Electrochemical profiling of natural furanocoumarins:
doi: https://doi.org/10.5599/admet.2199 1
ADMET & DMPK 0(0) (2024) 000-000; doi: https://doi.org/10.5599/admet.2199
Open Access : ISSN : 1848-7718
http://www.pub.iapchem.org/ojs/index.php/admet/index
Original scientific paper
Electrochemical profiling of natural furanocoumarins:
DNA interaction dynamics of oxypeucedanin and prantschimgin
Hüseyin Oğuzhan Kaya1, Gokay Albayrak2, Hasan Isbilir3, Fatma Kurul4, Sura Baykan5,
Yeni Wahyuni Hartati6 and Seda Nur Topkaya1,*
1Department of Analytical Chemistry, Faculty of Pharmacy, Izmir Katip Celebi University, Izmir, Türkiye
2Department of Pharmaceutical Botany, Faculty of Pharmacy, Izmir Katip Celebi University, Izmir, Türkiye
3Department of Nanoscience and Nanotechnology, Faculty of Science, Izmir Katip Celebi University, Izmir, Türkiye
4Izmir International Biomedicine and Genome Institute, Dokuz Eylul University, Izmir, Türkiye
5Department of Pharmaceutical Botany, Faculty of Pharmacy, Ege University, Izmir, Türkiye
6Department of Chemistry, Faculty of Mathematics and Natural Sciences, Universitas Padjadjaran, Indonesia
*Corresponding Author: E-mail: sedanur6@gmail.com; Tel.: + 90 232 329 35 35
Received: November 24, 2023; Revised: February 28, 2024; Published: March 7, 2024
Abstract
Background and purpose: In this study, we present an electrochemical sensor for the detection of
oxypeucedanin (Oxyp) and prantschimgin (Pra), two natural furanocoumarin derivatives. The determination
of the effects of these molecules on DNA is important to be potential drug candidates. Our research focused
on exploring the electrochemical behaviour of these compounds and their interaction with DNA.
Experimental approach: The electrochemical properties of Oxyp and Pra were systematically analyzed by
evaluating their oxidation currents. Changes in the oxidation currents and peak potentials of guanine bases
were monitored before and after interaction in the solution phase and at the electrode surface. Key results:
The limit of detection (LOD) and limit of quantitation (LOQ) for Oxyp were determined to be 1.3 and
4.3 µg/mL, respectively. For Pra, the LOD and LOQ were found to be 20 and 68 µg/mL, respectively. Stability
studies demonstrated that the Oxyp solution retained its oxidation capacity for over a month, whereas the
Pra solution retained its oxidation capacity for nearly 120 min. Our findings suggest that Oxyp interacts with
dsDNA, potentially through electrostatic interactions, showing promise as a potential drug candidate
targeting DNA. On the other hand, the interaction of Pra with dsDNA requires further exploration to fully
understand its mode of action. Conclusion: The electrochemical sensor developed in this study provides a
reliable and efficient method for detecting and analysing the interaction of these natural compounds with
dsDNA. Our research contributes to advancing the understanding of the interaction between natural
furanocoumarins and dsDNA, laying the groundwork for the design and development of novel and effective
DNA-targeted drugs.
©2024 by the authors. This article is an open-access article distributed under the terms and conditions of the Creative Commons
Attribution license (http://creativecommons.org/licenses/by/4.0/).
Keywords
Natural compounds; pencil graphite electrode; drug candidate; coumarins; electrochemical detection
Introduction
Natural compounds are predominantly considered potential drug candidates or lead compounds for novel
drug exploration in recent years [1]. This class of compounds includes coumarins, a group of secondary
metabolites synthesized in plants through the metabolic pathway known as the shikimate pathway. They are
divided into subunits based on their structural categories, such as simple coumarins, furanocoumarins,
https://doi.org/10.5599/admet.2199
https://doi.org/10.5599/admet.2199
http://www.pub.iapchem.org/ojs/index.php/admet/index
mailto:sedanur6@gmail.com
http://creativecommons.org/licenses/by/4.0/
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
2
pyranocoumarins, 4-phenylcoumarins, and benzocoumarins [2]. Coumarins have a lactone ring formed
structurally by benzene fused to an α-pyrone ring. The conjugated system in this structure possesses
numerous electrons and exhibits good charge-transport characteristics. Some coumarins serve as additives
in cosmetic products and aroma enhancers in tobacco and beverages [3]. They have been reported to possess
anti-oxidant, anticancer, anti-microbial, analgesic, anti-diabetic, anti-inflammatory, and neuroprotective
properties [4].
Two natural furanocoumarins, oxypeucedanin (Oxyp) and prantschimgin (Pra), have been identified in
various plant species, such as Angelica, Prangos, Ferulago, and Citrus [1]. They have been extensively studied
for their diverse biological activities, including anti-proliferative, cytotoxic, antispasmodic, anti-inflammatory,
anti-influenza, and anti-allergic effects. Oxyp is often synthesized by plants as a part of their defense
mechanisms or as secondary metabolites. While these compounds share some similarities, there are also
distinct differences that set them apart. For instance, they both originate from the same biosynthetic
pathway, the shikimate pathway, and possess a furanocoumarin skeleton. However, they are differentiated
by their substitutes. Oxyp has 4-((3,3-dimethyloxiranyl)methoxy) side chain, while Pra has a 2-propan-2-yl 3-
methylbut-2-enoate substitute and lacks a double bond in the furan ring. Oxyp has been extensively studied
for its antioxidant, anti-inflammatory, and anticancer properties [2]. On the other hand, Pra has been
investigated for its anti-bacterial, anti-viral, and anti-fungal activities [3]. The pharmacokinetic properties of
Oxyp and Pra have not been thoroughly studied. Further research is necessary to elucidate the
pharmacokinetic profiles of these compounds, contributing to a better understanding of their efficacy and
safety in various therapeutic applications. Moreover, comprehensive research is needed to understand and
compare the similarities and differences between Oxyp and Pra.
These compounds are isolated and purified from plant extracts using chromatographic techniques [4].
