Article Electrochemistry, 88(3), 224–229 (2020)

A Human Pathogen Capnocytophaga Ochracea Exhibits Current Producing Capability

Shu ZHANG,a,b,† Waheed MIRAN,c,† Divya NARADASU,c,d,† Siyi GUO,a and Akihiro OKAMOTOc,e,f,*

a Interfacial Energy Conversion Group, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japanb Section of Infection and Immunity, Norris Comprehensive Cancer Center, University of Southern California,925 W 34th Street, Los Angeles, CA 90089, USA

c International Center for Materials Nanoarchitectonics, National Institute for Materials Science,1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japan

d Department of Advanced Interdisciplinary Studies, RCAST, Graduate School of Engineering, The University of Tokyo,4-6-1 Komaba, Meguro-ku, Tokyo 153-8904, Japan

e Center for Sensor and Actuator Material, National Institute for Materials Science, 1-1 Namiki, Tsukuba, Ibaraki 305-0044, Japanf Graduate School of Chemical Sciences and Engineering, Hokkaido University, North 13 West 8, Kita-ku, Sapporo, Hokkaido 060-8628, Japan

* Corresponding author: OKAMOTO.Akihiro@nims.go.jp

ABSTRACTMicrobial extracellular electron transfer (EET) in diverse environments has gained increasing attention. However, the EET capability of oralpathogens and associated mechanisms has been scarcely studied. Here, our results suggest that the Capnocytophaga ochracea, anetiological human pathogen showed current production and demonstrated a rate enhancement of electron transport at a high cell-density.C. ochracea produced ~10-fold more current at an OD600 of 0.5 associated with twice a higher glucose consumption rate per cell, comparedto 0.1, measured in a three-electrode electrochemical system by single-potential amperometry at +0.2 V (vs Ag/AgCl [sat. KCl]). Duringcurrent production, the accumulation of the redox molecules on the electrode was observed at high OD600 compared to low OD600. Apartfrom cell released redox active product, externally added redox active additives enhanced the electron transport, suggesting the EETcapability of C. ochracea via electron mediator. A higher metabolic activity via single-cell assay (based on anabolic incorporation of 15NH4

+)in cells that did not attach to the electrode strongly suggests the EET rate enhancement through an electron mediator. As bacterialpopulations play a role in the pathogenesis of human infections such as periodontitis, our results suggest that population-induced EETmechanisms may facilitate in-vivo colonization of C. ochracea.

© The Author(s) 2020. Published by ECSJ. This is an open access article distributed under the terms of the Creative Commons Attribution 4.0 License (CC BY,http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse of the work in any medium provided the original work is properly cited. [DOI:10.5796/electrochemistry.20-00021]. Uploading "PDF file created by publishers" to institutional repositories or public websites is not permitted by the copyrightlicense agreement.

Keywords : extracellular electron transfer, oral pathogen, redox active molecules, electrochemical system

1. Introduction

Extracellular electron transfer (EET) mechanism has been well-studied in environmental microbes, such as Shewanella oneidensisMR-1 and Geobacter sulfurreducens PCA.1–4 Direct electrontransport by outer-membrane cytochromes and long-range electronshuttling processes are known as major mechanisms of EET in iron-reducing strains. Further, nanowire or conductive biofilm matricesalso mediate long-range electron transport, particularly in maturedbiofilms.5,6 Recently, distinct kinds of enzymes were identified inmarine sulfate-reducing bacteria Desulfovibrio ferrophilus IS57 andfood-borne pathogen Listeria monocytogenes.8 In particular, theidentification of flavin-based enzymes in L. monocytogenes, ahuman pathogen, implies divergent EET mechanisms in the humanmicrobiome. Further, Enterococcus faecalis9,10 and two otherisolates from the human gut share high homology with Klebsiellapneumoniae and Enterococcus avium, with respect to the 16S rRNAgene sequence,11 and are capable of EET. Their EET potential mayfacilitate their colonization in the host environment, similar to EETin the natural environment. Therefore, it is of great importance tounderstand the mechanism of EET at high cell-density.

Cell-density plays an important role in microbial physiology andecology. This feature profoundly affects the overall function ofmicrobial interactions in biofilms, such as metabolism, diversity, and

pathogenesis.12,13 In the 1970s, it was determined that bacteria couldsense the population-density of their and neighboring microorgan-isms and integrate this information into gene regulatory circuitsvia the production of diffusible signal molecules (autoinducers, i.e.,AI-2), which is now commonly known as ‘quorum sensing(QS)’.14,15 Bacteria in biofilms display a different phenotype throughQS in response to fluctuations in cell population and ultimatelymany physiological traits.16 However, the role of QS in microbialEET, especially in bacterial pathogens, is not well understood.

