E. Coli FepE Polysaccharide Co-Polymerase
Created by Anh Dao
The protein FepE (PDB ID: 3B8M) has been elucidated as a biosynthetic bacterial polysaccharide co-polymerase (PCP) commonly found in Escherichia coli. From its anchored position within the plasma membrane, FepE is known to function in the determination of chain length distribution of complex lipopolysaccharides (LPS) found on the bacterial cell surface. Within this role, it shares its niche with a class of PCP proteins comprising of more than 1,150 distinct polypeptide structures that can be categorically divided into three subsequent subclasses (PCP1, PCP2, and PCP3). These classes are cohesively linked by two common N-terminal and C-terminal transmembrane helices (identified as TM1 and TM2, respectively) bracketing a coiled-coil region as well as a long, central α-helix stalk penetrating 10Å into the periplasmic region. Nonetheless, the sequence similarity between these three discrete classes of PCP proteins - or even within a particular class - has been observed to be relatively low (Tocilj et al. 2008, Cuthbertson et al. 2009).
The PCP1 subclass that contains but is not limited to FepE is further divided into two subcategories: the PCP1a proteins (including FepE) and the PCP1b proteins. The former is known to function in the conferral of different modal chain lengths, most particularly during the biosynthesis of cell-surface LPS O-antigens (Tocilj et al. 2008), an important virulence factor involved in bacterial protection and communication relative to their host cells (Tsai et al. 2011).This determination of O-antigen chain length distribution is reliant on a Wzy-dependent pathway, and the specific properties of each modal chain are conferred by the individual nature of the PCP protein encoding it (Morona et al. 2009).
Though the genes involved in the synthesis and export of these bacterial surface LPS have been identified, the mechanistic details of this complex process, specifically those relating to the role of PCP proteins and FepE in particular, are still largely inscrutable (Tocilj et al. 2008). The overall process, however, is definitively known to be reliant upon the aforementioned Wzy-polymerase dependent pathway, which is in turn dependent on a multifunctional protein complex consisting of the inner proteins Wzx, Wzy, and Wzz. The process begins as sugar repeat units are synthesized within the cytoplasm of the cell and subsequently transported to the periplasmic side of the plasma membrane via the action of the flippase Wzx. Within the periplasm, these sugar units are aggregated into extended chains by a Wzy polymerase, then subsequently transported to the outer leaflet of the plasma membrane. Wzz proteins, including PCP proteins and FepE, the protein of interest, help to determine length distribution during this polymerization process through a poorly elucidated mechanism (Morona et al. 2009). Morona et al. 2009, however, has suggested that the absence of Wzz proteins results in "polysaccharide chain lengths show[ing] a stochastic distribution, decreasing in quantity as the chain grows longer, characteristic of a competition between elongation and transfer to an acceptor."
Past research by Kalynych et al. 2011 has also indicated that these LPS-mediating Wzz proteins share a high structural similarity within the class despite their low sequence identity (~15% to ~80%). A superimposition of a WzzB protein from S. flexneri and FepE from E. coli, for example, reveals strikingly similar adoption patterns relative to their three-dimensional folds.
Beyond the scope of O-antigen chain length distribution, PCP proteins are "also involved in enterobacterial common antigen (ECA) modal chain length regulation and biosynthesis and in capsule polysaccharide (CPS) and exopolysaccharide (EPS) biosynthesis (Papadopoulos et al. 2010)." Moreover, recent research has suggested a role for FepE in the bacterial acquisition of free iron through the ferric enterobactin (a high-affinity iron-chelating compound also known as enterochelin) transport pathway. Indeed, according to Raymond et al. 2003, FepE is part of a system of channel proteins that helps to facilitate the uptake of ferric enterobactin, beginning with recognition of the ferric enterobactin complex by the aptly-named FepA ferric enterobactin receptor. Enterobactin is subsequently transported by FepB through the cytoplasmic pores formed by FepG and FepD. FepC, in conjunction with our protein of interest FepE, are hypothesized to encode components that aid in the final processes of the ferric enterobactin intake. More specifically, Ozenberger et al. 1987 yielded experimental results that suggested that “FepE provide[s] functions which act… to form the ferric enterobactin-specific cytoplasmic membrane permease.” FepE then, is also hypothesized to play a critical role in the nutritional uptake and, by proxy, biological metabolism of microorganisms such as, but not necessarily limited to, E. coli.
Structurally, FepE protomers are known to consistently adopt an elongated, common fold. In total, the protomer secondary structure is composed of a total of eight α-helix chains and four β-sheets of varying lengths that can be methodically divided into two domains: 1) an α-β base domain spanning ~32Å wide and ~35Å deep, consisting of the aforementioned N-terminal section of the periplasmic region that acts as an attachment point for 2) a protruding helical hairpin domain (α6) spanning 100Å, or ~19 turns, long (Tocilj et al. 2008, Morona et al. 2009). The interactions between these two domains are stabilized by a group of conserved, hydrophobic residues between α6 and α2 (Morona et al. 2009). The overarching FepE protein structure is consisted of about 280 amino acids composing each individual protomer chain A, B, and C.
During protein folding, these discrete protein structures are thought to self-aggregate into a nonamer structure that extends approximately 100Å into the periplasm. This spindle-shaped oligomeric structure is open at the top to form an approximate bell pattern, further generating, in conjunction with the lipid of the inner membrane at its base, a solvent-filled cavity. From studies of the closely-related protein Wza, it is hypothesized that the water-filled channel would ensure the formation of hydrogen bonds between polar amino acid side chains and the hydroxyl groups of the polymer, a phenomenon that essentially lubricates the channel (Cuthbertson et al. 2009). The overall structure is embedded into the plasma membrane by its conserved TM1 and TM2 transmembrane helices (Tocilj et al. 2008). Taken together in context, these facts are consistent with FepE's hypothesized role as a component of a system of channel proteins that helps to facilitate the uptake of ferric enterobactin (Raymond et al. 2003) and form the ferric enterobactin-specific cytoplasmic membrane permease (Ozenberger et al. 1987).
The periplasmic α-helical domains of the FepE protein have been shown to act as the dominant functional regions. Mutations within this segment, particularly the highly-conserved Pro-Gly-rich motif overlapping the TM2 segment, have been shown to cripple function (Tocilj et al. 2008). The conserved proline residue located within the β-4 sheet that contributes to the structure of the base domain is furthermore hypothesized to play a critical role in orientating the periplasmic domain relative to the plasma membrane (Morona et al. 2009). Mutation of the specific, conserved Leu-273 residue to Ala-273 resulted in total loss of function, presumably via destabilization and subsequent degradation of the protein. Single-point mutations at Asp-225 and Glu-297 (Tocilj et al. 2008), as well as deletion of various sizes within residues 256 to 273 have also been shown to give rise to shorter O-antigen lengths (Morona et al. 2009). Conversely, mutations targeting the solvent-exposed parts of the oligomer had little to no effect on the overall structure of the protein (Cuthbertson et al. 2009).