Introduction and Background of the Project

Introduction

There is an urgent need for new treatments for Chagas disease. Existing medications lack effectiveness against chronic infection, require long regimens, and have several adverse effects. Given their integral roles in trypanosome signaling and low homology with human counterparts, phosphodiesterases (PDEs) have been posited as drug targets for Chagas disease. Given the paucity of identified targets and critical need for new mechanism-of-action drugs, these enzymes merit definitive evaluation followed by efficient identification and development of inhibitors.

Project objective

This project aims to validate PDEs as drug targets for Chagas disease and identify selective inhibitors using a computationally-enhanced screening cascade.

Project design

An accelerated drug development path will be sought by focusing on repurposing opportunities that can be rapidly progressed to clinical trials, complemented by screening for new chemical matter from Eisai’s compound library. Candidate inhibitors identified in machine-learning based virtual screens will be profiled experimentally and promising compounds advanced to animal studies.  

How can your partnership (project) address global health challenges?

There is an urgent need for development of new treatments for Chagas disease, which affects approximately 7 million people. The two currently approved medications, benznidazole and nifurtimox, were discovered over fifty years ago. These medications require long treatment courses (60-90 days) and cause adverse effects which often result in treatment discontinuation. Furthermore, they lack effectiveness against chronic infection, which is responsible for the majority of mortality and morbidity in Chagas disease. There is a paucity of new drugs under development, exacerbated by the recent failure of CYP51 inhibitors posaconazole and fosravuconazole in clinical trials.

This project seeks to address this pressing need by efficiently developing new compounds with the potential to achieve the target-product-profile for a new Chagas disease drug: a short, well-tolerated treatment that is effective against chronic infection.

What sort of innovation are you bringing in your project?

The project brings together three key elements to tackle the urgent need for new Chagas disease treatments:

First, by conducting conclusive validation of a potential drug target family. The lack of validated drug targets for Chagas disease means thorough exploration of putative targets is vital. Several lines of evidence point to phosphodiesterases (PDEs) as essential proteins in T. cruzi, therefore this project will definitively evaluate three PDE enzymes as drug targets using CRISPR/Cas9 technology.

Second, by harnessing advances in the use of artificial intelligence for drug discovery. Machine-learning models with high predictive power can rapidly screen large compound collections in silico to prioritize compounds for resource-intensive experimental follow-up. Both Eisai and UNLP will bring to bear advanced machine-learning tools in this project, in order to efficiently search chemical space for selective PDE inhibitors cost-effective manner.

Third, by employing a repurposing approach to expedite drug development. Repurposing of approved or shelved compounds with known human safety profiles is a proven strategy to reduce time and risk in drug development. The screening efforts in this project will focus on repurposing libraries to identify compounds with the potential for rapid progression to Phase II proof-of-concept studies. As a practical complement, Eisai’s chemical library will also be screened for new trypanocidal scaffolds.

The innovative approach of the collaboration team leverages the synergy of artificial intelligence, experimental screening capabilities and drug development expertise between a pharmaceutical company and academic investigators in an endemic country.

Role and Responsibility of Each Partner

Eisai will be the project lead and responsible for overall management. Eisai will develop and apply its machine learning capabilities to T. cruzi PDEs, and perform screening against the Eisai compound library. Eisai will further provide PK and safety expertise for evaluation of repurposing candidates, and lead medicinal chemistry activities.

At UNLP, Prof. Talevi’s group will conduct ligand-based machine-learning model development and screening of repurposing libraries, and perform molecular docking simulations. In addition, UNLP will oversee the target validation studies for PDE enzymes using CRISPR/Cas9, the biochemical and whole-cell activity assays, and the in vivo studies in mouse models of Chagas disease.  

UNLP will also contribute their experience as investigators in a country where Chagas disease is endemic.

Final Report

1. Project objective

This project aimed to validate phosphodiesterases (PDEs) as drug targets for Chagas disease and to identify selective inhibitors using a computationally-enhanced screening cascade.

2. Project design

The primary approach for confirmation of PDEs as drug targets for Chagas disease was through genetic knockout. This was conducted by generating Cas9-expressing parasite lines followed by transfection with sgRNAs targeting three PDE genes. An accelerated drug development path was sought by focusing on repurposing opportunities to allow rapid progression to clinical trials, complemented by screening for new chemical matter from Eisai’s compound library. Candidate inhibitors were identified in machine-learning based virtual screens and subsequently profiled experimentally. Promising compounds advanced to animal studies to assess in vivo efficacy.

3. Results, lessons learned

First, validation of a potential drug target family was conducted. There is a lack of validated drug targets for Chagas disease and therefore, thorough exploration of putative targets is vital. Several lines of evidence pointed to phosphodiesterases (PDEs) as essential proteins in T. cruzi, therefore this project genetically evaluated three PDE enzymes as drug targets using CRISPR/Cas9 technology.
This project also harnessed advances in the use of artificial intelligence for drug discovery. Machine-learning models with high predictive power can rapidly screen large compound collections in silico to prioritize compounds for resource-intensive experimental follow-up. Both Eisai and UNLP utilized machine-learning models in this project, and successfully and efficiently searched chemical space for selective PDE inhibitors in a cost-effective manner.
Finally, a repurposing approach was employed to expedite drug development. Repurposing of approved or shelved compounds with known human safety profiles is a proven strategy to reduce time and risk in drug development. The screening efforts in this project focused on repurposing libraries to identify compounds with the potential for rapid progression to Phase II proof-of-concept studies. As a practical complement, Eisai’s chemical library was also screened for new trypanocidal scaffolds. Promising compounds that emerged from these screens were then subjected to an experimental screening cascade for validation of their potential use as drugs in the treatment of Chagas disease.
The innovative approach of the collaboration team leveraged the synergy of artificial intelligence, experimental screening capabilities and drug development expertise between a pharmaceutical company and academic investigators in an endemic country.