Other analytical techniques, such as nuclear magnetic resonance (NMR) [5], mass spectrometry (MS) [6],
infrared (IR) spectroscopy, and ultraviolet-visible (UV-Vis) spectroscopy, [7] are also commonly employed
for the identification and characterization of these compounds. In one study, UPLC/MS/MS was employed
to determine Oxyp, one of the three major furanocoumarins in Angelica dahurica roots. The limit of
quantification (LOQ) for Oxyp was 1 ng/mL [8]. In another study, the detection of Oxyp was carried out
using the 1H-qNMR method, and the limit of detection (LOD) for the compound was found to be
0.149 mg/mL [9].
In another study, HPLC-DAD and HPLC-MS/MS, which combine a diode arrangement detector with mass
spectrometry, were used to detect Oxyp. The lower LOD was found to be 0.1 μg/mL [10]. Previously, Pra was
detected in Ferulago species by HPLC-DAD. LOD and LOQ for Pra were 0.264 and 0.879 μg/mL,
respectively [3]. In another study with Ferulago species, two coumarins, felamedin and Pra were analyzed
using HPLC. LOD and LOQ for Pra were determined as 0.0381 and 0.0177 mg/mL, respectively [11]. While
these methods are commonly used for detecting Oxyp and Pra, they have some drawbacks, such as requiring
burdensome pretreatment technology, large amounts of pure solvents, being time-consuming and labor-
intensive, and often involving expensive and complex processes. On the other hand, electrochemical
methods are advantageous as they overcome these disadvantages by providing a fast and reliable response,
simple procedures, cost-effectiveness, low analysis time, and high sensitivity and selectivity [12]. Considering
these advantages, electrochemical methods have gained prominence over the years in various fields for the
detection of numerous organic substances, drugs, and various analytes in biological samples.
This study represents the first investigation of the electrochemical properties of Oxyp and Pra and their
potential effects on double-stranded DNA (dsDNA). Initially, we explored the redox properties of Oxyp and
ADMET & DMPK 00(0) (2024) 000-000 Electrochemical profiling of natural furanocoumarins
doi: https://doi.org/10.5599/admet.2199 3
Pra, followed by investigating the changes in oxidation currents and peak potentials of the guanine bases
in dsDNA before and after interaction with these compounds. To optimize analytical signals, we carefully
adjusted various experimental parameters, including concentration, pH, scan rate, and immobilization
time. Furthermore, we assessed the stability of Oxyp and Pra at 4˚C. Notably, we successfully differentiated
the signals of Oxyp and Pra in the same voltammogram, a distinctive feature of our study. Additionally, we
utilized the SwissADME web tool to predict the in silico physicochemical properties, drug-likeness, and
lead-likeness profiles of Oxyp and Pra, providing valuable insights for drug discovery and development.
Furthermore, we calculated the toxicity values of these molecules with respect to their effect on DNA. The
originality of this research lies in the comprehensive investigation of the electrochemical properties of
Oxyp and Pra and their interactions with dsDNA, offering a novel perspective for drug research and
pharmaceutical endeavours.
Experimental
Apparatus
Differential pulse voltammetry (DPV) and cyclic voltammetry (CV) measurements were conducted using
an AUTOLAB potentiostat/galvanostat/impedance analyzer. The AUTOLAB was connected to a personal
computer via a USB cable and operated using Nova software. The experimental setup consisted of a three-
electrode system, with the working electrode being a pencil graphite electrode (PGE), a reference electrode
made of Ag/AgCl, and a counter electrode composed of a platinum wire. The graphite lead was held in a
Rotring T 0.5 mm pencil (Rotring, Germany). Pencil leads of HB grade, measuring 60 mm in length and 0.5 mm
in diameter, were purchased from a local bookstore for use in the experiment. Column chromatography was
carried out using silica gel 60 (70 to 230 mesh, Merck, Germany) and Lichroprep RP 18 (25 to 40 µm, Merck,
Germany). A Varian Oxford AS400 (1H:400 MHz and 13C:100 MHz) apparatus was utilized to generate 1D
NMR spectra. Mass spectra were obtained using a MALDI-TOF-MS spectrometer.
Chemicals
Ethanol absolute (99.9 %) and glacial acetic acid were purchased from Isolab Chemicals. Fish sperm
double-stranded DNA (dsDNA) was purchased from Sigma-Aldrich (Germany), and all other chemicals were
of high purity, obtained from Merck (Darmstadt, Germany), Tokyo Chemical Industry C. LTD. (Tokyo, Japan).
All solvents used were of high analytical grade and were employed without further purification. Buffers,
including 0.5 M acetate (ACB) at pH levels of 3.8, 4.8 and 5.6, 0.05 M phosphate (PBS) at pH 7.4, 0.1 M sodium
borate (BBS) at pH 9.8, along with 0.02 M NaCl and 0.05 M Tris-EDTA (TE) buffer at pH 8.0, were utilized in
the experiments.
Oxypeucedanin and prantschimgin
In our previous study, Prangos uechtritzii was obtained from the city of Konya of Türkiye [13]. The dried
and powdered roots of the plant were sequentially extracted with n-hexane, chloroform, and methanol using
an ultrasonic water bath for 24 h at room temperature. After filtration, the extracts were evaporated to
dryness one by one at 40 °C under low pressure, yielding n-hexane (21 g), chloroform (19 g), and methanol
(31 g) extracts. Column chromatography was used for fractionation and isolation studies. Following several
chromatographic column studies, Oxyp (15 mg) and Pra (68 mg) were isolated and identified using 1D NMR
and LC-MS in accordance with the literature [13]. The extracts and isolated molecules were subsequently
lyophilized and stored at -20 °C. The chemical structures of Oxyp and Pra are shown in Figure 1.
https://doi.org/10.5599/admet.2199
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
4
Figure 1. Chemical structure of oxypeucedanin and prantschimgin.
Experimental steps
Activation of working electrodes
Before activation, PGEs were trimmed to a length of 3 cm. Subsequently, all PGEs were activated by
applying a potential of +1.4 V for 30 seconds in an acetate buffer (ACB) at pH 4.8 to minimize background
current. The PGEs, treated in this electrochemical process, were then used as working electrodes in the
subsequent experiments.
Electrochemical characterization of oxypeucedanin and prantschimgin
Before proceeding to the interaction studies of Oxyp and Pra with dsDNA, the electrochemical properties
of both molecules were characterized.
