|
|
||||||||
Department of Animal and Range Sciences, North Dakota State University, Fargo 58105
| Abstract |
|---|
|
|
|---|
Key Words: By-products Cattle Fermentation Finishing Potatoes Tenderness
| Introduction |
|---|
|
|
|---|
Busboom et al. (2000)
and Nelson et al. (2000)
stated that some meat retailers, purveyors, and chefs in North American and Japan have the perception that barley- and potato-fed beef is softer, more watery, inferior in color, and less flavorful than corn-fed beef. However, inclusion of 10 and 20% ensiled potato pieces in corn- and barely-based diets had no effect on meat quality characteristics (Busboom et al., 2000
; Nelson et al., 2000
).
Adoption of steam peeling by the frozen potato industry produces a by-product with different characteristics (lower pH; higher CP) compared with alkaline peeling methods. To our knowledge, no study has investigated by-products from steam-peeled potato manufacturing in beef finishing diets on digestion and meat quality. The objectives were to evaluate effects of potato-processing waste (PW) inclusion level in high-concentrate beef cattle finishing diets on digestion, feedlot performance, and meat quality.
| Materials and Methods |
|---|
|
|
|---|
Steers and heifers were handled and cared for following protocols approved by the North Dakota State University Institutional Animal Care and Use Committee. All surgical procedures also followed approved institutional protocols.
Feedlot Study: Animals and Diets
One hundred twenty-five predominantly Charolais crossbred yearling heifers (365 ± 0.3 kg initial BW) were blocked by weight (five blocks) and allotted randomly to one of five dietary treatments (five pens/treatment). Initial weights were determined by a 3-d average during which heifers were fed a common diet (45% corn, 45% alfalfa hay, 5% concentrated separator by-product (CSB, desugared molasses), and 5% supplement) at 2% of BW. Heifers were implanted on d 1 with 14 mg of estradiol and 140 mg of trenbolone acetate (Revalor-H, Intervet, Millsboro, DE).
The potato waste (Table 1
) used was a slurry and was the by-product following steam peeling for frozen potato products production (13% DM) and was obtained from a commercial frozen potato-manufacturing facility located in North Dakota. The product contained steam-peeled potato waste, potato pieces, and other water-soluble material produced during the steam-peeling process. It did not contain partially cooked frozen potato products, which did not meet food-grade specifications. The final control diet contained 82% dry-rolled corn, 10% alfalfa hay, 5% CSB, and 3% supplement (Table 2
). Potato waste replaced 0, 10, 20, 30, and 40% of corn and CSB on a DM basis. The CSB was in the control diet to reduce dustiness and ingredient separation, which was not warranted in rations with potato waste and consequently replaced. Sauter et al. (1980)
indicates that feeding more than 25% of dietary DM as alkaline-peeled potato filter cake reduces daily gain. Nelson et al. (2000)
fed up to 20% ensiled potato pieces. We wanted to determine if steam-peeled potato waste could be fed at higher levels (up to 34% of dietary DM or 40% replacement of corn and CSB). Heifers were adapted for final finishing diets during the first 28 d of the trial. Heifers were initially fed 45% corn and(or) PW, 45% alfalfa hay, 5% CSB, and 5% supplement for 7 d. Every seventh day of the adaptation period, hay was decreased from 45 to 35, 25, and 15% (DM basis), such that on d 29 heifers were consuming the finishing diet. Diets were formulated to contain 27.5 mg monensin/kg, 11 mg tylosin/kg, and 0.44 mg melengestrol acetate/kg, and at least 13% CP, 0.7% Ca, 0.3% P, and 1% K (DM basis).
|
|
Carcass Data Collection and Muscle Sampling
The two heaviest blocks of heifers were slaughtered at 84 d and the remaining three blocks at 105 d. Final weights were calculated from hot carcass weight (HCW) divided by a 62% yield. At slaughter, HCW was recorded, and liver scores were evaluated to determine liver abscess severity (Brink et al., 1990
). After carcasses were chilled for 48 h, marbling; 12th rib fat thickness; longissimus muscle area; and percentage of kidney, pelvic, and heart fat (KPH) were determined.
