Toxin evolution and transmission in the soil environment

Many bacterial pathogens, even those for which humans or animals can serve as the natural or primary reservoir, spend a substantial amount of time outside the host body in the external environment. This suggests that the evolution of pathogens through HGT may not occur only in the host environment. For example, it is now well recognized that the selective pressure of xenobiotic pollutants in soil and water can lead to the acquisition by soil bacteria of plasmids encoding xenobiotic-degrading enzymes (Davison, 1999; Demaneche et al., 2001b; Ashelford et al., 2003). There is even some evidence that natural electro-transformation, as might occur during a thunder-lightning storm, might be a feasible mechanism for increasing the frequency of HGT in soil microcosms (Demaneche et al, 2001a).

Bacillus species are common spore-forming soil bacteria barely distinguishable at the genome sequence level (Helgason et al., 2000; Read et al., 2003; La Duc et al., 2004), yet this group of bacteria differs considerably in their virulence properties. B. subtilis is commonly used as a nonpathogenic laboratory model system for studying bacterial sporulation; B. thuringien-sis produces a number of insecticidal toxins widely used as pesticides in agriculture and more recently in genetically modified plants to confer insect resistance; B. cereus is a food-borne pathogen capable of causing human and animal gastrointestinal disease; and B. anthracis is a human and animal pathogen that causes anthrax disease and has received much attention recently because of its potential use as a bioterror agent.

The structural genes encoding the major virulence factors of B. anthracis responsible for anthrax, the anthrax lethal toxin and edema toxin genes (pag, lef, cya) and the poly-D-glutamate capsule biosynthetic genes (capBCA), reside on two large plasmids, pXO1 and pXO2, respectively (Okinaka et al., 1999a; Okinaka et al., 1999b). Loss of the pXO2 plasmid resulted in the greatly attenuated Sterne vaccine strain. Although it does not appear that these plasmids are self-transmissible, there are reports that suggest conjugative plas-mids from other Bacillus species might be able to supply the conjugal transfer functions in trans for these two virulence plasmids (Andrup et al., 1996; Pannucci et al., 2002). If this is true, then it is conceivable that other Bacillus species may serve as an environmental reservoir for the anthrax toxin genes. And, indeed, HGT is reported to be quite common among Bacillus species (La Duc et al., 2004), with an intriguing example being the finding that several of the B. anthracis pXO1 genes have homologues on the B. cereus chromosome Helgason et al., 2000) and that pXO1 contains 15 ORFs with sequence similarity to transposases, integrases, or recombinases, as well as a number of insertion (IS) elements with high homology to IS sequences found in B. thuringiensis and B. cereus (Okinaka et al., 1999a; Okinaka et al., 1999b). However, within the B. anthracis group there appears to be surprisingly little genetic variation, as illustrated by sequence analysis of the protective antigen toxin gene, pag, which revealed only five point mutations among 26 diverse B. anthracis isolates (Price et al., 1999). This finding was consistent with a previous report that used chromosomal markers (Keim et al., 1999), and may reflect a very recent origin of this species or other unknown population constrictions that warrant further study (Keim et al., 1999; Okinaka et al., 1999a; Okinaka et al., 1999b).

B. thuringiensis produces a considerable arsenal of toxins directed against insects and nematodes, with multiple toxin-encoding genes on plasmids and various mobilizable genetic elements on the chromosome (de Maagd et al., 2003). HGT, combined with recombination and shuffling between toxin genes (resulting in domain swapping) and sequence divergence, has yielded a wide range of host specificities for these insecticidal toxins (Lee et al., 1995; Bravo, 1997; de Maagd et al., 2001). The genes encoding the crystal protein toxins, for example, are frequently clustered on different transmissible plasmids or transposable elements (Schnepf et al, 1998; de Maagd et al, 2001), and conjugation between different strains has been observed in the soil environment and within insect guts (Thomas et al., 2001). Individual toxins have insec-ticidal activity only against a limited range of insect species, i.e., usually only within certain insect orders. The composite of the toxins produced by a particular strain thus defines the total insecticidal specificity and activity spectrum of that bacterial strain (de Maagd etal, 2001).

0 0

Post a comment