Accepted Manuscript

Title: An enzyme-free hydrogen peroxide sensor for

evaluation of probiotic potential of Enterococcus faecium

Authors: Mourad Braik, Lucian-Gabriel Zamfir, Lucian

Rotariu, Carmen Curutiu, Mariana Carmen Chifiriuc, Mounir
Ben Ali, Camelia Bala

PII: S0925-4005(18)31161-4
DOI: https://doi.org/10.1016/j.snb.2018.06.057
Reference: SNB 24894

To appear in: Sensors and Actuators B

Received date: 15-2-2018

Revised date: 7-6-2018
Accepted date: 12-6-2018

Please cite this article as: Braik M, Zamfir L-Gabriel, Rotariu L, Curutiu C, Chifiriuc
MC, Ali MB, Bala C, An enzyme-free hydrogen peroxide sensor for evaluation of
probiotic potential of Enterococcus faecium, Sensors and Actuators: B. Chemical
(2018), https://doi.org/10.1016/j.snb.2018.06.057

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
 An enzyme-free hydrogen peroxide sensor for evaluation of probiotic potential of
Enterococcus faecium

Mourad Braika,b, Lucian-Gabriel Zamfira,c, Lucian Rotariua,d,*, Carmen Curutiue, Mariana Carmen
Chifiriuce, Mounir Ben Alib,f, Camelia Balaa,d *

T
IP
a
LaborQ, University of Bucharest, 4-12 Regina Elisabeta Blvd., 030018 Bucharest, Romania

R
b
Higher Institute of Applied Sciences and Technology of Sousse, University of Sousse, Sousse, Tunisia;

SC
c
ICUB, University of Bucharest, 36-46 B-dul M. Kogalniceanu, 050107 Bucharest, Romania
d
Department of Analytical Chemistry, University of Bucharest, 4-12 Regina Elisabeta Blvd., 030018
Bucharest, Romania
U
N
e
Department of Botanics and Microbiology, Faculty of Biology, University of Bucharest, 1–3 Aleea
A
Portocalelor Str., 60101 Bucharest, Romania
f
Nanomisene Lab, LR16CRMN01, Center for Research on Microelectronics and Nanotechnology of
M

Sousse, Sousse, Tunisia

ED
PT

*corresponding author: Department of Analytical Chemistry, University of Bucharest, 4-12 Regina

Elisabeta Blvd., 030018 Bucharest, Romania tel./fax: +40 21 4104888:

E

email: camelia.bala@chimie.unibuc.ro
CC
A

1
Graphical abstract.

T
R IP
SC
Highlights

U
N
 simple and effective way to prepare composite based on GO/ERGO and [BMIM][BF4]
A
 enhanced electrochemical response towards hydrogen peroxide (1484 µA·mM-1·cm2)
M

 sensitive and selective determination of H2O2 in culture media of E.faecium

ED

Abstract
E PT

A novel strategy to realize a non-enzymatic hydrogen peroxide (H2O2) sensor based on reduced graphene
CC

oxide dispersed in ionic liquid for monitoring the exogenous production of H2O2 by Enterococcus
faecium is presented in this work. The electrodes functionalization steps were studied by electrochemical
impedance spectroscopy and cyclic voltammetry. FT-IR spectroscopy and scanning electron microscopy
A

were used to characterize the electrode materials. The amperometric measurements on reduced graphene
oxide-ionic liquid modified screen-printed electrode showed that the modified electrodes can be used to
detect H2O2 with a high sensitivity of 1,484±25 µA·mM-1·cm-2 and a low detection limit of 0.1 μM (for a
signal-to-noise ratio of 3). In addition, the electrodes exhibited an excellent reproducibility and long-term
stability as well as negligible interference from ascorbic acid, citric acid, urea and creatinine. The sensor
was successful in monitoring the H2O2 production from Enterococcus faecium 2980 clinical strain
2
cultivated in anaerobic and aerobic conditions. The good correlation of the proposed sensor with
spectrometric method indicates its very good potential in clinical and biochemical applications.

Keywords:

hydrogen peroxide, sensor, reduced graphene oxide, ionic liquid, probiotics

T
1. Introduction

R IP
SC
Hydrogen peroxide (H2O2) is an important compound in the metabolism of the living organisms [1]. It is
the product of many biochemical reactions catalyzed by oxidases enzymes. Being a very reactive species
and a precursor of other harmful oxygen species, hydrogen peroxide should be carefully monitored. A

U
higher level of hydrogen peroxide in biological fluids is associated with cells oxidative stress and its
N
determination would be useful in diagnosis of some types of cancer, inflammation or other diseases [2].
A
In some cases, the exogenous production of H2O2 is useful being one of the defense mechanisms of the
probiotics microorganisms against pathogens. There are probiotics species that are naturally present in the
M

human body. In the case of some deficiencies or use of an antibiotics treatment, supplement probiotics are
often used.
ED

The gastrointestinal tract is the host of some enterococci, among them the most common being
Enterococcus faecalis and Enterococcus faecium. What differentiates E. faecium is the ability to produce
PT

exogenous H2O2 at concentration levels that cause cellular damage. Literature reports that E. faecium
mediates the killing of some other bacterial pathogens or nematodes [3]. The use of glycerol, sorbitol or
E

mannitol in the cultivation media can increase the amount of the H2O2 produced by these bacteria [4].
CC

Fast and sensitive detection of H2O2 produced by probiotic strains is useful for studies and selection of
these microorganisms. Electrochemical sensors fulfil these requirements. Moreover, they are non-
destructive analytical tools that use small amounts of sample. Complexity of the biological samples and
A

culture media of microorganisms requires also selective electrochemical detection systems. The use of
bare classical electrodes for direct electrochemical oxidation or reduction of H2O2 is not recommended
because it requires high overpotentials [5]. Different approaches have been reported to overcome this
problem: one is referring to the chemical modification of electrodes surface and the other to biosensors
with horseradish peroxidase [6], myoglobin [7] or hemoglobin [8]. Although the use of biomolecules
leads to an improvement of the selectivity, some major drawbacks come upon, such as the difficulty of
3
the immobilization process, the relatively low stability of the bio-component and poor reproducibility.
Therefore, non-enzymatic sensors based on chemically modified electrodes have been further developed.
The most reported materials used for construction of H2O2 sensors, with high catalytic activity and good
stability are either metal complexes such as Prussian blue [9, 10], cobalt phthalocyanine [11],
metalloporphyrins [12] or metal nanoparticles [13, 14]. Moreover, different carbon-based materials such
graphene has attracted considerable attention in recent years, due to its unique features [15, 16]. The high
surface area of graphene has an important role in the electrochemical reactions by enhancing of the
electron transfer, which leads to a significant improvement of the electrochemical performances
compared to other carbon-based materials [17, 18]. All these advantages promote graphene to be a very

