Design

Biomarkers for Breast Cancer

A variety of biomarkers are associated with breast cancer, such as CA 15.3, MUCIN 1, and CA 27.29. These biomarkers are present in blood and saliva. Although no protein has yet been recommended as a satisfactory biomarker for the early detection of breast cancer, the three proteins mentioned above are being proposed as potential candidates. Therefore, we have sought to construct a proof-of-concept using these three biomarkers, in the hopes that future advances in the early detection of breast cancer may use the same technology and techniques as we do here. Furthermore, the proteins shown here have established reference ranges in literature, which makes them promising targets for future breast cancer diagnosis.

Aptamers

Aptamers are single-stranded oligonucleotides that can bind tightly and specifically to almost any molecule. Also known as synthetic antibodies, these short strands of DNA hold great promise as alternatives for traditional antibodies, and have the potential to be used in a variety of clinical settings.

At the beginning of this year, we settled on using a homogeneous immunoassay to fulfill our goal of a simple kit that can be easily administered in a clinical setting, offering a suitable alternative to costly and uncomfortable mammograms. Initially, we were intent on adapting lateral flow assays in pursuit of this goal; however, technical challenges and the relative sensitivity required to measure small fluctuations in biomarker concentration indicated that this was not the best course of action. Therefore, we instead turned to highly sensitive homogeneous immunoassays, such as the AlphaLISA test from Perkin-Elmer for inspiration.

We settled on using a homogeneous immunoassay due to the fact that such an assay does not require the time-intensive washing steps required by a traditional heterogeneous immunoassay, such as ELISA or DELFIA. Considering that this test kit is meant to be as convenient as possible, removing the long periods of time needed for incubation and washing greatly streamlines the procedure and makes it realistic for this to be administered at home or in a clinical setting. Furthermore, we decided to use aptamers instead of traditional antibodies due to their superior stability and ease of synthesis, thus allowing this proof-of-concept to be more feasibly implemented in the future. In addition, aptamers exhibit complicated structural changes upon binding with the substrate (in this case, the biomarker in question), which we were further able to exploit to create a light-switching fluorescent probe that fluoresce when bound to the biomarker of interest. One can then determine the original concentration of the biomarker by referencing a standard curve, similar to how one would quantify the data of an ELISA or Bradford assay. Furthermore, this form of liquid biopsy has been shown through experiments to offer satisfactory measurements and data.

For more information, please see our Human Practices page.

Selection of Aptamers from Literature

We reviewed the papers of Ferreira et al and Agnihotri et al to find aptamers and their binding affinities for Mucin 1 and CA 15.3 respectively. Both papers indicated that said aptamers were highly selective for the biomarkers of interest, and the sequences were relatively short and easy to synthesize de novo. However, there currently seems to be no published research regarding synthetic antibodies for CA 27.29. Therefore, we selected Mucin 1 and CA 15.3, as well as their corresponding aptamers, for this proof of concept.

Design of Aptamer Probe

For the design of the probe, we consulted with Dr. Kazunori Ikebukuro of Tokyo University of Agriculture and Technology and made extensive use of the aptamer modeling software UNAFold to design our probes.

Our probes were based on the FRET principle and fluorescence quenching using donor-acceptor pairs and heavily inspired by similar research by Yang et al and Wu et al, who designed FRET-based aptasensors for the detection of PDGF and Helicobacter pylori respectively. In both instances, the fluorophore-labeled aptamer resulted in detection that was highly specific and did not require a long incubation period, unlike ELISA or other immunoassays of similar specificity and sensitivity. And in both groups, fluorescence upon binding with the substrate was achieved by a conformational change of the aptamer, which changed the distance between the fluorophore and the quencher attached to the ends of the probe and resulted in a weakening of the FRET effect and increased fluorescence. The final fluorescence was then quantified using a well-plate reader under excitation and compared to a standard curve derived from standard solutions of the biomarker/cell of interest.

