Effects of Neolamarckia cadamba Leaf Extract on Dynamic Fermentation Characteristics and Bacterial Community of Stylosanthes guianensis Silage

: This study focused on exploring the effects of Neolamarckia cadamba leaf extract (NE) on the fermentation process and bacterial community of stylo ( Stylosanthes guianensis ) silage. Fresh raw materials of stylo were ensiled with 0%, 1%, and 2% NE, and various fermentation parameters, nitrogen components, and microbial compositions were analyzed at different time points (days 3, 7, 14, and 30) during the ensiling process. The experiment showed that, in comparison to the control group, incorporating NE into the ensiling process resulted in improved fermentation parameters, including increased lactic acid and acetic acid levels, as well as decreased pH, coliform population, and ammonia nitrogen concentration. Moreover, the relative abundances of Lactobacillus and Pediococcus were augmented, while the growth of Enterobacter was inhibited by the NE addition. These results suggest that NE has potential as a novel additive for silage, promoting a reduction in harmful bacteria and enhancements in the nutritional quality and fermentation characteristics of stylo silage.


Introduction
Stylo (Stylosanthes guianensis), also called "Tropical alfalfa", is a very popular legume forage native to Central and South America, and it is widely used in the tropics and subtropics for its high yield and protein content [1,2].A common way to preserve stylo in humid tropical and subtropical areas is through silage [3].Stylo silage quality is poor in the absence of additives, as reported by Liu et al. [4].High buffering capacity, low water-soluble carbohydrates (WSCs), and fewer lactic acid bacteria limit the fermentation of legume feed such as stylo [5].Lactic acid bacteria, molasses, cellulase, organic acids, and some new additives are used in silage to improve silage preservation [6][7][8].
Plant extracts are natural active substances derived from plants, with antioxidant, anti-inflammatory, and flora-regulating functions, and various applications in medicine, food, and other industries [9][10][11].Many studies have shown that plant extracts have good performance in improving the meat quality of animal products, enhancing the immune function of animals, and improving animal growth and development.They also have an excellent effect in reducing greenhouse gas emissions from animals, making them important feed additives [12][13][14].In addition, research has found that some plant extracts improve silage fermentation and bacterial communities.For example, ellagic acid enhances fermentation quality and protein retention, while ethanol extracts of oregano and thyme reduce mycotoxin levels in maize silage [15,16].
Typically found in tropical and subtropical regions, Neolamarckia cadamba is fastgrowing [17].Neolamarckia cadamba leaves have been studied as a woody forage in recent years due to their rapid growth, high protein content, and high biomass [18].Moreover, the leaves are rich in alkaloids, flavonoids, tannins, and phenolic acids, which have antioxidant, antibacterial, anticancer, and therapeutic effects on various diseases [19][20][21].Some phytochemicals, such as chlorogenic acid, dihydrocadambine, and β-sitosterol, have broad-spectrum antibacterial activity [22].These extracts have been reported to have significant antibacterial activity against Acinetobacter, Escherichia, Klebsiella, and Pseudomonas, which are also abundant in silage [23][24][25].As a result, the extraction of natural compounds from Neolamarckia cadamba leaves may be beneficial for improving silage quality.There is currently little research on the role of Neolamarckia cadamba leaf extract in silage, which requires further investigation.
In this study, it was hypothesized that the addition of Neolamarckia cadamba leaf extract (NE) would prevent undesirable fermentation during the ensiling of forage.NE contains rich active ingredients and biological activity, which can help improve silage quality.In order to experiment with this hypothesis, stylo underwent ensiling with either 0%, 1%, or 2% NE added.Assessments were conducted at various intervals (days 3, 7, 14, and 30) to assess the ensiling characteristics, nitrogen fractions, and bacterial community of the ensiling fermentation to evaluate the potential for improving stylo silage quality through NE addition.

