Description

Background Information

---

Chronic diseases often require lifelong pharmacological treatment, with consistent medication intake being critical to therapeutic success . Conditions such as psychiatric disorders and neurodegenerative diseases, including Parkinson's and Alzheimer's, frequently require strict commitment to complicated drug regimens. However, maintaining consistent medication intake, also known as medication adherence or medication loyalty, remains a major challenge for many patients .

The consequences of poor adherence are significant. According to the Medication Adherence Expertise Center of the Northern Netherlands (MAECON), an estimated 200,000 deaths annually in Europe alone are linked to insufficient medication adherence. Pfizer estimates that poor medication adherence leads to up to €125 billion in avoidable healthcare costs annually, placing a significant financial burden on healthcare systems . This issue is not confined to Europe. The World Health Organization (WHO) identifies medication non-adherence as a global health problem of alarming proportions. According to its report, adherence to long-term therapies: evidence for action (2003), the organization states that in developed countries, only 50% of patients with chronic diseases adhere to prescribed treatment regimens, a number that is even lower in many developing countries.

Parkinson's Disease (PD) is an example of a disease with suboptimal medication adherence. Especially in later stages of the disease, patients are expected to take medication several times a day, with complex and varying dosage regimens. This complexity contributes to non-compliance . Additional obstacles patients undergo are depressive symptoms, memory loss, low energy, and negative expectations of treatment. This can lead to a self-reinforcing cycle in which non-adherence to treatment guidelines worsens the underlying symptoms that contribute to it.

Our Solution

---

To address these challenges, we propose a self-sustained drug delivery system that eliminates the need for regular patient-administered medication. Specifically, we aim to engineer the commensal gut bacterium Pseudomonas alcaligenes to autonomously produce and secrete therapeutic compounds within the gastrointestinal tract of Danio rerio (zebrafish), serving as a proof-of-concept for future clinical applications!

Why Zebrafish?

The zebrafish is a well-established vertebrate model for studying human disease due to its genetic similarity to humans - approximately 70% of human genes have a zebrafish ortholog . Other advantages include its optical transparency, rapid development, and the ability for live imaging, which allows the study of in vivo cellular dynamics over the timespan of embryonic development . Zebrafish can be monitored from the start of life, since they develop ex utero from a fertilized egg. They are also particularly suitable for studying intestinal physiology, microbial colonization, and host-microbe interactions, given that several gut functions and immune genes are conserved between zebrafish and mammals . Furthermore, their use aligns with the principles of the 3Rs (Replacement, Reduction, and Refinement) in animal research, which focuses on replacing complex animals like higher vertebrates. The embryonic stages of zebrafish are considered a replacement and refinement . Until the fish become capable of independent feeding (five days post-fertilization - 5 dpf), they are considered embryos .

Lab Experiments

---

Engineered Function

Plasmid Creation

Our project utilizes a modular collection of newly designed genetic parts, assembled to enable fluorescence, secretion, uptake, metabolic production, and biosafety control. At the core of this collection, there are eight functional gene blocks encoding fluorescent proteins (fuGFP, mCherry), secretion and uptake tags (HlyA, TAT/R9), and a metabolic pathway for L-DOPA biosynthesis. These geneblocks were synthesized with Golden Gate-compatible overhangs and cloned into the broad-host-range pJUMP24-1A backbone, allowing efficient and flexible construction of higher-order assemblies. By combining different gene blocks, we generated plasmids that encode either fluorescent reporters, therapeutic production modules, or integrated kill switch systems.

Kill Switches

In order to maximize biosafety, we intend to implement two complementary kill switches. The first is a complete kill switch, removing all of the introduced modified bacteria. The addition of L-arabinose, a harmless sugar, triggers this kill switch. This provides full external control, enabling patients or clinicians to remove the engineered bacteria from the gut at any time. The second is a quorum-sensing (density-based) kill switch, which relies on LuxR/AHL signaling. When bacterial density becomes too high, the accumulated AHL dimerizes with LuxR, activating expression via the pLux promoter. Both the kill switches utilize the MazF microbial toxin, which poses no threat to humans. MazF cleaves single-stranded RNA, which inhibits protein synthesis and leads to cell death. The second kill switch prevents uncontrolled overgrowth and maintains a safe balance. Together, these systems combine external clearance with internal population control for patient safety and agency.

Fluorescent Proteins in the Gut

After the design, implementation, and validation of our DNA constructs in both E. coli and P. alcaligenes, we aim to assess their functionality in vivo. To achieve this, we utilize the zebrafish larvae model organism (before 5 dpf), which we feed with the transformed bacteria before imaging. More specifically, we investigate zebrafish gut colonization by assessing signal from the fluorescent protein construct-containing bacteria. More specifically, we will look into whether the engineer microbes can stably reside in the gut, followed by studying their localization and uptake dynamics within the gastrointestinal tract of the larvae.

Transwell System

To bridge the gap between our zebrafish proof-of-concept and potential application in humans, we tested whether our engineered bacteria could deliver therapeutic molecules across the intestinal barrier. For this purpose, we used the Caco-2 Transwell-based intestinal model developed by Floor et al. (2025)

In this system, Caco-2 cells are cultured under air–liquid interface (ALI) conditions with vasoactive intestinal peptide (VIP) added to the basolateral compartment. This combined ALI-VIP treatment induces the formation of a polarized epithelial monolayer covered by a robust mucus layer, closely mimicking the human intestinal lining. The model not only supports epithelial barrier formation and small-molecule permeability but also enables studies of commensal and pathogenic bacterial interactions with the mucus.

