Factors associated with microbial growth in ground beef extended with varying levels of textured soy protein

This study was conducted to examine various levels of soy extended ground beef chemically and microbiologically to determine the factor(s) responsible for accelerated microbial growth at refrigerator temperature (5°C). Three separate 11.36 kg batches of ground beef containing 25% fat were extended to 0, 10, 20 and 30% by weight with rehydrated, defatted, textured soy protein (TSP). Samples were packaged in 0.45 kg size styrofoam trays, wrapped with an oxygen semi-permeable film, placed in a laboratory refrigerator at 5°C and allowed to undergo spoilage during a 4 day storage period. Eight 50 g samples (one taken using sterile technique) were taken daily and analyzed for changes in bacterial numbers (standard plate count and coliforms), pH, titratable acidity, proximate analysis, ninhydrin reactive material (α-amino groups) and carbohydrates. Standard plate counts (SPG) and coliform counts differed (P < 0.05) with increasing levels of TSP. Increases in SPG and coliform numbers were observed as storage progressed from day 0 to day 4. At day 4, soy V 8 extended samples contained between 10⁷ and 10⁸ bacteria/g and were characterized by a sour odor. Replication differences were significant for SPG (P < 0.001) and coliform (P < 0.05) counts, respectively, pH values increased (P < 0.001) with increasing amounts of TSP. Significant differences were noted among replications for pH values and for titratable acidity. Titratable acidity increased (P < 0.01) as days of storage progressed. As soy level increased, no differences were noted for percent protein, but percent moisture and percent ash increased while percent fat decreased (P < 0.0 01). Percent ash decreased (P < 0.05) as storage continued from day 0 to day 4. Significant differences were noted among replications for all proximate analysis values except percent ash. As soy level increased, absorbance values for carbohydrates increased (P < 0.001) and ninhydrin reactive material increased (P < 0.05) as storage progressed. Differences among replications were significant for both carbohydrate and ninhydrin reactive material values.