Difference between revisions of "Team:Utrecht/Design"

Line 62: Line 62:
 
<p>We chose to base our biosensor on the bacterial chemotaxis system because it offers many opportunities to address the above considerations. Because it is a native bacterial system, it can be produced at low cost (requirement V) and responds rapidly to environmental input (requirement IV). </p>
 
<p>We chose to base our biosensor on the bacterial chemotaxis system because it offers many opportunities to address the above considerations. Because it is a native bacterial system, it can be produced at low cost (requirement V) and responds rapidly to environmental input (requirement IV). </p>
  
<figure style = "width:25%;float: right;">
+
<figure class = "zoom" style = "width:25%;float: right;">
 
<img width = 100% src="https://static.igem.org/mediawiki/2018/f/f7/T--Utrecht--2018-Figure1-ProjectDesign.svg" alt="BRET_Assay.png">
 
<img width = 100% src="https://static.igem.org/mediawiki/2018/f/f7/T--Utrecht--2018-Figure1-ProjectDesign.svg" alt="BRET_Assay.png">
 
<figcaption" style = "padding: 5%;">Figure 1: The Chemotaxis Pathway of <i>E. coli</i>. A) The active pathway. B) The inactive pathway.
 
<figcaption" style = "padding: 5%;">Figure 1: The Chemotaxis Pathway of <i>E. coli</i>. A) The active pathway. B) The inactive pathway.

Revision as of 11:14, 14 October 2018

Our project started with the observation that chemical contamination of surface water is an increasingly important problem. We felt that a biobased sensor would offer excellent opportunities for detection of such contaminants, and identified the bacterial chemotaxis pathway as the best point of entry to engineer a flexible and sensitive sensor. Below, we describe the design and implementation of our project.

Overview

The goal of our project is to engineer the chemotaxis pathway of the bacterium Escherichia coli to emit a visible light signal that lets us know whether specific contaminants are present in water, and at what levels.

Design

In the design phase of our project, the input from stakeholders was instrumental. In an iterative process of design and feedback, we came to a final design incorporating 5 requirements that our biosensor would have to fulfill in order to be successful.

Biosensor Design: Establishing requirements in concert with experts and stakeholders

Feedback sessions with stakeholders were instrumental in the design of our biosensor. At the start, we knew that chemical contamination of surface water is a severe threat to this valuable resource. The sources of contamination can be unexpected, like the dumping of chemical waste from illegal drug production. However, early sessions with stakeholders taught us that the most commonly encountered contamination is from common pharmaceuticals, like antidepressants. We therefore focused our efforts in this direction. In an iterative process of design and feedback, we identified the following 5 points as key requirements for our biosensor:

  1. accurate detection at different concentrations;
  2. clear and easily measurable detection signal;
  3. possibility for detecting diverse ligands;
  4. rapid signaling;
  5. low-cost system.


Chemotaxis: the optimal system to meet all biosensor requirements

We chose to base our biosensor on the bacterial chemotaxis system because it offers many opportunities to address the above considerations. Because it is a native bacterial system, it can be produced at low cost (requirement V) and responds rapidly to environmental input (requirement IV).

BRET_Assay.png Figure 1: The Chemotaxis Pathway of E. coli. A) The active pathway. B) The inactive pathway.

In their natural environment, the chemotaxis system is used by Escherichia coli bacteria to direct them towards the highest concentration of a ligand, or away from a harmful substance. This system uses a receptor that can be modified to detect a wealth of compounds (requirement III), (Shuangyu Bi et al. 2016). Furthermore, the sensitivity of this receptor can be fine tuned through methylation. Customizing the methylation states facilitates the ability to measure a broad concentration range (requirement II). Ultimately, the chemotaxis pathway controls the direction of rotation of whip-like proteins on the bacterial surface called flagella. When these flagella turn counterclockwise, they act in synergy causing the bacteria to swim coherently in one direction. But when they turn clockwise, the bacteria tumbles around, pointing it in a new direction.

When no ligand is bound to the chemotaxis receptor, it continuously activates a signal transduction pathway involving several proteins, including CheA, CheY, and CheZ (figure 1). CheA phosphorylates CheY, which subsequently translocates to the cell membrane where it binds the flagellar motor protein, thereby altering its rotational direction to the “running” state. After CheY is bound to the flagellar motor protein, it is subsequently dephosphorylated by CheZ (otherwise CheY would remain active indefinitely). Upon ligand binding, the receptor is inactivated, leading to an inactivated pathway and a rotational direction of the flagellar motor protein in the “tumbling” state. Since the binding of CheY and CheZ is linked to ligand binding of the receptor, measurement of the binding of these proteins using a Bioluminescence Resonance Energy Transfer (BRET) assay (described below) provides a quick and accurate indication of the concentration of ligand present (requirement I and II).

