Protocol

Characterizing the Anoxic Phenotype of

Pseudomonas putida Using a
Bioelectrochemical System
Bin Lai 1,2 , Anh Vu Nguyen 1 and Jens O Krömer 1,2, *
1 Systems Biotechnology group, Department of Solar Materials, Helmholtz Centre for Environmental
Research—UFZ, 04318 Leipzig, Germany; bin.lai@ufz.de (B.L.); anh-vu.nguyen@ufz.de (A.V.N.)
2 Center for Microbial Electrochemical Systems (CEMES), Advanced Water Management Center,
The University of Queensland, Brisbane QLD 4072, Australia
* Correspondence: jens.kroemer@ufz.de; Tel.: +49-341-235-4688




Received: 11 March 2019; Accepted: 27 March 2019; Published: 30 March 2019


Abstract: Industrial fermentation in aerobic processes is plagued by high costs due to gas transfer
limitations and substrate oxidation to CO2 . It has been a longstanding challenge to engineer an
obligate aerobe organism, such as Pseudomonas putida, into an anaerobe to facilitate its industrial
application. However, the progress in this field is limited, due to the poor understanding of the
constraints restricting its anoxic phenotype. In this paper, we provide a methodological description of
a novel cultivation technology for P. putida under anaerobic conditions, using the so-called microbial
electrochemical technology within a bioelectrochemical system. By using an electrode as the terminal
electron acceptor (mediated via redox chemicals), glucose catabolism could be activated without
oxygen present. This (i) provides an anoxic-producing platform for sugar acid production at high yield
and (ii) more importantly, enables systematic and quantitative characterizations of the phenotype of
P. putida in the absence of molecular oxygen. This unique electrode-based cultivation approach offers
a tool to understand and in turn engineer the anoxic phenotype of P. putida and possibly also other
obligate aerobes.

Keywords: Pseudomonas putida; Microbial electrochemical technology; bioelectrochemical system;

anoxic phenotype; anaerobic cultivation

1. Introduction
Pseudomonas putida (P. putida) is a promising industrial host for the production of harsh chemicals
due to its resilience against environmental stresses (e.g., solvents) that will quickly diminish other
industrial microorganisms [1–4]. This family, however, comprises obligate aerobes with oxygen
exclusively used as the terminal electron acceptor in nature. Compared to anaerobic processes,
oxygen-dependent metabolism can obtain higher energy yields, a high metabolic turnover and product
purity [5,6], but it also suffers significantly from the comparatively low product yield and high
operating cost for heat removal and oxygen supply [7,8]. These limitations pose a strong economic
barrier for the industrial application of P. putida, as well as other obligate aerobic strains and limits
feasible reactor sizes.
To overcome this bottleneck, it has been attempted to introduce a non-oxygen-dependent route
into P. putida for electron balance and thus to create an anaerobic mutant of P. putida. Nikel et al. [9]
engineered a fermentative pathway towards acetate and ethanol into P. putida KT2440 and another
group reported an approach of introducing a nitrate/nitrite-based respiration pathway in the same
strain [10]. Both systems were active in the recombinant strains; however, the mutants only showed

Methods Protoc. 2019, 2, 26; doi:10.3390/mps2020026 www.mdpi.com/journal/mps

Methods Protoc. 2019, 2, 26 2 of 10

limited metabolic activity and were not able to grow without oxygen. Identifying further engineering
targets to improve those strains is urgently required but is extremely difficult due to the poor
knowledge of the anaerobic constraints of an obligate aerobe (not limited to P. putida only) to date.
It is difficult to identify the key metabolic pathways (i.e., the potential targets for strain engineering)
since hardly any metabolic activity of P. putida is retained once oxygen is removed from the medium.
A novel cultivating method that could (even partially) activate anoxic metabolism of P. putida would
allow quantitative systems biology approaches to identify potential targets.
Over the last two decades, a technique termed microbial electrochemical technology has gained
increased interest. The fundamental principle is to use an electrode as the electron acceptor/donor for
microbial redox metabolism. Such a reactor is commonly named a bioelectrochemical system (BES) [11].
The ability to use a solid phase as electron acceptor via direct or indirect electron transfer route has
been found for many anaerobic or facultative anaerobic microorganisms such as Shewanella [12],
Geobacter [13], Pseudomonas aeruginosa [14] and more recently even gram-positive Listeria [15].
In this paper, we present a detailed protocol for the cultivation of P. putida in a BES reactor
without oxygen as well as the general results to be expected. The redox power from an anode and
a redox mediator can successfully drive the anoxic sugar metabolism of P. putida which then allows
further quantitative physiology characterizations. This electrode-based cultivation approach provides
a unique platform to quantitatively assess the anaerobic phenotype of P. putida cells, which cannot be
achieved by any other technique so far. We believe that this lays the basis for rationally developing an
anaerobically growing P. putida, and moreover, similar approaches may also be transferred to cultivate
other obligate aerobic microbes without oxygen present.

2. Experimental Design
The basic principle of this cultivation protocol is applying an anode as the final electron acceptor
for cellular redox metabolism. However, P. putida genetically have neither the extracellular iron
respiration system like Shewanella and Geobacter for direct electron transfer nor the pyocyanin
biosynthesis pathway existing in Pseudomonas aeruginosa for indirect electron transfer. Thus, to establish
an electron flow between the cells and anode, an artificial redox mediator has to be added in the
medium as an electron shuttle. Thermodynamically, the chemical should have a positive enough redox
potential to withdraw the electrons from cellular electron transport system, and the threshold was
found to be 0.207 V vs standard hydrogen electrode (SHE) [16].
Anaerobic growth of P. putida is not feasible so far, which is in fact the ultimate target of developing
this protocol. Therefore, an aerobic pre-culture preparation process is required to obtain enough
biomass for the BES cultivation and the subsequent quantitative physiology characterization. Figure 1
presents the general workflow of this protocol.
enough redox potential to withdraw the electrons from cellular electron transport system, and the
threshold was found to be 0.207 V vs standard hydrogen electrode (SHE) [16].
Anaerobic growth of P. putida is not feasible so far, which is in fact the ultimate target of
developing this protocol. Therefore, an aerobic pre-culture preparation process is required to obtain
Methods
enough Protoc. 2019, 2,
biomass for26the BES cultivation and the subsequent quantitative physiology characterization.
3 of 10

Figure 1 presents the general workflow of this protocol.

