Behavior of Bradford-reactive substances is consistent with predictions for glomalin

There is considerable controversy concerning detection in soils of the protein, glomalin, which is produced by arbuscular mycorrhizal fungi. Glomalin was originally defined as a substance that cross reacts with a monoclonal antibody formed against a substance in the cell walls of an arbuscular mycorrhizal fungus. Thus, one can use an immunological approach to detect glomalin. However, it was recently shown that other proteins cross react with the antibody. The other, more common, approach involves assay of soil protein using the Bradford reaction. This approach assumes that the Bradford assay is specific to protein, and that the assayed protein is largely glomalin, either because other proteins are in low concentration, or because the extraction process eliminates the possibility of their detection. These assumptions, however, have been called into question recently. One way to test whether the Bradford assay can be useful in quantifying glomalin is to determine whether the concentrations of Bradford-reactive substances are consistent with predictions for glomalin. For example, if recently produced glomalin is more labile than older glomalin, the concentrations of the two fractions should not be highly correlated. Moreover, when a contrast is established between mycorrhizal and nonmycorrhizal vegetation, recently produced glomalin should soon occur in higher concentration in soils supporting mycorrhizal vegetation. Older glomalin should not be found in higher concentrations in the soils of mycorrhizal vegetation until some time later. We tested these predictions by employing the Bradford assay during the course of a three-year field experiment in which canola (nonmycorrhizal) and soy (mycorrhizal) were grown in separate plots in year 1, both of which were followed by maize (mycorrhizal) in years 2 and 3. The correlation between the concentrations of fraction 1 Bradford-reactive substances (also known as easily extractable glomalin and frequently assumed to be recently produced) and fraction 2 (the more difficult-to-extract fraction, and frequently assumed to be older glomalin), was very poor. In year 1, the concentration of fraction 1 was significantly greater in soy plots than in canola plots. Finally, fraction 2 was only significantly higher in the former soy plots than in former canola plots in years 2 and 3. These data support the hypothesis that the Bradford assay was useful in detecting glomalin in this case. (C) 2012 Elsevier B.V. All rights reserved.