|
|
||||||||
Department of Animal Science, University of Arkansas, Fayetteville 72701
2 Correspondence:
phone 479-575-4840; fax 479-575-7294; E-mail:
japple{at}uark.edu.
| Abstract |
|---|
|
|
|---|
Key Words: Carcass Composition Feather Meal Growth Pigs Pork Quality
| Introduction |
|---|
|
|
|---|
Replacing soybean meal with up to 9% FM was shown to enhance carcass leanness without affecting performance (Chiba et al., 1995); however, when FM was the sole protein source in swine diets, weight gain and carcass cutability were depressed, even in pigs fed diets fortified with crystalline lysine (Chiba et al., 1996). van Heugten and van Kempen (2002) failed to note differences in pig performance and carcass traits when pigs were fed diets containing 2 to 8% FM, but performance was decreased by feeding diets supplemented with 10% FM. However, none of the aforementioned studies tested the effects of supplemental FM on pork quality. Therefore, the objectives of this research were to assess the effects of including FM in traditional cornsoybean meal diets (Exp. 1) or commercial cornsoybean meal and wheat middlings diets (Exp. 2) on live animal performance, carcass composition, and pork quality.
| Materials and Methods |
|---|
|
|
|---|
Animal care and all experimental procedures were approved by the University of Arkansas Interdepartmental Animal Care and Use Committee before the initiation of experiments.
Hydrolyzed FM containing 8% blood was contributed by Tyson Foods, Inc., Specialty Products Division, and was obtained from their protein plant in Noel, MO. Briefly, fresh poultry feathers were spread evenly on a conveyor, passed through a metal detector, and hydrolyzed under pressure (2.11 to 2.81 kg force/cm2) in a batch hydrolyzer for 30 min at 76.7°C. Feathers were hydrolyzed in a batch hydrolyzer to break keratin (long-chain proteins) into more digestible, smaller-chain proteins and to reduce microbial contamination. Blood was coagulated and added to the hydrolyzed feathers in the batch hydrolyzer to increase protein content of the product. The resulting product was then dried in a natural gas-fired, direct-contact dryer and milled through a mesh screen before being shipped to the University of Arkansas Feed Mill in Fayetteville. The FM contained 7% moisture, 81.2% CP (54.0% digestible protein), 11.1% crude fat, and 3.08 Mcal of ME/kg on a DM basis (Tyson Foods, Inc., Springdale, AR). The FM used in these experiments was analyzed for amino acids by an independent laboratory (Novus Int., Inc., St. Charles, MO), and was found to contain 7.36% leucine, 6.61% valine, 5.25% cysteine, 4.15% isoleucine, 3.93% threonine, 2.62% tyrosine, 2.30% lysine, 1.05% histidine, 0.62% methionine, and 0.44% tryptophan on a dry weight basis.
Animals and Diets.
Crossbred barrows and gilts used in both experiments were offspring of Yorkshire x Landrace dams mated to Duroc x Hampshire sires (The Pork Group, Inc., Rogers, AR). Prior to each experiment, pigs were moved approximately 100 m from the University of Arkansas Nursery to the University of Arkansas Swine Growing-Finishing Facility. In Exp. 1, 132 pigs (initial BW of 24.8 ± 3.6 kg) were blocked into four weight groups with 36 pigs in blocks 1 and 2, and 30 pigs in blocks 3 and 4. Pigs within each block were allotted randomly into pens with six pigs per pen in blocks 1 and 2, and five pigs per pen in blocks 3 and 4. Stratification across pens was based on sex and litter origin. A total of 24 pens were randomly assigned to one of three dietary treatments: 1) control cornsoybean meal starter, grower, and finisher diets devoid of FM; 2) control diets formulated with 3% FM; and 3) control diets formulated with 6% FM. Feather meal was substituted for soybean meal on an equal lysine basis at the expense of corn in the diets.
