Abstract
Effects of cellulase addition were assessed relative to the co-ensiling performance of air-dried corn stover (DCS) and cabbage waste (CW). The DCS and CW mixtures were co-ensiled with 0 to 0.3% of cellulase addition, and changes in composition, intermediates, and biological activity were characterized. The results showed that the addition of cellulase enhanced the decomposition of cellulose and hemicellulose by 2.51 to 6.93% and 3.41%, based on different dosages and compared with the control. Thus, the content of water-soluble carbohydrates increased. The acid content also increased from 5.8% for the control to the range 5.16 to 8.51% for the samples containing cellulase. Moreover, there was a shift from homolactic to heterolactic fermentation with prolonged ensiling time, coupled with the dominant lactic acid bacteria shifting from Paralactobacillus and Lactobacillus to more of Lactobacillus. Thus, the addition of cellulase improved the relative abundance of Lactobacillus. An assessment of fermentation quality, therefore, suggested that cellulase addition can improve the silage quality of DCS/CW during co-ensiling.
Download PDF
Full Article
Enhancing the Co-ensiling Performance of Corn Stover and Cabbage Waste via the Addition of Cellulase
Haiwei Ren,a Li Wang,a Yanan Sun, a Quanlin Zhao,a,b Yongming Sun,b Jinping Li,a and Bingyun Zhang a,*
Effects of cellulase addition were assessed relative to the co-ensiling performance of air-dried corn stover (DCS) and cabbage waste (CW). The DCS and CW mixtures were co-ensiled with 0 to 0.3% of cellulase addition, and changes in composition, intermediates, and biological activity were characterized. The results showed that the addition of cellulase enhanced the decomposition of cellulose and hemicellulose by 2.51 to 6.93% and 3.41%, based on different dosages and compared with the control. Thus, the content of water-soluble carbohydrates increased. The acid content also increased from 5.8% for the control to the range 5.16 to 8.51% for the samples containing cellulase. Moreover, there was a shift from homolactic to heterolactic fermentation with prolonged ensiling time, coupled with the dominant lactic acid bacteria shifting from Paralactobacillus and Lactobacillus to more of Lactobacillus. Thus, the addition of cellulase improved the relative abundance of Lactobacillus. An assessment of fermentation quality, therefore, suggested that cellulase addition can improve the silage quality of DCS/CW during co-ensiling.
Keywords: Cellulase; Ensiling quality; Microbial community; Lactic acid bacteria
Contact information: a: School of Life Science and Engineering, Lanzhou University of Technology / Key Laboratory of Complementary Energy System of Biomass and Solar Energy, Lanzhou 730050, P. R. China; b: Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangzhou 510640, China; * Corresponding author: 494467885@qq.com
INTRODUCTION
Over the past few decades, human activity has led to an increasing amount of agricultural waste, such as crop stalk, fruit, and vegetable waste. In agricultural production, an urgent need has emerged to solve the problem of waste disposal (Hillion et al. 2018). Agricultural residues are regarded as an attractive alternative for bioenergy production due to their wide availability, environmental friendliness, low cost, and renewability, and because they do not compete with food crops (Hess et al. 2009; Vervaeren et al. 2010). However, the drawback of seasonal production and short harvest times negatively affects the large-scale utilization of agricultural waste. Consequently, controlling the long-term preservation and real-time disposition of agricultural waste is indispensable for ensuring year-round availability and sustainable food throughout the year for bioenergy production.
Ensiling is an anaerobic wet storage method and represents a feasible and cost-effective way to supply year-round biomass (Gallegos et al. 2018). During the ensiling process, lactic acid bacteria (LAB) convert water-soluble carbohydrates (WSCs) into organic acids, reducing the pH to a range of 3.5 to 4.2, which inhibits the growth of undesirable microorganisms and preserves the organic matter of biomass (Janke et al. 2019). Satisfactory silage preservation depends on intrinsic biochemical characteristics, such as the moisture, carbohydrate content, and the buffering capacity of biomass and storage conditions, which includes temperature, time, and packing density. Among these, moisture content and WSCs are important for enhancing LAB activity and organic acid generation (Zhao et al. 2016). Studies have shown that ensiling with a moisture content ranging from 65% to 75%, and a rudimentary soluble carbohydrate content higher than 5% (dry matter, DM), is favorable for lactobacillus reproduction (Hillion et al. 2018). However, obtaining good silage quality on an industrial scale may is challenging, as the soluble sugar content in agro-industrial organic waste is insufficient for producing enough lactic acid. Thus, air-dried crop stalk is difficult to immediately ensile due to its low moisture, sugar, and high fiber content (Vervaeren et al. 2010). Although co-ensiling of dried crop stalk with other agro-industrial waste material, such as vegetable waste and sugar beet leaves, may achieve well-preserved silages, the natural cross-linked recalcitrant structure of dried crop stalk may hinder efficient lactic fermentation (Hillion et al. 2018; Ren et al. 2018; Chen et al. 2019).