Preparation of N. cadamba Leaf Powder
The leaves of N. cadamba were harvested from an experimental plot located at the South China Agricultural University (113 • 35 ′ E, 23 • 16 ′ N, Guangzhou, China), and all leaves were air-dried at 30 • C in fume hoods until they reached a constant weight.The dry leaves were ground using a Taisite Instrument Co. grinder (FW100, Tianjin, China) and sifted through a 1 mm screen.The powder was sealed in an enclosed bag and stored in a shaded place until the NE extraction.

Extraction of NE
The dry leaf powder was thoroughly mixed with 70% methanol solution (solid-toliquid ratio: 1:30) and extracted in an ultrasonic cleaner tank (JM-15D-40, Shenzhen, China).Afterward, the solvent was filtered using filter paper and the evaporation was carried out at 50 • C and −90 kPa.The concentrated extract was freeze-dried using a Biocool Experimental Instrument Co. lyophilizer (PILot5-8S, Beijing, China) to obtain the final NE extract.The NE extract was sealed in an enclosed bag and stored in a −20 • C refrigerator until silage experiments were conducted.

Stylo Silage Preparation
The fresh stylo (CIAT 184) was harvested from the above-mentioned experimental plot.The forage materials were chopped into pieces of around 2 cm using a forage chopper.The chopped stylo was mixed split into 3 treatments: no additives (CK), 1% NE (NE1), and 2% NE (NE2).NE was added to stylo based on fresh matter (FM) level.All additives were mixed homogenously with stylo, and then approximately 200 g of chopped stylo was put into an vacuum-sealed plastic bag (20 cm × 30 cm), as referenced in Zi et al. [26].In total, 36 bags (3 treatments × 4 ensiling periods × 3 replicates) were prepared and stored at room temperature (28 to 30 • C).For each treatment, three bags were randomly selected for the analysis of fermentation performance at room temperature after 3, 7, 14, and 30 days.

Microbial Counts and Chemical Composition Analysis
Referring to the methodology of Wang et al. [2], 20 g of each sample was randomly taken and combined with 180 mL of sterile saline solution and diluted stepwise to 10 −6 times to obtain 6 gradients of sample extract.The different bacteria and fungi in the samples were counted on specific agar media: Man-Rogosa-Sharpe agar for LAB, Violet Red Bile agar for coliform bacteria, and Rose Bengal agar for yeasts and molds.All the media used in the experiment were purchased from Guangdong Huankai Bio-tech.For every sample, 20 g was added to 80 mL of distilled water.After an overnight refrigeration at 4 • C, filter paper was used to filter the mixture.The pH, organic acid concentrations, and ammonia nitrogen (AN) were measured using the filtrate.The pH of the filtrate was determined using a pH meter.Levels of lactic acid (LA), acetic acid (AA), propionic acid (PA), and butyric acid (BA) were assessed using high-performance liquid chromatography to determine the concentration of organic acids.The determination of ammoniacal nitrogen was carried out following the procedure outlined by Broderick et al. [27].
For the purpose of determining dry matter (DM), each silage sample weighing about 100 g was dried in an oven at 65 • C for 48 h.For subsequent chemical analyses, dried silage samples were ground into powder to pass through a 1 mm screen.The determination of the DM loss was based on the method described by Köhler and colleagues [28].The anthrone method was used to quantify the content of water-soluble carbohydrates (WSCs) [29].Crude protein (CP) and true protein (TP) were analyzed using a Kjeldahl nitrogen analyzer(Kjeltec 2300 Auto-Analyser, FOSS Analytical AB, Hoganas, Sweden) and their difference was used to calculate non-protein nitrogen (NPN) content [30].The acid detergent fiber (ADF) and neutral detergent fiber (NDF) were analyzed according to the procedures by Van Soest et al. [31].

Bacterial Community Analyses
Following Yu et al. [32], bacterial DNA was extracted using a soil DNA kit (Magen, Guangzhou, China).The 16S rDNA V3-V4 region was amplified by PCR.According to the manufacturer's instructions, a DNA gel extraction kit (Axygen Biosciences, Union City, CA, USA) was used to extract the amplicons, and these were quantified in an ABI StepOnePlus Real-Time PCR System (Life Technologies, Foster City, CA, USA).The PCR products were purified and sequenced using the Illumina platform.Liu et al.'s [33] method was used to assemble the raw reads.

