Experimental design
The experiment was a completely randomized design with four treatments assigned to sixteen fermentation vessels in two eight-vessel RUSITEC apparatus with four replications for each treatment. The treatments were control diet (no SRU), control plus 0.28% SRU (U28), control plus 0.56% SRU (U56), and control diet was modified for substituting with 0.35% SRU for equavelant soybean protein. The diets were formulated to be isoenergetic, but protein concentration of U28 and U56 increased due to SRU addition and isonitrogenic between control and U35 diets (Table 1). The SRU product was provided by King Techina Feed Co., Ltd. (Hangzhou, China) and it was prepared based on a matrix of urea pills (87%) and palm oil and sodium carboxymethyl cellulose (13%). The release rate of SRU was 52.1% at 6 h and 71.3% at 12 h of incubation in artificial rumen fluid. The experiment was conducted as one period consisting of 15 days, including 8 days of adaptation and followed by 7 days for sample and data collection. The basal diet was prepared in the form of total mixed ration (TMR), and ground through a 4 mm-sieve. Approximately 10 g (dry matter, DM) of the basal diet was weighed into nylon bags (10 × 20 cm; pore size of 50 µm, Ankom Technology Corp., Macedon, NY, USA), and SRU was added to bags at the desired concentration followed by manual mix.
Table 1
Ingredient and chemical composition of experimental diets
| Diets |
Item | Control | U28 | U56 | MU35 |
Ingredient, % | | | | |
Corn silage | 27.1 | 27.1 | 27.1 | 27.1 |
Alfalfa hay | 11.4 | 11.4 | 11.4 | 11.4 |
Ort hay | 7.0 | 7.0 | 7.0 | 7.0 |
Steam flaked corn | 24.4 | 24.4 | 24.4 | 26.1 |
Soybean meal, 46% | 10.6 | 10.6 | 10.6 | 8.9 |
Canola meal | 2.8 | 2.8 | 2.8 | 2.8 |
Extrude soybean | 3.4 | 3.4 | 3.4 | 3.4 |
DDGS1 | 2.1 | 2.1 | 2.1 | 2.1 |
Dried beet pulp | 6.84 | 6.84 | 6.84 | 6.84 |
Mineral and vitamin | 1.82 | 1.82 | 1.82 | 1.82 |
Mycotoxin binder | 0.08 | 0.08 | 0.08 | 0.08 |
Soda | 0.57 | 0.57 | 0.57 | 0.57 |
Saturated fat | 1.90 | 1.90 | 1.90 | 1.90 |
Slow-released urea2 | 0 | 0.28 | 0.56 | 0.35 |
Chemical composition, % of DM | | | | |
DM | 91.2 | 91.2 | 91.2 | 91.3 |
OM | 92.4 | 92.4 | 92.4 | 92.5 |
CP | 16.2 | 16.9 | 17.7 | 16.2 |
NDF | 30.4 | 30.4 | 30.4 | 30.5 |
ADF | 19.0 | 19.0 | 19.0 | 19.0 |
Starch | 27.1 | 27.1 | 27.1 | 28.2 |
NEL3, Mcal/kg | 1.71 | 1.71 | 1.71 | 1.71 |
1DDGS = Distillers dried grains with solubles. |
2The product contains 13% of coating materials of palm oil and sodium carboxymethyl cellulose and 87% of urea. |
3NEL = Net energy for lactation. |
Inoculum donor
The experimental protocols were reviewed and approved by the Lethbridge Research and Development Centre Animal Care Committee, and cattle were handled in accordance with the guidelines of the Canadian Council on Animal Care [13].
Three ruminally cannulated Angus cross cows (averaging 668 ± 55.1 kg body weight) that fed a TMR similar to the diet used in Rusitec (containing 35% corn silage, 10% mixed hay, and 55% corn-based concentrate mix, DM basis) were used as inoculum donor. Two hours after morning feeding, solid rumen digesta and rumen liquid were collected from four locations within the rumen of each cow via rumen cannula. Contents were immediately pooled over in equal amount (4 L per cattle), filtered through four layers of cheesecloth into an insulated thermos and transported to the laboratory directly. Rumen liquid was well mixed, pH recorded and kept at 39°C in a water bath prior to introduction into fermenters.