For our project, we based the design of the probe upon this concept, and designed two different probes for the detection of Mucin 1 and CA 15.3. Using UNAFold, we were able to predict the conformational change of the aptamers upon binding with our biomarkers, and were able to select one aptamer for each protein that changed in such a way as to significantly weaken the FRET effect between the donor-acceptor pair attached to the ends upon binding, but not when suspended unbound.

We envision this project to be used in a clinical setting, where multiple samples may be tested at once (and thus lowering the cost).

For a more detailed discussion, please refer to the Modeling page.

Principles of FRET (Forster Resonance Energy Transfer)

For our project this year, we decided to employ the use of a light-switching aptamer probe, which will change shape and fluoresce under electromagnetic excitation when bound to the biomarker of interest. To that end, we have decided to make use of the Förster Resonance Energy Transfer principle, which allows for non-radiative energy transfers between two molecules, commonly known collectively as a donor-acceptor pair. Normally, FRET is often used in elucidating the structure of cellular components due to the effect’s extreme sensitivity to distance; beyond an angstrom distance of 100 Å, the interactions between the donor and acceptor greatly decrease. Therefore, monitoring FRET interactions is a useful way to gauge the distance between said donor and acceptor.

Donor-Acceptor Pairs

FRET is often used in the context of elucidating protein structure. This is done through the use of what are known as donor-acceptor pairs, which are molecules which exhibit FRET when within close proximity to each other. A subset of this general phenomenon involves a fluorophore and a quencher (instead of two fluorophores), whereby FRET is shown by the decrease in fluorescence by the fluorophore when in close proximity to the quencher, instead of the emission of light from the fluorophore that is not being excited. In our case, we are using a quencher and a fluorophore, which means that FRET interactions will decrease fluorescence of the fluorophore, while decreased FRET interactions will increase the fluorescence of the fluorophore. Below, we will outline the specific design for each probe.

Test

In order to test our aptamer probes, we first resuspended our aptamer probes into their buffer solutions. As an initial test, we resuspended the Mucin 1 S2.2 ssDNA aptasensor, and did not resuspend the CA15.3 Clone 2 and 4 dsDNA aptasensor.

We then performed our protocol to verify the FRET interactions between our aptamer probes and our Mucin 1 biomarker. Without the presence of the Mucin 1 biomarker, the aptamer probes should not fluoresce and we should get fluorescence values close to that of pure buffer solutions when we put it under the 96 wellplate reader. This is because the aptamer is still in its folded shape, and only opens up when it binds to the Mucin 1 biomarker. This causes the FRET phenomenon where the excited Cy5 dyes transfer their energy to the Black hole quenchers because they are close together. With the presence of the Mucin 1 biomarker, the aptamer probes should fluoresce, and we should be able to get a fluorescence reading under the 96 wellplate reader because the ends of the aptamer probes are too far for the FRET phenomenon to occur.

FRET Troubleshooting

Upon conducting our FRET experiment using the aptamer probes that came initially, we found that they did not fluoresce properly. Firstly, they were not fluorescing at the values we gathered from literature review, yet, that was still acceptable as part of our experimentation. However, the aptamer probes were fluorescing regardless of the presence of the Mucin 1 biomarker. This meant that the aptamer probes were:

 A. not manufactured properly so that there was no black hole quencher, leading to the absence of the FRET phenomenon
 B. were not folded properly so that the attached black hole quencher could not quench the energy emitted by the dye
 C. not binding to the biomarker, and the fluorescence value should be even higher when the aptamer probes are bound to the biomarker.

Upon the second day of experiments with newly resuspended aptamer probes, the aptamer probes did not fluoresce properly again. We omitted possibility B as we followed proper resuspension protocols, and hypothesized that it was possibility A. Thus, we reordered the aptamer probes, making sure that we had S2.2 aptamer probes with Cy5 dye on one end and BHQ1 on the other end. They arrived approximately a month later, and we were able to conduct the FRET experiment again.

