Human Practices | iGEM Hamburg 2025

• Department for clinical toxicology and specialist in internal medicine at the Technical University of Munich

"It is most important that treatment starts early"

• Treatment of amanitin intoxications has to start as fast as possible after ingestion
• α-Amanitin enters the enterohepatic circulation which makes it especially dangerous for the liver
• If you are a molecular biologist, an eight-amino-acid peptide seems quite small, but it is rather large for toxicologists!

As liver cells seem to be able to secrete α-amanitin before it enters the enterohepatic circulation, we plan to provide as much nanobody as is needed to bind the toxin in the nucleus and prevent it from blocking the polymerase. The cells will have enough time to excrete the toxin and are protected in the meantime. This approach can be combined with uptake blockers such as silibinin or indocyanine green, a promising antidote against α-amanitin which Prof. Eyer told us about.

---

From the recommended paper regarding indocyanine green, we learned that α-amanitin also affects mechanisms taking place in the cytosol. Thus, we incorporated restriction sites around the nuclear localization signal into the design of our nanobody construct. This way, we will be able to create nanobodies with and without nuclear localization.

---

As we learned about the existence of ELISA assays to detect amanitin in the urine of patients, we want to use those assays further down in our project to verify amanitin concentrations during different tests. Sadly, we couldn’t get access to the structure of the antibodies against amanitin used in those assays, which could have been interesting for our nanobody in silico design.

---

To get moreled inform n about nntoxications, we approached Prof. Dr. Florian Eyer for an interview, who is head of the department for clinical toxicology and specialist in internal medicine at the TU Munich. At his department, he sees many poisoned patients and many forms of poisonings regarding ethanol, drugs and medications, but also mushroom poisonings. Additionally, his department works closely together with the poison center of Munich to give advice on intoxications and performs drug testing and screenings when a patient enters with an unknown poisoning.

---

Prof. Eyer explained that for intoxications with one of the many Amanita species existing all over the world, the main treatment is symptomatic. It includes fluid resuscitation, administration of activated charcoal and treatment with the drug silibinin, which blocks the OATP1 transporter responsible for transport of α-amanitin into the liver cells. For countries who do not have access to silibinin, penicillin or N-acetylcysteine are the treatments of choice. A recent publication in Nature Communications also identified indocyanine green as an inhibitor for STT3B, an enzyme playing an important role for amanitin toxicity. However, it is crucial that those treatments are administered as fast as possible, as Prof. Eyer stated. Even if there is only the suspicion of an amanita intoxication, silibinin treatment is started right away. Although mushroom poisoning can be confused with other gastrointestinal problems, amanitxications are standing out, because the symptoms may not appear immediately, but only six to twelve hours after ingestion. These symptoms include severe vomiting and diarrhea, making administration of oral treatments like activated charcoal difficult. As the toxin enters the enterohepatic circulation after digestion and is secreted by hepatocytes into the bile, it is reabsorbed from the digestive tract into the liver. Some of it is excreted via the kidneys as well, and urine tests have been developed to test for the toxin up to 72 h after ingestion. Fortunately, Prof. Eyer told us, they usually do not have many severe cases in need of liver transplantation in Munich.

---

The poison center in Munich sees between 50 to 100 inquiries per year regarding mushroom poisonings, and Prof. Eyer estimated around 20 to 50 cases of α-amanitin poisonings per year in Germany. Case numbers largely depend on weather, as the mushroom grows best having a good mix of rainy, humid days and a few sunny, dry days.

---

In Prof. Eyer’s opinion, developing an antidote for α-amanitin could be rather tricky, as the molecule is quite large for a toxin. He suggested also looking into lipocalins or anticalins, synthetically developed peptides used for binding and blocking toxins. Additionally, he pointed out that solubility plays a large role for toxicity and absorption in general.

---

We want to thank Prof. Eyer very much for the fun interview and the helpful tips regarding literature and additional experts!

• Postdoc at immunology department of the university medical center Hamburg-Eppendorf (UKE)

"The good thing about nanobodies is that they have this expanded CDR3 loop which can be bent in the ways you want to"

• For controlled excretion of the toxin-nanobody complex from the body, activation of the complement system is crucial
• It is possible to use a construct of fused nanobodies for specific cell uptake, where the first nanobody targets the cell for delivery and the second one targets the toxin for treatment
• Nowadays, llama mice (genetically engineered mice with the immune system of llamas) can be usedimmunology laboratories for nanobody production

As Dr. Schäfer told us, Ni-NTA affinity chromatography is a typical apprch for purification of his-tagged nanobodies in immunology laboratories, which validated our choice of this tagging and purification method for our nanobody construct. Furthermore, the focus of the delivery part of our project got expanded by this enriching discussion. Besides our LNP-mediated uptake approach, we now additionally investigated the possibility to fuse a second nanobody to the one binding the toxin, which is able to directly penetrate the cell membrane. The fusion to other nanobodies could not only be helpful for cell penetration but for extending the half-life of nanobodies as well, which is another aspect we considered in our further research for a more advanced therapy version. Upon the recommendation to try the software Chai2 for the in silico development part, we reached out several times to gain access and run our calculations with this software. Unfortunately, no free version exists since we got no response to our outreach, no cooperation could be started.

