Introduction
Ensiling is a good conservation method for storing herbages, because it saves nutritional composition, and enhances forage palatability for livestock (Li et al., Reference Li, Ni, Zhang, Lin and Yang2018). During the ensiling, water soluble carbohydrates (WSC) are converted by lactic acid bacteria (LAB) into various organic acids, mainly lactic acid, thereby suppressing the populations and activity of undesirable microorganisms (McDonald et al., Reference McDonald, Henderson and Heron1991). In silage production, legume forage and oat (Avena sativa L.) account for large proportions as raw materials. Red clover (Trifolium pratense L.) and alfalfa (Medicago sativa L.) produce high protein yield, digestibility and palatability, and could fix nitrogen in soil. Besides, they have abundant mineral and vitamin contents, and low seasonal yield (Annicchiarico et al., Reference Annicchiarico, Barrett, Brummer, Julier and Marshall2015). The abovementioned attributes could result in a good milk production. Good palatability and large amounts of fermentable substrates can be found in oat silage. As we known, the fermentative products of grass are different from that in legumes. Previous studies reported that the major fermentative product in legume silage is acetic acid, and oat silage often produces high lactic acid contents (Buxton et al., Reference Buxton, Muck, Harrison, Buxton and O'Kiely2003). It is mainly related to the microbial and chemical compositions in fresh forages. In the aspect of chemical composition, high buffering capacity and low dry matter (DM) and WSC contents are found in legumes, while optimal moisture and abundant fermentable substrates exist in oats (McDonald et al., Reference McDonald, Henderson and Heron1991). In the aspect of microbial factors, the epiphytic microbiota on various herbages differed greatly (Duniere et al., Reference Duniere, Xu, Long, Elekwachi, Wang, Turkington, Forster and McAllister2017). However, there are few studies evaluating the influences of epiphytic microbiota from red clover and alfalfa on silage quality of oat.
In recent years, some researchers studied the impact of exogenous microbiota on fermentative products. The epiphytic microbiota and substrates were separated through the method of γ-ray irradiation sterilization in many forages, and the effects of epiphytic microbiota on ensiling characteristics and bacterial diversity were explored using the method of microbiota transplantation. The influence of epiphytic microbes from tropical forages (napiergrass, forage sorghum, Sudan grass, and whole-crop corn) on fermentation quality of irradiated Sudan grass silage were explored (Nazar et al., Reference Nazar, Wang, Zhao, Dong, Li, Ali Kaka and Shao2021), and the results showed that the epiphytic microbes from forage sorghum were effective in improving the fermentative products and bacterial community of Sudan grass silage. Moreover, the effects of epiphytic microbes from oat and Italian ryegrass on the ensiling characteristics and bacterial diversity in legume forage (red clover) silage (Wang et al., Reference Wang, Li, Zhao, Dong and Shao2022), and results showed that inoculating the epiphytic bacteria from oat can obviously increase the lactic acid contents, and decrease pH, acetic acid and ammonia nitrogen contents of red clover silage after 60 days of ensiling. Whereas, the impact of epiphytic microbes from red clover and alfalfa on oat silage was unknown.
In temperate areas, oat is a kind of principle grass in silage production. Red clover and alfalfa are also the typical legume forages. It is hypothesized that transplanting the epiphytic microbiota from alfalfa and red clover into oat silage can reconstitute a bacterial community with similar functions as found in red clover and alfalfa silages. This study aimed to evaluate the influences of epiphytic microbiota from red clover and alfalfa on fermentative products, bacterial community compositions and their metabolic pathways on oat silages.
