What Glues the Glue to the Cell Surface?

ABSTRACT

In the Caulobacterales, a highly adhesive polysaccharide called the holdfast mediates attachment to exogenous surfaces. The mechanism by which this polysaccharide is anchored to the cell envelope is not well defined. N. K. Chepkwony, G. G. Hardy, and Y. V. Brun (J Bacteriol 204:e00273-22, 2022, https://doi.org/10.1128/jb.00273-22) report the characterization of HfaE, a localized surface protein with amyloid-like properties that is required for robust holdfast anchoring. This study expands our understanding of the protein factors that attach a bacterial “superglue” to the surface of the cell.

TEXT

Cell attachment is the first step in the formation of adherent biofilm communities (1, 2). Many types of protein and polysaccharide cell surface structures that function as adhesins and facilitate attachment to biotic and abiotic substrates have evolved in the bacterial kingdom (3). These structures must be secreted and then must remain attached to the bacterial cell envelope to mediate adhesion. Several mechanisms that anchor protein-based adhesins to the envelope have been described. Pili and flagella often contribute to surface adherence; these protein polymers are anchored to the cell via a multiprotein trans-envelope base (4–6). Other proteinaceous adhesins are exported to the cell surface and remain attached to their export channel (e.g., LapA [7]), are covalently linked to the cell wall (e.g., fibronectin binding protein [FnBP] [8]), or nucleate on envelope-attached proteins to form amyloid-like fibers (e.g., curli [9]). The mechanisms by which polysaccharide-based adhesins are anchored to the cell surface are less well defined. Some surface polysaccharides are covalently linked to lipids anchored in the outer membrane, forming a capsule structure around the cell (10, 11). In other cases, outer membrane lectins anchor capsule polysaccharides to the cell surface (12).

The Alphaproteobacteria commonly elaborate polysaccharide adhesins known as holdfast, or unipolar polysaccharide (UPP), at one cell pole. These structures mediate nonspecific attachment to diverse surfaces and enable the formation of polar cell aggregates called rosettes. For example, Roseobacter spp. are ubiquitous colonizers of surfaces in marine environments that produce polar adhesins and form rosettes in culture (13–18). Polar polysaccharide adhesins that facilitate surface attachment and rosette formation have also been documented in free-living and plant-associated Hyphomicrobiales (e.g., Rhodopseudomonas palustris [19], Rhizobium leguminosarum [20], Agrobacterium tumefaciens [21–23], Prosthecomicrobium hirschii [24], and Hyphomicrobium sp. [25]), Sphingomonadales (26), and Caulobacterales (27–29). While these adhesins are a common trait of the Alphaproteobacteria, the chemical composition of the polysaccharide, the mechanisms of its anchoring to the cell, and the signals that regulate its synthesis vary across the members of this diverse Gram-negative class (19, 22, 29–32).

Holdfast-mediated adherence to surfaces has been studied extensively in the freshwater isolate Caulobacter crescentus, where it is reported to be the strongest known biological adhesive (33). When an external force is applied to remove an adherent Caulobacter cell from a surface, cells often rupture or detach from the holdfast before the holdfast is released from the exogenous substrate (27, 33). The C. crescentus holdfast enables attachment to a range of materials, including glass, mica, plastic, plant detritus, and the extremely hydrophobic microenvironment at air-liquid interfaces (27, 30, 34, 35). A remarkable feature of the holdfast is that while it adheres to nearly every substrate tested to date, including other holdfasts, attachment to the body of another C. crescentus cell is not observed (36). This indicates that some feature of the cell surface repels the strong, nonspecific adhesive properties of the holdfast. How then does the holdfast remain anchored to C. crescentus? What is special about the anchoring site at the pole? What glues this glue to the cell?

Holdfast-shedding mutants provided early insight into these questions. Over 30 years ago, John Smit and coworkers isolated strains with defects in holdfast anchoring (36). These mutant cells exhibited normal holdfast synthesis but were unable to efficiently hold on to the polysaccharide. The holdfast easily detached from these mutants, and cells exhibited defects in maintaining attachment to surfaces. Notably, mutants with defects in holdfast synthesis could bind holdfast synthesized by other cells at the site where the holdfast is normally elaborated (36). These results indicated that holdfast anchoring does not require holdfast synthesis and that holdfast binding functions are conferred by localized features of the envelope. Early electron micrographs of Caulobacter sp. and the closely related Asticcacaulis sp. revealed a “characteristic modification of the cortical layers of the cell at the site of holdfast secretion” (37) consistent with some type of localized anchoring structure in the envelope.

