AID-CARE:
Artificial Intelligence-aided Design of CAnnabinoid REceptor subtype 2 (CB2R) Multitarget ligands:
A new strategy for the treatment of Inflammation
The efficacy of drugs developed for the treatment of inflammation-based diseases may be potentiated through AI systems supporting drug-discovery processes by the improvement of the bioavailability of novel candidates and by the reduction of their side effects.
Cannabinoid receptor subtype 2 (CB2R) belongs to the endocannabinoid system (ECS), a complex system involved in body homeostasis. Increased endocannabinoid tone, due to CB2R direct or indirect activation leads to beneficial anti-inflammatory effects that can be used as a new strategy to tackle inflammation-based pathologies such as cancer, neurodegeneration, and pain.
DESCRIPTION
Cannabinoid receptor subtype 2 (CB2R) is part of the endocannabinoid system. The increased endocannabinoid tone, due to CB2R direct activation or to an indirect pathway as the inhibition of endocannabinoid degradation, may exert beneficial anti-inflammatory effects useful in several inflammatory-based pathologies.
The clinical management of inflammation still requires an optimal balance between effective actions and associated safety risks. The efficacy of drugs developed for the treatment of deregulated inflammation may be potentiated through AI systems, supporting drug-discovery processes by improving the bioavailability of novel candidates and by reducing their side effects.
Recently, we developed DeLA-Drug and ALPACA, deep learning and machine learning algorithms for the design and optimization of ligands active at CB2R, as a well-established target involved in inflammation-based diseases such as cancer and neurodegeneration.
Exploiting an AI-aided design of Multi-Target Directed Ligands (MTDLs) capable of considering physiochemical characteristics (i.e. stability and feasibility to a green-process synthesis) and important translational parameters such as the pharmacodynamic and pharmacokinetic features of the new compounds, the AID-CARE project will lead to the generation of new libraries that will be tested in different pre-clinical-studies.
AID-CARE represents a proof-of-concept of a new AI-driven workflow for Multi-Target Directed Ligands targeting CBR2 that can be adapted to enable a fast and more advantageous design-make-test-analyze cycle for the investigation of combinations of different targets.
AIM
Several pieces of evidence nowadays identified in inflammatory states the onset of different pathological conditions such as neurodegeneration, cancer, neuropathic pain, and SARS-Cov-2 disease.
AID-CARE project aims to speed up the development of new Multi-Target Directed Ligands (MTDLs) to tackle deregulated inflammation through an innovative AI-based approach (DeLa-Drug/ALPACA) combined with molecular docking studies.
EXPECTED RESULTS
- AI-driven identification of the chemical features of finely tuned multi-modal CB2R selectivity combined with receptor activation capability.
- Definition of the chemical space to improve stability and overcome pharmacokinetic liabilities.
- Pharmacological validation of the lead compounds in several inflammation-based disorders such as neuropathic pain, neurodegenerative diseases, and cancer.
- Validation of the AI-driven design proto-type for further drug discovery processes related to cannabinoid research.
At the end of the AID-CARE project, a minimum of 5 lead compounds are expected to be identified and pharmacologically characterized.
