Introduction

Our vision is the utilization of our system in water samples collected from the field as described in the Implementation section. However, the design of the project and the first experiments will be conducted in lab-made deltamethrin spiked water samples whose preparation is described in the Deltamethrin Sample Preparation part of the Experiments section.
In regards to the system itself, one of the most important aspects of a biosensing system is its sensitivity and selectivity to the targeted molecule. Especially, cell-free biosensing systems lack sensitivity due to the decrease of the signal transduction’s efficiency as the steps go on. Thus, our system contains three steps of signal amplification to increase the overall sensitivity and minimize the concentration of Deltamethrin in the water sample in which the pesticide is being detected.
Regarding the selectivity of the biosensor, we designed an aptamer that recognizes deltamethrin in the analyzed water sample and modified it to transduce this signal to the following factors of the biosensing system after the binding to deltamethrin. More specifically, the binding of dAPT (deltamethrin’s aptamer) with deltamethrin will cause a conformational change in the structure of dAPT, rendering it able to hybridize with the following DNA of the system called L1. This conformational change includes the release of a specific region of dAPT, region 1, which previously was hybridized with cDAPT, a small ssDNA that forms a duplex with dAPT in the absence of deltamethrin and gets excluded after dAPT-Deltamethrin binding. L1 molecule is a ssDNA molecule that can form a stem-loop structure by itself. This structure breaks after dAPT-L1 hybridization, causing the release of an L1 region, consisting of a*b*c*, that was previously involved in the stem-loop formation. With the help of this region, the L1 molecule, via the Toehold mediated strand displacement (TMSD) mechanism, forms a dsDNA after hybridization with the L2 molecule.
Considering that only one dAPT molecule needs to bind to deltamethrin for the formation of multiple L1-L2 duplexes, L1-L2 duplex formation is the first step of signal amplification. Cas12a catalyzes the next amplification step. In this system, Cas12a guided by its crRNA, will interact with the L1-L2 duplex and acquire its transnuclease activity. In this manner, it can hydrolyze the next and final DNA molecule of the system, the LINKER. LINKER is a modified ssDNA molecule that binds with a streptavidin-coated magnetic bead in its 5’ end and a Lipase enzyme in its 3’ end. The cleavage of the LINKER molecule is the second amplification step because just one Cas12a enzyme can hydrolyze multiple LINKER molecules, releasing many Lipase enzymes into the supernatant. Then, each one of these Lipase enzymes will transform their substrate into a final chromogenic product that will change the color of the supernatant. As we can understand, this is the last signal amplification step and the final step of the biosensing system.
Obviously, a biosensing system must have both the ability to activate in response to the target’s presence but also remain deactivated in response to the target’s absence. This is why we used magnetic beads in the biosensing system. After the incubation of Cas12a with the aforementioned factors of the detection system, we will concentrate all the magnetic beads into the bottom of the tube using a magnet and then, we will move the supernatant to a new tube. If deltamethrin is present, LINKER will be cleaved as described previously and lipase will be released into the supernatant. In this way, as we transfer the supernatant to the new tube we will transfer Lipase as well and as a result, the color change of the new solution will be possible after adding Lipase’s substrate. On the other hand, if deltamethrin is absent, LINKER will remain intact causing the removal of Lipase molecules from the supernatant as they are localized to the bottom of the tube with the magnetic beads. In this way, the new solution will not have Lipases, and thus retain its color, indicating the absence of deltamethrin.

Engineering Success

CYCLE 1: DNA aptamers docking for choosing the aptamer with the best scores for deltamethrin

CYCLE 1A: Aptamer selection and the structural similarities between Deltamethrin and lambda-cyhalothrin

Design

The binding of the dAPT aptamer and deltamethrin is the initial step in activating the detection system. Due to the binding, the aptamer's tertiary structure is altered and folded in a new conformation. In order to synthesize a fully selective aptamer, we initially considered using the Capture-SELEX method based on the work of Yuxia Yang et al., 2021, but the process was time-consuming and required expertise. Thus we contacted a company specialized in aptamers’ design, but the price was too high for any iGEM team to bear ($19,000). As a result, we considered whether we might use the aptamers' sequences provided for lambda-cyhalothrin binding in the work of Yuxia Yang et al., 2021. Since lambda-cyhalothrin and deltamethrin are both members of the pyrethroid pesticide family and share multiple structural characteristics, we consulted Mr. Eleftheriadis and decided to test the binding capabilities of the aforementioned aptamers with deltamethrin using the molecular docking procedure. More specifically, both of them are pyrethroid ester pesticides and they share a common ester skeleton with S-conformation in the asymmetric center but their difference lies on the substituents of the double bond. In deltamethrin, there are 2 bromo- groups, that are halogens but in lambda-cyhalothrin, there is one chloro- group, which is a halogen as well, and one trifluoromethyl group. In conclusion, with the exception of the groups of the double bond that differ between the two pesticides, their general structural conformation have many similarities.

