of fermentation, the AST of the LBL-treated silage was 16.8 h and 13.2 h longer, respectively, than those of the control silage (p < 0.001). At days 8, 16, and 32, the AST of the LBL-treated silage exceeded that of the control by 28.8 h, 31.2 h, and 76.4 h, respectively (p < 0.001). Evaluating the impact of storage duration revealed that from day 2 to day 32, the AST of the control silage increased by 16.8 h (p < 0.001), whereas the AST of the LBL-treated silage increased by 58 h (p < 0.001), representing a 3.4-fold improvement. The temperature dynamics of the silages (fermented for 2, 4, 8, 16 and 32 days) during the aerobic exposure period are presented in Figure 8.

At the end of the aerobic stability test, the silage treated with LBL exhibited a mean pH value that was 1.11 units lower than the untreated control (p < 0.001) (

Table 5. Effects of additive type and composition on aerobic stability parameters of maize silage exposed to air for 7 days.

Items	C	LBL	Mean	SE
AST, hours
2 days	32.40 gB	49.20 fgC	40.80 C	0.583
4 days	42.00 fgB	55.20 efC	48.60 C	0.583
8 days	69.60 deA	98.40 bcB	84.00 B	0.583
16 days	78.00 dA	109.20 bB	94.20 B	0.583
32 days	88.00 cdA	146.40 aA	117.60 A	0.583
Mean	62.16 b	91.92 a
pH value
2 days	8.49 aA	8.09 bcA	8.30 A	0.153
4 days	8.30 abB	7.83 cdA	8.10 A	0.153
8 days	7.90 cdB	6.90 eB	7.39 B	0.153
16 days	7.80 cdB	6.70 eB	7.28 B	0.153
32 days	7.58 dB	4.77 fC	6.22 C	0.153
Mean	8.01 a	6.90 b
Yeast, log10CFU/g FM
2 days	7.88 cdB	7.50 efA	7.69 AB	0.155
4 days	7.64 deB	7.37 efA	7.50 C	0.155
8 days	8.00 bcA	7.25 fA	7.62 BC	0.155
16 days	8.28 abA	7.39 fA	7.83 A	0.155
32 days	8.30 aA	7.24 fA	7.77 AB	0.155
Mean	8.02 a	7.35 b
Mold, log10CFU/g FM
2 days	6.69 efD	6.47 efBC	6.58 D	0.151
4 days	7.18 cdC	6.79 deBC	6.99 BC	0.151
8 days	7.46 cC	6.25 fB	6.85 CD	0.151
16 days	8.69 aA	7.37 cA	8.03 A	0.151
32 days	8.16 bB	6.30 efB	7.23 B	0.151
Mean	7.64 a	6.63 b

C–control; LBL–50:50 combination of Lentilactobacillus buchneri DSM22501 and Lactococcus lactis DSM11037; AST–aerobic stability test; FM–fresh matter; CFU–colony forming units; Yeast and mold, expressed as log₁₀ CFU g⁻¹ FM; 2–32 day–storage period of the silages at opening; SE–standard error; a–g–within a row, different letters indicate a significant difference (p < 0.001) between treatment groups; A–D–within a column, different letters indicate significant differences (p < 0.001) between days (2–32).

). After aerobic exposure, the LBL-treated silage fermented for 2, 4, 8, 16, and 32 days had pH values that were 0.40, 0.47, 1.00, 1.10, and 2.81 units lower, respectively, compared to the control (p < 0.001). Evaluating the effect of storage duration showed that from day 2 to day 32, the pH of control silage decreased by 0.91 units, whereas the pH of the LBL-treated silage decreased by 3.32 units (p < 0.001), representing a 3.4-fold greater reduction compared to the control. At the end of the aerobic stability test, the silage treated with LBL showed mean yeast and mold counts that were 8.3% and 13.2% lower, respectively, than those in the untreated control (C) (p < 0.001). After 2 and 4 days of fermentation, the LBL-treated silage exhibited yeast and mold counts that were 4.8% and 8.3%, and 3.2% and 5.4% lower, respectively, compared to the untreated C silage (p < 0.001). After 8 days of fermentation, the yeast and mold counts for the LBL-treated silage were 9.4% and 12.9% lower, respectively (p < 0.001), compared to the control. At 16 and 32 days of fermentation, the yeast and mold counts in the LBL-treated silage were 10.7% and 12.8%, and 15.2% and 22.8% lower, respectively, compared to the control (p < 0.001).

