Search
Author
Title
Vol.
Issue
Year
1st Page

Journal of Animal Science - Animal Nutrition

Effect of feeding dried distillers grains with solubles on ruminal biohydrogenation, intestinal fatty acid profile, and gut microbial diversity evaluated through DNA pyro-sequencing

 

This article in JAS

  1. Vol. 92 No. 2, p. 733-743
     
    Received: Oct 04, 2013
    Accepted: Dec 14, 2013
    Published: November 24, 2014


    2 Corresponding author(s): pkononoff2@unl.edu
 View
 Download
 Share

doi:10.2527/jas.2013-7223
  1. E. Castillo-Lopez*11,
  2. H.A. Ramirez Ramirez*,
  3. T. J. Klopfenstein*,
  4. C. L. Anderson*,
  5. N. D. Aluthge*,
  6. S. C. Fernando*,
  7. T. Jenkins and
  8. P. J. Kononoff 2
  1. Department of Animal Science, University of Nebraska-Lincoln, Lincoln 68583-0908
    Department of Animal & Veterinary Sciences, Clemson University, Clemson, SC 29634

Abstract

The objectives of this study were to evaluate the effect of dried distillers grains with solubles (DDGS) on ruminal biohydrogenation and duodenal flow of fatty acids, and to evaluate effects on the ruminal and duodenal microbial community using Roche 454 pyro-sequencing. Three crossbred steers (average BW 780 ± 137 kg) fitted with ruminal and duodenal cannulae were used in a 3-diet, 6-period crossover design. Animals were housed in individual free stalls and fed twice daily at 0700 and 1900 h. Diets (DM basis) were 1) CONTROL, 19.5% corn bran, 20% sorghum silage, 60% brome hay, 0.5% trace minerals, and 0.25% urea, but no DDGS; 2) LOW DDGS, inclusion of 9.75% DDGS replacing equal percentage of corn bran; 3) HIGH DDGS, inclusion of 19.5% DDGS completely replacing corn bran. Feed ingredients and duodenal digesta samples were analyzed for fatty acid composition. The DNA was extracted from isolated mixed ruminal bacterial samples and from intestinal digesta samples. The V1-V3 region of the 16S rRNA gene was sequenced, and bacterial phylogenetic analysis was conducted. Data were analyzed using the MIXED procedure of SAS. Biohydrogenation of C18:1 increased (P < 0.01) with DDGS inclusion; means were 68.3, 75.6, and 79.3 ± 4.3% for CONTROL, LOW DDGS, and HIGH DDGS, respectively. In the same order, means of biohydrogenation of C18:2 (P < 0.05) were 84.1, 91.5, and 93.3 ± 3.4%. Duodenal flow of total fatty acids increased (P < 0.01) with DDGS inclusion; means were 134, 168, and 223 ± 33 g/d for CONTROL, LOW DDGS, and HIGH DDGS, respectively. In the same order, means of C18:0 flow (P < 0.01) were 51, 86, and 121 ± 18 g/d. DDGS did not affect the predominant bacterial phyla in the gut, which were Bacteroidetes (P = 0.62) and Firmicutes (P = 0.71). However, the phylum Fibrobacteres decreased (P < 0.01) when DDGS was fed with means of 5.5, 6.0 and 3.7 ± 0.6% for CONTROL, LOW DDGS, and HIGH DDGS, respectively. Fibrobacteres were lower (P < 0.01) in isolated ruminal bacterial samples compared to duodenal digesta samples with means of 0.1 and 10.1 ± 0.6%, respectively. Overall, the inclusion of DDGS in diets increased ruminal biohydrogenation of C18:1 and C18:2, which increased duodenal flow of C18:0. In addition, the bacterial community of the rumen clustered separately from that of the duodenum suggesting different bacterial diversity between isolated ruminal bacteria and duodenal digesta.



INTRODUCTION

Dried distillers grains with solubles (DDGS) contain approximately 31% CP, 34% NDF, 12% fat, and 5% starch (Paz et al., 2013). Compared to common feed ingredients such as corn or soybean meal, DDGS contain greater amounts of PUFA, most notably linoleic acid. In the rumen, unsaturated fatty acids may be saturated by ruminal microbes through a process known as biohydrogenation (Harvatine and Allen, 2006). As a consequence, the fatty acid composition of ruminant food products (Palmquist et al., 1993; Jenkins et al., 2008) is affected by the interaction of the diet and the microbial community of the gut (Beam et al., 2000; Abdelqader et al., 2009).

Several studies have evaluated the impact of feeding DDGS on the bacterial community of the rumen (Callaway et al., 2010; Aldai et al., 2012). Recently, we have also reported that the inclusion of DDGS, at the expense of corn bran, reduced the flow of bacteria from the rumen (Castillo-Lopez et al., 2013). Given that the type of fatty acids present in ruminant food products may be affected by microbial metabolism (Harfoot and Hazlewood, 1988), there is a need to clearly understand how the microbial community and fatty acid profile of digesta changes when DDGS are fed. Although a number of bacteria have been identified to participate in ruminal biohydrogenation, including Butyrivibrio sp. (Polan et al., 1964), Megasphaera elsdenii (Kim et al., 2002), and Propionibacterium acnes (Devillard et al., 2006), the impact on the broader ruminal microbial community is still not clearly understood. The objectives of this study were to evaluate the effects of DDGS on ruminal biohydrogenation and duodenal flow of fatty acids, and to evaluate effects on the phylogenetic bacterial diversity of the rumen and duodenum. We hypothesized that the inclusion of DDGS in the diets would increase the flow of saturated fatty acids to the duodenum and cause a shift in the bacterial diversity of the rumen and duodenum.


MATERIALS AND METHODS

Steers were managed according to the guidelines stipulated by the University of Nebraska Animal Care and Use Committee. Research presented in this paper is part of a larger experiment evaluating the ruminal undegradable protein of DDGS and its impact on total mass flow of ruminal microorganisms to the small intestine, a portion of which has been previously reported (Castillo-Lopez et al., 2013).