Experimental procedure
Two RUSITEC apparatuses were used in this study with each equipped with eight 920 mL volume anaerobic fermenters, as described in previous study [14]. The fermenters were randomly chosen, and each fermenter was outfitted with a port for buffer input, and a port for effluent output. To begin the experiment, each fermenter was filled with 200 mL of McDougalls buffer (pH = 8.2) [15], 700 mL of strained rumen fluid, and two pre-labeled nylon bags with one containing 20 g of mixed solid rumen digesta from four donor cattle and one containing diet substrate with or without SRU addition. The fermenters were incubated in a water bath at 39ºC. After 24 h of incubation, the nylon bag containing solid rumen digesta was removed from fermenter and replaced by one new bag with substrate with or without SRU. Thereafter, one diet bag was replaced daily in the morning so that each bag will remain in the fermenter for 48 h except the last day when one bag in each vessel was removed after 24 h. Artificial saliva [15] was infused into the fermenter continuously by a peristaltic pump set (Model ISM 932D, Ismatec, Index Health and science GmbH, Wertheim, Germany) at a dilution rate of 2.9%/h. On d 8, the chemical composition of the artificial saliva was modified by mixing ammonium sulfate enriched with 15N ((15NH4)2SO4; Sigma Chemical Co., St. Louis, MO, USA; minimum 15N enrichment 1 g/L) into the infused buffer solution to label bacteria. Daily effluent was collected into a 2 L-volumetric flask, and gas was collected in a 2 L-bag (CurityR; Conviden Ltd., Mansfield, MA, USA). The effluent in each flask was preserved with 3 mL of a sodium azide solution to stop the microbial activity during the sample collection period.
Nutrient disappearance
Nutrient disappearance including DM, organic matter (OM), crude protein (CP), acid detergent fibre (ADF), neutral detergent fibre (NDF) and starch was measured from d 9 to 13 of the sampling period [16]. Feedbags were removed from fermenters after 48-h incubation and then hand-washed under running water until the water runoff was clear. Afterwards, feedbags were dried in oven at 55°C for 48 h for DM disappearance determination [17]. Feed residues from the same fermenter and collected over 5 days were pooled, and grounded through a 1 mm-screen using a Wiley mill (standard model 4 Arthur Thomas Co., Philadelphia, PA, USA) for chemical analysis. Part of grounded feed residue was further ground with a ball mill (Mixer Mill MM2000; Retsch, Haan, Germany) for total N and starch analysis. Ash content was analyzed by combustion samples at 550°C for 5 h (method 942.05) [17], and OM content was calculated as 100 minus the ash content. The NDF and ADF were determined on a VELP Fiber Digestion System (VELP Scientifica, Burlington, ON, Canada) using the method as described by Van Soest et al. [18] and AOAC (method 973.18) [17], respectively. Total N was determined using a combustion analyzer (NA 2100, Carlo Erba Instruments, Milan, Italy) according to Smith and Tabatabai [19], and CP was calculated as total N × 6.25. Starch was determined by enzymatic hydrolysis of α-linked glucose polymers [20]. Disappearance of DM, OM, CP, NDF, ADF, and starch was calculated as the differences between the amount of individual component in substrates before incubation and that in residues after 48 h of incubation. The chemical analysis was conducted in duplicate, and repeated if the CV for the replicate analysis was more than 5%.
Fermentation parameters, gas and methane production
Fermenter pH using a pH meter (Orion model 260A, Fisher Scientific, Toronto, Canada) and the volume of daily effluent were measured daily at the time of feedbag exchange. From d 9 through d 13, effluent (5 mL) was preserved with 1 mL of 25% metaphosphoric acid (w/v) for analysis of volatile fatty acid (VFA). Another subsample of effluent (5 mL) was preserved with 1 mL of 1% H2SO4 (vol/vol) for ammonia N (NH3-N) analysis. All samples were stored in frozen at -20°C until analysis. Concentration of VFA was quantified using a gas chromatograph (model 5890, Hewlett-Packard Lab, Palo Alto, CA, USA) equipped with a capillary column (30 m × 0.32 mm i.d., 1-µm phase thickness, Zebron ZB-FAAP, Phenomenex, Torrance, CA, USA) and flame ionization detection, with crotonic acid (trans-2-butenoic acid) used as internal standard. The concentrations (mmol/L) of total VFA were calibrated based on daily effluent volume (L/d) to determine daily production of total VFA. The NH3-N concentration was determined as described by Rhine et al. [21]. For protozoa enumeration, 1 mL of liquid was gently squeezed from the 48-h incubated feed bag and transferred to a screw-cap vial containing 1 mL of methyl green-formalin-saline solution on d 11 to 13. The samples were stored at room temperature and prevented from light until counting by light microscopy with a Levy-Hausser counting chamber (Hausser Scientific, Horsham, PA, USA).
Total gas production was measured daily using a gas meter (Model DM3A, Alexander-Wright, London, UK). From d 9 to 13, a volume of 20 mL gas was collected from each gas collection bag using a syringe into evacuated 6.8 mL-exetainers (Labco Ltd., Wycombe, Bucks, UK). Methane concentration were determined using a Varian 4900 gas chromatograph equipped with a GS-Carbon PLOT 30 m × 0.32 mm × 3 µm column and thermal conductivity detector (Agilent Technologies Canada Inc., Mississauga, ON, Canada) at an isothermal oven temperature of 35°C, with helium as the carrier gas (27 cm/s).
