Molecular profiling of microbial communities associated with seeds of Beta vulgaris subsp Vulgaris (sugar beet)

The composition of microbial communities on and within seeds may effect their storage and field performance, whether they are indigenous or applied as biocontrol agents. In this study, we have explored the usefulness of profiling small subunit ribosomal (SSR) gene fragments for studying the microflora associated with seeds. DNA was amplified by the polymerase chain reaction (PCR) and the amplicons separated using denaturing gradient gel clectrophoresis (DGGE). Primers targeting eukaryotic SSRs were used to investigate fungal communities, and primers targeting bacterial SSRs were employed to study the eubacterial microflora. As a case study, we attempted to profile the fungi and bacteria associated with seeds of Beta vulgaris (sugar beet) to permit an insight into the varying field performance of several well-characterised commercial seed lots. Serious interference with the microbial signals was observed from the plant's own nuclear 18S rRNA genes and chloroplast 16S rRNA genes using standard PCR conditions and DNA extracted from whole seeds as template. Hot-start and touchdown PCR made no appreciable improvement to these signals. Seed imbibition and dissection into operculum and fruit wall and true seed prior to DNA extraction improved signal recovery in the fruit fraction. With primer modification, bacteria and fungi were detected in an excess of plant DNA of 100:1 and 10:1, respectively. With this method, microbial communities on seeds could be profiled, however, it is likely that targeted depletion of plant rDNA targets will be a necessary extra step before this approach can be used to screen seeds routinely. (C) 2003 Elsevier B.V. All rights reserved.