human-practices
title picture

Design

Table of contents

Overview Our design Plastic-binding peptides Rubber-binding domain Plasmid controls

Overview

Our team aims to address the problem of microplastics in aquatic systems. With our project we want to capture the microplastics such as polythene (PE), polystyrene (PS) and polypropylene (PP) and rubber from abrasion of tires to enter our aquatic systems. To this end we use E. coli and overproduce Curli fibers, amyloid fibers representing the main protein component of the E. coli biofilm. We further modified the fibers by fusing peptides to the Curli fibers in order to improve the specification of our system towards certain plastic types.

biofilm capturing microplastics
Figure 1: Biofilm capturing microplastics. We overexpress the Curli fibers to form an E. coli biofilm. The biofilm is able to capture microplastics. By modifying the Curli fibers with peptides specifically binding to certain plastics, we are hoping to increase the amount of microplastics captured and to improve the specificity.

Curli fibers

Biofilm is a sticky and porous layer naturally created by E. coli colonies that serves as a protection. The main protein component of the biofilm are the Curli fibers, encoded by the csg genes. Two separate operons (csgBAC and csgDEFG) produce seven proteins (CsgA, CsgB, CsgC, CsgD, CsgE, CsgF, and CsgG) to form the Curli fibers.1 The CsgA is the main subunit of the extracellular Curli fibers; the other components are chaperones and membrane proteins that help the CsgA protein to reach the extracellular space and anchor to the outer membrane (Figure 2). CsgD is the master regulator of biofilm formation.

Schematic representation of the Csg proteins
Figure 2: Schematic representation of the Csg proteins involved in forming the Curli fibers.

Our design

pC3

Previous iGEM teams worked with Curli fibers. Most of the time, CsgA is overexpressed from a plasmid, while relying on the expression of CsgB, CsgC, CsgE, CsgF, and CsgG from the genome. However, these genes are not expressed under all conditions. Therefore, typically these experiments require a special nitrogen-limiting medium (M63) in order to induce expression of the csg operons. Moreover, the Curli production is limited by the proteins produced from the genome.
Instead, we chose to express all six csg genes from a plasmid and create a strain with the csg operons deleted from the genome (for the deletion strain, pB_002 check our contribution). The plasmid is called pC3 (Figure 3) and was given to us by Roberto Avendaño Vega who had created it previously for his research.
pC3 is a modified version of the plasmid pFM_1300.3 pC3 contains the csg operons with all the parts responsible for the creation of the Curli fiber. The promoter P(BBa_J23119), before the csgB gene activates transcription of the csgB, csgA and csgC genes. The promoter P(BBa_J23100) activates transcription of the csgE, csgF, csgG genes.

pC3
Figure 3: Plasmid map of pC3. It contains two csg operons under the control of constitutive promoters, an antibiotic resistance and the origin of replication. This figure was made using biorender.com

To ensure an accurate interpretation of our findings, we aimed to incorporate our plasmid pC3 in a cell strain already created by our assistant Roberto Avendaño Vega called M037 that did not contain the csg operon in the genome. In this way we are sure that the entirety of the biofilm production is due to the plasmid. Also, we used M037 without any plasmid as a negative control to have a reference point for the biofilm absence.

Plastic-binding peptides

Overview

Leveraging the natural adhesive properties of the biofilm, we demonstrated that we could capture microplastics by overexpressing the Curli fibers. (see section Results). After talking to experts (see section Integrated human practices), we concluded that to further improve the binding efficiency and to make our system more specific towards certain plastic types, we could functionalize the Curli fibers with peptides that have high affinity for different microplastics.

In the literature we identified peptide sequences that have been shown to bind to microplastics. The high affinity between the microplastics and the peptides is due to their hydrophobicity. There are also some charged groups such as the amino and carbonyl groups that can create an additional electrostatic interaction with the microplastic.

Our design

To increase the biofilm capacity to capture microplastic, we planned to fuse these peptide sequences to the csgA gene via a short flexible linker (Figure 4).

plasmid with peptides
Figure 4: Scheme of the plasmid for production of a functionalized biofilm. The original csgA sequence is fused to the Plastic-binding peptide sequence via a short linker.

