Final Report Summary - TRANSSIADRUG (Chagas' disease - Inhibitor discovery and optimization for Trypanosoma cruzi trans-sialidase)
Background: the TRANSSIADRUG project aimed at the identification and optimisation of inhibitors for the Trypanosoma cruzi trans-sialidase. T. cruzi is the causative agent of Chagas disease, a devastating chronic infection which afflicts millions in Latin America. Trans-sialidases are surface-anchored and secreted enzymes with multiple functions in the pathogen's virulence. The enzyme's activity blunts the host's immune response to the pathogen and is important for entry of the parasite in the host's cells. It therefore has potential as a drug target in this disease, for which only two relatively toxic drugs are currently available.
As a first step to validate whether blocking trans-sialidase activity is a viable therapeutic strategy, we set out to identify lead compounds for this target.
Strategy: to enable high-throughput screening of compound libraries to the trans-sialidase, we first set up extensive experiments to design a robust screening assay for this rather complex trans-glycosylation reaction. Previously used assays used radioactive substrate, which we wanted to avoid. Therefore, we evaluated an assay used at lower throughput earlier in our lab, making use of trans-sialidase acceptor substrate (lactose) labeled with a fluorescent group (8-aminoacridone; AMAC). Upon reaction, a negatively charged, sialylated product is formed (sialyl-lactose-AMAC) which can be separated from non-sialylated substrate through ion exchange chromatography. The formation of the fluorescent product can then be quantitated in the eluate of the column. While this assay has merit, when we evaluated its characteristics in a 96-well plate format, we found the dynamic range to be too limited to be useful in a HTP screening setting.
We then changed the strategy to a separation of substrate and product using C18 reversed phase solid phase extraction, which is feasible in 96-well plates. While more reproducible than the ion-exchange based assay, it was still not good enough to reliably detect partial inhibition of the enzyme (which is needed to pick up lead compounds). Moreover, a more in-depth study of the kinetics of this trans-glycosylation assay demonstrated that the formation of trans-glycosylation product was under kinetic control for a range of conditions: whereas trans-sialylation occurs initially in the reaction, later on the trans-glycosylation product can be hydrolysed because the relative turnover number of the enzyme for trans-sialylation and sialidase activity is roughly 4:1. While it is possible in reference settings to optimise the reaction conditions such that this confounding effect is very minimal, it is problematic in robust screening for inhibitors.
In a continuation of the project, we have subsequently resorted to a simpler assay system, i.e. screening for inhibition of the enzyme's sialidase activity. This assay was subsequently optimised for HTP screening and a library of 10,000 compounds was screened, resulting in the identification of several inhibitors with order-of-magnitude lower Ki than the inhibitors known sofar for the trans-sialidase. Moreover, we have adopted the autodock vina docking engine to computationally screen close to 1,000,000 drug-like compounds for their predicted binding to the trans-sialidase catalytic site. The best compounds of this endeavour were validated in our enzymatic assays and the very best have a Ki which is two orders of magnitude lower than the best inhibitors known for trans-sialidase at the start of the project (very low micromolar Ki). We are preparing the result for publication.
The lead compounds which we have now available will be tested in further studies for lack of toxicity and subsequently for their effect on disease course in animal models of Chagas disease.
As a first step to validate whether blocking trans-sialidase activity is a viable therapeutic strategy, we set out to identify lead compounds for this target.
Strategy: to enable high-throughput screening of compound libraries to the trans-sialidase, we first set up extensive experiments to design a robust screening assay for this rather complex trans-glycosylation reaction. Previously used assays used radioactive substrate, which we wanted to avoid. Therefore, we evaluated an assay used at lower throughput earlier in our lab, making use of trans-sialidase acceptor substrate (lactose) labeled with a fluorescent group (8-aminoacridone; AMAC). Upon reaction, a negatively charged, sialylated product is formed (sialyl-lactose-AMAC) which can be separated from non-sialylated substrate through ion exchange chromatography. The formation of the fluorescent product can then be quantitated in the eluate of the column. While this assay has merit, when we evaluated its characteristics in a 96-well plate format, we found the dynamic range to be too limited to be useful in a HTP screening setting.
We then changed the strategy to a separation of substrate and product using C18 reversed phase solid phase extraction, which is feasible in 96-well plates. While more reproducible than the ion-exchange based assay, it was still not good enough to reliably detect partial inhibition of the enzyme (which is needed to pick up lead compounds). Moreover, a more in-depth study of the kinetics of this trans-glycosylation assay demonstrated that the formation of trans-glycosylation product was under kinetic control for a range of conditions: whereas trans-sialylation occurs initially in the reaction, later on the trans-glycosylation product can be hydrolysed because the relative turnover number of the enzyme for trans-sialylation and sialidase activity is roughly 4:1. While it is possible in reference settings to optimise the reaction conditions such that this confounding effect is very minimal, it is problematic in robust screening for inhibitors.
In a continuation of the project, we have subsequently resorted to a simpler assay system, i.e. screening for inhibition of the enzyme's sialidase activity. This assay was subsequently optimised for HTP screening and a library of 10,000 compounds was screened, resulting in the identification of several inhibitors with order-of-magnitude lower Ki than the inhibitors known sofar for the trans-sialidase. Moreover, we have adopted the autodock vina docking engine to computationally screen close to 1,000,000 drug-like compounds for their predicted binding to the trans-sialidase catalytic site. The best compounds of this endeavour were validated in our enzymatic assays and the very best have a Ki which is two orders of magnitude lower than the best inhibitors known for trans-sialidase at the start of the project (very low micromolar Ki). We are preparing the result for publication.
The lead compounds which we have now available will be tested in further studies for lack of toxicity and subsequently for their effect on disease course in animal models of Chagas disease.