Figure 1. The general working flow for the cultivation and characterization of Pseudomonas putida in
bioelectrochemical system.

2.1. Reagents
• Potassium ferricyanide (Sigma-Aldrich, Munich, Germany; Cat. no. 244023)
• LB-agar plate: tryptone 10 g/L, yeast extract 5 g/L, NaCl 10 g/L and agar 17.5 g/L
• Trace solution: FeCl3 ·6H2 O 1.5 g/L, KI 0.18 g/L, H3 BO3 0.15 g/L, CoCl2 ·6H2 O 0.15 g/L,
MnCl2 ·4H2 O 0.12 g/L, Na2 MoO4 ·2H2 O 0.12 g/L, ZnSO4 ·7H2 O 0.12 g/L, CuSO4 ·5H2 O 0.03
g/L, EDTA (acid form) 10 g/L, and NiCl2 ·6H2 O 0.023 g/L.
• Potassium chloride solution: saturated concentration.
• DM9 medium: Na2 HPO4 6 g/L, KH2 PO4 3 g/L, NH4 Cl 1 g/L, MgSO4 ·7H2 O 0.10 g/L, 1 mL/L of
CaCl2 ·2H2 O stock solution (15 g/L), 1 mL/L of trace solution and glucose 5 g/L.
• Anode buffer (AB): Na2 HPO4 6 g/L, KH2 PO4 3 g/L, NH4 Cl 1 g/L, MgSO4 ·7H2 O 0.10 g/L, 1
mL/L of CaCl2 ·2H2 O stock solution (15 g/L), 1 mL/L of trace solution and glucose 1.5 g/L.
• Cathode buffer (CB): Na2 HPO4 6 g/L, KH2 PO4 3 g/L and NH4 Cl 1 g/L.

2.2. Equipment
• Potentiostat (Bio-logic science instrument, Seyssinet-Pariset, France; Cat. no.: VSP)
• Heating/refrigerated circulators (Julabo, Seelbach, Germany, Cat. no.: 9352506, 9312640)
• Magnetic stirring plate (LLG labware, Meckenheim, Germany; Cat. no.: 6.263.440)
• Oxygen sensor (PreSens, Regensburg, Germany; Cat. no.: OXY-4 mini)
• Gas flow meter (Cole-Parmer, Wertheim-Mondfel, Germany; Cat. no.: GZ-03216-08)
• Incubator (Infors AG, Basel, Switchzerland; Cat. No.: Multitron)
• Centrifuge (Eppendorf, Hamburg, Germany; Cat. no.: 5810R)
• Bioelectrochemical cells (custom-made, see Section 2.3 for details)

2.3. Bioelectrochemical System Configurations

The locally-made reactor is designed based on the following considerations: (i) it is sealable and
autoclavable to prevent contamination; (ii) it provides stable and defined conditions for the working
chamber (anodic chamber) where the cells will be cultivated; (iii) it has sufficient working volume for
Methods Protoc. 2019, 2, 26 4 of 10

multisampling and further analytics; and (iv) it is compatible for external plug-in sensors, e.g., pH,
pO2 , redox, etc.
In brief, we designed a double-jacketed cylinder-like glass reactor covered with polyether ether
ketone (PEEK)-made lids and plugs. The working chamber is separated from the counter chamber
(cathodic chamber) by an ion exchange membrane. The schematic diagram of the reactor is shown in
Figures 1 and 2 and described as below (the same order as numbered in Figure 1 step 2):
1. Glass vessel: working chamber, double-jacketed cylinder-like vessel with DN60 flat flange as
the open-top (DWK life sciences GmbH, Wertheim/main, Germany; Cat. no.: 21035340). NOTE:
Two open ports of GL14 screw size connected to the out-jacket chamber are for temperature
control, and the third open port of the same size connected to the inner working chamber is
for inoculation.
2. Lid: 100 mm diameter, fit DN60 flat flange and the corresponding quick release clamp (DWK
life sciences GmbH, Wertheim/main, Germany; Cat. no.: 10464691). Seven open ports were
designed on the lid for plugins, while six of them have the inner diameter of 12 mm and the left
(in the center) is of 2-mm diameter for working electrode cable. NOTE: All the measuring and
controlling units are plugged into the working chamber through the lid with screw plugs sealed
with O-rings. The lid is made from polyether ether ketone (PEEK) material because of its stability
and resistance against thermal, chemical and climate stresses.
3. Screw plugs: made from PEEK for the seven open ports on the lid. Four different designs are
used depending on the purpose: (i) plug with big hole (12 mm) in the center for glass tubes and
sensors; (ii) plug with no hole used as sealing stoppers for sparing ports; (iii) plug with small
hole (2 mm) only for working electrode cable; iv) plug with four small holes (3.2 mm each) for
gassing, sampling and feeding.
4. Bridging tube: a glass tube (out diameter of 12 mm) with the bottom-end sealed with porous
glass frit (pore size of < 1µm). NOTE: Not essential, but quite beneficial for two reasons: (i)
minimize the voltage drop; (ii) prevent reference electrode (Als, Tokyo, Japan; Cat. no.: 013503)
from being contaminated. The glass frit could also be fixed via heat shrink PTFE tube as well, but
this always falls off after autoclave.
5. Counter electrode compartment: a hollow glass tube (out diameter of 12 mm) with a GL14 screw
open end. NOTE: it is separated from the working chamber via cation exchange membrane using
GL14 open-top cap and O-ring.
6. Condensers: glass condenser with the flat neck of 12 mm diameter, the top open-end facing
outside the reactor is sealed with an autoclavable membrane filter (0.22 µm pore size). NOTE: it
is essential for balancing the reactor pressure and reducing water loss due to gas sparging.
Methods
7. Protoc. 2019,
Anode: 2, x
pretreated carbon cloth of 5 × 5 cm dimensions. 7 of 9

