Effects of Lifetime Copper, Zinc, and Manganese Supplementation and Source on Performance, Mineral Status, Immunity, and Carcass Characteristics of Feedlot Cattle1,2

Introduction

Trace minerals, such as Cu, Zn, and Mn, are required for normal tissue growth, homeostasis, enzyme function, and cell regulation and must be maintained within narrow concentrations in the body to ensure optimal growth, health, and performance in domestic livestock (Underwood and Suttle, 1999). Feedlot diets are typically fortified with trace minerals because feedstuffs utilized in feedlot rations normally contain lesser concentrations of essential trace minerals and/or may contain greater concentrations of known trace mineral antagonists (e.g., Mo, Fe, S, etc.). The chemical form of a supplemented trace mineral (oxide, sulfate, organic, etc.) has been reported to impact trace mineral status, possibly because of the differences in availability (Du et al., 1996), absorption (Kegley and Spears, 1994; Du et al., 1996) and/or metabolism (Spears, 1989; Nockels et al., 1993). Furthermore, physiological concentrations of Cu and Zn supplemented to feedlot cattle diets have been reported to decrease fat thickness (Cu; Ward and Spears, 1997), increase longissimus polyunsaturated fatty acids (Cu; Engle et al., 2000a), and increase carcass yield (Zn; Rust and Schlegel, 1993) and quality grade (Zn; Spears and Kegley, 2002).

Several experiments have focused on trace mineral supplementation during the cow-calf (Olson et al., 1999; Stanton et al., 2000) or feedlot (Ward and Spears, 1997; Engle et al., 2000c; Spears and Kegley, 2002; Rhoads et al., 2003) phases. However, the effects of lifetime (pre-natal through harvest) trace mineral supplementation and source on feedlot performance have not been well addressed. Because a large percentage of forages grazed by beef cattle in the US are deficient or marginally adequate in Cu and/or Zn, as well as marginal or high in Cu antagonists such as S, Fe, and Mo (Mortimer et al., 1999), it is possible for calves to enter the feedlot with a marginal trace mineral status. Therefore, the objectives of this experiment were to evaluate the effects of lifetime supplementation (at current NRC recommendations) and source (organic vs inorganic) of Cu, Zn, and Mn on feedlot performance, mineral status, immunity, carcass traits, and longissimus fatty acid profile in beef cattle during the growing and finishing phases.

Materials and Methods

Animal Procedures. Prior to the initiation of this experiment, all care, handling, and sampling of the animals herein was approved by the Colorado State University Animal Care and Use Committee. The effects of Cu, Zn, and Mn supplementation and source on the performance, immunity, and carcass traits of feedlot cattle were evaluated using two calf crops from the same cowherd over 2 consecutive yr. Calf crops consisted of crossbred calves (predominantly Red Angus × Charolais Fl hybrids) that originated from the Colorado State University Eastern Colorado Research Center (ECRC; Akron, CO). In total, 270 calves were transported to the Colorado State University Agricultural Research and Development Education Center Research Feedlot (Fort Collins, CO), where they were fed until harvest. One hundred forty head were included in yr 1 (219.5 ± 5.3 kg; 90 steers, 50 heifers), and 130 head were included in yr 2 (211.2 ± 5.3 kg; 80 steers, 50 heifers).

Prior to arrival to the feedlot in each year, calves and their dams were part of a 2-yr experiment evaluating the effects of trace mineral supplementation and source on grazing beef cow and calf performance (Ahola et al., 2004, 2005). In that experiment, beginning approximately 80 d prior to the average calving date of the cowherd in yr 1, dams were assigned to one of three treatments: 1) control (no supplemental Cu, Zn, or Mn), 2) organic (ORG; 50% organic and 50% inorganic Cu, Zn, and Mn); and 3) inorganic (ING; 100% inorganic CuSO^sub 4^, ZnSO^sub 4^, and MnSO^sub 4^) trace minerals. The ORG and ING ad libitum trace mineral treatments were formulated to supply 10 ppm Cu, 30 ppm Zn, and 40 ppm Mn daily. In both years, treatments were provided ad libitum in mineral feeders from 82 d (yr 1) and 81 d (yr 2) prior to the average calving date of the cowherd through 110 d (yr 1) and 135 d (yr 2) post-calving. When calves reached an average age of 90 d (yr 1) and 99 d (yr 2), calves were provided access to the same respective mineral treatments as their dams via creep feeders in each pasture. Calves continued to have access to mineral treatments exclusively until weaning at an average age of 185 d (yr 1) and 164 d (yr 2). In that experiment, as reported by Ahola et al. (2004), cows were maintained on native pastures that consisted primarily of blue grama (Bouteloua gracilis), prairie sandreed (Calamovilfa longifolia), and needle-and-thread grass (Stipa comata). Basal forage and water trace mineral concentrations were determined using samples collected from pasture, stored hay, and water sources. Trace mineral concentrations were as follows: pasture, 13.1 ppm Cu, 16.1 ppm Zn, and 36.6 ppm Mn; stored hay, 19.6 ppm Cu, 32.1 ppm Zn, and 52.2 ppm Mn; and water, <0.01 ppm Cu, <0.01 ppm Zn, and 0.08 ppm Mn.