Chemotaxis pathway modifications to address all considerations

To create a biosensor meeting all five listed requirements, the E. coli chemotaxis pathway was customized using a three step approach. First, we set up an assay to verify the extent to which one of the most important chemotaxis receptors, the Tar receptor, could be customized. This enables accurate detection and the possibility to detect diverse ligands (requirement III). Next, a method establishing the ability to measure the activity of this pathway was implemented, using a Bioluminescence Resonance Energy Transfer (BRET)-pair. BRET provides a clear and easily measurable detection signal (requirement II and IV). The last step of the approach facilitates accurate detection at different concentrations, by mimicking receptor methylation. This facilitates detection of different ligand concentrations (requirement I).

BRET_Assay.png Figure 2: BRET-pair activity. A) When CheZ and CheY do not interact, the distance between Luciferase and eYFP is too large for Luciferase to excite eYFP. As a consequence, only Luciferase emits light. B) Upon interaction of CheZ and CheY, Luciferase and eYFP are in close proximity. eYFP is excited by photons produced by luciferase, leading to luminescence of eYFP. BRET_Assay.png Figure 3: The implementation of the BRET-pair in the Chemotaxis pathway. A) the BRET-pair is able to get in close proximity when no ligand is bound to the receptor. B) Upon binding of ligand to the receptor, the pathway is inactivated. CheY and CheZ no longer bind, therefore YFP will not be excited by Luciferase.

Implementation: Creation of the biosensor though modification of the chemotaxis pathway

The implementation phase of our project involved the modification of a bacterial chemotaxis pathway at three points, in order to meet the 5 requirements outlined in the design phase.

  1. Measuring pathway activity. The activity of the chemotaxis pathway serves as a proxy for the concentration of a contaminant in water. To measure this activity, we developed a sensor based on the principle of Bioluminescence Resonance Energy Transfer (BRET). Two proteins in the chemotaxis pathway either bind or unbind depending on the presence of ligand, and we measure this binding/unbinding as a change in the color of visible light emitted by the bacteria.
  2. Detecting diverse ligands. In order to detect a range of ligands, DeTaXion will ultimately make use of ligand binding domains from different surface receptors, and further modify the ligand specificity of these domains based on prediction using binding-affinity prediction software such as Haddock. Within the scope of this project however, we test the possibility of swapping ligand binding domains between different receptors.
  3. Fine tuning pathway activity. In order to be able to measure a broad range of contaminant concentrations, we make use of an inherent feature of the chemotaxis pathway. Methylation on four different sites of the receptor alters the activity of the pathway at a given ligand concentration. By mimicking these methylations, we can alter the sensitivity of our system to a particular ligand, broadening the dynamic range of detection.

To create a biosensor meeting all five listed requirements, the E. coli chemotaxis pathway was customized using a three step approach. First, we developed a sensor based on the principle of Bioluminescence Resonance Energy Transfer (BRET). BRET provides a clear and easily measurable detection signal (requirement II and IV). Next, we tested the effectiveness of swapping ligand binding domains with the one of the chemotaxis receptors, the Tar receptor. This enables the possibility to detect diverse ligands (requirement III). The last step of the approach facilitates accurate detection at different concentrations (requirement I). We accomplish this by mimicking receptor methylation, which alters the sensitivity of the chemotaxis pathway.

Development of a BRET sensor

To measure the concentration of ligand present, the activity of wild type and modified Tar receptors has to be measured. Therefore, we designed a Bioluminescence Resonance Energy Transfer (BRET)-based sensor, inspired by a previously made chemotaxis BRET-pair (Cui et al. 2014). BRET is a technique where photons produced by a luciferase are used to excite a fluorophore. We opted to use Renilla luciferase (RLuc) since it produces photons that can excite eYFP. In addition, the substrate for this protein, coelenterazine, is permeable to the cell which makes it perfect for E. coli based bioassays. For energy transfer to occur, the RLuc and eYFP proteins must be less than 10 nm apart. Thus, by measuring the ratio of light emitted by the two molecules, we can detect whether they are physically close, as can be seen in figure 2.

By fusing the RLuc to CheZ and eYFP to CheY, the bioluminescence can be used as a proxy to demonstrate whether the chemotaxis proteins interact (Figure 3). When no ligand is bound to the Tar receptor, CheA is active and phosphorylates CheY. Because of this, the phosphorylated eYFP::CheY can interact with CheZ::Rluc, bringing RLuc and eYFP in close proximity of each other. This can be measured as a BRET signal when the RLuc substrate is added. Upon binding of a ligand to the Tar receptor, CheA becomes inactivated, leading to a decrease of the BRET signal. Importantly, the signal can be readily measured by using an affordable bioluminescence assay.

