Synergistic Effects of Exogenous Lactobacillus plantarum and Fibrolytic Enzymes on Fermentation Quality, Fiber Degradation, and In Vitro Digestibility of Napiergrass (Pennisetum purpureum) Silage

Abstract

This experiment aimed to evaluate the effects of exogenous L. plantarum, fibrolytic enzymes, and their combination on fermentation quality, structural carbohydrate components, and in vitro dry matter digestibility (IVDMD). The treatments included (i) no additives (control), (ii) L. plantarum (L), (iii) fibrolytic enzymes (E), and (iv) a combination of fibrolytic enzymes and L. plantarum (EL). After being fermented for 1, 3, 7, and 30 days, the silos were opened for subsequent analysis. L and EL increased the lactic acid content and decreased the pH value and NH₃-N content compared to silages without the addition of L. plantarum (p < 0.05). Compared to the control, enzymes alone or combined with L. plantarum improved enzymatic hydrolysis with higher water-soluble carbohydrates being retained at the early stage of ensiling; lower contents of NDF, ADF, hemicellulose, and cellulose were observed at the end of ensiling (p < 0.05). The IVDMD was improved in E and EL silage, and the highest IVDMD was observed in E. The L silage showed no significant difference in terms of the structural carbohydrate components or IVDMD compared to the control (p > 0.05). A principal component analysis showed that L. plantarum addition did not contribute to an increase in IVDMD, whereas LA fermentation was further enhanced when EL was synergistically involved.

1. Introduction

Napiergrass (Pennisetum purpureum) holds great promise for ruminant production in tropical and subtropical regions due to its nature as a perennial forage with substantial biomass. If not promptly disposed of after harvest, a considerable portion remains unused and inevitably leads to spoilage. Ensiling is an effective technique for preserving surplus fodder in large quantities and providing year-round feed supply for animals [1]. A lack of epiphytic lactic acid bacteria (LAB) hinders successful ensiling because LAB play a key role in silage acidification, fermenting water-soluble carbohydrates (WSCs) into lactic acid under anaerobic environments [2]. Inoculation with Lactobacillus plantarum (L. plantarum) facilitates facultative homofermentation, effectively increasing the lactic acid content and decreasing pH to inhibit the growth of harmful microorganisms, reducing the risk of feedstock spoilage and enabling the long-term preservation of rice straw silage [3]. Previous research confirmed that adding L. plantarum ensured quick fermentation in tall fescue on account of its extensive utilization of substrates and prolific lactic acid production [4].

In general, napiergrass, with its low WSC content, has demonstrated poor fermentation quality [5,6]. The stemmy structures of napiergrass, along with its high cell wall composition, pose a challenge to achieving proper compaction, leading to air being trapped in the silo [7]. As a result, plant respiration and aerobic microbial activity accelerate water-soluble carbohydrate consumption. Adding exogenous fibrolytic enzymes could accelerate structural carbohydrate degradation [8]. Various types of enzymes, such as endoglucanases, cellobiohydrolases, and β-glucosidases, are used in mixtures that act serially or synergistically to decompose crude cellulose [9]. In this case, hydrolyzed saccharide can serve as a fermentable substrate for LAB, addressing the shortage of WSC during ensiling [10].

Forage usage benefits from abundant plant biomass, particularly from napiergrass, wherein the content of structural carbohydrates exceeds 500 g kg⁻¹ DM after field wilting, and this concentration in the cell walls is difficult to digest due to the complex ultrastructure [11]. Adding fibrolytic enzymes significantly improves the digestibility by breaking down structural carbohydrates, such as cellulose and hemicellulose, into simpler sugars that are more accessible to microbial fermentation, thereby enhancing the nutrient availability for ruminants [12]. Wang et al. [4] demonstrated that the combination of L. plantarum and fibrolytic enzymes decreased the structural carbohydrate content and improved fermentation quality. However, other researchers reported that the natural barrier formed by the interaction of hemicellulose, lignin, and some neutral detergent-soluble components resist the microbial and enzymatic hydrolysis of cellulose [13]. It was hypothesized that fibrolytic enzymes combined with L. plantarum may synergistically improve fermentation quality. However, a thorough understanding of this combination’s effect on the structural carbohydrate degradation of napiergrass during ensiling remains to be elucidated.