A 7.62-cm (approximate) portion was removed from anterior end of the shortloins (NAMP #174; NAMP, 1997
) from the four heaviest blocks (n = 20) after carcasses were chilled for 48 h. Shortloin sections were vacuum packaged and transported in an ice-filled cooler to North Dakota State University. Shortloins were aged in the vacuum-packaged bags for 16 d postmortem at 4°C. After aging, shortloins were processed into 2.54-cm steaks. After steaks were exposed to air for approximately 15 min, a Minolta Chroma Meter CR-310 colorimeter (Minolta Corp., Ramsey, NJ) was used to record longissimus lean and subcutaneous fat L*, a*, and b* color space values. Steaks were weighed, individually vacuum packaged, and frozen at -40°C. Steaks were thawed at 4°C for 24 h before cooking for tenderness and sensory panel analysis.
Warner-Bratzler Shear Force Determination
One steak from each heifer was used for Warner-Bratzler shear force (WBS) evaluation (AMSA, 1995
). Steaks were thawed for 24 h at 2°C before cooking and thaw drip loss was determined. Steaks were oven broiled at 260°C until steaks reached an internal temperature of 71°C and allowed to cool to room temperature and then weighed to determine cooking loss. Seven to ten 1.27-cm cores from each steak were obtained parallel to the muscle fiber. Each core was sheared once on a Warner-Bratzler shear machine (G-E Electric Manufacturing Co., Manhattan, KS) using a 250 mm/min crosshead speed. The mean of six cores was used in the statistical analysis.
Trained Sensory Panel
Sensory panel analysis was conducted with a trained taste panel (AMSA, 1995
). Panelists were trained to evaluate initial tenderness, juiciness, sustained tenderness, and flavor intensity (Cross et al., 1978
). Panelists scored samples by placing marks on 10-cm lines labeled at each end (0 = extremely tough, dry, and bland and 10 = extremely tender, juicy, and intense beef flavor). A ruler was used to determine scores. Steaks were oven broiled as previously described with WBS evaluation. Sample size was 1.27 x 1.27 x 2.54 cm. The sensory panel was conducted for 10 d with two sessions per day at 0900 and 1500. Five samples were given per session and assigned randomly. Eight of 12 trained panelists were assigned to each session.
Digestion Study: Animals and Diets
Four ruminally, duodenally, and ileally cannulated Holstein steers (474.7 ± 26.6 kg initial BW) were used in a 4 x 4 Latin square. Steers were housed in an enclosed barn in individual tie stalls (1.5 m x 2.5 m). Animals were allowed ad libitum access to water and diets. Treatments included replacement of corn and CSB at 0, 13, 27, and 40% with PW (DM basis). The control diet contained 82% dry-rolled corn, 10% alfalfa hay, 5% CSB, and 3% supplement (Table 1
). Steers were fed twice daily at 12-h intervals. Each experimental period was 14 d in length with a 9-d adaptation period. Intake was measured for the last 5 d of the period. Samples of feed and orts were collected during the 5-d intake period. Steers were weighed at the beginning and end of each period.
Sample Collection
Chromic oxide (8 g) was placed in gelatin capsules and ruminally dosed at each feeding from d 7 to d 12. On d 11 and 12, duodenal, ileal, and fecal grab samples were collected at 4-h intervals. Collection times were advanced 2 h on d 12 to obtain samples that represented every even hour for a 24-h period. Fecal samples (200 g, fresh basis) were composited across sampling times within steer and period. Samples were stored frozen (-20°C) until analyses.
On d 13, Co-EDTA (200 mL; Uden et al., 1980
) was dosed intraruminally 2 h before morning feeding to determine liquid dilution rate and ruminal volume. Ruminal fluid samples were taken at 0, 2, 4, 6, 8, 10, and 12 h after feeding. Ruminal fluid was collected with a suction strainer, and pH was recorded with a Beckman 200 series portable pH meter (Beckman Instruments, Inc., Fullerton, CA). A 100-mL sample of ruminal fluid was retained, and 1 mL of 7.2 N H2SO4 was added to fluid. Samples were frozen (-20°C) for later analysis of NH3-N, Co, and VFA.
Ruminal evacuations were conducted before feeding on d 14 to determine DM and OM fill. Ruminal contents from each steer were removed, weighed, mixed by hand, and subsampled. One grab sample was taken for later analysis of DM, OM, CP, NDF, ADF, and starch. For isolation of bacterial cells and analysis of DM, OM, N, and purine, a 4-kg sample of rumen contents was taken, and 2 L of formaline/saline solution (3.7% formaldahyde/0.9% NaCl) was added (Zinn and Owens, 1986
). Samples were cooled for 2 to 4 h at 4°C and then frozen (-20°C).
