Molecular diversity of Bacillus anthracis in the Netherlands: Investigating the relationship to the worldwide population using whole-genome SNP discovery

Bacillus anthracis in South Africa, 1975-2013: are some lineages vanishing?

The anthrax-causing bacterium Bacillus anthracis comprises the genetic clades A, B, and C. In the northernmost part (Pafuri) of Kruger National Park (KNP), South Africa, both the common A and rare B strains clades occur. The B clade strains were reported to be dominant in Pafuri before 1991, while A clade strains occurred towards the central parts of KNP. The prevalence of B clade strains is currently much lower as only A clade strains have been isolated from 1992 onwards in KNP. In this study 319 B. anthracis strains were characterized with 31-loci multiple-locus variable-number tandem repeat analysis (MLVA-31). B clade strains from soil (n = 9) and a Tragelaphus strepsiceros carcass (n = 1) were further characterised by whole genome sequencing and compared to publicly available genomes. The KNP strains clustered in the B clade before 1991 into two dominant genotypes. South African strains cluster into a dominant genotype A.Br.005/006 consisting of KNP as well as the other anthrax endemic region, Northern Cape Province (NCP), South Africa. A few A.Br.001/002 strains from both endemic areas were also identified. Subclade A.Br.101 belonging to the A.Br.Aust94 lineage was reported in the NCP. The B-clade strains seems to be vanishing, while outbreaks in South Africa are caused mainly by the A.Br.005/006 genotypes as well as a few minor clades such as A.Br.001/002 and A.Br.101 present in NCP. This work confirmed the existence of the rare and vanishing B-clade strains that group in B.Br.001 branch with KrugerB and A0991 KNP strains.

Population genomics of Bacillus anthracis from an anthrax hyperendemic area reveals transmission processes across spatial scales and unexpected within-host diversity

Genomic sequencing has revolutionized our understanding of bacterial disease epidemiology, but remains underutilized for zoonotic pathogens in remote endemic settings. Anthrax, caused by the spore-forming bacterium Bacillus anthracis, remains a threat to human and animal health and rural livelihoods in low- and middle-income countries. While the global genomic diversity of B. anthracis has been well-characterized, there is limited information on how its populations are genetically structured at the scale at which transmission occurs, critical for understanding the pathogen's evolution and transmission dynamics. Using a uniquely rich dataset, we quantified genome-wide SNPs among 73 B. anthracis isolates derived from 33 livestock carcasses sampled over 1 year throughout the Ngorongoro Conservation Area, Tanzania, a region hyperendemic for anthrax. Genome-wide SNPs distinguished 22 unique B. anthracis genotypes (i.e. SNP profiles) within the study area. However, phylogeographical structure was lacking, as identical SNP profiles were found throughout the study area, likely the result of the long and variable periods of spore dormancy and long-distance livestock movements. Significantly, divergent genotypes were obtained from spatio-temporally linked cases and even individual carcasses. The high number of SNPs distinguishing isolates from the same host is unlikely to have arisen during infection, as supported by our simulation models. This points to an unexpectedly wide transmission bottleneck for B. anthracis, with an inoculum comprising multiple variants being the norm. Our work highlights that inferring transmission patterns of B. anthracis from genomic data will require analytical approaches that account for extended and variable environmental persistence, as well as co-infection.

Molecular genotyping of 15 B. anthracis strains isolated in Eastern Siberia and Far East

Bacillus anthracis is a pathogenic bacterium, which causes anthrax disease. The ability of this bacterium to form spores, which can be preserved in soil for decades and cause outbreaks later on, makes this pathogen a serious problem for veterinary and health services of many countries. Siberia is one of the most anthrax-influenced regions of Russia. In this research we report on the results of genotyping based on whole genome SNP analysis of 15 strains, isolated on the territory of Eastern Siberia and the Far East in 1956-2018. In this research, we sequenced 15 genomes of B. anthracis strains isolated from infected humans and animals, and from soil samples from the territory of Eastern Siberia and the Far East in the period from 1956 to 2018. We used genomic sequences obtained in this study and 219 B. anthracis genomes available in the international GenBank database to perform a comparative analysis. As a result we detected 6400 chromosomal SNPs which allowed to differentiate the studied strains. We built phylogenetic reconstruction of the global B. anthracis population based on the detected SNPs using the Maximum Likelihood Method and described genetic diversity of the strains isolated on the territory of Eastern Siberia and the Far East. Strains, isolated on this territory from 1956 to 2018 belong to 5 different genetic groups: "Ames", "STI", "Tsiankovskii", "Siberia" and "Asia". The greatest diversity of the strains is registered for two regions of the southern part of Eastern Siberia - Tyva and Buryatia. This research expands current understanding of genetic diversity of B. anthracis strains circulating on the territory of Russia.

