Crystals of LuxP and complexes thereof

Description

RELATED APPLICATION INFORMATION

This application claims priority to U.S. Provisional Application Ser. No. 60/314,705, filed Aug. 24, 2001, which is hereby incorporated by reference in its entirety.

This application is related to U.S. patent application Ser. No. 10/227,327 filed Aug. 22, 2002, which is hereby incorporated by reference in its entirety.

FEDERAL FUNDING

This invention was funded in part through grants from the National Institutes of Health, the National Science Foundation, and the Office of Naval Research. Therefore, the federal government has certain rights in this invention.

BACKGROUND OF THE INVENTION

1. Field of the Invention

This invention relates to the LuxP protein, and more particularly to a crystalline form of the LuxP protein having sufficient crystal quality to allow crystallographic data to be obtained, as well as to methods of structure-based drug design based on the crystallographic data.

2. Description of the Related Art

Cell-cell communication in bacteria occurs through the exchange of extracellular signaling compounds called autoinducers. This process, termed quorum sensing, allows bacterial populations to coordinate gene expression. Community cooperation likely enhances the effectiveness of processes including bioluminescence, virulence factor expression, antibiotic production, and biofilm development. Unlike other autoinducers, which are specific to particular species of bacteria, a recently discovered autoinducer (AI-2) is produced by a large number of bacterial species. AI-2 has been proposed to serve as a universal signal for inter-species communication.

AI-2 was originally identified in the bioluminescent marine bacterium Vibrio harveyi as one of two autoinducers that regulate light production in response to cell density. The synthase required for AI-2 production, LuxS, is widely conserved among gram-negative and -positive bacteria. Bacteria produce AI-2 from S-adenosylmethionine in several enzymatic steps, as shown in FIG. 1. Consumption of S-adenosylmethionine as a methyl donor produces S-adenosylhomocysteine, which undergoes hydrolysis catalyzed by the nucleosidase Pfs to yield adenine and S-ribosylhomocysteine. Subsequently LuxS catalyzes cleavage of S-ribosylhomocysteine to homocysteine and 4,5-dihydroxy-2,3-pentanedione, which can then cyclize either by itself or in an enzyme-mediated process.

Detection of AI-2 by V. harveyi involves two proteins, LuxP and LuxQ. LuxP belongs to a large family of periplasmic binding proteins whose members bind diverse ligands, while LuxQ is a two-component hybrid sensor kinase embedded in the bacterial inner membrane. It is believed that LuxP is the primary AI-2 receptor, see X. Chen, S. Schauder, N. Potier, A. Van Dorsselaer, I. Pelczer, B. Bassler, and F. Hughson, “Structural Identification of a Bacterial Quorum-Sensing Signal Containing Boron,” Nature, Vol. 415, pp. 545–549 (2002). In quorum sensing, the periplasmic LuxP-AI-2 complex likely interacts with LuxQ to transduce the autoinducer signal. However, heretofore LuxP has not been isolated in crystalline form suitable for structural determination by X-ray crystallography, and thus neither the crystal structure of LuxP nor the LuxP binding site for AI-2 were known prior to the instant invention.

SUMMARY OF THE INVENTION

A preferred embodiment provides a crystal comprising LuxP. The crystal may comprise a LuxP-ligand complex. Preferably, the ligand comprises boron and/or a furan moiety. AI-2 is an example of a preferred ligand.

Another preferred embodiment provides a method of using the crystal to identify whether a ligand binds to LuxP, comprising: obtaining the atomic coordinates in the crystal of at least a selected portion of LuxP; using the atomic coordinates to model the selected portion; identifying a potential ligand; and docking the potential ligand to the selected portion of LuxP. Preferably, the selected portion comprises an amino acid residue selected from the group consisting of Trp 82, Gln 77, Ser 79, Asp 267, Thr 266, Trp 289, Arg 310, Arg 215, and Asn 159 according to Table 1. Another preferred embodiment provides ligands identified by this method. A pharmaceutical composition comprising such a ligand is another preferred embodiment, and methods of treating bacterial infections by administering such pharmaceutical compositions to humans are another preferred embodiment.

These and other embodiments are described in greater detail below.

BRIEF DESCRIPTION OF THE DRAWINGS

Various aspects of the invention will be readily apparent from the following description and from the appended drawings, which are meant to illustrate and not to limit the invention, and wherein:

FIG. 1 illustrates several steps in the pathway of the production of AI-2 from S-adenosylmethionine.

FIG. 2 illustrates the structure of a LuxP-AI-2 complex.

