Simultaneous chemosensing of tryptophan and the bacterial signal molecule indole by boron doped diamond electrode

Article abstract of DOI:10.1016/j.electacta.2018.06.105

A simple and robust chemosensing approach using a boron-doped diamond (BDD) electrode has been developed and applied to analyze tryptophan (TRP) and indole during the growth of Escherichia coli in a complex growth medium. The bacterial enzyme tryptophanase catalyzes TRP to indole, an emerging signaling molecule. The process can now be monitored using electrochemistry, in a method far beyond the traditional identification protocols. Electroanalysis in a non-aqueous medium comprising acetonitrile (ACN) and tetrabutylammonium hexafluorophosphate (TBAH) is capable of separating the oxidation peak of TRP from that of indole. Mechanisms are postulated for the electrochemical oxidation of indole and TRP in ACN chosen because of its wider potential range, proton acceptor property, and solubilization of analytes. The electrochemical oxidation of TRP involves the elimination of two electrons. With a detection limit of 0.5 μM for both indole and TRP, this chemosensing approach is sufficient to monitor the level of these two biomolecules during the bacterial growth period.

Full text of DOI:10.1016/j.electacta.2018.06.105

Accepted Manuscript
Simultaneous chemosensing of tryptophan and the bacterial signal molecule indole
by boron doped diamond electrode
Alyah Buzid, F. Jerry Reen, Fergal O'Gara, Gerard P. McGlacken, Jeremy D.
Glennon, John H.T. Luong
PII:
S0013-4686(18)31393-8
10.1016/j.electacta.2018.06.105
EA 32098
DOI:
Reference:
To appear in: Electrochimica Acta
Received Date: 20 March 2018
Revised Date: 13 June 2018
Accepted Date: 14 June 2018
Please cite this article as: A. Buzid, F.J. Reen, F. O'Gara, G.P. McGlacken, J.D. Glennon, J.H.T. Luong,
Simultaneous chemosensing of tryptophan and the bacterial signal molecule indole by boron doped
diamond electrode, Electrochimica Acta (2018), doi: 10.1016/j.electacta.2018.06.105.
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.

ACCEPTED MANUSCRIPT
1
2
Simultaneous Chemosensing of Tryptophan and the Bacterial Signal
Molecule Indole by Boron Doped Diamond Electrode
3
a,b
c
d,e
b
4
5
Alyah Buzid , F. Jerry Reen , Fergal O’Gara , Gerard P. McGlacken , Jeremy D.
a,b,
a,b,
Glennon *, and John H. T. Luong
*
6
a
7
8
Innovative Chromatography Group, Irish Separation Science Cluster (ISSC) Ireland,
University College Cork, Cork, Ireland
b
9
0
School of Chemistry and Analytical & Biological Chemistry Research Facility (ABCRF),
1
1
University College Cork, College Road, Cork T12 YN60, Ireland
c
1
School of Microbiology, University College Cork, College Road, Cork T12 YN60, Ireland
d
1
1
2
3
BIOMERIT Research Centre, School of Microbiology, University College Cork, College
Road, Cork T12 YN60, Ireland
e
1
1
4
5
School of Biomedical Sciences, Curtin Health Innovation Research Institute, Curtin
University, Perth WA 6845, Australia
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
* Corresponding authors:
E-mail address: j.glennon@ucc.ie (J. D. Glennon) and j.luong@ucc.ie (J. H. T. Luong)
1

ACCEPTED MANUSCRIPT
1
ABSTRACT
2
3
4
5
6
7
8
9
0
1
2
3
4
A simple and robust chemosensing approach using a boron-doped diamond (BDD) electrode
has been developed and applied to analyze tryptophan (TRP) and indole during the growth of
Escherichia coli in a complex growth medium. The bacterial enzyme tryptophanase catalyzes
TRP to indole, an emerging signaling molecule. The process can now be monitored using
electrochemistry, in a method far beyond the traditional identification protocols.
Electroanalysis in a non-aqueous medium comprising acetonitrile (ACN) and
tetrabutylammonium hexafluorophosphate (TBAH) is capable of separating the oxidation
peak of TRP from that of indole. Mechanisms are postulated for the electrochemical
oxidation of indole and TRP in ACN chosen because of its wider potential range, proton
acceptor property, and solubilization of analytes. The electrochemical oxidation of TRP
involves the elimination of two electrons. With a detection limit of 0.5 µM for both indole
and TRP, this chemosensing approach is sufficient to monitor the level of these two
biomolecules during the bacterial growth period.
1
1
1
1
1
1
1
5
6
Keywords:
1
1
7
8
Indole; tryptophan; boron-doped diamond electrode; electrochemical detection; Escherichia
coli
1
2
9
0
1. Introduction
2
2
2
2
2
2
2
2
2
3
3
3
1
2
3
4
5
6
7
8
9
0
1
2
The indole moiety is a ubiquitous heterocyclic compound found in several natural products
and synthetic compounds [1], including the non-steroidal anti-inflammatory indomethacin,
and the anti-HIV drug delavirdine. Over 85 different gram-positive and gram-negative
bacteria synthesize a significant amount of indole. As an example, tryptophanase, a
cytoplasmic enzyme of Escherichia coli hydrolyzes tryptophan (TRP) to indole, pyruvate,
and ammonia [2]. This biomolecule has been implicated in biofilm formation, virulence, and
antibiotic resistance [3-7]. The essential amino acid TRP is an important precursor of
hormones, melatonin, vitamins, niacin, neurotransmitters, and serotonin [8, 9]. Low plasma
concentrations of TRP can be linked to insomnia, anxiety, and depression. Indole acts as an
extracellular signaling molecule [7], present in the human gastrointestinal tract, which
triggers responses in bacteria [10]. The E. coli indole signal increases epithelial-cell tight-
junction resistance and attenuates indicators of inflammation.
3
3
3
3
3
3
3
4
5
6
7
8
High-performance liquid chromatography (HPLC) and gas chromatography (GC) are
commonly used for accurate measurement of indole and TRP in beverage and biological
samples [11-13]. In addition to the complexity of special equipment, chromatographic
analysis of TRP in protein hydrolysates is more challenging as it is destroyed by acid
hydrolysis, even under optimal conditions used with other amino acids. These expensive and
labor-intensive methods are lab-based and not well suited to routine or spot analysis.
3
4
4
4
4
4
4
9
0
1
2
3
4
5
Levels of indole and its analogs in biological samples can be estimated using the simple
and rapid Kovac’s assay [4]. The commercially available Kovac’s or Ehrlich’s reagent
consists of isoamyl alcohol, para-dimethylaminobenzaldehyde (DMAB), and concentrated
hydrochloric acid. The indole reacts with DMAB to yield a red complex, which is used to
differentiate E. coli from other indole-negative bacteria, e.g., Pseudomonas aeruginosa [13].
However, the reaction color changes from pink to yellowish orange at high indole
concentrations, reducing the absorption maxima at 530 nm and thus the quantitation
2

