Application of whole genome sequence data in analyzing the molecular epidemiology of Shiga toxin-producing Escherichia coli O157:H7/H

A novel PCR-based genotyping method for Proteus mirabilis - Intergenic region polymorphism analysis

Proteus mirabilis is a predominant species in cases of food poisoning associated with meat products and is also an opportunistic pathogen causing numerous infections in humans. This study aimed to differentiate P. mirabilis isolates using intergenic region polymorphism analysis (IRPA). The IRPA typing scheme was developed to amplify polymorphic fragments in intergenic regions (IGRs). The presence, absence, or size change of amplified products were identified and utilized as genetic markers for rapid differentiation of strains. A total of 75 P. mirabilis isolates were isolated from 63 fresh poultry and pork samples were subtyped using the IRPA and ERIC-PCR methods, and their antibiotic resistance profiles were tested. The majority of P. mirabilis isolates showed resistance to tetracycline (85.3%), doxycycline (93.3%), chloramphenicol (82.7%), streptomycin (92.0%), spectinomycin (80.0%), trimethoprim (97.3%); trimethoprim-sulfalleth (82.7%), and erythromycin (100.0%). In contrast, resistance rates to ceftriaxon, cefoxitin, cefepime, and cefotaxim were lower at only 17.3%, 5.3%, 6.7%, and 13.3%, respectively, among P. mirabilis isolates. Eleven loci were selected for analysis of the genetic diversity of 75 P. mirabilis isolates. A combination of 4 loci was determined as the optimal combination. The results compared to those obtained using ERIC-PCR for the same isolates. The Simpson's index of diversity was 0.999 for IRPA and 0.923 for ERIC-PCR, indicating that IRPA has a higher discriminatory power than ERIC-PCR. The concordance between IRPA and ERIC-PCR methods was low, primarily because IRPA classified isolates from the same ERIC cluster into separate clusters due to its high resolution. The IRPA method presented in this study offers a rapid, simple, reproducible, and economical approach for genotyping P. mirabilis.

Clarification of relationship between single-nucleotide polymorphism panels of Shiga toxin-producing Escherichia coli O157:H7/H- strains

Eighty strains of enterohemorrhagic Escherichia coli O157:H7/H- were analyzed by three single-nucleotide polymorphism (SNP) panels using whole-genome sequencing data. The partial concordance of SNP types among the different SNP panels was observed on minimum spanning trees reconstructed with SNP data. As for lineage I/II strains, some of the clade 7 strains belonged to one unique SNP type as determined by three panels, suggesting that clade 7 should be divided into at least two genotypes, namely, the unique type and the rest. In addition, clade 8 contained two unique genotypes, which was consistent with the previous prediction. Similarly, for lineage II, clade 12 should be divided into three genotype strains. In contrast, many strains of several clades belonging to lineage I were clustered into the same node on each minimum spanning tree upon testing with the three SNP panels. Previous studies reported that lineage I diverged more recently than lineages I/II and II. Such low diversity in lineage I in this study may have arisen because this lineage has not accumulated SNPs because of its relatively recent divergence. Based on the concordance observed in this study, some of the previously published O157 genotype distribution data were successfully interpreted to clarify the clade distribution, which was well supported by previous literature.

Development of Single Nucleotide Polymorphism (SNP)-Based Triplex PCR Marker for Serotype-Specific Escherichia coli Detection

Single-nucleotide polymorphisms (SNPs) are one of the most common forms of genetic variation and as such are powerful tools for the identification of bacterial strains, their genetic diversity, phylogenetic analysis, and outbreak surveillance. In this study, we used 15 sets of SNP-containing primers to amplify and sequence the target Escherichia coli. Based on the combination of the 15-sequence primer sets, each SNP site encompassing forward and reverse primer sequences (620–919 bp) were aligned and an SNP-based marker was designed. Each SNP marker exists in at least two SNP sites at the 3′ end of each primer; one natural and the other artificially created by transition or transversion mutation. Thus, 12 sets of SNP primers (225–488 bp) were developed for validation by amplifying the target E. coli. Finally, a temperature gradient triplex PCR kit was designed to detect target E. coli strains. The selected primers were amplified in three genes (ileS, thrB, and polB), with fragment sizes of 401, 337, and 232 bp for E. coli O157:H7, E. coli, and E. coli O145:H28, respectively. This allele-specific SNP-based triplex primer assay provides serotype-specific detection of E. coli strains in one reaction tube. The developed marker would be used to diagnose, investigate, and control food-borne E. coli outbreaks.

Molecular epidemiological analysis of Mycobacterium tuberculosis modern Beijing genotype strains isolated in Chiba Prefecture over 10 years

INTRODUCTION

The prevalence of the phylogenetic groups of Mycobacterium tuberculosis Beijing genotype has been reported to be similar in different areas of Japan. However, recent reports from rural areas of Japan show a low prevalence of modern Beijing strains, suggesting that the distribution of modern Beijing strains may have changed recently. Therefore, multi-locus variable number of tandem repeats analysis (MLVA) and draft whole genome sequence (DWGS) analysis were carried out to investigate the prevalence of particular genotype strains.

METHODS

Nine hundred and ninety modern Beijing strains were studied using minimum spanning tree (MST) analysis and neighbor-net analysis of MLVA and WGS data.

