Ethan A. Merritt

Objectives: Methionyl-tRNA synthetase (MetRS) inhibitors are under investigation for the treatment of intestinal infections caused by Giardia lamblia . To properly analyze the therapeutic potential of the MetRS inhibitor 1717 , experimental tools including a robust cell-based assay and a murine model of infection were developed based on novel strains of G. lamblia that employ luciferase reporter systems to quantify viable parasites. Methods: Systematic screening of Giardia-specific promoters and luciferase variants led to the development of a strain expressing the click beetle green luciferase. Further modifying this strain to express NanoLuc created a dual reporter strain capable of quantifying parasites in both the trophozoite and cyst stages. These strains were used to develop a high throughput cell assay and a mouse infection model. A library of MetRS inhibitors was screened in the cell assay and 1717 was tested for efficacy in the mouse infection model. Results: Cell viability in in vitro compound screens was quantified via bioluminescence readouts while infection loads in mice were monitored with noninvasive whole-animal imaging and fecal analysis. Compound 1717 was effective in clearing mice of Giardia infection in 3 days at varying doses, which is supported by data from enzymatic and phenotypic cell assays. Conclusions: The new in vitro and in vivo assays based on luciferase expression by engineered G. lamblia strains are useful for the discovery and development of new therapeutics for giardiasis. MetRS inhibitors, as validated by 1717 , have promising anti-giardiasis properties that merit further study as alternative therapeutics.

Giardiasis is widely acknowledged to be a neglected disease in need of new therapeutics to address toxicity and resistance issues associated with the limited available treatment options. We examined seven protein kinases in the Giardia lamblia genome that are predicted to share an unusual structural feature in their active site. This feature, an expanded active site pocket resulting from an atypically small gatekeeper residue, confers sensitivity to “bumped” kinase inhibitors (BKIs), a class of compounds that has previously shown good pharmacological properties and minimal toxicity. An initial phenotypic screen for biological activity using a subset of an in-house BKI library found that 5 of the 36 compounds tested reduced trophozoite growth by at least 50% at a concentration of 5 μM. The cellular localization and the relative expression levels of the seven protein kinases of interest were determined after endogenously tagging the kinases. Essentiality of these kinases for parasite growth and infectivity were evaluated genetically using morpholino knockdown of protein expression to establish those that could be attractive targets for drug design. Two of the kinases were critical for trophozoite growth and attachment. Therefore, recombinant enzymes were expressed, purified and screened against a BKI library of >400 compounds in thermal stability assays in order to identify high affinity compounds. Compounds with substantial thermal stabilization effects on recombinant protein were shown to have good inhibition of cell growth in wild-type G. lamblia and metronidazole-resistant strains of G. lamblia. Our data suggest that BKIs are a promising starting point for the development of new anti-giardiasis therapeutics that do not overlap in mechanism with current drugs. The eukaryotic protozoan Giardia lamblia is the most commonly reported intestinal parasite worldwide. Current treatments used to treat giardiasis include metronidazole and other nitroimidazole derivatives. However, emergence of metronidazole-resistance strains and adverse reactions to the treatments suggest that alternative therapies against giardiasis are necessary. Here we identify a set of protein kinases in the Giardia genome that have an atypically small amino acid residue, called the gatekeeper residue, in the ATP binding pocket. Small gatekeeper residues are rare in mammalian kinases. We investigated whether this subset of kinases is necessary for parasite growth and proliferation and, if so, could they be targeted with a class of compounds called bumped kinase inhibitors (BKIs), designed to exploit the enlarged active site pocket made accessible by the small gatekeeper amino acid. Morpholino knockdown of two of the small gatekeeper kinases produced a distinctive phenotype characterized by defective cytokinesis. This phenotype was mimicked in Giardia cells treated with our most potent BKI. These results suggest that BKIs may be developed to selectively target small gatekeeper kinases in Giardia lamblia to provide a novel treatment option for giardiasis.

