METABOLISM AND NUTRITION The Effect of Dietary Supplementation with Copper Sulfate or Tribasic Copper Chloride on Broiler Performance, Relative Copper Bioavailability, and Dietary Prooxidant Activity1 R. D. MILES,*,2 S. F. O’KEEFE,† P. R. HENRY,‡ C. B. AMMERMAN,‡ and X. G. LUO3 Departments of *Dairy and Poultry Science, †Food Science and Human Nutrition, and ‡Animal Science, University of Florida, Gainesville, Florida 32611 ABSTRACT Three experiments were conducted to study Cu sulfate and tribasic Cu chloride (TBCC) as sources of supplemental Cu for poultry. In Experiment 1, 252 chicks were fed the basal corn-soybean meal diet (26 ppm Cu) supplemented with either 0, 150, 300, or 450 ppm Cu from Cu sulfate or TBCC for 21 d. Chicks fed 450 ppm Cu from sulfate had lower (P < 0.05) feed intake than those consuming other diets. Feeding supplemental Cu increased (P < 0.0001) liver Cu concentration linearly with increasing dietary Cu regardless of Cu source. The slopes of regression of log10 liver Cu on dietary Cu intake did not differ between sources (P > 0.10). Linear regression over nonzero dietary levels of log10 transformed liver Cu concentration (parts per million of DM) on analyzed total Cu intake (micrograms) resulted in a slope ratio estimate of 106 ± 19 for bioavailability of Cu from TBCC compared to 100 for that in Cu sulfate. In Experiment 2, a 42-d floor pen study was conducted with 1,260 broiler chicks given the basal cornsoybean meal diet supplemented with 0, 200, 400, or 600
ppm Cu from either feed-grade Cu sulfate or TBCC. Body weight and feed conversion did not differ in birds fed up to 400 ppm Cu from either source. Birds given 600 ppm Cu from either source had lower feed intake, poorer growth, and feed conversion (P < 0.0001). Liver Cu increased (P < 0.0001) linearly with increasing dietary Cu. Based on log10 liver Cu concentration, Cu in TBCC was 112% available compared to 100% for the standard Cu sulfate. In Experiment 3, Cu sources were added to broiler starter diets at concentrations of 25, 100, and 300 ppm Cu and diets were stored at an elevated temperature to examine the effect of particle size on oxidation. Diets were stored at 37 C for up to 20 d and samples were removed at 4-d intervals. At 300 ppm added Cu, oxidation in TBCC diets was lower (P < 0.0001) than oxidation in diets fortified with coarse Cu sulfate, even though TBCC modal diameter for particle size was almost seven times smaller. Oxidation promotion by Cu sulfate was much greater with fine than in coarse sized particles for all three fortification levels.
(Key words: copper, bioavailability, oxidation, liver, performance, broiler) 1998 Poultry Science 77:416–425
Determination of Cu bioavailability from supplemental sources with traditional methods using dietary concentrations of the element below requirement level and purified dietary ingredients has met with limited success. The inherently low values for absorption and retention of Cu, as well as the lack of a radioisotope with a convenient half-life, have served to complicate these assays. McNaughton et al. (1974) used liver Cu concentration as a criterion to determine relative bioavailability of several Cu sources; however, the added concentrations, 1 or 2 ppm, were less than or equal to the amount in the basal purified diet. In addition, feed intake data were not reported. Ammerman (1995) discussed some of the problems inherent with bioavaila-
INTRODUCTION The essentiality of Cu for poultry and livestock is well documented (Davis and Mertz, 1987). Potential sources of Cu for use by the commercial feed industry must be studied for bioavailability, as well as safety. Copper from various compounds has often been added to poultry diets as an antimicrobial agent at concentrations far in excess of the 8 ppm requirement established by the National Research Council (1994).
Received for publication January 10, 1997. Accepted for publication November 11, 1997. 1Florida Agricultural Experiment Station Journal Series Number R05524. 2To whom correspondence should be addressed:
[email protected] 3Present address: Institute of Animal Sciences, Chinese Academy of Agricultural Sciences, 2 Yuanmingyuan West Rd., Haidian, Beijing 100094, People’s Republic of China.
Abbreviation Key: AV = anisidine value; PV = peroxide value; TBCC = tribasic copper chloride.
