Solid tumors constitute the majority of cancer cases worldwide and account for most cancer-related deaths (Bray et al., 2024; Siegel et al., 2025). Unlike many blood cancers, solid tumors are protected by a hostile tumor microenvironment (TME): a dense extracellular matrix with immunosuppressive signals and cellular heterogeneity that block immune infiltration and complicate targeted therapies (Binnewies et al., 2018). These physical and biochemical barriers account for the limited clinical translation of immunotherapies that demonstrate efficacy in preclinical models of solid tumors, but fail in vivo.

A particularly challenging aspect of tumor targeting is that many key tumor cues are soluble ligands and cytokines that shape the local niche. Current receptor-based engineered cell therapies that attempt to sense such soluble signals risk systemic activation. By detecting a ligand in circulation or healthy tissue, they can produce on-target/off-tumor effects, excessive cytokine release and dangerous immune overreaction (Piraner et al., 2025). Still, targeting these soluble ligands is crucial for detecting and treating tumors that lack distinctive surface antigens. The result is a detrimental trade-off between sensitivity and safety.

---

Figure 1: Schematic overview of the PHOENICS cell. The cell (grey outline) is engineered to contain two receptor types - a stimulatory one (purple) carrying our SynKinase and an inhibitory one (brown) carrying our SynPhosphatase. When the cell encounters a positive ligand, the stimulatory receptor is activated, recruiting and phosphorylating the synthetic substrate through a proximity-based reaction. The phosphorylated SynSubstrate interacts with downstream circuit components and drives protein secretion at a fully post-translational level. Conversely, exposure to a negative ligand activates the inhibitory receptor, which brings the same SynSubstrate near SynPhosphatase, promoting its dephosphorylation and repressing downstream secretion.

PHOENICS, Phosphorylation Engineered Network for Inducible Controlled Secretion, is designed to resolve this dilemma! We engineered a synthetic immune cell that integrates phosphorylation-based logic gates with modular receptor architectures. This enables us to evaluate both a tumor's soluble signature and healthy-tissue-cues, computing precise AND/NOT logic at therapeutically relevant speed. The secretion of engineered payloads only occurs when the tumor signal is present and the inhibitory checkpoint is absent, which prevents off-site triggering. This powerful platform is modular, fast, and tuned to operate within the constraints of the TME. With PHOENICS, we embark on a journey, where we bring foundational advance into oncology. Our synthetic toolbox is designed to outsmart cancer rather than outgun it, enabling safer, more precise and adaptable cancer therapies that were previously out of reach.

Integrated Design

By combining extensive wet-lab validation with SPARC, our digital twin, we optimized receptor retargeting, binder generation, and circuit dynamics. This integration was crucial to experimentally demonstrate sensing of biologically relevant ligands such as VEGF and TNF-α, as well as secretion of IL-12, a powerful immune-stimulatory cytokine. These results highlight PHOENICS not only as a versatile signalling platform, but also as a step toward building synthetic immune cells as programmable systems that unite sensing, computation, and effector functions within a single cellular chassis.

With PHOENICS, we set out to create a modular plug-and-play toolbox for programmable cell therapies. Each unit, sensor, processor, and effector, is fully synthetic and orthogonal, allowing our circuits to function independently of native pathways. In contrast to CAR-based systems that rely on immune components, this ensures compatibility with various cellular chassis and avoids unintended interference with normal physiology.

Because our system can sense multiple ligands simultaneously and compute opposing kinase and phosphatase signals on a shared substrate, it is especially suited for hard-to-treat solid tumors with complex and dynamic TMEs. These are conditions where tumor and healthy tissues are difficult to distinguish and where therapies must adapt in real-time to rapidly changing signals. We envision PHOENICS as a platform for both off-the-shelf and precision-tailored therapies. Its orthogonal design opens possibilities for deployment in non-immune-cell chassis, while modularity by design allows patient-specific customization through receptor and payload combinations that match the tumor's molecular profile. To enable this level of precision, we needed a way to explore countless receptor-ligand pairings and circuit architectures before testing them experimentally. This is where SPARC comes into play.

---

Figure 2: Graphical Abtract of the SPARC framework. Our model incorporates a de novo binder generation software, BindCraft, followed by binder characterization, mathematical modelling of the PHOENICS system and an optimization scheme for cellular architectures.

We developed SPARC, the Simulator for Phosphorylation And Receptor Characterization, to circumvent expensive in vitro testing of all possible circuit architectures. By integrating mathematical modeling with protein binder design, this digital twin accelerates the design and optimization process of PHOENICS cells, and unlocks flexible target selection for diverse cancer applications.
As part of its integrated pipeline, SPARC leverages BindCraft, a cutting-edge deep learning tool for the design of de novo protein binders (Pacesa et al., 2025). Binder designs are ranked based on their binding properties, characterized by SPARC using molecular dynamics (MD) simulations (Govind Kumar et al., 2023). Experimental wet lab validation confirmed the accuracy of our predictions and demonstrated the models capabilities to generate new binders for efficient retargeting of PHOENICS cells.

