Periodic Reporting for period 1 - MystIMEC (Endothelial immunosuppressive mystery genes for alternative immunotherapy: artificial intelligence-driven target discovery and lipid nanoparticle/RNA-based target validation)
Reporting period: 2023-03-01 to 2025-08-31
MAIN GOAL
This project aims to identify and functionally characterise previously unstudied, immunomodulatory “mystery genes” expressed by endothelial cells (ECs) in disease, focusing on cancer and vascular inflammation (eg. diabetes, organ transplantation). Despite sequencing the entire human genome 21 years ago, it is estimated that 6,000-10,000 genes still lack any known function. These mystery genes represent a vast untapped resource for understanding biology and discovering novel drug targets. The project leverages in-house AI-driven discovery tools, a unique high-throughput lipid nanoparticle (LNP)/siRNA-based in vivo screening, and advanced humanized ex vivo models to systematically explore hidden biological interactions. The goal is to uncover novel druggable targets in ECs - especially plasma membrane and secreted proteins with immunoregulatory roles - that have been overlooked due to limited functional annotation.
SOCIETAL IMPORTANCE
ECs play a central yet underappreciated role in modulating immune responses in health and disease. By targeting immunoregulatory pathways in ECs, the project opens entirely new avenues for precision therapies in oncology and inflammatory disease-conditions that collectively affect hundreds of millions globally. Importantly, the methodologies developed not only enhance the precision of target validation but also significantly reduce costs and accelerate timelines in drug development, enhancing access and innovation.
OVERALL OBJECTIVES
• Discover & prioritise novel EC-expressed immunomodulatory genes using AI/ML tools
• Functionally validate top candidates through rapid in vivo screening with EC-selective LNPs
• Develop therapeutic modalities including monoclonal antibodies and nanobodies against validated targets
This project aims to identify and functionally characterise previously unstudied, immunomodulatory “mystery genes” expressed by endothelial cells (ECs) in disease, focusing on cancer and vascular inflammation (eg. diabetes, organ transplantation). Despite sequencing the entire human genome 21 years ago, it is estimated that 6,000-10,000 genes still lack any known function. These mystery genes represent a vast untapped resource for understanding biology and discovering novel drug targets. The project leverages in-house AI-driven discovery tools, a unique high-throughput lipid nanoparticle (LNP)/siRNA-based in vivo screening, and advanced humanized ex vivo models to systematically explore hidden biological interactions. The goal is to uncover novel druggable targets in ECs - especially plasma membrane and secreted proteins with immunoregulatory roles - that have been overlooked due to limited functional annotation.
SOCIETAL IMPORTANCE
ECs play a central yet underappreciated role in modulating immune responses in health and disease. By targeting immunoregulatory pathways in ECs, the project opens entirely new avenues for precision therapies in oncology and inflammatory disease-conditions that collectively affect hundreds of millions globally. Importantly, the methodologies developed not only enhance the precision of target validation but also significantly reduce costs and accelerate timelines in drug development, enhancing access and innovation.
OVERALL OBJECTIVES
• Discover & prioritise novel EC-expressed immunomodulatory genes using AI/ML tools
• Functionally validate top candidates through rapid in vivo screening with EC-selective LNPs
• Develop therapeutic modalities including monoclonal antibodies and nanobodies against validated targets
LONG-TERM VISION
The project aims to revolutionize our understanding of ECs as active immune regulators (IMECs) and to exploit this knowledge for the development of innovative vascular-targeted therapies. By uncovering the immunomodulatory roles of mystery genes expressed by ECs in disease, the project lays the foundation for new therapeutic strategies in cancer, diabetes, and inflammatory disorders - diseases that affect hundreds of millions globally.
WORK PERFORMED
We developed new AI/ML tools to identify and prioritize immunosuppressive and immunostimulatory genes (e.g. AI4TargetDicovery-v1.0 scMystYdentifier-v2.0 GenePrior-v2.0 DL4CPath). To discover novel EC phenotypes and standardize the nomenclature of EC phenotypes, a human mega single-EC atlas (SCVASCAT) was created by integrating published sc- and snRNA-seq datasets across healthy & disease conditions (covering most organs and including a number of diseases). Using stringent criteria, we identified core EC-subtype markers conserved in disease and highlighted tissue-specific signatures. We conducted a comparative analysis of tumor- and peri-tumor-derived ECs from matched samples, and observed consistent changes across tumors, highlighting a pro-angiogenic and immune-modulatory phenotype (Veys K, Cell Reports, in revision).