Oxyp: 1 mg/mL stock solution of Oxyp was prepared with dimethyl formamide (DMF). Subsequently, it was
diluted to a specific concentration with ACB (pH 4.8). Activated PGEs were immersed in this Oxyp solution and
left for 1 hour for passive adsorption. At the end of the duration, PGEs were rinsed with ACB to remove unbound
Oxyp from the electrode surface. Subsequently, DPV measurements were conducted. For CV measurements,
1 mg/mL stock solution of Oxyp was prepared with DMF. Subsequently, it was diluted to a specific concentration
with ACB (pH: 4.8), placed into the measurement cell and measurements were conducted.
Pra: 1 mg/mL stock solution was prepared using DMF. Subsequently, it was diluted to a specific concen-
tration with ACB (pH: 4.8). This Pra solution was placed into the measurement cell. DPV and CV
measurements were conducted by immersing activated PGEs into this solution.
Interaction
The interaction between Oxyp and Pra molecules with DNA was conducted both in the solution phase and
on the electrode surface.
At the electrode surface: Stock solution of dsDNA (1 mg/mL) was prepared with TE buffer and, then diluted
with ACB. The PGEs were immersed in the dsDNA solutions for 1 hour. After the immobilization, the
electrodes were rinsed with ACB to remove unbound dsDNA.
In the subsequent step, these dsDNA-coated electrodes were immersed in Oxyp and Pra solutions at
specific concentrations and left for 1 hour for interaction. At the end of the interaction, the electrodes were
rinsed with ACB to remove the non-adhered Oxyp and Pra to the surface. Finally, measurements were
conducted between 0.4 and 1.4 V using DPV.
In the solution phase: 1000 µg/mL of dsDNA and different concentrations of Oxyp and Pra were mixed in
a 1:1 ratio and placed in a container suitable for stirring. Then, the PGEs were immersed in the solution and
stirred at 650 rpm for 1 h at 37 °C. After the interaction of dsDNA and natural drug candidates, the PGEs were
washed with ACB.
ADMET & DMPK 00(0) (2024) 000-000 Electrochemical profiling of natural furanocoumarins
doi: https://doi.org/10.5599/admet.2199 5
Measurement
DPV measurements were performed in ACB at a scanning rate of 100 mV/s. CV measurements were
performed at a scanning rate of 100 mV/s and a time interval of 0.05 s between each potential sweep. The
experimental steps are illustrated in Figure 2.
Figure 2. Schematic representation of experimental steps: activation of PGE; Interaction of dsDNA with Oxyp
and Pra, and DPV measurements.
In silico physicochemical properties, drug-likeness, and lead-likeness prediction
The physicochemical properties, drug-likeness, and lead-likeness profiles of Oxyp and Pra were analyzed
using the SwissADME web tool (http://www.swissadme.ch/) [14]. By inputting the chemical structures of the
compounds, the tool provides valuable insights that aid in the evaluation of their pharmacokinetic profiles
and potential as drug candidates. The tool predicts various parameters for the compounds, including
molecular weight, numbers of hydrogen bond acceptor and donor topological polar surface area (TPSA),
Lipinski's rule of five compliances, gastrointestinal (GI) absorption, blood-brain barrier (BBB) permeability,
log P values, Ghose, Veber, Egan, and Muegge criteria. Additionally, it provides a bioavailability score, and
identifies PAINS and lead-likeness criteria.
Results and discussion
Electrochemical properties of oxypeucedanin and prantschimgin
In Figure 3, we present the redox properties of the compounds. A stock solution of Oxyp (1 mg/mL) was
prepared with DMF and diluted with ACB (pH:4.8). PGEs were activated and immersed in the Oxyp solution for
1 hour. The oxidation peak currents of Oxyp were subsequently measured with DPV in ACB (pH 4.8). Also, CV
measurements were taken directly by dipping activated electrodes in Oxyp solution prepared with ACB (pH 4.8).
As shown in Figure 3A and Figure 3B, Oxyp exhibits a single oxidation peak potential at approximately
+1.2 V. While attempts were made to investigate the reduction properties of Oxyp, a stable signal could not
be obtained (data not shown). Consequently, we decided to focus on the oxidation currents of Oxyp for the
rest of the study. A stock solution of Pra (1 mg/mL) was prepared with DMF and diluted with ACB (pH 4.8).
Unlike Oxyp, various parameters (concentration, immobilization time, stirring, etc.) were tested, but Pra did
https://doi.org/10.5599/admet.2199
http://www.swissadme.ch/
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
6
not immobilize on the electrode surface by passive adsorption. Therefore, this Pra solution was added to the
electrochemical measuring cell. Activated PGEs were immersed in these solutions and DPV measurements
were performed. As shown in Figure 3C, Pra exhibits a single oxidation peak potential at approximately
+1.4 V. Although attempts were made to investigate the reduction properties of Pra, a stable signal could not
be obtained either (data not shown).
Determining the exact sites where molecules undergo oxidation is challenging; however, some conclusions
can be made from the literature. It was observed that the oxidation signals of Oxyp and Pra were obtained at
peak potentials close to each other. Upon examining these common structures in the literature, it is noted that
the furan ring is electroactive and can be oxidized. The difference in the peak potentials of the oxidation signals
is presumed to arise from variations in the substituent groups attached to the furan ring.
Figure 3. (A) Differential pulse voltammograms and (B) cyclic voltammograms of blank and Oxyp-coated
electrodes. (DPV: +0.8 V to +1.4 V in ACB (pH 4.8) and CV: +0.65 V to +1.5 V at 100 mV/s scan rate with ACB
(pH 4.8)). (C) DPV of ACB (pH 4.8) (blank) and Pra solution (Pra) with a scan rate of 100 mV/s and a range of
+0.8 V to +1.4 V.
In Figure 4, optimization studies were conducted for Oxyp. In Figure 4A, two distinct methods for
immobilizing Oxyp onto the electrode surfaces were compared: passive adsorption, where activated
electrodes were dipped into the Oxyp solution, and agitation, involving dipping the electrodes with shaking.
Figure 4A demonstrates that passive adsorption led to higher oxidation currents of Oxyp compared to
agitation. Consequently, passive adsorption was selected for the subsequent phases of the study.
