Effects of Brewers’ Spent Grain Protein Hydrolysates on Gas Production, Ruminal Fermentation Characteristics, Microbial Protein Synthesis and Microbial Community in an Artificial Rumen Fed a High Grain Diet

Preparation of BSG protein hydrolysates

The BSG protein hydrolysates used in the current study were prepared following alkali extraction, acid precipitation and enzyme hydrolysis as previously reported [11, 13]. In brief, BSG was milled through a 0.5-mm screen, solubilized in 0.1 M NaOH at 20% (w/v) rate, followed by continuous stirring at 350 rpm at 50^oC for 2 h for protein extraction. Then the supernatant, collected by centrifuging at 8,000 × g for 15 min at 20^oC, was passed through an ultrafiltration membrane to obtain the protein fraction > 1 kDa, which was furtherly precipitated by acid (adjusting pH to 3.5) and centrifuged at 8,000 × g for 15 min at 20^oC, freeze dried. The obtained protein was dispersed in deionized water to reach a 5% (w/v) solution, hydrolyzed by 1% alcalase and flavourzyme at their optimum pH and temperatures (pH 8.0, 55^oC and pH 6.6, 50^oC, respectively). At the end of hydrolysis, all the hydrolysate solutions were adjusted to pH 7.0, heated at 95^oC for 5 min to inactive the enzyme, and centrifuged at 8,000 × g for 30 min to separate solubilized peptides and amino acids from the non-soluble substrates. The obtained hydrolysates with alcalase and flavourzyme were referred to AlcH and FlaH, respectively. The protein content and sample degree of protein hydrolysis were determined as described [11, 13].

Experimental design and treatments

The experiment was a completely randomized block design with four treatments assigned to sixteen fermentation vessels in two units of RUSITEC apparatus. Treatments were control (no hydrolysates, no antibiotics), 1% AlcH, 1% FlaH (diet dry matter; DM), and antibiotics (33 mg monensin + 11 mg tylosin/kg diet DM; positive control). The substrate was a high concentrate diet that contained 10% barley silage, 87% dry-rolled barley grain, and 3% vitamin and mineral supplement (DM basis; Table 1). The experiment lasted 15 d in duration, including 8 d for adaptation and followed by 7 d for sampling and data collection. The basal diet was prepared as total mixed ration, and grounded through a 4-mm sieve (Arthur Thomas Co., Philadelphia, PA, USA). Approximately 10 g (DM) of diet was weighed into nylon bags (10 × 20 cm; pore size of 50 µm, Ankom Technology Corp., Macedon, NY, USA). The protein hydrolysates and antibiotics were added to bags at the desired amount and manually mixed.

Inoculum donor

Three ruminally fistulated Angus cross cows (average, 768 ± 95.1 kg BW) fed a high grain diet containing 8.2% barley silage, 89.2% dry rolled barley grain, and 2.6% vitamin and mineral supplement (DM basis) were used as rumen inoculum donor. Two hours after morning feeding, solid and liquid contents were collected from four locations within the rumen of each cow via rumen cannula. The contents were immediately filtered through four layers of cheesecloth, pooled over in equal amount (4 L per cow), and recorded for pH before adding into fermenters. The experimental protocols were reviewed and approved by the Lethbridge Research and Development Centre Animal Care Committee, and the cows were handled in accordance with the guidelines of the Canadian Council on Animal Care [16].

Experimental procedure

The RUSITEC procedure were carried out using two units of RUSITEC apparatus, equipped with eight 920-mL anaerobic fermenters of each unit, as described previously [17, 18]. In brief, to initiate the fermentation, each fermenter was filled with 200 mL of McDougall’s buffer [19] and 700 mL of prepared rumen inoculum, and then one bag containing 20 g of prepared solid rumen digesta and one bag containing 10 g of experimental diet (DM basis) were placed in each fermenter at 0900 h on d 0. Fermenters were placed in a circulating water bath at 39 °C for incubation. After 24 h of incubation, the bag containing solid rumen digesta was replaced with a bag containing experimental diet. Thereafter, a bag was replaced daily, so that each bag remained in each fermenter for 48 h. During the daily feed-bag exchange, the fermenter was flushed with N₂ gas to maintain anaerobic condition in the fermenters. The artificial saliva was continuously infused into fermenters using a peristaltic pump (Model ISM 932D, Ismatec, Index Health and science GmbH, Wertheim, Germany) at a dilution rate of 2.9%/h. Effluent and fermentation gasses of each fermenter were collected, respectively, into a 2 L Erlenmeyer flask and a reusable 2 L gas-tight collection bags (CurityR; Conviden Ltd., Mansfield, MA, USA), and they were recorded at the feed bag exchange.

