Background information
Initially, as team we were looking out to develop a device which can work as one of the best point of care technology for detecting diseases. We soon realised that the technology we envisioned to develop can not only detect the microbes commonly responsible for causing sepsis like diseases but rather can detect any target molecule against which we can generate a highly specific and sensitive aptamer.
Our institutional iGEM committee’s head Prof. Anand Bachchawat also suggested to us that since our technology could detect any target molecule if an aptamer could be generated against it then why don’t we develop a “ Proof of Concept” that it could detect microbes which are more commonly available around us and are easier to work with.
We considered these aspects and decided to work with E. coli, S. typhimurium and Pseudomonas aeruginosa. Through the literature search, we found that these common microbes have already aptamers developed against them using whole cell-SELEX which are highly sensitive and specific.We decided to choose the microbes that had already available aptamer sequences in the literature as we understood that whole cell-SELEX process takes time to model an aptamer sequence and given the time frame of the competition we won’t be able to complete everything we wanted to exhibit if we are to begin with SELEX.
Our team wanted to make a system that would be better than the existing fluorescent-based detection methods due to its high false positive rate. We also wanted to quickly expand to the sample's multiple, simultaneous, multi-cohort detection. This was not possible with existing technologies with individual detection capabilities and data with differing error margins, which are later processed for a complete analysis of the disease. We solved all the above issues by implementing the processing system within the detection system using a neural chip that simultaneously detects and processes signals. It is also cheap, easy to mass manufacture, and accessible with minimal electrical equipment (in the proposed implementation a battery-powered reading box would suffice). The system would also have an error margin which would be easy to calculate against accurate world data, unlike data derived from individual detection, which have errors that cannot be easily calculated. Neurons are also susceptible to changes in the electrical fluctuations at the 70-100mV range, thus reducing the need for high-powered devices like spectrophotometers that also increase measurement and maintenance costs.
Further, our team member figured out later that the technology we envisioned developing the device and the neural chip we were developing for bio-computation could not perform the computation without dimension reduction. We could not have three inputs and three outputs if we had to use a neural chip to perform bio-computation and analyze the electrochemical changes that would happen upon the binding of microbes to the aptamer. The dimension reduction property would prove our chip's biocomputation ability and help reduce stray signals that would be generated and measured. This is why the chip was planned to have 3input signals corresponding to three different organisms but only two output signals. The interpretation of these signals would be based on the current output of the chip. The chip would present an overall signal lowering in both outputs when detecting the third organism. This ensured that the number of outputs and the interpreting device's involvement were minimized.
Hence, we further, our team member figured out later that the technology we were envisioning to develop the device and the neural chip we were developing for bio-computation cannot perform the computation without the dimension reduction. We cannot have three inputs and three outputs if we had to use neural chip to perform bio-computation anad analyse the electrochemical changes that would happen upon the binding of microbes to the aptamer. This would act as a biomarker for the fungi Penicillium chrysogenum.
Thus, we decided to work with E.coli MTCC 443, S. typhimurium MTCC , Penicillium chrysogenum as our chosen microbes.
Why Aptamers, Neural Chips and Electrochemical sensing?
Aptamers have emerged as a great tool in the recent years in the development of point of care diagnostics, drug delivery and therapeutics. Aptamers could be either ss DNA or RNA oligonucleotide sequences that adapt a secondary structure able to specifically bind to a target molecule. They are mainly generated by SELEX (Systematic Evolution of Ligands by Exponential Enrichment), which is a reiterative process starting with a large DNA or RNA library, and at each step, only high binding affinity sequences are separated from the library. One example is Silver Enhanced Fluorescence-Activated Cell Sorting. In this process, the target protein or the control/negative proteins (which are used to ensure specificity of the aptamer) are immobilized on polystyrene microspheres and added to the large DNA/RNA initial library, modified with a fluorophore. Silver decahedral nanoparticles conjugated to a fluorophore are also added, and then these NPs will interact with the labelled DNA and clump together. Through fluorescence-associated cell sorting, one can then choose the cells that had DNA clones that tightly bound to the protein.
