To prepare two experimental TMRs, a commercial concentrate mix (Nonghyup Feed, Co. Ltd., Miryang, Korea), alfalfa, corn, and rice grains were used as the main ingredients. The TMR containing 20% flaked corn was used as the control diet (Corn TMR) and the TMR containing 20% rice grains was used as the treatment diet (Rice TMR) for the in vitro and in vivo experiments. Both diets were formulated to meet the adequate metabolizable energy and metabolizable protein for growing beef cattle, with an average daily gain (ADG) of 1,000 g/d [8]. The formulation and chemical composition of the experimental diets are shown in Table 1. Before conducting the chemical analysis, all feed ingredients and experimental diets were dried at 60°C for 96 h and ground using a cyclone mill (Foss Tecator Cyclotec 1093, Foss, Hillerød, Denmark) fitted with a 1 mm screen. The dry matter (DM, #934.01), CP (#976.05), EE (#920.39), acid detergent fiber (#973.18), and ash (#942.05) were analyzed using AOAC international methods [9]. The crude protein was calculated by multiplying the nitrogen contents by 6.25. The total nitrogen was measured using a nitrogen combustion analyzer (Leco FP-528 Leco, St. Joseph, MI, USA) according to the Kjeldahl method. The neutral detergent fiber (aNDF) and lignin were analyzed by utilizing the methods described by Van Soest et al [10] to determine fiber contents. Heat-stable amylase (α-amylase) was used for aNDF estimation and expressed inclusive of residual ash. The TDN and net energy for maintenance of the experimental diets were estimated based on the equations in the NRC [8,11].

In vitro fermentation was carried out using rumen fluid collected from two cannulated Holsteins (body weight [BW]: 450±30 kg) before morning feeding at the Center for Agriculture Research, Pusan National University, Korea. Animals were fed a diet consisting of 600 g/kg of timothy hay and 400 g/kg of a commercial concentrate mix. The respective rumen fluids collected before the morning feeding time mixed together, transferred into a thermos bottle, and immediately transported to the laboratory. The rumen contents were filtered through 8 layers of cheesecloth and mixed with 2 volumes of in vitro rumen buffer solution, which was done in accordance with a modified method described by Goering and Van Soest [12] under strictly anaerobic conditions. Approximately 1 g of the ground experimental substrates was put into pre-weighed filter bags (F57, Ankom Technology, Macedon, NY, USA). All bags were heat-sealed and transferred into empty 500 mL Duran bottles. Five bottles were used for each dietary treatment and 2 bags were stored per bottle. Then, 250 mL of rumen fluid and buffer mixture was transferred to each bottle, accompanied by continuous flushing with O₂-free CO₂ gas. The bottles were sealed with caps and incubated on a rotary shaker (JSSI-300T, JS Research Inc., Gongju, Korea) for 48 h at 125 rpm and 39°C. After 48 h of incubation, the gas production, DM digestibility (IVDMD), CP digestibility (IVCPD), pH, ammonia nitrogen (NH₃-N), and concentration of volatile fatty acids (VFA) were measured. The gas production was measured at 3, 6, 12, 24, and 48 h using a pressure transducer (Sun Bee Instrument Inc., Seoul, Korea), as described by Theodorou et al [13]. Gas production profiles obtained during incubation were fitted to a simple exponential model [14], the equation for which is as follows:

where T is the time (h), L is the lag time (h), e is the exponential function, k_g is the fractional rate of gas production (h⁻¹), V_max is the theoretical maximum gas production (mL) after the asymptote is reached, and VT is the gas produced at time T (mL). After incubation, the bottle caps were removed; then, the bottles were immediately placed on ice to stop the fermentation. The filter bags were removed from the bottles and rinsed with flowing water until the water ran clear. The washed bags were dried at 55°C for 48 h, and then weighed to measure the IVDMD. The IVCPD was estimated using Kjeldahl nitrogen analysis. The aNDF content was assessed for the weighed bags using a modified version of the micro-NDF method proposed by Pell and Schofield [15] to evaluate aNDF digestibility (IVNDFD). The pH of the culture fluid was measured using a pH meter (FP20, Mettler Toledo, Columbus, OH, USA). The culture fluid was centrifuged at 15,000×g for 10 min at 4°C, and then stored at −20°C until the analysis of the VFA and NH₃-N concentrations.

For the VFA analysis, 200 μL of the supernatant was diluted with 800 μL of ethanol after centrifugation at 20,000 rcf for 15 min. The VFA concentration was measured using a gas chromatograph (Agilent 7890A, Agilent Technology, Santa Clara, CA, USA) equipped with a flame ionization detector and capillary column (Nukol Fused silica capillary column, 30 m×250 μm×0.25 μm, Supelco Inc., Bellefonte, PA, USA). The temperatures of the oven, injector, and detector were set at 90°C, 90 to 200°C, and 230°C, respectively. Nitrogen was used as the carrier gas, at a flow rate of 30 mL/min. The ammonia concentration was analyzed following the methods describe by Chaney and Marbach [16] with several modifications. After 2 μL of the supernatant was mixed with 100 μL of a phenol color reagent (50 g of phenol, 0.25 g of sodium nitroferricyanide, and 1 L of distilled water) and alkali hypochlorite (25 g of sodium hydroxide, 16.8 mL of sodium hydroxide, and 1 L of distilled water), the mixture was incubated in a 37°C water bath for 15 min. The ammonia concentration was determined by measuring the absorbance at 630 nm using a microplate reader (iMARK, Biorad, Hercules, CA, USA).

A total of 24 Hanwoo steers (9 months old; BW 259±13 kg) were randomly allocated to the following two dietary treatment groups: i) Corn TMR and ii) Rice TMR. Each treatment had 3 replicated pens with 4 steers per pen, which was equipped with a feeder and automated drinker. Therefore, each pen was considered as an experimental unit. All diets were prepared weekly and provided to the steers once per day at the feeding time (0800) ad libitum (targeting 10% refusal). The experiment lasted for 9 weeks. Feeds offered and refused in each pen were measured daily and calculated as the DM for daily DM intake (DMI). On the first and last days of the experiment, all steers were weighed to measure the ADG and feed conversion ratio (FCR).

The data from the in vitro and in vivo experiments were analyzed using the general linear model procedure of the SAS package 9.4 (SAS Institute Inc., Carey, NC, USA). The general linear model for the statistical analysis is as follows:

where y_ij is the jth observation in the ith treatment, μ is the overall mean, τ_i is the fixed effect of the ith treatment, and e_ij is the unexplained random effect on the jth observation for the ith treatment. Differences between treatments were analyzed using the Tukey test, with statistical significance declared at p<0.05 and a trend discussed at 0.05≤p<0.1.