Various additives, including lactic acid bacteria, enzyme, and molasses, have been used in ensiling to improve silage quality and break down the cell wall components (Li et al. 2014; Muck et al. 2018). Of these, fibrolytic enzymes can enhance LAB activity by promoting the release of sugar from lignocellulosic biomass, which improves ensiling performance at a lower pH and produces more lactic acid (Sun et al. 2011). Previous studies have confirmed that fibrolytic enzymes can increase the concentration of lactic acid and soluble carbohydrates and decrease the content of hemicellulose and cellulose in the silage (Lynch et al. 2015; Desta et al. 2016). Li et al. (2018) also reported that silage with added fibrolytic enzymes maintained lower levels of acid detergent lignin (ADL) throughout the ensiling process of Pennisetum sinese. In addition, fibrolytic enzymes can improve the digestibility of structural carbohydrates, especially neutral detergent fibers (NDF), during the first 24 h of incubation (Colombatto et al. 2004a; Romero et al. 2015). Therefore, before silage production, cellulolytic enzyme pretreatment is considered a beneficial biological pretreatment for lignocellulosic biomass for the production of bioenergy and biological products. This process can decompose the persistent structural matrix, change the composition of lignocellulose, and increase the conversion rate of cellulose (Sun et al. 2011; Desta et al. 2016; Wu et al. 2020). Thus, a better understanding of enzyme addition for altering the microbial communities in ensilage is needed in order to effectively improve silage quality.
Though reports have shown that both ensiling and the use of additives are effective strategies for the preservation and valorization of agricultural waste, and are a beneficial pretreatment method for downstream biorefining, to the best of our knowledge, there is a lack of research on the effects of additives applied during the ensiling of air-dried corn stover and cabbage waste. Therefore, the objectives of this study were to (1) determine the effect of cellulase addition for the co-ensiling performance of corn stover and cabbage waste in terms of fermentation characteristics and chemical composition, and (2) assess the evolution of microbial community diversity during co-ensiling with various cellulase dosages.
EXPERIMENTAL
Materials
Raw material and silage additives
Air-dried corn stover (DCS) was cultivated and harvested in October from local farms in Yuzhong County (latitude 35.34°N, longitude 103.49°E, and altitude 1874.4 m a.s.l., Gansu Province, China). The material was chopped mechanically into lengths of 10 to 20 mm. In this location, the average annual temperature is 7.9 °C, and the hottest and coldest monthly average temperatures are 35.8 °C (July) and −11 °C (January), respectively. In addition, the annual average precipitation is between 300 and 400 mm, the annual average solar radiation is 4800 MJ/m2, and the region receives an annual average of 2562.5 h of sunshine. The cabbage waste (CW) was obtained from the Qilihe vegetable market in Lanzhou (latitude 35.55°N, longitude 103.50˚E, altitude 1800.4 m a.s.L., Gansu Province, China) and cut into 20 mm × 20 mm pieces. The moisture content values of DCS and CW before ensiling were 10.23% and 91.41% on a fresh mass basis, respectively. The commercial cellulase (Ningxia Imperial Jade Biotechnology, Ltd., Yinchuan City, China), which consisted of β-glucanase, α-arazyme, α-galactosidase, β-galactosidase, and β-xylanase (> 10,000 U/g), was supplied as a freeze-dried powder.
