Fragment-based and structure-guided discovery of perforin inhibitors

Perforin is a pore-forming protein whose normal function enables cytotoxic T and natural killer (NK) cells to kill virus-infected and transformed cells. Conversely, unwanted perforin activity can also result in auto-immune attack, graft rejection and aberrant responses to pathogens. Perforin is critical for the function of the granule exocytosis cell death pathway and is therefore a target for drug development. In this study, by screening a fragment library using NMR and surface plasmon resonance, we identified 4,4-diaminodiphenyl sulfone (dapsone) as a perforin ligand. We also found that dapsone has modest (mM) inhibitory activity of perforin lytic activity in a red blood cell lysis assay in vitro. Sequential modification of this lead fragment, guided by structural knowledge of the ligand binding site and binding pose, and supported by SPR and ligand-detected 19F NMR, enabled the design of nanomolar inhibitors of the cytolytic activity of intact NK cells against various tumour cell targets. Interestingly, the ligands we developed were largely inert with respect to direct perforin-mediated red blood cell lysis but were very potent in the context of perforin's action on delivering granzymes in the immune synapse, the context in which it functions physiologically. Our work indicates that a fragment-based, structure-guided drug discovery strategy can be used to identify novel ligands that bind perforin. Moreover, these molecules have superior physicochemical properties and solubility compared to previous generations of perforin ligands.

Structure and Function Characterization of the a1a2 Motifs of Streptococcus pyogenes M Protein in Human Plasminogen Binding

Plasminogen (Plg)-binding M protein (PAM) is a group A streptococcal cell surface receptor that is crucial for bacterial virulence. Previous studies revealed that, by binding to the kringle 2 (KR2) domain of host Plg, the pathogen attains a proteolytic microenvironment on the cell surface that facilitates its dissemination from the primary infection site. Each of the PAM molecules in their dimeric assembly consists of two Plg binding motifs (called the a1 and a2 repeats). To date, the molecular interactions between the a1 repeat and KR2 have been structurally characterized, whereas the role of the a2 repeat is less well defined. Here, we report the 1.7-Å x-ray crystal structure of KR2 in complex with a monomeric PAM peptide that contains both the a1 and a2 motifs. The structure reveals how the PAM peptide forms key interactions simultaneously with two KR2 via the high-affinity lysine isosteres within the a1a2 motifs. Further studies, through combined mutagenesis and functional characterization, show that a2 is a stronger KR2 binder than a1, suggesting that these two motifs may play discrete roles in mediating the final PAM-Plg assembly.

A Cyclic Peptide Inhibitor of the iNOS-SPSB Protein-Protein Interaction as a Potential Anti-Infective Agent

SPRY domain- and SOCS box-containing proteins SPSB1, SPSB2, and SPSB4 interact with inducible nitric oxide synthase (iNOS), causing the iNOS to be polyubiquitinated and targeted for degradation. Inhibition of this interaction increases iNOS levels, and consequently cellular nitric oxide (NO) concentrations, and has been proposed as a potential strategy for killing intracellular pathogens. We previously described two DINNN-containing cyclic peptides (CP1 and CP2) as potent inhibitors of the murine SPSB-iNOS interaction. In this study, we report the crystal structures of human SPSB4 bound to CP1 and CP2 and human SPSB2 bound to CP2. We then used these structures to design a new inhibitor in which an intramolecular hydrogen bond was replaced with a hydrocarbon linkage to form a smaller macrocycle while maintaining the bound geometry of CP2 observed in the crystal structures. This resulting pentapeptide SPSB-iNOS inhibitor (CP3) has a reduced macrocycle ring size, fewer nonbinding residues, and includes additional conformational constraints. CP3 has a greater affinity for SBSB2 ( K_D = 7 nM as determined by surface plasmon resonance) and strongly inhibits the SPSB2-iNOS interaction in macrophage cell lysates. We have also determined the crystal structure of CP3 in complex with human SPSB2, which reveals the structural basis for the increased potency of CP3 and validates the original design.