Figure 1. Structure of deltamethrin

https://en.wikipedia.org/wiki/Deltamethrin

Figure 2. Structure of lambda-cyhalothrin

https://en.wikipedia.org/wiki/Cyhalothrin

Build

As we described previously, in the process of obtaining the three-dimensional structures of the pesticides we observed that they shared a similar structure with certain differences in their functional groups. In order to carry out the docking procedure we needed the 3D structures of both the aptamers and the pesticides (lambda-cyhalothrin and deltamethrin) as pdb files. Thus we retrieved the pesticides 3D structures from Pubchem database as .xml files and converted them to .pdb files through PyMol software. Another basic element was the determination of the three-dimensional structures of the aptamers. Therefore we reached out to the corresponding author of the paper, Yuangen Wu (Yuxia Yang et al., 2021) and requested the structures. The authors provided us with two of the thirty one 3D structures they listed in their publication, dAPT-1 and dAPT-1-39, as we named them. While observing these 3D structures in PyMol, we noticed that the structures are in fact created from RNA and not DNA molecules. After a series of meetings with professors specialized in bioinformatics and structural biology, we decided to design the structures ourselves as we had yet to elucidate the appropriate structures. Moreover, we needed to attach an extra sequence (S region, as mentioned at " Project Design" page) to the 5′ end of the existing sequences so that the results of the docking could be useful for later implementation into our system. The addition of the S region was necessary as the chances of it altering the aptamer’s structure and ameliorating its affinity with Deltamethrin were high. This way, the results of the docking would be more indicative of the actual binding capabilities of the aptamer to Deltamethrin. Initiating the design process, we decided at first to focus on the design of the 2 sequences with higher affinity to lambda-cyhalothrin listed by Yuxia Yang et al., (2021), since we expected them to also have a sufficient affinity with our target molecule. The design of the two structures from each one of the DNA sequences had to be conducted individually. The procedure consisted of the following: Every sequence used, along with the additional S sequence, had to be initially imported into the UNAFold Web Server in order to determine the appropriate 2D structure (based on those listed from Yuxia Yang et al. (2021)) and obtain the suitable FASTA file format ( .ct in our case). Afterwards, the .ct files were uploaded to the ct2dot web server environment from “RNAstructures” in order to convert the DNA structure into .txt files which contained the 2D structure in BRACKET FORM that was mandatory for the next step of the process. In the final step of the procedure the DNA sequence along with the 2D structure had to be imported into the DNA tertiary structure prediction method tool in order to obtain the desired 3D .pdb structures. With the structures in hand, we were able to proceed with the docking procedure.

Figure 3. 3D structure of l-cyhalothrin used in docking.(visualized at PyMol)

Figure 4. 3D structure of deltamethrin used in docking.(visualized at PyMol)

Test

Initiating the docking process, we imported to the MOE (Molecular Observing Environment) software the pdb files of the structure of the DNAs and pesticides. In its essence, the software mass-tests putative, sequence dependent DNA structures and attempts to attach the ligand at different possible positions so as to, ultimately, identify the five most stable receptor-ligand complexes. The algorithm employed by this software is beyond the scope of this description. Essentially, what we tested for was the interactions of deltamethrin and lambda-cyhalothrin with dAPT-1 and dAPT-1-39, respectively.

Learn

While comparing the interaction values of the two aptamers, we observed that the binding values of lambda-cyhalothrin are similar to the ones representing deltamethrin binding. The values listed below are DOCKING SCORES. These scores are affinity indexes, and more specifically represent the free energy of the complex forming while docking is taking place. More negative scores mean lower free energy of the referring complex suggesting high stability and higher affinity between receptor and ligand. More specifically, the values between dAPT-1, deltamethrin and lambda-cyhalothrin binding were -6.4055986 and -6.621511, respectively. On the other hand, the binding values of dAPT-1-39 with Deltamethrin and lambda-cyhalothrin were -5.993392 and -6.0832882, respectively. Taking the similarity of results to consideration, we concluded that the affinity of the aptamers is not limited to lambda-cyhalothrin but seems to extend to deltamethrin as well. However, the occurring values did not indicate, in our opinion, a sufficient affinity between neither dAPT-1 nor dAPT-1-39 with deltamethrin (as shown in our “experimental results'' page). Therefore, we concluded that designing DNA aptamers utilizing the remaining 29 DNA selective sequences listed by Yuxia Yang et al.(2021) and conducting a second cycle of docking using them, could potentially improve the scores and thus, help us choose aptamers with a better affinity for deltamethrin.

CYCLE 1B: Multiple aptamer sequences monitoring for deltamethrin binding

Design

In order to conduct the second cycle of the docking process, we needed to design the .pdb structures of the remaining twenty-nine aptamers.

Build

The design process for the remaining twentynine DNA sequences was conducted as mentioned before in the “ Build “ section of DNA APTAMERS DOCKING CYCLE 1A.

Figure 5. dAPT-15 2D structure (dG= -4,23 kcal mol-1)

Figure 6. dAPT-11 2D structure (dG= -6,22 kcal mol-1)

Figure 7. dAPT-11 3D(.pdb) structure used for docking

Figure 8. dAPT-15 3D(.pdb) structure used for docking.

Build

The same testing procedure was conducted as in CYCLE 1A. Every aptamer (from dAPT-2 to dAPT-30) was imported separately with each ligand and the final result was the docking score of the interaction between them.

Learn

After completing CYCLE 1B of the docking procedure, and analyzing the results, we concluded that the affinity of the remaining twenty-nine aptamer sequences extends to deltamethrin, besides l-cyhalothrin (according to the similarity of the values between the interaction of the two pesticides with the aptamers). Comparing the docking values between all the aptamers we chose two candidate aptamer sequences with the highest values, dAPT-15 and dAPT-11. Their values were the highest amongst the values of the 29 remaining aptamers and were comparable to the values by which Yuxia Yang et al., 2021 selected their aptamer for lambda-cyhalothrin detection. More specifically, the scores between dAPT-11, deltamethrin and lambda-cyhalothrin binding were -6.9090571 and -6.5811286, respectively. Similarly the values between dAPT-15, deltamethrin and lambda-cyhalothrin binding were -6.7261815 and -6.9343119, respectively. We observed that dAPT-11 bound with higher affinity to the target molecule; however, we had to validate these results in the laboratory before we could proceed to the final selection of aptamer In order to do so, we conducted a QCM-Biosensor experimental procedure which is presented in detail below.