The objective of this experiment was to evaluate the effect of exogenous fibrolytic enzymes, L. plantarum, and their combination on fermentation quality, structural carbohydrate components, and in vitro dry matter digestibility (IVDMD).

2. Materials and Methods

2.1. Ensiling Material and Silage Making

Napiergrass (Pennisetum purpureum) was grown in the experimental field of Nanjing Agricultural University located at 32.04°N, 118.85°E, at an elevation of 15 m a.m.s.l, with an annual mean temperature of 15.7 °C and an average yearly rainfall of 1021 mm. The soil in the field was characterized by the following properties: 19.4 g kg⁻¹ organic matter, 1.41 g kg⁻¹ total nitrogen, 10.8 mg kg⁻¹ available phosphorus, and 69.8 mg kg⁻¹ exchangeable potassium. To ensure the optimal growth, the planting density for the napiergrass was set at a row width of 100 cm and an intra-distance of 50 cm.

After harvested at the late vegetative stage, the napiergrass was cut to a theoretical size of 2 cm by using a fodder chopper; then, it was thoroughly mixed and was randomly divided into 4 parts. The following treatments were implemented in the experiment: (i) no additives (control); (ii) L. plantarum at 1 × 10⁶ cfu g⁻¹ (L); (iii) 0.1% fibrolytic enzymes (E); (iv) a combination of 0.1% fibrolytic enzymes and L. plantarum at 1 × 10⁶ cfu g⁻¹ (EL). The fibrolytic enzyme was a multi-enzyme complex (BIOCOM biotechnology Co., Ltd., Hangzhou, China), and the enzymes activity was quantified: xylanase 2000 U g⁻¹, feruloyl esterase 200 U g⁻¹ and cellulose 100 U g⁻¹. Both the fibrolytic enzymes and L. plantarum (Snow Brand Seed Co., Ltd., Sapporo, Hokkaido, Japan) were added based on the fresh material weight and diluted with sterile distilled water to achieve the desired concentrations. The processed material was homogenized and tightly packed into laboratory silos. The silos were hermetically sealed using threaded tops and plastic tapes and then maintained at room temperature (24 °C). After ensiling for 1, 3, 7, and 30 days, a total 48 silos (4 treatments × 4 ensiling days × 3 replicates) were opened and sampled for subsequent analyses.

2.2. Chemical Analysis

Silage was separated into three parts at the time of silo opening. The first part was dried at 65 °C for 48 h until it reached a constant weight to determine the dry matter (DM) content. The dried samples were then processed using a knife mill and subsequently sieved through a 1 mm sieve to obtain a finely ground powder. This powder was analyzed to determine neutral detergent fiber (NDF), acid detergent fiber (ADF), acid detergent lignin (ADL), ash, water-soluble carbohydrates (WSCs), total nitrogen (TN), and ether extract (EE). The WSC content was determined using a colorimetric method based on the reaction with anthrone reagent [14]. The TN content was analyzed via a Kjeldahl nitrogen analyzer (Kjeltec 8400, FOSS, Vastra Gotaland, Sweden). EE and ash were determined according to the method of the Association of Official Analytical Chemists [15]. NDF, ADF, and ADL were quantified following the protocol established by Van Soest et al. [16] employing an ANKOM filter bag and a fiber analyzer (ANKOM 200i, ANKOM Technologies, Inc., Fairport, NY, USA). The crude protein (CP) content was calculated as TN × 6.25, while cellulose, hemicellulose, and lignin contents were derived by NDF minus ADF, ADF minus ADL, and ADL minus ash, respectively.