The Single Disulfide-Directed β-Hairpin Fold. Dynamics, Stability, and Engineering

Grafting bioactive peptide sequences onto small cysteine-rich scaffolds is a promising strategy for enhancing their stability and value as novel peptide-based therapeutics. However, correctly folded disulfide-rich peptides can be challenging to produce by either recombinant or synthetic means. The single disulfide-directed β-hairpin (SDH) fold, first observed in contryphan-Vc1, provides a potential alternative to complex disulfide-rich scaffolds. We have undertaken recombinant production of full-length contryphan-Vc1 (rCon-Vc1[Z1Q]) and a truncated analogue (rCon-Vc1_1-22[Z1Q]), analyzed the backbone dynamics of rCon-Vc1[Z1Q], and probed the conformational and proteolytic stability of these peptides to evaluate the potential of contryphan-Vc1 as a molecular scaffold. Backbone ¹⁵N relaxation measurements for rCon-Vc1[Z1Q] indicate that the N-terminal domain of the peptide is ordered up to Thr19, whereas the remainder of the C-terminal region is highly flexible. The solution structure of truncated rCon-Vc1_1-22[Z1Q] was similar to that of the full-length peptide, indicating that the flexible C-terminus does not have any effect on the structured domain of the peptide. Contryphan-Vc1 exhibited excellent proteolytic stability against trypsin and chymotrypsin but was susceptible to pepsin digestion. We have investigated whether contryphan-Vc1 can accept a bioactive epitope while maintaining the structure of the peptide by introducing peptide sequences based on the DINNN motif of inducible nitric oxide synthase. We show that sCon-Vc1_1-22[NNN_12-14] binds to the iNOS-binding protein SPSB2 with an affinity of 1.3 μM while maintaining the SDH fold. This study serves as a starting point in utilizing the SDH fold as a peptide scaffold.

Structure and activity of contryphan-Vc2: Importance of the d-amino acid residue

In natural proteins and peptides, amino acids exist almost invariably as l-isomers. There are, however, several examples of naturally-occurring peptides containing d-amino acids. In this study we investigated the role of a naturally-occurring d-amino acid in a small peptide identified in the transcriptome of a marine cone snail. This peptide belongs to a family of peptides known as contryphans, all of which contain a single d-amino acid residue. The solution structure of this peptide was solved by NMR, but further investigations with molecular dynamics simulations suggest that its solution behaviour may be more dynamic than suggested by the NMR ensemble. Functional tests in mice uncovered a novel bioactivity, a depressive phenotype that contrasts with the hyperactive phenotypes typically induced by contryphans. Trp3 is important for bioactivity, but this role is independent of the chirality at this position. The d-chirality of Trp3 in this peptide was found to be protective against enzymatic degradation. Analysis by NMR and molecular dynamics simulations indicated an interaction of Trp3 with lipid membranes, suggesting the possibility of a membrane-mediated mechanism of action for this peptide.

Molecular Insights into the Interaction Between the SPRY Domain-Containing SOCS Box Protein SPSB2 and Peptides Based on the Binding Motif from iNOS

SPRY domain-containing SOCS box proteins SPSB1, 2, and 4 mediate the proteasomal degradation of inducible nitric oxide synthase (iNOS) and thereby modulate the amount of NO available for combating infectious organisms. A highly conserved Asp-Ile-Asn-Asn-Asn (DINNN) motif found at the N-terminus of iNOS binds to SPSB2 with nanomolar affinity. The design of specific and potent inhibitors of iNOS–SPSB interactions will be aided by a better understanding of the interactions of this DINNN sequence with SPSB2. Although crystal structures of SPSB complexes with DINNN peptides are available, aspects of the interaction between peptide and protein are still not fully understood. Here, our results from surface plasmon resonance and NMR spectroscopy indicate that residues flanking the DINNN motif, which make no direct contact with SPSB2 in the available crystal structures, nonetheless play an important role in enhancing the binding affinity to SPSB2, by up to 80-fold. Mutational analysis of the DINNN sequence showed that mutation of the Asp or the first Asn residue to Ala reduced the binding affinity by 200- or 600-fold respectively, whereas mutation of the third Asn made binding undetectable. Ala substitution of the second Asn residue caused a 30-fold drop in binding affinity. Substitution of the Ile had very little effect on the binding affinity and substitutions with bulky residues were tolerated. This provides an opportunity for further modification for therapeutic applications. These results highlight the complex interplay of peptide sequence and protein binding and inform efforts to design peptide therapeutics to disrupt the iNOS–SPSB interaction.

