The hFbpABC Transporter from Haemophilus influenzae Functions as a Binding-Protein-Dependent ABC Transporter with High Specificity and Affinity for Ferric Iron

To cause disease, many bacterial pathogens must compete for growth-essential iron within the extracellular environment of the human host (30, 33, 45). The majority of pathogenic bacteria employ siderophore-dependent iron acquisition systems in competition for host iron (47). These systems involve the use of nonproteinaceous iron-chelating compounds termed siderophores, which are produced and secreted into the environment (32). In gram-negative bacteria the uptake of iron-bound siderophores involves the expression of siderophore-specific outer membrane receptors and specific inner membrane binding protein-dependent ATP-binding cassette (ABC) transporters (14). These systems offer flexibility in the acquisition of iron from multiple sources; however, the expression of the numerous gene products involved in each specific siderophore-dependent transport pathway may be metabolically demanding. In contrast, Haemophilus influenzae and pathogenic Neisseria spp. (Neisseria meningitidis and N. gonorrhoeae) utilize a highly conserved siderophore-independent high-affinity iron acquisition system (31). This system employs specific surface receptors that directly bind host iron-binding proteins, transferrin (Tf) or lactoferrin (Lf) (37). Iron is extracted from the host proteins and transported into the periplasm through an energy-dependent TonB-mediated process. Transport of free (naked) iron from the periplasm to the cytosol is mediated via the FbpABC transporter, which is composed of a ferric ion binding protein (FbpA) and an inner membrane ABC transporter consisting of a membrane permease (FbpB) and an ATP-binding protein (FbpC) (1). Although bacteria utilizing this system may express several different host protein-specific outer membrane receptors, FbpABC is a convergence point in the acquisition of iron. By a strategy similar to that used in cloning the Serratia marcescens sfuABC operon (5), the H. influenzae hitABC and N. gonorrhoeae fbpABC operons were cloned by complementation of a siderophore-deficient (aroB) Escherichia coli strain for growth on nutrient agar containing 200 μM dipyridyl, an iron chelator (1, 2). Expression in this E. coli background has served as a model system with which to study the genetic and biochemical basis of FbpABC iron transport. Results of initial studies demonstrated that the transporter genes from these diverse bacteria exhibit a high level of homology, and a common nomenclature has been devised to designate the genetic and protein components of the transporters: for H. influenzae, the gene name is hitABC and the protein name is hFbpABC; for N. gonorrhoeae, the gene name is fbpABC and the protein name is nFbpABC. The FbpABC transporters are encoded by three-gene operons under negative regulatory control of the ferric uptake regulator (fur) (7, 15). The gene encoding the ferric ion binding protein (FbpA) is separated from the downstream two genes in the operon by a putative stem loop structure indicative of a rho-independent transcriptional terminator. This is consistent with the increase in expression of FbpA by several orders of magnitude compared with that of FbpB and FbpC (26). Biochemical analyses of nFbpA and hFbpA demonstrate that these proteins bind a single ferric (Fe3+) ion with high affinity and exhibit a characteristic spectroscopic profile (13, 34, 35). Interestingly, X-ray structural analyses of these proteins show that they bind iron in a manner remarkably similar to that of mammalian transferrin by using a common set of amino acid residues and employing a synergistic anion (10, 11). This FbpA binding mechanism results in an extremely high affinity for Fe3+, similar in scale to that of Tf (nFbpA, 2.4 × 1018 M−1; N-lobe hTf, 1.8 × 1017 M−1) (44, 49). However, this affinity is 10 to 12 orders of magnitude greater than the affinities exhibited by typical bacterial periplasmic binding proteins (PBPs) for their respective substrates (e.g., maltose binding protein for maltose; Kd = 1.6 × 10−6 M) (18). Based on sequence analysis, the nFbpB and hFbpB proteins are proposed highly hydrophobic proteins that function as membrane permeases within the context of the ABC transporters. The nFbpC and hFbpC proteins are proposed ATP-binding components of these transporters (31). The large number of recent studies on the FbpA ferric binding proteins (3, 8, 10, 16, 19, 22, 38-40, 44, 50) makes a critical investigation of the functional basis of FbpABC transport timely. This study focuses on functional investigation of the H. influenzae hFbpABC transporter through expression of the hitABC three-gene locus in E. coli. Recombinant expression in this background includes the use of an E. coli strain (H-1443) that has a deletion in the aroB gene, rendering it unable to synthesize the sole E. coli siderophore enterochelin (9). This background allows investigation of the hFbpABC system while controlling for endogenous iron uptake systems (e.g., FepABCD, FecABCDE, FeoABC, and MntH) (17, 25, 28, 41) by the use of high-affinity metal chelators, 2,2,-dipyridyl, and nitrilotriacetic acid. Using this model system, we established an assay for radiolabeled iron uptake in intact cells and generated apparent Michaelis-Menten Km and Vmax constants for hFbpABC transport. We then defined the energy requirements and metal specificity of hFbpABC by monitoring the growth-inhibitory effects of metals competing for hFbpABC-mediated transport. Finally, we investigated the impact of a single-amino-acid hFbpA mutation on hFbpABC-mediated transport and derived the functional basis of this effect. These studies form the basis of continuing investigations aimed at augmenting our understanding of FbpABC iron transport and its contribution to pathogenesis of diverse bacterial species.