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
accuracy. Evidently, a fast and accurate general method for simultaneous analysis of TRP and
indole in complex biological matrices is of importance for diversified applications, whereas
both HPLC and GC still play an important role as routine standard methods. Although
considerable efforts have been expended to develop electrochemical sensing for TRP, its
voltammetric response is not satisfactory due to the slow heterogeneous electron transfer.
Also, modified electrodes for TRP are still limited and suffer from poor reproducibility.
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
The boron-doped diamond (BDD) electrode is of particular importance for the analysis of
target analytes with limited aqueous solubility and oxidation potentials over + 1V [14-19] and
thus demonstrates potential as a basis for a novel sensing methodology for indole, TRP, and
their analogs. The single electrochemical detection of indole or TRP using the BDD electrode
was previously reported [20-22]. In this paper, we describe a chemosensing protocol for
quantitating TRP and indole in the most commonly used growth medium for E. coli, using a
bare BDD electrode. The detection is performed in a non-aqueous milieu, containing an
organic conducting salt, to alleviate electrode fouling due to the non-specific adsorption of
diversified biomolecules of the growth medium. This non-aqueous sensing protocol also
circumvents the subsequent oxidation and electrode surface fouling, arising due to oxidized
products stemming from TRP and indole. The simultaneous detection of TRP and indole is
successfully achieved in a non-aqueous media consisting of 0.2 M tetrabutylammonium
hexafluorophosphate in acetonitrile (TBAH/ACN). It is noteworthy that the simultaneous
detection of both target analytes proved unsuccessful in aqueous media. The proposed
method is assessed for its analytical performance and reproducibility, and a limit of detection
(LOD) is established. Application of the method was then demonstrated using a growth
medium of E. coli, as a relevant model study. Here, metabolization of TRP to indole was
monitored and quantified during bacterial growth.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
5
2
2
2
6
7
8
2. Experimental
2.1 Chemicals
2
3
3
3
3
3
3
3
3
3
9
0
1
2
3
4
5
6
7
8
Sodium phosphate monobasic, sodium phosphate dibasic, phosphoric acid, indole,
tryptophan (TRP), sodium hydroxide, potassium chloride, potassium hexacyanoferrate,
acetonitrile (ACN), methanol (CH OH). Kovac’s reagent for indoles, tetrabutylammonium
3
hexafluorophosphate (TBAH), and ethanol were purchased from Sigma-Aldrich (Dublin,
Ireland). 50 mM phosphate buffer solutions were prepared at pH 2 and pH 7 with ACN (80:
20, v/v) were used as aqueous buffers. ACN was used to support the solubility of the target
analytes. The stock solution of indole (2 mM) were prepared in ACN daily before use, while
the stock solution of TRP was prepared in CH OH: H O (70: 30, v/v) at 2 mM. Deionized
3
2
water (Millipore, Ireland) was utilized throughout the experiments, and all chemicals were of
the analytical grade.
3
4
9
0
2.2 Apparatus and measurements
4
4
4
4
4
4
1
2
3
4
5
6
All measurements were performed at room temperature using a CHI1040A electrochemical
workstation (CH Instrument, Austin, TX). The electrochemical cell consists of the BDD
working electrode (Windsor Scientific, Slough Berkshire, UK), a Pt wire counter electrode
(Sigma-Aldrich, Dublin, Ireland), and a silver/silver chloride (Ag/AgCl /3M KCl) reference
electrode (BAS Analytical Instruments, West Layette, IN). The BDD electrode was polished
using 0.3 and 0.05 µm alumina slurry with wet Nylon and MasterTex papers, respectively,
3

