Typhoid toxin reveals a new protein secretion mechanism
Typhoid toxin is a critical virulence factor of Salmonella Typhi, the cause of typhoid fever in humans 1. In experimental animals, this toxin can reproduce some of the acute, pathognomonic symptoms of typhoid fever. Typhoid toxin is a unique AB5 toxin in that it possesses two active ("A") enzymatic subunits linked to a single pentameric "B" subunit that targets these activities to specific cells and tissues 2. In addition, typhoid toxin has a rather unique biology since it is only expressed when S. Typhi is within mammalian cells, and it is subsequently exported to the extracellular environment by a specific vesicle trafficking process 3,4. This unique biology has hampered the ability to study the mechanisms by which the toxin is exported from the bacteria into the lumen of the Salmonella-containing vacuole since, until very recently, the only assay available to monitor toxin secretion was the visualization of the vesicle carrier intermediates in infected mammalian cells. Previous studies in our laboratory identified a gene, ttsA, which is essential for the secretion of typhoid toxin from bacterial cells 5. This gene, encoded within the same pathogenicity islet that contains the genes for the components of typhoid toxin, encodes a homolog of bacteriophage N-acetyl-β-D-muramidases. The recent discovery in our laboratory of the gene regulatory network that controls the expression of typhoid toxin within cells 6 has allowed the identification of in-vitro growth conditions that permit typhoid toxin expression. These growth conditions have allowed us to study the function of TtsA and, in the process, identified what we believe to be a novel mechanism of protein secretion described in our recent paper in Nature Microbiology.
These studies were triggered by our observation that, after in vitrogrowth of S.Typhi under typhoid toxin-inducing conditions, we were not able to detect typhoid toxin in culture supernatants or on the bacterial surface, despite wild type levels of expression of both, the typhoid toxin components and TtsA. This observation prompted us to examine in detail the location of typhoid toxin within bacterial cells after growth under these conditions. The three components of typhoid toxin contain amino terminal signals that are recognized by the sec machinery for their delivery to the periplasmic space where they assemble into the holotoxin. Therefore, we assumed that, after its expression and synthesis, typhoid toxin would be in this compartment. After extensive imaging analysis coupled to selective permeabilization and protease accessibility studies we discovered that typhoid toxin was indeed located in the periplasm of bacterial cells but with an intriguing caveat. Unlike resident periplasmic proteins, which are located between the inner membrane and the peptidoglycan (PG) layer (i. e. the cis side of PG), we found typhoid toxin located between the PG and the outer membrane (i. e. the trans side of the PG), a location that it could only reach in the presence of TtsA. In the absence of TtsA, typhoid toxin was found on the cis side of the PG along with resident periplasmic proteins. This is an intriguing observation since the trans side of the PG is not generally considered a compartment that harbors periplasmic proteins that are not linked to the PG or to the outer membrane. We discovered that TtsA mediates the translocation of TT across the PG exclusively at the bacterial poles. More importantly, we discovered that TtsA function requires the activity of the peptidoglycan-editing enzyme YcbB. This enzyme is an LD-transpeptidase, which mediates the cross-link between m-Dap residues of two adjacent stem peptides (the so called 3-3 or LD crosslinks) that link the glycan strands of the PG layer. This type of crosslinks is rare in exponentially grown bacteria, and consequently TtsA cannot exert its enzymatic activity in PG obtained from bacteria grown under those conditions. These results suggest that YcbB, which we found to be enriched at the poles, imparts localized PG modifications that are essential for TtsA activity.
At this point, although we had obtained major insight into the mechanism of translocation of typhoid toxin across the bacterial envelope, we were still in search for a mechanism by which the toxin would be ultimately released from the bacterial cells. In the case of all known two-step protein secretion systems in Gram-negative bacteria, after their sec–mediated transport to the periplasmic space, substrates are then engaged presumably in the cis side of the PG by secretion machinery components that moves them through the PG and the outer membrane for their release into the extracellular space. The components of the secretion machine are most often encoded within a common gene cluster with the substrates. However, no other genes required for typhoid toxin secretion are present within the pathogenicity islet that harbors ttsA. Periplasmic proteins can be released to the extracellular space in outer membrane vesicles (OMVs) shed from bacterial cells. However, we were not able to detect outer membrane vesicles containing typhoid toxin in S.Typhi grown under the in vitro conditions that lead to typhoid toxin expression. Importantly, in infected cells, the packaging of typhoid toxin into vesicle carrier intermediates requires the interaction of PltB with a glycan receptor on the luminal side of the membrane of the Salmonella-containing vacuole 4. The topology of potential toxin containing OMVs would be incompatible with this packaging mechanism, as PltB would not be available for interaction with its sorting receptor.