CYCLE 2: Determining the aptamer with the highest affinity for deltamethrin through acoustic Quartz Crystal Microbalance (QCM) biosensor experiments

Design

Following the computational docking of the aptamers, a comparative analysis was conducted to assess the binding energies of each aptamer in relation to their interaction with deltamethrin (delta). Subsequently, two aptamers exhibiting the most favorable and stable binding energies were selected, signifying their potential for enhanced interaction with deltamethrin. In light of this computational data and under the guidance of Dr. Gizeli's expertise, the decision was made to proceed with experimental validation using a Quartz Crystal Microbalance (QCM) biosensor. The primary objective was to empirically ascertain whether deltamethrin effectively binds to the selected aptamers and, furthermore, to identify which of the two aptamers yielded the most robust signal. To elucidate this process further, it is imperative to provide a concise explanation of the operational principles underlying the QCM biosensor. In the analytical procedures, an alternating voltage is applied to the surface of the piezoelectric material through attached electrodes, and various physical properties of the crystal are meticulously gauged. The fundamental physical quantity under scrutiny is the frequency of oscillation. As per the foundational law of energy conservation in physics, the energy preserved within these oscillations remains constant. Consequently, the introduction of additional mass onto the oscillating piezoelectric material results in a deceleration of the oscillatory process. Conversely, the removal of this mass from the piezoelectric material prompts an acceleration of the oscillations. Mathematically, the relationship between the oscillation frequency (∆f) and the mass (∆m) affixed to the surface of the piezoelectric material, considering the material's density (ρ), shear modulus (μ), and active surface area (A), is elegantly expressed by Sauerbrey in the following equation: ∆f = -2f02∆m / A(ρμ)1/2.

Figure 9: General principle of piezoelectric biosensors. A piezoelectric sensor is depicted as the gold cylinder while the bound mass as the green cylinder.

In most cases of QCM fabrication, the electrodes made on a quartz are composed from two metals. The underlayer serves for making a good mechanical and conductive contact between the electrode itself and quartz while the upper layer should be conductive as well as chemically stable to prevent oxidation and also suitable for connecting of the QCM into an electrical circuit (soldering or mechanical contact).

Figure 10: Appearance of a QCM sensor.A 5 MHz sensor based on a 14 mm quartz circle with gold 5mm wide electrodes on the opposite sites is depicted here.

The apparatus employed in our study is denoted as a Q-Sense instrument, equipped with four distinct gold chips/sensors. This configuration allowed for the execution of distinct reactions on each individual sensor, resulting in unique elemental compositions being present on each chip during every experimental iteration.

Build

The experimental protocol proceeded as follows: Subsequent to the installation of gold sensors within the Q-SENSE instrument and the configuration of the Q- sense software, a series of diverse solutions were systematically introduced. The initial solution comprised neutravidin, chosen for its binding affinity with the gold chip surface. Following this, a PBS (Phosphate-Buffered Saline) buffer solution was applied to effectively rinse away any unbound neutravidin. Importantly, this buffer rinse step was consistently performed after the introduction of each subsequent solution. Subsequently, the aptamer, which had been procured in a biotinylated form at the 3' end, was dissolved in deionized water (dH2O) and utilized at a concentration of 5 pmoles. It is noteworthy that the biotin moieties within the aptamers were intended to bind specifically with the neutravidin molecules, thereby facilitating the positioning of the aptamers atop the neutravidin layer immobilized on the gold chip. With the exception of a control sensor (S), wherein no aptamer solution was introduced, all other sensors received the aptamer-containing solution. Following this step, a PBS buffer rinse was administered once more. The introduction of deltamethrin was executed at a concentration of 0.05 milligrams per milliliter (mg/mL) and following the complete washing of the system with buffer, a fresh deltamethrin solution was introduced, this time at a concentration of 0.5 mg/mL.

TEST

Cycle 2.1: Detection of deltamethrin binding with aptamer dAPT15

Build

Initially, it is important to note that deltamethrin exhibits considerable hydrophobicity, rendering it poorly soluble in aqueous environments. In accordance with prior research and comparative literature analysis, our approach commenced with the dissolution of deltamethrin in absolute ethanol. Subsequently, the solution was subjected to a controlled temperature of 37 οC to augment solubility, thereby enhancing both dissolution kinetics and the resultant solution's optical clarity. Another step we designed involved injecting a chip with an amount of deltamethrin immediately after the neutravidin step to confirm that there is no non-specific binding between the two.

Test

Timeline PBS buffer pH 7.4 flow rate 50μl/min 30:05 S1-S4 neutravidin 0,2 mg/mL 34:23 S1-S4 buffer 2:08:19 s1 0,05 mg/ml deltamethrin 2:12:30 s1 buffer 2:24:06 s2 5pmol dAPT15, s1 control 2:27:52 s2 buffer (stopped flow accidentally) 2:39:01 start flow again 2:57:53 s1-s2 0,05 mg/mL deltamethrin 3:01:41 Buffer 3:14:10 s1-s2 0,5mg/mL deltamethrin 3:18:18 Buffer

Learn

Using ethanol alone as a solvent for deltamethrin didn't work as well as expected. We noticed that our signals weren't satisfactory, and deltamethrin didn't fully dissolve, making the solution cloudy. So, we looked into a different approach mentioned in scientific literature. We mixed equal parts of dimethyl sulfoxide (DMSO) and pure ethanol to dissolve the deltamethrin. Furthermore, we confirmed that there is no non-specific binding between delta and neutravidin while we observed a very strong signal at the 5 pmol concentration of the aptamer.