416
417
COPPER SULFATE VS TRIBASIC COPPER CHLORIDE FOR BROILERS
bility experiments in which the mineral content of the basal diet represents too large a proportion of the added amount of the mineral under investigation. Ledoux et al. (1991) demonstrated the use of liver Cu accumulation in response to high dietary Cu concentrations to estimate Cu source bioavailability for poultry. A major advantage of this method is that differences among sources can be detected with fewer observations. Other advantages of this method include the use of less expensive natural diets that allow animals to reach their full genetic growth potential and reduction of concern for contamination of diets and samples. Tribasic Cu chloride (TBCC) occurs naturally as the mineral, atacamite, and was first discovered in the Atacama Province of northern Chile in an area known as the Atacama Desert (Palache et al., 1944). This form of Cu is a secondary mineral that is formed by the oxidation of other Cu-containing deposits, especially under arid saline conditions. In the laboratory, TBCC can also be produced as crystals by heating Cu2O with a solution of FeCl3 in a sealed tube or by treating a mixed solution of NaCl with basic Cu nitrate (Palache et al., 1944). A major industrial source of Cu suitable for production of TBCC is the Cu remaining in the form of etchings arising from the manufacture of circuits. Very few, if any, studies have been conducted with poultry to determine the relative bioavailability of Cu in TBCC. Spears et al. (1997) reported that the Cu in TBCC was more available to steers than that of Cu sulfate when supplemented to diets high in the Cu antagonists, Mo and S. When evaluated in Cu-deficient steers fed diets low in Mo, the two Cu sources were of similar bioavailability. It is known that Cu can have prooxidant activity in meat systems (Love, 1985, 1987). Ke and Ackman (1976) reported that 10 ppm Cu sulfate greatly catalyzed the oxidation rate in mackerel lipids. The primary role of metals such as Cu is to accelerate the breakdown of hydroperoxides to radicals (Waters, 1971; Frankel, 1985). Sato and Hegarty (1971), however, have reported that Cu sulfate at 50 and 100 ppm inhibits lipid oxidation in a refrigerated cooked beef model system. Interestingly, Cu chloride at 150 ppm greatly inhibited oxidation. The TBA value (a measure of oxidation) decreased from 0.880 to 0.057 with the addition of 150 ppm CuCl2 (Sato and Hegarty, 1971). No literature reports of the oxidation potential of TBCC in mixed diets of poultry were found. The following three experiments were conducted to estimate the relative bioavailability and safety of TBCC and Cu sulfate for poultry and to study prooxidant activity in diets of both Cu sources at various particle sizes and dietary Cu concentrations.
MATERIALS AND METHODS
Experiment 1 An experiment was conducted to compare the bioavailability of Cu from TBCC [Cu2(OH)3Cl] with that of feed-
TABLE 1. Composition of basal diets Experiment 2 Item
Experiment 1
Starter
Grower
(%) Ingredient composition1 Ground yellow corn Soybean meal (48% CP) Dicalcium phosphate Ground limestone Salt, iodized DL-methionine Corn oil Microingredients2 Variable3 Coban4 Chemical composition5 Dry matter Cu, ppm 1As-fed
55.70 37.15 1.70 1.00 0.40 0.25 2.50 0.50 0.80 . . .
57.28 33.87 2.35 1.05 0.40 0.18 3.80 0.50 0.50 0.07
63.50 28.43 1.90 1.25 0.40 0.11 3.34 0.50 0.50 0.07
87.8 26.0
88.6 20.0
87.5 11.4
basis.
2Ingredients supplied per kilogram of diet: vitamin A palmitate, 6,600
IU; cholecalciferol, 2,200 IU; menadione dimethylpyrimidinol bisulfite, 2.2 mg; riboflavin, 4.4 mg; pantothenic acid, 13 mg; niacin, 40 mg; choline chloride, 500 mg; vitamin B12, 22 mg; biotin, 1 mg; ethoxyquin, 125 mg; manganese, 60 mg; iron, 50 mg; copper, 6 mg; zinc, 36 mg; iodine, 1.1 mg; selenium, 0.1 mg. 3Copper sources added in place of an equivalent weight of cornstarch in Experiment 1 and washed builders’ sand in Experiment 2. 4Elanco, Indianapolis, IN 46285. 5DM basis, by analysis.
grade Cu sulfate (CuSO4·5H2O). The basal corn-soybean meal diet (Table 1) containing 26 ppm Cu was formulated to meet the requirements of starting chicks (National Research Council, 1994). The Cu sources were added to the basal diet at 0, 150, 300, and 450 ppm Cu and fed for 21 d. There were six pens of six Ross × Ross chicks (three males and three females per pen) for the control treatment and seven pens of six chicks (three males and three females per pen) for each Cu-supplemented treatment for a total of 288 1-d-old chicks. Birds were housed in two Petersime brooder units with stainless steel feeders, waterers, and gates and maintained on a 24-h constant light schedule. Feed and tap water containing no detectable Cu, by analysis, were available for ad libitum consumption. At the end of the experiment, feed intake per pen was recorded and birds were individually weighed and then killed by cervical dislocation. The heaviest and lightest birds from each pen were discarded and livers were collected from the remaining four and frozen individually in heat sealed plastic bags for Cu analysis. In this experiment and Experiment 2, birds were handled in accordance with practices approved by the Office of the University of Florida Veterinarian. Copper concentrations in Cu sources were determined after refluxing in 1:1 (vol:vol) concentrated HNO3 and HCl for 4 h on a hotplate. Feed and liver samples were dried at 105 C for 12 h, livers were predigested in HNO3, then all samples were dry ashed at 550 C for 12 h and solubilized in HCl. Tap water was concentrated 10-fold by evaporation on a hotplate. Copper concentra-
418
MILES ET AL.
tions were determined by flame atomic absorption spectrophotometry4 (Anonymous, 1982). Standards were matched for acid and macroelement concentrations as needed and standard reference materials were included with samples.5 Solubility of 0.1 g of the Cu sources was determined in 100 mL of water, 0.4% HCl, 2% citric acid, and neutral ammonium citrate with constant stirring at 37 C (Watson et al., 1970). Particle size separation through Number 20, 100, and 200 mesh screens5 was done on approximately 10 g of each Cu source.