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
followed by sonication in ethanol for 5 min and deionized water for 10 min. It is important to
mention that the polishing effect on the BDD electrode is less effective compared to other
materials such as gold, platinum, copper, etc. Nevertheless, we have noticed that the
polishing can clean the BDD electrode and this practice has repeatedly been used in our labs
as exemplified by our previous publications [14-17, 19]. For electroanalysis in ACN, TBAH,
a quaternary ammonium salt, serves mainly used as an electrolyte. The salt consists positively
charged tetrabutylammonium and weakly basic hexafluorophosphate anion. Such chemically
inert species allows the salt to serve as an inert electrolyte over a wide potential range,
particularly with the BDD electrode in ACN.
1
1
0
1
2.3 The growth of Escherichia coli
1
1
1
1
1
1
1
1
2
2
2
3
4
5
6
7
8
9
0
1
The microorganism was grown shaking (150 rpm) in LB (Luria-Bertani) medium
without/with 5 mM exogenous TRP at 37 ˚C. This nutritionally rich medium is routinely used
to grow bacterial cultures. LB broth contains (per mL), 10 mg tryptone (a mixture of peptides
formed by the digestion of casein with the pancreatic enzyme, trypsin), 5 mg yeast extract (an
autolysate of yeast cells), and 5 mg NaCl. The whole-cell cultures are used for the monitoring
study. For the monitoring the production of indole, sample aliquots are taken at different time
intervals (4, 6, and 8 h). For the E. coli bacterial culture without addition of TRP, a sample
aliquot was taken at 8 h. The E. coli was grown is an aqueous media. Then, the bacterial
culture was diluted 100 times while the bacterial culture without TRP was diluted 50 times in
0.2 M TBAH/ACN. For the Kovac’s test, sample aliquots are taken at 8 and 30 h.
2
2
2
2
2
3
4
5
3. Results and discussion
3.1 Electrochemical characteristics of indole and tryptophan in aqueous media
2
2
2
2
3
3
3
3
6
7
8
9
0
1
2
3
As an aromatic heterocyclic compound, indole consists of a benzene ring fused to a pyrrole
ring. The C3 position of the pyrrole ring is more susceptible to substitution. Therefore, most
indole derivatives, e.g., TRP, indole propionic acid, and indole acetic acid, have a
corresponding substituent at this position. The electrochemical oxidation of indole in aqueous
media was previously investigated. In brief, the oxidation of indole is pH-dependent above
pH 3.3, and one electron is transferred following the liberation of a proton, whereas the
oxidation of indole is pH-independent below pH 3.3 and only an electron is transferred [23,
24].
3
3
3
3
3
3
4
4
4
5
6
7
8
9
0
1
In our initial attempts using indole and TRP (in aqueous media), similar characteristics
were observed. Electrochemical oxidation of indole at pH 2 and pH 7 gave cyclic
voltammograms (CVs) with a broad peak at a low oxidation potential ~ + 0.9 V and a second
peak at high oxidation potential ~ + 1.5 V. The peak of indole diminished rapidly after the
second scan and shifted to less positive values (Fig. 1a, c). The electrochemical behavior of
TRP in the aqueous media was similar to indole (Fig. S1a, c). The results obtained here
suggested the forming of polymer film on the surface on the BDD electrode after a
subsequent run of CVs at pH 2 and pH 7.
4
4
4
2
3
4
We suspect the electrochemical (EC) oxidation mechanism of indole and TRP in aqueous
media is that the nucleophilic attack of water on the radical formed in the first oxidation step
was capable of rapidly hydroxylating the indole molecule as also suggested by Enache and
4

ACCEPTED MANUSCRIPT
1
2
Oliveira-Brett [24]. The hydroxylated indole was then subject to further oxidation involving
the formation of a number of other plausible oxidation products.
3
4
5
6
The most prominent oxidation product might well be 7-hydroxyindole [25]. A plausible
pathway for the electrochemical oxidation of indole and TRP in aqueous media is suggested
in Scheme 1. The broad oxidation peak ~ + 1.5 V observed at pH 2 and pH 7 could be
attributed to the collective oxidation of indoles and higher oxidized derivatives.
7
8
9
0
The DPVs of indole obtained at pH 2 and pH 7 also showed one single peak at ~ + 0.7 V
with diminishing magnitude after the second and third scan, likely due to the adsorption of
oxidation products and subsequent electrode fouling (Fig. 1b, d). The DPVs behavior of TRP
obtained at pH 2 and pH 7 was similar to indole (Fig. S1b, d)
1
1
1
1
1
2
3
Interestingly, the DPV for the indole and TRP mixture at pH 2 and pH 7 revealed only one
oxidation peak (Fig. 2), because the oxidation potential of indole and TRP is similar at pH 2
and pH 7.
1
1
4
5
3.2 Electrochemical behavior of Indole in non-aqueous media
1
1
1
6
7
8
Overall, we suspected water in our aqueous media was acting to hydroxylate indole and
TRP, resulting in hydroxyindoles which bound and fouled the electrode surface. Thus, we
turned our attention to non-aqueous media.
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
The electrochemical characteristics of indole in the non-aqueous medium were then
investigated in 0.2 M TBAH/ACN using the bare BDD electrode. The CV obtained for indole
exhibited one single irreversible peak at + 1.17 V, followed by a very broad peak during the
first scan (Fig. 3a). The subsequent CV runs of indole (~ + 0.5 V to + 2 V) can result in the
formation of a weak polyindole film on the surface of the BDD electrode. As discussed in
section 3.2, the CV of indole in non-aqueous media shows one oxidation peak at +1.17 V
followed by a very broad peak at higher potential due to the further oxidation of the first peak
at +1.17 V (Fig. 3a). While for DPV, the scan was stopped at +1.5 V and did not go further to
a higher potential which minimized further oxidation of indole products which might form a
polymer (Fig. 3b). The indole oxidation peak occurred at a higher potential, decreased
modestly with subsequent scans, but became stable after 10 cycles of repeated scanning.
Similar behavior was observed with 1 mM indole for 10 repeated cycles. When the electrode
was subject to a fresh 0.2 M TBAH/ACN solution, an electroactive peak at + 1.2 V was
observed. The peak was not stable and decreased rapidly with repeated scanning (Fig. S2). As
shown in Fig. 3, the electrochemical oxidation of indole was carried out in non-aqueous
media. Therefore it oxidized at higher potential compared to the aqueous media and
electroanalysis of the organic solvent occurred at higher potential compared to water which
results in the electrochemical oxidation at higher potential. Note also that the response
3
−/4−
current to Fe(CN)6
of the resulting electrode was considerably lower compared to a clean
BDD electrode. The peak was also very broad and shifted to a higher potential, implying the
formation of a weak electroactive film on the electrode area (Fig. S3). Noticeably, radical
cations formed close to the anode could diffuse away from the electrode surface in this
condition. Remaining radicals would combine to form a polymer, particularly at high indole
concentration (> 1 mM). The chain length of polyindole would increase as electrooxidation
progressed, to cover the electrode’s active surface area. Nevertheless, the film was easily
delaminated from the BDD electrode by CV before each measurement.
5

