Throughout our project we were constantly looking to improve our design. To achieve this, we followed the four stages of the engineering design cycle:
design, build, test and learn. By implementing and reflecting on this engineering philosophy, not only were we able to improve our project as we moved forward, but we
were also able to learn from the process. Here, we outline each cycle of our project and how we moved from the design stage to the learning stage.
Design
To accomplish our goal, the detection of microplastics in water samples, we designed two different fusion proteins based on the anchor
peptide published by Rübsam et al. (2017). We chose pET45b as a vector plasmid because it could be provided in-house and because it contains a 6x His
tag for purification purposes. To be able to detect the bound anchor peptides, two different molecules were used. Our main construct is based on NanoLuc (NLuc), a
luciferase from the marine shrimp Oplophorus gracilirostris. The second construct is based on eGPF (enhanced green fluorescent protein). It is used
to simplify the proof of concept and as a control, given that an eGFP fusion protein has been successfully developed for microplastic detection by Ji et al. (2021).
The two constructs only differ in the reporter molecule.
To separate the anchor peptide and the eGFP or NLuc, respectively, a spacer sequence was designed. In their paper, Rübsam et al. used a sequence of ten alanine
residues as a spacer. We followed their lead and did the same but used different codons optimised for E. coli. Furthermore, we included a TEV cleavage site
to be able to cleave the fusion protein in anchor peptide and eGFP/NLuc later on. This has two advantages, firstly we can use the anchor peptide alone to build a
sandwich assay as described later and secondly, we would be able to use the separated eGFP or NLuc as a negative control since it should not bind to plastics.
For the sandwich binding assay, an Avi tag was included for biotinylation of the anchor peptide as described in later sections. An additional spacer of five
alanines was added between the Avi tag and the anchor peptide.
As part of our protein design we performed in silico analysis, visualising our folded fusion proteins in space. With structural prediction software, we were able to
observe our reporter molocules in relation to the anchor peptide and linker.
Build
Both constructs were developed in house. The protein sequence was inserted into the vector plasmid using restriction enzymes (SacI and HindIII) and in-fusion cloning.
Test
The ligation product was transformed into E. coli Top10 cells to select and amplify the plasmid. Several colonies were obtained. Colony PCR was performed and, after several attempts,
we were able to verify the sequence and cloning success through sequencing.
The predicted folding of our NLuc construct. In red you can see NLuc luciferase and in yellow the anchor peptide. These are joined by a linker in blue.
Learn
The cloning process presented unanticipated challenges. Following multiple attempts, with changes made to the protocol and extra controls, we were able to identify the error in our
ordered sequence insert. We took corrective action by procuring the correct sequence, which had been meticulously tailored with appropriately designed overhang sequences for
in-fusion cloning methodology. The in-fusion protocol allowed for rapid cloning with high accuracy.
Design
To express our protein we used E.coli BL21 and a lac operon induced by IPTG. The lacI gene was part of the vector backbone.
The start and stop codons were adapted according to the protein sequence and the desired tags on the plasmid. The 6xHis tag was included to
facilitate purification after protein expression and TEV cleavage respectively.
Build
Different incubation temperatures were chosen to find the optimum for protein expression of the fusion proteins. Purification methods were chosen because of their availability (batch vs. HPLC).
In order to determine the protein concentration with the photometer, the extinction coefficient was calculated using the ExPASy ProtParam web tool based on the protein sequence.
Test
The purified plasmid (after cloning) was transformed into E. coli BL21 and protein expression was induced with IPTG. To confirm protein expression, SDS-PAGE and Western blot were performed.
Sonication was used to lyse the cell, followed by batch purification or HPLC (ÄKTA) using a HisTrap column. Purification success was confirmed by SDS-PAGE and protein concentration was determined
with the photometer. TEV cleavage was performed using in-house produced enzymes and buffers followed by purification of the cleaved anchor peptide. Its success was investigated using SDS-PAGE.
Learn
Both fusion proteins were successfully expressed and purified. However, the TEV cleavage did not work as expected and due to logistical constraints we were not able to trouble shoot this in time.
Design
Specific binding with a polypropylene surface was tested by using our fusion proteins and the separated reporter molecule, eGFP and NLuc respectively, as a negative control.
The fusion proteins were designed to be cleaved at their TEV cleavage site to separate the anchor peptide from the linker connected to the reporter molecule. Without the anchor peptide,
it would not be able to bind polypropylene.
Build
To validate the binding of the anchor peptide to polypropylene we designed a binding assay to be analysed with a plate reader. Both constructs can be analysed with the plate reader due to its ability to measure luminescence,
as well as fluorescence intensity and absorbance. The TEV cleavage product, including the separated linker and eGFP/NLuc, could serve as a negative control to normalise for non-specific binding. The assay was designed to test
different concentrations of our fusion proteins to validate if it could be used for quantification purposes. The same assay was performed with polypropylene and polystyrene plates to confirm binding specificity to polypropylene.
Test
Both constructs were tested using 96 well plates made of polypropylene and polystyrene respectively. The signal was measured using the plate reader with programs measuring fluorescence intensity and luminescence to optimise detection.
The assay was repeated several times to eliminate errors. R Studio was used to analyse the data and to display the results in graphical form. In the figure above you can see the post-wash luminescence for different concentrations of our
NLuc fusion protein. The increase in luminescence following washing steps would suggest that our protein binds successfully.
Learn
Given that the TEV cleavage was not successful, our assay lacked a clear negative control. To confirm whether it is the anchor peptide that is responsible for the binding to polypropolene, further optimisation would be needed. Nevertheless,
by performing an innitial binding test with our NLuc fusion protein on polystyrene plates, we were able to confirm the specific binding of our fusion protein to polypropylene.
Design
The construct was designed to bind to polypropylene surfaces and with this we were able to design an assay for the detection of microplastics in water samples. To remove any unbound fusion
proteins we designed a sandwich assay. The constructs contain a TEV cleavage site to separate the anchor peptide from the reporter molecule connected to it and an Avi tag to enable biotinylation of the
anchor peptide to a magnetic bead.
Build
To build the sandwich assay, the anchor peptide had to be cleaved off and biotinylated using the enzyme BirA. Due to problems
from the manufacturers side we had to produce the enzyme in-house. Streptavidin coated magnetic beads were used to separate bound
protein from unbound protein. The signal could be detected using the plate reader and the fluorescence microscope.
Learn
We were not able to test the assay due to time constraints and logistic problems. Nevertheless we learnt a lot about assay design by speaking to
experts in the field.
Learn more about the experts we reached out to on our
human practices page.
Schematic overview of the sandwich assay principle. The separated and biotinylated anchor peptide (orange, biotin in green) is
immobilised on streptavidin coated magnetic beads (grey) and binds to the microplastic surface (light blue). The fusion protein binds
to the microplastic particle as well, creating the sandwich. Washing removes all unbound proteins and NLuc (dark blue) gives a detectable
luminescent signal after adding its substrate.