Methods
Co-ensiling of DCS and CW
The chopped DCS and CW were mixed thoroughly and divided into a total of 48 equivalent piles (3.23 kg of DCS and 9.87 kg of CW for each pile, wet basis) for experimental treatment according to our previously published methods (Ren et al. 2018). The three blended piles were immediately frozen at −20 °C for further analysis. The remaining 45 piles (5 ensiling time points × 3 dosage treatments × 3 replicates) were randomly assigned into the following treatment groups: (1) control without additives (CWA), (2) cellulase applied at a low dosage of 0.1% (w/w, fresh weight) (CLD), and (3) cellulase applied at a high dosage of 0.3% (w/w, fresh weight) (CHD). The cellulase for each treatment was dissolved in 10 mL of distilled water and uniformly sprayed on each pile of silage. For the control group, the same amount of distilled water was applied. All treatments were adjusted to an approximate DM content of 30%, compacted, and filled and sealed in 30 L cylindrical polyethylene plastic buckets (no headspace). All of the buckets were stored at 18 ± 1 °C and sampled after 30, 60, 90, 130, and 170 days.
Fermentation characteristics of silages
After opening the buckets on the designated ensiling day, a 50 g sample was homogenized with 450 mL of distilled water for 1 min (JJ-2; Guohua Instruments Co., Ltd., Suzhou, China). After filtration through four layers of medical gauze and centrifuged at 3,900 rpm for 10 min, the supernatant was obtained for pH measurements using a glass electrode pH meter (UB-7, DANFO, Beijing, China). Then, a portion of the supernatant was acidified by 50% H2SO4 (w/w) and filtered with a 0.22 μm dialyzer to analyze the organic acids (lactic, acetic, propionic, and butyric acid) with an Agilent 1260 high-performance liquid chromatography (HPLC) system (Agilent Technologies, Inc., Waldbronn, Germany). The system was fitted with a refractive index detector (RSpak KC-811, Shodex, Tokyo, Japan; eluent: 3 mmol/L HClO4, flow rate: 1 mL/min, injection volume: 5 μL, and temperature: 50˚C) (Li et al. 2019). After protein precipitation formed by trichloroacetic acid, ammonia nitrogen (AN) content was determined using the phenol-hypochlorite colorimetric method and expressed per total nitrogen (g/kg TN) (Ren et al. 2021).
Chemical composition analysis
The chemical compositions of the dry matter (DM), total nitrogen (TN), neutral detergent fiber (NDF), acid detergent fiber (ADF), and acid detergent lignin (ADL) were analyzed according to previously reported methods (Ni et al. 2017). Holocellulose (HoC) content was the sum of hemicellulose (HC) and cellulose (CL) content (Robinson et al. 2016). The biodegradation potential of the silage was estimated as the ratio of (hemicellulose + cellulose) to lignin (Vandecasteele et al. 2017). Thus, the carbohydrate content was extracted and used to measure the WSC content using the 3,5-dinitrosalicylic acid (DNS) method with glucose as the standard (Leyva et al. 2008). The TN was equivalent to the Kjeldahl nitrogen, which was determined by a Kjeldahl nitrogen analyzer (K9840, Hanon instruments Co., Ltd., Jinan, China) (Li et al. 2019). All analyses were performed in triplicate. A Flieg evaluation system was used to evaluate the quality of the silage by scoring the silage and by scoring the ratio of the lactic acid, acetic acid, and butyric acid to the total acid content (Agneessens et al. 2015).
Microbial Community Analysis
Metagenomic DNA extraction
Samples from the treatment of CWA and CHD were selected to analyze the dynamic changes in the microbial communities during the ensiling process, using high-throughput sequencing technology. A total of 20 g of samples was mixed with 180 mL of sterilized NaCl solution (0.85%) and then treated with a table concentrator at 120 rpm for 2 h. The suspension was then filtered to obtain the microbial cells for DNA extraction. Total DNA content was extracted using a water DNA extraction kit (Fuji Biotechnology Co., Ltd., Chengdu, China), and DNA quantity and quality were assessed by agarose gel electrophoresis. High-molecular-weight DNA samples (minimal concentration of 20 ng/μL) were used for sequencing.