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
The first DPV obtained for indole at a clean BDD showed a prominent oxidation peak,
followed by a broad peak at a higher potential. The second peak could be attributed to the
oxidation of an indole oxidation product, i.e., polyindole formed at the BDD surface during
the first scan. There was a minimal change of the peak amplitude when the electrode was
subject to the second and third scan, indicating minimal polymerization of indole radicals
within this scanning period (Fig. 3b). Based on the voltammograms observed for the BDD
electrode in 0.2 M TBAH/ACN and some pertinent literature information [25-28], a
mechanism is suggested for the electrochemical oxidation of indole to form a polyindole film
(Scheme 2a). A number of regioisomeric couplings to give multiple polyindoles are possible
[29]. The Pt electrodes used as a comparison because it has been used for indole detection in
non-aqueous media which is similar to our present work.
1
1
1
1
1
1
1
1
1
1
2
2
3
4
5
6
7
8
9
0
One consideration we had at this stage was the fate of the removed proton in a solvent such
as ACN. Water is of course highly miscible in ACN and probably participates to some degree
in hydrogen bonding (CN--H). The acid in acetonitrile will result in an acid-base equilibrium
[30]. The interaction of ACN with trifluoromethanesulfonic acid is known and gives rise to a
large number of products comprising both neutral and protonated ACN [31]. Also, ACN
serves as a hydrogen-bond acceptor in the reactions of phenyl 2,4,6-trinitrophenyl ether with
amines in benzene–acetonitrile mixtures [32]. Evaluation of such evidence leads to the
possibility that ACN could accept the H+ liberated. An alternative scenario involves the
adsorption of H+ onto the electrode, followed by the hydrogen evolution [33].
2
2
1
2
3.3 Electrochemical behavior of Tryptophan in non-aqueous media
2
2
2
2
2
2
2
3
3
3
3
3
3
3
4
5
6
7
8
9
0
1
2
3
4
5
As depicted in Scheme 1, TRP is an indole derivative with an α-carboxylic acid group at
the 3-position. The structural differences (compared to indole) of TRP appeared to prevent
minimum electropolymerization. Although the electrochemical oxidation of TRP in aqueous
media has been investigated extensively in the literature [24-29, 34-40]. Based on our
observations and literature precedent, a plausible mechanism for the electrochemical
oxidation of TRP in ACN is suggested in Scheme 2b. The CVs of the TRP in 0.2 M
TBAH/ACN show two oxidation peaks at ~ + 0.9 V and + 1.3 V. The oxidation potential of
TRP remains stable after 10 cycles of repeated scanning (Fig. S4a). Also, the DPVs of 10 µM
TRP displayed a small change in the peak current when the electrode was subjected to the
second and third scans (Fig. S4b). The electrode after 10 cycles of 20 µM TRP in 0.2 M
3
-/4-
TBAH/ACN was tested in Fe(CN)6
and the CVs showed that there is a slightly lower
current compared to a clean BDD electrode, indicating minimal polymerization of TRP
compared to indole (Fig. S5).
3
3
6
7
3.4 Simultaneous detection of Indole and Tryptophan
3
3
4
4
4
4
4
4
4
8
9
0
1
2
3
4
5
6
Considering the capability of the BDD in ACN to exhibit distinctive peaks of indole and
TRP, we hoped that both biomolecules could be detected together. The TRP and indole were
detected first individually and then a mixture of these two analytes was tested. The two peaks
were distinguished by referring the oxidation potential for each. The DPV of TRP and indole
detected individually is presented in Fig. 4a, b. The DPV was then performed on a mixture of
TRP and indole (Fig. 4c). As expected, the mixture exhibited two peaks with some degree of
overlap. If TRP was prepared in methanol, there was no noticeable effect of methanol on the
electrochemical response or the potential window. The final methanol percentage presented
in the final solution is 1.75 % (Fig. 4).
6

