Team:UZurich/Poster

Poster: UZurich

Plant Immunity Based Biosensing

Team: Erik Jung, Elio Marangi, Franka Butzbach, Jonas Trottmann, Michelle von Arx, Nicolas Fink, Philip Schulz, Tim Kurer
Principle investigator: Prof. Dr. Cyril Zipfel¹
Instructors: Dr. Isabel Monte¹, Dr. Kyle Bender¹
Advisors: Cauã Westmann², Dr. João Molino³

1: Department of Plant and Microbial Biology, University of Zurich
2: Department of Evolutionary Biology and Environmental Studies, University of Zurich
3: Department of Organic Chemistry, University of Zurich

Abstract

Bacterial contamination in water is a global issue that affects developing nations and first world countries. Even Switzerland, famed for its drinking water, faces over 400 cases of Legionnaires disease annually [1, 2]. But plants have been combating microbial pathogens for far longer than humans and we believe there is a lot to learn from them. That is why we are developing a biosensor based on plant pattern recognition receptors (PRRs). They are cell surface receptors of the plant immune system that dimerize in the presence of microbes. We designed a system based on EFR, CORE and their co-receptor BAK1, which recognize a broad spectrum of bacteria. We fused a split-luciferase to our receptors in order to quantify the total bacterial load of a water sample based on the luminescence-output [3]. We achieved the expression and localization of PRRs in yeast, which opens the door to future applications of PRRs as biosensors.

Introduction

It is known that our drinking water pipelines hosts diverse bacterial communities, unfortunately several among them being pathogenic to humans. If one considers the water quality found in developing countries, it should become evident that these countries face an even greater challenge when it comes to dealing with waterborne pathogens. We decided to face this challenge and designed a novel biosensor, while specifically focusing on cheap, rapid detection.

We were inspired by the natural microbial sensing circuits found in plants. Plants are continuously challenged by environmental stress and pathogens, so they rely on precise and fast detection mechanisms, to adapt their defense responses accordingly.

Figure 1: Plant pathogen detection with membrane localized PRRs. Created with BioRender.com

On the cell-surface sit so-called pattern recognition receptors (PRRs). These recognize pathogen- or microbe-associated molecular patterns (PAMPs or MAMPs), which then leads to further downstream signaling, finally inducing pathogen triggered immunity. Some receptors are targeted against more specific elicitors, while others will recognize highly conserved structures, present in a huge spectrum of microbes.

Our goal became not only to create a PRR-based microbial detection system, but also to lay the groundwork for future systems that wish to implement PRRs in microorganisms as biosensors. Localizing PRRs in non-plant membranes has not been achieved to date, it was the goal of this iGEM project to change this.

Problem

In the western world, we take safe drinking water for granted. But in many parts of the world, clean water is a luxury that is not always available [4]. It takes an enormous amount of effort to ensure that our water is safe for use and free of contamination. New technologies have the potential to make it easier and cheaper to test water and make clean water more broadly available. To aid in that effort, we aimed to develop a method which detects water contamination by designing a new type of biosensor. To do that, we made use of the plant immune system.

Legionella and other bacteria are particularly difficult to combat in water distribution systems, due to their ability to form biofilms, allowing them to better cope with biotic and abiotic stressors. Post-disinfection regrowth in water testing sites poses a great challenge. As such, many tests across the distribution system are necessary to ensure safe drinking water. Currently, many testing methods exist, but they all have drawbacks: either they are labor intensive, expensive, have low sensitivity or are not quantifiable.

Idea

We want to develop a biosensor that makes use of the abilities of PRRs to detect bacterial contamination in water. For that, we express PRRs in our detector organism, Saccharomyces cerevisiae. To detect as many bacteria as possible, we selected a number of PRRs that sense epitopes which are ubiquitous and highly conserved across a wide range of bacteria. To make pathogen detection measurable, we chose a luminescence-based output. Our chosen PPRs dimerize with their co-receptor BAK1 when they bind to their epitope. We make use of that dimerization by fusing one half of a split luciferase to the receptor and co-receptor. Upon ligand binding, the two luciferase halves are brought together, and thus its activity is restored. This produces a quantifiable luminescent output. This approach allows initial testing of water samples delivering a rapid visual output to measure microbial load. Our goal is to improve on these methods, by creating a bioreporter system that would be equally cheap and rapid but allow for greater quantification and simplicity than currently available systems.

