Compositions and methods of bioelectronic sensing

Abstract

This invention is directed to compositions and methods of bioelectronic sensing.

Description

COMPOSITIONS AND METHODS OF BIOELECTRONIC SENSING

[0001] For countries that permit incorporation by reference, all patents, patent applications and publications cited in this disclosure are hereby incorporated by reference in their entireties. In addition, any manufacturers’ instructions or catalogues for any products cited or mentioned herein are incorporated by reference. Documents incorporated by reference into this text, or any teachings therein, can be used in the practice of the present invention. Documents incorporated by reference into this text are not admitted being prior art.

[0002] This patent disclosure contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves any and all copyright rights.CROSS-REFERENCE TO RELATED APPLICATIONS

[0003] This application claims the benefit of U.S. Provisional Application No. 63 / 828,835, entitled “COMPOSITIONS AND METHODS OF BIOELECTRONIC SENSING”, filed June 23, 2025, and U.S. Provisional Application No. 63 / 682,083, entitled “COMPOSITIONS AND METHODS OF BIOELECTRONIC SENSING”, filed August 12, 2024. Each of these provisional applications is incorporated herein by reference in its entirety.GOVERNMENT INTERESTS

[0004] This invention was made with government support under W91 INF-22-1-0239 awarded by the Army Research Laboratory - Army Research Office. The government has certain rights in the invention.FIELD OF THE INVENTION

[0005] This invention is directed to biosensors and methods of use thereof.BACKGROUND OF THE INVENTION

[0006] Microbial extracellular electron transfer (EET) has been engineered by synthetic biology to develop bioelectronic sensors that can detect environmental stimuli and generate electrical signals. However, previous inventions have been restricted to a few EET model organisms and rely on complex engineering strategies, limiting the application scope of bioelectronic sensors. Moreover, while most biosensors can detect specific analytes, they oftenfail to distinguish between structural analogs of these analytes. Such examples include quinone and its derivatives, which play important roles as pharmaceuticals but can be toxic if overdosed. Thus, a sensitive and fast strategy is needed to selectively sense quinone derivatives for drug monitoring.SUMMARY OF THE INVENTION

[0007] Aspects of the invention are drawn towards an electroactive co-culture sensing system (e'COSENS) comprising a sender strain, a receiver strain, and a bioelectrochemical measurement apparatus, wherein the sender strain is a quinone-producing bacterium, and wherein the receiver strain is a quinone-dependent extracellular electron transfer (EET)- capable bacterium. In embodiments, the bioelectrochemical measurement apparatus comprises a bioelectrochemical cell and an electronic device for voltage and current monitoring. In embodiments, the sender strain is engineered to delete menD and is transformed with the menD gene on a plasmid with a stimulus-responsive genetic element. In embodiments, the sender strain is engineered further delete menA, ndh, or a combination thereof. In embodiments, the stimulus-responsive genetic element comprises an inducible transcription factor and an inducible promoter. In embodiments, the inducible transcription factor comprises leiR, arsR, oxyR, nisR and nisK, araCAMand araE, cymRAM, or a combination thereof. In embodiments, the promoter is selected from the group consisting of Pxyi / 2xteto, Panoci, Poxys, PntsA, PBAD, PqwRC, or a combination thereof. In embodiments, the receiver strain is genetically engineered to delete menA, ndh I, or a combination thereof. In embodiments, the receiver strain is a wild-type strain. In embodiments, the quinone-producing bacterium produces 1 ,4-dihyroxy-2 -naphthoic acid (DHNA). In embodiments, the sender strain is selected from the group consisting of Escherichia coll, Bacillus subtills, Vibrio natriegens, Shewanella oneidensls, Leuconostoc citreum, Staphylococcus aureus, Enterococcus faecalis, a strain from Example 6, or a combination thereof. In embodiments, the receiver strain is Lactiplantibacillus plantarum, a strain from Example 7, or a combination thereof.

[0008] Aspects of the disclosure are drawn towards a method of producing a co-culture sensing system (e'COSENS), comprising: obtaining a quinone-producing bacterium, and a quinone-dependent extracellular electron transfer (EET)-capable bacterium receiver strain; genetically modifying quinone-producing bacterium, thereby producing a sender strain; coculturing the sender strain and receiver strain, thereby producing a co-culture; and introducing the co-culture into a bioelectrochemical measurement apparatus, thereby producing a co- culture sensing system (e'COSENS). In embodiments, genetically modifying a quinone-producing bacterium comprises menD and transforming with a plasmid containing the menD gene with a stimulus-responsive genetic element. In embodiments, genetically modifying a quinone-producing bacterium further comprises deleting menA, ndh, or a combination thereof. In embodiments, the stimulus-responsive genetic element comprises an inducible transcription factor and an inducible promoter. In embodiments, the inducible transcription factor comprises tetR, arsR, oxyR, nisR and nlsK, a.raCAMand araE, cymRAM, or a combination thereof. In embodiments, the promoter comprises Pxy] / 2xteto, Parsocc, Prays, PimA, PBAD, Pcmxc, or a combination thereof. In embodiments, the method further comprises genetically modifying the quinone-dependent extracellular electron transfer (EET)-capable bacterium. In embodiments, genetically modifying the quinone-dependent EET-capable bacterium comprises deleting menA, ndhl, or a combination thereof. In embodiments, the quinone-producing bacterium produces l,4-dihyroxy-2-naphthoic acid (DHNA). In embodiments, the bioelectrochemical measurement apparatus comprises a bioelectrochemical cell and an electronic device for voltage and current monitoring. In embodiments, the sender strain is selected from the group consisting of Escherichia coll, Bacillus subtilis, Vibrio natriegens, Shewanella oneidensis, Leuconostoc citreum, Staphylococcus aureus, Enterococcus faecalis , a strain from Example 6, or a combination thereof. In embodiments, the receiver strain is selected from the group consisting ofLactiplantibacillus plantarum, a strain from Example 7, or a combination thereof.

[0009] In embodiments, tire e'COSENS comprises a single chamber, miniature microbial fuel cell.

[0010] Aspects of the disclosure are drawn towards a bioelectronic sensing system capable of translating environmental stimuli into electrical signals produced by a method described herein.

[0011] Another aspect of the present disclosure is drawn to a bioelectrochemical measurement apparatus comprising a miniature, single-chamber microbial fuel cell (MFC). In embodiments, the miniature single-chamber MFC comprises a clay-based proton exchange membrane (PEM). The clay-based PEM can comprise two opposing faces, and the clay-based PEM is disposed between an anode on one face and a cathode on the opposite face. The claybased PEM can comprise vermiculite clay. In embodiments, the clay-based PEM comprises an atomically thin 2D flake.

[0012] In certain embodiments of the bioelectrochemical measurement apparatus the anode comprises carbon. The anode can comprise a porous carbon fell material in contact with carbon ink that is disposed upon a polyethylene terephthalate sheet. In embodiments, the polyethylene terephthalate sheet comprises a hole, hole, gap, or passthrough that forms a well, and the porouscarbon felt is disposed within the well. In exemplary embodiments, the carbon felt is hydrophilic.

[0013] In embodiments of the bioelectrochemical measurement apparatus, the cathode comprises Ag / AgCl. The bioelectrochemical measurement apparatus can comprise a digital multimeter for voltage and current monitoring.

[0014] In embodiments, the miniature MFC comprises a width that is less than about 3.0 mm and a thickness that is less than about 0.5 cm.

[0015] Any of the e'COSENS or methods of the present disclosure can comprise any of the various bioelectrochemical measurement apparatuses disclosed herein.

[0016] Other objects and advantages of this invention will become readily apparent from the ensuing description.BRIEF DESCRIPTION OF THE FIGURES

[0017] FIG. 1 shows an exemplary schematic data and schematics showing a tailored EET pathway in L. plantarum improves signal orthogonality in response to DHNA. (Panel A) Schematic showing enzymatic reactions stimulated by environmental DHNA. Red arrows show that Ndh2 -associated EET uses DHNA as electron shuttles. Dash arrows indicate the exemplary side reactions that can occur when DHNA is present. (Panels B-C) Comparing the EET activity of / ,. plantarum wild type (WT), side reactions removed mutant (\dmkA\ndhl\ or kdmkAkndhl / 2 mutant by either (Panel B) iron reduction or (Panel C) anode reduction. DHNA (5 pM) was added in all cases. For anode reduction, the electrode was polarized at +0.2 V v.s. Ag / AgCl. Aj per cell represents the delta current density normalized to CFU. Aj of medium control was also divided by the mean CFU to bring values to the same magnitude for comparison. Data represent mean ± 1 s.d. for n = 3 (Panel B) or n = 2 (Panel C) biological replicates. Significant differences were determined by one-way ANOVA with Tukey’s test, ***p < 0.001, ****p < 0.0001.

[0018] FIG. 2 shows a non-limiting exemplary data and graphs indicating EET of L. plantarum shows high sensitivity and dynamic range when responding to environmental DHNA. (Panel A) Scanned image showing iron reduction levels when different density of cells was incubated with 0-500 pM DHNA for 12 h. The purple color represents Fe2+concentration in each well. Medium control indicates DHNA-induced abiotic iron reduction. (Panel B) Quantitative analysis of iron reduction levels in (Panel A). Biotic iron reduction of each well was calculated by subtracting abiotically produced Fe2' from biotically produced Fe2+and normalizing to CFU of the inoculated cell. Lines are guides to the eye. The inset shows anenlarged region for 0-5 pM DHNA. Data represent mean ± 1 s.d. for n = 3 biological replicates (Panel C) Iron reduction rate per cell (i.e. initial velocity of iron reduction from 0 h to 12 h normalized to CFU, see also Fig. 7) calibrated to DHNA concentration. Data were fitted into the Michaeiis-Menten equation. The inset shows an enlarged region for 0-5 uM DHNA.

[0019] FIG. 3 shows non-limiting, exemplary data indicating L. plantarum rapidly responds to DHNA with dose-dependent temporal dynamics. (Panel A) Current density produced by L. plantarum kdmkAbndhl strain upon exposure to different concentrations of DHNA in bioelectrochemical systems (BESs) within 24 h (right panel). An enlarged region showing rapid current response within 5 min (left panel). The black arrow' indicates DHNA injection. Data represent mean ± 1 s.d. for n = 2 biological replicates. (Panel B) Confidence to differentiate varied DHNA concentrations as a function of time. Confidence was determined by one-way ANOVA for current densities at each single time point. The inset show’s the calibration curve between current density and DHNA concentrations at 21 s. The data in the inset were fitted into the Michaeiis-Menten equation. (Panel C) Comparison of dose-dependent temporal dynamics in response to DHNA when tested in CDM or PBS. The hollow triangles represent the comparison of absolute level of current density (dQ / dt per cm2). The black dots represent the comparison of average current density (AQ / At per cm2). Insets showcase the comparison of average current density at 5 min, 5 h, 12 h, and 24 h. Each dot in the insets represents a different DHNA concentration. The mean data of n = 2 biological replicates for both CDM and PBS were used for comparison. (Panel D) Current density produced by L. plantarum upon exposure to stepwise increasing concentrations of DHNA. Medium control indicates the abiotic current upon DHNA injection. Data represent mean ± 1 s.d. for n = 2 biological replicates.

[0020] FIG. 4 show's non-limiting, exemplary schematics and data indicating Ndh2- dependent EET discriminates quinone structural analogs. (Panel A) Illustration showing Ndh2- dependent EET of L. plantarum selectively senses specific quinone derivatives from an environmental quinone pool. (Panel B) Current density produced by L. plantarum &dmkAt\ndhl or kdmkAkndhl / 2 mutant compared to medium control w’hen 1 pM of hydroquinone or 1,4-NQ derivatives w'ere added stepw'ise. The next compound was not injected until the current of the prior compound stabilized. Data represent mean ± 1 s.d. for n = 2 biological replicates. (Panel C) The differential current density w'as calculated by subtracting the mean Aj per cell of MmkAkndhl / 2 strain from that of \dmkA\ndhl strain. The shaded error bar represents propagated error. (Panel D) Differential average current density produced by L. plantarum w'hen 1 pM of DHNA, menadione, or both were added to the background ofinert quinone derivatives (hydroquinone, 1,4-NQ, juglone, DCDMNQ). Inert quinone derivatives were added at 1 pM in CDM and 0.01 pM in FaSSIF. Data represent mean ± propagated error for n = 2 biological replicates. (Panel E) Confidence to distinguish between DHNA, menadione, or both as a function of time. Confidence was determined by one-way ANOVA on data shown in (Panel C).

[0021] FIG. 5 shows non-limiting, exemplary data and schematics indicating the mechanism of L. plantarum as a selective quinone bioelectronic sensor. (Panel A) A four-step model denoting the reaction between quinone and Ndh2 of L. plantarum. Below each step shows the exemplary properties of a quinone that would determine whether this quinone will be selected for Ndh2-dependent EET. (Panel B) L. plantarum as a bioelectronic sensor to sense environmental quinones, process physicochemical information, and selectively output electrical signals in response to quinones that favor EET. (Panel C) Indicated L. plantarum Ndh2 i-409 structure, simulated quinone-Ndh2 binding, and the quinone binding pocket. (Panel D) Correlation between normalized EET and the quinone score. EQ / Q_ , quinone 1 e redox potential; , indicated binding energy between quinone and Ndh2; (cLogPo - cLogPo ), differential of the calculated partition coefficient between quinone and semiquinone. (Panel E) Lawsone (2-hydroxy-l,4-naphthoquinone) has an indicated quinone score greater than the negative control but smaller than DHNA. Without wishing to be bound by theory, lawsone can be EET positive. (Panel F) Differential average current density generated by L. plantarum when reacting with 1 pM menadione, DHNA, lawsone, or hydroquinone. Data represent mean ± propagated error for n = 2 biological replicates.

[0022] FIG. 6 shows non-limiting, exemplary data and schematics of L. plantarum in electrochemical systems. (Panel A) Microscope image of L. plantarum NCIMB8826. (Panel B) Schematic and image of the two-chamber bioelectrochemical system used to analyze the electroactivity of L. plantarum.

[0023] FIG. 7 shows non-limiting, exemplary graphs showing time-course iron reduction and calculation of the initial iron reduction rate. Panels (A-D) Concentration of Fe2produced by 0.25, 0.5, 1, or 2 OD of cells was measured at 0, 2. 4, 6, 8, 12 h after cells were mixed with iron(III) oxide and different concentrations of DHNA. To calculate the initial iron reduction rate, data of 0 h and 2 h were omitted for all DHNA concentrations and for all cell densities to improve the accuracy of linear regression. This is because the Fe2+concentrations were below the detection threshold at these two time points. The rest of the data were fitted by linearregression and the slope value was treated as the iron reduction rate. Data represent mean ± 1 s.d. for n=3 biological replicates.

[0024] FIG. 8 shows a non-limiting, exemplary’ data indicating a correlation between DHNA concentration and the increasing rate of current density. (Panel A) The current density- produced by L. plantarum kdmkAkndhl strain within 0 - 2 h in response to different concentrations of DHNA (based on the data shown in Fig. 3) was fitted by linear regression. The solid lines represent the result of linear regression. The slope value indicates the increasing rate of current density. (Panel B) The increasing rate of current density as a function of DHNA concentrations. The line is a guide to the eye. (Panel C) In the DHNA continuous sensing experiment shown in Fig. 3 Panel D, the linear regression of current density was performed on each adding interval. The increasing rate of current density was plotted against the accumulated DHNA concentration. The line is a guide to the eye. Data represent mean ± 1 s.d. for n=2 biological replicates.

[0025] FIG. 9 shows a non-limiting, exemplary data indicating L. plantarum responds to DHNA in PBS. (Panel A) Current density produced by L. plantarum \dmkA\ndhl strain upon exposure to different concentrations of DHNA with PBS as the assay buffer. The black arrow indicates DHNA injection. Data represent mean ± 1 s.d. for n=2 biological replicates. (Panel B) An enlarged region of (Panel A) shows a rapid current response within 5 minutes upon exposure to environmental DHNA in PBS. The dash black line indicates the time with 99% confidence (p-value < 0.001) to discriminate different DHNA concentrations. (Panel C) Current density as a function of DHNA concentration at the time when 99% confidence is reached. Data were fitted in the Michaelis-Menten equation.

[0026] FIG. 10 shows a non-limiting, exemplary data showing a comparison of current density produced by L. plantarum in response to DHNA in different media. Current densities at each time point were compared between CDM and PBS and the adjusted R2was reported. Insets show the comparison at 5 inin, 5 h, 12 h, and 24 h. Each dot in the insets represent a different DHNA concentration. The mean data of n=2 biological replicates for both CDM and PBS were used for comparison.

[0027] FIG. 11 shows a non-limiting, exemplary graph showing current spike upon DHNA addition results from abiotic DHNA oxidation. The amplitude of the current spike upon DHNA injection was calculated for the kdmkAbndhl mutant and the medium control. The spike current density was plotted against DHNA concentrations. No difference was observed between these two groups. Data represent mean ± 1 s.d. for n=2 biological replicates.

[0028] FIG. 12 shows non-limiting, exemplary graphs of current density generated by L. plantarum in response to different quinone derivations. (Panel A) The indicated quinone derivatives (1 pM) were injected at 0 h to either \dmkA\ndh 1 or MmkAAndh 1 / 2 strain. The two mutant strains produced differentiated current density when reacting with DHNA or menadione but not with other quinone derivatives. Data represent mean ± 1 s.d. for n=2 biological replicates. (Panel B) Lawsone can be utilized by Ndh2-dependent EET to produce current, in agreement with the model shown in Fig. 5 Panel E. Data represent mean ± 1 s.d. for n=2 biological replicates.

[0029] FIG. 13 shows non-limiting, exemplary data indicating average current density is less dependent of media than current density, allowing data comparison between different media. (Panel A) Data transition from current density (dQ / dt per cm2) to average current density (AQ / At per cm2) in CDM. (Panel B) Data transition from current density to average cunent density in FaSSIF. Data represent mean ± 1 s.d. for n=2 biological replicates.

[0030] FIG. 14 shows non-limiting, exemplary’ indicating Ndh2 structure and comparison. (Panel A) AlphaFold modeled Ndh2 structure of L. plantarum. The cyan region (1-409 aa) shows homology with known crystal structures of Ndh2. The orange region (409-635 aa) shows no homology to known protein. Truncated Ndh2i-409 was used for structural comparison and for AutoDock Vina simulation. (Panel B) Truncated Ndh2i-409 of L. plantarum superimposed upon Ndh2 of Caldalkalibacillus thermarum (PDB: 5KMS). (Panel C) Truncated Ndh2i-409 of L. plantarum superimposed upon Ndil of yeast (PDB: 4G73).

[0031] FIG. 15 shows non-limiting, exemplary cyclic voltammetry measurements of quinone 1 e' or 2 e72 H+redox potential. (Panel A) The 2 e72 H’ redox potential measurements of the represented quinone derivatives in an aqueous solution (MOPS buffer). DCDMNQ is redox irreversible in the aqueous solution despite the two well-separated redox peaks. (Panel B) Quinone 1 e" redox potential measurements in a non-aqueous solution (acetonitrile). Grayshadows indicate the redox peaks corresponding to 1 e" transfer. (Panel C) Quinone undergoes 2 e72 H electron transfer in aqueous environment and successive 1 e’ electron transfer in nonaqueous environment.

[0032] FIG. 16 shows without wishing to be bound by theory, non-limiting, exemplary 1 e' redox potential of DHNA. (Panel A) Summary of Hammett σparaconstants for each 2- substitutions of 1 ,4-naphthoquinone and the le' redox potentials for each derivative. The lc‘ redox potential for 1 ,4-naphthoquinone (2-H) and menadione (2-methyl) were measured in this paper (Fig. 14). The le' redox potentials for other derivatives were reported by Prince et al. (Prince et al., 2022), and were converted to compare with our measurements by a factor of -0.5V. The Hammett oPara constants were then plotted against le’ redox potentials and the le’ redox potential of DHNA (2-carboxyl) was indicated by using the fitted equation. (Panel B) Hammett Opara constants plotted against le’ redox potential of 2-substituted 1 ,4-naphthoquinones showing a linear relationship between these two parameters.

[0033] FIG. 17 show's non-limiting, exemplary relationship between EET and quinone physicochemical properties. Differential average current density plotted against (Panel A) quinone one electron redox potential, (Panel B) quinone two electrons redox potential, (Panel C) Autodock Vina modeled quinone-Ndh2 binding energy, (Panel D) calculated partition coefficient of quinone (cLogPq), (Panel E) calculated partition coefficient of semiquinone (cLogPq-), (Panel F) differential of the cLogP between quinone and semiquinone. NQ, naphthoquinone; HQ, hydroquinone.

[0034] FIG. 18 show's a comparison of techniques for quinone detection.

[0035] FIG. 19 show's a miniature bioelectronic device simplifies signal detection under one non-limiting, exemplary embodiment. A non-limiting, exemplary assembly schematic of the clay-based microbial fuel cell and a comparison of the size between the device and a coin is shown in Panel A. PEM, proton exchange membrane. Electrons generated through the DHNA-mediated EET in e- COSENS can accumulate in carbon felt or move from the carbon anode to the Ag / AgCl cathode via an external circuit, allowing for open circuit voltage (OCV) or short circuit current (SCA) measurements (Panel B). At the cathode, AgCl is reduced, and H+movement through the clay PEM maintains charge neutrality. The workflow' of signal detection (Panel C). Co-culture was inoculated into the sample or media and incubated for 1.5- 2 hours, and a 60 pL portion was then injected into the device for OCV or SCA measurements. OCV and SCA measurements for aTc detection in milk (Panel D). Differential (A) OCV wras reported by subtracting OCV at t = 0 from subsequent OCV values. SCA w as determined by linear sweep voltammetry at voltage = 0 vs. OCV. OCV measurement using a digital multimeter (Panel E), The positive terminal of the multimeter is connected to the carbon anode of the device, and the negative terminal is connected to the Ag / AgCl cathode. OCV measured by the digital multimeter allow s for aTc detection in milk. Data represent five biological replicates across two independent experiments on different days. Error bars in d, e represent mean ± s.e.m. P-values w'ere determined by tw'o-tailed unpaired t-test. **, P < 0.01.

[0036] FIG. 20 show's the clay-based MFC device allows for signal detection from both L. lactis-L. plantarum and E. coli-L. plantarum co-cultures under one exemplary, non-limitingembodiment. Photos of the device from the front, back, and sides (Panel A). Photo showing high-throughput manufacture of the clay-based MFC device (Panel B). Measurements of the open circuit voltage (OCV) and short circuit current (SCA) for EET-capable and incapable L. lactis-L. plantarum (Panel C) or E. coli-L. plantarum (Panel D) co-cultures. The OCV measured by the digital multimeter differentiates the EET-capable and incapable L. lactis-L. plantanim or E. coli-L. plantarum co-cultures (Panel E). Data represent three biological replicates or their mean ± s.e.m. P-values were determined by two-tailed impaired t-test. *, P < 0.05; **, P < 0.01 ; ***, P < 0.001.

[0037] FIG. 21 shows a non-limiting, exemplary mechanism of the electroactive co-culture sensing system (e'COSENS) under one embodiment. The e'COSENS system in this embodiment comprises a “sender” and a “receiver” bacterial strain. The sender synthesizes 1 ,4- dihydroxy-2-naphthoic acid (DHNA, precursor of vitamin K2) from the substrate chorismate through the seven-enzyme Men pathway. The rnenD gene was knocked out from the genome and expressed on plasmids under the control of analyte-inducible promoters to conditionally turn on DHNA biosynthesis for sensing purposes. The menA gene was also knocked out to prevent the prenylation of DHNA by prenyl diphosphate (PP). The secreted DHNA activates extracellular electron transfer (EET) in the receiver strain by mediating electron cycling between the membrane-bound type-II NADH:quinone oxidoreductase Ndh2 and the terminal electron acceptors, such as iron(III) oxide or an electrode. This results in the generation of electrical signals. The probiotic Lactiplantibacillus plantanim serves as the receiver in the exemplary system. DHNA0X: oxidized DHNA; DHNArej: reduced DHNA.

[0038] FIG. 22 shows non-limiting, and exemplary graphs and data indicating that potential sender and receiver candidates are widespread, and cell-free supernatants from phylogenetically distinct bacteria can activate EET in L. plantanim. Panel a shows an exemplary phylogenetic tree of bacteria with a complete DHNA biosynthesis pathway. A full list is provided in Example 6. Panel b shows an exemplary phylogenetic tree of Firmicutes possessing an EET locus24with orange-colored range indicating those with an incomplete or missing DHNA pathway. A full list is provided in Example 7. Panel c shows that Ndli2- dependent EET in L. plantanim can be activated by cell-free supernatants (CFS) from phylogenetically distinct bacteria with glucose, mannitol, or a combination thereof as the carbon source. The chemically defined medium (CDM) and CFS from Caulobacter crescentus (which lacks the DHNA pathway) serve as negative controls. CDM containing 2 uM DHNAsenes as the positive control. Results were reported as the Fe3concentration reduced by L. plantarum ndh2 knockout subtracted from that of ndh2 wild-type. Data represent mean ± s.d. of three biological replicates. P-values were determined by one-way ANOVA with Turkey’s post hoc test. ****, P < 0.0001 vs. CDM media control; ns, not significant.

[0039] FIG. 23 Shows non-limiting, and exemplary graphs and data indicating that coculturing L. plantarum with Lactococcus lactis or Escherichia coll produces electrical signals. Panel a provides a schematic showing L. lactis-L. plantarum co-culture reduces extracellular electron acceptors, such as iron(III) oxide or an electrode. In this exemplary embodiment, cocultured L. lactis and L. plantarum can reduce Fe3' to Fe2+(Panel b) and generate current in bioelectrochemical reactors (Panel c). Monocultures or co-cultures with mutants deficient in DHNA biosynthesis (L. lactis fanenD) and EET (Z. plantarum Endh2') show diminished electroactivity (Panel d). In this exemplary embodiment, E. coli-L. plant arum co-culture reduces extracellular electron acceptors, such as iron(III) oxide or an electrode (Panel e, f). In this embodiment, cell-free supernatant (CFS) from E. coll allows for L. plantarum to reduce Fe3' to Fe2+(Panel e), and the co-cultured E. coli and L. plantarum can produce current in bioelectrochemical reactors (Panel I). Monocultures or the mutated E. coli Amend) and L. plantarum Endh2 diminish iron reduction and current generation. For Panels c and f, cocultures were injected at 0 h, and differential (A) current density before and after cell injection was reported. Data represent mean ± s.d. of three (Panel b,e) or two (Panel c,f) biological replicates. P-values were determined by one-way ANOVA with Turkey’s post hoc test. **, P < 0.01; ****3p < 0.0001.

[0040] FIG. 24 shows non-limiting, and exemplary graphs and diagrams indicating that e" COSENS allow s modular, sensitive, and dose-dependent bioelectronic sensing. As shown in Panel a, L. lactis can be engineered to conditionally express rnenD and synthesize DHNA in response to nisin by activating the NisRK two-component system. DHNA subsequently activates EET in L. plantarum to produce electrical signals. As shown in Panel b, L. lactis-L. plantarum co-culture produces current upon nisin induction at 0 h (indicated by the arrow). The differential (A) current densities before and after nisin induction are proportional to the nisin concentrations (0.004-2.5 ng / mL). Panel c provides quantitative analysis of Acurrent density and the 95% confidence time of detection as a function of nisin concentration. As shown in Panel d, E. coli can be engineered to conditionally express menD and synthesize DHNA in response to cumate by inhibiting the CymRAMtranscriptional repressor. DHNA subsequently activates EET in L. plantarum to produce electrical signals. Panel e shows that an E. coli-L. plantarum co-culture produces current upon cumate induction at 0 h (indicated bythe arrow). The differential (A) current densities before and after nisin induction are proportional to the cumate concentrations (0.1-10 pM). Panel f shows quantitative analysis of Acurrent density and 95% confidence time of detection as a function of cumate concentration. Panels g and h show cyclic voltammetry analysis at 18 h (peak current) post nisin or cumate induction. Panel i shows that e'COSENS can be programmed to sense anhydrotetracycline (aTc), arsenite (NaAsCh), or hydrogen peroxide (H2O2). The results show Acurrent density at 12 h post-induction. Time-dependent data and plasmid designs can be found in FIG 32. Data in Panel g and Panel h are representative results of three biological replicates. Other data in panels represent mean ± s.d. of three biological replicates. P-values in i were determined by two-tailed unpaired t-test. **, P < 0.01.