T
attractive material for the design of high sensitive and selective sensors [10, 19] including detection of

IP
H2O2 [20], which have been recently reviewed [21]. However, the dispersion of graphene materials
remains the main issue for preparing the graphene-based composites for electrochemical applications.

R
Ionic liquids represent a promising modifier to design composite materials based on graphene for sensing

SC
applications [22]. There are works about electrochemical systems based on ionic liquids for reduction of
hydrogen peroxide (H2O2). Li et al. [23] reported electro-polymerization of Prussian blue (PB) on a

U
carbon ionic liquid electrode, while Wang et al. [24] electro-polymerized aniline and single-walled carbon
nanotubes on a platinum electrode in a room temperature ionic-liquid. Composite materials based on
N
dispersion of graphene oxide in ionic liquids have been reported to further improve the sensor
A
performances. They combine the advantages of the counterparts and were lately used for developing a
M

variety of electrochemical sensors and biosensors [10, 19]. Zhang et al. [10] demonstrated that the direct
dispersion of ILs in an aqueous solution of GO could be carried out. Chemical reduction of GO to RGO is
ED

time and reagent consuming [25]. An alternative to the chemical reduction is the on-site reduction of GO
to electrochemical reduced graphene oxide (ERGO) by cyclic voltammetry or amperometry [22]. The
advantage of this method to be very simple, fast and green was exploited in our work.
PT

Herein, we are reporting a novel sensor with ERGO and 1-butyl-3-methylimidazolium tetrafluoroborate
[BMIM][BF4] ionic liquid composite material for sensitive detection of H2O2 in culture media of E.
E

faecium 2980 clinical strain.

CC

2. Materials and methods

2.1. Reagents
A

Graphene oxide (GO) was purchased from DropSens (ref. GPHOX, 0.5-50 µm). The ionic liquid 1-butyl-
3-methylimidazolium tetrafluoroborate [BMIM][BF4], ethanol absolute, ≥99.8% and brain heart infusion
broth for microbiology (BHI broth) were purchased from Sigma–Aldrich. Nafion (5%) and H2O2 solution
(22%) (concentration determined by permanganometric titration) were obtained from Fluka. All other
chemicals were of analytical grade. All solutions were prepared in ultrapure water (Millipore, 18 MΩ

4
cm). Phosphate buffer 0.1 M, containing 0.1 M KCl, pH 7.0 was prepared by mixing solutions of
Na2HPO4, NaH2PO4 and KCl.

2.2. Equipments and materials

Cyclic voltammetry and amperometric measurements were carried out using an Autolab PGSTAT101,
while electrochemical impedance spectroscopy (EIS) experiments were performed with an Autolab
PGSTAT12 with FRA2 module (Eco Chemie, Utrecht, Netherlands). The screen–printed electrodes
(SPEs) with carbon working electrode (4 mm in diameter), silver pseudo-reference electrode and carbon
counter electrode (Dropsens DRP-C110) were used for electrochemical studies. All potentials were

T
reported vs. pseudo-reference silver electrode. The FT-IR spectra were recorded using a Nicolet 4700

IP
FTIR spectrophotometer (Thermo Scientific). The surface morphology of the modified electrodes was
studied using a field emission scanning electron microscope (FESEM) combined with a powerful focused

R
ion beam (FIB) model Auriga produced by Carl Zeiss Instruments. Accelerating voltage was 2 kV with a

SC
working distance of 4-5 mm.

2.3. Preparation of the modified electrodes

U
Before any modification, the screen-printed carbon electrodes (SPCEs) were electrochemically pretreated
N
by application of a potential of +1200 mV in 0.1 M sulfuric acid, for 2 min. A stable dispersion of GO in
A
ionic liquid without using any surfactants or/and stabilizers was successfully prepared by sonication.
Three different suspensions were prepared by sonicating 1 mg or 2 mg of GO with 20 µL [BMIM][BF4],
M

or 1 mg GO with 40 µL [BMIM][BF4]. A volume of 5µL from each suspension was mixed separately
with 10µL of 0.5 % Nafion (NAF) in ethanol. Nafion was used as a film-forming polymer to immobilize
ED

the GO-[BMIM][BF4] on the surface of the SPCE and to obtain a stable modified electrode [26]. A
volume of 3 µL from the obtained mixtures was dropped on the working electrodes and left overnight to
PT

dry at room temperature. The ERGO-[BMIM][BF4] modified SPCEs were prepared by electrochemical
reduction of GO directly on the electrode surface, as it was previously reported (cyclic voltammetry in
oxygen-free KCl 0.1 M, potential range between -1.7 V and 0.5 V, ν=10 mVs-1, 10 cycles) [27]. The
E

principle of ERGO-[BMIM][BF4]-NAF/SPCE sensors for detection of H2O2 is presented in Figure 1.