Animals and Experimental Design

Three British-bred crossbred steers fitted with ruminal and duodenal cannulae were used in this experiment, which was a 3-diet, 6-period crossover design. Duodenal cannulations were conducted as described previously (Streeter et al., 1991; Castillo-Lopez et al., 2013). During this study, steers received each of the 3 diets twice during six 19-d experimental periods. Animals averaged 780 ± 137 kg of BW throughout the trial and were housed in individual free box stalls. Diets (Table 1) were formulated to include 0, 9.75, or 19.5% DDGS (DM basis), which was obtained from a Nebraska-based corn-ethanol plant (Green Plains Renewable Energy, Central City, NE). The chemical composition of feed ingredients utilized in this experiment is listed in Table 2. These diets were formulated so that DDGS would simply replace corn bran and urea. Total mixed rations (TMR) were mixed by hand daily, and animals were fed 2 times per day at 0700 and 1900 h. The amount of feed offered was adjusted before the initiation of the trial by feeding a diet consisting of 80% forage and 20% concentrate to 95% of ad libitum intake during a 10-d period, followed by the experimental periods. Water was available for ad libitum consumption. Chromic oxide (Cr2O3) was used as a marker for the estimation of duodenal flow of DM (Hutton et al., 1971; Harvatine et al., 2002; Sylvester et al., 2005). Gelatin capsules (Torpac Inc., Fairfield, NJ) were filled with 7.5 g of Cr2O3, then dosed into the rumen via the ruminal cannula twice daily at 0700 and 1900 h during d 9 through 19 of each experimental period to provide a marker to estimate duodenal digesta flow.


View Full Table | Close Full ViewTable 1.

Ingredients and analyzed chemical composition of CONTROL, LOW dried distillers grains with solubles (DDGS), and HIGH DDGS diets

 
Diet1, % of DM
Ingredient CONTROL LOW DDGS HIGH DDGS
DDGS2 9.75 19.5
Corn bran 19.5 9.75
Sorghum silage 20.0 20.0 20.0
Brome hay 60.0 60.0 60.0
Trace minerals 0.50 0.50 0.50
Urea 0.25 0.125
Chemical composition
    DM, % as is 78.9 79.0 79.2
    CP 10.6 11.8 13.0
    NDF 66.2 62.9 59.6
    Starch 5.7 4.6 3.5
    Ether extract 1.7 2.6 3.6
    Ash 7.2 7.6 7.9
Fatty acids
    C12:0 0.01 0.01 0.01
    C14:0 0.01 0.01 0.01
    C16:0 0.21 0.27 0.35
    C18:0 0.04 0.05 0.06
    C18:1 0.15 0.29 0.43
    C18:2 0.36 0.66 0.96
    C18:3 0.13 0.14 0.15
    C20:0 0.02 0.02 0.02
    C22:0 0.02 0.02 0.02
    Other 0.68 0.70 0.71
    Total 1.63 2.17 2.70
1CONTROL contained (DM basis) no DDGS, 19.5% corn bran, 20% sorghum silage, 60% brome hay, 0.5% trace minerals, and 0.25% urea. LOW DDGS contained (DM basis) 9.75% DDGS replacing an equal percentage of corn bran. HIGH DDGS contained (DM basis) 19.5% DDGS completely replacing corn bran.
2DDGS obtained from Green Plains Renewable Energy (Central City, NE).

View Full Table | Close Full ViewTable 2.

Analyzed chemical composition (mean and SD) of feed ingredients used in the formulation of CONTROL, LOW dried distillers grains with solubles (DDGS), and HIGH DDGS diets1

 
Feedstuff, % of DM
Component DDGS2 Corn bran Sorghum Silage Brome hay
DM, % as is 90.7 ± 1.93 89.0 ± 0.70 35.9 ± 0.52 90.5 ± 1.22
CP 27.8 ± 0.76 11.8 ± 0.30 7.2 ± 0.30 10.2 ± 0.73
NDF 31.5 ± 0.70 65.4 ± 0.99 54.9 ± 0.47 70.8 ± 1.74
Starch 5.2 ± 0.22 16.5 ± 0.91 4.2 ± 0.24 2.7 ± 0.57
Ether extract 11.7 ± 0.51 2.0 ± 0.15 1.5 ± 0.27 1.7 ± 0.18
Ash 4.3 ± 0.12 1.0 ± 0.09 8.2 ± 0.25 8.2 ± 0.50
Fatty acids
    C12:0 0.006 ± 0.006 0.001 ± 0.005 0.017 ± 0.002 0.012 ± 0.001
    C14:0 0.012 ± 0.000 0.001 ± 0.001 0.012 ± 0.001 0.012 ± 0.001
    C16:0 1.09 ± 0.032 0.11 ± 0.013 0.17 ± 0.017 0.17 ± 0.011
    C18:0 0.17 ± 0.006 0.02 ± 0.004 0.04 ± 0.006 0.04 ± 0.003
    C18:1 1.95 ± 0.055 0.19 ± 0.023 0.05 ± 0.018 0.05 ± 0.002
    C18:2 4.42 ± 0.115 0.44 ± 0.051 0.13 ± 0.024 0.11 ± 0.004
    C18:3 0.13 ± 0.005 0.01 ± 0.004 0.14 ± 0.030 0.14 ± 0.012
    C20:0 0.036 ± 0.002 0.004 ± 0.002 0.029 ± 0.033 0.015 ± 0.002
    C22:0 0.022 ± 0.002 0.002 ± 0.002 0.016 ± 0.002 0.020 ± 0.002
    Other 0.31 ± 0.027 0.032 ± 0.020 0.396 ± 0.095 0.941 ± 0.080
    Total 8.17 ± 0.203 0.82 ± 0.080 1.0 ± 0.174 1.51 ± 0.074
1CONTROL contained (DM basis) no DDGS, 19.5% corn bran, 20% sorghum silage, 60% brome hay, 0.5% trace minerals, and 0.25% urea. LOW DDGS contained (DM basis) 9.75% DDGS replacing an equal percentage of corn bran. HIGH DDGS contained (DM basis) 19.5% DDGS completely replacing corn bran.
2DDGS obtained from Green Plains Renewable Energy (Central City, NE).

