Synthetic Immunology

𝗖𝗮𝗻 𝗺𝗮𝗰𝗵𝗶𝗻𝗲 𝗹𝗲𝗮𝗿𝗻𝗶𝗻𝗴 𝗵𝗲𝗹𝗽 𝗱𝗶𝘀𝗰𝗼𝘃𝗲𝗿 𝗹𝗶𝗳𝗲-𝘀𝗮𝘃𝗶𝗻𝗴 𝗮𝗻𝘁𝗶𝗯𝗼𝗱𝗶𝗲𝘀 𝗳𝗮𝘀𝘁𝗲𝗿? 𝗬𝗘𝗦. 𝗔𝗻𝗱 𝗵𝗲𝗿𝗲’𝘀 𝗵𝗼𝘄. Imagine a library, not of books, but of 1 billion synthetic antibodies. Now imagine that library being machine learning-ready, stripped of problematic sequences, barcoded for traceability, and optimised to mimic real human immune responses. That’s what Deepash Kothiwal et al. just did. And it’s a breakthrough. 👨🔬 They built a minimalist Fab yeast display library targeting the CDRH3 region—the “hotspot” for antigen recognition—while eliminating polyreactivity-causing motifs. 🔬 Then, they launched a parallel antibody discovery campaign against 10 tough targets: from PD-L1 and TIGIT to ROBO2 and Syncytin-2. The results? 🚀 ✅ Hundreds of antibodies with therapeutic-grade properties ✅ Binding confirmed by SPR, cell display, and IHC ✅ 486 antibodies rigorously tested ✅ Over 68,000 Fab sequences publicly released ✅ ML rescued potent antibodies missed during physical selection Oh, and they trained a model that accurately predicted cross-reactive antibodies from early selection data, paving the way for hybrid in silico + in vitro antibody discovery. This isn’t just academic. It’s a blueprint for how AI and structural biology can accelerate next-gen therapeutics, reliably, reproducibly, and at scale. 🔗 Full preprint: https://lnkd.in/du-zaCmx #AI #MachineLearning #AntibodyDiscovery #SyntheticBiology #StructuralBiology #ProteinEngineering #BiotechInnovation #OpenScience

Predicting TCR antigen specificity at proteome‑scale with synthetic immune cells and machine learning. https://lnkd.in/epX6uKyK I’m pleased to share our group's collaborative effort with Engimmune in this new pre‑print: Kai-Lin Hong, Beichen Gao, Lucas Stalder, Roy Ehling, Marton Horvath, Sarah Wehrle, Juliette Forster, Valentin Junet, Jakub Kucharczyk, PhD, Sébastien Lalevée, Marie-Charlotte Diringer, Qinmei Yang, Rodrigo Vazquez-Lombardi TCR therapies offer powerful precision — but also carry real safety risks due to unexpected peptide-HLA cross-reactivity. In this work, we combine: - Synthetic antigen-presenting cells, genomically engineered to stably present pHLA libraries - Fluorescent cytokine reporters for functional readouts in both APCs and T cells - Supervised ML models, trained on deep screening data, to predict specificity across the human proteome We applied this platform to therapeutic TCR candidates, including a clinically approved TCR drug, and uncovered previously unknown off-targets — some with substantial sequence dissimilarity to the intended target. This approach may provide a scalable, functionally grounded path to support safer TCR discovery and screening.

Researchers at MIT have developed a new way to reprogram a person’s own immune cells so they can recognize and attack a wide range of cancers. Traditional immunotherapies train T cells to target one specific marker found on certain tumors. This approach limits effectiveness because many cancers lack a single universal marker or evolve to hide it. The MIT team’s engineered cells use a modular system that lets them detect multiple cancer signals at once. Instead of chasing one protein, these “universal” immune cells can spot a broader profile of danger signs on tumor cells, allowing them to seek out and destroy many distinct cancer types regardless of their origin. The engineered cells are built from a normal class of white blood cells called macrophages, which are already tuned to engulf and digest abnormal cells. By adding synthetic receptors and control circuits inside these macrophages, the scientists gave them an enhanced ability to identify and break down cancer cells that normally escape detection. In laboratory and animal studies, the reprogrammed cells significantly slowed tumor growth across a diverse set of cancer models, showing promise as a flexible immunotherapy platform. While still early, this work points to a future where one form of cellular therapy could be effective against many cancers, offering hope beyond targeted treatments tied to specific tumor markers. Research Paper 📄 DOI: 10.1038/s41467-025-63863-8

Synthetic biologists have created a new approach to attacking tumors. They have engineered tumor-colonizing bacteria (probiotics) to produce synthetic targets in tumors that direct CAR-T cells to destroy newly highlighted cancer cells. The researchers have essentially created a universal CAR-T cell that attacks a universal antigen, by programming tumor-seeking bacteria to paint solid tumors with a synthetic marker that CAR-T cells can recognize. They expect that, with further improvements, this platform will enable treatment for any type of solid tumor without the need to identify a specific tumor antigen, thus avoiding the need to generate a customized CAR-T cell product for each type of cancer and each patient. The researchers continue to refine their work and hope to launch clinical trials to fully evaluate the platform's safety and efficacy in human patients.