Applications of (19)F-NMR in Fragment-Based Drug Discovery

(19)F-NMR has proved to be a valuable tool in fragment-based drug discovery. Its applications include screening libraries of fluorinated fragments, assessing competition among elaborated fragments and identifying the binding poses of promising hits. By observing fluorine in both the ligand and the target protein, useful information can be obtained on not only the binding pose but also the dynamics of ligand-protein interactions. These applications of (19)F-NMR will be illustrated in this review with studies from our fragment-based drug discovery campaigns against protein targets in parasitic and infectious diseases.

Structural Basis for Ca2+-mediated Interaction of the Perforin C2 Domain with Lipid Membranes

Natural killer cells and cytotoxic T-lymphocytes deploy perforin and granzymes to kill infected host cells. Perforin, secreted by immune cells, binds target membranes to form pores that deliver pro-apoptotic granzymes into the target cell. A crucial first step in this process is interaction of its C2 domain with target cell membranes, which is a calcium-dependent event. Some aspects of this process are understood, but many molecular details remain unclear. To address this, we investigated the mechanism of Ca(2+) and lipid binding to the C2 domain by NMR spectroscopy and x-ray crystallography. Calcium titrations, together with dodecylphosphocholine micelle experiments, confirmed that multiple Ca(2+) ions bind within the calcium-binding regions, activating perforin with respect to membrane binding. We have also determined the affinities of several of these binding sites and have shown that this interaction causes a significant structural rearrangement in CBR1. Thus, it is proposed that Ca(2+) binding at the weakest affinity site triggers changes in the C2 domain that facilitate its interaction with lipid membranes.

Structure and Functional Characterization of the Conserved JAK Interaction Region in the Intrinsically Disordered N-Terminus of SOCS5

SOCS5 can negatively regulate both JAK/STAT and EGF-receptor pathways and has therefore been implicated in regulating both the immune response and tumorigenesis. Understanding the molecular basis for SOCS5 activity may reveal novel ways to target key components of these signaling pathways. The N-terminal region of SOCS5 coordinates critical protein interactions involved in inhibition of JAK/STAT signaling, and a conserved region within the N-terminus of SOCS5 mediates direct binding to the JAK kinase domain. Here we have characterized the solution conformation of this conserved JAK interaction region (JIR) within the largely disordered N-terminus of SOCS5. Using nuclear magnetic resonance (NMR) chemical shift analysis, relaxation measurements, and NOE analysis, we demonstrate the presence of preformed structural elements in the JIR of mouse SOCS5 (mSOCS5175-244), consisting of an α-helix encompassing residues 224-233, preceded by a turn and an extended structure. We have identified a phosphorylation site (Ser211) within the JIR of mSOCS5 and have investigated the role of phosphorylation in modulating JAK binding using site-directed mutagenesis.

Promiscuous 2-aminothiazoles (PrATs): a frequent hitting scaffold

We have identified a class of molecules, known as 2-aminothiazoles (2-ATs), as frequent-hitting fragments in biophysical binding assays. This was exemplified by 4-phenylthiazol-2-amine being identified as a hit in 14/14 screens against a diverse range of protein targets, suggesting that this scaffold is a poor starting point for fragment-based drug discovery. This prompted us to analyze this scaffold in the context of an academic fragment library used for fragment-based drug discovery (FBDD) and two larger compound libraries used for high-throughput screening (HTS). This analysis revealed that such "promiscuous 2-aminothiazoles" (PrATs) behaved as frequent hitters under both FBDD and HTS settings, although the problem was more pronounced in the fragment-based studies. As 2-ATs are present in known drugs, they cannot necessarily be deemed undesirable, but the combination of their promiscuity and difficulties associated with optimizing them into a lead compound makes them, in our opinion, poor scaffolds for fragment libraries.