In Exp. 2, 120 pigs with an initial BW of 25.4 ± 0.1 kg were sorted into six weight blocks with 20 pigs per block. Pigs within each block were allotted randomly to pens (five pigs/pen), with stratification based on sex and litter origin. A total of 24 pens were randomly allotted to one of four dietary treatments: 1) positive control cornsoybean meal-based starter, grower, and finisher diets; 2) negative control cornsoybean meal- and wheat middlings-based starter, grower, and finisher diets; 3) negative control diets formulated with 3% FM; and 4) negative control diets formulated with 6% FM. In the FM-diets, FM was added to the negative control diets at the expense of wheat middlings.
Pigs in both experiments were fed a four-phase dietary program with transition from starter to grower-I, grower-I to grower-II, and grower-II to finisher phases occurring when average block weight reached 36.4, 68.2, and 90.9 kg, respectively. In each experiment, diets were formulated to be isolysinic (lysine content of starter, grower-I, grower-II, and finishing diets was 1.16, 0.90, 0.66, and 0.53%, respectively [Exp. 1, Table 1
], and 1.00, 0.91, 0.80, and 0.66%, respectively [Exp. 2, Table 2
]) and isocaloric (ME content of starter, grower-I, grower-II, and finishing diets was 3.40, 3.31, 3.32, and 3.30 Mcal/kg, respectively [Exp. 1, Table 1
], and 3.41, 3.35, 3.36, and 3.37 Mcal/kg, respectively [Exp. 2, Table 2
]). Furthermore, all diets met or exceeded NRC (1998) amino acid, energy, and other nutrient requirements for growing-finishing swine.
|
|
Carcass Data Collection.
In Exp. 1, when the lightest block of pigs averaged 109.1 kg, all pigs were transported approximately 309 km and harvested at a commercial pork-packing plant (Brown Packing Co., Little Rock, AR) after a 12-h rest period. Pigs were harvested according to industry-accepted procedures, and carcasses were conventionally chilled at 2°C for 24 h. Hot carcass weight was recorded and midline backfat depth opposite the first rib, last rib, and last lumbar vertebra was measured to calculate average backfat thickness. Carcasses were then fabricated into primal cuts according to Institutional Meat Purchase Specifications (IMPS) for Fresh Pork Products (USDA, 1995), and bone-in hams (IMPS #401) from left sides were collected and identified. All hams were individually weighed and the depth of fat beneath the butt face was measured. Hams were then paper-wrapped, boxed, and shipped in a refrigerated truck to the Louisiana State University Meat Laboratory (Baton Rouge, LA), where hams were analyzed for fat-free lean and total fat composition using a total-body electrical conductivity (TOBEC) unit (Meat Quality Inc., Springfield, IL). Ham TOBEC scans were also used to estimate carcass fat-free lean yield and total fat content using the prediction equations of Knowles et al. (1998).
When the lightest block of pigs in Exp. 2 averaged 108.8 kg, all pigs were transported approximately 402 km to a commercial pork harvest/fabrication plant (Excel Corp., Marshall, MO). Pigs were harvested after a 12-h rest period at the plant, fat and loin depths were measured online with a Fat-O-Meater automated probe (model S71; SFK Technology A/S, Cedar Rapids, IA) inserted between the 10th and 11th ribs at a distance of approximately 7 cm from the midline, and hot carcass weight was recorded. Carcasses were subsequently subjected to a conventional spray-chilling system for 24 h. Prior to fabrication, midline backfat depth opposite the first rib, last rib and last lumbar vertebra was recorded, and loins were marked between the 10th and 11th ribs in order to measure longissimus muscle area upon arrival at the University of Arkansas Red Meat Abattoir. Carcasses were then fabricated into primal cuts, and bone-in hams (IMPS #401) from left sides were analyzed for lean composition using a TOBEC unit. Prediction equations used to calculate ham lean composition (fat-free basis) from TOBEC scans cannot be presented because they are the intellectual property of Cargill Red Meat Sector (Wichita, KS). Bone-in pork loins (IMPS #410) from left sides were captured during fabrication and subsequently vacuum-packaged, boxed, loaded onto a refrigerated truck, and transported to the University of Arkansas for pork quality data collection.