FIG. 3 shows the structure of LuxP and AI-2 at a binding site.

FIG. 4 illustrates a route for the conversion of 4,5-dihydroxy-2,3-pentanedione to AI-2.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

A preferred embodiment provides a crystal comprising LuxP (SEQ ID NO: 1). In this context, those skilled in the art will understand that the term “crystal” refers to an ordered arrangement of atoms, the crystal having an overall size and quality sufficient for the elucidation of the atomic arrangement by X-ray crystallography. Preferably, the crystal diffracts X-rays to a resolution of greater than about 5.0 Angstroms (Å), more preferably greater than about 2.8 Å, even more preferably greater than about 1.5 Å. Those skilled in the art will understand that a resolution “greater than” a particular value means a resolution that numerically exceeds the recited value. For example, in the language of X-ray crystallography, a resolution of 2.8 Å is greater than a resolution of 5.0 Å. Crystals comprising LuxP are preferably prepared by the methods described in the Examples below. The atomic coordinates for LuxP are preferably determined by X-ray crystallography of a crystal comprising LuxP, preferably by the methods described in the Examples below. A set of atomic coordinates obtained by these methods for a crystal comprising LuxP appears in Table 1.

A preferred crystal comprises LuxP and a ligand. Preferably, the ligand comprises a furan moiety and/or a boron atom. In this context, those skilled in the art will understand that the term “ligand” refers to a molecule or ion that binds to LuxP. Preferably, binding between the ligand and LuxP occurs at a LuxP binding site. In this context, those skilled in the art will understand that the term “binding site” refers to a region of LuxP that favorably associates with a ligand, thus producing a LuxP-ligand complex in which the ligand binds relatively tightly to LuxP. Such strong binding may be produced, for example, when the shapes of the binding site and ligand are mutually compatible (e.g., “lock and key”), and/or when at least some of the ligand atoms are attracted to at least some of the LuxP atoms in the vicinity of the binding site by intermolecular forces, e.g., dipole-dipole interactions, Van der Waals attractions, hydrogen-bonding, etc.

Binding sites have significant utility in fields such as drug discovery. The association of natural ligands with the binding sites of their corresponding proteins, enzymes or receptors is the basis of many biological mechanisms of action. Similarly, many drugs exert their biological effects through association with the binding sites of proteins, enzymes, and receptors. Such associations may occur with all or any parts of the binding site. An understanding of such associations enables the design of drugs having more favorable associations with their target proteins, enzymes or receptors, and thus, improved biological effects. Therefore, this information is valuable in designing potential inhibitors of the binding sites of biologically important targets.

A preferred crystal comprises LuxP and AI-2, where the latter is an autoinducer that the bioluminescent marine bacterium Vibrio harveyi uses to regulate light production in response to cell density. It is believed that AI-2 has the following structure (I):

FIG. 2 shows a LuxP complex in which the compound represented by structure (I) (herein referred to as AI-2) occupies a LuxP binding site. FIG. 3 shows this binding site in greater detail. It is apparent from FIG. 3 that numerous polar interactions stabilize the two fused five-membered rings within the LuxP binding site. The positively charged side chains of Arg 215 and Arg 310 hydrogen-bond with three of the four borate oxygen atoms and likely stabilize the negative charge on the borate. As shown in FIG. 3, this binding site comprises at least one of the amino acid residues selected from the group consisting of Trp 82, Gln 77, Ser 79, Asp 267, Thr 266, Trp 289, Arg 310, Arg 215, and Asn 159, according to Table 1.

The LuxP X-ray crystallographic data shown in Table 1 may be used to identify whether a ligand binds to LuxP, and thus may be used for a variety of purposes, such as drug discovery. A preferred method comprises obtaining the atomic coordinates in the crystal of at least a selected portion of LuxP. Preferably, the selected portion comprises a binding site. More preferably, the selected portion comprises an amino acid residue selected from the group consisting of Trp 82, Gln 77, Ser 79, Asp 267, Thr 266, Trp 289, Arg 310, Arg 215, and Asn 159 according to Table 1. The atomic coordinates are preferably used to model the selected portion. Such modeling is preferably accomplished by storing crystallographic information about the selected portion on a computer and then using the computer to translate the atomic coordinates into the three-dimensional structure of the selected portion of LuxP. Computers and software suitable for carrying out these functions are commercially available. Preferred computer packages include Sybyl version 6.8 from Tripos, Inc. and MacroModel version 8.0 from Schrodinger Software. A potential ligand is then identified, and the likelihood of binding between the ligand and LuxP is determining by docking the potential ligand to the selected portion of LuxP. Such docking preferably involves computationally evaluating the ligand for its ability to bind with LuxP, preferably using the commercially available computational packages described above. Ligands that bind with LuxP are potential drug candidates. The LuxP structure encoded by the crystallographic data may be displayed in a graphical three-dimensional representation on a computer screen. This allows visual inspection of the structure, as well as visual inspection of the structure's association with the ligand. Preferably, a computer is used for the identifying of the potential ligand or the docking of the potential ligand to the binding site, or both.

