Kommedal Øyvind, Wilhelmsen Marianne Thulin, Skrede Steinar, Meisal Roger, Jakovljev Aleksandra, Gaustad Peter, Hermansen Nils Olav, Vik-Mo Einar, Solheim Ole, Ambur Ole Herman, Sæbø Øystein, Høstmælingen Christina Teisner, Helland Christian
Department of Microbiology, Haukeland University Hospital, Bergen, Norway
Department of Microbiology, Haukeland University Hospital, Bergen, Norway.
J Clin Microbiol. 2014 Jun;52(6):1990-7. doi: 10.1128/JCM.00346-14. Epub 2014 Mar 26.
Rapid development within the field of massive parallel sequencing (MPS) is about to bring this technology within reach for diagnostic microbiology laboratories. We wanted to explore its potential for improving diagnosis and understanding of polymicrobial infections, using bacterial brain abscesses as an example. We conducted a prospective nationwide study on bacterial brain abscesses. Fifty-two surgical samples were included over a 2-year period. The samples were categorized as either spontaneous intracerebral, spontaneous subdural, or postoperative. Bacterial 16S rRNA genes were amplified directly from the specimens and sequenced using Ion Torrent technology, with an average of 500,000 reads per sample. The results were compared to those from culture- and Sanger sequencing-based diagnostics. Compared to culture, MPS allowed for triple the number of bacterial identifications. Aggregatibacter aphrophilus, Fusobacterium nucleatum, and Streptococcus intermedius or combinations of them were found in all spontaneous polymicrobial abscesses. F. nucleatum was systematically detected in samples with anaerobic flora. The increased detection rate for Actinomyces spp. and facultative Gram-negative rods further revealed several species associations. We suggest that A. aphrophilus, F. nucleatum, and S. intermedius are key pathogens for the establishment of spontaneous polymicrobial brain abscesses. In addition, F. nucleatum seems to be important for the development of anaerobic flora. MPS can accurately describe polymicrobial specimens when a sufficient number of reads is used to compensate for unequal species concentrations and principles are defined to discard contaminant bacterial DNA in the subsequent data analysis. This will contribute to our understanding of how different types of polymicrobial infections develop.
大规模平行测序(MPS)领域的快速发展即将使诊断微生物学实验室能够使用这项技术。我们想以细菌性脑脓肿为例,探索其在改善多微生物感染诊断和理解方面的潜力。我们对细菌性脑脓肿进行了一项全国性前瞻性研究。在两年时间里纳入了52份手术样本。这些样本分为自发性脑内、自发性硬膜下或术后样本。细菌16S rRNA基因直接从标本中扩增,并使用Ion Torrent技术进行测序,每个样本平均有500,000条读数。将结果与基于培养和桑格测序的诊断结果进行比较。与培养相比,MPS能够鉴定出三倍数量的细菌。在所有自发性多微生物脓肿中均发现嗜沫聚集杆菌、具核梭杆菌和中间链球菌或它们的组合。在有厌氧菌群的样本中系统地检测到具核梭杆菌。放线菌属和兼性革兰氏阴性杆菌的检测率增加进一步揭示了几种菌种关联。我们认为嗜沫聚集杆菌、具核梭杆菌和中间链球菌是自发性多微生物脑脓肿形成的关键病原体。此外,具核梭杆菌似乎对厌氧菌群的发展很重要。当使用足够数量的读数来补偿物种浓度不均,并在后续数据分析中定义丢弃污染细菌DNA的原则时,MPS可以准确描述多微生物标本。这将有助于我们理解不同类型的多微生物感染是如何发展的。