In both the cow-calf and feedlot phases of this 2-yr experiment, inorganic trace minerals were supplemented as CuSO^sub 4^, ZnSO^sub 4^ and MnSO^sub 4^; organic trace minerals were provided from a commercially available mineral proteinate source (10% Cu, 15% Zn, 10% Mn; Bioplex® trace minerals; Alltech Inc., Nicholasville, KY). All supplemented minerals were formulated to meet NRC (1996) recommended daily concentrations. Composition of the trace mineral treatments provided to cows and calves has been reported (Ahola et al., 2004).

Growing and Finishing Phases. After shipment from ECRC to the Agricultural Research and Development Education Center (200 km) in both years, calves were individually weighed on 2 consecutive d; vaccinated (Cattle Master 4® and Bovishield®; Pfizer, Exton, PA); dewormed (Dectomax®; Pfizer); blocked by gender, BW, and previous pasture trace mineral treatment; and allotted to one of 15 pens (7 m × 40 m). All calves received the same respective treatments that they and their dam received during the cow-calf phase of the experiment. Feedlot treatments were 1) control (no supplemental Cu, Zn, or Mn), 2) ORG (33% organic and 67% inorganic Cu, Zn, and Mn), and 3) ING (100% inorganic CuSO^sub 4^, ZnSO^sub 4^, and MnSO^sub 4^) trace minerals.

Diets fed during the growing and finishing phases were formulated to meet or exceed NRC (1996) recommendations for all nutrients except Cu, Zn, and Mn (Table 1). The ORG and ING treatments were formulated to supply 10 ppm Cu, 30 ppm Zn, and 20 ppm Mn daily. The mineral analyses of the three trace mineral treatments fed during the feeding period are listed in Tables 2 and 3 for the growing and finishing phases, respectively. During the initial 56-d growing phase, calves were fed a corn silage-based diet. This was followed by a finishing phase where calves were gradually (over a 10-d period) switched over to a high-concentrate diet and fed for an additional period of 140 d (yr 1) and 145 or 181 d (yr 2). In yr 2, equal numbers of cattle per treatment were harvested after receiving the finishing diet for 145 d, and the remaining cattle were harvested after 181 d on the finishing diet. Cattle were fed once daily in the morning in amounts necessary to allow ad libitum access to feed.

Throughout both the growing and finishing phases of the feeding period, daily feed offerings were recorded and orts were measured every 28 d. From these data, ADG, DMI, and a BW gain-to-feed ratio (G:F) were determined for each pen during both the growing and finishing phases. A final BW at the end of the experiment was collected over 2 consecutive d immediately prior to harvest.

On d 28 of the growing phase in both years, all cattle were implanted [steers: Progesterone (200 mg) and estradiol benzoate (20 mg); Synovex-S®; Fort Dodge Animal Health, Fort Dodge, IA; heifers: trenbolone acetate (200 mg); Finaplix-H®; Intervet, Inc. Millsboro, DE]. All cattle were re-implanted on d 28 (yr 1) and d 56 (yr 2) of the finishing phase [steers: trenbolone acetate (80 mg) and estradiol benzoate (16 mg); Revalor-IS®; Intervet, Inc.; heifers: trenbalone acetate (200 mg); Finaplix-H®].

Mineral Status. To compare mineral status across treatments at different time points and throughout the entire feeding period, blood and liver samples were collected from a subgroup of randomly selected animals (three per pen). Blood was collected from the same animals every 28 d throughout the entire feeding phase via jugular venipuncture into heparinized trace mineral-free vacutainer tubes (Becton Dickinson Co., Franklin Lakes, NJ). Once collected, samples were placed on ice for approximately 2 h, transported to the laboratory, and centrifuged at 2000 × g for 15 min at room temperature. Plasma was removed and transferred to acidwashed polyethylene storage vials and stored at -20°C. Plasma was collected in both years; however, plasma concentrations of Cu and Zn were determined for yr 1 only. Plasma samples were analyzed for Cu and Zn concentrations as described by Ahola et al. (2004).

Liver biopsy samples were collected (from the same animals that were used for blood collection) on d 0 and 56 of the growing phase and on d 112 (yr 1) and 140 (yr 2) of the finishing phase using the true-cut technique described by Pearson and Craig (1980), as modified by Engle and Spears (2000b). Briefly, a 10-cm × 10-cm area was clipped of hair on the right side of each animal between the 11th and 12th ribs and scrubbed three times with iodine and 70% alcohol. Approximately 5 mL of 2% lidocaine hydrochloride (Abbott Laboratories, North Chicago, IL) was injected via a 20-ga × 2.5-cm needle between the 11th and 12th ribs on a line from the hip to the point of the shoulder. A small incision (approximately 1.0 cm) was made using a scalpel blade (number 11), and a core sample of liver tissue was collected using a modified Jan Shide bone marrow biopsy punch (0.5 cm × 14 cm; Sherwood Medical, St. Louis, MO). Following collection, samples were immediately rinsed with 0.01 M PBS (pH = 7.4) and placed into acidwashed polyethylene tubes, capped, placed on ice for approximately 4 h, and stored at -20°C. Liver samples were analyzed for Cu, Zn, and Mn concentrations as described by Ahola et al. (2004).