Nutrient disappearance

Disappearance of DM, OM, CP, acid detergent fiber (ADF), neutral detergent fiber (NDF) and starch was measured with 48-h incubated feed bags from d 9 to 13 of the sampling period. Bags were withdrawn from fermenters and washed manually under running cold water until the water was clear, and dried at 55 °C for 48 h (AOAC, 2005; method 930.15) to determine DM disappearance. Thereafter, the bag residues were pooled over 5 d by fermenter, and ground through a 1-mm sieve, for DM, OM, NDF and ADF analysis. A portion of ground sample was furtherly ball ground using a ball mill (Mixer Mill MM2000; Retsch, Haan, Germany) for total N and starch analyze. Disappearance of DM, OM, CP, NDF, ADF and starch was calculated as the differences between the amount of input and the amount remaining in residues.

Gas production and dissolved gases

Fermentation gasses were measured every 24 h using a gas meter (Model DM3A, Alexander-Wright, London, England, UK) from d 9 to 15 of the sampling period. The gas (20 mL) was sampled from each bag once daily for determining gas profiles. On d 14 and 15, fermentation liquids were sampled in duplicates (35 mL each) from each fermenter for measuring concentration of dissolved H₂ (dH₂) and dissolved CH₄ (dCH₄). The sampling procedure was carried out using two syringes as reported [20]. In brief, a 50 mL syringe that filled with 35 mL of fermentation liquids connected to a 20 mL syringe that filled with 5 mL of N₂ through a T tube with a valve. The N₂ was injected from the small syringe into the large syringe and the apparatus was vigorously shaken by hand. The entire gas phase was then transferred from the large syringe into the small one to determine the gas volume. Finally, the small syringe was removed from T tube and 6 mL of gas was sampled for both gas and dissolved gas analyses using a Varian 4900 Gas Chromatograph (Agilent Technologies Canada Inc., Mississauga, Ontario, Canada).

Fermentation parameters

The pH of fermentation fluid of each fermenter was measured daily using a pH meter (Orion model 260A, Fisher Scientific, Toronto, ON, Canada) at the time of feed-bag exchange. From d 9 to 13 of sampling period, effluent (5 mL) was sampled and preserved with 1 mL of 25% metaphosphoric acid for VFA analysis; another 5 mL of effluent were preserved with 1 mL of H₂SO₄ (1% vol/vol) for NH₃-N analysis. All samples were well mixed and frozen at -20 °C until analyzed. The production (mmol/d) of total VFA, individual VFA and NH₃-N were calibrated based on daily effluent volume.

Microbial protein synthesis

Bacteria in the fermenters were labeled using ¹⁵N. From d 7 to 15, 0.3 g/L (NH₄)₂SO₄ in McDougall’s buffer was replaced with 0.3 g/L ¹⁵N-enriched (NH₄)₂SO₄ (Sigma Chemical Co., St. Louis, MO, USA; minimum ¹⁵N enrichment 10.01 atom%). From d 9 to 15, the effluents were preserved by adding 3 mL of a sodium azide solution (20%; wt/vol) to each effluent flask. On d 14 and 15, daily volume of effluent from each fermenter was recorded and 35 mL of effluent was sampled and centrifuged at 20,000 × g, 4 °C for 30 min to isolate liquid-associated bacteria (LAB). The obtained pellets were washed with deionized water and centrifuged three times (20,000 × g, 30 min, 4 °C), suspended in distilled water, freezing and lyophilisation for determination of N and ¹⁵N.

Feed particle-associated (FPA) and feed particle-bound (FPB) bacterial fractions were measured from 48-h feed residues. After 48-h of incubation, feed bags were squeezed to expel excess liquids, placed individually in a plastic bag with 20 mL of McDougall’s buffer, and processed for 1 min using a Stomacher 400 Laboratory Blender (Seward Medical Ltd, London, UK). Then, the liquid from each feed bag was squeezed into a 50 mL centrifuge tube, the feed residues were washed twice with 10 mL of McDougall’s buffer in each wash, and the washed buffer was pooled over with the squeezed liquid to obtain the FPA bacterial fraction. The washed feed residues were considered as the FPB bacterial fraction. The obtained FPA samples were centrifuged at 500 × g, 4 °C for 10 min to remove large feed particles, the supernatant was centrifuged (20,000 × g, 30 min, 4 °C) to isolate bacterial pellets, which was furtherly processed as aforementioned. Washed feed residues (FPB fraction) were dried at 55 °C for 48 h, weighed for DM determination, and ball grinded (MM400; Retsch Inc., Newtown, PA, USA) for N and ¹⁵N analysis. Total microbial protein synthesis was estimated as the sum of LAB, FPA and FPB.