Using this setup, we aim to assess colonization and interaction of our engineered bacteria within the mucus layer. We will follow up with studies on the production and secretion of therapeutic outputs (L-DOPA or fluorescent proteins). Lastly, we will look into the transport across the epithelial barrier towards the basolateral compartment, which we plan to quantify using HPLC and other analytical chemistry methods.

Bacterial Production

Before moving into animal models for the therapeutic aspect of GutFeeling, we will first confirm that the engineered P. alcaligenes bacteria are also capable of producing L-DOPA in vitro. By feeding them simple precursors, usually found within the native environment of complex organisms (i.e. pyruvate, ammonia or pyrocatechol), our modified bacteria should be able to produce L-DOPA. This has been established in multiple bacterial strains . We will test L-DOPA synthesis under culture conditions and quantify production using HPLC and other analytical chemistry methods. Establishing this in vitro production is an important first step as it provides evidence that our system can generate our chosen compound before continuing to more complex zebrafish studies.

Modeling

---

To understand and predict the behavior of our engineered system under different conditions, our modeling team developed computational models that simulate bacterial growth, L-DOPA production, and metabolic fluxes. By comparing growth models, they evaluate how our plasmids influence growth and competitiveness. These models allow the team to extract important parameters, including growth rate and lag time, giving us insights into our constructs.

Beyond growth, the team extended established flux balance models of Pseudomonas putida to incorporate L-DOPA production, examining how resources such as pyruvate limit L-DOPA synthesis. Additionally, we modeled the biological effects of the kill switch on the metabolics of the cell. With this, we can make a prediction for the growth rate of our bacteria when the kill switch is induced. Combining this with the growth model will then predict the results we expect to see in the lab. These computational methods provide insights into the safety of our system, and computationally validate that our system is both safe and effective.

Lastly, our team also greatly expanded the flux balance model by adding gene product reactions for all known genes in Pseudomonas putida. Now we can incorporate differences in gene expression, gene networks and more in the model. For example, we can now test knock-outs of genes, to screen for ones that would increase yield or boost growth.

Human Practices

---

Synthetic biology is a controversial topic, and the actual implementation of live bacteriotherapy is years away from realization. Our Human Practices team investigates how our project could transition into society, using frameworks from transition science and innovation science. By applying the multi-level perspective (MLP) framework, they analyze how live bacteriotherapies for Parkinson's disease could fit into the broader healthcare landscape. This means examining not just the science, but also the market, industry, cultural, regulatory, and patient dimensions that shape whether such a treatment can succeed.

The team mapped the stakeholder network, whose needs drive innovation but also face barriers such as medication burden, financial access, and regulation. They used surveys to capture patient perspectives, finding that many see constant, self-sustaining treatment as a way to improve quality of life.

On the industry side, they assessed the pharmaceutical market, existing treatments, and potential challenges like reproducibility standards and hurdles for implementation of GMOs. They also reviewed cultural attitudes toward GMOs and living medicines, next to safety and containment requirements.

By combining these insights, the Human Practices team highlights both the drivers and barriers that will influence whether living biopharmaceuticals can become a sustainable treatment option. This structured approach ensures our project is not only scientifically important, but also socially and ethically responsible.

Education

---

Synthetic biology often raises concerns, especially when applied to therapeutics. From our surveys and interviews, we learned that hesitancy around ingesting engineered bacteria was less about outright opposition and more about unfamiliarity. People wanted to understand the science and the safeguards before feeling comfortable with it. Our education team addressed this by designing activities that make synthetic biology accessible, interactive, and age-appropriate.

With young children, we introduced synthetic biology through storytelling and play, letting them invent imaginative "fantasy bacteria" that reflected real-world concepts like gene regulation and biosafety. With teenagers, we aligned lessons with international curricula and developed The Plasmid Game, where students designed their own plasmids to solve global challenges, encouraging critical thinking, collaboration, and debate. At the university level, we organized seminars and discussions across disciplines, showing how synthetic biology connects to fields like geosciences, law, and economics.

Beyond classrooms, we explored creative ways to connect science with society. At Dutch Design Week, we collaborated with artists to translate our project into visual, emotional experiences. We also designed a Dungeons & Dragons one-shot adventure that introduces players to plasmid design and biosafety within a fantasy narrative.

By engaging audiences of all ages, our education efforts turned passive reception into active participation. More than teaching concepts, our goal was to build trust and curiosity, laying a foundation for the public acceptance of living medicines like GutFeeling.

Impact

---

Gutfeeling is pioneering the road towards live bacteriotherapies. Our proof-of-concept, the modification of P. alcaligenes to produce the Parkinson's medicine L-DOPA in zebrafish, aims to demonstrate the feasibility of using modified commensal bacteria for autonomous drug delivery. By addressing the issue of medication adherence, particularly in neurodegenerative conditions, this approach holds promise for reducing avoidable healthcare costs and improving patients' quality of life!

Our Teams

---