The identification of ligands having the potential to bind to LuxP is preferably carried out by considering at least two factors. First, preferred ligands can bind to at least a portion of a LuxP binding site. Non-covalent molecular interactions important in this association include hydrogen bonding, van der Waals interactions, hydrophobic interactions and electrostatic interactions. Second, preferred ligands can assume a conformation that allows them to associate with the LuxP binding site directly. Although certain portions of the ligand may not directly participate in these associations, those portions of the ligand may still influence the overall conformation of the ligand. This, in turn, may have a significant impact on potency. Such conformational requirements include the overall three-dimensional structure and orientation of the ligand in relation to all or a portion of the binding site, and/or the spacing between functional groups of a ligand comprising several functional groups that directly interact with a LuxP binding site. Thus, a ligand identified by the method described above is a preferred embodiment.

After docking (preferably by the computational methods described above) indicates that a particular ligand has the potential to bind to LuxP, the interaction of the indicated ligand is preferably studied by obtaining a sample of the potential ligand and exposing a prokaryotic cell to the sample. Such exposure may be for various reasons, e.g., toxicity testing. Preferably, contacting is carried out with a group of quorum-sensing prokaryotic cells, e.g., bacteria, to determine whether and to what extent the ligand affects quorum sensing.

The ligands identified as having the potential to bind to LuxP (“binding ligands”) are preferably administered to subjects in the form of pharmaceutical compositions comprising the binding ligand. The compound represented by structure (I) is an example of a binding ligand. A preferred mode of administration of the binding ligand is oral. Oral compositions preferably include an inert diluent and/or an edible carrier. The binding ligand can be enclosed in gelatin capsules or compressed into tablets. For the purpose of oral therapeutic administration, the binding ligand can be incorporated with excipients and used in the form of tablets, troches, or capsules. Pharmaceutically compatible binding agents and/or adjuvant materials can be included as part of the composition. The tablets, pills, capsules, troches and the like can contain any of the following ingredients, or compounds of a similar nature: a binder such as microcrystalline cellulose, gum tragacanth or gelatin; an excipient such as starch or lactose, a disintegrating agent such as alginic acid, Primogel, or corn starch; a lubricant such as magnesium stearate or Sterotes; a glidant such as colloidal silicon dioxide; a sweetening agent such as sucrose or saccharin; and/or a flavoring agent such as peppermint, methyl salicylate, or orange flavoring. When the dosage unit form is a capsule, it can contain, in addition to material of the above type, a liquid carrier such as a fatty oil. In addition, dosage unit forms can contain various other materials which modify the physical form of the dosage unit, for example, coatings of sugar, shellac, or other enteric agents. The ligand can be administered as a component of an elixir, suspension, syrup, wafer, chewing gum or the like. A syrup may contain, in addition to the ligand, sucrose as a sweetening agent and preservatives, dyes and colorings and flavors.

The binding ligand can also be mixed with other active materials that do not impair the desired action, or with materials that supplement the desired action, such as antibiotics. Preferred antibiotics for this purpose include aminoglycosides such as tobramycin, glycopeptides such as vancomycin, beta lactams such as amoxicillin, quinolones such as ciprofloxicin, macrolides such as azithromycin, tetracyclines, sulfonamides, trimethoprim-sulfamethoxazole, or chloramphenicol. Solutions or suspensions used for parenteral, intradermal, subcutaneous, or topical application can include the following components: a sterile diluent such as water for injection, saline solution, fixed oils, polyethylene glycols, glycerine, propylene glycol or other synthetic solvents; antibacterial agents such as benzyl alcohol or methyl parabens; chelating agents such as ethylenediaminetetraacetic acid (EDTA); buffers such as acetates, citrates or phosphates and agents for the adjustment of tonicity such as sodium chloride or dextrose. The parental preparation can be enclosed in ampoules, disposable syringes or multiple dose vials made of glass or plastic. If administered intravenously, preferred carriers are physiological saline or phosphate buffered saline (PBS).

