Comparison of Biolog and Phospholipid Fatty Acid Patterns to Detect Changes in Microbial Community

Abstract

To determine phenotypic composition of microbial communities in soil, community level physiological profiles by Biolog® GN microtiter plates, and phospholipid fatty acid (PLFA) analysis were applied. An incubation experiment was carried out to compare the methods for soils of different origin (Denmark, Germany, Italy) with and without maize straw application. The straw was placed either on the soil surface (mulching) or incorporated into the soil. Destructive samplings were carried out 0, 2, 4, 16, and 52 weeks after the application. Profiles of utilization capacity of carbon substrates and PLFAs were statistically analyzed by two multivariate methods: principal component analysis (PCA) and discriminant analysis (DA). The Biolog and PLFA patterns often showed the same trend but PLFA achieved more clear groupings. Selective growth of bacteria adapted to the conditions in Biolog microtiter wells might explain some differences between the methods. The basic recommendations to apply PCA (e.g. sufficient observations to variables ratio) were not always fulfilled. Discriminant analysis was applicable for data sets since grouping was known a priori, and led to better separation than PCA.