The impact of buffer pH and supporting electrolytes on Oxyp oxidation currents was investigated. In
Figure 4B, an Oxyp solution was prepared with different buffer solutions ranging from pH 3.8 to 9.8. The
highest Oxyp oxidation currents were recorded at pH 4.8. Consequently, pH 4.8 was selected as the optimal
solution for the preparation of Oxyp. Figure 4C illustrates the influence of the pH of the supporting electrolyte
on the oxidation currents of Oxyp. As the pH increased after 4.8, the amplitude of Oxyp oxidation currents
significantly decreased. Moreover, the Oxyp oxidation peak potentials shifted slightly to negative values,
indicating the involvement of protons in the oxidation process [15]. Since the highest oxidation peak was
achieved at pH 4.8, this value was chosen as the supporting electrolyte's pH for further investigations.
We also conducted a stability study to assess the maximum oxidation capacity of the Oxyp solution. The
Oxyp solution was prepared on day 0 and then stored at 4°C. As depicted in Figure 4D, the oxidation currents
of Oxyp decreased by approximately 32 % on the 18th day compared to day 0, demonstrating the decent stability
of Oxyp. On the 35th day of storage, the oxidation capacity of Oxyp is about 32 % compared to day 0. This
indicates that where it can be used as a potential drug, improvements should be considered to increase its
stability.
ADMET & DMPK 00(0) (2024) 000-000 Electrochemical profiling of natural furanocoumarins
doi: https://doi.org/10.5599/admet.2199 7
Figure 4. (A) DPV of oxidation currents of Oxyp with passive adsorption and agitating methods, scanning from
+0.8 V to +1.4 V at 100 mV/s. (B) DPV of Oxyp prepared with different buffer solutions. (C) DPV of oxidation
currents of Oxyp measured in different supporting electrolytes. (D) changes of Oxyp oxidation currents over a
35-day period at 4 °C, indicating the stability of Oxyp solution.
After studying the impact of pH on Oxyp oxidation currents, we explored the effect of scan rate on the
signal using CV in the range of 10-100 mV/s. As depicted in Figure 5A, Oxyp oxidation currents exhibited an
increase with higher scan rates. The Equation 1 shows the relationship between the anodic peak current (Ipa
and the scan rate (v).
Ipa = 0.3233 v + 9.2813 (R2 =0.9782) (1)
Figure 5B illustrates a linear relationship between the oxidation currents of Oxyp and the root of the scan
rate, suggesting a diffusion-controlled process. This finding indicates that the electrochemical behavior of
Oxyp is influenced by the rate at which species are transported to and from the electrode surface during the
redox process [16]. The equation is as follows:
Ip = 95.556 ν1/2 +7.3765 (R2 = 0.9766) (2)
Figure 5C shows the relationship between the log of peak potential and the log of the scan rate. The
equation is as follows:
log Ipa = 0.5921 log v + 2.2041 (R2 = 0.9859) (3)
As the scan rate increased, the peak potential of Oxyp shifted towards more positive values, indicating an
irreversible electrochemical process. This phenomenon occurs because the rate of the electrochemical
reaction is governed by the diffusion rate of the reactants to the electrode surface. The diffusion process
cannot keep up with the electrochemical reaction at higher scan rates, leading to an irreversible process [17].
This behavior aligns with findings from a previous study investigating the electrochemical detection of
furanocoumarin derivatives imperatorin and umbelliferon. In that study, the oxidation peak potentials of
imperatorin and umbelliferone also shifted towards more positive values with increasing scan rates, similar
to our observations for Oxyp [18].
https://doi.org/10.5599/admet.2199
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
8
According to Equation 3, the slope value is close to the theoretical value, e.g., 0.5, indicating that the
electrochemical process of Oxyp followed a diffusion-controlled mechanism. This result aligns with previous
research, confirming that the rate of the electrochemical reaction is primarily governed by the diffusion of
reactants to the electrode surface. The close correspondence between the experimental slope value and the
theoretical value of 0.5 provides strong evidence supporting the diffusion-controlled nature of the Oxyp
oxidation process, distinguishing it from an adsorption-controlled process [19].
Figure 5. (A) Examination of the effect of scan rate on the peak of Oxyp’s oxidation current. (B) Analysis of the
relationship between the scan rate root and peak current. (C) Investigation into the influence of scan rate on the
logarithm of peak current.
In Figure 6A, the effect of the supporting electrolyte's pH on the oxidation currents of Pra was examined.
As the pH increased, the oxidation current of the Pra exhibited a significant decrease. Additionally, the
oxidation peak potentials of Pra shifted towards more negative values, indicating the involvement of protons
in the oxidation process [15]. When the relationship between pH and oxidation peak potential is examined,
along with the structure of the molecule, and considering the possibility of oxidation occurring at the furan
ring, it is thought that during oxidation at the furan ring, 3 protons are involved per 2 electrons.
Among the tested pH values, the highest oxidation peak was observed at pH 4.8, which was selected for
further investigations. The anodic peak potential (Epa) of Pra demonstrated a linear decrease as pH increased
(Figure 6B). The relationship between Epa and pH is presented by Equation (4):
Epa= -0.0906 pH + 1.7687 (R2= 0.9769) (4)
Figure 6. (A) DPV showing the oxidation currents of Pra prepared with various supporting electrolytes. (B) Plot
illustrating the relationship between the oxidation peak currents of Pra and the pH of the supporting
electrolytes. (C) Percentage changes in the oxidation currents of Oxyp over a period of 120 min.
The slope value of 90.6 mV/pH from Equation (4) differs significantly from the optimal value of 59 mV/pH,
indicating an unequal transfer of protons and electrons in the reaction. This discrepancy could arise from
electrode process issues within the pH range, such as deprotonation or adsorption of oxidation products that
ADMET & DMPK 00(0) (2024) 000-000 Electrochemical profiling of natural furanocoumarins
doi: https://doi.org/10.5599/admet.2199 9
might block the electrode surface. Despite testing various potential ranges and scan rates, Pra's oxidation
currents were not observed in CV scan rate studies.
A stability study for Pra solution, similar to Oxyp, involved DPV measurements taken every 30 minutes for
120 minutes. In Figure 6C, the oxidation current of Pra decreased by approximately 60 % at 120 minutes
compared to the 0 minute. This signifies low stability for Pra, highlighting the need for stability-enhancing
formulations for potential drug use.