A classification framework for Bacillus anthracis defined by global genomic structure

Bacillus anthracis, the causative agent of anthrax, is a considerable global health threat affecting wildlife, livestock, and the general public. In this study, whole-genome sequence analysis of over 350 B. anthracis isolates was used to establish a new high-resolution global genotyping framework that is both biogeographically informative and compatible with multiple genomic assays. The data presented in this study shed new light on the diverse global dissemination of this species and indicate that many lineages may be uniquely suited to the geographic regions in which they are found. In addition, we demonstrate that plasmid genomic structure for this species is largely consistent with chromosomal population structure, suggesting vertical inheritance in this bacterium has contributed to its evolutionary persistence. This classification methodology is the first based on population genomic structure for this species and has potential use for local and broader institutions seeking to understand both disease outbreak origins and recent introductions. In addition, we provide access to a newly developed genotyping script as well as the full whole-genome sequence analyses output for this study, allowing future studies to rapidly employ and append their data in the context of this global collection. This framework may act as a powerful tool for public health agencies, wildlife disease laboratories, and researchers seeking to utilize and expand this classification scheme for further investigations into B. anthracis evolution.

Epidemiological and comparative genomic analysis of Bacillus anthracis isolated from northern Vietnam

To understand the epidemiological and genetic background of anthrax cases occurring in Vietnam from 2011 to 2015, we surveilled and genetically analyzed Bacillus anthracis isolated in the north of the country. Epidemiological surveillance showed that most human cutaneous anthrax cases occurred in association with animal dissection. Whole-genome sequences were obtained from six B. anthracis strains from human patients with cutaneous anthrax in the endemic area. Comparative genomic analysis showed that the genetic homogeneity among Vietnamese B. anthracis strains was very high. All Vietnamese B. anthracis strains belonged to the canSNP lineage of A.Br.011/009, which mostly consists of strains of the trans-Eurasian (TEA) group, including the most closely related strain, Carbosap. To clarify the genetic diversity of Vietnamese strains and strains belonging to A.Br.011/009 and A.Br.008/011 canSNP lineages, we applied a reference genome-based single-nucleotide polymorphism (SNP) and gene-by-gene genomic analysis (whole-genome MLST) strategy. The phylogeny from core genome SNPs revealed that the Vietnamese strains were positioned close to each other; moreover, several SNPs specific to Vietnamese B. anthracis were identified. Whole-genome MLST analysis revealed the differences in the number of SNPs between Vietnamese strains, which could enable discrimination at the strain level.

Phylogenomic structure of Bacillus anthracis isolates in the Northern Cape Province, South Africa revealed novel single nucleotide polymorphisms

Bacillus anthracis, the aetiological agent of anthrax, is regarded as a highly monomorphic pathogen that presents a low genetic diversity using standard molecular techniques. Whole genome sequencing and single nucleotide polymorphisms (SNPs) are definitive signatures for subtyping of B. anthracis. Here we employed whole genome single nucleotide polymorphism (wgSNP) analysis to investigate the genetic diversity of B. anthracis in the historically endemic region of Northern Cape Province (NCP), South Africa. Twenty-six isolates from anthrax outbreaks that occurred between 1998 and 2008/9 in NCP as well as from Namibia-South Africa Transfontier Conservation area and Botswana were compared to global B. anthracis genomes. Most NCP B. anthracis strains (n = 22) clustered in the A.Br.003/004 (A.Br.101) branch and are closely related to the Zimbabwe and Mozambique strains (A.Br.102 branch). A total of 4923 parsimony informative-SNPs accurately established the A.Br.003/004 phylogenetic relationships of the NCP isolates into two distinct sub-clades and SNP markers designated as A.Br.172 and A.Br.173 were developed. Other NCP strains (n = 2) grouped in the A.Br.001/002 (Sterne) branch while strains (n = 2) from the Namibia-South Africa Transfontier Conservation area and Botswana clustered in A.Br.005/006 (Ancient A) branch. The sequenced B. anthracis strains (A0094, A0096 and A0097) that clustered in the A.Br.064 (V770) clade were isolated from Vaalbos National Park and similar strains have not been isolated. The B. anthracis A0088 strain cluster with the NCP strains in the A.Br.003/004 (A.Br.172) SNP branch which has been isolated in NCP, South Africa. This study highlights the phylogenetic structure of NCP B. anthracis strains with distinctive SNP branches important for forensic tracing and novel SNP discovery purposes. The sequenced strains will serve as a means to further trace the dissemination of B. anthracis outbreaks in NCP, South Africa, and on the continent, as well as for forensic tracking on a global scale.

