1.B.12 The Autotransporter-1 (AT-1) Family

Pathogenic Gram-negative bacteria produce a diversity (over 700 sequenced autotransporters) of virulence factors which cross the cytoplasmic membrane via the Sec (general secretory) pathway (TC #3.A.5), and following cleavage of their N-terminal targetting sequence, they thereby enter the periplasm of the Gram-negative bacterial cell envelope (Benjelloun-Touimi et al., 1995; Finn et al., 1995; Jose et al., 1995; Suzuki et al., 1995). The C-terminal 250-300 amino acyl residues of proteins known as 'autotransporters' fold and insert into the outer membrane to give rise to β-barrel structures with 12 transmembrane β-strands (TMSs) (Loveless et al., 1997; Maurer et al., 1999; Oomen et al., 2004). Karuppiah et al. (2011) have reviewed channel formation in outer membrane translocons. Sauri et al. (2011) have concluded that efficient passenger secretion requires the β-domain that not only functions as a targeting device but also is directly involved in the translocation of the passenger to the cell surface. Leo et al. (2012) review these and other (putative) autotransporters.  Drobnak et al. 2015 proposed a unified, nomenclature for AT structural, functional and conserved sequence features. Several reports suggest that AT1 proteins may not alone translocate their passenger domains to the outer surface of the outer membrane, but instead may use the β-barrel assembly machinery (BAM complex) to accomplish this task (Peterson et al. 2018). This is confirmed in a review by van Ulsen et al. 2018.

Secretion of autotransporters from several organisms requires the outer membrane assembly factor YaeT (Jain and Goldberg, 2007). This structure may form an oligomeric (8-10 mer) pore through which the N-terminal virulence factor is transported to the extracellular milieu (Guyer et al., 2000; Veiga et al., 2002). Alternatively, the unfolded protein may pass through the β-barrel of the monomer, or another export complex such as the OmpIP (TC #1.B.33) system may export the passenger domain (Skillman et al., 2005; Bernstein, 2007). Pore formation in lipid bilayers by several of these autotransporter (AT) domains, e.g., that in BrkA (TC #1.B.12.2.3) and EspP of E. coli (TC #1.B.12.4.3), has been demonstrated (Shannon and Fernandez, 1999; Skillman et al., 2005). Following its export, the precursor virulence factor is usually (but not always) proteolytically digested to release a soluble protein that can promote virulence (St. Geme et al., 2000).

Following translocation, the passenger domains of some autotransporters are cleaved by an unknown mechanism. The passenger domain of the Escherichia coli O157:H7 autotransporter EspP is released in an autoproteolytic reaction. After purification, the uncleaved EspP precursor undergoes proteolytic processing in vitro (Dautin et al., 2007). An analysis of protein topology together with mutational studies strongly suggested that the reaction occurs inside the β-barrel and that two conserved residues, an aspartate within the β-domain (Asp(1120)) and an asparagine (Asn(1023)) at the P1 position of the cleavage junction, are essential for passenger domain cleavage. These residues are also essential for the proteolytic processing of two distantly related autotransporters. Asp and Asn probably form catalytic dyad that mediates self-cleavage through the cyclization of the asparagine. A similar mechanism has been proposed for the maturation of eukaryotic viral capsids.

The 3-D x-ray crystallography structure of the translocator domain of the autotransporter, NalP, of Neisseria meningitidis has been solved (Oomen et al., 2004). The 12-stranded β-barrel shows a central hydrophilic pore of 10 x 12.5 Å that is filled by an N-terminal α-helix. This domain has pore activity in vivo and in vitro. Oomen et al. (2004) propose that the unfolded passenger domain is transported through the hydrophilic channel in the β-barrel. They suggest alternatively that Omp85, required for outer membrane protein insertion, may play a role.

Structural data suggest that the diameter of the beta-barrel pore may not be sufficient to allow the passage of partly folded structures. Sauri et al., (2009) used a stalled translocation intermediate of the autotransporter, Hbp, to identify components involved in insertion and translocation of the protein across the outer membrane. At this intermediate stage the beta-domain was not inserted and folded as an integral beta-barrel in the outer membrane whereas part of the passenger was surface exposed. The intermediate copurified with the periplasmic chaperone SurA and subunits of the Bam (Omp85) complex that catalyze the insertion and assembly of outer membrane proteins (1.B.33). A critical role for this general machinery in the translocation of autotransporters across the outer membrane seems reasonable.

