Ryegrass is a cool-season grass widely used in ruminant feeding in southern Brazil (Cauduro et al., 2007), and in temperate and subtropical regions such as the United States, China, and Korea (Liu et al., 2020). The ensiling of grasses, such as ryegrass, can be a viable alternative for forage preservation during periods of scarcity or adverse weather conditions (Uslu et al., 2017). However, at the ideal stage of cutting for ensiling, these grasses generally have high moisture content (Borreani et al., 2018). In this context, field wilting emerges as an effective strategy to adjust the dry matter (DM) content of high-moisture forages at the time of ensiling.

The DM content of the forage has a direct influence on the silage fermentation process and, consequently, on its quality. Values below 30% DM, when associated with low concentration of soluble carbohydrates in grass silages, may favor undesirable fermentation such as those caused by microorganisms of the genus Clostridium (Muck et al., 2018), resulting in greater fermentative losses, reduced DM content, and excessive effluent production (Queiroz et al., 2018). In this context, wilting forage grasses prior to ensiling may help adjust the DM content and reduce losses. However, after cutting, cellular activities such as respiration continue to occur (Melo et al., 2023). Thus, the time between harvesting and silo sealing can influence the availability of substrates in the plant material, which are essential for proper fermentation.

The application of microbial additives during the ensiling process can enhance fermentation, especially under more challenging conditions (Liu et al., 2020), such as those encountered in wilted forage silages. The use of inoculants containing beneficial microorganisms that promote both fermentation and silage stability (Muck et al., 2018) can be an effective strategy for ensiling wilted ryegrass. Heterofermentative inoculants containing strains of Lentilactobacillus buchneri and other heterolactic bacteria, as the recently commercially available Lentilactobacillus hilgardii (Ferrero et al., 2021), act on the forage by promoting greater production of lactic and acetic acids, rapid pH decline, and reduced post-harvest proteolysis (Liu et al., 2020), potentially contributing to the prevention of heating and nutrient deterioration in wilted grasses.

Several studies in the literature have reported the use of enzymes in animal feeding with the aim of improving nutrient utilization by animals (Beauchemin et al., 2003; Antonio et al., 2018). Xylanase, for example, hydrolyzes glycosidic bonds, breaking them into simpler sugars and consequently increasing feed digestibility (Beauchemin et al., 2003). The combined use of xylanase and β-glucanase in animal diets has been shown to improve the digestibility of hay from different grasses (Antonio et al., 2018). In the forage preservation process, the use of enzymes such as xylanase and β-glucanase, in combination with the benefits of microbial inoculants, may enhance sugar release from the forage, promoting efficient fermentation, and contributing to greater stability and preservation of the ensiled material.

Thus, the objective of this study was to evaluate the effects of two different DM contents at ensiling, with or without the addition of commercial microbial inoculants, on the fermentation of wilted ryegrass (Lolium multiflorum L.) silages.

The chemical composition and microbiological analyses of the pre-wilted ryegrass before ensiling are presented in Table 1. The population of enterobacteria was high, exceeding 10⁷ CFU g⁻¹ in both harvests (32 and 45% DM) The chemical composition was similar between both harvests.

The composition of the pre-wilted ryegrass silages is shown in Table 2. Most of the variables analyzed were affected by the treatments (inoculants and harvest stage), as well as by their interaction. Inoculated silages showed higher DM content in the 32% DM harvest and lower pH in both harvests, indicating a more favorable fermentation profile. Silages with higher moisture content (32% DM) presented lower pH (3.97) and higher levels of lactic acid (21.5 g kg⁻¹ DM), as well as a greater population of lactic acid bacteria (8.38 log CFU g⁻¹). As a result, these silages had lower DM (8.6 g kg⁻¹ DM) and gas losses (8.1 g kg⁻¹ DM), no effluent production, and lower DM losses during aerobic exposure (7% DM) compared to silages harvested at 45% DM (Table 3). Acetic acid content was not influenced by the inoculants (Table 2). The INO inoculant treatment resulted in a lower 1,2-propanediol concentration at both DM levels (Table 2).