Immune Response and Health Status. An in vivo evaluation of cell-mediated immune (CMI) response was performed via the use of phytohemagglutinin (PHA; Sigma-Aldrich, St. Louis, MO) to stimulate an immune response (Fritz et al., 1990) using a select subset of animals (three per pen). On d 33 (yr 1) and 43 (yr 2) of the finishing phase, each animal was restrained in a squeeze chute, and a square 10-cm × 10-cm area of hair was clipped on the animal's left side immediately posterior to the scapula. Within the clipped area, in each of two separate sites approximately 2.5 cm apart, 0.1 mL of a PHA solution (150 µg PHA in 0.1 mL PBS) was injected intradermally with a 1-mL tuberculin syringe and 26-ga needle (Becton Dickinson Co.). The skin-fold thickness was measured (in mm) at both injection sites using skin-fold calipers (Vernier Type 6914; Scienceware, Pequannock, NJ) immediately prior to PHA injection (0 h) and at 4, 8, 12, 24, and 48 h post-injection. The two measurements collected on each animal were averaged, and the change in skin thickness was used to evaluate the CMI response to PHA.

Two primary humoral immune responses (during the growing and finishing phases) and one secondary humoral immune response (during the finishing phase only) were compared across treatments via administration of one or more foreign proteins into cattle. On d 28 (yr 1) and 31 (yr 2) of the growing phase, 5 mL of a 20% porcine red blood cell (PRBC) solution (Sigma-Aldrich) diluted in autoclaved PBS (pH = 7.4) was injected i.m. into the neck of a subset of randomly selected animals (three per pen) to elicit and evaluate a primary immune response during the growing phase. Blood was collected via jugular venipuncture into non-heparinized vacutainer tubes (Becton Dickinson Co.) immediately prior to PRBC injection (d 0), and on 7, 14, and 21 d post-injection from the same animals. Samples were stored on ice for approximately 2 h prior to being centrifuged at 2,000 × g for 15 min. Serum was harvested and stored in polyethylene tubes at -80°C until analyzed.

A secondary humoral immune response to PRBC was evaluated on d 84 (yr 1) and 80 (yr 2) of the finishing phase in 22 head (randomly selected, 1 to 2 head per pen) of the 45 animals previously used to evaluate a primary humoral immune response to PRBC. Five milliliters of a 20% PRBC solution, identical to that used to evaluate primary humoral immune response during the growing phase, was injected i.m. into the neck of each animal to elicit a secondary humoral immune response. Blood was collected via jugular venipuncture into non-heparinized vacutainer tubes immediately prior to the second PRBC injection (d 0) and at 7, 14, and 21 d post-injection from the same animals. Samples were handled as previously described.

Serum samples collected from animals administered PRBC were thawed, and antibody titers to PRBC were measured using a microtiter hemagglutination assay (Ferket and Qureshi, 1992). Concentrations of total Ig, IgG, and IgM specific for PRBC were determined.

A second evaluation of primary humoral immune response was performed during the finishing phase using ovalbumin (OVA) as the antigen (Ward et al., 1993). Of the original 45 head previously injected with PRBC to evaluate primary humoral immune response during the growing phase, 23 of those animals (randomly selected; 1 to 2 head per pen) were injected with OVA to stimulate a primary immune response during the finishing phase. A solution containing OVA (106 mg; Sigma-Aldrich), Freund's Incomplete Adjuvant (60 mL; Sigma-Aldrich), and PBS (60 mL) was injected into the neck of each animal both s.c. (2 mL) and i.m. (1 mL). A total of 4000 µg OVA was administered to each animal. Blood was collected via jugular venipuncture into non-heparinized vacutainer tubes immediately prior to the injection (d 0) and at 7, 14, and 21 d post-injection from the same animals. Samples were handled as described for the PRBC sampling for primary humoral immune response. After storage, samples were thawed, heated for 30 min at 56°C, and analyzed for antibody titers specific to OVA using an ELISA procedure (Engvall and Perlmann, 1972). Data were expressed as a total concentration of antibody titers to OVA.

Interferon gamma concentrations were determined using blood collected on d 0 and 56 into heparinized vacutainer tubes. After collection, samples were centrifuged at 2000 × g for 15 min, and plasma was harvested and stored at -80°C. After storage, interferon gamma concentrations were determined using a commercially available enzyme amplified sensitivity immunoassay (Bovine IFN-d EASIA; Biosource Europe S.A., Nivelles, Belgium).

