In silico analysis of genomes of Bacillus anthracis strains belonging to major genetic lineages

Full Text

Introduction

The global population of Bacillus anthracis is represented by three major genetic lineages (A, B, and C) including 14 canonical (canSNP) groups. Pathogen strains are distributed among them nonuniformly; representatives of lineage A prevail (about 90%), lineage B accounts for approximately 10%, while lineage C includes only 3 strains (less than 1%) [1, 2]. The distinct clade in the Bacillus cereus sensu lato group including B. anthracis is represented by Bacillus cereus biovar anthracis strains that can cause anthrax-like infection [3, 4]. The factors causing the nonuniform distribution of B. anthracis strains in different genetic lineages are still unknown. The question why strains of lineage A have become widely dispersed throughout the world, while strains of lineage B are limited in number and do not have a wide global distribution remains unanswered.

This nonuniformity can be explained by adaptive genetic differences, which have an impact on survival and reproduction in the environment or in the host body.

The reproduction efficiency in the host body depends on adaptive genetic differences of pathogenicity factors. The pathogenicity of B. anthracis depends on the main factors: Two binary exotoxins, lethal and edema factors, and the poly-γ-glutamic acid capsule [2]. Exotoxin components, lethal and edema factors, and the protective antigen (PA) are encoded by lef, cya and pagA genes located on the pXO1 plasmid [5]. The capBCDAE operon for capsule production is encoded by the pXO2 plasmid [2]. The loss of either of the two plasmids results in avirulence of strains. At the same time, strains carrying both plasmids significantly differ by their virulence as demonstrated by laboratory animals. The Carbosap vaccine strain attenuation not associated with the loss of pXO1 and pXO2 plasmids results from chromosomal deletions containing more than 50 genes, which have a known or proposed function that could be linked to virulence [6]. The detected differences in the virulence of strains producing a specific toxin and a typical capsule suggest the existence of additional pathogenicity factors. The role of such factors is claimed by many products of the anthrax microbe:

• the GerXC protein encoded by the gerXC plasmid gene and required for spore germination in vivo [7];
• phospholipase C (PlC gene) [8];
• nitric oxide synthase (nos gene) [9];
• bifunctional lysyl-phosphatidylglycerol flippase/synthase (mprF gene) [10];
• metalloprotease of the enhancin family (GBAA_RS16775 locus), homolog of which was first described in baculoviruses [11, 12];
• immune inhibitor A metalloprotease (inhA gene) [12];
• quorum-sensing signal molecule autoinducer LuxS (luxS gene) [13];
• anthorolysin O (ALO) (cholesterol-dependent cytolysin) (alo gene) [14];
• enterotoxin FM (EntFM) (entFM gene) [13].

Some B. anthracis strains have a set of phenotypic characteristics, which distinguish them from typical virulent strains; among them, the inability to grow in a minimal synthetic medium without tryptophan, and reduced virulence in rabbits. The genetic factors underlying these changes have not been identified [16].

There are publications reporting the effect of amino acid substitutions in the lethal factor protein on its catalytic activity and binding to PA [17–19]. The data were obtained during tests where mutations were introduced into the described genes to assess their impact on the virulence of mutant strains compared to wild-type strains. Allelic polymorphism for the PA gene that is represented by 6 allelic types in natural wild-type strains [20] was also described for other genes of pathogenicity factors [21–23]. However, the variability of chromosomal genes encoding products, for which the impact on the virulence of B. anthracis has been identified or assumed, remains unexplored. Quantitative and qualitative distinctive characteristics of polymorphisms inherent in genomes of strains belonging to certain genetic lineages have not been identified.

The molecular typing of B. anthracis using multi-virulence-locus sequence typing (MVLST) did not include chromosomal genes.

The urgent need for this study was dictated by the absence of the data that would shed a new light on the pathogenesis of anthrax infection, help identify potential targets for development of new products for treatment and prevention of this infection, contribute to the knowledge of the evolution of the anthrax-causing pathogen and methods of its molecular typing.

The aim of the study was to explore characteristics of genomes of different genetic lineages, which have a potential impact on the distribution of strains, by using the in silico analysis of a representative subset of B. anthracis strains.