Table 2

Variable	32% DM			Mean	45% DM			Mean	P-value1
	C	INO	INO+E		C	INO	INO+E		DM	Inoc	D × I
Dry matter – 105 oC (g kg−1)	343b	366a	359a	356	475	475	480	477	<0.0001	<0.0001	0.002
pH at opening (day 0)	4.19a	3.86b	3.91b	3.97	4.95a	4.08b	4.12b	4.38	<0.001	<0.001	<0.001
Crude protein (g kg−1 DM)	154	150	148	151	173a	162b	174a	170	<0.0001	0.002	0.007
NDF (g kg−1 DM)	527	531	523	527	548b	551b	572a	557	<0.0001	0.044	0.001
ADF (g kg−1 DM)	337	346	336	340	354	353	352	353	0.006	0.323	0.605
Ash (g kg−1 DM)	107a	97b	107a	103	101b	105a	106a	104	0.873	<0.001	<0.001
IVDMD (%)	70.7b	75.0a	75.1a	73.6	71.1ab	68.9b	71.3a	70.4	<0.0001	0.005	0.0001
IVOMD (%)	71.2b	75.5a	75.6a	74.1	71.6	69.9	71.9	71.1	<0.0001	0.003	0.0001
IVNDFD (%)	50.7b	57.2a	55.3a	54.4	52.2a	48.3b	53.7a	51.4	<0.001	<0.001	<0.001
Lactic acid (g kg−1 DM)	14.8b	21.8a	28.0a	21.5	5.9b	17.7a	18.8a	14.1	<0.001	<0.001	0.243
Acetic acid (g kg−1 DM)	3.5	3.0	3.7	3.4	3.4	3.0	2.7	3.0	0.004	0.723	0.347
Propionic acid (mg kg−1 DM)	49a	39b	38b	42	103a	41b	51b	65	<0.001	<0.001	<0.001
Butyric acid (mg kg−1 DM)	43a	6b	6b	15	12a	8b	9b	10	<0.001	<0.001	<0.001
Ethanol (g kg−1 DM)	0.44	0.69	0.47	0.53	0.99	0.80	0.53	0.77	0.810	0.222	0.046
1,2-Propanediol (mg kg−1 DM)	299a	114b	295a	236	160a	70b	114ab	114	<0.001	<0.001	0.008
Total LAB (log CFU g−1)	8.73a	7.84b	8.63a	8.38	8.12a	7.27c	7.60b	7.66	<0.001	<0.001	0.216
Enterobacteria (log CFU g−1)	4.37a	2.74b	2.20b	3.01	4.34a	3.44b	2.54c	3.44	0.085	<0.001	0.323
Yeasts (log CFU g−1)	<104	<104	<104	-	<104	<104	<104	-	-	-	-
Molds (log CFU g−1)	<104	<104	4.58	-	<104	<104	4.19	-	-	-	-

NDF - neutral detergent fiber; ADF - acid detergent fiber; IVDMD - in vitro digestibility of dry matter; IVOMD - in vitro digestibility of organic matter; IVNDFD - in vitro digestibility of neutral detergent fiber; LAB - lactic acid bacteria; CFU - colony forming units.