To this end, we studied the bacterial specie, Capnocytophagaochracea (C. ochracea), an oral pathogen that causes periodontaldisease and is also found in blood, contributing to opportunisticinfections such as sepsis and brain abscesses in immunocompro-mised patients.17,18 In the oral cavity, C. ochracea is prone to formbiofilms with gliding motility and also known to possess a QS luxSsystem.19 In addition, dehydrogenases-cytochromes in the cellularmembrane of C. ochracea were identified four decades ago;20

however, the biological function of these redox enzymes and thepotential EET capability of this microbial strain have yet to bestudied. Therefore, we conducted a whole-cell electrochemicalassay, single-potential amperometry, and differential pulse voltam-metry to study the EET capability and the effect of cell-densitydependency on current production. The microbial capability to usesoluble electron shuttling molecules was also examined in detailcombined with a cell-specific isotope, single-cell activity assay. TheEET mechanisms were further examined using a mutant strain of³Authors equally contributed to the study.

Electrochemistry Received: February 19, 2020

Accepted: March 17, 2020

Published online: April 21, 2020

The Electrochemical Society of Japan https://doi.org/10.5796/electrochemistry.20-00021

224

C. ochracea lacking the gene encoding the QS luxS system (termedas LKT7), which produced significantly fewer auto-inducers (AI-2)and substantially less biofilm formation than wild-type (WT).19

2. Materials and Methods

2.1 Bacterial growth media and culture conditionsCapnocytophaga ochracea ATCC 27872 was grown in 80mL of

DSMZ 340 medium supplied with 1 g/L sodium bicarbonate at37°C. The medium excluding glucose and hemin was autoclaved for15min at 121°C before use. Glucose and hemin were separatelyprepared, and filter-sterilized prior to inoculation. To maintain theanaerobic growth condition, 20min of N2/CO2 (80:20 v/v) gassparging was done prior to culture inoculation. In the electrochem-ical experiments, C. ochracea at the late exponential growth phaseor early stationary phase was used. The luxS gene (Coch_1216)deleted C. ochracea termed as LKT7 was previously constructedusing a PCR-ligation-mutation strategy,19 kindly provided by thegroup of Professor Kazuyuki Ishihara from Tokyo Dental Collegeand grown with the same conditions as wild type for our experiments.

2.2 Electrochemical cell operation and measurementsElectrochemical experiments were established using a single-

chamber three-electrode reactor in the anaerobic chamber (COYglovebox). Three electrodes include: 1- Working electrode, a glasssubstrate coated with tin-doped indium oxide (ITO) layer of 600 nmby spray pyrolysis deposition (SPD Laboratory, Inc., Japan); 2-Counter electrode, platinum wire of 0.1mm thickness; and 3-Reference electrode, Ag/AgCl (sat. KCl). Transparent ITO electrodewas used to confirm cellular attachment before the further analysis.As an electrolyte, a defined medium (DM) was prepared withfollowing composition (1L): 2.5 g NaHCO3, 0.08 g CaCl2·2H2O,1.0 g NH4Cl, 0.2 g MgCl2·6H2O, 10.0 g NaCl, 7.2 g HEPES, 0.5 gyeast extract and 10mM glucose as the carbon source. The pH of themedium was adjusted to 8.0. The sterilized electrolyte (4.8mL perreactor) was purged with N2 for 15min to remove dissolved oxygenin the reactor. Active Capnocytophaga bacteria, collected fromenrichment medium was washed twice with DM to remove nutrientsand resuspended in DM. Concentrated cell culture was added intothe electrochemical reactor with a final OD600 according to therequirement of the experiments. All electrochemical experimentswere carried out at 37°C without any agitation in the reactors.Single-potential amperometry and differential pulse voltammetrywere measured with an automatic polarization system (VMP3, BioLogic Company, France) under conditions such as 5.0-mV pulseincrements, 50-mV pulse amplitude, 300-ms pulse width, and a5.0-s pulse period.21 SOAS software, which is an open sourceprogram to analyze experimental electrochemical data was usedfor our analysis on differential pulse (DP) voltammograms. Thebackground current was subtracted by fitting the baseline fromregions sufficiently far from the peak assuming continuation of asimilar and smooth charging current throughout the peak region.22

Metabolites were measured as previously described.11 Glucoseconcentrations were measured by using a glucose assay kit (GAGO-20, Sigma-Aldrich) according to the manufacturer’s instructions.

