Background

Rapid expansion of fuel ethanol production has made available abundant supplies of distiller’s grains with solubles, which are well-suited as a substitute for cereal grains in finishing cattle diets. Several recently reported experiments have revealed that feeding distiller’s grains may have adverse effects on carcass value as a result of the tendency to produce carcasses with lower quality grades and(or) higher yield grades. The effects on quality grade have been most evident in flaked grain diets, but effects on yield grade are more-or-less independent of the type of grain that is fed. This may be a result of substituting starch as an energy source with non-starch carbohydrates, fats, and proteins when cereal grains are replaced by distiller’s grains. Researchers at Texas A&M University have proposed that the accumulation of intramuscular fat (i.e., marbling) may be influenced by the presence of energy substrates that yield glucose, as it appears that glucose may be the preferred energy source for the fat-producing cells that are located in this region of the muscle tissue. Starch is the primary energy substrate in grain, and when it is digested in the intestine it yields large quantities of glucose. This is not the case when distiller’s grains are substituted for starch-based feeds.

The manner in which grains are processed before feeding also has a significant impact on the proportion of starch that is converted to glucose. It is well-documented that steam flaking results in extensive digestion of starch within the rumen compared to dry rolling. Dry-rolled corn is less extensively digested in the rumen, thus allowing for increased starch supply to the intestine. This increases glucose absorption from the small intestine, which is believed to be directly tied to the synthesis of intramuscular fat. It is conceivable that partial substitution of steam-flaked corn with dry-rolled corn would increase glucose supply for intramuscular fat deposition, thus compensating for the reduced digestion in the rumen and improving quality grade when distiller’s grains are added to the diet.

Methodology

Yearling heifers (n=689) with an average initial body weight of 668 lbs were blocked by weight and randomly allotted, within block, to each of 28 feedlot pens. Experimental treatments consisting of four different feedlot diets were assigned randomly to pens such that each treatment appeared once within each group of 4 adjacent feedlot pens. Feedlot pens provided 180 ft2 of pen surface area per animal, and were equipped with fence-line feed bunks that provided 12-14 linear inches of bunk space per animal. All cattle were fed a series of step-up diets during the first 15-20 days after arrival in the feedlot (June, 2007). The final finishing diets (Table 1) were fed for the remaining 140 days of the finishing period (through October, 2007). The ingredient compositions of final treatment diets are shown in Table 1. Nearly half of the pens of cattle (12 of 28 pens) were sent to slaughter after 137 days on feed, while all other pens of cattle remained until 157 days on feed.

On the day of harvest, pens were weighed, loaded onto trucks and transported approximately 90 miles to a commercial abattoir in Emporia, Kansas. The incidence and severity of livers abscesses and hot carcass weight were determined on the day of harvest. Following a 24-hour chill, USDA yield grade; USDA quality grade; 12th-rib fat thickness; percentage kidney, pelvic and heart fat; rib-eye area; and marbling were determined for each carcass. 

At each slaughter point, 4 cattle were selected at random from each of 12 feedlot pens, and full rib sections (with the rib plate attached) were obtained from the right side of each carcass for subsequent evaluation of sensory attributes, retail display life, and compositional analyses. The ribs were individually identified and transported to the KSU Meats Laboratory for fabrication. Identity of two ribs was lost in transit, so the sensory and compositional analyses were limited to 94 carcasses.

Separation of the rib was performed as described by Hankins and Howe (U.S. Dept. of Agriculture, Technical Bulletin, No. 926, 1946) to yield 9th-10th-11th rib sections. These rib sections were blocked into groups of four, one from each treatment, and then physically separated into lean, fat, and bone. The separated portions were weighed, vacuum packed, and frozen for later chemical analysis. The separated lean and adipose portions were ground twice, then frozen with liquid N2, and pulverized to a powdery consistency to homogenize the sample. These samples were then used to evaluate fatty acid profiles of triglycerides and phospholipids, protein, moisture, ash, and vitamin E concentrations.