PCR amplification
Amplification of the V3-V4 hypervariable region of the 16S rRNA was achieved using 338F (5-ACTCCTACGGGAGGCAGCA-3) and 806R (5-GGACTACHVGGGTW TCTAAT-3) primers. Amplification of the targeted region was achieved using the following reaction chemistry: 5 μL of Gotaq Green master mix (Promega, Madison, WI, USA), 11.9 μL of DNase free water, 0.5 μL of MgCl2 (50 mM), 0.5 μL of deoxynucleotide triphosphates (10 mM), 1 μL of DNA forward and reverse primers (10 μM), and 5 μL of the DNA template were adjusted for all samples to a final average concentration of 1 ng/μL of total reaction volume. Reaction conditions for bacterial 16S amplification were as follows: 95 °C for 3 min, followed by 35 cycles of 95 °C for 45 s, 50 °C for 60 s, and 72 °C for 90 s. A final extension was conducted at 72 °C for 10 min (He et al. 2015). After a second examination with gel electrophoresis, the polymerase chain reaction (PCR) products were denatured with NaOH to generate single-strand DNA fragments, which were sequenced using an Illumina MiSeq platform (Personalbio Biotechnology, Ltd., Shanghai, China).
High-throughput sequencing of metagenomics DNA
To obtain the high-quality clean reads, quality filtering of the raw reads was performed under specific filtering conditions. Chimera sequences from the bacteria were removed using the UCHIME algorithm by comparing them with the Silva database. However, chimeral sequences of fungal materials were removed using the UCHIME algorithm by comparing them with the UNITE database. The effective sequences (reads) were counted and operational taxonomic units (OTUs) were clustered at 97%, at similar levels for performing the taxonomic analysis of different samples and obtaining microbial community composition at different levels. Alpha diversity analysis was conducted using QIIME software (http://qiime.org/) to analyze species richness (Chao1, ACE index) and diversity (Shannon and Simpson index). Beta diversities of both weighted and unweighted UniFrac were calculated using QIIME software. Subsequently, species with a relative abundance higher than 0.1% were selected for microbial community composition analysis.
Statistical Analysis
The experiment design consisted of a 3 × 5 factorial design with 3 dosages and 5 ensiling times. The statistical model follows,
Y = μ + αi + βj + (α × β)ij + εij, (1)
where Y represents the response variable, μ is the overall mean, αi is the effect of the ensiling period, βj is the effect of different additive dosages (treatments), (α × β)ij is the interaction effect between the ensiling period and additive dosages, and εij is the residual error (Li et al. 2020). Data were analyzed using the general linear model procedure from the statistical package for social science (SPSS 18.0, SPSS Inc., Chicago, IL, USA), and one-way analysis of variance (ANOVA) was determined using Tukey’s multiple comparison test at a significance level of 0.05.
RESULTS AND DISCUSSION
Comparison of Composition between the Mixtures and Raw Material
The chemical compositions of the DCS and CW are presented in Table 1. The dry matter of the DCS was 89.77 ± 0.01% (Table 1). DCS moisture content (10.23%) was lower than the proper value for ensiling (Adesogan and Newman 2014). Moreover, the WSC content of DCS was 8.05 ± 0.03%, which was lower than the essential WSC content for high-quality silage (8 to 10%). However, dry matter and WSC values in the CW were 8.59 ± 0.15% and 14.57 ± 0.02%, respectively. In general, the production of high-quality silage requires a raw material with 35 to 45% of DM content and 8 to 10% of WSC content (Agneessens et al. 2015). After mixing, the dry matter and WSC content of the mixtures were 29.56 ± 0.03% and 9.53 ± 0.20%, respectively. Compared with the chemical compositions of the raw materials, the mixture had more suitable moisture and WSC content, compared to DCS or CW alone, indicating that the mixtures had optimal compositions for lactic acid fermentation.
The contents of NDF, ADF, and ADL of DCS were 76.85%, 47.15%, and 8.12%, respectively (Table 1). After mixing, an increase of 22.54% in ADL content and decreases of 11.95% and 0.07% in NDF and ADF content were observed for the mixtures, respectively. Optimal cellulase content provides a sufficient source of lactic acid fermentation substrate, thereby promoting the reduction in pH and inhibiting harmful microorganisms (Li et al. 2020). Therefore, the co-ensiling of DCS and CW was proposed in this study as a simple and low-cost strategy for storing both waste products.