In our second round of FRET experiments, our aptamer probes fluoresced properly. Our advice to iGEM teams working with aptamer probes would be to verify their FRET interactions as soon as they arrive, as they take time to reorder and arrive.



For further information on our mathematical analysis of our FRET experiments, please view the Math Model page.

Analysis of FRET results

The Hill’s Equation is a useful equation to determine the amount of protein (biomarker) bound to the ligand (aptamer). Since it is theoretically unlikely that all of the biomarkers will bind to the aptamers, we will need a method to calculate the concentration of the biomarker that is bound to the aptamer. Only using the standard curve will result in an underestimation of the concentration of the biomarker. Hill’s Equation helps us determine the original concentration of biomarker by relating the biomarker concentration with the aptamer concentration and proportions of the pattern-biomarker complex (the dissociation constant). Although the dissociation constant was to be obtained from the Scatchard Plot, test results were inconclusive. Therefore, the Kd value was derived from 3D modeling through UNAFold. The Hill’s Equation is shown above. [PLn] is the concentration of protein (biomarker) bound to ligand (aptamer), [P0] is the total concentration of biomarkers, [L] is the total ligand (aptamer) concentration, and Kd is the apparent dissociation constant. The dissociation constant (Kd) describes the ability of a compound to dissociate or break down to its constituent components. In our case, it describes the ability of the biomarker to break away from the aptamer.

We were able to get a total of 12 total data points with calibration on 475nm excitation peak and 580—640 emission peak. We then were able to plot the line of best fit and find the equation of the standard curve. Above is the standard curve obtained from the serial dilution. The equation for the line of best fit is shown above the graph and the vertical error bars are depicted. To determine the bound concentration of the biomarker, the equation of the standard curve was initially rearranged in terms of the fluorescence value and then plugged into the Hill’s Equation. The Hill’s Equation was then used with the Kd found through 3D Modeling. Where F is the fluorescence value (RFU) from the standard curve, [L] is the aptamer concentration, and [PLn] is the concentration of the aptamer-biomarker complex. Using this equation, we are able to calculate the concentration of the biomarker bound to the aptamer.

Specificity Test

Furthermore, we conducted a specificity test using Bovine Serum Albumin instead of our Mucin 1 Biomarker. Our results showed that the fluorescence value did not change from pure buffer solution, meaning that the Mucin 1 S2.2 aptamer probes did not bind to the BSA. This confirms the specificity of the S2.2 aptamer probes to Mucin 1.

Shelf Life Test

Additionally, we conducted a shelf life test with the S2.2 aptamer probe, and we were able to verify that the shelf life of the aptamer probe was at least two weeks. Due to the unexpected manufacturing with our aptamer probes, we were unable to perform tests on our CA 15.3 dsDNA aptasensors. Additionally, please see our Notebook page for more details on what we did when performing our tests.

Design

In regards to the future applications of our project, we believe that this mechanism should not be further investigated until more concrete evidence is found on the levels of Mucin 1 biomarker patients and its correlation to having breast cancer. We explored various previous literature conducted on breast cancer patients, as well as literature on the detection of breast cancer through the ELISA method detecting levels of Mucin 1 biomarker. As shown in the box and whisker chart below, literature showed that the error bars of the standard mean error overlapped between breast cancer patients and healthy patients.

Thus, we determined that our method may not provide the most accurate assessment of who requires further screening due to risk of false positives—and alarming potential patients is a situation we’d like to avoid. For this reason, we propose to implement our screening method to detect diseases where any presence of biomarkers is an indication for the necessity of treatment (including hepatitis B virus). When it comes to breast cancer applications, we propose to apply this method in the future where technology for the identification of Mucin 1 biomarker concentrations in patients with breast cancer have improved.

Moreover, in the broad scope of the implications of our system, we believe that teams in the future can build off of our ssDNA and dsDNA aptamer probe designs, utilizing the donor-acceptor pairs of BHQ1 and Cy5, and BHQ2 and Cy3 dye.