The fatty neutral and polar lipids were separated according to the procedures described by Noci et al. (J. Anim. Sci., 2005; 83-1167-1178). Separation and quantification of fatty acid methyl esters in the separated portions was conducted on a flame ionization gas chromatograph (model 5890 series II, Hewlett Packard) fitted with a Supelco 2560 capillary column (100 m x 0.25 mm x 0.20 μfilm). Helium was used as the carrier gas, with a flow rate of 1.1 ml/min, initial temperature of 140◦C, followed by a 4◦C/min temperature increase to reach the final temperature of 240◦C. The final temperature was held for 15 minutes, while the injection and detector temperatures were maintained at 260◦C. The concentrations of fatty acids ranging from C8 to C24 were determined. Also from the pulverized lean and adipose tissue moisture, protein and ash were determined. Protein and moisture fractions were determined by AOAC official methods of analysis. Ash was determined on both portions by methods described by the AOAC official methods of analysis. Samples of the lean separated portion were sent to DSM Nutritional Products for determination of tocopherol (vitamin E) concentrations.

The Longissimus muscles were removed from the 6th, 7th, and 8th rib sections, weighed, vacuum packed, and wet aged for two weeks. At the conclusion of the 2-week aging process, the samples were patted dry with absorbent towels and weighed again to evaluate purge loss during storage. From these samples, steaks (1- in thick) were cut from most posterior end and used for the retail display study. Steaks were placed cut side up on 17 S white foam trays with a Dri Loc pad (Dri-Loc, Cryovac Sealed Air Corporation, Duncun, SC) and wrapped with PRV film (MAPAC M [23250 cc/m², 24 hrs, 72 gauge], Bordon Packaging and Industrial Products, North Andover, MA). The steaks were placed in a retail display case with a temperature of 3±1.6◦C for seven days. The display lighting consisted of 1614 ± 53.82 1x (45.72 ± 1.52 m candles) light intensity, 40 W Del Warm White 3000◦K (Phillips Lighting Company, Somerset, NJ). Instrument color was evaluated for CIE L * a * b* values for illuminant A and reflectance from 400 to 700 nm at 10 nm increments using a Hunter Miniscan XE Spectrophotometer (3.18 cm diameter aperture, 10◦ observer; Hunter Associates Laboratory, Reston, VA). On each day of evaluation, readings were taken from three locations on the Longissimus muscle, being careful to avoid intramuscular fat depots. Readings were averaged for statistical purposes. Steaks were rotated left to right and front to back twice daily. Once the 7-day display period was completed, the steaks were analyzed for lipid oxidation using a modified TBARS procedure (Witte, 1970). The steaks were chopped, then frozen with liquid N2 and pulverized to a powdery consistency in a Waring blender (Waring Products Division, Hartford, CT). In duplicate, 10 g of the frozen pulverized sample were weighted into mini-Waring blender cups (100 mL capacity). Fifteen mL of 7.2% cold perchloric acid and 20 mL cold distilled de-ionized water were added to precipitate the protein and extract malonaldehyde from the sample. Samples were then blended for 15 s and gravity filtered through Whatman no. 2 filter paper (Whatman International LTD., Maidstone, UK). Five mL 0.02 M thiobarbituric acid reagent solution (1.4415 g TBA and 500 mL de-ionized water) were then added to the filtrate and, mixed. The resulting mixture was stored for 18 to 24 h in complete darkness at room temperature to allow the color reaction to develop, after which absorbance was measured at 530 nm on a spectrometer (Spectronic 21, Bausch & Lomb, Rochester, NY). The resulting values were reported as mg of malonaldehyde/kg of steak.