---

Following our invitation to give a talk as part of this year´s summer school, Dr. Schäfer gave us insight on the ongoing research projects regarding nanobody therapeutics at the university medical center Hamburg-Eppendorf (UKE) as well as already existing treatments.

---

We learned that there are currently four FDA-approved treatments for therapeutic application of nanobodies. The majority of investigations in this field are done in the context of cancer treatment. Nanobodies are very well suited to be used as a building component in CAR-T cells. The nanobody is in this case the sensing domain, recognizing the tumor by a specific surface cell marker as for example the protein CD38, binding to it and thus mediating the tumor’s destruction via the T cells.

---

As part of the group’s ongoing research, bispecific nanobody constructs are being developed. Those are designed to bind different partners at the same time. For the application in cancer treatment, they are able to bind CD38 with one arm and natural killer cells (NK cells) with the other one. The NK cells are recruited for killing the tumor cells. The same approach is followed in their investigations on treatment of Duchenne Muscular Dystrophy. To treat this disease, gene therapy with the adeno-associated virus (AAV) as a vector is considered. For targeted transduction the DNA of a bispecific nanobody construct integrated in the AAV´s capsid is explored. The group is also working on nanobodies for simultaneous imaging with conventional antibodies during cancer treatment. Due to their small size, the fluorescently labeled nanobodies can still find an epitope to bind to.

---

In our discussion Dr. Schäfer pointed out the short half-life of nanobodies and their quick clearance through the kidneys. A possible way to prevent this is to use a construct consisting of several nanobodies. One of them could bind to albumin, an abundant blood protein, thereby extending the nanobodies half-life in the serum.

---

One of our most imminent questions is to investigate how the complex of nanobody and toxin can be cleared once they are bound together. Dr. Schäfer made clear that the nanobody binding alone is probably not sufficient for clearance from the serum. To activate the complement system, the formation of an immune complex is necessary.

---

We found out that the purification of nanobodies which were expressed with a his-tag is done with a Ni-NTA affinity chromatography, which we also planned to use, and that for the following lab-tests and experiments the his-tag is not removed. Furthermore, Dr. Schäfer recommended RFdiffusion and Chai2 for the in silico nanobody design which is a difficult task due to the modifications of the toxin and its bicyclic structure.

---

To thoroughly investigate specificity and clearance of a nanobody, it is inevitable to conduct animal studies which are not part of the framework of iGEM. This could be a thread to follow if the research on our project continues after the iGEM competition. We want to thank Dr. Schäfer for his very interesting presentation on medical applications of nanobodies and for the enriching discussion of our project.

---

"You guys picked a really really really hard problem"

• First you design a protein that binds, then one that binds with high affinity
• Non-proteins are hard to design binding proteins against
• Protein structure design is a powerful tool that is evolving every day

Our AI-based pipeline for designing nanobodies as binding proteins for alpha-Amanitin was refined with input from Prof. Schoeder. The use of different models capable of considering all atoms and non-protein targets helped us improve our modeling approach and reach promising candidates. The stringent endpoint for nanobody design was relaxed to better accommodate our goal of generating an expressable, therapeutically useful, and designable binding protein.

---

As we were facing initial problems with our in silico nanobody design, we met with Jun.-Prof. Dr. Clara Schoeder, who specializes in the development of immunotherapeutic drugs at the Institute of Drug Discovery at Leipzig University. On top of that she is an associated member of the Center for Scalable Data Analytics and Artificial Intelligence.

---

After we introduced her to our project, she assured us that we chose quite a hard problem to tackle, as most prediction methods work great with proteins nowadays. However, the moment you work on something else, like in our case a small and highly-modified peptide, the accuracy of prediction drops dramatically. Hearing about our troubles with designing a nanobody that binds to the toxin, she suggested considering other binders for the toxin too, such as helical proteins. In her experience, designing a helical protein to bind a small molecule worked well and the predictions are more reliable. But the binding is only the second of two problems: First, we need to actually make a protein, and then that protein has to bind to our toxin. So she advised us to take one step after the other.