Materials and methods

The study was performed using complete genomes of 49 diplasmid strains, including 19 B. anthracis isolates from the collection of pathogenic microorganisms from the Stavropol Research Anti-Plague Institute of Rospotrebnadzor and 30 isolates from the GenBank database, which belong to major genetic lineages A, B, C and 14 canSNP groups, as well as the genome of Bacillus cereus biovar anthracis strain CI.

GenBank¹ identifiers for genomes:

GCF_000008445.1, GCF_009831565.1, GCF_000167335.1, GCF_003063965.1, GCF_003064045.1, GCF_003860145.1, GCF_000793525.1, GCF_000832965.1, GCF_000310045.1, GCF_000167235.1, GCF_000534935.2, GCF_000258885.1, GCF_000278385.1, GCF_000832465.1, GCF_001273005.1, GCF_001273085.1, GCF_000167295.1, GCF_002896575.1, GCF_014249775.1, GCF_003227955.1, GCF_000831505.1, GCF_000832745.1, GCF_003064005.1, GCF_000008165.1, GCF_000583105.1, GCF_000833275.1, GCF_022221345.1, GCF_000743805.1, GCF_900014355.1, GCF_002356575.1, GCF_000143605.1.

The in silico analysis was performed using the genome of the B. anthracis Ames Ancestor strain (GenBank: NC_007530.2; NC_007322.2; NC_007323.3) as a reference genome. Polymorphisms were identified using BLAST, BLASTp, MEGA X, MAUVE, Tandem Repeat Finder programs. The alignment of pooled sequences of pathogenicity factor genes and the translation of nucleotide sequences of genes were performed using the MEGA X program. The whole-genome analysis of single nucleotide polymorphisms (SNPs) was performed using the Parsnp program from the Harvest Suite for multiple alignment of genomic sequences. The input data included genomes of 50 strains described above; they were aligned with the chromosomal nucleotide sequence of the reference genome of B. anthracis Ames Ancestor (GenBank: NC_007530.2) using Parsnp (parameters c -e -u -C 1000). The detected SNPs were extracted into a VCF file using HarvestTools v. 1.0. The edited file was used as an input file in HarvestTools to compile a FASTA file.

The phylogenetic tree was reconstructed by the neighbor-joining method in MEGA X in accordance with Tamura-Nei model; the bootstrapped confidence interval was based on 1,000 replicates; the goeBURST Full MST method and PHYLOViZ 2.0 program were used for identification of clonal complexes. The FigTree program was used for dendrogram visualization. The Hanter–Gaston discriminatory index was calculated in accordance with [24].

Results

Analysis of polymorphisms of the chromosomal region of the B. anthracis and Bacillus cereus biovar anthracis genome

The variability of the chromosomal region in B. anthracis and B. cereus biovar anthracis depended on SNPs, single-nucleotide repeats (SNR), tandem repeats, substitutions and indels (insertions/deletions). The main polymorphisms in chromosomal genomes were represented by SNPs (Table 1). Strains belonging to different genetic lineages had noticeable differences in the number of polymorphisms.

 

Table 1. Polymorphisms of chromosome genomes оf B. anthracis and B. cereus biovar anthracis strains

Major lineage	Strain	Quantity of polymorphisms comparing with B. anthracis strain Ames Ancestor
		SNP	SNR	tandem repeats	substitutions	indels	total
A	Australia 94	411	142	23	8	64	648
A	Vollum	609	233	32	12	68	954
B	SVA11	1693	418	73	16	109	2309
C	2002013094	2381	576	99	134	414	3604
B. cereus biovar anthracis	CI	76 714	1075	188	7857	1783	87 617

 

Chromosomal marker (specific) SNPs were identified for B. anthracis major genetic lineages (Table 2).

 

Table 2. Marker SNPs for chromosome genomes of B. anthracis

Major lineage	Quantity of marker SNP	Ratio marker SNPs/genome	SNP localization
			gene	intergene space
A	180	4,73	152	28
B	183	18,3	141	42
C	594	594	Not tested	Not tested

 

Strains of lineages B and C demonstrated significant differences in the relative amounts of marker SNPs compared to lineage A.

Most of the marker SNPs from lineages A and B were primarily located in housekeeping genes, out of which 4 genes were associated with sporulation and spore germination.