CC

Figure 1.
A

2.4. Cultivation of Enterococcus faecium 2980 clinical strain

Enterococcus faecium clinical strain isolated from a urine sample (E. faecium 2980) was used in this
study. The bacterial strain was cultivated in two different culture media: BHI broth and BHI broth + 0.2%
glycerol. The BHI broth was prepared by dissolving 37g of the BHI powder in 1 L of ultrapure water. The
culture medium was sterilized by autoclaving at 121°C for 15 minutes. The BHI culture medium with
glycerol was obtained by adding of 2 mL of sterile glycerol to 1 L of autoclaved BHI medium previously
5
brought to the room temperature. A volume of 300 mL of sterile BHI broth was added in two glass flasks,
while in another two flasks 300 mL of sterile BHI broth with glycerol were added. All four flasks were
inoculated with 30 mL of bacterial suspension corresponding to 1-3x108 CFU/mL (0.5 McFarland). The
bacterial suspensions were obtained from an overnight culture of the Enterococcus faecium 2980 strain
grown on nutritive agar. Two of the inoculated flasks (one with glycerol and another without) were
further incubated in aerobiosis, while the other two in anaerobiosis at 37oC. The anaerobiosis conditions
were achieved by adding a layer of paraffin oil (10 mL) over the culture medium. Over the incubation
period, the cultures were harvested at different time intervals by centrifugation at 6000 rpm for 10 min to
remove the bacterial cells and debris, and filtered through 0.22 µm filter to obtain a sterile supernatant.

T
The supernatant (approx. 10 mL) was collected in sterile flasks and the samples were further processed

IP
for the quantification of H2O2.

R
2.5. Enzymatic-spectrometric method for the detection of hydrogen peroxide

SC
The enzymatic-spectrometric method was used as reference method for detection of H2O2 and it was
described elsewhere [28]. Briefly, the method consists in oxidation of 2,2′-azino-di (3-ethyl

U
benzthiazoline-6-sulphonic acid) (ABTS) by the H2O2 in the presence of horseradish peroxidase. The
oxidized ABTS was monitored at 405 nm and the absorbance was correlated with the concentration of
N
H2O2 from the culture medium sample.
A
3. Results and discussion
M

3. FT-IR and SEM characterization

ED

The FTIR spectra were recorded for the GO, ERGO and [BMIM][BF4] precursors and for the composite
materials GO-[BMIM][BF4] and ERGO-[BMIM][BF4] and are shown in the Figure 2. Various oxygen
configurations were identified in the spectrum of GO (spectrum a), as follow: O-H stretching vibrations at
PT

3415 cm-1, C-O bending at 1060 cm-1, C-O stretching at 1226 cm-1 and C=O at 1612 cm-1. It can be
noticed a decrease of the peaks intensity from 1612 and 1226 cm-1 in the case of ERGO spectrum (b)
E

indicating the reduction of the oxygen-based functionalities [29]. The [BMIM][BF4] sample exhibited the
CC

following characteristic features (spectrum c): aliphatic asymmetric and symmetric (C–H) stretching
vibrations of the imidazolium ring appears at 3160 cm-1 and 2875 cm-1 respectively, C=C stretching
vibration and imidazole ring stretching vibration is located at 1572 cm -1. The band from 1164 cm-1 can be
A

attributed to the C–N stretching vibration of the imidazolium ring. In the GO-[BMIM][BF4] composite
spectrum (d) a slight shift toward lower wavenumbers was observed for the specific bands of the ionic
liquid, which were located at 1163, 1569, and 2871 cm-1. In the case of ERGO-[BMIM][BF4] spectrum
(e) the bands are located at 1166, 1568, and 2871 cm-1. That indicates that the  system of the
imidazolium ring interacts with the graphene by cation -  or π-π stacking [30].

6
 Figure 2.

The morphology of the surface of the SPCEs modified with GO-NAF, ERGO-NAF, GO-[BMIM][BF4]-
NAF and ERGO-[BMIM][BF4]-NAF was examined by FESEM and the recorded images are presented in
Figure 3. One can see distinct flakes in figure 3A and B, indicating the presence of GO and ERGO
respectively. The images of the composite materials that include the ionic liquid (figure 3 C and D) show
a more uniform surface topography. The interaction between [BMIM][BF4] and GO/ERGO leads to a
multilayer graphene structures. The graphene sheets are wrapped and stabilized in ionic liquid-Nafion
film. Figure 3C shows graphene layers orientated with a tilt angle on the surface of the electrode. That

T
can be explained by the presence of the butyl groups of the ionic liquid, which due to their hydrophobicity
tend to be adsorbed onto the electrode surface [31]. The charged moieties of the ionic liquid (imidazole

IP
rings) reside away from the electrode surface and interact with the negatively charged moieties of the GO.

R
Figure 3.

SC
3.2. Cyclic voltammetry study

U
The chemical modifications of the working electrodes surface were studied by cyclic voltammetry (CV)
N
in the potential range of −1000 to +800 mV at a sweep rate of 100 mV/s. There were prepared five types
of modified SPC electrodes. GO-NAF/SPCE, [BMIM][BF4]-NAF/SPCE, GO-[BMIM][BF4]-NAF/SPCE
A
were prepared by deposition of the appropriate mixture on the electrode surface. The on-site
M

electrochemical reduction of GO to ERGO generates the following electrodes: ERGO-NAF/SPCE and

ERGO-[BMIM][BF4]-NAF/SPCE. The electrochemical reduction of the oxygen functionalities of GO
ED

was performed at negative potentials [22, 32-34]. The electrochemical reduction process was carried on
by cyclic voltammetry and the data are shown in Figure S1 (Electronic Supplementary Material). One can
see that a reduction peak appears at a potential of -1200 mV and decrease considerably with each cycle.
PT

Reproducible voltammograms were obtained after 10 cycles, indicating that GO was completely reduced
to ERGO. The CV of the GO-[BMIM][BF4]-NAF/SPCE and ERGO-[BMIM][BF4]-NAF/SPCE modified
E

electrodes recorded in buffer and 10 mM H2O2 are shown in Figure 4 A. No reduction peak was observed
CC

in buffer solution (curves a and c). Both sensors present a reduction peaks of H 2O2 at -550 mV (curves b
and d) with a higher current in the case of ERGO-[BMIM][BF4]-NAF/SPCE (curve d). This can be
A

explained by the increase of the conductivity of the composite materials after electrochemical reduction
of the GO and it is correlated with the improvement of the charge transfer reported by impedance
measurements.
Figure. 4.
In order to study the effect of the scan rate on the reduction peak current of H 2O2, the current variation
was determined as function of the scan rate in the range from 5 to 300 mV s −1. The voltammograms are

7
presented in Figure 4 B. A linear relationship was obtained by plotting the peak current versus the square
root of the scan rate (inset Figure 4 B). That indicates an electrochemical reduction process controlled by
diffusion.

