Searching whole genome sequences for biochemical identification features of emerging and reemerging pathogenic Corynebacterium species

Abstract

Biochemical tests are traditionally used for bacterial identification at the species level in clinical microbiology laboratories. While biochemical profiles are generally efficient for the identification of the most important corynebacterial pathogen Corynebacterium diphtheriae, their ability to differentiate between biovars of this bacterium is still controversial. Besides, the unambiguous identification of emerging human pathogenic species of the genus Corynebacterium may be hampered by highly variable biochemical profiles commonly reported for these species, including Corynebacterium striatum, Corynebacterium amycolatum, Corynebacterium minutissimum, and Corynebacterium xerosis. In order to identify the genomic basis contributing for the biochemical variabilities observed in phenotypic identification methods of these bacteria, we combined a comprehensive literature review with a bioinformatics approach based on reconstruction of six specific biochemical reactions/pathways in 33 recently released whole genome sequences. We used data retrieved from curated databases (MetaCyc, PathoSystems Resource Integration Center (PATRIC), The SEED, TransportDB, UniProtKB) associated with homology searches by BLAST and profile Hidden Markov Models (HMMs) to detect enzymes participating in the various pathways and performed ab initio protein structure modeling and molecular docking to confirm specific results. We found a differential distribution among the various strains of genes that code for some important enzymes, such as beta-phosphoglucomutase and fructokinase, and also for individual components of carbohydrate transport systems, including the fructose-specific phosphoenolpyruvate-dependent sugar phosphotransferase (PTS) and the ribose-specific ATP-binging cassette (ABC) transporter. Horizontal gene transfer plays a role in the biochemical variability of the isolates, as some genes needed for sucrose fermentation were seen to be present in genomic islands. Noteworthy, using profile HMMs, we identified an enzyme with putative alpha-1,6-glycosidase activity only in some specific strains of C. diphtheriae and this may aid to understanding of the differential abilities to utilize glycogen and starch between the biovars.

Population genomics and antimicrobial resistance in Corynebacterium diphtheriae

“…Our genome-wide association study (GWAS) of accessory genes with the biovar phenotype revealed a strong association of a cluster of genes that includes spuA (DIP357; Additional file 4: Fig. S7) with biovar Gravis isolates, providing statistical support to the discovery of Santos et al [62]. This association was stronger within the Gravis lineage; in contrast within the Mitis lineage, few (5 out of 17) of the biovar Gravis isolates possessed spuA (Fig.…”

“…The nitrate reductase activity differentiates Mitis and Gravis isolates, which are positive, from Belfanti isolates, which are nitrate-negative. We found that the nitrate reduction narKGHJI gene cluster [62] was disrupted in three of the six isolates assigned to the biovar Belfanti: strains FRC0480 and FRC0481 had a G to A mutation at position 675 of the narG gene, leading to a stop codon, whereas in strain CIPA99, approximately 100 nucleotides were inserted at position 446 in narG. No molecular explanation was found for the lack of nitrate reductase ability of the three other Belfanti strains when scrutinizing the narKGHJI gene cluster and adjacent molybdenum cofactor biosynthesis genes [63].…”

“…Whereas previous work concluded on the absence of an association [11], our phylogenetic analyses reveal that Gravis and Mitis, the two main biovars of C. diphtheriae, are associated strongly with two phylogenetic lineages. Lineage Gravis appears to have acquired ancestrally a gene cluster comprising the extracellular glycogen debranching enzyme gene spuA [62]. Although the Gravis phenotype was associated with spuA within lineage Gravis, other genomic determinants of glycogen utilization remain to be discovered within the Mitis lineage.…”

“…Biovar Mitis and Gravis are distinguished by the ability to utilize glycogen (positive in Gravis, negative in Mitis). The spuA gene, which codes for a putative alpha-1,6-glycosidase, was reported as being specific for biovar Gravis isolates [62]. Our genome-wide association study (GWAS) of accessory genes with the biovar phenotype revealed a strong association of a cluster of genes that includes spuA (DIP357; Additional file 4: Fig.…”

Direct detection of Corynebacterium striatum, Corynebacterium propinquum, and Corynebacterium simulans in sputum samples by high-resolution melt curve analysis

“…Traditionally, these microorganisms have been routinely identified by biochemical tests using the API Coryne system (bioMérieux, Craponne, France) or the RapID CB PLUS system (Thermo Fisher Scientific, Waltham, MA, USA) in clinical microbiology laboratories [11,12]. However, these methods have low sensitivity, and are timeconsuming and unreliable for species identification, especially in the case of C. simulans, due to its similarities with C. striatum [13,14].…”