Figure 2: Our envisioned detection system. Created with BioRender.com

Bioreporters are a subcategory of biosensors, that utilize a whole organism as a base for a biochemical detection system. We believe a bioreporter approach would fulfill our needs perfectly, as such a system can reproduce itself and is thus extremely cheap to mass produce [5] and can be designed to possess a rapid and simple output. Additionally, through a well characterized output, biosensing systems can be made highly quantifiable. With these tools we hope to combat the prevalent problems that water testing currently faces.

Integrated Human Practices

We knew we wanted to tackle water contamination with a PRR-based biosensor - but the design of the system, which PRRs and what output to use, was determined by our interactions with experts, in the scope of human practices.

Our initial goal was a layman sensor with a colorimetric output, which is visible to the naked eye. This system would give a binary response to a threshold of bacteria in water.

However as we presented this idea to experts in biosensing, water safety and microbial detection - we learned that this implementation would not serve a purpose, due to many issues, key amongst them:

• Water samples always work against a massive background of harmless bacteria, water safety depends on low amounts of harmful bacteria
• Laymen have very few options of dealing with contamination, often only chlorination
• GMO biosensors cannot be disposed properly by laymen

However we also gained a lot of feedback that allowed us to adapt our project. We changed our goal and our system. It now functions as a quantifiable sensor that detects the total bacterial cell count of a sample. This may be used by experts as an early stage water quality indicator. We also exchanged our output for a luminescence system, gaining greater quantification at the cost of requiring a machine for readouts.

Science Communication and Education

A large focus of our human practices was improving our project itself, but education was also very important to us. We believe the currently available programm for synthetic biology doesn’t do justice to the scope and potential of this field. So we organized many events to help spread our knowledge of synthetic biology, all of which we have also made available to anyone, on our website.

Events:

• Implementation of Synthetic Biology in Switzerland - an interdisciplinary discussion: 3 interviews from 3 different points of views on synthetic biology
• Highschool Presentation: Introducing modularity and creative thinking in yynthetic biology
• High School Workshop: Learning basic lab techniques of synthetic biology: restriction digestion
• University Presentations of synthetic biology, iGEM and our project

To introduce people to the field of PRRs, we created a PRR Guide - providing the necessary background and resources for future iGEM teams and other researchers intending to make use of PRRs [6].

Design

Chassis

We chose Saccharomyces cerevisiae and Chlamydomonas reinhardtii, a single cell green algae, as our chassis. S. cerevisiae is one of the most used and best understood organisms in synthetic biology. Another appealing aspect is its ability to be lyophilized (freeze-dried) and thus could be stored for a long time at room temperature, facilitating its usage as a biosensor. Our second choice fell on C. reinhardtii, which is a popular organism for biosynthesis and has been used in previous iGEM projects, indicating its use as a potential chassis for our receptors.

The detector module

We chose one co-receptor (BAK1) and two PRRs: EFR and CORE, which recognize the bacterial elongation factor-Tu and cold shock protein respectively [7]. These receptors recognize an extremely broad spectrum of bacteria, with the elongation factor-Tu being highly conserved in all bacterial groups we have analyzed. The main receptors only dimerize with their co-receptor BAK1 in the presence of a specific bacterial epitope. These properties allow us to construct a PRR-based biosensor that detects the total bacterial load of a sample.

The reporter module

We have designed our system by attaching a split luciferase [8] to the intracellular domains of the receptors and co-receptors. In the presence of microbial contamination these receptors should dimerize and create a luminescent output. By integrating these receptors into S. cerevisiae, we can create a simple bioreporter that can be freeze-dried [9], stored and transported to remote locations with ease. By directly fusing our output to the receptor, we aim to obtain a luminescence signal that correlates directly with the number of bacteria present in a sample. Thus, a sample's bacterial load could be quickly and easily quantified using a luminometer or other related devices.

Figure 4: Figure 3: Constructs used to investigate dimerization via luminescence using the NanoBiT® luciferase system

Methodology

Cloning Strategy

For our cloning strategy Golden Gate Cloning [10, 11] was the obvious choice, as it enabled us to assemble our constructs from several fragments in a one-step process.
We used gentamicin and spectomicin as selectable markers [12, 13]. We assembled the fragments in a plasmid with a spectinomycin acetyltransferase and amplified the plasmids in an E. coli strain [13].
We would clone our Chlamydomonas constructs first into the yeast plasmids via Golden Gate assembly, and to get the fragments into the vector for Chlamydomonas, we used a technique called In-Fusion® cloning developed by Takara Bio [14]. Whenever fragments had been cloned into a plasmid, the success of the reaction was confirmed by sequencing the plasmids to ensure the right fragments were inserted into the vectors correctly and no mutations in the sequence have occurred. After having prepared all the constructs, our chassis were transformed with the respective plasmids.