3.3. Electrochemical impedance spectroscopy

The EIS measurements were carried on in order to determine the ability of the electrode materials to
transfer and exchange charge. Such electron exchange ability is strongly influenced by the electronic
conductivity of the composite material. The modified electrodes were designed to detect H2O2, a neutral
molecule and therefore a non-charged redox couple, benzoquinone/hydroquinone was selected to carry on

T
the impedance measurements. Figure 5 shows the Nyquist diagrams for EIS measurements of SPCE bare,

IP
[BMIM][BF4]-NAF/SPCE, GO-[BMIM][BF4]/SPCE, ERGO-[BMIM][BF4]/SPCE, GO-[BMIM][BF4]-
NAF/SPCE and ERGO-[BMIM][BF4]-NAF/SPCE modified electrodes in 5 mM benzoquinone

R
(BQ)/hydroquinone (HQ) redox couple across the frequency range from 105 to 0.1 Hz. The equivalent

SC
electrical circuit used to fit the experimental data consists in an active electrolyte resistance (RS) in series
with the parallel combination of the constant phase element (CPE) and the resistance of the charge

U
transfer (Rct). The fitting parameters for all electrodes are presented in the Electronic Supplementary
Material, Table S1. The highest resistances of the charge transfer (Rct) was obtained for the bare SPCE
N
(17.156±0.087 kΩ) and GO-NAF/SPCE (1.378±0.042 kΩ). The large amount of oxygen-containing
A
functional groups on the surface of the GO affects the electronic conductivity of the GO-NAF composite
and can explain the value of the Rct. A decreasing of the Rct to 145±12 Ω was observed in the case of GO-
M

[BMIM][BF4]-NAF/SPCE, which can be attributed to the high conductivity of the [BMIM][BF4]. The
Nyquist plot of [BMIM][BF4]-NAF/SPCE (graph d) reveals an even lower resistance to the charge
ED

transfer in the presence of BQ/HQ redox couple (Rct = 109.28 Ω). These results clearly show that GO
based composite has a lower conductivity and does not enhance the electron transfer at the surface of the
PT

modified electrodes.

Figure 5.
E

The electrochemical reduction of the GO leads to an important improvement of the conductivity as it can
CC

be seen in the Figure 5 for ERGO-NAF/SPCE (graph c) and ERGO-[BMIM][BF4]-NAF/SPCE (graph f).
The Rct decrease from 1303±36 Ω in the absence of the ionic liquid to 91±9 Ω in the presence of
A

[BMIM][BF4]. Moreover, it can be clearly seen that, among the all modified electrodes, the lowest Rct
was obtained for the ERGO-[BMIM][BF4]-NAF/SPCE sensor. That can be explained by the large active
surface area and good electrical conductivity of the ERGO combined with the good ionic conductivity of
the [BMIM][BF4]. The advantages of the counterparts recommend the ERGO-[BMIM][BF4]-NAF
composite material to be used as electrode material for detection of H2O2.

8
3.4. Chronoamperometry

The chronoamperometry technique is employed for the investigation of electrode processes at the surface
of the chemically modified electrodes. In this work, the chronoamperometry was used to estimate the
diffusion coefficient (D) of hydrogen peroxide at the surface of the ERGO-[BMIM][BF4]-NAF/SPCE
and the catalytic rate constant. Figure 6 shows the chronoamperograms of the ERGO-[BMIM][BF4]-
NAF/SPCE modified electrode recorded in PBS solution and 0.5, 1 and 2.5 mM H 2O2 at an applied
potential of -550 mV. The diffusion coefficient value was estimated according to Cottrell equation [35,
36]:
1

T
n FAD 2 C
i 1 1 eq. 1

IP
π 2
t 2

where n is the electron number, A is the surface area of the working electrode, F is the Faraday constant,

R
C represents the H2O2 concentrations and t is the time in second.

SC
Figure 6.

The slopes of the representation of i versus t-1/2 plots for different concentrations of hydrogen peroxide
U
were used to estimate the diffusion coefficient as 2.27 10-5 cm2 s-1. This is approximately one order of
N
magnitude higher than other reports from the literature [12, 37]. For instance a diffusion coefficient for
A
H2O2 of 2.34 10-6 cm2 s-1 was calculated for a glassy carbon electrode modified with MWCNTs and
reactive blue 19 [12], while a value of 6.70 10-6 cm2 s-1 was reported for a Pt electrode modified with
M

MWCNTs and Pt nanoparticles [37]. Determination of the catalytic rate constant (kcat) was also realized
from the chronoamperometric measurements. The catalytic current (Icat) is dominated by the rate of
ED

electrocatalyzed reduction of hydrogen peroxide and the rate constant for the chemical reaction between
hydrogen peroxide and the surface of the modified electrode is determined according to the method
PT

described in the literature [38].

k cat C t  2
1 1
i cat i buffer  π 2
eq. 2
E

where icat and ibuffer are the currents in the presence and in the absence of H2O2.
CC

The plot icat/ibuffer vs. t1/2 for H2O2 concentration of 2.5 mM showed a linear dependence (inset Figure 6),
from where the value of kcat was calculated to be 2.06 102 L·mol-1 s-1. This value is much higher
A

compared to a SPE modified with polyamic acid (0.293 L·mol-1·s-1) [39] or graphene quantum dots-
chitosan/methylene blue modified GCE (44.5 L·mol-1·s-1) [40] indicating a faster reaction at the surface
of the one reported in this work.