Steaks (1-inch thick) also were cut to be used for sensory analyses, which were performed by the Sensory Analysis Center in the Department of Foods and Nutrition, Kansas State University, Manhattan, KS. The sensory steaks were evaluated for 10 attributes: initial tenderness; juiciness; chewiness; mealy texture; fiber awareness; residual connective tissue; beef flavor; blood/serumy flavor; metallic flavor, and rancid flavor. Each steak was evaluated by a five-member, professional panel using a 15-point scale graded in increments of 0.5. For flavor profiling, the steaks were thawed and cooked to an internal temperature of 160 ◦F on an electric broiler and cut into ½ x ½ x 1 inch samples. The samples were presented along with reference samples. Also, from these same steaks, cooking loss was measured by taking an initial raw weight and then weighing the cooked steak.

Formation of heterocyclic amines was achieved by cooking rib steaks (1-in thick) on a Teflon® covered electric grill with a temperature controller (Toastmaster, Denver, CO) at 400◦F for 5.0 minutes per side. Temperature profile of the grill surface was measured using a surface probe thermometer (Barnant Company, Barrington, IL) prior to the cooking. The steaks were placed in the middle section of the grill each time to ensure they were cooked under similar conditions and temperature. The internal temperature profile of the steaks was monitored using a thermocouple thermometer (Barnant Company, Barrington, IL). After cooking, the steaks were chilled in the refrigerator and 2 mm of the surface was sliced off with a commercial grade meat slicer. The surface material was then ground and used for extraction of heterocyclic amines (HCA). The ground samples were stored frozen (-18◦C) if not assayed immediately. Three grams of the previously ground samples were homogenized in 12 ml of 1 M sodium hydroxide and mixed thoroughly with Extrelut refill material and loaded onto an empty Extrelut column. Bond elut PRS tubes were coupled to the Extrelut column and the HCAs were eluted to PRS with 60 mL ethyl acetate using a Supelco Visiprep SPE vacuum manifold (Sigma Aldrich St. Louis MO). The PRS tubes were preconditioned with 7 mL ethyl acetate. The PRS was dried under a stream of nitrogen and rinsed with 6 mL of 0.1 M HCl, 15 mL of methanol/0.1 M HCl (45:55 v/v) and 2 mL of distilled water. The PRS tubes were then coupled to 100 mg C-18 tubes which had been pre conditioned with 1 mL methanol and 10 mL water. The HCAs were concentrated on the C-18 tubes by passing 20 mL of 0.5 M ammonium acetate (pH 8) through the PRS tubes. The C-18 tubes were then rinsed with 2 mL of distilled water and dried under a stream of nitrogen. The HCAs were eluted from the C-18 tubes into 4 mL vials with 1 mL of methanol/ammonium hydroxide (9:1, v/v), concentrated until dry, and dissolved in 25 μL of methanol. The flow rate was 1 mL/min throughout the extraction. The Extrelut columns and refill material were obtained from VWR (West Chester, PA) and the PRS and C-18 tubes from Varian Inc (Palo Alto CA). The final extract (25 μL) was analyzed on HP1090A, series II HPLC (Agilent Technologies, Palo Alto, CA) coupled with a photodiode array UV-visible detector (HP 1040) and an HP 1046A programmable fluorescence detector. The column used was TSK gel ODS-80 TM column (25 cm x 4.6 mm x 5 μm, Tosohass, Montgomeryville, PA) with a mobile phase of 0.01 M triethylamine pH 3.6 (A) and acetonitrile (B). The HCAs were separated using a linear gradient starting with 95% A, 5% B, change to 75% A, 25% B in 30 min, flow rate of 1 mL/min at a column temperature of 40°C. After 30 min, the mobile phase returned to its original ratio (95% A, 5% B) for 10 min to allow the column to re-equilibrate before the next injection. The UV detector was set at 252 nm for IQx, and MeIQx, while the fluorescence detector was programmed accordingly to the excitation/emission wavelengths of 229 and 437 for PhIP. The HCA were confirmed by comparing the retention times and the UV absorbance spectrum of each peak with retention times and library spectra acquired from standard solutions.