Red blood cell (RBC) superoxide dismutase (SOD) enzyme activity was determined from samples collected on d 0 and 28 of the growing phase. Heparinized vacutainer tubes were used to collect blood, which was centrifuged at 2000 × g for 15 min, and 1 mL of RBC was collected from each sample and lysed in 4 mL of cold dH^sub 2^O. Eysed RBC samples were stored at -80°C and later analyzed for SOD activity using a spectrophotometric assay (Bioxytech SOD-525®; Oxis Health Products, Inc., Portland, OR) read at 525 nm by a Spectronic Genesys 5® spectrophotometer (Thermo Electron Corp., Woburn, MA). Hemoglobin concentrations were determined using a hemoglobin reagent set (Pointe Scientific, Lincoln Park, MI) so that SOD activity data could be expressed per milligram of hemoglobin.

Throughout the entire feeding period, cattle were visually monitored daily to detect signs of morbidity. Cattle observed displaying clinical symptoms of morbidity were transported to a working facility, and a rectal temperature was collected. If an animal's rectal temperature was >39.4°C, it was considered morbid and treated accordingly. Based on visual diagnosis by trained personnel, morbid cattle were classified as morbid because of a respiratory disease or other reasons. It was also noted if an animal was being treated for the first time or retreated for signs of morbidity.

Harvest and Carcass Data. At the conclusion of the feeding period, all cattle were transported 55 km and harvested at a commercial abattoir. In yr 1, all cattle were harvested after receiving the finishing diet for 140 d. In yr 2, equal numbers of steers and heifers per treatment (the heaviest from each treatment) were harvested after receiving the finishing diet for 145 d (70 animals); the remaining 60 animals were harvested after receiving the finishing diet for 181 d. After harvest, carcasses were chilled for approximately 28 h (yr 1) and 36 h (yr 2) before complete carcass data were collected prior to carcass fabrication. Carcass data included hot carcass weight; dressing percentage; longissimus area; 12th rib fat; percentage of kidney, pelvic, and heart fat; calculated yield grade; bone maturity; lean maturity; and marbling score. For analysis of percentage of lipid and fatty acid composition, in both years a longissimus sample (approximately 100 g) encompassing the entire surface of the ribeye was collected from the face of the loin section from the same side of each chilled carcass (at the time of carcass data collection). Samples were placed into plastic bags, stored on ice, transported to the laboratory, and stored at -80°C. Immediately after thawing at room temperature, subcutaneous fat was trimmed from each sample prior to grinding the trimmed longissimus sample in a small food processor (Cuisinart, Stamford, CT). After grinding, 1 g of the ground homogenate was weighed onto oven-dried filter paper, folded, oven-dried for 24 h at 100°C, and reweighed to calculate percentage DM. Samples were then placed into a Soxhlet apparatus (Pyrex, Corning, Inc., Corning, NY), and lipid was extracted via ether distillation (AOAC, 1975). Samples were then removed from the apparatus, air-dried under a fume hood for approximately 12 h, oven-dried at 100°C for 24 h, and weighed to determine lipid content percentage. To analyze for fatty acid composition, duplicate 1-g subsamples of longissimus muscle (ground homogenate) were used for lipid extraction (Engle et al., 2000a). Methyl ester derivatives of the fatty acids extracted from longissimus muscle were prepared in duplicate using a combination of NaOCH^sub 3^ followed by HCl/ methanol (Kramer et al., 1997). After methylation, the methyl ester solution was dried under N^sub 2^ gas and reconstituted with 0.5 mL hexane. Fatty acid composition of longissimus muscle was determined via a gas chromatograph (GC; 6890 Series; Agilent Technologies, Wilmington, DE), as described by Engle et al. (2000a). The GC was equipped with a 6B90 series injector (Agilent Technologies) and flame ionization detector and was fitted with a fused silica capillary column (100 m × 0.25 mm i.d.; SP-2560; Supelco Inc., Bellefonte, PA). Samples were injected using the split mode with He as the carrier gas and a split ratio of 100:1 at 180°C. Data collected and reported were the normalized area percentages of fatty acids.

Statistical Analyses. Performance (BW, ADG, DMI, and G:F), immune (interferon gamma, SOD, and humoral and CMI responses), and mineral status data (liver and plasma mineral concentrations) were all assessed for the growing and finishing phases separately using a restricted maximum likelihood-based, mixed effects model repeat measures analysis (PROC MIXED®; SAS Inst. Inc., Cary, NC). Pen within year × treatment was included as a random variable, and the autoregressive covariate design was used when variation was homogeneous and time intervals were equal. Autoregressive with heterogeneous variation was used when variation was heterogeneous, and a spatial power covariance structure was used when time intervals were not equal. Initial models for performance, health status, immune response, and mineral status included the fixed effects of treatment, sex, time, year, and all possible interactions. Observational health data, including morbidity rate, were assessed using ?^sup 2^ analysis (PROC FREQ®; SAS Inst. Inc.). Data for carcass characteristics, percentage of lipid, and fatty acid content were analyzed using least squares analysis of variance models (PROC MIXED®; SAS Inst. Inc.). Initial models for carcass characteristics, percentage of lipid, and fatty acid content included the fixed effects of treatment, sex, year, harvest date, and all possible interactions. For all analyses, pen (replicate) was used as the experimental unit. When an interaction was not significant, it was removed, and the model was reduced. If the year × treatment interaction was not significant, data were pooled across years. Differences among means were determined using preplanned single degree of freedom contrasts; comparisons made were 1) control vs trace mineral supplemented cattle and 2) ORG vs ING.