Analysis of SNPs of pathogenicity factor genes

The analysis included 19 variable genes encoding products associated with pathogenicity (Table 3).

 

Table 3. SNPs in pathogenicity factors genes of B. anthracis strains of different lineages

Gene	Ratio quantity of SNPs/genome in strains of lineages
	A (n = 38)		B (n = 10)		C (n = 1)		B. cereus biovar anthracis (n = 1)
	total	nonsynonimic	total	nonsynonimic	total	nonsynonimic	total	nonsynonimic
cya	7/0,18	4/0,1	3/0,3	2/0,2	6	4	6	4
lef	2/0,05	2/0,05	2/0,2	2/0,2	4	3	10	8
pagA	4/0,1	2/0,05	3/0,3	2/0,2	3	2	4	3
atxA	1/0,02	0	0	0	0	0	1	1
capA	3/0,07	2/0,05	1/0,1	1/0,1	1	1	1	0
capC	2/0,05	0	0	0	1	1	1	0
capD	3/0,07	3/0,07	2/0,2	0	3	3	4	4
acpA	1/0,02	0	1/0,1	1/0,1	1	1	3	2
ger XC	1/0,02	1/0,02	2/0,2	2/0,2	1	1	1	1
mprF	2/0,05	1/0,02	2/0,2	2/0,2	2	1	23	1
entFM	2/0,05	2/0,05	1/0,1	1/0,1	0	0	17	5
GBAA_RS16775	1/0,02	1/0,02	2/0,2	2/0,2	0	0	132	31
plC	0	0	1/0,1	1/0,1	1/0,1	1/0,1	9	3
alo	3/0,07	3/0,073	0	0	0	0	57	23
nos	0	0	2/0,2	2/0,2	0	0	32	9
luxS	0	0	0	0	1	1	8	0
trpA	0	0	3/0,3	1/0,1	0	0	18	5
trpD	0	0	1/0,1	1/0,1	1	0	15	4
GBAA_RS06415 (trpG)	0	0	1/0,1	1/0,1	1	1	14	5

 

The gene variability depended on the presence of SNPs, VNTRs and INDELs. The largest number of all SNPs and nonsynonymous SNPs per one genome was found in the B. cereus biovar anthracis strain followed by B. anthracis strains of lineage C, lineage B, lineage A. The deletion in the trpA gene resulted in the formation of pseudogenes and the absence of functional proteins in some strains of lineage B. In the B. cereus biovar anthracis strain, the frameshift mutation caused the formation of the pseudogene and absence of the Mobl relaxase (capsule-production plasmid plC-XO2). There are noticeable differences in the number of SNPs in different genes and in strains of different genetic lineages.

Analysis of the polymorphism of pathogenicity factor proteins in strains of different genetic lineages

The lethal factor. In the lethal factor protein, substitutions E709G and E681K are located within domain 4, which contains the catalytic center, at the distance of 10 and 37 amino acids, respectively, from the zinc binding site; substitutions A299T, L298M and R543Q are located in domain 2, substitutions E66K and V246I — in domain 1.

The edema factor. All strains of lineage B had substitutions D180G and 318T. The strain of lineage C and B. cereus biovar anthracis strain CI have substitutions K278E, I318T and N789K. The B. cereus var. anthracis strain CI has substitution V694A. Substitutions D180G and K278E are located within the PAB domain, I318T — in the CA segment of the ACD core domain, V694A and N789K — in the helical region of the edema factor.

The protective antigen. Strains of lineage A were found to have substitutions A600V and P565S; strains of lineage B had substitutions I433V and A600V; the strain of lineage C and B. cereus biovar anthracis strain CI had substitutions S66P and A600V. B. cereus biovar anthracis strain CI had substitution S290I. Substitution S66P was located in domain 1, substitution I433V — in domain 2, substitution P565S – in domain 3; substitution A600V — in domain 4 in the receptor-binding region (L595–T735).