ACCEPTED MANUSCRIPT
1
2
3
4
5
In contrast, only one single peak was noted when the experiment was performed in
aqueous electrolytes at pH 2 and pH 7 (Fig. 2). Thus, 0.2 M TBAH/ACN was chosen as the
optimum media for selective detection of indole and TRP. Consequently, the DPV in ACN
was used to establish the calibration curves and detect indole and TRP during the growth of
E. coli.
6
7
8
9
0
1
2
Notice also that the first oxidation potential of TRP was shifted by 0.03 V to less positive
value while the oxidation potential of indole was shifted by 0.028 V to a higher positive value
in the standard mixture compared to their individual values. The peak area of the first
-6
oxidation peak of TRP was slightly less by 0.11x10 in the standard mixture and the peak
-6
1
1
1
area of indole was also decreased by 0.48x10 in the standard mixture. The values of the
oxidation potentials and the peak areas for the individual and the standard mixture were
summarized in Table S1.
1
1
3
4
3.5 Calibration curve of Indole and Tryptophan
1
1
1
1
1
2
2
2
2
2
2
5
6
7
8
9
0
1
2
3
4
5
The DPVs of indole and TRP at different concentrations are shown in Fig. 5 and the first
DPV was taken for the calibration curve. For indole with a single peak, the peak area was
integrated and correlated with concentration. The peak potential of indole was concentration-
dependent and the peak area was increased with increasing concentration resulting in a slight
shift of the potential (Fig. 5a). A limit of detection (LOD) of 0.5 µM and linearity up to 100
2
µM with a correlation coefficient of R = 0.931 was obtained (Fig. 5a, inset). The LOD was
established as the signal response observable at the lowest concentration over the background
signal (S/N = 3). It should be noted that several literature reports claim very low LODs, based
on the estimated slope of the calibration curve. The claimed LOD is often at least ten-fold
below the lowest concentration used to establish the calibration curve. This method is
generally unsatisfactory and the values are rarely achieved using “real-world” samples.
2
2
2
2
3
3
3
3
3
6
7
8
9
0
1
2
3
4
Similar results were estimated for TRP. However, the shoulder of the second peak was
deconvoluted and subtracted from the total peak area (OriginPro, 2017, OriginLab). The
correlation coefficient of the calibrated linear curve was R = 0.987 with the final methanol
2
percentage presented in the final solution was 0.07 –0.7 % (Fig. 5b). Thomas et al. [41]
reported two linear dynamic ranges for TRP: 0.6–9.0 µM and 10.0–100.0 µM for the carbon
paste electrode modified with MWCNTs using the drop cast method. The LOD was reported
-7
as 4.20 ± 0.58 x 10 M, about ten-fold below the lowest concentration used to test the
electrode response. In reality, the electrode’s response was very similar for both 0.5 µM and
1 µM of TRP as clearly reported in the paper.
3
3
3
3
3
4
4
4
4
4
5
6
7
8
9
0
1
2
3
4
Turning our attention to a biological sample, the calibration of indole and TRP in a
bacterial growth medium (LB) is considered very challenging, since this complex matrix
contains various proteins and other biomolecules (Fig. 6 and insets). The final methanol
percentage presented in the final solution was 0.35 –3.5 % (Fig. 6b). As expected, the DPV
responses of both analytes in a spiked LB medium, were noticeably lower, with a detection
limit of 22 and 4.28 µM for indole and TRP, respectively. Notice also that the oxidation
signals of both indole and TRP in LB media were lower than ACN. The calibration curves of
indole and TRP were also carried out in the LB media, and the curves show linearity, which
illustrated the measurement was quantitative. After each measurement, the BDD electrode
was cleaned in 0.5 M H SO .
2
4
4
4
5
6
From a practical aspect, fecal indole concentrations, for example, from healthy adults vary
widely from 0.30 mM to 6.64 mM [4]. In the laboratory, enterotoxigenic E. coli strain
7

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
H10407 produces over 3.3 mM indole during a 24-h period in the presence of 5 mM TRP [4].
It appeared to us that the detection limit using BDD electrodes in our work would be
sufficient for the analysis of indole in such biological samples. The plasma TRP
concentrations in case of obsessive-compulsive disorder and major depressive disorder range
from 50 to 70 µM [42], significantly higher than the LOD of the bare BDD electrode.
However, it is present with two metabolites, 5-hydroxytryptophan, and serotonin at much
lower concentrations (ng/mL) in patients affected with different forms of amenorrhea [43].
This opens up a new challenge for chemosensing, i.e., to target a minute quantity of these two
molecules together with TRP.
1
1
0
1
3.6 Monitoring the level of Tryptophan and Indole during the growth of E. coli
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
As mentioned in the Experimental Section, 5 mM of TRP was added to the growth
medium, and this added amount deserves a brief discussion here. In E. coli, the enzyme
tryptophanase produces indole from TRP and the final yield of indole depends on the amount
of exogenous TRP. The enzyme converts TRP into an equal amount of indole, up to almost 5
mM. The DPVs for the E. coli cell broth collected at 4, 6, and 8 h were shown in Fig. 7. As
expected, the level of TRP decreased corresponding to increasing indole levels. These
concentrations were estimated as ~ 4, 3, and 2 mM for TRP and ~ 0.45, 0.92, and 2.5 mM for
indole in the cell culture at 4, 6, and 8 h, respectively. Moreover, the DPV recorded in the
growth media without the addition of TRP is presented in Fig. 7. However, the potential of
TRP and indole in E. coli was shifted slightly to less positive value by 0.03 V and 0.116 V,
respectively compared to the value obtained using the standards. Also, the concentrations of
TRP and indole in E.coli were determined by the standard addition method. Thus, the
corresponding concentration consumption of TRP resulting from metabolism by E. coli were
1, 2, and 3 mM during the bacterial growth. It was also anticipated that an appreciable
amount of TRP was conjugated with intracellular protein and not available for detection.
Nevertheless, judging from the detection limit and its applicability in complex media, this
chemosensing approach is remarkably efficient and applicable and will find applications in
diversified and important bioprocesses and biological samples. In addition, E. coli grown in
LB media supplemented with 5 mM TRP for 8 h and 30 h were tested with Kovac’s reagent.
As shown in Fig. S6, positive results were shown with the bacterial culture grown for 8 h and
30 h, whereas negative results were observed with the LB media alone, due to the absence of
indole.
3
3
4
5
4. Conclusions
3
3
3
3
4
4
4
4
4
4
4
4
6
7
8
9
0
1
2
3
4
5
6
7
The simultaneous electrochemical detection of indole and TRP was attempted in both
aqueous and non-aqueous media on the bare BDD electrode. The results show electroanalysis
using the BDD electrode in ACN together with TBAH was capable of distinguishing the
oxidation peaks of the indole and TRP mixture with a detection limit of 0.5 µM for both
biomolecules. In contrast, the simultaneous detection of both analytes was not successful
aqueous media because the oxidation potentials of indole and TRP are similar. Non-aqueous
chemosensing with a broad potential window is more suitable for the analysis of the target
analytes and alleviates electrode fouling compared to aqueous media. The sensing protocol
was demonstrated for monitoring the levels of TRP decrease and indole increase during the
growth of E. coli in a complex medium. The sensing approach with BDD also circumvented
electrode fouling during the electrochemical oxidation of electropolymerizable targets as well
as its exposure to a complex medium.
8

