Probing inhibitors binding to human urokinase crystals by Raman microscopy: implications for compound screening

Inhibition of urokinase activity represents a promising target for antimetastatic therapy for several types of tumor. The present study sets out to investigate the potential of Raman spectroscopy for defining the molecular details of inhibitor binding to this enzyme, with emphasis on single crystal studies. It is demonstrated that high quality Raman spectra from a series of five inhibitors bound individually to the active site of human urokinase can be obtained in situ from urokinase single crystals in hanging drops by using a Raman microscope. After recording the spectrum of the free crystal, a solution of inhibitor containing an amidine functional group on a naphthalene ring was added, and the spectrum of the crystal-inhibitor complex was obtained. The resulting difference Raman spectrum contained only vibrational modes due to bound inhibitor, originating from the protonated group, i.e., the amidinium moiety, as well as naphthalene ring modes and features from other functionalities that made up each inhibitor. The identification of the amidinium modes was placed on a quantitative basis by experimental and theoretical work on naphthamidine compounds. For the protonated group, -C-(NH2)(2)(+), the symmetric stretch occurs near 1520 cm(-1), and a less intense antisymmetric mode appears in the Raman spectra near 1680 cm(-1). The presence of vibrational modes near 1520 cm(-1) in each of the Raman difference spectra of the five complexes examined unambiguously identifies the protonated form of the amidinium group in the active site. Several advantages were found for single crystal experiments over solution studies of inhibitor-enzyme complexes, and these are discussed. The use of single crystals permits competitive binding experiments that cannot be undertaken in solution in any kind of homogeneous assay format. The Raman difference spectrum for a single crystal that had been exposed to equimolar amounts of all five inhibitors in the hanging drop showed only the Raman signature of the compound with the lowest K(i). These findings suggest that the Raman approach may offer a route in the screening of compounds in drug design applications as well as an adjunct to crystallographic analysis.

Automated crystal mounting and data collection for protein crystallography

To increase the efficiency of diffraction data collection for protein crystallographic studies, an automated system designed to store frozen protein crystals, mount them sequentially, align them to the X-ray beam, collect complete data sets, and return the crystals to storage has been developed. Advances in X-ray data collection technology including more brilliant X-ray sources, improved focusing optics, and faster-readout detectors have reduced diffraction data acquisition times from days to hours at a typical protein crystallography laboratory [1,2]. In addition, the number of high-brilliance synchrotron X-ray beam lines dedicated to macromolecular crystallography has increased significantly, and data collection times at these facilities can be routinely less than an hour per crystal. Because the number of protein crystals that may be collected in a 24 hr period has substantially increased, unattended X-ray data acquisition, including automated crystal mounting and alignment, is a desirable goal for protein crystallography. The ability to complete X-ray data collection more efficiently should impact a number of fields, including the emerging structural genomics field [3], structure-directed drug design, and the newly developed screening by X-ray crystallography [4], as well as small molecule applications.

Discovering novel ligands for macromolecules using X-ray crystallographic screening

The need to decrease the time scale for clinical compound discovery has led to innovations at several stages in the process, including genomics/proteomics for target identification, ultrahigh-throughput screening for lead identification, and structure-based drug design and combinatorial chemistry for lead optimization. A critical juncture in the process is the identification of a proper lead compound, because a poor choice may generate costly difficulties at later stages. Lead compounds are commonly identified from high-throughput screens of large compound libraries, derived from known substrates/inhibitors, or identified in computational prescreeusing X-ray crystal structures. Structural information is often consulted to efficiently optimize leads, but under the current paradigm, such data require preidentification and confirmation of compound binding. Here, we describe a new X-ray crystallography-driven screening technique that combines the steps of lead identification, structural assessment, and optimization. The method is rapid, efficient, and high-throughput, and it results in detailed crystallographic structure information. The utility of the method is demonstrated in the discovery and optimization of a new orally available class of urokinase inhibitors for the treatment of cancer.

Structure-directed discovery of potent non-peptidic inhibitors of human urokinase that access a novel binding subsite

BACKGROUND

Human urokinase-type plasminogen activator has been implicated in the regulation and control of basement membrane and interstitial protein degradation. Because of its role in tissue remodeling, urokinase is a central player in the disease progression of cancer, making it an attractive target for design of an anticancer clinical agent: Few urokinase inhibitors have been described, which suggests that discovery of such a compound is in the early stages. Towards integrating structural data into this process, a new human urokinase crystal form amenable to structure-based drug design has been used to discover potent urokinase inhibitors.

RESULTS

On the basis of crystallographic data, 2-naphthamidine was chosen as the lead scaffold for structure-directed optimization. This co-crystal structure shows the compound binding at the primary specificity pocket of the trypsin-like protease and at a novel binding subsite that is accessible from the 8-position of 2-napthamidine. This novel subsite was characterized and used to design two compounds with very different 8-substituents that inhibit urokinase with K(i) values of 30-40 nM.