RESULTS

An MST of M. tuberculosis Beijing genotype strains reconstructed from 12 loci-MLVA data showed two large complexes with the J₁₂-0006 MLVA pattern. In one of the complexes, strains with the pECT07 pattern produced by 24 loci-MLVA and its SLVs were most prevalent. DWGS analysis was carried out for pECT07 and its SLV strains. Neighbor-net and MST analyses of the DWGS data showed that pECT07 and its SLV strains were grouped in separate clusters. When all the combinations of two of the tested strains were analyzed, MST analysis showed that only 9 (1.7%) of the 528 pairs of tested strains had 5 or less SNPs.

CONCLUSIONS

The results of this study suggested that pECT07 and its variants were prevalent among M. tuberculosis modern Beijing strains in Chiba Prefecture, but the prevalence of those strains may not have been due to an earlier large-scale latent outbreak.

Whole Genome Analysis Detects the Emergence of a Single Salmonella enterica Serovar Chester Clone in Japan's Kanto Region

In Japan's Kanto region, the number of Salmonella enterica serovar Chester infections increased temporarily between 2014 and 2016. Concurrently with this temporal increase in the Kanto region, S. Chester isolates belonging to one clonal group were causing repetitive outbreaks in Europe. A recent study reported that the European outbreaks were associated with travelers who had been exposed to contaminated food in Morocco, possibly seafood. Because Japan imports a large amount of seafood from Morocco, we aimed to establish whether the temporal increase in S. Chester infections in the Kanto region was associated with imported Moroccan seafood. Short sequence reads from the whole-genome sequencing of 47 S. Chester isolates from people in the Kanto region (2014-2016), and the additional genome sequences from 58 isolates from the European outbreaks, were analyzed. The reads were compared with the complete genome sequence from a S. Chester reference strain, and 347 single nucleotide polymorphisms (SNPs) were identified. These SNPs were used in this study. Cluster and Bayesian cluster analyses showed that the Japanese and European isolates fell into two different clusters. Therefore, Φ PT and I A S values were calculated to evaluate genetic differences between these clusters. The results revealed that the Japanese and European isolates were genetically distinct populations. Our root-to-tip analysis showed that the Japanese isolates originating from one clone had accumulated mutations, suggesting that an emergence of this organism occurred. A minimum spanning tree analysis demonstrated no correlation between genetic and geographical distances in the Japanese isolates, suggesting that the emergence of the serovar in the Kanto region did not involve person-to-person contact; rather, it occurred through food consumption. The d N /d S ratio indicated that the Japanese strain has evolved under positive selection pressure. Generally, a population of bacterial clones in a reservoir faces negative selection pressure. Therefore, the Japanese strain must have existed outside of any reservoir during its emergence. In conclusion, S. Chester isolates originating from one clone probably emerged in the Kanto region via the consumption of contaminated foods other than imported Moroccan seafood. The emerging strain may have not established a reservoir for survival in the food supply chain resulting in its disappearance after 2017.

Top-Down Proteomic Identification of Shiga Toxin 1 and 2 from Pathogenic Escherichia coli Using MALDI-TOF-TOF Tandem Mass Spectrometry

Shiga-toxin-producing Escherichia coli (STEC) are a burden on agriculture and a threat to public health. Rapid methods are needed to identify STEC strains and characterize the Shiga toxin (Stx) they produce. We analyzed three STEC strains for Stx expression, using antibiotic induction, matrix-assisted laser desorption/ionization time-of-flight-time-of-flight (MALDI-TOF-TOF) mass spectrometry, and top-down proteomic analysis. E. coli O157:H- strain 493/89 is a clinical isolate linked to an outbreak of hemolytic uremic syndrome (HUS) in Germany in the late 1980s. E. coli O145:H28 strains RM12367-C1 and RM14496-C1 were isolated from an agricultural region in California. The stx operon of the two environmental strains were determined by whole genome sequencing (WGS). STEC strain 493/89 expressed Shiga toxin 2a (Stx2a) as identified by tandem mass spectrometry (MS/MS) of its B-subunit that allowed identification of the type and subtype of the toxin. RM12367-C1 also expressed Stx2a as identified by its B-subunit. RM14496-C1 expressed Shiga toxin 1a (Stx1a) as identified from its B-subunit. The B-subunits of Stx1 and Stx2 both have an intramolecular disulfide bond. MS/MS was obtained on both the disulfide-bond-intact and disulfide-bond-reduced B-subunit, with the latter being used for top-down proteomic identification. Top-down proteomic analysis was consistent with WGS.

Isolation of Salmonella enterica serovar Agona strains and their similarities to strains derived from a clone caused a serovar shift in broilers

Salmonella enterica serovar Agona strains isolated from human cases were compared to strains that were derived from a clone caused a serovar shift in broilers. Pulsed field gel electrophoresis (PFGE) analysis with XbaI or BlnI digestion showed that three of seven strains from human case strains and most of the 81 strains from broilers were clustered in single complex in a minimum spanning tree (MST) reconstructed from the PFGE data. All the strains from human cases and 22 randomly selected strains from broilers were also analyzed by whole genome sequencing (WGS). Analysis of single nucleotide polymorphism (SNP) in the S. Agona core genes showed that four strains from human cases and all the strains from broilers were clustered in a maximum likelihood phylogenetic tree (ML tree) and an MST. These results indicated that the strains derived from the clone caused the serovar shift had already spread to humans. PFGE analysis with XbaI showed that four strains from broilers did not cluster with the other strains in an MST, though all those strains clustered in an ML tree and an MST reconstructed from SNP data. Moreover, three strains from broilers did not cluster in an MST reconstructed from PFGE with BlnI digestion, though those strains clustered in an ML tree and an MST reconstructed from SNP data. Therefore, it was suggested that S. Agona strains derived from a particular clone could not be traced by PFGE analysis but can be investigated by WGS analysis.