At approximately 48 h postmortem, pork loins were cut between the 10th and 11th ribs, and the area of the longissimus muscle was traced onto acetate paper (Bee Paper Co. Inc., Wayne, NJ). The area of the longissimus muscle was measured at a later time using a compensating planimeter. The longissimus muscle chops were removed from the posterior portion of the loin in the following order: 1) 2.5-cm-thick loin chop; 2) 3.8-cm-thick loin chop; 3) 2.5-cm-thick loin chop; and 4) 3.8-cm-thick loin chop.
After a 45-min bloom period at 4°C, the 2.5-cm-thick longissimus muscle chops were visually evaluated for marbling (1 = devoid [1% intramuscular lipid] to 10 = abundant [10% intramuscular lipid]; NPPC, 1999), firmness (1 = very soft and watery to 5 = very firm and dry; NPPC, 1991), and color based on both the American (1 = pale, pinkish gray to 6 = dark purplish red; NPPC, 1999) and Japanese color standards (Nakai et al., 1975). Commission International de lEclairage L*, a*, and b* values (CIE, 1976) were determined from a mean of four random readings (two readings for each 2.5-cm-thick chop) made with a Hunter MiniScan XE (model 45/0-L; Hunter Associates Laboratory, Reston, VA) using illuminate C and a 10° standard observer. The saturation index, or chroma (C*), was calculated as C* = (a*2 + b*2)1/2 and is a measure of the total color, or vividness of the color, of the longissimus muscle. After quality data collection, both chops were wrapped in white, polycoated, heavy-weight freezer paper (Paper Con, Dallas, TX), and frozen at -20°C before cooking and Warner-Bratzler shear force determinations.
The two 3.8-cm-thick chops were used for drip loss determination following the suspension procedure of Honikel et al. (1986). A 3.8-cm core was removed from each 3.8-cm-thick chop, weighed, and suspended on a fishhook (barb removed) mounted to the lid of a plastic container (46 x 66 x 38 cm deep Dur-X Food Box; Rubbermaid Commercial Products LLC, Winchester, VA). Containers were sealed and stored at 2°C for 48 h. Then, each core was removed from its fishhook, blotted with a paper towel, and reweighed. The loss in weight due to drip and evaporation was divided by the original weight, multiplied by 100, and reported as a percentage of drip loss. Additionally, 2 g of longissimus muscle from each chop after core removal was homogenized in 20 mL of distilled, deionized water. The pH of the homogenate was measured with a temperature-compensating combination electrode (model 300731.1; Denver Instrument Co., Arvada, CO) attached to a pH/ion/FET-meter (model AP25; Denver Instrument Co.).
Longissimus muscle chops were thawed for 16 h at 2°C, weighed, and then cooked to an internal temperature of 71°C in a commercial convection oven (Zephaire E model; Blodgett Oven Co., Burlington, VT) preheated to 165°C. Internal temperature was monitored with Teflon-coated thermocouple wires (Type T; Omega Engineering, Inc., Stamford, CT) placed into the geometric center of each longissimus muscle chop and attached to a multi-channel data logger (model 245A; VAS Engineering Inc., San Diego, CA). Chops were turned once during the cooking process when the internal temperature reached 35°C. Immediately after removal from the oven, chops were blotted dry on paper towels and weighed, and the difference between precooked and cooked weights was used to calculate percentage of cooking loss. Chops were allowed to cool to room temperature, and five 1.27-cm diameter cores were removed parallel to the muscle fiber orientation. Then, each core was sheared once through the center with a Warner-Bratzler shear force device attached to an Instron Universal Testing Machine (model 4466; Instron Corp., Canton, MA) with a 55-kg tension/compression load cell and a crosshead speed of 250 mm/min.
Data Analyses.
Animal performance, carcass composition, and pork quality data from both experiments were analyzed as a randomized complete block design with pen as the experimental unit and blocks based on initial BW. Analysis of variance was generated using the GLM procedure of SAS (SAS Inst., Inc., Cary, NC), and the residual error was used to test the main effect of dietary treatment in both experiments. Linear and quadratic polynomials were used to detect the response of replacing soybean meal (Exp. 1) and wheat middlings (Exp. 2) with FM in the diet. Additionally, a contrast was included in the statistical model to more accurately compare the positive (0% wheat middlings) and negative (17 to 18% wheat middlings) control diets in Exp. 2.