Statistical Analysis
Two-way analysis of variance was performed using SPSS v 22.0 software (SPSS Inc., Chicago, IL, USA) to assess the impact of treatments, storage days, and their interaction on stylo silage.Multiple comparisons were performed using Duncan's test, with p < 0.05 designating a significant difference.A free web tool (http://www.omicshare.com/tools) was used to analyze the sequencing data and perform correlation analyses, accessed on 10 December 2023.The figures were made using GraphPad Prism 9.3 and OriginPro 9.1 software and further enhanced using Adobe Illustrator 2022 v26.0.1.

Characteristics of Raw Stylo Materials Prior to Ensiling
In Table 1, the characteristics of raw stylo materials are shown.The DM content of stylo at harvest was around 32% FM, and the CP and WSC contents were approximately 10% and 1% DM, respectively.The quantity of lactic acid bacteria (LAB) was similar to that of coliform bacteria, with both exceeding 4 log10 colony-forming units (cfu) g −1 FM.

Fermentation Quality and Microbial Population
Table 2 shows that NE and storage days significantly affected DM, DM loss, pH, and counts of LAB and coliform (p < 0.05).NE caused a significant decrease (p < 0.05) in pH levels and coliform counts on each day of storage.Storage days resulted in significantly increased DM loss and decreased coliform counts in silage at each treatment (p < 0.05).There were no molds or yeasts found in the silage during the ensiling process.The LAB count was lower in the treatment group after 30 days of ensiling compared to that after 3 days of ensiling.NE had no significant influence (p > 0.05) on DM content after 30 days of silage.Note: Significant differences in the same column followed by different lowercase letters (p < 0.05); significant differences in the same row followed by different capital letters (p < 0.05); T: treatments; D: storage days; SEM: standard error of the mean; the same as below.

Organic Acids and Nutritional Quality
The effects of NE on organic acids and nutritional quality are shown in Table 3.In addition to crude protein and true protein content, significant effects of NE and storage days on the content of LA, AA, AN, NPN, NDF, and ADF were observed in stylo silage.NE significantly boosted (p < 0.05) LA and AA contents and dramatically reduced (p < 0.05) the AN content of the silage with each storage day.Storage days dramatically boosted (p < 0.05) the AN content of the silage with each treatment.After 30 days of silage, the addition of 2% NE tended to increase the TP content and decrease the NDF and ADF contents compared to the control.In addition, no PA or BA content was detected.

