OVERVIEW

The team was determined to actualise APTASTELES and deliver a kit that could be close to the final kit. For this, we aimed to reach a proof of concept stage and determine the feasibility of our designs.

Hence we set out to ask ourselves the following questions and address them for each module of our project:

1. microRNA detection 


• • Using the FASTmiR designs modified for a given miRNA, can we detect fold change of these miRNAs?
• • Which FASTmiR design is sensitive to the miRNA, based on the split position of the miRNA?
• • What is the detectable concentration of miRNA by a functional FASTmiR? 


2. Protein and Hormone detection


• • What is the threshold concentration of biomarker required to detect a fluorescence readout?
• • Which one of the approaches devised is more effective?


3. Chip design 


• • Which design would be the best to observe the fluorescence emission from the chip?

To answer all these questions, we utilised a combination of in-silico modelling and experimental tools and provided a proof of concept.

TESTING THE FASTmiR DESIGNS FOR DIFFERENT microRNAs

We used modified Spinach aptamer as our RNA aptamer and designed it with our target miRNA complementary sequence. These designs were validated using in-silico data from various online servers and packages like NUPACK [1], RNAfold [2] and RNAcofold [3].

Based on the in-silico results, we picked all sequences based on their minimum free energy and characterised them via wet lab experiments. The following sequences are the best-identified sequences:

Figure 1. The following table shows how we prioritised designs based on their minimum free energy.

Experimental Workflow

Figure 2. Flowchart showing POC experimental workflow to test FASTmiR designs.

We used single-stranded DNA (referred to as miDNA) instead of miRNA strands for our proof of concept. This approach was taken due to its cost-effectiveness, ease of use and time constraints. Various literature also confirms the effectiveness of this approach [4,5].

We were able to test fluorescence for all designs and found promising results from FASTmiR-146a-D4 and FASTmiR-222-D2 sequences:

FASTmiR-146a-D4(BBa_K4438108)

Figure 3 and 4.Depicting the fluorescence readout of FASTmiR-146a-D4 (design 4) after 30 mins and 1 hour incubation respectively. Control 1: DFHBI dye + buffer; Control 2: DFHBI dye + buffer + FASTmiR; Exp: DFHBI dye + buffer + FASTmiR + miDNA

Figure 5. Table explaining the fold change for fluorescence readout for FASTmiR-146a-D4 after 30 mins incubation.

Figure 6. Table explaining the fluorescence readout for FASTmiR-146a-D4 after 1-hour incubation.

Explanation: After 30 minutes of incubation with 5X sensor buffer and dye, there was a dramatic shift in fluorescence intensity when the FASTmiR-146a sensor was increased in concentration. An average fold-change of 1.327 was observed after the addition of miDNA-146a throughout a range of DNA concentrations, with a maximum fold-change of 3.319 observed at a DNA concentration of 1.8 μM and a FASTmiR-146a concentration of 90 nM.

While after 1 hour of incubation with 5X sensor buffer and dye, an average fold-change of 1.41 was observed after the addition of miDNA-146a throughout a range of DNA concentrations, with a maximum fold-change of 3.311 observed at a DNA concentration of 1.8 μM and a FASTmiR-146a concentration of 90 nM.

FASTmiR-222-D2

Figure 7 and 8. Depicting fluorescence readout of FASTmiR-222-D2 after 30 mins and 1 hour incubation respectively.Control 1: DFHBI dye + buffer; Control 2: DFHBI dye + buffer + FASTmiR; Exp: DFHBI dye + buffer + FASTmiR + miDNA

Figure 10. Table explaining fluorescence readout for FASTmiR-222-D2 after 1-hour incubation.

Explanation: After 30mins of incubation with 5X sensor buffer and dye, there was a dramatic shift in fluorescence intensity when the FASTmiR-222 sensor was increased in concentration. An average fold-change of 2.69 was observed after the addition of miDNA-222 throughout a range of DNA concentrations, with a maximum fold-change of 6.048 observed at a DNA concentration of 1 μM and a FASTmiR-222 concentration of 70 nM.

After 1 hour, an average fold-change of 3.07 was observed after the addition of miDNA-222 throughout a range of DNA concentrations, with a maximum fold-change of 7.423 observed at a DNA concentration of 1 μM and a FASTmiR-222 concentration of 70 nM.

TESTING THE DUAL APTAMER SYSTEM FOR PROTEINS AND HORMONES

We developed the dual aptamer system keeping in mind the specificity and adaptability we required for different proteins and hormones, just by making minor changes to the design. So we went on a quest to design and test different strategies of trigger design for a given biomarker-specific aptamer. In addition to designs for two of the biomarkers, testosterone and C-reactive protein, we also used progesterone and quinine as they are structurally and functionally more characterised.

We built and edited our aptamer designs through simulations via  NUPACK [1], Duplexfold[6], RNAfold [2], and RNACofold [3]. Each detection system requires its unique aptamers. Hence adaptations of many available aptamers from the literature were taken and incorporated into our system. The designed aptamers for each type of biomarker which we analysed in-silico can be found on Model. 