| Results |
|---|
|
|
|---|
During the starter phase, ADG decreased quadratically (P < 0.06), and gain:feed decreased quadratically (P < 0.01), as FM was increased in swine diets (Table 3
). Pig performance was not (P > 0.10) affected by FM inclusion during the grower-I phase, and ADG and ADFI did not differ (P > 0.10) across dietary treatments during the grower-II phase; however, gain:feed tended to increase linearly (P < 0.07) as FM increased from 0 to 6% in the diet. Inclusion of FM in the finisher diets did not (P > 0.10) impact pig performance, nor were overall ADG, ADFI, and gain:feed affected (P > 0.10) by FM level in the diet.
|
|
During the starter phase, pigs fed the positive control diet (devoid of wheat middlings) had greater ADG (P < 0.003) and gain:feed (P < 0.02) than pigs fed the negative control diet containing 17.75% wheat middlings (Table 5
). Moreover, pigs fed the positive control diet were heavier (P < 0.005) than pigs fed the negative control diet at the end of the starter phase. Pig performance did not differ (P > 0.10) among dietary treatments during the grower-I and -II phases, and dietary treatment had no (P > 0.10) impact on pig performance during the finishing phase or across the entire study.
|
|
|
| Discussion |
|---|
|
|
|---|
The effect of dietary FM on poultry performance is well documented. Early research demonstrated that broiler performance was unaffected when up to 4% of their diet was supplemented with FM; however, 5 to 8% dietary FM induced lysine, methionine, histidine, and tryptophan deficiencies, resulting in depressed broiler performance (Moran et al., 1966; Luong and Payne, 1977; MacAlpine and Payne, 1977). Yet, when synthetic amino acids were used to remedy the associated deficiencies, up to 10% FM could be included in the diet without impacting weight gain and feed efficiency of broilers (Baker et al., 1981; Cabel et al., 1987; Cupo and Cartwright, 1991). Additionally, Eissler and Firman (1996) concluded that up to 6% FM could be included in isolysinic turkey diets without affecting performance.
Combs et al. (1958) reported that growth rate was reduced when pigs were fed diets containing 7.5 and 10% FM, but the performance of pigs fed diets containing 5% FM was similar to those fed the basal diet of corn and soybean meal. Moreover, they observed that when diets containing 7.5% FM were supplemented with synthetic lysine, growth rate and feed efficiency were improved to levels similar to pigs fed the control diet (Combs et al., 1958). Chiba et al. (1995) found that performance of pigs fed diets containing 7.5 or 15% FM (replacing soybean meal on an equal lysine basis) was similar to the performance of pigs fed the control diets; however, in a subsequent study, when soybean meal was replaced by FM in the diet, ADG and gain:feed decreased linearly as dietary FM increased from 0 (12% soybean meal) to 12% (0% soybean meal; Chiba et al., 1996). When diets containing 9% FM were supplemented with synthetic lysine to achieve a dietary lysine level of 0.73%, pig performance remained below that of pigs fed the control cornsoybean meal diets; still, Chiba et al. (1996) concluded that up to 9% FM could be incorporated in the diet without adversely affecting performance. More recently, van Heugten and van Kempen (2002) evaluated inclusion of 0, 2, 4, 6, 8, and 10% FM in the finishing diets of swine and found that pig performance was unaffected by dietary inclusion of up to 8% FM, but diets containing 10% FM reduced ADG and gain:feed compared to their contemporaries fed control (0% FM) diets.
With the exception of the quadratic relationship between pig performance and FM inclusion level noted in Exp. 1, dietary FM did not significantly impact ADG, ADFI, or gain:feed in growing-finishing pigs, which is consistent with previously published findings (Combs et al., 1958; Chiba et al., 1995; van Heugten and van Kempen, 2002). Therefore, FM may be included in swine diets at a level of 6% without any detrimental effects on live animal performance.