4. Discussion

4.1. Characteristics of Forage Prior to Ensiling

The water-soluble carbohydrates, crude protein content, and LAB number influence the rate of pH decrease during the early stages of ensiling, which is critical for successful silage fermentation. Additionally, the fermentability coefficient and LAB number also affect the rate of pH decline during this phase [6,30,31]. In the present study, the forage exhibited low buffer capacity, and the calculated fermentability coefficient was moderate (38.0). Consequently, the whole-plant maize forage was classified as being easy to moderately difficult to ensile [32]. The epiphytic LAB count in fresh whole-plant maize was 4.97 log₁₀CFU g⁻¹ FM, indicating that ensiling should not be difficult. However, the presence of yeast and mold at levels of 5.75 and 5.52 log₁₀CFU g⁻¹ of fresh matter suggests a risk of rapid aerobic degradation and low aerobic stability of the silage. This highlights the potential challenges of ensiling without inoculants, emphasizing the necessity of adding LAB inoculants to enhance silage aerobic stability. Therefore, Lentilactobacillus buchneri—a heterofermentative LAB known for enhancing aerobic stability through the production of acetic acid and 1,2-propanediol, which inhibit spoilage microorganisms—and Lactococcus lactis DSM11037—a relatively novel LAB strain with high oxygen-scavenging capacity that enables it to dominate the early fermentation phase, rapidly lower pH, and suppress undesirable aerobic microbes—were used in this study.

4.1.1. Nutritional Composition and Fermentation Characteristics of Silages upon Opening

The heterofermentative lactic acid bacteria Lentilactobacillus buchneri has been widely used as a silage inoculant due to its ability to improve aerobic stability through the production of acetic acid and 1,2-propanediol, which inhibit spoilage microorganisms and delay aerobic deterioration [22,23]. In contrast, Lactococcus lactis DSM11037 is a homofermentative LAB strain characterized by its strong oxygen-scavenging capacity and rapid pH reduction, making it highly effective during the initial stages of fermentation [24]. The complementary activity of these two strains in our study likely supported a more controlled and efficient fermentation, which contributed to the improved preservation of DM, DMc, and CP contents. While a decline in DM and DMc was observed over time, as expected during ensiling, silages treated with the dual-strain inoculant (LBL) consistently maintained higher DM and DMc values compared to the untreated control, particularly after 16 and 32 days of storage. This suggests that enhanced fermentation quality reduced the nutrient losses typically associated with uncontrolled microbial activity. Notably, previous concerns that heterofermentative pathways might lead to greater DM losses were not supported in our findings, aligning with the meta-analysis by Kleinschmit and Kung [9]. It can be assumed that the LBL mixture effectively outcompeted the epiphytic LAB microflora, guiding the fermentation process toward a more favorable direction. During the silage process, crude protein degradation is unavoidable, and some proteins are converted to non-protein-N, especially ammonia-N under the action of plant enzymes and microorganisms [33]. In the present study, crude protein content declined over time, with a more pronounced reduction in the control silages. The inoculated silages (LBL) maintained higher crude protein levels across all (2, 4, 8, 16, and 32 days) storage durations. At day 32 of fermentation, the crude protein level of the LBL-treated silage was 14% higher (p < 0.01) compared to the control, indicating the effectiveness of LBL treatment. This could be attributed to reduced proteolysis in the inoculated silage compared to the control [34], further supported by the 20% lower (p < 0.001) average ammonia-N concentration observed in the LBL silages.