Cycle 2.2: Detection of Deltamethrin binding with aptamer dAPT15- change of the solution

Build

After the unsuccessful dissolution of delta in pure ethanol we decided to try another alternative. Based on the literature, DMSO is also considered a solvent for this pesticide. So we dissolved the delta in a DMSO-pure ethanol solution with a ratio of 1:1. The final concentration of delta in the solution was 1mg/mL and with successive dilutions we created solutions of 0.5 and 0.05mg/mL delta. Which were re-diluted 100X and 10X respectively with PBS. In addition, a new step added was that before the introduction of delta, we tested a mixture we created which contained DMSO, pure ethanol and PBS with a ratio of 0.5:0.5:99 respectively. This was done to confirm that the signals we might get would not be due to non-specific binding of the solvents. Also, we used two different concentrations of aptamer dAPT15, 2.5pmol and 5pmol .

Test

Timeline PBS buffer pH 7.4, flow rate 50μl/min 20:22 S1-S4 Neutravidin 0.2 mg/mL 25:19 Buffer PBS 1:10:24 dAPT15 S1 Control, S2/S4 2,5pmol, S3 5pmol, 200μL each 1:14:48 Buffer PBS 2:15:19 1:100 buffer (0.5% EtOH:0.5 DMSO:99%PBS) 2:20:24 Buffer PBS 2:32:22 0,05 mg/mL (0.5% EtOH : 0.5%DMSO:99%PBS) deltamethrin 2:37:30 Buffer PBS 2:48:46 1:10 buffer (5% EtOH:5% DMSO:90%PBS) 2:54:09 Buffer PBS 3:05:51 0,5mg/mL (0.5% EtOH : 0.5%DMSO:99%PBS) deltamethrin 3:10:57 Buffer

Learn

The change of the solvents resulted in much better outcomes. Deltamethrin dissolved more effectively in DMSO-etOH (1:1), and there were no pesticide particles suspended in the solution, even without heating. Furthermore using 2 different concentrations of delta and dAPT15, we observed that at the lower concentration the sensor did not give any noticeable signal. Therefore, in subsequent experiments we continued with the highest concentration of 0.5 mg/mL of delta and 5 pmol concentration for the aptamer dAPT15.

Figure 11: General principle of piezoelectric biosensors. A piezoelectric sensor is depicted as the gold cylinder while the bound mass as the green cylinder.

Cycle 2.3: Detection of Deltamethrin binding with aptamer dAPT11

Build

Following the completion of the initial two experiments and the subsequent determination of the optimal delta solution conditions, we embarked on a parallel experimental procedure for the dAPT11 aptamer. Given our limited knowledge regarding its binding properties with deltamethrin, we opted to explore two distinct aptamer concentrations. Consequently, among the four available gold chips, one was designated as the control sensor, with no DNA introduced. Two chips were designated for the introduction of DNA at a concentration of 2.5 picomoles (pmol), while the remaining chip received the aptamer at a concentration of 5 pmol. Also, deltamethrin was introduced at two varying concentrations: 0.5 and 0.05 milligrams per milliliter (mg/mL). These solutions had been previously prepared and stored at 4°C one week prior to the experiment.

Test

The same procedure as above applied for dAPT15 was followed except that the aptamer concentrations applied to each chip were different (2.5 pmol and 5 pmol). It's important to note that the deltamethrin solutions utilized had been prepared one week prior to the experiment. All other parameters, encompassing delta solvents, delta concentrations, and the operational conditions of the machinery, remained consistent and unaltered.

Timeline PBS buffer pH 7.4, flow rate 50 μl/min 12:28 s1-s4 neutravidin 0,2 mg/mL 17:46 s1-s4 buffer PBS pH 7,4 44:05 s1: buffer / s2+s4: 2.5pmol dAPT11 / s3: 5pmol dAPT11 49:56 S1-S4 buffer PBS 1:11:45 s1-s4 buffer 1:100 (0.5% EtOH:0.5 DMSO:99%PBS) 1:16:41 s1-s4 buffer PBS 1:33:11 s1-s4 0,05 mg/mL (0.5% EtOH : 0.5%DMSO:99%PBS) deltamethrin 1:38:16 s1-s4 buffer PBS 2:00:40 s1-s4 buffer 1:10 (0.5% EtOH:0.5 DMSO:99%PBS) 2:05:33 s1-s4 buffer PBS 2:20:01 s1-s4 0.5 mg/mL (0.5% EtOH : 0.5%DMSO:99%PBS) deltamethrin 2:28:25 s1-s4 buffer PBS

Learn

From the above experiment, it was confirmed that the delta concentration that gives a signal for the dAPT11 aptamer is that of 0.5 mg/mL, while the concentration of 5 pmol seemed more suitable for the dAPT11 aptamer, similarly to the dAPT15 aptamer. Finally, some small repeating signals were observed on sensors which piqued our interest.

Cycle 2.4: Comparison of deltamethrin binding to aptamers dAPT15 and dAPT11 and the "delta fridge" test

Build

The occurrence of sporadic, yet discernible signals within certain sensors with the emergence of atypical response curves in others, prompted an investigation into the underlying causative factors. This perturbation was notably correlated with the utilization of pre-prepared delta solutions that had been securely stored for one week at a refrigerated temperature of 4°C. To comprehensively address this phenomenon, we instituted an experimental strategy that encompassed the inclusion of two distinct control groups. One control group was subjected to a freshly prepared deltamethrin solution, while the second control group received the refrigerated solution. Moreover, as part of the experimental design, 5 picomoles (pmol) of dAPT11 and 5 pmol of dAPT15 were introduced alongside our sensors (representing 2 out of the total 4 sensors) to enable comparative analysis of the outcomes.