The peptides are short enough that they do not disturb the proper folding of the Curli fibers.4 The idea of this design is that the peptide stretches out of the beta-sheet structure of the Curli fibers and has the possibility to interact with the external environment (Figure 5).

Functionalizing the curli fibers with plastic-binding peptides
Figure 5: Functionalizing the Curli fibers with Plastic-binding peptides. The CsgA constituting the Curli fibers is fused to a linker and a peptide capable of binding one of the three chosen microplastics (PP,PS,PE).

We have selected three plastic types (PP, PS, PE) and selected peptides from the literature that bind to these plastics. For each plastic type we have chosen to test five different peptides with different lengths and hydrophobic values (Table1).

Table 1: List of the peptides we used to functionalize our biofilm with length and hydrophobicity. This is a subset of a library described by Woo and colleagues.5
table peptide

Initially we planned to include the peptide sequences by amplifying two fragments by PCR. One PCR used a long primer containing a sequence complementary to the csgA gene, the sequences of the linker the peptide and another region complementary to the template, followed by Gibson assembly.

primer1
Figure 6: Primer design to include a linker and peptide into the original plasmid.

This assembly strategy later proved to be ineffective (see section Engineering success), therefore we planned an alternative cloning scheme to include the peptide sequences in our plasmid. We wanted to avoid primers with excessive length, in order to limit the risk of homodimer formation We first created a linear fragment (“template”) containing the linker and the specific peptide, which we amplified with two primers carrying homology regions for the backbone. In this way, all primers were shorter than 60 nucleotides (Figure 7).

primer2
Figure 7: This primer design includes a template containing the linker and the specific peptide. For this template to be amplified we used a forward primer (PR_041) which is the same for every peptide as it does not amplify the peptide region and a reverse primer (PR_peptide_R) which is peptide specific. To amplify pC3, the backbone we use reverse primer PR_042 and forward primer PR_043.

Rubber-binding domain

Overview

During the project we learned that in addition to plastics, particles from car tires can also cause serious damage to aquatic ecosystems (see section Integrated human practices). Road drains do not have any kind of filtration mechanism and the water, instead of flowing into the water purification systems, goes directly into the lake (Figure 8). We decided to expand our project in order to tackle this problem, by functionalizing our biofilm so that it would be able to capture particles from tire wear.During the project we learned that in addition to plastics, particles from car tires can also cause serious damage to aquatic ecosystems (see section Integrated human practices). Road drains do not have any kind of filtration mechanism and the water, instead of flowing into the water purification systems, goes directly into the lake (Figure 8). We decided to expand our project in order to tackle this problem, by functionalizing our biofilm so that it would be able to capture particles from tire wear.

Where tyres go
Figure 8: Particles from car tires go to the road drains that do not have any filtration mechanism. The water also does not go to the water purification station, but ends up directly in our aquatic systems.

By searching the literature, we identified a potential candidate that could attach to the silicone elastomer of which car tires are made of. It is a protein (here called Rubber-binding protein), encoded by the ALS3 gene in Candida albicans.6 This yeast is very often associated with biofilm formation on bioprosthetic materials such as catheters. Als3 is considered the most important adhesin for this adherence. Catheters are also made of silicone elastomer. Therefore we hope this protein may also have an affinity for rubber particles from tires.

We started designing a plasmid for producing a biofilm specialized to capture rubber, but we quickly realized that we could not use the exact same system as for the Plastic-binding peptides. The size of the Rubber-binding protein is 435 aa. If we would fuse that to the CsgA protein, the resulting fusion-protein would be too big to cross the pore formed by CsgG in the cell membrane and could not be exported.4

To solve this issue, we decided to split the system in two parts that would later on reconstitute a functional biofilm thanks to the SpyTag/SpyCatcher system.

How does the SpyTag/SpyCatcher work?

The SpyTag/SpyCatcher system allows an irreversible conjugation of proteins: the peptide SpyTag reacts with the protein SpyCacther and forms an intramolecular isopeptide bond between the two proteins. We decided to leverage this key-lock mechanism by combining the SpyCatcher with the Rubber-binding protein and the SpyTag with the CsgA subunit. Upon reaction between the two parts, the Curli fibers would be decorated with the Rubber-binding domain (Figure 9).