Comparative genomics and phylogenetic analysis of Bacillus anthracis strains isolated from domestic animals in Japan

Anthrax, caused by Bacillus anthracis, is a severe zoonosis with a great impact on both human and animal health. In the present study, we identified the phylogenetic relationships among 16 Japanese strains of B. anthracis, including eight bovine strains, two equine strains, five swine strains, and one former vaccine strain, using in silico canonical single nucleotide polymorphism (canSNP) and core genome SNP analyses. The results of our in silico canSNP analysis suggest that these 16 Japanese strains could be divided into four lineages: i) one equine strain in A.Br.Ames, ii) one equine and six bovine strains in A.Br.001/002, iii) five swine and one bovine strain in A.Br.Aust94, and iv) one bovine and one vaccine strain in A.Br.008/011. A comparison with non-Japanese B. anthracis strains revealed a total of 3787 SNPs identified from the whole genome sequences of the Japanese strains; these SNP data were subjected to a phylogenetic analysis using the maximum parsimony (MP) method. Our core genome SNP analysis was also able to detect differences of a few chromosomal SNPs across clonal strains from the same cases that had different storage and passage histories. Additionally, our whole genome SNP analysis clearly indicated that the Japanese swine anthrax cases of 1982 were caused by at least three independent strains; however, their phylogeny revealed no clear relationship with swine strains from other countries. The bovine strain belonging to the A.Br.008/011 lineage differed from a former Japanese vaccine strain by only 12 SNPs. Together with the phylogenic results and epidemiological circumstances, the diversity of strains reveals that the B. anthracis available in Japan probably resulted from multiple relatively recent import events, rather than reflecting the persistence of a more ancient ecologically established group.

The identification of novel single nucleotide polymorphisms to assist in mapping the spread of Bacillus anthracis across the Southern Caucasus

Anthrax is common as a zoonotic disease in the southern Caucasus area including parts of Turkey and Georgia. In this region, population genetics of the etiological agent Bacillus anthracis comprises, where known, the major canonical single nucleotide polymorphism (canSNP) groups A.Br.Aust94 and A.Br.008/009 of the pathogen's global phylogeny, respectively. Previously, isolates of B. anthracis from Turkey have been genotyped predominantly by multi locus variable number of tandem repeat analysis (MLVA) or canSNP typing. While whole genome sequencing is the future gold standard, it is currently still costly. For that reason we were interested in identifying novel SNPs which could assist in further distinguishing closely related isolates using low cost assay platforms. In this study we sequenced the genomes of seven B. anthracis strains collected from the Kars province of Eastern Anatolia in Turkey and discovered new SNPs which allowed us to assign these and other geographically related strains to three novel branches of the major A-branch canSNP-group (A.Br.) Aust94. These new branches were named Kafkas-Geo 1-3 and comprised isolates from the Kars region and the neighboring republic of Georgia suggesting a common ancestry. The novel SNPs identified in this study connect the population genetics of B. anthracis in the South Caucasus and Turkey and will likely assist efforts to map the spread of the pathogen across this region.

Pathogenicity, population genetics and dissemination of Bacillus anthracis

Bacillus anthracis, the etiological agent of anthrax, procures its particular virulence by a capsule and two AB type toxins: the lethal factor LF and the edema factor EF. These toxins primarily disable immune cells. Both toxins are translocated to the host cell by the adhesin-internalin subunit called protective antigen PA. PA enables LF to reach intra-luminal vesicles, where it remains active for long periods. Subsequently, LF translocates to non-infected cells, leading to inefficient late therapy of anthrax. B. anthracis undergoes slow evolution because it alternates between vegetative and long spore phases. Full genome sequence analysis of a large number of worldwide strains resulted in a robust evolutionary reconstruction of this bacterium, showing that B. anthracis is split in three main clades: A, B and C. Clade A efficiently disseminated worldwide underpinned by human activities including heavy intercontinental trade of goat and sheep hair. Subclade A.Br.WNA, which is widespread in the Northern American continent, is estimated to have split from clade A reaching the Northern American continent in the late Pleistocene epoch via the former Bering Land Bridge and further spread from Northwest southwards. An alternative hypothesis is that subclade A.Br.WNA. evolved from clade A.Br.TEA tracing it back to strains from Northern France that were assumingly dispatched by European explorers that settled along the St. Lawrence River. Clade B established mostly in Europe along the alpine axis where it evolved in association with local cattle breeds and hence displays specific geographic subclusters. Sequencing technologies are also used for forensic applications to trace unintended or criminal acts of release of B. anthracis. Under natural conditions, B. anthracis generally affects domesticated and wild ruminants in arid ecosystems. The more recently discovered B. cereus biovar anthracis spreads in tropical forests, where it threatens particularly endangered primate populations.