Experiment 2 This experiment was conducted with 1,260 1-d-old Ross × Ross broiler chicks to study performance and liver Cu accumulation after 42 d of supplementation with feedgrade Cu sulfate or TBCC and verify bioavailability estimates determined in Experiment 1. The feather-sexed chicks, which had been vaccinated against Marek’s disease and Infectious Bursal Disease, were assigned to floor pens (0.13 m2 per bird) in a randomized complete block design. New wood shavings at a depth of approximately 8 cm were used as litter material and each pen contained one hanging heat lamp, one tube feeder, and one Plasson waterer. There were 18 birds per pen and 10 pens (5 with males and 5 with females) per dietary treatment. Tap water containing no detectable Cu and feed were available for ad libitum consumption and lighting was continuous. The corn-soybean meal basal starter (0 to 3 wk) and grower (3 to 6 wk) diets (Table 1) contained 20.0 and 11.4 ppm Cu (DM basis), respectively. Dietary treatments were the basal diet supplemented with 0, 200, 400, or 600 ppm added Cu from either source. Feed consumption, body weight, and total mortality were recorded for each pen and feed conversion calculated. At the end of the experiment, three birds selected randomly from each pen were killed by cervical dislocation and livers were collected individually and frozen in heat-sealed plastic bags for Cu analysis as described in Experiment 1.
Experiment 3 This experiment was conducted to compare the stability over time of poultry diets supplemented with feed-grade Cu sulfate and TBCC. Coarse Cu Sulfate. A sample of commercial feedgrade material used in Experiments 1 and 2 was sieved to remove fines. The modal particle size was 455 mm and 95% confidence limits were 85 to 1,270 mm. Fine Cu Sulfate. The material from the sieve operation above was manually ground with a mortar and pestle.
4Model 5000 with AS-50 autosampler, Perkin-Elmer, Norwalk, CT 06859-0156. 5National Institute of Standards and Technology, Gaithersburg, MD 20899-0001.
This resulted in a relatively wide distribution of particle sizes with a modal particle size of 11.9 mm and a 95% confidence interval from 0.52 to 74.3 mm. TBCC. The TBCC had a modal particle size of 67 mm and a 95% confidence interval from 33 to 151 mm. Copper was added to the Experiment 2 starter diet sources at 25, 100, and 300 ppm. Coarse Cu sulfate was tested to evaluate the significance of particle size as a factor in prooxidant activity. Oxidative Stability. The oxidative stability of the different diets was determined using procedures described by O’Keefe et al. (1993). The diets were placed in beakers that were distributed randomly in a 37 C oven for the 20-d experimental period. At 4-d intervals, two samples from each treatment were removed and placed in glass containers, and the head space gas was flushed with nitrogen. Containers were closed and frozen at –30 C. Lipid was extracted from samples using the method of Bligh and Dyer (1959). The solvent was removed at low temperature (< 45 C). The peroxide value (PV) was determined on the extracted oil using the AOCS method Cd 8-53 (Walker, 1989). The anisidine value (AV) was measured using the official IUPAC method (Paquot, 1979).
Statistical Analyses In Experiment 1, data were analyzed by least squares analysis of variance using the General Linear Models procedure of the SAS Institute (1988). The model included the main effects of Cu source and dietary Cu concentration and their interaction. Pen was used as the experimental unit. For liver Cu concentration and body weight only, the mean of the four birds used for liver samples was used in the analysis. Liver Cu concentrations exhibited variance heterogeneity and were subjected to log10 transformation prior to analysis. Results of previous studies with Cu in chicks (Ledoux et al., 1989a,b; 1991) indicated that liver Cu uptake could not be assumed to be linear down to the zero added Cu level. Therefore, separate linear regressions were fitted over the nonzero levels within each source according to the procedure described by Freund and Littell (1981). Feed intake differed for the dietary treatments, so Cu intake rather than dietary Cu concentration was used as the independent variable for regression analysis. The mean pen log10 transformed liver Cu concentration was regressed on added Cu intake (average pen feed intake × added dietary Cu concentration), calculated total Cu intake (average pen feed intake × calculated total dietary Cu concentration (basal + added), and analyzed total Cu intake (average pen feed intake × analyzed dietary Cu concentration). Performance data in Experiment 2 were analyzed by least squares analysis of variance using the General Linear Models procedure (SAS Institute, 1988). The model included the main effects of Cu treatment, block, sex, and suitable interactions as error terms. Pen was used as the experimental unit. Liver Cu concentrations exhibited variance heterogeneity and were subjected to log10 transformation prior to analysis. When block was found to be nonsignificant for liver Cu concentration, another
419
COPPER SULFATE VS TRIBASIC COPPER CHLORIDE FOR BROILERS
model was fitted using Cu source, dietary Cu concentration, and sex as main effects and their interactions. Separate linear regressions were fitted over nonzero levels within each source according to the procedure described by Freund and Littell (1981). Feed intake differed by dietary treatment and sex of birds, so Cu intake as well as added and total dietary Cu concentrations were used as independent variables for regression analysis. Data in Experiment 3 were analyzed as a completely randomized design with a factorial arrangement of treatments that included two Cu sources, three Cu concentrations, six time periods, and their respective interactions. The ANOVA was done using Statistics for Windows (version 4.5).6 When treatment effects were significant, mean separation was conducted using an LSD test.
RESULTS
Copper Samples The physical and chemical characteristics of the Cu sources are given in Table 2. Analyzed Cu concentrations were 25.4 and 55.6% for Cu sulfate and TBCC, respectively (Table 2). The sulfate was highly soluble in all solvents, but the chloride was insoluble in water. The Cu sulfate had a greater percentage of larger particles than TBCC. All of the TBCC sample passed a No. 20 sieve and 98.5% was retained on a No. 100 sieve, indicating uniformity of particle size. Neither source had significant contamination with other mineral elements.