The comparative analysis was performed using PA genes and proteins in all live spore vaccines based on strains Carbosap, 34F2_Sterne, A16R, Tsiankovskii-1, STI-1, 55VNIIViM, 228/8 and Brazilian vaccinal as well as in chemical vaccines based on PA of the avirulent V770-NP-1R strain (U.S. vaccines AVA (or BioThrax) and AV7909) and based on PA of vaccine strain 34F2 Sterne (vaccine AVP; United Kingdom). All the strains belonged to lineage A. Strains V770-NP-1R, Carbosap and all the strains of the Russian origin (Tsiankovskii-1, STI-1, 55VNIIViM, 228/8) had substitutions in gene C→T at the position 195 and 1799; in addition, the Tsiankovskii-1 strain had substitution 981 A→T. The same strains had substitution A600V in PA domain 4 in the receptor-binding region. No substitutions were detected in PA gene and protein in strains 34F2 Sterne, Brazilian vaccinal and A16R.

Capsular polyglutamate synthetase CapA: strains of lineage A had substitution Q399K; strains of lineage B had substitution T345A; lineage C strains had substitution V156L.

Capsule biosynthesis protein CapC: substitution T80M in strains of lineage C and B. cereus strains.

CapD gamma-glutamyl transferase: strains of lineage A had substitutions I4M, V266I and S381E; strains of lineage B did not have any substitutions; the strain of lineage C and B. cereus biovar anthracis strain CI had substitutions H70Y, K223E and F379I; B. cereus biovar anthracis strain CI had substitution G499D. Substitution H70Y was located in the L strand, substitution F379I — in the S strand.

Capsule synthesis transcriptional regulator AcpA: strains of lineages B and C and B. cereus biovar anthracis strain CI had substitution E285K; B. cereus also had substitution Y354H.

Anthrax toxin trans-activating transcriptional regulator AtxA: substitution I188N in B. cereus biovar anthracis strain CI.

Anthrolysin O (ALO): strains of the A.Br.Vollum group had substitution S422F; strains of the A.Br.Aust94 group, which were isolated in South Africa, had substitution N221T; ; strains of the A.Br.005/006 group had substitution V416G.

Metalloprotease of the enhancin family: strains of lineage A had substitution P631S; strains of lineage B — substitution L139F; strains of lineage C — substitution D444E, the protein of B. cereus biovar anthracis strain CI had 29 substitutions.

GerXC spore germination protein: lineages A and B had substitution H29R; lineage B — substitution T35I, lineage C and B. cereus biovar anthracis — substitution E219G.

Protein production autoinducer-2, LuxS: lineage C had substitution D111G.

Bifunctional lysyl-phosphatidylglycerol flippase/synthetase MprF: lineage A had substitution H631R; lineage B — substitution L289F; lineages B and C as well as B. cereus biovar anthracis — substitution V424I.

Nitric oxide synthase NOS: lineage B — substitutions Q288H and I348F; B. cereus biovar anthracis — 9 substitutions.

Phospholipase C: strains of lineages B and C as well as B. cereus biovar anthracis had substitution H194Y; B. cereus biovar anthracis — had substitutions N20S and A59V.

Tryptophan synthase alpha subunit: lineage B — substitutions T222K and a non-functional protein in some strains of this lineage due to the deletion and formation of a pseudogene; B. cereus biovar anthracis — 5 substitutions.

Anthranilate phosphoribosyl transferase TrpD: lineage B — substitution N300S; B. cereus biovar anthracis — 4 substitutions.

Aminodeoxychorismate/anthranilate synthase component II, TrpG: Lineages B and C — substitutionsN300S; B. cereus biovar anthracis — 5 substitutions. H70Y is located in the L strand, F379I — in the S strand.

The obtained data on substitutions in plasmid genes and proteins of pathogenicity factors correlate with the data reported in earlier [22, 23]. Substitutions in genes of additional pathogenicity factors having chromosomal localization have been described by us for the first time.

The largest number of significant substitutions was identified in genes of enhancin and anthorolysin O in B. cereus biovar anthracis strain CI. In total, 19 pathogenicity factor genes had 15 significant substitutions in 38 strains of lineage A, 20 substitutions in 10 strains of lineage B, 20 substitutions in 1 strain of lineage C, 102 substitutions in B. cereus biovar anthracis strain CI. The consistency of pattern is observed: Strains from major lineages B and C, let alone, the B. cereus biovar anthracis strain, had significantly larger numbers of amino acid substitutions in pathogenicity factor proteins, which can change their functional activity, than strains from lineage A.