Schematic representation of the SpyTag/SpyCatcher system
Figure 9: Schematic representation of the SpyTag/SpyCatcher system applied to our project. The SpyTag protein, located on the surface of Curli fiber, will interact with the SpyCatcher protein that is bound to the Rubber-binding protein.

Our design

We decided to synthesize the two components separately, by inserting two different plasmids into the appropriate strains.

The first plasmid is identical to our plasmid for the plastic-binding peptides, with the only difference that, instead of the linker + peptide sequence, the csgA gene is fused to the SpyTag sequence. This plasmid, which we call pFM_tag (Figure 10), was kindly provided by one of our assistants (Roberto Avendaño Vega).

pFM_tag
Figure 10: Schematic representation of the plasmide pFM_tag, containing csgA-SpyTag.

In order to assemble the second plasmid, we started from a backbone (pN2_013, kindly provided by Roberto Avendaño Vega) that contained the SpyCatcher and the lacZ gene and we ordered the sequence of the ALS3 gene through IDT. We designed primers to replace the lacZ gene with the Rubber-binding domain. We refer to this plasmid as pFM_RB.

pN2 pFM
Figure 11: Schematic representation to compare the starting plasmid PN2_013 (left) and the final product of our design pFM_RB (right)

For the expression of the Rubber-binding domain, we decided to use the strain BL21, that is specifically constructed for high-level expression of recombinant proteins.2 Our working strain PB_002 (strain lacking the csg operon) will instead express the SpyCatcher domain fused to the Curli fibers.

As the cell is not capable of exporting the Rubber-binding protein outside, we planned to culture the two cell strains separately, then to lyse the BL21 strain in order to release the SpyCatcher-Rubber-binding domain. Upon incubation with the PB_002 strain, the SpyTag/SpyCatcher reaction should irreversibly combine the Rubber-binding domain with our Curli fibers. The resulting biofilm would be functionalized and suitable for capturing rubber (Figure 12).

spy biofilm
Figure 12: Schematic representation of the production of a biofilm functionalized for capturing rubber. The two strains, PB_002 and BL21, will be grown separately, then BL21 will be lysed and its content will be put in contact with PB_002. The reaction between SpyTag and SpyCatcher will attach the Rubber-binding domain to the Curli fibers.

Control plasmids

IPTG-inducible Curli expression

We also wanted to develop a plasmid that would allow us to control the overexpression of Curli fibers using Isopropyl β-d-1-thiogalactopyranoside (IPTG). We decided to create the plasmid by modifying an already existing plasmid, called pFM_1300.3 This plasmid encodes the csg operons (containing csgABC and csgDEF genes), with promoters that respond to light and IPTG, respectively. There is also the lacI sequence, necessary for IPTG induction. We wanted to replace the light-inducible promoter with an IPTG-inducible promoter.

IPT inducible promoter
Figure 13: Schematic representation of construction of pFM_indu_ctrl, by replacing the light inducible promoter by the PLac IPTG inducible promoter.

Positive and negative controls

In order to ensure the appropriate interpretation of our results, starting from the IPTG-inducible plasmid, we wanted to create a positive and a negative control to our collection of plasmids for biofilm formation (Figure 13).
For the positive control, we designed a circuit that would constitutively drive biofilm production. We decided to replace the IPTG inducible promoters with constitutive promoters available in the iGEM registry: Bba_23100 for csgBAC and Bba_23119 for csgGEF. This plasmid would have been very similar to pC3, but would have been created by us. We planned to add the promoter sequences through Gibson assembly using primer extensions.
For the negative control, we designed a circuit that would not result in biofilm production, but that would mimic as much as possible the metabolic burden experienced by our other strains. In this case, we decided to modify the positive control plasmid and remove the csgA gene.
map of the positive and negative control

Figure 14 Map of of positive and negative control plasmids. The positive control has all six genes responsible for the formation of the biofilm, but instead of inducible promoters the plasmid contains 2 constitutive promoters. It also does not contain the LacI. The negative control is a modification where we planned to remove the csgA gene.

References