After revealing the electrochemical characteristics of Oxyp and Pra, a concentration investigation was
conducted. Various Oxyp concentrations ranging from 0.75 to 15 µg/mL were prepared, and their oxidation
currents were measured with DPV (Figure 7A). The oxidation currents exhibited an increase with rising Oxyp
concentration. Figure 7B shows that the oxidation currents of Oxyp exhibit a linear relationship between 0.75
and 15 µg/mL. The LOD was determined using the formula (3×standard deviation (SD)/(slope of calibration
curve), and the LOQ was determined using the formula (10×SD)/(slope of calibration curve), where sd
corresponds to the standard deviation calculated by regression analysis for a signal-to-noise ratio of 3. The
study revealed that the LOD was 1.3 µg/mL, and the LOQ was 4.3 µg/mL for Oxyp.
Different Pra concentrations ranging from 60 to 120 µg/mL were prepared, and their oxidation currents
were measured with DPV (Figure 7C). The oxidation currents showed an increase with higher Pra concen-
tration. Figure 7D illustrates the linear relationship between the oxidation current of Pra and concentrations
ranging from 60 to 120 µg/mL. LOD and LOQ for Pra were determined as 20 and 68 µg/mL, respectively.
Figure 7. (A) DPV and (B) calibration plot for the different concentrations of Oxyp (0.75 to 15 μg/mL). (C) DPV
and (D) calibration plot for the different concentrations of Pra (60 to 120 µg/mL).
Ensuring the sensor's selectivity for Oxyp and Pra is crucial. To assess this, a mixed solution containing
both substances was prepared and individually distinguished. Initial solutions of 400 µg/mL Oxyp and Pra
were prepared and mixed equally to form a 200 µg/mL solution of each, and separate 200 µg/mL solutions
were also prepared. DPV measurements were conducted within the range of 0.8 to 1.7 V. Figure 8 shows the
https://doi.org/10.5599/admet.2199
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
10
successful detection of distinct oxidation signals for Oxyp and Pra within the mixture, confirming the sensor's
high selectivity. Minor reduction in oxidation peak height and more positive peak potentials suggest a
potential interaction between Oxyp and Pra. When examining drug-DNA interaction studies in the literature,
the shift of peak potential values towards a more positive value as a result of interaction indicates a possible
intercalative interaction [20].
Figure 8. DPV obtained in different solutions (DPV: + 0.8 V to + 1.7 V at 100 mV/s scan rate with 0.5 s interval
time).
The interaction of Oxyp with dsDNA
The assessment of the interaction potency between potential drug molecules and DNA constitutes a
crucial aspect of pharmaceutical investigations. In our study, we tracked the oxidation currents associated
with guanine and adenine bases in the solution phase and at the electrode interface before and after
interaction with Oxyp and Pra. For the solution-phase interaction exploration, we mixed 1 mg/mL of dsDNA
with varied concentrations of Oxyp solution, maintaining a 1:1 ratio. Consequently, this yielded a final
solution containing 500 µg/mL of dsDNA, alongside the targeted Oxyp concentration. Following this, the
electrodes were immersed in 100 µL of the interaction solution, and the tubes were placed onto a thermal
shaker for 1 hour. Upon conclusion of this incubation period, we proceeded to quantify the oxidation currents
of the guanine and adenine bases utilizing DPV, both in the presence and absence of Oxyp.
We prepared Oxyp solutions with varying concentrations from 5 to 80 µg/mL (equivalent to 17.5 to
279.5 µM), which were subsequently combined with dsDNA in the solution phase. The oxidation currents
corresponding to guanine were then quantified using DPV in both the presence and absence of Oxyp. As
depicted in Figure 9, the oxidation currents of guanine exhibited a linear decrease as the concentration of
Oxyp increased. The most significant decrease in the guanine oxidation current was observed at an Oxyp
concentration of 80 µg/mL (279.5 µM). Consequently, this concentration was chosen for subsequent
experiments.
In Figure 10A, an oxidation signal was detected at approximately +1.22 V following the interaction
between Oxyp and dsDNA in the solution phase. This signal was positioned between the oxidation signal of
adenine at around +1.25 V and the Oxyp oxidation signal at approximately +1.20 V. We surmise that these
two oxidation signals (adenine and Oxyp) converged post-interaction, leading us to exclude the adenine
oxidation signal from our interaction assessment.
ADMET & DMPK 00(0) (2024) 000-000 Electrochemical profiling of natural furanocoumarins
doi: https://doi.org/10.5599/admet.2199 11
Potential, V log (COxyp / M)
Figure 9. (A) DPV from 0.9 to 1.4 V at 100 mV/s of scan rate for dsDNA before and after interaction with Oxyp
in solution phase. (B) Graph showing the linear relationship of log (Icomplex /(IDNA-Icomplex)) vs. log COxyp in the
range of 17.47 to 279.45 µM.
Figure 10. (A) DPV of Oxyp and dsDNA before and after interaction in the solution phase. (B) Histograms
depicting the average oxidation current values of the guanine base before and after interaction with Oxyp in
the solution phase. (C) DPV of Oxyp and dsDNA before and after interaction at the electrode surface. (D)
Histograms illustrating the average oxidation current values of the guanine base before and after interaction
with Oxyp at the electrode surface.
As depicted in Figure 10B, the oxidation peaks current of guanine exhibited a significant reduction
following the interaction. Specifically, the guanine oxidation peak current experienced a decrease of 63.6 %
after the interaction. Prior to the interaction, the guanine oxidation peak potential resided around +1.04 V,
and following the interaction, it shifted slightly in the negative direction to approximately +1.02 V. This
observation aligns with existing literature, suggesting an electrostatic nature of the interaction between Oxyp
and dsDNA [21]. For electrode surface interaction, we prepared a 500 µg/mL dsDNA solution in TE and ACB.
PGEs were immersed in this solution for 30 minutes, rinsed with ACB three times, and then dipped in
80 µg/mL Oxyp solution for 1 hour, followed by another ACB rinse.
As depicted in Figure 10C, following the interaction of Oxyp with dsDNA at the electrode surface, an
oxidation signal at approximately +1.22 V was noted, positioned between the adenine oxidation signal at
lo
g
(I
co
m
p
le
x
/(
I D
N
A
-I
co
m
p
le
x)
)
https://doi.org/10.5599/admet.2199
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
12
around +1.25 V and the Oxyp oxidation signal at about +1.20 V. Similar to the interaction in the solution phase,
it is presumed that these two oxidation signals (adenine and Oxyp) merged after the interaction. In Figure 10D,
a notable reduction in the oxidation peak current of guanine is evident following the interaction at the electrode
surface. Specifically, the guanine oxidation peak current displayed a reduction of 59 % post-interaction.
