Cleave a Septum, Leave a Cell: Bdellovibrio bacteriovorus Secretes a Specialized Lytic Transglycosylase to Clear Prey Cell Septum Obstruction

ABSTRACT

Predatory microbes like Bdellovibrio feed on other bacteria by invading their periplasm, replicating within the bacterial shell that is now a feeding trough, and ultimately lysing the prey and disseminating. A new study by E. J. Banks, C. Lambert, S. Mason, J. Tyson, et al. (J Bacteriol 205:e00475-22, 2023, https://doi.org/10.1128/jb.00475-22) highlights the great lengths to which Bdellovibrio must go to affect host cell remodeling: A secreted cell wall lytic enzyme with specificity for the host septal cell wall maximizes the size of the attacker's meal and the size of the restaurant in which it can spread out. This study provides novel insights into bacterial predator-prey dynamics and showcases elegant co-option of an endogenous cell wall turnover enzyme refurbished as a warhead to enhance prey consumption.

TEXT

So, Nat'ralists observe, a Flea

Hath smaller Fleas that on him prey,

And these have smaller yet to bite 'em,

And so proceed ad infinitum…

Johnathan Swift, “On Poetry: A Rhapsody,” 1733

Bdellovibrio bacteriovorus (“leech-like Vibrio that eats bacteria,” by rough translation) is a Gram-negative curved rod whose main distinguishing feature is that it is able to live as a pericellular parasite in other Gram-negative bacteria (1). While this bacterium is a fascinating study system in and of itself, B. bacteriovorus has also recently received additional attention as a possible antibiotic alternative, since development of resistance against predation is extremely rare (2) and B. bacteriovorus targets Gram-negative bacteria, which require novel antimicrobial therapy approaches with particular urgency (3).

The life cycle of B. bacteriovorus consists of several stages that require the well-orchestrated activity of multiple enzymes (4, 5). First, the attacker attaches to the outer membrane (OM) of prey cells in a poorly understood way. Next, the prey cell periplasm is invaded, followed by resealing of the OM (6); this process is equally poorly understood. The next phase of the B. bacteriovorus life cycle is conversion of the prey cell into a stable, rounded bag of food, which allows the predator cell to elongate and ultimately divide in the prey periplasm, while sucking the juice out of the cytoplasm. The main obstruction to this part of the life cycle in the periplasm of Gram-negative bacteria is the bacterial cell wall, which consists mostly of the polysaccharide peptidoglycan (PG). PG forms a tight mesh of polysaccharide strands cross-linked by small peptide side chains and provides shape and structural integrity to the cell. For B. bacteriovorus, however, PG is simply an annoyance, preventing the formation of the “bdelloplast,” the nutrient-rich niche in the shape of a spherical prey cell that the attacker requires for proliferation and signaling to kin that this niche is occupied (7). Prey PG thus needs to be processed via the concerted action of multiple cell wall-modifying enzymes secreted by the “leech.” Interestingly, B. bacteriovorus is not content to simply destroy the cell wall but sculpts it to decorate its new home. First, d,d-endopeptidases (EPs) cleave the PG cross-links, resulting in relaxation of the PG layer (7, 8). A side effect of this cleavage, however, is potential destabilization of the PG network, which could be detrimental to Bdellovibrio if the prey cell lyses before the life cycle is completed. Next, l,d-transpeptidases are produced, which form new cross-links that presumably stabilize the spherical prey structure (6). Additional modifications include a highly unusual mechanism of tethering modified prey PG to the prey OM (9) (replacing the endogenous “Braun’s lipoprotein”) and large-scale deacetylation to give prey PG a unique signature (10), as safeguards to keep PG lytic enzymes specifically active on prey PG to avoid self-destruction (other immunity factors for endogenous PG-modifying enzymes are also produced [11]).

Cell wall lytic enzymes (collectively referred to as autolysins) ordinarily fulfill important physiological functions. Homologs of the EPs that B. bacteriovorus uses for prey PG modification, for example, have important roles in cell elongation in multiple bacteria (12). The study by Banks et al. (13) examines another group of autolysins, the lytic transglycosylases (LTGs) (14). Similar to EPs, LTGs play a role in cellular growth, albeit in a poorly defined way (15). As is the case with most autolysins, elucidating LTG function is not an easy undertaking, due to the high degree of redundancy these enzymes typically exhibit. Indeed, B. bacteriovorus is at the high end of redundancy, containing 13 LTG genes (in comparison, the Gram-negative model organism Escherichia coli contains only 8). Might the surplus LTGs be specialized enzymes for the predator-prey interaction? Intriguingly, those authors found that three MltA-like LTGs are upregulated during predation, pointing toward an involvement in prey attack. Banks et al. (13) then systematically mutated MltA-like enzymes and conducted careful predator-prey interaction studies. Those experiments revealed that prey cells infected with a mutant B. bacteriovorus strain lacking one such LTG assumed a never-before-seen morphology termed a “dumbbell,” i.e., a cell that seems to try to form a spherical shape but is bound by a central constriction belt where the septum once was. The authors then demonstrated that dumbbells were indeed formed only by prey cells that were in the midst of septation during attack and that the MltA homolog localized to and preferentially cleaved septal prey PG. On a population level, there appears to be a need to remove septal PG, the mltA mutant exhibited reduced predatorial efficiency. While prey cell exit was not found to be affected by the ability to cleave the prey septum, septum cleavage likely facilitates access to the cytoplasm of the whole cell, maximizing replication space and nourishment and thus optimally preparing progeny for dissemination. Therefore, B. bacteriovorus co-opted several typically endogenous PG lytic enzymes (EPs and LTGs) for prey PG modification.

What makes this LTG highly unusual, however, is that it is secreted into the extracellular milieu despite containing a classic lipoprotein signature. The authors conclusively demonstrated that MltA accumulates in the prey periplasm away from the predator cells. As ever so often with nonmodel organisms, this study thus provides a taste for future elucidation of exotic biology; both septal localization of MltA and its mysterious secretion pathway do not rely on known factors, because there is neither a known septal PG binding domain (e.g., the sporulation-related repeat [SPOR] domain [16]) in the LTG nor a known secretion signal. The authors propose the possibility that secretion is effected by the controlled release of LTG-containing OM vesicles (OMVs). Whether the production of OMVs is an intentional, well-controlled process or simply a more-or-less accidental blebbing in other bacteria is still controversial (17). Speculative as it may be, a specialized role for OMVs in LTG secretion might point toward an active process, at least in B. bacteriovorus. Alternatively, a completely novel secretion mechanism might be at play here. Thus, whether by saving the world as a future antibiotic or by simply providing us with the joy of exploration of novel biology, Bdellovibrio remains a cornucopia of surprises and a model well worth studying.

The views expressed in this article do not necessarily reflect the views of the journal or of ASM.