To test which components of the receptors are necessary for its functionality, we made 3 constructs without the intracellular kinase domain (eEFR, eBAK, CORE) and two (BAK-, BAK+) retaining the intracellular kinase domain. Additionally we tested whether a yeast signalling peptide (mating factor alpha) was necessary by attaching it to all receptors except for BAK+ which retained the native A. thaliana signalling peptide.

Experimental Design

We came up with a three step strategy to test our PRR constructs in the chassis:

1. Expression: To check if the PRRs are expressed correctly, we assembled constructs were the PRRs were fused to YFP. That way, we can easily check for correct expression by measuring fluorescence in the samples with a plate reader or flow cytometry
2. Localization: To investigate the localization of the receptors we imaged the same YFP samples with a confocal microscope. If the constructs are trafficked correctly to the membrane, we should be able to observe the YFP fluorescence in a sphere-like structure at the cell membrane
3. Dimerization: Finally, we wanted to see if our system is functional. . Thus, we transformed our chassis with two plasmids: one containing the EFR PRR, and one containing the BAK1 co-receptor. Both of these parts were fused to one half of the split NanoLuc® luciferase. That way, if the PRR and co-receptor dimerize due to the presence of the elicitor epitope (EF-Tu in the case of EFR), the luciferase halves are brought together and the function of the luciferase enzyme is restored. We can then measure the luminescence intensity in the samples

Expression

We assembled PRR constructs containing a YFP-tag. We then measured fluorescence levels in S. cerevisiae, transformed with these constructs, and compared them to an untransformed control group. We measured fluorescence in a plate reader assay, and also using flow cytometry. Both experiments confirmed higher fluorescence levels in the transformed samples than in the control, indicating that the PRR constructs are expressed.

Figure 5: Fluorescence levels in S. cerevisiae expressing YFP tagged PRR constructs as measured with a plate reader

Figure 5 shows higher fluorescence levels for all YFP-tagged PRR constructs except the CORE receptor. Especially BAK-, eBAK and EFR show high fluorescence levels, indicating higher levels of expression.

Figure 6: Fluorescence levels in S. cerevisiae expressing YFP tagged PRR constructs measured by Fluorescence-activated cell sorting

To confirm the results we also measured the fluorescence with a flow cytometer. Figure 6 shows the distributions of fluorescence levels in 200000 measured cells per sample. Again, all samples except CORE show higher fluorescence levels than the control.

Localization

We then imaged the transformed samples with a confocal microscope to see if the constructs are localized correctly at the cell membrane. The membrane was stained with FM4-64. BAK-, eBAK and EFR (Figure 9, 10, 11, which have the yeast alpha factor signal peptide) all show correct localization at the membrane in some cells. Although BAK-, which retained the kinase domain, also showed up in what seemed like the vacuoles of the yeast cells. BAK+ still has the signal peptide from A. thaliana and we clearly demonstrated that this is not sufficient for correct trafficking to the cell membrane (Figure 8).

Fig. 8: BAK+ shows moderate YFP fluorescence which is not co-localized with FM4-64, indicating that it is not localized at the membrane

Fig. 9: BAK- shows strong YFP fluorescence with little co-localization with FM4-64, indicating that some gets trafficked to the membrane

Fig. 10: Most eBAK cells show strong fluorescence, clear co-localization with FM4-64 in a ring structure in about 1% of cells

Fig. 11: EFR co-localization at the membrane in about 25% of cells. Dense clumps in the cell indicate that many constructs still get stuck

Dimerization

After confirming good expression and localization, we assembled the composite parts with the split NanoLuc® attached to test receptor/co-receptor dimerization. These two parts were then co-expressed in S. cerevisiae. With both receptor and co-receptor present at the cell membrane, addition of the EFR’s elicitor elf18 should induce dimerization and increase activity of the split NanoLuc® system, resulting in higher luminescence. We performed multiple dimerization assays under varying conditions and test parameters.

