Eric T. Baldwin

Crystal Structure of the Herpes Simplex Virus 1 DNA Polymerase

Herpesviruses are the second leading cause of human viral diseases. Herpes Simplex Virus types 1 and 2 and Varicella-zoster virus produce neurotropic infections such as cutaneous and genital herpes, chickenpox, and shingles. Infections of a lymphotropic nature are caused by cytomegalovirus, HSV-6, HSV-7, and Epstein-Barr virus producing lymphoma, carcinoma, and congenital abnormalities. Yet another series of serious health problems are posed by infections in immunocompromised individuals. Common therapies for herpes viral infections employ nucleoside analogs, such as Acyclovir, and target the viral DNA polymerase, essential for viral DNA replication. Although clinically useful, this class of drugs exhibits a narrow antiviral spectrum, and resistance to these agents is an emerging problem for disease management. A better understanding of herpes virus replication will help the development of new safe and effective broad spectrum anti-herpetic drugs that fill an unmet need. Here, we present the first crystal structure of a herpesvirus polymerase, the Herpes Simplex Virus type 1 DNA polymerase, at 2.7 Å resolution. The structural similarity of this polymerase to other α polymerases has allowed us to construct high confidence models of a replication complex of the polymerase and of Acyclovir as a DNA chain terminator. We propose a novel inhibition mechanism in which a representative of a series of non-nucleosidic viral polymerase inhibitors, the 4-oxo-dihydroquinolines, binds at the polymerase active site interacting non-covalently with both the polymerase and the DNA duplex.

Structure of HIV-1 protease with KNI-272, a tight-binding transition-state analog containing allophenylnorstatine.

BACKGROUND HIV-1 protease (HIV PR), an aspartic protease, cleaves Phe-Pro bonds in the Gag and Gag-Pol viral polyproteins. Substrate-based peptide mimics constitute a major class of inhibitors of HIV PR presently being developed for AIDS treatment. One such compound, KNI-272, which incorporates allophenylnorstatine (Apns)-thioproline (Thp) in place of Phe-Pro, has potent antiviral activity and is undergoing clinical trials. The structure of the enzyme-inhibitor complex should lead to an understanding of the structural basis for its tight binding properties and provide a framework for interpreting the emerging resistance to this drug. RESULTS The three-dimensional crystal structure of KNI-272 bound to HIV PR has been determined to 2.0 A resolution and used to analyze structure-activity data and drug resistance for the Arg8-->Gln and ILe84-->Val mutations in HIV PR. The conformationally constrained Apns-Thp linkage is favorably recognized in its low energy trans conformation, which results in a symmetric mode of binding to the active-site aspartic acids and also explains the unusual preference of HIV PR for the S, or syn, hydroxyl group of the Apns residue. The inhibitor recognizes the enzyme via hydrogen bonds to three bridging water molecules, including one that is coordinated directly to the catalytic Asp125 residue. CONCLUSIONS The structure of the HIV PR/KNI-272 complex illustrates the importance of limiting the conformational degrees of freedom and of using protein-bound water molecules for building potent inhibitors. The binding mode of HIV PR inhibitors can be predicted from the stereochemical relationship between adjacent hydroxyl-bearing and side chain bearing carbon atoms of the P1 substituent. Our structure also provides a framework for designing analogs targeted to drug-resistant mutant enzymes.

Human liver cathepsin D: Purification, crystallization and preliminary X-ray diffraction analysis of a lysosomal enzyme☆

The two-chain form of active cathepsin D, a glycosylated, lysosomal aspartic proteinase, has been isolated from human liver. Isoelectric focusing revealed two major species of enzyme that differed by approximately 0.2 pI unit. Crystals suitable for X-ray diffraction analysis were prepared from acidic solutions using precipitation with ammonium sulfate. The hexagonal crystals diffracted X-rays to beyond 3.1 A resolution and belonged to space group P6(1) (or P6(5)) with cell constants a = b = 125.9 A, c = 104.1 A, gamma = 120.0 degrees. The crystals likely contain two molecules in the asymmetric unit, giving a solvent content of 56% (v/w). Biochemical analysis of crystals indicated that both isoforms were present in approximately equimolar proportions. Full structure determination of the enzyme is underway.