Inclusion of FM in the starter diets of growing-finishing pigs appeared to negatively impact ADG and gain:feed, especially in Exp. 2; still, the severity of the depression in performance in Exp. 2 may be more of a response to the inclusion level of wheat middlings in starter diets than an effect of FM inclusion. In Exp. 1, no wheat middlings were included in the starter diets, and a quadratic relationship between FM inclusion level and performance was noted. However, in Exp. 2, ADG and gain:feed were substantially reduced in pigs fed the negative control diet (17.75% wheat middlings) compared to pigs fed the positive control diet devoid of wheat middlings, whereas performance of pigs fed the high-wheat middlings diets containing FM was similar to that of pigs fed the negative control diet. This is consistent with Erickson et al. (1985), who reported that ADG and gain:feed decreased and feed intake increased linearly as the inclusion level of wheat middlings increased from 0 to 30% in starter diets of growing-finishing swine. Moreover, Han et al. (1998) found that feeding more than 10% wheat middlings reduced ADG during the first 3 wk of feeding in weanling pigs, but performance from 3 wk to finishing was not affected by dietary levels of wheat middlings, which is in agreement with the results of the present study.
Carcass Composition.
Beneficial effects of feeding diets containing FM on carcass composition have been observed in swine (Chiba et al., 1995) and broilers (Cabel et al., 1987; Cupo and Cartwright, 1991). Chiba et al. (1995) reported that dietary FM effectively reduced 10th-rib and average backfat depths and increased longissimus muscle area, resulting in higher estimates of carcass muscle composition. However, in a subsequent study, carcass traits were similar among pigs fed 0, 3, 6, and 9% FM, but, at an inclusion rate of 12%, FM actually reduced carcass cutability (Chiba et al., 1996). On the other hand, van Heugten and van Kempen (2002) reported that ultrasound backfat measurements tended to increase with increasing levels of dietary FM, but did not affect longissimus muscle area or lean muscle yields, which is similar to results from Exp. 1.
Pork Quality.
To our knowledge, this is the first study to characterize the effects of dietary FM on pork quality measures. Forty-eight-hour muscle pH increased linearly as FM increased in the negative control diets; therefore, one would expect that pork color would also get lighter and less vivid, which it did. Nonetheless, pork from pigs fed diets containing 3% FM received higher Japanese color scores than pork of pigs fed the negative control and 6% FM diets. Redness (a*) and yellowness (b*) values increased with increasing FM; a* values were intermediate to those of pork from pigs fed either the positive or negative control diets, whereas b* values did not differ from those of pork from pigs fed the cornsoybean meal (positive control) diets. Therefore, dietary inclusion of up to 6% FM did not adversely affect pork color or water-holding capacity.
Dietary wheat middlings inclusion level appeared to have had a greater impact on pork quality than altering dietary FM. For instance, ultimate (48-h) muscle pH was lower for pigs fed the negative control diet compared to their contemporaries fed the positive control diet, and pork color of pigs fed the high wheat middlings (negative control) diets was redder (higher a* value), more yellow (higher b* value), and more vivid (higher chroma value) than the color of pork from pigs fed the cornsoybean meal (positive control) diet devoid of wheat middlings. Although not statistically significant, shear force values of cooked pork from pigs fed the negative control diets were more than 0.2 kg less than those of pork from pigs fed the positive control diets; thus, the observed linear increase in shear force values with increasing levels of FM inclusion may actually be a result of the reduction in wheat middling composition of the diets rather than an increase in dietary FM. The same could also be noted about the linear relationships between dietary FM and American color scores, L*, a*, b*, and chroma values; however, the lack of previously published information on the effects of wheat middlings and/or FM on pork quality hampers the ability to discern "cause-and-effect" between the two feedstuffs.
| Implications |
|---|
|
|
|---|
| Footnotes |
|---|
Received for publication May 30, 2002. Accepted for publication September 19, 2002.
| Literature Cited |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. C. Divakala, L. I. Chiba, R. B. Kamalakar, S. P. Rodning, E. G. Welles, K. A. Cummins, J. Swann, F. Cespedes, and R. L. Payne Amino acid supplementation of hydrolyzed feather meal diets for finisher pigs J Anim Sci, April 1, 2009; 87(4): 1270 - 1281. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. W. Ssu, M. C. Brumm, and P. S. Miller Effect of feather meal on barrow performance J Anim Sci, September 1, 2004; 82(9): 2588 - 2595. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |