We evaluated different sample preparation methods for one‐dimensional proton nuclear magnetic resonance metabolomics analysis of beef cattle ruminal fluid and serum. After collecting and pooling samples from five bulls (all of which received the same diet and were subjected to the same management conditions), we prepared the samples with the following methods: metabolite extraction using methanol and chloroform (M + C); metabolite extraction using methanol, chloroform and acetone (M + C + A); filtering with a 3‐kDa filter; and filtering with a 2‐μm filter. We evaluated the methods based on the spectral quality, reproducibility, ease of processing, speed and relative cost. The use of 2‐μm filters did not produce good‐spectra quality for serum samples, but it was a promising method for ruminal fluid. The yields of metabolites were similar for the the solvent extractions (22 metabolites for ruminal fluid and 29 for serum samples), but lower compared with the filter methods (31 metabolites for ruminal fluid and 47 for serum samples). The filters resulted in a higher concentration of metabolites and better reproducibility. In conclusion, the 2‐μm and 3‐kDa filter methods represent a suitable way to prepare biological samples from beef cattle for nuclear magnetic resonance – based metabolomics.

Background: Bos taurus indicus cattle is known to be temperamental and to produce beef with greater variability in terms of quality compared to beef of Bos taurus taurus. Cattle adaptability and resilience are of great importance to sustain beef production worldwide. Objective: The study aimed to understand early post-mortem metabolites among muscles with different fiber types profile of calm and excitable Nellore, as well as its relationship with fragmentation of beef aged up to 28 d. Methods: Animals were evaluated based on chute score and exit velocity to calculate a temperament index, which was used to classify them as calm or excitable. At slaughter, the pH and temperature declines of Triceps brachii (TB) and Longissimus lumborum (LL) were measured, muscles were sampled, and aged up to 28 d. Metabolites were determined, and sarcomere length and myofibrillar fragmentation index (MFI) were quantified. Metabolomics data were analyzed using a multivariate approach, while other traits were investigated through ANOVA. Results: The pH decline was affected by all three fixed effects investigated (temperament × muscle × time post-mortem: p = 0.016), while temperature decline was affected by muscle × time (p < 0.001). Metabolites differed among muscles and cattle temperament, with excitable cattle showing greater taurine abundance in LL, as well as greater creatine in TB 1 h post-mortem, based on the volcano plot. Sarcomere length and MFI results revealed faster and limited tenderization in excitable cattle beef. Conclusions: Altogether, results emphasized the upregulation of mitochondrial enzymes and reduced tenderization as determinants of inferior beef quality after prolonged aging in excitable cattle.