Collection of Duodenal Digesta and Feed Samples

Duodenal digesta contents (200 mL) were collected from each steer every 4 h and placed in 250-mL Nalgene bottles (Thermo Scientific Inc., Waltham, MA) during d 16 through 19 of each period. Collection time was advanced 1 h in subsequent collection days, so that every 1-h interval in a 24-h period was represented with 6 samples per day and a total of 24 samples per steer per period. Collected duodenal digesta samples were then composited by steer, within period, and immediately frozen at –20°C for subsequent analysis for chemical and fatty acid composition, and for later DNA extraction, DNA pyro-sequencing, and bacterial phylogenetic analyses. Samples of individual feed ingredients as well as samples of each TMR were also collected twice daily immediately after feeding on d 16 through 19 of each period and frozen at –20°C for later analyses of chemical and fatty acid composition.

Collection of Mixed Ruminal Bacterial Samples

On d 18 and 19 of each experimental period, 1.5 L of whole ruminal digesta contents were taken by hand from the caudal ventral sac, cranial ventral sac, and 2 samples from the feed mat in the dorsal rumen of each steer at 1000 and 1600 h (d 18) and 1200 and 1800 h (d 19). Then, mixed ruminal bacteria were isolated according to the procedure described by Hristov et al. (2005). Briefly, collected whole ruminal contents were composited by steer and squeezed through 2 layers of cheesecloth and the filtrate was retained. Solids remaining on the cheesecloth were added to a volume of cold buffer (McDougall, 1948) equal to the volume of filtrate, and shaken manually in a screw-capped jar to dislodge the ruminal microorganisms loosely associated with feed particles. This suspension was then squeezed through 2 layers of cheesecloth and the 2 filtrates were combined (vol/vol) and preserved with 5% (vol/vol) formalin. From this sample, bacteria were harvested via differential centrifugation (Hristov and Broderick, 1996) with an initial low-speed centrifugation at 400 × g for 5 min at 4°C and a subsequent high-speed centrifugation of the supernatant at 20,000 × g for 15 min at 4°C. Samples were maintained on ice while being processed. The supernatant was then discarded, and the isolated bacterial pellets collected on both days were composited by steer, within period and frozen at –20°C for later DNA extraction, DNA pyro-sequencing, and bacterial phylogenetic analyses.

Ruminal Fluid pH Measurements

Ruminal fluid pH was measured according to the protocol described by Penner et al. (2006). Briefly, during d 16 through 19 of each period ruminal fluid pH was measured once every minute using wireless pH meters (Dascor Inc., Escondido, CA). The pH meter was placed into the ventral sac of the rumen of each steer. Each pH meter contained a data logger, 9-V battery, and an electrode cable housed in a watertight capsule constructed out of polyvinyl chloride. Each pH electrode was enclosed in a weighted, polyvinyl chloride cover that maintained the electrode in the ventral sac of the rumen. Measurements of ruminal fluid pH were averaged by steer across the 4 collection days, so that a period of 24 h was represented in each period. From these values, minimum, maximum, and mean ruminal fluid pH was calculated for each steer. The time and area below the mean ruminal fluid pH are also presented. Wireless ruminal fluid pH probes were calibrated in buffers pH 7 and 4 before inserting them into the rumen and after downloading pH data at the end of each period. The initialization of the data logger and the transfer of data were performed by a Microsoft Windows-compatible software package supplied by Dascor (M1b version 6.1.2h).

Chemical Analysis of Feed and Duodenal Digesta Samples

Collected feed ingredients and TMR samples were dried for 48 h at 60°C in a forced air oven, ground to pass through a 1-mm screen (Wiley mill, Arthur A. Thomas Co., Philadelphia, PA), and analyzed for chemical composition by an external laboratory (Cumberland Valley Analytical Services, Hagerstown, MD), which included DM (method 930.15; AOAC, 2000), N (method 990.03; AOAC, 2000; Leco FP-528 Nitrogen Combustion Analyzer, Leco Corp. St. Joseph, MI), NDF (Van Soest et al., 1991), starch (Hall, 2009), ether extract (method 2003.05; AOAC, 2006) and ash (method 942.05; AOAC, 2000). Nutrient composition of the diets (Table 1) was calculated based on analysis of individual feed ingredients and the rate of inclusion in the diet.

For the determination of ruminal nutrient digestibility, subsamples of collected duodenal contents were lyophilized and ground to pass through a 1-mm screen using a Wiley mill (Arthur H. Thomas, Philadelphia, PA). Then, samples were analyzed for DM (100°C for 24 h), N (method 990.03; AOAC, 2000; Leco FP-528 Nitrogen Combustion Analyzer), NDF (Van Soest et al., 1991), starch (Hall, 2009), and ash (method 942.05; AOAC, 2000). Approximately 0.5 g of ground duodenal samples were also analyzed for Cr2O3 by an external laboratory (Servi-Tech Laboratories, Hastings, NE) by sample digestion in 10 mL nitric acid and 3 mL peroxide, with the addition of 4 mL of hydrochloric acid, followed by analysis by inductively coupled plasma. Duodenal flows were calculated as described by Erasmus et al. (1992). Briefly, the flow of DM was calculated by dividing the dosing rate (g/d) of Cr2O3 by Cr2O3 concentration in ground duodenal samples. Ruminal digestibility coefficients were then calculated based on the intake and the duodenal flows of each nutrient (Harvatine et al., 2002).