Microbial community

Microbial community of FPA samples were assessed through high-throughput sequencing. Total DNA was extracted from FPA samples (30 mg) using a QIAamp Fast DNA stool mini kit (Qiagen, Hilden, Germany), according to the manufacturer’s instructions. The quality and quantity of extracted DNA was measured using a NanoDrop spectrophotometer (Thermo Fisher Scientific, Waltham, MA, USA), then the extracted DNA was stored at -20 °C until sequencing. The V4 hypervariable region of the archaeal and bacterial 16S rRNA gene was amplified using the modified 515-F and 806-R primers, with the PCR conditions and sequencing steps carried out as previously described by [21]. Briefly, the 16S rRNA gene amplicons were generated using a two-step PCR, and then the amplicons were subjected to Illumina paired-end library preparation, cluster generation, and sequenced on an Illumina MiSeq instrument (Illumina, Inc., San Diego, CA, USA).

The obtained 16S rRNA gene sequencing raw data were processed using QIME2 [22] and the R-package DADA2 (Version 1.4) denoise method as described. In brief, after removing primer sequences and truncating both the forward and reverse reads at 225 bp, quality control was done for the reads using the QIME2, with chimeric sequences identified and removed. Then, the richness (number of OTUs) and diversity (Shannon index) were calculated and non-metric multidimensional scaling (NMDS) was performed based on Bray-Curtis similarity distances using R packages vegan (Version 2.4.4; [23]) and phyloseq (Version 1.20.0; [24]). Fold change of ruminal bacterial at genus level with a threshold of 5% was analyzed using R-package Deseq2 [25].

Chemical analysis

The chemical analyze of the feeds and feed residues were conducted in duplicate, and analysis was repeated when the CV for the replicate analysis was more than 5%. Analytical DM was measured by oven drying at 135 °C for 2 h (AOAC, 2005; method 930.15) [26] and ash content was analyzed by combustion samples at 550 °C for 5 h (AOAC, 2005; method 942.05). The NDF and ADF were determined on a VELP Fiber Digestion System (VELP Scientifica, Burlington, Ontario, Canada) using the method of Van Soest et al. [27] and AOAC (2005; method 973.18), respectively. Total N of feed and residue samples and ¹⁵N of LAB, FPA and FPB samples were analyzed using combustion analyzer (NA 2100, Carlo Erba Instruments, Milan, Italy), and CP was calculated as total N × 6.25. Starch was determined by enzymatic hydrolysis of α-linked glucose polymers as reported previously [28]. Concentration of VFA and NH₃-N in the effluent was determined using a gas chromatograph (model 5890, Hewlett-Packard Lab, Palo Alto, CA, USA).

Statistical analysis

Data were analyzed according to a completely randomized block design using the MIXED procedure of SAS (Version 16.0.0, SAS Inst. Inc., Cary, NC, USA), with treatment considered as fixed effect, day of sampling as repeated measures, and the fermenter and RUSITEC apparatus as random effects. For the repeated measures, various covariance structures were tested with the final structure chosen based on the lowest Akaike’s information criteria value. Results are reported as least squares means, which were compared using the Tukey correction for multiple comparisons. Significance among treatments was declared at P ≤ 0.05 and a trend at 0.05 < P ≤ 0.10 unless otherwise stated.

