Use of Single Nucleotide Polymorphisms in the plcR Gene for Specific Identification of Bacillus anthracis

Abstract

A TaqMan-minor groove binding assay designed around a nonsense mutation in the plcR gene was used to genotype Bacillus anthracis, B. cereus, and B. thuringiensis isolates. The assay differentiated B. anthracis from these genetic near-neighbors and determined that the nonsense mutation is ubiquitous across 89 globally and genetically diverse B. anthracis strains.

The genetic similarities among the pathogenic, spore-forming soil bacteria Bacillus cereus, B. thuringiensis, and B. anthracis have resulted in the suggestion that they be considered members of the same species (3). Interestingly, these bacteria exhibit phenotypic differences and express virulence in diverse ways. B. cereus and B. thuringiensis are opportunistic pathogens in mammals due to the secretion of nonspecific virulence factors, such as hemolysins, the expression of which is regulated by the transcriptional activator PlcR (8). In B. anthracis, PlcR is inactivated due to a nonsense mutation in the plcR gene (1), and its virulence in mammals is attributed to the expression of specific toxins under the control of the AtxA regulator (2).

The nonsense mutation in the plcR gene of B. anthracis may represent an evolutionarily stable, species-specific marker. Research by Mignot et al. (8), in which a functional PlcR was expressed in B. anthracis, demonstrated that PlcR- and AtxA-controlled regulons were incompatible, as plcR expression interfered with sporulation in B. anthracis. Since sporulation is a critical component of the ecology of B. anthracis, the authors speculated that a functional PlcR is counterselected in this species. Recent sequence comparisons of the plcR genes of two phylogenetically distinct B. anthracis lineages revealed the same nonsense mutation in the plcR gene (9), providing additional evidence to support the species specificity of this mutation.

To initially test the utility of the nonsense mutation in plcR as a species-specific marker for B. anthracis, we examined the plcR gene fragments that surround the nonsense mutation in several Bacillus spp. The strains examined included nine genetically diverse B. anthracis strains, nine B. cereus strains, six B. thuringiensis strains, and one unidentified near-neighbor (TET 2b-3) (4). Sequences obtained either from GenBank or from sequencing efforts in our laboratory were compared using MegAlign (Fig. 1). The nonsense mutation was present in all nine of the B. anthracis sequences and was absent in the 16 near-neighbor sequences.

FIG. 1.

Sequence alignment of plcR gene fragments from B. anthracis and genetic near-neighbors. The numbered lines indicate the sequences of (line 1) the primers and probes used in the assay; (line 2) B. anthracis strains Ames, Vollum, A2012, A1055, AUS94, CNEVA9066, Kruger, Sterne, and WNA6153; (line 3) B. cereus strains 3A (GenBank accession no. AY785766) and S2-8 and B. thuringensis strain HD1011; (line 4) B. thuringensis strain 97-27 (AY785771); (line 5) near-neighbor strain TET 2b-3; (line 6) B. cereus strain AH-527 (AY785767); (line 7) B. cereus strain D17 (AY785768) and B. thuringensis strains HD682, HD571, and HD44; (line 8) B. cereus strains F3502/72 (AY785769) and R6; (line 9) B. cereus strain F2-1 (AY785770); and (line 10) B. cereus strains R4 and ATCC 33018 and B. thuringensis strain HD1012 (AY785772). Light shading indicates areas of polymorphism that are detected in the assay. Darker shading indicates nucleotide differences between near-neighbors and B. anthracis. * indicates the nonsense mutation.

Based upon these sequences, we designed a TaqMan-minor groove binding (MGB) allelic discrimination assay around the nonsense mutation. The TaqMan-MGB probes were designed using Primer Express software (Applied Biosystems, Foster City, CA). One probe was designed to specifically hybridize to the B. anthracis sequence (5′-VIC-CAAAGCGCTTATTCGTATT-3′-MGB), and the other was designed to hybridize to the alternate allele (5′-FAM-AAAGCGCTTCTTCGTATT-3′-MGB) (Fig. 1 shows probe locations). Real-time PCRs were conducted in 10.0-μl reaction mixtures that contained 600 nM of both forward (5′-CCAATCAATGTCATACTATTAATTTGACAC-3′) and reverse (5′-ATGCAAAAGCATTATACTTGGACAAT-3′) primers (Fig. 1 shows primer locations), 250 nM of each probe, 1× Invitrogen Platinum qPCR SuperMix-UDG, and 1.0 μl of template. Thermal cycling was performed on an ABI 7900 HT sequence detection system (Applied Biosystems) under the following conditions: 50°C for 2 min, 95°C for 2 min, and 40 cycles of 95°C for 15 s and 60°C for 1 min.

To further evaluate the nonsense mutation in plcR as a species-specific marker for B. anthracis, we used the assay described above to genotype a collection of B. anthracis strains representing 89 unique genetic lineages (6). In addition, we genotyped 29 strains that were identified by amplified fragment length polymorphism analysis as genetic near-neighbors of B. anthracis (4) (Table 1 shows strain list). All of the B. anthracis isolates supported amplification and were shown to have the plcR nonsense mutation genotype (T allele). Not surprisingly, genetic near-neighbors that had mutations in the priming site either failed to exhibit amplification or amplified with lower efficiency relative to the four strains that had complete sequence identity to B. anthracis except for the nonsense mutation (Table 1; Fig. 1). Of the 29 near-neighbors, 16 failed to exhibit amplification and the remaining 13 exhibited the G allele genotype (Table 1). The presence of the G allele in 5 of the 16 isolates that failed to amplify in the assay was confirmed via sequencing with flanking primers (Fig. 1).