3.5. Amperometric sensing of H2O2

9
Amperometric measurements were carried on in an electrochemical cell of 5 mL in the optimal
experimental conditions mentioned above, at an applied potential of -550 mV. The ratio between the
counterparts of the ERGO-[BMIM][BF4] composite material can affect the H2O2 sensor performances. In
order to optimize the amount of ERGO, there were prepared three modified ERGO-[BMIM][BF4]-
NAF/SPCE electrodes that contain 2, 4 and 7.6% ERGO and the electrochemical behavior was studied by
amperometry. All the measurements were performed in triplicates. The sensor based on 2% ERGO has an
extended linear range from 5 to 130 µM H2O2 and a sensitivity of 462±10 µA·mM-1·cm2. The electrode
prepared with 4% ERGO presented the highest sensitivity of 1,484±25 µA·mM-1·cm2 in the range 1-150
µM H2O2. One can observe an increase of the sensitivity and a low background current allowing us to

T
calculate a detection limit of 0.1 µM. A higher amount of ERGO (7.6%) did not improve the

IP
performances of the sensor. On the contrary, the sensitivity was lower of 1,006±21 µA·mM-1·cm2 on a
narrow linear range 40-120 µM H2O2. Another linear range was identified for this electrode between 7

R
and 45 µM H2O2 with a lower sensitivity of 348±10 µA·mM-1·cm2. All sensors have a rapid response

SC
with a response time of approx. 10 s for the whole range of concentrations.
The performances of the H2O2 sensor based on 4% ERGO was compared with other hydrogen peroxide

U
sensors reported in the literature and the results are illustrated in Table S2 (Electronic Supplementary
Material). The ERGO-[BMIM][BF4]-NAF/SPCE sensor shows the highest sensitivity compared to other
N
sensors based on composite materials and one of the lowest detection limit. Sensitivity of the H2O2
A
sensors reported in table S2 varies from 31 [41], to 1395 µA·mM-1·cm-2 [42], which are lower that the
M

value reported in this work. The very high sensitivity of the sensor makes it suitable for quantification of
low variation of H2O2 concentration in biological sample
ED

3.6. Interferences studies and real samples analysis

Electroactive species such as ascorbic acid, citric acid, urea or creatinine are the main interference species
PT

that can be found in real samples. The effect of the potential interfering compounds on the H2O2 sensor
response was studied by amperometry in artificial serum at an applied potential of -550 mV. The artificial
E

urine contains 7 mmol/L KH2PO4, 7 mmol/L K2HPO4, 2.5 mmol/L CaCl2, 2 mmol/L MgCl2; 90 mmol/L
CC

NaCl, 25 mmol/L NH4Cl; 10 mmol/L Na2SO4, 25 mmol/L NaHCO3, 0.005 mml/L FeSO4, 7 mmol/L
creatinine; 170 mmol/L urea, 2 mmol/L citric acid, 1.1 mmol/L lactic acid and 0.4 mmol/L uric acid,
according to data published elsewhere [43]. The experiments were performed as follow: the sensor
A

response was recorded successively after a first addition of 0.1 mM H2O2; followed by an addition of 0.1
mM of the interfering species and finally by a second addition of 0.1 mM H2O2. Comparing the
amperometric responses of the electroactive species and H2O2 before and after addition of the interfering
compound, it can be concluded that the analytical signal recorded for 0.1 mM H2O2 was not significantly
affected by the presence of the tested compounds. The signal change was below 5% in all cases. The very

10
high sensitivity of the developed sensor makes it suitable for H2O2 detection in biological samples since
the limit between the normal and pathologic level of H2O2 in urine ranges between 5 and 150 µM [44-46].

The H2O2 production by E. faecium 2980 clinical strain was studied in aerobic and anaerobic conditions
in the presence and the absence of glycerol. It is expected to have a higher concentration of H2O2 in the
aerobic medium in the presence of glycerol due to an increase of the activity of the glycerol-3-phosphate
oxidase, which uses molecular oxygen to produce dihydroxyacetone and hydrogen peroxide [47].

Samples of culture medium were collected every hour in the first 8 hours of incubation. The samples were
diluted with PBS (1:1 volume ratio) and the H2O2 concentration was determined with the electrochemical

T
sensor. All the measurements were performed in triplicates. The results presented in the Fig. 7 show a

IP
very low concentration of H2O2 in the first 2 hours of incubation followed by an increase of its
concentration between 2 h and 6 h of incubation. One can see an increase of the amount of H2O2 in the

R
anaerobic condition if the medium contain glycerol the maximum being reached after 5 hours of

SC
incubation. These results can be explained by the consumption of the oxygen dissolved in the culture
medium and of the one present at the air-liquid interface. The H2O2 concentration reaches more or less the

U
same level in the case of both anaerobic media for an incubation longer than 6 hours. That indicates that
all oxygen from the culture flask was consumed. A significant increase of the amount of H2O2 produced
N
in the case of the aerobic culture in the presence of glycerol after 6 h of incubation. The result confirms
A
the mechanism proposed by others for the production of the H2O2 [47]. One can observe a decrease of the
M

H2O2 concentration by increasing the incubation time more than 6 hours, which can be attributed to the
low stability of this compound at the incubation temperature of 37oC and a lower reaction rate of the
enzymatic production of H2O2. On the other hand, the synthesis of NADH peroxidase that can decompose
ED

H2O2 can be responsible for the decrease of H2O2 concentration for longer incubation time [3].

Figure 7.
PT

Table 1 shows the results obtained by electrochemical and enzymatic-spectrometric method for E.
E

faecium culture in aerobic medium with glycerol. The good correlation between two methods indicate that
H2O2 sensor proposed in this work can be successfully used for detection of H2O2 in complex
CC

biochemical media. It has the advantage of being fast, sensitive, non-destructive and cheaper compared
with enzymatic-spectrometric method.
A

Table 1.

11
4. Conclusion

In this work a new H2O2 sensor based on ERGO-[BMIM][BF4] composite material was developed,
characterized and tested for detection of the H2O2 production by E. faecium clinical strain. The FT-IR
spectra confirmed the existence of interaction between the ERGO and [BMIM][BF4]. The impedance
measurements confirmed the excellent electron transfer at the surface of the SPCE electrode modified
with ERGO-[BMIM][BF4]-NAF, while the high catalytic constant and diffusion coefficient determined

T
from chronoamperometric measurements underline the enhanced electrochemical properties of this

IP
composite material towards H2O2 detection.