BSG protein hydrolysates of AlcH and FlaH

The protein content of BSG protein extract was 70%, which was much higher than that of the original BSG substrate (Table 1). The present value was comparable to a recent study that combination of enzyme and ultrasonication, resulting in 69.8% of BSG protein content [29]; whereas, it was greater than pre-treated BSG with carbohydrases followed by direct hydrolysis using proteolytic enzymes (63.1%) [30]. The Degree of hydrolysis (DH) of the protein extracted from BSG was greater (P < 0.01) when it was treated with alcalase (34.7%) than with flavourzyme (11.9%) and both protein hydrolysates showed good solubility in water. The DH expresses the number of peptide bonds cleaved as a percentage of total number of peptide bonds in the substrate; the higher DH can be expected to have lower molecular weight peptides and vice versa. Bamdad et al. [11] showed good antioxidant activities of barley hordein and glutelin enzymatic hydrolysates. Especially the flavourzyme and alcalase hydrolysates had superior DPPH (1,1-diphenyl-2-picryl hydrazyl) free radical scavenging activity, Fe²⁺-chelating ability and superoxide radical scavenging capacity. The peptides high in hydrophobic and non-polar amino acid residues were most likely responsible for the antioxidant effects [11, 13]. As the hydrophobic and non-polar amino acids are mostly remained in BSG after brewing processing [11], it was hypothesized that the BSG hydrolysates may have beneficial effects in rumen feeds based on their antioxidant activities.

Effects of AlcH and FlaH on nutrient disappearance and fermentation characteristics

Supplementation of FlaH reduced (P < 0.01) the disappearance of DM, OM, CP and starch as compared to the control and AlcH, which had no difference in the nutrient disappearances (Table 2). The reduction of DM disappearance by FlaH consists with the results obtained in our previous study using batch culture technique. We found that the DM disappearance of barley grain linearly decreased from 76.4, 70.7 to 63.8% as the inclusion rate of FlaH increased from 0, 5 to 10% (unpublished data). The reduction in the disappearances of DM and OM by FlaH was primarily resulted from the decreased of CP and starch. It appeared that the effect of AlcH and FlaH on the nutrient disappearances was not associated with their antioxidant activities measured in the present study. In fact, if we admit that the increased the DPPH scavenging or Fe-chelating activity adversely affected the nutrient disappearance as observed with FlaH, the nutrient disappearance with AlcH would be less than that with FlaH because of the greater DPPH scavenging or Fe-chelating activity of AlcH than FlaH. These results suggest that the hydrolysate with chemically determined antioxidant activity may have different response under rumen fermentation condition. Furthermore, the fermentation under current Rusitec condition would be less in oxidative stress compared with in vivo where the oxygen can frequently enter the rumen with feeds and rumination. We also suggest that the hydrolysates may have more like antimicrobial than antioxidant activity in the fermentation. The different bioactivities of AlcH and FlaH in rumen fermentation would be expected, as the activity of protein hydrolysates depends on the specificity of the protease used, hydrolysis conditions and degree of hydrolysis [13]. The flavourzyme is an endo- and exopeptidase enzyme mixture, and mainly produce small peptides and free amino acids, while the alcalase is an endo-protease and mainly generate small- and medium-sized peptides [13]. Furtherly, about 6% of protein in the total protein input was hydrolysate origin as a result of adding FlaH, the decreased CP disappearance with FlaH could be due to the resistance of hydrolysate to microbial degradation. The resistance is related to the amino acid sequences and structures, especially peptides with Gly-Gly, Pro-X or X-Pro residues at the N-terminus, or with modification of N-terminal amino groups [31, 32]. Thus, it is predicted that FlaH generated by flavourzyme is more resistant to rumen microbial digestion, and possess strong bioactivity in modulating rumen fermentation, including reduce the activity of proteolytic and starch utilization microbes. The disappearance of NDF and ADF were not affected by supplementation of either AlcH or FlaH, suggesting that both protein hydrolysates had limited effect on the activity of fibrolytic microbes.

The pH in the fermenters remained constant throughout the whole experimental period, with slightly higher (P < 0.01) pH with Ant than other treatments (average, 5.76; Table 3). Production of VFA (mmol/d) and individual VFA molar proportion as well as acetate to propionate ratio were not affected with adding AlcH or FlaH except for lower proportion of butyrate with FlaH than control and AlcH. The similar VFA production is consistent with biologically minor difference in the OM disappearance despite of statistical difference between FlaH and control or AlcH. However, AlcH and FlaH had greater (P < 0.01) NH₃-N production than control. The greater NH₃-N production with FlaH is not expected because of decreased CP degradability by FlaH. It was probably because they provided certain amount of peptides and free amino acid to fulfil partial nitrogen requirements of ruminal microbes. It was reported that ruminal microbes preferred to use peptides or amino acids as a source of nitrogen or as a source of energy, thus lead to accumulation of NH₃-N [33]. We speculate that the supplementation of AlcH and FlaH can promote the deaminative activity of high ammonia producing bacteria. Ammonia is mainly produced by the low activity species, but proliferation of the high ammonia producing bacteria will be a problem if certain diets were fed [31]. It was reported that a variety of ruminal bacteria produced ammonia from protein hydrolysates, with strains of Bacteroides ruminicola, Megasphaera elsdenii, and Selenomonas ruminantium being the most active [34].