CRITICAL
NOTE: KEY STEP: Values of
component chosen in stepsystem.
the whole 21–22 areThecritical to minimize
raw carbon cloth isevaporation of water
quite hydrophobic
during
on operation.
the surface andForthusthe conditions
needs given above,
to be pretreated priorwetomeasured ~0.09mL water
use. The pretreatment wouldshould
method be lost
per hour.hydrophilicity
increase This value is essential for the
of the carbon quantitative
cloth evaluation
and also give (e.g., carbon and
a high overpotential electron
for oxygen balance)
evolution.
of the
The methods tested are either soaking overnight (about 16 h, 40 ◦ C, 200 rpm of
whole system.
successful
23. Connect the
shaking) workingcyclic
or running electrode, counter electrode
voltammetry and reference
for 100 cycles in 2 mM electrode
cetrimonium to the potentiostat.
bromide (CTAB)
24. Start the potentiostat and apply a chronoamperometry method to the BES
solution (Supplementary Figure S1). Identical performance can be observed for both approaches. reactor with the
following
The workingparameters:
potential working
window, electrode
defined aspotential (Ewe) of
no significant “0.5 V”
oxygen vs Ag/AgCl/KCl
evolution , current (I)
observedsat(otherwise
recording
oxygen of every
would dI of 100 through
be produced µ A and electrochemically
every dt of 10 min.oxidation of H2 O), for the treated carbon
NOTE:
cloth working
can be overelectrode potential
0.8 V (vs SHE), needs
which to be high
is sufficient forenough
oxidizingto efficiently oxidize
the mediators usedthe mediator,
(discussion
which can
below). Thebeunsuccessful
determined by cyclic voltammetry
pretreatment method;
method tested was0.5 V is determined
electrolysis in acids,forwhich
ferricyanide.
gave a
This value
strong needs tonoise
background be adjusted
from the when
blankchanging
electrodemediators, for instanceFigure
(see supplementary 0.3 V for
S1).[Co(bpy)3]Cl3).
8.25. Record current
Membrane: cationfor exchange
5–8 h formembrane
the fresh AB medium (over
(Membranes the dayINC.,
international time Ringwood,
of day 2), and a flat
NJ, USA;
background
Cat. signal should
no.: CMI-7000), 12-mmbe achieved
diameter before
disc. the It
NOTE: next step;to seal the bottom end of the counter
is used
NOTE: Quality
electrode control I:(the
compartment The one
current recorded
plugged intoon
thethe potentiostat
working for blank
chamber) AB medium
to separate should
it from the
be around 0 µ A fluctuated within a maximum of ±10 µ A range. Larger noise background is likely
coming from the improper electric connections and/or contaminants present in the buffer or
working electrode surface, which needs to be checked.
26. Inject a final concentration of 1 mM ferricyanide (about 3–5 mL of stock solution) into the
working chamber, recording the current for 14–16 h (overnight from day 2 to day 3), and a flat
Methods Protoc. 2019, 2, 26 5 of 10

working chamber, using a GL14 open-top cap and o-ring fixed on the counter electrode tube.
Other types of membranes can also be used but the pH gradient cross the membrane and gas
permeability need to be considered.
9. Electric connecting wire: titanium wire of 0.5-mm diameter (Advent, Oxford, England; Cat.
no.: TI5555). NOTE: Titanium is selected for electrode connections, because it is nontoxic to
microorganisms and highly stable against the electric, chemical and biologic stress.
10. Cathode: stainless steel mesh (Membranes international INC., Ringwood, NJ, USA; Cat. no.:
CMI-7000). NOTE: Other electric-conductive materials can also be used, such as platinum, carbon
or graphite. In general, an electrode with low overpotential for hydrogen evolution or oxygen
reduction will be recommended to improve the energy efficiency of the whole reactor. Changing
cathode
Methods material
Protoc. 2019, 2, x will not affect the microbe cultivation in the working chamber. Some other 6 of 9
reactors modified from a conventional stir-tank bioreactor may also be used if available [17,18].
solvent (e.g., ethanol
Those reactors or expensive
are more methanol) but
andcan
sterile
givedistilled watercontrol
more precise respectively.
of such Reference
as the masselectrode
transfer
cannot be sonicated
in the liquid phase, and should
which only
can be be for
used washed gently
further withengineering
process ethanol andandwater, or be maintained
optimization.
following the manufacturer’s instructions.

Figure 2.
Figure 2. Pictures
Picturesof ofthe
theassembled
assembledBES BESreactor.
reactor.(A) top-view;
(A) top-view;(B)(B)
bottom
bottom view of the
view lid; lid;
of the (C) and (D)
(C) and
overview
(D) fromfrom
overview the side. 1, working
the side. electrode
1, working electricelectric
electrode wire; 2,wire;
top-end of the bridging
2, top-end tube; 3, top-end
of the bridging tube; 3,
of the counter
top-end electrode
of the counter compartment;
electrode 4, four-port
compartment; plugplug
4, four-port for for
gassing,
gassing, sampling
samplingandandfeeding;
feeding; 5,
5,
condenser; 6, bottom-end of the counter electrode compartment, sealed by
condenser; 6, bottom-end of the counter electrode compartment, sealed by an ion exchange membrane; an ion exchange
membrane;
7, bottom-end7, bottom-end of the
of the bridging bridging
tube, sealed tube, sealedglass
by porous by porous glass
frit (pore sizefrit
of(pore size of < 1 µ m).
< 1 µm).

3.
3.3.Procedure
BES Cultivation (Time for Completion: x Days, from Day 2 until Experiment Terminated)
Followingthe
14. Transfer theautoclaved
general workflow given
reactor to in Figureclean
a sterilized 1, thebench.
detailed procedure of cultivating wild-type
Pseudomonas putida is presented below. All media are
15. Remove the liquids inside the working chamber and refill 300sterilized by mL
either
ABautoclaving
buffer; or membrane
filtration, and all work
CRITICAL should
STEP: be done intoa check
It is important sterilized
thatenvironment, unless
there is no bubble otherwise
blocking stated.exchange
the cation
membrane (Nr. 6 in the Figure 2 B) while filling the medium. This is the first of the two most
3.1. Pre-Culture Preparation (Time for Completion: 3 Days, from Day 1 to Day 3)
common reasons that will break the electric circuit.
NOTE:
16. Fix Maintain electrode
the reference consistenton
procedures, especially
the bridging tube. considering the incubation times are essential
to minimize batch to
CRITICAL batchCheck
STEP: variations. Differences
the buffer in the metabolic
level inside status
the reference of the precultures
electrode will
and refill with
increase variability in the BES performance.
saturated KCl solution if necessary. This is the second most common reason that will break the
circuit. In addition, some KCl crystals should be visible inside the reference electrode and
bridging tube, to guarantee the saturation during the experiment.
17. Place the counter electrode inside the counter electrode compartment.
18. Open the gassing tube and connect to a sterile syringe membrane filter (0.22 µ m pore size, PTFE).
19. Transfer the reactor to the working bench and place it on the magnetic stir plate, setup the
stirring speed of 400 rpm.
NOTE: Stirring speed can be adjusted but has to be consistent for the all batches, unless mass
Methods Protoc. 2019, 2, 26 6 of 10