[0041] FIG. 25 provides non-limiting, and exemplary graphs, data, and photos indicating that e'COSENS allows for sensing in food, environmental, and human-relevant environments. Panel a shows certain non-limiting, exemplary physicochemical properties of milk, bayou water, and artificial saliva samples. TOC, total organic carbon. Transmittance w’as determined at 509 nm, the emission wavelength of green fluorescent protein. Panel b provides a photo of the 5 mL-volume bioelectrochemical reactors. N? gas, nitrogen gas; WE, working electrode; CE, counter electrode; RE, reference electrode; PEM, proton exchange membrane. As shown in panel C, the L. lactis-L. plantarum co-culture detects aTc (5 ng / mL) in milk, and the E. coli- L. plantarum co-culture detects NaAsO2(2.5 pM) and H2O2(25 pM) in bayou water and artificial saliva, respectively. The arrows indicate analyte injection. The differential (A) current before and after analyte addition was reported. The asterisks indicate 95% (*, P < 0.05) or 99% (**, P < 0.01) confidence time of detection. Data represent three biological replicates. Panel d Testing L. lactis-L. plantarum co-culture in the gut microbiota from three human donors. The large intestinal microbiota was cultivated in continuous-flow' minibioreactor arrays’3. Panel e: L. lactis-L. plantarum co-culture can sense nisin (2.5 ng / mL) and produce distinguishable cunent within the three gut microbiota samples. Arrows I, II, and III indicate gut microbiota, co-culture, and nisin / buffer injection. Panel f: Statistic analysis of the current levels in Panel E at 360 min. Data represent mean ± s.e.m. of the six biological replicates obtained from three donors. P-values were determined by two-tailed unpaired t-test (Panel c) or one-way ANOVA with Turkey’s post hoc test (Panel I). *, P < 0.05; **, P < 0.01; ns, not significant.

[0042] FIG. 26 show's non-limiting, and exemplary schematics, graphs, data, and photos indicate miniature bioelectronic device simplifies signal detection. Panel a provides a nonlimiting, exemplary assembly schematic of a clay-based microbial fuel cell and a comparisonof the size between the device and a coin under one embodiment. PEM, proton exchange membrane. Panel b is a schematic showing that electrons generated through the DHNA- mediated EET in e’COSENS can accumulate in carbon felt or move from the carbon anode to the Ag, AgCl cathode via an external circuit, allowing for open circuit voltage (OCV) or short circuit current (SCA) measurements. At the cathode, AgCl can be reduced, and H+movement through the clay PEM maintains charge neutrality. Panel c provides an exemplary workflow of signal detection. In this embodiment, co-culture was inoculated into the sample or media and incubated for 1.5-2 hours, and a 60 pL portion was then injected into the device for OCV or SCA measurements. Panel d: OCV and SCA measurements for aTc detection in milk. Differential (A) OCV was reported by subtracting OCV at t = 0 from subsequent OCV values. SCA was determined by linear sweep voltammetry at voltage = 0 vs. OCV. e, OCV measurement using a digital multimeter. In this embodiment, the positive terminal of the multimeter is connected to the carbon anode of the device, and the negative terminal is connected to the Ag / AgCl cathode. OCV measured by the digital multimeter allows for aTc detection in milk. Data represent five biological replicates across two independent experiments on different days. Error bars in (Panels d, e) represent mean ± s.e.m. P-values were determined by two-tailed unpaired t-test. **, P < 0.01.

[0043] FIG. 27 show's non-limiting, and exemplary growth of bacterial strains in chemically defined media containing 0.5% glucose or 1% mannitol. The optical density at 600 nm (ODaoo) of the overnight-grown bacterial cultures was determined by a plate reader. Data represent mean ± s.d. of three biological replicates.

[0044] FIG. 28 shows non-limiting, and exemplary graphs, data, and diagrams indicating that optimization of co-culture inoculation ratio, and co-cultures produces electrical current under microaerobic conditions. Panel a: Iron reduction signal fold change at various inoculation ratios of sender and receiver. The signal fold change represents the Fe2+concentration reduced by the co-culture divided by the sum of background Fe2concentration reduced by the sender or receiver monocultures. An optimal inoculation ratio is ODeoo = 0.05- 0.1 for the sender (L. lactis) and ODeoo = 0.1-0.5 for the receiver {L. plantarum) . Data represent the average of three biological replicates. As shown in panels b and c, both L. lactis - L. plantarum (Panel b) and E. coli - L. plantarum (Panel c) co-cultures produce current under microaerobic conditions (w'ithout N2 gas sparging). The current production is diminished in monocultures or when menD is knocked out in the sender and ndh2 is knocked out in the receiver. Data represent mean ± s.d. of two independent biological replicates.

[0045] FIG. 29 provides non-limiting, and exemplary data indicating growth curves of sender and receiver in co-cultures under anaerobic or microaerobic conditions. Growth and survival of the sender and receiver were determined by CFU counting at the indicated hours under anaerobic (Panel a, b) or microaerobic (Panel c, d) conditions in bioelectrochemical reactors. L. lactis and E. coli were labeled with mCherry, and L. plantarum was labeled with super folded GFP (sfGFP). Images of CFU samples are representative of three biological replicates. The quantitative results represent mean ± s.d. of three biological replicates.

[0046] FIG. 30 shows non-limiting and exemplary graphs and data indicating that e" COSENS’s sensitivity' can be optimized by tuning the ribosome binding site (RBS), and the analyte-induced current production depends on MenD expression and Ndh2-dependent EET. Panel a, d: The translation initiation rate (TIR) of degenerate RBS (NNGGNGN) is predicted using the RBS library calculator70’71. Four RBSs with varied TIRs were tested to tune nisin- inducible MenD expression in L. lactis (Panel a) or cumate-inducible MenD expression in E. coli (Panel d). Panel b, e: RBSs with varied TIRs result in different iron reduction fold changes upon nisin (Panel b) or cumate (Panel e) induction. The menD wild-type (genomic expression of menD) serves as the positive control, and the menD knockout serves as the negative control. Iron reduction levels were normalized to the ODeoo of the sender strains. Panel c, f: Quantitative analysis of iron reduction levels with or without 2.5 ng / mL nisin (Panel c) or 50 uM cumate (Panel f) induction. RBSs with the lowest TIR ( RBS / .M and RBS / w) resulted in the highest fold change and were selected for subsequent tests. Panel g. h : Current production upon nisin (2.5 ng / mL) or cumate ( 10 pM) induction necessitates functional co-cultures. L. lactis or E. coli carrying an empty vector or L. plantarum Endh2 mutant diminished current production. All data represent mean ± s.d. of three biological replicates. P-values in Panel c and Panel f were determined by one-way ANOVA with Turkey’s post hoc test. *. P < 0.05; **, P < 0.01 ; ***, P < 0.001; ns, not significant.

[0047] FIG. 31 provides non-limiting, and exemplary graphs indicating the impacts of exogenous redox-active molecules on sensing. Panels a,b: Various redox-active molecules at 250 nM were sequentially added to the L. lactis-L. plantarum (Panel a) or E. coli-L. plantarum (Panel b) co-cultures at an interval of 15 min before nisin or cumate induction. Co-cultures not exposed to redox-active molecules served as the controls. PYO, pyocyanin; PCA, phenazine- I -carboxylic acid; FAD, flavin adenine dinucleotide; FMN, flavin mononucleotide. (Panel c, d) Influences of exogenous DHNA at 50 or 1 nM on current production of L. lactis-L. plantarum (Panel c) or A. coli-L. plantarum (Panel d) upon induction. DHNA of 50 nM inhibits the nisin-induced current production in L. lactis-L. plantarum co-culture. Panel e: Schematicshows that DHNA can allosterically inhibit MenD. All data represent mean ± s.d. of three biological replicates.

[0048] FIG. 32 shows non-limiting, and exemplary data indicating construction, optimization, and characterization of the aTc, arsenite, and H2O2sensors. Panel a: The anhydrotetracycline (aTc) sensing genetic circuit in L. lactis and optimization of menD expression by tuning RBS strength. In this embodiment, the medium-strength RBSrn resulted in the highest iron reduction fold change in response to varying concentrations of aTc and was selected for subsequent tests. Panel b: Current production of the L. lactis-L. plantarum coculture in response to 5 ng / mL aTc. Panel c: The arsenite sensing genetic circuit in A’, coli and optimization of arsR expression by tuning RBS strength. Synthetic RBS (sRBS) was designed using the RBS calculator70,71. The medium-strength sRBSarsR2resulted in the highest iron reduction fold change in response to varying concentrations of arsenite (in the form of NaAsCh) and was selected for subsequent tests. Panel d: Current production of the E. coli-L. plantarum co-culture in response to 2.5 pM arsenite. Panel e: The hydrogen peroxide (H2O2) sensing genetic circuit in E. coli and iron reduction levels in response to varying concentrations of H2O2. Panel f: Current production of the E. coli-L. plantarum co-culture in response to 25 pM H2O2. Arrows in Panels b-f indicate inducer or buffer injection. All data represent mean ± s.d. of three biological replicates.

[0049] FIG. 33 provides non-limiting, and exemplary co-culture growth analysis and current production of e’COSENS in milk, bayou water, and artificial saliva. Panel a: CFU analysis of co-cultures at 0, 4-, 8-, I2-, and 24-hours post-inoculation into the milk, bayou water, or artificial saliva samples. Panels b, d, f: No difference in current production wras observed between the induced and buffer control groups if / .. lactis or E. coli carried an empty vector or ndh2 was knocked out in L. plantarum. Arrows indicate inducer or buffer injection. Panels c, e, g: Comparison of current production between deficient and functional co-cultures at 120 min (milk) or 360 min (bayou water and artificial saliva). All data represent three biological replicates or their mean ± s.d. P-values w ere determined by one-w ay ANOVA with Turkey’s post hoc test. **, P < 0.01 ; ***, P < 0.001; ns, not significant.

[0050] FIG. 34 shows non-limiting, and exemplary graphs and data indicating growth analysis of L. lactis and L. plantarum co-culture in the gut microbiota and composition of gut microbiota. Panel a, b, c: CFU analysis of the co-culture at 0, 4-, 8-, 12-, and 24-hours postinoculation into the gut microbiola from the three donors. Glucose Ml 7 plates containing 10 pg / mL erythromycin w'ere used for dilution plating to isolate L. lactis and L. plantarum from the gut microbiota. L. plantarum was labeled with sfGFP, and L. lactis was labeled wnthmCherry. The non-fluorescent colonies are eryihromyciii-resislani gut microbes. Images of CFU samples are representative of three biological replicates. The quantitative results represent mean ± s.d. of three biological replicates. Panel d: Metagenomic analysis of the gut microbiota composition from the three donors.

[0051] FIG. 35 shows non-limiting, and exemplary graphs and photos indicating how the clay-based MFC device allows for signal detection from both L. lactis-L. plantarum and E. coli-L. plantarum co-cultures. Panel a shows photos of the device under one exemplary embodiment from the front, back, and sides. Panel b provides a photo showing high-throughput manufacture of the clay-based MFC device. Panels c & d: Measurements of the open circuit voltage (OCV) and short circuit current (SCA) for EET-capable and incapable L. lactis-L. plantarum (Panel c) or E. coli-L. plantanm (Panel d) co-cultures. Panel e: The OCV measured by the digital multimeter differentiates the EET-capable and incapable L. lactis-L. plantarum or E. coli-L. plantarum co-cultures. Data represent three biological replicates or their mean ± s.e.m. P-values were determined by two-tailed unpaired t-test. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

[0052] FIG. 36 shows non-limiting, and exemplary graphs showing optimization of coculture media. The medium was optimized to 1) support the growth of the sender strains (L. lactis or E. coli) for efficient transcriptional regulation in sensing; 2) allow the receiver strain (Z. plantarum)' to maintain metabolic activity without requiring growth; 3) keep the medium composition as simple as possible to minimize the electrochemical background. Panels a, b: Specific growth rate (p) of L. lactis or L. plantanm in blended mCDM and mM 17 media under anaerobic (Panel a) or aerobic (Panel b) conditions. A medium composed of mCDM and mM17 in a ratio of 20:1 was selected for co-culture, as it provides the simplest composition to support / .. lactis growth while maintaining the activity of L. plantarum. Panels c, d: Specific growth rate (p) of E. coli or L. plantanm in M9-derived media under anaerobic (Panel c) or aerobic (Panel d) conditions. mM9-CDM was made by substituting the MOPS buffer and salts in mCDM (Table 10) with M9 salts. Minimal mM9 (with vitamins) was made by supplementing IX Wolfe’s vitamin solution to the minimal mM9. mM9 was made by adding 1% mannitol to the M9 salt solution. The minimal mM9 (without vitamins) was selected for co-culture, as it offers the simplest composition to support E. coli growth while maintaining the activity of L. plantarum. All data represent mean ± s.d. of three biological replicates.

[0053] FIG. 37 shows a schematic representation of a bacterium transferring electrons on metals (Panel A) or electrodes (Panel B) via extracellular electron transfer under one nonlimiting, exemplary embodiment.

[0054] FIG. 38 provides a schematic showing that EET is capable of producing electrical power.

[0055] FIG. 39 shows a schematic representation of current output in the absence (Panel A) and presence (Panel B) of an analyte.

[0056] FIG. 40 is a schematic showing that quinone-dependent EET in Lactiplantibacillus plantarum in one embodiment.

[0057] FIG. 41 provides an exemplary, non-limiting embodiment showing that / .. plantation depends on exogenous DHNA for EET.

[0058] FIG. 42 shows a schematic of an exemplary Bioelectronic Co-culture Sensing System (e-COSENS) under one embodiment of the present disclosure.

[0059] FIG. 43 is an exemplary schematic showing a non-liming means for synthetical control of DHNA production in a DHNA producer under one embodiment.

[0060] FIG. 44 provides an exemplary DHNA synthesis pathway and certain genes involved in the synthesis along with data show ing that DHNA synthesis is regulated by reverse flux and negative allosteric regulation.

[0061] FIG. 45 provides non-limiting, exemplary schematics and data showing that one embodiment of the presently disclosed system is modular and compatible to existing genetic circuits. Briefly, these schematics and data show that, in the present system, transcription factors can be manipulated to allow for the detection of carbon source (arabinose) (Panel A) and redox oxygen species (H2O2) (Panel B).

[0062] FIG 46 is a non-limiting, exemplary diagram showing that living cells can be integrated with electronics to allow7for microbial bioelectronic sensors under various embodiments.

[0063] FIG. 47 provides an exemplary flow7diagram illustrating that, in various embodiments, bioelectric sensors can be developed by using analyte-responsive genetic elements to regulate the expression of MenD in a feeder strain.

[0064] FIG. 48 provides a schematic showing a top perspective (Panel A) and exploded view7(Panel B) of a miniature bioelectronic device under one embodiment of the present disclosure.DETAILED DESCRIPTION OF THE INVENTION

[0065] Abbreviations and Definitions

[0066] Detailed descriptions of one or more preferred embodiments are provided herein. It is to be understood, however, that the present invention can be embodied in various forms. Therefore, specific details disclosed herein are not to be interpreted as limiting, but rather as a basis for the claims and as a representative basis for teaching one skilled in the art to employ the present invention in any appropriate manner.

[0067] The singular forms “a”, “an” and “the” include plural reference unless the context clearly dictates otherwise. The use of the word “a” or “an” when used in conjunction with the term “comprising” in the claims and / or the specification can mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.”

[0068] Wherever any of the phrases “for example,” “such as,” “including” and the like are used herein, the phrase “and without limitation” is understood to follow unless explicitly stated otherwise. Similarly, “an example,” “exemplary” and the like are understood to be nonlimiting.

[0069] The term “substantially” allows for deviations from the descriptor that do not negatively impact the intended purpose. Descriptive terms are understood to be modified by the term “substantially” even if the word “substantially” is not explicitly recited.

[0070] The terms “comprising” and “including” and “having” and “involving” (and similarly “comprises”, “includes,” “has,” and “involves”) and the like are used interchangeably and have the same meaning. Specifically, each of the terms is defined consistent with the common United States patent law definition of “comprising” and is therefore interpreted to be an open term meaning “at least the following,” and is also interpreted not to exclude additional features, limitations, aspects, etc. Thus, for example, “a process involving steps a, b, and c” means that the process includes at least steps a, b and c. Wherever the terms “a” or “an” are used, “one or more” is understood, unless such interpretation is nonsensical in context.

[0071] The term “about” is used herein to mean approximately, roughly, around, or in the region of. When the term “about” is used in conjunction with a numerical range, it modifies that range by extending the boundaries above and below the numerical values set forth. In general, the term “about” is used herein to modify a numerical value above and below the stated value by a variance of 20 percent up or down (higher or lower).

[0072] Described herein is an electroactive co-culture sensing system (e’COSENS) to improve upon existing biosensing technologies by generating electrical signals, which can be delected by simple electronic components, offering a cost-effective solution for biosensing. Additionally, e’COSENS offers modularity and customizability, allowing for easy replacement of the sender and receiver strains with different bacterial species tailored to specificapplications. Moreover, e'COSENS standardizes the conversion of chemical signals into electrical signals, bypassing extensive engineering, and allowing for broader application beyond native electroactive bacteria in diverse environmental settings.

[0073] Aspects of the invention are drawn towards an electroactive co-culture sensing system (e'COSENS) comprising a sender strain, a receiver strain, and a bioelectrochemical measurement apparatus. As used herein, the term electroactive co-culture sensing system can refer to a co-culture that can sense an environmental stimulus or analyte and generate electrical signals. For example, the term electroactive co-culture sensing system can refer to a sender strain and a receiver strain co-culture that can sense an environmental stimulus or analyte and generate electrical signals detectable by a bioelectrochemical measurement apparatus.

[0074] In embodiments, the bioelectrochemical measurement apparatus can comprise a bioelectrochemical cell and an electronic device (e.g., a potentiostat or a current / voltage meter) to monitor the electrical signals, including voltage and current. For example, the bioelectrochemical cell can comprise a working chamber, a counter chamber, a working electrode, a counter electrode, a reference electrode, and a cation exchange membrane. However, the reference electrode and counter chamber are not always required.

[0075] As used herein, the term, “sender strain” can refer to a bacterial strain that can produce a quinone in response to an environmental stimulus or analyte. In embodiments, the sender strain is a quinone-producing bacterium. In embodiments, the sender strain can be engineered to delete menD and can be further transformed with the menD gene on a plasmid w ith a stimulus-responsive genetic element.

[0076] As used herein, the term “stimulus-responsive genetic element” can refer to a genetic element that is activated by an environmental stimulus. For example, upon activation of the genetic element, the response can be the production of a quinone. For example, stimulus- responsive genetic element can comprise an inducible transcription factor and an inducible promoter. In embodiments, when combined with an insulator, a ribosome binding site (RBS) and a terminator, the complete genetic circuit can regulate menD expression on a plasmid. Without washing to be bound by theory, any genetic elements that can regulate protein expression levels or modulate protein activity can be used to control menD. Genetic elements can comprise riboswitches, aptamers, CRISPR-based tools, RNA interference (RNAi), split protein switches, protein degradation tags, or other genetic elements known in the art. For example, the genetic element can comprise telR, arsR. oxyR, nisR and nlsK. araCAMand araE, cymR4M, Pxyi / 2xreio,PcymRc, or a combination thereof. For example, the genetic element can comprise a transcription factor selected from the group consisting of leiR.arsR, oxyR, nisR and nisK, araCAMand araE, cymRiM, or a combination thereof. For example, the genetic element can comprise a promoter selected from the group consisting of Pxyi / 2xteto, Panoc2, Poxtf, Pnfc.4, PBAD, PcymRc, or a combination thereof.

[0077] In embodiments, the sender strain can be engineered to delete menA, ndh, or a combination thereof. Deletion of menA can be used to prevent or reduce derivatization of DHNA in the sender strain in the absence of the activation of the genetic element. In embodiments, deletion of ndh prevents the sender strain from performing EET in the absence of activation of the genetic element. Deletion of menA, ndh, or both can be used to enhance DHNA-derived electrical signals (menA deletion), to reduce the background signals (ndh deletion).

[0078] In embodiments, the sender strain is a bacterium with a DHNA synthesis pathway. For example, the bacterium can produce 1 ,4-dihyroxy-2-naphthoic acid (DHNA), dimethylmenaquinone (DMK), menaquinone, or a combination thereof. In embodiments, the sender strain can be selected from the group consisting of Escherichia coli, Bacillus subtilis, Vibrio natriegens, Shewanella oneidensis, Leuconostoc citreum, Staphylococcus aureus, Enterococcus faecalis, a strain from Example 6, or a combination thereof.

[0079] In embodiments, the term “receiver strain” can refer to a quinone-dependent extracellular electron transfer (EET)-capable bacterium. For example, the receiver strain can produce an electrical output. For example, the receiver strain can produce an electrical output using an environmental quinone as an electron shuttle. Quinone, as used herein can refer to any quinone derivative, quinone precursor, a quinone, or a combination thereof. In certain embodiments, quinone refers to any reduced or modified derivative within the quinoid family. Quinones, as used herein, can refer to any compound that comprises a conjugated carbonyl / aromatic system from the quinone backbone. In certain non-limiting, exemplary embodiments, the quinone can be selected from the group consisting of DHNA, DMK, menaquinone, hydroquinone 1,4-NQ, menadione, juglone, DCDMQ, or a combination thereof. For example, the receiver strain can use the quinone as an electron shuttle through the Ndh2- dependent EET pathway. In embodiments, the receiver strain can be a wild-type quinonedependent EET-capable strain. In embodiments, the receiver strain can be genetically engineered to delete menA, ndhl, or a combination thereof. In embodiments, the receiver strain is Lactiplantibacillus plantarum, a strain selected from Example 7, or a combination thereof.

[0080] In certain non-limiting, exemplary embodiments, the presently disclosed sensors, systems, and methods can permit the detection of quinones at a wide concentration range. In embodiments, the presently disclosed sensors, systems, and methods the DHNA at aconcentration of at least 5 nM. In embodiments, the presently disclosed sensors, systems, and methods can detect DHNA at a concentration range of at least 500 pM. The electrical output signals can be proportional to the concentration of quinone being detected.

[0081] Various embodiments of the present disclosure improve analyte selectivity in cells that can be adapted to sense and distinguish between different redox-active molecules. In certain embodiments, the presently disclosed sensors, systems, and methods can selectively distinguish between different quinone structural analogues. For example, the presently disclosed sensors, systems, and methods can sense and distinguish menadione, DHNA, lawsone, or a combination thereof amongst other quinone structural analogs. Embodiments include a whole-cell biosensor and methods of use thereof configured for selectivity and specificity (Fig. 18). By way of non-limiting example, certain embodiments can sense a subset of quinones - 1 ,4-naphthoquinones, and selectively discriminate analogs, such as DHNA, menadione, and lawsone, based on the differential increments of the current density over time.

[0082] Aspects of the disclosure are drawn towards producing a co-cultuie sensing system (e’COSENS). the method comprising: obtaining a quinone-producing bacterium, and a quinone-dependent extracellular electron transfer (EET)-capable bacterium; genetically modifying quinone-producing bacterium, thereby producing a sender strain; genetically modifying a quinone-dependent extracellular electron transfer (EET)-capable bacterium, thereby producing a receiver strain; co-culturing the sender strain and receiver strain, thereby producing a co-culture; and introducing the co-culture into a bioelectrochemical measurement apparatus, thereby producing a co-culture sensing system (e'COSENS).

[0083] In various embodiments, the co-culture conditions sustain the viability of both sender and receiver strains, support optimal EET activity, and minimize the electrochemical background. In one exemplary embodiment, the media used in the co-culture conditions supports growth of E. coli-L. plantarum and L. lactis-L. plantarum co-cultures. In embodiments, the growth media comprises mannitol. In such embodiments, mannitol can serve as a carbon source to entrance quinone synthesis, EET, or a combination thereof.

[0084] In embodiments, the ratio of sender-to-receiver strains is up to about 1 : 50. The ratio of sender to-receiver strains can be up to about 1 :20. In certain embodiments, , the ratio of sender-to-receiver strains is about 1 : 1. The ratio of sender-to-receiver strains can be between about 1 : 1 and 1 : 15. In embodiments, the sender-to-receiver ratio is between about 1 :2 to about 1 : 10. , the ratio sender-to-receiver strains can be about 1 :2, about 1:3, about 1 :4, about 1 :5, about 1 :6, about 1 :7, about 1 :8, about 1 :9, or about 1 : 10.

[0085] In embodiments, genetically modifying a quinone-producing bacterium comprises deleting mend from the quinone-producing bacterium and transforming quinone-producing bacterium with a plasmid containing the menD gene with a stimulus-responsive genetic element. In embodiments, genetically modifying a quinone-producing bacterium further comprises deleting menA, ndh, or a combination thereof.

[0086] In embodiments, the stimulus-responsive genetic element comprises an inducible transcription factor and an inducible promoter. For example, the inducible transcription factor comprises tetR, arsR, oxyR, nisR and rrisK, araCAMand araE, cymRAil, or a combination thereof. For example, the inducible promoter comprises PXyi / 2xfeto, P ’arsoc2, Poxys, PniSa, PBAD, PcymRc, or a combination thereof.

[0087] In embodiments, the method further comprises genetically modifying the quinonedependent extracellular electron transfer (EET)-capable bacterium. In embodiments, genetically modifying the quinone-dependent EET-capable bacterium comprises deleting menA, ndhl, or a combination thereof from the quinone-dependent EET-capable bacterium.

[0088] In embodiments, the quinone-producing bacterium produces 1 ,4-dihyroxy-2- naphthoic acid (DHNA).

[0089] In embodiments, the bioelectrochemical measurement apparatus comprises a bioelectrochemical cell and an electronic device for voltage and current monitoring.

[0090] In embodiments, the sender strain is Escherichia coli, Bacillus subtilis, Vibrio natriegens, Shewanella oneidensis, Leuconostoc citreum, Lactococcus laclis, Enterococcus faecalis, a strain selected from Example 6, or a combination thereof.

[0091] In embodiments, tire receiver strain is Lactiplantibacillus plantarum, a strain selected from Example 7, or a combination thereof.

[0092] A bioelectronic sensing system capable of translating environmental stimuli into electrical signals produced by the methods described herein.

[0093] A bioelectronic sensor or bioelectronic measurement apparatus for detecting pharmacologically relevant quinones and associated methods are also disclosed herein. Certain embodiments relate to a bio-electronic hybrid device for portable and low-cost bioelectronic sensing.

[0094] FIG. 48A provides a schematic showing a top perspective view of a miniature bioelectronic measurement apparatus 100 under one embodiment of the present disclosure. FIG. 48b shows an exploded view of the miniature bioelectronic measurement apparatus 100 of the FIG. 48A embodiment. In various embodiments, the bioelectrochemical measurement apparatus used in the present sensors, systems, or methods comprises a miniature microbialfuel cell (MFC) 100. In one exemplary' embodiment, the presently disclosed biosensor is a miniature clay-based microbial fuel MFC 100. In one non-limiting, exemplary embodiment, MFC comprises a single-chamber MFC 100.

[0095] In various embodiments, the single-chamber MFC 100 comprises an anode 110, a cathode 150, and a proton exchange membrane (PEM) 140. In certain embodiments, the cathode 150 comprises an Ag / AgCl-coated polyester sheet 152. In embodiments, the anode 1 10 comprises a carbon felt 120. The carbon felt 120 can be hydrophilic. The anode can comprise a porous carbon felt 120. In embodiments, the anode comprises a hole, gap, or passthrough 114. The hole, gap, or passthrough 114 can be configured to receive bacteria or other inputs into the miniature MFC 100.

[0096] The MFC 100 can comprise a well 130 with a hole, gap, passthrough, or notch 134 disposed therein. In embodiments, the hole, gap, passthrough , or notch 134 is configured to receive and hold the carbon felt 120 therein.

[0097] In one embodiment, the well 130 comprises acrylic. In embodiments the well 130 is about 19 mm wide, about 19 mm long, about 3 mm thick, or a combination thereof. In some embodiments, the width of the well 130 ranges from about 14.0 mm to about 24.0 mm. The well 130 can comprise a width of about 14.0 mm, about 15.0 mm, about 16.0 mm, about 17.0 mm, about 18.0 mm, about 19.0 mm, about 20.0 mm, about 21.0 mm, about 22.0 mm, about 23.0 mm, or about 24.0 mm. In various embodiments, the well 130 comprises a length of between about 14.0 mm to about 24.0 mm, inclusive. The well 130 can comprise of length of about 14.0 mm, about 15.0 mm, about 16.0 mm, about 17.0 mm, about 18.0 mm, about 19.0 mm, about 20.0 mm, about 21.0 mm, about 22.0 mm, about 23.0 mm, or about 24.0 mm. In various embodiments, the thickness of the well 130 is between about 1.0 mm to about 6.0 mm, inclusive. The thickness of the well 130 can be about 1.0 mm, about 2.0 mm, about 3.0 mm, about 4.0 mm, about 5.0 mm, or about 6.0 mm. In embodiments the thickness of the well 130 is substantially the same or similar to the thickness of the carbon felt 120.

[0098] In one embodiment, the cathode 150 is created by screen-printing an Ag / AgCl ink onto a polyester sheet 152 and allowed to dry.

[0099] The electrodes 110, 150 can be separated by the PEM 140. As can be seen in FIG. 48B, the PEM 140 can be sandwiched between an Ag / AgCl cathode 150 on one side and a carbon anode 110 on the other side.

