A microbiological method for assessing the nutritional value of proteins

Within groups of similar proteins-for example whale-meat meals or fish meals of different manufacturethere may be large differences in nutritive quality but only relatively small differences in amino acid composition. From chemical and microbiological tests (Bunyan & Price, 1960; Ford, 1962) it appears that the differences in quality may not be specifically related to any one amino acid but reflect corresponding differences in the availability of several amino acids. Thus in twelve whale-meat meals of sensibly uniform amino acid composition but widely differing nutritive quality, values obtained for available lysine, methionine, tryptophan, leucine and arginine were closely correlated with each other and also with rat assay values representing overall protein quality (Ford, 1962). Any one set of values would serve to grade these meals in order of nutritive qualitywhether available lysine measured chemically by the dinitrofluorobenzene procedure of Carpenter (1960), or available methionine, tryptophan, leucine or arginine measured microbiologically as described by Ford (1962). We cannot infer from this that these chemical and microbiological values represent quantitatively the ' biological availability' of the amino acids. They must first be checked by comparison with the results of direct growth tests with animals. This will be a laborious undertaking but some beginnings have been made. Thus, Carpenter, March, Milner & Campbell (1963) have developed a chick growth assay for lysine, and found fair agreement between chemical and biological estimates of the available lysine in five fish meals. Miller, Carpenter & Morgan (1963) compared the results of growth assays with chicks for methionine in several meat meals and fish meals with values for available methionine obtained in microbiological tests. They found that the Streptococcus zymogenes procedure of Ford (1962) gave uniformly lower results than the chick assays. On increasing the concentration of papain used for the preliminary digestion the microbiological assay values increased, to agree closely with the chick results. The coefficient of correlation was then 0.97 (P < 0.001 ; ten samples). It is an open question whether these higher values on which the chick and microbiological tests agreed represented all the methionine that was potentially available. The microbiological results were maximum values in the sense that they were not further increased by treating the test samples with even greater amounts of papain. They might con-