Cannabis sativa L. as a Feed Source in Backyard Rabbit Production

Objective 1

Hemp processing facilities within the United States of America that used cold pressing as the method of oil extraction were contacted to obtain hempseed meal (HSM) samples. Processors were asked to provide distinct samples from as many batches as possible. All hemp seed was collected from C. sativa plants containing less than 0.3% THC per the 2018 Farm Bill (United States Congress, 2018).

Upon receipt, HSM samples were dried in a forced-air oven at 55^oC until weight loss ceased. Samples were then ground to pass through a 2-mm screen in a Wiley mill (Arthur H. Thomas Co., Philadelphia, PA, USA); a subsample was ground to pass thorugh a 1-mm screen.

The experimental design for determining source of variaiton was a completely randomized design. A total of four processing facilities and 15 HSM batches were represented in the experiment. Four laboratory replicates of each batch × source combination were assayed.

Carbon and nitrogen concentrations were determined by combustion using the Dumas total combustion method in a Leco Cornerstone CN 828 (Elementar Americas, Mt. Laurel, NJ, USA; Method 990.09; AOAC, 2000). Nitrogen concentration was used to calculate CP content by multiplying by 6.25 (AOAC, 1984). Neutral detergent fiber (NDF), acid detergent fiber (ADF), and acid detergent lignin (ADL) were measured with the ANKOM²⁰⁰ Fiber Analyzer (ANKOM Technology Corporation, Fairport, NY, USA) using modified methods (Vogel et al., 1999) originally described by Van Soest and Robertson (1980). Acid detergent lignin was measured by the sulfuric acid (Method 973.18; AOAC, 2000). An ANKOM Daisy^II incubator (ANKOM Technology Corporation, Fairport, NY, USA) was used to determine in vitro true digestiblity (IVTD) using a modification (Vogel et al., 1999) of the in sacco disappearance method originally described by Lowrey (1969). Rumen fluid was collected from a cannulated steer at the Texas A&M AgriLife Research and Extension Center in Stephenville TX. The steer was offered ad libitum access to ‘Coastal’ bermudagrass (Cynodon dactylon [L.] Pers.) hay with a minumum 120 g CP kg^-1 DM.

The experimental design for each of the in vitro dry matter digestibility (IVDMD) experiments in which the effect of HSM inclusion on IVDMD was evaluated was a generalized complete block design. Block was designated as a single rumen fluid collection, and each experiment included two block. Each dietary treatment was replicated three times within each block, with the flask serving as the experimental unit.

The HSM used in rations for the in vitro experiments was a composite of the 15 batches collected from four oil processing factories (336 g CP kg^-1 DM). Steam flaked corn (Zea mays L.), bermudagrass hay, alfalfa (Medicago sativa L.), and soybean (Glycine max L.) meal were purchased at a local feed store. All ingredients were ground to pass through a 2-mm screen in a Wiley mill (Arthur H. Thomas Co., Philadelphia, PA, USA). In the first experiment, HSM replaced the concentrate (steam-flaked corn) in a 60:40 forage-to-concentrate ratio ration at 0, 250, 500, 750, and 1000 g kg^-1. The forage in this ration was bermudgrass hay. In the second experiment, isonitrogenous rations were formulated in which HSM replaced soybean meal and alfalfa at 0, 250, 500, and 750 g kg^-1. Modified methods of the procedure originally described by Tilley and Terry (1963) were utilized. For each experiment, 0.5 g samples were weighed into 125 mL Erlenmeyer flasks with 40 mL of a 1:5 ratio of buffer solution B and buffer solution A (ANKOM Technology, 2017). Ten mL of CO₂-flushed rumen fluid collected from a ruminally-fistulated steer was then added to each flask, and flasks were sealed with a topper and placed in a Fisherbrand Isotemp Shaking Water Bath (Thermo Fisher Scientific, Newington, NH, USA) at 39°C and agitated at 30 RPM for 48 h. Then, 2 mL HCl and 0.5 g pepsin were added to each flask and toppers were removed. After 48 h, samples were filtered through P8 filter paper and dried for 2 h at 105°C.

Data were analyzed using SAS v. 9.4 (SAS Institute, Inc., Cary, NC). Prior to analysis, raw data were tested using the NORMAL option of PROC UNIVARIATE to ensure data normality. Normality was assumed when Shapiro-Wilk’s W met or exceeded 0.9 (Shapiro and Wilk, 1965; Royston, 1992).

For the experiment evaluating the source of variation in nutritive value, response variables were analyzed using the linear mixed models procedure (PROC MIXED) in SAS using the COVTEST option for random effects models. Random effects included source (processor), batch within source, and replicate within batch by source (replicate was understood to represent the laboratory replicate and not a statistical replicate).

For the IVDMD experiements, response variables were analyzed using the generalized linear mixed models procedure (PROC GLIMMIX) in SAS. The fixed effect was treatment, and the random effects were block and treatment by block.

For all experiments, denominator degrees of freedom were adjusted using the Kenward-Roger approximation method (Kenward and Roger, 1997). The α-level for mean differences was set at 0.05. When interactions had P < α, the interaction was discussed; otherwise, main effects were discussed. Means separations were performed based on F-protected t-tests using Tukey-Kramer’s HSD.

Objective 2

Hempseed meal samples used in Objective 1 were composited for use in Objective 2 (336 g CP kg^-1 DM). Steam flaked corn (Zea mays L.), bermudagrass hay, alfalfa (Medicago sativa L.), and soybean (Glycine max L.) meal were purchased at a local feed store. All ingredients were ground to pass through a 2-mm screen in a Wiley mill (Arthur H. Thomas Co., Philadelphia, PA, USA). Isonitrogenous rations were formulated in which HSM replaced soybean meal and alfalfa at 0, 250, 500, and 750 g kg^-1. Rations were pelleted for ease of feeding

Weaned Californian meat rabbits (n = 32) were allocated randomly to eight cages (four animals/cage) in a generalized complete block design. Cages were allocated randomly to one of four levels of protein replacement. Rabbits were fed for 35 d from weaning to harvest. Rabbits were weighed and health checked weekly. Feed samples were collected daily and composited weekly. Orts and feces were collected in the final seven days of each period to calculate measures of digestibility.

Data were analyzed using SAS v. 9.4 (SAS Institute, Inc., Cary, NC). Prior to analysis, raw data were tested using the NORMAL option of PROC UNIVARIATE to ensure data normality. Normality was assumed when Shapiro-Wilk’s W met or exceeded 0.9 (Shapiro and Wilk, 1965; Royston, 1992).

Response variables were analyzed using the generalized linear mixed models procedure (PROC GLIMMIX) in SAS. The fixed effect was treatment, and the random effects were period, treatment by period, animal within cage by period, and sex (for individual animal responses). Denominator degrees of freedom were adjusted using the Kenward-Roger approximation method (Kenward and Roger, 1997). The α-level for mean differences was set at 0.05. When interactions had P < α, the interaction was discussed; otherwise, main effects were discussed. Means separations were performed based on F-protected t-tests using Tukey-Kramer’s HSD.

Outreach Plan

The primary mode of outreach for this project was presentation of research results at the Annual Meeting of the American Society of Animal Science.