Sequence typing based on SNPs in pathogenicity factors genes

A total of 409 phylogenetically significant SNPs and 33 genotypes of pathogenicity factors were identified in 19 pathogenicity factor gene of 49 B. anthracis strains and 1 B. cereus biovar anthracis strain identified.

For the multi-virulence-locus sequence typing (MVLST), the Hunter–Gaston discriminatory index was 0.9633, being higher than the respective index for canSNP-typing (0.9056), coming close to the index for WGS-SNP typing (0.9869). Similar results were demonstrated by the comparative analysis of the MVLST and whole-genome SNP typing effectiveness [25]. Compared to canSNP and coreWGS-SNP typing, the MVLST typing based on SNPs of pathogenicity factor genes can be used for chromosomal and plasmid genes.

The dendrogram based on SNPs of pathogenicity factor genes also shows three universally recognized major genetic lineages: A, B, C of B. anthracis and the clade of the B. cereus biovar anthracis strain; the latter is basic for all the three major genetic B. anthracis lineages, while the clade of lineage C is basic for lineages B and C (Fig. 1). These findings support the idea of the evolution of B. anthracis from its predecessor B. cereus to lineage C and then to lineages B and C; they can also suggest that variability of pathogenicity factor genes can be seen as a driving force of the evolution.

 

Fig. 1. Comparison of the phylogenetic reconstruction of the results received using the multilocus sequence typing and coreWGS-SNP.

 

The phylogenetic relationships of strains, which were reconstructed using multilocus sequence typing, differed from those demonstrated by whole-genome SNP typing.

The canSNP group B.Br.001/002 and B.Br.Kruger clustering conformity was observed using multilocus sequence typing and whole-genome SNP typing.

The comparison of the dendrograms constructed with multilocus sequence typing and whole-genome SNP typing of the same strains (Fig. 1) showed that some strains belonging to other canSNP groups are clustered with strains from groups that do not conform to their canSNP-group affiliation.

Identification of clonal complexes using multilocus sequence typing

The identification of clonal complexes (CC) using multilocus sequence typing resulted in identification of 33 genotypes (GTs) comprising 5 clonal complexes for 48 B. anthracis strains of major lineages A and B as well as two distinct GTs for B. anthracis strains of lineage C and B. cereus bv anthracis (Fig. 2). CC1 is the most numerous complex consisting of 12 GTs of 20 strains.

 

Fig. 2. Clonal complexes of B. anthracis strains. The dendrogram was constructed using multilocus sequence typing and the goeBURST Full MST algorithm in the PHYLOViZ 2.0 program; numbers on the clades correspond to genetic distances.

 

CC1 includes GTs of 18 strains of 7 groups out of 14 canSNP groups from major genetic B. anthracis lineage A. CC2 consists of 2 GTs of the A.Br.008/011 group and serves as an intermediate link between CC1, CC4 having strains of lineage A and CC5having strains of lineage B. CC3 includes 7 GTs from A.Br.Aust94, A.Br.001/002 and A.Br.Ames groups. CC4 consists of 5 GTs of the A.Br.008/011 group of major lineage A. CC5 includes 7 GTs of 10 strains from all canSNP groups of major B. anthracis lineage B.

In the clonal complexes, the genetic distances between GTs range from 1 to 6 units. All B. anthracis strains of lineage A are separated from strains of lineage B by the distance of 17 units (from GT17 to GT43), from the strain of lineage C — by the distance of 24 units (between GT7 and GT49), from the B. cereus bv anthracis strain (GT50) — by the distance of 340 units from the B. anthracis strain of lineage C (GT49).

The analysis of the clonal complexes confirms that B. anthracis strains are classified into three major genetic lineages and are distributed by genotypes of pathogenicity factor genes, which do not demonstrate complete conformity to canonical SNP groups. The analysis confirms the evolution of pathogenicity factors from B. cereus biovar anthracis to B. anthracis of lineage C and then to lineages A and B, thus making it possible to identify genetic groups different from canonical SNP groups.

