Macroscopic control of cell electrophysiology through ion channel expression
Curation statements for this article:-
Curated by eLife
eLife assessment
This manuscript will be of interest to those working on non-neuronal bioelectricity, particular synthetic biologists and bioengineers. The primary contribution is the ability to leverage engineered gene circuits to control cellular membrane potential. We find issue, however, with the presentation of the data in this work as electrical communication since the synchronous behavior largely arises from external chemical stimuli.
This article has been Reviewed by the following groups
Listed in
- Evaluated articles (eLife)
- Synthetic Biology (eLife)
Abstract
Cells convert electrical signals into chemical outputs to facilitate the active transport of information across larger distances. This electrical-to-chemical conversion requires a tightly regulated expression of ion channels. Alterations of ion channel expression provide landmarks of numerous pathological diseases, such as cardiac arrhythmia, epilepsy, or cancer. Although the activity of ion channels can be locally regulated by external light or chemical stimulus, it remains challenging to coordinate the expression of ion channels on extended spatial–temporal scales. Here, we engineered yeast Saccharomyces cerevisiae to read and convert chemical concentrations into a dynamic potassium channel expression. A synthetic dual-feedback circuit controls the expression of engineered potassium channels through phytohormones auxin and salicylate to produce a macroscopically coordinated pulses of the plasma membrane potential. Our study provides a compact experimental model to control electrical activity through gene expression in eukaryotic cell populations setting grounds for various cellular engineering, synthetic biology, and potential therapeutic applications.
Article activity feed
-
-
Author Response
Reviewer #1 (Public Review):
In this manuscript the authors describe an approach for controlling cellular membrane potential using engineered gene circuits via ion channel expression. Specifically, the authors use microfluidics to track S. cerevisiae gene expression and plasma membrane potential (PMP) in single cells over time. They first establish a small engineered gene circuit capable of producing excitable gene expression dynamics through the combination of positive and negative feedback, tracking expression using GFP (Figure 1). Though not especially novel or complex, the data quality is high in Figure 1 and the results are convincing. Note that the circuit is excitable and not oscillatory; it is being driven periodically by a chemical inducer. I think the authors could have done a better job justifying the use …
Author Response
Reviewer #1 (Public Review):
In this manuscript the authors describe an approach for controlling cellular membrane potential using engineered gene circuits via ion channel expression. Specifically, the authors use microfluidics to track S. cerevisiae gene expression and plasma membrane potential (PMP) in single cells over time. They first establish a small engineered gene circuit capable of producing excitable gene expression dynamics through the combination of positive and negative feedback, tracking expression using GFP (Figure 1). Though not especially novel or complex, the data quality is high in Figure 1 and the results are convincing. Note that the circuit is excitable and not oscillatory; it is being driven periodically by a chemical inducer. I think the authors could have done a better job justifying the use of an excitable engineered gene circuit system, since you could get a similar result by just driving a promoter with the equivalent time course of inducer.
We restructured the manuscript by presenting the open-loop version of our synthetic circuit and demonstrate that closed loop system integrating feedbacks performs significantly better than its open-loop version (revised Figures 1 and 3). This open-loop system is based on Mar proteins that can synchronizes gene expression on extended spatiotemporal scales (PerezGarcia et al., Nat Comm, 2021). Other driven systems (i.e., TetR, AraC, LacI) can temporally synchronize gene expression in single bacteria cells to successive cycles of inducer. However, over time these bacterial systems build substantial delays in phases between cells, partially due to noise that ultimately led to desynchrony between individual cells even though they tend to follow the common inducer. This is clearly not the case in Mar-based systems (Perez-Garcia et al., Nat Comm, 2021) as eukaryotic cells synchronize to each other under guidance of common environmental stimuli with neglectable phase drift. Furthermore, in revised version we show that dual feedback strategy provides a robust solution to control ion channel expression and associated changes in PMP (see Conclusions lines 231-237).
