Final Report Summary - TRYPOBASE (Nucleobase derivatives as drugs against trypanosomal diseases)
Executive summary:
Leishmaniasis, African trypanosomiasis and Chagas disease are neglected diseases responsible for substantial global morbidity, mortality, and economic adversity, and in most countries where they are endemic, existing strategies for control and treatment are either failing or under serious threat. New tools for combating pathogenic protozoa and the development and exploitation of new drug targets are required. The main objective of this project was the identification of novel compounds for the treatment of the leishmaniases and trypanosomiases.
A two-pronged approach was proposed to discover leads for the treatment of these diseases by targeting nucleoside / nucleotide metabolism.
(1) A phenotypic approach exploring the potential of large collections of novel nucleobase derivatives against parasites.
(2) A target-based approach specifically centred on the development of inhibitors of the enzyme deoxyuridine triphosphate nucleotidohydrolase (dUTPase).
The ultimate aim was to identify compounds that have potent antiparasitic activity and drug-like properties. Expertise in parasite biology, medicinal chemistry and pharmacokinetics has been combined in order to render compounds that exhibit potent activity against protozoa of the Trypanosomatidae family, have drug-like properties and show proof of concept in appropriate models of disease. Our prime objective was to provide leads for optimisation studies and hopefully amenable to clinical development after further work, to give compounds suitable for use in developing countries.
During the course of the project the main activities have been:
(i) screening activities of several chemical compound classes in order to establish antiprotozoal activity for further development;
(ii) medicinal chemistry around confirmed hits and derivatives in order to provide a view of SAR and identify candidates for in vivo studies;
(iii) generate 3D information of kinetoplastid dUTPases and potential drug targets of interest bound to selected ligands;
(iv) obtain information regarding the mode of action of selected chemical classes highly active in vitro;
(v) generation of in vivo pharmacokinetic data with selected compounds;
(vi) evaluation of in vivo activity of leads with good pharmacokinetic profiles.
Several contributions can be highlighted as major findings of TRYPOBASE. We have significantly advanced in our basic understanding of dUTPases and identified novel potential targets and hit compounds active against Kinetoplastid protozoa. Unique methodology has been exploited for establishing mode of action of active compounds. Importantly, as a result of this multidisciplinary effort, we have identified a chemical series that shows potent activity both in vitro and in vivo against Trypanosoma cruzi and drug-like properties. This promising series now needs to proceed through lead optimisation in order to identify analogues that have the potential to rapidly progress down the development pathway toward clinical candidates. The ultimate objective of the project has been fulfilled. We have rendered a highly promising new class of compounds that through future development may provide a new drug for the treatment of Chagas disease.
Project context and objectives:
The protozoan diseases, leishmaniasis, African trypanosomiasis and Chagas disease are responsible for extensive global morbidity, mortality, and economic adversity, and in most countries existing strategies for control are either failing or under serious risk. New tools for combating pathogenic protozoa rely on development and exploitation of new drug targets and/or vaccine candidates as well as efforts at vector control, all of which require detailed understanding of the biology of the pathogen. These diseases are mainly limited to tropical and sub-tropical regions, particularly affecting the third world. However, co-infection of leishmaniasis and HIV is a problem in the European Union (EU). Climate changes, epidemic outbreaks, severe drug resistance, changes in the habitats of parasites and vectors, globalization, commerce and transport all underscore the importance of research in this area.
Various species and sub-species of Leishmania cause leishmaniasis, threatening about 350 million people and giving rise to about 2 million clinical cases each year of which about 25 % are due to the fatal disease visceral leishmaniasis. Trypanosoma brucei rhodesiense and Trypanosoma brucei gambiense cause African trypanosomiasis, with epidemics raging in a number of countries in sub-Saharan Africa, causing an estimated 50 000 cases annually. Trypanosoma cruzi causes Chagas disease; currently 16 - 18 million people are infected with about 100 million at risk from the disease.
Vaccine development is problematic for these diseases. In some cases, vector control has been effective in reducing the incidence of the disease. However, chemotherapy remains the major means for control of these parasitic protozoa, but the current therapeutic agents are unsatisfactory. There are reported cases of resistance to drugs for all of the above diseases. In addition, the available drugs show lack of clinical effectiveness and side effects.
Purine and pyrimidine antimetabolites have been used with great success in the chemotherapy of a wide range of diseases. Examples of the many nucleobases and nucleosides with therapeutic qualities include 6-thioguanine, 5-fluorouracil, and 6-mercaptopurine in leukaemia and cancer therapy; aciclovir, ganciclovir, ribavirin, 3'-azido-3'-deoxythymidine (AZT) and 3'-thia-2',3'-dideoxycytidine (3TC) against viruses; and allopurinol and pyrimethamine against protozoan infections.
The nucleotide metabolism of parasitic protozoa of the genera Leishmania and Trypanosoma present several unique features. These organisms are incapable of de novo synthesis of purines while pyrimidines are made available through both de novo synthesis and salvage. These distinctive characteristics pave the way for evidence-based generation of drugs that exclusively inhibit trypanosomes.
The enzyme deoxyuridine triphosphate nucleotidohydrolase (dUTPase) (EC 3.6.1.23) catalyses the hydrolysis of dUTP to dUMP and PPi and has a dual role in pyrimidine nucleotide metabolism. Firstly, it provides the substrate for thymidylate synthase allowing the formation of dTTP. Secondly, it is responsible for keeping dUTP intracellular levels low, avoiding its mis-incorporation into nascent DNA. The ability to specifically remove the non-canonical nucleotide, dUTP, from the dNTP pool indicates the importance of dUTPase in the preservation of genetic integrity. The enzyme is ubiquitous and is essential for growth and virulence where this has been investigated. It is a drug target that has been validated in several organisms including Trypanosoma. Trypanosomes exhibit a dimeric form of dUTPase that has an entirely different 3D structure and is an analogue, not a homologue, of trimericdUTPases such as the human enzyme. The protozoa lack a trimericdUTPase. Homologues of the trypanosomal enzyme have been identified in the genomes of several Gram-positive bacteria and their phages. DimericdUTPases differ profoundly from the classical trimeric human dUTPase in a number of features. Members of the consortium have fully characterized dimericdUTPases from T. cruzi and L. major. Detailed kinetic studies have shown that the trypanosomal enzymes present exclusive properties such as the capacity to hydrolyse both the nucleoside diphosphate and triphosphate and product inhibition. Structure determinations of the both the native and dUDP-enzyme complex from T. cruzi have revealed a novel protein fold that displays no sequence or structural similarities to previously described dUTPases. The structure is predominantly helical. The molecular unit is a dimer with two active sites. Nucleotide binding promotes extensive structural rearrangements and secondary structure rearrangement. In the apo form, the active site is open; upon substrate binding, the enzyme undergoes a large conformational change and closes around the substrate.
The nature of the binding site is distinctly different from that in the trimeric (including human) dUTPases which are made up of predominantly I²-sheets. In both holo-dimeric and trimeric enzymes the base is buried in the complex, and the ribose also buried to a lesser extent: for both moieties the environment is different in the two families. The very significant structural differences between the dimeric and trimeric enzymes suggested that it would be possible to derive selective inhibitors of the dimeric enzymes.
Work performed by the consortium has shown that the dimeric enzyme is essential for growth in both procyclics and bloodstream forms of T. brucei using RNAi. Thus in this parasite dUTPase is a nuclear enzyme necessary for growth and strong down-regulation caused decreased cell proliferation and increased intracellular levels of dUTP. dUTPase-depleted procyclics presented augmented levels of uracil-DNA glycosylase activity, the first step in base excision repair. The knockdown of activity gave rise to defects in G2/M progression and alterations in cytokinesis. Multiple parasites with a single enlarged nucleus plus one kinetoplast were visualized together with an enhanced population of anucleated cells or zoids.
This project was built on several achievements and observations in the area of nucleotide metabolism:
-Pyrimidine and purine metabolism exhibits unique features in trypanosomes.
-The identification of a unique enzyme involved in pyrimidine metabolism restricted to trypanosomes and essential for viability: the dimeric all-dUTPase.
-Collections of compounds were available through the consortium for antiprotozoal screening and lead identification.
-The consortium brought together a combination of excellent expertise for drug discovery.
The core objective was the identification of new nucleobase derivatives for the treatment of the leishmaniases and trypanosomiases based on the observation that nucleoside / nucleotide metabolism is an attractive route for therapeutic intervention.