Discussion

The number of all types of polymorphisms in the chromosomal region of genomes in B. anthracis strains of lineages B and C was 1.6-13.4 times and in the B. cereus var. anthracis strain 5-150 times as large as their number in B. anthracis strains of lineage A. Especially significant differences were observed in the numbers of substitutions, SNPs, and indels in the genome of the B. cereus var. anthracis strain, which were 785.7, 150 and 27 times as large as their numbers in B. anthracis strains of lineage A. These differences can be explained by the conformity between the chromosomal region of the genome of B. cereus var. anthracis strains and the genome of representatives of the B. cereus sensu lato group, except for B. anthracis, while pCI-XO1 and pCI-XO2 plasmids do not differ significantly from pXO1 and pXO2 plasmids of B. anthracis [4].

The number of chromosomal specific marker SNPs per genome had an inverse relationship with the number of lineage strains; for the only strain of major lineage C, this number was 24 times as large as the number for 10 strains of lineage B and exceeded 170 times the number for 38 strains of lineage A. This phenomenon can result from a longer evolution history involving accumulation of mutations in genetic lineages B and C compared to lineage A. As SNPs were located primarily in housekeeping genes, there is a high probability that these mutations were significant and could affect the ecological adaptation of lineages B and C, consequently, their limited spread.

The variability of 19 pathogenicity factor genes, out of which 9 were encoded by pXO1 and pXO2 plasmids of B. anthracis or pCI-XO1 and pCI-XO2 of B. cereus biovar anthracis, and another 10 — by a chromosome, was manifested by the presence of SNPs and indels. It should be noted that no indels have been detected in B. anthracis strains from lineage A, except for the strain 2000031008, as opposed to other lineages. In strains of lineages B and C, insertion ATATAGATA in the acpA gene caused the insertion of 3 amino acids NID (asparagine-isoleucine-aspartic acid) in the capsule synthesis transcriptional regulator AcpA. This insertion, a tandem repeat unit, was described as a new VNTR-locus first in our publication [26] and later by other authors [23]. The deletion in the 107–124 bp region in the trpA gene turned it into a pseudogene, deprived the TrpA tryptophan synthase alpha subunit of 35–40 N-terminal amino acids and made the enzyme nonfunctional in most of the strains of lineage B, which we have studied. These effects could explain the dependence of these strains on tryptophan [16].

PA is the main immunogenic component of anthrax vaccines. All vaccine strains of live vaccines as well as strains, PA of which is used in chemical vaccines, belonged to major genetic lineage A. Vaccine strains of Russian origin, strains V770-NP-1R and Carbosap have the alanine-to-valine substitution in the receptor-binding region of PA domain 4, which is absent in vaccine strains from China, Brazil and in the 34F2 Sterne strain used for vaccination of livestock in Western countries. These data can be useful in development of new anthrax vaccines.

The global dominance of genotypes of lineage A serves as a proof of a bid reproductive success (and, consequently, adaptation) and considerable dispersion over large distances [27]. Smith et al. assume that strains of lineage A, though not lineage B, have a hypothetical ability to cause latent infection in animals, which is associated with their global distribution and the restricted distribution of lineage B. The comparison of isolates from lineages A and B from South Africa showed that strains A were adapted to more diverse environments than strains B, which were restricted by narrower environments [28]. The limited number and restricted geographical distribution of rarer lineages can be caused by higher adaptation costs, which are associated with niche specialization [29].

The genotypes from lineage C and, to a lesser extent, from lineage B, apparently have a very poor adaptation compared to the genotypes from lineage A. Indeed, the clade of lineage C is characterized by significantly slower rates of evolution than the clade of lineage A, implying the smaller number of infectious cycles in nature [1].

The differences between evolutionary lineages in terms of susceptible hosts can also contribute to differences in their distribution. Strains of the B.Br.CNEVA group of lineage B were reported only in France, Southern Germany, Switzerland, Northern Italy, Bosnia and Herzegovina, Croatia, Slovenia, Slovakia and Poland. They constitute a transalpine axis represented by valley grasslands with rich meadows, traditionally focusing on specific breeds of cattle, which were kept isolated and were not exchanged for centuries. Such geographic isolation might have provided favorable environment for spore survival and replication of B. anthracis of the B.Br.CNEVA group [2, 30].