The second part was submerged in deionized water at a ratio of 1:3 (w/v) and maintained at 4 °C for 1 day. The water extract was filtered through two layers of gauze and filter paper. The filtrate was analyzed for pH, buffering capacity (BC), and the concentrations of ammonia nitrogen (NH₃-N) and organic acids. The pH value was measured using a glass electrode pH meter (HANNA pH 211, Hanna Instruments Italia Srl, Padova, Italy) to measure. The BC was measured according to the method reported by Tucker et al. [17]. The NH₃-N content was measured via the phenol sodium–hypochlorite reaction [18]. An Agilent HPLC 1260 system (Agilent Technologies, Inc., Santa Clara, CA, USA) fitted with a refractive index detector and a Carbomix^® H-NP5 column (Sepax Technologies, Inc., Santa Clara, CA, USA) was employed to determine the contents of organic acids. The mobile phase consisted of 2.5 mmol L⁻¹ H₂SO₄ at a flow rate of 0.5 mL min⁻¹, and the column temperature was maintained at 55 °C to optimize separation.

The third part was prepared by homogenizing it with a sterilized saline solution (0.85% NaCl) in ratio of 1:9 (w/v), which was followed by serial dilution through 6-fold dilution steps. The resulting diluted samples were plated onto specific media for microbial incubation. After incubation at 30 °C for 2–3 days, lactic acid bacteria (LAB), aerobic bacteria (AB), and yeasts were counted on plates of Man, Rogosa and Sharpe agar (Difco Laboratories, Detroit, MI, USA), nutrient agar (Sinopharm Chemical Reagent Co., Ltd., Shanghai, China), and potato dextrose agar (Sincere Biotech Co., Ltd., Shanghai, China) mediums, respectively. All microbiological data were log₁₀ transformed for statistical analysis.

2.3. Measurement of IVDMD

Rumen fluid sampling was performed according to the guidelines issued by the Animal Welfare and Ethical Committee of the University. Rumen fluid was collected from two steers cannulated in the rumen before morning feeding. Before incubation, the rumen fluid was filtered through 4 layers of gauze and blended with buffer at a ratio of 1:2 (v/v). Pre-warmed (39 °C) serum bottles (Shuniu Glass Instrument Co., Ltd., Chengdu, China) were prepared to contain the 30 days silage samples, which were packed into the nylon bag (ANKOM Technology, Macedon, NY, USA) and flushed with CO₂ to keep an anaerobic condition. The bottles were sealed with rubber plugs and then manually shaken and incubated in a water bath at 39 °C for 72 h. At 6 h intervals, the headspace gas production was expelled using a needle. Three additional empty filter bags served as blanks. After incubation, the nylon bags were rinsed with distilled water and dried at 65 °C to a consistent weight before being weighed to determine in vitro dry matter digestibility (IVDMD).

2.4. Statistical Analyses

One-way analysis of variance (ANOVA) was used to analyze the in vitro dry matter digestibility, and the fermentation profiles, chemical compositions and microorganism count were analyzed by two-way analysis of ANOVA subjected to the general linear model (GLM) procedure of SAS software version 9.2 (SAS Institute, Cary, NC, USA):

Y_ij = μ + T_i + D_j + (T × D)_ij + e_ij

Here, Y_ij represents the dependent variable; μ represents the overall mean; T_i denotes the effect of treatments; D_j indicates the effect of ensiling days; (T × D)_ij represents the interaction between treatments and ensiling days; and e_ij represents the residual error term. Significance was defined as p < 0.05. Least-squares means were adjusted using Tukey’s multiple comparison method.

To analyze the correlation between the profiles and fermentative variables, principal component analysis (PCA) with biplot visualization was constructed employing Origin 2018 (OriginLab Corporation, Northampton, MA, USA).

