Key Points Fragment-based drug design is based on screening smaller numbers of compounds (typically several thousand) in the hopes of finding low-affinity fragments ( K d values in the high micromolar to millimolar range), in contrast to conventional high-throughput screening (HTS), which attempts to evaluate as many compounds as technologically possible (typically a million or more) in the hopes of finding relatively potent drug leads ( K d values ideally less than 1 μM). The combination of broader sampling of the potential chemical universe than HTS and increased hit rates for molecules of low complexity makes fragment-based screening a powerful tool for lead generation. Fragment-based screening is also less prone to artefacts as the low-molecular-mass compounds tend to be more soluble and the methods of detection are simpler and more robust. Two-dimensional, isotope-edited nuclear magnetic resonance (NMR) spectroscopy was the first approach used in fragment-based drug design. It is well suited to this purpose as NMR chemical shifts are exquisitely sensitive to ligand binding, and problems with compound interference can be solved by spectral editing. During the past decade, the popularity of fragment-based screening has grown at a remarkable rate in both industry and academia. A range of different strategies have been developed, including alternative NMR-based approaches that obviate the need for isotope labelling, approaches based on X-ray-crystallography and fragment tethering, which are discussed here. The ability to obtain NMR or X-ray crystal structures on fragment leads has a dramatic influence on the success of fragment-based drug design. The successful applications of fragment-based drug design have provided ample support that the use of fragments could, in many cases, be the most direct route to the best achievable balance between potency and pharmacokinetics.

All reported GnRH receptor mutants (causing human hypogonadotropic hypogonadism) are misfolded proteins that cannot traffic to the plasma membrane. Pharmacoperones correct misfolding and rescue mutants, routing them to the plasma membrane where they regain function. Because pharmacoperones are often peptidomimetic antagonists, these must be removed for receptor function after rescue; in vivo this necessitates pulsatile pharmacoperone administration. As an antecedent to in vivo studies, we determined whether pharmacoperones need to be present at the time of synthesis or whether previously misfolded proteins could be refolded and rescued. Accordingly, we blocked either protein synthesis or intra-cellular transport. Biochemical and morphological studies using 12 mutants and 10 pharmacoperones representing three different chemical classes show that previously synthesized mutant proteins, retained by the quality control system (QCS), are rescued by pharmacoperones, showing that pharmacoperone administration in vivo likely need not consider whether the target protein is being synthesized at the time of drug administration.

The design and synthesis of a series of 11,12-cyclic carbamate derivatives of 6-O-methylerythromycin A that are novel, nonpeptide LHRH antagonists, is described. The macrolide antagonist 1, discovered during a screen of our chemical repository, was compared to a macrocyclic peptide antagonist 2 using molecular modeling, thus providing a model for the design of more potent antagonists. Medicinal chemistry efforts to find a replacement for cladinose at position 3 of the erythronolide core provided a series of oxazolidinone carbamates that were equally as active as the cladinose-containing parent macrolides. The descladinose LHRH antagonist 14 has 1-2 nM affinity for both rat and human LHRH receptors and is a potent inhibitor of LH release (pA2 = 8.76) in vitro. In vivo, 14 was found to produce a dose-dependent suppression of LH in male castrate rats via both i.v. and p.o. dosing.

Antibacterial SAR for a series of macrolides derived from erythromycin A that are potent LHRH antagonists was developed in an attempt to eliminate the antibiotic activities of these compounds. Increasing the size of the alkyl substituents on the desosamine 3'-amine resulted in potent LHRH antagonists that were inactive against staphylococcal bacteria strains, and were significantly (>10-fold) less active against streptococcal bacteria strains. Complete elimination of antibacterial activities could be achieved by replacement of one or both methyl groups on the 3'-amine with a large alkyl substituent.

Here we describe the three- dimensional crystal structures of human glucocorticoid receptor ligand- binding domain ( GR- LBD) in complex with the antagonist RU486 at 2.3 Angstrom resolution and with the agonist dexamethasone ligand together with a coactivator peptide at 2.8 Angstrom. The RU- 486 structure was solved in several different crystal forms, two with helix 12 intact ( GR1 and GR3) and one with a protease- digested C terminus ( GR2). In GR1, part of helix 12 is in a position that covers the co- activator pocket, whereas in the GR3, domain swapping is seen between the crystallographically identical subunits in the GR dimer. An arm consisting of the end of helix 11 and beyond stretches out from one molecule, and helix 12 binds to the other LBD, partly blocking the coactivator pocket of that molecule. This type of GR-LBD dimer has not been described before but might be an artifact from crystallization. Furthermore, the subunits of the GR3 dimers are covalently connected via a disulfide bond between the Cys- 736 residues in the two molecules. All three RU- 486 GR- LBD structures show that GR has a very flexible region between the end of helix 11 and the end of helix 12.