The authors then use a similar approach to produce excitable expression of the bacterial ion channel KcsA, tracking membrane voltage using the voltage-sensitive dye ThT rather than GFP fluorescence (Figure 2). The experimental results in this figure are more novel as the authors are now using the expression of a heterologous ion channel to dynamically control plasma membrane potential. While fairly convincing, I think there are a few experimental controls that would make these results even more convincing. It is also unclear why the authors are now using power spectra to display observed frequencies compared to the much more intuitive histograms used in Figure 1.
Now we use violin plots with period distributions consistently in all figures to ease the comparison between scenarios.
Finally, the authors move on to use a similar excitable engineered gene circuit approach to produce inducible control of the K1 toxin which influences the native potassium channel TOK1 rather than the heterologous ion channel KcsA (Figure 3). I have a similar reaction to this figure as with Figure 2: the results are novel and interesting but would benefit from more experimental controls. Additionally, the image data shown in Figure 3b is very unclear and could be expanded and improved.
In revised version we have decided to remove K1 toxin data as we are aware that we cannot modulate K1 degradation rate due to its extracellular nature. Instead, we have decided to perform additional experiments in which we directly plugged our circuit to TOK1 native potassium channel to demonstrate that our feedback-integrating synthetic circuit is capable of controlling TOK1 dosage and associated PMP changes (revised Figure 3, and lines 209-220). We believe these new data make more direct connection between synthetic circuits phytohormones and native channel expression than presented earlier K1-based scenario.
Overall, in my opinion the claims in the abstract and title are a bit strong. I would deemphasize global coordination and "synchronous electrical signaling" since the authors are driving a global inducer. To make the claim of synchronous signaling I would want to see spatial data for cells near vs. far from K1 toxin producing cells in Figure 3 along with estimates of inducer/flow timescale vs. expression/diffusion of K1 toxin. As I read the manuscript, I see that most of the synchronicity comes from the fact that all cells are experiencing a global inducer concentration.
We agree with the Reviewer, synchronicity and global coordination comes from phytohormone sensing feedback circuit that is guided by cyclic environmental changes. We have revised definition of synchronous signaling as suggested, focusing on the macroscopic synchronization of ion channel expression achieved by external modulation, which is the key message coming from this work.
Reviewer #2 (Public Review):
The authors present a novel method to induce electrical signaling through an artificial chemical circuit in yeast which is an unconventional approach that could enable extremely interesting, future experiments. I appreciate that the authors created a computer model that mathematically predicts how the relationship between their two chemical stimulants interact with their two chosen receptors, IacR/MarR, could produce such effects. Their experimental validations clearly demonstrated control over phase that is directly related to the chemical stimulation. In addition, in the three scenarios in which they tested their circuit showed clear promise as the phase difference between spatially distant yeast communities was ~10%. Interestingly, indirect TOK1 expression through K1 toxin gives a nice example of inter-strain coupling, although the synchronization was weaker than in the other cases. Overall, the method is sound as a way to chemically stimulate yeast cultures to produce synchronous electrical activity. However, it is important to point out that this synchronicity is not produced by colony-colony communication (i.e., self-organized), but by a global chemical drive of the constructed gene-expression circuit.
The greatest limitation of the study lies in the presentation (not the science). There are two significant examples of this. First, the authors state this study 'provides a robust synthetic transcriptional toolbox' towards chemo-electrical coupling. In order to be a toolbox, more effort needs to be put into helping others use this approach. However little detail is given about methodological choices, circuit mechanisms in relation to the rest of the cell, nor how this method would be used outside of the demonstrated use case. Second, the authors stress that this method is 'non-invasive', but I fail to see how the presented methodology could be considered non-invasive, in in-vivo applications, as gene circuits are edited and a reliable way to chemically stimulate a large population of cells would be needed. It may be that I misunderstood their claim as the presentation of method and data were not done in a way that led to easy comprehension, but this needs to be addressed specifically and described.
