Study of the bacterial diversity of foods: PCR-DGGE versus LH-PCR

- Bacterial microbiota in food was explored by LH-PCR, PCR-DGGE and viable counts.
- The total microbial DNA and RNA were extracted directly from foods.
- Only partial overlapping of bacteria was found using LH-PCR and PCR-DGGE.
- A higher sensitivity of LH-PCR relative to PCR-DGGE can be hypothesized.
- The LH-PCR database needs to be expanded.

The present study compared two culture-independent methods, polymerase chain reaction denaturing gradient gel electrophoresis (PCR-DGGE) and length-heterogeneity polymerase chain reaction (LH-PCR), for their ability to reveal food bacterial microbiota. Total microbial DNA and RNA were extracted directly from fourteen fermented and unfermented foods, and domain A of the variable regions V1 and V2 of the 16S rRNA gene was analyzed through LH-PCR and PCR-DGGE. Finally, the outline of these analyses was compared with bacterial viable counts obtained after bacterial growth on suitable selective media.For the majority of the samples, RNA-based PCR-DGGE revealed species that the DNA-based PCR-DGGE was not able to highlight. When analyzing either DNA or RNA, LH-PCR identified several lactic acid bacteria (LAB) and coagulase negative cocci (CCN) species that were not identified by PCR-DGGE. This phenomenon was particularly evident in food samples with viable loads < 5.0 Log cfu g− 1. Furthermore, LH-PCR was able to detect a higher number of peaks in the analyzed food matrices relative to species identified by PCR-DGGE. In light of these findings, it may be suggested that LH-PCR shows greater sensitivity than PCR-DGGE. However, PCR-DGGE detected some other species (LAB included) that were not detected by LH-PCR. Therefore, certain LH-PCR peaks not attributed to known species within the LH-PCR database could be solved by comparing them with species identified by PCR-DGGE. Overall, this study also showed that LH-PCR is a promising method for use in the food microbiology field, indicating the necessity to expand the LH-PCR database, which is based, up to now, mainly on LAB isolates from dairy products.