[0100] In embodiments, tire PEM 140 comprises a clay material. For example, the clay can be a vermiculite clay material. In embodiments, the PEM 140 is atomically thin. In various embodiments, the PEM 140 comprises a thermally expanded clay material. The PEM can be asubstantially 2D clay-based structure. The PEM 140 can be exfoliated from natural vermiculite clay. In certain embodiments, the clay PEM 140 is made by exfoliating a thermally expanded natural vermiculite clay into atomically thin 2D flakes using a 20% aqueous HQ solution. An aqueous dispersion of these clay nanosheets can be vacuum-filtered onto a cellulose nitrate membrane to fabricate a freestanding lamellar membrane. To make the anode 1 10, carbon ink 1 10 can be screen-printed onto a Polyethylene Terephthalate (PET) sheet 112, with a hole, gap, or passthrough 114. In embodiments, the hole, gap, or passthrough is about 1 mm in diameter. A carbon felt 120 material can then be attached to the carbon ink 110, and the ink can be allowed to dry.

[0101] In embodiments, the diameter of the carbon felt is about 6 mm. The carbon felt can have a diameter of up to about 10 mm. hi certain embodiments, the diameter of the carbon felt is less than about 1 mm. The diameter of the carbon felt can be about 1 nun, about 2 mm, about 3 mm, about 4 mm, about 5 mm, about 6 nun, about 7 mm, or about 8 m. In certain embodiments, the thickness of the carbon felt is about 3.18 mm. The carbon felt can have a thickness of up to about 5 mm. In embodiments, the carbon felt has a thickness of less than about 1 mm. The carbon felt can have a thickness of between about 1 mm and about 5 nun, between about 2 mm and about 4 mm, or between about 3.05 and 3.95 mm. In embodiments, the thickness of the carbon felt about 3.0 nun, about 3.1 mm, about 3.2 mm, about 3.3 mm, about 3.4 mm, about 3.5 mm, about 3.6 mm, about 3.7 mm, about 3.8 mm, about 3.9 mm, or about 4.0 mm. in embodiments, the thickness of the carbon felt is about 3.10 mm, about 3. 1 1 mm, about 3.12 mm, about 3.13 mm, about 3.14 mm, about 3.15 mm, about 3.16 mm, about 3.17 mm, about 3.18 mm, about 3.19 mm, or about 3.20 mm.

[0102] In various embodiments, a digital multimeter can be used to assess electrical output. A potentiostat can be used for obtaining electrochemical measurements. For electrochemical measurement using the potentiostat under one non-limiting, exemplary embodiment, the working electrode can be connected to the carbon anode 1 10 of the device 100, and the reference electrode can be joined with the counter electrode and connected to the Ag / AgCl cathode 150 of the device 100.

[0103] In embodiments, the miniature MFC 100 disclosed herein can be formulated by placing a clay membrane 140 on top of the cathode 150. The well 130 can be affixed on the top of the clay membrane 140. In embodiments, the well 130 is affixed using glue. The carbon felt 120 anode can be fitted into the hole, gap, passthrough, or notch 134 of tire well 130.

[0104] Under certain embodiments, the screen-printed MFC device 100 can be treated with UV, -'ozone to make carbon felt 120 hydrophilic. The carbon felt 120 can be treated with UV / ozone for up to at least about 30 minutes.

[0105] The bio-electronic sensor or bioelectronic measurement apparatus 100 can be compact (centimeter-scale) and lightweight. In embodiments, the miniature MFC or bioelectronic measurement apparatus 100 comprises a width that is less than about 3 cm. The bio-electronic sensor can comprise a width that is less than about 2.5 cm. In embodiments, the width of the bio-electronic sensor is about 2 cm. The width of the bio-electronic sensor can be up to about 2.5 cm, up to about 2.25 cm, up to about 2.00 cm, up to about 1.75 cm, up to about 1 .5 cm, up to about 1.25 cm, up to about 1.00 cm, or up to about 0.75 cm. In embodiments, the height of the bioelectronic sensor is less than about 1 cm. The height of the sensor can be about 0.5 cm. In embodiments, the height of the bio-electronic sensor is up to about 0.75 cm, about 0.70 cm, about 0.65 cm, about 0.60 cm, about 0.55 cm, about 0.50 cm, about 0.45 cm, about 0.40 cm, about 0.35 cm, about 0.30 cm, about 0.25 cm, about 0.20 cm, about 0.15 cm, about 0.10 cm, or about 0.05 cm.

[0106] Non-limiting, exemplary aspects of the present disclosure include commercial testing kits for rapid field detection of various environmental factors, such as pollutants in environmental samples, contaminations in food, and biomarkers in biological fluids. Additional non-limiting, exemplary aspects of the present disclosure relate to wearable, implantable, or digestible bioelectronic devices for point-of-care and continuous health monitoring.

[0107] Certain embodiments permit sensing of quinones in a living host. Embodiments permit sensing of quinones in a mammalian host. Under one embodiment, the presently disclosed sensors, systems, and methods permit detection of quinones in intestinal fluid.

[0108] In various embodiments, the presently disclosed sensors, systems, and methods can utilize the EET pathway to produce colorimetric (reducing iron) or amperometric signals (reducing electrode).

[0109] In various embodiments of the presently disclosed sensors, systems, and methods, perform their respective sensing functions without changing the environment within which the sensors or systems are deployed or within which the methods are performed. By way of nonlimiting example, after sensing the quinone via the presently disclosed sensors, systems, or methods, the quinone is re-oxidized by the electrode such that there is no consumption of donors / acceptors or oxyanions during the sensing process.

[0110] The present disclosure further provides for systems and method of improving analyte selectivity in whole cells that can be adapted to sense redox-active molecules.

[0111] Other Embodiments

[0112] While the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended c laims. Other aspects, advantages, and modifications are within the scope of the following claims.

[0113] The invention will be further described in the following examples, which do not limit the scope of the invention described in the claims.EXAMPLES

[0114] Examples are provided below to facilitate a more complete understanding of the invention. The following examples illustrate the exemplary modes of making and practicing the invention. However, the scope of the invention is not limited to specific embodiments disclosed in these Examples, which are for purposes of illustration only, since alternative methods can be utilized to obtain similar results.EXAMPLE 1

[0115] Synthetic microbial co-cultures for modular bioelectronic sensing

[0116] We have developed e’COSENS (electroactive co-culture sensing system), a modular whole-cell bioelectronic sensing platform that can translate environmental stimuli into electrical signals. This system harnesses cooperation between two engineered bacterial strains: a sender strain that detects analytes through genetic circuits and consequently produces quinone mediators, and a receiver strain that uses quinone as an electron mediator to generate electrical current through extracellular electron transfer (EET). The inherent modularity allows easy replacement of the sender bacterial strain, the receiver bacterial strain, and the genetic circuits tailored for specific applications. We have demonstrated the use of e’COSENS to sense antibiotics, heavy metals, redox oxygen species, antimicrobial peptides, and common inducer molecules. This versatile bioelectronic sensing platform shows promise for applications in food, environment, and health monitoring.

[0117] This invention can be classified into either a new process, composition of matter, or a device.

[0118] New Process: genetic engineering processes to transform wild-type bacterial candidates into the sender or receiver strains suitable for bioelectronic sensing.

[0119] Composition of matter: combine a quinone-producing bacterium harboring stimulus-responsive genetic circuits with an electricity-generating bacterium to create synthetic co-cultures for bioelectronic sensing.

[0120] Device: a living device based on synthetic microbial consortium that can convert input of interests into electrical outputs.

[0121] This invention presents e’COSENS (electroactive co-culture sensing system), a modular bioelectronic sensing system capable of translating environmental stimuli into electrical signals. We introduce a simple yet efficient approach to engineering bioelectronic sensors by leveraging electronic communication between a quinone-producing bacterium anda quinone-dependent EET-capable bacterium. Our approach has implications for versatile, low- cost bioelectronic sensing in food, environment, and health industries.

[0122] First, we select a DHNA ( 1 ,4-dihydroxy-2-naphthoic acid, a menaquinone precursor)-producing bacterium and engineer it into a sender strain that selectively produces DHNA in response to an environmental stimulus. Our survey identifies 2238 bacteria across diverse phyla harboring the intact DHNA synthesis pathway; these bacteria are all qualified as exemplary sender strains. We have demonstrated our approach with two sender strains: Escherichia coli and Lactococcus lactis. Subsequent genetic engineering is performed on the sender strain to delete men A (encodes for 1 ,4-dihydroxy-2 -naphthoate octaprenyl transferase), menD (encodes for 2-succinyl-5-enolpyruvyl-6-hydroxy-3-cyclohexene- 1 -carboxylate synthase), and ndh (encodes for Type II NADH dehydrogenases). By deleting menA from the genome, DHNA is not further derivatized. Deletion of ndh prevents the sender strain from performing EET. Deletion of menD disrupts the natural regulated synthesis of DHNA, so that this reaction can be conditionally complemented. The sender strain is then transformed into a sensor by complementing the menD gene back on a plasmid under tire control of a stimulus- responsive genetic element, such as an inducible transcription factor. This genetic circuit allows the sender strain to only produce DHNA in response to a specific environmental stimulus.

[0123] Second, we select a receiver strain capable of performing extracellular electron transfer (EET) and generating electrical current in the presence of exogenous DHNA. Our survey identifies 324 bacteria within the Firmicutes phylum which can be capable of EET when exogenous DHNA is present, with a notable example being the probiotic gut bacterium Lactipkmtibacillus plantarum. Genetic modifications are conducted to delete menA (encodes for l,4-dihydroxy-2-naphthoate octaprenyltransferase) and ndhl (encodes for type-II NADH dehydrogenase) from the genome to eliminate DHNA-consuming pathways and improve signal orthogonality. These modifications ensure signal detection through the DHNA-mediated Ndh2 -dependent EET pathway in the receiver strain.

[0124] Finally, the sender and receiver strains are combined to create an electroactive co-culture for bioelectronic sensing. When an environmental stimulus is present, it triggers the menD expression and DHNA synthesis in tire sender strain, w'hich subsequently activates EET in the receiver strain, resulting in the generation of electrical current. The electrical current is monitored by a potentiostat connected to three-electrode bioelectrochemical systems (BES) with a carbon felt working electrode, a titanium wire counter electrode, and an Ag / AgCl reference electrode. The system is inherently modular and customizable; numerous suchelectroactive co-cultures can be developed by pairing one of the 2238 DHNA-producing bacteria with one of the 324 DHNA-dependent EET-capable bacteria. We have shown E. coll - L. plantarum co-culture and L. lactis - L. plantarum co-culture and changed transcription factors to allow the detection of heavy metal (arsenite), antibiotics (anhydrotetracycline), redox oxygen species (H2O2), antimicrobial peptide (nisin), carbon source (arabinose), and common inducer molecule (cumate). We have also shown that these co-cultures can sense and respond in food, environmental, and simulated human samples, including milk, bayou water, artificial saliva, and in vitro cultivated human gut microbiota.

[0125] This invention can be deployed for scenarios that necessitate a fast, low-cost, and user-friendly sensor, such as pollution monitoring in wastewater, food safety and quality control, or health monitoring and diagnosis. For example, and without wishing to be bound bytheory, one non-limiting usage can be to load samples into a microbial-fuel-cell-based miniature bioelectronic device that contains the e'COSENS and monitor electrical signals triggered by a target analyte via an external voltage / current meter. In addition, and without being bound by theory, e’COSENS as disclosed herein can be integrated into a compact bioelectronic pill to sense-and-respond in situ and wirelessly transduce the signals to an external device, such as a smart phone. Furthermore, in certain embodiments, e'COSENS can be used to fundamentally investigate interspecies electron transfer and its physiological implications.

[0126] Non-limiting surprising aspects described herein comprise:

[0127] Generation of electrical signals: Unlike existing biosensors that primarily use optical signals such as fluorescence, e'COSENS generates electrical signals, which allows detection using miniaturized and low-cost electronic components.

[0128] Modularity: e'COSENS employs cooperation between tw'O bacterial strains. This allows easy replacement of the sender and the receiver bacterial strain wnth different bacterial species suitable for specific applications.

[0129] Standardized signal conversion: The employment of a receiver strain simplifies and standardizes the conversion of chemical signals into electrical signals. This is distinct from existing approaches for bioelectronic sensor development which require extensive engineering on electron transfer pathways in native or non-native electroactive bacteria.

[0130] Versatility: The successful development of bioelectronic sensors by pairing phylogenelically distinct bacteria such as E. coli and L. plantarum highlights e'COSENS’s versatility, indicating the application of e'COSENS across diverse environmental settings.

[0131] Without wishing to be bound by theory, encapsulating the bioelectronic sensor on the electrode can isolate it from environmental microbiota and minimize interference. Integrating the bioelectronic sensor into a device casing can further protect it from environmental factors. Employing ratiometric measurement with an EET-incapable negative control can also address interferences and improve the signal-to-noise ratio.

[0132] References

[0133] Tejedor-Sanz, S., Stevens. E.T., Li. S.. Finnegan, P., Nelson, J., Knoesen, A., Light, S.H., Ajo-Franklin, C.M., Marco, M.L. Extracellular electron transfer increases fermentation in lactic acid bacteria via a hybrid metabolism. eLife 11, e70684 (2022)

[0134] Stevens, E.T., Beeck, W.V., Blackburn, B., Tejedor-Sanz, S., Rasmussen, A.R.M., Mevers, E., Ajo-Franklin, C.M., Marco, M.L. Lactiplantibacillus plantarum uses ecologically relevant, exogenous quinones for extracellular electron transfer. mBio, e02234- 23 (2023)

[0135] Tolar, J.G., Li, S., Ajo-Franklin, C.M., 2022. The Differing Roles of Flavins and Quinones in Extracellular Electron Transfer in Lactiplantibacillus plantarum. Applied and Environmental Microbiology 89, e01313-22 (2023)

[0136] Li, S., Tavares, C. D. G., Tolar, J. G. & Ajo-Franklin, C. Selective bioelectronic sensing of pharmacologically relevant quinones using extracellular electron transfer in Lactiplantibacillus plantarum. Biosensors and Bioelectronics 243, 115762 (2024)

[0137] Atkinson, J.T.. Su, L., Zhang, X., Bennett, G.N., Silberg, J.J., Ajo-Franklin, C.M. Real-time bioelectronic sensing of environmental contaminants. Nature 611, 548-553 (2022)

[0138] Mimee, M., Nadeau, P., Hayward, A., Carim, S., Flanagan, S., Jerger, L., Collins, J., McDonnell, S.. Swartwout, R., Citorik, R.J., Bulovic, V., Langer, R., Traverse, G., Chandrakasan, A.P., Lu, T.K. An digestible bacterial-electronic system to monitor gastrointestinal health. Science 360, 915-918 (2018)

[0139] Inda-Webb, M. E. et al. Sub- 1.4 cm3 capsule for detecting labile inflammatory' biomarkers in situ. Nature 620, 386-392 (2023).EXAMPLE 2

[0140] Selective bioelectronic sensing of pharmacologically relevant quinones using extracellular electron transfer in Lactiplantibacillus plantarum

[0141] Abstract

[0142] Redox-active small molecules containing quinone functional groups play important roles as pharmaceuticals, but can be toxic if overdosed. Despite the need for a fast and quantitative method to detect quinone and its derivatives, current sensing strategies are often slow and struggle to differentiate between structural analogs. Leveraging the discovery that microorganisms use certain quinones to perform extracellular electron transfer (EET), we investigated the use of Lactiplanlibacillus plantarum as a whole-cell bioelectronic sensor to selectively sense quinone analogs. By tailoring the native EET pathway in L. plantarum, we allowed for quantitative quinone sensing of 1 ,4-dihydroxy-2-naphthoic acid (DHNA) - a gut bifidogenic growth stimulator. We found that L. plantarum can respond to environmental DHNA within seconds, producing concentration-dependent electrical signals. This sensing capacity was robust in different assay media and allowed for continuous monitoring of DHNA concentrations. In a simulated gut environment containing a mixed pool of quinone derivatives, this tailored EET pathway can selectively sense pharmacologically relevant quinone analogs, such as DHNA and menadione, amongst other structurally similar quinone derivatives. We also developed a multivariate model to describe the mechanism behind this selectivity and found a correlation between quinone physicochemical properties and the corresponding electrical signals. Our work presents a new concept to selectively sense quinone using whole-cell bioelectronic sensors and opens the possibility of using probiotic A, plantarum for bioelectronic applications in human health.

[0143] Introduction

[0144] Redox-active small molecules facilitate interactions between organisms and their environment. For example, redox-active molecules are tools for interspecies cooperation and competition (Fritts et al., 2021), vehicles for organisms to gain nutrients and conserve energy from their environment (Glasser et al., 2017), and messengers for engineering artificial cell-cell communication (VanArsdale et al., 2020). A particularly significant set of redox-active small molecules contain the quinone functional group. These molecules, referred to as quinone, play a crucial role in biological energy transduction and have been widely used as pharmaceuticals to combat pathogens, tumors, and to reduce inflammation (Pereyra et al., 2019; Bolton and Dunlap, 2017). However, adverse effects, including toxicity, appear at high doses or long-term exposure (Bolton et al., 2000; Bolton and Dunlap, 2017). Additionally, toxic quinone derivatives released from industrial products or combustion of fossil fuels have contributed to air. water, and soil pollution, damaging global health (Cao et al., 2022; Kumagai et al., 2012). Therefore, we need a rapid sensing approach to monitor quinones and mitigate their toxic effects.

[0145] Existing strategies for detecting a specific quinone in biological samples typically involve high-performance liquid chromatography (HPLC) or liquid chromatographymass spectrometry (LCMS) (Mevers et al., 2019; Takebayashi et al., 2008; Pei et al., 2018). However, these analytical methods use expensive equipment that requires highly trained operators to perform calibration and sample preparation before measurement. While electrochemical methods can also be used to detect quinones, they need to be coupled with chromatography to differentiate between quinone derivatives (Babula et al., 2009). Other techniques, including fluorescence spectroscopy (Cory and McKnight, 2005; Mazzulli et al.,2016), colorimetric assay (Fukuda et al., 2019; Irreverre and Sullivan, 1941), or enzymatic biosensors (Abdullah et al., 2007; Maleki et al., 2017; Shan et al., 2008) offer a faster detection time at lower cost but are inefficient in distinguishing quinone analogs. Consequently, cunent technologies (summarized in Fig. 18) do not meet the demand for rapid and selective detection of quinones.

[0146] An emerging sensing technology that is fast, uses inexpensive instrumentation, and operates in complex samples is whole-cell bio-electronic sensors. These sensors harness exoelectrogenic microbes to sense chemicals and generate electrical signals through a multi-step process known as extracellular electron transfer (EET) (Bird et al., 2021; TerAvest and Ajo- Franklin, 2016). EET can be rewired to sense target analytes by transcriptionally controlling the expression of electron transfer proteins with an analyte-inducible promoter (Golitsch et al., 2013; Karbelkar et al., 2021 ; Khan et al., 2020; Webster et al., 2014; West et al., 2017; Zhou et al., 2021). While this strategy show's high specificity, it suffers from limited analyte range and slow7response time (hours to days). To sense substrates required for electron transfer, such as electron donors (Zhou et al., 2017), terminal electron acceptors (Atci et al., 2016; Baruch el al., 2021), or oxyanions involved in assimilatory reduction (Atkinson et al., 2022), the substratespecific oxidoreductase can be introduced or deleted to control electron flux through EET. While this strategy can yield a real-time response (minutes to hours), it consumes the analyte, which can lead to unwanted changes in the environment. Another strategy involves engineering a protein in the EET pathway to be allosterically regulated by the analyte of interest. This strategy has recently been used to sense an endocrine disruptor in real-time (minutes) (Atkinson et al., 2019, 2022), but it requires a known ligand-binding domain for the targeted analyte. Lastly, redox-active analytes can be detected by shuttling electrons from cellular metabolism to generate an electrical current. This strategy has been used to sense riboflavin (Yu et al., 2017) and pyocyanin (Yang et al.,2017), but it remains unknown whether their analogs can be selectively distinguished. Thus,these existing limitations in bio-electronic sensing call for new strategies that are fast, sensitive, and suitable for selective detection of molecules that have diverse analogs, such as quinones.

[0147] Recent discoveries of the diverse ways EET is carried out in nature provide inspiration for new sensing strategies. Together with Light and Portnoy (Light et al., 2018) and Marco (Tejedor-Sanz et al., 2022), we recently discovered an EET pathway that is widely found in gram-positive Firmicutes, including a probiotic gut bacterium - Lactiplantibacillus plantarum (Fig. 6 Panel A). This lactic acid bacterium can grow by reducing extracellular- electron acceptors such as insoluble iron oxides and an electrode in the presence of exogenous quinone through membranebound oxidoreductases (Tejedor-Sanz et al., 2022; Tolar et al., 2022). Here, we leverage this new understanding to develop L. plantarum as a bioelectronic sensor for pharmacologically relevant quinones. We first examined its sensitivity, speed, and dose-dependence in response to DHNA (1,4- dihydroxy-2-naphthoic acid), a menaquinone precursor that can reduce gut inflammation and stimulate bifidogenic growth in the gut (Cheng et al., 2017; Fukiunoto et al., 2014; ISAWA et al., 2002). We then expanded its ability to sense other quinone derivatives and found that L. plantation can differentiate between different analogs, such as DHNA and menadione, in a mixed quinone pool and generate distinguishable electrical signals. Finally, we tested the quinone bioelectronic sensor in a simulated gut fluid and explored the mechanism behind the quinone selectivity by creating a multivariate model. This work demonstrates a new strategy for rapidly and selectively sensing quinones and reveals the use of L. plantation as a probiotic exoelectrogen for health-related applications.

[0148] Results

[0149] In embodiments, to develop L. plantarum as a bioelectronic sensor to sense quinone, the output electrical signal can be proportional to the concentration of the quinone. We have shown that Ndh2, a membrane-bound type II NADH dehydrogenase, is required for EET in L. plantarum (Tejedor-Sanz et al., 2022). The addition of exogenous quinones, such as DHNA, triggers Ndh2 to oxidize intracellular NADH and reduce exogenous quinones thereby directly or indirectly (via PplA and / or EetA) transferring electrons from the cytosol to the extracellular electron acceptors (Tejedor-Sanz et al., 2022; Tolar et al., 2022) (Fig. 1 Panel A). The sequential reaction can be described as below:Ndh2NADH + DHNAOX- > NAD++ DHNAred(Equation 1)

[0150] Without wishing to be bound by theory, at steady state, the overall EET rate is determined by equation (1) and follows Michaelis-Menten kinetics (equations (3)- (5) in Materials and Methods). Therefore, the DHNA concentration is directly linked to EET,demonstrating the possibility of using L. plantarum to quantitatively sense quinone. To experimentally test this working model, it is required that the observ ed electrical signal is solely driven by Ndh2. However, enzymes that are not essential for EET, including DmkA, an octaprenyltransferase that converts DHNA to dimethylmenaquinone (DMK) (Light et al., 2018), and Ndhl, a type II NADH dehydrogenase that has a role in the aerobic-respiration- like response of L. plantarum (Brooijmans et al., 2009), also speculated to react with DHNA. These reactions can interfere with DHNA reduction driven by Ndh2 (Fig. 1 panel A).

[0151] We explored if this interference can be eliminated by generating a tailored EET pathway. To do so, we deleted dmkA and ndhl from L. plantarum NCIMB8826. A mutant with an additional ndh2 deletion served as the control to confirm the role that Ndh2 plays in EET. We compared the EET activity of the mutants by testing their ability7to reduce either iron (III) oxide ( Fe ' to Fe2+) or anode (produce current) in the presence of 5 pM DHNA. The MmkANndhl strain showed slightly decreased levels of iron reduction and current generation compared with the wildtype strain, likely because of branched or secondary’ EET pathways associated with DmkA or Ndhl. The additional ndh2 knockout significantly disrupted the EET activity (Fig. 1 panels B and C). Without wishing to be bound by theory, the majority of the electrical signal is driven by Ndh2 whereas DmkA and Ndhl are dispensable for EET activity7. Thus, we used tire INdmkANndhl mutant strain in the following experiments to test how L. plantarum responds to environmental quinones.

[0152] 2.2 L. plantarum sensitively responds to a wide range of DHNA concentrations following Michaelis-Menten kinetics

[0153] L. plantarum can serve as bioelectronic sensors for quinones, we can first determine the sensitivity, dynamics, and cell-dependence of the EET in response to physiologically- relevant concentrations of DHNA. Therapeutically, 5 nM DHNA produced by food-derived propionibacteria can stimulate the growth of bifidobacteria in vitro (ISAWA et al., 2002), 0.1 pM DHNA exhibits anti-inflammatory7activity7in vitro (Fukumoto et al., 2014), and 2 mg / kg DHNA inhibits colitis hi mice model (Fukumoto et al., 2014). However, a high concentration of DHNA (50-100 pM) induces significant cytotoxicity in mouse and human cell lines (Cheng et al., 2017). Under well-controlled laboratory conditions, microbes can synthesize DHNA to even higher concentrations, ranging from 1.8 to 236 pM (Eom et al., 2012; Furuichi et al., 2006, 2007). To determine how sensitively L. plantarum responds to environmental DHNA, we incubated the NdmkANndhl strain with varied DHNA concentrations and measured iron reduction levels at several time points over 12 h. We observed that L. plantarum can react with DHNA over a IO6concentration range from 5 nM to 500 pM (Fig.2 A and B), indicating a highly sensitive EET machinery with a large dynamic range. To analyze if EET can be calibrated to DHNA concentration following Michaelis-Menten kinetics, we calculated the iron reduction rate per cell over 12 h for each DHNA concentration (Fig. 7). The data can be fitted into the Michaelis-Menten equation (Fig. 2C) (equations (4) and (5) in the Materials and Methods), which confirms our working model. Moreover, the iron reduction rate remained constant over the experimental period, as can be seen from the linear increase in iron reduction levels (Fig. 7), implying adequate electron donor and acceptor over the experimental period to maintain electron transfer rate mediated by DHNA. These data indicate that L. plantanim can sen e as a quantitative bioelectronic sensor for DHNA.

[0154] The Michaelis-Menten equation also implies a correlation between Ndh2 concentration and EET rate (equations (4) and (5) in the Materials and Methods). Since the total amount of Ndh2 in a given culture volume can depend on the number of cells, we sought to explore how cell density' affects iron reduction and to determine an optimized cell density' that is suitable for sensing. We found that, for all the tested DHNA concentrations (0-500 pM), the iron reduction level at 12 h and the iron reduction rate over 12 h both decreased when the optical density (OD) of cells increased (Fig. 2 panels B and C). This effect can be attributed to the decreased carbon source per cell as the cell density' increased, which would decrease the NADH concentration and rate of iron reduction on a per cell basis. Since OD = 0.25 showed the fastest EET rate in response to DHNA (Fig. 2 panels B and C), this fixed cell density was chosen for quinone sensing.

[0155] 2.3 Temporal dynamics of EET reflect the concentration of environmentalDHNA in real-time

[0156] Bioelectronic sensors can be created by' integrating EET capable cells into bioelectrochemical systems (BESs) - hybrid biological and electro-chemical systems that allow continuous monitoring of electrical signals (Fig. 6 Panel B). The response speed of bioelectronic sensors can vary' substantially, from minutes (Atkinson et al., 2022) to hours (Golitsch et al., 2013; Webster et al., 2014), depending on the sensing strategy. To investigate how rapidly L. plantarum responds to environmental quinones, we inoculated the MmkANvdhl strain at the OD = 0.25 into the BESs and monitored the temporal dynamics of the electrical signal when physiological concentrations of DHNA (0.05-5 pM) were injected. We probed the electrical signal by calculating the difference in current density before and after DHNA injection produced on a per cell basis (A / per cell, a single BES contains ~1010cells). As a first test, the nutrient-rich chemically defined media (CDM) was used as the medium to support growth. Immediately upon the injection of DHNA into BESs, we observed a transient current spike(Fig. 3A). This transient spike was due to the abiotic oxidation of DHNA by the electrode, as a similar spike caused by DHNA was also observed in our previous study when heat-killed cells were inoculated (Tolar et al., 2022). Following the spike, the DHNA triggered EET within seconds, and a higher DHNA concentration resulted in a higher current magnitude (Fig. 3 panel A, left). The biotic current produced by EET then gradually increased over 24 h (Fig. 3A, right). We found that the current density per L. plant arum cell can be calibrated to DHNA concentrations with 99% confidence (p < 0.01) within 21s, highlighting the real-time response of the DHNA-induced EET (Fig. 3B). To ftirther dissect the temporal dynamics, we used linear regression to calculate the initial rate of current increase within 2 h (Fig. 8 Panel A). The increasing rate is also correlated with DHNA concentration (Fig.8 Panel B). Together, these data indicate that DHNA can dose-dependently activate EET, and the temporal dynamics of current produced by EET reflect DHNA concentrations in real time.

