Toward in Situ, Electrochemical Monitoring of Extracellular H₂O₂ and NO in Cell Cultures Using Graphene Ink and Its Heterostructure with Iron-Sulfide

Reactive oxygen and nitrogen species (RO/NS), such as hydrogen peroxide (H₂O₂), nitric oxide (NO), and the less stable molecules ONOO^-, O₂·^-, and OH·, play crucial roles in human physiology, including cellular signal transduction, immune response against foreign species, aging, diabetes, and cancer.^1–4 In particular, the local concentration of RO/NS plays a critical role in regulation of cancer cells.^5–7 Thus, a better understanding of the spatial and temporal production of RO/NS is important for elucidating their contributions in tumor development and progression, as well as in other biological contexts. Developing label-free, scalable, and low-cost sensors to detect these species could enable early diagnosis and minimization of the deleterious effects associated with these diseases. Given the redox-active nature of H₂O₂ and NO (stable, model RO/NS molecules), electrochemical sensors offer many advantages for their detection. Electrochemical sensors can provide label-free and rapid readout with signal transduction that is conducive to integrated circuit technology, thereby positioning them at the forefront of next-generation, in vitro diagnostics and therapeutic drug screening assays.

In this work, we have developed electrochemical sensors based on graphene ink for detecting the oxidation of a model RNS - nitric oxide - and graphene ink/iron-sulfide (FeS_x) for the reduction of a model ROS - hydrogen peroxide. Using cyclic voltammetry (CV) with a NO donor molecule, spermine NONOate,⁸ an oxidation wave can be seen at 0.6 V vs. Ag/AgCl (Figure 1a) with graphene. NO concentrations down to ~ 5 nM are detectable using graphene. Further modification of graphene with FeS_x, in order to leverage Fenton chemistry, enables the detection of H₂O₂ reduction down to ~ 5 pM (Figure 1b). Furthermore, we tested the graphene/FeS_x sensors for monitoring the leptin-induced, temporal ROS release profile of MDA-MB-231 breast cancer cells. A clear spike in the electrochemical signal is seen after 10 mins of incubation with 100 ng/ml leptin, followed by a slight decline and plateau in the signal over 60 mins. A similar trend is found using the DCFDA fluorescent assay (general ROS assay), and the trend agrees well with previous reports,⁹ thus demonstrating the efficacy of graphene/FeS_x for monitoring complex biological samples.

In order to demonstrate the sensor application for in situ monitoring of release of cellular RO/NS, substrate biocompatibility was evaluated by examining cell adhesion, proliferation, and oxidative stress for graphene and graphene/FeS_x materials. Both tumorigenic (MDA-MB-231) and healthy (MCF-10A) breast cell lines were studied. Figures 1c-e show optical images of the MDA-MB-231 cells grown on each material. In general, only minor differences in cell morphology are seen for cells grown on graphene and graphene/FeS_x compared with control (bare tissue culture plate), indicating good adhesion to the materials. Furthermore, the cellular proliferation and oxidative stress was quantified using MTS and DCFDA assays, respectively. Both materials show slightly decreased viability compared with the control, although a good degree of viability is still maintained (~ 70% ± 5% for most conditions). Little increase in ROS generation is seen for the normal cells on either of the materials; however, for tumorigenic cells, the graphene/FeS_x especially shows a marked increase in ROS levels. Our results indicate that overall both materials are biocompatible using tumorigenic and healthy breast cell lines, with the graphene ink being more conducive to cell growth compared to graphene/FeS_x.

In situ sensing of NO released by stimulated cells and control experiments using NO-inhibiting agents are in progress. The sensor results will be compared with NO detection assays. The studies on graphene and graphene/FeS_x demonstrate their ability for in situ, label-free electrochemical measurements of model ROS and RNS molecules. Further translating these sensors to an on-chip, array-formatted platform can enable high-throughput, real-time monitoring of cellular response to therapeutics and environmental stressors, thereby providing a better understanding of how RO/NS contribute to cancer development and progression.

(1) Finkel, T., J. Cell Biol. 2011, 194 (1), 7–15.

(2) Darley-Usmar, V.; et al., Pharm. Res. 1996, 13 (5), 649–662.

(3) Adams, L.; et al., Exp. Biol. Med. 2015, 240 (6), 711–717.

(4) Reuter, S.; et al., Free Radic. Biol. Med. 2010, 49 (11), 1603–1616.

(5) Fukumura, D.; et al., Nat. Rev. Cancer 2006, 6 (7), 521–534.

(6) Burdon, R. H.; et al., Free Radic. Res. Commun. 1990, 11 (1–3), 65–76.

(7) Galadari, S.; et al., Free Radic. Biol. Med. 2017, 104, 144–164.

(8) Ramamurthi, A.; et al., Chem. Res. Toxicol. 1997, 10 (4), 408–413.

(9) Mahbouli, S.; et al., Oncol. Rep. 2017, 38, 3254–3264.

Figure 1