**1. Introduction**

Gynecological cancers are life-endangering malignancies of the female reproductive system, discriminated as cervical, ovarian, uterine, vaginal, vulvar, and fallopian tube cancer after the a ffected anatomical organ. These diseases are estimated to cause approximately 110,000 new cases and 33,000 deaths in the United States alone on an annual basis [1].

The successful treatment of cancers requires a combination of invasive and non-invasive processes such as surgery, chemotherapy and radiation, considering e fficient treatment and patient's quality of life [2]. Chemotherapy is a major constituent of the multidisciplinary cancer therapy, evolved throughout the years from a single active compound remedy to a multifactor approach, applying combined regimens towards personalized medicine. One of the most frequently used chemotherapeutic agents in cancer therapy is 5-fluorouracil (5-FU), an analogue of uracil [3,4]. The anticancer activity of 5-FU originates from the inhibition of thymidylate synthase (TS) activity during the S phase of the

cell cycle and its incorporation into RNA and DNA of tumor cells [5,6]. Furthermore, 5-FU promotes cell death by generating mitochondrial ROS in the p53-dependent pathway [7–9] and by inducing apoptosis through the activation of a cascade of caspases 1, 3 and 8 [10].

Cancer cells have the ability to prevent programmed cell death by disorganizing tissue homeostasis, the balance between cell proliferation and cell death [11]. Therefore, a fundamental target of conventional chemotherapy is the activation of endocellular signaling mechanisms involved in cell death pathways, in particular, those mediating apoptosis. 5-fluorouracil is an e ffective pharmaceutical agen<sup>t</sup> reported to initiate apoptotic processes against a number of malignancies such as colorectal [12], oral [13], breast [14], head and neck [15], gastric [16] and cervical carcinomas [17,18]. A crucial element of apoptotic cell death is caspase-3, a cysteine protease that catalyzes a number of key endocellular proteins [19]. Human caspases are a group of eleven endoproteases, caspases 1–10 and caspase 14, controlling cell regulatory networks of inflammation and cell death [20]. The executioner caspase-3 is activated from initiator caspases, the outcome of apoptotic cell signaling driven from either mitochondrial cytochrome c release or cell death receptor activation [21]. The anticancer e fficacy of 5-FU, when administered alone or in combination with other agents, is associated with caspase-3 function as reported in several in vivo and in vitro studies regarding colorectal [22], gastric [23], pancreatic [24] and gynecological malignancies [25,26].

Cell culture monitoring can be used as indicator for the response to di fferent chemotherapy options. The use of biosensors, in particular bioelectric and electrochemical sensors in the analysis of antineoplastic drugs has increased in importance over the last years [27,28]. In this context, a critical marker for monitoring cancer cells di fferentiation within a cell population is the superoxide anion. This molecule is mainly a by-product of the oxidative phosphorylation of the mitochondria electron transport chain. Initially, it is released to the mitochondrial matrix, where it is converted immediately to hydrogen peroxide. Mitochondrial hydrogen peroxide can then di ffuse to both the cytosol and the nucleus and either interact with other free radical species, modulate signaling cascades or cause cellular damage. Together with other free radical species, superoxide has been found to mediate the development and/or survival of cancer cells and tumors, both in vivo and in vitro [29–31]. This property of superoxide has led researchers to propose the regulation of cellular redox status as a novel, critical and highly e fficient cancer therapeutic strategy [32,33], utilizing superoxide dismutase along with other antioxidant systems [34,35].

Chemotherapy is associated with oxidative stress alterations that a ffect vital cellular processes such as cell cycle progression and drug-induced apoptosis [36–38]. The scope of this study is to investigate whether it is possible to employ a biosensor-based approach to detect in a non-invasive way the superoxide levels generated by cervical cancer cells after exposure to the anticancer agen<sup>t</sup> 5-fluorouracil. For this purpose, di fferent 5-FU concentrations were tested towards the HeLa cervical cancer line for 24 and 48 h. After the determination of 5-FU's cytotoxic activity cellular stress markers such as mitochondrial superoxide and caspase-3 levels were determined. Superoxide levels were determined in parallel in the culture medium with an advanced cell-based bioelectric biosensor, an approach which has been previously applied to monitoring superoxide levels in cultures of di fferentiating neuronal cells [39]. In this way, we demonstrate that it is possible to access in a high throughput, non-invasive way the in vitro efficacy of target anticancer compounds with a cytotoxic principle based on free radical production.