---

Regarding immunogenicity of such helical proteins, she pointed out that for our application, which is thought to be used only once in an individual, using a recombinant helical protein should not lead to an antibody-mediated immune response after just one administration. A big problem we faced were the modified amino acids in the toxin, which cannot be used in the protein design process of RoseTTAFold diffusion (RFdiffusion). Prof. Schoeder suggested using RFdiffusion All-Atom or Boltzdesign instead, tools which can also be used for structure design but are capable of considering non-protein ligands. Another complication could be different conformations in which our toxin can exist. When designing our binding protein, we have to make sure that the structure for alpha-Amanitin we are using does actually exist. As we are using a structure resolved by X-ray crystallography, this is likely not a problem for us. Additionally, as free and bound compounds exist in equilibriums, we can “force” the compound to adapt the preferable conformation for our binder.

---

In Prof. Schoeder’s opinion, we should definitely consider docking for testing our nanobody structure, as performing multiple rounds of docking can be used to tell apart good models from bad ones using the RMSD. We are quite free in the choice of our docking program, as they are performing in a similar way, while some are easier to set up than others. However, a warning remains: if we manage to design a protein that binds to our toxin, we should expect it to bind quite poorly, at least at first. It is highly unlikely to design a fantastic binder with high affinity right from the start, and many publications dealing with this are testing a lot of structures in the lab before something works out.

---

For further help, Prof. Schoeder gave us the contact of one of her PhD students who works with the methods we planned on using and who will help us along the way. We would like to thank Prof. Schoeder very much for the practical tipps on tools and experimental methods, and the great interview we had!

• PhD molecular biologist
• independent entrepreneur organizing Science Slams in all of Germany

"Why do you want to write a children’s book? Why should someone read it?"

• It’s important to know who your target audience for science communication is and why you want to reach out te in the first place
• If you want to reach a lot of people with your work, social media is a good way to go
• Finding established partners for science communication and outreach, such as the German initiative Jugend forscht (“youth researching”), could be helpful to bring a physical book to interested people

Following Dr. Offe’s tips, we reached out to various organizations for science communication, especially for students, such as the MINTforum Hamburg (a cooperation of initiatives for children and students interested in STEM). We also discussed the audience we want to target in general regarding age, previous knowledge and interest in scientific topics. As we want our book to be available to everyone, we decided against the option of taking donations in exchange for downloads, and linked the PDF here on our wiki.

---

During our Summer School "Challenges of Synthetic Biology: From Theory to Practice", Dr. Julia Offe held a workshop on science communication, especially through science slams, short and creative presentations about scientific topics targeting a wide audience. In the context of creating a book on nanobodies for a general audience, we approached her for advice on which stakeholders to engage anlearning to scientific research.

---

Dr. Offe first caught us with a simple question: why do we even want to spread the information on nanobodies? Why should someone read our book, what would they get from it? This simple question already showed us that a lot of thought can go into science communication, not just on how to convey information, but also on why. We went on to discuss our target audience. Not only should the language be age-appropriate, but it can also be helpful to reach out to people with a basic level of interest in scientific topics. This could be students who engage in scientific projects outside of school or other people interested in sciences in their daily lives. Dr. Offe suggested reaching out to institutes responsible for creating textbooks for students or initiatives such as Jugend forscht, a competition for students in Germany participating in science projects. She noted that social media can be great for reaching a lot of people with relatively low effort. To spread physical copies of our book, she suggested handing them out to students in schools as well. Or, of course, we could participate in a science slam and talk about nanobodies.

---

This interview showed us that there is much more to writing a children’s book than just finding words and pictures.

---

We thank Dr. Offe for her time and valuable tips!

• University of Nova Gorica, Slovenia

"I mean, it's not a problem of weeks, of months, but of years, if you have to figure this one out."

• It’s not enough to find a high affinity binding nanobody. It also needs to be stable, not aggregating, expressable in E.coli AND must be able to be packaged into liposomes.
• Choose the problems at hand to work on, not multiple ones at different stages downstream of the whole process.
• Find ways to reduce the search space by multiple angles.

The feedback of Dr. Kropivšek and Prof. deMarco helped us assess the range of our search for a high affinity binding nanobody. Prof. deMarco was adamant to keep the developability/expressability of final designs in mind, so we did implement AI models in our in silico workflow, trained to predict solubility, fold stability and the probability to successfully express the nanobodies in prokaryotes. Additionally we made an effort to obtain a nanobody library to parallelize our in silico search with a screening method, but with Prof. deMarco’s input stopped this path in favor of concentrating on our initial design and validation efforts.

---

We were seeking a nanobody library—a collection of phages, each one displaying one single nanobody and containing the corresponding phagemid encoding the nanobody sequence—and identified Prof. Ario de Marco as a potential collaborator. After discussing our project's timeline, which likely would not accommodate the full screening, optimization, and validation process required to identify a high-affinity nanobody against alpha-amanitin, Prof. de Marco agreed to meet with us.

---

He was joined by his colleague Dr. Klara Kropivšek, a research fellow specializing in the application of machine learning to scientific research.

