Fatty acid patterns of phospholipids and lipopolysaccharides in the characterisation of microbial communities in soil: a review

 This review discusses the analysis of whole-community phospholipid fatty acid (PLFA) profiles and the composition of lipopolysaccharides in order to assess the microbial biomass and the community structure in soils. For the determination of soil microbial biomass a good correlation was obtained between the total amount of PLFAs and the microbial biomass measured with methods commonly used for determinations such as total adenylate content and substrate-induced respiration. Generally, after the application of multivariate statistical analyses, whole-community fatty acid profiles indicate which communities are similar or different. However, in most cases, the organisms accounting for similarity or difference cannot be determined, and therefore artefacts could not be excluded. The fatty acids used to determine the biomass vary from those which determine the community structure. Specific attention has to be paid when choosing extraction methods in order to avoid the liberation of fatty acids from non-living organic material and deposits, and to exclude the non-target selection of lipids from living organisms, as well. By excluding the fatty acids which were presumed to be common and widespread prior to multivariate statistical analysis, estimates were improved considerably. Results from principal component analysis showed that determining the levels of fatty acids present in both low and high concentrations is essential in order to correctly identify microorganisms and accurately classify them into taxonomically defined groups. The PLFA technique has been used to elucidate different strategies employed by microorganisms to adapt to changed environmental conditions under wide ranges of soil types, management practices, climatic origins and different perturbations. It has been proposed that the classification of PLFAs into a number of chemically different subgroups should simplify the evaluating procedure and improve the assessment of soil microbial communities, since then only the subgroups assumed to be involved in key processes would be investigated. Received: 24 August 1998