Materials and methods
Preparing inoculum and making silage
Alfalfa (AL, Medicago sativa L. cv. Sanditi), oat (OT, Avena sativa L. cv. Magnum) and red clover (RC, Trifolium pratense L. cv. Badong) were planted (seeded on 25th October 2018 and harvested on 15th May 2019) in Nanjing (119°10′ E, 31°36′ N). The average elevation, annual temperature and precipitation were 43.1 m, 16.0°C, and 1099 mm, respectively. The soil type is Lixisols, based on the classification of the World Reference Base for Soil Resources. The pH of the test soil (10~15 cm layer) was 8.0, and the available potassium, available phosphorus and organic matter contents were 90.1, 9.4 and 7.9 g/kg, respectively. No fertilizers were used before seeding or during planting, and weeds were removed every two weeks. Red clover and alfalfa were both harvested at early-blooming stage. Oat was harvested at heading stage, similar to the reports of Li et al. (Reference Li, Tang, Liao, Li, Chen, Lu, Huang, Chen and Gou2021) and Tahir et al. (Reference Tahir, Li, Xin, Wang, Chen, Zhong, Zhang, Liu, He, Wen and Yan2023). The collected fresh materials were chopped into 2~3 cm by a cutter, and prepared for silage making.
The inoculum of the epiphytic microbiota on forages was prepared firstly. Due to the 10% loss in making microbiota inoculum, the epiphytic inoculum on fresh material (400 g) was collected from more fresh material (444 g) using a horizontal shaker (1 h, 120 rpm, 20°C) and Ringer solution with Tween-80 (0.5 ml/l). After centrifuging, the residues were collected as inoculum and preserved in a fridge (−20°C). After cutting fresh materials, the 400 g-fresh OT was loaded into the plastic vacuum bag of specific size (45 × 32 cm), followed by compaction (packing density: 94.4 g of DM per bag) to remove air with a vacuum sealer. In total, 72 experimental silos (4 inoculum × 6 storage periods × 3 replicates) were prepared. The sealed silos were taken to a company (Xiyue Irradiation Technology) for γ-ray irradiation (30 kGy for 10 min with 60Co). Then, the sterilized OT was added by the sterilized deionized-water (STOT), and the inoculum from epiphytic microbes on RC (OTRC), OT (OTOT), and AL (OTAL), respectively. After inoculation, 72 silos were preserved at room temperature (24~27°C). Triplicate silos of each treatment were opened for analysis after different ensiling days (1, 3, 7, 15, 30 and 60).
Chemical composition and microbial population analysis
At sampling, the wet silage samples were mixed firstly. Then, a part of subsample (25.0 g) was mixed with 100 ml deionized water, and filtered by sterile gauze and filter paper (pore size: 20–30 μm). The obtained filtrates were used for further analyses. The buffering capacity of raw materials was tested (Playne and McDonald, Reference Playne and McDonald1966). The pH of silage was determined by the electrode pH meter. The NH3-N of silage was determined using the method of phenol-hypochlorite as reported by Broderick and Kang (Reference Broderick and Kang1980). The lactic, acetic, propionic and butyric acid and ethanol concentrations were tested using an Agilent HPLC 1260 system with a column (Carbomix H-NP5) and a detector (Refractive Index Detector) and the eluent (H2SO4 2.5 mmol/l, speed: 0.5 ml/min, Temperature: 55°C). A subsample of silage or fresh material samples (200 g) were used to test the dry matter (DM) content in the oven (48 h at 60°C). After grinding, the water soluble carbohydrate (WSC), total nitrogen (TN), and fibre compositions were determined by the methods of Thomas (Reference Thomas1977), Krishnamoorthy et al. (Reference Krishnamoorthy, Muscato, Sniffen and Van Soest1982), and Khan et al. (Reference Khan, Tewoldebrhan, Zom, Cone and Hendriks2012), respectively. Another subsample of silage or fresh samples (10.0 g) mixed with sterilized saline (90 ml) was utilized for counting microbial populations, and the residual filtered liquid (about 85 ml) in conical flask was collected and stored at −80°C and prepared for DNA extraction and sequencing analyses. The LAB colonies were cultured on MRS medium in the anaerobic incubator at 30°C for 48 h. Yeasts, aerobic bacteria, and Enterobacteriaceae were cultured on potato dextrose agar, nutrient agar, and Violet Red Bile Glucose agar respectively at 37°C for 24 h (Enterobacteroaceae) or 72 h (yeasts and aerobic bacteria) under aerobic condition.