Table 3

Variable	32% DM			Mean	45% DM			Mean	P-value1
	C	INO	INO+E		C	INO	INO+E		DM	Inoc	D × I
DM losses (g kg−1 DM)	37.2a	−21.6c	10.1b	8.6	38.4b	60.7a	53.6ab	50.9	<0.001	0.011	<0.001
Gas losses (g kg−1 DM)	36.7a	−21.9c	9.3b	8.1	37.3b	60.5a	53.6ab	50.5	<0.001	0.014	<0.001
Effluent (kg/ton wet basis)	0.51	0.25	0.76	0.51	1.16	0.29	0.00	0.48	0.637	0.194	0.053
Aerobic stability - AS (hour)	386	324	349	353	228b	366a	388a	327	0.119	0.021	<0.001
Max. Temp. (oC)	32.0	31.5	34.0	32.5	33.8	30.3	28.5	30.9	0.179	0.335	0.061
Hour to TMax (hour)	447	405	416	422	315b	431a	460a	402	0.219	0.034	0.001
DMLae (g kg−1 DM)	43.2	47.4	114.2	70.0	263.4a	71.6b	70.9b	135.3	0.008	0.011	<0.001
pH day 2 of AS	4.19a	3.84b	3.86b	3.95	5.10a	4.02b	4.06b	4.39	<0.001	<0.001	<0.001
pH day 4 of AS	4.16a	3.88b	3.92b	3.97	4.97a	4.07b	4.16b	4.40	<0.001	<0.001	<0.001
pH day 6 of AS	4.11a	3.86b	3.88b	3.94	4.76a	4.14b	4.10b	4.33	<0.001	<0.001	<0.001
pH day 8 of AS	4.07a	3.87b	3.92b	3.94	4.70a	4.08b	4.13b	4.30	<0.001	<0.001	<0.001
pH day 10 of AS	4.09	3.90	4.14	4.04	5.11a	4.41b	4.17b	4.57	<0.001	<0.001	<0.001
pH day 18 of AS	4.14	6.01	6.65	5.71	7.50	5.77	4.60	5.96	0.440	0.878	<0.001
pH day 20 of AS	7.34	8.19	8.40	8.02	8.65	6.08	6.21	6.98	0.045	0.319	0.009

C - Wilted ryegrass silage control; INO - Wilted ryegrass silage with Lentilactobacillus hilgardii CNCM I-4785 (1×10⁵ CFU g⁻¹ forage), Lentilactobacillus buchneri NCIMB 40788 (1×10⁵ CFU g⁻¹ forage), and Pediococcus pentosaceus NCIMB 12455 (1×10⁵ CFU g⁻¹ forage); INO+E - Wilted ryegrass silage with L. hilgardii (7.5×10⁴ CFU g⁻¹ forage), L. buchneri (7.5×10⁴ CFU g⁻¹ forage), and P. pentosaceus (1×10⁵ CFU g⁻¹ forage), plus enzymes β-glucanase (5,750 IU g⁻¹ of product) and xylanase (30,000 IU g⁻¹ of product).

Ryegrass silages harvested at 45% DM showed crude protein content approximately two percentage points higher than those harvested at 32% DM (Table 2). On the other hand, at 32% DM, the inoculants INO and INO+E significantly increased the in vitro estimates of digestibility for DM, OM, and NDF. However, at 45% DM, both inoculants showed similar digestibility coefficients to the control silage (Table 2).

The counts of enterobacteria in 32% DM silages were 37.3% and 49.6% lower for INO and INO+E treatments, respectively, compared with the control silage. However, in 45% DM silages, the INO+E inoculant showed a stronger effect on reducing enterobacteria populations (Table 3). Yeast and mold counts were lower than expected, with no colonies observed at the lowest dilution tested (10⁻⁴). Silages treated with INO+E showed a slight tendency toward higher fungal colony numbers; however, no statistics could be performed.

The 32% DM silages presented lower estimates of DM, gas, and effluent losses than the 45% DM treatments (Table 3). In the 32% DM silages, the INO inoculant showed negative values for gas (−21.9 g kg⁻¹ DM) and DM losses (−21.6 g kg⁻¹ DM). At 45% DM, the treatment INO increased DM losses (60.7, 53.6, and 38.4 g kg⁻¹ DM for INO, INO+E, and control, respectively) and gas losses (60.5, 53.6, and 37.3 g kg⁻¹ DM, respectively).