Experiment 1 Chicks fed 450 ppm Cu as sulfate had lower (P < 0.05) feed intakes than those fed the other diets (Table 3). Feed intake of chicks consuming TBCC at all dietary concentrations was similar to that of chicks fed the control diet. Copper sulfate and copper acetate at 450 ppm added Cu have been reported to decrease feed intake of chicks, but Cu carbonate and Cu oxide had no effect (Ledoux et al., 1991). Feeding Cu carbonate and oxide also resulted in less uptake of Cu by the liver than from feeding Cu sulfate, which was not the case with TBCC in this experiment. Feeding supplemental Cu increased (P < 0.0001) liver Cu concentrations in chicks regardless of the Cu source. Liver Cu concentrations increased linearly (P < 0.0001) with increasing dietary Cu. The lack of a significant interaction indicated that the slopes representing the two sources over increasing concentrations did not differ (P > 0.10). Estimates of relative bioavailability of Cu in TBCC compared with 100% for Cu sulfate ranged from 90 to 106% depending upon the independent variable used in
6StatSoft
Inc., Tulsa, OK 74101.
TABLE 2. Physical and chemical characteristics of copper sources Item
CuSO4·5H2O
Cu2(OH)3Cl (%)
Solubility1 H2O 0.4% HCl 2% Citric acid Neutral ammonium citrate Particle size2 +20 –20 + 100 –100 + 200 –200 Chemical constituents Cu Zn Fe Mn Ca Mg P Physical appearance
99 100 100 100 33.3 65.9 0.8 0.0 25.4 18.0 . . .3 . . . 142.0 37.0 . . . Light blue, crystals
0.10) by sex.
420
MILES ET AL. TABLE 3. Effect of copper source and dietary copper concentration on performance and liver copper concentration of broiler chicks fed 21 d, Experiment 11
Cu source Control CuSO4·5H2O
Cu2(OH)3Cl
Added Cu 0 150 300 450 150 300 450
Dietary Cu2
Feed intake
(ppm) 26 175 341 529 161 305 468
831 785 767 695 786 839 820
ANOVA Cu source (S) Cu concentration (C) S × C
± ± ± ± ± ± ±
Body weight 14.5 23.7 14.8 21.8 25.6 17.3 30.0
0.0006 0.1252 0.0243
(g) 605 ± 13.1 595 ± 18.1 568 ± 12.2 485 ± 20.3 594 ± 12.3 633 ± 9.3 619 ± 23.4 Significance 0.0001 0.0090 0.0008
Liver Cu, DM basis (ppm) 18 ± 0.54 25 ± 2.36 102 ± 17.1 185 ± 30.1 20 ± 1.90 72 ± 11.4 204 ± 25.5 0.00013 0.0001 0.2163
1Each value represents the mean ± SE of six pens for control and seven pens for all Cu treatments (six birds per pen for feed intake and body weight and four birds per pen for liver Cu). 2DM basis, by analysis. 3ANOVA of log 10 transformed liver Cu concentrations.
Feeding supplemental Cu increased (P < 0.0001) liver Cu concentrations of chicks regardless of the Cu source (Table 6). The probability value for the Type III SS for main effect of source was 0.88, indicating that effect of source was the difference between the control and either source and not between the two sources. Liver Cu concentrations increased linearly (P < 0.0001) with increasing dietary Cu. The lack of a significant interaction between Cu source and dietary Cu concentration indicated that slopes of response lines representing the two sources over increasing concentrations did not differ (P > 0.10). Although the female birds consumed less feed than the male birds, there was no difference (P > 0.10) in liver Cu concentration between sexes. Concentrations of liver Cu, however, averaged 271 and 254 ppm (DM basis) for females and males, respectively. Therefore, regressions based on daily Cu intake rather than dietary Cu concentration had poorer fits to a linear model. There were no interactions (P > 0.10) for liver Cu concentration between sex and other main effects. Regression equations of log10 transformed liver Cu concentration on dietary Cu over the nonzero addition levels are found in Table 7. There was excellent fit to the linear model (r2 = 0.93 and 0.88) observed with equations
using added dietary Cu concentration as the independent variable. The resulting slope-ratio calculation estimated a relative value of 104.2% for the chloride compared to 100% for the sulfate standard. However, as feed intake varied among treatments, regressions based on Cu intake rather than dietary Cu concentration are also presented in Table 7, although they display poorer fits to the linear model because females had an inverse relationship between feed intake and liver Cu concentration. The equations for the latter two comparisons gave a relative bioavailability estimate of 112% for Cu from TBCC compared to 100% for the standard. The estimated relative bioavailability for chicks of Cu from TBCC was equal to that from feed-grade Cu sulfate. The decrease in feed intake in birds given 600 ppm Cu as TBCC may have been due to a slightly rancid odor in the feed (observed toward the end of the experiment) and/or taste of this diet, but liver Cu was still similar to that in birds given Cu sulfate.
Experiment 3 Initial evaluations of the stored feed samples were made subjectively using three sensory panelists that had
TABLE 4. Linear regression of log10 transformed liver copper concentrations on various independent variables over non-zero levels and estimated relative bioavailability of copper sources for broiler chicks fed 21 d, Experiment 11
Independent variable
Cu source
Intercept
Slope ± SE (×10–6)
Added Cu intake, mg
Sulfate Chloride Sulfate Chloride Sulfate Chloride
0.949 0.869 0.864 0.798 0.967 0.858
4.13 3.78 4.15 3.75 3.48 3.69
Calculated total Cu intake, mg Analyzed total Cu intake, mg 1Each
equation represents 21 pens with four chicks per pen.
