Bird husbandry and management
Lohmann Indian River chicks sourced from a commercial hatchery and feather sexed were used for sex-separate experiments. Prior to incubation, all eggs collected from broiler breeders were of a similar age. Male (4,500 chicks) and female (4,800 chicks) broilers were placed in floor pens in a closed sided house. Each experiment utilized 130 floor pens measuring 1.44 x 1.70 m and was conducted from 22 to 35 d of age. At d 22, both male and female bird numbers were adjusted to 34 (13.9 birds per m2) and 36 (14.7 birds per m2), respectively.
The closed sided wall facility used radiant heating during brooding, negative pressure ventilation for air exchange during brooding, and tunnel ventilation for cooling during the 22 to 35 d experimental periods. Each pen was equipped with one tube feeder, one nipple drinker line, and new rice hulls for bedding. The light procedure followed the primary breeder recommendation for Lohmann Indian River broilers . Feed and water were provided ad libitum. Mortality were obtained daily and their weights were recorded to calculate adjusted feed conversion by adding mortality weight to pen body weight at day 35. Bird care followed research protocol and ethics guidelines established by the JAPFA Animal Care Committee. Any mortality exceeding the primary breeder recommendation required flock oversight from a veterinarian.
Experimental design and diets
Both male and female chicks were offered a proprietary common starter feed in crumbled form from 1 to 21 d of age. A basic test diet primarily composed of corn, soybean meal, and peanut meal was formulated to create 13 treatment variations in BCAA that were administered from 22 to 35 d of age and shown in Table 1. In both male and female experiments, varying levels of L-valine, L-isoleucine, and L-leucine were used to create the 3-factor treatments of the Box-Behnken experimental design. Two fillers were used (i.e., palm olein and washed sand) so that the 13 BCAA treatments would be equal in energy and diet total volume. The 13 BCAA treatments were expressed on a fecal digestible basis relative to dietary lysine. Each treatment was replicated ten times in each experiment. All treatment diets were pelleted after mixing with a conditioning temperature range from 78 to 80 C. Composite samples of pelleted dietary treatments were obtained, ground, hydrolyzed in acid, and analyzed for amino acids .
Measurements for male and female experiments followed identical procedures. Pen body weight was obtained at 22 and 35 d and feed intake was measured from the 22 to 35 d period. Mortality was measured daily. Feed conversion (as FCR) was calculated for the 22 to 35 d period and represented total pen feed intake divided by total pen body weight gain. In each experiment at d 35, two birds per pen were randomly selected, weighed, and processed manually (260 birds total or 20 birds per treatment in each experiment). Once dressed carcasses were obtained, abdominal fat was removed and breast meat was manually excised from hot carcasses. Processing measurements consisted of carcass yield, and abdominal fat percentage, and total breast yield relative to body weight.
Pen was the experimental unit for all analyses. The Box-Behnken design has been shown to be an effective multivariate design to create dietary amino acid surface plots while requiring less experimental units (i.e., pens) than a complete factorial analyses , which was employed to assess the BCAA concurrently. The three level (-1, 0, +1) three-factor Box-Behnken design was analyzed using the PROC RSREG procedure  for linear, quadratic, and interactive effects. Response surface plots were constructed for parameters analyzed with P < 0.10.