This approximation was broad but a strategy was devised that allowed the consortium to develop its focus as the project progressed and to have back-up strategies. The final aim was to synthesise a compound series that showed efficacy in vivo for at least one of the trypanosomatid diseases, with the compounds having drug-like properties.
Thus, a two-pronged approach was proposed to discover new leads for the treatment of leishmaniasis and trypanosomiasis targeting nucleoside / nucleotide metabolism. Both approaches were explored bearing in mind that drug discovery has a high-failure rate. Initial screening covered all three parasites (Leishmaniadonovani, T. brucei and T. cruzi), and at a later phase, focus was placed on the most promising compounds.
Approach 1: The phenotypic approach: The consortium aimed at exploring the potential of large collection of novel nucleobase / nucleoside derivatives against trypanosomal diseases.
Approach 2: The target-based approach: Part of the effort was specifically centred on the development of inhibitors of the enzyme deoxyuridine triphosphate nucleotidohydrolase. The trypanosomal enzyme shows structural and functional characteristics which differ profoundly from the mammalian counterpart. The aim was to identify potent inhibitors active against parasitic protozoa, active in rodent models of infection and with drug-like properties.
Thus, the main expected outcome of the project was the identification of nucleobase derivatives that present activity in vitro against protozoa of the Trypanosomatidae family and drug-like properties in order to provide compounds amenable to further development. A goal was to identify a chemical series that showed efficacy in an animal model of infection in at least one of the trypanosomatid diseases. In addition, it was anticipated that project would contribute significantly to the understanding of dUTPases and nucleotide metabolism in the Trypanosomatidae. While the basic features of this metabolic pathway have been identified, especially since the unravelling of the genomes of these pathogens, certain features relative to the requirements of the synthesis de novo, transport peculiarities and aspects of salvage remained unknown.
Project results:
The proposed scientific activities were organised as six work packages (WPs).
1. High throughput screening and hit identification
2. Design and synthesis of compounds
3. Screening activities
4. ADME-TOX
5. X-ray crystallography
6. Intracellular and molecular mode of action studies.
WP1. High throughput library screening and hit identification
This WP has combined the following screening activities:
Enzyme screening
In order to identify a chemical starting point to prepare inhibitors of the trypanosomatid deoxyuridine nucleotidohydrolase we carried out a high throughput screen of a 62 000 compound library at the University of Dundee and the MEDIVIR library against Trypanosoma cruzi dUTPase. The collection of proprietary compounds including nucleoside and nucleobase derivatives provided by MEDIVIR AB was also put into a screen for identification of dimeric dUTPase inhibitors (Granada). The Dundee screen yielded a number of hits, or compounds that showed some inhibition of the parasite enzyme. Work was carried out to validate these hits, which involved re-purchasing hits and re-assaying them. In some cases, analogues of the initial hits were also purchased and made. Compounds were tested in an orthogonal assay against the enzyme at the CSIC Granada. Lack of inhibition on both the enzyme and parasite growth justified focus of the project beyond the midterm review on promising hits arising only from the phenotypic assay.
Thus, screening activities supported the notion that the enzyme has low druggability and inhibition with small bioavailable molecules is problematic. One of the enzyme primary hits obtained in Dundee was found to inhibit the growth of the parasite Trypanosoma cruzi, although inhibition of deoxyuridine triphosphate nucleotidohydrolase was not confirmed. A phenotypic optimisation of this series was performed. A number of compounds were prepared and assayed. To date, the compound series will require very significant optimisation if it is to be pursued further. Work on this series was put on hold when a more promising series was identified by phenotypic screening (below).
Indeed modelling studies and 3D information (York) show that the enzyme has a very polar active site. In one final set of experiments, the enzyme was screened with the Dundee fragment library and some weak hits were identified. However University of York was not able to obtain robust structural information on binding of these fragments to the enzyme further supporting that inhibitor design for this class of enzymes is challenging although there is still room for further work (see WP5).
Phenotypic screening
A proprietary chemical library, containing a collection of nucleoside and nucleobase derivatives provided by MEDIVIR AB was put into a medium throughput screen for lead identification. This collection included 6979 compounds that have been developed over the years. Compounds were tested against the clinically relevant forms of T. cruzi, T.brucei rhodesiense and L. donovani in Basel (Swiss TPH).
All compounds were assayed using established methodologies for activity determination against the clinically relevant forms of the protozoan parasites: the trypomastigote form of Trypanosoma brucei; the amastigote form of Trypanosoma cruzi cultured in L6-cells; and the amastigote forms of Leishmania donovani. The compounds were also screened against a mammalian cell line (L6-cell) as a measure of cytotoxicity. Compounds showing IC50's less than 20 g/ml against axenic amastigotes of L. donovani were tested against intracellular amstigotes.
Results from the screening were completed 21 October 2009 and structures and data were presented at the project meeting in York 13 November 2009. In this initial screen, 41 compounds inhibited the growth of T. b. rhodesiense more than 50 %, 42 compounds inhibited the growth of T. cruzi and 31 compounds inhibited the growth of L. donovani (axenic amastigotes). Of these, a total of 78 compounds were available for hit confirmation and IC50 determination has been completed. 10 compounds showed good and selective activity (IC50 less than 0.2 g/ml) against T.b.rhodesiense and 11 compounds showed moderate activity (IC50 between 0.2 g/ml and 1.5 g/ml).When the L. donovani hits were followed up in the more relevant macrophage assay, 6 compounds had IC50 less than 10 µg/ml and one compound an IC50 of 2 µg/ml. (Reference compound miltefosine IC50 1.5-2 µg/ml). For T.b.rhodesiense mainly nucleoside analogs emerged as hits.
For T. cruzi, seven hit compounds were of initial interest, with three belonging to the same structural class. One of the seven compounds was selected for further SAR optimisation and the continued SAR investigation of this compound eventually resulted in compounds with low nM T. cruzi activity and in vivo activity. Initial hits were confirmed.
In addition, IC50 values against the three parasites were determined of a collection of nucleobase derivatives provided by the group of Dr Maria Jesus Perez from the Instituto de Quimica Medica CSIC, Madrid. This activity did not lead to the identification of new hits.
WP2. Design and synthesis of compounds
Medicinal chemistry and hit to lead optimisation
Hit compounds obtained in dUTPase screening
A small amount of compounds were designed and synthesised for testing in the phenotypic and the target-based assay of the dimeric dUTPases in Dundee. These activities did not allow for the identification of compounds that justified further optimisation.
Hit compounds obtained the phenotypic screen
As mentioned above, high through screening of the Medivir library identified hit compounds that had promising activity against T. cruzi.
The promising T. cruzi hit compound from the phenotypic screening of Medivir's library was selected for SAR optimization in a medicinal chemistry program by Medivir and Syngene.
The hit compound was drug-like with desirable characteristics. Medivir/ Syngene explored the SAR space by systematic variations of compound features and properties, starting late 2009. This SAR mapping established a solid foundation, clarifying important SAR requirements for potent and selective inhibition of T cruzi in vitro. For increased effort at this promising stage, University of Dundee joined the Medivir / Syngene SAR optimisation program in May - June 2011, with additional medicinal chemistry and DMPK. Combined work has resulted in potent compounds with low nM T. cruzi activity in vitro. Potent in vivo activity has been also demonstrated in a proof of concept study.
Therefore in May - June 2011, Dundee, Medivir and Syngene joined efforts in following up this hit series. The chemistry was driven by the results from screening compounds. Decision making on progression of particular compounds was based on specific criteria relate to potency, selectivity, solubility and metabolic stability. A significant break-through was made when the activity against the parasite was increased by approximately 20-fold. Dundee, Medivir and Syngene, then focused on improving the pharmacokinetic properties of the molecule. Eventually several molecules were prepared that had activity and pharmacokinetic properties suitable for evaluation in a rodent model of disease. One of these molecules showed good activity against the rodent model of Chagas disease, both parentally and orally. Further work will be required to optimise this series further, but this marks a key goal of the project. Parasite screening for these series were carried out by the Swiss Tropical and Public Health Institute and Granada, York and Montevideo have participated in studies aimed at establishing potential mode of action.