Laboratory Analysis
Feed, orts, and fecal samples were dried at 50°C in a forced-air oven for 48 h. Dried samples were ground with a Wiley mill to pass a 2-mm screen. Samples of feed and orts were composited within each period for analysis. Duodenal, ileal, and ruminal grab samples were lyophilized and ground with a coffee grinder (Braun Aromatic KSM2, Boston, MA). Diet, ruminal, duodenal, ileal, and fecal samples were analyzed for DM, OM, ADF, and N (AOAC, 1997
). Analysis of diet, ruminal, duodenal, ileal, and fecal NDF were conducted by the method of Robertson and Van Soest (1991)
. Diet, ruminal, duodenal, ileal, and fecal samples were analyzed for starch (Herrera-Saldana and Huber, 1989
). Fractional rumen passage (Kp) and digestion (Kd) rates of starch were calculated as the ratio of duodenal starch flow (passage; g/h) and ruminal disappearance (digestion; g/h) to ruminal fill (g), respectively.
Duodenal, ileal, and fecal samples were analyzed for Cr (Czarnocki et al., 1961
). Sample Cr concentration was used in conjunction with nutrient concentration to calculate site of digestion and digestibility (Merchen, 1988
). Approximately 3,000 g of duodenal fluid and 1,800 g of ileal fluid were taken during each collection period from each animal.
Ruminal fluid was thawed and centrifuged (10,000 x g; 10 min). Five milliliters of supernatant was mixed with 1 mL of 25% (wt/vol) metaphosphoric acid and centrifuged (10,000 x g; 10 min). Resulting supernatant was taken for ruminal VFA analyses, which were determined by gas chromatography (Shimadzu GC 9A) using a packed column (SP-1200; Supelco, Bellefonte, PA) and 2-ethyl butyric acid as the internal standard (Goetsch and Galyean, 1983
). A small portion (0.6 g) of the lyophilized duodenal samples was reconstituted to 3% (wt/vol) in 0.1 N HCl, vortexed, and centrifuged at 20,000 x g for 20 min (Hannah et al., 1991
). Supernatant for duodenal and first ruminal centrifugation was taken for NH3-N analysis (Broderick and Kang 1980
). Ruminal cobalt concentration was determined after centrifugation (18,000 x g; 20 min) using atomic absorption spectrometry (model 3030B, Perkin Elmer, Inc., Wellesley, MA). Fluid dilution rate was calculated by regressing the natural logarithm of Co concentrations against time (Warner and Stacy, 1968
).
Ruminal bacteria were isolated from a 4-kg sample of ruminal contents. Ruminal contents were blended, and the mixture strained through two layers of cheesecloth. Feed particles and protozoa in ruminal samples were removed via centrifugation at 1,000 x g for 10 min. Bacteria were separated from supernatant by centrifuging at 20,000 x g for 20 min. Isolated bacteria were frozen and analyzed for DM (lyophilized), ash, N (as previously described), and purines (Zinn and Owens, 1986
). True ruminal OM digestion was calculated by correcting OM apparently digested in the rumen with microbial OM flow to the duodenum.
Statistical Analysis
Feedlot data were analyzed as a randomized complete block design (Steel and Torrie, 1980
) using Mixed procedures of SAS (Ver. 6.12; SAS Inst., Inc., Cary, NC). The experimental unit for feedlot performance and carcass characteristics was pen (n = 5), and for instrumental tenderness, color, and sensory panel data the unit was individual animal (n = 20). Linear, quadratic, and cubic contrasts were used to compare levels of PW. Distributions of quality grade, tenderness, and liver scores were analyzed with chi-squared analysis of the FREQ procedure of SAS.
Digestion and flow data were analyzed as a 4 x 4 Latin square using the GLM procedure of SAS. The model included effects for period, steer, and diet. Data over time were analyzed as a repeated measures design (Gill and Hafs, 1971
). The model included effects for period, steer, diet, time, diet x time, and steer x period x diet. The three-way interaction was used in the error term to test for effect of diet. Linear, quadratic, and cubic contrasts were used to compare levels of PW.