Our observation that addition of low amounts of detergent was required to visualize typhoid toxin by immunofluorescence on the bacterial surface suggested to us the possibility that other membrane disrupting compounds that are consistently encountered by SalmonellaTyphi during infection (e. g. antimicrobial peptides, bile salts, etc.) may be able to trigger the release of the toxin from its location in close proximity to the outer membrane. Indeed, we found that addition of antimicrobial peptides or bile salts at concentrations way below those needed for growth inhibition triggered the rapid release of up to ~20% of the total typhoid toxin pool in wild-type S. Typhi but not in its isogenic ∆ttsAmutant grown under the same conditions. LC-MS/MS analysis of cell-free supernatants from wild type and ∆ttsAS.Typhi after stimulation with the different agonists indicated that the typhoid toxin release is specific as our analysis did not detect any other resident periplasmic proteins. The protein secretion system we have described provides the pathogen with a mechanism to release pre-synthesized toxin molecules upon the reception of environmental cues (e. g. antimicrobial peptides, bile salts) whose presence may define the very specific environment in which the toxin must exert its function.
Genomic analysis indicate that this novel secretion mechanism is present in other bacteria as homologs of TtsA can be found encoded in the immediate vicinity of toxins or large extracellular enzymes in several bacterial genomes 5. For example, in Zymomonas mobilisa homolog of TtsA, ZlyS, has been reported to be required for the release of an extracellular levansucrase 7, and that addition of certainamphiphilic compounds to Z. mobiliscultures, including some alcohols, stimulates the release of this enzyme 8. As the activity of the levansucrase eventually leads to the generation of alcohol, it is tempting to hypothesize that in Z. mobilisthis protein secretion system may have evolved so that the accumulation of the final product of the metabolic pathway stimulates the release of a critical enzyme for that metabolic pathway.
Jorge E. Galanand Tobias Geiger
Department of Microbial Pathogenesis
Yale University School of Medicine
1 Galán, J. Typhoid toxin provides a window into typhoid fever and the biology of Salmonella Typhi. Proc Natl Acad Sci U S A.113, 6338-6344 ( 2016 ).
2 Song, J., Gao, X. & Galan, J. E. Structure and function of the Salmonella Typhi chimaeric A(2)B(5) typhoid toxin. Nature499, 350-354 (2013).
3 Spano, S., Ugalde, J. E. & Galan, J. E. Delivery of a Salmonella Typhi exotoxin from a host intracellular compartment. Cell Host Microbe3, 30-38, doi:10.1016/j.chom.2007.11.001 (2008).
4 Chang, S., Song, J. & Galán, J. Receptor-Mediated Sorting of Typhoid Toxin during Its Export from Salmonella Typhi-Infected Cells. Cell Host Microbe20, 682-689 (2016).
5 Hodak, H. & Galán, J. Salmonella Typhi homolog of bacteriophage muramidases controls Typhoid toxin secretion. EMBO Reports14, 95-102 (2013).
6 Fowler, C. & Galán, J. Decoding a Salmonella Typhi Regulatory Network that Controls Typhoid Toxin Expression within Human Cells. Cell Host Microbe23, 65-76 (2018).
7 Kondo, Y.et al.Cloning and characterization of a pair of genes that stimulate the production and secretion of Zymomonas mobilis extracellular levansucrase and invertase. Biosci Biotechnol Biochem.58, 526-530. (1994).
8 Zikmanis, P., Shakirova, L., Baltkalne, M., Andersone, I. & Lilija Auzina, L. The effect of amphiphilic compounds on the secretion of levansucrase by Zymomonas mobilis. Process Biochemistry40, 3723–3731 (2005).