Despite increasingly detailed knowledge of the biochemical processes involved in the determination of meat quality traits, robust models, using biochemical characteristics of the muscle to predict future meat quality, lack. The neglecting of various aspects of the model paradigm may explain this. First, preslaughter stress has a major impact on meat quality and varies according to slaughter context and individuals. Yet, it is rarely taken into account in meat quality models. Second, phenotypic similarity does not imply similarity in the underlying biological causes, and several models may be needed to explain a given phenotype. Finally, the implications of the complexity of biological systems are discussed: a homeostatic equilibrium can be reached in countless ways, involving thousands of interacting processes and molecules at different levels of the organism, changing over time and differing between animals. Consequently, even a robust model may explain a significant part, but not all of the variability between individuals.

Tenderness and intramuscular fat content are key attributes for beef sensory qualities. Recently some proteomic analysis revealed several proteins which are considered as good biomarkers of these quality traits. This study focuses on the analysis of 20 of these proteins representative of several biological functions: muscle structure and ultrastructure, muscle energetic metabolism, cellular stress and apoptosis. The relative abundance of the proteins was measured by Reverse Phase Protein Array (RPPA) in five muscles known to have different tenderness and intramuscular lipid contents: Longissimus thoracis (LT), Semimembranosus (SM), Rectus abdominis (RA), Triceps brachii (TB) and Semitendinosus (ST). The main results showed a muscle type effect on 16 among the 20 analyzed proteins. They revealed differences in protein abundance depending on the contractile and metabolic properties of the muscles. The RA muscle was the most different by 11 proteins differentially abundant comparatively to the four other muscles. Among these 11 proteins, six were less abundant namely enolase 3 (ENO3), phosphoglucomutase 1 (PGK1), aldolase (ALDOA), myosin heavy chain IIX (MyHC-IIX), fast myosin light chain 1 (MLC1F), triosephosphate isomerase 1 (TPI1) and five more abundant: Heat shock protein (HSP27, HSP70-1A1, αB-crystallin (CRYAB), troponin T slow (TNNT1), and aldolase dehydrogenase 1 (ALDH1A1). Four proteins: HSP40, four and a half LIM domains protein 1 (FHL1), glycogen phosphorylase B (PYGB) and malate dehydrogenase (MDH1) showed the same abundance whatever the muscle. The correlations observed between the 20 proteins in all the five muscles were used to construct a correlation network. The proteins the most connected with the others were in the following order MyHC-IIX, CRYAB, TPI1, PGK1, ALDH1A1, HSP27 and TNNT1. This knowledge is important for understanding the biological functions related to beef tenderness and intramuscular fat content.