WP3. Screening activities
Phenotypic screening activities
Summary of phenotypic screening activities performed during hit to lead optimization
Over 400 compounds coming from the medicinal chemistry programs have been tested against the 3 parasites and for cytotoxicity during the length of the project in Basel. 12 compounds showed good and selective activity against T. b. rhodesiense (IC50 less than 0.2 µg/ml, selectivity index (SI) ¥ 100). 53 compounds showed activity against axenic amastigotes of L. donovani (IC50 less than 5 µg/ml, SI greater than 10). These compounds were tested against intracellular amastigotes of L. donovani (macrophage assay). 2 compounds showed moderate activity (IC50 between 2 and 5 µg/ml, SI greater than 5). Most of the tested compounds lack selectivity, only 2 compounds showed moderate activity (IC50 greater than 2 µg/ml, SI greater than 5). A total of 38 compounds so far have been made that showed good and selective activity against T. cruzi (IC50 less than 1 µg/ml, SI ¥ 50).
In the second half of the project the medicinal chemistry program was mainly focused on the improvement of activity against T. cruzi to obtain proof of principle in vivo. To select the best T. cruzi active compounds, IC90 values were determined for selected compounds and the parasite growth inhibition curves compared. Considering the DMPK and in vitro test data, two compounds were selected for in vivo studies. Both compounds showed in vivo efficacy in a T. cruzi acute mouse model.
Enzyme inhibitor screening
Multiple compounds coming from the Dundee HTS against dUTPase were tested for inhibition of T. cruzi and Leishmania dUTPase in a spectrophotometric proton release assay specifically designed for this class of enzymes (Granada) and in fluorimetric ligand binding assays in order to establish potency (York and Granada). A subset was also tested against human dUTPase for selectivity. Additionally, compounds coming from previous collections were tested. In total, 126 compounds have been tested in Granada. Unfortunately, most of the compounds were not confirmed as inhibitors and further development was not prioritised.
Thymidine kinase has been also explored as a novel drug target. Thus we have thoroughly characterised the Leishmania major recombinant enzyme (Granada) and established properties regarding substrate specificity and inhibition profiles. A collection of 47 thymidine analogues was tested against the enzyme. Due to lack of significant inhibition and antiprotozoal activity, this avenue was not further pursued.
Screening in vivo in animal models of infection
Testing in vivo was performed for specific compounds of the T. cruzi active chemical series that exhibited a good PK profile and metabolic stability. Very good efficacy and proof of concept was obtained in an acute mouse model.
WP4. ADME-Tox
In the T. cruzi structure activity optimisation phase, all new compounds were sent to partner 2 for determination of in vitro DMPK properties and selected compounds were tested for in vivo DMPK. The resulting information has allowed for the selection of two candidates for in vivo studies.
WP5. X-ray crystallography
The aim of partner 3 (York) was to determine the crystal structures of T. brucei dUTPase and co-crystals of T. cruzi, T. brucei and L. major dUTPases with ligands to establish the requirements for inhibitor binding and to provide information that would aid in improved inhibitor design. In addition the crystallization of other proteins identified as potential targets of compounds arising in the phenotypic screens was addressed.
Main achievements are:
1. dUTPases:
3D structures of all three dimeric dUTPases in complex with natural substrate analogues.
Proof of mechanism with attack on the beta-phosphate by structure, fluorescence and NMR.
First structure of an inhibitory ligand bound to the open conformation.
2. Other targets:
Structures of T. brucei adenosine kinase and leucyl aminopeptidase with ligands.
Preliminary expression of thymidylate kinase in baculovirus.
Dimeric dUTPases from L. major, T. brucei and T. cruzi.
L. major dUTPase.
No crystals were obtained for the open form. For the closed form, new complexes were crystallised with dUpNpp, the product dUMP and deoxyuridine (dU). The enzyme binds different numbers of metals dependent upon the identity of the substrate. The proposed nucleophilic water molecule is evident. Crystals of the dUMP and dU complexes were both in the completely closed conformation. Tryptophan fluorescence quenching showed that dUpNpp binds to the enzyme with a Kd of approximately 1.8 µM in the presence of Mg2+. PPi had little effect upon dUMP binding but had a major effect upon dU binding confirming that the presence of negative charge in the phosphate binding region is vital.
T. brucei dUTPase
The apo structure was similar to but more open than for T. cruzi. The closed form was very similar those of the other dimeric dUTPases with identical arrangement of the 5 substrate binding motifs and similar metal binding.
T. cruzi dUTPase
Previous open and closed structures had been obtained by partner 1 but good-diffracting were consistently harder to produce for this species. The structure of the dUpNpp complex was solved with different numbers of metals bound in the two active sites.
Ligands provided by partner 2 for crystallisation screening
Partner 2 first provided a series of compounds with uracil rings with carbon linkers of varying lengths to a phosphate group for co-crystallisation and assay of binding strength using tryptophan fluorescence quenching. The assay revealed very weak binding with Kd's of several hundred micromolar and no crystals were obtained. Partner 2 then provided a series of compounds identified as possible T. cruzi dUTPase inhibitors in high throughput screening, but these compounds had low solubility in water and extensive crystallisation trials gave no useful quality crystals. Partner 2 performed fragment screening using SPR to identify new compounds and the best hits were provided for screening. Crystals were obtained with the T. brucei enzyme but the structures were all of the open form and all but one lacked any bound ligand. The structure of T. brucei dUTPase with this ligand at 2.8 resolution revealed that it binds to the open form but the ligand density was weak. This hit occurred at the end of the contract and could be usefully pursued. Crystals of the T. cruzi enzyme with this compound bound were not sufficiently good to allow data collection.
Transition state analogue studies
Aluminium and magnesium fluorides were successfully used as planar transition state analogues to validate the mechanism. Tryptophan fluorescence quenching with the T. brucei enzyme clearly showed that AlF3 has a dramatic effect upon binding of dUMP but not of dU suggesting that the proposed attack by water on the phosphate was correct. The increase in affinity for dUMP in the presence of AlF3 and PO43- was followed up using MgF3- as a transition state analogue yielding similar results. Diffracting crystals of both the L. major and T. brucei dUTPases were obtained in the presence of dUMP, AlF3 and PO42- and provided structural confirmation of the mechanism. Structures of T. brucei dUTPase with dUMP and both [AlF3-OPO3] and MgF3- reveal the interactions between the protein and the planar transition state. 31P-NMR was used to probe the products of the dUTPase reaction and confirmed attack on the beta-phosphate highlighting this important difference between the trypanosomal and human proteins.
2. Other targets.
Thymidine (TK) and thymidylate (TMPK) kinases
With partner 1, these kinases involved in nucleotide metabolism from the three TriTryps were identified as potential drug targets. Extensive expression trials in E. coli focused on the TK set, but in spite of there being structures of homologues in the PDB, these proved impossible to express in stable soluble form. In attempts to improve the yield of L. major TK and to obtain soluble protein expression of TKs of T. brucei and T. cruzi, 30 different constructs were cloned and tested for expression with the baculovirus system in collaboration with the Oxford Protein Production Facility (OPPF). Some successfully expressed on a small scale and are being up-scaled to bigger culture volumes subsequent to the end of the contract.
Adenosine kinase (AK)
AK was proposed as a potential target at a consortium meeting as it activates the adenosine analogue cordycepin, highly active against T. brucei. The structure of TbAK was solved in the apo open form and closed form with adenosine and AMPPNP, which echoes the binding mode of the alternative substrate cordycepin. This gives direction for possible modifications of cordycepin that might improve binding affinity and resistance to deamination. Structural work is on-going. The structure was independently reported by another group (Kuettel, S. et al., 2011).
Leucyl aminopeptidases
At the consortium meeting in Brussels in month 37, the leucyl aminopeptidase (LAP) was proposed as a potential target for hit compounds by partner 7. Partner 3 decided to clone and express the LAPs of all three Tritryps with the aim of determining their crystal structures. The scope was broadened to include a second putative endopeptidase (AP2) and a puromycin aminopeptidase (PSA), from all three organisms. The LAP2 from all three TriTryps were successfully cloned and over-expressed and crystals obtained. The structure of T. brucei LAP was solved in its apo form to a resolution and bound to the inhibitor actinonin. There are no changes of conformation upon ligand binding. An independent structure of T. brucei LAP has recently been deposited in the PDB by the Structural Genomics Consortium (PDB 4efd). While other partners subsequently established that LAP was not the target of hit compounds, the LAP project is being actively pursued by partner 3 subsequent to the end of the contract.