3. Results

3.1. Chemical and Microbial Composition of Napiergrass

As shown in

Table 1. Chemical and microbial composition of fresh napiergrass.

Items	Value
Chemical composition
DM (g kg−1 FW)	378
WSC (g kg−1 DM)	90.2
CP (g kg−1 DM)	55.1
EE (g kg−1 DM)	42.6
Ash (g kg−1 DM)	68.8
Buffering capacity (mEq kg−1 DM)	102
NDF (g kg−1 DM)	541
ADF (g kg−1 DM)	389
ADL (g kg−1 DM)	114
Hemicellulose (g kg−1 DM)	152
Cellulose (g kg−1 DM)	275
Lignin (g kg−1 DM)	46.5
microbial composition
Lactic acid bacteria (log10 CFU g−1 FW)	3.17
Aerobic bacteria (log10 CFU g−1 FW)	7.54
Yeast (log10 CFU g−1 FW)	5.60

DM, dry matter; FW, fresh weight; WSC, water-soluble carbohydrate; CP, crude protein; EE, ether extract; NDF, neutral detergent fiber; ADF, acid detergent fiber; ADL, acid detergent lignin.

, the fresh napiergrass had a DM content of 378 g kg⁻¹ FW and WSC of 90.2 g kg⁻¹ DM. The contents of cellulose, hemicellulose, and lignin were 275, 152 and 46.5 g kg⁻¹ DM, respectively. The BC level was 102 mEq kg⁻¹ DM. The aerobic bacteria count was 7.54 log₁₀ CFU g⁻¹ FW, whereas the counts of LAB and yeast were below 1 × 10⁶ log₁₀ CFU g⁻¹ FW.

3.2. Dynamics Changes in Fermentation Profiles and Structural Carbohydrates Compositions During Ensiling

As presented in

Table 2. Changes in pH, and the contents of dry matter, lactic acid, WSC, and NH₃-N of napiergrass during ensiling.

Items	Treatments	Days of Ensiling				Mean	SEM	p-Value
		1	3	7	30			T	D	T × D
DM (g kg−1 FW)	Control	373	360	354	346	358AB	2.31	<0.001	<0.001	0.078
	L	371	364	360	357	363A
	E	378	363	353	341	355B
	EL	376	370	367	361	369A
	Mean	375a	364b	359bc	351c
pH	Control	5.22	4.78	4.71	4.56	4.82A	0.106	<0.001	<0.001	0.306
	L	4.94	4.79	4.63	4.14	4.63A
	E	5.12	4.85	4.57	4.43	4.74A
	EL	4.51	4.23	3.81	3.76	4.08B
	Mean	4.95a	4.66b	4.43	4.22c
Lactic acid (g kg−1 DM)	Control	18.5Bb	20.0Bb	22.4Ca	24.1Ca	21.3	2.60	<0.001	<0.001	<0.001
	L	21.3Ac	25.1Bbc	28.2Bb	33.1Ba	26.9
	E	17.4Bb	18.9Cb	24.1BCa	25.9Ca	21.6
	EL	21.1Ad	33.5Ac	42.0Ab	58.2Aa	38.7
	Mean	19.6	24.4	29.2	35.3
WSC (g kg−1 DM)	Control	78.6ABa	68.1Ab	52.5Ac	33.9Ad	58.3	5.22	<0.001	<0.001	0.001
	L	74.4Ba	54.5Bb	40.7Bc	24.6Bd	48.6
	E	87.6Aa	70.3Ab	54.9Ac	31.2Ad	61.0
	EL	84.2Aa	64.2ABb	42.0Bc	22.3Bd	53.2
	Mean	81.2	64.3	42.0	22.3
NH3-N (g kg−1 TN)	Control	26.7d	42.1Ac	55.4Ab	75.2Aa	49.9	4.08	<0.001	<0.001	<0.001
	L	28.3c	34.6ABbc	40.1Bb	62.5Ba	41.4
	E	27.8d	37.9ABc	45.0Bb	76.3Aa	46.8
	EL	26.7c	31.2Bc	39.4Bb	57.2Ba	38.6
	Mean	27.4	36.5	45.0	67.8