Test

Timeline 1 (exp1) PBS buffer pH 7.4, flow rate 50 μl/min 11:38 Neutravidin 0,2 mg/mL 16:19 Buffer PBS 49:30 dAPT11: 5 pmol/ CT15 5pmol/ control fridge/ control 54:11 Buffer PBS 1:27:25 DMSO:EtOH (1:1) 10X PBS 1:31:57 Buffer PBS 1:52:26 Delta 0,5 mg/mL (S1-S4/ S3: delta fridge) 1:57:06 Buffer PBS Timeline 2 (exp2) 50ul/min, PBS, pH= 7,4 13:13 Neutravidin 0,2mg/mL 18:09 BUFFER PBS 29:25 S1-dAPT11/S2-CONTROL/S4-dAPT15 33:30 BUFFER PBS 57:22 DMSO:EtOH (1:1) 10X PBS 1:02:01 BUFFER PBS 1:19:48 DELTA 0.5mg/mL 1:23:53 BUFFER PBS Timeline 3 (exp3) PBS pH=7.4, flow rate 50 μL/min 10:21 Neutravidin 0,2 mg/mL 14:16 BUFFER PBS 26:22 S1-dAPT15/ S2-dAPT11/S3-CONTROL 30:37 BUFFER 41:11 DMSO:EtOH (1:1) 10X PBS 45:34 BUFFER PBS 1:00:55 DELTA 0,5 mg/mL 1:05:02 BUFFER PBS

Learn

This triplet of experiments showed us that both aptamers, when compared during the experimental procedure, show similar binding to deltamethrin. The signals we received sometimes ranged from 160 Hz to 400 Hz, which may be explained by the fact that the experiments were performed on different pumps. However, since the signals in each experiment were not disproportionate to each other, we considered that we can interpret them in the above way. Additionally, compared to the control where appreciable signals were observed, we inferred that a proportion of the signals was due to non-specific binding of delta to either the sensor’s gold or neutravidin since the neutravidin surface was in direct contact with the dissolved delta. Contrary to this, in the sensors with aptamers the neutravidin is covered by the aptamers attached to it, through bonds with biotin at the 3'end.

Learn

The description provided above encapsulates an iteration of the entire design-build-test-learn cycle, while the testing consists of 4 experimental cycles where each one helped to optimize the protocol of the next one. Overall, the experiments were conducted to determine which aptamer would exhibit the greatest selectivity for binding to deltamethrin. To this end, separate experiments were conducted using dAPT11 and dAPT15 on different days and at varying concentrations. Furthermore, additional trials were executed, involving the concurrent use of both aptamers on the same day, at identical concentrations, under uniform experimental conditions. Notably, the results from the latter set of experiments revealed that the signals generated by the aptamers were closely aligned with each other, distinguishing them from the control group. It became evident that dAPT11 exhibited marginally larger signals and it gave signals in the initial experiments that were more interpretable. For instance, in an experiment where the aptamer was tested at concentrations of 2.5 pmol and 5 pmol, the signal disparity was notably more pronounced than in the case of dAPT15, where the signals at 2.5 pmol and 5 pmol were closely matched, contrary to expectations. Hence, the evidence and insights gleaned from these experiments led us to the conclusion that, based on the aforementioned observations, dAPT11 emerged as the preferred aptamer for integration into our system.

CYCLE 3: DNA polyacrylamide gel for identification of duplex and hairpin structures

Design

Before proceeding to building the entire system, we firstly individually tested the function of its constituent parts. One of the most important steps regarding the sensitivity of the biosensor lies in the hybridization of L1 and L2 molecules with the subsequent creation of a L1-L2 duplex. A necessary condition for this reaction is firstly the hybridization between dAPT and L1. We tested the formation of the aforementioned hybridizations through the conduction of an electrophoresis in polyacrylamide gel. The aptamer is being used in a duplex formation dAPT/cdAPT, so the hybridization between dAPT and L1 in the absence of deltamethrin will only be possible after the thermal denaturation of dAPT/cdAPT duplex and the hairpin structure of L1 molecule. More specifically, the S (a, b, c) region of dAPT binds to the complementary a* b* c* regions of L1 and thus structure dAPT-L1 is created. Subsequently, inside the actual biosensing system via the Toehold mediated strand displacement (TMSD) process, the L1 molecule hybridizes with the L2 molecule to produce a dsDNA, through the hybridization of b, c, d*,c* regions of L2 to L1 b*, c*, d, c regions, respectively. However, in order to run the electrophoresis we incubated only L1 and L2 molecules without the previous components of the system. In this case, L1-L2 hybridization requires the thermal denaturation of the L1 and L2 molecules for the breaking of the hairpin structures and for the release of the subregions that participate in the hybridization. Because of the small size of L1 and L2 (55b and 46b respectively), a highly sensitive variation of a DNA Polyacrylamide Gel Electrophoresis technique was used to indicate the hybridization between dAPT and L1 and between L1 and L2.

CYCLE 3A: Identification of duplex and hairpin structures

Build

L1, L2 and dAPT were all placed in 1ml tubes mixed with TE buffer, according to the concentrations provided in the table below. Additionally, we used L1, L2 as controls in separate tubes.

Table 1. Samples loaded in the wells.

The four mixtures were incubated in a water bath at 94oC for 4 minutes so that the double-stranded DNAs were denatured into single strands. Hybridization was achieved by naturally reducing the temperature of the water until it approached room temperature. Following, the samples were spinned down and were ready to be loaded into the Polyacrylamide Gel. The concentration of the polyacrylamide matrix used for a gel is a crucial factor that influences how DNA fragments are separated during DNA electrophoresis. Based on the small size range of the DNA fragments L1 and L2, we chose a 12% acrylamide gel as it is the optimal choice for a 25-150 (bp) resolution. 4.8 ml 30% Acrylamide (29:1), 4.8 ml H20, 200μl APS, 2.4 ML 1X TBE, 10 μl TEMED and DNA Safe Stain were mixed for the gel. We added 3 μl of the 6X loading buffer into each tube. Each gel well was fully filled with the entire solution amount of each tube. In order to prevent the system’s overheating as we increased the voltage after the first 15 minutes, we had to place the tray on ice.

Test

The electrophoresis was completed after 30 minutes and the pictures of the gel were captured using UV radiation. Αlthough the bands for L1 and L2 respectively were depicted as expected, with L1 being higher (closer to the negative electrode) than L2 due to slightly larger size, the band that would theoretically indicate the hybridization of L1 and L2 was slightly higher than the band of L2, but lower (closer to the positive electrode) than the corresponding band of L1. In the column in which well, we placed L1 with dAPT there was only one band with similar mobility to the one of L1-L2.

Learn

The above results were not the expected ones. Generally, if a DNA sample containing various DNA fragments is loaded and run through the gel, a pattern where smaller DNA fragments with higher mobility that move farther down the gel than larger fragments will be observed. That is why we expected that the band of the hybridized L1 and L2 would be higher than the non-hybridized bands of L1 and L2. The same result was expected in the column in which we run the 4th mixture (dAPT-L1) due to the dAPT-L1 duplex formation. In our attempt to explain the observed results, we initially hypothesized that no hybridization took place in any of the columns. In this case, however, in each column there should have been multiple clearly visible bands that correspond to the molecules placed in each well. For instance, two bands, one for each DNA, should have appeared in the well where L1 and L2 were loaded together. Therefore, in order to explain the depicted bands we consulted Prof. Spilianakis, an expert in the field of Molecular Biology. Mr. Spilianakis explained that the bands might actually indicate hybridization between L1-L2 and dAPT-L1, contrary to what we initially thought. He noted that the separation of DNAs in a DNA PAGE depends not only on the molecular weight and size of the DNA molecules but also on their structure. In our case, L1 and L2 molecules adopt a hairpin structure by themselves which can lead to lower mobility than the estimated one based solely on their size. Furthermore, the L1-L2 duplex is completely linear due to the complete hybridization of the 2 molecules. A linear structure could be in fact responsible for the higher mobility of L1-L2 duplex than the estimated mobility based solely on its size. As a result, the L1-L2 duplex as well as the dAPT-L1 duplex could have an intermediate mobility compared to the mobility of L1 and L2 molecules. In order to reinforce this explanation, we asked him how it is possible to not see any other band even if hybridization actually occurs. We thought that as a reaction, hybridization can not have a 100% efficiency, meaning that there should be some molecules of L1 and L2 that don’t participate in the binding and thus creating bands in the column with the same mobility of L1 and L2 columns, respectively. He replied that due to the higher melting temperature of the L1-L2 duplex, this confirmation will occur much faster than the hairpin structures of L1 and L2 in the same high temperatures. This means that a single L1 or L2 molecule, even if it’s not hybridized at first, due to its lower TM will be linear at high temperature and will be more likely hybridized until the temperature reaches the room’s temperature than remain as a single strand and adopt its hairpin structure. Finally, he advised to use a gel with 8% polyacrylamide. An 8% gel has larger pores compared to a 12% gel and can possibly allow the movement of the DNA molecules to be mostly affected by their size and not their structure, as a hairpin structured ssDNA will have no problem moving through the large pores of an 8% gel.

CYCLE 3B: Polyacrylamide gel concentration correction for more size-based separation by differential mobility

Build

L1, L2 and DAPT were all placed in tubes of 1000μl mixed with TE buffer as in the previous experiment, according to the combinations of the table below. We used L1, L2 as controls in separate tubes.

The procedure of incubation was the same. However the gel mixture was at 8% polyacrylamide concentration and not 12%, as Mr. Spilianakis advised. This time 3.2ml 30% Acrylamide (29:1), 6.4ml H20, 200μl APS, 2.4 mL 1X TBE, 10 μl TEMED and DNA Safe Stain were mixed. Additionally, the amount of loading buffer used was reduced (as indicated by the table above) in order to avoid the dark-appearing smears and enhance the sharpness of the bands.

Test

The occurring bands were exactly the same as the ones that occurred in the first experiment.

Learn

After the repetition of the same experiment with 8% gel and reduced amount of dye, the replicable results have led us to adopt the Prof Spilianakis’ hypothesis that the hybridization of L1 and L2 and DAPT with L1 was successful, as described in the Learn Section of CYCLE 3A.

CYCLE 4: Streptavidin coated magnetic beads (SA-MABs) synthesis procedure

Design

Every detection system should be able to turn on and off in response to the presence or absence of the targeted molecule. Our Wet Lab team, after thorough research, found that using streptavidin coated magnetic beads (SA-MBs) to achieve this responsiveness, is ideal in our case. The SA-MBs were integrated in the system in such a way in which they can remove lipase, the chromogenic enzyme that depict by its function deltamethrin's presence, from the solution. To do so, they must be linked with lipase in a way in which they can also be separated if deltamethrin is present. The LINKER DNA molecule is responsible for connecting the magnetic beads with Lipase and the same molecule gets hydrolyzed if deltamethrin is present. In this way, SA-MBs and Lipase are separated. One of the easiest ways to connect a streptavidin coated magnetic bead with a DNA molecule is through the biotin-streptavidin interaction. For this reason, we designed our LINKER molecule and the magnetic beads to contain biotin in the 5’ end and be streptavidin-coated, respectively.

Build

We followed a modified version of the protocol provided by Peijun et al. (2012), as described in detail in the Experiments section. The adjustment of the protocol was based on many factors, such as the reagents and equipment readily available in the lab as well as the desired outcome of the synthesis. The first step was the synthesis of the magnetic beads that are in fact magnetite aggregates. To create them, we mixed 4.46g FeCl2 and 1.6g FeCl3 and diluted them in dH2O. The aggregation of magnetite molecules requires alkaline conditions at a pH of around 10, so after stirring, we added NH4OH dropwise . However, a neutral pH is necessary for streptavidin binding to the magnetic beads. Therefore, after their formation, we repeatedly washed the mixture to lower the pH back to neutral (pH around 7). Then, 10ml of the stock solution of magnetic beads was mixed with 37μl of streptavidin and stirred overnight at 4oC. The next day, the mixture was washed to remove any excess unreacted streptavidin molecules.

Test

After conducting the experiment once, we realized that our magnetic beads had lost almost all their magnetic properties because there was no significant impact on the mixture after close contact with a magnet. Their color remained black, so there was no problem with the magnetic beads alone, because the black color only occurs if magnetite aggregates are present that are by default magnetic.

Learn Stage

We hypothesized two reasons for this magnetic loss. Firstly, streptavidin causes this deficit after coating the aggregates. This might seem logical, but the literature has mentioned several times that streptavidin coating can not change magnetic properties by itself. Another possible cause could be the high concentration of the magnetic beads used in the incubation and the high volumes of both the magnetic beads and the streptavidin that were used. We noticed that after fractionating a small quantity of the stock of magnetic beads, which was then diluted with dH2O, the occurring solution was almost instantly cleared after putting it next to a magnet due to the localization of the beads in the tube’s wall next to the magnet. This effect was weak in the stock solution. Therefore, we made two dilutions of the stock solution, one at 1/20 and another at 1/100 after mixing 50μl and 10μl of the stock solution with 950μl and 990μl of dH2O, respectively. Then, we added only 2μl of streptavidin to both diluted solutions and followed the same protocol.

CYCLE 5: Conjugation procedure and crosslinking verification

Design

If the biosensoxawdr detects deltamethrin in the water sample, we will observe a color change in the supernatant. We decided to use an enzyme that will change the color of the water to be immense and easily observable. More specifically, we used Lipase which converts p-nitrophenyl ester to 4-nitrophenol which is the substance that will give the supernatant its yellow color. Furthermore, LINKER will be linked to Lipase in its 3’ end and to the magnetic bead in its 5’ end. If the water contains deltamethrin, LINKER will get hydrolyzed, and by using a magnet, the magnetic bead and a part of the LINKER will get precipitated in the bottom of the tube. However, Lipase and the rest of the LINKER will remain in the supernatant. The supernatant will then be moved to a new tube and after the addition of p-nitrophenyl its color will change. However, if deltamethrin is absent, LINKER will remain intact. As a result, after removing the magnetic beads with the magnet, we will also remove Lipase because it is attached to the bead via the LINKER. As a result, the transferred aliquot of the supernatant will retain its color after the addition of Lipase substrate.

Build

The formation of an amide and a thioether bond allows the conjugation of LINKER with Lipase. Before the linkage, both lipase and LINKER should get activated. An important step is mixing the DNA, sodium phosphate buffer with pH 5,5 and TCEP. After 1 hour at room temperature, they get centrifuged, the 3' Thiol Modifier C3 S-S attachment of LINKER will break and the required thiol group will be formed. For the activation of Lipase, a purification protocol is followed in order for some salts to be removed and the lipase to be condensed. More specifically, the lipase is washed and then the supernatant is purified 3 times in the centrifuge using Amicon-10k filters and Buffer A solution (pH of 7.3). Then, Lipase is mixed on the shaker with sulfo-SMCC in a buffer containing sodium phosphate buffer and NaCl for 1 hour at ambient conditions. Then, the solution is centrifuged, the excess of insoluble SMCC is discarded, and the supernatant is purified using Amicon filters. For the conjugation of LINKER-Enzyme, the LINKER with the activated thiol group is mixed and incubated with the activated enzyme for 48 hours at room temperature. Sulfo-SMCC is applied for specific bioconjugates via one- or two-step crosslinking reactions. More specifically, SMCC is an amine-to-sulfhydryl crosslinker that contains NHS-ester and maleimide reactive groups at opposite ends of a medium-length cyclohexane spacer arm and it can react towards amino and sulfhydryl groups. From the one side, the ester-group side, will be conjugated with the enzyme, via the formation of a covalent amide bond. The ester will be converted into an amide by an aminolysis reaction with the N-Terminal peptide. We followed the protocol provided by (reference) as described in detail in the Experiments section. Besides amine-reactive compounds, those having chemical groups that form bonds with sulfhydryls (–SH) are the most common cross-linkers and modification reagents. The maleimide group reacts specifically with sulfhydryl groups when the pH of the reaction mixture is between 6.5 and 7.5; the result is formation of a stable thioether linkage that is not reversible. So, the Linker that now has an available thiol group, will be also conjugated with the SMCC but from the maleimide side (the opposite side).