In a preferred embodiment, the binding ligand is prepared with carriers that protect the ligand against rapid elimination from the body, such as a controlled release formulation, including implants and microencapsulated delivery systems. Biodegradable, biocompatible polymers can be used, such as ethylene vinyl acetate, polyanhydrides, polyglycolic acid, collagen, polyorthoesters, and polylactic acid. Methods for the preparation of such formulations are known to those skilled in the art.

Pharmaceutical compositions are preferably administered to subjects, preferably humans, in an amount that is therapeutically effective to treat a bacterial infection. Therapeutically effective amounts can be determined by those skilled in the art by such methods as clinical trials. Dosage may be adjusted in individual cases as required to achieve the desired degree of target bacterial regulation. Sustained release dosages and infusions are specifically contemplated. Pharmaceutical compositions can be administered by any appropriate route for systemic, local or topical delivery, for example, orally, parenterally, intravenously, intradermally, subcutaneously, buccally, intranasally, by inhalation, vaginally, rectally or topically, in liquid or solid form. Methods of administering the compounds described herein may be by specific dose or by controlled release vehicles.

The pharmaceutical composition may be administered at once, or may be divided into a number of smaller doses to be administered at varying intervals of time. It is to be further understood that for any particular subject, specific dosage regimens should be adjusted over time according to the individual need and the professional judgment of the person administering or supervising the administration of the binding ligand, and that the concentration ranges set forth herein are exemplary only and are not intended to limit the scope or practice of the claimed methods.

The instant invention is not bound by any theory of operation. The following discussion is provided for the benefit of those skilled in the art, and does not limit the scope of the claims.

FIG. 4 illustrates a chemically straightforward route connecting 4,5-dihydroxy-2,3-pentanedione, the product of the LuxS-catalyzed reaction, to AI-2. The cyclic form of 4,5-dihydroxy-2,3-pentanedione (“pro-AI-2”), can react with borate to form a cyclic borate diester. Boric acid is widely available in the biosphere; for example, the average boric acid concentration in seawater is approximately 0.4 mM. Borate, with a pKa of 9.2, is present largely as undissociated boric acid at neutral pH. Furthermore, furanosyl borate diesters similar in structure to AI-2 are relatively stable. Surprisingly, furanoid rings expected to form stable borate addition products are rare in nature with only two, ribose and apiose, known to have this configuration in their physiological derivatives. In addition to AI-2, other borate adducts may be produced from cyclic derivatives of 4,5-dihydroxy-2,3-pentanedione.

Boric acid has a dramatic effect on AI-2 signaling in V. harveyi, as shown through use of a LuxP⁺, LuxS⁺ reporter strain that produces its own AI-2. Endogenously produced AI-2 stimulates light production, but only after a delay during which the AI-2 concentration builds to a threshold level. During this delay period, the addition of the activity produced by the in vitro reaction of S-adenosylhomocysteine with Pfs and LuxS induces light production. It has been found that addition of 1 mM boric acid also results in substantial induction. Boric acid concentrations as low as 10 μM cause significant (10-fold) induction. It has also been found that boric acid has no effect on a LuxS⁻ strain that cannot synthesize 4,5-dihydroxy-2,3-pentanedio latter strains show wild-type responses to the other V. harveyi autoinducer (AI-1) under these experimental conditions. Therefore, the induction of bioluminescence by boric acid is specific for the AI-2 detection system. The strong induction by boric acid in the bioassay is consistent with the proposed chemical structure for AI-2 shown in formula (1).

The structure shown in formula (I) is a novel furanosyl borate diester. By contrast, previously characterized autoinducers are acyl homoserine lactones, modified oligopeptides or quinolones. One of the most unexpected features of the proposed AI-2 structure is the presence of a boron atom. Boron has previously been found in a small number of polyketide antibiotics. It is also known to be essential for vascular plants and several other organisms including cyanobacteria; however, in no case is its functional role well understood. The results discussed herein provide evidence for a biochemically defined function for boron in bacterial quorum sensing.

Whereas acyl homoserine lactone and oligopeptide autoinducers are used for communication within a bacterial species, AI-2 is proposed to be a universal signal that facilitates inter-species communication. LuxS enzymes from a variety of bacteria produce AI-2 activity, presumably through the generation of a common intermediate, pro-AI-2. It is apparent that LuxP, the AI-2 sensor in V. harveyi, binds a borate diester of pro-AI-2. Since the biochemical machinery for synthesizing pro-AI-2 is broadly conserved, and borate is widely available, it is believed that AI-2 is produced and detected by diverse bacterial species.

