Non-enzymatic and highly sensitive lactose detection utilizing graphene field-effect transistors
Introduction
Graphene field-effect transistors (G-FETs) have been employed as ultrasensitive, label-free sensors for the electrical detection of a variety of analytes (Justino et al., 2017; Peña-Bahamonde et al., 2018). By functionalizing the graphene surface with a biological recognition element, the sensor response is engineered to be highly specific (Matsumoto et al., 2014; Fu et al., 2017). Graphene is an attractive nanomaterial for FET biosensors due to its high conductivity and ease of functionalization (Suvarnaphaet and Pechprasarn, 2017). Every atom in a graphene sheet is in contact with the solution environment and responds to external electrostatic fluctuations, making G-FETs highly sensitive with limits of detection (LoD) for proteins or DNA at attomolar to femtomolar concentrations (Kim et al., 2013; Cai et al., 2015; Lei et al., 2017; Campos et al., 2019).
G-FETs have also been employed as sensors for glucose (Viswanathan et al., 2015), fructose (Zhao et al., 2019), and lactose (Nguyen et al., 2016). In these electrochemical biosensors, the sensor response is generated from a change in hydrogen peroxide upon oxidation of the target carbohydrate, catalyzed with enzymes or metal NPs attached to the graphene. Despite recent achievements in G-FET carbohydrate biosensor development, their LoD has been limited to nanomolar or micromolar levels (Liao et al., 2013; Cagang et al., 2016).
G-FET carbohydrate biosensor development could take advantage of the natural occurring proteins that bind specific carbohydrates, namely lectins. Lectins recognize the terminal or penultimate sugar residue, their glycosidic linkages (α or β) and possibly associated functional groups (e.g. acetyl). While few lectins bind glycans with medically useful affinity, i.e. at least nanomolar affinity, a vast majority bind with low affinity in the micromolar range (Varki et al., 2017). This is primarily because the non-covalent lectin-glycan interactions are mediated by weak H-bonds or van der Waals contacts. However, nature compensates this handicap with multivalent interactions involving oligomeric carbohydrate recognition domains (CRDs) that simultaneously interact with several glycans on the same cell surface, thus leading to an avidity effect and consequently to tight and specific binding (Drickamer, 1995; Lee and Lee, 1995; Kiessling and Pohl, 1996). The paramount challenge is to detect a very low concentration of glycans (nM concentration or even lower) with a single lectin. This task can be achieved by exploiting the advantages of a G-FET sensor.
Herein, we choose human galectin 3 (hGal-3) among all the natural lectins since hGal-3 is very well studied and extensive structural information is currently available. This structural information is essential to plan the mutations for the functionalization of the G-FET device. The C-terminal CRD of hGal-3, which shows close structural homology between the galectins, specifically binds to β-galactoside residues. Among the β-galactosides, hGal-3 binds the disaccharide lactose with a dissociation constant in the range of 150–250 μM (Saraboji et al., 2012). Lactose is composed of galactose and glucose subunits and is a very important dietary disaccharide found naturally in the milk of most mammals.
A system that combines a lectin such as Gal-3 and a G-FET sensor will be a powerful tool to recognize carbohydrates with high specificity and sensitivity. In this work, we choose to test our biosensor for lactose which is the minimum glycan unit that Gal-3 can bind with lowest affinity. This choice allows us to understand how effective the system is at detecting low affinity binders.
Section snippets
Graphene FET fabrication
Monolayer graphene was synthesized using a chemical vapor deposition (CVD) system, transferred to Si:SiO2 substrates using a lamination procedure described in (Shivayogimath et al., 2019), then patterned into G-FETs using photolithography and metal e-beam evaporation (Section S1). The graphene was confirmed to be a monolayer thin film using Raman spectroscopy (Fig. S1).