Bacterial diversity analyses
The raw materials (ALFM, OTFM, RCFM) and silage on day 3 (OTOT-3, OTAL-3, OTRC-3), and 60 (OTOT-60, OTAL-60, OTRC-60) were chosen to describe the bacterial compositions and their functions through the NGS (Next Generation Sequencing) technology. The total genomic DNAs in samples were extracted by the DNA extraction kit (MP Biomedicals, Santa Ana, CA, USA) according to the manufacturer's protocol (Wang et al., Reference Wang, Sun, Zhao, Dong, Li, Nazar and Shao2020a). The spectrophotometry (optical density: 260/280 nm ratio) and agarose gel electrophoresis (1%) were used to check the quality and quantity of extracted DNA. The 338F and 806R primers were used to amplify the V3–V4 regions of bacterial 16S ribosomal RNA. The AxyPrep DNA Gel Extraction Kit (Axygen Biosciences, Union City, CA, USA) was used to purify the PCR products, and QuantiFluor™-ST (Promega, USA) was applied to quantified based on the manufacturer's protocol. The DNA paired-end sequencing was conducted on the platform of Illumina MiSeq PE300. The bacterial compositions were analysed on levels of Genus and Phylum under Silva 138 (>75% of confidence). The Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway classification and Tax4Fun tool were used to predict the bacterial functions (Kathrin et al., Reference Kathrin, Bernd, Rolf and Peter2015). The accession number was PRJNA993592 after uploading the raw sequencing data into the SRA (Sequence Read Archive) of NCBI.
Data analyses
The microbial populations were evaluated as colony-forming units (cfu)/g based on the fresh weight (FW) and changed to log10 cfu/g FW. The Statistical Packages for the Social Sciences (SPSS) software was applied to reflect the differences of data. The comparisons between sterilized OT (n = 3) and fresh OT (n = 3) were analysed using one-way analysis of variance (ANOVA):
where Yi is the dependent variable; μ is the least square mean; Gi is the impact of gamma-ray irradiation (i = 2, without vs. with); and eij is the residual error term.
Effects of storage period and microbiota treatments were analysed by two-way analysis of variance, with Tukey's multiple comparison. The data of fermentation quality was analysed using the general linear model procedure based on the model for a factorial treatment design as following:
where Yij is the dependent variable; μ is overall mean; Di is the impact of storage period (i = 6, storage period after 1, 3, 7, 15, 30 and 60 days); Tj is the impact of microbiota (j = 3, epiphytic microbiota from oat, alfalfa, and red clover); (D × T)ij is the impact of interaction between storage period and epiphytic microbiota; and eijk is the residual error term.
Data regarding the proportions of KEGG pathways was examined using one-way ANOVA. The data difference was examined by the Tukey's multiple comparisons. Difference was regarding as significant at P < 0.05.
Results
In the current study, the sterile OT had similar (P > 0.05) pH, contents of buffering capacity, DM, WSC, crude protein, neutral and acid detergent fibre with fresh OT. The epiphytic microorganisms (culturable) were not detected in sterile OT (Table 1).
Chemical and microbial compositions of fresh and sterile oat before ensiling
DM, dry matter; FW, fresh weight; mEq, milligram equivalent; cfu, colony-forming units; ND, not detected; s.e.m., standard error of means.
As shown in Fig. 1(a), all the rarefaction curves presented a trend of first rising and then stabilizing. Higher sequencing depth (>28 000) just produced a few OTUs. Moreover, the Shannon curve levels (Fig. 1(b)) kept stable at the initial stage.
Rarefaction curves (a), Shannon curves (b), phylum (c) and genus (d) level compositions of the bacterial community in fresh materials and oat silages. OTFM, fresh material of oat; ALFM, fresh material of alfalfa; RCFM, fresh material of red clover; OTOT, sterile oat inoculated by epiphytic bacteria from oat; OTAL, sterile oat inoculated by epiphytic bacteria from alfalfa; OTRC, sterile oat inoculated by epiphytic bacteria from red clover; 3, 3 days of ensiling; 60, 60 days of ensiling.
The Proteobacteria and Firmicutes were predominant in fresh OT (49.7%; 48.3%) and fresh AL (56.2%; 41.8%) groups, while the predominant phylum in fresh RC was only Proteobacteria (91.7%) (Fig. 1(c)). After ensiling, the proportions of Firmicutes were enhanced in all inoculated silages.