Note: T, effect of additives; D, effect of ensiling day; T × D, interaction between additives and ensiling day; means with different superscripts in the same row (a–d) or column (A–C) differed at p < 0.05. Abbreviations: DM, dry matter; FW, fresh weight; WSC, water-soluble carbohydrate; NH₃-N, ammonia–nitrogen; TN, total nitrogen; SEM, standard error of the mean; L, L. plantarum; E, 0.1% fibrolytic enzymes; EL, combination of 0.1% fibrolytic enzymes and L. plantarum.

and

Table 3. Changes in volatile acids and LA/AA of napiergrass during ensiling.

Items	Treatments	Days of Ensiling				Mean	SEM	p-Value
		1	3	7	30			T	D	T × D
Acetic acid (g kg−1 DM)	Control	7.44c	10.8b	12.3ab	14.4Ba	11.2	0.855	0.011	<0.001	<0.001
	L	6.98c	8.95bc	10.2b	13.2Ba	9.83
	E	7.58c	9.83bc	11.8b	15.2Ba	11.1
	EL	6.25c	8.14bc	9.88b	18.6Aa	10.7
	Mean	7.06	9.43	11.0	15.4
LA/AA	Control	2.49	1.85	1.82	1.67	1.96C	0.220	<0.001	0.005	0.080
	L	3.05	2.80	2.76	2.51	2.78B
	E	2.30	1.92	2.04	1.70	1.99C
	EL	3.38	4.12	4.25	3.13	3.72A
	Mean	2.81a	2.67a	2.72a	2.25b
Propionic acid (g kg−1 DM)	Control	ND	0.39Bb	0.59b	0.88Ba	0.62	0.098	<0.001	<0.001	<0.001
	L	ND	0.20C	0.33	0.35C	0.29
	E	ND	0.66Ab	0.70b	1.29Aa	0.88
	EL	ND	ND	ND	ND
	Mean		0.42	0.54	0.84
Butyric acid (g kg−1 DM)	Control	ND	ND	0.22ABb	0.68a	0.45	0.070	<0.001	<0.001	<0.001
	L	ND	ND	0.08B	0.19	0.14
	E	ND	ND	0.26Ab	0.94a	0.60
	EL	ND	ND	ND	ND
	Mean			0.19	0.60

Note: T, effect of additives; D, effect of ensiling day; T × D, interaction between additives and ensiling day; means with different superscripts in the same row (a–c) or column (A–C) differed at p < 0.05. Abbreviations: DM, dry matter; LA, lactic acid; AA, acetic acid; ND, no detected; SEM, standard error of the mean; L, L. plantarum; E, 0.1% fibrolytic enzymes; EL, combination of 0.1% fibrolytic enzymes and L. plantarum.

, the effects of treatments and ensiling day influenced all fermentation parameters (p < 0.05), while their interaction had no significant effect on DM (p = 0.078), pH (p = 0.306) and the lactic/acetic acid ratio (p = 0.080). With the increase in ensiling days, the DM content, WSC content, and pH value linearly (p < 0.05) decreased, while LA, AA and NH₃-N contents increased significantly (p < 0.05). The L and EL silages increased significantly (p < 0.05) the LA content and lactic/acetic acid ratio as well as decreased the pH, NH₃-N, and WSC contents as compared to the control and E silages. EL had the highest DM, LA contents, and LA/AA ratio among all silages (p < 0.05). Propionic and butyric acid were detected in trace amounts in all silages.

Changes in the structural carbohydrates composition of napiergrass during ensiling are presented in

Table 4. Changes in NDF, ADF, ADL, and ash of napiergrass during ensiling.