R
A high sensitivity of 1484±25 µA·mM-1·cm-2 and a low detection limit of 0.1µM for the H2O2 detection

SC
were determined using the optimal amount of ERGO of 4% in the composite material. Moreover, the
sensor exhibited an excellent selectivity to H2O2 in the presence of potential interfering compounds that
allowed us to use it for monitoring the level of H2O2 in culture media for E. faecium. Our study revealed

U
that the presence of the glycerol as carbon source in the culture medium dramatically increase the
N
production of the H2O2 in aerobic condition. Compared to enzymatic-spectrometric method the proposed
A
sensor presents many advantages allowing fast, sensitive and non-destructive detection of the analyte. It
can be successfully used for selection of the probiotic microorganisms based on their ability to produce
M

H2O2. Moreover, due to the analytical performances, the use of this sensor can be extended to a large
variety of biological samples. Quantification of low variation of H2O2 concentration can be achieved and
ED

it is useful especially when the difference between normal and pathologic level is situated in a narrow
concentration range.
PT

Acknowledgements
E
CC
A

This work was supported by a grant of Ministry of Research and Innovation, CNCS - UEFISCDI, project
number PN-III-P4-ID-PCE-2016-0288, within PNCDI III. M. Braik acknowledges the secondment under
FP7 Marie Curie PIRSES_GA_2012-318053. L.-G. Zamfir was supported by a fellowship at the Research
Institute of the University of Bucharest (ICUB). SEM images were accomplished with the help of Dr.
Virgil Marinescu from the National Institute for R&D in Electrical Engineering, Bucharest.

12
References

[1] H.J. Forman, Hydrogen Peroxide: The good, the bad, and the ugly, in: G. Valacchi, P.A. Davis (Eds.),
Oxidants in Biology, Springer 2008, p. 328.
[2] S. Andreescu, M. Hepel, Oxidative stress: Diagnostics, Prevention, and Therapy, American Chemical

T
Society 2011, pp. 83-98.
[3] T.I. Moy, E. Mylonakis, S.B. Calderwood, F.M. Ausubel, Cytotoxicity of hydrogen peroxide

IP
produced by Enterococcus faecium, Infect. Immun. 72 (2004) 4512-4520.

R
[4] R. Whittenbury, The differentiation of Streptococcus faecalis and Streptococcus faecium, J. Gen.

SC
Microbiol. 38 (1965) 279-287.
[5] W. Chen, S. Cai, Q.-Q. Ren, W. Wen, Y.-D. Zhao, Recent advances in electrochemical sensing for
hydrogen peroxide: a review, Analyst 137 (2012) 49-58.

U
[6] Y. Wang, Z. Wang, Y. Rui, M. Li, Horseradish peroxidase immobilization on carbon nanodots/CoFe
N
layered double hydroxides: direct electrochemistry and hydrogen peroxide sensing, Biosens. Bioelectron.
A
64 (2015) 57-62.
[7] W. Yang, Y. Li, Y. Bai, C. Sun, Hydrogen peroxide biosensor based on myoglobin/colloidal gold
M

nanoparticles immobilized on glassy carbon electrode by a Nafion film, Sens. Actuators B Chem. 115
(2006) 42-48.
ED

[8] M. Baccarin, B.C. Janegitz, R. Berté, F.C. Vicentini, C.E. Banks, O. Fatibello-Filho, et al., Direct
electrochemistry of hemoglobin and biosensing for hydrogen peroxide using a film containing silver
PT

nanoparticles and poly(amidoamine) dendrimer, Mat. Sci. Eng. C 58 (2016) 97-102.

[9] T.S.T. Balamurugan, V. Mani, C.-C. Hsieh, S.-T. Huang, T.-K. Peng, H.-Y. Lin, Real-time tracking
E

and quantification of endogenous hydrogen peroxide production in living cells using graphenated carbon
nanotubes supported Prussian blue cubes, Sens. Actuators B Chem. 257 (2018) 220-227.
CC

[10] B. Zhang, W. Ning, J. Zhang, X. Qiao, J. Zhang, J. He, et al., Stable dispersions of reduced graphene
oxide in ionic liquids, J. Mat. Chem. 20 (2010) 5401-5403.
A

[11] A.A. Ensafi, F. Rezaloo, B. Rezaei, Electrochemical sensor based on porous silicon/silver
nanocomposite for the determination of hydrogen peroxide, Sens. Actuators B Chem. 231 (2016) 239-
244.
[12] N. Nasirizadeh, Z. Shekari, A. Nazari, M. Tabatabaee, Fabrication of a novel electrochemical sensor
for determination of hydrogen peroxide in different fruit juice samples, Journal of Food and Drug
Analysis 24 (2016) 72-82.
13
[13] S. Berbeć, S. Żołądek, A. Jabłońska, B. Pałys, Electrochemically reduced graphene oxide on gold
nanoparticles modified with a polyoxomolybdate film. Highly sensitive non-enzymatic electrochemical
detection of H2O2, Sens. Actuators B Chem. 258 (2018) 745-756.
[14] S. Chakraborty, C. Retna Raj, Pt nanoparticle-based highly sensitive platform for the enzyme-free
amperometric sensing of H2O2, Biosens. Bioelectron. 24 (2009) 3264-3268.
[15] D.A. Dikin, S. Stankovich, E.J. Zimney, R.D. Piner, G.H.B. Dommett, G. Evmenenko, et al.,
Preparation and characterization of graphene oxide paper, Nature 448 (2007) 457-460.
[16] S. Woo, Y.-R. Kim, T.D. Chung, Y. Piao, H. Kim, Synthesis of a graphene–carbon nanotube
composite and its electrochemical sensing of hydrogen peroxide, Eletrochim. Acta 59 (2012) 509-514.

T
[17] D.A.C. Brownson, C.E. Banks, Graphene electrochemistry: an overview of potential applications,

IP
Analyst 135 (2010) 2768-2778.
[18] K.R. Ratinac, W. Yang, J.J. Gooding, P. Thordarson, F. Braet, Graphene and related materials in

R
electrochemical eensing, Electroanal. 23 (2011) 803-826.

