Engineering

Amplification of target ctDNA

These design cycles are dedicated to the development of target ctDNA amplification from the choice of method, design of parts to experimenting and troubleshooting our chosen method: recombinase polymerase amplification (RPA).

---

Cycle 1

Design 1

Choosing method of amplification

To address the low concentration of ctDNA in the blood, amplification of target region of the mutation (exon 19 of the EGFR gene on the 7th chromosome) must be performed before the detection mechanism is applied.

However, to limit the use of infrastructure, keeping the test a point of care test, the test must be done in an isothermal condition, therefore, not requiring expensive thermocyclers. Additionally, as non-small cell lung cancer, like nearly all other types of cancers, can stem from multiple variations of mutations, the amplification method should be multiplexable, allowing for simultaneous amplification of various targets.

From these conditions, some plausible amplification methods to be used were identified with advantages and disadvantages displayed on the table:

Amplification method	Multiplexable	Isothermal amplification
PCR	✓	x
Loop mediated isothermal amplification (LAMP)	✓*	✓
Recombinase polymerase amplification (RPA)	✓	✓
Rolling circle amplification (RCA)	✓	✓

LAMP is only multiplexable in a specifically designed assay (multiplexable LAMP) with more complicated designs and require separate reactions

Due to LAMP’s complication in making it multiplexable, RCA and RPA seem to be promising methods. However, RCA can only detect single stranded nucleic acids; therefore, it is more suitable for RNA detection rather than dsDNA structure of ctDNA. As a result, RPA was chosen as the method of amplification for further investigations.

---

Target sequences

Target sequences was designed using sequences adapted from the Genome Data Viewer from National Library of Medicine. From these sequences, two analyte sequences were designed: wildtype (BBa_25FC1YHP) and mutation (BBa_25OK44KF) variation. The mutation variation was adapted from Xu and Xing, 2019 (doi:10.1039/C9RA06758B).

---

Primer designs

For RPA, primers must be designed to detect the presence of the target ctDNA. As mutation being focused is a 23 bp deletion mutation on exon 19 of the EGFR gene, primers detecting the presence of the mutation is designed to overlay the deleted region of the exon, aiming that the 3’ region will not be able to bind to the wildtype sequence and thus cannot initiate the amplification process.

For RPA, primers were designed to be 30–35 bp long allowing recombinase proteins to bind to the primer and carry out the double strand invasion. Other criteria are as follows:

• Primer design 3’ start with at least one GC pair → increase the yield
• The Tm of 3’ end that goes over the mutation sequence has to be lower than 37 degrees less than 10 bp (to prevent 3’ binding to the wildtype sequence)
• Aim at around 60 degrees melting temperature
• Primers cannot overlap or have potential overlapping sequence that could lead to primer dimers
• Secondary structure does not show likely self-binding

With these conditions, 4 forward primers were designed with varying 3’ length that goes over the deleted sequences: 4bp, 5bp, 6bp, and 8bp. Additionally, a reverse primer (BBa_250CQDQM) was designed downstream of the mutation sequence. For positive control, a primer that is specific for the wildtype sequence was also designed with the primer mostly located within the deleted sequence.

The sequences of the designed primers are as follows:

• Wildtype forward primer: ATTCCCGTCGCTATCAAGGAATTAAGAGAAGC
• Reverse primer: GATGTGGAGATGAGCAGGGTCTAGAGCAGAG
• 4bp overlap forward primer: GGTGAGAAAGTTAAAATTCCCGTCGCTATCAACGAA
• 5bp overlap forward primer: GAGAAAGTTAAAATTCCCGTCGCTATCAACGAAA
• 6bp overlap forward primer: GAAAGTTAAAATTCCCGTCGCTATCAACGAAAG
• 8bp overlap forward primer: GTTAAAATTCCCGTCGCTATCAACGAAAGCC

Each forward primer was also checked for possible primer dimer with reverse primer as well as off-target binding on the target sequence. With these simulations, the 5bp overlap forward primer (also registered as BBa_25ZSP4AQ on the iGEM registry) shows highest specificity.

---

Build 1

Primers and analyte sequences were synthesized by IDT as an oligo and gBlock respectively. Primers were then resuspended in DNase free water at 37˚C and shaking at 300 rpm for 5 minutes to create a 100 µM stock solution. Analyte sequences were resuspended in DNase free water at 50˚C and shaking at 300 rpm for 30 minutes to create a 10 ng/µL stock solution.

Proteins and buffer for RPA was ordered as a Lyo-ready Enzymes and Reagents for Recombinase Polymerase Amplification (RPA) from ThermoFisher Scientific.