Table 1. Chemical Compositions of Air-dried Corn Stover and Cabbage Waste
Effect of Cellulase on Co-ensiling Quality
Changes in lignocellulosic composition during the ensiling process
The effects of cellulase on the mixture composition during the ensiling process are listed in Table 2. With prolonged silage time, the NDF, ADF, and ADL content in the CWA groups ranged from 67.67 ± 0.39% to 74.52 ± 0.15%, 44.01 ± 1.09% to 48.04 ± 0.53%, and 4.65 ± 0.07% to 10.98 ± 0.85%, respectively, during ensiling periods of 30 to 170 days. Compared with the CWA group, CLD and CHD treatments had lower HC and CL content (P < 0.05), with decreases of 13.55%, 26.60%, and 44.33% in NDF, ADF, and ADL content, respectively. Similarly, compared with the CWA group, decreases of 17.31% and 9.1% in CL and HoC content were observed for the CLD and CHD treatment groups, respectively. Thus, the addition of cellulase could effectively promote the degradation of fiber content, and a higher dosage of cellulase could effectively reduce CL, ADF, and ADL contents. Cellulase can also promote fiber degradation, elevating the WSC production by enzymolysis and acid solubilization, which increases the availability of fermentation substrates for LAB to produce lactic acid (Li et al. 2019; Hu et al. 2021). The present results were in good agreement with a previous study by Kholif et al. (2017), where NDF, ADF, and CL contents were significantly decreased after ensiling. Based on the composition changes, the biodegradation potential of the various samples was assessed, and the results are presented in Table 2. The biodegradation potentials of the samples increased after 170 days of long-term co-ensiling. The CHD group met this requirement, but the CLD group did not obtain a similar result.
To further understand the effects of cellulase on DCS and CW co-ensiling, WSC content was quantified, and the results are presented in Fig. 1. WSC content in all three treatments decreased significantly (P < 0.05) during the first 30 days of co-ensiling, with descent percentages of 79.64%, 72.19%, and 60.13% in groups CWA, CLD, and CHD, respectively. This was attributed to the quick growth and proliferation of aerobic microorganisms, with abundant fermentation substrates during the initial stage. With more ensiling time, the CLD and CHD groups exhibited increasing trends in WSC content after 30 and 90 days, respectively. Moreover, the CLD and CHD groups showed significantly higher amounts of WSC than CWA during the storage period of 30 to 170 days, which was consistent with the report that cellulase indirectly provided fermentable sugars from the degradation of NDF, ADF, and HC, and thus required more time (Hess et al. 2009). Weinberg and Chen (2013) suggested that cell-wall degradation enzymes can release fermentable sugars and cause a decrease in fiber content, which was responsible for the observed enhanced digestibility. Furthermore, the WCS content in the CLD group was significantly higher (p < 0.05) than in the CHD group between 60 and 170 days of co-ensiling. This indicated that a low dose of cellulase was beneficial for the preservation of WSC.
Fig. 1. Dynamic effects of cellulase dosage on WSC content: CWA contained no cellulase, CLD contained a low dose of cellulase (0.1%), and CHD contained a high dose of cellulase (0.3%). The different lowercase letters indicate significant differences at p < 0.05 between different groups at the same time, and different capital letters indicate significant differences at p < 0.05 for the same groups at different times.
Table 2. Compositions of Air-dried Corn Stover and Cabbage Waste Mixtures during Co-ensiling (%DM)
CWA: no cellulase addition; CLD: low dosage of cellulase (0.1%); CHD: high dosage of cellulase (0.3%); NDF: neutral detergent fiber; ADF: acid detergent fiber; ADL: acid detergent lignin; HC: hemicellulose; CL: cellulose; and HoC: holocellulose.
Same column with different lowercase letters indicates significant differences at p < 0.05 between different groups at the same time, and the same column with different capital letters indicates significant differences at p < 0.05 for the same groups at different times.
Changes in pH and organic acid during the ensiling process
The pH values of the samples during co-ensiling were measured and are presented in Fig. 2. The results showed that the pH of all three groups decreased with prolonged ensiling time. The pH values of the CWA, CLD, and CHD groups after 30 days were 4.15, 4.08, and 3.91, respectively. In addition, the pH values of CWA, CLD, and CHD decreased to 3.92, 3.72, and 3.8 after 170 days of ensiling time, respectively. Furthermore, the pH values of the CLD and CHD groups were significantly (p < 0.05) lower than the CWA group, indicating that the addition of cellulase significantly reduced the pH of the silage. Thus, a low dose of cellulase (0.1%) showed better performance in maintaining a stable low pH compared to a higher (0.3%) dosage, which was consistent with the results reported by Colombatto et al. (2004b). The decrease in pH also promoted the degradation of lignocellulose and the production of soluble carbohydrate content, as indicated by the changes in WSC, NDF, ADF, and HC content. Sargsyan et al. (2016) reported that a long-term acidic environment with a low pH can loosen the lignocellulosic structure of fibrous polymers, which can help decompose the cell wall (cellulose and hemicellulose) due to the H+ from the organic acid and the release of soluble carbohydrates.