Comparative genomics of Bacillus anthracis from the wool industry highlights polymorphisms of lineage A.Br.Vollum

BACKGROUND

With the advent of affordable next-generation sequencing (NGS) technologies, major progress has been made in the understanding of the population structure and evolution of the B. anthracis species. Here we report the use of whole genome sequencing and computer-based comparative analyses to characterize six strains belonging to the A.Br.Vollum lineage. These strains were isolated in Switzerland, in 1981, during iterative cases of anthrax involving workers in a textile plant processing cashmere wool from the Indian subcontinent.

RESULTS

We took advantage of the hundreds of currently available B. anthracis genomes in public databases, to investigate the genetic diversity existing within the A.Br.Vollum lineage and to position the six Swiss isolates into the worldwide B. anthracis phylogeny. Thirty additional genomes related to the A.Br.Vollum group were identified by whole-genome single nucleotide polymorphism (SNP) analysis, including two strains forming a new evolutionary branch at the basis of the A.Br.Vollum lineage. This new phylogenetic lineage (termed A.Br.H9401) splits off the branch leading to the A.Br.Vollum group soon after its divergence to the other lineages of the major A clade (i.e. 6 SNPs). The available dataset of A.Br.Vollum genomes were resolved into 2 distinct groups. Isolates from the Swiss wool processing facility clustered together with two strains from Pakistan and one strain of unknown origin isolated from yarn. They were clearly differentiated (69 SNPs) from the twenty-five other A.Br.Vollum strains located on the branch leading to the terminal reference strain A0488 of the lineage. Novel analytic assays specific to these new subgroups were developed for the purpose of rapid molecular epidemiology.

CONCLUSIONS

Whole genome SNP surveys greatly expand upon our knowledge on the sub-structure of the A.Br.Vollum lineage. Possible origin and route of spread of this lineage worldwide are discussed.

Unexpected genomic relationships between Bacillus anthracis strains from Bangladesh and Central Europe

The zoonosis anthrax caused by the bacterium Bacillus anthracis has a broad geographical distribution. Active enzootic areas are typically located away from central and northern Europe where cases of the disease occur only sporadically and in limited numbers. In contrast, a few out of the 64 districts of Bangladesh are hyper-endemic for anthrax and there the disease causes major losses in live-stock. In this study we genotyped eight strains of B. anthracis collected from the districts of Sirajganj and Tangail in 2013. All these strains belonged to canSNP group A.Br.001/002 Sterne differing only in a few of 31 tandem-repeat (MLVA)-markers. Whole genome sequences were obtained from five of these strains and compared with genomic information of B. anthracis strains originating from various geographical locations. Characteristic signatures were detected defining two "Bangladesh" clusters potentially useful for rapid molecular epidemiology. From this data high-resolution PCR assays were developed and subsequently tested on additional isolates from Bangladesh and Central Europe. Remarkably, this comparative genomic analysis focusing on SNP-discovery revealed a close genetic relationship between these strains from Bangladesh and historic strains collected between 1991 and 2008 in The Netherlands and Germany, respectively. Possible explanations for these phylogenetic relationships are discussed.

Whole-genome sequencing investigation of animal-skin-drum-associated UK anthrax cases reveals evidence of mixed populations and relatedness to a US case

There have been two anthrax cases affecting people that played and/or made animal-skin drums in the UK during the last 10 years, with single fatal occurrences in Scotland in 2006 and London in 2008. Investigations by the Health Protection Agency (now Public Health England) employing multi-locus-variable number tandem repeat analysis had previously linked the clinical cases to spores associated with animal skins and drums the patients had been in contact with. In this study, whole-genome sequencing of 23 Bacillus anthracis isolates harvested during the investigations was performed. High-quality draft assemblies of these genomes provided greater characterization of the B. anthracis strains present and placed them all upon a new branch of the global phylogeny. Although closely related, the clinical isolates from the two events, and another isolated from a drum-skin-associated case in New York in 2006, were distinct from each other. Multiple distinct genotypes were found during both investigations, implying either multiple contamination events or a single heterogeneous contamination. One environmental isolate from the Scottish incident was more closely related to London isolates than to the other Scottish isolates. As B. anthracis of this subgroup was present at both geographically and temporally distinct events, it may be more widespread at the source of contamination. All isolates were distinct from currently characterized West African strains, despite this being the likely origin of the drums and hides, therefore adding to our knowledge of B. anthracis diversity in the region.