Electrical Resistance as a measure of Growth |
|
As stated before, to connect the different parts of our circuit we use the marionettes inducing system [3]. Between the processor and the output we use the ——— (OC6) inducer, which is produced by the processor To avoid a ion interference with the other redox-based pathway we used for the input, we preferred to use a lysis system to connect electrically the output of our circuit. The idea was to view the number of cells in the medium as a resistance value of the medium, allowing us to measure it using a potensiostat, thus connecting the output our circuit with an electrically measurable unit |
|
As shown in [1], growing bacteria’s capture ions present in their media, and, in doing so, decrease the electrical conductivity of the medium, thus, the more there is the more the resistance decreases. The goal of our experiences was to connect OD measurements to a resistance value and show a linear relation between the increase of OD and the decrease of media resistance. Our aim was to achieve a reproductible measurements method that could allow us to compare different resistances each measured at different OD, despite the change in media composition caused by the change in bacteria concentration. |
Materials and methods |
|
We took a liquid culture of a bacterial colony (overnight grown in LB + Kanamycin ) and made multiple dilution to mimic multiple values of lysis of the colony. Each of the dilutions were characterized via an OD measurement, to have a precise value per dilution that would serve as a backbone for the comparaison of the resistance measurements. The resistance measurement were done using constant voltammetry, with the IO-Rodeo Open Source Potensiostat[2]. We would give an input voltage, and read a outputted current after it had been affected by the media. This was then computed thanks to the Ohm Law as a media resistance. The whole process was automated using a python code. All measurements of every OD were done at the same parameters (inputted voltage, duration of electrocution, and current reading range) to be able to compare them. |
|
However, reactions happening between redox molecules potentially inside of the medium are affected by the number of bacteria which release or absorb molecules from their environment. Furthermore, redox reaction behave differently at different concentrations, and with different voltages and different bacterial concentration (visible with the OD), we might trigger different types of reactions involving different redox couples. Knowing that we can’t compare different media resistances, i.e different number of ions flowing, if we are not dealing with the same types of ions, w therefore needed to find a way to compute the right value at which perform our constant voltammetry in order to have reproductive and comparable result at different OD, so at different bacterial concentration. |
Results |
|
To compute this right value, we did a linear sweep of voltages, ranging from -1.5mV to 1.5mV. We then plotted the evolution of the resistance in regards to the voltage (fig 1), and pinpointed an interval where the relation between the resistance and the different ODs was proportional (fig 2). We now had a value at which we had a resistance value of the media that was proportional to its OD, connecting reliably bacterial concentration that was determined by the expression of the lysis gène, which in itself was controlled by the value given by the processor. We therefore can assess the production of **HCL4** (molecule produced by the processor) using a resistance value, electronically measured and thus connectable to other electronics. However, there were no clear monotonic relation between all 17 ODs at any voltage. We then had to scan all this data to find the biggest range where the maximum of OD were behaving in a monotonic way. The rightness of the candidates then needed to be assessed by comparing their standard deviation to see if the data was fit to pull conclusions from. After those manipulations we found one range where the resistance was strictly increasing as the OD increased. The figure 2.a. shows the evolutions of the media resistance in this range for selected OD that held this monotonic relation, and the figure 2.b. only keeps the range at which the standard deviation is bellow 5%. Finally to clearly show this monotonic relation, we plotted the different ODs as a function of the media resistance, and as shown in the figure 3., we do have a clear strict increase of media resistance as the OD increases when we do the measurements at -0.45mV. |
fig1 Voltage sweep for 17 value of Optical Density
fig2.a Evolution of the resistance of different media at different bacterial concentration (OD) as a function of voltage
fig2.b Evolution of the resistance of different media at different bacterial concentration (OD) as a function of voltage (Zoomed in on the section with the lowest standart deviation)
Our system isn’t reliable for low bacterial concentrations, but can clearly electrically detect differences of OD (above 0.5 where the blank was set for pure LB). This shows that with a better tuning of our system and the development of an algorithms that mimics the whole process that we went though to find the right voltage at which preform the measurements, we can measure the strength of a genetic output using a lysis system at a population level.
Using this output visualization method we can sense the response of our processor, and thus complete our circuit that takes a input voltage, processes it, and finally gives a response we can now quantify.
Fig3 Evolution of the resistance at -045mV as a function of media bacterial concentration (OD)
References |
1. Meyer, A.J. *et al.* (2019) ‘Escherichia coli “Marionette” strains with 12 highly optimized small-molecule sensors’, *Nature Chemical Biology*, 15(2), pp. 196–204. Available at: https://doi.org/10.1038/s41589-018-0168-3
2. Tschirhart, T. et al. (2016) ‘Electrochemical Measurement of the β-Galactosidase Reporter from Live Cells: A Comparison to the Miller Assay’, ACS Synthetic Biology, 5(1), pp. 28–35. Available at: https://doi.org/10.1021/acssynbio.5b00073.
3. Din, M.O. *et al.* (2020) ‘Interfacing gene circuits with microelectronics through engineered population dynamics’, *Science Advances*, 6(21), p. eaaz8344. Available at: https://doi.org/10.1126/sciadv.aaz8344