The control silages exhibited slightly higher ADF values than the LBL-treated silages, suggesting a potential beneficial effect of inoculation on fiber degradation. This aligns with the previous studies that reported an increase in fiber fractions (ADF and aNDF) in silage following inoculation [35,36]. Cai et al. [37] observed increased aNDF levels in LAB-inoculated corn stover silage during ensiling, highlighting its potential as a viable silage resource to address animal feed challenges. In the current experiment, DM losses increased significantly over time, with the highest values observed at 16 and 32 days. The inoculated silage (LBL) consistently exhibited lower DM losses compared to the control (1.72% vs. 3.41%, p < 0.001), indicating improved DM recovery. Similar DM losses in maize silage inoculated with Lentilactobacillus buchneri at various concentrations was reported by Ranjit et al. [38]. Likewise, Oude Elferink et al. [22] observed DM losses of only 1.4% in laboratory silos of maize silage inoculated with different strains of Lentilactobacillus buchneri. The speed of acidification and the maintenance of a low pH during ensiling are critical properties of LAB inoculants in producing high-quality silages. McDonald et al. [39] and Xu et al. [40] suggested that microorganisms used for silage preparations should have a uniform fermentation route, which not only rapidly ferments to produce the maximum amount of lactic acid and quickly lowers the pH to inhibit other microorganisms but also has the ability to withstand acid. In our study, the LBL-inoculated silages exhibited greater acid-producing ability compared to the non-inoculated silages. The mean pH values of the LBL-inoculated silages were significantly lower than those of the non-inoculated silages (4.10 vs. 4.27, p < 0.001). Additionally, the strains of Lentilactobacillus buchneri and Lactococcus lactis performed well in producing lactic acid and acetic acid relative to epiphytic LAB. It appears that Lactococcus lactis, due to its oxygen-scavenging ability, played a crucial role in initiating early fermentation, which supported the observed results. Specifically, it was particularly effective in inhibiting Clostridia, reducing pH, and producing lactic acid and acetic acid [24] faster than the control silage. The significant pH reduction in the LBL-inoculated maize silages at the start of fermentation (4.55 vs. 4.39, p < 0.001) compared to the non-inoculated silages—and continuously thereafter—can be linked to the dominance of Lentilactobacillus buchneri and Lactococcus lactis, which rapidly ferment WSC to synthesize LA, resulting in a rapid pH decrease. As expected, LA concentration increased more rapidly in the LBL-inoculated maize silages than in the non-inoculated silages. On the other hand, significantly higher mean AA contents were detected in the LBL-inoculated silages compared to the contol silages (1.05 vs. 1.00% DM, p < 0.001). Until now, heterofermentative LAB have been the most prevalent LAB group, synthesizing significant amounts of AA from LA conversion during fermentation [6,41]. Moreover, during the anaerobic phase of silage fermentation, lactic acid is gradually converted to acetic acid by heterofermentative lactic acid bacteria, such as Lentilactobacillus buchneri [42]. The LBL-treated silage effectively inhibited butyric acid production, reduced ethanol accumulation, and lowered NH₃-N levels, indicating improved fermentation efficiency and enhanced protein preservation due to reduced protein breakdown. The frequently observed higher ethanol concentration in silages inoculated with Lentilactobacillus buchneri, compared to other silage types [43], was confirmed by evaluating the contrast (p < 0.01) between the control (mean concentration: 1.01% DM) and LBL (mean concentration: 0.65% DM) treatments.

Higher BA concentrations can be found in poorly fermented silages, caused by clostridial fermentation, which results in reduced nutritional value due to the breakdown of soluble nutrients [43]. Our findings also suggest lower yeast activity and decreased protein degradation.