± ± ± ± ± ±
0.58 0.38 0.59 0.38 0.50 0.38
r2
Standard deviation
Relative value ± SE
0.73 0.84 0.72 0.83 0.72 0.84
0.215 0.183 0.217 0.186 0.219 0.184
100 91.5 ± 15.8 100 90.4 ± 15.8 100 106.0 ± 18.7
421
COPPER SULFATE VS TRIBASIC COPPER CHLORIDE FOR BROILERS TABLE 5. Effect of copper source, dietary copper concentration, and sex on performance and liver copper concentration of broilers fed 42 d, Experiment 21 Cu source
Added Cu2
Control
(ppm) 0
CuSO4·5H2O
200 400 600
Cu2(OH)3Cl
200 400 600
Sex Male Female Male Female Male Female Male Female Male Female Male Female Male Female
Pooled SE ANOVA Treatment (T) Replicate (R) T × R Sex (G) T × G 1Each 2Basal
Feed intake 3,725 3,169 3,683 3,035 3,434 3,032 3,325 2,848 3,591 3,194 3,550 3,182 2,779 2,432 17.6
Body weight (g) 2,139 1,780 2,134 1,707 1,986 1,684 1,543 1,394 2,060 1,787 1,966 1,741 1,335 1,187 12.9
Feed conversion Mortality (g:g) 1.74 1.78 1.73 1.78 1.73 1.80 2.16 2.04 1.74 1.79 1.81 1.83 2.10 2.06 0.010
(n) 4 4 1 3 4 1 3 2 3 0 3 1 1 7
0.0001 0.1524 0.6483 0.7126 0.1713
0.1456 0.5191 0.8269 0.9172 0.0853
Probability 0.0001 0.2497 0.3851 0.0001 0.2491
0.0001 0.5428 0.9474 0.0005 0.0605
value represents the mean of five pens (18 birds per pen). starter diet contained 20 ppm Cu and the basal grower diet contained 11.4 ppm Cu.
extensive experience with oxidation odors. Blind sniffing showed that “painty” odors began by Day 4 in the samples containing fine Cu sulfate and were more predominant for higher concentrations of Cu. No differences in odor between TBCC and coarse Cu sulfate samples were noted at any storage period. The painty odors originate from hexanal, which is a product of linoleic (C18:2n6) acid oxidation (Frankel, 1985). Changes in PV are shown in Table 8. The PV measures the hydroperoxide and other peroxides in the oxidizing samples. Peroxides are transitory and will eventually degrade to secondary oxidation products. The fine Cu sulfate at all levels caused a greater increase in PV than TBCC or coarse Cu sulfate, but by Day 16 the PV were either lower (300 ppm, P = 0.003) or not different (P > 0.05) from the other samples. The decline in PV was expected and does not represent a lower level of oxidation. The AV measures aldehydes (Paquot, 1979) (Table 9). Because aldehydes are secondary oxidation products, the AV is usually less sensitive at initial oxidation. In agreement with the PV, fine Cu sulfate caused the fastest rate of oxidation. The AV for TBCC samples were lower than coarse Cu sulfate at 300 ppm at Day 8 and later, but the difference was significant only at Day 16 (P = 0.013) and 20 (P = 0.005). There were no significant differences in AV between TBCC and coarse Cu sulfate samples at 100 and 25 ppm added Cu. It was noteworthy that the TBCC promoted less oxidation than coarse Cu sulfate at 300 ppm added Cu, as the modal particle size of TBCC was lower (67 mm) than Cu sulfate (455 mm). Because the surface area of a sphere equals D2p (where D is the diameter and p = 3.1416), the modal surface area of TBCC would be 46 times smaller
TABLE 6. Effect of copper source, dietary copper concentration, and sex on liver copper concentration of broilers fed 42 d, Experiment 21
Cu source
Added Cu2 Sex
Control
(ppm) 0
CuSO4·5H2O
200 400 600
Cu2(OH)3Cl
200 400 600
ANOVA Source (S) Concentration (C) S × C Sex (G) S × G C × G S × C × G 1Each
Male Female Male Female Male Female Male Female Male Female Male Female Male Female
Liver Cu, DM basis (ppm) 14.1 ± 1.1 13.9 ± .91 17.2 ± 1.6 19.0 ± 1.8 98.4 ± 25 105.7 ± 29 675.7 ± 84 749.5 ± 163 18.5 ± 2.3 16.6 ± 1.2 155.3 ± 43 48.9 ± 7.6 799.1 ± 130 940.4 ± 287 Probability 0.00013 0.0001 0.6562 0.1907 0.1536 0.0869 0.1271
value represents the mean of five pens (three birds per pen). starter diet contained 20 ppm Cu and the basal grower diet contained 11.4 ppm Cu. 3ANOVA of log 10 transformed liver Cu concentrations. 2Basal
422
MILES ET AL.
