Animal management and measurements

The Fort Keogh Livestock and Range Research Laboratory Institutional Animal Care and Use Committee approved all animal handling and experimental procedures used in the present study.

Twenty crossbred ruminally cannulated cows (577 ± 13 kg) were stratified by age (5 ± 0·4 years old) and prior lactation history (three cows did not raise a calf). Cows were assigned to one of two forms of Co supplementation: (1) 12·5 mg Co supplied as 27·2 mg of CC (n = 11 cows) (2) 12·5 mg Co supplied as 50 mg of Co glucoheptonate (CGH, n = 8 cows; Zinpro Corp., Eden Prairie, MN, USA). On the first day of treatment supplementation, one cow in the CGH treatment group was given a CC bolus, thus was moved to the CC treatment group, increasing this group to 11 cows. One cow in the CGH treatment group was found to be a true outlier, thus was removed from the study decreasing this group to eight cows. Diets met or exceed the dietary requirements for Co of 10 mg/kg (NRC 2000). Cobalt concentrations were below assay detection limits for the grass hay provided in the diet and verified by two independent commercial laboratory analyses. To alleviate weaning stress, calves were weaned 14 days prior to starting the trial, at which time cows were relocated from the native range to a common dry lot pen and fed a chopped grass hay diet, free choice (Table 1). Seven days prior to the start of the trial, cows were assigned randomly to an individual pen (6 × 10 m²) with ad libitum access to water. Water analysis revealed undetectable concentrations of Co and no other known antagonistic factors were detected. Throughout the trial all cows were fed a basal diet of chopped grass hay at a rate of 0·02% of body weight and 57 g of Fort Keogh Range Mineral top-dressed into each bunk once daily (07·00 h; Table 1). Cows were managed in individual pens for 62 days, which included an acclimation (AC), treatment (TR) and residual (RE) period. During the AC period (days 0–17), all cows were fed only the basal diet with no supplemental Co. On days 18–44 (TR period) cows received one of the two Co treatments daily, 2 h after feeding (09·00 h). Cobalt supplement was administered directly into the rumen via the rumen cannula in a gelatin capsule (Torpac, Inc. Fairfield, NJ, USA). From days 45 to 62 (RE period), cows received no Co supplementation to determine any carry-over effects from the previous TR period.

Table 1.

Chemical composition of hay and top dressed Fort Keogh Range Mineral fed throughout the 62-day experiment

DM, dry matter; CP, crude protein; NDF, neutral detergent fibre; ADF, acid detergent fibre.

* n/d: below detectable assay limits (confirmed by two independent commercial laboratories).

To estimate diet digestibility, in situ neutral detergent fibre (NDF) disappearance (ISNDFD) and in situ organic matter (OM) disappearance (ISOMD) were measured. Grass hay samples (same source as basal diet) were collected from stockpiled hay and ground to pass through a 2 mm screen (Model 4 Wiley mill, Arthur H. Thomas Co, Philadelphia, PA, USA). Dacron bags (10 × 20 cm²: 50 µm porosity: Ankom Technology Corp., Fairport, NY, USA) were filled individually with 5 g of ground grass hay and sealed. During the last 4 days of each period (AC d 14, TR d 41 and RE d 58) a 96 h in situ was initiated. At each incubation time (96, 48 and 24 h), five filled Dacron bags along with a sealed blank bag were placed in a 35 × 45 cm² polyester mesh bag (Household Essentials, Hazelwood, MO, USA) anchored with ~1 m of string and a rubber stopper. The bags were then placed into the rumen in the vicinity of the mat and liquid interface and all removed at end of the 96 h interval. The empty bag was used as a correction factor for ruminal contamination entering the bags. The empty bag residue weight was subtracted from each sample bag at the same incubation time. Upon removal from the rumen, all the bags were subjected to an initial rinse by submerging them in a bucket filled with cold tap water to arrest microbial fermentation. Five filled bags and one empty bag were not inserted into the rumen, but were subjected to the cold water rinse (Wiley et al. Reference Wiley, Petersen, Clark, Ansotegui and Lodman1991) and designated 0 h bags. All bags were transported to the lab and rinsed individually under cold tap water until the effluent was clear, after which bags were frozen (–20 °C), lyophilized and weighed. Residue remaining in the bag was analysed for dry matter (DM), OM and NDF using the methods of Van Soest et al. (Reference Van Soest, Robertson and Lewis1991). Disappearance of NDF was determined with a batch processor (ANKOM 200 fibre analyser, ANKOM Technology, Fairport, NY, USA).