Ieva and Bernstein (2009) showed that the insertion of a small linker into the passenger domain of the E. coli autotransporter EspP (1.B.12.4.3) effectively creates a translocation intermediate by transiently stalling translocation near the site of insertion. Residues adjacent to the stall point interact with BamA, a component of a heterooligomeric complex (Bam complex) that catalyzes OM protein assembly (1.A.33). Residues closer to the EspP N terminus interact with the periplasmic chaperones SurA and Skp. The EspP-BamA interaction was short-lived and could be detected only when passenger domain translocation was stalled. Molecular chaperones may thus prevent misfolding of the passenger domain before its secretion, and the Bam complex may catalyze both the integration of the beta domain into the OM and the translocation of the passenger domain across the OM in a C- to N-terminal direction.

The crystal structure of the autotransporter, Hbp (Tsh) of E. coli (TC #1.B.12.4.2), has been solved at 2.2 Å resolution. The hemoglobin proteases passenger domain proved to have the largest parallel α-helical structure yet solved (Otto et al., 2005). This structure is not likely to be applicable to all passenger domains of AT family members since these may possess any of a variety of functions.

Although the C-terminal autotransporter (AT) domains are all homologous, they are extremely diverse in sequence. Moreover, the N-terminal virulence factor domains are not all homologous. These various protein domains can (1) catalyze proteolysis, (2) serve as adhesins, (3) mediate actin-promoted bacterial motility or (4) serve as cytotoxins to animal cells. The intact protein, prior to processing, can vary in size between 418 amino acyl residues and 3705 residues. A few proteins appear to consist only of the AT domain. Such proteins might reasonably transport non-covalently linked proteins. A lack of specificity for the protein transported has been demonstrated for some autotransporters (Lattemann et al., 2000). Some unlinked autotransporters have been predicted to consist of 19 rather than 12 β-stranded barrels (Henderson et al., 2000).

The β-subunit of Flu (TC #1.B.12.1.3) (the AT domain) has been shown to transport the α-subunit (obtained by processing the intact Flu protein). The β-subunit can be used to display many foreign antigens, including whole protein domains, on the bacterial cell surface. This antigen expression system can be used in a wide range of proteobacteria. (Henderson et al., 1997). The EspP (TC# 1.B.12.4.3) β-domain and an embedded polypeptide segment appear to be integrated into the outer membrane as a single pre-formed unit (Ieva et al., 2008). At least some outer membrane proteins probably acquire tertiary structure prior to their membrane integration.

Autotransporters from a wide variety of rod-shaped pathogens, including IcsA and SepA of Shigella flexneri, AIDA-I of diffusely adherent Escherichia coli, and BrkA of Bordetella pertussis, are localized to the bacterial pole (Jain et al., 2006). Restriction of autotransporters to the pole is dependent on the presence of a complete lipopolysaccharide (LPS), consistent with known effects of LPS composition on membrane fluidity. Newly synthesized and secreted BrkA is polar even in the presence of truncated LPS, and all autotransporters examined are polar in the cytoplasm prior to secretion. Autotransporter secretion probably occurrs at the poles of rod-shaped gram-negative organisms. Moreover, NalP, an autotransporter of spherically shaped Neisseria meningitidis, contains the molecular information to localize to the pole of Escherichia coli. In N. meningitidis, NalP is secreted at distinct sites around the cell (Jain et al., 2006).

Adhesins of Campylobacter (1.B.12.10.1) contain repeat sequences that are homologous to repeat sequences in AT2 proteins and the toxins of TC# 1.C.11.1.4, 1.C.57.3.4 and 1.C.75.1.1, members of the RTX superfamily, as well as other toxins in these families, and TolA (2.C.1.2.1). These repeat sequences probably mediate protein-protein interacts and comprise parts of toxins.

C-terminal domains having an N-terminal α-helix and a β-barrel appear to constitute functional transport units for the translocation of peptides and immunoglobulin domains with disulfide bonds (Marín et al., 2010). In vivo and in vitro analyses showed that multimerization is not a conserved feature in AT C-terminal domains. Deletion of the conserved α-helix severely impairs β-barrel folding and OM insertion and thereby blocks passenger domain secretion. These observations suggest that the AT β-barrel without its α-helix cannot form a stable hydrophilic channel in the OM for protein translocation. 

When translocation to the cell surface is blocked, the AT passenger domain remains unfolded in the periplasm.  AT secretion is a kinetically controlled, non-equilibrium process coupled to folding of the passenger on the outer surface of the cell envelope (Drobnak et al. 2015).  Passenger folding is therefore presumed to be a driving force for OM translocation, but possibly another energy source is required to initiate the process.  The TamA/TamB proteins of the Translocation and assembly module (TAM) complex may catalyzed autotransport protein export (Heinz et al. 2015).

The generalized transport reaction catalyzed by AT domains is:

Protein virulence factor (periplasm) → protein virulence factor (external milieu)