Results and Discussion

In yr 2, one steer died from bloat at the beginning of the growing phase, shortly after arrival to the feedlot. All data collected from this steer were removed prior to any statistical analyses. Results of the cow-calf (pre-weaning) phase of this experiment have been reported elsewhere (Ahola et al., 2004).

Growing Phase Performance. During the initial 56-d growing phase, no year × treatment interactions were present (P>0.10) for initial or final BW, ADG, or DMI (Table 4). Initial BW was similar (P>0.10) across treatments. Calf weaning BW, collected at the ranch (ECRC), were affected by both trace mineral supplementation and source (Ahola et al., 2004); however, at the time of weaning on the ranch (prior to shipment to the feedlot), equal numbers of calves across treatments were culled. Culling decisions were based predominantly on poor pre-weaning performance (i.e. light weaning BW, generally unthrifty, or poor-performing calves). Culled calves were never incorporated into the feedlot phase of the experiment. At the end of the growing phase (d 56), no BW differences were observed between supplemented and control cattle (P=0.72) or between ORG- and ING-supplemented cattle (P=0.79). Similarly, neither ADG nor DMI were affected by either trace mineral supplementation (P=0.60 and P=0.75, respectively) or source (P= 0.86 and P=0.25, respectively) throughout the growing phase. There was a tendency (P=0.10) for a year × treatment interaction for G:F; therefore, G:F are reported separately for each year. In yr 1, no difference (P= 0.92) was observed for G:F between supplemented and non-supplemented control cattle, although ORG-supplemented cattle had a greater (P<0.03) G:F than did ING-supplemented cattle. In yr 2, no differences were observed for G:F between control and supplemented cattle (P=0.70) or between ORG and ING cattle (P=0.47).

The combined effects of lifetime Cu, Zn, and Mn supplementation and source on growing phase performance have not been well addressed in previous beef cattle trace mineral research. The majority of trace mineral experiments in feedlots have focused on the supplementation of a single trace mineral (e.g., Cu or Zn) only during the growing and/or finishing phases. The current experiment evaluated the combined effects of Cu, Zn, and Mn supplementation and source during the pre-natal, neonatal (pre-weaning), and post-weaning (feedlot) periods on ultimate growing and finishing phase performance. Therefore, it is not possible to determine the individual effects of Cu, Zn, or Mn.

George et al. (1997) evaluated the combined effects of Cu, Zn, Mn, and Co supplementation and source on weaned heifer calf performance during a 42-d growing phase using trace mineral treatments of 1) inorganic at NRC (1984) recommended concentrations, 2) organic at NRC (1984), and 3) organic at three times NRC (1984) recommendations for the first 14 d and then at NRC (1984) recommended concentrations for the remainder of the growing phase. Those researchers reported that DMI, ADG, and G:F were not affected by either supplementation or source; however, heifers receiving the organic form at three times the NRC (1984) recommendation had greater ADG and G:F compared with those receiving either inorganic or organic minerals at NRC (1984) recommended concentrations.

Although not directly comparable with our data, the effects of either Cu or Zn supplementation and/or source on growing phase performance have been evaluated extensively. The majority of researchers have concluded that performance during the growing phase is not affected by Cu supplementation (Ward et al., 1993; Engle and Spears, 2000b, 2001; Engle et al., 2000c; Lee et al., 2002) or source (Engle and Spears, 2000b; Lee et al., 2002). Relative to the greater G:F observed in ORG- vs ING-supplemented cattle in the current experiment, Ward et al. (1993) reported that feed efficiency was greater during the first 21 d of a 98-d growing phase in steers receiving supplemental Cu as CuSO^sub 4^ than in steers receiving Cu-lysine. Effects of Zn supplementation and source on growing phase performance have been variable. Results have included greater ADG in steers (Spears and Kegley, 2002), greater G:F and ADG in lambs and heifers (Spears, 1989), and greater ADG and DMI in steers (Rust and Schlegel, 1993) receiving supplemental Zn compared with non-supplemented controls. However, no effect of Zn source on growing phase performance has also been reported (Spears, 1989; Engle et al., 1997; Spears and Kegley, 2002).

No benefit of lifetime supplementation of Cu, Zn, and Mn was observed on growing phase performance in the current experiment, although this has not been well addressed in the literature. Because cattle receiving the control treatment in the current experiment were able to maintain liver mineral concentrations above deficient levels (Tables 6 and 7), it is likely that any potential benefit of trace mineral supplementation was not detectable, as adequate amounts of the trace minerals were most likely available from the basal diet throughout the growing period. Trace mineral concentrations of the pre-weaning basal diet have been reported (Ahola et al., 2004).