P6. Intracellular and molecular mode of action studies:
1. Studies aimed at understanding the mode of action of series active against Trypanosoma cruzi have been accomplished. Several approaches have been used:
a) Capture of protein targets by proteomics using affinity chromatography. After using immobilised benznidazole as proof of concept for capture experiments, we received compounds from the consortium for immobilisation and the identification of the eluted proteins by mass spectrometry (Montevideo). In addition, an analysis of differentially expressed proteins upon drug exposure was studied by 2D electrophoresis and MALDI TOF/TOF.
b) Metabolomics. The metabolic profiling in response to selected compounds was studied in Trypanosoma cruzi (Montevideo).
c) Induction of resistance. Parasites resistant to the most active compounds were generated (Granada and Montevideo).
d) Selected compounds active against T. cruzi have been evaluate to establish static/cidal mode of killing (Granada).
e) Evaluation of inhibition of aminopeptidases as potential targets of specific hit compounds has been accomplished (Granada).
2. Studies on the mode of action of adenosine derivatives exhibiting activity against Trypanosoma brucei
Here a small number of adenosine derivatives made in the medicinal chemistry programme (Syngene) were tested for deamination by adenosine deaminase and phosphorylation by adenosine kinase (Granada). IC50 values and enzymatic data of metabolic conversion of these compounds suggested that a significant amount of work would be necessary for optimization. This avenue was not further pursued yet provided interesting information regarding structural requirements for deamination and phosphorylation of adenosine analogues.
Potential impact:
This project has directly addressed research in neglected infectious diseases and the development of new tools to control infections due to parasites of the Trypanosomatidae family. Other overall objectives covered include translational research in major infectious diseases and innovative therapeutic approaches and interventions.
The project was focused initially on all three diseases (Trypanosomiasis, Chagas disease, Leishmaniasis) and the main objective was to develop new interventions by means of the development of new drugs.
Due to the nature of these diseases, it is expected that the research outcome will be applicable and affordable for use in disease endemic countries and contribute to reduce incidence and impact of neglected tropical diseases. In addition, the project has contributed significantly to the strengthening of the capacity building of endemic countries for understanding and control of these diseases through networking and training scientists through the participation of partners from India and Uruguay.
The effort was clearly a goal oriented public-private partnership set up for drug discovery and aimed at translating research for human health: basic discoveries will be used in the development of practical benefits in order to ultimately improve the quality of life. Discovery and development of new chemotherapeutic tools through partnerships with the private sector is a major tool to develop new strategies of control.
The project included multidisciplinary discovery activities, the development and testing of new therapeutic interventions in vitro and in vivo, lead optimisation, investigation of drug metabolism, pharmacokinetics and preliminary safety of several new chemical entities before achieving the final candidate drug (CD) target profile anti-viral nucleoside analogues, some of reported toxicity.
Over the duration of the project, several young researchers have been involved in the implementation of research activities. New knowledge and techniques have been incorporated in the undergraduate and post-graduate training of the university partners on drug design and development and have been used by the CSIC and SWISS TPH in their training programmes.
Social and economical impact on American trypanosomiasis
Here we have rendered promising leads to be optimised for the treatment of Chagas disease. American trypanosomiasis (Chagas disease) occurs throughout Mexico and central and southern America, and continues to pose a serious threat to health in many countries of the region. Furthermore, due to recent significant population migrations from endemic countries towards developed countries, the threat of the disease is expanding to reach areas outside the traditional geographic boundaries; this is especially relevant in the context of transmission through blood transfusion. The overall prevalence of human Trypanosoma cruzi infection is estimated at 16 - 18 million cases. Approximately 120 million people, i.e. 25 % of the inhabitants of Latin America, are at risk of contracting the infection.
Two drugs, nifurtimox and benznidazole, are capable of curing at least 50 % of recent infections. These products are active in the acute and short-term (up to a few years) chronic phase. However, they have little or no activity in long-term chronic forms of the disease. In addition, both drugs have serious and frequent side-effects so their use is limited. The results presented here may contribute to the development of new control tools, such as safer drugs.
This international research cooperation project in the area of human health will help to stimulate global competitiveness, and is directly related to some of Europe's international commitments. Likewise cooperation will provide knowledge to partners about institutions in other parts of the world, help deliver European health policies, and create a favourable environment for other forms of international alliances. Finally this international research and development project has contributed to the production of global public goods and help to close the gap between different countries in the world.
A reduction of poverty by improving health conditions has for a long time been one of the commitments of the EU. However, investing in research of neglected diseases is not only an act of solidarity but also a global need due to their potential threat. Factors such as high mobility of people, centralised food handling and climate change contribute to an increased global vulnerability. This is manifested in outbreaks and rapid spread of infectious diseases across all continents and regardless of income levels. New epidemics have been observed in the last 20 years, and more emerging or re-emerging diseases (resulting from antimicrobial resistance) are expected. There is an urgent need for research in these illnesses.
Dissemination of project results
Protection of promising compounds by patent is required and this is currently under consideration by all partners. This will place some delay on publication. Specific contacts are being made with the private sector for further development of the promising lead chemical series active against Trypanosoma cruzi. Then results regarding promising compounds will be presented through publication in leading, international scientific journals and at international meetings on chemotherapy and drug design. Results concerning dUTPases are already published or under consideration for publication. Manuscripts concerning novel molecular drug targets are in preparation.
Exploitation of results
The project has given rise to potential discoveries in several areas. As already indicated it was a primary goal of the project to identify new leads and to give rise to potential therapeutic agents for treatment of diseases produced by the Trypanosomatidae. Several categories of compounds have been identified some of which would require substantial optimisation. Here we have only optimised the most promising considering the resources available. Since the promising lead compounds active against T. cruzi both in vitro and in vivo require further structural optimisation, future funding sources must be identified.
We have contributed significantly to the knowledge of the structure and function of dUTPases. While high affinity ligands with antiprotozoal activity have not been identified, high resolution structures are available to the scientific community for further efforts in inhibitor design. The information is also relevant to the understanding of the structure-function of the all-nucleotidohydrolase family of enzymes. Other advances include the characterisation of novel potential drug targets in the area of nucleotide metabolism and the application of state of the art methodology for understanding mode of action of selected compounds.
There is an urgent need for the development of new drugs for the treatment of Chagas disease. New, more effective drugs could have a significant impact on the lives of millions of people, particularly in South America. The compound series that has been developed has the potential for exciting further development. This will require further optimisation, particularly in terms of solubility and metabolic stability and evaluation in a whole variety of efficacy, pharmacokinetic and safety pharmacology assays. A patenting strategy is being agreed between the partners. We are actively investigating mechanisms to ensure that the compound series is developed further and are seeking funding sources.
Project website: http://www.ipb.csic.es/trypobase/index.html
Coordinator partner1
Dolores Gonzalez Pacanowska
Instituto de Parasitologia y Biomedicina 'Lopez-Neyra'
Consejo Superior de Investigaciones Cientificas
Avda. del Conocimiento s/n
Parque Tecnologico de Ciencias de la Salud
18100 - Armilla .Granada,
Spain
E-mail: dgonzalez@ipb.csic.es
Partner 2
Ian Gilbert
College of Life Sciences
University of Dundee
Sir James Black Centre
Dundee DD1 5EH,
UK
E-mail: I.H.Gilbert@dundee.ac.uk
Partner 3
Keith Wilson
Structure Biology Laboratory
Department of Chemistry
University of York
Heslington
York YO10 5YW,
UK
E-mail: keith@ysbl.york.ac.uk
Partner 4
Reto Brun
Swiss Tropical and Public Health Institute
Parasite Chemotherapy
Socinstr. 57
P.O. Box
CH-4002 Basel,
Switzerland
E-mail: reto.brun@unibas.ch
Partner 5
Nils Gunnar Johansson
Medivir AB
Lunastigen 7
S-141 44 Huddinge,
Sweden
E-mail: NilsGunnar.Johansson@medivir.com
Partner 6
Ashis Baran Mandal
Syngene Intl Ltd.
