Cargando…

637. Whole-Genome Sequencing (WGS) of Isolates from Resolving (RB) and Persistent (PB) Methicillin-Resistant Staphylococcus aureus Bacteremia

BACKGROUND: The basis for PB in MRSA is poorly understood. We tested the hypothesis that PB isolates will exhibit distinct genetic variation vs. resolving bacteremia (RB). METHODS: MRSA isolates from an existing cohort of prospectively enrolled patients with PB (n = 81) and RB (n = 81) were matched...

Descripción completa

Detalles Bibliográficos
Autores principales: Sharma-Kuinkel, Batu K, Medie, Felix, Ruffin, Felicia, Wattam, Rebecca, Shukla, Maulik, Davis, James, Yeaman, Michael, Fowler, Vance G
Formato: Online Artículo Texto
Lenguaje:English
Publicado: Oxford University Press 2019
Materias:
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6810939/
http://dx.doi.org/10.1093/ofid/ofz360.705
Descripción
Sumario:BACKGROUND: The basis for PB in MRSA is poorly understood. We tested the hypothesis that PB isolates will exhibit distinct genetic variation vs. resolving bacteremia (RB). METHODS: MRSA isolates from an existing cohort of prospectively enrolled patients with PB (n = 81) and RB (n = 81) were matched by sex, age, dialysis, diabetes mellitus, and presence of any implantable medical device. WGS was performed via MiSeq 2x150 runs (Illumina®) with a targeted depth of 150-fold at Duke Genome Sequencing Core (DUGSIM). The genome sequence was analyzed, assembled, and annotated in PATRIC. RESULTS: No genetic signature was associated with PB or RB isolates at the genome-wide significance level. Phylogenetic analysis separated the genomes into two distinct clades [Clade 1 (C1): 77 genomes; predominantly CC5; Clade 2 (C2): 74 genomes; predominantly CC8; Figure 1]. Most isolates in C1 genome carried SCC Mec type II (2A) vs. the SCC Mec type IV (2B) in C2. C1 and C2 had a total of 3602 (407 unique only to C1) and 3563 (368 unique only to C2) protein families, respectively, with 3195 families shared across the two groups. Out of the 407 unique C1 protein families, majority were not shared across all the genomes and only 28 were found in more than 40 of the genomes. A detailed look of 28 protein families showed that majority were found to be in 3 distinct genomic regions (R1—5.6 kb, 7 genes, also contains mecI; R2—18.2 kb; 16 genes; R3—6.3 kb, 6 genes, 2 predicted erythromycin resistance gene). Furthermore, 12 of C1 genomes lack these unique regions. Most of the genomes in C1 (Other than 12 genomes) carry genes responsible for erythromycin and clindamycin resistance. Out of the 368 protein families unique to C2, only 3 are widely shared (2 hypothetical proteins and 1 two-component transcriptional regulator. Patient characteristics were not associated with clade type. CONCLUSION: The basis for the PB phenotype in patients with MRSA bacteremia is not due to sequence-based variation in the bacterial genome. [Image: see text] DISCLOSURES: All authors: No reported disclosures.