We can observe that only aptamers that bind to the cell or protein of interest are selected from the initial library. This is the first round, where enrichment occurs. All other rounds are negative selection, where the protein or cell of interest reacts with silver nanoparticles, and the non-binding DNA sequences are directed to the waste container via an electric current. The collection container sequence is then amplified via PCR, and ssDNA or RNA aptamers have been generated as a result. They can then be sequenced via standard sequencing methods. Aptamers have the advantage that any molecules can be targeted, and there are no immunizations of animals required. This results in a high affinity and high specificity receptor that can be selected for and then synthesized completely in vitro. Moreover, since aptamers are just a DNA sequence, they can be ordered from DNA synthesis companies.
Fortunately, the sequences we wanted to work with were highly specific and sensitive with KD value in nM range and were available under IDT’s sponsorship programme.
Fluorescence detection using Aptamers was initially considered, but our team soon realized a number of problems with this method. Aptamer being a small molecule is prone to more molecular motion than traditional antigen-antibody detection systems. This makes the fluophore and the quencher interaction more prone to misfiring. This increases the false positive rate in such systems. We also considered the problems faced by the use of spectroscopes and image processing of biological data. Spectroscopes have a error percentage of 5% at best condition. This makes the data highly unsuitable for further processing as additional processing will further add error to the measurement.
Also most spectroscopes in a developing country like ours, do not have regular maintenance. We were made aware of this issue by food scientists working in Verka, during our vist. This has made the food industry trust less on spectroscopic data for bacteria detection and prefer culturing the bacteria to check for its presence. This is a time consuming process that is prone to environmental contamination.
To solve all of these problems we decided to use a property of the aptamer which is linear, easy to measure, shows large range changes, scalable and devoid of further processing steps. We hence decided to use the impedance change property of Aptamers and interpret out results using a neural chip, which solves all the above problems. The chip processes the data from the aptamer and directly outputs a easily quantifiable signal, generated from an Ensemble of neurons. The neurons are also more sensitive than traditional current detection systems, firing at 70-100mV range. The trainability of the network also allowed us to build the compuational system within the detection system rather than taking the data separately and processing it. This reduces dependency on big and expensive machinery and is easily deployable in the field. The results are instantaneous and can be monitored in real time. The neural chip itself is reusable, cheap to manufacture and easy to dispose. It contains no heavy metal contaminants, other than the wiring of the chip, which itself can be reused to make more chips. It also requires no maintenance, other than a normal temperature of 37*C, and once a supply chain can be established will be a perfect tool for rapid, cheap and sustainable multicohort detection.
For E. coli ATCC generic strain 25922 (MTCC 443) , according to the studies, whole bacterial cell is used as a target and P12-31 is highly specific and sensitive aptamer towards it. Its Kd value is 87.03± 17.32, Bmax value is 0.56±0.04 and around 10113 no. of binding sites are available for it. Fluorescently labeled aptamers label the surface of E. coli cells, as viewed by fluorescent microscopy. Specificity tests with twelve different bacterial species showed that one of the aptamers–called P12-31—is highly specific for E. coli.
Sequence of P12-31 aptamer is given below which is obtained by Whole-Cell SELEX:
5'CCCTCCGGGGGGGGGGGTCATCGGGATACCTGGTAAGGATACCCTCCGGGGGG
GTCATCGGGATACCTGGTAAGGATA 3'
Structure:
5′ATAGGAGTCACGACGACCAGAAAGTAATGCCCGGTAGTTATTCAAAGATGAGTAGGAA
AAGATATGTGCGTCTACCTCTTGACTAAT3′
Structure of the ST2P Aptamer:
EIS measurement is a powerful technique for analyzing the physiochemical changes at the interface (i.e., between the bulk solution and solid electrode) arising from the formation of insulating film incurred by the interaction of the analytes with their probing molecules immobilized on electrode surfaces. Measurement of charge transfer resistance RcT the impedance increases across the interfacial layer EIS provides a rapid, sensitive, and nondestructive technique for detecting various analytes, including biomolecules that would otherwise be difficult to quantify.
