Periodic Reporting for period 1 - NeuRoPROBE (Probing (Orphan) Nuclear Receptors in Neurodegeneration)
Reporting period: 2022-05-01 to 2024-10-31
Neurodegeneration occurring in many pathologies such as Parkinson’s disease (PD), Alzheimer’s disease (AD), and other dementias is a severe health burden with major impact on patients, relatives and societies. It affects millions of people in the world, but treatment options are insufficient and mainly address symptoms underlining the urgent need for new effective therapies to halt and counteract neuronal loss. Due to several failures, research for a cure in neurodegenerative diseases has decreased, however, and new anti-neurodegenerative mechanisms are needed. Although neurodegenerative diseases arise from different underlying causes like misfolded amyloid proteins in AD and α-synuclein in PD, their pathomechanisms overlap suggesting that new interventions against neurodegeneration rather than the individual diseases can be developed. For example, neuroinflammation, dysfunctional energy supply and protein degradation processes as well as insufficient regeneration occur in multiple neurodegenerative pathologies. NeuRoPROBE aims to validate new molecular targets for novel therapeutic approaches in neurodegeneration. It focuses on the ligand-sensing transcription factors nuclear receptor-related 1 (Nurr1) and tailless homologue (TLX) which appear to have critical roles in neuronal health, but their pharmacological control and target validation have not been established. Nurr1is found in (mainly dopaminergic) neurons where it seems to have a protective and anti-neuroinflammatory role. Reduced Nurr1 expression/activity has been observed in neurodegenerative diseases including AD and PD patients and preclinical animal studies support great therapeutic potential. TLX is almost exclusively expressed in neural stem cells (NSCs) and appears to be responsible for maintaining NSCs proliferation to enable neurogenesis. Observations from patients and animal studies point to critical roles of TLX in learning, cognitive function and mental health. Therapeutic modulation of the transcription factors Nurr1 and TLX may thus evolve as a very attractive novel strategy to counteract neurodegeneration but pharmacological control of their activity and hence target validation for therapeutic approaches is hindered by the lack of potent and selective synthetic modulators. NeuRoPROBE aims to close this gap by developing high-quality chemical toolboxes and applying them in phenotypic models of neurodegeneration to study pharmacological control of Nurr1 and TLX.
NeuRoPROBE has focused in the first project phase on (i) identification of chemical starting matter for TLX and Nurr1 modulators and extensive structural optimization, (ii) development of deep-learning-based tools to aid ligand design and structural optimization, and (iii) development of phenotypic in vitro models of neuroinflammation/-degeneration to study effects of Nurr1 and TLX modulation.
Using systematic, structure-guided and AI-driven design approaches NeuRoPROBE has obtained four potent Nurr1 agonist scaffolds two of which already comprise chemical probe quality. Final optimization, development of structurally matched negative control compounds and comprehensive characterization are ongoing. Inverse agonist development to pharmacologically block Nurr1 activity has also progressed. Despite lower potency, a set of three inverse Nurr1 agonist scaffolds is available as early chemical tool and further optimization is ongoing. Similarly, systematic approaches and evolution-inspired combinatorial chemistry have enabled the development of two potent and selective TLX agonist scaffolds. Final optimization to chemical probe characteristics is in progress and inverse TLX agonist development is proceeding at an earlier stage.
Successful chemical tool development in NeuRoPROBE has benefitted from deep learning tools for molecular design. We have advanced the use of chemical language models (CLM) based on SMILES to be applied in very low data scenarios, for multi-target design and for structural optimization within a chemical series. Further refinement and extension of these models is in progress.
Application of chemical tools developed by NeuRoPROBE in phenotypic experiments has already provided further evidence for the great potential of Nurr1. The currently most advanced Nurr1 agonist has been applied in a midbrain organoid model based on induced pluripotent stem cells (iPSC). To mimic PD, organoids were generated from iPSC bearing a gain-of-function LRRK2 mutant (G2019S), which is among the most prevalent genetic causes of PD. Nurr1 agonist treatment rescued expression of the key dopamine synthesis gene tyrosine hydroxylase in mutant organoids to the levels of isogenic controls thus further supporting therapeutic potential of Nurr1 activation in PD. Moreover, preliminary phenotypic experiments with inverse Nurr1 agonists revealed that Nurr1 inhibition sensitizes neuronal cells against neurotoxic agents underlining the neuroprotective role of Nurr1.