---

Test 1

RPA protocol (described in the Protocol section) was run with 10ng of mutation analyte sequence, all designed RPA mutation specific forward primer and reverse primer. Samples were then analysed using gel electrophoresis as shown below:

All primers show successful amplification of the mutation sequence with 5bp primer showing the highest level of amplification due to the highest intensity band. However, there are bands at 500 bp which was not expected (bands around 300 bp is expected to be the input analyte). Additionally, 6 bp overlap primer shows less specificity than other primers, showing multiple bands.

---

Learn 1

All forward primers designed for detection of mutation sequence were successful in amplifying the mutation analyte, using RPA. However, there could be off target binding resulting in other amplicon as shown by the 500 bp band.

The next step is to verify whether the amplicon at 300 bp is truly the input analyte by decreasing the input amount. Additionally, the specificity against wildtype primer should be assessed by reacting the mutation primers with wildtype inputs.

---

Cycle 2

Design 2

To check for specificity of the mutation primers to the mutation input, each forward primer and reverse primer were reacted with wildtype analyte, mutation analyte, and water as a negative control.

Additionally, to test the limit of detection and confirm the 300 bp band being the analyte, the input was designed to be 1ng per reaction.

---

Build 2

1 ng of each analyte was prepared from the stock solution and reacted using the same RPA protocol described in the protocol section.

---

Test 2

RPA was used for 9 reactions with primers and input. The samples were run in the gel with the following order:

1. 1kb plus ladder
2. 4bp Overlap Primer + wildtype input
3. 4bp Overlap Primer + mutation input
4. 4bp Overlap Primer + no input
5. 5bp Overlap Primer + wildtype input
6. 5bp Overlap Primer + mutation input
7. 5bp Overlap Primer + no input
8. 8bp Overlap Primer + wildtype input
9. 8bp Overlap Primer + mutation input
10. 8bp Overlap Primer + no input

As seen in the gel, the expected size amplicon has the highest intensity in the mutation input reaction. The 300 bp band is no longer present confirming the fact that it was the input material. However, the 500 bp band is still present in all 3 reactions with the mutation input.

Wildtype and negative control shows a positive result. For wildtype, the result indicates the low specificity of the primer and RPA process in general. However, due to the positive result in the negative control, no real conclusion could be drawn, as it could be results of contamination.

The smeared faint bands in all of the reactions could be a result of protein-DNA complexes which results in a varying band results despite the single amplicon.

---

Learn 2

From this testing, it can be noticed that the primers and RPA in general might have a low specificity to the mutation. Additionally, it can be seen that RPA is prone to contamination. However, the bands are not clear due to possible protein-DNA complexes that can form after RPA which is directly loaded into the gel. Therefore, purification is needed before full conclusion can be drawn.

---

Cycle 3

Design 3

To get a conclusion on the specificity of the primers and RPA process, purification of amplicons from proteins involved needs to be carried out. No specific purification kit for RPA has been developed; however, the use of PCR purification kit should be sufficient to create a solution of concentrated amplicons, due to the processes’ similar nature of amplicon and proteins.

Additionally, the incubation temperature was designed to be increased to 45˚C as well as applying shaking at 300 rpm with the aim to increase specificity of the primer.

---

Build 3

Purelink PCR purification kit was ordered from ThermoFisher Scientific to be used with in purification process of the amplicons.

---

Test 3

RPA protocol was run for 5bp Overlap Forward Primer as it shows the best specificity in dry lab and first design cycle. Additionally, varying input mass was used to determine the effect of input material on the reaction.

Purelink PCR Purification kit was then further used to purify the product as described in the protocol section.

The order of the sample ran in the gel are as follows:

1. 100 – 300 bp DNA ladder
2. RPA sample with 0 ng of mutated DNA template input
3. RPA sample with 0.5 ng of mutated DNA template input
4. RPA sample with 1 ng of mutated DNA template input
5. RPA sample with 5 ng of mutated DNA template input
6. DNA input (5bp Overlap Forward Primer, Reverse primer, mutated DNA template input)
7. 100 ng of wildtype DNA template
8. 100 ng of mutated DNA template

The use of PCR purification kit clearly improves the clarity of gel bands with bands being more sharp and smears being absent. However, the negative control still shows a positive result which could be an indicator of a primer dimer. Additionally, 500 bp band is still present.

---

Learn 3

PCR purification kit seems to be essential in testing RPA product in gel electrophoresis. However, due to the presence of the positive result in the negative control, primer dimer could exist. To truly verify the design of the primers, they could be run in PCR, if specificity could not be reached in PCR, other processes will need to be implemented to improve the specificity of the diagnosis.

---

Cycle 4

Design 4

To run the PCR, melting temperature was obtained and appropriate cycling temperature was decided based on the parameter.

Additionally, to reduce the risk of contamination, both mutation and wild type input template was amplified using forward primer (BBa_25WP03JR) and reverse primer (BBa_25RZ50SR) specifically designed for PCR. These primers are situated further upstream and downstream than the forward and reverse primers for RPA and have shorter length of around 20 bp.