| Results and Discussion |
|---|
|
|
|---|
Heifer performance and apparent dietary energy concentration are shown in Table 3
. Dry matter intake decreased (linear, P = 0.007) from 10 to 40% PW. Final BW decreased from 0 to 30% and then increased at 40% addition of PW (linear, quadratic, and cubic; P < 0.01). Average daily gain decreased from 0 to 30% treatments, then increased at 40% PW (quadratic; P < 0.01). Feed efficiency (gain:feed) followed the pattern of ADG, first decreasing from 0 to 30% and then increasing at 40% PW (linear and quadratic, P = 0.001). Sauter et al. (1980)
noted a 14% decrease in DMI, a 17% decrease in ADG, and a 5% decrease in efficiency with inclusion of 50% potato by-product as compared with 25% potato by-product in barley-based finishing diets. Our results differ with previous studies evaluating filter cake (Heinemann and Dyer, 1972
), where up to 40% addition potato by-product did not affect intake, gain, or efficiency of a barley-based finishing diet. Diets lower in energy content were used for these two studies (30% roughage), which likely explains the difference between data. However, Nelson et al. (2000)
reported increased DMI and ADG with a 10% addition of ensiled potato pieces, in corn-based diets with 7% roughage. It is possible that some of the decrease we observed in DMI with high inclusion levels of PW was the result of degradation in the bunk due to its high moisture content. However, weather data collected at an automated weather collection site located in close proximity to the feedlot, indicated that temperatures were within normal ranges throughout the duration of the feeding period. In addition, the daytime maximum temperature only exceeded 32°C on 5 d of the 104-d trial, and average temperature during the trial was 15°C (NDAWN, 2000
).
|
Hot carcass weight decreased (linear, quadratic, and cubic; P < 0.01) with increasing PW, was lowest at 30%, and was greatest at the 0% inclusion level (Table 4
). Addition of PW decreased 12th rib fat thickness (linear, P = 0.04). Linear (P = 0.02) and cubic (P = 0.06) decreases were observed for longissimus muscle area, with the larger longissimus muscle area being associated with 10% PW. Yield grade (YG) was greatest at 0% PW but was similar from 10% to 40% PW (quadratic, P = 0.09). In comparing 25 and 50% addition of potato by-product, Sauter et al. (1980)
reported decreased fat thickness, ribeye area (REA), and YG. In contrast, Nelson et al. (2000)
observed no difference in HCW, REA, fat thickness, or YG with addition of ensiled potato pieces up to 20% of the finishing diet. Addition of PW did not affect marbling, KPH, or liver score (P > 0.20). Marbling was similar for diets in studies containing 0 to 20% ensiled potato pieces (Nelson et al., 2000
) and 0 to 50% of potato slurry (Sauter et al., 1980
). Taken together, carcass characteristics indicate lower NE intake of heifers consuming greater than 10% PW.
|
Meat Quality Characteristics
One shortloin was not retrieved from the packing plant from the 10% PW treatment; therefore, 99 steaks were used in the analysis. Subcutaneous fat color space (L* = 76.6 ± 0.8; a* = 10.7 ± 0.4; b* = 10.4 ± 0.4) and longissimus lean color space (L* = 41.0 ± 0.5; a* = 24.3 ± 0.3; b* = 9.8 ± 0.3) were not different (P
0.15) among treatments and were similar to data reported by Page et al. (2001)
for beef heifers. No effect (P
0.20) was observed for thaw-drip (3.82 ± 0.30%) or cooking loss (29.0 ± 0.70%). Warner-Bratzler shear force tended to be greater for 30% (3.27 kg) than for 40% (2.75 kg) PW (cubic, P = 0.07). Distribution of WBS was not different among treatments (P > 0.4, 0.46%, 1 to 2 kg; 44.5%, 2 to 3 kg; 41.4%, 3 to 4 kg; and 10.1%, 4 to 5 kg) and two steaks were tough (WBS above 4.5 kg). Huffman et al. (1996)
reported 98% consumer acceptability of steaks with a WBS value of 4.1 kg or less, which suggests that steaks on all treatments were acceptable.
Initial tenderness responded cubically (cubic, P = 0.02), first increasing, then decreasing, and then increasing as PW level increased (Table 5
). Overall tenderness tended to increase with the inclusion of PW similar to the manner in which initial tenderness responded (linear, P = 0.08; cubic, P = 0.10). Juiciness and flavor intensity sensory panel scores were not different (P > 0.3) among treatments. Busboom et al. (2000)
also reported that inclusion of ensiled potato pieces in barley- and corn-based finishing diets did not affect thaw-drip loss; cooking loss; WBS values; or trained sensory panel tenderness, juiciness, or flavor intensity.