#### **2. Materials and Methods**

#### *2.1. Cell Line and Culture Conditions*

HeLa (ATCC ® CCL-2 ™) and Vero cell lines were originally purchased from the American Type Culture Collection (ATCC) (Manassas, VA, USA). The cells were cultured in Dulbecco's Modified Eagle Medium (BiochromGmbh, Berlin, Germany) supplemented with 10% fetal bovine serum (Thermo Fisher Scientific, Waltham, MA, USA), 2 mM L-glutamine, 0.5 mM sodium pyruvate and 1% penicillin/streptomycin, all procured from Biowest (Biowest, Nuaillé, France). Cells were incubated in a 5% CO2 incubator (HF90Air jacketed CO2 Incubator, Heal Force Bio-Meditech Holdings Limited, Shanghai, China) at 37 ◦C for proliferation.

#### *2.2. MTT Cell Proliferation Assay*

Cells were plated in transparent flat-bottom 96-well plates (SPL Life Sciences Co Ltd., Naechon-Myeon, Korea) at two di fferent population densities according to two incubation time intervals (24 and 48 h): 10<sup>3</sup> and 8 × 10<sup>3</sup> cells per well, both supplemented with 100 μL of culture medium. The next day the medium was replaced with 200 μL of medium supplemented with 1% FBS that contained the di fferent 5-fluorouracil (5-FU) concentrations. Cells not treated with 5-FU were considered as control (0). The cytotoxic agen<sup>t</sup> 5-FU was initially diluted in dimethyl-sulfoxide. After 24 and 48 h incubation 0.5 mg/mL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) (Duchefa Biochemie, Haarlem, the Netherlands) was added in each well. After 3 h in culture, the MTT-containing medium was removed and cells were solubilized with 200 μL dimethyl sulfoxide (DMSO). Cell viability was determined by measuring the absorbance at 560 nm wavelength, using a PowerWave 240 microplate photometer (Biotek, Winooski, VT, USA). The results were expressed as the percentage of absorbance values compared to control and were assessed to determine the changes in viability.

#### *2.3. Measurement of Mitochondrial Superoxide Production*

Mitochondria-targeted MitoSOX ™ Red fluorogenic dye (Thermo Fisher Scientific, Rockford, IL, USA) was used to measure mitochondrial superoxide accumulation according to the manufacturer's instructions. Briefly, cells were seeded in 96-well black plates (at the same densities indicated at the MTT viability assessment) with clear bottom and were left for 24 and 48 h incubation with the bioactive compounds. Antimycin-A (50 μM) was used as a positive control. After overnight incubation, the medium was aspirated, and cells were incubated for 10 min at 37 ◦C in 0.2 mL of measurement bu ffer containing 5 μM MitoSOX ™ Red. Then, the cells were washed twice with PBS. MitoSOX ™ fluorescence was measured at 510 nm excitation and 580 nm emission wavelengths on an Infinite M200PRO multimode microplate reader (Tecan Group Ltd., Männedorf, Switzerland). MitoSOX ™ fluorescence was adjusted based on total protein content as determined by the Bradford assay [40] measured at an absorbance of 595 nm. The results were expressed as the percentage of adjusted values compared to control.

#### *2.4. Measurement of Caspase Activity*

The activity of caspase-3 enzymes was measured in cell lysates by a colorimetric assay kit (CASP-3-C, SIGMA-ALDRICH, Saint Louis, MO, USA), that is based on the hydrolysis of the compound acetyl-Asp-Glu-Val-Asp p-nitroanilide (Ac-DEVD-pNA) (A 2559, Sigma-Aldrich, Saint Louis, MO, USA) by caspase-3. The reaction releases the p-nitroaniline moiety (p-NA) that absorbs at 405 nm. Cells were pre-incubated with 5 μM doxorubicin hydrochloride (DOX) (Sigma–Aldrich, Deisenhofen, Germany) for 3 h for caspase cascade initiation [41,42]. Samples (10 μL) were tested both with and without the caspase-3 inhibitor acetyl-Asp-Glu-Val-Asp-al (Ac-DEVD-CHO, 20 μM final concentration), in a total reaction volume of 100 μL in 96-well plates. The substrate Ac-DEVD-pNA concentration was 200 μM and the assay was performed at 37 ◦C for 90 min. The caspase-3 specific activity was expressed as the percentage of μmol of p-nitroaniline released per min per μg of total protein values compared to control.