💡 Synthetic T Cells Redefining Precision Medicine 🔬 This new groundbreaking study introduces synthetic suppressor T cells that deliver immune protection precisely where it’s needed. 🔬 What’s New? Researchers engineered CD4+ T cells with synthetic Notch (synNotch) circuits that activate upon encountering specific antigens. These cells: ✔️ Release anti-inflammatory agents such as IL-10, TGFβ1, and PD-L1. ✔️ Act as cytokine sinks, consuming IL-2 via CD25 to halt inflammatory responses. 📊 Highlights and Results: 1️⃣ Local Immune Suppression: ◼️ The best-performing suppressor cells combined TGFβ1 + CD25, mimicking natural regulatory T cells. ◼️ These cells blocked cytotoxic CD4+ and CD8+ T cells in both in vitro models and animal studies. 2️⃣ Cancer Immunotherapy: Synthetic suppressor T cells were used to build a "NOT gate" logic circuit. ◼️ Outcome: Healthy, cross-reactive tissues were shielded from CAR T cell attacks, while tumors remained targeted. ◼️ Key data: Tumor volumes reduced significantly in targeted areas without affecting surrounding tissues. 3️⃣ Organ Transplants: Protected pancreatic islet transplants in a Type 1 diabetes model, maintaining endocrine function without systemic immunosuppression. ◼️ Bioluminescence tracking: Transplants treated with suppressor T cells remained functional for 6+ weeks. ◼️ Insulin secretion: Functional beta cells continued to respond to glucose stimulation. 4️⃣ Advanced Cytokine Control: ◼️ Synthetic circuits reduced local IL-2 levels, impairing T cell proliferation and degranulation. ◼️ Enhanced production of TGFβ1 created a positive feedback loop, increasing suppressor T cell efficacy over time. 💡 Broader Applications: These engineered cells could transform treatments across various domains: ✔️ Autoimmune diseases: Protect tissues like the brain in multiple sclerosis or pancreatic islets in diabetes. ✔️ Organ transplants: Prevent rejection while preserving graft function. ✔️ Cancer therapy: Expand CAR T cell applications by minimizing off-target toxicities in solid tumors. 🔧 Technological Innovation: ◼️ Modular design enables customization for diverse diseases. ◼️ Combines immune suppression with regenerative potential, e.g., delivering repair factors alongside suppression. ◼️ Leverages paracrine signaling to protect nearby cells even with heterogeneous antigen expression. 🚀 Future Directions: Synthetic biology offers a powerful toolkit to dissect and reprogram immune responses. These findings pave the way for precision immune modulation, balancing immune attack and suppression with spatial and temporal precision. 🔗 Data-Driven Insights: Tumor models showed selective immune suppression, preserving tumor-killing by CAR T cells. Functional grafts maintained glucose-responsive insulin production, verified by C-peptide levels in vivo. #SyntheticBiology #CellTherapy #Immunotherapy #AutoimmuneDisease #CART

