Comparison of the Lowry and the Bradford protein assays as applied for protein estimation of membrane-containing fractions.

Although the Bradford protein estimation assay has found wide distribution, it has a number of drawbacks, which in some cases have been shown to produce erroneous results upon comparison to other, more precise, methods for protein estimation. It was found that the underestimation of the protein content of membrane-containing fractions cannot be overcome by pretreatment with NaOH or the detergents employed (Triton X-100, sodium dodecyl sulfate, 3[(3-cholamidopropyl)-dimethylammonio]propanesulfonic acid) and the protein estimates obtained do not agree with estimates obtained by the Lowry assay. Upon storage of fractions at -20 degrees C there is a considerable loss of dye binding activity, varying in accordance with the membrane content of the fractions, reaching up to 58% in the case of membrane-enriched fractions stored at -20 degrees C for 15 days. Pretreatment with the employed agents brought about an equal increase of dye binding capacity, specific for the individual fractions; however, none of these agents could recover the dye binding activity lost during several days of storage at -20 degrees C. It is suggested that the straightforward Bradford procedure has a rather limited scope of application, particularly concerning membrane-containing samples, and requires preliminary studies to determine its applicability according to the nature of the biological material examined.