Aerobic stability (AS) was high for all samples (Table 3 and Figure 1). At 45% DM, the inoculants significantly enhanced AS. Silages harvested at 32% DM were more stable than those at 45% DM, with no inoculant effect. Control silages heated only after 10 days of aerobic exposure, although visible signs of spoilage and unpleasant odor (not evaluated) were observed after six days. In contrast, silages treated with INO and INO+E at 45% DM retained a typical silage odor even after 16 days of exposure at 25 °C (not evaluated). The DM losses under aerobic exposure (DMLae) were higher in 45% DM silages, and the inoculants were effective in reducing them compared with the 32% DM silages.

Figure 1 C32 and C45 - Wilted ryegrass silage control with 32% and 45% DM, respectively; INO-32 and INO-45 - Wilted ryegrass silage with Lentilactobacillus hilgardii CNCM I-4785 (1×10⁵ CFU g⁻¹ forage), Lentilactobacillus buchneri NCIMB 40788 (1×10⁵ CFU g⁻¹ forage), and Pediococcus pentosaceus NCIMB 12455 (1×10⁵ CFU g⁻¹ forage) with 32% and 45% DM, respectively; INO+E-32 and INO+E-45 - Wilted ryegrass silage with L. hilgardii (7.5×10⁴ CFU g⁻¹ forage), L. buchneri (7.5×10⁴ CFU g⁻¹ forage), and P. pentosaceus (1×10⁵ CFU g⁻¹ forage), plus enzymes β-glucanase (5,750 IU g⁻¹ of product) and xylanase (30,000 IU g⁻¹ of product), with 32% and 45% DM, respectively.

The pH of the silages remained stable until the 10th day of aerobic exposure. Due to a technical failure, no samples were collected for pH analysis between the 10th and 18th days. After 18 days, pH values showed high variability among replicates, with only the INO-45 and INO+E-45 treatments maintaining average pH values close to 5.0 (Figure 2).

Figure 2 C32 and C45 - Wilted ryegrass silage control with 32% and 45% DM, respectively; INO-32 and INO-45 - Wilted ryegrass silage with Lentilactobacillus hilgardii CNCM I-4785 (1×10⁵ CFU g⁻¹ forage), Lentilactobacillus buchneri NCIMB 40788 (1×10⁵ CFU g⁻¹ forage), and Pediococcus pentosaceus NCIMB 12455 (1×10⁵ CFU g⁻¹ forage) with 32% and 45% DM, respectively; INO+E-32 and INO+E-45 - Wilted ryegrass silage with L. hilgardii (7.5×10⁴ CFU g⁻¹ forage), L. buchneri (7.5×10⁴ CFU g⁻¹ forage), and P. pentosaceus (1×10⁵ CFU g⁻¹ forage), plus enzymes β-glucanase (5,750 IU g⁻¹ of product) and xylanase (30,000 IU g⁻¹ of product), with 32% and 45% DM, respectively.

Traditionally, farmers wilt ryegrass for silage until 45-50% of DM. Interestingly, few studies have evaluated lower DM content as a goal for wilted silages. In this trial, the 32% DM treatments seem to produce better quality silages over a shorter period.

The chemical composition of the fresh pre-wilted ryegrass evaluated in this study was similar to the values reported by Conaghan et al. (2010). The high population of enterobacteria in the fresh forage may be associated with dust contamination during the wilting process under field conditions. Wang et al. (2020) reported counts of up to 8.16 log CFU g^-1 of enterobacteria in fresh Italian ryegrass forage. However, despite the high counts of enterobacteria in the fresh forage, the fermentation process contributed to the preservation of the ensiled material, as it reduced the counts of enterobacteria, yeasts, and molds, while increasing the population of lactic acid bacteria, regardless of the treatment.

The higher DM content and lower pH values observed at silo opening in inoculated silages harvested at 32% DM, compared with the control silage, indicate a more favorable fermentative profile of the material. These results are consistent with the higher lactic acid concentrations, increased lactic acid bacteria (LAB) population, and lower butyric acid concentrations observed in these treatments (Table 2).