TABLE 7. Linear regression of log10 transformed liver copper concentration on various independent variables over non-zero levels and estimated relative bioavailability of copper sources for broiler chicks fed 42 d, Experiment 21 Independent variable
Cu Source
Intercept
Slope ± SE (×10–6) r2
Standard deviation Relative value ± SE
Added Cu, ppm
Sulfate Chloride Sulfate Chloride Sulfate Chloride
0.432 0.358 0.414 0.288 0.352 0.223
3,940 ± 201 4,106 ± 289 1.747 ± 0.15 1.965 ± 0.28 1.744 ± 0.15 1.955 ± 0.29
0.180 0.259 0.278 0.447 0.283 0.456
Added Cu intake, mg Total Cu intake, mg 1Each
0.93 0.88 0.84 0.64 0.83 0.62
100 104.2 ± 15.8 100 112.5 ± 18.7 100 112.1 ± 15.8
equation represents 30 pens (three birds per pen).
than that of coarse Cu sulfate, assuming spherical particle geometry.
DISCUSSION As stated by Fairweather-Tait (1987), bioavailability should not be considered to be an inherent property that is a characteristic of a mineral source, but rather an experimental estimate that reflects absorption and utilization under conditions of that specific test. Generalizations and attempts to establish a unique bioavailability value for a mineral source can lead to misconceptions regarding the use of supplemental sources of elements. The design of a bioavailability experiment, dietary considerations, and method used to calculate the estimate can all have a profound effect on the bioavailability value for a particular mineral source (Ammerman, 1995). Estimates of relative Cu bioavailability for supplemental sources have been reported to be similar
whether determined at low or high dietary concentrations. Ledoux et al. (1991) reported a value of 67% for Cu carbonate relative to 100% for Cu acetate based on the liver Cu concentration of chicks consuming diets that contained from 150 to 450 ppm supplemental Cu. Using these same sources, Zanetti et al. (1991) reported a value of 66% for the Cu in the carbonate relative to 100% for that in Cu acetate based on liver Cu concentration of chicks supplemented with 5, 10, or 20 ppm Cu in the diet. The analyzed Cu concentration of the basal diet in this experiment was approximately 5 ppm. Aoyagi and Baker (1993b) obtained a value of 145% for Cu in Cu chloride compared with 100% for that in Cu sulfate based on liver Cu accumulation in chicks supplemented with from 50 to 200 ppm Cu. In a previous experiment (Aoyagi and Baker, 1993a), these authors reported a value of 142% for the Cu in Cu chloride compared with 100% for that in Cu sulfate in Cu-deficient birds supplemented with 0.5 or 1 ppm Cu and using bile Cu concentration as the dependent variable. Considering
TABLE 8. Effect of copper source and copper concentration on peroxide values of lipid extracted from stored diets, Experiment 31 Time of storage Cu sample2 Sulfate – C Sulfate – F TBCC Sulfate – C Sulfate – F TBCC Sulfate – C Sulfate – F TBCC Pooled SE ANOVA Source (S) Concentration (C) S × C Time (T) S × T C × T S × C × T 1Each
Added Cu
0 d
4 d
(ppm) 25 25 25 100 100 100 300 300 300
2.0 2.2 1.4 1.7 3.5 1.9 2.0 2.0 1.9
24.8 57.9 22.8 22.8 240.6 26.5 27.4 270.0 27.9
8 d (meq/kg) 30.7 211.9 36.1 40.8 129.4 35.5 56.5 70.3 58.7 2.95 Probability 0.0001 0.0007 0.0002 0.0001 0.0001 0.0001 0.0001
16 d
20 d
37.3 96.5 63.0 66.0 48.9 58.7 133.1 22.2 109.5
29.4 74.2 26.5 44.2 36.4 37.9 99.5 29.4 62.5
value represents the mean of two samples per subtreatment group. C and F indicate coarse and fine sulfate and TBCC is tribasic Cu chloride.
2Where
423
COPPER SULFATE VS TRIBASIC COPPER CHLORIDE FOR BROILERS TABLE 9. Effect of copper source and copper concentration on anisidine values1 of lipid extracted from stored diets, Experiment 32 Time of storage Cu sample3 Sulfate – C Sulfate – F TBCC Sulfate – C Sulfate – F TBCC Sulfate – C Sulfate – F TBCC Pooled SE ANOVA Source (S) Concentration (C) S × C Time (T) S × T C × T S × C × T
Added Cu
0 d
4 d
(ppm) 25 25 25 100 100 100 300 300 300
37 34 25 24 25 35 27 25 19
83 128 87 94 342 99 124 684 116
8 d (Anisidine Value) 147 421 142 168 819 152 268 898 199 76.2 Probability 0.0001 0.0001 0.0666 0.0001 0.0001 0.0424 0.0066
16 d
20 d
143 289 112 212 291 191 316 307 238
151 326 151 195 272 162 303 310 208
1Expressed as 100 × absorbance in a 1-cm cell with a solution of 1 g oil/100 mL of mixture of solvent and reagent. 2Each value represents the mean of two samples per subtreatment group. 3Where C and F indicate coarse and fine sulfate and TBCC is tribasic Cu chloride.
