Electrochemical Response of Saccharomyces cerevisiae Corresponds to Cell Viability upon Exposure to Dioclea reflexa Seed Extracts and Antifungal Drugs

Dioclea reflexa bioactive compounds have been shown to contain antioxidant properties. The extracts from the same plant are used in traditional medical practices to treat various diseases with impressive outcomes. In this study, ionic mobility in Saccharomyces cerevisiae cells in the presence of D. reflexa seed extracts was monitored using electrochemical detection methods to link cell death to ionic imbalance. Cells treated with ethanol, methanol, and water extracts were studied using cyclic voltammetry and cell counting to correlate electrochemical behavior and cell viability, respectively. The results were compared with cells treated with pore-forming Amphotericin b (Amp b), as well as Fluconazole (Flu) and the antimicrobial drug Rifampicin (Rif). The D. reflexa seed water extract (SWE) revealed higher anodic peak current with 58% cell death. Seed methanol extract (SME) and seed ethanol extract (SEE) recorded 31% and 22% cell death, respectively. Among the three control drugs, Flu revealed the highest cell death of about 64%, whereas Amp b and Rif exhibited cell deaths of 35% and 16%, respectively, after 8 h of cell growth. It was observed that similar to SWE, there was an increase in the anodic peak current in the presence of different concentrations of Amp b, which also correlated with enhanced cell death. It was concluded from this observation that Amp b and SWE might follow similar mechanisms to inhibit cell growth. Thus, the individual bioactive compounds from the water extracts of D. reflexa seeds could further be purified and tested to validate their potential therapeutic application. The strategy to link electrochemical behavior to biochemical responses could be a simple, fast, and robust screening technique for new drug targets and to understand the mechanism of action of such drugs against disease models.


Introduction
Electrochemical detection of drugs that interact with most biological systems is an important strategy to understand cellular stresses that cause cell death [1][2][3]. Saccharomyces cerevisiae (S. cerevisiae) shares the complex internal cell structure of animal cells and serves as an ideal model for conducting

Solvent Extraction
Five grams of the seed powder was mixed with 30 mL of each solvent (70% ethanol, 70% methanol, and 100% deionized water, all solvents were obtained from Sigma-Aldrich, St. Louis, MO, USA with 99.9% purity). The mixture was then rotated on an orbital mixer for 48 h. It was later removed and then allowed to settle. The supernatants from all the extractions were freeze-dried, and the resulting powder was reconstituted with 1000 µL of 70% ethanol. UV-VIS absorption measurements were done using a JENWAY, 6705 UV-Vis Spectrophotometer (Cole-Parmer, Staffordshire, UK).

Cell Viability Measurements Using Trypan Blue Based Assay
A stock solution of 1 mg/mL of the drug or the extracts were prepared separately using dimethyl sulfoxide (≥99.7% purity) (Sigma-Aldrich, St. Louis, MO, USA). The final drug or extract concentration in 200 µL of cells ranges from 5-30 µg/mL [23]. The cells were incubated with the drugs or the extracts in time intervals ranging from 20 min to 8 h. Twenty microliters of cells were added to 20 µL of 0.2% trypan blue, prepared in PBS at pH = 7.2 and mixed thoroughly. After which 20 µL of the resulting solution was pipetted and then deposited onto the counting chamber for the cell viability studies using a Nexcelom Cellometer (Nexcelom Bioscience, Lawrence, MA, USA).
Electrochemical detection of the cells was done using cyclic voltammetry under steady-state conditions. A CheapStat potentiostat device (IO Rodeo, Pasadena, CA, USA) was used in all experiments. Interdigitated Gold Electrodes (IDEs)/Microelectrodes was purchased from Metrohm, DropSens (Llanero, Spain) and composed of two interdigitated electrodes with two connection tracks all made of gold on a glass substrate. The design of the Interdigitated electrodes allows two electrodes to fuse together, and as a result, the distance between two electrodes is reduced. The electrodes were thoroughly cleaned and polished before each measurement. The potentiostat was held at open circuit prior to each scan, and the cyclic voltammograms were obtained by scanning from 690 mV to 970 mV at a scan rate of 10 mV/s. Notably, the position of the voltammogram on the current axis gave an immediate indication of the proportions of each quantification of the redox form [17].