4.1.2. Microbial and Aerobic Stability Characteristics

The enumeration of yeast and mold in silage is useful because high yeast counts are typically associated with elevated ethanol concentrations and are often inversely related to aerobic stability. This is particularly important for corn-based crops [44,45]. In our experiment, LBL-treated silages showed reduced yeast and mold growth in maize silage across all fermentation periods (2, 4, 8, 16, and 32 days). On the day 32 of fermentation, the yeast and mold counts in the LBL-treated silage were lower by 3.1 times and 54.3%, respectively (p < 0.001), compared to the control (C) silage, indicating that Lactococcus lactis was more effective in oxygen removal during the initial phase of the fermentation process [24]. Longer storage times further suppressed microbial growth, with the LBL treatment showing the most pronounced effect. In contrast, non-inoculated silage retained higher microbial counts, indicating a greater risk of spoilage. According to Muck et al. [6], when silage is exposed to air, considerable changes in the chemical composition occur, including a significant increase in pH and a marked rise in temperature due to oxygen exposure. In our experiment, aerobic stability increased over time in both treatments when comparing silages fermented for 2 and 32 days, with a 55.6 h increase in the control (C) silage and a 97 h increase in the LBL-treated silage (p < 0.001), indicating that the silage became more stable with longer storage and highlighting the impact of LBL treatment.

The most relevant and significant effects of additive application were observed in the yeast count and temperature dynamics during the aerobic exposure period. The LBL treatment resulted in a lower temperature increase across all periods of silage exposure to air, reflecting improved stability. The LBL treatment more effectively maintained pH levels throughout the aerobic exposure period (mean pH of 6.90 in LBL vs. 8.01 in C, p < 0.001), as well as lower yeast and mold counts (mean yeast and mold counts of 7.35 and 6.63, respectively, in LBL vs. 8.02 and 7.64 in the control, respectively, p < 0.001), further supporting enhanced aerobic stability. According to previous studies [24,26], the combination of Lentilactobacillus buchneri and Lactococcus lactis effectively reduced yeast populations, allowing for a delay of 24 h before silo closing, compared to untreated silage. A similar trend was observed in the present trial, where the yeast counts began to decline as early as 2 days after ensiling.

These findings support the hypothesis that oxygen scavenging during the initial phase of fermentation plays a key role in reducing aerobic microorganism populations, accelerating fermentation, and promoting early acetic acid production, thereby enhancing resistance to aerobic spoilage upon air exposure. Importantly, this suggests that the prevention of aerobic deterioration is not solely dependent on acetic acid concentration but also linked to controlling aerobic microbial populations early in the fermentation process [46]. This observation aligns with the results of Hindrichsen et al. [24] and Jatkauskas et al. [26], indicating that limiting aerobic microorganisms from the outset, combined with acetic acid production, provides a dual protective effect against spoilage when silage is exposed to air. Aerobic stability indicators notably improved with longer silage fermentation time. Numerous studies [47,48,49] have demonstrated that inoculation with Lentilactobacillus buchneri, either alone or in combination with other LAB, consistently inhibits yeast growth and subsequently enhances the aerobic stability of silage.

5. Conclusions

The application of an LBL inoculant at a rate of 10⁵ CFU g⁻¹ of whole-plant maize significantly improved the fermentation quality and preservation efficiency of silage over a 32-day storage period. Compared to the untreated control, the LBL-treated silage exhibited higher dry matter, improved DM and CP contents, reduced DM losses by two times (1.72% vs. 3.41%), and lower concentrations of undesirable fermentation products, such as butyric acid, ethanol, and ammonia-N. The LBL inoculant limited bacterial growth in whole-plant maze silage by 68.1% and preserved crude protein levels that were 10.8% higher throughout the 32-day storage period. Targeted inoculation not only accelerates early acidification but also effectively controls aerobic microbial populations, enhancing silage aerobic stability. The use of additive LBL has the potential to minimize the formation of climate-relevant volatile organic compounds in silage.