Biocon Park, Plot 2 and 3,
Bommasandra Industrial Estate - Phase-IV
Bommasandra-Jigani Link Road,
Bangalore 560 099,
India
E-mail: ashis.mandal@syngeneintl.com
Partner 7
Carlos Robello
Unidad de Biologia Molecular
Institut Pasteur de Montevideo
Mataojo 2020 CP11400
Montevideo, Uruguay
E-mail: robello@pasteur.edu.uy
Leishmaniasis, African trypanosomiasis and Chagas disease are neglected diseases responsible for substantial global morbidity, mortality, and economic adversity, and in most countries where they are endemic, existing strategies for control and treatment are either failing or under serious threat. New tools for combating pathogenic protozoa and the development and exploitation of new drug targets are required. The main objective of this project was the identification of novel compounds for the treatment of the leishmaniases and trypanosomiases.
A two-pronged approach was proposed to discover leads for the treatment of these diseases by targeting nucleoside / nucleotide metabolism.
(1) A phenotypic approach exploring the potential of large collections of novel nucleobase derivatives against parasites.
(2) A target-based approach specifically centred on the development of inhibitors of the enzyme deoxyuridine triphosphate nucleotidohydrolase (dUTPase).
The ultimate aim was to identify compounds that have potent antiparasitic activity and drug-like properties. Expertise in parasite biology, medicinal chemistry and pharmacokinetics has been combined in order to render compounds that exhibit potent activity against protozoa of the Trypanosomatidae family, have drug-like properties and show proof of concept in appropriate models of disease. Our prime objective was to provide leads for optimisation studies and hopefully amenable to clinical development after further work, to give compounds suitable for use in developing countries.
During the course of the project the main activities have been:
(i) screening activities of several chemical compound classes in order to establish antiprotozoal activity for further development;
(ii) medicinal chemistry around confirmed hits and derivatives in order to provide a view of SAR and identify candidates for in vivo studies;
(iii) generate 3D information of kinetoplastid dUTPases and potential drug targets of interest bound to selected ligands;
(iv) obtain information regarding the mode of action of selected chemical classes highly active in vitro;
(v) generation of in vivo pharmacokinetic data with selected compounds;
(vi) evaluation of in vivo activity of leads with good pharmacokinetic profiles.
Several contributions can be highlighted as major findings of TRYPOBASE. We have significantly advanced in our basic understanding of dUTPases and identified novel potential targets and hit compounds active against Kinetoplastid protozoa. Unique methodology has been exploited for establishing mode of action of active compounds. Importantly, as a result of this multidisciplinary effort, we have identified a chemical series that shows potent activity both in vitro and in vivo against Trypanosoma cruzi and drug-like properties. This promising series now needs to proceed through lead optimisation in order to identify analogues that have the potential to rapidly progress down the development pathway toward clinical candidates. The ultimate objective of the project has been fulfilled. We have rendered a highly promising new class of compounds that through future development may provide a new drug for the treatment of Chagas disease.
Project context and objectives:
The protozoan diseases, leishmaniasis, African trypanosomiasis and Chagas disease are responsible for extensive global morbidity, mortality, and economic adversity, and in most countries existing strategies for control are either failing or under serious risk. New tools for combating pathogenic protozoa rely on development and exploitation of new drug targets and/or vaccine candidates as well as efforts at vector control, all of which require detailed understanding of the biology of the pathogen. These diseases are mainly limited to tropical and sub-tropical regions, particularly affecting the third world. However, co-infection of leishmaniasis and HIV is a problem in the European Union (EU). Climate changes, epidemic outbreaks, severe drug resistance, changes in the habitats of parasites and vectors, globalization, commerce and transport all underscore the importance of research in this area.
Various species and sub-species of Leishmania cause leishmaniasis, threatening about 350 million people and giving rise to about 2 million clinical cases each year of which about 25 % are due to the fatal disease visceral leishmaniasis. Trypanosoma brucei rhodesiense and Trypanosoma brucei gambiense cause African trypanosomiasis, with epidemics raging in a number of countries in sub-Saharan Africa, causing an estimated 50 000 cases annually. Trypanosoma cruzi causes Chagas disease; currently 16 - 18 million people are infected with about 100 million at risk from the disease.
Vaccine development is problematic for these diseases. In some cases, vector control has been effective in reducing the incidence of the disease. However, chemotherapy remains the major means for control of these parasitic protozoa, but the current therapeutic agents are unsatisfactory. There are reported cases of resistance to drugs for all of the above diseases. In addition, the available drugs show lack of clinical effectiveness and side effects.
Purine and pyrimidine antimetabolites have been used with great success in the chemotherapy of a wide range of diseases. Examples of the many nucleobases and nucleosides with therapeutic qualities include 6-thioguanine, 5-fluorouracil, and 6-mercaptopurine in leukaemia and cancer therapy; aciclovir, ganciclovir, ribavirin, 3'-azido-3'-deoxythymidine (AZT) and 3'-thia-2',3'-dideoxycytidine (3TC) against viruses; and allopurinol and pyrimethamine against protozoan infections.
The nucleotide metabolism of parasitic protozoa of the genera Leishmania and Trypanosoma present several unique features. These organisms are incapable of de novo synthesis of purines while pyrimidines are made available through both de novo synthesis and salvage. These distinctive characteristics pave the way for evidence-based generation of drugs that exclusively inhibit trypanosomes.
The enzyme deoxyuridine triphosphate nucleotidohydrolase (dUTPase) (EC 3.6.1.23) catalyses the hydrolysis of dUTP to dUMP and PPi and has a dual role in pyrimidine nucleotide metabolism. Firstly, it provides the substrate for thymidylate synthase allowing the formation of dTTP. Secondly, it is responsible for keeping dUTP intracellular levels low, avoiding its mis-incorporation into nascent DNA. The ability to specifically remove the non-canonical nucleotide, dUTP, from the dNTP pool indicates the importance of dUTPase in the preservation of genetic integrity. The enzyme is ubiquitous and is essential for growth and virulence where this has been investigated. It is a drug target that has been validated in several organisms including Trypanosoma. Trypanosomes exhibit a dimeric form of dUTPase that has an entirely different 3D structure and is an analogue, not a homologue, of trimericdUTPases such as the human enzyme. The protozoa lack a trimericdUTPase. Homologues of the trypanosomal enzyme have been identified in the genomes of several Gram-positive bacteria and their phages. DimericdUTPases differ profoundly from the classical trimeric human dUTPase in a number of features. Members of the consortium have fully characterized dimericdUTPases from T. cruzi and L. major. Detailed kinetic studies have shown that the trypanosomal enzymes present exclusive properties such as the capacity to hydrolyse both the nucleoside diphosphate and triphosphate and product inhibition. Structure determinations of the both the native and dUDP-enzyme complex from T. cruzi have revealed a novel protein fold that displays no sequence or structural similarities to previously described dUTPases. The structure is predominantly helical. The molecular unit is a dimer with two active sites. Nucleotide binding promotes extensive structural rearrangements and secondary structure rearrangement. In the apo form, the active site is open; upon substrate binding, the enzyme undergoes a large conformational change and closes around the substrate.
The nature of the binding site is distinctly different from that in the trimeric (including human) dUTPases which are made up of predominantly I²-sheets. In both holo-dimeric and trimeric enzymes the base is buried in the complex, and the ribose also buried to a lesser extent: for both moieties the environment is different in the two families. The very significant structural differences between the dimeric and trimeric enzymes suggested that it would be possible to derive selective inhibitors of the dimeric enzymes.
Work performed by the consortium has shown that the dimeric enzyme is essential for growth in both procyclics and bloodstream forms of T. brucei using RNAi. Thus in this parasite dUTPase is a nuclear enzyme necessary for growth and strong down-regulation caused decreased cell proliferation and increased intracellular levels of dUTP. dUTPase-depleted procyclics presented augmented levels of uracil-DNA glycosylase activity, the first step in base excision repair. The knockdown of activity gave rise to defects in G2/M progression and alterations in cytokinesis. Multiple parasites with a single enlarged nucleus plus one kinetoplast were visualized together with an enhanced population of anucleated cells or zoids.
This project was built on several achievements and observations in the area of nucleotide metabolism:
-Pyrimidine and purine metabolism exhibits unique features in trypanosomes.
-The identification of a unique enzyme involved in pyrimidine metabolism restricted to trypanosomes and essential for viability: the dimeric all-dUTPase.
-Collections of compounds were available through the consortium for antiprotozoal screening and lead identification.
-The consortium brought together a combination of excellent expertise for drug discovery.
The core objective was the identification of new nucleobase derivatives for the treatment of the leishmaniases and trypanosomiases based on the observation that nucleoside / nucleotide metabolism is an attractive route for therapeutic intervention.