1. Steak a Pseudomonas putida cryo stock on LB-agar plate. NOTE: Single colonies should be visible
on the plate after incubation (i) to reduce the heterogeneity of colonies picked for liquid culture
and (ii) to check and avoid the contamination on the plate.
2. Incubate at 30 ◦ C for 24–36 h. NOTE: longer incubation time will cause a longer lag phase for
liquid culture growth.
3. Pick up a single colony and inoculate in a baffled Erlenmeyer shaking flask containing
DM9 medium. CRITICAL STEP: To improve sufficient oxygen supply, cotton stoppers or
membrane vent caps (not silicon-based stoppers) should be used for the gas exchange (see
supplementary Figure S2); moreover, the flasks should be filled with medium of <= 20% (identical
amount for all batches) of the nominal volume of the flask.
4. Incubate the liquid culture overnight (15–16 h);
5. Collect the cell pellets by centrifugation at 30 ◦ C, 7000 g, 10 min;
6. Resuspend the cell pellets in 15 mL AM buffer and suck the solution in a syringe for immediate
injection (Section 3.3) NOTE: Precultures should be timed in a way that cells can be harvested
once the reactors are ready for injection (see below)

3.2. Reactor Assembly (Time for Completion: 2 Days, from Day 1 to Day 2)
NOTE: This step is running simultaneously with step 3.1. Normally, it is to assemble the reactor
after step 1 and then do the autoclave overnight (simultaneously with step 2).

7. Assemble the BES reactor accordingly, as the final picture shows in Figure 2. NOTE: The sampling
tube on the four-port plug should reach close to the stir bar, while the gassing tube should be
above the liquid phase (it is only to gas the headspace). A new working electrode should be used
for each batch.
8. OPTIONAL STEP: plug pH, pO2 or redox sensors into the two spare ports on the lid if necessary.
NOTE: Some autoclavable pH sensors were found to be quite sensitive to the electric field after
autoclaving. The value can be shifted dramatically (up to 1 pH unit), or become unresponsive
after applying the desired potential on the working electrode. This issue can be avoided by
sterilizing the pH sensor with 80% ethanol instead of autoclaving.
9. Fill 50 mL CB buffer into the working chamber. NOTE: For autoclaving purpose. Not AB buffer,
because glucose is not autoclavable together with salt.
10. Fill the counter electrode tube with CB buffer until full.
11. Fill the bridge tube with saturated KCl solution until full.
Methods
12. Protoc.the
Check 2019,
gas2, xtightness of the assembled reactor. 7 of 9

CRITICAL
CRITICALSTEP: STEP:Values chosen
The reactor in step
should be21–22 are critical
gas-tight to preventto minimize
the entryevaporation
of oxygen from of water
the
during operation.
outside atmosphereFor intothetheconditions
reactor and given above, we
to minimize themeasured ~0.09mL water
risk of contamination would
during be lost
operation.
per hour. This
Procedure valuethe
to check is essential for theconnect
gas tightness: quantitative evaluation
the gassing tube(e.g., carbon
to gas line, and
blockelectron balance)
the condenser
of the
and whole
open the system.
sampling tube. The filled-in buffer should be pushed out of the sampling tube if the
23. reactor
Connectisthe
air working electrode,
tight. Otherwise, counter
please electrode
check and reference
the O-rings electrode to the potentiostat.
for each connection.
13. Autoclave the assembled reactor. NOTE: reference electrode andtothe
24. Start the potentiostat and apply a chronoamperometry method thestainless-steel
BES reactor with the
counter
followingshould
electrode parameters:
not beworking
autoclaved.electrode potential
If necessary, (Ewe)cloth
carbon of “0.5andV”stainless-steel
vs Ag/AgCl/KCl sat, current
electrode can be(I)
recording
cleaned byoftwo-step
every dIsonication
of 100 µ Ain and every(e.g.,
solvent dt of ethanol
10 min. or methanol) and sterile distilled water
NOTE: working
respectively. electrode
Reference potential
electrode needsbe
cannot tosonicated
be high enough to efficiently
and should only be oxidize
washedthe mediator,
gently with
which can
ethanol andbewater,
determined by cyclic voltammetry
or be maintained following themethod; 0.5 V is instructions.
manufacturer’s determined for ferricyanide.
This value needs to be adjusted when changing mediators, for instance 0.3 V for [Co(bpy)3]Cl3).
3.3. BES Cultivation
25. Record current(Time for Completion:
for 5–8 x Days,
h for the fresh ABfrom Day 2 until
medium (overExperiment
the day timeTerminated)
of day 2), and a flat
background signal should be achieved before
14. Transfer the autoclaved reactor to a sterilized clean bench. the next step;
NOTE: Quality
15. Remove control
the liquids I: The
inside thecurrent
working recorded
chamberonand therefill
potentiostat
300 mL AB for buffer;
blank AB medium should
be around 0 µ A fluctuated within a maximum of ±10 µ A range. Larger noise background is likely
coming from the improper electric connections and/or contaminants present in the buffer or
working electrode surface, which needs to be checked.
26. Inject a final concentration of 1 mM ferricyanide (about 3–5 mL of stock solution) into the
working chamber, recording the current for 14–16 h (overnight from day 2 to day 3), and a flat
background signal should be achieved before the next step.
Methods Protoc. 2019, 2, 26 7 of 10
Methods Protoc. 2019, 2, x 7 of 9

