A synthetic mammalian electro-genetic transcription circuit

Abstract

Electric signal processing has evolved to manage rapid information transfer in neuronal networks and muscular contraction in multicellular organisms and controls the most sophisticated man-built devices. Using a synthetic biology approach to assemble electronic parts with genetic control units engineered into mammalian cells, we designed an electric power-adjustable transcription control circuit able to integrate the intensity of a direct current over time, to translate the amplitude or frequency of an alternating current into an adjustable genetic readout or to modulate the beating frequency of primary heart cells. Successful miniaturization of the electro-genetic devices may pave the way for the design of novel hybrid electro-genetic implants assembled from electronic and genetic parts.

Figure 1.

Synthetic mammalian cell-based electro-genetic device. (a) Circuit diagram of the electro-genetic device. DC applied to the input device results in the electrochemical conversion of ethanol into acetaldehyde, which enables the acetaldehyde-dependent transactivator AlcR to bind and induce transcription from its cognate promoter P_AIR which triggers transcription of the human placental SEAP. SEAP subsequently catalyses the production of coloured p-nitrophenolate, which is quantified photometrically at 405 nm by a photodiode and converted into an electric output signal. AlcR, acetaldehyde-inducible transactivator; pA, polyadenylation signal; P_AIR, AlcR-responsive promoter; P_LTR, murine stem cell virus 5′ long terminal repeat-derived promoter. (b) Characterization of the CPU. The CPU was connected to DC of different intensities for 1 h, and the resulting acetaldehyde concentration as well as the electric output signal were quantified after 24 h. (c) Correlation between DC input and corresponding voltage. (d) Characterization of the output interface. SEAP activity was plotted against the electric signal measured by the output interface.

Figure 2.

Electro-genetic circuits. (a) Mammalian cell-based integrator. The CPU was connected to DC of different intensities and for various periods of time; the product of time and current intensity was kept constant (t × I = constant). The resulting acetaldehyde levels and electric output signals were scored after 24 h. (b) Mammalian cell-based AM detector. The CPU was connected to an AC of 50 Hz and different intensities for 1 h and the acetaldehyde levels as well as the electric output current were quantified after 24 h. (c) Mammalian cell-based FM detector. The CPU was connected to an AC of 50 mA and different frequencies for 1 h and the acetaldehyde levels as well as the electric output current were quantified after 48 h.

Figure 3.

Mammalian cell-based frequency generator. (a) Circuit diagram of the cell-based frequency generator. DC power converts ethanol into acetaldehyde, which dose-dependently triggers expression of the BMP-2 in engineered rat cardiomyocytes (_AIRNRC-BMP-2) and increases the contraction frequency (tachycardia). (b) The beating frequency of cardiomyocytes is recorded as a function of the input current and acetaldehyde concentration using a CMOS-based HD-MEA. NC: negative control, mock-transduced cardiomyocytes; PC: positive control, cells transduced for constitutive BMP-2 expression. (c) HD-MEA-based analysis of the electrogenic behaviour of NRCs engineered for electro-inducible acetaldehyde-responsive BMP-2 expression. The dataset shown as example was recorded at a direct input current of 50 mA corresponding to a beating frequency of 2.1 Hz. Detailed activation map illustrating the average signal shape of the 121 selected electrodes during 10 s. Average signal shape over all 121 channels. Raster plot showing a dot for each contraction on each channel over time. Inter-burst interval or inter-beat interval used to calculate the average beating frequency and beating frequency variation. (d) Zoom-in of two selected bursts/beats on different electrodes. (e) Long signal trace showing the synchronized contraction frequency on five selected electrodes or channels.

Figure 3.

Mammalian cell-based frequency generator. (a) Circuit diagram of the cell-based frequency generator. DC power converts ethanol into acetaldehyde, which dose-dependently triggers expression of the BMP-2 in engineered rat cardiomyocytes (_AIRNRC-BMP-2) and increases the contraction frequency (tachycardia). (b) The beating frequency of cardiomyocytes is recorded as a function of the input current and acetaldehyde concentration using a CMOS-based HD-MEA. NC: negative control, mock-transduced cardiomyocytes; PC: positive control, cells transduced for constitutive BMP-2 expression. (c) HD-MEA-based analysis of the electrogenic behaviour of NRCs engineered for electro-inducible acetaldehyde-responsive BMP-2 expression. The dataset shown as example was recorded at a direct input current of 50 mA corresponding to a beating frequency of 2.1 Hz. Detailed activation map illustrating the average signal shape of the 121 selected electrodes during 10 s. Average signal shape over all 121 channels. Raster plot showing a dot for each contraction on each channel over time. Inter-burst interval or inter-beat interval used to calculate the average beating frequency and beating frequency variation. (d) Zoom-in of two selected bursts/beats on different electrodes. (e) Long signal trace showing the synchronized contraction frequency on five selected electrodes or channels.

Figure 3.

Mammalian cell-based frequency generator. (a) Circuit diagram of the cell-based frequency generator. DC power converts ethanol into acetaldehyde, which dose-dependently triggers expression of the BMP-2 in engineered rat cardiomyocytes (_AIRNRC-BMP-2) and increases the contraction frequency (tachycardia). (b) The beating frequency of cardiomyocytes is recorded as a function of the input current and acetaldehyde concentration using a CMOS-based HD-MEA. NC: negative control, mock-transduced cardiomyocytes; PC: positive control, cells transduced for constitutive BMP-2 expression. (c) HD-MEA-based analysis of the electrogenic behaviour of NRCs engineered for electro-inducible acetaldehyde-responsive BMP-2 expression. The dataset shown as example was recorded at a direct input current of 50 mA corresponding to a beating frequency of 2.1 Hz. Detailed activation map illustrating the average signal shape of the 121 selected electrodes during 10 s. Average signal shape over all 121 channels. Raster plot showing a dot for each contraction on each channel over time. Inter-burst interval or inter-beat interval used to calculate the average beating frequency and beating frequency variation. (d) Zoom-in of two selected bursts/beats on different electrodes. (e) Long signal trace showing the synchronized contraction frequency on five selected electrodes or channels.

Figure 4.

Design and characterization of the mCPU. (a) Miniaturized input device. The input device consists of a PVDF hollow fiber membrane filled with an ethanol (17%, v/v)-containing hydrogel (16.7 µl) connected to platinum electrodes at both ends. (b) mCPU validation. The mCPU consisting of the miniaturized input device, a CPU (Figure 1a) and an enzymatic–optical output device (Figure 1a) was exposed to increasing DC for 1 h and the resulting electric output signal was scored after 24 h.