Sensitive detection of <i>Ralstonia solanacearum</i> in soil: a comparison of different detection techniques

Abstract

The sensitivity and specificity of various methods were compared for routine detection of Ralstonia solanacearum in a sandy loam soil. Populations fewer than 10 2 CFU per g soil were detected by dilution plating on a modified semiselective medium (SMSA). In comparison, a tomato bioassay was shown consistently to detect populations at or greater than 7·5 × 10 5 CFU per g soil. An indirect enzyme‐linked immunosorbent assay (ELISA) was as sensitive as the tomato bioassay, but detected as few as 10 4 CFU per g soil when the suspension was first incubated in SMSA broth prior to testing. Detection using a nested polymerase chain reaction (PCR) was equally as sensitive as that using culture on SMSA agar, but only when the infested soil sample was first enriched overnight in SMSA broth prior to the nested PCR. Longer incubation periods in SMSA broth also increased the sensitivity of pathogen detection using a conventional PCR method, permitting detection of as few as 10 2 CFU per g soil after 60 h enrichment in SMSA broth. When evaluated using naturally infected field soils in Nepal, isolation of R. solanacearum on SMSA was reliable only when pathogen populations were higher than those of saprophytic soilborne bacteria. As few as 5 × 10 2 CFU of R. solanacearum per g were recovered from naturally infested soil, whereas the sensitivity of indirect ELISA was 10 6 CFU g −1 .