CRITICAL
CRITICAL STEP: STEP: Values chosen to
It is important incheck
step 21–22that there are critical
is no bubbleto minimize
blockingevaporation
the cation exchange of water
during operation.
membrane (Nr. 6 in Forthethe conditions
Figure 2 B) while givenfillingabove, thewe measured
medium. This~0.09mL water
is the first would
of the twobe lost
most
per hour.reasons
common This value thatiswill
essential
breakfor thethe quantitative
electric circuit. evaluation (e.g., carbon and electron balance)
16. ofProtoc.
Methods
Fix the
thewhole
2019, 2,system.
reference x electrode on the bridging tube. 7 of 9
23. Connect the working electrode, counter electrode and reference electrode to the potentiostat.
24. Start CRITICAL
CRITICAL STEP:
the potentiostat STEP:Values Check
and chosen
applythe abufferin step 21–22
level
chronoamperometry insideare critical
themethod to minimize
reference to electrode
the BESevaporation
and refill
reactor of water
with with
the
during
saturated operation.
following KCl For the
solutionworking
parameters: conditions
if necessary.electrode given above,
Thispotential
is the second we measured
(Ewe) ofmost “0.5 common ~0.09mL water
reason that
V” vs Ag/AgCl/KCl would
sat,will be lost
break
current (I)
per
the hour. This
circuit.
recording valuedIisof
ofInevery
addition, essential
some
100 µ A KClfor
and the quantitative
crystals
every should
dt of 10 be evaluation
min. visible inside(e.g., carbon and electron
the reference electrode balance)
and
of
NOTE:the whole
bridging tube,system.
working toelectrode
guaranteepotential the saturationneeds during
to be high the enough
experiment. to efficiently oxidize the mediator,
23. Place
17. Connect
whichthe canthe beworking
counter electrode,
electrode
determined byinside counter
cyclic electrode
thevoltammetry
counter and reference
electrode
method; 0.5 V electrode
compartment. is determined to thefor potentiostat.
ferricyanide.
24. Start
This the
value potentiostat
needs to be and
adjustedapply whena chronoamperometry
changing mediators,
18. Open the gassing tube and connect to a sterile syringe membrane filter (0.22 µm pore size, method
for to
instance the
0.3 BES
V forreactor
[Co(bpy) with 3]Cl
PTFE).the
3).