G-FET devices were decorated with gold nanoparticles (Au NPs) using a direct-current (DC) magnetron-sputtering inert-gas
Au NP decorated graphene and protein characterization
From AFM images taken after Au NP decoration of graphene sheets using the magnetron sputtering system (Fig. 2a), we see an even distribution of Au NPs is present over the monolayer graphene. Root-mean-squared (RMS) roughness was ~1.1 nm. From estimating the NPs density at approximately 5 x 1010 cm−2, each graphene device will present ~5 x 108 binding sites to the biological solution. The covalent bond between gold and sulfur on the mutant M249C CRD strongly attaches the protein to the Au NPs on
Conclusion
In summary, we have developed a Au NP-decorated G-FET biosensor capable of label-free and ultrasensitive detection of lactose using a liquid gate measurement. The sensor is functionalized using an engineered CRD hGal-3 mutant that attaches to the Au NP-decorated graphene surface via a thiol bond. Affinity binding of lactose to the CRD functionalized G-FET n-dopes the graphene, causing a negative shift in the minimum voltage of the G-FET when measured in a liquid-gate configuration. This shift
CRediT authorship contribution statement
Eric Danielson: Conceptualization, Methodology, Investigation, Writing - original draft, Writing - review & editing. Mirco Dindo: Methodology, Resources, Investigation, Writing - original draft, Writing - review & editing. Alexander J. Porkovich: Resources, Investigation, Writing - review & editing. Pawan Kumar: Methodology, Investigation, Writing - review & editing. Zhenwei Wang: Methodology, Investigation, Writing - review & editing. Prashant Jain: Resources, Investigation, Writing - review &
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
This work was supported by funding from the Okinawa Institute of Science and Technology Graduate University (OIST). We are grateful for the help and support provided by Dr. Alexander Badrutdinov and Dr. Hyung-Been Kang from the Engineering Support section of the Research Support Division at OIST. We thank the OIST Imaging Section for providing access to the Zeiss LSM 510 fluorescent microscope and Dr. Shinya Komoto for support. We are also grateful for data visualization help provided by Pavel
References (33)
- et al.
Anal. Chim. Acta
(2016) - et al.
Biosens. Bioelectron.
(2015) - et al.
Biosens. Bioelectron.
(2020) - et al.
Biosens. Bioelectron.
(2013) - et al.
Trac. Trends Anal. Chem.
(2017) - et al.
Chem. Biol.
(1996) - et al.
Biosens. Bioelectron.
(2017) - et al.
Curr. Appl. Phys.
(2016) - et al.
Mater. Today
(2015) - et al.
J. Electroanal. Chem.
(2019)
ACS Sens.
Acta Crystallogr. D
J. Am. Chem. Soc.
Adv. Mater.
Nat. Struct. Biol.
Adv. Mater.
Cited by (21)
Recent advances in nanomaterials based biosensors
2023, TrAC - Trends in Analytical ChemistryElectrochemical biosensors for biomolecules
2023, Handbook of Biomolecules: Fundamentals, Properties and ApplicationsA comprehensive review on graphene FET bio-sensors and their emerging application in DNA/RNA sensing & rapid Covid-19 detection
2023, Measurement: Journal of the International Measurement ConfederationCitation Excerpt :Lactose sensing is also significant because it is necessary to estimate its level in food and dairy products in order to prevent disease conditions like lactose intolerance. In the year 2020, E. Danielson et al [69] et al reported the first Gr-FET based non enzymatic lactose sensor. The fabricated liquid-gated Gr-FET was decorated with gold nanoparticles.
Future aspects of biosensor-based devices in disease detection
2022, Advanced Biosensors for Virus Detection: Smart Diagnostics to Combat SARS-CoV-2Systematic review on lectin-based electrochemical biosensors for clinically relevant carbohydrates and glycoconjugates
2021, Colloids and Surfaces B: BiointerfacesCitation Excerpt :The bienzymatic biosensing signal was negatively influenced by the co-existence of uric (76% for 2.5 ×10-5 mol L1) and ascorbic acid (23% for 5.0 ×10-6 mol L1) due to the high working potential, but the competitive analytical characteristics were significantly improved by using a Nafion® conductive membrane, decreasing the interferences to 22% and 4.5% for uric and ascorbic acid, respectively. Alternative non-enzymatic electrochemical biosensors to determine free carbohydrates can also be constructed with lectins, as reported by Danielson et al. [46]. These researchers proposed a highly sensitive field-effect transistor device based on graphene decorated with gold nanoparticles (AuNPs) for lactose detection in a liquid-gate measurement configuration.