ACCEPTED MANUSCRIPT
1
Acknowledgments
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
This research was financially supported by SFI/EI Technology Innovation Development
Award (TIDA) (SFI/12/TIDA/B2405). FJR, GPM, and FOG acknowledge support from
Enterprise Ireland (CF-2017-0757-P). FOG acknowledges the grants awarded by the
European Commission (FP7-PEOPLE-2013-ITN, 607786; FP7-KBBE-2012-6, CP-TP-
312184; FP7-KBBE-2012-6, 311975; OCEAN 2011-2, 287589; Marie Curie 256596; EU-
634486), Science Foundation Ireland (SSPC-2, 12/RC/2275; 13/TIDA/B2625;
12/TIDA/B2411; 12/TIDA/B2405; 14/TIDA/2438; 15/TIDA/2977), the Department of
Agriculture and Food (FIRM/RSF/CoFoRD; FIRM 08/RDC/629; FIRM 1/F009/MabS;
FIRM 13/F/516), the Irish Research Council for Science, Engineering and Technology
(PD/2011/2414; GOIPG/2014/647), the Health Research Board/Irish Thoracic Society
(MRCG-2014-6), the Marine Institute (Beaufort award C2CRA 2007/082) and Teagasc
(Walsh Fellowship 2013). JDG thanks the Science Foundation Ireland (08/SRC/B1412) for
research funding of the Irish Separation Science Cluster (ISSC) under the Strategic Research
Cluster Programme. GPM acknowledges supports by the Science Foundation Ireland
(SFI/12/IP/1315, SFI/09/RFP/CHS2353, and SSPC2 12/RC/2275), the Irish Research
Council (GOIPG/2013/336) and UCC for a Strategic Research Fund Ph.D. Studentship.
1
1
1
1
1
1
1
1
1
1
2
2
2
8
9
0
1
2
Appendix A. Supplementary data
Supplementary data to this article can be found at
References
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
[1] G.W. Gribble, Introduction, Indole ring synthesis, John Wiley & Sons, Ltd 2016, pp. 1-
38.
[2] W.A. Newton, Y. Morino, E.E. Snell, Properties of crystalline tryptophanase, J. Biol.
Chem. 240 (1965) 1211-1218.
[3] T. Bansal, D. Englert, J. Lee, M. Hegde, T.K. Wood, A. Jayaraman, Differential effects of
epinephrine, norepinephrine, and indole on Escherichia coli O157:H7 chemotaxis,
colonization, and gene expression, Infect. Immun. 75 (2007) 4597-4607.
[4] C. Darkoh, C. Chappell, C. Gonzales, P. Okhuysen, A rapid and specific method for the
detection of indole in complex biological samples, Appl. Environ. Microbiol. 81 (2015)
8093-8097.
[5] J. Domka, J. Lee, T.K. Wood, YliH (BssR) and YceP (BssS) regulate Escherichia coli K-
12 biofilm formation by influencing cell signaling, Appl. Environ. Microbiol. 72 (2006)
2449-2459.
[6] H. Hirakawa, Y. Inazumi, T. Masaki, T. Hirata, A. Yamaguchi, Indole induces the
expression of multidrug exporter genes in Escherichia coli, Mol. Microbiol. 55 (2005) 1113-
1126.
[7] J.H. Lee, J. Lee, Indole as an intercellular signal in microbial communities, FEMS
Microbiol. Rev. 34 (2010) 426-444.
[8] I. Gonzalez-Burgos, E. Olvera-Cortes, A.R. Del Angel-Meza, A. Feria-Velasco, Serotonin
involvement in the spontaneous alternation ability: a behavioral study in tryptophan-restricted
rats, Neurosci. Lett. 190 (1995) 143-145.
[9] D.M. Richard, M.A. Dawes, C.W. Mathias, A. Acheson, N. Hill-Kapturczak, D.M.
Dougherty, L-Tryptophan: basic metabolic functions, behavioral research and therapeutic
indications, Int. J. Tryptophan Res. 2 (2009) 45-60.
9

