Current Response

Overview of Characterization in Bioelectrochemical Systems

As described in the Chassis section, S. oneidensis MR-1 is capable of shuttling electrons through the Mtr pathway to reduce extracellular metals because of the negative free energy change associated with these redox reactions. Thus, to encourage Shewanella to transfer electrons to an electrode, we poise the potential of an electrode in a three electrode system, controlled by a potentiostat, so that electron transfer is energetically favorable to the organism. In short, a potentiostat works by setting the potential of a working electrode (WE) with respect to a Ag/AgCl reference electrode (RE) by injecting current through a counter electrode (CE). (To read about how we made our own field-deployable potentiostats, read our drylab components page.) These electrodes can be seen in the schematic representation of our single-compartment bioelectrochemical reactors shown to the right.

Because we are interested in continuous monitoring of contaminants, we characterized the current response to analyte of all engineered strains by operating reactors in continuous flow setup such that steady-state current output could be reached. In general, all experiments were performed in bench-scale reactors provided by the Angenent Lab—with a constant fluid volume of 120 mL and a consistent electrode-surface area. All characterization experiments began at an analyte concentration of zero, while media was fed to the system at a constant rate of 18 mL/min. Once reached steady state was reached—for a period of greater than three system retention times—the analyte concentration in the feed tank was increased so that a new steady state—corresponding to the higher analyte concentration—could be reached. By iteratively repeating this process, we were able to characterize the current response of our reporter strains to either arsenic-containing compounds or naphthalene, as appropriate.

After initial characterization of our arsenic- and naphthalene-sensing strains, we have shown that our arsenic sensor works as expected, producing higher levels of steady-state current in response to arsenite salts. However, more trials are needed in order to construct a reliable calibration curve.

First Lessons Learned: Control Reactors

In order to enhance the field-deployability of our final device, we initially decided to feed our reactors with LB, since a very concentrated LB source fed at a low flow rate could sustain our field reactors for extended periods of time without taking up much physical space. However, upon setting up control reactors—both in batch and continuous flow operation—we discovered that wild type S. oneidensis MR-1 produced significantly less current when fed with LB than M4—a commonly used media for Shewanella-inoculated bioelectrochemical systems, as illustrated for batch operation by the figure below.
Fig. 1. Current production over time is plotted for batch reactors inoculated with wildtype S. oneidensis MR-1 growing on M4 media (blue) and LB media (green). Maximum current production from M4-fed Shewanella is much greater than that from LB-fed.

When operated in a continuous flow setup, we observed the steady state current production from an LB fed reactor to be within the background noise of the setupM4—i.e., an un-incoluated reactor was indistinguishable from a reactor inoculated with wildtype S. oneidensis MR-1. Because of this, we chose to use M4 media for all future characterization experiments, since optimization of signal-to-noise ratio is essential in the development of any sensing system.

Arsenic Sensing

We chose to initially characterize our arsenic-sensing Shewanella by dosing with sodium arsenite, since the organism's native arsenate reductase activity would introduce a confounding variable in part characterization. After this initial characterization , we have shown that current is, indeed, upregulated in response to our analyte of interest. However, we should emphasize that this data is preliminary: Because of the care we took to establish a thorough Standard Operation Procedure for the handling of arsenic, we only had time for a single characterization trial, shown below in Fig. 2. Additionally, we plan on characterizing in response to both arsenate salts (to determine whether arsenate reductase activity is indeed confounding) and antimonite (to estimate the likelyhood of false positives).
Fig. 3. Current production over time is plotted for continuous flow reactors inoculated with our arsenic-sensing strain (JG700+p14k). Transient phases corresponding to arsenite concentrations of 100 μM (blue) and 500 μM (green)are shown. Before a potentiostat channel restart at time zero, basal current was recorded at approximately 4 μA. We report induced current in percent of this basal response.

It is also important to point out that we have not normalized any data to a per-cell basis. Because we are interested in a field-deployable system that produces greater absolute current in response to analyte, we did not record either optical density or colony forming units over time. However, while not a rigorous, we have observed a visible decrease in biomass for increasing concentrations of arsenite—as would be expected. Therefore, it is likely that an a per-cell basis, our arsenic sensing strains produce much more current that Fig. 2 would suggest. We plan on repeating characterization experiments in response to arsenite—both to generate a reliable transfer function relating current to arsenite concentration and to better understand the relationship between specific growth rate and MtrB production as a function of analyte concentration.

Naphthalene & Salicylate Sensing

Because we did not have time to confirm successful conjugation of our naphthalene-degrading plasmid into Shewanella, we focused our efforts on observing the response to salicylate of our naphthalene-sensing precursor strain (JG700 + pSAL). This strain contains the molecular machinery requisite to respond to salicylate, which—in our final system—is a metabolite produced as a result of naphthalene catabolism via the enzymes encoded by the nah operon. (To read more about our naphthalene sensor design, see our DNA Assembly page.)
Fig. 3. Current production over time is plotted for continuous flow reactors inoculated with our salicylate reporter strain (blue) and wildtype S. oneidensis MR-1 (green). Both duration of transient period and value of saturating current are approximately equal for reactors corresponding to both strains.

As seen above, our un-induced salicylate sensing strain produces approximately the same steady state current as wildtype Shewanella, suggesting that leaky expression of MtrB from uninduced promoter activity is enough to saturate current output. This can be understood in terms of the complete Mtr pathway, since metal-reduction activity requires a functional complex of MtrA, MtrB, and MtrC—as described in the Chassis section page. Since there is only so much MtrC and MtrA in the cell, increasing MtrB activity saturates as free MtrC and MtrA is sequestered and localized.

Ferrozine Assays

We are also in the process of conducting ferrozine assays to demonstrate the functionality of our reporter strains at the system level. This assay quantifies the reduction of iron (III) to iron (II), an energetically favorable process for Shewanella expressing the Mtr pathway. We will test each of our reporter strains at various analyte concentrations, incubated with iron(III) in an anaerobic environment for extended periods of time. Increased reduction of iron (III) is a proxy for increased current production in bioreactors, and can thus be used as a reliable indicator that our sensing strains are functioning at the system level.