Visual assessments of meat surface structure and meat colour to predict beef tenderness between five South African beef breeds

Abstract

Visual measurements of colour are the ideal standards for the determination of meat quality at the point of sale. Thus far, consumers rely on colour alone to make a purchase decision. The use of visual surface structures to differentiate tender from tough meat could serve as a new technology to assist consumers in predicting meat tenderness at the point of sale. According to the South African Red Meat Carcass Classification system, carcasses are classified according to animal age, fatness class, confirmation class and bruising or damage of the carcass. This system does not clearly indicate to the consumer the physical, compositional and sensory characteristics of the meat, and has limitations in classifying carcasses into very tender, tender, less tender, and slight tender. The use of visual structural measurements as a tool to assist in classifying meat into the categories of tender and tough could be a valuable technology. The objectives of this study were to determine the possibility to predict beef tenderness with experienced vision, determine the possibility of an association between colour, surface structure (morphology), and tenderness; as well as to determine genotypic differences in meat colour, morphological structure, and resulting shelf life. To achieve these aims, the study was conducted in two phases, with Phase 1 being an exploratory phase, after which the findings were implemented from the first phase into the second phase to allow for in-depth analyses. The beef breeds Brahman (Br) (Bos indicus), Nguni (N) (Sanga type), Angus (A) (British Bos Taurus), Charolais (C) (European Bos Taurus) and Bonsmara (Bo) (composite) were used, with 10 steers per genotype per phase. The animals were finished off on a feedlot diet for a period of between 90-110 days at the ARC-API feedlot and were slaughtered at the ARC-API abattoir when they reached a live weight which would produce a carcass of Class A (no permanent incisors), and fat class 2 to 3 (1-≤5 mm) (South African Beef Classification System). After exsanguination, the carcasses were halved. The right sides were electrically stimulated for 20 s (400 V peak, 5 ms pulses at 15 pulses/s) and entered the cold rooms (± 2°C) within 60 min after slaughter (ES). The left sides were placed in a room with a controlled temperature of 10°C for six hours, thereafter in cold rooms at ±2°C (NS). Temperature and pH and muscle energy samples were taken at 1, 3, 6 and 24 hrs from the m. longissimus dorsi. Steaks were sampled at the m. longissimus dorsi at 24 hours post mortem (pm). One set of steaks were aged for three and nine days pm in polystyrene plates covered with polypropylene cling wrap (PP) at 6°C in a display cabinet; while another set of steaks were aged for 14 and 20 days pm in vacuum bags at 1 4°C in a cold room for Phase 1. For Phase 2 of the study, steaks were aged for three days pm in polystyrene plates covered with polypropylene cling wrap at 6°C in a display cabinet and the other steaks were aged for 9, 14 and 20 days pm in vacuum bags at 1 4°C in a cold room. The change of packaging for nine days pm was based on the lipid peroxidation results as was measured by the TBARS assay, the nine days pm steaks had higher TBARS, and the steaks had already started to develop moulds, discolouration and bad odour. The fresh, aged steaks were analysed for visual colour, marbling, fibre separation, texture and structure integrity using a 10 member trained panel. Other analyses included tenderness measurements (e.g. Warner Bratzler shear force, sarcomere length, myofibril fragment length, connective tissue characteristics, and fibre detachment), colour measurements (Minolta CIE. L*, a* , b*, myoglobin derivatives), protein and lipid denaturation, drip loss and water holding capacity. Correlation coefficients were established between visual colour measurements and instrumental colour measurements, visual tenderness measurements, instrumental tenderness measurements; and between visual tenderness measurements and instrumental colour measurements. Results of this study revealed that the evaluation of meat colour by the panel correlated very well with instrumental colour (CIE. L*, a*, b*) and very little with myoglobin and its derivatives (oxymyoglobin, metmyoglobin and deoxymyoglobin). The visual panel was able to differentiate meat colour between the breeds. Good correlations were found between instrumental tenderness measurements (WBSF) and visual tenderness measurements, which included the fibre separation, structure integrity and marbling. Visual texture showed very low correlations with WBSF measurements. The use of visual meat colour to predict meat tenderness showed very low correlations with WBSF. The study also revealed breed differences in meat colour. Nguni breed was found to produce darker meat than Angus, Bonsmara, Brahman and Charolais. The Charolais breed produced lighter meat during Phase 1 and darker meat during Phase 2. This was due to the harsh weather experienced during Phase 2. In conclusion, visual analysis of colour can accurately be used to evaluate the quality of meat colour. Fibre separation, structure integrity and marbling could be used to visually predict meat tenderness, but only when training is provided. Visual texture did not show any potential as a reliable visual tenderness attribute, therefore cannot be used to predict meat tenderness. Visual analysis of meat colour cannot be used to predict meat tenderness.