The supplementation of antibiotics (P < 0.01) decreased disappearance of DM, OM, CP, NDF, ADF and starch compared with other treatments. Therefore, lower (P < 0.01) production of total VFA and higher (P < 0.01) fermenter pH was observed in Ant treatment. Furthermore, Ant altered the individual VFA proportion as compared to the other treatments, with lower (P < 0.01) molar proportion of acetate, butyrate, BCVFA and caproate, and higher (P < 0.01) molar proportion of propionate and valerate, thus led to lower (P < 0.01) acetate to propionate ratio. Our results confirm the general known mode of action of monensin in the rumen [35]. Additionally, supplementation of monensin and tylosin reduced (P < 0.01) the production of ammonia, which was in alignment with previous reports that ionophores could decrease ammonia production by suppressing high ammonia producing microbial population, as well as the peptidelytic and deaminative activity of the bacteria that grow in the presence of ionophores [35]. Although the FlaH had similar effects on reducing CP and starch disappearance as Ant did, fibre disappearance and fermentation pattern were not altered. It is likely that mode of action in the rumen is different between monensin and FlaH. The addition of FlaH protected protein and starch from rumen fermentation, which can let more protein enter the small intestine for digestion and alleviate risk of rumen acidosis.

Effects of AlcH and FlaH on gas and dissolved gas production

H₂ was produced during the oxidizing of reduced electron carries (such as NADH and ferredoxin) by membrane-bound hydrogenases of some rumen microbes, which plays an important role in maintaining the oxidation-reduction homeostasis to guarantee rumen anaerobic fermentation [36]. However, the activity of membrane-bound hydrogenase will be quickly inhibited by high H₂ pressure, while the methanogenic Archaea can efficiently use H₂ to reduce CO₂ to keep a low ruminal H₂ pressure, but at the price of producing CH₄ [37]. Therefore, either reducing the production of H₂ or finding alternative sinks for H₂ are theoretically feasible to reduce CH₄ emission. It is now clear that ionophore antibiotics, like monensin, can decrease CH₄ production via inhibiting growth of H₂ producing bacteria, without causing side effects to succinate- and propionate-producing bacteria [35]. In the current study, fermentation liquid and gas were analyzed for measuring dissolved gas (dH₂ and dCH₄) and gas (H₂ and CH₄) production, respectively. Production of dissolved gas and total gas (L/d) as well as the production of dH₂ were not affected by treatments, whereas the production of dCH₄ (% of dGas, mg/d or mg/g digested DM), and CH₄ (% of gas) were lower (P < 0.01) with FlaH than with control and AlcH (Table 4). Supplementation of either AlcH or FlaH considerably decreased (P < 0.01) the production of H₂, expressed as µg/d or µg/g digested DM. Adding monensin and tylosin consistently decreased (P < 0.05) the production of CH₄, H₂ and dCH₄ compared with other treatments. The reduced production of H₂ by adding FlaH (-49%) is of interest, and it consistent with the decrease in dCH₄ production and in the proportion of CH₄ in total gas. However, the molar proportion of propionate was unchanged by adding FlaH despite of the decreased in production of CH₄ and H₂. The decrease in OM disappearance with FlaH may be unable to produce sufficient H₂ to increase propionate production. These results suggest that the FlaH may especially target at methanogenesis. The decreased production of H₂ with AlcH without altering the production of CH₄ and propionate is not clear. Anyway, the OM disappearance was not affected by AlcH. The present results confirmed the effect of monensin on reducing CH₄ and H₂ production and increasing propionate production [35]. Although the FlaH showed reduction of CH₄ and H₂ as well as the reduction in nutrient disappearance, the magnitude of the reduction by FlaH was much less than by monensin, and in particularly, the FlaH did not increase propionate production and alter fermentation pattern. Therefore, it suggests lower activity of FlaH than monensin and different mode of action between the two additives. Future work also needs to be carried out to determine relationships between their physicochemical and techno-functional properties.