---

During our brief discussion, Dr. Kropivšek shared her own challenges with generating high-quality docking results using both classical and AI-based methods for nanobody-epitope docking. She cautioned us to interpret docking scores carefully, as AI-based models do not always yield the most probable binding modes, often favoring plausible but less likely options compared to classical algorithms.

---

Prof. de Marco highlighted that completing our project within the available timeframe would be challenging. As a result, much of the conversation focused on strategies to narrow the project scope and identify ways to prioritize key objectives.

---

One alternative Prof. de Marco suggested was to design a binder, such as a small molecule or nanobody, against RNA polymerase II, the target of alpha-amanitin. This strategy could sidestep some of the difficulties associated with in silico design against alpha-amanitin and potentially simplify both binder selection and subsequent nanobody development.

---

Dr. Kropivšek also introduced the idea of reducing the candidate pool. By filtering the in silico-generated sequences based on specific predicted characteristics—such as stability, yields, and suitability for packaging into liposomes—we could dramatically accelerate the screening process. She emphasized that our main bottleneck is the successful incorporation of nanobodies into lipid nanoparticles. Therefore, she advised integrating downstream application requirements into our initial filtering criteria to ensure our search is both focused and goal-oriented.

---

Overall, the discussion converged on the importance of defining a specific problem to solve, proving a core concept, and maintaining focus, rather than attempting to advance every aspect of the project simultaneously.

---

Additionally, it was recommended to consider developability at every stage of in silico design—not merely as a constraint, but as a way to streamline and refine the search process.

---

Finally, we received advice on expressing nanobodies in E. coli, optimized as application-specific reagents after their molecular fusion with tags suitable for effective and site-specific functionalization, and employing various methods for characterizing the purified proteins.

• Principal investigator and group leader at the University Medical Center Hamburg-Eppendorf
• HCTI department of medicine (UKE HCTI)

• Cell culture is a reliable and established scientific tool and is often used in drug development or toxicity testing. It enables researchers to study questions in a controlled environment with regulated parameters. This reduces variability compared to whole organisms and allows experiments to be repeated consistently. It also allows the investigation of a single cell type without interference from systemic factors in an organism
• Working with cell lines requires experiment-specific definitions and testing. Cells need time to grow into the correct state and require regular splitting and care. Additionally, setting up experiments with new compounds is challenging and time-consuming. Toxicity, concentration, and readouts all need to be defined and tested.
• Various liver models are in use today, ranging from the established human liver cancer cell line HepG2 to primary cells from mice or humans, as well as 3D organoids.

Together with Dr. Schwinge and her team, we designed a theoretical experimental workflow to test and characterize the effect of a nanobody antidote. Within this workflow, we discussed how the toxicity and efficiency of the nanobody antidote could be assessed in an in vitro cell culture setting. We considered both advantages (high reproducibility and availability of tools) and challenges (time requirements and possible readouts). Our discussions on available liver models, analyses, and visualization devices—such as fluorescence microscopy and fluorescence-activated cell sorting (FACS)—enabled us to assemble a series of potential experiments. For example, we discussed using an ELISA assay to validate our nanobody’s binding effect to the toxin with an anti-GFP nanobody and GFP. Since this system had already been used as a test candidate in the wet lab for our expression and production pipeline, it would also allow direct comparison with results obtained for the actual toxin. We also openly addressed potential practical challenges, such as identifying the correct concentrations, managing possible nanobody toxicity, and determining appropriate incubation times at each step. Since the generation of the nanobody has not yet been finalized and establishing the cell culture would require additional time, these experiments could not yet be carried out in practice.

---

During project planning, we discussed ways to test and characterize our antidote’s effect later on. As α-amanitin mainly targets the liver, we reached out to a group at the University Medical Center Hamburg-Eppendorf led by Dr. Dorothee Schwinge. The group works with liver cell lines as well as primary cells and liver organoids and agreed on helping us to set up a theoretical workflow for testing various conditions and validating delivery and localization of our antidote. We were joined by many of Dr. Schwinge’s PhD students who helped with ideas and input as well.

---

Dr. Schwinge agreed with us that we wouldn’t be able to perform any practical experiments in the remaining time for our projects, as we hadn’t produced any nanobodies yet. The human cancer liver cell lines we are allowed to work with in iGEM context need one to two weeks from thawing to being usable alone, and establishing cell culture needs testing and additional time. As we were already in the final weeks of our project time, we would not be able to properly establish and conduct our experiments practically. Instead, we opted for creating a series of thought experiments. In collaboration with Dr. Schwinge and her PhD students, we considered different options for cell types as models and delivery strategies for our nanobody. We also discussed the possibility of initially testing binding and affinity in medium or human blood, since it is more readily available at the Medical Center. Our approach was to first identify toxin concentrations causing cell death after specific incubation periods. We then aimed to evaluate changes in lethality following incubation with our nanobody. Additionally, we would need to find ways to visualize and analyse our hypotheses regarding localization of our nanobody in the nucleus and delivery, entry and release of the nanobody into the cells. Dr. Schwinge discussed different strategies with us regarding availability of methods as well as high reproducibility. She also offered us to join one of her PhD students if time allowed to learn how to cultivate and split the cells they use. Additionally, she suggested checking if the toxin is metabolized in some way before reaching its lethal form in the body. We were recommended to contact the group led by Dr. Anna Mann which we reached out to already to ask for nanobodies able to be taken up by hepatocytes. Once we finished our theoretical workflow, Dr. Schwinge gave us feedback and tips on our ideas.