Based on the in-silico results, we picked different types of designs and approaches and characterised them via wet lab experiments. Amongst the sequences for designed for Approach 1 (requires phi29 polymerase), the following sequences are the best characterised one: (Refer Blueprint). Using this approach, we identified Design 2 for Testosterone detection.

Figure 11. Table of Parts : testosterone detecting dual aptamer (Approach 1, Design 2)

Experimental Workflow 

Figure 12.Workflow to test our designs for Approach 1

As per the above workflow, we used various concentrations of testosterone propionate to validate our designs. 

Figure 13. Depicting fluorescence readout of T6-D1 and T6-D2 after 30 mins and 1 hour incubation respectively.Dye: DFHBI dye + buffer; T6LA1: DFHBI dye + buffer + Trigger1-T6 Aptamer + Testosterone; T6LA2: DFHBI dye + buffer +Trigger2-T6 Aptamer + Testosterone

Figure 14. Fold change of fluorescence at various testosterone concentrations.

Explanation: After 30  minutes of incubation with 5X sensor buffer and dye, there was a shift in fluorescence intensity for 1mM  testosterone compared to lower testosterone concentrations.  An maximum fold-change of 4.23 was observed in the sample with 1mM testosterone, while the range of fold-change at lower concentrations was 1.15 to 1.34 folds.

This experiment paved the way for further optimization of testosterone-sensing dual aptamer systems. Our designs proved successful in detecting the hormone, however, further experiments could help us achieve a more sensitive system. 

SELEX & MAWS 

To provide a proof of concept for the biomarkers, for which there haven’t been any aptamers characterised,we decided to raise aptamer using in-vitro and in-silico techniques for prolidase and irisin. We planned to use SELEX (Systematic Evolution of Ligands by Exponential enrichment) and MAWS-Heidelberg 2015 (Making Aptamers Without SELEX) to identify aptamers for the above-mentioned proteins. However, we couldn’t generate any aptamers due to various technical issues. We thank our collaborators for NU-Kazakhstan for their help with understanding how to use MAWS software.

TESTING OUR PRELIMINARY PROTOTYPE

Developing a prototype for the kit that takes into account the parameters required to achieve our final goal took a lot of iterations. To determine levels of the ten selected biomarkers simultaneously, in a small volume of blood serum stands as a challenge to achieve an accurate detection.

(Please refer to Hardware to know more about the iterations in kit design.)

For our preliminary prototypes, we considered the following parameters:

• • The minimum volume of sample required
• • Method of sample collection and processing
• • Time taken by FASTmiR and Dual aptamer biosensors to provide a readout
• • Reagents and enzymes needed, along with their requirement for detection at a given time

Keeping in mind the proof of concept and the implementation of the kit, we came up with a design that would be feasible to make and test in the lab. 

Figure 15. With the help of Dr Mampallil, we used a 3-D printer to make our prototype and tested the design

Experimental Workflow

We initially tested the design by adding approximately 2 mL of water to the storage chamber and observed that it stayed there. Later, on pressing the chamber the water was emptied into the reaction chamber.

Figure 16. Experimental workflow for testing the proof of concept kit design

Result

Figure 17. & 18. Before fluorescence and After fluorescence

We observed the fluorescence emission from the reaction chamber. Due to the hydrophobicity of the chambers, the dye stayed in the storage chamber. When we pressed the chamber, it was directed to the reaction chamber. These results indicate that one part of the chip design we plan to implement is functioning as expected.

Conclusions

We believe this is closer to the actual kit we want to implement. This also helped us to understand the limitation of our design to hold the exact amount of content and we plan to make improvements to overcome this. These results align with the proof of concept and help us answer questions we asked ourselves in the beginning.

REFERENCES

1. J. N. Zadeh, C. D. Steenberg, J. S. Bois, B. R. Wolfe, M. B. Pierce, A. R. Khan, R. M. Dirks, N. A. Pierce. NUPACK: analysis and design of nucleic acid systems. J Comput Chem, 32:170–173, 2011.
2. Bellaousov, S., Reuter, J. S., Seetin, M. G., & Mathews, D. H. (2013). RNAstructure: web servers for RNA secondary structure prediction and analysis. Nucleic acids research, 41(W1), W471-W474.
3. Denman, R. B. (1993). Using RNAFOLD to predict the activity of small catalytic RNAs. Biotechniques, 15(6), 1090-1095.
4. Soni, R., Sharma, D., Krishna, A. M., Sathiri, J., & Sharma, A. (2019). A highly efficient Baby Spinach-based minimal modified sensor (BSMS) for nucleic acid analysis. Organic & Biomolecular Chemistry, 17(30), 7222-722
5. Lesnik, E. A., & Freier, S. M. (1995). Relative thermodynamic stability of DNA, RNA, and DNA: RNA hybrid duplexes: relationship with base composition and structure. Biochemistry, 34(34), 10807-10815.
6. Reuter, J. S., & Mathews, D. H. (2010). RNAstructure: software for RNA secondary structure prediction and analysis. BMC Bioinformatics. 11,129.