---

Build 4

PCR primers were ordered and synthesized as oligos from IDT, and Taq polymerase from ThermoFisher scientific was used in PCR reactions.

---

Test 4

PCR was run for all primers with mutated, wildtype inputs, and negative control using water instead of input.

The order run on the gel from left to right is always the same primer pair with wildtype template, mutation template then the negative control in groups. Then the variation between groups from left to right is as follows with the annealing temperature specified:

1. Wildtype forward primer + reverse primer (72°C)
2. 4bp overlap primer + reverse primer (72°C)
3. 5bp overlap primer + reverse primer (71°C)
4. 6bp overlap primer + reverse primer (70°C)
5. 8bp overlap primer + reverse primer (72°C)
6. Forward template primer + reverse template primer (61°C)

From the gel diagram, it can be seen that the primers designed for mutations all show one clear band after amplification by PCR with mutated input. However, the primers were not able to reach desired specificity even with PCR: only wildtype forward primer had the desired specificity to the wildtype input. With PCR, the negative controls show negative results; therefore, the positive results shown before could be due to the sensitivity of RPA to contamination or possible off-target binding that could occur at lower temperature.

To further verify the design of the primers and to ensure the purity of the template inputs PCR amplified input templates and the 5bp PCR amplified samples were run at high input quantity through gel electrophoresis then purified using the Purelink Quick Gel Extraction Kit. The extracted samples were then sent for Sanger sequencing to verify the output material.

---

Learn 4

From this design cycle, it was concluded that, even with higher temperature of PCR, none of the primers could reach the desired specificity. However, the one clean band of amplicons in the mutation and results from Sanger sequencing matching the region of interest mean that the mutation primers are successful at amplifying the region.

As a result, the further processing of the amplicon to improve specificity should be implemented before the readout process to avoid false positives due to off-target amplification.

---

Readout

Design 1

Choosing readout method

After amplification, the amplicon has to go through reactions or process to create a readout indicating positive or negative presence of mutated ctDNA. Some readout methods were identified that is compatible in sensing RPA-amplified DNA regions.

Aptamer detection mechanism such as split-lettuce detection mechanism, broccoli and spinach aptamers CRISPR Cas12 in combination with molecular beacons Lateral Flow assay (LFA)

However, to limit cost and infrastructure, the need of extra large equipment apparatus such as a fluorescence detector is not ideal. Additionally, to streamline the workflow of the diagnostic kit, the aim is to create a one-pot assay; therefore, all the components of the readout reaction should be compatible with those of RPA. From these conditions, a table showing how each process meet the criteria are shown.

Readout process	No need for extra apparatus	Compatible with RPA
DNA aptamer detection	x	✓
Cas12a + molecular beacon	x	x
Lateral Flow Assay	✓	✓

Due to it meeting the most criteria, LFA was chosen as the readout method for further investigation.

---

Designing LFA mechanism

To implement LFA in the workflow, the primers are designed to have a tagged 5’ end with biomolecules namely FAM for the forward primers and biotin for the reverse primer.

The test strip was then designed to contain immobilized anti-biotin antibody while loading solution that RPA product will be reacted with will contain anti-FAM antibody with heavy chain tagged with gold nanoparticles to create a colorimetric readout.

When the amplicon is present, a complete amplicon with biotin and FAM tagged 5’ ends will be synthesized allowing the amplicon to bind to immobilized anti-biotin antibody and gold-tagged anti-FAM antibody creating a band on the lateral flow strip, showing positive results.

Unfortunately, due to time constraint, the readout process could not be built or tested.

---

Amplicon Processing

Design 1

Due to the inability to achieve desired specificity in RPA, the RPA product will have to go through processing to eliminate false positive result in the LFA. Two processes were identified including the use of a d-spacer probe and 5’ blocker on the forward primer in combination with endonuclease IV and the use of CRISPR-Cas9 protein.

After reviewing existing research, the use of CRISPR-Cas9 seems to be more feasible and well-researched, showing high specificity. Additionally, both target analyte have feasible PAM sites for CRISPR-Cas9 to bind and initiate its endonucleatic activity.

The use of LFA probe still suffers from the long sequence required in binding to recombinase protein, resulting in higher melting temperature, thus more likely off-target binding. Additionally, one bp mismatch could happen in off-target regions which would result in cleavage of a 5’ blocker and false positive results.

To implement in the workflow, the CRISPR-Cas9 can be designed to have sgRNA that is specific to deleted region of the wildtype sequence and incubated in the blood sample before the amplification and initiate its activity. This will prevent the successful amplification of wildtype samples as well as cleave any amplicons that were successfully amplified during RPA dissociating the molecular tags and resulting in a negative readout.

Due to time constraint, this process unfortunately could not be built or tested.