2.3 Scanning electron microscopy (SEM)At the end of electrochemical experiments, the ITO electrodes

were collected for SEM sample preparation. The ITO slides werefirst fixed with 2.5% (v/v) glutaraldehyde and followed by gradientdehydration. Briefly, fixed samples were washed three times by PBSbuffer, followed by dehydration in (v/v) 25%, 50%, 75%, 90%, and95% ethanol, and three times with t-butanol and then overnightfreeze-dried under vacuum. The vacuum freeze-dried slides werecoated with evaporated platinum and observed via a Keyence VE-9800 microscope.

2.4 Fluorescence spectroscopyFlavin secretion by C. ochracea was tested by measuring the

fluorescence intensity of cell-free supernatant in an anaerobic quartzcuvette. JASCO-FDT 538 (Jasco corporation, Japan) instrument wasused for the fluorescence spectrometry measurements. Flavin wasthe target secretion molecule and therefore excitation wavelengthpertaining to flavin, i.e., 450 nm was selected for fluorescencespectroscopy. Control analysis of deaerated 3mL DM in the presenceand absence of 5 µM riboflavin was subjected to fluorescence andemission spectrum of oxidized riboflavin was measured by excitationat 450 nm. Emission peak intensity at 530 nm was measured.

2.5 Sample preparation for nanoscale secondary ion massspectroscopy (NanoSIMS)

Prior to NanoSIMS analysis, for the electrode non-attached cells,supernatant from the reactor was removed without disturbing theelectrode attached cells, and the cells collected from the supernatantwere placed on a new ITO electrode for NanoSIMS analysis. Forelectrode attached cells, the ITO surface was washed several timesand only the firmly attached cells on ITO were analyzed byNanoSIMS. Samples of each condition were fixed in 2.5% (v/v)glutaraldehyde phosphate medium and dehydrated in an ethanolgradient and t-butanol as previously described.23 CAMECANanoSIMS 50L system (CAMECA, Gennevilliers, France) wasused to analyze the biofilm sample. Briefly, a Cs+ beam approachedsample and irradiated four secondary ions (12C¹, 13C¹, 12C14N¹, and12C15N¹) emitted from the sample surface. Four secondary-ionimages of each sample were recorded with a raster size of 25 ©25µm2 and analyzed by Fiji (version 1.0) for the isotopeassimilation calculation. Given C. ochracea cells tend to formweb-like morphology,19 to fully recover the isotope signals, thesignals for each NanoSIMS image were averaged. Using the plugin,Open MIMS Image, all signal regions, including clusters of cellswere selected from each image (regions of interest, ROI) and theirpixel values were quantified for isotopic ratio analysis. Average andstandard deviation were calculated based on the number of ROIs.The ratio of 13C and 15N was calculated as follows:

13C- ratio ¼13C-pixel

ð13Cþ 12CÞ-pixel � 100% ð1Þ

15N- ratio ¼12C15N-pixel

ð12C15Nþ 12C14NÞ-pixel � 100% ð2Þ

3. Results and Discussion

3.1 Enhanced current production capability in C. ochracea athigher cell-density

To address the possibility of current production and to evaluatecell-density behavior during the current generation of C. ochracea, athree-electrode electrochemical system was used. Single-potentialamperometry was performed at +0.2V using the DM as anelectrolyte in the presence of 10mM glucose. Upon the addition ofbacterial cells into the reactor to a final OD600 of 0.5, an immediatecurrent increase was observed, reaching approximately 60 nAcm¹2

after 5 h (Fig. 1A and Fig. S1A). Such low current is also observedin other weak electricigens.24 A significantly less extent, but asimilar time course of current production was observed even in theabsence of glucose, indicating that approximately 50% of currentproduction was derived from glucose oxidation and the remainingcurrent from nutrients such as yeast extract present in the medium(Fig. S1A). Accordingly, we confirmed a significant decrease inglucose concentration during current production (Fig. S1B). Theseresults strongly suggest that C. ochracea is an EET-capablebacterium with current production associated with glucose oxidation.

We next examined the cell-density dependency on currentproduction in C. ochracea at different OD600 from 0.1 to 1.0 under

Electrochemistry, 88(3), 224–229 (2020)

225

the same electrochemical condition as mentioned above. At ODs of0.1 and 0.2, current production did not significantly differ from thebackground current (Fig. 1A), while cellular attachment on theelectrode surface was confirmed by SEM (Fig. S2A). In contrast,when the cell-density was at an OD600 of 0.5, current production was³10-fold higher than 0.1. Furthermore, although the cell-densitywas sufficient to cover the whole electrode surface at an OD600 of0.5 (Fig. S2B), a higher current was produced at an OD600 of 1.0.Also, a significant increase in charge transfer to the anode at higherODs was observed, which was calculated for 12 h by subtracting thecharge generated in the absence of glucose from that generated inthe presence of glucose by C. ochracea at their respective cell-densities (Fig. 1B).