Figure 12: Luminescence levels in samples coexpressing EFR and BAK1, CORE and BAK1, and an untransformed control. Measurement started 30 minutes after addition of elicitor

In summary, the split parts were correctly expressed, since the transformed sample showed much higher luminescence levels than the untransformed control. We were very happy to see the NanoLuc® system working in action. However, the dimerization - and with it NanoLuc® activity - did not seem to be driven by the detection of the elf18 epitope, the ligand for EFR. We are not sure why that is the case. In plants, PRRs have helper proteins, so called negative regulators, that suppress dimerization when no ligand is present. The absence of these factors might explain the high activity in the absence of the elicitor in our trials.

Conclusion

Our team successfully assembled several different composite parts to achieve our goal of using PRRs in a biosensing system. The main goal of these parts was to introduce the PRRs into our host organism S. cerevisiae and be able to check that expression, localization, and dimerization of our receptor/co-receptor system was functional. To achieve this, we first constructed two main types of our PRRs:

• The PRR fused to YFP, so we can check correct expression and localization using fluorescence measurements and microscopy.
• The PRR fused to the NanoLuc® SmallBit, and the BAK1 co-receptor fused to NanoLuc® LargeBit. These two parts are then co-expressed in S. cerevisiae to test dimerization using luminescence measurements.

We successfully demonstrated expression and localization of our PRR constructs. During the NanoLuc® dimerization assays, we observed that the functionality of the luciferase enzyme was restored from the two split halves. However, the dimerization of the NanoLuc® parts was not driven by ligand-dependent interaction of the receptors.

Proposed Implementation

The current design of our system would serve as a simple method for detecting the total bacterial cell count (bacterial load) of a sample. This can be done simply by adding the sample and luciferase substrate to a tube or microplate and inserting it into a luminometer. By having our sensors pre-packaged directly in a tube we ensure simplicity of operation and reduce the risk of escape for both our organism and the contaminated water sample.

Figure 13: Our envisioned detection system. Created with BioRender.com

Our system is low cost, has low labour intensity and is rapid. This makes it suited for taking many samples, for example across a distribution network. Due to its limited sensitivity and nonspecific output, it is not suited for making a final decision. Instead it serves as an early detection method to monitor changes in water quality and trace issues, such as bacterial regrowth and biofilms, to their source.

Future prospects

Figure 14: Future prospects. Created in BioRender.com

Our current implementation is a proof of concept. However, PRR-based biosensing has far greater future uses. This is due to the modularity of our system: The biochemical similarity amongst PRRs should allow for the interchange of both the extracellular receptor and the output rather simply.

As more PRRs are being discovered and engineered[15], the specificity and sensitivity of this system can be adapted to suit more precise needs, while still retaining its aforementioned advantages.

Alternatively the output can be adapted to suit the desired application, two possibilities being:

• Beta-galactosidase for a colorimetric, visible to the naked eye signal
• A signalling cascade and transcription factors for a more sensitive or sustainable system.

Contribution

From 23 new basic parts that we added to the iGEM registry, we further built 28 composite parts. It is worth highlighting this inclusive and well characterized collection of biological parts, offering a great base for future researchers.
The development of a new application for plant immunity receptors through the study and re-engineering of plant pattern recognition receptors (PRRs) has been an exciting and enriching journey. Establishing an endogenous plant system in yeast (especially a transmembrane-based one) was demanding. It posed many challenges in the lab and introduced us to a whole new conceptual and technical framework. Part of this framework includes the plasmid design, so that our constructs may be expressed and properly located. At the end of this project, we were very happy to have obtained initial results that pointed towards the establishment of PRR systems in yeast for biosensing applications.
Since we are introducing components that have never been used in iGEM before, we wanted to go beyond the design and characterization of new BioBricks and give this system the spotlight it deserves. We hope our efforts on expressing plant immunity proteins in yeast and discussing their potential implications for water sensing, inspire future research teams to work on this topic.
We are convinced that further experiments will lead to our goal of epitope-induced dimerization. With our results and efforts we hope to contribute to the establishment of the PRRs biosensing application and are excited to see the further development of our efforts.

Acknowledgements

University of Zurich, Faculty of Science
Department of Plant and Microbial Biology
Prof. Cyril Zipfel Group (Molecular and Cellular Plant Physiology)

We would like to highlight the sponsors, their support enabled our research and ultimately the participation in the iGEM competition.