𝐖𝐡𝐚𝐭 𝐢𝐟 𝐲𝐨𝐮 𝐜𝐨𝐮𝐥𝐝 𝐟𝐢𝐧𝐞-𝐭𝐮𝐧𝐞 𝐓 𝐜𝐞𝐥𝐥𝐬 𝐥𝐢𝐤𝐞 𝐚 𝐬𝐲𝐦𝐩𝐡𝐨𝐧𝐲, 𝐧𝐨𝐭 𝐚 𝐬𝐨𝐥𝐨? 𝐒𝐜𝐢𝐞𝐧𝐭𝐢𝐬𝐭𝐬 𝐡𝐚𝐯𝐞 𝐝𝐞𝐯𝐞𝐥𝐨𝐩𝐞𝐝 𝐚 𝐧𝐞𝐰 𝐰𝐚𝐲 𝐭𝐨 𝐜𝐫𝐞𝐚𝐭𝐞 𝐡𝐢𝐠𝐡𝐥𝐲 𝐜𝐮𝐬𝐭𝐨𝐦𝐢𝐬𝐞𝐝 𝐓 𝐜𝐞𝐥𝐥𝐬 that can target disease more precisely. Instead of ending up with a messy mix of partly edited cells, this method helps researchers keep only the ones with all the right edits. No extra labels or tags needed! 𝐓𝐡𝐞 𝐬𝐞𝐜𝐫𝐞𝐭 𝐥𝐢𝐞𝐬 𝐢𝐧 𝐚 𝐜𝐥𝐞𝐯𝐞𝐫 𝐭𝐨𝐨𝐥 𝐜𝐚𝐥𝐥𝐞𝐝 𝐒𝐄𝐄𝐃-𝐒𝐞𝐥𝐞𝐜𝐭𝐢𝐨𝐧. It uses synthetic "disruptors" to link the integration of a new gene with the removal of a native protein. That means you can tell which cells have been fully edited just by checking what’s missing on their surface. 𝐓𝐡𝐞 𝐭𝐞𝐚𝐦 𝐭𝐞𝐬𝐭𝐞𝐝 𝐭𝐡𝐢𝐬 𝐚𝐩𝐩𝐫𝐨𝐚𝐜𝐡 𝐨𝐧 𝐭𝐡𝐫𝐞𝐞 𝐜𝐫𝐢𝐭𝐢𝐜𝐚𝐥 𝐭𝐚𝐫𝐠𝐞𝐭𝐬 𝐫𝐞𝐥𝐚𝐭𝐞𝐝 𝐭𝐨 𝐓 𝐜𝐞𝐥𝐥 𝐟𝐮𝐧𝐜𝐭𝐢𝐨𝐧: 🟠 Specificity (what the T cell recognises) 🟠 Co-receptors (how it responds) 🟠 MHC expression (how it communicates with other immune cells) 𝐓𝐡𝐞 𝐫𝐞𝐬𝐮𝐥𝐭𝐬? ✅ 98% purity for individual edits ✅ 90% purity even when doing six changes at once (three knock-ins and three knockouts) ✅ Fully compatible with clinical-grade manufacturing workflows 𝐓𝐡𝐢𝐬 𝐦𝐢𝐠𝐡𝐭 𝐬𝐞𝐞𝐦 𝐥𝐢𝐤𝐞 𝐚 𝐭𝐞𝐜𝐡𝐧𝐢𝐜𝐚𝐥 𝐭𝐰𝐞𝐚𝐤, but it opens the door to faster, cleaner, and more precise production of next-generation cell therapies. Imagine the impact for cancer, autoimmunity, and even infectious disease treatments. 💡 𝐖𝐡𝐚𝐭 𝐬𝐭𝐚𝐧𝐝𝐬 𝐨𝐮𝐭 𝐭𝐨 𝐦𝐞 is how this tackles one of the most frustrating bottlenecks in cell therapy manufacturing, which is purity without complexity. Exactly what the field needs to move from promise to product. 👣 Follow me for more insights on cell engineering, biotech breakthroughs, and the science behind tomorrow’s therapies. https://lnkd.in/eqymad4p

"Studies using antigen-presenting systems at the single-cell and ensemble levels can provide complementary insights into T-cell signaling and activation. Although crucial for advancing basic immunology and immunotherapy, there is a notable absence of synthetic material toolkits that examine T cells at both levels, and especially those capable of single-molecule-level manipulation. Here we devise a biomimetic antigen-presenting system (bAPS) for single-cell stimulation and ensemble modulation of T-cell recognition. Our bAPS uses hexapod heterostructures composed of a submicrometer cubic hematite core (α-Fe2O3) and nanostructured silica branches with diverse surface modifications. At single-molecule resolution, we show T-cell activation by a single agonist peptide-loaded major histocompatibility complex; distinct T-cell receptor (TCR) responses to structurally similar peptides that differ by only one amino acid; and the superior antigen recognition sensitivity of TCRs compared with that of chimeric antigen receptors (CARs). We also demonstrate how the magnetic field-induced rotation of hexapods amplifies the immune responses in suspended T and CAR-T cells. In addition, we establish our bAPS as a precise and scalable method for identifying stimulatory antigen-specific TCRs at the single-cell level. Thus, our multimodal bAPS represents a unique biointerface tool for investigating T-cell recognition, signaling and function." Interesting paper detailing a single cell method for investigating T-cell receptor (TCR) and chimeric antigen receptors (CARs) antigen recognition through the use of covalently modified nanostructures bearing single agonist peptide-loaded major histocompatibility (MHC) complex. The investigators showed their method allows for high-throughput identification of stimulatory CD8+ antigen-specific TCRs at the single-cell level. Paper and research by @Xiaodan Huang and larger team at The University of Chicago Pritzker School of Medicine https://lnkd.in/e956Yy88