By comparing the percentage reduction in the guanine oxidation signal due to interaction in both the
solution phase and at the electrode surface, it is apparent that the interaction of Oxyp and dsDNA is more
pronounced in the solution phase, albeit with a slight disparity at the electrode surface. Similar to the
interaction studies in the solution phase, the guanine oxidation peak potential, initially situated at
approximately +1.04 V before interaction, shifts slightly more negatively to around +1.02 V after interaction.
This outcome provides further affirmation of an electrostatic mode of interaction.
The interaction of Pra with dsDNA
To assess the interaction between dsDNA and Pra in the solution phase, we mixed 1 mg/mL of dsDNA with
varying concentrations of Pra solution at a 1:1 ratio. This yielded a final solution containing 500 µg/mL of dsDNA
and the desired Pra concentration. Subsequently, 100 µL of this interaction solution was utilized, and the
electrodes were immersed. The tubes were then subjected to a thermal shaker for 1 hour. Finally, the oxidation
currents of guanine and adenine bases were measured with DPV both in the presence and absence of Pra.
As depicted in Figure 11A, no interaction was observed between dsDNA and Pra in the solution phase.
This is evident from the nearly unchanged oxidation current of the guanine base following the interaction.
Despite exploring various parameters, including different concentrations of dsDNA and Pra, as well as varying
interaction times (data not shown), the outcome remained consistent.
Figure 11. DPV of Oxyp and dsDNA before and after interaction (A) in the solution phase and (B) at the electrode surface.
To assess the interaction between dsDNA and Pra at the electrode surface, a 500 µg/mL of dsDNA solution
was prepared using TE and ACB. PGEs were immersed in this solution for 30 minutes, followed by a single
rinse with ACB. This procedure was repeated thrice. Subsequently, PGEs were immersed in the desired
concentration of Pra solution for 1 hour and then rinsed with ACB. As illustrated in Figure 11B, no interaction
was observed between dsDNA and Pra at the electrode surface, which is similar to the solution phase. This
conclusion is based on the unchanged oxidation current of the guanine base after the interaction. Despite
examining various parameters such as dsDNA and Pra concentrations, as well as interaction time (data not
shown), the interaction did not manifest.
The toxicity assessment
The toxicity value (S / %) of Oxyp on dsDNA was determined using Equation 5 after the interaction in the
solution phase and at the electrode surface:
ADMET & DMPK 00(0) (2024) 000-000 Electrochemical profiling of natural furanocoumarins
doi: https://doi.org/10.5599/admet.2199 13
S = (Sa/Sb)100 (5)
S / % Adenine peak current change
Sa / µA Magnitude of the guanine current upon interaction with Oxyp
Sb / µA Magnitude of the guanine current before interaction with Oxyp
Using Equation 5, we determined the S values to be 36 and 38 % for the solution phase and electrode surface,
respectively. According to the literature, if the S value is greater than 85 %, the substance is considered non-
toxic [22]. If the value ranges between 50 and 85 %, the molecule is moderately toxic to DNA, and if the value
is below 50 %, it is considered toxic to DNA. Our results indicated that the Oxyp molecule could be toxic to DNA.
Since there was no interaction between dsDNA and Pra, it was inferred that Pra did not exhibit toxicity towards
dsDNA. Toxicity assessment could not be conducted due to the absence of signal change.
In silico physicochemical properties, drug-likeness, and lead-likeness prediction
In silico evaluation of physicochemical, drug-likeness, and lead-likeness properties for Oxy and Pra is
presented in Table 1.
Table 1. In silico physicochemical properties and drug-likeness prediction of Oxy and Pra.
Properties Oxyp Pra
MW 286.28 g/mol 328.40 g/mol
MLOGP 1.39 2.64
HBA 5 5
HBD 0 0
TPSA 0.6511 nm2 0.6574 nm2
XLOGP3 2.24 3.93
GI Absorption High High
BBB Yes Yes
Lipinski Yes; 0 violation Yes; 0 violation
Ghose Yes Yes
Veber Yes Yes
Egan Yes Yes
Muegge Yes Yes
Bioavailability Source 0.55 0.55
PAINS 0 alert 0 alert
Leadlikeness Yes Yes
The parameters were described by using Swiss ADME. MW: molecular weight; HBA: hydrogen-bond acceptor; HBD: hydrogen-bond
donor; TPSA: topological polar surface area; MLOGP and XLOGP3: octanol-water distribution coefficients, BBB: the blood-brain
barrier.
Lipinski, Ghose, Vever, Egan and Muegge rules are guidelines used in drug discovery to assess the drug-
likeness of compounds based on factors such as molecular weight, lipophilicity, hydrogen bond donors and
acceptors, polar surface area, and number of rotatable bonds, with the aim of predicting properties related
to oral bioavailability, blood-brain barrier penetration, and overall pharmacokinetic behavior. PAINS is
abbreviation for "Pan-Assay Interference Compounds." These are chemical substances commonly found in
drug discovery screening tests that can lead to false positive results. The molecular weights of the compounds
ranged from 286.28 to 328.40 g/mol. MLOGP and XLOGP3 values fell within the ranges of 1.39 to 2.64 and
2.24 to 3.93, respectively. The compounds exhibited 5 H-bond acceptors and 0 donors. TPSA values ranged
from 0.6511 to 0.6574 nm2 (65.11 to 65.74 Å2). Both compounds displayed high gastrointestinal absorption
and adhered to Lipinski’s rules based on these parameters [23]. Neither compound violated drug-likeness
criteria from the literature as assessed by the web tool [24,25]. Furthermore, they showed no alerts against
PAINS criteria [26]. Oxyp and Pra exhibited lead-likeness properties, indicating their potential as lead
compounds for further synthesis studies.
https://doi.org/10.5599/admet.2199
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
14
Conclusions
The electrochemical investigation has shed light on Oxyp and Pra behavior, unveiling potential avenues for
DNA-targeting therapeutics. Evaluating their electrochemical properties and therapeutic potential on dsDNA
provides a basis for drug discovery. This study pioneers Oxyp's and Pra's electrochemical detection and
interaction with dsDNA. Oxyp's electrochemical properties were characterized, revealing an irreversible redox
process at +1.2 V and a diffusion-controlled mechanism. LOD and LOQ were 1.3 and 4.3 µg/mL. We explored
Oxyp's interaction with dsDNA in solution and at the electrode surface, tracking guanine's oxidation peak.