Impedance is usually expressed as a Nyquist plot. A typical Nyquist plot consists of a semicircle in the high-frequency region representative of the electron transfer limited process and a straight line in the low-frequency region associated with the diffusion-controlled process. In fig:3, The bare GC showed a small semicircle in the high-frequency region with the calculated RCT value of 3300Ω. After the electrochemical deposition of GO/PANI, the Nyquist plot for EIS spectra shows the decrease in RCT value showing the reduction in the surface resistance, therefore providing a more effective platform for developing aptasensor.3 The Nyquist plot of the electrode where the aptamer molecule shows an increase in the RCT value to 11000, showing the successful assembly of the aptamer layer on the modified GC. Aptamer acts as a barrier to the anion electrolyte's restriction [Fe(CN)6]-3/-4 to the electrode surface.4
The Nyquist response shows a vast increase in the RCT value to 40000 Ω showing the complete coverage of E. coli by the biosensor.
Similar observation can be observed for CV (fig:5) and EIS response (fig:6) for developing the Penicillin aptasensor. On another GC, a Change in the RCT from 1400 Ω to 1100 Ω after deposition shows the GO/PANI layer formation on the GC surface. EIS curves also show the change in RCT value for penicillin aptasensor and penicillin, which is seen by the RCT value or the semicircle size of the EIS response. The RCT values for penicillin aptasensor of around 1500 Ω shows the SAM of aptamer onto the deposited GO/PANI surface, the RCT value changes to 3600 Ω after dropcasting of penicillin onto modified aptasensor GC. Hence, the sensitivity of the developed aptasensor towards penicillin is good.
For the measurement of the sensitivity of the E.Coli aptasensor for the given E.coli strain, the concentration of the E.coli was varied in two different ratios in terms of optical density, i.e., 0.7OD and 0.07OD.
The experiment was performed in two different GCs with two different E.coli concentrations. The previous experiments for the development of aptasensor were the same, and the GCs CV (Figures 7 and 8 ) and EIS (Figures 9 and 10) responses were recorded.
For GC with E.coli concentration of 0.7OD, the CV responses were recorded (Figure: 7). The change in CV curves area with the going from bare GC to the E.coli assembled form was observed. As seen in figure: 7, the CV responses show a decrease in the current value with the given potential window. Figure 8 shows the CV of the 0.07 OD E.coli, showing a similar response as that of 0.7OD E.coli. Still, the CV curve for the 0.7 OD E.coli assembled aptasensor is more horizontal than the other concentration of 0.07 OD E.coli. This shows the current is hindered more in 0.7OD E.coli than the other concentration ratio.
Nyquist plot for 0.7OD (figure: 9) and 0.07OD (figure: 10) concentration of E.coli shows the immense RCT value for 0.7OD concentration than 0.07OD concentration, i.e., 5000 Ω in former and 13800 Ω in later, this is the inference of the increase in bacterial assembly on the electrode surface.
To ensure that the physical hardware provided as little resistance to the incoming signals, to the neural chip we use Ti-Au layered input and output electrodes. Titanium provides structural support to the electrode and the Gold layer makes the electrode bio-safe and highly conductive. We however suspected that the electrode may show slight variability in its resistance, due to oxidation of exposed titanium. To ensure that the change in resistance was minimal, we kept a record of the resistance values over 8 days for the input electrodes of a selected chip, plated without liquid media. We report that the change in the electrode resistance was minimal, and stabilised over the span of 8days. This indicates that the sputtering was successful and the gold layer was preventing oxidation of Titanium underneath.
Measuring resistance of electrodes
Resistance values
We modelled our neural chip to be most sensitive around a current range of 150mA. A current drop below 100mA would virtually cause the neurons around the input electrodes to stop firing. We simulated these conditions using Arduino, where initially all input of the electrodes were kept at 160mA and then selectively switched off inputs by lowering their current strength to 50mA.
We used a bifocal fluorescence microscope to image the neuron-like cells. We used it in bright field mode and captured the pictures using an attached camera.
Copyright © 2022 — Content on this site is licensed under a Creative Commons Attribution 4.0 International license.