We apologize Reviewer for a potential misunderstanding. By ‘non-invasive’ we meant that such systems would not need, for instance, the surgical installation of light components to control ion channel activity. Nonetheless, we have removed these confusing sentences from the revised manuscript.
The rational for using Mar-based system with feedback strategy data has been now presented in more structured and comprehensive way across the revised manuscript to demonstrate benefits from integrating feedback as well as potential of such systems for excitable dynamics with noise-filtering capability and faster responsiveness. We also show how system can be coupled to native potassium channels, opening ways to integrate synthetic circuit into other organisms.
In terms of classifying the synchronicity, while phase difference among communities was the key indicator of synchronization, there were little data exploring other aspects of coupled waveforms, nor a discussion into potential drawbacks. For example, phase may be aligned while other properties such as amplitude and typical wave-shape measures may differ. As this is presented as a method meant for adoption in other labs, a more rigorous analytical approach was expected.
In the revised manuscript, we have analyzed synchronicity using several different approaches:
(1) we calculate cumulative autocorrelations of response between communities.
(2) to complement autocorrelation analysis, we developed a quantitative metric of ‘synchrony index’ defined as 1 - R where R is the ratio of differences in subsequent ThT peak positions amongst cell communities (phase) to expected period. This metrics describes how well synchronized are fungi colonies with each other under guidance of the common environmental signal.
(3) we analyzed amplitudes and peak widths for all presented scenarios and we conclude that while periods and peak widths are robust across communities there is noticeable variation in amplitudes (i.e. Figure 3E).
We therefore believe that this multistep quantitative approach is rigorous in identifying oscillatory signal characteristics.
Reviewer #3 (Public Review):
We are enthusiastic about this paper. It demonstrates controlled expression of ion channels, which itself is impressive. Going a step further, the authors show that through their control over ion channel expression, they can dynamically manipulate membrane potential in yeast. This chemical to electrophysiological conversion opens up new opportunities for synthetic biology, for example development of synthetic signaling systems or biological electrochemical interfaces. We believe that control of ion channel expression and hence membrane potential through external stimuli can be emphasized more strongly in the report. The experimental time-lapse data were also high quality. We have two major critiques on the paper, which we will discuss below.
First, we do not believe the analyses used supports the authors' claims that chemical or electrical signals are propagating from cell-to-cell. The text makes this claim indirectly and directly. For example, in lines 139-141, the authors describe the observed membrane potential dynamics as "indicative of the effective communication of electrical messages within the populations". There are similar remarks in lines 144 and 154-156. The claim of electrical communication is further established by Figure 2 supplement 3, which is a spatial signal propagation model. As far as we can tell, this model describes a system different from the one implemented in the paper.
Second, it is not clear why the excitable dynamics of the circuit are so important or if the circuit constructed does in fact exhibit excitable dynamics. Certainly, the mathematical model has excitable dynamics. However, not enough data demonstrates that the biological implementation is in an excitable regime. For example, where in the parameter space of Figure 1 supplement 1 does the biological circuit lie? If the circuit has excitable dynamics, then the authors should observe something like Figure 1 supplement 1B in response to a nonoscillating input. Do they observe that? Do they observe a refractory period? Even if the circuit as constructed is not excitable, we don't think that's a major problem because it is not central to what we believe is the most important part of this work - controlling ion channel expression and hence membrane potential with external chemical stimuli.
We thank Reviewer for encouraging comments and positive evaluation of our work.
-
eLife assessment
This manuscript will be of interest to those working on non-neuronal bioelectricity, particular synthetic biologists and bioengineers. The primary contribution is the ability to leverage engineered gene circuits to control cellular membrane potential. We find issue, however, with the presentation of the data in this work as electrical communication since the synchronous behavior largely arises from external chemical stimuli.