This approximation was broad but a strategy was devised that allowed the consortium to develop its focus as the project progressed and to have back-up strategies. The final aim was to synthesise a compound series that showed efficacy in vivo for at least one of the trypanosomatid diseases, with the compounds having drug-like properties.
Thus, a two-pronged approach was proposed to discover new leads for the treatment of leishmaniasis and trypanosomiasis targeting nucleoside / nucleotide metabolism. Both approaches were explored bearing in mind that drug discovery has a high-failure rate. Initial screening covered all three parasites (Leishmaniadonovani, T. brucei and T. cruzi), and at a later phase, focus was placed on the most promising compounds.
Approach 1: The phenotypic approach: The consortium aimed at exploring the potential of large collection of novel nucleobase / nucleoside derivatives against trypanosomal diseases.
Approach 2: The target-based approach: Part of the effort was specifically centred on the development of inhibitors of the enzyme deoxyuridine triphosphate nucleotidohydrolase. The trypanosomal enzyme shows structural and functional characteristics which differ profoundly from the mammalian counterpart. The aim was to identify potent inhibitors active against parasitic protozoa, active in rodent models of infection and with drug-like properties.
Thus, the main expected outcome of the project was the identification of nucleobase derivatives that present activity in vitro against protozoa of the Trypanosomatidae family and drug-like properties in order to provide compounds amenable to further development. A goal was to identify a chemical series that showed efficacy in an animal model of infection in at least one of the trypanosomatid diseases. In addition, it was anticipated that project would contribute significantly to the understanding of dUTPases and nucleotide metabolism in the Trypanosomatidae. While the basic features of this metabolic pathway have been identified, especially since the unravelling of the genomes of these pathogens, certain features relative to the requirements of the synthesis de novo, transport peculiarities and aspects of salvage remained unknown.
Project results:
The proposed scientific activities were organised as six work packages (WPs).
1. High throughput screening and hit identification
2. Design and synthesis of compounds
3. Screening activities
4. ADME-TOX
5. X-ray crystallography
6. Intracellular and molecular mode of action studies.
WP1. High throughput library screening and hit identification
This WP has combined the following screening activities:
Enzyme screening
In order to identify a chemical starting point to prepare inhibitors of the trypanosomatid deoxyuridine nucleotidohydrolase we carried out a high throughput screen of a 62 000 compound library at the University of Dundee and the MEDIVIR library against Trypanosoma cruzi dUTPase. The collection of proprietary compounds including nucleoside and nucleobase derivatives provided by MEDIVIR AB was also put into a screen for identification of dimeric dUTPase inhibitors (Granada). The Dundee screen yielded a number of hits, or compounds that showed some inhibition of the parasite enzyme. Work was carried out to validate these hits, which involved re-purchasing hits and re-assaying them. In some cases, analogues of the initial hits were also purchased and made. Compounds were tested in an orthogonal assay against the enzyme at the CSIC Granada. Lack of inhibition on both the enzyme and parasite growth justified focus of the project beyond the midterm review on promising hits arising only from the phenotypic assay.
Thus, screening activities supported the notion that the enzyme has low druggability and inhibition with small bioavailable molecules is problematic. One of the enzyme primary hits obtained in Dundee was found to inhibit the growth of the parasite Trypanosoma cruzi, although inhibition of deoxyuridine triphosphate nucleotidohydrolase was not confirmed. A phenotypic optimisation of this series was performed. A number of compounds were prepared and assayed. To date, the compound series will require very significant optimisation if it is to be pursued further. Work on this series was put on hold when a more promising series was identified by phenotypic screening (below).
Indeed modelling studies and 3D information (York) show that the enzyme has a very polar active site. In one final set of experiments, the enzyme was screened with the Dundee fragment library and some weak hits were identified. However University of York was not able to obtain robust structural information on binding of these fragments to the enzyme further supporting that inhibitor design for this class of enzymes is challenging although there is still room for further work (see WP5).
Phenotypic screening
A proprietary chemical library, containing a collection of nucleoside and nucleobase derivatives provided by MEDIVIR AB was put into a medium throughput screen for lead identification. This collection included 6979 compounds that have been developed over the years. Compounds were tested against the clinically relevant forms of T. cruzi, T.brucei rhodesiense and L. donovani in Basel (Swiss TPH).
All compounds were assayed using established methodologies for activity determination against the clinically relevant forms of the protozoan parasites: the trypomastigote form of Trypanosoma brucei; the amastigote form of Trypanosoma cruzi cultured in L6-cells; and the amastigote forms of Leishmania donovani. The compounds were also screened against a mammalian cell line (L6-cell) as a measure of cytotoxicity. Compounds showing IC50's less than 20 g/ml against axenic amastigotes of L. donovani were tested against intracellular amstigotes.
Results from the screening were completed 21 October 2009 and structures and data were presented at the project meeting in York 13 November 2009. In this initial screen, 41 compounds inhibited the growth of T. b. rhodesiense more than 50 %, 42 compounds inhibited the growth of T. cruzi and 31 compounds inhibited the growth of L. donovani (axenic amastigotes). Of these, a total of 78 compounds were available for hit confirmation and IC50 determination has been completed. 10 compounds showed good and selective activity (IC50 less than 0.2 g/ml) against T.b.rhodesiense and 11 compounds showed moderate activity (IC50 between 0.2 g/ml and 1.5 g/ml).When the L. donovani hits were followed up in the more relevant macrophage assay, 6 compounds had IC50 less than 10 µg/ml and one compound an IC50 of 2 µg/ml. (Reference compound miltefosine IC50 1.5-2 µg/ml). For T.b.rhodesiense mainly nucleoside analogs emerged as hits.
For T. cruzi, seven hit compounds were of initial interest, with three belonging to the same structural class. One of the seven compounds was selected for further SAR optimisation and the continued SAR investigation of this compound eventually resulted in compounds with low nM T. cruzi activity and in vivo activity. Initial hits were confirmed.
In addition, IC50 values against the three parasites were determined of a collection of nucleobase derivatives provided by the group of Dr Maria Jesus Perez from the Instituto de Quimica Medica CSIC, Madrid. This activity did not lead to the identification of new hits.
WP2. Design and synthesis of compounds
Medicinal chemistry and hit to lead optimisation
Hit compounds obtained in dUTPase screening
A small amount of compounds were designed and synthesised for testing in the phenotypic and the target-based assay of the dimeric dUTPases in Dundee. These activities did not allow for the identification of compounds that justified further optimisation.
Hit compounds obtained the phenotypic screen
As mentioned above, high through screening of the Medivir library identified hit compounds that had promising activity against T. cruzi.
The promising T. cruzi hit compound from the phenotypic screening of Medivir's library was selected for SAR optimization in a medicinal chemistry program by Medivir and Syngene.
The hit compound was drug-like with desirable characteristics. Medivir/ Syngene explored the SAR space by systematic variations of compound features and properties, starting late 2009. This SAR mapping established a solid foundation, clarifying important SAR requirements for potent and selective inhibition of T cruzi in vitro. For increased effort at this promising stage, University of Dundee joined the Medivir / Syngene SAR optimisation program in May - June 2011, with additional medicinal chemistry and DMPK. Combined work has resulted in potent compounds with low nM T. cruzi activity in vitro. Potent in vivo activity has been also demonstrated in a proof of concept study.
Therefore in May - June 2011, Dundee, Medivir and Syngene joined efforts in following up this hit series. The chemistry was driven by the results from screening compounds. Decision making on progression of particular compounds was based on specific criteria relate to potency, selectivity, solubility and metabolic stability. A significant break-through was made when the activity against the parasite was increased by approximately 20-fold. Dundee, Medivir and Syngene, then focused on improving the pharmacokinetic properties of the molecule. Eventually several molecules were prepared that had activity and pharmacokinetic properties suitable for evaluation in a rodent model of disease. One of these molecules showed good activity against the rodent model of Chagas disease, both parentally and orally. Further work will be required to optimise this series further, but this marks a key goal of the project. Parasite screening for these series were carried out by the Swiss Tropical and Public Health Institute and Granada, York and Montevideo have participated in studies aimed at establishing potential mode of action.