Structure of the Antifungal Drugs and Schematic of the Study
The chemical structures of amphotericin (Amp b), fluconazole (Flu), and rifampicin (Rif) are shown in Figure 1. Amp b is a polyene with seven adjoining trans double bonds. Flu is a synthetic triazole with fungistatic activity, whereas Rif is a semisynthetic antibiotic obtained from Streptomyces. The stepwise procedure in this work was to probe the mechanism of action of the antimicrobial drugs and plant extracts, as shown in Figure 2. First, the drug/constituents of the plant extracts were used to target the membrane environment and cause membrane depolarization, leading to the formation of pores with an increased permeability to protons and monovalent ions such as Na + and K + . The ionic transfer was captured through electrochemical detection followed by cell viability measurements to determine the correlation between ionic mobility across the biological membrane and cell death. at a scan rate of 10 mV/s. Notably, the position of the voltammogram on the current axis gave an immediate indication of the proportions of each quantification of the redox form [17].

Structure of the Antifungal Drugs and Schematic of the Study
The chemical structures of amphotericin (Amp b), fluconazole (Flu), and rifampicin (Rif) are shown in Figure 1. Amp b is a polyene with seven adjoining trans double bonds. Flu is a synthetic triazole with fungistatic activity, whereas Rif is a semisynthetic antibiotic obtained from Streptomyces. The stepwise procedure in this work was to probe the mechanism of action of the antimicrobial drugs and plant extracts, as shown in Figure 2. First, the drug/constituents of the plant extracts were used to target the membrane environment and cause membrane depolarization, leading to the formation of pores with an increased permeability to protons and monovalent ions such as Na + and K + . The ionic transfer was captured through electrochemical detection followed by cell viability measurements to determine the correlation between ionic mobility across the biological membrane and cell death.  how its e le ctrochemical re sponse (using a miniature ele ctrode) correlates to cell viability , as captured by the ce ll counting de vice (Ce llometer).

UV-VIS Spectrophotometry Studies
The UV/VIS monitoring of the extracts showed that SEE and SME were very efficient in removing the bioactive compounds, whereas SWE revealed the least as shown in Figure 3. As expected, the lower absorbance value from the water extracts was presumably due to the fact that either most of the hydrophobic compounds could not be extracted into the aqueous phase, or the bioactive compounds were not UV/VIS active. Methanol and ethanol, however, were more efficient in extracting the bioactive molecules resulting in higher pea k absorbance intensities. Nonetheless, the major UV/VIS absorbance wavelengths were in the same range (290 -293 nm) for all the extracts and confirmed previous studies of D. reflexa-that extracts had unique wavelength characteristics in the presence of antioxidants, phenolic compounds, alkaloids, flavonoids, cinnamaldehydes, benzene, and lignin derivatives [14,23,24].

UV-VIS Spectrophotometry Studies
The UV/VIS monitoring of the extracts showed that SEE and SME were very efficient in removing the bioactive compounds, whereas SWE revealed the least as shown in Figure 3. As expected, the lower absorbance value from the water extracts was presumably due to the fact that either most of the hydrophobic compounds could not be extracted into the aqueous phase, or the bioactive compounds were not UV/VIS active. Methanol and ethanol, however, were more efficient in extracting the bioactive molecules resulting in higher peak absorbance intensities. Nonetheless, the major UV/VIS absorbance wavelengths were in the same range (290-293 nm) for all the extracts and confirmed previous studies of D. reflexa-that extracts had unique wavelength characteristics in the presence of antioxidants, phenolic compounds, alkaloids, flavonoids, cinnamaldehydes, benzene, and lignin derivatives [14,23,24]. how its e le ctrochemical re sponse (using a miniature ele ctrode) correlates to cell viability , as captured by the ce ll counting de vice (Ce llometer).

