The 16S rRNA lung microbiome in mechanically ventilated patients: a methodological study.

Purpose: Characterization of the respiratory tract bacterial microbiome is in its infancy when compared to the gut microbiota. To limit bias mandates a robust methodology. Specific amplification of the hypervariable (V) region of the 16SrRNA gene is a crucial step. Differences in accuracy exist for one V region to another depending on the sampled environment. We aimed to assess the impact of the primer sequences targeting the V4 region currently used for gut microbiota studies in respiratory samples. Materials and methods: The original 515 F-806R primer pair targets the V4 region of the 16SrRNA gene. We compared two different 515 F-806R primer pairs before Illumina 250 paired-end sequencing for bacterial microbiome analyses of respiratory samples from critically-ill ventilated patients. "S-V4" for "Stringent V4" primer pair is used in two ongoing international projects "the Integrative Human microbiome project (iHMP)" and "the Earth microbiome project (EMP)." "R-V4" for "Relaxed V4" primer pair has been modified to reduce biases against specific environmental taxa. The optimal method was determined by concordance with conventional microbiology. Results: Twenty samples from three patients who developed a ventilator-associated pneumonia (VAP) and four who did not (control ventilated patients) were sequenced. Highly different results were obtained. "S-V4" provided the best agreement with the conventional microbiology for endotracheal aspirate: 89% as compared to 56% for "R-V4." The main difference related to poor Enterobacteriaceae detection with "R-V4" primers. Conclusions: Accuracy of the bacterial lung microbiome composition was highly dependent on the primers used for amplification of the 16 s rRNA hypervariable sequence. This work validates for future lung microbiome studies the use of the 515 F-806R "S-V4" primer pair associated to Illumina® MiSeq paired-end sequencing.