C. ochracea exhibited that the current production is not propor-tional to the cell concentration unlike Shewanella oneidensis MR1,a model EET-capable microbe in the environment. The currentproduction of Shewanella came from the electrode-attached singlecell using a direct electron transport mechanism via outer-membranec-type cytochromes.25 However, electron transport of C. ochraceawas enhanced at a certain threshold of cell-density, not following thesame mechanism of direct electron transport to the electrode asin Shewanella. A possible explanation is that Shewanella usesnon-fermentable carbon sources and so the metabolic activity ofShewanella might be low when it cannot use an electron acceptor.Therefore, cells that are not able to perform direct electron transportvia outer membrane cytochromes to the electrode (or able to availthemselves of a mediator) would be expected to be relativelymetabolically inactive. In contrast, C. ochracea can ferment glucoseand so that the cells regardless of the direct electron transport

mechanism to the electrode should also have considerable metabolicactivity and impact the current production.

Glucose concentration was measured after 8 h for OD600 of 0.1and 0.5 during current production under the same condition asFig. 1A. A 10-fold higher glucose consumption rate at OD600 of 0.5compared to 0.1 showed that the metabolism of C. ochracea wassubstantially activated at high cell-density (Fig. 1C). The electro-chemically detected quantity of cell released redox active product onthe electrode by differential pulse voltammetry also showed asignificant increase with OD (Fig. 1D and Fig. S3). However, themagnitude of difference in DP voltammogram peaks for low OD andhigh OD is not same as the current production and glucoseconsumptions, because DP voltammogram analysis reflects theinformation about redox active molecules, either coming frombacterial surface electron transfer agent and/or soluble mediators.DP voltammogram measured at OD600 of 0.1 and 0.5 showed aslight difference in the peak potential (Ep) shift approximately from¹0.35 (OD600 of 0.1) to ¹0.30 (OD600 of 0.5)V (Fig. 1D). Thesedata suggest that current enhancement at high cell-density resultsfrom both metabolic activation and the increase in cell releasedredox active product having peak potential of approximately¹0.30V.

3.2 Population-induced EET mechanism via cell releasedredox active product in C. ochracea

To examine whether C. ochracea could use the cell releasedredox active product detected in DP voltammogram in its currentproduction, we performed supernatant swapping with fresh mediumin the reactor at an OD600 of 0.5, after current production was

-0.6 -0.4 -0.2 0.0 0.20

3

6

9

12

15 OD600:0.1 OD600:0.5

delta

I (μ

A)

E (V vs Ag/AgCl [sat. KCl])

0

20

40

60

80

100

1.0 OD 0.2 OD 0.5 OD

Cur

rent

den

sity

(nA

cm

-2)

Time (h)

0.1 OD WO/Glu 0.1 OD

0 3 6 9 12

0.1 OD 0.5 OD 0.00

0.06

0.12

0.18

0.24

0.30

Glu

cose

con

sum

ptio

n ra

te (m

M h

-1)

Cells added

B)A)

D)C)0.1

OD

0.2OD

0.5 O

D

1.0OD

0

1

2

3

4

5

Tota

l cou

lom

bs (m

C)

Figure 1. Population-induced EET capability and metabolism activation of C. ochracea. (A) Current vs time measurements conducted inanaerobic reactors equipped with ITO electrodes (surface area: 3.14 cm2) poised at +0.2V vs Ag/AgCl (sat. KCl) in the absence (red dashedline) and presence of 10mM glucose and the OD600 of 0.1, 0.2, 0.5 and 1.0 represent the initial cell densities added to the reactors. Arrowposition indicates the time of cell addition in the electrochemical reactor. (B) Total coulombs generated by C. ochracea with different initialcell densities as in Fig. 1A, calculated by subtracting the charge generated in the absence of glucose from that generated in the presence ofglucose for 12 hours. (C) Glucose consumption rate based on 8 h sampling from anaerobic reactors of C. ochracea measured at OD600 of 0.1and 0.5 with initial glucose concentration of 10mM. Data values represent the mean « standard deviations from two independentexperiments. (D) Baseline subtracted differential pulse (DP) voltammogram of C. ochracea measured for anaerobic reactors at OD600 of 0.1and 0.5.

Electrochemistry, 88(3), 224–229 (2020)

226

saturated at approximately 8 h. The removal of planktonic cellsdecreased the current production (Fig. 2A), suggesting that anelectron mediator may contribute to the current production at theOD600 of 0.5. We confirmed the presence of cell released redoxactive product at low and high OD600 by voltammetric analysis ofthe cell-free supernatant (cells were removed by centrifugation)collected from the reactor. The DP voltammogram of the cell-freesupernatant solution from OD600 of 0.5 showed a 7-times higherpeak current (¦I) compared to OD600 of 0.2. Furthermore, while anOD600 of 0.2 showed a single peak at Ep of ¹0.20V, an oxidativesignal in the case of OD600 at 0.5 contained multiple peaks atdifferent Ep values (Fig. 2B). These significant differences inoxidative signals suggest that the accumulated cell released redoxactive product at OD600 of 0.5 is critical for the current enhancement.