UV-VIS Spectrophotometry Studies
The UV/VIS monitoring of the extracts showed that SEE and SME were very efficient in removing the bioactive compounds, whereas SWE revealed the least as shown in Figure 3. As expected, the lower absorbance value from the water extracts was presumably due to the fact that either most of the hydrophobic compounds could not be extracted into the aqueous phase, or the bioactive compounds were not UV/VIS active. Methanol and ethanol, however, were more efficient in extracting the bioactive molecules resulting in higher pea k absorbance intensities. Nonetheless, the major UV/VIS absorbance wavelengths were in the same range (290 -293 nm) for all the extracts and confirmed previous studies of D. reflexa-that extracts had unique wavelength characteristics in the presence of antioxidants, phenolic compounds, alkaloids, flavonoids, cinnamaldehydes, benzene, and lignin derivatives [14,23,24].  , and seed methanol extract (SME) revealed intense peaks all centered around 293 nm. The SWE also indicated the least peak intensity, implying a low extraction efficiency with water.

Cyclic Voltammetry and Cell Viability Studies of S. cerevisiae Cells Treated with Extracts
To test the electrochemical behavior and redox activity of the extracts, cyclic voltammetry analysis was conducted using interdigitated gold electrodes (IDEs), (Metrohm, DropSens). Briefly, the IDEs were composed of two interdigitated electrodes with two connection tracks on a glass substrate and offered several advantages, such as working with low volumes of samples and avoiding tedious polishing of solid electrodes. There were no redox peaks observed from the bare electrodes, as shown in Figure 4A (CONT, yellow), however, all the extracts showed quasi-reversible oxidation processes in 0.1% DMSO with current values that ranged from 0.10 to 0.18 mA at a scan rate of 10 mV/s, as shown in Figure 4A for SWE, red; SME, blue; and SEE, black. The SWE exhibited a higher oxidative peak current, which was shifted to the left, probably indicating that most of the bioactive compounds were oxidative species compared to those in the SEE and SME extracts. The corresponding concentration-dependent cell viability studies were conducted for each extract, and the results are shown in Figure 4B. It was observed that SWE (red) demonstrated the most cell death, followed by SME (blue) and SEE (black), and the untreated cells CONT (yellow) exhibited the least cell viability with concentrations up to 30 µg/mL and an incubation time of 8 h. It was noted that prolonging incubation beyond 8 h resulted in programmed cell death, and the cell counter continually indicated error messages.
bioactive compounds were oxidative species compared to those in the SEE and SME extracts. The corresponding concentration-dependent cell viability studies were conducted for each extract, and the results are shown in Figure 4B. It was observed that SWE (red) demonstrated the most cell death, followed by SME (blue) and SEE (black), and the untreated cells CONT (yellow) exhibited the least cell viability with concentrations up to 30 µg/mL and an incubation time of 8 h. It was noted that prolonging incubation beyond 8 h resulted in programmed cell death, and the cell counter continually indicated error messages.
When each extract was tested on S. cerevisiae cell lines, with similar concentrations ranging from 5 to 30 µg/mL, distinct anodic peak currents were recorded at 15 µg/mL for the SWE extract (purple), as shown in Figure 5A. Also, a noticeable redox activity was observed in the presence of SME (blue) and SEE (green) compared to the untreated cells (CONT, yellow) and the medium in which the cells were cultured (MED, red). The metabolites in the media also recorded modest peak currents in the same concentration range as shown in Figure 5A (MED, red). The results were interpreted in terms of one or more biological processes including increased biological membrane porosity in the presence of the extracts with SWE, exhibiting the most influx of ions at a 15 µg/mL extract concentration or the release of reactive oxygen species (ROS) as a result of the presence of the extracts. We also correlated ionic leakage to cell death by conducting cell viability studies with an extract concentration of 15 µg/mL and with cells incubated for 8 h, as shown in Figure 5B. SWE (purple) revealed cell death of about 57%, whereas SME (blue) and SEE (green) recorded about 31% and 22%, respectively, at the same concentration. Cell death was recorded in an increasing order, SEE < SME < SWE, which correlated with the oxidative peak current in the same order, suggesting that electrochemical responses from the cells might have resulted in cellular stress, leading to the highest cell death in the presence of the D. reflexa cell extract (especially SWE). When each extract was tested on S. cerevisiae cell lines, with similar concentrations ranging from 5 to 30 µg/mL, distinct anodic peak currents were recorded at 15 µg/mL for the SWE extract (purple), as shown in Figure 5A. Also, a noticeable redox activity was observed in the presence of SME (blue) and SEE (green) compared to the untreated cells (CONT, yellow) and the medium in which the cells were cultured (MED, red). The metabolites in the media also recorded modest peak currents in the same concentration range as shown in Figure 5A (MED, red). The results were interpreted in terms of one or more biological processes including increased biological membrane porosity in the presence of the extracts with SWE, exhibiting the most influx of ions at a 15 µg/mL extract concentration or the release of reactive oxygen species (ROS) as a result of the presence of the extracts. We also correlated ionic leakage to cell death by conducting cell viability studies with an extract concentration of 15 µg/mL and with cells incubated for 8 h, as shown in Figure 5B. SWE (purple) revealed cell death of about 57%, whereas SME (blue) and SEE (green) recorded about 31% and 22%, respectively, at the same concentration. Cell death was recorded in an increasing order, SEE < SME < SWE, which correlated with the oxidative peak current in the same order, suggesting that electrochemical responses from the cells might have resulted in cellular stress, leading to the highest cell death in the presence of the D. reflexa cell extract (especially SWE). Pe rce nt cell viability at 15µg/mL e xtract concentration with data re corded from 20 min to 8 h using trypan blue as a staining dye . SWE, purple ; SME, blue ; and SEE, gre e n we re compared to the untre ated ce lls (CONT, ye llow) afte r 8 h. Re producibility of the data was analyzed using triplicate me asurements.