[0157] To test the EET dynamics when cells were in a resting, rather than growing state, we repeated the above experiment in phosphate buffered saline (PBS), which lacks nutrients for protein synthesis and cell proliferation. Again, EET was dose-dependently activated by DHNA within seconds after DHNA injection (Figs. S4A-B). The varied DHNA concentrations can be differentiated with 99% confidence within 36 s (Fig. 9 Panel C). Interestingly, the biotic current reached a significantly smaller plateau (~5 x 10-4vs. -2 x 10-2fA / cm2per cell) in less time (-7.5 h vs. -24 h) under these resting conditions, indicating that cell proliferation and-'or gene expression increase EET. Nonetheless, these results demonstrate that L. plantarum can sense and respond to DHNA in a dose-dependent manner in both nutrient-rich and nutrient-poor media. Effective bioelectronic sensors accurately report the concentration of analyte independent of other physicochemical variations in the sample. To rigorously investigate the reliability and reproducibility of DHNA sensing in different media, we compared the current produced by tire L. plantarum MmkA\ndhl strain in PBS or CDM in response to different concentrations of DHNA, and used adjusted R2to determine the data similarity. In the course of 24 h post DHNA injection, we found a consistent moderate to high adjusted R2between these two datasets (Fig. 10), indicating that the temporal dynamics of current produced by EET can reproducibly reflect DHNA concentrations irrespective of media components. However, the adjusted R2started to fluctuate over 24h after reaching the plateau within 5 min (Fig. 10). This indicates that, although data from different media can be aligned, the absolute level of current density (dQ / dt per cm2) for a fixed DHNA concentration is still dependent on whether the cells are in PBS or CDM. As an alternative, we calculated the average current density (AQ / At per cm2) to focus on the temporal evolution of current change rather than the absolute curreni level. We observed an improved consistencybetween the two datasets by using this analysis (Fig. 3 panel C). We conclude that, although the absolute level of current density is suitable for short-term sensing, the average cunent density is more suitable for long-term monitoring of DHNA. This provides a methodology for data comparison in bioelectronic sensing.

[0158] We next sought to explore if / ., plantarum can continuously sense environmental quinone by adding incremental concentrations of DHNA (0.01-10 pM) in a stepwise maimer with CDM as the assay medium. With every successive addition of a higher concentration of DHNA at a 1 - h interval, we observed a steeper current density in response (Fig. 3 panel D). The speed of the cunent increase con-elates with the accumulated DHNA concentrations (Fig. 8 Panel C). Again, there is a transient current spike induced by DHNA injection starting at 1 pM DHNA (Fig. 3 panel D). The spike magnitude showed no difference between EET and medium control, confirming this signal arises from the abiotic reaction between DHNA and electrode (Fig. 11). Following the spike, the current for medium control gradually went down, indicating the increased current for the tsdmkAkndhl strain was solely driven by EET (Fig. 3 Panel D). These results indicated that the L. plantarum can continuously sense DHNA, adjust its EET activity accordingly, and report the DHNA concentration as electrical signals in real time.

[0159] 2.4 L. plantarum selectively senses pharmacological-relevant quinones in a complex environment

[0160] In the preceding experiments, we have reported on the sensitivity, rapidity, dynamics, reproducibility, and continuity of DHNA sensing by L. plantarum. However, quinones encompass a large group of redox-active aromatic compounds derived from core structural motifs, with DHNA belonging to the class of 1 ,4-naphthoquinone (1,4-NQ) derivatives. It has been reported that L. plant arum can utilize different 1,4- NQ derivatives to support EET (Stevens et al., 2023). However, the selectivity among these derivatives remains unknown. Therefore, we investigated if L. plantarum can discriminate specific quinone derivatives from an environmental quinone pool containing diverse structural analogs, and thus, behave as a selective quinone bioelectronic sensor (Fig. 4 panel A). We tested six quinone derivatives (one 1 ,4-benzoquinone and five 1,4-NQ derivatives) that have pharmacological effects. These derivatives were hydroquinone (depigmenting agent) (Schwartz et al., 2022), 1 ,4- NQ (anti-cancer agent) (Kayashima et al., 2009), DHNA (anti-inflammation agent, bifidogenic growth stimulator) (ISAWA et al., 2002; Fukumoto et al., 2014; Cheng et al., 2017), menadione (vitamin K3) (Hirota el al., 2013), juglone (herbicide) (A. K. M. Mominul Islam and Joshua R. Widhalm, 2020), and DCDMQ (anti-cancer agent) (Kanaan et al., 2009). Without wishing to be bound by theory, Ndh2 can play a role in quinone recognition. We injected the six quinonederivatives individually to either the AdmkAhndhl or the kdmkANndhl / 2 strains and monitored the differential current density produced by these two mutants. Intriguingly, the MmkA&ndhl strain produced more current when reacting with DHNA or menadione compared to the hdmkAkndhl / 2 strain, while no difference was observed when reacting with other quinone derivatives (Fig. 12 Panel A). This observation indicates thatNdh2 can specifically recognize DHNA and menadione to support EET.[00161 [ To test if L. plantarum can selectively sense DHNA and menadione in a continuous way, we injected the six quinone derivatives sequentially into the same batch of cells and monitored the temporal dynamics of current density. All quinone derivatives were injected at the same concentration ( 1 pM). When hydroquinone, 1,4-NQ, juglone, or DCDMNQ were sequentially added, the two mutants produced a similar current in response (Fig. 4 Panel B). However, when DHNA and menadione w'ere further injected, the MmkAkndh i strain produced more current than the &dmkAkndhl / 2 strain (Fig. 4 Panel B). The rather flat current of the kdmkA&ndhl strain in response to DHNA was due to the influence of the still decreasing current from the prior injections. And we attributed the drop in current after 28 h to the decreasing pH in the medium, which was also noticed in our prior study (Tejedor-Sanz et al., 2022).

[0162] To isolate the signal for Ndh2-dependent EET, we calculated the differential current density' by subtracting the current produced by the \dmkA\ndhl / 2 strain from that produced by the kdmkAkndhl strain (Fig. 4 Panel C). The Ndh2 -independent EET does not significantly affect the differential current density, while the Ndh2-dependent EET results in an increased differential current density. Moreover, when injected at the same concentration, menadione resulted in a more rapid increase in the differential current density than DHNA. In this way', DEINA and menadione can be further differentiated (Fig. 4 panel C). Thus, although several quinone structural analogs were present in the same environment, the Ndh2-dependent EET in L. plantarum can continuously and selectively sense EET-supportive quinone derivatives and produce distinguishable signals.[00163J We further explored if L. plantarum can distinguish DEINA and menadione from the other quinones when they were all injected together. In addition to CDM, we also tested in a fasted state simulated intestinal fluid (FaSSIF) to explore L. plantarum's sensing ability’ in a simulated gut environment (Dahlgren et al., 2021). The inert quinone derivatives (hydroquinone, juglone, DCDMNQ, 1,4-NQ) w'ere added at a lower concentration (0.1 pM) in FaSSIF than in CDM (1 pM) to mimic the physiological quinone background in the human gut. To these background media, we then added DHNA, menadione, or both and calculated the differential average cunent density' to neutralize signal fluctuations and analyze the Ndh2-dependent EET (Fig. 13). In bothmedia, we observed the highest differential average current density when DHNA and menadione were added together, followed by menadione and DHNA alone (Fig. 4D). This data demonstrates that DHNA and menadione have an additive effect in activating Ndh2-dependent EET. Thus, L. plantarum can produce distinguishable electrical signals in response to different quinone combinations in a mixed quinone pool. The time needed to differentiate DHNA, menadione, and the mixture with 95% confidence was 37.3 min and 1.58 min in CDM and FaSSIF, respectively (Fig. 4E). The longer time needed in CDM is attributed to the transient Ndh2-independent EET induced by the inert quinone derivatives that masked the signals from Ndh2-dependent EET (Fig. 13). In an environment with lower background, such as in FaSSIF, the time needed to differentiate DHNA and menadione is ~20-fold lower.

[0164] 2.5 Mechanism of L. plantarum as a selective quinone bioelectronic sensor

[0165] Based on all our results, we sought to dissect the fundamental principles of how L. plantarum behaves as a selective bioelectronic sensor. We treated the reaction between quinone and the Ndh2-dependent EET pathway as a series of four successive steps: 1) quinone entering the cell membrane, 2) binding to Ndh2, 3) transferring electrons from the co-factor flavin adenine dinucleotide (FAD), 4) semi or hydroquinone leaving Ndh2 and transferring electrons through the rest of the EET pathway (Fig. 5 panel A). Without wishing to be bound by theory, quinones can be selected by L. plantarum for EET through these four steps based on their physicochemical properties (Fig. 5 panel B). These properties include lipophilicity / hydrophilicity, which determines if quinone can freely move in and out of the cell membrane, binding affinity, which determines how tight quinone binds to Ndh2, and redox potential, which determines the tendency to acquire and lose electrons.

[0166] To quantitatively analyze these properties, we determined the following three parameters for each quinone derivative: the calculated partition coefficient (cLogP) (Materials and Methods) (Leo et al., 1975); the modeled binding free energy between quinone and Ndh2 (Materials and Methods, Fig. 5 Panel C, 14) (Trott and Olson, 2010); and either the experimentally measured or indicated redox potential (Materials and Methods, Figs. 15 and 16). Paramenters are shown in Table 1. We then plotted these parameters against the differential average current density induced by each quinone derivative (based on the data shown in Fig. 12 Panel A) to analyze the correlations between EET and quinone parameters (Fig. 17). For redox potential, previous work shows that quinones prefer to undergo one electron transfer (le‘) in an aprotic environment (quinone^semiquinone, Fig. 15 Panel C), such as in the cell membrane, and revealed a correlation between quinone le’ redox potential and biophotocurrent generated bypurple bacteria (Grattieri et al., 2019). Indeed, we found a weak correlation between Ndh2- dependent EET and le' redox potential (adj R2= 0.33) (Fig. 17 Panel A), while no correlation was observed for 2H+ / 2e' redox potential (adj R2= 0.13) (Fig. 17 Panel B). However, le' redox potential alone fails to explain why juglone, 1,4-NQ, and hydroquinone, which have a lower le' redox potential than DHNA, cannot be utilized by EET. For quinone-Ndh2 interaction, we first used AlphaFold to model the Ndh2 structure for L. plantarum (Fig. 14) (Jumper et al., 2021). Then, we used AutoDock Vina to simulate quinone-Ndh2 binding (Trott and Olson, 2010). We were able to model a similar quinone binding pocket in the Ndh2 structure of L. plantarum compared to the homologs in other bacteria or yeast (Feng et al., 2012; Heikal et al., 2014), demonstrating the reliability of our simulation methods (Fig. 5 Panel C). However, no obvious correlation was found between the binding energy and EET (adj R2=-0.23) (Fig. 17 Panel C). For lipophihcity / hydrophilicity, the cLogP value of quinone or semiquinone also fails to con'elate wdth EET (adj R2= -0.23, -0.24, respectively) (Figs. 17 Panels D and E). Thus, we concluded that a single parameter is insufficient to describe the relationship between quinone and Ndh2- dependent EET. This finding is not w'holly surprising given that our mechanistic model indicates all of these three parameters would affect EET.

[0167] Table 1 : Quinone parameters summary11cLogP wus modeled by ChemDraw 21.0.0IIImodeled le" redox potential. See also Fig. 16.IVConverted from Prince et al.(Prince et al., 2022) n.d., not determined

[0168] Building upon prior work (Grattieri et al., 2019; Guo et al., 2017; Rhodes et al., 2021), we performed a multivariate analysis by integrating all parameters. To decrease the variate number, we calculated the differential of cLogP between quinone and semiquinone and treated it as a single variate (cLogPq - cLogPo-) (Fig. 17 Panel F). We first normalized tire current density and all the variates to the unsubstituted 1,4-NQ (i.e., all variates for 1,4-NQ = 1). Then, multivariate regression was performed and we termed the model output as “quinone score”. We observed a positive correlation between the quinone score and the normalized EET (adj R2= 0.6) (Fig. 5D). indicating that the multivariate model is able to interpret the quinone-EET relationship. However, some of the outliers cannot be explained by the model, indicating that there are still parameters that our model does not cover. For example, 1,4-NQ has a relatively high quinone score but cannot be utilized by Ndh2-dependent EET.

[0169] Nonetheless, to test the predictive power of our model, modeled the quinone score for lawsone (2-hydroxy-l,4-naphthoquinone) - a natural hair dye present in henna plants with a relatively low le’ redox potential (Bhuiyan et al., 2017; Prince et al., 2022). The quinone score for lawsone is higher when compared to the EET-negative control but lower than those for DHNA and menadione (Fig. 5E). This score indicates that L. plantarum can use lawsone as an electron shuttle in its Ndh2-dependent EET pathway, but will generate a lower current compared to DHNA or menadione. Indeed, when lawsone was supplied, the kdmkAkndhl strain produced more current than the \dmkA\ndhl / 2 strain (Fig. 12 Panel B). When comparing the differential average current densities, the magnitude is higher for lawsone than that for hydroquinone (negative control) but lower than those for DHNA or menadione (Fig. 5F). Thus, our multivariate model can identify new quinone derivatives that can be sensed by Ndh2-d ependent EET. Using this model, the electrical signals produced by L. plantarum can be decoded to distinguish between quinone structural analogs.

[0170] 2.6 Discussion

[0171] Here, we investigated the potential of L. plantarum to serve as a bioelectronic sensor for quinones. Using a genome-modified L. plantarum strain, we demonstrated its ability to sensitively, rapidly, and continuously sense varying concentrations of DHNA in both nutrient-rich and nutrient-poor media following Michaelis-Menlen kinetics. Further analysis revealed that the Ndh2-dependent EET in L. plantarum can selectively sense menadione.DHNA, and lawsone, amongst other quinone structural analogs, and we created a multivariate model to partially explain this selectivity.

[0172] Our multivariate model reveals general principles underlying L. plantarum's ability to selectively sense quinones. The model indicates that Ndh2 -dependent EET mainly discriminates based on le‘ redox potential of quinones, followed by its binding affinity to Ndh2, and least on its lipophilicity' or hydrophilicity (Fig. 5 Panel D). The positive and negative coefficients of the multivariate equation also imply that, in general, quinones with a higher tendency to lose electrons (lower le" redox potential), higher binding affinity to Ndh2, and a smaller differential in lipophilicity / hydrophilicity (so that can easily move in and out the cell membrane) are more likely to be utilized by L . plantarum for EET. There are a few other models that examine the structure-function relationships between quinone and EET (Grattieri et al., 2019; Rhodes et al., 2021). Our model goes beyond these prior models by not only considering quinone’s physicochemical properties but also the biological interaction between quinone and a specific reductase (Ndh2). However, our model is limited by being based on a relatively small set of 1 ,4-naphthoquinones with similar physicochemical properties.

[0173] Our study reveals a new whole-cell bioelectronic sensor chassis - L. plantarum, and a new strategy to sense quinones. State-of-the-art whole-cell bioelectronic sensors rely on Shewanella oneidensis (Golitsch et al., 2013; Webster et al., 2014; West et al., 2017), Geobacter sidfurreducens (Atci et al., 2016), or lab-engineered Escherichia coll (Atkinson et al.. 2022), and are limited in their applications to environmental sensing. By demonstrating that L. plantarum, a probiotic gut microbe, can sense quinones in a simulated intestinal fluid (Fig. 4C), our study expands the applications of bioelectronic sensors to include sensing within mammalian hosts. Moreover, we provide a new approach to sense quinones through a specific EET pathway to produce colori- metric (reducing iron) or amperometric signals (reducing electrode). Quinones are known to mediate electron transfers in different exoelectrogens (Franza and Gaudu, 2022), and a prior study explored different 1,4-benzoquinone derivatives to mediate photocurrent generated by purple bacteria (Grattieri et al., 2019). However, these quinones have not been characterized as analytes that can be selectively sensed by exoelectrogens. This is because a single quinone can simultaneously involve in multiple electron transfer pathways, regulating cellular metabolism and physiology at the systems level, which makes signal quantification and interpretation difficult (Franza and Gaudu, 2022). In contrast, our work established that quinones are utilized by L. planlarum through Ndh2, and the deletion of this specific NADH- dehydrogenase diminishes EET (Tejedor-Sanz et al., 2022; Tolar et al., 2022) (Fig. 1). This prior understanding of the EET mechanism allowed us to focus on signals generated by Ndh2 andquantify signals via Michaelis-Menten kinetics (Fig. 2). Since the quinone is re-oxidized by the electrode, the process of sensing does not change the environment. This is in contrast to other EET-based bioelectronic sensors that consume donors / acceptors or oxyanions during the sensing process (Atci et al., 2016; Atkinson et al., 2022; Zhou et al., 2017). The recycling of quinone also amplifies the signals, making the limited detection concentration of our bioelectronic sensor (0.005 or 0.01 pM DHNA by iron reduction assay or electrochemical assay, respectively) 100 to 200-fold lower than HPLC (1 pM DHNA) (Takebayashi et al., 2008).

[0174] Our work also demonstrates a new concept for improving analyte selectivity in whole cells that can be adapted to sense redox-active molecules. Whole-cell biosensors offer advantages such as long-term operation and the ability to function in complex environments, but many of them depend on analyte-specific receptors that modulate transcription or other cellular activities. This can be problematic when dealing with structural analogs, as many receptors must be created to detect each individual analyte (d’Oelsnitz et al., 2022). In contrast to specific sensing, selective sensing can be achieved through the use of nonspecific sensing elements, such as nanoparticles, quantum dots, peptides, or nonspecific transcriptional factors, to simultaneously detect a class of analytes (Peveler et al., 2016; Graham et al., 2020; Ceto' et al., 2012). However, these sensors must be carefully arranged in a complex array to generate unique “fingerprints” to identify individual components (Ceto' et al., 2012; Graham et al., 2020; Peveler et al., 2016). Our quinone bioelectronic sensor represents a new type of whole-cell biosensor that offers both selectivity and specificity (Fig. 18). It specifically senses a subset of quinones - 1,4- naphthoquinones, and selectively discriminates analogs, such as DHNA, menadione, and lawsone, based on the differential increments of the current density over time (Figs. 4 panel C and 5 panel F). This difference in the output signal is a result of the differing kinetics of Ndh2- dependent reduction of these analogs, which can be described using a multivariate model (Fig. 5 panel D). It is worth noting that the current study uses the same concentration (1 u.M) for each individual quinone in the mixed pool. Thus, it remains to be explored whether one can distinguish between quinone analogs when mixed at different concentration ratios. One way to address this challenge is to build a machine learning model based on the dynamic s of the cunent response to precisely identify individual components at different concentrations (Ceto' et al., 2012; Graham et al., 2020).

[0175] Moreover, the quinone bioelectronic sensor in this study leverages the native Ndh2 of L. plantarum and shows a preference for sensing 1,4- naphthoquinones (Figs. 4 Panel B, 12), whereas Ndh2 homologs in other organisms show broader binding to ubiquinones or menaquinones (Feng et al., 2012; Heikal et al., 2014). It can be possible to use protein evolutionto modify Ndh2 of L. plantarum to sense other types of quinone derivatives, such as quinonederived cancer drugs or air pollutants, for rapid drug monitoring and pollution control. For practical detection in the field with a smaller volume of samples (~pl), the quinone sensor can be integrated into miniature bioelectrochemical systems (Zhou et al., 2017) or ingestible capsules (Mimee et al., 2018). Hence, this work opens the possibility of using L. plantarum as a probiotic chassis for EET-based bioelectronic applications in the fields of environment, food, and medicine.

[0176] 2.7 Conclusions

[0177] Rapid and accurate detection of chemical analogs will allow a deeper understanding of the structure-activity relationship, aiding in drug development, toxicity assessment, and quality control across industries. Herein, we present selective bioelectronic sensing of quinone analogs that leverages differing EET kinetics derived from promiscuous reactions between quinone analogs and the oxidoreductase Ndh2. This new sensing strategy' holds inherent speed and sensitivity owing to fast occurring electron transfer in whole cells and is adaptable for the selective detection of other redox-active molecules. Although it remains a challenge to identify individual analogs and report their concentrations from a quinone mixture, our multivariant model points a wav forward to using key descriptors to decipher the intermixed signals. Our work showcases that whole-cell bioelectronic sensors can function as analytical machinery for complex sensing of chemical analogs, which will serve as powerful tools for environmental monitoring.

[0178] 3. Materials and Methods

[0179] 3.1 Strains and culture conditions

[0180] Strains used in this study are listed in Table 4. Escherichia coli 10- beta (New England Biolabs) was used for plasmid construction. Lactiplantibacillus plantarum NCIMB8826 was used as the parental strain for all the experiments. L. plantarum strains were grown in commercial MRS (HiMedia) from glycerol stocks at 37 °C without shaking. To grow a large volume of L. plantarum for a subsequent iron reduction assay or a bioelectrochemical assay, cells were subcultured in modified MRS containing 1% (w / v) mannitol and grown at 37 °C without shaking (De MAN et al., 1960; Tejedor-Sanz et al., 2022). In bioelectrochemical systems, Chemically Defined Medium (CDM) containing 1% (w / v) mannitol (Table 2) was used as the culture medium for L. plantarum. For testing in a bio-relevant environment, the Fasted State Simulated Intestinal Fluid (FaSSIF, Biorelevant, UK) was prepared as indicated in Table 3. Forboth iron reduction and bioelectrochemical assays, cells were tested at 30 °C under anaerobic conditions.[001811 Table 2: Chemically defined medium recipe

[0182] Table 3: Fasted state simulated intestinal fluid recipe00183] Table 4: Strains used in this example

[0185] L. plantarum SL046 (kdmkA&ndhl} and SL052 (kdmkAkndhl \ndh2) strains were constructed by using the CRISPR-Cas9 toolbox developed by Huang et al. (2019). The crRNAs targeting each gene were designed by using CRISPOR (http: / / crispor.tefor.net / '). The sgRNA fragment and the homologous arms flanking each gene were cloned into theApal-Xbal digested pHSP02 backbone by Gibson assembly to create the editing plasmids pSTS04, pSL47, pSL51 that target dmkA, ndhl, ndh2, respectively (primers and oligos are listed in Table 6). For CRISPR editing. L. plantarum NCIMB8826 strain harboring helper plasmid pLHOl was induced with 100 ng / ml Sakain P peptide (GenScript) for RecE / T expression and was subsequently prepared as competent cells. The editing plasmid was then delivered into L. plantarum by electroporation. The transformed cells were spread on MRS plates containing 10 pg / mL erythromycin and 10 ug / mL chloramphenicol to screen for the deletion mutants. We found that the edited L. plantamm mutant does not retain the helper plasmid pLHO 1 and cannot grow with antibiotic selection in liquid MRS culture. Thus, for sequential editing, we inoculated the L. plantarum mutant in antibiotics-free MRS for two passages and screened for the colonies with both the helper plasmid and the editing plasmid being cured. Then, we electroporated the helper plasmid back to the mutant strain for the next round of gene deletion.

[0186] Table 5: Plasmids used in this example00187] Table 6: Primer used in this example

[0188] *Gray color shows the overhang of each primer for assembly

[0189] 3.3 Michaelis-Menten kinetics

[0190] Without wishing to be bound by theory', DHNA reduction by Ndh2 can follow Michaelis- Menten kinetics when the intracellular NADH is sufficient to support EET. DHNA oxidation by terminal electron acceptor can be either direct or indirect (involving intermediate steps via PplA and / or EetA) (Tolar et al., 2022). When an insoluble terminal electron acceptor is used (such as metal iron or electrode), and the DHNA oxidation is fast enough, this step can be treated as a pseudo-first-order reaction. Under this assumption, the apparent EET rate (VEET, the rate that terminal electron acceptors are reduced) is equal to DHNA oxidation rate following first- order kinetics. At the quasi-steady-state, the concentration of reduced DHNA does not change. This means DHNA reduction rate is equal to DHNA oxidation rate: 3)(Equation 4) where X is the first-order rate constant, Vmax is the maximum DHNA reduction rate, Kmis the Michaelis constant, kcalis the turnover number of Ndh2. This implies that the apparent EET rate(VEET) is a Michaelis- Menten function of oxidized DHNA concentration. Because it is unpractical to measure the concentration of oxidized DHNA at steady-state, we assume [DHNAox] » [DHNAred] when DHNA oxidation is fast enough and treat the total DHNA concentration [DHNAtotai] as an approximation of the oxidized DHNA concentration: (Equation 5)1

[0191] Therefore, we used equation (5) to fit the experimental data into the Michaelis-Menten equation to create the calibration curve between the EET rate and DHNA concentrations. Notably, Hill function (the general form of the Michaelis-Menten equation) has been validated to describe the calibration dependences of biosensors (Kurganov et al., 2001 ). All data fitting and analysis were performed in either OriginLab or MATLAB.

[0192] 3.4 Iron reduction assay

[0193] After overnight growth, cells were harvested at 4000 rpm, 4 °C, for 15 min. Cells were washed twice with lx PBS and resuspended in lx PBS to ODeoo = 4. CFU was then measured bydilution plating. In an anaerobic chamber (Whitley A45 anaerobic workstation), the concentrated cells were inoculated in a 96-deep well plate (cells were series diluted if necessary) and mixed in a 1: 1 ratio with a 2x iron master mixture prepared in PBS. The final composition of the reaction mixture was: 2 mM iron (III) oxide nanoparticles (<50 nm, Sigma-Aldrich), 2 mM ferrozine, 1% (m / v) mannitol, indicated concentration of DHNA, and indicated cell ODeoo. After incubating at 30 °C for indicated hours, a 150 pl aliquot of the reaction mixture was harvested, and 100 pl supernatant was collected after centrifuging at 4000 rpm for 10 min. The absorbance of the supernatant was measured at 562 nm by a plate reader (Tecan Spark). The concentration of Fe2+in each sample was calculated by a standard curve prepared with FeSOr (0-0.4 mM, 2-fold increase) and was normalized to CFU. The scanned image showing the color range was taken by a desktop scanner in a 96-well white bottom plate.

[0194] 3.5 Bioelectrochemical system construction

[0195] We used waler-jacketed dual-chamber bioelectrochemical systems (BESs) (Adams & Chittenden Scientific Glass) and a VMP-300 potentiostat (BioLogic) for all the bioelectrochemical measurements. The anodic chamber of the BES contained an Ag / AgCl reference electrode filled with 3M KC1 (CHI 11, CH Instruments) and a 6.35-mm-thick graphite felt working electrode with a 16-mm radius size (Alfa Aesar) threaded to a 0.5-mm radius titanium wire (Alfa Aesar). The cathodic chamber contained a titanium wire as the counter electrode. The anodic chamber was separated from the cathodic chamber by using a cation membrane (CMI-7000, Membranes International).

[0196] The BESs were filled with ddH?O and sterilized by autoclaving. The media in the anodic chamber was then replaced by 1 10 ml lx CDM or PBS containing 1% mannitol. The media in the cathodic chamber was replaced by lx M9 salts (BD Difico). A magnetic stir bar was placed in the anodic chamber for continuous stirring at 220 rpm (IKA RO10 Magnetic Stirrers). N? gas was purged continuously into the anodic chamber to maintain anaerobic conditions. The BESs were kept at 30 °C by connecting the water-jackets to an ECO ES4 heating circulator (Lauda- Brinkmann). To cany out bioelectrochemical measurements, the working electrode was poised at -j-0.2 V versus Ag / AgCl and clironoamperometry was carried out to record the current every 36 s. Once the current stabilized, the cells were inoculated for the subsequent bioelectrochemical analysis.

[0197] 3.6 Bioelectrochemical assay with quinone addition

[0198] Overnight cultures of L. plantarwn were harv ested at 4000 rpm, 4 °C, for 15 min. Cells were washed twice and resuspended in lx PBS to OD&oo = 9. 16 per ml. A 3 ml aliquot of the cell resuspension was injected into the BESs so that the final ODfioo = 0.25 per ml. To measure CFU, a 200 pl aliquot of sample was taken from each BES safer cells were homogeneously stirred. The CFU was then determined by dilution plating. The quinone of interest was injected after the current had stabilized.

[0199] DHNA, hydroquinone. 1 ,4-naphthoquinone, juglone, DCDMNQ, menadione, and lawsone (Sigma-Aldrich) were prepared freshly in 100% dimethyl sulfoxide (DMSO). A 200 pl aliquot of a 550x stock quinone solution with varied concentrations was then injected into the BESs containing 1 10 ml medium for the desired final concentration (lx). The current was recorded every Is to monitor the transient current change upon quinone addition.

[0200] 3.7 Redox potential measurement or indication

[0201] Cyclic voltammetry (CV) measurement of quinone le’ or 2e72ET redox potential was conducted in a single chamber electrochemical cell with a glassy carbon working electrode and a Pt mesh counter electrode. All solutions were sparged with N2 prior to the measurement and N2 was continuously blown over the solution during the CV scan to maintain an anaerobic condition. The 2e72H+redox potential was measured in 100 mM MOPS buffer containing 100 pM quinone of interest. CV was performed v.s. Saturated Calomel Electrode (SCE) reference electrode at a scan rate of 50 mV / s. The le‘ redox potential was measured in acetonitrile containing 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) as the electrolyte (Huynh et al., 2016). Each quinone of interest was tested at 10 mM. Ag / Ag;was used as the reference electrode. Ferrocene of 5 mM was added as the internal standard. CV was performed at a scan rate of 50 mV / s.

[0202] The le’ redox potential for DHNA was not experimentally accessible due to the lack of purified oxidized DHNA. We adapted and modified the method reported by Prince et al. to correlate the Hammett oparaconstants with 1 e' redox potentials (Prince et al., 2022). We then used the trend line to indicate the 1 e" redox potential for DHNA (see Fig. 17).

