Communications

Overview

We had science communications during different stages of our project. Firstly, we designed a socially-oriented questionnaire to find out how much the public knows about Shigella. Then, through expert interviews, we managed to gain insights into our project feasibility and possible future directions. Finally, we participated in on-campus communication to let more people know about Shigella and surface display technology.

Socially-Oriented Questionnaire

Before the experiment, we designed and published a questionnaire about antibiotic treatment and dysentery on social networks, and obtained more than 500 responses from people of different regions, age groups, and education levels in China. We found that many people, although aware of dysentery, do not know how it is transmitted or the bacteria that causes it. This makes us realize the importance of science popularization. Meanwhile, people have different levels of concern about antibiotic resistance, which means we must focus on antibiotic therapy replacement.

Interview with Dr. Cao

We interviewed Dr. Huansheng Cao, Assistant Professor of Environmental Science at Duke Kunshan University for suggestions on our initial lab plans.

In the interview, team members Yuxin Wang and Chang Shen presented the main design—bacterial surface display—to consult Dr. Cao on its feasibility. After confirming the rich research foundations of our plan, he encouraged us to add computational elements to our lab design.

As a scientist working in computational biology and bioinformatics, Dr. Cao pointed us to Alphafold2 to enhance the binding affinity of our nanobody and the antigen unit. As a well-trained biologist, he also suggested fitting our design to more than one model organisms, which shows integration in our drug design.

The interview was full of enthusiasm, so we were able to come across many other topics. We discussed the trends in bioinformatics and biotechnology, including bacteria-based bioelectricity, and the relationship between human longevity and gut microbiota.

This interview was both encouraging and informative, and contributed greatly to the first stage of human practice.

Interview with Dr. Beckford

After we successfully predicted the 3d structure of the fusion protein-antibody complex by using Robetta and Alphafold2, the focus shifted to how the fusion protein will influence the binding (docking) between antibody and antigen. We plan to get some insights about this from computer modeling before the wet lab begins.

We interviewed Dr. Floyd Beckford, Professor of Chemistry at Duke Kunshan University for suggestions on our protein docking simulation.

In the beginning, team leader Lisa Wu introduced the synthetic biology part of our project. Modeling team leader Zhenyu Xu introduced the progress of protein structure prediction and the need for protein docking simulation.

As a scientist working in modern medicine, Prof. Floyd pointed us to some useful docking software: Autodock, ADFR suite, GOLD, and Maestro. These pieces of software are powerful at predicting the docking scenario and are used by many researchers, with many online teaching resources accessible.

To improve the speed of computer modeling, Prof. Floyd emphasized the preprocessing of protein structures. Namely, the antigen structure. He suggested that since we only care about the binding scenario, we should delete the floating molecules outside the binding sites and keep the inside ones. He also pointed out that virtual screening could help if we want to do more mutations to the binding sites. Concerning the fusion protein-antibody complex models, since they are the optimal prediction given by Robetta and Alphafold2, we assume these models do not need preprocessing anymore.

For docking, he suggested we try rigid docking first, and then flexible docking if we want to get closer to reality.

Interview with Dr. Neel Joshi

Introduction

Dr. Neel Joshi is an expert in bioengineering with tracks in human disease and new therapies. In his 2019 and 2021 literature, he and his team described genetically engineering E. coli Nissle 1917 to treat some gastrointestinal diseases like bowel inflammation, which provide essential references for our project. To further understand the logic and advantages of this bioengineered model, we invited Dr. Joshi for an interview.

Interview Content

Q: What stimulated you and your team to develop the surface-displaced single-domain antibodies? What practical problems do you think this model can solve?

A: Although nanobodies have great potential in treating gastrointestinal diseases, it is somewhat hard to deliver those nanobodies directly to the digestive tract. Therefore, the probiotics can serve as a scaffold to make the delivery of nanobodies easy.

Q: But why did you choose to display the nanobodies on the surface? Why not directly make the bacteria synthesize nanobodies and release them?

A: The surface display resolution has the property of multi-valency. By displaying the nanobodies on the surface, we would have the potential to space the binding domains together to bind to a pathogen surface. It could increase binding affinity and decrease the dissociation constant. It can also diffuse in different directions and increase the local concentration.

Q: What about the method of drug delivery? That is to say, how do you decide to send the engineered probiotics to the human gastrointestinal tract? And could you explain why you use that delivery method?

A: We only proposed an approach of micro-biological therapy. But to deliver the bacteria, I think it can be freeze-dried to produce a freeze-dried product. Because a large portion of the geographical locations that suffered from these gut problems are poor areas, and those areas do not have the condition to store the bacteria that require a very low temperature. The freeze-dried products do not require that strict storage conditions, so they are more available in those areas.

Q: Why pBbB8k plasmid backbone with kanamycin selection marker? (what's the difference between pUC19, which is our choice, and pBbB8K?)

A: pBbB8k is a rather arbitrary choice, we simply chose it because it was used due to antibiotics incompatibility in the previous experiment. When doing in vivo studies, we engineered plasmids from the original E. coli Nissle genome, the pMUT plasmids. These plasmids are more stable than the pBbB8K. For pUC19, I don't see problems with the expression of the pUC19 plasmid. However, you do need to consider if your antibiotic-resistant marker in the plasmid is compatible with the Shigella testing. If Shigella is killed due to the antibiotic selecting E. coli, it can be a problem.

Q: What are the setbacks of the Probiotic-associated therapeutic curli hybrids (PATCH) system?