Cyclic Voltammetry Studies of S. cerevisiae in the Presence of Antifungal Drugs
The extract data was compared to Amp b because the antifungal ability of the drug to cause cell death has been linked to cellular stress and ionic leakage, with numerous electrochemical studies suggesting a correlation between membrane permeability of ions to the drug mechani sm of action [24,25]. As shown in Figure 6A, cyclic voltammetry measurements of treated cells with drugs were compared to the cells alone. Amp b (blue) treated cells with concentrations ranging from 5 to 30 µg/mL showed elevated anodic responses, indicating ionic leakage from a porous membrane, as previously demonstrated [24,25]. Similar treatment with Flu (purple) showed no dramatic changes in anodic peak current in the pr esence of the drug, and this observation was the same when the cells were treated with Rif (green) after eight hours of drug exposure, compared to the Amp b current response within the margin of statistical error. Prior to the treatment of the cells with the drugs, cyclic voltammetry analysis indicated that Amp B was not redox active, yet its mechanism of action could create pores in the cellular membrane to enhance the influx of ions in and out of the membrane, indicating membrane-mediated effects (data not shown). Thus, it was confirmed that Flu and Rif were not membrane-medicated, and, therefore, membrane permeation of ions was very minimal as revealed in the data. These observations attested to earlier studies that the mechanism of action of these drugs was through different pathways, but they could still demonstrate some level of electrochemical responses [26,27].
The second phase was to correlate the electrochemical response of each drug to cell viability, as shown in Figure 6B. The cell viability percentage was recorded using trypan blue as the staining agent from 20 min to 8 h in the presence of each drug, as was performed with the extracts. The cell viability results after 20 min did not show a drastic change in cell death when treated with the drugs. However, there was a drastic change in cell viability of about 64% in the presence of Flu (purple) after eight hours. Amp b (blue) showed cell death of about 35%, whereas Rif (green) recorded about 16% cell death at concentrations of 15 µg/mL. Previous studies on Flu indicated that cell death was not membrane-mediated, and was, therefore, used as a control fungistatic drug, whereas Rif served as an antibacterial drug that has no antifungal properties. It was, therefore, concluded from this study that alterations of the endogenous membrane due to the presence of Amp b might have resulted in an enhanced influx of ions, and this correlated to the increased cell death that has already been documented [22,28]. As previous findings have indicated, many biological membranes have electrochemical characteristics, which is important for the generation of electron transfer in living systems [29][30][31][32]. Thus, endogenous chemical imbalances and an increased influx of drugs in the cell