Statistical Analysis
Data were analyzed using the Mixed Procedure of the Statistical Analysis System (Version 9.1, Cary, NC) with the fixed effects of weight block, DDGS, DRC, and the interaction between DRC and DDGS.

Findings  

Feedlot Performance and Carcass Characteristics
Performance and carcass quality data are shown in Table 2. The two weight blocks of heifers were fed for 137 and 157 days, respectively. We found that average daily gain, feed intake, and feed conversion efficiency were similar among treatment groups (P > 0.10). Likewise, there were no differences among treatments with respect to quality grade, yield grade, 12th rib fat thickness, KPH, incidence of liver abscess, or total carcass value. There was a tendency for cattle fed DRC to produce heavier carcasses (721 vs 728 lb for 0 and 25% DRC, respectively; P < 0.12). Likewise, cattle fed dry-rolled corn had larger rib-eye size (P < 0.10) compared to their counterparts without DRC. Total monetary value of carcasses was not different among treatments (P > 0.30), though values were numerically lower for cattle fed distiller’s grains.

Carcass Composition
In addition to routine carcass measurements, we evaluated carcass composition, sensory attributes, storage characteristics, lipid peroxidation, fatty acids profiles, and vitamin E concentrations of tissues using a subset of 94 animals. Composition of the 94 carcasses used in this portion of the experiment are shown in Table 3. The attributes of carcasses selected for compositional assays closely reflect those of the larger population.

Physically separated components of the 9th-10th-11th rib sections are summarized in Table 4. We evaluated both the actual separated components of the rib section, as well as the carcass percentages of various components that were predicted using the regression equation of Hankins and Howe (1946). Overall carcass fat was approximately 28.5% when averaged across treatments. There were no differences among treatments with respect to separable portions of lean, fat, and bone (P > 0.10). The percentages of protein, moisture, and ether extract also were similar among diets. Using the equations of Hankins and Howe (U.S. Dept. of Agriculture, Technical Bulletin, No. 926, 1946), we found that the percentages of lean, bone, and adipose tissue of the total carcass were not different among treatments. 

Sensory Attributes
Longissimus steaks from the 6th, 7th and 8th rib sections were wet aged for two weeks and then used for sensory analyses by a trained panel. We found no differences among treatments in terms of initial tenderness, juiciness, fiber awareness, residual connective tissue, bloody/serumy flavor, or rancid flavor. Overall, we found that steaks from cattle fed DDGS were not noticeably different from those derived from cattle fed diets without DDGS (P > 0.10). We did observe that DRC tended to influence several sensory attributes, but the magnitude of these differences is likely not of practical importance.

Retail Display Lift and Lipid Oxidation
Longissimus steaks representing the different dietary treatments yielded comparable color changes throughout a 7-day simulated retail display period (Figures 1, 2, 3, and 4). There were no significant interactions between levels of DRC and DDGS (P > 0.01), and steaks derived from cattle fed diets with 0 and 25% DRC did not differ with respect to L*, a*, b*, or saturation indices (P > 0.10). Likewise, steaks from cattle fed 0 and 25% DDGS underwent comparable changes in color over the 7-day display period (P > 0.10). Likewise, there were no interactions between DRC and DDGS for TBARS values. TBARS tended (P = 0.11) to be greater when DDGS were included in the diet, but the magnitude of these differences apparently was not sufficient to cause detectable differences in rancid flavors or to induce changes in shelf stability (color change) of beef.

Loss of Weight During Storage (Purge Loss) and During Cooking
We found that there were no differences among treatments with respect to purge loss (P > 0.10) during the 14-day wet aging period (Figure 6). However, we did observe that the addition of DRC to finishing diets (Figure 7) increased weight loss during cooking (P < 0.05), though these differences were quite modest. Purge loss and shrink during cooking were unaffected by the use of distiller’s grains in the diet (P > 0.15).