[0203] 3.8 Quinone-Ndh2 binding simulation

[0204] The three-dimensional structure of Ndlt2 from L. plantarum was indicated by AlphaFold (Jumper et al., 2021). The truncated 1^409 aa region was used for molecular docking (Fig. 15). Quinone-Ndh2 docking was performed by using AutoDock Vina (Trott and Olson, 2010) with the searching exhaustiveness of 9 and the energy range of 24 (kcal / mol). The grid box size was set to 24 A x 34 A x 10 A and centered at position (8.885, -7.046, 6.636) in the coordinate space of the receptor. This space is demonstrated to be the quinone binding pocket according to previous studies (Feng et al., 2012; Heikal et al., 2014). The model with the highest binding energy and a proper posing of quinone in the binding pocket was chosen for further analysis.

[0205] 3.9 cLogP determination

[0206] The cLogP value for quinone or semiquinone was calculated by using ChemDraw 21.0.0.

[0207] References cited in this example

[0208] Abdullah, J., Ahmad, M., Heng, L.Y., Karuppiah, N., Sidek, H., 2007. An optical biosensor based on immobilization of laccase and MBTH in stacked films for the detection of catechol. Sensors 7, 2238-2250. https: / / doi.org / 10.3390 / s7102238.

[0209] Atci, E., Babauta, J.T., Sultana, S.T., Beyenal, H., 2016. Microbiosensor for the detection of acetate in electrode-respiring biofilms. Biosens. Bioelectron. 81, 517-523. https: / / doi.org / 10.1016 / j.bios.2016.03.027.

[0210] Atkinson, J.T., Campbell, I.J., Thomas, E.E.. Bonitatibus, S.C., Elliott, S.J., Bennett, G.N., Silberg, J.J., 2019. Metalloprotein switches that display chemical-dependent electron transfer in cells. Nat. Chem. Biol. 15, 189-195. https: / / doi.org / 10.1038 / s41589- 018-0192-3.

[0211] Atkinson, J.T., Su, L., Zhang, X., Bennett, G.N., Silberg, J.J., Ajo-Franklin, C.M., 2022.

[0212] Real-time bioelectronic sensing of environmental contaminants. Nature 611, 548—553. https: / / doi.org / 10. 1038 / s41586-022-05356-y.

[0213] Babula, P., Adam, V., Havel, L., Kizek, R., 2009. Noteworthy secondary metabolites naphthoquinones - their occurrence, pharmacological properties and analysis. CPA 5, 47-68. https: / / doi.org / 10.2174 / 157341209787314936.

[0214] Baruch, M., Tejedor-Sanz, S., Su, L., Ajo-Franklin, C.M., 2021. Electronic control of redox reactions inside Escherichia coli using a genetic module. PLoS One 16, e0258380. https: / / doi.org / 10.137 l / journal.pone.0258380.

[0215] Bhuiyan, M.A.R.. Islam, A., Ali, A.. Islam, M.N., 2017. Color and chemical constitution of natural dye hernia (Lawsonia inemiis L) and its application in the coloration of textiles. J. Clean. Prod. 167, 14-22. https: / / doi.Org / 10.1016 / j.jclepro.2017.08.142.

[0216] Bird, L.J., Kundu, B.B., Tschirhart, T., Corts, A.D., Su, L., Gralnick, J.A., Ajo- Franklin, C. M., Glaven, S.M., 2021. Engineering wired life: synthetic biology for electroactive bacteria. ACS Synth. Biol. 10, 2808-2823. https: / / doi.org / 10.1021 / acssynbio.1 c00335.

[0217] Bolton, J.L., Dunlap, T., 2017. Formation and biological targets of quinones: cytotoxic versus cytoprotective effects. Chem. Res. Toxicol. 30, 13-37. https: / / doi.org / 10.102 l / acs.chemrestox.6b00256.

[0218] Bolton, J.L., Trush, M.A., Penning, T.M., Dryhurst, G., Monks, T.J., 2000. Role of quinones in toxicology. Chem. Res. Toxicol. 13, 135-160. https: / / doi.org / 10.1021 / tx9902082.

[0219] Brooijmans, R.J.W., Vos, W.M. de, Hugenholtz, J., 2009. Lactobacillus plantarum WCFS1 electron transport chains. Appl. Environ. Microbiol. 75, 3580-3585. https: / / doi.org / 10.1128 / AEM.00147-09.

[0220] Cao, G„ Wang, W., Zhang, J., Wu, P„ Zhao, X., Yang, Z„ Hu, D„ Cai, Z., 2022. New evidence of rubber-derived quinones in water, air, and soil. Environ. Sci. Technol. 56, 4142-4150. https: / / doi.org / 10.1021 / acs.est.lc07376.

[0221] Ceto', X., C espedes, F., Pividori, M.I., Guti'errez, J.M., del Valle, M., 2012. Resolution of phenolic antioxidant mixtures employing a voltammetric bio-electronic tongue.

[0222] Analyst 137, 349-356. https: / / doi.org / 10.1039 / ClAN15456G.

[0223] Cheng, Y., Jin, U.-H., Davidson, L.A., Chapkin, R.S., Jayaraman, A., Tamamis, P., Orr, A., Allred, C., Denison, M.S., Soshilov, A., Weaver, E., Safe, S., 2017. Editor’s highlight: microbial-derived 1 ,4-Dihydroxy-2-naphthoic acid and related compounds as arylhydrocarbon receptor agonists / 'antagonists: structure-activity relationships and receptor modeling. Toxicol. Sci. 155, 458-473. https: / / doi.org / 10.1093 / toxsci / kfw230.

[0224] Cory, R.M., McKnight, D.M., 2005. Fluorescence spectroscopy reveals ubiquitous presence of oxidized and reduced quinones in dissolved organic matter. Environ. Sci. TechnoL 39, 8142-8149. https: / / doi.org / 10.1021 / es0506962.

[0225] Dahlgren, D., Venczel, M., Ridoux, J.-P., Skjo'ld, C., Miillertz, A., Holm, R., Augustijns, P., Hellstro ’m, P.M., Lennem' as, H., 2021. Fasted and fed state human duodenal fluids: characterization, drug solubility, and comparison to simulated fluids and with human bioavailability. Eur. J. Pharni. Biopharm. 163, 240-251. https: / / doi.org / 10.1016 / j.ejpb.2021.04.005.

[0226] De Man, J.C., Rogosa, M„ Sharpe, M.E., 1960. A medium for the cultivation of lactobacilli. J. Appl. Bacteriol. 23, 130-135. https: / / doi.org / 10.1111 / j.1365- 2672.1960 ,tb00188.x.

[0227] d’Oelsnitz, S., Kim, W., Burkholder, N.T., Javanmardi, K., Thyer, R., Zhang, Y., Alper, H. S., Ellington, A.D., 2022. Using fungible biosensors to evolve improved alkaloid biosyntheses. Nat. Chem. Biol. 18, 981-989. https: / / doi.org / 10.1038 / s41589-022- 01072-w.

[0228] Eom, J.-E., Kwon, S.-C., Moon, G.-S., 2012. Detection of l,4-dihydroxy-2- naphthoic acid from commercial makgeolli products. Prev Nutr Food Sci 17, 83-86. https: / / doi.org / 10.3746 / pnf.2012.17.1.083.

[0229] Feng. Y„ Li, W„ Li. J., Wang, J., Ge, J„ Xu. D., Liu, Y„ Wu, K., Zeng, Q„ Wu, L-W.,

[0230] Tian, C., Zhou, B., Yang, M., 2012. Structural insight into the type-II mitochondrial NADH dehydrogenases. Nature 491, 478-482. https: / / doi.org / 10.1038 / nature! 1541.

[0231] Franza, T., Gaudu, P., 2022. Quinones: more than electron shuttles. Res. Microbiol. 173, 103953 https: / / doi.Org / 10.1016 / j.resmic.2022.103953.

[0232] Fritts, R.K., McCully, A.L., McKinlay, J.B., 2021. Extracellular metabolism sets the table for microbial cross-feeding. Microbiol. Mol. Biol. Rev. 85, e00135 https: / / doi.org / 10.1128 / MMBR.00135-20, 20.

[0233] Fukuda, M., Qianjun, L., Kishikawa, N., Ohyama, K., Kuroda, N., 2019. Development of ultrafast colorimetric microplate assay method for ubiquinone utilizing the redox cycle of the quinone. Microchem. J. 150, 104104 https: / / doi.Org / 10.1016 / j. microc.2019.104104.

[0234] Fukumoto, S., Toshimitsu, T., Matsuoka, S., Maruyama, A., Oh-oka, K., Takamura, T., Nakamura, Y., Ishimaru, K., Fujii-Kuriyama, Y., Ikegami, S., Itou, H., Nakao, A., 2014. Identification of a probiotic bacteria-derived activator of the aryl hydrocarbon receptor that inhibits colitis. Immunol. Cell Biol. 92, 460—465. https: / / doi.org / 10.1038 / icb.2014.2.

[0235] Furuichi, K., Hojo, K., Katakura, Y., Ninomiya, K., Shioya, S., 2006. Aerobic culture of Propionibacterium freudenreichii ET-3 can increase production ratio of 1.4-dihy- droxy-2-naphthoic acid to menaquinone. J. Biosci. Bioeng. 101 , 464—470. https: / / doi.org / 10.1263 / jbb .101 .464.

[0236] Furuichi, K., Katakura, Y., Ninomiya, K., Shioya, S., 2007. Enhancement ofI,4-dihy- droxy-2 -naphthoic acid production by propionibacterium freudenreichii ET-3 fed- batch culture. Appl. Environ. Microbiol. 73, 3137-3143. https: / / doi.org / 10.1128 / AEM.01307-06.

[0237] Glasser, N.R., Saunders, S.H., Newman, D.K., 2017. The colorful world of extracellular electron shuttles. Annu. Rev. Microbiol. 71 , 731-751. https : / / doi . org / 10.1 146 / annurev-micro-090816-093913.

[0238] Golitsch, F., Bucking, C., Gescher, J., 2013. Proof of principle for an engineered microbial biosensor based on Shewanella oneidensis outer membrane protein complexes.

[0239] Biosens. Bioelectron. 47. 285-291. https: / / doi.org / 10. 1016 / j .bios .2013.03.010.

[0240] Graham, G., Csicsery, N., Stasiowski, E., Thouvenin, G., Mather, W.H., Ferry, M., Cookson, S., Hasty, J., 2020. Genome-scale transcriptional dynamics and environmental biosensing. Proc. Natl. Acad. Sci. USA 117, 3301-3306. https: / / doi. org / 10.1073 / pnas.19130031 17.

[0241] Grattieri, M., Rhodes, Z., Hickey, D.P., Beaver, K., Minteer, S.D., 2019. Understanding biophotocurrent generation in photosynthetic purple bacteria. ACS Catal. 9,

[0242] 867-873. https: / / doi.org / 10.1021 / acscatal.8b04464.

[0243] Guo, J.-Y.. Minko, Y., Santiago, C.B., Sigman, M.S., 2017. Developing comprehensive computational parameter sets to describe the performance of pyridine- oxazoline and related ligands. ACS Catal. 7, 4144-4151. https: / / doi.org / 10.1021 / acscatal.7b00739.

[0244] Heikal, A., Nakalani, Y., Dunn, E., Weimar, M.R., Day, C.L., Baker, E.N., Lott,J.S., Sazanov, L.A., Cook, G.M., 2014. Structure of the bacterial type II NADHdehydrogenase: a monotopic membrane protein with an essential role in energy generation. Mol. Microbiol. 91, 950-964. https: / 7doi.org'T0.1111 / 'mmi.12507.

[0245] Hirota, Y., Tsugawa, N., Nakagawa, K., Suhara, Y., Tanaka, K., Uchino, Y., Takeuchi, A., Sawada, N., Kamao, M., Wada, A., Okitsu, T., Okano, T., 2013. Menadione (vitamin K3) is a catabolic product of oral phylloquinone (vitamin KI) in the intestine and a circulating precursor of tissue menaquinone-4 (vitamin K2) in rats. J. Biol. Chem. 288, 33071- 33080. https: / / doi.org / T0.1074 / jbc.Ml 13.477356.

[0246] Huang, H., Song, X., Yang, S., 2019. Development of a RecE / T-assisted CRISPR-cas9 toolbox for lactobacillus. Biotechnol. J. 14, 1800690 https: / / doi.org / 10.1002 / biot.201800690.

[0247] Huynh, M.T., Anson, C.W., Cavell, A.C., Stahl, S.S., Hammes-Schiffer, S., 2016. Quinone 1 e- and 2 e- / 2 H+ reduction potentials: identification and analysis of deviations from systematic scaling relationships. J. Am. Chem. Soc. 138, 15903-15910. https: / / doi.org / 10.1021 / jacs.6b05797.

[0248] Irreverre, F., Sullivan, M.X., 1941. A colorimetric test for vitamin KI . Science 94, 497-498. https: / / doi.org / 10. 1126 / science.94.2447.497.b.

[0249] Isawa, K., Hojo, K., Yoda, N., Kamiyama, T., Makino, S., Saito, M., Sugano, H., Mizoguchi, C., Kurama, S., Shibasaki, M., Endo, N., Sato, Y., 2002. Isolation and identification of a new bifidogenic growth stimulator produced by propionibacterium freudenreichii ET-3. Biosci., Biotechnol., Biochem. 66, 679-681. https : / / doi . org / 10.1271 / bbb .66.679.

[0250] Jumper, J., Evans, R., Pritzel, A., Green, T., Figurnov, M., Ronneberger, O., Tunyasuvunakool, K., Bates, R., ZT'dek, A., Potapenko, A., Bridgland, A., Meyer, C.. Kohl, S.A.A., Ballard, A.J., Cowie, A., Romera-Paredes, B., Nikolov, S., Jain, R.,

[0251] Adler, J., Back, T., Petersen, S., Reiman, D., Clancy, E., Zielinski, M., Steinegger, M., Pacholska, M., Berghammer, T., Bodenstein, S., Silver, D., Vinyals, O., Senior, A.W., Kavukcuoglu, K., Kohli, P.. Hassabis, D., 2021. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583-589. https: / / doi.org / 10, 1038 / s41586- 021-03819-2.

[0252] Kanaan, Y.M., Das, J.R., Bakare, O., Enwerem, N.M,, Berhe, S„ Bevene, D., Williams, V., Zhou, Y„ Copeland, R.L., 2009, Biological evaluation of 2,3-dichloro-5,8- dimethoxy- 1 ,4-naphthoquinone as an anti-breast cancer agent. Anticancer Res. 29, 191-199.

[0253] Karbelkar, A.A., Reynolds, E.E., Ahimark, R., Furst, A.L., 2021. A microbial electrochemical technology to detect and degrade organophosphate pesticides. ACS Cent. Sci. 7, 1718-1727. https: / / d0i.0rg / l 0. 1021 / acscentsci. lc00931.

[0254] Kayashima, T., Mori, M., Yoshida, H., Mizushina, Y._ Matsubara, K., 2009.1 ,4- Naphthoquinone is a potent inhibitor of human cancer cell growth and angiogenesis. Cancer Lett. 278, 34—40. https : / / doi.org / 10 , 1016 / i , canlet.2008 , 12 , 020.

[0255] Khan, A., Salama, E.-S., Chen, Z., Ni, H., Zhao, S., Zhou, T., Pei, Y., Sani, R.K., Ling, Z., Liu, P., Li, X., 2020. A novel biosensor for zinc detection based on microbial fuel cell system. Biosens. Bioelectron. 147, 111763 https: / / d0i.0rg / l 0. 1016 / j . bios.2019, 1 11763.

[0256] Kumagai, Y., Shinkai, Y., Miura, T., Cho, A.K., 2012. The chemical biology of naphthoquinones and its environmental implications. Annu. Rev. Pharmacol. Toxicol. 52, 221- 247. https: / / doi.org / 10,1146 / annurev-pharmtox-O 10611-134517.

[0257] Kurganov, B.I., Lobanov, A.V., Borisov, I.A., Reshetilov, A.N., 2001 . Criterion for Hill equation validity for description of biosensor calibration curves. Anal. Chim. Acta 427, 1 1-19. https: / / doi.org / 10, 1016 / S0003-2670( 00)01167-3.

[0258] Leo, A., Jow, P.Y.C., Silipo, C., Hansch, C., 1975. Calculation of hydrophobic constant (log P) from .pi. and f constants. J. Med. Chem. 18, 865-868. https: / / doi.org / 10.1021 / jm00243a00L

[0259] Light, S.H., Su, L., Rivera-Lugo, R., Cornejo, J. A., Louie, A., lavarone, A.T., Ajo- Franklin, C.M., Portnoy, D.A., 2018. A flavin-based extracellular electron transfer mechanism in diverse Gram-positive bacteria. Nature 562, 140-144. https: / / doi. org / 10, 1038 / s41586-018-0498-z.

[0260] Maleki, N., Kashanian, S., Maleki, E., Nazari, M., 2017. A novel enzyme based biosensor for catechol detection in water samples using artificial neural network. Biochem. Eng. J. 128, 1-11. https: / / d0i.0rg / l 0, 1016 / j .bej.2017,09,005.

[0261] Mazzulli, J.R., Burbulla, L.F., Krainc, D., Ischiropoulos, H., 2016. Detection of free and protein-bound ortho-quinones by near-infrared fluorescence. Anal. Chem. 88, 2399- 2405. https: / / 'doi.org / 10, 102 l / acs.analchem,5b04420.

[0262] Mevers, E., Su, L., Pishchany, G., Baruch, M., Cornejo, J., Hobert, E., Dimise, E., Ajo- Franklin, C.M., Clardy, J., 2019. An elusive electron shuttle from a facultative anaerobe. Elife 8, e48054. https: / / doi.org / 10.7554 / eLife.48054.

[0263] Mimee, M., Nadeau, P„ Hayward, A., Carim, S., Flanagan, S., Jerger, L., Collins, J., McDonnell, S., Swartwout, R., Citorik, R.J., Bulovi'c, V., Langer, R., Traverse, G.,Chandrakasan, A.P., Lu, T.K., 2018. An ingestible bacterial-electronic system to monitor gastrointestinal health. Science 360, 915-918. htps: / / doi.Org / 10.l 126 / science.aas9315.

[0264] Mominul Islam, A.K.M., Widhalm, Joshua R., 2020. Agricultural uses of juglone: opportunities and challenges. Agronomy 10, 1500. https : / 7doi . or g / 10,3390 / agronomyl0101500.

[0265] Pei, J., Wang, Y., Yu, K., 2018. Sensitive determination of quinones by high- performance liquid chromatography-electrospray ionization-tandem mass spectrometry with methanol derivatization. Anal. Sci. 34, 335-339. https: / / doi.org / 10.21 16 / analsci .34.335.

[0266] Pereyra, C.E., Dantas, R.F., Ferreira, S.B., Gomes, L.P., Silva Jr., F.P., 2019. The diverse mechanisms and anticancer potential of naphthoquinones. Cancer Cell Int. 19, 207. https: / / doi.org / 10. 1186 / s 12935-019-0925-8.

[0267] Peveler, W.J., Yazdani, M., Rotello, V.M., 2016. Selectivity and specificity: pros and cons in sensing. ACS Sens. 1, 1282-1285. https: / / doi.org / 10, 1021 / acssensors ,6b00564. Prince, R.C., Dutton, P.L., Gunner, M.R., 2022. The aprotic electrochemistry of quinones.

[0268] Biochim. Biophys. Acta Bioenerg. 1863, 148558 https: / / doi.org / 10, 1016 / i , bbabio.2022, 148558.

[0269] Rhodes, Z., Simoska, O., Dantanarayana, A., Stevenson, K.J., Minteer, S.D., 2021. Using structure-function relationships to understand the mechanism of phenazinemediated extracellular electron transfer in Escherichia coli. iScience 24, 103033. https : / / doi.org / 10.1016 / i .isci.2021. 103033.

[0270] Schwartz, C., Jan. A., Zito, P.M.. 2022, Hydroquinone. In: StatPearls. StatPearls Publishing, Treasure Island (FL.

[0271] Shan. D., Li, Q., Xue, H., Cosnier, S., 2008. A highly reversible and sensitive tyrosinase inhibition-based amperometric biosensor for benzoic acid monitoring. Sensor.

[0272] Actuator. B Chem. 134, 1016-1021. https: / / doi.Org / 10.1016 / i.snb.2008.07.006.

[0273] Stevens, E.T., Beeck, W.V., Blackburn, B., Tejedor-Sanz, S., Rasmussen, A.R.M., Mevers, E., Ajo-Franklin, C.M., Marco, M.L., 2023. Lactiplantibacillus Plantarum Uses Ecologically Relevant, Exogenous Quinones for Extracellular Electron Transfer. mBio, e02234-23. doi: 10.1128 / mbio.02234-23.

[0274] Takebayashi, J., Nagata, J., Yamada, K., 2008. Improved analytical precision of 1,4- Dihydroxy-2-naphthoic acid by high performance liquid chromatography using dithiothreitol as mobile phase additive. FSTR 14, 509-512. https: / / doi.org'' 10,3136 / fstr, 14.509.

[0275] Tejedor-Sanz, S., Stevens, E.T., Li, S., Finnegan, P., Nelson, J., Knoesen, A., Light, S.H., Ajo-Franklin, C.M., Marco, M.L., 2022. Extracellular electron transfer increases fermentation in lactic acid bacteria via a hybrid metabolism. Elife 11 , e70684. https: / / doi.org / 10.7554 / eLife.70684.

[0276] TerAvest, M.A., Ajo-Franklin, C.M., 2016. Transforming exoelectrogens for biotechnology using synthetic biology. Biotechnol. Bioeng. 113, 687-697. https: / / doi.org / 10.10Q2 / bit.25723.

[0277] Tolar, J.G., Li, S., Ajo-Franklin, C.M., 2022. The differing roles of flavins and quinones in extracellular electron transfer in Lactiplantibacillus plantarum. Appl. Environ.

[0278] Microbiol. 89„ e01313-e01322. https: / / doi.org / 10. 1128 / aem.O1313-22.

[0279] Trott. O., Olson, A. J., 2010. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization and multithreading.

[0280] J. Comput. Chem. 31, 455. https: / / doi.org / 10.1002 / jcc.21334.

[0281] VanArsdale, E., Pitzer, J., Payne, G.F., Bentley, W.E., 2020. Redox electrochemistry to interrogate and control biomolecular communication. iScience 23, 101545. https: / / doi.org / 10, 1016 / j .isci.2020, 101545.

[0282] Webster, D.P., TerAvest, M.A., Doud, D.F.R., Chakravorty, A., Holmes, E.C., Radens, C. M., Sureka, S., Gralnick, J.A., Angenent, L.T., 2014. An arsenic-specific biosensor with genetically engineered Shewanella oneidensis in a bioelectrochemical system. Biosens. Bioelectron. 62, 320-324. https: / / doi.Org / 10.1016 / j.bios.2014.07.003.

[0283] West, E.A., Jain, A., Gralnick, J. A., 2017. Engineering a native inducible expression system in Shewanella oneidensis to control extracellular electron transfer. ACS Synth. Biol. 6, 1627—1634. https: / / doi.org / T0.1021 / acssynbio.6b00349.

[0284] Yang. Y., Yu, Y.-Y., Wang, Y.-Z., Zhang, C.-L., Wang, J.-X., Fang. Z., Lv. H., Zhong, J.-J.,

[0285] Yong, Y.-C., 2017. Amplification of electrochemical signal by a whole-cell redox reactivation module for ultrasensitive detection of pyocyanin. Biosens. Bioelectron. 98, 338-344. https: / / doi.org / T0.1016 / j.bios.2017.07.008.

[0286] Yu, Y.-Y., Wang, J.-X., Si, R.-W., Yang, Y„ Zhang, C.-L., Yong, Y.-C., 2017. Sensitive amperometric detection of riboflavin with a whole-cell electrochemical sensor. Anal.Chim. Acta 985, 148-154. https: / / doi.Org / 10.1016 / j.aca.2017.06.053.

[0287] Zhou, A.Y., Baruch, M., Ajo-Franklin, C.M., Maharbiz, M.M., 2017. A portable bioelectronic sensing system (BESSY) for environmental deployment incorporatingdifferential microbial sensing in miniaturized reactors. PLoS One 12, eO 184994. https: / / doi.org / 10.1371 / joumal. pone.0184994.

[0288] Zhou, T., Li, R., Zhang, S., Zhao, S., Sharma, M., Kulshrestha, S., Khan, A., Kakade, A., Han, H., Niu, Y., Li, X., 2021. A copper-specific microbial fuel cell biosensor based on riboflavin biosynthesis of engineered Escherichia coli. Biotechnol. Bioeng. 118, 210-222. https: / / doi.org / 10.1002 / bit.27563.

[0289] Huang, H., Song, X., Yang, S., 2019. Development of a RecE / T-Assisted CRISPR-Cas9 Toolbox for Lactobacillus. Biotechnology Journal 14, 1800690. https: / / doi.org / 10. 1002 / biot.201800690

[0290] Prince, R.C., Dutton, P.L., Gunner, M.R., 2022. The aprotic electrochemistry of quinones. Biochimica et Biophysica Acta (BBA) - Bioenergetics 1863, 148558. https: / / doi.org / 10. 1016 / j .bbabio.2022.148558

[0291] Tejedor-Sanz, S., Stevens, E.T., Li, S., Finnegan, P., Nelson, J., Knoesen, A., Light, S.H., Ajo-Franklin, C.M.. Marco, M.L., 2022. Extracellular electron transfer increases fermentation in lactic acid bacteria via a hybrid metabolism. eLife 11, e70684. https : / / doi . org / 10.7554 / eLife .70684

[0292] Trott, O., Olson, A.J., 2010. AutoDock Vina: improving the speed and accuracy of docking with a new scoring function, efficient optimization and multithreading. Journal of computational chemistry 31, 455. https: / / doi.org / 10.1002 / jcc.21334EXAMPLE 4

[0293] Portable bioeleetronic sensing using engineered microbes and miniature microbial fuel cells

[0294] In various embodiments, disclosed here is a low-cost, portable bioeleetronic sensing platform to detect environmental stimuli by integrating engineered microbial sensors, miniature microbial fuel cells (MFCs), and an exemplary strategy for electrical signal detection. Microbial bioeleetronic sensors, which translate chemical information into electrical signals, can require bulky, specialized, and expensive (>S10K) instrumentation for signal detection. To overcome certain limitations, in one embodiment the present disclosure relates to a compact, cost-effective (<$3) centimeter-scale MFC that allows for electrical signal readout using a handheld digital multimeter. Without being bond by theory, this approach reduces both the complexity and cost associated with microbial bioeleetronic sensing, making it more accessible for field applications. In the presently disclosed embodiment, the integrated bio-electronic hybrid system for small molecule detection is applied in real-world samples.Given its modular design, affordability, and ease of use, our platform can be readily reengineered and broadly adopted for commercial bioelectronic applications in environmental and health monitoring.

[0295] In one aspect, the present disclosure can includes process, a composition of matter, a device, in any combination thereof.

[0296] In one embodiment, the present disclosure relates to a bio-electronic hybrid device for portable and low-cost bioelectronic sensing.

[0297] In another aspect, the present disclosure relates to a method to detect electrical signals (voltage and current) from electroactive microbes and microbial bioelectronic sensors using MFCs and a digital multimeter.

[0298] In yet another aspect, the present disclosure presents an improved utility of certain microbial sensors and MFC devices.

[0299] Non-Limiting, Exemplary Materials and Methods:

[0300] Microbial fuel cell device manufacture under one embodiment

[0301] In one non-limiting, exemplary embodiment, the single-chamber MFC can be constructed using an Ag / AgCl-coated polyester sheet as the cathode and a porous carbon felt as the anode. The electrodes can be separated by a clay proton exchange membrane (PEM). The clay PEM can be made by exfoliating the thermally expanded natural vermiculite clay into atomically thin 2D flakes using a 20% aqueous HC1 solution. An aqueous dispersion of these clay nanosheets can be vacuum-filtered onto a cellulose nitrate membrane to fabricate a freestanding lamellar membrane. To make the anode, carbon ink (Kayaku C-250J) can be screen-printed onto a Polyethylene Terephthalate (PET) sheet, with a 1 mm hole punched at the center. A 6 mm-diameter, 3.18 mm-thick carbon felt round can then be attached to the wet carbon ink, and the ink can be allowed to dry. To make the cathode, Ag / AgCl ink (Kayaku AGCL-675) can be screen-printed onto a polyester sheet and allowed to dry. To assemble the device, a 10 mm diameter clay membrane can be placed on top of the cathode, and an acrylic well ( 19 mm wide x 19 mm long x 3 mm thick) can be affixed on the top of the clay membrane using glue. The carbon felt anode can then fitted into the acrylic well. The hole at the top of the anode can be designed to introduce bacteria into the device.