Cyclic Voltammetry Studies of S. cerevisiae in the Presence of Antifungal Drugs
The extract data was compared to Amp b because the antifungal ability of the drug to cause cell death has been linked to cellular stress and ionic leakage, with numerous electrochemical studies suggesting a correlation between membrane permeability of ions to the drug mechanism of action [24,25]. As shown in Figure 6A, cyclic voltammetry measurements of treated cells with drugs were compared to the cells alone. Amp b (blue) treated cells with concentrations ranging from 5 to 30 µg/mL showed elevated anodic responses, indicating ionic leakage from a porous membrane, as previously demonstrated [24,25]. Similar treatment with Flu (purple) showed no dramatic changes in anodic peak current in the presence of the drug, and this observation was the same when the cells were treated with Rif (green) after eight hours of drug exposure, compared to the Amp b current response within the margin of statistical error. Prior to the treatment of the cells with the drugs, cyclic voltammetry analysis indicated that Amp B was not redox active, yet its mechanism of action could create pores in the cellular membrane to enhance the influx of ions in and out of the membrane, indicating membrane-mediated effects (data not shown). Thus, it was confirmed that Flu and Rif were not membrane-medicated, and, therefore, membrane permeation of ions was very minimal as revealed in the data. These observations attested to earlier studies that the mechanism of action of these drugs was through different pathways, but they could still demonstrate some level of electrochemical responses [26,27].
The second phase was to correlate the electrochemical response of each drug to cell viability, as shown in Figure 6B. The cell viability percentage was recorded using trypan blue as the staining agent from 20 min to 8 h in the presence of each drug, as was performed with the extracts. The cell viability results after 20 min did not show a drastic change in cell death when treated with the drugs. However, there was a drastic change in cell viability of about 64% in the presence of Flu (purple) after eight hours. Amp b (blue) showed cell death of about 35%, whereas Rif (green) recorded about 16% cell death at concentrations of 15 µg/mL. Previous studies on Flu indicated that cell death was not membrane-mediated, and was, therefore, used as a control fungistatic drug, whereas Rif served as an antibacterial drug that has no antifungal properties. It was, therefore, concluded from this study that alterations of the endogenous membrane due to the presence of Amp b might have resulted in an enhanced influx of ions, and this correlated to the increased cell death that has already been documented [22,28]. As previous findings have indicated, many biological membranes have electrochemical characteristics, which is important for the generation of electron transfer in living systems [29][30][31][32]. Thus, endogenous chemical imbalances and an increased influx of drugs in the cell can result in cellular stress, leading to cell death. Any new molecule or drug entity with the ability to increase membrane permeability of ions is likely to increase cellular stress. While it is acknowledged that cell death may show different mechanism of inhibition, it is expected from this study that plant extracts, or isolated bioactive compounds from plants, can be used to study cellular stress and correlate their activity to cell death using both electrochemical and cell viability studies [33,34]. can result in cellular stress, leading to cell death. Any new molecule or drug entity with the ability to increase membrane permeability of ions is likely to increase cellular stress. While it is acknowledged that cell death may show different mechanism of inhibition, it is expected from this study that plant extracts, or isolated bioactive compounds from plants, can be used to study cellular stress and correlate their activity to cell death using both electrochemical and cell viability studies [33,34].

Discussion
There are numerous redox mediators that are used to either study cell redox activity or develop biosensors for many biological systems. For example, S. cerevisiae has several redox centers, such as [Fe(CN)6] 3− /[Fe(CN)6] 4− and NAD(P)H/NAD(P ) + , which can be targeted by hydrophilic/hydrophobic molecules, as extensively discussed previously by Rawson et al. [35,36]. In addition, electrochemical monitoring can be conducted w hen reactive oxygen species are released as a result of drug-induced action [32,[37][38][39][40]. Reactive oxygen species is a term used to describe oxygen species, including superoxide anion radical (O2 •− ) and hydrogen peroxide (H2O2), and they can cause cytotoxic and antimicrobial effects in most organisms. Thus, there are at least three main mechanisms of drug action including, but not limited to, membrane depolarization leading to an influx of ions (as in the case of Amp b), targeting of redox centers, and/or the release of ROS in S. cerevisiae [28,35,36,41]. Amp b has been used to treat fungal infection, and the mechanism through which the antifungal drug kills infected cells is well-characterized using both biophysical and microbiological techniques [22,25]. The antifungal drug binds to ergosterol and forms pores at the cell membrane, causing the loss of ions and leading to depolarization of the membrane [13,21]. This mechanism can cause an enhanced oxidation potential, which can be captured through electrochemical detection , as already observed in our studies as well as studies from other groups [25,36].
In the case of the two other mechanisms, an enhanced anodic response can result from either targeting redox mediator centers or generating one or more ROS species as a result of drug binding, as depicted below: H2O2 ⟶ O2 (g) + 2H + (aq) + 2e − . (2) Whereas the reduction peaks seemed to deviate from a typical redox reaction, the research suggested that the effect of Amp b (35%) or the plant extracts (57%) on S. cerevisiae viability, which corresponded to the quasi-reversible oxidation process, inevitably supported a general claim that