---

We’d like to thank her and her PhD students very much for the helpful input and inspiration they gave us!

• Nanobody scientist
• Berking Theranostics
• Berking Bioscience

"Nobody expected to find antibodies in dromedars that were much smaller than conventional antibodies"

• Nanobodies are small, stable, and can bind targets inaccessible to conventional antibodies.
• Humanization of nanobodies is important for therapeutic applications to reduce immunogenicity.
• Production in bacteria or yeast is efficient and scalable.
• Nanobodies can be used for both diagnostics and therapeutics (theranostics).

Alejandro describing nanobodies, their mode of binding and their functions in diagnostics and therapeutics in such a creative and fun way sparked our idea to write a small book for children and non-scientists about nanobodies. The metaphors and pictures he used were already so fitting and vivid that we could transfer them easily into drawings and descriptions. We were really infected by his enthusiasm (no pun intended) and wanted to pass it on to others as well. Additionally, as Alejandro mentioned that nanobodies are usually humanized for therapeutic application to make them invisible to the immune system, we looked for humanized scaffolds in our in silico design approach. This way, our designed nanobody would hopefully be non-immunogenic as well.

---

At some point during the beginning of our project, we gained and instagram follower with the handle "alpaca_buddha", showing lots of handsome alpacas on their page. Curiously, we reached out and made the acquaintance of Alejandro Rojas-Fernandez. He is a nanobody scientist, proud owner of 27 "ugly" male alpacas, as he calls them, and founder and CEO of two companies working with nanobodies. Alejandro started his talk with the discovery of nanobodies in 1989. In Belgium, students were isolating antibodies from dromedar blood, when they found strange biochemical features, which were smaller than all antibodies known before. Those were later confirmed to be a new species of antibodies, called heavy-chain antibodies with single variable regions, also known as single domain antibodies, (VHH) or nanobodies. They are lacking the light chain antibodies possess and rather than acting like tweezers to hold and bind antigens, they rely on an elongated CDR3 to bind. Advantages of nanobodies, as Alejandro mentioned, are their small size enabling them to effectively penetrate tissues, their high stability regarding pH and temperature changes and their capabilities of binding small pockets or structures that can't be reached by antibodies with high affinities in the range of nM to pM. For example, nanobodies unfold at around 60 °C, and can be heated up to 90 °C for twenty minutes, and afterwards they will just refold and be good as new. Additionally, nanobodies can be produced recombinantly in bacteria or yeast with up to 20 g/L. After their discovery, nanobodies developed by immunizing camelids with antigens got patented by Raymond Hamers, and this patent only expired in 2017. In the same year, Alejandro bought his herd of alpacas. He started to work on the development of nanobodies leading to the degradation of proteins, and later tackled local and global viruses, for example Hanta virus, Nipah virus and Corona virus during the pandemic. Today, back in Hamburg, Berking Theranostics uses nanobodies to target tissue cells that infiltrate solid tumors. Those nanobodies can be conjugated to radiating material destroying the tumor, or labeling molecules for visualization, and are called theranostics, a mix of therapeutics and diagnostics. His other company, Berking Bioscience, aims to bring organic and sustainable therapies to the oceans. Regarding our project, Alejandro pointed out how horses are used to develop antidotes against snake venoms, by repeatedly immunizing them with increasing doses of the toxins. This could be possible with our toxin as well.