the above finding, it must be kept in mind that the chloride form of Cu used in these experiments by the above mentioned researchers is not the same as that used in the present study (TBCC). Relative bioavailability of Cu chloride has consistently been equal to or exceeded that for Cu sulfate. In addition to the experiments discussed above, Norvell et al. (1974) reported a value of 110% for Cu from Cu chloride compared with 100% for Cu sulfate when based on liver Cu concentration in chicks fed 720 ppm added Cu for 8 wk. In later experiments with the same sources and dietary concentrations, Norvell et al. (1975) reported that Cu from Cu chloride was 106% as available as that in their standard source, Cu sulfate. All of those experiments were three-point assays with a control diet containing approximately 15 ppm Cu, and are in agreement with results obtained in the present studies. The National Research Council (1994) lists the toxic concentrations of inorganic elements and compounds for poultry. Toxicity, as defined in that publication, is any adverse effect on performance. Also stated is that the toxicity of a mineral is influenced by the nature of the compound in which it is present. Because many different factors affect the quantity of a element needed to produce toxicity, diverse toxic effects have been reported for any given mineral (National Research Council, 1994). Dietary concentrations of Cu as high as 480 ppm as Cu sulfate have been fed to laying hens without any adverse effect on egg production (Thomas and Goatcher, 1976). Jensen (1975) fed 800 ppm Cu in Cu sulfate to turkeys in a practical diet and reported no adverse effects. The present study was designed to evaluate the safety of TBCC by comparing its effect to
Cu sulfate when supplying the same dietary concentration of Cu. Its lack of a detrimental effect on feed intake may indicate that this source is less toxic for chicks than Cu sulfate. Body weight of chicks fed 450 ppm Cu as sulfate was also lower (P < 0.05) than that of chicks fed all other diets (Table 3). Body weight of birds fed all concentrations of TBCC did not differ from those fed the basal diet. Overall mortality for the study was 2.08%, which was within the expected loss for birds up to 3 wk of age (Ensminger, 1980). The bioavailability of Cu in TBCC was equal to that of Cu sulfate for chicks and the detrimental effect of Cu sulfate on feed intake was not observed with TBCC. Thus, from a standpoint of both bioavailability and safety of the product, it appears that TBCC can be used by the feed industry as a source of supplemental Cu. Anisidine and peroxide values of the diets were strongly affected by time, Cu source, and dietary Cu concentration in the present study. Both AV and PV measure oxidation. Peroxide value measures the first products, hydroperoxides and peroxides, which are transient and decompose to aldehydes and ketones, whereas AV measures these secondary oxidation products. The results for PV and AV were comparable for TBCC and Cu sulfate. As Cu level increased, the extent of oxidation increased. The fine Cu sulfate catalyzed oxidation to a much greater extent than coarse Cu sulfate. The average particle size in TBCC was smaller than coarse Cu sulfate, but the extent of oxidation was similar or lower, especially for the 300 ppm treatment. Because of the significant effect of particle size on oxidation for Cu sulfate, it is likely that larger particles of TBCC would promote oxidation much less than equivalent-sized Cu sulfate.
424
MILES ET AL.
As observed in this experiment, fine particle size in Cu sulfate results in much greater oxidation in poultry diets. Because experiments with different particle sizes have not been conducted for TBCC, it is unclear whether the same effect will be seen. Commercial Cu sulfate will undoubtedly contain more fines than the coarse Cu sulfate used in this study, which was sieved. Therefore, it is likely that the TBCC used in this study will promote even less oxidation than feed-grade Cu sulfate. Further work is required to validate this hypothesis. There is a void in the literature on Cu-induced oxidation in poultry diets. Copper has been shown to be a prooxidant (Ke and Ackman, 1976; Love, 1987) or antioxidant (Sato and Hegarty, 1971) in meat systems. However, since this report, the antioxidant effect has never been substantiated. Both Cu sulfate and TBCC were prooxidant in the present study. The amount of water in the system may have an effect on observed rates of oxidation. Because TBCC is not water-soluble, this may explain why the smaller particles of TBCC did not promote the rate of oxidation observed with large particles of Cu sulfate. Further work is needed to clearly compare the effects of Cu from different sources and particle sizes on oxidation. Results from the present study show that particle size must be similar for comparisons to be valid. Commercially available Cu sulfate containing fines may promote oxidation to a greater extent than an equivalent or comparable product with fines removed. Results of the present experiments indicate that the Cu in TBCC is as available and as safe for broilers as that in feed-grade Cu sulfate. Supplementation of diets for poultry with Cu from TBCC at concentrations commonly recommended in the commercial industry should not cause any adverse effect on bird performance.
ACKNOWLEDGMENTS The authors wish to acknowledge Microingredients, Indianapolis, IN 46231 for supplying TBCC and funds in support of this research and Southeastern Minerals, Bainbridge, GA 31717 for supplying Cu sulfate.
REFERENCES Ammerman, C. B., 1995. Methods for estimation of mineral bioavailability. Pages 83–94 in: Bioavailability of Nutrients for Animals: Amino Acids, Minerals, and Vitamins. C. B. Ammerman, D. H. Baker, and A. J. Lewis, ed. Academic Press, Inc., San Diego, CA. Anonymous, 1982. Analytical Methods for Atomic Absorption Spectrophotometry. The Perkin-Elmer Corp., Norwalk, CT. Aoyagi, S., and D. H. Baker, 1993a. Bioavailability of copper in analytical-grade and feed-grade inorganic copper sources when fed to provide copper at levels below the chick’s requirement. Poultry Sci. 72:1075–1083. Aoyagi, S., and D. H. Baker, 1993b. Nutritional evaluation of a copper-methionine complex for chicks. Poultry Sci. 72: 2309–2315.