Vitamin E Concentration
Vitamin E was evaluated on the lean portion of the separated 9th-10th-11th rib section, and results are summarized in Figure 8. Diet had no effect (P > 0.20) on vitamin E content of lean tissue. This observation is supported by the fact that there were no differences in retail display life among the four dietary treatments.

Fatty Acid Analyses
Fatty acid analyses are summarized for the triglyceride and phospholipid fractions of the fat and lean fractions of the 9th-10th-11th rib section in Tables 6, 7, 8, and 9. There were no interactions between DRC and DDGS with respect to triglycerides extracted from the separable fat portions of the rib (P > 0.05), except for C21:0 (P < 0.01). Similarly, there were few interactions between DRC and DDGS in terms of changes in fatty acid profiles of neutral lipids (i.e., triglycerides). Notable exceptions include C14:1 and C23:0 in the fat that was extracted from the separated lean fraction (presumably this was largely marbling), which were lower when DDGS were added to diets without DRC, but higher when DDGS were added to diets with DRC. In spite of the statistical significance of these effects, the biological relevance is limited, as the magnitude of change is quite small and they constitute minor fatty acids with no known specific biological activities.

Feeding DRC resulted in small but measureable increases in C12:0, C14:0, and C21:0, and a compensatory decrease in C18:1n9 from triglycerides. The magnitude of these changes was relatively modest. The increase in myristic acid (C14:0) generally is not positive, as this is one of the key fatty acids associated with plaque formation in atherosclerosis. However, the change was relatively small, and was apparent only in fat extracted from the separated lean.

Feeding DDGS resulted in a number of changes in the proportions of fatty acids that appeared in the triglycerides extracted from the separated fat and lean portions of the rib. Generally speaking, the C18:1 fatty acids decreased in response to feeding DDGS, while the proportions of C18:0 and C18:2 increased, including the trans-10, cis-12 isomer of conjugated linoleic acid. The proportion of C16:0 in fat extracted from the separated lean fraction also was significantly decreased, which generally is positive.

Diet had relatively little impact on the proportions of fatty acids within the phospholipid fraction of separated lean and fat. As with the triglycerides, the greatest impact (quantitatively speaking) was in the substitution of C18:2 fatty acids for C18:1 fatty acids in phospholipids. Overall, the changes in fatty acid profiles of steaks derived from cattle fed the different diets were, as expected, quite modest. The distiller’s grains used in this experiment were derived entirely from corn, so the fatty acid profiles of the diets should have been similar.

Heterocyclic Amines (HCA) in Steaks
The values for HCA were calculated on a cooked weight basis, and are expressed as a concentration of HCA in parts per billion (Figure 9). They were measured on a ppb and were compared based on the concentration of HCAs present. We found no differences with the addition of DRC or DDGS. With the lipid oxidation values of steaks from cattle fed DDGS having higher values, this posed a possibility of having increased HCA concentrations, however this was found no to be true. 

Implications  

Based on performance and carcass values in this study, DDGS can effectively substitute for 25% of steam-flaked corn with no deleterious consequences. Substitution of flaked grain with DDGS would therefore be dictated by availability and relative cost of these ingredients. Substitution of flaked corn with dry-rolled corn would reduce overall costs associated with grain processing, which can be achieved while maintaining performance and carcass characteristics that are at least as good as that which can be achieved feeding diets based solely on flaked grain. Additionally, this research indicates that distiller’s grains and dry-rolled corn can substitute for flaked corn with no detrimental effects on carcass composition of beef cattle. These data also suggest that DDGS can be added to SFC or DRC diets without impacting the composition of the carcass, especially the fat content. Ultimately, it appears that partial substitution of SFC with DRC and(or) DDGS has little effect on flavor attributes of beef. The absence of differences between diets is positive, indicating that DRC and DDGS can be substituted for SFC with no negative impact on display attributes of fresh meat products. Finally, the research suggests that the industry can feed DDGS without increasing the amount of carcinogenic compounds that are released when cooking beef from cattle fed DDGS.