WP3. Screening activities
Phenotypic screening activities
Summary of phenotypic screening activities performed during hit to lead optimization
Over 400 compounds coming from the medicinal chemistry programs have been tested against the 3 parasites and for cytotoxicity during the length of the project in Basel. 12 compounds showed good and selective activity against T. b. rhodesiense (IC50 less than 0.2 µg/ml, selectivity index (SI) ¥ 100). 53 compounds showed activity against axenic amastigotes of L. donovani (IC50 less than 5 µg/ml, SI greater than 10). These compounds were tested against intracellular amastigotes of L. donovani (macrophage assay). 2 compounds showed moderate activity (IC50 between 2 and 5 µg/ml, SI greater than 5). Most of the tested compounds lack selectivity, only 2 compounds showed moderate activity (IC50 greater than 2 µg/ml, SI greater than 5). A total of 38 compounds so far have been made that showed good and selective activity against T. cruzi (IC50 less than 1 µg/ml, SI ¥ 50).
In the second half of the project the medicinal chemistry program was mainly focused on the improvement of activity against T. cruzi to obtain proof of principle in vivo. To select the best T. cruzi active compounds, IC90 values were determined for selected compounds and the parasite growth inhibition curves compared. Considering the DMPK and in vitro test data, two compounds were selected for in vivo studies. Both compounds showed in vivo efficacy in a T. cruzi acute mouse model.
Enzyme inhibitor screening
Multiple compounds coming from the Dundee HTS against dUTPase were tested for inhibition of T. cruzi and Leishmania dUTPase in a spectrophotometric proton release assay specifically designed for this class of enzymes (Granada) and in fluorimetric ligand binding assays in order to establish potency (York and Granada). A subset was also tested against human dUTPase for selectivity. Additionally, compounds coming from previous collections were tested. In total, 126 compounds have been tested in Granada. Unfortunately, most of the compounds were not confirmed as inhibitors and further development was not prioritised.
Thymidine kinase has been also explored as a novel drug target. Thus we have thoroughly characterised the Leishmania major recombinant enzyme (Granada) and established properties regarding substrate specificity and inhibition profiles. A collection of 47 thymidine analogues was tested against the enzyme. Due to lack of significant inhibition and antiprotozoal activity, this avenue was not further pursued.
Screening in vivo in animal models of infection
Testing in vivo was performed for specific compounds of the T. cruzi active chemical series that exhibited a good PK profile and metabolic stability. Very good efficacy and proof of concept was obtained in an acute mouse model.
WP4. ADME-Tox
In the T. cruzi structure activity optimisation phase, all new compounds were sent to partner 2 for determination of in vitro DMPK properties and selected compounds were tested for in vivo DMPK. The resulting information has allowed for the selection of two candidates for in vivo studies.
WP5. X-ray crystallography
The aim of partner 3 (York) was to determine the crystal structures of T. brucei dUTPase and co-crystals of T. cruzi, T. brucei and L. major dUTPases with ligands to establish the requirements for inhibitor binding and to provide information that would aid in improved inhibitor design. In addition the crystallization of other proteins identified as potential targets of compounds arising in the phenotypic screens was addressed.
Main achievements are:
1. dUTPases:
3D structures of all three dimeric dUTPases in complex with natural substrate analogues.
Proof of mechanism with attack on the beta-phosphate by structure, fluorescence and NMR.
First structure of an inhibitory ligand bound to the open conformation.
2. Other targets:
Structures of T. brucei adenosine kinase and leucyl aminopeptidase with ligands.
Preliminary expression of thymidylate kinase in baculovirus.
Dimeric dUTPases from L. major, T. brucei and T. cruzi.
L. major dUTPase.
No crystals were obtained for the open form. For the closed form, new complexes were crystallised with dUpNpp, the product dUMP and deoxyuridine (dU). The enzyme binds different numbers of metals dependent upon the identity of the substrate. The proposed nucleophilic water molecule is evident. Crystals of the dUMP and dU complexes were both in the completely closed conformation. Tryptophan fluorescence quenching showed that dUpNpp binds to the enzyme with a Kd of approximately 1.8 µM in the presence of Mg2+. PPi had little effect upon dUMP binding but had a major effect upon dU binding confirming that the presence of negative charge in the phosphate binding region is vital.
T. brucei dUTPase
The apo structure was similar to but more open than for T. cruzi. The closed form was very similar those of the other dimeric dUTPases with identical arrangement of the 5 substrate binding motifs and similar metal binding.
T. cruzi dUTPase
Previous open and closed structures had been obtained by partner 1 but good-diffracting were consistently harder to produce for this species. The structure of the dUpNpp complex was solved with different numbers of metals bound in the two active sites.
Ligands provided by partner 2 for crystallisation screening
Partner 2 first provided a series of compounds with uracil rings with carbon linkers of varying lengths to a phosphate group for co-crystallisation and assay of binding strength using tryptophan fluorescence quenching. The assay revealed very weak binding with Kd's of several hundred micromolar and no crystals were obtained. Partner 2 then provided a series of compounds identified as possible T. cruzi dUTPase inhibitors in high throughput screening, but these compounds had low solubility in water and extensive crystallisation trials gave no useful quality crystals. Partner 2 performed fragment screening using SPR to identify new compounds and the best hits were provided for screening. Crystals were obtained with the T. brucei enzyme but the structures were all of the open form and all but one lacked any bound ligand. The structure of T. brucei dUTPase with this ligand at 2.8 resolution revealed that it binds to the open form but the ligand density was weak. This hit occurred at the end of the contract and could be usefully pursued. Crystals of the T. cruzi enzyme with this compound bound were not sufficiently good to allow data collection.
Transition state analogue studies
Aluminium and magnesium fluorides were successfully used as planar transition state analogues to validate the mechanism. Tryptophan fluorescence quenching with the T. brucei enzyme clearly showed that AlF3 has a dramatic effect upon binding of dUMP but not of dU suggesting that the proposed attack by water on the phosphate was correct. The increase in affinity for dUMP in the presence of AlF3 and PO43- was followed up using MgF3- as a transition state analogue yielding similar results. Diffracting crystals of both the L. major and T. brucei dUTPases were obtained in the presence of dUMP, AlF3 and PO42- and provided structural confirmation of the mechanism. Structures of T. brucei dUTPase with dUMP and both [AlF3-OPO3] and MgF3- reveal the interactions between the protein and the planar transition state. 31P-NMR was used to probe the products of the dUTPase reaction and confirmed attack on the beta-phosphate highlighting this important difference between the trypanosomal and human proteins.
2. Other targets.
Thymidine (TK) and thymidylate (TMPK) kinases
With partner 1, these kinases involved in nucleotide metabolism from the three TriTryps were identified as potential drug targets. Extensive expression trials in E. coli focused on the TK set, but in spite of there being structures of homologues in the PDB, these proved impossible to express in stable soluble form. In attempts to improve the yield of L. major TK and to obtain soluble protein expression of TKs of T. brucei and T. cruzi, 30 different constructs were cloned and tested for expression with the baculovirus system in collaboration with the Oxford Protein Production Facility (OPPF). Some successfully expressed on a small scale and are being up-scaled to bigger culture volumes subsequent to the end of the contract.
Adenosine kinase (AK)
AK was proposed as a potential target at a consortium meeting as it activates the adenosine analogue cordycepin, highly active against T. brucei. The structure of TbAK was solved in the apo open form and closed form with adenosine and AMPPNP, which echoes the binding mode of the alternative substrate cordycepin. This gives direction for possible modifications of cordycepin that might improve binding affinity and resistance to deamination. Structural work is on-going. The structure was independently reported by another group (Kuettel, S. et al., 2011).
Leucyl aminopeptidases
At the consortium meeting in Brussels in month 37, the leucyl aminopeptidase (LAP) was proposed as a potential target for hit compounds by partner 7. Partner 3 decided to clone and express the LAPs of all three Tritryps with the aim of determining their crystal structures. The scope was broadened to include a second putative endopeptidase (AP2) and a puromycin aminopeptidase (PSA), from all three organisms. The LAP2 from all three TriTryps were successfully cloned and over-expressed and crystals obtained. The structure of T. brucei LAP was solved in its apo form to a resolution and bound to the inhibitor actinonin. There are no changes of conformation upon ligand binding. An independent structure of T. brucei LAP has recently been deposited in the PDB by the Structural Genomics Consortium (PDB 4efd). While other partners subsequently established that LAP was not the target of hit compounds, the LAP project is being actively pursued by partner 3 subsequent to the end of the contract.