High ratios of Pseudomonas were observed in ALFM (14.5%), RCFM (11.6%) and OTFM (4.29%) (Fig. 1(d)). RCFM had a high proportion (30.0%) of Methylobacterium and a high ratio (13.4%) of Sphingomonas. After 3 days of ensiling, OTOT (31.5%) and OTAL (40.5%) had higher proportions of Lactobacillus than OTRC (3.44%). A higher ratio of Hafnia-Obesumbacterium was found in OTRC (54.5%) than OTOT (9.50%) and OTAL (5.88%) on day 3. The proportions of Lactococcus were similar among the OTOT (18.7%), OTAL (13.4%) and OTRC (20.6%) after 3 days of ensiling, although RCFM had a low abundance (0.02%) of Lactococcus. OTAL (20.8%) had higher abundances of Enterobacteriaceae than OTOT (7.40%) after 60 days. After 3 days, OTAL (7.99%) and OTOT (5.89%) had higher abundances of Weissella than OTRC (0.44%).
The STOT group remained unfermented during the storage period (Table 2). High concentration of lactic acid was rapidly accumulated, and pH declined (P < 0.05) in fermented groups at early stage. The lactic acid concentrations declined (P < 0.05) in all fermented groups between day 30 and day 60. Higher (P < 0.05) contents of lactic acid were found in OTAL (36.5 g/kg DM) than OTOT (21.6 g/kg DM) and OTRC (18.6 g/kg DM) after 3 days. On day 60, OTOT had higher (P < 0.05) lactic acid contents and lower pH than OTAL and OTRC. The acetic acid concentrations in fermented groups gradually accumulated and were higher than 41.5 g/kg DM after 60 days. OTAL and OTOT had lower (P < 0.05) acetic acid concentrations than OTRC after 60 days. The ethanol concentrations were less than 15.0 g/kg DM in fermented groups. STOT had stable (P > 0.05) dry matter contents during the entire ensiling process (Table 3). The dry matter contents in fermented groups declined (P < 0.05) during ensiling. The NH3-N contents in the STOT group gradually increased (from 2.70 to 14.3 g/kg TN) during fermentation (Table 3). Low NH3-N contents (<100 g/kg TN) were observed in all fermented groups. On day 60, lower (P < 0.05) NH3-N concentrations were found in OTAL in comparison to OTOT and OTRC.
Effects of exogenous microbiota on pH, organic acid and ethanol contents in oat silages
DM, dry matter; STOT, sterile oat; OTOT, sterile oat inoculated by epiphytic microbiota from oat; OTAL, sterile oat inoculated by epiphytic microbiota from alfalfa; OTRC, sterile oat inoculated by epiphytic microbiota from red clover; s.e.m., standard error of means; T, microbiota; D, ensiling days; T × D, the interaction between microbiota and ensiling days.
A−F Means in the same row with different capital letter differed (P < 0.05).
a−d Means in the same column with different lowercase differed (P < 0.05).
Effects of exogenous microbiota on chemical compositions in oat silages
DM, dry matter; FW, fresh weight; TN, total nitrogen; STOT, sterile oat; OTOT, sterile oat inoculated by epiphytic microbiota from oat; OTAL, sterile oat inoculated by epiphytic microbiota from alfalfa; OTRC, sterile oat inoculated by epiphytic microbiota from red clover; s.e.m., standard error of means; T, microbiota; D, ensiling days; T × D, the interaction between microbiota and ensiling days.
A−F Means in the same row with different capital letter differed (P < 0.05).
a−d Means in the same column with different lowercase differed (P < 0.05).
Before ensiling, the various pathway levels were observed in raw materials (Fig. 2). After 3 days, OTAL and OTOT had higher (P < 0.05) proportions of ‘Cellular Process’ and ‘Metabolism’, while lower proportions of ‘Genetic Information Processing’ than OTRC. On day 60, a low difference of metabolic pathways on the first level was observed among all treatments.