EXAMPLES

LuxP (residues 24–365) from Vibrio harveyi was cloned into the pGEX-4T expression vector (Amersham Pharmacia) with an engineered thrombin cleavage site in between glutathione-S-transferase (GST) and the N-terminus of LuxP. The vector was transformed into E. coli strain BL21 for overexpression of GST-LuxP fusion protein upon induction with IPTG. After cell lysis, the GST-LuxP fusion protein was purified through use of a glutathione-agarose (Amersham Pharmacia) column. After elution, the fusion protein was digested with thrombin overnight on ice and further purified using an 8-ml MonoQ column (Amersham Pharmacia). The eluent was salt-adjusted and then concentrated through use of a pressure cell (Amicon) to yield a stock solution of 8 mg/mL LuxP in 0.05 M Tris-HCl, 0.15 M NaCl, pH 8.0.

The hanging drop vapor diffusion method was used for the crystallization. Crystals were grown by equilibrating a mixture of 1.0 μL of LuxP AI-2 solution and 1 μL of 0.1 M Tris-HCl, 16% PEG 4000, 18% glycerol (v/v), pH 8.5 at room temperature, against the same buffer/precipitant solution in the reservoir. The crystals are in space group P2₁, with cell parameters of a=42.3 , b=77.5, c=52.0 Å and β=96.4°. X-ray diffraction data on the native protein were collected at beamline X12C of the National Synchrotron Light Source at the Brookhaven National Laboratories through use of a Brandeis CCD detector at 100 K. Data on heavy atom derivatives were collected through use of R-AXIS detectors at 100K. Diffraction data were all processed with the HKL program suite, see Otwinowski, Z. & Minor, W., Methods Enzymol., vol. 276, pp. 965–969 (1996). Heavy atom sites were located through use of the program Solve, see Terwilliger, T. C. & Berendzen, J., Acta Crystllogr. D, vol. 55, pp. 849–861 (1999). Multiple isomorphous phasing was done through use of MLPHARE and DM from the CCP4 suite. Model building and refinement were done through use of O, see Jones, T. A. et al., Acta Crystallogr. A, vol. 47, pp. 110–119 (1991), and CNS, see Brunger, A. T., et al. Acta Crystallogr. D, vol. 54, pp. 905–921 (1998), respectively.

The X-ray crystal structure of recombinant V. harveyi LuxP, overproduced in the LuxS⁺ E. coli strain BL21, was determined by multiple isomorphous replacement at 2.8 Å resolution and refined to 1.5 Å resolution, see Table 1. The assignment of atom type at each position was dictated by both the valence and the local chemical environment within the protein. Thus, tetrahedrally substituted atoms were deemed to be carbon atoms, whereas atoms within hydrogen-bonding distance of at least one (but generally two or three) polar protein atoms or buried water molecules were deemed to be oxygen atoms. Weak bond length and bond angle restraints were applied to the ligand during subsequent refinement. The resulting structural model for the LuxP-AI-2 complex has an R_cryst and R_free of 0.21 and 0.24, respectively, and exhibits excellent geometry. No orientational or chemical heterogeneity of the buried ligand is evident in the crystallographic electron density.

Nuclear magnetic resonance (NMR) spectroscopy was used to confirm that the ligand contains boron. The ¹¹B NMR spectrum of the LuxP-AI-2 complex shows a peak signifying the presence of boron. The chemical shift for this peak (6.2 ppm) is within the range (3.9–6.2 ppm) observed for borate esters of carbohydrate 1,2-diols. For comparison, borate esters of 1,3-diols and boric acid display chemical shifts of 0.3–0.9 and 18.8 ppm, respectively. The ¹¹B NMR peak disappeared after ultrafiltration to remove protein. Furthermore, it was not observed in apo-LuxP overproduced in LuxS⁻ BL21 E. coli.

The mass of the LuxP-AI-2 complex was determined through use of electrospray ionization mass spectrometry (ESI-MS). Under optimized conditions, peaks were observed for both LuxP and LuxP-AI-2 at molecular weights that differ by 194.2±3.0 Da. This mass increment agrees closely with the molecular weight of the AI-2 structure shown in formula (1) (192.9 Da).

It will be appreciated by those skilled in the art that various omissions, additions and modifications may be made to the processes described above without departing from the scope of the invention, and all such modifications and changes are intended to fall within the scope of the invention, as defined by the appended claims.