Finishing Phase Performance. No year × treatment interactions were detected (P>0.10) for initial or final BW, ADG, or DMI, so data for both years were combined (Table 5). Neither initial nor final BW differed between control and supplemented cattle (P=0.72) or between ORG and ING cattle (P=0.78). Similarly, no effect of trace mineral supplementation or source was present for ADG (P=0.75 and P=0.84, respectively). There was a tendency (P<0.10) for greater DMI in supplemented vs non-supplemented controls during the finishing phase, but no effect (P=0.28) of trace mineral source on DMI was observed. A year × treatment interaction was present (P<0.04) for finishing phase G:F. In yr 1, non-supplemented control cattle had a greater (P<0.01) G:F than did supplemented cattle; ORG cattle had a greater (P<0.04) G:F than ING cattle. However in yr 2, no differences were observed between control and supplemented cattle (P=0.87) or between ORG and ING cattle (P= 0.69).

Reported results on the impact of simultaneous Cu, Zn, and Mn supplementation during the finishing phase on beef cattle performance are limited. However, the effect of concentration and source of four trace minerals (Cu, Zn, Mn, and Co) on finishing phase performance has been evaluated previously. Steers received Cu, Zn, Mn, and Co in an organic form at NRC (1996) recommended concentrations or at 1.5 times NRC (1996) recommendations or received Cu, Zn, Mn, and Co in an inorganic form at 1.5 or 3.0 times NRC (1996) recommendations throughout a 198- or 230-d finishing period (Rhoads et al., 2003). In contrast to the current experiment, the researchers reported that DMI was greater in steers receiving the organic form at NRC (1996) recommendations vs the other three treatments.

Relative to the effect of Cu supplementation on finishing phase performance, Cu can have a positive (Ward and Spears, 1997; Engle et al, 2000c), negative (Engle and Spears, 2000b), or no effect (Engle and Spears, 2000a; Engle and Spears, 2001; Engle et al., 2000b) on performance compared with non-supplemented controls. It has also been reported that Cu source had no effect on finishing phase performance (Engle and Spears, 2000b), but led to greater final BW and a tendency for greater ADG in steers receiving organic Cu vs inorganic Cu (Lee et al., 2002). Zinc supplementation in finishing cattle led to a tendency (P<0.10) for greater ADG (Rust and Schlegel, 1993), and no effect on ADG, DMI, or G:F (Spears and Kegley, 2002) vs non-supplemented controls; DMI decreased as Zn concentration increased (Malcolm-Callis et al., 2000). Relative to Zn source, there was tendency for ADG (P=0.10) and G:F (P=0.07) to be greater in cattle supplemented with organic Zn than in cattle supplemented with inorganic Zn (Spears and Kegley, 2002); in contrast ADG was greater in cattle receiving ZnSO^sub 4^ than in cattle receiving Zn-methionine (Nunnery et al., 1996). Several researchers have also reported that Zn source had no impact on performance (Greene et al., 1988; Rust and Schlegel, 1993; Malcolm-Callis et al., 2000).

Reaching an objective conclusion about the impact of lifetime Cu, Zn, and Mn supplementation (alone or in combination) on finishing phase performance is difficult. The effect of individual trace mineral supplementation on finishing phase performance has been extremely variable, but appears to be limited. In the current experiment, the tendency for trace mineral supplementation to improve DMI and the inconsistent effects of trace mineral supplementation and source on G:F (present in yr 1, but not in yr 2) may be false or may reflect subtle differences in feed intake and/or daily gain.

Plasma Mineral Status. Plasma Cu and Zn concentrations, determined from samples collected at four points during the growing and finishing phases of yr 1, are listed in Table 6. Neither trace mineral supplementation nor source affected (P>0.10) plasma Cu concentrations at any point during the feeding period in yr 1. Plasma Zn concentration tended (P<0.08) to be greater in supplemented cattle only at the end of the finishing phase; trace mineral source did not impact (P>0.10) plasma Zn concentrations at any point during the feeding period. Utilizing plasma Zn concentration as a method of classifying Zn status may be inaccurate if cattle are not truly Zn deficient (Hambidge et al., 1986; Underwood and Suttle, 1999).

All treatments (including the nonsupplemented controls) were able to maintain plasma concentrations of both Cu and Zn above levels considered deficient (0.6 mg Cu/L and 0.4 mg Zn/L; Puis, 1994). Therefore, throughout the duration of this experiment (both cow-calf and feedlot phases), it appears that cattle receiving the control treatment were able to consume adequate amounts of Cu and Zn via the basal diet alone.

