Kiel Matthias, Sagory-Zalkind Pierre, Miganeh Céline, Stork Christoph, Leimbach Andreas, Sekse Camilla, Mellmann Alexander, Rechenmann François, Dobrindt Ulrich
Institute of Hygiene, University of Münster, Münster, Germany.
Genostar Bioinformatics, Montbonnot-Saint-Martin, France.
Front Microbiol. 2018 Jun 26;9:1321. doi: 10.3389/fmicb.2018.01321. eCollection 2018.
It would be desirable to have an unambiguous scheme for the typing of Shiga toxin-producing (STEC) isolates to subpopulations. Such a scheme should take the high genomic plasticity of into account and utilize the stratification of STEC into subgroups, based on serotype or phylogeny. Therefore, our goal was to identify specific marker combinations for improved classification of STEC subtypes. We developed and evaluated two bioinformatic pipelines for genomic marker identification from larger sets of bacterial genome sequences. Pipeline A performed all-against-all BLASTp analyses of gene products predicted in STEC genome test sets against a set of control genomes. Pipeline B identified STEC marker genes by comparing the STEC core proteome and the "pan proteome" of a non-STEC control group. Both pipelines defined an overlapping, but not identical set of discriminative markers for different STEC subgroups. Differential marker prediction resulted from differences in genome assembly, ORF finding and inclusion cut-offs in both workflows. Based on the output of the pipelines, we defined new specific markers for STEC serogroups and phylogenetic groups frequently associated with outbreaks and cases of foodborne illnesses. These included STEC serogroups O157, O26, O45, O103, O111, O121, and O145, Shiga toxin-positive enteroaggregative O104:H4, and HUS-associated sequence type (ST)306. We evaluated these STEC marker genes for their presence in whole genome sequence data sets. Based on the identified discriminative markers, we developed a multiplex PCR (mPCR) approach for detection and typing of the targeted STEC. The specificity of the mPCR primer pairs was verified using well-defined clinical STEC isolates as well as isolates from the ECOR, DEC, and HUSEC collections. The application of the STEC mPCR for food analysis was tested with inoculated milk. In summary, we evaluated two different strategies to screen large genome sequence data sets for discriminative markers and implemented novel marker genes found in this genome-wide approach into a DNA-based typing tool for STEC that can be used for the characterization of STEC from clinical and food samples.
拥有一种明确的方案来将产志贺毒素大肠杆菌（STEC）分离株分型到亚群中将是很理想的。这样的方案应该考虑到其高基因组可塑性，并基于血清型或系统发育将STEC分层到亚组中。因此，我们的目标是识别特定的标记组合，以改进STEC亚型的分类。我们开发并评估了两种生物信息学流程，用于从更大的细菌基因组序列集中识别基因组标记。流程A对STEC基因组测试集中预测的基因产物与一组对照基因组进行全对全的BLASTp分析。流程B通过比较STEC核心蛋白质组和非STEC对照组的“泛蛋白质组”来识别STEC标记基因。两个流程都为不同的STEC亚组定义了一组重叠但不相同的鉴别标记。差异标记预测源于两个工作流程中基因组组装、开放阅读框查找和包含阈值的差异。基于流程的输出，我们为经常与食源性疾病暴发和病例相关的STEC血清群和系统发育群定义了新的特异性标记。这些包括STEC血清群O157、O26、O45、O103、O111、O121和O145、产志贺毒素的肠聚集性O104:H4以及与溶血尿毒综合征相关的序列型（ST）306。我们评估了这些STEC标记基因在全基因组序列数据集中的存在情况。基于所识别的鉴别标记，我们开发了一种多重PCR（mPCR）方法用于检测和分型目标STEC。使用明确的临床STEC分离株以及来自ECOR、DEC和HUSEC集合的分离株验证了mPCR引物对的特异性。用接种牛奶测试了STEC mPCR在食品分析中的应用。总之，我们评估了两种不同的策略来筛选大基因组序列数据集以寻找鉴别标记，并将在这种全基因组方法中发现的新标记基因应用到一种基于DNA的STEC分型工具中，该工具可用于临床和食品样本中STEC的特征分析。