Microbial Community of Stylo Silage during Ensiling
Table 4 displays the bacterial diversity found in stylo silage.All the treatments had good coverage values exceeding 0.995.In comparison to the CK group, the NE treatment group exhibited higher Chao1 and Ace indices at various storage days, while the Shannon and Simpson indices were comparatively lower.Figure 1 illustrates the differences in the bacterial populations under the various treatments.With longer ensiling days, the variances in the bacterial community between the three treatments became apparent.The NE-treated samples were clearly separated from the CK samples.Principle components 1 (PC1) and 2 (PC2) were partially responsible for 66.42% and 17.94% of the variation.Figure 2 illustrates the distribution of bacterial communities in the stylo silage, both at the phylum and genus levels.The bacterial communities had similar phylum-level components, including mainly Firmicutes, Proteobacteria, Cyanobacteria, and Bacteroidetes.Among all groups, Firmicutes were the most abundant bacteria (47.46-61.28%),and their population increased when treated with NE.Proteobacteria were the second most abundant (24.77-38.46%),but decreased when treated with NE.As the ensiling days were progressed, there was a decrease in the overall level of Cyanobacteria.Lactobacillus (4.51-36.73%),Weissella (6.68-27.05%),Enterobacter (7.59-18.52%),Pediococcus (8.05-22.77%),and Lactococcus (3.13-13.04%)were the five most dominant bacteria in this study.Enterobacter and Weissella were the two predominant bacterial genera in the CK group at D3, D7, and Figure 2 illustrates the distribution of bacterial communities in the stylo silage, both at the phylum and genus levels.The bacterial communities had similar phylum-level components, including mainly Firmicutes, Proteobacteria, Cyanobacteria, and Bacteroidetes.Among all groups, Firmicutes were the most abundant bacteria (47.46-61.28%),and their population increased when treated with NE.Proteobacteria were the second most abundant (24.77-38.46%),but decreased when treated with NE.As the ensiling days were progressed, there was a decrease in the overall level of Cyanobacteria.Lactobacillus (4.51-36.73%),Weissella (6.68-27.05%),Enterobacter (7.59-18.52%),Pediococcus (8.05-22.77%),and Lactococcus (3.13-13.04%)were the five most dominant bacteria in this study.Enterobacter and Weissella were the two predominant bacterial genera in the CK group at D3, D7, and D14 of silage.After 30 days of silage, Enterobacter (18.52%) and Lactobacillus (19.49%) were the two predominant bacterial genera in the CK group.One of the most dominant bacterial genera in the NE group throughout the first 7 days of ensiling was Weissella.Beginning on the 14th day of silage, Pediococcus and Lactobacillus became the dominant bacterial genera in the NE group.Figure 3 clearly demonstrates a decrease in the proportion of Weissella and Lactococcus, and an increase in Lactobacillus, Enterobacter, and Pediococcus at D30 compared to D3.Throughout the fermentation process, Pediococcus and Lactobacillus were more abundant in the NE group compared to the CK group, while Enterobacter and Lactococcus were less abundant in the NE group than in the CK group.Meanwhile, starting from day 7, the presence of NE resulted in a decline in Weissella abundance.As the number of days increased, Lactobacillus showed an increasing trend, while Lactococcus and Weissella showed a decreasing trend.Throughout the fermentation process, the NE treatment had a lower abundance of Enterobacter and a higher abundance of Pediococcus than the CK treatment.
D14 of silage.After 30 days of silage, Enterobacter (18.52%) and Lactobacillus (19.49%) were the two predominant bacterial genera in the CK group.One of the most dominant bacterial genera in the NE group throughout the first 7 days of ensiling was Weissella.Beginning on the 14th day of silage, Pediococcus and Lactobacillus became the dominant bacterial genera in the NE group.Figure 3 clearly demonstrates a decrease in the proportion of Weissella and Lactococcus, and an increase in Lactobacillus, Enterobacter, and Pediococcus at D30 compared to D3.Throughout the fermentation process, Pediococcus and Lactobacillus were more abundant in the NE group compared to the CK group, while Enterobacter and Lactococcus were less abundant in the NE group than in the CK group.Meanwhile, starting from day 7, the presence of NE resulted in a decline in Weissella abundance.As the number of days increased, Lactobacillus showed an increasing trend, while Lactococcus and Weissella showed a decreasing trend.Throughout the fermentation process, the NE treatment had a lower abundance of Enterobacter and a higher abundance of Pediococcus than the CK treatment.A correlation heatmap displaying the relationship between the main bacterial genera and silage characteristics is presented in Figure 4. Enterobacter, Weissella, Pantoea, Raoultella, and Lactococcus all showed a substantial negative connection (p < 0.001) with pH, but Pediococcus and Lactobacillus showed a positive association (p < 0.001) with pH.Enterobacter and AN had a considerably positive correlation (p < 0.001), while Weissella and Pseudomonas showed a different significantly negative correlation with AN, respectively.LA had a negative connection (p < 0.001) with Enterobacter, Weissella, Pantoea, Raoultella, and Lactococcus, but a substantial positive correlation (p < 0.001) with Pediococcus and Lactobacillus.AA had a different significantly negative correlation with Lactococcus, Raoultella, and Enterobacter, but a positive correlation (p < 0.05) with Lactobacillus.A correlation heatmap displaying the relationship between the main bacterial genera and silage characteristics is presented in Figure 4. Enterobacter, Weissella, Pantoea, Raoultella, and Lactococcus all showed a substantial negative connection (p < 0.001) with pH, but Pediococcus and Lactobacillus showed a positive association (p < 0.001) with pH.Enterobacter and AN had a considerably positive correlation (p < 0.001), while Weissella and Pseudomonas showed a different significantly negative correlation with AN, respectively.LA had a negative connection (p < 0.001) with Enterobacter, Weissella, Pantoea, Raoultella, and Lactococcus, but a substantial positive correlation (p < 0.001) with Pediococcus and Lactobacillus.AA had a different significantly negative correlation with Lactococcus, Raoultella, and Enterobacter, but a positive correlation (p < 0.05) with Lactobacillus.