CONCLUSIONS

Utilization of a novel subsite yielded two potent urokinase inhibitors even though this site has not been widely used in inhibitor optimization with other trypsin-like proteases, such as those reported for thrombin or factor Xa. The extensive binding pockets present at the substrate-binding groove of these other proteins are blocked by unique insertion loops in urokinase, thus necessitating the utilization of additional binding subsites. Successful implementation of this strategy and characterization of the novel site provides a significant step towards the discovery of an anticancer clinical agent.

Thrombin inhibitor design: X-ray and solution studies provide a novel P1 determinant

The crystal structures of proflavin and 6-fluorotryptamine thrombin have been completed showing binding of both ligands at the active site S1 pocket. The structure of proflavin:thrombin was confirmatory, while the structure of 6-fluorotryptamine indicated a novel binding mode at the thrombin active site. Furthermore, speculation that the sodium atom identified in an extended solvent channel beneath the S pocket may play a role in binding of these ligands was investigated by direct proflavin titrations as well as chromogenic activity measurements as a function of sodium concentration at constant ionic strength. These results suggested a linkage between the sodium site and the S1 pocket. This observation could be due to a simple ionic interaction between Asp189 and the sodium ion or a more complicated structural rearrangement of the thrombin S1 pocket. Finally, the unique binding mode of 6-fluorotryptamine provides ideas toward the design of a neutrally charged thrombin inhibitor.

Atomic structures of two nitroxide spin labels complexed with human thrombin: comparison with solution studies

Crystal structures of thrombin complexed with two spin labels called para-V, 4-(2,2,5,5-tetramethylpyrrolidine-1-oxyl)-p-(fluorosulfonyl) benzamidine, and meta-V, 3-(2,2,5,5-tetramethyl-pyrrolidine1-oxyl)-m-(fluorosulfonyl) benzamidine, have been completed at 2.0 and 3.0 A resolution, respectively. Previous electron spin resonance studies with these labels gave rise to a low-resolution "topography map" of thrombin's extended active site. These labels monitor two distinct areas of the thrombin active site: (1) an apolar binding site which manifests itself in an biphasic activation/inhibition effect on thrombin activity and (2) a region sensitive to alpha-thrombin autoproteolytic cleavage(s) to gamma-thrombin (Arg75-Tyr76 and/or Arg77A-Asn78, and Lys149E-Gly150, chymotrypsin numbering). Para-V was found to bind along the substrate binding cleft, while meta-V was found to bind both at the substrate primary specificity pocket and at a site which interacts with the gamma-cleavage loop. These studies reaffirm that accurate information may be gained from solution studies and indicates the complementarity of solid-state studies.

Crystal structure of ErmC', an rRNA methyltransferase which mediates antibiotic resistance in bacteria

The prevalent mechanism of bacterial resistance to erythromycin and other antibiotics of the macrolide-lincosamide-streptogramin B group (MLS) is methylation of the 23S rRNA component of the 50S subunit in bacterial ribosomes. This sequence-specific methylation is catalyzed by the Erm group of methyltransferases (MTases). They are found in several strains of pathogenic bacteria, and ErmC is the most studied member of this class. The crystal structure of ErmC' (a naturally occurring variant of ErmC) from Bacillus subtilis has been determined at 3.0 A resolution by multiple anomalous diffraction phasing methods. The structure consists of a conserved alpha/beta amino-terminal domain which binds the cofactor S-adenosyl-l-methionine (SAM), followed by a smaller, alpha-helical RNA-recognition domain. The beta-sheet structure of the SAM-binding domain is well-conserved between the DNA, RNA, and small-molecule MTases. However, the C-terminal nucleic acid binding domain differs from the DNA-binding domains of other MTases and is unlike any previously reported RNA-recognition fold. A large, positively charged, concave surface is found at the interface of the N- and C-terminal domains and is proposed to form part of the protein-RNA interaction surface. ErmC' exhibits the conserved structural motifs previously found in the SAM-binding domain of other methyltransferases. A model of SAM bound to ErmC' is presented which is consistent with the motif conservation among MTases.

Structure-based understanding of ligand affinity using human thrombin as a model system

Kinetic study of a series of compounds containing the thrombin-directed peptide D-Phe-ProboroArg-OH had indicated that the structure of the N-terminal blocking group may be correlated with binding [Kettner, C., Mersinger, L., & Knabb, R. (1990) J. Biol. Chem. 265, 18289-18297]. In order to further study this phenomenon, a second series of compounds that contains a C-terminal methyl ester in place of the boronic acid was synthesized, binding measured, and the three-dimensional structure in complex with human thrombin determined by X-ray crystallography. Incubation of Ac-D-Phe-Pro-Arg-OMe, Boc-D-Phe-Pro-Arg-OMe, and H-D-Phe-Pro-Arg-OMe resulted in the formation of thrombin-product complexes within the crystal. Ki values for the corresponding products (free carboxylic acids) were 60 +/- 12 microM, 7.8 +/- 0.1 microM, 0.58 +/- 0.02 microM, respectively, indicating that the nature of the N-terminal blocking group has a significant effect on affinity. Examination of the crystal structures indicated that the higher affinity of the H-D-Phe peptide is due to rearrangement of one residue comprising the S3 site (Glu192) in order to maximize electrostatic interactions with the "NH3(+)-" of H-D-Phe. The relative affinity of Boc-D-Phe-Pro-Arg-OH is due to favorable hydrophobic interactions between thrombin and the bulky butyl group. However, this results in less favorable binding of Arg-P1 in the oxyanion hole as shown by long hydrogen-bonding distances. This work gave rise to some general observations applicable to structure-based drug design: (1) altering the structure of an inhibitor at one site can affect binding at an unchanged distal site; (2) minor alteration of inhibitor structure can lead to small, but significant reorganization of neighboring protein structure; (3) these unexpected reorganizations can define alternate binding motifs.