Discovery of new acylaminopyridines as GSK-3 inhibitors by a structure guided in-depth exploration of chemical space around a pyrrolopyridinone core.

Glycogen synthase kinase-3 (GSK-3) has been proposed to play a crucial role in the pathogenesis of many diseases including cancer, stroke, bipolar disorders, diabetes and neurodegenerative diseases. GSK-3 inhibition has been a major area of pharmaceutical interest over the last two decades. A plethora of reports appeared recently on selective inhibitors and their co-crystal structures in GSK-3β. We identified several series of promising new GSK-3β inhibitors from a coherent design around a pyrrolopyridinone core structure. A systematic exploration of the chemical space around the central spacer led to potent single digit and sub-nanomolar GSK-3β inhibitors. When dosed orally in a transgenic mouse model of Alzheimers disease (AD), an exemplary compound showed significant lowering of Tau phosphorylation at one of the GSK-3 phosphorylating sites, Ser396. X-ray crystallography greatly aided in validating the binding hypotheses.

Conserved core substructures in the overlay of protein-ligand complexes

The method of conserved core substructure matching (CSM) for the overlay of protein-ligand complexes is described. The method relies upon distance geometry to align structurally similar substructures without regard to sequence similarity onto substructures from a reference protein empirically selected to include key determinants of binding site location and geometry. The error in ligand position is reduced in reoriented ensembles generated with CSM when compared to other overlay methods. Since CSM can only succeed when the selected core substructure is geometrically conserved, misalignments only rarely occur. The method may be applied to reliably overlay large numbers of protein-ligand complexes in a way that optimizes ligand position at a specific binding site or subsite or to align structures from large and diverse protein families where the conserved binding site is localized to only a small portion of either protein. Core substructures may be complex and must be chosen with care. We have created a database of empirically selected core substructures to demonstrate the utility of CSM alignment of ligand binding sites in important drug targets. A Web-based interface can be used to apply CSM to align large collections of protein-ligand complexes for use in drug design using these substructures or to evaluate the use of alternative core substructures that may then be shared with the larger user community. Examples show the benefit of CSM in the practice of structure-based drug design.

Physical methods to determine the binding mode of putative ligands for hepatitis C virus NS3 helicase.

Several small molecules identified by high-throughput screening (HTS) were evaluated for their ability to bind to a nonstructural protein 3 (NS3) helicase from hepatitis C virus (HCV). Equilibrium dissociation constants (K(d)s) of the compounds for this helicase were determined using several techniques including an assay measuring the kinetics of isothermal enzyme denaturation at several concentrations of the test molecule. Effects of two nonhydrolyzable ATP analogs on helicase denaturation were measured as controls using the isothermal denaturation (ITD) assay. Two compounds, 4-(2,4-dimethylphenyl)-2,7,8-trimethyl-4,5-quinolinediamine and 2-phenyl-N-(5-piperazin-1-ylpentyl)quinazolin-4-amine, were identified from screening that inhibited the enzyme and had low micromolar dissociation constants for NS3 helicase in the ITD assay. Low micromolar affinity of the quinolinediamine to helicase was also confirmed by nuclear magnetic resonance experiments. Unfortunately, isothermal titration calorimetry (ITC) experiments indicated that a more water-soluble analog bound to the 47/23-mer oligonucleotide helicase substrate with low micromolar affinity as did the substituted quinazolinamine. There was no further interest in these templates as helicase inhibitors due to the nonspecific binding to enzyme and substrate. A combination of physical methods was required to discern the mode of action of compounds identified by HTS and remove undesirable lead templates from further consideration.