Characteristics of Raw Stylo Materials
The CP content of stylo in the current study was similar to the findings published by Liu and colleagues [34], but less than those of Da et al. [35].The contents of NDF and ADF

Characteristics of Raw Stylo Materials
The CP content of stylo in the current study was similar to the findings published by Liu and colleagues [34], but less than those of Da et al. [35].The contents of NDF and ADF were higher than the results reported by Bureenok et al. [36].This variation could be due to planting conditions, location, and climate.The DM content of stylo was 32.23%, which is in line with the dry matter content requirement for good silage [37].The ideal DM content promotes lactic acid bacteria growth and inhibits harmful microorganisms, such as Clostridium, from fermenting [38].In addition, the main parameters affecting the quality of silage feed fermentation are WSC content and the quantity of lactic acid bacteria.In general, well-preserved silage feed requires lactic acid bacteria ≥ 5 lg cfu•g −1 FM and a WSC content over 5% DM in raw materials [39,40].The DM content of the raw materials for stylo was relatively optimal for the ensiling process.However, the lactic acid bacteria count was below the standard for quality silage, with an abundance of harmful microorganisms like Escherichia coli, and low WSC content.These factors might lead to an inadequate fermentation of stylo silage, necessitating the use of silage additives.

Microbial Population and Quality of Fermentation in Stylo Silage
Dry matter losses occurred during each of the stages of the ensiling process [41].This primarily occurred due to plant respiration during the initial stages of silage and microbial metabolism during the middle and later stages [42].Consequently, it was evident that as the duration of silage increased, the rate of dry matter loss also increased in all treated groups.After 30 days of ensiling, adding NE resulted in a reduction in DM loss and was beneficial for feed preservation in comparison with the CK group.Throughout the ensiling time, groups treated with NE had lower pH values than the CK group, with NE2 causing the greatest pH reduction of all the treatments.According to the findings by Ni et al. [43], pH decline mostly happened within the first 7 to 14 days of ensiling, and no additional substantial pH decrease was seen with extended ensiling time.A pH of less than 4.2 is frequently used to determine if silage is well fermented.In this experiment, the pH of the direct silage of stylo was above 5.0, while the pH of the NE treatment groups was below 5.0 but above 4.2.This could be because lowering the pH value is challenging due to the large buffering capacity [5].However, the pH of the NE group was within the acceptable pH range (4.3-5.0) for producing leguminous forage silage [44].The number of lactic acid bacteria almost reached a maximum on the 7th day of silage.The low pH environment may prevent some acid-tolerant lactic acid bacteria from growing, while the absence of fermentable substrate may prevent lactic acid bacteria from being active [45,46].Following 30 days of ensiling, the inclusion of NE notably enhanced the quantity of lactic acid microbes and facilitated the buildup of lactic acid and acetic acid, which also corresponded to a relatively low pH.Coliform bacteria declined as the duration of ensiling progressed, which was in line with Zou et al.'s [15] findings.In addition, there was a reduction in the number of coliforms with the addition of NE.This might be due to the bacteriostatic effect of NE on coliform [24].In this experiment, the counts of both molds and yeasts were lower than 2 lg cfu•g −1 FM.This may be due to the vacuum-sealed plastic bag creating a good vacuum environment, making the anaerobic conditions in the fermentation environment more adequate, and thus inhibiting the growth of molds and yeasts.Alternatively, it could be because they could not compete with other bacteria in the fermentation environment.

