Difference between revisions of "Template:Team:Utrecht/MainBody"

This year, Utrecht University participates in the iGEM for the first time. We aim to create a cheap DNA detection kit for disease diagnosis that is easy to use and does not rely on complicated sequencing technologies.

Disease diagnosis is of great importance for healthcare. In developing countries, diagnoses often have to be made based on limited information, even though accurate disease determination based on pathogen specific DNA sequences is possible through sequencing technologies. These technologies, however, require specialised equipment and expertise that simply is not available everywhere. The OUTCASST two-component system and detection kit hopes to alleviate this problem.

The OUTCASST two-component system consists of two proteins, expressed to the membrane of a dryable cell. One of the proteins is a Cas9-fusion and the other contains Cpf1. Both proteins can be given a guide RNA that makes it bind to a specific, user-chosen, complementary sequence. When both proteins bind a DNA fragment from a sample, possibly containing pathogen DNA, they co-localize, so that a transcription factor is released intracellularly which then induces an intracellular reporter mechanism such as a dye or fluorescent signal.

The architecture we use for OUTCASST is inspired by the Modular Extracellular Sensor Architecture (MESA) (Daringer, N. M., Dudek, R. M., Schwarz, K. A., & Leonard, J. N., 2014: Modular extracellular sensor architecture for engineering mammalian cell-based devices. ACS synthetic biology, 3(12), 892-902, http://pubs.acs.org/doi/abs/10.1021/sb400128g) (Schwarz, K. A., Daringer, N. M., Dolberg, T. B., & Leonard, J. N. 2017: Rewiring human cellular input-output using modular extracellular sensors. Nature chemical biology, 13(2), 202-209, http://www.nature.com/nchembio/journal/v13/n2/abs/nchembio.2253.html?foxtrotcallback=true). Because of this, MESA first needed to be replicated to verify whether the final product could work as intended. A successful replication will serve as an indication that OUTCASST would work, as well as provide data we can use to compare and correct models of the system.

The first few weeks, we tried to replicate the MESA construct using Luciferase-GFP fusion protein as the output signal and dsRED as a transfection control. After a Skype Call with the MESA authors we decided to use YFP as the output signal and BFP as a transfection control to prevent signal overlap, equivalent to the ones they used.  

MESA is a protein structure with an extracellular domain, a transmembrane domain and an intracellular domain. It can be used as a sensor by having two different protein chains with, on the extracellular region, an element to cause dimerization and, on the intracellular region, a combination of a protease on one chain and a transcription factor on the other. The transcription factor can be cleaved off by the protease when the two chains dimerize, either through a ligand or randomly. Subsequently, the transcription factor travels to the nucleus where it induces transcription of a reporter (Figure 1).

The MESA constructs that inspired OUTCASST, and therefore need to be verified in this chapter, are V2-MESA-35F-M-tTA and V2-MESA-35F-TEV. These chains have an extracellular domain which binds vascular endothelial growth factor (VEGF) with an intracellular region each. For V2-MESA-35F-M-tTA, the  intracellular region is the tetracycline-controlled transactivator (tTA) and a Tobacco Etch Virus (TEV) protease for V2-MESA-35F-TEV. pL3-TRE-LucGFP-2L  and pBI-MCS-EYFP were used as reporter plasmids, which express luciferase-GFP fusion protein and yellow fluorescent protein (YFP), respectively.  

A Cre reporter has constituently active dsRED and was used as transfection control. In our later experiments with YFP, blue fluorescent protein (BFP) was used as a transfection control to avoid spectrum overlap.

Replicating this system is of major importance to our final DNA Biosensor design because we need to verify that the same approach would work for our system as well. Finally, we would compare it to model data and to benchmark output. Unfortunately, we could not reach this stage.

We seeded HEK293T in a 24-well plate with mEF media and 1% penicillin-streptomycin. 24 h post-seeding the cells were transfected according to Table 1 in the supplement. The amounts of the plasmid in ng were the same between the wells for all different plasmids aside from the reporter plasmids and controls. Per well 180 ng of V2-MESA-35F-M-tTA; 15 ng of V2-MESA-35F-TEV and 25 ng of either pSLQ-Set1-BFP or Cre reporter was used. For reporter plasmids pL3-TRE-LucGFP-2L and pSLQ-Set1-BFP, four different amounts were used, namely 250 ng, 275 ng, 300 ng and 350 ng. Wells where the total amount of plasmid was less than 500 ng were supplemented with random  DNA up to a total of 500 ng.

The media was refreshed 24 h post transfection and 250 ng of VEGF was added per mL of media. The results were obtained by FACS (BD FacsAria Fusion machine) approximately 18 h later.

<b>Figure 2.</b> 350 ng GFP. (Left) Plot before adding VEGF. (Right) Plot after adding VEGF. [GFP] is GFP activity.

<b>Figure 3.</b> 300 ng YFP. (Left) Plot before adding VEGF. (Right) Plot after adding VEGF. [Q2] is YFP activity.  

<b>Figure 4.</b> 300 ng YFP. (Left) with protease chain. (Right) without protease chain. [Q2] is YFP activity.  

The goal of these experiments was to verify the MESA system in our own lab. We had little success in the matter. In the original article the authors managed to achieve a doubling of output when the ligand was added, while we only achieved 25% more signal at best. The output also was inconsistent. Any number of reasons can be the cause of this. The most likely explanation is the difference in amount of plasmid we added compared to the original authors. Due to the costs of transfection, less plasmid of each kind was used. Though the absolute amount of plasmid was reduced, the ratio of the TC plasmid and PC plasmid was maintained. The ratio between the TC/PC plasmids and the reporter plasmid were not maintained however. A possible cause for the lack of output might be that too little of reporter plasmid was used. This however seems unlikely, since in our latest experiments the amounts of reporter plasmid were similar to the amounts the original authors used. Rather, it could be that the amount of TC/PC plasmids used was too little compared to reporter plasmid. Another difference is the method of transfection; we used lipofectamine for transfectio n. This, coupled with the differences in plasmid amounts could have some influence on the results.

The results were inconsistent. Even between duplicates we found large differences in outcome. Possible explanations include, for example, errors during preparation of the samples and differences in the duration under trypsin. However, the latter can not explain the inconsistency between duplicates.

The controls of the MESA system, with only the target chain without the protease chain, showed a similar level of signal as the samples with both chains, as was seen in figure 4 and 5. This could imply that the MESA plasmids that were used are not functional, as it is expected that the target chain on it’s own will show minimal background signalling. Alternatively it could imply that the reporter is very leaky, showing little differences between TC, TC/PC and TC/PC with VEGF. Additionally, we used mEF media for cell culturing, which also contains fetal bovine serum. Fetal bovine serum can contain VEGF, although we would assume this amount is negligible compared to the amount of VEGF we add.  

To produce the OUTCASST system, catalytically inactive Cas9 and Cpf1 need to be expressed on the extracellular domain of the MESA construct instead of the original extracellular VEGF binding domain. The first step in this process is to make dead versions of Cas9 and Cpf1, from here on out designated as dCas9 and dCpf1 respectively, by introducing mutations. This way, the two proteins won’t be able to cut the DNA strands in separate pieces and are only able to bind the DNA. When our target DNA remains in one piece it makes co-localization of the two transmembrane proteins possible. For the OUTCASST system, we substituted dCas9 for the extracellular domain of the MESA chain with the Tobacco Etch Virus (TEV) protease and we substituted dCpf1 to the MESA chain with the tetracycline-controlled transactivator (tTA). Lastly, we wanted to test the OUTCASST system for functionality and optimize it. However, we ran into difficulties while substituting the extracellulair domains of MESA for dCas9 and dCpf1 and therefore did not manage to test the functionality of the OUTCASST system.

All results were integrated in the Excel template file provided by iGEM. In figure 1 the relative fluorescence expression per cell of the diluted cultures is plotted against the time for all devices i.e. plasmid constructs. Here, it is visible that device 1 has very deviating values relative to the other devices. In figure 2 test device 1 is omitted to get a better view at the other devices. Here it is visible that test devices 3, 5 and 6 follow closer to the negative control, while test devices 2 and 4 follow closer to the positive control.

Since safety is a very important aspect in synthetic biology, we collaborated with the RIVM National Institute for Public Health and Environment of the Netherlands. They encouraged us to think about safety before we act. We came to the conclusion that safety has different meanings for different stakeholders. Here we describe the most important points for these stakeholders. Besides safety, we’ve also thought about the societal impact of our tool and the possible ethical issues involved. The information we gathered was summarized in an infographic, which we used to talk to the general public about synthetic biology and safety at an event organized by the RIVM.

Since we are working with GMO’s we have to follow several safety regulations to ensure the safety of our team members whilst working on OUTCASST. Our team’s lab management worked closely with drs. Fraukje Bitter-van Asma, the Occupational Health and Safety & Environment Expert of Utrecht University. She helped us determine which safety forms and permits needed to be filled out, filed and requested as well as which university guidelines and emergency measures were in place in case of calamities and to prevent health risks to both team members working on the lab and the environment.

Since the HEK293T cells are too fragile to use in the device, we opt to use air-dried cells from the anhydrobiotic insect, Polypedilum vanderplanki. To make the use of these cells as safe as possible, the design of our tool is going to be a closed system, wherein everything is present and only the blood sample will be applied. There will also be several mechanisms and kill-switches incorporated in the detecting cells. This way, the cells are physically separated from both the user and patients and this minimizes the chance of survival of these cells outside of the system.  

There are 2 main ethical considerations of usage of the OUTCASST system:

<b>3.1 Issues arising from use of GMO</b><br><br>

However, the OUTCASST toolkit consists of genetically modified cells stored in a closed box environment. This closed box environment contains the cells in an isolated environment, minimizing the risk of our genetically modified cells escaping into the environment. When disposed of correctly, in line with protocols for disposal of genetically modified material, then there should be no chance of contamination or escape of our gene into the natural environment.

<b>3.2 Issues arising from DNA detection</b><br><br>

Altogether, we gathered information from many different fields and perspectives. All of these views help us to optimize our design to use OUTCASST as a tool for diagnosing Chagas disease. Since we want our tool to be easy to use in the field and rural areas, there are different aspects that should be taken into account. Robustness and resistance of the toolkit to temperature fluctuations and humidity are chief among these. In addition, it is important not to rely on material and storage containers such as fridges or freezers (Marit de Wit, Doctors without borders).  

There are also some things in the OUTCASST toolkit that need to be changed in comparison to the experimental approach in order to prepare the system for diagnosing Chagas disease. One of these things is the use of HEK293T cells, which need a very stable environment to stay alive. In the eventual tool, we will need to use cells that are more robust to environment fluctuations yet still cannot survive outside of the device (Patrick van Zon and Pieter-Jaap Krijtenburg, University Medical Center of Utrecht: genome diagnostics.) We also used a fluorescence signal as output in the experiments, which requires a fluorescence microscope to analyse the test results. To avoid the need of these and other equipment, we would ideally use an output signal in the form of visible light or, more promisingly, a change of color that is visible to the naked eye. Another thing we should keep in mind is the time it takes to get the results from our test device (Marit de Wit, Doctors without Borders).  