Fatty Acid Analysis of Feed and Duodenal Samples

Individual feed ingredients and subsamples of duodenal digesta were analyzed for fatty acid composition at Clemson University (Clemson, SC). Fatty acid analysis was based on the direct transesterification procedure of Sukhija and Palmquist (1988). Fatty acids were separated on a 30 m × 0.25 mm × 0.2 μm film thickness SP2380 capillary column (Supelco Inc., Bellefonte, PA). The column oven was programmed from an initial temperature of 150°C held for 2 min, increased at a rate of 2°C per min, and then held at a final temperature of 220°C for 10 min. Injector and detector temperatures were maintained at 250°C. Heptadecanoic acid was added as an internal standard to all samples (Fotouhi and Jenkins, 1992; Jenkins and Adams, 2002; Kelzer et al., 2009; Maia et al., 2012). Helium was used as a carrier gas, and verification of peak identity was established by comparison of peak retention times to known standards.

Calculations of Ruminal Biohydrogenation of Fatty Acids

Two methods were utilized for the calculation of ruminal biohydrogenation of C18:1, C18:2, and C18:3 fatty acids. The first method was as described by Lundy et al. (2004) and Loor et al. (2004). This approach is based on the total grams of fatty acid disappearing between the mouth and duodenum (consumed–duodenal flow) as a percentage of fatty acid consumed. The second method was conducted according to the formula described by Wu et al. (1991): biohydrogenation (%) = 100 – 100 × (individual unsaturated C18/total C18 in duodenal digesta)/(individual unsaturated C18/total C18 in feed).

Evaluation of the Gut Bacterial Diversity Through DNA Pyro-sequencing

DNA Extraction.

To evaluate ruminal and duodenal bacterial diversity, DNA was extracted from collected mixed ruminal bacterial samples and from intestinal digesta samples using an automated DNA extraction protocol at University of Nebraska Core for Applied Genomics and Ecology (CAGE) facilities as described by Martinez et al. (2009). Briefly, samples were thawed and 100 μL of each were transferred to sterile bead beating tubes (Biospec Products, Bartlesville, OK) containing 300 mg of zirconium beads (0.1 mm). Samples were washed 3 times in chilled PBS followed by centrifugation at 8,000 × g for 5 min at room temperature. Pellets were suspended in 100 μL of lysis buffer (200 mM NaCl, 100 mM Tris, 20 mM EDTA, 20 mg/mL lysozyme, pH 8.0) and incubated at 37°C for 30 min. Buffer ASL (1.6 mL) from the QIAamp DNA Stool Mini Kit (Qiagen, Valencia, CA) was added to each sample after the samples were homogenized in a MiniBeadbeater-8 (BioSpec Products) for 2 min at maximum speed. The DNA was isolated from 1.2 mL of the supernatants using the QIAamp DNA Stool Mini Kit, following the manufacturer’s instructions. The concentration of DNA in each sample was measured using a NanoDrop ND-1000 Spectrophotometer (NanoDrop Technologies, Inc. Wilmington, DE) and stored at –20°C until sequencing.

DNA Pyro-sequencing and Bacterial Phylogenetic Analysis.

The V1-V3 region of the 16S rRNA gene was amplified using bar-coded fusion primers (Benson et al., 2010) with the Roche-454 sequencing adapters. Robust PCR products were obtained from the 17 duodenal digesta samples and 17 mixed ruminal bacterial samples. The PCR products were quantified based on their staining intensity using the image acquisition software Genesnap (Syngene USA, Frederick, MD). The PCR products were combined in equal amounts and gel-purified using the QIAquick Gel Extraction Kit (Qiagen). Pooled and gel-purified amplicon products were sequenced using the Roche-454 GS FLX Titanium chemistry at University of Nebraska CAGE facilities. Raw reads were filtered according to length and quality criteria. The resulting high quality reads, which amounted to a total of 462,669 sequences, were parsed into sample-barcoded bins and subsequent analysis was completed using the Quantitative Insights Into Microbial Ecology (QIIME) bioinformatic pipeline (Caporaso et al., 2010). Briefly, chimeras were removed and sequences clustered into operational taxonomic units (OTU) at 97% sequence similarity using UCHIME (Edgar et al., 2011) and USEARCH (www.drive5.com/usearch/), respectively. Representative sequences from each OTU were then subjected to taxonomic classification using the Ribosomal Database Project “CLASSIFIER” (Wang et al., 2007) algorithm (trained using Greengenes database 12_10). To minimize animal to animal variation and to represent the shared OTU within each diet, a core measurable microbiome (CMM) was defined for each diet within each sampling site. A CMM allows identification of the microbial community influenced by the treatment sorting through animal to animal variation. The hypothesis is that, if the treatment affects the microbial community, this effect should be present across multiple animals on the same treatment. Therefore, the CMM allows identification of the effects that might otherwise be hidden in the data (Benson et al., 2010). In the current experiment we were unable to collect digesta samples from 1 steer consuming LOW DDGS during the last experimental period. An OTU was considered as part of the CMM if a given OTU was present in 66% of the samples collected from the rumen or duodenum of animals fed the CONTROL or HIGH DDGS diets or in 60% of the samples collected from the rumen or duodenum of animals fed the LOW DDGS diet. Collectively, the CMM was composed of 2485 OTU. The abundance of previously reported butyrate producers and microbial species identified in biohydrogenation (Polan et al., 1964; Verhulst et al., 1985) were selected from the CMM for further analysis. Both the Ribosomal Database Project (Cole et al., 2009) classifications and BLAST (MegaBLASTN, default parameters) were used for identification at the species level. Only the top hit with significant e-values were considered in identification of species identity.

Statistical Analysis

Data were analyzed with the MIXED procedure of SAS (Version 9.1; SAS Inst. Inc., Cary, NC) as a 3-diet, 6-period crossover design assuming diet and period as fixed effects and steer as a random effect. Using the CONTRAST statement of SAS, linear effects of diets were tested. Diet means are presented as least square means, and the largest SEM is reported. Significance was declared at P ≤ 0.05, and tendency was declared if 0.05 < P ≤ 0.15.