Microbial protein synthesis
Microbial protein synthesis was estimated as the sum of microbial biomass in the form of liquid-associated bacteria (LAB), feed particle-associated (FPA) and feed particle-bound (FPB) bacterial fractions. On d 14 and 15, the daily total effluent for each fermenter was measured and a subsample (250 mL) was centrifuged (20,000×g, 30 min, 4°C) for isolation of LAB. The resulting pellets were washed with PBS and centrifuged (20,000×g, 30 min, 4°C) three times prior to suspension in distilled water, and stored in frozen. at the meantime, feedbags followed by 48 h of incubation, were put in a specific plastic bag with 20 mL of McDougall’s buffer and processed for 1 min in a Stomacher 400 Laboratory Blender (Seward Medical Ltd, London, UK). Feed residues were hand washed twice with 10 mL of McDougall’s buffer. All processed liquid was collected in centrifuge tubes, centrifuged (500×g, 10 min, 4°C) to remove large feed particles, and the supernatant was centrifuged (20,000×g, 30 min, 4°C) following the same procedure as for the LAB for obtaining FPA pellet. Washed feed residues (FPB fraction) were dried at 55°C for 48 h and weighed for amount of solid DM determination. The LAB and FPA pellets were freeze-dried and the samples (LAB, FPA and FPB) were ball ground (MM400; Retsch Inc., Newtown, PA, USA) for analysis of N and 15N by combustion analysis coupled to a mass spectrometer (NA1500, Carlo Erba Instruments, Milan, Italy).
Microbial community
Each of FPA and LAB sample were pooled from d 14 and 15 collection, and total microbial DNA was extracted using the Qiagen DNeasy PowerLyzer PowerSoil kit (Qiagen Inc., Mississauga, ON, Canada) according to manufacturer’s instructions. The concentration of DNA was measured using the Qubit dsDNA BR Assay Kit (Thermo Fisher Scientific Inc., Waltham, MA, USA) with a Qubit 2.0 Fluorometer (Thermo Fisher Scientific Inc., Waltham, MA, USA). Negative extraction controls were included in duplicate for control of extraction contamination. The extracted DNA was stored at -20℃ until sequencing.
All PCR amplification and sequencing steps were carried out at Genome Quebec (Montreal, QC, Canada). Libraries of 16S rRNA gene sequence were generated using a two-step PCR protocol. The V4 region of the 16S rRNA gene was amplified using the universal bacterial and archaeal primers 515-F (GTGYCAGCMGCCGCGGTAA) and 806-R (GGACTACNVGGGTWTCTAAT) in the first step of PCR [22]. A unique 10-bp barcode and Illumina (Illumina, San Diego, CA, USA) adapter sequences were added at the 5’ end of each amplicon in the second PCR step. The 16S rRNA gene amplicons were quantified using a Quant-iT PicoGreen dsDNA assay kit (Invitrogen, Burlington, ON, Canada), pooled in equimolar ratios, and then purified with AMPure XP beads (Beckman Coulter, Mississauga, ON, Canada). Sequencing of 16S rRNA gene amplicons was carried out according to manufacturer’s instructions using an Illumina MiSeq (2 × 250) and the MiSeq Reagent Kit v2 (500 cycles; Illumina).
Sequencing quality was checked with FastQC 0.11.5 and MultiQC 1.0 [23]. De-noised reads were used to construct amplicon sequence variants (SVs) using QIIME2 [24]. Analysis of 16S rRNA gene sequences was processed and analyzed within the QIIME2 and the R package DADA2 (Version 1.4). Sequences were then assigned to operational taxonomic units (OTUs) at 97% similarity using an open-reference OTU picking method and the Greengenes database v13_8 (). In this method, sequences that were less than 97% similar to those in the Greengenes database were clustered into OTUs using the de novo approach and USEARCH. The Shannon diversity index and observed OTU were calculated in QIIME2 and Bray-Curtis dissimilarities were assessed using the R packages vegan v. 2.4.4 [25] and phyloseq v. 1.20.0 [26]. Alpha diversity was estimated by observed OTUs and Shannon diversity index by using QIIME2. Beta diversity was performed to evaluate differences in overall bacterial communities by non-metric multidimensional scaling (NMDS), based on Bray-Curtis dissimilarities, using the vegan package of the R software suite. The significance of between-groups differentiation on Bray-Curtis dissimilarity was assessed by PERMANOVA using the adonis function of the R package vegan with 999 permutations. Differentially abundant OTUs between treated group and control group were identified with a threshold of 5% using DESeq2 [27].
Statistical analysis
Data were analyzed as repeated measures according to a completely randomized design using the MIXED procedure of SAS (Version 16.0.0, SAS Inst. Inc., Cary, NC, USA), with treatment as fixed effect, day of sampling as repeated measures, and fermenter and RUSITEC apparatus as random effect. For the repeated measures, various covariance structures were tested with the final structure chosen based on the minimum Akaike’s information criteria value. The protozoa count data were normalized by log10 transformation prior to statistical analysis. Data were tested for normality of variance. Orthogonal polynomial contrasts were performed to test for linear and quadratic responses to SRU at different addition level (i.e., control, U28, U56). Contrasts were generated to compare the control and MU35. The differences were declared significant at P ≤ 0.05 and trends at 0.05 < P ≤ 0.10.