A glutamic acid specific serine protease utilizes a novel histidine triad in substrate binding

Proteases specific for cleavage after acidic residues have been implicated in several disease states, including epidermolysis, inflammation, and viral processing. A serine protease with specificity toward glutamic acid substrates (Glu-SGP) has been crystallized in the presence of a tetrapeptide ligand and its structure determined and refined to an R-factor of 17% at 2.0-A resolution. This structure provides an initial description of the design of proteolytic specificity for negatively charged residues. While the overall fold of Glu-SGP closely resembles that observed in the pancreatic-type serine proteases, stabilization of the negatively charged substrate when bound to this protein appears to involve a more extensive part of the protease than previously observed. The substrate carboxylate is bound to a histidine side chain, His213, which provides the primary electrostatic compensation of the negative charge on the substrate, and to two serine hydroxyls, Ser192 and Ser216. Glu-SGP displays maximum activity at pH 8.3, and assuming normal pKa's, the glutamate side chain and His213 will be negatively charged and neutral, respectively, at this pH. In order for His213 to carry a positive charge at the optimal pH, its pKa will have to be raised by at least two units. An alternative mechanism for substrate charge compensation is suggested that involves a novel histidine triad, His213, His199, and His228, not observed in any other serine protease. The C-terminal alpha-helix, ubiquitous to all pancreatic-type proteases, is directly linked to this histidine triad and may also play a role in substrate stabilization.(ABSTRACT TRUNCATED AT 250 WORDS)

Conformational similarities between one-chain and two-chain tissue plasminogen activator (t-PA): implications to the activation mechanism on one-chain t-PA

Tissue plasminogen activator (t-PA) is an exceptional serine protease, because unlike most other serine protease zymogens single-chain tissue plasminogen activator (sct-PA) possesses a substantial amount of proteolytic activity. The unusual reaction of sct-PA afforded the opportunity to directly compare the active site environment of sct-PA and two-chain tissue plasminogen activator (tct-PA) in solution through the application of a series of nitroxide spin labels and fluorophores. These labels, which have been previously shown to covalently label the catalytic serine of other serine proteases, inactivated both sct-PA and tct-PA. The labels can be divided into two classes: those which form tetrahedral complexes (sulfonates) and those which form trigonal complexes (anthranilates). Those which formed tetrahedral complexes were found to be insensitive to structural differences between sct-PA and tct-PA at the active site. In contrast, those which formed trigonal complexes could differentiate and monitor the sct-PA to tct-PA conversion by fluorescence spectroscopy. Models of the structure of sct-PA and tct-PA were constructed on the basis of the known X-ray structures of other serine protease zymogen and active enzyme forms. One of the nitroxide spin labels was modeled into the sct-PA and tct-PA structures in two possible orientations, both of which could be sensitive to structural differences between sct-PA and tct-PA. These models formed the structural rationale used to explain the results obtained with the "tetrahedral" and "trigonal" probes, as well as to offer a possible explanation for the unique reactivity of sct-PA.

Subtle differences in active site structure between bovine and human thrombins: ESR and fluorescence studies

The primary structures of bovine and human alpha-thrombins are highly homologous yet their x-ray structures are not yet complete enough to distinguish differences. In order to probe and compare their dynamic conformations in solution, we examined bovine and human alpha-thrombins with a series of active site directed fluorosulfonylphenyl spin labeled inhibitors and fluorophores which probe a region within 10-15 A of the catalytic serine residue. Overall, the nitroxide moieties were more immobilized in the bovine vs human derivatives reflecting either more apolar binding regions or steric obstructions to the motion of the nitroxide in bovine thrombin. Most of the labels which distinguish indole (apolar ligand) binding in human thrombin were found to display similar interactions in bovine thrombin, although slight differences in the general topography of this region were suggested. The two active site directed fluorophores, dansyl fluoride and p-nitrophenyl anthranilate showed differences in both lambda emmax and lambda exmax of the complexes with bovine and human-alpha-thrombin, respectively, Several of the effects observed i.e., ligand binding (indole or benzamidine) and the subtle hydrophobic interactions between the nitroxide moiety and the protein active site would be difficult to assess from an x-ray structure determination alone.