Items	Treatments	Days of Ensiling				Mean	SEM	p-Value
		1	3	7	30			T	D	T × D
NDF (g kg−1 DM)	Control	539Aa	534Aa	530Aab	518Ab	530	11.9	<0.001	<0.001	<0.001
	L	536Aa	533Aa	530Aa	501Ab	521
	E	510Ba	486Bab	461Bb	382Bc	461
	EL	512Ba	497Bab	477Bb	395Bc	470
	Mean	524	513	500	449
ADF (g kg−1 DM)	Control	386	382	375A	369A	379	6.22	<0.001	<0.001	<0.001
	L	384	383	377A	358A	373
	E	379a	368ab	351Bb	305Bc	351
	EL	376a	370a	356Ba	312Bb	354
	Mean	381	376	365	336
ADL (g kg−1 DM)	Control	114	113	114	110	113A	0.931	<0.001	0.001	0.163
	L	113	115	112	104	111AB
	E	109	112	109	106	109B
	EL	111	110	108	105	109B
	Mean	112a	113a	111a	106b
Ash (g kg−1 DM)	Control	68.3	69.6	69.1	66.2	68.3	0.759	0.267	0.101	0.573
	L	65.9	65.9	67.8	62.5	65.5
	E	66.7	69.4	67.4	64.8	67.1
	EL	65.5	68.5	65.4	64.8	66.1
	Mean	66.6	68.4	67.4	64.6

Note: T, effect of additives; D, effect of ensiling day; T × D, interaction between additives and ensiling day; means with different superscripts in the same row (a–c) or column (A,B) differed at p < 0.05. Abbreviations: DM, dry matter; NDF, neutral detergent fiber; ADF, acid detergent fiber; ADL, acid detergent lignin; SEM, standard error of the mean; L, L. plantarum; E, 0.1% fibrolytic enzymes; EL, combination of 0.1% fibrolytic enzymes and L. plantarum.

and

Table 5. Changes in the structural carbohydrates of napiergrass during ensiling.

Items	Treatments	Days of Ensiling				Mean	SEM	p-Value
		1	3	7	30			T	D	T × D
Hemicellulose (g kg−1 DM)	Control	153	152	155A	148A	152	6.49	<0.001	<0.001	0.006
	L	152	149	153A	141A	149
	E	131a	118a	110Ba	76.5Bb	109
	EL	136a	128a	121Ba	83.5Bb	117
	Mean	143	137	135	112
Cellulose (g kg−1 DM)	Control	272	269	261AB	259A	265	5.76	<0.001	<0.001	<0.001
	L	273	269	265A	259A	266
	E	270a	256ab	242Cb	199Bc	242
	EL	265a	260a	248BCa	206Bb	245
	Mean	270	264	254	231
Lignin (g kg−1 DM)	Control	46.0	43.4	44.9	43.9	44.6	0.870	0.074	0.357	0.729
	L	45.4	49.0	44.4	43.4	45.6
	E	42.4	42.7	42.1	41.6	42.2
	EL	45.8	41.6	42.6	40.8	42.7
	Mean	44.9	44.2	43.5	42.4

Note: T, the effect of additives; D, the effect of the ensiling day; T × D, the interaction between additives and ensiling day; means with different superscripts in the same row (a–c) or column (A–C) differed at p < 0.05. Abbreviations: DM, dry matter; SEM, standard error of the mean; L, L. plantarum; E, 0.1% fibrolytic enzymes; EL, combination of 0.1% fibrolytic enzymes and L. plantarum.

. The contents of NDF, ADF, hemicellulose, and cellulose decreased significantly (p < 0.001), while a constant trend in ash and lignin was observed in all silages (p > 0.05). The E and EL silages exhibited a rapid decrease in NDF, ADF, hemicellulose, and cellulose contents after 7 days of ensiling, and they were significantly lower than the control and L silages at the end of ensiling (p < 0.05).