Using these, we prioritized ~60 high-potential mystery genes with predicted immunoregulatory functions. To validate these targets, we designed a novel endothelial-specific lipid nanoparticle (LNP) platform enabling in vivo gene silencing. By incorporating siRNAs targeting the mystery gene, we can very cheaply (<€100/mouse) and quickly (<24 hours) silence the expression of our mystery genes in ECs in vivo – compared to €10K- €50K and 1-5 years to create a conditional knockout mouse. More than half of the prioritized targets showed therapeutic benefit in preclinical models (>50% lung tumor growth inhibition).
After selecting targets with sufficient therapeutic effect, we use our in-house developed ‘gene prioritization’ tool to quickly gather multiple molecular features (3D structure, expression pattern, disease association, subcellular localization, etc.) of our targets.
MAIN ACHIEVEMENTS
• Developed and use of novel AI/ML tools for rapid immune target discovery.
• Created a new EC-selective LNP-siRNA delivery system for fast in vivo validation (in lung diseases).
• Discovered and validated ~30 novel immune-modulatory EC targets.
• Built technically demanding but highly translational ex vivo human disease models (e.g. perfused tumor-on-chip systems).
• Initiated generation of therapeutic nanobodies & monoclonal antibodies for top candidates.
NEXT STEPS
Continue high-throughput in vivo validation of remaining mystery genes and expand therapeutic screening using nanobodies & monoclonal antibodies. Concurrently, scaling up disease models and exploring patent and trademark protection to facilitate clinical translation and engage industrial partners.
The project aims to revolutionize our understanding of ECs as active immune regulators (IMECs) and to exploit this knowledge for the development of innovative vascular-targeted therapies. By uncovering the immunomodulatory roles of mystery genes expressed by ECs in disease, the project lays the foundation for new therapeutic strategies in cancer, diabetes, and inflammatory disorders - diseases that affect hundreds of millions globally.
WORK PERFORMED
We developed new AI/ML tools to identify and prioritize immunosuppressive and immunostimulatory genes (e.g. AI4TargetDicovery-v1.0 scMystYdentifier-v2.0 GenePrior-v2.0 DL4CPath). To discover novel EC phenotypes and standardize the nomenclature of EC phenotypes, a human mega single-EC atlas (SCVASCAT) was created by integrating published sc- and snRNA-seq datasets across healthy & disease conditions (covering most organs and including a number of diseases). Using stringent criteria, we identified core EC-subtype markers conserved in disease and highlighted tissue-specific signatures. We conducted a comparative analysis of tumor- and peri-tumor-derived ECs from matched samples, and observed consistent changes across tumors, highlighting a pro-angiogenic and immune-modulatory phenotype (Veys K, Cell Reports, in revision).
Using these, we prioritized ~60 high-potential mystery genes with predicted immunoregulatory functions. To validate these targets, we designed a novel endothelial-specific lipid nanoparticle (LNP) platform enabling in vivo gene silencing. By incorporating siRNAs targeting the mystery gene, we can very cheaply (<€100/mouse) and quickly (<24 hours) silence the expression of our mystery genes in ECs in vivo – compared to €10K- €50K and 1-5 years to create a conditional knockout mouse. More than half of the prioritized targets showed therapeutic benefit in preclinical models (>50% lung tumor growth inhibition).
After selecting targets with sufficient therapeutic effect, we use our in-house developed ‘gene prioritization’ tool to quickly gather multiple molecular features (3D structure, expression pattern, disease association, subcellular localization, etc.) of our targets.
MAIN ACHIEVEMENTS
• Developed and use of novel AI/ML tools for rapid immune target discovery.
• Created a new EC-selective LNP-siRNA delivery system for fast in vivo validation (in lung diseases).
• Discovered and validated ~30 novel immune-modulatory EC targets.
• Built technically demanding but highly translational ex vivo human disease models (e.g. perfused tumor-on-chip systems).
• Initiated generation of therapeutic nanobodies & monoclonal antibodies for top candidates.
NEXT STEPS
Continue high-throughput in vivo validation of remaining mystery genes and expand therapeutic screening using nanobodies & monoclonal antibodies. Concurrently, scaling up disease models and exploring patent and trademark protection to facilitate clinical translation and engage industrial partners.