Organic Acids and Nitrogen Fractions of Stylo Silage
Organic acids in silage are produced mainly by the metabolic activity of microorganisms.Lactic acid bacteria use WSCs to manufacture lactic acid, but some heterofermentative lactic acid bacteria use them to produce acetic acid [47].In the present study, LA accumulated with increasing days, indicating that lactic acid bacteria dominated the fermentation.Following 30 days of silage, compared to the CK group, LA and AA contents were high in the NE-treated group.This suggests that the incorporation of NE had a positive impact on the metabolic activities of lactic acid bacteria.Notably, no PA or BA levels were identified.This result was also found in the study by Zou et al. [15].It is possible that the activities of Clostridium and Propionibacterium were suppressed in the silage environment of this experiment, and therefore PA and BA were not detected.Clostridium metabolism produces butyric acid, which causes DM loss and reduced protein value, while propionic acid production by propionic acid bacteria increases aerobic stability [48].Therefore, it is beneficial for silage samples to be without butyric acid.Although the absence of propionic acid may result in reduced aerobic stability, the presence of acetic acid may compensate for this [44].
It is well known that plant protein hydrolysis in silage is unavoidable.Proteases and microbes hydrolyze proteins to produce free amino acids, peptides, amides, amines, and ammonia [49].The CP and TP contents of the treatments tended to decrease with increasing fermenting days.However, the NE treatment group effectively increased the TP content at D30 of the silage.There was a relatively low efficiency of non-protein nitrogen utilization by ruminants compared to the true protein utilization efficiency [50].The addition of NE tended to reduce the NPN content, but not significantly.Limiting the hydrolysis of proteins could improve the fermentation quality to a certain extent.The ratio of ammonium nitrogen to total nitrogen reflected the degree of protein hydrolysis in the silage [51].The ratio of ammonia-N to total nitrogen increased with prolonged ensilage time, but it was consistently lower in the NE group than in the CK group.Following 30 days of fermentation, the CK group had an ammonium nitrogen content of 8.27% TN, which exceeded the criterion for quality silage material with an ammoniacal nitrogen content below 8% TN [52].With the addition of NE, the ammonia-N content was lower than 5% TN in all cases.Furthermore, the NE2 group's true protein content showed a significant increase.This suggests that adding NE helped preserve stylo silage proteins to some extent.This might be attributed to the high concentration of polyphenols, such as tannins, in NE [53].Tannins affect the deamidation of free amino acids, reduce the conversion of free amino acids, and inhibit the production of ammonia-N, which together restrict proteolysis in ensiled forage [54].It is worth noting that the addition of NE resulted in a reduction in NDF and ADF content, suggesting that NE might help improve silage digestibility [55].

Bacterial Diversity of Stylo Silage
The composition and structure of the silage bacterial community have a major impact on the quality of silage fermentation.Understanding the silage fermentation process and the alterations in the bacterial community can be achieved through the analysis of the bacterial community.Nowadays, 16S rDNA sequencing technology is widely used for investigating bacterial communities, including community diversity and microbiological identification.Good's coverage values in this study reached over 99%, suggesting that the majority of microorganisms were sufficiently caught through sequencing.The result was a good representation of the bacterial community.In stylo silage, NE application resulted in an increase in Chao1 and Ace, but a decrease in Shannon and Simpson.This suggests a decline in diversity but an increase in the richness of the bacterial community.Ogunade et al. [56] discovered that microbial communities are less diverse when dominant bacteria are more abundant.It could be inferred that NE increases the richness of dominant bacterial genera and hinders the growth of harmful bacteria.The alterations in bacterial communities were additionally shown by the β-diversity analysis.In stylo silage, samples from the CK group appeared to be isolated from those from the NE-treated groups, indicating that NE had a notable impact on the bacterial community.The addition of NE increased the samples' discreteness, and there was some sample overlap between the two groups that received NE treatment.Similar outcomes were likewise obtained with the addition of GA in the study conducted by He et al. [50].On the other hand, the samples from the CK group at various times were tightly grouped within a small range.It is proposed that the addition of NE during ensiling is thought to influence the successional process of the bacterial community in stylo silage.

