Intraspecific Variation in Small-Subunit rRNA Sequences in GenBank: Why Single Sequences May Not Adequately Represent Prokaryotic Taxa

Small-subunit rRNA (SSU rRNA) sequencing is a powerful tool to detect, identify, and classify prokaryotic organisms, and there is currently an explosion of SSU rRNA sequencing in the microbiology community. We report unexpectedly high levels of intraspecific variation (within and between strains) of prokaryote SSU rRNA sequences deposited in GenBank. A total of 82% of the prokaryote species with two published SSU rRNA sequences had more variable positions than a 0.1% random sequencing error would predict, and 48% of these sequence pairs had more variable positions than predicted by a 1.0% random sequencing error. Other sources of sequence variability must account for some of this intraspecific variation. Given these results, phylogenetic studies and biodiversity estimates obtained by using prokaryotic SSU rRNA sequences cannot proceed under the assumption that rRNA sequences of single operons from single isolates adequately represent their taxa. Sequencing SSU rRNA molecules from multiple operons and multiple isolates is highly recommended to obtain meaningful phylogenetic hypotheses, as is careful attention to accurate strain identification.