Technical problems and opportunities in using vegetable proteins in dairy products

Physicochemical and acid gelation properties of commercial UHT-treated plant-based milk substitutes and lactose free bovine milk

Physicochemical and acid gelation properties of UHT-treated commercial soy, oat, quinoa, rice and lactose-free bovine milks were studied. The separation profiles were determined using a LUMiSizer dispersion analyser. Soy, rice and quinoa milks formed both cream and sediment layers, while oat milk sedimented but did not cream. Bovine milk was very stable to separation while all plant milks separated at varying rates; rice and oat milks being the most unstable products. Particle sizes in plant-based milk substitutes, expressed as volume mean diameters (d4.3), ranged from 0.55μm (soy) to 2.08μm (quinoa) while the average size in bovine milk was 0.52μm. Particles of plant-based milk substitutes were significantly more polydisperse compared to those of bovine milk. Upon acidification with glucono-δ-lactone (GDL), bovine, soy and quinoa milks formed structured gels with maximum storage moduli of 262, 187 and 105Pa, respectively while oat and rice milks did not gel. In addition to soy products currently on the market, quinoa may have potential in dairy-type food applications.

Rapid turbidimetric detection of milk powder adulteration with plant proteins

Development of assays to screen milk for economically motivated adulteration with foreign proteins has been stalled since 2008 due to strong international reactions to the melamine poisoning incident in China and the surveillance emphasis placed on low molecular weight nitrogen-rich adulterants. New screening assays are still needed to detect high molecular weight foreign protein adulterants and characterize this understudied potential risk. A rapid turbidimetric method was developed to screen milk powder for adulteration with insoluble plant proteins. Milk powder samples spiked with 0.03-3% by weight of soy, pea, rice, and wheat protein isolates were extracted in 96-well plates, and resuspended pellet solution absorbance was measured. Limits of detection ranged from 100 to 200 μg, or 0.1-0.2% of the sample weight, and adulterant pellets were visually apparent even at ∼0.1%. Extraction recoveries ranged from 25 to 100%. Assay sensitivity and simplicity indicate that it would be ideally suitable to rapidly screen milk samples in resource poor environments where adulteration with plant protein is suspected.