Effects of AlcH and FlaH on microbial protein synthesis and microbial community

Production of total microbial N (LAB + FPB + FPA) and LAB was less (P < 0.05) with FlaH than control and AlcH, whereas the production of FPA was greater (P < 0.01) with AlcH and FlaH without difference in FPB among treatments (Table 5). Supplementation of monensin and tylosin produced greater (P < 0.01) FPA but less LAB and total microbial N compared to control and AlcH without differing with FlaH except for LAB, which was less (P < 0.01) with Ant than FlaH. Microbial N production efficiency did not differ among treatments. Although the total microbial N production reduced with adding FlaH or Ant, proportion of attached microbial biomass (FPA + FPB) in the total biomass was greater for FlaH (30.1%) or Ant (34.4%) than control (24.4%). Our results suggested that adding FlaH or Ant benefits to microbial colonization to feed particle. As the supplementation of AlcH and FlaH to basal diet increased the microbial population in the FPA fraction, which was interesting because the microbial attachment is essential to the feed digestion. Thus, the FPA samples were selected for high-throughput sequencing to assess the microbial community. Neither numbers of OTUs (Fig. 1A) nor Shannon diversity indices (Fig. 1B) of FPA microbial community were affected by supplementing AlcH and FlaH. Whereas, both indices were reduced (P < 0.05) by Ant. Similarly, the results of NMDS analysis indicated that there was no clustering of the FPA microbial community in control, FlaH and AlcH treatments, but the Ant treatment had more clustered community structure (Fig. 2). The study on the effects of BSG protein hydrolysates on ruminal microbiome is lacking. A study applied a blend of plant essential oils with antioxidant activity demonstrated no effect on ruminal microbiome [38]. Meanwhile, our results that monensin greatly reduced the diversity of FPA microbial community were in agreement with in vitro [39] or in vivo [38, 40] studies.

Analysis of the taxonomic composition revealed that AlcH and FlaH had no affect on the relative abundance (RA) of bacteria at phylum level (Fig. 3A). Whereas, the addition of antibiotics reduced (P < 0.05) the RA of Firmicutes and Bacteroidetes, and enhanced (P < 0.05) the RA of Proteobacteria at phylum level. For all treatments, phyla of Firmicutes, Bacteroidetes and Proteobacteria were the most predominant phyla, accounting for approximately 90% of the microbial biomass in FPA, which was consistent with the sequencing results of liquid associated microbial community of in vitro [38] and in vivo studies [39, 40]. This suggested that phyla of Firmicutes, Bacteroidetes and Proteobacteria were predominant in both liquid and particle associated ruminal microbiome. While at the genus level, Prevotella, Succinivibrio, Selenomonas, Shuttleworthia, Schwartzia, Bifidobacterium, Lactobacillus, Dialister and Anaerobiospirillum were among the 10 most abundant genera in FPA microbial community (Fig. 3B) and should be considered as the “core bacteria” associated with feed particle. The FlaH enhanced (P < 0.05) the RA of Prevotella, but reduced the RA of Selenomonas, Shuttleworthia, Bifidobacterium and Dialister as compared to the control. The AlcH also had less (P < 0.05) RA of Schwartzia and Bifidobacterium than control. Notably, Ant reduced (P < 0.05) the RA of Prevotella, Selenomonas, Shuttleworthia, Bifidobacterium, Lactobacillus and Dialister, and increased RA of Succinivibrio compared with control. Further Log2 fold change analysis found that several genera from phyla Actinobacteria, Bacteroidetes, Firmicutes and Proteobacteria had 5% more change (enhanced or reduced) than control (Fig. 4). Furthermore, genera of Shuttleworthia, Selenomonas, Lactobacillus and Schwartzia from phyla Firmicutes were especially susceptible to FlaH, AlcH and Ant supplementation. These results indicated that FlaH and AlcH had less antibacterial effects as compared to Ant. As F. succinogenes, R. albus, R. flavefaciens, P. ruminicola, E. cellulosolvens, and E. ruminantium are major rumen fibrolytic bacteria [41], the present results indicated that application of BSG protein hydrolysates had no obvious detrimental effect on fibrolytic bacteria. This was supported by no treatment effect of FlaH and AlcH on the disappearances of NDF and ADF.

