 |
Protein Crystallography, Tropical Diseases, and Structure-Based Drug Design
Summary: Wim Hol is interested in protein crystallography for structure-based design of drugs for tropical diseases.
The major goals of our laboratory are to unravel the three-dimensional structures of key protein molecules; to explore the relationships among protein structure, function, and dynamics; and to exploit this insight for the design of new medically relevant molecules, in particular for the treatment of infectious tropical diseases.
Sleeping Sickness and Leishmaniasis
A number of important tropical diseases are caused by a group of unicellular flagellates, the trypanosomatids. Specifically, Trypanosoma brucei causes sleeping sickness in Africa, T. cruzi is responsible for Chagas' disease in Latin America, and a dozen or more Leishmania species cause a variety of leishmaniases throughout the tropics and subtropics. Despite the entirely different diseases due to these organisms, they share many fascinating cellular and biochemical characteristics. One of these is the presence of glycosomes, unique microbodies that contain large quantities of glycolytic enzymes as well as smaller amounts of other proteins. Another is the presence of a unique macromolecular machine, the editosome.
We have recently solved the high-resolution structures of the first two editosome enzymes: the RNA-editing ligase (REL1) and the RNA-editing terminal uridylyl transferase (RET2), both with various nucleotides bound.
Since the sleeping sickness parasite in the human host is entirely dependent on glycolysis, we have targeted a number of glycosomal glycolytic enzymes to explore their potential for the design and synthesis of high-affinity and selective inhibitors that block either enzymatic activity or the import of these proteins into the organelle. In collaboration with Paul Michels and Fred Opperdoes (Christian de Duve Institute of Cellular and Molecular Pathology, or ICP, Brussels), and Christophe Verlinde, Michael Gelb, Wesley Van Voorhis, and Fred Buckner (University of Washington), we have increased the affinity of ligands for T. brucei and L. mexicana glyceraldehyde-3-phosphate dehydrogenase (GAPDH) enzyme by four to five orders of magnitude. Recent crystal structures of T. brucei GAPDH-inhibitor complexes showed that the enzyme undergoes surprisingly large conformational shifts upon binding the ligand. In the case of glycerol-3-phosphate dehydrogenase (GPDH), long-wavelength anomalous scattering effects were used to unravel multiple binding modes of ligands in the cofactor-binding cleft. These protein-inhibitor structures provide a platform for further improvements in affinity and selectivity.
All glycosomal proteins are nuclear-encoded and subsequently imported into the glycosome by peroxins, a 23-member group of associating proteins. We have elucidated the crystal structure of a fragment of T. brucei peroxin 5 (PEX5), the major glycosomal target signal receptor protein. This fragment consists of three tetratrico-peptide repeats (TPRs). TPR motifs are supposed to be all antiparallel helices, but the third TPR motif in our structure appears to be a perfectly straight single helix. This suggests the possibility of large, jackknife-type motions in PEX5 and other TPR-containing proteins. We have also demonstrated that the three similar motifs of PEX5 responsible for the interaction with PEX14 have surprisingly different characteristics.
Tuberculosis
A high-resolution structure of the related iron-dependent regulator IdeR from Mycobacterium tuberculosis was determined in its fully activated form—displaying the wedge position of the metal-binding SH3 domain. In collaboration with Randall Holmes (University of Colorado Health Sciences Center, Denver) and Craig Beeson (Medical University of South Carolina), we are investigating ways to discover synthetic molecules that can "deregulate the regulator." The newest structures focus on the DosR regulator, involved in the hypoxic response of M. tuberculosis.
Crystal structures were also determined for two key enzymes of M. tuberculosis: dihydrofolate reductase (DHFR) and dihydropteroate synthase (DHPS), in collaboration with Worachart Sirawaraporn (Mahidol University, Bangkok). The structure of M. tuberculosis DHFR was determined in complex with a dozen inhibitors of the folate-binding site, with diamino pyrimidine as an exquisitely precisely bound "anchor." The different inhibitors showed how various ring systems attached to the anchor by linkers of different lengths occupy different positions farther away from the catalytic center.
The M. tuberculosis DHPS enzyme is the target of sulfa drugs in many bacteria and of the important antileprosy drug dapsone in M. leprae. DHPS is a dimer with an eight-stranded, parallel β-barrel topology. The most characteristic feature of DHPS is perhaps the presence of numerous long loops, with some drug-resistant mutations residing in these loops 10–20 Å from plausible substrate-binding sites. The mechanism of DHPS action remains shrouded in mystery, yet it seems clear that large motions of the loops during catalysis are essential for the functioning of these enzymes.
DNA Topology and DNA Repair
In collaboration with James Champoux (University of Washington), we have made major progress in understanding the way in which cells maintain the proper topology of DNA, by elucidating three-dimensional structures of human topoisomerase I (topo I) in covalent and noncovalent complex with DNA. Recently the crystal structure of human tyrosyl-DNA phosphodiesterase (TDP), an enzyme involved in DNA repair, was determined. TDP has a most unusual substrate—DNA covalently bound to topo I. TDP cleaves the phosphodiester bond between a tyrosine residue and a 3' DNA phosphate. Insight into the enzyme's mechanism has been obtained by the structure determination of a quaternary complex of enzyme, vanadate, oligo nucleotide, and tyrosine-containing peptide.
Cholera Toxin and Heat-Labile Enterotoxin
Cholera toxin, the major virulence factor of Vibrio cholerae, is closely related to heat-labile enterotoxin of enterotoxigenic Escherichia coli. These toxins are responsible for significant numbers of deaths among young children in developing countries. The receptor-binding site has been well characterized by a 1.25-Å-resolution structure of cholera toxin's B pentamer in complex with five pentasaccharides from the receptor GM1. We are exploring this information for inhibitor design in collaboration with Erkang Fan and Christophe Verlinde (University of Washington), which has resulted in the synthesis of large pentavalent and decavalent ligands that display enhanced affinity for the pentamers by four to six orders of magnitude compared to monomeric ligands. Two of these were "macroligands" cocrystallized with the target protein, yielding structures of up to 1.5-Å resolution and revealing essential aspects of their binding mode. In collaboration with Randall Holmes, we have made significant progress in understanding the activation mechanism of the A subunit of these toxins by unraveling the structure of a constitutively active cholera toxin variant and of the complex of the toxin's A1 subunit with the human G protein ARF6.
Cholera toxin and heat-labile enterotoxin are secreted by the multiprotein type II secretion system. Recently we have solved structures of several components of this large machinery, including EpsE, the "secretion ATPase" of the system, and domains of EpsL and EpsM. These studies are carried out in collaboration with Maria Sandkvist (University of Maryland School of Medicine) and Michael Bagdasarian (Michigan State University).
Other Projects
Additional projects include structural studies on cephalosporin acylase, phosphodiesterases, the giant pyruvate dehydrogenase multienzyme complex, and key proteins from the major malaria parasite, Plasmodium falciparum. Specifically, the structure of peptide deformylase (PDF), a key metalloenzyme from P. falciparum and a likely drug target, has recently been elucidated—both with and without inhibitors bound to the active site. In collaboration with the groups of Dehua Pei (Ohio State University) and Christophe Verlinde, we have designed novel cyclic PDF inhibitors with block bacterial PDFs in the 100-picomolar range.
Part of the work described above was also supported by a grant from the National Institutes of Health. Major NIH support has also been obtained for Structural Genomics of Pathogenic Protozoa (SGPP), a collaboration with a dozen other groups across the country.
|
|
|
|