We next examined the effect of external redox active additives(which are also essential nutrients in many dietary products), such asmenadione (Vitamin K3, VK3) and riboflavin (Vitamin B2, VB2),on current production. The addition of VK3 and VB2 to the reactorwith high OD significantly increased current production over 10-fold (Fig. 3A) compared to the absence of redox active additives. Inaddition to Ep of ¹0.20V assignable to the cell surface protein,sharp oxidative peaks were observed with Ep at ¹0.40V and+0.30V in the presence of VB2 and VK3 in DP voltammogram,respectively (Fig. 3B). Therefore, the population-induced EETmechanism was dominated by the electron mediator mechanism,and C. ochracea likely accumulated cell released redox activeproduct as shown in the schematic (Fig. 4). Further, becauseCapnocytophaga is a flavobacterium26 that secretes flavins, wemeasured the fluorescence spectrum of the cell-free supernatantcollected from the electrochemical reactor during EET to detect the

presence of secreted flavins (Fig. S4). However, no difference wasobserved between fresh DM and cell-free supernatant, indicatingthat C. ochracea produced insignificant amounts of flavins in theelectrochemical reactor, confirming that flavin was not involvedin the current generation by C. ochracea and there is likely aninvolvement of other cell released redox active product.

The larger current production at high OD600 0.5 was associatedwith both higher glucose metabolic activity (Fig. 1C) and the higherconcentration of cell released redox active product secretioncompared to low OD (Fig. 1C, 2B). To this end, we analyzed themetabolic activity of bacterial cells in the electrochemical cell bynanoscale secondary ion mass spectrometry (NanoSIMS). Nano-SIMS is a robust tool to visualize and quantify the incorporation oflabeled substrates in a single cell.27,28 Therefore, we used 13Clabeled glucose and 15N labeled ammonium chloride (15NH4Cl) toinvestigate the anabolic activity coupled to glucose metabolismin C. ochracea. In the single-potential amperometry condition,C. ochracea harvested from the electrode surface presented less 13Cand 15N anabolic activity than non-attached cells in the supernatant(Fig. 5A and B). The lower activity of the attached cells may bederived from the low nutrient or glucose concentration at thebottom of the cell accumulation. Subsequently, the 13C/Ctotal (%)and 15N/Ntotal (%) in the non-electrode attached C. ochracea werecompared in the presence of external redox active additives, VB2and VK3 (Fig. 5C). It can be seen that high activity was moreabundant in electrode non-attached cells in comparison to electrodeattached cells, supporting the significance of non-attached cells onmetabolic activity and ultimately enhanced the EET.

It is important to mention that electron mediators affectinteractions with the bacterial surface and eventually enhance the

0

14

28

42

56

70

Cur

rent

den

sity

(nA

cm

-2)

Time (h)

A)

Cells added(OD600:0.5)

After fresh DM exchange

B)

0.0

0.4

0.8

1.2

1.6

2.0 OD600:0.2 (filtered supernatant) OD600:0.5 (filtered supernatant)

delta

I (μ

A)

E (V vs Ag/AgCl [sat. KCl])

0 5 10 15 20

-0.6 -0.4 -0.2 0.0 0.2

Figure 2. Electron mediating capability of C. ochracea and cellreleased redox active product accumulation. (A) Supernatantreplacement during the current production of C. ochracea at thepoised potential of +0.2V vs Ag/AgCl (sat. KCl). At indicatedtimes, the medium was removed and replaced with fresh definedmedium (DM) containing 10mM glucose led to a decrease incurrent production. (B) Baseline subtracted differential pulse (DP)voltammogram showing the peak potential of cell-free supernatantscollected from OD600 of 0.2 and OD600 of 0.5 reactors.

0

300

600

900

1200

1500 Control 10. μM VB2 20. μM VB2 10. μM VK3

Cur

rent

den

sity

(nA

cm

-2)

Time (h)

Cellsadded

0 4 8 12 16

-0.9 -0.6 -0.3 0.0 0.3 0.64

8

12

16

20 Control 10. μM VB2 20. μM VB2 10. μM VK3

delta

I (μ

A)

E (V vs Ag/AgCl [sat. KCl])

A)

B)

Figure 3. Impact of external redox active additives on electro-chemical measurements of C. ochracea cells. (A) Current vs timemeasurements by C. ochracea in the absence (control) and thepresence of external redox active additives i.e., Vitamin K3 (VK3)and Vitamin B2 (VB2). Line break in 20 µM VB2 profile is due todifferential pulse (DP) voltammogram measurements at that point.(B) DP voltammogram of C. ochracea in the presence and absence(control) of redox active additives (VB2, VK3).