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
[10] T. Bansal, R.C. Alaniz, T.K. Wood, A. Jayaraman, The bacterial signal indole increases
epithelial-cell tight-junction resistance and attenuates indicators of inflammation, Proc. Natl.
Acad. Sci. U.S.A. 107 (2010) 228-233.
[11] C.A. Gallagher, L.B. Hough, S.M. Keefner, A. Seyed-Mozaffari, S. Archer, S.D. Glick,
Identification and quantification of the indole alkaloid ibogaine in biological samples by gas
chromatography-mass spectrometry, Biochem. Pharmacol. 49 (1995) 73-79.
[12] J.W.H. Yong, L. Ge, W.S. Wong, Z. Ma, S.N. Tan, Analyses of indole compounds in
sugar cane (saccharum officinarum l.) juice by high performance liquid chromatography and
liquid chromatography-mass spectrometry after solid-phase extraction, Separations 4 (2017)
1-12.
[13] Y.H. Zhu, Y. Yang, Z.X. Zhou, G.R. Li, M. Jiang, C. Zhang, S.Q. Chen, Direct
determination of free tryptophan contents in soy sauces and its application as an index of soy
sauce adulteration, Food Chem 118 (2010) 159-162.
[14] A. Buzid, J.H.T. Luong, F.J. Reen, F. O'Gara, J.D. Glennon, G.P. McGlacken, Rapid
electrochemical detection of Pseudomonas aeruginosa signaling molecules by boron-doped
diamond electrode, Methods Mol. Biol. 1673 (2018) 107-116.
[15] A. Buzid, E.O. Muimhneachain, F.J. Reen, P.E. Hayes, L.M. Pardo, F. Shang, F. O'Gara,
J. Sperry, J.H.T. Luong, J.D. Glennon, G.P. McGlacken, Synthesis and electrochemical
detection of a thiazolyl-indole natural product isolated from the nosocomial pathogen
Pseudomonas aeruginosa, Anal. Bioanal. Chem. 408 (2016) 6361-6367.
[16] A. Buzid, F.J. Reen, V.K. Langsi, E.Ó. Muimhneacháin, F. O'Gara, G.P. McGlacken,
J.H.T. Luong, J.D. Glennon, Direct and rapid electrochemical detection of Pseudomonas
aeruginosa quorum signaling molecules in bacterial cultures and cystic fibrosis sputum
samples through cationic surfactant‐assisted membrane disruption, ChemElectroChem 4
(2017) 533-541.
[17] A. Buzid, F. Shang, F.J. Reen, E.O. Muimhneachain, S.L. Clarke, L. Zhou, J.H.T.
Luong, F. O'Gara, G.P. McGlacken, J.D. Glennon, Molecular signature of Pseudomonas
aeruginosa with simultaneous nanomolar detection of quorum sensing signaling molecules at
a boron-doped diamond electrode, Sci. Rep. 6 (2016) 30001.
[18] F. Shang, E.O. Muimhneachain, F.J. Reen, A. Buzid, F. O'Gara, J.H.T. Luong, J.D.
Glennon, G.P. McGlacken, One step preparation and electrochemical analysis of IQS, a cell-
cell communication signal in the nosocomial pathogen Pseudomonas aeruginosa, Bioorg.
Med. Chem. Lett. 24 (2014) 4703-4707.
[19] L. Zhou, J.D. Glennon, J.H.T. Luong, F.J. Reen, F. O'Gara, C. McSweeney, G.P.
McGlacken, Detection of the Pseudomonas Quinolone Signal (PQS) by cyclic voltammetry
and amperometry using a boron doped diamond electrode, Chem. Commun. 47 (2011)
10347-10349.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
[20] D.K. Belghiti, E. Scorsone, J. de Sanoit, P. Bergonzo, Simultaneous detection of indole
and 3-methylindole using boron-doped diamond electrodes, Phys. Status Solidi 213 (2016)
2662-2671.
[21] J.S. Foord, A. Chatterjee, Electrochemical detection of indoles at diamond electrodes,
Phys. Status Solidi 202 (2005) 2110-2115.
[22] Q. Wang, A. Vasilescu, P. Subramanian, A. Vezeanu, V. Andrei, Y. Coffinier, M.S. Li,
R. Boukherroub, S. Szunerits, Simultaneous electrochemical detection of tryptophan and
tyrosine using boron-doped diamond and diamond nanowire electrodes, Electrochem
Commun 35 (2013) 84-87.
[23] J. Bunnett, R. Jones, Names for hydrogen atoms, ions, and groups, and for reactions
involving them (Recommendations 1988), Pure Appl. Chem. 60 (1988) 1115-1116.
[24] T.A. Enache, A.M. Oliveira‐Brett, Pathways of electrochemical oxidation of indolic
compounds, Electroanalysis 23 (2011) 1337-1344.
10

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
[25] K. Somers, E. Kryachko, A. Ceulemans, Theoretical study of indole: protonation,
indolyl radical, tautomers of indole, and its interaction with water, Chem. Phys. 301 (2004)
61-79.
[26] K.M. Choi, J.H. Jang, H.W. Rhee, K.H. Kim, A study of polyindole perchlorate (PIP)
prepared by electropolymerization, J. Appl. Polym. Sci. 46 (1992) 1695-1706.
[27] K.M. Choi, C.Y. Kim, K.H. Kim, Polymerization mechanism and physicochemical
properties of electrochemically prepared polyindole tetrafluoroborate, J. Phys. Chem. A 96
(1992) 3782-3788.
[28] R.J. Waltman, A. Diaz, J. Bargon, Substituent effects in the electropolymerization of
aromatic heterocyclic compounds, J. Phys. Chem. 88 (1984) 4343-4346.
[29] M. Saraji, A. Bagheri, Electropolymerization of indole and study of electrochemical
behavior of the polymer in aqueous solutions, Synth. Met. 98 (1998) 57-63.
[30] M. Kilpatrick, M.L. Kilpatrick, Relative acid strengths in acetonitrile, Chem. Rev. 13
(1933) 131-137.
[31] G.E. Salnikov, A.M. Genaev, V.G. Vasiliev, V.G. Shubin, Interaction of acetonitrile
with trifluoromethanesulfonic acid: unexpected formation of a wide variety of structures,
Org. Biomol. Chem. 10 (2012) 2282-2288.
[32] O. Banjoko, I.A. Babatunde, Catalytic effects of hydrogen-bond acceptor solvent on
nucleophilic aromatic substitution reactions in non-polar aprotic solvent: reactions of phenyl
2, 4, 6-trinitrophenyl ether with amines in benzene–acetonitrile mixtures, Tetrahedron 61
(2005) 8035-8040.
[33] M.N. Jackson, Y. Surendranath, Donor-dependent kinetics of interfacial proton-coupled
electron transfer, J. Am. Chem. Soc. 138 (2016) 3228-3234.
[34] V. Brabec, V. Mornstein, Electrochemical behaviour of proteins at graphite electrodes:
II. Electrooxidation of amino acids, Biophys. Chem. 12 (1980) 159-165.
[35] M.R.M. Domingues, P. Domingues, A. Reis, C. Fonseca, F.M. Amado, A.J. Ferrer-
Correia, Identification of oxidation products and free radicals of tryptophan by mass
spectrometry, J. Am. Soc. Mass Spectrom. 14 (2003) 406-416.
[36] B. Malfoy, J. Reynaud, Electrochemical investigations of amino acids at solid
electrodes: part II. Amino acids containing no sulfur atoms: tryptophan, tyrosine, histidine
and derivatives, J. Electroanal. Chem. Interfac. 114 (1980) 213-223.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
[37] N. Nguyen, M.Z. Wrona, G. Dryhurst, Electrochemical oxidation of tryptophan, J.
Electroanal. Chem. Interfac. 199 (1986) 101-126.
[38] C.J. Nielsen, R. Stotz, G.T. Cheek, R.F. Nelson, The anodic oxidation of indole in
acetonitrile an anomalous product, J. Electroanal. Chem. Interfac. 90 (1978) 127-130.
[39] P. Pandey, R. Prakash, Electrochemical synthesis of polyindole and its evaluation for
rechargeable battery applications, J. Electrochem. Soc. 145 (1998) 999-1003.
[40] L. Shagun, I. Dorofeev, V. Smirnov, I. Tokareva, M. Voronkov, Synthesis of poly (2, 2′,
3, 3′-indole), Russ. Chem. Bull. 61 (2012) 2360-2362.
[41] T. Thomas, R.J. Mascarenhas, O.J. D'Souza, P. Martis, J. Dalhalle, B.E. Swamy, Multi-
walled carbon nanotube modified carbon paste electrode as a sensor for the amperometric
detection of L-tryptophan in biological samples, J. Colloid Interface Sci. 402 (2013) 223-229.
[42] L. Bellodi, S. Erzegovesi, L. Bianchi, V. Lucini, R. Conca, A. Lucca, Plasma tryptophan
levels and tryptophan/neutral amino acid ratios in obsessive-compulsive patients with and
without depression, Psychiatry Res. 69 (1997) 9-15.
[43] A. Tagliamonte, G. Biggio, L. Vargiu, G.L. Gessa, Free tryptophan in serum controls
brain tryptophan level and serotonin synthesis, Life Sci. 12 (1973) 277-287.
4
4
8
9
11