As was stated earlier, a lot of variables need to be kept constant to keep the HEK293T cells alive. Because of this, it is not feasible to use these cells in our design. Instead, we opt to use air-dried cells from the anhydrobiotic insect, Polypelidum vanderplanki, which can be stored at room temperature for 251 days and can restart proliferating again after rehydration (Watanabe K, Imanishi S, Akiduki G, Cornette R, Okuda T. Air-dried cells from the anhydrobiotic insect, Polypedilum vanderplanki, can survive long term preservation at room temperature and retain proliferation potential after rehydration. Cryobiology. 2016 Aug 31;73(1):93-8). This way the shelf life of our tool can also be prolonged. To prevent the risk of our GMO getting out in the environment, several mechanisms and kill-switches will be incorporated in the cells, so they can only survive in our closed box system, in their resurgent state. This can be done by manipulating the metabolism, so that the cells can’t produce a crucial substance for survival, which will be added in the toolkit medium. In case the cells get out of the toolkit, they will die because of the absence of the crucial substance.

These were the preparation steps and the real diagnosis can start now. First off, a blood sample has to be taken from a patient that might be infected with Chagas disease. To prevent the blood from clotting, heparin or EDTA can be added to the sample. The blood sample can then be introduced to the tool, after which the device needs to be sealed. To get access to the parasite DNA, all cells need to be lysed, including the red blood cells. This is done with a lysis buffer, a hypotonic solution.  

The first thing we still need to consider is the blood sample size needed to perform the test. From the patiënts aspect it would be best to take as little as possible. A smaller blood sample would also mean that the device can be made smaller, which in turn also makes the production costs for one test cheaper. However, there needs to be enough pathogen DNA in the blood sample to make sure that the test gives the good results. It would be possible to pretreat a larger sample to concentrate it before applying, increasing the chance of correct diagnosis, but this would again require skilled professionals and materials.

We have also thought about a question that was raised at the University Medical Center at the Cancer department. The question was why we wanted to express our system on the membrane of eukaryotic cells and not just express it intracellularly in bacteria. Then a blood sample could be added and the bacteria can be heat shocked to get the pathogen DNA intracellular, activating the binding of the two proteins. In this case, there would be a loss of the amplification step, since the transcription factor is then able to activate the reporter gene without a signal or cleavage of the transcription factor. Since we don’t know what the minimum amount of blood needed is, we wanted to design it in the way we can get the most signal, which is to include the amplification step. If it would prove that this amplification step is not needed, we could also just put the proteins in the tool and use a split reporter. On the other hand, the tool would not rely on use of living cells, which would make the use of our tool a whole lot safer  

We have also heard that the tool should have a low incidence of false positive and negative results and that our device should distinguish DNA strands with one different base pair. We want to take this information into account to decide the target DNA. There are two possibilities from which we can choose. The first option would be to permit certain mutations in the target DNA, to prevent getting a false negative result in some cases. The second option would be to use a very conserved domain as target DNA and don’t allow any mismatches. From our perspective, we think the second option would be more suitable, since the specificity in our system is a very valuable aspect of the design. We have chosen to use the satellite DNA, which is present in the <i>T. cruzi</i> parasite as a 195 base pair repeat with about 100,000 copies (Aldert Bart, Academical Medical Center Amsterdam: Clinical molecular parasitologist).

Science can have an impact on the world in many ways. With our project, we are not only trying to make a difference by creating a diagnostic tool, but by reaching out to the public we hope to make science accessible for everyone. We try to achieve this by collaborating with ‘de Kennis van Nu’, a Dutch platform that brings different scientific themes to the general public in an understandable way. They aim to make science accessible to everyone, old and young, and encourage everyone to be curious and bring out the scientist in themselves! On this platform we explain the formation of Utrecht’s very first team, our design and how we are trying to solve healthcare problems. Through our whole iGEM experience, they will follow us from lab bench to Boston.

<li />Part Ba_K2351009 (sCas9)<br>

BioBrick Part Ba_K2351009 (sCas9) has been validated. This BioBrick is very special since our team successfully secreted CRISPR-associated proteins from HEK293 cells. To our knowledge, this is the first time that Cas9 has ever been expressed outside of the cell.