The benefits of heterolactic LAB on silage fermentation and preservation are well documented in the literature (McDonald et al., 1991; Muck et al., 2018). The main advantage of intensified LAB activity is the production of compounds that enhance silage preservation, resulting in reduced or negligible fermentative losses. Among these compounds, lactic acid stands out due to its low pKa value (3.86), which is the lowest among the acids commonly found in silage, thereby promoting a rapid decline in pH (Muck et al., 2018). pH is one of the key factors influencing silage fermentation quality (Wang et al., 2020). Many undesirable microorganisms in silage, such as certain fungi, exhibit limited activity in low-pH environments (McDonald et al., 1991). Furthermore, higher acetic acid concentrations may contribute to the inhibition of undesirable microorganisms, such as yeasts, due to its well-known antifungal action (Pordeus et al., 2023). Our results demonstrate that the evaluated inoculants were effective in controlling the activity of undesirable microorganisms, such as enterobacteria, and promoted higher concentrations of lactic and acetic acids while reducing butyric acid levels contributing to the preservation of the ensiled material, particularly in wilted ryegrass silages harvested at 32% DM.

As previously mentioned, some cellular activities continue to occur in plants even after cutting, which can influence the concentration of essential compounds for fermentation, such as water-soluble carbohydrates (Melo et al., 2023). The ryegrass used in the 45% DM treatment remained in the field for an additional 24 h to reach the DM content, which led to higher CP and NDF content, likely as result of the loss of soluble carbohydrates plant respiration under field conditions (not evaluated). The 45% DM silages presented higher pH and lower acid production than the 32% DM silages, and the inoculants were less effective in improving fermentation in drier silages, possibly due to reduced substrate availability. Nevertheless, even under these conditions, silages treated with inoculants showed lower pH values than the untreated untreated control, which can be attributed greater lactic acid production. These results indicate that inoculants may also be useful in high-DM forages, which are generally more susceptible to suboptimal fermentation. The lower DM losses, reduced gas production, and absence of effluent during the fermentation process, as well as the lower DM losses during aerobic exposure observed in the 32% DM silages, can be attributed to the greater activity of desirable microorganisms and the metabolites produced by their actions, which play a crucial role both during fermentation and after exposure of the silage to air (Gheller et al., 2021). These results indicate clear advantages for farmers, since keeping the forage in the field for a shorter period represents a lower risk of losses due to adverse weather conditions (Uslu et al., 2017). Moreover, shorter wilting periods contribute to nutrient preservation, which influences the fermentation process and, consequently, the aerobic stability (AS) of the silages, extending the safe utilization period after silo opening.

Higher DM content forages may present compaction difficulties at the time of ensiling, resulting in greater residual oxygen (Kung et al., 2018), which contributes to increased losses both during the fermentation process and after air exposure. However, this effect was isolated in the present study, as similar bulk densities were used for both DM levels. The 45% DM silages exhibited, on average, sixfold greater DM losses during fermentation compared with the 32% DM silages, as well as higher DM losses during aerobic exposure and a shorter aerobic stability period. A negative interaction between DM content and pH, associated with longer field wilting periods results in a reduction of water-soluble nutrients, increased contamination by undesirable microorganisms, and lower fermentative capacity (Uslu et al., 2017). These factors may explain the higher pH values and greater DM losses observed in 45% DM silages.

Even at 45% DM, the inoculants used extended silage aerobic stability by 138 and 160 h for the INO and INO+E treatments, respectively, and reduced DM losses after air exposure by 19.2%. In addition, up to the 10th day of aerobic exposure no significant changes in silage pH were observed in silages treated with the inoculants, regardless of DM content. From a practical standpoint, these results are highly relevant for farmers, as they indicate an extended utilization window for the preserved feed.

