This study evaluated multilocus sequence analysis (MLSA) using five protein-coding housekeeping genes (dnaJ, gyrB, hsp60, recA, rpoB) and 16S rRNA for identifying and classifying Bacteroides species. Researchers analyzed 38 strains representing 35 Bacteroides species using neighbor-joining, maximum-likelihood, and maximum-parsimony phylogenetic methods. The housekeeping genes, particularly gyrB and dnaJ, provided significantly better species discrimination than 16S rRNA alone. Analysis of concatenated sequences from all six genes produced robust phylogenetic relationships with almost complete resolution of Bacteroides species for the first time. The study defined Bacteroides species as sharing at least 97.5% sequence similarity across the analyzed gene fragments. MLSA demonstrated superior resolution to single-gene approaches, with concatenated sequences of just three genes (dnaJ, gyrB, hsp60) producing comparable results to the full six-gene analysis. This research establishes MLSA as a valuable alternative to DNA-DNA hybridization for accurate species identification and circumscription within the genus Bacteroides.

Key findings

• Housekeeping genes (gyrB, dnaJ) provide significantly better species discrimination than 16S rRNA gene sequences for Bacteroides identification
• Concatenated sequences of six genes produced robust, nearly fully resolved phylogenetic relationships among 35 Bacteroides species, achieving almost complete species resolution for the first time
• A threshold of 97.5% sequence similarity across five protein-coding housekeeping genes and 16S rRNA can define Bacteroides species boundaries
• Among housekeeping genes tested, gyrB was most informative with 50.6% parsimony-informative sites, compared to 19.6% for 16S rRNA
• MLSA using concatenated sequences of just three genes (dnaJ, gyrB, hsp60) produced phylogenetic results essentially equivalent to the six-gene analysis