Structural and Biochemical Investigations of the Heme Loading Mechanisms in Hemophores HasAp from Pseudomonas aeruginosa and HasAyp from Yersinia pestis

Abstract

Under iron-limiting conditions, several Gram-negative pathogens secrete hemophore (HasA) protein that binds heme and deliver it to the outer membrane receptors in bacteria. Hemophore proteins have been structurally and biochemically characterized in their heme-free (apo) and heme-bound (holo) forms in Serratia marcescens (HasAs) and Pseudomonas aeruginosa (HasAp). Hemophores have unique structural features which are distinguished from other known proteins. The structure of HasAp (and HasAs), are composed of an "α-helical wall" on one side and a "β-sheet wall" on another. Heme is bound between the two extended loops, the Y75 loop harboring the conserved Tyr75 ligand and the H32 loop harboring the His32 ligand. The major difference between the apo- and holo-HasAp is the large conformational rearrangements of the H32 loop upon heme binding which relocates His32 side chain ~ 30 Å from open to closed form. It has been established through kinetics experiments that heme first loads onto the Y75 loop where it is coordinated by Tyr75, followed by the closure of the H32 loop and coordination of heme-iron ion by His32. Multiple sequence alignment of known hemophore sequences from Gram-negative bacteria reveals the conservation of Tyr75 but not the His32; instead in Yersinia species Gln32 is present at a similar position. To investigate the role played by the Gln32 in heme coordination, we structurally characterized hemophore from Yersinia pestis (HasAyp) in apo- and holo-form. Surprisingly, the Q32 loop in apo-HasAyp is already in closed conformation and the heme is only coordinated by one protein provided ligand, Tyr75 from the Y75 loop. Further, heme loading is fast compared to the HasAp and occurs with minimal structural changes. Analyzing the available structural information from hemophores (HasAp, HasAyp and HasAs), we looked for the interactions that stabilized the open H32 loop conformation in apo-HasAp. Our analysis revealed the presence of Arginine at position 33 (Arg33) anchoring the H32 loop to the body of the protein. In order to investigate the role played by the Arg33, we replaced it with Alanine (Ala) and carried out mutagenesis investigations. Results from X-ray crystallography, solution NMR and molecular dynamics simulations clearly indicate that single replacement of Arg33 to Ala shifts the open H32 loop conformation to a closed conformation which is very similar to the H32 loop in the wt holo-HasAp. The structure of Y75 loop is conserved with Tyr75 and its hydrogen bond partner His83 (His81 in HasAyp). To investigate the proposed requirement for the Tyr75 - His83 unit for heme coordination by Tyr75, we conducted structural and biochemical study of Y75A and His83A HasAp mutants. The high resolution crystal structure of Y75A and H83A mutants show the original α+β protein fold similar to the wt protein. In the absence of Tyr75 in Y75A mutant, His83 do not coordinate the heme-iron ion instead, a formate ion occupies the same position as that of the side chain of Tyr75, coordinating heme iron ion by one of its oxygen atoms. In H83A mutants, Tyr75 and His32 coordinate the heme-iron ion even at the pH 5.4. Despite the severe changes made in the Y75 loop residues, kinetic analysis suggests that the heme loads with biphasic kinetics similar to the wt HasAp. Taken together, heme loading onto to the Y75 loop is mainly driven by hydrophobic, pi-pi interactions onto the Y75 loop.