Fig. 2. Dynamic effects of cellulase dosage on fermentation pH values, where CWA contained no cellulase, CLD contained a low dosage of cellulase (0.1%), and CHD contained a high dosage of cellulase (0.3%). The different lowercase letters indicate significant differences at p < 0.05 between the different groups at the same time, while the different capital letters indicate significant differences at p < 0. 05 for the same groups at different times.
Concentration changes in the organic acids during the ensiling process are listed in Table 3. The concentration of LA in all three treated groups initially increased, followed by some decreases, which was further followed by increases in their maximum values at 60 days. It should be noted that the concentrations of LA in the CLD and CHD groups were higher than in CWA, while the maximum value was obtained in the CHD group, except at 130 days. The results showed that the addition of cellulase enhanced the production of LA, which was consistent with the relative abundance of LA bacteria in Table 3. The concentration of acetic acid (AA) in CWA and CHD increased steadily, while in the CLD group, it showed a transient increase (30–90 days) followed by a decrease. On day 30, both CLD and CHD groups had higher AA content than CWA. However, between 60 and 130 days, the concentration of AA in the CHD group remained lower than in the CWA group.
Moreover, the concentrations of butyric acid (BA) in the CLD and CHD groups were higher than in the CWA group, while propionic acid (PPA) and BA content were less than 0.5% of the total concentration of organic acid during the entire process. These results were consistent with the requirements for high-quality ensiling (Agneessens et al. 2015). The ethanol (EA) content of CWA decreased continuously, while in the CLD and CHD groups, it increased. Based on the calculated contents of the organic acids, the ratios of lactic acid/acetic acid (LA/AA) and lactic acid/total organic acid (LA/TOA) decreased from 7.44 and 0.82 (30 days) to 4.06 and 0.75 (30 days), respectively, for all three groups during co-ensiling. These ratios were higher in CHD and CLD than in CWA.
Ammonia nitrogen (AN) content represents protein degradation, and a decline in AN content indicates lower CP consumption and better preservation of silage (Luo et al. 2021). Thus, as shown in Fig. 3, the AN value in the CWA group was stable from 30 to 90 days, followed by a significant decrease (p < 0.05) after 130 days, and then stabilized afterward. The AN values of the CLD group were stable from 1.20 to 1.25 between 30 and 60 days, which significantly (p < 0.05) increased on day 90 and stabilized afterward. In the CHD group, the AN values increased first to the maximum (2.83 ± 0.38) after 90 days, and then decreased. Furthermore, CLD and CHD had significantly (p < 0.05) lower AN values than the CWA group between 30 and 60 days, which increased and stabilized on day 90, and finally significantly (p < 0.05) increased on day 170. These results indicated that the addition of cellulase reduced protein decomposition during the co-ensiling stage between the 30- and 60-day period. Although the AN values increased after 90 days, the values were still within the recommended range (less than 10%) for high-quality ensiling (Ni et al. 2017). Generally, a successful ensiling process requires that the concentrations of LA, AA, and PPA be no more than 60%, less than 4%, and 1.5% of the total organic acid content, respectively. In addition, a ratio of LA/AA higher than two and a concentration of BA close to zero is required (Colombatto et al. 2004b). Filya (2003) reported that a high-quality ensiling process should have high LA (> 60%), low BA (< 2%), and low AN content (< 9%), and a low pH value (< 4.2). Thus, it was concluded that the co-ensiling of DCS and CW achieved good long-term storage performance (170 days), and the addition of cellulase promoted silage quality.
Fig. 3. Dynamic effects of cellulase dosage on the ammonia nitrogen (AN) content, where CWA had no cellulase addition, CLD contained a low dosage of cellulase (0.1%), and CHD contained a high dosage of cellulase (0.3%). The different lowercase letters indicate significant differences at p < 0.05 between the different groups at the same time, while the different capital letters indicate significant differences at p < 0.05 for the same groups at different times.
Table 3. Dynamic Effects of Cellulase Dosage on the Formation Patterns of Low-molecular Fermentation Products (mg/g DM)