RESULTS AND DISCUSSION

Animals and Ruminal Fluid pH

During the last period, 1 animal was removed from the experiment because of a dramatic decrease in DMI and a suspected health problem. We did not statistically account for carry-over effects which could influence responses in the next period (Morris, 1999); doing so would require a more complex statistical model. To maintain the statistical power using such an approach would require the use of more animals, which was not practical. We are reasonably confident that carry-over effects were minor because: 1) mature animals were used to ensure a stable physiological state across the experiment; 2) diets met or exceeded nutrient requirements; and 3) a 15-d adaptation period was used to allow both the animal and rumen microflora to adequately adapt before samples were collected.

Table 3 lists ruminal fluid pH measurements. In addition, Fig. 1 illustrates pH averaged by diet across d 16 through 19 of experimental periods. Mean ruminal fluid pH was not affected (P = 0.88) by amount of DDGS and averaged 6.4 ± 0.09 across diets. In addition, the time (P = 0.68) and area (P = 0.98) below pH 6.4 were not affected and averaged 567 ± 362 min/d and 80 ± 61 pH × min/d, respectively. Ruminal fluid pH was highest just before feeding and declined for approximately 4 to 5 h before gradually increasing to pre-feeding values. Overall, ruminal fluid pH in this study was in the normal range and should not have had detrimental effects on ruminal fermentation and ruminal microbial growth (Russell et al., 1979; Russell and Rychlik, 2001).


View Full Table | Close Full ViewTable 3.

Intake, duodenal flow, ruminal nutrient digestibilities and ruminal fluid pH of steers fed increasing amounts of dried distillers grains with solubles (DDGS)

 
Diet1
Item CONTROL LOW DDGS HIGH DDGS SEM P-value,Linear
Intake, g/d
    DM 10,500 10,500 10,500
    CP 1,107 1,249 1,368 10.3 < 0.01
    NDF 6,955 6,607 6,259 7.6 < 0.01
    Starch 602 487 370 4.8 < 0.01
Duodenal flow, g/d
    DM 5,406 5,305 5,541 1,025 0.87
    CP 953 1,030 1,168 201 0.12
    NDF 3,012 2,863 2,862 560 0.74
    Starch 36 44 46 7.1 0.26
Ruminal digestibility, %
    DM 50.8 54.9 53.0 4.7 0.73
    CP 53.3 50.3 40.6 6.1 0.12
    NDF 56.8 59.3 54.3 4.7 0.41
    Starch 93.9 90.6 87.4 1.6 0.01
Ruminal fluid pH
    Minimum 6.13 6.18 6.16 0.12 0.86
    Maximum 6.62 6.61 6.66 0.07 0.65
    Mean 6.41 6.37 6.42 0.09 0.88
Time
    pH < 6.4, min/d 493 520 688 362 0.68
Area
    pH < 6.4, pH × min/d 76 86 77 61 0.98
1CONTROL contained (DM basis) no DDGS, 19.5% corn bran, 20% sorghum silage, 60% brome hay, 0.5% trace minerals, and 0.25% urea. LOW DDGS contained (DM basis) 9.75% DDGS replacing an equal percentage of corn bran. HIGH DDGS contained (DM basis) 19.5% DDGS completely replacing corn bran.
Figure 1.
Figure 1.

Ruminal fluid pH variation with time after feeding for steers fed twice daily and consuming CONTROL (0% diet DM; dotted line), LOW dried distillers grains with solubles (DDGS; 9.5% of diet DM; solid black line) and HIGH DDGS (19.5% of diet DM; solid grey line). 1Feeding times were 0 h (0700) and 12 h (1900).

 

Intake, Duodenal Flow, and Ruminal Digestibility of Nutrients

Table 3 lists intakes, duodenal flows, and ruminal digestibilities of DM, CP, NDF, and starch. By design, average DMI was similar (10.5 kg/d) for all diets. Intake of CP increased (P < 0.01) with DDGS inclusion, but intake of NDF and starch decreased (P < 0.01) with increasing amounts of DDGS. The dietary inclusion of DDGS had no effect (P < 0.87) on duodenal DM flow, with an average of 5.42 ± 1.0 kg/d. However, total duodenal CP flow tended (P = 0.12) to increase with increasing DDGS. Duodenal flows of NDF (P = 0.74) and starch (P = 0.22) were not affected by DDGS inclusion amount. The inclusion of DDGS in diets did not affect the ruminal digestibility of DM (P = 0.73) and NDF (P = 0.41) with averages across diets of 52.9 ± 4.7% and 56.8 ± 4.7%, respectively. However, the ruminal digestibility of CP (P = 0.12) tended to decrease with means of 53.2, 50.3, and 40.6 ± 6.1% for CONTROL, LOW DDGS, and HIGH DDGS. In addition, the ruminal digestibility of starch tended to decrease (P = 0.11), means were 93.9, 90.6, and 87.4 ± 1.5 for CONTROL, LOW DDGS, and HIGH DDGS, respectively.

Similar duodenal flows of DM were expected because there were no differences in DMI across diets. Our observations of ruminal DM digestibility are comparable to Faulkner et al. (1985), who reported that ruminal DM digestibilities ranged from 52.5 to 55.8% for steers consuming 80% cornstalks and 20% supplement. The increase in duodenal CP flow with increasing DDGS in diets was likely due to greater CP content in DDGS-containing diets compared to CONTROL (10.6, 11.9, and 13.0 ± 1.2% CP for CONTROL, LOW DDGS, and HIGH DDGS, respectively) and due to more RUP supplied by DDGS compared to corn bran (Herold, 1999). With the increase of DDGS in diets, the proportion of duodenal CP represented by microbial protein decreased (Castillo-Lopez et al., 2013). The decrease in NDF and starch intake with increasing amounts of DDGS is attributed to the lower concentrations of these nutrients in DDGS compared to corn bran (Table 2). Similar to our findings, ruminal digestibility of NDF has been reported to range from 55 to 62% for steers consuming a diet consisting of 69% fescue hay and 30% concentrate (Lardy et al., 1993).

