Monitoring tetracycline through a solid-state nanopore sensor

Antibiotics as emerging environmental contaminants, are widely used in both human and veterinary medicines. A solid-state nanopore sensing method is reported in this article to detect Tetracycline, which is based on Tet-off and Tet-on systems. rtTA (reverse tetracycline-controlled trans-activator) and TRE (Tetracycline Responsive Element) could bind each other under the action of Tetracycline to form one complex. When the complex passes through nanopores with 8 ~ 9 nanometers in diameter, we could detect the concentrations of Tet from 2 ng/mL to 2000 ng/mL. According to the Logistic model, we could define three growth zones of Tetracycline for rtTA and TRE. The slow growth zone is 0–39.5 ng/mL. The rapid growth zone is 39.5−529.7 ng/mL. The saturated zone is > 529.7 ng/mL. Compared to the previous methods, the nanopore sensor could detect and quantify these different kinds of molecule at the single-molecule level.


. Experimental setup
The fluidic cell was made by PMMA (Polymethyl Methacrylate). Each part of the fluidic cell includes a 200 L chamber, Cis chamber and Trans chamber. The nanopore device fabricated by dielectric breakdown was mounted in between the two parts of the fluid cell. This nanopore was the only one channel, which connects the Cis and Trans chambers. An Axopatch 200B (Molecular Devices, CA) was used to apply voltages and detect ionic current through two Ag/AgCl electrodes that were put into each of the chambers respectively. The signal was digitizedby Axon Digidata 1550 with a low-pass 10 KHz filter. Data were acquired and analyzed with the following softwares, pClamp 10.0 (Molecular Devices) and Matlab-based program.
The mean values from amplitude and dwell time histograms (over hundreds of events)were obtained with Gaussian and exponential functions. All of this experiment was performed at room temperature. Every Si 3 N 4 chip was immersed in piranha solution for 20 minutes before nanopore fabrication. This process was performed toclean the Si 3 N 4 chip and prevent the generation of air bubbles. Before nanopore experiments, rtTA and TRE fragment were put in reaction buffer (deionized water).
Then 0-2uL Tet was put in it (estimated concentration of rtTA was 25000 ng/mL, estimated concentration TRE was 14000 ng/mL, while that of Tet was between 0 ng/mL and 20000 ng/mL), which the total volume of them is 20 uL, and then let them stand for 60 minutes at room temperature.The Logistic mode fits well with our data.

Discriminate rtTA and TRE fragment separately
The Si 3 N 4 nanopore around 8.5 nm in diameter with 10-nm thickness was used in this experiment. The resistance of this nanopore was around 25 ΜΩ. Figure   It is clear that the events were clustered together for all of them from the figures. In order to clearly describe the characteristics of data distribution, we used exponential and Gaussian fittings for the histograms of dwell time and amplitude, respectively. The value of t 1 (about 0.31s) is almost two times larger than that of t 2 (0.15s), whereas t 1 and t 2 are the attenuation factors of exponential fitting results of rtTA and TRE fragment. As can be seen from Fig. S5(e) and (f), there were one peak (pk 1 ) around 250pA for rtTA and two peaks (pk 21 and pk 22 ) that one is around 250pA, the other is around 430pA for TRE fragment. It is clear that the events were clustered together for all of them from the figures. We could obviously discriminate the rtTA and TRE fragment based on the dwell time and amplitude values.

Before and after adding Tet into the mixed rtTA and TRE solutions
We tested the mixed rtTA and TRE fragment with nanopore firstly with 8.5 nm nanopore in diameter. After adding Tet to this buffer solution, the signal was changed significantly. All types of this variation trend are showed in Figure S6. There are two different areas showed in Fig. S6(a): the blue area and the red area, which represented the events of putting rtTA and TRE fragment in one buffer solution before and after adding Tet. There are almost no overlaps in these two different areas.To analyze data accurately, exponential and Gaussian modes were adopted to fit the distributions of dwell time and amplitude. From Fig. S6(c) and (d), the value of t 4 (about 3.7s) was much larger than that of t 3 (about 0.3s). Figure S6(b) shows the amplitude distribution of before and after adding Tet: the position of pk 3 is around 310 pA and the position of pk 4 is around 475pA. rtTA and TRE fragment combined each other under the action of Tet and produced a much larger complex. This complex could stay too much longer in nanopore and produce a higher blocked current signal. From the results of EMSA, rtTA interacted with TRE fragment even without Tet, while rtTAbinded more with TRE fragment in present of Tet. This result was observed more clearly in our nanopore experiment. The signals of rtTA were obviously different from that of TRE fragment when the experiments were performed individually. There are two peaks for TRE fragment only in Fig. S5(f).It does not happen when rtTA and TRE fragment were added into one buffer solution simultaneously. A single combined signal that has 6 higher amplitude was observed in Fig. S6(b).The interaction between rtTA and TRE may just allow TRE to get through the nanopore in one fixed type, which may be the reason of producing different signals. After adding Tet into this solution, there is a significant change in both amplitude and dwell time. This reason is clear: rtTA and TRE could combine to each other to form a much bigger complex under the action of Tet, and this complex would stay too much longer time in nanopore and produce much higher amplitude. The insert shows the details of circled areas at the left-bottom corner of this picture.
The Logistic mode fits well with our data. We got four special concentrations of Tet: EC05 (concentration for 5% of maximal effect,39.5 ng/ml),EC20 (concentration for 20% of maximal effect, 99.8 ng/ml), EC50 (concentration for 50% of maximal effect, 230.0ng/ml), EC80 (concentration for 80% of maximal effect, 529.7 ng/ml). Tet. We defined area1 and area2 as the peak areas for peak1 and peak2. The overlapping region (area3) between area1 and area2 could bring large errors when the ratio between peak1 and peak2 is calculated.