19. following
25. Transfer
Record current parameters:
the reactor for to5–8 working
thehworking electrode
for the benchfresh AB potential
and medium (E
place it on(over
we ) of “0.5 V” vs
the day stir
the magnetic Ag/AgCl/KCl
time of day
plate, setup 2),the
sat , current
and (I)
a flat
stirring
recording
background of every
signal dI of
should 100 be µ A and
achieved everybeforedt ofthe10 min.
next
speed of 400 rpm. NOTE: Stirring speed can be adjusted but has to be consistent for the all step;
NOTE: Quality
batches, working
unless mass electrode
control transfer potential
I: The currentneeds
parameters recorded
areto tobeon
behighthe enough
potentiostat
studied. Highto efficiently
for blank
stirring oxidize
speed ABresults the mediator,
medium in should
a high
which
be around
mass can
transfer be
0 µA determined
fluctuated
rate in the system.by cyclic voltammetry method; 0.5 V is
within a maximum of ±10 µ A range. Larger noise background is likely determined for ferricyanide.
This
coming
20. Connect valuetheneeds
from the to be adjusted
improper
double-jacketed when
electric
chamber changing
connections
to mediators,
a recirculating and/orheating for instance
contaminants
water bath 0.3(set
V for
present toin[Co(bpy)
30the◦ C).buffer3]Cl3or
NOTE: ).
25. It
Record
working current
electrode
is important for 5–8
to surface,
check theh for
whichthe fresh
needs to be
temperature AB medium
dropchecked. (over the day time of
while running multiple reactors in parallel. day 2), and a flat
26. Abackground
maximum number (normally 4–6) of reactorsnext
Inject a final signal should
concentration be ofachieved
1 mM before
ferricyanide the step;be
(about
should 3–5determined
mL of stock forsolution)
a given heating into the
NOTE:
working Quality
chamber,
circulator system. control I:
recording The current
the currentrecorded
for on
14–16 the
h potentiostat
(overnight fromfor blank
day 2 AB
to medium
day 3), and should
a flat
be around
background 0 µ A fluctuated
signal should within a maximum of ±10 µ A range. Larger noise background is likely
21. Connect the condenser to abe achieved before
recirculating chillerthe (set next
to 6step.
◦ C).
coming
NOTE: from
Quality the improper
control II: electric
The signal connections
should mostand/or likely contaminants
fluctuate present0–10 in the buffer 7 ofor
Methods
22. Protoc. 2019,
Connect the 2, x
gassing port to the nitrogen gasline (flow rate set up to bebetween20–30 mL/min). µ A, with a9
working electrode
maximum up to 20surface, which needs
µ A. Otherwise, to be checked.
the system needs to be checked.
26.
27. Inject
Inoculate a final
CRITICAL
CRITICAL the BESconcentration
STEP:
STEP:reactor
Values
Values with of 1 mM
the
chosen in ferricyanide
biomass stepfrom 21–223.1, are (about
record
critical
critical 3–5
current
to mL of stock
until
to minimize
minimize solution)
manually
evaporation
evaporation stopped;into
of
of the
water
water
workingoperation.
28. during
OPTIONAL chamber,
STEP:For recording
Check the the
the conditions
the conditions current
dissolved given
givenforabove,
oxygen 14–16
above,level hwe
we (overnight
with from
plugged-in
measured
measured ~0.09mL
~0.09mL day
pO2 2sensor,
to day
water
water 3), and
and
would
would the be
be alost
flat
value
lost
background
should
per
per hour.
hour. beThis
This signal
0. value should be achieved before
value is essential for the quantitative evaluation the next step.
evaluation (e.g., (e.g., carbon
carbon and and electron
electron balance)
balance)
NOTE:
29. of
OPTIONAL Quality
the whole STEP:control
system. II: The signal
Control/monitor theshould
pH value most of likely fluctuate between 0–10 µ A, with a
the system.
maximum
NOTE:
23. Connect
23. theup
While to 20 µ A.
multi-BES
working Otherwise,
electrode,
electrode, arethe
reactorscounter
counter system
controlled
electrode
electrode needs
inandand to
parallel, be checked.
one possible
reference
reference electrode
electrode issue
tothe
to forpotentiostat.
the the pH system
potentiostat.
27. Inoculate
is the the
24. Start
Start the BES
cross-interference
the reactor
potentiostat and with
between
and the biomass
applydifferent from 3.1,
reactors due to the record current
electrical until manually
connection ofstopped;
thewith
systems
24. potentiostat apply aa chronoamperometry
chronoamperometry method
method to the
to the BES reactor
BES reactor with the
the
28. following
OPTIONAL
through the STEP: Check
potentiostat.
parameters: the dissolved
The
working change
electrode of oxygen
applied
potential level
electrode
(E with
) of plugged-in
potential
“0.5 V” vs in pOone2 sensor,
Ag/AgCl/KClreactor andcan, the value
possibly
current (I)
following parameters: working electrode potential (Ewe ) of “0.5 V” vs Ag/AgCl/KClsat , current
we sat
should
affect
recording be
the 0.
pH sensor
of every dI of in another reactor. Solutions to this issue can be galvanically isolating
(I) recording of every dI 100
of 100 µ AµA andand every everydt of dt 10of min.
10 min. NOTE: working electrode potential
29. OPTIONAL
different
NOTE: pH
working STEP:
channels Control/monitor
electrodeor choosing
potential the pH
optical
needs pH
to value of enough
measurements.
be high the system. to efficiently oxidize the mediator,
needs to be high enough to efficiently oxidize the mediator, which can be determined by cyclic
NOTE:
30. voltammetry
Sample While
the multi-BES
reactor once reactors
a day are
(or controlled
more in
frequently) parallel,
along one the possible
batch issue for the pH system
which can bemethod; determined 0.5 Vby cyclic voltammetry
is determined method;This
for ferricyanide. 0.5 Vvalue needs tocollect
and
is determined befor supernatant
ferricyanide.
adjusted when
is
Thisthevalue
and/or cross-interference
cell pellets
needs todepending
be between
adjusted on different
the
when analytic reactors
changing purposes; due to thefor
mediators, electrical
instance connection
0.3 V for of the systems
[Co(bpy) 3]Cl3).
changing mediators, for instance 0.3 V for [Co(bpy)3 ]Cl3 ).
31. through the potentiostat.
Apply quantitative approachesThe change
forfresh of AB
analyzingapplied electrode
genotype, potential
phenotype in
and/oroneof reactor
metabolism. canand possibly
25. Record current for 5–8 h for the fresh AB medium (over the day time of day 2), and aa flat
25. Record current for 5–8 h for the medium (over the day time day 2), flat
affect the
background signalpH sensor
signal should in
shouldbe another
beachieved reactor.
achievedbefore Solutions
beforethe thenext to
nextstep; this issue can be galvanically
step;NOTE: Quality control I: The current isolating
background
different
4. Expected
NOTE: pH channels
Results
Quality control orI: choosing
The current optical
recordedpH measurements.
on the potentiostat for blank AB medium should
recorded on the potentiostat for blank AB medium should be around 0 µA fluctuated within a
30. be
Sample
around the0 µreactor
A once awithin
fluctuated day (or a more frequently)
maximum of ±10 µ Aalong
range. the batchnoise
Larger and background
collect supernatant is likely
The described
maximum of ±10 protocol
µA range. hasLarger
been experimentally
noise background validated
is likely for comingcultivating
from thedifferent
improperP.electric putida
and/or
coming cell
from pellets dependingelectric
the improper on the connections
analytic purposes; and/or contaminants present
strains including
connections type-strains
and/or as well aspresent
contaminants recombinant
in the bufferstrains. or Typical
working raw results
electrode for ainshort-term
surface, the
which buffer BES
needs or
31. Apply
working quantitative
electrode approaches
surface, which for analyzing
needs to be genotype,
checked. phenotype and/or metabolism.
batchtowithout
be checked. pH control using the type-strain Pseudomonas putida F1 are presented in Figure 3.
26.
26. Inject
The a final
increasing
Inject a final concentration
current detected
concentration of 1ofmM 1by mM ferricyanide
potentiostat
ferricyanide (about
for biotic
(about 3–5 mL 3–5
condition mL confirms
of stock ofsolution)
stock solution)
the into the
intoestablishment
the working
4. Expected
working Results
chamber, recording the current for 14–16 h (overnight from day 2 to day 3), and a flat
of electron
chamber, flux from thethe
recording cells to electrode.
current for 14–16Glucose,h (overnight as being
from the day only
2 to dayuseable
3), and carbon
a flat and electron
background
background
Thepresent
source
signal described
should signal
in bethe should
protocol
system,has
achieved be achieved
been
is consumed
before before
theexperimentally
nextby step. the
theNOTE: next
cells, step.
validated
Qualityfor
further cultivating
demonstrated
control different
by
II: The signal theshould P. putida
decrease most of
likely fluctuate between 0–10 µA, with a maximum up to 20 µA. Otherwise, the system needs toa
NOTE:
strains
glucose Quality
including
concentration control
type-strains
in II:
theas The
well
BES signal
as should
recombinant
reactor most
quantified strains. likely
by fluctuate
Typical
HPLC. raw between
results
Accompanying for 0–10
a µ
short-term
the A, with BES
glucose
batchmaximum
without
consumption,
be checked. pHupcontrol
organicto 20acids
µ A.
usingOtherwise,
arethe accumulated the system
type-strain needs
Pseudomonas
in the broth to putida
be
and checked.
F1 are
thus, pH presented
is dropping. in Figure The 3.major
27. Inoculate
The is
product the
increasing BES
2-ketogluconic reactor
current with
detected
acid with theby
thebiomass
potentiostat
carbonfrom from
molar 3.1,
for record
biotic current
condition until
confirmsmanually
aceticthe stopped;
establishment
27. Inoculate the BES reactor with the biomass 3.1,yield
record of > 90%,until
current and manually acid is the only
stopped;
28.
of OPTIONAL
electron
significant flux STEP:
from
by-product. the Check
cells the
to dissolved
electrode. oxygen
Glucose, level
as with
being plugged-in
the only pO
useable 2 sensor,
carbon and
and the value
electron
28. OPTIONAL STEP:No Check significant
the dissolved glucose oxidation
oxygen level withcouldplugged-in
be detected pOif2 sensor,
there isand no the mediator
value
should
source
and/or presentbe
electric 0.
in the system,
power applied. is consumed by the cells, further demonstrated by the decrease of
should be 0.
29. OPTIONAL
glucose concentration
As expected, STEP:
the biomassControl/monitor
in thedensity
BES reactor the pH
is also value of the
quantified by system.
HPLC. Accompanying theobserved.
glucose
29. OPTIONAL STEP: Control/monitor thedecreasing,
pH value and of the no anaerobic
system. growthWhile
NOTE: can be multi-BES
NOTE:
consumption,
The sugar While
organic
consumption multi-BES
acids reactors
are are
accumulated
rateinisparallel,
much lower controlled
in the
(only < issue in parallel,
broth
5%) compared and one possible
thus, pH issue
is for
dropping. the pH
The system
major
reactors are controlled one possible for the pHtosystem the value is the reported for aerobic
cross-interference
is the
cultivations. cross-interference
product is 2-ketogluconic
Moreover, theacid between different
with indicate
products the carbon reactors
the molar due
carbonyield to the
metabolic electrical
of > 90%,activity connection
andisacetic
mainlyacid of the systems
is the only
constrained to
the through
significant
periplasmic the space,
by-product.potentiostat.
No
while The
the change
significant of applied
glucose
cytoplasmic oxidation
central electrode
couldpotential
metabolism be detected in one
is only reactor
ifpartially
there canmediator
is activated
no possibly (to
affect
and/or
produce the power
electric
acetic pH
acid). sensor
applied.in another
Nevertheless, reactor. Solutions
a complete anaerobictoconsumptionthis issue can of be
sugargalvanically
is achieved isolating
for an
different pH channels or choosing optical pH measurements.
As expected, the biomass density is also decreasing, and no anaerobic growth can be observed.
30. Sample
The sugar the reactor rate
consumption onceisamuch day (or lowermore (onlyfrequently)
< 5%) compared along the to batch
the value andreported
collect supernatant
for aerobic
and/or cell
cultivations. pellets depending
Moreover, the products on indicate
the analytic the purposes;
carbon metabolic activity is mainly constrained to
Methods Protoc. 2019, 2, 26 8 of 10