More susceptible to FlaH and AlcH for genera of Shuttleworthia, Selenomonas, Lactobacillus and Schwartzia from phyla Firmicutes and genera of Bifidobacterium from phyla Actinobacteria, suggested that the BSG protein hydrolysates with antioxidant activity were more potent to Gram-positive bacteria than to Gram-negative bacteria. The underlined mechanism is unknown, but it seemed that this difference might be attributed to the lack of the protective outer membrane in Gram-positive bacteria. As essential oils with antioxidant activity, had a stronger antibacterial activity than other essential oils [42], we speculated that the antibacterial effects of BSG protein hydrolysates might be due to their antioxidant activities. The antioxidant activity of protein hydrolysates relies on the enzymes and methods used during BSG protein hydrolysis. Similarly, differences in effectiveness of monensin on ruminal Gram-positive and Gram-negative bacteria were also observed in the current study, and confirm the monensin sensitivity to Gram-positive bacteria [38].

Rumen microbial community was usually dominated by Prevotella at genus level when fed high grain diets [43]. Members of Prevotella are able to utilise various nutrients such as starch, proteins and non-cellulosic polysaccharides [44, 45], and to convert lactate into propionate to prevent accumulation of lactate [40]. Moreover, as Prevotella are also H₂-consuming bacteria in addition to Selenomonas [46], the reduced H₂ with FlaH and AlcH was likely due to the increased relative abundance of genus Prevotella. Meanwhile, members of Prevotella are known as the predominant proteolytic bacteria with a great diversity of extracellular proteolytic activities in the rumen [31, 47], thus, the decreased disappearance of CP by supplementing monensin could be explained by decreased relative abundance of Prevotella. Currently, the probiotics such as direct-fed microbials are often developed from the genera Lactobacillus and Bifidobacterium [48], and the reduced relative abundance of Lactobacillus and Bifidobacterium with Ant suggested a disadvantage of applying monensin in the diet. The genus Schwartzia are asaccharolytic and can ferment succinic acid to produce propionic acid [49]. It was reported that bacteria affiliated with Schwartzia were negatively correlated with methane emissions [50], which was contrary to our results that monensin reduced the RA of Schwartzia, as well as methane emission.

There is limited information on the genus Shuttleworthia, and its function in the rumen is almost unknown [43]. Recent studies found that genus Shuttleworthia was digesta-adherent rumen bacteria in dairy and beef cattle [51], and was a starch and sugar fermenter [52]. It was reported that supplementation of phytogenic compounds decreased the RA of Shuttleworthia in dry cow [52]. In the current study, the decreased Shuttleworthia by BSG hydrolysates or Ant may explain the decreased starch digestibility with FlaH and Ant. A recent study showed negative correlation of Shuttleworthia with lactate and NH₃-N concentration [53], thus explained the greater NH₃-N production with BSG hydrolysates.

Bacteria from genus Selenomonas, besides of degrading starch and cellulose, play critical role in maintaining normal rumen fermentation through converting lactate and succinate into propionate and reducing lactate accumulation [54–56] and maintaining low H₂ concentration as H₂-consumer [57, 58]. As fumarate and nitrate reducers, Selenomonas were proven a significant H₂ sinks in sheep [59]. Ruminants fed high starch diets often result in great rumen dH₂ concentration, and would promote the growth of S. ruminantium for disposing electrons derived from fermentation [57]. Therefore, in the present study, the reduced H₂ production by applying BSG protein hydrolysates or monensin is consistent with the declined relative abundance of Selenomonas.

Members of genus Succinivibrio can ferment both starch and cellulose into succinate, and succinate then fermented by Selenomonas and other bacteria into propionate via succinate pathway, which is the primary propionate producing pathway in rumen [56]. Therefore, the abundances of members of Succinivibrio in the rumen has been reported to be positively associated with feed efficiency of ruminants [60, 61]. The increased RA of Succinivibrio with monensin in the present study, was correlated with the increased propionate proportion and the increased fermenter pH. Our results agreed with previous report that monensin can lead to an increase in abundance of succinate producers [38], and the report that Succinivibrio is positively correlated to the ruminal pH [55, 62]. The increased pH with Ant would be partly due to the increased RA of genus Dialister, which were reported to play a role in altering the buffering capacity of rumen fluid [63]. Furtherly, although the effects of monensin on methanogens were not determined, the application of monensin reduced the H₂ production and CH₄ emission. Our results supported that the decreased methane production with Ant most likely ascribed to decrease in nutrient availability for methanogenesis by acting on other rumen microorganisms instead of a direct effect of monensin on the methanogens [38].