Changes of KEGG metabolic pathways on the first level obtained with Tax4Fun in different groups. KEGG, Kyoto Encyclopedia of Genes and Genomes; OTOT, sterile oat inoculated by epiphytic bacteria from oat; OTAL, sterile oat inoculated by epiphytic bacteria from alfalfa; OTRC, sterile oat inoculated by epiphytic bacteria from red clover. *, 0.01 < P < 0.05; **, 0.001 < P < 0.01; ***, P < 0.001.
Discussion
Chemical compositions and microbial populations of fresh and sterile oat
The WSC contents of raw material should be higher than 50 g/kg DM in quality silage (Wang et al., Reference Wang, Li, Zhao, Dong and Shao2022). Herein, the high WSC contents of fresh OT indicated fresh OT could be utilized to assess the effects of epiphytic microbiota on fermentative products, because high WSC contents were adequate for the microbial growth during ensiling. Moreover, sterile OT had similar chemical components with fresh OT, suggesting that the selected irradiation conditions were suitable because the chemical compositions of sterile OT were not changed by irradiation. More importantly, we did not observe any microorganism in different kinds of cultured medium in sterile OT group, reflecting the epiphytic microbes of herbages were killed by the γ-ray irradiation.
Alpha diversity
The rarefaction curves presented a trend of first rising and then stabilizing. It was suggested that the amount of sequencing was optimal, and can show the profile and diversity of bacterial compositions in samples (Wang et al., Reference Wang, Zhao, Dong, Li and Shao2020b). Higher sequencing depth (>28 000) just produced a few OTUs that had no impact on assessing the bacterial community of samples. Moreover, the stable Shannon curves at the early stage suggested the depth of sequencing was adequate to show the bacterial compositions.
Bacterial community compositions
The Proteobacteria and Firmicutes were predominant in fresh OT and AL groups. Proteobacteria was critical in accelerating nitrogen and carbon cycles, and degrading organic matter (Ma et al., Reference Ma, Fang, Sun, Han, He and Huang2018). In anaerobic environment, Firmicutes could use the acid hydrolytic function to produce different kinds of enzymes (Wang et al., Reference Wang, Sun, Zhao, Dong, Li, Nazar and Shao2020a). In contrast, the major phylum in fresh RC was Proteobacteria (91.7%), which may be due to the various growth environments and chemical compositions of herbages (Wang et al., Reference Wang, Zhao, Dong, Li and Shao2020b). After ensiling, the proportions of Firmicutes were enhanced in fermented silages. It was probably because the proliferation of Firmicutes preferred an anaerobic and acidic environment during ensiling (Keshri et al., Reference Keshri, Chen, Pinto, Kroupitski, Weinberg and Sela Saldinger2018).
High ratios of Pseudomonas were observed in ALFM, RCFM and OTFM. Pseudomonas could proliferate extensively under the anaerobic condition, and their existence in silage was undesirable, probably because of producing biogenic amines, leading to the reduction of nutritive value (Duniere et al., Reference Duniere, Sindou, Chaucheyras-Durand, Chevallier and Thévenot-Sergentet2013). RCFM had a high proportion (30.0%) of Methylobacterium, which have been found in a variety of silages (Ogunade et al., Reference Ogunade, Jiang, Pech Cervantes, Kim, Oliveira, Vyas, Weinberg, Jeong and Adesogan2018; Wang et al., Reference Wang, Chen, Wang, He, Zhou and Yang2019). RCFM also had a high ratio (13.4%) of Sphingomonas, which belong to Gram-negative, pathogenic and aerobic bacteria, and can metabolize many xenobiotic compounds (Wang et al., Reference Wang, Li, Zhao, Dong and Shao2022). The difference in the bacterial community compositions on fresh forages could be affected by the environmental factors of forage growth. Previous studies reported that the humidity and rainfall can change the epiphytic bacterial compositions on fresh maize (Guan et al., Reference Guan, Yan, Li, Li, Shuai, Feng and Zhang2018), and the ensiling characteristics were highly related to the activity, populations and species of epiphytic microorganisms, which can be influenced by geographical location, weather and fertilization condition (Wang et al., Reference Wang, Zhao, Dong, Li and Shao2020b).