between different reactors due to the electrical connection of the systems through the potentiostat.
The change of applied electrode potential in one reactor can possibly affect the pH sensor in
another reactor. Solutions to this issue can be galvanically isolating different pH channels or
choosing optical pH measurements.
30. Sample the reactor once a day (or more frequently) along the batch and collect supernatant
and/or cell pellets depending on the analytic purposes;
31. Apply quantitative approaches for analyzing genotype, phenotype and/or metabolism.

4. Expected Results
Methods Protoc. 2019, 2, x 8 of 9
The described protocol has been experimentally validated for cultivating different P. putida strains
obligate aerobe,
including providing
type-strains as well aassystem to quantitatively
recombinant investigate
strains. Typical and
raw results for identify the BES
a short-term metabolic
batch
constraints
without pH limiting the anaerobic
control using metabolism
the type-strain of P. putida
Pseudomonas using
putida systems
F1 are biology
presented approaches
in Figure 3.

Figure 3. Typical
Figure 3. Typicalraw
raw results
resultsofof the
the short-batch
short-batch BES BES cultivation of Pseudomonas
cultivation of Pseudomonas putida
putidaF1F1 with
with glucose
glucose
as
as the sole carbon source (averaged from triple reactors). Ferricyanide of 1mM is used as the mediator
the sole carbon source (averaged from triple reactors). Ferricyanide of 1mM is used as the mediator
and
and the
the working
working electrode
electrodepotential
potentialisissetsetatat0.5
0.5V Vvsvs
Ag/AgCl/KCl
Ag/AgCl/KClsat . Right Y axis indicates the
sat. Right Y axis indicates the
concentrations of different compounds in the broth quantified by
concentrations of different compounds in the broth quantified by HPLC. They HPLC. They needneed
to beto
adjusted with
be adjusted
awith
concentrating factor because of the water loss caused by gas flushing (0.09 mL/h)
a concentrating factor because of the water loss caused by gas flushing (0.09 mL/h) while while conducting
carbon and electron
conducting carbon andbalance (bothbalance
electron balanced for balanced
(both the presented data)
for the based on
presented those
data) data.
based on those data.

The increasing current detected by potentiostat for biotic condition confirms the establishment
Supplementary Materials: The following files are available online at www.mdpi.com/xxx/s1, Figure S1: Cyclic
of electron flux from the cells to electrode. Glucose, as being the only useable carbon and electron
voltammetry characterization of carbon cloths with different pre-treatment methods, Figure S2: Growth
source present in the system, is consumed by the cells, further demonstrated by the decrease of glucose
property of Pseudomonas putida F1 in shake flasks with different gas-permeable stoppers in minimum mineral
concentration
medium. in the BES reactor quantified by HPLC. Accompanying the glucose consumption, organic
acids are accumulated in the broth and thus, pH is dropping. The major product is 2-ketogluconic acid
Author Contributions: B.L. and J.O.K. designed and developed the protocol. B.L. and A.V.N. prepared the
with the carbon molar yield of > 90%, and acetic acid is the only significant by-product. No significant
manuscript draft, and J.O.K. reviewed and revised the paper
glucose oxidation could be detected if there is no mediator and/or electric power applied.
Funding: This research
As expected, the received
biomassno external
density isfunding
also decreasing, and no anaerobic growth can be observed.
The sugar consumption rate is much lower (only
Conflicts of Interest: The authors declare no conflict of <interest.
5%) compared to the value reported for aerobic
cultivations. Moreover, the products indicate the carbon metabolic activity is mainly constrained to
the periplasmic space, while the cytoplasmic central metabolism is only partially activated (to produce
References
acetic acid). Nevertheless, a complete anaerobic consumption of sugar is achieved for an obligate
1. Nikel, P.I.; de Lorenzo, V. Pseudomonas putida as a functional chassis for industrial biocatalysis: From native
aerobe, providing a system to quantitatively investigate and identify the metabolic constraints limiting
biochemistry to trans-metabolism. Metab. Eng. 2018, 50, 142–155.
the anaerobic metabolism of P. putida using systems biology approaches
2. Wackett, L.P. Pseudomonas putida--a versatile biocatalyst. Nat. Biotechnol. 2003, 21, 136–138.
1. Nikel, P.I.; Chavarría, M.; Danchin, A.; de Lorenzo, V. From dirt to industrial applications: Pseudomonas
putida as a synthetic biology chassis for hosting harsh biochemical reactions. Curr. Opin. Chem. Biol. 2016,
34, 20–29.
4. Loeschcke, A.; Thies, S. Pseudomonas putida—A versatile host for the production of natural products. Appl.
Microbiol. Biotechnol. 2015, 99, 6197–6214.
Methods Protoc. 2019, 2, 26 9 of 10