-
Reviewer #1 (Public Review):
In this manuscript the authors describe an approach for controlling cellular membrane potential using engineered gene circuits via ion channel expression. Specifically, the authors use microfluidics to track S. cerevisiae gene expression and plasma membrane potential (PMP) in single cells over time. They first establish a small engineered gene circuit capable of producing excitable gene expression dynamics through the combination of positive and negative feedback, tracking expression using GFP (Figure 1). Though not especially novel or complex, the data quality is high in Figure 1 and the results are convincing. Note that the circuit is excitable and not oscillatory; it is being driven periodically by a chemical inducer. I think the authors could have done a better job justifying the use of an excitable …
Reviewer #1 (Public Review):
In this manuscript the authors describe an approach for controlling cellular membrane potential using engineered gene circuits via ion channel expression. Specifically, the authors use microfluidics to track S. cerevisiae gene expression and plasma membrane potential (PMP) in single cells over time. They first establish a small engineered gene circuit capable of producing excitable gene expression dynamics through the combination of positive and negative feedback, tracking expression using GFP (Figure 1). Though not especially novel or complex, the data quality is high in Figure 1 and the results are convincing. Note that the circuit is excitable and not oscillatory; it is being driven periodically by a chemical inducer. I think the authors could have done a better job justifying the use of an excitable engineered gene circuit system, since you could get a similar result by just driving a promoter with the equivalent time course of inducer. The authors then use a similar approach to produce excitable expression of the bacterial ion channel KcsA, tracking membrane voltage using the voltage-sensitive dye ThT rather than GFP fluorescence (Figure 2). The experimental results in this figure are more novel as the authors are now using the expression of a heterologous ion channel to dynamically control plasma membrane potential. While fairly convincing, I think there are a few experimental controls that would make these results even more convincing. It is also unclear why the authors are now using power spectra to display observed frequencies compared to the much more intuitive histograms used in Figure 1. Finally, the authors move on to use a similar excitable engineered gene circuit approach to produce inducible control of the K1 toxin which influences the native potassium channel TOK1 rather than the heterologous ion channel KcsA (Figure 3). I have a similar reaction to this figure as with Figure 2: the results are novel and interesting but would benefit from more experimental controls. Additionally, the image data shown in Figure 3b is very unclear and could be expanded and improved.
Overall, in my opinion the claims in the abstract and title are a bit strong. I would de-emphasize global coordination and "synchronous electrical signaling" since the authors are driving a global inducer. To make the claim of synchronous signaling I would want to see spatial data for cells near vs. far from K1 toxin producing cells in Figure 3 along with estimates of inducer/flow timescale vs. expression/diffusion of K1 toxin. As I read the manuscript, I see that most of the synchronicity comes from the fact that all cells are experiencing a global inducer concentration.
-
Reviewer #2 (Public Review):
The authors present a novel method to induce electrical signaling through an artificial chemical circuit in yeast which is an unconventional approach that could enable extremely interesting, future experiments. I appreciate that the authors created a computer model that mathematically predicts how the relationship between their two chemical stimulants interact with their two chosen receptors, IacR/MarR, could produce such effects. Their experimental validations clearly demonstrated control over phase that is directly related to the chemical stimulation. In addition, in the three scenarios in which they tested their circuit showed clear promise as the phase difference between spatially distant yeast communities was ~10%. Interestingly, indirect TOK1 expression through K1 toxin gives a nice example of …
Reviewer #2 (Public Review):
The authors present a novel method to induce electrical signaling through an artificial chemical circuit in yeast which is an unconventional approach that could enable extremely interesting, future experiments. I appreciate that the authors created a computer model that mathematically predicts how the relationship between their two chemical stimulants interact with their two chosen receptors, IacR/MarR, could produce such effects. Their experimental validations clearly demonstrated control over phase that is directly related to the chemical stimulation. In addition, in the three scenarios in which they tested their circuit showed clear promise as the phase difference between spatially distant yeast communities was ~10%. Interestingly, indirect TOK1 expression through K1 toxin gives a nice example of inter-strain coupling, although the synchronization was weaker than in the other cases. Overall, the method is sound as a way to chemically stimulate yeast cultures to produce synchronous electrical activity. However, it is important to point out that this synchronicity is not produced by colony-colony communication (i.e., self-organized), but by a global chemical drive of the constructed gene-expression circuit.