---

For his alpacas, they never use toxic components for immunization, but only fragments of dangerous proteins or inactivated viruses. The animals are immunized four times over the course of eight weeks with increasing doses and monitored before and after the immunizations. Then a small sample of blood is taken, from which the cells expressing antibodies and nanobodies are purified, and the alpaca is left alone for one year. As different lymph nodes hold different lymphocytes, immunization can lead to generation of different nanobody families with slightly different binding sites on the same antigen. He also points out that nanobodies generally struggle more to bind linear peptides, since they lack the "tweezer effect" found in conventional antibodies. However, if the peptide provides a 3D structure with some kind of pocket to bind, it should still be manageable. Nanobodies are produced recombinantly by isolating lymphocytes and preparing cDNA libraries for the respective nanobodies. The genetic information of the nanobody is sequenced and then used in bacterial display technologies to isolate nanobodies that bind best to the respective antigen. Alejandro uses a technique with beads conjugated to the antigen to pull down bacteria displaying the right nanobodies combined with a high-content microscopy analysis to identify high-affinity specific nanobodies. Currently, there is one treatment fully approved by the FDA using nanobodies, with three more in clinical trials. As the patent on nanobodies was only lifted in 2017 and clinical development can take up to ten years, not many treatments are on the market yet. Alejandro also mentioned that nanobodies are not new in Hamburg. Professor Koch-Nolte from the University Medical Center Hamburg-Eppendorf generated mice that carry the immune system of alpacas. He aims to further continue the implementation of nanobodies for the benefit of human and planetary health. So, we are just right on time with our project! We thank Alejandro so much for the inspiring and hilarious interview and wish him all the best with his many ideas in the future!

• Technical University of Munich - Toxicological Laboratory

"Amatoxin is probably the mushroom toxin that scares doctors the most."

• The death cap is one of many mushrooms containing amatoxins, but it can have the highest concentrations of α-amanitin with up to 6.8 mg per gram of dry weight
• Silibinin treatment works very well in Germany, while some other countries don't use silibinin as a treatment, but for example N-Acetylcystein
• Identifying death caps is not too hard if you have the right source of information and know how to use it. Only knowledge about mushrooms protects against mushroom poisonings!

From Ms. Haberl, we learned a lot of detailed facts about identification of the death cap mushroom, as well as numbers and facts regarding toxin concentrations in different mushroom species. We use this information on our wiki to explain risks regarding the death cap. As silibinin seems to be quite expensive as a therapeutic for poisonings, we aim to make our antidote as affordable as possible. Since we picked E. coli to produce our nanobody, we hope that upscaling for higher production yields will be easy and low-cost later. Ms. Haberl also gave us the contact information of Bühlmann Laboratories AG, the company producing ELISA kits against α-amanitin. We reached out to them, but unfortunately they were not able to give us more information on the antibodies used for detecting the toxin.

---

To get to know our target better, we interviewed Bettina Haberl, a mycologist and chemical-technical assistant at the toxicological lab of the Technical University of Munich. She informed us that the world of mushrooms, and especially their classification, is an ever-changing topic where lots of questions are still unanswered at the moment. The death cap belongs to the genus Amanita, which consists of roughly 20 species in Germany. Other mushrooms contain α-amanitin as well, in Germany mainly the genera Lepiota and Galerina. Taken together, there are roughly 9 known species of amanitin-containing mushrooms in Germany. Identifying mushrooms can be hard in general, as their appearance can vary a lot through their development. Mushrooms of the genus Amanita form mycorrhizae with plants. The species Amanita phalloides grows in deciduous forests with oak and beech trees. But it can also occur in parks or even big gardens. Mushrooms usually appear between July and October, strongly depending on the weather and temperature. The hotter and more dry it is, the less mushrooms appear. To identify mushrooms, it is vital to not just cut them, but dig them out of the ground. The death cap has a thick, bulb-like volva that is often hidden by leaves. Another characteristic feature is a hanging ring around the stem, as well as white lamellae. In the case of the Amanita virosa, the destroying angel, this can be used to distinguish it from the portobello mushroom which has brown lamellae. The cap of the death cap is green or greenish with radial stripes. It is often confused with mushrooms like green-cracking russula or other Russula species. Most patients suffering from mushroom poisonings are amateur mushroom pickers and family members who are served meals containing the harvest. Accidental poisonings, like toddlers eating parts of mushrooms, don't happen that often and are usually harmless. If poisoning with amatoxins do occur, doctors and experts are most alert and start treatment as early as possible. Even small amounts of the death cap, as much as one fruiting body with up to 6.8 mg of toxin per gram of dry weight, can be enough to kill an adult. Interestingly, the concentration of the toxin decreases over the growth period. Younger mushrooms have the highest concentrations, while older mushrooms are more and more watered down. The treatment of poisonings with amanitin-containing mushrooms varies from country to country. While Germany uses silibinin, other countries like Italy favor N-acetylcysteine. Ms. Haberl informed us that fortunately, she hasn't seen a patient in need of liver transplantation for quite a while with silibinin as a form of treatment. The downside is that silibinin is quite expensive, with costs around a thousand euros per dose. Not every country has financial resources for this, and not every hospital can buy and store silibinin. But with 1.5 to 2% of total calls, mushroom poisonings don't play a major role at the poison control center. The number one reason for mushroom poisonings, as Ms. Haberl tells us, is simply eating spoiled mushrooms, where bacteria cause gastrointestinal symptoms. Reason number two - mushrooms containing thermolabile toxins are not cooked thoroughly enough. α-Amanitin is a peptide consisting of eight amino acids, which is quite unusual for a mushroom toxin, as Ms. Haberl told us. Most known toxins, such as muscimol or psilocybin, are small molecules. So what would help against death cap poisonings? Probably learning how to use identification books and apps, in Ms. Haberl's opinion. They contain lots of useful information and especially apps carry a load of pictures and data. Only knowing the specific features and their combinations of toxic mushrooms protects against accidental poisonings. This knowledge can't be obtained by only looking at pictures and skipping descriptions. Especially skipping information about confusion with toxic species can quickly lead to dangerous situations. Or less severe, but still very unpleasant stomach problems. If you are interested in gathering mushrooms, you can reach out to mycological consultation centers, join a mycologist on mycological excursions or visit "identification nights" of mycology organizations. We thank Ms. Haberl very much for the fun and informational interview!