On day 3, OTOT and OTAL had higher proportions of Lactobacillus than OTRC. Lactobacillus could consume glucose to produce the lactic acids. During fermentation, Lactobacillus can grow and proliferate rapidly, producing lactic acid to lower the pH of silages, and undesirable microorganisms are suppressed (Duniere et al., Reference Duniere, Sindou, Chaucheyras-Durand, Chevallier and Thévenot-Sergentet2013). Hence, the high proportions of Lactobacillus could be responsible for the higher lactic acid production in OTOT-3 and OTAL-3 than OTRC-3. More importantly, a higher ratio of Hafnia-Obesumbacterium was found in OTRC than OTOT and OTAL after 3 days. As enterobacteria, Hafnia-Obesumbacterium could enhance the proteolytic activity in silage (Wang et al., Reference Wang, Sun, Zhao, Dong, Li, Nazar and Shao2020a). It is consistent with our findings that OTRC had higher NH3-N contents than other groups on day 3. Whereas, some commercial LAB inoculants cannot restrict the proliferation of Hafnia-Obesumbacterium (Zhao et al., Reference Zhao, Yang, Wang, Fan and Wang2021). Hence, more experiments related to Hafnia-Obesumbacterium need to be studied.
On day 3, the proportions of Lactococcus were similar among the OTOT, OTAL and OTRC, although RCFM had a low abundance of Lactococcus. It is suggested that the populations of Lactococcus was more related to the chemical components of fermentable substrates during fermentation, and the adequate WSC contents of forages can highly benefit the proliferation of Lactococcus during fermentation, even if their numbers were lower in fresh forages. Furthermore, OTAL had higher ratios of Enterobacteriaceae than OTOT after 60 days. Enterobacteriaceae is undesirable during fermentation, because they could live in an anaerobic and weak acidic environment, compete with lactic acid bacteria for the substrates, and use lactic acid and WSC, leading to the dry matter and nutrient loss (Sun et al., Reference Sun, Jiang, Ling, Na, Xu, Vyas, Adesogan and Xue2021). Thus, the decrease of silage quality in OTAL on day 60 may be due to the higher abundance of Enterobacteriaceae. On day 3, OTAL and OTOT had more abundant Weissella than OTRC. Weissella could convert WSC to lactic and acetic acids, however the acidic environment is undesirable for their proliferation (Graf et al., Reference Graf, Ulrich, Idler and Klocke2016). The high proportions of Weissella in OTAL-3 and OTOT-3 might be related to the rapid accumulation of lactic acid in OTAL and OTOT groups during the early stage of ensiling.
Ensiling characteristics of oat silage
The STOT group was not fermented during the storage period, suggesting the suitable dose of γ-ray irradiation can separate the microbial community and chemical compositions of herbages. A large amount of lactic acid was quickly accumulated and pH declined in the fermented silages at the initial period of fermentation. After chopping, forages could release the plant juice and provide sufficient WSC for LAB. Moreover, lactic acid concentrations declined in all fermented groups between day 30 and 60, which may be correlated with the decreased numbers and activity of LAB and large amounts of undesirable microorganisms evidenced by the high contents of NH3-N and acetic acid in silages. Higher lactic acid contents in OTAL than OTOT and OTRC might be related to the higher abundances of Weissella and Lactobacillus in OTAL after 3 days of ensiling. Weissella can promote the lactic acid accumulation at the initial period of fermentation, and Lactobacillus is often predominant in silage whereby its capacity of rapidly producing lactic acid (Wang et al., Reference Wang, Li, Zhao, Dong and Shao2022). On day 60, OTOT had lower pH and higher lactic acid concentrations compared to OTAL and OTRC, which may be closely related to the high proportion of Lactobacillus in OTOT.