Expression and isotopic labelling of the potassium channel blocker ShK toxin as a thioredoxin fusion protein in bacteria

The polypeptide toxin ShK is a potent blocker of Kv1.3 potassium channels, which play a crucial role in the activation of human effector memory T-cells (T(EM)). Selective blockers constitute valuable therapeutic leads for the treatment of autoimmune diseases mediated by T(EM) cells, such as multiple sclerosis, rheumatoid arthritis, and type-1 diabetes. We have established a recombinant peptide expression system in order to generate isotopically-labelled ShK and various ShK analogues for in-depth biophysical and pharmacological studies. ShK was expressed as a thioredoxin fusion protein in Escherichia coli BL21 (DE3) cells and purified initially by Ni²⁺ iminodiacetic acid affinity chromatography. The fusion protein was cleaved with enterokinase and purified to homogeneity by reverse-phase HPLC. NMR spectra of ¹⁵N-labelled ShK were similar to those reported previously for the unlabelled synthetic peptide, confirming that recombinant ShK was correctly folded. Recombinant ShK blocked Kv1.3 channels with a K(d) of 25 pM and inhibited the proliferation of human and rat T lymphocytes with a preference for T(EM) cells, with similar potency to synthetic ShK in all assays. This expression system also enables the efficient production of ¹⁵N-labelled ShK for NMR studies of peptide dynamics and of the interaction of ShK with Kv1.3 channels.

Conformational changes in a plant ketol-acid reductoisomerase upon Mg(2+) and NADPH binding as revealed by two crystal structures

Ketol-acid reductoisomerase (KARI; EC 1.1.1.86) is an enzyme in the branched-chain amino acid biosynthesis pathway where it catalyzes the conversion of 2-acetolactate into (2R)-2,3-dihydroxy-3-isovalerate or the conversion of 2-aceto-2-hydroxybutyrate into (2R,3R)-2,3-dihydroxy-3-methylvalerate. KARI catalyzes two reactions-alkyl migration and reduction-and requires Mg(2+) and NADPH for activity. To date, the only reported structures for a plant KARI are those of the spinach enzyme-Mn(2+)-(phospho)ADP ribose-(2R,3R)-2,3-dihydroxy-3-methylvalerate complex and the spinach KARI-Mg(2)(+)-NADPH-N-hydroxy-N-isopropyloxamate complex, where N-hydroxy-N-isopropyloxamate is a predicted transition-state analog. These studies demonstrated that the enzyme consists of two domains, N-domain and C-domain, with the active site at the interface of these domains. Here, we have determined the structures of the rice KARI-Mg(2+) and rice KARI-Mg(2)(+)-NADPH complexes to 1.55 A and 2.80 A resolutions, respectively. In comparing the structures of all the complexes, several differences are observed. Firstly, the N-domain is rotated up to 15 degrees relative to the C-domain, expanding the active site by up to 4 A. Secondly, an alpha-helix in the C-domain that includes residues V510-T519 and forms part of the active site moves by approximately 3.9 A upon binding of NADPH. Thirdly, the 15 C-terminal amino acid residues in the rice KARI-Mg(2+) complex are disordered. In the rice KARI-Mg(2)(+)-NADPH complex and the spinach KARI structures, many of the 15 residues bind to NADPH and the N-domain and cover the active site. Fourthly, the location of the metal ions within the active site can vary by up to 2.7 A. The new structures allow us to propose that an induced-fit mechanism operates to (i) allow substrate to enter the active site, (ii) close over the active site during catalysis, and (iii) open the active site to facilitate product release.

Human tartrate-resistant acid phosphatase becomes an effective ATPase upon proteolytic activation

Proteolytic cleavage in an exposed loop of human tartrate-resistant acid phosphatase (TRAcP) with trypsin leads to a significant increase in activity. At each pH value between 3.25 and 8.0 the cleaved enzyme is more active. Substrate specificity is also influenced by proteolysis. Only the cleaved form is able to hydrolyze unactivated substrates efficiently, and at pH >6 cleaved TRAcP acquires a marked preference for ATP. The cleaved enzyme also has altered sensitivity to inhibitors. Interestingly, the magnitude and mode of inhibition by fluoride depends not only on the proteolytic state but also pH. The combined kinetic data imply a role of the loop residue D158 in catalysis in the cleaved enzyme. Notably, at low pH this residue may act as a proton donor for the leaving group. In this respect the mechanism of cleaved TRAcP resembles that of sweet potato purple acid phosphatase.