[0302] Signal detection from engineered microbial bioelectronic sensors using the MFC device under one embodiment

[0303] The screen-printed MFC devices can be treated with UV / ozone for 30 min to make carbon felt hydrophilic. A 16.7 ill. exponential-phase microbial sensor cell (e.g., e-COSENSco-cultures in the below results) can be added to a 1 .5 mL microcentrifuge tube containing 500 uL of testing media or real-world samples. The environmental stimuli (e.g., aTc in the below results) or blank ddH2O can be added to the respective tubes before inoculating microbial sensor cells. All tubes (three replicates for each group) can be incubated at 30 °C. After 1.5-2 hours, a 60 pL portion can be taken and slowly injected into the device using an insulin needle (BD). For electrochemical measurement using the potentiostat, the working electrode can be connected to the carbon anode of the device, and the reference electrode can be joined with the counter electrode and connected to the Ag / AgCl cathode of the device. Open circuit voltage (OCV) can be recorded every 3 s. To measure short circuit current (SCA), the voltage (E) can be swept from E = OCV to E = 0 vs. OCV at a scan rate of 10 mV / s. The current at E = 0 vs. OCV can be taken as the SCA. For OCV measurement using the digital multimeter (Fluke 280), the positive terminal (source) can be connected to the carbon anode, and the negative terminal (drain) was connected to the Ag / AgCl cathode. The OCV was measured as direct voltage (DC) in mV mode.

[0304] Results:

[0305] Prior electrical signal detection for microbial bioelectronic sensors was achieved using glassy electrochemical reactors and high-precision potentiostat; these instruments are cumbersome, high-cost, and have very limited portability. In one embodiment, to allow for low-cost, portable, and energy-friendly bioelectronic sensing, an electroactive microbial coculture sensing system (e-COSENS) can be integrated into a miniature clay-based microbial fuel cell (MFC) device for detecting electrical signals (Fig. 19a). The e-COSENS generates electric signals upon detection of environmental stimuli through l,4-dihydroxy-2-naphthoic acid (DHNA)-mediated extracellular electron transfer (EET). The device can be fabricated by sandwiching a proton exchange membrane (PEM) between a screenprinted Ag / AgCl cathode and a carbon anode stacked with a 6 mm carbon felt round (Fig. 19a). In this specific nonlimiting exemplary embodiment, the PEM was exfoliated from natural vermiculite clay, showing biocompatibility and ion conductivity of up to 2x 10-2 S / 'cm. Manufactured in Boston and shipped to Houston for testing, the devices maintained good performance, demonstrating chemical and mechanical stability. Moreover, in embodiments, a single device costs less than three dollars and can be manufactured in a high-throughput way (Fig. 20b).

[0306] We delected electrical signals from e-COSENS using the clay-based MFC device by measuring the open circuit voltage (OCV) and short circuit current (SCA) (Fig. 19b). The OCV can measure the potential difference between anode and cathode; a faster drop of OCV isanticipated for DHNA-mediated EET because DEINA reduction accumulates electrons on the anode, leading to a more rapid decrease in anodic potential compared to the EET-deflcient control. When the circuit is closed, EET would allow more electron transfer from the anode to the cathode, resulting in higher SCA. To prepare for detection, in the present embodiment, the exponential-phase microbial coculture cells were incubated for 1.5-2 hours to allow DHNA secretion. Then, 60 pL of the cocultures were injected into the acrylic well of the device for OCV and SCA measurements (Fig. 19c). A more rapid drop of OCV and a higher SCA for EET-capable co-cultures such as Lactococcus lactis and Lactiplantibacillus plantarum as well as Escherichia coll and Lactiplantibacillus plantarum co-cultures were observed when compared to those containing EET-deficient cocultures (those containing L. plantarum Andh2) (Fig. 20c, d). We next tested whether the joint utility of e-COSENS and the MFC would allow bioelectronic sensing. For this embodiment, we characterized an engineered L. Jactis-L. plantarum co-culture to detect anhydrotetracy cline (aTc, an antibiotic derivative) in food (milk) samples. The faster drop of OCV and higher SCA reflected the detection of aTc in the milk sample (Fig. 19d). These results demonstrate that the mini bioelectronic device supports signal detection from e-COSENS using microliter-scale samples.

[0307] Using the present embodiment, we also demonstrated using a digital multimeter to measure electrical signals (Fig. 19e). By connecting the positive terminal of the multimeter to the carbon anode of the device and the negative terminal to the Ag / AgCl cathode, the multimeter displayed lower OCV values for EET-capable co-cultures compared to deficient controls (Fig. 20e) and also reflected aTc sensing (Fig. 19e). Without being bound by theory, these data highlight the use of the integrated microbial sensors and clay-based MFC for portable bioelectronic sensing with a digital multimeter as a simple and accessible method for electrical signal readout.

[0308] Non-limiting, exemplary uses and applications of the present disclosure'.

[0309] In various embodiments and without wishing to be bound by theory, the present disclosure has broad applications in environmental sensing, food safety, medical diagnostics, and other areas.

[0310] In embodiments, the presently disclosed integrated system combines the advantages of microbial sensors (self-repair, self-assemble, sensitive, specific) and miniature bioelectronic devices (low-cost, low-energy, easy-to-use, portable), making this joint system suitable for sensing applications in resource-limited sellings.

[0311] Exemplary applications include but are not limited to:• Accessible & portable field testing: Without being bound by theory, with its simple electrical readout using a digital multimeter, the device can be integrated into commercial testing kits for rapid field detection of various environmental factors, such as pollutants in environmental samples, contaminations in food, and biomarkers in biological fluids.• Biocompatible electronic devices: Without being bound by theory, given that, under certain embodiments, the system is constructed using biocompatible materials (clay proton exchange membrane) and generally recognized as safe (GRAS) microbes, the present disclosure can be further advanced to build wearable, implantable, or digestible bioelectronics for point-of-care and continuous health monitoring.• Platform for fundamental research: Without being bound by theory, in addition to sensing applications, the under certain embodiments, the presently disclosed system can serve as a research tool to study the fundamental microbial extracellular electron transfer phenomenon.

[0312] Non-limiting, exemplary benefits of the present disclosure:

[0313] Microbial bioelectronic sensors hold promise for fast and cost-efficient sensing applications. However, current microbial bioelectronic sensors often rely on glassy electrochemical reactors and high-precision potentiostats or source meters for electrical signal detection. These instruments are cumbersome, expensive, and require specialized expertise to operate, making them impractical for field applications. As a result, microbial bioelectronic sensors have not been widely adopted by other biosensors, such as those that output optical signals.

[0314] In various non-limiting embodiments and without wishing to be bound by theory’, the presently disclosed system overcomes these challenges by integrating microbial bioelectronic sensors wdth a low-cost, miniaturized microbial fuel cell (MFC) and a simplified electrical signal detection method, with the following non-limiting, exemplary advantages:• Miniaturized and portable: Unlike bulky glassy electrochemical reactors, in certain embodiments of the presently disclosure, the MFC-based sensing platform is compact (centimeter-scale) and lightweight, making it highly portable for on-site and field applications.• Low-cost and scalable: In embodiments, the presently disclosed device is manufactured using inexpensive materials that are significantly cheaper thantraditional electrochemical sensing setups and can be manufactured in a high- throughput manner.• No need for specialized equipment: Tn certain embodiments, the presently disclosed device allows for electrical signal detection using a simple, handheld digital multimeter, making the system user-friendly.• Simple operation: In embodiments, the device allows direct microbial sensors + sample injection into the MFC without needing complex electrode preparation and calibration.• Versatile: In certain embodiments, the system can be re-engineered for different sensing applications by genetically modifying the microbial sensors to detect various environmental analytes.

[0315] There is a strong market demand for a low-cost, portable, and easy-to-use bioelectronic sensing platform that allows for real-time monitoring of environmental and health parameters. Current microbial bioelectronic sensors have barriers to commercialization due to their reliance on bulky and expensive electrochemical instrumentation. The present disclosure overcomes these challenges by integrating microbial bioelectronic sensors into miniature microbial fuel ceils and allowing signal detection with a simple digital multimeter. This approach offers a simple, low-cost, and user-friendly solution, making bioelectronic sensing more practical for widespread applications.

[0316] Non-limiting, exemplary aspects of the present disclosure include commercial testing kits for rapid field detection of various environmental factors, such as pollutants in environmental samples, contaminations in food, and biomarkers in biological fluids. Additional non-limiting, exemplary aspects of the present disclosure relate to wearable, implantable, or digestible bioelectronic devices for point-of-care and continuous health monitoring.

[0317] In embodiments, the present disclosure can be useful for biotechnology companies specializing in biosensor development or microbial engineering; food safety company, healthcare and diagnosis company, or environmental monitoring companies seeking faster detection technologies.

[0318] In certain embodiments and without being bound by theory, the presently disclosed microbial fuel cell device is scalable, can be manufactured in a high-throughput manner, and the engineered microbes can self-assemble and self-replicate in large culture media.

[0319] Non -limiting, exemplary’ modifications

[0320] Without being bound by theory, due to its miniaturized design, under certain embodiments, the presently disclosed device generates relatively low electrical signals. Without being bound by theory, this can be addressed by further engineering the sensing microbes to enhance their EET activity. Moreover, in certain embodiments, signal amplification techniques like Organic Electrochemical Transistors (OECTs) can be integrated into the system to enhance the signal fold change and sensitivity.

[0321] In embodiments, the microbial sensors require pre-growth and preparation before use in the MFC, which can introduce delays in testing. Without being bound by theory, to streamline this process, engineered microbial cells can be lyophilized at their exponential growth phase and rehydrated upon usage, allowing faster deployment.

[0322] In certain embodiments, the electrode materials used in the MFC device, such as Ag / AgCl, could degrade when exposed to oxygen over extended periods, potentially affecting performance. Without being bound by theory, to extend shelf life and maintain stability’, the devices can be vacuum-sealed during storage and transport.

[0323] References cited in this example

[0324] Tejedor-Sanz, S., Stevens, E.T., Li, S., Finnegan, P., Nelson, J., Knoesen, A., Light, S.H., Ajo-Franklin, C.M., Marco, M.L. Extracellular electron transfer increases fermentation in lactic acid bacteria via a hybrid metabolism. eLife 1 1, e70684 (2022).

[0325] Tolar, J.G., Li, S., Ajo-Franklin, C.M., 2022. The Differing Roles of Flavins and Quinones in Extracellular Electron Transfer in Lactiplantibacillus plantarum. Applied and Environmental Microbiology 89, e01313-22 (2023).

[0326] Li, S., Tavares, C. D. G., Tolar, J. G. & Ajo-Franklin, C. Selective bioelectronic sensing of pharmacologically relevant quinones using extracellular electron transfer in Lactiplantibacillus plantarum. Biosensors and Bioelectronics 243, 115762 (2024).

[0327] Li, S., Zuo, X., Carpenter, M. D., Verduzco, R., Ajo-Franklin, C. M. Microbial bioelectronic sensors for environmental monitoring. Nature Reviews Bioengineering 3:30-49 (2025).

[0328] Li, S., Zhang, J., Ajo-Franklin, C. M., Igoshin, O. A. The growth benefits and toxicity of quinone synthesis are balanced by a dual regulatory’ mechanism and substrate limitations. bioRxiv 10.1 101 / 2025.02. 15.638467 (2025).

[0329] Saha, K., Deka, J., Gogoi, R. K., Datta, K.K.R., Raidongia, K. Applications of lamellar membranes reconstructed from clay mineral-based nanosheets: a review. ACS Applied Nano Materials 5 (11), 15972-15999 (2022).

[0330] Bora, B. R., Nath, N.. Dey, M., Saha, K., Raidongia, K, Assembly of natural clay minerals as highly robust evaporation-driven power generator. ACS Applied Energy Materials 7, 6507-6514 (2024).

[0331] Atkinson, J.T.. Su, L., Zhang, X., Bennett, G.N., Silberg, J.J., Ajo-Franklin, C.M. Real-time bioelectronic sensing of environmental contaminants. Nature 611 , 548-553 (2022).

[0332] Mimee, M., Nadeau, P., Hayward, A., Carim, S., Flanagan, S., Jerger, L., Collins, J., McDonnell, S., Swartwout, R., Citorik, R.J., Bulovic, V., Langer, R., Traverse, G., Chandrakasan, A.P., Lu, T.K. An ingestible bacterial-electronic system to monitor gastrointestinal health. Science 360, 915-918 (2018).

[0333] Inda-Webb, M. E. et al. Sub- 1.4 cm3 capsule for detecting labile inflammatory biomarkers in situ. Nature 620, 386-392 (2023).EXAMPLE 5

[0334] Synthetic microbial co-cultures for modular bioelectronic sensing in diverse environments

[0335] Abstract

[0336] Human disruption of ecosystems poses a significant threat to global health, driving the need for low-cost, low-power, and easily deployable sensors for environmental and health monitoring. Microbial bioelectronic sensors are well-suited as they generate electrical signals and can be integrated into compact electronic devices for field deployment over extended periods. However, current engineering strategies for bioelectronic sensors lack modularity, are limited to a few microbial chassis, and depend on specialized instruments for signal detection. Here, we present the electroactive cp-culture sensing system (e’COSENS), a plug-and-play platform for bioelectronic sensor development. This system comprises a “sender” bacterium that produces electron mediators in response to analytes and a “receiver” bacterium that utilizes tire electron mediators to generate electrical signals via extracellular electron transfer (EET). By modularly swapping the sender bacterium and its associated genetic sensing elements, we achieved bioelectronic sensing of metals, small molecules, and peptides in distinct environmental, food, and human-relevant settings. Moreover, we designed a centimeter-sized bioelectronic device that allows for low-cost, portable signal readout from e’COSENS using a household digital multimeter. The e’COSENS platform greatly simplifies the bioelectronic sensor design and opens unprecedented potential for bioelectronic sensor applications.

[0337] Introduction61

[0338] The rapidly expanding global human footprint has caused severe pollution of the environmental and food chains1, necessitating continuous monitoring of pollutants and health parameters using low-cost and deployable sensing techniques. Synthetic biology’ has engineered microbes as whole-cell biosensors to detect chemicals in complex environments over extended periods through analyte-inducible genetic circuits and generate optical2, acoustic3, or electrical signals4. Among these, bioelectronic sensors — which generate electrical signals — rely on low-cost and low-power electronic components for signal measurement4and can detect analytes within minutes of contact even in opaque environments3. Bioelectronic sensors have been developed to sense pollutants5-7and pharmacologically relevant molecules89, and can be integrated into miniature electronic devices for use in urban waterways5,10-13, sediments14 13, or on human skin16, showing promise as on-demand tools for sensing in urban settings.

[0339] However, the development of whole-cell biosensors with electrical output lags far behind other biosensor techniques, such as fluorescent biosensors2, due to complex engineering strategies and the shortage of suitable microbial chassis4. Current approaches modulate the on-off activity of extracellular electron transfer (EET) pathways in native electroactive microbes, such as the mesophilic, freshwater-dwelling Shewanella oneidensis6'1or the oxygen-sensitive Geobacter sulfurreducens^ . These approaches restrict the application range of bioelectronic sensors due to the narrow environmental tolerance of native electroactive microbes and limited availability of genetic tools18. Synthetic biology allows heterologous expression of multi-component EET pathways in non-native electroactive hosts, such as the model organism Escherichia coli5’{9or the marine bacterium Marinobacter atlanticus2® . However, these strategies have only succeeded after tedious trial and error yet yield electrical signals far below native systems21-23. Furthermore, detecting electrical signals historically requires specialized instruments, such as potentiostats or source meters, which hiders the broader adoption of bioelectronic sensors. Thus, simpler and more efficient engineering and measurement strategies are needed to fully unlock the potential of bioelectronic sensors.

[0340] Here, we present the electroactive co-culture sensing system (e’COSENS), a plug-and-play platform for creating bioelectronic sensors (Fig. 21). This system comprises a “sender” bacterium engineered to produce an electron mediator (e.g., l,4-dihydroxy-2- naphthoic acid, DHNA) in response to specific analytes, and a “receiver” bacterium capable of using the electron mediator (such as DHNA) to generate electrical signals through a DHNA- mediated EET pathway2425(Fig. 21). This division of functions allows non-electroactive bacteria to be modularly integrated for versatile bioelectronic sensing. By employing thegenerally-recognized-as-safe Lactiplantibacillus plantarum as a universal receiver36, e' COSENS demonstrates robust functionality across diverse environmental and human-relevant settings. We also integrate e'COSENS into a miniature microbial fuel cell device, which requires only a household digital multimeter for readout, highlighting the possibility of tailoring e’COSENS for portable biosensing applications.

[0341] Results

[0342] "Sender” and ‘‘receiver ” candidates are widespread

[0343] In embodiments, the e'COSENS platform is designed based on a quinone- mediated EET pathway possessed by many Gram-positive bacteria24’25. One component of this system is the membrane- bound type-II NADH:quinone oxidoreductase Ndh2, which uses quinones (e.g., DHNA) to shuttle electrons from the cytosolic NADH pool to extracellular electron acceptors, such as iron(III) oxide or electrodes (Fig. 21). Studies have shown that certain quinone auxotrophs, such as L. plantarum, can perform Ndh2-dependent EET with exogenous DHNA8’25-27or when incubated with the spent media of a quinone-producing bacterium28. Certain embodiments leverage this quinone cross-feeding phenomenon to create sender-receiver co-cultures for EET. In one non-limiting exemplary’ embodiment, the sender bacterium can produce DHNA and the receiver bacterium comprises a DHNA auxotroph that is capable of DHNA-mediated EET. By way of non-limiting example, the co-culture can permit programmable EET for activity sensing. As shown in tire exemplary embodiment of Fig. 21, such a co-culture can be generated by controlling the expression of a DHNA biosynthesis enzyme (e.g. MenD). For instance, DHNA biosynthesis can be controlled on plasmids29.29

[0344] We first screened for bacterial candidates qualified to serve as the sender and the receiver in e'COSENS. In the sender bacterium, DHNA is biosynthesized from the substrate chorismate through a conserved seven-enzyme pathway (MenF, D, H, C, E, B, I) and is subsequently prenylated (MenA + prenyl diphosphate) and demethylated (MenG) to form vitamin K230(Fig. 21). To survey the ubiquitousness of the DHNA biosynthesis pathway, we obtained the identifiers for each of the seven enzymes from the annotated reaction or protein family databases and searched for matched protein entries in the UniProtKB protein sequence database. We identified 2238 phylogenetically distinct bacteria with an intact DHNA biosynthesis pathway (Fig. 22 Panel a. Example 6), indicating that the sender candidates are widespread. To survey the potential receiver candidates, we focused on a subset of Firmicutes that were previously predicted to possess a quinone-mediated EET locus24. Among the 383 putative EET-capable species in Firmicutes, 324 of them (85%) were found to lack a partial or the entire DHNA biosynthesis pathway (Fig. 22 Panel b, Example 7). This implies that asubstantial portion of Firmicutes depend on exogenous quinones for EET, which qualifies them as potential receiver candidates.

[0345] We next experimentally validated the sender and receiver candidates. We chose L. plantarum as the receiver due to its DHNA-mediated EET activity25,2 ,28and its generally- recognized-as-safe status26. Specifically, we used a mutant strain with a dmkA knockout (encoding DHNA octaprenyltransferase) to eliminate DHNA consumption and an ndhl knockout (encoding NADH.quinone oxidoreductase not essential for EET) to improve electron transfer flux through Ndh2s. Hereafter in the present example only, L. plantarum AdmkAAndhl is simply referred to as “Z. plantarum ”. To identify potential senders, we inoculated L. plantarum into cell-free supernatant (CFS) collected from 15 putative DHNA-producing bacterial strains across 11 phylogenetically distinct species, as well as a DHNA-null control bacterium (Caulobacter crescentus). We also tested glucose and mannitol as carbon sources, as variations in carbon sources can influence quinone synthesis levels31. We then quantified the Ndh2-dependent EET by subtracting the iron(III) oxide reduction level of the L. plantarum \ndh2 from that of the ndh2 wild-type. No EET activity was observed with the CFS from the negative control C. crescentus EET. In comparison, EET was activated by the CFS from 12 out of 15 putative DHNA-producing strains when using glucose and by 1 1 out of 15 strains when using mannitol as the carbon source (Fig. 22 Panel c). While mannitol supported less bacterial growth compared to glucose (Fig. 27), the CFS derived from mannitol-based media activated stronger Ndh2-dependent EET (Fig. 22 Panel c). These results confirm that diverse DHNA-producing bacterial strains can activate EET in the receiver L. plantarum, and different carbon sources can modulate the level of EET activity.

[0346] Synthetic co-cultures exhibit electroactivity in a DHNA-mediated Ndh2- d ependent manner

[0347] Among the bacterial strains whose CFS activated EET in L. plantarum, we selected Lactococcus lactis subsp. lactis KF 147 and E. coli BL21 as senders to construct cocultures with L. plantarum (Fig. 23 Panel a, d). / .. lactis is a Gram-positive, generally- recognized-as-safe bacterium with promise in food and medical applications32, and E. coli is the Gram-negative model bacterium for synthetic biology. To demonstrate that existing biosensors can be modularly integrated into e'COSENS for bioelectronic sensing, we leveraged an E. coli BL21 “Marionette” strain that carries a highly optimized genome-integrated sensor array that can respond to diverse small molecules3j. To enhance DHNA biosynthesis, we eliminated DHNA-consumption pathways by knocking out men A (encoding DHNA octaprenyltransferase) in both L. lactis and E. coli. Additionally, we knocked out noxAB(encoding type-II NADHiquinone oxidoreductases) in L. lactis to ensure DHNA only interacts with Ndh2 in L. plantarum. Hereafter in this Example 5 only, “A. lactis'’'' and “E. coir imply that these strains contain the respective genomic mutations (Table 7).

[0348] To establish functional co-cultures, we optimized media composition and inoculation ratios. The ideal co-cuhurc conditions should sustain the viability of both sender and receiver strains, support optimal EET activity, and minimize the electrochemical background. We developed mannitol-based minimal media for E. coli-L. plant arum and L. lactis-L. plantarum co-cultures, respectively (FIG. 36); these media support growths of both strains, with mannitol as the carbon source to enhance quinone synthesis and EET. We next determined that a sender-to-receiver ratio of 1:2 to 1 : 10 allowed for maximum fold change in EET using the iron reduction assay (Fig. 28 Panel a). However, E. coli was found incompatible with the iron reduction assay due to a high Fe2+background, likely resulting from intracellular iron acquisition during E. coli growth34. As a result, we initially optimized inoculation ratios using L. lactis-L. plantarum and found these ratios applicable to E. coli-L. plantarum.

[0349] We then tested if co-cultures can reduce iron(III) oxide and produce electrical current via a DHNA-mediated Ndh2-dependent mechanism. When inoculated with iron(III) oxide, monocultures of L. lactis and L. plantarum and co-cultures deficient in DHNA biosynthesis (L. lactis EmenD) or EET (L. plantarum EndhE) showed no iron reduction activity. Only the functional L. lactis-L. plantarum co-culture reduced a significant amount of iron(lll) oxide (Fig. 23 Panel b). Similarly, when A. plantarum was inoculated into CFS from E. coli, the iron reduction level lowered if E. coli was incapable of DHNA biosynthesis (EmenD) or L. plantarum was deficient in EET {Endh2') (Fig. 23 Panel e). These results indicate that both DHNA biosynthesis in L. lactis or E. coli and EET in L. plantarum are essential for extracellular reduction. To test if the co-cultures produce current, we inoculated both L. lactis-L. plantarum and E. coli-L. plantarum co-cultures into bioelectrochemical reactors and performed chronoamperometry to monitor current35. The co-cultures can transfer electrons to the electrode, resulting in current generation under anaerobic (Fig. 23 Panel c, f) or microaerobic (Fig. 28 Panel b, c) conditions when both DHNA biosynthesis and EET were functional (Fig. 23 Panel c, f). While we observed competition between the sender and receiver strains over a 72-hour co-culture period (Fig. 29), the interspecies interaction did not impair electroactivity. Taken together, these results confirm that sender-receiver co-cultures can exhibit electroactivity, which mechanistically depends on DHNA biosynthesis in L. lactis or E. coli and Ndh2-dependent EET in L. plantarum.

[0350] DHNA-mediated interspecies signaling can be rewired for bioelectronic sensing

[0351] Having established electroactive co-cultures, we sought to rewire DHNA- mediated EET to be analyte-dependent for bioelectronic sensing. We reprogramed the senders to synthesize DHNA only in response to specific environmental stimuli. To achieve this, we knocked out the menD gene, which encodes the key regulatory node in the DHNA biosynthesis pathway29from the sender bacterium, and controlled menD expression on plasmids under analyte-inducible transcriptional factors (Fig. 21). We first tested the nisin- and cumate- inducible systems in L. lactis and E. coll, respectively. Nisin is an antimicrobial peptide produced by L. lactis for food preservation that can be sensed by the genomically encoded NisRK two-component system , leading to menD transcription from the PnisA promoter (Fig. 24 Panel a). Cumate is an inducer molecule that can inhibit CymRAMfrom repressing the PcymRc promoter33, therefore initiating menD transcription (Fig. 24 Panel d).

[0352] We optimized the genetic circuits for maximal sensing fold change and evaluated e’COSENS’s sensing capability in electrochemical reactors. To reduce basal expression, we screened degenerate ribosome binding sites (RBSs) with varying translation initiation rates (TIRs) for menD expression (Fig. 30 Panel a, d). RBSs with the lowest TIRs yielded the largest fold changes in iron reduction in response to nisin (Fig. 30 Panel b, c) or cumate (Fig. 30 Panel e, 1) and were thus selected for subsequent use. We then inoculated cocultures in electrochemical reactors and monitored current generation upon nisin or cumate induction. As controls, co-cultures with the empty-vector senders or EET-deficient L. plantarum had no response to inducers (Fig. 30 Panel g, h). The functional L. lactis-L. plantarum (Fig. 30 Panel b) and E. coli-L. plantarum (Fig. 24 Panel e) co-cultures responded to inducers and generated differential (A) current densities proportional to the inducer concentrations (Fig. 24 Panel c, f). The time required to significantly differentiate (P<0.05) sensing signals from the background was negatively proportional to the inducer concentrations, with the shortest time being 0.63 and 1.01 hours for nisin and cumate sensing, respectively (Fig. 24 Panel c, f). These results indicate that by controlling DHNA-mediated EET, synthetic co-cultures can be rewired for quantitative bioelectronic sensing, and the sensing ability necessitates both DHNA biosynthesis and EET functionalities.

[0353] Bioelectronic sensors have also used voltametric techniques to detect redoxactive signaling molecules like DHNA:"4'’: this led us to compare the signal intensity from direct voltametric measurement of DHNA with that from DHNA-mediated EET. To do so, we compared cyclic voltammetry (CV) profiles of co-cultures comprising EET-capable (ndh2WT) or deficient (\ndh2) L. plantarum after nisin or cumate induction. When EET was deficient, CV was able to detect a DHNA catalytic wave around a midpoint of -25 mV (vs. Ag / AgCl) for E. coli but not for L. lactis (Fig. 24 Panel g, h). However, when EET was functional, such a catalytic wave became detectable for L. lactis and was significantly more prominent for coli (Fig. 24 Panel g, h). At 200 mV (the potential for chronoamperometry), the current density was 2.0- or 3.3-fold higher for EET-capable E. coli-L. plantarum or L. lactis-L. plantarum co-cultures, respectively, compared to deficient controls. This increase in current can result from continuous redox-cycling of DHNA between the electrode and L. plantarum's EET pathway, which amplifies the electrical signals. Thus, e’COSENS offers advantages over direct CV detection by providing signal amplification and greater sensitivity.

[0354] As many environmental molecules are redox-active, wre asked whether these background molecules would interfere with DHNA-mediated sensing. We sequentially- exposed co-cultures to 250 nM microbe- or plant-derived redox-active molecules, including heme, pyocyanin, phenazine- 1 -carboxylic acid, riboflavin, flavin adenine dinucleotide, flavin mononucleotide, and phylloquinone. Although some background current was observed and the electroactivity of L. lactis-L. plantarum was slightly impaired, these redox-active molecules did not affect current generation in response to nisin or cumate induction (Fig. 31 Panel a, b). Additionally, we tested the influence of environmental DHNA and observed that 50 nM DHNA induced a high background current, which inhibited nisin sensing in L. lactis-L. plantarum (Fig. 31 Panel c). However, cumate sensing in E. coli-L. plantarum was not affected despite the DHNA background current (Fig. 30 Panel d). Without being bound by theory, in one exemplary embodiment, this repression is due, at least in part, to exogenous DHNA allosterically inhibiting MenD with a half-maximal inhibitory concentration varying from about 50 nM to 4 pM in different bacterial species41 42, which would suppress endogenous DHNA biosynthesis (Fig. 30 Panel e). These results indicate that, while exogenous DHNA influences e’COSENS to varying extents depending on the sender bacterium, e’COSENS is resilient against many common redox-active molecules.

