Universal bacterial identification by mass spectrometry of 16S ribosomal RNA cleavage products

The public availability of over 180,000 bacterial 16S ribosomal RNA (rRNA) sequences has facilitated microbial identification and classification using nucleic acid hybridization and other molecular approaches. Species-specific PCR, microarrays, and in situ hybridization are based on the presence of unique subsequences in the target sequence and therefore require prior knowledge of what organisms are likely to be present in a sample. Mass spectrometry is not limited by a pre-synthesized inventory of probe/primer sequences. It has already been demonstrated that organism identification can be recovered from mass spectra using various methods including base-specific cleavage of nucleic acids. The feasibility of broad bacterial identification by comparing such mass spectral patterns to predictive databases derived from virtually all previously sequenced strains has yet to be demonstrated, however. Herein, we present universal bacterial identification by base-specific cleavage, mass spectrometry, and an efficient coincidence function for rapid spectral scoring against a large database of predicted "mass catalogs". Using this approach in conjunction with universal PCR of the 16S rDNA gene, four bacterial isolates and an uncultured clone were successfully identified against a database of predicted cleavage products derived from over 47,000 16S rRNA sequences representing all major bacterial taxa. At present, the conventional DNA isolation and PCR steps require approximately 2 h, while subsequent transcription, enzymatic cleavage, mass spectrometric analysis, and database comparison require less than 45 min. All steps are amenable to high-throughput implementation.