The lower counts of enterobacteria observed in silages treated with INO and INO+E, both at 32 and 45% DM, represent a relevant finding, considering that these microorganisms are potentially pathogenic and compromise silage quality (Queiroz et al., 2018). In addition to directly competing with LAB for water-soluble carbohydrates, enterobacteria are associated with greater DM losses during the fermentation process (McDonald et al., 1991). In a meta-analysis, no significant effect of wilting was detected on LAB, yeast, or aerobic bacteria populations (Ridla et al., 2024). However, the authors report that wilting is related to a general reduction in microbial activity in silages, with a significant decline in undesirable microorganisms. In the present study, the enterobacteria, yeast, and mold counts in wilted ryegrass prior to ensiling were substantially reduced after fermentation, indicating that the ensiling process was effective in suppressing the activity of these undesirable microorganisms.

After aerobic exposure, yeasts and other fungi resume their activity, consuming carbohydrates and nutrients from the silage, which results in gas and heat production and leads to material spoilage (Borreani et al., 2018). However, the observation that control silages exhibited a temperature increase only after 10 days of aerobic exposure despite visible spoilage and unpleasant odor from the sixth day onward, suggests that temperature alone may not be a reliable indicator of aerobic deterioration in wilted grass silages. Diepersloot et al. (2022) 240 h of temperature-based aerobic stability in wilted bermudagrass silages. Gomes et al. (2019) reported pH changes of wilted oat silages after 8 days of air exposure. However, these authors did not report visual or olfactory indicators of silage deterioration.

The main goal of ensiling is to preserve the nutritional value of the conserved forage as much as possible (McDonald et al., 1991). The action of enzymes on cellulose and hemicellulose during the fermentation process can reduce the resistance of the NDF fraction (Muck et al., 2018), making it more digestible. However, the lower NDF contents and higher digestibility of this fraction observed in 32% DM silages, compared to 45% DM silages, can be explained by greater acid hydrolysis of hemicellulose, favored by the lower pH achieved in inoculated silages relative to the control. The inoculated 32% DM silages showed higher NDF, OM, and DM digestibility compared with the control silages, whereas no additional effects of the enzymes present in the INO+E inoculant were observed.

At 45% DM silages treated with INO+E showed higher NDF content, but digestibility coefficients greater than those of INO and similar to the control. These findings suggest that the enzymes used (xylanase and β-glucanase) promoted partial polysaccharide breakdown without increasing solubilization in neutral detergent. The study published by Antonio et al. (2018), which evaluated the use of enzymes in hay conservation (a drier material than wilted forage), demonstrated that xylanase, either alone or in combination with glucanase and cellulases, increased NDF and dry matter digestibility. Together, these results suggest that enzymatic activity may be more pronounced in higher-pH substrates, such as the 45% DM silages compared to the 32% silages.

The results of this study indicate that the traditional DM content applied by farmers (around 45% DM) for ensiling wilted forages may not always represent the most advantageous strategy, especially under unfavorable climatic conditions. It was observed that silage with a lower DM content (32%) required less time in the field to reach the target DM level, exhibited lower fermentative losses, improved chemical composition, and greater resistance to aerobic deterioration after silo opening. These findings highlight the importance of reassessing established management practices based on scientific evidence. Therefore, dissemination of this information through extension services is essential to support improved decision-making by farmers when producing ryegrass silage or haylage.

The treatment with 32% DM resulted in higher-quality silages, requiring less time in the field, with a more favorable fermentation profile and reduced losses during both the fermentation process and after silo opening. The use of inoculants reduced silage pH compared to the control. At 32% DM, inoculants enhanced DM and NDF digestibility, as well as increased lactic acid concentration. In silages with 45% DM, inoculants also increased lactic acid content and significantly improved aerobic stability. These results highlight the importance of setting an appropriate DM content for ensiling, and using additives as key strategies for producing silages with higher nutritional value and improved stability after silo opening.