Reduction in guanine's oxidation current post-interaction suggests Oxyp-dsDNA interaction. A shift towards
more negative guanine oxidation potential indicates electrostatic interaction. Oxyp's toxicity towards dsDNA
was assessed, revealing potential toxicity. Additionally, we investigated Pra's electrochemical behavior and
dsDNA interaction. Unlike Oxyp, Pra did not exhibit significant interaction with dsDNA neither in solution nor at
the electrode surface. The oxidation of guanine remained unchanged, indicating an absence of binding affinity.
This selectivity reveals unique molecule-DNA interactions. Though Pra lacked interaction under our conditions,
its therapeutic value merits further exploration. In conclusion, this study elucidates the electrochemical
properties of Oxyp and Pra and their interactions with dsDNA. Oxyp demonstrates interaction potential with
possible toxicity, while Pra exhibits limited interaction. These findings contribute to our understanding of
bioactive molecule-DNA interactions, guiding innovative drug development and therapeutic strategies.
Conflict of interest: The authors whose names are listed in the manuscript certify that they have no affiliations
with or involvement in any organization or entity with any financial interest or non-financial interest in the
subject matter or materials discussed in this manuscript..
References
[1] G. Sevin, E. Alan, S. Demir, G. Albayrak, T. Demiroz, G. Yetik-Anacak, S. Baykan. Comparative
evaluation of relaxant effects of three prangos species on mouse corpus cavernosum: Chemical
characterization and the relaxant mechanisms of action of P. pabularia and (+)-oxypeucedanin. J
Ethnopharmacol 284 (2022) 114823. https://doi.org/10.1016/j.jep.2021.114823.
[2] M.C. Zheng, W.T. Tang, L.L. Yu, X.J. Qian, J. Ren, J.J. Li, W.W. Rong, J.X. Li, Q. Zhu. Preclinical
Pharmacokinetics and Bioavailability of Oxypeucedanin in Rats after Single Intravenous and Oral
Administration. Molecules 27 (2022). https://doi.org/10.3390/molecules27113570.
[3] S. Karakaya. Identification and Quantification of Coumarins in Four Ferulago Species (Apiaceae)
Growing in Turkey by HPLC-DAD. Journal of Pharmaceutical Sciences and Experimental Pharmacology
2017 (2017). https://doi.org/10.29199/PSEP.101016.
[4] J. Mottaghipisheh. Oxypeucedanin: Chemotaxonomy, Isolation, and Bioactivities. Plants (Basel) 10
(2021). https://doi.org/10.3390/plants10081577.
[5] L. Yang, Q. Li, Y. Feng, D. Qiu. Simultaneous Determination of Three Coumarins in Angelica dahurica
by (1)H-qNMR Method: A Fast and Validated Method for Crude Drug Quality Control. J Anal Methods
Chem 2020 (2020) 8987560. https://doi.org/10.1155/2020/8987560.
[6] G. Zhao, C. Peng, W. Du, S. Wang. Pharmacokinetic study of eight coumarins of Radix Angelicae
Dahuricae in rats by gas chromatography-mass spectrometry. Fitoterapia 89 (2013) 250-256.
https://doi.org/10.1016/j.fitote.2013.06.007.
[7] G.J. Li, H.J. Wu, Y. Wang, W.L. Hung, R.L. Rouseff. Determination of citrus juice coumarins,
furanocoumarins and methoxylated flavones using solid phase extraction and HPLC with photodiode
array and fluorescence detection. Food Chem 271 (2019) 29-38.
https://doi.org/10.1016/j.foodchem.2018.07.130.
[8] Y.-H. Hwang, H.J. Yang, J.Y. Ma. Simultaneous Determination of Three Furanocoumarins by
UPLC/MS/MS: Application to Pharmacokinetic Study of Angelica dahurica Radix after Oral
https://doi.org/10.1016/j.jep.2021.114823
https://doi.org/10.3390/molecules27113570
https://doi.org/10.29199/PSEP.101016
https://doi.org/10.3390/plants10081577
https://doi.org/10.1155/2020/8987560
https://doi.org/10.1016/j.fitote.2013.06.007
https://doi.org/10.1016/j.foodchem.2018.07.130
ADMET & DMPK 00(0) (2024) 000-000 Electrochemical profiling of natural furanocoumarins
doi: https://doi.org/10.5599/admet.2199 15
Administration to Normal and Experimental Colitis-Induced Rats. Molecules : A Journal of Synthetic
Chemistry and Natural Product Chemistry 22 (2017).
[9] L. Yang, Q. Li, Y. Feng, D. Qiu. Simultaneous Determination of Three Coumarins in Angelica
dahurica by 1H-qNMR Method: A Fast and Validated Method for Crude Drug Quality
Control. Journal of Analytical Methods in Chemistry 2020 (2020) 8987560.
https://doi.org/10.1155/2020/8987560.
[10] L. Tang, Y. Wang, J. Yang, L. Shu, H. Ding, Y. Yao, Y. Li. Enantioseparation and determination of
oxypeucedanin and its application to a stereoselective analysis in Angelica Dahuricae Radix and
pharmacokinetic study of rats. J Chromatogr B Analyt Technol Biomed Life Sci 1207 (2022) 123355.
https://doi.org/10.1016/j.jchromb.2022.123355.
[11] F. Bakar-Ates, B. Hoti, I. Gurbuz, T. Gunbatan, H. Duman, C.S. Kilic. The cytotoxic and apoptotic
effects of Ferulago W. Koch extracts on various cancer cell lines. Turkish Journal of Biochemistry 46
(2021) 281-291. https://doi.org/doi:10.1515/tjb-2020-0225.
[12] H.R.S. Lima, J.S. da Silva, E.A. de Oliveira Farias, P.R.S. Teixeira, C. Eiras, L.César C. Nunes.
Electrochemical sensors and biosensors for the analysis of antineoplastic drugs. Biosensors and
Bioelectronics 108 (2018) 27-37. https://doi.org/https://doi.org/10.1016/j.bios.2018.02.034.
