Poster Presentation Australian Society for Microbiology Annual Scientific Meeting 2017

Genomic analysis of Acinetobacter baumannii isolated from 93 bacteremia patients in Taiwan (#223)

Tzu-Wen Huang 1 , Yi-Tzu Lee 2 , Shu-Chen Kuo 3 , Yen-Ming Liu 4 , Tsai-Lien Liao 4 , Chang-Phone Fung 5 , Shih-Feng Tsai 4 , Te-Li Chen 6
  1. Taipei Medical University, Taipei, Taiwan
  2. Emergency Department, Taipei Veterans General Hospital, Taipei, Taiwan
  3. National Institute of Infectious Diseases and Vaccinology, National Health Research Institutes, Zhunan, Taiwan
  4. Institute of Molecular and Genomic Medicine, National Health Research Institutes, Zhunan, Taiwan
  5. Division of Infectious Diseases, Department of Medicine, Taipei Veterans General Hospital, Taipei, Taiwan
  6. Institute of Clinical Medicine, School of Medicine, National Yang-Ming University, Taipei, Taiwan

Acinetobacter baumannii is an important healthcare-associated pathogen. It causes pneumonia, surgical and bloodstream infections, and contributes to the increasing incidence of multidrug resistance globally. Here, we randomly selected 93 multidrug-resistant isolates of Acinetobacter baumannii from bacteremia patients at Taipei Veterans General Hospital (T-VGH, Taiwan) collected from 1998 to 2008. Whole genome sequences of 93 isolates were obtained simultaneously using the MiSeq platform and were assembled de no by the CLC bio software. In silico analysis of multilocus sequence typing (MLST) showed that 95% (89/93) isolates could be identified, in which 66% (59/89) belong to the global dominant lineage (ST-2 or named International Clone II, IC-II); whereas 4 assigned to ST-20, which was International Clone I (IC-I). For detection of antimicrobial resistance genes (ARGs), 65% (61/93) isolates contained the armA gene conferring aminoglycosides resistance and most of them carried two or more classes of ARGs. Comparing their capsule gene cluster reported capsule (KL) types, 57% (53/93) of them could be classified into 7 KL types. Among them, KL2 was a dominant type (20/53), and KL3 (10/53) and KL10 (9/53) ran second. Phylogenetic analysis showed that no dominant ST or KL types were found among bloodstream isolates. The phylogenetic clusters generated with whole-genome data were correlated with ST and KL types. Isolates (30/93) excluding from IC-I and IC-II were mainly clustered into two subgroups, ST-10 with unidentified KL types and diverse ST and unknown KL types. The genomic profiles of MLST, KL types, ARGs and their phylogeny revealed the diversity of Acinetobacter baumannii from bloodstream isolates. Our study identified that a significant fraction (40%) of them did not belong to the known KL types, and 24% of the isolates remains unclear for the genetic determinants of the multidrug resistance. Thus, the clinical and genomic information of this collection is valuable in uncovering new resistance mechanisms and mining the possible virulence factors associated with bloodstream infection of Acinetobacter baumannii.