• Platinum sponsors: University of Zurich (UZH), UZH Science Alumni
• Gold sponsors: UZH President’s Services, Microsynth, IDT, Promega
• Silver sponsors: Benchling, SnapGene, Praxis Dr. Sylvia Baumann Kurer

Poster authors are the iGEM team members as stated in the project summary. The icons of our poster were created by Michelle von Arx. The BioRender.com schemes were created by Erik Jung. Philip Schulz implemented the poster in the browser. The content is written mainly by Franka Butzbach, Timothy Kurer and Philip Schulz. For proofing and editing Elio Marangi, Jonas Trottmann and Nicolas Fink helped .

Refererences

Abstract
[1] (in German) www.bag.admin.ch/bag/de/home/zahlen-und-statistiken/zahlen-zu-infektionskrankheiten.exturl.html/aHR0cHM6Ly9tZWxkZXN5c3RlbWUuYmFnYXBwcy5jaC9pbmZyZX/BvcnRpbmcvZGF0ZW5kZXRhaWxzL2QvbGVnaW9uZWxsYS5odG1s/P3dlYmdyYWI9aWdub3Jl.html
[2] (in German) www.bag.admin.ch/bag/de/home/krankheiten/krankheiten-im-ueberblick/legionellose.html
[3] Wang F.Z. et al., Split nano luciferase complementation for probing protein-protein interactions in plant cells, J. Integr. Plant Biol. 2019; (Published online November 22, 2019. doi.org/10.1111/jipb.12891)
Problem
[4] www.who.int/water_sanitation_health/water-quality/en/
Idea
[5] van der Meer, J., Belkin, S. Where microbiology meets microengineering: design and applications of reporter bacteria. Nat Rev Microbiol 8, 511–522 (2010). doi.org/10.1038/nrmicro2392
Human practices
[6] 2020.igem.org/wiki/images/f/f1/T--UZurich--prr-guide.pdf Design
[7] Kunze, G., Zipfel, C. et al. The N Terminus of Bacterial Elongation Factor Tu Elicits InnateImmunity in Arabidopsis Plants, The Plant Cell, Vol. 16, 3496–3507, December 2004, doi: 10.1105/tpc.104.026765
[8] Wang F.Z. et al., Split nano luciferase complementation for probing protein-protein interactions in plant cells, J. Integr. Plant Biol. 2019; (Published online November 22, 2019. doi.org/10.1111/jipb.12891)
[9] J.-F. Berny, and G. L. Hennebert. "Viability and Stability of Yeast Cells and Filamentous Fungus Spores during Freeze-Drying: Effects of Protectants and Cooling Rates." Mycologia 83, no. 6 (1991): 805-15. Accessed October 26, 2020. doi:10.2307/3760439
Methodology
[10] Chiasson, D., Giménez-Oya, V., Bircheneder, M., Bachmaier, S., Studtrucker, T., Ryan, J., Sollweck, K., Leonhardt, H., Boshart, M., Dietrich, P. and Parniske, M., 2019. A unified multi-kingdom Golden Gate cloning platform. Scientific reports, 9(1), pp.1-12
[11] Engler, C., & Marillonnet, S. (2014). Golden Gate cloning. Methods in molecular biology (Clifton, N.J.), 1116, 119–131. doi: 10.1007/978-1-62703-764-8_9
[12] Vardanyan, R. S., & Hruby, V. J. (2006). 32 - Antibiotics. In R. S. Vardanyan & V. J. Hruby (Eds.), Synthesis of Essential Drugs (pp. 425–498). Elsevier. doi: 10.1016/B978-044452166-8/50032-7
[13] Phizicky, E. M., & Fields, S. (1995). Protein-protein interactions: methods for detection and analysis. Microbiological reviews, 59(1), 94–123
[14] https://www.takarabio.com/products/cloning/in-fusion-cloning
Future Implementation
[15]Albert M, Jehle AK, Mueller K, Eisele C, Lipschis M, Felix G. Arabidopsis thaliana pattern recognition receptors for bacterial elongation factor Tu and flagellin can be combined to form functional chimeric receptors. J Biol Chem. 2010 Jun 18;285(25):19035-42. doi: 10.1074/jbc.M110.124800. Epub 2010 Apr 21. PMID: 20410299; PMCID: PMC2885181.