X-Ray Structure of a Tethered Dimer for HIV-1 Protease

HIV-1 protease (HIV PR) is essential for the replication of human immunodeficiency virus and is responsible for the processing of gag and gag-pol polyproteins. Thus, HIV PR has been the target of intensive efforts to design inhibitors [1–4]. The enzyme is a homodimeric aspartic protease, and its active site is formed by the monomer-monomer interface and exhibits two-fold rotational symmetry. In this respect, HIV PR differs from the single chain, cellular aspartic proteases whose active sites are formed by the interaction of two homologous, but non-identical, domains. Because of its dimeric nature, mutations in HIV PR are expressed in a symmetric pair-wise fashion involving both chains. To facilitate studies of single site mutations, tethered HIV PR dimers have been constructed using recombinant methods [5,6]. In these studies, two identical HIV PR genes were placed in tandem, separated by a short spacer region, such that the C-terminus of subunit 1 would be covalently linked via the spacer to the N-terminus of subunit 2. This design was based on the fact that the N- and C-termini of different subunits are located within several A of each other near the surface of the native enzyme. While biochemical studies indicated no major differences in the activities of tethered and two-chain HIV PR [5,6], we wished to examine the structure of the single chain enzyme as a basis for further studies on single site mutants. We report here the crystal structure of a recombinant tethered dimer complexed with a C2 symmetry-based inhibitor A-76928. A fuller account of this structure will appear elsewhere[9].

STRUCTURE OF HUMAN CATHEPSIN D: COMPARISON OF INHIBITOR BINDING AND SUBDOMAIN DISPLACEMENT WITH OTHER ASPARTIC PROTEASES

Cathepsin D (EC 3.4.23.5) (CatD) is an intracellular aspartic protease (AP) that is normally found in the lysosomes of higher eukaryotes. CatD shares two structural features with lysosomal enzymes that distinguish it from most secreted, extracellular APs. First, the mature enzyme from humans is found as a two-chain enzyme as a result of post-translational cleavage and removal of an insertion loop in the N-domain. Second, CatD contains phosphorylated, N-linked oligosaccharides that target the enzyme to lysosomes via man-nose-6-phosphate receptors (MPR). Interest in CatD as a target for drug design results from its association with several biological processes of therapeutic significance, particularly breast cancer and Alzheimer’s disease (Table 1). Recent studies have implicated CatD in the processing of (β-amyloid precursor protein to promote amyloid plaque formation in Alzheimer’s brain. Numerous studies of primary breast cancers have demonstrated that elevated levels of CatD were correlated with an increased risk of metastasis and shorter relapse-free survival. High levels of CatD produced in the vicinity of the growing tumor may degrade the extracellular matrix, either directly or indirectly by potentiating the activity of other tissue proteases, such as collagenase, and thereby promote tumor growth and metastasis. While a role for CatD or other APs in metastasis has been questioned by recent results that show a negative effect of pepstatin A in an in vitro invasion assay using breast cancer-derived MCF7 cells, new studies have revealed that elevated levels of CatD secretion in breast cancers correlate with tumor aggressiveness.

Structure of HIV-1 protease with KNI-272: A transition state mimetic inhibitor containing allophenylnorstatine

Inhibitors of human immunodeficiency virus type 1 protease (HIV PR) are the subject of wide interest for the development of therapeutic drugs against AIDS [1–4]. The crystal structures of numerous HIV PR/inhibitor complexes have been solved to aid the process of inhibitor design and to rationalize the structure-activity relationships (SAR) for various classes of inhibitors [5–7]. Incorporation of allophenylnorstatine (APNS) [(2S,3S)-3-amino-2-hydroxy-4-phenylbutyric acid] into peptidomimetics can lead to highly potent inhibitors of HIV PR [8]. However, no structure of APNS containing inhibitor complexes has been reported. Recently, KNI-272, a conformationally-constrained inhibitor containing APNS (Figure la), has been shown to possess good anti-HIV activity [9]. As a step towards understanding the high affinity of this novel class of inhibitors, we determined the three dimensional structure of KNI-272 bound to HIV PR.

Molecular Dynamics of HIV-1 Protease in Complex with a Difluoroketone-Containing Inhibitor: Implications for the Catalytic Mechanism

The structures of three aspartic proteases, penicillopepsin, endothiapepsin and rhizopuspepsin, have been determined in complexes with difluorostatone (DFS)-containing peptide-based inhibitors [1–3]. In all three structures, the hydrated core of the inhibitors display similar conformations and interactions with the active site aspartate groups. One of the gem-hydroxyls is positioned between both aspartates, the other one interacts with a single aspartate, and the fluorines do not interact with either of them. Since one of the hydroxyl groups is located near the position of the active site water observed in the native structures, these inhibitors have been proposed as analogs of transition state intermediates of peptide bond hydrolysis.