[13] G. Albayrak, S. Demir. New coumarin glycosides from endemic Prangos heyniae H. Duman & M.F.
Watson. 37 (2023) 227-239. https://doi.org/10.1080/14786419.2021.1961138.
[14] [14] A. Daina, O. Michielin, V. Zoete. SwissADME: a free web tool to evaluate pharmacokinetics, drug-
likeness and medicinal chemistry friendliness of small molecules. Sci Rep 7 (2017) 42717.
https://doi.org/10.1038/srep42717.
[15] S. Aftab, S. Kurbanoglu, G. Ozcelikay, N.K. Bakirhan, A. Shah, S.A. Ozkan. Carbon quantum dots co-
catalyzed with multiwalled carbon nanotubes and silver nanoparticles modified nanosensor for the
electrochemical assay of anti-HIV drug Rilpivirine. Sensors and Actuators B: Chemical 285 (2019) 571-
583. https://doi.org/https://doi.org/10.1016/j.snb.2019.01.094.
[16] M.A. Mohamed, D.M. El-Gendy, N. Ahmed, C.E. Banks, N.K. Allam. 3D spongy graphene-modified
screen-printed sensors for the voltammetric determination of the narcotic drug codeine. Biosens
Bioelectron 101 (2018) 90-95. https://doi.org/10.1016/j.bios.2017.10.020.
[17] A.M. Fekry, M. Shehata, S.M. Azab, A. Walcarius. Voltammetric detection of caffeine in
pharmacological and beverages samples based on simple nano- Co (II, III) oxide modified carbon
paste electrode in aqueous and micellar media. Sensors and Actuators B: Chemical 302 (2020)
127172. https://doi.org/https://doi.org/10.1016/j.snb.2019.127172.
[18] J. Chen, B. Fu, T. Liu, Z. Yan, K. Li. A simple and fast method for the simultaneous determination of
Imperatorin, Isoimperatorin and Umbelliferone at β-CD–GO composite film modified GCE. Sensors
and Actuators B: Chemical 260 (2018) 650-656.
https://doi.org/https://doi.org/10.1016/j.snb.2017.12.064.
[19] P. Jakóbczyk, A. Dettlaff, G. Skowierzak, T. Ossowski, J. Ryl, R. Bogdanowicz. Enhanced stability of
electrochemical performance of few-layer black phosphorus electrodes by noncovalent adsorption of
1,4-diamine-9,10-anthraquinone. Electrochimica Acta 416 (2022) 140290.
https://doi.org/https://doi.org/10.1016/j.electacta.2022.140290.
[20] M. Sirajuddin, S. Ali, A. Badshah. Drug–DNA interactions and their study by UV–Visible, fluorescence
spectroscopies and cyclic voltametry. Journal of Photochemistry and Photobiology B: Biology 124
(2013) 1-19. https://doi.org/https://doi.org/10.1016/j.jphotobiol.2013.03.013.
[21] S. Bilge, B. Dogan-Topal, T. Taskin Tok, E.B. Atici, A. Sınağ, S.A. Ozkan. Investigation of the interaction
between anticancer drug ibrutinib and double-stranded DNA by electrochemical and molecular
docking techniques. Microchemical Journal 180 (2022) 107622.
https://doi.org/https://doi.org/10.1016/j.microc.2022.107622.
[22] G. Bagni, D. Osella, E. Sturchio, M. Mascini. Deoxyribonucleic acid (DNA) biosensors for
environmental risk assessment and drug studies. Anal Chim Acta 573-574 (2006) 81-89.
https://doi.org/10.1016/j.aca.2006.03.085.
https://doi.org/10.5599/admet.2199
https://doi.org/10.1155/2020/8987560
https://doi.org/10.1016/j.jchromb.2022.123355
https://doi.org/doi:10.1515/tjb-2020-0225
https://doi.org/https:/doi.org/10.1016/j.bios.2018.02.034
https://doi.org/10.1080/14786419.2021.1961138
https://doi.org/10.1038/srep42717
https://doi.org/https:/doi.org/10.1016/j.snb.2019.01.094
https://doi.org/10.1016/j.bios.2017.10.020
https://doi.org/https:/doi.org/10.1016/j.snb.2019.127172
https://doi.org/https:/doi.org/10.1016/j.snb.2017.12.064
https://doi.org/https:/doi.org/10.1016/j.electacta.2022.140290
https://doi.org/https:/doi.org/10.1016/j.jphotobiol.2013.03.013
https://doi.org/https:/doi.org/10.1016/j.microc.2022.107622
https://doi.org/10.1016/j.aca.2006.03.085
H. O. Kaya et al. ADMET & DMPK 00(0) (2024) 000-000
16
[23] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney. Experimental and computational approaches to
estimate solubility and permeability in drug discovery and development settings. Advanced Drug
Delivery Reviews 23 (1997) 3-25. https://doi.org/https://doi.org/10.1016/S0169-409X(96)00423-1.
[24] I. Muegge, S.L. Heald, D. Brittelli. Simple Selection Criteria for Drug-like Chemical Matter. Journal of
Medicinal Chemistry 44 (2001) 1841-1846. https://doi.org/10.1021/jm015507e.
[25] A.K. Ghose, V.N. Viswanadhan, J.J. Wendoloski. A Knowledge-Based Approach in Designing
Combinatorial or Medicinal Chemistry Libraries for Drug Discovery. 1. A Qualitative and Quantitative
Characterization of Known Drug Databases. Journal of Combinatorial Chemistry 1 (1999) 55-68.
https://doi.org/10.1021/cc9800071.
[26] J.B. Baell, G.A. Holloway. New Substructure Filters for Removal of Pan Assay Interference Compounds
(PAINS) from Screening Libraries and for Their Exclusion in Bioassays. Journal of Medicinal Chemistry
53 (2010) 2719-2740. https://doi.org/10.1021/jm901137j.
©2024 by the authors; licensee IAPC, Zagreb, Croatia. This article is an open-access article distributed under the terms and
conditions of the Creative Commons Attribution license (http://creativecommons.org/licenses/by/3.0/)
https://doi.org/https:/doi.org/10.1016/S0169-409X(96)00423-1
https://doi.org/10.1021/jm015507e
https://doi.org/10.1021/cc9800071
https://doi.org/10.1021/jm901137j
http://creativecommons.org/licenses/by/3.0/