ACCEPTED MANUSCRIPT
1
2
3
4
5
6
7
8
9
0
Fig. 1. CVs (a, b) were obtained for 10 continuous cycles of 20 µM indole whereas
differential pulse voltammetry (DPVs) (b, d) of 10 µM indole were recorded for three
repeated runs on bare BDD electrode vs. Ag/AgCl. (a) and (b) using 50 mM phosphate
buffer, pH 2 with 20 % acetonitrile , ACN (80: 20, v/v) whereas (c) and (d) using 50 mM
phosphate buffer, pH 7 with 20 % (ACN, 80: 20, v/v).
1
1
1
1
2
Scheme 1. Electrochemical oxidation of (a) indole and (b) tryptophan (TRP) in an aqueous
medium.
1
1
1
3
4
5
Fig. 2. A mixture of 50 µM each of indole and TRP using (a) 50 mM phosphate buffer, pH 2
with 20 % ACN (80: 20, v/v) and (b) 50 mM phosphate buffer, pH 7 with 20 % ACN (80: 20,
v/v) on the bare BDD electrode vs. Ag/AgCl.
1
1
1
6
7
8
Fig. 3. (a) CVs of 20 µM indole obtained for 10 continuous cycles and (b) DPVs of 10 µM
indole recorded for three repeated runs on the bare BDD electrode vs. Ag/AgCl. The
detection was achieved in 0.2 M tetrabutylammonium hexafluorophosphate (TBAH)/ACN.
1
2
9
0
Scheme 2. Electrochemical oxidation of (a) indole and (b) tryptophan in a non-aqueous
medium.
2
2
2
2
1
2
3
4
Fig. 4. (a) DPVs of in the absence (dotted lines) and presence (solid lines) of 50 µM TRP, (b)
DPV of 50 µM indole, and (c) a representative DPVs obtained for a mixture of TRP and
indole (50 µM each). The detection was achieved on the bare BDD electrode using 0.2 M
TBAH/ACN.
2
2
2
2
2
3
3
5
6
7
8
9
0
1
Fig. 5. (a) DPVs of different indole concentrations on the bare BDD electrode (1 µM –100
µM) and the calibration plot of indole (insert); intercept =7.84 x 10-7 ± µA and slope =1.64
x 10-7 ± µA/µM (95 % of confidence interval). (b) DPVs of different TRP concentrations on
the bare BDD electrode (2 µM – 20 µM) and the calibration plot of TRP (insert); intercept =
1.85 x 10-7 ± µA and slope = 1.1 x 10-7 ± µA/µM (95 % of confidence interval), using 0.2 M
TBAH/ACN. The error bar was estimated from three different values of the signal response
for each analyte concentration.
3
3
3
3
3
3
3
3
2
3
4
5
6
7
8
9
Fig. 6. Calibration curves of indole and TRP in LB media. (a) DPVs of different indole
concentrations on the bare BDD electrode (10 µM –100 µM) and the calibration plot of
indole in LB media (insert); intercept = -1.21 x 10-6 ± µA and slope = 8.4 x 10-8 ± µA/µM (
95 % of confidence interval). (b) DPVs of different TRP concentrations on the bare BDD
electrode (10 µM – 100 µM) and the calibration plot of TRP in LB media (insert); intercept =
7.5 x 10-7 ± µA and slope = 4.14 x 10-8 ± µA/µM (95 % of confidence interval), using 0.2 M
TBAH/ACN. A blank LB medium is the control presented as dotted lines. The error bar was
estimated from three different values of the signal response for each analyte concentration.
4
4
4
0
1
2
Fig. 7. Monitoring the production of indole in the E. coli bacterial culture as a function of
time and DPV of the E. coli bacterial culture grown for 8 h without the addition of TRP. The
detection was achieved on the bare BDD electrode vs. Ag/AgCl.
12

ACCEPTED MANUSCRIPT
1
2
13

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT