Species identification and genotyping of Citrobacter spp. using genes with high nucleotide diversity

Citrobacter species are facultative anaerobic Gram-negative bacilli found in a wide range of habitats. Recently, an increasing number of nosocomial infections caused by Citrobacter species have been reported, including strains exhibiting high antimicrobial resistance. In clinical laboratory testing, Citrobacter species, such as Citrobacter freundii and Citrobacter braakii, are often collectively handled as the Citrobacter freundii complex. This can hamper species identification and genetic lineage estimation, which are necessary for estimating the distribution mechanism and evaluating the degree of antimicrobial resistance associated with each Citrobacter species. In this study, we investigated gene combinations that allow Citrobacter species identification and genetic lineage estimation using as few genes as possible. Genome sequences from 18 species were collected from GenBank, yielding a total of 453 genomes. Following gene annotation, nucleotide diversity was calculated for 1,113 genes contained in all the genomes. Genotypes of seven genes with the highest nucleotide diversity were confirmed, and their distributions among Citrobacter genomes were investigated. Except for Citrobacter werkmanii and Citrobacter cronae, the genotype of the highest-nucleotide-diversity gene (groups_3152) showed species-specific distribution, whereas combinations of the top two gene genotypes exhibited genetic lineage-specific distributions. Considering that Citrobacter werkmanii and Citrobacter cronae genomes are highly similar and difficult to distinguish using conventional genotyping methods, these findings suggest that combinations of the two genes with the highest nucleotide diversity can be used for Citrobacter species identification and genetic lineage classification.IMPORTANCEThe molecular epidemiology of individual Citrobacter species remains poorly understood because clinical laboratories often report these organisms collectively as the Citrobacter freundii complex. Multilocus sequence typing, which relies on seven conserved housekeeping genes, can also be limited when target loci are absent or deleted in some genomes. In this study, we propose a simple typing strategy based on two genes with high nucleotide diversity (HND) for both species identification and genetic lineage classification of Citrobacter species. Using a single HND gene enables species identification, while the combination of two genes allows lineage classification, substantially reducing costs compared with whole-genome sequencing. This approach may facilitate implementation in routine clinical laboratories and promote molecular epidemiological studies of individual Citrobacter species.