Surface Adhesins that Contribute to Biofilm Structure 311 Autotransporter Adhesins

In most Gram-negative bacteria, the translocation of proteins toward the extracellular medium requires crossing of the cytoplasmic membrane, the periplasm, and the outer membrane. This translocation can be achieved through at least six different secretion pathways (Economou et al. 2006). Among them, the type V secretion pathway enables a family of proteins to reach the surface with a very limited number of accessory secretion factors because most information necessary to the translocation process is contained within the secreted protein itself. These proteins, which can therefore carry out their own transport to the outer membrane, are called autotransported or autotransporter proteins.

3.1.1.1 The Type V Secretion Pathway

Pohlner and collaborators were the first to describe a model of translocation of the Neisseria gonorrhoeae IgA1 protease by the type V secretion pathway (Pohlner et al. 1987). Since then, many autotransporter proteins have been described. Each of these proteins displays a modular structure with four characteristic domains:

(1) an N-terminal signal peptide that allows translocation through the cytoplasmic membrane via the Sec general pathway; (2) a passenger alpha domain that provides functionality to the secreted proteins and is exposed to the cell surface or released in the extracellular medium; (3) a linker necessary for translocation of the passenger domain through the outer membrane; and (4) a C-terminal beta-domain forming a transmembrane pore (Henderson et al. 2004). The study of the mechanisms of type V secretion showed that, once exported through the cytoplasmic membrane via the mechanism of Sec secretion, the signal peptide is then cleaved by a peptidase that releases a mature protein into the periplasm. The beta-domain of the now periplasmic protein is then thought to insert spontaneously and form a transmembrane beta-barrel pore, enabling translocation of the passenger domain onto the surface of the cell. Though this was initially thought to be an autonomous process, some accessory factors such as the Omp85 protein are nevertheless involved in insertion of autotransporter proteins into the outer membrane (Voulhoux et al. 2003; Oomen et al. 2004). However, Omp85 also seems to be required for other outer membrane protein insertions and is therefore not specific to the type V secretion pathway. The passenger domain can then either undergo self autocatalytic cleavage or remain attached to the external membrane (Henderson et al. 1998). Autotransporter proteins were identified in most Gram-negative bacteria and were classified into three families according to the function carried by their passenger alpha domain: proteases such as the IgA1 protein, esterases, such as ApeE of Salmonella typhimurium and adhesins (Henderson et al. 1998).

3.1.1.2 Antigen 43

In 1980, B. Diderichsen identified an E. coli gene whose mutation affects various phenotypes associated with the surface properties of the bacteria. He showed that a change in this gene, called flu, prevented flocculation of bacteria at the bottom of a tube and modified the morphology of colonies on agar plates (Diderichsen 1980). The flu gene encodes antigen 43 (Ag43), a major outer membrane protein found in most commensal and pathogenic E. coli. Although E. coli K-12 has only one copy of flu, most other strains of E. coli have several copies of this gene.

Ag43 is a self-recognizing surface autotransporter protein that does not seem to be involved in non-specific initial adhesion to abiotic surfaces, but rather, promotes cell-to-cell adhesion (Kjaergaard et al. 2000a). While, in liquid culture, this property leads to autoaggregation and clump formation rapidly followed by bacterial sedimentation, it also facilitates bacteria-bacteria adhesion and leads to the three-dimensional development of the biofilm (Owen et al. 1996; Henderson et al. 1997a; Hasman et al. 1999; Kjaergaard et al. 2000a; Schembri et al. 2003a). When expressed in different species, Ag43 can also be used to promote mixed biofilm formation between different bacteria, for example, between E. coli and Pseudomonas aeruginosa (Kjaergaard et al. 2000a, 2000b).

Hence, Ag43 seems to play a key role in biofilm maturation on abiotic surfaces, but also in eukaryotic cells, where a recently discovered glycosylation reaction on

Ag43 could play a role, suggesting a link between expression of Ag43 and the ability of pathogenic E. coli to adhere to and form biofilm-like structures on epithelial cells (Anderson et al. 2003; Justice et al. 2004; Sherlock et al. 2006). Several groups have consistently demonstrated that there is a correlation between an increased level of flu expression and E. coli capacity to form a biofilm on different abiotic surfaces (Danese et al. 2000a; Kjaergaard et al. 2000a, 2000b; Beloin et al. 2006). Moreover, global gene expression studies showed that the biofilm lifestyle is often associated with increased expression of flu as compared to planktonic growth (Schembri et al. 2003b).

3.1.1.3 AidA and TibA Proteins

AidA adhesin from diarrhea-causing E. coli and TibA adhesin/invasin associated with some enterotoxigenic E. coli are two glycosylated surface proteins involved in bacterial adhesion to a variety of eukaryotic cells. Both AidA and TibA are autotransporter proteins sharing approximately 25% identity at the sequence level with Ag43, and they play a role in the virulence of different pathogenic E. coli strains. Recent studies have shown that, in addition to autoaggregation, expression of these proteins also promotes biofilm formation on abiotic surfaces (Sherlock et al. 2004, 2005). Though they are quite different with respect to size, glycosyla-tion, and processing, these three proteins share common properties: all are self-associating proteins that cause bacterial aggregation and enhance biofilm formation. They can also interact with each other via heterologous interactions, promoting the formation of mixed bacterial aggregates. Based on these properties, Klemm and co-workers proposed that they be classified together in a subgroup termed SAAT, for self-associating autotransporters (Klemm et al. 2006).

3.1.2 Exploring the Adhesin Potential of E. coli

Genetic analyses have revealed the diversity of E. coli adhesins contributing either to colonization or biofilm maturation. Few, if any, of these adhesion factors are absolutely required for biofilm formation; instead, they can be replaced by alternative adhesion factors. A recent study demonstrated that four previously uncharacterized E. coli genes (yfaL, yeeJ, ypjA, and ycgV), sharing homologies with the autotransporter adhesin Ag43 and biofilm-associated proteins (Bap) (Cucarella et al. 2001; Latasa et al. 2005), lead to clear adhesion and a biofilm phe-notype when expressed from a chromosomally introduced inducible promoter. Deletion of genes coding for these putative adhesins does not significantly alter the adhesion phenotype of the wild type MG1655 E. coli K-12 strain under laboratory conditions, indicating that these genes may be cryptic (Roux et al. 2005).

Those studies demonstrated that E. coli K-12 probably possesses a large and partly unexplored arsenal of surface adhesins with different binding specificities that are expressed under specific physiological conditions, possibly in response to different environmental cues. This adhesion potential is likely to be even greater in some pathogenic E. coli isolates, which not only often have a larger genome than E. coli K-12 and therefore express new types of adhesin or fimbrial structures (Buckles et al. 2004), but also carry several plasmids that can contribute to adherence to cells and abiotic surfaces (Perna et al. 2001; Dobrindt et al. 2002; Welch et al. 2002; Dudley et al. 2006).

Was this article helpful?

0 0

Post a comment