Our project has delivered several key breakthroughs that significantly advance current knowledge and innovation capacity in vascular biology and immunotherapy:
SCIENTIFIC ADVANCES
Uncovering a previously unappreciated immunoregulatory role of endothelial cells (IMECs) across multiple diseases. Using large-scale AI-guided screening and novel EC-specific in vivo validation tools, we identified and functionally characterized >30 mystery genes with immune-modulating properties. This shifts the paradigm in vascular biology from a passive barrier model to an active immune regulation framework, opening new therapeutic avenues.
TECHNOLOGICAL INNOVATION
We developed an integrated, scalable platform that combines:
• AI/ML-based gene prioritization,
• EC-specific LNP delivery for siRNA-based gene silencing in vivo,
• Humanized, perfused 3D-disease models for translational validation.
These tools reduce costs, accelerate novel discoveries, and improve the precision of preclinical validation compared to traditional methods (e.g. conditional knockout mouse generation).
COMMERCIALISATION & IPR SUPPORT
We are exploring opportunities to file patents.
• Methodological approach: filing a patent on our holistic approach ‘from AI to pharmacological tool compound testing’
• LNP formulation: investigating the impact of various LNP formulations on biodistribution and gene silencing efficacy in in vivo animal models. Through iterative optimization, we have developed an in-house LNP formulation by modifying an established formulation. Ongoing work focuses on testing different lipid compositions to identify the optimal formulation for selective/ greater EC uptake
• In-house AI tool commercialization: developing our in-house AI tools into commercial service platforms
FUTURE KEY INVESTMENTS
To fully realize the societal and economic impact of this work, the following enabling steps will be essential:
• Investment in therapeutic development. We consider to apply for EIC Transition funding based on our awarded ERC PoC project
• Support for patenting, trademarking, and early-stage IP valorization (in collaboration with VIB TTO)
• Industrial partnerships for drug development, manufacturing, and clinical testing
• Access to regulatory guidance, mainly for siRNA/LNP and antibody-based therapies
• Establishment of standards for novel human disease models to support broader validation
RESULTS (CURRENT STAGE):
• ~60 mystery immune targets prioritized; ~30 validated in vivo
• New EC-selective LNP delivery platform established
• First therapeutic nanobodies in development; monoclonal antibody campaigns initiated
• AI tools validated and available for broader use
SCIENTIFIC ADVANCES
Uncovering a previously unappreciated immunoregulatory role of endothelial cells (IMECs) across multiple diseases. Using large-scale AI-guided screening and novel EC-specific in vivo validation tools, we identified and functionally characterized >30 mystery genes with immune-modulating properties. This shifts the paradigm in vascular biology from a passive barrier model to an active immune regulation framework, opening new therapeutic avenues.
TECHNOLOGICAL INNOVATION
We developed an integrated, scalable platform that combines:
• AI/ML-based gene prioritization,
• EC-specific LNP delivery for siRNA-based gene silencing in vivo,
• Humanized, perfused 3D-disease models for translational validation.
These tools reduce costs, accelerate novel discoveries, and improve the precision of preclinical validation compared to traditional methods (e.g. conditional knockout mouse generation).
COMMERCIALISATION & IPR SUPPORT
We are exploring opportunities to file patents.
• Methodological approach: filing a patent on our holistic approach ‘from AI to pharmacological tool compound testing’
• LNP formulation: investigating the impact of various LNP formulations on biodistribution and gene silencing efficacy in in vivo animal models. Through iterative optimization, we have developed an in-house LNP formulation by modifying an established formulation. Ongoing work focuses on testing different lipid compositions to identify the optimal formulation for selective/ greater EC uptake
• In-house AI tool commercialization: developing our in-house AI tools into commercial service platforms
FUTURE KEY INVESTMENTS
To fully realize the societal and economic impact of this work, the following enabling steps will be essential:
• Investment in therapeutic development. We consider to apply for EIC Transition funding based on our awarded ERC PoC project
• Support for patenting, trademarking, and early-stage IP valorization (in collaboration with VIB TTO)
• Industrial partnerships for drug development, manufacturing, and clinical testing
• Access to regulatory guidance, mainly for siRNA/LNP and antibody-based therapies
• Establishment of standards for novel human disease models to support broader validation
RESULTS (CURRENT STAGE):
• ~60 mystery immune targets prioritized; ~30 validated in vivo
• New EC-selective LNP delivery platform established
• First therapeutic nanobodies in development; monoclonal antibody campaigns initiated
• AI tools validated and available for broader use