TABLE 1.

List of Bacillus sp. strains examined using the assay developed in this study

Speciesa,b	Strainb	plcR gene fragment sequenced	Avg threshold cyclee	TaqMan result (allele)
BA	89 diverse strainsc	2	26.0f	+ (T)
BC	ATCC 4342	NA	No Amp	−
BC	ATCC 14579	NA	No Amp	−
BC	D17	7	No Amp	− (G)g
BC	F3-27	NA	No Amp	−
BC	F3502/72	8	27.1	+ (G)
BC	R6	8	27.6	+ (G)
BC	ATCC 33018	10	38.5	+ (G)
BC	D5	NA	28.4	+ (G)
BC	3A	3	24.8	+ (G)
BC	S2-8	3	27.0	+ (G)
BC	F3350/87	NA	No Amp	−
BC	S2-4	NA	34.9	+ (G)
BC	R4	10	34.7	+ (G)
BC	F2-1	9	No Amp	− (G)g
BC	AH 527	6	30.0	+ (G)
BT	HD 1015	NA	No Amp	−
BT	HD 681	NA	No Amp	−
BT	HD 288	NA	No Amp	−
BT	HD 526	NA	No Amp	−
BT	97-27	4	33.0	+ (G)
BT	HD 1011	3	26.7	+ (G)
BT	HD 571	7	No Amp	− (G)g
BT	HD 682	7	No Amp	− (G)g
BT	HD 974	NA	No Amp	−
BT	HD 44	7	No Amp	− (G)g
BT	HD 30	NA	No Amp	−
BT	HD 1012	10	33.0	+ (G)
BT	HD 50	NA	No Amp	−
UNK	TET-2B	5	33.3	+ (G)

^aBA, B. anthracis; BC, B. cereus; BT, B. thuringiensis; UNK, unknown Bacillus spp.

^bSpecies and strain designations according to reference 4.

^cThe 89 diverse B. anthracis strains are described in reference 6.

^dAs represented in Fig 1. NA, strain not sequenced.

^eInput, 10 pg, average of triplicate cycle threshold values. No Amp, no amplification.

^fAverage of triplicate cycle threshold values from 10-pg input of B. anthracis Ames strain.

^gplcR genotypes (Fig. 1) were determined via sequencing using flanking primers.

To test the limit of detection of the assay, we utilized a dilution series generated from DNAs from three diverse B. anthracis isolates (Ames [A0462], Kruger B1 [A0442], and Vollum [A0488]). DNA was quantified using a Pico Green assay, and template levels ranging from 100 pg to 10.0 fg were used in the plcR TaqMan assay. The assay reliably detected and genotyped Bacillus anthracis DNA template at levels as low as 100 fg, with 10-fg samples exhibiting sporadic amplification (Fig. 2).

FIG. 2.

Results of triplicate analysis of 10-fold serial dilutions of DNA from B. anthracis strain A0442. The average cycle threshold values across the three replicates were as follows: 100 pg, 21.4; 10 pg, 24.7; 1 pg, 28.2; 100 fg, 31.2; 10 fg, 34.3. The average cycle threshold values for strains A0488 and A0462 were similar (data not shown), although amplification at the 10-fg level was not consistent.

Our data provide further evidence that the nonsense mutation in the plcR gene of B. anthracis is an evolutionarily stable, species-specific marker. Although additional genetic changes, such as deletions, could produce a nonfunctional PlcR in B. anthracis and potentially cause false-negative results in our assay, this was not observed. The presence of this mutation in the 89 genetically diverse B. anthracis lineages examined here, as well as the known genetic homogeneity of the species (5), limits the likelihood of alternate genetic mechanisms for plcR inactivation in B. anthracis. The recent findings of Slamti et al. (9), which demonstrated that this specific plcR nonsense mutation was not responsible for the nonhemolytic properties of B. cereus and B. thuringiensis strains, further support the concept that this nonsense mutation is a defining or canonical single nucleotide polymorphism (7) for B. anthracis.

The real-time assay presented here represents a potentially valuable diagnostic tool in the event of a future bioterrorist attack. From a biodefense perspective, diagnostic assays allowing rapid and specific identification of B. anthracis are critical to initiate appropriate first-response actions, such as remediation measures and prophylactic therapies. As our assay targets a well-characterized, biologically relevant single nucleotide polymorphism, it limits the likelihood of false-negative or -positive results, which can lead to misallocation of resources during an attack scenario. Furthermore, this assay is amenable to high-throughput real-time PCR platforms that are currently used in homeland defense initiatives, such as BioWatch.

In summary, our results indicate that the plcR nonsense mutation is ubiquitous in globally and genetically diverse B. anthracis isolates and, thereby, represents an excellent target for diagnostic assays. Future studies will involve genotyping more extensive collections of B. anthracis and genetic near-neighbors, as well as the optimization and validation of this assay for the specific, low-level detection of B. anthracis in complex environmental samples.