• University of Hamburg - Research Unit for Organismic Botany and Mycology

"Mushrooms vary greatly from region to region in terms of toxicity and the composition of their toxins"

• The mechanisms and roles of fungal compounds, especially small peptides called effectors, are complex and not fully understood.
• Mushroom toxicity, such as that of Amanita, varies regionally and is likely driven by evolutionary responses to environmental pressures like insect predation

Prof. Begerow’s field of study does not include the death cap per se, but he still gave us a nice overview and recommended other experts to us. Following his tips, we reached out to various other mycologists.

---

From several different sides, we got the recommendation to talk to Prof. Dr. Dominik Begerow about the death cap and its characteristics. He specializes in research on smut fungi (Basidiomycota, Ustilaginomycotina), which are plant parasites. The group’s focus is on the evolution, host specificity, morphology, and genetics of these fungi rather than on their metabolites. These fungi do not produce fruiting bodies like our death cap does, but we still got useful tips for our project.

---

Regarding the toxic death cap, Prof. Begerow explained that it is a typical forest fungus forming mycorrhizal relationships, and emphasizes the need for detailed identification, including fruiting body characteristics, color, and spore size—best conducted via microscopy or by specialists.

---

Parasitic fungi can also produce toxins, often proteins or peptides (effectors), which interact at various biological levels, sometimes reprogramming host cells. Identifying the specific functions of these peptides is complex and slow because their individual effects are usually minor, and there are hundreds per fungus.

---

Control of these plant pathogens is limited because most are not hazardous to humans—primary interventions focus on agricultural crops like wheat and maize. In some cultures, infected maize is even consumed as a delicacy.

---

Much of the research remains fundamental rather than applied. Toxicity of mushrooms varies by geographic region, influenced by local environmental pressures like insect predation, leading to variability in toxin concentrations and composition. These evolutionary adaptations likely serve as defense mechanisms. Amanita toxins such as alpha-amanitin are notably potent, with symptoms appearing hours after ingestion; reliable statistics on poisoning incidents are scarce. For mushroom identification in Hamburg, Prof. Begerow recommended colleagues at the university, e.g. Cornelia Heinze and local organizations e.g., Pilzberatung Hamburg. We thank Prof. Begerow for his time and the interesting interview!

• PI of our team and group leader of the lab hosting our iGEM team
• Secondary PI and biosecurity advisor of our iGEM team
• work safety at the University of Hamburg
• Key Account Manager of Bühlmann Laboratories AG
• forensic toxicology at the University Medical Center Hamburg-Eppendorf
• Biological Chemistry at the Technical University of Berlin
• A contact at Heidelberg Pharma

• Our biggest challenge in planning lab work with the toxin was finding out about waste management and proper disposal
• The risks of handling toxins depend largely on the quantities that are used
• Experimental workflows, including numbers of repetitions, need to be thoroughly planned to ensure that minimal amounts of toxin are purchased and stored

We sat down with Dr. Himmel to discuss potential risks and dual-use concerns regarding our project. We incorporated the results of our discussions into iGEM’s safety forms and reached out to additional safety experts he mentioned. With the help of Prof. Dr. Kolbe, a specialist for work safety at the University of Hamburg, Ms. Klug, Prof. Dr. Süßmuth and other contacts from academia and industry, we performed a thorough risk assessment and prepared new operating instructions detailing the handling of α-amanitin as well as proper storage and especially safe disposal of the toxin.

---

Who?