Combinatorial and structure-based medicinal chemistry strategies were used together to advance a lead compound with an activity of K(i) = 58 microM via a potency enhancement of >70 000-fold to an analogue with an activity of K(i) = 0.8 nM against influenza neuraminidase (A/Tokyo/67). Lead optimization was initiated using molecular modeling and combinatorial chemistry. Protein crystal structures revealed that inconsistent structure-activity relationship (SAR) data resulted from different binding orientations of the inhibitor core five-membered rings from one series to another. Binding modes for a series of compounds showed up to a 180 degrees variation in orientation of the five-membered ring within the active site. Potent analogues were only achieved with chemical series that were observed to bind in the same orientation and yielded consistent SAR. In one series, consistent binding was obtained by an unprecedented occupation of a negatively charged binding pocket by a neutral methyl ester unit. The structural rationale for this novel SAR variation, based on protein crystallographic data, is given.

We expressed a test system of wild-type (WT) rat (r) and human (h) gonadotropin-releasing hormone (GnRH) receptors (GnRHRs), including naturally occurring (13) and manufactured (five) “loss-of-function” mutants of the GnRHR. These were used to assess the ability of different GnRH peptidomimetics to rescue defective GnRHR mutants and determine their effect on the level of membrane expression of the WT receptors. Among the manufactured mutants were the shortest rGnRHR C-terminal truncation mutant that resulted in receptor loss-of-function (des325–327-rGnRHR), two nonfunctional deletion mutants (des237–241-rGnRHR and des260–265-rGnRHR), two nonfunctional Cys mutants (C229A-rGnRHR and C278A-rGnRHR); the naturally occurring mutants included all 13 full-length GnRHR point mutations reported to date that result in full or partial human hypogonadotropic hypogonadism. The 10 peptidomimetics assessed as potential rescue molecules (“pharmacoperones”) are from three differing chemical pedigrees (indoles, quinolones, and erythromycin-derived macrolides) and were originally developed as GnRH peptidomimetic antagonists. These structures were selected for this study because of their predicted ability to permeate the cell membrane and interact with a defined affinity with the GnRH receptor. All peptidomimetics studied with an IC50 value (for hGnRHR) ≤2.3 nM had measurable efficacy in rescuing GnRHR mutants, and within a single chemical class, this ability correlated to these IC50 values. Erythromycin-derived macrolides with IC50 values as high as 669.5 nM showed efficacy as rescue compounds. The ability to rescue a particular receptor was a reasonable predictor of the ability to rescue others, even across species lines, although particular mutants could not be rescued by any of the drugs tested.

Here we describe the three-dimensional crystal structures of human glucocorticoid receptor ligand-binding domain (GR-LBD) in complex with the antagonist RU-486 at 2.3 Å resolution and with the agonist dexamethasone ligand together with a coactivator peptide at 2.8 Å. The RU-486 structure was solved in several different crystal forms, two with helix 12 intact (GR1 and GR3) and one with a protease-digested C terminus (GR2). In GR1, part of helix 12 is in a position that covers the co-activator pocket, whereas in the GR3, domain swapping is seen between the crystallographically identical subunits in the GR dimer. An arm consisting of the end of helix 11 and beyond stretches out from one molecule, and helix 12 binds to the other LBD, partly blocking the coactivator pocket of that molecule. This type of GR-LBD dimer has not been described before but might be an artifact from crystallization. Furthermore, the subunits of the GR3 dimers are covalently connected via a disulfide bond between the Cys-736 residues in the two molecules. All three RU-486 GR-LBD structures show that GR has a very flexible region between the end of helix 11 and the end of helix 12.