Conclusion

There are significant differences in the numbers of polymorphisms in genomes of representatives of major genetic B. anthracis lineages A, B and C. Strains of the most geographically restricted lineage C had 4.5 times and strains of lineage B, which is also quite restricted, 3 times as many types of polymorphisms compared to strains of numerous lineage A. The dominating location of nucleotide substitutions, including significant substitutions, within housekeeping genes and pathogenicity factors could contribute to changes in functions of respective proteins. The expansion of lineage A can be explained by its advantages over lineages B and C, which were established during the evolution. Lineage C, which is evolutionary older and is primary in relation to lineages B and A as well as least adapted, is restricted in distribution by negative selection. The evolution of B. anthracis, which is associated with the variability of pathogenicity factors, helps identify genetic groups that are different from canonical SNP groups. MVLST having a good discriminatory power can be used as an additional method of molecular typing of the anthrax pathogen, making it possible to differentiate strains using pathogenicity determinant.

In our study, we, for the first time, found that strains of lineages B and C have significantly higher numbers of polymorphisms in genomes, including pathogenicity factor genes, compared to lineage A. We identified significant substitutions in chromosomal and plasmid genes, which have a potential effect on virulence, and demonstrated a high discriminatory power of MVLST) using the analysis of SNPs of 19 pathogenicity factor genes. For the first time, the mechanism underlying the tryptophan dependance in some B. anthracis strains of lineage B was identified, demonstrating its association with mutations in the gene of the tryptophan synthase alpha subunit.

The variability of genes associated with spore germination and spore formation should be studied further, as they can also have an impact on adaptation and distribution of the genetic lineages of B. anthracis.

Thus, the dominance of major genetic lineage A of B. anthracis can be explained, among other versions, by a significantly smaller numbers of genomic mutations compared to B and, especially, C lineages as well as by its better adaptation to external and host environments.

Funding source. This study was not supported by any external sources of funding.

Conflict of interest. The authors declare no apparent or potential conflicts of interest related to the publication of this article.

 

¹ URL: ftp://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/833/275/

About the authors

Evgeny I. Eremenko

Stavropol Plague Control Research Institute

Author for correspondence.
Email: ejer@mail.ru
ORCID iD: 0000-0002-1117-1185
SPIN-code: 3966-6884
Scopus Author ID: 6701369832
ResearcherId: AAJ-7406-2020

D. Sci. (Med.), Prof, Chief Scientist, Anthrax laboratory

Russian Federation, Stavropol

Grigorii A. Pechkovskii

Stavropol Plague Control Research Institute

Email: grigorii.pechkovskii@gmail.com
ORCID iD: 0000-0001-7033-9972

junior researcher, Anthrax laboratory

Russian Federation, Stavropol

Alla G. Ryazanova

Stavropol Plague Control Research Institute

Email: anthraxlab.stv@mail.ru
ORCID iD: 0000-0002-5196-784X

Cand. Sci. (Med.), Head, Anthrax laboratory

Russian Federation, Stavropol

Sergey V. Pisarenko

Stavropol Plague Control Research Institute

Email: pisarenko_sv@mail.ru
ORCID iD: 0000-0001-6458-6790

Cand. Sci. (Chem.), leading researcher, Laboratory of biochemistry

Russian Federation, Stavropol

Dmitry A. Kovalev

Stavropol Plague Control Research Institute

Email: kovalev_da.stv@list.ru
ORCID iD: 0000-0002-9366-5647

Cand. Sci. (Chem.), Head, Laboratory of biochemistry

Russian Federation, Stavropol

Lyudmila Yu. Aksenova

Stavropol Plague Control Research Institute

Email: anthraxlab.stv@mail.ru
ORCID iD: 0000-0002-7744-3112

Cand. Sci. (Med.), senior researcher, Anthrax laboratory

Russian Federation, Stavropol

Ol’ga V. Semenova

Stavropol Plague Control Research Institute

Email: anthraxlab.stv@mail.ru
ORCID iD: 0000-0003-0274-898X

Cand. Sci. (Biol.), researcher, Anthrax laboratory

Russian Federation, Stavropol

Alexander N. Kulichenko

Stavropol Plague Control Research Institute

Email: kulichenko_an@list.ru
ORCID iD: 0000-0002-9362-3949

D. Sci. (Med.), Prof., Academician of RAS, Director

Russian Federation, Stavropol

References

Supplementary files