Detection of a range of ligands

DeTaXion will ultimately make use of ligand binding domains from different surface receptors, and further modify the ligand specificity of these domains based on prediction using binding-affinity prediction software such as Haddock. Within the scope of this project however, we test the possibility of swapping ligand binding domains to change the type of ligand detected. We fused the Tar ligand binding domain (LBD) to the two-component copper sensitive (CusS) receptor. This new hybrid receptor protein detects aspartate instead of copper, and serves as a proof of concept for the ability to create hybrid receptors by exchange of LBDs from two different receptors. The functionality of the fusion protein was verified by using the downstream Cuss pathway.

When a ligand specific for the LBD is bound to the receptor, a signal is transduced to the response regulator protein (CusR) which binds to and activates a copper sensitive promoter. Normally, the CusR promoter activates genes involved in copper resistance protecting the bacterium against this potentially harmful transition metal.

In this experiment, the promoter region was coupled to a red fluorescent protein (RFP) sequence. Thus, when aspartate is bound to the receptor, CusR is activated as well as the downstream promoter and RFP is transcribed. This way, the expression of RFP serves as an indicator for a functional Tar-CusS receptor. RFP was used as indicator as it is easily measured using fluorescent light microscopy. The custom receptor was ordered at Integrated DNA Technologies (IDT) while the CusR promotor and RFP were transformed from the iGEM biobrick kit.

Fine tuning of sensor sensitivity

Jos van der Vosse (TNO) suggested quantifiable measurements. We therefore decided to implement an additional feature into our biosensor, facilitating the measurement of multiple ranges of concentrations. This method is based on the methylation state of the Tar receptor, which influences its sensitivity to ligands.

In the unmethylated state, the receptor transduces the ligand binding signal to an inactivated pathway (Figure 4A and 4B). Upon receptor methylation, the conformational change required to transduce the inactivation signal, is energetically less favorable, resulting in lower occurrence. Therefore, when the same amount of ligand is present, the level of pathway inactivation is lower than in the unmethylated receptor state. Since the Tar receptor has four acidic residues that can be methylated (Q295, E302, Q309, E491), there are in total sixteen different combinations of methylation states, each one resulting in a slightly different sensitivity (Figure 4C).

BRET_Assay.png BRET_Assay.png BRET_Assay.png Figure 4: Chemotaxis-coupled BRET activity. (A) Represents a situation in which no ligand is present and where the chemotaxis pathway is active. In this case phosphorylated CheY and CheZ interact, inducing a BRET associated luminescence. (B) Represents a situation with a high concentration ligand present, leading to an inactivated pathway. BRET associated luminescence is eliminated. (C) The same concentration of ligand is present as in B. However, since the receptors are methylated, more ligand has to be added in order for the pathway to be inactivated. This results in less YFP luminescence and relatively more luciferase luminescence, resulting in light tending more to blue then to green

Application of methylation in our biosensor

The methylation of glutamine (Q) or glutamic acid (E) residues can be mimicked by mutating these residues to alanine (A) (Krembel et al. 2015). The different possible combinations of methylation mimicking alanine mutations yields a wide range of potential sensitivities, which we predicted with our computational ODE-model. As a proof of concept, we chose three different methylation states:

  1. The wild type state, the unmethylated Tar receptor (QEQE, most sensitive);
  2. A Tar receptor with one mimicked methylation site (QEAE, average sensitivity);
  3. A Tar receptor with two mimicked methylation sites (QEAA, least sensitive).

The Tar receptor that we received from a collaboration with iGEM team Groningen was mutated using site-directed mutagenesis and expressed in E. coli strains UU1250 and VS181, which do not express any chemotaxis receptors. In addition, VS181 does not express the methyltransferases CheW and CheB, resulting in a fixed methylation state. The E. coli strains were kindly provided by Shuangyu Bi from the lab of V. Sourjik at the Max Planck Institute for Terrestrial Microbiology in Marburg.

Next, the sensitivity of E. coli for aspartate could be measured. This was achieved by addition of a range of aspartate concentrations and subsequent luminescence measurements by using the self-designed BRET-pair. Our model can be verified and corrected based on these measurements. BRET measurements were validated using a FRET-pair received from Marburg.

Cui B, Wang Y, Song Y, et al. (2014) Bioluminescence Resonance Energy Transfer System for Measuring Dynamic Protein-Protein Interactions in Bacteria. mBio. 5(3):e01050-14. DOI:10.1128/mBio.01050-14.
Krembel A, Colin R, Sourjik V (2015) Importance of Multiple Methylation Sites in Escherichia coli Chemotaxis. PLoS ONE 10(12): e0145582.
Shuangyu Bi, Abiola M. Pollard, et al. (2016) Engineering Hybrid Chemotaxis Receptors in Bacteria. ACS Synthetic Biology 5(9), 989-1001. DOI: 10.1021/acssynbio.6b00053