Bligh, E. G., and W. J. Dyer, 1959. A rapid method of total lipid extraction and purification. Can. J. Biochem. Physiol. 37: 911–917. Davis, G. K., and W. Mertz, 1987. Copper. Pages 301–364 in: Trace Elements in Human and Animal Nutrition. 5th ed. Vol. 1. W. Mertz, ed. Academic Press, New York, NY. Ensminger, M. E., 1980. Poultry Science. The Interstate Printers and Publishers, Inc., Danville, IL. Fairweather-Tait, S. J., 1987. The concept of bioavailability as it relates to iron nutrition. Nutr. Res. 7:319–325. Frankel, E. N., 1985. Pages 1–37 in: Flavor Chemistry of Fats and Oils. D. B. Min and T. H. Smouse, ed. American Oil Chemists’ Society, Champaign, IL. Freund, R. J., and R. C. Littell, 1981. SAS for Linear Models. SAS Institute Inc., Cary, NC. Jensen, L. S., 1975. Modification of a selenium toxicity in chicks by dietary silver and copper. J. Nutr. 105:769–775. Ke, P. J., and R. G. Ackman, 1976. Metal-catalyzed oxidation in mackerel skin and meat lipid. J. Am. Oil Chem. Soc. 53: 636–640. Ledoux, D. R., P. R. Henry, C. B. Ammerman, and R. D. Miles, 1989a. Effect of dietary copper on tissue mineral composition as an estimate of copper bioavailability in broiler chicks. Nutr. Rep. Int. 39:1117–1126. Ledoux, D. R., P. R. Henry, C. B. Ammerman, and R. D. Miles, 1989b. Effect of dietary copper and age on tissue mineral composition of broiler-type chicks as a bioassay of inorganic copper sources. Nutr. Rep. Int. 40:53–66. Ledoux, D. R., P. R. Henry, C. B. Ammerman, P. V. Rao, and R. D. Miles, 1991. Estimation of the relative bioavailability of inorganic copper sources for chicks using tissue uptake of copper. J. Anim. Sci. 69:215–222. Love, J., 1985. Factors affecting lipid oxidation—Metal catalysts and chelators. Pages 61–78 in: Flavor Chemistry of Fats and Oils. D. B. Min and T. H. Smouse, ed. American Oil Chemists’ Society Press, Champaign, IL. Love, J. A., 1987. Mechanism of iron catalysis of lipid oxidation in warmed-over flavor of meat. Pages 19–40 in: WarmedOver Flavor of Meat. A. J. St. Angelo and M. E. Bailey, ed. Academic Press Inc, New York, NY. McNaughton, J. L., E. J. Day, B. C. Dilworth, and B. D. Lott, 1974. Iron and copper availability from various sources. Poultry Sci. 53:1325–1330. National Research Council, 1994. Nutrient Requirements of Poultry. 9th rev. ed. National Academy Press, Washington, DC. Norvell, M. J., M. C. Thomas, W. D. Goatcher, D. A. Gable, and C. C. Calvert, 1974. Some effects of high dietary levels of various salts of copper in broiler chickens. Pages 367–372 in: Trace Substances in Environmental Health. Vol. 8. D. D. Hemphill, ed. University of Missouri, Columbia, MO. Norvell, M. J., D. A. Gable, and M. C. Thomas, 1975. Effects of feeding high levels of various copper salts to broiler chickens. Pages 457–464 in: Trace Substances in Environmental Health. Vol. 9. D. D. Hemphill, ed. University of Missouri, Columbia, MO. O’Keefe, S. F., V. A. Wiley, and D. A. Knauft, 1993. Comparison of oxidative stability of high- and normaloleic peanut oils. J. Am. Oil Chem. Soc. 70:489–492. Palache, C., H. Berman, and C. Fronde, 1944. The System of Mineralogy of James Dwight Dana and Edward Salisbury Dana, Yale University 1837–1892. 7th ed. Vol. II. Halides, nitrates, borates, carbonates, sulfates, phosphates, arsenates, tungstates, molybdates, etc. John Wiley and Sons, New York, NY.
COPPER SULFATE VS TRIBASIC COPPER CHLORIDE FOR BROILERS Paquot, C., 1979. Standard Methods for the Analysis of Oils, Fats and Derivatives, Part 1, 6th ed., Method 2.201. International Union of Pure and Applied Chemists, Pergamon Press, New York, NY. SAS Institute, 1988. SAS User’s Guide: Statistics. SAS Institute Inc., Cary, NC. Sato, K., and G. R. Hegarty, 1971. Warmed-over flavor in cooked meats. J. Food Sci. 36:1098–1102. Spears, J. W., E. B. Kegley, L. A. Mullis, and T. A. Wise, 1997. Bioavailability of copper chloride in cattle. J. Anim. Sci. 75 (Suppl. 1):265. (Abstr.) Thomas, M. C., and W. D. Goatcher, 1976. High levels of copper in layer diets. Poultry Sci. 55:1608. (Abstr.)
425
Walker, R. O., 1989. Official Methods and Recommended Practices of the American Oil Chemists Society, Method Cd 8-53. American Oil Chemists Society, Champaign, IL. Waters, W. A., 1971. The kinetics and mechanism of metal catalyzed autooxidation. J. Am. Oil Chem. Soc. 48:427–433. Watson, L. T., C. B. Ammerman, S. M. Miller, and R. H. Harms, 1970. Biological assay of inorganic manganese for chicks. Poultry Sci. 49:1548–1554. Zanetti, M. A., P. R. Henry, C. B. Ammerman, and R. D. Miles, 1991. Estimation of the relative bioavailability of copper sources in chicks fed on conventional dietary amounts. Br. Poult. Sci. 32:583–588.