Electrochemistry, 88(3), 224–229 (2020)

227

electron transfer rate to maintain the intracellular redox environ-ments, i.e., redox homeostasis.29 Metabolism-associated electrontransfer of C. ochracea could act in concert to promote pathoge-nicity and biofilm development. Moreover, EET linked topopulation-level phenotypes can be critical during microbialcolonization and biofilm formation. Therefore, crosstalk betweenEET and QS is likely a strategy against environmental stresses byC. ochracea.

3.3 EET enhancement of C. ochracea at high cell-density isnot regulated by quorum sensing gene luxS

The enhanced current generation due to higher concentration cellreleased redox active product at higher cell-density by C. ochraceasuggests an EET-linked mechanism triggered by a cell-densitythreshold as observed in QS mechanisms.14 To further investigate theinvolvement of QS regulation in the EET rate enhancement ofC. ochracea at high cell-density, we compared the current productionof C. ochracea (WT) and its mutant strain (LKT7). LKT7, lackingthe luxS gene, is a QS-deficient strain, which produces significantlyfewer auto-inducers (AI-2) and less biofilm formation than WT, butwith a similar growth rate.19 Current generation measurements usingthe same conditions (OD600 of 0.5) revealed that LKT7 was alsocapable of producing significant anodic current, comparable to WT(Fig. 6A). Hence, these deletion factors hardly pertained to the EETcapability of C. ochracea. DP voltammogram results showed thepossible involvement of the same redox enzyme in electron transfer,as notable peak potential at almost the same Ep were observed in bothWT and LKT7 (Fig. 6B). These results suggest that the known luxSgene-based QS in C. ochracea was unlikely to contribute to EETenhancement at high cell-density. However, the possible involvementof another QS mechanism cannot be ruled out.

In oral microbiota, it has been reported that Porphyromonasgingivalis enhanced biofilm formation of Fusobacterium nucleatumby releasing diffusible molecules, although AI-2 was not involved.30

Moreover, the release of soluble molecules from C. ochracea playeda significant role in biofilm formation by F. nucleatum. Further,induction of coaggregation between F. nucleatum and C. ochraceahas been reported.31 However, the soluble molecules released byC. ochracea in our study have redox properties and may play acritical role both in oral polymicrobial biofilms and the associationof other niche bacteria (independent of QS). The cell-densitydependency for metabolic activation and redox mediator secretionsuggests that a different, but similar, regulatory pathway other thanQS luxS in C. ochracea. It indicates a possible new mode ofpopulation-dependent QS mechanism, leading to enhanced meta-bolic activation and cell released redox active product. There aremany other types of QS mechanism have been identified in Gram-negative bacteria, such as LuxI/LuxR,32 LasI/LasR, RhlI/RhlR,PQS,33 and quorum quenching (QQ) enzymes.34 Therefore, othertypes of QS mechanisms in enhancing EET at higher cell-density isalso possible. Moreover, the pathogenesis of oral bacteria has beenwell studied; yet, their microbial electrical activities, in particular,the capability of extracellular electron transportation have not beenreported regarding population-level phenotypes. Therefore, theactivities of C. ochracea and other extracellularly electron-trans-ferring organisms should be reexamined to develop pathogenicitycontrol models.

B)A)

cellscells

Attached Non-attached

C)

Attached

Non-at

tache

d

Attache

d(VB2)

Non-atta

ched(VB2)

Attached

(VK3)

Non-at

tache

d (VK3)

0

1

2

3

4

5

Isot

ope

assi

mila

tion

(%)

13C/Ctotal (%)15N/Ntotal (%)

low

high

Figure 5. NanoSIMS analysis of C. ochracea in the presence andabsence of redox active additives. NanoSIMS images of electrodeattached cells (A) and non-attached cells (B) showing the 15N/Ntotal

(%) assimilation. Arrows indicating the rod-shaped C. ochraceacells (scale bars: 5 µm). Warmer colored cells are more enriched in15N/Ntotal (%), which corresponds with the higher levels of anabolicactivity. (C) Bar graph showing the isotopic assimilation of 13C/Ctotal (%) and 15N/Ntotal (%) in electrode surface attached and non-attached cells analyzed via NanoSIMS analysis. Given C. ochraceacells tend to form web-like morphology, to fully recover the isotopesignals, the signals for each NanoSIMS image were averaged. Datavalues representing the mean « standard deviations from twoindependent experiments and a similar tendency was observed inmore than four individual experiments.