#### *2.5. HeLa Cell Total Protein Extraction*

For total protein extraction, cells were lysed in assay bu ffer (20 mM HEPES, pH 7.4, 2 mM EDTA, 0.1% CHAPS, 5 mM DTT) containing protease inhibitors (11873580001; Roche Diagnostics; Mannheim, Germany). Protein concentration was determined by the Bradford assay [40]. Briefly, the medium was

removed from cells cultured in six-well plates and cells were washed with PBS. Then, 50 μL of assay buffer was added in each well and cells were left for 20 min in −20 ◦C. After incubation, lysed cells were centrifuged at 15,000× *g* for 15 min at 4 ◦C and the supernatants were transferred to Eppendorf tubes. The cytoplasmic proteins were maintained at −80 ◦C until use.

#### *2.6. Creation of Membrane-Engineered Cells and Bioensor Fabrication (Vero-SOD)*

Membrane-engineered mammalian cells were created by the electroinsertion of the enzyme superoxide dismutase (SOD) into the membrane of Vero cell fibroblasts following the protocol of Moschopoulou et al. [43]. Initially, cells at a density of 3 × 10<sup>6</sup> mL−<sup>1</sup> were centrifuged at 1000 rpm for 2 min and the pellet was resuspended in PBS (pH 7.4). Afterwards, cells were incubated with 1500 U·mL−<sup>1</sup> CuZnSOD (EC1.15.1.1) for 20 min at 4 ◦C and the mixture was transferred to electroporator (Eppendorf Eporator, Eppendorf AG, Germany) cuvettes. Electroinsertion was performed by applying four pulses of an electric field at 1800 V·cm<sup>−</sup>1. Then, cells were centrifuged at 1000 rpm for 2 min and resuspended in cell culture medium. Finally, the sensors were fabricated by mixing 1 volume of Vero-SOD cells with 2 volumes of 4% (w/v) sodium alginate solution and was added dropwise with the use of a 22G syringe in 0.8 M CaCl2. Cells were immobilized in calcium alginate, forming beads containing 75 × 10<sup>3</sup> cells per bead with an approximate diameter of 2 mm. As already reported [39,43], the membrane potential of membrane-engineered Vero cell fibroblasts is affected by the interactions of electroinserted SOD molecules and superoxide anions, producing measurable changes in the membrane potential.

#### *2.7. Biosensor Setup for Recording Superoxide Concentration and Data Processing*

For recording the signal and processing of data, the PMD-1608FSA/D card (Measurement Computing, Norton, MA, USA) recording device and the software InstaCal (Measurement Computing) were used, respectively. A two-electrode system (working and reference) was connected to the device. These silver electrodes were electrochemically coated with an AgCl layer. A cell-bearing bead was attached to the working electrode while a cell-free bead was connected to the reference electrode. For each assay, both beads (sensor system, Figure 1) were immersed into the well containing adherent cells [39] and the response of each biosensor potential was achieved within 100 s after its sinking into the culture medium. The biosensor was calibrated with known superoxide concentration produced by the oxidation of xanthine by the xanthine oxidase. The range of xanthine concentration was from 1 pM to 10 nM and the xanthine oxidase was 100 mU/mL [43]. Each response was expressed as the average of the cellular membrane potential of each assay, which has been calibrated to correspond to relative changes in superoxide concentration.

**Figure 1.** Cell-based biosensor system configuration.

## *2.8. Statistical Analysis*

Each experiment was repeated independently three times for each treatment with <sup>n</sup>=5. Significance testing in comparisons was based on Student's *t*-tests for pairs. *p*-values < 0.05 were considered to be statistically significant.