Intake, Duodenal Flow, and Ruminal Biohydrogenation of Fatty Acids

Intake, duodenal flow, and ruminal biohydrogenation of fatty acids are listed in Table 4. Total fatty acid intake increased (P < 0.01) with increasing amounts of DDGS in diets. Likewise, intake of total C18 fatty acids consisting of C18:0, C18:1, C18:2, and C18:3 increased (P < 0.01) when DDGS was included in diets.


View Full Table | Close Full ViewTable 4.

Intake and duodenal flow of total and C18 fatty acids, and biohydrogenation of unsaturated fatty acids for steers fed increasing amounts of dried distillers grains with solubles (DDGS)

 
Diet1
Item CONTROL LOW DDGS HIGH DDGS SEM2 P-value, Linear
Intake, g/d
    Total fatty acids 171.2 228.0 283.5 57.7 < 0.01
    C18:0 4.2 5.1 6.3 1.2 < 0.01
    C18:1n9 17.0 29.2 43.0 7.6 < 0.01
    C18:2n6 37.8 69.6 100.8 17.5 < 0.01
    C18:3n3 13.6 14.6 15.7 3.4 0.01
    Total C183 72.4 118.7 165.9 29.7 < 0.01
Duodenal flow, g/d
    Total FA 133.8 167.8 223.1 32.9 < 0.01
    C18:0 50.5 86.4 121.2 17.9 < 0.01
    Trans C18:1 6.5 10.8 19.2 2.6 < 0.01
    C18:1n9 5.6 7.1 9.2 1.2 0.01
    C18:2n6 6.0 5.9 6.7 1.1 0.62
    C18:3n3 2.2 2.0 2.1 0.3 0.35
    Total C184 70.8 112.7 158.6 22.3 < 0.01
Biohydrogenation, %
    C18:1n95 68.3 75.6 79.3 4.3 < 0.01
    C18:1n96 67.2 75.4 78.1 3.3 < 0.01
    C18:2n65 84.1 91.5 93.3 3.4 0.02
    C18:2n66 83.5 90.8 93.0 1.6 < 0.01
    C18:3n35 82.8 86.3 84.7 1.6 0.34
    C18:3n36 84.5 85.1 84.8 1.0 0.76
1CONTROL contained (DM basis) no DDGS, 19.5% corn bran, 20% sorghum silage, 60% brome hay, 0.5% trace minerals, and 0.25% urea. LOW DDGS contained (DM basis) 9.75% DDGS replacing an equal percentage of corn bran. HIGH DDGS contained (DM basis) 19.5% DDGS completely replacing corn bran.
2The largest SEM among diets is shown.
3Consisting of C18:0, C18:1n9, C18:2n6, and C18:3n3.
4Consisting of C18:0, trans C18:1, C18:1n9, C18:2n6, and C18:3n3.
5Percentage biohydrogenation based on total grams of fatty acid disappearing between the mouth and duodenum divided by the fatty acid intake (Lundy et al., 2004; Loor et al., 2004).
6Percentage biohydrogenation as calculated by Wu et al. (1991). Bio-hydrogenation (%) = 100– 100 × (individual unsaturated C18/total C18 in duodenal digesta)/(individual unsaturated C18/total C18 in feed).

Duodenal flow of total fatty acids increased (P < 0.01) when DDGS were included in diets. Likewise, duodenal flow of total C18 fatty acids consisting of C18:0, C18:1, C18:2, C18:3, and trans C18:1 increased (P < 0.01). However, duodenal flow of C18:2 (P = 0.62) and C18:3 (P = 0.35) was not affected by increasing amounts of DDGS, with averages across diets of 6.2 ± 1.1 g/d and 2.1 ± 0.3 g/d, respectively. Furthermore, duodenal flow of trans C18:1 fatty acids, consisting of t6/t8 c18:1; t9 c18:1; t10 c18:1; t11 c18:1, and t12 c18:1, increased (P < 0.01) with the addition of DDGS to diets; mean flows of trans C18:1 fatty acids were 6.5, 10.8, and 19.2 ± 2.6 g/d for CONTROL, LOW DDGS, and HIGH DDGS, respectively.

Extensive ruminal trans fatty acid formation in the rumen has been reported elsewhere in steers (Vander Pol et al., 2009) and dairy cattle (Janicek et al., 2008) consuming DDGS. In agreement with our observations, reports indicate that trans fatty acids are formed during biohydrogenation of dietary PUFA in the rumen, predominantly C18:2 (Harfoot, 1978; Tanaka and Shigeno, 1976). Despite differences in the intake of C18:2, duodenal flow of C18:2 was similar across diets, and this observation reflects its intensive microbial metabolism and biohydrogenation in the rumen.

When ruminal biohydrogenation was calculated using total grams of fatty acid disappearing between the mouth and duodenum, biohydrogenation of C18:1 increased (P < 0.01) with the addition of DDGS to diets, means were 68.3, 75.6, and 79.3 ± 4.3% for CONTROL, LOW DDGS, and HIGH DDGS, respectively. In addition, biohydrogenation of C18:2 increased (P = 0.02) with the inclusion of DDGS in diets; means were 84.1, 91.5, and 93.3 ± 3.4% for CONTROL, LOW DDGS, and HIGH DDGS, respectively. However, biohydrogenation of C18:3 was not affected (P = 0.34) by diet and averaged 84.6 ± 1.6% across diets. Likewise, when ruminal biohydrogenation was calculated using an approach that corrects for losses in C18 flow to the duodenum, biohydrogenation of C18:1 increased (P < 0.01) with increasing amounts of DDGS with means of 67.2, 75.4, and 78.1 ± 3.3% for CONTROL, LOW DDGS, and HIGH DDGS, respectively. In addition, biohydrogenation of C18:2 increased (P < 0.01), but biohydrogenation of C18:3 was not affected (P = 0.76) and averaged 84.8 ± 1.0% across diets.