The genome of the human intestinal parasite Giardia lamblia contains only a single aminoacyl-tRNA synthetase gene for each amino acid. The Giardia prolyl-tRNA synthetase gene product was originally misidentified as a dual-specificity Pro/Cys enzyme, in part owing to its unexpectedly high off-target activation of cysteine, but is now believed to be a normal representative of the class of archaeal/eukaryotic prolyl-tRNA synthetases. The 2.2 Å resolution crystal structure of the G. lamblia enzyme presented here is thus the first structure determination of a prolyl-tRNA synthetase from a eukaryote. The relative occupancies of substrate (proline) and product (prolyl-AMP) in the active site are consistent with half-of-the-sites reactivity, as is the observed biphasic thermal denaturation curve for the protein in the presence of proline and MgATP. However, no corresponding induced asymmetry is evident in the structure of the protein. No thermal stabilization is observed in the presence of cysteine and ATP. The implied low affinity for the off-target activation product cysteinyl-AMP suggests that translational fidelity in Giardia is aided by the rapid release of misactivated cysteine.

The growth of diffracting crystals from purified proteins is often a major bottleneck in determining structures of biological and medical interest. ThePROSPEROweb server, http://skuld.bmsc.washington.edu/prospero, is intended both to provide a means of organizing the potentially large numbers of experimental characterizations measured from such proteins, and to provide useful guidance for structural biologists who have succeeded in purifying their target protein but have reached an impasse in the difficult and poorly understood process of turning purified protein into well diffracting crystals. These researchers need to decide which of many possible rescue options are worth pursuing, given finite resources. This choice is even more crucial when attempting to solve high-priority but relatively difficult structures of eukaryotic proteins. The site currently uses theHyGX1predictor, which was trained and validated on protein samples from pathogenic protozoa (eukaryotes) using results from six types of experiment.PROSPEROallows users to store, analyze and display multiple results for each sample, to group samples into projects, and to share results and predictions with collaborators.

In choosing and refining any crystallographic structural model, there is tension between the desire to extract the most detailed information possible and the necessity to describe no more than what is justified by the observed data. A more complex model is not necessarily a better model. Thus, it is important to validate the choice of parameters as well as validating their refined values. One recurring task is to choose the best model for describing the displacement of each atom about its mean position. At atomic resolution one has the option of devoting six model parameters (a `thermal ellipsoid') to describe the displacement of each atom. At medium resolution one typically devotes at most one model parameter per atom to describe the same thing (a `B factor'). At very low resolution one cannot justify the use of even one parameter per atom. Furthermore, this aspect of the structure may be described better by an explicit model of bulk displacements, the most common of which is the translation/libration/screw (TLS) formalism, rather than by assigning some number of parameters to each atom individually. One can sidestep this choice between atomic displacement parameters and TLS descriptions by including both treatments in the same model, but this is not always statistically justifiable. The choice of which treatment is best for a particular structure refinement at a particular resolution can be guided by general considerations of the ratio of model parameters to the number of observations and by specific statistics such as the Hamilton R-factor ratio test.

Diseases caused by the apicomplexan protozoans Toxoplasma gondii and Cryptosporidium parvum are a major health concern. The life cycle of these parasites is regulated by a family of calcium-dependent protein kinases (CDPKs) that have no direct homologues in the human host. Fortuitously, CDPK1 from both parasites contains a rare glycine gatekeeper residue adjacent to the ATP-binding pocket. This has allowed creation of a series of C3-substituted pyrazolopyrimidine compounds that are potent inhibitors selective for CDPK1 over a panel of human kinases. Here we demonstrate that selectivity is further enhanced by modification of the scaffold at the C1 position. The explanation for this unexpected result is provided by crystal structures of the inhibitors bound to CDPK1 and the human kinase c-SRC. Furthermore, the insight gained from these studies was applied to transform an alternative ATP-competitive scaffold lacking potency and selectivity for CDPK1 into a low nanomolar inhibitor of this enzyme with no activity against SRC.

Crystallographic structural models for macromolecules have typically included an isotropic displacement parameter B(iso) for each atom. In cases where the structural model instead includes anisotropic displacement parameters U(ij), the derived quantity B(eq) can be substituted for B(iso) for many purposes. B(eq) is not, however, the best predictor of the value B(iso) that would hypothetically have been obtained by direct refinement of an isotropic model. A new entity B(est) is proposed that represents an estimate for B(iso) that minimizes the Kullback-Leibler divergence from a paired anisotropic model. In general B(est)/B(eq) < 1, with the difference between the two values becoming larger for atoms that are more anisotropic. Although this difference does not affect direct refinement of either isotropic or anisotropic models, it is relevant to any analysis that compares isotropic and anisotropic models of the same underlying structure. In particular, it may lead to improved selection of multi-group TLS models based on analysis of an initial isotropic refinement.