|
Addition of PW decreased DMI relative to BW linearly and quadratically (P < 0.01, Table 6
) from 2.23 to 1.83%. Similar results were reported by Stanhope et al. (1980)
when potato-processing residue replaced barley in finishing diets from 0 to 60%; however, DMI averaged only 1.1% of BW in that study. Onwubuemeli et al. (1985)
reported decreased intake in Holstein steers when PW replaced corn at 0, 10, 20, and 30% of the diet. Liquid dilution rate tended to increase from 6.81 to 9.69%/h (linear, P = 0.11) with increasing level of PW (Table 6
). Dietary DM decreased from 87.9 to 28.0% with an increasing level of PW. The increase in liquid dilution rate could be explained by increased water consumption, which was not measured in this trial. Holzer et al. (1975)
reported that total water consumption increased as dietary moisture increased from 10 to 75%. Ruminal liquid volume and DM fill were not different among treatments (P > 0.60; Table 6
). Ruminal starch fill (1.4 ± 0.3 kg), starch Kp (2.6 ± 1.3%/hr), and starch Kd (21.0 ± 4.6%/hr) also were not different (P > 0.24) among treatments. The starch Kd value is apparent and does not account for microbial contribution to duodenal
-linked glucose (Branco et al., 1999
). Furthermore, ruminal starch pool size from ruminal evacuations conducted 12 h after feeding would be smaller than mean ruminal starch.
|
There were no time x treatment interactions for any ruminal characteristics; therefore, treatment means averaged across time are presented and discussed. Ruminal pH was not affected by addition of PW (P
0.22; Table 6
). Onwubuemeli et al. (1985)
reported an increase in ruminal pH with the addition of steam-peeled potato waste (0, 10, 20, and 30% of diet DM) in steers fed lactating dairy cow diets. Ruminal NH3-N concentration was lowest at 0% PW at 7.91 mM and increased (quadratic and cubic, P
0.07) with PW addition. The 13% level of PW had the greatest ruminal ammonia concentration. In contrast to our results, ruminal NH3-N decreased linearly when PW replaced high-moisture corn in lactating dairy cow diets (Onwubuemeli et al., 1985
). It appears that increasing potato waste replacement (above 13%) may decrease ruminal ammonia in both experiments. Onwubuemeli et al. (1985)
suggested potato starch fermentation rate was faster than cornstarch, causing a decrease in ruminal ammonia. We did not measure starch fermentation rate.
Increasing level of PW increased total ruminal VFA concentration (linear, P < 0.001; quadratic, P = 0.03; Table 6
). Acetate proportions were not different among treatments (P
0.15). Propionate decreased with increasing PW (linear and cubic, P
0.003), but was greatest at 0 and 27% levels (28.7 and 28.1 mol/100 mol, respectively). Butyrate increased with the addition of PW (linear; P = 0.06). Acetate-to-propionate ratios were lower for the 0 and 27% PW treatments than for the 13 and 40% PW treatments (cubic, P = 0.02). It is difficult to explain these changes. Onwubuemeli et al. (1985)
reported decreased total ruminal VFA and acetate concentrations, increased ruminal propionate concentrations, and decreased acetate:propionate when PW replaced high-moisture corn.
Digestion
The greatest OM intake (Table 7
) was at 0% PW and decreased with the addition of PW (quadratic, P = 0.04), which follows DMI. Apparent ruminal OM digestion for the 0% level was 58.9%, which is similar to 56% for dry-rolled corn diets (Zinn, 1990
; Zinn et al., 1995
). True ruminal OM disappearance was greatest at 0% and tended (linear, P = 0.08; quadratic, P = 0.06) to decrease with the addition of PW but was not different when expressed relative to intake. Small and large intestinal OM disappearance was not different among treatments (P > 0.15). Our apparent total-tract OM disappearance for control diet (86%) was greater than reported in previous studies, 78% (Zinn, 1990
) and 76% (Zinn et al., 1995
).
|
|
|
| Implications |
|---|
|
|
|---|
| Footnotes |
|---|
2 Current address: Department of Animal Sciences, Washington State University, Pullman, WA 99164. ![]()
4 Current address: Department of Animal Science and Industry, Kansas State University, Weber Hall, Manhattan 66506. ![]()
3 Correspondence: 177 Hultz Hall (phone: 701-231-7660; fax: 701-231-7590; E-mail: glardy{at}ndsuext.nodak.edu).
Received for publication January 30, 2003. Accepted for publication July 14, 2003.
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
J. I. Szasz, C. W. Hunt, O. A. Turgeon Jr., P. A. Szasz, and K. A. Johnson Effects of pasteurization of potato slurry by-product fed in corn-or barley-based beef finishing diets J Anim Sci, December 1, 2005; 83(12): 2806 - 2814. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |