High-resolution melting PCR assay as a powerful tool for the epidemiological surveillance of tularemia in Western Europe.

In Europe, tularemia is caused by Francisella tularensis subsp. holarctica and is a sporadic disease affecting mainly wildlife animals and humans. Classification of this species relies on canonical single nucleotide polymorphisms (canSNPs). Four main clades have been described for F. tularensis subsp. holarctica: B.4, B.6, B.12 and B.16. Phylogeographic studies have shown that clade B.6 is predominant in Western Europe and B.12 in Eastern and Central Europe. Based on this global phylogeny, we aimed to design a molecular typing assay for all genetic subclades of subclade B.11, which is the predominant subclade in clade B.6. We designed high-resolution melting (HRM) primers for the screening of 109 canSNPs divided in seven orders of discrimination for the molecular epidemiology analysis and tracking of Francisella tularensis subsp. holarctica in Western Europe.

Phylogeography and Genetic Diversity of Francisella tularensis subsp. holarctica in France (1947-2018).

In France, tularemia is caused by Francisella tularensis subsp. holarctica and is a sporadic disease affecting mainly wildlife animals and humans. F. tularensis species presents low genetic diversity that remains poorly described in France, as only a few genomes of isolates from the country are available so far. The objective of this study was to characterize the genetic diversity of F. tularensis in France and describe the phylogenetic distribution of isolates through whole-genome sequencing and molecular typing. Whole genomes of 350 strains of human or animal origin, collected from 1947 to 2018 in France and neighboring countries, were sequenced. A preliminary classification using the established canonical single nucleotide polymorphism (canSNP) nomenclature was performed. All isolates from France (except four) belonged to clade B.44, previously described in Western Europe. To increase the resolution power, a whole-genome SNP analysis was carried out. We were able to accurately reconstruct the population structure according to the global phylogenetic framework, and highlight numerous novel subclades. Whole-genome SNP analysis identified 87 new canSNPs specific to these subclades, among which 82 belonged to clade B.44. Identifying genomic features that are specific to sublineages is highly relevant in epidemiology and public health. We highlighted a large number of clusters among a single clade (B.44), which shows for the first time some genetic diversity among F. tularensis isolates from France, and the star phylogeny observed in clade B.44-subclades revealed that F. tularensis biodiversity in the country is relatively recent and resulted from clonal expansion of a single population. No association between clades and hosts or clinical forms of the disease was detected, but spatiotemporal clusters were identified for the first time in France. This is consistent with the hypothesis of persistence of F. tularensis strains found in Western Europe in the environment, associated with slow replication rates. Moreover, the presence of identical genotypes across long periods of time, and across long distances, supports this hypothesis but also suggests long-distance dispersal of the bacterium.

First Draft Genome Sequences of Three Strains of Francisella tularensis subsp. holarctica, Isolated from Hares and a Tick in France.

Here, we report the complete genome sequences of three strains of Francisella tularensis subsp. holarctica (11-789-5S, 11-935-13S, and 11-930-9S), isolated from brown hares and a tick during a tularemia outbreak in France, where tularemia is endemic.

Spatio-temporal dynamics of tularemia in French wildlife: 2002-2013.

Tularemia, caused by Francisella tularensis, is endemic in France. The surveillance of this disease in wildlife is operated by the SAGIR Network and by the National Reference Laboratory for Tularemia. Wild animals found dead or dying collected by the SAGIR network are necropsied and when tularemia is suspected culture and/or PCR are performed to confirm the diagnosis. The aim of this study was to present the results of tularemia surveillance in wildlife and to investigate the spatial and temporal pattern of tularemia observed between the 2002-2003 and 2012-2013 hunting seasons in French wildlife. Fourty-one to 121 cases were collected each hunting season for a total of 693 confirmed cases and 46 additional suspected cases. The main species affected was the European Brown Hare (Lepus europaeus) but 4 rabbits (Oryctolagus cuniculus), 2 roe deer (Capreolus capreolus) and one wild boar (Sus scrofa) were also found positive. The Standard Mortality Ratio was mapped and Kulldorff's retrospective space-time scan statistic was implemented and allowed the detection of several clusters: the nationwide outbreak of 2007-2008; 2 clusters in northern and central-western France in high hare-abundance areas and another in North-eastern France where the abundance of hares is low. Our results confirm the usefulness of brown hare as a sentinel of environmental risk.

Genotyping of French Bacillus anthracis strains based on 31-loci multi locus VNTR analysis: epidemiology, marker evaluation, and update of the internet genotype database.

BACKGROUND: Bacillus anthracis is known to have low genetic variability. In spite of this lack of diversity, multiple-locus variable-number tandem repeat (VNTR) analysis (MLVA) and single nucleotide polymorphisms (SNPs) including the canonical SNPs assay (canSNP) have proved to be highly effective to differentiate strains. Five different MLVA schemes based on a collection of 31 VNTR loci (MLVA8, MLVA15, MLVA20, MLVA25 and MLVA31) with increased resolving power have been described. RESULTS: MLVA31 was applied to characterize the French National Reference Laboratory collection. The total collection of 130 strains is resolved in 35 genotypes. The 119 veterinary and environmental strains collection in France were resolved into 26 genotypes belonging to three canSNP lineages and four MLVA clonal complexes (CCs) with particular geographical clustering. A subset of seven loci (MLVA7) is proposed to constitute a first line assay. The loci are compatible with moderate resolution equipment such as agarose gel electrophoresis and show a good congruence value with MLVA31. The associated MLVA and SNP data was imported together with published genotyping data by taking advantage of major enhancements to the MLVAbank software and web site. CONCLUSIONS: The present report provides a wide coverage of the genetic diversity of naturally occurring B. anthracis strains in France as can be revealed by MLVA. The data obtained suggests that once such coverage is achieved, it becomes possible to devise optimized first-line MLVA assays comprising a sufficiently low number of loci to be typed either in one multiplex PCR or on agarose gels. Such a selection of seven loci is proposed here, and future similar investigations in additional countries will indicate to which extend the same selection can be used worldwide as a common minimum set. It is hoped that this approach will contribute to an efficient and low-cost routine surveillance of important pathogens for biosecurity such as B. anthracis.

Use of high-resolution melting and melting temperature-shift assays for specific detection and identification of Bacillus anthracis based on single nucleotide discrimination.

Single nucleotide polymorphisms (SNPs) are important diagnostic markers for the detection and differentiation of Bacillus anthracis. High-Resolution Melting (HRM) and Melting Temperature (Tm)-shift methods are two approaches that enable SNP detection without the need for expensive labeled probes. We evaluated the potential diagnostic capability of those methods to discriminate B. anthracis from the other members of the B. cereus group. Two assays targeting B. anthracis-specific SNPs in the plcR and gyrA genes were designed for each method and used to genotype a panel of 155 Bacilli strains. All B. anthracis isolates (n=65) were correctly and unambiguously identified. Assays also proved to be appropriate for the direct genotyping of biological samples. They could reliably detect B. anthracis in contaminated organs containing as little as 10(3)CFU/ml, corresponding to a few genome equivalents per reaction. The HRM and Tm-shift applications described here represent valuable tools for specific identification of B. anthracis at reduced cost.