A: The first is the safety concerns of curli fibers. It has been known that curli fibers induce various types of diseases, including neurological diseases, etc. The other is inherent disadvantages for E. coli to produce curli fibers in vivo. However, we have been able to show the effective production of curli fiber in vivo by measuring the amount of curli fibers coming from the fecal samples.

Q: What if the E. coli went out of control and grow infinitely in the gut, disrupting the original gastrointestinal microorganism environment?

A: This is a very popular question when it comes to distributing microbes to the gut. Actually, we could use some of the auxotrophic strains to limit the growth of the bacteria. And I think there are many "switches" that can turn on the suicide program of those bacteria. But in this case, I think E. coli Nissle 1917 is a very safe bacteria, it has already been identified as a harmless bacterium, so there should be fewer safety issues of this strain.

Q&A Section

Q: Is it possible to apply magnetic triggers to Shigella infection and cooperate this into the PATCH system?

A: Magnetic triggers are mostly applied in cancer treatment. This is because, in cancer, you know the spot of the disease. In bacterial infection, it's hard to locate.

Q: In Barta et al., the authors described the different indexes of the nanobodies, including binding domains and neutralization effects. What is your rationale when choosing nanobodies?

A: It really depends on how you generate the nanobodies. When different parts of the pathogen are used in Alpaca immunizations, the nanobodies will vary. When displaying the nanobodies, some work and others don't. We assume this is because of folding problems, but we did not figure out the exact reason.

Q: What do you think about sequence optimization in silica? Do you have experience working with this? Do you have any suggestions?

A: I would suggest generating different mutations in the sequence and do some docking analysis. Then generate the mutations and do wet labs.

Q: Our project aims to tackle the antibiotic resistance problem worldwide. Would the PATCH system generate drug resistance issues?

A: This is possible. For example, a mutation in the binding domain of the antigen might cause binding problems. However, I would say the selective pressure by the nanobody is smaller compared to that of the antibiotic. This is because antibiotic kills the pathogen, while PATCH only binds to the pathogen, keeping it from invasive activities.

Q: We are worried that the proteins in the GI tract environment would coat the surface of the probiotic, which would cause dysfunction of the nanobodies. How did your team encounter this problem?

A: The animal model was effective, which was sufficient to show that the GI tract environment did not inhibit the binding activity. The curli production is continuous in vivo, which provides an adequate chance for the binding to occur. However, the proteases in the GI tract are something to consider.

Summary and Reflection

Dr. Joshi provided lots of perspectives from a brand-new aspect, which were extra inspirable for the whole team. First, Dr. Joshi elaborated on the practical problems that this model could potentially resolve. That the probiotic expression system could make the delivery of nanobodies easier expanded our understanding of the model as the delivery problem does not only exist in the step of bacteria delivery, it starts from the delivery of the essential nanobodies.

Next, the freeze-dry delivery approach gives a less strict storage condition of this drug, making it more available in poor areas, which corresponds with our assumed delivery method. As for the safety problem that the public is most interested in, Dr. Joshi proposed using some molecular switches to eliminate those bacteria in addition to using safe strains like Nissle, giving important clues and indications for our project design.

In the Q&A section, questions regarding drug localization, drug resistance, choice of nanobodies were mentioned by team members. Dr. Joshi gave many valuable suggestions depending on the questions. Overall, the interview was successful, and we were able to arm ourselves with adequate knowledge for the wet lab.

Interview with SINOVAC

See Entrepreneurship.

On-Campus Communication

As an undergraduate team, we participated in the 2022's College Student Innovation and Entrepreneurship Training Program—a national training program for college students on innovation and entrepreneurship. In this way, we can better promote the topic of synthetic biology to the DKU community and attract more fresh blood to participate in innovation. We made posters and participated in project exhibitions, and were regularly inspected by the jury on campus.

During the final pitch, Lisa Wu and Lingrui Lai made the presentation about our project of Shigella treatment via synthetic biology methods. Lingrui went through the basic concept of Shigella and synthetic biology, then, the current challenges of Shigella treatment and our suggestions.
Lisa introduced the methodology involved and techniques used in our project. In the end, they stressed the significance of our project.

Judgers include Kai Huang (Division Chair of Natural and Applied Sciences of DKU), Joohyun Lee (Assistant professor of biology of DKU), Anastasia Tsigkou (Associate Professor of Biology of DKU), Ming-Chun Huang (Associate professor of data and computational science of DKU), Liqi Ren (Associate Director for Innovation and Entrepreneurship).

After the presentation, Prof. Lee and Dr. Tsigkou presented valuable questions and suggestions.

Prof. Lee recognized the difficulty of antibody therapy and informed us of the clear restrictions of FDI approval. We assured him that the nanobodies used in our project are thoroughly researched, and that we would continue to modify the project to enhance biosafety. Dr. Tsigkou asked us about the binding affinity test and false positive. This was a question that did not come to us then, so we answered that the detection might be conducted through fluorescence microscopy.

After discussing with our PI, Prof. Linfeng Huang, we decided that we would use the pull-down method in the detection of binding affinity. We would conduct the experiment by pulling down on both sides of the nanobody-antigen complex. For false positives, DKU iGEM member Lingrui Lai suggested isotype testing.

Summary

Throughout the project, we never stopped science communications with different groups. The social questionnaire allowed us to identify the need for the program; the school publicity has attracted a batch of fresh blood to join the research and innovation of synthetic biology; the expert interviews enabled us to reflect on the feasibility and future directions of our lab design. These science communications helped us become more clear about the direction of the experiment and the significance of this project.