Nice or nasty: Protein translocation between bacteria and the different forms of response

Bacterial cells are now understood to live in close-knit communities where they can experience complete interdependence, as well as fierce competition with their neighbors. In extreme cases, cells can commit suicide for the benefit of the surrounding bacteria (1). In contrast, cells have also evolved a complex array of different ways to kill one another (2). Evolutionary theory can successfully make predictions about when cells should be nice to their neighbors and when they should be nasty. A key prediction is that cooperative behavior is more strongly favored when it is directed at cells that share genes in common, and that antagonistic behavior should be directed toward nonrelatives. An article in PNAS (3), however, shows how the same behavior can be nice or nasty, depending on the identity of the cell toward which it is targeted.

Garcia et al. (3) study a mechanism of protein translocation between bacteria known to result in growth inhibition (Fig. 1A) (4). Two bacteria need to be in direct contact for one bacterium to deliver a protein to its neighbor. Upon protein uptake by the recipient bacterium, the protein acts toxic and inhibits its growth. Bacteria that make the toxic protein themselves are immune and not inhibited in growth even if they receive the toxic protein from a neighboring bacterium (Fig. 1B). The advantage of toxin translocation results from targeting bacteria sensitive to the toxin and inhibiting their growth. Translocation of toxin to immune bacteria might seem like an inefficient mechanism and even a waste of toxin: that is, unless immune bacteria respond to the receipt of a toxic protein in a way that is beneficial for the community, such as by the formation of biofilms.

Fig. 1.

Graphic depiction of two different responses of bacteria to the translocation of a catalytically active toxin via the contact-dependent growth inhibition system. (A) Illustration of the consequences of toxin translocation into a bacterium lacking a cognate immunity protein. The recipient bacterium is inhibited in growth, indicated by gray color. (B) The translocation of a catalytically active toxin among two bacteria that are in possession of the cognate immunity protein similar to B. thailandensis analyzed in Garcia et al. (3). Receipt of the toxin changes the transcription of genes associated with the formation of biofilms and putative antiprokaryotic activity, indicated by the green color.

Contact-dependent protein translocation is known to also play a crucial role in the ability of microbial communities to form biofilms. Biofilms are composed of extracellular polysaccharides, nucleic acids, proteins, and lipids (5). Bacteria embedded in biofilms benefit from adhesion, resistance to antimicrobial compounds, and better access to nutrients (5). In their study, Garcia et al. (3) describe the formation of bacterial aggregates and the production of extracellular polymers in dependency on the translocation of catalytically active toxin between immune bacteria of the species Burkholderia thailandensis. These observations are in support of previous findings describing the requirement of toxin translocation for the formation of biofilms (6, 7). The requirement of the contact-dependent protein delivery system for the formation of biofilms has also been shown in Escherichia coli (8), although by a different mechanism than that described by Garcia et al. (3). In E. coli, the long protein filaments carrying the toxic domain also mediate adhesion themselves. Uptake of the toxin into the recipient bacterium and catalytic activity of the toxin, like in the case of B. thailandensis, are not required for biofilm formation in E. coli. Adhesion and biofilm formation are thus phenotypes found among bacteria that either use the contact-dependent protein delivery system to attach to each other, or by modifying gene expression in the recipient bacterium upon toxin internalization.

In their study, Garcia et al. (3) take a global approach in analyzing changes in gene expression in response to toxin translocation. The authors identify genes associated with biofilm formation and putative antiprokaryotic activity to be up-regulated upon protein translocation. As referred to in the discussion by Garcia et al., biofilm formation and antiprokaryotic activity have recently been described in the context of a bacterial response to competition and danger (9–11). The activation of an antiprokaryotic mechanism has been reported in response to lysis of clonal bacteria upon an attack (11). The findings described by Garcia et al. (3) add two aspects to these reports. First, the authors see a collective response of biofilms and antiprokaryotic mechanisms. Second, the responding B. thailandensis bacteria are immune to the toxin and respond to clonal bacteria that are still alive. In addition to genes up-regulated in response to protein translocation, Garcia et al. also observe that the gene encoding the translocated toxin is down-regulated. Protein translocation thus serves as a negative feedback loop across bacteria. This finding advances our understanding of the regulation of the system that is known to be activated stochastically (12) and in response to bacterial density (13). The data on the differential expression of genes support the phenotypic changes among bacteria-translocating toxin, described above, and expand our understanding of the collective response beyond the phenotype of agglutination and adhesion that has triggered the interest of Garcia et al. (3) in studying the transcriptional profile.

Garcia et al. (3) further investigate that the transcriptional changes are not limited to a response to clonal bacteria. This observation is relevant to our understanding of how bacteria target their behavior toward other cells based on the presence or absence of the gene locus encoding the social behavior. Immunity to toxins translocated in direct contact is mediated by an immunity protein that binds and inactivates the toxin (14). The toxin and immunity protein are encoded in two adjacent genes and are genetically linked (15). The genes are passed on among clonal bacteria by inheritance and are likely distributed independently of inheritance via horizontal gene transfer (16). Bacteria that carry the toxin and immunity protein, whether they are clonal or not, are treated equally and referred to as “kin” or “kind.” Bacteria that do not carry the immunity protein are eliminated. This form of discrimination, whitch mediates directional behavior toward bacteria via one allele, is referred to as the “greenbeard effect” (17, 18). Contact-dependent inhibition systems are recognized to fulfill the criteria of a greenbeard allele (19), which is followed up on in this publication (3). Specifically, Garcia et al. show that B. thailandensis change their gene expression in response to a toxin homologous to their own but derived from a different species, similarly to if they were receiving the toxin from a clonal bacterium.

The genes underlying contact-dependent protein translocation have the potential to evolve rapidly and are subject to dynamic changes (16, 20). The ability of the translocated proteins

Garcia et al. show that B. thailandensis change their gene expression in response to a toxin homologous to their own but derived from a different species, similarly to if they were receiving the toxin from a clonal bacterium.

to kill bacteria lacking the cognate immunity protein has been observed for a variety of translocated proteins in various bacterial species (4, 12, 15, 16). Garcia et al. (3) observe that the response of immune B. thailandensis to protein translocation among bacteria is rather specific to the toxin of B. thailandensis and not found with other toxins translocated by the same mechanism. Nevertheless, the observed response is likely the result of an evolutionary process resulting in an advantage of bacteria responding to toxin transfer among immune bacteria. Rearrangement of toxin- and immunity-protein encoding genes within genomes and via horizontal gene transfer might result in the constant generation of a pool of strains that differ in these genes. Survival of a particular strain likely depends on the fitness advantage the toxin confers to the strain. The ability of the toxin to inhibit relevant competitors in a given environment might be one criterion for selection of a particular toxin. The findings by Garcia et al. (3) might provide another reason as to why B. thailandensis is found with a contact-dependent delivery system equipped with this particular toxin and immunity protein, an added benefit to immune bacteria from toxin translocation. Garcia et al. introduce the term “contact-dependent signaling” for the response to protein-translocation among kind bacteria and suggest using the term “contact-dependent inhibition” when toxin translocation to nonkind results in growth inhibition. Determining the mechanism of how the transferred protein initiates the change in gene expression in the receiving bacterium is currently under investigation by Garcia et al. The findings will be an important piece of the puzzle to further understand interactions between B. thailandensis bacteria and will ultimately reveal the evolutionary driver of the response to the translocation of toxin among kind.