SC
[19] F. Valentini, D. Roscioli, M. Carbone, V. Conte, B. Floris, E.M. Bauer, et al., Graphene and ionic
liquids new gel paste electrodes for caffeic acid quantification, Sens. Actuators B Chem. 212 (2015) 248-
255.
U
[20] S. Cui, Y. Li, D. Deng, L. Zeng, X. Yan, J. Qian, et al., Photo-reduction assisted synthesis of
N
MnO2/reduced graphene oxide/P25 for electrochemical detection of hydrogen peroxide, RSC Advances 6
A
(2016) 2632-2640.
M

[21] R. Zhang, W. Chen, Recent advances in graphene-based nanomaterials for fabricating

electrochemical hydrogen peroxide sensors, Biosens. Bioelectron. 89 (2017) 249-268.
ED

[22] J. Ping, Y. Wang, K. Fan, J. Wu, Y. Ying, Direct electrochemical reduction of graphene oxide on
ionic liquid doped screen-printed electrode and its electrochemical biosensing application, Biosens.
Bioelectron. 28 (2011) 204-209.
PT

[23] X.L. Y. Li, X. Zeng, Y. Liu, X. Liu, W. Wei, S. Luo, Nonenzymatic hydrogen peroxide sensor based
on a Prussian Blue-modified carbon ionic liquid electrode, Microchim Acta 165 (2009) 393-398.
E

[24] T.M.S. Chang, Biomedical Applications of Immobilized Enzymes and Proteins: Springer US; 1977.
CC

[25] T.-Y. Huang, J.-H. Huang, H.-Y. Wei, K.-C. Ho, C.-W. Chu, rGO/SWCNT composites as novel
electrode materials for electrochemical biosensing, Biosens. Bioelectron. 43 (2013) 173-179.
[26] H. Chen, Y. Wang, Y. Liu, Y. Wang, L. Qi, S. Dong, Direct electrochemistry and electrocatalysis of
A

horseradish peroxidase immobilized in Nafion-RTIL composite film, Electrochem. Commun. 9 (2007)

469-474.
[27] O.-M. Istrate, L. Rotariu, C. Bala, Electrochemical determination of NADH using screen printed
carbon electrodes modified with reduced graphene oxide and poly(allylamine hydrochloride), Microchim.
Acta 183 (2015) 57-65.

14
[28] H.E. Müller, Detection of hydrogen peroxide produced by microorganisms on an ABTS peroxidase
medium, Zentralblatt für Bakteriologie, Mikrobiologie und Hygiene. Series A: Medical Microbiology,
Infectious Diseases, Virology, Parasitology 259 (1985) 151-154.
[29] R. Ramachandran, S. Felix, G.M. Joshi, B.P.C. Raghupathy, S.K. Jeong, A.N. Grace, Synthesis of
graphene platelets by chemical and electrochemical route, Mater. Res. Bull. 48 (2013) 3834-3842.
[30] N. Maleki, A. Safavi, F. Tajabadi, High-Performance carbon composite electrode based on an ionic
liquid as a binder, Anal. Chem. 78 (2006) 3820-3826.
[31] W. Liu, X. Yan, J. Lang, Q. Xue, Electrochemical behavior of graphene nanosheets in
alkylimidazolium tetrafluoroborate ionic liquid electrolytes: influences of organic solvents and the alkyl

T
chains, J. Mater. Chem. 21 (2011) 13205-13212.

IP
[32] H.-L. Guo, X.-F. Wang, Q.-Y. Qian, F.-B. Wang, X.-H. Xia, A green approach to the synthesis of
graphene nanosheets, ACS Nano 3 (2009) 2653-2659.

R
[33] S.Y. Toh, K.S. Loh, S.K. Kamarudin, W.R.W. Daud, Graphene production via electrochemical

SC
reduction of graphene oxide: synthesis and characterisation, Chem. Eng. J. 251 (2014) 422-434.
[34] X. Wang, X. Zhang, Electrochemical co-reduction synthesis of graphene/nano-gold composites and

U
its application to electrochemical glucose biosensor, Eletrochim. Acta 112 (2013) 774-782.
[35] A.J. Bard, L. R.Faulkner, Electrochemical Methods: Fundamentals and Application, New York: John
N
Wiley; 1980
A
[36] L. Rotariu, L.-G. Zamfir, C. Bala, Low potential thiocholine oxidation at carbon nanotube-ionic
M

liquid gel sensor, Sens. Actuators B Chem. 150 (2010) 73-79.

[37] Z. Miao, D. Zhang, Q. Chen, Non-enzymatic hydrogen peroxide sensors based on multi-wall carbon
ED

nanotube/Pt nanoparticle nanohybrids, Materials 7 (2014) 2945.

[38] H. Razmi, E. Habibi, Nanomolar detection of hydrogen peroxide at a new polynuclear cluster of tin
pentacyanonitrosylferrate nanoparticle-modified carbon ceramic electrode, Anal. Biochem. 392 (2009)
PT

126-132.
[39] L.-G. Zamfir, L. Rotariu, V.E. Marinescu, X.T. Simelane, P.G.L. Baker, E.I. Iwuoha, et al., Non-
E

enzymatic polyamic acid sensors for hydrogen peroxide detection, Sens. Actuators B Chem. 226 (2016)
CC

525-533.
[40] F. Mollarasouli, K. Asadpour-Zeynali, S. Campuzano, P. Yáñez-Sedeño, J.M. Pingarrón, Non-
enzymatic hydrogen peroxide sensor based on graphene quantum dots-chitosan/methylene blue hybrid
A

nanostructures, Electrochim. Acta 246 (2017) 303-314.

[41] S.Z. Bas, Gold nanoparticle functionalized graphene oxide modified platinum electrode for hydrogen
peroxide and glucose sensing, Mat. Lett. 150 (2015) 20-23.

15
[42] M.-Y. Hua, H.-C. Chen, C.-K. Chuang, R.-Y. Tsai, J.-L. Jeng, H.-W. Yang, et al., The intrinsic
redox reactions of polyamic acid derivatives and their application in hydrogen peroxide sensor, Biomat.
32 (2011) 4885-4895.
[43] T. Brooks, C.W. Keevil, A simple artificial urine for the growth of urinary pathogens, Lett Appl
Microbiol. 24 (1997) 203-206.
[44] X. Chen, G. Wu, Z. Cai, M. Oyama, X. Chen, Advances in enzyme-free electrochemical sensors for
hydrogen peroxide, glucose, and uric acid, Microchim. Acta 181 (2014) 689-705.
[45] B. Halliwell, M.V. Clement, L.H. Long, Hydrogen peroxide in the human body, FEBS Letters 486
(2000) 10-13.