On days 17, 44 and 61 of the corresponding AC, TR and RE periods, blood samples were collected via coccygeal venipuncture (Corvac, Sherwood Medical, St Louis, MO, USA) at 4 and 6 h after feeding. Blood samples were centrifuged at 1500  g for 30 min; serum was decanted and stored at –20 °C until the analysis. Cobalamin was assayed by Ronald Kincaid's laboratory (Department of Animal Sciences, Washington State University, Pullman, WA, USA) using a Dualcount Solid Phase Vitamin B₁₂/Folic Acid procedure (Siemens Healthcare Diagnostics, Los Angeles, CA, USA. Cat # KDSP1, interassay CV 4·2% and intraassay CV 3·3%) and expressed as pg/ml.

Particulate-associated carboxymethylcellulase (CMCase) activity was measured on residue from in situ bags incubated at 96, 48, 24 and 0 h for all three periods of cellulolytic activity in response to differing sources of Co supplementation. A single dacron bag/cow/incubation period was pulled randomly after lyophilization for the CMCase analysis. Approximately 0·4 g of residue was weighed into a 50 ml centrifuge tube, after which 8 ml of 10 mm sodium phosphate buffer (pH 6·8) containing 20 µg/ml of lysozyme (Sigma Aldrich, St. Louis, MO, USA) was added. Following an addition of 1 ml of carbon tetrachloride, the mixture was vortexed and incubated in a 37 °C water bath for 3 h. The tubes were then centrifuged at 29 000  g at 4 °C for 15 min and supernatant was decanted and frozen (–20 °C) prior to analysis. The CMCase assay has been previously adapted to utilize a 96-well micro plate system (Xiao et al. Reference Xiao, Storms and Tsang2004, Reference Xiao, Storms and Tsang2005). Thirty micro litres of the sample supernatant, in duplicate, was pipetted into a 96-well polymerase chain reaction (PCR) plate and an addition of 30 µl of 2% (w/v) sodium carboxymethylcellulose (Sigma Aldrich, St. Louis, MO, USA) containing 0·1 mg/ml of thirmersol was added to each well. The 96-well plate was incubated utilizing a thermocycler, programmed to heat to 50 °C for 30 min followed by cooling and holding at 20 °C. Following incubation, 60 µl of 3,5-dinitrosalicylic acid reagent (Miller et al. Reference Miller, Blum, Glennon and Burton1960) was added to each well. To develop colour, the plate was incubated again in the thermocycler programmed to heat to 95 °C for 5 min followed by cooling to 4 °C for 1 min and holding at 20 °C (King et al. Reference King, Donnelly, Bergstrom, Walker and Gibson2009). Following colour development, a 100 µl aliquot from each duplicated sample was transferred to the wells of a flat-bottom 96-well (Costar, Corning INC, Corning, NY, USA). Absorbance was measured on a plate reader (Synergy HT, Bio TEK Instruments INC, Winooski, VT, USA) at 540 nm. D-glucose was used as the standard, and CMCase activity was expressed as μm glucose released/g OM/min. Activity of CMCase was corrected by subtracting the background glucose activity from every sample (Bhatti & Firkins Reference Bhatti and Firkins1995).

Statistical analysis

Data were analysed using Proc Mixed procedures of SAS (Version 9.3, SAS Institute Inc., Cary, NC, USA). The experiment was a completely randomized design with three periods (AC, TR and RE). The restricted maximum-likelihood method was used for estimating the variance components and degree of freedom was adjusted using the Kenward–Roger option. The model included the fixed effects of treatment and period (TR and RE) and their interaction. Data were analysed by hour. The covariance structure used was variance components. The Repeated statement included period with the subject being cow. Corresponding measurements obtained in the acclimation period were used as covariates in the treatment and residual period. Significance was set at P ⩽ 0·05, while trends were considered between P = 0·05 and 0·10.