ITO

ITO[low OD; limited EET]

Cell

Electron mediator

Red Ox

[High OD; EnhancedEET by “Mediator”]

Figure 4. Schematic of possible electron transport mechanismby C. Ochracea cells. At low OD the EET is limited by theconcentration of cell released redox active product, whereas athigher ODs the concentration is increased which acts as an electronmediator. Cell-released mediator attaches to the cell and transferthe electrons to ITO via reduction (Red) and oxidation (Ox) statecycling.

Electrochemistry, 88(3), 224–229 (2020)

228

4. Conclusions

Our electrochemical analysis results suggested that C. ochracea ahuman pathogen has the EET capability with population-inducedEET paradigm and metabolic activation. Interestingly, the cell-density dependent electron transport observed in this study isindependent of the QS regulation luxS. Importantly, the EET ratewas enhanced at higher cell densities and single-cell anabolicactivity showed that electrode non-attached cells played a very vitalrole in EET enhancement. These distinct mechanisms are likelycritical during microbial colonization and biofilm formation ofC. ochracea in the human environment. Therefore, the presentfindings of cell-density dependent mechanisms may provide criticalaspects for suppressing the microbial activity, e.g., medicalapplication and drug design. Identification of cell released redoxactive product and proteins would be an interesting future study.

Supporting Information

The Supporting Information is available on the website at DOI:https://doi.org/10.5796/electrochemistry.20-00021.

Author Contributions

This work was conceived by S.Z., and A.O. S.Z., D.N., W.M.,and S.G. performed the experiments and collected data. All authorscontributed intellectually to the analysis and interpretation of thedata. W.M., D.N., S.Z., and A.O wrote the manuscript.

Acknowledgments

We thank Prof. Ishihara Kazuyuki at Tokyo Dental College forproviding the mutant strain of LKT7. This work was financially

supported by a Grant-in-Aid from the Japan Society for Promotionof Science (JSPS) KAKENHI Grant Number 17H04969 to A.O.,and the Japan Agency for Medical Research and Development(AMED) (19gm6010002h0004 to A.O.).

Conflict of Interests

Authors declared that there is no conflict of interest involved inthis work.

References

1. Y. A. Gorby, S. Yanina, J. S. McLean, K. M. Rosso, D. Moyles, A. Dohnalkova,T. J. Beveridge, I. S. Chang, B. H. Kim, K. S. Kim, D. E. Culley, S. B. Reed, M. F.Romine, D. A. Saffarini, E. A. Hill, L. Shi, D. A. Elias, D. W. Kennedy, G.Pinchuk, K. Watanabe, S. Ishii, B. Logan, K. H. Nealson, and J. K. Fredrickson,Proc. Natl. Acad. Sci. U.S.A., 103, 11358 (2006).

2. G. Reguera, K. D. McCarthy, T. Mehta, J. S. Nicoll, M. T. Tuominen, and D. R.Lovley, Nature, 435, 1098 (2005).

3. C. R. Myers and K. H. Nealson, Science, 240, 1319 (1988).4. H. Richter, K. McCarthy, K. P. Nevin, J. P. Johnson, V. M. Rotello, and D. R.

Lovley, Langmuir, 24, 4376 (2008).5. G. Reguera, Proc. Natl. Acad. Sci. U.S.A., 115, 5632 (2018).6. K. Kondo, A. Okamoto, K. Hashimoto, and R. Nakamura, Langmuir, 31, 7427

(2015).7. X. Deng, N. Dohmae, K. H. Nealson, K. Hashimoto, and A. Okamoto, Sci. Adv., 4,

eaao5682 (2018).8. S. H. Light, L. Su, R. Rivera-Lugo, J. A. Cornejo, A. Louie, A. T. Iavarone, C. M.

Ajo-Franklin, and D. A. Portnoy, Nature, 562, 140 (2018).9. D. Keogh, L. N. Lam, L. E. Doyle, A. Matysik, S. Pavagadhi, S. Umashankar,

P. M. Low, J. L. Dale, Y. Song, S. P. Ng, C. B. Boothroyd, G. M. Dunny, S. Swarup,R. B. H. Williams, E. Marsili, and K. A. Kline, MBio, 9, e00626-17 (2018).

10. G. Pankratova, D. Leech, L. Gorton, and L. Hederstedt, Biochemistry, 57, 4597(2018).

11. D. Naradasu, W. Miran, M. Sakamoto, and A. Okamoto, Front. Microbiol., 9,3267 (2019).

12. S. Freilich, R. Zarecki, O. Eilam, E. S. Segal, C. S. Henry, M. Kupiec, U. Gophna,R. Sharan, and E. Ruppin, Nat. Commun., 2, 589 (2011).