Using systematic, structure-guided and AI-driven design approaches NeuRoPROBE has obtained four potent Nurr1 agonist scaffolds two of which already comprise chemical probe quality. Final optimization, development of structurally matched negative control compounds and comprehensive characterization are ongoing. Inverse agonist development to pharmacologically block Nurr1 activity has also progressed. Despite lower potency, a set of three inverse Nurr1 agonist scaffolds is available as early chemical tool and further optimization is ongoing. Similarly, systematic approaches and evolution-inspired combinatorial chemistry have enabled the development of two potent and selective TLX agonist scaffolds. Final optimization to chemical probe characteristics is in progress and inverse TLX agonist development is proceeding at an earlier stage.
Successful chemical tool development in NeuRoPROBE has benefitted from deep learning tools for molecular design. We have advanced the use of chemical language models (CLM) based on SMILES to be applied in very low data scenarios, for multi-target design and for structural optimization within a chemical series. Further refinement and extension of these models is in progress.
Application of chemical tools developed by NeuRoPROBE in phenotypic experiments has already provided further evidence for the great potential of Nurr1. The currently most advanced Nurr1 agonist has been applied in a midbrain organoid model based on induced pluripotent stem cells (iPSC). To mimic PD, organoids were generated from iPSC bearing a gain-of-function LRRK2 mutant (G2019S), which is among the most prevalent genetic causes of PD. Nurr1 agonist treatment rescued expression of the key dopamine synthesis gene tyrosine hydroxylase in mutant organoids to the levels of isogenic controls thus further supporting therapeutic potential of Nurr1 activation in PD. Moreover, preliminary phenotypic experiments with inverse Nurr1 agonists revealed that Nurr1 inhibition sensitizes neuronal cells against neurotoxic agents underlining the neuroprotective role of Nurr1.
The interim results of NeuRoPROBE indicate that pharmacological Nurr1 inhibition is detrimental for neuronal health while activation could restore levels of the important dopaminergic marker tyrosine hydroxylase. While substantial extension and further validation of these preliminary in vitro results will be needed, these observations provide further evidence that Nurr1 has a neuroprotective role and support therapeutic potential of its pharmacological activation in neurodegenerative diseases. NeuRoPROBE will provide the tools to broadly validate these findings and contribute to further in vitro confirmation. Additionally, Nurr1 ligand scaffolds developed by NeuRoPROBE may serve as very valuable starting points to develop Nurr1 agonist drugs.
NeuRoPROBE is refining and prospectively applying deep learning models for molecular design in the chemical tool development efforts with considerable success. These machine learning models can design innovative bioactive molecules. Substantial improvements achieved by NeuRoPROBE enable their application in low data scenarios for orphan targets and, most importantly, to accelerate structural optimization. These achievements will resonate broadly in medicinal chemistry and early drug discovery.
Beyond its focus on validating Nurr1 and TLX as targets for new interventions in neurodegenerative diseases, NeuRoPROBE is contributing to the development of chemogenomics libraries for the nuclear receptor family. A set of 69 chemogenomics compounds comprehensively covering the 19 nuclear receptors of the NR1 family has been developed and made available to the public as a highly valuable tool to explore therapeutic potential of these proteins. Further libraries covering the NR2, NR3 and NR4 families are being developed and the chemical tools for Nurr1 (NR4A2) and TLX (NR2E1) developed by NeuRoPROBE contribute significantly to these efforts. Application of the NR1 chemogenomics set in phenotypic experiments has revealed beneficial effects of several NR1 receptors on neuroinflammation and autophagy which are both related to neurodegeneration. These efforts thus contribute to NeuRoPROBE's overall objective of finding new therapeutic approaches to neurodegenerative diseases.
NeuRoPROBE is refining and prospectively applying deep learning models for molecular design in the chemical tool development efforts with considerable success. These machine learning models can design innovative bioactive molecules. Substantial improvements achieved by NeuRoPROBE enable their application in low data scenarios for orphan targets and, most importantly, to accelerate structural optimization. These achievements will resonate broadly in medicinal chemistry and early drug discovery.
Beyond its focus on validating Nurr1 and TLX as targets for new interventions in neurodegenerative diseases, NeuRoPROBE is contributing to the development of chemogenomics libraries for the nuclear receptor family. A set of 69 chemogenomics compounds comprehensively covering the 19 nuclear receptors of the NR1 family has been developed and made available to the public as a highly valuable tool to explore therapeutic potential of these proteins. Further libraries covering the NR2, NR3 and NR4 families are being developed and the chemical tools for Nurr1 (NR4A2) and TLX (NR2E1) developed by NeuRoPROBE contribute significantly to these efforts. Application of the NR1 chemogenomics set in phenotypic experiments has revealed beneficial effects of several NR1 receptors on neuroinflammation and autophagy which are both related to neurodegeneration. These efforts thus contribute to NeuRoPROBE's overall objective of finding new therapeutic approaches to neurodegenerative diseases.