P6. Intracellular and molecular mode of action studies:
1. Studies aimed at understanding the mode of action of series active against Trypanosoma cruzi have been accomplished. Several approaches have been used:
a) Capture of protein targets by proteomics using affinity chromatography. After using immobilised benznidazole as proof of concept for capture experiments, we received compounds from the consortium for immobilisation and the identification of the eluted proteins by mass spectrometry (Montevideo). In addition, an analysis of differentially expressed proteins upon drug exposure was studied by 2D electrophoresis and MALDI TOF/TOF.
b) Metabolomics. The metabolic profiling in response to selected compounds was studied in Trypanosoma cruzi (Montevideo).
c) Induction of resistance. Parasites resistant to the most active compounds were generated (Granada and Montevideo).
d) Selected compounds active against T. cruzi have been evaluate to establish static/cidal mode of killing (Granada).
e) Evaluation of inhibition of aminopeptidases as potential targets of specific hit compounds has been accomplished (Granada).
2. Studies on the mode of action of adenosine derivatives exhibiting activity against Trypanosoma brucei
Here a small number of adenosine derivatives made in the medicinal chemistry programme (Syngene) were tested for deamination by adenosine deaminase and phosphorylation by adenosine kinase (Granada). IC50 values and enzymatic data of metabolic conversion of these compounds suggested that a significant amount of work would be necessary for optimization. This avenue was not further pursued yet provided interesting information regarding structural requirements for deamination and phosphorylation of adenosine analogues.
Potential impact:
This project has directly addressed research in neglected infectious diseases and the development of new tools to control infections due to parasites of the Trypanosomatidae family. Other overall objectives covered include translational research in major infectious diseases and innovative therapeutic approaches and interventions.
The project was focused initially on all three diseases (Trypanosomiasis, Chagas disease, Leishmaniasis) and the main objective was to develop new interventions by means of the development of new drugs.
Due to the nature of these diseases, it is expected that the research outcome will be applicable and affordable for use in disease endemic countries and contribute to reduce incidence and impact of neglected tropical diseases. In addition, the project has contributed significantly to the strengthening of the capacity building of endemic countries for understanding and control of these diseases through networking and training scientists through the participation of partners from India and Uruguay.
The effort was clearly a goal oriented public-private partnership set up for drug discovery and aimed at translating research for human health: basic discoveries will be used in the development of practical benefits in order to ultimately improve the quality of life. Discovery and development of new chemotherapeutic tools through partnerships with the private sector is a major tool to develop new strategies of control.
The project included multidisciplinary discovery activities, the development and testing of new therapeutic interventions in vitro and in vivo, lead optimisation, investigation of drug metabolism, pharmacokinetics and preliminary safety of several new chemical entities before achieving the final candidate drug (CD) target profile anti-viral nucleoside analogues, some of reported toxicity.
Over the duration of the project, several young researchers have been involved in the implementation of research activities. New knowledge and techniques have been incorporated in the undergraduate and post-graduate training of the university partners on drug design and development and have been used by the CSIC and SWISS TPH in their training programmes.
Social and economical impact on American trypanosomiasis
Here we have rendered promising leads to be optimised for the treatment of Chagas disease. American trypanosomiasis (Chagas disease) occurs throughout Mexico and central and southern America, and continues to pose a serious threat to health in many countries of the region. Furthermore, due to recent significant population migrations from endemic countries towards developed countries, the threat of the disease is expanding to reach areas outside the traditional geographic boundaries; this is especially relevant in the context of transmission through blood transfusion. The overall prevalence of human Trypanosoma cruzi infection is estimated at 16 - 18 million cases. Approximately 120 million people, i.e. 25 % of the inhabitants of Latin America, are at risk of contracting the infection.
Two drugs, nifurtimox and benznidazole, are capable of curing at least 50 % of recent infections. These products are active in the acute and short-term (up to a few years) chronic phase. However, they have little or no activity in long-term chronic forms of the disease. In addition, both drugs have serious and frequent side-effects so their use is limited. The results presented here may contribute to the development of new control tools, such as safer drugs.
This international research cooperation project in the area of human health will help to stimulate global competitiveness, and is directly related to some of Europe's international commitments. Likewise cooperation will provide knowledge to partners about institutions in other parts of the world, help deliver European health policies, and create a favourable environment for other forms of international alliances. Finally this international research and development project has contributed to the production of global public goods and help to close the gap between different countries in the world.
A reduction of poverty by improving health conditions has for a long time been one of the commitments of the EU. However, investing in research of neglected diseases is not only an act of solidarity but also a global need due to their potential threat. Factors such as high mobility of people, centralised food handling and climate change contribute to an increased global vulnerability. This is manifested in outbreaks and rapid spread of infectious diseases across all continents and regardless of income levels. New epidemics have been observed in the last 20 years, and more emerging or re-emerging diseases (resulting from antimicrobial resistance) are expected. There is an urgent need for research in these illnesses.
Dissemination of project results
Protection of promising compounds by patent is required and this is currently under consideration by all partners. This will place some delay on publication. Specific contacts are being made with the private sector for further development of the promising lead chemical series active against Trypanosoma cruzi. Then results regarding promising compounds will be presented through publication in leading, international scientific journals and at international meetings on chemotherapy and drug design. Results concerning dUTPases are already published or under consideration for publication. Manuscripts concerning novel molecular drug targets are in preparation.
Exploitation of results
The project has given rise to potential discoveries in several areas. As already indicated it was a primary goal of the project to identify new leads and to give rise to potential therapeutic agents for treatment of diseases produced by the Trypanosomatidae. Several categories of compounds have been identified some of which would require substantial optimisation. Here we have only optimised the most promising considering the resources available. Since the promising lead compounds active against T. cruzi both in vitro and in vivo require further structural optimisation, future funding sources must be identified.
We have contributed significantly to the knowledge of the structure and function of dUTPases. While high affinity ligands with antiprotozoal activity have not been identified, high resolution structures are available to the scientific community for further efforts in inhibitor design. The information is also relevant to the understanding of the structure-function of the all-nucleotidohydrolase family of enzymes. Other advances include the characterisation of novel potential drug targets in the area of nucleotide metabolism and the application of state of the art methodology for understanding mode of action of selected compounds.
There is an urgent need for the development of new drugs for the treatment of Chagas disease. New, more effective drugs could have a significant impact on the lives of millions of people, particularly in South America. The compound series that has been developed has the potential for exciting further development. This will require further optimisation, particularly in terms of solubility and metabolic stability and evaluation in a whole variety of efficacy, pharmacokinetic and safety pharmacology assays. A patenting strategy is being agreed between the partners. We are actively investigating mechanisms to ensure that the compound series is developed further and are seeking funding sources.
Project website: http://www.ipb.csic.es/trypobase/index.html
Coordinator partner1
Dolores Gonzalez Pacanowska
Instituto de Parasitologia y Biomedicina 'Lopez-Neyra'
Consejo Superior de Investigaciones Cientificas
Avda. del Conocimiento s/n
Parque Tecnologico de Ciencias de la Salud
18100 - Armilla .Granada,
Spain
E-mail: dgonzalez@ipb.csic.es
Partner 2
Ian Gilbert
College of Life Sciences
University of Dundee
Sir James Black Centre
Dundee DD1 5EH,
UK
E-mail: I.H.Gilbert@dundee.ac.uk
Partner 3
Keith Wilson
Structure Biology Laboratory
Department of Chemistry
University of York
Heslington
York YO10 5YW,
UK
E-mail: keith@ysbl.york.ac.uk
Partner 4
Reto Brun
Swiss Tropical and Public Health Institute
Parasite Chemotherapy
Socinstr. 57
P.O. Box
CH-4002 Basel,
Switzerland
E-mail: reto.brun@unibas.ch
Partner 5
Nils Gunnar Johansson
Medivir AB
Lunastigen 7
S-141 44 Huddinge,
Sweden
E-mail: NilsGunnar.Johansson@medivir.com
Partner 6
Ashis Baran Mandal
Syngene Intl Ltd.
Biocon Park, Plot 2 and 3,
Bommasandra Industrial Estate - Phase-IV
Bommasandra-Jigani Link Road,
Bangalore 560 099,
India
E-mail: ashis.mandal@syngeneintl.com
Partner 7
Carlos Robello
Unidad de Biologia Molecular
Institut Pasteur de Montevideo
Mataojo 2020 CP11400
Montevideo, Uruguay
E-mail: robello@pasteur.edu.uy