The increased intake of total fatty acids and total C18 fatty acids, most notably unsaturated fatty acids, with the inclusion of DDGS was a result of the higher concentration of fat contained in DDGS compared to corn bran (Table 2). During the dry milling process, after removing the starch for ethanol production (Spiehs et al., 2002; Xu et al., 2010), the concentration of fat in DDGS reaches values of approximately 12% (Paz et al., 2013). As a consequence, when DDGS was included in the diets, the high concentrations of unsaturated fatty acids available in the rumen, notably C18:1 and C18:2, likely stimulated an increase in fatty acid biohydrogenation, and as a consequence an increase in the flow of saturated fatty acid (C18:0) to the duodenum. The reason for the lack of an effect of DDGS inclusion on C18:3 biohydrogenation is unclear, but Beam et al. (2000) suggested that the different extent and rate of biohydrogenation of unsaturated C18 fatty acids may reflect the activity of different microbial enzymes. Our results agree with previous reports (Duckett et al., 2002; Jenkins et al., 2008; Duckett and Gillis, 2010) in that the inclusion of unsaturated fatty acids in rations formulated for beef cattle increases the intake and ruminal biohydrogenation of unsaturated fatty acids. Further studies on ruminal biohydrogenation are warranted because the ration composition and the fatty acid metabolism by the ruminal microbial community affect the fatty acid composition of ruminant food products (Palmquist et al., 1993; Beam et al., 2000; Jenkins et al., 2008; Abdelqader et al., 2009).

Bacterial Diversity

Results obtained through the Roche-454 pyro-sequencing platform were utilized to evaluate the effect of diet on the bacterial community structure of the mixed ruminal bacterial samples and the duodenal digesta samples. The use of pyrosequencing has been reported to be effective for phylogenetic analysis at the genera (Callaway et al., 2010) and species levels (Vahjen et al., 2011). A total of 361,726 reads were generated after quality control and chimera removal, which resulted in an average of 10,639 reads per sample. The distribution of phyla and sub-phyla among the diets revealed no clustering by diet within each location (rumen or duodenum). However, significant structuring of the microbial community was seen by location where the rumen and the duodenal microbial communities clustered separately (Fig. 2). The microbial communities within the rumen and duodenum were dominated by Bacteroidetes and Firmicutes (Table 5). In addition to these 2 phyla, phylum Fibrobacteres had high numbers in the duodenum, but was very low in the rumen.

Figure 2.
Figure 2.

Principal Coordinate Analysis (PCoA) of the core measureable microbiome (CMM) revealing separate clusters between the bacterial communities found in mixed ruminal bacteria collected from whole ruminal contents and duodenal digesta samples of steers consuming CONTROL, LOW dried distillers grains with solubles (DDGS), and HIGH DDGS; rum: mixed ruminal bacterial samples, duo: duodenal digesta samples.

 

View Full Table | Close Full ViewTable 5.

Percentage of dominant Operational Taxonomic Units (OTU) and phyla in the core measurable microbiome (CMM) of mixed ruminal bacteria collected from whole ruminal contents and duodenal digesta of steers increasing amounts of dried distillers grains with solubles (DDGS)

 
Diet1 and location2
OTU Classification CONTROL LOW DDGS HIGH DDGS P-value
Identification Phylum Genus/species Rum Duo Rum Duo Rum Duo SEM3 Linear Interaction4
Most abundant
    19 Fibrobacteres Fibrobacter succinogenes 0.02 2.08 0.03 1.31 0.01 1.53 0.40 0.41 0.52
    102,674 Bacteroidetes Prevotella sp. 0.73 0.16 0.51 0.25 1.30 0.23 0.48 0.47 0.68
    21 Firmicutes Succiniclasticum sp. 0.46 0.48 0.26 0.31 0.30 0.33 0.17 0.18 0.90
    22 Fusobacteria Fusobacterium necrophorum 0.00 0.73 0.00 0.49 0.01 1.26 0.46 0.50 0.67
    24 Firmicutes Ruminococcus flavefaciens 0.14 0.60 0.11 0.05 0.18 0.14 0.28 0.38 0.50
    123 Bacteroidetes Paludibacter sp. 0.02 2.08 0.03 1.31 0.01 1.53 0.40 0.41 0.52
    79 Firmicutes Anaerostipes sp. 0.06 0.07 0.11 0.19 0.22 0.70 0.18 0.02 0.32
Biohydrogenating
    67 Firmicutes Butyrivibrio sp. 0.22 0.03 0.11 0.05 0.20 0.04 0.03 0.76 0.12
    820 Firmicutes Clostridium sp. 0.12 0.07 0.19 0.01 0.10 0.03 0.04 0.52 0.28
Total number of OTU 920 868 1,120 911 938 792 86 0.67 0.54
Total Bacteroidetes 44.10 40.77 43.75 41.01 44.79 44.26 4.58 0.62 0.94
Total Firmicutes 44.83 31.42 47.78 29.92 45.77 33.13 3.57 0.71 0.76
Total Fibrobacteres 0.06 10.93 0.13 11.95 0.16 7.28 0.56 < 0.01 < 0.01
Other phyla5 10.99 16.86 8.31 17.10 9.27 15.31 1.69 0.29 0.61
Ratio of Firmicutes:Bacteroidetes 1.12 0.82 1.27 0.73 1.26 0.76 0.25 0.87 0.86
1CONTROL contained (DM basis) no DDGS, 19.5% corn bran, 20% sorghum silage, 60% brome hay, 0.5% trace minerals and 0.25% urea. LOW DDGS contained (DM basis) 9.75% DDGS replacing an equal percentage of corn bran. HIGH DDGS contained (DM basis) 19.5% DDGS completely replacing corn bran.
2Rum: rumen, Duo: duodenum.
3The largest SEM among diets is shown.
4Interaction of diet by location.
5Other phyla include unclassified phyla, Actinobacteria, Armatimonadetes, Chlorolexi, Cyanobacteria, Elusimicrobia, Fusobacteria, Lentisphaerae, Planctomycetes, Protoebacteria, SR1, Spirochaetes, Synergistetes, TM7, Tenericutes, Verrucomicrobia and WPS-2.