T
[46] L.H. Long, P.J. Evans, B. Halliwell, Hydrogen peroxide in human urine: implications for antioxidant

IP
defense and redox regulation, Biochem. Biophys. Res. Comm. 262 (1999) 605-609.
[47] T.W. Esders, C.A. Michrina, Purification and properties of L-alpha-glycerophosphate oxidase from

R
Streptococcus faecium ATCC 12755, J. Biol. Chem. 254 (1979) 2710-2715.

SC
U
N
A
M
ED
E PT
CC
A

16
Authors’ biographies

Mourad Braik received his Ph. D degree from the Faculty of Sciences of Tunis (FST), University of
Tunis EL Manar in 2015. His main current interests are: development and characterization of
electrochemical micro/nano-structured (bio) chemical sensors and photovoltaic cells.

Lucian-Gabriel Zamfir graduate in 2010, the Master Biomolecules from University of Bucharest,
Romania. He received Ph.D degree in 2013 from University of Bucharest. His research work is been
mainly focused on the design of new electrochemical sensors and biosensors.

Lucian Rotariu received a PhD degree in analytical chemistry from University of Bucharest, Romania,
in 1998. He is currently associate professor at Faculty of Chemistry, University of Bucharest, Romania.
His research activity is mainly focused on development of new electrochemical sensors and biosensors

T
for environmental, agro-food and medical applications, as well as surfaces functionalization for
immobilization of biomolecules.

IP
Camelia Bala serves as full professor of analytical chemistry in University of Bucharesty where she

R
began her academic career in 1990. She received her doctoral degree in Chemistry from the University of
Bucharest (1997). She is currently Head of the R&D Center LaborQ and coordinates the research group

SC
working in analytical electrochemistry, nanostructured electrochemical interfaces, design and
characterization of electrochemical sensors and biosensors, bioanalytical chemistry.

Mounir Ben Ali is a full professor at the High Institute of Applied Science and Technology of Sousse

U
(ISSAT, Tunisia) working mainly in the field of Biosensors and Bioelectronics. He has
published/presented over 80 papers/communications in the area of chemical and biosensors devices,
N
nonmaterial’s and their applications. His main current interests are: simulation and characterization of
electrochemical micro/nano-structured (bio) chemical sensors.
A

Mariana Carmen Chifiriuc, is professor of microbiology and immunology at the University of

M

Bucharest. Her main research focuses on phenotypic, genotypic and epidemiological investigation of
antibiotic resistance and microbial virulence; development of new anti-infective strategies against
antibiotic resistant bacteria (synthetic and natural compounds, probiotics, bacteriophages, nanomaterials).
ED
E PT
CC
A

17
Figure Captions

Figure 1. The preparation and working principle of the ERGO-[BMIM][BF4]-NAF/SPCE hydrogen

peroxide sensor.

Figure 2. FT-IR spectra of (a) GO, (b) ERGO, (c) [BMIM][BF4], (d) GO-[BMIM][BF4] and (e) ERGO-
[BMIM][BF4].

T
Figure 3. SEM images of (a) GO-NAF/SPCE, (b) ERGO-NAF/SPCE, (c) GO-[BMIM][BF4]-NAF/SPCE

IP
and (d) ERGO-[BMIM][BF4]-NAF/SPCE electrodes.

R
Figure 4. A. Cyclic voltammograms of GO-[BMIM][BF4]-NAF/SPCE in presence of (a) buffer, (b) 5
mM H2O2 and of ERGO-[BMIM][BF4]-NAF/SPCE in presence of (c) buffer (d) 5 mM H2O2. (SR=50

SC
mV/s); B. Effect of the scan rate on the reduction peak current of H2O2 (inset: plot of the peak current
versus the square root of the scan rate).

U
Figure 5. Nyquist plots of (a) bare SPCE, (b) GO-NAF/SPCE, (c) ERGO-NAF/SPCE (d) [BMIM][BF4]-
N
NAF/SPCE, (e) GO-[BMIM][BF4]-NAF/SPCE and (f) ERGO-[BMIM][BF4]-NAF/SPCE in 5 mM of
A
benzoquinone/hydroquinone.
M

Figure 6. Chronoamperograms of ERGO-[BMIM][BF4]-NAF/SPE sensor in (a) buffer, (b) 0.5 mM, (c) 1
mM, (d) 2.5 mM H2O2 at applied potential of -0.55 V. Inset plot: iH2O2/iPBS vs. t1/2 derived from
ED

chronoamperometric data for 5 mM H2O2.

Figure 7. Determination of H2O2 concentration in different culture media of E. faecium 2980 (■ –

PT

aerobic; ● – aerobic + glycerol; ▲- anaerobic; ▼ – anaerobic + glycerol). n=3

E
CC
A

18
 Fig 1

A
CC
EPT
ED
M
A
N
U
SC
RIP
T

19
 Fig 2

A
CC
EPT
ED
M
A
N
U
SC
RIP
T

20
 Fig 3

A
CC
EPT
ED
M
A
N
U
SC
RIP
T

21
 Fig 4

A
CC
EPT
ED
M
A
N
U
SC
RIP
T

22
 Fig 5

A
CC
EPT
ED
M
A
N
U
SC
RIP
T

23
 Fig 6

A
CC
EPT
ED
M
A
N
U
SC
RIP
T

24
 Fig 7

A
CC
EPT
ED
M
A
N
U
SC
RIP
T

25
 Table 1. Determination of the concentration of H2O2 produced by E. faecium 2980.

H2O2 concentration (µM)*

Incubation time
ERGO-[BMIM][BF4]-
(hours) spectrometric method
NAF/SPCE
3 11.5±0.7 12.0 ±1.0
4 21.5±1.3 19.2±1.3
5 67.9±3.5 69.6±3.1
6 128±13 125±10
7 12.5±1.1 12.4±0.3
8 16.5±1.3 15.5±1.1

T
* n=3

R IP
SC
U
N
A
M
ED
E PT
CC
A

26