Treatments and ensiling days had a significant (p < 0.001) effect on the microorganism counts (

Table 6. Changes in microorganism of napiergrass during ensiling.

Items	Treatments	Days of Ensiling				Mean	SEM	p-Value
		1	3	7	30			T	D	T × D
LAB (log10 CFU g−1 FW)	Control	3.51Bc	3.89Bbc	4.22Bb	5.25Ca	4.22	0.364	<0.001	<0.001	<0.001
	L	5.68Ac	6.62Ab	7.37Aa	6.27Bbc	6.49
	E	3.32Bc	4.05Bb	4.59Bb	5.44Ca	4.35
	EL	5.81Ac	6.74Ab	7.55Aa	7.84Aa	6.99
	Mean	4.58	5.33	5.93	6.20
Aerobic bacteria (log10 CFU g−1 FW)	Control	7.42a	5.08Ab	4.70Ab	4.52Ab	5.43	0.373	<0.001	<0.001	0.003
	L	7.01a	4.66ABb	4.03ABbc	3.63Bc	4.83
	E	7.29a	5.23Ab	4.41Ac	4.12ABc	5.26
	EL	7.33a	4.01Bb	3.42Bb	2.65Cc	4.35
	Mean	7.26	4.75	4.14	3.73
Yeasts (log10 CFU g−1 FW)	Control	5.52	3.08	2.41	2.13	3.29AB	0.352	<0.001	<0.001	0.118
	L	5.40	2.56	2.00	2.00	2.99B
	E	5.61	3.37	2.48	2.39	3.46A
	EL	5.36	2.33	2.00	2.00	2.92B
	Mean	5.47a	2.84b	2.22bc	2.13c

Note: T, the effect of additives; D, the effect of the ensiling day; T × D, the interaction between additives and ensiling day; means with different superscripts in the same row (a–c) or column (A–C) differed at p < 0.05. Abbreviations: LAB, lactic acid bacteria; CFU, colony-forming unit; FW, fresh weight; SEM, standard error of the mean; L, L. plantarum; E, 0.1% fibrolytic enzymes; EL, combination of 0.1% fibrolytic enzymes and L. plantarum.

). L and EL silages had significantly higher LAB and lower aerobic bacteria and yeasts counts compared to control and E silages.

The structural carbohydrates compositions and in vitro dry matter digestibility of the 30-day silages are presented in Figure 1. The distribution of hemicellulose plus cellulose, and lignin ranged 86.9–90.3% and from 13.1–9.70% across all treatments, respectively. The IVDMD in the E (66.9%) and EL (65.5%) silages was significantly higher (p < 0.05) than the control (55.5%) and L (56.9%) silages.

3.3. Correlation of Fermentative Profiles and Parameters in Silages

The principal component analysis graph is presented in Figure 2. A distinct separation was observed with L, E, and EL. Meanwhile, the control counterparts were spread out around the right half of the graph. The symbols representing each treatment were grouped together. LAB, LA, and AA were positively correlated and defined the EL group while showing a negative correlation with pH. NDF, ADF, hemicellulose, cellulose, and lignin show very sharp angles, highly signifying a positive correlation with a negative correlation observed with IVDMD.

4. Discussion

4.1. Fermentation Profile and Microbial Counts

During ensiling, the increase in lactic acid content and rapid decrease in pH were associated with WSC utilization by lactic acid bacteria (LAB). With a WSC content of 90.2 g kg⁻¹ DM, napiergrass meets the requirements for lactic acid fermentation, as a WSC content above 60 g kg⁻¹ DM is deemed sufficient for quality silage [6]. L and EL increased the lactic acid content and decreased the pH, which could attributed to the exogenous LAB that facilitated lactic acid fermentation and improved fermentation quality. Significantly higher DM content was also observed in L and EL silages, which was related to higher LA and lower pH that inhibited the growth of undesirable microbes and decreased nutrient substrate loss during ensiling. Although E silage exhibited higher WSC content compared to other silages during the early stages of ensiling, E silage presented the lower lactic acid content and high pH value, which was less efficient than L. plantarum in enhancing the fermentation quality; this may be ascribed to the short of epiphytic LAB population in napiergrass.

