Introduction and Background of the Project

Introduction

High-throughput screening against the malaria parasite Plasmodium falciparum has identified a wealth of new chemical matter with potent antimalarial activity. However, the majority of hits derive from cell-killing phenotypic screens, and there is no knowledge of their mode of action (MoA). Although techniques exist to subsequently identify MoA, these are time- and labour-intensive, with no guarantee to yield identified targets.

This project aims to develop a combined high-throughput growth inhibition assay and cell painting platform for antimalarial drug discovery to classify the MoA of newly confirmed active compounds derived from high throughput screenings. Based on high-content imaging of Plasmodium-infected red blood cells, the platform will allow the clustering of compounds based on their biological effect on the parasite, and aid decision-making on whether to bring a compound series forward in the pipeline early on according to the novelty of its MoA while simultaneously providing information on its effect on parasite growth inhibition.

Project objective

The project ultimately aims to deliver a new high-throughput and information-rich platform for informing and classifying antimalarial modes of action, and highlighting novel compound phenotypes. This proposal seeks to leverage advances in cellular imaging and machine learning-led pattern recognition to develop a robust, reproducible method to deliver information on a compound’s biological impact (whether its MoA or pathway is novel or known) in synchrony with the confirmation of growth inhibition and thus allow clustering on both chemistry and biology, potentially saving months in the context of Hit Generation.

Project design

The project relies on high-content imaging and subsequent analysis of drug-treated Plasmodium falciparum parasites. The initial assay development phase will optimise methodologies for staining, fixation and imaging of parasite-infected red blood cells, including both healthy parasites and those treated with a pilot set of compounds with defined MoA. This will allow preliminary development of artificial intelligence (AI) models to classify parasite morphology across the 48 hour lifecycle, as well as the impact of drug-treatment. Once treatment and imaging parameters have been optimised, data collection will be performed with an expanded set of compounds covering a diverse range of MoA, in order to refine and validate the development of AI models for pattern recognition. AI models will ultimately be packaged into a cloud-based, user-friendly application so that images generated by researchers can be analysed without specialist AI knowledge.

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

Malaria caused approximately 263 million new cases and 597,000 deaths in 2023,¹ predominantly in children under five years of age and pregnant women. Furthermore, the emergence of resistance to artemisinin and partner drugs, in Southeast Asia and more recently in East Africa, poses a threat to malaria control and elimination. Thus, there is a pressing need for the development and approval of new therapies, ideally with novel MoA that are not subject to existing resistance mechanisms present in the field. This project aims to develop a compound profiling platform to rapidly identify the MoA of antimalarial compounds. This will enable a focus on chemical series with the highest priority for drug discovery, and potential triage of those with unwanted MoA, using state-of-the-art parasitology, imaging, and AI analysis tools to reduce the time and cost of deconvolving the MoA of phenotypic hits from screening.

What sort of innovation are you bringing in your project?

Cell Painting approaches, which use a combination of fluorescent stains and reporters to illuminate cellular structures and organelles, together with unbiased quantification of these cellular features, have been demonstrated to facilitate drug discovery and MoA studies in a variety of disease models. The project will bring together a multi-disciplinary team with expertise in drug discovery research, malaria parasitology and high-content imaging, and AI image analysis.

Role and Responsibility of Each Partner

The project is a multi-disciplinary collaboration between Medicines for Malaria Venture (MMV), the University of Dundee, and LPIXEL. MMV will lead on the curation of compounds with diverse MoA for conducting the assay, University of Dundee will perform the parasite culture optimisation, compound exposure, and high-content imaging and analysis, and LPIXEL will develop AI models for MoA deconvolution and implement them on a cloud-based platform.

Others (including references if necessary)

1.World Health Organisation (2024) World Malaria Report 2024 Available at: https://www.who.int/teams/global-malaria-programme/reports/world-malaria-report-2024 (Accessed 19 February 2025)

* Graph, Chart etc. can be attached to the document and if you would like to attach photos, please submit those photos with the Development Partner Profile. Please note that high-resolution photos (JPEG) are necessary. 

Red blood cells infected with Plasmodium falciparum. Parasites express green fluorescent protein (GFP), nuclei are stained with Hoechst, and the red blood cell membrane labelled with wheat-germ agglutinin coupled to AlexaFluor633.