Bacterial Abundance in Stylo Silage
Proteobacteria, Firmicutes, and Cyanobacteria were the three phyla that predominated in all of the stylo samples in this investigation, which is in line with other research on stylo silage [15].During the fermentation of silage, Firmicutes and Proteobacteria were the most prevalent phyla [57].After ensiling for 30 days, the relative abundance of Firmicutes rose.According to the report by Keshri et al. [58], in settings of anaerobic and acidic fermentation, Firmicutes might grow and reproduce more readily.Adding NE reduced the abundance of Proteobacteria, which can utilize lactic acid to cause nutrition loss [59].The use of NE may therefore be beneficial in maintaining silage nutrition.Similar findings were made in the current investigation as well as by He et al. [55], who discovered that extended fermentation decreased the relative abundance of Cyanobacteria.
There was found to be a distinction in the bacterial communities during ensiling between the control and NE-treated groups, according to the further analysis of the bacterial population at the genus level.In the early stages of fermentation under anaerobic conditions, lactic acid bacteria, such as Lactobacillus, Pediococcus, Weissella, and Lactococcus in the silage, converted WSCs to lactic acid and inhibited enterobacteria, clostridia, and other microorganisms by lowering the pH.This reduced proteolysis and fermentation DM losses [47].Lactococcus and Weissella were common lactic acid bacteria abundant during the early stages of fermentation.However, when the pH of the silage dropped, the acid-tolerant Lactobacillus progressively took over as the dominant genus [39].In this investigation, the abundance of Lactococcus and Weissella reduced with fermentation, whereas Lactobacillus eventually became the dominant genus.This demonstrated that extremely acid-tolerant, homofermentative lactic acid bacteria promote the later phases of fermentation.Lactobacillus abundance increased significantly with the addition of NE, probably because NE is rich in polyphenols such as phenolic acids and flavonoids, which promote lactic acid bacterial growth and lactic acid concentration [60,61].It is worth mentioning that Pediococcus was more prevalent after NE was added.This may be an indication that the pH may continue to decrease.Enterobacter are undesirable microbes in silage because they compete with lactic acid bacteria for fermentation substrate to produce ammonia-N, causing spoilage and lowering nutritional quality [62].The Enterobacter abundance in the CK group increased, while that in the NE group was much lower than that in the CK group.This suggests that NE could successfully stop Enterobacter from growing and proliferating.Additionally, the abundance of Enterobacter decreased as the amount of NE increased.The NE group's decreased ammonia-N content was partially explained by the decline in Enterobacter abundance.Pantoea is also undesirable in silage, and is commonly found in stylo silage [3].Pantoea, however, has reportedly been shown to lower the pH and ammonia-N concentration of silage [56].More research is needed to determine Pantoea's role in stylo silage, despite the fact that its relative abundance in the experiment was quite low.In theory, Methylobacterium would be inhibited in prolonged silage due to anaerobic and low pH environments [63].Methylobacterium was found in every sample, which could be a result of the stylo silage's comparatively high pH (>4.5).Wang et al. [2] also reported finding Kosakonia in stylo silage, which was classified from the genus Enterobacter.Furthermore, this was found to have a positive impact on the distribution of nitrogen by converting molecular nitrogen to ammonia-N, and mostly making protein after that [64].Due to the potential for biogenic amine production, Pseudomonas may be undesirable in silage [65].The good news is that Pseudomonas is not abundant in stylo silage.