Discussion
There are numerous redox mediators that are used to either study cell redox activity or develop biosensors for many biological systems. For example, S. cerevisiae has several redox centers, such as [Fe(CN) 6 ] 3− /[Fe(CN) 6 ] 4− and NAD(P)H/NAD(P) + , which can be targeted by hydrophilic/hydrophobic molecules, as extensively discussed previously by Rawson et al. [35,36]. In addition, electrochemical monitoring can be conducted when reactive oxygen species are released as a result of drug-induced action [32,[37][38][39][40]. Reactive oxygen species is a term used to describe oxygen species, including superoxide anion radical (O 2 •− ) and hydrogen peroxide (H 2 O 2 ), and they can cause cytotoxic and antimicrobial effects in most organisms. Thus, there are at least three main mechanisms of drug action including, but not limited to, membrane depolarization leading to an influx of ions (as in the case of Amp b), targeting of redox centers, and/or the release of ROS in S. cerevisiae [28,35,36,41]. Amp b has been used to treat fungal infection, and the mechanism through which the antifungal drug kills infected cells is well-characterized using both biophysical and microbiological techniques [22,25]. The antifungal drug binds to ergosterol and forms pores at the cell membrane, causing the loss of ions and leading to depolarization of the membrane [13,21]. This mechanism can cause an enhanced oxidation potential, which can be captured through electrochemical detection, as already observed in our studies as well as studies from other groups [25,36].
In the case of the two other mechanisms, an enhanced anodic response can result from either targeting redox mediator centers or generating one or more ROS species as a result of drug binding, as depicted below: NADPH ←→ Whereas the reduction peaks seemed to deviate from a typical redox reaction, the research suggested that the effect of Amp b (35%) or the plant extracts (57%) on S. cerevisiae viability, which corresponded to the quasi-reversible oxidation process, inevitably supported a general claim that Amp b (57%) and the plant extracts (SWE, SME, SEE at 57%, 31%, and 22%, respectively) behave similarly and have the ability to create ionic pores in fungal cells that leads to their death. Our data was reproducible because it used triplicate measurements from the same extract or from different extracts of the same plant, with error margins of less than 5% confidence. The selectivity towards the extracts was validated using the non-membrane-targeted antimicrobial drug, Rif, with the extracts, revealing up to 57% antifungal activity compared to 16% of the antimicrobial drug. These studies provide a sensitive sensing method that can electrochemically detect plant extracts, causing either depolarization of membranes, accumulation of ROS, or targeting redox centers in S. cerevisiae cell cultures. Whereas the fabrication of microelectrodes for electrochemical studies has been going on for one or two decades in various fields, there is limited application in the natural product field. The limited evidence of the application of electrochemical methods in natural products screening, as shown in Table 1, might be due to the high cost of operation [42]. As shown in Table 1, our method offers a simple and straightforward screening platform to select active natural product pools in a fast and robust fashion prior to undertaking further validation studies [43,44].

Conclusions
At a concentration of 15 µg/mL, S. cerevisiae cell lines revealed unique oxidation peak responses at 0.34, 0.25, and 0.23 in the presence of SWE, SME, and SEE of the D. reflexa extracts, respectively, correlating to cell death of 57%, 31%, and 22% in that order. The results were comparable to Amp b on S. cerevisiae cell death, where the highest oxidation peak current was directly related to inhibitory effects. Thus, one or many bioactive compounds in D. reflexa might induce cell death through similar mechanisms as Amp b. Future work will focus on the individual bioactive compounds from the water extracts of D. reflexa seeds to validate their potential therapeutic application. In conclusion, the current studies have highlighted the robustness of electrochemical detection in monitoring cell death using microliter volumes of sample.