13. A. G. Patterson, S. A. Jackson, C. Taylor, G. B. Evans, G. P. C. Salmond, R.Przybilski, R. H. J. Staals, and P. C. Fineran, Mol. Cell, 64, 1102 (2016).

14. K. H. Nealson, Arch. Microbiol., 112, 73 (1977).15. D. G. Davies, M. R. Parsek, J. P. Pearson, B. H. Iglewski, J. W. Costerton, and

E. P. Greenberg, Science, 280, 295 (1998).16. H. Gu, S. Hou, C. Yongyat, S. De Tore, and D. Ren, Langmuir, 29, 11145 (2013).17. S. S. Desai, R. A. Harrison, and M. D. Murphy, J. Infect., 54, e107 (2007).18. H. K. Wang, Y. C. Chen, L. J. Teng, C. C. Hung, M. L. Chen, S. H. Du, H. J. Pan,

P. R. Hsueh, and S. C. Chang, J. Clin. Microbiol., 45, 645 (2007).19. K. Hosohama-Saito, E. Kokubu, K. Okamoto-Shibayama, D. Kita, A. Katakura,

and K. Ishihara, PLoS One, 11, e0147114 (2016).20. T. T. Lillich and R. Calmes, Arch. Oral Biol., 24, 699 (1979).21. A. Okamoto, K. Hashimoto, K. H. Nealson, and R. Nakamura, Proc. Natl. Acad.

Sci. U.S.A., 110, 7856 (2013).22. V. Fourmond, K. Hoke, H. A. Heering, C. Baffert, F. Leroux, P. Bertrand, and C.

Leger, Bioelectrochemistry, 76, 141 (2009).23. J. Saito, K. Hashimoto, and A. Okamoto, Electrochemistry, 85, 444 (2017).24. L. E. Doyle and E. Marsili, Bioresour. Technol., 258, 354 (2018).25. R. Nakamura, F. Kai, A. Okamoto, G. J. Newton, and K. Hashimoto, Angew.

Chem., Int. Ed. Engl., 48, 508 (2009).26. D. Kita, S. Shibata, Y. Kikuchi, E. Kokubu, K. Nakayama, A. Saito, and K.

Ishihara, Appl. Environ. Microbiol., 82, 1756 (2016).27. S. E. McGlynn, G. L. Chadwick, C. P. Kempes, and V. J. Orphan, Nature, 526,

531 (2015).28. A. R. Sheik, E. E. L. Muller, J.-N. Audinot, L. A. Lebrun, P. Grysan, C. Guignard,

and P. Wilmes, ISME J., 10, 1274 (2016).29. R. Moscoviz, C. Flayac, E. Desmond-Le Quéméner, E. Trably, and N. Bernet, Sci.

Rep., 7, 7766 (2017).30. Y. Saito, R. Fujii, K. I. Nakagawa, H. K. Kuramitsu, K. Okuda, and K. Ishihara,

Oral Microbiol. Immunol., 23, 1 (2008).31. T. Okuda, K. Okuda, E. Kokubu, T. Kawana, A. Saito, and K. Ishihara, Anaerobe,

18, 157 (2012).32. D. Paul, J. Gopal, M. Kumar, and M. Manikandan, Chem. Biol. Interact., 280, 86

(2018).33. A. Kariminik, M. Baseri-Salehi, and B. Kheirkhah, Immunol. Lett., 190, 1 (2017).34. M. López, A. Rueda, J. P. Florido, L. Blasco, L. Fernández-García, R. Trastoy, F.

Fernández-Cuenca, L. Martínez-Martínez, J. Vila, A. Pascual, G. Bou, and M.Tomas, Sci. Rep., 8, 2523 (2018).

0

25

50

75

100C

urre

nt d

ensi

ty (n

A c

m-2)

Time (h)

A)

Cells added(OD600:0.5)

0 4 8 12 16

-0.6 -0.4 -0.2 0.0 0.20

3

6

9

12

15 WT-OD600:0.5 LKT7-OD600:0.5

delta

I (μ

A)

E (V vs Ag/AgCl [sat. KCl])

B)

Figure 6. Effect of luxS gene deletion on EET of C. ochracea.(A) Current vs time profiles of C. ochracea WT and mutant strainLKT7 (lacking quorum sensing gene luxS) at OD600 of 0.5 at thepoised potential of +0.2V vs Ag/AgCl (sat. KCl). (B) Baselinesubtracted differential pulse (DP) voltammogram of C. ochraceaWT and mutant strain- LKT7 at OD600 of 0.5.

Electrochemistry, 88(3), 224–229 (2020)

229