ReviewAnthrax molecular epidemiology and forensics: using the appropriate marker for different evolutionary scales
Section snippets
Background and introduction
Molecular typing of Bacillus anthracis has provided important insights into bioterrorism or biowarfare related events. Recently, analyses performed by the Centers for Disease Control and Prevention (CDC) and the Keim Genetics Laboratory determined that the initial victim of the 2001 anthrax letter attacks had been infected with the Ames strain of anthrax (Hoffmaster et al., 2002, Read et al., 2002). As Ames is a commonly used laboratory strain and is rare in nature (Keim et al., 2000), the
B. anthracis exhibits low genetic diversity
Pathogens that emerge through a population bottleneck (likely a single cell) initially form groups of genetically identical organisms or clones. Over evolutionary time, mutations will eventually generate genetic heterogeneity within these organisms and provide the basis for distinguishing among individual lineages. For this reason, when it is found that a pathogen has little or no genetic diversity it is assumed to have recently emerged. B. anthracis appears to fit this definition as it
Genetic diversity and mutation rates
Genetic diversity, a compound measure that includes the number of allelic states as well as their frequency distribution within the population, is affected by four processes: mutation, selection, genetic drift and recombination. Over evolutionary time, mutations provide the raw material of diversity by creating additional alleles. However, this process eventually approaches a maximum value, where novel mutations are offset by recurrent mutations (Fig. 1). Selection, genetic drift and
Canonical markers: single nucleotide polymorphisms (SNPs)
SNPs occur at very low frequencies in the B. anthracis genome but they can be readily discovered using intensive sampling methods. As we describe above, the genome of B. anthracis is relatively homogenous and point mutations that form SNPs occur at very low rates. In comparison to other genetic markers, SNPs are exceedingly rare among even distantly related B. anthracis isolates and, therefore, would seem to have limited subtyping capacity. Fortunately, despite their rarity, thousands of SNPs
High resolution analysis: multiple locus VNTR analysis (MLVA)
Amplified fragment length polymorphism (AFLP) analysis was an important first step in the molecular characterization of B. anthracis and led to the discovery and use of VNTRs in this pathogen (Keim et al., 1997). Rare variation was observed within the AFLP markers, which allowed successful differentiation among some B. anthracis isolates and identification of novel genetic lineages. Even this limited diversity in the AFLP markers was an improvement over other molecular typing methods (Harrell
Highest resolution analysis: single nucleotide repeats (SNR)
Genetic differentiation among very closely related individuals requires the use of molecular markers that exhibit very high diversity and, therefore, have very high mutation rates. Single nucleotide repeats are a class of VNTRs that have been well characterized in the human genome, where they have been shown to exhibit extreme mutability (Mori et al., 2001, Zhang et al., 2001). These regions are mutational hot spots due to high occurrences of slipped-strand mispairing, which reduces the
Maximizing phylogenetic accuracy across evolutionary scales in B. anthracis using PHRANA (progressive hierarchical resolving assays using nucleic acids)
The ultimate utility of SNPs, VNTRs and SNRs as informative genetic markers is largely dependent on their mutation rates, as well as the population being examined (Fig. 3). If marker mutation rates are extremely low, then polymorphisms in those markers will only be detected in highly diverse populations and less diverse populations will appear relatively monomorphic. Conversely, if mutation rates are high, diversity will quickly arise at these loci, enabling differentiation among even closely
Conclusion
The genomes of pathogenic bacteria are relatively large, providing the opportunity to selectively choose optimal strategies for molecular subtyping. This involves the inevitable trade-off between selecting markers with high genetic resolution and selecting markers that accurately describes large evolutionary relationships. A common solution for trying to overcome this trade-off is to combine data from several marker types that exhibit variable levels of genetic diversity and, therefore,
Acknowledgements
This work was supported by grants from National Institutes of Health—General Medical Sciences, Department of Energy—Chem Bio Non Proliferation program, National Science Foundation and the Cowden Endowment for Microbiology. We also thank our collaborators at TIGR (Jacques Ravel and Timothy Read) and NAU (Joseph Busch, Ryan Easterday, Sergey Kachur, Rebecca Leadem, Shane Rhoton, Tatum Simonson, Daniel Solomon, Jana U’Ren and Shaylan Zanecki) for allowing us to discuss the implications of
References (48)
- et al.
Mutation rate in human microsatellites: influence of the structure and length of the tandem repeat
Am. J. Hum. Genet.
(1998) Estimating character weights during tree search
Cladistics
(1993)- et al.
Base stacking and even/odd behavior of hairpin loops in DNA triplet repeat slippage and expansion with DNA polymerase
J. Biol. Chem.
(2000) - et al.
Differentiation of Bacillus anthracis and other ‘Bacillus cereus group’ bacteria using IS231-derived sequences
FEMS Microbiol. Lett.
(1995) - et al.
Length of CTG.CAG repeats determines the influence of mismatch repair on genetic instability
J. Mol. Biol.
(2000) - et al.
Modeling bacterial evolution with comparative-genome-based marker systems: application to Mycobacterium tuberculosis evolution and pathogenesis
J. Bacteriol.
(2003) - et al.
Identification of a region of genetic variability among Bacillus anthracis strains and related species
J. Bacteriol.
(1996) - et al.
Mutation rate in the hypervariable vntr g3 (d7s22) is affected by allele length and a flanking DNA sequence polymorphism near the repeat array
Am. J. Hum. Genet.
(1996) - et al.
Influences of array size and homogeneity on minisatellite mutation
EMBO J.
(1998) - et al.
Clinicopathologic characteristics related to the high variability of coding mononucleotide repeat sequences in tumors with high-microsatellite instability
Oncol. Rep.
(2003)