In addition, the analysis at 97% nucleotide similarity of the V1-V3 sequences revealed the presence of 2,485 OTU within the CMM, where 894 ± 64, 1,015 ± 68, and 865 ± 64 OTU were present for CONTROL, LOW DDGS, and HIGH DDGS, respectively. Table 5 lists the percentage of dominant OTU in each diet for each sampling site.

Effect of Diet on the Bacterial Community.

The analysis revealed no effect of diet on the presence of Bacteroidetes (P = 0.62), Firmicutes (P = 0.71), or the ratio of Firmicutes:Bacteroidetes (P = 0.87). However, the phylum Fibrobacteres was negatively affected (P < 0.01) by the inclusion of DDGS with means averaged across sampling sites of 5.51 ± 0.36, 6.04 ± 0.40, and 3.72 ± 0.36% for CONTROL, LOW DDGS, and HIGH DDGS, respectively. The presence of the majority of dominant OTU were not affected by diet, and belonged to Fibrobacter succinogenes (P = 0.41), Prevotella sp. (P = 0.47), Succiniclasticum sp. (P = 0.18), Fusobacterium necrophorum (P = 0.50), and Ruminococcus flavefaciens (P = 0.38). These results agree with Russell (2002), who indicated that Fibrobacter succinogenes, Prevotella sp., and Ruminococcus flavefaciens are among the predominant bacteria in the rumen. Our observations, however, contrast the reports of Callaway et al. (2010) and Ramirez-Ramirez et al. (2012) who observed a decrease in the ratio of Firmicutes:Bacteroidetes in cattle consuming increasing amounts of DDGS. We suggest that in these cases the decrease in the ratio was observed because of the higher inclusion rates of DDGS in the rations fed by those researchers, specifically 50 and 30% of the diet DM. Another factor that may have contributed to this difference is the ingredient composition of diets, more specifically the higher amounts of concentrate and higher levels of fat used by Ramirez-Ramirez et al. (2012) and Callaway et al. (2010). Our results suggest that the phylum Fibrobacteres was more abundant in diets containing corn bran compared to diets containing DDGS. In spite of the observed reduction of the phylum Fibrobacteres, the presence of the cellulose digesting bacteria Fibrobacter succinogenes (Stewart and Bryant, 1988) and Ruminococcus flavefaciens (Fondevila and Dehority, 1996) was not affected with increasing amounts of DDGS; interestingly the digestion of NDF was similar across diets (Table 3).

In the current study, similar to the observations of Aldai et al. (2012), in which increasing amounts of corn and wheat-based DDGS were used, and Ramirez-Ramirez et al. (2012), where 2 amounts of corn-based DDGS were used, the abundance of Butyrivibrio sp. (P = 0.77) and Clostridium sp. (P = 0.52) were not affected by diet. This indicates that biohydrogenation in the rumen can be maintained or increased without affecting the abundance of the bacteria involved in this process.

Bacterial Community of the Rumen Compared to the Duodenum.

Microbial communities from samples collected in the rumen and duodenum clustered separately (Fig. 2). More specifically, the phylum Fibrobacteres was lower (P < 0.01) in the rumen compared to the duodenum, with means averaged across diets of 0.13 and 10.1 ± 0.56% for the rumen and duodenum, respectively (Table 5). Bacteria identified in intestinal contents could originate from lysed cells as they pass through the abomasum (Waghorn et al., 1990; Koenig et al., 1997; Hristov, 2007), and the function of the duodenal population is likely different than that of the ruminal bacterial community in terms of its role in feed digestion.

The clustering of the bacterial communities by location suggests differences in phylogenetic composition between isolated mixed ruminal bacteria and duodenal digesta. It is important to emphasize that this observation may be confounded by the methodology utilized during sample collection. Specifically, the differential centrifugation (Hristov et al., 2005) used for the isolation of mixed ruminal bacterial samples would lead to samples that contain bacteria found in the liquid fraction and bacteria loosely attached to feed particles. However, the isolation protocol likely excluded bacteria tightly associated to feed particles, which may have contributed to the lower Fibrobacteres found in mixed ruminal bacteria. To fully establish if these differences occur, in vivo future research should involve the extraction of bacterial DNA directly from whole ruminal contents (Fernando et al., 2010) and this would likely be a better approach to describe the bacterial community flowing from the rumen compared to bacterial DNA extracted from ruminal fluid (Callaway et al., 2010; Ramirez-Ramirez et al., 2012).

Conclusions

Overall, the inclusion of DDGS in the diets fed to steers increased ruminal biohydrogenation of unsaturated fatty acids and the flow of saturated fatty acid to the duodenum. Contrasting our expectations, the presence of the 2 largest phyla, Bacteroidetes and Firmicutes, were unaffected by dietary DDGS inclusion. However, the phylum Fibrobacteres was negatively affected by DDGS inclusion when averaged across sampling sites. In addition, dominant OTU in the CMM of the gut were not affected by diet. However, we detected that the bacterial community of the rumen clustered separately from that of the duodenum. It is possible that bacterial OTU detected in lower proportions in ruminal samples are those found tightly associated with feed particles, and therefore were not isolated during the collection of mixed ruminal bacterial samples from whole ruminal contents.

 

References

Footnotes


Comments
Be the first to comment.



Please log in to post a comment.
*Society members, certified professionals, and authors are permitted to comment.