Staphylococcal peroxidase inhibitor: structure/function analysis and studies on host-species specificity

Abstract

Human neutrophils, the most abundant leukocytes in circulation, serve as the primary responders in the innate immune defense against bacterial pathogens. Following the opsonization of bacteria by antibodies or complement fragments, neutrophils execute killing through both extracellular traps (NETs) and intracellular phagocytosis. While neutrophils employ a diverse arsenal of granule-resident defenses including proteases, lysozymes, and antimicrobial peptides, their oxidative killing system is paramount to bacterial clearance. This oxidative process is centered on the heme-containing enzyme myeloperoxidase (MPO), the most abundant protein within the neutrophil. Myeloperoxidase as a critical component of the antibacterial arsenal of neutrophils, consumes H₂O₂ as an oxidant to convert halogen and pseudohalogen anions into cytotoxic hypohalous acids. Following phagocytosis by neutrophils, the human pathogen Staphylococcus aureus secretes a potent myeloperoxidase inhibitory protein, called SPIN, as part of its immune evasion repertoire. The matured S. aureus SPIN polypeptide consists of only 73 residues yet contains two functional domains: whereas the 60 residue C-terminal helical bundle domain is responsible for MPO binding, the 13 residue N-terminal domain is required to inhibit MPO. Previous studies informed understanding of the SPIN N-terminal domain, but comparatively little was known about the helical domain insofar as the contribution of individual residues is concerned. To address this limitation, we carried out a residue-level structure/function investigation on the helical bundle domain of S. aureus SPIN. Using sequence conservation and existing structures of SPIN bound to human MPO, we selected residues L49, E50, H51, E52, Y55, and Y75 for interrogation by site-directed mutagenesis. We found that loss of L49 or E52 reduced SPIN activity by roughly an order of magnitude, but that loss of Y55 or H51 caused progressively greater loss of inhibitory potency. Direct binding studies by SPR showed that loss of inhibitory potency in these SPIN mutants resulted from a diminished initial interaction between the inhibitor and MPO. Together, our studies provided new insights into the structure/function relationships of SPIN and identified positions Y55 and H51 as critical determinants of SPIN function. S. aureus SPIN relies on its C-terminal α-helical bundle domain to mediate initial binding to MPO but requires a disordered N-terminal region to fold into a [beta]-hairpin conformation to inhibit MPO activity. To further investigate the structure/function relationship of SPIN, we introduced two cysteine residues into its N-terminal region to trap SPIN in its MPO-bound conformation and characterized the modified protein, which we refer to here as SPIN-CYS. Although control experiments confirmed the presence of the disulfide bond in SPIN-CYS, solution structure determination revealed that the N-terminal region of SPIN-CYS adopted a physically constrained series of lariat-like structures rather than a well-defined [beta]-hairpin. Nevertheless, SPIN-CYS exhibited a gain in inhibitory potency against human MPO when compared to wild-type SPIN. This gain of function persisted even in the presence of deleterious mutations within the C-terminal [alpha]-helical bundle domain. SPR studies showed that the gain in potency arose through an increase in apparent affinity of SPIN-CYS for MPO, which was driven primarily by an increased association rate with MPO when compared to wild-type SPIN. Together, this work provided new information on the coupled binding and folding events required to manifest biological activity of this unusual MPO inhibitor. S. aureus is primarily a human pathogen, and its SPIN protein cannot inhibit MPO from many non-human species. Evaluation of the inhibitory potency of SPIN orthologs against hMPO, revealed that these proteins also exhibited significantly weaker potency compared to SPIN-aureus. This raised questions as to whether staphylococci that prefer alternative mammalian species might produce SPIN proteins with binding preferences for MPO from those hosts. We tested this hypothesis by isolating canine MPO (cMPO) from individual and pooled abscess fluid samples and exploring its inhibition profile by different SPIN orthologs. Consistent in both trials, our results showed that SPIN proteins from known canine pathogens bind more tightly to and inhibit canine MPO more potently than human MPO. We also found that the increased affinity and potency arose from increases in the second order (i.e. association rate constant (ka)) rate constant. Comparison of the prototypic of these inhibitors, SPIN-delphini, bound to canine MPO when compared to human MPO, explains the physical basis for our observation and provides support for our hypothesis. This work alongside our previous studies on SPIN/MPO interaction can be developed into a valuable model system for understanding the structure/function principles that underlie host-specific evolution of virulence proteins.