To ensure that our cell is specifically activated by ligand profiles in the TME, SPARC quantifies the phosphorylation output through an integrated mathematical model of receptor dimerization and the protein circuit (Lu & Wang, 2017; Yang et al., 2025). Simulated dose-response curves closely align with our wet lab data, demonstrating SPARC's predictive accuracy. Building on this, we created a digital database that maps the behavior space of all possible component combinations, enabling rapid prototyping of optimal circuit architectures for various solid tumors.
SPARC was created in several iterations, each highly benefiting from the input of experts and stakeholders in the field. With our model we can efficiently explore and optimize PHOENICS cell designs, providing a valuable tool for the scientific community. SPARC improves the precision of our system and ensures that it can be rapidly adapted to different TMEs, offering a safe and scalable solution for the future of personalized cancer treatments.

Building on the insights gained from our computational and wet lab efforts, we successfully established PHOENICS as a versatile signaling toolbox, not only for our project, but for the entire iGEM community. Our goal was to make every element of PHOENICS accessible, standardized, and reusable, providing a foundation that future teams can easily build upon.
To achieve this, we assembled a comprehensive part collection covering all functional layers of the system: sensors, processing units, effectors, and adaptors. Each component was individually characterized to determine activity, specificity, and limitations, and then validated in combination to ensure robust performance within integrated circuits. Computational insights from SPARC complemented this process, defining binding properties, kinetics, and overall circuit behavior.

sensory repertoire includes synthetic GPCRs and MESA receptors capable of detecting small molecules, soluble proteins, or surface-bound antigens. These receptors are directly fused to kinases or phosphatases, acting as modular input units that translate extracellular cues into phosphorylation events.
At the core of the collection are our processing units - synthetic kinases, phosphatases, substrates, and phospho-binders. Together, they form the computational engine of PHOENICS and represent a standardized set of reusable signaling modules for synthetic phosphorylation circuits. Their benchmarking ensures broad applicability, empowering future teams to design and adapt circuits for their own biological questions.
To translate phosphorylation into action, we developed both transcriptional outputs and a rapid secretion platform, where proteins are pre-synthesized, retained in the ER, and released only upon activation.

the core, we showcased the extensibility of PHOENICS with additional modules: an in silico designed binder against GDF15, a temperature-responsive MELT input, and an ATP-sensitive GPCR variant (P2Y11), demonstrating how the framework can easily extend to physical and metabolic signals.
All components are fully documented and open-access, forming not only the foundation of PHOENICS but also a shared resource for the synthetic biology community a coherent, standardized, and ready-to-use library for building next-generation phosphorylation-based signaling systems. By openly sharing our designs and data, we aim to empower others to create, adapt, and expand upon PHOENICS. This spirit of transparency and collaboration is what drives our educational mission: fostering and inspiring the community to turn innovation into real-world impact.

We are building a community of innovators. Our mission is to empower today's scientists and spark the imagination of tomorrow's pioneers, bridging the translational gap through education, collaboration, and dialogue.

Heidelberg is home to phenomenal fundamental research. Yet, a crucial focus is often missing in today's scientists' education: how results actually translate into treatments that effectively improve patients’ quality of life. Our program exists to bridge this translational gap. We build upon Heidelberg’s innovation ecosystem to provide the support networks and educational backdrop for young scientists to think beyond the laboratory. Taking a guiding role, we aim to equip the next generation of innovators with the knowledge and network to turn discovery into impact.

We began by asking a simple question: what expertise does it truly take to bring research into the clinic? Our conversations with leaders in the field revealed that most students are unaware of the translational process. We learned about the critical importance of topics like proactive communication, patient involvement, and industry collaboration, as well as how to navigate cultural barriers, ethical considerations, and funding landscapes.
Therefore, we brought precisely these topics into focus. We organized a panel discussion featuring key decision-makers and pioneers, including Prof. Dr. Otmar Wiestler, President of the Helmholtz Association, and Prof. Dr. Eva Winkler, Deputy Chair of the German Ethics Council. The goal was to discuss these issues at eye level, inspire students to broaden their perspective, and feel empowered to act and innovate on their own ideas.

While a local innovation ecosystem already exists, the support network is highly fragmented and difficult to navigate. To give bright minds better access to the regional infrastructure, we collaborated with startup incubators, innovation clusters, and venture capitalists to create a comprehensive roadmap on the innovation landscape. It serves as a practical guide, helping young scientists connect to the right partners at each stage of development.

foster this next generation of scientists, we expanded the Summer School that was first established by last year's iGEM Team together with the German Cancer Research Center. We gave high school students the opportunity to step into a lab and perform their first experiments. To cement the course as a lasting and accessible learning tool , we developed a complementary interactive learning platform featuring illustrated lab protocols, calculators, and module quizzes. Integrating feedback from students and teachers, we created instructional videos that showcase fundamental lab techniques. This learning portal is now ready to be used by iGEM teams and educators worldwide, and is designed to be expanded with further courses as future iGEM teams build on this foundation.

In parallel to our Summer School, we organized a Friday Lecture series and various workshops guiding students in designing their own synthetic cell, while also promoting proactive science communication to the broad public with an interactive workshop in the heart of Heidelberg’s Old Town. Our motto throughout this journey has been simple: Educate others and Inspire to Act. And that is what we will keep doing, because the future of translation starts with young scientists daring to learn, to question, and to build.