• Prof. Dr. Michael Kolbe, PI of our team and group leader of the lab hosting our iGEM team (www.helmholtz-hzi.de/en/persons/prof-dr-michael-kolbe/)
• Dr. Mirko Himmel, Secondary PI and biosecurity advisor of our iGEM team; he is principal investigator and senior project coordinator at the Department for Microbiology and Biotechnology at the University of Hamburg as well as Associate Senior Researcher at the Stockholm International Peace Research Institute within the Armament and Disarmament research area (www.sipri.org/about/bios/dr-mirko-himmel)
• A specialist for work safety at the University of Hamburg (www.uni-hamburg.de)
• Daniela Klug, Key Account Manager of Bühlmann Laboratories AG in Southern Germany (www.buhlmannlabs.ch/)
• A member of the forensic toxicology at the University Medical Center Hamburg-Eppendorf
• Prof. Dr. Roderich Süßmuth, chair of Biological Chemistry at the Technical University of Berlin (www.tu.berlin/en/biochemie/about-us/management)
• A contact at Heidelberg Pharma (www.heidelberg-pharma.com)

We first sat together with one of our PI’s, Dr. Mirko Himmel, to discuss potential risks and safety measures regarding our project. With his expertise on biosafety, biosecurity, and preventive arms control of biological weapons, he gave us impulses on risks that could arise from designing a nanobody against α-amanitin. For example, we discussed if the nanobody could accidentally bind intracellular structures and lead to unwanted side effects. Effective translocation of the nanobody is achieved by methods which could be perceived as delivery mechanisms for highly toxic substances usable by malicious actors to create great harm.

In order to get more detailed information about regulations regarding lab work with toxins, Dr. Himmel recommended talking to safety experts, for example a specialist for work safety at the University of Hamburg. For calculations on how much toxin we would need to purchase, we should think of which amounts were needed for experimental repetitions, and how many repetitions we wanted to do. He also suggested purchasing commercial test kits to be able to detect the toxin. Our entire team participated in a workshop led by Dr. Himmel about concerns regarding dual-use and misuse of knowledge and science. In general, we came to the conclusion that our project holds little risks apart from tests requiring the toxin. We only use synthetic biology to produce our nanobody, but don’t plan to release or use any modified organisms after the protein production. So, there is no immediate misuse potential in terms of drug delivery. Unfortunately, we cannot substitute the toxin with a less dangerous substance to test our nanobody, since binding of nanobodies is highly specific and could already be impossible with a slightly different molecule.

As no one had ever worked with α-amanitin before in our lab, we next sought to create a thorough risk assessment and operating instructions, which could be used as a basis for our safety training. Our PI, Prof. Dr. Michael Kolbe, suggested contacting a specialist for work safety at the University as well. She gave us information on how to create risk assessments, and pointed out important aspects such as keeping the access to the toxin restricted and prioritizing hand hygiene in addition to wearing gloves.

Based on the valuable input by Prof. Dr. Kolbe and the work safety specialist, we created a risk assessment and operating instructions for α-amanitin. During the preparation of our operating instructions, we relied on safety data sheets from commercial vendors. Interestingly, these documents did not detail procedures for proper disposal of contaminated waste, as those regulations vary between countries and even regions. As there were no specialists on dealing with α-amanitin-contaminated waste in particular at our university or lab, we researched groups or companies dealing with amanitin waste in Germany to get tips on proper disposal. Bettina Haberl, expert for mycology and toxicology, pointed us towards Daniela Klug who is key account manager of Bühlmann Laboratories AG, a company producing detection kits for α-amanitin. We reached out to her, but unfortunately she couldn’t help us out further as their lab only handles minimal amounts of the toxin that do not fall under special regulations.

We then reached out to a member of the forensic toxicology at the University Medical Center Hamburg-Eppendorf. They described how their lab detects amanitin in urine samples of patients. As these samples only contain the toxin in ng/mL ranges, they are disposed of as potentially infectious material, which does not apply to our project. We also got feedback from a contact at Heidelberg Pharma, a biopharmaceutical company working on antibody-drug conjugates based on amanitin. They informed us that their lab disposes of amanitin-contaminated solid waste like gloves or pipette tips in labeled barrels which are handled by the local specialized waste disposal company. Liquid waste is handled as non-halogenated organic waste.

Finally, we contacted Prof. Dr. Roderich Süßmuth, chair of Biological Chemistry at the Technical University of Berlin. His group discovered a way for the synthetic production of α-amanitin to produce large enough amounts for therapeutic applications such as cancer therapies (Yao, G. et al. Iodine-Mediated Tryptathionine Formation Facilitates the Synthesis of Amanitins. J. Am. Chem. Soc. 143, 14322–14331 (2021). ). Prof. Dr. Süßmuth explained that the toxin can be inactivated by acidic hydrolysis, and pointed out that as we would be using minimal amounts as well, this would be sufficient to safely dispose of the toxin. He also warned to not use hollow needles while working with the toxin and preventing the formation of dust. We incorporated all the information collected in our operating instructions.

Ultimately, we did not work with α-amanitin in our experiments due to time issues. Nevertheless, we are glad to have researched a lot about work safety and gained new insights in preparing and planning scientific experiments. We’d like to thank all our experts for their time and help!