In the intervening decades, the holdfast-shedding mutants led to the identification and characterization of three envelope-associated holdfast anchor proteins, HfaA, HfaB, and HfaD, in C. crescentus (38–41). Disruption of the genes encoding any of these proteins leads to anchoring defects as evidenced by holdfast shedding, although the ΔhfaB mutant has the most profound defect. All three proteins are localized to the site of holdfast attachment. HfaB is required for the proper localization of HfaA and HfaD, but the reverse is not true. HfaB is an outer membrane protein that is thought to translocate HfaA and HfaD, which have amyloid-like characteristics, to the cell surface. Recent cryo-electron microscopy data beautifully localize HfaB to the periplasmic side of the outer membrane and HfaA and HfaD to the external face of the outer membrane (42). These data provide further evidence that HfaABD together form an outer membrane-spanning complex that anchors the holdfast polysaccharide to the tip of the cell. However, it was postulated that other holdfast attachment factors remained to be discovered. This hypothesis stemmed from the observation that the disruption of HfaB led to a more severe attachment defect than the disruption of both extracellular proteins HfaA and HfaD. These results suggested additional roles for HfaB in the export of additional anchoring factors and/or in directly contributing to polysaccharide anchoring.

In this issue, Chepkwony and colleagues report the characterization of the attachment factor HfaE (43). This team hypothesized that another anchoring factor(s) exists and that it may be encoded in the chromosomal neighborhood of known holdfast anchoring genes. As such, they explored the genomic regions surrounding the hfaABD genes of C. crescentus and a closely related marine species, Hirschia baltica, and identified hfaE. While hfaE is separated from the other anchor genes in C. crescentus, it is adjacent to the anchor cluster in H. baltica and is broadly conserved in the neighborhood of hfaABD in other holdfast-producing Caulobacterales. Chepkwony et al. first demonstrated that hfaE is required for robust surface attachment and holdfast anchoring in both species. This result is consistent with the recent identification of hfaE as a holdfast anchoring factor in a genome-scale selection to identify C. crescentus adhesion genes (44). Localization studies by Chepkwony et al. provide further evidence that HfaE participates in holdfast anchoring with HfaA, HfaB, and HfaD. Specifically, HfaE localizes to the site of holdfast attachment, colocalizes with the holdfast, and requires the holdfast secretion machinery (HfsDAB), but not holdfast synthesis, for proper localization (43). Furthermore, HfaE has amyloid-like properties: it exists in a high-molecular-weight, SDS-resistant form and forms extracellular fibers. HfaE localization at the cell pole requires both the translocation protein HfaB and the other extracellular proteins HfaA and HfaD. However, HfsE is stable and forms high-molecular-weight structures in the absence of these other proteins, a feature that distinguishes HfaE from HfaA and HfaD. The latter proteins also exhibit amyloid-like properties and require each other, but not HfaE, for localization and stabilization.

hfaABD and hfaE are conserved in Caulobacterales but are not evident in the genomes of other Alphaproteobacteria that synthesize polar polysaccharide adhesins (14, 19, 21). In fact, both hfaA and hfaE have been previously noted as signature genes in the order Caulobacterales (45). Thus, while polar polysaccharide adhesins are a common feature of the Alphaproteobacteria, the use of these particular Hfa anchoring proteins to “glue” the polysaccharide to the cell surface is unique to this order. The mechanism(s) by which other Alphaproteobacteria anchor their respective polar adhesins to the cell envelope is not known.

Until quite recently, hfaE was a gene of unknown function with a primary structure that held no functional clues or predicted domain families. The characterization of HfaE by Chepkwony et al. extends our understanding of the molecular players required to anchor holdfast polysaccharide to the cell pole. From a structural standpoint, the characteristics of HfaE are quite interesting and generate many questions. Can HfaE spontaneously polymerize into filaments, or does it require a nucleating protein to facilitate multimerization? Do HfaE filaments have a regular repeating structure, or are these large multimers a result of nonspecific aggregation? How does HfaE function in holdfast anchoring: does it directly bind and stabilize the holdfast polysaccharide, or does it function to stabilize a holdfast anchoring protein complex?

Furthermore, the molecular interactions that govern holdfast anchoring to the cell remain undefined. Cryo-electron tomography images show that the C. crescentus S-layer proteins are assembled between the outer membrane and the holdfast polysaccharide (42, 46). Do the Hfa proteins interact with the proteinaceous S-layer? Do Hfa polymers extend through and beyond the S-layer? How does the S-layer facilitate or impinge on interactions between the anchor proteins and the holdfast polysaccharide? The Caulobacter S-layer is anchored to the cell through interactions with the O-chain of lipopolysaccharide (LPS) (47, 48). Mutation of genes involved in LPS biosynthesis can bypass the need for the secreted anchoring proteins HfaA, HfaD, and HfaE (43, 49), but the molecular basis of the compensatory effects conferred by these LPS mutations is not clear. Our understanding of the interactions among the anchor proteins, lipopolysaccharide, S-layer, and holdfast polysaccharide remains quite limited. Importantly, we now have a more complete understanding of the proteins required for holdfast anchoring and can begin to address this process in biochemical and biophysical detail.

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