Analysis of the Relationship between the Microbial Community Structure and the Quality of Fermentation in Stylo Silage
The fermentation of silage is a complex biological process heavily influenced by the activity of lactic acid bacteria, which are essential in promoting the production of lactic acid and helping to lower the pH levels of silage [66].Just like the study mentioned earlier, in this research, we observed an inverse relationship between pH levels and the population of lactic acid bacteria, particularly Lactobacillus and Pediococcus.These two organisms are homologous fermentation lactic acid bacteria that excel in fermenting and producing acids, as well as resisting acidic environments [67,68].A strong positive correlation was observed between pH levels and the presence of Weissella and Lactococcus in this study.Weissella is a weak acid-producing heterologous fermenter, while Lactococcus is less acid-tolerant than Lactobacillus.Both undergo growth inhibition late in fermentation [58,69].Enterobacter was positively correlated with AN content.This is due to the fact that, in silage, Enterobacter is in competition with lactic acid bacteria for the production of ammoniacal nitrogen [62].Of particular interest was the inverse correlation between the quantities of Weissella and Pseudomonas and AN content.Pseudomonas is one of the undesirable genera in silage, and this study showed the ability to reduce AN content.Consequently, further exploration into the function of Pseudomonas in the fermentation process of silage is necessary.

Conclusions
This experiment was designed with three treatments, four silage time periods, and three replications to investigate the effects of NE on the dynamics of fermentation quality in stylo.The study demonstrated a decrease in pH value, ammonia-N content, and coliform bacteria count in stylo silage when NE was added.Conversely, there was an increase in lactic acid and acetic acid levels.Additionally, the abundance of Enterobacter decreased while lactate-producing bacteria increased.These findings indicate that NE may serve as an innovative supplement for enhancing the fermentation quality of stylo silage.Specifically, the addition of 2% NE showed superior efficacy in enhancing the silage quality of stylo.

Figure 3 .
Figure 3. Changes in the relative abundance of five dominant bacterial genera.(A-E) represent changes in relative abundance of Lactobacillus, Lactococcus, Weissella, Enterobacter and Pediococcus, respectively.(d) represents storage days.

Figure 3 .
Figure 3. Changes in the relative abundance of five dominant bacterial genera.(A-E) represent changes in relative abundance of Lactobacillus, Lactococcus, Weissella, Enterobacter and Pediococcus, respectively.(d) represents storage days.Fermentation 2024, 10, x FOR PEER REVIEW 9 of 16

Figure 4 .
Figure 4.The correlation between microorganisms and fermentation parameters at the genus level.Corresponding value of middle heat map is Spearman correlation coefficient r, which ranges between 1 and −1; r < 0 indicates a negative correlation (blue) and r > 0 indicates a positive correlation (yellow).* significance at p < 0.05; ** significance at p < 0.01; *** significance at p < 0.001.

Figure 4 .
Figure 4.The correlation between microorganisms and fermentation parameters at the genus level.Corresponding value of middle heat map is Spearman correlation coefficient r, which ranges between 1 and −1; r < 0 indicates a negative correlation (blue) and r > 0 indicates a positive correlation (yellow).* significance at p < 0.05; ** significance at p < 0.01; *** significance at p < 0.001.

Author Contributions:
Conceptualization, Q.Z.; investigation, P.H. and M.C.; software, D.C.; validation, M.Z., W.Z. and X.L.; data curation, P.H. and D.C.; writing-original draft preparation, P.H.; writing-review and editing, P.H., M.C. and M.Z.; visualization, W.Z. and X.L.; supervision, H.H.; project administration, Q.Z.All authors have read and agreed to the published version of the manuscript.Funding: This research was funded by Chimelong Group Research Projects (F110020230509), National Natural Science Foundation of China (32371771) and the Science and Technology Major Project of Inner Mongolia Autonomous Region of China to the State Key Laboratory of Reproductive Regulation and Breeding of Grassland Livestock (2021ZD0048).Institutional Review Board Statement: Not applicable.Informed Consent Statement: Not applicable.

Table 2 .
The effect of NE on the dynamic fermentation characteristics and microbial population of stylo silage.

Table 3 .
The effect of NE on organic acids and nutritional quality of stylo silage.

Table 4 .
Alpha diversity of bacterial community of ensiled stylo.

Table 4 .
Alpha diversity of bacterial community of ensiled stylo.