The greatest limitation of the study lies in the presentation (not the science). There are two significant examples of this. First, the authors state this study 'provides a robust synthetic transcriptional toolbox' towards chemo-electrical coupling. In order to be a toolbox, more effort needs to be put into helping others use this approach. However little detail is given about methodological choices, circuit mechanisms in relation to the rest of the cell, nor how this method would be used outside of the demonstrated use case. Second, the authors stress that this method is 'non-invasive', but I fail to see how the presented methodology could be considered non-invasive, in in-vivo applications, as gene circuits are edited and a reliable way to chemically stimulate a large population of cells would be needed. It may be that I misunderstood their claim as the presentation of method and data were not done in a way that led to easy comprehension, but this needs to be addressed specifically and described.
In terms of classifying the synchronicity, while phase difference among communities was the key indicator of synchronization, there were little data exploring other aspects of coupled waveforms, nor a discussion into potential drawbacks. For example, phase may be aligned while other properties such as amplitude and typical wave-shape measures may differ. As this is presented as a method meant for adoption in other labs, a more rigorous analytical approach was expected.
-
Reviewer #3 (Public Review):
We are enthusiastic about this paper. It demonstrates controlled expression of ion channels, which itself is impressive. Going a step further, the authors show that through their control over ion channel expression, they can dynamically manipulate membrane potential in yeast. This chemical to electrophysiological conversion opens up new opportunities for synthetic biology, for example development of synthetic signaling systems or biological electrochemical interfaces. We believe that control of ion channel expression and hence membrane potential through external stimuli can be emphasized more strongly in the report. The experimental time-lapse data were also high quality. We have two major critiques on the paper, which we will discuss below.
First, we do not believe the analyses used supports the authors' …
Reviewer #3 (Public Review):
We are enthusiastic about this paper. It demonstrates controlled expression of ion channels, which itself is impressive. Going a step further, the authors show that through their control over ion channel expression, they can dynamically manipulate membrane potential in yeast. This chemical to electrophysiological conversion opens up new opportunities for synthetic biology, for example development of synthetic signaling systems or biological electrochemical interfaces. We believe that control of ion channel expression and hence membrane potential through external stimuli can be emphasized more strongly in the report. The experimental time-lapse data were also high quality. We have two major critiques on the paper, which we will discuss below.
First, we do not believe the analyses used supports the authors' claims that chemical or electrical signals are propagating from cell-to-cell. The text makes this claim indirectly and directly. For example, in lines 139-141, the authors describe the observed membrane potential dynamics as "indicative of the effective communication of electrical messages within the populations". There are similar remarks in lines 144 and 154-156. The claim of electrical communication is further established by Figure 2 supplement 3, which is a spatial signal propagation model. As far as we can tell, this model describes a system different from the one implemented in the paper.
Second, it is not clear why the excitable dynamics of the circuit are so important or if the circuit constructed does in fact exhibit excitable dynamics. Certainly, the mathematical model has excitable dynamics. However, not enough data demonstrates that the biological implementation is in an excitable regime. For example, where in the parameter space of Figure 1 supplement 1 does the biological circuit lie? If the circuit has excitable dynamics, then the authors should observe something like Figure 1 supplement 1B in response to a non-oscillating input. Do they observe that? Do they observe a refractory period? Even if the circuit as constructed is not excitable, we don't think that's a major problem because it is not central to what we believe is the most important part of this work - controlling ion channel expression and hence membrane potential with external chemical stimuli.
-
-