Finally, the process of making the trimeric complex requires the additional preparation of the streptavidin-coated magnetic beads. Firstly, the beads are resuspended using Buffer A solution as described in the Experiments Section. The interaction between streptavidin and biotin allows SA-MBs to connect to the LINKER-Lipase complex, completing the cross-linkage process. Therefore, the LINKER-Lipase complex is mixed with the solution of streptavidin-coated magnetic beads and remains in the shaker for half an hour at room temperature. After those steps, the complex consisting of MBs-LINKER-Lipase is formed.

Test

During the above experiments we kept 10 µl of the enzyme used to ensure the enzyme's functionality by observing a color change after the addition of its substrate. After mixing this amount of enzyme with 10μl of its substrate, a color change was observed as the supernatant changed from transparent to yellowish; thus confirming that the enzyme was indeed functional. Subsequently, after joining the magnetic beads with the LINKER-Lipase complex, we also wanted to confirm that the formation of the trimeric complex was adequately completed. So, in the solution containing these three parts, we used a magnet to pull this triple complex to the wall and took 10μl of supernatant to see if it would cause a color change when it's being added to the substrate solution. We expected no color change in substrate solution as the beads and Lipase would be pulled to the wall in contact with the magnet resulting in the complete removal of Lipase from the supernatant. Indeed, 10 µl of the supernatant were mixed with an equivalent amount of substrate and no color change was observed indicating the successful crosslinking between SA-MBs-LINKER-Lipase.

Learn

From the conduction of the above cycle, we verified the successful crosslinking between the 3 factors which was a necessary step before proceeding to the next part of the system.

Figure 17: The left tube contains the LINKER-Lipase solution mixed with the substrate solution. The color change of the solution indicates Lipase’s functionality.

CYCLE 6: Biosensor implementation into deltamethrin-spiked samples and the characterization of its functionality

Design

During all the previous steps, our objective was to see if the individual elements of the experiment were working properly. For example, we started by choosing the most selective aptamer via the Docking procedure and the QCM biosensor, we checked the hybridizations of dAPT-L1 and L1-L2 molecules via PAGE and verified the binding of the magnetic beads with lipase by observing the color change of the supernatant solution. The only part of the system that could not be determined whether it was working or not was the activation of Cas12a. Since all the confirmatory steps were carried out, the final experiment was the actual detection of the pesticide in lab-made deltamethrin-spiked samples to test the functionality of the biosensing system as a whole.

Build

For the preparation of deltamethrin samples, a stock solution of 5 mg/ml was initially formed which would be further diluted during the preparation of the total system solutions. 20 µl of deltamethrin stock sample was mixed with dAPT, L1 and L2 DNA molecules and diluted in Tris-HCl and deionised water to a final volume of 200 µl and a final concentration of 0.5 mg/ml. We set deltamethrin concentration to 0.5mg/ml in the deltamethrin-dAPT solution as this concentration was indicated as the ideal one for dAPT-delta interaction by the QCM experiments. Afterwards, Cas12a enzyme sample was prepared. Cas12a is responsible for the release of lipase into the supernatant caused by the hydrolysis of LINKER and the subsequent change in color of the substrate solution. More specifically, 10μl of 10X NEB buffer r2.1 were mixed with 75μl of 1μM Cas12a and 2μl of 100μM crRNA. Finally, dAPT-deltamethrin, Cas12a and SA-MBs-LINKER-Lipase solutions were mixed and incubated. This would allow the interaction of particular factors such as L1-L2 hybridization and the induction of several reactions such as the hydrolysis of LINKER by Cas12a. In the last part of the detection system described above we use a magnet to attach the magnetic beads to the walls so that they can be separated from the rest of the solution leaving lipase in the supernatant. Afterwards, the supernatant was transferred in a new eppendorf and 10μl of it were added in 10μl substrate solution.

Test

After the incubation of the aforementioned solutions, we performed magnetic separation to the mixture causing the localization of the SA-MBs to the tube’s wall next to the magnet. As we have described in the design section of this cycle, we implemented the biosensor to a deltamethrin-spiked water sample. This means that we expected a positive result from the sensor which would be depicted in a color change of the substrate solution after adding 10μl of supernatant. However, no color change was observed.

Learn

The fact that there was no color change in the substrate’s solution implies the absence of Lipase in the supernatant. This in turn means that Lipase was also localized with the SA-MBs during the magnetic separation stage which could only be possible if LINKER remained intact. As can be easily understood, Cas12a for some reason was not activated resulting in the non-hydrolysis of LINKER. The inability for Cas12a activation could in fact lie on several preceding steps of the system. However, as mentioned above, after a series of experiments in QCM, we have seen that the aptamer can indeed bind with deltamethrin and that L1-L2 is most likely hybridized as established during the DNA polyacrylamide gel electrophoresis procedure. Additionally, we didn't only confirm that lipase was functional but also that the SA-MBs-LINKER-Lipase binding was successful. The successful formation of this trimeric complex rules out the possibility of problems occurring during the system signal transduction after the hydrolytic action of Cas12a. As a result, we concluded that the most probable reason for the non-activation of Cas12a was its insufficient interaction with crRNA which is necessary for Cas12a normal function. In general, RNA molecules are easily degradable due to their physicochemical properties but also the function of RNases. Therefore, it is highly possible that during the handling of crRNA it got damaged and rendered non-functional. Another possible cause could be the inhibition of L1-L2 duplex formation as the dAPT-Deltamethrin solution contained small amount of DMSO and ethanol which are strong organic solvents that can interfere with hydrogen bond formation.