Fortuna Haviv, Wesley Dwight, Bradley Crawford, Rolf Swenson, Milan Bruncko, Michele Kaminski, Kaneyoshi Kato, Yoshihiro Sugiura, Lisa Frey, Gilbert Diaz, Gary Bammert, Eugene N. Bush, Leslie Besecke, Kurt Mohning, Jason Segreti, Mary Spangler, Craig Wegner, and Jonathan Greer Pharmaceutical Products Division, Abbott Laboratories, Abbott Park, IL 60064-3500, U.S.A.; Pharmaceutical Development Division, Takeda Chemical Industries, Osaka Japan; and TAP Pharmaceutical Inc., Abbott Park, IL 60064-3500, U.S.A.

Gonadotropin-releasing hormone (GnRH) stimulates the synthesis and secretion of the gonadotropins that maintain the reproductive axis in mammals. Efforts have focused an the characterization of novel, nonpeptidic orally active antagonists of the GnRH receptor. An erythromycin A derivative, A-198401 (11-deoxy-11-[carboxy (3,4-dichlorophenethyl) amino]-3-O-[4-(S)-methyl-oxazolidin-2-one] carbamoyl-5-O-(3 ' -N-desmethyl-3 ' -N-cyclopropylmrthyl) desosaminyl-6-O-methyl-erythronolide A 11,12-(cyclic carbamate), showed nanomolar affinity for the human (CHO-21) and rat GnRH receptors in vitro (pK values 8.7 +/- 0.2 and 9.2 +/- 0.14, respectively). In a functional in vitro assay, A-198401 inhibited leuprolide-induced release of luteinizing hormone (LH) from cultured rat pituitary cells with a pA(2) Value of 8.8. Intravenous (IV) dosing of A-198401 in castrate male rats produced a significant dose-dependent suppression of LH production with an ED80 value of 5.26 mg/kg. Sustained testosterone (T) suppression was observed after IV dosing of A-198401 in the intact rat, with a 10-mg/kg-dose producing 9-24 h suppression. Analysis of the IV and PO data indicate that A-198401 has a bioavailability of 15%. A-198401 is a novel nonpeptide GnRH antagonist that produces significant and sustained suppression of LH and T production in animal models when dosed either IV or PO and may provide the basis for a therapeutic GnRH antagonist for the clinical treatment of reproductive hormone-dependent diseases. (C) 2001 Wiley-Liss, Inc.

Summary mounting, and (3) automated crystal alignment. To increase the efficiency of diffraction data collection for protein crystallographic studies, an automated system designed Crystal Storage to store frozen protein crystals, mount them sequentially, align In order to maximize crystal stability, the crystal samples are them to the X-ray beam, collect complete data sets, and return stored in a custom rack that is maintained at cryogenic temthe crystals to storage has been developed. peratures by immersion in a liquid nitrogen bath. The protein Advances in X-ray data collection technology including more crystals are manually mounted onto custom crystal holders brilliant X-ray sources, improved focusing optics, and faster- equipped with CryoLoops purchased from Hampton Research readout detectors have reduced diffraction data acquisition (Laguna Niguel, CA) and immediately transferred into the custimes from days to hours at a typical protein crystallography tom rack using a specially designed tool that mimics the robot laboratory [1, 2]. In addition, the number of high-brilliance syn- arm (Figure 1). The sample rack (Figure 2) contains 63 crystal chrotron X-ray beam lines dedicated to macromolecular crys- positions and is composed of a 9 3 7 rectangular array. At tallography has increased significantly, and data collection each position, a tapered passageway serves to guide the tool times at these facilities can be routinely less than an hour per and crystal sample to a magnetic base at the bottom of the crystal. Because the number of protein crystals that may be passageway. The rack is designed to fit precisely onto the collected in a 24 hr period has substantially increased, unat- system deck into the bottom of an insulated sample dewar tended X-ray data acquisition, including automated crystal that is equipped with a liquid nitrogen refill system (Figure 2). mounting and alignment, is a desirable goal for protein crystal- The sample dewar is open at the top to provide access to the lography. The ability to complete X-ray data collection more samples by the robot arm. This system both preserves the efficiently should impact a number of fields, including the crystals at liquid nitrogen temperatures and presents the proemerging structural genomics field [3], structure-directed drug tein crystals in an ordered array for robotic crystal mounting. design, and the newly developed screening by X-ray crystallog- Frosting in and around the sample dewar is minimized by mainraphy [4], as well as small molecule applications. taining the humidity in the room as low as possible. As the liquid nitrogen evaporates, a blanket of nitrogen gas forms Design Considerations above the samples. This provides a barrier to any remaining

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 pier 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.