For high dry matter silage, the ideal pH for stable fermentation typically ranges from 4.5 to 5.0 [19]. A greater quality silage was observed in EL, which was characterized by the lowest NH₃-N content, pH, and yeast counts, and the highest LA content. This synergistic interaction of LAB with fibrolytic enzymes improved fermentation quality by breaking down complex structural carbohydrates into water-soluble carbohydrates, making them more available for LAB fermentation and creating a more acidic environment [20]. All silages increased acetic acid (AA) content during ensiling, which may result from pentose released from hemicellulose degradation, stimulating heterolactic fermentation, and a linear decrease in the LA/AA ratio without L. plantarum indicating a shift from homofermentation to heterofermentation [21].

4.2. Structural Carbohydrates Degradation and Digestibility

In this experiment, all treated silages reduced the contents of structural carbohydrates compositions compared to the control, which was in line with the work of Liu et al. [22] and Zhao et al. [23]. In L silage, the higher LA content maintained low pH levels, and acid hydrolysis facilitated the decomposition of the cell wall, leading to a numerical decrease in NDF, ADF, and hemicellulose contents. A similar research indicated that the significant reduction in hemicellulose and NDF levels was attributed to the acid-induced breakdown of structural carbohydrates during ensiling [24].

E decreased the hemicellulose and cellulose significantly as compared with control. Fibrolytic enzymes, including xylanase, cellulase, and feruloyl esterase, directly hydrolyze structural carbohydrates into monosaccharides, providing more fermentable substrates [25]. However, the EL silage exhibited a slight numerical increase in hemicellulose and cellulose compared to E. This could be due to the higher lactic acid and lower pH than E, which could potentially inhibit fibrolytic enzyme activity. Several studies have reported that fibrolytic enzymes had different optimal pH ranges [26,27].

In vitro dry matter digestibility is negatively correlated with structural carbohydrates content, particularly lignin, which is a complex phenolic polymer that resists microbial and enzymatic degradation, providing rigidity to the plant cell wall and limiting access to cellulose and hemicellulose [28]. Silages treated with fibrolytic enzymes had a higher IVDMD than those treated with L. plantarum. This can be attributed to enzymolysis, which reduced the proportion of hemicellulose and cellulose, thereby mitigating the negative impact of lignin on digestibility. This suggested that a higher lignin content does not necessarily imply a lower IVDMD if the cellulose and hemicellulose fractions are degraded sufficiently.

4.3. Correlation Between Treatments and Fermentative Parameters

Principal component analysis (PCA) illustrated the correlation between treatments and fermentative parameters after 30 days of ensiling. The correlation in each profile was defined by the angle between the arrows pointing at two variables, where sharp angles stipulated positive correlations, orthogonal angles stipulated no correlations, and obtuse angles stipulated negative correlations [29]. The PCA biplot clearly separated the different treatment groups, EL showing the most distinct clustering, indicating a significant alteration in EL and fermentative profiles compared to the control. The positive correlation between LAB counts, lactic acid (LA), and acetic acid (AA) with EL; these suggested that the combined treatment of fibrolytic enzymes and L. plantarum resulted in a higher accumulation of fermentation products, and the findings also underscore the effectiveness in improving the fermentation quality. IVDMD is characterized by a positive correlation with the E and EL group, indicating that fibrolytic enzymes facilitate the degradation of structural carbohydrates, leading to the disruption of the plant cell wall matrix. This disruption not only liberates fermentable substrates but also accelerates the digestion of associated macronutrients.

5. Conclusions

Combining L. plantarum and fibrolytic enzymes significantly enhanced the fermentation quality, and the IVDMD was improved in silage with fibrolytic enzymes evidently.