[0355] Having shown that e-COSENS can detect common inducers, we sought to probe its modularity to detect other analytes, including anhydrotetracycline (aTc, a derivative of the antibiotic tetracycline), arsenite (NaAsO2, a heavy metal contamination in water), and hydrogen peroxide (H2O2, an inflammation biomarker). We used established transcriptional systems: tetR-Pxy i / 2xteio for aTc sensing in L. lactis45, arsR-ParsOC2 for arsenite sensing in E. coli44, and oxyR-Poxys for H2O2sensing in E. coli45(Fig. 32 Panel a, c, e). With optimized RBSs, we detected significant iron reduction at 1.3-10 pM aTc, 0.01-10 pM arsenite (the WorldHealth Organization limit of 10 ppm or ~0.1 pM in drinking water), and 1-100 pM H2O2(Fig. 32 Panel a, c, e). When tested in bioelectrochemical reactors, both L. lactis-L. plantarum and E. coli-L. plantarum produced current in response to the respective analytes (Fig. 24 Panel i, Fig. 32 Panel b, d, f). However, a relatively high background was observed for arsenite and H2O2sensing circuits, demonstrating that further optimization is needed to reduce basal expression. Nonetheless, these results highlight e’COSENS's unprecedented modularity for bioelectronic sensing, allowing flexible switching of the sender bacterium and associated genetic circuits to detect diverse analytes.

[0356] e~COSENS functions in diverse environments and within microbial communities

[0357] We next explored whether e’COSENS can be applied in real-world environments. As L. lactis and L. plant arum are GRAS species used for food and health applications32,46, we tested if their co-culture can sense aTc in milk and produce electrical signals, showing potential for food quality monitoring. Engineered E. coll modalities have shown promise for applications in various environments and mammalian hosts6,47-49; we examined whether its co-culture with L. plantarum can be applied for pollutant and biomarker monitoring by testing arsenite and H2O2bioelectronic sensing in bayou water and artificial saliva. Milk was obtained from a local grocery store, and bayou water was collected from the Brays Bayou in the Houston area. Artificial saliva mimics natural saliva's chemical and physical characteristics’0and was used as a substitute for saliva to allow for testing in larger quantities. These samples have distinct physicochemical properties, varying in total organic carbon (TOC), pH, electrical resistance, and light transmittance (Fig. 25 Panel a). Notably, milk exhibited 0% transmittance for light at 509 nm (1 cm path length), which is the emission wavelength of green fluorescent protein (GFP), making it an opaque environment where GFP- based biosensors would be ineffective. Moreover, the samples represent a carbon source level varying over a factor of 104, with TOC of 5 x 104mg / L for milk to 5 mg / niL for bayou water.

[0358] To evaluate e'COSEN S ’ s functionality in these environments, we inoculated the exponential-phase L. lactis-L. plantarum or E. coli-L. plantarum co-cultures into bioelectrochemical reactors containing 5 mL samples spiked with respective analytes (Fig. 25 Panel b). The L. lactis-L. plantarum co-culture detected aTc (5 ng / mL) in milk with 95% and 99% confidence within 21 and 25.8 minutes (Fig. 25 Panel c). Despite a higher background, the arsenite (2.5 pM) was detected by E. coli-L. plantarum co-culture with 95% and 99% confidence in 273 and 347 minutes (Fig. 25 Panel c). Similarly, the H2O2(25 pM) can also be sensed by E. coli-L. plantarum co-culture with 95% and 99% confidence in 207 and 253 minutes (Fig. 25 Panel c). The detection time varied across different environments and co-cultures, potentially affected by the basal current level, the nutrient availability, and the cell viability (Fig. 33 Panel a). The fastest response was achieved in the nutrient-rich milk sample (Fig. 25 Panel c) with the low basal expression aTc-sensing genetic circuit (Fig. 32 Panel a, b). Moreover, no differential current was observed if L. lactis or E. colt carried empty vectors or if ndh2 was knocked out from L. plantarum (Fig. 33 Panel b-g), indicating that sensing relied on functional co-cultures. These results demonstrate that 1) electrical signals can be detected in opaque environments (milk) where optical biosensors are inoperable; 2) the universal receiver L. plantarum is applicable in distinct environments, which is not surprising given L. plantarum’s nomadic lifestyle and adaptability to diverse habitats31; 3) by pairing different senders with L. plantarum, e'COSENS can be tailored for bioelectronic sensing in diverse environments.

[0359] Since L. lactis and L. plantarum are present in the human gastrointestinal tract32, we additionally asked whether their co-cuiture can be used for bioelectronic sensing within the human gut microbiota. Using continuous- flow minibioreactor arrays (MBRAs)53, we cultivated microbiota from the large intestine contents of three distinct human donors. We then inoculated each microbiota along with L. lactis-L. plantarum co-culture into bioelectrochemical reactors to test for nisin sensing (Fig. 25 Panel d). The co-culture survived (Fig. 34 Panel a-c) and detected nisin in the three microbiota samples, producing current distinguishable from the EET-deficient and buffer controls (Fig. 25 Panel e, f). Donor I exhibited the lowest background current, while donors 2 and 3 showed a moderately high background (Fig. 25 Panel e). The variations in background levels can be attributed to the different microbiota composition across tire three donors (Fig. 34 Panel d). These results show that e'COSENS is effective within complex microbial communities despite potential interference from neighboring microbes.

[0360] A miniature bioelectronic device under non-limiting, exemplary embodiments allows for portable signal detection

[0361] Prior electrical signal detection was achieved using glassy electrochemical reactors and high-precision potentiostat (Methods); these instruments are cumbersome, high- cost, and have very limited portability. To allow low-cost, portable, and energy-friendly bioelectronic sensing under one embodiment, we disclosed herein is a miniature clay-based microbial fuel cell (MFC) device for detecting electrical signals from e'COSENS (Fig. 26 Panel a, Fig. 35 Panel a). In one embodiment, the device is fabricated by sandwiching a proton exchange membrane (PEM) between a screen-printed Ag / AgCl cathode and a carbon anode stacked with a 6 mm carbon felt round (Fig. 26 Panel a). The PEM can be exfoliated from natural vermiculite clay, showing high biocompatibility and extraordinary ion conductivity ofup to 2* 10-2S cm54. Manufactured in Boston and shipped to Houston for testing, the devices maintained good performance, demonstrating chemical and mechanical stability. Moreover, a single device costs less than three dollars and can be manufactured in a high-throughput way (Fig. 35 Panel b).

[0362] We detected electrical signals from e'COSENS using the clay-based MFC device by measuring the open circuit voltage (OCV) and short circuit current (SCA) (Fig. 26 Panel b). The OCV measures the potential difference between anode and cathode; a faster drop of OCV is anticipated for DHNA-mediated EET because DHNA reduction accumulates electrons on the anode, leading to a more rapid decrease in anodic potential compared to tire EET-deficient control. When the circuit is closed, EET would allow more electron transfer from the anode to the cathode, resulting in higher SCA. To prepare for detection, the exponential-phase co-cultures were incubated for 1.5-2 hours to allow DHNA secretion. Then, 60 pL of the co-cultures were injected into the acrylic well of the device for OCV and SCA measurements (Fig. 26 Panel c). We observed a more rapid drop of OCV and a higher SCA for EET-capable L. lactis-L. plantarum and E. coli-L. plamarum co-cultures when compared to those containing L. plantarum \ndh2 (Fig. 35 Panel c, d). Consistently, when tested with aTc sensing in milk withL. lactis-L. plantarum (Fig. 26 Panel c), the faster drop of OCV and higher SCA reflected the detection of aTc (Fig. 26 Panel d). These results demonstrate that the mini bioelectronic device supports signal detection from e'COSENS using microliter-scale samples.

[0363] We also demonstrated using a digital multimeter to measure electrical signals (Fig. 26 Panel e). By connecting the positive terminal of the multimeter to the carbon anode of tire device and the negative terminal to the Ag / AgCl cathode, the multimeter displayed lower OCV values for EET-capable co-cultures compared to deficient controls (Fig. 35 Panel e) and also reflected aTc sensing (Fig. 26 Panel e). This highlights the potential of e'COSENS for portable bioelectronic sensing with a digital multimeter as a simple and accessible method for electrical signal readout.

[0364] Discussion

[0365] In this study, we developed e'COSENS, a modular bacterial co-culture platform for designing bioelectronic sensors. In various embodiments, the presently disclosed system comprises of a “sender” bacterium that produces DHNA in response to target analytes and a “receiver” bacterium that generates electrical signals through DHNA-mediated EET. We showcased the versatility of e'COSENS by co-culturing two phylogenetically distinct senders, L. lactis and E. coll, with the universal receiver L. plantarum. By leveraging genetic sensing circuits functional in L. lactis, E. coli, or a combination thereof we demonstrated the detectionof nisin, aTc, cumate, arsenite, and H2.O2in lab-based media, milk, bayou water, artificial saliva, and human microbiome. Moreover, we designed a miniature MFC device for low-cost, low-energy, and portable signal detection using a commercial digital multimeter.

[0366] The modularity of e'COSENS overcomes several bottlenecks for bioelectronic sensor development, including limited availability of chassis and complex engineering strategies. For example, the presently disclosed system removes the need to tediously engineer electroactive bacteria to create bioelectronic sensors. Instead, a wide range of non-electroactive bacteria (such as those that produce DHNA (Fig. 22 Panel a, c)) can be coupled with a receiver bacterium for standardized electrical signal conversion. Further, it significantly broadens the range of detectable analytes by allowing flexible selection of the sender bacterium and associated genetic sensing elements55. This includes microbial chassis with large genetic toolboxes, such as but not limited to E. coif3'-6, L. laciiE1'1, Vibrio natriegens36, Bacillus subtilis36(Fig. 22 Panel c). Moreover, e'COSENS opens up new environments w-here bioelectronic sensors can be deployed. With the generally-recognized-as-safe L. lactis-L. plantarum co-culture, we demonstrated efficient use of bioelectronic sensors in food products and human-relevant contexts (Fig. 25). Without being bound by theory, the presently disclosed approach extends to other contexts, such as by pairing the multi-habitat-adaptable receiver L. plantarum66with appropriate senders, such as V. natriegens for sensing in marine environments61or B. subtilis for soil and plant applications62(Fig. 22 Panel c). With these advantages, e'COSENS can allow the creation of customized bioelectronic sensors for specific applications.

[0367] In embodiments, certain non-limiting, exemplary modifications can be made to the e'COSENS platform. By w ay of example, to address environmental DHNA’s attenuation on e'COSENS inducibility (Fig. 31 Panel c, d), due to allosteric inhibition on MenD (Fig. 31 Panel e), a DHNA-insensitive MenD mutant can be created to eliminate this inhibitory effect41,42. In one embodiment, to minimize background electrical signals caused by environmental DHNA (Fig. 31 Panel c, d) and other microbes in the community (Fig. 25 Panel e), the co-culture can be encapsulated in a hydrogel matrix5'63, which would enrich the sensing signals around the electrodes, reducing the impact of noise and enhancing the signal-to-noise ratio. Additionally, carbon source availability and type can also impact sensor performance (Fig. 22 Panel c and Fig. 27). When applying e'COSENS in the field under certain embodiments, carbon source can be provided during sample incubation (Fig. 26 Panel b) or incorporated along with an encapsulation matrix. To stabilize DHNA biosynthesis and sensor output across different carbon sources, the putative carbon source-sensitive promoter of themen operon64can be genomically replaced with a constitutive promoter to allow consistent transcription of the men operon.

[0368] In various embodiments, the e'COSENS platform permits the combination of biological and electronic systems for sensing technologies. As disclosed in various embodiments herein, e'COSENS’s can generate electrical energy during sensing, which can allow for self-powered bioelectronic sensors ideal for long-term and remote monitoring0’. With endeavors in electronic device miniaturization (Fig. 26), in certain embodiments, e'COSENS can be incorporated into various compact electronic modalities, such as paper-based electrochemical systems10 11 16or ingestible electronic capsules48-49. Without being bound by theory, these bacterial-electronic hybrid systems will offer valuable tools for effective monitoring of environment, food, and health parameters.

[0369] Methods

[0370] Strains

[0371] A list of strains is provided in Table 7. Escherichia coli DI 15a (NEB) was used for all molecular cloning in this Example. The receiver strain Lactiplantibacillus plantation NCIMB8826 EdmltAAndhl , and the EET null L. plantarum EdmkAtsndhlEndh2 have been previously reported8.

[0372] Table 7 | Exemplary List of strains in this Example

[0373] To develop the sender Lactococcus lactis, the parental strain KF 147 was genome-modified to knock out menA. noxAB, and menD using double-crossover homologous recombination. Upstream and downstream homologous arms (about 600 base pairs each) flanking the targeted gene were inserted into the Notl-EcoRI digested suicide plasmid pRV30066. The resulting plasmids were electroporated into L. lactis, and genome-integrated mutants were selected on 0.5% glucose Ml 7 (gM17) agar plates containing 5 pg / mL erythromycin at 30 °C and subsequently verified by colony PCR. A single genome-integrated colony was inoculated in 3 mL gM17 without antibiotics at 30 °C and passaged daily at a 1 : 1000 dilution. Starting from day 6, cells were spread on gM17 plates without antibiotics, and the following day, replica-plated on gM17 agar plates containing 5 pg''mL erythromycin. Erythromycin-sensitive colonies were picked, and gene deletion was confirmed by colony PCR and DNA sequencing.

[0374] To develop the sender E. coli, the BL21 “Marionette” strain with a “sensor array” integrated into the genome33was used as the parental strain and was genome-modified to knock out menA and menD using CR1SPR-Cas9 assisted genome editing6'. The donor DNA containing upstream and downstream homologous arms (about 500 base pairs each) flanking the targeted gene was synthesized as a linear double-stranded DNA fragment (Twist Bioscience). Single-guide RNA (sgRNA) specific to the targeted gene was introduced into the pTargetF backbone (Addgene, #62226). The E. coli strain harboring pEcCas (Addgene, #73227) was grown overnight and then subcultured in 25 mL lysogeny broth (LB) containing 50 pg / mL kanamycin with an initial ODeoo = 0.05. When ODeoo reached 0.1 , 40 mM arabinose was added to induce X-Red recombinase expression. Cells were harvested at an ODeoo of 0.4- 0.6 by centrifugation and made electrocompetent. Competent cells (50 pL) were electroporatedwith 400 ng linear donor DNA and 100 ng pTargetF (containing gene-specific sgRNA). Colonies were selected on LB agar plates containing 50 pg / mL kanamycin and 50 pg / mL spectinomycin at 37 °C. Positive mutants were then verified by colony PCR and DNA sequencing. pTargetF and pEcCas were cured by growing cells in the presence of 10 mM rhamnose or 5 g / L glucose and 10 g / L sucrose, respectively, according to the previous study67.

[0375] Culture conditions

[0376] All molecular cloning experiments were performed in LB or terrific broth. To prepare L. lactis-L. plantarum co-cultures, L. plantarum was initially grown in 3 mL commercial MRS (deMan, Rogosa, and Sharpe) broth at 37 °C without shaking for 15 h, and an aliquot was subcultured (1: 100 v / v) into a desired volume of 1% mannitol MRS broth (mMRS) (Table 12) at 37 °C without shaking for 15 h. L. lactis was initially grown in 3 mL 0.5% glucose M17 (gM17) at 30 °C without shaking for 15 h, and an aliquot was subcultured (1: 100 v / v) in a desired volume of 1 % mannitol M 17 (mM 17) at 30 °C without shaking for 15 h. For iron reduction assay, evaluation of co-culture electroactivity, and bioelectronic sensing in lab-based media, the stationary-phase cultures of L. lactis and L. plantarum were washed twice in IX phosphate-buffered saline (PBS) and resuspended in a blended medium containing 1% mannitol chemically defined medium (mCDM) (Table 10) and mM17 in a 20: 1 volume ratio. This medium supportedlactis growth for sensing while maintaining a low electrochemical background. Co-cultures were created by mixing the two resuspended strains at the indicated ODeoo. For bioelectronic sensing in milk and gut microbiota samples, stationary-phase L. lactis and L. plantarum were subcultured in mM 17 or mMRS, respectively, at an initial ODeoo of 0.1 until reaching the exponential phase (OD6oo = 0.4-0.6). Cells wrere then washed once with PBS and mixed in mM17 at the indicated OD&oo for injection into samples. When L. lactis or L. plantarum carried erythromycin resistance plasmids, all media and samples above (except gut microbiota samples) were supplemented with 10 pg / mL erythromycin.

[0377] To prepare E. coli-L. plantarum co-culture, L. plantarum was grown following the same procedure described herein. E. coli was initially grown in 3 mL LB at 37 °C with 200 rpm shaking for 15 h, and an aliquot was subcultured (1 : 100 v / v) m a desired volume of M9 minimal medium containing 1% mannitol (minimal mM9) (Table 11) at 37 °C with 200 rpm shaking for 15 h. To evaluate co-culture electroactivity and bioelectronic sensing in lab-based media, stationary-phase cultures of E. coli and L. plantarum were washed twice in 1 X PBS and resuspended in minimal mM9. Co-culture was created by mixing the two resuspended strainsat the indicated ODeoo. For bioelectronic sensing in bayou water and artificial saliva samples, stationary-phase E. coli and L. plantarum were subcultured in minimal mM9 or mMRS at an initial ODeoo of 0.05 or 0.1, respectively, until reaching the exponential phase (ODeoo = 0.4- 0.6). Cells were then washed once with PBS and mixed at the indicated ODcoo in minimal mM9 (2.5% mannitol) for injection into samples. When E. coli or L. plantarum carried chloramphenicol resistance plasmids, 34 pg / mL or 10 pg / mL chloramphenicol was added to the monoculture medium of E. coli or L. plantarum, respectively. Due to the redox activity of chloramphenicol, a low concentration of 3 ug / mL was used in the co-culture medium and samples for electrochemical analysis.

[0378] Plasmid construction

[0379] A non-limiting, exemplary list of plasmids used in this Example is provided inTable 8.

[0380] Table 8 | Exemplary List of plasmids used in this Example

[0381] The sequence of exemplary genetic parts used in this Example is provided inTable 9.

[0383] Plasmids for L. lactis were constructed using a backbone derived from the shuttle vector pECGMC3 (Addgene #75441). For nisin sensing, the NisKR two-component system is natively expressed in L. lactis KF14760. Thepromoter and ribosome binding site (RBS) were synthesized as a DNA fragment (Twist Bioscience) based on the sequence from pNZ804868, and cloned upstream of the menD gene amplified from the genome of L. lactis KF 147 using Golden Gate assembly69. To tune RBS strength, a 30 bp sequence upstream of the menD start codon was designed to contain four degenerate nucleotides in RBS (NNGGNGN), and an RBS library with varying translation initiation rate (TIR) was generated using the RBS library Calculator70-71. Four RBS variants with TIR ranging from 10- to 104were selected for experimental testing. For aTc sensing, the letR-?tetR-Di-?xyn2\t&o cassette43was synthesized as a DNA fragment (Twist Bioscience). The same RBS library used for P»&.4 was placed after P^,,- 2X,t,o to optimize menD expression.

[0384] Plasmids for E. coll were constructed using a medium-copy backbone containing a chloramphenicol-resistant gene and p! 5A origin. For cumate sensing, the PcymRC promoter33. SarJ insulator33, a synthetic RBS designed by RBS Calculator70-71, and the menD gene amplified from the genome of E. coli BL21 were integrated into the backbone using Golden Gate assembly. A degenerate RBS (NNGGNGN) library was designed, and four RBS variants (TIR ranging from 10!to 103) were selected to optimize menD expression. For arsenite and hydrogen peroxide sensing, the P.H99-RBS3o-a.rs'R~DT54-P;;,.soc-?44and P->ron-RBSoxvx-oxy / ?- DT54-P<«i / 5cassettes were synthesized and assembled with SarJ and a weak RBS from the degenerate RBS library to drive menD expression. DT5456is a double terminator that prevents cross-transcription. To optimize arsenite sensing, three synthetic RBSs (TIR = 102-10- ) were tested to replace RBSso and rune arsR expression.

[0385] Plasmids for L. plantarum were constructed using a backbone derived from pSIP40372. An ampicillin-resistant gene ampR was inserted into the backbone for molecular cloning. The erythromycin-resistant gene ermR was replaced by chloramphenicol-resistant gene cat for co-culture with E. coli.

[0386] Bioinformatic and phylogenetic analysis

[0387] Identifiers for the seven enzymes in the DHNA biosynthesis pathway were obtained from the annotated reaction database Rhea73, as w ell as the protein family databases InterPro74and NCBIfam75. A list of identifiers for each enzyme is listed in Table 14. These identifiers were used to search for matching protein entries in the protein sequence database UniProtKB (assessed on September 18, 2023). The NCBIfam identifier (TIGR00369) for menl (l ,4-dihydroxy-2-naphthoyl-CoA hydrolase) also includes other hotdog fold thioesterases, andprotein entries were filtered to retain those specifically annotated as 1 ,4-dihydroxy-2- naphthoyl-CoA hydrolase / thioesterase. The resulting protein entries were sorted by organism name using an in-house MATLAB program. The taxa of organisms with no missed genes in the DHNA biosynthesis pathway were obtained from the Genome Taxonomy Database76(assessed on November 8, 2023), and the phylogenetic trees were visualized using iTOL (https: / / itol.embi.de / ). To identify potential EET-capable Firmicutes with incomplete DHNA biosynthesis pathways, previously identified Firmicutes harboring the EET locus24were compared with organisms with no missed DHNA biosynthesis genes.

[0388] Iron reduction assays for cell-free supernatants or co-culture

[0389] L. plantarum and L. lactis were resuspended in the mCDMimM 17=20: 1 medium to ODeoo of 2 or 0.2, respectively. In a 96-deep well plate, a 100 tiL portion of each cell resuspension was mixed along with 200 pL of reagent mixture containing 4 mM iron(III) oxide nanoparticles (Sigma) and 4 mM ferrozine (Sigma) prepared in mCDM-mM17. The plate was covered with aluminum foil and incubated in an anaerobic chamber (Whitley A45) at 30 °C with 150 rpm shaking. After 2 hours, 100 pL of supernatant was collected to measure the absorbance at 562 nm (Tecan Spark plate reader). The Fe2 :concentration w as calculated using a standard curve prepared with ferrous sulfate. To test sensing plasmids, exponential-phase L. lactis -was induced with indicated concentrations of inducers in a 96-deep w'ell plate and incubated for 15 h. Due to the difficulties in normalizing ODeoo on the 96-deep well plate, cells were resuspended in 450 pL of mCDM-mM17, and a 13.3 pL portion was diluted to 100 pL to obtain an approximate ODeoo of 0.2. Final concentrations of Fe2+were normalized to the actual ODeoo of L. lactis in each well.

[0390] E. coli induced a high background in the iron reduction assay, and thus cell-free supernatant (CFS) was tested w’ith L. plantarum. L. plantarum was resuspended in minimal mM9 to ODeoo of 4. In a 96-deep w'ell plate, 160 pL of E. coli CFS w'as mixed w'ith 32 pL of 20 mM iron(III) oxide nanoparticles, 32 pL of 20 mM ferrozine, and 224 pL of L. plantarum resuspension. The plate wras covered with aluminum foil and incubated under the same conditions as described herein. Sensing plasmids were tested by inducing exponential-phase E. coli with indicated concentrations of inducers in a 96-deep wrell plate overnight for 15 h. The final Fe2+concentration was normalized to ODeoo of overnight-grown cells in each well.

[0391] To compare the CFSs from different species, cells were grown in their outgrowth medium (Table 7) before subculturing in mCDM or 1% glucose CDM (gCDM) for 15 h. L. plantarum was subcultured in mMRS or commercial (glucose) MRS for 15 h beforebeing washed and resuspended in mCDM or gCDM to ODsoo of 4 to test with mCDM- or gCDM-derived CFS, respectively. CFS iron reduction assay followed the same procedure as described for E. coll.

[0392] Bioelectrochemical analysis of co-cultures in lab-based media

[0393] Co-cultures electroactivity and inducibility were tested in water-jacketed, two- chamber bioelectrochemical reactors (Adams & Chittenden Scientific Glass). The anodic and cathodic chambers were separated by a cation exchange membrane (CMI-7000). The anodic chamber comprised an Ag / AgCl reference electrode (CH instrument) refilled with 3M KC1 saturated with silver chloride (Sigma), a 6.35-mm-tliick graphite felt working electrode with a 3.2 cm radius (Alfa Aesar) threaded through a 0.5-mm radius titanium wire (Alfa Aesar), and 1 10 mL of medium (mCDM :mM 17=20: 1 for L. lactis-L. plantarum co-culture or minimal mM9 for E. eoli-L. plantarum co-culture). The cathodic chamber comprised a 0.5-mm radius titanium wire as the counter electrode and 110 mL of M9 buffer. The reactors were placed on a magnetic stir station (IKA RO 10) with the anodic chamber solution stirred at 200 rpm using a magnetic stir bar. The water jackets were connected to a heating circulator (Lauda ECO E4S), and the temperature was maintained at 30 °C. For experiments under anaerobic conditions, nitrogen (N2) gas was continuously sparged into the anodic chamber.

[0394] Electrochemical measurements were carried out using a VSP-300 potentiostat (BioLogic). For chronoamperometry, the working electrode was biased at +0.2 V vs. Ag / AgCl, and the current was recorded every 36 s. Once the current stabilized, 2 mL sender cells (L. lactis or E. coli) and 2 mL receiver cells (L. plantation) were injected into the anodic chamber at a final ODeoo of 0.08 or 0.2, respectively. A 1 1 O u. L 1000-times concentrated inducer stock or blank buffer was injected 4 h post-cell injection for / .. lactis or 6 h for E. coli. The interval time was to allow cell growth and background current stabilization. For cyclic voltammetry, the working electrode voltage was swept within the indicated range at a scan rate of 2 mV / s.

[0395] Real-world and human-relevant sample collection and characterization

[0396] The bayou water sample was collected from the Houston area, and large solids were allowed to precipitate overnight at 4 °C before taking the supernatant for measurement. The whole milk was purchased from a local grocery store and stored at 4 °C before use. The artificial saliva was purchased, from Sigma (catalog #SAE0149) and stored at 4CC before use. All samples were not filtered for electrochemical analysis of co-cultures. The pH of each sample was measured with a pH meter (Mettler Toledo). The absorbance (A) at 509 nrn wasdetermined using a UV-Vis spectrophotometer (Agilent) with a 1 cm path-length quartz cuvette. The transmittance (T) was calculated using T = 10A(2-A). The absorbance for the milk sample was saturated, and the transmittance was near zero. To determine total organic carbon (TOC), all samples were filtered through a 0.22 gm filter (the milk sample was diluted 10* times before filtering). TOC was measured using a TOC-vcsh analyzer (Shimadzu). The sample resistance was determined by the current interrupt method using a three-electrode electrochemical system. Briefly, a 10 nA current was applied to the sample, and the uncompensated resistance (Ru) was calculated based on the ratio of the measured voltage and current (Rv.= AE / AI).

[0397] Human gut microbiota collection, cultivation, and metagenomic analysis

[0398] Human large intestines were obtained through the LifeGift organ donation program at the Texas Medical Center (Houston, TX, USA). All organ donors were adults without gastrointestinal disease, surgery, or trauma. Donors testing positive for hepatitis B or C, HIV, or COVID were excluded. The whole intestines were transported on ice w ithin 1 hour of removal. Fresh large intestinal contents were resuspended in cold, sterile PBS with 20% glycerol to a final concentration of 100 mg / mL, and aliquots were stored at -80 °C until use. To cultivate large intestinal microbiotas, frozen resuspensions were thawed under anaerobic conditions. A 4 mL resuspension was inoculated into each well (6 wells total) of the minibioreactor arrays53pre-filled with 15 ml of BRM3 media (Table 13). Microbiota cultures were incubated for 16 h before initiating a continuous flow of fresh BRM3 media. After 5 days of continuous flowy microbiota cultures were collected from each reactor. All culturing wras conducted in an anaerobic chamber with an atmosphere of 5% H2, 5% CO2, and 90% N? at 37°C.

[0399] For metagenomic analysis, cultures from the six wells of each donor’s microbiota were pooled into two groups for sequencing. Metagenomic data were analyzed using lab-developed software and ATIMA (Agile Toolkit for Incisive Microbial Analysis) developed by the Baylor College of Medicine. The remaining microbiota samples were combined for the bioelectronic sensing experiment.

[0400] Bioelectrochemical analysis of co-cultures in real-world and human-relevant samples

[0401] Bioelectronic sensing in milk, bayou water, artificial saliva, or gut microbiota was evaluated in 5 mL-volume two-chamber bioelectrochemical reactors (Adams & Chittenden Scientific Glass). The anodic and cathodic chambers were separated by a Nafion-212 perfluorinated membrane (pre-soaked in KC1, Sigma). The working electrode was a 0.7 cm-wide, 1 cm-long, 3.18 mm-thick graphite felt (Alfa Aesar) threaded with a 0.5-mm radius titanium wire (Alfa Aesar). The reference electrode was Ag / AgCl in 3M KC1 saturated with silver chloride. The counter electrode was a 0.5-mm radius titanium wire. The anodic chamber was filled with 5 mL samples (or CDM for subsequent gut microbiota injection). The cathodic chamber was filled with 5 mL M9 buffer. The reactors were placed in an incubator at 30 °C and sparged with N? gas. Current was monitored by chronoamperometry every 36 s, and the sample-derived background current was allowed to stabilize for 1 h. Exponential-phase sender and receiver cells were first mixed to create co-cultures. A 250 uL mixture of L. lactis-L. plantarum co-culture in mM17 or a 500 pL mixture of E. coli-L. plantarum co-culture in minimal mM9 (2.5% mannitol) was injected into the anodic chambers for a final sender ODeoo of 0.15 and receiver ODeoo of 0.3. After 20 min, 25 yL 200-times concentrated inducer stock or blank buffer was injected into the anodic chambers.

