Validation of fluorescent in situ hybridization combined with flow cytometry for assessing interindividual variation in the composition of human fecal microflora during long-term storage of samples

Abstract

This work was conducted to assess the accuracy of in situ hybridization to show differences in human microflora composition between volunteers and to optimize the storage of fecal samples to allow delayed analysis of gut microflora composition in humans. Fecal samples from 25 healthy subjects (14 women, 11 men aged 24-51) were collected. The samples were fixed in 4% Paraformaldehyde (PFA) solution at 4°C overnight and stored at -70°C. Twenty samples were analysed to quantify the variation due to interindividual differences in the composition of fecal microflora. The five remaining samples were stored either after PFA fixation or directly frozen at -70°C and were monitored on a 12-month period. The fecal microflora was analysed by in situ hybridization combined with flow cytometry detection. Ribosomal RNA-targeted probes were used to assess the relative proportions of four phylogenetic groups: Clostridium coccoides-Eubacterium rectale (Erec 482), Bacteroides (Bac 303), Faecalibacterium prausnitzii (Fprau 645) and Bifidobacterium (Bif 164). Our results demonstrated that the method used is adapted to detect significant differences in fecal microflora composition in humans. Moreover, samples stored in PFA solution demonstrated a stable composition even after 8 months of storage. Conversely, frozen samples were less stable as the Bifidobacterium and C. coccoides-E. rectale groups showed significant differences after 2 months of storage. In conclusion, the fecal microflora composition can be analysed up to 8 months after 4% PFA fixation and storage at -70°C. It represents an extended time compared with the 2-month period currently recommended. This will give more flexibility for applying this technology in epidemiological studies including a large number of samples. © 2004 Elsevier B.V. All rights reserved.