Design and Engineering of SynNotch CARs SynNotch CAR, or synthetic Notch chimeric antigen receptor, represents a next-generation approach to CAR-T cell therapy designed to enhance selectivity and control of T cell activation. While traditional CARs activate T cells immediately upon binding to the target antigen, SynNotch CAR uses a modular two-step activation process to target tumor cells with high specificity while minimizing off-target effects. This innovative system exploits the Notch signaling pathway to enable T cells to respond only when precise tumor antigen sequences are detected. The SynNotch CAR system consists of two key components: the synNotch receptor and the traditional CAR. The synNotch receptor recognizes an antigen on the surface of the tumor cell and upon binding releases a transcription factor into the T cell nucleus, which activates the expression of a second component (usually a CAR or another therapeutic gene). This two-step activation ensures that T cells only activate the CAR when encountering specific combinations of antigens on cancer cells, resulting in a highly targeted response. This design offers significant advantages in terms of safety and effectiveness. For example, in solid tumors where normal cells share some antigens with tumor cells, SynNotch CAR can prevent T cells from attacking healthy tissue. By requiring the presence of two or more antigens, the SynNotch CAR system greatly reduces the risk of on-target, off-tumor toxicity. Furthermore, it provides real-time control of T cell activation, as the system is activated only in specific tumor microenvironments. SynNotch CAR has also shown promise in preclinical models, particularly in treating heterogeneous tumors that require a high degree of specificity. This innovative approach provides an additional layer of control, allowing responses to be tailored to the unique characteristics of individual tumors and represents a major advance in precision cancer immunotherapy. References [1] Fei Teng et al., Signal Transduction and Targeted Therapy 2024 (https://lnkd.in/eSDZXeCi) [2] Ming-Ru Lu et al., Nature Review Cancer 2019 (DOI: 10.1038/s41568-019-0121-0) #SynNotchCAR #CARTCells #CancerImmunotherapy #PrecisionMedicine #SyntheticBiology #ImmunoOncology #CancerResearch #TargetedTherapies #MedicalInnovation #TumorTargeting

Lab-Grown Human Immune System Model Uncovers Weakened Response in Cancer Patients. Georgia Tech. November 12, 2024 Excerpt: The miniature models of human immune organoids mimic the environment where immune cells learn to recognize and attack harmful invaders and respond to vaccines. The organoids are important new tools for studying immune function in cancer, their use is likely to accelerate vaccine development, predict disease treatment response for patients, and accelerate clinical trials. “Our synthetic hydrogels create a breakthrough environment for human immune organoids, allowing us to model antibody production more precisely, and for longer duration,” said Ankur Singh, Professor in George W. Woodruff School of Mechanical Engineering and professor in Wallace H. Coulter Department of Biomedical Engineering at Georgia Tech and Emory. Note: “For the first time, we can recreate and sustain complex immunological processes in a synthetic gel, using blood, effectively tracking B cell responses,” he added. “This is a gamechanger for understanding and treating immune vulnerabilities in patients with lymphoma who have undergone cancer treatment — and hopefully other disorders too.” Led by Singh, the team created lab-grown immune systems that mimic human tonsils and lymph node tissue to study immune responses. The research findings, published in the journal Nature Materials (link enclosed), included investigators from Emory University, Children’s Hospital Atlanta, and Vanderbilt University. Designing a Tiny Immune System Model: The researchers were inspired to address a critical issue in biomedical science: the poor success rate of translating preclinical findings from animal models into effective clinical outcomes, in immunity, infection, and vaccine responses. “Animal models are valuable for many types of research, and often fail to accurately mirror human immune biology, disease mechanisms, and treatment responses,” said Monica (Zhe) Zhong, a Bioengineering Ph.D. student and the paper’s first author. “We designed a new model that replicates the unique complexity of human immune biology across molecular, cellular, tissue, and system levels.” The team used synthetic hydrogels to recreate a microenvironment where B cells from human blood and tonsils mature and produce antibodies. Utilizing organoids for individual patients helps predict response to infection. The models enable researchers to control and test immune responses under various conditions. The team discovered not all tissue sources are the same, and tonsil cells struggled with longevity. Game-Changing Technology: This research is of interest to infectious disease researchers, cancer researchers, immunologists, and healthcare professionals dedicated to improving patient outcomes. An individual researcher can make hundreds of organoids in a single sitting. The model’s capability to target different populations increases its usability for vaccine and therapeutic testing.