[0402] To test sensing in gut microbiota, the gut microbiota samples were washed three times with Nj-sparged IX PBS and resuspended in N2-sparged IX PBS. N2 sparging removed dissolved oxygen from PBS. A 100 uL resuspension was injected into the anodic chamber for a final ODsoo of 0.25. After 1.5 h, a 250 uL mixture of L. lactis-L. plantarum co-culture in mM17 was injected. Nisin or blank buffer was injected 20 min after co-culture injection.

[0403] Screen-printed bioelectronic device manufacture]00404] The single-chamber MFC was constructed using an Ag / AgCl-coated polyethylene terephthalate (PET) sheet as the cathode and a porous carbon felt as the anode. The electrodes wrere separated by a clay proton exchange membrane (PEM). The clay PEM was made by exfoliating the thermally expanded natural vermiculite clay into atomically thin 2D flakes using a 20% aqueous HC1 solution. An aqueous dispersion of these clay nanosheets was vacuum-filtered onto a cellulose nitrate membrane to fabricate a freestanding lamellar membrane. To make the anode, carbon ink (Kayaku C-250J) was screen-printed onto a PET sheet, with a 1 mm hole punched at the center. A 6 mm-diameter, 3.18 mm-thick carbon felt round was then attached to the wet carbon ink, and the ink was allowed to dry. To make the cathode, Ag / AgCl ink (Kayaku AGCL-675) was screen-printed onto a PET sheet and allowed to dry. To assemble the device, a 10 mm-diameter clay membrane was placed on top of the cathode, and an acrylic well (19 mm wide x 19 mm long x 3 mm thick) was affixed on the top of the clay membrane using glue. The carbon felt anode was then fitted into tire acrylic well. The hole at the top of the anode was designed to introduce bacteria into the device.

[0405] Screen-printed bioelectronic device signal measurements

[0406] The screen-printed devices were treated with UV / ozone for 30 min to make carbon felt hydrophilic. Exponential-phase sender and receiver cells were washed and mixed to create co-cultures. A 16.7 pL co-culture mixture was added to a 1.5 mL microcentrifuge tube containing 500 pL of testing media or samples (milk) so that the final sender ODeoo was 1 .5 and receiver ODeoo was 3.0. For milk samples, 2.5 pL of 1 pg / mL aTc or blank ddFFO was added to the respective tubes before inoculating co-cultures. All tubes (three replicates for each group) were incubated at 30 °C. After 1.5-2 hours, a 60 pL portion was taken and slowly injected into the device using an insulin needle (BD). For electrochemical measurement using the potentiostat, the working electrode was connected to the carbon anode of the device, and the reference electrode was joined with the counter electrode and connected to the Ag / AgCl cathode of the device. Open circuit voltage (OCV) was recorded every 3 s. To measure short circuit current (SCA), the voltage (E) was swept from E = OCV to E = 0 vs. OCV at a scan rate of 10 mV / s. The current at E = 0 vs. OCV was taken as the SCA. For OCV measurement using the digital multimeter (Fluke 280), the positive terminal (source) was connected to the carbon anode, and the negative terminal (drain) was connected to the Ag / AgCl cathode. The OCV was measured as direct voltage (DC) in mV mode.

[0407] Colony-forming unit counting

[0408] The sender L. lactis or E. coll were labeled with mCherry fluorescent protein, and the receiver L. plantarum was labeled with super folded green fluorescent protein (sfGFP). Co-culture samples were taken at the indicated hours and w’ere 10-fold serially diluted from 101to 106times. A 10 pL portion from each dilution was drop-plated on a 0.5% glucose M17 plate containing 10 pg / mL erythromycin (for L. lactis-L. plantarum)' or 10 pg / mL chloramphenicol (for E. coli-L. plantarum) and incubated at 30 °C overnight Colonies w ere counted based on fluorescence.

[0409] Statistics

[0410] Statistical analyses were performed using the Origin Pro software (OriginLab). The p values were determined by two-tailed impaired Student’s t-tests, or one-w'ay analysis of variance w ith Tukey’s post hoc test. Three biological replicates or more wrere defaulted. When two biological replicates were tested due to technical restraints, the experiments were repeated at least three times.

[0411] Data Availability

[0412] Genetic constructions are available from Addgene upon publication. Source data are provided upon publication.

[0413]

[0414] Table 10 [ Chemically defined medium (CDM) in this Exemplary EmbodimentpH adjusted to 6.5

[0415] Table II | Exemplary M9 mannitol minimal medium (minimal mM9) used in this Example

[0416] Table 12 | Non-Limiting Mannitol MRS medium (mMRS) used in the presentExample*TIGR00369 describes hotdog fold thioesterases and was filtered to retain those specifically annotated as 1, 4-dihydroxy-2-naphthoyl-CoA hydrolase / thioesterase

[0419] References Used in this Example

[0420] 1. Fuller, R. et al. Pollution and health: a progress update. The LancetPlanetary^ Health 6, e535-e547 (2022).

[0421] 2. Inda, M. E. & Lu, T. K. Microbes as Biosensors. Annual Review ofMicrobiology 74, 337-359 (2020).

[0422] 3. Maresca, D. et al. Biomolecular Ultrasound and Sonogenetics. AnnualReview of Chemical and Biomolecular Engineering 9 , 229-252 (2018).

[0423] 4. Li, S., Zuo, X., Carpenter, M. D., Verduzco, R. & Ajo-Franklin, C. M.Microbial bioelectronic sensors for environmental monitoring. Nat Rev Bioeng 1-20 (2024) doi: 10.1038 / s44222-024-00233-x.

[0424] 5. Atkinson, J. T. et al. Real-time bioelectronic sensing of environmental contaminants. Nature 611, 548-553 (2022).

[0425] 6. Karbelkar, A. A., Reynolds, E. E., Ahimark, R. & Furst, A. L. AMicrobial Electrochemical Technology to Detect and Degrade Organophosphate Pesticides. ACS Cent. Sei. 7, 1718-1727 (2021).

[0426] 7. Webster, D. P. et al. An arsenic-specific biosensor with genetically engineered Shewanella oneidensis in a bioelectrochemical system. Biosensors and Bioelectronics 62, 320-324 (2014).

[0427] 8. Li, S., De Groote Tavares, C., Tolar, J. G. & Ajo-Franklin, C. M.Selective bioelectronic sensing of pharmacologically relevant quinones using extracellular electron transfer in Lactiplantibacillus plantarum. Biosensors and Bioelectronics 243, 115762 (2024).

[0428] 9. Gao, Y., Ryu, J., Liu, L. & Choi, S. A simple, inexpensive, and rapid method to assess antibiotic effectiveness against exoelectrogenic bacteria. Biosensors and Bioelectronics 168, 1 12518 (2020).

[0429] 10. Cho, J. H., Gao, Y., Ryu, J. & Choi, S. Portable, Disposable, Paper-Based Microbial Fuel Cell Sensor Utilizing Freeze-Dried Bacteria for In Situ Water Quality Monitoring. A CS Omega 5, 13940-13947 (2020).

[0430] 11 . Chouler, J., Cruz-Izquierdo, A., Rengaraj, S., Scott, J. L. & Di Lorenzo,M. A screen-printed paper microbial ftiel cell biosensor for detection of toxic compounds in water. Biosensors and Bioelectronics 102, 49-56 (2018).

[0431] 12. Pasternak, G., Greenman, J. & leropoulos, I. Self-powered, autonomousBiological Oxygen Demand biosensor for online water quality monitoring. Sens Actuators B Chem 244, 815-822 (2017).

[0432] 13. Adekunle, A., Rickwood, C. & Tartakovsky, B. Online monitoring of heavy metal-related toxicity using flow-through and floating microbial fuel cell biosensors. Environ Monit Assess 192, 52 (2020).

[0433] 14. Liu, L., Lu, Y ., Zhong, W., Meng. L. & Deng, H. On-line monitoring of repeated copper pollutions using sediment microbial fuel ceil based sensors in the field environment. Science of The Total Environment 748, 141544 (2020).

[0434] 15. Song, N. et al. Development of a sediment microbial fuel cell-based biosensor for simultaneous online monitoring of dissolved oxygen concentrations along vari ous depths in lake water. Science of The Total Environment 673, 272-280 (2019).

[0435] 16. Mohammadifar, M. & Choi, S. A Portable and VisualElectrobiochemical Sensor for Lactate Monitoring in Sweat, in 2018 IEEE 12th International Conference on Nano / Molecular Medicine and Engineering (NANOMED) 73-77 (Waikiki Beach, HI, USA, 2018). doi: 10.1109 / NANOMED.2018.8641665.

[0436] 17. Ueki, T., Nevin, K. P., Woodard, T. L. & Lovley, D. R. Genetic switches and related tools for controlling gene expression and electrical outputs of Geobacter sulfurreducens. Journal of Industrial Microbiology and Biotechnology 43, 1561-1575 (2016).

[0437] 18. Bird, L. J. et al. Engineering Wired Life: Synthetic Biology forElectroactive Bacteria. ACS Synth. Biol. 10, 2808-2823 (2021).

[0438] 19. Jensen, H. M. el al. Engineering of a synthetic electron conduit in living cells. Proc Natl Acad Sci USA 107. 19213-19218 (2010).

[0439] 20. Bird, L. J. el al. Marine Biofilm Engineered to Produce Current inResponse to Small Molecules. ACS Synth. Biol. 12, 1007-1020 (2023).

[0440] 21 . Goldbeck, C. P. et al. Tuning promoter strengths for improved synthesis and function of electron conduits in Escherichia coli. ACS Synth Biol 2, 150-159 (2013).

[0441] 22. Su, L., Fukushima, T. & Ajo-Franklin, C. M. A hybrid cyt c maturation system enhances the bioelectrical performance of engineered Escherichia coli by improving the rate-limiting step. Biosensors and Bioelectronics 165, 112312 (2020).

[0442] 23. Su, L. et al. Modifying Cytochrome c Maturation Can Increase theBioelectronic Performance of Engineered Escherichia coli. ACS Synth. Biol. 9, 115-124 (2020).

[0443] 24. Light, S. H. et al. A flavin-based extracellular electron transfer mechanism in diverse Gram-positive bacteria. Nature 562, 140-144 (2018).

[0444] 25. Tejedor-Sanz, S. et al. Extracellular electron transfer increases fermentation in lactic acid bacteria via a hybrid metabolism. eLife 11, e70684 (2022).

[0445] 26. Seddik, H. A. et al. Lactobacillus plantarum and Its Probiotic and FoodPotentialities. Probiotics Antimicrob Proteins 9, 111-122 (2017).

[0446] 27. Tolar, J. G., Li, S. & Ajo-Franklin, C. M. The Differing Roles of Flavins and Quinones in Extracellular Electron Transfer in Lactiplantibacillus plantarum. Applied and Environmental Microbiology 89, eO 1313-22 (2022).

[0447] 28. Stevens, E. T. et al. Lactiplantibacillus plantarum uses ecologically relevant, exogenous quinones for extracellular electron transfer. mBio 14, e02234-23 (2023).

[0448] 29. Li, S., Zhang, J., Ajo-Franklin, C, M. & Igoshin, O. A. The growth benefits and toxicity of quinone synthesis are balanced by a dual regulatory mechanism and substrate limitations. 2025.02.15.638467 Preprint at https: / / doi.org / 10.1101 / 2025.02.15.638467 (2025) .

[0449] 30. Kang, M.-J., Baek, K.-R., Lee, Y.-R., Kim, G.-H. & Seo, S.-O.Production of Vitamin K by Wild-Type and Engineered Microorganisms. Microorganisms 10, 554 (2022).

[0450] 31 . Liu, Y., van Bennekom, E. O., Zhang, Y., Abee, T. & Smid, E. J. Long- chain vitamin K2 production in Lactococcus lactis is influenced by temperature, carbon source, aeration and mode of energy metabolism. Microbial Cell Factories 18, 129 (2019).

[0451] 32. Song, A. A.-L., In, L. L. A., Lim, S. H. E. & Rahim, R. A. A review onLactococcus lactis: from food to factory. Microbial Cell Factories 16, 55 (2017).

[0452] 33. Meyer, A. J., Segall-Shapiro, T. H., Glassey, E., Zhang, J. & Voigt, C.A. Escherichia coli “Marionette” strains with 12 highly optimized small-molecule sensors. Nat Chem Biol 15, 196-204 (2019).

[0453] 34. Andrews, S. C., Robinson, A. K. & Rodriguez-Quinones, F. Bacterial iron homeostasis. FEMS Microbiology Reviews 27, 215-237 (2003).

[0454] 35. Alba, R. A. C., Li, S., Kundu, B. B„ Ajo-Franklin, C. M. & Cai, R.Characterizing Mediated Extracellular Electron Transfer in Lactic Acid Bacteria with a Three- Electrode, Two-Chamber Bioelectrochemical System. Journal of Visualized Experiments e67204 (2024) doi: 10.3791 / 67204.

[0455] 36. de Ruyter, P. G., Kuipers, O. P. & de Vos, W. M. Controlled gene expression systems for Lactococcus lactis with the food-grade inducer nisin. Appl Environ Microbiol 62, 3662-3667 (1996).

[0456] 37. VanArsdale, E. et al. A Coculture Based Tyrosine-TyrosinaseElectrochemical Gene Circuit for Connecting Cellular Communication with Electronic Networks. ACS Synth. Biol. 9, 11 17-1 128 (2020).

[0457] 38. Tschirhart, T. et al. Electrochemical Measurement of the [LGalaclosidase Reporter from Live Cells: A Comparison to the Miller Assay. ACS Synth. Biol. 5, 28-35 (2016).

[0458] 39. Zhou, T. et al. A copper-specific microbial fuel cell biosensor based on riboflavin biosynthesis of engineered Escherichia coli. Biotechnology and Bioengineering 118, 210-222 (2021).

[0459] 40. Khan, A. et al. A novel biosensor for zinc detection based on microbial fuel cell system. Biosensors and Bioelectronics 147, 11 1763 (2020).

[0460] 41. Bashiri, G. et al. Allosteric regulation of menaquinone (vitamin K2) biosynthesis in the human pathogen Mycobacterium tuberculosis. J Biol Chem 295, 3759-3770 (2020).

[0461] 42. Stanborough, T. et al. Allosteric inhibition of Staphylococcus aureusMenD by 1 ,4-dihydroxy naphthoic acid: a feedback inhibition mechanism of the menaquinone biosynthesis pathway. Philosophical Transactions of the Royal Society B: Biological Sciences 378, 20220035 (2023).

[0462] 43. Markakiou, S., Neves, A. R., Zeidan, A. A. & Gaspar, P. Development of a Tetracycline-Inducible System for Conditional Gene Expression in Lactococcus lactis and Streptococcus thermophilus. Microbiology’ Spectrum 11, e00668-23 (2023).

[0463] 44. Chen, S.-Y., Zhang, Y., Li, R„ Wang, B. & Ye, B.-C. De Novo Design of the ArsR Regulated Pars Promoter Enables a Highly Sensitive Whole-Cell Biosensor for Arsenic Contamination. Anal. Chem. 94, 7210-7218 (2022).

[0464] 45. Rubens, J. R., Selvaggio, G. & Lu, T. K. Synthetic mixed-signal computation in living cells. Nat Common 7, 1 1658 (2016).

[0465] 46. Mays, Z. J. & Nair, N. U. Synthetic biology in probiotic lactic acid bacteria: At the frontier of living therapeutics. Current Opinion in Biotechnology’ 53, 224-231 (2018).

[0466] 47. Kotula, J. W. et al. Programmable bacteria detect and record an environmental signal in the mammalian gut. Proc. Natl. Acad. Sei. U.S.A. Ill, 4838-4843 (2014).

[0467] 48. Mimee, M. et al. An ingestible bacterial-electronic system to monitor gastrointestinal health. Science 360, 915—918 (2018).

[0468] 49. Inda-Webb, M. E, et al. Sub- 1.4 cm3 capsule for detecting labile inflammatory biomarkers in situ. Nature 620, 386-392 (2023).

[0469] 50. J Pytko-Polonczyk, J., Jakubik, A., Przeklasa-Bierowiec, A. &Muszynska, B. Artificial saliva and its use in biological experiments. J Physiol Pharmacol 68, 807-813 (2017).

[0470] 51. Martino, M. E. et al. Nomadic lifestyle of Lactobacillus plantarum revealed by comparative genomics of 54 strains isolated from different habitats. Environmental Microbiology 18, 4974-4989 (2016).

[0471] 52. De Filippis, F., Pasolli, E. & Ercolini, D. The food-gut axis: lactic acid bacteria and their link to food, the gut microbiome and human health. FEMS Microbiology Reviews 44, 454 (2020).

[0472] 53. Auchtung, J. M., Robinson, C. D. & Britton, R. A. Cultivation of stable, reproducible microbial communities from different fecal donors using minibioreactor arrays (MBRAs). Microbiome 3, 42 (2015).

[0473] 54. Bora, B. R., Nath, N.. Dey, M., Saha, K. & Raidongia, K. Assembly ofNatural Clay Minerals as Highly Robust Evaporation-Driven Power Generator. ACS Appl. Energy? Mater. 1, 6507-6514 (2024).

[0474] 55. Del Valle, I. el al. Translating New Synthetic Biology Advances forBiosensing Into the Earth and Environmental Sciences. Front. Microbiol. 11, 618373 (2021).

[0475] 56. Park, Y., Espah Borujeni, A., Gorochowski, T. E., Shin, J. & Voigt, C.A. Precision design of stable genetic circuits carried in highly-insulated E. coli genomic landing pads. Molecular Systems Biology 16, e9584 (2020).

[0476] 57. Kok, J. et al. The Evolution of gene regulation research in Lactococcus lactis. FEMS Microbiology Reviews 41, S220-S243 (2017).

[0477] 58. Tschirhart, T. et al. Synthetic Biology Tools for the Fast-GrowingMarine Bacterium Vibrio natriegens. ACS Synth. Biol. 8, 2069-2079 (2019).

[0478] 59. Liu, Y., Liu, L., Li, J., Du, G. & Chen, J. Synthetic Biology Toolbox and Chassis Development in Bacillus subtilis. Trends in Biotechnology 37, 548-562 (2019).

[0479] 60. Siezen, R. J. et al. Genome-scale diversity and niche adaptation analysis of Lactococcus lactis by comparative genome hybridization using multi-strain arrays. Microb Biotechnol 4, 383-402 (201 1).

[0480] 61 . Hoff, J. et al. Vibrio natriegens : an ultrafast-growing marine bacterium as emerging synthetic biology chassis. Environmental Microbiology 22, 4394-4408 (2020).

[0481] 62. Su, Y., Liu, C., Fang, H. & Zhang, D. Bacillus subtilis: a universal cell factory for industry', agriculture, biomaterials and medicine. Microbial Cell Factories 19, 173 (2020).

[0482] 63. Tang, T.-C. et al. Hydrogel-based biocontainment of bacteria for continuous sensing and computation. Nature Chemical Biology 17, 724-731 (2021 ).

[0483] 64. Qin, X. & Taber, H. W. Transcriptional regulation of the Bacillus subtilis menpl promoter. Journal of Bacteriology 178, 705-713 (1996).

[0484] 65. Grattieri, M. & Minteer, S. D. Self-Powered Biosensors. ACS Sens. 3,44-53 (2018).

[0485] 66. Leloup, L., Ehrlich, S. D., Zagorec, M. & Morel-Deville, F. Singlecrossover integration in the Lactobacillus sake chromosome and insertional inactivation of the ptsl and lacL genes. Appl Environ Microbiol 63, 21 17-2123 (1997).

[0486] 67. Li, Q. et al. A modified pCas / pTargetF system for CRISPR-Cas9- assisted genome editing in Escherichia coli. Acta Biochimica et Biophysica Sinica 53, 620-627 (2021).

[0487] 68. Kuipers, O. P., de Ruyter, P. G. G. A., Kleerebezem, M. & de Vos, W.M. Quorum sensing-controlled gene expression in lactic acid bacteria. Journal of Biotechnology 64, 15-21 (1998).

[0488] 69. Engler, C., Kandzia, R. & Marillonnet, S. A One Pot, One Step,Precision Cloning Method with High Throughput Capability. PLoS ONE 3, e3647 (2008).

[0489] 70. Salis, H. M., Mirsky, E. A. & Voigt, C. A. Automated design of synthetic ribosome binding sites to control protein expression. Nat Biotechnol 27, 946-950 (2009).

[0490] 71. Reis, A. C. & Salis, H. M. An Automated Model Test System forSystematic Development and Improvement of Gene Expression Models. ACS Synth. Biol. 9, 3145-3156 (2020).

[0491] 72. Sorvig, E., Mathiesen, G., Naterstad, K., Eijsink, V. G. H. & Axelsson,L. High-level, inducible gene expression in Lactobacillus sakei and Lactobacillus plantarum using versatile expression vectors. Microbiology' (Reading, Engl.) 151, 2439-2449 (2005).

[0492] 73. Bansal, P. et al. Rhea, the reaction knowledgebase in 2022. NucleicAcids Res 50, D693-D700 (2022).

[0493] 74. Hunter, S. et al. InterPro: the integrative protein signature database.Nucleic Acids Research 37, D211 (2008).

[0494] 75. Li, W. et al. RefSeq: expanding the Prokaryotic Genome AnnotationPipeline reach with protein family model curation. Nucleic Acids Res 49, D1020-D1028 (2021 ).

[0495] 76. Parks, D. H. et al. A standardized bacterial taxonomy based on genome phylogeny substantially revises the tree of life. Nat Biotechnol 36, 996-1004 (2018).

[0496] Additional References used in this Example1 . Li, S., De Groote Tavares, C., Tolar, J. G. & Ajo-Franklin, C. M. Selective bioelectronic sensing of pharmacologically relevant quinones using extracellular electron transfer in Lactiplantibacillus plantarum. Biosensors and Bioelectronics 243, 115762 (2024).2. Meyer, A. J., Segall-Shapiro, T. H., Glassey, E., Zhang, J. & Voigt, C. A. Escherichia coli “Marionette” strains with 12 highly optimized small-molecule sensors. Nat Chem Biol 15, 196-204 (2019).EXAMPLE 6

[0497] Non-Limiting. Exemplary Sender Strains.EXAMPLE 7

[0498] Non-Limiting, Exemplary Receiver Strains.EQUIVALENTS

[0499] Those skilled in the art will recognize, or be able to ascertain, using no more than routine experimentation, numerous equivalents to the specific substances and procedures described herein. Such equivalents are considered to be within the scope of this invention, and are covered by the following claims.

Claims

CLAIMSWhat is Claimed is:

1. An electroactive co-culture sensing system (e'COSENS) comprising a sender strain, a receiver strain, and a bioelectrochemical measurement apparatus, wherein the sender strain is a quinone-producing bacterium, and wherein the receiver strain is a quinonedependent extracellular electron transfer (EET)-capable bacterium.

2. The e'COSENS of claim 1, wherein the bioelectrochemical measurement apparatus comprises a bioelectrochemical cell and an electronic device for voltage and current monitoring.

3. The e'COSENS of claim 1 , wherein the sender strain is engineered to delete menD and is transformed with the menD gene on a plasmid with a stimulus-responsive genetic element.

4. The e'COSENS of claim 3, wherein the sender strain is engineered further delete menA, ndh, or a combination thereof.

5. The e'COSENS of claim 3, wherein the stimulus-responsive genetic element comprises an inducible transcription factor and an inducible promoter.

6. The e'COSENS of claim 5, wherein the inducible transcription factor comprises tetR, arsR, oxyR. nisR and nisK, araCAMand araE, cymRAM, or a combination thereof.

7. The e'COSENS of claim 5, wherein the promoter is selected from the group consisting of l?xyi'2xtetO, P apsOC2^ PavyS, Pfi / sJ, PBAD, PCTOTAC, or a combination thereof.

8. The e'COSENS of claim 1, wherein the receiver strain is genetically engineered to delete menA, ndhl, or a combination thereof.

9. The e'COSENS of claim 1, wherein the receiver strain is a wild-type strain.

10. The e'COSENS of claim 1 , wherein the quinone-producing bacterium produces 1,4- dihyroxy-2-naphthoic acid (DHNA).

11. The e'COSENS of claim 1, wherein the sender strain is selected from the group consisting of Escherichia coli, Bacillus subtlis, Vibrio natriegens, Shewanella oneidensis, Leuconostoc citreum, Lactococcus lactis. Enterococcus faecalis, a strain from Example 6, or a combination thereof.

12. The e'COSENS of claim 1 , wherein the receiver strain is Lactiplantibacillus plantarum, a strain from Example 7, or a combination thereof.

13. A method of producing a co-culture sensing system (e'COSENS), the method comprising: obtaining a quinone-producing bacterium, and a quinone-dependent extracellular electron transfer (EET)-capable bacterium receiver strain; genetically modifying quinone-producing bacterium, thereby producing a sender strain; co-culturing the sender strain and receiver strain, thereby producing a co-culture; and introducing the co-culture into a bioelectrochemical measurement apparatus, thereby producing a co-culture sensing system (e'COSENS).

14. The method of claim 13, wherein genetically modifying a quinone-producing bacterium comprises deleting menD and transforming with a plasmid containing the menD gene with a stimulus-responsive genetic element.

15. The method of claim 14, wherein genetically modifying a quinone-producing bacterium further comprises deleting men A, ndh, or a combination thereof.

16. The method of claim 13, wherein the stimulus-responsive genetic element comprises an inducible transcription factor and an inducible promoter.

17. The method of claim 16, wherein the inducible transcription factor comprises ieiR, arsR, oxyR, nisR and nisK, araC**1and araE, cymRM1, or a combination thereof.

18. The method of claim 16, wherein the promoter comprises PxyiQxteto, Varsoct, P^-s, P»«,4, PBAD, Pomsc, or a combination thereof.

19. The method of claim 13, wherein the method further comprises genetically modifying the quinone-dependent extracellular electron transfer (EET)-capable bacterium.

20. The method of claim 19, wherein genetically modifying the quinone-dependent EET- capable bacterium comprises deleting menA, ndhl, or a combination thereof.

21. The method of claim 13, wherein the quinone-producing bacterium produces 1 ,4- dihyroxy-2 -naphthoic acid (DHNA).

22. The method of claim 13, wherein the bioelectrochemical measurement apparatus comprises a bioelectrochemical cell and an electronic device for voltage and current monitoring.

23. The method of 13, wherein the sender strain is selected from the group consisting of Escherichia coli, Bacillus subtilis, Vibrio natriegens, Shewanella oneidensis, Leuconostoc citreum. Lactococcus lactis, Enterococcus faecalis, a strain from Example 6, or a combination thereof.

24. The method of claim 13, wherein the receiver strain is selected from the group consisting of Lactiplantibacillus plantarum, a strain from Example 7, or a combination thereof.

25. A bioelectronic sensing system capable of translating environmental stimuli into electrical signals produced by the method of claim 13.

26. The system of claim 1, wherein the e'COSENS comprises a single chamber, miniature microbial fuel cell.

27. The method of claim 13 , wherein the e'COSENS comprises a single chamber, miniature microbial fuel cell.

28. The bioelectric sensing system of claim 25, wherein the e'COSENS comprises a single chamber, miniature microbial fuel cell.

29. A bioelectrocheniical measurement apparatus comprising a miniature, single-chamber microbial fuel cell (MFC), wherein the miniature single-chamber MFC comprises a clay-based proton exchange membrane (PEM) with two opposing faces, and the claybased PEM is disposed between an anode on one face and a cathode on the opposite face.

30. The bioelectrochemical measurement apparatus of claim 29. wherein the clay-based PEM comprises vermiculite clay.

31. The bioelectrochemical measurement apparatus of claim 29, wherein the clay-based PEM comprises an atomically thin 2D flake.

32. The bioelectrochemical measurement apparatus of claim 29, wherein the anode comprises carbon.

33. The bioelectrochemical measurement apparatus of claim 32, wherein the anode comprises a porous carbon felt material in contact with carbon ink that is disposed upon a polyethylene terephthalate sheet; the polyethylene terephthalate sheet comprising a hole, gap, or passthrough that forms a well, and the porous carbon felt is disposed within the well.

34. The bioelectrochemical measurement apparatus of claim 33, wherein the carbon felt is hydrophilic.

35. The bioelectrochemical measurement apparatus of any one of claims 29-34, wherein the cathode comprises Ag / AgCl.

36. The bioelectrochemical measurement apparatus of any one of claims 29-35 further comprising a digital multimeter for voltage and current monitoring.

37. The bioelectrochemical measurement apparatus of any one of claims 29-36, wherein the miniature MFC comprises a width that is less than about 3.0 mm and a thickness that is less than about 0.5 cm.

38. The e'COSENS of any one or more of claims 1-12, comprising the bioelectrochemical measurement apparatus of any one of claims 29-37.

39. The method of any one or more of claims 13-24, comprising the bioelectrochemical measurement apparatus of any one of claims 29-37.

Images