Liver Mineral Status. No year × treatment interactions were present (P>0.10) for liver concentrations of Cu, Zn, or Mn (Table 7). At every time point, supplemented cattle had greater (P<0.01) liver Cu concentrations than non-supplemented controls. However, liver Cu concentrations were not different (P>0.10) between ORG and ING cattle at any time during the experiment. Liver Zn concentration was not affected by either Zn supplementation (P>0.10) or source (P>0.10) during the growing or finishing phases. However, liver Zn concentrations did decline numerically during the feeding period from the initial liver biopsy collection (on d 0 of growing phase) to the final liver biopsy collection (on d 112 and 140 of the finishing phase, yr 1 and yr 2, respectively). Liver Mn concentration was affected (P<0.02) by trace mineral supplementation, but only at the end of the feeding period when supplemented cattle had greater (P<0.02) liver Mn concentration than non-supplemented controls. There was no effect (P>0.10) of trace mineral source on liver Mn concentration at any point during the feeding period.

Neither liver Cu nor Mn concentration was affected by trace mineral source when Cu, Zn, Mn, and Co were supplemented to cattle during the receiving and finishing phases (Rhoads et al., 2003). However, those researchers reported that liver Zn concentration was greater in steers receiving the organic form of Zn at 1.5 times NRC (1996) recommendations vs steers receiving the inorganic form of Zn at the same concentration.

Although differences in average liver Cu and Mn concentrations were observed in the current experiment, all treatments (including the non-supplemented controls) were able to maintain liver concentrations of Cu, Zn, and Mn at levels considered to be adequate (25 mg Cu/kg DM, 25 mg Zn/kg DM, and 2.5 mg Mn/kg DM; Puls, 1994). Therefore, throughout the duration of this experiment (during both the cow-calf and feedlot phases), cattle receiving the non-supplemented control treatment were apparently able to consume adequate amounts of Cu, Zn, and Mn via the basal diet.

Immunity and Health. The effect of lifetime trace mineral supplementation and source on health status during the growing and finishing phases was evaluated using several methods, including SOD activity, interferon gamma concentration, immune response, and morbidity rate. There were no year × treatment interactions (P>0.10) for SOD activity or interferon gamma concentration. Activity of the SOD enzyme, a Cu/Zn-dependent metalloenzyme responsible for catalyzing the conversion of a superoxide anion into oxygen and hydrogen peroxide, did not differ between supplemented and control cattle (P= 0.85) or between ORG- and ING-supplemented cattle (P=0.84; data not shown). In addition, neither trace mineral supplementation nor source (P>0.10) affected interferon gamma concentration (data not shown).

Limited literature has evaluated the combined effects of Cu, Zn, and Mn on SOD activity; however, experiments have evaluated the effect of Cu alone on the enzyme. Supplementation of Cu did not affect SOD activity in feedlot cattle (Dorton et al., 2003; Ward et al., 1993); however, greater SOD activity has been reported in Cu-supplemented cattle on d 0, 28, and 56 of the growing phase but not after d 56 (Ward and Spears, 1997). Because a reduction in SOD activity typically only occurs during extended Cu deficiency (Paynter, 1987), the lack of an effect of trace mineral supplementation or source on SOD activity observed in the present experiment is not unusual, as cattle were above deficiency levels of Cu throughout the experiment (Puis, 1994).

There was no year × treatment interaction (P>0.10) for CMI to PHA. An in vivo CMI response to PHA was observed in all treatments when evaluated at 4 h after injection with PHA; however, during the 48-h period following stimulation of a CMI response, the skin-fold thickness in non-supplemented control cattle did not differ (P=0.18) from that of supplemented cattle, nor did ORG-supplemented cattle differ from ING-supplemented cattle (P=0.35; data not shown).

When Cu, Zn, Mn, and Co were combined and supplemented to stressed weaned heifer calves, no effect of concentration or source of trace minerals on CMI response at 12 or 24 h after PHA injection on d 7 after arrival to the feedlot was reported (George et al., 1997). However, those researchers did report a lesser CMI response at 48 h after PHA injection in calves supplemented with the organic form at three times the NRC (1984) recommended concentrations compared with calves receiving the inorganic form at NRC (1984) recommended concentrations. Those researchers also reported that 21 d after arrival at the feedlot, greater and longer CMI responses to PHA at 12, 24, and 48 h were observed in calves supplemented with organic trace minerals at three times the NRC (1984) recommended concentrations vs calves receiving the trace minerals in an organic or inorganic form at NRC (1984) recommended concentrations. Of the calves receiving Cu, Zn, Mn, and Co at NRC (1984) recommended concentrations, calves receiving the organic form had a greater CMI response to PHA than the inorganic form, but only at 48 h post-PHA injection (George et al., 1997).

There were no year × treatment interactions (P>0.10) for primary humoral immune response to PRBC or OVA. In response to the first PRBC injection during the growing phase, there was no effect of trace mineral supplementation on IgG (P=0.87), IgM (P=0.25), or total Ig (P=0.79) concentrations, and no effect of trace mineral source on IgG (P=0.77), IgM (P=0.42), or total Ig (P=0.18) concentrations (data not shown). Similarly, concentrations of primary antibody titers raised to OVA during the finishing phase were not impacted by either trace mineral supplementation (P=0.16) or source (P=0.55).