In this study, the acetic acid concentrations in fermented silages gradually accumulated and were higher than 41.5 g/kg DM after 60 days. The increased tendency of acetic acid during ensiling was highly correlated with the activity of hetero-fermentative lactic acid bacteria strains, enterobacteria and Propionibacterium during fermentation (McDonald et al., Reference McDonald, Henderson and Heron1991). OTAL and OTOT had lower acetic acid concentrations than OTRC on day 60. It is probably because the numbers and activity of acetic acid-producing microbes were restricted by the rapid acidification in OTAL and OTOT. Large production of ethanol of silages can result in higher DM and energy losses during fermentation. The proliferation of yeast in silages can lead to a large accumulation of ethanol (30~40 g/kg DM) (Kung et al., Reference Kung, Shaver, Grant and Schmidt2018). Herein, the ethanol concentrations were lower than 15.0 g/kg DM in fermented groups, indicating the ethanol contents were mainly produced by some bacteria (e.g. hetero-fermentative LAB, enterobacteria). Enterobacteriaceae can consume lactic acid and fermentable substrates to produce ethanol (Sun et al., Reference Sun, Jiang, Ling, Na, Xu, Vyas, Adesogan and Xue2021), and hetero-fermentative lactic acid bacteria strains could produce CO2, ethanol, lactic acid and acetic acid (Borreani et al., Reference Borreani, Tabacco, Schmidt, Holmes and Muck2018).
The dry matter contents decreased in the inoculated silages during fermentation, which is related to the microbial metabolism. After 60 days, the higher dry matter content in OTAL than OTRC could be related to the rapid decline of pH in OTAL, limiting the proliferation of undesirable microorganisms and preserving silage nutrients. Moreover, the NH3-N contents in STOT were probably related to the plant enzyme. A small amount of NH3-N (<100 g/kg TN) was found in silages, reflecting good silage quality (McDonald et al., Reference McDonald, Edwards, Greenhalgh and Morgan2002). After 60 days, the lower (P < 0.05) NH3-N contents in OTAL than OTOT and OTRC were mainly due to the large accumulation of lactic acid and decline of pH at the initial period of ensiling in OTAL, suppressing the enzyme activity of plants and microorganisms.
Functions of bacterial community
As a kind of bioinformatics tool, KEGG could be utilized to know the functionality and activity of cells and organisms (Mao and Kanehisa, Reference Mao and Kanehisa2012). The information regarding the functions of bacterial communities in silages are conducive for us to understand the fermentation process. Hence, KEGG was used to assess the impact of epiphytic microorganisms on functional changes in oat silages.
Prior to ensiling, the various pathway levels in raw materials were related to the richness and diversity of epiphytic microbiota. After 3 days, OTAL and OTOT had higher proportions of ‘Metabolism’ and ‘Cellular Process’, and lower proportions of ‘Genetic Information Processing’ than OTRC. Based on the fermentation products in different groups, it was suggested that the epiphytic microbes from alfalfa improved the silage quality of RC. This could be due to the changes of cell characteristics, and limitation of signal transduction and membrane transport of undesirable bacteria. After 60 days of fermentation, low variation of metabolic pathways on the first level is observed among various groups, which might be related to the developed internal environment during the final ensiling period.
Conclusion
The ensiling characteristics of oat were particularly influenced by the compositions and activity of epiphytic microbes. The Enterobacteriaceae and hetero-fermentative lactic acid bacteria were mainly responsible for the higher ethanol contents in oat silage. The effects of epiphytic microbiota on nutritive value and aerobic deterioration of oat silage should be further studied in the future experiments. Our results provide more insights into the role of epiphytic microbiota from forages on silage fermentation products but hard to be applied in the silage production due to the strict experimental conditions.
Author contributions
Siran Wang and Tao Shao conceived and designed the study. Siran Wang, Jie Zhao, Zhihao Dong, and Junfeng Li conducted data gathering. Siran Wang, Niaz Ali Kaka, Mudasir Nazar, and Ziqun Lin performed statistical analyses. Siran Wang wrote the article.
Funding statement
This study was financially supported by National Natural Science Foundation of China (32301500), Joint Fund for Regional Innovation and Development of National Natural Science Foundation of China (U20A2003), National Natural Science Foundation of China (32171690), Research and demonstration of mixed silage technology for forage in high-altitude pastoral areas (XZ202301YD0012C).
Competing interests
We certify that there is no conflict of interest with any financial organization regarding the material discussed in the manuscript.
Ethical standards
This article does not contain any studies with human participants or animals performed by any of the authors.