Supplementary Materials: The following files are available online at http://www.mdpi.com/2409-9279/2/2/

26/s1, Figure S1: Cyclic voltammetry characterization of carbon cloths with different pre-treatment methods,
Figure S2: Growth property of Pseudomonas putida F1 in shake flasks with different gas-permeable stoppers in
minimum mineral medium.
Author Contributions: B.L. and J.O.K. designed and developed the protocol. B.L. and A.V.N. prepared the
manuscript draft, and J.O.K. reviewed and revised the paper
Funding: This research received no external funding.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Nikel, P.I.; de Lorenzo, V. Pseudomonas putida as a functional chassis for industrial biocatalysis: From native
biochemistry to trans-metabolism. Metab. Eng. 2018, 50, 142–155. [CrossRef] [PubMed]
2. Wackett, L.P. Pseudomonas putida—A versatile biocatalyst. Nat. Biotechnol. 2003, 21, 136–138. [CrossRef]
[PubMed]
3. Nikel, P.I.; Chavarría, M.; Danchin, A.; de Lorenzo, V. From dirt to industrial applications: Pseudomonas
putida as a synthetic biology chassis for hosting harsh biochemical reactions. Curr. Opin. Chem. Biol. 2016, 34,
20–29. [CrossRef] [PubMed]
4. Loeschcke, A.; Thies, S. Pseudomonas putida—A versatile host for the production of natural products.
Appl. Microbiol. Biotechnol. 2015, 99, 6197–6214. [CrossRef] [PubMed]
5. Rich, P.R. The molecular machinery of keilin’s respiratory chain. Biochem. Soc. Trans. 2003, 31, 1095–1105.
[CrossRef] [PubMed]
6. Chen, X.; Alonso, A.P.; Allen, D.K.; Reed, J.L.; Shachar-Hill, Y. Synergy between 13 C-metabolic flux analysis
and flux balance analysis for understanding metabolic adaptation to anaerobiosis in e. Coli. Metab. Eng. 2011,
13, 38–48. [CrossRef] [PubMed]
7. Tran, Q.H.; Unden, G. Changes in the proton potential and the cellular energetics of escherichia coli during
growth by aerobic and anaerobic respiration or by fermentation. Eur. J. Biochem. 1998, 251, 538–543.
[CrossRef] [PubMed]
8. Noor, E.; Eden, E.; Milo, R.; Alon, U. Central carbon metabolism as a minimal biochemical walk between
precursors for biomass and energy. Mol. Cell 2010, 39, 809–820. [CrossRef] [PubMed]
9. Nikel, P.I.; de Lorenzo, V. Engineering an anaerobic metabolic regime in pseudomonas putida kt2440 for the
anoxic biodegradation of 1,3-dichloroprop-1-ene. Metab. Eng. 2013, 15, 98–112. [CrossRef] [PubMed]
10. Steen, A.; Ütkür, F.Ö.; Borrero-de Acuña, J.M.; Bunk, B.; Roselius, L.; Bühler, B.; Jahn, D.; Schobert, M.
Construction and characterization of nitrate and nitrite respiring pseudomonas putida kt2440 strains for anoxic
biotechnical applications. J. Biotechnol. 2013, 163, 155–165. [CrossRef] [PubMed]
11. Schröder, U.; Harnisch, F.; Angenent, L.T. Microbial electrochemistry and technology: Terminology and
classification. Energy Environ. Sci. 2015, 8, 513–519. [CrossRef]
12. Park, D.H.; Zeikus, J.G. Impact of electrode composition on electricity generation in a single-compartment
fuel cell using shewanella putrefaciens. Appl. Microbiol. Biotechnol. 2002, 59, 58–61. [PubMed]
13. Reguera, G.; McCarthy, K.D.; Mehta, T.; Nicoll, J.S.; Tuominen, M.T.; Lovley, D.R. Extracellular electron
transfer via microbial nanowires. Nature 2005, 435, 1098–1101. [CrossRef] [PubMed]
14. Rabaey, K.; Boon, N.; Hofte, M.; Verstraete, W. Microbial phenazine production enhances electron transfer in
biofuel cells. Environ. Sci. Technol. 2005, 39, 3401–3408. [CrossRef] [PubMed]
15. Light, S.H.; Su, L.; Rivera-Lugo, R.; Cornejo, J.A.; Louie, A.; Iavarone, A.T.; Ajo-Franklin, C.M.; Portnoy, D.A.
A flavin-based extracellular electron transfer mechanism in diverse gram-positive bacteria. Nature 2018, 562,
140–144. [CrossRef] [PubMed]
16. Lai, B.; Yu, S.; Bernhardt, P.V.; Rabaey, K.; Virdis, B.; Krömer, J.O. Anoxic metabolism and biochemical
production in pseudomonas putida f1 driven by a bioelectrochemical system. Biotechnol. Biofuels 2016, 9, 39.
[CrossRef] [PubMed]
Methods Protoc. 2019, 2, 26 10 of 10

17. Hintermayer, S.; Yu, S.; Krömer, J.O.; Weuster-Botz, D. Anodic respiration of pseudomonas putida kt2440 in a
stirred-tank bioreactor. Biochem. Eng. J. 2016, 115, 1–13. [CrossRef]
18. Rosa, L.F.M.; Hunger, S.; Gimkiewicz, C.; Zehnsdorf, A.; Harnisch, F. Paving the way for bioelectrotechnology:
Integrating electrochemistry into bioreactors. Eng. Life Sci. 2017, 17, 77–85. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).