Although limited research has evaluated the combined effect of Cu, Zn, and Mn on humoral immune response, previous researchers have reported that primary immune response to PRBC was not affected by Cu supplementation in 70-d-old calves (Ward et al., 1997) and was less in non-stressed calves receiving supplemental Cu and greater in stressed calves receiving supplemental Cu vs non-supplemented controls (Ward and Spears, 1999). Growing calves receiving inorganic Cu had greater IgG and total Ig concentrations than calves supplemented with organic Cu (Dorton et al., 2003). Copper-supplemented cattle did not respond differently than non-supplemented controls to an injection of OVA (Ward et al., 1993); however, a greater immune response to OVA has been reported in Cu-supplemented cattle vs non-supplemented controls (Dorton et al., 2003; Ward and Spears, 1999) and in calves receiving supplemental Cu as organic vs inorganic on d 14 and 21 after injection with OVA (Dorton et al., 2003).

There was no year × treatment interaction (P>0.10) for secondary humoral immune response to PRBC. In the evaluation of secondary immune response to PRBC during the finishing phase, concentrations of IgG, IgM, and total Ig were not impacted by trace mineral supplementation (P= 0.58, P=0.42, and P=0.14, respectively; data not shown). Similarly, concentrations of IgG and IgM were not affected by trace mineral source (P=0.62 and P=0.22, respectively); however, there was a tendency (P<0.07) for greater total Ig concentration in ORG-supplemented cattle compared with ING-supplemented cattle (data not shown).

There were no year × treatment interactions (P>0.10) for any of the morbidity rate variables evaluated, and trace mineral supplementation (P>0.10) did not influence the number of cattle treated or re-treated for symptoms of a respiratory disease or treated for ailments other than respiratory disease during the feeding period. The combination of a lesser morbidity rate (only 14.8% of cattle were treated for symptoms of a respiratory disease), and the fact that morbidity data are binomial, might have prevented the observation of differences across treatments for morbidity rate.

Grotelueschen et al. (2001) evaluated lifetime supplementation of trace minerals (Cu, Zn, Mn, and Co) on calf morbidity with 641 cow-calf pairs. Those researchers reported that fewer organic and non-supplemented calves were treated during the first 28 d compared with the inorganic treatment, fewer organic calves were treated throughout the feeding period compared with inorganic calves, and organic calves were re-treated less than inorganic and non-supplemented control calves during the first 28-d period and throughout the entire feeding period. When Zn source was evaluated, twice as many steers were treated in the basal and lesser organic Zn treatments compared with the greater organic and inorganic Zn treatments (Galyean et al., 1995).

Carcass Characteristics. A summarization of carcass characteristics for both years is presented in Table 8. No carcass variables differed between non-supplemented control and supplemented cattle (P>0.10) or between ORG and ING cattle (P>0.10).

No year × treatment interaction was present (P=0.63) for longissimus lipid content (data not shown). Longissimus lipid content was similar between control and supplemented cattle (P=0.69) and was similar between ORG and ING cattle (P=0.15). This is consistent with the absence of a trace mineral supplementation or source effect on marbling score (Table 8).

Fatty Acid Profile. The results of fatty acid analysis of longissimus samples are listed in Table 9. No year × treatment interactions were observed for any of the fatty acids or combinations analyzed, so data were pooled across years. Neither supplementation nor source of trace minerals impacted (P>0.10) fatty acid composition of longissimus samples. At concentrations 1.5 times NRC (1996) recommendations, cattle receiving inorganic trace minerals had greater BW percentages of C18:1 cis and C20:4 than cattle receiving organic trace minerals; however, the BW percentage of C18:2 cis-9 trans-11 was greater in cattle receiving the organic form than in cattle receiving the inorganic form (Rhoads et al., 2003).

It is unclear why neither supplementation nor source affected fatty acid profile, unlike previous research. The absence of any fatty acid differences might have been due to breed differences between the current experiment and previous research, differences in trace mineral concentrations and/or sources supplemented to cattle, and/or differences in duration of supplementation.

Implications

The combined effects of lifetime (prenatal through harvest) Cu, Zn, and Mn supplementation and source on feedlot performance, mineral status, health, immune response, and carcass traits have not been well addressed in the literature. Experiments that have evaluated individual trace minerals such as Cu or Zn indicate that both supplementation and source can impact mineral status, performance, health, and carcass characteristics during the feeding phase. However, when trace minerals were supplemented to cattle from the late fetal stage through harvest in the current experiment, only liver Cu and Mn concentrations and the G:F (in yr 1 only) were affected. It appears that lifetime supplementation and source of Cu, Zn, and Mn has limited impact on the performance, health, and carcass characteristics of feedlot cattle when trace mineral stores can be maintained above levels considered to be deficient. As long as there are no deficiencies, there appears to be a range of acceptable supplementation concentrations and types.

© 2005 American Registry of Professional Animal Scientists Provided by ProQuest LLC. All Rights Reserved.